From a7c4a547ea1de2e9d6cb82263962aaff3ba67ff5 Mon Sep 17 00:00:00 2001
From: Simon Wunderlich <sw@simonwunderlich.de>
Date: Mon, 25 Nov 2019 09:06:16 +0100
Subject: batman-adv: Start new development cycle

Signed-off-by: Simon Wunderlich <sw@simonwunderlich.de>
---
 net/batman-adv/main.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index c7b340ddd0e7..fd8c0728ddc7 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -13,7 +13,7 @@
 #define BATADV_DRIVER_DEVICE "batman-adv"
 
 #ifndef BATADV_SOURCE_VERSION
-#define BATADV_SOURCE_VERSION "2019.5"
+#define BATADV_SOURCE_VERSION "2020.0"
 #endif
 
 /* B.A.T.M.A.N. parameters */
-- 
cgit v1.2.3-59-g8ed1b


From f4191c6d80b92caf21d779429c51dd52e2a16d70 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Thu, 14 Nov 2019 22:12:37 +0100
Subject: batman-adv: Strip dots from variable macro kerneldoc

The commit 43756e347f21 ("scripts/kernel-doc: Add support for named
variable macro arguments") changed the handling of variable macro
parameters. The three dots of the argument must no longer be added to the
kernel doc. The support for the old format is scheduled to be removed in
the future.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Simon Wunderlich <sw@simonwunderlich.de>
---
 net/batman-adv/log.h | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/net/batman-adv/log.h b/net/batman-adv/log.h
index 741cfa3719ff..41fd900121c5 100644
--- a/net/batman-adv/log.h
+++ b/net/batman-adv/log.h
@@ -74,7 +74,7 @@ __printf(2, 3);
  * @bat_priv: the bat priv with all the soft interface information
  * @ratelimited: whether output should be rate limited
  * @fmt: format string
- * @arg...: variable arguments
+ * @arg: variable arguments
  */
 #define _batadv_dbg(type, bat_priv, ratelimited, fmt, arg...)		\
 	do {								\
@@ -98,7 +98,7 @@ static inline void _batadv_dbg(int type __always_unused,
  * batadv_dbg() - Store debug output without ratelimiting
  * @type: type of debug message
  * @bat_priv: the bat priv with all the soft interface information
- * @arg...: format string and variable arguments
+ * @arg: format string and variable arguments
  */
 #define batadv_dbg(type, bat_priv, arg...) \
 	_batadv_dbg(type, bat_priv, 0, ## arg)
@@ -107,7 +107,7 @@ static inline void _batadv_dbg(int type __always_unused,
  * batadv_dbg_ratelimited() - Store debug output with ratelimiting
  * @type: type of debug message
  * @bat_priv: the bat priv with all the soft interface information
- * @arg...: format string and variable arguments
+ * @arg: format string and variable arguments
  */
 #define batadv_dbg_ratelimited(type, bat_priv, arg...) \
 	_batadv_dbg(type, bat_priv, 1, ## arg)
@@ -116,7 +116,7 @@ static inline void _batadv_dbg(int type __always_unused,
  * batadv_info() - Store message in debug buffer and print it to kmsg buffer
  * @net_dev: the soft interface net device
  * @fmt: format string
- * @arg...: variable arguments
+ * @arg: variable arguments
  */
 #define batadv_info(net_dev, fmt, arg...)				\
 	do {								\
@@ -130,7 +130,7 @@ static inline void _batadv_dbg(int type __always_unused,
  * batadv_err() - Store error in debug buffer and print it to kmsg buffer
  * @net_dev: the soft interface net device
  * @fmt: format string
- * @arg...: variable arguments
+ * @arg: variable arguments
  */
 #define batadv_err(net_dev, fmt, arg...)				\
 	do {								\
-- 
cgit v1.2.3-59-g8ed1b


From 69fed4ce2e9a1686894bc23e1523a7ca1a8c74f6 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sat, 16 Nov 2019 08:28:58 +0100
Subject: batman-adv: Fix typo metAdata

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Simon Wunderlich <sw@simonwunderlich.de>
---
 net/batman-adv/types.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 47718a82eaf2..bdf9827b5f63 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -457,7 +457,7 @@ struct batadv_orig_node {
 	/**
 	 * @tt_lock: prevents from updating the table while reading it. Table
 	 *  update is made up by two operations (data structure update and
-	 *  metdata -CRC/TTVN-recalculation) and they have to be executed
+	 *  metadata -CRC/TTVN-recalculation) and they have to be executed
 	 *  atomically in order to avoid another thread to read the
 	 *  table/metadata between those.
 	 */
@@ -1011,7 +1011,7 @@ struct batadv_priv_tt {
 	/**
 	 * @commit_lock: prevents from executing a local TT commit while reading
 	 *  the local table. The local TT commit is made up by two operations
-	 *  (data structure update and metdata -CRC/TTVN- recalculation) and
+	 *  (data structure update and metadata -CRC/TTVN- recalculation) and
 	 *  they have to be executed atomically in order to avoid another thread
 	 *  to read the table/metadata between those.
 	 */
-- 
cgit v1.2.3-59-g8ed1b


From 8d985555ddaa6735c5d1f3d6e4d4d9cb47f27878 Mon Sep 17 00:00:00 2001
From: Alagu Sankar <alagusankar@silex-india.com>
Date: Fri, 15 Nov 2019 09:20:58 +0200
Subject: ath10k: enable RX bundle receive for sdio

The existing implementation of initiating multiple sdio transfers for
receive bundling is slowing down the receive speed. Combining the
transfers using a bundle method would be ideal.

The transmission utilization ratio for sdio bus for small packet is
slow, because the space and time cost for sdio bus is same for large
length packet and small length packet. So the speed of data for large
length packet is higher than small length.

Test result of different length of data:
data packet(byte)   cost time(us)   calculated rate(Mbps)
      256               28                73
      512               33               124
     1024               35               234
     1792               45               318
    14336              168               682
    28672              333               688
    57344              660               695

Tested with QCA6174 SDIO with firmware
WLAN.RMH.4.4.1-00017-QCARMSWPZ-1

Signed-off-by: Alagu Sankar <alagusankar@silex-india.com>
Signed-off-by: Wen Gong <wgong@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/sdio.c | 110 +++++++++++++++++++++------------
 drivers/net/wireless/ath/ath10k/sdio.h |  11 +++-
 2 files changed, 79 insertions(+), 42 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/sdio.c b/drivers/net/wireless/ath/ath10k/sdio.c
index 120200a93bcc..7f25079dde2d 100644
--- a/drivers/net/wireless/ath/ath10k/sdio.c
+++ b/drivers/net/wireless/ath/ath10k/sdio.c
@@ -24,6 +24,8 @@
 #include "trace.h"
 #include "sdio.h"
 
+#define ATH10K_SDIO_VSG_BUF_SIZE	(32 * 1024)
+
 /* inlined helper functions */
 
 static inline int ath10k_sdio_calc_txrx_padded_len(struct ath10k_sdio *ar_sdio,
@@ -484,11 +486,11 @@ out:
 	return ret;
 }
 
-static int ath10k_sdio_mbox_alloc_pkt_bundle(struct ath10k *ar,
-					     struct ath10k_sdio_rx_data *rx_pkts,
-					     struct ath10k_htc_hdr *htc_hdr,
-					     size_t full_len, size_t act_len,
-					     size_t *bndl_cnt)
+static int ath10k_sdio_mbox_alloc_bundle(struct ath10k *ar,
+					 struct ath10k_sdio_rx_data *rx_pkts,
+					 struct ath10k_htc_hdr *htc_hdr,
+					 size_t full_len, size_t act_len,
+					 size_t *bndl_cnt)
 {
 	int ret, i;
 
@@ -529,6 +531,7 @@ static int ath10k_sdio_mbox_rx_alloc(struct ath10k *ar,
 	size_t full_len, act_len;
 	bool last_in_bundle;
 	int ret, i;
+	int pkt_cnt = 0;
 
 	if (n_lookaheads > ATH10K_SDIO_MAX_RX_MSGS) {
 		ath10k_warn(ar,
@@ -572,20 +575,19 @@ static int ath10k_sdio_mbox_rx_alloc(struct ath10k *ar,
 			 */
 			size_t bndl_cnt;
 
-			ret = ath10k_sdio_mbox_alloc_pkt_bundle(ar,
-								&ar_sdio->rx_pkts[i],
-								htc_hdr,
-								full_len,
-								act_len,
-								&bndl_cnt);
+			ret = ath10k_sdio_mbox_alloc_bundle(ar,
+							    &ar_sdio->rx_pkts[pkt_cnt],
+							    htc_hdr,
+							    full_len,
+							    act_len,
+							    &bndl_cnt);
 
 			if (ret) {
 				ath10k_warn(ar, "alloc_bundle error %d\n", ret);
 				goto err;
 			}
 
-			n_lookaheads += bndl_cnt;
-			i += bndl_cnt;
+			pkt_cnt += bndl_cnt;
 			/*Next buffer will be the last in the bundle */
 			last_in_bundle = true;
 		}
@@ -597,7 +599,7 @@ static int ath10k_sdio_mbox_rx_alloc(struct ath10k *ar,
 		if (htc_hdr->flags & ATH10K_HTC_FLAGS_RECV_1MORE_BLOCK)
 			full_len += ATH10K_HIF_MBOX_BLOCK_SIZE;
 
-		ret = ath10k_sdio_mbox_alloc_rx_pkt(&ar_sdio->rx_pkts[i],
+		ret = ath10k_sdio_mbox_alloc_rx_pkt(&ar_sdio->rx_pkts[pkt_cnt],
 						    act_len,
 						    full_len,
 						    last_in_bundle,
@@ -606,9 +608,11 @@ static int ath10k_sdio_mbox_rx_alloc(struct ath10k *ar,
 			ath10k_warn(ar, "alloc_rx_pkt error %d\n", ret);
 			goto err;
 		}
+
+		pkt_cnt++;
 	}
 
-	ar_sdio->n_rx_pkts = i;
+	ar_sdio->n_rx_pkts = pkt_cnt;
 
 	return 0;
 
@@ -622,59 +626,76 @@ err:
 	return ret;
 }
 
-static int ath10k_sdio_mbox_rx_packet(struct ath10k *ar,
-				      struct ath10k_sdio_rx_data *pkt)
+static int ath10k_sdio_mbox_rx_fetch(struct ath10k *ar)
 {
 	struct ath10k_sdio *ar_sdio = ath10k_sdio_priv(ar);
+	struct ath10k_sdio_rx_data *pkt = &ar_sdio->rx_pkts[0];
 	struct sk_buff *skb = pkt->skb;
 	struct ath10k_htc_hdr *htc_hdr;
 	int ret;
 
 	ret = ath10k_sdio_readsb(ar, ar_sdio->mbox_info.htc_addr,
 				 skb->data, pkt->alloc_len);
-	if (ret)
-		goto out;
 
-	/* Update actual length. The original length may be incorrect,
-	 * as the FW will bundle multiple packets as long as their sizes
-	 * fit within the same aligned length (pkt->alloc_len).
-	 */
-	htc_hdr = (struct ath10k_htc_hdr *)skb->data;
-	pkt->act_len = le16_to_cpu(htc_hdr->len) + sizeof(*htc_hdr);
-	if (pkt->act_len > pkt->alloc_len) {
-		ath10k_warn(ar, "rx packet too large (%zu > %zu)\n",
-			    pkt->act_len, pkt->alloc_len);
-		ret = -EMSGSIZE;
-		goto out;
+	if (ret) {
+		ar_sdio->n_rx_pkts = 0;
+		ath10k_sdio_mbox_free_rx_pkt(pkt);
+		return ret;
 	}
 
-	skb_put(skb, pkt->act_len);
-
-out:
+	htc_hdr = (struct ath10k_htc_hdr *)skb->data;
+	pkt->act_len = le16_to_cpu(htc_hdr->len) + sizeof(*htc_hdr);
 	pkt->status = ret;
+	skb_put(skb, pkt->act_len);
 
 	return ret;
 }
 
-static int ath10k_sdio_mbox_rx_fetch(struct ath10k *ar)
+static int ath10k_sdio_mbox_rx_fetch_bundle(struct ath10k *ar)
 {
 	struct ath10k_sdio *ar_sdio = ath10k_sdio_priv(ar);
+	struct ath10k_sdio_rx_data *pkt;
+	struct ath10k_htc_hdr *htc_hdr;
 	int ret, i;
+	u32 pkt_offset, virt_pkt_len;
 
+	virt_pkt_len = 0;
+	for (i = 0; i < ar_sdio->n_rx_pkts; i++)
+		virt_pkt_len += ar_sdio->rx_pkts[i].alloc_len;
+
+	if (virt_pkt_len > ATH10K_SDIO_VSG_BUF_SIZE) {
+		ath10k_warn(ar, "sdio vsg buffer size limit: %d\n", virt_pkt_len);
+		ret = -E2BIG;
+		goto err;
+	}
+
+	ret = ath10k_sdio_readsb(ar, ar_sdio->mbox_info.htc_addr,
+				 ar_sdio->vsg_buffer, virt_pkt_len);
+	if (ret) {
+		ath10k_warn(ar, "failed to read bundle packets: %d", ret);
+		goto err;
+	}
+
+	pkt_offset = 0;
 	for (i = 0; i < ar_sdio->n_rx_pkts; i++) {
-		ret = ath10k_sdio_mbox_rx_packet(ar,
-						 &ar_sdio->rx_pkts[i]);
-		if (ret)
-			goto err;
+		pkt = &ar_sdio->rx_pkts[i];
+		htc_hdr = (struct ath10k_htc_hdr *)(ar_sdio->vsg_buffer + pkt_offset);
+		pkt->act_len = le16_to_cpu(htc_hdr->len) + sizeof(*htc_hdr);
+
+		skb_put_data(pkt->skb, htc_hdr, pkt->act_len);
+		pkt->status = 0;
+		pkt_offset += pkt->alloc_len;
 	}
 
 	return 0;
 
 err:
 	/* Free all packets that was not successfully fetched. */
-	for (; i < ar_sdio->n_rx_pkts; i++)
+	for (i = 0; i < ar_sdio->n_rx_pkts; i++)
 		ath10k_sdio_mbox_free_rx_pkt(&ar_sdio->rx_pkts[i]);
 
+	ar_sdio->n_rx_pkts = 0;
+
 	return ret;
 }
 
@@ -717,7 +738,10 @@ static int ath10k_sdio_mbox_rxmsg_pending_handler(struct ath10k *ar,
 			 */
 			*done = false;
 
-		ret = ath10k_sdio_mbox_rx_fetch(ar);
+		if (ar_sdio->n_rx_pkts > 1)
+			ret = ath10k_sdio_mbox_rx_fetch_bundle(ar);
+		else
+			ret = ath10k_sdio_mbox_rx_fetch(ar);
 
 		/* Process fetched packets. This will potentially update
 		 * n_lookaheads depending on if the packets contain lookahead
@@ -2020,6 +2044,12 @@ static int ath10k_sdio_probe(struct sdio_func *func,
 		goto err_core_destroy;
 	}
 
+	ar_sdio->vsg_buffer = devm_kmalloc(ar->dev, ATH10K_SDIO_VSG_BUF_SIZE, GFP_KERNEL);
+	if (!ar_sdio->vsg_buffer) {
+		ret = -ENOMEM;
+		goto err_core_destroy;
+	}
+
 	ar_sdio->irq_data.irq_en_reg =
 		devm_kzalloc(ar->dev, sizeof(struct ath10k_sdio_irq_enable_regs),
 			     GFP_KERNEL);
diff --git a/drivers/net/wireless/ath/ath10k/sdio.h b/drivers/net/wireless/ath/ath10k/sdio.h
index b8c7ac0330bd..8d5b09f8b4f3 100644
--- a/drivers/net/wireless/ath/ath10k/sdio.h
+++ b/drivers/net/wireless/ath/ath10k/sdio.h
@@ -138,8 +138,8 @@ struct ath10k_sdio_irq_proc_regs {
 	u8 rx_lookahead_valid;
 	u8 host_int_status2;
 	u8 gmbox_rx_avail;
-	__le32 rx_lookahead[2];
-	__le32 rx_gmbox_lookahead_alias[2];
+	__le32 rx_lookahead[2 * ATH10K_HIF_MBOX_NUM_MAX];
+	__le32 int_status_enable;
 };
 
 struct ath10k_sdio_irq_enable_regs {
@@ -196,6 +196,13 @@ struct ath10k_sdio {
 	struct ath10k *ar;
 	struct ath10k_sdio_irq_data irq_data;
 
+	/* temporary buffer for sdio read.
+	 * It is allocated when probe, and used for receive bundled packets,
+	 * the read for bundled packets is not parallel, so it does not need
+	 * protected.
+	 */
+	u8 *vsg_buffer;
+
 	/* temporary buffer for BMI requests */
 	u8 *bmi_buf;
 
-- 
cgit v1.2.3-59-g8ed1b


From 224776520ead69e9e85e33e5eb8c705c3552c4e1 Mon Sep 17 00:00:00 2001
From: Wen Gong <wgong@codeaurora.org>
Date: Fri, 15 Nov 2019 09:21:01 +0200
Subject: ath10k: change max RX bundle size from 8 to 32 for sdio

The max bundle size support by firmware is 32, change it from 8 to 32
will help performance. This results in significant performance
improvement on RX path.

The real max rx bundle is decided in ath10k_htc_wait_target(),
it is the min value of HTC_HOST_MAX_MSG_PER_RX_BUNDLE and the value reported
from firmware. So this change shouldn't cause any regressions with other
hardware supported by ath10k.

Tested with QCA6174 SDIO with firmware WLAN.RMH.4.4.1-00017-QCARMSWPZ-1.

Signed-off-by: Wen Gong <wgong@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/htc.h  | 21 +++++++++++++++++++--
 drivers/net/wireless/ath/ath10k/sdio.c |  4 ++--
 drivers/net/wireless/ath/ath10k/sdio.h |  4 ++--
 3 files changed, 23 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/htc.h b/drivers/net/wireless/ath/ath10k/htc.h
index f55d3caec61f..16102175e07a 100644
--- a/drivers/net/wireless/ath/ath10k/htc.h
+++ b/drivers/net/wireless/ath/ath10k/htc.h
@@ -12,6 +12,7 @@
 #include <linux/bug.h>
 #include <linux/skbuff.h>
 #include <linux/timer.h>
+#include <linux/bitfield.h>
 
 struct ath10k;
 
@@ -39,7 +40,7 @@ struct ath10k;
  * 4-byte aligned.
  */
 
-#define HTC_HOST_MAX_MSG_PER_RX_BUNDLE        8
+#define HTC_HOST_MAX_MSG_PER_RX_BUNDLE        32
 
 enum ath10k_htc_tx_flags {
 	ATH10K_HTC_FLAG_NEED_CREDIT_UPDATE = 0x01,
@@ -49,9 +50,25 @@ enum ath10k_htc_tx_flags {
 enum ath10k_htc_rx_flags {
 	ATH10K_HTC_FLAGS_RECV_1MORE_BLOCK = 0x01,
 	ATH10K_HTC_FLAG_TRAILER_PRESENT = 0x02,
-	ATH10K_HTC_FLAG_BUNDLE_MASK     = 0xF0
 };
 
+#define ATH10K_HTC_FLAG_BUNDLE_MASK GENMASK(7, 4)
+
+/* bits 2-3 are for extra bundle count bits 4-5 */
+#define ATH10K_HTC_BUNDLE_EXTRA_MASK GENMASK(3, 2)
+#define ATH10K_HTC_BUNDLE_EXTRA_SHIFT 4
+
+static inline unsigned int ath10k_htc_get_bundle_count(u8 flags)
+{
+	unsigned int count, extra_count;
+
+	count = FIELD_GET(ATH10K_HTC_FLAG_BUNDLE_MASK, flags);
+	extra_count = FIELD_GET(ATH10K_HTC_BUNDLE_EXTRA_MASK, flags) <<
+		ATH10K_HTC_BUNDLE_EXTRA_SHIFT;
+
+	return count + extra_count;
+}
+
 struct ath10k_htc_hdr {
 	u8 eid; /* @enum ath10k_htc_ep_id */
 	u8 flags; /* @enum ath10k_htc_tx_flags, ath10k_htc_rx_flags */
diff --git a/drivers/net/wireless/ath/ath10k/sdio.c b/drivers/net/wireless/ath/ath10k/sdio.c
index 7f25079dde2d..23a6176fbde6 100644
--- a/drivers/net/wireless/ath/ath10k/sdio.c
+++ b/drivers/net/wireless/ath/ath10k/sdio.c
@@ -24,7 +24,7 @@
 #include "trace.h"
 #include "sdio.h"
 
-#define ATH10K_SDIO_VSG_BUF_SIZE	(32 * 1024)
+#define ATH10K_SDIO_VSG_BUF_SIZE	(64 * 1024)
 
 /* inlined helper functions */
 
@@ -494,7 +494,7 @@ static int ath10k_sdio_mbox_alloc_bundle(struct ath10k *ar,
 {
 	int ret, i;
 
-	*bndl_cnt = FIELD_GET(ATH10K_HTC_FLAG_BUNDLE_MASK, htc_hdr->flags);
+	*bndl_cnt = ath10k_htc_get_bundle_count(htc_hdr->flags);
 
 	if (*bndl_cnt > HTC_HOST_MAX_MSG_PER_RX_BUNDLE) {
 		ath10k_warn(ar,
diff --git a/drivers/net/wireless/ath/ath10k/sdio.h b/drivers/net/wireless/ath/ath10k/sdio.h
index 8d5b09f8b4f3..00bd4caa7fa1 100644
--- a/drivers/net/wireless/ath/ath10k/sdio.h
+++ b/drivers/net/wireless/ath/ath10k/sdio.h
@@ -89,10 +89,10 @@
  * to the maximum value (HTC_HOST_MAX_MSG_PER_RX_BUNDLE).
  *
  * in this case the driver must allocate
- * (HTC_HOST_MAX_MSG_PER_RX_BUNDLE * HTC_HOST_MAX_MSG_PER_RX_BUNDLE) skb's.
+ * (HTC_HOST_MAX_MSG_PER_RX_BUNDLE * 2) skb's.
  */
 #define ATH10K_SDIO_MAX_RX_MSGS \
-	(HTC_HOST_MAX_MSG_PER_RX_BUNDLE * HTC_HOST_MAX_MSG_PER_RX_BUNDLE)
+	(HTC_HOST_MAX_MSG_PER_RX_BUNDLE * 2)
 
 #define ATH10K_FIFO_TIMEOUT_AND_CHIP_CONTROL   0x00000868u
 #define ATH10K_FIFO_TIMEOUT_AND_CHIP_CONTROL_DISABLE_SLEEP_OFF 0xFFFEFFFF
-- 
cgit v1.2.3-59-g8ed1b


From 67654b26c903de9949202f23f4864a74a09e22d5 Mon Sep 17 00:00:00 2001
From: Wen Gong <wgong@codeaurora.org>
Date: Fri, 15 Nov 2019 09:21:03 +0200
Subject: ath10k: add workqueue for RX path of sdio

For RX, it has two parts, one is to read data from sdio, another
is to indicate the packets to upper stack. Recently it has only
one thread to do all RX things, it results that it is sequential
for RX and low throughout, change RX to parallel for the two parts
will increase throughout.

This patch move the indication to a workqueue, it results in
significant performance improvement on RX path.

Udp rx throughout is 200Mbps without this patch, and it arrives
400Mbps with this patch.

Tested with QCA6174 SDIO with firmware
WLAN.RMH.4.4.1-00017-QCARMSWPZ-1

Signed-off-by: Wen Gong <wgong@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/core.h   |  1 +
 drivers/net/wireless/ath/ath10k/htt_rx.c |  3 +++
 drivers/net/wireless/ath/ath10k/sdio.c   | 38 +++++++++++++++++++++++++++++---
 drivers/net/wireless/ath/ath10k/sdio.h   |  5 +++++
 4 files changed, 44 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h
index af68eb5d0776..c5407f5080b2 100644
--- a/drivers/net/wireless/ath/ath10k/core.h
+++ b/drivers/net/wireless/ath/ath10k/core.h
@@ -124,6 +124,7 @@ struct ath10k_skb_cb {
 struct ath10k_skb_rxcb {
 	dma_addr_t paddr;
 	struct hlist_node hlist;
+	u8 eid;
 };
 
 static inline struct ath10k_skb_cb *ATH10K_SKB_CB(struct sk_buff *skb)
diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c
index d95b63f133ab..edda6dd93476 100644
--- a/drivers/net/wireless/ath/ath10k/htt_rx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_rx.c
@@ -2235,7 +2235,10 @@ static bool ath10k_htt_rx_proc_rx_ind_hl(struct ath10k_htt *htt,
 
 	hdr = (struct ieee80211_hdr *)skb->data;
 	qos = ieee80211_is_data_qos(hdr->frame_control);
+
 	rx_status = IEEE80211_SKB_RXCB(skb);
+	memset(rx_status, 0, sizeof(*rx_status));
+
 	rx_status->chains |= BIT(0);
 	if (rx->ppdu.combined_rssi == 0) {
 		/* SDIO firmware does not provide signal */
diff --git a/drivers/net/wireless/ath/ath10k/sdio.c b/drivers/net/wireless/ath/ath10k/sdio.c
index 23a6176fbde6..c7d09b07a382 100644
--- a/drivers/net/wireless/ath/ath10k/sdio.c
+++ b/drivers/net/wireless/ath/ath10k/sdio.c
@@ -419,6 +419,7 @@ static int ath10k_sdio_mbox_rx_process_packets(struct ath10k *ar,
 	struct ath10k_htc *htc = &ar->htc;
 	struct ath10k_sdio_rx_data *pkt;
 	struct ath10k_htc_ep *ep;
+	struct ath10k_skb_rxcb *cb;
 	enum ath10k_htc_ep_id id;
 	int ret, i, *n_lookahead_local;
 	u32 *lookaheads_local;
@@ -464,10 +465,16 @@ static int ath10k_sdio_mbox_rx_process_packets(struct ath10k *ar,
 		if (ret)
 			goto out;
 
-		if (!pkt->trailer_only)
-			ep->ep_ops.ep_rx_complete(ar_sdio->ar, pkt->skb);
-		else
+		if (!pkt->trailer_only) {
+			cb = ATH10K_SKB_RXCB(pkt->skb);
+			cb->eid = id;
+
+			skb_queue_tail(&ar_sdio->rx_head, pkt->skb);
+			queue_work(ar->workqueue_aux,
+				   &ar_sdio->async_work_rx);
+		} else {
 			kfree_skb(pkt->skb);
+		}
 
 		/* The RX complete handler now owns the skb...*/
 		pkt->skb = NULL;
@@ -1317,6 +1324,28 @@ static void __ath10k_sdio_write_async(struct ath10k *ar,
 	ath10k_sdio_free_bus_req(ar, req);
 }
 
+/* To improve throughput use workqueue to deliver packets to HTC layer,
+ * this way SDIO bus is utilised much better.
+ */
+static void ath10k_rx_indication_async_work(struct work_struct *work)
+{
+	struct ath10k_sdio *ar_sdio = container_of(work, struct ath10k_sdio,
+						   async_work_rx);
+	struct ath10k *ar = ar_sdio->ar;
+	struct ath10k_htc_ep *ep;
+	struct ath10k_skb_rxcb *cb;
+	struct sk_buff *skb;
+
+	while (true) {
+		skb = skb_dequeue(&ar_sdio->rx_head);
+		if (!skb)
+			break;
+		cb = ATH10K_SKB_RXCB(skb);
+		ep = &ar->htc.endpoint[cb->eid];
+		ep->ep_ops.ep_rx_complete(ar, skb);
+	}
+}
+
 static void ath10k_sdio_write_async_work(struct work_struct *work)
 {
 	struct ath10k_sdio *ar_sdio = container_of(work, struct ath10k_sdio,
@@ -2087,6 +2116,9 @@ static int ath10k_sdio_probe(struct sdio_func *func,
 	for (i = 0; i < ATH10K_SDIO_BUS_REQUEST_MAX_NUM; i++)
 		ath10k_sdio_free_bus_req(ar, &ar_sdio->bus_req[i]);
 
+	skb_queue_head_init(&ar_sdio->rx_head);
+	INIT_WORK(&ar_sdio->async_work_rx, ath10k_rx_indication_async_work);
+
 	dev_id_base = FIELD_GET(QCA_MANUFACTURER_ID_BASE, id->device);
 	switch (dev_id_base) {
 	case QCA_MANUFACTURER_ID_AR6005_BASE:
diff --git a/drivers/net/wireless/ath/ath10k/sdio.h b/drivers/net/wireless/ath/ath10k/sdio.h
index 00bd4caa7fa1..9acaf0229cf2 100644
--- a/drivers/net/wireless/ath/ath10k/sdio.h
+++ b/drivers/net/wireless/ath/ath10k/sdio.h
@@ -187,6 +187,9 @@ struct ath10k_sdio {
 	struct ath10k_sdio_bus_request bus_req[ATH10K_SDIO_BUS_REQUEST_MAX_NUM];
 	/* free list of bus requests */
 	struct list_head bus_req_freeq;
+
+	struct sk_buff_head rx_head;
+
 	/* protects access to bus_req_freeq */
 	spinlock_t lock;
 
@@ -213,6 +216,8 @@ struct ath10k_sdio {
 	struct list_head wr_asyncq;
 	/* protects access to wr_asyncq */
 	spinlock_t wr_async_lock;
+
+	struct work_struct async_work_rx;
 };
 
 static inline struct ath10k_sdio *ath10k_sdio_priv(struct ath10k *ar)
-- 
cgit v1.2.3-59-g8ed1b


From efd2f4c5016bd24f33357487adcf873359116e9a Mon Sep 17 00:00:00 2001
From: Kalle Valo <kvalo@codeaurora.org>
Date: Fri, 15 Nov 2019 09:21:09 +0200
Subject: ath10k: sdio: cosmetic cleanup

Do some cosmetic cleanup while reviewing the files. No functional changes.

Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/sdio.c | 21 +++++++++------------
 1 file changed, 9 insertions(+), 12 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/sdio.c b/drivers/net/wireless/ath/ath10k/sdio.c
index c7d09b07a382..347e33909584 100644
--- a/drivers/net/wireless/ath/ath10k/sdio.c
+++ b/drivers/net/wireless/ath/ath10k/sdio.c
@@ -541,10 +541,8 @@ static int ath10k_sdio_mbox_rx_alloc(struct ath10k *ar,
 	int pkt_cnt = 0;
 
 	if (n_lookaheads > ATH10K_SDIO_MAX_RX_MSGS) {
-		ath10k_warn(ar,
-			    "the total number of pkgs to be fetched (%u) exceeds maximum %u\n",
-			    n_lookaheads,
-			    ATH10K_SDIO_MAX_RX_MSGS);
+		ath10k_warn(ar, "the total number of pkgs to be fetched (%u) exceeds maximum %u\n",
+			    n_lookaheads, ATH10K_SDIO_MAX_RX_MSGS);
 		ret = -ENOMEM;
 		goto err;
 	}
@@ -553,10 +551,8 @@ static int ath10k_sdio_mbox_rx_alloc(struct ath10k *ar,
 		htc_hdr = (struct ath10k_htc_hdr *)&lookaheads[i];
 		last_in_bundle = false;
 
-		if (le16_to_cpu(htc_hdr->len) >
-		    ATH10K_HTC_MBOX_MAX_PAYLOAD_LENGTH) {
-			ath10k_warn(ar,
-				    "payload length %d exceeds max htc length: %zu\n",
+		if (le16_to_cpu(htc_hdr->len) > ATH10K_HTC_MBOX_MAX_PAYLOAD_LENGTH) {
+			ath10k_warn(ar, "payload length %d exceeds max htc length: %zu\n",
 				    le16_to_cpu(htc_hdr->len),
 				    ATH10K_HTC_MBOX_MAX_PAYLOAD_LENGTH);
 			ret = -ENOMEM;
@@ -567,8 +563,7 @@ static int ath10k_sdio_mbox_rx_alloc(struct ath10k *ar,
 		full_len = ath10k_sdio_calc_txrx_padded_len(ar_sdio, act_len);
 
 		if (full_len > ATH10K_SDIO_MAX_BUFFER_SIZE) {
-			ath10k_warn(ar,
-				    "rx buffer requested with invalid htc_hdr length (%d, 0x%x): %d\n",
+			ath10k_warn(ar, "rx buffer requested with invalid htc_hdr length (%d, 0x%x): %d\n",
 				    htc_hdr->eid, htc_hdr->flags,
 				    le16_to_cpu(htc_hdr->len));
 			ret = -EINVAL;
@@ -590,12 +585,14 @@ static int ath10k_sdio_mbox_rx_alloc(struct ath10k *ar,
 							    &bndl_cnt);
 
 			if (ret) {
-				ath10k_warn(ar, "alloc_bundle error %d\n", ret);
+				ath10k_warn(ar, "failed to allocate a bundle: %d\n",
+					    ret);
 				goto err;
 			}
 
 			pkt_cnt += bndl_cnt;
-			/*Next buffer will be the last in the bundle */
+
+			/* next buffer will be the last in the bundle */
 			last_in_bundle = true;
 		}
 
-- 
cgit v1.2.3-59-g8ed1b


From 2246c215659060b214c9cc7fcd4abec95874ae5a Mon Sep 17 00:00:00 2001
From: Kalle Valo <kvalo@codeaurora.org>
Date: Fri, 15 Nov 2019 09:21:11 +0200
Subject: ath10k: sdio: remove struct ath10k_sdio_rx_data::status

It seems to be unused.

Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/sdio.c | 2 --
 drivers/net/wireless/ath/ath10k/sdio.h | 1 -
 2 files changed, 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/sdio.c b/drivers/net/wireless/ath/ath10k/sdio.c
index 347e33909584..2bf96e9665cd 100644
--- a/drivers/net/wireless/ath/ath10k/sdio.c
+++ b/drivers/net/wireless/ath/ath10k/sdio.c
@@ -649,7 +649,6 @@ static int ath10k_sdio_mbox_rx_fetch(struct ath10k *ar)
 
 	htc_hdr = (struct ath10k_htc_hdr *)skb->data;
 	pkt->act_len = le16_to_cpu(htc_hdr->len) + sizeof(*htc_hdr);
-	pkt->status = ret;
 	skb_put(skb, pkt->act_len);
 
 	return ret;
@@ -687,7 +686,6 @@ static int ath10k_sdio_mbox_rx_fetch_bundle(struct ath10k *ar)
 		pkt->act_len = le16_to_cpu(htc_hdr->len) + sizeof(*htc_hdr);
 
 		skb_put_data(pkt->skb, htc_hdr, pkt->act_len);
-		pkt->status = 0;
 		pkt_offset += pkt->alloc_len;
 	}
 
diff --git a/drivers/net/wireless/ath/ath10k/sdio.h b/drivers/net/wireless/ath/ath10k/sdio.h
index 9acaf0229cf2..33195f49acab 100644
--- a/drivers/net/wireless/ath/ath10k/sdio.h
+++ b/drivers/net/wireless/ath/ath10k/sdio.h
@@ -126,7 +126,6 @@ struct ath10k_sdio_rx_data {
 	bool part_of_bundle;
 	bool last_in_bundle;
 	bool trailer_only;
-	int status;
 };
 
 struct ath10k_sdio_irq_proc_regs {
-- 
cgit v1.2.3-59-g8ed1b


From e01cc82c4d1ec3bddcbb7cd991cf5dc0131ed9a1 Mon Sep 17 00:00:00 2001
From: Wen Gong <wgong@codeaurora.org>
Date: Fri, 15 Nov 2019 09:21:32 +0200
Subject: ath10k: correct the tlv len of ath10k_wmi_tlv_op_gen_config_pno_start

the tlv len is set to the total len of the wmi cmd, it will trigger
firmware crash, correct the tlv len.

Tested with QCA6174 SDIO with firmware
WLAN.RMH.4.4.1-00017-QCARMSWP-1 and QCA6174
PCIE with firmware WLAN.RM.4.4.1-00110-QCARMSWPZ-1.

Fixes: ce834e280f2f875 ("ath10k: support NET_DETECT WoWLAN feature")
Signed-off-by: Wen Gong <wgong@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/wmi-tlv.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath10k/wmi-tlv.c b/drivers/net/wireless/ath/ath10k/wmi-tlv.c
index 69a1ec53df29..7b358484940e 100644
--- a/drivers/net/wireless/ath/ath10k/wmi-tlv.c
+++ b/drivers/net/wireless/ath/ath10k/wmi-tlv.c
@@ -3707,6 +3707,7 @@ ath10k_wmi_tlv_op_gen_config_pno_start(struct ath10k *ar,
 	struct wmi_tlv *tlv;
 	struct sk_buff *skb;
 	__le32 *channel_list;
+	u16 tlv_len;
 	size_t len;
 	void *ptr;
 	u32 i;
@@ -3764,10 +3765,12 @@ ath10k_wmi_tlv_op_gen_config_pno_start(struct ath10k *ar,
 	/* nlo_configured_parameters(nlo_list) */
 	cmd->no_of_ssids = __cpu_to_le32(min_t(u8, pno->uc_networks_count,
 					       WMI_NLO_MAX_SSIDS));
+	tlv_len = __le32_to_cpu(cmd->no_of_ssids) *
+		sizeof(struct nlo_configured_parameters);
 
 	tlv = ptr;
 	tlv->tag = __cpu_to_le16(WMI_TLV_TAG_ARRAY_STRUCT);
-	tlv->len = __cpu_to_le16(len);
+	tlv->len = __cpu_to_le16(tlv_len);
 
 	ptr += sizeof(*tlv);
 	nlo_list = ptr;
-- 
cgit v1.2.3-59-g8ed1b


From d58f466a5dee814b7820aff9aaa393b3782559d9 Mon Sep 17 00:00:00 2001
From: Wen Gong <wgong@codeaurora.org>
Date: Fri, 15 Nov 2019 09:21:35 +0200
Subject: ath10k: add large size for BMI download data for SDIO

Download firmware time cost of SDIO is too long, it is about 480ms,
add large size 2048 bytes for BMI download for SDIO chip, its time
cost will reduced to 240ms.

This will optimize the download firmware time cost.

Tested with QCA6174 SDIO with firmware
WLAN.RMH.4.4.1-00017-QCARMSWP-1.

Signed-off-by: Wen Gong <wgong@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/bmi.c  | 52 +++++++++++++++++++++++++++++++++-
 drivers/net/wireless/ath/ath10k/bmi.h  | 10 +++++++
 drivers/net/wireless/ath/ath10k/core.c |  1 +
 drivers/net/wireless/ath/ath10k/hw.h   |  3 ++
 drivers/net/wireless/ath/ath10k/sdio.c |  2 +-
 5 files changed, 66 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/bmi.c b/drivers/net/wireless/ath/ath10k/bmi.c
index 95dc4be82e5c..ea908107581d 100644
--- a/drivers/net/wireless/ath/ath10k/bmi.c
+++ b/drivers/net/wireless/ath/ath10k/bmi.c
@@ -346,6 +346,52 @@ int ath10k_bmi_execute(struct ath10k *ar, u32 address, u32 param, u32 *result)
 	return 0;
 }
 
+static int ath10k_bmi_lz_data_large(struct ath10k *ar, const void *buffer, u32 length)
+{
+	struct bmi_cmd *cmd;
+	u32 hdrlen = sizeof(cmd->id) + sizeof(cmd->lz_data);
+	u32 txlen;
+	int ret;
+	size_t buf_len;
+
+	ath10k_dbg(ar, ATH10K_DBG_BMI, "large bmi lz data buffer 0x%pK length %d\n",
+		   buffer, length);
+
+	if (ar->bmi.done_sent) {
+		ath10k_warn(ar, "command disallowed\n");
+		return -EBUSY;
+	}
+
+	buf_len = sizeof(*cmd) + BMI_MAX_LARGE_DATA_SIZE - BMI_MAX_DATA_SIZE;
+	cmd = kzalloc(buf_len, GFP_KERNEL);
+	if (!cmd)
+		return -ENOMEM;
+
+	while (length) {
+		txlen = min(length, BMI_MAX_LARGE_DATA_SIZE - hdrlen);
+
+		WARN_ON_ONCE(txlen & 3);
+
+		cmd->id          = __cpu_to_le32(BMI_LZ_DATA);
+		cmd->lz_data.len = __cpu_to_le32(txlen);
+		memcpy(cmd->lz_data.payload, buffer, txlen);
+
+		ret = ath10k_hif_exchange_bmi_msg(ar, cmd, hdrlen + txlen,
+						  NULL, NULL);
+		if (ret) {
+			ath10k_warn(ar, "unable to write to the device\n");
+			return ret;
+		}
+
+		buffer += txlen;
+		length -= txlen;
+	}
+
+	kfree(cmd);
+
+	return 0;
+}
+
 int ath10k_bmi_lz_data(struct ath10k *ar, const void *buffer, u32 length)
 {
 	struct bmi_cmd cmd;
@@ -430,7 +476,11 @@ int ath10k_bmi_fast_download(struct ath10k *ar,
 	if (trailer_len > 0)
 		memcpy(trailer, buffer + head_len, trailer_len);
 
-	ret = ath10k_bmi_lz_data(ar, buffer, head_len);
+	if (ar->hw_params.bmi_large_size_download)
+		ret = ath10k_bmi_lz_data_large(ar, buffer, head_len);
+	else
+		ret = ath10k_bmi_lz_data(ar, buffer, head_len);
+
 	if (ret)
 		return ret;
 
diff --git a/drivers/net/wireless/ath/ath10k/bmi.h b/drivers/net/wireless/ath/ath10k/bmi.h
index ef3bdba43bed..f6fadcbdd86e 100644
--- a/drivers/net/wireless/ath/ath10k/bmi.h
+++ b/drivers/net/wireless/ath/ath10k/bmi.h
@@ -45,6 +45,15 @@
 			sizeof(u32) + \
 			sizeof(u32))
 
+/* Maximum data size used for large BMI transfers */
+#define BMI_MAX_LARGE_DATA_SIZE	2048
+
+/* len = cmd + addr + length */
+#define BMI_MAX_LARGE_CMDBUF_SIZE (BMI_MAX_LARGE_DATA_SIZE + \
+			sizeof(u32) + \
+			sizeof(u32) + \
+			sizeof(u32))
+
 /* BMI Commands */
 
 enum bmi_cmd_id {
@@ -258,6 +267,7 @@ int ath10k_bmi_write_memory(struct ath10k *ar, u32 address,
 int ath10k_bmi_execute(struct ath10k *ar, u32 address, u32 param, u32 *result);
 int ath10k_bmi_lz_stream_start(struct ath10k *ar, u32 address);
 int ath10k_bmi_lz_data(struct ath10k *ar, const void *buffer, u32 length);
+
 int ath10k_bmi_fast_download(struct ath10k *ar, u32 address,
 			     const void *buffer, u32 length);
 int ath10k_bmi_read_soc_reg(struct ath10k *ar, u32 address, u32 *reg_val);
diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c
index 4f76ba5d78a9..0d345852c04a 100644
--- a/drivers/net/wireless/ath/ath10k/core.c
+++ b/drivers/net/wireless/ath/ath10k/core.c
@@ -189,6 +189,7 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = {
 		.num_wds_entries = 0x20,
 		.uart_pin_workaround = true,
 		.tx_stats_over_pktlog = false,
+		.bmi_large_size_download = true,
 	},
 	{
 		.id = QCA6174_HW_2_1_VERSION,
diff --git a/drivers/net/wireless/ath/ath10k/hw.h b/drivers/net/wireless/ath/ath10k/hw.h
index 35a362329a4f..21b7a2a873b0 100644
--- a/drivers/net/wireless/ath/ath10k/hw.h
+++ b/drivers/net/wireless/ath/ath10k/hw.h
@@ -613,6 +613,9 @@ struct ath10k_hw_params {
 	/* target supporting fw download via diag ce */
 	bool fw_diag_ce_download;
 
+	/* target supporting fw download via large size BMI */
+	bool bmi_large_size_download;
+
 	/* need to set uart pin if disable uart print, workaround for a
 	 * firmware bug
 	 */
diff --git a/drivers/net/wireless/ath/ath10k/sdio.c b/drivers/net/wireless/ath/ath10k/sdio.c
index 2bf96e9665cd..3e63adfe3e9c 100644
--- a/drivers/net/wireless/ath/ath10k/sdio.c
+++ b/drivers/net/wireless/ath/ath10k/sdio.c
@@ -2082,7 +2082,7 @@ static int ath10k_sdio_probe(struct sdio_func *func,
 		goto err_core_destroy;
 	}
 
-	ar_sdio->bmi_buf = devm_kzalloc(ar->dev, BMI_MAX_CMDBUF_SIZE, GFP_KERNEL);
+	ar_sdio->bmi_buf = devm_kzalloc(ar->dev, BMI_MAX_LARGE_CMDBUF_SIZE, GFP_KERNEL);
 	if (!ar_sdio->bmi_buf) {
 		ret = -ENOMEM;
 		goto err_core_destroy;
-- 
cgit v1.2.3-59-g8ed1b


From 501d4152b0184b7f0a4df03a1dc748b381410a11 Mon Sep 17 00:00:00 2001
From: Jeffrey Hugo <jeffrey.l.hugo@gmail.com>
Date: Fri, 15 Nov 2019 09:21:43 +0200
Subject: ath10k: Handle when FW doesn't support QMI_WLFW_HOST_CAP_REQ_V01

Firmware with the build id QC_IMAGE_VERSION_STRING=WLAN.HL.1.0.2-XXXX does
not support the QMI_WLFW_HOST_CAP_REQ_V01 message and will return the
QMI not supported error to the ath10k driver.  Since not supporting this
message is not fatal to the firmware nor the ath10k driver, lets catch
this particular scenario and ignore it so that we can still bring up
wifi services successfully.

Signed-off-by: Jeffrey Hugo <jeffrey.l.hugo@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/qmi.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath10k/qmi.c b/drivers/net/wireless/ath/ath10k/qmi.c
index a0ba07b85362..ed69e9775143 100644
--- a/drivers/net/wireless/ath/ath10k/qmi.c
+++ b/drivers/net/wireless/ath/ath10k/qmi.c
@@ -635,7 +635,9 @@ static int ath10k_qmi_host_cap_send_sync(struct ath10k_qmi *qmi)
 	if (ret < 0)
 		goto out;
 
-	if (resp.resp.result != QMI_RESULT_SUCCESS_V01) {
+	/* older FW didn't support this request, which is not fatal */
+	if (resp.resp.result != QMI_RESULT_SUCCESS_V01 &&
+	    resp.resp.error != QMI_ERR_NOT_SUPPORTED_V01) {
 		ath10k_err(ar, "host capability request rejected: %d\n", resp.resp.error);
 		ret = -EINVAL;
 		goto out;
-- 
cgit v1.2.3-59-g8ed1b


From 23b5156a856b6342c8a1ef7045d387c00cdb17c5 Mon Sep 17 00:00:00 2001
From: Wen Gong <wgong@codeaurora.org>
Date: Fri, 8 Nov 2019 17:27:06 +0800
Subject: ath10k: add NL80211_FEATURE_ND_RANDOM_MAC_ADDR for NLO

Add NL80211_FEATURE_ND_RANDOM_MAC_ADDR for NLO will enable the random
mac address for netdetect case.
iw command:
iw phy0 wowlan enable net-detect net-detect
randomize=AA:7B:A1:AC:B2:41/FF:FF:FF:FF:FF:FF interval 5000 delay 30
freqs 2412 matches ssid foo.
After suspend, DUT will send probe request with mac AA:7B:A1:AC:B2:41.

WCN3990, QCA9377, QCA6174 PCI also support this feature.

Tested with QCA6174 SDIO with firmware
WLAN.RMH.4.4.1-00029.

Signed-off-by: Wen Gong <wgong@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/mac.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index 83cc8778ca1e..6c6eb9341e6d 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -8908,6 +8908,7 @@ int ath10k_mac_register(struct ath10k *ar)
 			WMI_PNO_MAX_SCHED_SCAN_PLAN_INT;
 		ar->hw->wiphy->max_sched_scan_plan_iterations =
 			WMI_PNO_MAX_SCHED_SCAN_PLAN_ITRNS;
+		ar->hw->wiphy->features |= NL80211_FEATURE_ND_RANDOM_MAC_ADDR;
 	}
 
 	ar->hw->vif_data_size = sizeof(struct ath10k_vif);
-- 
cgit v1.2.3-59-g8ed1b


From f8a595a87e93a33a10879f4b856be818d2f53c84 Mon Sep 17 00:00:00 2001
From: Jeffrey Hugo <jeffrey.l.hugo@gmail.com>
Date: Wed, 13 Nov 2019 07:40:16 -0800
Subject: ath10k: Fix qmi init error handling

When ath10k_qmi_init() fails, the error handling does not free the irq
resources, which causes an issue if we EPROBE_DEFER as we'll attempt to
(re-)register irqs which are already registered.

Fix this by doing a power off since we just powered on the hardware, and
freeing the irqs as error handling.

Fixes: ba94c753ccb4 ("ath10k: add QMI message handshake for wcn3990 client")
Signed-off-by: Jeffrey Hugo <jeffrey.l.hugo@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/snoc.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath10k/snoc.c b/drivers/net/wireless/ath/ath10k/snoc.c
index 16177497bba7..7e85c4916e7f 100644
--- a/drivers/net/wireless/ath/ath10k/snoc.c
+++ b/drivers/net/wireless/ath/ath10k/snoc.c
@@ -1563,13 +1563,16 @@ static int ath10k_snoc_probe(struct platform_device *pdev)
 	ret = ath10k_qmi_init(ar, msa_size);
 	if (ret) {
 		ath10k_warn(ar, "failed to register wlfw qmi client: %d\n", ret);
-		goto err_core_destroy;
+		goto err_power_off;
 	}
 
 	ath10k_dbg(ar, ATH10K_DBG_SNOC, "snoc probe\n");
 
 	return 0;
 
+err_power_off:
+	ath10k_hw_power_off(ar);
+
 err_free_irq:
 	ath10k_snoc_free_irq(ar);
 
-- 
cgit v1.2.3-59-g8ed1b


From 319c2b71041f324e06dbee03c1011106cf746795 Mon Sep 17 00:00:00 2001
From: Jeffrey Hugo <jeffrey.l.hugo@gmail.com>
Date: Wed, 13 Nov 2019 07:46:46 -0800
Subject: ath10k: Handle "invalid" BDFs for msm8998 devices

When the BDF download QMI message has the end field set to 1, it signals
the end of the transfer, and triggers the firmware to do a CRC check.  The
BDFs for msm8998 devices fail this check, yet the firmware is happy to
still use the BDF.  It appears that this error is not caught by the
downstream drive by concidence, therefore there are production devices
in the field where this issue needs to be handled otherwise we cannot
support wifi on them.  So, attempt to detect this scenario as best we can
and treat it as non-fatal.

Signed-off-by: Jeffrey Hugo <jeffrey.l.hugo@gmail.com>
Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/qmi.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath10k/qmi.c b/drivers/net/wireless/ath/ath10k/qmi.c
index ed69e9775143..7b524b614f22 100644
--- a/drivers/net/wireless/ath/ath10k/qmi.c
+++ b/drivers/net/wireless/ath/ath10k/qmi.c
@@ -279,7 +279,15 @@ static int ath10k_qmi_bdf_dnld_send_sync(struct ath10k_qmi *qmi)
 		if (ret < 0)
 			goto out;
 
-		if (resp.resp.result != QMI_RESULT_SUCCESS_V01) {
+		/* end = 1 triggers a CRC check on the BDF.  If this fails, we
+		 * get a QMI_ERR_MALFORMED_MSG_V01 error, but the FW is still
+		 * willing to use the BDF.  For some platforms, all the valid
+		 * released BDFs fail this CRC check, so attempt to detect this
+		 * scenario and treat it as non-fatal.
+		 */
+		if (resp.resp.result != QMI_RESULT_SUCCESS_V01 &&
+		    !(req->end == 1 &&
+		      resp.resp.result == QMI_ERR_MALFORMED_MSG_V01)) {
 			ath10k_err(ar, "failed to download board data file: %d\n",
 				   resp.resp.error);
 			ret = -EINVAL;
-- 
cgit v1.2.3-59-g8ed1b


From 7005eafc1f1cb0a525631c2214f760d6fb636ebb Mon Sep 17 00:00:00 2001
From: Wen Gong <wgong@codeaurora.org>
Date: Thu, 14 Nov 2019 19:19:31 +0800
Subject: ath10k: report rssi of each chain to mac80211 for sdio

iw command only show rssi without each chain's rssi on sdio
iw wlan0 station dump
Station a0:40:a0:93:3e:de (on wlan0)
signal:         -82 dBm
signal avg:     -82 dBm

after this patch, it will show each chain's rssi on sdio
Station a0:40:a0:93:3e:de (on wlan0)
signal:         -82 [-84, -88] dBm
signal avg:     -82 [-84, -87] dBm

For QCA6174 PCIe, the ppdu have the correct rssi of each chain, it
indicate rssi of rx data by ath10k_htt_rx_h_signal. For sdio chip, the
rssi of each chain stored in rx management reported by firmware, the
ath10k_wmi_tlv_op_pull_mgmt_rx_ev which used for tlv wmi will get the
rssi of each chain and stored them in wmi_mgmt_rx_ev_arg, then indicate
them to mac80211. For non-tlv wmi chip, it will not get the rssi of each
chain and not indicate to mac80211, for non-tlv wmi chip, this patch will
not have impact. For tlv wmi chip, if the rssi of chain in mgmt is valid,
it will be indicate to mac80211, tested with QCA6174 PCIe/SDIO, the rssi
of 2 chain in mgmt is valid.

rssi of chains in mgmt of QCA6174 SDIO:
92096.652780: ath10k:ath10k_log_warn: ath10k_sdio mmc1:0001:1 rssi[0]:70
92096.657324: ath10k:ath10k_log_warn: ath10k_sdio mmc1:0001:1 rssi[1]:68
92096.662009: ath10k:ath10k_log_warn: ath10k_sdio mmc1:0001:1 rssi[2]:128
92096.666647: ath10k:ath10k_log_warn: ath10k_sdio mmc1:0001:1 rssi[3]:128

rssi of chains in mgmt of QCA6174 PCIe:
[ 1581.049816] ath10k_pci 0000:02:00.0: mgmt rssi[0]:17
[ 1581.049818] ath10k_pci 0000:02:00.0: mgmt rssi[1]:22
[ 1581.049821] ath10k_pci 0000:02:00.0: mgmt rssi[2]:128
[ 1581.049823] ath10k_pci 0000:02:00.0: mgmt rssi[3]:128

after apply this patch, the iw's rssi of PCIe do not changed, result is
same with before.

iw wlan0 station dump of QCA6174 PCIe:
Station 6c:e8:73:b8:92:dc (on wlan0)
        signal:         -70 [-77, -72] dBm
        signal avg:     -69 [-78, -72] dBm

iw wlan-5000mhz station dump of QCA9984 PCIe
connected with 2 client which has 2 chain:
Station 70:48:0f:1f:1a:b2 (on wlan-5000mhz)
        signal:         -47 [-55, -48, -87, -88] dBm
        signal avg:     -42 [-50, -43, -83, -86] dBm
Station ac:c1:ee:39:e3:83 (on wlan-5000mhz)
        signal:         -43 [-46, -45, -79, -84] dBm
        signal avg:     -43 [-46, -46, -82, -83] dBm

Tested with QCA6174 SDIO with firmware WLAN.RMH.4.4.1-00017-QCARMSWP-1.
Tested with QCA6174 PCIe with firmware WLAN.RM.4.4.1-00110-QCARMSWP-1.
Tested with QCA9984 PCIe with firmware 10.4-3.9.0.2-00040.

Signed-off-by: Wen Gong <wgong@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/htt_rx.c  |  1 -
 drivers/net/wireless/ath/ath10k/wmi-tlv.c |  5 ++++-
 drivers/net/wireless/ath/ath10k/wmi.c     | 18 ++++++++++++++++--
 drivers/net/wireless/ath/ath10k/wmi.h     |  2 ++
 4 files changed, 22 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c
index edda6dd93476..0f1f71e43b9b 100644
--- a/drivers/net/wireless/ath/ath10k/htt_rx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_rx.c
@@ -2239,7 +2239,6 @@ static bool ath10k_htt_rx_proc_rx_ind_hl(struct ath10k_htt *htt,
 	rx_status = IEEE80211_SKB_RXCB(skb);
 	memset(rx_status, 0, sizeof(*rx_status));
 
-	rx_status->chains |= BIT(0);
 	if (rx->ppdu.combined_rssi == 0) {
 		/* SDIO firmware does not provide signal */
 		rx_status->signal = 0;
diff --git a/drivers/net/wireless/ath/ath10k/wmi-tlv.c b/drivers/net/wireless/ath/ath10k/wmi-tlv.c
index 7b358484940e..4e68debda9bf 100644
--- a/drivers/net/wireless/ath/ath10k/wmi-tlv.c
+++ b/drivers/net/wireless/ath/ath10k/wmi-tlv.c
@@ -841,7 +841,7 @@ static int ath10k_wmi_tlv_op_pull_mgmt_rx_ev(struct ath10k *ar,
 	const struct wmi_tlv_mgmt_rx_ev *ev;
 	const u8 *frame;
 	u32 msdu_len;
-	int ret;
+	int ret, i;
 
 	tb = ath10k_wmi_tlv_parse_alloc(ar, skb->data, skb->len, GFP_ATOMIC);
 	if (IS_ERR(tb)) {
@@ -865,6 +865,9 @@ static int ath10k_wmi_tlv_op_pull_mgmt_rx_ev(struct ath10k *ar,
 	arg->phy_mode = ev->phy_mode;
 	arg->rate = ev->rate;
 
+	for (i = 0; i < ARRAY_SIZE(ev->rssi); i++)
+		arg->rssi[i] = ev->rssi[i];
+
 	msdu_len = __le32_to_cpu(arg->buf_len);
 
 	if (skb->len < (frame - skb->data) + msdu_len) {
diff --git a/drivers/net/wireless/ath/ath10k/wmi.c b/drivers/net/wireless/ath/ath10k/wmi.c
index 9f564e2b7a14..13f7531d945d 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.c
+++ b/drivers/net/wireless/ath/ath10k/wmi.c
@@ -2463,10 +2463,10 @@ int ath10k_wmi_event_mgmt_rx(struct ath10k *ar, struct sk_buff *skb)
 	u32 rx_status;
 	u32 channel;
 	u32 phy_mode;
-	u32 snr;
+	u32 snr, rssi;
 	u32 rate;
 	u16 fc;
-	int ret;
+	int ret, i;
 
 	ret = ath10k_wmi_pull_mgmt_rx(ar, skb, &arg);
 	if (ret) {
@@ -2525,6 +2525,20 @@ int ath10k_wmi_event_mgmt_rx(struct ath10k *ar, struct sk_buff *skb)
 
 	status->freq = ieee80211_channel_to_frequency(channel, status->band);
 	status->signal = snr + ATH10K_DEFAULT_NOISE_FLOOR;
+
+	BUILD_BUG_ON(ARRAY_SIZE(status->chain_signal) != ARRAY_SIZE(arg.rssi));
+
+	for (i = 0; i < ARRAY_SIZE(status->chain_signal); i++) {
+		status->chains &= ~BIT(i);
+		rssi = __le32_to_cpu(arg.rssi[i]);
+		ath10k_dbg(ar, ATH10K_DBG_MGMT, "mgmt rssi[%d]:%d\n", i, arg.rssi[i]);
+
+		if (rssi != ATH10K_INVALID_RSSI && rssi != 0) {
+			status->chain_signal[i] = ATH10K_DEFAULT_NOISE_FLOOR + rssi;
+			status->chains |= BIT(i);
+		}
+	}
+
 	status->rate_idx = ath10k_mac_bitrate_to_idx(sband, rate / 100);
 
 	hdr = (struct ieee80211_hdr *)skb->data;
diff --git a/drivers/net/wireless/ath/ath10k/wmi.h b/drivers/net/wireless/ath/ath10k/wmi.h
index 74adce1dd3a9..972d53d77654 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.h
+++ b/drivers/net/wireless/ath/ath10k/wmi.h
@@ -6786,6 +6786,7 @@ struct wmi_peer_delete_resp_ev_arg {
 	struct wmi_mac_addr peer_addr;
 };
 
+#define WMI_MGMT_RX_NUM_RSSI 4
 struct wmi_mgmt_rx_ev_arg {
 	__le32 channel;
 	__le32 snr;
@@ -6794,6 +6795,7 @@ struct wmi_mgmt_rx_ev_arg {
 	__le32 buf_len;
 	__le32 status; /* %WMI_RX_STATUS_ */
 	struct wmi_mgmt_rx_ext_info ext_info;
+	__le32 rssi[WMI_MGMT_RX_NUM_RSSI];
 };
 
 struct wmi_ch_info_ev_arg {
-- 
cgit v1.2.3-59-g8ed1b


From ea0c3e2a470271573d480f48a67e74633566a317 Mon Sep 17 00:00:00 2001
From: Linus Lüssing <ll@simonwunderlich.de>
Date: Fri, 15 Nov 2019 11:56:12 +0100
Subject: ath10k: fix RX of frames with broken FCS in monitor mode
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

So far, frames were forwarded regardless of the FCS correctness leading
to userspace applications listening on the monitor mode interface to
receive potentially broken frames, even with the "fcsfail" flag unset.

By default, with the "fcsfail" flag of a monitor mode interface
unset, frames with FCS errors should be dropped. With this patch, the
fcsfail flag is taken into account correctly.

Tested-on: QCA4019 firmware-5-ct-full-community-12.bin-lede.011

Cc: Simon Wunderlich <sw@simonwunderlich.de>
Signed-off-by: Linus Lüssing <ll@simonwunderlich.de>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/core.h   | 1 +
 drivers/net/wireless/ath/ath10k/debug.c  | 2 ++
 drivers/net/wireless/ath/ath10k/htt_rx.c | 7 +++++++
 3 files changed, 10 insertions(+)

diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h
index c5407f5080b2..5101bf2b5b15 100644
--- a/drivers/net/wireless/ath/ath10k/core.h
+++ b/drivers/net/wireless/ath/ath10k/core.h
@@ -1181,6 +1181,7 @@ struct ath10k {
 
 	struct {
 		/* protected by data_lock */
+		u32 rx_crc_err_drop;
 		u32 fw_crash_counter;
 		u32 fw_warm_reset_counter;
 		u32 fw_cold_reset_counter;
diff --git a/drivers/net/wireless/ath/ath10k/debug.c b/drivers/net/wireless/ath/ath10k/debug.c
index 04c50a26a4f4..e000677ac516 100644
--- a/drivers/net/wireless/ath/ath10k/debug.c
+++ b/drivers/net/wireless/ath/ath10k/debug.c
@@ -1094,6 +1094,7 @@ static const char ath10k_gstrings_stats[][ETH_GSTRING_LEN] = {
 	"d_rts_good",
 	"d_tx_power", /* in .5 dbM I think */
 	"d_rx_crc_err", /* fcs_bad */
+	"d_rx_crc_err_drop", /* frame with FCS error, dropped late in kernel */
 	"d_no_beacon",
 	"d_tx_mpdus_queued",
 	"d_tx_msdu_queued",
@@ -1193,6 +1194,7 @@ void ath10k_debug_get_et_stats(struct ieee80211_hw *hw,
 	data[i++] = pdev_stats->rts_good;
 	data[i++] = pdev_stats->chan_tx_power;
 	data[i++] = pdev_stats->fcs_bad;
+	data[i++] = ar->stats.rx_crc_err_drop;
 	data[i++] = pdev_stats->no_beacons;
 	data[i++] = pdev_stats->mpdu_enqued;
 	data[i++] = pdev_stats->msdu_enqued;
diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c
index 0f1f71e43b9b..9faecfa5f9d7 100644
--- a/drivers/net/wireless/ath/ath10k/htt_rx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_rx.c
@@ -1285,6 +1285,13 @@ static void ath10k_process_rx(struct ath10k *ar, struct sk_buff *skb)
 
 	status = IEEE80211_SKB_RXCB(skb);
 
+	if (!(ar->filter_flags & FIF_FCSFAIL) &&
+	    status->flag & RX_FLAG_FAILED_FCS_CRC) {
+		ar->stats.rx_crc_err_drop++;
+		dev_kfree_skb_any(skb);
+		return;
+	}
+
 	ath10k_dbg(ar, ATH10K_DBG_DATA,
 		   "rx skb %pK len %u peer %pM %s %s sn %u %s%s%s%s%s%s %srate_idx %u vht_nss %u freq %u band %u flag 0x%x fcs-err %i mic-err %i amsdu-more %i\n",
 		   skb,
-- 
cgit v1.2.3-59-g8ed1b


From 5b1413f00b5beb9f5fed94e43ea0c497d5db9633 Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.king@canonical.com>
Date: Fri, 15 Nov 2019 12:09:53 +0000
Subject: wil6210: fix break that is never reached because of zero'ing of a
 retry counter

There is a check on the retry counter invalid_buf_id_retry that is always
false because invalid_buf_id_retry is initialized to zero on each iteration
of a while-loop.  Fix this by initializing the retry counter before the
while-loop starts.

Addresses-Coverity: ("Logically dead code")
Fixes: b4a967b7d0f5 ("wil6210: reset buff id in status message after completion")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Maya Erez <merez@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/wil6210/txrx_edma.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/wil6210/txrx_edma.c b/drivers/net/wireless/ath/wil6210/txrx_edma.c
index 778b63be6a9a..02548d40253c 100644
--- a/drivers/net/wireless/ath/wil6210/txrx_edma.c
+++ b/drivers/net/wireless/ath/wil6210/txrx_edma.c
@@ -869,6 +869,7 @@ static struct sk_buff *wil_sring_reap_rx_edma(struct wil6210_priv *wil,
 	u8 data_offset;
 	struct wil_rx_status_extended *s;
 	u16 sring_idx = sring - wil->srings;
+	int invalid_buff_id_retry;
 
 	BUILD_BUG_ON(sizeof(struct wil_rx_status_extended) > sizeof(skb->cb));
 
@@ -882,9 +883,9 @@ again:
 	/* Extract the buffer ID from the status message */
 	buff_id = le16_to_cpu(wil_rx_status_get_buff_id(msg));
 
+	invalid_buff_id_retry = 0;
 	while (!buff_id) {
 		struct wil_rx_status_extended *s;
-		int invalid_buff_id_retry = 0;
 
 		wil_dbg_txrx(wil,
 			     "buff_id is not updated yet by HW, (swhead 0x%x)\n",
-- 
cgit v1.2.3-59-g8ed1b


From dae0978d4174422b4a2aa08a96252a1e60d40ef5 Mon Sep 17 00:00:00 2001
From: Kalle Valo <kvalo@codeaurora.org>
Date: Fri, 15 Nov 2019 11:28:25 +0200
Subject: dt: bindings: net: add qcom,ath11k.yaml

ath11k is a driver for Qualcomm IEEE 802.11ax devices. Add a
bindings document for the driver, first documenting IPQ8074 which is the
only device ath11k currently supports.

Signed-off-by: Anilkumar Kolli <akolli@codeaurora.org>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 .../bindings/net/wireless/qcom,ath11k.yaml         | 273 +++++++++++++++++++++
 1 file changed, 273 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/net/wireless/qcom,ath11k.yaml

diff --git a/Documentation/devicetree/bindings/net/wireless/qcom,ath11k.yaml b/Documentation/devicetree/bindings/net/wireless/qcom,ath11k.yaml
new file mode 100644
index 000000000000..a1717db36dba
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/wireless/qcom,ath11k.yaml
@@ -0,0 +1,273 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+# Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/net/wireless/qcom,ath11k.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Qualcomm Technologies ath11k wireless devices Generic Binding
+
+maintainers:
+  - Kalle Valo <kvalo@codeaurora.org>
+
+description: |
+  These are dt entries for Qualcomm Technologies, Inc. IEEE 802.11ax
+  devices, for example like AHB based IPQ8074.
+
+properties:
+  compatible:
+    const: qcom,ipq8074-wifi
+
+  reg:
+    maxItems: 1
+
+  interrupts:
+    items:
+      - description: misc-pulse1 interrupt events
+      - description: misc-latch interrupt events
+      - description: sw exception interrupt events
+      - description: watchdog interrupt events
+      - description: interrupt event for ring CE0
+      - description: interrupt event for ring CE1
+      - description: interrupt event for ring CE2
+      - description: interrupt event for ring CE3
+      - description: interrupt event for ring CE4
+      - description: interrupt event for ring CE5
+      - description: interrupt event for ring CE6
+      - description: interrupt event for ring CE7
+      - description: interrupt event for ring CE8
+      - description: interrupt event for ring CE9
+      - description: interrupt event for ring CE10
+      - description: interrupt event for ring CE11
+      - description: interrupt event for ring host2wbm-desc-feed
+      - description: interrupt event for ring host2reo-re-injection
+      - description: interrupt event for ring host2reo-command
+      - description: interrupt event for ring host2rxdma-monitor-ring3
+      - description: interrupt event for ring host2rxdma-monitor-ring2
+      - description: interrupt event for ring host2rxdma-monitor-ring1
+      - description: interrupt event for ring reo2ost-exception
+      - description: interrupt event for ring wbm2host-rx-release
+      - description: interrupt event for ring reo2host-status
+      - description: interrupt event for ring reo2host-destination-ring4
+      - description: interrupt event for ring reo2host-destination-ring3
+      - description: interrupt event for ring reo2host-destination-ring2
+      - description: interrupt event for ring reo2host-destination-ring1
+      - description: interrupt event for ring rxdma2host-monitor-destination-mac3
+      - description: interrupt event for ring rxdma2host-monitor-destination-mac2
+      - description: interrupt event for ring rxdma2host-monitor-destination-mac1
+      - description: interrupt event for ring ppdu-end-interrupts-mac3
+      - description: interrupt event for ring ppdu-end-interrupts-mac2
+      - description: interrupt event for ring ppdu-end-interrupts-mac1
+      - description: interrupt event for ring rxdma2host-monitor-status-ring-mac3
+      - description: interrupt event for ring rxdma2host-monitor-status-ring-mac2
+      - description: interrupt event for ring rxdma2host-monitor-status-ring-mac1
+      - description: interrupt event for ring host2rxdma-host-buf-ring-mac3
+      - description: interrupt event for ring host2rxdma-host-buf-ring-mac2
+      - description: interrupt event for ring host2rxdma-host-buf-ring-mac1
+      - description: interrupt event for ring rxdma2host-destination-ring-mac3
+      - description: interrupt event for ring rxdma2host-destination-ring-mac2
+      - description: interrupt event for ring rxdma2host-destination-ring-mac1
+      - description: interrupt event for ring host2tcl-input-ring4
+      - description: interrupt event for ring host2tcl-input-ring3
+      - description: interrupt event for ring host2tcl-input-ring2
+      - description: interrupt event for ring host2tcl-input-ring1
+      - description: interrupt event for ring wbm2host-tx-completions-ring3
+      - description: interrupt event for ring wbm2host-tx-completions-ring2
+      - description: interrupt event for ring wbm2host-tx-completions-ring1
+      - description: interrupt event for ring tcl2host-status-ring
+
+
+  interrupt-names:
+    items:
+      - const: misc-pulse1
+      - const: misc-latch
+      - const: sw-exception
+      - const: watchdog
+      - const: ce0
+      - const: ce1
+      - const: ce2
+      - const: ce3
+      - const: ce4
+      - const: ce5
+      - const: ce6
+      - const: ce7
+      - const: ce8
+      - const: ce9
+      - const: ce10
+      - const: ce11
+      - const: host2wbm-desc-feed
+      - const: host2reo-re-injection
+      - const: host2reo-command
+      - const: host2rxdma-monitor-ring3
+      - const: host2rxdma-monitor-ring2
+      - const: host2rxdma-monitor-ring1
+      - const: reo2ost-exception
+      - const: wbm2host-rx-release
+      - const: reo2host-status
+      - const: reo2host-destination-ring4
+      - const: reo2host-destination-ring3
+      - const: reo2host-destination-ring2
+      - const: reo2host-destination-ring1
+      - const: rxdma2host-monitor-destination-mac3
+      - const: rxdma2host-monitor-destination-mac2
+      - const: rxdma2host-monitor-destination-mac1
+      - const: ppdu-end-interrupts-mac3
+      - const: ppdu-end-interrupts-mac2
+      - const: ppdu-end-interrupts-mac1
+      - const: rxdma2host-monitor-status-ring-mac3
+      - const: rxdma2host-monitor-status-ring-mac2
+      - const: rxdma2host-monitor-status-ring-mac1
+      - const: host2rxdma-host-buf-ring-mac3
+      - const: host2rxdma-host-buf-ring-mac2
+      - const: host2rxdma-host-buf-ring-mac1
+      - const: rxdma2host-destination-ring-mac3
+      - const: rxdma2host-destination-ring-mac2
+      - const: rxdma2host-destination-ring-mac1
+      - const: host2tcl-input-ring4
+      - const: host2tcl-input-ring3
+      - const: host2tcl-input-ring2
+      - const: host2tcl-input-ring1
+      - const: wbm2host-tx-completions-ring3
+      - const: wbm2host-tx-completions-ring2
+      - const: wbm2host-tx-completions-ring1
+      - const: tcl2host-status-ring
+
+  qcom,rproc:
+    $ref: /schemas/types.yaml#definitions/phandle
+    description:
+      DT entry of q6v5-wcss remoteproc driver.
+      Phandle to a node that can contain the following properties
+        * compatible
+        * reg
+        * reg-names
+
+required:
+  - compatible
+  - reg
+  - interrupts
+  - interrupt-names
+  - qcom,rproc
+
+additionalProperties: false
+
+examples:
+  - |
+
+    q6v5_wcss: q6v5_wcss@CD00000 {
+        compatible = "qcom,ipq8074-wcss-pil";
+        reg = <0xCD00000 0x4040>,
+              <0x4AB000 0x20>;
+        reg-names = "qdsp6",
+                    "rmb";
+    };
+
+    wifi0: wifi@c000000 {
+        compatible = "qcom,ipq8074-wifi";
+        reg = <0xc000000 0x2000000>;
+        interrupts = <0 320 1>,
+                     <0 319 1>,
+                     <0 318 1>,
+                     <0 317 1>,
+                     <0 316 1>,
+                     <0 315 1>,
+                     <0 314 1>,
+                     <0 311 1>,
+                     <0 310 1>,
+                     <0 411 1>,
+                     <0 410 1>,
+                     <0 40 1>,
+                     <0 39 1>,
+                     <0 302 1>,
+                     <0 301 1>,
+                     <0 37 1>,
+                     <0 36 1>,
+                     <0 296 1>,
+                     <0 295 1>,
+                     <0 294 1>,
+                     <0 293 1>,
+                     <0 292 1>,
+                     <0 291 1>,
+                     <0 290 1>,
+                     <0 289 1>,
+                     <0 288 1>,
+                     <0 239 1>,
+                     <0 236 1>,
+                     <0 235 1>,
+                     <0 234 1>,
+                     <0 233 1>,
+                     <0 232 1>,
+                     <0 231 1>,
+                     <0 230 1>,
+                     <0 229 1>,
+                     <0 228 1>,
+                     <0 224 1>,
+                     <0 223 1>,
+                     <0 203 1>,
+                     <0 183 1>,
+                     <0 180 1>,
+                     <0 179 1>,
+                     <0 178 1>,
+                     <0 177 1>,
+                     <0 176 1>,
+                     <0 163 1>,
+                     <0 162 1>,
+                     <0 160 1>,
+                     <0 159 1>,
+                     <0 158 1>,
+                     <0 157 1>,
+                     <0 156 1>;
+        interrupt-names = "misc-pulse1",
+                          "misc-latch",
+                          "sw-exception",
+                          "watchdog",
+                          "ce0",
+                          "ce1",
+                          "ce2",
+                          "ce3",
+                          "ce4",
+                          "ce5",
+                          "ce6",
+                          "ce7",
+                          "ce8",
+                          "ce9",
+                          "ce10",
+                          "ce11",
+                          "host2wbm-desc-feed",
+                          "host2reo-re-injection",
+                          "host2reo-command",
+                          "host2rxdma-monitor-ring3",
+                          "host2rxdma-monitor-ring2",
+                          "host2rxdma-monitor-ring1",
+                          "reo2ost-exception",
+                          "wbm2host-rx-release",
+                          "reo2host-status",
+                          "reo2host-destination-ring4",
+                          "reo2host-destination-ring3",
+                          "reo2host-destination-ring2",
+                          "reo2host-destination-ring1",
+                          "rxdma2host-monitor-destination-mac3",
+                          "rxdma2host-monitor-destination-mac2",
+                          "rxdma2host-monitor-destination-mac1",
+                          "ppdu-end-interrupts-mac3",
+                          "ppdu-end-interrupts-mac2",
+                          "ppdu-end-interrupts-mac1",
+                          "rxdma2host-monitor-status-ring-mac3",
+                          "rxdma2host-monitor-status-ring-mac2",
+                          "rxdma2host-monitor-status-ring-mac1",
+                          "host2rxdma-host-buf-ring-mac3",
+                          "host2rxdma-host-buf-ring-mac2",
+                          "host2rxdma-host-buf-ring-mac1",
+                          "rxdma2host-destination-ring-mac3",
+                          "rxdma2host-destination-ring-mac2",
+                          "rxdma2host-destination-ring-mac1",
+                          "host2tcl-input-ring4",
+                          "host2tcl-input-ring3",
+                          "host2tcl-input-ring2",
+                          "host2tcl-input-ring1",
+                          "wbm2host-tx-completions-ring3",
+                          "wbm2host-tx-completions-ring2",
+                          "wbm2host-tx-completions-ring1",
+                          "tcl2host-status-ring";
+        qcom,rproc = <&q6v5_wcss>;
+    };
-- 
cgit v1.2.3-59-g8ed1b


From d5c65159f2895379e11ca13f62feabe93278985d Mon Sep 17 00:00:00 2001
From: Kalle Valo <kvalo@codeaurora.org>
Date: Sat, 23 Nov 2019 09:58:40 +0200
Subject: ath11k: driver for Qualcomm IEEE 802.11ax devices

ath11k is a new driver for Qualcomm IEEE 802.11ax devices, first
supporting only IPQ8074 SoC using the shared memory AHB bus. ath11k
uses mac80211 and supports AP, Station and Mesh modes.

Even though ath11k has some similar code as with ath10k (especially
the WMI layer) it was concluded to be simpler to have a "clean start"
for ath11k code base and not try to share the code with ath10k. This
makes maintenance easier and avoids major changes in ath10k, which
would have significantly increased the risk of regressions in existing
setups.

Even though the driver is very similar with ath10k but there are major
differences as well. The datapath is completely different. ath11k
supports multiple MACs, called "soc" in the firmware interface. And
there's only one WMI interface to support.

Currently ath11k supports only IEEE 802.11ac mode, but patches for
802.11ax are available and they will be submitted after ath11k is
accepted to upstream.

The firmware images are available from ath11k-firmware repository but
they will be also submitted to linux-firmware:

https://github.com/kvalo/ath11k-firmware

This was tested with firmware version WLAN.HK.2.1.0.1-00629-QCAHKSWPL_SILICONZ-1.

The driver has had multiple authors who are listed in alphabetical
order below.

Signed-off-by: Anilkumar Kolli <akolli@codeaurora.org>
Signed-off-by: Bhagavathi Perumal S <bperumal@codeaurora.org>
Signed-off-by: Ganesh Sesetti <gseset@codeaurora.org>
Signed-off-by: Govindaraj Saminathan <gsamin@codeaurora.org>
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Julia Lawall <julia.lawall@lip6.fr>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Signed-off-by: Karthikeyan Periyasamy <periyasa@codeaurora.org>
Signed-off-by: kbuild test robot <lkp@intel.com>
Signed-off-by: Maharaja Kennadyrajan <mkenna@codeaurora.org>
Signed-off-by: Manikanta Pubbisetty <mpubbise@codeaurora.org>
Signed-off-by: Miles Hu <milehu@codeaurora.org>
Signed-off-by: Muna Sinada <msinada@codeaurora.org>
Signed-off-by: Pradeep Kumar Chitrapu <pradeepc@codeaurora.org>
Signed-off-by: Rajkumar Manoharan <rmanohar@codeaurora.org>
Signed-off-by: Sathishkumar Muruganandam <murugana@codeaurora.org>
Signed-off-by: Shashidhar Lakkavalli <slakkavalli@datto.com>
Signed-off-by: Sriram R <srirrama@codeaurora.org>
Signed-off-by: Sven Eckelmann <seckelmann@datto.com>
Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@codeaurora.org>
Signed-off-by: Venkateswara Naralasetty <vnaralas@codeaurora.org>
---
 drivers/net/wireless/ath/Kconfig                  |    1 +
 drivers/net/wireless/ath/Makefile                 |    1 +
 drivers/net/wireless/ath/ath11k/Kconfig           |   35 +
 drivers/net/wireless/ath/ath11k/Makefile          |   26 +
 drivers/net/wireless/ath/ath11k/ahb.c             | 1004 ++++
 drivers/net/wireless/ath/ath11k/ahb.h             |   35 +
 drivers/net/wireless/ath/ath11k/ce.c              |  805 +++
 drivers/net/wireless/ath/ath11k/ce.h              |  178 +
 drivers/net/wireless/ath/ath11k/core.c            |  772 +++
 drivers/net/wireless/ath/ath11k/core.h            |  823 +++
 drivers/net/wireless/ath/ath11k/debug.c           | 1060 ++++
 drivers/net/wireless/ath/ath11k/debug.h           |  278 +
 drivers/net/wireless/ath/ath11k/debug_htt_stats.c | 4432 ++++++++++++++++
 drivers/net/wireless/ath/ath11k/debug_htt_stats.h | 1620 ++++++
 drivers/net/wireless/ath/ath11k/debugfs_sta.c     |  538 ++
 drivers/net/wireless/ath/ath11k/dp.c              |  898 ++++
 drivers/net/wireless/ath/ath11k/dp.h              | 1526 ++++++
 drivers/net/wireless/ath/ath11k/dp_rx.c           | 4143 +++++++++++++++
 drivers/net/wireless/ath/ath11k/dp_rx.h           |   84 +
 drivers/net/wireless/ath/ath11k/dp_tx.c           |  943 ++++
 drivers/net/wireless/ath/ath11k/dp_tx.h           |   40 +
 drivers/net/wireless/ath/ath11k/hal.c             | 1124 ++++
 drivers/net/wireless/ath/ath11k/hal.h             |  897 ++++
 drivers/net/wireless/ath/ath11k/hal_desc.h        | 2469 +++++++++
 drivers/net/wireless/ath/ath11k/hal_rx.c          | 1221 +++++
 drivers/net/wireless/ath/ath11k/hal_rx.h          |  352 ++
 drivers/net/wireless/ath/ath11k/hal_tx.c          |  226 +
 drivers/net/wireless/ath/ath11k/hal_tx.h          |   80 +
 drivers/net/wireless/ath/ath11k/htc.c             |  773 +++
 drivers/net/wireless/ath/ath11k/htc.h             |  313 ++
 drivers/net/wireless/ath/ath11k/hw.h              |  127 +
 drivers/net/wireless/ath/ath11k/mac.c             | 5427 ++++++++++++++++++++
 drivers/net/wireless/ath/ath11k/mac.h             |  145 +
 drivers/net/wireless/ath/ath11k/peer.c            |  236 +
 drivers/net/wireless/ath/ath11k/peer.h            |   35 +
 drivers/net/wireless/ath/ath11k/qmi.c             | 2435 +++++++++
 drivers/net/wireless/ath/ath11k/qmi.h             |  445 ++
 drivers/net/wireless/ath/ath11k/reg.c             |  701 +++
 drivers/net/wireless/ath/ath11k/reg.h             |   35 +
 drivers/net/wireless/ath/ath11k/rx_desc.h         | 1212 +++++
 drivers/net/wireless/ath/ath11k/testmode.c        |  199 +
 drivers/net/wireless/ath/ath11k/testmode.h        |   29 +
 drivers/net/wireless/ath/ath11k/testmode_i.h      |   50 +
 drivers/net/wireless/ath/ath11k/trace.c           |    9 +
 drivers/net/wireless/ath/ath11k/trace.h           |  110 +
 drivers/net/wireless/ath/ath11k/wmi.c             | 5632 +++++++++++++++++++++
 drivers/net/wireless/ath/ath11k/wmi.h             | 4664 +++++++++++++++++
 47 files changed, 48188 insertions(+)
 create mode 100644 drivers/net/wireless/ath/ath11k/Kconfig
 create mode 100644 drivers/net/wireless/ath/ath11k/Makefile
 create mode 100644 drivers/net/wireless/ath/ath11k/ahb.c
 create mode 100644 drivers/net/wireless/ath/ath11k/ahb.h
 create mode 100644 drivers/net/wireless/ath/ath11k/ce.c
 create mode 100644 drivers/net/wireless/ath/ath11k/ce.h
 create mode 100644 drivers/net/wireless/ath/ath11k/core.c
 create mode 100644 drivers/net/wireless/ath/ath11k/core.h
 create mode 100644 drivers/net/wireless/ath/ath11k/debug.c
 create mode 100644 drivers/net/wireless/ath/ath11k/debug.h
 create mode 100644 drivers/net/wireless/ath/ath11k/debug_htt_stats.c
 create mode 100644 drivers/net/wireless/ath/ath11k/debug_htt_stats.h
 create mode 100644 drivers/net/wireless/ath/ath11k/debugfs_sta.c
 create mode 100644 drivers/net/wireless/ath/ath11k/dp.c
 create mode 100644 drivers/net/wireless/ath/ath11k/dp.h
 create mode 100644 drivers/net/wireless/ath/ath11k/dp_rx.c
 create mode 100644 drivers/net/wireless/ath/ath11k/dp_rx.h
 create mode 100644 drivers/net/wireless/ath/ath11k/dp_tx.c
 create mode 100644 drivers/net/wireless/ath/ath11k/dp_tx.h
 create mode 100644 drivers/net/wireless/ath/ath11k/hal.c
 create mode 100644 drivers/net/wireless/ath/ath11k/hal.h
 create mode 100644 drivers/net/wireless/ath/ath11k/hal_desc.h
 create mode 100644 drivers/net/wireless/ath/ath11k/hal_rx.c
 create mode 100644 drivers/net/wireless/ath/ath11k/hal_rx.h
 create mode 100644 drivers/net/wireless/ath/ath11k/hal_tx.c
 create mode 100644 drivers/net/wireless/ath/ath11k/hal_tx.h
 create mode 100644 drivers/net/wireless/ath/ath11k/htc.c
 create mode 100644 drivers/net/wireless/ath/ath11k/htc.h
 create mode 100644 drivers/net/wireless/ath/ath11k/hw.h
 create mode 100644 drivers/net/wireless/ath/ath11k/mac.c
 create mode 100644 drivers/net/wireless/ath/ath11k/mac.h
 create mode 100644 drivers/net/wireless/ath/ath11k/peer.c
 create mode 100644 drivers/net/wireless/ath/ath11k/peer.h
 create mode 100644 drivers/net/wireless/ath/ath11k/qmi.c
 create mode 100644 drivers/net/wireless/ath/ath11k/qmi.h
 create mode 100644 drivers/net/wireless/ath/ath11k/reg.c
 create mode 100644 drivers/net/wireless/ath/ath11k/reg.h
 create mode 100644 drivers/net/wireless/ath/ath11k/rx_desc.h
 create mode 100644 drivers/net/wireless/ath/ath11k/testmode.c
 create mode 100644 drivers/net/wireless/ath/ath11k/testmode.h
 create mode 100644 drivers/net/wireless/ath/ath11k/testmode_i.h
 create mode 100644 drivers/net/wireless/ath/ath11k/trace.c
 create mode 100644 drivers/net/wireless/ath/ath11k/trace.h
 create mode 100644 drivers/net/wireless/ath/ath11k/wmi.c
 create mode 100644 drivers/net/wireless/ath/ath11k/wmi.h

diff --git a/drivers/net/wireless/ath/Kconfig b/drivers/net/wireless/ath/Kconfig
index 56616d988c96..d6df17d312cd 100644
--- a/drivers/net/wireless/ath/Kconfig
+++ b/drivers/net/wireless/ath/Kconfig
@@ -62,5 +62,6 @@ source "drivers/net/wireless/ath/ar5523/Kconfig"
 source "drivers/net/wireless/ath/wil6210/Kconfig"
 source "drivers/net/wireless/ath/ath10k/Kconfig"
 source "drivers/net/wireless/ath/wcn36xx/Kconfig"
+source "drivers/net/wireless/ath/ath11k/Kconfig"
 
 endif
diff --git a/drivers/net/wireless/ath/Makefile b/drivers/net/wireless/ath/Makefile
index ee2b2431e5a3..8e4ae9de5ae4 100644
--- a/drivers/net/wireless/ath/Makefile
+++ b/drivers/net/wireless/ath/Makefile
@@ -7,6 +7,7 @@ obj-$(CONFIG_AR5523)		+= ar5523/
 obj-$(CONFIG_WIL6210)		+= wil6210/
 obj-$(CONFIG_ATH10K)		+= ath10k/
 obj-$(CONFIG_WCN36XX)		+= wcn36xx/
+obj-$(CONFIG_ATH11K)		+= ath11k/
 
 obj-$(CONFIG_ATH_COMMON)	+= ath.o
 
diff --git a/drivers/net/wireless/ath/ath11k/Kconfig b/drivers/net/wireless/ath/ath11k/Kconfig
new file mode 100644
index 000000000000..cfab4fb86aef
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/Kconfig
@@ -0,0 +1,35 @@
+# SPDX-License-Identifier: BSD-3-Clause-Clear
+config ATH11K
+	tristate "Qualcomm Technologies 802.11ax chipset support"
+	depends on MAC80211 && HAS_DMA
+	depends on REMOTEPROC
+	depends on ARCH_QCOM || COMPILE_TEST
+	select ATH_COMMON
+	select QCOM_QMI_HELPERS
+	---help---
+	  This module adds support for Qualcomm Technologies 802.11ax family of
+	  chipsets.
+
+	  If you choose to build a module, it'll be called ath11k.
+
+config ATH11K_DEBUG
+	bool "QCA ath11k debugging"
+	depends on ATH11K
+	---help---
+	  Enables debug support
+
+	  If unsure, say Y to make it easier to debug problems.
+
+config ATH11K_DEBUGFS
+	bool "QCA ath11k debugfs support"
+	depends on ATH11K && DEBUG_FS
+	---help---
+	  Enable ath11k debugfs support
+
+	  If unsure, say Y to make it easier to debug problems.
+
+config ATH11K_TRACING
+	bool "ath11k tracing support"
+	depends on ATH11K && EVENT_TRACING
+	---help---
+	  Select this to use ath11k tracing infrastructure.
diff --git a/drivers/net/wireless/ath/ath11k/Makefile b/drivers/net/wireless/ath/ath11k/Makefile
new file mode 100644
index 000000000000..a91d75c1cfeb
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/Makefile
@@ -0,0 +1,26 @@
+# SPDX-License-Identifier: BSD-3-Clause-Clear
+obj-$(CONFIG_ATH11K) += ath11k.o
+ath11k-y += core.o \
+	    hal.o \
+	    hal_tx.o \
+	    hal_rx.o \
+	    ahb.o \
+	    wmi.o \
+	    mac.o \
+	    reg.o \
+	    htc.o \
+	    qmi.o \
+	    dp.o  \
+	    dp_tx.o \
+	    dp_rx.o \
+	    debug.o \
+	    ce.o \
+	    peer.o
+
+ath11k-$(CONFIG_ATH11K_DEBUGFS) += debug_htt_stats.o
+ath11k-$(CONFIG_MAC80211_DEBUGFS) += debugfs_sta.o
+ath11k-$(CONFIG_NL80211_TESTMODE) += testmode.o
+ath11k-$(CONFIG_ATH11K_TRACING) += trace.o
+
+# for tracing framework to find trace.h
+CFLAGS_trace.o := -I$(src)
diff --git a/drivers/net/wireless/ath/ath11k/ahb.c b/drivers/net/wireless/ath/ath11k/ahb.c
new file mode 100644
index 000000000000..f80173b8afc6
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/ahb.c
@@ -0,0 +1,1004 @@
+// SPDX-License-Identifier: BSD-3-Clause-Clear
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/of_device.h>
+#include <linux/of.h>
+#include <linux/dma-mapping.h>
+#include "ahb.h"
+#include "debug.h"
+#include <linux/remoteproc.h>
+
+static const struct of_device_id ath11k_ahb_of_match[] = {
+	/* TODO: Should we change the compatible string to something similar
+	 * to one that ath10k uses?
+	 */
+	{ .compatible = "qcom,ipq8074-wifi",
+	  .data = (void *)ATH11K_HW_IPQ8074,
+	},
+	{ }
+};
+
+MODULE_DEVICE_TABLE(of, ath11k_ahb_of_match);
+
+/* Target firmware's Copy Engine configuration. */
+static const struct ce_pipe_config target_ce_config_wlan[] = {
+	/* CE0: host->target HTC control and raw streams */
+	{
+		.pipenum = __cpu_to_le32(0),
+		.pipedir = __cpu_to_le32(PIPEDIR_OUT),
+		.nentries = __cpu_to_le32(32),
+		.nbytes_max = __cpu_to_le32(2048),
+		.flags = __cpu_to_le32(CE_ATTR_FLAGS),
+		.reserved = __cpu_to_le32(0),
+	},
+
+	/* CE1: target->host HTT + HTC control */
+	{
+		.pipenum = __cpu_to_le32(1),
+		.pipedir = __cpu_to_le32(PIPEDIR_IN),
+		.nentries = __cpu_to_le32(32),
+		.nbytes_max = __cpu_to_le32(2048),
+		.flags = __cpu_to_le32(CE_ATTR_FLAGS),
+		.reserved = __cpu_to_le32(0),
+	},
+
+	/* CE2: target->host WMI */
+	{
+		.pipenum = __cpu_to_le32(2),
+		.pipedir = __cpu_to_le32(PIPEDIR_IN),
+		.nentries = __cpu_to_le32(32),
+		.nbytes_max = __cpu_to_le32(2048),
+		.flags = __cpu_to_le32(CE_ATTR_FLAGS),
+		.reserved = __cpu_to_le32(0),
+	},
+
+	/* CE3: host->target WMI */
+	{
+		.pipenum = __cpu_to_le32(3),
+		.pipedir = __cpu_to_le32(PIPEDIR_OUT),
+		.nentries = __cpu_to_le32(32),
+		.nbytes_max = __cpu_to_le32(2048),
+		.flags = __cpu_to_le32(CE_ATTR_FLAGS),
+		.reserved = __cpu_to_le32(0),
+	},
+
+	/* CE4: host->target HTT */
+	{
+		.pipenum = __cpu_to_le32(4),
+		.pipedir = __cpu_to_le32(PIPEDIR_OUT),
+		.nentries = __cpu_to_le32(256),
+		.nbytes_max = __cpu_to_le32(256),
+		.flags = __cpu_to_le32(CE_ATTR_FLAGS | CE_ATTR_DIS_INTR),
+		.reserved = __cpu_to_le32(0),
+	},
+
+	/* CE5: target->host Pktlog */
+	{
+		.pipenum = __cpu_to_le32(5),
+		.pipedir = __cpu_to_le32(PIPEDIR_IN),
+		.nentries = __cpu_to_le32(32),
+		.nbytes_max = __cpu_to_le32(2048),
+		.flags = __cpu_to_le32(0),
+		.reserved = __cpu_to_le32(0),
+	},
+
+	/* CE6: Reserved for target autonomous hif_memcpy */
+	{
+		.pipenum = __cpu_to_le32(6),
+		.pipedir = __cpu_to_le32(PIPEDIR_INOUT),
+		.nentries = __cpu_to_le32(32),
+		.nbytes_max = __cpu_to_le32(65535),
+		.flags = __cpu_to_le32(CE_ATTR_FLAGS),
+		.reserved = __cpu_to_le32(0),
+	},
+
+	/* CE7 used only by Host */
+	{
+		.pipenum = __cpu_to_le32(7),
+		.pipedir = __cpu_to_le32(PIPEDIR_OUT),
+		.nentries = __cpu_to_le32(32),
+		.nbytes_max = __cpu_to_le32(2048),
+		.flags = __cpu_to_le32(CE_ATTR_FLAGS),
+		.reserved = __cpu_to_le32(0),
+	},
+
+	/* CE8 target->host used only by IPA */
+	{
+		.pipenum = __cpu_to_le32(8),
+		.pipedir = __cpu_to_le32(PIPEDIR_INOUT),
+		.nentries = __cpu_to_le32(32),
+		.nbytes_max = __cpu_to_le32(65535),
+		.flags = __cpu_to_le32(CE_ATTR_FLAGS),
+		.reserved = __cpu_to_le32(0),
+	},
+
+	/* CE9 host->target HTT */
+	{
+		.pipenum = __cpu_to_le32(9),
+		.pipedir = __cpu_to_le32(PIPEDIR_OUT),
+		.nentries = __cpu_to_le32(32),
+		.nbytes_max = __cpu_to_le32(2048),
+		.flags = __cpu_to_le32(CE_ATTR_FLAGS),
+		.reserved = __cpu_to_le32(0),
+	},
+
+	/* CE10 target->host HTT */
+	{
+		.pipenum = __cpu_to_le32(10),
+		.pipedir = __cpu_to_le32(PIPEDIR_INOUT_H2H),
+		.nentries = __cpu_to_le32(0),
+		.nbytes_max = __cpu_to_le32(0),
+		.flags = __cpu_to_le32(CE_ATTR_FLAGS),
+		.reserved = __cpu_to_le32(0),
+	},
+
+	/* CE11 Not used */
+	{
+		.pipenum = __cpu_to_le32(0),
+		.pipedir = __cpu_to_le32(0),
+		.nentries = __cpu_to_le32(0),
+		.nbytes_max = __cpu_to_le32(0),
+		.flags = __cpu_to_le32(CE_ATTR_FLAGS),
+		.reserved = __cpu_to_le32(0),
+	},
+};
+
+/* Map from service/endpoint to Copy Engine.
+ * This table is derived from the CE_PCI TABLE, above.
+ * It is passed to the Target at startup for use by firmware.
+ */
+static const struct service_to_pipe target_service_to_ce_map_wlan[] = {
+	{
+		.service_id = __cpu_to_le32(ATH11K_HTC_SVC_ID_WMI_DATA_VO),
+		.pipedir = __cpu_to_le32(PIPEDIR_OUT),	/* out = UL = host -> target */
+		.pipenum = __cpu_to_le32(3),
+	},
+	{
+		.service_id = __cpu_to_le32(ATH11K_HTC_SVC_ID_WMI_DATA_VO),
+		.pipedir = __cpu_to_le32(PIPEDIR_IN),	/* in = DL = target -> host */
+		.pipenum = __cpu_to_le32(2),
+	},
+	{
+		.service_id = __cpu_to_le32(ATH11K_HTC_SVC_ID_WMI_DATA_BK),
+		.pipedir = __cpu_to_le32(PIPEDIR_OUT),	/* out = UL = host -> target */
+		.pipenum = __cpu_to_le32(3),
+	},
+	{
+		.service_id = __cpu_to_le32(ATH11K_HTC_SVC_ID_WMI_DATA_BK),
+		.pipedir = __cpu_to_le32(PIPEDIR_IN),	/* in = DL = target -> host */
+		.pipenum = __cpu_to_le32(2),
+	},
+	{
+		.service_id = __cpu_to_le32(ATH11K_HTC_SVC_ID_WMI_DATA_BE),
+		.pipedir = __cpu_to_le32(PIPEDIR_OUT),	/* out = UL = host -> target */
+		.pipenum = __cpu_to_le32(3),
+	},
+	{
+		.service_id = __cpu_to_le32(ATH11K_HTC_SVC_ID_WMI_DATA_BE),
+		.pipedir = __cpu_to_le32(PIPEDIR_IN),	/* in = DL = target -> host */
+		.pipenum = __cpu_to_le32(2),
+	},
+	{
+		.service_id = __cpu_to_le32(ATH11K_HTC_SVC_ID_WMI_DATA_VI),
+		.pipedir = __cpu_to_le32(PIPEDIR_OUT),	/* out = UL = host -> target */
+		.pipenum = __cpu_to_le32(3),
+	},
+	{
+		.service_id = __cpu_to_le32(ATH11K_HTC_SVC_ID_WMI_DATA_VI),
+		.pipedir = __cpu_to_le32(PIPEDIR_IN),	/* in = DL = target -> host */
+		.pipenum = __cpu_to_le32(2),
+	},
+	{
+		.service_id = __cpu_to_le32(ATH11K_HTC_SVC_ID_WMI_CONTROL),
+		.pipedir = __cpu_to_le32(PIPEDIR_OUT),	/* out = UL = host -> target */
+		.pipenum = __cpu_to_le32(3),
+	},
+	{
+		.service_id = __cpu_to_le32(ATH11K_HTC_SVC_ID_WMI_CONTROL),
+		.pipedir = __cpu_to_le32(PIPEDIR_IN),	/* in = DL = target -> host */
+		.pipenum = __cpu_to_le32(2),
+	},
+	{
+		.service_id = __cpu_to_le32(ATH11K_HTC_SVC_ID_WMI_CONTROL_MAC1),
+		.pipedir = __cpu_to_le32(PIPEDIR_OUT),	/* out = UL = host -> target */
+		.pipenum = __cpu_to_le32(7),
+	},
+	{
+		.service_id = __cpu_to_le32(ATH11K_HTC_SVC_ID_WMI_CONTROL_MAC1),
+		.pipedir = __cpu_to_le32(PIPEDIR_IN),	/* in = DL = target -> host */
+		.pipenum = __cpu_to_le32(2),
+	},
+	{
+		.service_id = __cpu_to_le32(ATH11K_HTC_SVC_ID_WMI_CONTROL_MAC2),
+		.pipedir = __cpu_to_le32(PIPEDIR_OUT),	/* out = UL = host -> target */
+		.pipenum = __cpu_to_le32(9),
+	},
+	{
+		.service_id = __cpu_to_le32(ATH11K_HTC_SVC_ID_WMI_CONTROL_MAC2),
+		.pipedir = __cpu_to_le32(PIPEDIR_IN),	/* in = DL = target -> host */
+		.pipenum = __cpu_to_le32(2),
+	},
+	{
+		.service_id = __cpu_to_le32(ATH11K_HTC_SVC_ID_RSVD_CTRL),
+		.pipedir = __cpu_to_le32(PIPEDIR_OUT),	/* out = UL = host -> target */
+		.pipenum = __cpu_to_le32(0),
+	},
+	{
+		.service_id = __cpu_to_le32(ATH11K_HTC_SVC_ID_RSVD_CTRL),
+		.pipedir = __cpu_to_le32(PIPEDIR_IN),	/* in = DL = target -> host */
+		.pipenum = __cpu_to_le32(1),
+	},
+	{ /* not used */
+		.service_id = __cpu_to_le32(ATH11K_HTC_SVC_ID_TEST_RAW_STREAMS),
+		.pipedir = __cpu_to_le32(PIPEDIR_OUT),	/* out = UL = host -> target */
+		.pipenum = __cpu_to_le32(0),
+	},
+	{ /* not used */
+		.service_id = __cpu_to_le32(ATH11K_HTC_SVC_ID_TEST_RAW_STREAMS),
+		.pipedir = __cpu_to_le32(PIPEDIR_IN),	/* in = DL = target -> host */
+		.pipenum = __cpu_to_le32(1),
+	},
+	{
+		.service_id = __cpu_to_le32(ATH11K_HTC_SVC_ID_HTT_DATA_MSG),
+		.pipedir = __cpu_to_le32(PIPEDIR_OUT),	/* out = UL = host -> target */
+		.pipenum = __cpu_to_le32(4),
+	},
+	{
+		.service_id = __cpu_to_le32(ATH11K_HTC_SVC_ID_HTT_DATA_MSG),
+		.pipedir = __cpu_to_le32(PIPEDIR_IN),	/* in = DL = target -> host */
+		.pipenum = __cpu_to_le32(1),
+	},
+	{
+		.service_id = __cpu_to_le32(ATH11K_HTC_SVC_ID_PKT_LOG),
+		.pipedir = __cpu_to_le32(PIPEDIR_IN),	/* in = DL = target -> host */
+		.pipenum = __cpu_to_le32(5),
+	},
+
+	/* (Additions here) */
+
+	{ /* terminator entry */ }
+};
+
+#define ATH11K_IRQ_CE0_OFFSET 4
+
+static const char *irq_name[ATH11K_IRQ_NUM_MAX] = {
+	"misc-pulse1",
+	"misc-latch",
+	"sw-exception",
+	"watchdog",
+	"ce0",
+	"ce1",
+	"ce2",
+	"ce3",
+	"ce4",
+	"ce5",
+	"ce6",
+	"ce7",
+	"ce8",
+	"ce9",
+	"ce10",
+	"ce11",
+	"host2wbm-desc-feed",
+	"host2reo-re-injection",
+	"host2reo-command",
+	"host2rxdma-monitor-ring3",
+	"host2rxdma-monitor-ring2",
+	"host2rxdma-monitor-ring1",
+	"reo2ost-exception",
+	"wbm2host-rx-release",
+	"reo2host-status",
+	"reo2host-destination-ring4",
+	"reo2host-destination-ring3",
+	"reo2host-destination-ring2",
+	"reo2host-destination-ring1",
+	"rxdma2host-monitor-destination-mac3",
+	"rxdma2host-monitor-destination-mac2",
+	"rxdma2host-monitor-destination-mac1",
+	"ppdu-end-interrupts-mac3",
+	"ppdu-end-interrupts-mac2",
+	"ppdu-end-interrupts-mac1",
+	"rxdma2host-monitor-status-ring-mac3",
+	"rxdma2host-monitor-status-ring-mac2",
+	"rxdma2host-monitor-status-ring-mac1",
+	"host2rxdma-host-buf-ring-mac3",
+	"host2rxdma-host-buf-ring-mac2",
+	"host2rxdma-host-buf-ring-mac1",
+	"rxdma2host-destination-ring-mac3",
+	"rxdma2host-destination-ring-mac2",
+	"rxdma2host-destination-ring-mac1",
+	"host2tcl-input-ring4",
+	"host2tcl-input-ring3",
+	"host2tcl-input-ring2",
+	"host2tcl-input-ring1",
+	"wbm2host-tx-completions-ring3",
+	"wbm2host-tx-completions-ring2",
+	"wbm2host-tx-completions-ring1",
+	"tcl2host-status-ring",
+};
+
+#define ATH11K_TX_RING_MASK_0 0x1
+#define ATH11K_TX_RING_MASK_1 0x2
+#define ATH11K_TX_RING_MASK_2 0x4
+
+#define ATH11K_RX_RING_MASK_0 0x1
+#define ATH11K_RX_RING_MASK_1 0x2
+#define ATH11K_RX_RING_MASK_2 0x4
+#define ATH11K_RX_RING_MASK_3 0x8
+
+#define ATH11K_RX_ERR_RING_MASK_0 0x1
+
+#define ATH11K_RX_WBM_REL_RING_MASK_0 0x1
+
+#define ATH11K_REO_STATUS_RING_MASK_0 0x1
+
+#define ATH11K_RXDMA2HOST_RING_MASK_0 0x1
+#define ATH11K_RXDMA2HOST_RING_MASK_1 0x2
+#define ATH11K_RXDMA2HOST_RING_MASK_2 0x4
+
+#define ATH11K_HOST2RXDMA_RING_MASK_0 0x1
+#define ATH11K_HOST2RXDMA_RING_MASK_1 0x2
+#define ATH11K_HOST2RXDMA_RING_MASK_2 0x4
+
+#define ATH11K_RX_MON_STATUS_RING_MASK_0 0x1
+#define ATH11K_RX_MON_STATUS_RING_MASK_1 0x2
+#define ATH11K_RX_MON_STATUS_RING_MASK_2 0x4
+
+const u8 ath11k_tx_ring_mask[ATH11K_EXT_IRQ_GRP_NUM_MAX] = {
+	ATH11K_TX_RING_MASK_0,
+	ATH11K_TX_RING_MASK_1,
+	ATH11K_TX_RING_MASK_2,
+};
+
+const u8 rx_mon_status_ring_mask[ATH11K_EXT_IRQ_GRP_NUM_MAX] = {
+	0, 0, 0, 0,
+	ATH11K_RX_MON_STATUS_RING_MASK_0,
+	ATH11K_RX_MON_STATUS_RING_MASK_1,
+	ATH11K_RX_MON_STATUS_RING_MASK_2,
+};
+
+const u8 ath11k_rx_ring_mask[ATH11K_EXT_IRQ_GRP_NUM_MAX] = {
+	0, 0, 0, 0, 0, 0, 0,
+	ATH11K_RX_RING_MASK_0,
+	ATH11K_RX_RING_MASK_1,
+	ATH11K_RX_RING_MASK_2,
+	ATH11K_RX_RING_MASK_3,
+};
+
+const u8 ath11k_rx_err_ring_mask[ATH11K_EXT_IRQ_GRP_NUM_MAX] = {
+	ATH11K_RX_ERR_RING_MASK_0,
+};
+
+const u8 ath11k_rx_wbm_rel_ring_mask[ATH11K_EXT_IRQ_GRP_NUM_MAX] = {
+	ATH11K_RX_WBM_REL_RING_MASK_0,
+};
+
+const u8 ath11k_reo_status_ring_mask[ATH11K_EXT_IRQ_GRP_NUM_MAX] = {
+	ATH11K_REO_STATUS_RING_MASK_0,
+};
+
+const u8 ath11k_rxdma2host_ring_mask[ATH11K_EXT_IRQ_GRP_NUM_MAX] = {
+	ATH11K_RXDMA2HOST_RING_MASK_0,
+	ATH11K_RXDMA2HOST_RING_MASK_1,
+	ATH11K_RXDMA2HOST_RING_MASK_2,
+};
+
+const u8 ath11k_host2rxdma_ring_mask[ATH11K_EXT_IRQ_GRP_NUM_MAX] = {
+	ATH11K_HOST2RXDMA_RING_MASK_0,
+	ATH11K_HOST2RXDMA_RING_MASK_1,
+	ATH11K_HOST2RXDMA_RING_MASK_2,
+};
+
+/* enum ext_irq_num - irq numbers that can be used by external modules
+ * like datapath
+ */
+enum ext_irq_num {
+	host2wbm_desc_feed = 16,
+	host2reo_re_injection,
+	host2reo_command,
+	host2rxdma_monitor_ring3,
+	host2rxdma_monitor_ring2,
+	host2rxdma_monitor_ring1,
+	reo2host_exception,
+	wbm2host_rx_release,
+	reo2host_status,
+	reo2host_destination_ring4,
+	reo2host_destination_ring3,
+	reo2host_destination_ring2,
+	reo2host_destination_ring1,
+	rxdma2host_monitor_destination_mac3,
+	rxdma2host_monitor_destination_mac2,
+	rxdma2host_monitor_destination_mac1,
+	ppdu_end_interrupts_mac3,
+	ppdu_end_interrupts_mac2,
+	ppdu_end_interrupts_mac1,
+	rxdma2host_monitor_status_ring_mac3,
+	rxdma2host_monitor_status_ring_mac2,
+	rxdma2host_monitor_status_ring_mac1,
+	host2rxdma_host_buf_ring_mac3,
+	host2rxdma_host_buf_ring_mac2,
+	host2rxdma_host_buf_ring_mac1,
+	rxdma2host_destination_ring_mac3,
+	rxdma2host_destination_ring_mac2,
+	rxdma2host_destination_ring_mac1,
+	host2tcl_input_ring4,
+	host2tcl_input_ring3,
+	host2tcl_input_ring2,
+	host2tcl_input_ring1,
+	wbm2host_tx_completions_ring3,
+	wbm2host_tx_completions_ring2,
+	wbm2host_tx_completions_ring1,
+	tcl2host_status_ring,
+};
+
+static void ath11k_ahb_kill_tasklets(struct ath11k_base *ab)
+{
+	int i;
+
+	for (i = 0; i < CE_COUNT; i++) {
+		struct ath11k_ce_pipe *ce_pipe = &ab->ce.ce_pipe[i];
+
+		if (ath11k_ce_get_attr_flags(i) & CE_ATTR_DIS_INTR)
+			continue;
+
+		tasklet_kill(&ce_pipe->intr_tq);
+	}
+}
+
+static void ath11k_ahb_ext_grp_disable(struct ath11k_ext_irq_grp *irq_grp)
+{
+	int i;
+
+	for (i = 0; i < irq_grp->num_irq; i++)
+		disable_irq_nosync(irq_grp->ab->irq_num[irq_grp->irqs[i]]);
+}
+
+static void __ath11k_ahb_ext_irq_disable(struct ath11k_base *ab)
+{
+	struct sk_buff *skb;
+	int i;
+
+	for (i = 0; i < ATH11K_EXT_IRQ_GRP_NUM_MAX; i++) {
+		struct ath11k_ext_irq_grp *irq_grp = &ab->ext_irq_grp[i];
+
+		ath11k_ahb_ext_grp_disable(irq_grp);
+
+		napi_synchronize(&irq_grp->napi);
+		napi_disable(&irq_grp->napi);
+
+		while ((skb = __skb_dequeue(&irq_grp->pending_q)))
+			dev_kfree_skb_any(skb);
+	}
+}
+
+static void ath11k_ahb_ext_grp_enable(struct ath11k_ext_irq_grp *irq_grp)
+{
+	int i;
+
+	for (i = 0; i < irq_grp->num_irq; i++)
+		enable_irq(irq_grp->ab->irq_num[irq_grp->irqs[i]]);
+}
+
+static void ath11k_ahb_setbit32(struct ath11k_base *ab, u8 bit, u32 offset)
+{
+	u32 val;
+
+	val = ath11k_ahb_read32(ab, offset);
+	ath11k_ahb_write32(ab, offset, val | BIT(bit));
+}
+
+static void ath11k_ahb_clearbit32(struct ath11k_base *ab, u8 bit, u32 offset)
+{
+	u32 val;
+
+	val = ath11k_ahb_read32(ab, offset);
+	ath11k_ahb_write32(ab, offset, val & ~BIT(bit));
+}
+
+static void ath11k_ahb_ce_irq_enable(struct ath11k_base *ab, u16 ce_id)
+{
+	const struct ce_pipe_config *ce_config;
+
+	ce_config = &target_ce_config_wlan[ce_id];
+	if (__le32_to_cpu(ce_config->pipedir) & PIPEDIR_OUT)
+		ath11k_ahb_setbit32(ab, ce_id, CE_HOST_IE_ADDRESS);
+
+	if (__le32_to_cpu(ce_config->pipedir) & PIPEDIR_IN) {
+		ath11k_ahb_setbit32(ab, ce_id, CE_HOST_IE_2_ADDRESS);
+		ath11k_ahb_setbit32(ab, ce_id + CE_HOST_IE_3_SHIFT,
+				    CE_HOST_IE_3_ADDRESS);
+	}
+}
+
+static void ath11k_ahb_ce_irq_disable(struct ath11k_base *ab, u16 ce_id)
+{
+	const struct ce_pipe_config *ce_config;
+
+	ce_config = &target_ce_config_wlan[ce_id];
+	if (__le32_to_cpu(ce_config->pipedir) & PIPEDIR_OUT)
+		ath11k_ahb_clearbit32(ab, ce_id, CE_HOST_IE_ADDRESS);
+
+	if (__le32_to_cpu(ce_config->pipedir) & PIPEDIR_IN) {
+		ath11k_ahb_clearbit32(ab, ce_id, CE_HOST_IE_2_ADDRESS);
+		ath11k_ahb_clearbit32(ab, ce_id + CE_HOST_IE_3_SHIFT,
+				      CE_HOST_IE_3_ADDRESS);
+	}
+}
+
+static void ath11k_ahb_sync_ce_irqs(struct ath11k_base *ab)
+{
+	int i;
+	int irq_idx;
+
+	for (i = 0; i < CE_COUNT; i++) {
+		if (ath11k_ce_get_attr_flags(i) & CE_ATTR_DIS_INTR)
+			continue;
+
+		irq_idx = ATH11K_IRQ_CE0_OFFSET + i;
+		synchronize_irq(ab->irq_num[irq_idx]);
+	}
+}
+
+static void ath11k_ahb_sync_ext_irqs(struct ath11k_base *ab)
+{
+	int i, j;
+	int irq_idx;
+
+	for (i = 0; i < ATH11K_EXT_IRQ_GRP_NUM_MAX; i++) {
+		struct ath11k_ext_irq_grp *irq_grp = &ab->ext_irq_grp[i];
+
+		for (j = 0; j < irq_grp->num_irq; j++) {
+			irq_idx = irq_grp->irqs[j];
+			synchronize_irq(ab->irq_num[irq_idx]);
+		}
+	}
+}
+
+static void ath11k_ahb_ce_irqs_enable(struct ath11k_base *ab)
+{
+	int i;
+
+	for (i = 0; i < CE_COUNT; i++) {
+		if (ath11k_ce_get_attr_flags(i) & CE_ATTR_DIS_INTR)
+			continue;
+		ath11k_ahb_ce_irq_enable(ab, i);
+	}
+}
+
+static void ath11k_ahb_ce_irqs_disable(struct ath11k_base *ab)
+{
+	int i;
+
+	for (i = 0; i < CE_COUNT; i++) {
+		if (ath11k_ce_get_attr_flags(i) & CE_ATTR_DIS_INTR)
+			continue;
+		ath11k_ahb_ce_irq_disable(ab, i);
+	}
+}
+
+int ath11k_ahb_start(struct ath11k_base *ab)
+{
+	ath11k_ahb_ce_irqs_enable(ab);
+	ath11k_ce_rx_post_buf(ab);
+
+	return 0;
+}
+
+void ath11k_ahb_ext_irq_enable(struct ath11k_base *ab)
+{
+	int i;
+
+	for (i = 0; i < ATH11K_EXT_IRQ_GRP_NUM_MAX; i++) {
+		struct ath11k_ext_irq_grp *irq_grp = &ab->ext_irq_grp[i];
+
+		napi_enable(&irq_grp->napi);
+		ath11k_ahb_ext_grp_enable(irq_grp);
+	}
+}
+
+void ath11k_ahb_ext_irq_disable(struct ath11k_base *ab)
+{
+	__ath11k_ahb_ext_irq_disable(ab);
+	ath11k_ahb_sync_ext_irqs(ab);
+}
+
+void ath11k_ahb_stop(struct ath11k_base *ab)
+{
+	if (!test_bit(ATH11K_FLAG_CRASH_FLUSH, &ab->dev_flags))
+		ath11k_ahb_ce_irqs_disable(ab);
+	ath11k_ahb_sync_ce_irqs(ab);
+	ath11k_ahb_kill_tasklets(ab);
+	del_timer_sync(&ab->rx_replenish_retry);
+	ath11k_ce_cleanup_pipes(ab);
+}
+
+int ath11k_ahb_power_up(struct ath11k_base *ab)
+{
+	int ret;
+
+	ret = rproc_boot(ab->tgt_rproc);
+	if (ret)
+		ath11k_err(ab, "failed to boot the remote processor Q6\n");
+
+	return ret;
+}
+
+void ath11k_ahb_power_down(struct ath11k_base *ab)
+{
+	rproc_shutdown(ab->tgt_rproc);
+}
+
+static void ath11k_ahb_init_qmi_ce_config(struct ath11k_base *ab)
+{
+	struct ath11k_qmi_ce_cfg *cfg = &ab->qmi.ce_cfg;
+
+	cfg->tgt_ce = (u8 *)target_ce_config_wlan;
+	cfg->tgt_ce_len = sizeof(target_ce_config_wlan);
+
+	cfg->svc_to_ce_map = (u8 *)target_service_to_ce_map_wlan;
+	cfg->svc_to_ce_map_len = sizeof(target_service_to_ce_map_wlan);
+}
+
+static void ath11k_ahb_free_ext_irq(struct ath11k_base *ab)
+{
+	int i, j;
+
+	for (i = 0; i < ATH11K_EXT_IRQ_GRP_NUM_MAX; i++) {
+		struct ath11k_ext_irq_grp *irq_grp = &ab->ext_irq_grp[i];
+
+		for (j = 0; j < irq_grp->num_irq; j++)
+			free_irq(ab->irq_num[irq_grp->irqs[j]], irq_grp);
+	}
+}
+
+static void ath11k_ahb_free_irq(struct ath11k_base *ab)
+{
+	int irq_idx;
+	int i;
+
+	for (i = 0; i < CE_COUNT; i++) {
+		if (ath11k_ce_get_attr_flags(i) & CE_ATTR_DIS_INTR)
+			continue;
+		irq_idx = ATH11K_IRQ_CE0_OFFSET + i;
+		free_irq(ab->irq_num[irq_idx], &ab->ce.ce_pipe[i]);
+	}
+
+	ath11k_ahb_free_ext_irq(ab);
+}
+
+static void ath11k_ahb_ce_tasklet(unsigned long data)
+{
+	struct ath11k_ce_pipe *ce_pipe = (struct ath11k_ce_pipe *)data;
+
+	ath11k_ce_per_engine_service(ce_pipe->ab, ce_pipe->pipe_num);
+
+	ath11k_ahb_ce_irq_enable(ce_pipe->ab, ce_pipe->pipe_num);
+}
+
+static irqreturn_t ath11k_ahb_ce_interrupt_handler(int irq, void *arg)
+{
+	struct ath11k_ce_pipe *ce_pipe = arg;
+
+	ath11k_ahb_ce_irq_disable(ce_pipe->ab, ce_pipe->pipe_num);
+
+	tasklet_schedule(&ce_pipe->intr_tq);
+
+	return IRQ_HANDLED;
+}
+
+static int ath11k_ahb_ext_grp_napi_poll(struct napi_struct *napi, int budget)
+{
+	struct ath11k_ext_irq_grp *irq_grp = container_of(napi,
+						struct ath11k_ext_irq_grp,
+						napi);
+	struct ath11k_base *ab = irq_grp->ab;
+	int work_done;
+
+	work_done = ath11k_dp_service_srng(ab, irq_grp, budget);
+	if (work_done < budget) {
+		napi_complete_done(napi, work_done);
+		ath11k_ahb_ext_grp_enable(irq_grp);
+	}
+
+	if (work_done > budget)
+		work_done = budget;
+
+	return work_done;
+}
+
+static irqreturn_t ath11k_ahb_ext_interrupt_handler(int irq, void *arg)
+{
+	struct ath11k_ext_irq_grp *irq_grp = arg;
+
+	ath11k_ahb_ext_grp_disable(irq_grp);
+
+	napi_schedule(&irq_grp->napi);
+
+	return IRQ_HANDLED;
+}
+
+static int ath11k_ahb_ext_irq_config(struct ath11k_base *ab)
+{
+	int i, j;
+	int irq;
+	int ret;
+
+	for (i = 0; i < ATH11K_EXT_IRQ_GRP_NUM_MAX; i++) {
+		struct ath11k_ext_irq_grp *irq_grp = &ab->ext_irq_grp[i];
+		u32 num_irq = 0;
+
+		irq_grp->ab = ab;
+		irq_grp->grp_id = i;
+		init_dummy_netdev(&irq_grp->napi_ndev);
+		netif_napi_add(&irq_grp->napi_ndev, &irq_grp->napi,
+			       ath11k_ahb_ext_grp_napi_poll, NAPI_POLL_WEIGHT);
+		__skb_queue_head_init(&irq_grp->pending_q);
+
+		for (j = 0; j < ATH11K_EXT_IRQ_NUM_MAX; j++) {
+			if (ath11k_tx_ring_mask[i] & BIT(j)) {
+				irq_grp->irqs[num_irq++] =
+					wbm2host_tx_completions_ring1 - j;
+			}
+
+			if (ath11k_rx_ring_mask[i] & BIT(j)) {
+				irq_grp->irqs[num_irq++] =
+					reo2host_destination_ring1 - j;
+			}
+
+			if (ath11k_rx_err_ring_mask[i] & BIT(j))
+				irq_grp->irqs[num_irq++] = reo2host_exception;
+
+			if (ath11k_rx_wbm_rel_ring_mask[i] & BIT(j))
+				irq_grp->irqs[num_irq++] = wbm2host_rx_release;
+
+			if (ath11k_reo_status_ring_mask[i] & BIT(j))
+				irq_grp->irqs[num_irq++] = reo2host_status;
+
+			if (j < MAX_RADIOS) {
+				if (ath11k_rxdma2host_ring_mask[i] & BIT(j)) {
+					irq_grp->irqs[num_irq++] =
+						rxdma2host_destination_ring_mac1
+						- ath11k_core_get_hw_mac_id(ab, j);
+				}
+
+				if (ath11k_host2rxdma_ring_mask[i] & BIT(j)) {
+					irq_grp->irqs[num_irq++] =
+						host2rxdma_host_buf_ring_mac1
+						- ath11k_core_get_hw_mac_id(ab, j);
+				}
+
+				if (rx_mon_status_ring_mask[i] & BIT(j)) {
+					irq_grp->irqs[num_irq++] =
+						ppdu_end_interrupts_mac1 -
+						ath11k_core_get_hw_mac_id(ab, j);
+					irq_grp->irqs[num_irq++] =
+						rxdma2host_monitor_status_ring_mac1 -
+						ath11k_core_get_hw_mac_id(ab, j);
+				}
+			}
+		}
+		irq_grp->num_irq = num_irq;
+
+		for (j = 0; j < irq_grp->num_irq; j++) {
+			int irq_idx = irq_grp->irqs[j];
+
+			irq = platform_get_irq_byname(ab->pdev,
+						      irq_name[irq_idx]);
+			ab->irq_num[irq_idx] = irq;
+			irq_set_status_flags(irq, IRQ_NOAUTOEN);
+			ret = request_irq(irq, ath11k_ahb_ext_interrupt_handler,
+					  IRQF_TRIGGER_RISING,
+					  irq_name[irq_idx], irq_grp);
+			if (ret) {
+				ath11k_err(ab, "failed request_irq for %d\n",
+					   irq);
+			}
+		}
+	}
+
+	return 0;
+}
+
+static int ath11k_ahb_config_irq(struct ath11k_base *ab)
+{
+	int irq, irq_idx, i;
+	int ret;
+
+	/* Configure CE irqs */
+	for (i = 0; i < CE_COUNT; i++) {
+		struct ath11k_ce_pipe *ce_pipe = &ab->ce.ce_pipe[i];
+
+		if (ath11k_ce_get_attr_flags(i) & CE_ATTR_DIS_INTR)
+			continue;
+
+		irq_idx = ATH11K_IRQ_CE0_OFFSET + i;
+
+		tasklet_init(&ce_pipe->intr_tq, ath11k_ahb_ce_tasklet,
+			     (unsigned long)ce_pipe);
+		irq = platform_get_irq_byname(ab->pdev, irq_name[irq_idx]);
+		ret = request_irq(irq, ath11k_ahb_ce_interrupt_handler,
+				  IRQF_TRIGGER_RISING, irq_name[irq_idx],
+				  ce_pipe);
+		if (ret)
+			return ret;
+
+		ab->irq_num[irq_idx] = irq;
+	}
+
+	/* Configure external interrupts */
+	ret = ath11k_ahb_ext_irq_config(ab);
+
+	return ret;
+}
+
+int ath11k_ahb_map_service_to_pipe(struct ath11k_base *ab, u16 service_id,
+				   u8 *ul_pipe, u8 *dl_pipe)
+{
+	const struct service_to_pipe *entry;
+	bool ul_set = false, dl_set = false;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(target_service_to_ce_map_wlan); i++) {
+		entry = &target_service_to_ce_map_wlan[i];
+
+		if (__le32_to_cpu(entry->service_id) != service_id)
+			continue;
+
+		switch (__le32_to_cpu(entry->pipedir)) {
+		case PIPEDIR_NONE:
+			break;
+		case PIPEDIR_IN:
+			WARN_ON(dl_set);
+			*dl_pipe = __le32_to_cpu(entry->pipenum);
+			dl_set = true;
+			break;
+		case PIPEDIR_OUT:
+			WARN_ON(ul_set);
+			*ul_pipe = __le32_to_cpu(entry->pipenum);
+			ul_set = true;
+			break;
+		case PIPEDIR_INOUT:
+			WARN_ON(dl_set);
+			WARN_ON(ul_set);
+			*dl_pipe = __le32_to_cpu(entry->pipenum);
+			*ul_pipe = __le32_to_cpu(entry->pipenum);
+			dl_set = true;
+			ul_set = true;
+			break;
+		}
+	}
+
+	if (WARN_ON(!ul_set || !dl_set))
+		return -ENOENT;
+
+	return 0;
+}
+
+static int ath11k_ahb_probe(struct platform_device *pdev)
+{
+	struct ath11k_base *ab;
+	const struct of_device_id *of_id;
+	struct resource *mem_res;
+	void __iomem *mem;
+	int ret;
+
+	of_id = of_match_device(ath11k_ahb_of_match, &pdev->dev);
+	if (!of_id) {
+		dev_err(&pdev->dev, "failed to find matching device tree id\n");
+		return -EINVAL;
+	}
+
+	mem_res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!mem_res) {
+		dev_err(&pdev->dev, "failed to get IO memory resource\n");
+		return -ENXIO;
+	}
+
+	mem = devm_ioremap_resource(&pdev->dev, mem_res);
+	if (IS_ERR(mem)) {
+		dev_err(&pdev->dev, "ioremap error\n");
+		return PTR_ERR(mem);
+	}
+
+	ret = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
+	if (ret) {
+		dev_err(&pdev->dev, "failed to set 32-bit consistent dma\n");
+		return ret;
+	}
+
+	ab = ath11k_core_alloc(&pdev->dev);
+	if (!ab) {
+		dev_err(&pdev->dev, "failed to allocate ath11k base\n");
+		return -ENOMEM;
+	}
+
+	ab->pdev = pdev;
+	ab->hw_rev = (enum ath11k_hw_rev)of_id->data;
+	ab->mem = mem;
+	ab->mem_len = resource_size(mem_res);
+	platform_set_drvdata(pdev, ab);
+
+	ret = ath11k_hal_srng_init(ab);
+	if (ret)
+		goto err_core_free;
+
+	ret = ath11k_ce_alloc_pipes(ab);
+	if (ret) {
+		ath11k_err(ab, "failed to allocate ce pipes: %d\n", ret);
+		goto err_hal_srng_deinit;
+	}
+
+	ath11k_ahb_init_qmi_ce_config(ab);
+
+	ret = ath11k_ahb_config_irq(ab);
+	if (ret) {
+		ath11k_err(ab, "failed to configure irq: %d\n", ret);
+		goto err_ce_free;
+	}
+
+	ret = ath11k_core_init(ab);
+	if (ret) {
+		ath11k_err(ab, "failed to init core: %d\n", ret);
+		goto err_ce_free;
+	}
+
+	return 0;
+
+err_ce_free:
+	ath11k_ce_free_pipes(ab);
+
+err_hal_srng_deinit:
+	ath11k_hal_srng_deinit(ab);
+
+err_core_free:
+	ath11k_core_free(ab);
+	platform_set_drvdata(pdev, NULL);
+
+	return ret;
+}
+
+static int ath11k_ahb_remove(struct platform_device *pdev)
+{
+	struct ath11k_base *ab = platform_get_drvdata(pdev);
+
+	reinit_completion(&ab->driver_recovery);
+
+	if (test_bit(ATH11K_FLAG_RECOVERY, &ab->dev_flags))
+		wait_for_completion_timeout(&ab->driver_recovery,
+					    ATH11K_AHB_RECOVERY_TIMEOUT);
+
+	set_bit(ATH11K_FLAG_UNREGISTERING, &ab->dev_flags);
+	cancel_work_sync(&ab->restart_work);
+
+	ath11k_core_deinit(ab);
+	ath11k_ahb_free_irq(ab);
+
+	ath11k_hal_srng_deinit(ab);
+	ath11k_ce_free_pipes(ab);
+	ath11k_core_free(ab);
+	platform_set_drvdata(pdev, NULL);
+
+	return 0;
+}
+
+static struct platform_driver ath11k_ahb_driver = {
+	.driver         = {
+		.name   = "ath11k",
+		.of_match_table = ath11k_ahb_of_match,
+	},
+	.probe  = ath11k_ahb_probe,
+	.remove = ath11k_ahb_remove,
+};
+
+int ath11k_ahb_init(void)
+{
+	return platform_driver_register(&ath11k_ahb_driver);
+}
+
+void ath11k_ahb_exit(void)
+{
+	platform_driver_unregister(&ath11k_ahb_driver);
+}
diff --git a/drivers/net/wireless/ath/ath11k/ahb.h b/drivers/net/wireless/ath/ath11k/ahb.h
new file mode 100644
index 000000000000..93f46dfe22df
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/ahb.h
@@ -0,0 +1,35 @@
+/* SPDX-License-Identifier: BSD-3-Clause-Clear */
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+#ifndef ATH11K_AHB_H
+#define ATH11K_AHB_H
+
+#include "core.h"
+
+#define ATH11K_AHB_RECOVERY_TIMEOUT (3 * HZ)
+struct ath11k_base;
+
+static inline u32 ath11k_ahb_read32(struct ath11k_base *ab, u32 offset)
+{
+	return ioread32(ab->mem + offset);
+}
+
+static inline void ath11k_ahb_write32(struct ath11k_base *ab, u32 offset, u32 value)
+{
+	iowrite32(value, ab->mem + offset);
+}
+
+void ath11k_ahb_ext_irq_enable(struct ath11k_base *ab);
+void ath11k_ahb_ext_irq_disable(struct ath11k_base *ab);
+int ath11k_ahb_start(struct ath11k_base *ab);
+void ath11k_ahb_stop(struct ath11k_base *ab);
+int ath11k_ahb_power_up(struct ath11k_base *ab);
+void ath11k_ahb_power_down(struct ath11k_base *ab);
+int ath11k_ahb_map_service_to_pipe(struct ath11k_base *ab, u16 service_id,
+				   u8 *ul_pipe, u8 *dl_pipe);
+
+int ath11k_ahb_init(void);
+void ath11k_ahb_exit(void);
+
+#endif
diff --git a/drivers/net/wireless/ath/ath11k/ce.c b/drivers/net/wireless/ath/ath11k/ce.c
new file mode 100644
index 000000000000..a6572b414303
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/ce.c
@@ -0,0 +1,805 @@
+// SPDX-License-Identifier: BSD-3-Clause-Clear
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#include "dp_rx.h"
+#include "debug.h"
+
+static const struct ce_attr host_ce_config_wlan[] = {
+	/* CE0: host->target HTC control and raw streams */
+	{
+		.flags = CE_ATTR_FLAGS,
+		.src_nentries = 16,
+		.src_sz_max = 2048,
+		.dest_nentries = 0,
+	},
+
+	/* CE1: target->host HTT + HTC control */
+	{
+		.flags = CE_ATTR_FLAGS,
+		.src_nentries = 0,
+		.src_sz_max = 2048,
+		.dest_nentries = 512,
+		.recv_cb = ath11k_htc_rx_completion_handler,
+	},
+
+	/* CE2: target->host WMI */
+	{
+		.flags = CE_ATTR_FLAGS,
+		.src_nentries = 0,
+		.src_sz_max = 2048,
+		.dest_nentries = 512,
+		.recv_cb = ath11k_htc_rx_completion_handler,
+	},
+
+	/* CE3: host->target WMI (mac0) */
+	{
+		.flags = CE_ATTR_FLAGS,
+		.src_nentries = 32,
+		.src_sz_max = 2048,
+		.dest_nentries = 0,
+	},
+
+	/* CE4: host->target HTT */
+	{
+		.flags = CE_ATTR_FLAGS | CE_ATTR_DIS_INTR,
+		.src_nentries = 2048,
+		.src_sz_max = 256,
+		.dest_nentries = 0,
+	},
+
+	/* CE5: target->host pktlog */
+	{
+		.flags = CE_ATTR_FLAGS,
+		.src_nentries = 0,
+		.src_sz_max = 2048,
+		.dest_nentries = 512,
+		.recv_cb = ath11k_dp_htt_htc_t2h_msg_handler,
+	},
+
+	/* CE6: target autonomous hif_memcpy */
+	{
+		.flags = CE_ATTR_FLAGS | CE_ATTR_DIS_INTR,
+		.src_nentries = 0,
+		.src_sz_max = 0,
+		.dest_nentries = 0,
+	},
+
+	/* CE7: host->target WMI (mac1) */
+	{
+		.flags = CE_ATTR_FLAGS,
+		.src_nentries = 32,
+		.src_sz_max = 2048,
+		.dest_nentries = 0,
+	},
+
+	/* CE8: target autonomous hif_memcpy */
+	{
+		.flags = CE_ATTR_FLAGS,
+		.src_nentries = 0,
+		.src_sz_max = 0,
+		.dest_nentries = 0,
+	},
+
+	/* CE9: host->target WMI (mac2) */
+	{
+		.flags = CE_ATTR_FLAGS,
+		.src_nentries = 32,
+		.src_sz_max = 2048,
+		.dest_nentries = 0,
+	},
+
+	/* CE10: target->host HTT */
+	{
+		.flags = CE_ATTR_FLAGS,
+		.src_nentries = 0,
+		.src_sz_max = 2048,
+		.dest_nentries = 512,
+		.recv_cb = ath11k_htc_rx_completion_handler,
+	},
+
+	/* CE11: Not used */
+	{
+		.flags = CE_ATTR_FLAGS,
+		.src_nentries = 0,
+		.src_sz_max = 0,
+		.dest_nentries = 0,
+	},
+};
+
+static int ath11k_ce_rx_buf_enqueue_pipe(struct ath11k_ce_pipe *pipe,
+					 struct sk_buff *skb, dma_addr_t paddr)
+{
+	struct ath11k_base *ab = pipe->ab;
+	struct ath11k_ce_ring *ring = pipe->dest_ring;
+	struct hal_srng *srng;
+	unsigned int write_index;
+	unsigned int nentries_mask = ring->nentries_mask;
+	u32 *desc;
+	int ret;
+
+	lockdep_assert_held(&ab->ce.ce_lock);
+
+	write_index = ring->write_index;
+
+	srng = &ab->hal.srng_list[ring->hal_ring_id];
+
+	spin_lock_bh(&srng->lock);
+
+	ath11k_hal_srng_access_begin(ab, srng);
+
+	if (unlikely(ath11k_hal_srng_src_num_free(ab, srng, false) < 1)) {
+		ret = -ENOSPC;
+		goto exit;
+	}
+
+	desc = ath11k_hal_srng_src_get_next_entry(ab, srng);
+	if (!desc) {
+		ret = -ENOSPC;
+		goto exit;
+	}
+
+	ath11k_hal_ce_dst_set_desc(desc, paddr);
+
+	ring->skb[write_index] = skb;
+	write_index = CE_RING_IDX_INCR(nentries_mask, write_index);
+	ring->write_index = write_index;
+
+	pipe->rx_buf_needed--;
+
+	ret = 0;
+exit:
+	ath11k_hal_srng_access_end(ab, srng);
+
+	spin_unlock_bh(&srng->lock);
+
+	return ret;
+}
+
+static int ath11k_ce_rx_post_pipe(struct ath11k_ce_pipe *pipe)
+{
+	struct ath11k_base *ab = pipe->ab;
+	struct sk_buff *skb;
+	dma_addr_t paddr;
+	int ret = 0;
+
+	if (!(pipe->dest_ring || pipe->status_ring))
+		return 0;
+
+	spin_lock_bh(&ab->ce.ce_lock);
+	while (pipe->rx_buf_needed) {
+		skb = dev_alloc_skb(pipe->buf_sz);
+		if (!skb) {
+			ret = -ENOMEM;
+			goto exit;
+		}
+
+		WARN_ON_ONCE(!IS_ALIGNED((unsigned long)skb->data, 4));
+
+		paddr = dma_map_single(ab->dev, skb->data,
+				       skb->len + skb_tailroom(skb),
+				       DMA_FROM_DEVICE);
+		if (unlikely(dma_mapping_error(ab->dev, paddr))) {
+			ath11k_warn(ab, "failed to dma map ce rx buf\n");
+			dev_kfree_skb_any(skb);
+			ret = -EIO;
+			goto exit;
+		}
+
+		ATH11K_SKB_RXCB(skb)->paddr = paddr;
+
+		ret = ath11k_ce_rx_buf_enqueue_pipe(pipe, skb, paddr);
+
+		if (ret) {
+			ath11k_warn(ab, "failed to enqueue rx buf: %d\n", ret);
+			dma_unmap_single(ab->dev, paddr,
+					 skb->len + skb_tailroom(skb),
+					 DMA_FROM_DEVICE);
+			dev_kfree_skb_any(skb);
+			goto exit;
+		}
+	}
+
+exit:
+	spin_unlock_bh(&ab->ce.ce_lock);
+	return ret;
+}
+
+static int ath11k_ce_completed_recv_next(struct ath11k_ce_pipe *pipe,
+					 struct sk_buff **skb, int *nbytes)
+{
+	struct ath11k_base *ab = pipe->ab;
+	struct hal_srng *srng;
+	unsigned int sw_index;
+	unsigned int nentries_mask;
+	u32 *desc;
+	int ret = 0;
+
+	spin_lock_bh(&ab->ce.ce_lock);
+
+	sw_index = pipe->dest_ring->sw_index;
+	nentries_mask = pipe->dest_ring->nentries_mask;
+
+	srng = &ab->hal.srng_list[pipe->status_ring->hal_ring_id];
+
+	spin_lock_bh(&srng->lock);
+
+	ath11k_hal_srng_access_begin(ab, srng);
+
+	desc = ath11k_hal_srng_dst_get_next_entry(ab, srng);
+	if (!desc) {
+		ret = -EIO;
+		goto err;
+	}
+
+	*nbytes = ath11k_hal_ce_dst_status_get_length(desc);
+	if (*nbytes == 0) {
+		ret = -EIO;
+		goto err;
+	}
+
+	*skb = pipe->dest_ring->skb[sw_index];
+	pipe->dest_ring->skb[sw_index] = NULL;
+
+	sw_index = CE_RING_IDX_INCR(nentries_mask, sw_index);
+	pipe->dest_ring->sw_index = sw_index;
+
+	pipe->rx_buf_needed++;
+err:
+	ath11k_hal_srng_access_end(ab, srng);
+
+	spin_unlock_bh(&srng->lock);
+
+	spin_unlock_bh(&ab->ce.ce_lock);
+
+	return ret;
+}
+
+static void ath11k_ce_recv_process_cb(struct ath11k_ce_pipe *pipe)
+{
+	struct ath11k_base *ab = pipe->ab;
+	struct sk_buff *skb;
+	struct sk_buff_head list;
+	unsigned int nbytes, max_nbytes;
+	int ret;
+
+	__skb_queue_head_init(&list);
+	while (ath11k_ce_completed_recv_next(pipe, &skb, &nbytes) == 0) {
+		max_nbytes = skb->len + skb_tailroom(skb);
+		dma_unmap_single(ab->dev, ATH11K_SKB_RXCB(skb)->paddr,
+				 max_nbytes, DMA_FROM_DEVICE);
+
+		if (unlikely(max_nbytes < nbytes)) {
+			ath11k_warn(ab, "rxed more than expected (nbytes %d, max %d)",
+				    nbytes, max_nbytes);
+			dev_kfree_skb_any(skb);
+			continue;
+		}
+
+		skb_put(skb, nbytes);
+		__skb_queue_tail(&list, skb);
+	}
+
+	while ((skb = __skb_dequeue(&list))) {
+		ath11k_dbg(ab, ATH11K_DBG_AHB, "rx ce pipe %d len %d\n",
+			   pipe->pipe_num, skb->len);
+		pipe->recv_cb(ab, skb);
+	}
+
+	ret = ath11k_ce_rx_post_pipe(pipe);
+	if (ret && ret != -ENOSPC) {
+		ath11k_warn(ab, "failed to post rx buf to pipe: %d err: %d\n",
+			    pipe->pipe_num, ret);
+		mod_timer(&ab->rx_replenish_retry,
+			  jiffies + ATH11K_CE_RX_POST_RETRY_JIFFIES);
+	}
+}
+
+static struct sk_buff *ath11k_ce_completed_send_next(struct ath11k_ce_pipe *pipe)
+{
+	struct ath11k_base *ab = pipe->ab;
+	struct hal_srng *srng;
+	unsigned int sw_index;
+	unsigned int nentries_mask;
+	struct sk_buff *skb;
+	u32 *desc;
+
+	spin_lock_bh(&ab->ce.ce_lock);
+
+	sw_index = pipe->src_ring->sw_index;
+	nentries_mask = pipe->src_ring->nentries_mask;
+
+	srng = &ab->hal.srng_list[pipe->src_ring->hal_ring_id];
+
+	spin_lock_bh(&srng->lock);
+
+	ath11k_hal_srng_access_begin(ab, srng);
+
+	desc = ath11k_hal_srng_src_reap_next(ab, srng);
+	if (!desc) {
+		skb = ERR_PTR(-EIO);
+		goto err_unlock;
+	}
+
+	skb = pipe->src_ring->skb[sw_index];
+
+	pipe->src_ring->skb[sw_index] = NULL;
+
+	sw_index = CE_RING_IDX_INCR(nentries_mask, sw_index);
+	pipe->src_ring->sw_index = sw_index;
+
+err_unlock:
+	spin_unlock_bh(&srng->lock);
+
+	spin_unlock_bh(&ab->ce.ce_lock);
+
+	return skb;
+}
+
+static void ath11k_ce_send_done_cb(struct ath11k_ce_pipe *pipe)
+{
+	struct ath11k_base *ab = pipe->ab;
+	struct sk_buff *skb;
+
+	while (!IS_ERR(skb = ath11k_ce_completed_send_next(pipe))) {
+		if (!skb)
+			continue;
+
+		dma_unmap_single(ab->dev, ATH11K_SKB_CB(skb)->paddr, skb->len,
+				 DMA_TO_DEVICE);
+		dev_kfree_skb_any(skb);
+	}
+}
+
+static int ath11k_ce_init_ring(struct ath11k_base *ab,
+			       struct ath11k_ce_ring *ce_ring,
+			       int ce_id, enum hal_ring_type type)
+{
+	struct hal_srng_params params = { 0 };
+	int ret;
+
+	params.ring_base_paddr = ce_ring->base_addr_ce_space;
+	params.ring_base_vaddr = ce_ring->base_addr_owner_space;
+	params.num_entries = ce_ring->nentries;
+
+	switch (type) {
+	case HAL_CE_SRC:
+		if (!(CE_ATTR_DIS_INTR & host_ce_config_wlan[ce_id].flags))
+			params.intr_batch_cntr_thres_entries = 1;
+		break;
+	case HAL_CE_DST:
+		params.max_buffer_len = host_ce_config_wlan[ce_id].src_sz_max;
+		if (!(host_ce_config_wlan[ce_id].flags & CE_ATTR_DIS_INTR)) {
+			params.intr_timer_thres_us = 1024;
+			params.flags |= HAL_SRNG_FLAGS_LOW_THRESH_INTR_EN;
+			params.low_threshold = ce_ring->nentries - 3;
+		}
+		break;
+	case HAL_CE_DST_STATUS:
+		if (!(host_ce_config_wlan[ce_id].flags & CE_ATTR_DIS_INTR)) {
+			params.intr_batch_cntr_thres_entries = 1;
+			params.intr_timer_thres_us = 0x1000;
+		}
+		break;
+	default:
+		ath11k_warn(ab, "Invalid CE ring type %d\n", type);
+		return -EINVAL;
+	}
+
+	/* TODO: Init other params needed by HAL to init the ring */
+
+	ret = ath11k_hal_srng_setup(ab, type, ce_id, 0, &params);
+	if (ret < 0) {
+		ath11k_warn(ab, "failed to setup srng: %d ring_id %d\n",
+			    ret, ce_id);
+		return ret;
+	}
+	ce_ring->hal_ring_id = ret;
+
+	return 0;
+}
+
+static struct ath11k_ce_ring *
+ath11k_ce_alloc_ring(struct ath11k_base *ab, int nentries, int desc_sz)
+{
+	struct ath11k_ce_ring *ce_ring;
+	dma_addr_t base_addr;
+
+	ce_ring = kzalloc(struct_size(ce_ring, skb, nentries), GFP_KERNEL);
+	if (ce_ring == NULL)
+		return ERR_PTR(-ENOMEM);
+
+	ce_ring->nentries = nentries;
+	ce_ring->nentries_mask = nentries - 1;
+
+	/* Legacy platforms that do not support cache
+	 * coherent DMA are unsupported
+	 */
+	ce_ring->base_addr_owner_space_unaligned =
+		dma_alloc_coherent(ab->dev,
+				   nentries * desc_sz + CE_DESC_RING_ALIGN,
+				   &base_addr, GFP_KERNEL);
+	if (!ce_ring->base_addr_owner_space_unaligned) {
+		kfree(ce_ring);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	ce_ring->base_addr_ce_space_unaligned = base_addr;
+
+	ce_ring->base_addr_owner_space = PTR_ALIGN(
+			ce_ring->base_addr_owner_space_unaligned,
+			CE_DESC_RING_ALIGN);
+	ce_ring->base_addr_ce_space = ALIGN(
+			ce_ring->base_addr_ce_space_unaligned,
+			CE_DESC_RING_ALIGN);
+
+	return ce_ring;
+}
+
+static int ath11k_ce_alloc_pipe(struct ath11k_base *ab, int ce_id)
+{
+	struct ath11k_ce_pipe *pipe = &ab->ce.ce_pipe[ce_id];
+	const struct ce_attr *attr = &host_ce_config_wlan[ce_id];
+	int nentries;
+	int desc_sz;
+
+	pipe->attr_flags = attr->flags;
+
+	if (attr->src_nentries) {
+		pipe->send_cb = ath11k_ce_send_done_cb;
+		nentries = roundup_pow_of_two(attr->src_nentries);
+		desc_sz = ath11k_hal_ce_get_desc_size(HAL_CE_DESC_SRC);
+		pipe->src_ring = ath11k_ce_alloc_ring(ab, nentries, desc_sz);
+		if (!pipe->src_ring)
+			return -ENOMEM;
+	}
+
+	if (attr->dest_nentries) {
+		pipe->recv_cb = attr->recv_cb;
+		nentries = roundup_pow_of_two(attr->dest_nentries);
+		desc_sz = ath11k_hal_ce_get_desc_size(HAL_CE_DESC_DST);
+		pipe->dest_ring = ath11k_ce_alloc_ring(ab, nentries, desc_sz);
+
+		if (!pipe->dest_ring)
+			return -ENOMEM;
+
+		desc_sz = ath11k_hal_ce_get_desc_size(HAL_CE_DESC_DST_STATUS);
+		pipe->status_ring = ath11k_ce_alloc_ring(ab, nentries, desc_sz);
+		if (!pipe->status_ring)
+			return -ENOMEM;
+	}
+
+	return 0;
+}
+
+void ath11k_ce_per_engine_service(struct ath11k_base *ab, u16 ce_id)
+{
+	struct ath11k_ce_pipe *pipe = &ab->ce.ce_pipe[ce_id];
+
+	if (pipe->send_cb)
+		pipe->send_cb(pipe);
+
+	if (pipe->recv_cb)
+		ath11k_ce_recv_process_cb(pipe);
+}
+
+void ath11k_ce_poll_send_completed(struct ath11k_base *ab, u8 pipe_id)
+{
+	struct ath11k_ce_pipe *pipe = &ab->ce.ce_pipe[pipe_id];
+
+	if ((pipe->attr_flags & CE_ATTR_DIS_INTR) && pipe->send_cb)
+		pipe->send_cb(pipe);
+}
+
+int ath11k_ce_send(struct ath11k_base *ab, struct sk_buff *skb, u8 pipe_id,
+		   u16 transfer_id)
+{
+	struct ath11k_ce_pipe *pipe = &ab->ce.ce_pipe[pipe_id];
+	struct hal_srng *srng;
+	u32 *desc;
+	unsigned int write_index, sw_index;
+	unsigned int nentries_mask;
+	int ret = 0;
+	u8 byte_swap_data = 0;
+	int num_used;
+
+	/* Check if some entries could be regained by handling tx completion if
+	 * the CE has interrupts disabled and the used entries is more than the
+	 * defined usage threshold.
+	 */
+	if (pipe->attr_flags & CE_ATTR_DIS_INTR) {
+		spin_lock_bh(&ab->ce.ce_lock);
+		write_index = pipe->src_ring->write_index;
+
+		sw_index = pipe->src_ring->sw_index;
+
+		if (write_index >= sw_index)
+			num_used = write_index - sw_index;
+		else
+			num_used = pipe->src_ring->nentries - sw_index +
+				   write_index;
+
+		spin_unlock_bh(&ab->ce.ce_lock);
+
+		if (num_used > ATH11K_CE_USAGE_THRESHOLD)
+			ath11k_ce_poll_send_completed(ab, pipe->pipe_num);
+	}
+
+	if (test_bit(ATH11K_FLAG_CRASH_FLUSH, &ab->dev_flags))
+		return -ESHUTDOWN;
+
+	spin_lock_bh(&ab->ce.ce_lock);
+
+	write_index = pipe->src_ring->write_index;
+	nentries_mask = pipe->src_ring->nentries_mask;
+
+	srng = &ab->hal.srng_list[pipe->src_ring->hal_ring_id];
+
+	spin_lock_bh(&srng->lock);
+
+	ath11k_hal_srng_access_begin(ab, srng);
+
+	if (unlikely(ath11k_hal_srng_src_num_free(ab, srng, false) < 1)) {
+		ath11k_hal_srng_access_end(ab, srng);
+		ret = -ENOBUFS;
+		goto err_unlock;
+	}
+
+	desc = ath11k_hal_srng_src_get_next_reaped(ab, srng);
+	if (!desc) {
+		ath11k_hal_srng_access_end(ab, srng);
+		ret = -ENOBUFS;
+		goto err_unlock;
+	}
+
+	if (pipe->attr_flags & CE_ATTR_BYTE_SWAP_DATA)
+		byte_swap_data = 1;
+
+	ath11k_hal_ce_src_set_desc(desc, ATH11K_SKB_CB(skb)->paddr,
+				   skb->len, transfer_id, byte_swap_data);
+
+	pipe->src_ring->skb[write_index] = skb;
+	pipe->src_ring->write_index = CE_RING_IDX_INCR(nentries_mask,
+						       write_index);
+
+	ath11k_hal_srng_access_end(ab, srng);
+
+	spin_unlock_bh(&srng->lock);
+
+	spin_unlock_bh(&ab->ce.ce_lock);
+
+	return 0;
+
+err_unlock:
+	spin_unlock_bh(&srng->lock);
+
+	spin_unlock_bh(&ab->ce.ce_lock);
+
+	return ret;
+}
+
+static void ath11k_ce_rx_pipe_cleanup(struct ath11k_ce_pipe *pipe)
+{
+	struct ath11k_base *ab = pipe->ab;
+	struct ath11k_ce_ring *ring = pipe->dest_ring;
+	struct sk_buff *skb;
+	int i;
+
+	if (!(ring && pipe->buf_sz))
+		return;
+
+	for (i = 0; i < ring->nentries; i++) {
+		skb = ring->skb[i];
+		if (!skb)
+			continue;
+
+		ring->skb[i] = NULL;
+		dma_unmap_single(ab->dev, ATH11K_SKB_RXCB(skb)->paddr,
+				 skb->len + skb_tailroom(skb), DMA_FROM_DEVICE);
+		dev_kfree_skb_any(skb);
+	}
+}
+
+void ath11k_ce_cleanup_pipes(struct ath11k_base *ab)
+{
+	struct ath11k_ce_pipe *pipe;
+	int pipe_num;
+
+	for (pipe_num = 0; pipe_num < CE_COUNT; pipe_num++) {
+		pipe = &ab->ce.ce_pipe[pipe_num];
+		ath11k_ce_rx_pipe_cleanup(pipe);
+
+		/* Cleanup any src CE's which have interrupts disabled */
+		ath11k_ce_poll_send_completed(ab, pipe_num);
+
+		/* NOTE: Should we also clean up tx buffer in all pipes? */
+	}
+}
+
+void ath11k_ce_rx_post_buf(struct ath11k_base *ab)
+{
+	struct ath11k_ce_pipe *pipe;
+	int i;
+	int ret;
+
+	for (i = 0; i < CE_COUNT; i++) {
+		pipe = &ab->ce.ce_pipe[i];
+		ret = ath11k_ce_rx_post_pipe(pipe);
+		if (ret) {
+			if (ret == -ENOSPC)
+				continue;
+
+			ath11k_warn(ab, "failed to post rx buf to pipe: %d err: %d\n",
+				    i, ret);
+			mod_timer(&ab->rx_replenish_retry,
+				  jiffies + ATH11K_CE_RX_POST_RETRY_JIFFIES);
+
+			return;
+		}
+	}
+}
+
+void ath11k_ce_rx_replenish_retry(struct timer_list *t)
+{
+	struct ath11k_base *ab = from_timer(ab, t, rx_replenish_retry);
+
+	ath11k_ce_rx_post_buf(ab);
+}
+
+int ath11k_ce_init_pipes(struct ath11k_base *ab)
+{
+	struct ath11k_ce_pipe *pipe;
+	int i;
+	int ret;
+
+	for (i = 0; i < CE_COUNT; i++) {
+		pipe = &ab->ce.ce_pipe[i];
+
+		if (pipe->src_ring) {
+			ret = ath11k_ce_init_ring(ab, pipe->src_ring, i,
+						  HAL_CE_SRC);
+			if (ret) {
+				ath11k_warn(ab, "failed to init src ring: %d\n",
+					    ret);
+				/* Should we clear any partial init */
+				return ret;
+			}
+
+			pipe->src_ring->write_index = 0;
+			pipe->src_ring->sw_index = 0;
+		}
+
+		if (pipe->dest_ring) {
+			ret = ath11k_ce_init_ring(ab, pipe->dest_ring, i,
+						  HAL_CE_DST);
+			if (ret) {
+				ath11k_warn(ab, "failed to init dest ring: %d\n",
+					    ret);
+				/* Should we clear any partial init */
+				return ret;
+			}
+
+			pipe->rx_buf_needed = pipe->dest_ring->nentries ?
+					      pipe->dest_ring->nentries - 2 : 0;
+
+			pipe->dest_ring->write_index = 0;
+			pipe->dest_ring->sw_index = 0;
+		}
+
+		if (pipe->status_ring) {
+			ret = ath11k_ce_init_ring(ab, pipe->status_ring, i,
+						  HAL_CE_DST_STATUS);
+			if (ret) {
+				ath11k_warn(ab, "failed to init dest status ing: %d\n",
+					    ret);
+				/* Should we clear any partial init */
+				return ret;
+			}
+
+			pipe->status_ring->write_index = 0;
+			pipe->status_ring->sw_index = 0;
+		}
+	}
+
+	return 0;
+}
+
+void ath11k_ce_free_pipes(struct ath11k_base *ab)
+{
+	struct ath11k_ce_pipe *pipe;
+	int desc_sz;
+	int i;
+
+	for (i = 0; i < CE_COUNT; i++) {
+		pipe = &ab->ce.ce_pipe[i];
+
+		if (pipe->src_ring) {
+			desc_sz = ath11k_hal_ce_get_desc_size(HAL_CE_DESC_SRC);
+			dma_free_coherent(ab->dev,
+					  pipe->src_ring->nentries * desc_sz +
+					  CE_DESC_RING_ALIGN,
+					  pipe->src_ring->base_addr_owner_space,
+					  pipe->src_ring->base_addr_ce_space);
+			kfree(pipe->src_ring);
+			pipe->src_ring = NULL;
+		}
+
+		if (pipe->dest_ring) {
+			desc_sz = ath11k_hal_ce_get_desc_size(HAL_CE_DESC_DST);
+			dma_free_coherent(ab->dev,
+					  pipe->dest_ring->nentries * desc_sz +
+					  CE_DESC_RING_ALIGN,
+					  pipe->dest_ring->base_addr_owner_space,
+					  pipe->dest_ring->base_addr_ce_space);
+			kfree(pipe->dest_ring);
+			pipe->dest_ring = NULL;
+		}
+
+		if (pipe->status_ring) {
+			desc_sz =
+			  ath11k_hal_ce_get_desc_size(HAL_CE_DESC_DST_STATUS);
+			dma_free_coherent(ab->dev,
+					  pipe->status_ring->nentries * desc_sz +
+					  CE_DESC_RING_ALIGN,
+					  pipe->status_ring->base_addr_owner_space,
+					  pipe->status_ring->base_addr_ce_space);
+			kfree(pipe->status_ring);
+			pipe->status_ring = NULL;
+		}
+	}
+}
+
+int ath11k_ce_alloc_pipes(struct ath11k_base *ab)
+{
+	struct ath11k_ce_pipe *pipe;
+	int i;
+	int ret;
+	const struct ce_attr *attr;
+
+	spin_lock_init(&ab->ce.ce_lock);
+
+	for (i = 0; i < CE_COUNT; i++) {
+		attr = &host_ce_config_wlan[i];
+		pipe = &ab->ce.ce_pipe[i];
+		pipe->pipe_num = i;
+		pipe->ab = ab;
+		pipe->buf_sz = attr->src_sz_max;
+
+		ret = ath11k_ce_alloc_pipe(ab, i);
+		if (ret) {
+			/* Free any parial successful allocation */
+			ath11k_ce_free_pipes(ab);
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
+/* For Big Endian Host, Copy Engine byte_swap is enabled
+ * When Copy Engine does byte_swap, need to byte swap again for the
+ * Host to get/put buffer content in the correct byte order
+ */
+void ath11k_ce_byte_swap(void *mem, u32 len)
+{
+	int i;
+
+	if (IS_ENABLED(CONFIG_CPU_BIG_ENDIAN)) {
+		if (!mem)
+			return;
+
+		for (i = 0; i < (len / 4); i++) {
+			*(u32 *)mem = swab32(*(u32 *)mem);
+			mem += 4;
+		}
+	}
+}
+
+int ath11k_ce_get_attr_flags(int ce_id)
+{
+	if (ce_id >= CE_COUNT)
+		return -EINVAL;
+
+	return host_ce_config_wlan[ce_id].flags;
+}
diff --git a/drivers/net/wireless/ath/ath11k/ce.h b/drivers/net/wireless/ath/ath11k/ce.h
new file mode 100644
index 000000000000..6174eac84410
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/ce.h
@@ -0,0 +1,178 @@
+/* SPDX-License-Identifier: BSD-3-Clause-Clear */
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#ifndef ATH11K_CE_H
+#define ATH11K_CE_H
+
+#define CE_COUNT 12
+
+/* Byte swap data words */
+#define CE_ATTR_BYTE_SWAP_DATA 2
+
+/* no interrupt on copy completion */
+#define CE_ATTR_DIS_INTR		8
+
+/* Host software's Copy Engine configuration. */
+#ifdef __BIG_ENDIAN
+#define CE_ATTR_FLAGS CE_ATTR_BYTE_SWAP_DATA
+#else
+#define CE_ATTR_FLAGS 0
+#endif
+
+/* Threshold to poll for tx completion in case of Interrupt disabled CE's */
+#define ATH11K_CE_USAGE_THRESHOLD 32
+
+void ath11k_ce_byte_swap(void *mem, u32 len);
+
+/*
+ * Directions for interconnect pipe configuration.
+ * These definitions may be used during configuration and are shared
+ * between Host and Target.
+ *
+ * Pipe Directions are relative to the Host, so PIPEDIR_IN means
+ * "coming IN over air through Target to Host" as with a WiFi Rx operation.
+ * Conversely, PIPEDIR_OUT means "going OUT from Host through Target over air"
+ * as with a WiFi Tx operation. This is somewhat awkward for the "middle-man"
+ * Target since things that are "PIPEDIR_OUT" are coming IN to the Target
+ * over the interconnect.
+ */
+#define PIPEDIR_NONE		0
+#define PIPEDIR_IN		1 /* Target-->Host, WiFi Rx direction */
+#define PIPEDIR_OUT		2 /* Host->Target, WiFi Tx direction */
+#define PIPEDIR_INOUT		3 /* bidirectional */
+#define PIPEDIR_INOUT_H2H	4 /* bidirectional, host to host */
+
+/* CE address/mask */
+#define CE_HOST_IE_ADDRESS	0x00A1803C
+#define CE_HOST_IE_2_ADDRESS	0x00A18040
+#define CE_HOST_IE_3_ADDRESS	CE_HOST_IE_ADDRESS
+
+#define CE_HOST_IE_3_SHIFT	0xC
+
+#define CE_RING_IDX_INCR(nentries_mask, idx) (((idx) + 1) & (nentries_mask))
+
+#define ATH11K_CE_RX_POST_RETRY_JIFFIES 50
+
+struct ath11k_base;
+
+/* Establish a mapping between a service/direction and a pipe. */
+struct service_to_pipe {
+	__le32 service_id;
+	__le32 pipedir;
+	__le32 pipenum;
+};
+
+/*
+ * Configuration information for a Copy Engine pipe.
+ * Passed from Host to Target during startup (one per CE).
+ *
+ * NOTE: Structure is shared between Host software and Target firmware!
+ */
+struct ce_pipe_config {
+	__le32 pipenum;
+	__le32 pipedir;
+	__le32 nentries;
+	__le32 nbytes_max;
+	__le32 flags;
+	__le32 reserved;
+};
+
+struct ce_attr {
+	/* CE_ATTR_* values */
+	unsigned int flags;
+
+	/* #entries in source ring - Must be a power of 2 */
+	unsigned int src_nentries;
+
+	/*
+	 * Max source send size for this CE.
+	 * This is also the minimum size of a destination buffer.
+	 */
+	unsigned int src_sz_max;
+
+	/* #entries in destination ring - Must be a power of 2 */
+	unsigned int dest_nentries;
+
+	void (*recv_cb)(struct ath11k_base *, struct sk_buff *);
+};
+
+#define CE_DESC_RING_ALIGN 8
+
+struct ath11k_ce_ring {
+	/* Number of entries in this ring; must be power of 2 */
+	unsigned int nentries;
+	unsigned int nentries_mask;
+
+	/* For dest ring, this is the next index to be processed
+	 * by software after it was/is received into.
+	 *
+	 * For src ring, this is the last descriptor that was sent
+	 * and completion processed by software.
+	 *
+	 * Regardless of src or dest ring, this is an invariant
+	 * (modulo ring size):
+	 *     write index >= read index >= sw_index
+	 */
+	unsigned int sw_index;
+	/* cached copy */
+	unsigned int write_index;
+
+	/* Start of DMA-coherent area reserved for descriptors */
+	/* Host address space */
+	void *base_addr_owner_space_unaligned;
+	/* CE address space */
+	u32 base_addr_ce_space_unaligned;
+
+	/* Actual start of descriptors.
+	 * Aligned to descriptor-size boundary.
+	 * Points into reserved DMA-coherent area, above.
+	 */
+	/* Host address space */
+	void *base_addr_owner_space;
+
+	/* CE address space */
+	u32 base_addr_ce_space;
+
+	/* HAL ring id */
+	u32 hal_ring_id;
+
+	/* keep last */
+	struct sk_buff *skb[0];
+};
+
+struct ath11k_ce_pipe {
+	struct ath11k_base *ab;
+	u16 pipe_num;
+	unsigned int attr_flags;
+	unsigned int buf_sz;
+	unsigned int rx_buf_needed;
+
+	void (*send_cb)(struct ath11k_ce_pipe *);
+	void (*recv_cb)(struct ath11k_base *, struct sk_buff *);
+
+	struct tasklet_struct intr_tq;
+	struct ath11k_ce_ring *src_ring;
+	struct ath11k_ce_ring *dest_ring;
+	struct ath11k_ce_ring *status_ring;
+};
+
+struct ath11k_ce {
+	struct ath11k_ce_pipe ce_pipe[CE_COUNT];
+	/* Protects rings of all ce pipes */
+	spinlock_t ce_lock;
+};
+
+void ath11k_ce_cleanup_pipes(struct ath11k_base *ab);
+void ath11k_ce_rx_replenish_retry(struct timer_list *t);
+void ath11k_ce_per_engine_service(struct ath11k_base *ab, u16 ce_id);
+int ath11k_ce_send(struct ath11k_base *ab, struct sk_buff *skb, u8 pipe_id,
+		   u16 transfer_id);
+void ath11k_ce_rx_post_buf(struct ath11k_base *ab);
+int ath11k_ce_init_pipes(struct ath11k_base *ab);
+int ath11k_ce_alloc_pipes(struct ath11k_base *ab);
+void ath11k_ce_free_pipes(struct ath11k_base *ab);
+int ath11k_ce_get_attr_flags(int ce_id);
+void ath11k_ce_poll_send_completed(struct ath11k_base *ab, u8 pipe_id);
+#endif
diff --git a/drivers/net/wireless/ath/ath11k/core.c b/drivers/net/wireless/ath/ath11k/core.c
new file mode 100644
index 000000000000..b0c71b5879b1
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/core.c
@@ -0,0 +1,772 @@
+// SPDX-License-Identifier: BSD-3-Clause-Clear
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/remoteproc.h>
+#include <linux/firmware.h>
+#include "ahb.h"
+#include "core.h"
+#include "dp_tx.h"
+#include "debug.h"
+
+unsigned int ath11k_debug_mask;
+module_param_named(debug_mask, ath11k_debug_mask, uint, 0644);
+MODULE_PARM_DESC(debug_mask, "Debugging mask");
+
+static const struct ath11k_hw_params ath11k_hw_params = {
+	.name = "ipq8074",
+	.fw = {
+		.dir = IPQ8074_FW_DIR,
+		.board_size = IPQ8074_MAX_BOARD_DATA_SZ,
+		.cal_size =  IPQ8074_MAX_CAL_DATA_SZ,
+	},
+};
+
+/* Map from pdev index to hw mac index */
+u8 ath11k_core_get_hw_mac_id(struct ath11k_base *ab, int pdev_idx)
+{
+	switch (pdev_idx) {
+	case 0:
+		return 0;
+	case 1:
+		return 2;
+	case 2:
+		return 1;
+	default:
+		ath11k_warn(ab, "Invalid pdev idx %d\n", pdev_idx);
+		return ATH11K_INVALID_HW_MAC_ID;
+	}
+}
+
+static int ath11k_core_create_board_name(struct ath11k_base *ab, char *name,
+					 size_t name_len)
+{
+	/* Note: bus is fixed to ahb. When other bus type supported,
+	 * make it to dynamic.
+	 */
+	scnprintf(name, name_len,
+		  "bus=ahb,qmi-chip-id=%d,qmi-board-id=%d",
+		  ab->qmi.target.chip_id,
+		  ab->qmi.target.board_id);
+
+	ath11k_dbg(ab, ATH11K_DBG_BOOT, "boot using board name '%s'\n", name);
+
+	return 0;
+}
+
+static const struct firmware *ath11k_fetch_fw_file(struct ath11k_base *ab,
+						   const char *dir,
+						   const char *file)
+{
+	char filename[100];
+	const struct firmware *fw;
+	int ret;
+
+	if (file == NULL)
+		return ERR_PTR(-ENOENT);
+
+	if (dir == NULL)
+		dir = ".";
+
+	snprintf(filename, sizeof(filename), "%s/%s", dir, file);
+	ret = firmware_request_nowarn(&fw, filename, ab->dev);
+	ath11k_dbg(ab, ATH11K_DBG_BOOT, "boot fw request '%s': %d\n",
+		   filename, ret);
+
+	if (ret)
+		return ERR_PTR(ret);
+	ath11k_warn(ab, "Downloading BDF: %s, size: %zu\n",
+		    filename, fw->size);
+
+	return fw;
+}
+
+void ath11k_core_free_bdf(struct ath11k_base *ab, struct ath11k_board_data *bd)
+{
+	if (!IS_ERR(bd->fw))
+		release_firmware(bd->fw);
+
+	memset(bd, 0, sizeof(*bd));
+}
+
+static int ath11k_core_parse_bd_ie_board(struct ath11k_base *ab,
+					 struct ath11k_board_data *bd,
+					 const void *buf, size_t buf_len,
+					 const char *boardname,
+					 int bd_ie_type)
+{
+	const struct ath11k_fw_ie *hdr;
+	bool name_match_found;
+	int ret, board_ie_id;
+	size_t board_ie_len;
+	const void *board_ie_data;
+
+	name_match_found = false;
+
+	/* go through ATH11K_BD_IE_BOARD_ elements */
+	while (buf_len > sizeof(struct ath11k_fw_ie)) {
+		hdr = buf;
+		board_ie_id = le32_to_cpu(hdr->id);
+		board_ie_len = le32_to_cpu(hdr->len);
+		board_ie_data = hdr->data;
+
+		buf_len -= sizeof(*hdr);
+		buf += sizeof(*hdr);
+
+		if (buf_len < ALIGN(board_ie_len, 4)) {
+			ath11k_err(ab, "invalid ATH11K_BD_IE_BOARD length: %zu < %zu\n",
+				   buf_len, ALIGN(board_ie_len, 4));
+			ret = -EINVAL;
+			goto out;
+		}
+
+		switch (board_ie_id) {
+		case ATH11K_BD_IE_BOARD_NAME:
+			ath11k_dbg_dump(ab, ATH11K_DBG_BOOT, "board name", "",
+					board_ie_data, board_ie_len);
+
+			if (board_ie_len != strlen(boardname))
+				break;
+
+			ret = memcmp(board_ie_data, boardname, strlen(boardname));
+			if (ret)
+				break;
+
+			name_match_found = true;
+			ath11k_dbg(ab, ATH11K_DBG_BOOT,
+				   "boot found match for name '%s'",
+				   boardname);
+			break;
+		case ATH11K_BD_IE_BOARD_DATA:
+			if (!name_match_found)
+				/* no match found */
+				break;
+
+			ath11k_dbg(ab, ATH11K_DBG_BOOT,
+				   "boot found board data for '%s'", boardname);
+
+			bd->data = board_ie_data;
+			bd->len = board_ie_len;
+
+			ret = 0;
+			goto out;
+		default:
+			ath11k_warn(ab, "unknown ATH11K_BD_IE_BOARD found: %d\n",
+				    board_ie_id);
+			break;
+		}
+
+		/* jump over the padding */
+		board_ie_len = ALIGN(board_ie_len, 4);
+
+		buf_len -= board_ie_len;
+		buf += board_ie_len;
+	}
+
+	/* no match found */
+	ret = -ENOENT;
+
+out:
+	return ret;
+}
+
+static int ath11k_core_fetch_board_data_api_n(struct ath11k_base *ab,
+					      struct ath11k_board_data *bd,
+					      const char *boardname)
+{
+	size_t len, magic_len;
+	const u8 *data;
+	char *filename = ATH11K_BOARD_API2_FILE;
+	size_t ie_len;
+	struct ath11k_fw_ie *hdr;
+	int ret, ie_id;
+
+	if (!bd->fw)
+		bd->fw = ath11k_fetch_fw_file(ab,
+					      ab->hw_params.fw.dir,
+					      filename);
+	if (IS_ERR(bd->fw))
+		return PTR_ERR(bd->fw);
+
+	data = bd->fw->data;
+	len = bd->fw->size;
+
+	/* magic has extra null byte padded */
+	magic_len = strlen(ATH11K_BOARD_MAGIC) + 1;
+	if (len < magic_len) {
+		ath11k_err(ab, "failed to find magic value in %s/%s, file too short: %zu\n",
+			   ab->hw_params.fw.dir, filename, len);
+		ret = -EINVAL;
+		goto err;
+	}
+
+	if (memcmp(data, ATH11K_BOARD_MAGIC, magic_len)) {
+		ath11k_err(ab, "found invalid board magic\n");
+		ret = -EINVAL;
+		goto err;
+	}
+
+	/* magic is padded to 4 bytes */
+	magic_len = ALIGN(magic_len, 4);
+	if (len < magic_len) {
+		ath11k_err(ab, "failed: %s/%s too small to contain board data, len: %zu\n",
+			   ab->hw_params.fw.dir, filename, len);
+		ret = -EINVAL;
+		goto err;
+	}
+
+	data += magic_len;
+	len -= magic_len;
+
+	while (len > sizeof(struct ath11k_fw_ie)) {
+		hdr = (struct ath11k_fw_ie *)data;
+		ie_id = le32_to_cpu(hdr->id);
+		ie_len = le32_to_cpu(hdr->len);
+
+		len -= sizeof(*hdr);
+		data = hdr->data;
+
+		if (len < ALIGN(ie_len, 4)) {
+			ath11k_err(ab, "invalid length for board ie_id %d ie_len %zu len %zu\n",
+				   ie_id, ie_len, len);
+			return -EINVAL;
+		}
+
+		switch (ie_id) {
+		case ATH11K_BD_IE_BOARD:
+			ret = ath11k_core_parse_bd_ie_board(ab, bd, data,
+							    ie_len,
+							    boardname,
+							    ATH11K_BD_IE_BOARD);
+			if (ret == -ENOENT)
+				/* no match found, continue */
+				break;
+			else if (ret)
+				/* there was an error, bail out */
+				goto err;
+			/* either found or error, so stop searching */
+			goto out;
+		}
+
+		/* jump over the padding */
+		ie_len = ALIGN(ie_len, 4);
+
+		len -= ie_len;
+		data += ie_len;
+	}
+
+out:
+	if (!bd->data || !bd->len) {
+		ath11k_err(ab,
+			   "failed to fetch board data for %s from %s/%s\n",
+			   boardname, ab->hw_params.fw.dir, filename);
+		ret = -ENODATA;
+		goto err;
+	}
+
+	return 0;
+
+err:
+	ath11k_core_free_bdf(ab, bd);
+	return ret;
+}
+
+static int ath11k_core_fetch_board_data_api_1(struct ath11k_base *ab,
+					      struct ath11k_board_data *bd)
+{
+	bd->fw = ath11k_fetch_fw_file(ab,
+				      ab->hw_params.fw.dir,
+				      ATH11K_DEFAULT_BOARD_FILE);
+	if (IS_ERR(bd->fw))
+		return PTR_ERR(bd->fw);
+
+	bd->data = bd->fw->data;
+	bd->len = bd->fw->size;
+
+	return 0;
+}
+
+#define BOARD_NAME_SIZE 100
+int ath11k_core_fetch_bdf(struct ath11k_base *ab, struct ath11k_board_data *bd)
+{
+	char boardname[BOARD_NAME_SIZE];
+	int ret;
+
+	ret = ath11k_core_create_board_name(ab, boardname, BOARD_NAME_SIZE);
+	if (ret) {
+		ath11k_err(ab, "failed to create board name: %d", ret);
+		return ret;
+	}
+
+	ab->bd_api = 2;
+	ret = ath11k_core_fetch_board_data_api_n(ab, bd, boardname);
+	if (!ret)
+		goto success;
+
+	ab->bd_api = 1;
+	ret = ath11k_core_fetch_board_data_api_1(ab, bd);
+	if (ret) {
+		ath11k_err(ab, "failed to fetch board-2.bin or board.bin from %s\n",
+			   ab->hw_params.fw.dir);
+		return ret;
+	}
+
+success:
+	ath11k_dbg(ab, ATH11K_DBG_BOOT, "using board api %d\n", ab->bd_api);
+	return 0;
+}
+
+static void ath11k_core_stop(struct ath11k_base *ab)
+{
+	if (!test_bit(ATH11K_FLAG_CRASH_FLUSH, &ab->dev_flags))
+		ath11k_qmi_firmware_stop(ab);
+	ath11k_ahb_stop(ab);
+	ath11k_wmi_detach(ab);
+
+	/* De-Init of components as needed */
+}
+
+static int ath11k_core_soc_create(struct ath11k_base *ab)
+{
+	int ret;
+
+	ret = ath11k_qmi_init_service(ab);
+	if (ret) {
+		ath11k_err(ab, "failed to initialize qmi :%d\n", ret);
+		return ret;
+	}
+
+	ret = ath11k_debug_soc_create(ab);
+	if (ret) {
+		ath11k_err(ab, "failed to create ath11k debugfs\n");
+		goto err_qmi_deinit;
+	}
+
+	ret = ath11k_ahb_power_up(ab);
+	if (ret) {
+		ath11k_err(ab, "failed to power up :%d\n", ret);
+		goto err_debugfs_reg;
+	}
+
+	return 0;
+
+err_debugfs_reg:
+	ath11k_debug_soc_destroy(ab);
+err_qmi_deinit:
+	ath11k_qmi_deinit_service(ab);
+	return ret;
+}
+
+static void ath11k_core_soc_destroy(struct ath11k_base *ab)
+{
+	ath11k_debug_soc_destroy(ab);
+	ath11k_dp_free(ab);
+	ath11k_reg_free(ab);
+	ath11k_qmi_deinit_service(ab);
+}
+
+static int ath11k_core_pdev_create(struct ath11k_base *ab)
+{
+	int ret;
+
+	ret = ath11k_debug_pdev_create(ab);
+	if (ret) {
+		ath11k_err(ab, "failed to create core pdev debugfs: %d\n", ret);
+		return ret;
+	}
+
+	ret = ath11k_mac_create(ab);
+	if (ret) {
+		ath11k_err(ab, "failed to create new hw device with mac80211 :%d\n",
+			   ret);
+		goto err_pdev_debug;
+	}
+
+	ret = ath11k_dp_pdev_alloc(ab);
+	if (ret) {
+		ath11k_err(ab, "failed to attach DP pdev: %d\n", ret);
+		goto err_mac_destroy;
+	}
+
+	return 0;
+
+err_mac_destroy:
+	ath11k_mac_destroy(ab);
+
+err_pdev_debug:
+	ath11k_debug_pdev_destroy(ab);
+
+	return ret;
+}
+
+static void ath11k_core_pdev_destroy(struct ath11k_base *ab)
+{
+	ath11k_mac_unregister(ab);
+	ath11k_ahb_ext_irq_disable(ab);
+	ath11k_dp_pdev_free(ab);
+	ath11k_debug_pdev_destroy(ab);
+}
+
+static int ath11k_core_start(struct ath11k_base *ab,
+			     enum ath11k_firmware_mode mode)
+{
+	int ret;
+
+	ret = ath11k_qmi_firmware_start(ab, mode);
+	if (ret) {
+		ath11k_err(ab, "failed to attach wmi: %d\n", ret);
+		return ret;
+	}
+
+	ret = ath11k_wmi_attach(ab);
+	if (ret) {
+		ath11k_err(ab, "failed to attach wmi: %d\n", ret);
+		goto err_firmware_stop;
+	}
+
+	ret = ath11k_htc_init(ab);
+	if (ret) {
+		ath11k_err(ab, "failed to init htc: %d\n", ret);
+		goto err_wmi_detach;
+	}
+
+	ret = ath11k_ahb_start(ab);
+	if (ret) {
+		ath11k_err(ab, "failed to start HIF: %d\n", ret);
+		goto err_wmi_detach;
+	}
+
+	ret = ath11k_htc_wait_target(&ab->htc);
+	if (ret) {
+		ath11k_err(ab, "failed to connect to HTC: %d\n", ret);
+		goto err_hif_stop;
+	}
+
+	ret = ath11k_dp_htt_connect(&ab->dp);
+	if (ret) {
+		ath11k_err(ab, "failed to connect to HTT: %d\n", ret);
+		goto err_hif_stop;
+	}
+
+	ret = ath11k_wmi_connect(ab);
+	if (ret) {
+		ath11k_err(ab, "failed to connect wmi: %d\n", ret);
+		goto err_hif_stop;
+	}
+
+	ret = ath11k_htc_start(&ab->htc);
+	if (ret) {
+		ath11k_err(ab, "failed to start HTC: %d\n", ret);
+		goto err_hif_stop;
+	}
+
+	ret = ath11k_wmi_wait_for_service_ready(ab);
+	if (ret) {
+		ath11k_err(ab, "failed to receive wmi service ready event: %d\n",
+			   ret);
+		goto err_hif_stop;
+	}
+
+	ret = ath11k_wmi_cmd_init(ab);
+	if (ret) {
+		ath11k_err(ab, "failed to send wmi init cmd: %d\n", ret);
+		goto err_hif_stop;
+	}
+
+	ret = ath11k_wmi_wait_for_unified_ready(ab);
+	if (ret) {
+		ath11k_err(ab, "failed to receive wmi unified ready event: %d\n",
+			   ret);
+		goto err_hif_stop;
+	}
+
+	ret = ath11k_dp_tx_htt_h2t_ver_req_msg(ab);
+	if (ret) {
+		ath11k_err(ab, "failed to send htt version request message: %d\n",
+			   ret);
+		goto err_hif_stop;
+	}
+
+	return 0;
+
+err_hif_stop:
+	ath11k_ahb_stop(ab);
+err_wmi_detach:
+	ath11k_wmi_detach(ab);
+err_firmware_stop:
+	ath11k_qmi_firmware_stop(ab);
+
+	return ret;
+}
+
+int ath11k_core_qmi_firmware_ready(struct ath11k_base *ab)
+{
+	int ret;
+
+	ret = ath11k_ce_init_pipes(ab);
+	if (ret) {
+		ath11k_err(ab, "failed to initialize CE: %d\n", ret);
+		return ret;
+	}
+
+	ret = ath11k_dp_alloc(ab);
+	if (ret) {
+		ath11k_err(ab, "failed to init DP: %d\n", ret);
+		return ret;
+	}
+
+	mutex_lock(&ab->core_lock);
+	ret = ath11k_core_start(ab, ATH11K_FIRMWARE_MODE_NORMAL);
+	if (ret) {
+		ath11k_err(ab, "failed to start core: %d\n", ret);
+		goto err_dp_free;
+	}
+
+	ret = ath11k_core_pdev_create(ab);
+	if (ret) {
+		ath11k_err(ab, "failed to create pdev core: %d\n", ret);
+		goto err_core_stop;
+	}
+	ath11k_ahb_ext_irq_enable(ab);
+	mutex_unlock(&ab->core_lock);
+
+	return 0;
+
+err_core_stop:
+	ath11k_core_stop(ab);
+err_dp_free:
+	ath11k_dp_free(ab);
+	return ret;
+}
+
+static int ath11k_core_reconfigure_on_crash(struct ath11k_base *ab)
+{
+	int ret;
+
+	mutex_lock(&ab->core_lock);
+	ath11k_ahb_ext_irq_disable(ab);
+	ath11k_dp_pdev_free(ab);
+	ath11k_ahb_stop(ab);
+	ath11k_wmi_detach(ab);
+	mutex_unlock(&ab->core_lock);
+
+	ath11k_dp_free(ab);
+	ath11k_hal_srng_deinit(ab);
+
+	ab->free_vdev_map = (1LL << (ab->num_radios * TARGET_NUM_VDEVS)) - 1;
+
+	ret = ath11k_hal_srng_init(ab);
+	if (ret)
+		return ret;
+
+	clear_bit(ATH11K_FLAG_CRASH_FLUSH, &ab->dev_flags);
+
+	ret = ath11k_core_qmi_firmware_ready(ab);
+	if (ret)
+		goto err_hal_srng_deinit;
+
+	clear_bit(ATH11K_FLAG_RECOVERY, &ab->dev_flags);
+
+	return 0;
+
+err_hal_srng_deinit:
+	ath11k_hal_srng_deinit(ab);
+	return ret;
+}
+
+void ath11k_core_halt(struct ath11k *ar)
+{
+	struct ath11k_base *ab = ar->ab;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	ar->num_created_vdevs = 0;
+
+	ath11k_mac_scan_finish(ar);
+	ath11k_mac_peer_cleanup_all(ar);
+	cancel_delayed_work_sync(&ar->scan.timeout);
+	cancel_work_sync(&ar->regd_update_work);
+
+	rcu_assign_pointer(ab->pdevs_active[ar->pdev_idx], NULL);
+	synchronize_rcu();
+	INIT_LIST_HEAD(&ar->arvifs);
+	idr_init(&ar->txmgmt_idr);
+}
+
+static void ath11k_core_restart(struct work_struct *work)
+{
+	struct ath11k_base *ab = container_of(work, struct ath11k_base, restart_work);
+	struct ath11k *ar;
+	struct ath11k_pdev *pdev;
+	int i, ret = 0;
+
+	spin_lock_bh(&ab->base_lock);
+	ab->stats.fw_crash_counter++;
+	spin_unlock_bh(&ab->base_lock);
+
+	for (i = 0; i < ab->num_radios; i++) {
+		pdev = &ab->pdevs[i];
+		ar = pdev->ar;
+		if (!ar || ar->state == ATH11K_STATE_OFF)
+			continue;
+
+		ieee80211_stop_queues(ar->hw);
+		ath11k_mac_drain_tx(ar);
+		complete(&ar->scan.started);
+		complete(&ar->scan.completed);
+		complete(&ar->peer_assoc_done);
+		complete(&ar->install_key_done);
+		complete(&ar->vdev_setup_done);
+		complete(&ar->bss_survey_done);
+
+		wake_up(&ar->dp.tx_empty_waitq);
+		idr_for_each(&ar->txmgmt_idr,
+			     ath11k_mac_tx_mgmt_pending_free, ar);
+		idr_destroy(&ar->txmgmt_idr);
+	}
+
+	wake_up(&ab->wmi_sc.tx_credits_wq);
+	wake_up(&ab->peer_mapping_wq);
+
+	ret = ath11k_core_reconfigure_on_crash(ab);
+	if (ret) {
+		ath11k_err(ab, "failed to reconfigure driver on crash recovery\n");
+		return;
+	}
+
+	for (i = 0; i < ab->num_radios; i++) {
+		pdev = &ab->pdevs[i];
+		ar = pdev->ar;
+		if (!ar || ar->state == ATH11K_STATE_OFF)
+			continue;
+
+		mutex_lock(&ar->conf_mutex);
+
+		switch (ar->state) {
+		case ATH11K_STATE_ON:
+			ar->state = ATH11K_STATE_RESTARTING;
+			ath11k_core_halt(ar);
+			ieee80211_restart_hw(ar->hw);
+			break;
+		case ATH11K_STATE_OFF:
+			ath11k_warn(ab,
+				    "cannot restart radio %d that hasn't been started\n",
+				    i);
+			break;
+		case ATH11K_STATE_RESTARTING:
+			break;
+		case ATH11K_STATE_RESTARTED:
+			ar->state = ATH11K_STATE_WEDGED;
+			/* fall through */
+		case ATH11K_STATE_WEDGED:
+			ath11k_warn(ab,
+				    "device is wedged, will not restart radio %d\n", i);
+			break;
+		}
+		mutex_unlock(&ar->conf_mutex);
+	}
+	complete(&ab->driver_recovery);
+}
+
+int ath11k_core_init(struct ath11k_base *ab)
+{
+	struct device *dev = ab->dev;
+	struct rproc *prproc;
+	phandle rproc_phandle;
+	int ret;
+
+	if (of_property_read_u32(dev->of_node, "qcom,rproc", &rproc_phandle)) {
+		ath11k_err(ab, "failed to get q6_rproc handle\n");
+		return -ENOENT;
+	}
+
+	prproc = rproc_get_by_phandle(rproc_phandle);
+	if (!prproc) {
+		ath11k_err(ab, "failed to get rproc\n");
+		return -EINVAL;
+	}
+	ab->tgt_rproc = prproc;
+	ab->hw_params = ath11k_hw_params;
+
+	ret = ath11k_core_soc_create(ab);
+	if (ret) {
+		ath11k_err(ab, "failed to create soc core: %d\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+void ath11k_core_deinit(struct ath11k_base *ab)
+{
+	mutex_lock(&ab->core_lock);
+
+	ath11k_core_pdev_destroy(ab);
+	ath11k_core_stop(ab);
+
+	mutex_unlock(&ab->core_lock);
+
+	ath11k_ahb_power_down(ab);
+	ath11k_mac_destroy(ab);
+	ath11k_core_soc_destroy(ab);
+}
+
+void ath11k_core_free(struct ath11k_base *ab)
+{
+	kfree(ab);
+}
+
+struct ath11k_base *ath11k_core_alloc(struct device *dev)
+{
+	struct ath11k_base *ab;
+
+	ab = kzalloc(sizeof(*ab), GFP_KERNEL);
+	if (!ab)
+		return NULL;
+
+	init_completion(&ab->driver_recovery);
+
+	ab->workqueue = create_singlethread_workqueue("ath11k_wq");
+	if (!ab->workqueue)
+		goto err_sc_free;
+
+	mutex_init(&ab->core_lock);
+	spin_lock_init(&ab->base_lock);
+
+	INIT_LIST_HEAD(&ab->peers);
+	init_waitqueue_head(&ab->peer_mapping_wq);
+	init_waitqueue_head(&ab->wmi_sc.tx_credits_wq);
+	INIT_WORK(&ab->restart_work, ath11k_core_restart);
+	timer_setup(&ab->rx_replenish_retry, ath11k_ce_rx_replenish_retry, 0);
+	ab->dev = dev;
+
+	return ab;
+
+err_sc_free:
+	kfree(ab);
+	return NULL;
+}
+
+static int __init ath11k_init(void)
+{
+	int ret;
+
+	ret = ath11k_ahb_init();
+	if (ret)
+		printk(KERN_ERR "failed to register ath11k ahb driver: %d\n",
+		       ret);
+	return ret;
+}
+module_init(ath11k_init);
+
+static void __exit ath11k_exit(void)
+{
+	ath11k_ahb_exit();
+}
+module_exit(ath11k_exit);
+
+MODULE_DESCRIPTION("Driver support for Qualcomm Technologies 802.11ax wireless chip");
+MODULE_LICENSE("Dual BSD/GPL");
diff --git a/drivers/net/wireless/ath/ath11k/core.h b/drivers/net/wireless/ath/ath11k/core.h
new file mode 100644
index 000000000000..95447d1be104
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/core.h
@@ -0,0 +1,823 @@
+/* SPDX-License-Identifier: BSD-3-Clause-Clear */
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#ifndef ATH11K_CORE_H
+#define ATH11K_CORE_H
+
+#include <linux/types.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/bitfield.h>
+#include "qmi.h"
+#include "htc.h"
+#include "wmi.h"
+#include "hal.h"
+#include "dp.h"
+#include "ce.h"
+#include "mac.h"
+#include "hw.h"
+#include "hal_rx.h"
+#include "reg.h"
+
+#define SM(_v, _f) (((_v) << _f##_LSB) & _f##_MASK)
+
+#define ATH11K_TX_MGMT_NUM_PENDING_MAX	512
+
+#define ATH11K_TX_MGMT_TARGET_MAX_SUPPORT_WMI 64
+
+/* Pending management packets threshold for dropping probe responses */
+#define ATH11K_PRB_RSP_DROP_THRESHOLD ((ATH11K_TX_MGMT_TARGET_MAX_SUPPORT_WMI * 3) / 4)
+
+#define ATH11K_INVALID_HW_MAC_ID	0xFF
+
+enum ath11k_supported_bw {
+	ATH11K_BW_20	= 0,
+	ATH11K_BW_40	= 1,
+	ATH11K_BW_80	= 2,
+	ATH11K_BW_160	= 3,
+};
+
+enum wme_ac {
+	WME_AC_BE,
+	WME_AC_BK,
+	WME_AC_VI,
+	WME_AC_VO,
+	WME_NUM_AC
+};
+
+#define ATH11K_HT_MCS_MAX	7
+#define ATH11K_VHT_MCS_MAX	9
+#define ATH11K_HE_MCS_MAX	11
+
+static inline enum wme_ac ath11k_tid_to_ac(u32 tid)
+{
+	return (((tid == 0) || (tid == 3)) ? WME_AC_BE :
+		((tid == 1) || (tid == 2)) ? WME_AC_BK :
+		((tid == 4) || (tid == 5)) ? WME_AC_VI :
+		WME_AC_VO);
+}
+
+struct ath11k_skb_cb {
+	dma_addr_t paddr;
+	u8 eid;
+	struct ath11k *ar;
+	struct ieee80211_vif *vif;
+} __packed;
+
+struct ath11k_skb_rxcb {
+	dma_addr_t paddr;
+	bool is_first_msdu;
+	bool is_last_msdu;
+	bool is_continuation;
+	struct hal_rx_desc *rx_desc;
+	u8 err_rel_src;
+	u8 err_code;
+	u8 mac_id;
+	u8 unmapped;
+};
+
+enum ath11k_hw_rev {
+	ATH11K_HW_IPQ8074,
+};
+
+enum ath11k_firmware_mode {
+	/* the default mode, standard 802.11 functionality */
+	ATH11K_FIRMWARE_MODE_NORMAL,
+
+	/* factory tests etc */
+	ATH11K_FIRMWARE_MODE_FTM,
+};
+
+#define ATH11K_IRQ_NUM_MAX 52
+#define ATH11K_EXT_IRQ_GRP_NUM_MAX 11
+#define ATH11K_EXT_IRQ_NUM_MAX	16
+
+extern const u8 ath11k_reo_status_ring_mask[ATH11K_EXT_IRQ_GRP_NUM_MAX];
+extern const u8 ath11k_tx_ring_mask[ATH11K_EXT_IRQ_GRP_NUM_MAX];
+extern const u8 ath11k_rx_ring_mask[ATH11K_EXT_IRQ_GRP_NUM_MAX];
+extern const u8 ath11k_rx_err_ring_mask[ATH11K_EXT_IRQ_GRP_NUM_MAX];
+extern const u8 ath11k_rx_wbm_rel_ring_mask[ATH11K_EXT_IRQ_GRP_NUM_MAX];
+extern const u8 ath11k_rxdma2host_ring_mask[ATH11K_EXT_IRQ_GRP_NUM_MAX];
+extern const u8 ath11k_host2rxdma_ring_mask[ATH11K_EXT_IRQ_GRP_NUM_MAX];
+extern const u8 rx_mon_status_ring_mask[ATH11K_EXT_IRQ_GRP_NUM_MAX];
+
+struct ath11k_ext_irq_grp {
+	struct ath11k_base *ab;
+	u32 irqs[ATH11K_EXT_IRQ_NUM_MAX];
+	u32 num_irq;
+	u32 grp_id;
+	struct napi_struct napi;
+	struct net_device napi_ndev;
+	/* Queue of pending packets, not expected to be accessed concurrently
+	 * to avoid locking overhead.
+	 */
+	struct sk_buff_head pending_q;
+};
+
+#define HEHANDLE_CAP_PHYINFO_SIZE       3
+#define HECAP_PHYINFO_SIZE              9
+#define HECAP_MACINFO_SIZE              5
+#define HECAP_TXRX_MCS_NSS_SIZE         2
+#define HECAP_PPET16_PPET8_MAX_SIZE     25
+
+#define HE_PPET16_PPET8_SIZE            8
+
+/* 802.11ax PPE (PPDU packet Extension) threshold */
+struct he_ppe_threshold {
+	u32 numss_m1;
+	u32 ru_mask;
+	u32 ppet16_ppet8_ru3_ru0[HE_PPET16_PPET8_SIZE];
+};
+
+struct ath11k_he {
+	u8 hecap_macinfo[HECAP_MACINFO_SIZE];
+	u32 hecap_rxmcsnssmap;
+	u32 hecap_txmcsnssmap;
+	u32 hecap_phyinfo[HEHANDLE_CAP_PHYINFO_SIZE];
+	struct he_ppe_threshold   hecap_ppet;
+	u32 heop_param;
+};
+
+#define MAX_RADIOS 3
+
+enum {
+	WMI_HOST_TP_SCALE_MAX   = 0,
+	WMI_HOST_TP_SCALE_50    = 1,
+	WMI_HOST_TP_SCALE_25    = 2,
+	WMI_HOST_TP_SCALE_12    = 3,
+	WMI_HOST_TP_SCALE_MIN   = 4,
+	WMI_HOST_TP_SCALE_SIZE   = 5,
+};
+
+enum ath11k_scan_state {
+	ATH11K_SCAN_IDLE,
+	ATH11K_SCAN_STARTING,
+	ATH11K_SCAN_RUNNING,
+	ATH11K_SCAN_ABORTING,
+};
+
+enum ath11k_dev_flags {
+	ATH11K_CAC_RUNNING,
+	ATH11K_FLAG_CORE_REGISTERED,
+	ATH11K_FLAG_CRASH_FLUSH,
+	ATH11K_FLAG_RAW_MODE,
+	ATH11K_FLAG_HW_CRYPTO_DISABLED,
+	ATH11K_FLAG_BTCOEX,
+	ATH11K_FLAG_RECOVERY,
+	ATH11K_FLAG_UNREGISTERING,
+	ATH11K_FLAG_REGISTERED,
+};
+
+enum ath11k_monitor_flags {
+	ATH11K_FLAG_MONITOR_ENABLED,
+};
+
+struct ath11k_vif {
+	u32 vdev_id;
+	enum wmi_vdev_type vdev_type;
+	enum wmi_vdev_subtype vdev_subtype;
+	u32 beacon_interval;
+	u32 dtim_period;
+	u16 ast_hash;
+	u16 tcl_metadata;
+	u8 hal_addr_search_flags;
+	u8 search_type;
+
+	struct ath11k *ar;
+	struct ieee80211_vif *vif;
+
+	u16 tx_seq_no;
+	struct wmi_wmm_params_all_arg wmm_params;
+	struct list_head list;
+	union {
+		struct {
+			u32 uapsd;
+		} sta;
+		struct {
+			/* 127 stations; wmi limit */
+			u8 tim_bitmap[16];
+			u8 tim_len;
+			u32 ssid_len;
+			u8 ssid[IEEE80211_MAX_SSID_LEN];
+			bool hidden_ssid;
+			/* P2P_IE with NoA attribute for P2P_GO case */
+			u32 noa_len;
+			u8 *noa_data;
+		} ap;
+	} u;
+
+	bool is_started;
+	bool is_up;
+	u32 aid;
+	u8 bssid[ETH_ALEN];
+	struct cfg80211_bitrate_mask bitrate_mask;
+	int num_legacy_stations;
+	int rtscts_prot_mode;
+	int txpower;
+};
+
+struct ath11k_vif_iter {
+	u32 vdev_id;
+	struct ath11k_vif *arvif;
+};
+
+struct ath11k_rx_peer_stats {
+	u64 num_msdu;
+	u64 num_mpdu_fcs_ok;
+	u64 num_mpdu_fcs_err;
+	u64 tcp_msdu_count;
+	u64 udp_msdu_count;
+	u64 other_msdu_count;
+	u64 ampdu_msdu_count;
+	u64 non_ampdu_msdu_count;
+	u64 stbc_count;
+	u64 beamformed_count;
+	u64 mcs_count[HAL_RX_MAX_MCS + 1];
+	u64 nss_count[HAL_RX_MAX_NSS];
+	u64 bw_count[HAL_RX_BW_MAX];
+	u64 gi_count[HAL_RX_GI_MAX];
+	u64 coding_count[HAL_RX_SU_MU_CODING_MAX];
+	u64 tid_count[IEEE80211_NUM_TIDS + 1];
+	u64 pream_cnt[HAL_RX_PREAMBLE_MAX];
+	u64 reception_type[HAL_RX_RECEPTION_TYPE_MAX];
+	u64 rx_duration;
+};
+
+#define ATH11K_HE_MCS_NUM       12
+#define ATH11K_VHT_MCS_NUM      10
+#define ATH11K_BW_NUM           4
+#define ATH11K_NSS_NUM          4
+#define ATH11K_LEGACY_NUM       12
+#define ATH11K_GI_NUM           4
+#define ATH11K_HT_MCS_NUM       32
+
+enum ath11k_pkt_rx_err {
+	ATH11K_PKT_RX_ERR_FCS,
+	ATH11K_PKT_RX_ERR_TKIP,
+	ATH11K_PKT_RX_ERR_CRYPT,
+	ATH11K_PKT_RX_ERR_PEER_IDX_INVAL,
+	ATH11K_PKT_RX_ERR_MAX,
+};
+
+enum ath11k_ampdu_subfrm_num {
+	ATH11K_AMPDU_SUBFRM_NUM_10,
+	ATH11K_AMPDU_SUBFRM_NUM_20,
+	ATH11K_AMPDU_SUBFRM_NUM_30,
+	ATH11K_AMPDU_SUBFRM_NUM_40,
+	ATH11K_AMPDU_SUBFRM_NUM_50,
+	ATH11K_AMPDU_SUBFRM_NUM_60,
+	ATH11K_AMPDU_SUBFRM_NUM_MORE,
+	ATH11K_AMPDU_SUBFRM_NUM_MAX,
+};
+
+enum ath11k_amsdu_subfrm_num {
+	ATH11K_AMSDU_SUBFRM_NUM_1,
+	ATH11K_AMSDU_SUBFRM_NUM_2,
+	ATH11K_AMSDU_SUBFRM_NUM_3,
+	ATH11K_AMSDU_SUBFRM_NUM_4,
+	ATH11K_AMSDU_SUBFRM_NUM_MORE,
+	ATH11K_AMSDU_SUBFRM_NUM_MAX,
+};
+
+enum ath11k_counter_type {
+	ATH11K_COUNTER_TYPE_BYTES,
+	ATH11K_COUNTER_TYPE_PKTS,
+	ATH11K_COUNTER_TYPE_MAX,
+};
+
+enum ath11k_stats_type {
+	ATH11K_STATS_TYPE_SUCC,
+	ATH11K_STATS_TYPE_FAIL,
+	ATH11K_STATS_TYPE_RETRY,
+	ATH11K_STATS_TYPE_AMPDU,
+	ATH11K_STATS_TYPE_MAX,
+};
+
+struct ath11k_htt_data_stats {
+	u64 legacy[ATH11K_COUNTER_TYPE_MAX][ATH11K_LEGACY_NUM];
+	u64 ht[ATH11K_COUNTER_TYPE_MAX][ATH11K_HT_MCS_NUM];
+	u64 vht[ATH11K_COUNTER_TYPE_MAX][ATH11K_VHT_MCS_NUM];
+	u64 he[ATH11K_COUNTER_TYPE_MAX][ATH11K_HE_MCS_NUM];
+	u64 bw[ATH11K_COUNTER_TYPE_MAX][ATH11K_BW_NUM];
+	u64 nss[ATH11K_COUNTER_TYPE_MAX][ATH11K_NSS_NUM];
+	u64 gi[ATH11K_COUNTER_TYPE_MAX][ATH11K_GI_NUM];
+};
+
+struct ath11k_htt_tx_stats {
+	struct ath11k_htt_data_stats stats[ATH11K_STATS_TYPE_MAX];
+	u64 tx_duration;
+	u64 ba_fails;
+	u64 ack_fails;
+};
+
+struct ath11k_per_ppdu_tx_stats {
+	u16 succ_pkts;
+	u16 failed_pkts;
+	u16 retry_pkts;
+	u32 succ_bytes;
+	u32 failed_bytes;
+	u32 retry_bytes;
+};
+
+struct ath11k_sta {
+	struct ath11k_vif *arvif;
+
+	/* the following are protected by ar->data_lock */
+	u32 changed; /* IEEE80211_RC_* */
+	u32 bw;
+	u32 nss;
+	u32 smps;
+
+	struct work_struct update_wk;
+	struct ieee80211_tx_info tx_info;
+	struct rate_info txrate;
+	struct rate_info last_txrate;
+	u64 rx_duration;
+	u8 rssi_comb;
+	struct ath11k_htt_tx_stats *tx_stats;
+	struct ath11k_rx_peer_stats *rx_stats;
+};
+
+#define ATH11K_NUM_CHANS 41
+#define ATH11K_MAX_5G_CHAN 173
+
+enum ath11k_state {
+	ATH11K_STATE_OFF,
+	ATH11K_STATE_ON,
+	ATH11K_STATE_RESTARTING,
+	ATH11K_STATE_RESTARTED,
+	ATH11K_STATE_WEDGED,
+	/* Add other states as required */
+};
+
+/* Antenna noise floor */
+#define ATH11K_DEFAULT_NOISE_FLOOR -95
+
+struct ath11k_fw_stats {
+	struct dentry *debugfs_fwstats;
+	u32 pdev_id;
+	u32 stats_id;
+	struct list_head pdevs;
+	struct list_head vdevs;
+	struct list_head bcn;
+};
+
+struct ath11k_dbg_htt_stats {
+	u8 type;
+	u8 reset;
+	struct debug_htt_stats_req *stats_req;
+	/* protects shared stats req buffer */
+	spinlock_t lock;
+};
+
+struct ath11k_debug {
+	struct dentry *debugfs_pdev;
+	struct ath11k_dbg_htt_stats htt_stats;
+	u32 extd_tx_stats;
+	struct ath11k_fw_stats fw_stats;
+	struct completion fw_stats_complete;
+	bool fw_stats_done;
+	u32 extd_rx_stats;
+	u32 pktlog_filter;
+	u32 pktlog_mode;
+	u32 pktlog_peer_valid;
+	u8 pktlog_peer_addr[ETH_ALEN];
+};
+
+struct ath11k_per_peer_tx_stats {
+	u32 succ_bytes;
+	u32 retry_bytes;
+	u32 failed_bytes;
+	u16 succ_pkts;
+	u16 retry_pkts;
+	u16 failed_pkts;
+	u32 duration;
+	u8 ba_fails;
+	bool is_ampdu;
+};
+
+#define ATH11K_FLUSH_TIMEOUT (5 * HZ)
+
+struct ath11k_vdev_stop_status {
+	bool stop_in_progress;
+	u32  vdev_id;
+};
+
+struct ath11k {
+	struct ath11k_base *ab;
+	struct ath11k_pdev *pdev;
+	struct ieee80211_hw *hw;
+	struct ieee80211_ops *ops;
+	struct ath11k_pdev_wmi *wmi;
+	struct ath11k_pdev_dp dp;
+	u8 mac_addr[ETH_ALEN];
+	u32 ht_cap_info;
+	u32 vht_cap_info;
+	struct ath11k_he ar_he;
+	enum ath11k_state state;
+	struct {
+		struct completion started;
+		struct completion completed;
+		struct completion on_channel;
+		struct delayed_work timeout;
+		enum ath11k_scan_state state;
+		bool is_roc;
+		int vdev_id;
+		int roc_freq;
+		bool roc_notify;
+	} scan;
+
+	struct {
+		struct ieee80211_supported_band sbands[NUM_NL80211_BANDS];
+	} mac;
+	unsigned long dev_flags;
+	unsigned int filter_flags;
+	unsigned long monitor_flags;
+	u32 min_tx_power;
+	u32 max_tx_power;
+	u32 txpower_limit_2g;
+	u32 txpower_limit_5g;
+	u32 txpower_scale;
+	u32 power_scale;
+	u32 chan_tx_pwr;
+	u32 num_stations;
+	u32 max_num_stations;
+	bool monitor_present;
+	/* To synchronize concurrent synchronous mac80211 callback operations,
+	 * concurrent debugfs configuration and concurrent FW statistics events.
+	 */
+	struct mutex conf_mutex;
+	/* protects the radio specific data like debug stats, ppdu_stats_info stats,
+	 * vdev_stop_status info, scan data, ath11k_sta info, ath11k_vif info,
+	 * channel context data, survey info, test mode data.
+	 */
+	spinlock_t data_lock;
+
+	struct list_head arvifs;
+	/* should never be NULL; needed for regular htt rx */
+	struct ieee80211_channel *rx_channel;
+
+	/* valid during scan; needed for mgmt rx during scan */
+	struct ieee80211_channel *scan_channel;
+
+	u8 cfg_tx_chainmask;
+	u8 cfg_rx_chainmask;
+	u8 num_rx_chains;
+	u8 num_tx_chains;
+	/* pdev_idx starts from 0 whereas pdev->pdev_id starts with 1 */
+	u8 pdev_idx;
+	u8 lmac_id;
+
+	struct completion peer_assoc_done;
+
+	int install_key_status;
+	struct completion install_key_done;
+
+	int last_wmi_vdev_start_status;
+	struct ath11k_vdev_stop_status vdev_stop_status;
+	struct completion vdev_setup_done;
+
+	int num_peers;
+	int max_num_peers;
+	u32 num_started_vdevs;
+	u32 num_created_vdevs;
+
+	struct idr txmgmt_idr;
+	/* protects txmgmt_idr data */
+	spinlock_t txmgmt_idr_lock;
+	atomic_t num_pending_mgmt_tx;
+
+	/* cycle count is reported twice for each visited channel during scan.
+	 * access protected by data_lock
+	 */
+	u32 survey_last_rx_clear_count;
+	u32 survey_last_cycle_count;
+
+	/* Channel info events are expected to come in pairs without and with
+	 * COMPLETE flag set respectively for each channel visit during scan.
+	 *
+	 * However there are deviations from this rule. This flag is used to
+	 * avoid reporting garbage data.
+	 */
+	bool ch_info_can_report_survey;
+	struct survey_info survey[ATH11K_NUM_CHANS];
+	struct completion bss_survey_done;
+
+	struct work_struct regd_update_work;
+
+	struct work_struct wmi_mgmt_tx_work;
+	struct sk_buff_head wmi_mgmt_tx_queue;
+
+	struct ath11k_per_peer_tx_stats peer_tx_stats;
+	struct list_head ppdu_stats_info;
+	u32 ppdu_stat_list_depth;
+
+	struct ath11k_per_peer_tx_stats cached_stats;
+	u32 last_ppdu_id;
+	u32 cached_ppdu_id;
+#ifdef CONFIG_ATH11K_DEBUGFS
+	struct ath11k_debug debug;
+#endif
+	bool dfs_block_radar_events;
+};
+
+struct ath11k_band_cap {
+	u32 max_bw_supported;
+	u32 ht_cap_info;
+	u32 he_cap_info[2];
+	u32 he_mcs;
+	u32 he_cap_phy_info[PSOC_HOST_MAX_PHY_SIZE];
+	struct ath11k_ppe_threshold he_ppet;
+};
+
+struct ath11k_pdev_cap {
+	u32 supported_bands;
+	u32 ampdu_density;
+	u32 vht_cap;
+	u32 vht_mcs;
+	u32 he_mcs;
+	u32 tx_chain_mask;
+	u32 rx_chain_mask;
+	u32 tx_chain_mask_shift;
+	u32 rx_chain_mask_shift;
+	struct ath11k_band_cap band[NUM_NL80211_BANDS];
+};
+
+struct ath11k_pdev {
+	struct ath11k *ar;
+	u32 pdev_id;
+	struct ath11k_pdev_cap cap;
+	u8 mac_addr[ETH_ALEN];
+};
+
+struct ath11k_board_data {
+	const struct firmware *fw;
+	const void *data;
+	size_t len;
+};
+
+/* IPQ8074 HW channel counters frequency value in hertz */
+#define IPQ8074_CC_FREQ_HERTZ 320000
+
+struct ath11k_soc_dp_rx_stats {
+	u32 err_ring_pkts;
+	u32 invalid_rbm;
+	u32 rxdma_error[HAL_REO_ENTR_RING_RXDMA_ECODE_MAX];
+	u32 reo_error[HAL_REO_DEST_RING_ERROR_CODE_MAX];
+	u32 hal_reo_error[DP_REO_DST_RING_MAX];
+};
+
+/* Master structure to hold the hw data which may be used in core module */
+struct ath11k_base {
+	enum ath11k_hw_rev hw_rev;
+	struct platform_device *pdev;
+	struct device *dev;
+	struct ath11k_qmi qmi;
+	struct ath11k_wmi_base wmi_sc;
+	struct completion fw_ready;
+	struct rproc *tgt_rproc;
+	int num_radios;
+	/* HW channel counters frequency value in hertz common to all MACs */
+	u32 cc_freq_hz;
+
+	struct ath11k_htc htc;
+
+	struct ath11k_dp dp;
+
+	void __iomem *mem;
+	unsigned long mem_len;
+
+	const struct ath11k_hif_ops *hif_ops;
+
+	struct ath11k_ce ce;
+	struct timer_list rx_replenish_retry;
+	struct ath11k_hal hal;
+	/* To synchronize core_start/core_stop */
+	struct mutex core_lock;
+	/* Protects data like peers */
+	spinlock_t base_lock;
+	struct ath11k_pdev pdevs[MAX_RADIOS];
+	struct ath11k_pdev __rcu *pdevs_active[MAX_RADIOS];
+	struct ath11k_hal_reg_capabilities_ext hal_reg_cap[MAX_RADIOS];
+	unsigned long long free_vdev_map;
+	struct list_head peers;
+	wait_queue_head_t peer_mapping_wq;
+	u8 mac_addr[ETH_ALEN];
+	bool wmi_ready;
+	u32 wlan_init_status;
+	int irq_num[ATH11K_IRQ_NUM_MAX];
+	struct ath11k_ext_irq_grp ext_irq_grp[ATH11K_EXT_IRQ_GRP_NUM_MAX];
+	struct napi_struct *napi;
+	struct ath11k_targ_cap target_caps;
+	u32 ext_service_bitmap[WMI_SERVICE_EXT_BM_SIZE];
+	bool pdevs_macaddr_valid;
+	int bd_api;
+	struct ath11k_hw_params hw_params;
+	const struct firmware *cal_file;
+
+	/* Below regd's are protected by ab->data_lock */
+	/* This is the regd set for every radio
+	 * by the firmware during initializatin
+	 */
+	struct ieee80211_regdomain *default_regd[MAX_RADIOS];
+	/* This regd is set during dynamic country setting
+	 * This may or may not be used during the runtime
+	 */
+	struct ieee80211_regdomain *new_regd[MAX_RADIOS];
+
+	/* Current DFS Regulatory */
+	enum ath11k_dfs_region dfs_region;
+#ifdef CONFIG_ATH11K_DEBUGFS
+	struct dentry *debugfs_soc;
+	struct dentry *debugfs_ath11k;
+#endif
+	struct ath11k_soc_dp_rx_stats soc_stats;
+
+	unsigned long dev_flags;
+	struct completion driver_recovery;
+	struct workqueue_struct *workqueue;
+	struct work_struct restart_work;
+	struct {
+		/* protected by data_lock */
+		u32 fw_crash_counter;
+	} stats;
+};
+
+struct ath11k_fw_stats_pdev {
+	struct list_head list;
+
+	/* PDEV stats */
+	s32 ch_noise_floor;
+	/* Cycles spent transmitting frames */
+	u32 tx_frame_count;
+	/* Cycles spent receiving frames */
+	u32 rx_frame_count;
+	/* Total channel busy time, evidently */
+	u32 rx_clear_count;
+	/* Total on-channel time */
+	u32 cycle_count;
+	u32 phy_err_count;
+	u32 chan_tx_power;
+	u32 ack_rx_bad;
+	u32 rts_bad;
+	u32 rts_good;
+	u32 fcs_bad;
+	u32 no_beacons;
+	u32 mib_int_count;
+
+	/* PDEV TX stats */
+	/* Num HTT cookies queued to dispatch list */
+	s32 comp_queued;
+	/* Num HTT cookies dispatched */
+	s32 comp_delivered;
+	/* Num MSDU queued to WAL */
+	s32 msdu_enqued;
+	/* Num MPDU queue to WAL */
+	s32 mpdu_enqued;
+	/* Num MSDUs dropped by WMM limit */
+	s32 wmm_drop;
+	/* Num Local frames queued */
+	s32 local_enqued;
+	/* Num Local frames done */
+	s32 local_freed;
+	/* Num queued to HW */
+	s32 hw_queued;
+	/* Num PPDU reaped from HW */
+	s32 hw_reaped;
+	/* Num underruns */
+	s32 underrun;
+	/* Num PPDUs cleaned up in TX abort */
+	s32 tx_abort;
+	/* Num MPDUs requed by SW */
+	s32 mpdus_requed;
+	/* excessive retries */
+	u32 tx_ko;
+	/* data hw rate code */
+	u32 data_rc;
+	/* Scheduler self triggers */
+	u32 self_triggers;
+	/* frames dropped due to excessive sw retries */
+	u32 sw_retry_failure;
+	/* illegal rate phy errors	*/
+	u32 illgl_rate_phy_err;
+	/* wal pdev continuous xretry */
+	u32 pdev_cont_xretry;
+	/* wal pdev tx timeouts */
+	u32 pdev_tx_timeout;
+	/* wal pdev resets */
+	u32 pdev_resets;
+	/* frames dropped due to non-availability of stateless TIDs */
+	u32 stateless_tid_alloc_failure;
+	/* PhY/BB underrun */
+	u32 phy_underrun;
+	/* MPDU is more than txop limit */
+	u32 txop_ovf;
+
+	/* PDEV RX stats */
+	/* Cnts any change in ring routing mid-ppdu */
+	s32 mid_ppdu_route_change;
+	/* Total number of statuses processed */
+	s32 status_rcvd;
+	/* Extra frags on rings 0-3 */
+	s32 r0_frags;
+	s32 r1_frags;
+	s32 r2_frags;
+	s32 r3_frags;
+	/* MSDUs / MPDUs delivered to HTT */
+	s32 htt_msdus;
+	s32 htt_mpdus;
+	/* MSDUs / MPDUs delivered to local stack */
+	s32 loc_msdus;
+	s32 loc_mpdus;
+	/* AMSDUs that have more MSDUs than the status ring size */
+	s32 oversize_amsdu;
+	/* Number of PHY errors */
+	s32 phy_errs;
+	/* Number of PHY errors drops */
+	s32 phy_err_drop;
+	/* Number of mpdu errors - FCS, MIC, ENC etc. */
+	s32 mpdu_errs;
+};
+
+struct ath11k_fw_stats_vdev {
+	struct list_head list;
+
+	u32 vdev_id;
+	u32 beacon_snr;
+	u32 data_snr;
+	u32 num_tx_frames[WLAN_MAX_AC];
+	u32 num_rx_frames;
+	u32 num_tx_frames_retries[WLAN_MAX_AC];
+	u32 num_tx_frames_failures[WLAN_MAX_AC];
+	u32 num_rts_fail;
+	u32 num_rts_success;
+	u32 num_rx_err;
+	u32 num_rx_discard;
+	u32 num_tx_not_acked;
+	u32 tx_rate_history[MAX_TX_RATE_VALUES];
+	u32 beacon_rssi_history[MAX_TX_RATE_VALUES];
+};
+
+struct ath11k_fw_stats_bcn {
+	struct list_head list;
+
+	u32 vdev_id;
+	u32 tx_bcn_succ_cnt;
+	u32 tx_bcn_outage_cnt;
+};
+
+void ath11k_peer_unmap_event(struct ath11k_base *ab, u16 peer_id);
+void ath11k_peer_map_event(struct ath11k_base *ab, u8 vdev_id, u16 peer_id,
+			   u8 *mac_addr, u16 ast_hash);
+struct ath11k_peer *ath11k_peer_find(struct ath11k_base *ab, int vdev_id,
+				     const u8 *addr);
+struct ath11k_peer *ath11k_peer_find_by_addr(struct ath11k_base *ab,
+					     const u8 *addr);
+struct ath11k_peer *ath11k_peer_find_by_id(struct ath11k_base *ab, int peer_id);
+int ath11k_core_qmi_firmware_ready(struct ath11k_base *ab);
+int ath11k_core_init(struct ath11k_base *ath11k);
+void ath11k_core_deinit(struct ath11k_base *ath11k);
+struct ath11k_base *ath11k_core_alloc(struct device *dev);
+void ath11k_core_free(struct ath11k_base *ath11k);
+int ath11k_core_fetch_bdf(struct ath11k_base *ath11k,
+			  struct ath11k_board_data *bd);
+void ath11k_core_free_bdf(struct ath11k_base *ab, struct ath11k_board_data *bd);
+
+void ath11k_core_halt(struct ath11k *ar);
+u8 ath11k_core_get_hw_mac_id(struct ath11k_base *ab, int pdev_idx);
+
+static inline const char *ath11k_scan_state_str(enum ath11k_scan_state state)
+{
+	switch (state) {
+	case ATH11K_SCAN_IDLE:
+		return "idle";
+	case ATH11K_SCAN_STARTING:
+		return "starting";
+	case ATH11K_SCAN_RUNNING:
+		return "running";
+	case ATH11K_SCAN_ABORTING:
+		return "aborting";
+	}
+
+	return "unknown";
+}
+
+static inline struct ath11k_skb_cb *ATH11K_SKB_CB(struct sk_buff *skb)
+{
+	return (struct ath11k_skb_cb *)&IEEE80211_SKB_CB(skb)->driver_data;
+}
+
+static inline struct ath11k_skb_rxcb *ATH11K_SKB_RXCB(struct sk_buff *skb)
+{
+	BUILD_BUG_ON(sizeof(struct ath11k_skb_rxcb) > sizeof(skb->cb));
+	return (struct ath11k_skb_rxcb *)skb->cb;
+}
+
+static inline struct ath11k_vif *ath11k_vif_to_arvif(struct ieee80211_vif *vif)
+{
+	return (struct ath11k_vif *)vif->drv_priv;
+}
+
+#endif /* _CORE_H_ */
diff --git a/drivers/net/wireless/ath/ath11k/debug.c b/drivers/net/wireless/ath/ath11k/debug.c
new file mode 100644
index 000000000000..c27fffd13a5d
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/debug.c
@@ -0,0 +1,1060 @@
+// SPDX-License-Identifier: BSD-3-Clause-Clear
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#include <linux/vmalloc.h>
+#include "core.h"
+#include "debug.h"
+#include "wmi.h"
+#include "hal_rx.h"
+#include "dp_tx.h"
+#include "debug_htt_stats.h"
+#include "peer.h"
+
+void ath11k_info(struct ath11k_base *ab, const char *fmt, ...)
+{
+	struct va_format vaf = {
+		.fmt = fmt,
+	};
+	va_list args;
+
+	va_start(args, fmt);
+	vaf.va = &args;
+	dev_info(ab->dev, "%pV", &vaf);
+	/* TODO: Trace the log */
+	va_end(args);
+}
+
+void ath11k_err(struct ath11k_base *ab, const char *fmt, ...)
+{
+	struct va_format vaf = {
+		.fmt = fmt,
+	};
+	va_list args;
+
+	va_start(args, fmt);
+	vaf.va = &args;
+	dev_err(ab->dev, "%pV", &vaf);
+	/* TODO: Trace the log */
+	va_end(args);
+}
+
+void ath11k_warn(struct ath11k_base *ab, const char *fmt, ...)
+{
+	struct va_format vaf = {
+		.fmt = fmt,
+	};
+	va_list args;
+
+	va_start(args, fmt);
+	vaf.va = &args;
+	dev_warn_ratelimited(ab->dev, "%pV", &vaf);
+	/* TODO: Trace the log */
+	va_end(args);
+}
+
+#ifdef CONFIG_ATH11K_DEBUG
+void __ath11k_dbg(struct ath11k_base *ab, enum ath11k_debug_mask mask,
+		  const char *fmt, ...)
+{
+	struct va_format vaf;
+	va_list args;
+
+	va_start(args, fmt);
+
+	vaf.fmt = fmt;
+	vaf.va = &args;
+
+	if (ath11k_debug_mask & mask)
+		dev_printk(KERN_DEBUG, ab->dev, "%pV", &vaf);
+
+	/* TODO: trace log */
+
+	va_end(args);
+}
+
+void ath11k_dbg_dump(struct ath11k_base *ab,
+		     enum ath11k_debug_mask mask,
+		     const char *msg, const char *prefix,
+		     const void *buf, size_t len)
+{
+	char linebuf[256];
+	size_t linebuflen;
+	const void *ptr;
+
+	if (ath11k_debug_mask & mask) {
+		if (msg)
+			__ath11k_dbg(ab, mask, "%s\n", msg);
+
+		for (ptr = buf; (ptr - buf) < len; ptr += 16) {
+			linebuflen = 0;
+			linebuflen += scnprintf(linebuf + linebuflen,
+						sizeof(linebuf) - linebuflen,
+						"%s%08x: ",
+						(prefix ? prefix : ""),
+						(unsigned int)(ptr - buf));
+			hex_dump_to_buffer(ptr, len - (ptr - buf), 16, 1,
+					   linebuf + linebuflen,
+					   sizeof(linebuf) - linebuflen, true);
+			dev_printk(KERN_DEBUG, ab->dev, "%s\n", linebuf);
+		}
+	}
+}
+
+#endif
+
+#ifdef CONFIG_ATH11K_DEBUGFS
+static void ath11k_fw_stats_pdevs_free(struct list_head *head)
+{
+	struct ath11k_fw_stats_pdev *i, *tmp;
+
+	list_for_each_entry_safe(i, tmp, head, list) {
+		list_del(&i->list);
+		kfree(i);
+	}
+}
+
+static void ath11k_fw_stats_vdevs_free(struct list_head *head)
+{
+	struct ath11k_fw_stats_vdev *i, *tmp;
+
+	list_for_each_entry_safe(i, tmp, head, list) {
+		list_del(&i->list);
+		kfree(i);
+	}
+}
+
+static void ath11k_fw_stats_bcn_free(struct list_head *head)
+{
+	struct ath11k_fw_stats_bcn *i, *tmp;
+
+	list_for_each_entry_safe(i, tmp, head, list) {
+		list_del(&i->list);
+		kfree(i);
+	}
+}
+
+static void ath11k_debug_fw_stats_reset(struct ath11k *ar)
+{
+	spin_lock_bh(&ar->data_lock);
+	ar->debug.fw_stats_done = false;
+	ath11k_fw_stats_pdevs_free(&ar->debug.fw_stats.pdevs);
+	ath11k_fw_stats_vdevs_free(&ar->debug.fw_stats.vdevs);
+	spin_unlock_bh(&ar->data_lock);
+}
+
+void ath11k_debug_fw_stats_process(struct ath11k_base *ab, struct sk_buff *skb)
+{
+	struct ath11k_fw_stats stats = {};
+	struct ath11k *ar;
+	struct ath11k_pdev *pdev;
+	bool is_end;
+	static unsigned int num_vdev, num_bcn;
+	size_t total_vdevs_started = 0;
+	int i, ret;
+
+	INIT_LIST_HEAD(&stats.pdevs);
+	INIT_LIST_HEAD(&stats.vdevs);
+	INIT_LIST_HEAD(&stats.bcn);
+
+	ret = ath11k_wmi_pull_fw_stats(ab, skb, &stats);
+	if (ret) {
+		ath11k_warn(ab, "failed to pull fw stats: %d\n", ret);
+		goto free;
+	}
+
+	rcu_read_lock();
+	ar = ath11k_mac_get_ar_by_pdev_id(ab, stats.pdev_id);
+	if (!ar) {
+		rcu_read_unlock();
+		ath11k_warn(ab, "failed to get ar for pdev_id %d: %d\n",
+			    stats.pdev_id, ret);
+		goto free;
+	}
+
+	spin_lock_bh(&ar->data_lock);
+
+	if (stats.stats_id == WMI_REQUEST_PDEV_STAT) {
+		list_splice_tail_init(&stats.pdevs, &ar->debug.fw_stats.pdevs);
+		ar->debug.fw_stats_done = true;
+		goto complete;
+	}
+
+	if (stats.stats_id == WMI_REQUEST_VDEV_STAT) {
+		if (list_empty(&stats.vdevs)) {
+			ath11k_warn(ab, "empty vdev stats");
+			goto complete;
+		}
+		/* FW sends all the active VDEV stats irrespective of PDEV,
+		 * hence limit until the count of all VDEVs started
+		 */
+		for (i = 0; i < ab->num_radios; i++) {
+			pdev = rcu_dereference(ab->pdevs_active[i]);
+			if (pdev && pdev->ar)
+				total_vdevs_started += ar->num_started_vdevs;
+		}
+
+		is_end = ((++num_vdev) == total_vdevs_started ? true : false);
+
+		list_splice_tail_init(&stats.vdevs,
+				      &ar->debug.fw_stats.vdevs);
+
+		if (is_end) {
+			ar->debug.fw_stats_done = true;
+			num_vdev = 0;
+		}
+		goto complete;
+	}
+
+	if (stats.stats_id == WMI_REQUEST_BCN_STAT) {
+		if (list_empty(&stats.bcn)) {
+			ath11k_warn(ab, "empty bcn stats");
+			goto complete;
+		}
+		/* Mark end until we reached the count of all started VDEVs
+		 * within the PDEV
+		 */
+		is_end = ((++num_bcn) == ar->num_started_vdevs ? true : false);
+
+		list_splice_tail_init(&stats.bcn,
+				      &ar->debug.fw_stats.bcn);
+
+		if (is_end) {
+			ar->debug.fw_stats_done = true;
+			num_bcn = 0;
+		}
+	}
+complete:
+	complete(&ar->debug.fw_stats_complete);
+	rcu_read_unlock();
+	spin_unlock_bh(&ar->data_lock);
+
+free:
+	ath11k_fw_stats_pdevs_free(&stats.pdevs);
+	ath11k_fw_stats_vdevs_free(&stats.vdevs);
+	ath11k_fw_stats_bcn_free(&stats.bcn);
+}
+
+static int ath11k_debug_fw_stats_request(struct ath11k *ar,
+					 struct stats_request_params *req_param)
+{
+	struct ath11k_base *ab = ar->ab;
+	unsigned long timeout, time_left;
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	/* FW stats can get split when exceeding the stats data buffer limit.
+	 * In that case, since there is no end marking for the back-to-back
+	 * received 'update stats' event, we keep a 3 seconds timeout in case,
+	 * fw_stats_done is not marked yet
+	 */
+	timeout = jiffies + msecs_to_jiffies(3 * HZ);
+
+	ath11k_debug_fw_stats_reset(ar);
+
+	reinit_completion(&ar->debug.fw_stats_complete);
+
+	ret = ath11k_wmi_send_stats_request_cmd(ar, req_param);
+
+	if (ret) {
+		ath11k_warn(ab, "could not request fw stats (%d)\n",
+			    ret);
+		return ret;
+	}
+
+	time_left =
+	wait_for_completion_timeout(&ar->debug.fw_stats_complete,
+				    1 * HZ);
+	if (!time_left)
+		return -ETIMEDOUT;
+
+	for (;;) {
+		if (time_after(jiffies, timeout))
+			break;
+
+		spin_lock_bh(&ar->data_lock);
+		if (ar->debug.fw_stats_done) {
+			spin_unlock_bh(&ar->data_lock);
+			break;
+		}
+		spin_unlock_bh(&ar->data_lock);
+	}
+	return 0;
+}
+
+static int ath11k_open_pdev_stats(struct inode *inode, struct file *file)
+{
+	struct ath11k *ar = inode->i_private;
+	struct ath11k_base *ab = ar->ab;
+	struct stats_request_params req_param;
+	void *buf = NULL;
+	int ret;
+
+	mutex_lock(&ar->conf_mutex);
+
+	if (ar->state != ATH11K_STATE_ON) {
+		ret = -ENETDOWN;
+		goto err_unlock;
+	}
+
+	buf = vmalloc(ATH11K_FW_STATS_BUF_SIZE);
+	if (!buf) {
+		ret = -ENOMEM;
+		goto err_unlock;
+	}
+
+	req_param.pdev_id = ar->pdev->pdev_id;
+	req_param.vdev_id = 0;
+	req_param.stats_id = WMI_REQUEST_PDEV_STAT;
+
+	ret = ath11k_debug_fw_stats_request(ar, &req_param);
+	if (ret) {
+		ath11k_warn(ab, "failed to request fw pdev stats: %d\n", ret);
+		goto err_free;
+	}
+
+	ath11k_wmi_fw_stats_fill(ar, &ar->debug.fw_stats, req_param.stats_id,
+				 buf);
+
+	file->private_data = buf;
+
+	mutex_unlock(&ar->conf_mutex);
+	return 0;
+
+err_free:
+	vfree(buf);
+
+err_unlock:
+	mutex_unlock(&ar->conf_mutex);
+	return ret;
+}
+
+static int ath11k_release_pdev_stats(struct inode *inode, struct file *file)
+{
+	vfree(file->private_data);
+
+	return 0;
+}
+
+static ssize_t ath11k_read_pdev_stats(struct file *file,
+				      char __user *user_buf,
+				      size_t count, loff_t *ppos)
+{
+	const char *buf = file->private_data;
+	size_t len = strlen(buf);
+
+	return simple_read_from_buffer(user_buf, count, ppos, buf, len);
+}
+
+static const struct file_operations fops_pdev_stats = {
+	.open = ath11k_open_pdev_stats,
+	.release = ath11k_release_pdev_stats,
+	.read = ath11k_read_pdev_stats,
+	.owner = THIS_MODULE,
+	.llseek = default_llseek,
+};
+
+static int ath11k_open_vdev_stats(struct inode *inode, struct file *file)
+{
+	struct ath11k *ar = inode->i_private;
+	struct stats_request_params req_param;
+	void *buf = NULL;
+	int ret;
+
+	mutex_lock(&ar->conf_mutex);
+
+	if (ar->state != ATH11K_STATE_ON) {
+		ret = -ENETDOWN;
+		goto err_unlock;
+	}
+
+	buf = vmalloc(ATH11K_FW_STATS_BUF_SIZE);
+	if (!buf) {
+		ret = -ENOMEM;
+		goto err_unlock;
+	}
+
+	req_param.pdev_id = ar->pdev->pdev_id;
+	/* VDEV stats is always sent for all active VDEVs from FW */
+	req_param.vdev_id = 0;
+	req_param.stats_id = WMI_REQUEST_VDEV_STAT;
+
+	ret = ath11k_debug_fw_stats_request(ar, &req_param);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to request fw vdev stats: %d\n", ret);
+		goto err_free;
+	}
+
+	ath11k_wmi_fw_stats_fill(ar, &ar->debug.fw_stats, req_param.stats_id,
+				 buf);
+
+	file->private_data = buf;
+
+	mutex_unlock(&ar->conf_mutex);
+	return 0;
+
+err_free:
+	vfree(buf);
+
+err_unlock:
+	mutex_unlock(&ar->conf_mutex);
+	return ret;
+}
+
+static int ath11k_release_vdev_stats(struct inode *inode, struct file *file)
+{
+	vfree(file->private_data);
+
+	return 0;
+}
+
+static ssize_t ath11k_read_vdev_stats(struct file *file,
+				      char __user *user_buf,
+				      size_t count, loff_t *ppos)
+{
+	const char *buf = file->private_data;
+	size_t len = strlen(buf);
+
+	return simple_read_from_buffer(user_buf, count, ppos, buf, len);
+}
+
+static const struct file_operations fops_vdev_stats = {
+	.open = ath11k_open_vdev_stats,
+	.release = ath11k_release_vdev_stats,
+	.read = ath11k_read_vdev_stats,
+	.owner = THIS_MODULE,
+	.llseek = default_llseek,
+};
+
+static int ath11k_open_bcn_stats(struct inode *inode, struct file *file)
+{
+	struct ath11k *ar = inode->i_private;
+	struct ath11k_vif *arvif;
+	struct stats_request_params req_param;
+	void *buf = NULL;
+	int ret;
+
+	mutex_lock(&ar->conf_mutex);
+
+	if (ar->state != ATH11K_STATE_ON) {
+		ret = -ENETDOWN;
+		goto err_unlock;
+	}
+
+	buf = vmalloc(ATH11K_FW_STATS_BUF_SIZE);
+	if (!buf) {
+		ret = -ENOMEM;
+		goto err_unlock;
+	}
+
+	req_param.stats_id = WMI_REQUEST_BCN_STAT;
+	req_param.pdev_id = ar->pdev->pdev_id;
+
+	/* loop all active VDEVs for bcn stats */
+	list_for_each_entry(arvif, &ar->arvifs, list) {
+		if (!arvif->is_up)
+			continue;
+
+		req_param.vdev_id = arvif->vdev_id;
+		ret = ath11k_debug_fw_stats_request(ar, &req_param);
+		if (ret) {
+			ath11k_warn(ar->ab, "failed to request fw bcn stats: %d\n", ret);
+			goto err_free;
+		}
+	}
+
+	ath11k_wmi_fw_stats_fill(ar, &ar->debug.fw_stats, req_param.stats_id,
+				 buf);
+
+	/* since beacon stats request is looped for all active VDEVs, saved fw
+	 * stats is not freed for each request until done for all active VDEVs
+	 */
+	spin_lock_bh(&ar->data_lock);
+	ath11k_fw_stats_bcn_free(&ar->debug.fw_stats.bcn);
+	spin_unlock_bh(&ar->data_lock);
+
+	file->private_data = buf;
+
+	mutex_unlock(&ar->conf_mutex);
+	return 0;
+
+err_free:
+	vfree(buf);
+
+err_unlock:
+	mutex_unlock(&ar->conf_mutex);
+	return ret;
+}
+
+static int ath11k_release_bcn_stats(struct inode *inode, struct file *file)
+{
+	vfree(file->private_data);
+
+	return 0;
+}
+
+static ssize_t ath11k_read_bcn_stats(struct file *file,
+				     char __user *user_buf,
+				     size_t count, loff_t *ppos)
+{
+	const char *buf = file->private_data;
+	size_t len = strlen(buf);
+
+	return simple_read_from_buffer(user_buf, count, ppos, buf, len);
+}
+
+static const struct file_operations fops_bcn_stats = {
+	.open = ath11k_open_bcn_stats,
+	.release = ath11k_release_bcn_stats,
+	.read = ath11k_read_bcn_stats,
+	.owner = THIS_MODULE,
+	.llseek = default_llseek,
+};
+
+static ssize_t ath11k_read_simulate_fw_crash(struct file *file,
+					     char __user *user_buf,
+					     size_t count, loff_t *ppos)
+{
+	const char buf[] =
+		"To simulate firmware crash write one of the keywords to this file:\n"
+		"`assert` - this will send WMI_FORCE_FW_HANG_CMDID to firmware to cause assert.\n"
+		"`hw-restart` - this will simply queue hw restart without fw/hw actually crashing.\n";
+
+	return simple_read_from_buffer(user_buf, count, ppos, buf, strlen(buf));
+}
+
+/* Simulate firmware crash:
+ * 'soft': Call wmi command causing firmware hang. This firmware hang is
+ * recoverable by warm firmware reset.
+ * 'hard': Force firmware crash by setting any vdev parameter for not allowed
+ * vdev id. This is hard firmware crash because it is recoverable only by cold
+ * firmware reset.
+ */
+static ssize_t ath11k_write_simulate_fw_crash(struct file *file,
+					      const char __user *user_buf,
+					      size_t count, loff_t *ppos)
+{
+	struct ath11k_base *ab = file->private_data;
+	struct ath11k_pdev *pdev;
+	struct ath11k *ar = ab->pdevs[0].ar;
+	char buf[32] = {0};
+	ssize_t rc;
+	int i, ret, radioup;
+
+	for (i = 0; i < ab->num_radios; i++) {
+		pdev = &ab->pdevs[i];
+		ar = pdev->ar;
+		if (ar && ar->state == ATH11K_STATE_ON) {
+			radioup = 1;
+			break;
+		}
+	}
+	/* filter partial writes and invalid commands */
+	if (*ppos != 0 || count >= sizeof(buf) || count == 0)
+		return -EINVAL;
+
+	rc = simple_write_to_buffer(buf, sizeof(buf) - 1, ppos, user_buf, count);
+	if (rc < 0)
+		return rc;
+
+	/* drop the possible '\n' from the end */
+	if (buf[*ppos - 1] == '\n')
+		buf[*ppos - 1] = '\0';
+
+	if (radioup == 0) {
+		ret = -ENETDOWN;
+		goto exit;
+	}
+
+	if (!strcmp(buf, "assert")) {
+		ath11k_info(ab, "simulating firmware assert crash\n");
+		ret = ath11k_wmi_force_fw_hang_cmd(ar,
+						   ATH11K_WMI_FW_HANG_ASSERT_TYPE,
+						   ATH11K_WMI_FW_HANG_DELAY);
+	} else {
+		ret = -EINVAL;
+		goto exit;
+	}
+
+	if (ret) {
+		ath11k_warn(ab, "failed to simulate firmware crash: %d\n", ret);
+		goto exit;
+	}
+
+	ret = count;
+
+exit:
+	return ret;
+}
+
+static const struct file_operations fops_simulate_fw_crash = {
+	.read = ath11k_read_simulate_fw_crash,
+	.write = ath11k_write_simulate_fw_crash,
+	.open = simple_open,
+	.owner = THIS_MODULE,
+	.llseek = default_llseek,
+};
+
+static ssize_t ath11k_write_enable_extd_tx_stats(struct file *file,
+						 const char __user *ubuf,
+						 size_t count, loff_t *ppos)
+{
+	struct ath11k *ar = file->private_data;
+	u32 filter;
+	int ret;
+
+	if (kstrtouint_from_user(ubuf, count, 0, &filter))
+		return -EINVAL;
+
+	mutex_lock(&ar->conf_mutex);
+
+	if (ar->state != ATH11K_STATE_ON) {
+		ret = -ENETDOWN;
+		goto out;
+	}
+
+	if (filter == ar->debug.extd_tx_stats) {
+		ret = count;
+		goto out;
+	}
+
+	ar->debug.extd_tx_stats = filter;
+	ret = count;
+
+out:
+	mutex_unlock(&ar->conf_mutex);
+	return ret;
+}
+
+static ssize_t ath11k_read_enable_extd_tx_stats(struct file *file,
+						char __user *ubuf,
+						size_t count, loff_t *ppos)
+
+{
+	char buf[32] = {0};
+	struct ath11k *ar = file->private_data;
+	int len = 0;
+
+	mutex_lock(&ar->conf_mutex);
+	len = scnprintf(buf, sizeof(buf) - len, "%08x\n",
+			ar->debug.extd_tx_stats);
+	mutex_unlock(&ar->conf_mutex);
+
+	return simple_read_from_buffer(ubuf, count, ppos, buf, len);
+}
+
+static const struct file_operations fops_extd_tx_stats = {
+	.read = ath11k_read_enable_extd_tx_stats,
+	.write = ath11k_write_enable_extd_tx_stats,
+	.open = simple_open
+};
+
+static ssize_t ath11k_write_extd_rx_stats(struct file *file,
+					  const char __user *ubuf,
+					  size_t count, loff_t *ppos)
+{
+	struct ath11k *ar = file->private_data;
+	struct htt_rx_ring_tlv_filter tlv_filter = {0};
+	u32 enable, rx_filter = 0, ring_id;
+	int ret;
+
+	if (kstrtouint_from_user(ubuf, count, 0, &enable))
+		return -EINVAL;
+
+	mutex_lock(&ar->conf_mutex);
+
+	if (ar->state != ATH11K_STATE_ON) {
+		ret = -ENETDOWN;
+		goto exit;
+	}
+
+	if (enable > 1) {
+		ret = -EINVAL;
+		goto exit;
+	}
+
+	if (enable == ar->debug.extd_rx_stats) {
+		ret = count;
+		goto exit;
+	}
+
+	if (enable) {
+		rx_filter =  HTT_RX_FILTER_TLV_FLAGS_MPDU_START;
+		rx_filter |= HTT_RX_FILTER_TLV_FLAGS_PPDU_START;
+		rx_filter |= HTT_RX_FILTER_TLV_FLAGS_PPDU_END;
+		rx_filter |= HTT_RX_FILTER_TLV_FLAGS_PPDU_END_USER_STATS;
+		rx_filter |= HTT_RX_FILTER_TLV_FLAGS_PPDU_END_USER_STATS_EXT;
+		rx_filter |= HTT_RX_FILTER_TLV_FLAGS_PPDU_END_STATUS_DONE;
+
+		tlv_filter.rx_filter = rx_filter;
+		tlv_filter.pkt_filter_flags0 = HTT_RX_FP_MGMT_FILTER_FLAGS0;
+		tlv_filter.pkt_filter_flags1 = HTT_RX_FP_MGMT_FILTER_FLAGS1;
+		tlv_filter.pkt_filter_flags2 = HTT_RX_FP_CTRL_FILTER_FLASG2;
+		tlv_filter.pkt_filter_flags3 = HTT_RX_FP_CTRL_FILTER_FLASG3 |
+			HTT_RX_FP_DATA_FILTER_FLASG3;
+	} else {
+		tlv_filter = ath11k_mac_mon_status_filter_default;
+	}
+
+	ring_id = ar->dp.rx_mon_status_refill_ring.refill_buf_ring.ring_id;
+	ret = ath11k_dp_tx_htt_rx_filter_setup(ar->ab, ring_id, ar->dp.mac_id,
+					       HAL_RXDMA_MONITOR_STATUS,
+					       DP_RX_BUFFER_SIZE, &tlv_filter);
+
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to set rx filter for moniter status ring\n");
+		goto exit;
+	}
+
+	ar->debug.extd_rx_stats = enable;
+	ret = count;
+exit:
+	mutex_unlock(&ar->conf_mutex);
+	return ret;
+}
+
+static ssize_t ath11k_read_extd_rx_stats(struct file *file,
+					 char __user *ubuf,
+					 size_t count, loff_t *ppos)
+{
+	struct ath11k *ar = file->private_data;
+	char buf[32];
+	int len = 0;
+
+	mutex_lock(&ar->conf_mutex);
+	len = scnprintf(buf, sizeof(buf) - len, "%d\n",
+			ar->debug.extd_rx_stats);
+	mutex_unlock(&ar->conf_mutex);
+
+	return simple_read_from_buffer(ubuf, count, ppos, buf, len);
+}
+
+static const struct file_operations fops_extd_rx_stats = {
+	.read = ath11k_read_extd_rx_stats,
+	.write = ath11k_write_extd_rx_stats,
+	.open = simple_open,
+};
+
+static ssize_t ath11k_debug_dump_soc_rx_stats(struct file *file,
+					      char __user *user_buf,
+					      size_t count, loff_t *ppos)
+{
+	struct ath11k_base *ab = file->private_data;
+	struct ath11k_soc_dp_rx_stats *soc_stats = &ab->soc_stats;
+	int len = 0, i, retval;
+	const int size = 4096;
+	static const char *rxdma_err[HAL_REO_ENTR_RING_RXDMA_ECODE_MAX] = {
+			"Overflow", "MPDU len", "FCS", "Decrypt", "TKIP MIC",
+			"Unencrypt", "MSDU len", "MSDU limit", "WiFi parse",
+			"AMSDU parse", "SA timeout", "DA timeout",
+			"Flow timeout", "Flush req"};
+	static const char *reo_err[HAL_REO_DEST_RING_ERROR_CODE_MAX] = {
+			"Desc addr zero", "Desc inval", "AMPDU in non BA",
+			"Non BA dup", "BA dup", "Frame 2k jump", "BAR 2k jump",
+			"Frame OOR", "BAR OOR", "No BA session",
+			"Frame SN equal SSN", "PN check fail", "2k err",
+			"PN err", "Desc blocked"};
+
+	char *buf;
+
+	buf = kzalloc(size, GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+
+	len += scnprintf(buf + len, size - len, "SOC RX STATS:\n\n");
+	len += scnprintf(buf + len, size - len, "err ring pkts: %u\n",
+			 soc_stats->err_ring_pkts);
+	len += scnprintf(buf + len, size - len, "Invalid RBM: %u\n\n",
+			 soc_stats->invalid_rbm);
+	len += scnprintf(buf + len, size - len, "RXDMA errors:\n");
+	for (i = 0; i < HAL_REO_ENTR_RING_RXDMA_ECODE_MAX; i++)
+		len += scnprintf(buf + len, size - len, "%s: %u\n",
+				 rxdma_err[i], soc_stats->rxdma_error[i]);
+
+	len += scnprintf(buf + len, size - len, "\nREO errors:\n");
+	for (i = 0; i < HAL_REO_DEST_RING_ERROR_CODE_MAX; i++)
+		len += scnprintf(buf + len, size - len, "%s: %u\n",
+				 reo_err[i], soc_stats->reo_error[i]);
+
+	len += scnprintf(buf + len, size - len, "\nHAL REO errors:\n");
+	len += scnprintf(buf + len, size - len,
+			 "ring0: %u\nring1: %u\nring2: %u\nring3: %u\n",
+			 soc_stats->hal_reo_error[0],
+			 soc_stats->hal_reo_error[1],
+			 soc_stats->hal_reo_error[2],
+			 soc_stats->hal_reo_error[3]);
+
+	if (len > size)
+		len = size;
+	retval = simple_read_from_buffer(user_buf, count, ppos, buf, len);
+	kfree(buf);
+
+	return retval;
+}
+
+static const struct file_operations fops_soc_rx_stats = {
+	.read = ath11k_debug_dump_soc_rx_stats,
+	.open = simple_open,
+	.owner = THIS_MODULE,
+	.llseek = default_llseek,
+};
+
+int ath11k_debug_pdev_create(struct ath11k_base *ab)
+{
+	ab->debugfs_soc = debugfs_create_dir(ab->hw_params.name, ab->debugfs_ath11k);
+
+	if (IS_ERR_OR_NULL(ab->debugfs_soc)) {
+		if (IS_ERR(ab->debugfs_soc))
+			return PTR_ERR(ab->debugfs_soc);
+		return -ENOMEM;
+	}
+
+	debugfs_create_file("simulate_fw_crash", 0600, ab->debugfs_soc, ab,
+			    &fops_simulate_fw_crash);
+
+	debugfs_create_file("soc_rx_stats", 0600, ab->debugfs_soc, ab,
+			    &fops_soc_rx_stats);
+
+	return 0;
+}
+
+void ath11k_debug_pdev_destroy(struct ath11k_base *ab)
+{
+	debugfs_remove_recursive(ab->debugfs_ath11k);
+	ab->debugfs_ath11k = NULL;
+}
+
+int ath11k_debug_soc_create(struct ath11k_base *ab)
+{
+	ab->debugfs_ath11k = debugfs_create_dir("ath11k", NULL);
+
+	if (IS_ERR_OR_NULL(ab->debugfs_ath11k)) {
+		if (IS_ERR(ab->debugfs_ath11k))
+			return PTR_ERR(ab->debugfs_ath11k);
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+void ath11k_debug_soc_destroy(struct ath11k_base *ab)
+{
+	debugfs_remove_recursive(ab->debugfs_soc);
+	ab->debugfs_soc = NULL;
+}
+
+void ath11k_debug_fw_stats_init(struct ath11k *ar)
+{
+	struct dentry *fwstats_dir = debugfs_create_dir("fw_stats",
+							ar->debug.debugfs_pdev);
+
+	ar->debug.fw_stats.debugfs_fwstats = fwstats_dir;
+
+	/* all stats debugfs files created are under "fw_stats" directory
+	 * created per PDEV
+	 */
+	debugfs_create_file("pdev_stats", 0600, fwstats_dir, ar,
+			    &fops_pdev_stats);
+	debugfs_create_file("vdev_stats", 0600, fwstats_dir, ar,
+			    &fops_vdev_stats);
+	debugfs_create_file("beacon_stats", 0600, fwstats_dir, ar,
+			    &fops_bcn_stats);
+
+	INIT_LIST_HEAD(&ar->debug.fw_stats.pdevs);
+	INIT_LIST_HEAD(&ar->debug.fw_stats.vdevs);
+	INIT_LIST_HEAD(&ar->debug.fw_stats.bcn);
+
+	init_completion(&ar->debug.fw_stats_complete);
+}
+
+static ssize_t ath11k_write_pktlog_filter(struct file *file,
+					  const char __user *ubuf,
+					  size_t count, loff_t *ppos)
+{
+	struct ath11k *ar = file->private_data;
+	struct htt_rx_ring_tlv_filter tlv_filter = {0};
+	u32 rx_filter = 0, ring_id, filter, mode;
+	u8 buf[128] = {0};
+	int ret;
+	ssize_t rc;
+
+	mutex_lock(&ar->conf_mutex);
+	if (ar->state != ATH11K_STATE_ON) {
+		ret = -ENETDOWN;
+		goto out;
+	}
+
+	rc = simple_write_to_buffer(buf, sizeof(buf) - 1, ppos, ubuf, count);
+	if (rc < 0) {
+		ret = rc;
+		goto out;
+	}
+	buf[rc] = '\0';
+
+	ret = sscanf(buf, "0x%x %u", &filter, &mode);
+	if (ret != 2) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	if (filter) {
+		ret = ath11k_wmi_pdev_pktlog_enable(ar, filter);
+		if (ret) {
+			ath11k_warn(ar->ab,
+				    "failed to enable pktlog filter %x: %d\n",
+				    ar->debug.pktlog_filter, ret);
+			goto out;
+		}
+	} else {
+		ret = ath11k_wmi_pdev_pktlog_disable(ar);
+		if (ret) {
+			ath11k_warn(ar->ab, "failed to disable pktlog: %d\n", ret);
+			goto out;
+		}
+	}
+
+#define HTT_RX_FILTER_TLV_LITE_MODE \
+			(HTT_RX_FILTER_TLV_FLAGS_PPDU_START | \
+			HTT_RX_FILTER_TLV_FLAGS_PPDU_END | \
+			HTT_RX_FILTER_TLV_FLAGS_PPDU_END_USER_STATS | \
+			HTT_RX_FILTER_TLV_FLAGS_PPDU_END_USER_STATS_EXT | \
+			HTT_RX_FILTER_TLV_FLAGS_PPDU_END_STATUS_DONE | \
+			HTT_RX_FILTER_TLV_FLAGS_MPDU_START)
+
+	if (mode == ATH11K_PKTLOG_MODE_FULL) {
+		rx_filter = HTT_RX_FILTER_TLV_LITE_MODE |
+			    HTT_RX_FILTER_TLV_FLAGS_MSDU_START |
+			    HTT_RX_FILTER_TLV_FLAGS_MSDU_END |
+			    HTT_RX_FILTER_TLV_FLAGS_MPDU_END |
+			    HTT_RX_FILTER_TLV_FLAGS_PACKET_HEADER |
+			    HTT_RX_FILTER_TLV_FLAGS_ATTENTION;
+	} else if (mode == ATH11K_PKTLOG_MODE_LITE) {
+		rx_filter = HTT_RX_FILTER_TLV_LITE_MODE;
+	}
+
+	tlv_filter.rx_filter = rx_filter;
+	if (rx_filter) {
+		tlv_filter.pkt_filter_flags0 = HTT_RX_FP_MGMT_FILTER_FLAGS0;
+		tlv_filter.pkt_filter_flags1 = HTT_RX_FP_MGMT_FILTER_FLAGS1;
+		tlv_filter.pkt_filter_flags2 = HTT_RX_FP_CTRL_FILTER_FLASG2;
+		tlv_filter.pkt_filter_flags3 = HTT_RX_FP_CTRL_FILTER_FLASG3 |
+					       HTT_RX_FP_DATA_FILTER_FLASG3;
+	}
+
+	ring_id = ar->dp.rx_mon_status_refill_ring.refill_buf_ring.ring_id;
+	ret = ath11k_dp_tx_htt_rx_filter_setup(ar->ab, ring_id, ar->dp.mac_id,
+					       HAL_RXDMA_MONITOR_STATUS,
+					       DP_RX_BUFFER_SIZE, &tlv_filter);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to set rx filter for moniter status ring\n");
+		goto out;
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI, "pktlog filter %d mode %s\n",
+		   filter, ((mode == ATH11K_PKTLOG_MODE_FULL) ? "full" : "lite"));
+
+	ar->debug.pktlog_filter = filter;
+	ar->debug.pktlog_mode = mode;
+	ret = count;
+
+out:
+	mutex_unlock(&ar->conf_mutex);
+	return ret;
+}
+
+static ssize_t ath11k_read_pktlog_filter(struct file *file,
+					 char __user *ubuf,
+					 size_t count, loff_t *ppos)
+
+{
+	char buf[32] = {0};
+	struct ath11k *ar = file->private_data;
+	int len = 0;
+
+	mutex_lock(&ar->conf_mutex);
+	len = scnprintf(buf, sizeof(buf) - len, "%08x %08x\n",
+			ar->debug.pktlog_filter,
+			ar->debug.pktlog_mode);
+	mutex_unlock(&ar->conf_mutex);
+
+	return simple_read_from_buffer(ubuf, count, ppos, buf, len);
+}
+
+static const struct file_operations fops_pktlog_filter = {
+	.read = ath11k_read_pktlog_filter,
+	.write = ath11k_write_pktlog_filter,
+	.open = simple_open
+};
+
+static ssize_t ath11k_write_simulate_radar(struct file *file,
+					   const char __user *user_buf,
+					   size_t count, loff_t *ppos)
+{
+	struct ath11k *ar = file->private_data;
+	int ret;
+
+	ret = ath11k_wmi_simulate_radar(ar);
+	if (ret)
+		return ret;
+
+	return count;
+}
+
+static const struct file_operations fops_simulate_radar = {
+	.write = ath11k_write_simulate_radar,
+	.open = simple_open
+};
+
+int ath11k_debug_register(struct ath11k *ar)
+{
+	struct ath11k_base *ab = ar->ab;
+	char pdev_name[5];
+	char buf[100] = {0};
+
+	snprintf(pdev_name, sizeof(pdev_name), "%s%d", "mac", ar->pdev_idx);
+
+	ar->debug.debugfs_pdev = debugfs_create_dir(pdev_name, ab->debugfs_soc);
+
+	if (IS_ERR_OR_NULL(ar->debug.debugfs_pdev)) {
+		if (IS_ERR(ar->debug.debugfs_pdev))
+			return PTR_ERR(ar->debug.debugfs_pdev);
+
+		return -ENOMEM;
+	}
+
+	/* Create a symlink under ieee80211/phy* */
+	snprintf(buf, 100, "../../ath11k/%pd2", ar->debug.debugfs_pdev);
+	debugfs_create_symlink("ath11k", ar->hw->wiphy->debugfsdir, buf);
+
+	ath11k_debug_htt_stats_init(ar);
+
+	ath11k_debug_fw_stats_init(ar);
+
+	debugfs_create_file("ext_tx_stats", 0644,
+			    ar->debug.debugfs_pdev, ar,
+			    &fops_extd_tx_stats);
+	debugfs_create_file("ext_rx_stats", 0644,
+			    ar->debug.debugfs_pdev, ar,
+			    &fops_extd_rx_stats);
+	debugfs_create_file("pktlog_filter", 0644,
+			    ar->debug.debugfs_pdev, ar,
+			    &fops_pktlog_filter);
+
+	if (ar->hw->wiphy->bands[NL80211_BAND_5GHZ]) {
+		debugfs_create_file("dfs_simulate_radar", 0200,
+				    ar->debug.debugfs_pdev, ar,
+				    &fops_simulate_radar);
+		debugfs_create_bool("dfs_block_radar_events", 0200,
+				    ar->debug.debugfs_pdev,
+				    &ar->dfs_block_radar_events);
+	}
+
+	return 0;
+}
+
+void ath11k_debug_unregister(struct ath11k *ar)
+{
+}
+#endif /* CONFIG_ATH11K_DEBUGFS */
diff --git a/drivers/net/wireless/ath/ath11k/debug.h b/drivers/net/wireless/ath/ath11k/debug.h
new file mode 100644
index 000000000000..aef33f83c9b1
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/debug.h
@@ -0,0 +1,278 @@
+/* SPDX-License-Identifier: BSD-3-Clause-Clear */
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#ifndef _ATH11K_DEBUG_H_
+#define _ATH11K_DEBUG_H_
+
+#include "hal_tx.h"
+#include "trace.h"
+
+enum ath11k_debug_mask {
+	ATH11K_DBG_AHB		= 0x00000001,
+	ATH11K_DBG_WMI		= 0x00000002,
+	ATH11K_DBG_HTC		= 0x00000004,
+	ATH11K_DBG_DP_HTT	= 0x00000008,
+	ATH11K_DBG_MAC		= 0x00000010,
+	ATH11K_DBG_BOOT		= 0x00000020,
+	ATH11K_DBG_QMI		= 0x00000040,
+	ATH11K_DBG_DATA		= 0x00000080,
+	ATH11K_DBG_MGMT		= 0x00000100,
+	ATH11K_DBG_REG		= 0x00000200,
+	ATH11K_DBG_TESTMODE	= 0x00000400,
+	ATH11k_DBG_HAL		= 0x00000800,
+	ATH11K_DBG_ANY		= 0xffffffff,
+};
+
+/* htt_dbg_ext_stats_type */
+enum ath11k_dbg_htt_ext_stats_type {
+	ATH11K_DBG_HTT_EXT_STATS_RESET                      =  0,
+	ATH11K_DBG_HTT_EXT_STATS_PDEV_TX                    =  1,
+	ATH11K_DBG_HTT_EXT_STATS_PDEV_RX                    =  2,
+	ATH11K_DBG_HTT_EXT_STATS_PDEV_TX_HWQ                =  3,
+	ATH11K_DBG_HTT_EXT_STATS_PDEV_TX_SCHED              =  4,
+	ATH11K_DBG_HTT_EXT_STATS_PDEV_ERROR                 =  5,
+	ATH11K_DBG_HTT_EXT_STATS_PDEV_TQM                   =  6,
+	ATH11K_DBG_HTT_EXT_STATS_TQM_CMDQ                   =  7,
+	ATH11K_DBG_HTT_EXT_STATS_TX_DE_INFO                 =  8,
+	ATH11K_DBG_HTT_EXT_STATS_PDEV_TX_RATE               =  9,
+	ATH11K_DBG_HTT_EXT_STATS_PDEV_RX_RATE               =  10,
+	ATH11K_DBG_HTT_EXT_STATS_PEER_INFO                  =  11,
+	ATH11K_DBG_HTT_EXT_STATS_TX_SELFGEN_INFO            =  12,
+	ATH11K_DBG_HTT_EXT_STATS_TX_MU_HWQ                  =  13,
+	ATH11K_DBG_HTT_EXT_STATS_RING_IF_INFO               =  14,
+	ATH11K_DBG_HTT_EXT_STATS_SRNG_INFO                  =  15,
+	ATH11K_DBG_HTT_EXT_STATS_SFM_INFO                   =  16,
+	ATH11K_DBG_HTT_EXT_STATS_PDEV_TX_MU                 =  17,
+	ATH11K_DBG_HTT_EXT_STATS_ACTIVE_PEERS_LIST          =  18,
+	ATH11K_DBG_HTT_EXT_STATS_PDEV_CCA_STATS             =  19,
+	ATH11K_DBG_HTT_EXT_STATS_TWT_SESSIONS               =  20,
+	ATH11K_DBG_HTT_EXT_STATS_REO_RESOURCE_STATS         =  21,
+	ATH11K_DBG_HTT_EXT_STATS_TX_SOUNDING_INFO           =  22,
+
+	/* keep this last */
+	ATH11K_DBG_HTT_NUM_EXT_STATS,
+};
+
+struct debug_htt_stats_req {
+	bool done;
+	u8 pdev_id;
+	u8 type;
+	u8 peer_addr[ETH_ALEN];
+	struct completion cmpln;
+	u32 buf_len;
+	u8 buf[0];
+};
+
+#define ATH11K_HTT_STATS_BUF_SIZE (1024 * 512)
+
+#define ATH11K_FW_STATS_BUF_SIZE (1024 * 1024)
+
+#define ATH11K_HTT_PKTLOG_MAX_SIZE 2048
+
+enum ath11k_pktlog_filter {
+	ATH11K_PKTLOG_RX		= 0x000000001,
+	ATH11K_PKTLOG_TX		= 0x000000002,
+	ATH11K_PKTLOG_RCFIND		= 0x000000004,
+	ATH11K_PKTLOG_RCUPDATE		= 0x000000008,
+	ATH11K_PKTLOG_EVENT_SMART_ANT	= 0x000000020,
+	ATH11K_PKTLOG_EVENT_SW		= 0x000000040,
+	ATH11K_PKTLOG_ANY		= 0x00000006f,
+};
+
+enum ath11k_pktlog_mode {
+	ATH11K_PKTLOG_MODE_LITE = 1,
+	ATH11K_PKTLOG_MODE_FULL = 2,
+};
+
+enum ath11k_pktlog_enum {
+	ATH11K_PKTLOG_TYPE_TX_CTRL      = 1,
+	ATH11K_PKTLOG_TYPE_TX_STAT      = 2,
+	ATH11K_PKTLOG_TYPE_TX_MSDU_ID   = 3,
+	ATH11K_PKTLOG_TYPE_RX_STAT      = 5,
+	ATH11K_PKTLOG_TYPE_RC_FIND      = 6,
+	ATH11K_PKTLOG_TYPE_RC_UPDATE    = 7,
+	ATH11K_PKTLOG_TYPE_TX_VIRT_ADDR = 8,
+	ATH11K_PKTLOG_TYPE_RX_CBF       = 10,
+	ATH11K_PKTLOG_TYPE_RX_STATBUF   = 22,
+	ATH11K_PKTLOG_TYPE_PPDU_STATS   = 23,
+	ATH11K_PKTLOG_TYPE_LITE_RX      = 24,
+};
+
+__printf(2, 3) void ath11k_info(struct ath11k_base *ab, const char *fmt, ...);
+__printf(2, 3) void ath11k_err(struct ath11k_base *ab, const char *fmt, ...);
+__printf(2, 3) void ath11k_warn(struct ath11k_base *ab, const char *fmt, ...);
+
+extern unsigned int ath11k_debug_mask;
+
+#ifdef CONFIG_ATH11K_DEBUG
+__printf(3, 4) void __ath11k_dbg(struct ath11k_base *ab,
+				 enum ath11k_debug_mask mask,
+				 const char *fmt, ...);
+void ath11k_dbg_dump(struct ath11k_base *ab,
+		     enum ath11k_debug_mask mask,
+		     const char *msg, const char *prefix,
+		     const void *buf, size_t len);
+#else /* CONFIG_ATH11K_DEBUG */
+static inline int __ath11k_dbg(struct ath11k_base *ab,
+			       enum ath11k_debug_mask dbg_mask,
+			       const char *fmt, ...)
+{
+	return 0;
+}
+
+static inline void ath11k_dbg_dump(struct ath11k_base *ab,
+				   enum ath11k_debug_mask mask,
+				   const char *msg, const char *prefix,
+				   const void *buf, size_t len)
+{
+}
+#endif /* CONFIG_ATH11K_DEBUG */
+
+#ifdef CONFIG_ATH11K_DEBUGFS
+int ath11k_debug_soc_create(struct ath11k_base *ab);
+void ath11k_debug_soc_destroy(struct ath11k_base *ab);
+int ath11k_debug_pdev_create(struct ath11k_base *ab);
+void ath11k_debug_pdev_destroy(struct ath11k_base *ab);
+int ath11k_debug_register(struct ath11k *ar);
+void ath11k_debug_unregister(struct ath11k *ar);
+void ath11k_dbg_htt_ext_stats_handler(struct ath11k_base *ab,
+				      struct sk_buff *skb);
+void ath11k_debug_fw_stats_process(struct ath11k_base *ab, struct sk_buff *skb);
+
+void ath11k_debug_fw_stats_init(struct ath11k *ar);
+int ath11k_dbg_htt_stats_req(struct ath11k *ar);
+
+static inline bool ath11k_debug_is_pktlog_lite_mode_enabled(struct ath11k *ar)
+{
+	return (ar->debug.pktlog_mode == ATH11K_PKTLOG_MODE_LITE);
+}
+
+static inline bool ath11k_debug_is_pktlog_rx_stats_enabled(struct ath11k *ar)
+{
+	return (!ar->debug.pktlog_peer_valid && ar->debug.pktlog_mode);
+}
+
+static inline bool ath11k_debug_is_pktlog_peer_valid(struct ath11k *ar, u8 *addr)
+{
+	return (ar->debug.pktlog_peer_valid && ar->debug.pktlog_mode &&
+		ether_addr_equal(addr, ar->debug.pktlog_peer_addr));
+}
+
+static inline int ath11k_debug_is_extd_tx_stats_enabled(struct ath11k *ar)
+{
+	return ar->debug.extd_tx_stats;
+}
+
+static inline int ath11k_debug_is_extd_rx_stats_enabled(struct ath11k *ar)
+{
+	return ar->debug.extd_rx_stats;
+}
+#else
+static inline int ath11k_debug_soc_create(struct ath11k_base *ab)
+{
+	return 0;
+}
+
+static inline void ath11k_debug_soc_destroy(struct ath11k_base *ab)
+{
+}
+
+static inline int ath11k_debug_pdev_create(struct ath11k_base *ab)
+{
+	return 0;
+}
+
+static inline void ath11k_debug_pdev_destroy(struct ath11k_base *ab)
+{
+}
+
+static inline int ath11k_debug_register(struct ath11k *ar)
+{
+	return 0;
+}
+
+static inline void ath11k_debug_unregister(struct ath11k *ar)
+{
+}
+
+static inline void ath11k_dbg_htt_ext_stats_handler(struct ath11k_base *ab,
+						    struct sk_buff *skb)
+{
+}
+
+static inline void ath11k_debug_fw_stats_process(struct ath11k_base *ab,
+						 struct sk_buff *skb)
+{
+}
+
+static inline void ath11k_debug_fw_stats_init(struct ath11k *ar)
+{
+}
+
+static inline int ath11k_debug_is_extd_tx_stats_enabled(struct ath11k *ar)
+{
+	return 0;
+}
+
+static inline int ath11k_debug_is_extd_rx_stats_enabled(struct ath11k *ar)
+{
+	return 0;
+}
+
+static inline int ath11k_dbg_htt_stats_req(struct ath11k *ar)
+{
+	return 0;
+}
+
+static inline bool ath11k_debug_is_pktlog_lite_mode_enabled(struct ath11k *ar)
+{
+	return false;
+}
+
+static inline bool ath11k_debug_is_pktlog_rx_stats_enabled(struct ath11k *ar)
+{
+	return false;
+}
+
+static inline bool ath11k_debug_is_pktlog_peer_valid(struct ath11k *ar, u8 *addr)
+{
+	return false;
+}
+#endif /* CONFIG_ATH11K_DEBUGFS */
+
+#ifdef CONFIG_MAC80211_DEBUGFS
+void ath11k_sta_add_debugfs(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+			    struct ieee80211_sta *sta, struct dentry *dir);
+void
+ath11k_accumulate_per_peer_tx_stats(struct ath11k_sta *arsta,
+				    struct ath11k_per_peer_tx_stats *peer_stats,
+				    u8 legacy_rate_idx);
+void ath11k_update_per_peer_stats_from_txcompl(struct ath11k *ar,
+					       struct sk_buff *msdu,
+					       struct hal_tx_status *ts);
+#else /* !CONFIG_MAC80211_DEBUGFS */
+static inline void
+ath11k_accumulate_per_peer_tx_stats(struct ath11k_sta *arsta,
+				    struct ath11k_per_peer_tx_stats *peer_stats,
+				    u8 legacy_rate_idx)
+{
+}
+
+static inline void
+ath11k_update_per_peer_stats_from_txcompl(struct ath11k *ar,
+					  struct sk_buff *msdu,
+					  struct hal_tx_status *ts)
+{
+}
+
+#endif /* CONFIG_MAC80211_DEBUGFS*/
+
+#define ath11k_dbg(ar, dbg_mask, fmt, ...)			\
+do {								\
+	if (ath11k_debug_mask & dbg_mask)			\
+		__ath11k_dbg(ar, dbg_mask, fmt, ##__VA_ARGS__);	\
+} while (0)
+
+#endif /* _ATH11K_DEBUG_H_ */
diff --git a/drivers/net/wireless/ath/ath11k/debug_htt_stats.c b/drivers/net/wireless/ath/ath11k/debug_htt_stats.c
new file mode 100644
index 000000000000..27b301bc1a1b
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/debug_htt_stats.c
@@ -0,0 +1,4432 @@
+// SPDX-License-Identifier: BSD-3-Clause-Clear
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#include <linux/vmalloc.h>
+#include "core.h"
+#include "dp_tx.h"
+#include "dp_rx.h"
+#include "debug.h"
+#include "debug_htt_stats.h"
+
+#define HTT_DBG_OUT(buf, len, fmt, ...) \
+			scnprintf(buf, len, fmt "\n", ##__VA_ARGS__)
+
+#define HTT_MAX_STRING_LEN 256
+#define HTT_MAX_PRINT_CHAR_PER_ELEM 15
+
+#define HTT_TLV_HDR_LEN 4
+
+#define ARRAY_TO_STRING(out, arr, len)							\
+	do {										\
+		int index = 0; u8 i;							\
+		for (i = 0; i < len; i++) {						\
+			index += snprintf(out + index, HTT_MAX_STRING_LEN - index,	\
+					  " %u:%u,", i, arr[i]);			\
+			if (index < 0 || index >= HTT_MAX_STRING_LEN)			\
+				break;							\
+		}									\
+	} while (0)
+
+static inline void htt_print_stats_string_tlv(const void *tag_buf,
+					      u16 tag_len,
+					      struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_stats_string_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	u8  i;
+	u16 index = 0;
+	char data[HTT_MAX_STRING_LEN] = {0};
+
+	tag_len = tag_len >> 2;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_STATS_STRING_TLV:");
+
+	for (i = 0; i < tag_len; i++) {
+		index += snprintf(&data[index],
+				HTT_MAX_STRING_LEN - index,
+				"%.*s", 4, (char *)&(htt_stats_buf->data[i]));
+		if (index >= HTT_MAX_STRING_LEN)
+			break;
+	}
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "data = %s\n", data);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_tx_pdev_stats_cmn_tlv(const void *tag_buf,
+						   struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_pdev_stats_cmn_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_PDEV_STATS_CMN_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mac_id = %u",
+			   htt_stats_buf->mac_id__word & 0xFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "hw_queued = %u",
+			   htt_stats_buf->hw_queued);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "hw_reaped = %u",
+			   htt_stats_buf->hw_reaped);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "underrun = %u",
+			   htt_stats_buf->underrun);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "hw_paused = %u",
+			   htt_stats_buf->hw_paused);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "hw_flush = %u",
+			   htt_stats_buf->hw_flush);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "hw_filt = %u",
+			   htt_stats_buf->hw_filt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_abort = %u",
+			   htt_stats_buf->tx_abort);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mpdu_requeued = %u",
+			   htt_stats_buf->mpdu_requed);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_xretry = %u",
+			   htt_stats_buf->tx_xretry);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "data_rc = %u",
+			   htt_stats_buf->data_rc);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mpdu_dropped_xretry = %u",
+			   htt_stats_buf->mpdu_dropped_xretry);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "illegal_rate_phy_err = %u",
+			   htt_stats_buf->illgl_rate_phy_err);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "cont_xretry = %u",
+			   htt_stats_buf->cont_xretry);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_timeout = %u",
+			   htt_stats_buf->tx_timeout);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "pdev_resets = %u",
+			   htt_stats_buf->pdev_resets);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "phy_underrun = %u",
+			   htt_stats_buf->phy_underrun);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "txop_ovf = %u",
+			   htt_stats_buf->txop_ovf);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "seq_posted = %u",
+			   htt_stats_buf->seq_posted);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "seq_failed_queueing = %u",
+			   htt_stats_buf->seq_failed_queueing);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "seq_completed = %u",
+			   htt_stats_buf->seq_completed);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "seq_restarted = %u",
+			   htt_stats_buf->seq_restarted);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mu_seq_posted = %u",
+			   htt_stats_buf->mu_seq_posted);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "seq_switch_hw_paused = %u",
+			   htt_stats_buf->seq_switch_hw_paused);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "next_seq_posted_dsr = %u",
+			   htt_stats_buf->next_seq_posted_dsr);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "seq_posted_isr = %u",
+			   htt_stats_buf->seq_posted_isr);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "seq_ctrl_cached = %u",
+			   htt_stats_buf->seq_ctrl_cached);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mpdu_count_tqm = %u",
+			   htt_stats_buf->mpdu_count_tqm);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "msdu_count_tqm = %u",
+			   htt_stats_buf->msdu_count_tqm);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mpdu_removed_tqm = %u",
+			   htt_stats_buf->mpdu_removed_tqm);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "msdu_removed_tqm = %u",
+			   htt_stats_buf->msdu_removed_tqm);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mpdus_sw_flush = %u",
+			   htt_stats_buf->mpdus_sw_flush);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mpdus_hw_filter = %u",
+			   htt_stats_buf->mpdus_hw_filter);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mpdus_truncated = %u",
+			   htt_stats_buf->mpdus_truncated);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mpdus_ack_failed = %u",
+			   htt_stats_buf->mpdus_ack_failed);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mpdus_expired = %u",
+			   htt_stats_buf->mpdus_expired);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mpdus_seq_hw_retry = %u",
+			   htt_stats_buf->mpdus_seq_hw_retry);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ack_tlv_proc = %u",
+			   htt_stats_buf->ack_tlv_proc);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "coex_abort_mpdu_cnt_valid = %u",
+			   htt_stats_buf->coex_abort_mpdu_cnt_valid);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "coex_abort_mpdu_cnt = %u",
+			   htt_stats_buf->coex_abort_mpdu_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_total_ppdus_tried_ota = %u",
+			   htt_stats_buf->num_total_ppdus_tried_ota);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_data_ppdus_tried_ota = %u",
+			   htt_stats_buf->num_data_ppdus_tried_ota);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "local_ctrl_mgmt_enqued = %u",
+			   htt_stats_buf->local_ctrl_mgmt_enqued);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "local_ctrl_mgmt_freed = %u",
+			   htt_stats_buf->local_ctrl_mgmt_freed);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "local_data_enqued = %u",
+			   htt_stats_buf->local_data_enqued);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "local_data_freed = %u",
+			   htt_stats_buf->local_data_freed);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mpdu_tried = %u",
+			   htt_stats_buf->mpdu_tried);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "isr_wait_seq_posted = %u",
+			   htt_stats_buf->isr_wait_seq_posted);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_active_dur_us_low = %u",
+			   htt_stats_buf->tx_active_dur_us_low);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_active_dur_us_high = %u\n",
+			   htt_stats_buf->tx_active_dur_us_high);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_pdev_stats_urrn_tlv_v(const void *tag_buf,
+				   u16 tag_len,
+				   struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_pdev_stats_urrn_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char urrn_stats[HTT_MAX_STRING_LEN] = {0};
+	u16 num_elems = min_t(u16, (tag_len >> 2), HTT_TX_PDEV_MAX_URRN_STATS);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_PDEV_STATS_URRN_TLV_V:");
+
+	ARRAY_TO_STRING(urrn_stats, htt_stats_buf->urrn_stats, num_elems);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "urrn_stats = %s\n", urrn_stats);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_pdev_stats_flush_tlv_v(const void *tag_buf,
+				    u16 tag_len,
+				    struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_pdev_stats_flush_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char flush_errs[HTT_MAX_STRING_LEN] = {0};
+	u16 num_elems = min_t(u16, (tag_len >> 2), HTT_TX_PDEV_MAX_FLUSH_REASON_STATS);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_PDEV_STATS_FLUSH_TLV_V:");
+
+	ARRAY_TO_STRING(flush_errs, htt_stats_buf->flush_errs, num_elems);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "flush_errs = %s\n", flush_errs);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_pdev_stats_sifs_tlv_v(const void *tag_buf,
+				   u16 tag_len,
+				   struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_pdev_stats_sifs_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char sifs_status[HTT_MAX_STRING_LEN] = {0};
+	u16 num_elems = min_t(u16, (tag_len >> 2), HTT_TX_PDEV_MAX_SIFS_BURST_STATS);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_PDEV_STATS_SIFS_TLV_V:");
+
+	ARRAY_TO_STRING(sifs_status, htt_stats_buf->sifs_status, num_elems);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sifs_status = %s\n",
+			   sifs_status);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_pdev_stats_phy_err_tlv_v(const void *tag_buf,
+				      u16 tag_len,
+				      struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_pdev_stats_phy_err_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char phy_errs[HTT_MAX_STRING_LEN] = {0};
+	u16 num_elems = min_t(u16, (tag_len >> 2), HTT_TX_PDEV_MAX_PHY_ERR_STATS);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_PDEV_STATS_PHY_ERR_TLV_V:");
+
+	ARRAY_TO_STRING(phy_errs, htt_stats_buf->phy_errs, num_elems);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "phy_errs = %s\n", phy_errs);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_pdev_stats_sifs_hist_tlv_v(const void *tag_buf,
+					u16 tag_len,
+					struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_pdev_stats_sifs_hist_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char sifs_hist_status[HTT_MAX_STRING_LEN] = {0};
+	u16 num_elems = min_t(u16, (tag_len >> 2), HTT_TX_PDEV_MAX_SIFS_BURST_HIST_STATS);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "HTT_TX_PDEV_STATS_SIFS_HIST_TLV_V:");
+
+	ARRAY_TO_STRING(sifs_hist_status, htt_stats_buf->sifs_hist_status, num_elems);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sifs_hist_status = %s\n",
+			   sifs_hist_status);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_pdev_stats_tx_ppdu_stats_tlv_v(const void *tag_buf,
+					    struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_pdev_stats_tx_ppdu_stats_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "HTT_TX_PDEV_STATS_TX_PPDU_STATS_TLV_V:");
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_data_ppdus_legacy_su = %u",
+			   htt_stats_buf->num_data_ppdus_legacy_su);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_data_ppdus_ac_su = %u",
+			   htt_stats_buf->num_data_ppdus_ac_su);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_data_ppdus_ax_su = %u",
+			   htt_stats_buf->num_data_ppdus_ax_su);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_data_ppdus_ac_su_txbf = %u",
+			   htt_stats_buf->num_data_ppdus_ac_su_txbf);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_data_ppdus_ax_su_txbf = %u\n",
+			   htt_stats_buf->num_data_ppdus_ax_su_txbf);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_pdev_stats_tried_mpdu_cnt_hist_tlv_v(const void *tag_buf,
+						  u16 tag_len,
+						  struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_pdev_stats_tried_mpdu_cnt_hist_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char tried_mpdu_cnt_hist[HTT_MAX_STRING_LEN] = {0};
+	u32  num_elements = ((tag_len - sizeof(htt_stats_buf->hist_bin_size)) >> 2);
+	u32  required_buffer_size = HTT_MAX_PRINT_CHAR_PER_ELEM * num_elements;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "HTT_TX_PDEV_STATS_TRIED_MPDU_CNT_HIST_TLV_V:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "TRIED_MPDU_CNT_HIST_BIN_SIZE : %u",
+			   htt_stats_buf->hist_bin_size);
+
+	if (required_buffer_size < HTT_MAX_STRING_LEN) {
+		ARRAY_TO_STRING(tried_mpdu_cnt_hist,
+				htt_stats_buf->tried_mpdu_cnt_hist,
+				num_elements);
+		len += HTT_DBG_OUT(buf + len, buf_len - len, "tried_mpdu_cnt_hist = %s\n",
+				   tried_mpdu_cnt_hist);
+	} else {
+		len += HTT_DBG_OUT(buf + len, buf_len - len,
+				   "INSUFFICIENT PRINT BUFFER\n");
+	}
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_hw_stats_intr_misc_tlv(const void *tag_buf,
+						    struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_hw_stats_intr_misc_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char hw_intr_name[HTT_STATS_MAX_HW_INTR_NAME_LEN + 1] = {0};
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_HW_STATS_INTR_MISC_TLV:");
+	memcpy(hw_intr_name, &(htt_stats_buf->hw_intr_name[0]),
+	       HTT_STATS_MAX_HW_INTR_NAME_LEN);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "hw_intr_name = %s ", hw_intr_name);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mask = %u",
+			   htt_stats_buf->mask);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "count = %u\n",
+			   htt_stats_buf->count);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_hw_stats_wd_timeout_tlv(const void *tag_buf,
+				  struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_hw_stats_wd_timeout_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char hw_module_name[HTT_STATS_MAX_HW_MODULE_NAME_LEN + 1] = {0};
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_HW_STATS_WD_TIMEOUT_TLV:");
+	memcpy(hw_module_name, &(htt_stats_buf->hw_module_name[0]),
+	       HTT_STATS_MAX_HW_MODULE_NAME_LEN);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "hw_module_name = %s ",
+			   hw_module_name);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "count = %u",
+			   htt_stats_buf->count);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_hw_stats_pdev_errs_tlv(const void *tag_buf,
+						    struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_hw_stats_pdev_errs_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_HW_STATS_PDEV_ERRS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mac_id = %u",
+			   htt_stats_buf->mac_id__word & 0xFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_abort = %u",
+			   htt_stats_buf->tx_abort);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_abort_fail_count = %u",
+			   htt_stats_buf->tx_abort_fail_count);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_abort = %u",
+			   htt_stats_buf->rx_abort);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_abort_fail_count = %u",
+			   htt_stats_buf->rx_abort_fail_count);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "warm_reset = %u",
+			   htt_stats_buf->warm_reset);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "cold_reset = %u",
+			   htt_stats_buf->cold_reset);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_flush = %u",
+			   htt_stats_buf->tx_flush);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_glb_reset = %u",
+			   htt_stats_buf->tx_glb_reset);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_txq_reset = %u",
+			   htt_stats_buf->tx_txq_reset);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_timeout_reset = %u\n",
+			   htt_stats_buf->rx_timeout_reset);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_msdu_flow_stats_tlv(const void *tag_buf,
+						 struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_msdu_flow_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_MSDU_FLOW_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "last_update_timestamp = %u",
+			   htt_stats_buf->last_update_timestamp);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "last_add_timestamp = %u",
+			   htt_stats_buf->last_add_timestamp);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "last_remove_timestamp = %u",
+			   htt_stats_buf->last_remove_timestamp);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "total_processed_msdu_count = %u",
+			   htt_stats_buf->total_processed_msdu_count);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "cur_msdu_count_in_flowq = %u",
+			   htt_stats_buf->cur_msdu_count_in_flowq);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sw_peer_id = %u",
+			   htt_stats_buf->sw_peer_id);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_flow_no = %u",
+			   htt_stats_buf->tx_flow_no__tid_num__drop_rule & 0xFFFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tid_num = %u",
+			   (htt_stats_buf->tx_flow_no__tid_num__drop_rule & 0xF0000) >>
+			   16);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "drop_rule = %u",
+			   (htt_stats_buf->tx_flow_no__tid_num__drop_rule & 0x100000) >>
+			   20);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "last_cycle_enqueue_count = %u",
+			   htt_stats_buf->last_cycle_enqueue_count);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "last_cycle_dequeue_count = %u",
+			   htt_stats_buf->last_cycle_dequeue_count);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "last_cycle_drop_count = %u",
+			   htt_stats_buf->last_cycle_drop_count);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "current_drop_th = %u\n",
+			   htt_stats_buf->current_drop_th);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_tx_tid_stats_tlv(const void *tag_buf,
+					      struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_tid_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char tid_name[MAX_HTT_TID_NAME + 1] = {0};
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_TID_STATS_TLV:");
+	memcpy(tid_name, &(htt_stats_buf->tid_name[0]), MAX_HTT_TID_NAME);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tid_name = %s ", tid_name);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sw_peer_id = %u",
+			   htt_stats_buf->sw_peer_id__tid_num & 0xFFFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tid_num = %u",
+			   (htt_stats_buf->sw_peer_id__tid_num & 0xFFFF0000) >> 16);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_sched_pending = %u",
+			   htt_stats_buf->num_sched_pending__num_ppdu_in_hwq & 0xFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_ppdu_in_hwq = %u",
+			   (htt_stats_buf->num_sched_pending__num_ppdu_in_hwq &
+			   0xFF00) >> 8);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tid_flags = 0x%x",
+			   htt_stats_buf->tid_flags);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "hw_queued = %u",
+			   htt_stats_buf->hw_queued);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "hw_reaped = %u",
+			   htt_stats_buf->hw_reaped);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mpdus_hw_filter = %u",
+			   htt_stats_buf->mpdus_hw_filter);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "qdepth_bytes = %u",
+			   htt_stats_buf->qdepth_bytes);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "qdepth_num_msdu = %u",
+			   htt_stats_buf->qdepth_num_msdu);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "qdepth_num_mpdu = %u",
+			   htt_stats_buf->qdepth_num_mpdu);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "last_scheduled_tsmp = %u",
+			   htt_stats_buf->last_scheduled_tsmp);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "pause_module_id = %u",
+			   htt_stats_buf->pause_module_id);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "block_module_id = %u\n",
+			   htt_stats_buf->block_module_id);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_tx_tid_stats_v1_tlv(const void *tag_buf,
+						 struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_tid_stats_v1_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char tid_name[MAX_HTT_TID_NAME + 1] = {0};
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_TID_STATS_V1_TLV:");
+	memcpy(tid_name, &(htt_stats_buf->tid_name[0]), MAX_HTT_TID_NAME);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tid_name = %s ", tid_name);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sw_peer_id = %u",
+			   htt_stats_buf->sw_peer_id__tid_num & 0xFFFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tid_num = %u",
+			   (htt_stats_buf->sw_peer_id__tid_num & 0xFFFF0000) >> 16);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_sched_pending = %u",
+			   htt_stats_buf->num_sched_pending__num_ppdu_in_hwq & 0xFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_ppdu_in_hwq = %u",
+			   (htt_stats_buf->num_sched_pending__num_ppdu_in_hwq &
+			   0xFF00) >> 8);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tid_flags = 0x%x",
+			   htt_stats_buf->tid_flags);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "max_qdepth_bytes = %u",
+			   htt_stats_buf->max_qdepth_bytes);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "max_qdepth_n_msdus = %u",
+			   htt_stats_buf->max_qdepth_n_msdus);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rsvd = %u",
+			   htt_stats_buf->rsvd);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "qdepth_bytes = %u",
+			   htt_stats_buf->qdepth_bytes);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "qdepth_num_msdu = %u",
+			   htt_stats_buf->qdepth_num_msdu);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "qdepth_num_mpdu = %u",
+			   htt_stats_buf->qdepth_num_mpdu);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "last_scheduled_tsmp = %u",
+			   htt_stats_buf->last_scheduled_tsmp);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "pause_module_id = %u",
+			   htt_stats_buf->pause_module_id);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "block_module_id = %u",
+			   htt_stats_buf->block_module_id);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "allow_n_flags = 0x%x",
+			   htt_stats_buf->allow_n_flags);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sendn_frms_allowed = %u\n",
+			   htt_stats_buf->sendn_frms_allowed);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_rx_tid_stats_tlv(const void *tag_buf,
+					      struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_rx_tid_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char tid_name[MAX_HTT_TID_NAME + 1] = {0};
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_RX_TID_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sw_peer_id = %u",
+			   htt_stats_buf->sw_peer_id__tid_num & 0xFFFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tid_num = %u",
+			   (htt_stats_buf->sw_peer_id__tid_num & 0xFFFF0000) >> 16);
+	memcpy(tid_name, &(htt_stats_buf->tid_name[0]), MAX_HTT_TID_NAME);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tid_name = %s ", tid_name);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "dup_in_reorder = %u",
+			   htt_stats_buf->dup_in_reorder);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "dup_past_outside_window = %u",
+			   htt_stats_buf->dup_past_outside_window);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "dup_past_within_window = %u",
+			   htt_stats_buf->dup_past_within_window);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rxdesc_err_decrypt = %u\n",
+			   htt_stats_buf->rxdesc_err_decrypt);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_counter_tlv(const void *tag_buf,
+					 struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_counter_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char counter_name[HTT_MAX_STRING_LEN] = {0};
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_COUNTER_TLV:");
+
+	ARRAY_TO_STRING(counter_name,
+			htt_stats_buf->counter_name,
+			HTT_MAX_COUNTER_NAME);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "counter_name = %s ", counter_name);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "count = %u\n",
+			   htt_stats_buf->count);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_peer_stats_cmn_tlv(const void *tag_buf,
+						struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_peer_stats_cmn_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_PEER_STATS_CMN_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ppdu_cnt = %u",
+			   htt_stats_buf->ppdu_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mpdu_cnt = %u",
+			   htt_stats_buf->mpdu_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "msdu_cnt = %u",
+			   htt_stats_buf->msdu_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "pause_bitmap = %u",
+			   htt_stats_buf->pause_bitmap);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "block_bitmap = %u",
+			   htt_stats_buf->block_bitmap);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "last_rssi = %d",
+			   htt_stats_buf->rssi);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "enqueued_count = %llu",
+			   htt_stats_buf->peer_enqueued_count_low |
+			   ((u64)htt_stats_buf->peer_enqueued_count_high << 32));
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "dequeued_count = %llu",
+			   htt_stats_buf->peer_dequeued_count_low |
+			   ((u64)htt_stats_buf->peer_dequeued_count_high << 32));
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "dropped_count = %llu",
+			   htt_stats_buf->peer_dropped_count_low |
+			   ((u64)htt_stats_buf->peer_dropped_count_high << 32));
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "transmitted_ppdu_bytes = %llu",
+			   htt_stats_buf->ppdu_transmitted_bytes_low |
+			   ((u64)htt_stats_buf->ppdu_transmitted_bytes_high << 32));
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ttl_removed_count = %u",
+			   htt_stats_buf->peer_ttl_removed_count);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "inactive_time = %u\n",
+			   htt_stats_buf->inactive_time);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_peer_details_tlv(const void *tag_buf,
+					      struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_peer_details_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_PEER_DETAILS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "peer_type = %u",
+			   htt_stats_buf->peer_type);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sw_peer_id = %u",
+			   htt_stats_buf->sw_peer_id);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "vdev_id = %u",
+			   htt_stats_buf->vdev_pdev_ast_idx & 0xFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "pdev_id = %u",
+			   (htt_stats_buf->vdev_pdev_ast_idx & 0xFF00) >> 8);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ast_idx = %u",
+			   (htt_stats_buf->vdev_pdev_ast_idx & 0xFFFF0000) >> 16);
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "mac_addr = %02x:%02x:%02x:%02x:%02x:%02x",
+			   htt_stats_buf->mac_addr.mac_addr_l32 & 0xFF,
+			   (htt_stats_buf->mac_addr.mac_addr_l32 & 0xFF00) >> 8,
+			   (htt_stats_buf->mac_addr.mac_addr_l32 & 0xFF0000) >> 16,
+			   (htt_stats_buf->mac_addr.mac_addr_l32 & 0xFF000000) >> 24,
+			   (htt_stats_buf->mac_addr.mac_addr_h16 & 0xFF),
+			   (htt_stats_buf->mac_addr.mac_addr_h16 & 0xFF00) >> 8);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "peer_flags = 0x%x",
+			   htt_stats_buf->peer_flags);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "qpeer_flags = 0x%x\n",
+			   htt_stats_buf->qpeer_flags);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_tx_peer_rate_stats_tlv(const void *tag_buf,
+						    struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_peer_rate_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char str_buf[HTT_MAX_STRING_LEN] = {0};
+	char *tx_gi[HTT_TX_PEER_STATS_NUM_GI_COUNTERS];
+	u8 j;
+
+	for (j = 0; j < HTT_TX_PEER_STATS_NUM_GI_COUNTERS; j++)
+		tx_gi[j] = kmalloc(HTT_MAX_STRING_LEN, GFP_ATOMIC);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_PEER_RATE_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_ldpc = %u",
+			   htt_stats_buf->tx_ldpc);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rts_cnt = %u",
+			   htt_stats_buf->rts_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ack_rssi = %u",
+			   htt_stats_buf->ack_rssi);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->tx_mcs,
+			HTT_TX_PDEV_STATS_NUM_MCS_COUNTERS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_mcs = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->tx_su_mcs,
+			HTT_TX_PDEV_STATS_NUM_MCS_COUNTERS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_su_mcs = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->tx_mu_mcs,
+			HTT_TX_PDEV_STATS_NUM_MCS_COUNTERS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_mu_mcs = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf,
+			htt_stats_buf->tx_nss,
+			HTT_TX_PDEV_STATS_NUM_SPATIAL_STREAMS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_nss = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf,
+			htt_stats_buf->tx_bw,
+			HTT_TX_PDEV_STATS_NUM_BW_COUNTERS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_bw = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->tx_stbc,
+			HTT_TX_PDEV_STATS_NUM_MCS_COUNTERS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_stbc = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->tx_pream,
+			HTT_TX_PDEV_STATS_NUM_PREAMBLE_TYPES);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_pream = %s ", str_buf);
+
+	for (j = 0; j < HTT_TX_PEER_STATS_NUM_GI_COUNTERS; j++) {
+		ARRAY_TO_STRING(tx_gi[j],
+				htt_stats_buf->tx_gi[j],
+				HTT_TX_PEER_STATS_NUM_MCS_COUNTERS);
+		len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_gi[%u] = %s ",
+				j, tx_gi[j]);
+	}
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf,
+			htt_stats_buf->tx_dcm,
+			HTT_TX_PDEV_STATS_NUM_DCM_COUNTERS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_dcm = %s\n", str_buf);
+
+	for (j = 0; j < HTT_TX_PEER_STATS_NUM_GI_COUNTERS; j++)
+		kfree(tx_gi[j]);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_rx_peer_rate_stats_tlv(const void *tag_buf,
+						    struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_rx_peer_rate_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	u8 j;
+	char *rssi_chain[HTT_RX_PEER_STATS_NUM_SPATIAL_STREAMS];
+	char *rx_gi[HTT_RX_PEER_STATS_NUM_GI_COUNTERS];
+	char str_buf[HTT_MAX_STRING_LEN] = {0};
+
+	for (j = 0; j < HTT_RX_PEER_STATS_NUM_SPATIAL_STREAMS; j++)
+		rssi_chain[j] = kmalloc(HTT_MAX_STRING_LEN, GFP_ATOMIC);
+
+	for (j = 0; j < HTT_RX_PEER_STATS_NUM_GI_COUNTERS; j++)
+		rx_gi[j] = kmalloc(HTT_MAX_STRING_LEN, GFP_ATOMIC);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_RX_PEER_RATE_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "nsts = %u",
+			   htt_stats_buf->nsts);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_ldpc = %u",
+			   htt_stats_buf->rx_ldpc);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rts_cnt = %u",
+			   htt_stats_buf->rts_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rssi_mgmt = %u",
+			   htt_stats_buf->rssi_mgmt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rssi_data = %u",
+			   htt_stats_buf->rssi_data);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rssi_comb = %u",
+			   htt_stats_buf->rssi_comb);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->rx_mcs,
+			HTT_RX_PDEV_STATS_NUM_MCS_COUNTERS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_mcs = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->rx_nss,
+			HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_nss = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->rx_dcm,
+			HTT_RX_PDEV_STATS_NUM_DCM_COUNTERS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_dcm = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->rx_stbc,
+			HTT_RX_PDEV_STATS_NUM_MCS_COUNTERS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_stbc = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->rx_bw,
+			HTT_RX_PDEV_STATS_NUM_BW_COUNTERS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_bw = %s ", str_buf);
+
+	for (j = 0; j < HTT_RX_PEER_STATS_NUM_SPATIAL_STREAMS; j++) {
+		ARRAY_TO_STRING(rssi_chain[j], htt_stats_buf->rssi_chain[j],
+				HTT_RX_PEER_STATS_NUM_BW_COUNTERS);
+		len += HTT_DBG_OUT(buf + len, buf_len - len, "rssi_chain[%u] = %s ",
+				   j, rssi_chain[j]);
+	}
+
+	for (j = 0; j < HTT_RX_PEER_STATS_NUM_GI_COUNTERS; j++) {
+		ARRAY_TO_STRING(rx_gi[j], htt_stats_buf->rx_gi[j],
+				HTT_RX_PDEV_STATS_NUM_MCS_COUNTERS);
+		len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_gi[%u] = %s ",
+				j, rx_gi[j]);
+	}
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->rx_pream,
+			HTT_RX_PDEV_STATS_NUM_PREAMBLE_TYPES);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_pream = %s\n", str_buf);
+
+	for (j = 0; j < HTT_RX_PEER_STATS_NUM_SPATIAL_STREAMS; j++)
+		kfree(rssi_chain[j]);
+
+	for (j = 0; j < HTT_RX_PEER_STATS_NUM_GI_COUNTERS; j++)
+		kfree(rx_gi[j]);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_hwq_mu_mimo_sch_stats_tlv(const void *tag_buf,
+				       struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_hwq_mu_mimo_sch_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_HWQ_MU_MIMO_SCH_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mu_mimo_sch_posted = %u",
+			   htt_stats_buf->mu_mimo_sch_posted);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mu_mimo_sch_failed = %u",
+			   htt_stats_buf->mu_mimo_sch_failed);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mu_mimo_ppdu_posted = %u\n",
+			   htt_stats_buf->mu_mimo_ppdu_posted);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_hwq_mu_mimo_mpdu_stats_tlv(const void *tag_buf,
+					struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_hwq_mu_mimo_mpdu_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "HTT_TX_HWQ_MU_MIMO_MPDU_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mu_mimo_mpdus_queued_usr = %u",
+			   htt_stats_buf->mu_mimo_mpdus_queued_usr);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mu_mimo_mpdus_tried_usr = %u",
+			   htt_stats_buf->mu_mimo_mpdus_tried_usr);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mu_mimo_mpdus_failed_usr = %u",
+			   htt_stats_buf->mu_mimo_mpdus_failed_usr);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mu_mimo_mpdus_requeued_usr = %u",
+			   htt_stats_buf->mu_mimo_mpdus_requeued_usr);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mu_mimo_err_no_ba_usr = %u",
+			   htt_stats_buf->mu_mimo_err_no_ba_usr);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mu_mimo_mpdu_underrun_usr = %u",
+			   htt_stats_buf->mu_mimo_mpdu_underrun_usr);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mu_mimo_ampdu_underrun_usr = %u\n",
+			   htt_stats_buf->mu_mimo_ampdu_underrun_usr);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_hwq_mu_mimo_cmn_stats_tlv(const void *tag_buf,
+				       struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_hwq_mu_mimo_cmn_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_HWQ_MU_MIMO_CMN_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mac_id = %u",
+			   htt_stats_buf->mac_id__hwq_id__word & 0xFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "hwq_id = %u\n",
+			   (htt_stats_buf->mac_id__hwq_id__word & 0xFF00) >> 8);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_hwq_stats_cmn_tlv(const void *tag_buf, struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_hwq_stats_cmn_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	/* TODO: HKDBG */
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_HWQ_STATS_CMN_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mac_id = %u",
+			   htt_stats_buf->mac_id__hwq_id__word & 0xFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "hwq_id = %u",
+			   (htt_stats_buf->mac_id__hwq_id__word & 0xFF00) >> 8);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "xretry = %u",
+			   htt_stats_buf->xretry);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "underrun_cnt = %u",
+			   htt_stats_buf->underrun_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "flush_cnt = %u",
+			   htt_stats_buf->flush_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "filt_cnt = %u",
+			   htt_stats_buf->filt_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "null_mpdu_bmap = %u",
+			   htt_stats_buf->null_mpdu_bmap);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "user_ack_failure = %u",
+			   htt_stats_buf->user_ack_failure);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ack_tlv_proc = %u",
+			   htt_stats_buf->ack_tlv_proc);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sched_id_proc = %u",
+			   htt_stats_buf->sched_id_proc);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "null_mpdu_tx_count = %u",
+			   htt_stats_buf->null_mpdu_tx_count);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mpdu_bmap_not_recvd = %u",
+			   htt_stats_buf->mpdu_bmap_not_recvd);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_bar = %u",
+			   htt_stats_buf->num_bar);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rts = %u",
+			   htt_stats_buf->rts);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "cts2self = %u",
+			   htt_stats_buf->cts2self);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "qos_null = %u",
+			   htt_stats_buf->qos_null);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mpdu_tried_cnt = %u",
+			   htt_stats_buf->mpdu_tried_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mpdu_queued_cnt = %u",
+			   htt_stats_buf->mpdu_queued_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mpdu_ack_fail_cnt = %u",
+			   htt_stats_buf->mpdu_ack_fail_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mpdu_filt_cnt = %u",
+			   htt_stats_buf->mpdu_filt_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "false_mpdu_ack_count = %u",
+			   htt_stats_buf->false_mpdu_ack_count);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "txq_timeout = %u\n",
+			   htt_stats_buf->txq_timeout);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_hwq_difs_latency_stats_tlv_v(const void *tag_buf,
+					  u16 tag_len,
+					  struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_hwq_difs_latency_stats_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	u16 data_len = min_t(u16, (tag_len >> 2), HTT_TX_HWQ_MAX_DIFS_LATENCY_BINS);
+	char difs_latency_hist[HTT_MAX_STRING_LEN] = {0};
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "HTT_TX_HWQ_DIFS_LATENCY_STATS_TLV_V:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "hist_intvl = %u",
+			htt_stats_buf->hist_intvl);
+
+	ARRAY_TO_STRING(difs_latency_hist, htt_stats_buf->difs_latency_hist,
+			data_len);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "difs_latency_hist = %s\n",
+			difs_latency_hist);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_hwq_cmd_result_stats_tlv_v(const void *tag_buf,
+					u16 tag_len,
+					struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_hwq_cmd_result_stats_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	u16 data_len;
+	char cmd_result[HTT_MAX_STRING_LEN] = {0};
+
+	data_len = min_t(u16, (tag_len >> 2), HTT_TX_HWQ_MAX_CMD_RESULT_STATS);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "HTT_TX_HWQ_CMD_RESULT_STATS_TLV_V:");
+
+	ARRAY_TO_STRING(cmd_result, htt_stats_buf->cmd_result, data_len);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "cmd_result = %s\n", cmd_result);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_hwq_cmd_stall_stats_tlv_v(const void *tag_buf,
+				       u16 tag_len,
+				       struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_hwq_cmd_stall_stats_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	u16 num_elems;
+	char cmd_stall_status[HTT_MAX_STRING_LEN] = {0};
+
+	num_elems = min_t(u16, (tag_len >> 2), HTT_TX_HWQ_MAX_CMD_STALL_STATS);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_HWQ_CMD_STALL_STATS_TLV_V:");
+
+	ARRAY_TO_STRING(cmd_stall_status, htt_stats_buf->cmd_stall_status, num_elems);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "cmd_stall_status = %s\n",
+			   cmd_stall_status);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_hwq_fes_result_stats_tlv_v(const void *tag_buf,
+					u16 tag_len,
+					struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_hwq_fes_result_stats_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	u16 num_elems;
+	char fes_result[HTT_MAX_STRING_LEN] = {0};
+
+	num_elems = min_t(u16, (tag_len >> 2), HTT_TX_HWQ_MAX_FES_RESULT_STATS);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "HTT_TX_HWQ_FES_RESULT_STATS_TLV_V:");
+
+	ARRAY_TO_STRING(fes_result, htt_stats_buf->fes_result, num_elems);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fes_result = %s\n", fes_result);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_hwq_tried_mpdu_cnt_hist_tlv_v(const void *tag_buf,
+					   u16 tag_len,
+					   struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_hwq_tried_mpdu_cnt_hist_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char tried_mpdu_cnt_hist[HTT_MAX_STRING_LEN] = {0};
+	u32  num_elements = ((tag_len -
+			    sizeof(htt_stats_buf->hist_bin_size)) >> 2);
+	u32  required_buffer_size = HTT_MAX_PRINT_CHAR_PER_ELEM * num_elements;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "HTT_TX_HWQ_TRIED_MPDU_CNT_HIST_TLV_V:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "TRIED_MPDU_CNT_HIST_BIN_SIZE : %u",
+			   htt_stats_buf->hist_bin_size);
+
+	if (required_buffer_size < HTT_MAX_STRING_LEN) {
+		ARRAY_TO_STRING(tried_mpdu_cnt_hist,
+				htt_stats_buf->tried_mpdu_cnt_hist,
+				num_elements);
+		len += HTT_DBG_OUT(buf + len, buf_len - len,
+				   "tried_mpdu_cnt_hist = %s\n",
+				   tried_mpdu_cnt_hist);
+	} else {
+		len += HTT_DBG_OUT(buf + len, buf_len - len,
+				   "INSUFFICIENT PRINT BUFFER ");
+	}
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_hwq_txop_used_cnt_hist_tlv_v(const void *tag_buf,
+					  u16 tag_len,
+					  struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_hwq_txop_used_cnt_hist_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char txop_used_cnt_hist[HTT_MAX_STRING_LEN] = {0};
+	u32 num_elements = tag_len >> 2;
+	u32  required_buffer_size = HTT_MAX_PRINT_CHAR_PER_ELEM * num_elements;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "HTT_TX_HWQ_TXOP_USED_CNT_HIST_TLV_V:");
+
+	if (required_buffer_size < HTT_MAX_STRING_LEN) {
+		ARRAY_TO_STRING(txop_used_cnt_hist,
+				htt_stats_buf->txop_used_cnt_hist,
+				num_elements);
+		len += HTT_DBG_OUT(buf + len, buf_len - len, "txop_used_cnt_hist = %s\n",
+				   txop_used_cnt_hist);
+	} else {
+		len += HTT_DBG_OUT(buf + len, buf_len - len,
+				   "INSUFFICIENT PRINT BUFFER ");
+	}
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_tx_sounding_stats_tlv(const void *tag_buf,
+						   struct debug_htt_stats_req *stats_req)
+{
+	s32 i;
+	const struct htt_tx_sounding_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	const u32 *cbf_20 = htt_stats_buf->cbf_20;
+	const u32 *cbf_40 = htt_stats_buf->cbf_40;
+	const u32 *cbf_80 = htt_stats_buf->cbf_80;
+	const u32 *cbf_160 = htt_stats_buf->cbf_160;
+
+	if (htt_stats_buf->tx_sounding_mode == HTT_TX_AC_SOUNDING_MODE) {
+		len += HTT_DBG_OUT(buf + len, buf_len - len,
+				   "\nHTT_TX_AC_SOUNDING_STATS_TLV:\n");
+		len += HTT_DBG_OUT(buf + len, buf_len - len,
+				   "ac_cbf_20 = IBF : %u, SU_SIFS : %u, SU_RBO : %u, MU_SIFS : %u, MU_RBO : %u ",
+				   cbf_20[HTT_IMPLICIT_TXBF_STEER_STATS],
+				   cbf_20[HTT_EXPLICIT_TXBF_SU_SIFS_STEER_STATS],
+				   cbf_20[HTT_EXPLICIT_TXBF_SU_RBO_STEER_STATS],
+				   cbf_20[HTT_EXPLICIT_TXBF_MU_SIFS_STEER_STATS],
+				   cbf_20[HTT_EXPLICIT_TXBF_MU_RBO_STEER_STATS]);
+		len += HTT_DBG_OUT(buf + len, buf_len - len,
+				   "ac_cbf_40 = IBF : %u, SU_SIFS : %u, SU_RBO : %u, MU_SIFS : %u, MU_RBO : %u",
+				   cbf_40[HTT_IMPLICIT_TXBF_STEER_STATS],
+				   cbf_40[HTT_EXPLICIT_TXBF_SU_SIFS_STEER_STATS],
+				   cbf_40[HTT_EXPLICIT_TXBF_SU_RBO_STEER_STATS],
+				   cbf_40[HTT_EXPLICIT_TXBF_MU_SIFS_STEER_STATS],
+				   cbf_40[HTT_EXPLICIT_TXBF_MU_RBO_STEER_STATS]);
+		len += HTT_DBG_OUT(buf + len, buf_len - len,
+				   "ac_cbf_80 = IBF : %u, SU_SIFS : %u, SU_RBO : %u, MU_SIFS : %u, MU_RBO : %u",
+				   cbf_80[HTT_IMPLICIT_TXBF_STEER_STATS],
+				   cbf_80[HTT_EXPLICIT_TXBF_SU_SIFS_STEER_STATS],
+				   cbf_80[HTT_EXPLICIT_TXBF_SU_RBO_STEER_STATS],
+				   cbf_80[HTT_EXPLICIT_TXBF_MU_SIFS_STEER_STATS],
+				   cbf_80[HTT_EXPLICIT_TXBF_MU_RBO_STEER_STATS]);
+		len += HTT_DBG_OUT(buf + len, buf_len - len,
+				   "ac_cbf_160 = IBF : %u, SU_SIFS : %u, SU_RBO : %u, MU_SIFS : %u, MU_RBO : %u",
+				   cbf_160[HTT_IMPLICIT_TXBF_STEER_STATS],
+				   cbf_160[HTT_EXPLICIT_TXBF_SU_SIFS_STEER_STATS],
+				   cbf_160[HTT_EXPLICIT_TXBF_SU_RBO_STEER_STATS],
+				   cbf_160[HTT_EXPLICIT_TXBF_MU_SIFS_STEER_STATS],
+				   cbf_160[HTT_EXPLICIT_TXBF_MU_RBO_STEER_STATS]);
+
+		for (i = 0; i < HTT_TX_PDEV_STATS_NUM_AC_MUMIMO_USER_STATS; i++) {
+			len += HTT_DBG_OUT(buf + len, buf_len - len,
+					   "Sounding User %u = 20MHz: %u, 40MHz : %u, 80MHz: %u, 160MHz: %u ",
+					   i,
+					   htt_stats_buf->sounding[0],
+					   htt_stats_buf->sounding[1],
+					   htt_stats_buf->sounding[2],
+					   htt_stats_buf->sounding[3]);
+		}
+	} else if (htt_stats_buf->tx_sounding_mode == HTT_TX_AX_SOUNDING_MODE) {
+		len += HTT_DBG_OUT(buf + len, buf_len - len,
+				   "\nHTT_TX_AX_SOUNDING_STATS_TLV:\n");
+		len += HTT_DBG_OUT(buf + len, buf_len - len,
+				   "ax_cbf_20 = IBF : %u, SU_SIFS : %u, SU_RBO : %u, MU_SIFS : %u, MU_RBO : %u ",
+				   cbf_20[HTT_IMPLICIT_TXBF_STEER_STATS],
+				   cbf_20[HTT_EXPLICIT_TXBF_SU_SIFS_STEER_STATS],
+				   cbf_20[HTT_EXPLICIT_TXBF_SU_RBO_STEER_STATS],
+				   cbf_20[HTT_EXPLICIT_TXBF_MU_SIFS_STEER_STATS],
+				   cbf_20[HTT_EXPLICIT_TXBF_MU_RBO_STEER_STATS]);
+		len += HTT_DBG_OUT(buf + len, buf_len - len,
+				   "ax_cbf_40 = IBF : %u, SU_SIFS : %u, SU_RBO : %u, MU_SIFS : %u, MU_RBO : %u",
+				   cbf_40[HTT_IMPLICIT_TXBF_STEER_STATS],
+				   cbf_40[HTT_EXPLICIT_TXBF_SU_SIFS_STEER_STATS],
+				   cbf_40[HTT_EXPLICIT_TXBF_SU_RBO_STEER_STATS],
+				   cbf_40[HTT_EXPLICIT_TXBF_MU_SIFS_STEER_STATS],
+				   cbf_40[HTT_EXPLICIT_TXBF_MU_RBO_STEER_STATS]);
+		len += HTT_DBG_OUT(buf + len, buf_len - len,
+				   "ax_cbf_80 = IBF : %u, SU_SIFS : %u, SU_RBO : %u, MU_SIFS : %u, MU_RBO : %u",
+				   cbf_80[HTT_IMPLICIT_TXBF_STEER_STATS],
+				   cbf_80[HTT_EXPLICIT_TXBF_SU_SIFS_STEER_STATS],
+				   cbf_80[HTT_EXPLICIT_TXBF_SU_RBO_STEER_STATS],
+				   cbf_80[HTT_EXPLICIT_TXBF_MU_SIFS_STEER_STATS],
+				   cbf_80[HTT_EXPLICIT_TXBF_MU_RBO_STEER_STATS]);
+		len += HTT_DBG_OUT(buf + len, buf_len - len,
+				   "ax_cbf_160 = IBF : %u, SU_SIFS : %u, SU_RBO : %u, MU_SIFS : %u, MU_RBO : %u",
+				   cbf_160[HTT_IMPLICIT_TXBF_STEER_STATS],
+				   cbf_160[HTT_EXPLICIT_TXBF_SU_SIFS_STEER_STATS],
+				   cbf_160[HTT_EXPLICIT_TXBF_SU_RBO_STEER_STATS],
+				   cbf_160[HTT_EXPLICIT_TXBF_MU_SIFS_STEER_STATS],
+				   cbf_160[HTT_EXPLICIT_TXBF_MU_RBO_STEER_STATS]);
+
+		for (i = 0; i < HTT_TX_PDEV_STATS_NUM_AX_MUMIMO_USER_STATS; i++) {
+			len += HTT_DBG_OUT(buf + len, buf_len - len,
+					   "Sounding User %u = 20MHz: %u, 40MHz : %u, 80MHz: %u, 160MHz: %u ",
+					   i,
+					   htt_stats_buf->sounding[0],
+					   htt_stats_buf->sounding[1],
+					   htt_stats_buf->sounding[2],
+					   htt_stats_buf->sounding[3]);
+		}
+	}
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_selfgen_cmn_stats_tlv(const void *tag_buf,
+				   struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_selfgen_cmn_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_SELFGEN_CMN_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mac_id = %u",
+			   htt_stats_buf->mac_id__word & 0xFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "su_bar = %u",
+			   htt_stats_buf->su_bar);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rts = %u",
+			   htt_stats_buf->rts);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "cts2self = %u",
+			   htt_stats_buf->cts2self);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "qos_null = %u",
+			   htt_stats_buf->qos_null);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "delayed_bar_1 = %u",
+			   htt_stats_buf->delayed_bar_1);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "delayed_bar_2 = %u",
+			   htt_stats_buf->delayed_bar_2);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "delayed_bar_3 = %u",
+			   htt_stats_buf->delayed_bar_3);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "delayed_bar_4 = %u",
+			   htt_stats_buf->delayed_bar_4);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "delayed_bar_5 = %u",
+			   htt_stats_buf->delayed_bar_5);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "delayed_bar_6 = %u",
+			   htt_stats_buf->delayed_bar_6);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "delayed_bar_7 = %u\n",
+			   htt_stats_buf->delayed_bar_7);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_selfgen_ac_stats_tlv(const void *tag_buf,
+				  struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_selfgen_ac_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_SELFGEN_AC_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ac_su_ndpa = %u",
+			   htt_stats_buf->ac_su_ndpa);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ac_su_ndp = %u",
+			   htt_stats_buf->ac_su_ndp);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ac_mu_mimo_ndpa = %u",
+			   htt_stats_buf->ac_mu_mimo_ndpa);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ac_mu_mimo_ndp = %u",
+			   htt_stats_buf->ac_mu_mimo_ndp);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ac_mu_mimo_brpoll_1 = %u",
+			   htt_stats_buf->ac_mu_mimo_brpoll_1);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ac_mu_mimo_brpoll_2 = %u",
+			   htt_stats_buf->ac_mu_mimo_brpoll_2);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ac_mu_mimo_brpoll_3 = %u\n",
+			   htt_stats_buf->ac_mu_mimo_brpoll_3);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_selfgen_ax_stats_tlv(const void *tag_buf,
+				  struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_selfgen_ax_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_SELFGEN_AX_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_su_ndpa = %u",
+			   htt_stats_buf->ax_su_ndpa);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_su_ndp = %u",
+			   htt_stats_buf->ax_su_ndp);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_mu_mimo_ndpa = %u",
+			   htt_stats_buf->ax_mu_mimo_ndpa);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_mu_mimo_ndp = %u",
+			   htt_stats_buf->ax_mu_mimo_ndp);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_mu_mimo_brpoll_1 = %u",
+			   htt_stats_buf->ax_mu_mimo_brpoll_1);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_mu_mimo_brpoll_2 = %u",
+			   htt_stats_buf->ax_mu_mimo_brpoll_2);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_mu_mimo_brpoll_3 = %u",
+			   htt_stats_buf->ax_mu_mimo_brpoll_3);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_mu_mimo_brpoll_4 = %u",
+			   htt_stats_buf->ax_mu_mimo_brpoll_4);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_mu_mimo_brpoll_5 = %u",
+			   htt_stats_buf->ax_mu_mimo_brpoll_5);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_mu_mimo_brpoll_6 = %u",
+			   htt_stats_buf->ax_mu_mimo_brpoll_6);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_mu_mimo_brpoll_7 = %u",
+			   htt_stats_buf->ax_mu_mimo_brpoll_7);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_basic_trigger = %u",
+			   htt_stats_buf->ax_basic_trigger);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_bsr_trigger = %u",
+			   htt_stats_buf->ax_bsr_trigger);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_mu_bar_trigger = %u",
+			   htt_stats_buf->ax_mu_bar_trigger);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_mu_rts_trigger = %u\n",
+			   htt_stats_buf->ax_mu_rts_trigger);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_selfgen_ac_err_stats_tlv(const void *tag_buf,
+				      struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_selfgen_ac_err_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_SELFGEN_AC_ERR_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ac_su_ndp_err = %u",
+			   htt_stats_buf->ac_su_ndp_err);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ac_su_ndpa_err = %u",
+			   htt_stats_buf->ac_su_ndpa_err);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ac_mu_mimo_ndpa_err = %u",
+			   htt_stats_buf->ac_mu_mimo_ndpa_err);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ac_mu_mimo_ndp_err = %u",
+			   htt_stats_buf->ac_mu_mimo_ndp_err);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ac_mu_mimo_brp1_err = %u",
+			   htt_stats_buf->ac_mu_mimo_brp1_err);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ac_mu_mimo_brp2_err = %u",
+			   htt_stats_buf->ac_mu_mimo_brp2_err);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ac_mu_mimo_brp3_err = %u\n",
+			   htt_stats_buf->ac_mu_mimo_brp3_err);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_selfgen_ax_err_stats_tlv(const void *tag_buf,
+				      struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_selfgen_ax_err_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_SELFGEN_AX_ERR_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_su_ndp_err = %u",
+			   htt_stats_buf->ax_su_ndp_err);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_su_ndpa_err = %u",
+			   htt_stats_buf->ax_su_ndpa_err);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_mu_mimo_ndpa_err = %u",
+			   htt_stats_buf->ax_mu_mimo_ndpa_err);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_mu_mimo_ndp_err = %u",
+			   htt_stats_buf->ax_mu_mimo_ndp_err);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_mu_mimo_brp1_err = %u",
+			   htt_stats_buf->ax_mu_mimo_brp1_err);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_mu_mimo_brp2_err = %u",
+			   htt_stats_buf->ax_mu_mimo_brp2_err);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_mu_mimo_brp3_err = %u",
+			   htt_stats_buf->ax_mu_mimo_brp3_err);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_mu_mimo_brp4_err = %u",
+			   htt_stats_buf->ax_mu_mimo_brp4_err);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_mu_mimo_brp5_err = %u",
+			   htt_stats_buf->ax_mu_mimo_brp5_err);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_mu_mimo_brp6_err = %u",
+			   htt_stats_buf->ax_mu_mimo_brp6_err);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_mu_mimo_brp7_err = %u",
+			   htt_stats_buf->ax_mu_mimo_brp7_err);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_basic_trigger_err = %u",
+			   htt_stats_buf->ax_basic_trigger_err);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_bsr_trigger_err = %u",
+			   htt_stats_buf->ax_bsr_trigger_err);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_mu_bar_trigger_err = %u",
+			   htt_stats_buf->ax_mu_bar_trigger_err);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_mu_rts_trigger_err = %u\n",
+			   htt_stats_buf->ax_mu_rts_trigger_err);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_pdev_mu_mimo_sch_stats_tlv(const void *tag_buf,
+					struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_pdev_mu_mimo_sch_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	u8 i;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "HTT_TX_PDEV_MU_MIMO_SCH_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mu_mimo_sch_posted = %u",
+			   htt_stats_buf->mu_mimo_sch_posted);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mu_mimo_sch_failed = %u",
+			   htt_stats_buf->mu_mimo_sch_failed);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mu_mimo_ppdu_posted = %u\n",
+			   htt_stats_buf->mu_mimo_ppdu_posted);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "11ac MU_MIMO SCH STATS:");
+
+	for (i = 0; i < HTT_TX_PDEV_STATS_NUM_AC_MUMIMO_USER_STATS; i++)
+		len += HTT_DBG_OUT(buf + len, buf_len - len,
+				   "ac_mu_mimo_sch_nusers_%u = %u",
+				   i, htt_stats_buf->ac_mu_mimo_sch_nusers[i]);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "\n11ax MU_MIMO SCH STATS:");
+
+	for (i = 0; i < HTT_TX_PDEV_STATS_NUM_AX_MUMIMO_USER_STATS; i++)
+		len += HTT_DBG_OUT(buf + len, buf_len - len,
+				   "ax_mu_mimo_sch_nusers_%u = %u",
+				   i, htt_stats_buf->ax_mu_mimo_sch_nusers[i]);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "\n11ax OFDMA SCH STATS:");
+
+	for (i = 0; i < HTT_TX_PDEV_STATS_NUM_OFDMA_USER_STATS; i++)
+		len += HTT_DBG_OUT(buf + len, buf_len - len,
+				   "ax_ofdma_sch_nusers_%u = %u",
+				   i, htt_stats_buf->ax_ofdma_sch_nusers[i]);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_pdev_mu_mimo_mpdu_stats_tlv(const void *tag_buf,
+					 struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_pdev_mpdu_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	if (htt_stats_buf->tx_sched_mode == HTT_STATS_TX_SCHED_MODE_MU_MIMO_AC) {
+		if (!htt_stats_buf->user_index)
+			len += HTT_DBG_OUT(buf + len, buf_len - len,
+					   "HTT_TX_PDEV_MU_MIMO_AC_MPDU_STATS:\n");
+
+		if (htt_stats_buf->user_index <
+		    HTT_TX_PDEV_STATS_NUM_AC_MUMIMO_USER_STATS) {
+			len += HTT_DBG_OUT(buf + len, buf_len - len,
+					   "ac_mu_mimo_mpdus_queued_usr_%u = %u",
+					   htt_stats_buf->user_index,
+					   htt_stats_buf->mpdus_queued_usr);
+			len += HTT_DBG_OUT(buf + len, buf_len - len,
+					   "ac_mu_mimo_mpdus_tried_usr_%u = %u",
+					   htt_stats_buf->user_index,
+					   htt_stats_buf->mpdus_tried_usr);
+			len += HTT_DBG_OUT(buf + len, buf_len - len,
+					   "ac_mu_mimo_mpdus_failed_usr_%u = %u",
+					   htt_stats_buf->user_index,
+					   htt_stats_buf->mpdus_failed_usr);
+			len += HTT_DBG_OUT(buf + len, buf_len - len,
+					   "ac_mu_mimo_mpdus_requeued_usr_%u = %u",
+					   htt_stats_buf->user_index,
+					   htt_stats_buf->mpdus_requeued_usr);
+			len += HTT_DBG_OUT(buf + len, buf_len - len,
+					   "ac_mu_mimo_err_no_ba_usr_%u = %u",
+					   htt_stats_buf->user_index,
+					   htt_stats_buf->err_no_ba_usr);
+			len += HTT_DBG_OUT(buf + len, buf_len - len,
+					   "ac_mu_mimo_mpdu_underrun_usr_%u = %u",
+					   htt_stats_buf->user_index,
+					   htt_stats_buf->mpdu_underrun_usr);
+			len += HTT_DBG_OUT(buf + len, buf_len - len,
+					   "ac_mu_mimo_ampdu_underrun_usr_%u = %u\n",
+					   htt_stats_buf->user_index,
+					   htt_stats_buf->ampdu_underrun_usr);
+		}
+	}
+
+	if (htt_stats_buf->tx_sched_mode == HTT_STATS_TX_SCHED_MODE_MU_MIMO_AX) {
+		if (!htt_stats_buf->user_index)
+			len += HTT_DBG_OUT(buf + len, buf_len - len,
+					   "HTT_TX_PDEV_MU_MIMO_AX_MPDU_STATS:\n");
+
+		if (htt_stats_buf->user_index <
+		    HTT_TX_PDEV_STATS_NUM_AX_MUMIMO_USER_STATS) {
+			len += HTT_DBG_OUT(buf + len, buf_len - len,
+					   "ax_mu_mimo_mpdus_queued_usr_%u = %u",
+					   htt_stats_buf->user_index,
+					   htt_stats_buf->mpdus_queued_usr);
+			len += HTT_DBG_OUT(buf + len, buf_len - len,
+					   "ax_mu_mimo_mpdus_tried_usr_%u = %u",
+					   htt_stats_buf->user_index,
+					   htt_stats_buf->mpdus_tried_usr);
+			len += HTT_DBG_OUT(buf + len, buf_len - len,
+					   "ax_mu_mimo_mpdus_failed_usr_%u = %u",
+					   htt_stats_buf->user_index,
+					   htt_stats_buf->mpdus_failed_usr);
+			len += HTT_DBG_OUT(buf + len, buf_len - len,
+					   "ax_mu_mimo_mpdus_requeued_usr_%u = %u",
+					   htt_stats_buf->user_index,
+					   htt_stats_buf->mpdus_requeued_usr);
+			len += HTT_DBG_OUT(buf + len, buf_len - len,
+					   "ax_mu_mimo_err_no_ba_usr_%u = %u",
+					   htt_stats_buf->user_index,
+					   htt_stats_buf->err_no_ba_usr);
+			len += HTT_DBG_OUT(buf + len, buf_len - len,
+					   "ax_mu_mimo_mpdu_underrun_usr_%u = %u",
+					   htt_stats_buf->user_index,
+					   htt_stats_buf->mpdu_underrun_usr);
+			len += HTT_DBG_OUT(buf + len, buf_len - len,
+					   "ax_mu_mimo_ampdu_underrun_usr_%u = %u\n",
+					   htt_stats_buf->user_index,
+					   htt_stats_buf->ampdu_underrun_usr);
+		}
+	}
+
+	if (htt_stats_buf->tx_sched_mode == HTT_STATS_TX_SCHED_MODE_MU_OFDMA_AX) {
+		if (!htt_stats_buf->user_index)
+			len += HTT_DBG_OUT(buf + len, buf_len - len,
+					   "HTT_TX_PDEV_AX_MU_OFDMA_MPDU_STATS:\n");
+
+		if (htt_stats_buf->user_index < HTT_TX_PDEV_STATS_NUM_OFDMA_USER_STATS) {
+			len += HTT_DBG_OUT(buf + len, buf_len - len,
+					   "ax_mu_ofdma_mpdus_queued_usr_%u = %u",
+					   htt_stats_buf->user_index,
+					   htt_stats_buf->mpdus_queued_usr);
+			len += HTT_DBG_OUT(buf + len, buf_len - len,
+					   "ax_mu_ofdma_mpdus_tried_usr_%u = %u",
+					   htt_stats_buf->user_index,
+					   htt_stats_buf->mpdus_tried_usr);
+			len += HTT_DBG_OUT(buf + len, buf_len - len,
+					   "ax_mu_ofdma_mpdus_failed_usr_%u = %u",
+					   htt_stats_buf->user_index,
+					   htt_stats_buf->mpdus_failed_usr);
+			len += HTT_DBG_OUT(buf + len, buf_len - len,
+					   "ax_mu_ofdma_mpdus_requeued_usr_%u = %u",
+					   htt_stats_buf->user_index,
+					   htt_stats_buf->mpdus_requeued_usr);
+			len += HTT_DBG_OUT(buf + len, buf_len - len,
+					   "ax_mu_ofdma_err_no_ba_usr_%u = %u",
+					   htt_stats_buf->user_index,
+					   htt_stats_buf->err_no_ba_usr);
+			len += HTT_DBG_OUT(buf + len, buf_len - len,
+					   "ax_mu_ofdma_mpdu_underrun_usr_%u = %u",
+					   htt_stats_buf->user_index,
+					   htt_stats_buf->mpdu_underrun_usr);
+			len += HTT_DBG_OUT(buf + len, buf_len - len,
+					   "ax_mu_ofdma_ampdu_underrun_usr_%u = %u\n",
+					   htt_stats_buf->user_index,
+					   htt_stats_buf->ampdu_underrun_usr);
+		}
+	}
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_sched_txq_cmd_posted_tlv_v(const void *tag_buf,
+				     u16 tag_len,
+				     struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_sched_txq_cmd_posted_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char sched_cmd_posted[HTT_MAX_STRING_LEN] = {0};
+	u16 num_elements = min_t(u16, (tag_len >> 2), HTT_TX_PDEV_SCHED_TX_MODE_MAX);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_SCHED_TXQ_CMD_POSTED_TLV_V:");
+
+	ARRAY_TO_STRING(sched_cmd_posted, htt_stats_buf->sched_cmd_posted,
+			num_elements);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sched_cmd_posted = %s\n",
+			   sched_cmd_posted);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_sched_txq_cmd_reaped_tlv_v(const void *tag_buf,
+				     u16 tag_len,
+				     struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_sched_txq_cmd_reaped_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char sched_cmd_reaped[HTT_MAX_STRING_LEN] = {0};
+	u16 num_elements = min_t(u16, (tag_len >> 2), HTT_TX_PDEV_SCHED_TX_MODE_MAX);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_SCHED_TXQ_CMD_REAPED_TLV_V:");
+
+	ARRAY_TO_STRING(sched_cmd_reaped, htt_stats_buf->sched_cmd_reaped,
+			num_elements);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sched_cmd_reaped = %s\n",
+			   sched_cmd_reaped);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_sched_txq_sched_order_su_tlv_v(const void *tag_buf,
+					 u16 tag_len,
+					 struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_sched_txq_sched_order_su_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char sched_order_su[HTT_MAX_STRING_LEN] = {0};
+	/* each entry is u32, i.e. 4 bytes */
+	u32 sched_order_su_num_entries =
+		min_t(u32, (tag_len >> 2), HTT_TX_PDEV_NUM_SCHED_ORDER_LOG);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "HTT_SCHED_TXQ_SCHED_ORDER_SU_TLV_V:");
+
+	ARRAY_TO_STRING(sched_order_su, htt_stats_buf->sched_order_su,
+			sched_order_su_num_entries);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sched_order_su = %s\n",
+			   sched_order_su);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_sched_txq_sched_ineligibility_tlv_v(const void *tag_buf,
+					      u16 tag_len,
+					      struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_sched_txq_sched_ineligibility_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char sched_ineligibility[HTT_MAX_STRING_LEN] = {0};
+	/* each entry is u32, i.e. 4 bytes */
+	u32 sched_ineligibility_num_entries = tag_len >> 2;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "HTT_SCHED_TXQ_SCHED_INELIGIBILITY_V:");
+
+	ARRAY_TO_STRING(sched_ineligibility, htt_stats_buf->sched_ineligibility,
+			sched_ineligibility_num_entries);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sched_ineligibility = %s\n",
+			   sched_ineligibility);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_pdev_stats_sched_per_txq_tlv(const void *tag_buf,
+					  struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_pdev_stats_sched_per_txq_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "HTT_TX_PDEV_STATS_SCHED_PER_TXQ_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mac_id = %u",
+			   htt_stats_buf->mac_id__txq_id__word & 0xFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "txq_id = %u",
+			   (htt_stats_buf->mac_id__txq_id__word & 0xFF00) >> 8);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sched_policy = %u",
+			   htt_stats_buf->sched_policy);
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "last_sched_cmd_posted_timestamp = %u",
+			   htt_stats_buf->last_sched_cmd_posted_timestamp);
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "last_sched_cmd_compl_timestamp = %u",
+			   htt_stats_buf->last_sched_cmd_compl_timestamp);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sched_2_tac_lwm_count = %u",
+			   htt_stats_buf->sched_2_tac_lwm_count);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sched_2_tac_ring_full = %u",
+			   htt_stats_buf->sched_2_tac_ring_full);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sched_cmd_post_failure = %u",
+			   htt_stats_buf->sched_cmd_post_failure);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_active_tids = %u",
+			   htt_stats_buf->num_active_tids);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_ps_schedules = %u",
+			   htt_stats_buf->num_ps_schedules);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sched_cmds_pending = %u",
+			   htt_stats_buf->sched_cmds_pending);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_tid_register = %u",
+			   htt_stats_buf->num_tid_register);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_tid_unregister = %u",
+			   htt_stats_buf->num_tid_unregister);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_qstats_queried = %u",
+			   htt_stats_buf->num_qstats_queried);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "qstats_update_pending = %u",
+			   htt_stats_buf->qstats_update_pending);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "last_qstats_query_timestamp = %u",
+			   htt_stats_buf->last_qstats_query_timestamp);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_tqm_cmdq_full = %u",
+			   htt_stats_buf->num_tqm_cmdq_full);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_de_sched_algo_trigger = %u",
+			   htt_stats_buf->num_de_sched_algo_trigger);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_rt_sched_algo_trigger = %u",
+			   htt_stats_buf->num_rt_sched_algo_trigger);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_tqm_sched_algo_trigger = %u",
+			   htt_stats_buf->num_tqm_sched_algo_trigger);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "notify_sched = %u\n",
+			   htt_stats_buf->notify_sched);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "dur_based_sendn_term = %u\n",
+			   htt_stats_buf->dur_based_sendn_term);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_stats_tx_sched_cmn_tlv(const void *tag_buf,
+						    struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_stats_tx_sched_cmn_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_STATS_TX_SCHED_CMN_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mac_id = %u",
+			   htt_stats_buf->mac_id__word & 0xFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "current_timestamp = %u\n",
+			   htt_stats_buf->current_timestamp);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_tqm_gen_mpdu_stats_tlv_v(const void *tag_buf,
+				      u16 tag_len,
+				      struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_tqm_gen_mpdu_stats_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char gen_mpdu_end_reason[HTT_MAX_STRING_LEN] = {0};
+	u16 num_elements = min_t(u16, (tag_len >> 2),
+				 HTT_TX_TQM_MAX_LIST_MPDU_END_REASON);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_TQM_GEN_MPDU_STATS_TLV_V:");
+
+	ARRAY_TO_STRING(gen_mpdu_end_reason, htt_stats_buf->gen_mpdu_end_reason,
+			num_elements);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "gen_mpdu_end_reason = %s\n",
+			   gen_mpdu_end_reason);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_tqm_list_mpdu_stats_tlv_v(const void *tag_buf,
+				       u16 tag_len,
+				       struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_tqm_list_mpdu_stats_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char list_mpdu_end_reason[HTT_MAX_STRING_LEN] = {0};
+	u16 num_elems = min_t(u16, (tag_len >> 2), HTT_TX_TQM_MAX_LIST_MPDU_END_REASON);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "HTT_TX_TQM_LIST_MPDU_STATS_TLV_V:");
+
+	ARRAY_TO_STRING(list_mpdu_end_reason, htt_stats_buf->list_mpdu_end_reason,
+			num_elems);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "list_mpdu_end_reason = %s\n",
+			   list_mpdu_end_reason);
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_tqm_list_mpdu_cnt_tlv_v(const void *tag_buf,
+				     u16 tag_len,
+				     struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_tqm_list_mpdu_cnt_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char list_mpdu_cnt_hist[HTT_MAX_STRING_LEN] = {0};
+	u16 num_elems = min_t(u16, (tag_len >> 2),
+			      HTT_TX_TQM_MAX_LIST_MPDU_CNT_HISTOGRAM_BINS);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_TQM_LIST_MPDU_CNT_TLV_V:");
+
+	ARRAY_TO_STRING(list_mpdu_cnt_hist, htt_stats_buf->list_mpdu_cnt_hist,
+			num_elems);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "list_mpdu_cnt_hist = %s\n",
+			   list_mpdu_cnt_hist);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_tqm_pdev_stats_tlv_v(const void *tag_buf,
+				  struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_tqm_pdev_stats_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_TQM_PDEV_STATS_TLV_V:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "msdu_count = %u",
+			   htt_stats_buf->msdu_count);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mpdu_count = %u",
+			   htt_stats_buf->mpdu_count);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "remove_msdu = %u",
+			   htt_stats_buf->remove_msdu);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "remove_mpdu = %u",
+			   htt_stats_buf->remove_mpdu);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "remove_msdu_ttl = %u",
+			   htt_stats_buf->remove_msdu_ttl);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "send_bar = %u",
+			   htt_stats_buf->send_bar);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "bar_sync = %u",
+			   htt_stats_buf->bar_sync);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "notify_mpdu = %u",
+			   htt_stats_buf->notify_mpdu);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sync_cmd = %u",
+			   htt_stats_buf->sync_cmd);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "write_cmd = %u",
+			   htt_stats_buf->write_cmd);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "hwsch_trigger = %u",
+			   htt_stats_buf->hwsch_trigger);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ack_tlv_proc = %u",
+			   htt_stats_buf->ack_tlv_proc);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "gen_mpdu_cmd = %u",
+			   htt_stats_buf->gen_mpdu_cmd);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "gen_list_cmd = %u",
+			   htt_stats_buf->gen_list_cmd);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "remove_mpdu_cmd = %u",
+			   htt_stats_buf->remove_mpdu_cmd);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "remove_mpdu_tried_cmd = %u",
+			   htt_stats_buf->remove_mpdu_tried_cmd);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mpdu_queue_stats_cmd = %u",
+			   htt_stats_buf->mpdu_queue_stats_cmd);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mpdu_head_info_cmd = %u",
+			   htt_stats_buf->mpdu_head_info_cmd);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "msdu_flow_stats_cmd = %u",
+			   htt_stats_buf->msdu_flow_stats_cmd);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "remove_msdu_cmd = %u",
+			   htt_stats_buf->remove_msdu_cmd);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "remove_msdu_ttl_cmd = %u",
+			   htt_stats_buf->remove_msdu_ttl_cmd);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "flush_cache_cmd = %u",
+			   htt_stats_buf->flush_cache_cmd);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "update_mpduq_cmd = %u",
+			   htt_stats_buf->update_mpduq_cmd);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "enqueue = %u",
+			   htt_stats_buf->enqueue);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "enqueue_notify = %u",
+			   htt_stats_buf->enqueue_notify);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "notify_mpdu_at_head = %u",
+			   htt_stats_buf->notify_mpdu_at_head);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "notify_mpdu_state_valid = %u",
+			   htt_stats_buf->notify_mpdu_state_valid);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sched_udp_notify1 = %u",
+			   htt_stats_buf->sched_udp_notify1);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sched_udp_notify2 = %u",
+			   htt_stats_buf->sched_udp_notify2);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sched_nonudp_notify1 = %u",
+			   htt_stats_buf->sched_nonudp_notify1);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sched_nonudp_notify2 = %u\n",
+			   htt_stats_buf->sched_nonudp_notify2);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_tx_tqm_cmn_stats_tlv(const void *tag_buf,
+						  struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_tqm_cmn_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_TQM_CMN_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mac_id = %u",
+			   htt_stats_buf->mac_id__word & 0xFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "max_cmdq_id = %u",
+			   htt_stats_buf->max_cmdq_id);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "list_mpdu_cnt_hist_intvl = %u",
+			   htt_stats_buf->list_mpdu_cnt_hist_intvl);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "add_msdu = %u",
+			   htt_stats_buf->add_msdu);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "q_empty = %u",
+			   htt_stats_buf->q_empty);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "q_not_empty = %u",
+			   htt_stats_buf->q_not_empty);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "drop_notification = %u",
+			   htt_stats_buf->drop_notification);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "desc_threshold = %u\n",
+			   htt_stats_buf->desc_threshold);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_tx_tqm_error_stats_tlv(const void *tag_buf,
+						    struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_tqm_error_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_TQM_ERROR_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "q_empty_failure = %u",
+			   htt_stats_buf->q_empty_failure);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "q_not_empty_failure = %u",
+			   htt_stats_buf->q_not_empty_failure);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "add_msdu_failure = %u\n",
+			   htt_stats_buf->add_msdu_failure);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_tx_tqm_cmdq_status_tlv(const void *tag_buf,
+						    struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_tqm_cmdq_status_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_TQM_CMDQ_STATUS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mac_id = %u",
+			   htt_stats_buf->mac_id__cmdq_id__word & 0xFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "cmdq_id = %u\n",
+			   (htt_stats_buf->mac_id__cmdq_id__word & 0xFF00) >> 8);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sync_cmd = %u",
+			   htt_stats_buf->sync_cmd);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "write_cmd = %u",
+			   htt_stats_buf->write_cmd);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "gen_mpdu_cmd = %u",
+			   htt_stats_buf->gen_mpdu_cmd);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mpdu_queue_stats_cmd = %u",
+			   htt_stats_buf->mpdu_queue_stats_cmd);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mpdu_head_info_cmd = %u",
+			   htt_stats_buf->mpdu_head_info_cmd);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "msdu_flow_stats_cmd = %u",
+			   htt_stats_buf->msdu_flow_stats_cmd);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "remove_mpdu_cmd = %u",
+			   htt_stats_buf->remove_mpdu_cmd);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "remove_msdu_cmd = %u",
+			   htt_stats_buf->remove_msdu_cmd);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "flush_cache_cmd = %u",
+			   htt_stats_buf->flush_cache_cmd);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "update_mpduq_cmd = %u",
+			   htt_stats_buf->update_mpduq_cmd);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "update_msduq_cmd = %u\n",
+			   htt_stats_buf->update_msduq_cmd);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_de_eapol_packets_stats_tlv(const void *tag_buf,
+					struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_de_eapol_packets_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "HTT_TX_DE_EAPOL_PACKETS_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "m1_packets = %u",
+			   htt_stats_buf->m1_packets);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "m2_packets = %u",
+			   htt_stats_buf->m2_packets);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "m3_packets = %u",
+			   htt_stats_buf->m3_packets);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "m4_packets = %u",
+			   htt_stats_buf->m4_packets);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "g1_packets = %u",
+			   htt_stats_buf->g1_packets);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "g2_packets = %u\n",
+			   htt_stats_buf->g2_packets);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_de_classify_failed_stats_tlv(const void *tag_buf,
+					  struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_de_classify_failed_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "HTT_TX_DE_CLASSIFY_FAILED_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ap_bss_peer_not_found = %u",
+			   htt_stats_buf->ap_bss_peer_not_found);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ap_bcast_mcast_no_peer = %u",
+			   htt_stats_buf->ap_bcast_mcast_no_peer);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sta_delete_in_progress = %u",
+			   htt_stats_buf->sta_delete_in_progress);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ibss_no_bss_peer = %u",
+			   htt_stats_buf->ibss_no_bss_peer);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "invalid_vdev_type = %u",
+			   htt_stats_buf->invalid_vdev_type);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "invalid_ast_peer_entry = %u",
+			   htt_stats_buf->invalid_ast_peer_entry);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "peer_entry_invalid = %u",
+			   htt_stats_buf->peer_entry_invalid);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ethertype_not_ip = %u",
+			   htt_stats_buf->ethertype_not_ip);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "eapol_lookup_failed = %u",
+			   htt_stats_buf->eapol_lookup_failed);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "qpeer_not_allow_data = %u",
+			   htt_stats_buf->qpeer_not_allow_data);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fse_tid_override = %u",
+			   htt_stats_buf->fse_tid_override);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ipv6_jumbogram_zero_length = %u",
+			   htt_stats_buf->ipv6_jumbogram_zero_length);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "qos_to_non_qos_in_prog = %u\n",
+			   htt_stats_buf->qos_to_non_qos_in_prog);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_de_classify_stats_tlv(const void *tag_buf,
+				   struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_de_classify_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_DE_CLASSIFY_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "arp_packets = %u",
+			   htt_stats_buf->arp_packets);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "igmp_packets = %u",
+			   htt_stats_buf->igmp_packets);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "dhcp_packets = %u",
+			   htt_stats_buf->dhcp_packets);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "host_inspected = %u",
+			   htt_stats_buf->host_inspected);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "htt_included = %u",
+			   htt_stats_buf->htt_included);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "htt_valid_mcs = %u",
+			   htt_stats_buf->htt_valid_mcs);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "htt_valid_nss = %u",
+			   htt_stats_buf->htt_valid_nss);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "htt_valid_preamble_type = %u",
+			   htt_stats_buf->htt_valid_preamble_type);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "htt_valid_chainmask = %u",
+			   htt_stats_buf->htt_valid_chainmask);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "htt_valid_guard_interval = %u",
+			   htt_stats_buf->htt_valid_guard_interval);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "htt_valid_retries = %u",
+			   htt_stats_buf->htt_valid_retries);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "htt_valid_bw_info = %u",
+			   htt_stats_buf->htt_valid_bw_info);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "htt_valid_power = %u",
+			   htt_stats_buf->htt_valid_power);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "htt_valid_key_flags = 0x%x",
+			   htt_stats_buf->htt_valid_key_flags);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "htt_valid_no_encryption = %u",
+			   htt_stats_buf->htt_valid_no_encryption);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fse_entry_count = %u",
+			   htt_stats_buf->fse_entry_count);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fse_priority_be = %u",
+			   htt_stats_buf->fse_priority_be);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fse_priority_high = %u",
+			   htt_stats_buf->fse_priority_high);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fse_priority_low = %u",
+			   htt_stats_buf->fse_priority_low);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fse_traffic_ptrn_be = %u",
+			   htt_stats_buf->fse_traffic_ptrn_be);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fse_traffic_ptrn_over_sub = %u",
+			   htt_stats_buf->fse_traffic_ptrn_over_sub);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fse_traffic_ptrn_bursty = %u",
+			   htt_stats_buf->fse_traffic_ptrn_bursty);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fse_traffic_ptrn_interactive = %u",
+			   htt_stats_buf->fse_traffic_ptrn_interactive);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fse_traffic_ptrn_periodic = %u",
+			   htt_stats_buf->fse_traffic_ptrn_periodic);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fse_hwqueue_alloc = %u",
+			   htt_stats_buf->fse_hwqueue_alloc);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fse_hwqueue_created = %u",
+			   htt_stats_buf->fse_hwqueue_created);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fse_hwqueue_send_to_host = %u",
+			   htt_stats_buf->fse_hwqueue_send_to_host);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mcast_entry = %u",
+			   htt_stats_buf->mcast_entry);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "bcast_entry = %u",
+			   htt_stats_buf->bcast_entry);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "htt_update_peer_cache = %u",
+			   htt_stats_buf->htt_update_peer_cache);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "htt_learning_frame = %u",
+			   htt_stats_buf->htt_learning_frame);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fse_invalid_peer = %u",
+			   htt_stats_buf->fse_invalid_peer);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mec_notify = %u\n",
+			   htt_stats_buf->mec_notify);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_de_classify_status_stats_tlv(const void *tag_buf,
+					  struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_de_classify_status_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "HTT_TX_DE_CLASSIFY_STATUS_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "eok = %u",
+			   htt_stats_buf->eok);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "classify_done = %u",
+			   htt_stats_buf->classify_done);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "lookup_failed = %u",
+			   htt_stats_buf->lookup_failed);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "send_host_dhcp = %u",
+			   htt_stats_buf->send_host_dhcp);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "send_host_mcast = %u",
+			   htt_stats_buf->send_host_mcast);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "send_host_unknown_dest = %u",
+			   htt_stats_buf->send_host_unknown_dest);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "send_host = %u",
+			   htt_stats_buf->send_host);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "status_invalid = %u\n",
+			   htt_stats_buf->status_invalid);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_de_enqueue_packets_stats_tlv(const void *tag_buf,
+					  struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_de_enqueue_packets_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "HTT_TX_DE_ENQUEUE_PACKETS_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "enqueued_pkts = %u",
+			htt_stats_buf->enqueued_pkts);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "to_tqm = %u",
+			htt_stats_buf->to_tqm);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "to_tqm_bypass = %u\n",
+			htt_stats_buf->to_tqm_bypass);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_de_enqueue_discard_stats_tlv(const void *tag_buf,
+					  struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_de_enqueue_discard_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "HTT_TX_DE_ENQUEUE_DISCARD_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "discarded_pkts = %u",
+			   htt_stats_buf->discarded_pkts);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "local_frames = %u",
+			   htt_stats_buf->local_frames);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "is_ext_msdu = %u\n",
+			   htt_stats_buf->is_ext_msdu);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_tx_de_compl_stats_tlv(const void *tag_buf,
+						   struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_de_compl_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_DE_COMPL_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tcl_dummy_frame = %u",
+			   htt_stats_buf->tcl_dummy_frame);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tqm_dummy_frame = %u",
+			   htt_stats_buf->tqm_dummy_frame);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tqm_notify_frame = %u",
+			   htt_stats_buf->tqm_notify_frame);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fw2wbm_enq = %u",
+			   htt_stats_buf->fw2wbm_enq);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tqm_bypass_frame = %u\n",
+			   htt_stats_buf->tqm_bypass_frame);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_de_fw2wbm_ring_full_hist_tlv(const void *tag_buf,
+					  u16 tag_len,
+					  struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_de_fw2wbm_ring_full_hist_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char fw2wbm_ring_full_hist[HTT_MAX_STRING_LEN] = {0};
+	u16  num_elements = tag_len >> 2;
+	u32  required_buffer_size = HTT_MAX_PRINT_CHAR_PER_ELEM * num_elements;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "HTT_TX_DE_FW2WBM_RING_FULL_HIST_TLV");
+
+	if (required_buffer_size < HTT_MAX_STRING_LEN) {
+		ARRAY_TO_STRING(fw2wbm_ring_full_hist,
+				htt_stats_buf->fw2wbm_ring_full_hist,
+				num_elements);
+		len += HTT_DBG_OUT(buf + len, buf_len - len,
+				   "fw2wbm_ring_full_hist = %s\n",
+				   fw2wbm_ring_full_hist);
+	} else {
+		len += HTT_DBG_OUT(buf + len, buf_len - len,
+				   "INSUFFICIENT PRINT BUFFER ");
+	}
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_tx_de_cmn_stats_tlv(const void *tag_buf, struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_de_cmn_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_DE_CMN_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mac_id = %u",
+			   htt_stats_buf->mac_id__word & 0xFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tcl2fw_entry_count = %u",
+			   htt_stats_buf->tcl2fw_entry_count);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "not_to_fw = %u",
+			   htt_stats_buf->not_to_fw);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "invalid_pdev_vdev_peer = %u",
+			   htt_stats_buf->invalid_pdev_vdev_peer);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tcl_res_invalid_addrx = %u",
+			   htt_stats_buf->tcl_res_invalid_addrx);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "wbm2fw_entry_count = %u",
+			   htt_stats_buf->wbm2fw_entry_count);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "invalid_pdev = %u\n",
+			   htt_stats_buf->invalid_pdev);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_ring_if_stats_tlv(const void *tag_buf,
+					       struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_ring_if_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char low_wm_hit_count[HTT_MAX_STRING_LEN] = {0};
+	char high_wm_hit_count[HTT_MAX_STRING_LEN] = {0};
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_RING_IF_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "base_addr = %u",
+			   htt_stats_buf->base_addr);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "elem_size = %u",
+			   htt_stats_buf->elem_size);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_elems = %u",
+			   htt_stats_buf->num_elems__prefetch_tail_idx & 0xFFFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "prefetch_tail_idx = %u",
+			   (htt_stats_buf->num_elems__prefetch_tail_idx &
+			   0xFFFF0000) >> 16);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "head_idx = %u",
+			   htt_stats_buf->head_idx__tail_idx & 0xFFFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tail_idx = %u",
+			   (htt_stats_buf->head_idx__tail_idx & 0xFFFF0000) >> 16);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "shadow_head_idx = %u",
+			   htt_stats_buf->shadow_head_idx__shadow_tail_idx & 0xFFFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "shadow_tail_idx = %u",
+			   (htt_stats_buf->shadow_head_idx__shadow_tail_idx &
+			   0xFFFF0000) >> 16);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_tail_incr = %u",
+			   htt_stats_buf->num_tail_incr);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "lwm_thresh = %u",
+			   htt_stats_buf->lwm_thresh__hwm_thresh & 0xFFFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "hwm_thresh = %u",
+			   (htt_stats_buf->lwm_thresh__hwm_thresh & 0xFFFF0000) >> 16);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "overrun_hit_count = %u",
+			   htt_stats_buf->overrun_hit_count);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "underrun_hit_count = %u",
+			   htt_stats_buf->underrun_hit_count);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "prod_blockwait_count = %u",
+			   htt_stats_buf->prod_blockwait_count);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "cons_blockwait_count = %u",
+			   htt_stats_buf->cons_blockwait_count);
+
+	ARRAY_TO_STRING(low_wm_hit_count, htt_stats_buf->low_wm_hit_count,
+			HTT_STATS_LOW_WM_BINS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "low_wm_hit_count = %s ",
+			   low_wm_hit_count);
+
+	ARRAY_TO_STRING(high_wm_hit_count, htt_stats_buf->high_wm_hit_count,
+			HTT_STATS_HIGH_WM_BINS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "high_wm_hit_count = %s\n",
+			   high_wm_hit_count);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_ring_if_cmn_tlv(const void *tag_buf,
+					     struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_ring_if_cmn_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_RING_IF_CMN_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mac_id = %u",
+			   htt_stats_buf->mac_id__word & 0xFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_records = %u\n",
+			   htt_stats_buf->num_records);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_sfm_client_user_tlv_v(const void *tag_buf,
+						   u16 tag_len,
+						   struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_sfm_client_user_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char dwords_used_by_user_n[HTT_MAX_STRING_LEN] = {0};
+	u16 num_elems = tag_len >> 2;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_SFM_CLIENT_USER_TLV_V:");
+
+	ARRAY_TO_STRING(dwords_used_by_user_n,
+			htt_stats_buf->dwords_used_by_user_n,
+			num_elems);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "dwords_used_by_user_n = %s\n",
+			   dwords_used_by_user_n);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_sfm_client_tlv(const void *tag_buf,
+					    struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_sfm_client_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_SFM_CLIENT_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "client_id = %u",
+			   htt_stats_buf->client_id);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "buf_min = %u",
+			   htt_stats_buf->buf_min);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "buf_max = %u",
+			   htt_stats_buf->buf_max);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "buf_busy = %u",
+			   htt_stats_buf->buf_busy);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "buf_alloc = %u",
+			   htt_stats_buf->buf_alloc);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "buf_avail = %u",
+			   htt_stats_buf->buf_avail);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_users = %u\n",
+			   htt_stats_buf->num_users);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_sfm_cmn_tlv(const void *tag_buf,
+					 struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_sfm_cmn_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_SFM_CMN_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mac_id = %u",
+			   htt_stats_buf->mac_id__word & 0xFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "buf_total = %u",
+			   htt_stats_buf->buf_total);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mem_empty = %u",
+			   htt_stats_buf->mem_empty);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "deallocate_bufs = %u",
+			   htt_stats_buf->deallocate_bufs);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_records = %u\n",
+			   htt_stats_buf->num_records);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_sring_stats_tlv(const void *tag_buf,
+					     struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_sring_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_SRING_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mac_id = %u",
+			   htt_stats_buf->mac_id__ring_id__arena__ep & 0xFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ring_id = %u",
+			   (htt_stats_buf->mac_id__ring_id__arena__ep & 0xFF00) >> 8);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "arena = %u",
+			   (htt_stats_buf->mac_id__ring_id__arena__ep & 0xFF0000) >> 16);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ep = %u",
+			   (htt_stats_buf->mac_id__ring_id__arena__ep & 0x1000000) >> 24);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "base_addr_lsb = 0x%x",
+			   htt_stats_buf->base_addr_lsb);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "base_addr_msb = 0x%x",
+			   htt_stats_buf->base_addr_msb);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ring_size = %u",
+			   htt_stats_buf->ring_size);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "elem_size = %u",
+			   htt_stats_buf->elem_size);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_avail_words = %u",
+			   htt_stats_buf->num_avail_words__num_valid_words & 0xFFFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_valid_words = %u",
+			   (htt_stats_buf->num_avail_words__num_valid_words &
+			   0xFFFF0000) >> 16);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "head_ptr = %u",
+			   htt_stats_buf->head_ptr__tail_ptr & 0xFFFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tail_ptr = %u",
+			   (htt_stats_buf->head_ptr__tail_ptr & 0xFFFF0000) >> 16);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "consumer_empty = %u",
+			   htt_stats_buf->consumer_empty__producer_full & 0xFFFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "producer_full = %u",
+			   (htt_stats_buf->consumer_empty__producer_full &
+			   0xFFFF0000) >> 16);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "prefetch_count = %u",
+			   htt_stats_buf->prefetch_count__internal_tail_ptr & 0xFFFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "internal_tail_ptr = %u\n",
+			   (htt_stats_buf->prefetch_count__internal_tail_ptr &
+			   0xFFFF0000) >> 16);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_sring_cmn_tlv(const void *tag_buf,
+					   struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_sring_cmn_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_SRING_CMN_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_records = %u\n",
+			   htt_stats_buf->num_records);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_tx_pdev_rate_stats_tlv(const void *tag_buf,
+						    struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_tx_pdev_rate_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	u8 j;
+	char str_buf[HTT_MAX_STRING_LEN] = {0};
+	char *tx_gi[HTT_TX_PEER_STATS_NUM_GI_COUNTERS];
+
+	for (j = 0; j < HTT_TX_PEER_STATS_NUM_GI_COUNTERS; j++)
+		tx_gi[j] = kmalloc(HTT_MAX_STRING_LEN, GFP_ATOMIC);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_PDEV_RATE_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mac_id = %u",
+			   htt_stats_buf->mac_id__word & 0xFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_ldpc = %u",
+			   htt_stats_buf->tx_ldpc);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ac_mu_mimo_tx_ldpc = %u",
+			   htt_stats_buf->ac_mu_mimo_tx_ldpc);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_mu_mimo_tx_ldpc = %u",
+			   htt_stats_buf->ax_mu_mimo_tx_ldpc);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ofdma_tx_ldpc = %u",
+			   htt_stats_buf->ofdma_tx_ldpc);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rts_cnt = %u",
+			   htt_stats_buf->rts_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rts_success = %u",
+			   htt_stats_buf->rts_success);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ack_rssi = %u",
+			   htt_stats_buf->ack_rssi);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "Legacy CCK Rates: 1 Mbps: %u, 2 Mbps: %u, 5.5 Mbps: %u, 11 Mbps: %u",
+			   htt_stats_buf->tx_legacy_cck_rate[0],
+			   htt_stats_buf->tx_legacy_cck_rate[1],
+			   htt_stats_buf->tx_legacy_cck_rate[2],
+			   htt_stats_buf->tx_legacy_cck_rate[3]);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "Legacy OFDM Rates: 6 Mbps: %u, 9 Mbps: %u, 12 Mbps: %u, 18 Mbps: %u\n"
+			   "                   24 Mbps: %u, 36 Mbps: %u, 48 Mbps: %u, 54 Mbps: %u",
+			   htt_stats_buf->tx_legacy_ofdm_rate[0],
+			   htt_stats_buf->tx_legacy_ofdm_rate[1],
+			   htt_stats_buf->tx_legacy_ofdm_rate[2],
+			   htt_stats_buf->tx_legacy_ofdm_rate[3],
+			   htt_stats_buf->tx_legacy_ofdm_rate[4],
+			   htt_stats_buf->tx_legacy_ofdm_rate[5],
+			   htt_stats_buf->tx_legacy_ofdm_rate[6],
+			   htt_stats_buf->tx_legacy_ofdm_rate[7]);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->tx_mcs,
+			HTT_TX_PDEV_STATS_NUM_MCS_COUNTERS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_mcs = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->ac_mu_mimo_tx_mcs,
+			HTT_TX_PDEV_STATS_NUM_MCS_COUNTERS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ac_mu_mimo_tx_mcs = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->ax_mu_mimo_tx_mcs,
+			HTT_TX_PDEV_STATS_NUM_MCS_COUNTERS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_mu_mimo_tx_mcs = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->ofdma_tx_mcs,
+			HTT_TX_PDEV_STATS_NUM_MCS_COUNTERS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ofdma_tx_mcs = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->tx_nss,
+			HTT_TX_PDEV_STATS_NUM_SPATIAL_STREAMS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_nss = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->ac_mu_mimo_tx_nss,
+			HTT_TX_PDEV_STATS_NUM_SPATIAL_STREAMS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ac_mu_mimo_tx_nss = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->ax_mu_mimo_tx_nss,
+			HTT_TX_PDEV_STATS_NUM_SPATIAL_STREAMS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_mu_mimo_tx_nss = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->ofdma_tx_nss,
+			HTT_TX_PDEV_STATS_NUM_SPATIAL_STREAMS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ofdma_tx_nss = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->tx_bw,
+			HTT_TX_PDEV_STATS_NUM_BW_COUNTERS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_bw = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->ac_mu_mimo_tx_bw,
+			HTT_TX_PDEV_STATS_NUM_BW_COUNTERS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ac_mu_mimo_tx_bw = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->ax_mu_mimo_tx_bw,
+			HTT_TX_PDEV_STATS_NUM_BW_COUNTERS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ax_mu_mimo_tx_bw = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->ofdma_tx_bw,
+			HTT_TX_PDEV_STATS_NUM_BW_COUNTERS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ofdma_tx_bw = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->tx_stbc,
+			HTT_TX_PDEV_STATS_NUM_MCS_COUNTERS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_stbc = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->tx_pream,
+			HTT_TX_PDEV_STATS_NUM_PREAMBLE_TYPES);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_pream = %s ", str_buf);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HE LTF: 1x: %u, 2x: %u, 4x: %u",
+			   htt_stats_buf->tx_he_ltf[1],
+			   htt_stats_buf->tx_he_ltf[2],
+			   htt_stats_buf->tx_he_ltf[3]);
+
+	/* SU GI Stats */
+	for (j = 0; j < HTT_TX_PDEV_STATS_NUM_GI_COUNTERS; j++) {
+		ARRAY_TO_STRING(tx_gi[j], htt_stats_buf->tx_gi[j],
+				HTT_TX_PDEV_STATS_NUM_MCS_COUNTERS);
+		len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_gi[%u] = %s ",
+				   j, tx_gi[j]);
+	}
+
+	/* AC MU-MIMO GI Stats */
+	for (j = 0; j < HTT_TX_PDEV_STATS_NUM_GI_COUNTERS; j++) {
+		ARRAY_TO_STRING(tx_gi[j], htt_stats_buf->ac_mu_mimo_tx_gi[j],
+				HTT_TX_PDEV_STATS_NUM_MCS_COUNTERS);
+		len += HTT_DBG_OUT(buf + len, buf_len - len,
+				   "ac_mu_mimo_tx_gi[%u] = %s ",
+				   j, tx_gi[j]);
+	}
+
+	/* AX MU-MIMO GI Stats */
+	for (j = 0; j < HTT_TX_PDEV_STATS_NUM_GI_COUNTERS; j++) {
+		ARRAY_TO_STRING(tx_gi[j], htt_stats_buf->ax_mu_mimo_tx_gi[j],
+				HTT_TX_PDEV_STATS_NUM_MCS_COUNTERS);
+		len += HTT_DBG_OUT(buf + len, buf_len - len,
+				   "ax_mu_mimo_tx_gi[%u] = %s ",
+				   j, tx_gi[j]);
+	}
+
+	/* DL OFDMA GI Stats */
+	for (j = 0; j < HTT_TX_PDEV_STATS_NUM_GI_COUNTERS; j++) {
+		ARRAY_TO_STRING(tx_gi[j], htt_stats_buf->ofdma_tx_gi[j],
+				HTT_TX_PDEV_STATS_NUM_MCS_COUNTERS);
+		len += HTT_DBG_OUT(buf + len, buf_len - len, "ofdma_tx_gi[%u] = %s ",
+				   j, tx_gi[j]);
+	}
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->tx_dcm,
+			HTT_TX_PDEV_STATS_NUM_DCM_COUNTERS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_dcm = %s\n", str_buf);
+
+	for (j = 0; j < HTT_TX_PEER_STATS_NUM_GI_COUNTERS; j++)
+		kfree(tx_gi[j]);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_rx_pdev_rate_stats_tlv(const void *tag_buf,
+						    struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_rx_pdev_rate_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	u8 j;
+	char *rssi_chain[HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS];
+	char *rx_gi[HTT_RX_PDEV_STATS_NUM_GI_COUNTERS];
+	char str_buf[HTT_MAX_STRING_LEN] = {0};
+
+	for (j = 0; j < HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS; j++)
+		rssi_chain[j] = kmalloc(HTT_MAX_STRING_LEN, GFP_ATOMIC);
+
+	for (j = 0; j < HTT_RX_PDEV_STATS_NUM_GI_COUNTERS; j++)
+		rx_gi[j] = kmalloc(HTT_MAX_STRING_LEN, GFP_ATOMIC);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_RX_PDEV_RATE_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mac_id = %u",
+			   htt_stats_buf->mac_id__word & 0xFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "nsts = %u",
+			   htt_stats_buf->nsts);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_ldpc = %u",
+			   htt_stats_buf->rx_ldpc);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rts_cnt = %u",
+			   htt_stats_buf->rts_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rssi_mgmt = %u",
+			   htt_stats_buf->rssi_mgmt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rssi_data = %u",
+			   htt_stats_buf->rssi_data);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rssi_comb = %u",
+			   htt_stats_buf->rssi_comb);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rssi_in_dbm = %d",
+			   htt_stats_buf->rssi_in_dbm);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->rx_mcs,
+			HTT_RX_PDEV_STATS_NUM_MCS_COUNTERS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_mcs = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->rx_nss,
+			HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_nss = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->rx_dcm,
+			HTT_RX_PDEV_STATS_NUM_DCM_COUNTERS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_dcm = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->rx_stbc,
+			HTT_RX_PDEV_STATS_NUM_MCS_COUNTERS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_stbc = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->rx_bw,
+			HTT_RX_PDEV_STATS_NUM_BW_COUNTERS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_bw = %s ", str_buf);
+
+	for (j = 0; j < HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS; j++) {
+		ARRAY_TO_STRING(rssi_chain[j], htt_stats_buf->rssi_chain[j],
+				HTT_RX_PDEV_STATS_NUM_BW_COUNTERS);
+		len += HTT_DBG_OUT(buf + len, buf_len - len, "rssi_chain[%u] = %s ",
+				   j, rssi_chain[j]);
+	}
+
+	for (j = 0; j < HTT_RX_PDEV_STATS_NUM_GI_COUNTERS; j++) {
+		ARRAY_TO_STRING(rx_gi[j], htt_stats_buf->rx_gi[j],
+				HTT_RX_PDEV_STATS_NUM_MCS_COUNTERS);
+		len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_gi[%u] = %s ",
+				   j, rx_gi[j]);
+	}
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->rx_pream,
+			HTT_RX_PDEV_STATS_NUM_PREAMBLE_TYPES);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_pream = %s", str_buf);
+
+	for (j = 0; j < HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS; j++)
+		kfree(rssi_chain[j]);
+
+	for (j = 0; j < HTT_RX_PDEV_STATS_NUM_GI_COUNTERS; j++)
+		kfree(rx_gi[j]);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_11ax_su_ext = %u",
+			   htt_stats_buf->rx_11ax_su_ext);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_11ac_mumimo = %u",
+			   htt_stats_buf->rx_11ac_mumimo);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_11ax_mumimo = %u",
+			   htt_stats_buf->rx_11ax_mumimo);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_11ax_ofdma = %u",
+			   htt_stats_buf->rx_11ax_ofdma);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "txbf = %u",
+			   htt_stats_buf->txbf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->rx_legacy_cck_rate,
+			HTT_RX_PDEV_STATS_NUM_LEGACY_CCK_STATS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_legacy_cck_rate = %s ",
+			   str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->rx_legacy_ofdm_rate,
+			HTT_RX_PDEV_STATS_NUM_LEGACY_OFDM_STATS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_legacy_ofdm_rate = %s ",
+			   str_buf);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_active_dur_us_low = %u",
+			   htt_stats_buf->rx_active_dur_us_low);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_active_dur_us_high = %u\n",
+			   htt_stats_buf->rx_active_dur_us_high);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_rx_soc_fw_stats_tlv(const void *tag_buf,
+						 struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_rx_soc_fw_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_RX_SOC_FW_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fw_reo_ring_data_msdu = %u",
+			   htt_stats_buf->fw_reo_ring_data_msdu);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fw_to_host_data_msdu_bcmc = %u",
+			   htt_stats_buf->fw_to_host_data_msdu_bcmc);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fw_to_host_data_msdu_uc = %u",
+			   htt_stats_buf->fw_to_host_data_msdu_uc);
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "ofld_remote_data_buf_recycle_cnt = %u",
+			   htt_stats_buf->ofld_remote_data_buf_recycle_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "ofld_remote_free_buf_indication_cnt = %u",
+			   htt_stats_buf->ofld_remote_free_buf_indication_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "ofld_buf_to_host_data_msdu_uc = %u",
+			   htt_stats_buf->ofld_buf_to_host_data_msdu_uc);
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "reo_fw_ring_to_host_data_msdu_uc = %u",
+			   htt_stats_buf->reo_fw_ring_to_host_data_msdu_uc);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "wbm_sw_ring_reap = %u",
+			   htt_stats_buf->wbm_sw_ring_reap);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "wbm_forward_to_host_cnt = %u",
+			   htt_stats_buf->wbm_forward_to_host_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "wbm_target_recycle_cnt = %u",
+			   htt_stats_buf->wbm_target_recycle_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "target_refill_ring_recycle_cnt = %u",
+			   htt_stats_buf->target_refill_ring_recycle_cnt);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_rx_soc_fw_refill_ring_empty_tlv_v(const void *tag_buf,
+					    u16 tag_len,
+					    struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_rx_soc_fw_refill_ring_empty_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char refill_ring_empty_cnt[HTT_MAX_STRING_LEN] = {0};
+	u16 num_elems = min_t(u16, (tag_len >> 2), HTT_RX_STATS_REFILL_MAX_RING);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "HTT_RX_SOC_FW_REFILL_RING_EMPTY_TLV_V:");
+
+	ARRAY_TO_STRING(refill_ring_empty_cnt,
+			htt_stats_buf->refill_ring_empty_cnt,
+			num_elems);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "refill_ring_empty_cnt = %s\n",
+			   refill_ring_empty_cnt);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_rx_soc_fw_refill_ring_num_rxdma_err_tlv_v(const void *tag_buf,
+						    u16 tag_len,
+						    struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_rx_soc_fw_refill_ring_num_rxdma_err_tlv_v *htt_stats_buf =
+		tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char rxdma_err_cnt[HTT_MAX_STRING_LEN] = {0};
+	u16 num_elems = min_t(u16, (tag_len >> 2), HTT_RX_RXDMA_MAX_ERR_CODE);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "HTT_RX_SOC_FW_REFILL_RING_NUM_RXDMA_ERR_TLV_V:");
+
+	ARRAY_TO_STRING(rxdma_err_cnt,
+			htt_stats_buf->rxdma_err,
+			num_elems);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rxdma_err = %s\n",
+			   rxdma_err_cnt);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_rx_soc_fw_refill_ring_num_reo_err_tlv_v(const void *tag_buf,
+						  u16 tag_len,
+						  struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_rx_soc_fw_refill_ring_num_reo_err_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char reo_err_cnt[HTT_MAX_STRING_LEN] = {0};
+	u16 num_elems = min_t(u16, (tag_len >> 2), HTT_RX_REO_MAX_ERR_CODE);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "HTT_RX_SOC_FW_REFILL_RING_NUM_REO_ERR_TLV_V:");
+
+	ARRAY_TO_STRING(reo_err_cnt,
+			htt_stats_buf->reo_err,
+			num_elems);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "reo_err = %s\n",
+			   reo_err_cnt);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_rx_reo_debug_stats_tlv_v(const void *tag_buf,
+				   struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_rx_reo_resource_stats_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_RX_REO_RESOURCE_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sample_id = %u",
+			   htt_stats_buf->sample_id);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "total_max = %u",
+			   htt_stats_buf->total_max);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "total_avg = %u",
+			   htt_stats_buf->total_avg);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "total_sample = %u",
+			   htt_stats_buf->total_sample);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "non_zeros_avg = %u",
+			   htt_stats_buf->non_zeros_avg);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "non_zeros_sample = %u",
+			   htt_stats_buf->non_zeros_sample);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "last_non_zeros_max = %u",
+			   htt_stats_buf->last_non_zeros_max);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "last_non_zeros_min %u",
+			   htt_stats_buf->last_non_zeros_min);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "last_non_zeros_avg %u",
+			   htt_stats_buf->last_non_zeros_avg);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "last_non_zeros_sample %u\n",
+			   htt_stats_buf->last_non_zeros_sample);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_rx_soc_fw_refill_ring_num_refill_tlv_v(const void *tag_buf,
+						 u16 tag_len,
+						 struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_rx_soc_fw_refill_ring_num_refill_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char refill_ring_num_refill[HTT_MAX_STRING_LEN] = {0};
+	u16 num_elems = min_t(u16, (tag_len >> 2), HTT_RX_STATS_REFILL_MAX_RING);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "HTT_RX_SOC_FW_REFILL_RING_NUM_REFILL_TLV_V:");
+
+	ARRAY_TO_STRING(refill_ring_num_refill,
+			htt_stats_buf->refill_ring_num_refill,
+			num_elems);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "refill_ring_num_refill = %s\n",
+			   refill_ring_num_refill);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_rx_pdev_fw_stats_tlv(const void *tag_buf,
+						  struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_rx_pdev_fw_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char fw_ring_mgmt_subtype[HTT_MAX_STRING_LEN] = {0};
+	char fw_ring_ctrl_subtype[HTT_MAX_STRING_LEN] = {0};
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_RX_PDEV_FW_STATS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mac_id = %u",
+			   htt_stats_buf->mac_id__word & 0xFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ppdu_recvd = %u",
+			   htt_stats_buf->ppdu_recvd);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mpdu_cnt_fcs_ok = %u",
+			   htt_stats_buf->mpdu_cnt_fcs_ok);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mpdu_cnt_fcs_err = %u",
+			   htt_stats_buf->mpdu_cnt_fcs_err);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tcp_msdu_cnt = %u",
+			   htt_stats_buf->tcp_msdu_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tcp_ack_msdu_cnt = %u",
+			   htt_stats_buf->tcp_ack_msdu_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "udp_msdu_cnt = %u",
+			   htt_stats_buf->udp_msdu_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "other_msdu_cnt = %u",
+			   htt_stats_buf->other_msdu_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fw_ring_mpdu_ind = %u",
+			   htt_stats_buf->fw_ring_mpdu_ind);
+
+	ARRAY_TO_STRING(fw_ring_mgmt_subtype,
+			htt_stats_buf->fw_ring_mgmt_subtype,
+			HTT_STATS_SUBTYPE_MAX);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fw_ring_mgmt_subtype = %s ",
+			   fw_ring_mgmt_subtype);
+
+	ARRAY_TO_STRING(fw_ring_ctrl_subtype,
+			htt_stats_buf->fw_ring_ctrl_subtype,
+			HTT_STATS_SUBTYPE_MAX);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fw_ring_ctrl_subtype = %s ",
+			   fw_ring_ctrl_subtype);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fw_ring_mcast_data_msdu = %u",
+			   htt_stats_buf->fw_ring_mcast_data_msdu);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fw_ring_bcast_data_msdu = %u",
+			   htt_stats_buf->fw_ring_bcast_data_msdu);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fw_ring_ucast_data_msdu = %u",
+			   htt_stats_buf->fw_ring_ucast_data_msdu);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fw_ring_null_data_msdu = %u",
+			   htt_stats_buf->fw_ring_null_data_msdu);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fw_ring_mpdu_drop = %u",
+			   htt_stats_buf->fw_ring_mpdu_drop);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ofld_local_data_ind_cnt = %u",
+			   htt_stats_buf->ofld_local_data_ind_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "ofld_local_data_buf_recycle_cnt = %u",
+			   htt_stats_buf->ofld_local_data_buf_recycle_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "drx_local_data_ind_cnt = %u",
+			   htt_stats_buf->drx_local_data_ind_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "drx_local_data_buf_recycle_cnt = %u",
+			   htt_stats_buf->drx_local_data_buf_recycle_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "local_nondata_ind_cnt = %u",
+			   htt_stats_buf->local_nondata_ind_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "local_nondata_buf_recycle_cnt = %u",
+			   htt_stats_buf->local_nondata_buf_recycle_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fw_status_buf_ring_refill_cnt = %u",
+			   htt_stats_buf->fw_status_buf_ring_refill_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fw_status_buf_ring_empty_cnt = %u",
+			   htt_stats_buf->fw_status_buf_ring_empty_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fw_pkt_buf_ring_refill_cnt = %u",
+			   htt_stats_buf->fw_pkt_buf_ring_refill_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fw_pkt_buf_ring_empty_cnt = %u",
+			   htt_stats_buf->fw_pkt_buf_ring_empty_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fw_link_buf_ring_refill_cnt = %u",
+			   htt_stats_buf->fw_link_buf_ring_refill_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fw_link_buf_ring_empty_cnt = %u",
+			   htt_stats_buf->fw_link_buf_ring_empty_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "host_pkt_buf_ring_refill_cnt = %u",
+			   htt_stats_buf->host_pkt_buf_ring_refill_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "host_pkt_buf_ring_empty_cnt = %u",
+			   htt_stats_buf->host_pkt_buf_ring_empty_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mon_pkt_buf_ring_refill_cnt = %u",
+			   htt_stats_buf->mon_pkt_buf_ring_refill_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mon_pkt_buf_ring_empty_cnt = %u",
+			   htt_stats_buf->mon_pkt_buf_ring_empty_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "mon_status_buf_ring_refill_cnt = %u",
+			   htt_stats_buf->mon_status_buf_ring_refill_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mon_status_buf_ring_empty_cnt = %u",
+			   htt_stats_buf->mon_status_buf_ring_empty_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mon_desc_buf_ring_refill_cnt = %u",
+			   htt_stats_buf->mon_desc_buf_ring_refill_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mon_desc_buf_ring_empty_cnt = %u",
+			   htt_stats_buf->mon_desc_buf_ring_empty_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mon_dest_ring_update_cnt = %u",
+			   htt_stats_buf->mon_dest_ring_update_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mon_dest_ring_full_cnt = %u",
+			   htt_stats_buf->mon_dest_ring_full_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_suspend_cnt = %u",
+			   htt_stats_buf->rx_suspend_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_suspend_fail_cnt = %u",
+			   htt_stats_buf->rx_suspend_fail_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_resume_cnt = %u",
+			   htt_stats_buf->rx_resume_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_resume_fail_cnt = %u",
+			   htt_stats_buf->rx_resume_fail_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_ring_switch_cnt = %u",
+			   htt_stats_buf->rx_ring_switch_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_ring_restore_cnt = %u",
+			   htt_stats_buf->rx_ring_restore_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_flush_cnt = %u",
+			   htt_stats_buf->rx_flush_cnt);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_recovery_reset_cnt = %u\n",
+			   htt_stats_buf->rx_recovery_reset_cnt);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_rx_pdev_fw_ring_mpdu_err_tlv_v(const void *tag_buf,
+					 struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_rx_pdev_fw_ring_mpdu_err_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char fw_ring_mpdu_err[HTT_MAX_STRING_LEN] = {0};
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "HTT_RX_PDEV_FW_RING_MPDU_ERR_TLV_V:");
+
+	ARRAY_TO_STRING(fw_ring_mpdu_err,
+			htt_stats_buf->fw_ring_mpdu_err,
+			HTT_RX_STATS_RXDMA_MAX_ERR);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fw_ring_mpdu_err = %s\n",
+			   fw_ring_mpdu_err);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_rx_pdev_fw_mpdu_drop_tlv_v(const void *tag_buf,
+				     u16 tag_len,
+				     struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_rx_pdev_fw_mpdu_drop_tlv_v *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char fw_mpdu_drop[HTT_MAX_STRING_LEN] = {0};
+	u16 num_elems = min_t(u16, (tag_len >> 2), HTT_RX_STATS_FW_DROP_REASON_MAX);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_RX_PDEV_FW_MPDU_DROP_TLV_V:");
+
+	ARRAY_TO_STRING(fw_mpdu_drop,
+			htt_stats_buf->fw_mpdu_drop,
+			num_elems);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "fw_mpdu_drop = %s\n", fw_mpdu_drop);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_rx_pdev_fw_stats_phy_err_tlv(const void *tag_buf,
+				       struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_rx_pdev_fw_stats_phy_err_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	char phy_errs[HTT_MAX_STRING_LEN] = {0};
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_RX_PDEV_FW_STATS_PHY_ERR_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mac_id__word = %u",
+			   htt_stats_buf->mac_id__word);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "tota_phy_err_nct = %u",
+			   htt_stats_buf->total_phy_err_cnt);
+
+	ARRAY_TO_STRING(phy_errs,
+			htt_stats_buf->phy_err,
+			HTT_STATS_PHY_ERR_MAX);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "phy_errs = %s\n", phy_errs);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_pdev_cca_stats_hist_tlv(const void *tag_buf,
+				  struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_pdev_cca_stats_hist_v1_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "\nHTT_PDEV_CCA_STATS_HIST_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "chan_num = %u",
+			   htt_stats_buf->chan_num);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_records = %u",
+			   htt_stats_buf->num_records);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "valid_cca_counters_bitmap = 0x%x",
+			   htt_stats_buf->valid_cca_counters_bitmap);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "collection_interval = %u\n",
+			   htt_stats_buf->collection_interval);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "HTT_PDEV_STATS_CCA_COUNTERS_TLV:(in usec)");
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "|  tx_frame|   rx_frame|   rx_clear| my_rx_frame|        cnt| med_rx_idle| med_tx_idle_global|   cca_obss|");
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_pdev_stats_cca_counters_tlv(const void *tag_buf,
+				      struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_pdev_stats_cca_counters_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "|%10u| %10u| %10u| %11u| %10u| %11u| %18u| %10u|",
+			   htt_stats_buf->tx_frame_usec,
+			   htt_stats_buf->rx_frame_usec,
+			   htt_stats_buf->rx_clear_usec,
+			   htt_stats_buf->my_rx_frame_usec,
+			   htt_stats_buf->usec_cnt,
+			   htt_stats_buf->med_rx_idle_usec,
+			   htt_stats_buf->med_tx_idle_global_usec,
+			   htt_stats_buf->cca_obss_usec);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_print_hw_stats_whal_tx_tlv(const void *tag_buf,
+						  struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_hw_stats_whal_tx_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_HW_STATS_WHAL_TX_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "mac_id = %u",
+			   htt_stats_buf->mac_id__word & 0xFF);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "last_unpause_ppdu_id = %u",
+			   htt_stats_buf->last_unpause_ppdu_id);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "hwsch_unpause_wait_tqm_write = %u",
+			   htt_stats_buf->hwsch_unpause_wait_tqm_write);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "hwsch_dummy_tlv_skipped = %u",
+			   htt_stats_buf->hwsch_dummy_tlv_skipped);
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "hwsch_misaligned_offset_received = %u",
+			   htt_stats_buf->hwsch_misaligned_offset_received);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "hwsch_reset_count = %u",
+			   htt_stats_buf->hwsch_reset_count);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "hwsch_dev_reset_war = %u",
+			   htt_stats_buf->hwsch_dev_reset_war);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "hwsch_delayed_pause = %u",
+			   htt_stats_buf->hwsch_delayed_pause);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "hwsch_long_delayed_pause = %u",
+			   htt_stats_buf->hwsch_long_delayed_pause);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sch_rx_ppdu_no_response = %u",
+			   htt_stats_buf->sch_rx_ppdu_no_response);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sch_selfgen_response = %u",
+			   htt_stats_buf->sch_selfgen_response);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sch_rx_sifs_resp_trigger= %u\n",
+			   htt_stats_buf->sch_rx_sifs_resp_trigger);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_pdev_stats_twt_sessions_tlv(const void *tag_buf,
+				      struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_pdev_stats_twt_sessions_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_PDEV_STATS_TWT_SESSIONS_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "pdev_id = %u",
+			   htt_stats_buf->pdev_id);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "num_sessions = %u\n",
+			   htt_stats_buf->num_sessions);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_pdev_stats_twt_session_tlv(const void *tag_buf,
+				     struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_pdev_stats_twt_session_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_PDEV_STATS_TWT_SESSION_TLV:");
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "vdev_id = %u",
+			   htt_stats_buf->vdev_id);
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "peer_mac = %02x:%02x:%02x:%02x:%02x:%02x",
+			   htt_stats_buf->peer_mac.mac_addr_l32 & 0xFF,
+			   (htt_stats_buf->peer_mac.mac_addr_l32 & 0xFF00) >> 8,
+			   (htt_stats_buf->peer_mac.mac_addr_l32 & 0xFF0000) >> 16,
+			   (htt_stats_buf->peer_mac.mac_addr_l32 & 0xFF000000) >> 24,
+			   (htt_stats_buf->peer_mac.mac_addr_h16 & 0xFF),
+			   (htt_stats_buf->peer_mac.mac_addr_h16 & 0xFF00) >> 8);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "flow_id_flags = %u",
+			   htt_stats_buf->flow_id_flags);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "dialog_id = %u",
+			   htt_stats_buf->dialog_id);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "wake_dura_us = %u",
+			   htt_stats_buf->wake_dura_us);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "wake_intvl_us = %u",
+			   htt_stats_buf->wake_intvl_us);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "sp_offset_us = %u\n",
+			   htt_stats_buf->sp_offset_us);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void
+htt_print_pdev_obss_pd_stats_tlv_v(const void *tag_buf,
+				   struct debug_htt_stats_req *stats_req)
+{
+	const struct htt_pdev_obss_pd_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "OBSS Tx success PPDU = %u",
+			   htt_stats_buf->num_obss_tx_ppdu_success);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "OBSS Tx failures PPDU = %u\n",
+			   htt_stats_buf->num_obss_tx_ppdu_failure);
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static inline void htt_htt_stats_debug_dump(const u32 *tag_buf,
+					    struct debug_htt_stats_req *stats_req)
+{
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
+	u32 tlv_len = 0, i = 0, word_len = 0;
+
+	tlv_len  = FIELD_GET(HTT_TLV_LEN, *tag_buf) + HTT_TLV_HDR_LEN;
+	word_len = (tlv_len % 4) == 0 ? (tlv_len / 4) : ((tlv_len / 4) + 1);
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "============================================");
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "HKDBG TLV DUMP: (tag_len=%u bytes, words=%u)",
+			   tlv_len, word_len);
+
+	for (i = 0; i + 3 < word_len; i += 4) {
+		len += HTT_DBG_OUT(buf + len, buf_len - len,
+				   "0x%08x 0x%08x 0x%08x 0x%08x",
+				   tag_buf[i], tag_buf[i + 1],
+				   tag_buf[i + 2], tag_buf[i + 3]);
+	}
+
+	if (i + 3 == word_len) {
+		len += HTT_DBG_OUT(buf + len, buf_len - len, "0x%08x 0x%08x 0x%08x ",
+				tag_buf[i], tag_buf[i + 1], tag_buf[i + 2]);
+	} else if (i + 2 == word_len) {
+		len += HTT_DBG_OUT(buf + len, buf_len - len, "0x%08x 0x%08x ",
+				tag_buf[i], tag_buf[i + 1]);
+	} else if (i + 1 == word_len) {
+		len += HTT_DBG_OUT(buf + len, buf_len - len, "0x%08x ",
+				tag_buf[i]);
+	}
+	len += HTT_DBG_OUT(buf + len, buf_len - len,
+			   "============================================");
+
+	if (len >= buf_len)
+		buf[buf_len - 1] = 0;
+	else
+		buf[len] = 0;
+
+	stats_req->buf_len = len;
+}
+
+static int ath11k_dbg_htt_ext_stats_parse(struct ath11k_base *ab,
+					  u16 tag, u16 len, const void *tag_buf,
+					  void *user_data)
+{
+	struct debug_htt_stats_req *stats_req = user_data;
+
+	switch (tag) {
+	case HTT_STATS_TX_PDEV_CMN_TAG:
+		htt_print_tx_pdev_stats_cmn_tlv(tag_buf, stats_req);
+		break;
+	case HTT_STATS_TX_PDEV_UNDERRUN_TAG:
+		htt_print_tx_pdev_stats_urrn_tlv_v(tag_buf, len, stats_req);
+		break;
+	case HTT_STATS_TX_PDEV_SIFS_TAG:
+		htt_print_tx_pdev_stats_sifs_tlv_v(tag_buf, len, stats_req);
+		break;
+	case HTT_STATS_TX_PDEV_FLUSH_TAG:
+		htt_print_tx_pdev_stats_flush_tlv_v(tag_buf, len, stats_req);
+		break;
+	case HTT_STATS_TX_PDEV_PHY_ERR_TAG:
+		htt_print_tx_pdev_stats_phy_err_tlv_v(tag_buf, len, stats_req);
+		break;
+	case HTT_STATS_TX_PDEV_SIFS_HIST_TAG:
+		htt_print_tx_pdev_stats_sifs_hist_tlv_v(tag_buf, len, stats_req);
+		break;
+
+	case HTT_STATS_TX_PDEV_TX_PPDU_STATS_TAG:
+		htt_print_tx_pdev_stats_tx_ppdu_stats_tlv_v(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_PDEV_TRIED_MPDU_CNT_HIST_TAG:
+		htt_print_tx_pdev_stats_tried_mpdu_cnt_hist_tlv_v(tag_buf, len,
+								  stats_req);
+		break;
+
+	case HTT_STATS_STRING_TAG:
+		htt_print_stats_string_tlv(tag_buf, len, stats_req);
+		break;
+
+	case HTT_STATS_TX_HWQ_CMN_TAG:
+		htt_print_tx_hwq_stats_cmn_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_HWQ_DIFS_LATENCY_TAG:
+		htt_print_tx_hwq_difs_latency_stats_tlv_v(tag_buf, len, stats_req);
+		break;
+
+	case HTT_STATS_TX_HWQ_CMD_RESULT_TAG:
+		htt_print_tx_hwq_cmd_result_stats_tlv_v(tag_buf, len, stats_req);
+		break;
+
+	case HTT_STATS_TX_HWQ_CMD_STALL_TAG:
+		htt_print_tx_hwq_cmd_stall_stats_tlv_v(tag_buf, len, stats_req);
+		break;
+
+	case HTT_STATS_TX_HWQ_FES_STATUS_TAG:
+		htt_print_tx_hwq_fes_result_stats_tlv_v(tag_buf, len, stats_req);
+		break;
+
+	case HTT_STATS_TX_HWQ_TRIED_MPDU_CNT_HIST_TAG:
+		htt_print_tx_hwq_tried_mpdu_cnt_hist_tlv_v(tag_buf, len, stats_req);
+		break;
+
+	case HTT_STATS_TX_HWQ_TXOP_USED_CNT_HIST_TAG:
+		htt_print_tx_hwq_txop_used_cnt_hist_tlv_v(tag_buf, len, stats_req);
+		break;
+	case HTT_STATS_TX_TQM_GEN_MPDU_TAG:
+		htt_print_tx_tqm_gen_mpdu_stats_tlv_v(tag_buf, len, stats_req);
+		break;
+
+	case HTT_STATS_TX_TQM_LIST_MPDU_TAG:
+		htt_print_tx_tqm_list_mpdu_stats_tlv_v(tag_buf, len, stats_req);
+		break;
+
+	case HTT_STATS_TX_TQM_LIST_MPDU_CNT_TAG:
+		htt_print_tx_tqm_list_mpdu_cnt_tlv_v(tag_buf, len, stats_req);
+		break;
+
+	case HTT_STATS_TX_TQM_CMN_TAG:
+		htt_print_tx_tqm_cmn_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_TQM_PDEV_TAG:
+		htt_print_tx_tqm_pdev_stats_tlv_v(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_TQM_CMDQ_STATUS_TAG:
+		htt_print_tx_tqm_cmdq_status_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_DE_EAPOL_PACKETS_TAG:
+		htt_print_tx_de_eapol_packets_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_DE_CLASSIFY_FAILED_TAG:
+		htt_print_tx_de_classify_failed_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_DE_CLASSIFY_STATS_TAG:
+		htt_print_tx_de_classify_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_DE_CLASSIFY_STATUS_TAG:
+		htt_print_tx_de_classify_status_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_DE_ENQUEUE_PACKETS_TAG:
+		htt_print_tx_de_enqueue_packets_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_DE_ENQUEUE_DISCARD_TAG:
+		htt_print_tx_de_enqueue_discard_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_DE_FW2WBM_RING_FULL_HIST_TAG:
+		htt_print_tx_de_fw2wbm_ring_full_hist_tlv(tag_buf, len, stats_req);
+		break;
+
+	case HTT_STATS_TX_DE_CMN_TAG:
+		htt_print_tx_de_cmn_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_RING_IF_TAG:
+		htt_print_ring_if_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_PDEV_MU_MIMO_STATS_TAG:
+		htt_print_tx_pdev_mu_mimo_sch_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_SFM_CMN_TAG:
+		htt_print_sfm_cmn_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_SRING_STATS_TAG:
+		htt_print_sring_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_RX_PDEV_FW_STATS_TAG:
+		htt_print_rx_pdev_fw_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_RX_PDEV_FW_RING_MPDU_ERR_TAG:
+		htt_print_rx_pdev_fw_ring_mpdu_err_tlv_v(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_RX_PDEV_FW_MPDU_DROP_TAG:
+		htt_print_rx_pdev_fw_mpdu_drop_tlv_v(tag_buf, len, stats_req);
+		break;
+
+	case HTT_STATS_RX_SOC_FW_STATS_TAG:
+		htt_print_rx_soc_fw_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_RX_SOC_FW_REFILL_RING_EMPTY_TAG:
+		htt_print_rx_soc_fw_refill_ring_empty_tlv_v(tag_buf, len, stats_req);
+		break;
+
+	case HTT_STATS_RX_SOC_FW_REFILL_RING_NUM_REFILL_TAG:
+		htt_print_rx_soc_fw_refill_ring_num_refill_tlv_v(
+				tag_buf, len, stats_req);
+		break;
+	case HTT_STATS_RX_REFILL_RXDMA_ERR_TAG:
+		htt_print_rx_soc_fw_refill_ring_num_rxdma_err_tlv_v(
+				tag_buf, len, stats_req);
+		break;
+
+	case HTT_STATS_RX_REFILL_REO_ERR_TAG:
+		htt_print_rx_soc_fw_refill_ring_num_reo_err_tlv_v(
+				tag_buf, len, stats_req);
+		break;
+
+	case HTT_STATS_RX_REO_RESOURCE_STATS_TAG:
+		htt_print_rx_reo_debug_stats_tlv_v(
+				tag_buf, stats_req);
+		break;
+	case HTT_STATS_RX_PDEV_FW_STATS_PHY_ERR_TAG:
+		htt_print_rx_pdev_fw_stats_phy_err_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_PDEV_RATE_STATS_TAG:
+		htt_print_tx_pdev_rate_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_RX_PDEV_RATE_STATS_TAG:
+		htt_print_rx_pdev_rate_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_PDEV_SCHEDULER_TXQ_STATS_TAG:
+		htt_print_tx_pdev_stats_sched_per_txq_tlv(tag_buf, stats_req);
+		break;
+	case HTT_STATS_TX_SCHED_CMN_TAG:
+		htt_print_stats_tx_sched_cmn_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_PDEV_MPDU_STATS_TAG:
+		htt_print_tx_pdev_mu_mimo_mpdu_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_SCHED_TXQ_CMD_POSTED_TAG:
+		htt_print_sched_txq_cmd_posted_tlv_v(tag_buf, len, stats_req);
+		break;
+
+	case HTT_STATS_RING_IF_CMN_TAG:
+		htt_print_ring_if_cmn_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_SFM_CLIENT_USER_TAG:
+		htt_print_sfm_client_user_tlv_v(tag_buf, len, stats_req);
+		break;
+
+	case HTT_STATS_SFM_CLIENT_TAG:
+		htt_print_sfm_client_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_TQM_ERROR_STATS_TAG:
+		htt_print_tx_tqm_error_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_SCHED_TXQ_CMD_REAPED_TAG:
+		htt_print_sched_txq_cmd_reaped_tlv_v(tag_buf, len, stats_req);
+		break;
+
+	case HTT_STATS_SRING_CMN_TAG:
+		htt_print_sring_cmn_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_SOUNDING_STATS_TAG:
+		htt_print_tx_sounding_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_SELFGEN_AC_ERR_STATS_TAG:
+		htt_print_tx_selfgen_ac_err_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_SELFGEN_CMN_STATS_TAG:
+		htt_print_tx_selfgen_cmn_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_SELFGEN_AC_STATS_TAG:
+		htt_print_tx_selfgen_ac_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_SELFGEN_AX_STATS_TAG:
+		htt_print_tx_selfgen_ax_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_SELFGEN_AX_ERR_STATS_TAG:
+		htt_print_tx_selfgen_ax_err_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_HWQ_MUMIMO_SCH_STATS_TAG:
+		htt_print_tx_hwq_mu_mimo_sch_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_HWQ_MUMIMO_MPDU_STATS_TAG:
+		htt_print_tx_hwq_mu_mimo_mpdu_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_HWQ_MUMIMO_CMN_STATS_TAG:
+		htt_print_tx_hwq_mu_mimo_cmn_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_HW_INTR_MISC_TAG:
+		htt_print_hw_stats_intr_misc_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_HW_WD_TIMEOUT_TAG:
+		htt_print_hw_stats_wd_timeout_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_HW_PDEV_ERRS_TAG:
+		htt_print_hw_stats_pdev_errs_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_COUNTER_NAME_TAG:
+		htt_print_counter_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_TID_DETAILS_TAG:
+		htt_print_tx_tid_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_TID_DETAILS_V1_TAG:
+		htt_print_tx_tid_stats_v1_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_RX_TID_DETAILS_TAG:
+		htt_print_rx_tid_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_PEER_STATS_CMN_TAG:
+		htt_print_peer_stats_cmn_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_PEER_DETAILS_TAG:
+		htt_print_peer_details_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_PEER_MSDU_FLOWQ_TAG:
+		htt_print_msdu_flow_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_PEER_TX_RATE_STATS_TAG:
+		htt_print_tx_peer_rate_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_PEER_RX_RATE_STATS_TAG:
+		htt_print_rx_peer_rate_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_TX_DE_COMPL_STATS_TAG:
+		htt_print_tx_de_compl_stats_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_PDEV_CCA_1SEC_HIST_TAG:
+	case HTT_STATS_PDEV_CCA_100MSEC_HIST_TAG:
+	case HTT_STATS_PDEV_CCA_STAT_CUMULATIVE_TAG:
+		htt_print_pdev_cca_stats_hist_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_PDEV_CCA_COUNTERS_TAG:
+		htt_print_pdev_stats_cca_counters_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_WHAL_TX_TAG:
+		htt_print_hw_stats_whal_tx_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_PDEV_TWT_SESSIONS_TAG:
+		htt_print_pdev_stats_twt_sessions_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_PDEV_TWT_SESSION_TAG:
+		htt_print_pdev_stats_twt_session_tlv(tag_buf, stats_req);
+		break;
+
+	case HTT_STATS_SCHED_TXQ_SCHED_ORDER_SU_TAG:
+		htt_print_sched_txq_sched_order_su_tlv_v(tag_buf, len, stats_req);
+		break;
+
+	case HTT_STATS_SCHED_TXQ_SCHED_INELIGIBILITY_TAG:
+		htt_print_sched_txq_sched_ineligibility_tlv_v(tag_buf, len, stats_req);
+		break;
+
+	case HTT_STATS_PDEV_OBSS_PD_TAG:
+		htt_print_pdev_obss_pd_stats_tlv_v(tag_buf, stats_req);
+		break;
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+void ath11k_dbg_htt_ext_stats_handler(struct ath11k_base *ab,
+				      struct sk_buff *skb)
+{
+	struct ath11k_htt_extd_stats_msg *msg;
+	struct debug_htt_stats_req *stats_req;
+	struct ath11k *ar;
+	u32 len;
+	u64 cookie;
+	int ret;
+	u8 pdev_id;
+
+	msg = (struct ath11k_htt_extd_stats_msg *)skb->data;
+	cookie = msg->cookie;
+
+	if (FIELD_GET(HTT_STATS_COOKIE_MSB, cookie) != HTT_STATS_MAGIC_VALUE) {
+		ath11k_warn(ab, "received invalid htt ext stats event\n");
+		return;
+	}
+
+	pdev_id = FIELD_GET(HTT_STATS_COOKIE_LSB, cookie);
+	rcu_read_lock();
+	ar = ath11k_mac_get_ar_by_pdev_id(ab, pdev_id);
+	rcu_read_unlock();
+	if (!ar) {
+		ath11k_warn(ab, "failed to get ar for pdev_id %d\n", pdev_id);
+		return;
+	}
+
+	stats_req = ar->debug.htt_stats.stats_req;
+	if (!stats_req)
+		return;
+
+	spin_lock_bh(&ar->debug.htt_stats.lock);
+	if (stats_req->done) {
+		spin_unlock_bh(&ar->debug.htt_stats.lock);
+		return;
+	}
+	stats_req->done = true;
+	spin_unlock_bh(&ar->debug.htt_stats.lock);
+
+	len = FIELD_GET(HTT_T2H_EXT_STATS_INFO1_LENGTH, msg->info1);
+	ret = ath11k_dp_htt_tlv_iter(ab, msg->data, len,
+				     ath11k_dbg_htt_ext_stats_parse,
+				     stats_req);
+	if (ret)
+		ath11k_warn(ab, "Failed to parse tlv %d\n", ret);
+
+	complete(&stats_req->cmpln);
+}
+
+static ssize_t ath11k_read_htt_stats_type(struct file *file,
+					  char __user *user_buf,
+					  size_t count, loff_t *ppos)
+{
+	struct ath11k *ar = file->private_data;
+	char buf[32];
+	size_t len;
+
+	len = scnprintf(buf, sizeof(buf), "%u\n", ar->debug.htt_stats.type);
+
+	return simple_read_from_buffer(user_buf, count, ppos, buf, len);
+}
+
+static ssize_t ath11k_write_htt_stats_type(struct file *file,
+					   const char __user *user_buf,
+					   size_t count, loff_t *ppos)
+{
+	struct ath11k *ar = file->private_data;
+	u8 type;
+	int ret;
+
+	ret = kstrtou8_from_user(user_buf, count, 0, &type);
+	if (ret)
+		return ret;
+
+	if (type >= ATH11K_DBG_HTT_NUM_EXT_STATS)
+		return -E2BIG;
+
+	if (type == ATH11K_DBG_HTT_EXT_STATS_RESET ||
+	    type == ATH11K_DBG_HTT_EXT_STATS_PEER_INFO)
+		return -EPERM;
+
+	ar->debug.htt_stats.type = type;
+
+	ret = count;
+
+	return ret;
+}
+
+static const struct file_operations fops_htt_stats_type = {
+	.read = ath11k_read_htt_stats_type,
+	.write = ath11k_write_htt_stats_type,
+	.open = simple_open,
+	.owner = THIS_MODULE,
+	.llseek = default_llseek,
+};
+
+static int ath11k_prep_htt_stats_cfg_params(struct ath11k *ar, u8 type,
+					    const u8 *mac_addr,
+					    struct htt_ext_stats_cfg_params *cfg_params)
+{
+	if (!cfg_params)
+		return -EINVAL;
+
+	switch (type) {
+	case ATH11K_DBG_HTT_EXT_STATS_PDEV_TX_HWQ:
+	case ATH11K_DBG_HTT_EXT_STATS_TX_MU_HWQ:
+		cfg_params->cfg0 = HTT_STAT_DEFAULT_CFG0_ALL_HWQS;
+		break;
+	case ATH11K_DBG_HTT_EXT_STATS_PDEV_TX_SCHED:
+		cfg_params->cfg0 = HTT_STAT_DEFAULT_CFG0_ALL_TXQS;
+		break;
+	case ATH11K_DBG_HTT_EXT_STATS_TQM_CMDQ:
+		cfg_params->cfg0 = HTT_STAT_DEFAULT_CFG0_ALL_CMDQS;
+		break;
+	case ATH11K_DBG_HTT_EXT_STATS_PEER_INFO:
+		cfg_params->cfg0 = HTT_STAT_PEER_INFO_MAC_ADDR;
+		cfg_params->cfg0 |= FIELD_PREP(GENMASK(15, 1),
+					HTT_PEER_STATS_REQ_MODE_FLUSH_TQM);
+		cfg_params->cfg1 = HTT_STAT_DEFAULT_PEER_REQ_TYPE;
+		cfg_params->cfg2 |= FIELD_PREP(GENMASK(7, 0), mac_addr[0]);
+		cfg_params->cfg2 |= FIELD_PREP(GENMASK(15, 8), mac_addr[1]);
+		cfg_params->cfg2 |= FIELD_PREP(GENMASK(23, 16), mac_addr[2]);
+		cfg_params->cfg2 |= FIELD_PREP(GENMASK(31, 24), mac_addr[3]);
+		cfg_params->cfg3 |= FIELD_PREP(GENMASK(7, 0), mac_addr[4]);
+		cfg_params->cfg3 |= FIELD_PREP(GENMASK(15, 8), mac_addr[5]);
+		break;
+	case ATH11K_DBG_HTT_EXT_STATS_RING_IF_INFO:
+	case ATH11K_DBG_HTT_EXT_STATS_SRNG_INFO:
+		cfg_params->cfg0 = HTT_STAT_DEFAULT_CFG0_ALL_RINGS;
+		break;
+	case ATH11K_DBG_HTT_EXT_STATS_ACTIVE_PEERS_LIST:
+		cfg_params->cfg0 = HTT_STAT_DEFAULT_CFG0_ACTIVE_PEERS;
+		break;
+	case ATH11K_DBG_HTT_EXT_STATS_PDEV_CCA_STATS:
+		cfg_params->cfg0 = HTT_STAT_DEFAULT_CFG0_CCA_CUMULATIVE;
+		break;
+	case ATH11K_DBG_HTT_EXT_STATS_TX_SOUNDING_INFO:
+		cfg_params->cfg0 = HTT_STAT_DEFAULT_CFG0_ACTIVE_VDEVS;
+		break;
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+int ath11k_dbg_htt_stats_req(struct ath11k *ar)
+{
+	struct debug_htt_stats_req *stats_req = ar->debug.htt_stats.stats_req;
+	u8 type = stats_req->type;
+	u64 cookie = 0;
+	int ret, pdev_id = ar->pdev->pdev_id;
+	struct htt_ext_stats_cfg_params cfg_params = { 0 };
+
+	init_completion(&stats_req->cmpln);
+
+	stats_req->done = false;
+	stats_req->pdev_id = pdev_id;
+
+	cookie = FIELD_PREP(HTT_STATS_COOKIE_MSB, HTT_STATS_MAGIC_VALUE) |
+		 FIELD_PREP(HTT_STATS_COOKIE_LSB, pdev_id);
+
+	ret = ath11k_prep_htt_stats_cfg_params(ar, type, stats_req->peer_addr,
+					       &cfg_params);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to set htt stats cfg params: %d\n", ret);
+		return ret;
+	}
+
+	ret = ath11k_dp_tx_htt_h2t_ext_stats_req(ar, type, &cfg_params, cookie);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to send htt stats request: %d\n", ret);
+		mutex_unlock(&ar->conf_mutex);
+		return ret;
+	}
+
+	while (!wait_for_completion_timeout(&stats_req->cmpln, 3 * HZ)) {
+		spin_lock_bh(&ar->debug.htt_stats.lock);
+		if (!stats_req->done) {
+			stats_req->done = true;
+			spin_unlock_bh(&ar->debug.htt_stats.lock);
+			ath11k_warn(ar->ab, "stats request timed out\n");
+			return -ETIMEDOUT;
+		}
+		spin_unlock_bh(&ar->debug.htt_stats.lock);
+	}
+
+	return 0;
+}
+
+static int ath11k_open_htt_stats(struct inode *inode, struct file *file)
+{
+	struct ath11k *ar = inode->i_private;
+	struct debug_htt_stats_req *stats_req;
+	u8 type = ar->debug.htt_stats.type;
+	int ret;
+
+	if (type == ATH11K_DBG_HTT_EXT_STATS_RESET)
+		return -EPERM;
+
+	stats_req = vzalloc(sizeof(*stats_req) + ATH11K_HTT_STATS_BUF_SIZE);
+	if (!stats_req)
+		return -ENOMEM;
+
+	mutex_lock(&ar->conf_mutex);
+	ar->debug.htt_stats.stats_req = stats_req;
+	stats_req->type = type;
+	ret = ath11k_dbg_htt_stats_req(ar);
+	mutex_unlock(&ar->conf_mutex);
+	if (ret < 0)
+		goto out;
+
+	file->private_data = stats_req;
+	return 0;
+out:
+	vfree(stats_req);
+	return ret;
+}
+
+static int ath11k_release_htt_stats(struct inode *inode, struct file *file)
+{
+	vfree(file->private_data);
+	return 0;
+}
+
+static ssize_t ath11k_read_htt_stats(struct file *file,
+				     char __user *user_buf,
+				     size_t count, loff_t *ppos)
+{
+	struct debug_htt_stats_req *stats_req = file->private_data;
+	char *buf;
+	u32 length = 0;
+
+	buf = stats_req->buf;
+	length = min_t(u32, stats_req->buf_len, ATH11K_HTT_STATS_BUF_SIZE);
+	return simple_read_from_buffer(user_buf, count, ppos, buf, length);
+}
+
+static const struct file_operations fops_dump_htt_stats = {
+	.open = ath11k_open_htt_stats,
+	.release = ath11k_release_htt_stats,
+	.read = ath11k_read_htt_stats,
+	.owner = THIS_MODULE,
+	.llseek = default_llseek,
+};
+
+static ssize_t ath11k_read_htt_stats_reset(struct file *file,
+					   char __user *user_buf,
+					   size_t count, loff_t *ppos)
+{
+	struct ath11k *ar = file->private_data;
+	char buf[32];
+	size_t len;
+
+	len = scnprintf(buf, sizeof(buf), "%u\n", ar->debug.htt_stats.reset);
+
+	return simple_read_from_buffer(user_buf, count, ppos, buf, len);
+}
+
+static ssize_t ath11k_write_htt_stats_reset(struct file *file,
+					    const char __user *user_buf,
+					    size_t count, loff_t *ppos)
+{
+	struct ath11k *ar = file->private_data;
+	u8 type;
+	struct htt_ext_stats_cfg_params cfg_params = { 0 };
+	int ret;
+
+	ret = kstrtou8_from_user(user_buf, count, 0, &type);
+	if (ret)
+		return ret;
+
+	if (type >= ATH11K_DBG_HTT_NUM_EXT_STATS ||
+	    type == ATH11K_DBG_HTT_EXT_STATS_RESET)
+		return -E2BIG;
+
+	mutex_lock(&ar->conf_mutex);
+	cfg_params.cfg0 = HTT_STAT_DEFAULT_RESET_START_OFFSET;
+	cfg_params.cfg1 = 1 << (cfg_params.cfg0 + type);
+	ret = ath11k_dp_tx_htt_h2t_ext_stats_req(ar,
+						 ATH11K_DBG_HTT_EXT_STATS_RESET,
+						 &cfg_params,
+						 0ULL);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to send htt stats request: %d\n", ret);
+		mutex_unlock(&ar->conf_mutex);
+		return ret;
+	}
+
+	ar->debug.htt_stats.reset = type;
+	mutex_unlock(&ar->conf_mutex);
+
+	ret = count;
+
+	return ret;
+}
+
+static const struct file_operations fops_htt_stats_reset = {
+	.read = ath11k_read_htt_stats_reset,
+	.write = ath11k_write_htt_stats_reset,
+	.open = simple_open,
+	.owner = THIS_MODULE,
+	.llseek = default_llseek,
+};
+
+void ath11k_debug_htt_stats_init(struct ath11k *ar)
+{
+	spin_lock_init(&ar->debug.htt_stats.lock);
+	debugfs_create_file("htt_stats_type", 0600, ar->debug.debugfs_pdev,
+			    ar, &fops_htt_stats_type);
+	debugfs_create_file("htt_stats", 0400, ar->debug.debugfs_pdev,
+			    ar, &fops_dump_htt_stats);
+	debugfs_create_file("htt_stats_reset", 0600, ar->debug.debugfs_pdev,
+			    ar, &fops_htt_stats_reset);
+}
diff --git a/drivers/net/wireless/ath/ath11k/debug_htt_stats.h b/drivers/net/wireless/ath/ath11k/debug_htt_stats.h
new file mode 100644
index 000000000000..618f1946bf49
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/debug_htt_stats.h
@@ -0,0 +1,1620 @@
+/* SPDX-License-Identifier: BSD-3-Clause-Clear */
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#ifndef DEBUG_HTT_STATS_H
+#define DEBUG_HTT_STATS_H
+
+#define HTT_STATS_COOKIE_LSB    GENMASK_ULL(31, 0)
+#define HTT_STATS_COOKIE_MSB    GENMASK_ULL(63, 32)
+#define HTT_STATS_MAGIC_VALUE   0xF0F0F0F0
+
+enum htt_tlv_tag_t {
+	HTT_STATS_TX_PDEV_CMN_TAG                           = 0,
+	HTT_STATS_TX_PDEV_UNDERRUN_TAG                      = 1,
+	HTT_STATS_TX_PDEV_SIFS_TAG                          = 2,
+	HTT_STATS_TX_PDEV_FLUSH_TAG                         = 3,
+	HTT_STATS_TX_PDEV_PHY_ERR_TAG                       = 4,
+	HTT_STATS_STRING_TAG                                = 5,
+	HTT_STATS_TX_HWQ_CMN_TAG                            = 6,
+	HTT_STATS_TX_HWQ_DIFS_LATENCY_TAG                   = 7,
+	HTT_STATS_TX_HWQ_CMD_RESULT_TAG                     = 8,
+	HTT_STATS_TX_HWQ_CMD_STALL_TAG                      = 9,
+	HTT_STATS_TX_HWQ_FES_STATUS_TAG                     = 10,
+	HTT_STATS_TX_TQM_GEN_MPDU_TAG                       = 11,
+	HTT_STATS_TX_TQM_LIST_MPDU_TAG                      = 12,
+	HTT_STATS_TX_TQM_LIST_MPDU_CNT_TAG                  = 13,
+	HTT_STATS_TX_TQM_CMN_TAG                            = 14,
+	HTT_STATS_TX_TQM_PDEV_TAG                           = 15,
+	HTT_STATS_TX_TQM_CMDQ_STATUS_TAG                    = 16,
+	HTT_STATS_TX_DE_EAPOL_PACKETS_TAG                   = 17,
+	HTT_STATS_TX_DE_CLASSIFY_FAILED_TAG                 = 18,
+	HTT_STATS_TX_DE_CLASSIFY_STATS_TAG                  = 19,
+	HTT_STATS_TX_DE_CLASSIFY_STATUS_TAG                 = 20,
+	HTT_STATS_TX_DE_ENQUEUE_PACKETS_TAG                 = 21,
+	HTT_STATS_TX_DE_ENQUEUE_DISCARD_TAG                 = 22,
+	HTT_STATS_TX_DE_CMN_TAG                             = 23,
+	HTT_STATS_RING_IF_TAG                               = 24,
+	HTT_STATS_TX_PDEV_MU_MIMO_STATS_TAG                 = 25,
+	HTT_STATS_SFM_CMN_TAG                               = 26,
+	HTT_STATS_SRING_STATS_TAG                           = 27,
+	HTT_STATS_RX_PDEV_FW_STATS_TAG                      = 28,
+	HTT_STATS_RX_PDEV_FW_RING_MPDU_ERR_TAG              = 29,
+	HTT_STATS_RX_PDEV_FW_MPDU_DROP_TAG                  = 30,
+	HTT_STATS_RX_SOC_FW_STATS_TAG                       = 31,
+	HTT_STATS_RX_SOC_FW_REFILL_RING_EMPTY_TAG           = 32,
+	HTT_STATS_RX_SOC_FW_REFILL_RING_NUM_REFILL_TAG      = 33,
+	HTT_STATS_TX_PDEV_RATE_STATS_TAG                    = 34,
+	HTT_STATS_RX_PDEV_RATE_STATS_TAG                    = 35,
+	HTT_STATS_TX_PDEV_SCHEDULER_TXQ_STATS_TAG           = 36,
+	HTT_STATS_TX_SCHED_CMN_TAG                          = 37,
+	HTT_STATS_TX_PDEV_MUMIMO_MPDU_STATS_TAG             = 38,
+	HTT_STATS_SCHED_TXQ_CMD_POSTED_TAG                  = 39,
+	HTT_STATS_RING_IF_CMN_TAG                           = 40,
+	HTT_STATS_SFM_CLIENT_USER_TAG                       = 41,
+	HTT_STATS_SFM_CLIENT_TAG                            = 42,
+	HTT_STATS_TX_TQM_ERROR_STATS_TAG                    = 43,
+	HTT_STATS_SCHED_TXQ_CMD_REAPED_TAG                  = 44,
+	HTT_STATS_SRING_CMN_TAG                             = 45,
+	HTT_STATS_TX_SELFGEN_AC_ERR_STATS_TAG               = 46,
+	HTT_STATS_TX_SELFGEN_CMN_STATS_TAG                  = 47,
+	HTT_STATS_TX_SELFGEN_AC_STATS_TAG                   = 48,
+	HTT_STATS_TX_SELFGEN_AX_STATS_TAG                   = 49,
+	HTT_STATS_TX_SELFGEN_AX_ERR_STATS_TAG               = 50,
+	HTT_STATS_TX_HWQ_MUMIMO_SCH_STATS_TAG               = 51,
+	HTT_STATS_TX_HWQ_MUMIMO_MPDU_STATS_TAG              = 52,
+	HTT_STATS_TX_HWQ_MUMIMO_CMN_STATS_TAG               = 53,
+	HTT_STATS_HW_INTR_MISC_TAG                          = 54,
+	HTT_STATS_HW_WD_TIMEOUT_TAG                         = 55,
+	HTT_STATS_HW_PDEV_ERRS_TAG                          = 56,
+	HTT_STATS_COUNTER_NAME_TAG                          = 57,
+	HTT_STATS_TX_TID_DETAILS_TAG                        = 58,
+	HTT_STATS_RX_TID_DETAILS_TAG                        = 59,
+	HTT_STATS_PEER_STATS_CMN_TAG                        = 60,
+	HTT_STATS_PEER_DETAILS_TAG                          = 61,
+	HTT_STATS_PEER_TX_RATE_STATS_TAG                    = 62,
+	HTT_STATS_PEER_RX_RATE_STATS_TAG                    = 63,
+	HTT_STATS_PEER_MSDU_FLOWQ_TAG                       = 64,
+	HTT_STATS_TX_DE_COMPL_STATS_TAG                     = 65,
+	HTT_STATS_WHAL_TX_TAG                               = 66,
+	HTT_STATS_TX_PDEV_SIFS_HIST_TAG                     = 67,
+	HTT_STATS_RX_PDEV_FW_STATS_PHY_ERR_TAG              = 68,
+	HTT_STATS_TX_TID_DETAILS_V1_TAG                     = 69,
+	HTT_STATS_PDEV_CCA_1SEC_HIST_TAG                    = 70,
+	HTT_STATS_PDEV_CCA_100MSEC_HIST_TAG                 = 71,
+	HTT_STATS_PDEV_CCA_STAT_CUMULATIVE_TAG              = 72,
+	HTT_STATS_PDEV_CCA_COUNTERS_TAG                     = 73,
+	HTT_STATS_TX_PDEV_MPDU_STATS_TAG                    = 74,
+	HTT_STATS_PDEV_TWT_SESSIONS_TAG                     = 75,
+	HTT_STATS_PDEV_TWT_SESSION_TAG                      = 76,
+	HTT_STATS_RX_REFILL_RXDMA_ERR_TAG                   = 77,
+	HTT_STATS_RX_REFILL_REO_ERR_TAG                     = 78,
+	HTT_STATS_RX_REO_RESOURCE_STATS_TAG                 = 79,
+	HTT_STATS_TX_SOUNDING_STATS_TAG                     = 80,
+	HTT_STATS_TX_PDEV_TX_PPDU_STATS_TAG                 = 81,
+	HTT_STATS_TX_PDEV_TRIED_MPDU_CNT_HIST_TAG           = 82,
+	HTT_STATS_TX_HWQ_TRIED_MPDU_CNT_HIST_TAG            = 83,
+	HTT_STATS_TX_HWQ_TXOP_USED_CNT_HIST_TAG             = 84,
+	HTT_STATS_TX_DE_FW2WBM_RING_FULL_HIST_TAG           = 85,
+	HTT_STATS_SCHED_TXQ_SCHED_ORDER_SU_TAG              = 86,
+	HTT_STATS_SCHED_TXQ_SCHED_INELIGIBILITY_TAG         = 87,
+	HTT_STATS_PDEV_OBSS_PD_TAG                          = 88,
+
+	HTT_STATS_MAX_TAG,
+};
+
+#define HTT_STATS_MAX_STRING_SZ32            4
+#define HTT_STATS_MACID_INVALID              0xff
+#define HTT_TX_HWQ_MAX_DIFS_LATENCY_BINS     10
+#define HTT_TX_HWQ_MAX_CMD_RESULT_STATS      13
+#define HTT_TX_HWQ_MAX_CMD_STALL_STATS       5
+#define HTT_TX_HWQ_MAX_FES_RESULT_STATS      10
+
+enum htt_tx_pdev_underrun_enum {
+	HTT_STATS_TX_PDEV_NO_DATA_UNDERRUN           = 0,
+	HTT_STATS_TX_PDEV_DATA_UNDERRUN_BETWEEN_MPDU = 1,
+	HTT_STATS_TX_PDEV_DATA_UNDERRUN_WITHIN_MPDU  = 2,
+	HTT_TX_PDEV_MAX_URRN_STATS                   = 3,
+};
+
+#define HTT_TX_PDEV_MAX_FLUSH_REASON_STATS     71
+#define HTT_TX_PDEV_MAX_SIFS_BURST_STATS       9
+#define HTT_TX_PDEV_MAX_SIFS_BURST_HIST_STATS  10
+#define HTT_TX_PDEV_MAX_PHY_ERR_STATS          18
+#define HTT_TX_PDEV_SCHED_TX_MODE_MAX          4
+#define HTT_TX_PDEV_NUM_SCHED_ORDER_LOG        20
+
+#define HTT_RX_STATS_REFILL_MAX_RING         4
+#define HTT_RX_STATS_RXDMA_MAX_ERR           16
+#define HTT_RX_STATS_FW_DROP_REASON_MAX      16
+
+/* Bytes stored in little endian order */
+/* Length should be multiple of DWORD */
+struct htt_stats_string_tlv {
+	u32 data[0]; /* Can be variable length */
+} __packed;
+
+/* == TX PDEV STATS == */
+struct htt_tx_pdev_stats_cmn_tlv {
+	u32 mac_id__word;
+	u32 hw_queued;
+	u32 hw_reaped;
+	u32 underrun;
+	u32 hw_paused;
+	u32 hw_flush;
+	u32 hw_filt;
+	u32 tx_abort;
+	u32 mpdu_requed;
+	u32 tx_xretry;
+	u32 data_rc;
+	u32 mpdu_dropped_xretry;
+	u32 illgl_rate_phy_err;
+	u32 cont_xretry;
+	u32 tx_timeout;
+	u32 pdev_resets;
+	u32 phy_underrun;
+	u32 txop_ovf;
+	u32 seq_posted;
+	u32 seq_failed_queueing;
+	u32 seq_completed;
+	u32 seq_restarted;
+	u32 mu_seq_posted;
+	u32 seq_switch_hw_paused;
+	u32 next_seq_posted_dsr;
+	u32 seq_posted_isr;
+	u32 seq_ctrl_cached;
+	u32 mpdu_count_tqm;
+	u32 msdu_count_tqm;
+	u32 mpdu_removed_tqm;
+	u32 msdu_removed_tqm;
+	u32 mpdus_sw_flush;
+	u32 mpdus_hw_filter;
+	u32 mpdus_truncated;
+	u32 mpdus_ack_failed;
+	u32 mpdus_expired;
+	u32 mpdus_seq_hw_retry;
+	u32 ack_tlv_proc;
+	u32 coex_abort_mpdu_cnt_valid;
+	u32 coex_abort_mpdu_cnt;
+	u32 num_total_ppdus_tried_ota;
+	u32 num_data_ppdus_tried_ota;
+	u32 local_ctrl_mgmt_enqued;
+	u32 local_ctrl_mgmt_freed;
+	u32 local_data_enqued;
+	u32 local_data_freed;
+	u32 mpdu_tried;
+	u32 isr_wait_seq_posted;
+
+	u32 tx_active_dur_us_low;
+	u32 tx_active_dur_us_high;
+};
+
+/* NOTE: Variable length TLV, use length spec to infer array size */
+struct htt_tx_pdev_stats_urrn_tlv_v {
+	u32 urrn_stats[0]; /* HTT_TX_PDEV_MAX_URRN_STATS */
+};
+
+/* NOTE: Variable length TLV, use length spec to infer array size */
+struct htt_tx_pdev_stats_flush_tlv_v {
+	u32 flush_errs[0]; /* HTT_TX_PDEV_MAX_FLUSH_REASON_STATS */
+};
+
+/* NOTE: Variable length TLV, use length spec to infer array size */
+struct htt_tx_pdev_stats_sifs_tlv_v {
+	u32 sifs_status[0]; /* HTT_TX_PDEV_MAX_SIFS_BURST_STATS */
+};
+
+/* NOTE: Variable length TLV, use length spec to infer array size */
+struct htt_tx_pdev_stats_phy_err_tlv_v {
+	u32  phy_errs[0]; /* HTT_TX_PDEV_MAX_PHY_ERR_STATS */
+};
+
+/* NOTE: Variable length TLV, use length spec to infer array size */
+struct htt_tx_pdev_stats_sifs_hist_tlv_v {
+	u32 sifs_hist_status[0]; /* HTT_TX_PDEV_SIFS_BURST_HIST_STATS */
+};
+
+struct htt_tx_pdev_stats_tx_ppdu_stats_tlv_v {
+	u32 num_data_ppdus_legacy_su;
+	u32 num_data_ppdus_ac_su;
+	u32 num_data_ppdus_ax_su;
+	u32 num_data_ppdus_ac_su_txbf;
+	u32 num_data_ppdus_ax_su_txbf;
+};
+
+/* NOTE: Variable length TLV, use length spec to infer array size .
+ *
+ *  Tried_mpdu_cnt_hist is the histogram of MPDUs tries per HWQ.
+ *  The tries here is the count of the  MPDUS within a PPDU that the
+ *  HW had attempted to transmit on  air, for the HWSCH Schedule
+ *  command submitted by FW.It is not the retry attempts.
+ *  The histogram bins are  0-29, 30-59, 60-89 and so on. The are
+ *   10 bins in this histogram. They are defined in FW using the
+ *  following macros
+ *  #define WAL_MAX_TRIED_MPDU_CNT_HISTOGRAM 9
+ *  #define WAL_TRIED_MPDU_CNT_HISTOGRAM_INTERVAL 30
+ */
+struct htt_tx_pdev_stats_tried_mpdu_cnt_hist_tlv_v {
+	u32 hist_bin_size;
+	u32 tried_mpdu_cnt_hist[0]; /* HTT_TX_PDEV_TRIED_MPDU_CNT_HIST */
+};
+
+/* == SOC ERROR STATS == */
+
+/* =============== PDEV ERROR STATS ============== */
+#define HTT_STATS_MAX_HW_INTR_NAME_LEN 8
+struct htt_hw_stats_intr_misc_tlv {
+	/* Stored as little endian */
+	u8 hw_intr_name[HTT_STATS_MAX_HW_INTR_NAME_LEN];
+	u32 mask;
+	u32 count;
+};
+
+#define HTT_STATS_MAX_HW_MODULE_NAME_LEN 8
+struct htt_hw_stats_wd_timeout_tlv {
+	/* Stored as little endian */
+	u8 hw_module_name[HTT_STATS_MAX_HW_MODULE_NAME_LEN];
+	u32 count;
+};
+
+struct htt_hw_stats_pdev_errs_tlv {
+	u32    mac_id__word; /* BIT [ 7 :  0] : mac_id */
+	u32    tx_abort;
+	u32    tx_abort_fail_count;
+	u32    rx_abort;
+	u32    rx_abort_fail_count;
+	u32    warm_reset;
+	u32    cold_reset;
+	u32    tx_flush;
+	u32    tx_glb_reset;
+	u32    tx_txq_reset;
+	u32    rx_timeout_reset;
+};
+
+struct htt_hw_stats_whal_tx_tlv {
+	u32 mac_id__word;
+	u32 last_unpause_ppdu_id;
+	u32 hwsch_unpause_wait_tqm_write;
+	u32 hwsch_dummy_tlv_skipped;
+	u32 hwsch_misaligned_offset_received;
+	u32 hwsch_reset_count;
+	u32 hwsch_dev_reset_war;
+	u32 hwsch_delayed_pause;
+	u32 hwsch_long_delayed_pause;
+	u32 sch_rx_ppdu_no_response;
+	u32 sch_selfgen_response;
+	u32 sch_rx_sifs_resp_trigger;
+};
+
+/* ============ PEER STATS ============ */
+struct htt_msdu_flow_stats_tlv {
+	u32 last_update_timestamp;
+	u32 last_add_timestamp;
+	u32 last_remove_timestamp;
+	u32 total_processed_msdu_count;
+	u32 cur_msdu_count_in_flowq;
+	u32 sw_peer_id;
+	u32 tx_flow_no__tid_num__drop_rule;
+	u32 last_cycle_enqueue_count;
+	u32 last_cycle_dequeue_count;
+	u32 last_cycle_drop_count;
+	u32 current_drop_th;
+};
+
+#define MAX_HTT_TID_NAME 8
+
+/* Tidq stats */
+struct htt_tx_tid_stats_tlv {
+	/* Stored as little endian */
+	u8     tid_name[MAX_HTT_TID_NAME];
+	u32 sw_peer_id__tid_num;
+	u32 num_sched_pending__num_ppdu_in_hwq;
+	u32 tid_flags;
+	u32 hw_queued;
+	u32 hw_reaped;
+	u32 mpdus_hw_filter;
+
+	u32 qdepth_bytes;
+	u32 qdepth_num_msdu;
+	u32 qdepth_num_mpdu;
+	u32 last_scheduled_tsmp;
+	u32 pause_module_id;
+	u32 block_module_id;
+	u32 tid_tx_airtime;
+};
+
+/* Tidq stats */
+struct htt_tx_tid_stats_v1_tlv {
+	/* Stored as little endian */
+	u8 tid_name[MAX_HTT_TID_NAME];
+	u32 sw_peer_id__tid_num;
+	u32 num_sched_pending__num_ppdu_in_hwq;
+	u32 tid_flags;
+	u32 max_qdepth_bytes;
+	u32 max_qdepth_n_msdus;
+	u32 rsvd;
+
+	u32 qdepth_bytes;
+	u32 qdepth_num_msdu;
+	u32 qdepth_num_mpdu;
+	u32 last_scheduled_tsmp;
+	u32 pause_module_id;
+	u32 block_module_id;
+	u32 tid_tx_airtime;
+	u32 allow_n_flags;
+	u32 sendn_frms_allowed;
+};
+
+struct htt_rx_tid_stats_tlv {
+	u32 sw_peer_id__tid_num;
+	u8 tid_name[MAX_HTT_TID_NAME];
+	u32 dup_in_reorder;
+	u32 dup_past_outside_window;
+	u32 dup_past_within_window;
+	u32 rxdesc_err_decrypt;
+	u32 tid_rx_airtime;
+};
+
+#define HTT_MAX_COUNTER_NAME 8
+struct htt_counter_tlv {
+	u8 counter_name[HTT_MAX_COUNTER_NAME];
+	u32 count;
+};
+
+struct htt_peer_stats_cmn_tlv {
+	u32 ppdu_cnt;
+	u32 mpdu_cnt;
+	u32 msdu_cnt;
+	u32 pause_bitmap;
+	u32 block_bitmap;
+	u32 current_timestamp;
+	u32 peer_tx_airtime;
+	u32 peer_rx_airtime;
+	s32 rssi;
+	u32 peer_enqueued_count_low;
+	u32 peer_enqueued_count_high;
+	u32 peer_dequeued_count_low;
+	u32 peer_dequeued_count_high;
+	u32 peer_dropped_count_low;
+	u32 peer_dropped_count_high;
+	u32 ppdu_transmitted_bytes_low;
+	u32 ppdu_transmitted_bytes_high;
+	u32 peer_ttl_removed_count;
+	u32 inactive_time;
+};
+
+struct htt_peer_details_tlv {
+	u32 peer_type;
+	u32 sw_peer_id;
+	u32 vdev_pdev_ast_idx;
+	struct htt_mac_addr mac_addr;
+	u32 peer_flags;
+	u32 qpeer_flags;
+};
+
+enum htt_stats_param_type {
+	HTT_STATS_PREAM_OFDM,
+	HTT_STATS_PREAM_CCK,
+	HTT_STATS_PREAM_HT,
+	HTT_STATS_PREAM_VHT,
+	HTT_STATS_PREAM_HE,
+	HTT_STATS_PREAM_RSVD,
+	HTT_STATS_PREAM_RSVD1,
+
+	HTT_STATS_PREAM_COUNT,
+};
+
+#define HTT_TX_PEER_STATS_NUM_MCS_COUNTERS        12
+#define HTT_TX_PEER_STATS_NUM_GI_COUNTERS          4
+#define HTT_TX_PEER_STATS_NUM_DCM_COUNTERS         5
+#define HTT_TX_PEER_STATS_NUM_BW_COUNTERS          4
+#define HTT_TX_PEER_STATS_NUM_SPATIAL_STREAMS      8
+#define HTT_TX_PEER_STATS_NUM_PREAMBLE_TYPES       HTT_STATS_PREAM_COUNT
+
+struct htt_tx_peer_rate_stats_tlv {
+	u32 tx_ldpc;
+	u32 rts_cnt;
+	u32 ack_rssi;
+
+	u32 tx_mcs[HTT_TX_PEER_STATS_NUM_MCS_COUNTERS];
+	u32 tx_su_mcs[HTT_TX_PEER_STATS_NUM_MCS_COUNTERS];
+	u32 tx_mu_mcs[HTT_TX_PEER_STATS_NUM_MCS_COUNTERS];
+	/* element 0,1, ...7 -> NSS 1,2, ...8 */
+	u32 tx_nss[HTT_TX_PEER_STATS_NUM_SPATIAL_STREAMS];
+	/* element 0: 20 MHz, 1: 40 MHz, 2: 80 MHz, 3: 160 and 80+80 MHz */
+	u32 tx_bw[HTT_TX_PEER_STATS_NUM_BW_COUNTERS];
+	u32 tx_stbc[HTT_TX_PEER_STATS_NUM_MCS_COUNTERS];
+	u32 tx_pream[HTT_TX_PEER_STATS_NUM_PREAMBLE_TYPES];
+
+	/* Counters to track number of tx packets in each GI
+	 * (400us, 800us, 1600us & 3200us) in each mcs (0-11)
+	 */
+	u32 tx_gi[HTT_TX_PEER_STATS_NUM_GI_COUNTERS][HTT_TX_PEER_STATS_NUM_MCS_COUNTERS];
+
+	/* Counters to track packets in dcm mcs (MCS 0, 1, 3, 4) */
+	u32 tx_dcm[HTT_TX_PEER_STATS_NUM_DCM_COUNTERS];
+
+};
+
+#define HTT_RX_PEER_STATS_NUM_MCS_COUNTERS        12
+#define HTT_RX_PEER_STATS_NUM_GI_COUNTERS          4
+#define HTT_RX_PEER_STATS_NUM_DCM_COUNTERS         5
+#define HTT_RX_PEER_STATS_NUM_BW_COUNTERS          4
+#define HTT_RX_PEER_STATS_NUM_SPATIAL_STREAMS      8
+#define HTT_RX_PEER_STATS_NUM_PREAMBLE_TYPES       HTT_STATS_PREAM_COUNT
+
+struct htt_rx_peer_rate_stats_tlv {
+	u32 nsts;
+
+	/* Number of rx ldpc packets */
+	u32 rx_ldpc;
+	/* Number of rx rts packets */
+	u32 rts_cnt;
+
+	u32 rssi_mgmt; /* units = dB above noise floor */
+	u32 rssi_data; /* units = dB above noise floor */
+	u32 rssi_comb; /* units = dB above noise floor */
+	u32 rx_mcs[HTT_RX_PEER_STATS_NUM_MCS_COUNTERS];
+	/* element 0,1, ...7 -> NSS 1,2, ...8 */
+	u32 rx_nss[HTT_RX_PEER_STATS_NUM_SPATIAL_STREAMS];
+	u32 rx_dcm[HTT_RX_PEER_STATS_NUM_DCM_COUNTERS];
+	u32 rx_stbc[HTT_RX_PEER_STATS_NUM_MCS_COUNTERS];
+	/* element 0: 20 MHz, 1: 40 MHz, 2: 80 MHz, 3: 160 and 80+80 MHz */
+	u32 rx_bw[HTT_RX_PEER_STATS_NUM_BW_COUNTERS];
+	u32 rx_pream[HTT_RX_PEER_STATS_NUM_PREAMBLE_TYPES];
+	/* units = dB above noise floor */
+	u8 rssi_chain[HTT_RX_PEER_STATS_NUM_SPATIAL_STREAMS]
+		     [HTT_RX_PEER_STATS_NUM_BW_COUNTERS];
+
+	/* Counters to track number of rx packets in each GI in each mcs (0-11) */
+	u32 rx_gi[HTT_RX_PEER_STATS_NUM_GI_COUNTERS]
+		 [HTT_RX_PEER_STATS_NUM_MCS_COUNTERS];
+};
+
+enum htt_peer_stats_req_mode {
+	HTT_PEER_STATS_REQ_MODE_NO_QUERY,
+	HTT_PEER_STATS_REQ_MODE_QUERY_TQM,
+	HTT_PEER_STATS_REQ_MODE_FLUSH_TQM,
+};
+
+enum htt_peer_stats_tlv_enum {
+	HTT_PEER_STATS_CMN_TLV       = 0,
+	HTT_PEER_DETAILS_TLV         = 1,
+	HTT_TX_PEER_RATE_STATS_TLV   = 2,
+	HTT_RX_PEER_RATE_STATS_TLV   = 3,
+	HTT_TX_TID_STATS_TLV         = 4,
+	HTT_RX_TID_STATS_TLV         = 5,
+	HTT_MSDU_FLOW_STATS_TLV      = 6,
+
+	HTT_PEER_STATS_MAX_TLV       = 31,
+};
+
+/* =========== MUMIMO HWQ stats =========== */
+/* MU MIMO stats per hwQ */
+struct htt_tx_hwq_mu_mimo_sch_stats_tlv {
+	u32 mu_mimo_sch_posted;
+	u32 mu_mimo_sch_failed;
+	u32 mu_mimo_ppdu_posted;
+};
+
+struct htt_tx_hwq_mu_mimo_mpdu_stats_tlv {
+	u32 mu_mimo_mpdus_queued_usr;
+	u32 mu_mimo_mpdus_tried_usr;
+	u32 mu_mimo_mpdus_failed_usr;
+	u32 mu_mimo_mpdus_requeued_usr;
+	u32 mu_mimo_err_no_ba_usr;
+	u32 mu_mimo_mpdu_underrun_usr;
+	u32 mu_mimo_ampdu_underrun_usr;
+};
+
+struct htt_tx_hwq_mu_mimo_cmn_stats_tlv {
+	u32 mac_id__hwq_id__word;
+};
+
+/* == TX HWQ STATS == */
+struct htt_tx_hwq_stats_cmn_tlv {
+	u32 mac_id__hwq_id__word;
+
+	/* PPDU level stats */
+	u32 xretry;
+	u32 underrun_cnt;
+	u32 flush_cnt;
+	u32 filt_cnt;
+	u32 null_mpdu_bmap;
+	u32 user_ack_failure;
+	u32 ack_tlv_proc;
+	u32 sched_id_proc;
+	u32 null_mpdu_tx_count;
+	u32 mpdu_bmap_not_recvd;
+
+	/* Selfgen stats per hwQ */
+	u32 num_bar;
+	u32 rts;
+	u32 cts2self;
+	u32 qos_null;
+
+	/* MPDU level stats */
+	u32 mpdu_tried_cnt;
+	u32 mpdu_queued_cnt;
+	u32 mpdu_ack_fail_cnt;
+	u32 mpdu_filt_cnt;
+	u32 false_mpdu_ack_count;
+
+	u32 txq_timeout;
+};
+
+/* NOTE: Variable length TLV, use length spec to infer array size */
+struct htt_tx_hwq_difs_latency_stats_tlv_v {
+	u32 hist_intvl;
+	/* histogram of ppdu post to hwsch - > cmd status received */
+	u32 difs_latency_hist[0]; /* HTT_TX_HWQ_MAX_DIFS_LATENCY_BINS */
+};
+
+/* NOTE: Variable length TLV, use length spec to infer array size */
+struct htt_tx_hwq_cmd_result_stats_tlv_v {
+	/* Histogram of sched cmd result */
+	u32 cmd_result[0]; /* HTT_TX_HWQ_MAX_CMD_RESULT_STATS */
+};
+
+/* NOTE: Variable length TLV, use length spec to infer array size */
+struct htt_tx_hwq_cmd_stall_stats_tlv_v {
+	/* Histogram of various pause conitions */
+	u32 cmd_stall_status[0]; /* HTT_TX_HWQ_MAX_CMD_STALL_STATS */
+};
+
+/* NOTE: Variable length TLV, use length spec to infer array size */
+struct htt_tx_hwq_fes_result_stats_tlv_v {
+	/* Histogram of number of user fes result */
+	u32 fes_result[0]; /* HTT_TX_HWQ_MAX_FES_RESULT_STATS */
+};
+
+/* NOTE: Variable length TLV, use length spec to infer array size
+ *
+ *  The hwq_tried_mpdu_cnt_hist is a  histogram of MPDUs tries per HWQ.
+ *  The tries here is the count of the  MPDUS within a PPDU that the HW
+ *  had attempted to transmit on  air, for the HWSCH Schedule command
+ *  submitted by FW in this HWQ .It is not the retry attempts. The
+ *  histogram bins are  0-29, 30-59, 60-89 and so on. The are 10 bins
+ *  in this histogram.
+ *  they are defined in FW using the following macros
+ *  #define WAL_MAX_TRIED_MPDU_CNT_HISTOGRAM 9
+ *  #define WAL_TRIED_MPDU_CNT_HISTOGRAM_INTERVAL 30
+ */
+struct htt_tx_hwq_tried_mpdu_cnt_hist_tlv_v {
+	u32 hist_bin_size;
+	/* Histogram of number of mpdus on tried mpdu */
+	u32 tried_mpdu_cnt_hist[0]; /* HTT_TX_HWQ_TRIED_MPDU_CNT_HIST */
+};
+
+/* NOTE: Variable length TLV, use length spec to infer array size
+ *
+ * The txop_used_cnt_hist is the histogram of txop per burst. After
+ * completing the burst, we identify the txop used in the burst and
+ * incr the corresponding bin.
+ * Each bin represents 1ms & we have 10 bins in this histogram.
+ * they are deined in FW using the following macros
+ * #define WAL_MAX_TXOP_USED_CNT_HISTOGRAM 10
+ * #define WAL_TXOP_USED_HISTOGRAM_INTERVAL 1000 ( 1 ms )
+ */
+struct htt_tx_hwq_txop_used_cnt_hist_tlv_v {
+	/* Histogram of txop used cnt */
+	u32 txop_used_cnt_hist[0]; /* HTT_TX_HWQ_TXOP_USED_CNT_HIST */
+};
+
+/* == TX SELFGEN STATS == */
+struct htt_tx_selfgen_cmn_stats_tlv {
+	u32 mac_id__word;
+	u32 su_bar;
+	u32 rts;
+	u32 cts2self;
+	u32 qos_null;
+	u32 delayed_bar_1; /* MU user 1 */
+	u32 delayed_bar_2; /* MU user 2 */
+	u32 delayed_bar_3; /* MU user 3 */
+	u32 delayed_bar_4; /* MU user 4 */
+	u32 delayed_bar_5; /* MU user 5 */
+	u32 delayed_bar_6; /* MU user 6 */
+	u32 delayed_bar_7; /* MU user 7 */
+};
+
+struct htt_tx_selfgen_ac_stats_tlv {
+	/* 11AC */
+	u32 ac_su_ndpa;
+	u32 ac_su_ndp;
+	u32 ac_mu_mimo_ndpa;
+	u32 ac_mu_mimo_ndp;
+	u32 ac_mu_mimo_brpoll_1; /* MU user 1 */
+	u32 ac_mu_mimo_brpoll_2; /* MU user 2 */
+	u32 ac_mu_mimo_brpoll_3; /* MU user 3 */
+};
+
+struct htt_tx_selfgen_ax_stats_tlv {
+	/* 11AX */
+	u32 ax_su_ndpa;
+	u32 ax_su_ndp;
+	u32 ax_mu_mimo_ndpa;
+	u32 ax_mu_mimo_ndp;
+	u32 ax_mu_mimo_brpoll_1; /* MU user 1 */
+	u32 ax_mu_mimo_brpoll_2; /* MU user 2 */
+	u32 ax_mu_mimo_brpoll_3; /* MU user 3 */
+	u32 ax_mu_mimo_brpoll_4; /* MU user 4 */
+	u32 ax_mu_mimo_brpoll_5; /* MU user 5 */
+	u32 ax_mu_mimo_brpoll_6; /* MU user 6 */
+	u32 ax_mu_mimo_brpoll_7; /* MU user 7 */
+	u32 ax_basic_trigger;
+	u32 ax_bsr_trigger;
+	u32 ax_mu_bar_trigger;
+	u32 ax_mu_rts_trigger;
+};
+
+struct htt_tx_selfgen_ac_err_stats_tlv {
+	/* 11AC error stats */
+	u32 ac_su_ndp_err;
+	u32 ac_su_ndpa_err;
+	u32 ac_mu_mimo_ndpa_err;
+	u32 ac_mu_mimo_ndp_err;
+	u32 ac_mu_mimo_brp1_err;
+	u32 ac_mu_mimo_brp2_err;
+	u32 ac_mu_mimo_brp3_err;
+};
+
+struct htt_tx_selfgen_ax_err_stats_tlv {
+	/* 11AX error stats */
+	u32 ax_su_ndp_err;
+	u32 ax_su_ndpa_err;
+	u32 ax_mu_mimo_ndpa_err;
+	u32 ax_mu_mimo_ndp_err;
+	u32 ax_mu_mimo_brp1_err;
+	u32 ax_mu_mimo_brp2_err;
+	u32 ax_mu_mimo_brp3_err;
+	u32 ax_mu_mimo_brp4_err;
+	u32 ax_mu_mimo_brp5_err;
+	u32 ax_mu_mimo_brp6_err;
+	u32 ax_mu_mimo_brp7_err;
+	u32 ax_basic_trigger_err;
+	u32 ax_bsr_trigger_err;
+	u32 ax_mu_bar_trigger_err;
+	u32 ax_mu_rts_trigger_err;
+};
+
+/* == TX MU STATS == */
+#define HTT_TX_PDEV_STATS_NUM_AC_MUMIMO_USER_STATS 4
+#define HTT_TX_PDEV_STATS_NUM_AX_MUMIMO_USER_STATS 8
+#define HTT_TX_PDEV_STATS_NUM_OFDMA_USER_STATS    74
+
+struct htt_tx_pdev_mu_mimo_sch_stats_tlv {
+	/* mu-mimo sw sched cmd stats */
+	u32 mu_mimo_sch_posted;
+	u32 mu_mimo_sch_failed;
+	/* MU PPDU stats per hwQ */
+	u32 mu_mimo_ppdu_posted;
+	/*
+	 * Counts the number of users in each transmission of
+	 * the given TX mode.
+	 *
+	 * Index is the number of users - 1.
+	 */
+	u32 ac_mu_mimo_sch_nusers[HTT_TX_PDEV_STATS_NUM_AC_MUMIMO_USER_STATS];
+	u32 ax_mu_mimo_sch_nusers[HTT_TX_PDEV_STATS_NUM_AX_MUMIMO_USER_STATS];
+	u32 ax_ofdma_sch_nusers[HTT_TX_PDEV_STATS_NUM_OFDMA_USER_STATS];
+};
+
+struct htt_tx_pdev_mu_mimo_mpdu_stats_tlv {
+	u32 mu_mimo_mpdus_queued_usr;
+	u32 mu_mimo_mpdus_tried_usr;
+	u32 mu_mimo_mpdus_failed_usr;
+	u32 mu_mimo_mpdus_requeued_usr;
+	u32 mu_mimo_err_no_ba_usr;
+	u32 mu_mimo_mpdu_underrun_usr;
+	u32 mu_mimo_ampdu_underrun_usr;
+
+	u32 ax_mu_mimo_mpdus_queued_usr;
+	u32 ax_mu_mimo_mpdus_tried_usr;
+	u32 ax_mu_mimo_mpdus_failed_usr;
+	u32 ax_mu_mimo_mpdus_requeued_usr;
+	u32 ax_mu_mimo_err_no_ba_usr;
+	u32 ax_mu_mimo_mpdu_underrun_usr;
+	u32 ax_mu_mimo_ampdu_underrun_usr;
+
+	u32 ax_ofdma_mpdus_queued_usr;
+	u32 ax_ofdma_mpdus_tried_usr;
+	u32 ax_ofdma_mpdus_failed_usr;
+	u32 ax_ofdma_mpdus_requeued_usr;
+	u32 ax_ofdma_err_no_ba_usr;
+	u32 ax_ofdma_mpdu_underrun_usr;
+	u32 ax_ofdma_ampdu_underrun_usr;
+};
+
+#define HTT_STATS_TX_SCHED_MODE_MU_MIMO_AC  1
+#define HTT_STATS_TX_SCHED_MODE_MU_MIMO_AX  2
+#define HTT_STATS_TX_SCHED_MODE_MU_OFDMA_AX 3
+
+struct htt_tx_pdev_mpdu_stats_tlv {
+	/* mpdu level stats */
+	u32 mpdus_queued_usr;
+	u32 mpdus_tried_usr;
+	u32 mpdus_failed_usr;
+	u32 mpdus_requeued_usr;
+	u32 err_no_ba_usr;
+	u32 mpdu_underrun_usr;
+	u32 ampdu_underrun_usr;
+	u32 user_index;
+	u32 tx_sched_mode; /* HTT_STATS_TX_SCHED_MODE_xxx */
+};
+
+/* == TX SCHED STATS == */
+/* NOTE: Variable length TLV, use length spec to infer array size */
+struct htt_sched_txq_cmd_posted_tlv_v {
+	u32 sched_cmd_posted[0]; /* HTT_TX_PDEV_SCHED_TX_MODE_MAX */
+};
+
+/* NOTE: Variable length TLV, use length spec to infer array size */
+struct htt_sched_txq_cmd_reaped_tlv_v {
+	u32 sched_cmd_reaped[0]; /* HTT_TX_PDEV_SCHED_TX_MODE_MAX */
+};
+
+/* NOTE: Variable length TLV, use length spec to infer array size */
+struct htt_sched_txq_sched_order_su_tlv_v {
+	u32 sched_order_su[0]; /* HTT_TX_PDEV_NUM_SCHED_ORDER_LOG */
+};
+
+enum htt_sched_txq_sched_ineligibility_tlv_enum {
+	HTT_SCHED_TID_SKIP_SCHED_MASK_DISABLED = 0,
+	HTT_SCHED_TID_SKIP_NOTIFY_MPDU,
+	HTT_SCHED_TID_SKIP_MPDU_STATE_INVALID,
+	HTT_SCHED_TID_SKIP_SCHED_DISABLED,
+	HTT_SCHED_TID_SKIP_TQM_BYPASS_CMD_PENDING,
+	HTT_SCHED_TID_SKIP_SECOND_SU_SCHEDULE,
+
+	HTT_SCHED_TID_SKIP_CMD_SLOT_NOT_AVAIL,
+	HTT_SCHED_TID_SKIP_NO_ENQ,
+	HTT_SCHED_TID_SKIP_LOW_ENQ,
+	HTT_SCHED_TID_SKIP_PAUSED,
+	HTT_SCHED_TID_SKIP_UL,
+	HTT_SCHED_TID_REMOVE_PAUSED,
+	HTT_SCHED_TID_REMOVE_NO_ENQ,
+	HTT_SCHED_TID_REMOVE_UL,
+	HTT_SCHED_TID_QUERY,
+	HTT_SCHED_TID_SU_ONLY,
+	HTT_SCHED_TID_ELIGIBLE,
+	HTT_SCHED_INELIGIBILITY_MAX,
+};
+
+/* NOTE: Variable length TLV, use length spec to infer array size */
+struct htt_sched_txq_sched_ineligibility_tlv_v {
+	/* indexed by htt_sched_txq_sched_ineligibility_tlv_enum */
+	u32 sched_ineligibility[0];
+};
+
+struct htt_tx_pdev_stats_sched_per_txq_tlv {
+	u32 mac_id__txq_id__word;
+	u32 sched_policy;
+	u32 last_sched_cmd_posted_timestamp;
+	u32 last_sched_cmd_compl_timestamp;
+	u32 sched_2_tac_lwm_count;
+	u32 sched_2_tac_ring_full;
+	u32 sched_cmd_post_failure;
+	u32 num_active_tids;
+	u32 num_ps_schedules;
+	u32 sched_cmds_pending;
+	u32 num_tid_register;
+	u32 num_tid_unregister;
+	u32 num_qstats_queried;
+	u32 qstats_update_pending;
+	u32 last_qstats_query_timestamp;
+	u32 num_tqm_cmdq_full;
+	u32 num_de_sched_algo_trigger;
+	u32 num_rt_sched_algo_trigger;
+	u32 num_tqm_sched_algo_trigger;
+	u32 notify_sched;
+	u32 dur_based_sendn_term;
+};
+
+struct htt_stats_tx_sched_cmn_tlv {
+	/* BIT [ 7 :  0]   :- mac_id
+	 * BIT [31 :  8]   :- reserved
+	 */
+	u32 mac_id__word;
+	/* Current timestamp */
+	u32 current_timestamp;
+};
+
+/* == TQM STATS == */
+#define HTT_TX_TQM_MAX_GEN_MPDU_END_REASON          16
+#define HTT_TX_TQM_MAX_LIST_MPDU_END_REASON         16
+#define HTT_TX_TQM_MAX_LIST_MPDU_CNT_HISTOGRAM_BINS 16
+
+/* NOTE: Variable length TLV, use length spec to infer array size */
+struct htt_tx_tqm_gen_mpdu_stats_tlv_v {
+	u32 gen_mpdu_end_reason[0]; /* HTT_TX_TQM_MAX_GEN_MPDU_END_REASON */
+};
+
+/* NOTE: Variable length TLV, use length spec to infer array size */
+struct htt_tx_tqm_list_mpdu_stats_tlv_v {
+	u32 list_mpdu_end_reason[0]; /* HTT_TX_TQM_MAX_LIST_MPDU_END_REASON */
+};
+
+/* NOTE: Variable length TLV, use length spec to infer array size */
+struct htt_tx_tqm_list_mpdu_cnt_tlv_v {
+	u32 list_mpdu_cnt_hist[0];
+			/* HTT_TX_TQM_MAX_LIST_MPDU_CNT_HISTOGRAM_BINS */
+};
+
+struct htt_tx_tqm_pdev_stats_tlv_v {
+	u32 msdu_count;
+	u32 mpdu_count;
+	u32 remove_msdu;
+	u32 remove_mpdu;
+	u32 remove_msdu_ttl;
+	u32 send_bar;
+	u32 bar_sync;
+	u32 notify_mpdu;
+	u32 sync_cmd;
+	u32 write_cmd;
+	u32 hwsch_trigger;
+	u32 ack_tlv_proc;
+	u32 gen_mpdu_cmd;
+	u32 gen_list_cmd;
+	u32 remove_mpdu_cmd;
+	u32 remove_mpdu_tried_cmd;
+	u32 mpdu_queue_stats_cmd;
+	u32 mpdu_head_info_cmd;
+	u32 msdu_flow_stats_cmd;
+	u32 remove_msdu_cmd;
+	u32 remove_msdu_ttl_cmd;
+	u32 flush_cache_cmd;
+	u32 update_mpduq_cmd;
+	u32 enqueue;
+	u32 enqueue_notify;
+	u32 notify_mpdu_at_head;
+	u32 notify_mpdu_state_valid;
+	/*
+	 * On receiving TQM_FLOW_NOT_EMPTY_STATUS from TQM, (on MSDUs being enqueued
+	 * the flow is non empty), if the number of MSDUs is greater than the threshold,
+	 * notify is incremented. UDP_THRESH counters are for UDP MSDUs, and NONUDP are
+	 * for non-UDP MSDUs.
+	 * MSDUQ_SWNOTIFY_UDP_THRESH1 threshold    - sched_udp_notify1 is incremented
+	 * MSDUQ_SWNOTIFY_UDP_THRESH2 threshold    - sched_udp_notify2 is incremented
+	 * MSDUQ_SWNOTIFY_NONUDP_THRESH1 threshold - sched_nonudp_notify1 is incremented
+	 * MSDUQ_SWNOTIFY_NONUDP_THRESH2 threshold - sched_nonudp_notify2 is incremented
+	 *
+	 * Notify signifies that we trigger the scheduler.
+	 */
+	u32 sched_udp_notify1;
+	u32 sched_udp_notify2;
+	u32 sched_nonudp_notify1;
+	u32 sched_nonudp_notify2;
+};
+
+struct htt_tx_tqm_cmn_stats_tlv {
+	u32 mac_id__word;
+	u32 max_cmdq_id;
+	u32 list_mpdu_cnt_hist_intvl;
+
+	/* Global stats */
+	u32 add_msdu;
+	u32 q_empty;
+	u32 q_not_empty;
+	u32 drop_notification;
+	u32 desc_threshold;
+};
+
+struct htt_tx_tqm_error_stats_tlv {
+	/* Error stats */
+	u32 q_empty_failure;
+	u32 q_not_empty_failure;
+	u32 add_msdu_failure;
+};
+
+/* == TQM CMDQ stats == */
+struct htt_tx_tqm_cmdq_status_tlv {
+	u32 mac_id__cmdq_id__word;
+	u32 sync_cmd;
+	u32 write_cmd;
+	u32 gen_mpdu_cmd;
+	u32 mpdu_queue_stats_cmd;
+	u32 mpdu_head_info_cmd;
+	u32 msdu_flow_stats_cmd;
+	u32 remove_mpdu_cmd;
+	u32 remove_msdu_cmd;
+	u32 flush_cache_cmd;
+	u32 update_mpduq_cmd;
+	u32 update_msduq_cmd;
+};
+
+/* == TX-DE STATS == */
+/* Structures for tx de stats */
+struct htt_tx_de_eapol_packets_stats_tlv {
+	u32 m1_packets;
+	u32 m2_packets;
+	u32 m3_packets;
+	u32 m4_packets;
+	u32 g1_packets;
+	u32 g2_packets;
+};
+
+struct htt_tx_de_classify_failed_stats_tlv {
+	u32 ap_bss_peer_not_found;
+	u32 ap_bcast_mcast_no_peer;
+	u32 sta_delete_in_progress;
+	u32 ibss_no_bss_peer;
+	u32 invalid_vdev_type;
+	u32 invalid_ast_peer_entry;
+	u32 peer_entry_invalid;
+	u32 ethertype_not_ip;
+	u32 eapol_lookup_failed;
+	u32 qpeer_not_allow_data;
+	u32 fse_tid_override;
+	u32 ipv6_jumbogram_zero_length;
+	u32 qos_to_non_qos_in_prog;
+};
+
+struct htt_tx_de_classify_stats_tlv {
+	u32 arp_packets;
+	u32 igmp_packets;
+	u32 dhcp_packets;
+	u32 host_inspected;
+	u32 htt_included;
+	u32 htt_valid_mcs;
+	u32 htt_valid_nss;
+	u32 htt_valid_preamble_type;
+	u32 htt_valid_chainmask;
+	u32 htt_valid_guard_interval;
+	u32 htt_valid_retries;
+	u32 htt_valid_bw_info;
+	u32 htt_valid_power;
+	u32 htt_valid_key_flags;
+	u32 htt_valid_no_encryption;
+	u32 fse_entry_count;
+	u32 fse_priority_be;
+	u32 fse_priority_high;
+	u32 fse_priority_low;
+	u32 fse_traffic_ptrn_be;
+	u32 fse_traffic_ptrn_over_sub;
+	u32 fse_traffic_ptrn_bursty;
+	u32 fse_traffic_ptrn_interactive;
+	u32 fse_traffic_ptrn_periodic;
+	u32 fse_hwqueue_alloc;
+	u32 fse_hwqueue_created;
+	u32 fse_hwqueue_send_to_host;
+	u32 mcast_entry;
+	u32 bcast_entry;
+	u32 htt_update_peer_cache;
+	u32 htt_learning_frame;
+	u32 fse_invalid_peer;
+	/*
+	 * mec_notify is HTT TX WBM multicast echo check notification
+	 * from firmware to host.  FW sends SA addresses to host for all
+	 * multicast/broadcast packets received on STA side.
+	 */
+	u32    mec_notify;
+};
+
+struct htt_tx_de_classify_status_stats_tlv {
+	u32 eok;
+	u32 classify_done;
+	u32 lookup_failed;
+	u32 send_host_dhcp;
+	u32 send_host_mcast;
+	u32 send_host_unknown_dest;
+	u32 send_host;
+	u32 status_invalid;
+};
+
+struct htt_tx_de_enqueue_packets_stats_tlv {
+	u32 enqueued_pkts;
+	u32 to_tqm;
+	u32 to_tqm_bypass;
+};
+
+struct htt_tx_de_enqueue_discard_stats_tlv {
+	u32 discarded_pkts;
+	u32 local_frames;
+	u32 is_ext_msdu;
+};
+
+struct htt_tx_de_compl_stats_tlv {
+	u32 tcl_dummy_frame;
+	u32 tqm_dummy_frame;
+	u32 tqm_notify_frame;
+	u32 fw2wbm_enq;
+	u32 tqm_bypass_frame;
+};
+
+/*
+ *  The htt_tx_de_fw2wbm_ring_full_hist_tlv is a histogram of time we waited
+ *  for the fw2wbm ring buffer.  we are requesting a buffer in FW2WBM release
+ *  ring,which may fail, due to non availability of buffer. Hence we sleep for
+ *  200us & again request for it. This is a histogram of time we wait, with
+ *  bin of 200ms & there are 10 bin (2 seconds max)
+ *  They are defined by the following macros in FW
+ *  #define ENTRIES_PER_BIN_COUNT 1000  // per bin 1000 * 200us = 200ms
+ *  #define RING_FULL_BIN_ENTRIES (WAL_TX_DE_FW2WBM_ALLOC_TIMEOUT_COUNT /
+ *                               ENTRIES_PER_BIN_COUNT)
+ */
+struct htt_tx_de_fw2wbm_ring_full_hist_tlv {
+	u32 fw2wbm_ring_full_hist[0];
+};
+
+struct htt_tx_de_cmn_stats_tlv {
+	u32   mac_id__word;
+
+	/* Global Stats */
+	u32   tcl2fw_entry_count;
+	u32   not_to_fw;
+	u32   invalid_pdev_vdev_peer;
+	u32   tcl_res_invalid_addrx;
+	u32   wbm2fw_entry_count;
+	u32   invalid_pdev;
+};
+
+/* == RING-IF STATS == */
+#define HTT_STATS_LOW_WM_BINS      5
+#define HTT_STATS_HIGH_WM_BINS     5
+
+struct htt_ring_if_stats_tlv {
+	u32 base_addr; /* DWORD aligned base memory address of the ring */
+	u32 elem_size;
+	u32 num_elems__prefetch_tail_idx;
+	u32 head_idx__tail_idx;
+	u32 shadow_head_idx__shadow_tail_idx;
+	u32 num_tail_incr;
+	u32 lwm_thresh__hwm_thresh;
+	u32 overrun_hit_count;
+	u32 underrun_hit_count;
+	u32 prod_blockwait_count;
+	u32 cons_blockwait_count;
+	u32 low_wm_hit_count[HTT_STATS_LOW_WM_BINS];
+	u32 high_wm_hit_count[HTT_STATS_HIGH_WM_BINS];
+};
+
+struct htt_ring_if_cmn_tlv {
+	u32 mac_id__word;
+	u32 num_records;
+};
+
+/* == SFM STATS == */
+/* NOTE: Variable length TLV, use length spec to infer array size */
+struct htt_sfm_client_user_tlv_v {
+	/* Number of DWORDS used per user and per client */
+	u32 dwords_used_by_user_n[0];
+};
+
+struct htt_sfm_client_tlv {
+	/* Client ID */
+	u32 client_id;
+	/* Minimum number of buffers */
+	u32 buf_min;
+	/* Maximum number of buffers */
+	u32 buf_max;
+	/* Number of Busy buffers */
+	u32 buf_busy;
+	/* Number of Allocated buffers */
+	u32 buf_alloc;
+	/* Number of Available/Usable buffers */
+	u32 buf_avail;
+	/* Number of users */
+	u32 num_users;
+};
+
+struct htt_sfm_cmn_tlv {
+	u32 mac_id__word;
+	/* Indicates the total number of 128 byte buffers
+	 * in the CMEM that are available for buffer sharing
+	 */
+	u32 buf_total;
+	/* Indicates for certain client or all the clients
+	 * there is no dowrd saved in SFM, refer to SFM_R1_MEM_EMPTY
+	 */
+	u32 mem_empty;
+	/* DEALLOCATE_BUFFERS, refer to register SFM_R0_DEALLOCATE_BUFFERS */
+	u32 deallocate_bufs;
+	/* Number of Records */
+	u32 num_records;
+};
+
+/* == SRNG STATS == */
+struct htt_sring_stats_tlv {
+	u32 mac_id__ring_id__arena__ep;
+	u32 base_addr_lsb; /* DWORD aligned base memory address of the ring */
+	u32 base_addr_msb;
+	u32 ring_size;
+	u32 elem_size;
+
+	u32 num_avail_words__num_valid_words;
+	u32 head_ptr__tail_ptr;
+	u32 consumer_empty__producer_full;
+	u32 prefetch_count__internal_tail_ptr;
+};
+
+struct htt_sring_cmn_tlv {
+	u32 num_records;
+};
+
+/* == PDEV TX RATE CTRL STATS == */
+#define HTT_TX_PDEV_STATS_NUM_MCS_COUNTERS        12
+#define HTT_TX_PDEV_STATS_NUM_GI_COUNTERS          4
+#define HTT_TX_PDEV_STATS_NUM_DCM_COUNTERS         5
+#define HTT_TX_PDEV_STATS_NUM_BW_COUNTERS          4
+#define HTT_TX_PDEV_STATS_NUM_SPATIAL_STREAMS      8
+#define HTT_TX_PDEV_STATS_NUM_PREAMBLE_TYPES       HTT_STATS_PREAM_COUNT
+#define HTT_TX_PDEV_STATS_NUM_LEGACY_CCK_STATS     4
+#define HTT_TX_PDEV_STATS_NUM_LEGACY_OFDM_STATS    8
+#define HTT_TX_PDEV_STATS_NUM_LTF                  4
+
+#define HTT_TX_NUM_OF_SOUNDING_STATS_WORDS \
+	(HTT_TX_PDEV_STATS_NUM_BW_COUNTERS * \
+	 HTT_TX_PDEV_STATS_NUM_AX_MUMIMO_USER_STATS)
+
+struct htt_tx_pdev_rate_stats_tlv {
+	u32 mac_id__word;
+	u32 tx_ldpc;
+	u32 rts_cnt;
+	/* RSSI value of last ack packet (units = dB above noise floor) */
+	u32 ack_rssi;
+
+	u32 tx_mcs[HTT_TX_PDEV_STATS_NUM_MCS_COUNTERS];
+
+	u32 tx_su_mcs[HTT_TX_PDEV_STATS_NUM_MCS_COUNTERS];
+	u32 tx_mu_mcs[HTT_TX_PDEV_STATS_NUM_MCS_COUNTERS];
+
+	/* element 0,1, ...7 -> NSS 1,2, ...8 */
+	u32 tx_nss[HTT_TX_PDEV_STATS_NUM_SPATIAL_STREAMS];
+	/* element 0: 20 MHz, 1: 40 MHz, 2: 80 MHz, 3: 160 and 80+80 MHz */
+	u32 tx_bw[HTT_TX_PDEV_STATS_NUM_BW_COUNTERS];
+	u32 tx_stbc[HTT_TX_PDEV_STATS_NUM_MCS_COUNTERS];
+	u32 tx_pream[HTT_TX_PDEV_STATS_NUM_PREAMBLE_TYPES];
+
+	/* Counters to track number of tx packets
+	 * in each GI (400us, 800us, 1600us & 3200us) in each mcs (0-11)
+	 */
+	u32 tx_gi[HTT_TX_PDEV_STATS_NUM_GI_COUNTERS][HTT_TX_PDEV_STATS_NUM_MCS_COUNTERS];
+
+	/* Counters to track packets in dcm mcs (MCS 0, 1, 3, 4) */
+	u32 tx_dcm[HTT_TX_PDEV_STATS_NUM_DCM_COUNTERS];
+	/* Number of CTS-acknowledged RTS packets */
+	u32 rts_success;
+
+	/*
+	 * Counters for legacy 11a and 11b transmissions.
+	 *
+	 * The index corresponds to:
+	 *
+	 * CCK: 0: 1 Mbps, 1: 2 Mbps, 2: 5.5 Mbps, 3: 11 Mbps
+	 *
+	 * OFDM: 0: 6 Mbps, 1: 9 Mbps, 2: 12 Mbps, 3: 18 Mbps,
+	 *       4: 24 Mbps, 5: 36 Mbps, 6: 48 Mbps, 7: 54 Mbps
+	 */
+	u32 tx_legacy_cck_rate[HTT_TX_PDEV_STATS_NUM_LEGACY_CCK_STATS];
+	u32 tx_legacy_ofdm_rate[HTT_TX_PDEV_STATS_NUM_LEGACY_OFDM_STATS];
+
+	u32 ac_mu_mimo_tx_ldpc;
+	u32 ax_mu_mimo_tx_ldpc;
+	u32 ofdma_tx_ldpc;
+
+	/*
+	 * Counters for 11ax HE LTF selection during TX.
+	 *
+	 * The index corresponds to:
+	 *
+	 * 0: unused, 1: 1x LTF, 2: 2x LTF, 3: 4x LTF
+	 */
+	u32 tx_he_ltf[HTT_TX_PDEV_STATS_NUM_LTF];
+
+	u32 ac_mu_mimo_tx_mcs[HTT_TX_PDEV_STATS_NUM_MCS_COUNTERS];
+	u32 ax_mu_mimo_tx_mcs[HTT_TX_PDEV_STATS_NUM_MCS_COUNTERS];
+	u32 ofdma_tx_mcs[HTT_TX_PDEV_STATS_NUM_MCS_COUNTERS];
+
+	u32 ac_mu_mimo_tx_nss[HTT_TX_PDEV_STATS_NUM_SPATIAL_STREAMS];
+	u32 ax_mu_mimo_tx_nss[HTT_TX_PDEV_STATS_NUM_SPATIAL_STREAMS];
+	u32 ofdma_tx_nss[HTT_TX_PDEV_STATS_NUM_SPATIAL_STREAMS];
+
+	u32 ac_mu_mimo_tx_bw[HTT_TX_PDEV_STATS_NUM_BW_COUNTERS];
+	u32 ax_mu_mimo_tx_bw[HTT_TX_PDEV_STATS_NUM_BW_COUNTERS];
+	u32 ofdma_tx_bw[HTT_TX_PDEV_STATS_NUM_BW_COUNTERS];
+
+	u32 ac_mu_mimo_tx_gi[HTT_TX_PDEV_STATS_NUM_GI_COUNTERS]
+			    [HTT_TX_PDEV_STATS_NUM_MCS_COUNTERS];
+	u32 ax_mu_mimo_tx_gi[HTT_TX_PDEV_STATS_NUM_GI_COUNTERS]
+			    [HTT_TX_PDEV_STATS_NUM_MCS_COUNTERS];
+	u32 ofdma_tx_gi[HTT_TX_PDEV_STATS_NUM_GI_COUNTERS]
+		       [HTT_TX_PDEV_STATS_NUM_MCS_COUNTERS];
+};
+
+/* == PDEV RX RATE CTRL STATS == */
+#define HTT_RX_PDEV_STATS_NUM_LEGACY_CCK_STATS     4
+#define HTT_RX_PDEV_STATS_NUM_LEGACY_OFDM_STATS    8
+#define HTT_RX_PDEV_STATS_NUM_MCS_COUNTERS        12
+#define HTT_RX_PDEV_STATS_NUM_GI_COUNTERS          4
+#define HTT_RX_PDEV_STATS_NUM_DCM_COUNTERS         5
+#define HTT_RX_PDEV_STATS_NUM_BW_COUNTERS          4
+#define HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS      8
+#define HTT_RX_PDEV_STATS_NUM_PREAMBLE_TYPES       HTT_STATS_PREAM_COUNT
+
+struct htt_rx_pdev_rate_stats_tlv {
+	u32 mac_id__word;
+	u32 nsts;
+
+	u32 rx_ldpc;
+	u32 rts_cnt;
+
+	u32 rssi_mgmt; /* units = dB above noise floor */
+	u32 rssi_data; /* units = dB above noise floor */
+	u32 rssi_comb; /* units = dB above noise floor */
+	u32 rx_mcs[HTT_RX_PDEV_STATS_NUM_MCS_COUNTERS];
+	/* element 0,1, ...7 -> NSS 1,2, ...8 */
+	u32 rx_nss[HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS];
+	u32 rx_dcm[HTT_RX_PDEV_STATS_NUM_DCM_COUNTERS];
+	u32 rx_stbc[HTT_RX_PDEV_STATS_NUM_MCS_COUNTERS];
+	/* element 0: 20 MHz, 1: 40 MHz, 2: 80 MHz, 3: 160 and 80+80 MHz */
+	u32 rx_bw[HTT_RX_PDEV_STATS_NUM_BW_COUNTERS];
+	u32 rx_pream[HTT_RX_PDEV_STATS_NUM_PREAMBLE_TYPES];
+	u8 rssi_chain[HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS]
+		     [HTT_RX_PDEV_STATS_NUM_BW_COUNTERS];
+					/* units = dB above noise floor */
+
+	/* Counters to track number of rx packets
+	 * in each GI in each mcs (0-11)
+	 */
+	u32 rx_gi[HTT_RX_PDEV_STATS_NUM_GI_COUNTERS][HTT_RX_PDEV_STATS_NUM_MCS_COUNTERS];
+	s32 rssi_in_dbm; /* rx Signal Strength value in dBm unit */
+
+	u32 rx_11ax_su_ext;
+	u32 rx_11ac_mumimo;
+	u32 rx_11ax_mumimo;
+	u32 rx_11ax_ofdma;
+	u32 txbf;
+	u32 rx_legacy_cck_rate[HTT_RX_PDEV_STATS_NUM_LEGACY_CCK_STATS];
+	u32 rx_legacy_ofdm_rate[HTT_RX_PDEV_STATS_NUM_LEGACY_OFDM_STATS];
+	u32 rx_active_dur_us_low;
+	u32 rx_active_dur_us_high;
+};
+
+/* == RX PDEV/SOC STATS == */
+struct htt_rx_soc_fw_stats_tlv {
+	u32 fw_reo_ring_data_msdu;
+	u32 fw_to_host_data_msdu_bcmc;
+	u32 fw_to_host_data_msdu_uc;
+	u32 ofld_remote_data_buf_recycle_cnt;
+	u32 ofld_remote_free_buf_indication_cnt;
+
+	u32 ofld_buf_to_host_data_msdu_uc;
+	u32 reo_fw_ring_to_host_data_msdu_uc;
+
+	u32 wbm_sw_ring_reap;
+	u32 wbm_forward_to_host_cnt;
+	u32 wbm_target_recycle_cnt;
+
+	u32 target_refill_ring_recycle_cnt;
+};
+
+/* NOTE: Variable length TLV, use length spec to infer array size */
+struct htt_rx_soc_fw_refill_ring_empty_tlv_v {
+	u32 refill_ring_empty_cnt[0]; /* HTT_RX_STATS_REFILL_MAX_RING */
+};
+
+/* NOTE: Variable length TLV, use length spec to infer array size */
+struct htt_rx_soc_fw_refill_ring_num_refill_tlv_v {
+	u32 refill_ring_num_refill[0]; /* HTT_RX_STATS_REFILL_MAX_RING */
+};
+
+/* RXDMA error code from WBM released packets */
+enum htt_rx_rxdma_error_code_enum {
+	HTT_RX_RXDMA_OVERFLOW_ERR                           = 0,
+	HTT_RX_RXDMA_MPDU_LENGTH_ERR                        = 1,
+	HTT_RX_RXDMA_FCS_ERR                                = 2,
+	HTT_RX_RXDMA_DECRYPT_ERR                            = 3,
+	HTT_RX_RXDMA_TKIP_MIC_ERR                           = 4,
+	HTT_RX_RXDMA_UNECRYPTED_ERR                         = 5,
+	HTT_RX_RXDMA_MSDU_LEN_ERR                           = 6,
+	HTT_RX_RXDMA_MSDU_LIMIT_ERR                         = 7,
+	HTT_RX_RXDMA_WIFI_PARSE_ERR                         = 8,
+	HTT_RX_RXDMA_AMSDU_PARSE_ERR                        = 9,
+	HTT_RX_RXDMA_SA_TIMEOUT_ERR                         = 10,
+	HTT_RX_RXDMA_DA_TIMEOUT_ERR                         = 11,
+	HTT_RX_RXDMA_FLOW_TIMEOUT_ERR                       = 12,
+	HTT_RX_RXDMA_FLUSH_REQUEST                          = 13,
+	HTT_RX_RXDMA_ERR_CODE_RVSD0                         = 14,
+	HTT_RX_RXDMA_ERR_CODE_RVSD1                         = 15,
+
+	/* This MAX_ERR_CODE should not be used in any host/target messages,
+	 * so that even though it is defined within a host/target interface
+	 * definition header file, it isn't actually part of the host/target
+	 * interface, and thus can be modified.
+	 */
+	HTT_RX_RXDMA_MAX_ERR_CODE
+};
+
+/* NOTE: Variable length TLV, use length spec to infer array size */
+struct htt_rx_soc_fw_refill_ring_num_rxdma_err_tlv_v {
+	u32 rxdma_err[0]; /* HTT_RX_RXDMA_MAX_ERR_CODE */
+};
+
+/* REO error code from WBM released packets */
+enum htt_rx_reo_error_code_enum {
+	HTT_RX_REO_QUEUE_DESC_ADDR_ZERO                     = 0,
+	HTT_RX_REO_QUEUE_DESC_NOT_VALID                     = 1,
+	HTT_RX_AMPDU_IN_NON_BA                              = 2,
+	HTT_RX_NON_BA_DUPLICATE                             = 3,
+	HTT_RX_BA_DUPLICATE                                 = 4,
+	HTT_RX_REGULAR_FRAME_2K_JUMP                        = 5,
+	HTT_RX_BAR_FRAME_2K_JUMP                            = 6,
+	HTT_RX_REGULAR_FRAME_OOR                            = 7,
+	HTT_RX_BAR_FRAME_OOR                                = 8,
+	HTT_RX_BAR_FRAME_NO_BA_SESSION                      = 9,
+	HTT_RX_BAR_FRAME_SN_EQUALS_SSN                      = 10,
+	HTT_RX_PN_CHECK_FAILED                              = 11,
+	HTT_RX_2K_ERROR_HANDLING_FLAG_SET                   = 12,
+	HTT_RX_PN_ERROR_HANDLING_FLAG_SET                   = 13,
+	HTT_RX_QUEUE_DESCRIPTOR_BLOCKED_SET                 = 14,
+	HTT_RX_REO_ERR_CODE_RVSD                            = 15,
+
+	/* This MAX_ERR_CODE should not be used in any host/target messages,
+	 * so that even though it is defined within a host/target interface
+	 * definition header file, it isn't actually part of the host/target
+	 * interface, and thus can be modified.
+	 */
+	HTT_RX_REO_MAX_ERR_CODE
+};
+
+/* NOTE: Variable length TLV, use length spec to infer array size */
+struct htt_rx_soc_fw_refill_ring_num_reo_err_tlv_v {
+	u32 reo_err[0]; /* HTT_RX_REO_MAX_ERR_CODE */
+};
+
+/* == RX PDEV STATS == */
+#define HTT_STATS_SUBTYPE_MAX     16
+
+struct htt_rx_pdev_fw_stats_tlv {
+	u32 mac_id__word;
+	u32 ppdu_recvd;
+	u32 mpdu_cnt_fcs_ok;
+	u32 mpdu_cnt_fcs_err;
+	u32 tcp_msdu_cnt;
+	u32 tcp_ack_msdu_cnt;
+	u32 udp_msdu_cnt;
+	u32 other_msdu_cnt;
+	u32 fw_ring_mpdu_ind;
+	u32 fw_ring_mgmt_subtype[HTT_STATS_SUBTYPE_MAX];
+	u32 fw_ring_ctrl_subtype[HTT_STATS_SUBTYPE_MAX];
+	u32 fw_ring_mcast_data_msdu;
+	u32 fw_ring_bcast_data_msdu;
+	u32 fw_ring_ucast_data_msdu;
+	u32 fw_ring_null_data_msdu;
+	u32 fw_ring_mpdu_drop;
+	u32 ofld_local_data_ind_cnt;
+	u32 ofld_local_data_buf_recycle_cnt;
+	u32 drx_local_data_ind_cnt;
+	u32 drx_local_data_buf_recycle_cnt;
+	u32 local_nondata_ind_cnt;
+	u32 local_nondata_buf_recycle_cnt;
+
+	u32 fw_status_buf_ring_refill_cnt;
+	u32 fw_status_buf_ring_empty_cnt;
+	u32 fw_pkt_buf_ring_refill_cnt;
+	u32 fw_pkt_buf_ring_empty_cnt;
+	u32 fw_link_buf_ring_refill_cnt;
+	u32 fw_link_buf_ring_empty_cnt;
+
+	u32 host_pkt_buf_ring_refill_cnt;
+	u32 host_pkt_buf_ring_empty_cnt;
+	u32 mon_pkt_buf_ring_refill_cnt;
+	u32 mon_pkt_buf_ring_empty_cnt;
+	u32 mon_status_buf_ring_refill_cnt;
+	u32 mon_status_buf_ring_empty_cnt;
+	u32 mon_desc_buf_ring_refill_cnt;
+	u32 mon_desc_buf_ring_empty_cnt;
+	u32 mon_dest_ring_update_cnt;
+	u32 mon_dest_ring_full_cnt;
+
+	u32 rx_suspend_cnt;
+	u32 rx_suspend_fail_cnt;
+	u32 rx_resume_cnt;
+	u32 rx_resume_fail_cnt;
+	u32 rx_ring_switch_cnt;
+	u32 rx_ring_restore_cnt;
+	u32 rx_flush_cnt;
+	u32 rx_recovery_reset_cnt;
+};
+
+#define HTT_STATS_PHY_ERR_MAX 43
+
+struct htt_rx_pdev_fw_stats_phy_err_tlv {
+	u32 mac_id__word;
+	u32 total_phy_err_cnt;
+	/* Counts of different types of phy errs
+	 * The mapping of PHY error types to phy_err array elements is HW dependent.
+	 * The only currently-supported mapping is shown below:
+	 *
+	 * 0 phyrx_err_phy_off Reception aborted due to receiving a PHY_OFF TLV
+	 * 1 phyrx_err_synth_off
+	 * 2 phyrx_err_ofdma_timing
+	 * 3 phyrx_err_ofdma_signal_parity
+	 * 4 phyrx_err_ofdma_rate_illegal
+	 * 5 phyrx_err_ofdma_length_illegal
+	 * 6 phyrx_err_ofdma_restart
+	 * 7 phyrx_err_ofdma_service
+	 * 8 phyrx_err_ppdu_ofdma_power_drop
+	 * 9 phyrx_err_cck_blokker
+	 * 10 phyrx_err_cck_timing
+	 * 11 phyrx_err_cck_header_crc
+	 * 12 phyrx_err_cck_rate_illegal
+	 * 13 phyrx_err_cck_length_illegal
+	 * 14 phyrx_err_cck_restart
+	 * 15 phyrx_err_cck_service
+	 * 16 phyrx_err_cck_power_drop
+	 * 17 phyrx_err_ht_crc_err
+	 * 18 phyrx_err_ht_length_illegal
+	 * 19 phyrx_err_ht_rate_illegal
+	 * 20 phyrx_err_ht_zlf
+	 * 21 phyrx_err_false_radar_ext
+	 * 22 phyrx_err_green_field
+	 * 23 phyrx_err_bw_gt_dyn_bw
+	 * 24 phyrx_err_leg_ht_mismatch
+	 * 25 phyrx_err_vht_crc_error
+	 * 26 phyrx_err_vht_siga_unsupported
+	 * 27 phyrx_err_vht_lsig_len_invalid
+	 * 28 phyrx_err_vht_ndp_or_zlf
+	 * 29 phyrx_err_vht_nsym_lt_zero
+	 * 30 phyrx_err_vht_rx_extra_symbol_mismatch
+	 * 31 phyrx_err_vht_rx_skip_group_id0
+	 * 32 phyrx_err_vht_rx_skip_group_id1to62
+	 * 33 phyrx_err_vht_rx_skip_group_id63
+	 * 34 phyrx_err_ofdm_ldpc_decoder_disabled
+	 * 35 phyrx_err_defer_nap
+	 * 36 phyrx_err_fdomain_timeout
+	 * 37 phyrx_err_lsig_rel_check
+	 * 38 phyrx_err_bt_collision
+	 * 39 phyrx_err_unsupported_mu_feedback
+	 * 40 phyrx_err_ppdu_tx_interrupt_rx
+	 * 41 phyrx_err_unsupported_cbf
+	 * 42 phyrx_err_other
+	 */
+	u32 phy_err[HTT_STATS_PHY_ERR_MAX];
+};
+
+/* NOTE: Variable length TLV, use length spec to infer array size */
+struct htt_rx_pdev_fw_ring_mpdu_err_tlv_v {
+	/* Num error MPDU for each RxDMA error type  */
+	u32 fw_ring_mpdu_err[0]; /* HTT_RX_STATS_RXDMA_MAX_ERR */
+};
+
+/* NOTE: Variable length TLV, use length spec to infer array size */
+struct htt_rx_pdev_fw_mpdu_drop_tlv_v {
+	/* Num MPDU dropped  */
+	u32 fw_mpdu_drop[0]; /* HTT_RX_STATS_FW_DROP_REASON_MAX */
+};
+
+#define HTT_PDEV_CCA_STATS_TX_FRAME_INFO_PRESENT               (0x1)
+#define HTT_PDEV_CCA_STATS_RX_FRAME_INFO_PRESENT               (0x2)
+#define HTT_PDEV_CCA_STATS_RX_CLEAR_INFO_PRESENT               (0x4)
+#define HTT_PDEV_CCA_STATS_MY_RX_FRAME_INFO_PRESENT            (0x8)
+#define HTT_PDEV_CCA_STATS_USEC_CNT_INFO_PRESENT              (0x10)
+#define HTT_PDEV_CCA_STATS_MED_RX_IDLE_INFO_PRESENT           (0x20)
+#define HTT_PDEV_CCA_STATS_MED_TX_IDLE_GLOBAL_INFO_PRESENT    (0x40)
+#define HTT_PDEV_CCA_STATS_CCA_OBBS_USEC_INFO_PRESENT         (0x80)
+
+struct htt_pdev_stats_cca_counters_tlv {
+	/* Below values are obtained from the HW Cycles counter registers */
+	u32 tx_frame_usec;
+	u32 rx_frame_usec;
+	u32 rx_clear_usec;
+	u32 my_rx_frame_usec;
+	u32 usec_cnt;
+	u32 med_rx_idle_usec;
+	u32 med_tx_idle_global_usec;
+	u32 cca_obss_usec;
+};
+
+struct htt_pdev_cca_stats_hist_v1_tlv {
+	u32    chan_num;
+	/* num of CCA records (Num of htt_pdev_stats_cca_counters_tlv)*/
+	u32    num_records;
+	u32    valid_cca_counters_bitmap;
+	u32    collection_interval;
+
+	/* This will be followed by an array which contains the CCA stats
+	 * collected in the last N intervals,
+	 * if the indication is for last N intervals CCA stats.
+	 * Then the pdev_cca_stats[0] element contains the oldest CCA stats
+	 * and pdev_cca_stats[N-1] will have the most recent CCA stats.
+	 * htt_pdev_stats_cca_counters_tlv cca_hist_tlv[1];
+	 */
+};
+
+struct htt_pdev_stats_twt_session_tlv {
+	u32 vdev_id;
+	struct htt_mac_addr peer_mac;
+	u32 flow_id_flags;
+
+	/* TWT_DIALOG_ID_UNAVAILABLE is used
+	 * when TWT session is not initiated by host
+	 */
+	u32 dialog_id;
+	u32 wake_dura_us;
+	u32 wake_intvl_us;
+	u32 sp_offset_us;
+};
+
+struct htt_pdev_stats_twt_sessions_tlv {
+	u32 pdev_id;
+	u32 num_sessions;
+	struct htt_pdev_stats_twt_session_tlv twt_session[0];
+};
+
+enum htt_rx_reo_resource_sample_id_enum {
+	/* Global link descriptor queued in REO */
+	HTT_RX_REO_RESOURCE_GLOBAL_LINK_DESC_COUNT_0           = 0,
+	HTT_RX_REO_RESOURCE_GLOBAL_LINK_DESC_COUNT_1           = 1,
+	HTT_RX_REO_RESOURCE_GLOBAL_LINK_DESC_COUNT_2           = 2,
+	/*Number of queue descriptors of this aging group */
+	HTT_RX_REO_RESOURCE_BUFFERS_USED_AC0                   = 3,
+	HTT_RX_REO_RESOURCE_BUFFERS_USED_AC1                   = 4,
+	HTT_RX_REO_RESOURCE_BUFFERS_USED_AC2                   = 5,
+	HTT_RX_REO_RESOURCE_BUFFERS_USED_AC3                   = 6,
+	/* Total number of MSDUs buffered in AC */
+	HTT_RX_REO_RESOURCE_AGING_NUM_QUEUES_AC0               = 7,
+	HTT_RX_REO_RESOURCE_AGING_NUM_QUEUES_AC1               = 8,
+	HTT_RX_REO_RESOURCE_AGING_NUM_QUEUES_AC2               = 9,
+	HTT_RX_REO_RESOURCE_AGING_NUM_QUEUES_AC3               = 10,
+
+	HTT_RX_REO_RESOURCE_STATS_MAX                          = 16
+};
+
+struct htt_rx_reo_resource_stats_tlv_v {
+	/* Variable based on the Number of records. HTT_RX_REO_RESOURCE_STATS_MAX */
+	u32 sample_id;
+	u32 total_max;
+	u32 total_avg;
+	u32 total_sample;
+	u32 non_zeros_avg;
+	u32 non_zeros_sample;
+	u32 last_non_zeros_max;
+	u32 last_non_zeros_min;
+	u32 last_non_zeros_avg;
+	u32 last_non_zeros_sample;
+};
+
+/* == TX SOUNDING STATS == */
+
+enum htt_txbf_sound_steer_modes {
+	HTT_IMPLICIT_TXBF_STEER_STATS                = 0,
+	HTT_EXPLICIT_TXBF_SU_SIFS_STEER_STATS        = 1,
+	HTT_EXPLICIT_TXBF_SU_RBO_STEER_STATS         = 2,
+	HTT_EXPLICIT_TXBF_MU_SIFS_STEER_STATS        = 3,
+	HTT_EXPLICIT_TXBF_MU_RBO_STEER_STATS         = 4,
+	HTT_TXBF_MAX_NUM_OF_MODES                    = 5
+};
+
+enum htt_stats_sounding_tx_mode {
+	HTT_TX_AC_SOUNDING_MODE                      = 0,
+	HTT_TX_AX_SOUNDING_MODE                      = 1,
+};
+
+struct htt_tx_sounding_stats_tlv {
+	u32 tx_sounding_mode; /* HTT_TX_XX_SOUNDING_MODE */
+	/* Counts number of soundings for all steering modes in each bw */
+	u32 cbf_20[HTT_TXBF_MAX_NUM_OF_MODES];
+	u32 cbf_40[HTT_TXBF_MAX_NUM_OF_MODES];
+	u32 cbf_80[HTT_TXBF_MAX_NUM_OF_MODES];
+	u32 cbf_160[HTT_TXBF_MAX_NUM_OF_MODES];
+	/*
+	 * The sounding array is a 2-D array stored as an 1-D array of
+	 * u32. The stats for a particular user/bw combination is
+	 * referenced with the following:
+	 *
+	 *          sounding[(user* max_bw) + bw]
+	 *
+	 * ... where max_bw == 4 for 160mhz
+	 */
+	u32 sounding[HTT_TX_NUM_OF_SOUNDING_STATS_WORDS];
+};
+
+struct htt_pdev_obss_pd_stats_tlv {
+	u32        num_obss_tx_ppdu_success;
+	u32        num_obss_tx_ppdu_failure;
+};
+
+void ath11k_debug_htt_stats_init(struct ath11k *ar);
+#endif
diff --git a/drivers/net/wireless/ath/ath11k/debugfs_sta.c b/drivers/net/wireless/ath/ath11k/debugfs_sta.c
new file mode 100644
index 000000000000..b392117eff3c
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/debugfs_sta.c
@@ -0,0 +1,538 @@
+// SPDX-License-Identifier: BSD-3-Clause-Clear
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#include <linux/vmalloc.h>
+
+#include "core.h"
+#include "peer.h"
+#include "debug.h"
+
+void
+ath11k_accumulate_per_peer_tx_stats(struct ath11k_sta *arsta,
+				    struct ath11k_per_peer_tx_stats *peer_stats,
+				    u8 legacy_rate_idx)
+{
+	struct rate_info *txrate = &arsta->txrate;
+	struct ath11k_htt_tx_stats *tx_stats;
+	int gi, mcs, bw, nss;
+
+	if (!arsta->tx_stats)
+		return;
+
+	tx_stats = arsta->tx_stats;
+	gi = FIELD_GET(RATE_INFO_FLAGS_SHORT_GI, arsta->txrate.flags);
+	mcs = txrate->mcs;
+	bw = txrate->bw;
+	nss = txrate->nss - 1;
+
+#define STATS_OP_FMT(name) tx_stats->stats[ATH11K_STATS_TYPE_##name]
+
+	if (txrate->flags & RATE_INFO_FLAGS_HE_MCS) {
+		STATS_OP_FMT(SUCC).he[0][mcs] += peer_stats->succ_bytes;
+		STATS_OP_FMT(SUCC).he[1][mcs] += peer_stats->succ_pkts;
+		STATS_OP_FMT(FAIL).he[0][mcs] += peer_stats->failed_bytes;
+		STATS_OP_FMT(FAIL).he[1][mcs] += peer_stats->failed_pkts;
+		STATS_OP_FMT(RETRY).he[0][mcs] += peer_stats->retry_bytes;
+		STATS_OP_FMT(RETRY).he[1][mcs] += peer_stats->retry_pkts;
+	} else if (txrate->flags & RATE_INFO_FLAGS_VHT_MCS) {
+		STATS_OP_FMT(SUCC).vht[0][mcs] += peer_stats->succ_bytes;
+		STATS_OP_FMT(SUCC).vht[1][mcs] += peer_stats->succ_pkts;
+		STATS_OP_FMT(FAIL).vht[0][mcs] += peer_stats->failed_bytes;
+		STATS_OP_FMT(FAIL).vht[1][mcs] += peer_stats->failed_pkts;
+		STATS_OP_FMT(RETRY).vht[0][mcs] += peer_stats->retry_bytes;
+		STATS_OP_FMT(RETRY).vht[1][mcs] += peer_stats->retry_pkts;
+	} else if (txrate->flags & RATE_INFO_FLAGS_MCS) {
+		STATS_OP_FMT(SUCC).ht[0][mcs] += peer_stats->succ_bytes;
+		STATS_OP_FMT(SUCC).ht[1][mcs] += peer_stats->succ_pkts;
+		STATS_OP_FMT(FAIL).ht[0][mcs] += peer_stats->failed_bytes;
+		STATS_OP_FMT(FAIL).ht[1][mcs] += peer_stats->failed_pkts;
+		STATS_OP_FMT(RETRY).ht[0][mcs] += peer_stats->retry_bytes;
+		STATS_OP_FMT(RETRY).ht[1][mcs] += peer_stats->retry_pkts;
+	} else {
+		mcs = legacy_rate_idx;
+
+		STATS_OP_FMT(SUCC).legacy[0][mcs] += peer_stats->succ_bytes;
+		STATS_OP_FMT(SUCC).legacy[1][mcs] += peer_stats->succ_pkts;
+		STATS_OP_FMT(FAIL).legacy[0][mcs] += peer_stats->failed_bytes;
+		STATS_OP_FMT(FAIL).legacy[1][mcs] += peer_stats->failed_pkts;
+		STATS_OP_FMT(RETRY).legacy[0][mcs] += peer_stats->retry_bytes;
+		STATS_OP_FMT(RETRY).legacy[1][mcs] += peer_stats->retry_pkts;
+	}
+
+	if (peer_stats->is_ampdu) {
+		tx_stats->ba_fails += peer_stats->ba_fails;
+
+		if (txrate->flags & RATE_INFO_FLAGS_HE_MCS) {
+			STATS_OP_FMT(AMPDU).he[0][mcs] +=
+			peer_stats->succ_bytes + peer_stats->retry_bytes;
+			STATS_OP_FMT(AMPDU).he[1][mcs] +=
+			peer_stats->succ_pkts + peer_stats->retry_pkts;
+		} else if (txrate->flags & RATE_INFO_FLAGS_MCS) {
+			STATS_OP_FMT(AMPDU).ht[0][mcs] +=
+			peer_stats->succ_bytes + peer_stats->retry_bytes;
+			STATS_OP_FMT(AMPDU).ht[1][mcs] +=
+			peer_stats->succ_pkts + peer_stats->retry_pkts;
+		} else {
+			STATS_OP_FMT(AMPDU).vht[0][mcs] +=
+			peer_stats->succ_bytes + peer_stats->retry_bytes;
+			STATS_OP_FMT(AMPDU).vht[1][mcs] +=
+			peer_stats->succ_pkts + peer_stats->retry_pkts;
+		}
+		STATS_OP_FMT(AMPDU).bw[0][bw] +=
+			peer_stats->succ_bytes + peer_stats->retry_bytes;
+		STATS_OP_FMT(AMPDU).nss[0][nss] +=
+			peer_stats->succ_bytes + peer_stats->retry_bytes;
+		STATS_OP_FMT(AMPDU).gi[0][gi] +=
+			peer_stats->succ_bytes + peer_stats->retry_bytes;
+		STATS_OP_FMT(AMPDU).bw[1][bw] +=
+			peer_stats->succ_pkts + peer_stats->retry_pkts;
+		STATS_OP_FMT(AMPDU).nss[1][nss] +=
+			peer_stats->succ_pkts + peer_stats->retry_pkts;
+		STATS_OP_FMT(AMPDU).gi[1][gi] +=
+			peer_stats->succ_pkts + peer_stats->retry_pkts;
+	} else {
+		tx_stats->ack_fails += peer_stats->ba_fails;
+	}
+
+	STATS_OP_FMT(SUCC).bw[0][bw] += peer_stats->succ_bytes;
+	STATS_OP_FMT(SUCC).nss[0][nss] += peer_stats->succ_bytes;
+	STATS_OP_FMT(SUCC).gi[0][gi] += peer_stats->succ_bytes;
+
+	STATS_OP_FMT(SUCC).bw[1][bw] += peer_stats->succ_pkts;
+	STATS_OP_FMT(SUCC).nss[1][nss] += peer_stats->succ_pkts;
+	STATS_OP_FMT(SUCC).gi[1][gi] += peer_stats->succ_pkts;
+
+	STATS_OP_FMT(FAIL).bw[0][bw] += peer_stats->failed_bytes;
+	STATS_OP_FMT(FAIL).nss[0][nss] += peer_stats->failed_bytes;
+	STATS_OP_FMT(FAIL).gi[0][gi] += peer_stats->failed_bytes;
+
+	STATS_OP_FMT(FAIL).bw[1][bw] += peer_stats->failed_pkts;
+	STATS_OP_FMT(FAIL).nss[1][nss] += peer_stats->failed_pkts;
+	STATS_OP_FMT(FAIL).gi[1][gi] += peer_stats->failed_pkts;
+
+	STATS_OP_FMT(RETRY).bw[0][bw] += peer_stats->retry_bytes;
+	STATS_OP_FMT(RETRY).nss[0][nss] += peer_stats->retry_bytes;
+	STATS_OP_FMT(RETRY).gi[0][gi] += peer_stats->retry_bytes;
+
+	STATS_OP_FMT(RETRY).bw[1][bw] += peer_stats->retry_pkts;
+	STATS_OP_FMT(RETRY).nss[1][nss] += peer_stats->retry_pkts;
+	STATS_OP_FMT(RETRY).gi[1][gi] += peer_stats->retry_pkts;
+
+	tx_stats->tx_duration += peer_stats->duration;
+}
+
+void ath11k_update_per_peer_stats_from_txcompl(struct ath11k *ar,
+					       struct sk_buff *msdu,
+					       struct hal_tx_status *ts)
+{
+	struct ath11k_base *ab = ar->ab;
+	struct ath11k_per_peer_tx_stats *peer_stats = &ar->cached_stats;
+	struct ath11k_peer *peer;
+	struct ath11k_sta *arsta;
+	struct ieee80211_sta *sta;
+	u16 rate;
+	u8 rate_idx;
+	int ret;
+
+	rcu_read_lock();
+	spin_lock_bh(&ab->base_lock);
+	peer = ath11k_peer_find_by_id(ab, ts->peer_id);
+	if (!peer || !peer->sta) {
+		ath11k_warn(ab, "failed to find the peer\n");
+		spin_unlock_bh(&ab->base_lock);
+		rcu_read_unlock();
+		return;
+	}
+
+	sta = peer->sta;
+	arsta = (struct ath11k_sta *)sta->drv_priv;
+
+	memset(&arsta->txrate, 0, sizeof(arsta->txrate));
+
+	if (ts->pkt_type == HAL_TX_RATE_STATS_PKT_TYPE_11A ||
+	    ts->pkt_type == HAL_TX_RATE_STATS_PKT_TYPE_11B) {
+		ret = ath11k_mac_hw_ratecode_to_legacy_rate(ts->mcs,
+							    ts->pkt_type,
+							    &rate_idx,
+							    &rate);
+		if (ret < 0) {
+			spin_unlock_bh(&ab->base_lock);
+			rcu_read_unlock();
+			return;
+		}
+		arsta->txrate.legacy = rate;
+	} else if (ts->pkt_type == HAL_TX_RATE_STATS_PKT_TYPE_11N) {
+		if (ts->mcs > 7) {
+			ath11k_warn(ab, "Invalid HT mcs index %d\n", ts->mcs);
+			spin_unlock_bh(&ab->base_lock);
+			rcu_read_unlock();
+			return;
+		}
+
+		arsta->txrate.mcs = ts->mcs + 8 * (arsta->last_txrate.nss - 1);
+		arsta->txrate.flags = RATE_INFO_FLAGS_MCS;
+		if (ts->sgi)
+			arsta->txrate.flags |= RATE_INFO_FLAGS_SHORT_GI;
+	} else if (ts->pkt_type == HAL_TX_RATE_STATS_PKT_TYPE_11AC) {
+		if (ts->mcs > 9) {
+			ath11k_warn(ab, "Invalid VHT mcs index %d\n", ts->mcs);
+			spin_unlock_bh(&ab->base_lock);
+			rcu_read_unlock();
+			return;
+		}
+
+		arsta->txrate.mcs = ts->mcs;
+		arsta->txrate.flags = RATE_INFO_FLAGS_VHT_MCS;
+		if (ts->sgi)
+			arsta->txrate.flags |= RATE_INFO_FLAGS_SHORT_GI;
+	} else {
+		/*TODO: update HE rates */
+	}
+
+	arsta->txrate.nss = arsta->last_txrate.nss;
+	arsta->txrate.bw = ts->bw;
+
+	ath11k_accumulate_per_peer_tx_stats(arsta, peer_stats, rate_idx);
+	spin_unlock_bh(&ab->base_lock);
+	rcu_read_unlock();
+}
+
+static ssize_t ath11k_dbg_sta_dump_tx_stats(struct file *file,
+					    char __user *user_buf,
+					    size_t count, loff_t *ppos)
+{
+	struct ieee80211_sta *sta = file->private_data;
+	struct ath11k_sta *arsta = (struct ath11k_sta *)sta->drv_priv;
+	struct ath11k *ar = arsta->arvif->ar;
+	struct ath11k_htt_data_stats *stats;
+	static const char *str_name[ATH11K_STATS_TYPE_MAX] = {"succ", "fail",
+							      "retry", "ampdu"};
+	static const char *str[ATH11K_COUNTER_TYPE_MAX] = {"bytes", "packets"};
+	int len = 0, i, j, k, retval = 0;
+	const int size = 2 * 4096;
+	char *buf;
+
+	buf = kzalloc(size, GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+
+	mutex_lock(&ar->conf_mutex);
+
+	spin_lock_bh(&ar->data_lock);
+	for (k = 0; k < ATH11K_STATS_TYPE_MAX; k++) {
+		for (j = 0; j < ATH11K_COUNTER_TYPE_MAX; j++) {
+			stats = &arsta->tx_stats->stats[k];
+			len += scnprintf(buf + len, size - len, "%s_%s\n",
+					 str_name[k],
+					 str[j]);
+			len += scnprintf(buf + len, size - len,
+					 " HE MCS %s\n",
+					 str[j]);
+			for (i = 0; i < ATH11K_HE_MCS_NUM; i++)
+				len += scnprintf(buf + len, size - len,
+						 "  %llu ",
+						 stats->he[j][i]);
+			len += scnprintf(buf + len, size - len, "\n");
+			len += scnprintf(buf + len, size - len,
+					 " VHT MCS %s\n",
+					 str[j]);
+			for (i = 0; i < ATH11K_VHT_MCS_NUM; i++)
+				len += scnprintf(buf + len, size - len,
+						 "  %llu ",
+						 stats->vht[j][i]);
+			len += scnprintf(buf + len, size - len, "\n");
+			len += scnprintf(buf + len, size - len, " HT MCS %s\n",
+					 str[j]);
+			for (i = 0; i < ATH11K_HT_MCS_NUM; i++)
+				len += scnprintf(buf + len, size - len,
+						 "  %llu ", stats->ht[j][i]);
+			len += scnprintf(buf + len, size - len, "\n");
+			len += scnprintf(buf + len, size - len,
+					" BW %s (20,40,80,160 MHz)\n", str[j]);
+			len += scnprintf(buf + len, size - len,
+					 "  %llu %llu %llu %llu\n",
+					 stats->bw[j][0], stats->bw[j][1],
+					 stats->bw[j][2], stats->bw[j][3]);
+			len += scnprintf(buf + len, size - len,
+					 " NSS %s (1x1,2x2,3x3,4x4)\n", str[j]);
+			len += scnprintf(buf + len, size - len,
+					 "  %llu %llu %llu %llu\n",
+					 stats->nss[j][0], stats->nss[j][1],
+					 stats->nss[j][2], stats->nss[j][3]);
+			len += scnprintf(buf + len, size - len,
+					 " GI %s (0.4us,0.8us,1.6us,3.2us)\n",
+					 str[j]);
+			len += scnprintf(buf + len, size - len,
+					 "  %llu %llu %llu %llu\n",
+					 stats->gi[j][0], stats->gi[j][1],
+					 stats->gi[j][2], stats->gi[j][3]);
+			len += scnprintf(buf + len, size - len,
+					 " legacy rate %s (1,2 ... Mbps)\n  ",
+					 str[j]);
+			for (i = 0; i < ATH11K_LEGACY_NUM; i++)
+				len += scnprintf(buf + len, size - len, "%llu ",
+						 stats->legacy[j][i]);
+			len += scnprintf(buf + len, size - len, "\n");
+		}
+	}
+
+	len += scnprintf(buf + len, size - len,
+			 "\nTX duration\n %llu usecs\n",
+			 arsta->tx_stats->tx_duration);
+	len += scnprintf(buf + len, size - len,
+			"BA fails\n %llu\n", arsta->tx_stats->ba_fails);
+	len += scnprintf(buf + len, size - len,
+			"ack fails\n %llu\n", arsta->tx_stats->ack_fails);
+	spin_unlock_bh(&ar->data_lock);
+
+	if (len > size)
+		len = size;
+	retval = simple_read_from_buffer(user_buf, count, ppos, buf, len);
+	kfree(buf);
+
+	mutex_unlock(&ar->conf_mutex);
+	return retval;
+}
+
+static const struct file_operations fops_tx_stats = {
+	.read = ath11k_dbg_sta_dump_tx_stats,
+	.open = simple_open,
+	.owner = THIS_MODULE,
+	.llseek = default_llseek,
+};
+
+static ssize_t ath11k_dbg_sta_dump_rx_stats(struct file *file,
+					    char __user *user_buf,
+					    size_t count, loff_t *ppos)
+{
+	struct ieee80211_sta *sta = file->private_data;
+	struct ath11k_sta *arsta = (struct ath11k_sta *)sta->drv_priv;
+	struct ath11k *ar = arsta->arvif->ar;
+	struct ath11k_rx_peer_stats *rx_stats = arsta->rx_stats;
+	int len = 0, i, retval = 0;
+	const int size = 4096;
+	char *buf;
+
+	if (!rx_stats)
+		return -ENOENT;
+
+	buf = kzalloc(size, GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+
+	mutex_lock(&ar->conf_mutex);
+	spin_lock_bh(&ar->ab->base_lock);
+
+	len += scnprintf(buf + len, size - len, "RX peer stats:\n");
+	len += scnprintf(buf + len, size - len, "Num of MSDUs: %llu\n",
+			 rx_stats->num_msdu);
+	len += scnprintf(buf + len, size - len, "Num of MSDUs with TCP L4: %llu\n",
+			 rx_stats->tcp_msdu_count);
+	len += scnprintf(buf + len, size - len, "Num of MSDUs with UDP L4: %llu\n",
+			 rx_stats->udp_msdu_count);
+	len += scnprintf(buf + len, size - len, "Num of MSDUs part of AMPDU: %llu\n",
+			 rx_stats->ampdu_msdu_count);
+	len += scnprintf(buf + len, size - len, "Num of MSDUs not part of AMPDU: %llu\n",
+			 rx_stats->non_ampdu_msdu_count);
+	len += scnprintf(buf + len, size - len, "Num of MSDUs using STBC: %llu\n",
+			 rx_stats->stbc_count);
+	len += scnprintf(buf + len, size - len, "Num of MSDUs beamformed: %llu\n",
+			 rx_stats->beamformed_count);
+	len += scnprintf(buf + len, size - len, "Num of MPDUs with FCS ok: %llu\n",
+			 rx_stats->num_mpdu_fcs_ok);
+	len += scnprintf(buf + len, size - len, "Num of MPDUs with FCS error: %llu\n",
+			 rx_stats->num_mpdu_fcs_err);
+	len += scnprintf(buf + len, size - len,
+			 "GI: 0.8us %llu 0.4us %llu 1.6us %llu 3.2us %llu\n",
+			 rx_stats->gi_count[0], rx_stats->gi_count[1],
+			 rx_stats->gi_count[2], rx_stats->gi_count[3]);
+	len += scnprintf(buf + len, size - len,
+			 "BW: 20Mhz %llu 40Mhz %llu 80Mhz %llu 160Mhz %llu\n",
+			 rx_stats->bw_count[0], rx_stats->bw_count[1],
+			 rx_stats->bw_count[2], rx_stats->bw_count[3]);
+	len += scnprintf(buf + len, size - len, "BCC %llu LDPC %llu\n",
+			 rx_stats->coding_count[0], rx_stats->coding_count[1]);
+	len += scnprintf(buf + len, size - len,
+			 "preamble: 11A %llu 11B %llu 11N %llu 11AC %llu 11AX %llu\n",
+			 rx_stats->pream_cnt[0], rx_stats->pream_cnt[1],
+			 rx_stats->pream_cnt[2], rx_stats->pream_cnt[3],
+			 rx_stats->pream_cnt[4]);
+	len += scnprintf(buf + len, size - len,
+			 "reception type: SU %llu MU_MIMO %llu MU_OFDMA %llu MU_OFDMA_MIMO %llu\n",
+			 rx_stats->reception_type[0], rx_stats->reception_type[1],
+			 rx_stats->reception_type[2], rx_stats->reception_type[3]);
+	len += scnprintf(buf + len, size - len, "TID(0-15) Legacy TID(16):");
+	for (i = 0; i <= IEEE80211_NUM_TIDS; i++)
+		len += scnprintf(buf + len, size - len, "%llu ", rx_stats->tid_count[i]);
+	len += scnprintf(buf + len, size - len, "\nMCS(0-11) Legacy MCS(12):");
+	for (i = 0; i < HAL_RX_MAX_MCS + 1; i++)
+		len += scnprintf(buf + len, size - len, "%llu ", rx_stats->mcs_count[i]);
+	len += scnprintf(buf + len, size - len, "\nNSS(1-8):");
+	for (i = 0; i < HAL_RX_MAX_NSS; i++)
+		len += scnprintf(buf + len, size - len, "%llu ", rx_stats->nss_count[i]);
+	len += scnprintf(buf + len, size - len, "\nRX Duration:%llu ",
+			 rx_stats->rx_duration);
+	len += scnprintf(buf + len, size - len, "\n");
+
+	spin_unlock_bh(&ar->ab->base_lock);
+
+	if (len > size)
+		len = size;
+	retval = simple_read_from_buffer(user_buf, count, ppos, buf, len);
+	kfree(buf);
+
+	mutex_unlock(&ar->conf_mutex);
+	return retval;
+}
+
+static const struct file_operations fops_rx_stats = {
+	.read = ath11k_dbg_sta_dump_rx_stats,
+	.open = simple_open,
+	.owner = THIS_MODULE,
+	.llseek = default_llseek,
+};
+
+static int
+ath11k_dbg_sta_open_htt_peer_stats(struct inode *inode, struct file *file)
+{
+	struct ieee80211_sta *sta = inode->i_private;
+	struct ath11k_sta *arsta = (struct ath11k_sta *)sta->drv_priv;
+	struct ath11k *ar = arsta->arvif->ar;
+	struct debug_htt_stats_req *stats_req;
+	int ret;
+
+	stats_req = vzalloc(sizeof(*stats_req) + ATH11K_HTT_STATS_BUF_SIZE);
+	if (!stats_req)
+		return -ENOMEM;
+
+	mutex_lock(&ar->conf_mutex);
+	ar->debug.htt_stats.stats_req = stats_req;
+	stats_req->type = ATH11K_DBG_HTT_EXT_STATS_PEER_INFO;
+	memcpy(stats_req->peer_addr, sta->addr, ETH_ALEN);
+	ret = ath11k_dbg_htt_stats_req(ar);
+	mutex_unlock(&ar->conf_mutex);
+	if (ret < 0)
+		goto out;
+
+	file->private_data = stats_req;
+	return 0;
+out:
+	vfree(stats_req);
+	return ret;
+}
+
+static int
+ath11k_dbg_sta_release_htt_peer_stats(struct inode *inode, struct file *file)
+{
+	vfree(file->private_data);
+	return 0;
+}
+
+static ssize_t ath11k_dbg_sta_read_htt_peer_stats(struct file *file,
+						  char __user *user_buf,
+						  size_t count, loff_t *ppos)
+{
+	struct debug_htt_stats_req *stats_req = file->private_data;
+	char *buf;
+	u32 length = 0;
+
+	buf = stats_req->buf;
+	length = min_t(u32, stats_req->buf_len, ATH11K_HTT_STATS_BUF_SIZE);
+	return simple_read_from_buffer(user_buf, count, ppos, buf, length);
+}
+
+static const struct file_operations fops_htt_peer_stats = {
+	.open = ath11k_dbg_sta_open_htt_peer_stats,
+	.release = ath11k_dbg_sta_release_htt_peer_stats,
+	.read = ath11k_dbg_sta_read_htt_peer_stats,
+	.owner = THIS_MODULE,
+	.llseek = default_llseek,
+};
+
+static ssize_t ath11k_dbg_sta_write_peer_pktlog(struct file *file,
+						const char __user *buf,
+						size_t count, loff_t *ppos)
+{
+	struct ieee80211_sta *sta = file->private_data;
+	struct ath11k_sta *arsta = (struct ath11k_sta *)sta->drv_priv;
+	struct ath11k *ar = arsta->arvif->ar;
+	int ret, enable;
+
+	mutex_lock(&ar->conf_mutex);
+
+	if (ar->state != ATH11K_STATE_ON) {
+		ret = -ENETDOWN;
+		goto out;
+	}
+
+	ret = kstrtoint_from_user(buf, count, 0, &enable);
+	if (ret)
+		goto out;
+
+	ar->debug.pktlog_peer_valid = enable;
+	memcpy(ar->debug.pktlog_peer_addr, sta->addr, ETH_ALEN);
+
+	/* Send peer based pktlog enable/disable */
+	ret = ath11k_wmi_pdev_peer_pktlog_filter(ar, sta->addr, enable);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to set peer pktlog filter %pM: %d\n",
+			    sta->addr, ret);
+		goto out;
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI, "peer pktlog filter set to %d\n",
+		   enable);
+	ret = count;
+
+out:
+	mutex_unlock(&ar->conf_mutex);
+	return ret;
+}
+
+static ssize_t ath11k_dbg_sta_read_peer_pktlog(struct file *file,
+					       char __user *ubuf,
+					       size_t count, loff_t *ppos)
+{
+	struct ieee80211_sta *sta = file->private_data;
+	struct ath11k_sta *arsta = (struct ath11k_sta *)sta->drv_priv;
+	struct ath11k *ar = arsta->arvif->ar;
+	char buf[32] = {0};
+	int len;
+
+	mutex_lock(&ar->conf_mutex);
+	len = scnprintf(buf, sizeof(buf), "%08x %pM\n",
+			ar->debug.pktlog_peer_valid,
+			ar->debug.pktlog_peer_addr);
+	mutex_unlock(&ar->conf_mutex);
+
+	return simple_read_from_buffer(ubuf, count, ppos, buf, len);
+}
+
+static const struct file_operations fops_peer_pktlog = {
+	.write = ath11k_dbg_sta_write_peer_pktlog,
+	.read = ath11k_dbg_sta_read_peer_pktlog,
+	.open = simple_open,
+	.owner = THIS_MODULE,
+	.llseek = default_llseek,
+};
+
+void ath11k_sta_add_debugfs(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+			    struct ieee80211_sta *sta, struct dentry *dir)
+{
+	struct ath11k *ar = hw->priv;
+
+	if (ath11k_debug_is_extd_tx_stats_enabled(ar))
+		debugfs_create_file("tx_stats", 0400, dir, sta,
+				    &fops_tx_stats);
+	if (ath11k_debug_is_extd_rx_stats_enabled(ar))
+		debugfs_create_file("rx_stats", 0400, dir, sta,
+				    &fops_rx_stats);
+
+	debugfs_create_file("htt_peer_stats", 0400, dir, sta,
+			    &fops_htt_peer_stats);
+
+	debugfs_create_file("peer_pktlog", 0644, dir, sta,
+			    &fops_peer_pktlog);
+}
diff --git a/drivers/net/wireless/ath/ath11k/dp.c b/drivers/net/wireless/ath/ath11k/dp.c
new file mode 100644
index 000000000000..aa2b28189cda
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/dp.c
@@ -0,0 +1,898 @@
+// SPDX-License-Identifier: BSD-3-Clause-Clear
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#include <linux/kfifo.h>
+#include "core.h"
+#include "dp_tx.h"
+#include "hal_tx.h"
+#include "debug.h"
+#include "dp_rx.h"
+#include "peer.h"
+
+static void ath11k_dp_htt_htc_tx_complete(struct ath11k_base *ab,
+					  struct sk_buff *skb)
+{
+	dev_kfree_skb_any(skb);
+}
+
+void ath11k_dp_peer_cleanup(struct ath11k *ar, int vdev_id, const u8 *addr)
+{
+	struct ath11k_base *ab = ar->ab;
+	struct ath11k_peer *peer;
+
+	/* TODO: Any other peer specific DP cleanup */
+
+	spin_lock_bh(&ab->base_lock);
+	peer = ath11k_peer_find(ab, vdev_id, addr);
+	if (!peer) {
+		ath11k_warn(ab, "failed to lookup peer %pM on vdev %d\n",
+			    addr, vdev_id);
+		spin_unlock_bh(&ab->base_lock);
+		return;
+	}
+
+	ath11k_peer_rx_tid_cleanup(ar, peer);
+	spin_unlock_bh(&ab->base_lock);
+}
+
+int ath11k_dp_peer_setup(struct ath11k *ar, int vdev_id, const u8 *addr)
+{
+	struct ath11k_base *ab = ar->ab;
+	u32 reo_dest;
+	int ret;
+
+	/* NOTE: reo_dest ring id starts from 1 unlike mac_id which starts from 0 */
+	reo_dest = ar->dp.mac_id + 1;
+	ret = ath11k_wmi_set_peer_param(ar, addr, vdev_id,
+					WMI_PEER_SET_DEFAULT_ROUTING,
+					DP_RX_HASH_ENABLE | (reo_dest << 1));
+
+	if (ret) {
+		ath11k_warn(ab, "failed to set default routing %d peer :%pM vdev_id :%d\n",
+			    ret, addr, vdev_id);
+		return ret;
+	}
+
+	ret = ath11k_peer_rx_tid_setup(ar, addr, vdev_id,
+				       HAL_DESC_REO_NON_QOS_TID, 1, 0);
+	if (ret) {
+		ath11k_warn(ab, "failed to setup rxd tid queue for non-qos tid %d\n",
+			    ret);
+		return ret;
+	}
+
+	ret = ath11k_peer_rx_tid_setup(ar, addr, vdev_id, 0, 1, 0);
+	if (ret) {
+		ath11k_warn(ab, "failed to setup rxd tid queue for tid 0 %d\n",
+			    ret);
+		return ret;
+	}
+
+	/* TODO: Setup other peer specific resource used in data path */
+
+	return 0;
+}
+
+void ath11k_dp_srng_cleanup(struct ath11k_base *ab, struct dp_srng *ring)
+{
+	if (!ring->vaddr_unaligned)
+		return;
+
+	dma_free_coherent(ab->dev, ring->size, ring->vaddr_unaligned,
+			  ring->paddr_unaligned);
+
+	ring->vaddr_unaligned = NULL;
+}
+
+int ath11k_dp_srng_setup(struct ath11k_base *ab, struct dp_srng *ring,
+			 enum hal_ring_type type, int ring_num,
+			 int mac_id, int num_entries)
+{
+	struct hal_srng_params params = { 0 };
+	int entry_sz = ath11k_hal_srng_get_entrysize(type);
+	int max_entries = ath11k_hal_srng_get_max_entries(type);
+	int ret;
+
+	if (max_entries < 0 || entry_sz < 0)
+		return -EINVAL;
+
+	if (num_entries > max_entries)
+		num_entries = max_entries;
+
+	ring->size = (num_entries * entry_sz) + HAL_RING_BASE_ALIGN - 1;
+	ring->vaddr_unaligned = dma_alloc_coherent(ab->dev, ring->size,
+						   &ring->paddr_unaligned,
+						   GFP_KERNEL);
+	if (!ring->vaddr_unaligned)
+		return -ENOMEM;
+
+	ring->vaddr = PTR_ALIGN(ring->vaddr_unaligned, HAL_RING_BASE_ALIGN);
+	ring->paddr = ring->paddr_unaligned + ((unsigned long)ring->vaddr -
+		      (unsigned long)ring->vaddr_unaligned);
+
+	params.ring_base_vaddr = ring->vaddr;
+	params.ring_base_paddr = ring->paddr;
+	params.num_entries = num_entries;
+
+	switch (type) {
+	case HAL_REO_DST:
+		params.intr_batch_cntr_thres_entries =
+					HAL_SRNG_INT_BATCH_THRESHOLD_RX;
+		params.intr_timer_thres_us = HAL_SRNG_INT_TIMER_THRESHOLD_RX;
+		break;
+	case HAL_RXDMA_BUF:
+	case HAL_RXDMA_MONITOR_BUF:
+	case HAL_RXDMA_MONITOR_STATUS:
+		params.low_threshold = num_entries >> 3;
+		params.flags |= HAL_SRNG_FLAGS_LOW_THRESH_INTR_EN;
+		params.intr_batch_cntr_thres_entries = 0;
+		params.intr_timer_thres_us = HAL_SRNG_INT_TIMER_THRESHOLD_RX;
+		break;
+	case HAL_WBM2SW_RELEASE:
+		if (ring_num < 3) {
+			params.intr_batch_cntr_thres_entries =
+					HAL_SRNG_INT_BATCH_THRESHOLD_TX;
+			params.intr_timer_thres_us =
+					HAL_SRNG_INT_TIMER_THRESHOLD_TX;
+			break;
+		}
+		/* follow through when ring_num >= 3 */
+		/* fall through */
+	case HAL_REO_EXCEPTION:
+	case HAL_REO_REINJECT:
+	case HAL_REO_CMD:
+	case HAL_REO_STATUS:
+	case HAL_TCL_DATA:
+	case HAL_TCL_CMD:
+	case HAL_TCL_STATUS:
+	case HAL_WBM_IDLE_LINK:
+	case HAL_SW2WBM_RELEASE:
+	case HAL_RXDMA_DST:
+	case HAL_RXDMA_MONITOR_DST:
+	case HAL_RXDMA_MONITOR_DESC:
+	case HAL_RXDMA_DIR_BUF:
+		params.intr_batch_cntr_thres_entries =
+					HAL_SRNG_INT_BATCH_THRESHOLD_OTHER;
+		params.intr_timer_thres_us = HAL_SRNG_INT_TIMER_THRESHOLD_OTHER;
+		break;
+	default:
+		ath11k_warn(ab, "Not a valid ring type in dp :%d\n", type);
+		return -EINVAL;
+	}
+
+	ret = ath11k_hal_srng_setup(ab, type, ring_num, mac_id, &params);
+	if (ret < 0) {
+		ath11k_warn(ab, "failed to setup srng: %d ring_id %d\n",
+			    ret, ring_num);
+		return ret;
+	}
+
+	ring->ring_id = ret;
+
+	return 0;
+}
+
+static void ath11k_dp_srng_common_cleanup(struct ath11k_base *ab)
+{
+	struct ath11k_dp *dp = &ab->dp;
+	int i;
+
+	ath11k_dp_srng_cleanup(ab, &dp->wbm_desc_rel_ring);
+	ath11k_dp_srng_cleanup(ab, &dp->tcl_cmd_ring);
+	ath11k_dp_srng_cleanup(ab, &dp->tcl_status_ring);
+	for (i = 0; i < DP_TCL_NUM_RING_MAX; i++) {
+		ath11k_dp_srng_cleanup(ab, &dp->tx_ring[i].tcl_data_ring);
+		ath11k_dp_srng_cleanup(ab, &dp->tx_ring[i].tcl_comp_ring);
+	}
+	ath11k_dp_srng_cleanup(ab, &dp->reo_reinject_ring);
+	ath11k_dp_srng_cleanup(ab, &dp->rx_rel_ring);
+	ath11k_dp_srng_cleanup(ab, &dp->reo_except_ring);
+	ath11k_dp_srng_cleanup(ab, &dp->reo_cmd_ring);
+	ath11k_dp_srng_cleanup(ab, &dp->reo_status_ring);
+}
+
+static int ath11k_dp_srng_common_setup(struct ath11k_base *ab)
+{
+	struct ath11k_dp *dp = &ab->dp;
+	struct hal_srng *srng;
+	int i, ret;
+
+	ret = ath11k_dp_srng_setup(ab, &dp->wbm_desc_rel_ring,
+				   HAL_SW2WBM_RELEASE, 0, 0,
+				   DP_WBM_RELEASE_RING_SIZE);
+	if (ret) {
+		ath11k_warn(ab, "failed to set up wbm2sw_release ring :%d\n",
+			    ret);
+		goto err;
+	}
+
+	ret = ath11k_dp_srng_setup(ab, &dp->tcl_cmd_ring, HAL_TCL_CMD, 0, 0,
+				   DP_TCL_CMD_RING_SIZE);
+	if (ret) {
+		ath11k_warn(ab, "failed to set up tcl_cmd ring :%d\n", ret);
+		goto err;
+	}
+
+	ret = ath11k_dp_srng_setup(ab, &dp->tcl_status_ring, HAL_TCL_STATUS,
+				   0, 0, DP_TCL_STATUS_RING_SIZE);
+	if (ret) {
+		ath11k_warn(ab, "failed to set up tcl_status ring :%d\n", ret);
+		goto err;
+	}
+
+	for (i = 0; i < DP_TCL_NUM_RING_MAX; i++) {
+		ret = ath11k_dp_srng_setup(ab, &dp->tx_ring[i].tcl_data_ring,
+					   HAL_TCL_DATA, i, 0,
+					   DP_TCL_DATA_RING_SIZE);
+		if (ret) {
+			ath11k_warn(ab, "failed to set up tcl_data ring (%d) :%d\n",
+				    i, ret);
+			goto err;
+		}
+
+		ret = ath11k_dp_srng_setup(ab, &dp->tx_ring[i].tcl_comp_ring,
+					   HAL_WBM2SW_RELEASE, i, 0,
+					   DP_TX_COMP_RING_SIZE);
+		if (ret) {
+			ath11k_warn(ab, "failed to set up tcl_comp ring ring (%d) :%d\n",
+				    i, ret);
+			goto err;
+		}
+
+		srng = &ab->hal.srng_list[dp->tx_ring[i].tcl_data_ring.ring_id];
+		ath11k_hal_tx_init_data_ring(ab, srng);
+	}
+
+	ret = ath11k_dp_srng_setup(ab, &dp->reo_reinject_ring, HAL_REO_REINJECT,
+				   0, 0, DP_REO_REINJECT_RING_SIZE);
+	if (ret) {
+		ath11k_warn(ab, "failed to set up reo_reinject ring :%d\n",
+			    ret);
+		goto err;
+	}
+
+	ret = ath11k_dp_srng_setup(ab, &dp->rx_rel_ring, HAL_WBM2SW_RELEASE,
+				   3, 0, DP_RX_RELEASE_RING_SIZE);
+	if (ret) {
+		ath11k_warn(ab, "failed to set up rx_rel ring :%d\n", ret);
+		goto err;
+	}
+
+	ret = ath11k_dp_srng_setup(ab, &dp->reo_except_ring, HAL_REO_EXCEPTION,
+				   0, 0, DP_REO_EXCEPTION_RING_SIZE);
+	if (ret) {
+		ath11k_warn(ab, "failed to set up reo_exception ring :%d\n",
+			    ret);
+		goto err;
+	}
+
+	ret = ath11k_dp_srng_setup(ab, &dp->reo_cmd_ring, HAL_REO_CMD,
+				   0, 0, DP_REO_CMD_RING_SIZE);
+	if (ret) {
+		ath11k_warn(ab, "failed to set up reo_cmd ring :%d\n", ret);
+		goto err;
+	}
+
+	srng = &ab->hal.srng_list[dp->reo_cmd_ring.ring_id];
+	ath11k_hal_reo_init_cmd_ring(ab, srng);
+
+	ret = ath11k_dp_srng_setup(ab, &dp->reo_status_ring, HAL_REO_STATUS,
+				   0, 0, DP_REO_STATUS_RING_SIZE);
+	if (ret) {
+		ath11k_warn(ab, "failed to set up reo_status ring :%d\n", ret);
+		goto err;
+	}
+
+	ath11k_hal_reo_hw_setup(ab);
+
+	return 0;
+
+err:
+	ath11k_dp_srng_common_cleanup(ab);
+
+	return ret;
+}
+
+static void ath11k_dp_scatter_idle_link_desc_cleanup(struct ath11k_base *ab)
+{
+	struct ath11k_dp *dp = &ab->dp;
+	struct hal_wbm_idle_scatter_list *slist = dp->scatter_list;
+	int i;
+
+	for (i = 0; i < DP_IDLE_SCATTER_BUFS_MAX; i++) {
+		if (!slist[i].vaddr)
+			continue;
+
+		dma_free_coherent(ab->dev, HAL_WBM_IDLE_SCATTER_BUF_SIZE_MAX,
+				  slist[i].vaddr, slist[i].paddr);
+		slist[i].vaddr = NULL;
+	}
+}
+
+static int ath11k_dp_scatter_idle_link_desc_setup(struct ath11k_base *ab,
+						  int size,
+						  u32 n_link_desc_bank,
+						  u32 n_link_desc,
+						  u32 last_bank_sz)
+{
+	struct ath11k_dp *dp = &ab->dp;
+	struct dp_link_desc_bank *link_desc_banks = dp->link_desc_banks;
+	struct hal_wbm_idle_scatter_list *slist = dp->scatter_list;
+	u32 n_entries_per_buf;
+	int num_scatter_buf, scatter_idx;
+	struct hal_wbm_link_desc *scatter_buf;
+	int align_bytes, n_entries;
+	dma_addr_t paddr;
+	int rem_entries;
+	int i;
+	int ret = 0;
+	u32 end_offset;
+
+	n_entries_per_buf = HAL_WBM_IDLE_SCATTER_BUF_SIZE /
+			    ath11k_hal_srng_get_entrysize(HAL_WBM_IDLE_LINK);
+	num_scatter_buf = DIV_ROUND_UP(size, HAL_WBM_IDLE_SCATTER_BUF_SIZE);
+
+	if (num_scatter_buf > DP_IDLE_SCATTER_BUFS_MAX)
+		return -EINVAL;
+
+	for (i = 0; i < num_scatter_buf; i++) {
+		slist[i].vaddr = dma_alloc_coherent(ab->dev,
+						    HAL_WBM_IDLE_SCATTER_BUF_SIZE_MAX,
+						    &slist[i].paddr, GFP_KERNEL);
+		if (!slist[i].vaddr) {
+			ret = -ENOMEM;
+			goto err;
+		}
+	}
+
+	scatter_idx = 0;
+	scatter_buf = slist[scatter_idx].vaddr;
+	rem_entries = n_entries_per_buf;
+
+	for (i = 0; i < n_link_desc_bank; i++) {
+		align_bytes = link_desc_banks[i].vaddr -
+			      link_desc_banks[i].vaddr_unaligned;
+		n_entries = (DP_LINK_DESC_ALLOC_SIZE_THRESH - align_bytes) /
+			     HAL_LINK_DESC_SIZE;
+		paddr = link_desc_banks[i].paddr;
+		while (n_entries) {
+			ath11k_hal_set_link_desc_addr(scatter_buf, i, paddr);
+			n_entries--;
+			paddr += HAL_LINK_DESC_SIZE;
+			if (rem_entries) {
+				rem_entries--;
+				scatter_buf++;
+				continue;
+			}
+
+			rem_entries = n_entries_per_buf;
+			scatter_idx++;
+			scatter_buf = slist[scatter_idx].vaddr;
+		}
+	}
+
+	end_offset = (scatter_buf - slist[scatter_idx].vaddr) *
+		     sizeof(struct hal_wbm_link_desc);
+	ath11k_hal_setup_link_idle_list(ab, slist, num_scatter_buf,
+					n_link_desc, end_offset);
+
+	return 0;
+
+err:
+	ath11k_dp_scatter_idle_link_desc_cleanup(ab);
+
+	return ret;
+}
+
+static void
+ath11k_dp_link_desc_bank_free(struct ath11k_base *ab,
+			      struct dp_link_desc_bank *link_desc_banks)
+{
+	int i;
+
+	for (i = 0; i < DP_LINK_DESC_BANKS_MAX; i++) {
+		if (link_desc_banks[i].vaddr_unaligned) {
+			dma_free_coherent(ab->dev,
+					  link_desc_banks[i].size,
+					  link_desc_banks[i].vaddr_unaligned,
+					  link_desc_banks[i].paddr_unaligned);
+			link_desc_banks[i].vaddr_unaligned = NULL;
+		}
+	}
+}
+
+static int ath11k_dp_link_desc_bank_alloc(struct ath11k_base *ab,
+					  struct dp_link_desc_bank *desc_bank,
+					  int n_link_desc_bank,
+					  int last_bank_sz)
+{
+	struct ath11k_dp *dp = &ab->dp;
+	int i;
+	int ret = 0;
+	int desc_sz = DP_LINK_DESC_ALLOC_SIZE_THRESH;
+
+	for (i = 0; i < n_link_desc_bank; i++) {
+		if (i == (n_link_desc_bank - 1) && last_bank_sz)
+			desc_sz = last_bank_sz;
+
+		desc_bank[i].vaddr_unaligned =
+					dma_alloc_coherent(ab->dev, desc_sz,
+							   &desc_bank[i].paddr_unaligned,
+							   GFP_KERNEL);
+		if (!desc_bank[i].vaddr_unaligned) {
+			ret = -ENOMEM;
+			goto err;
+		}
+
+		desc_bank[i].vaddr = PTR_ALIGN(desc_bank[i].vaddr_unaligned,
+					       HAL_LINK_DESC_ALIGN);
+		desc_bank[i].paddr = desc_bank[i].paddr_unaligned +
+				     ((unsigned long)desc_bank[i].vaddr -
+				      (unsigned long)desc_bank[i].vaddr_unaligned);
+		desc_bank[i].size = desc_sz;
+	}
+
+	return 0;
+
+err:
+	ath11k_dp_link_desc_bank_free(ab, dp->link_desc_banks);
+
+	return ret;
+}
+
+void ath11k_dp_link_desc_cleanup(struct ath11k_base *ab,
+				 struct dp_link_desc_bank *desc_bank,
+				 u32 ring_type, struct dp_srng *ring)
+{
+	ath11k_dp_link_desc_bank_free(ab, desc_bank);
+
+	if (ring_type != HAL_RXDMA_MONITOR_DESC) {
+		ath11k_dp_srng_cleanup(ab, ring);
+		ath11k_dp_scatter_idle_link_desc_cleanup(ab);
+	}
+}
+
+static int ath11k_wbm_idle_ring_setup(struct ath11k_base *ab, u32 *n_link_desc)
+{
+	struct ath11k_dp *dp = &ab->dp;
+	u32 n_mpdu_link_desc, n_mpdu_queue_desc;
+	u32 n_tx_msdu_link_desc, n_rx_msdu_link_desc;
+	int ret = 0;
+
+	n_mpdu_link_desc = (DP_NUM_TIDS_MAX * DP_AVG_MPDUS_PER_TID_MAX) /
+			   HAL_NUM_MPDUS_PER_LINK_DESC;
+
+	n_mpdu_queue_desc = n_mpdu_link_desc /
+			    HAL_NUM_MPDU_LINKS_PER_QUEUE_DESC;
+
+	n_tx_msdu_link_desc = (DP_NUM_TIDS_MAX * DP_AVG_FLOWS_PER_TID *
+			       DP_AVG_MSDUS_PER_FLOW) /
+			      HAL_NUM_TX_MSDUS_PER_LINK_DESC;
+
+	n_rx_msdu_link_desc = (DP_NUM_TIDS_MAX * DP_AVG_MPDUS_PER_TID_MAX *
+			       DP_AVG_MSDUS_PER_MPDU) /
+			      HAL_NUM_RX_MSDUS_PER_LINK_DESC;
+
+	*n_link_desc = n_mpdu_link_desc + n_mpdu_queue_desc +
+		      n_tx_msdu_link_desc + n_rx_msdu_link_desc;
+
+	ret = ath11k_dp_srng_setup(ab, &dp->wbm_idle_ring,
+				   HAL_WBM_IDLE_LINK, 0, 0, *n_link_desc);
+	if (ret) {
+		ath11k_warn(ab, "failed to setup wbm_idle_ring: %d\n", ret);
+		return ret;
+	}
+	return ret;
+}
+
+int ath11k_dp_link_desc_setup(struct ath11k_base *ab,
+			      struct dp_link_desc_bank *link_desc_banks,
+			      u32 ring_type, struct hal_srng *srng,
+			      u32 n_link_desc)
+{
+	u32 tot_mem_sz;
+	u32 n_link_desc_bank, last_bank_sz;
+	u32 entry_sz, align_bytes, n_entries;
+	u32 paddr;
+	u32 *desc;
+	int i, ret;
+
+	if (n_link_desc & (n_link_desc - 1))
+		n_link_desc = 1 << fls(n_link_desc);
+
+	tot_mem_sz = n_link_desc * HAL_LINK_DESC_SIZE;
+	tot_mem_sz += HAL_LINK_DESC_ALIGN;
+
+	if (tot_mem_sz <= DP_LINK_DESC_ALLOC_SIZE_THRESH) {
+		n_link_desc_bank = 1;
+		last_bank_sz = tot_mem_sz;
+	} else {
+		n_link_desc_bank = tot_mem_sz /
+				   (DP_LINK_DESC_ALLOC_SIZE_THRESH -
+				    HAL_LINK_DESC_ALIGN);
+		last_bank_sz = tot_mem_sz %
+			       (DP_LINK_DESC_ALLOC_SIZE_THRESH -
+				HAL_LINK_DESC_ALIGN);
+
+		if (last_bank_sz)
+			n_link_desc_bank += 1;
+	}
+
+	if (n_link_desc_bank > DP_LINK_DESC_BANKS_MAX)
+		return -EINVAL;
+
+	ret = ath11k_dp_link_desc_bank_alloc(ab, link_desc_banks,
+					     n_link_desc_bank, last_bank_sz);
+	if (ret)
+		return ret;
+
+	/* Setup link desc idle list for HW internal usage */
+	entry_sz = ath11k_hal_srng_get_entrysize(ring_type);
+	tot_mem_sz = entry_sz * n_link_desc;
+
+	/* Setup scatter desc list when the total memory requirement is more */
+	if (tot_mem_sz > DP_LINK_DESC_ALLOC_SIZE_THRESH &&
+	    ring_type != HAL_RXDMA_MONITOR_DESC) {
+		ret = ath11k_dp_scatter_idle_link_desc_setup(ab, tot_mem_sz,
+							     n_link_desc_bank,
+							     n_link_desc,
+							     last_bank_sz);
+		if (ret) {
+			ath11k_warn(ab, "failed to setup scatting idle list descriptor :%d\n",
+				    ret);
+			goto fail_desc_bank_free;
+		}
+
+		return 0;
+	}
+
+	spin_lock_bh(&srng->lock);
+
+	ath11k_hal_srng_access_begin(ab, srng);
+
+	for (i = 0; i < n_link_desc_bank; i++) {
+		align_bytes = link_desc_banks[i].vaddr -
+			      link_desc_banks[i].vaddr_unaligned;
+		n_entries = (link_desc_banks[i].size - align_bytes) /
+			    HAL_LINK_DESC_SIZE;
+		paddr = link_desc_banks[i].paddr;
+		while (n_entries &&
+		       (desc = ath11k_hal_srng_src_get_next_entry(ab, srng))) {
+			ath11k_hal_set_link_desc_addr((struct hal_wbm_link_desc *)desc,
+						      i, paddr);
+			n_entries--;
+			paddr += HAL_LINK_DESC_SIZE;
+		}
+	}
+
+	ath11k_hal_srng_access_end(ab, srng);
+
+	spin_unlock_bh(&srng->lock);
+
+	return 0;
+
+fail_desc_bank_free:
+	ath11k_dp_link_desc_bank_free(ab, link_desc_banks);
+
+	return ret;
+}
+
+int ath11k_dp_service_srng(struct ath11k_base *ab,
+			   struct ath11k_ext_irq_grp *irq_grp,
+			   int budget)
+{
+	struct napi_struct *napi = &irq_grp->napi;
+	int grp_id = irq_grp->grp_id;
+	int work_done = 0;
+	int i = 0;
+	int tot_work_done = 0;
+
+	while (ath11k_tx_ring_mask[grp_id] >> i) {
+		if (ath11k_tx_ring_mask[grp_id] & BIT(i))
+			ath11k_dp_tx_completion_handler(ab, i);
+		i++;
+	}
+
+	if (ath11k_rx_err_ring_mask[grp_id]) {
+		work_done = ath11k_dp_process_rx_err(ab, napi, budget);
+		budget -= work_done;
+		tot_work_done += work_done;
+		if (budget <= 0)
+			goto done;
+	}
+
+	if (ath11k_rx_wbm_rel_ring_mask[grp_id]) {
+		work_done = ath11k_dp_rx_process_wbm_err(ab,
+							 napi,
+							 budget);
+		budget -= work_done;
+		tot_work_done += work_done;
+
+		if (budget <= 0)
+			goto done;
+	}
+
+	if (ath11k_rx_ring_mask[grp_id]) {
+		for (i = 0; i <  ab->num_radios; i++) {
+			if (ath11k_rx_ring_mask[grp_id] & BIT(i)) {
+				work_done = ath11k_dp_process_rx(ab, i, napi,
+								 &irq_grp->pending_q,
+								 budget);
+				budget -= work_done;
+				tot_work_done += work_done;
+			}
+			if (budget <= 0)
+				goto done;
+		}
+	}
+
+	if (rx_mon_status_ring_mask[grp_id]) {
+		for (i = 0; i <  ab->num_radios; i++) {
+			if (rx_mon_status_ring_mask[grp_id] & BIT(i)) {
+				work_done =
+				ath11k_dp_rx_process_mon_rings(ab,
+							       i, napi,
+							       budget);
+				budget -= work_done;
+				tot_work_done += work_done;
+			}
+			if (budget <= 0)
+				goto done;
+		}
+	}
+
+	if (ath11k_reo_status_ring_mask[grp_id])
+		ath11k_dp_process_reo_status(ab);
+
+	for (i = 0; i < ab->num_radios; i++) {
+		if (ath11k_rxdma2host_ring_mask[grp_id] & BIT(i)) {
+			work_done = ath11k_dp_process_rxdma_err(ab, i, budget);
+			budget -= work_done;
+			tot_work_done += work_done;
+		}
+
+		if (budget <= 0)
+			goto done;
+
+		if (ath11k_host2rxdma_ring_mask[grp_id] & BIT(i)) {
+			struct ath11k_pdev_dp *dp = &ab->pdevs[i].ar->dp;
+			struct dp_rxdma_ring *rx_ring = &dp->rx_refill_buf_ring;
+
+			ath11k_dp_rxbufs_replenish(ab, i, rx_ring, 0,
+						   HAL_RX_BUF_RBM_SW3_BM,
+						   GFP_ATOMIC);
+		}
+	}
+	/* TODO: Implement handler for other interrupts */
+
+done:
+	return tot_work_done;
+}
+
+void ath11k_dp_pdev_free(struct ath11k_base *ab)
+{
+	struct ath11k *ar;
+	int i;
+
+	for (i = 0; i < ab->num_radios; i++) {
+		ar = ab->pdevs[i].ar;
+		ath11k_dp_rx_pdev_free(ab, i);
+		ath11k_debug_unregister(ar);
+		ath11k_dp_rx_pdev_mon_detach(ar);
+	}
+}
+
+int ath11k_dp_pdev_alloc(struct ath11k_base *ab)
+{
+	struct ath11k *ar;
+	struct ath11k_pdev_dp *dp;
+	int ret;
+	int i;
+
+	for (i = 0; i <  ab->num_radios; i++) {
+		ar = ab->pdevs[i].ar;
+		dp = &ar->dp;
+		dp->mac_id = i;
+		idr_init(&dp->rx_refill_buf_ring.bufs_idr);
+		spin_lock_init(&dp->rx_refill_buf_ring.idr_lock);
+		atomic_set(&dp->num_tx_pending, 0);
+		init_waitqueue_head(&dp->tx_empty_waitq);
+		idr_init(&dp->rx_mon_status_refill_ring.bufs_idr);
+		spin_lock_init(&dp->rx_mon_status_refill_ring.idr_lock);
+		idr_init(&dp->rxdma_mon_buf_ring.bufs_idr);
+		spin_lock_init(&dp->rxdma_mon_buf_ring.idr_lock);
+	}
+
+	/* TODO:Per-pdev rx ring unlike tx ring which is mapped to different AC's */
+	for (i = 0; i < ab->num_radios; i++) {
+		ar = ab->pdevs[i].ar;
+		ret = ath11k_dp_rx_pdev_alloc(ab, i);
+		if (ret) {
+			ath11k_warn(ab, "failed to allocate pdev rx for pdev_id :%d\n",
+				    i);
+			goto err;
+		}
+		ret = ath11k_dp_rx_pdev_mon_attach(ar);
+		if (ret) {
+			ath11k_warn(ab, "failed to initialize mon pdev %d\n",
+				    i);
+			goto err;
+		}
+	}
+
+	return 0;
+
+err:
+	ath11k_dp_pdev_free(ab);
+
+	return ret;
+}
+
+int ath11k_dp_htt_connect(struct ath11k_dp *dp)
+{
+	struct ath11k_htc_svc_conn_req conn_req;
+	struct ath11k_htc_svc_conn_resp conn_resp;
+	int status;
+
+	memset(&conn_req, 0, sizeof(conn_req));
+	memset(&conn_resp, 0, sizeof(conn_resp));
+
+	conn_req.ep_ops.ep_tx_complete = ath11k_dp_htt_htc_tx_complete;
+	conn_req.ep_ops.ep_rx_complete = ath11k_dp_htt_htc_t2h_msg_handler;
+
+	/* connect to control service */
+	conn_req.service_id = ATH11K_HTC_SVC_ID_HTT_DATA_MSG;
+
+	status = ath11k_htc_connect_service(&dp->ab->htc, &conn_req,
+					    &conn_resp);
+
+	if (status)
+		return status;
+
+	dp->eid = conn_resp.eid;
+
+	return 0;
+}
+
+static void ath11k_dp_update_vdev_search(struct ath11k_vif *arvif)
+{
+	/* Enable AddrY (SA based search) for STA mode. All other modes it
+	 * is going to be AddrX (DA based search). For STA mode, set search
+	 * type based on AST value.
+	 */
+	switch (arvif->vdev_type) {
+	case WMI_VDEV_TYPE_STA:
+		arvif->hal_addr_search_flags = HAL_TX_ADDRY_EN;
+		arvif->search_type = HAL_TX_ADDR_SEARCH_INDEX;
+		break;
+	case WMI_VDEV_TYPE_AP:
+	case WMI_VDEV_TYPE_IBSS:
+		arvif->hal_addr_search_flags = HAL_TX_ADDRX_EN;
+		arvif->search_type = HAL_TX_ADDR_SEARCH_DEFAULT;
+		break;
+	case WMI_VDEV_TYPE_MONITOR:
+	default:
+		return;
+	}
+}
+
+void ath11k_dp_vdev_tx_attach(struct ath11k *ar, struct ath11k_vif *arvif)
+{
+	arvif->tcl_metadata |= FIELD_PREP(HTT_TCL_META_DATA_TYPE, 1) |
+			       FIELD_PREP(HTT_TCL_META_DATA_VDEV_ID,
+					  arvif->vdev_id) |
+			       FIELD_PREP(HTT_TCL_META_DATA_PDEV_ID,
+					  ar->pdev->pdev_id);
+
+	/* set HTT extension valid bit to 0 by default */
+	arvif->tcl_metadata &= ~HTT_TCL_META_DATA_VALID_HTT;
+
+	ath11k_dp_update_vdev_search(arvif);
+}
+
+static int ath11k_dp_tx_pending_cleanup(int buf_id, void *skb, void *ctx)
+{
+	struct ath11k_base *ab = (struct ath11k_base *)ctx;
+	struct sk_buff *msdu = skb;
+
+	dma_unmap_single(ab->dev, ATH11K_SKB_CB(msdu)->paddr, msdu->len,
+			 DMA_TO_DEVICE);
+
+	dev_kfree_skb_any(msdu);
+
+	return 0;
+}
+
+void ath11k_dp_free(struct ath11k_base *ab)
+{
+	struct ath11k_dp *dp = &ab->dp;
+	int i;
+
+	ath11k_dp_link_desc_cleanup(ab, dp->link_desc_banks,
+				    HAL_WBM_IDLE_LINK, &dp->wbm_idle_ring);
+
+	ath11k_dp_srng_common_cleanup(ab);
+
+	ath11k_dp_reo_cmd_list_cleanup(ab);
+
+	for (i = 0; i < DP_TCL_NUM_RING_MAX; i++) {
+		spin_lock_bh(&dp->tx_ring[i].tx_idr_lock);
+		idr_for_each(&dp->tx_ring[i].txbuf_idr,
+			     ath11k_dp_tx_pending_cleanup, ab);
+		idr_destroy(&dp->tx_ring[i].txbuf_idr);
+		spin_unlock_bh(&dp->tx_ring[i].tx_idr_lock);
+
+		spin_lock_bh(&dp->tx_ring[i].tx_status_lock);
+		kfifo_free(&dp->tx_ring[i].tx_status_fifo);
+		spin_unlock_bh(&dp->tx_ring[i].tx_status_lock);
+	}
+
+	/* Deinit any SOC level resource */
+}
+
+int ath11k_dp_alloc(struct ath11k_base *ab)
+{
+	struct ath11k_dp *dp = &ab->dp;
+	struct hal_srng *srng = NULL;
+	size_t size = 0;
+	u32 n_link_desc = 0;
+	int ret;
+	int i;
+
+	dp->ab = ab;
+
+	INIT_LIST_HEAD(&dp->reo_cmd_list);
+	INIT_LIST_HEAD(&dp->reo_cmd_cache_flush_list);
+	spin_lock_init(&dp->reo_cmd_lock);
+
+	ret = ath11k_wbm_idle_ring_setup(ab, &n_link_desc);
+	if (ret) {
+		ath11k_warn(ab, "failed to setup wbm_idle_ring: %d\n", ret);
+		return ret;
+	}
+
+	srng = &ab->hal.srng_list[dp->wbm_idle_ring.ring_id];
+
+	ret = ath11k_dp_link_desc_setup(ab, dp->link_desc_banks,
+					HAL_WBM_IDLE_LINK, srng, n_link_desc);
+	if (ret) {
+		ath11k_warn(ab, "failed to setup link desc: %d\n", ret);
+		return ret;
+	}
+
+	ret = ath11k_dp_srng_common_setup(ab);
+	if (ret)
+		goto fail_link_desc_cleanup;
+
+	size = roundup_pow_of_two(DP_TX_COMP_RING_SIZE);
+
+	for (i = 0; i < DP_TCL_NUM_RING_MAX; i++) {
+		idr_init(&dp->tx_ring[i].txbuf_idr);
+		spin_lock_init(&dp->tx_ring[i].tx_idr_lock);
+		dp->tx_ring[i].tcl_data_ring_id = i;
+
+		spin_lock_init(&dp->tx_ring[i].tx_status_lock);
+		ret = kfifo_alloc(&dp->tx_ring[i].tx_status_fifo, size,
+				  GFP_KERNEL);
+		if (ret)
+			goto fail_cmn_srng_cleanup;
+	}
+
+	for (i = 0; i < HAL_DSCP_TID_MAP_TBL_NUM_ENTRIES_MAX; i++)
+		ath11k_hal_tx_set_dscp_tid_map(ab, i);
+
+	/* Init any SOC level resource for DP */
+
+	return 0;
+
+fail_cmn_srng_cleanup:
+	ath11k_dp_srng_common_cleanup(ab);
+
+fail_link_desc_cleanup:
+	ath11k_dp_link_desc_cleanup(ab, dp->link_desc_banks,
+				    HAL_WBM_IDLE_LINK, &dp->wbm_idle_ring);
+
+	return ret;
+}
diff --git a/drivers/net/wireless/ath/ath11k/dp.h b/drivers/net/wireless/ath/ath11k/dp.h
new file mode 100644
index 000000000000..4db71173db62
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/dp.h
@@ -0,0 +1,1526 @@
+/* SPDX-License-Identifier: BSD-3-Clause-Clear */
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#ifndef ATH11K_DP_H
+#define ATH11K_DP_H
+
+#include <linux/kfifo.h>
+#include "hal_rx.h"
+
+struct ath11k_base;
+struct ath11k_peer;
+struct ath11k_dp;
+struct ath11k_vif;
+struct hal_tcl_status_ring;
+struct ath11k_ext_irq_grp;
+
+struct dp_rx_tid {
+	u8 tid;
+	u32 *vaddr;
+	dma_addr_t paddr;
+	u32 size;
+	u32 ba_win_sz;
+	bool active;
+};
+
+#define DP_REO_DESC_FREE_TIMEOUT_MS 1000
+
+struct dp_reo_cache_flush_elem {
+	struct list_head list;
+	struct dp_rx_tid data;
+	unsigned long ts;
+};
+
+struct dp_reo_cmd {
+	struct list_head list;
+	struct dp_rx_tid data;
+	int cmd_num;
+	void (*handler)(struct ath11k_dp *, void *,
+			enum hal_reo_cmd_status status);
+};
+
+struct dp_srng {
+	u32 *vaddr_unaligned;
+	u32 *vaddr;
+	dma_addr_t paddr_unaligned;
+	dma_addr_t paddr;
+	int size;
+	u32 ring_id;
+};
+
+struct dp_rxdma_ring {
+	struct dp_srng refill_buf_ring;
+	struct idr bufs_idr;
+	/* Protects bufs_idr */
+	spinlock_t idr_lock;
+	int bufs_max;
+};
+
+struct dp_tx_ring {
+	u8 tcl_data_ring_id;
+	struct dp_srng tcl_data_ring;
+	struct dp_srng tcl_comp_ring;
+	struct idr txbuf_idr;
+	u32 num_tx_pending;
+	/* Protects txbuf_idr and num_pending */
+	spinlock_t tx_idr_lock;
+	DECLARE_KFIFO_PTR(tx_status_fifo, struct hal_wbm_release_ring);
+	/* lock to protect tx_status_fifo because tx_status_fifo can be
+	 * accessed concurrently.
+	 */
+	spinlock_t tx_status_lock;
+};
+
+struct ath11k_pdev_mon_stats {
+	u32 status_ppdu_state;
+	u32 status_ppdu_start;
+	u32 status_ppdu_end;
+	u32 status_ppdu_compl;
+	u32 status_ppdu_start_mis;
+	u32 status_ppdu_end_mis;
+	u32 status_ppdu_done;
+	u32 dest_ppdu_done;
+	u32 dest_mpdu_done;
+	u32 dest_mpdu_drop;
+	u32 dup_mon_linkdesc_cnt;
+	u32 dup_mon_buf_cnt;
+};
+
+struct dp_link_desc_bank {
+	void *vaddr_unaligned;
+	void *vaddr;
+	dma_addr_t paddr_unaligned;
+	dma_addr_t paddr;
+	u32 size;
+};
+
+/* Size to enforce scatter idle list mode */
+#define DP_LINK_DESC_ALLOC_SIZE_THRESH 0x200000
+#define DP_LINK_DESC_BANKS_MAX 8
+
+#define DP_RX_DESC_COOKIE_INDEX_MAX		0x3ffff
+#define DP_RX_DESC_COOKIE_POOL_ID_MAX		0x1c0000
+#define DP_RX_DESC_COOKIE_MAX	\
+	(DP_RX_DESC_COOKIE_INDEX_MAX | DP_RX_DESC_COOKIE_POOL_ID_MAX)
+#define DP_NOT_PPDU_ID_WRAP_AROUND 20000
+
+enum ath11k_dp_ppdu_state {
+	DP_PPDU_STATUS_START,
+	DP_PPDU_STATUS_DONE,
+};
+
+struct ath11k_mon_data {
+	struct dp_link_desc_bank link_desc_banks[DP_LINK_DESC_BANKS_MAX];
+	struct hal_rx_mon_ppdu_info mon_ppdu_info;
+
+	u32 mon_ppdu_status;
+	u32 mon_last_buf_cookie;
+	u64 mon_last_linkdesc_paddr;
+	u16 chan_noise_floor;
+
+	struct ath11k_pdev_mon_stats rx_mon_stats;
+	/* lock for monitor data */
+	spinlock_t mon_lock;
+	struct sk_buff_head rx_status_q;
+};
+
+struct ath11k_pdev_dp {
+	u32 mac_id;
+	atomic_t num_tx_pending;
+	wait_queue_head_t tx_empty_waitq;
+	struct dp_srng reo_dst_ring;
+	struct dp_rxdma_ring rx_refill_buf_ring;
+	struct dp_srng rxdma_err_dst_ring;
+	struct dp_srng rxdma_mon_dst_ring;
+	struct dp_srng rxdma_mon_desc_ring;
+
+	struct dp_rxdma_ring rxdma_mon_buf_ring;
+	struct dp_rxdma_ring rx_mon_status_refill_ring;
+	struct ieee80211_rx_status rx_status;
+	struct ath11k_mon_data mon_data;
+};
+
+#define DP_NUM_CLIENTS_MAX 64
+#define DP_AVG_TIDS_PER_CLIENT 2
+#define DP_NUM_TIDS_MAX (DP_NUM_CLIENTS_MAX * DP_AVG_TIDS_PER_CLIENT)
+#define DP_AVG_MSDUS_PER_FLOW 128
+#define DP_AVG_FLOWS_PER_TID 2
+#define DP_AVG_MPDUS_PER_TID_MAX 128
+#define DP_AVG_MSDUS_PER_MPDU 4
+
+#define DP_RX_HASH_ENABLE	0 /* Disable hash based Rx steering */
+
+#define DP_BA_WIN_SZ_MAX	256
+
+#define DP_TCL_NUM_RING_MAX	3
+
+#define DP_IDLE_SCATTER_BUFS_MAX 16
+
+#define DP_WBM_RELEASE_RING_SIZE	64
+#define DP_TCL_DATA_RING_SIZE		512
+#define DP_TX_COMP_RING_SIZE		8192
+#define DP_TX_IDR_SIZE			(DP_TX_COMP_RING_SIZE << 1)
+#define DP_TCL_CMD_RING_SIZE		32
+#define DP_TCL_STATUS_RING_SIZE		32
+#define DP_REO_DST_RING_MAX		4
+#define DP_REO_DST_RING_SIZE		2048
+#define DP_REO_REINJECT_RING_SIZE	32
+#define DP_RX_RELEASE_RING_SIZE		1024
+#define DP_REO_EXCEPTION_RING_SIZE	128
+#define DP_REO_CMD_RING_SIZE		128
+#define DP_REO_STATUS_RING_SIZE		256
+#define DP_RXDMA_BUF_RING_SIZE		4096
+#define DP_RXDMA_REFILL_RING_SIZE	2048
+#define DP_RXDMA_ERR_DST_RING_SIZE	1024
+#define DP_RXDMA_MON_STATUS_RING_SIZE	1024
+#define DP_RXDMA_MONITOR_BUF_RING_SIZE	4096
+#define DP_RXDMA_MONITOR_DST_RING_SIZE	2048
+#define DP_RXDMA_MONITOR_DESC_RING_SIZE	4096
+
+#define DP_RX_BUFFER_SIZE	2048
+#define DP_RX_BUFFER_ALIGN_SIZE	128
+
+#define DP_RXDMA_BUF_COOKIE_BUF_ID	GENMASK(17, 0)
+#define DP_RXDMA_BUF_COOKIE_PDEV_ID	GENMASK(20, 18)
+
+#define DP_HW2SW_MACID(mac_id) ((mac_id) ? ((mac_id) - 1) : 0)
+#define DP_SW2HW_MACID(mac_id) ((mac_id) + 1)
+
+#define DP_TX_DESC_ID_MAC_ID  GENMASK(1, 0)
+#define DP_TX_DESC_ID_MSDU_ID GENMASK(18, 2)
+#define DP_TX_DESC_ID_POOL_ID GENMASK(20, 19)
+
+struct ath11k_dp {
+	struct ath11k_base *ab;
+	enum ath11k_htc_ep_id eid;
+	struct completion htt_tgt_version_received;
+	u8 htt_tgt_ver_major;
+	u8 htt_tgt_ver_minor;
+	struct dp_link_desc_bank link_desc_banks[DP_LINK_DESC_BANKS_MAX];
+	struct dp_srng wbm_idle_ring;
+	struct dp_srng wbm_desc_rel_ring;
+	struct dp_srng tcl_cmd_ring;
+	struct dp_srng tcl_status_ring;
+	struct dp_srng reo_reinject_ring;
+	struct dp_srng rx_rel_ring;
+	struct dp_srng reo_except_ring;
+	struct dp_srng reo_cmd_ring;
+	struct dp_srng reo_status_ring;
+	struct dp_tx_ring tx_ring[DP_TCL_NUM_RING_MAX];
+	struct hal_wbm_idle_scatter_list scatter_list[DP_IDLE_SCATTER_BUFS_MAX];
+	struct list_head reo_cmd_list;
+	struct list_head reo_cmd_cache_flush_list;
+	/* protects access to reo_cmd_list and reo_cmd_cache_flush_list */
+	spinlock_t reo_cmd_lock;
+};
+
+/* HTT definitions */
+
+#define HTT_TCL_META_DATA_TYPE			BIT(0)
+#define HTT_TCL_META_DATA_VALID_HTT		BIT(1)
+
+/* vdev meta data */
+#define HTT_TCL_META_DATA_VDEV_ID		GENMASK(9, 2)
+#define HTT_TCL_META_DATA_PDEV_ID		GENMASK(11, 10)
+#define HTT_TCL_META_DATA_HOST_INSPECTED	BIT(12)
+
+/* peer meta data */
+#define HTT_TCL_META_DATA_PEER_ID		GENMASK(15, 2)
+
+#define HTT_TX_WBM_COMP_STATUS_OFFSET 8
+
+/* HTT tx completion is overlayed in wbm_release_ring */
+#define HTT_TX_WBM_COMP_INFO0_STATUS		GENMASK(12, 9)
+#define HTT_TX_WBM_COMP_INFO0_REINJECT_REASON	GENMASK(16, 13)
+#define HTT_TX_WBM_COMP_INFO0_REINJECT_REASON	GENMASK(16, 13)
+
+#define HTT_TX_WBM_COMP_INFO1_ACK_RSSI		GENMASK(31, 24)
+
+struct htt_tx_wbm_completion {
+	u32 info0;
+	u32 info1;
+	u32 info2;
+	u32 info3;
+} __packed;
+
+enum htt_h2t_msg_type {
+	HTT_H2T_MSG_TYPE_VERSION_REQ		= 0,
+	HTT_H2T_MSG_TYPE_SRING_SETUP		= 0xb,
+	HTT_H2T_MSG_TYPE_RX_RING_SELECTION_CFG	= 0xc,
+	HTT_H2T_MSG_TYPE_EXT_STATS_CFG		= 0x10,
+	HTT_H2T_MSG_TYPE_PPDU_STATS_CFG		= 0x11,
+};
+
+#define HTT_VER_REQ_INFO_MSG_ID		GENMASK(7, 0)
+
+struct htt_ver_req_cmd {
+	u32 ver_reg_info;
+} __packed;
+
+enum htt_srng_ring_type {
+	HTT_HW_TO_SW_RING,
+	HTT_SW_TO_HW_RING,
+	HTT_SW_TO_SW_RING,
+};
+
+enum htt_srng_ring_id {
+	HTT_RXDMA_HOST_BUF_RING,
+	HTT_RXDMA_MONITOR_STATUS_RING,
+	HTT_RXDMA_MONITOR_BUF_RING,
+	HTT_RXDMA_MONITOR_DESC_RING,
+	HTT_RXDMA_MONITOR_DEST_RING,
+	HTT_HOST1_TO_FW_RXBUF_RING,
+	HTT_HOST2_TO_FW_RXBUF_RING,
+	HTT_RXDMA_NON_MONITOR_DEST_RING,
+};
+
+/* host -> target  HTT_SRING_SETUP message
+ *
+ * After target is booted up, Host can send SRING setup message for
+ * each host facing LMAC SRING. Target setups up HW registers based
+ * on setup message and confirms back to Host if response_required is set.
+ * Host should wait for confirmation message before sending new SRING
+ * setup message
+ *
+ * The message would appear as follows:
+ *
+ * |31            24|23    20|19|18 16|15|14          8|7                0|
+ * |--------------- +-----------------+----------------+------------------|
+ * |    ring_type   |      ring_id    |    pdev_id     |     msg_type     |
+ * |----------------------------------------------------------------------|
+ * |                          ring_base_addr_lo                           |
+ * |----------------------------------------------------------------------|
+ * |                         ring_base_addr_hi                            |
+ * |----------------------------------------------------------------------|
+ * |ring_misc_cfg_flag|ring_entry_size|            ring_size              |
+ * |----------------------------------------------------------------------|
+ * |                         ring_head_offset32_remote_addr_lo            |
+ * |----------------------------------------------------------------------|
+ * |                         ring_head_offset32_remote_addr_hi            |
+ * |----------------------------------------------------------------------|
+ * |                         ring_tail_offset32_remote_addr_lo            |
+ * |----------------------------------------------------------------------|
+ * |                         ring_tail_offset32_remote_addr_hi            |
+ * |----------------------------------------------------------------------|
+ * |                          ring_msi_addr_lo                            |
+ * |----------------------------------------------------------------------|
+ * |                          ring_msi_addr_hi                            |
+ * |----------------------------------------------------------------------|
+ * |                          ring_msi_data                               |
+ * |----------------------------------------------------------------------|
+ * |         intr_timer_th            |IM|      intr_batch_counter_th     |
+ * |----------------------------------------------------------------------|
+ * |          reserved        |RR|PTCF|        intr_low_threshold         |
+ * |----------------------------------------------------------------------|
+ * Where
+ *     IM = sw_intr_mode
+ *     RR = response_required
+ *     PTCF = prefetch_timer_cfg
+ *
+ * The message is interpreted as follows:
+ * dword0  - b'0:7   - msg_type: This will be set to
+ *                     HTT_H2T_MSG_TYPE_SRING_SETUP
+ *           b'8:15  - pdev_id:
+ *                     0 (for rings at SOC/UMAC level),
+ *                     1/2/3 mac id (for rings at LMAC level)
+ *           b'16:23 - ring_id: identify which ring is to setup,
+ *                     more details can be got from enum htt_srng_ring_id
+ *           b'24:31 - ring_type: identify type of host rings,
+ *                     more details can be got from enum htt_srng_ring_type
+ * dword1  - b'0:31  - ring_base_addr_lo: Lower 32bits of ring base address
+ * dword2  - b'0:31  - ring_base_addr_hi: Upper 32bits of ring base address
+ * dword3  - b'0:15  - ring_size: size of the ring in unit of 4-bytes words
+ *           b'16:23 - ring_entry_size: Size of each entry in 4-byte word units
+ *           b'24:31 - ring_misc_cfg_flag: Valid only for HW_TO_SW_RING and
+ *                     SW_TO_HW_RING.
+ *                     Refer to HTT_SRING_SETUP_RING_MISC_CFG_RING defs.
+ * dword4  - b'0:31  - ring_head_off32_remote_addr_lo:
+ *                     Lower 32 bits of memory address of the remote variable
+ *                     storing the 4-byte word offset that identifies the head
+ *                     element within the ring.
+ *                     (The head offset variable has type u32.)
+ *                     Valid for HW_TO_SW and SW_TO_SW rings.
+ * dword5  - b'0:31  - ring_head_off32_remote_addr_hi:
+ *                     Upper 32 bits of memory address of the remote variable
+ *                     storing the 4-byte word offset that identifies the head
+ *                     element within the ring.
+ *                     (The head offset variable has type u32.)
+ *                     Valid for HW_TO_SW and SW_TO_SW rings.
+ * dword6  - b'0:31  - ring_tail_off32_remote_addr_lo:
+ *                     Lower 32 bits of memory address of the remote variable
+ *                     storing the 4-byte word offset that identifies the tail
+ *                     element within the ring.
+ *                     (The tail offset variable has type u32.)
+ *                     Valid for HW_TO_SW and SW_TO_SW rings.
+ * dword7  - b'0:31  - ring_tail_off32_remote_addr_hi:
+ *                     Upper 32 bits of memory address of the remote variable
+ *                     storing the 4-byte word offset that identifies the tail
+ *                     element within the ring.
+ *                     (The tail offset variable has type u32.)
+ *                     Valid for HW_TO_SW and SW_TO_SW rings.
+ * dword8  - b'0:31  - ring_msi_addr_lo: Lower 32bits of MSI cfg address
+ *                     valid only for HW_TO_SW_RING and SW_TO_HW_RING
+ * dword9  - b'0:31  - ring_msi_addr_hi: Upper 32bits of MSI cfg address
+ *                     valid only for HW_TO_SW_RING and SW_TO_HW_RING
+ * dword10 - b'0:31  - ring_msi_data: MSI data
+ *                     Refer to HTT_SRING_SETUP_RING_MSC_CFG_xxx defs
+ *                     valid only for HW_TO_SW_RING and SW_TO_HW_RING
+ * dword11 - b'0:14  - intr_batch_counter_th:
+ *                     batch counter threshold is in units of 4-byte words.
+ *                     HW internally maintains and increments batch count.
+ *                     (see SRING spec for detail description).
+ *                     When batch count reaches threshold value, an interrupt
+ *                     is generated by HW.
+ *           b'15    - sw_intr_mode:
+ *                     This configuration shall be static.
+ *                     Only programmed at power up.
+ *                     0: generate pulse style sw interrupts
+ *                     1: generate level style sw interrupts
+ *           b'16:31 - intr_timer_th:
+ *                     The timer init value when timer is idle or is
+ *                     initialized to start downcounting.
+ *                     In 8us units (to cover a range of 0 to 524 ms)
+ * dword12 - b'0:15  - intr_low_threshold:
+ *                     Used only by Consumer ring to generate ring_sw_int_p.
+ *                     Ring entries low threshold water mark, that is used
+ *                     in combination with the interrupt timer as well as
+ *                     the the clearing of the level interrupt.
+ *           b'16:18 - prefetch_timer_cfg:
+ *                     Used only by Consumer ring to set timer mode to
+ *                     support Application prefetch handling.
+ *                     The external tail offset/pointer will be updated
+ *                     at following intervals:
+ *                     3'b000: (Prefetch feature disabled; used only for debug)
+ *                     3'b001: 1 usec
+ *                     3'b010: 4 usec
+ *                     3'b011: 8 usec (default)
+ *                     3'b100: 16 usec
+ *                     Others: Reserverd
+ *           b'19    - response_required:
+ *                     Host needs HTT_T2H_MSG_TYPE_SRING_SETUP_DONE as response
+ *           b'20:31 - reserved:  reserved for future use
+ */
+
+#define HTT_SRNG_SETUP_CMD_INFO0_MSG_TYPE	GENMASK(7, 0)
+#define HTT_SRNG_SETUP_CMD_INFO0_PDEV_ID	GENMASK(15, 8)
+#define HTT_SRNG_SETUP_CMD_INFO0_RING_ID	GENMASK(23, 16)
+#define HTT_SRNG_SETUP_CMD_INFO0_RING_TYPE	GENMASK(31, 24)
+
+#define HTT_SRNG_SETUP_CMD_INFO1_RING_SIZE			GENMASK(15, 0)
+#define HTT_SRNG_SETUP_CMD_INFO1_RING_ENTRY_SIZE		GENMASK(23, 16)
+#define HTT_SRNG_SETUP_CMD_INFO1_RING_LOOP_CNT_DIS		BIT(25)
+#define HTT_SRNG_SETUP_CMD_INFO1_RING_FLAGS_MSI_SWAP		BIT(27)
+#define HTT_SRNG_SETUP_CMD_INFO1_RING_FLAGS_HOST_FW_SWAP	BIT(28)
+#define HTT_SRNG_SETUP_CMD_INFO1_RING_FLAGS_TLV_SWAP		BIT(29)
+
+#define HTT_SRNG_SETUP_CMD_INTR_INFO_BATCH_COUNTER_THRESH	GENMASK(14, 0)
+#define HTT_SRNG_SETUP_CMD_INTR_INFO_SW_INTR_MODE		BIT(15)
+#define HTT_SRNG_SETUP_CMD_INTR_INFO_INTR_TIMER_THRESH		GENMASK(31, 16)
+
+#define HTT_SRNG_SETUP_CMD_INFO2_INTR_LOW_THRESH	GENMASK(15, 0)
+#define HTT_SRNG_SETUP_CMD_INFO2_PRE_FETCH_TIMER_CFG	BIT(16)
+#define HTT_SRNG_SETUP_CMD_INFO2_RESPONSE_REQUIRED	BIT(19)
+
+struct htt_srng_setup_cmd {
+	u32 info0;
+	u32 ring_base_addr_lo;
+	u32 ring_base_addr_hi;
+	u32 info1;
+	u32 ring_head_off32_remote_addr_lo;
+	u32 ring_head_off32_remote_addr_hi;
+	u32 ring_tail_off32_remote_addr_lo;
+	u32 ring_tail_off32_remote_addr_hi;
+	u32 ring_msi_addr_lo;
+	u32 ring_msi_addr_hi;
+	u32 msi_data;
+	u32 intr_info;
+	u32 info2;
+} __packed;
+
+/* host -> target FW  PPDU_STATS config message
+ *
+ * @details
+ * The following field definitions describe the format of the HTT host
+ * to target FW for PPDU_STATS_CFG msg.
+ * The message allows the host to configure the PPDU_STATS_IND messages
+ * produced by the target.
+ *
+ * |31          24|23          16|15           8|7            0|
+ * |-----------------------------------------------------------|
+ * |    REQ bit mask             |   pdev_mask  |   msg type   |
+ * |-----------------------------------------------------------|
+ * Header fields:
+ *  - MSG_TYPE
+ *    Bits 7:0
+ *    Purpose: identifies this is a req to configure ppdu_stats_ind from target
+ *    Value: 0x11
+ *  - PDEV_MASK
+ *    Bits 8:15
+ *    Purpose: identifies which pdevs this PPDU stats configuration applies to
+ *    Value: This is a overloaded field, refer to usage and interpretation of
+ *           PDEV in interface document.
+ *           Bit   8    :  Reserved for SOC stats
+ *           Bit 9 - 15 :  Indicates PDEV_MASK in DBDC
+ *                         Indicates MACID_MASK in DBS
+ *  - REQ_TLV_BIT_MASK
+ *    Bits 16:31
+ *    Purpose: each set bit indicates the corresponding PPDU stats TLV type
+ *        needs to be included in the target's PPDU_STATS_IND messages.
+ *    Value: refer htt_ppdu_stats_tlv_tag_t <<<???
+ *
+ */
+
+struct htt_ppdu_stats_cfg_cmd {
+	u32 msg;
+} __packed;
+
+#define HTT_PPDU_STATS_CFG_MSG_TYPE		GENMASK(7, 0)
+#define HTT_PPDU_STATS_CFG_PDEV_ID		GENMASK(16, 9)
+#define HTT_PPDU_STATS_CFG_TLV_TYPE_BITMASK	GENMASK(31, 16)
+
+enum htt_ppdu_stats_tag_type {
+	HTT_PPDU_STATS_TAG_COMMON,
+	HTT_PPDU_STATS_TAG_USR_COMMON,
+	HTT_PPDU_STATS_TAG_USR_RATE,
+	HTT_PPDU_STATS_TAG_USR_MPDU_ENQ_BITMAP_64,
+	HTT_PPDU_STATS_TAG_USR_MPDU_ENQ_BITMAP_256,
+	HTT_PPDU_STATS_TAG_SCH_CMD_STATUS,
+	HTT_PPDU_STATS_TAG_USR_COMPLTN_COMMON,
+	HTT_PPDU_STATS_TAG_USR_COMPLTN_BA_BITMAP_64,
+	HTT_PPDU_STATS_TAG_USR_COMPLTN_BA_BITMAP_256,
+	HTT_PPDU_STATS_TAG_USR_COMPLTN_ACK_BA_STATUS,
+	HTT_PPDU_STATS_TAG_USR_COMPLTN_FLUSH,
+	HTT_PPDU_STATS_TAG_USR_COMMON_ARRAY,
+	HTT_PPDU_STATS_TAG_INFO,
+	HTT_PPDU_STATS_TAG_TX_MGMTCTRL_PAYLOAD,
+
+	/* New TLV's are added above to this line */
+	HTT_PPDU_STATS_TAG_MAX,
+};
+
+#define HTT_PPDU_STATS_TAG_DEFAULT (BIT(HTT_PPDU_STATS_TAG_COMMON) \
+				   | BIT(HTT_PPDU_STATS_TAG_USR_COMMON) \
+				   | BIT(HTT_PPDU_STATS_TAG_USR_RATE) \
+				   | BIT(HTT_PPDU_STATS_TAG_SCH_CMD_STATUS) \
+				   | BIT(HTT_PPDU_STATS_TAG_USR_COMPLTN_COMMON) \
+				   | BIT(HTT_PPDU_STATS_TAG_USR_COMPLTN_ACK_BA_STATUS) \
+				   | BIT(HTT_PPDU_STATS_TAG_USR_COMPLTN_FLUSH) \
+				   | BIT(HTT_PPDU_STATS_TAG_USR_COMMON_ARRAY))
+
+/* HTT_H2T_MSG_TYPE_RX_RING_SELECTION_CFG Message
+ *
+ * details:
+ *    HTT_H2T_MSG_TYPE_RX_RING_SELECTION_CFG message is sent by host to
+ *    configure RXDMA rings.
+ *    The configuration is per ring based and includes both packet subtypes
+ *    and PPDU/MPDU TLVs.
+ *
+ *    The message would appear as follows:
+ *
+ *    |31       26|25|24|23            16|15             8|7             0|
+ *    |-----------------+----------------+----------------+---------------|
+ *    |   rsvd1   |PS|SS|     ring_id    |     pdev_id    |    msg_type   |
+ *    |-------------------------------------------------------------------|
+ *    |              rsvd2               |           ring_buffer_size     |
+ *    |-------------------------------------------------------------------|
+ *    |                        packet_type_enable_flags_0                 |
+ *    |-------------------------------------------------------------------|
+ *    |                        packet_type_enable_flags_1                 |
+ *    |-------------------------------------------------------------------|
+ *    |                        packet_type_enable_flags_2                 |
+ *    |-------------------------------------------------------------------|
+ *    |                        packet_type_enable_flags_3                 |
+ *    |-------------------------------------------------------------------|
+ *    |                         tlv_filter_in_flags                       |
+ *    |-------------------------------------------------------------------|
+ * Where:
+ *     PS = pkt_swap
+ *     SS = status_swap
+ * The message is interpreted as follows:
+ * dword0 - b'0:7   - msg_type: This will be set to
+ *                    HTT_H2T_MSG_TYPE_RX_RING_SELECTION_CFG
+ *          b'8:15  - pdev_id:
+ *                    0 (for rings at SOC/UMAC level),
+ *                    1/2/3 mac id (for rings at LMAC level)
+ *          b'16:23 - ring_id : Identify the ring to configure.
+ *                    More details can be got from enum htt_srng_ring_id
+ *          b'24    - status_swap: 1 is to swap status TLV
+ *          b'25    - pkt_swap:  1 is to swap packet TLV
+ *          b'26:31 - rsvd1:  reserved for future use
+ * dword1 - b'0:16  - ring_buffer_size: size of bufferes referenced by rx ring,
+ *                    in byte units.
+ *                    Valid only for HW_TO_SW_RING and SW_TO_HW_RING
+ *        - b'16:31 - rsvd2: Reserved for future use
+ * dword2 - b'0:31  - packet_type_enable_flags_0:
+ *                    Enable MGMT packet from 0b0000 to 0b1001
+ *                    bits from low to high: FP, MD, MO - 3 bits
+ *                        FP: Filter_Pass
+ *                        MD: Monitor_Direct
+ *                        MO: Monitor_Other
+ *                    10 mgmt subtypes * 3 bits -> 30 bits
+ *                    Refer to PKT_TYPE_ENABLE_FLAG0_xxx_MGMT_xxx defs
+ * dword3 - b'0:31  - packet_type_enable_flags_1:
+ *                    Enable MGMT packet from 0b1010 to 0b1111
+ *                    bits from low to high: FP, MD, MO - 3 bits
+ *                    Refer to PKT_TYPE_ENABLE_FLAG1_xxx_MGMT_xxx defs
+ * dword4 - b'0:31 -  packet_type_enable_flags_2:
+ *                    Enable CTRL packet from 0b0000 to 0b1001
+ *                    bits from low to high: FP, MD, MO - 3 bits
+ *                    Refer to PKT_TYPE_ENABLE_FLAG2_xxx_CTRL_xxx defs
+ * dword5 - b'0:31  - packet_type_enable_flags_3:
+ *                    Enable CTRL packet from 0b1010 to 0b1111,
+ *                    MCAST_DATA, UCAST_DATA, NULL_DATA
+ *                    bits from low to high: FP, MD, MO - 3 bits
+ *                    Refer to PKT_TYPE_ENABLE_FLAG3_xxx_CTRL_xxx defs
+ * dword6 - b'0:31 -  tlv_filter_in_flags:
+ *                    Filter in Attention/MPDU/PPDU/Header/User tlvs
+ *                    Refer to CFG_TLV_FILTER_IN_FLAG defs
+ */
+
+#define HTT_RX_RING_SELECTION_CFG_CMD_INFO0_MSG_TYPE	GENMASK(7, 0)
+#define HTT_RX_RING_SELECTION_CFG_CMD_INFO0_PDEV_ID	GENMASK(15, 8)
+#define HTT_RX_RING_SELECTION_CFG_CMD_INFO0_RING_ID	GENMASK(23, 16)
+#define HTT_RX_RING_SELECTION_CFG_CMD_INFO0_SS		BIT(24)
+#define HTT_RX_RING_SELECTION_CFG_CMD_INFO0_PS		BIT(25)
+
+#define HTT_RX_RING_SELECTION_CFG_CMD_INFO1_BUF_SIZE	GENMASK(15, 0)
+
+enum htt_rx_filter_tlv_flags {
+	HTT_RX_FILTER_TLV_FLAGS_MPDU_START		= BIT(0),
+	HTT_RX_FILTER_TLV_FLAGS_MSDU_START		= BIT(1),
+	HTT_RX_FILTER_TLV_FLAGS_RX_PACKET		= BIT(2),
+	HTT_RX_FILTER_TLV_FLAGS_MSDU_END		= BIT(3),
+	HTT_RX_FILTER_TLV_FLAGS_MPDU_END		= BIT(4),
+	HTT_RX_FILTER_TLV_FLAGS_PACKET_HEADER		= BIT(5),
+	HTT_RX_FILTER_TLV_FLAGS_PER_MSDU_HEADER		= BIT(6),
+	HTT_RX_FILTER_TLV_FLAGS_ATTENTION		= BIT(7),
+	HTT_RX_FILTER_TLV_FLAGS_PPDU_START		= BIT(8),
+	HTT_RX_FILTER_TLV_FLAGS_PPDU_END		= BIT(9),
+	HTT_RX_FILTER_TLV_FLAGS_PPDU_END_USER_STATS	= BIT(10),
+	HTT_RX_FILTER_TLV_FLAGS_PPDU_END_USER_STATS_EXT	= BIT(11),
+	HTT_RX_FILTER_TLV_FLAGS_PPDU_END_STATUS_DONE	= BIT(12),
+};
+
+enum htt_rx_mgmt_pkt_filter_tlv_flags0 {
+	HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS0_ASSOC_REQ		= BIT(0),
+	HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS0_ASSOC_REQ		= BIT(1),
+	HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS0_ASSOC_REQ		= BIT(2),
+	HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS0_ASSOC_RESP		= BIT(3),
+	HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS0_ASSOC_RESP		= BIT(4),
+	HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS0_ASSOC_RESP		= BIT(5),
+	HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS0_REASSOC_REQ	= BIT(6),
+	HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS0_REASSOC_REQ	= BIT(7),
+	HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS0_REASSOC_REQ	= BIT(8),
+	HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS0_REASSOC_RESP	= BIT(9),
+	HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS0_REASSOC_RESP	= BIT(10),
+	HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS0_REASSOC_RESP	= BIT(11),
+	HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS0_PROBE_REQ		= BIT(12),
+	HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS0_PROBE_REQ		= BIT(13),
+	HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS0_PROBE_REQ		= BIT(14),
+	HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS0_PROBE_RESP		= BIT(15),
+	HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS0_PROBE_RESP		= BIT(16),
+	HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS0_PROBE_RESP		= BIT(17),
+	HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS0_PROBE_TIMING_ADV	= BIT(18),
+	HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS0_PROBE_TIMING_ADV	= BIT(19),
+	HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS0_PROBE_TIMING_ADV	= BIT(20),
+	HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS0_RESERVED_7		= BIT(21),
+	HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS0_RESERVED_7		= BIT(22),
+	HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS0_RESERVED_7		= BIT(23),
+	HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS0_BEACON		= BIT(24),
+	HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS0_BEACON		= BIT(25),
+	HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS0_BEACON		= BIT(26),
+	HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS0_ATIM		= BIT(27),
+	HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS0_ATIM		= BIT(28),
+	HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS0_ATIM		= BIT(29),
+};
+
+enum htt_rx_mgmt_pkt_filter_tlv_flags1 {
+	HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS1_DISASSOC		= BIT(0),
+	HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS1_DISASSOC		= BIT(1),
+	HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS1_DISASSOC		= BIT(2),
+	HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS1_AUTH		= BIT(3),
+	HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS1_AUTH		= BIT(4),
+	HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS1_AUTH		= BIT(5),
+	HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS1_DEAUTH		= BIT(6),
+	HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS1_DEAUTH		= BIT(7),
+	HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS1_DEAUTH		= BIT(8),
+	HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS1_ACTION		= BIT(9),
+	HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS1_ACTION		= BIT(10),
+	HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS1_ACTION		= BIT(11),
+	HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS1_ACTION_NOACK	= BIT(12),
+	HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS1_ACTION_NOACK	= BIT(13),
+	HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS1_ACTION_NOACK	= BIT(14),
+	HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS1_RESERVED_15	= BIT(15),
+	HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS1_RESERVED_15	= BIT(16),
+	HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS1_RESERVED_15	= BIT(17),
+};
+
+enum htt_rx_ctrl_pkt_filter_tlv_flags2 {
+	HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_RESERVED_1	= BIT(0),
+	HTT_RX_MD_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_RESERVED_1	= BIT(1),
+	HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_RESERVED_1	= BIT(2),
+	HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_RESERVED_2	= BIT(3),
+	HTT_RX_MD_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_RESERVED_2	= BIT(4),
+	HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_RESERVED_2	= BIT(5),
+	HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_TRIGGER	= BIT(6),
+	HTT_RX_MD_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_TRIGGER	= BIT(7),
+	HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_TRIGGER	= BIT(8),
+	HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_RESERVED_4	= BIT(9),
+	HTT_RX_MD_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_RESERVED_4	= BIT(10),
+	HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_RESERVED_4	= BIT(11),
+	HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_BF_REP_POLL	= BIT(12),
+	HTT_RX_MD_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_BF_REP_POLL	= BIT(13),
+	HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_BF_REP_POLL	= BIT(14),
+	HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_VHT_NDP	= BIT(15),
+	HTT_RX_MD_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_VHT_NDP	= BIT(16),
+	HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_VHT_NDP	= BIT(17),
+	HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_FRAME_EXT	= BIT(18),
+	HTT_RX_MD_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_FRAME_EXT	= BIT(19),
+	HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_FRAME_EXT	= BIT(20),
+	HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_WRAPPER	= BIT(21),
+	HTT_RX_MD_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_WRAPPER	= BIT(22),
+	HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_WRAPPER	= BIT(23),
+	HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS2_BAR		= BIT(24),
+	HTT_RX_MD_CTRL_PKT_FILTER_TLV_FLAGS2_BAR		= BIT(25),
+	HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS2_BAR		= BIT(26),
+	HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS2_BA			= BIT(27),
+	HTT_RX_MD_CTRL_PKT_FILTER_TLV_FLAGS2_BA			= BIT(28),
+	HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS2_BA			= BIT(29),
+};
+
+enum htt_rx_ctrl_pkt_filter_tlv_flags3 {
+	HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS3_PSPOLL		= BIT(0),
+	HTT_RX_MD_CTRL_PKT_FILTER_TLV_FLAGS3_PSPOLL		= BIT(1),
+	HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS3_PSPOLL		= BIT(2),
+	HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS3_RTS		= BIT(3),
+	HTT_RX_MD_CTRL_PKT_FILTER_TLV_FLAGS3_RTS		= BIT(4),
+	HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS3_RTS		= BIT(5),
+	HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS3_CTS		= BIT(6),
+	HTT_RX_MD_CTRL_PKT_FILTER_TLV_FLAGS3_CTS		= BIT(7),
+	HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS3_CTS		= BIT(8),
+	HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS3_ACK		= BIT(9),
+	HTT_RX_MD_CTRL_PKT_FILTER_TLV_FLAGS3_ACK		= BIT(10),
+	HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS3_ACK		= BIT(11),
+	HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS3_CFEND		= BIT(12),
+	HTT_RX_MD_CTRL_PKT_FILTER_TLV_FLAGS3_CFEND		= BIT(13),
+	HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS3_CFEND		= BIT(14),
+	HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS3_CFEND_ACK		= BIT(15),
+	HTT_RX_MD_CTRL_PKT_FILTER_TLV_FLAGS3_CFEND_ACK		= BIT(16),
+	HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS3_CFEND_ACK		= BIT(17),
+};
+
+enum htt_rx_data_pkt_filter_tlv_flasg3 {
+	HTT_RX_FP_DATA_PKT_FILTER_TLV_FLASG3_MCAST	= BIT(18),
+	HTT_RX_MD_DATA_PKT_FILTER_TLV_FLASG3_MCAST	= BIT(19),
+	HTT_RX_MO_DATA_PKT_FILTER_TLV_FLASG3_MCAST	= BIT(20),
+	HTT_RX_FP_DATA_PKT_FILTER_TLV_FLASG3_UCAST	= BIT(21),
+	HTT_RX_MD_DATA_PKT_FILTER_TLV_FLASG3_UCAST	= BIT(22),
+	HTT_RX_MO_DATA_PKT_FILTER_TLV_FLASG3_UCAST	= BIT(23),
+	HTT_RX_FP_DATA_PKT_FILTER_TLV_FLASG3_NULL_DATA	= BIT(24),
+	HTT_RX_MD_DATA_PKT_FILTER_TLV_FLASG3_NULL_DATA	= BIT(25),
+	HTT_RX_MO_DATA_PKT_FILTER_TLV_FLASG3_NULL_DATA	= BIT(26),
+};
+
+#define HTT_RX_FP_MGMT_FILTER_FLAGS0 \
+	(HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS0_ASSOC_REQ \
+	| HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS0_ASSOC_RESP \
+	| HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS0_REASSOC_REQ \
+	| HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS0_REASSOC_RESP \
+	| HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS0_PROBE_REQ \
+	| HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS0_PROBE_RESP \
+	| HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS0_PROBE_TIMING_ADV \
+	| HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS0_BEACON \
+	| HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS0_ATIM)
+
+#define HTT_RX_MD_MGMT_FILTER_FLAGS0 \
+	(HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS0_ASSOC_REQ \
+	| HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS0_ASSOC_RESP \
+	| HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS0_REASSOC_REQ \
+	| HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS0_REASSOC_RESP \
+	| HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS0_PROBE_REQ \
+	| HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS0_PROBE_RESP \
+	| HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS0_PROBE_TIMING_ADV \
+	| HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS0_BEACON \
+	| HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS0_ATIM)
+
+#define HTT_RX_MO_MGMT_FILTER_FLAGS0 \
+	(HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS0_ASSOC_REQ \
+	| HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS0_ASSOC_RESP \
+	| HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS0_REASSOC_REQ \
+	| HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS0_REASSOC_RESP \
+	| HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS0_PROBE_REQ \
+	| HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS0_PROBE_RESP \
+	| HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS0_PROBE_TIMING_ADV \
+	| HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS0_BEACON \
+	| HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS0_ATIM)
+
+#define HTT_RX_FP_MGMT_FILTER_FLAGS1 (HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS1_DISASSOC \
+				     | HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS1_AUTH \
+				     | HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS1_DEAUTH \
+				     | HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS1_ACTION \
+				     | HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS1_ACTION_NOACK)
+
+#define HTT_RX_MD_MGMT_FILTER_FLAGS1 (HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS1_DISASSOC \
+				     | HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS1_AUTH \
+				     | HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS1_DEAUTH \
+				     | HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS1_ACTION \
+				     | HTT_RX_MD_MGMT_PKT_FILTER_TLV_FLAGS1_ACTION_NOACK)
+
+#define HTT_RX_MO_MGMT_FILTER_FLAGS1 (HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS1_DISASSOC \
+				     | HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS1_AUTH \
+				     | HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS1_DEAUTH \
+				     | HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS1_ACTION \
+				     | HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS1_ACTION_NOACK)
+
+#define HTT_RX_FP_CTRL_FILTER_FLASG2 (HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_WRAPPER \
+				     | HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS2_BAR \
+				     | HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS2_BA)
+
+#define HTT_RX_MD_CTRL_FILTER_FLASG2 (HTT_RX_MD_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_WRAPPER \
+				     | HTT_RX_MD_CTRL_PKT_FILTER_TLV_FLAGS2_BAR \
+				     | HTT_RX_MD_CTRL_PKT_FILTER_TLV_FLAGS2_BA)
+
+#define HTT_RX_MO_CTRL_FILTER_FLASG2 (HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_WRAPPER \
+				     | HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS2_BAR \
+				     | HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS2_BA)
+
+#define HTT_RX_FP_CTRL_FILTER_FLASG3 (HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS3_PSPOLL \
+				     | HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS3_RTS \
+				     | HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS3_CTS \
+				     | HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS3_ACK \
+				     | HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS3_CFEND \
+				     | HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS3_CFEND_ACK)
+
+#define HTT_RX_MD_CTRL_FILTER_FLASG3 (HTT_RX_MD_CTRL_PKT_FILTER_TLV_FLAGS3_PSPOLL \
+				     | HTT_RX_MD_CTRL_PKT_FILTER_TLV_FLAGS3_RTS \
+				     | HTT_RX_MD_CTRL_PKT_FILTER_TLV_FLAGS3_CTS \
+				     | HTT_RX_MD_CTRL_PKT_FILTER_TLV_FLAGS3_ACK \
+				     | HTT_RX_MD_CTRL_PKT_FILTER_TLV_FLAGS3_CFEND \
+				     | HTT_RX_MD_CTRL_PKT_FILTER_TLV_FLAGS3_CFEND_ACK)
+
+#define HTT_RX_MO_CTRL_FILTER_FLASG3 (HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS3_PSPOLL \
+				     | HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS3_RTS \
+				     | HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS3_CTS \
+				     | HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS3_ACK \
+				     | HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS3_CFEND \
+				     | HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS3_CFEND_ACK)
+
+#define HTT_RX_FP_DATA_FILTER_FLASG3 (HTT_RX_FP_DATA_PKT_FILTER_TLV_FLASG3_MCAST \
+				     | HTT_RX_FP_DATA_PKT_FILTER_TLV_FLASG3_UCAST \
+				     | HTT_RX_FP_DATA_PKT_FILTER_TLV_FLASG3_NULL_DATA)
+
+#define HTT_RX_MD_DATA_FILTER_FLASG3 (HTT_RX_MD_DATA_PKT_FILTER_TLV_FLASG3_MCAST \
+				     | HTT_RX_MD_DATA_PKT_FILTER_TLV_FLASG3_UCAST \
+				     | HTT_RX_MD_DATA_PKT_FILTER_TLV_FLASG3_NULL_DATA)
+
+#define HTT_RX_MO_DATA_FILTER_FLASG3 (HTT_RX_MO_DATA_PKT_FILTER_TLV_FLASG3_MCAST \
+				     | HTT_RX_MO_DATA_PKT_FILTER_TLV_FLASG3_UCAST \
+				     | HTT_RX_MO_DATA_PKT_FILTER_TLV_FLASG3_NULL_DATA)
+
+#define HTT_RX_MON_FP_MGMT_FILTER_FLAGS0 \
+		(HTT_RX_FP_MGMT_FILTER_FLAGS0 | \
+		HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS0_RESERVED_7)
+
+#define HTT_RX_MON_MO_MGMT_FILTER_FLAGS0 \
+		(HTT_RX_MO_MGMT_FILTER_FLAGS0 | \
+		HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS0_RESERVED_7)
+
+#define HTT_RX_MON_FP_MGMT_FILTER_FLAGS1 \
+		(HTT_RX_FP_MGMT_FILTER_FLAGS1 | \
+		HTT_RX_FP_MGMT_PKT_FILTER_TLV_FLAGS1_RESERVED_15)
+
+#define HTT_RX_MON_MO_MGMT_FILTER_FLAGS1 \
+		(HTT_RX_MO_MGMT_FILTER_FLAGS1 | \
+		HTT_RX_MO_MGMT_PKT_FILTER_TLV_FLAGS1_RESERVED_15)
+
+#define HTT_RX_MON_FP_CTRL_FILTER_FLASG2 \
+		(HTT_RX_FP_CTRL_FILTER_FLASG2 | \
+		HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_RESERVED_1 | \
+		HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_RESERVED_2 | \
+		HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_TRIGGER | \
+		HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_RESERVED_4 | \
+		HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_BF_REP_POLL | \
+		HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_VHT_NDP | \
+		HTT_RX_FP_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_FRAME_EXT)
+
+#define HTT_RX_MON_MO_CTRL_FILTER_FLASG2 \
+		(HTT_RX_MO_CTRL_FILTER_FLASG2 | \
+		HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_RESERVED_1 | \
+		HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_RESERVED_2 | \
+		HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_TRIGGER | \
+		HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_RESERVED_4 | \
+		HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_BF_REP_POLL | \
+		HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_VHT_NDP | \
+		HTT_RX_MO_CTRL_PKT_FILTER_TLV_FLAGS2_CTRL_FRAME_EXT)
+
+#define HTT_RX_MON_FP_CTRL_FILTER_FLASG3 HTT_RX_FP_CTRL_FILTER_FLASG3
+
+#define HTT_RX_MON_MO_CTRL_FILTER_FLASG3 HTT_RX_MO_CTRL_FILTER_FLASG3
+
+#define HTT_RX_MON_FP_DATA_FILTER_FLASG3 HTT_RX_FP_DATA_FILTER_FLASG3
+
+#define HTT_RX_MON_MO_DATA_FILTER_FLASG3 HTT_RX_MO_DATA_FILTER_FLASG3
+
+#define HTT_RX_MON_FILTER_TLV_FLAGS \
+		(HTT_RX_FILTER_TLV_FLAGS_MPDU_START | \
+		HTT_RX_FILTER_TLV_FLAGS_PPDU_START | \
+		HTT_RX_FILTER_TLV_FLAGS_PPDU_END | \
+		HTT_RX_FILTER_TLV_FLAGS_PPDU_END_USER_STATS | \
+		HTT_RX_FILTER_TLV_FLAGS_PPDU_END_USER_STATS_EXT | \
+		HTT_RX_FILTER_TLV_FLAGS_PPDU_END_STATUS_DONE)
+
+#define HTT_RX_MON_FILTER_TLV_FLAGS_MON_STATUS_RING \
+		(HTT_RX_FILTER_TLV_FLAGS_MPDU_START | \
+		HTT_RX_FILTER_TLV_FLAGS_PPDU_START | \
+		HTT_RX_FILTER_TLV_FLAGS_PPDU_END | \
+		HTT_RX_FILTER_TLV_FLAGS_PPDU_END_USER_STATS | \
+		HTT_RX_FILTER_TLV_FLAGS_PPDU_END_USER_STATS_EXT | \
+		HTT_RX_FILTER_TLV_FLAGS_PPDU_END_STATUS_DONE)
+
+#define HTT_RX_MON_FILTER_TLV_FLAGS_MON_BUF_RING \
+		(HTT_RX_FILTER_TLV_FLAGS_MPDU_START | \
+		HTT_RX_FILTER_TLV_FLAGS_MSDU_START | \
+		HTT_RX_FILTER_TLV_FLAGS_RX_PACKET | \
+		HTT_RX_FILTER_TLV_FLAGS_MSDU_END | \
+		HTT_RX_FILTER_TLV_FLAGS_MPDU_END | \
+		HTT_RX_FILTER_TLV_FLAGS_PACKET_HEADER | \
+		HTT_RX_FILTER_TLV_FLAGS_PER_MSDU_HEADER | \
+		HTT_RX_FILTER_TLV_FLAGS_ATTENTION)
+
+struct htt_rx_ring_selection_cfg_cmd {
+	u32 info0;
+	u32 info1;
+	u32 pkt_type_en_flags0;
+	u32 pkt_type_en_flags1;
+	u32 pkt_type_en_flags2;
+	u32 pkt_type_en_flags3;
+	u32 rx_filter_tlv;
+} __packed;
+
+struct htt_rx_ring_tlv_filter {
+	u32 rx_filter; /* see htt_rx_filter_tlv_flags */
+	u32 pkt_filter_flags0; /* MGMT */
+	u32 pkt_filter_flags1; /* MGMT */
+	u32 pkt_filter_flags2; /* CTRL */
+	u32 pkt_filter_flags3; /* DATA */
+};
+
+/* HTT message target->host */
+
+enum htt_t2h_msg_type {
+	HTT_T2H_MSG_TYPE_VERSION_CONF,
+	HTT_T2H_MSG_TYPE_RX_ADDBA	= 0x5,
+	HTT_T2H_MSG_TYPE_PKTLOG		= 0x8,
+	HTT_T2H_MSG_TYPE_SEC_IND	= 0xb,
+	HTT_T2H_MSG_TYPE_PEER_MAP	= 0x1e,
+	HTT_T2H_MSG_TYPE_PEER_UNMAP	= 0x1f,
+	HTT_T2H_MSG_TYPE_PPDU_STATS_IND = 0x1d,
+	HTT_T2H_MSG_TYPE_EXT_STATS_CONF = 0x1c,
+};
+
+#define HTT_TARGET_VERSION_MAJOR 3
+
+#define HTT_T2H_MSG_TYPE		GENMASK(7, 0)
+#define HTT_T2H_VERSION_CONF_MINOR	GENMASK(15, 8)
+#define HTT_T2H_VERSION_CONF_MAJOR	GENMASK(23, 16)
+
+struct htt_t2h_version_conf_msg {
+	u32 version;
+} __packed;
+
+#define HTT_T2H_PEER_MAP_INFO_VDEV_ID	GENMASK(15, 8)
+#define HTT_T2H_PEER_MAP_INFO_PEER_ID	GENMASK(31, 16)
+#define HTT_T2H_PEER_MAP_INFO1_MAC_ADDR_H16	GENMASK(15, 0)
+#define HTT_T2H_PEER_MAP_INFO1_HW_PEER_ID	GENMASK(31, 16)
+#define HTT_T2H_PEER_MAP_INFO2_AST_HASH_VAL	GENMASK(15, 0)
+#define HTT_T2H_PEER_MAP_INFO2_NEXT_HOP_M	BIT(16)
+#define HTT_T2H_PEER_MAP_INFO2_NEXT_HOP_S	16
+
+struct htt_t2h_peer_map_event {
+	u32 info;
+	u32 mac_addr_l32;
+	u32 info1;
+	u32 info2;
+} __packed;
+
+#define HTT_T2H_PEER_UNMAP_INFO_VDEV_ID	HTT_T2H_PEER_MAP_INFO_VDEV_ID
+#define HTT_T2H_PEER_UNMAP_INFO_PEER_ID	HTT_T2H_PEER_MAP_INFO_PEER_ID
+#define HTT_T2H_PEER_UNMAP_INFO1_MAC_ADDR_H16 \
+					HTT_T2H_PEER_MAP_INFO1_MAC_ADDR_H16
+#define HTT_T2H_PEER_MAP_INFO1_NEXT_HOP_M HTT_T2H_PEER_MAP_INFO2_NEXT_HOP_M
+#define HTT_T2H_PEER_MAP_INFO1_NEXT_HOP_S HTT_T2H_PEER_MAP_INFO2_NEXT_HOP_S
+
+struct htt_t2h_peer_unmap_event {
+	u32 info;
+	u32 mac_addr_l32;
+	u32 info1;
+} __packed;
+
+struct htt_resp_msg {
+	union {
+		struct htt_t2h_version_conf_msg version_msg;
+		struct htt_t2h_peer_map_event peer_map_ev;
+		struct htt_t2h_peer_unmap_event peer_unmap_ev;
+	};
+} __packed;
+
+/* ppdu stats
+ *
+ * @details
+ * The following field definitions describe the format of the HTT target
+ * to host ppdu stats indication message.
+ *
+ *
+ * |31                         16|15   12|11   10|9      8|7            0 |
+ * |----------------------------------------------------------------------|
+ * |    payload_size             | rsvd  |pdev_id|mac_id  |    msg type   |
+ * |----------------------------------------------------------------------|
+ * |                          ppdu_id                                     |
+ * |----------------------------------------------------------------------|
+ * |                        Timestamp in us                               |
+ * |----------------------------------------------------------------------|
+ * |                          reserved                                    |
+ * |----------------------------------------------------------------------|
+ * |                    type-specific stats info                          |
+ * |                     (see htt_ppdu_stats.h)                           |
+ * |----------------------------------------------------------------------|
+ * Header fields:
+ *  - MSG_TYPE
+ *    Bits 7:0
+ *    Purpose: Identifies this is a PPDU STATS indication
+ *             message.
+ *    Value: 0x1d
+ *  - mac_id
+ *    Bits 9:8
+ *    Purpose: mac_id of this ppdu_id
+ *    Value: 0-3
+ *  - pdev_id
+ *    Bits 11:10
+ *    Purpose: pdev_id of this ppdu_id
+ *    Value: 0-3
+ *     0 (for rings at SOC level),
+ *     1/2/3 PDEV -> 0/1/2
+ *  - payload_size
+ *    Bits 31:16
+ *    Purpose: total tlv size
+ *    Value: payload_size in bytes
+ */
+
+#define HTT_T2H_PPDU_STATS_INFO_PDEV_ID GENMASK(11, 10)
+#define HTT_T2H_PPDU_STATS_INFO_PAYLOAD_SIZE GENMASK(31, 16)
+
+struct ath11k_htt_ppdu_stats_msg {
+	u32 info;
+	u32 ppdu_id;
+	u32 timestamp;
+	u32 rsvd;
+	u8 data[0];
+} __packed;
+
+struct htt_tlv {
+	u32 header;
+	u8 value[0];
+} __packed;
+
+#define HTT_TLV_TAG			GENMASK(11, 0)
+#define HTT_TLV_LEN			GENMASK(23, 12)
+
+enum HTT_PPDU_STATS_BW {
+	HTT_PPDU_STATS_BANDWIDTH_5MHZ   = 0,
+	HTT_PPDU_STATS_BANDWIDTH_10MHZ  = 1,
+	HTT_PPDU_STATS_BANDWIDTH_20MHZ  = 2,
+	HTT_PPDU_STATS_BANDWIDTH_40MHZ  = 3,
+	HTT_PPDU_STATS_BANDWIDTH_80MHZ  = 4,
+	HTT_PPDU_STATS_BANDWIDTH_160MHZ = 5, /* includes 80+80 */
+	HTT_PPDU_STATS_BANDWIDTH_DYN    = 6,
+};
+
+#define HTT_PPDU_STATS_CMN_FLAGS_FRAME_TYPE_M	GENMASK(7, 0)
+#define HTT_PPDU_STATS_CMN_FLAGS_QUEUE_TYPE_M	GENMASK(15, 8)
+/* bw - HTT_PPDU_STATS_BW */
+#define HTT_PPDU_STATS_CMN_FLAGS_BW_M		GENMASK(19, 16)
+
+struct htt_ppdu_stats_common {
+	u32 ppdu_id;
+	u16 sched_cmdid;
+	u8 ring_id;
+	u8 num_users;
+	u32 flags; /* %HTT_PPDU_STATS_COMMON_FLAGS_*/
+	u32 chain_mask;
+	u32 fes_duration_us; /* frame exchange sequence */
+	u32 ppdu_sch_eval_start_tstmp_us;
+	u32 ppdu_sch_end_tstmp_us;
+	u32 ppdu_start_tstmp_us;
+	/* BIT [15 :  0] - phy mode (WLAN_PHY_MODE) with which ppdu was transmitted
+	 * BIT [31 : 16] - bandwidth (in MHz) with which ppdu was transmitted
+	 */
+	u16 phy_mode;
+	u16 bw_mhz;
+} __packed;
+
+#define HTT_PPDU_STATS_USER_RATE_INFO0_USER_POS_M	GENMASK(3, 0)
+#define HTT_PPDU_STATS_USER_RATE_INFO0_MU_GROUP_ID_M	GENMASK(11, 4)
+
+#define HTT_PPDU_STATS_USER_RATE_INFO1_RESP_TYPE_VALD_M	BIT(0)
+#define HTT_PPDU_STATS_USER_RATE_INFO1_PPDU_TYPE_M	GENMASK(5, 1)
+
+#define HTT_PPDU_STATS_USER_RATE_FLAGS_LTF_SIZE_M	GENMASK(1, 0)
+#define HTT_PPDU_STATS_USER_RATE_FLAGS_STBC_M		BIT(2)
+#define HTT_PPDU_STATS_USER_RATE_FLAGS_HE_RE_M		BIT(3)
+#define HTT_PPDU_STATS_USER_RATE_FLAGS_TXBF_M		GENMASK(7, 4)
+#define HTT_PPDU_STATS_USER_RATE_FLAGS_BW_M		GENMASK(11, 8)
+#define HTT_PPDU_STATS_USER_RATE_FLAGS_NSS_M		GENMASK(15, 12)
+#define HTT_PPDU_STATS_USER_RATE_FLAGS_MCS_M		GENMASK(19, 16)
+#define HTT_PPDU_STATS_USER_RATE_FLAGS_PREAMBLE_M	GENMASK(23, 20)
+#define HTT_PPDU_STATS_USER_RATE_FLAGS_GI_M		GENMASK(27, 24)
+#define HTT_PPDU_STATS_USER_RATE_FLAGS_DCM_M		BIT(28)
+#define HTT_PPDU_STATS_USER_RATE_FLAGS_LDPC_M		BIT(29)
+
+#define HTT_USR_RATE_PREAMBLE(_val) \
+		FIELD_GET(HTT_PPDU_STATS_USER_RATE_FLAGS_PREAMBLE_M, _val)
+#define HTT_USR_RATE_BW(_val) \
+		FIELD_GET(HTT_PPDU_STATS_USER_RATE_FLAGS_BW_M, _val)
+#define HTT_USR_RATE_NSS(_val) \
+		FIELD_GET(HTT_PPDU_STATS_USER_RATE_FLAGS_NSS_M, _val)
+#define HTT_USR_RATE_MCS(_val) \
+		FIELD_GET(HTT_PPDU_STATS_USER_RATE_FLAGS_MCS_M, _val)
+#define HTT_USR_RATE_GI(_val) \
+		FIELD_GET(HTT_PPDU_STATS_USER_RATE_FLAGS_GI_M, _val)
+
+#define HTT_PPDU_STATS_USER_RATE_RESP_FLAGS_LTF_SIZE_M		GENMASK(1, 0)
+#define HTT_PPDU_STATS_USER_RATE_RESP_FLAGS_STBC_M		BIT(2)
+#define HTT_PPDU_STATS_USER_RATE_RESP_FLAGS_HE_RE_M		BIT(3)
+#define HTT_PPDU_STATS_USER_RATE_RESP_FLAGS_TXBF_M		GENMASK(7, 4)
+#define HTT_PPDU_STATS_USER_RATE_RESP_FLAGS_BW_M		GENMASK(11, 8)
+#define HTT_PPDU_STATS_USER_RATE_RESP_FLAGS_NSS_M		GENMASK(15, 12)
+#define HTT_PPDU_STATS_USER_RATE_RESP_FLAGS_MCS_M		GENMASK(19, 16)
+#define HTT_PPDU_STATS_USER_RATE_RESP_FLAGS_PREAMBLE_M		GENMASK(23, 20)
+#define HTT_PPDU_STATS_USER_RATE_RESP_FLAGS_GI_M		GENMASK(27, 24)
+#define HTT_PPDU_STATS_USER_RATE_RESP_FLAGS_DCM_M		BIT(28)
+#define HTT_PPDU_STATS_USER_RATE_RESP_FLAGS_LDPC_M		BIT(29)
+
+struct htt_ppdu_stats_user_rate {
+	u8 tid_num;
+	u8 reserved0;
+	u16 sw_peer_id;
+	u32 info0; /* %HTT_PPDU_STATS_USER_RATE_INFO0_*/
+	u16 ru_end;
+	u16 ru_start;
+	u16 resp_ru_end;
+	u16 resp_ru_start;
+	u32 info1; /* %HTT_PPDU_STATS_USER_RATE_INFO1_ */
+	u32 rate_flags; /* %HTT_PPDU_STATS_USER_RATE_FLAGS_ */
+	/* Note: resp_rate_info is only valid for if resp_type is UL */
+	u32 resp_rate_flags; /* %HTT_PPDU_STATS_USER_RATE_RESP_FLAGS_ */
+} __packed;
+
+#define HTT_PPDU_STATS_TX_INFO_FLAGS_RATECODE_M		GENMASK(7, 0)
+#define HTT_PPDU_STATS_TX_INFO_FLAGS_IS_AMPDU_M		BIT(8)
+#define HTT_PPDU_STATS_TX_INFO_FLAGS_BA_ACK_FAILED_M	GENMASK(10, 9)
+#define HTT_PPDU_STATS_TX_INFO_FLAGS_BW_M		GENMASK(13, 11)
+#define HTT_PPDU_STATS_TX_INFO_FLAGS_SGI_M		BIT(14)
+#define HTT_PPDU_STATS_TX_INFO_FLAGS_PEERID_M		GENMASK(31, 16)
+
+#define HTT_TX_INFO_IS_AMSDU(_flags) \
+			FIELD_GET(HTT_PPDU_STATS_TX_INFO_FLAGS_IS_AMPDU_M, _flags)
+#define HTT_TX_INFO_BA_ACK_FAILED(_flags) \
+			FIELD_GET(HTT_PPDU_STATS_TX_INFO_FLAGS_BA_ACK_FAILED_M, _flags)
+#define HTT_TX_INFO_RATECODE(_flags) \
+			FIELD_GET(HTT_PPDU_STATS_TX_INFO_FLAGS_RATECODE_M, _flags)
+#define HTT_TX_INFO_PEERID(_flags) \
+			FIELD_GET(HTT_PPDU_STATS_TX_INFO_FLAGS_PEERID_M, _flags)
+
+struct htt_tx_ppdu_stats_info {
+	struct htt_tlv tlv_hdr;
+	u32 tx_success_bytes;
+	u32 tx_retry_bytes;
+	u32 tx_failed_bytes;
+	u32 flags; /* %HTT_PPDU_STATS_TX_INFO_FLAGS_ */
+	u16 tx_success_msdus;
+	u16 tx_retry_msdus;
+	u16 tx_failed_msdus;
+	u16 tx_duration; /* united in us */
+} __packed;
+
+enum  htt_ppdu_stats_usr_compln_status {
+	HTT_PPDU_STATS_USER_STATUS_OK,
+	HTT_PPDU_STATS_USER_STATUS_FILTERED,
+	HTT_PPDU_STATS_USER_STATUS_RESP_TIMEOUT,
+	HTT_PPDU_STATS_USER_STATUS_RESP_MISMATCH,
+	HTT_PPDU_STATS_USER_STATUS_ABORT,
+};
+
+#define HTT_PPDU_STATS_USR_CMPLTN_CMN_FLAGS_LONG_RETRY_M	GENMASK(3, 0)
+#define HTT_PPDU_STATS_USR_CMPLTN_CMN_FLAGS_SHORT_RETRY_M	GENMASK(7, 4)
+#define HTT_PPDU_STATS_USR_CMPLTN_CMN_FLAGS_IS_AMPDU_M		BIT(8)
+#define HTT_PPDU_STATS_USR_CMPLTN_CMN_FLAGS_RESP_TYPE_M		GENMASK(12, 9)
+
+#define HTT_USR_CMPLTN_IS_AMPDU(_val) \
+	    FIELD_GET(HTT_PPDU_STATS_USR_CMPLTN_CMN_FLAGS_IS_AMPDU_M, _val)
+#define HTT_USR_CMPLTN_LONG_RETRY(_val) \
+	    FIELD_GET(HTT_PPDU_STATS_USR_CMPLTN_CMN_FLAGS_LONG_RETRY_M, _val)
+#define HTT_USR_CMPLTN_SHORT_RETRY(_val) \
+	    FIELD_GET(HTT_PPDU_STATS_USR_CMPLTN_CMN_FLAGS_SHORT_RETRY_M, _val)
+
+struct htt_ppdu_stats_usr_cmpltn_cmn {
+	u8 status;
+	u8 tid_num;
+	u16 sw_peer_id;
+	/* RSSI value of last ack packet (units = dB above noise floor) */
+	u32 ack_rssi;
+	u16 mpdu_tried;
+	u16 mpdu_success;
+	u32 flags; /* %HTT_PPDU_STATS_USR_CMPLTN_CMN_FLAGS_LONG_RETRIES*/
+} __packed;
+
+#define HTT_PPDU_STATS_ACK_BA_INFO_NUM_MPDU_M	GENMASK(8, 0)
+#define HTT_PPDU_STATS_ACK_BA_INFO_NUM_MSDU_M	GENMASK(24, 9)
+#define HTT_PPDU_STATS_ACK_BA_INFO_TID_NUM	GENMASK(3, 0)
+
+struct htt_ppdu_stats_usr_cmpltn_ack_ba_status {
+	u32 ppdu_id;
+	u16 sw_peer_id;
+	u16 reserved0;
+	u32 info; /* %HTT_PPDU_STATS_USR_CMPLTN_CMN_INFO_ */
+	u16 current_seq;
+	u16 start_seq;
+	u32 success_bytes;
+} __packed;
+
+struct htt_ppdu_stats_usr_cmn_array {
+	struct htt_tlv tlv_hdr;
+	u32 num_ppdu_stats;
+	/* tx_ppdu_stats_info is filled by multiple struct htt_tx_ppdu_stats_info
+	 * elements.
+	 * tx_ppdu_stats_info is variable length, with length =
+	 *     number_of_ppdu_stats * sizeof (struct htt_tx_ppdu_stats_info)
+	 */
+	struct htt_tx_ppdu_stats_info tx_ppdu_info[0];
+} __packed;
+
+struct htt_ppdu_user_stats {
+	u16 peer_id;
+	u32 tlv_flags;
+	bool is_valid_peer_id;
+	struct htt_ppdu_stats_user_rate rate;
+	struct htt_ppdu_stats_usr_cmpltn_cmn cmpltn_cmn;
+	struct htt_ppdu_stats_usr_cmpltn_ack_ba_status ack_ba;
+};
+
+#define HTT_PPDU_STATS_MAX_USERS	8
+#define HTT_PPDU_DESC_MAX_DEPTH	16
+
+struct htt_ppdu_stats {
+	struct htt_ppdu_stats_common common;
+	struct htt_ppdu_user_stats user_stats[HTT_PPDU_STATS_MAX_USERS];
+};
+
+struct htt_ppdu_stats_info {
+	u32 ppdu_id;
+	struct htt_ppdu_stats ppdu_stats;
+	struct list_head list;
+};
+
+/**
+ * @brief target -> host packet log message
+ *
+ * @details
+ * The following field definitions describe the format of the packet log
+ * message sent from the target to the host.
+ * The message consists of a 4-octet header,followed by a variable number
+ * of 32-bit character values.
+ *
+ * |31                         16|15  12|11   10|9    8|7            0|
+ * |------------------------------------------------------------------|
+ * |        payload_size         | rsvd |pdev_id|mac_id|   msg type   |
+ * |------------------------------------------------------------------|
+ * |                              payload                             |
+ * |------------------------------------------------------------------|
+ *   - MSG_TYPE
+ *     Bits 7:0
+ *     Purpose: identifies this as a pktlog message
+ *     Value: HTT_T2H_MSG_TYPE_PKTLOG
+ *   - mac_id
+ *     Bits 9:8
+ *     Purpose: identifies which MAC/PHY instance generated this pktlog info
+ *     Value: 0-3
+ *   - pdev_id
+ *     Bits 11:10
+ *     Purpose: pdev_id
+ *     Value: 0-3
+ *     0 (for rings at SOC level),
+ *     1/2/3 PDEV -> 0/1/2
+ *   - payload_size
+ *     Bits 31:16
+ *     Purpose: explicitly specify the payload size
+ *     Value: payload size in bytes (payload size is a multiple of 4 bytes)
+ */
+struct htt_pktlog_msg {
+	u32 hdr;
+	u8 payload[0];
+};
+
+/**
+ * @brief host -> target FW extended statistics retrieve
+ *
+ * @details
+ * The following field definitions describe the format of the HTT host
+ * to target FW extended stats retrieve message.
+ * The message specifies the type of stats the host wants to retrieve.
+ *
+ * |31          24|23          16|15           8|7            0|
+ * |-----------------------------------------------------------|
+ * |   reserved   | stats type   |   pdev_mask  |   msg type   |
+ * |-----------------------------------------------------------|
+ * |                   config param [0]                        |
+ * |-----------------------------------------------------------|
+ * |                   config param [1]                        |
+ * |-----------------------------------------------------------|
+ * |                   config param [2]                        |
+ * |-----------------------------------------------------------|
+ * |                   config param [3]                        |
+ * |-----------------------------------------------------------|
+ * |                         reserved                          |
+ * |-----------------------------------------------------------|
+ * |                        cookie LSBs                        |
+ * |-----------------------------------------------------------|
+ * |                        cookie MSBs                        |
+ * |-----------------------------------------------------------|
+ * Header fields:
+ *  - MSG_TYPE
+ *    Bits 7:0
+ *    Purpose: identifies this is a extended stats upload request message
+ *    Value: 0x10
+ *  - PDEV_MASK
+ *    Bits 8:15
+ *    Purpose: identifies the mask of PDEVs to retrieve stats from
+ *    Value: This is a overloaded field, refer to usage and interpretation of
+ *           PDEV in interface document.
+ *           Bit   8    :  Reserved for SOC stats
+ *           Bit 9 - 15 :  Indicates PDEV_MASK in DBDC
+ *                         Indicates MACID_MASK in DBS
+ *  - STATS_TYPE
+ *    Bits 23:16
+ *    Purpose: identifies which FW statistics to upload
+ *    Value: Defined by htt_dbg_ext_stats_type (see htt_stats.h)
+ *  - Reserved
+ *    Bits 31:24
+ *  - CONFIG_PARAM [0]
+ *    Bits 31:0
+ *    Purpose: give an opaque configuration value to the specified stats type
+ *    Value: stats-type specific configuration value
+ *           Refer to htt_stats.h for interpretation for each stats sub_type
+ *  - CONFIG_PARAM [1]
+ *    Bits 31:0
+ *    Purpose: give an opaque configuration value to the specified stats type
+ *    Value: stats-type specific configuration value
+ *           Refer to htt_stats.h for interpretation for each stats sub_type
+ *  - CONFIG_PARAM [2]
+ *    Bits 31:0
+ *    Purpose: give an opaque configuration value to the specified stats type
+ *    Value: stats-type specific configuration value
+ *           Refer to htt_stats.h for interpretation for each stats sub_type
+ *  - CONFIG_PARAM [3]
+ *    Bits 31:0
+ *    Purpose: give an opaque configuration value to the specified stats type
+ *    Value: stats-type specific configuration value
+ *           Refer to htt_stats.h for interpretation for each stats sub_type
+ *  - Reserved [31:0] for future use.
+ *  - COOKIE_LSBS
+ *    Bits 31:0
+ *    Purpose: Provide a mechanism to match a target->host stats confirmation
+ *        message with its preceding host->target stats request message.
+ *    Value: LSBs of the opaque cookie specified by the host-side requestor
+ *  - COOKIE_MSBS
+ *    Bits 31:0
+ *    Purpose: Provide a mechanism to match a target->host stats confirmation
+ *        message with its preceding host->target stats request message.
+ *    Value: MSBs of the opaque cookie specified by the host-side requestor
+ */
+
+struct htt_ext_stats_cfg_hdr {
+	u8 msg_type;
+	u8 pdev_mask;
+	u8 stats_type;
+	u8 reserved;
+} __packed;
+
+struct htt_ext_stats_cfg_cmd {
+	struct htt_ext_stats_cfg_hdr hdr;
+	u32 cfg_param0;
+	u32 cfg_param1;
+	u32 cfg_param2;
+	u32 cfg_param3;
+	u32 reserved;
+	u32 cookie_lsb;
+	u32 cookie_msb;
+} __packed;
+
+/* htt stats config default params */
+#define HTT_STAT_DEFAULT_RESET_START_OFFSET 0
+#define HTT_STAT_DEFAULT_CFG0_ALL_HWQS 0xffffffff
+#define HTT_STAT_DEFAULT_CFG0_ALL_TXQS 0xffffffff
+#define HTT_STAT_DEFAULT_CFG0_ALL_CMDQS 0xffff
+#define HTT_STAT_DEFAULT_CFG0_ALL_RINGS 0xffff
+#define HTT_STAT_DEFAULT_CFG0_ACTIVE_PEERS 0xff
+#define HTT_STAT_DEFAULT_CFG0_CCA_CUMULATIVE 0x10
+#define HTT_STAT_DEFAULT_CFG0_ACTIVE_VDEVS 0xff
+
+/* HTT_DBG_EXT_STATS_PEER_INFO
+ * PARAMS:
+ * @config_param0:
+ *  [Bit0] - [0] for sw_peer_id, [1] for mac_addr based request
+ *  [Bit15 : Bit 1] htt_peer_stats_req_mode_t
+ *  [Bit31 : Bit16] sw_peer_id
+ * @config_param1:
+ *  peer_stats_req_type_mask:32 (enum htt_peer_stats_tlv_enum)
+ *   0 bit htt_peer_stats_cmn_tlv
+ *   1 bit htt_peer_details_tlv
+ *   2 bit htt_tx_peer_rate_stats_tlv
+ *   3 bit htt_rx_peer_rate_stats_tlv
+ *   4 bit htt_tx_tid_stats_tlv/htt_tx_tid_stats_v1_tlv
+ *   5 bit htt_rx_tid_stats_tlv
+ *   6 bit htt_msdu_flow_stats_tlv
+ * @config_param2: [Bit31 : Bit0] mac_addr31to0
+ * @config_param3: [Bit15 : Bit0] mac_addr47to32
+ *                [Bit31 : Bit16] reserved
+ */
+#define HTT_STAT_PEER_INFO_MAC_ADDR BIT(0)
+#define HTT_STAT_DEFAULT_PEER_REQ_TYPE 0x7f
+
+/* Used to set different configs to the specified stats type.*/
+struct htt_ext_stats_cfg_params {
+	u32 cfg0;
+	u32 cfg1;
+	u32 cfg2;
+	u32 cfg3;
+};
+
+/**
+ * @brief target -> host extended statistics upload
+ *
+ * @details
+ * The following field definitions describe the format of the HTT target
+ * to host stats upload confirmation message.
+ * The message contains a cookie echoed from the HTT host->target stats
+ * upload request, which identifies which request the confirmation is
+ * for, and a single stats can span over multiple HTT stats indication
+ * due to the HTT message size limitation so every HTT ext stats indication
+ * will have tag-length-value stats information elements.
+ * The tag-length header for each HTT stats IND message also includes a
+ * status field, to indicate whether the request for the stat type in
+ * question was fully met, partially met, unable to be met, or invalid
+ * (if the stat type in question is disabled in the target).
+ * A Done bit 1's indicate the end of the of stats info elements.
+ *
+ *
+ * |31                         16|15    12|11|10 8|7   5|4       0|
+ * |--------------------------------------------------------------|
+ * |                   reserved                   |    msg type   |
+ * |--------------------------------------------------------------|
+ * |                         cookie LSBs                          |
+ * |--------------------------------------------------------------|
+ * |                         cookie MSBs                          |
+ * |--------------------------------------------------------------|
+ * |      stats entry length     | rsvd   | D|  S |   stat type   |
+ * |--------------------------------------------------------------|
+ * |                   type-specific stats info                   |
+ * |                      (see htt_stats.h)                       |
+ * |--------------------------------------------------------------|
+ * Header fields:
+ *  - MSG_TYPE
+ *    Bits 7:0
+ *    Purpose: Identifies this is a extended statistics upload confirmation
+ *             message.
+ *    Value: 0x1c
+ *  - COOKIE_LSBS
+ *    Bits 31:0
+ *    Purpose: Provide a mechanism to match a target->host stats confirmation
+ *        message with its preceding host->target stats request message.
+ *    Value: LSBs of the opaque cookie specified by the host-side requestor
+ *  - COOKIE_MSBS
+ *    Bits 31:0
+ *    Purpose: Provide a mechanism to match a target->host stats confirmation
+ *        message with its preceding host->target stats request message.
+ *    Value: MSBs of the opaque cookie specified by the host-side requestor
+ *
+ * Stats Information Element tag-length header fields:
+ *  - STAT_TYPE
+ *    Bits 7:0
+ *    Purpose: identifies the type of statistics info held in the
+ *        following information element
+ *    Value: htt_dbg_ext_stats_type
+ *  - STATUS
+ *    Bits 10:8
+ *    Purpose: indicate whether the requested stats are present
+ *    Value: htt_dbg_ext_stats_status
+ *  - DONE
+ *    Bits 11
+ *    Purpose:
+ *        Indicates the completion of the stats entry, this will be the last
+ *        stats conf HTT segment for the requested stats type.
+ *    Value:
+ *        0 -> the stats retrieval is ongoing
+ *        1 -> the stats retrieval is complete
+ *  - LENGTH
+ *    Bits 31:16
+ *    Purpose: indicate the stats information size
+ *    Value: This field specifies the number of bytes of stats information
+ *       that follows the element tag-length header.
+ *       It is expected but not required that this length is a multiple of
+ *       4 bytes.
+ */
+
+#define HTT_T2H_EXT_STATS_INFO1_LENGTH   GENMASK(31, 16)
+
+struct ath11k_htt_extd_stats_msg {
+	u32 info0;
+	u64 cookie;
+	u32 info1;
+	u8 data[0];
+} __packed;
+
+struct htt_mac_addr {
+	u32 mac_addr_l32;
+	u32 mac_addr_h16;
+};
+
+static inline void ath11k_dp_get_mac_addr(u32 addr_l32, u16 addr_h16, u8 *addr)
+{
+	if (IS_ENABLED(CONFIG_CPU_BIG_ENDIAN)) {
+		addr_l32 = swab32(addr_l32);
+		addr_h16 = swab16(addr_h16);
+	}
+
+	memcpy(addr, &addr_l32, 4);
+	memcpy(addr + 4, &addr_h16, ETH_ALEN - 4);
+}
+
+int ath11k_dp_service_srng(struct ath11k_base *ab,
+			   struct ath11k_ext_irq_grp *irq_grp,
+			   int budget);
+int ath11k_dp_htt_connect(struct ath11k_dp *dp);
+void ath11k_dp_vdev_tx_attach(struct ath11k *ar, struct ath11k_vif *arvif);
+void ath11k_dp_free(struct ath11k_base *ab);
+int ath11k_dp_alloc(struct ath11k_base *ab);
+int ath11k_dp_pdev_alloc(struct ath11k_base *ab);
+void ath11k_dp_pdev_free(struct ath11k_base *ab);
+int ath11k_dp_tx_htt_srng_setup(struct ath11k_base *ab, u32 ring_id,
+				int mac_id, enum hal_ring_type ring_type);
+int ath11k_dp_peer_setup(struct ath11k *ar, int vdev_id, const u8 *addr);
+void ath11k_dp_peer_cleanup(struct ath11k *ar, int vdev_id, const u8 *addr);
+void ath11k_dp_srng_cleanup(struct ath11k_base *ab, struct dp_srng *ring);
+int ath11k_dp_srng_setup(struct ath11k_base *ab, struct dp_srng *ring,
+			 enum hal_ring_type type, int ring_num,
+			 int mac_id, int num_entries);
+void ath11k_dp_link_desc_cleanup(struct ath11k_base *ab,
+				 struct dp_link_desc_bank *desc_bank,
+				 u32 ring_type, struct dp_srng *ring);
+int ath11k_dp_link_desc_setup(struct ath11k_base *ab,
+			      struct dp_link_desc_bank *link_desc_banks,
+			      u32 ring_type, struct hal_srng *srng,
+			      u32 n_link_desc);
+
+#endif
diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c b/drivers/net/wireless/ath/ath11k/dp_rx.c
new file mode 100644
index 000000000000..9491a479fab5
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
@@ -0,0 +1,4143 @@
+// SPDX-License-Identifier: BSD-3-Clause-Clear
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#include <linux/ieee80211.h>
+#include "core.h"
+#include "debug.h"
+#include "hal_desc.h"
+#include "hw.h"
+#include "dp_rx.h"
+#include "hal_rx.h"
+#include "dp_tx.h"
+#include "peer.h"
+
+static u8 *ath11k_dp_rx_h_80211_hdr(struct hal_rx_desc *desc)
+{
+	return desc->hdr_status;
+}
+
+static enum hal_encrypt_type ath11k_dp_rx_h_mpdu_start_enctype(struct hal_rx_desc *desc)
+{
+	if (!(__le32_to_cpu(desc->mpdu_start.info1) &
+	    RX_MPDU_START_INFO1_ENCRYPT_INFO_VALID))
+		return HAL_ENCRYPT_TYPE_OPEN;
+
+	return FIELD_GET(RX_MPDU_START_INFO2_ENC_TYPE,
+			 __le32_to_cpu(desc->mpdu_start.info2));
+}
+
+static u8 ath11k_dp_rx_h_mpdu_start_decap_type(struct hal_rx_desc *desc)
+{
+	return FIELD_GET(RX_MPDU_START_INFO5_DECAP_TYPE,
+			 __le32_to_cpu(desc->mpdu_start.info5));
+}
+
+static bool ath11k_dp_rx_h_attn_msdu_done(struct hal_rx_desc *desc)
+{
+	return !!FIELD_GET(RX_ATTENTION_INFO2_MSDU_DONE,
+			   __le32_to_cpu(desc->attention.info2));
+}
+
+static bool ath11k_dp_rx_h_attn_first_mpdu(struct hal_rx_desc *desc)
+{
+	return !!FIELD_GET(RX_ATTENTION_INFO1_FIRST_MPDU,
+			   __le32_to_cpu(desc->attention.info1));
+}
+
+static bool ath11k_dp_rx_h_attn_l4_cksum_fail(struct hal_rx_desc *desc)
+{
+	return !!FIELD_GET(RX_ATTENTION_INFO1_TCP_UDP_CKSUM_FAIL,
+			   __le32_to_cpu(desc->attention.info1));
+}
+
+static bool ath11k_dp_rx_h_attn_ip_cksum_fail(struct hal_rx_desc *desc)
+{
+	return !!FIELD_GET(RX_ATTENTION_INFO1_IP_CKSUM_FAIL,
+			   __le32_to_cpu(desc->attention.info1));
+}
+
+static bool ath11k_dp_rx_h_attn_is_decrypted(struct hal_rx_desc *desc)
+{
+	return (FIELD_GET(RX_ATTENTION_INFO2_DCRYPT_STATUS_CODE,
+			  __le32_to_cpu(desc->attention.info2)) ==
+		RX_DESC_DECRYPT_STATUS_CODE_OK);
+}
+
+static u32 ath11k_dp_rx_h_attn_mpdu_err(struct hal_rx_desc *desc)
+{
+	u32 info = __le32_to_cpu(desc->attention.info1);
+	u32 errmap = 0;
+
+	if (info & RX_ATTENTION_INFO1_FCS_ERR)
+		errmap |= DP_RX_MPDU_ERR_FCS;
+
+	if (info & RX_ATTENTION_INFO1_DECRYPT_ERR)
+		errmap |= DP_RX_MPDU_ERR_DECRYPT;
+
+	if (info & RX_ATTENTION_INFO1_TKIP_MIC_ERR)
+		errmap |= DP_RX_MPDU_ERR_TKIP_MIC;
+
+	if (info & RX_ATTENTION_INFO1_A_MSDU_ERROR)
+		errmap |= DP_RX_MPDU_ERR_AMSDU_ERR;
+
+	if (info & RX_ATTENTION_INFO1_OVERFLOW_ERR)
+		errmap |= DP_RX_MPDU_ERR_OVERFLOW;
+
+	if (info & RX_ATTENTION_INFO1_MSDU_LEN_ERR)
+		errmap |= DP_RX_MPDU_ERR_MSDU_LEN;
+
+	if (info & RX_ATTENTION_INFO1_MPDU_LEN_ERR)
+		errmap |= DP_RX_MPDU_ERR_MPDU_LEN;
+
+	return errmap;
+}
+
+static u16 ath11k_dp_rx_h_msdu_start_msdu_len(struct hal_rx_desc *desc)
+{
+	return FIELD_GET(RX_MSDU_START_INFO1_MSDU_LENGTH,
+			 __le32_to_cpu(desc->msdu_start.info1));
+}
+
+static u8 ath11k_dp_rx_h_msdu_start_sgi(struct hal_rx_desc *desc)
+{
+	return FIELD_GET(RX_MSDU_START_INFO3_SGI,
+			 __le32_to_cpu(desc->msdu_start.info3));
+}
+
+static u8 ath11k_dp_rx_h_msdu_start_rate_mcs(struct hal_rx_desc *desc)
+{
+	return FIELD_GET(RX_MSDU_START_INFO3_RATE_MCS,
+			 __le32_to_cpu(desc->msdu_start.info3));
+}
+
+static u8 ath11k_dp_rx_h_msdu_start_rx_bw(struct hal_rx_desc *desc)
+{
+	return FIELD_GET(RX_MSDU_START_INFO3_RECV_BW,
+			 __le32_to_cpu(desc->msdu_start.info3));
+}
+
+static u32 ath11k_dp_rx_h_msdu_start_freq(struct hal_rx_desc *desc)
+{
+	return __le32_to_cpu(desc->msdu_start.phy_meta_data);
+}
+
+static u8 ath11k_dp_rx_h_msdu_start_pkt_type(struct hal_rx_desc *desc)
+{
+	return FIELD_GET(RX_MSDU_START_INFO3_PKT_TYPE,
+			 __le32_to_cpu(desc->msdu_start.info3));
+}
+
+static u8 ath11k_dp_rx_h_msdu_start_nss(struct hal_rx_desc *desc)
+{
+	u8 mimo_ss_bitmap = FIELD_GET(RX_MSDU_START_INFO3_MIMO_SS_BITMAP,
+				      __le32_to_cpu(desc->msdu_start.info3));
+
+	return hweight8(mimo_ss_bitmap);
+}
+
+static u8 ath11k_dp_rx_h_msdu_end_l3pad(struct hal_rx_desc *desc)
+{
+	return FIELD_GET(RX_MSDU_END_INFO2_L3_HDR_PADDING,
+			 __le32_to_cpu(desc->msdu_end.info2));
+}
+
+static bool ath11k_dp_rx_h_msdu_end_first_msdu(struct hal_rx_desc *desc)
+{
+	return !!FIELD_GET(RX_MSDU_END_INFO2_FIRST_MSDU,
+			   __le32_to_cpu(desc->msdu_end.info2));
+}
+
+static bool ath11k_dp_rx_h_msdu_end_last_msdu(struct hal_rx_desc *desc)
+{
+	return !!FIELD_GET(RX_MSDU_END_INFO2_LAST_MSDU,
+			   __le32_to_cpu(desc->msdu_end.info2));
+}
+
+static void ath11k_dp_rx_desc_end_tlv_copy(struct hal_rx_desc *fdesc,
+					   struct hal_rx_desc *ldesc)
+{
+	memcpy((u8 *)&fdesc->msdu_end, (u8 *)&ldesc->msdu_end,
+	       sizeof(struct rx_msdu_end));
+	memcpy((u8 *)&fdesc->attention, (u8 *)&ldesc->attention,
+	       sizeof(struct rx_attention));
+	memcpy((u8 *)&fdesc->mpdu_end, (u8 *)&ldesc->mpdu_end,
+	       sizeof(struct rx_mpdu_end));
+}
+
+static u32 ath11k_dp_rxdesc_get_mpdulen_err(struct hal_rx_desc *rx_desc)
+{
+	struct rx_attention *rx_attn;
+
+	rx_attn = &rx_desc->attention;
+
+	return FIELD_GET(RX_ATTENTION_INFO1_MPDU_LEN_ERR,
+			 __le32_to_cpu(rx_attn->info1));
+}
+
+static u32 ath11k_dp_rxdesc_get_decap_format(struct hal_rx_desc *rx_desc)
+{
+	struct rx_msdu_start *rx_msdu_start;
+
+	rx_msdu_start = &rx_desc->msdu_start;
+
+	return FIELD_GET(RX_MSDU_START_INFO2_DECAP_FORMAT,
+			 __le32_to_cpu(rx_msdu_start->info2));
+}
+
+static u8 *ath11k_dp_rxdesc_get_80211hdr(struct hal_rx_desc *rx_desc)
+{
+	u8 *rx_pkt_hdr;
+
+	rx_pkt_hdr = &rx_desc->msdu_payload[0];
+
+	return rx_pkt_hdr;
+}
+
+static bool ath11k_dp_rxdesc_mpdu_valid(struct hal_rx_desc *rx_desc)
+{
+	u32 tlv_tag;
+
+	tlv_tag = FIELD_GET(HAL_TLV_HDR_TAG,
+			    __le32_to_cpu(rx_desc->mpdu_start_tag));
+
+	return tlv_tag == HAL_RX_MPDU_START ? true : false;
+}
+
+static u32 ath11k_dp_rxdesc_get_ppduid(struct hal_rx_desc *rx_desc)
+{
+	return __le16_to_cpu(rx_desc->mpdu_start.phy_ppdu_id);
+}
+
+/* Returns number of Rx buffers replenished */
+int ath11k_dp_rxbufs_replenish(struct ath11k_base *ab, int mac_id,
+			       struct dp_rxdma_ring *rx_ring,
+			       int req_entries,
+			       enum hal_rx_buf_return_buf_manager mgr,
+			       gfp_t gfp)
+{
+	struct hal_srng *srng;
+	u32 *desc;
+	struct sk_buff *skb;
+	int num_free;
+	int num_remain;
+	int buf_id;
+	u32 cookie;
+	dma_addr_t paddr;
+
+	req_entries = min(req_entries, rx_ring->bufs_max);
+
+	srng = &ab->hal.srng_list[rx_ring->refill_buf_ring.ring_id];
+
+	spin_lock_bh(&srng->lock);
+
+	ath11k_hal_srng_access_begin(ab, srng);
+
+	num_free = ath11k_hal_srng_src_num_free(ab, srng, true);
+	if (!req_entries && (num_free > (rx_ring->bufs_max * 3) / 4))
+		req_entries = num_free;
+
+	req_entries = min(num_free, req_entries);
+	num_remain = req_entries;
+
+	while (num_remain > 0) {
+		skb = dev_alloc_skb(DP_RX_BUFFER_SIZE +
+				    DP_RX_BUFFER_ALIGN_SIZE);
+		if (!skb)
+			break;
+
+		if (!IS_ALIGNED((unsigned long)skb->data,
+				DP_RX_BUFFER_ALIGN_SIZE)) {
+			skb_pull(skb,
+				 PTR_ALIGN(skb->data, DP_RX_BUFFER_ALIGN_SIZE) -
+				 skb->data);
+		}
+
+		paddr = dma_map_single(ab->dev, skb->data,
+				       skb->len + skb_tailroom(skb),
+				       DMA_FROM_DEVICE);
+		if (dma_mapping_error(ab->dev, paddr))
+			goto fail_free_skb;
+
+		spin_lock_bh(&rx_ring->idr_lock);
+		buf_id = idr_alloc(&rx_ring->bufs_idr, skb, 0,
+				   rx_ring->bufs_max * 3, gfp);
+		spin_unlock_bh(&rx_ring->idr_lock);
+		if (buf_id < 0)
+			goto fail_dma_unmap;
+
+		desc = ath11k_hal_srng_src_get_next_entry(ab, srng);
+		if (!desc)
+			goto fail_idr_remove;
+
+		ATH11K_SKB_RXCB(skb)->paddr = paddr;
+
+		cookie = FIELD_PREP(DP_RXDMA_BUF_COOKIE_PDEV_ID, mac_id) |
+			 FIELD_PREP(DP_RXDMA_BUF_COOKIE_BUF_ID, buf_id);
+
+		num_remain--;
+
+		ath11k_hal_rx_buf_addr_info_set(desc, paddr, cookie, mgr);
+	}
+
+	ath11k_hal_srng_access_end(ab, srng);
+
+	spin_unlock_bh(&srng->lock);
+
+	return req_entries - num_remain;
+
+fail_idr_remove:
+	spin_lock_bh(&rx_ring->idr_lock);
+	idr_remove(&rx_ring->bufs_idr, buf_id);
+	spin_unlock_bh(&rx_ring->idr_lock);
+fail_dma_unmap:
+	dma_unmap_single(ab->dev, paddr, skb->len + skb_tailroom(skb),
+			 DMA_FROM_DEVICE);
+fail_free_skb:
+	dev_kfree_skb_any(skb);
+
+	ath11k_hal_srng_access_end(ab, srng);
+
+	spin_unlock_bh(&srng->lock);
+
+	return req_entries - num_remain;
+}
+
+static int ath11k_dp_rxdma_buf_ring_free(struct ath11k *ar,
+					 struct dp_rxdma_ring *rx_ring)
+{
+	struct ath11k_pdev_dp *dp = &ar->dp;
+	struct sk_buff *skb;
+	int buf_id;
+
+	spin_lock_bh(&rx_ring->idr_lock);
+	idr_for_each_entry(&rx_ring->bufs_idr, skb, buf_id) {
+		idr_remove(&rx_ring->bufs_idr, buf_id);
+		/* TODO: Understand where internal driver does this dma_unmap of
+		 * of rxdma_buffer.
+		 */
+		dma_unmap_single(ar->ab->dev, ATH11K_SKB_RXCB(skb)->paddr,
+				 skb->len + skb_tailroom(skb), DMA_FROM_DEVICE);
+		dev_kfree_skb_any(skb);
+	}
+
+	idr_destroy(&rx_ring->bufs_idr);
+	spin_unlock_bh(&rx_ring->idr_lock);
+
+	rx_ring = &dp->rx_mon_status_refill_ring;
+
+	spin_lock_bh(&rx_ring->idr_lock);
+	idr_for_each_entry(&rx_ring->bufs_idr, skb, buf_id) {
+		idr_remove(&rx_ring->bufs_idr, buf_id);
+		/* XXX: Understand where internal driver does this dma_unmap of
+		 * of rxdma_buffer.
+		 */
+		dma_unmap_single(ar->ab->dev, ATH11K_SKB_RXCB(skb)->paddr,
+				 skb->len + skb_tailroom(skb), DMA_BIDIRECTIONAL);
+		dev_kfree_skb_any(skb);
+	}
+
+	idr_destroy(&rx_ring->bufs_idr);
+	spin_unlock_bh(&rx_ring->idr_lock);
+	return 0;
+}
+
+static int ath11k_dp_rxdma_pdev_buf_free(struct ath11k *ar)
+{
+	struct ath11k_pdev_dp *dp = &ar->dp;
+	struct dp_rxdma_ring *rx_ring = &dp->rx_refill_buf_ring;
+
+	ath11k_dp_rxdma_buf_ring_free(ar, rx_ring);
+
+	rx_ring = &dp->rxdma_mon_buf_ring;
+	ath11k_dp_rxdma_buf_ring_free(ar, rx_ring);
+
+	rx_ring = &dp->rx_mon_status_refill_ring;
+	ath11k_dp_rxdma_buf_ring_free(ar, rx_ring);
+	return 0;
+}
+
+static int ath11k_dp_rxdma_ring_buf_setup(struct ath11k *ar,
+					  struct dp_rxdma_ring *rx_ring,
+					  u32 ringtype)
+{
+	struct ath11k_pdev_dp *dp = &ar->dp;
+	int num_entries;
+
+	num_entries = rx_ring->refill_buf_ring.size /
+		      ath11k_hal_srng_get_entrysize(ringtype);
+
+	rx_ring->bufs_max = num_entries;
+	ath11k_dp_rxbufs_replenish(ar->ab, dp->mac_id, rx_ring, num_entries,
+				   HAL_RX_BUF_RBM_SW3_BM, GFP_KERNEL);
+	return 0;
+}
+
+static int ath11k_dp_rxdma_pdev_buf_setup(struct ath11k *ar)
+{
+	struct ath11k_pdev_dp *dp = &ar->dp;
+	struct dp_rxdma_ring *rx_ring = &dp->rx_refill_buf_ring;
+
+	ath11k_dp_rxdma_ring_buf_setup(ar, rx_ring, HAL_RXDMA_BUF);
+
+	rx_ring = &dp->rxdma_mon_buf_ring;
+	ath11k_dp_rxdma_ring_buf_setup(ar, rx_ring, HAL_RXDMA_MONITOR_BUF);
+
+	rx_ring = &dp->rx_mon_status_refill_ring;
+	ath11k_dp_rxdma_ring_buf_setup(ar, rx_ring, HAL_RXDMA_MONITOR_STATUS);
+
+	return 0;
+}
+
+static void ath11k_dp_rx_pdev_srng_free(struct ath11k *ar)
+{
+	struct ath11k_pdev_dp *dp = &ar->dp;
+
+	ath11k_dp_srng_cleanup(ar->ab, &dp->rx_refill_buf_ring.refill_buf_ring);
+	ath11k_dp_srng_cleanup(ar->ab, &dp->reo_dst_ring);
+	ath11k_dp_srng_cleanup(ar->ab, &dp->rxdma_err_dst_ring);
+	ath11k_dp_srng_cleanup(ar->ab, &dp->rx_mon_status_refill_ring.refill_buf_ring);
+	ath11k_dp_srng_cleanup(ar->ab, &dp->rxdma_mon_buf_ring.refill_buf_ring);
+}
+
+static int ath11k_dp_rx_pdev_srng_alloc(struct ath11k *ar)
+{
+	struct ath11k_pdev_dp *dp = &ar->dp;
+	struct dp_srng *srng = NULL;
+	int ret;
+
+	ret = ath11k_dp_srng_setup(ar->ab,
+				   &dp->rx_refill_buf_ring.refill_buf_ring,
+				   HAL_RXDMA_BUF, 0,
+				   dp->mac_id, DP_RXDMA_BUF_RING_SIZE);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to setup rx_refill_buf_ring\n");
+		return ret;
+	}
+
+	ret = ath11k_dp_srng_setup(ar->ab, &dp->reo_dst_ring, HAL_REO_DST,
+				   dp->mac_id, dp->mac_id,
+				   DP_REO_DST_RING_SIZE);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to setup reo_dst_ring\n");
+		return ret;
+	}
+
+	ret = ath11k_dp_srng_setup(ar->ab, &dp->rxdma_err_dst_ring,
+				   HAL_RXDMA_DST, 0, dp->mac_id,
+				   DP_RXDMA_ERR_DST_RING_SIZE);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to setup rxdma_err_dst_ring\n");
+		return ret;
+	}
+
+	srng = &dp->rx_mon_status_refill_ring.refill_buf_ring;
+	ret = ath11k_dp_srng_setup(ar->ab,
+				   srng,
+				   HAL_RXDMA_MONITOR_STATUS, 0, dp->mac_id,
+				   DP_RXDMA_MON_STATUS_RING_SIZE);
+	if (ret) {
+		ath11k_warn(ar->ab,
+			    "failed to setup rx_mon_status_refill_ring\n");
+		return ret;
+	}
+	ret = ath11k_dp_srng_setup(ar->ab,
+				   &dp->rxdma_mon_buf_ring.refill_buf_ring,
+				   HAL_RXDMA_MONITOR_BUF, 0, dp->mac_id,
+				   DP_RXDMA_MONITOR_BUF_RING_SIZE);
+	if (ret) {
+		ath11k_warn(ar->ab,
+			    "failed to setup HAL_RXDMA_MONITOR_BUF\n");
+		return ret;
+	}
+
+	ret = ath11k_dp_srng_setup(ar->ab, &dp->rxdma_mon_dst_ring,
+				   HAL_RXDMA_MONITOR_DST, 0, dp->mac_id,
+				   DP_RXDMA_MONITOR_DST_RING_SIZE);
+	if (ret) {
+		ath11k_warn(ar->ab,
+			    "failed to setup HAL_RXDMA_MONITOR_DST\n");
+		return ret;
+	}
+
+	ret = ath11k_dp_srng_setup(ar->ab, &dp->rxdma_mon_desc_ring,
+				   HAL_RXDMA_MONITOR_DESC, 0, dp->mac_id,
+				   DP_RXDMA_MONITOR_DESC_RING_SIZE);
+	if (ret) {
+		ath11k_warn(ar->ab,
+			    "failed to setup HAL_RXDMA_MONITOR_DESC\n");
+		return ret;
+	}
+
+	return 0;
+}
+
+void ath11k_dp_reo_cmd_list_cleanup(struct ath11k_base *ab)
+{
+	struct ath11k_dp *dp = &ab->dp;
+	struct dp_reo_cmd *cmd, *tmp;
+	struct dp_reo_cache_flush_elem *cmd_cache, *tmp_cache;
+
+	spin_lock_bh(&dp->reo_cmd_lock);
+	list_for_each_entry_safe(cmd, tmp, &dp->reo_cmd_list, list) {
+		list_del(&cmd->list);
+		dma_unmap_single(ab->dev, cmd->data.paddr,
+				 cmd->data.size, DMA_BIDIRECTIONAL);
+		kfree(cmd->data.vaddr);
+		kfree(cmd);
+	}
+
+	list_for_each_entry_safe(cmd_cache, tmp_cache,
+				 &dp->reo_cmd_cache_flush_list, list) {
+		list_del(&cmd_cache->list);
+		dma_unmap_single(ab->dev, cmd_cache->data.paddr,
+				 cmd_cache->data.size, DMA_BIDIRECTIONAL);
+		kfree(cmd_cache->data.vaddr);
+		kfree(cmd_cache);
+	}
+	spin_unlock_bh(&dp->reo_cmd_lock);
+}
+
+static void ath11k_dp_reo_cmd_free(struct ath11k_dp *dp, void *ctx,
+				   enum hal_reo_cmd_status status)
+{
+	struct dp_rx_tid *rx_tid = ctx;
+
+	if (status != HAL_REO_CMD_SUCCESS)
+		ath11k_warn(dp->ab, "failed to flush rx tid hw desc, tid %d status %d\n",
+			    rx_tid->tid, status);
+
+	dma_unmap_single(dp->ab->dev, rx_tid->paddr, rx_tid->size,
+			 DMA_BIDIRECTIONAL);
+	kfree(rx_tid->vaddr);
+}
+
+static void ath11k_dp_reo_cache_flush(struct ath11k_base *ab,
+				      struct dp_rx_tid *rx_tid)
+{
+	struct ath11k_hal_reo_cmd cmd = {0};
+	unsigned long tot_desc_sz, desc_sz;
+	int ret;
+
+	tot_desc_sz = rx_tid->size;
+	desc_sz = ath11k_hal_reo_qdesc_size(0, HAL_DESC_REO_NON_QOS_TID);
+
+	while (tot_desc_sz > desc_sz) {
+		tot_desc_sz -= desc_sz;
+		cmd.addr_lo = lower_32_bits(rx_tid->paddr + tot_desc_sz);
+		cmd.addr_hi = upper_32_bits(rx_tid->paddr);
+		ret = ath11k_dp_tx_send_reo_cmd(ab, rx_tid,
+						HAL_REO_CMD_FLUSH_CACHE, &cmd,
+						NULL);
+		if (ret)
+			ath11k_warn(ab,
+				    "failed to send HAL_REO_CMD_FLUSH_CACHE, tid %d (%d)\n",
+				    rx_tid->tid, ret);
+	}
+
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.addr_lo = lower_32_bits(rx_tid->paddr);
+	cmd.addr_hi = upper_32_bits(rx_tid->paddr);
+	cmd.flag |= HAL_REO_CMD_FLG_NEED_STATUS;
+	ret = ath11k_dp_tx_send_reo_cmd(ab, rx_tid,
+					HAL_REO_CMD_FLUSH_CACHE,
+					&cmd, ath11k_dp_reo_cmd_free);
+	if (ret) {
+		ath11k_err(ab, "failed to send HAL_REO_CMD_FLUSH_CACHE cmd, tid %d (%d)\n",
+			   rx_tid->tid, ret);
+		dma_unmap_single(ab->dev, rx_tid->paddr, rx_tid->size,
+				 DMA_BIDIRECTIONAL);
+		kfree(rx_tid->vaddr);
+	}
+}
+
+static void ath11k_dp_rx_tid_del_func(struct ath11k_dp *dp, void *ctx,
+				      enum hal_reo_cmd_status status)
+{
+	struct ath11k_base *ab = dp->ab;
+	struct dp_rx_tid *rx_tid = ctx;
+	struct dp_reo_cache_flush_elem *elem, *tmp;
+
+	if (status == HAL_REO_CMD_DRAIN) {
+		goto free_desc;
+	} else if (status != HAL_REO_CMD_SUCCESS) {
+		/* Shouldn't happen! Cleanup in case of other failure? */
+		ath11k_warn(ab, "failed to delete rx tid %d hw descriptor %d\n",
+			    rx_tid->tid, status);
+		return;
+	}
+
+	elem = kzalloc(sizeof(*elem), GFP_ATOMIC);
+	if (!elem)
+		goto free_desc;
+
+	elem->ts = jiffies;
+	memcpy(&elem->data, rx_tid, sizeof(*rx_tid));
+
+	spin_lock_bh(&dp->reo_cmd_lock);
+	list_add_tail(&elem->list, &dp->reo_cmd_cache_flush_list);
+	spin_unlock_bh(&dp->reo_cmd_lock);
+
+	/* Flush and invalidate aged REO desc from HW cache */
+	spin_lock_bh(&dp->reo_cmd_lock);
+	list_for_each_entry_safe(elem, tmp, &dp->reo_cmd_cache_flush_list,
+				 list) {
+		if (time_after(jiffies, elem->ts +
+			       msecs_to_jiffies(DP_REO_DESC_FREE_TIMEOUT_MS))) {
+			list_del(&elem->list);
+			spin_unlock_bh(&dp->reo_cmd_lock);
+
+			ath11k_dp_reo_cache_flush(ab, &elem->data);
+			kfree(elem);
+			spin_lock_bh(&dp->reo_cmd_lock);
+		}
+	}
+	spin_unlock_bh(&dp->reo_cmd_lock);
+
+	return;
+free_desc:
+	dma_unmap_single(ab->dev, rx_tid->paddr, rx_tid->size,
+			 DMA_BIDIRECTIONAL);
+	kfree(rx_tid->vaddr);
+}
+
+static void ath11k_peer_rx_tid_delete(struct ath11k *ar,
+				      struct ath11k_peer *peer, u8 tid)
+{
+	struct ath11k_hal_reo_cmd cmd = {0};
+	struct dp_rx_tid *rx_tid = &peer->rx_tid[tid];
+	int ret;
+
+	if (!rx_tid->active)
+		return;
+
+	cmd.flag = HAL_REO_CMD_FLG_NEED_STATUS;
+	cmd.addr_lo = lower_32_bits(rx_tid->paddr);
+	cmd.addr_hi = upper_32_bits(rx_tid->paddr);
+	cmd.upd0 |= HAL_REO_CMD_UPD0_VLD;
+	ret = ath11k_dp_tx_send_reo_cmd(ar->ab, rx_tid,
+					HAL_REO_CMD_UPDATE_RX_QUEUE, &cmd,
+					ath11k_dp_rx_tid_del_func);
+	if (ret) {
+		ath11k_err(ar->ab, "failed to send HAL_REO_CMD_UPDATE_RX_QUEUE cmd, tid %d (%d)\n",
+			   tid, ret);
+		dma_unmap_single(ar->ab->dev, rx_tid->paddr, rx_tid->size,
+				 DMA_BIDIRECTIONAL);
+		kfree(rx_tid->vaddr);
+	}
+
+	rx_tid->active = false;
+}
+
+void ath11k_peer_rx_tid_cleanup(struct ath11k *ar, struct ath11k_peer *peer)
+{
+	int i;
+
+	for (i = 0; i <= IEEE80211_NUM_TIDS; i++)
+		ath11k_peer_rx_tid_delete(ar, peer, i);
+}
+
+static int ath11k_peer_rx_tid_reo_update(struct ath11k *ar,
+					 struct ath11k_peer *peer,
+					 struct dp_rx_tid *rx_tid,
+					 u32 ba_win_sz, u16 ssn)
+{
+	struct ath11k_hal_reo_cmd cmd = {0};
+	int ret;
+
+	cmd.addr_lo = lower_32_bits(rx_tid->paddr);
+	cmd.addr_hi = upper_32_bits(rx_tid->paddr);
+	cmd.flag = HAL_REO_CMD_FLG_NEED_STATUS;
+	cmd.upd0 = HAL_REO_CMD_UPD0_BA_WINDOW_SIZE |
+		   HAL_REO_CMD_UPD0_SSN;
+	cmd.ba_window_size = ba_win_sz;
+	cmd.upd2 = FIELD_PREP(HAL_REO_CMD_UPD2_SSN, ssn);
+
+	ret = ath11k_dp_tx_send_reo_cmd(ar->ab, rx_tid,
+					HAL_REO_CMD_UPDATE_RX_QUEUE, &cmd,
+					NULL);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to update rx tid queue, tid %d (%d)\n",
+			    rx_tid->tid, ret);
+		return ret;
+	}
+
+	rx_tid->ba_win_sz = ba_win_sz;
+
+	return 0;
+}
+
+static void ath11k_dp_rx_tid_mem_free(struct ath11k_base *ab,
+				      const u8 *peer_mac, int vdev_id, u8 tid)
+{
+	struct ath11k_peer *peer;
+	struct dp_rx_tid *rx_tid;
+
+	spin_lock_bh(&ab->base_lock);
+
+	peer = ath11k_peer_find(ab, vdev_id, peer_mac);
+	if (!peer) {
+		ath11k_warn(ab, "failed to find the peer to free up rx tid mem\n");
+		goto unlock_exit;
+	}
+
+	rx_tid = &peer->rx_tid[tid];
+	if (!rx_tid->active)
+		goto unlock_exit;
+
+	dma_unmap_single(ab->dev, rx_tid->paddr, rx_tid->size,
+			 DMA_BIDIRECTIONAL);
+	kfree(rx_tid->vaddr);
+
+	rx_tid->active = false;
+
+unlock_exit:
+	spin_unlock_bh(&ab->base_lock);
+}
+
+int ath11k_peer_rx_tid_setup(struct ath11k *ar, const u8 *peer_mac, int vdev_id,
+			     u8 tid, u32 ba_win_sz, u16 ssn)
+{
+	struct ath11k_base *ab = ar->ab;
+	struct ath11k_peer *peer;
+	struct dp_rx_tid *rx_tid;
+	u32 hw_desc_sz;
+	u32 *addr_aligned;
+	void *vaddr;
+	dma_addr_t paddr;
+	int ret;
+
+	spin_lock_bh(&ab->base_lock);
+
+	peer = ath11k_peer_find(ab, vdev_id, peer_mac);
+	if (!peer) {
+		ath11k_warn(ab, "failed to find the peer to set up rx tid\n");
+		spin_unlock_bh(&ab->base_lock);
+		return -ENOENT;
+	}
+
+	rx_tid = &peer->rx_tid[tid];
+	/* Update the tid queue if it is already setup */
+	if (rx_tid->active) {
+		paddr = rx_tid->paddr;
+		ret = ath11k_peer_rx_tid_reo_update(ar, peer, rx_tid,
+						    ba_win_sz, ssn);
+		spin_unlock_bh(&ab->base_lock);
+		if (ret) {
+			ath11k_warn(ab, "failed to update reo for rx tid %d\n", tid);
+			return ret;
+		}
+
+		ret = ath11k_wmi_peer_rx_reorder_queue_setup(ar, vdev_id,
+							     peer_mac, paddr,
+							     tid, 1, ba_win_sz);
+		if (ret)
+			ath11k_warn(ab, "failed to send wmi command to update rx reorder queue, tid :%d (%d)\n",
+				    tid, ret);
+		return ret;
+	}
+
+	rx_tid->tid = tid;
+
+	rx_tid->ba_win_sz = ba_win_sz;
+
+	/* TODO: Optimize the memory allocation for qos tid based on the
+	 * the actual BA window size in REO tid update path.
+	 */
+	if (tid == HAL_DESC_REO_NON_QOS_TID)
+		hw_desc_sz = ath11k_hal_reo_qdesc_size(ba_win_sz, tid);
+	else
+		hw_desc_sz = ath11k_hal_reo_qdesc_size(DP_BA_WIN_SZ_MAX, tid);
+
+	vaddr = kzalloc(hw_desc_sz + HAL_LINK_DESC_ALIGN - 1, GFP_KERNEL);
+	if (!vaddr) {
+		spin_unlock_bh(&ab->base_lock);
+		return -ENOMEM;
+	}
+
+	addr_aligned = PTR_ALIGN(vaddr, HAL_LINK_DESC_ALIGN);
+
+	ath11k_hal_reo_qdesc_setup(addr_aligned, tid, ba_win_sz, ssn);
+
+	paddr = dma_map_single(ab->dev, addr_aligned, hw_desc_sz,
+			       DMA_BIDIRECTIONAL);
+
+	ret = dma_mapping_error(ab->dev, paddr);
+	if (ret) {
+		spin_unlock_bh(&ab->base_lock);
+		goto err_mem_free;
+	}
+
+	rx_tid->vaddr = vaddr;
+	rx_tid->paddr = paddr;
+	rx_tid->size = hw_desc_sz;
+	rx_tid->active = true;
+
+	spin_unlock_bh(&ab->base_lock);
+
+	ret = ath11k_wmi_peer_rx_reorder_queue_setup(ar, vdev_id, peer_mac,
+						     paddr, tid, 1, ba_win_sz);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to setup rx reorder queue, tid :%d (%d)\n",
+			    tid, ret);
+		ath11k_dp_rx_tid_mem_free(ab, peer_mac, vdev_id, tid);
+	}
+
+	return ret;
+
+err_mem_free:
+	kfree(vaddr);
+
+	return ret;
+}
+
+int ath11k_dp_rx_ampdu_start(struct ath11k *ar,
+			     struct ieee80211_ampdu_params *params)
+{
+	struct ath11k_base *ab = ar->ab;
+	struct ath11k_sta *arsta = (void *)params->sta->drv_priv;
+	int vdev_id = arsta->arvif->vdev_id;
+	int ret;
+
+	ret = ath11k_peer_rx_tid_setup(ar, params->sta->addr, vdev_id,
+				       params->tid, params->buf_size,
+				       params->ssn);
+	if (ret)
+		ath11k_warn(ab, "failed to setup rx tid %d\n", ret);
+
+	return ret;
+}
+
+int ath11k_dp_rx_ampdu_stop(struct ath11k *ar,
+			    struct ieee80211_ampdu_params *params)
+{
+	struct ath11k_base *ab = ar->ab;
+	struct ath11k_peer *peer;
+	struct ath11k_sta *arsta = (void *)params->sta->drv_priv;
+	int vdev_id = arsta->arvif->vdev_id;
+	dma_addr_t paddr;
+	bool active;
+	int ret;
+
+	spin_lock_bh(&ab->base_lock);
+
+	peer = ath11k_peer_find(ab, vdev_id, params->sta->addr);
+	if (!peer) {
+		ath11k_warn(ab, "failed to find the peer to stop rx aggregation\n");
+		spin_unlock_bh(&ab->base_lock);
+		return -ENOENT;
+	}
+
+	paddr = peer->rx_tid[params->tid].paddr;
+	active = peer->rx_tid[params->tid].active;
+
+	ath11k_peer_rx_tid_delete(ar, peer, params->tid);
+
+	spin_unlock_bh(&ab->base_lock);
+
+	if (!active)
+		return 0;
+
+	ret = ath11k_wmi_peer_rx_reorder_queue_setup(ar, vdev_id,
+						     params->sta->addr, paddr,
+						     params->tid, 1, 1);
+	if (ret)
+		ath11k_warn(ab, "failed to send wmi to delete rx tid %d\n",
+			    ret);
+
+	return ret;
+}
+
+static int ath11k_get_ppdu_user_index(struct htt_ppdu_stats *ppdu_stats,
+				      u16 peer_id)
+{
+	int i;
+
+	for (i = 0; i < HTT_PPDU_STATS_MAX_USERS - 1; i++) {
+		if (ppdu_stats->user_stats[i].is_valid_peer_id) {
+			if (peer_id == ppdu_stats->user_stats[i].peer_id)
+				return i;
+		} else {
+			return i;
+		}
+	}
+
+	return -EINVAL;
+}
+
+static int ath11k_htt_tlv_ppdu_stats_parse(struct ath11k_base *ab,
+					   u16 tag, u16 len, const void *ptr,
+					   void *data)
+{
+	struct htt_ppdu_stats_info *ppdu_info;
+	struct htt_ppdu_user_stats *user_stats;
+	int cur_user;
+	u16 peer_id;
+
+	ppdu_info = (struct htt_ppdu_stats_info *)data;
+
+	switch (tag) {
+	case HTT_PPDU_STATS_TAG_COMMON:
+		if (len < sizeof(struct htt_ppdu_stats_common)) {
+			ath11k_warn(ab, "Invalid len %d for the tag 0x%x\n",
+				    len, tag);
+			return -EINVAL;
+		}
+		memcpy((void *)&ppdu_info->ppdu_stats.common, ptr,
+		       sizeof(struct htt_ppdu_stats_common));
+		break;
+	case HTT_PPDU_STATS_TAG_USR_RATE:
+		if (len < sizeof(struct htt_ppdu_stats_user_rate)) {
+			ath11k_warn(ab, "Invalid len %d for the tag 0x%x\n",
+				    len, tag);
+			return -EINVAL;
+		}
+
+		peer_id = ((struct htt_ppdu_stats_user_rate *)ptr)->sw_peer_id;
+		cur_user = ath11k_get_ppdu_user_index(&ppdu_info->ppdu_stats,
+						      peer_id);
+		if (cur_user < 0)
+			return -EINVAL;
+		user_stats = &ppdu_info->ppdu_stats.user_stats[cur_user];
+		user_stats->peer_id = peer_id;
+		user_stats->is_valid_peer_id = true;
+		memcpy((void *)&user_stats->rate, ptr,
+		       sizeof(struct htt_ppdu_stats_user_rate));
+		user_stats->tlv_flags |= BIT(tag);
+		break;
+	case HTT_PPDU_STATS_TAG_USR_COMPLTN_COMMON:
+		if (len < sizeof(struct htt_ppdu_stats_usr_cmpltn_cmn)) {
+			ath11k_warn(ab, "Invalid len %d for the tag 0x%x\n",
+				    len, tag);
+			return -EINVAL;
+		}
+
+		peer_id = ((struct htt_ppdu_stats_usr_cmpltn_cmn *)ptr)->sw_peer_id;
+		cur_user = ath11k_get_ppdu_user_index(&ppdu_info->ppdu_stats,
+						      peer_id);
+		if (cur_user < 0)
+			return -EINVAL;
+		user_stats = &ppdu_info->ppdu_stats.user_stats[cur_user];
+		user_stats->peer_id = peer_id;
+		user_stats->is_valid_peer_id = true;
+		memcpy((void *)&user_stats->cmpltn_cmn, ptr,
+		       sizeof(struct htt_ppdu_stats_usr_cmpltn_cmn));
+		user_stats->tlv_flags |= BIT(tag);
+		break;
+	case HTT_PPDU_STATS_TAG_USR_COMPLTN_ACK_BA_STATUS:
+		if (len <
+		    sizeof(struct htt_ppdu_stats_usr_cmpltn_ack_ba_status)) {
+			ath11k_warn(ab, "Invalid len %d for the tag 0x%x\n",
+				    len, tag);
+			return -EINVAL;
+		}
+
+		peer_id =
+		((struct htt_ppdu_stats_usr_cmpltn_ack_ba_status *)ptr)->sw_peer_id;
+		cur_user = ath11k_get_ppdu_user_index(&ppdu_info->ppdu_stats,
+						      peer_id);
+		if (cur_user < 0)
+			return -EINVAL;
+		user_stats = &ppdu_info->ppdu_stats.user_stats[cur_user];
+		user_stats->peer_id = peer_id;
+		user_stats->is_valid_peer_id = true;
+		memcpy((void *)&user_stats->ack_ba, ptr,
+		       sizeof(struct htt_ppdu_stats_usr_cmpltn_ack_ba_status));
+		user_stats->tlv_flags |= BIT(tag);
+		break;
+	}
+	return 0;
+}
+
+int ath11k_dp_htt_tlv_iter(struct ath11k_base *ab, const void *ptr, size_t len,
+			   int (*iter)(struct ath11k_base *ar, u16 tag, u16 len,
+				       const void *ptr, void *data),
+			   void *data)
+{
+	const struct htt_tlv *tlv;
+	const void *begin = ptr;
+	u16 tlv_tag, tlv_len;
+	int ret = -EINVAL;
+
+	while (len > 0) {
+		if (len < sizeof(*tlv)) {
+			ath11k_err(ab, "htt tlv parse failure at byte %zd (%zu bytes left, %zu expected)\n",
+				   ptr - begin, len, sizeof(*tlv));
+			return -EINVAL;
+		}
+		tlv = (struct htt_tlv *)ptr;
+		tlv_tag = FIELD_GET(HTT_TLV_TAG, tlv->header);
+		tlv_len = FIELD_GET(HTT_TLV_LEN, tlv->header);
+		ptr += sizeof(*tlv);
+		len -= sizeof(*tlv);
+
+		if (tlv_len > len) {
+			ath11k_err(ab, "htt tlv parse failure of tag %hhu at byte %zd (%zu bytes left, %hhu expected)\n",
+				   tlv_tag, ptr - begin, len, tlv_len);
+			return -EINVAL;
+		}
+		ret = iter(ab, tlv_tag, tlv_len, ptr, data);
+		if (ret == -ENOMEM)
+			return ret;
+
+		ptr += tlv_len;
+		len -= tlv_len;
+	}
+	return 0;
+}
+
+static u8 ath11k_bw_to_mac80211_bw(u8 bw)
+{
+	u8 ret = 0;
+
+	switch (bw) {
+	case ATH11K_BW_20:
+		ret = RATE_INFO_BW_20;
+		break;
+	case ATH11K_BW_40:
+		ret = RATE_INFO_BW_40;
+		break;
+	case ATH11K_BW_80:
+		ret = RATE_INFO_BW_80;
+		break;
+	case ATH11K_BW_160:
+		ret = RATE_INFO_BW_160;
+		break;
+	}
+
+	return ret;
+}
+
+static u32 ath11k_bw_to_mac80211_bwflags(u8 bw)
+{
+	u32 bwflags = 0;
+
+	switch (bw) {
+	case ATH11K_BW_40:
+		bwflags = IEEE80211_TX_RC_40_MHZ_WIDTH;
+		break;
+	case ATH11K_BW_80:
+		bwflags = IEEE80211_TX_RC_80_MHZ_WIDTH;
+		break;
+	case ATH11K_BW_160:
+		bwflags = IEEE80211_TX_RC_160_MHZ_WIDTH;
+		break;
+	}
+
+	return bwflags;
+}
+
+static void
+ath11k_update_per_peer_tx_stats(struct ath11k *ar,
+				struct htt_ppdu_stats *ppdu_stats, u8 user)
+{
+	struct ath11k_base *ab = ar->ab;
+	struct ath11k_peer *peer;
+	struct ieee80211_sta *sta;
+	struct ath11k_sta *arsta;
+	struct htt_ppdu_stats_user_rate *user_rate;
+	struct ieee80211_chanctx_conf *conf = NULL;
+	struct ath11k_per_peer_tx_stats *peer_stats = &ar->peer_tx_stats;
+	struct htt_ppdu_user_stats *usr_stats = &ppdu_stats->user_stats[user];
+	struct htt_ppdu_stats_common *common = &ppdu_stats->common;
+	int ret;
+	u8 flags, mcs, nss, bw, sgi, rate_idx = 0;
+	u32 succ_bytes = 0;
+	u16 rate = 0, succ_pkts = 0;
+	u32 tx_duration = 0;
+	bool is_ampdu = false;
+
+	if (!usr_stats)
+		return;
+
+	if (!(usr_stats->tlv_flags & BIT(HTT_PPDU_STATS_TAG_USR_RATE)))
+		return;
+
+	if (usr_stats->tlv_flags & BIT(HTT_PPDU_STATS_TAG_USR_COMPLTN_COMMON))
+		is_ampdu =
+			HTT_USR_CMPLTN_IS_AMPDU(usr_stats->cmpltn_cmn.flags);
+
+	if (usr_stats->tlv_flags &
+	    BIT(HTT_PPDU_STATS_TAG_USR_COMPLTN_ACK_BA_STATUS)) {
+		succ_bytes = usr_stats->ack_ba.success_bytes;
+		succ_pkts = FIELD_GET(HTT_PPDU_STATS_ACK_BA_INFO_NUM_MSDU_M,
+				      usr_stats->ack_ba.info);
+	}
+
+	if (common->fes_duration_us)
+		tx_duration = common->fes_duration_us;
+
+	user_rate = &usr_stats->rate;
+	flags = HTT_USR_RATE_PREAMBLE(user_rate->rate_flags);
+	bw = HTT_USR_RATE_BW(user_rate->rate_flags) - 2;
+	nss = HTT_USR_RATE_NSS(user_rate->rate_flags) + 1;
+	mcs = HTT_USR_RATE_MCS(user_rate->rate_flags);
+	sgi = HTT_USR_RATE_GI(user_rate->rate_flags);
+
+	/* Note: If host configured fixed rates and in some other special
+	 * cases, the broadcast/management frames are sent in different rates.
+	 * Firmware rate's control to be skipped for this?
+	 */
+
+	if (flags == WMI_RATE_PREAMBLE_VHT && mcs > 9) {
+		ath11k_warn(ab, "Invalid VHT mcs %hhd peer stats",  mcs);
+		return;
+	}
+
+	if (flags == WMI_RATE_PREAMBLE_HT && (mcs > 7 || nss < 1)) {
+		ath11k_warn(ab, "Invalid HT mcs %hhd nss %hhd peer stats",
+			    mcs, nss);
+		return;
+	}
+
+	if (flags == WMI_RATE_PREAMBLE_CCK || flags == WMI_RATE_PREAMBLE_OFDM) {
+		ret = ath11k_mac_hw_ratecode_to_legacy_rate(mcs,
+							    flags,
+							    &rate_idx,
+							    &rate);
+		if (ret < 0)
+			return;
+	}
+
+	rcu_read_lock();
+	spin_lock_bh(&ab->base_lock);
+	peer = ath11k_peer_find_by_id(ab, usr_stats->peer_id);
+
+	if (!peer || !peer->sta) {
+		spin_unlock_bh(&ab->base_lock);
+		rcu_read_unlock();
+		return;
+	}
+
+	sta = peer->sta;
+	arsta = (struct ath11k_sta *)sta->drv_priv;
+
+	memset(&arsta->txrate, 0, sizeof(arsta->txrate));
+	memset(&arsta->tx_info.status, 0, sizeof(arsta->tx_info.status));
+
+	switch (flags) {
+	case WMI_RATE_PREAMBLE_OFDM:
+		arsta->txrate.legacy = rate;
+		if (arsta->arvif && arsta->arvif->vif)
+			conf = rcu_dereference(arsta->arvif->vif->chanctx_conf);
+		if (conf && conf->def.chan->band == NL80211_BAND_5GHZ)
+			arsta->tx_info.status.rates[0].idx = rate_idx - 4;
+		break;
+	case WMI_RATE_PREAMBLE_CCK:
+		arsta->txrate.legacy = rate;
+		arsta->tx_info.status.rates[0].idx = rate_idx;
+		if (mcs > ATH11K_HW_RATE_CCK_LP_1M &&
+		    mcs <= ATH11K_HW_RATE_CCK_SP_2M)
+			arsta->tx_info.status.rates[0].flags |=
+					IEEE80211_TX_RC_USE_SHORT_PREAMBLE;
+		break;
+	case WMI_RATE_PREAMBLE_HT:
+		arsta->txrate.mcs = mcs + 8 * (nss - 1);
+		arsta->tx_info.status.rates[0].idx = arsta->txrate.mcs;
+		arsta->txrate.flags = RATE_INFO_FLAGS_MCS;
+		arsta->tx_info.status.rates[0].flags |= IEEE80211_TX_RC_MCS;
+		if (sgi) {
+			arsta->txrate.flags |= RATE_INFO_FLAGS_SHORT_GI;
+			arsta->tx_info.status.rates[0].flags |=
+					IEEE80211_TX_RC_SHORT_GI;
+		}
+		break;
+	case WMI_RATE_PREAMBLE_VHT:
+		arsta->txrate.mcs = mcs;
+		ieee80211_rate_set_vht(&arsta->tx_info.status.rates[0], mcs, nss);
+		arsta->txrate.flags = RATE_INFO_FLAGS_VHT_MCS;
+		arsta->tx_info.status.rates[0].flags |= IEEE80211_TX_RC_VHT_MCS;
+		if (sgi) {
+			arsta->txrate.flags |= RATE_INFO_FLAGS_SHORT_GI;
+			arsta->tx_info.status.rates[0].flags |=
+						IEEE80211_TX_RC_SHORT_GI;
+		}
+		break;
+	}
+
+	arsta->txrate.nss = nss;
+	arsta->txrate.bw = ath11k_bw_to_mac80211_bw(bw);
+	arsta->tx_info.status.rates[0].flags |= ath11k_bw_to_mac80211_bwflags(bw);
+
+	memcpy(&arsta->last_txrate, &arsta->txrate, sizeof(struct rate_info));
+
+	if (succ_pkts) {
+		arsta->tx_info.flags = IEEE80211_TX_STAT_ACK;
+		arsta->tx_info.status.rates[0].count = 1;
+		ieee80211_tx_rate_update(ar->hw, sta, &arsta->tx_info);
+	}
+
+	memset(peer_stats, 0, sizeof(*peer_stats));
+
+	peer_stats->succ_pkts = succ_pkts;
+	peer_stats->succ_bytes = succ_bytes;
+	peer_stats->is_ampdu = is_ampdu;
+	peer_stats->duration = tx_duration;
+	peer_stats->ba_fails =
+		HTT_USR_CMPLTN_LONG_RETRY(usr_stats->cmpltn_cmn.flags) +
+		HTT_USR_CMPLTN_SHORT_RETRY(usr_stats->cmpltn_cmn.flags);
+
+	if (ath11k_debug_is_extd_tx_stats_enabled(ar))
+		ath11k_accumulate_per_peer_tx_stats(arsta,
+						    peer_stats, rate_idx);
+
+	spin_unlock_bh(&ab->base_lock);
+	rcu_read_unlock();
+}
+
+static void ath11k_htt_update_ppdu_stats(struct ath11k *ar,
+					 struct htt_ppdu_stats *ppdu_stats)
+{
+	u8 user;
+
+	for (user = 0; user < HTT_PPDU_STATS_MAX_USERS - 1; user++)
+		ath11k_update_per_peer_tx_stats(ar, ppdu_stats, user);
+}
+
+static
+struct htt_ppdu_stats_info *ath11k_dp_htt_get_ppdu_desc(struct ath11k *ar,
+							u32 ppdu_id)
+{
+	struct htt_ppdu_stats_info *ppdu_info = NULL;
+
+	spin_lock_bh(&ar->data_lock);
+	if (!list_empty(&ar->ppdu_stats_info)) {
+		list_for_each_entry(ppdu_info, &ar->ppdu_stats_info, list) {
+			if (ppdu_info && ppdu_info->ppdu_id == ppdu_id) {
+				spin_unlock_bh(&ar->data_lock);
+				return ppdu_info;
+			}
+		}
+
+		if (ar->ppdu_stat_list_depth > HTT_PPDU_DESC_MAX_DEPTH) {
+			ppdu_info = list_first_entry(&ar->ppdu_stats_info,
+						     typeof(*ppdu_info), list);
+			list_del(&ppdu_info->list);
+			ar->ppdu_stat_list_depth--;
+			ath11k_htt_update_ppdu_stats(ar, &ppdu_info->ppdu_stats);
+			kfree(ppdu_info);
+		}
+	}
+	spin_unlock_bh(&ar->data_lock);
+
+	ppdu_info = kzalloc(sizeof(*ppdu_info), GFP_KERNEL);
+	if (!ppdu_info)
+		return NULL;
+
+	spin_lock_bh(&ar->data_lock);
+	list_add_tail(&ppdu_info->list, &ar->ppdu_stats_info);
+	ar->ppdu_stat_list_depth++;
+	spin_unlock_bh(&ar->data_lock);
+
+	return ppdu_info;
+}
+
+static int ath11k_htt_pull_ppdu_stats(struct ath11k_base *ab,
+				      struct sk_buff *skb)
+{
+	struct ath11k_htt_ppdu_stats_msg *msg;
+	struct htt_ppdu_stats_info *ppdu_info;
+	struct ath11k *ar;
+	int ret;
+	u8 pdev_id;
+	u32 ppdu_id, len;
+
+	msg = (struct ath11k_htt_ppdu_stats_msg *)skb->data;
+	len = FIELD_GET(HTT_T2H_PPDU_STATS_INFO_PAYLOAD_SIZE, msg->info);
+	pdev_id = FIELD_GET(HTT_T2H_PPDU_STATS_INFO_PDEV_ID, msg->info);
+	ppdu_id = msg->ppdu_id;
+
+	rcu_read_lock();
+	ar = ath11k_mac_get_ar_by_pdev_id(ab, pdev_id);
+	if (!ar) {
+		ret = -EINVAL;
+		goto exit;
+	}
+
+	if (ath11k_debug_is_pktlog_lite_mode_enabled(ar))
+		trace_ath11k_htt_ppdu_stats(ar, skb->data, len);
+
+	ppdu_info = ath11k_dp_htt_get_ppdu_desc(ar, ppdu_id);
+	if (!ppdu_info) {
+		ret = -EINVAL;
+		goto exit;
+	}
+
+	ppdu_info->ppdu_id = ppdu_id;
+	ret = ath11k_dp_htt_tlv_iter(ab, msg->data, len,
+				     ath11k_htt_tlv_ppdu_stats_parse,
+				     (void *)ppdu_info);
+	if (ret) {
+		ath11k_warn(ab, "Failed to parse tlv %d\n", ret);
+		goto exit;
+	}
+
+exit:
+	rcu_read_unlock();
+
+	return ret;
+}
+
+static void ath11k_htt_pktlog(struct ath11k_base *ab, struct sk_buff *skb)
+{
+	struct htt_pktlog_msg *data = (struct htt_pktlog_msg *)skb->data;
+	struct ath11k *ar;
+	u32 len;
+	u8 pdev_id;
+
+	len = FIELD_GET(HTT_T2H_PPDU_STATS_INFO_PAYLOAD_SIZE, data->hdr);
+
+	if (len > ATH11K_HTT_PKTLOG_MAX_SIZE) {
+		ath11k_warn(ab, "htt pktlog buffer size %d, expected < %d\n",
+			    len,
+			    ATH11K_HTT_PKTLOG_MAX_SIZE);
+		return;
+	}
+
+	pdev_id = FIELD_GET(HTT_T2H_PPDU_STATS_INFO_PDEV_ID, data->hdr);
+	pdev_id = DP_HW2SW_MACID(pdev_id);
+	ar = ab->pdevs[pdev_id].ar;
+
+	trace_ath11k_htt_pktlog(ar, data->payload, len);
+}
+
+void ath11k_dp_htt_htc_t2h_msg_handler(struct ath11k_base *ab,
+				       struct sk_buff *skb)
+{
+	struct ath11k_dp *dp = &ab->dp;
+	struct htt_resp_msg *resp = (struct htt_resp_msg *)skb->data;
+	enum htt_t2h_msg_type type = FIELD_GET(HTT_T2H_MSG_TYPE, *(u32 *)resp);
+	u16 peer_id;
+	u8 vdev_id;
+	u8 mac_addr[ETH_ALEN];
+	u16 peer_mac_h16;
+	u16 ast_hash;
+
+	ath11k_dbg(ab, ATH11K_DBG_DP_HTT, "dp_htt rx msg type :0x%0x\n", type);
+
+	switch (type) {
+	case HTT_T2H_MSG_TYPE_VERSION_CONF:
+		dp->htt_tgt_ver_major = FIELD_GET(HTT_T2H_VERSION_CONF_MAJOR,
+						  resp->version_msg.version);
+		dp->htt_tgt_ver_minor = FIELD_GET(HTT_T2H_VERSION_CONF_MINOR,
+						  resp->version_msg.version);
+		complete(&dp->htt_tgt_version_received);
+		break;
+	case HTT_T2H_MSG_TYPE_PEER_MAP:
+		vdev_id = FIELD_GET(HTT_T2H_PEER_MAP_INFO_VDEV_ID,
+				    resp->peer_map_ev.info);
+		peer_id = FIELD_GET(HTT_T2H_PEER_MAP_INFO_PEER_ID,
+				    resp->peer_map_ev.info);
+		peer_mac_h16 = FIELD_GET(HTT_T2H_PEER_MAP_INFO1_MAC_ADDR_H16,
+					 resp->peer_map_ev.info1);
+		ath11k_dp_get_mac_addr(resp->peer_map_ev.mac_addr_l32,
+				       peer_mac_h16, mac_addr);
+		ast_hash = FIELD_GET(HTT_T2H_PEER_MAP_INFO2_AST_HASH_VAL,
+				     resp->peer_map_ev.info1);
+		ath11k_peer_map_event(ab, vdev_id, peer_id, mac_addr, ast_hash);
+		break;
+	case HTT_T2H_MSG_TYPE_PEER_UNMAP:
+		peer_id = FIELD_GET(HTT_T2H_PEER_UNMAP_INFO_PEER_ID,
+				    resp->peer_unmap_ev.info);
+		ath11k_peer_unmap_event(ab, peer_id);
+		break;
+	case HTT_T2H_MSG_TYPE_PPDU_STATS_IND:
+		ath11k_htt_pull_ppdu_stats(ab, skb);
+		break;
+	case HTT_T2H_MSG_TYPE_EXT_STATS_CONF:
+		ath11k_dbg_htt_ext_stats_handler(ab, skb);
+		break;
+	case HTT_T2H_MSG_TYPE_PKTLOG:
+		ath11k_htt_pktlog(ab, skb);
+		break;
+	default:
+		ath11k_warn(ab, "htt event %d not handled\n", type);
+		break;
+	}
+
+	dev_kfree_skb_any(skb);
+}
+
+static int ath11k_dp_rx_msdu_coalesce(struct ath11k *ar,
+				      struct sk_buff_head *msdu_list,
+				      struct sk_buff *first, struct sk_buff *last,
+				      u8 l3pad_bytes, int msdu_len)
+{
+	struct sk_buff *skb;
+	struct ath11k_skb_rxcb *rxcb = ATH11K_SKB_RXCB(first);
+	struct hal_rx_desc *ldesc;
+	int space_extra;
+	int rem_len;
+	int buf_len;
+
+	if (!rxcb->is_continuation) {
+		skb_put(first, HAL_RX_DESC_SIZE + l3pad_bytes + msdu_len);
+		skb_pull(first, HAL_RX_DESC_SIZE + l3pad_bytes);
+		return 0;
+	}
+
+	if (WARN_ON_ONCE(msdu_len <= (DP_RX_BUFFER_SIZE -
+			 (HAL_RX_DESC_SIZE + l3pad_bytes)))) {
+		skb_put(first, HAL_RX_DESC_SIZE + l3pad_bytes + msdu_len);
+		skb_pull(first, HAL_RX_DESC_SIZE + l3pad_bytes);
+		return 0;
+	}
+
+	ldesc = (struct hal_rx_desc *)last->data;
+	rxcb->is_first_msdu = ath11k_dp_rx_h_msdu_end_first_msdu(ldesc);
+	rxcb->is_last_msdu = ath11k_dp_rx_h_msdu_end_last_msdu(ldesc);
+
+	/* MSDU spans over multiple buffers because the length of the MSDU
+	 * exceeds DP_RX_BUFFER_SIZE - HAL_RX_DESC_SIZE. So assume the data
+	 * in the first buf is of length DP_RX_BUFFER_SIZE - HAL_RX_DESC_SIZE.
+	 */
+	skb_put(first, DP_RX_BUFFER_SIZE);
+	skb_pull(first, HAL_RX_DESC_SIZE + l3pad_bytes);
+
+	space_extra = msdu_len - (DP_RX_BUFFER_SIZE + skb_tailroom(first));
+	if (space_extra > 0 &&
+	    (pskb_expand_head(first, 0, space_extra, GFP_ATOMIC) < 0)) {
+		/* Free up all buffers of the MSDU */
+		while ((skb = __skb_dequeue(msdu_list)) != NULL) {
+			rxcb = ATH11K_SKB_RXCB(skb);
+			if (!rxcb->is_continuation) {
+				dev_kfree_skb_any(skb);
+				break;
+			}
+			dev_kfree_skb_any(skb);
+		}
+		return -ENOMEM;
+	}
+
+	/* When an MSDU spread over multiple buffers attention, MSDU_END and
+	 * MPDU_END tlvs are valid only in the last buffer. Copy those tlvs.
+	 */
+	ath11k_dp_rx_desc_end_tlv_copy(rxcb->rx_desc, ldesc);
+
+	rem_len = msdu_len -
+		  (DP_RX_BUFFER_SIZE - HAL_RX_DESC_SIZE - l3pad_bytes);
+	while ((skb = __skb_dequeue(msdu_list)) != NULL && rem_len > 0) {
+		rxcb = ATH11K_SKB_RXCB(skb);
+		if (rxcb->is_continuation)
+			buf_len = DP_RX_BUFFER_SIZE - HAL_RX_DESC_SIZE;
+		else
+			buf_len = rem_len;
+
+		if (buf_len > (DP_RX_BUFFER_SIZE - HAL_RX_DESC_SIZE)) {
+			WARN_ON_ONCE(1);
+			dev_kfree_skb_any(skb);
+			return -EINVAL;
+		}
+
+		skb_put(skb, buf_len + HAL_RX_DESC_SIZE);
+		skb_pull(skb, HAL_RX_DESC_SIZE);
+		skb_copy_from_linear_data(skb, skb_put(first, buf_len),
+					  buf_len);
+		dev_kfree_skb_any(skb);
+
+		rem_len -= buf_len;
+		if (!rxcb->is_continuation)
+			break;
+	}
+
+	return 0;
+}
+
+static struct sk_buff *ath11k_dp_rx_get_msdu_last_buf(struct sk_buff_head *msdu_list,
+						      struct sk_buff *first)
+{
+	struct sk_buff *skb;
+	struct ath11k_skb_rxcb *rxcb = ATH11K_SKB_RXCB(first);
+
+	if (!rxcb->is_continuation)
+		return first;
+
+	skb_queue_walk(msdu_list, skb) {
+		rxcb = ATH11K_SKB_RXCB(skb);
+		if (!rxcb->is_continuation)
+			return skb;
+	}
+
+	return NULL;
+}
+
+static int ath11k_dp_rx_retrieve_amsdu(struct ath11k *ar,
+				       struct sk_buff_head *msdu_list,
+				       struct sk_buff_head *amsdu_list)
+{
+	struct sk_buff *msdu = skb_peek(msdu_list);
+	struct sk_buff *last_buf;
+	struct ath11k_skb_rxcb *rxcb;
+	struct ieee80211_hdr *hdr;
+	struct hal_rx_desc *rx_desc, *lrx_desc;
+	u16 msdu_len;
+	u8 l3_pad_bytes;
+	u8 *hdr_status;
+	int ret;
+
+	if (!msdu)
+		return -ENOENT;
+
+	rx_desc = (struct hal_rx_desc *)msdu->data;
+	hdr_status = ath11k_dp_rx_h_80211_hdr(rx_desc);
+	hdr = (struct ieee80211_hdr *)hdr_status;
+	/* Process only data frames */
+	if (!ieee80211_is_data(hdr->frame_control)) {
+		__skb_unlink(msdu, msdu_list);
+		dev_kfree_skb_any(msdu);
+		return -EINVAL;
+	}
+
+	do {
+		__skb_unlink(msdu, msdu_list);
+		last_buf = ath11k_dp_rx_get_msdu_last_buf(msdu_list, msdu);
+		if (!last_buf) {
+			ath11k_warn(ar->ab,
+				    "No valid Rx buffer to access Atten/MSDU_END/MPDU_END tlvs\n");
+			ret = -EIO;
+			goto free_out;
+		}
+
+		rx_desc = (struct hal_rx_desc *)msdu->data;
+		lrx_desc = (struct hal_rx_desc *)last_buf->data;
+
+		if (!ath11k_dp_rx_h_attn_msdu_done(lrx_desc)) {
+			ath11k_warn(ar->ab, "msdu_done bit in attention is not set\n");
+			ret = -EIO;
+			goto free_out;
+		}
+
+		rxcb = ATH11K_SKB_RXCB(msdu);
+		rxcb->rx_desc = rx_desc;
+		msdu_len = ath11k_dp_rx_h_msdu_start_msdu_len(rx_desc);
+		l3_pad_bytes = ath11k_dp_rx_h_msdu_end_l3pad(lrx_desc);
+
+		if (!rxcb->is_continuation) {
+			skb_put(msdu, HAL_RX_DESC_SIZE + l3_pad_bytes + msdu_len);
+			skb_pull(msdu, HAL_RX_DESC_SIZE + l3_pad_bytes);
+		} else {
+			ret = ath11k_dp_rx_msdu_coalesce(ar, msdu_list,
+							 msdu, last_buf,
+							 l3_pad_bytes, msdu_len);
+			if (ret) {
+				ath11k_warn(ar->ab,
+					    "failed to coalesce msdu rx buffer%d\n", ret);
+				goto free_out;
+			}
+		}
+		__skb_queue_tail(amsdu_list, msdu);
+
+		/* Should we also consider msdu_cnt from mpdu_meta while
+		 * preparing amsdu list?
+		 */
+		if (rxcb->is_last_msdu)
+			break;
+	} while ((msdu = skb_peek(msdu_list)) != NULL);
+
+	return 0;
+
+free_out:
+	dev_kfree_skb_any(msdu);
+	__skb_queue_purge(amsdu_list);
+
+	return ret;
+}
+
+static void ath11k_dp_rx_h_csum_offload(struct sk_buff *msdu)
+{
+	struct ath11k_skb_rxcb *rxcb = ATH11K_SKB_RXCB(msdu);
+	bool ip_csum_fail, l4_csum_fail;
+
+	ip_csum_fail = ath11k_dp_rx_h_attn_ip_cksum_fail(rxcb->rx_desc);
+	l4_csum_fail = ath11k_dp_rx_h_attn_l4_cksum_fail(rxcb->rx_desc);
+
+	msdu->ip_summed = (ip_csum_fail || l4_csum_fail) ?
+			  CHECKSUM_NONE : CHECKSUM_UNNECESSARY;
+}
+
+static int ath11k_dp_rx_crypto_mic_len(struct ath11k *ar,
+				       enum hal_encrypt_type enctype)
+{
+	switch (enctype) {
+	case HAL_ENCRYPT_TYPE_OPEN:
+	case HAL_ENCRYPT_TYPE_TKIP_NO_MIC:
+	case HAL_ENCRYPT_TYPE_TKIP_MIC:
+		return 0;
+	case HAL_ENCRYPT_TYPE_CCMP_128:
+		return IEEE80211_CCMP_MIC_LEN;
+	case HAL_ENCRYPT_TYPE_CCMP_256:
+		return IEEE80211_CCMP_256_MIC_LEN;
+	case HAL_ENCRYPT_TYPE_GCMP_128:
+	case HAL_ENCRYPT_TYPE_AES_GCMP_256:
+		return IEEE80211_GCMP_MIC_LEN;
+	case HAL_ENCRYPT_TYPE_WEP_40:
+	case HAL_ENCRYPT_TYPE_WEP_104:
+	case HAL_ENCRYPT_TYPE_WEP_128:
+	case HAL_ENCRYPT_TYPE_WAPI_GCM_SM4:
+	case HAL_ENCRYPT_TYPE_WAPI:
+		break;
+	}
+
+	ath11k_warn(ar->ab, "unsupported encryption type %d for mic len\n", enctype);
+	return 0;
+}
+
+static int ath11k_dp_rx_crypto_param_len(struct ath11k *ar,
+					 enum hal_encrypt_type enctype)
+{
+	switch (enctype) {
+	case HAL_ENCRYPT_TYPE_OPEN:
+		return 0;
+	case HAL_ENCRYPT_TYPE_TKIP_NO_MIC:
+	case HAL_ENCRYPT_TYPE_TKIP_MIC:
+		return IEEE80211_TKIP_IV_LEN;
+	case HAL_ENCRYPT_TYPE_CCMP_128:
+		return IEEE80211_CCMP_HDR_LEN;
+	case HAL_ENCRYPT_TYPE_CCMP_256:
+		return IEEE80211_CCMP_256_HDR_LEN;
+	case HAL_ENCRYPT_TYPE_GCMP_128:
+	case HAL_ENCRYPT_TYPE_AES_GCMP_256:
+		return IEEE80211_GCMP_HDR_LEN;
+	case HAL_ENCRYPT_TYPE_WEP_40:
+	case HAL_ENCRYPT_TYPE_WEP_104:
+	case HAL_ENCRYPT_TYPE_WEP_128:
+	case HAL_ENCRYPT_TYPE_WAPI_GCM_SM4:
+	case HAL_ENCRYPT_TYPE_WAPI:
+		break;
+	}
+
+	ath11k_warn(ar->ab, "unsupported encryption type %d\n", enctype);
+	return 0;
+}
+
+static int ath11k_dp_rx_crypto_icv_len(struct ath11k *ar,
+				       enum hal_encrypt_type enctype)
+{
+	switch (enctype) {
+	case HAL_ENCRYPT_TYPE_OPEN:
+	case HAL_ENCRYPT_TYPE_CCMP_128:
+	case HAL_ENCRYPT_TYPE_CCMP_256:
+	case HAL_ENCRYPT_TYPE_GCMP_128:
+	case HAL_ENCRYPT_TYPE_AES_GCMP_256:
+		return 0;
+	case HAL_ENCRYPT_TYPE_TKIP_NO_MIC:
+	case HAL_ENCRYPT_TYPE_TKIP_MIC:
+		return IEEE80211_TKIP_ICV_LEN;
+	case HAL_ENCRYPT_TYPE_WEP_40:
+	case HAL_ENCRYPT_TYPE_WEP_104:
+	case HAL_ENCRYPT_TYPE_WEP_128:
+	case HAL_ENCRYPT_TYPE_WAPI_GCM_SM4:
+	case HAL_ENCRYPT_TYPE_WAPI:
+		break;
+	}
+
+	ath11k_warn(ar->ab, "unsupported encryption type %d\n", enctype);
+	return 0;
+}
+
+static void ath11k_dp_rx_h_undecap_nwifi(struct ath11k *ar,
+					 struct sk_buff *msdu,
+					 u8 *first_hdr,
+					 enum hal_encrypt_type enctype,
+					 struct ieee80211_rx_status *status)
+{
+	struct ieee80211_hdr *hdr;
+	size_t hdr_len;
+	u8 da[ETH_ALEN];
+	u8 sa[ETH_ALEN];
+
+	/* pull decapped header and copy SA & DA */
+	hdr = (struct ieee80211_hdr *)msdu->data;
+	ether_addr_copy(da, ieee80211_get_DA(hdr));
+	ether_addr_copy(sa, ieee80211_get_SA(hdr));
+	skb_pull(msdu, ieee80211_hdrlen(hdr->frame_control));
+
+	/* push original 802.11 header */
+	hdr = (struct ieee80211_hdr *)first_hdr;
+	hdr_len = ieee80211_hdrlen(hdr->frame_control);
+
+	if (!(status->flag & RX_FLAG_IV_STRIPPED)) {
+		memcpy(skb_push(msdu,
+				ath11k_dp_rx_crypto_param_len(ar, enctype)),
+		       (void *)hdr + hdr_len,
+		       ath11k_dp_rx_crypto_param_len(ar, enctype));
+	}
+
+	memcpy(skb_push(msdu, hdr_len), hdr, hdr_len);
+
+	/* original 802.11 header has a different DA and in
+	 * case of 4addr it may also have different SA
+	 */
+	hdr = (struct ieee80211_hdr *)msdu->data;
+	ether_addr_copy(ieee80211_get_DA(hdr), da);
+	ether_addr_copy(ieee80211_get_SA(hdr), sa);
+}
+
+static void ath11k_dp_rx_h_undecap_raw(struct ath11k *ar, struct sk_buff *msdu,
+				       enum hal_encrypt_type enctype,
+				       struct ieee80211_rx_status *status,
+				       bool decrypted)
+{
+	struct ath11k_skb_rxcb *rxcb = ATH11K_SKB_RXCB(msdu);
+	struct ieee80211_hdr *hdr;
+	size_t hdr_len;
+	size_t crypto_len;
+
+	if (!rxcb->is_first_msdu ||
+	    !(rxcb->is_first_msdu && rxcb->is_last_msdu)) {
+		WARN_ON_ONCE(1);
+		return;
+	}
+
+	skb_trim(msdu, msdu->len - FCS_LEN);
+
+	if (!decrypted)
+		return;
+
+	hdr = (void *)msdu->data;
+
+	/* Tail */
+	if (status->flag & RX_FLAG_IV_STRIPPED) {
+		skb_trim(msdu, msdu->len -
+			 ath11k_dp_rx_crypto_mic_len(ar, enctype));
+
+		skb_trim(msdu, msdu->len -
+			 ath11k_dp_rx_crypto_icv_len(ar, enctype));
+	} else {
+		/* MIC */
+		if (status->flag & RX_FLAG_MIC_STRIPPED)
+			skb_trim(msdu, msdu->len -
+				 ath11k_dp_rx_crypto_mic_len(ar, enctype));
+
+		/* ICV */
+		if (status->flag & RX_FLAG_ICV_STRIPPED)
+			skb_trim(msdu, msdu->len -
+				 ath11k_dp_rx_crypto_icv_len(ar, enctype));
+	}
+
+	/* MMIC */
+	if ((status->flag & RX_FLAG_MMIC_STRIPPED) &&
+	    !ieee80211_has_morefrags(hdr->frame_control) &&
+	    enctype == HAL_ENCRYPT_TYPE_TKIP_MIC)
+		skb_trim(msdu, msdu->len - IEEE80211_CCMP_MIC_LEN);
+
+	/* Head */
+	if (status->flag & RX_FLAG_IV_STRIPPED) {
+		hdr_len = ieee80211_hdrlen(hdr->frame_control);
+		crypto_len = ath11k_dp_rx_crypto_param_len(ar, enctype);
+
+		memmove((void *)msdu->data + crypto_len,
+			(void *)msdu->data, hdr_len);
+		skb_pull(msdu, crypto_len);
+	}
+}
+
+static void *ath11k_dp_rx_h_find_rfc1042(struct ath11k *ar,
+					 struct sk_buff *msdu,
+					 enum hal_encrypt_type enctype)
+{
+	struct ath11k_skb_rxcb *rxcb = ATH11K_SKB_RXCB(msdu);
+	struct ieee80211_hdr *hdr;
+	size_t hdr_len, crypto_len;
+	void *rfc1042;
+	bool is_amsdu;
+
+	is_amsdu = !(rxcb->is_first_msdu && rxcb->is_last_msdu);
+	hdr = (struct ieee80211_hdr *)ath11k_dp_rx_h_80211_hdr(rxcb->rx_desc);
+	rfc1042 = hdr;
+
+	if (rxcb->is_first_msdu) {
+		hdr_len = ieee80211_hdrlen(hdr->frame_control);
+		crypto_len = ath11k_dp_rx_crypto_param_len(ar, enctype);
+
+		rfc1042 += hdr_len + crypto_len;
+	}
+
+	if (is_amsdu)
+		rfc1042 += sizeof(struct ath11k_dp_amsdu_subframe_hdr);
+
+	return rfc1042;
+}
+
+static void ath11k_dp_rx_h_undecap_eth(struct ath11k *ar,
+				       struct sk_buff *msdu,
+				       u8 *first_hdr,
+				       enum hal_encrypt_type enctype,
+				       struct ieee80211_rx_status *status)
+{
+	struct ieee80211_hdr *hdr;
+	struct ethhdr *eth;
+	size_t hdr_len;
+	u8 da[ETH_ALEN];
+	u8 sa[ETH_ALEN];
+	void *rfc1042;
+
+	rfc1042 = ath11k_dp_rx_h_find_rfc1042(ar, msdu, enctype);
+	if (WARN_ON_ONCE(!rfc1042))
+		return;
+
+	/* pull decapped header and copy SA & DA */
+	eth = (struct ethhdr *)msdu->data;
+	ether_addr_copy(da, eth->h_dest);
+	ether_addr_copy(sa, eth->h_source);
+	skb_pull(msdu, sizeof(struct ethhdr));
+
+	/* push rfc1042/llc/snap */
+	memcpy(skb_push(msdu, sizeof(struct ath11k_dp_rfc1042_hdr)), rfc1042,
+	       sizeof(struct ath11k_dp_rfc1042_hdr));
+
+	/* push original 802.11 header */
+	hdr = (struct ieee80211_hdr *)first_hdr;
+	hdr_len = ieee80211_hdrlen(hdr->frame_control);
+
+	if (!(status->flag & RX_FLAG_IV_STRIPPED)) {
+		memcpy(skb_push(msdu,
+				ath11k_dp_rx_crypto_param_len(ar, enctype)),
+		       (void *)hdr + hdr_len,
+		       ath11k_dp_rx_crypto_param_len(ar, enctype));
+	}
+
+	memcpy(skb_push(msdu, hdr_len), hdr, hdr_len);
+
+	/* original 802.11 header has a different DA and in
+	 * case of 4addr it may also have different SA
+	 */
+	hdr = (struct ieee80211_hdr *)msdu->data;
+	ether_addr_copy(ieee80211_get_DA(hdr), da);
+	ether_addr_copy(ieee80211_get_SA(hdr), sa);
+}
+
+static void ath11k_dp_rx_h_undecap(struct ath11k *ar, struct sk_buff *msdu,
+				   struct hal_rx_desc *rx_desc,
+				   enum hal_encrypt_type enctype,
+				   struct ieee80211_rx_status *status,
+				   bool decrypted)
+{
+	u8 *first_hdr;
+	u8 decap;
+
+	first_hdr = ath11k_dp_rx_h_80211_hdr(rx_desc);
+	decap = ath11k_dp_rx_h_mpdu_start_decap_type(rx_desc);
+
+	switch (decap) {
+	case DP_RX_DECAP_TYPE_NATIVE_WIFI:
+		ath11k_dp_rx_h_undecap_nwifi(ar, msdu, first_hdr,
+					     enctype, status);
+		break;
+	case DP_RX_DECAP_TYPE_RAW:
+		ath11k_dp_rx_h_undecap_raw(ar, msdu, enctype, status,
+					   decrypted);
+		break;
+	case DP_RX_DECAP_TYPE_ETHERNET2_DIX:
+		ath11k_dp_rx_h_undecap_eth(ar, msdu, first_hdr,
+					   enctype, status);
+		break;
+	case DP_RX_DECAP_TYPE_8023:
+		/* TODO: Handle undecap for these formats */
+		break;
+	}
+}
+
+static void ath11k_dp_rx_h_mpdu(struct ath11k *ar,
+				struct sk_buff_head *amsdu_list,
+				struct hal_rx_desc *rx_desc,
+				struct ieee80211_rx_status *rx_status)
+{
+	struct ieee80211_hdr *hdr;
+	enum hal_encrypt_type enctype;
+	struct sk_buff *last_msdu;
+	struct sk_buff *msdu;
+	struct ath11k_skb_rxcb *last_rxcb;
+	bool is_decrypted;
+	u32 err_bitmap;
+	u8 *qos;
+
+	if (skb_queue_empty(amsdu_list))
+		return;
+
+	hdr = (struct ieee80211_hdr *)ath11k_dp_rx_h_80211_hdr(rx_desc);
+
+	/* Each A-MSDU subframe will use the original header as the base and be
+	 * reported as a separate MSDU so strip the A-MSDU bit from QoS Ctl.
+	 */
+	if (ieee80211_is_data_qos(hdr->frame_control)) {
+		qos = ieee80211_get_qos_ctl(hdr);
+		qos[0] &= ~IEEE80211_QOS_CTL_A_MSDU_PRESENT;
+	}
+
+	is_decrypted = ath11k_dp_rx_h_attn_is_decrypted(rx_desc);
+	enctype = ath11k_dp_rx_h_mpdu_start_enctype(rx_desc);
+
+	/* Some attention flags are valid only in the last MSDU. */
+	last_msdu = skb_peek_tail(amsdu_list);
+	last_rxcb = ATH11K_SKB_RXCB(last_msdu);
+
+	err_bitmap = ath11k_dp_rx_h_attn_mpdu_err(last_rxcb->rx_desc);
+
+	/* Clear per-MPDU flags while leaving per-PPDU flags intact. */
+	rx_status->flag &= ~(RX_FLAG_FAILED_FCS_CRC |
+			     RX_FLAG_MMIC_ERROR |
+			     RX_FLAG_DECRYPTED |
+			     RX_FLAG_IV_STRIPPED |
+			     RX_FLAG_MMIC_STRIPPED);
+
+	if (err_bitmap & DP_RX_MPDU_ERR_FCS)
+		rx_status->flag |= RX_FLAG_FAILED_FCS_CRC;
+
+	if (err_bitmap & DP_RX_MPDU_ERR_TKIP_MIC)
+		rx_status->flag |= RX_FLAG_MMIC_ERROR;
+
+	if (is_decrypted)
+		rx_status->flag |= RX_FLAG_DECRYPTED | RX_FLAG_MMIC_STRIPPED |
+				   RX_FLAG_MIC_STRIPPED | RX_FLAG_ICV_STRIPPED;
+
+	skb_queue_walk(amsdu_list, msdu) {
+		ath11k_dp_rx_h_csum_offload(msdu);
+		ath11k_dp_rx_h_undecap(ar, msdu, rx_desc,
+				       enctype, rx_status, is_decrypted);
+	}
+}
+
+static void ath11k_dp_rx_h_rate(struct ath11k *ar, struct hal_rx_desc *rx_desc,
+				struct ieee80211_rx_status *rx_status)
+{
+	struct ieee80211_supported_band *sband;
+	enum rx_msdu_start_pkt_type pkt_type;
+	u8 bw;
+	u8 rate_mcs, nss;
+	u8 sgi;
+	bool is_cck;
+
+	pkt_type = ath11k_dp_rx_h_msdu_start_pkt_type(rx_desc);
+	bw = ath11k_dp_rx_h_msdu_start_rx_bw(rx_desc);
+	rate_mcs = ath11k_dp_rx_h_msdu_start_rate_mcs(rx_desc);
+	nss = ath11k_dp_rx_h_msdu_start_nss(rx_desc);
+	sgi = ath11k_dp_rx_h_msdu_start_sgi(rx_desc);
+
+	switch (pkt_type) {
+	case RX_MSDU_START_PKT_TYPE_11A:
+	case RX_MSDU_START_PKT_TYPE_11B:
+		is_cck = (pkt_type == RX_MSDU_START_PKT_TYPE_11B);
+		sband = &ar->mac.sbands[rx_status->band];
+		rx_status->rate_idx = ath11k_mac_hw_rate_to_idx(sband, rate_mcs,
+								is_cck);
+		break;
+	case RX_MSDU_START_PKT_TYPE_11N:
+		rx_status->encoding = RX_ENC_HT;
+		if (rate_mcs > ATH11K_HT_MCS_MAX) {
+			ath11k_warn(ar->ab,
+				    "Received with invalid mcs in HT mode %d\n",
+				     rate_mcs);
+			break;
+		}
+		rx_status->rate_idx = rate_mcs + (8 * (nss - 1));
+		if (sgi)
+			rx_status->enc_flags |= RX_ENC_FLAG_SHORT_GI;
+		rx_status->bw = ath11k_bw_to_mac80211_bw(bw);
+		break;
+	case RX_MSDU_START_PKT_TYPE_11AC:
+		rx_status->encoding = RX_ENC_VHT;
+		rx_status->rate_idx = rate_mcs;
+		if (rate_mcs > ATH11K_VHT_MCS_MAX) {
+			ath11k_warn(ar->ab,
+				    "Received with invalid mcs in VHT mode %d\n",
+				     rate_mcs);
+			break;
+		}
+		rx_status->nss = nss;
+		if (sgi)
+			rx_status->enc_flags |= RX_ENC_FLAG_SHORT_GI;
+		rx_status->bw = ath11k_bw_to_mac80211_bw(bw);
+		break;
+	case RX_MSDU_START_PKT_TYPE_11AX:
+		rx_status->rate_idx = rate_mcs;
+		if (rate_mcs > ATH11K_HE_MCS_MAX) {
+			ath11k_warn(ar->ab,
+				    "Received with invalid mcs in HE mode %d\n",
+				    rate_mcs);
+			break;
+		}
+		rx_status->encoding = RX_ENC_HE;
+		rx_status->nss = nss;
+		rx_status->bw = ath11k_bw_to_mac80211_bw(bw);
+		break;
+	}
+}
+
+static void ath11k_dp_rx_h_ppdu(struct ath11k *ar, struct hal_rx_desc *rx_desc,
+				struct ieee80211_rx_status *rx_status)
+{
+	u8 channel_num;
+
+	rx_status->freq = 0;
+	rx_status->rate_idx = 0;
+	rx_status->nss = 0;
+	rx_status->encoding = RX_ENC_LEGACY;
+	rx_status->bw = RATE_INFO_BW_20;
+
+	rx_status->flag |= RX_FLAG_NO_SIGNAL_VAL;
+
+	channel_num = ath11k_dp_rx_h_msdu_start_freq(rx_desc);
+
+	if (channel_num >= 1 && channel_num <= 14) {
+		rx_status->band = NL80211_BAND_2GHZ;
+	} else if (channel_num >= 36 && channel_num <= 173) {
+		rx_status->band = NL80211_BAND_5GHZ;
+	} else {
+		ath11k_warn(ar->ab, "Unsupported Channel info received %d\n",
+			    channel_num);
+		return;
+	}
+
+	rx_status->freq = ieee80211_channel_to_frequency(channel_num,
+							 rx_status->band);
+
+	ath11k_dp_rx_h_rate(ar, rx_desc, rx_status);
+}
+
+static void ath11k_dp_rx_process_amsdu(struct ath11k *ar,
+				       struct sk_buff_head *amsdu_list,
+				       struct ieee80211_rx_status *rx_status)
+{
+	struct sk_buff *first;
+	struct ath11k_skb_rxcb *rxcb;
+	struct hal_rx_desc *rx_desc;
+	bool first_mpdu;
+
+	if (skb_queue_empty(amsdu_list))
+		return;
+
+	first = skb_peek(amsdu_list);
+	rxcb = ATH11K_SKB_RXCB(first);
+	rx_desc = rxcb->rx_desc;
+
+	first_mpdu = ath11k_dp_rx_h_attn_first_mpdu(rx_desc);
+	if (first_mpdu)
+		ath11k_dp_rx_h_ppdu(ar, rx_desc, rx_status);
+
+	ath11k_dp_rx_h_mpdu(ar, amsdu_list, rx_desc, rx_status);
+}
+
+static char *ath11k_print_get_tid(struct ieee80211_hdr *hdr, char *out,
+				  size_t size)
+{
+	u8 *qc;
+	int tid;
+
+	if (!ieee80211_is_data_qos(hdr->frame_control))
+		return "";
+
+	qc = ieee80211_get_qos_ctl(hdr);
+	tid = *qc & IEEE80211_QOS_CTL_TID_MASK;
+	snprintf(out, size, "tid %d", tid);
+
+	return out;
+}
+
+static void ath11k_dp_rx_deliver_msdu(struct ath11k *ar, struct napi_struct *napi,
+				      struct sk_buff *msdu)
+{
+	struct ieee80211_rx_status *status;
+	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)msdu->data;
+	char tid[32];
+
+	status = IEEE80211_SKB_RXCB(msdu);
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_DATA,
+		   "rx skb %pK len %u peer %pM %s %s sn %u %s%s%s%s%s%s%s %srate_idx %u vht_nss %u freq %u band %u flag 0x%x fcs-err %i mic-err %i amsdu-more %i\n",
+		   msdu,
+		   msdu->len,
+		   ieee80211_get_SA(hdr),
+		   ath11k_print_get_tid(hdr, tid, sizeof(tid)),
+		   is_multicast_ether_addr(ieee80211_get_DA(hdr)) ?
+							"mcast" : "ucast",
+		   (__le16_to_cpu(hdr->seq_ctrl) & IEEE80211_SCTL_SEQ) >> 4,
+		   (status->encoding == RX_ENC_LEGACY) ? "legacy" : "",
+		   (status->encoding == RX_ENC_HT) ? "ht" : "",
+		   (status->encoding == RX_ENC_VHT) ? "vht" : "",
+		   (status->encoding == RX_ENC_HE) ? "he" : "",
+		   (status->bw == RATE_INFO_BW_40) ? "40" : "",
+		   (status->bw == RATE_INFO_BW_80) ? "80" : "",
+		   (status->bw == RATE_INFO_BW_160) ? "160" : "",
+		   status->enc_flags & RX_ENC_FLAG_SHORT_GI ? "sgi " : "",
+		   status->rate_idx,
+		   status->nss,
+		   status->freq,
+		   status->band, status->flag,
+		   !!(status->flag & RX_FLAG_FAILED_FCS_CRC),
+		   !!(status->flag & RX_FLAG_MMIC_ERROR),
+		   !!(status->flag & RX_FLAG_AMSDU_MORE));
+
+	/* TODO: trace rx packet */
+
+	ieee80211_rx_napi(ar->hw, NULL, msdu, napi);
+}
+
+static void ath11k_dp_rx_pre_deliver_amsdu(struct ath11k *ar,
+					   struct sk_buff_head *amsdu_list,
+					   struct ieee80211_rx_status *rxs)
+{
+	struct sk_buff *msdu;
+	struct sk_buff *first_subframe;
+	struct ieee80211_rx_status *status;
+
+	first_subframe = skb_peek(amsdu_list);
+
+	skb_queue_walk(amsdu_list, msdu) {
+		/* Setup per-MSDU flags */
+		if (skb_queue_empty(amsdu_list))
+			rxs->flag &= ~RX_FLAG_AMSDU_MORE;
+		else
+			rxs->flag |= RX_FLAG_AMSDU_MORE;
+
+		if (msdu == first_subframe) {
+			first_subframe = NULL;
+			rxs->flag &= ~RX_FLAG_ALLOW_SAME_PN;
+		} else {
+			rxs->flag |= RX_FLAG_ALLOW_SAME_PN;
+		}
+		rxs->flag |= RX_FLAG_SKIP_MONITOR;
+
+		status = IEEE80211_SKB_RXCB(msdu);
+		*status = *rxs;
+	}
+}
+
+static void ath11k_dp_rx_process_pending_packets(struct ath11k_base *ab,
+						 struct napi_struct *napi,
+						 struct sk_buff_head *pending_q,
+						 int *quota, u8 mac_id)
+{
+	struct ath11k *ar;
+	struct sk_buff *msdu;
+	struct ath11k_pdev *pdev;
+
+	if (skb_queue_empty(pending_q))
+		return;
+
+	ar = ab->pdevs[mac_id].ar;
+
+	rcu_read_lock();
+	pdev = rcu_dereference(ab->pdevs_active[mac_id]);
+
+	while (*quota && (msdu = __skb_dequeue(pending_q))) {
+		if (!pdev) {
+			dev_kfree_skb_any(msdu);
+			continue;
+		}
+
+		ath11k_dp_rx_deliver_msdu(ar, napi, msdu);
+		(*quota)--;
+	}
+	rcu_read_unlock();
+}
+
+int ath11k_dp_process_rx(struct ath11k_base *ab, int mac_id,
+			 struct napi_struct *napi, struct sk_buff_head *pending_q,
+			 int budget)
+{
+	struct ath11k *ar = ab->pdevs[mac_id].ar;
+	struct ath11k_pdev_dp *dp = &ar->dp;
+	struct ieee80211_rx_status *rx_status = &dp->rx_status;
+	struct dp_rxdma_ring *rx_ring = &dp->rx_refill_buf_ring;
+	struct hal_srng *srng;
+	struct hal_rx_meta_info meta_info;
+	struct sk_buff *msdu;
+	struct sk_buff_head msdu_list;
+	struct sk_buff_head amsdu_list;
+	struct ath11k_skb_rxcb *rxcb;
+	u32 *rx_desc;
+	int buf_id;
+	int num_buffs_reaped = 0;
+	int quota = budget;
+	int ret;
+	bool done = false;
+
+	/* Process any pending packets from the previous napi poll.
+	 * Note: All msdu's in this pending_q corresponds to the same mac id
+	 * due to pdev based reo dest mapping and also since each irq group id
+	 * maps to specific reo dest ring.
+	 */
+	ath11k_dp_rx_process_pending_packets(ab, napi, pending_q, &quota,
+					     mac_id);
+
+	/* If all quota is exhausted by processing the pending_q,
+	 * Wait for the next napi poll to reap the new info
+	 */
+	if (!quota)
+		goto exit;
+
+	__skb_queue_head_init(&msdu_list);
+
+	srng = &ab->hal.srng_list[dp->reo_dst_ring.ring_id];
+
+	spin_lock_bh(&srng->lock);
+
+	ath11k_hal_srng_access_begin(ab, srng);
+
+try_again:
+	while ((rx_desc = ath11k_hal_srng_dst_get_next_entry(ab, srng))) {
+		memset(&meta_info, 0, sizeof(meta_info));
+		ath11k_hal_rx_parse_dst_ring_desc(ab, rx_desc, &meta_info);
+
+		buf_id = FIELD_GET(DP_RXDMA_BUF_COOKIE_BUF_ID,
+				   meta_info.msdu_meta.cookie);
+		spin_lock_bh(&rx_ring->idr_lock);
+		msdu = idr_find(&rx_ring->bufs_idr, buf_id);
+		if (!msdu) {
+			ath11k_warn(ab, "frame rx with invalid buf_id %d\n",
+				    buf_id);
+			spin_unlock_bh(&rx_ring->idr_lock);
+			continue;
+		}
+
+		idr_remove(&rx_ring->bufs_idr, buf_id);
+		spin_unlock_bh(&rx_ring->idr_lock);
+
+		rxcb = ATH11K_SKB_RXCB(msdu);
+		dma_unmap_single(ab->dev, rxcb->paddr,
+				 msdu->len + skb_tailroom(msdu),
+				 DMA_FROM_DEVICE);
+
+		num_buffs_reaped++;
+
+		if (meta_info.push_reason !=
+		    HAL_REO_DEST_RING_PUSH_REASON_ROUTING_INSTRUCTION) {
+			/* TODO: Check if the msdu can be sent up for processing */
+			dev_kfree_skb_any(msdu);
+			ab->soc_stats.hal_reo_error[dp->reo_dst_ring.ring_id]++;
+			continue;
+		}
+
+		rxcb->is_first_msdu = meta_info.msdu_meta.first;
+		rxcb->is_last_msdu = meta_info.msdu_meta.last;
+		rxcb->is_continuation = meta_info.msdu_meta.continuation;
+		rxcb->mac_id = mac_id;
+		__skb_queue_tail(&msdu_list, msdu);
+
+		/* Stop reaping from the ring once quota is exhausted
+		 * and we've received all msdu's in the the AMSDU. The
+		 * additional msdu's reaped in excess of quota here would
+		 * be pushed into the pending queue to be processed during
+		 * the next napi poll.
+		 * Note: More profiling can be done to see the impact on
+		 * pending_q and throughput during various traffic & density
+		 * and how use of budget instead of remaining quota affects it.
+		 */
+		if (num_buffs_reaped >= quota && rxcb->is_last_msdu &&
+		    !rxcb->is_continuation) {
+			done = true;
+			break;
+		}
+	}
+
+	/* Hw might have updated the head pointer after we cached it.
+	 * In this case, even though there are entries in the ring we'll
+	 * get rx_desc NULL. Give the read another try with updated cached
+	 * head pointer so that we can reap complete MPDU in the current
+	 * rx processing.
+	 */
+	if (!done && ath11k_hal_srng_dst_num_free(ab, srng, true)) {
+		ath11k_hal_srng_access_end(ab, srng);
+		goto try_again;
+	}
+
+	ath11k_hal_srng_access_end(ab, srng);
+
+	spin_unlock_bh(&srng->lock);
+
+	if (!num_buffs_reaped)
+		goto exit;
+
+	/* Should we reschedule it later if we are not able to replenish all
+	 * the buffers?
+	 */
+	ath11k_dp_rxbufs_replenish(ab, mac_id, rx_ring, num_buffs_reaped,
+				   HAL_RX_BUF_RBM_SW3_BM, GFP_ATOMIC);
+
+	rcu_read_lock();
+	if (!rcu_dereference(ab->pdevs_active[mac_id])) {
+		__skb_queue_purge(&msdu_list);
+		goto rcu_unlock;
+	}
+
+	if (test_bit(ATH11K_CAC_RUNNING, &ar->dev_flags)) {
+		__skb_queue_purge(&msdu_list);
+		goto rcu_unlock;
+	}
+
+	while (!skb_queue_empty(&msdu_list)) {
+		__skb_queue_head_init(&amsdu_list);
+		ret = ath11k_dp_rx_retrieve_amsdu(ar, &msdu_list, &amsdu_list);
+		if (ret) {
+			if (ret == -EIO) {
+				ath11k_err(ab, "rx ring got corrupted %d\n", ret);
+				__skb_queue_purge(&msdu_list);
+				/* Should stop processing any more rx in
+				 * future from this ring?
+				 */
+				goto rcu_unlock;
+			}
+
+			/* A-MSDU retrieval got failed due to non-fatal condition,
+			 * continue processing with the next msdu.
+			 */
+			continue;
+		}
+
+		ath11k_dp_rx_process_amsdu(ar, &amsdu_list, rx_status);
+
+		ath11k_dp_rx_pre_deliver_amsdu(ar, &amsdu_list, rx_status);
+		skb_queue_splice_tail(&amsdu_list, pending_q);
+	}
+
+	while (quota && (msdu = __skb_dequeue(pending_q))) {
+		ath11k_dp_rx_deliver_msdu(ar, napi, msdu);
+		quota--;
+	}
+
+rcu_unlock:
+	rcu_read_unlock();
+exit:
+	return budget - quota;
+}
+
+static void ath11k_dp_rx_update_peer_stats(struct ath11k_sta *arsta,
+					   struct hal_rx_mon_ppdu_info *ppdu_info)
+{
+	struct ath11k_rx_peer_stats *rx_stats = arsta->rx_stats;
+	u32 num_msdu;
+
+	if (!rx_stats)
+		return;
+
+	num_msdu = ppdu_info->tcp_msdu_count + ppdu_info->tcp_ack_msdu_count +
+		   ppdu_info->udp_msdu_count + ppdu_info->other_msdu_count;
+
+	rx_stats->num_msdu += num_msdu;
+	rx_stats->tcp_msdu_count += ppdu_info->tcp_msdu_count +
+				    ppdu_info->tcp_ack_msdu_count;
+	rx_stats->udp_msdu_count += ppdu_info->udp_msdu_count;
+	rx_stats->other_msdu_count += ppdu_info->other_msdu_count;
+
+	if (ppdu_info->preamble_type == HAL_RX_PREAMBLE_11A ||
+	    ppdu_info->preamble_type == HAL_RX_PREAMBLE_11B) {
+		ppdu_info->nss = 1;
+		ppdu_info->mcs = HAL_RX_MAX_MCS;
+		ppdu_info->tid = IEEE80211_NUM_TIDS;
+	}
+
+	if (ppdu_info->nss > 0 && ppdu_info->nss <= HAL_RX_MAX_NSS)
+		rx_stats->nss_count[ppdu_info->nss - 1] += num_msdu;
+
+	if (ppdu_info->mcs <= HAL_RX_MAX_MCS)
+		rx_stats->mcs_count[ppdu_info->mcs] += num_msdu;
+
+	if (ppdu_info->gi < HAL_RX_GI_MAX)
+		rx_stats->gi_count[ppdu_info->gi] += num_msdu;
+
+	if (ppdu_info->bw < HAL_RX_BW_MAX)
+		rx_stats->bw_count[ppdu_info->bw] += num_msdu;
+
+	if (ppdu_info->ldpc < HAL_RX_SU_MU_CODING_MAX)
+		rx_stats->coding_count[ppdu_info->ldpc] += num_msdu;
+
+	if (ppdu_info->tid <= IEEE80211_NUM_TIDS)
+		rx_stats->tid_count[ppdu_info->tid] += num_msdu;
+
+	if (ppdu_info->preamble_type < HAL_RX_PREAMBLE_MAX)
+		rx_stats->pream_cnt[ppdu_info->preamble_type] += num_msdu;
+
+	if (ppdu_info->reception_type < HAL_RX_RECEPTION_TYPE_MAX)
+		rx_stats->reception_type[ppdu_info->reception_type] += num_msdu;
+
+	if (ppdu_info->is_stbc)
+		rx_stats->stbc_count += num_msdu;
+
+	if (ppdu_info->beamformed)
+		rx_stats->beamformed_count += num_msdu;
+
+	if (ppdu_info->num_mpdu_fcs_ok > 1)
+		rx_stats->ampdu_msdu_count += num_msdu;
+	else
+		rx_stats->non_ampdu_msdu_count += num_msdu;
+
+	rx_stats->num_mpdu_fcs_ok += ppdu_info->num_mpdu_fcs_ok;
+	rx_stats->num_mpdu_fcs_err += ppdu_info->num_mpdu_fcs_err;
+
+	arsta->rssi_comb = ppdu_info->rssi_comb;
+	rx_stats->rx_duration += ppdu_info->rx_duration;
+	arsta->rx_duration = rx_stats->rx_duration;
+}
+
+static struct sk_buff *ath11k_dp_rx_alloc_mon_status_buf(struct ath11k_base *ab,
+							 struct dp_rxdma_ring *rx_ring,
+							 int *buf_id, gfp_t gfp)
+{
+	struct sk_buff *skb;
+	dma_addr_t paddr;
+
+	skb = dev_alloc_skb(DP_RX_BUFFER_SIZE +
+			    DP_RX_BUFFER_ALIGN_SIZE);
+
+	if (!skb)
+		goto fail_alloc_skb;
+
+	if (!IS_ALIGNED((unsigned long)skb->data,
+			DP_RX_BUFFER_ALIGN_SIZE)) {
+		skb_pull(skb, PTR_ALIGN(skb->data, DP_RX_BUFFER_ALIGN_SIZE) -
+			 skb->data);
+	}
+
+	paddr = dma_map_single(ab->dev, skb->data,
+			       skb->len + skb_tailroom(skb),
+			       DMA_BIDIRECTIONAL);
+	if (unlikely(dma_mapping_error(ab->dev, paddr)))
+		goto fail_free_skb;
+
+	spin_lock_bh(&rx_ring->idr_lock);
+	*buf_id = idr_alloc(&rx_ring->bufs_idr, skb, 0,
+			    rx_ring->bufs_max, gfp);
+	spin_unlock_bh(&rx_ring->idr_lock);
+	if (*buf_id < 0)
+		goto fail_dma_unmap;
+
+	ATH11K_SKB_RXCB(skb)->paddr = paddr;
+	return skb;
+
+fail_dma_unmap:
+	dma_unmap_single(ab->dev, paddr, skb->len + skb_tailroom(skb),
+			 DMA_BIDIRECTIONAL);
+fail_free_skb:
+	dev_kfree_skb_any(skb);
+fail_alloc_skb:
+	return NULL;
+}
+
+int ath11k_dp_rx_mon_status_bufs_replenish(struct ath11k_base *ab, int mac_id,
+					   struct dp_rxdma_ring *rx_ring,
+					   int req_entries,
+					   enum hal_rx_buf_return_buf_manager mgr,
+					   gfp_t gfp)
+{
+	struct hal_srng *srng;
+	u32 *desc;
+	struct sk_buff *skb;
+	int num_free;
+	int num_remain;
+	int buf_id;
+	u32 cookie;
+	dma_addr_t paddr;
+
+	req_entries = min(req_entries, rx_ring->bufs_max);
+
+	srng = &ab->hal.srng_list[rx_ring->refill_buf_ring.ring_id];
+
+	spin_lock_bh(&srng->lock);
+
+	ath11k_hal_srng_access_begin(ab, srng);
+
+	num_free = ath11k_hal_srng_src_num_free(ab, srng, true);
+
+	req_entries = min(num_free, req_entries);
+	num_remain = req_entries;
+
+	while (num_remain > 0) {
+		skb = ath11k_dp_rx_alloc_mon_status_buf(ab, rx_ring,
+							&buf_id, gfp);
+		if (!skb)
+			break;
+		paddr = ATH11K_SKB_RXCB(skb)->paddr;
+
+		desc = ath11k_hal_srng_src_get_next_entry(ab, srng);
+		if (!desc)
+			goto fail_desc_get;
+
+		cookie = FIELD_PREP(DP_RXDMA_BUF_COOKIE_PDEV_ID, mac_id) |
+			 FIELD_PREP(DP_RXDMA_BUF_COOKIE_BUF_ID, buf_id);
+
+		num_remain--;
+
+		ath11k_hal_rx_buf_addr_info_set(desc, paddr, cookie, mgr);
+	}
+
+	ath11k_hal_srng_access_end(ab, srng);
+
+	spin_unlock_bh(&srng->lock);
+
+	return req_entries - num_remain;
+
+fail_desc_get:
+	spin_lock_bh(&rx_ring->idr_lock);
+	idr_remove(&rx_ring->bufs_idr, buf_id);
+	spin_unlock_bh(&rx_ring->idr_lock);
+	dma_unmap_single(ab->dev, paddr, skb->len + skb_tailroom(skb),
+			 DMA_BIDIRECTIONAL);
+	dev_kfree_skb_any(skb);
+	ath11k_hal_srng_access_end(ab, srng);
+	spin_unlock_bh(&srng->lock);
+
+	return req_entries - num_remain;
+}
+
+static int ath11k_dp_rx_reap_mon_status_ring(struct ath11k_base *ab, int mac_id,
+					     int *budget, struct sk_buff_head *skb_list)
+{
+	struct ath11k *ar = ab->pdevs[mac_id].ar;
+	struct ath11k_pdev_dp *dp = &ar->dp;
+	struct dp_rxdma_ring *rx_ring = &dp->rx_mon_status_refill_ring;
+	struct hal_srng *srng;
+	void *rx_mon_status_desc;
+	struct sk_buff *skb;
+	struct ath11k_skb_rxcb *rxcb;
+	struct hal_tlv_hdr *tlv;
+	u32 cookie;
+	int buf_id;
+	dma_addr_t paddr;
+	u8 rbm;
+	int num_buffs_reaped = 0;
+
+	srng = &ab->hal.srng_list[rx_ring->refill_buf_ring.ring_id];
+
+	spin_lock_bh(&srng->lock);
+
+	ath11k_hal_srng_access_begin(ab, srng);
+	while (*budget) {
+		*budget -= 1;
+		rx_mon_status_desc =
+			ath11k_hal_srng_src_peek(ab, srng);
+		if (!rx_mon_status_desc)
+			break;
+
+		ath11k_hal_rx_buf_addr_info_get(rx_mon_status_desc, &paddr,
+						&cookie, &rbm);
+		if (paddr) {
+			buf_id = FIELD_GET(DP_RXDMA_BUF_COOKIE_BUF_ID, cookie);
+
+			spin_lock_bh(&rx_ring->idr_lock);
+			skb = idr_find(&rx_ring->bufs_idr, buf_id);
+			if (!skb) {
+				ath11k_warn(ab, "rx monitor status with invalid buf_id %d\n",
+					    buf_id);
+				spin_unlock_bh(&rx_ring->idr_lock);
+				continue;
+			}
+
+			idr_remove(&rx_ring->bufs_idr, buf_id);
+			spin_unlock_bh(&rx_ring->idr_lock);
+
+			rxcb = ATH11K_SKB_RXCB(skb);
+
+			dma_sync_single_for_cpu(ab->dev, rxcb->paddr,
+						skb->len + skb_tailroom(skb),
+						DMA_FROM_DEVICE);
+
+			dma_unmap_single(ab->dev, rxcb->paddr,
+					 skb->len + skb_tailroom(skb),
+					 DMA_BIDIRECTIONAL);
+
+			tlv = (struct hal_tlv_hdr *)skb->data;
+			if (FIELD_GET(HAL_TLV_HDR_TAG, tlv->tl) !=
+					HAL_RX_STATUS_BUFFER_DONE) {
+				ath11k_hal_srng_src_get_next_entry(ab, srng);
+				continue;
+			}
+
+			__skb_queue_tail(skb_list, skb);
+		}
+
+		skb = ath11k_dp_rx_alloc_mon_status_buf(ab, rx_ring,
+							&buf_id, GFP_ATOMIC);
+
+		if (!skb) {
+			ath11k_hal_rx_buf_addr_info_set(rx_mon_status_desc, 0, 0,
+							HAL_RX_BUF_RBM_SW3_BM);
+			num_buffs_reaped++;
+			break;
+		}
+		rxcb = ATH11K_SKB_RXCB(skb);
+
+		cookie = FIELD_PREP(DP_RXDMA_BUF_COOKIE_PDEV_ID, mac_id) |
+			 FIELD_PREP(DP_RXDMA_BUF_COOKIE_BUF_ID, buf_id);
+
+		ath11k_hal_rx_buf_addr_info_set(rx_mon_status_desc, rxcb->paddr,
+						cookie, HAL_RX_BUF_RBM_SW3_BM);
+		ath11k_hal_srng_src_get_next_entry(ab, srng);
+		num_buffs_reaped++;
+	}
+	ath11k_hal_srng_access_end(ab, srng);
+	spin_unlock_bh(&srng->lock);
+
+	return num_buffs_reaped;
+}
+
+int ath11k_dp_rx_process_mon_status(struct ath11k_base *ab, int mac_id,
+				    struct napi_struct *napi, int budget)
+{
+	struct ath11k *ar = ab->pdevs[mac_id].ar;
+	enum hal_rx_mon_status hal_status;
+	struct sk_buff *skb;
+	struct sk_buff_head skb_list;
+	struct hal_rx_mon_ppdu_info ppdu_info;
+	struct ath11k_peer *peer;
+	struct ath11k_sta *arsta;
+	int num_buffs_reaped = 0;
+
+	__skb_queue_head_init(&skb_list);
+
+	num_buffs_reaped = ath11k_dp_rx_reap_mon_status_ring(ab, mac_id, &budget,
+							     &skb_list);
+	if (!num_buffs_reaped)
+		goto exit;
+
+	while ((skb = __skb_dequeue(&skb_list))) {
+		memset(&ppdu_info, 0, sizeof(ppdu_info));
+		ppdu_info.peer_id = HAL_INVALID_PEERID;
+
+		if (ath11k_debug_is_pktlog_rx_stats_enabled(ar))
+			trace_ath11k_htt_rxdesc(ar, skb->data, DP_RX_BUFFER_SIZE);
+
+		hal_status = ath11k_hal_rx_parse_mon_status(ab, &ppdu_info, skb);
+
+		if (ppdu_info.peer_id == HAL_INVALID_PEERID ||
+		    hal_status != HAL_RX_MON_STATUS_PPDU_DONE) {
+			dev_kfree_skb_any(skb);
+			continue;
+		}
+
+		rcu_read_lock();
+		spin_lock_bh(&ab->base_lock);
+		peer = ath11k_peer_find_by_id(ab, ppdu_info.peer_id);
+
+		if (!peer || !peer->sta) {
+			ath11k_warn(ab, "failed to find the peer with peer_id %d\n",
+				    ppdu_info.peer_id);
+			spin_unlock_bh(&ab->base_lock);
+			rcu_read_unlock();
+			dev_kfree_skb_any(skb);
+			continue;
+		}
+
+		arsta = (struct ath11k_sta *)peer->sta->drv_priv;
+		ath11k_dp_rx_update_peer_stats(arsta, &ppdu_info);
+
+		if (ath11k_debug_is_pktlog_peer_valid(ar, peer->addr))
+			trace_ath11k_htt_rxdesc(ar, skb->data, DP_RX_BUFFER_SIZE);
+
+		spin_unlock_bh(&ab->base_lock);
+		rcu_read_unlock();
+
+		dev_kfree_skb_any(skb);
+	}
+exit:
+	return num_buffs_reaped;
+}
+
+static int ath11k_dp_rx_link_desc_return(struct ath11k_base *ab,
+					 u32 *link_desc,
+					 enum hal_wbm_rel_bm_act action)
+{
+	struct ath11k_dp *dp = &ab->dp;
+	struct hal_srng *srng;
+	u32 *desc;
+	int ret = 0;
+
+	srng = &ab->hal.srng_list[dp->wbm_desc_rel_ring.ring_id];
+
+	spin_lock_bh(&srng->lock);
+
+	ath11k_hal_srng_access_begin(ab, srng);
+
+	desc = ath11k_hal_srng_src_get_next_entry(ab, srng);
+	if (!desc) {
+		ret = -ENOBUFS;
+		goto exit;
+	}
+
+	ath11k_hal_rx_msdu_link_desc_set(ab, (void *)desc, (void *)link_desc,
+					 action);
+
+exit:
+	ath11k_hal_srng_access_end(ab, srng);
+
+	spin_unlock_bh(&srng->lock);
+
+	return ret;
+}
+
+static void ath11k_dp_rx_frag_h_mpdu(struct ath11k *ar,
+				     struct sk_buff *msdu,
+				     struct hal_rx_desc *rx_desc,
+				     struct ieee80211_rx_status *rx_status)
+{
+	u8 rx_channel;
+	enum hal_encrypt_type enctype;
+	bool is_decrypted;
+	u32 err_bitmap;
+
+	is_decrypted = ath11k_dp_rx_h_attn_is_decrypted(rx_desc);
+	enctype = ath11k_dp_rx_h_mpdu_start_enctype(rx_desc);
+	err_bitmap = ath11k_dp_rx_h_attn_mpdu_err(rx_desc);
+
+	if (err_bitmap & DP_RX_MPDU_ERR_FCS)
+		rx_status->flag |= RX_FLAG_FAILED_FCS_CRC;
+
+	if (err_bitmap & DP_RX_MPDU_ERR_TKIP_MIC)
+		rx_status->flag |= RX_FLAG_MMIC_ERROR;
+
+	rx_status->encoding = RX_ENC_LEGACY;
+	rx_status->bw = RATE_INFO_BW_20;
+
+	rx_status->flag |= RX_FLAG_NO_SIGNAL_VAL;
+
+	rx_channel = ath11k_dp_rx_h_msdu_start_freq(rx_desc);
+
+	if (rx_channel >= 1 && rx_channel <= 14) {
+		rx_status->band = NL80211_BAND_2GHZ;
+	} else if (rx_channel >= 36 && rx_channel <= 173) {
+		rx_status->band = NL80211_BAND_5GHZ;
+	} else {
+		ath11k_warn(ar->ab, "Unsupported Channel info received %d\n",
+			    rx_channel);
+		return;
+	}
+
+	rx_status->freq = ieee80211_channel_to_frequency(rx_channel,
+							 rx_status->band);
+	ath11k_dp_rx_h_rate(ar, rx_desc, rx_status);
+
+	/* Rx fragments are received in raw mode */
+	skb_trim(msdu, msdu->len - FCS_LEN);
+
+	if (is_decrypted) {
+		rx_status->flag |= RX_FLAG_DECRYPTED | RX_FLAG_MIC_STRIPPED;
+		skb_trim(msdu, msdu->len -
+			 ath11k_dp_rx_crypto_mic_len(ar, enctype));
+	}
+}
+
+static int
+ath11k_dp_process_rx_err_buf(struct ath11k *ar, struct napi_struct *napi,
+			     int buf_id, bool frag)
+{
+	struct ath11k_pdev_dp *dp = &ar->dp;
+	struct dp_rxdma_ring *rx_ring = &dp->rx_refill_buf_ring;
+	struct ieee80211_rx_status rx_status = {0};
+	struct sk_buff *msdu;
+	struct ath11k_skb_rxcb *rxcb;
+	struct ieee80211_rx_status *status;
+	struct hal_rx_desc *rx_desc;
+	u16 msdu_len;
+
+	spin_lock_bh(&rx_ring->idr_lock);
+	msdu = idr_find(&rx_ring->bufs_idr, buf_id);
+	if (!msdu) {
+		ath11k_warn(ar->ab, "rx err buf with invalid buf_id %d\n",
+			    buf_id);
+		spin_unlock_bh(&rx_ring->idr_lock);
+		return -EINVAL;
+	}
+
+	idr_remove(&rx_ring->bufs_idr, buf_id);
+	spin_unlock_bh(&rx_ring->idr_lock);
+
+	rxcb = ATH11K_SKB_RXCB(msdu);
+	dma_unmap_single(ar->ab->dev, rxcb->paddr,
+			 msdu->len + skb_tailroom(msdu),
+			 DMA_FROM_DEVICE);
+
+	if (!frag) {
+		/* Process only rx fragments below, and drop
+		 * msdu's indicated due to error reasons.
+		 */
+		dev_kfree_skb_any(msdu);
+		return 0;
+	}
+
+	rcu_read_lock();
+	if (!rcu_dereference(ar->ab->pdevs_active[ar->pdev_idx])) {
+		dev_kfree_skb_any(msdu);
+		goto exit;
+	}
+
+	if (test_bit(ATH11K_CAC_RUNNING, &ar->dev_flags)) {
+		dev_kfree_skb_any(msdu);
+		goto exit;
+	}
+
+	rx_desc = (struct hal_rx_desc *)msdu->data;
+	msdu_len = ath11k_dp_rx_h_msdu_start_msdu_len(rx_desc);
+	skb_put(msdu, HAL_RX_DESC_SIZE + msdu_len);
+	skb_pull(msdu, HAL_RX_DESC_SIZE);
+
+	ath11k_dp_rx_frag_h_mpdu(ar, msdu, rx_desc, &rx_status);
+
+	status = IEEE80211_SKB_RXCB(msdu);
+
+	*status = rx_status;
+
+	ath11k_dp_rx_deliver_msdu(ar, napi, msdu);
+
+exit:
+	rcu_read_unlock();
+	return 0;
+}
+
+int ath11k_dp_process_rx_err(struct ath11k_base *ab, struct napi_struct *napi,
+			     int budget)
+{
+	struct hal_rx_msdu_meta meta[HAL_NUM_RX_MSDUS_PER_LINK_DESC];
+	struct dp_link_desc_bank *link_desc_banks;
+	enum hal_rx_buf_return_buf_manager rbm;
+	int tot_n_bufs_reaped, quota, ret, i;
+	int n_bufs_reaped[MAX_RADIOS] = {0};
+	struct hal_rx_meta_info meta_info;
+	struct dp_rxdma_ring *rx_ring;
+	struct dp_srng *reo_except;
+	u32 desc_bank, num_msdus;
+	struct hal_srng *srng;
+	struct ath11k_dp *dp;
+	void *link_desc_va;
+	int buf_id, mac_id;
+	struct ath11k *ar;
+	dma_addr_t paddr;
+	u32 *desc;
+	bool is_frag;
+
+	tot_n_bufs_reaped = 0;
+	quota = budget;
+
+	dp = &ab->dp;
+	reo_except = &dp->reo_except_ring;
+	link_desc_banks = dp->link_desc_banks;
+
+	srng = &ab->hal.srng_list[reo_except->ring_id];
+
+	spin_lock_bh(&srng->lock);
+
+	ath11k_hal_srng_access_begin(ab, srng);
+
+	while (budget &&
+	       (desc = ath11k_hal_srng_dst_get_next_entry(ab, srng))) {
+		ab->soc_stats.err_ring_pkts++;
+		ret = ath11k_hal_desc_reo_parse_err(ab, desc, &paddr,
+						    &desc_bank);
+		if (ret) {
+			ath11k_warn(ab, "failed to parse error reo desc %d\n",
+				    ret);
+			continue;
+		}
+		link_desc_va = link_desc_banks[desc_bank].vaddr +
+			       (paddr - link_desc_banks[desc_bank].paddr);
+		ath11k_hal_rx_msdu_link_info_get(link_desc_va, &num_msdus, meta,
+						 &rbm);
+		if (rbm != HAL_RX_BUF_RBM_WBM_IDLE_DESC_LIST &&
+		    rbm != HAL_RX_BUF_RBM_SW3_BM) {
+			ab->soc_stats.invalid_rbm++;
+			ath11k_warn(ab, "invalid return buffer manager %d\n", rbm);
+			ath11k_dp_rx_link_desc_return(ab, desc,
+						      HAL_WBM_REL_BM_ACT_REL_MSDU);
+			continue;
+		}
+
+		memset(&meta_info, 0, sizeof(meta_info));
+		ath11k_hal_rx_parse_dst_ring_desc(ab, desc, &meta_info);
+
+		is_frag = meta_info.mpdu_meta.frag;
+
+		/* Return the link desc back to wbm idle list */
+		ath11k_dp_rx_link_desc_return(ab, desc,
+					      HAL_WBM_REL_BM_ACT_PUT_IN_IDLE);
+
+		for (i = 0; i < num_msdus; i++) {
+			buf_id = FIELD_GET(DP_RXDMA_BUF_COOKIE_BUF_ID,
+					   meta[i].cookie);
+
+			mac_id = FIELD_GET(DP_RXDMA_BUF_COOKIE_PDEV_ID,
+					   meta[i].cookie);
+
+			ar = ab->pdevs[mac_id].ar;
+
+			if (!ath11k_dp_process_rx_err_buf(ar, napi, buf_id,
+							  is_frag)) {
+				n_bufs_reaped[mac_id]++;
+				tot_n_bufs_reaped++;
+			}
+		}
+
+		if (tot_n_bufs_reaped >= quota) {
+			tot_n_bufs_reaped = quota;
+			goto exit;
+		}
+
+		budget = quota - tot_n_bufs_reaped;
+	}
+
+exit:
+	ath11k_hal_srng_access_end(ab, srng);
+
+	spin_unlock_bh(&srng->lock);
+
+	for (i = 0; i <  ab->num_radios; i++) {
+		if (!n_bufs_reaped[i])
+			continue;
+
+		ar = ab->pdevs[i].ar;
+		rx_ring = &ar->dp.rx_refill_buf_ring;
+
+		ath11k_dp_rxbufs_replenish(ab, i, rx_ring, n_bufs_reaped[i],
+					   HAL_RX_BUF_RBM_SW3_BM, GFP_ATOMIC);
+	}
+
+	return tot_n_bufs_reaped;
+}
+
+static void ath11k_dp_rx_null_q_desc_sg_drop(struct ath11k *ar,
+					     int msdu_len,
+					     struct sk_buff_head *msdu_list)
+{
+	struct sk_buff *skb, *tmp;
+	struct ath11k_skb_rxcb *rxcb;
+	int n_buffs;
+
+	n_buffs = DIV_ROUND_UP(msdu_len,
+			       (DP_RX_BUFFER_SIZE - HAL_RX_DESC_SIZE));
+
+	skb_queue_walk_safe(msdu_list, skb, tmp) {
+		rxcb = ATH11K_SKB_RXCB(skb);
+		if (rxcb->err_rel_src == HAL_WBM_REL_SRC_MODULE_REO &&
+		    rxcb->err_code == HAL_REO_DEST_RING_ERROR_CODE_DESC_ADDR_ZERO) {
+			if (!n_buffs)
+				break;
+			__skb_unlink(skb, msdu_list);
+			dev_kfree_skb_any(skb);
+			n_buffs--;
+		}
+	}
+}
+
+static int ath11k_dp_rx_h_null_q_desc(struct ath11k *ar, struct sk_buff *msdu,
+				      struct ieee80211_rx_status *status,
+				      struct sk_buff_head *msdu_list)
+{
+	struct sk_buff_head amsdu_list;
+	u16 msdu_len;
+	struct hal_rx_desc *desc = (struct hal_rx_desc *)msdu->data;
+	u8 l3pad_bytes;
+	struct ath11k_skb_rxcb *rxcb = ATH11K_SKB_RXCB(msdu);
+
+	msdu_len = ath11k_dp_rx_h_msdu_start_msdu_len(desc);
+
+	if ((msdu_len + HAL_RX_DESC_SIZE) > DP_RX_BUFFER_SIZE) {
+		/* First buffer will be freed by the caller, so deduct it's length */
+		msdu_len = msdu_len - (DP_RX_BUFFER_SIZE - HAL_RX_DESC_SIZE);
+		ath11k_dp_rx_null_q_desc_sg_drop(ar, msdu_len, msdu_list);
+		return -EINVAL;
+	}
+
+	if (!ath11k_dp_rx_h_attn_msdu_done(desc)) {
+		ath11k_warn(ar->ab,
+			    "msdu_done bit not set in null_q_des processing\n");
+		__skb_queue_purge(msdu_list);
+		return -EIO;
+	}
+
+	/* Handle NULL queue descriptor violations arising out a missing
+	 * REO queue for a given peer or a given TID. This typically
+	 * may happen if a packet is received on a QOS enabled TID before the
+	 * ADDBA negotiation for that TID, when the TID queue is setup. Or
+	 * it may also happen for MC/BC frames if they are not routed to the
+	 * non-QOS TID queue, in the absence of any other default TID queue.
+	 * This error can show up both in a REO destination or WBM release ring.
+	 */
+
+	__skb_queue_head_init(&amsdu_list);
+
+	rxcb->is_first_msdu = ath11k_dp_rx_h_msdu_end_first_msdu(desc);
+	rxcb->is_last_msdu = ath11k_dp_rx_h_msdu_end_last_msdu(desc);
+
+	l3pad_bytes = ath11k_dp_rx_h_msdu_end_l3pad(desc);
+
+	if ((HAL_RX_DESC_SIZE + l3pad_bytes + msdu_len) > DP_RX_BUFFER_SIZE)
+		return -EINVAL;
+
+	skb_put(msdu, HAL_RX_DESC_SIZE + l3pad_bytes + msdu_len);
+	skb_pull(msdu, HAL_RX_DESC_SIZE + l3pad_bytes);
+
+	ath11k_dp_rx_h_ppdu(ar, desc, status);
+
+	__skb_queue_tail(&amsdu_list, msdu);
+
+	ath11k_dp_rx_h_mpdu(ar, &amsdu_list, desc, status);
+
+	/* Please note that caller will having the access to msdu and completing
+	 * rx with mac80211. Need not worry about cleaning up amsdu_list.
+	 */
+
+	return 0;
+}
+
+static bool ath11k_dp_rx_h_reo_err(struct ath11k *ar, struct sk_buff *msdu,
+				   struct ieee80211_rx_status *status,
+				   struct sk_buff_head *msdu_list)
+{
+	struct ath11k_skb_rxcb *rxcb = ATH11K_SKB_RXCB(msdu);
+	bool drop = false;
+
+	ar->ab->soc_stats.reo_error[rxcb->err_code]++;
+
+	switch (rxcb->err_code) {
+	case HAL_REO_DEST_RING_ERROR_CODE_DESC_ADDR_ZERO:
+		if (ath11k_dp_rx_h_null_q_desc(ar, msdu, status, msdu_list))
+			drop = true;
+		break;
+	default:
+		/* TODO: Review other errors and process them to mac80211
+		 * as appropriate.
+		 */
+		drop = true;
+		break;
+	}
+
+	return drop;
+}
+
+static void ath11k_dp_rx_h_tkip_mic_err(struct ath11k *ar, struct sk_buff *msdu,
+					struct ieee80211_rx_status *status)
+{
+	u16 msdu_len;
+	struct hal_rx_desc *desc = (struct hal_rx_desc *)msdu->data;
+	u8 l3pad_bytes;
+	struct ath11k_skb_rxcb *rxcb = ATH11K_SKB_RXCB(msdu);
+
+	rxcb->is_first_msdu = ath11k_dp_rx_h_msdu_end_first_msdu(desc);
+	rxcb->is_last_msdu = ath11k_dp_rx_h_msdu_end_last_msdu(desc);
+
+	l3pad_bytes = ath11k_dp_rx_h_msdu_end_l3pad(desc);
+	msdu_len = ath11k_dp_rx_h_msdu_start_msdu_len(desc);
+	skb_put(msdu, HAL_RX_DESC_SIZE + l3pad_bytes + msdu_len);
+	skb_pull(msdu, HAL_RX_DESC_SIZE + l3pad_bytes);
+
+	ath11k_dp_rx_h_ppdu(ar, desc, status);
+
+	status->flag |= (RX_FLAG_MMIC_STRIPPED | RX_FLAG_MMIC_ERROR |
+			 RX_FLAG_DECRYPTED);
+
+	ath11k_dp_rx_h_undecap(ar, msdu, desc,
+			       HAL_ENCRYPT_TYPE_TKIP_MIC, status, false);
+}
+
+static bool ath11k_dp_rx_h_rxdma_err(struct ath11k *ar,  struct sk_buff *msdu,
+				     struct ieee80211_rx_status *status)
+{
+	struct ath11k_skb_rxcb *rxcb = ATH11K_SKB_RXCB(msdu);
+	bool drop = false;
+
+	ar->ab->soc_stats.rxdma_error[rxcb->err_code]++;
+
+	switch (rxcb->err_code) {
+	case HAL_REO_ENTR_RING_RXDMA_ECODE_TKIP_MIC_ERR:
+		ath11k_dp_rx_h_tkip_mic_err(ar, msdu, status);
+		break;
+	default:
+		/* TODO: Review other rxdma error code to check if anything is
+		 * worth reporting to mac80211
+		 */
+		drop = true;
+		break;
+	}
+
+	return drop;
+}
+
+static void ath11k_dp_rx_wbm_err(struct ath11k *ar,
+				 struct napi_struct *napi,
+				 struct sk_buff *msdu,
+				 struct sk_buff_head *msdu_list)
+{
+	struct ath11k_skb_rxcb *rxcb = ATH11K_SKB_RXCB(msdu);
+	struct ieee80211_rx_status rxs = {0};
+	struct ieee80211_rx_status *status;
+	bool drop = true;
+
+	switch (rxcb->err_rel_src) {
+	case HAL_WBM_REL_SRC_MODULE_REO:
+		drop = ath11k_dp_rx_h_reo_err(ar, msdu, &rxs, msdu_list);
+		break;
+	case HAL_WBM_REL_SRC_MODULE_RXDMA:
+		drop = ath11k_dp_rx_h_rxdma_err(ar, msdu, &rxs);
+		break;
+	default:
+		/* msdu will get freed */
+		break;
+	}
+
+	if (drop) {
+		dev_kfree_skb_any(msdu);
+		return;
+	}
+
+	status = IEEE80211_SKB_RXCB(msdu);
+	*status = rxs;
+
+	ath11k_dp_rx_deliver_msdu(ar, napi, msdu);
+}
+
+int ath11k_dp_rx_process_wbm_err(struct ath11k_base *ab,
+				 struct napi_struct *napi, int budget)
+{
+	struct ath11k *ar;
+	struct ath11k_dp *dp = &ab->dp;
+	struct dp_rxdma_ring *rx_ring;
+	struct hal_rx_wbm_rel_info err_info;
+	struct hal_srng *srng;
+	struct sk_buff *msdu;
+	struct sk_buff_head msdu_list[MAX_RADIOS];
+	struct ath11k_skb_rxcb *rxcb;
+	u32 *rx_desc;
+	int buf_id, mac_id;
+	int num_buffs_reaped[MAX_RADIOS] = {0};
+	int total_num_buffs_reaped = 0;
+	int ret, i;
+
+	for (i = 0; i < MAX_RADIOS; i++)
+		__skb_queue_head_init(&msdu_list[i]);
+
+	srng = &ab->hal.srng_list[dp->rx_rel_ring.ring_id];
+
+	spin_lock_bh(&srng->lock);
+
+	ath11k_hal_srng_access_begin(ab, srng);
+
+	while (budget) {
+		rx_desc = ath11k_hal_srng_dst_get_next_entry(ab, srng);
+		if (!rx_desc)
+			break;
+
+		ret = ath11k_hal_wbm_desc_parse_err(ab, rx_desc, &err_info);
+		if (ret) {
+			ath11k_warn(ab,
+				    "failed to parse rx error in wbm_rel ring desc %d\n",
+				    ret);
+			continue;
+		}
+
+		buf_id = FIELD_GET(DP_RXDMA_BUF_COOKIE_BUF_ID, err_info.cookie);
+		mac_id = FIELD_GET(DP_RXDMA_BUF_COOKIE_PDEV_ID, err_info.cookie);
+
+		ar = ab->pdevs[mac_id].ar;
+		rx_ring = &ar->dp.rx_refill_buf_ring;
+
+		spin_lock_bh(&rx_ring->idr_lock);
+		msdu = idr_find(&rx_ring->bufs_idr, buf_id);
+		if (!msdu) {
+			ath11k_warn(ab, "frame rx with invalid buf_id %d pdev %d\n",
+				    buf_id, mac_id);
+			spin_unlock_bh(&rx_ring->idr_lock);
+			continue;
+		}
+
+		idr_remove(&rx_ring->bufs_idr, buf_id);
+		spin_unlock_bh(&rx_ring->idr_lock);
+
+		rxcb = ATH11K_SKB_RXCB(msdu);
+		dma_unmap_single(ab->dev, rxcb->paddr,
+				 msdu->len + skb_tailroom(msdu),
+				 DMA_FROM_DEVICE);
+
+		num_buffs_reaped[mac_id]++;
+		total_num_buffs_reaped++;
+		budget--;
+
+		if (err_info.push_reason !=
+		    HAL_REO_DEST_RING_PUSH_REASON_ERR_DETECTED) {
+			dev_kfree_skb_any(msdu);
+			continue;
+		}
+
+		rxcb->err_rel_src = err_info.err_rel_src;
+		rxcb->err_code = err_info.err_code;
+		rxcb->rx_desc = (struct hal_rx_desc *)msdu->data;
+		__skb_queue_tail(&msdu_list[mac_id], msdu);
+	}
+
+	ath11k_hal_srng_access_end(ab, srng);
+
+	spin_unlock_bh(&srng->lock);
+
+	if (!total_num_buffs_reaped)
+		goto done;
+
+	for (i = 0; i <  ab->num_radios; i++) {
+		if (!num_buffs_reaped[i])
+			continue;
+
+		ar = ab->pdevs[i].ar;
+		rx_ring = &ar->dp.rx_refill_buf_ring;
+
+		ath11k_dp_rxbufs_replenish(ab, i, rx_ring, num_buffs_reaped[i],
+					   HAL_RX_BUF_RBM_SW3_BM, GFP_ATOMIC);
+	}
+
+	rcu_read_lock();
+	for (i = 0; i <  ab->num_radios; i++) {
+		if (!rcu_dereference(ab->pdevs_active[i])) {
+			__skb_queue_purge(&msdu_list[i]);
+			continue;
+		}
+
+		ar = ab->pdevs[i].ar;
+
+		if (test_bit(ATH11K_CAC_RUNNING, &ar->dev_flags)) {
+			__skb_queue_purge(&msdu_list[i]);
+			continue;
+		}
+
+		while ((msdu = __skb_dequeue(&msdu_list[i])) != NULL)
+			ath11k_dp_rx_wbm_err(ar, napi, msdu, &msdu_list[i]);
+	}
+	rcu_read_unlock();
+done:
+	return total_num_buffs_reaped;
+}
+
+int ath11k_dp_process_rxdma_err(struct ath11k_base *ab, int mac_id, int budget)
+{
+	struct ath11k *ar = ab->pdevs[mac_id].ar;
+	struct dp_srng *err_ring = &ar->dp.rxdma_err_dst_ring;
+	struct dp_rxdma_ring *rx_ring = &ar->dp.rx_refill_buf_ring;
+	struct dp_link_desc_bank *link_desc_banks = ab->dp.link_desc_banks;
+	struct hal_srng *srng;
+	struct hal_rx_msdu_meta meta[HAL_NUM_RX_MSDUS_PER_LINK_DESC];
+	enum hal_rx_buf_return_buf_manager rbm;
+	enum hal_reo_entr_rxdma_ecode rxdma_err_code;
+	struct ath11k_skb_rxcb *rxcb;
+	struct sk_buff *skb;
+	struct hal_reo_entrance_ring *entr_ring;
+	void *desc;
+	int num_buf_freed = 0;
+	int quota = budget;
+	dma_addr_t paddr;
+	u32 desc_bank;
+	void *link_desc_va;
+	int num_msdus;
+	int i;
+	int buf_id;
+
+	srng = &ab->hal.srng_list[err_ring->ring_id];
+
+	spin_lock_bh(&srng->lock);
+
+	ath11k_hal_srng_access_begin(ab, srng);
+
+	while (quota-- &&
+	       (desc = ath11k_hal_srng_dst_get_next_entry(ab, srng))) {
+		ath11k_hal_rx_reo_ent_paddr_get(ab, desc, &paddr, &desc_bank);
+
+		entr_ring = (struct hal_reo_entrance_ring *)desc;
+		rxdma_err_code =
+			FIELD_GET(HAL_REO_ENTR_RING_INFO1_RXDMA_ERROR_CODE,
+				  entr_ring->info1);
+		ab->soc_stats.rxdma_error[rxdma_err_code]++;
+
+		link_desc_va = link_desc_banks[desc_bank].vaddr +
+			       (paddr - link_desc_banks[desc_bank].paddr);
+		ath11k_hal_rx_msdu_link_info_get(link_desc_va, &num_msdus, meta,
+						 &rbm);
+
+		for (i = 0; i < num_msdus; i++) {
+			buf_id = FIELD_GET(DP_RXDMA_BUF_COOKIE_BUF_ID,
+					   meta[i].cookie);
+
+			spin_lock_bh(&rx_ring->idr_lock);
+			skb = idr_find(&rx_ring->bufs_idr, buf_id);
+			if (!skb) {
+				ath11k_warn(ab, "rxdma error with invalid buf_id %d\n",
+					    buf_id);
+				spin_unlock_bh(&rx_ring->idr_lock);
+				continue;
+			}
+
+			idr_remove(&rx_ring->bufs_idr, buf_id);
+			spin_unlock_bh(&rx_ring->idr_lock);
+
+			rxcb = ATH11K_SKB_RXCB(skb);
+			dma_unmap_single(ab->dev, rxcb->paddr,
+					 skb->len + skb_tailroom(skb),
+					 DMA_FROM_DEVICE);
+			dev_kfree_skb_any(skb);
+
+			num_buf_freed++;
+		}
+
+		ath11k_dp_rx_link_desc_return(ab, desc,
+					      HAL_WBM_REL_BM_ACT_PUT_IN_IDLE);
+	}
+
+	ath11k_hal_srng_access_end(ab, srng);
+
+	spin_unlock_bh(&srng->lock);
+
+	if (num_buf_freed)
+		ath11k_dp_rxbufs_replenish(ab, mac_id, rx_ring, num_buf_freed,
+					   HAL_RX_BUF_RBM_SW3_BM, GFP_ATOMIC);
+
+	return budget - quota;
+}
+
+void ath11k_dp_process_reo_status(struct ath11k_base *ab)
+{
+	struct ath11k_dp *dp = &ab->dp;
+	struct hal_srng *srng;
+	struct dp_reo_cmd *cmd, *tmp;
+	bool found = false;
+	u32 *reo_desc;
+	u16 tag;
+	struct hal_reo_status reo_status;
+
+	srng = &ab->hal.srng_list[dp->reo_status_ring.ring_id];
+
+	memset(&reo_status, 0, sizeof(reo_status));
+
+	spin_lock_bh(&srng->lock);
+
+	ath11k_hal_srng_access_begin(ab, srng);
+
+	while ((reo_desc = ath11k_hal_srng_dst_get_next_entry(ab, srng))) {
+		tag = FIELD_GET(HAL_SRNG_TLV_HDR_TAG, *reo_desc);
+
+		switch (tag) {
+		case HAL_REO_GET_QUEUE_STATS_STATUS:
+			ath11k_hal_reo_status_queue_stats(ab, reo_desc,
+							  &reo_status);
+			break;
+		case HAL_REO_FLUSH_QUEUE_STATUS:
+			ath11k_hal_reo_flush_queue_status(ab, reo_desc,
+							  &reo_status);
+			break;
+		case HAL_REO_FLUSH_CACHE_STATUS:
+			ath11k_hal_reo_flush_cache_status(ab, reo_desc,
+							  &reo_status);
+			break;
+		case HAL_REO_UNBLOCK_CACHE_STATUS:
+			ath11k_hal_reo_unblk_cache_status(ab, reo_desc,
+							  &reo_status);
+			break;
+		case HAL_REO_FLUSH_TIMEOUT_LIST_STATUS:
+			ath11k_hal_reo_flush_timeout_list_status(ab, reo_desc,
+								 &reo_status);
+			break;
+		case HAL_REO_DESCRIPTOR_THRESHOLD_REACHED_STATUS:
+			ath11k_hal_reo_desc_thresh_reached_status(ab, reo_desc,
+								  &reo_status);
+			break;
+		case HAL_REO_UPDATE_RX_REO_QUEUE_STATUS:
+			ath11k_hal_reo_update_rx_reo_queue_status(ab, reo_desc,
+								  &reo_status);
+			break;
+		default:
+			ath11k_warn(ab, "Unknown reo status type %d\n", tag);
+			continue;
+		}
+
+		spin_lock_bh(&dp->reo_cmd_lock);
+		list_for_each_entry_safe(cmd, tmp, &dp->reo_cmd_list, list) {
+			if (reo_status.uniform_hdr.cmd_num == cmd->cmd_num) {
+				found = true;
+				list_del(&cmd->list);
+				break;
+			}
+		}
+		spin_unlock_bh(&dp->reo_cmd_lock);
+
+		if (found) {
+			cmd->handler(dp, (void *)&cmd->data,
+				     reo_status.uniform_hdr.cmd_status);
+			kfree(cmd);
+		}
+
+		found = false;
+	}
+
+	ath11k_hal_srng_access_end(ab, srng);
+
+	spin_unlock_bh(&srng->lock);
+}
+
+void ath11k_dp_rx_pdev_free(struct ath11k_base *ab, int mac_id)
+{
+	struct ath11k *ar = ab->pdevs[mac_id].ar;
+
+	ath11k_dp_rx_pdev_srng_free(ar);
+	ath11k_dp_rxdma_pdev_buf_free(ar);
+}
+
+int ath11k_dp_rx_pdev_alloc(struct ath11k_base *ab, int mac_id)
+{
+	struct ath11k *ar = ab->pdevs[mac_id].ar;
+	struct ath11k_pdev_dp *dp = &ar->dp;
+	u32 ring_id;
+	int ret;
+
+	ret = ath11k_dp_rx_pdev_srng_alloc(ar);
+	if (ret) {
+		ath11k_warn(ab, "failed to setup rx srngs\n");
+		return ret;
+	}
+
+	ret = ath11k_dp_rxdma_pdev_buf_setup(ar);
+	if (ret) {
+		ath11k_warn(ab, "failed to setup rxdma ring\n");
+		return ret;
+	}
+
+	ring_id = dp->rx_refill_buf_ring.refill_buf_ring.ring_id;
+	ret = ath11k_dp_tx_htt_srng_setup(ab, ring_id, mac_id, HAL_RXDMA_BUF);
+	if (ret) {
+		ath11k_warn(ab, "failed to configure rx_refill_buf_ring %d\n",
+			    ret);
+		return ret;
+	}
+
+	ring_id = dp->rxdma_err_dst_ring.ring_id;
+	ret = ath11k_dp_tx_htt_srng_setup(ab, ring_id, mac_id, HAL_RXDMA_DST);
+	if (ret) {
+		ath11k_warn(ab, "failed to configure rxdma_err_dest_ring %d\n",
+			    ret);
+		return ret;
+	}
+
+	ring_id = dp->rxdma_mon_buf_ring.refill_buf_ring.ring_id;
+	ret = ath11k_dp_tx_htt_srng_setup(ab, ring_id,
+					  mac_id, HAL_RXDMA_MONITOR_BUF);
+	if (ret) {
+		ath11k_warn(ab, "failed to configure rxdma_mon_buf_ring %d\n",
+			    ret);
+		return ret;
+	}
+	ret = ath11k_dp_tx_htt_srng_setup(ab,
+					  dp->rxdma_mon_dst_ring.ring_id,
+					  mac_id, HAL_RXDMA_MONITOR_DST);
+	if (ret) {
+		ath11k_warn(ab, "failed to configure rxdma_mon_dst_ring %d\n",
+			    ret);
+		return ret;
+	}
+	ret = ath11k_dp_tx_htt_srng_setup(ab,
+					  dp->rxdma_mon_desc_ring.ring_id,
+					  mac_id, HAL_RXDMA_MONITOR_DESC);
+	if (ret) {
+		ath11k_warn(ab, "failed to configure rxdma_mon_dst_ring %d\n",
+			    ret);
+		return ret;
+	}
+	ring_id = dp->rx_mon_status_refill_ring.refill_buf_ring.ring_id;
+	ret = ath11k_dp_tx_htt_srng_setup(ab, ring_id, mac_id,
+					  HAL_RXDMA_MONITOR_STATUS);
+	if (ret) {
+		ath11k_warn(ab,
+			    "failed to configure mon_status_refill_ring %d\n",
+			    ret);
+		return ret;
+	}
+	return 0;
+}
+
+static void ath11k_dp_mon_set_frag_len(u32 *total_len, u32 *frag_len)
+{
+	if (*total_len >= (DP_RX_BUFFER_SIZE - sizeof(struct hal_rx_desc))) {
+		*frag_len = DP_RX_BUFFER_SIZE - sizeof(struct hal_rx_desc);
+		*total_len -= *frag_len;
+	} else {
+		*frag_len = *total_len;
+		*total_len = 0;
+	}
+}
+
+static
+int ath11k_dp_rx_monitor_link_desc_return(struct ath11k *ar,
+					  void *p_last_buf_addr_info,
+					  u8 mac_id)
+{
+	struct ath11k_pdev_dp *dp = &ar->dp;
+	struct dp_srng *dp_srng;
+	void *hal_srng;
+	void *src_srng_desc;
+	int ret = 0;
+
+	dp_srng = &dp->rxdma_mon_desc_ring;
+	hal_srng = &ar->ab->hal.srng_list[dp_srng->ring_id];
+
+	ath11k_hal_srng_access_begin(ar->ab, hal_srng);
+
+	src_srng_desc = ath11k_hal_srng_src_get_next_entry(ar->ab, hal_srng);
+
+	if (src_srng_desc) {
+		struct ath11k_buffer_addr *src_desc =
+				(struct ath11k_buffer_addr *)src_srng_desc;
+
+		*src_desc = *((struct ath11k_buffer_addr *)p_last_buf_addr_info);
+	} else {
+		ath11k_dbg(ar->ab, ATH11K_DBG_DATA,
+			   "Monitor Link Desc Ring %d Full", mac_id);
+		ret = -ENOMEM;
+	}
+
+	ath11k_hal_srng_access_end(ar->ab, hal_srng);
+	return ret;
+}
+
+static
+void ath11k_dp_rx_mon_next_link_desc_get(void *rx_msdu_link_desc,
+					 dma_addr_t *paddr, u32 *sw_cookie,
+					 void **pp_buf_addr_info)
+{
+	struct hal_rx_msdu_link *msdu_link =
+			(struct hal_rx_msdu_link *)rx_msdu_link_desc;
+	struct ath11k_buffer_addr *buf_addr_info;
+	u8 rbm = 0;
+
+	buf_addr_info = (struct ath11k_buffer_addr *)&msdu_link->buf_addr_info;
+
+	ath11k_hal_rx_buf_addr_info_get(buf_addr_info, paddr, sw_cookie, &rbm);
+
+	*pp_buf_addr_info = (void *)buf_addr_info;
+}
+
+static int ath11k_dp_pkt_set_pktlen(struct sk_buff *skb, u32 len)
+{
+	if (skb->len > len) {
+		skb_trim(skb, len);
+	} else {
+		if (skb_tailroom(skb) < len - skb->len) {
+			if ((pskb_expand_head(skb, 0,
+					      len - skb->len - skb_tailroom(skb),
+					      GFP_ATOMIC))) {
+				dev_kfree_skb_any(skb);
+				return -ENOMEM;
+			}
+		}
+		skb_put(skb, (len - skb->len));
+	}
+	return 0;
+}
+
+static void ath11k_hal_rx_msdu_list_get(struct ath11k *ar,
+					void *msdu_link_desc,
+					struct hal_rx_msdu_list *msdu_list,
+					u16 *num_msdus)
+{
+	struct hal_rx_msdu_details *msdu_details = NULL;
+	struct rx_msdu_desc *msdu_desc_info = NULL;
+	struct hal_rx_msdu_link *msdu_link = NULL;
+	int i;
+	u32 last = FIELD_PREP(RX_MSDU_DESC_INFO0_LAST_MSDU_IN_MPDU, 1);
+	u32 first = FIELD_PREP(RX_MSDU_DESC_INFO0_FIRST_MSDU_IN_MPDU, 1);
+	u8  tmp  = 0;
+
+	msdu_link = (struct hal_rx_msdu_link *)msdu_link_desc;
+	msdu_details = &msdu_link->msdu_link[0];
+
+	for (i = 0; i < HAL_RX_NUM_MSDU_DESC; i++) {
+		if (FIELD_GET(BUFFER_ADDR_INFO0_ADDR,
+			      msdu_details[i].buf_addr_info.info0) == 0) {
+			msdu_desc_info = &msdu_details[i - 1].rx_msdu_info;
+			msdu_desc_info->info0 |= last;
+			;
+			break;
+		}
+		msdu_desc_info = &msdu_details[i].rx_msdu_info;
+
+		if (!i)
+			msdu_desc_info->info0 |= first;
+		else if (i == (HAL_RX_NUM_MSDU_DESC - 1))
+			msdu_desc_info->info0 |= last;
+		msdu_list->msdu_info[i].msdu_flags = msdu_desc_info->info0;
+		msdu_list->msdu_info[i].msdu_len =
+			 HAL_RX_MSDU_PKT_LENGTH_GET(msdu_desc_info->info0);
+		msdu_list->sw_cookie[i] =
+			FIELD_GET(BUFFER_ADDR_INFO1_SW_COOKIE,
+				  msdu_details[i].buf_addr_info.info1);
+		tmp = FIELD_GET(BUFFER_ADDR_INFO1_RET_BUF_MGR,
+				msdu_details[i].buf_addr_info.info1);
+		msdu_list->rbm[i] = tmp;
+	}
+	*num_msdus = i;
+}
+
+static u32 ath11k_dp_rx_mon_comp_ppduid(u32 msdu_ppdu_id, u32 *ppdu_id,
+					u32 *rx_bufs_used)
+{
+	u32 ret = 0;
+
+	if ((*ppdu_id < msdu_ppdu_id) &&
+	    ((msdu_ppdu_id - *ppdu_id) < DP_NOT_PPDU_ID_WRAP_AROUND)) {
+		*ppdu_id = msdu_ppdu_id;
+		ret = msdu_ppdu_id;
+	} else if ((*ppdu_id > msdu_ppdu_id) &&
+		((*ppdu_id - msdu_ppdu_id) > DP_NOT_PPDU_ID_WRAP_AROUND)) {
+		/* mon_dst is behind than mon_status
+		 * skip dst_ring and free it
+		 */
+		*rx_bufs_used += 1;
+		*ppdu_id = msdu_ppdu_id;
+		ret = msdu_ppdu_id;
+	}
+	return ret;
+}
+
+static void ath11k_dp_mon_get_buf_len(struct hal_rx_msdu_desc_info *info,
+				      bool *is_frag, u32 *total_len,
+				      u32 *frag_len, u32 *msdu_cnt)
+{
+	if (info->msdu_flags & RX_MSDU_DESC_INFO0_MSDU_CONTINUATION) {
+		if (!*is_frag) {
+			*total_len = info->msdu_len;
+			*is_frag = true;
+		}
+		ath11k_dp_mon_set_frag_len(total_len,
+					   frag_len);
+	} else {
+		if (*is_frag) {
+			ath11k_dp_mon_set_frag_len(total_len,
+						   frag_len);
+		} else {
+			*frag_len = info->msdu_len;
+		}
+		*is_frag = false;
+		*msdu_cnt -= 1;
+	}
+}
+
+static u32
+ath11k_dp_rx_mon_mpdu_pop(struct ath11k *ar,
+			  void *ring_entry, struct sk_buff **head_msdu,
+			  struct sk_buff **tail_msdu, u32 *npackets,
+			  u32 *ppdu_id)
+{
+	struct ath11k_pdev_dp *dp = &ar->dp;
+	struct ath11k_mon_data *pmon = (struct ath11k_mon_data *)&dp->mon_data;
+	struct dp_rxdma_ring *rx_ring = &dp->rxdma_mon_buf_ring;
+	struct sk_buff *msdu = NULL, *last = NULL;
+	struct hal_rx_msdu_list msdu_list;
+	void *p_buf_addr_info, *p_last_buf_addr_info;
+	struct hal_rx_desc *rx_desc;
+	void *rx_msdu_link_desc;
+	dma_addr_t paddr;
+	u16 num_msdus = 0;
+	u32 rx_buf_size, rx_pkt_offset, sw_cookie;
+	u32 rx_bufs_used = 0, i = 0;
+	u32 msdu_ppdu_id = 0, msdu_cnt = 0;
+	u32 total_len = 0, frag_len = 0;
+	bool is_frag, is_first_msdu;
+	bool drop_mpdu = false;
+	struct ath11k_skb_rxcb *rxcb;
+	struct hal_reo_entrance_ring *ent_desc =
+			(struct hal_reo_entrance_ring *)ring_entry;
+	int buf_id;
+
+	ath11k_hal_rx_reo_ent_buf_paddr_get(ring_entry, &paddr,
+					    &sw_cookie, &p_last_buf_addr_info,
+					    &msdu_cnt);
+
+	if (FIELD_GET(HAL_REO_ENTR_RING_INFO1_RXDMA_PUSH_REASON,
+		      ent_desc->info1) ==
+		      HAL_REO_DEST_RING_PUSH_REASON_ERR_DETECTED) {
+		u8 rxdma_err =
+			FIELD_GET(HAL_REO_ENTR_RING_INFO1_RXDMA_ERROR_CODE,
+				  ent_desc->info1);
+		if (rxdma_err == HAL_REO_ENTR_RING_RXDMA_ECODE_FLUSH_REQUEST_ERR ||
+		    rxdma_err == HAL_REO_ENTR_RING_RXDMA_ECODE_MPDU_LEN_ERR ||
+		    rxdma_err == HAL_REO_ENTR_RING_RXDMA_ECODE_OVERFLOW_ERR) {
+			drop_mpdu = true;
+			pmon->rx_mon_stats.dest_mpdu_drop++;
+		}
+	}
+
+	is_frag = false;
+	is_first_msdu = true;
+
+	do {
+		if (pmon->mon_last_linkdesc_paddr == paddr) {
+			pmon->rx_mon_stats.dup_mon_linkdesc_cnt++;
+			return rx_bufs_used;
+		}
+
+		rx_msdu_link_desc =
+			(void *)pmon->link_desc_banks[sw_cookie].vaddr +
+			(paddr - pmon->link_desc_banks[sw_cookie].paddr);
+
+		ath11k_hal_rx_msdu_list_get(ar, rx_msdu_link_desc, &msdu_list,
+					    &num_msdus);
+
+		for (i = 0; i < num_msdus; i++) {
+			u32 l2_hdr_offset;
+
+			if (pmon->mon_last_buf_cookie == msdu_list.sw_cookie[i]) {
+				ath11k_dbg(ar->ab, ATH11K_DBG_DATA,
+					   "i %d last_cookie %d is same\n",
+					   i, pmon->mon_last_buf_cookie);
+				drop_mpdu = true;
+				pmon->rx_mon_stats.dup_mon_buf_cnt++;
+				continue;
+			}
+			buf_id = FIELD_GET(DP_RXDMA_BUF_COOKIE_BUF_ID,
+					   msdu_list.sw_cookie[i]);
+
+			spin_lock_bh(&rx_ring->idr_lock);
+			msdu = idr_find(&rx_ring->bufs_idr, buf_id);
+			spin_unlock_bh(&rx_ring->idr_lock);
+			if (!msdu) {
+				ath11k_dbg(ar->ab, ATH11K_DBG_DATA,
+					   "msdu_pop: invalid buf_id %d\n", buf_id);
+				break;
+			}
+			rxcb = ATH11K_SKB_RXCB(msdu);
+			if (!rxcb->unmapped) {
+				dma_unmap_single(ar->ab->dev, rxcb->paddr,
+						 msdu->len +
+						 skb_tailroom(msdu),
+						 DMA_FROM_DEVICE);
+				rxcb->unmapped = 1;
+			}
+			if (drop_mpdu) {
+				ath11k_dbg(ar->ab, ATH11K_DBG_DATA,
+					   "i %d drop msdu %p *ppdu_id %x\n",
+					   i, msdu, *ppdu_id);
+				dev_kfree_skb_any(msdu);
+				msdu = NULL;
+				goto next_msdu;
+			}
+
+			rx_desc = (struct hal_rx_desc *)msdu->data;
+
+			rx_pkt_offset = sizeof(struct hal_rx_desc);
+			l2_hdr_offset = ath11k_dp_rx_h_msdu_end_l3pad(rx_desc);
+
+			if (is_first_msdu) {
+				if (!ath11k_dp_rxdesc_mpdu_valid(rx_desc)) {
+					drop_mpdu = true;
+					dev_kfree_skb_any(msdu);
+					msdu = NULL;
+					pmon->mon_last_linkdesc_paddr = paddr;
+					goto next_msdu;
+				}
+
+				msdu_ppdu_id =
+					ath11k_dp_rxdesc_get_ppduid(rx_desc);
+
+				if (ath11k_dp_rx_mon_comp_ppduid(msdu_ppdu_id,
+								 ppdu_id,
+								 &rx_bufs_used))
+					return rx_bufs_used;
+				pmon->mon_last_linkdesc_paddr = paddr;
+				is_first_msdu = false;
+			}
+			ath11k_dp_mon_get_buf_len(&msdu_list.msdu_info[i],
+						  &is_frag, &total_len,
+						  &frag_len, &msdu_cnt);
+			rx_buf_size = rx_pkt_offset + l2_hdr_offset + frag_len;
+
+			ath11k_dp_pkt_set_pktlen(msdu, rx_buf_size);
+
+			if (!(*head_msdu))
+				*head_msdu = msdu;
+			else if (last)
+				last->next = msdu;
+
+			last = msdu;
+next_msdu:
+			pmon->mon_last_buf_cookie = msdu_list.sw_cookie[i];
+			rx_bufs_used++;
+			spin_lock_bh(&rx_ring->idr_lock);
+			idr_remove(&rx_ring->bufs_idr, buf_id);
+			spin_unlock_bh(&rx_ring->idr_lock);
+		}
+
+		ath11k_dp_rx_mon_next_link_desc_get(rx_msdu_link_desc, &paddr,
+						    &sw_cookie,
+						    &p_buf_addr_info);
+
+		if (ath11k_dp_rx_monitor_link_desc_return(ar,
+							  p_last_buf_addr_info,
+							  dp->mac_id))
+			ath11k_dbg(ar->ab, ATH11K_DBG_DATA,
+				   "dp_rx_monitor_link_desc_return failed");
+
+		p_last_buf_addr_info = p_buf_addr_info;
+
+	} while (paddr && msdu_cnt);
+
+	if (last)
+		last->next = NULL;
+
+	*tail_msdu = msdu;
+
+	if (msdu_cnt == 0)
+		*npackets = 1;
+
+	return rx_bufs_used;
+}
+
+static void ath11k_dp_rx_msdus_set_payload(struct sk_buff *msdu)
+{
+	u32 rx_pkt_offset, l2_hdr_offset;
+
+	rx_pkt_offset = sizeof(struct hal_rx_desc);
+	l2_hdr_offset = ath11k_dp_rx_h_msdu_end_l3pad((struct hal_rx_desc *)msdu->data);
+	skb_pull(msdu, rx_pkt_offset + l2_hdr_offset);
+}
+
+static struct sk_buff *
+ath11k_dp_rx_mon_merg_msdus(struct ath11k *ar,
+			    u32 mac_id, struct sk_buff *head_msdu,
+			    struct sk_buff *last_msdu,
+			    struct ieee80211_rx_status *rxs)
+{
+	struct sk_buff *msdu, *mpdu_buf, *prev_buf;
+	u32 decap_format, wifi_hdr_len;
+	struct hal_rx_desc *rx_desc;
+	char *hdr_desc;
+	u8 *dest;
+	struct ieee80211_hdr_3addr *wh;
+
+	mpdu_buf = NULL;
+
+	if (!head_msdu)
+		goto err_merge_fail;
+
+	rx_desc = (struct hal_rx_desc *)head_msdu->data;
+
+	if (ath11k_dp_rxdesc_get_mpdulen_err(rx_desc))
+		return NULL;
+
+	decap_format = ath11k_dp_rxdesc_get_decap_format(rx_desc);
+
+	ath11k_dp_rx_h_ppdu(ar, rx_desc, rxs);
+
+	if (decap_format == DP_RX_DECAP_TYPE_RAW) {
+		ath11k_dp_rx_msdus_set_payload(head_msdu);
+
+		prev_buf = head_msdu;
+		msdu = head_msdu->next;
+
+		while (msdu) {
+			ath11k_dp_rx_msdus_set_payload(msdu);
+
+			prev_buf = msdu;
+			msdu = msdu->next;
+		}
+
+		prev_buf->next = NULL;
+
+		skb_trim(prev_buf, prev_buf->len - HAL_RX_FCS_LEN);
+	} else if (decap_format == DP_RX_DECAP_TYPE_NATIVE_WIFI) {
+		__le16 qos_field;
+		u8 qos_pkt = 0;
+
+		rx_desc = (struct hal_rx_desc *)head_msdu->data;
+		hdr_desc = ath11k_dp_rxdesc_get_80211hdr(rx_desc);
+
+		/* Base size */
+		wifi_hdr_len = sizeof(struct ieee80211_hdr_3addr);
+		wh = (struct ieee80211_hdr_3addr *)hdr_desc;
+
+		if (ieee80211_is_data_qos(wh->frame_control)) {
+			struct ieee80211_qos_hdr *qwh =
+					(struct ieee80211_qos_hdr *)hdr_desc;
+
+			qos_field = qwh->qos_ctrl;
+			qos_pkt = 1;
+		}
+		msdu = head_msdu;
+
+		while (msdu) {
+			rx_desc = (struct hal_rx_desc *)msdu->data;
+			hdr_desc = ath11k_dp_rxdesc_get_80211hdr(rx_desc);
+
+			if (qos_pkt) {
+				dest = skb_push(msdu, sizeof(__le16));
+				if (!dest)
+					goto err_merge_fail;
+				memcpy(dest, hdr_desc, wifi_hdr_len);
+				memcpy(dest + wifi_hdr_len,
+				       (u8 *)&qos_field, sizeof(__le16));
+			}
+			ath11k_dp_rx_msdus_set_payload(msdu);
+			prev_buf = msdu;
+			msdu = msdu->next;
+		}
+		dest = skb_put(prev_buf, HAL_RX_FCS_LEN);
+		if (!dest)
+			goto err_merge_fail;
+
+		ath11k_dbg(ar->ab, ATH11K_DBG_DATA,
+			   "mpdu_buf %pK mpdu_buf->len %u",
+			   prev_buf, prev_buf->len);
+	} else {
+		ath11k_dbg(ar->ab, ATH11K_DBG_DATA,
+			   "decap format %d is not supported!\n",
+			   decap_format);
+		goto err_merge_fail;
+	}
+
+	return head_msdu;
+
+err_merge_fail:
+	if (mpdu_buf && decap_format != DP_RX_DECAP_TYPE_RAW) {
+		ath11k_dbg(ar->ab, ATH11K_DBG_DATA,
+			   "err_merge_fail mpdu_buf %pK", mpdu_buf);
+		/* Free the head buffer */
+		dev_kfree_skb_any(mpdu_buf);
+	}
+	return NULL;
+}
+
+static int ath11k_dp_rx_mon_deliver(struct ath11k *ar, u32 mac_id,
+				    struct sk_buff *head_msdu,
+				    struct sk_buff *tail_msdu,
+				    struct napi_struct *napi)
+{
+	struct ath11k_pdev_dp *dp = &ar->dp;
+	struct sk_buff *mon_skb, *skb_next, *header;
+	struct ieee80211_rx_status *rxs = &dp->rx_status, *status;
+
+	mon_skb = ath11k_dp_rx_mon_merg_msdus(ar, mac_id, head_msdu,
+					      tail_msdu, rxs);
+
+	if (!mon_skb)
+		goto mon_deliver_fail;
+
+	header = mon_skb;
+
+	rxs->flag = 0;
+	do {
+		skb_next = mon_skb->next;
+		if (!skb_next)
+			rxs->flag &= ~RX_FLAG_AMSDU_MORE;
+		else
+			rxs->flag |= RX_FLAG_AMSDU_MORE;
+
+		if (mon_skb == header) {
+			header = NULL;
+			rxs->flag &= ~RX_FLAG_ALLOW_SAME_PN;
+		} else {
+			rxs->flag |= RX_FLAG_ALLOW_SAME_PN;
+		}
+		rxs->flag |= RX_FLAG_ONLY_MONITOR;
+
+		status = IEEE80211_SKB_RXCB(mon_skb);
+		*status = *rxs;
+
+		ath11k_dp_rx_deliver_msdu(ar, napi, mon_skb);
+		mon_skb = skb_next;
+	} while (mon_skb && (mon_skb != tail_msdu));
+	rxs->flag = 0;
+
+	return 0;
+
+mon_deliver_fail:
+	mon_skb = head_msdu;
+	while (mon_skb) {
+		skb_next = mon_skb->next;
+		dev_kfree_skb_any(mon_skb);
+		mon_skb = skb_next;
+	}
+	return -EINVAL;
+}
+
+static void ath11k_dp_rx_mon_dest_process(struct ath11k *ar, u32 quota,
+					  struct napi_struct *napi)
+{
+	struct ath11k_pdev_dp *dp = &ar->dp;
+	struct ath11k_mon_data *pmon = (struct ath11k_mon_data *)&dp->mon_data;
+	void *ring_entry;
+	void *mon_dst_srng;
+	u32 ppdu_id;
+	u32 rx_bufs_used;
+	struct ath11k_pdev_mon_stats *rx_mon_stats;
+	u32	 npackets = 0;
+
+	mon_dst_srng = &ar->ab->hal.srng_list[dp->rxdma_mon_dst_ring.ring_id];
+
+	if (!mon_dst_srng) {
+		ath11k_warn(ar->ab,
+			    "HAL Monitor Destination Ring Init Failed -- %pK",
+			    mon_dst_srng);
+		return;
+	}
+
+	spin_lock_bh(&pmon->mon_lock);
+
+	ath11k_hal_srng_access_begin(ar->ab, mon_dst_srng);
+
+	ppdu_id = pmon->mon_ppdu_info.ppdu_id;
+	rx_bufs_used = 0;
+	rx_mon_stats = &pmon->rx_mon_stats;
+
+	while ((ring_entry = ath11k_hal_srng_dst_peek(ar->ab, mon_dst_srng))) {
+		struct sk_buff *head_msdu, *tail_msdu;
+
+		head_msdu = NULL;
+		tail_msdu = NULL;
+
+		rx_bufs_used += ath11k_dp_rx_mon_mpdu_pop(ar, ring_entry,
+							  &head_msdu,
+							  &tail_msdu,
+							  &npackets, &ppdu_id);
+
+		if (ppdu_id != pmon->mon_ppdu_info.ppdu_id) {
+			pmon->mon_ppdu_status = DP_PPDU_STATUS_START;
+			ath11k_dbg(ar->ab, ATH11K_DBG_DATA,
+				   "dest_rx: new ppdu_id %x != status ppdu_id %x",
+				   ppdu_id, pmon->mon_ppdu_info.ppdu_id);
+			break;
+		}
+		if (head_msdu && tail_msdu) {
+			ath11k_dp_rx_mon_deliver(ar, dp->mac_id, head_msdu,
+						 tail_msdu, napi);
+			rx_mon_stats->dest_mpdu_done++;
+		}
+
+		ring_entry = ath11k_hal_srng_dst_get_next_entry(ar->ab,
+								mon_dst_srng);
+	}
+	ath11k_hal_srng_access_end(ar->ab, mon_dst_srng);
+
+	spin_unlock_bh(&pmon->mon_lock);
+
+	if (rx_bufs_used) {
+		rx_mon_stats->dest_ppdu_done++;
+		ath11k_dp_rxbufs_replenish(ar->ab, dp->mac_id,
+					   &dp->rxdma_mon_buf_ring,
+					   rx_bufs_used,
+					   HAL_RX_BUF_RBM_SW3_BM, GFP_ATOMIC);
+	}
+}
+
+static void ath11k_dp_rx_mon_status_process_tlv(struct ath11k *ar,
+						u32 quota,
+						struct napi_struct *napi)
+{
+	struct ath11k_pdev_dp *dp = &ar->dp;
+	struct ath11k_mon_data *pmon = (struct ath11k_mon_data *)&dp->mon_data;
+	struct hal_rx_mon_ppdu_info *ppdu_info;
+	struct sk_buff *status_skb;
+	u32 tlv_status = HAL_TLV_STATUS_BUF_DONE;
+	struct ath11k_pdev_mon_stats *rx_mon_stats;
+
+	ppdu_info = &pmon->mon_ppdu_info;
+	rx_mon_stats = &pmon->rx_mon_stats;
+
+	if (pmon->mon_ppdu_status != DP_PPDU_STATUS_START)
+		return;
+
+	while (!skb_queue_empty(&pmon->rx_status_q)) {
+		status_skb = skb_dequeue(&pmon->rx_status_q);
+
+		tlv_status = ath11k_hal_rx_parse_mon_status(ar->ab, ppdu_info,
+							    status_skb);
+		if (tlv_status == HAL_TLV_STATUS_PPDU_DONE) {
+			rx_mon_stats->status_ppdu_done++;
+			pmon->mon_ppdu_status = DP_PPDU_STATUS_DONE;
+			ath11k_dp_rx_mon_dest_process(ar, quota, napi);
+			pmon->mon_ppdu_status = DP_PPDU_STATUS_START;
+		}
+		dev_kfree_skb_any(status_skb);
+	}
+}
+
+static int ath11k_dp_mon_process_rx(struct ath11k_base *ab, int mac_id,
+				    struct napi_struct *napi, int budget)
+{
+	struct ath11k *ar = ab->pdevs[mac_id].ar;
+	struct ath11k_pdev_dp *dp = &ar->dp;
+	struct ath11k_mon_data *pmon = (struct ath11k_mon_data *)&dp->mon_data;
+	int num_buffs_reaped = 0;
+
+	num_buffs_reaped = ath11k_dp_rx_reap_mon_status_ring(ar->ab, dp->mac_id, &budget,
+							     &pmon->rx_status_q);
+	if (num_buffs_reaped)
+		ath11k_dp_rx_mon_status_process_tlv(ar, budget, napi);
+
+	return num_buffs_reaped;
+}
+
+int ath11k_dp_rx_process_mon_rings(struct ath11k_base *ab, int mac_id,
+				   struct napi_struct *napi, int budget)
+{
+	struct ath11k *ar = ab->pdevs[mac_id].ar;
+	int ret = 0;
+
+	if (test_bit(ATH11K_FLAG_MONITOR_ENABLED, &ar->monitor_flags))
+		ret = ath11k_dp_mon_process_rx(ab, mac_id, napi, budget);
+	else
+		ret = ath11k_dp_rx_process_mon_status(ab, mac_id, napi, budget);
+	return ret;
+}
+
+static int ath11k_dp_rx_pdev_mon_status_attach(struct ath11k *ar)
+{
+	struct ath11k_pdev_dp *dp = &ar->dp;
+	struct ath11k_mon_data *pmon = (struct ath11k_mon_data *)&dp->mon_data;
+
+	skb_queue_head_init(&pmon->rx_status_q);
+
+	pmon->mon_ppdu_status = DP_PPDU_STATUS_START;
+
+	memset(&pmon->rx_mon_stats, 0,
+	       sizeof(pmon->rx_mon_stats));
+	return 0;
+}
+
+int ath11k_dp_rx_pdev_mon_attach(struct ath11k *ar)
+{
+	struct ath11k_pdev_dp *dp = &ar->dp;
+	struct ath11k_mon_data *pmon = &dp->mon_data;
+	struct hal_srng *mon_desc_srng = NULL;
+	struct dp_srng *dp_srng;
+	int ret = 0;
+	u32 n_link_desc = 0;
+
+	ret = ath11k_dp_rx_pdev_mon_status_attach(ar);
+	if (ret) {
+		ath11k_warn(ar->ab, "pdev_mon_status_attach() failed");
+		return ret;
+	}
+
+	dp_srng = &dp->rxdma_mon_desc_ring;
+	n_link_desc = dp_srng->size /
+		ath11k_hal_srng_get_entrysize(HAL_RXDMA_MONITOR_DESC);
+	mon_desc_srng =
+		&ar->ab->hal.srng_list[dp->rxdma_mon_desc_ring.ring_id];
+
+	ret = ath11k_dp_link_desc_setup(ar->ab, pmon->link_desc_banks,
+					HAL_RXDMA_MONITOR_DESC, mon_desc_srng,
+					n_link_desc);
+	if (ret) {
+		ath11k_warn(ar->ab, "mon_link_desc_pool_setup() failed");
+		return ret;
+	}
+	pmon->mon_last_linkdesc_paddr = 0;
+	pmon->mon_last_buf_cookie = DP_RX_DESC_COOKIE_MAX + 1;
+	spin_lock_init(&pmon->mon_lock);
+	return 0;
+}
+
+static int ath11k_dp_mon_link_free(struct ath11k *ar)
+{
+	struct ath11k_pdev_dp *dp = &ar->dp;
+	struct ath11k_mon_data *pmon = &dp->mon_data;
+
+	ath11k_dp_link_desc_cleanup(ar->ab, pmon->link_desc_banks,
+				    HAL_RXDMA_MONITOR_DESC,
+				    &dp->rxdma_mon_desc_ring);
+	return 0;
+}
+
+int ath11k_dp_rx_pdev_mon_detach(struct ath11k *ar)
+{
+	ath11k_dp_mon_link_free(ar);
+	return 0;
+}
diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.h b/drivers/net/wireless/ath/ath11k/dp_rx.h
new file mode 100644
index 000000000000..026820fda568
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.h
@@ -0,0 +1,84 @@
+/* SPDX-License-Identifier: BSD-3-Clause-Clear */
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+#ifndef ATH11K_DP_RX_H
+#define ATH11K_DP_RX_H
+
+#include "core.h"
+#include "rx_desc.h"
+#include "debug.h"
+
+#define DP_RX_MPDU_ERR_FCS			BIT(0)
+#define DP_RX_MPDU_ERR_DECRYPT			BIT(1)
+#define DP_RX_MPDU_ERR_TKIP_MIC			BIT(2)
+#define DP_RX_MPDU_ERR_AMSDU_ERR		BIT(3)
+#define DP_RX_MPDU_ERR_OVERFLOW			BIT(4)
+#define DP_RX_MPDU_ERR_MSDU_LEN			BIT(5)
+#define DP_RX_MPDU_ERR_MPDU_LEN			BIT(6)
+#define DP_RX_MPDU_ERR_UNENCRYPTED_FRAME	BIT(7)
+
+enum dp_rx_decap_type {
+	DP_RX_DECAP_TYPE_RAW,
+	DP_RX_DECAP_TYPE_NATIVE_WIFI,
+	DP_RX_DECAP_TYPE_ETHERNET2_DIX,
+	DP_RX_DECAP_TYPE_8023,
+};
+
+struct ath11k_dp_amsdu_subframe_hdr {
+	u8 dst[ETH_ALEN];
+	u8 src[ETH_ALEN];
+	__be16 len;
+} __packed;
+
+struct ath11k_dp_rfc1042_hdr {
+	u8 llc_dsap;
+	u8 llc_ssap;
+	u8 llc_ctrl;
+	u8 snap_oui[3];
+	__be16 snap_type;
+} __packed;
+
+int ath11k_dp_rx_ampdu_start(struct ath11k *ar,
+			     struct ieee80211_ampdu_params *params);
+int ath11k_dp_rx_ampdu_stop(struct ath11k *ar,
+			    struct ieee80211_ampdu_params *params);
+void ath11k_peer_rx_tid_cleanup(struct ath11k *ar, struct ath11k_peer *peer);
+int ath11k_peer_rx_tid_setup(struct ath11k *ar, const u8 *peer_mac, int vdev_id,
+			     u8 tid, u32 ba_win_sz, u16 ssn);
+void ath11k_dp_htt_htc_t2h_msg_handler(struct ath11k_base *ab,
+				       struct sk_buff *skb);
+int ath11k_dp_rx_pdev_alloc(struct ath11k_base *ab, int pdev_idx);
+void ath11k_dp_rx_pdev_free(struct ath11k_base *ab, int pdev_idx);
+void ath11k_dp_reo_cmd_list_cleanup(struct ath11k_base *ab);
+void ath11k_dp_process_reo_status(struct ath11k_base *ab);
+int ath11k_dp_process_rxdma_err(struct ath11k_base *ab, int mac_id, int budget);
+int ath11k_dp_rx_process_wbm_err(struct ath11k_base *ab,
+				 struct napi_struct *napi, int budget);
+int ath11k_dp_process_rx_err(struct ath11k_base *ab, struct napi_struct *napi,
+			     int budget);
+int ath11k_dp_process_rx(struct ath11k_base *ab, int mac_id,
+			 struct napi_struct *napi, struct sk_buff_head *pending_q,
+			 int budget);
+int ath11k_dp_rxbufs_replenish(struct ath11k_base *ab, int mac_id,
+			       struct dp_rxdma_ring *rx_ring,
+			       int req_entries,
+			       enum hal_rx_buf_return_buf_manager mgr,
+			       gfp_t gfp);
+int ath11k_dp_htt_tlv_iter(struct ath11k_base *ab, const void *ptr, size_t len,
+			   int (*iter)(struct ath11k_base *ar, u16 tag, u16 len,
+				       const void *ptr, void *data),
+			   void *data);
+int ath11k_dp_rx_process_mon_rings(struct ath11k_base *ab, int mac_id,
+				   struct napi_struct *napi, int budget);
+int ath11k_dp_rx_process_mon_status(struct ath11k_base *ab, int mac_id,
+				    struct napi_struct *napi, int budget);
+int ath11k_dp_rx_mon_status_bufs_replenish(struct ath11k_base *ab, int mac_id,
+					   struct dp_rxdma_ring *rx_ring,
+					   int req_entries,
+					   enum hal_rx_buf_return_buf_manager mgr,
+					   gfp_t gfp);
+int ath11k_dp_rx_pdev_mon_detach(struct ath11k *ar);
+int ath11k_dp_rx_pdev_mon_attach(struct ath11k *ar);
+
+#endif /* ATH11K_DP_RX_H */
diff --git a/drivers/net/wireless/ath/ath11k/dp_tx.c b/drivers/net/wireless/ath/ath11k/dp_tx.c
new file mode 100644
index 000000000000..51a02b8e66cc
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/dp_tx.c
@@ -0,0 +1,943 @@
+// SPDX-License-Identifier: BSD-3-Clause-Clear
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#include "core.h"
+#include "dp_tx.h"
+#include "debug.h"
+#include "hw.h"
+
+/* NOTE: Any of the mapped ring id value must not exceed DP_TCL_NUM_RING_MAX */
+static const u8
+ath11k_txq_tcl_ring_map[ATH11K_HW_MAX_QUEUES] = { 0x0, 0x1, 0x2, 0x2 };
+
+static enum hal_tcl_encap_type
+ath11k_dp_tx_get_encap_type(struct ath11k_vif *arvif, struct sk_buff *skb)
+{
+	/* TODO: Determine encap type based on vif_type and configuration */
+	return HAL_TCL_ENCAP_TYPE_NATIVE_WIFI;
+}
+
+static void ath11k_dp_tx_encap_nwifi(struct sk_buff *skb)
+{
+	struct ieee80211_hdr *hdr = (void *)skb->data;
+	u8 *qos_ctl;
+
+	if (!ieee80211_is_data_qos(hdr->frame_control))
+		return;
+
+	qos_ctl = ieee80211_get_qos_ctl(hdr);
+	memmove(skb->data + IEEE80211_QOS_CTL_LEN,
+		skb->data, (void *)qos_ctl - (void *)skb->data);
+	skb_pull(skb, IEEE80211_QOS_CTL_LEN);
+
+	hdr = (void *)skb->data;
+	hdr->frame_control &= ~__cpu_to_le16(IEEE80211_STYPE_QOS_DATA);
+}
+
+static u8 ath11k_dp_tx_get_tid(struct sk_buff *skb)
+{
+	struct ieee80211_hdr *hdr = (void *)skb->data;
+
+	if (!ieee80211_is_data_qos(hdr->frame_control))
+		return HAL_DESC_REO_NON_QOS_TID;
+	else
+		return skb->priority & IEEE80211_QOS_CTL_TID_MASK;
+}
+
+static enum hal_encrypt_type ath11k_dp_tx_get_encrypt_type(u32 cipher)
+{
+	switch (cipher) {
+	case WLAN_CIPHER_SUITE_WEP40:
+		return HAL_ENCRYPT_TYPE_WEP_40;
+	case WLAN_CIPHER_SUITE_WEP104:
+		return HAL_ENCRYPT_TYPE_WEP_104;
+	case WLAN_CIPHER_SUITE_TKIP:
+		return HAL_ENCRYPT_TYPE_TKIP_MIC;
+	case WLAN_CIPHER_SUITE_CCMP:
+		return HAL_ENCRYPT_TYPE_CCMP_128;
+	case WLAN_CIPHER_SUITE_CCMP_256:
+		return HAL_ENCRYPT_TYPE_CCMP_256;
+	case WLAN_CIPHER_SUITE_GCMP:
+		return HAL_ENCRYPT_TYPE_GCMP_128;
+	case WLAN_CIPHER_SUITE_GCMP_256:
+		return HAL_ENCRYPT_TYPE_AES_GCMP_256;
+	default:
+		return HAL_ENCRYPT_TYPE_OPEN;
+	}
+}
+
+int ath11k_dp_tx(struct ath11k *ar, struct ath11k_vif *arvif,
+		 struct sk_buff *skb)
+{
+	struct ath11k_base *ab = ar->ab;
+	struct ath11k_dp *dp = &ab->dp;
+	struct hal_tx_info ti = {0};
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
+	struct ath11k_skb_cb *skb_cb = ATH11K_SKB_CB(skb);
+	struct hal_srng *tcl_ring;
+	struct ieee80211_hdr *hdr = (void *)skb->data;
+	struct dp_tx_ring *tx_ring;
+	u8 cached_desc[HAL_TCL_DESC_LEN];
+	void *hal_tcl_desc;
+	u8 pool_id;
+	u8 hal_ring_id;
+	int ret;
+
+	if (test_bit(ATH11K_FLAG_CRASH_FLUSH, &ar->ab->dev_flags))
+		return -ESHUTDOWN;
+
+	if (!ieee80211_is_data(hdr->frame_control))
+		return -ENOTSUPP;
+
+	pool_id = skb_get_queue_mapping(skb) & (ATH11K_HW_MAX_QUEUES - 1);
+	ti.ring_id = ath11k_txq_tcl_ring_map[pool_id];
+
+	tx_ring = &dp->tx_ring[ti.ring_id];
+
+	spin_lock_bh(&tx_ring->tx_idr_lock);
+	ret = idr_alloc(&tx_ring->txbuf_idr, skb, 0,
+			DP_TX_IDR_SIZE - 1, GFP_ATOMIC);
+	spin_unlock_bh(&tx_ring->tx_idr_lock);
+
+	if (ret < 0)
+		return -ENOSPC;
+
+	ti.desc_id = FIELD_PREP(DP_TX_DESC_ID_MAC_ID, ar->pdev_idx) |
+		     FIELD_PREP(DP_TX_DESC_ID_MSDU_ID, ret) |
+		     FIELD_PREP(DP_TX_DESC_ID_POOL_ID, pool_id);
+	ti.encap_type = ath11k_dp_tx_get_encap_type(arvif, skb);
+	ti.meta_data_flags = arvif->tcl_metadata;
+
+	if (info->control.hw_key)
+		ti.encrypt_type =
+			ath11k_dp_tx_get_encrypt_type(info->control.hw_key->cipher);
+	else
+		ti.encrypt_type = HAL_ENCRYPT_TYPE_OPEN;
+
+	ti.addr_search_flags = arvif->hal_addr_search_flags;
+	ti.search_type = arvif->search_type;
+	ti.type = HAL_TCL_DESC_TYPE_BUFFER;
+	ti.pkt_offset = 0;
+	ti.lmac_id = ar->lmac_id;
+	ti.bss_ast_hash = arvif->ast_hash;
+	ti.dscp_tid_tbl_idx = 0;
+
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
+		ti.flags0 |= FIELD_PREP(HAL_TCL_DATA_CMD_INFO1_IP4_CKSUM_EN, 1) |
+			     FIELD_PREP(HAL_TCL_DATA_CMD_INFO1_UDP4_CKSUM_EN, 1) |
+			     FIELD_PREP(HAL_TCL_DATA_CMD_INFO1_UDP6_CKSUM_EN, 1) |
+			     FIELD_PREP(HAL_TCL_DATA_CMD_INFO1_TCP4_CKSUM_EN, 1) |
+			     FIELD_PREP(HAL_TCL_DATA_CMD_INFO1_TCP6_CKSUM_EN, 1);
+	}
+
+	if (ieee80211_vif_is_mesh(arvif->vif))
+		ti.flags1 |= FIELD_PREP(HAL_TCL_DATA_CMD_INFO2_MESH_ENABLE, 1);
+
+	ti.flags1 |= FIELD_PREP(HAL_TCL_DATA_CMD_INFO2_TID_OVERWRITE, 1);
+
+	ti.tid = ath11k_dp_tx_get_tid(skb);
+
+	switch (ti.encap_type) {
+	case HAL_TCL_ENCAP_TYPE_NATIVE_WIFI:
+		ath11k_dp_tx_encap_nwifi(skb);
+		break;
+	case HAL_TCL_ENCAP_TYPE_RAW:
+		/*  TODO: for CHECKSUM_PARTIAL case in raw mode, HW checksum offload
+		 *	  is not applicable, hence manual checksum calculation using
+		 *	  skb_checksum_help() is needed
+		 */
+	case HAL_TCL_ENCAP_TYPE_ETHERNET:
+	case HAL_TCL_ENCAP_TYPE_802_3:
+		/* TODO: Take care of other encap modes as well */
+		ret = -EINVAL;
+		goto fail_remove_idr;
+	}
+
+	ti.paddr = dma_map_single(ab->dev, skb->data, skb->len, DMA_TO_DEVICE);
+	if (dma_mapping_error(ab->dev, ti.paddr)) {
+		ath11k_warn(ab, "failed to DMA map data Tx buffer\n");
+		ret = -ENOMEM;
+		goto fail_remove_idr;
+	}
+
+	ti.data_len = skb->len;
+	skb_cb->paddr = ti.paddr;
+	skb_cb->vif = arvif->vif;
+	skb_cb->ar = ar;
+
+	memset(cached_desc, 0, HAL_TCL_DESC_LEN);
+
+	ath11k_hal_tx_cmd_desc_setup(ab, cached_desc, &ti);
+
+	hal_ring_id = tx_ring->tcl_data_ring.ring_id;
+	tcl_ring = &ab->hal.srng_list[hal_ring_id];
+
+	spin_lock_bh(&tcl_ring->lock);
+
+	ath11k_hal_srng_access_begin(ab, tcl_ring);
+
+	hal_tcl_desc = (void *)ath11k_hal_srng_src_get_next_entry(ab, tcl_ring);
+	if (!hal_tcl_desc) {
+		/* NOTE: It is highly unlikely we'll be running out of tcl_ring
+		 * desc because the desc is directly enqueued onto hw queue.
+		 * So add tx packet throttling logic in future if required.
+		 */
+		ath11k_hal_srng_access_end(ab, tcl_ring);
+		spin_unlock_bh(&tcl_ring->lock);
+		ret = -ENOMEM;
+		goto fail_unmap_dma;
+	}
+
+	ath11k_hal_tx_desc_sync(cached_desc, hal_tcl_desc);
+
+	ath11k_hal_srng_access_end(ab, tcl_ring);
+
+	spin_unlock_bh(&tcl_ring->lock);
+
+	spin_lock_bh(&tx_ring->tx_idr_lock);
+	tx_ring->num_tx_pending++;
+	spin_unlock_bh(&tx_ring->tx_idr_lock);
+
+	atomic_inc(&ar->dp.num_tx_pending);
+
+	return 0;
+
+fail_unmap_dma:
+	dma_unmap_single(ab->dev, ti.paddr, ti.data_len, DMA_TO_DEVICE);
+
+fail_remove_idr:
+	spin_lock_bh(&tx_ring->tx_idr_lock);
+	idr_remove(&tx_ring->txbuf_idr,
+		   FIELD_GET(DP_TX_DESC_ID_MSDU_ID, ti.desc_id));
+	spin_unlock_bh(&tx_ring->tx_idr_lock);
+
+	return ret;
+}
+
+static void ath11k_dp_tx_free_txbuf(struct ath11k_base *ab, u8 mac_id,
+				    int msdu_id,
+				    struct dp_tx_ring *tx_ring)
+{
+	struct ath11k *ar;
+	struct sk_buff *msdu;
+	struct ath11k_skb_cb *skb_cb;
+
+	spin_lock_bh(&tx_ring->tx_idr_lock);
+	msdu = idr_find(&tx_ring->txbuf_idr, msdu_id);
+	if (!msdu) {
+		ath11k_warn(ab, "tx completion for unknown msdu_id %d\n",
+			    msdu_id);
+		spin_unlock_bh(&tx_ring->tx_idr_lock);
+		return;
+	}
+
+	skb_cb = ATH11K_SKB_CB(msdu);
+
+	idr_remove(&tx_ring->txbuf_idr, msdu_id);
+	tx_ring->num_tx_pending--;
+	spin_unlock_bh(&tx_ring->tx_idr_lock);
+
+	dma_unmap_single(ab->dev, skb_cb->paddr, msdu->len, DMA_TO_DEVICE);
+	dev_kfree_skb_any(msdu);
+
+	ar = ab->pdevs[mac_id].ar;
+	if (atomic_dec_and_test(&ar->dp.num_tx_pending))
+		wake_up(&ar->dp.tx_empty_waitq);
+}
+
+static void
+ath11k_dp_tx_htt_tx_complete_buf(struct ath11k_base *ab,
+				 struct dp_tx_ring *tx_ring,
+				 struct ath11k_dp_htt_wbm_tx_status *ts)
+{
+	struct sk_buff *msdu;
+	struct ieee80211_tx_info *info;
+	struct ath11k_skb_cb *skb_cb;
+	struct ath11k *ar;
+
+	spin_lock_bh(&tx_ring->tx_idr_lock);
+	msdu = idr_find(&tx_ring->txbuf_idr, ts->msdu_id);
+	if (!msdu) {
+		ath11k_warn(ab, "htt tx completion for unknown msdu_id %d\n",
+			    ts->msdu_id);
+		spin_unlock_bh(&tx_ring->tx_idr_lock);
+		return;
+	}
+
+	skb_cb = ATH11K_SKB_CB(msdu);
+	info = IEEE80211_SKB_CB(msdu);
+
+	ar = skb_cb->ar;
+
+	idr_remove(&tx_ring->txbuf_idr, ts->msdu_id);
+	tx_ring->num_tx_pending--;
+	spin_unlock_bh(&tx_ring->tx_idr_lock);
+
+	if (atomic_dec_and_test(&ar->dp.num_tx_pending))
+		wake_up(&ar->dp.tx_empty_waitq);
+
+	dma_unmap_single(ab->dev, skb_cb->paddr, msdu->len, DMA_TO_DEVICE);
+
+	memset(&info->status, 0, sizeof(info->status));
+
+	if (ts->acked) {
+		if (!(info->flags & IEEE80211_TX_CTL_NO_ACK)) {
+			info->flags |= IEEE80211_TX_STAT_ACK;
+			info->status.ack_signal = ATH11K_DEFAULT_NOISE_FLOOR +
+						  ts->ack_rssi;
+			info->status.is_valid_ack_signal = true;
+		} else {
+			info->flags |= IEEE80211_TX_STAT_NOACK_TRANSMITTED;
+		}
+	}
+
+	ieee80211_tx_status(ar->hw, msdu);
+}
+
+static void
+ath11k_dp_tx_process_htt_tx_complete(struct ath11k_base *ab,
+				     void *desc, u8 mac_id,
+				     u32 msdu_id, struct dp_tx_ring *tx_ring)
+{
+	struct htt_tx_wbm_completion *status_desc;
+	struct ath11k_dp_htt_wbm_tx_status ts = {0};
+	enum hal_wbm_htt_tx_comp_status wbm_status;
+
+	status_desc = desc + HTT_TX_WBM_COMP_STATUS_OFFSET;
+
+	wbm_status = FIELD_GET(HTT_TX_WBM_COMP_INFO0_STATUS,
+			       status_desc->info0);
+
+	switch (wbm_status) {
+	case HAL_WBM_REL_HTT_TX_COMP_STATUS_OK:
+	case HAL_WBM_REL_HTT_TX_COMP_STATUS_DROP:
+	case HAL_WBM_REL_HTT_TX_COMP_STATUS_TTL:
+		ts.acked = (wbm_status == HAL_WBM_REL_HTT_TX_COMP_STATUS_OK);
+		ts.msdu_id = msdu_id;
+		ts.ack_rssi = FIELD_GET(HTT_TX_WBM_COMP_INFO1_ACK_RSSI,
+					status_desc->info1);
+		ath11k_dp_tx_htt_tx_complete_buf(ab, tx_ring, &ts);
+		break;
+	case HAL_WBM_REL_HTT_TX_COMP_STATUS_REINJ:
+	case HAL_WBM_REL_HTT_TX_COMP_STATUS_INSPECT:
+		ath11k_dp_tx_free_txbuf(ab, mac_id, msdu_id, tx_ring);
+		break;
+	case HAL_WBM_REL_HTT_TX_COMP_STATUS_MEC_NOTIFY:
+		/* This event is to be handled only when the driver decides to
+		 * use WDS offload functionality.
+		 */
+		break;
+	default:
+		ath11k_warn(ab, "Unknown htt tx status %d\n", wbm_status);
+		break;
+	}
+}
+
+static void ath11k_dp_tx_cache_peer_stats(struct ath11k *ar,
+					  struct sk_buff *msdu,
+					  struct hal_tx_status *ts)
+{
+	struct ath11k_per_peer_tx_stats *peer_stats = &ar->cached_stats;
+
+	if (ts->try_cnt > 1) {
+		peer_stats->retry_pkts += ts->try_cnt - 1;
+		peer_stats->retry_bytes += (ts->try_cnt - 1) * msdu->len;
+
+		if (ts->status != HAL_WBM_TQM_REL_REASON_FRAME_ACKED) {
+			peer_stats->failed_pkts += 1;
+			peer_stats->failed_bytes += msdu->len;
+		}
+	}
+}
+
+static void ath11k_dp_tx_complete_msdu(struct ath11k *ar,
+				       struct sk_buff *msdu,
+				       struct hal_tx_status *ts)
+{
+	struct ath11k_base *ab = ar->ab;
+	struct ieee80211_tx_info *info;
+	struct ath11k_skb_cb *skb_cb;
+
+	if (WARN_ON_ONCE(ts->buf_rel_source != HAL_WBM_REL_SRC_MODULE_TQM)) {
+		/* Must not happen */
+		return;
+	}
+
+	skb_cb = ATH11K_SKB_CB(msdu);
+
+	dma_unmap_single(ab->dev, skb_cb->paddr, msdu->len, DMA_TO_DEVICE);
+
+	rcu_read_lock();
+
+	if (!rcu_dereference(ab->pdevs_active[ar->pdev_idx])) {
+		dev_kfree_skb_any(msdu);
+		goto exit;
+	}
+
+	if (!skb_cb->vif) {
+		dev_kfree_skb_any(msdu);
+		goto exit;
+	}
+
+	info = IEEE80211_SKB_CB(msdu);
+	memset(&info->status, 0, sizeof(info->status));
+
+	/* skip tx rate update from ieee80211_status*/
+	info->status.rates[0].idx = -1;
+
+	if (ts->status == HAL_WBM_TQM_REL_REASON_FRAME_ACKED &&
+	    !(info->flags & IEEE80211_TX_CTL_NO_ACK)) {
+		info->flags |= IEEE80211_TX_STAT_ACK;
+		info->status.ack_signal = ATH11K_DEFAULT_NOISE_FLOOR +
+					  ts->ack_rssi;
+		info->status.is_valid_ack_signal = true;
+	}
+
+	if (ts->status == HAL_WBM_TQM_REL_REASON_CMD_REMOVE_TX &&
+	    (info->flags & IEEE80211_TX_CTL_NO_ACK))
+		info->flags |= IEEE80211_TX_STAT_NOACK_TRANSMITTED;
+
+	if (ath11k_debug_is_extd_tx_stats_enabled(ar)) {
+		if (ts->flags & HAL_TX_STATUS_FLAGS_FIRST_MSDU) {
+			if (ar->last_ppdu_id == 0) {
+				ar->last_ppdu_id = ts->ppdu_id;
+			} else if (ar->last_ppdu_id == ts->ppdu_id ||
+				   ar->cached_ppdu_id == ar->last_ppdu_id) {
+				ar->cached_ppdu_id = ar->last_ppdu_id;
+				ar->cached_stats.is_ampdu = true;
+				ath11k_update_per_peer_stats_from_txcompl(ar, msdu, ts);
+				memset(&ar->cached_stats, 0,
+				       sizeof(struct ath11k_per_peer_tx_stats));
+			} else {
+				ar->cached_stats.is_ampdu = false;
+				ath11k_update_per_peer_stats_from_txcompl(ar, msdu, ts);
+				memset(&ar->cached_stats, 0,
+				       sizeof(struct ath11k_per_peer_tx_stats));
+			}
+			ar->last_ppdu_id = ts->ppdu_id;
+		}
+
+		ath11k_dp_tx_cache_peer_stats(ar, msdu, ts);
+	}
+
+	/* NOTE: Tx rate status reporting. Tx completion status does not have
+	 * necessary information (for example nss) to build the tx rate.
+	 * Might end up reporting it out-of-band from HTT stats.
+	 */
+
+	ieee80211_tx_status(ar->hw, msdu);
+
+exit:
+	rcu_read_unlock();
+}
+
+void ath11k_dp_tx_completion_handler(struct ath11k_base *ab, int ring_id)
+{
+	struct ath11k *ar;
+	struct ath11k_dp *dp = &ab->dp;
+	int hal_ring_id = dp->tx_ring[ring_id].tcl_comp_ring.ring_id;
+	struct hal_srng *status_ring = &ab->hal.srng_list[hal_ring_id];
+	struct sk_buff *msdu;
+	struct hal_wbm_release_ring tx_status;
+	struct hal_tx_status ts;
+	struct dp_tx_ring *tx_ring = &dp->tx_ring[ring_id];
+	u32 *desc;
+	u32 msdu_id;
+	u8 mac_id;
+
+	spin_lock_bh(&status_ring->lock);
+
+	ath11k_hal_srng_access_begin(ab, status_ring);
+
+	spin_lock_bh(&tx_ring->tx_status_lock);
+	while (!kfifo_is_full(&tx_ring->tx_status_fifo) &&
+	       (desc = ath11k_hal_srng_dst_get_next_entry(ab, status_ring))) {
+		ath11k_hal_tx_status_desc_sync((void *)desc,
+					       (void *)&tx_status);
+		kfifo_put(&tx_ring->tx_status_fifo, tx_status);
+	}
+
+	if ((ath11k_hal_srng_dst_peek(ab, status_ring) != NULL) &&
+	    kfifo_is_full(&tx_ring->tx_status_fifo)) {
+		/* TODO: Process pending tx_status messages when kfifo_is_full() */
+		ath11k_warn(ab, "Unable to process some of the tx_status ring desc because status_fifo is full\n");
+	}
+
+	spin_unlock_bh(&tx_ring->tx_status_lock);
+
+	ath11k_hal_srng_access_end(ab, status_ring);
+	spin_unlock_bh(&status_ring->lock);
+
+	spin_lock_bh(&tx_ring->tx_status_lock);
+	while (kfifo_get(&tx_ring->tx_status_fifo, &tx_status)) {
+		memset(&ts, 0, sizeof(ts));
+		ath11k_hal_tx_status_parse(ab, &tx_status, &ts);
+
+		mac_id = FIELD_GET(DP_TX_DESC_ID_MAC_ID, ts.desc_id);
+		msdu_id = FIELD_GET(DP_TX_DESC_ID_MSDU_ID, ts.desc_id);
+
+		if (ts.buf_rel_source == HAL_WBM_REL_SRC_MODULE_FW) {
+			ath11k_dp_tx_process_htt_tx_complete(ab,
+							     (void *)&tx_status,
+							     mac_id, msdu_id,
+							     tx_ring);
+			continue;
+		}
+
+		spin_lock_bh(&tx_ring->tx_idr_lock);
+		msdu = idr_find(&tx_ring->txbuf_idr, msdu_id);
+		if (!msdu) {
+			ath11k_warn(ab, "tx completion for unknown msdu_id %d\n",
+				    msdu_id);
+			spin_unlock_bh(&tx_ring->tx_idr_lock);
+			continue;
+		}
+		idr_remove(&tx_ring->txbuf_idr, msdu_id);
+		tx_ring->num_tx_pending--;
+		spin_unlock_bh(&tx_ring->tx_idr_lock);
+
+		ar = ab->pdevs[mac_id].ar;
+
+		if (atomic_dec_and_test(&ar->dp.num_tx_pending))
+			wake_up(&ar->dp.tx_empty_waitq);
+
+		/* TODO: Locking optimization so that tx_completion for an msdu
+		 * is not called with tx_status_lock acquired
+		 */
+		ath11k_dp_tx_complete_msdu(ar, msdu, &ts);
+	}
+	spin_unlock_bh(&tx_ring->tx_status_lock);
+}
+
+int ath11k_dp_tx_send_reo_cmd(struct ath11k_base *ab, struct dp_rx_tid *rx_tid,
+			      enum hal_reo_cmd_type type,
+			      struct ath11k_hal_reo_cmd *cmd,
+			      void (*cb)(struct ath11k_dp *, void *,
+					 enum hal_reo_cmd_status))
+{
+	struct ath11k_dp *dp = &ab->dp;
+	struct dp_reo_cmd *dp_cmd;
+	struct hal_srng *cmd_ring;
+	int cmd_num;
+
+	cmd_ring = &ab->hal.srng_list[dp->reo_cmd_ring.ring_id];
+	cmd_num = ath11k_hal_reo_cmd_send(ab, cmd_ring, type, cmd);
+
+	/* reo cmd ring descriptors has cmd_num starting from 1 */
+	if (cmd_num <= 0)
+		return -EINVAL;
+
+	if (!cb)
+		return 0;
+
+	/* Can this be optimized so that we keep the pending command list only
+	 * for tid delete command to free up the resoruce on the command status
+	 * indication?
+	 */
+	dp_cmd = kzalloc(sizeof(*dp_cmd), GFP_ATOMIC);
+
+	if (!dp_cmd)
+		return -ENOMEM;
+
+	memcpy(&dp_cmd->data, rx_tid, sizeof(struct dp_rx_tid));
+	dp_cmd->cmd_num = cmd_num;
+	dp_cmd->handler = cb;
+
+	spin_lock_bh(&dp->reo_cmd_lock);
+	list_add_tail(&dp_cmd->list, &dp->reo_cmd_list);
+	spin_unlock_bh(&dp->reo_cmd_lock);
+
+	return 0;
+}
+
+static int
+ath11k_dp_tx_get_ring_id_type(struct ath11k_base *ab,
+			      int mac_id, u32 ring_id,
+			      enum hal_ring_type ring_type,
+			      enum htt_srng_ring_type *htt_ring_type,
+			      enum htt_srng_ring_id *htt_ring_id)
+{
+	int lmac_ring_id_offset = 0;
+	int ret = 0;
+
+	switch (ring_type) {
+	case HAL_RXDMA_BUF:
+		lmac_ring_id_offset = mac_id * HAL_SRNG_RINGS_PER_LMAC;
+		if (!(ring_id == (HAL_SRNG_RING_ID_WMAC1_SW2RXDMA0_BUF +
+				  lmac_ring_id_offset) ||
+		    ring_id == (HAL_SRNG_RING_ID_WMAC1_SW2RXDMA1_BUF +
+				lmac_ring_id_offset))) {
+			ret = -EINVAL;
+		}
+		*htt_ring_id = HTT_RXDMA_HOST_BUF_RING;
+		*htt_ring_type = HTT_SW_TO_HW_RING;
+		break;
+	case HAL_RXDMA_DST:
+		*htt_ring_id = HTT_RXDMA_NON_MONITOR_DEST_RING;
+		*htt_ring_type = HTT_HW_TO_SW_RING;
+		break;
+	case HAL_RXDMA_MONITOR_BUF:
+		*htt_ring_id = HTT_RXDMA_MONITOR_BUF_RING;
+		*htt_ring_type = HTT_SW_TO_HW_RING;
+		break;
+	case HAL_RXDMA_MONITOR_STATUS:
+		*htt_ring_id = HTT_RXDMA_MONITOR_STATUS_RING;
+		*htt_ring_type = HTT_SW_TO_HW_RING;
+		break;
+	case HAL_RXDMA_MONITOR_DST:
+		*htt_ring_id = HTT_RXDMA_MONITOR_DEST_RING;
+		*htt_ring_type = HTT_HW_TO_SW_RING;
+		break;
+	case HAL_RXDMA_MONITOR_DESC:
+		*htt_ring_id = HTT_RXDMA_MONITOR_DESC_RING;
+		*htt_ring_type = HTT_SW_TO_HW_RING;
+		break;
+	default:
+		ath11k_warn(ab, "Unsupported ring type in DP :%d\n", ring_type);
+		ret = -EINVAL;
+	}
+	return ret;
+}
+
+int ath11k_dp_tx_htt_srng_setup(struct ath11k_base *ab, u32 ring_id,
+				int mac_id, enum hal_ring_type ring_type)
+{
+	struct htt_srng_setup_cmd *cmd;
+	struct hal_srng *srng = &ab->hal.srng_list[ring_id];
+	struct hal_srng_params params;
+	struct sk_buff *skb;
+	u32 ring_entry_sz;
+	int len = sizeof(*cmd);
+	dma_addr_t hp_addr, tp_addr;
+	enum htt_srng_ring_type htt_ring_type;
+	enum htt_srng_ring_id htt_ring_id;
+	int ret = 0;
+
+	skb = ath11k_htc_alloc_skb(ab, len);
+	if (!skb)
+		return -ENOMEM;
+
+	memset(&params, 0, sizeof(params));
+	ath11k_hal_srng_get_params(ab, srng, &params);
+
+	hp_addr = ath11k_hal_srng_get_hp_addr(ab, srng);
+	tp_addr = ath11k_hal_srng_get_tp_addr(ab, srng);
+
+	if (ath11k_dp_tx_get_ring_id_type(ab, mac_id, ring_id,
+					  ring_type, &htt_ring_type,
+					  &htt_ring_id))
+		goto err_free;
+
+	skb_put(skb, len);
+	cmd = (struct htt_srng_setup_cmd *)skb->data;
+	cmd->info0 = FIELD_PREP(HTT_SRNG_SETUP_CMD_INFO0_MSG_TYPE,
+				HTT_H2T_MSG_TYPE_SRING_SETUP);
+	if (htt_ring_type == HTT_SW_TO_HW_RING ||
+	    htt_ring_type == HTT_HW_TO_SW_RING)
+		cmd->info0 |= FIELD_PREP(HTT_SRNG_SETUP_CMD_INFO0_PDEV_ID,
+					 DP_SW2HW_MACID(mac_id));
+	else
+		cmd->info0 |= FIELD_PREP(HTT_SRNG_SETUP_CMD_INFO0_PDEV_ID,
+					 mac_id);
+	cmd->info0 |= FIELD_PREP(HTT_SRNG_SETUP_CMD_INFO0_RING_TYPE,
+				 htt_ring_type);
+	cmd->info0 |= FIELD_PREP(HTT_SRNG_SETUP_CMD_INFO0_RING_ID, htt_ring_id);
+
+	cmd->ring_base_addr_lo = params.ring_base_paddr &
+				 HAL_ADDR_LSB_REG_MASK;
+
+	cmd->ring_base_addr_hi = (u64)params.ring_base_paddr >>
+				 HAL_ADDR_MSB_REG_SHIFT;
+
+	ret = ath11k_hal_srng_get_entrysize(ring_type);
+	if (ret < 0)
+		return -EINVAL;
+
+	ring_entry_sz = ret;
+
+	ring_entry_sz >>= 2;
+	cmd->info1 = FIELD_PREP(HTT_SRNG_SETUP_CMD_INFO1_RING_ENTRY_SIZE,
+				ring_entry_sz);
+	cmd->info1 |= FIELD_PREP(HTT_SRNG_SETUP_CMD_INFO1_RING_SIZE,
+				 params.num_entries * ring_entry_sz);
+	cmd->info1 |= FIELD_PREP(HTT_SRNG_SETUP_CMD_INFO1_RING_FLAGS_MSI_SWAP,
+				 !!(params.flags & HAL_SRNG_FLAGS_MSI_SWAP));
+	cmd->info1 |= FIELD_PREP(
+			HTT_SRNG_SETUP_CMD_INFO1_RING_FLAGS_TLV_SWAP,
+			!!(params.flags & HAL_SRNG_FLAGS_DATA_TLV_SWAP));
+	cmd->info1 |= FIELD_PREP(
+			HTT_SRNG_SETUP_CMD_INFO1_RING_FLAGS_HOST_FW_SWAP,
+			!!(params.flags & HAL_SRNG_FLAGS_RING_PTR_SWAP));
+	if (htt_ring_type == HTT_SW_TO_HW_RING)
+		cmd->info1 |= HTT_SRNG_SETUP_CMD_INFO1_RING_LOOP_CNT_DIS;
+
+	cmd->ring_head_off32_remote_addr_lo = hp_addr & HAL_ADDR_LSB_REG_MASK;
+	cmd->ring_head_off32_remote_addr_hi = (u64)hp_addr >>
+					      HAL_ADDR_MSB_REG_SHIFT;
+
+	cmd->ring_tail_off32_remote_addr_lo = tp_addr & HAL_ADDR_LSB_REG_MASK;
+	cmd->ring_tail_off32_remote_addr_hi = (u64)tp_addr >>
+					      HAL_ADDR_MSB_REG_SHIFT;
+
+	cmd->ring_msi_addr_lo = 0;
+	cmd->ring_msi_addr_hi = 0;
+	cmd->msi_data = 0;
+
+	cmd->intr_info = FIELD_PREP(
+			HTT_SRNG_SETUP_CMD_INTR_INFO_BATCH_COUNTER_THRESH,
+			params.intr_batch_cntr_thres_entries * ring_entry_sz);
+	cmd->intr_info |= FIELD_PREP(
+			HTT_SRNG_SETUP_CMD_INTR_INFO_INTR_TIMER_THRESH,
+			params.intr_timer_thres_us >> 3);
+
+	cmd->info2 = 0;
+	if (params.flags & HAL_SRNG_FLAGS_LOW_THRESH_INTR_EN) {
+		cmd->info2 = FIELD_PREP(
+				HTT_SRNG_SETUP_CMD_INFO2_INTR_LOW_THRESH,
+				params.low_threshold);
+	}
+
+	ret = ath11k_htc_send(&ab->htc, ab->dp.eid, skb);
+	if (ret)
+		goto err_free;
+
+	return 0;
+
+err_free:
+	dev_kfree_skb_any(skb);
+
+	return ret;
+}
+
+#define HTT_TARGET_VERSION_TIMEOUT_HZ (3 * HZ)
+
+int ath11k_dp_tx_htt_h2t_ver_req_msg(struct ath11k_base *ab)
+{
+	struct ath11k_dp *dp = &ab->dp;
+	struct sk_buff *skb;
+	struct htt_ver_req_cmd *cmd;
+	int len = sizeof(*cmd);
+	int ret;
+
+	init_completion(&dp->htt_tgt_version_received);
+
+	skb = ath11k_htc_alloc_skb(ab, len);
+	if (!skb)
+		return -ENOMEM;
+
+	skb_put(skb, len);
+	cmd = (struct htt_ver_req_cmd *)skb->data;
+	cmd->ver_reg_info = FIELD_PREP(HTT_VER_REQ_INFO_MSG_ID,
+				       HTT_H2T_MSG_TYPE_VERSION_REQ);
+
+	ret = ath11k_htc_send(&ab->htc, dp->eid, skb);
+	if (ret) {
+		dev_kfree_skb_any(skb);
+		return ret;
+	}
+
+	ret = wait_for_completion_timeout(&dp->htt_tgt_version_received,
+					  HTT_TARGET_VERSION_TIMEOUT_HZ);
+	if (ret == 0) {
+		ath11k_warn(ab, "htt target version request timed out\n");
+		return -ETIMEDOUT;
+	}
+
+	if (dp->htt_tgt_ver_major != HTT_TARGET_VERSION_MAJOR) {
+		ath11k_err(ab, "unsupported htt major version %d supported version is %d\n",
+			   dp->htt_tgt_ver_major, HTT_TARGET_VERSION_MAJOR);
+		return -ENOTSUPP;
+	}
+
+	return 0;
+}
+
+int ath11k_dp_tx_htt_h2t_ppdu_stats_req(struct ath11k *ar, u32 mask)
+{
+	struct ath11k_base *ab = ar->ab;
+	struct ath11k_dp *dp = &ab->dp;
+	struct sk_buff *skb;
+	struct htt_ppdu_stats_cfg_cmd *cmd;
+	int len = sizeof(*cmd);
+	u8 pdev_mask;
+	int ret;
+
+	skb = ath11k_htc_alloc_skb(ab, len);
+	if (!skb)
+		return -ENOMEM;
+
+	skb_put(skb, len);
+	cmd = (struct htt_ppdu_stats_cfg_cmd *)skb->data;
+	cmd->msg = FIELD_PREP(HTT_PPDU_STATS_CFG_MSG_TYPE,
+			      HTT_H2T_MSG_TYPE_PPDU_STATS_CFG);
+
+	pdev_mask = 1 << (ar->pdev_idx);
+	cmd->msg |= FIELD_PREP(HTT_PPDU_STATS_CFG_PDEV_ID, pdev_mask);
+	cmd->msg |= FIELD_PREP(HTT_PPDU_STATS_CFG_TLV_TYPE_BITMASK, mask);
+
+	ret = ath11k_htc_send(&ab->htc, dp->eid, skb);
+	if (ret) {
+		dev_kfree_skb_any(skb);
+		return ret;
+	}
+
+	return 0;
+}
+
+int ath11k_dp_tx_htt_rx_filter_setup(struct ath11k_base *ab, u32 ring_id,
+				     int mac_id, enum hal_ring_type ring_type,
+				     int rx_buf_size,
+				     struct htt_rx_ring_tlv_filter *tlv_filter)
+{
+	struct htt_rx_ring_selection_cfg_cmd *cmd;
+	struct hal_srng *srng = &ab->hal.srng_list[ring_id];
+	struct hal_srng_params params;
+	struct sk_buff *skb;
+	int len = sizeof(*cmd);
+	enum htt_srng_ring_type htt_ring_type;
+	enum htt_srng_ring_id htt_ring_id;
+	int ret = 0;
+
+	skb = ath11k_htc_alloc_skb(ab, len);
+	if (!skb)
+		return -ENOMEM;
+
+	memset(&params, 0, sizeof(params));
+	ath11k_hal_srng_get_params(ab, srng, &params);
+
+	if (ath11k_dp_tx_get_ring_id_type(ab, mac_id, ring_id,
+					  ring_type, &htt_ring_type,
+					  &htt_ring_id))
+		goto err_free;
+
+	skb_put(skb, len);
+	cmd = (struct htt_rx_ring_selection_cfg_cmd *)skb->data;
+	cmd->info0 = FIELD_PREP(HTT_RX_RING_SELECTION_CFG_CMD_INFO0_MSG_TYPE,
+				HTT_H2T_MSG_TYPE_RX_RING_SELECTION_CFG);
+	if (htt_ring_type == HTT_SW_TO_HW_RING ||
+	    htt_ring_type == HTT_HW_TO_SW_RING)
+		cmd->info0 |=
+			FIELD_PREP(HTT_RX_RING_SELECTION_CFG_CMD_INFO0_PDEV_ID,
+				   DP_SW2HW_MACID(mac_id));
+	else
+		cmd->info0 |=
+			FIELD_PREP(HTT_RX_RING_SELECTION_CFG_CMD_INFO0_PDEV_ID,
+				   mac_id);
+	cmd->info0 |= FIELD_PREP(HTT_RX_RING_SELECTION_CFG_CMD_INFO0_RING_ID,
+				 htt_ring_id);
+	cmd->info0 |= FIELD_PREP(HTT_RX_RING_SELECTION_CFG_CMD_INFO0_SS,
+				 !!(params.flags & HAL_SRNG_FLAGS_MSI_SWAP));
+	cmd->info0 |= FIELD_PREP(HTT_RX_RING_SELECTION_CFG_CMD_INFO0_PS,
+				 !!(params.flags & HAL_SRNG_FLAGS_DATA_TLV_SWAP));
+
+	cmd->info1 = FIELD_PREP(HTT_RX_RING_SELECTION_CFG_CMD_INFO1_BUF_SIZE,
+				rx_buf_size);
+	cmd->pkt_type_en_flags0 = tlv_filter->pkt_filter_flags0;
+	cmd->pkt_type_en_flags1 = tlv_filter->pkt_filter_flags1;
+	cmd->pkt_type_en_flags2 = tlv_filter->pkt_filter_flags2;
+	cmd->pkt_type_en_flags3 = tlv_filter->pkt_filter_flags3;
+	cmd->rx_filter_tlv = tlv_filter->rx_filter;
+
+	ret = ath11k_htc_send(&ab->htc, ab->dp.eid, skb);
+	if (ret)
+		goto err_free;
+
+	return 0;
+
+err_free:
+	dev_kfree_skb_any(skb);
+
+	return ret;
+}
+
+int
+ath11k_dp_tx_htt_h2t_ext_stats_req(struct ath11k *ar, u8 type,
+				   struct htt_ext_stats_cfg_params *cfg_params,
+				   u64 cookie)
+{
+	struct ath11k_base *ab = ar->ab;
+	struct ath11k_dp *dp = &ab->dp;
+	struct sk_buff *skb;
+	struct htt_ext_stats_cfg_cmd *cmd;
+	int len = sizeof(*cmd);
+	int ret;
+
+	skb = ath11k_htc_alloc_skb(ab, len);
+	if (!skb)
+		return -ENOMEM;
+
+	skb_put(skb, len);
+
+	cmd = (struct htt_ext_stats_cfg_cmd *)skb->data;
+	memset(cmd, 0, sizeof(*cmd));
+	cmd->hdr.msg_type = HTT_H2T_MSG_TYPE_EXT_STATS_CFG;
+
+	cmd->hdr.pdev_mask = 1 << ar->pdev->pdev_id;
+
+	cmd->hdr.stats_type = type;
+	cmd->cfg_param0 = cfg_params->cfg0;
+	cmd->cfg_param1 = cfg_params->cfg1;
+	cmd->cfg_param2 = cfg_params->cfg2;
+	cmd->cfg_param3 = cfg_params->cfg3;
+	cmd->cookie_lsb = lower_32_bits(cookie);
+	cmd->cookie_msb = upper_32_bits(cookie);
+
+	ret = ath11k_htc_send(&ab->htc, dp->eid, skb);
+	if (ret) {
+		ath11k_warn(ab, "failed to send htt type stats request: %d",
+			    ret);
+		dev_kfree_skb_any(skb);
+		return ret;
+	}
+
+	return 0;
+}
+
+int ath11k_dp_tx_htt_monitor_mode_ring_config(struct ath11k *ar, bool reset)
+{
+	struct ath11k_pdev_dp *dp = &ar->dp;
+	struct htt_rx_ring_tlv_filter tlv_filter = {0};
+	int ret = 0, ring_id = 0;
+
+	ring_id = dp->rxdma_mon_buf_ring.refill_buf_ring.ring_id;
+
+	if (!reset) {
+		tlv_filter.rx_filter = HTT_RX_MON_FILTER_TLV_FLAGS_MON_BUF_RING;
+		tlv_filter.pkt_filter_flags0 =
+					HTT_RX_MON_FP_MGMT_FILTER_FLAGS0 |
+					HTT_RX_MON_MO_MGMT_FILTER_FLAGS0;
+		tlv_filter.pkt_filter_flags1 =
+					HTT_RX_MON_FP_MGMT_FILTER_FLAGS1 |
+					HTT_RX_MON_MO_MGMT_FILTER_FLAGS1;
+		tlv_filter.pkt_filter_flags2 =
+					HTT_RX_MON_FP_CTRL_FILTER_FLASG2 |
+					HTT_RX_MON_MO_CTRL_FILTER_FLASG2;
+		tlv_filter.pkt_filter_flags3 =
+					HTT_RX_MON_FP_CTRL_FILTER_FLASG3 |
+					HTT_RX_MON_MO_CTRL_FILTER_FLASG3 |
+					HTT_RX_MON_FP_DATA_FILTER_FLASG3 |
+					HTT_RX_MON_MO_DATA_FILTER_FLASG3;
+	}
+
+	ret = ath11k_dp_tx_htt_rx_filter_setup(ar->ab, ring_id, dp->mac_id,
+					       HAL_RXDMA_MONITOR_BUF,
+					       DP_RXDMA_REFILL_RING_SIZE,
+					       &tlv_filter);
+	if (ret)
+		return ret;
+
+	ring_id = dp->rx_mon_status_refill_ring.refill_buf_ring.ring_id;
+	if (!reset)
+		tlv_filter.rx_filter =
+				HTT_RX_MON_FILTER_TLV_FLAGS_MON_STATUS_RING;
+	else
+		tlv_filter = ath11k_mac_mon_status_filter_default;
+
+	ret = ath11k_dp_tx_htt_rx_filter_setup(ar->ab, ring_id, dp->mac_id,
+					       HAL_RXDMA_MONITOR_STATUS,
+					       DP_RXDMA_REFILL_RING_SIZE,
+					       &tlv_filter);
+	return ret;
+}
diff --git a/drivers/net/wireless/ath/ath11k/dp_tx.h b/drivers/net/wireless/ath/ath11k/dp_tx.h
new file mode 100644
index 000000000000..f8a9f9c8e444
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/dp_tx.h
@@ -0,0 +1,40 @@
+/* SPDX-License-Identifier: BSD-3-Clause-Clear */
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#ifndef ATH11K_DP_TX_H
+#define ATH11K_DP_TX_H
+
+#include "core.h"
+#include "hal_tx.h"
+
+struct ath11k_dp_htt_wbm_tx_status {
+	u32 msdu_id;
+	bool acked;
+	int ack_rssi;
+};
+
+int ath11k_dp_tx_htt_h2t_ver_req_msg(struct ath11k_base *ab);
+int ath11k_dp_tx(struct ath11k *ar, struct ath11k_vif *arvif,
+		 struct sk_buff *skb);
+void ath11k_dp_tx_completion_handler(struct ath11k_base *ab, int ring_id);
+int ath11k_dp_tx_send_reo_cmd(struct ath11k_base *ab, struct dp_rx_tid *rx_tid,
+			      enum hal_reo_cmd_type type,
+			      struct ath11k_hal_reo_cmd *cmd,
+			      void (*func)(struct ath11k_dp *, void *,
+					   enum hal_reo_cmd_status));
+
+int ath11k_dp_tx_htt_h2t_ppdu_stats_req(struct ath11k *ar, u32 mask);
+int
+ath11k_dp_tx_htt_h2t_ext_stats_req(struct ath11k *ar, u8 type,
+				   struct htt_ext_stats_cfg_params *cfg_params,
+				   u64 cookie);
+int ath11k_dp_tx_htt_monitor_mode_ring_config(struct ath11k *ar, bool reset);
+
+int ath11k_dp_tx_htt_rx_filter_setup(struct ath11k_base *ab, u32 ring_id,
+				     int mac_id, enum hal_ring_type ring_type,
+				     int rx_buf_size,
+				     struct htt_rx_ring_tlv_filter *tlv_filter);
+
+#endif
diff --git a/drivers/net/wireless/ath/ath11k/hal.c b/drivers/net/wireless/ath/ath11k/hal.c
new file mode 100644
index 000000000000..b58ac11c2747
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/hal.c
@@ -0,0 +1,1124 @@
+// SPDX-License-Identifier: BSD-3-Clause-Clear
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+#include <linux/dma-mapping.h>
+#include "ahb.h"
+#include "hal_tx.h"
+#include "debug.h"
+#include "hal_desc.h"
+
+static const struct hal_srng_config hw_srng_config[] = {
+	/* TODO: max_rings can populated by querying HW capabilities */
+	{ /* REO_DST */
+		.start_ring_id = HAL_SRNG_RING_ID_REO2SW1,
+		.max_rings = 4,
+		.entry_size = sizeof(struct hal_reo_dest_ring) >> 2,
+		.lmac_ring = false,
+		.ring_dir = HAL_SRNG_DIR_DST,
+		.reg_start = {
+			HAL_SEQ_WCSS_UMAC_REO_REG + HAL_REO1_RING_BASE_LSB,
+			HAL_SEQ_WCSS_UMAC_REO_REG + HAL_REO1_RING_HP,
+		},
+		.reg_size = {
+			HAL_REO2_RING_BASE_LSB - HAL_REO1_RING_BASE_LSB,
+			HAL_REO2_RING_HP - HAL_REO1_RING_HP,
+		},
+		.max_size = HAL_REO_REO2SW1_RING_BASE_MSB_RING_SIZE,
+	},
+	{ /* REO_EXCEPTION */
+		/* Designating REO2TCL ring as exception ring. This ring is
+		 * similar to other REO2SW rings though it is named as REO2TCL.
+		 * Any of theREO2SW rings can be used as exception ring.
+		 */
+		.start_ring_id = HAL_SRNG_RING_ID_REO2TCL,
+		.max_rings = 1,
+		.entry_size = sizeof(struct hal_reo_dest_ring) >> 2,
+		.lmac_ring = false,
+		.ring_dir = HAL_SRNG_DIR_DST,
+		.reg_start = {
+			HAL_SEQ_WCSS_UMAC_REO_REG + HAL_REO_TCL_RING_BASE_LSB,
+			HAL_SEQ_WCSS_UMAC_REO_REG + HAL_REO_TCL_RING_HP,
+		},
+		.max_size = HAL_REO_REO2TCL_RING_BASE_MSB_RING_SIZE,
+	},
+	{ /* REO_REINJECT */
+		.start_ring_id = HAL_SRNG_RING_ID_SW2REO,
+		.max_rings = 1,
+		.entry_size = sizeof(struct hal_reo_entrance_ring) >> 2,
+		.lmac_ring = false,
+		.ring_dir = HAL_SRNG_DIR_SRC,
+		.reg_start = {
+			HAL_SEQ_WCSS_UMAC_REO_REG + HAL_SW2REO_RING_BASE_LSB,
+			HAL_SEQ_WCSS_UMAC_REO_REG + HAL_SW2REO_RING_HP,
+		},
+		.max_size = HAL_REO_SW2REO_RING_BASE_MSB_RING_SIZE,
+	},
+	{ /* REO_CMD */
+		.start_ring_id = HAL_SRNG_RING_ID_REO_CMD,
+		.max_rings = 1,
+		.entry_size = (sizeof(struct hal_tlv_hdr) +
+			sizeof(struct hal_reo_get_queue_stats)) >> 2,
+		.lmac_ring = false,
+		.ring_dir = HAL_SRNG_DIR_SRC,
+		.reg_start = {
+			HAL_SEQ_WCSS_UMAC_REO_REG + HAL_REO_CMD_RING_BASE_LSB,
+			HAL_SEQ_WCSS_UMAC_REO_REG + HAL_REO_CMD_HP,
+		},
+		.max_size = HAL_REO_CMD_RING_BASE_MSB_RING_SIZE,
+	},
+	{ /* REO_STATUS */
+		.start_ring_id = HAL_SRNG_RING_ID_REO_STATUS,
+		.max_rings = 1,
+		.entry_size = (sizeof(struct hal_tlv_hdr) +
+			sizeof(struct hal_reo_get_queue_stats_status)) >> 2,
+		.lmac_ring = false,
+		.ring_dir = HAL_SRNG_DIR_DST,
+		.reg_start = {
+			HAL_SEQ_WCSS_UMAC_REO_REG +
+				HAL_REO_STATUS_RING_BASE_LSB,
+			HAL_SEQ_WCSS_UMAC_REO_REG + HAL_REO_STATUS_HP,
+		},
+		.max_size = HAL_REO_STATUS_RING_BASE_MSB_RING_SIZE,
+	},
+	{ /* TCL_DATA */
+		.start_ring_id = HAL_SRNG_RING_ID_SW2TCL1,
+		.max_rings = 3,
+		.entry_size = (sizeof(struct hal_tlv_hdr) +
+			     sizeof(struct hal_tcl_data_cmd)) >> 2,
+		.lmac_ring = false,
+		.ring_dir = HAL_SRNG_DIR_SRC,
+		.reg_start = {
+			HAL_SEQ_WCSS_UMAC_TCL_REG + HAL_TCL1_RING_BASE_LSB,
+			HAL_SEQ_WCSS_UMAC_TCL_REG + HAL_TCL1_RING_HP,
+		},
+		.reg_size = {
+			HAL_TCL2_RING_BASE_LSB - HAL_TCL1_RING_BASE_LSB,
+			HAL_TCL2_RING_HP - HAL_TCL1_RING_HP,
+		},
+		.max_size = HAL_SW2TCL1_RING_BASE_MSB_RING_SIZE,
+	},
+	{ /* TCL_CMD */
+		.start_ring_id = HAL_SRNG_RING_ID_SW2TCL_CMD,
+		.max_rings = 1,
+		.entry_size = (sizeof(struct hal_tlv_hdr) +
+			     sizeof(struct hal_tcl_gse_cmd)) >> 2,
+		.lmac_ring =  false,
+		.ring_dir = HAL_SRNG_DIR_SRC,
+		.reg_start = {
+			HAL_SEQ_WCSS_UMAC_TCL_REG + HAL_TCL_RING_BASE_LSB,
+			HAL_SEQ_WCSS_UMAC_TCL_REG + HAL_TCL_RING_HP,
+		},
+		.max_size = HAL_SW2TCL1_CMD_RING_BASE_MSB_RING_SIZE,
+	},
+	{ /* TCL_STATUS */
+		.start_ring_id = HAL_SRNG_RING_ID_TCL_STATUS,
+		.max_rings = 1,
+		.entry_size = (sizeof(struct hal_tlv_hdr) +
+			     sizeof(struct hal_tcl_status_ring)) >> 2,
+		.lmac_ring = false,
+		.ring_dir = HAL_SRNG_DIR_DST,
+		.reg_start = {
+			HAL_SEQ_WCSS_UMAC_TCL_REG +
+				HAL_TCL_STATUS_RING_BASE_LSB,
+			HAL_SEQ_WCSS_UMAC_TCL_REG + HAL_TCL_STATUS_RING_HP,
+		},
+		.max_size = HAL_TCL_STATUS_RING_BASE_MSB_RING_SIZE,
+	},
+	{ /* CE_SRC */
+		.start_ring_id = HAL_SRNG_RING_ID_CE0_SRC,
+		.max_rings = 12,
+		.entry_size = sizeof(struct hal_ce_srng_src_desc) >> 2,
+		.lmac_ring = false,
+		.ring_dir = HAL_SRNG_DIR_SRC,
+		.reg_start = {
+			(HAL_SEQ_WCSS_UMAC_CE0_SRC_REG +
+			 HAL_CE_DST_RING_BASE_LSB),
+			HAL_SEQ_WCSS_UMAC_CE0_SRC_REG + HAL_CE_DST_RING_HP,
+		},
+		.reg_size = {
+			(HAL_SEQ_WCSS_UMAC_CE1_SRC_REG -
+			 HAL_SEQ_WCSS_UMAC_CE0_SRC_REG),
+			(HAL_SEQ_WCSS_UMAC_CE1_SRC_REG -
+			 HAL_SEQ_WCSS_UMAC_CE0_SRC_REG),
+		},
+		.max_size = HAL_CE_SRC_RING_BASE_MSB_RING_SIZE,
+	},
+	{ /* CE_DST */
+		.start_ring_id = HAL_SRNG_RING_ID_CE0_DST,
+		.max_rings = 12,
+		.entry_size = sizeof(struct hal_ce_srng_dest_desc) >> 2,
+		.lmac_ring = false,
+		.ring_dir = HAL_SRNG_DIR_SRC,
+		.reg_start = {
+			(HAL_SEQ_WCSS_UMAC_CE0_DST_REG +
+			 HAL_CE_DST_RING_BASE_LSB),
+			HAL_SEQ_WCSS_UMAC_CE0_DST_REG + HAL_CE_DST_RING_HP,
+		},
+		.reg_size = {
+			(HAL_SEQ_WCSS_UMAC_CE1_DST_REG -
+			 HAL_SEQ_WCSS_UMAC_CE0_DST_REG),
+			(HAL_SEQ_WCSS_UMAC_CE1_DST_REG -
+			 HAL_SEQ_WCSS_UMAC_CE0_DST_REG),
+		},
+		.max_size = HAL_CE_DST_RING_BASE_MSB_RING_SIZE,
+	},
+	{ /* CE_DST_STATUS */
+		.start_ring_id = HAL_SRNG_RING_ID_CE0_DST_STATUS,
+		.max_rings = 12,
+		.entry_size = sizeof(struct hal_ce_srng_dst_status_desc) >> 2,
+		.lmac_ring = false,
+		.ring_dir = HAL_SRNG_DIR_DST,
+		.reg_start = {
+			(HAL_SEQ_WCSS_UMAC_CE0_DST_REG +
+			 HAL_CE_DST_STATUS_RING_BASE_LSB),
+			(HAL_SEQ_WCSS_UMAC_CE0_DST_REG +
+			 HAL_CE_DST_STATUS_RING_HP),
+		},
+		.reg_size = {
+			(HAL_SEQ_WCSS_UMAC_CE1_DST_REG -
+			 HAL_SEQ_WCSS_UMAC_CE0_DST_REG),
+			(HAL_SEQ_WCSS_UMAC_CE1_DST_REG -
+			 HAL_SEQ_WCSS_UMAC_CE0_DST_REG),
+		},
+		.max_size = HAL_CE_DST_STATUS_RING_BASE_MSB_RING_SIZE,
+	},
+	{ /* WBM_IDLE_LINK */
+		.start_ring_id = HAL_SRNG_RING_ID_WBM_IDLE_LINK,
+		.max_rings = 1,
+		.entry_size = sizeof(struct hal_wbm_link_desc) >> 2,
+		.lmac_ring = false,
+		.ring_dir = HAL_SRNG_DIR_SRC,
+		.reg_start = {
+			(HAL_SEQ_WCSS_UMAC_WBM_REG +
+			 HAL_WBM_IDLE_LINK_RING_BASE_LSB),
+			(HAL_SEQ_WCSS_UMAC_WBM_REG + HAL_WBM_IDLE_LINK_RING_HP),
+		},
+		.max_size = HAL_WBM_IDLE_LINK_RING_BASE_MSB_RING_SIZE,
+	},
+	{ /* SW2WBM_RELEASE */
+		.start_ring_id = HAL_SRNG_RING_ID_WBM_SW_RELEASE,
+		.max_rings = 1,
+		.entry_size = sizeof(struct hal_wbm_release_ring) >> 2,
+		.lmac_ring = false,
+		.ring_dir = HAL_SRNG_DIR_SRC,
+		.reg_start = {
+			(HAL_SEQ_WCSS_UMAC_WBM_REG +
+			 HAL_WBM_RELEASE_RING_BASE_LSB),
+			(HAL_SEQ_WCSS_UMAC_WBM_REG + HAL_WBM_RELEASE_RING_HP),
+		},
+		.max_size = HAL_SW2WBM_RELEASE_RING_BASE_MSB_RING_SIZE,
+	},
+	{ /* WBM2SW_RELEASE */
+		.start_ring_id = HAL_SRNG_RING_ID_WBM2SW0_RELEASE,
+		.max_rings = 4,
+		.entry_size = sizeof(struct hal_wbm_release_ring) >> 2,
+		.lmac_ring = false,
+		.ring_dir = HAL_SRNG_DIR_DST,
+		.reg_start = {
+			(HAL_SEQ_WCSS_UMAC_WBM_REG +
+			 HAL_WBM0_RELEASE_RING_BASE_LSB),
+			(HAL_SEQ_WCSS_UMAC_WBM_REG + HAL_WBM0_RELEASE_RING_HP),
+		},
+		.reg_size = {
+			(HAL_WBM1_RELEASE_RING_BASE_LSB -
+			 HAL_WBM0_RELEASE_RING_BASE_LSB),
+			(HAL_WBM1_RELEASE_RING_HP - HAL_WBM0_RELEASE_RING_HP),
+		},
+		.max_size = HAL_WBM2SW_RELEASE_RING_BASE_MSB_RING_SIZE,
+	},
+	{ /* RXDMA_BUF */
+		.start_ring_id = HAL_SRNG_RING_ID_WMAC1_SW2RXDMA0_BUF,
+		.max_rings = 2,
+		.entry_size = sizeof(struct hal_wbm_buffer_ring) >> 2,
+		.lmac_ring = true,
+		.ring_dir = HAL_SRNG_DIR_SRC,
+		.max_size = HAL_RXDMA_RING_MAX_SIZE,
+	},
+	{ /* RXDMA_DST */
+		.start_ring_id = HAL_SRNG_RING_ID_WMAC1_RXDMA2SW0,
+		.max_rings = 1,
+		.entry_size = sizeof(struct hal_reo_entrance_ring) >> 2,
+		.lmac_ring = true,
+		.ring_dir = HAL_SRNG_DIR_DST,
+		.max_size = HAL_RXDMA_RING_MAX_SIZE,
+	},
+	{ /* RXDMA_MONITOR_BUF */
+		.start_ring_id = HAL_SRNG_RING_ID_WMAC1_SW2RXDMA2_BUF,
+		.max_rings = 1,
+		.entry_size = sizeof(struct hal_wbm_buffer_ring) >> 2,
+		.lmac_ring = true,
+		.ring_dir = HAL_SRNG_DIR_SRC,
+		.max_size = HAL_RXDMA_RING_MAX_SIZE,
+	},
+	{ /* RXDMA_MONITOR_STATUS */
+		.start_ring_id = HAL_SRNG_RING_ID_WMAC1_SW2RXDMA1_STATBUF,
+		.max_rings = 1,
+		.entry_size = sizeof(struct hal_wbm_buffer_ring) >> 2,
+		.lmac_ring = true,
+		.ring_dir = HAL_SRNG_DIR_SRC,
+		.max_size = HAL_RXDMA_RING_MAX_SIZE,
+	},
+	{ /* RXDMA_MONITOR_DST */
+		.start_ring_id = HAL_SRNG_RING_ID_WMAC1_RXDMA2SW1,
+		.max_rings = 1,
+		.entry_size = sizeof(struct hal_reo_entrance_ring) >> 2,
+		.lmac_ring = true,
+		.ring_dir = HAL_SRNG_DIR_DST,
+		.max_size = HAL_RXDMA_RING_MAX_SIZE,
+	},
+	{ /* RXDMA_MONITOR_DESC */
+		.start_ring_id = HAL_SRNG_RING_ID_WMAC1_SW2RXDMA1_DESC,
+		.max_rings = 1,
+		.entry_size = sizeof(struct hal_wbm_buffer_ring) >> 2,
+		.lmac_ring = true,
+		.ring_dir = HAL_SRNG_DIR_SRC,
+		.max_size = HAL_RXDMA_RING_MAX_SIZE,
+	},
+	{ /* RXDMA DIR BUF */
+		.start_ring_id = HAL_SRNG_RING_ID_RXDMA_DIR_BUF,
+		.max_rings = 1,
+		.entry_size = 8 >> 2, /* TODO: Define the struct */
+		.lmac_ring = true,
+		.ring_dir = HAL_SRNG_DIR_SRC,
+		.max_size = HAL_RXDMA_RING_MAX_SIZE,
+	},
+};
+
+static int ath11k_hal_alloc_cont_rdp(struct ath11k_base *ab)
+{
+	struct ath11k_hal *hal = &ab->hal;
+	size_t size;
+
+	size = sizeof(u32) * HAL_SRNG_RING_ID_MAX;
+	hal->rdp.vaddr = dma_alloc_coherent(ab->dev, size, &hal->rdp.paddr,
+					    GFP_KERNEL);
+	if (!hal->rdp.vaddr)
+		return -ENOMEM;
+
+	return 0;
+}
+
+static void ath11k_hal_free_cont_rdp(struct ath11k_base *ab)
+{
+	struct ath11k_hal *hal = &ab->hal;
+	size_t size;
+
+	if (!hal->rdp.vaddr)
+		return;
+
+	size = sizeof(u32) * HAL_SRNG_RING_ID_MAX;
+	dma_free_coherent(ab->dev, size,
+			  hal->rdp.vaddr, hal->rdp.paddr);
+	hal->rdp.vaddr = NULL;
+}
+
+static int ath11k_hal_alloc_cont_wrp(struct ath11k_base *ab)
+{
+	struct ath11k_hal *hal = &ab->hal;
+	size_t size;
+
+	size = sizeof(u32) * HAL_SRNG_NUM_LMAC_RINGS;
+	hal->wrp.vaddr = dma_alloc_coherent(ab->dev, size, &hal->wrp.paddr,
+					    GFP_KERNEL);
+	if (!hal->wrp.vaddr)
+		return -ENOMEM;
+
+	return 0;
+}
+
+static void ath11k_hal_free_cont_wrp(struct ath11k_base *ab)
+{
+	struct ath11k_hal *hal = &ab->hal;
+	size_t size;
+
+	if (!hal->wrp.vaddr)
+		return;
+
+	size = sizeof(u32) * HAL_SRNG_NUM_LMAC_RINGS;
+	dma_free_coherent(ab->dev, size,
+			  hal->wrp.vaddr, hal->wrp.paddr);
+	hal->wrp.vaddr = NULL;
+}
+
+static void ath11k_hal_ce_dst_setup(struct ath11k_base *ab,
+				    struct hal_srng *srng, int ring_num)
+{
+	const struct hal_srng_config *srng_config = &hw_srng_config[HAL_CE_DST];
+	u32 addr;
+	u32 val;
+
+	addr = HAL_CE_DST_RING_CTRL +
+	       srng_config->reg_start[HAL_SRNG_REG_GRP_R0] +
+	       ring_num * srng_config->reg_size[HAL_SRNG_REG_GRP_R0];
+	val = ath11k_ahb_read32(ab, addr);
+	val &= ~HAL_CE_DST_R0_DEST_CTRL_MAX_LEN;
+	val |= FIELD_PREP(HAL_CE_DST_R0_DEST_CTRL_MAX_LEN,
+			  srng->u.dst_ring.max_buffer_length);
+	ath11k_ahb_write32(ab, addr, val);
+}
+
+static void ath11k_hal_srng_dst_hw_init(struct ath11k_base *ab,
+					struct hal_srng *srng)
+{
+	struct ath11k_hal *hal = &ab->hal;
+	u32 val;
+	u64 hp_addr;
+	u32 reg_base;
+
+	reg_base = srng->hwreg_base[HAL_SRNG_REG_GRP_R0];
+
+	if (srng->flags & HAL_SRNG_FLAGS_MSI_INTR) {
+		ath11k_ahb_write32(ab, reg_base +
+				       HAL_REO1_RING_MSI1_BASE_LSB_OFFSET,
+				   (u32)srng->msi_addr);
+
+		val = FIELD_PREP(HAL_REO1_RING_MSI1_BASE_MSB_ADDR,
+				 ((u64)srng->msi_addr >>
+				  HAL_ADDR_MSB_REG_SHIFT)) |
+		      HAL_REO1_RING_MSI1_BASE_MSB_MSI1_ENABLE;
+		ath11k_ahb_write32(ab, reg_base +
+				       HAL_REO1_RING_MSI1_BASE_MSB_OFFSET, val);
+
+		ath11k_ahb_write32(ab,
+				   reg_base + HAL_REO1_RING_MSI1_DATA_OFFSET,
+				   srng->msi_data);
+	}
+
+	ath11k_ahb_write32(ab, reg_base, (u32)srng->ring_base_paddr);
+
+	val = FIELD_PREP(HAL_REO1_RING_BASE_MSB_RING_BASE_ADDR_MSB,
+			 ((u64)srng->ring_base_paddr >>
+			  HAL_ADDR_MSB_REG_SHIFT)) |
+	      FIELD_PREP(HAL_REO1_RING_BASE_MSB_RING_SIZE,
+			 (srng->entry_size * srng->num_entries));
+	ath11k_ahb_write32(ab, reg_base + HAL_REO1_RING_BASE_MSB_OFFSET, val);
+
+	val = FIELD_PREP(HAL_REO1_RING_ID_RING_ID, srng->ring_id) |
+	      FIELD_PREP(HAL_REO1_RING_ID_ENTRY_SIZE, srng->entry_size);
+	ath11k_ahb_write32(ab, reg_base + HAL_REO1_RING_ID_OFFSET, val);
+
+	/* interrupt setup */
+	val = FIELD_PREP(HAL_REO1_RING_PRDR_INT_SETUP_INTR_TMR_THOLD,
+			 (srng->intr_timer_thres_us >> 3));
+
+	val |= FIELD_PREP(HAL_REO1_RING_PRDR_INT_SETUP_BATCH_COUNTER_THOLD,
+			  (srng->intr_batch_cntr_thres_entries *
+			   srng->entry_size));
+
+	ath11k_ahb_write32(ab,
+			   reg_base + HAL_REO1_RING_PRODUCER_INT_SETUP_OFFSET,
+			   val);
+
+	hp_addr = hal->rdp.paddr +
+		  ((unsigned long)srng->u.dst_ring.hp_addr -
+		   (unsigned long)hal->rdp.vaddr);
+	ath11k_ahb_write32(ab, reg_base + HAL_REO1_RING_HP_ADDR_LSB_OFFSET,
+			   hp_addr & HAL_ADDR_LSB_REG_MASK);
+	ath11k_ahb_write32(ab, reg_base + HAL_REO1_RING_HP_ADDR_MSB_OFFSET,
+			   hp_addr >> HAL_ADDR_MSB_REG_SHIFT);
+
+	/* Initialize head and tail pointers to indicate ring is empty */
+	reg_base = srng->hwreg_base[HAL_SRNG_REG_GRP_R2];
+	ath11k_ahb_write32(ab, reg_base, 0);
+	ath11k_ahb_write32(ab, reg_base + HAL_REO1_RING_TP_OFFSET, 0);
+	*srng->u.dst_ring.hp_addr = 0;
+
+	reg_base = srng->hwreg_base[HAL_SRNG_REG_GRP_R0];
+	val = 0;
+	if (srng->flags & HAL_SRNG_FLAGS_DATA_TLV_SWAP)
+		val |= HAL_REO1_RING_MISC_DATA_TLV_SWAP;
+	if (srng->flags & HAL_SRNG_FLAGS_RING_PTR_SWAP)
+		val |= HAL_REO1_RING_MISC_HOST_FW_SWAP;
+	if (srng->flags & HAL_SRNG_FLAGS_MSI_SWAP)
+		val |= HAL_REO1_RING_MISC_MSI_SWAP;
+	val |= HAL_REO1_RING_MISC_SRNG_ENABLE;
+
+	ath11k_ahb_write32(ab, reg_base + HAL_REO1_RING_MISC_OFFSET, val);
+}
+
+static void ath11k_hal_srng_src_hw_init(struct ath11k_base *ab,
+					struct hal_srng *srng)
+{
+	struct ath11k_hal *hal = &ab->hal;
+	u32 val;
+	u64 tp_addr;
+	u32 reg_base;
+
+	reg_base = srng->hwreg_base[HAL_SRNG_REG_GRP_R0];
+
+	if (srng->flags & HAL_SRNG_FLAGS_MSI_INTR) {
+		ath11k_ahb_write32(ab, reg_base +
+				       HAL_TCL1_RING_MSI1_BASE_LSB_OFFSET,
+				   (u32)srng->msi_addr);
+
+		val = FIELD_PREP(HAL_TCL1_RING_MSI1_BASE_MSB_ADDR,
+				 ((u64)srng->msi_addr >>
+				  HAL_ADDR_MSB_REG_SHIFT)) |
+		      HAL_TCL1_RING_MSI1_BASE_MSB_MSI1_ENABLE;
+		ath11k_ahb_write32(ab, reg_base +
+				       HAL_TCL1_RING_MSI1_BASE_MSB_OFFSET,
+				   val);
+
+		ath11k_ahb_write32(ab, reg_base +
+				       HAL_TCL1_RING_MSI1_DATA_OFFSET,
+				   srng->msi_data);
+	}
+
+	ath11k_ahb_write32(ab, reg_base, (u32)srng->ring_base_paddr);
+
+	val = FIELD_PREP(HAL_TCL1_RING_BASE_MSB_RING_BASE_ADDR_MSB,
+			 ((u64)srng->ring_base_paddr >>
+			  HAL_ADDR_MSB_REG_SHIFT)) |
+	      FIELD_PREP(HAL_TCL1_RING_BASE_MSB_RING_SIZE,
+			 (srng->entry_size * srng->num_entries));
+	ath11k_ahb_write32(ab, reg_base + HAL_TCL1_RING_BASE_MSB_OFFSET, val);
+
+	val = FIELD_PREP(HAL_REO1_RING_ID_ENTRY_SIZE, srng->entry_size);
+	ath11k_ahb_write32(ab, reg_base + HAL_TCL1_RING_ID_OFFSET, val);
+
+	/* interrupt setup */
+	/* NOTE: IPQ8074 v2 requires the interrupt timer threshold in the
+	 * unit of 8 usecs instead of 1 usec (as required by v1).
+	 */
+	val = FIELD_PREP(HAL_TCL1_RING_CONSR_INT_SETUP_IX0_INTR_TMR_THOLD,
+			 srng->intr_timer_thres_us);
+
+	val |= FIELD_PREP(HAL_TCL1_RING_CONSR_INT_SETUP_IX0_BATCH_COUNTER_THOLD,
+			  (srng->intr_batch_cntr_thres_entries *
+			   srng->entry_size));
+
+	ath11k_ahb_write32(ab,
+			   reg_base + HAL_TCL1_RING_CONSR_INT_SETUP_IX0_OFFSET,
+			   val);
+
+	val = 0;
+	if (srng->flags & HAL_SRNG_FLAGS_LOW_THRESH_INTR_EN) {
+		val |= FIELD_PREP(HAL_TCL1_RING_CONSR_INT_SETUP_IX1_LOW_THOLD,
+				  srng->u.src_ring.low_threshold);
+	}
+	ath11k_ahb_write32(ab,
+			   reg_base + HAL_TCL1_RING_CONSR_INT_SETUP_IX1_OFFSET,
+			   val);
+
+	if (srng->ring_id != HAL_SRNG_RING_ID_WBM_IDLE_LINK) {
+		tp_addr = hal->rdp.paddr +
+			  ((unsigned long)srng->u.src_ring.tp_addr -
+			   (unsigned long)hal->rdp.vaddr);
+		ath11k_ahb_write32(ab,
+				   reg_base + HAL_TCL1_RING_TP_ADDR_LSB_OFFSET,
+				   tp_addr & HAL_ADDR_LSB_REG_MASK);
+		ath11k_ahb_write32(ab,
+				   reg_base + HAL_TCL1_RING_TP_ADDR_MSB_OFFSET,
+				   tp_addr >> HAL_ADDR_MSB_REG_SHIFT);
+	}
+
+	/* Initialize head and tail pointers to indicate ring is empty */
+	reg_base = srng->hwreg_base[HAL_SRNG_REG_GRP_R2];
+	ath11k_ahb_write32(ab, reg_base, 0);
+	ath11k_ahb_write32(ab, reg_base + HAL_TCL1_RING_TP_OFFSET, 0);
+	*srng->u.src_ring.tp_addr = 0;
+
+	reg_base = srng->hwreg_base[HAL_SRNG_REG_GRP_R0];
+	val = 0;
+	if (srng->flags & HAL_SRNG_FLAGS_DATA_TLV_SWAP)
+		val |= HAL_TCL1_RING_MISC_DATA_TLV_SWAP;
+	if (srng->flags & HAL_SRNG_FLAGS_RING_PTR_SWAP)
+		val |= HAL_TCL1_RING_MISC_HOST_FW_SWAP;
+	if (srng->flags & HAL_SRNG_FLAGS_MSI_SWAP)
+		val |= HAL_TCL1_RING_MISC_MSI_SWAP;
+
+	/* Loop count is not used for SRC rings */
+	val |= HAL_TCL1_RING_MISC_MSI_LOOPCNT_DISABLE;
+
+	val |= HAL_TCL1_RING_MISC_SRNG_ENABLE;
+
+	ath11k_ahb_write32(ab, reg_base + HAL_TCL1_RING_MISC_OFFSET, val);
+}
+
+static void ath11k_hal_srng_hw_init(struct ath11k_base *ab,
+				    struct hal_srng *srng)
+{
+	if (srng->ring_dir == HAL_SRNG_DIR_SRC)
+		ath11k_hal_srng_src_hw_init(ab, srng);
+	else
+		ath11k_hal_srng_dst_hw_init(ab, srng);
+}
+
+static int ath11k_hal_srng_get_ring_id(struct ath11k_base *ab,
+				       enum hal_ring_type type,
+				       int ring_num, int mac_id)
+{
+	const struct hal_srng_config *srng_config = &hw_srng_config[type];
+	int ring_id;
+
+	if (ring_num >= srng_config->max_rings) {
+		ath11k_warn(ab, "invalid ring number :%d\n", ring_num);
+		return -EINVAL;
+	}
+
+	ring_id = srng_config->start_ring_id + ring_num;
+	if (srng_config->lmac_ring)
+		ring_id += mac_id * HAL_SRNG_RINGS_PER_LMAC;
+
+	if (WARN_ON(ring_id >= HAL_SRNG_RING_ID_MAX))
+		return -EINVAL;
+
+	return ring_id;
+}
+
+int ath11k_hal_srng_get_entrysize(u32 ring_type)
+{
+	const struct hal_srng_config *srng_config;
+
+	if (WARN_ON(ring_type >= HAL_MAX_RING_TYPES))
+		return -EINVAL;
+
+	srng_config = &hw_srng_config[ring_type];
+
+	return (srng_config->entry_size << 2);
+}
+
+int ath11k_hal_srng_get_max_entries(u32 ring_type)
+{
+	const struct hal_srng_config *srng_config;
+
+	if (WARN_ON(ring_type >= HAL_MAX_RING_TYPES))
+		return -EINVAL;
+
+	srng_config = &hw_srng_config[ring_type];
+
+	return (srng_config->max_size / srng_config->entry_size);
+}
+
+void ath11k_hal_srng_get_params(struct ath11k_base *ab, struct hal_srng *srng,
+				struct hal_srng_params *params)
+{
+	params->ring_base_paddr = srng->ring_base_paddr;
+	params->ring_base_vaddr = srng->ring_base_vaddr;
+	params->num_entries = srng->num_entries;
+	params->intr_timer_thres_us = srng->intr_timer_thres_us;
+	params->intr_batch_cntr_thres_entries =
+		srng->intr_batch_cntr_thres_entries;
+	params->low_threshold = srng->u.src_ring.low_threshold;
+	params->flags = srng->flags;
+}
+
+dma_addr_t ath11k_hal_srng_get_hp_addr(struct ath11k_base *ab,
+				       struct hal_srng *srng)
+{
+	if (!(srng->flags & HAL_SRNG_FLAGS_LMAC_RING))
+		return 0;
+
+	if (srng->ring_dir == HAL_SRNG_DIR_SRC)
+		return ab->hal.wrp.paddr +
+		       ((unsigned long)srng->u.src_ring.hp_addr -
+			(unsigned long)ab->hal.wrp.vaddr);
+	else
+		return ab->hal.rdp.paddr +
+		       ((unsigned long)srng->u.dst_ring.hp_addr -
+			 (unsigned long)ab->hal.rdp.vaddr);
+}
+
+dma_addr_t ath11k_hal_srng_get_tp_addr(struct ath11k_base *ab,
+				       struct hal_srng *srng)
+{
+	if (!(srng->flags & HAL_SRNG_FLAGS_LMAC_RING))
+		return 0;
+
+	if (srng->ring_dir == HAL_SRNG_DIR_SRC)
+		return ab->hal.rdp.paddr +
+		       ((unsigned long)srng->u.src_ring.tp_addr -
+			(unsigned long)ab->hal.rdp.vaddr);
+	else
+		return ab->hal.wrp.paddr +
+		       ((unsigned long)srng->u.dst_ring.tp_addr -
+			(unsigned long)ab->hal.wrp.vaddr);
+}
+
+u32 ath11k_hal_ce_get_desc_size(enum hal_ce_desc type)
+{
+	switch (type) {
+	case HAL_CE_DESC_SRC:
+		return sizeof(struct hal_ce_srng_src_desc);
+	case HAL_CE_DESC_DST:
+		return sizeof(struct hal_ce_srng_dest_desc);
+	case HAL_CE_DESC_DST_STATUS:
+		return sizeof(struct hal_ce_srng_dst_status_desc);
+	}
+
+	return 0;
+}
+
+void ath11k_hal_ce_src_set_desc(void *buf, dma_addr_t paddr, u32 len, u32 id,
+				u8 byte_swap_data)
+{
+	struct hal_ce_srng_src_desc *desc = (struct hal_ce_srng_src_desc *)buf;
+
+	desc->buffer_addr_low = paddr & HAL_ADDR_LSB_REG_MASK;
+	desc->buffer_addr_info =
+		FIELD_PREP(HAL_CE_SRC_DESC_ADDR_INFO_ADDR_HI,
+			   ((u64)paddr >> HAL_ADDR_MSB_REG_SHIFT)) |
+		FIELD_PREP(HAL_CE_SRC_DESC_ADDR_INFO_BYTE_SWAP,
+			   byte_swap_data) |
+		FIELD_PREP(HAL_CE_SRC_DESC_ADDR_INFO_GATHER, 0) |
+		FIELD_PREP(HAL_CE_SRC_DESC_ADDR_INFO_LEN, len);
+	desc->meta_info = FIELD_PREP(HAL_CE_SRC_DESC_META_INFO_DATA, id);
+}
+
+void ath11k_hal_ce_dst_set_desc(void *buf, dma_addr_t paddr)
+{
+	struct hal_ce_srng_dest_desc *desc =
+		(struct hal_ce_srng_dest_desc *)buf;
+
+	desc->buffer_addr_low = paddr & HAL_ADDR_LSB_REG_MASK;
+	desc->buffer_addr_info =
+		FIELD_PREP(HAL_CE_DEST_DESC_ADDR_INFO_ADDR_HI,
+			   ((u64)paddr >> HAL_ADDR_MSB_REG_SHIFT));
+}
+
+u32 ath11k_hal_ce_dst_status_get_length(void *buf)
+{
+	struct hal_ce_srng_dst_status_desc *desc =
+		(struct hal_ce_srng_dst_status_desc *)buf;
+	u32 len;
+
+	len = FIELD_GET(HAL_CE_DST_STATUS_DESC_FLAGS_LEN, desc->flags);
+	desc->flags &= ~HAL_CE_DST_STATUS_DESC_FLAGS_LEN;
+
+	return len;
+}
+
+void ath11k_hal_set_link_desc_addr(struct hal_wbm_link_desc *desc, u32 cookie,
+				   dma_addr_t paddr)
+{
+	desc->buf_addr_info.info0 = FIELD_PREP(BUFFER_ADDR_INFO0_ADDR,
+					       (paddr & HAL_ADDR_LSB_REG_MASK));
+	desc->buf_addr_info.info1 = FIELD_PREP(BUFFER_ADDR_INFO1_ADDR,
+					       ((u64)paddr >> HAL_ADDR_MSB_REG_SHIFT)) |
+				    FIELD_PREP(BUFFER_ADDR_INFO1_RET_BUF_MGR, 1) |
+				    FIELD_PREP(BUFFER_ADDR_INFO1_SW_COOKIE, cookie);
+}
+
+u32 *ath11k_hal_srng_dst_peek(struct ath11k_base *ab, struct hal_srng *srng)
+{
+	lockdep_assert_held(&srng->lock);
+
+	if (srng->u.dst_ring.tp != srng->u.dst_ring.cached_hp)
+		return (srng->ring_base_vaddr + srng->u.dst_ring.tp);
+
+	return NULL;
+}
+
+u32 *ath11k_hal_srng_dst_get_next_entry(struct ath11k_base *ab,
+					struct hal_srng *srng)
+{
+	u32 *desc;
+
+	lockdep_assert_held(&srng->lock);
+
+	if (srng->u.dst_ring.tp == srng->u.dst_ring.cached_hp)
+		return NULL;
+
+	desc = srng->ring_base_vaddr + srng->u.dst_ring.tp;
+
+	srng->u.dst_ring.tp = (srng->u.dst_ring.tp + srng->entry_size) %
+			      srng->ring_size;
+
+	return desc;
+}
+
+int ath11k_hal_srng_dst_num_free(struct ath11k_base *ab, struct hal_srng *srng,
+				 bool sync_hw_ptr)
+{
+	u32 tp, hp;
+
+	lockdep_assert_held(&srng->lock);
+
+	tp = srng->u.dst_ring.tp;
+
+	if (sync_hw_ptr) {
+		hp = *srng->u.dst_ring.hp_addr;
+		srng->u.dst_ring.cached_hp = hp;
+	} else {
+		hp = srng->u.dst_ring.cached_hp;
+	}
+
+	if (hp >= tp)
+		return (hp - tp) / srng->entry_size;
+	else
+		return (srng->ring_size - tp + hp) / srng->entry_size;
+}
+
+/* Returns number of available entries in src ring */
+int ath11k_hal_srng_src_num_free(struct ath11k_base *ab, struct hal_srng *srng,
+				 bool sync_hw_ptr)
+{
+	u32 tp, hp;
+
+	lockdep_assert_held(&srng->lock);
+
+	hp = srng->u.src_ring.hp;
+
+	if (sync_hw_ptr) {
+		tp = *srng->u.src_ring.tp_addr;
+		srng->u.src_ring.cached_tp = tp;
+	} else {
+		tp = srng->u.src_ring.cached_tp;
+	}
+
+	if (tp > hp)
+		return ((tp - hp) / srng->entry_size) - 1;
+	else
+		return ((srng->ring_size - hp + tp) / srng->entry_size) - 1;
+}
+
+u32 *ath11k_hal_srng_src_get_next_entry(struct ath11k_base *ab,
+					struct hal_srng *srng)
+{
+	u32 *desc;
+	u32 next_hp;
+
+	lockdep_assert_held(&srng->lock);
+
+	/* TODO: Using % is expensive, but we have to do this since size of some
+	 * SRNG rings is not power of 2 (due to descriptor sizes). Need to see
+	 * if separate function is defined for rings having power of 2 ring size
+	 * (TCL2SW, REO2SW, SW2RXDMA and CE rings) so that we can avoid the
+	 * overhead of % by using mask (with &).
+	 */
+	next_hp = (srng->u.src_ring.hp + srng->entry_size) % srng->ring_size;
+
+	if (next_hp == srng->u.src_ring.cached_tp)
+		return NULL;
+
+	desc = srng->ring_base_vaddr + srng->u.src_ring.hp;
+	srng->u.src_ring.hp = next_hp;
+
+	/* TODO: Reap functionality is not used by all rings. If particular
+	 * ring does not use reap functionality, we need not update reap_hp
+	 * with next_hp pointer. Need to make sure a separate function is used
+	 * before doing any optimization by removing below code updating
+	 * reap_hp.
+	 */
+	srng->u.src_ring.reap_hp = next_hp;
+
+	return desc;
+}
+
+u32 *ath11k_hal_srng_src_reap_next(struct ath11k_base *ab,
+				   struct hal_srng *srng)
+{
+	u32 *desc;
+	u32 next_reap_hp;
+
+	lockdep_assert_held(&srng->lock);
+
+	next_reap_hp = (srng->u.src_ring.reap_hp + srng->entry_size) %
+		       srng->ring_size;
+
+	if (next_reap_hp == srng->u.src_ring.cached_tp)
+		return NULL;
+
+	desc = srng->ring_base_vaddr + next_reap_hp;
+	srng->u.src_ring.reap_hp = next_reap_hp;
+
+	return desc;
+}
+
+u32 *ath11k_hal_srng_src_get_next_reaped(struct ath11k_base *ab,
+					 struct hal_srng *srng)
+{
+	u32 *desc;
+
+	lockdep_assert_held(&srng->lock);
+
+	if (srng->u.src_ring.hp == srng->u.src_ring.reap_hp)
+		return NULL;
+
+	desc = srng->ring_base_vaddr + srng->u.src_ring.hp;
+	srng->u.src_ring.hp = (srng->u.src_ring.hp + srng->entry_size) %
+			      srng->ring_size;
+
+	return desc;
+}
+
+u32 *ath11k_hal_srng_src_peek(struct ath11k_base *ab, struct hal_srng *srng)
+{
+	lockdep_assert_held(&srng->lock);
+
+	if (((srng->u.src_ring.hp + srng->entry_size) % srng->ring_size) ==
+	    srng->u.src_ring.cached_tp)
+		return NULL;
+
+	return srng->ring_base_vaddr + srng->u.src_ring.hp;
+}
+
+void ath11k_hal_srng_access_begin(struct ath11k_base *ab, struct hal_srng *srng)
+{
+	lockdep_assert_held(&srng->lock);
+
+	if (srng->ring_dir == HAL_SRNG_DIR_SRC)
+		srng->u.src_ring.cached_tp =
+			*(volatile u32 *)srng->u.src_ring.tp_addr;
+	else
+		srng->u.dst_ring.cached_hp = *srng->u.dst_ring.hp_addr;
+}
+
+/* Update cached ring head/tail pointers to HW. ath11k_hal_srng_access_begin()
+ * should have been called before this.
+ */
+void ath11k_hal_srng_access_end(struct ath11k_base *ab, struct hal_srng *srng)
+{
+	lockdep_assert_held(&srng->lock);
+
+	/* TODO: See if we need a write memory barrier here */
+	if (srng->flags & HAL_SRNG_FLAGS_LMAC_RING) {
+		/* For LMAC rings, ring pointer updates are done through FW and
+		 * hence written to a shared memory location that is read by FW
+		 */
+		if (srng->ring_dir == HAL_SRNG_DIR_SRC)
+			*srng->u.src_ring.hp_addr = srng->u.src_ring.hp;
+		else
+			*srng->u.dst_ring.tp_addr = srng->u.dst_ring.tp;
+	} else {
+		if (srng->ring_dir == HAL_SRNG_DIR_SRC) {
+			ath11k_ahb_write32(ab,
+					   (unsigned long)srng->u.src_ring.hp_addr -
+					   (unsigned long)ab->mem,
+					   srng->u.src_ring.hp);
+		} else {
+			ath11k_ahb_write32(ab,
+					   (unsigned long)srng->u.dst_ring.tp_addr -
+					   (unsigned long)ab->mem,
+					   srng->u.dst_ring.tp);
+		}
+	}
+}
+
+void ath11k_hal_setup_link_idle_list(struct ath11k_base *ab,
+				     struct hal_wbm_idle_scatter_list *sbuf,
+				     u32 nsbufs, u32 tot_link_desc,
+				     u32 end_offset)
+{
+	struct ath11k_buffer_addr *link_addr;
+	int i;
+	u32 reg_scatter_buf_sz = HAL_WBM_IDLE_SCATTER_BUF_SIZE / 64;
+
+	link_addr = (void *)sbuf[0].vaddr + HAL_WBM_IDLE_SCATTER_BUF_SIZE;
+
+	for (i = 1; i < nsbufs; i++) {
+		link_addr->info0 = sbuf[i].paddr & HAL_ADDR_LSB_REG_MASK;
+		link_addr->info1 = FIELD_PREP(
+				HAL_WBM_SCATTERED_DESC_MSB_BASE_ADDR_39_32,
+				(u64)sbuf[i].paddr >> HAL_ADDR_MSB_REG_SHIFT) |
+				FIELD_PREP(
+				HAL_WBM_SCATTERED_DESC_MSB_BASE_ADDR_MATCH_TAG,
+				BASE_ADDR_MATCH_TAG_VAL);
+
+		link_addr = (void *)sbuf[i].vaddr +
+			     HAL_WBM_IDLE_SCATTER_BUF_SIZE;
+	}
+
+	ath11k_ahb_write32(ab,
+			   HAL_SEQ_WCSS_UMAC_WBM_REG + HAL_WBM_R0_IDLE_LIST_CONTROL_ADDR,
+			   FIELD_PREP(HAL_WBM_SCATTER_BUFFER_SIZE, reg_scatter_buf_sz) |
+			   FIELD_PREP(HAL_WBM_LINK_DESC_IDLE_LIST_MODE, 0x1));
+	ath11k_ahb_write32(ab,
+			   HAL_SEQ_WCSS_UMAC_WBM_REG + HAL_WBM_R0_IDLE_LIST_SIZE_ADDR,
+			   FIELD_PREP(HAL_WBM_SCATTER_RING_SIZE_OF_IDLE_LINK_DESC_LIST,
+				      reg_scatter_buf_sz * nsbufs));
+	ath11k_ahb_write32(ab,
+			   HAL_SEQ_WCSS_UMAC_WBM_REG +
+			   HAL_WBM_SCATTERED_RING_BASE_LSB,
+			   FIELD_PREP(BUFFER_ADDR_INFO0_ADDR,
+				      sbuf[0].paddr & HAL_ADDR_LSB_REG_MASK));
+	ath11k_ahb_write32(ab,
+			   HAL_SEQ_WCSS_UMAC_WBM_REG +
+			   HAL_WBM_SCATTERED_RING_BASE_MSB,
+			   FIELD_PREP(
+				HAL_WBM_SCATTERED_DESC_MSB_BASE_ADDR_39_32,
+				(u64)sbuf[0].paddr >> HAL_ADDR_MSB_REG_SHIFT) |
+				FIELD_PREP(
+				HAL_WBM_SCATTERED_DESC_MSB_BASE_ADDR_MATCH_TAG,
+				BASE_ADDR_MATCH_TAG_VAL));
+
+	/* Setup head and tail pointers for the idle list */
+	ath11k_ahb_write32(ab,
+			   HAL_SEQ_WCSS_UMAC_WBM_REG +
+			   HAL_WBM_SCATTERED_DESC_PTR_HEAD_INFO_IX0,
+			   FIELD_PREP(BUFFER_ADDR_INFO0_ADDR,
+				      sbuf[nsbufs - 1].paddr));
+	ath11k_ahb_write32(ab,
+			   HAL_SEQ_WCSS_UMAC_WBM_REG +
+			   HAL_WBM_SCATTERED_DESC_PTR_HEAD_INFO_IX1,
+			   FIELD_PREP(
+				HAL_WBM_SCATTERED_DESC_MSB_BASE_ADDR_39_32,
+				((u64)sbuf[nsbufs - 1].paddr >>
+				 HAL_ADDR_MSB_REG_SHIFT)) |
+			   FIELD_PREP(HAL_WBM_SCATTERED_DESC_HEAD_P_OFFSET_IX1,
+				      (end_offset >> 2)));
+	ath11k_ahb_write32(ab,
+			   HAL_SEQ_WCSS_UMAC_WBM_REG +
+			   HAL_WBM_SCATTERED_DESC_PTR_HEAD_INFO_IX0,
+			   FIELD_PREP(BUFFER_ADDR_INFO0_ADDR,
+				      sbuf[0].paddr));
+
+	ath11k_ahb_write32(ab,
+			   HAL_SEQ_WCSS_UMAC_WBM_REG +
+			   HAL_WBM_SCATTERED_DESC_PTR_TAIL_INFO_IX0,
+			   FIELD_PREP(BUFFER_ADDR_INFO0_ADDR,
+				      sbuf[0].paddr));
+	ath11k_ahb_write32(ab,
+			   HAL_SEQ_WCSS_UMAC_WBM_REG +
+			   HAL_WBM_SCATTERED_DESC_PTR_TAIL_INFO_IX1,
+			   FIELD_PREP(
+				HAL_WBM_SCATTERED_DESC_MSB_BASE_ADDR_39_32,
+				((u64)sbuf[0].paddr >> HAL_ADDR_MSB_REG_SHIFT)) |
+			   FIELD_PREP(HAL_WBM_SCATTERED_DESC_TAIL_P_OFFSET_IX1,
+				      0));
+	ath11k_ahb_write32(ab,
+			   HAL_SEQ_WCSS_UMAC_WBM_REG +
+			   HAL_WBM_SCATTERED_DESC_PTR_HP_ADDR,
+			   2 * tot_link_desc);
+
+	/* Enable the SRNG */
+	ath11k_ahb_write32(ab,
+			   HAL_SEQ_WCSS_UMAC_WBM_REG +
+			   HAL_WBM_IDLE_LINK_RING_MISC_ADDR, 0x40);
+}
+
+int ath11k_hal_srng_setup(struct ath11k_base *ab, enum hal_ring_type type,
+			  int ring_num, int mac_id,
+			  struct hal_srng_params *params)
+{
+	struct ath11k_hal *hal = &ab->hal;
+	const struct hal_srng_config *srng_config = &hw_srng_config[type];
+	struct hal_srng *srng;
+	int ring_id;
+	u32 lmac_idx;
+	int i;
+	u32 reg_base;
+
+	ring_id = ath11k_hal_srng_get_ring_id(ab, type, ring_num, mac_id);
+	if (ring_id < 0)
+		return ring_id;
+
+	srng = &hal->srng_list[ring_id];
+
+	srng->ring_id = ring_id;
+	srng->ring_dir = srng_config->ring_dir;
+	srng->ring_base_paddr = params->ring_base_paddr;
+	srng->ring_base_vaddr = params->ring_base_vaddr;
+	srng->entry_size = srng_config->entry_size;
+	srng->num_entries = params->num_entries;
+	srng->ring_size = srng->entry_size * srng->num_entries;
+	srng->intr_batch_cntr_thres_entries =
+				params->intr_batch_cntr_thres_entries;
+	srng->intr_timer_thres_us = params->intr_timer_thres_us;
+	srng->flags = params->flags;
+	spin_lock_init(&srng->lock);
+
+	for (i = 0; i < HAL_SRNG_NUM_REG_GRP; i++) {
+		srng->hwreg_base[i] = srng_config->reg_start[i] +
+				      (ring_num * srng_config->reg_size[i]);
+	}
+
+	memset(srng->ring_base_vaddr, 0,
+	       (srng->entry_size * srng->num_entries) << 2);
+
+	/* TODO: Add comments on these swap configurations */
+	if (IS_ENABLED(CONFIG_CPU_BIG_ENDIAN))
+		srng->flags |= HAL_SRNG_FLAGS_MSI_SWAP | HAL_SRNG_FLAGS_DATA_TLV_SWAP |
+			       HAL_SRNG_FLAGS_RING_PTR_SWAP;
+
+	reg_base = srng->hwreg_base[HAL_SRNG_REG_GRP_R2];
+
+	if (srng->ring_dir == HAL_SRNG_DIR_SRC) {
+		srng->u.src_ring.hp = 0;
+		srng->u.src_ring.cached_tp = 0;
+		srng->u.src_ring.reap_hp = srng->ring_size - srng->entry_size;
+		srng->u.src_ring.tp_addr = (void *)(hal->rdp.vaddr + ring_id);
+		srng->u.src_ring.low_threshold = params->low_threshold *
+						 srng->entry_size;
+		if (srng_config->lmac_ring) {
+			lmac_idx = ring_id - HAL_SRNG_RING_ID_LMAC1_ID_START;
+			srng->u.src_ring.hp_addr = (void *)(hal->wrp.vaddr +
+						   lmac_idx);
+			srng->flags |= HAL_SRNG_FLAGS_LMAC_RING;
+		} else {
+			srng->u.src_ring.hp_addr =
+				(u32 *)((unsigned long)ab->mem + reg_base);
+		}
+	} else {
+		/* During initialization loop count in all the descriptors
+		 * will be set to zero, and HW will set it to 1 on completing
+		 * descriptor update in first loop, and increments it by 1 on
+		 * subsequent loops (loop count wraps around after reaching
+		 * 0xffff). The 'loop_cnt' in SW ring state is the expected
+		 * loop count in descriptors updated by HW (to be processed
+		 * by SW).
+		 */
+		srng->u.dst_ring.loop_cnt = 1;
+		srng->u.dst_ring.tp = 0;
+		srng->u.dst_ring.cached_hp = 0;
+		srng->u.dst_ring.hp_addr = (void *)(hal->rdp.vaddr + ring_id);
+		if (srng_config->lmac_ring) {
+			/* For LMAC rings, tail pointer updates will be done
+			 * through FW by writing to a shared memory location
+			 */
+			lmac_idx = ring_id - HAL_SRNG_RING_ID_LMAC1_ID_START;
+			srng->u.dst_ring.tp_addr = (void *)(hal->wrp.vaddr +
+						   lmac_idx);
+			srng->flags |= HAL_SRNG_FLAGS_LMAC_RING;
+		} else {
+			srng->u.dst_ring.tp_addr =
+				(u32 *)((unsigned long)ab->mem + reg_base +
+					(HAL_REO1_RING_TP - HAL_REO1_RING_HP));
+		}
+	}
+
+	if (srng_config->lmac_ring)
+		return ring_id;
+
+	ath11k_hal_srng_hw_init(ab, srng);
+
+	if (type == HAL_CE_DST) {
+		srng->u.dst_ring.max_buffer_length = params->max_buffer_len;
+		ath11k_hal_ce_dst_setup(ab, srng, ring_num);
+	}
+
+	return ring_id;
+}
+
+int ath11k_hal_srng_init(struct ath11k_base *ab)
+{
+	struct ath11k_hal *hal = &ab->hal;
+	int ret;
+
+	memset(hal, 0, sizeof(*hal));
+
+	hal->srng_config = hw_srng_config;
+
+	ret = ath11k_hal_alloc_cont_rdp(ab);
+	if (ret)
+		goto err_hal;
+
+	ret = ath11k_hal_alloc_cont_wrp(ab);
+	if (ret)
+		goto err_free_cont_rdp;
+
+	return 0;
+
+err_free_cont_rdp:
+	ath11k_hal_free_cont_rdp(ab);
+
+err_hal:
+	return ret;
+}
+
+void ath11k_hal_srng_deinit(struct ath11k_base *ab)
+{
+	ath11k_hal_free_cont_rdp(ab);
+	ath11k_hal_free_cont_wrp(ab);
+}
diff --git a/drivers/net/wireless/ath/ath11k/hal.h b/drivers/net/wireless/ath/ath11k/hal.h
new file mode 100644
index 000000000000..5b13ccdf39e4
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/hal.h
@@ -0,0 +1,897 @@
+/* SPDX-License-Identifier: BSD-3-Clause-Clear */
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#ifndef ATH11K_HAL_H
+#define ATH11K_HAL_H
+
+#include "hal_desc.h"
+#include "rx_desc.h"
+
+struct ath11k_base;
+
+#define HAL_LINK_DESC_SIZE			(32 << 2)
+#define HAL_LINK_DESC_ALIGN			128
+#define HAL_NUM_MPDUS_PER_LINK_DESC		6
+#define HAL_NUM_TX_MSDUS_PER_LINK_DESC		7
+#define HAL_NUM_RX_MSDUS_PER_LINK_DESC		6
+#define HAL_NUM_MPDU_LINKS_PER_QUEUE_DESC	12
+#define HAL_MAX_AVAIL_BLK_RES			3
+
+#define HAL_RING_BASE_ALIGN	8
+
+#define HAL_WBM_IDLE_SCATTER_BUF_SIZE_MAX	32704
+/* TODO: Check with hw team on the supported scatter buf size */
+#define HAL_WBM_IDLE_SCATTER_NEXT_PTR_SIZE	8
+#define HAL_WBM_IDLE_SCATTER_BUF_SIZE (HAL_WBM_IDLE_SCATTER_BUF_SIZE_MAX - \
+				       HAL_WBM_IDLE_SCATTER_NEXT_PTR_SIZE)
+
+#define HAL_DSCP_TID_MAP_TBL_NUM_ENTRIES_MAX	48
+#define HAL_DSCP_TID_TBL_SIZE			24
+
+/* calculate the register address from bar0 of shadow register x */
+#define SHADOW_BASE_ADDRESS			0x00003024
+#define SHADOW_NUM_REGISTERS				36
+
+/* WCSS Relative address */
+#define HAL_SEQ_WCSS_UMAC_REO_REG		0x00a38000
+#define HAL_SEQ_WCSS_UMAC_TCL_REG		0x00a44000
+#define HAL_SEQ_WCSS_UMAC_CE0_SRC_REG		0x00a00000
+#define HAL_SEQ_WCSS_UMAC_CE0_DST_REG		0x00a01000
+#define HAL_SEQ_WCSS_UMAC_CE1_SRC_REG		0x00a02000
+#define HAL_SEQ_WCSS_UMAC_CE1_DST_REG		0x00a03000
+#define HAL_SEQ_WCSS_UMAC_WBM_REG		0x00a34000
+
+/* SW2TCL(x) R0 ring configuration address */
+#define HAL_TCL1_RING_CMN_CTRL_REG		0x00000014
+#define HAL_TCL1_RING_DSCP_TID_MAP		0x0000002c
+#define HAL_TCL1_RING_BASE_LSB			0x00000510
+#define HAL_TCL1_RING_BASE_MSB			0x00000514
+#define HAL_TCL1_RING_ID			0x00000518
+#define HAL_TCL1_RING_MISC			0x00000520
+#define HAL_TCL1_RING_TP_ADDR_LSB		0x0000052c
+#define HAL_TCL1_RING_TP_ADDR_MSB		0x00000530
+#define HAL_TCL1_RING_CONSUMER_INT_SETUP_IX0	0x00000540
+#define HAL_TCL1_RING_CONSUMER_INT_SETUP_IX1	0x00000544
+#define HAL_TCL1_RING_MSI1_BASE_LSB		0x00000558
+#define HAL_TCL1_RING_MSI1_BASE_MSB		0x0000055c
+#define HAL_TCL1_RING_MSI1_DATA			0x00000560
+#define HAL_TCL2_RING_BASE_LSB			0x00000568
+#define HAL_TCL_RING_BASE_LSB			0x00000618
+
+#define HAL_TCL1_RING_MSI1_BASE_LSB_OFFSET \
+		(HAL_TCL1_RING_MSI1_BASE_LSB - HAL_TCL1_RING_BASE_LSB)
+#define HAL_TCL1_RING_MSI1_BASE_MSB_OFFSET \
+		(HAL_TCL1_RING_MSI1_BASE_MSB - HAL_TCL1_RING_BASE_LSB)
+#define HAL_TCL1_RING_MSI1_DATA_OFFSET \
+		(HAL_TCL1_RING_MSI1_DATA - HAL_TCL1_RING_BASE_LSB)
+#define HAL_TCL1_RING_BASE_MSB_OFFSET \
+		(HAL_TCL1_RING_BASE_MSB - HAL_TCL1_RING_BASE_LSB)
+#define HAL_TCL1_RING_ID_OFFSET \
+		(HAL_TCL1_RING_ID - HAL_TCL1_RING_BASE_LSB)
+#define HAL_TCL1_RING_CONSR_INT_SETUP_IX0_OFFSET \
+		(HAL_TCL1_RING_CONSUMER_INT_SETUP_IX0 - HAL_TCL1_RING_BASE_LSB)
+#define HAL_TCL1_RING_CONSR_INT_SETUP_IX1_OFFSET \
+		(HAL_TCL1_RING_CONSUMER_INT_SETUP_IX1 - HAL_TCL1_RING_BASE_LSB)
+#define HAL_TCL1_RING_TP_ADDR_LSB_OFFSET \
+		(HAL_TCL1_RING_TP_ADDR_LSB - HAL_TCL1_RING_BASE_LSB)
+#define HAL_TCL1_RING_TP_ADDR_MSB_OFFSET \
+		(HAL_TCL1_RING_TP_ADDR_MSB - HAL_TCL1_RING_BASE_LSB)
+#define HAL_TCL1_RING_MISC_OFFSET \
+		(HAL_TCL1_RING_MISC - HAL_TCL1_RING_BASE_LSB)
+
+/* SW2TCL(x) R2 ring pointers (head/tail) address */
+#define HAL_TCL1_RING_HP			0x00002000
+#define HAL_TCL1_RING_TP			0x00002004
+#define HAL_TCL2_RING_HP			0x00002008
+#define HAL_TCL_RING_HP				0x00002018
+
+#define HAL_TCL1_RING_TP_OFFSET \
+		(HAL_TCL1_RING_TP - HAL_TCL1_RING_HP)
+
+/* TCL STATUS ring address */
+#define HAL_TCL_STATUS_RING_BASE_LSB		0x00000720
+#define HAL_TCL_STATUS_RING_HP			0x00002030
+
+/* REO2SW(x) R0 ring configuration address */
+#define HAL_REO1_GEN_ENABLE			0x00000000
+#define HAL_REO1_DEST_RING_CTRL_IX_2		0x0000000c
+#define HAL_REO1_DEST_RING_CTRL_IX_3		0x00000010
+#define HAL_REO1_RING_BASE_LSB			0x0000029c
+#define HAL_REO1_RING_BASE_MSB			0x000002a0
+#define HAL_REO1_RING_ID			0x000002a4
+#define HAL_REO1_RING_MISC			0x000002ac
+#define HAL_REO1_RING_HP_ADDR_LSB		0x000002b0
+#define HAL_REO1_RING_HP_ADDR_MSB		0x000002b4
+#define HAL_REO1_RING_PRODUCER_INT_SETUP	0x000002c0
+#define HAL_REO1_RING_MSI1_BASE_LSB		0x000002e4
+#define HAL_REO1_RING_MSI1_BASE_MSB		0x000002e8
+#define HAL_REO1_RING_MSI1_DATA			0x000002ec
+#define HAL_REO2_RING_BASE_LSB			0x000002f4
+#define HAL_REO1_AGING_THRESH_IX_0		0x00000564
+#define HAL_REO1_AGING_THRESH_IX_1		0x00000568
+#define HAL_REO1_AGING_THRESH_IX_2		0x0000056c
+#define HAL_REO1_AGING_THRESH_IX_3		0x00000570
+
+#define HAL_REO1_RING_MSI1_BASE_LSB_OFFSET \
+		(HAL_REO1_RING_MSI1_BASE_LSB - HAL_REO1_RING_BASE_LSB)
+#define HAL_REO1_RING_MSI1_BASE_MSB_OFFSET \
+		(HAL_REO1_RING_MSI1_BASE_MSB - HAL_REO1_RING_BASE_LSB)
+#define HAL_REO1_RING_MSI1_DATA_OFFSET \
+		(HAL_REO1_RING_MSI1_DATA - HAL_REO1_RING_BASE_LSB)
+#define HAL_REO1_RING_BASE_MSB_OFFSET \
+		(HAL_REO1_RING_BASE_MSB - HAL_REO1_RING_BASE_LSB)
+#define HAL_REO1_RING_ID_OFFSET (HAL_REO1_RING_ID - HAL_REO1_RING_BASE_LSB)
+#define HAL_REO1_RING_PRODUCER_INT_SETUP_OFFSET \
+		(HAL_REO1_RING_PRODUCER_INT_SETUP - HAL_REO1_RING_BASE_LSB)
+#define HAL_REO1_RING_HP_ADDR_LSB_OFFSET \
+		(HAL_REO1_RING_HP_ADDR_LSB - HAL_REO1_RING_BASE_LSB)
+#define HAL_REO1_RING_HP_ADDR_MSB_OFFSET \
+		(HAL_REO1_RING_HP_ADDR_MSB - HAL_REO1_RING_BASE_LSB)
+#define HAL_REO1_RING_MISC_OFFSET (HAL_REO1_RING_MISC - HAL_REO1_RING_BASE_LSB)
+
+/* REO2SW(x) R2 ring pointers (head/tail) address */
+#define HAL_REO1_RING_HP			0x00003038
+#define HAL_REO1_RING_TP			0x0000303c
+#define HAL_REO2_RING_HP			0x00003040
+
+#define HAL_REO1_RING_TP_OFFSET	(HAL_REO1_RING_TP - HAL_REO1_RING_HP)
+
+/* REO2TCL R0 ring configuration address */
+#define HAL_REO_TCL_RING_BASE_LSB		0x000003fc
+
+/* REO2TCL R2 ring pointer (head/tail) address */
+#define HAL_REO_TCL_RING_HP			0x00003058
+
+/* REO CMD R0 address */
+#define HAL_REO_CMD_RING_BASE_LSB		0x00000194
+
+/* REO CMD R2 address */
+#define HAL_REO_CMD_HP				0x00003020
+
+/* SW2REO R0 address */
+#define HAL_SW2REO_RING_BASE_LSB		0x000001ec
+
+/* SW2REO R2 address */
+#define HAL_SW2REO_RING_HP			0x00003028
+
+/* CE ring R0 address */
+#define HAL_CE_DST_RING_BASE_LSB		0x00000000
+#define HAL_CE_DST_STATUS_RING_BASE_LSB		0x00000058
+#define HAL_CE_DST_RING_CTRL			0x000000b0
+
+/* CE ring R2 address */
+#define HAL_CE_DST_RING_HP			0x00000400
+#define HAL_CE_DST_STATUS_RING_HP		0x00000408
+
+/* REO status address */
+#define HAL_REO_STATUS_RING_BASE_LSB		0x00000504
+#define HAL_REO_STATUS_HP			0x00003070
+
+/* WBM Idle R0 address */
+#define HAL_WBM_IDLE_LINK_RING_BASE_LSB		0x00000860
+#define HAL_WBM_IDLE_LINK_RING_MISC_ADDR	0x00000870
+#define HAL_WBM_R0_IDLE_LIST_CONTROL_ADDR	0x00000048
+#define HAL_WBM_R0_IDLE_LIST_SIZE_ADDR		0x0000004c
+#define HAL_WBM_SCATTERED_RING_BASE_LSB		0x00000058
+#define HAL_WBM_SCATTERED_RING_BASE_MSB		0x0000005c
+#define HAL_WBM_SCATTERED_DESC_PTR_HEAD_INFO_IX0 0x00000068
+#define HAL_WBM_SCATTERED_DESC_PTR_HEAD_INFO_IX1 0x0000006c
+#define HAL_WBM_SCATTERED_DESC_PTR_TAIL_INFO_IX0 0x00000078
+#define HAL_WBM_SCATTERED_DESC_PTR_TAIL_INFO_IX1 0x0000007c
+#define HAL_WBM_SCATTERED_DESC_PTR_HP_ADDR	 0x00000084
+
+/* WBM Idle R2 address */
+#define HAL_WBM_IDLE_LINK_RING_HP		0x000030b0
+
+/* SW2WBM R0 release address */
+#define HAL_WBM_RELEASE_RING_BASE_LSB		0x000001d8
+
+/* SW2WBM R2 release address */
+#define HAL_WBM_RELEASE_RING_HP			0x00003018
+
+/* WBM2SW R0 release address */
+#define HAL_WBM0_RELEASE_RING_BASE_LSB		0x00000910
+#define HAL_WBM1_RELEASE_RING_BASE_LSB		0x00000968
+
+/* WBM2SW R2 release address */
+#define HAL_WBM0_RELEASE_RING_HP		0x000030c0
+#define HAL_WBM1_RELEASE_RING_HP		0x000030c8
+
+/* TCL ring feild mask and offset */
+#define HAL_TCL1_RING_BASE_MSB_RING_SIZE		GENMASK(27, 8)
+#define HAL_TCL1_RING_BASE_MSB_RING_BASE_ADDR_MSB	GENMASK(7, 0)
+#define HAL_TCL1_RING_ID_ENTRY_SIZE			GENMASK(7, 0)
+#define HAL_TCL1_RING_MISC_MSI_LOOPCNT_DISABLE		BIT(1)
+#define HAL_TCL1_RING_MISC_MSI_SWAP			BIT(3)
+#define HAL_TCL1_RING_MISC_HOST_FW_SWAP			BIT(4)
+#define HAL_TCL1_RING_MISC_DATA_TLV_SWAP		BIT(5)
+#define HAL_TCL1_RING_MISC_SRNG_ENABLE			BIT(6)
+#define HAL_TCL1_RING_CONSR_INT_SETUP_IX0_INTR_TMR_THOLD   GENMASK(31, 16)
+#define HAL_TCL1_RING_CONSR_INT_SETUP_IX0_BATCH_COUNTER_THOLD GENMASK(14, 0)
+#define HAL_TCL1_RING_CONSR_INT_SETUP_IX1_LOW_THOLD	GENMASK(15, 0)
+#define HAL_TCL1_RING_MSI1_BASE_MSB_MSI1_ENABLE		BIT(8)
+#define HAL_TCL1_RING_MSI1_BASE_MSB_ADDR		GENMASK(7, 0)
+#define HAL_TCL1_RING_CMN_CTRL_DSCP_TID_MAP_PROG_EN	BIT(17)
+#define HAL_TCL1_RING_FIELD_DSCP_TID_MAP		GENMASK(31, 0)
+#define HAL_TCL1_RING_FIELD_DSCP_TID_MAP0		GENMASK(2, 0)
+#define HAL_TCL1_RING_FIELD_DSCP_TID_MAP1		GENMASK(5, 3)
+#define HAL_TCL1_RING_FIELD_DSCP_TID_MAP2		GENMASK(8, 6)
+#define HAL_TCL1_RING_FIELD_DSCP_TID_MAP3		GENMASK(11, 9)
+#define HAL_TCL1_RING_FIELD_DSCP_TID_MAP4		GENMASK(14, 12)
+#define HAL_TCL1_RING_FIELD_DSCP_TID_MAP5		GENMASK(17, 15)
+#define HAL_TCL1_RING_FIELD_DSCP_TID_MAP6		GENMASK(20, 18)
+#define HAL_TCL1_RING_FIELD_DSCP_TID_MAP7		GENMASK(23, 21)
+
+/* REO ring feild mask and offset */
+#define HAL_REO1_RING_BASE_MSB_RING_SIZE		GENMASK(27, 8)
+#define HAL_REO1_RING_BASE_MSB_RING_BASE_ADDR_MSB	GENMASK(7, 0)
+#define HAL_REO1_RING_ID_RING_ID			GENMASK(15, 8)
+#define HAL_REO1_RING_ID_ENTRY_SIZE			GENMASK(7, 0)
+#define HAL_REO1_RING_MISC_MSI_SWAP			BIT(3)
+#define HAL_REO1_RING_MISC_HOST_FW_SWAP			BIT(4)
+#define HAL_REO1_RING_MISC_DATA_TLV_SWAP		BIT(5)
+#define HAL_REO1_RING_MISC_SRNG_ENABLE			BIT(6)
+#define HAL_REO1_RING_PRDR_INT_SETUP_INTR_TMR_THOLD	GENMASK(31, 16)
+#define HAL_REO1_RING_PRDR_INT_SETUP_BATCH_COUNTER_THOLD GENMASK(14, 0)
+#define HAL_REO1_RING_MSI1_BASE_MSB_MSI1_ENABLE		BIT(8)
+#define HAL_REO1_RING_MSI1_BASE_MSB_ADDR		GENMASK(7, 0)
+#define HAL_REO1_GEN_ENABLE_FRAG_DST_RING		GENMASK(25, 23)
+#define HAL_REO1_GEN_ENABLE_AGING_LIST_ENABLE		BIT(2)
+#define HAL_REO1_GEN_ENABLE_AGING_FLUSH_ENABLE		BIT(3)
+
+/* CE ring bit field mask and shift */
+#define HAL_CE_DST_R0_DEST_CTRL_MAX_LEN			GENMASK(15, 0)
+
+#define HAL_ADDR_LSB_REG_MASK				0xffffffff
+
+#define HAL_ADDR_MSB_REG_SHIFT				32
+
+/* WBM ring bit field mask and shift */
+#define HAL_WBM_LINK_DESC_IDLE_LIST_MODE		BIT(1)
+#define HAL_WBM_SCATTER_BUFFER_SIZE			GENMASK(10, 2)
+#define HAL_WBM_SCATTER_RING_SIZE_OF_IDLE_LINK_DESC_LIST GENMASK(31, 16)
+#define HAL_WBM_SCATTERED_DESC_MSB_BASE_ADDR_39_32	GENMASK(7, 0)
+#define HAL_WBM_SCATTERED_DESC_MSB_BASE_ADDR_MATCH_TAG	GENMASK(31, 8)
+
+#define HAL_WBM_SCATTERED_DESC_HEAD_P_OFFSET_IX1	GENMASK(20, 8)
+#define HAL_WBM_SCATTERED_DESC_TAIL_P_OFFSET_IX1	GENMASK(20, 8)
+
+#define BASE_ADDR_MATCH_TAG_VAL 0x5
+
+#define HAL_REO_REO2SW1_RING_BASE_MSB_RING_SIZE		0x000fffff
+#define HAL_REO_REO2TCL_RING_BASE_MSB_RING_SIZE		0x000fffff
+#define HAL_REO_SW2REO_RING_BASE_MSB_RING_SIZE		0x0000ffff
+#define HAL_REO_CMD_RING_BASE_MSB_RING_SIZE		0x0000ffff
+#define HAL_REO_STATUS_RING_BASE_MSB_RING_SIZE		0x0000ffff
+#define HAL_SW2TCL1_RING_BASE_MSB_RING_SIZE		0x000fffff
+#define HAL_SW2TCL1_CMD_RING_BASE_MSB_RING_SIZE		0x000fffff
+#define HAL_TCL_STATUS_RING_BASE_MSB_RING_SIZE		0x0000ffff
+#define HAL_CE_SRC_RING_BASE_MSB_RING_SIZE		0x0000ffff
+#define HAL_CE_DST_RING_BASE_MSB_RING_SIZE		0x0000ffff
+#define HAL_CE_DST_STATUS_RING_BASE_MSB_RING_SIZE	0x0000ffff
+#define HAL_WBM_IDLE_LINK_RING_BASE_MSB_RING_SIZE	0x0000ffff
+#define HAL_SW2WBM_RELEASE_RING_BASE_MSB_RING_SIZE	0x0000ffff
+#define HAL_WBM2SW_RELEASE_RING_BASE_MSB_RING_SIZE	0x000fffff
+#define HAL_RXDMA_RING_MAX_SIZE				0x0000ffff
+
+#define HAL_RX_DESC_SIZE (sizeof(struct hal_rx_desc))
+
+/* Add any other errors here and return them in
+ * ath11k_hal_rx_desc_get_err().
+ */
+
+enum hal_srng_ring_id {
+	HAL_SRNG_RING_ID_REO2SW1 = 0,
+	HAL_SRNG_RING_ID_REO2SW2,
+	HAL_SRNG_RING_ID_REO2SW3,
+	HAL_SRNG_RING_ID_REO2SW4,
+	HAL_SRNG_RING_ID_REO2TCL,
+	HAL_SRNG_RING_ID_SW2REO,
+
+	HAL_SRNG_RING_ID_REO_CMD = 8,
+	HAL_SRNG_RING_ID_REO_STATUS,
+
+	HAL_SRNG_RING_ID_SW2TCL1 = 16,
+	HAL_SRNG_RING_ID_SW2TCL2,
+	HAL_SRNG_RING_ID_SW2TCL3,
+	HAL_SRNG_RING_ID_SW2TCL4,
+
+	HAL_SRNG_RING_ID_SW2TCL_CMD = 24,
+	HAL_SRNG_RING_ID_TCL_STATUS,
+
+	HAL_SRNG_RING_ID_CE0_SRC = 32,
+	HAL_SRNG_RING_ID_CE1_SRC,
+	HAL_SRNG_RING_ID_CE2_SRC,
+	HAL_SRNG_RING_ID_CE3_SRC,
+	HAL_SRNG_RING_ID_CE4_SRC,
+	HAL_SRNG_RING_ID_CE5_SRC,
+	HAL_SRNG_RING_ID_CE6_SRC,
+	HAL_SRNG_RING_ID_CE7_SRC,
+	HAL_SRNG_RING_ID_CE8_SRC,
+	HAL_SRNG_RING_ID_CE9_SRC,
+	HAL_SRNG_RING_ID_CE10_SRC,
+	HAL_SRNG_RING_ID_CE11_SRC,
+
+	HAL_SRNG_RING_ID_CE0_DST = 56,
+	HAL_SRNG_RING_ID_CE1_DST,
+	HAL_SRNG_RING_ID_CE2_DST,
+	HAL_SRNG_RING_ID_CE3_DST,
+	HAL_SRNG_RING_ID_CE4_DST,
+	HAL_SRNG_RING_ID_CE5_DST,
+	HAL_SRNG_RING_ID_CE6_DST,
+	HAL_SRNG_RING_ID_CE7_DST,
+	HAL_SRNG_RING_ID_CE8_DST,
+	HAL_SRNG_RING_ID_CE9_DST,
+	HAL_SRNG_RING_ID_CE10_DST,
+	HAL_SRNG_RING_ID_CE11_DST,
+
+	HAL_SRNG_RING_ID_CE0_DST_STATUS = 80,
+	HAL_SRNG_RING_ID_CE1_DST_STATUS,
+	HAL_SRNG_RING_ID_CE2_DST_STATUS,
+	HAL_SRNG_RING_ID_CE3_DST_STATUS,
+	HAL_SRNG_RING_ID_CE4_DST_STATUS,
+	HAL_SRNG_RING_ID_CE5_DST_STATUS,
+	HAL_SRNG_RING_ID_CE6_DST_STATUS,
+	HAL_SRNG_RING_ID_CE7_DST_STATUS,
+	HAL_SRNG_RING_ID_CE8_DST_STATUS,
+	HAL_SRNG_RING_ID_CE9_DST_STATUS,
+	HAL_SRNG_RING_ID_CE10_DST_STATUS,
+	HAL_SRNG_RING_ID_CE11_DST_STATUS,
+
+	HAL_SRNG_RING_ID_WBM_IDLE_LINK = 104,
+	HAL_SRNG_RING_ID_WBM_SW_RELEASE,
+	HAL_SRNG_RING_ID_WBM2SW0_RELEASE,
+	HAL_SRNG_RING_ID_WBM2SW1_RELEASE,
+	HAL_SRNG_RING_ID_WBM2SW2_RELEASE,
+	HAL_SRNG_RING_ID_WBM2SW3_RELEASE,
+
+	HAL_SRNG_RING_ID_UMAC_ID_END = 127,
+	HAL_SRNG_RING_ID_LMAC1_ID_START,
+
+	HAL_SRNG_RING_ID_WMAC1_SW2RXDMA0_BUF = HAL_SRNG_RING_ID_LMAC1_ID_START,
+	HAL_SRNG_RING_ID_WMAC1_SW2RXDMA1_BUF,
+	HAL_SRNG_RING_ID_WMAC1_SW2RXDMA2_BUF,
+	HAL_SRNG_RING_ID_WMAC1_SW2RXDMA0_STATBUF,
+	HAL_SRNG_RING_ID_WMAC1_SW2RXDMA1_STATBUF,
+	HAL_SRNG_RING_ID_WMAC1_RXDMA2SW0,
+	HAL_SRNG_RING_ID_WMAC1_RXDMA2SW1,
+	HAL_SRNG_RING_ID_WMAC1_SW2RXDMA1_DESC,
+	HAL_SRNG_RING_ID_RXDMA_DIR_BUF,
+
+	HAL_SRNG_RING_ID_LMAC1_ID_END = 143
+};
+
+/* SRNG registers are split into two groups R0 and R2 */
+#define HAL_SRNG_REG_GRP_R0	0
+#define HAL_SRNG_REG_GRP_R2	1
+#define HAL_SRNG_NUM_REG_GRP    2
+
+#define HAL_SRNG_NUM_LMACS      3
+#define HAL_SRNG_REO_EXCEPTION  HAL_SRNG_RING_ID_REO2SW1
+#define HAL_SRNG_RINGS_PER_LMAC (HAL_SRNG_RING_ID_LMAC1_ID_END - \
+				 HAL_SRNG_RING_ID_LMAC1_ID_START)
+#define HAL_SRNG_NUM_LMAC_RINGS (HAL_SRNG_NUM_LMACS * HAL_SRNG_RINGS_PER_LMAC)
+#define HAL_SRNG_RING_ID_MAX    (HAL_SRNG_RING_ID_UMAC_ID_END + \
+				 HAL_SRNG_NUM_LMAC_RINGS)
+
+enum hal_ring_type {
+	HAL_REO_DST,
+	HAL_REO_EXCEPTION,
+	HAL_REO_REINJECT,
+	HAL_REO_CMD,
+	HAL_REO_STATUS,
+	HAL_TCL_DATA,
+	HAL_TCL_CMD,
+	HAL_TCL_STATUS,
+	HAL_CE_SRC,
+	HAL_CE_DST,
+	HAL_CE_DST_STATUS,
+	HAL_WBM_IDLE_LINK,
+	HAL_SW2WBM_RELEASE,
+	HAL_WBM2SW_RELEASE,
+	HAL_RXDMA_BUF,
+	HAL_RXDMA_DST,
+	HAL_RXDMA_MONITOR_BUF,
+	HAL_RXDMA_MONITOR_STATUS,
+	HAL_RXDMA_MONITOR_DST,
+	HAL_RXDMA_MONITOR_DESC,
+	HAL_RXDMA_DIR_BUF,
+	HAL_MAX_RING_TYPES,
+};
+
+#define HAL_RX_MAX_BA_WINDOW	256
+
+#define HAL_DEFAULT_REO_TIMEOUT_USEC		(40 * 1000)
+
+/**
+ * enum hal_reo_cmd_type: Enum for REO command type
+ * @CMD_GET_QUEUE_STATS: Get REO queue status/stats
+ * @CMD_FLUSH_QUEUE: Flush all frames in REO queue
+ * @CMD_FLUSH_CACHE: Flush descriptor entries in the cache
+ * @CMD_UNBLOCK_CACHE: Unblock a descriptor's address that was blocked
+ *      earlier with a 'REO_FLUSH_CACHE' command
+ * @CMD_FLUSH_TIMEOUT_LIST: Flush buffers/descriptors from timeout list
+ * @CMD_UPDATE_RX_REO_QUEUE: Update REO queue settings
+ */
+enum hal_reo_cmd_type {
+	HAL_REO_CMD_GET_QUEUE_STATS     = 0,
+	HAL_REO_CMD_FLUSH_QUEUE         = 1,
+	HAL_REO_CMD_FLUSH_CACHE         = 2,
+	HAL_REO_CMD_UNBLOCK_CACHE       = 3,
+	HAL_REO_CMD_FLUSH_TIMEOUT_LIST  = 4,
+	HAL_REO_CMD_UPDATE_RX_QUEUE     = 5,
+};
+
+/**
+ * enum hal_reo_cmd_status: Enum for execution status of REO command
+ * @HAL_REO_CMD_SUCCESS: Command has successfully executed
+ * @HAL_REO_CMD_BLOCKED: Command could not be executed as the queue
+ *			 or cache was blocked
+ * @HAL_REO_CMD_FAILED: Command execution failed, could be due to
+ *			invalid queue desc
+ * @HAL_REO_CMD_RESOURCE_BLOCKED:
+ * @HAL_REO_CMD_DRAIN:
+ */
+enum hal_reo_cmd_status {
+	HAL_REO_CMD_SUCCESS		= 0,
+	HAL_REO_CMD_BLOCKED		= 1,
+	HAL_REO_CMD_FAILED		= 2,
+	HAL_REO_CMD_RESOURCE_BLOCKED	= 3,
+	HAL_REO_CMD_DRAIN		= 0xff,
+};
+
+struct hal_wbm_idle_scatter_list {
+	dma_addr_t paddr;
+	struct hal_wbm_link_desc *vaddr;
+};
+
+struct hal_srng_params {
+	dma_addr_t ring_base_paddr;
+	u32 *ring_base_vaddr;
+	int num_entries;
+	u32 intr_batch_cntr_thres_entries;
+	u32 intr_timer_thres_us;
+	u32 flags;
+	u32 max_buffer_len;
+	u32 low_threshold;
+
+	/* Add more params as needed */
+};
+
+enum hal_srng_dir {
+	HAL_SRNG_DIR_SRC,
+	HAL_SRNG_DIR_DST
+};
+
+/* srng flags */
+#define HAL_SRNG_FLAGS_MSI_SWAP			0x00000008
+#define HAL_SRNG_FLAGS_RING_PTR_SWAP		0x00000010
+#define HAL_SRNG_FLAGS_DATA_TLV_SWAP		0x00000020
+#define HAL_SRNG_FLAGS_LOW_THRESH_INTR_EN	0x00010000
+#define HAL_SRNG_FLAGS_MSI_INTR			0x00020000
+#define HAL_SRNG_FLAGS_LMAC_RING		0x80000000
+
+#define HAL_SRNG_TLV_HDR_TAG		GENMASK(9, 1)
+#define HAL_SRNG_TLV_HDR_LEN		GENMASK(25, 10)
+
+/* Common SRNG ring structure for source and destination rings */
+struct hal_srng {
+	/* Unique SRNG ring ID */
+	u8 ring_id;
+
+	/* Ring initialization done */
+	u8 initialized;
+
+	/* Interrupt/MSI value assigned to this ring */
+	int irq;
+
+	/* Physical base address of the ring */
+	dma_addr_t ring_base_paddr;
+
+	/* Virtual base address of the ring */
+	u32 *ring_base_vaddr;
+
+	/* Number of entries in ring */
+	u32 num_entries;
+
+	/* Ring size */
+	u32 ring_size;
+
+	/* Ring size mask */
+	u32 ring_size_mask;
+
+	/* Size of ring entry */
+	u32 entry_size;
+
+	/* Interrupt timer threshold - in micro seconds */
+	u32 intr_timer_thres_us;
+
+	/* Interrupt batch counter threshold - in number of ring entries */
+	u32 intr_batch_cntr_thres_entries;
+
+	/* MSI Address */
+	dma_addr_t msi_addr;
+
+	/* MSI data */
+	u32 msi_data;
+
+	/* Misc flags */
+	u32 flags;
+
+	/* Lock for serializing ring index updates */
+	spinlock_t lock;
+
+	/* Start offset of SRNG register groups for this ring
+	 * TBD: See if this is required - register address can be derived
+	 * from ring ID
+	 */
+	u32 hwreg_base[HAL_SRNG_NUM_REG_GRP];
+
+	/* Source or Destination ring */
+	enum hal_srng_dir ring_dir;
+
+	union {
+		struct {
+			/* SW tail pointer */
+			u32 tp;
+
+			/* Shadow head pointer location to be updated by HW */
+			volatile u32 *hp_addr;
+
+			/* Cached head pointer */
+			u32 cached_hp;
+
+			/* Tail pointer location to be updated by SW - This
+			 * will be a register address and need not be
+			 * accessed through SW structure
+			 */
+			u32 *tp_addr;
+
+			/* Current SW loop cnt */
+			u32 loop_cnt;
+
+			/* max transfer size */
+			u16 max_buffer_length;
+		} dst_ring;
+
+		struct {
+			/* SW head pointer */
+			u32 hp;
+
+			/* SW reap head pointer */
+			u32 reap_hp;
+
+			/* Shadow tail pointer location to be updated by HW */
+			u32 *tp_addr;
+
+			/* Cached tail pointer */
+			u32 cached_tp;
+
+			/* Head pointer location to be updated by SW - This
+			 * will be a register address and need not be accessed
+			 * through SW structure
+			 */
+			u32 *hp_addr;
+
+			/* Low threshold - in number of ring entries */
+			u32 low_threshold;
+		} src_ring;
+	} u;
+};
+
+/* Interrupt mitigation - Batch threshold in terms of numer of frames */
+#define HAL_SRNG_INT_BATCH_THRESHOLD_TX 256
+#define HAL_SRNG_INT_BATCH_THRESHOLD_RX 128
+#define HAL_SRNG_INT_BATCH_THRESHOLD_OTHER 1
+
+/* Interrupt mitigation - timer threshold in us */
+#define HAL_SRNG_INT_TIMER_THRESHOLD_TX 1000
+#define HAL_SRNG_INT_TIMER_THRESHOLD_RX 500
+#define HAL_SRNG_INT_TIMER_THRESHOLD_OTHER 1000
+
+/* HW SRNG configuration table */
+struct hal_srng_config {
+	int start_ring_id;
+	u16 max_rings;
+	u16 entry_size;
+	u32 reg_start[HAL_SRNG_NUM_REG_GRP];
+	u16 reg_size[HAL_SRNG_NUM_REG_GRP];
+	u8 lmac_ring;
+	enum hal_srng_dir ring_dir;
+	u32 max_size;
+};
+
+/**
+ * enum hal_rx_buf_return_buf_manager
+ *
+ * @HAL_RX_BUF_RBM_WBM_IDLE_BUF_LIST: Buffer returned to WBM idle buffer list
+ * @HAL_RX_BUF_RBM_WBM_IDLE_DESC_LIST: Descriptor returned to WBM idle
+ *	descriptor list.
+ * @HAL_RX_BUF_RBM_FW_BM: Buffer returned to FW
+ * @HAL_RX_BUF_RBM_SW0_BM: For Tx completion -- returned to host
+ * @HAL_RX_BUF_RBM_SW1_BM: For Tx completion -- returned to host
+ * @HAL_RX_BUF_RBM_SW2_BM: For Tx completion -- returned to host
+ * @HAL_RX_BUF_RBM_SW3_BM: For Rx release -- returned to host
+ */
+
+enum hal_rx_buf_return_buf_manager {
+	HAL_RX_BUF_RBM_WBM_IDLE_BUF_LIST,
+	HAL_RX_BUF_RBM_WBM_IDLE_DESC_LIST,
+	HAL_RX_BUF_RBM_FW_BM,
+	HAL_RX_BUF_RBM_SW0_BM,
+	HAL_RX_BUF_RBM_SW1_BM,
+	HAL_RX_BUF_RBM_SW2_BM,
+	HAL_RX_BUF_RBM_SW3_BM,
+};
+
+#define HAL_SRNG_DESC_LOOP_CNT		0xf0000000
+
+#define HAL_REO_CMD_FLG_NEED_STATUS		BIT(0)
+#define HAL_REO_CMD_FLG_STATS_CLEAR		BIT(1)
+#define HAL_REO_CMD_FLG_FLUSH_BLOCK_LATER	BIT(2)
+#define HAL_REO_CMD_FLG_FLUSH_RELEASE_BLOCKING	BIT(3)
+#define HAL_REO_CMD_FLG_FLUSH_NO_INVAL		BIT(4)
+#define HAL_REO_CMD_FLG_FLUSH_FWD_ALL_MPDUS	BIT(5)
+#define HAL_REO_CMD_FLG_FLUSH_ALL		BIT(6)
+#define HAL_REO_CMD_FLG_UNBLK_RESOURCE		BIT(7)
+#define HAL_REO_CMD_FLG_UNBLK_CACHE		BIT(8)
+
+/* Should be matching with HAL_REO_UPD_RX_QUEUE_INFO0_UPD_* feilds */
+#define HAL_REO_CMD_UPD0_RX_QUEUE_NUM		BIT(8)
+#define HAL_REO_CMD_UPD0_VLD			BIT(9)
+#define HAL_REO_CMD_UPD0_ALDC			BIT(10)
+#define HAL_REO_CMD_UPD0_DIS_DUP_DETECTION	BIT(11)
+#define HAL_REO_CMD_UPD0_SOFT_REORDER_EN	BIT(12)
+#define HAL_REO_CMD_UPD0_AC			BIT(13)
+#define HAL_REO_CMD_UPD0_BAR			BIT(14)
+#define HAL_REO_CMD_UPD0_RETRY			BIT(15)
+#define HAL_REO_CMD_UPD0_CHECK_2K_MODE		BIT(16)
+#define HAL_REO_CMD_UPD0_OOR_MODE		BIT(17)
+#define HAL_REO_CMD_UPD0_BA_WINDOW_SIZE		BIT(18)
+#define HAL_REO_CMD_UPD0_PN_CHECK		BIT(19)
+#define HAL_REO_CMD_UPD0_EVEN_PN		BIT(20)
+#define HAL_REO_CMD_UPD0_UNEVEN_PN		BIT(21)
+#define HAL_REO_CMD_UPD0_PN_HANDLE_ENABLE	BIT(22)
+#define HAL_REO_CMD_UPD0_PN_SIZE		BIT(23)
+#define HAL_REO_CMD_UPD0_IGNORE_AMPDU_FLG	BIT(24)
+#define HAL_REO_CMD_UPD0_SVLD			BIT(25)
+#define HAL_REO_CMD_UPD0_SSN			BIT(26)
+#define HAL_REO_CMD_UPD0_SEQ_2K_ERR		BIT(27)
+#define HAL_REO_CMD_UPD0_PN_ERR			BIT(28)
+#define HAL_REO_CMD_UPD0_PN_VALID		BIT(29)
+#define HAL_REO_CMD_UPD0_PN			BIT(30)
+
+/* Should be matching with HAL_REO_UPD_RX_QUEUE_INFO1_* feilds */
+#define HAL_REO_CMD_UPD1_VLD			BIT(16)
+#define HAL_REO_CMD_UPD1_ALDC			GENMASK(18, 17)
+#define HAL_REO_CMD_UPD1_DIS_DUP_DETECTION	BIT(19)
+#define HAL_REO_CMD_UPD1_SOFT_REORDER_EN	BIT(20)
+#define HAL_REO_CMD_UPD1_AC			GENMASK(22, 21)
+#define HAL_REO_CMD_UPD1_BAR			BIT(23)
+#define HAL_REO_CMD_UPD1_RETRY			BIT(24)
+#define HAL_REO_CMD_UPD1_CHECK_2K_MODE		BIT(25)
+#define HAL_REO_CMD_UPD1_OOR_MODE		BIT(26)
+#define HAL_REO_CMD_UPD1_PN_CHECK		BIT(27)
+#define HAL_REO_CMD_UPD1_EVEN_PN		BIT(28)
+#define HAL_REO_CMD_UPD1_UNEVEN_PN		BIT(29)
+#define HAL_REO_CMD_UPD1_PN_HANDLE_ENABLE	BIT(30)
+#define HAL_REO_CMD_UPD1_IGNORE_AMPDU_FLG	BIT(31)
+
+/* Should be matching with HAL_REO_UPD_RX_QUEUE_INFO2_* feilds */
+#define HAL_REO_CMD_UPD2_SVLD			BIT(10)
+#define HAL_REO_CMD_UPD2_SSN			GENMASK(22, 11)
+#define HAL_REO_CMD_UPD2_SEQ_2K_ERR		BIT(23)
+#define HAL_REO_CMD_UPD2_PN_ERR			BIT(24)
+
+#define HAL_REO_DEST_RING_CTRL_HASH_RING_MAP	GENMASK(31, 8)
+
+struct ath11k_hal_reo_cmd {
+	u32 addr_lo;
+	u32 flag;
+	u32 upd0;
+	u32 upd1;
+	u32 upd2;
+	u32 pn[4];
+	u16 rx_queue_num;
+	u16 min_rel;
+	u16 min_fwd;
+	u8 addr_hi;
+	u8 ac_list;
+	u8 blocking_idx;
+	u16 ba_window_size;
+	u8 pn_size;
+};
+
+enum hal_pn_type {
+	HAL_PN_TYPE_NONE,
+	HAL_PN_TYPE_WPA,
+	HAL_PN_TYPE_WAPI_EVEN,
+	HAL_PN_TYPE_WAPI_UNEVEN,
+};
+
+enum hal_ce_desc {
+	HAL_CE_DESC_SRC,
+	HAL_CE_DESC_DST,
+	HAL_CE_DESC_DST_STATUS,
+};
+
+struct hal_reo_status_header {
+	u16 cmd_num;
+	enum hal_reo_cmd_status cmd_status;
+	u16 cmd_exe_time;
+	u32 timestamp;
+};
+
+struct hal_reo_status_queue_stats {
+	u16 ssn;
+	u16 curr_idx;
+	u32 pn[4];
+	u32 last_rx_queue_ts;
+	u32 last_rx_dequeue_ts;
+	u32 rx_bitmap[8]; /* Bitmap from 0-255 */
+	u32 curr_mpdu_cnt;
+	u32 curr_msdu_cnt;
+	u16 fwd_due_to_bar_cnt;
+	u16 dup_cnt;
+	u32 frames_in_order_cnt;
+	u32 num_mpdu_processed_cnt;
+	u32 num_msdu_processed_cnt;
+	u32 total_num_processed_byte_cnt;
+	u32 late_rx_mpdu_cnt;
+	u32 reorder_hole_cnt;
+	u8 timeout_cnt;
+	u8 bar_rx_cnt;
+	u8 num_window_2k_jump_cnt;
+};
+
+struct hal_reo_status_flush_queue {
+	bool err_detected;
+};
+
+enum hal_reo_status_flush_cache_err_code {
+	HAL_REO_STATUS_FLUSH_CACHE_ERR_CODE_SUCCESS,
+	HAL_REO_STATUS_FLUSH_CACHE_ERR_CODE_IN_USE,
+	HAL_REO_STATUS_FLUSH_CACHE_ERR_CODE_NOT_FOUND,
+};
+
+struct hal_reo_status_flush_cache {
+	bool err_detected;
+	enum hal_reo_status_flush_cache_err_code err_code;
+	bool cache_controller_flush_status_hit;
+	u8 cache_controller_flush_status_desc_type;
+	u8 cache_controller_flush_status_client_id;
+	u8 cache_controller_flush_status_err;
+	u8 cache_controller_flush_status_cnt;
+};
+
+enum hal_reo_status_unblock_cache_type {
+	HAL_REO_STATUS_UNBLOCK_BLOCKING_RESOURCE,
+	HAL_REO_STATUS_UNBLOCK_ENTIRE_CACHE_USAGE,
+};
+
+struct hal_reo_status_unblock_cache {
+	bool err_detected;
+	enum hal_reo_status_unblock_cache_type unblock_type;
+};
+
+struct hal_reo_status_flush_timeout_list {
+	bool err_detected;
+	bool list_empty;
+	u16 release_desc_cnt;
+	u16 fwd_buf_cnt;
+};
+
+enum hal_reo_threshold_idx {
+	HAL_REO_THRESHOLD_IDX_DESC_COUNTER0,
+	HAL_REO_THRESHOLD_IDX_DESC_COUNTER1,
+	HAL_REO_THRESHOLD_IDX_DESC_COUNTER2,
+	HAL_REO_THRESHOLD_IDX_DESC_COUNTER_SUM,
+};
+
+struct hal_reo_status_desc_thresh_reached {
+	enum hal_reo_threshold_idx threshold_idx;
+	u32 link_desc_counter0;
+	u32 link_desc_counter1;
+	u32 link_desc_counter2;
+	u32 link_desc_counter_sum;
+};
+
+struct hal_reo_status {
+	struct hal_reo_status_header uniform_hdr;
+	u8 loop_cnt;
+	union {
+		struct hal_reo_status_queue_stats queue_stats;
+		struct hal_reo_status_flush_queue flush_queue;
+		struct hal_reo_status_flush_cache flush_cache;
+		struct hal_reo_status_unblock_cache unblock_cache;
+		struct hal_reo_status_flush_timeout_list timeout_list;
+		struct hal_reo_status_desc_thresh_reached desc_thresh_reached;
+	} u;
+};
+
+/**
+ * HAL context to be used to access SRNG APIs (currently used by data path
+ * and transport (CE) modules)
+ */
+struct ath11k_hal {
+	/* HAL internal state for all SRNG rings.
+	 */
+	struct hal_srng srng_list[HAL_SRNG_RING_ID_MAX];
+
+	/* SRNG configuration table */
+	const struct hal_srng_config *srng_config;
+
+	/* Remote pointer memory for HW/FW updates */
+	struct {
+		u32 *vaddr;
+		dma_addr_t paddr;
+	} rdp;
+
+	/* Shared memory for ring pointer updates from host to FW */
+	struct {
+		u32 *vaddr;
+		dma_addr_t paddr;
+	} wrp;
+
+	/* Available REO blocking resources bitmap */
+	u8 avail_blk_resource;
+
+	u8 current_blk_index;
+
+	/* shadow register configuration */
+	u32 shadow_reg_addr[SHADOW_NUM_REGISTERS];
+	int num_shadow_reg_configured;
+};
+
+u32 ath11k_hal_reo_qdesc_size(u32 ba_window_size, u8 tid);
+void ath11k_hal_reo_qdesc_setup(void *vaddr, int tid, u32 ba_window_size,
+				u32 start_seqtype);
+void ath11k_hal_reo_init_cmd_ring(struct ath11k_base *ab,
+				  struct hal_srng *srng);
+void ath11k_hal_reo_hw_setup(struct ath11k_base *ab);
+void ath11k_hal_setup_link_idle_list(struct ath11k_base *ab,
+				     struct hal_wbm_idle_scatter_list *sbuf,
+				     u32 nsbufs, u32 tot_link_desc,
+				     u32 end_offset);
+
+dma_addr_t ath11k_hal_srng_get_tp_addr(struct ath11k_base *ab,
+				       struct hal_srng *srng);
+dma_addr_t ath11k_hal_srng_get_hp_addr(struct ath11k_base *ab,
+				       struct hal_srng *srng);
+void ath11k_hal_set_link_desc_addr(struct hal_wbm_link_desc *desc, u32 cookie,
+				   dma_addr_t paddr);
+u32 ath11k_hal_ce_get_desc_size(enum hal_ce_desc type);
+void ath11k_hal_ce_src_set_desc(void *buf, dma_addr_t paddr, u32 len, u32 id,
+				u8 byte_swap_data);
+void ath11k_hal_ce_dst_set_desc(void *buf, dma_addr_t paddr);
+u32 ath11k_hal_ce_dst_status_get_length(void *buf);
+int ath11k_hal_srng_get_entrysize(u32 ring_type);
+int ath11k_hal_srng_get_max_entries(u32 ring_type);
+void ath11k_hal_srng_get_params(struct ath11k_base *ab, struct hal_srng *srng,
+				struct hal_srng_params *params);
+u32 *ath11k_hal_srng_dst_get_next_entry(struct ath11k_base *ab,
+					struct hal_srng *srng);
+u32 *ath11k_hal_srng_dst_peek(struct ath11k_base *ab, struct hal_srng *srng);
+int ath11k_hal_srng_dst_num_free(struct ath11k_base *ab, struct hal_srng *srng,
+				 bool sync_hw_ptr);
+u32 *ath11k_hal_srng_src_peek(struct ath11k_base *ab, struct hal_srng *srng);
+u32 *ath11k_hal_srng_src_get_next_reaped(struct ath11k_base *ab,
+					 struct hal_srng *srng);
+u32 *ath11k_hal_srng_src_reap_next(struct ath11k_base *ab,
+				   struct hal_srng *srng);
+u32 *ath11k_hal_srng_src_get_next_entry(struct ath11k_base *ab,
+					struct hal_srng *srng);
+int ath11k_hal_srng_src_num_free(struct ath11k_base *ab, struct hal_srng *srng,
+				 bool sync_hw_ptr);
+void ath11k_hal_srng_access_begin(struct ath11k_base *ab,
+				  struct hal_srng *srng);
+void ath11k_hal_srng_access_end(struct ath11k_base *ab, struct hal_srng *srng);
+int ath11k_hal_srng_setup(struct ath11k_base *ab, enum hal_ring_type type,
+			  int ring_num, int mac_id,
+			  struct hal_srng_params *params);
+int ath11k_hal_srng_init(struct ath11k_base *ath11k);
+void ath11k_hal_srng_deinit(struct ath11k_base *ath11k);
+
+#endif
diff --git a/drivers/net/wireless/ath/ath11k/hal_desc.h b/drivers/net/wireless/ath/ath11k/hal_desc.h
new file mode 100644
index 000000000000..35e191b77271
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/hal_desc.h
@@ -0,0 +1,2469 @@
+/* SPDX-License-Identifier: BSD-3-Clause-Clear */
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+#ifndef ATH11K_HAL_DESC_H
+#define ATH11K_HAL_DESC_H
+
+#define BUFFER_ADDR_INFO0_ADDR         GENMASK(31, 0)
+
+#define BUFFER_ADDR_INFO1_ADDR         GENMASK(7, 0)
+#define BUFFER_ADDR_INFO1_RET_BUF_MGR  GENMASK(10, 8)
+#define BUFFER_ADDR_INFO1_SW_COOKIE    GENMASK(31, 11)
+
+struct ath11k_buffer_addr {
+	u32 info0;
+	u32 info1;
+} __packed;
+
+/* ath11k_buffer_addr
+ *
+ * info0
+ *		Address (lower 32 bits) of the msdu buffer or msdu extension
+ *		descriptor or Link descriptor
+ *
+ * addr
+ *		Address (upper 8 bits) of the msdu buffer or msdu extension
+ *		descriptor or Link descriptor
+ *
+ * return_buffer_manager (RBM)
+ *		Consumer: WBM
+ *		Producer: SW/FW
+ *		Indicates to which buffer manager the buffer or MSDU_EXTENSION
+ *		descriptor or link descriptor that is being pointed to shall be
+ *		returned after the frame has been processed. It is used by WBM
+ *		for routing purposes.
+ *
+ *		Values are defined in enum %HAL_RX_BUF_RBM_
+ *
+ * sw_buffer_cookie
+ *		Cookie field exclusively used by SW. HW ignores the contents,
+ *		accept that it passes the programmed value on to other
+ *		descriptors together with the physical address.
+ *
+ *		Field can be used by SW to for example associate the buffers
+ *		physical address with the virtual address.
+ */
+
+enum hal_tlv_tag {
+	HAL_MACTX_CBF_START                    =   0 /* 0x0 */,
+	HAL_PHYRX_DATA                         =   1 /* 0x1 */,
+	HAL_PHYRX_CBF_DATA_RESP                =   2 /* 0x2 */,
+	HAL_PHYRX_ABORT_REQUEST                =   3 /* 0x3 */,
+	HAL_PHYRX_USER_ABORT_NOTIFICATION      =   4 /* 0x4 */,
+	HAL_MACTX_DATA_RESP                    =   5 /* 0x5 */,
+	HAL_MACTX_CBF_DATA                     =   6 /* 0x6 */,
+	HAL_MACTX_CBF_DONE                     =   7 /* 0x7 */,
+	HAL_MACRX_CBF_READ_REQUEST             =   8 /* 0x8 */,
+	HAL_MACRX_CBF_DATA_REQUEST             =   9 /* 0x9 */,
+	HAL_MACRX_EXPECT_NDP_RECEPTION         =  10 /* 0xa */,
+	HAL_MACRX_FREEZE_CAPTURE_CHANNEL       =  11 /* 0xb */,
+	HAL_MACRX_NDP_TIMEOUT                  =  12 /* 0xc */,
+	HAL_MACRX_ABORT_ACK                    =  13 /* 0xd */,
+	HAL_MACRX_REQ_IMPLICIT_FB              =  14 /* 0xe */,
+	HAL_MACRX_CHAIN_MASK                   =  15 /* 0xf */,
+	HAL_MACRX_NAP_USER                     =  16 /* 0x10 */,
+	HAL_MACRX_ABORT_REQUEST                =  17 /* 0x11 */,
+	HAL_PHYTX_OTHER_TRANSMIT_INFO16        =  18 /* 0x12 */,
+	HAL_PHYTX_ABORT_ACK                    =  19 /* 0x13 */,
+	HAL_PHYTX_ABORT_REQUEST                =  20 /* 0x14 */,
+	HAL_PHYTX_PKT_END                      =  21 /* 0x15 */,
+	HAL_PHYTX_PPDU_HEADER_INFO_REQUEST     =  22 /* 0x16 */,
+	HAL_PHYTX_REQUEST_CTRL_INFO            =  23 /* 0x17 */,
+	HAL_PHYTX_DATA_REQUEST                 =  24 /* 0x18 */,
+	HAL_PHYTX_BF_CV_LOADING_DONE           =  25 /* 0x19 */,
+	HAL_PHYTX_NAP_ACK                      =  26 /* 0x1a */,
+	HAL_PHYTX_NAP_DONE                     =  27 /* 0x1b */,
+	HAL_PHYTX_OFF_ACK                      =  28 /* 0x1c */,
+	HAL_PHYTX_ON_ACK                       =  29 /* 0x1d */,
+	HAL_PHYTX_SYNTH_OFF_ACK                =  30 /* 0x1e */,
+	HAL_PHYTX_DEBUG16                      =  31 /* 0x1f */,
+	HAL_MACTX_ABORT_REQUEST                =  32 /* 0x20 */,
+	HAL_MACTX_ABORT_ACK                    =  33 /* 0x21 */,
+	HAL_MACTX_PKT_END                      =  34 /* 0x22 */,
+	HAL_MACTX_PRE_PHY_DESC                 =  35 /* 0x23 */,
+	HAL_MACTX_BF_PARAMS_COMMON             =  36 /* 0x24 */,
+	HAL_MACTX_BF_PARAMS_PER_USER           =  37 /* 0x25 */,
+	HAL_MACTX_PREFETCH_CV                  =  38 /* 0x26 */,
+	HAL_MACTX_USER_DESC_COMMON             =  39 /* 0x27 */,
+	HAL_MACTX_USER_DESC_PER_USER           =  40 /* 0x28 */,
+	HAL_EXAMPLE_USER_TLV_16                =  41 /* 0x29 */,
+	HAL_EXAMPLE_TLV_16                     =  42 /* 0x2a */,
+	HAL_MACTX_PHY_OFF                      =  43 /* 0x2b */,
+	HAL_MACTX_PHY_ON                       =  44 /* 0x2c */,
+	HAL_MACTX_SYNTH_OFF                    =  45 /* 0x2d */,
+	HAL_MACTX_EXPECT_CBF_COMMON            =  46 /* 0x2e */,
+	HAL_MACTX_EXPECT_CBF_PER_USER          =  47 /* 0x2f */,
+	HAL_MACTX_PHY_DESC                     =  48 /* 0x30 */,
+	HAL_MACTX_L_SIG_A                      =  49 /* 0x31 */,
+	HAL_MACTX_L_SIG_B                      =  50 /* 0x32 */,
+	HAL_MACTX_HT_SIG                       =  51 /* 0x33 */,
+	HAL_MACTX_VHT_SIG_A                    =  52 /* 0x34 */,
+	HAL_MACTX_VHT_SIG_B_SU20               =  53 /* 0x35 */,
+	HAL_MACTX_VHT_SIG_B_SU40               =  54 /* 0x36 */,
+	HAL_MACTX_VHT_SIG_B_SU80               =  55 /* 0x37 */,
+	HAL_MACTX_VHT_SIG_B_SU160              =  56 /* 0x38 */,
+	HAL_MACTX_VHT_SIG_B_MU20               =  57 /* 0x39 */,
+	HAL_MACTX_VHT_SIG_B_MU40               =  58 /* 0x3a */,
+	HAL_MACTX_VHT_SIG_B_MU80               =  59 /* 0x3b */,
+	HAL_MACTX_VHT_SIG_B_MU160              =  60 /* 0x3c */,
+	HAL_MACTX_SERVICE                      =  61 /* 0x3d */,
+	HAL_MACTX_HE_SIG_A_SU                  =  62 /* 0x3e */,
+	HAL_MACTX_HE_SIG_A_MU_DL               =  63 /* 0x3f */,
+	HAL_MACTX_HE_SIG_A_MU_UL               =  64 /* 0x40 */,
+	HAL_MACTX_HE_SIG_B1_MU                 =  65 /* 0x41 */,
+	HAL_MACTX_HE_SIG_B2_MU                 =  66 /* 0x42 */,
+	HAL_MACTX_HE_SIG_B2_OFDMA              =  67 /* 0x43 */,
+	HAL_MACTX_DELETE_CV                    =  68 /* 0x44 */,
+	HAL_MACTX_MU_UPLINK_COMMON             =  69 /* 0x45 */,
+	HAL_MACTX_MU_UPLINK_USER_SETUP         =  70 /* 0x46 */,
+	HAL_MACTX_OTHER_TRANSMIT_INFO          =  71 /* 0x47 */,
+	HAL_MACTX_PHY_NAP                      =  72 /* 0x48 */,
+	HAL_MACTX_DEBUG                        =  73 /* 0x49 */,
+	HAL_PHYRX_ABORT_ACK                    =  74 /* 0x4a */,
+	HAL_PHYRX_GENERATED_CBF_DETAILS        =  75 /* 0x4b */,
+	HAL_PHYRX_RSSI_LEGACY                  =  76 /* 0x4c */,
+	HAL_PHYRX_RSSI_HT                      =  77 /* 0x4d */,
+	HAL_PHYRX_USER_INFO                    =  78 /* 0x4e */,
+	HAL_PHYRX_PKT_END                      =  79 /* 0x4f */,
+	HAL_PHYRX_DEBUG                        =  80 /* 0x50 */,
+	HAL_PHYRX_CBF_TRANSFER_DONE            =  81 /* 0x51 */,
+	HAL_PHYRX_CBF_TRANSFER_ABORT           =  82 /* 0x52 */,
+	HAL_PHYRX_L_SIG_A                      =  83 /* 0x53 */,
+	HAL_PHYRX_L_SIG_B                      =  84 /* 0x54 */,
+	HAL_PHYRX_HT_SIG                       =  85 /* 0x55 */,
+	HAL_PHYRX_VHT_SIG_A                    =  86 /* 0x56 */,
+	HAL_PHYRX_VHT_SIG_B_SU20               =  87 /* 0x57 */,
+	HAL_PHYRX_VHT_SIG_B_SU40               =  88 /* 0x58 */,
+	HAL_PHYRX_VHT_SIG_B_SU80               =  89 /* 0x59 */,
+	HAL_PHYRX_VHT_SIG_B_SU160              =  90 /* 0x5a */,
+	HAL_PHYRX_VHT_SIG_B_MU20               =  91 /* 0x5b */,
+	HAL_PHYRX_VHT_SIG_B_MU40               =  92 /* 0x5c */,
+	HAL_PHYRX_VHT_SIG_B_MU80               =  93 /* 0x5d */,
+	HAL_PHYRX_VHT_SIG_B_MU160              =  94 /* 0x5e */,
+	HAL_PHYRX_HE_SIG_A_SU                  =  95 /* 0x5f */,
+	HAL_PHYRX_HE_SIG_A_MU_DL               =  96 /* 0x60 */,
+	HAL_PHYRX_HE_SIG_A_MU_UL               =  97 /* 0x61 */,
+	HAL_PHYRX_HE_SIG_B1_MU                 =  98 /* 0x62 */,
+	HAL_PHYRX_HE_SIG_B2_MU                 =  99 /* 0x63 */,
+	HAL_PHYRX_HE_SIG_B2_OFDMA              = 100 /* 0x64 */,
+	HAL_PHYRX_OTHER_RECEIVE_INFO           = 101 /* 0x65 */,
+	HAL_PHYRX_COMMON_USER_INFO             = 102 /* 0x66 */,
+	HAL_PHYRX_DATA_DONE                    = 103 /* 0x67 */,
+	HAL_RECEIVE_RSSI_INFO                  = 104 /* 0x68 */,
+	HAL_RECEIVE_USER_INFO                  = 105 /* 0x69 */,
+	HAL_MIMO_CONTROL_INFO                  = 106 /* 0x6a */,
+	HAL_RX_LOCATION_INFO                   = 107 /* 0x6b */,
+	HAL_COEX_TX_REQ                        = 108 /* 0x6c */,
+	HAL_DUMMY                              = 109 /* 0x6d */,
+	HAL_RX_TIMING_OFFSET_INFO              = 110 /* 0x6e */,
+	HAL_EXAMPLE_TLV_32_NAME                = 111 /* 0x6f */,
+	HAL_MPDU_LIMIT                         = 112 /* 0x70 */,
+	HAL_NA_LENGTH_END                      = 113 /* 0x71 */,
+	HAL_OLE_BUF_STATUS                     = 114 /* 0x72 */,
+	HAL_PCU_PPDU_SETUP_DONE                = 115 /* 0x73 */,
+	HAL_PCU_PPDU_SETUP_END                 = 116 /* 0x74 */,
+	HAL_PCU_PPDU_SETUP_INIT                = 117 /* 0x75 */,
+	HAL_PCU_PPDU_SETUP_START               = 118 /* 0x76 */,
+	HAL_PDG_FES_SETUP                      = 119 /* 0x77 */,
+	HAL_PDG_RESPONSE                       = 120 /* 0x78 */,
+	HAL_PDG_TX_REQ                         = 121 /* 0x79 */,
+	HAL_SCH_WAIT_INSTR                     = 122 /* 0x7a */,
+	HAL_SCHEDULER_TLV                      = 123 /* 0x7b */,
+	HAL_TQM_FLOW_EMPTY_STATUS              = 124 /* 0x7c */,
+	HAL_TQM_FLOW_NOT_EMPTY_STATUS          = 125 /* 0x7d */,
+	HAL_TQM_GEN_MPDU_LENGTH_LIST           = 126 /* 0x7e */,
+	HAL_TQM_GEN_MPDU_LENGTH_LIST_STATUS    = 127 /* 0x7f */,
+	HAL_TQM_GEN_MPDUS                      = 128 /* 0x80 */,
+	HAL_TQM_GEN_MPDUS_STATUS               = 129 /* 0x81 */,
+	HAL_TQM_REMOVE_MPDU                    = 130 /* 0x82 */,
+	HAL_TQM_REMOVE_MPDU_STATUS             = 131 /* 0x83 */,
+	HAL_TQM_REMOVE_MSDU                    = 132 /* 0x84 */,
+	HAL_TQM_REMOVE_MSDU_STATUS             = 133 /* 0x85 */,
+	HAL_TQM_UPDATE_TX_MPDU_COUNT           = 134 /* 0x86 */,
+	HAL_TQM_WRITE_CMD                      = 135 /* 0x87 */,
+	HAL_OFDMA_TRIGGER_DETAILS              = 136 /* 0x88 */,
+	HAL_TX_DATA                            = 137 /* 0x89 */,
+	HAL_TX_FES_SETUP                       = 138 /* 0x8a */,
+	HAL_RX_PACKET                          = 139 /* 0x8b */,
+	HAL_EXPECTED_RESPONSE                  = 140 /* 0x8c */,
+	HAL_TX_MPDU_END                        = 141 /* 0x8d */,
+	HAL_TX_MPDU_START                      = 142 /* 0x8e */,
+	HAL_TX_MSDU_END                        = 143 /* 0x8f */,
+	HAL_TX_MSDU_START                      = 144 /* 0x90 */,
+	HAL_TX_SW_MODE_SETUP                   = 145 /* 0x91 */,
+	HAL_TXPCU_BUFFER_STATUS                = 146 /* 0x92 */,
+	HAL_TXPCU_USER_BUFFER_STATUS           = 147 /* 0x93 */,
+	HAL_DATA_TO_TIME_CONFIG                = 148 /* 0x94 */,
+	HAL_EXAMPLE_USER_TLV_32                = 149 /* 0x95 */,
+	HAL_MPDU_INFO                          = 150 /* 0x96 */,
+	HAL_PDG_USER_SETUP                     = 151 /* 0x97 */,
+	HAL_TX_11AH_SETUP                      = 152 /* 0x98 */,
+	HAL_REO_UPDATE_RX_REO_QUEUE_STATUS     = 153 /* 0x99 */,
+	HAL_TX_PEER_ENTRY                      = 154 /* 0x9a */,
+	HAL_TX_RAW_OR_NATIVE_FRAME_SETUP       = 155 /* 0x9b */,
+	HAL_EXAMPLE_STRUCT_NAME                = 156 /* 0x9c */,
+	HAL_PCU_PPDU_SETUP_END_INFO            = 157 /* 0x9d */,
+	HAL_PPDU_RATE_SETTING                  = 158 /* 0x9e */,
+	HAL_PROT_RATE_SETTING                  = 159 /* 0x9f */,
+	HAL_RX_MPDU_DETAILS                    = 160 /* 0xa0 */,
+	HAL_EXAMPLE_USER_TLV_42                = 161 /* 0xa1 */,
+	HAL_RX_MSDU_LINK                       = 162 /* 0xa2 */,
+	HAL_RX_REO_QUEUE                       = 163 /* 0xa3 */,
+	HAL_ADDR_SEARCH_ENTRY                  = 164 /* 0xa4 */,
+	HAL_SCHEDULER_CMD                      = 165 /* 0xa5 */,
+	HAL_TX_FLUSH                           = 166 /* 0xa6 */,
+	HAL_TQM_ENTRANCE_RING                  = 167 /* 0xa7 */,
+	HAL_TX_DATA_WORD                       = 168 /* 0xa8 */,
+	HAL_TX_MPDU_DETAILS                    = 169 /* 0xa9 */,
+	HAL_TX_MPDU_LINK                       = 170 /* 0xaa */,
+	HAL_TX_MPDU_LINK_PTR                   = 171 /* 0xab */,
+	HAL_TX_MPDU_QUEUE_HEAD                 = 172 /* 0xac */,
+	HAL_TX_MPDU_QUEUE_EXT                  = 173 /* 0xad */,
+	HAL_TX_MPDU_QUEUE_EXT_PTR              = 174 /* 0xae */,
+	HAL_TX_MSDU_DETAILS                    = 175 /* 0xaf */,
+	HAL_TX_MSDU_EXTENSION                  = 176 /* 0xb0 */,
+	HAL_TX_MSDU_FLOW                       = 177 /* 0xb1 */,
+	HAL_TX_MSDU_LINK                       = 178 /* 0xb2 */,
+	HAL_TX_MSDU_LINK_ENTRY_PTR             = 179 /* 0xb3 */,
+	HAL_RESPONSE_RATE_SETTING              = 180 /* 0xb4 */,
+	HAL_TXPCU_BUFFER_BASICS                = 181 /* 0xb5 */,
+	HAL_UNIFORM_DESCRIPTOR_HEADER          = 182 /* 0xb6 */,
+	HAL_UNIFORM_TQM_CMD_HEADER             = 183 /* 0xb7 */,
+	HAL_UNIFORM_TQM_STATUS_HEADER          = 184 /* 0xb8 */,
+	HAL_USER_RATE_SETTING                  = 185 /* 0xb9 */,
+	HAL_WBM_BUFFER_RING                    = 186 /* 0xba */,
+	HAL_WBM_LINK_DESCRIPTOR_RING           = 187 /* 0xbb */,
+	HAL_WBM_RELEASE_RING                   = 188 /* 0xbc */,
+	HAL_TX_FLUSH_REQ                       = 189 /* 0xbd */,
+	HAL_RX_MSDU_DETAILS                    = 190 /* 0xbe */,
+	HAL_TQM_WRITE_CMD_STATUS               = 191 /* 0xbf */,
+	HAL_TQM_GET_MPDU_QUEUE_STATS           = 192 /* 0xc0 */,
+	HAL_TQM_GET_MSDU_FLOW_STATS            = 193 /* 0xc1 */,
+	HAL_EXAMPLE_USER_CTLV_32               = 194 /* 0xc2 */,
+	HAL_TX_FES_STATUS_START                = 195 /* 0xc3 */,
+	HAL_TX_FES_STATUS_USER_PPDU            = 196 /* 0xc4 */,
+	HAL_TX_FES_STATUS_USER_RESPONSE        = 197 /* 0xc5 */,
+	HAL_TX_FES_STATUS_END                  = 198 /* 0xc6 */,
+	HAL_RX_TRIG_INFO                       = 199 /* 0xc7 */,
+	HAL_RXPCU_TX_SETUP_CLEAR               = 200 /* 0xc8 */,
+	HAL_RX_FRAME_BITMAP_REQ                = 201 /* 0xc9 */,
+	HAL_RX_FRAME_BITMAP_ACK                = 202 /* 0xca */,
+	HAL_COEX_RX_STATUS                     = 203 /* 0xcb */,
+	HAL_RX_START_PARAM                     = 204 /* 0xcc */,
+	HAL_RX_PPDU_START                      = 205 /* 0xcd */,
+	HAL_RX_PPDU_END                        = 206 /* 0xce */,
+	HAL_RX_MPDU_START                      = 207 /* 0xcf */,
+	HAL_RX_MPDU_END                        = 208 /* 0xd0 */,
+	HAL_RX_MSDU_START                      = 209 /* 0xd1 */,
+	HAL_RX_MSDU_END                        = 210 /* 0xd2 */,
+	HAL_RX_ATTENTION                       = 211 /* 0xd3 */,
+	HAL_RECEIVED_RESPONSE_INFO             = 212 /* 0xd4 */,
+	HAL_RX_PHY_SLEEP                       = 213 /* 0xd5 */,
+	HAL_RX_HEADER                          = 214 /* 0xd6 */,
+	HAL_RX_PEER_ENTRY                      = 215 /* 0xd7 */,
+	HAL_RX_FLUSH                           = 216 /* 0xd8 */,
+	HAL_RX_RESPONSE_REQUIRED_INFO          = 217 /* 0xd9 */,
+	HAL_RX_FRAMELESS_BAR_DETAILS           = 218 /* 0xda */,
+	HAL_TQM_GET_MPDU_QUEUE_STATS_STATUS    = 219 /* 0xdb */,
+	HAL_TQM_GET_MSDU_FLOW_STATS_STATUS     = 220 /* 0xdc */,
+	HAL_TX_CBF_INFO                        = 221 /* 0xdd */,
+	HAL_PCU_PPDU_SETUP_USER                = 222 /* 0xde */,
+	HAL_RX_MPDU_PCU_START                  = 223 /* 0xdf */,
+	HAL_RX_PM_INFO                         = 224 /* 0xe0 */,
+	HAL_RX_USER_PPDU_END                   = 225 /* 0xe1 */,
+	HAL_RX_PRE_PPDU_START                  = 226 /* 0xe2 */,
+	HAL_RX_PREAMBLE                        = 227 /* 0xe3 */,
+	HAL_TX_FES_SETUP_COMPLETE              = 228 /* 0xe4 */,
+	HAL_TX_LAST_MPDU_FETCHED               = 229 /* 0xe5 */,
+	HAL_TXDMA_STOP_REQUEST                 = 230 /* 0xe6 */,
+	HAL_RXPCU_SETUP                        = 231 /* 0xe7 */,
+	HAL_RXPCU_USER_SETUP                   = 232 /* 0xe8 */,
+	HAL_TX_FES_STATUS_ACK_OR_BA            = 233 /* 0xe9 */,
+	HAL_TQM_ACKED_MPDU                     = 234 /* 0xea */,
+	HAL_COEX_TX_RESP                       = 235 /* 0xeb */,
+	HAL_COEX_TX_STATUS                     = 236 /* 0xec */,
+	HAL_MACTX_COEX_PHY_CTRL                = 237 /* 0xed */,
+	HAL_COEX_STATUS_BROADCAST              = 238 /* 0xee */,
+	HAL_RESPONSE_START_STATUS              = 239 /* 0xef */,
+	HAL_RESPONSE_END_STATUS                = 240 /* 0xf0 */,
+	HAL_CRYPTO_STATUS                      = 241 /* 0xf1 */,
+	HAL_RECEIVED_TRIGGER_INFO              = 242 /* 0xf2 */,
+	HAL_REO_ENTRANCE_RING                  = 243 /* 0xf3 */,
+	HAL_RX_MPDU_LINK                       = 244 /* 0xf4 */,
+	HAL_COEX_TX_STOP_CTRL                  = 245 /* 0xf5 */,
+	HAL_RX_PPDU_ACK_REPORT                 = 246 /* 0xf6 */,
+	HAL_RX_PPDU_NO_ACK_REPORT              = 247 /* 0xf7 */,
+	HAL_SCH_COEX_STATUS                    = 248 /* 0xf8 */,
+	HAL_SCHEDULER_COMMAND_STATUS           = 249 /* 0xf9 */,
+	HAL_SCHEDULER_RX_PPDU_NO_RESPONSE_STATUS = 250 /* 0xfa */,
+	HAL_TX_FES_STATUS_PROT                 = 251 /* 0xfb */,
+	HAL_TX_FES_STATUS_START_PPDU           = 252 /* 0xfc */,
+	HAL_TX_FES_STATUS_START_PROT           = 253 /* 0xfd */,
+	HAL_TXPCU_PHYTX_DEBUG32                = 254 /* 0xfe */,
+	HAL_TXPCU_PHYTX_OTHER_TRANSMIT_INFO32  = 255 /* 0xff */,
+	HAL_TX_MPDU_COUNT_TRANSFER_END         = 256 /* 0x100 */,
+	HAL_WHO_ANCHOR_OFFSET                  = 257 /* 0x101 */,
+	HAL_WHO_ANCHOR_VALUE                   = 258 /* 0x102 */,
+	HAL_WHO_CCE_INFO                       = 259 /* 0x103 */,
+	HAL_WHO_COMMIT                         = 260 /* 0x104 */,
+	HAL_WHO_COMMIT_DONE                    = 261 /* 0x105 */,
+	HAL_WHO_FLUSH                          = 262 /* 0x106 */,
+	HAL_WHO_L2_LLC                         = 263 /* 0x107 */,
+	HAL_WHO_L2_PAYLOAD                     = 264 /* 0x108 */,
+	HAL_WHO_L3_CHECKSUM                    = 265 /* 0x109 */,
+	HAL_WHO_L3_INFO                        = 266 /* 0x10a */,
+	HAL_WHO_L4_CHECKSUM                    = 267 /* 0x10b */,
+	HAL_WHO_L4_INFO                        = 268 /* 0x10c */,
+	HAL_WHO_MSDU                           = 269 /* 0x10d */,
+	HAL_WHO_MSDU_MISC                      = 270 /* 0x10e */,
+	HAL_WHO_PACKET_DATA                    = 271 /* 0x10f */,
+	HAL_WHO_PACKET_HDR                     = 272 /* 0x110 */,
+	HAL_WHO_PPDU_END                       = 273 /* 0x111 */,
+	HAL_WHO_PPDU_START                     = 274 /* 0x112 */,
+	HAL_WHO_TSO                            = 275 /* 0x113 */,
+	HAL_WHO_WMAC_HEADER_PV0                = 276 /* 0x114 */,
+	HAL_WHO_WMAC_HEADER_PV1                = 277 /* 0x115 */,
+	HAL_WHO_WMAC_IV                        = 278 /* 0x116 */,
+	HAL_MPDU_INFO_END                      = 279 /* 0x117 */,
+	HAL_MPDU_INFO_BITMAP                   = 280 /* 0x118 */,
+	HAL_TX_QUEUE_EXTENSION                 = 281 /* 0x119 */,
+	HAL_RX_PEER_ENTRY_DETAILS              = 282 /* 0x11a */,
+	HAL_RX_REO_QUEUE_REFERENCE             = 283 /* 0x11b */,
+	HAL_RX_REO_QUEUE_EXT                   = 284 /* 0x11c */,
+	HAL_SCHEDULER_SELFGEN_RESPONSE_STATUS  = 285 /* 0x11d */,
+	HAL_TQM_UPDATE_TX_MPDU_COUNT_STATUS    = 286 /* 0x11e */,
+	HAL_TQM_ACKED_MPDU_STATUS              = 287 /* 0x11f */,
+	HAL_TQM_ADD_MSDU_STATUS                = 288 /* 0x120 */,
+	HAL_RX_MPDU_LINK_PTR                   = 289 /* 0x121 */,
+	HAL_REO_DESTINATION_RING               = 290 /* 0x122 */,
+	HAL_TQM_LIST_GEN_DONE                  = 291 /* 0x123 */,
+	HAL_WHO_TERMINATE                      = 292 /* 0x124 */,
+	HAL_TX_LAST_MPDU_END                   = 293 /* 0x125 */,
+	HAL_TX_CV_DATA                         = 294 /* 0x126 */,
+	HAL_TCL_ENTRANCE_FROM_PPE_RING         = 295 /* 0x127 */,
+	HAL_PPDU_TX_END                        = 296 /* 0x128 */,
+	HAL_PROT_TX_END                        = 297 /* 0x129 */,
+	HAL_PDG_RESPONSE_RATE_SETTING          = 298 /* 0x12a */,
+	HAL_MPDU_INFO_GLOBAL_END               = 299 /* 0x12b */,
+	HAL_TQM_SCH_INSTR_GLOBAL_END           = 300 /* 0x12c */,
+	HAL_RX_PPDU_END_USER_STATS             = 301 /* 0x12d */,
+	HAL_RX_PPDU_END_USER_STATS_EXT         = 302 /* 0x12e */,
+	HAL_NO_ACK_REPORT                      = 303 /* 0x12f */,
+	HAL_ACK_REPORT                         = 304 /* 0x130 */,
+	HAL_UNIFORM_REO_CMD_HEADER             = 305 /* 0x131 */,
+	HAL_REO_GET_QUEUE_STATS                = 306 /* 0x132 */,
+	HAL_REO_FLUSH_QUEUE                    = 307 /* 0x133 */,
+	HAL_REO_FLUSH_CACHE                    = 308 /* 0x134 */,
+	HAL_REO_UNBLOCK_CACHE                  = 309 /* 0x135 */,
+	HAL_UNIFORM_REO_STATUS_HEADER          = 310 /* 0x136 */,
+	HAL_REO_GET_QUEUE_STATS_STATUS         = 311 /* 0x137 */,
+	HAL_REO_FLUSH_QUEUE_STATUS             = 312 /* 0x138 */,
+	HAL_REO_FLUSH_CACHE_STATUS             = 313 /* 0x139 */,
+	HAL_REO_UNBLOCK_CACHE_STATUS           = 314 /* 0x13a */,
+	HAL_TQM_FLUSH_CACHE                    = 315 /* 0x13b */,
+	HAL_TQM_UNBLOCK_CACHE                  = 316 /* 0x13c */,
+	HAL_TQM_FLUSH_CACHE_STATUS             = 317 /* 0x13d */,
+	HAL_TQM_UNBLOCK_CACHE_STATUS           = 318 /* 0x13e */,
+	HAL_RX_PPDU_END_STATUS_DONE            = 319 /* 0x13f */,
+	HAL_RX_STATUS_BUFFER_DONE              = 320 /* 0x140 */,
+	HAL_BUFFER_ADDR_INFO                   = 321 /* 0x141 */,
+	HAL_RX_MSDU_DESC_INFO                  = 322 /* 0x142 */,
+	HAL_RX_MPDU_DESC_INFO                  = 323 /* 0x143 */,
+	HAL_TCL_DATA_CMD                       = 324 /* 0x144 */,
+	HAL_TCL_GSE_CMD                        = 325 /* 0x145 */,
+	HAL_TCL_EXIT_BASE                      = 326 /* 0x146 */,
+	HAL_TCL_COMPACT_EXIT_RING              = 327 /* 0x147 */,
+	HAL_TCL_REGULAR_EXIT_RING              = 328 /* 0x148 */,
+	HAL_TCL_EXTENDED_EXIT_RING             = 329 /* 0x149 */,
+	HAL_UPLINK_COMMON_INFO                 = 330 /* 0x14a */,
+	HAL_UPLINK_USER_SETUP_INFO             = 331 /* 0x14b */,
+	HAL_TX_DATA_SYNC                       = 332 /* 0x14c */,
+	HAL_PHYRX_CBF_READ_REQUEST_ACK         = 333 /* 0x14d */,
+	HAL_TCL_STATUS_RING                    = 334 /* 0x14e */,
+	HAL_TQM_GET_MPDU_HEAD_INFO             = 335 /* 0x14f */,
+	HAL_TQM_SYNC_CMD                       = 336 /* 0x150 */,
+	HAL_TQM_GET_MPDU_HEAD_INFO_STATUS      = 337 /* 0x151 */,
+	HAL_TQM_SYNC_CMD_STATUS                = 338 /* 0x152 */,
+	HAL_TQM_THRESHOLD_DROP_NOTIFICATION_STATUS = 339 /* 0x153 */,
+	HAL_TQM_DESCRIPTOR_THRESHOLD_REACHED_STATUS = 340 /* 0x154 */,
+	HAL_REO_FLUSH_TIMEOUT_LIST             = 341 /* 0x155 */,
+	HAL_REO_FLUSH_TIMEOUT_LIST_STATUS      = 342 /* 0x156 */,
+	HAL_REO_TO_PPE_RING                    = 343 /* 0x157 */,
+	HAL_RX_MPDU_INFO                       = 344 /* 0x158 */,
+	HAL_REO_DESCRIPTOR_THRESHOLD_REACHED_STATUS = 345 /* 0x159 */,
+	HAL_SCHEDULER_RX_SIFS_RESPONSE_TRIGGER_STATUS = 346 /* 0x15a */,
+	HAL_EXAMPLE_USER_TLV_32_NAME           = 347 /* 0x15b */,
+	HAL_RX_PPDU_START_USER_INFO            = 348 /* 0x15c */,
+	HAL_RX_RXPCU_CLASSIFICATION_OVERVIEW   = 349 /* 0x15d */,
+	HAL_RX_RING_MASK                       = 350 /* 0x15e */,
+	HAL_WHO_CLASSIFY_INFO                  = 351 /* 0x15f */,
+	HAL_TXPT_CLASSIFY_INFO                 = 352 /* 0x160 */,
+	HAL_RXPT_CLASSIFY_INFO                 = 353 /* 0x161 */,
+	HAL_TX_FLOW_SEARCH_ENTRY               = 354 /* 0x162 */,
+	HAL_RX_FLOW_SEARCH_ENTRY               = 355 /* 0x163 */,
+	HAL_RECEIVED_TRIGGER_INFO_DETAILS      = 356 /* 0x164 */,
+	HAL_COEX_MAC_NAP                       = 357 /* 0x165 */,
+	HAL_MACRX_ABORT_REQUEST_INFO           = 358 /* 0x166 */,
+	HAL_MACTX_ABORT_REQUEST_INFO           = 359 /* 0x167 */,
+	HAL_PHYRX_ABORT_REQUEST_INFO           = 360 /* 0x168 */,
+	HAL_PHYTX_ABORT_REQUEST_INFO           = 361 /* 0x169 */,
+	HAL_RXPCU_PPDU_END_INFO                = 362 /* 0x16a */,
+	HAL_WHO_MESH_CONTROL                   = 363 /* 0x16b */,
+	HAL_L_SIG_A_INFO                       = 364 /* 0x16c */,
+	HAL_L_SIG_B_INFO                       = 365 /* 0x16d */,
+	HAL_HT_SIG_INFO                        = 366 /* 0x16e */,
+	HAL_VHT_SIG_A_INFO                     = 367 /* 0x16f */,
+	HAL_VHT_SIG_B_SU20_INFO                = 368 /* 0x170 */,
+	HAL_VHT_SIG_B_SU40_INFO                = 369 /* 0x171 */,
+	HAL_VHT_SIG_B_SU80_INFO                = 370 /* 0x172 */,
+	HAL_VHT_SIG_B_SU160_INFO               = 371 /* 0x173 */,
+	HAL_VHT_SIG_B_MU20_INFO                = 372 /* 0x174 */,
+	HAL_VHT_SIG_B_MU40_INFO                = 373 /* 0x175 */,
+	HAL_VHT_SIG_B_MU80_INFO                = 374 /* 0x176 */,
+	HAL_VHT_SIG_B_MU160_INFO               = 375 /* 0x177 */,
+	HAL_SERVICE_INFO                       = 376 /* 0x178 */,
+	HAL_HE_SIG_A_SU_INFO                   = 377 /* 0x179 */,
+	HAL_HE_SIG_A_MU_DL_INFO                = 378 /* 0x17a */,
+	HAL_HE_SIG_A_MU_UL_INFO                = 379 /* 0x17b */,
+	HAL_HE_SIG_B1_MU_INFO                  = 380 /* 0x17c */,
+	HAL_HE_SIG_B2_MU_INFO                  = 381 /* 0x17d */,
+	HAL_HE_SIG_B2_OFDMA_INFO               = 382 /* 0x17e */,
+	HAL_PDG_SW_MODE_BW_START               = 383 /* 0x17f */,
+	HAL_PDG_SW_MODE_BW_END                 = 384 /* 0x180 */,
+	HAL_PDG_WAIT_FOR_MAC_REQUEST           = 385 /* 0x181 */,
+	HAL_PDG_WAIT_FOR_PHY_REQUEST           = 386 /* 0x182 */,
+	HAL_SCHEDULER_END                      = 387 /* 0x183 */,
+	HAL_PEER_TABLE_ENTRY                   = 388 /* 0x184 */,
+	HAL_SW_PEER_INFO                       = 389 /* 0x185 */,
+	HAL_RXOLE_CCE_CLASSIFY_INFO            = 390 /* 0x186 */,
+	HAL_TCL_CCE_CLASSIFY_INFO              = 391 /* 0x187 */,
+	HAL_RXOLE_CCE_INFO                     = 392 /* 0x188 */,
+	HAL_TCL_CCE_INFO                       = 393 /* 0x189 */,
+	HAL_TCL_CCE_SUPERRULE                  = 394 /* 0x18a */,
+	HAL_CCE_RULE                           = 395 /* 0x18b */,
+	HAL_RX_PPDU_START_DROPPED              = 396 /* 0x18c */,
+	HAL_RX_PPDU_END_DROPPED                = 397 /* 0x18d */,
+	HAL_RX_PPDU_END_STATUS_DONE_DROPPED    = 398 /* 0x18e */,
+	HAL_RX_MPDU_START_DROPPED              = 399 /* 0x18f */,
+	HAL_RX_MSDU_START_DROPPED              = 400 /* 0x190 */,
+	HAL_RX_MSDU_END_DROPPED                = 401 /* 0x191 */,
+	HAL_RX_MPDU_END_DROPPED                = 402 /* 0x192 */,
+	HAL_RX_ATTENTION_DROPPED               = 403 /* 0x193 */,
+	HAL_TXPCU_USER_SETUP                   = 404 /* 0x194 */,
+	HAL_RXPCU_USER_SETUP_EXT               = 405 /* 0x195 */,
+	HAL_CE_SRC_DESC                        = 406 /* 0x196 */,
+	HAL_CE_STAT_DESC                       = 407 /* 0x197 */,
+	HAL_RXOLE_CCE_SUPERRULE                = 408 /* 0x198 */,
+	HAL_TX_RATE_STATS_INFO                 = 409 /* 0x199 */,
+	HAL_CMD_PART_0_END                     = 410 /* 0x19a */,
+	HAL_MACTX_SYNTH_ON                     = 411 /* 0x19b */,
+	HAL_SCH_CRITICAL_TLV_REFERENCE         = 412 /* 0x19c */,
+	HAL_TQM_MPDU_GLOBAL_START              = 413 /* 0x19d */,
+	HAL_EXAMPLE_TLV_32                     = 414 /* 0x19e */,
+	HAL_TQM_UPDATE_TX_MSDU_FLOW            = 415 /* 0x19f */,
+	HAL_TQM_UPDATE_TX_MPDU_QUEUE_HEAD      = 416 /* 0x1a0 */,
+	HAL_TQM_UPDATE_TX_MSDU_FLOW_STATUS     = 417 /* 0x1a1 */,
+	HAL_TQM_UPDATE_TX_MPDU_QUEUE_HEAD_STATUS = 418 /* 0x1a2 */,
+	HAL_REO_UPDATE_RX_REO_QUEUE            = 419 /* 0x1a3 */,
+	HAL_CE_DST_DESC			       = 420 /* 0x1a4 */,
+	HAL_TLV_BASE                           = 511 /* 0x1ff */,
+};
+
+#define HAL_TLV_HDR_TAG		GENMASK(9, 1)
+#define HAL_TLV_HDR_LEN		GENMASK(25, 10)
+
+#define HAL_TLV_ALIGN	4
+
+struct hal_tlv_hdr {
+	u32 tl;
+	u8 value[0];
+} __packed;
+
+#define RX_MPDU_DESC_INFO0_MSDU_COUNT		GENMASK(7, 0)
+#define RX_MPDU_DESC_INFO0_SEQ_NUM		GENMASK(19, 8)
+#define RX_MPDU_DESC_INFO0_FRAG_FLAG		BIT(20)
+#define RX_MPDU_DESC_INFO0_MPDU_RETRY		BIT(21)
+#define RX_MPDU_DESC_INFO0_AMPDU_FLAG		BIT(22)
+#define RX_MPDU_DESC_INFO0_BAR_FRAME		BIT(23)
+#define RX_MPDU_DESC_INFO0_VALID_PN		BIT(24)
+#define RX_MPDU_DESC_INFO0_VALID_SA		BIT(25)
+#define RX_MPDU_DESC_INFO0_SA_IDX_TIMEOUT	BIT(26)
+#define RX_MPDU_DESC_INFO0_VALID_DA		BIT(27)
+#define RX_MPDU_DESC_INFO0_DA_MCBC		BIT(28)
+#define RX_MPDU_DESC_INFO0_DA_IDX_TIMEOUT	BIT(29)
+#define RX_MPDU_DESC_INFO0_RAW_MPDU		BIT(30)
+
+struct rx_mpdu_desc {
+	u32 info0; /* %RX_MPDU_DESC_INFO */
+	u32 meta_data;
+} __packed;
+
+/* rx_mpdu_desc
+ *		Producer: RXDMA
+ *		Consumer: REO/SW/FW
+ *
+ * msdu_count
+ *		The number of MSDUs within the MPDU
+ *
+ * mpdu_sequence_number
+ *		The field can have two different meanings based on the setting
+ *		of field 'bar_frame'. If 'bar_frame' is set, it means the MPDU
+ *		start sequence number from the BAR frame otherwise it means
+ *		the MPDU sequence number of the received frame.
+ *
+ * fragment_flag
+ *		When set, this MPDU is a fragment and REO should forward this
+ *		fragment MPDU to the REO destination ring without any reorder
+ *		checks, pn checks or bitmap update. This implies that REO is
+ *		forwarding the pointer to the MSDU link descriptor.
+ *
+ * mpdu_retry_bit
+ *		The retry bit setting from the MPDU header of the received frame
+ *
+ * ampdu_flag
+ *		Indicates the MPDU was received as part of an A-MPDU.
+ *
+ * bar_frame
+ *		Indicates the received frame is a BAR frame. After processing,
+ *		this frame shall be pushed to SW or deleted.
+ *
+ * valid_pn
+ *		When not set, REO will not perform a PN sequence number check.
+ *
+ * valid_sa
+ *		Indicates OLE found a valid SA entry for all MSDUs in this MPDU.
+ *
+ * sa_idx_timeout
+ *		Indicates, at least 1 MSDU within the MPDU has an unsuccessful
+ *		MAC source address search due to the expiration of search timer.
+ *
+ * valid_da
+ *		When set, OLE found a valid DA entry for all MSDUs in this MPDU.
+ *
+ * da_mcbc
+ *		Field Only valid if valid_da is set. Indicates at least one of
+ *		the DA addresses is a Multicast or Broadcast address.
+ *
+ * da_idx_timeout
+ *		Indicates, at least 1 MSDU within the MPDU has an unsuccessful
+ *		MAC destination address search due to the expiration of search
+ *		timer.
+ *
+ * raw_mpdu
+ *		Field only valid when first_msdu_in_mpdu_flag is set. Indicates
+ *		the contents in the MSDU buffer contains a 'RAW' MPDU.
+ */
+
+enum hal_rx_msdu_desc_reo_dest_ind {
+	HAL_RX_MSDU_DESC_REO_DEST_IND_TCL,
+	HAL_RX_MSDU_DESC_REO_DEST_IND_SW1,
+	HAL_RX_MSDU_DESC_REO_DEST_IND_SW2,
+	HAL_RX_MSDU_DESC_REO_DEST_IND_SW3,
+	HAL_RX_MSDU_DESC_REO_DEST_IND_SW4,
+	HAL_RX_MSDU_DESC_REO_DEST_IND_RELEASE,
+	HAL_RX_MSDU_DESC_REO_DEST_IND_FW,
+};
+
+#define RX_MSDU_DESC_INFO0_FIRST_MSDU_IN_MPDU	BIT(0)
+#define RX_MSDU_DESC_INFO0_LAST_MSDU_IN_MPDU	BIT(1)
+#define RX_MSDU_DESC_INFO0_MSDU_CONTINUATION	BIT(2)
+#define RX_MSDU_DESC_INFO0_MSDU_LENGTH		GENMASK(16, 3)
+#define RX_MSDU_DESC_INFO0_REO_DEST_IND		GENMASK(21, 17)
+#define RX_MSDU_DESC_INFO0_MSDU_DROP		BIT(22)
+#define RX_MSDU_DESC_INFO0_VALID_SA		BIT(23)
+#define RX_MSDU_DESC_INFO0_SA_IDX_TIMEOUT	BIT(24)
+#define RX_MSDU_DESC_INFO0_VALID_DA		BIT(25)
+#define RX_MSDU_DESC_INFO0_DA_MCBC		BIT(26)
+#define RX_MSDU_DESC_INFO0_DA_IDX_TIMEOUT	BIT(27)
+
+#define HAL_RX_MSDU_PKT_LENGTH_GET(val)		\
+	(FIELD_GET(RX_MSDU_DESC_INFO0_MSDU_LENGTH, (val)))
+
+struct rx_msdu_desc {
+	u32 info0;
+	u32 rsvd0;
+} __packed;
+
+/* rx_msdu_desc
+ *
+ * first_msdu_in_mpdu
+ *		Indicates first msdu in mpdu.
+ *
+ * last_msdu_in_mpdu
+ *		Indicates last msdu in mpdu. This flag can be true only when
+ *		'Msdu_continuation' set to 0. This implies that when an msdu
+ *		is spread out over multiple buffers and thus msdu_continuation
+ *		is set, only for the very last buffer of the msdu, can the
+ *		'last_msdu_in_mpdu' be set.
+ *
+ *		When both first_msdu_in_mpdu and last_msdu_in_mpdu are set,
+ *		the MPDU that this MSDU belongs to only contains a single MSDU.
+ *
+ * msdu_continuation
+ *		When set, this MSDU buffer was not able to hold the entire MSDU.
+ *		The next buffer will therefor contain additional information
+ *		related to this MSDU.
+ *
+ * msdu_length
+ *		Field is only valid in combination with the 'first_msdu_in_mpdu'
+ *		being set. Full MSDU length in bytes after decapsulation. This
+ *		field is still valid for MPDU frames without A-MSDU. It still
+ *		represents MSDU length after decapsulation Or in case of RAW
+ *		MPDUs, it indicates the length of the entire MPDU (without FCS
+ *		field).
+ *
+ * reo_destination_indication
+ *		The id of the reo exit ring where the msdu frame shall push
+ *		after (MPDU level) reordering has finished. Values are defined
+ *		in enum %HAL_RX_MSDU_DESC_REO_DEST_IND_.
+ *
+ * msdu_drop
+ *		Indicates that REO shall drop this MSDU and not forward it to
+ *		any other ring.
+ *
+ * valid_sa
+ *		Indicates OLE found a valid SA entry for this MSDU.
+ *
+ * sa_idx_timeout
+ *		Indicates, an unsuccessful MAC source address search due to
+ *		the expiration of search timer for this MSDU.
+ *
+ * valid_da
+ *		When set, OLE found a valid DA entry for this MSDU.
+ *
+ * da_mcbc
+ *		Field Only valid if valid_da is set. Indicates the DA address
+ *		is a Multicast or Broadcast address for this MSDU.
+ *
+ * da_idx_timeout
+ *		Indicates, an unsuccessful MAC destination address search due
+ *		to the expiration of search timer fot this MSDU.
+ */
+
+enum hal_reo_dest_ring_buffer_type {
+	HAL_REO_DEST_RING_BUFFER_TYPE_MSDU,
+	HAL_REO_DEST_RING_BUFFER_TYPE_LINK_DESC,
+};
+
+enum hal_reo_dest_ring_push_reason {
+	HAL_REO_DEST_RING_PUSH_REASON_ERR_DETECTED,
+	HAL_REO_DEST_RING_PUSH_REASON_ROUTING_INSTRUCTION,
+};
+
+enum hal_reo_dest_ring_error_code {
+	HAL_REO_DEST_RING_ERROR_CODE_DESC_ADDR_ZERO,
+	HAL_REO_DEST_RING_ERROR_CODE_DESC_INVALID,
+	HAL_REO_DEST_RING_ERROR_CODE_AMPDU_IN_NON_BA,
+	HAL_REO_DEST_RING_ERROR_CODE_NON_BA_DUPLICATE,
+	HAL_REO_DEST_RING_ERROR_CODE_BA_DUPLICATE,
+	HAL_REO_DEST_RING_ERROR_CODE_FRAME_2K_JUMP,
+	HAL_REO_DEST_RING_ERROR_CODE_BAR_2K_JUMP,
+	HAL_REO_DEST_RING_ERROR_CODE_FRAME_OOR,
+	HAL_REO_DEST_RING_ERROR_CODE_BAR_OOR,
+	HAL_REO_DEST_RING_ERROR_CODE_NO_BA_SESSION,
+	HAL_REO_DEST_RING_ERROR_CODE_FRAME_SN_EQUALS_SSN,
+	HAL_REO_DEST_RING_ERROR_CODE_PN_CHECK_FAILED,
+	HAL_REO_DEST_RING_ERROR_CODE_2K_ERR_FLAG_SET,
+	HAL_REO_DEST_RING_ERROR_CODE_PN_ERR_FLAG_SET,
+	HAL_REO_DEST_RING_ERROR_CODE_DESC_BLOCKED,
+	HAL_REO_DEST_RING_ERROR_CODE_MAX,
+};
+
+#define HAL_REO_DEST_RING_INFO0_QUEUE_ADDR_HI		GENMASK(7, 0)
+#define HAL_REO_DEST_RING_INFO0_BUFFER_TYPE		BIT(8)
+#define HAL_REO_DEST_RING_INFO0_PUSH_REASON		GENMASK(10, 9)
+#define HAL_REO_DEST_RING_INFO0_ERROR_CODE		GENMASK(15, 11)
+#define HAL_REO_DEST_RING_INFO0_RX_QUEUE_NUM		GENMASK(31, 16)
+
+#define HAL_REO_DEST_RING_INFO1_REORDER_INFO_VALID	BIT(0)
+#define HAL_REO_DEST_RING_INFO1_REORDER_OPCODE		GENMASK(4, 1)
+#define HAL_REO_DEST_RING_INFO1_REORDER_SLOT_IDX	GENMASK(12, 5)
+
+#define HAL_REO_DEST_RING_INFO2_RING_ID			GENMASK(27, 20)
+#define HAL_REO_DEST_RING_INFO2_LOOPING_COUNT		GENMASK(31, 28)
+
+struct hal_reo_dest_ring {
+	struct ath11k_buffer_addr buf_addr_info;
+	struct rx_mpdu_desc rx_mpdu_info;
+	struct rx_msdu_desc rx_msdu_info;
+	u32 queue_addr_lo;
+	u32 info0; /* %HAL_REO_DEST_RING_INFO0_ */
+	u32 info1; /* %HAL_REO_DEST_RING_INFO1_ */
+	u32 rsvd0;
+	u32 rsvd1;
+	u32 rsvd2;
+	u32 rsvd3;
+	u32 rsvd4;
+	u32 rsvd5;
+	u32 info2; /* %HAL_REO_DEST_RING_INFO2_ */
+} __packed;
+
+/* hal_reo_dest_ring
+ *
+ *		Producer: RXDMA
+ *		Consumer: REO/SW/FW
+ *
+ * buf_addr_info
+ *		Details of the physical address of a buffer or MSDU
+ *		link descriptor.
+ *
+ * rx_mpdu_info
+ *		General information related to the MPDU that is passed
+ *		on from REO entrance ring to the REO destination ring.
+ *
+ * rx_msdu_info
+ *		General information related to the MSDU that is passed
+ *		on from RXDMA all the way to to the REO destination ring.
+ *
+ * queue_addr_lo
+ *		Address (lower 32 bits) of the REO queue descriptor.
+ *
+ * queue_addr_hi
+ *		Address (upper 8 bits) of the REO queue descriptor.
+ *
+ * buffer_type
+ *		Indicates the type of address provided in the buf_addr_info.
+ *		Values are defined in enum %HAL_REO_DEST_RING_BUFFER_TYPE_.
+ *
+ * push_reason
+ *		Reason for pushing this frame to this exit ring. Values are
+ *		defined in enum %HAL_REO_DEST_RING_PUSH_REASON_.
+ *
+ * error_code
+ *		Valid only when 'push_reason' is set. All error codes are
+ *		defined in enum %HAL_REO_DEST_RING_ERROR_CODE_.
+ *
+ * rx_queue_num
+ *		Indicates the REO MPDU reorder queue id from which this frame
+ *		originated.
+ *
+ * reorder_info_valid
+ *		When set, REO has been instructed to not perform the actual
+ *		re-ordering of frames for this queue, but just to insert
+ *		the reorder opcodes.
+ *
+ * reorder_opcode
+ *		Field is valid when 'reorder_info_valid' is set. This field is
+ *		always valid for debug purpose as well.
+ *
+ * reorder_slot_idx
+ *		Valid only when 'reorder_info_valid' is set.
+ *
+ * ring_id
+ *		The buffer pointer ring id.
+ *		0 - Idle ring
+ *		1 - N refers to other rings.
+ *
+ * looping_count
+ *		Indicates the number of times the producer of entries into
+ *		this ring has looped around the ring.
+ */
+
+enum hal_reo_entr_rxdma_ecode {
+	HAL_REO_ENTR_RING_RXDMA_ECODE_OVERFLOW_ERR,
+	HAL_REO_ENTR_RING_RXDMA_ECODE_MPDU_LEN_ERR,
+	HAL_REO_ENTR_RING_RXDMA_ECODE_FCS_ERR,
+	HAL_REO_ENTR_RING_RXDMA_ECODE_DECRYPT_ERR,
+	HAL_REO_ENTR_RING_RXDMA_ECODE_TKIP_MIC_ERR,
+	HAL_REO_ENTR_RING_RXDMA_ECODE_UNECRYPTED_ERR,
+	HAL_REO_ENTR_RING_RXDMA_ECODE_MSDU_LEN_ERR,
+	HAL_REO_ENTR_RING_RXDMA_ECODE_MSDU_LIMIT_ERR,
+	HAL_REO_ENTR_RING_RXDMA_ECODE_WIFI_PARSE_ERR,
+	HAL_REO_ENTR_RING_RXDMA_ECODE_AMSDU_PARSE_ERR,
+	HAL_REO_ENTR_RING_RXDMA_ECODE_SA_TIMEOUT_ERR,
+	HAL_REO_ENTR_RING_RXDMA_ECODE_DA_TIMEOUT_ERR,
+	HAL_REO_ENTR_RING_RXDMA_ECODE_FLOW_TIMEOUT_ERR,
+	HAL_REO_ENTR_RING_RXDMA_ECODE_FLUSH_REQUEST_ERR,
+	HAL_REO_ENTR_RING_RXDMA_ECODE_MAX,
+};
+
+#define HAL_REO_ENTR_RING_INFO0_QUEUE_ADDR_HI		GENMASK(7, 0)
+#define HAL_REO_ENTR_RING_INFO0_MPDU_BYTE_COUNT		GENMASK(21, 8)
+#define HAL_REO_ENTR_RING_INFO0_DEST_IND		GENMASK(26, 22)
+#define HAL_REO_ENTR_RING_INFO0_FRAMELESS_BAR		BIT(27)
+
+#define HAL_REO_ENTR_RING_INFO1_RXDMA_PUSH_REASON	GENMASK(1, 0)
+#define HAL_REO_ENTR_RING_INFO1_RXDMA_ERROR_CODE	GENMASK(6, 2)
+
+struct hal_reo_entrance_ring {
+	struct ath11k_buffer_addr buf_addr_info;
+	struct rx_mpdu_desc rx_mpdu_info;
+	u32 queue_addr_lo;
+	u32 info0; /* %HAL_REO_ENTR_RING_INFO0_ */
+	u32 info1; /* %HAL_REO_ENTR_RING_INFO1_ */
+	u32 info2; /* %HAL_REO_DEST_RING_INFO2_ */
+
+} __packed;
+
+/* hal_reo_entrance_ring
+ *
+ *		Producer: RXDMA
+ *		Consumer: REO
+ *
+ * buf_addr_info
+ *		Details of the physical address of a buffer or MSDU
+ *		link descriptor.
+ *
+ * rx_mpdu_info
+ *		General information related to the MPDU that is passed
+ *		on from REO entrance ring to the REO destination ring.
+ *
+ * queue_addr_lo
+ *		Address (lower 32 bits) of the REO queue descriptor.
+ *
+ * queue_addr_hi
+ *		Address (upper 8 bits) of the REO queue descriptor.
+ *
+ * mpdu_byte_count
+ *		An approximation of the number of bytes received in this MPDU.
+ *		Used to keeps stats on the amount of data flowing
+ *		through a queue.
+ *
+ * reo_destination_indication
+ *		The id of the reo exit ring where the msdu frame shall push
+ *		after (MPDU level) reordering has finished. Values are defined
+ *		in enum %HAL_RX_MSDU_DESC_REO_DEST_IND_.
+ *
+ * frameless_bar
+ *		Indicates that this REO entrance ring struct contains BAR info
+ *		from a multi TID BAR frame. The original multi TID BAR frame
+ *		itself contained all the REO info for the first TID, but all
+ *		the subsequent TID info and their linkage to the REO descriptors
+ *		is passed down as 'frameless' BAR info.
+ *
+ *		The only fields valid in this descriptor when this bit is set
+ *		are queue_addr_lo, queue_addr_hi, mpdu_sequence_number,
+ *		bar_frame and peer_meta_data.
+ *
+ * rxdma_push_reason
+ *		Reason for pushing this frame to this exit ring. Values are
+ *		defined in enum %HAL_REO_DEST_RING_PUSH_REASON_.
+ *
+ * rxdma_error_code
+ *		Valid only when 'push_reason' is set. All error codes are
+ *		defined in enum %HAL_REO_ENTR_RING_RXDMA_ECODE_.
+ *
+ * ring_id
+ *		The buffer pointer ring id.
+ *		0 - Idle ring
+ *		1 - N refers to other rings.
+ *
+ * looping_count
+ *		Indicates the number of times the producer of entries into
+ *		this ring has looped around the ring.
+ */
+
+#define HAL_REO_CMD_HDR_INFO0_CMD_NUMBER	GENMASK(15, 0)
+#define HAL_REO_CMD_HDR_INFO0_STATUS_REQUIRED	BIT(16)
+
+struct hal_reo_cmd_hdr {
+	u32 info0;
+} __packed;
+
+#define HAL_REO_GET_QUEUE_STATS_INFO0_QUEUE_ADDR_HI	GENMASK(7, 0)
+#define HAL_REO_GET_QUEUE_STATS_INFO0_CLEAR_STATS	BIT(8)
+
+struct hal_reo_get_queue_stats {
+	struct hal_reo_cmd_hdr cmd;
+	u32 queue_addr_lo;
+	u32 info0;
+	u32 rsvd0[6];
+} __packed;
+
+/* hal_reo_get_queue_stats
+ *		Producer: SW
+ *		Consumer: REO
+ *
+ * cmd
+ *		Details for command execution tracking purposes.
+ *
+ * queue_addr_lo
+ *		Address (lower 32 bits) of the REO queue descriptor.
+ *
+ * queue_addr_hi
+ *		Address (upper 8 bits) of the REO queue descriptor.
+ *
+ * clear_stats
+ *		Clear stats settings. When set, Clear the stats after
+ *		generating the status.
+ *
+ *		Following stats will be cleared.
+ *		Timeout_count
+ *		Forward_due_to_bar_count
+ *		Duplicate_count
+ *		Frames_in_order_count
+ *		BAR_received_count
+ *		MPDU_Frames_processed_count
+ *		MSDU_Frames_processed_count
+ *		Total_processed_byte_count
+ *		Late_receive_MPDU_count
+ *		window_jump_2k
+ *		Hole_count
+ */
+
+#define HAL_REO_FLUSH_QUEUE_INFO0_DESC_ADDR_HI		GENMASK(7, 0)
+#define HAL_REO_FLUSH_QUEUE_INFO0_BLOCK_DESC_ADDR	BIT(8)
+#define HAL_REO_FLUSH_QUEUE_INFO0_BLOCK_RESRC_IDX	GENMASK(10, 9)
+
+struct hal_reo_flush_queue {
+	struct hal_reo_cmd_hdr cmd;
+	u32 desc_addr_lo;
+	u32 info0;
+	u32 rsvd0[6];
+} __packed;
+
+#define HAL_REO_FLUSH_CACHE_INFO0_CACHE_ADDR_HI		GENMASK(7, 0)
+#define HAL_REO_FLUSH_CACHE_INFO0_FWD_ALL_MPDUS		BIT(8)
+#define HAL_REO_FLUSH_CACHE_INFO0_RELEASE_BLOCK_IDX	BIT(9)
+#define HAL_REO_FLUSH_CACHE_INFO0_BLOCK_RESRC_IDX	GENMASK(11, 10)
+#define HAL_REO_FLUSH_CACHE_INFO0_FLUSH_WO_INVALIDATE	BIT(12)
+#define HAL_REO_FLUSH_CACHE_INFO0_BLOCK_CACHE_USAGE	BIT(13)
+#define HAL_REO_FLUSH_CACHE_INFO0_FLUSH_ALL		BIT(14)
+
+struct hal_reo_flush_cache {
+	struct hal_reo_cmd_hdr cmd;
+	u32 cache_addr_lo;
+	u32 info0;
+	u32 rsvd0[6];
+} __packed;
+
+#define HAL_TCL_DATA_CMD_INFO0_DESC_TYPE	BIT(0)
+#define HAL_TCL_DATA_CMD_INFO0_EPD		BIT(1)
+#define HAL_TCL_DATA_CMD_INFO0_ENCAP_TYPE	GENMASK(3, 2)
+#define HAL_TCL_DATA_CMD_INFO0_ENCRYPT_TYPE	GENMASK(7, 4)
+#define HAL_TCL_DATA_CMD_INFO0_SRC_BUF_SWAP	BIT(8)
+#define HAL_TCL_DATA_CMD_INFO0_LNK_META_SWAP	BIT(9)
+#define HAL_TCL_DATA_CMD_INFO0_SEARCH_TYPE	BIT(12)
+#define HAL_TCL_DATA_CMD_INFO0_ADDRX_EN		BIT(14)
+#define HAL_TCL_DATA_CMD_INFO0_ADDRY_EN		BIT(15)
+#define HAL_TCL_DATA_CMD_INFO0_CMD_NUM		GENMASK(31, 16)
+
+#define HAL_TCL_DATA_CMD_INFO1_DATA_LEN		GENMASK(15, 0)
+#define HAL_TCL_DATA_CMD_INFO1_IP4_CKSUM_EN	BIT(16)
+#define HAL_TCL_DATA_CMD_INFO1_UDP4_CKSUM_EN	BIT(17)
+#define HAL_TCL_DATA_CMD_INFO1_UDP6_CKSUM_EN	BIT(18)
+#define HAL_TCL_DATA_CMD_INFO1_TCP4_CKSUM_EN	BIT(19)
+#define HAL_TCL_DATA_CMD_INFO1_TCP6_CKSUM_EN	BIT(20)
+#define HAL_TCL_DATA_CMD_INFO1_TO_FW		BIT(21)
+#define HAL_TCL_DATA_CMD_INFO1_PKT_OFFSET	GENMASK(31, 23)
+
+#define HAL_TCL_DATA_CMD_INFO2_BUF_TIMESTAMP	GENMASK(18, 0)
+#define HAL_TCL_DATA_CMD_INFO2_BUF_T_VALID	BIT(19)
+#define HAL_TCL_DATA_CMD_INFO2_MESH_ENABLE	BIT(20)
+#define HAL_TCL_DATA_CMD_INFO2_TID_OVERWRITE	BIT(21)
+#define HAL_TCL_DATA_CMD_INFO2_TID		GENMASK(25, 22)
+#define HAL_TCL_DATA_CMD_INFO2_LMAC_ID		GENMASK(27, 26)
+
+#define HAL_TCL_DATA_CMD_INFO3_DSCP_TID_TABLE_IDX	GENMASK(5, 0)
+#define HAL_TCL_DATA_CMD_INFO3_SEARCH_INDEX		GENMASK(25, 6)
+#define HAL_TCL_DATA_CMD_INFO3_CACHE_SET_NUM		GENMASK(29, 26)
+
+#define HAL_TCL_DATA_CMD_INFO4_RING_ID			GENMASK(27, 20)
+#define HAL_TCL_DATA_CMD_INFO4_LOOPING_COUNT		GENMASK(31, 28)
+
+enum hal_encrypt_type {
+	HAL_ENCRYPT_TYPE_WEP_40,
+	HAL_ENCRYPT_TYPE_WEP_104,
+	HAL_ENCRYPT_TYPE_TKIP_NO_MIC,
+	HAL_ENCRYPT_TYPE_WEP_128,
+	HAL_ENCRYPT_TYPE_TKIP_MIC,
+	HAL_ENCRYPT_TYPE_WAPI,
+	HAL_ENCRYPT_TYPE_CCMP_128,
+	HAL_ENCRYPT_TYPE_OPEN,
+	HAL_ENCRYPT_TYPE_CCMP_256,
+	HAL_ENCRYPT_TYPE_GCMP_128,
+	HAL_ENCRYPT_TYPE_AES_GCMP_256,
+	HAL_ENCRYPT_TYPE_WAPI_GCM_SM4,
+};
+
+enum hal_tcl_encap_type {
+	HAL_TCL_ENCAP_TYPE_RAW,
+	HAL_TCL_ENCAP_TYPE_NATIVE_WIFI,
+	HAL_TCL_ENCAP_TYPE_ETHERNET,
+	HAL_TCL_ENCAP_TYPE_802_3 = 3,
+};
+
+enum hal_tcl_desc_type {
+	HAL_TCL_DESC_TYPE_BUFFER,
+	HAL_TCL_DESC_TYPE_EXT_DESC,
+};
+
+enum hal_wbm_htt_tx_comp_status {
+	HAL_WBM_REL_HTT_TX_COMP_STATUS_OK,
+	HAL_WBM_REL_HTT_TX_COMP_STATUS_DROP,
+	HAL_WBM_REL_HTT_TX_COMP_STATUS_TTL,
+	HAL_WBM_REL_HTT_TX_COMP_STATUS_REINJ,
+	HAL_WBM_REL_HTT_TX_COMP_STATUS_INSPECT,
+	HAL_WBM_REL_HTT_TX_COMP_STATUS_MEC_NOTIFY,
+};
+
+struct hal_tcl_data_cmd {
+	struct ath11k_buffer_addr buf_addr_info;
+	u32 info0;
+	u32 info1;
+	u32 info2;
+	u32 info3;
+	u32 info4;
+} __packed;
+
+/* hal_tcl_data_cmd
+ *
+ * buf_addr_info
+ *		Details of the physical address of a buffer or MSDU
+ *		link descriptor.
+ *
+ * desc_type
+ *		Indicates the type of address provided in the buf_addr_info.
+ *		Values are defined in enum %HAL_REO_DEST_RING_BUFFER_TYPE_.
+ *
+ * epd
+ *		When this bit is set then input packet is an EPD type.
+ *
+ * encap_type
+ *		Indicates the encapsulation that HW will perform. Values are
+ *		defined in enum %HAL_TCL_ENCAP_TYPE_.
+ *
+ * encrypt_type
+ *		Field only valid for encap_type: RAW
+ *		Values are defined in enum %HAL_ENCRYPT_TYPE_.
+ *
+ * src_buffer_swap
+ *		Treats source memory (packet buffer) organization as big-endian.
+ *		1'b0: Source memory is little endian
+ *		1'b1: Source memory is big endian
+ *
+ * link_meta_swap
+ *		Treats link descriptor and Metadata as big-endian.
+ *		1'b0: memory is little endian
+ *		1'b1: memory is big endian
+ *
+ * search_type
+ *		Search type select
+ *		0 - Normal search, 1 - Index based address search,
+ *		2 - Index based flow search
+ *
+ * addrx_en
+ * addry_en
+ *		Address X/Y search enable in ASE correspondingly.
+ *		1'b0: Search disable
+ *		1'b1: Search Enable
+ *
+ * cmd_num
+ *		This number can be used to match against status.
+ *
+ * data_length
+ *		MSDU length in case of direct descriptor. Length of link
+ *		extension descriptor in case of Link extension descriptor.
+ *
+ * *_checksum_en
+ *		Enable checksum replacement for ipv4, udp_over_ipv4, ipv6,
+ *		udp_over_ipv6, tcp_over_ipv4 and tcp_over_ipv6.
+ *
+ * to_fw
+ *		Forward packet to FW along with classification result. The
+ *		packet will not be forward to TQM when this bit is set.
+ *		1'b0: Use classification result to forward the packet.
+ *		1'b1: Override classification result & forward packet only to fw
+ *
+ * packet_offset
+ *		Packet offset from Metadata in case of direct buffer descriptor.
+ *
+ * buffer_timestamp
+ * buffer_timestamp_valid
+ *		Frame system entrance timestamp. It shall be filled by first
+ *		module (SW, TCL or TQM) that sees the frames first.
+ *
+ * mesh_enable
+ *		For raw WiFi frames, this indicates transmission to a mesh STA,
+ *		enabling the interpretation of the 'Mesh Control Present' bit
+ *		(bit 8) of QoS Control.
+ *		For native WiFi frames, this indicates that a 'Mesh Control'
+ *		field is present between the header and the LLC.
+ *
+ * hlos_tid_overwrite
+ *
+ *		When set, TCL shall ignore the IP DSCP and VLAN PCP
+ *		fields and use HLOS_TID as the final TID. Otherwise TCL
+ *		shall consider the DSCP and PCP fields as well as HLOS_TID
+ *		and choose a final TID based on the configured priority
+ *
+ * hlos_tid
+ *		HLOS MSDU priority
+ *		Field is used when HLOS_TID_overwrite is set.
+ *
+ * lmac_id
+ *		TCL uses this LMAC_ID in address search, i.e, while
+ *		finding matching entry for the packet in AST corresponding
+ *		to given LMAC_ID
+ *
+ *		If LMAC ID is all 1s (=> value 3), it indicates wildcard
+ *		match for any MAC
+ *
+ * dscp_tid_table_num
+ *		DSCP to TID mapping table number that need to be used
+ *		for the MSDU.
+ *
+ * search_index
+ *		The index that will be used for index based address or
+ *		flow search. The field is valid when 'search_type' is  1 or 2.
+ *
+ * cache_set_num
+ *
+ *		Cache set number that should be used to cache the index
+ *		based search results, for address and flow search. This
+ *		value should be equal to LSB four bits of the hash value of
+ *		match data, in case of search index points to an entry which
+ *		may be used in content based search also. The value can be
+ *		anything when the entry pointed by search index will not be
+ *		used for content based search.
+ *
+ * ring_id
+ *		The buffer pointer ring ID.
+ *		0 refers to the IDLE ring
+ *		1 - N refers to other rings
+ *
+ * looping_count
+ *
+ *		A count value that indicates the number of times the
+ *		producer of entries into the Ring has looped around the
+ *		ring.
+ *
+ *		At initialization time, this value is set to 0. On the
+ *		first loop, this value is set to 1. After the max value is
+ *		reached allowed by the number of bits for this field, the
+ *		count value continues with 0 again.
+ *
+ *		In case SW is the consumer of the ring entries, it can
+ *		use this field to figure out up to where the producer of
+ *		entries has created new entries. This eliminates the need to
+ *		check where the head pointer' of the ring is located once
+ *		the SW starts processing an interrupt indicating that new
+ *		entries have been put into this ring...
+ *
+ *		Also note that SW if it wants only needs to look at the
+ *		LSB bit of this count value.
+ */
+
+#define HAL_TCL_DESC_LEN sizeof(struct hal_tcl_data_cmd)
+
+enum hal_tcl_gse_ctrl {
+	HAL_TCL_GSE_CTRL_RD_STAT,
+	HAL_TCL_GSE_CTRL_SRCH_DIS,
+	HAL_TCL_GSE_CTRL_WR_BK_SINGLE,
+	HAL_TCL_GSE_CTRL_WR_BK_ALL,
+	HAL_TCL_GSE_CTRL_INVAL_SINGLE,
+	HAL_TCL_GSE_CTRL_INVAL_ALL,
+	HAL_TCL_GSE_CTRL_WR_BK_INVAL_SINGLE,
+	HAL_TCL_GSE_CTRL_WR_BK_INVAL_ALL,
+	HAL_TCL_GSE_CTRL_CLR_STAT_SINGLE,
+};
+
+/* hal_tcl_gse_ctrl
+ *
+ * rd_stat
+ *		Report or Read statistics
+ * srch_dis
+ *		Search disable. Report only Hash.
+ * wr_bk_single
+ *		Write Back single entry
+ * wr_bk_all
+ *		Write Back entire cache entry
+ * inval_single
+ *		Invalidate single cache entry
+ * inval_all
+ *		Invalidate entire cache
+ * wr_bk_inval_single
+ *		Write back and invalidate single entry in cache
+ * wr_bk_inval_all
+ *		Write back and invalidate entire cache
+ * clr_stat_single
+ *		Clear statistics for single entry
+ */
+
+#define HAL_TCL_GSE_CMD_INFO0_CTRL_BUF_ADDR_HI		GENMASK(7, 0)
+#define HAL_TCL_GSE_CMD_INFO0_GSE_CTRL			GENMASK(11, 8)
+#define HAL_TCL_GSE_CMD_INFO0_GSE_SEL			BIT(12)
+#define HAL_TCL_GSE_CMD_INFO0_STATUS_DEST_RING_ID	BIT(13)
+#define HAL_TCL_GSE_CMD_INFO0_SWAP			BIT(14)
+
+#define HAL_TCL_GSE_CMD_INFO1_RING_ID			GENMASK(27, 20)
+#define HAL_TCL_GSE_CMD_INFO1_LOOPING_COUNT		GENMASK(31, 28)
+
+struct hal_tcl_gse_cmd {
+	u32 ctrl_buf_addr_lo;
+	u32 info0;
+	u32 meta_data[2];
+	u32 rsvd0[2];
+	u32 info1;
+} __packed;
+
+/* hal_tcl_gse_cmd
+ *
+ * ctrl_buf_addr_lo, ctrl_buf_addr_hi
+ *		Address of a control buffer containing additional info needed
+ *		for this command execution.
+ *
+ * gse_ctrl
+ *		GSE control operations. This includes cache operations and table
+ *		entry statistics read/clear operation. Values are defined in
+ *		enum %HAL_TCL_GSE_CTRL.
+ *
+ * gse_sel
+ *		To select the ASE/FSE to do the operation mention by GSE_ctrl.
+ *		0: FSE select 1: ASE select
+ *
+ * status_destination_ring_id
+ *		TCL status ring to which the GSE status needs to be send.
+ *
+ * swap
+ *		Bit to enable byte swapping of contents of buffer.
+ *
+ * meta_data
+ *		Meta data to be returned in the status descriptor
+ */
+
+enum hal_tcl_cache_op_res {
+	HAL_TCL_CACHE_OP_RES_DONE,
+	HAL_TCL_CACHE_OP_RES_NOT_FOUND,
+	HAL_TCL_CACHE_OP_RES_TIMEOUT,
+};
+
+#define HAL_TCL_STATUS_RING_INFO0_GSE_CTRL		GENMASK(3, 0)
+#define HAL_TCL_STATUS_RING_INFO0_GSE_SEL		BIT(4)
+#define HAL_TCL_STATUS_RING_INFO0_CACHE_OP_RES		GENMASK(6, 5)
+#define HAL_TCL_STATUS_RING_INFO0_MSDU_CNT		GENMASK(31, 8)
+
+#define HAL_TCL_STATUS_RING_INFO1_HASH_IDX		GENMASK(19, 0)
+
+#define HAL_TCL_STATUS_RING_INFO2_RING_ID		GENMASK(27, 20)
+#define HAL_TCL_STATUS_RING_INFO2_LOOPING_COUNT		GENMASK(31, 28)
+
+struct hal_tcl_status_ring {
+	u32 info0;
+	u32 msdu_byte_count;
+	u32 msdu_timestamp;
+	u32 meta_data[2];
+	u32 info1;
+	u32 rsvd0;
+	u32 info2;
+} __packed;
+
+/* hal_tcl_status_ring
+ *
+ * gse_ctrl
+ *		GSE control operations. This includes cache operations and table
+ *		entry statistics read/clear operation. Values are defined in
+ *		enum %HAL_TCL_GSE_CTRL.
+ *
+ * gse_sel
+ *		To select the ASE/FSE to do the operation mention by GSE_ctrl.
+ *		0: FSE select 1: ASE select
+ *
+ * cache_op_res
+ *		Cache operation result. Values are defined in enum
+ *		%HAL_TCL_CACHE_OP_RES_.
+ *
+ * msdu_cnt
+ * msdu_byte_count
+ *		MSDU count of Entry and MSDU byte count for entry 1.
+ *
+ * hash_indx
+ *		Hash value of the entry in case of search failed or disabled.
+ */
+
+#define HAL_CE_SRC_DESC_ADDR_INFO_ADDR_HI	GENMASK(7, 0)
+#define HAL_CE_SRC_DESC_ADDR_INFO_HASH_EN	BIT(8)
+#define HAL_CE_SRC_DESC_ADDR_INFO_BYTE_SWAP	BIT(9)
+#define HAL_CE_SRC_DESC_ADDR_INFO_DEST_SWAP	BIT(10)
+#define HAL_CE_SRC_DESC_ADDR_INFO_GATHER	BIT(11)
+#define HAL_CE_SRC_DESC_ADDR_INFO_LEN		GENMASK(31, 16)
+
+#define HAL_CE_SRC_DESC_META_INFO_DATA		GENMASK(15, 0)
+
+#define HAL_CE_SRC_DESC_FLAGS_RING_ID		GENMASK(27, 20)
+#define HAL_CE_SRC_DESC_FLAGS_LOOP_CNT		HAL_SRNG_DESC_LOOP_CNT
+
+struct hal_ce_srng_src_desc {
+	u32 buffer_addr_low;
+	u32 buffer_addr_info; /* %HAL_CE_SRC_DESC_ADDR_INFO_ */
+	u32 meta_info; /* %HAL_CE_SRC_DESC_META_INFO_ */
+	u32 flags; /* %HAL_CE_SRC_DESC_FLAGS_ */
+} __packed;
+
+/*
+ * hal_ce_srng_src_desc
+ *
+ * buffer_addr_lo
+ *		LSB 32 bits of the 40 Bit Pointer to the source buffer
+ *
+ * buffer_addr_hi
+ *		MSB 8 bits of the 40 Bit Pointer to the source buffer
+ *
+ * toeplitz_en
+ *		Enable generation of 32-bit Toeplitz-LFSR hash for
+ *		data transfer. In case of gather field in first source
+ *		ring entry of the gather copy cycle in taken into account.
+ *
+ * src_swap
+ *		Treats source memory organization as big-endian. For
+ *		each dword read (4 bytes), the byte 0 is swapped with byte 3
+ *		and byte 1 is swapped with byte 2.
+ *		In case of gather field in first source ring entry of
+ *		the gather copy cycle in taken into account.
+ *
+ * dest_swap
+ *		Treats destination memory organization as big-endian.
+ *		For each dword write (4 bytes), the byte 0 is swapped with
+ *		byte 3 and byte 1 is swapped with byte 2.
+ *		In case of gather field in first source ring entry of
+ *		the gather copy cycle in taken into account.
+ *
+ * gather
+ *		Enables gather of multiple copy engine source
+ *		descriptors to one destination.
+ *
+ * ce_res_0
+ *		Reserved
+ *
+ *
+ * length
+ *		Length of the buffer in units of octets of the current
+ *		descriptor
+ *
+ * fw_metadata
+ *		Meta data used by FW.
+ *		In case of gather field in first source ring entry of
+ *		the gather copy cycle in taken into account.
+ *
+ * ce_res_1
+ *		Reserved
+ *
+ * ce_res_2
+ *		Reserved
+ *
+ * ring_id
+ *		The buffer pointer ring ID.
+ *		0 refers to the IDLE ring
+ *		1 - N refers to other rings
+ *		Helps with debugging when dumping ring contents.
+ *
+ * looping_count
+ *		A count value that indicates the number of times the
+ *		producer of entries into the Ring has looped around the
+ *		ring.
+ *
+ *		At initialization time, this value is set to 0. On the
+ *		first loop, this value is set to 1. After the max value is
+ *		reached allowed by the number of bits for this field, the
+ *		count value continues with 0 again.
+ *
+ *		In case SW is the consumer of the ring entries, it can
+ *		use this field to figure out up to where the producer of
+ *		entries has created new entries. This eliminates the need to
+ *		check where the head pointer' of the ring is located once
+ *		the SW starts processing an interrupt indicating that new
+ *		entries have been put into this ring...
+ *
+ *		Also note that SW if it wants only needs to look at the
+ *		LSB bit of this count value.
+ */
+
+#define HAL_CE_DEST_DESC_ADDR_INFO_ADDR_HI		GENMASK(7, 0)
+#define HAL_CE_DEST_DESC_ADDR_INFO_RING_ID		GENMASK(27, 20)
+#define HAL_CE_DEST_DESC_ADDR_INFO_LOOP_CNT		HAL_SRNG_DESC_LOOP_CNT
+
+struct hal_ce_srng_dest_desc {
+	u32 buffer_addr_low;
+	u32 buffer_addr_info; /* %HAL_CE_DEST_DESC_ADDR_INFO_ */
+} __packed;
+
+/* hal_ce_srng_dest_desc
+ *
+ * dst_buffer_low
+ *		LSB 32 bits of the 40 Bit Pointer to the Destination
+ *		buffer
+ *
+ * dst_buffer_high
+ *		MSB 8 bits of the 40 Bit Pointer to the Destination
+ *		buffer
+ *
+ * ce_res_4
+ *		Reserved
+ *
+ * ring_id
+ *		The buffer pointer ring ID.
+ *		0 refers to the IDLE ring
+ *		1 - N refers to other rings
+ *		Helps with debugging when dumping ring contents.
+ *
+ * looping_count
+ *		A count value that indicates the number of times the
+ *		producer of entries into the Ring has looped around the
+ *		ring.
+ *
+ *		At initialization time, this value is set to 0. On the
+ *		first loop, this value is set to 1. After the max value is
+ *		reached allowed by the number of bits for this field, the
+ *		count value continues with 0 again.
+ *
+ *		In case SW is the consumer of the ring entries, it can
+ *		use this field to figure out up to where the producer of
+ *		entries has created new entries. This eliminates the need to
+ *		check where the head pointer' of the ring is located once
+ *		the SW starts processing an interrupt indicating that new
+ *		entries have been put into this ring...
+ *
+ *		Also note that SW if it wants only needs to look at the
+ *		LSB bit of this count value.
+ */
+
+#define HAL_CE_DST_STATUS_DESC_FLAGS_HASH_EN		BIT(8)
+#define HAL_CE_DST_STATUS_DESC_FLAGS_BYTE_SWAP		BIT(9)
+#define HAL_CE_DST_STATUS_DESC_FLAGS_DEST_SWAP		BIT(10)
+#define HAL_CE_DST_STATUS_DESC_FLAGS_GATHER		BIT(11)
+#define HAL_CE_DST_STATUS_DESC_FLAGS_LEN		GENMASK(31, 16)
+
+#define HAL_CE_DST_STATUS_DESC_META_INFO_DATA		GENMASK(7, 0)
+#define HAL_CE_DST_STATUS_DESC_META_INFO_RING_ID	GENMASK(27, 20)
+#define HAL_CE_DST_STATUS_DESC_META_INFO_LOOP_CNT	HAL_SRNG_DESC_LOOP_CNT
+
+struct hal_ce_srng_dst_status_desc {
+	u32 flags; /* %HAL_CE_DST_STATUS_DESC_FLAGS_ */
+	u32 toeplitz_hash0;
+	u32 toeplitz_hash1;
+	u32 meta_info; /* HAL_CE_DST_STATUS_DESC_META_INFO_ */
+} __packed;
+
+/* hal_ce_srng_dst_status_desc
+ *
+ * ce_res_5
+ *		Reserved
+ *
+ * toeplitz_en
+ *
+ * src_swap
+ *		Source memory buffer swapped
+ *
+ * dest_swap
+ *		Destination  memory buffer swapped
+ *
+ * gather
+ *		Gather of multiple copy engine source descriptors to one
+ *		destination enabled
+ *
+ * ce_res_6
+ *		Reserved
+ *
+ * length
+ *		Sum of all the Lengths of the source descriptor in the
+ *		gather chain
+ *
+ * toeplitz_hash_0
+ *		32 LS bits of 64 bit Toeplitz LFSR hash result
+ *
+ * toeplitz_hash_1
+ *		32 MS bits of 64 bit Toeplitz LFSR hash result
+ *
+ * fw_metadata
+ *		Meta data used by FW
+ *		In case of gather field in first source ring entry of
+ *		the gather copy cycle in taken into account.
+ *
+ * ce_res_7
+ *		Reserved
+ *
+ * ring_id
+ *		The buffer pointer ring ID.
+ *		0 refers to the IDLE ring
+ *		1 - N refers to other rings
+ *		Helps with debugging when dumping ring contents.
+ *
+ * looping_count
+ *		A count value that indicates the number of times the
+ *		producer of entries into the Ring has looped around the
+ *		ring.
+ *
+ *		At initialization time, this value is set to 0. On the
+ *		first loop, this value is set to 1. After the max value is
+ *		reached allowed by the number of bits for this field, the
+ *		count value continues with 0 again.
+ *
+ *		In case SW is the consumer of the ring entries, it can
+ *		use this field to figure out up to where the producer of
+ *		entries has created new entries. This eliminates the need to
+ *		check where the head pointer' of the ring is located once
+ *		the SW starts processing an interrupt indicating that new
+ *		entries have been put into this ring...
+ *
+ *		Also note that SW if it wants only needs to look at the
+ *			LSB bit of this count value.
+ */
+
+#define HAL_TX_RATE_STATS_INFO0_VALID		BIT(0)
+#define HAL_TX_RATE_STATS_INFO0_BW		GENMASK(2, 1)
+#define HAL_TX_RATE_STATS_INFO0_PKT_TYPE	GENMASK(6, 3)
+#define HAL_TX_RATE_STATS_INFO0_STBC		BIT(7)
+#define HAL_TX_RATE_STATS_INFO0_LDPC		BIT(8)
+#define HAL_TX_RATE_STATS_INFO0_SGI		GENMASK(10, 9)
+#define HAL_TX_RATE_STATS_INFO0_MCS		GENMASK(14, 11)
+#define HAL_TX_RATE_STATS_INFO0_OFDMA_TX	BIT(15)
+#define HAL_TX_RATE_STATS_INFO0_TONES_IN_RU	GENMASK(27, 16)
+
+enum hal_tx_rate_stats_bw {
+	HAL_TX_RATE_STATS_BW_20,
+	HAL_TX_RATE_STATS_BW_40,
+	HAL_TX_RATE_STATS_BW_80,
+	HAL_TX_RATE_STATS_BW_160,
+};
+
+enum hal_tx_rate_stats_pkt_type {
+	HAL_TX_RATE_STATS_PKT_TYPE_11A,
+	HAL_TX_RATE_STATS_PKT_TYPE_11B,
+	HAL_TX_RATE_STATS_PKT_TYPE_11N,
+	HAL_TX_RATE_STATS_PKT_TYPE_11AC,
+	HAL_TX_RATE_STATS_PKT_TYPE_11AX,
+};
+
+enum hal_tx_rate_stats_sgi {
+	HAL_TX_RATE_STATS_SGI_08US,
+	HAL_TX_RATE_STATS_SGI_04US,
+	HAL_TX_RATE_STATS_SGI_16US,
+	HAL_TX_RATE_STATS_SGI_32US,
+};
+
+struct hal_tx_rate_stats {
+	u32 info0;
+	u32 tsf;
+} __packed;
+
+struct hal_wbm_link_desc {
+	struct ath11k_buffer_addr buf_addr_info;
+} __packed;
+
+/* hal_wbm_link_desc
+ *
+ *	Producer: WBM
+ *	Consumer: WBM
+ *
+ * buf_addr_info
+ *		Details of the physical address of a buffer or MSDU
+ *		link descriptor.
+ */
+
+enum hal_wbm_rel_src_module {
+	HAL_WBM_REL_SRC_MODULE_TQM,
+	HAL_WBM_REL_SRC_MODULE_RXDMA,
+	HAL_WBM_REL_SRC_MODULE_REO,
+	HAL_WBM_REL_SRC_MODULE_FW,
+	HAL_WBM_REL_SRC_MODULE_SW,
+};
+
+enum hal_wbm_rel_desc_type {
+	HAL_WBM_REL_DESC_TYPE_REL_MSDU,
+	HAL_WBM_REL_DESC_TYPE_MSDU_LINK,
+	HAL_WBM_REL_DESC_TYPE_MPDU_LINK,
+	HAL_WBM_REL_DESC_TYPE_MSDU_EXT,
+	HAL_WBM_REL_DESC_TYPE_QUEUE_EXT,
+};
+
+/* hal_wbm_rel_desc_type
+ *
+ * msdu_buffer
+ *	The address points to an MSDU buffer
+ *
+ * msdu_link_descriptor
+ *	The address points to an Tx MSDU link descriptor
+ *
+ * mpdu_link_descriptor
+ *	The address points to an MPDU link descriptor
+ *
+ * msdu_ext_descriptor
+ *	The address points to an MSDU extension descriptor
+ *
+ * queue_ext_descriptor
+ *	The address points to an TQM queue extension descriptor. WBM should
+ *	treat this is the same way as a link descriptor.
+ */
+
+enum hal_wbm_rel_bm_act {
+	HAL_WBM_REL_BM_ACT_PUT_IN_IDLE,
+	HAL_WBM_REL_BM_ACT_REL_MSDU,
+};
+
+/* hal_wbm_rel_bm_act
+ *
+ * put_in_idle_list
+ *	Put the buffer or descriptor back in the idle list. In case of MSDU or
+ *	MDPU link descriptor, BM does not need to check to release any
+ *	individual MSDU buffers.
+ *
+ * release_msdu_list
+ *	This BM action can only be used in combination with desc_type being
+ *	msdu_link_descriptor. Field first_msdu_index points out which MSDU
+ *	pointer in the MSDU link descriptor is the first of an MPDU that is
+ *	released. BM shall release all the MSDU buffers linked to this first
+ *	MSDU buffer pointer. All related MSDU buffer pointer entries shall be
+ *	set to value 0, which represents the 'NULL' pointer. When all MSDU
+ *	buffer pointers in the MSDU link descriptor are 'NULL', the MSDU link
+ *	descriptor itself shall also be released.
+ */
+
+#define HAL_WBM_RELEASE_INFO0_REL_SRC_MODULE		GENMASK(2, 0)
+#define HAL_WBM_RELEASE_INFO0_BM_ACTION			GENMASK(5, 3)
+#define HAL_WBM_RELEASE_INFO0_DESC_TYPE			GENMASK(8, 6)
+#define HAL_WBM_RELEASE_INFO0_FIRST_MSDU_IDX		GENMASK(12, 9)
+#define HAL_WBM_RELEASE_INFO0_TQM_RELEASE_REASON	GENMASK(16, 13)
+#define HAL_WBM_RELEASE_INFO0_RXDMA_PUSH_REASON		GENMASK(18, 17)
+#define HAL_WBM_RELEASE_INFO0_RXDMA_ERROR_CODE		GENMASK(23, 19)
+#define HAL_WBM_RELEASE_INFO0_REO_PUSH_REASON		GENMASK(25, 24)
+#define HAL_WBM_RELEASE_INFO0_REO_ERROR_CODE		GENMASK(30, 26)
+#define HAL_WBM_RELEASE_INFO0_WBM_INTERNAL_ERROR	BIT(31)
+
+#define HAL_WBM_RELEASE_INFO1_TQM_STATUS_NUMBER		GENMASK(23, 0)
+#define HAL_WBM_RELEASE_INFO1_TRANSMIT_COUNT		GENMASK(30, 24)
+
+#define HAL_WBM_RELEASE_INFO2_ACK_FRAME_RSSI		GENMASK(7, 0)
+#define HAL_WBM_RELEASE_INFO2_SW_REL_DETAILS_VALID	BIT(8)
+#define HAL_WBM_RELEASE_INFO2_FIRST_MSDU		BIT(9)
+#define HAL_WBM_RELEASE_INFO2_LAST_MSDU			BIT(10)
+#define HAL_WBM_RELEASE_INFO2_MSDU_IN_AMSDU		BIT(11)
+#define HAL_WBM_RELEASE_INFO2_FW_TX_NOTIF_FRAME		BIT(12)
+#define HAL_WBM_RELEASE_INFO2_BUFFER_TIMESTAMP		GENMASK(31, 13)
+
+#define HAL_WBM_RELEASE_INFO3_PEER_ID			GENMASK(15, 0)
+#define HAL_WBM_RELEASE_INFO3_TID			GENMASK(19, 16)
+#define HAL_WBM_RELEASE_INFO3_RING_ID			GENMASK(27, 20)
+#define HAL_WBM_RELEASE_INFO3_LOOPING_COUNT		GENMASK(31, 28)
+
+#define HAL_WBM_REL_HTT_TX_COMP_INFO0_STATUS		GENMASK(12, 9)
+#define HAL_WBM_REL_HTT_TX_COMP_INFO0_REINJ_REASON	GENMASK(16, 13)
+#define HAL_WBM_REL_HTT_TX_COMP_INFO0_EXP_FRAME		BIT(17)
+
+struct hal_wbm_release_ring {
+	struct ath11k_buffer_addr buf_addr_info;
+	u32 info0;
+	u32 info1;
+	u32 info2;
+	struct hal_tx_rate_stats rate_stats;
+	u32 info3;
+} __packed;
+
+/* hal_wbm_release_ring
+ *
+ *	Producer: SW/TQM/RXDMA/REO/SWITCH
+ *	Consumer: WBM/SW/FW
+ *
+ * HTT tx status is overlayed on wbm_release ring on 4-byte words 2, 3, 4 and 5
+ * for software based completions.
+ *
+ * buf_addr_info
+ *	Details of the physical address of the buffer or link descriptor.
+ *
+ * release_source_module
+ *	Indicates which module initiated the release of this buffer/descriptor.
+ *	Values are defined in enum %HAL_WBM_REL_SRC_MODULE_.
+ *
+ * bm_action
+ *	Field only valid when the field return_buffer_manager in
+ *	Released_buff_or_desc_addr_info indicates:
+ *		WBM_IDLE_BUF_LIST / WBM_IDLE_DESC_LIST
+ *	Values are defined in enum %HAL_WBM_REL_BM_ACT_.
+ *
+ * buffer_or_desc_type
+ *	Field only valid when WBM is marked as the return_buffer_manager in
+ *	the Released_Buffer_address_info. Indicates that type of buffer or
+ *	descriptor is being released. Values are in enum %HAL_WBM_REL_DESC_TYPE.
+ *
+ * first_msdu_index
+ *	Field only valid for the bm_action release_msdu_list. The index of the
+ *	first MSDU in an MSDU link descriptor all belonging to the same MPDU.
+ *
+ * tqm_release_reason
+ *	Field only valid when Release_source_module is set to release_source_TQM
+ *	Release reasons are defined in enum %HAL_WBM_TQM_REL_REASON_.
+ *
+ * rxdma_push_reason
+ * reo_push_reason
+ *	Indicates why rxdma/reo pushed the frame to this ring and values are
+ *	defined in enum %HAL_REO_DEST_RING_PUSH_REASON_.
+ *
+ * rxdma_error_code
+ *	Field only valid when 'rxdma_push_reason' set to 'error_detected'.
+ *	Values are defined in enum %HAL_REO_ENTR_RING_RXDMA_ECODE_.
+ *
+ * reo_error_code
+ *	Field only valid when 'reo_push_reason' set to 'error_detected'. Values
+ *	are defined in enum %HAL_REO_DEST_RING_ERROR_CODE_.
+ *
+ * wbm_internal_error
+ *	Is set when WBM got a buffer pointer but the action was to push it to
+ *	the idle link descriptor ring or do link related activity OR
+ *	Is set when WBM got a link buffer pointer but the action was to push it
+ *	to the buffer descriptor ring.
+ *
+ * tqm_status_number
+ *	The value in this field is equal to tqm_cmd_number in TQM command. It is
+ *	used to correlate the statu with TQM commands. Only valid when
+ *	release_source_module is TQM.
+ *
+ * transmit_count
+ *	The number of times the frame has been transmitted, valid only when
+ *	release source in TQM.
+ *
+ * ack_frame_rssi
+ *	This field is only valid when the source is TQM. If this frame is
+ *	removed as the result of the reception of an ACK or BA, this field
+ *	indicates the RSSI of the received ACK or BA frame.
+ *
+ * sw_release_details_valid
+ *	This is set when WMB got a 'release_msdu_list' command from TQM and
+ *	return buffer manager is not WMB. WBM will then de-aggregate all MSDUs
+ *	and pass them one at a time on to the 'buffer owner'.
+ *
+ * first_msdu
+ *	Field only valid when SW_release_details_valid is set.
+ *	When set, this MSDU is the first MSDU pointed to in the
+ *	'release_msdu_list' command.
+ *
+ * last_msdu
+ *	Field only valid when SW_release_details_valid is set.
+ *	When set, this MSDU is the last MSDU pointed to in the
+ *	'release_msdu_list' command.
+ *
+ * msdu_part_of_amsdu
+ *	Field only valid when SW_release_details_valid is set.
+ *	When set, this MSDU was part of an A-MSDU in MPDU
+ *
+ * fw_tx_notify_frame
+ *	Field only valid when SW_release_details_valid is set.
+ *
+ * buffer_timestamp
+ *	Field only valid when SW_release_details_valid is set.
+ *	This is the Buffer_timestamp field from the
+ *	Timestamp in units of 1024 us
+ *
+ * struct hal_tx_rate_stats rate_stats
+ *	Details for command execution tracking purposes.
+ *
+ * sw_peer_id
+ * tid
+ *	Field only valid when Release_source_module is set to
+ *	release_source_TQM
+ *
+ *	1) Release of msdu buffer due to drop_frame = 1. Flow is
+ *	not fetched and hence sw_peer_id and tid = 0
+ *
+ *	buffer_or_desc_type = e_num 0
+ *	MSDU_rel_buffertqm_release_reason = e_num 1
+ *	tqm_rr_rem_cmd_rem
+ *
+ *	2) Release of msdu buffer due to Flow is not fetched and
+ *	hence sw_peer_id and tid = 0
+ *
+ *	buffer_or_desc_type = e_num 0
+ *	MSDU_rel_buffertqm_release_reason = e_num 1
+ *	tqm_rr_rem_cmd_rem
+ *
+ *	3) Release of msdu link due to remove_mpdu or acked_mpdu
+ *	command.
+ *
+ *	buffer_or_desc_type = e_num1
+ *	msdu_link_descriptortqm_release_reason can be:e_num 1
+ *	tqm_rr_rem_cmd_reme_num 2 tqm_rr_rem_cmd_tx
+ *	e_num 3 tqm_rr_rem_cmd_notxe_num 4 tqm_rr_rem_cmd_aged
+ *
+ *	This field represents the TID from the TX_MSDU_FLOW
+ *	descriptor or TX_MPDU_QUEUE descriptor
+ *
+ * rind_id
+ *	For debugging.
+ *	This field is filled in by the SRNG module.
+ *	It help to identify the ring that is being looked
+ *
+ * looping_count
+ *	A count value that indicates the number of times the
+ *	producer of entries into the Buffer Manager Ring has looped
+ *	around the ring.
+ *
+ *	At initialization time, this value is set to 0. On the
+ *	first loop, this value is set to 1. After the max value is
+ *	reached allowed by the number of bits for this field, the
+ *	count value continues with 0 again.
+ *
+ *	In case SW is the consumer of the ring entries, it can
+ *	use this field to figure out up to where the producer of
+ *	entries has created new entries. This eliminates the need to
+ *	check where the head pointer' of the ring is located once
+ *	the SW starts processing an interrupt indicating that new
+ *	entries have been put into this ring...
+ *
+ *	Also note that SW if it wants only needs to look at the
+ *	LSB bit of this count value.
+ */
+
+/**
+ * enum hal_wbm_tqm_rel_reason - TQM release reason code
+ * @HAL_WBM_TQM_REL_REASON_FRAME_ACKED: ACK or BACK received for the frame
+ * @HAL_WBM_TQM_REL_REASON_CMD_REMOVE_MPDU: Command remove_mpdus initiated by SW
+ * @HAL_WBM_TQM_REL_REASON_CMD_REMOVE_TX: Command remove transmitted_mpdus
+ *	initiated by sw.
+ * @HAL_WBM_TQM_REL_REASON_CMD_REMOVE_NOTX: Command remove untransmitted_mpdus
+ *	initiated by sw.
+ * @HAL_WBM_TQM_REL_REASON_CMD_REMOVE_AGED_FRAMES: Command remove aged msdus or
+ *	mpdus.
+ * @HAL_WBM_TQM_REL_REASON_CMD_REMOVE_RESEAON1: Remove command initiated by
+ *	fw with fw_reason1.
+ * @HAL_WBM_TQM_REL_REASON_CMD_REMOVE_RESEAON2: Remove command initiated by
+ *	fw with fw_reason2.
+ * @HAL_WBM_TQM_REL_REASON_CMD_REMOVE_RESEAON3: Remove command initiated by
+ *	fw with fw_reason3.
+ */
+enum hal_wbm_tqm_rel_reason {
+	HAL_WBM_TQM_REL_REASON_FRAME_ACKED,
+	HAL_WBM_TQM_REL_REASON_CMD_REMOVE_MPDU,
+	HAL_WBM_TQM_REL_REASON_CMD_REMOVE_TX,
+	HAL_WBM_TQM_REL_REASON_CMD_REMOVE_NOTX,
+	HAL_WBM_TQM_REL_REASON_CMD_REMOVE_AGED_FRAMES,
+	HAL_WBM_TQM_REL_REASON_CMD_REMOVE_RESEAON1,
+	HAL_WBM_TQM_REL_REASON_CMD_REMOVE_RESEAON2,
+	HAL_WBM_TQM_REL_REASON_CMD_REMOVE_RESEAON3,
+};
+
+struct hal_wbm_buffer_ring {
+	struct ath11k_buffer_addr buf_addr_info;
+};
+
+enum hal_desc_owner {
+	HAL_DESC_OWNER_WBM,
+	HAL_DESC_OWNER_SW,
+	HAL_DESC_OWNER_TQM,
+	HAL_DESC_OWNER_RXDMA,
+	HAL_DESC_OWNER_REO,
+	HAL_DESC_OWNER_SWITCH,
+};
+
+enum hal_desc_buf_type {
+	HAL_DESC_BUF_TYPE_TX_MSDU_LINK,
+	HAL_DESC_BUF_TYPE_TX_MPDU_LINK,
+	HAL_DESC_BUF_TYPE_TX_MPDU_QUEUE_HEAD,
+	HAL_DESC_BUF_TYPE_TX_MPDU_QUEUE_EXT,
+	HAL_DESC_BUF_TYPE_TX_FLOW,
+	HAL_DESC_BUF_TYPE_TX_BUFFER,
+	HAL_DESC_BUF_TYPE_RX_MSDU_LINK,
+	HAL_DESC_BUF_TYPE_RX_MPDU_LINK,
+	HAL_DESC_BUF_TYPE_RX_REO_QUEUE,
+	HAL_DESC_BUF_TYPE_RX_REO_QUEUE_EXT,
+	HAL_DESC_BUF_TYPE_RX_BUFFER,
+	HAL_DESC_BUF_TYPE_IDLE_LINK,
+};
+
+#define HAL_DESC_REO_OWNED		4
+#define HAL_DESC_REO_QUEUE_DESC		8
+#define HAL_DESC_REO_QUEUE_EXT_DESC	9
+#define HAL_DESC_REO_NON_QOS_TID	16
+
+#define HAL_DESC_HDR_INFO0_OWNER	GENMASK(3, 0)
+#define HAL_DESC_HDR_INFO0_BUF_TYPE	GENMASK(7, 4)
+#define HAL_DESC_HDR_INFO0_DBG_RESERVED	GENMASK(31, 8)
+
+struct hal_desc_header {
+	u32 info0;
+} __packed;
+
+struct hal_rx_mpdu_link_ptr {
+	struct ath11k_buffer_addr addr_info;
+} __packed;
+
+struct hal_rx_msdu_details {
+	struct ath11k_buffer_addr buf_addr_info;
+	struct rx_msdu_desc rx_msdu_info;
+} __packed;
+
+#define HAL_RX_MSDU_LNK_INFO0_RX_QUEUE_NUMBER		GENMASK(15, 0)
+#define HAL_RX_MSDU_LNK_INFO0_FIRST_MSDU_LNK		BIT(16)
+
+struct hal_rx_msdu_link {
+	struct hal_desc_header desc_hdr;
+	struct ath11k_buffer_addr buf_addr_info;
+	u32 info0;
+	u32 pn[4];
+	struct hal_rx_msdu_details msdu_link[6];
+} __packed;
+
+struct hal_rx_reo_queue_ext {
+	struct hal_desc_header desc_hdr;
+	u32 rsvd;
+	struct hal_rx_mpdu_link_ptr mpdu_link[15];
+} __packed;
+
+/* hal_rx_reo_queue_ext
+ *	Consumer: REO
+ *	Producer: REO
+ *
+ * descriptor_header
+ *	Details about which module owns this struct.
+ *
+ * mpdu_link
+ *	Pointer to the next MPDU_link descriptor in the MPDU queue.
+ */
+
+enum hal_rx_reo_queue_pn_size {
+	HAL_RX_REO_QUEUE_PN_SIZE_24,
+	HAL_RX_REO_QUEUE_PN_SIZE_48,
+	HAL_RX_REO_QUEUE_PN_SIZE_128,
+};
+
+#define HAL_RX_REO_QUEUE_RX_QUEUE_NUMBER		GENMASK(15, 0)
+
+#define HAL_RX_REO_QUEUE_INFO0_VLD			BIT(0)
+#define HAL_RX_REO_QUEUE_INFO0_ASSOC_LNK_DESC_COUNTER	GENMASK(2, 1)
+#define HAL_RX_REO_QUEUE_INFO0_DIS_DUP_DETECTION	BIT(3)
+#define HAL_RX_REO_QUEUE_INFO0_SOFT_REORDER_EN		BIT(4)
+#define HAL_RX_REO_QUEUE_INFO0_AC			GENMASK(6, 5)
+#define HAL_RX_REO_QUEUE_INFO0_BAR			BIT(7)
+#define HAL_RX_REO_QUEUE_INFO0_RETRY			BIT(8)
+#define HAL_RX_REO_QUEUE_INFO0_CHECK_2K_MODE		BIT(9)
+#define HAL_RX_REO_QUEUE_INFO0_OOR_MODE			BIT(10)
+#define HAL_RX_REO_QUEUE_INFO0_BA_WINDOW_SIZE		GENMASK(18, 11)
+#define HAL_RX_REO_QUEUE_INFO0_PN_CHECK			BIT(19)
+#define HAL_RX_REO_QUEUE_INFO0_EVEN_PN			BIT(20)
+#define HAL_RX_REO_QUEUE_INFO0_UNEVEN_PN		BIT(21)
+#define HAL_RX_REO_QUEUE_INFO0_PN_HANDLE_ENABLE		BIT(22)
+#define HAL_RX_REO_QUEUE_INFO0_PN_SIZE			GENMASK(24, 23)
+#define HAL_RX_REO_QUEUE_INFO0_IGNORE_AMPDU_FLG		BIT(25)
+
+#define HAL_RX_REO_QUEUE_INFO1_SVLD			BIT(0)
+#define HAL_RX_REO_QUEUE_INFO1_SSN			GENMASK(12, 1)
+#define HAL_RX_REO_QUEUE_INFO1_CURRENT_IDX		GENMASK(20, 13)
+#define HAL_RX_REO_QUEUE_INFO1_SEQ_2K_ERR		BIT(21)
+#define HAL_RX_REO_QUEUE_INFO1_PN_ERR			BIT(22)
+#define HAL_RX_REO_QUEUE_INFO1_PN_VALID			BIT(31)
+
+#define HAL_RX_REO_QUEUE_INFO2_MPDU_COUNT		GENMASK(6, 0)
+#define HAL_RX_REO_QUEUE_INFO2_MSDU_COUNT		(31, 7)
+
+#define HAL_RX_REO_QUEUE_INFO3_TIMEOUT_COUNT		GENMASK(9, 4)
+#define HAL_RX_REO_QUEUE_INFO3_FWD_DUE_TO_BAR_CNT	GENMASK(15, 10)
+#define HAL_RX_REO_QUEUE_INFO3_DUPLICATE_COUNT		GENMASK(31, 10)
+
+#define HAL_RX_REO_QUEUE_INFO4_FRAME_IN_ORD_COUNT	GENMASK(23, 0)
+#define HAL_RX_REO_QUEUE_INFO4_BAR_RECVD_COUNT		GENMASK(31, 24)
+
+#define HAL_RX_REO_QUEUE_INFO5_LATE_RX_MPDU_COUNT	GENMASK(11, 0)
+#define HAL_RX_REO_QUEUE_INFO5_WINDOW_JUMP_2K		GENMASK(15, 12)
+#define HAL_RX_REO_QUEUE_INFO5_HOLE_COUNT		GENMASK(31, 16)
+
+struct hal_rx_reo_queue {
+	struct hal_desc_header desc_hdr;
+	u32 rx_queue_num;
+	u32 info0;
+	u32 info1;
+	u32 pn[4];
+	u32 last_rx_enqueue_timestamp;
+	u32 last_rx_dequeue_timestamp;
+	u32 next_aging_queue[2];
+	u32 prev_aging_queue[2];
+	u32 rx_bitmap[8];
+	u32 info2;
+	u32 info3;
+	u32 info4;
+	u32 processed_mpdus;
+	u32 processed_msdus;
+	u32 processed_total_bytes;
+	u32 info5;
+	u32 rsvd[3];
+	struct hal_rx_reo_queue_ext ext_desc[0];
+} __packed;
+
+/* hal_rx_reo_queue
+ *
+ * descriptor_header
+ *	Details about which module owns this struct. Note that sub field
+ *	Buffer_type shall be set to receive_reo_queue_descriptor.
+ *
+ * receive_queue_number
+ *	Indicates the MPDU queue ID to which this MPDU link descriptor belongs.
+ *
+ * vld
+ *	Valid bit indicating a session is established and the queue descriptor
+ *	is valid.
+ * associated_link_descriptor_counter
+ *	Indicates which of the 3 link descriptor counters shall be incremented
+ *	or decremented when link descriptors are added or removed from this
+ *	flow queue.
+ * disable_duplicate_detection
+ *	When set, do not perform any duplicate detection.
+ * soft_reorder_enable
+ *	When set, REO has been instructed to not perform the actual re-ordering
+ *	of frames for this queue, but just to insert the reorder opcodes.
+ * ac
+ *	Indicates the access category of the queue descriptor.
+ * bar
+ *	Indicates if BAR has been received.
+ * retry
+ *	Retry bit is checked if this bit is set.
+ * chk_2k_mode
+ *	Indicates what type of operation is expected from Reo when the received
+ *	frame SN falls within the 2K window.
+ * oor_mode
+ *	Indicates what type of operation is expected when the received frame
+ *	falls within the OOR window.
+ * ba_window_size
+ *	Indicates the negotiated (window size + 1). Max of 256 bits.
+ *
+ *	A value 255 means 256 bitmap, 63 means 64 bitmap, 0 (means non-BA
+ *	session, with window size of 0). The 3 values here are the main values
+ *	validated, but other values should work as well.
+ *
+ *	A BA window size of 0 (=> one frame entry bitmat), means that there is
+ *	no additional rx_reo_queue_ext desc. following rx_reo_queue in memory.
+ *	A BA window size of 1 - 105, means that there is 1 rx_reo_queue_ext.
+ *	A BA window size of 106 - 210, means that there are 2 rx_reo_queue_ext.
+ *	A BA window size of 211 - 256, means that there are 3 rx_reo_queue_ext.
+ * pn_check_needed, pn_shall_be_even, pn_shall_be_uneven, pn_handling_enable,
+ * pn_size
+ *	REO shall perform the PN increment check, even number check, uneven
+ *	number check, PN error check and size of the PN field check.
+ * ignore_ampdu_flag
+ *	REO shall ignore the ampdu_flag on entrance descriptor for this queue.
+ *
+ * svld
+ *	Sequence number in next field is valid one.
+ * ssn
+ *	 Starting Sequence number of the session.
+ * current_index
+ *	Points to last forwarded packet
+ * seq_2k_error_detected_flag
+ *	REO has detected a 2k error jump in the sequence number and from that
+ *	moment forward, all new frames are forwarded directly to FW, without
+ *	duplicate detect, reordering, etc.
+ * pn_error_detected_flag
+ *	REO has detected a PN error.
+ */
+
+#define HAL_REO_UPD_RX_QUEUE_INFO0_QUEUE_ADDR_HI		GENMASK(7, 0)
+#define HAL_REO_UPD_RX_QUEUE_INFO0_UPD_RX_QUEUE_NUM		BIT(8)
+#define HAL_REO_UPD_RX_QUEUE_INFO0_UPD_VLD			BIT(9)
+#define HAL_REO_UPD_RX_QUEUE_INFO0_UPD_ASSOC_LNK_DESC_CNT	BIT(10)
+#define HAL_REO_UPD_RX_QUEUE_INFO0_UPD_DIS_DUP_DETECTION	BIT(11)
+#define HAL_REO_UPD_RX_QUEUE_INFO0_UPD_SOFT_REORDER_EN		BIT(12)
+#define HAL_REO_UPD_RX_QUEUE_INFO0_UPD_AC			BIT(13)
+#define HAL_REO_UPD_RX_QUEUE_INFO0_UPD_BAR			BIT(14)
+#define HAL_REO_UPD_RX_QUEUE_INFO0_UPD_RETRY			BIT(15)
+#define HAL_REO_UPD_RX_QUEUE_INFO0_UPD_CHECK_2K_MODE		BIT(16)
+#define HAL_REO_UPD_RX_QUEUE_INFO0_UPD_OOR_MODE			BIT(17)
+#define HAL_REO_UPD_RX_QUEUE_INFO0_UPD_BA_WINDOW_SIZE		BIT(18)
+#define HAL_REO_UPD_RX_QUEUE_INFO0_UPD_PN_CHECK			BIT(19)
+#define HAL_REO_UPD_RX_QUEUE_INFO0_UPD_EVEN_PN			BIT(20)
+#define HAL_REO_UPD_RX_QUEUE_INFO0_UPD_UNEVEN_PN		BIT(21)
+#define HAL_REO_UPD_RX_QUEUE_INFO0_UPD_PN_HANDLE_ENABLE		BIT(22)
+#define HAL_REO_UPD_RX_QUEUE_INFO0_UPD_PN_SIZE			BIT(23)
+#define HAL_REO_UPD_RX_QUEUE_INFO0_UPD_IGNORE_AMPDU_FLG		BIT(24)
+#define HAL_REO_UPD_RX_QUEUE_INFO0_UPD_SVLD			BIT(25)
+#define HAL_REO_UPD_RX_QUEUE_INFO0_UPD_SSN			BIT(26)
+#define HAL_REO_UPD_RX_QUEUE_INFO0_UPD_SEQ_2K_ERR		BIT(27)
+#define HAL_REO_UPD_RX_QUEUE_INFO0_UPD_PN_ERR			BIT(28)
+#define HAL_REO_UPD_RX_QUEUE_INFO0_UPD_PN_VALID			BIT(29)
+#define HAL_REO_UPD_RX_QUEUE_INFO0_UPD_PN			BIT(30)
+
+#define HAL_REO_UPD_RX_QUEUE_INFO1_RX_QUEUE_NUMBER		GENMASK(15, 0)
+#define HAL_REO_UPD_RX_QUEUE_INFO1_VLD				BIT(16)
+#define HAL_REO_UPD_RX_QUEUE_INFO1_ASSOC_LNK_DESC_COUNTER	GENMASK(18, 17)
+#define HAL_REO_UPD_RX_QUEUE_INFO1_DIS_DUP_DETECTION		BIT(19)
+#define HAL_REO_UPD_RX_QUEUE_INFO1_SOFT_REORDER_EN		BIT(20)
+#define HAL_REO_UPD_RX_QUEUE_INFO1_AC				GENMASK(22, 21)
+#define HAL_REO_UPD_RX_QUEUE_INFO1_BAR				BIT(23)
+#define HAL_REO_UPD_RX_QUEUE_INFO1_RETRY			BIT(24)
+#define HAL_REO_UPD_RX_QUEUE_INFO1_CHECK_2K_MODE		BIT(25)
+#define HAL_REO_UPD_RX_QUEUE_INFO1_OOR_MODE			BIT(26)
+#define HAL_REO_UPD_RX_QUEUE_INFO1_PN_CHECK			BIT(27)
+#define HAL_REO_UPD_RX_QUEUE_INFO1_EVEN_PN			BIT(28)
+#define HAL_REO_UPD_RX_QUEUE_INFO1_UNEVEN_PN			BIT(29)
+#define HAL_REO_UPD_RX_QUEUE_INFO1_PN_HANDLE_ENABLE		BIT(30)
+#define HAL_REO_UPD_RX_QUEUE_INFO1_IGNORE_AMPDU_FLG		BIT(31)
+
+#define HAL_REO_UPD_RX_QUEUE_INFO2_BA_WINDOW_SIZE		GENMASK(7, 0)
+#define HAL_REO_UPD_RX_QUEUE_INFO2_PN_SIZE			GENMASK(9, 8)
+#define HAL_REO_UPD_RX_QUEUE_INFO2_SVLD				BIT(10)
+#define HAL_REO_UPD_RX_QUEUE_INFO2_SSN				GENMASK(22, 11)
+#define HAL_REO_UPD_RX_QUEUE_INFO2_SEQ_2K_ERR			BIT(23)
+#define HAL_REO_UPD_RX_QUEUE_INFO2_PN_ERR			BIT(24)
+#define HAL_REO_UPD_RX_QUEUE_INFO2_PN_VALID			BIT(25)
+
+struct hal_reo_update_rx_queue {
+	struct hal_reo_cmd_hdr cmd;
+	u32 queue_addr_lo;
+	u32 info0;
+	u32 info1;
+	u32 info2;
+	u32 pn[4];
+} __packed;
+
+#define HAL_REO_UNBLOCK_CACHE_INFO0_UNBLK_CACHE		BIT(0)
+#define HAL_REO_UNBLOCK_CACHE_INFO0_RESOURCE_IDX	GENMASK(2, 1)
+
+struct hal_reo_unblock_cache {
+	struct hal_reo_cmd_hdr cmd;
+	u32 info0;
+	u32 rsvd[7];
+} __packed;
+
+enum hal_reo_exec_status {
+	HAL_REO_EXEC_STATUS_SUCCESS,
+	HAL_REO_EXEC_STATUS_BLOCKED,
+	HAL_REO_EXEC_STATUS_FAILED,
+	HAL_REO_EXEC_STATUS_RESOURCE_BLOCKED,
+};
+
+#define HAL_REO_STATUS_HDR_INFO0_STATUS_NUM	GENMASK(15, 0)
+#define HAL_REO_STATUS_HDR_INFO0_EXEC_TIME	GENMASK(25, 16)
+#define HAL_REO_STATUS_HDR_INFO0_EXEC_STATUS	GENMASK(27, 26)
+
+struct hal_reo_status_hdr {
+	u32 info0;
+	u32 timestamp;
+} __packed;
+
+/* hal_reo_status_hdr
+ *		Producer: REO
+ *		Consumer: SW
+ *
+ * status_num
+ *		The value in this field is equal to value of the reo command
+ *		number. This field helps to correlate the statuses with the REO
+ *		commands.
+ *
+ * execution_time (in us)
+ *		The amount of time REO took to excecute the command. Note that
+ *		this time does not include the duration of the command waiting
+ *		in the command ring, before the execution started.
+ *
+ * execution_status
+ *		Execution status of the command. Values are defined in
+ *		enum %HAL_REO_EXEC_STATUS_.
+ */
+#define HAL_REO_GET_QUEUE_STATS_STATUS_INFO0_SSN		GENMASK(11, 0)
+#define HAL_REO_GET_QUEUE_STATS_STATUS_INFO0_CUR_IDX		GENMASK(19, 12)
+
+#define HAL_REO_GET_QUEUE_STATS_STATUS_INFO1_MPDU_COUNT		GENMASK(6, 0)
+#define HAL_REO_GET_QUEUE_STATS_STATUS_INFO1_MSDU_COUNT		GENMASK(31, 7)
+
+#define HAL_REO_GET_QUEUE_STATS_STATUS_INFO2_TIMEOUT_COUNT	GENMASK(9, 4)
+#define HAL_REO_GET_QUEUE_STATS_STATUS_INFO2_FDTB_COUNT		GENMASK(15, 10)
+#define HAL_REO_GET_QUEUE_STATS_STATUS_INFO2_DUPLICATE_COUNT	GENMASK(31, 16)
+
+#define HAL_REO_GET_QUEUE_STATS_STATUS_INFO3_FIO_COUNT		GENMASK(23, 0)
+#define HAL_REO_GET_QUEUE_STATS_STATUS_INFO3_BAR_RCVD_CNT	GENMASK(31, 24)
+
+#define HAL_REO_GET_QUEUE_STATS_STATUS_INFO4_LATE_RX_MPDU	GENMASK(11, 0)
+#define HAL_REO_GET_QUEUE_STATS_STATUS_INFO4_WINDOW_JMP2K	GENMASK(15, 12)
+#define HAL_REO_GET_QUEUE_STATS_STATUS_INFO4_HOLE_COUNT		GENMASK(31, 16)
+
+#define HAL_REO_GET_QUEUE_STATS_STATUS_INFO5_LOOPING_CNT	GENMASK(31, 28)
+
+struct hal_reo_get_queue_stats_status {
+	struct hal_reo_status_hdr hdr;
+	u32 info0;
+	u32 pn[4];
+	u32 last_rx_enqueue_timestamp;
+	u32 last_rx_dequeue_timestamp;
+	u32 rx_bitmap[8];
+	u32 info1;
+	u32 info2;
+	u32 info3;
+	u32 num_mpdu_frames;
+	u32 num_msdu_frames;
+	u32 total_bytes;
+	u32 info4;
+	u32 info5;
+} __packed;
+
+/* hal_reo_get_queue_stats_status
+ *		Producer: REO
+ *		Consumer: SW
+ *
+ * status_hdr
+ *		Details that can link this status with the original command. It
+ *		also contains info on how long REO took to execute this command.
+ *
+ * ssn
+ *		Starting Sequence number of the session, this changes whenever
+ *		window moves (can be filled by SW then maintained by REO).
+ *
+ * current_index
+ *		Points to last forwarded packet.
+ *
+ * pn
+ *		Bits of the PN number.
+ *
+ * last_rx_enqueue_timestamp
+ * last_rx_dequeue_timestamp
+ *		Timestamp of arrival of the last MPDU for this queue and
+ *		Timestamp of forwarding an MPDU accordingly.
+ *
+ * rx_bitmap
+ *		When a bit is set, the corresponding frame is currently held
+ *		in the re-order queue. The bitmap  is Fully managed by HW.
+ *
+ * current_mpdu_count
+ * current_msdu_count
+ *		The number of MPDUs and MSDUs in the queue.
+ *
+ * timeout_count
+ *		The number of times REO started forwarding frames even though
+ *		there is a hole in the bitmap. Forwarding reason is timeout.
+ *
+ * forward_due_to_bar_count
+ *		The number of times REO started forwarding frames even though
+ *		there is a hole in the bitmap. Fwd reason is reception of BAR.
+ *
+ * duplicate_count
+ *		The number of duplicate frames that have been detected.
+ *
+ * frames_in_order_count
+ *		The number of frames that have been received in order (without
+ *		a hole that prevented them from being forwarded immediately).
+ *
+ * bar_received_count
+ *		The number of times a BAR frame is received.
+ *
+ * mpdu_frames_processed_count
+ * msdu_frames_processed_count
+ *		The total number of MPDU/MSDU frames that have been processed.
+ *
+ * total_bytes
+ *		An approximation of the number of bytes received for this queue.
+ *
+ * late_receive_mpdu_count
+ *		The number of MPDUs received after the window had already moved
+ *		on. The 'late' sequence window is defined as
+ *		(Window SSN - 256) - (Window SSN - 1).
+ *
+ * window_jump_2k
+ *		The number of times the window moved more than 2K
+ *
+ * hole_count
+ *		The number of times a hole was created in the receive bitmap.
+ *
+ * looping_count
+ *		A count value that indicates the number of times the producer of
+ *		entries into this Ring has looped around the ring.
+ */
+
+#define HAL_REO_STATUS_LOOP_CNT			GENMASK(31, 28)
+
+#define HAL_REO_FLUSH_QUEUE_INFO0_ERR_DETECTED	BIT(0)
+#define HAL_REO_FLUSH_QUEUE_INFO0_RSVD		GENMASK(31, 1)
+#define HAL_REO_FLUSH_QUEUE_INFO1_RSVD		GENMASK(27, 0)
+
+struct hal_reo_flush_queue_status {
+	struct hal_reo_status_hdr hdr;
+	u32 info0;
+	u32 rsvd0[21];
+	u32 info1;
+} __packed;
+
+/* hal_reo_flush_queue_status
+ *		Producer: REO
+ *		Consumer: SW
+ *
+ * status_hdr
+ *		Details that can link this status with the original command. It
+ *		also contains info on how long REO took to execute this command.
+ *
+ * error_detected
+ *		Status of blocking resource
+ *
+ *		0 - No error has been detected while executing this command
+ *		1 - Error detected. The resource to be used for blocking was
+ *		    already in use.
+ *
+ * looping_count
+ *		A count value that indicates the number of times the producer of
+ *		entries into this Ring has looped around the ring.
+ */
+
+#define HAL_REO_FLUSH_CACHE_STATUS_INFO0_IS_ERR			BIT(0)
+#define HAL_REO_FLUSH_CACHE_STATUS_INFO0_BLOCK_ERR_CODE		GENMASK(2, 1)
+#define HAL_REO_FLUSH_CACHE_STATUS_INFO0_FLUSH_STATUS_HIT	BIT(8)
+#define HAL_REO_FLUSH_CACHE_STATUS_INFO0_FLUSH_DESC_TYPE	GENMASK(11, 9)
+#define HAL_REO_FLUSH_CACHE_STATUS_INFO0_FLUSH_CLIENT_ID	GENMASK(15, 12)
+#define HAL_REO_FLUSH_CACHE_STATUS_INFO0_FLUSH_ERR		GENMASK(17, 16)
+#define HAL_REO_FLUSH_CACHE_STATUS_INFO0_FLUSH_COUNT		GENMASK(25, 18)
+
+struct hal_reo_flush_cache_status {
+	struct hal_reo_status_hdr hdr;
+	u32 info0;
+	u32 rsvd0[21];
+	u32 info1;
+} __packed;
+
+/* hal_reo_flush_cache_status
+ *		Producer: REO
+ *		Consumer: SW
+ *
+ * status_hdr
+ *		Details that can link this status with the original command. It
+ *		also contains info on how long REO took to execute this command.
+ *
+ * error_detected
+ *		Status for blocking resource handling
+ *
+ *		0 - No error has been detected while executing this command
+ *		1 - An error in the blocking resource management was detected
+ *
+ * block_error_details
+ *		only valid when error_detected is set
+ *
+ *		0 - No blocking related errors found
+ *		1 - Blocking resource is already in use
+ *		2 - Resource requested to be unblocked, was not blocked
+ *
+ * cache_controller_flush_status_hit
+ *		The status that the cache controller returned on executing the
+ *		flush command.
+ *
+ *		0 - miss; 1 - hit
+ *
+ * cache_controller_flush_status_desc_type
+ *		Flush descriptor type
+ *
+ * cache_controller_flush_status_client_id
+ *		Module who made the flush request
+ *
+ *		In REO, this is always 0
+ *
+ * cache_controller_flush_status_error
+ *		Error condition
+ *
+ *		0 - No error found
+ *		1 - HW interface is still busy
+ *		2 - Line currently locked. Used for one line flush command
+ *		3 - At least one line is still locked.
+ *		    Used for cache flush command.
+ *
+ * cache_controller_flush_count
+ *		The number of lines that were actually flushed out
+ *
+ * looping_count
+ *		A count value that indicates the number of times the producer of
+ *		entries into this Ring has looped around the ring.
+ */
+
+#define HAL_REO_UNBLOCK_CACHE_STATUS_INFO0_IS_ERR	BIT(0)
+#define HAL_REO_UNBLOCK_CACHE_STATUS_INFO0_TYPE		BIT(1)
+
+struct hal_reo_unblock_cache_status {
+	struct hal_reo_status_hdr hdr;
+	u32 info0;
+	u32 rsvd0[21];
+	u32 info1;
+} __packed;
+
+/* hal_reo_unblock_cache_status
+ *		Producer: REO
+ *		Consumer: SW
+ *
+ * status_hdr
+ *		Details that can link this status with the original command. It
+ *		also contains info on how long REO took to execute this command.
+ *
+ * error_detected
+ *		0 - No error has been detected while executing this command
+ *		1 - The blocking resource was not in use, and therefore it could
+ *		    not be unblocked.
+ *
+ * unblock_type
+ *		Reference to the type of unblock command
+ *		0 - Unblock a blocking resource
+ *		1 - The entire cache usage is unblock
+ *
+ * looping_count
+ *		A count value that indicates the number of times the producer of
+ *		entries into this Ring has looped around the ring.
+ */
+
+#define HAL_REO_FLUSH_TIMEOUT_STATUS_INFO0_IS_ERR		BIT(0)
+#define HAL_REO_FLUSH_TIMEOUT_STATUS_INFO0_LIST_EMPTY		BIT(1)
+
+#define HAL_REO_FLUSH_TIMEOUT_STATUS_INFO1_REL_DESC_COUNT	GENMASK(15, 0)
+#define HAL_REO_FLUSH_TIMEOUT_STATUS_INFO1_FWD_BUF_COUNT	GENMASK(31, 16)
+
+struct hal_reo_flush_timeout_list_status {
+	struct hal_reo_status_hdr hdr;
+	u32 info0;
+	u32 info1;
+	u32 rsvd0[20];
+	u32 info2;
+} __packed;
+
+/* hal_reo_flush_timeout_list_status
+ *		Producer: REO
+ *		Consumer: SW
+ *
+ * status_hdr
+ *		Details that can link this status with the original command. It
+ *		also contains info on how long REO took to execute this command.
+ *
+ * error_detected
+ *		0 - No error has been detected while executing this command
+ *		1 - Command not properly executed and returned with error
+ *
+ * timeout_list_empty
+ *		When set, REO has depleted the timeout list and all entries are
+ *		gone.
+ *
+ * release_desc_count
+ *		Producer: SW; Consumer: REO
+ *		The number of link descriptor released
+ *
+ * forward_buf_count
+ *		Producer: SW; Consumer: REO
+ *		The number of buffers forwarded to the REO destination rings
+ *
+ * looping_count
+ *		A count value that indicates the number of times the producer of
+ *		entries into this Ring has looped around the ring.
+ */
+
+#define HAL_REO_DESC_THRESH_STATUS_INFO0_THRESH_INDEX		GENMASK(1, 0)
+#define HAL_REO_DESC_THRESH_STATUS_INFO1_LINK_DESC_COUNTER0	GENMASK(23, 0)
+#define HAL_REO_DESC_THRESH_STATUS_INFO2_LINK_DESC_COUNTER1	GENMASK(23, 0)
+#define HAL_REO_DESC_THRESH_STATUS_INFO3_LINK_DESC_COUNTER2	GENMASK(23, 0)
+#define HAL_REO_DESC_THRESH_STATUS_INFO4_LINK_DESC_COUNTER_SUM	GENMASK(23, 0)
+
+struct hal_reo_desc_thresh_reached_status {
+	struct hal_reo_status_hdr hdr;
+	u32 info0;
+	u32 info1;
+	u32 info2;
+	u32 info3;
+	u32 info4;
+	u32 rsvd0[17];
+	u32 info5;
+} __packed;
+
+/* hal_reo_desc_thresh_reached_status
+ *		Producer: REO
+ *		Consumer: SW
+ *
+ * status_hdr
+ *		Details that can link this status with the original command. It
+ *		also contains info on how long REO took to execute this command.
+ *
+ * threshold_index
+ *		The index of the threshold register whose value got reached
+ *
+ * link_descriptor_counter0
+ * link_descriptor_counter1
+ * link_descriptor_counter2
+ * link_descriptor_counter_sum
+ *		Value of the respective counters at generation of this message
+ *
+ * looping_count
+ *		A count value that indicates the number of times the producer of
+ *		entries into this Ring has looped around the ring.
+ */
+
+#endif /* ATH11K_HAL_DESC_H */
diff --git a/drivers/net/wireless/ath/ath11k/hal_rx.c b/drivers/net/wireless/ath/ath11k/hal_rx.c
new file mode 100644
index 000000000000..2de4b388db20
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/hal_rx.c
@@ -0,0 +1,1221 @@
+// SPDX-License-Identifier: BSD-3-Clause-Clear
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#include "ahb.h"
+#include "debug.h"
+#include "hal.h"
+#include "hal_tx.h"
+#include "hal_rx.h"
+#include "hal_desc.h"
+
+static void ath11k_hal_reo_set_desc_hdr(struct hal_desc_header *hdr,
+					u8 owner, u8 buffer_type, u32 magic)
+{
+	hdr->info0 = FIELD_PREP(HAL_DESC_HDR_INFO0_OWNER, owner) |
+		     FIELD_PREP(HAL_DESC_HDR_INFO0_BUF_TYPE, buffer_type);
+
+	/* Magic pattern in reserved bits for debugging */
+	hdr->info0 |= FIELD_PREP(HAL_DESC_HDR_INFO0_DBG_RESERVED, magic);
+}
+
+static int ath11k_hal_reo_cmd_queue_stats(struct hal_tlv_hdr *tlv,
+					  struct ath11k_hal_reo_cmd *cmd)
+{
+	struct hal_reo_get_queue_stats *desc;
+
+	tlv->tl = FIELD_PREP(HAL_TLV_HDR_TAG, HAL_REO_GET_QUEUE_STATS) |
+		  FIELD_PREP(HAL_TLV_HDR_LEN, sizeof(*desc));
+
+	desc = (struct hal_reo_get_queue_stats *)tlv->value;
+	memset(&desc->queue_addr_lo, 0,
+	       (sizeof(*desc) - sizeof(struct hal_reo_cmd_hdr)));
+
+	desc->cmd.info0 &= ~HAL_REO_CMD_HDR_INFO0_STATUS_REQUIRED;
+	if (cmd->flag & HAL_REO_CMD_FLG_NEED_STATUS)
+		desc->cmd.info0 |= HAL_REO_CMD_HDR_INFO0_STATUS_REQUIRED;
+
+	desc->queue_addr_lo = cmd->addr_lo;
+	desc->info0 = FIELD_PREP(HAL_REO_GET_QUEUE_STATS_INFO0_QUEUE_ADDR_HI,
+				 cmd->addr_hi);
+	if (cmd->flag & HAL_REO_CMD_FLG_STATS_CLEAR)
+		desc->info0 |= HAL_REO_GET_QUEUE_STATS_INFO0_CLEAR_STATS;
+
+	return FIELD_GET(HAL_REO_CMD_HDR_INFO0_CMD_NUMBER, desc->cmd.info0);
+}
+
+static int ath11k_hal_reo_cmd_flush_cache(struct ath11k_hal *hal, struct hal_tlv_hdr *tlv,
+					  struct ath11k_hal_reo_cmd *cmd)
+{
+	struct hal_reo_flush_cache *desc;
+	u8 avail_slot = ffz(hal->avail_blk_resource);
+
+	if (cmd->flag & HAL_REO_CMD_FLG_FLUSH_BLOCK_LATER) {
+		if (avail_slot >= HAL_MAX_AVAIL_BLK_RES)
+			return -ENOSPC;
+
+		hal->current_blk_index = avail_slot;
+	}
+
+	tlv->tl = FIELD_PREP(HAL_TLV_HDR_TAG, HAL_REO_FLUSH_CACHE) |
+		  FIELD_PREP(HAL_TLV_HDR_LEN, sizeof(*desc));
+
+	desc = (struct hal_reo_flush_cache *)tlv->value;
+	memset(&desc->cache_addr_lo, 0,
+	       (sizeof(*desc) - sizeof(struct hal_reo_cmd_hdr)));
+
+	desc->cmd.info0 &= ~HAL_REO_CMD_HDR_INFO0_STATUS_REQUIRED;
+	if (cmd->flag & HAL_REO_CMD_FLG_NEED_STATUS)
+		desc->cmd.info0 |= HAL_REO_CMD_HDR_INFO0_STATUS_REQUIRED;
+
+	desc->cache_addr_lo = cmd->addr_lo;
+	desc->info0 = FIELD_PREP(HAL_REO_FLUSH_CACHE_INFO0_CACHE_ADDR_HI,
+				 cmd->addr_hi);
+
+	if (cmd->flag & HAL_REO_CMD_FLG_FLUSH_FWD_ALL_MPDUS)
+		desc->info0 |= HAL_REO_FLUSH_CACHE_INFO0_FWD_ALL_MPDUS;
+
+	if (cmd->flag & HAL_REO_CMD_FLG_FLUSH_BLOCK_LATER) {
+		desc->info0 |= HAL_REO_FLUSH_CACHE_INFO0_BLOCK_CACHE_USAGE;
+		desc->info0 |=
+			FIELD_PREP(HAL_REO_FLUSH_CACHE_INFO0_BLOCK_RESRC_IDX,
+				   avail_slot);
+	}
+
+	if (cmd->flag & HAL_REO_CMD_FLG_FLUSH_NO_INVAL)
+		desc->info0 |= HAL_REO_FLUSH_CACHE_INFO0_FLUSH_WO_INVALIDATE;
+
+	if (cmd->flag & HAL_REO_CMD_FLG_FLUSH_ALL)
+		desc->info0 |= HAL_REO_FLUSH_CACHE_INFO0_FLUSH_ALL;
+
+	return FIELD_GET(HAL_REO_CMD_HDR_INFO0_CMD_NUMBER, desc->cmd.info0);
+}
+
+static int ath11k_hal_reo_cmd_update_rx_queue(struct hal_tlv_hdr *tlv,
+					      struct ath11k_hal_reo_cmd *cmd)
+{
+	struct hal_reo_update_rx_queue *desc;
+
+	tlv->tl = FIELD_PREP(HAL_TLV_HDR_TAG, HAL_REO_UPDATE_RX_REO_QUEUE) |
+		  FIELD_PREP(HAL_TLV_HDR_LEN, sizeof(*desc));
+
+	desc = (struct hal_reo_update_rx_queue *)tlv->value;
+	memset(&desc->queue_addr_lo, 0,
+	       (sizeof(*desc) - sizeof(struct hal_reo_cmd_hdr)));
+
+	desc->cmd.info0 &= ~HAL_REO_CMD_HDR_INFO0_STATUS_REQUIRED;
+	if (cmd->flag & HAL_REO_CMD_FLG_NEED_STATUS)
+		desc->cmd.info0 |= HAL_REO_CMD_HDR_INFO0_STATUS_REQUIRED;
+
+	desc->queue_addr_lo = cmd->addr_lo;
+	desc->info0 =
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO0_QUEUE_ADDR_HI,
+			   cmd->addr_hi) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO0_UPD_RX_QUEUE_NUM,
+			   !!(cmd->upd0 & HAL_REO_CMD_UPD0_RX_QUEUE_NUM)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO0_UPD_VLD,
+			   !!(cmd->upd0 & HAL_REO_CMD_UPD0_VLD)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO0_UPD_ASSOC_LNK_DESC_CNT,
+			   !!(cmd->upd0 & HAL_REO_CMD_UPD0_ALDC)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO0_UPD_DIS_DUP_DETECTION,
+			   !!(cmd->upd0 & HAL_REO_CMD_UPD0_DIS_DUP_DETECTION)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO0_UPD_SOFT_REORDER_EN,
+			   !!(cmd->upd0 & HAL_REO_CMD_UPD0_SOFT_REORDER_EN)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO0_UPD_AC,
+			   !!(cmd->upd0 & HAL_REO_CMD_UPD0_AC)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO0_UPD_BAR,
+			   !!(cmd->upd0 & HAL_REO_CMD_UPD0_BAR)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO0_UPD_RETRY,
+			   !!(cmd->upd0 & HAL_REO_CMD_UPD0_RETRY)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO0_UPD_CHECK_2K_MODE,
+			   !!(cmd->upd0 & HAL_REO_CMD_UPD0_CHECK_2K_MODE)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO0_UPD_OOR_MODE,
+			   !!(cmd->upd0 & HAL_REO_CMD_UPD0_OOR_MODE)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO0_UPD_BA_WINDOW_SIZE,
+			   !!(cmd->upd0 & HAL_REO_CMD_UPD0_BA_WINDOW_SIZE)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO0_UPD_PN_CHECK,
+			   !!(cmd->upd0 & HAL_REO_CMD_UPD0_PN_CHECK)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO0_UPD_EVEN_PN,
+			   !!(cmd->upd0 & HAL_REO_CMD_UPD0_EVEN_PN)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO0_UPD_UNEVEN_PN,
+			   !!(cmd->upd0 & HAL_REO_CMD_UPD0_UNEVEN_PN)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO0_UPD_PN_HANDLE_ENABLE,
+			   !!(cmd->upd0 & HAL_REO_CMD_UPD0_PN_HANDLE_ENABLE)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO0_UPD_PN_SIZE,
+			   !!(cmd->upd0 & HAL_REO_CMD_UPD0_PN_SIZE)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO0_UPD_IGNORE_AMPDU_FLG,
+			   !!(cmd->upd0 & HAL_REO_CMD_UPD0_IGNORE_AMPDU_FLG)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO0_UPD_SVLD,
+			   !!(cmd->upd0 & HAL_REO_CMD_UPD0_SVLD)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO0_UPD_SSN,
+			   !!(cmd->upd0 & HAL_REO_CMD_UPD0_SSN)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO0_UPD_SEQ_2K_ERR,
+			   !!(cmd->upd0 & HAL_REO_CMD_UPD0_SEQ_2K_ERR)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO0_UPD_PN_VALID,
+			   !!(cmd->upd0 & HAL_REO_CMD_UPD0_PN_VALID)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO0_UPD_PN,
+			   !!(cmd->upd0 & HAL_REO_CMD_UPD0_PN));
+
+	desc->info1 =
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO1_RX_QUEUE_NUMBER,
+			   cmd->rx_queue_num) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO1_VLD,
+			   !!(cmd->upd1 & HAL_REO_CMD_UPD1_VLD)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO1_ASSOC_LNK_DESC_COUNTER,
+			   FIELD_GET(HAL_REO_CMD_UPD1_ALDC, cmd->upd1)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO1_DIS_DUP_DETECTION,
+			   !!(cmd->upd1 & HAL_REO_CMD_UPD1_DIS_DUP_DETECTION)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO1_SOFT_REORDER_EN,
+			   !!(cmd->upd1 & HAL_REO_CMD_UPD1_SOFT_REORDER_EN)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO1_AC,
+			   FIELD_GET(HAL_REO_CMD_UPD1_AC, cmd->upd1)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO1_BAR,
+			   !!(cmd->upd1 & HAL_REO_CMD_UPD1_BAR)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO1_CHECK_2K_MODE,
+			   !!(cmd->upd1 & HAL_REO_CMD_UPD1_CHECK_2K_MODE)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO1_RETRY,
+			   !!(cmd->upd1 & HAL_REO_CMD_UPD1_RETRY)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO1_OOR_MODE,
+			   !!(cmd->upd1 & HAL_REO_CMD_UPD1_OOR_MODE)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO1_PN_CHECK,
+			   !!(cmd->upd1 & HAL_REO_CMD_UPD1_PN_CHECK)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO1_EVEN_PN,
+			   !!(cmd->upd1 & HAL_REO_CMD_UPD1_EVEN_PN)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO1_UNEVEN_PN,
+			   !!(cmd->upd1 & HAL_REO_CMD_UPD1_UNEVEN_PN)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO1_PN_HANDLE_ENABLE,
+			   !!(cmd->upd1 & HAL_REO_CMD_UPD1_PN_HANDLE_ENABLE)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO1_IGNORE_AMPDU_FLG,
+			   !!(cmd->upd1 & HAL_REO_CMD_UPD1_IGNORE_AMPDU_FLG));
+
+	if (cmd->pn_size == 24)
+		cmd->pn_size = HAL_RX_REO_QUEUE_PN_SIZE_24;
+	else if (cmd->pn_size == 48)
+		cmd->pn_size = HAL_RX_REO_QUEUE_PN_SIZE_48;
+	else if (cmd->pn_size == 128)
+		cmd->pn_size = HAL_RX_REO_QUEUE_PN_SIZE_128;
+
+	if (cmd->ba_window_size < 1)
+		cmd->ba_window_size = 1;
+
+	if (cmd->ba_window_size == 1)
+		cmd->ba_window_size++;
+
+	desc->info2 =
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO2_BA_WINDOW_SIZE,
+			   cmd->ba_window_size - 1) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO2_PN_SIZE, cmd->pn_size) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO2_SVLD,
+			   !!(cmd->upd2 & HAL_REO_CMD_UPD2_SVLD)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO2_SSN,
+			   FIELD_GET(HAL_REO_CMD_UPD2_SSN, cmd->upd2)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO2_SEQ_2K_ERR,
+			   !!(cmd->upd2 & HAL_REO_CMD_UPD2_SEQ_2K_ERR)) |
+		FIELD_PREP(HAL_REO_UPD_RX_QUEUE_INFO2_PN_ERR,
+			   !!(cmd->upd2 & HAL_REO_CMD_UPD2_PN_ERR));
+
+	return FIELD_GET(HAL_REO_CMD_HDR_INFO0_CMD_NUMBER, desc->cmd.info0);
+}
+
+int ath11k_hal_reo_cmd_send(struct ath11k_base *ab, struct hal_srng *srng,
+			    enum hal_reo_cmd_type type,
+			    struct ath11k_hal_reo_cmd *cmd)
+{
+	struct hal_tlv_hdr *reo_desc;
+	int ret;
+
+	spin_lock_bh(&srng->lock);
+
+	ath11k_hal_srng_access_begin(ab, srng);
+	reo_desc = (struct hal_tlv_hdr *)ath11k_hal_srng_src_get_next_entry(ab, srng);
+	if (!reo_desc) {
+		ret = -ENOBUFS;
+		goto out;
+	}
+
+	switch (type) {
+	case HAL_REO_CMD_GET_QUEUE_STATS:
+		ret = ath11k_hal_reo_cmd_queue_stats(reo_desc, cmd);
+		break;
+	case HAL_REO_CMD_FLUSH_CACHE:
+		ret = ath11k_hal_reo_cmd_flush_cache(&ab->hal, reo_desc, cmd);
+		break;
+	case HAL_REO_CMD_UPDATE_RX_QUEUE:
+		ret = ath11k_hal_reo_cmd_update_rx_queue(reo_desc, cmd);
+		break;
+	case HAL_REO_CMD_FLUSH_QUEUE:
+	case HAL_REO_CMD_UNBLOCK_CACHE:
+	case HAL_REO_CMD_FLUSH_TIMEOUT_LIST:
+		ath11k_warn(ab, "Unsupported reo command %d\n", type);
+		ret = -ENOTSUPP;
+		break;
+	default:
+		ath11k_warn(ab, "Unknown reo command %d\n", type);
+		ret = -EINVAL;
+		break;
+	}
+
+out:
+	ath11k_hal_srng_access_end(ab, srng);
+	spin_unlock_bh(&srng->lock);
+
+	return ret;
+}
+
+void ath11k_hal_rx_buf_addr_info_set(void *desc, dma_addr_t paddr,
+				     u32 cookie, u8 manager)
+{
+	struct ath11k_buffer_addr *binfo = (struct ath11k_buffer_addr *)desc;
+	u32 paddr_lo, paddr_hi;
+
+	paddr_lo = lower_32_bits(paddr);
+	paddr_hi = upper_32_bits(paddr);
+	binfo->info0 = FIELD_PREP(BUFFER_ADDR_INFO0_ADDR, paddr_lo);
+	binfo->info1 = FIELD_PREP(BUFFER_ADDR_INFO1_ADDR, paddr_hi) |
+		       FIELD_PREP(BUFFER_ADDR_INFO1_SW_COOKIE, cookie) |
+		       FIELD_PREP(BUFFER_ADDR_INFO1_RET_BUF_MGR, manager);
+}
+
+void ath11k_hal_rx_buf_addr_info_get(void *desc, dma_addr_t *paddr,
+				     u32 *cookie, u8 *rbm)
+{
+	struct ath11k_buffer_addr *binfo = (struct ath11k_buffer_addr *)desc;
+
+	*paddr =
+		(((u64)FIELD_GET(BUFFER_ADDR_INFO1_ADDR, binfo->info1)) << 32) |
+		FIELD_GET(BUFFER_ADDR_INFO0_ADDR, binfo->info0);
+	*cookie = FIELD_GET(BUFFER_ADDR_INFO1_SW_COOKIE, binfo->info1);
+	*rbm = FIELD_GET(BUFFER_ADDR_INFO1_RET_BUF_MGR, binfo->info1);
+}
+
+void ath11k_hal_rx_msdu_link_info_get(void *link_desc, u32 *num_msdus,
+				      struct hal_rx_msdu_meta *meta,
+				      enum hal_rx_buf_return_buf_manager *rbm)
+{
+	struct hal_rx_msdu_link *link = (struct hal_rx_msdu_link *)link_desc;
+	struct hal_rx_msdu_details *msdu;
+	int i;
+
+	*num_msdus = HAL_NUM_RX_MSDUS_PER_LINK_DESC;
+
+	msdu = &link->msdu_link[0];
+	*rbm = FIELD_GET(BUFFER_ADDR_INFO1_RET_BUF_MGR,
+			 msdu->buf_addr_info.info1);
+
+	for (i = 0; i < *num_msdus; i++) {
+		msdu = &link->msdu_link[i];
+
+		if (!FIELD_GET(BUFFER_ADDR_INFO0_ADDR,
+			       msdu->buf_addr_info.info0)) {
+			*num_msdus = i;
+			break;
+		}
+		meta->msdu_len = FIELD_GET(RX_MSDU_DESC_INFO0_MSDU_LENGTH,
+					   msdu->rx_msdu_info.info0);
+		meta->first = !!(msdu->rx_msdu_info.info0 &
+				 RX_MSDU_DESC_INFO0_FIRST_MSDU_IN_MPDU);
+		meta->last = !!(msdu->rx_msdu_info.info0 &
+				RX_MSDU_DESC_INFO0_LAST_MSDU_IN_MPDU);
+		meta->continuation = !!(msdu->rx_msdu_info.info0 &
+					RX_MSDU_DESC_INFO0_MSDU_CONTINUATION);
+		meta->cookie = FIELD_GET(BUFFER_ADDR_INFO1_SW_COOKIE,
+					 msdu->buf_addr_info.info1);
+		meta++;
+	}
+}
+
+int ath11k_hal_desc_reo_parse_err(struct ath11k_base *ab, u32 *rx_desc,
+				  dma_addr_t *paddr, u32 *desc_bank)
+{
+	struct hal_reo_dest_ring *desc = (struct hal_reo_dest_ring *)rx_desc;
+	enum hal_reo_dest_ring_push_reason push_reason;
+	enum hal_reo_dest_ring_error_code err_code;
+
+	push_reason = FIELD_GET(HAL_REO_DEST_RING_INFO0_PUSH_REASON,
+				desc->info0);
+	err_code = FIELD_GET(HAL_REO_DEST_RING_INFO0_ERROR_CODE,
+			     desc->info0);
+	ab->soc_stats.reo_error[err_code]++;
+
+	if (push_reason != HAL_REO_DEST_RING_PUSH_REASON_ERR_DETECTED &&
+	    push_reason != HAL_REO_DEST_RING_PUSH_REASON_ROUTING_INSTRUCTION) {
+		ath11k_warn(ab, "expected error push reason code, received %d\n",
+			    push_reason);
+		return -EINVAL;
+	}
+
+	if (FIELD_GET(HAL_REO_DEST_RING_INFO0_BUFFER_TYPE, desc->info0) !=
+	    HAL_REO_DEST_RING_BUFFER_TYPE_LINK_DESC) {
+		ath11k_warn(ab, "expected buffer type link_desc");
+		return -EINVAL;
+	}
+
+	ath11k_hal_rx_reo_ent_paddr_get(ab, rx_desc, paddr, desc_bank);
+
+	return 0;
+}
+
+void ath11k_hal_rx_parse_dst_ring_desc(struct ath11k_base *ab, u32 *rx_desc,
+				       struct hal_rx_meta_info *meta_info)
+{
+	struct hal_reo_dest_ring *desc = (struct hal_reo_dest_ring *)rx_desc;
+	struct rx_mpdu_desc *mpdu = &desc->rx_mpdu_info;
+	struct rx_msdu_desc *msdu = &desc->rx_msdu_info;
+	struct hal_rx_mpdu_meta *meta_mpdu = &meta_info->mpdu_meta;
+	struct hal_rx_msdu_meta *meta_msdu = &meta_info->msdu_meta;
+
+	meta_info->push_reason = FIELD_GET(HAL_REO_DEST_RING_INFO0_PUSH_REASON,
+					   desc->info0);
+
+	meta_mpdu->msdu_cnt = FIELD_GET(RX_MPDU_DESC_INFO0_MSDU_COUNT,
+					mpdu->info0);
+	meta_mpdu->seq_num = FIELD_GET(RX_MPDU_DESC_INFO0_SEQ_NUM, mpdu->info0);
+	meta_mpdu->frag = !!(mpdu->info0 & RX_MPDU_DESC_INFO0_FRAG_FLAG);
+	meta_mpdu->retry = !!(mpdu->info0 & RX_MPDU_DESC_INFO0_MPDU_RETRY);
+	meta_mpdu->ampdu = !!(mpdu->info0 & RX_MPDU_DESC_INFO0_AMPDU_FLAG);
+	meta_mpdu->raw = !!(mpdu->info0 & RX_MPDU_DESC_INFO0_RAW_MPDU);
+	meta_mpdu->peer_meta = mpdu->meta_data;
+
+	meta_msdu->cookie = FIELD_GET(BUFFER_ADDR_INFO1_SW_COOKIE,
+				      desc->buf_addr_info.info1);
+	meta_msdu->msdu_len = FIELD_GET(RX_MSDU_DESC_INFO0_MSDU_LENGTH,
+					msdu->info0);
+	meta_msdu->first =
+		!!(msdu->info0 & RX_MSDU_DESC_INFO0_FIRST_MSDU_IN_MPDU);
+	meta_msdu->last =
+		!!(msdu->info0 & RX_MSDU_DESC_INFO0_LAST_MSDU_IN_MPDU);
+	meta_msdu->continuation =
+		!!(msdu->info0 & RX_MSDU_DESC_INFO0_MSDU_CONTINUATION);
+}
+
+int ath11k_hal_wbm_desc_parse_err(struct ath11k_base *ab, void *desc,
+				  struct hal_rx_wbm_rel_info *rel_info)
+{
+	struct hal_wbm_release_ring *wbm_desc = desc;
+	enum hal_wbm_rel_desc_type type;
+	enum hal_wbm_rel_src_module rel_src;
+
+	type = FIELD_GET(HAL_WBM_RELEASE_INFO0_DESC_TYPE,
+			 wbm_desc->info0);
+	/* We expect only WBM_REL buffer type */
+	if (type != HAL_WBM_REL_DESC_TYPE_REL_MSDU) {
+		WARN_ON(1);
+		return -EINVAL;
+	}
+
+	rel_src = FIELD_GET(HAL_WBM_RELEASE_INFO0_REL_SRC_MODULE,
+			    wbm_desc->info0);
+	if (rel_src != HAL_WBM_REL_SRC_MODULE_RXDMA &&
+	    rel_src != HAL_WBM_REL_SRC_MODULE_REO)
+		return -EINVAL;
+
+	if (FIELD_GET(BUFFER_ADDR_INFO1_RET_BUF_MGR,
+		      wbm_desc->buf_addr_info.info1) != HAL_RX_BUF_RBM_SW3_BM) {
+		ab->soc_stats.invalid_rbm++;
+		return -EINVAL;
+	}
+
+	rel_info->cookie = FIELD_GET(BUFFER_ADDR_INFO1_SW_COOKIE,
+				     wbm_desc->buf_addr_info.info1);
+	rel_info->err_rel_src = rel_src;
+	if (rel_src == HAL_WBM_REL_SRC_MODULE_REO) {
+		rel_info->push_reason =
+			FIELD_GET(HAL_WBM_RELEASE_INFO0_REO_PUSH_REASON,
+				  wbm_desc->info0);
+		rel_info->err_code =
+			FIELD_GET(HAL_WBM_RELEASE_INFO0_REO_ERROR_CODE,
+				  wbm_desc->info0);
+	} else {
+		rel_info->push_reason =
+			FIELD_GET(HAL_WBM_RELEASE_INFO0_RXDMA_PUSH_REASON,
+				  wbm_desc->info0);
+		rel_info->err_code =
+			FIELD_GET(HAL_WBM_RELEASE_INFO0_RXDMA_ERROR_CODE,
+				  wbm_desc->info0);
+	}
+
+	rel_info->first_msdu = FIELD_GET(HAL_WBM_RELEASE_INFO2_FIRST_MSDU,
+					 wbm_desc->info2);
+	rel_info->last_msdu = FIELD_GET(HAL_WBM_RELEASE_INFO2_LAST_MSDU,
+					wbm_desc->info2);
+	return 0;
+}
+
+void ath11k_hal_rx_reo_ent_paddr_get(struct ath11k_base *ab, void *desc,
+				     dma_addr_t *paddr, u32 *desc_bank)
+{
+	struct ath11k_buffer_addr *buff_addr = desc;
+
+	*paddr = ((u64)(FIELD_GET(BUFFER_ADDR_INFO1_ADDR, buff_addr->info1)) << 32) |
+		  FIELD_GET(BUFFER_ADDR_INFO0_ADDR, buff_addr->info0);
+
+	*desc_bank = FIELD_GET(BUFFER_ADDR_INFO1_SW_COOKIE, buff_addr->info1);
+}
+
+void ath11k_hal_rx_msdu_link_desc_set(struct ath11k_base *ab, void *desc,
+				      void *link_desc,
+				      enum hal_wbm_rel_bm_act action)
+{
+	struct hal_wbm_release_ring *dst_desc = desc;
+	struct hal_wbm_release_ring *src_desc = link_desc;
+
+	dst_desc->buf_addr_info = src_desc->buf_addr_info;
+	dst_desc->info0 |= FIELD_PREP(HAL_WBM_RELEASE_INFO0_REL_SRC_MODULE,
+				      HAL_WBM_REL_SRC_MODULE_SW) |
+			   FIELD_PREP(HAL_WBM_RELEASE_INFO0_BM_ACTION, action) |
+			   FIELD_PREP(HAL_WBM_RELEASE_INFO0_DESC_TYPE,
+				      HAL_WBM_REL_DESC_TYPE_MSDU_LINK);
+}
+
+void ath11k_hal_reo_status_queue_stats(struct ath11k_base *ab, u32 *reo_desc,
+				       struct hal_reo_status *status)
+{
+	struct hal_tlv_hdr *tlv = (struct hal_tlv_hdr *)reo_desc;
+	struct hal_reo_get_queue_stats_status *desc =
+		(struct hal_reo_get_queue_stats_status *)tlv->value;
+
+	status->uniform_hdr.cmd_num =
+				FIELD_GET(HAL_REO_STATUS_HDR_INFO0_STATUS_NUM,
+					  desc->hdr.info0);
+	status->uniform_hdr.cmd_status =
+				FIELD_GET(HAL_REO_STATUS_HDR_INFO0_EXEC_STATUS,
+					  desc->hdr.info0);
+
+	ath11k_dbg(ab, ATH11k_DBG_HAL, "Queue stats status:\n");
+	ath11k_dbg(ab, ATH11k_DBG_HAL, "header: cmd_num %d status %d\n",
+		   status->uniform_hdr.cmd_num,
+		   status->uniform_hdr.cmd_status);
+	ath11k_dbg(ab, ATH11k_DBG_HAL, "ssn %ld cur_idx %ld\n",
+		   FIELD_GET(HAL_REO_GET_QUEUE_STATS_STATUS_INFO0_SSN,
+			     desc->info0),
+		   FIELD_GET(HAL_REO_GET_QUEUE_STATS_STATUS_INFO0_CUR_IDX,
+			     desc->info0));
+	ath11k_dbg(ab, ATH11k_DBG_HAL, "pn = [%08x, %08x, %08x, %08x]\n",
+		   desc->pn[0], desc->pn[1], desc->pn[2], desc->pn[3]);
+	ath11k_dbg(ab, ATH11k_DBG_HAL, "last_rx: enqueue_tstamp %08x dequeue_tstamp %08x\n",
+		   desc->last_rx_enqueue_timestamp,
+		   desc->last_rx_dequeue_timestamp);
+	ath11k_dbg(ab, ATH11k_DBG_HAL, "rx_bitmap [%08x %08x %08x %08x %08x %08x %08x %08x]\n",
+		   desc->rx_bitmap[0], desc->rx_bitmap[1], desc->rx_bitmap[2],
+		   desc->rx_bitmap[3], desc->rx_bitmap[4], desc->rx_bitmap[5],
+		   desc->rx_bitmap[6], desc->rx_bitmap[7]);
+	ath11k_dbg(ab, ATH11k_DBG_HAL, "count: cur_mpdu %ld cur_msdu %ld\n",
+		   FIELD_GET(HAL_REO_GET_QUEUE_STATS_STATUS_INFO1_MPDU_COUNT,
+			     desc->info1),
+		   FIELD_GET(HAL_REO_GET_QUEUE_STATS_STATUS_INFO1_MSDU_COUNT,
+			     desc->info1));
+	ath11k_dbg(ab, ATH11k_DBG_HAL, "fwd_timeout %ld fwd_bar %ld dup_count %ld\n",
+		   FIELD_GET(HAL_REO_GET_QUEUE_STATS_STATUS_INFO2_TIMEOUT_COUNT,
+			     desc->info2),
+		   FIELD_GET(HAL_REO_GET_QUEUE_STATS_STATUS_INFO2_FDTB_COUNT,
+			     desc->info2),
+		   FIELD_GET(HAL_REO_GET_QUEUE_STATS_STATUS_INFO2_DUPLICATE_COUNT,
+			     desc->info2));
+	ath11k_dbg(ab, ATH11k_DBG_HAL, "frames_in_order %ld bar_rcvd %ld\n",
+		   FIELD_GET(HAL_REO_GET_QUEUE_STATS_STATUS_INFO3_FIO_COUNT,
+			     desc->info3),
+		   FIELD_GET(HAL_REO_GET_QUEUE_STATS_STATUS_INFO3_BAR_RCVD_CNT,
+			     desc->info3));
+	ath11k_dbg(ab, ATH11k_DBG_HAL, "num_mpdus %d num_msdus %d total_bytes %d\n",
+		   desc->num_mpdu_frames, desc->num_msdu_frames,
+		   desc->total_bytes);
+	ath11k_dbg(ab, ATH11k_DBG_HAL, "late_rcvd %ld win_jump_2k %ld hole_cnt %ld\n",
+		   FIELD_GET(HAL_REO_GET_QUEUE_STATS_STATUS_INFO4_LATE_RX_MPDU,
+			     desc->info4),
+		   FIELD_GET(HAL_REO_GET_QUEUE_STATS_STATUS_INFO4_WINDOW_JMP2K,
+			     desc->info4),
+		   FIELD_GET(HAL_REO_GET_QUEUE_STATS_STATUS_INFO4_HOLE_COUNT,
+			     desc->info4));
+	ath11k_dbg(ab, ATH11k_DBG_HAL, "looping count %ld\n",
+		   FIELD_GET(HAL_REO_GET_QUEUE_STATS_STATUS_INFO5_LOOPING_CNT,
+			     desc->info5));
+}
+
+int ath11k_hal_reo_process_status(u8 *reo_desc, u8 *status)
+{
+	struct hal_tlv_hdr *tlv = (struct hal_tlv_hdr *)reo_desc;
+	struct hal_reo_status_hdr *hdr;
+
+	hdr = (struct hal_reo_status_hdr *)tlv->value;
+	*status = FIELD_GET(HAL_REO_STATUS_HDR_INFO0_EXEC_STATUS, hdr->info0);
+
+	return FIELD_GET(HAL_REO_STATUS_HDR_INFO0_STATUS_NUM, hdr->info0);
+}
+
+void ath11k_hal_reo_flush_queue_status(struct ath11k_base *ab, u32 *reo_desc,
+				       struct hal_reo_status *status)
+{
+	struct hal_tlv_hdr *tlv = (struct hal_tlv_hdr *)reo_desc;
+	struct hal_reo_flush_queue_status *desc =
+		(struct hal_reo_flush_queue_status *)tlv->value;
+
+	status->uniform_hdr.cmd_num =
+				FIELD_GET(HAL_REO_STATUS_HDR_INFO0_STATUS_NUM,
+					  desc->hdr.info0);
+	status->uniform_hdr.cmd_status =
+				FIELD_GET(HAL_REO_STATUS_HDR_INFO0_EXEC_STATUS,
+					  desc->hdr.info0);
+	status->u.flush_queue.err_detected =
+		FIELD_GET(HAL_REO_FLUSH_QUEUE_INFO0_ERR_DETECTED,
+			  desc->info0);
+}
+
+void ath11k_hal_reo_flush_cache_status(struct ath11k_base *ab, u32 *reo_desc,
+				       struct hal_reo_status *status)
+{
+	struct ath11k_hal *hal = &ab->hal;
+	struct hal_tlv_hdr *tlv = (struct hal_tlv_hdr *)reo_desc;
+	struct hal_reo_flush_cache_status *desc =
+		(struct hal_reo_flush_cache_status *)tlv->value;
+
+	status->uniform_hdr.cmd_num =
+				FIELD_GET(HAL_REO_STATUS_HDR_INFO0_STATUS_NUM,
+					  desc->hdr.info0);
+	status->uniform_hdr.cmd_status =
+				FIELD_GET(HAL_REO_STATUS_HDR_INFO0_EXEC_STATUS,
+					  desc->hdr.info0);
+
+	status->u.flush_cache.err_detected =
+			FIELD_GET(HAL_REO_FLUSH_CACHE_STATUS_INFO0_IS_ERR,
+				  desc->info0);
+	status->u.flush_cache.err_code =
+		FIELD_GET(HAL_REO_FLUSH_CACHE_STATUS_INFO0_BLOCK_ERR_CODE,
+			  desc->info0);
+	if (!status->u.flush_cache.err_code)
+		hal->avail_blk_resource |= BIT(hal->current_blk_index);
+
+	status->u.flush_cache.cache_controller_flush_status_hit =
+		FIELD_GET(HAL_REO_FLUSH_CACHE_STATUS_INFO0_FLUSH_STATUS_HIT,
+			  desc->info0);
+
+	status->u.flush_cache.cache_controller_flush_status_desc_type =
+		FIELD_GET(HAL_REO_FLUSH_CACHE_STATUS_INFO0_FLUSH_DESC_TYPE,
+			  desc->info0);
+	status->u.flush_cache.cache_controller_flush_status_client_id =
+		FIELD_GET(HAL_REO_FLUSH_CACHE_STATUS_INFO0_FLUSH_CLIENT_ID,
+			  desc->info0);
+	status->u.flush_cache.cache_controller_flush_status_err =
+		FIELD_GET(HAL_REO_FLUSH_CACHE_STATUS_INFO0_FLUSH_ERR,
+			  desc->info0);
+	status->u.flush_cache.cache_controller_flush_status_cnt =
+		FIELD_GET(HAL_REO_FLUSH_CACHE_STATUS_INFO0_FLUSH_COUNT,
+			  desc->info0);
+}
+
+void ath11k_hal_reo_unblk_cache_status(struct ath11k_base *ab, u32 *reo_desc,
+				       struct hal_reo_status *status)
+{
+	struct ath11k_hal *hal = &ab->hal;
+	struct hal_tlv_hdr *tlv = (struct hal_tlv_hdr *)reo_desc;
+	struct hal_reo_unblock_cache_status *desc =
+		(struct hal_reo_unblock_cache_status *)tlv->value;
+
+	status->uniform_hdr.cmd_num =
+				FIELD_GET(HAL_REO_STATUS_HDR_INFO0_STATUS_NUM,
+					  desc->hdr.info0);
+	status->uniform_hdr.cmd_status =
+				FIELD_GET(HAL_REO_STATUS_HDR_INFO0_EXEC_STATUS,
+					  desc->hdr.info0);
+
+	status->u.unblock_cache.err_detected =
+			FIELD_GET(HAL_REO_UNBLOCK_CACHE_STATUS_INFO0_IS_ERR,
+				  desc->info0);
+	status->u.unblock_cache.unblock_type =
+			FIELD_GET(HAL_REO_UNBLOCK_CACHE_STATUS_INFO0_TYPE,
+				  desc->info0);
+
+	if (!status->u.unblock_cache.err_detected &&
+	    status->u.unblock_cache.unblock_type ==
+	    HAL_REO_STATUS_UNBLOCK_BLOCKING_RESOURCE)
+		hal->avail_blk_resource &= ~BIT(hal->current_blk_index);
+}
+
+void ath11k_hal_reo_flush_timeout_list_status(struct ath11k_base *ab,
+					      u32 *reo_desc,
+					      struct hal_reo_status *status)
+{
+	struct hal_tlv_hdr *tlv = (struct hal_tlv_hdr *)reo_desc;
+	struct hal_reo_flush_timeout_list_status *desc =
+		(struct hal_reo_flush_timeout_list_status *)tlv->value;
+
+	status->uniform_hdr.cmd_num =
+				FIELD_GET(HAL_REO_STATUS_HDR_INFO0_STATUS_NUM,
+					  desc->hdr.info0);
+	status->uniform_hdr.cmd_status =
+				FIELD_GET(HAL_REO_STATUS_HDR_INFO0_EXEC_STATUS,
+					  desc->hdr.info0);
+
+	status->u.timeout_list.err_detected =
+			FIELD_GET(HAL_REO_FLUSH_TIMEOUT_STATUS_INFO0_IS_ERR,
+				  desc->info0);
+	status->u.timeout_list.list_empty =
+			FIELD_GET(HAL_REO_FLUSH_TIMEOUT_STATUS_INFO0_LIST_EMPTY,
+				  desc->info0);
+
+	status->u.timeout_list.release_desc_cnt =
+		FIELD_GET(HAL_REO_FLUSH_TIMEOUT_STATUS_INFO1_REL_DESC_COUNT,
+			  desc->info1);
+	status->u.timeout_list.fwd_buf_cnt =
+		FIELD_GET(HAL_REO_FLUSH_TIMEOUT_STATUS_INFO1_FWD_BUF_COUNT,
+			  desc->info1);
+}
+
+void ath11k_hal_reo_desc_thresh_reached_status(struct ath11k_base *ab,
+					       u32 *reo_desc,
+					       struct hal_reo_status *status)
+{
+	struct hal_tlv_hdr *tlv = (struct hal_tlv_hdr *)reo_desc;
+	struct hal_reo_desc_thresh_reached_status *desc =
+		(struct hal_reo_desc_thresh_reached_status *)tlv->value;
+
+	status->uniform_hdr.cmd_num =
+				FIELD_GET(HAL_REO_STATUS_HDR_INFO0_STATUS_NUM,
+					  desc->hdr.info0);
+	status->uniform_hdr.cmd_status =
+				FIELD_GET(HAL_REO_STATUS_HDR_INFO0_EXEC_STATUS,
+					  desc->hdr.info0);
+
+	status->u.desc_thresh_reached.threshold_idx =
+		FIELD_GET(HAL_REO_DESC_THRESH_STATUS_INFO0_THRESH_INDEX,
+			  desc->info0);
+
+	status->u.desc_thresh_reached.link_desc_counter0 =
+		FIELD_GET(HAL_REO_DESC_THRESH_STATUS_INFO1_LINK_DESC_COUNTER0,
+			  desc->info1);
+
+	status->u.desc_thresh_reached.link_desc_counter1 =
+		FIELD_GET(HAL_REO_DESC_THRESH_STATUS_INFO2_LINK_DESC_COUNTER1,
+			  desc->info2);
+
+	status->u.desc_thresh_reached.link_desc_counter2 =
+		FIELD_GET(HAL_REO_DESC_THRESH_STATUS_INFO3_LINK_DESC_COUNTER2,
+			  desc->info3);
+
+	status->u.desc_thresh_reached.link_desc_counter_sum =
+		FIELD_GET(HAL_REO_DESC_THRESH_STATUS_INFO4_LINK_DESC_COUNTER_SUM,
+			  desc->info4);
+}
+
+void ath11k_hal_reo_update_rx_reo_queue_status(struct ath11k_base *ab,
+					       u32 *reo_desc,
+					       struct hal_reo_status *status)
+{
+	struct hal_tlv_hdr *tlv = (struct hal_tlv_hdr *)reo_desc;
+	struct hal_reo_status_hdr *desc =
+		(struct hal_reo_status_hdr *)tlv->value;
+
+	status->uniform_hdr.cmd_num =
+				FIELD_GET(HAL_REO_STATUS_HDR_INFO0_STATUS_NUM,
+					  desc->info0);
+	status->uniform_hdr.cmd_status =
+				FIELD_GET(HAL_REO_STATUS_HDR_INFO0_EXEC_STATUS,
+					  desc->info0);
+}
+
+u32 ath11k_hal_reo_qdesc_size(u32 ba_window_size, u8 tid)
+{
+	u32 num_ext_desc;
+
+	if (ba_window_size <= 1) {
+		if (tid != HAL_DESC_REO_NON_QOS_TID)
+			num_ext_desc = 1;
+		else
+			num_ext_desc = 0;
+	} else if (ba_window_size <= 105) {
+		num_ext_desc = 1;
+	} else if (ba_window_size <= 210) {
+		num_ext_desc = 2;
+	} else {
+		num_ext_desc = 3;
+	}
+
+	return sizeof(struct hal_rx_reo_queue) +
+		(num_ext_desc * sizeof(struct hal_rx_reo_queue_ext));
+}
+
+void ath11k_hal_reo_qdesc_setup(void *vaddr, int tid, u32 ba_window_size,
+				u32 start_seq)
+{
+	struct hal_rx_reo_queue *qdesc = (struct hal_rx_reo_queue *)vaddr;
+	struct hal_rx_reo_queue_ext *ext_desc;
+
+	memset(qdesc, 0, sizeof(*qdesc));
+
+	ath11k_hal_reo_set_desc_hdr(&qdesc->desc_hdr, HAL_DESC_REO_OWNED,
+				    HAL_DESC_REO_QUEUE_DESC,
+				    REO_QUEUE_DESC_MAGIC_DEBUG_PATTERN_0);
+
+	qdesc->rx_queue_num = FIELD_PREP(HAL_RX_REO_QUEUE_RX_QUEUE_NUMBER, tid);
+
+	qdesc->info0 =
+		FIELD_PREP(HAL_RX_REO_QUEUE_INFO0_VLD, 1) |
+		FIELD_PREP(HAL_RX_REO_QUEUE_INFO0_ASSOC_LNK_DESC_COUNTER, 1) |
+		FIELD_PREP(HAL_RX_REO_QUEUE_INFO0_AC, ath11k_tid_to_ac(tid));
+
+	if (ba_window_size < 1)
+		ba_window_size = 1;
+
+	if (ba_window_size == 1 && tid != HAL_DESC_REO_NON_QOS_TID)
+		ba_window_size++;
+
+	if (ba_window_size == 1)
+		qdesc->info0 |= FIELD_PREP(HAL_RX_REO_QUEUE_INFO0_RETRY, 1);
+
+	qdesc->info0 |= FIELD_PREP(HAL_RX_REO_QUEUE_INFO0_BA_WINDOW_SIZE,
+				   ba_window_size - 1);
+
+	/* TODO: Set Ignore ampdu flags based on BA window size and/or
+	 * AMPDU capabilities
+	 */
+	qdesc->info0 |= FIELD_PREP(HAL_RX_REO_QUEUE_INFO0_IGNORE_AMPDU_FLG, 1);
+
+	qdesc->info1 |= FIELD_PREP(HAL_RX_REO_QUEUE_INFO1_SVLD, 0);
+
+	if (start_seq <= 0xfff)
+		qdesc->info1 = FIELD_PREP(HAL_RX_REO_QUEUE_INFO1_SSN,
+					  start_seq);
+
+	if (tid == HAL_DESC_REO_NON_QOS_TID)
+		return;
+
+	ext_desc = qdesc->ext_desc;
+
+	/* TODO: HW queue descriptors are currently allocated for max BA
+	 * window size for all QOS TIDs so that same descriptor can be used
+	 * later when ADDBA request is recevied. This should be changed to
+	 * allocate HW queue descriptors based on BA window size being
+	 * negotiated (0 for non BA cases), and reallocate when BA window
+	 * size changes and also send WMI message to FW to change the REO
+	 * queue descriptor in Rx peer entry as part of dp_rx_tid_update.
+	 */
+	memset(ext_desc, 0, 3 * sizeof(*ext_desc));
+	ath11k_hal_reo_set_desc_hdr(&ext_desc->desc_hdr, HAL_DESC_REO_OWNED,
+				    HAL_DESC_REO_QUEUE_EXT_DESC,
+				    REO_QUEUE_DESC_MAGIC_DEBUG_PATTERN_1);
+	ext_desc++;
+	ath11k_hal_reo_set_desc_hdr(&ext_desc->desc_hdr, HAL_DESC_REO_OWNED,
+				    HAL_DESC_REO_QUEUE_EXT_DESC,
+				    REO_QUEUE_DESC_MAGIC_DEBUG_PATTERN_2);
+	ext_desc++;
+	ath11k_hal_reo_set_desc_hdr(&ext_desc->desc_hdr, HAL_DESC_REO_OWNED,
+				    HAL_DESC_REO_QUEUE_EXT_DESC,
+				    REO_QUEUE_DESC_MAGIC_DEBUG_PATTERN_3);
+}
+
+void ath11k_hal_reo_init_cmd_ring(struct ath11k_base *ab,
+				  struct hal_srng *srng)
+{
+	struct hal_srng_params params;
+	struct hal_tlv_hdr *tlv;
+	struct hal_reo_get_queue_stats *desc;
+	int i, cmd_num = 1;
+	int entry_size;
+	u8 *entry;
+
+	memset(&params, 0, sizeof(params));
+
+	entry_size = ath11k_hal_srng_get_entrysize(HAL_REO_CMD);
+	ath11k_hal_srng_get_params(ab, srng, &params);
+	entry = (u8 *)params.ring_base_vaddr;
+
+	for (i = 0; i < params.num_entries; i++) {
+		tlv = (struct hal_tlv_hdr *)entry;
+		desc = (struct hal_reo_get_queue_stats *)tlv->value;
+		desc->cmd.info0 =
+			FIELD_PREP(HAL_REO_CMD_HDR_INFO0_CMD_NUMBER, cmd_num++);
+		entry += entry_size;
+	}
+}
+
+void ath11k_hal_reo_hw_setup(struct ath11k_base *ab)
+{
+	u32 reo_base = HAL_SEQ_WCSS_UMAC_REO_REG;
+	u32 val;
+
+	val = ath11k_ahb_read32(ab, reo_base + HAL_REO1_GEN_ENABLE);
+
+	val &= ~HAL_REO1_GEN_ENABLE_FRAG_DST_RING;
+	val |= FIELD_PREP(HAL_REO1_GEN_ENABLE_FRAG_DST_RING,
+			  HAL_SRNG_RING_ID_REO2SW1) |
+	       FIELD_PREP(HAL_REO1_GEN_ENABLE_AGING_LIST_ENABLE, 1) |
+	       FIELD_PREP(HAL_REO1_GEN_ENABLE_AGING_FLUSH_ENABLE, 1);
+	ath11k_ahb_write32(ab, reo_base + HAL_REO1_GEN_ENABLE, val);
+
+	ath11k_ahb_write32(ab, reo_base + HAL_REO1_AGING_THRESH_IX_0,
+			   HAL_DEFAULT_REO_TIMEOUT_USEC);
+	ath11k_ahb_write32(ab, reo_base + HAL_REO1_AGING_THRESH_IX_1,
+			   HAL_DEFAULT_REO_TIMEOUT_USEC);
+	ath11k_ahb_write32(ab, reo_base + HAL_REO1_AGING_THRESH_IX_2,
+			   HAL_DEFAULT_REO_TIMEOUT_USEC);
+	ath11k_ahb_write32(ab, reo_base + HAL_REO1_AGING_THRESH_IX_3,
+			   HAL_DEFAULT_REO_TIMEOUT_USEC);
+}
+
+static enum hal_rx_mon_status
+ath11k_hal_rx_parse_mon_status_tlv(struct ath11k_base *ab,
+				   struct hal_rx_mon_ppdu_info *ppdu_info,
+				   u32 tlv_tag, u8 *tlv_data)
+{
+	u32 info0, info1;
+
+	switch (tlv_tag) {
+	case HAL_RX_PPDU_START: {
+		struct hal_rx_ppdu_start *ppdu_start =
+			(struct hal_rx_ppdu_start *)tlv_data;
+
+		ppdu_info->ppdu_id =
+			FIELD_GET(HAL_RX_PPDU_START_INFO0_PPDU_ID,
+				  __le32_to_cpu(ppdu_start->info0));
+		ppdu_info->chan_num = __le32_to_cpu(ppdu_start->chan_num);
+		ppdu_info->ppdu_ts = __le32_to_cpu(ppdu_start->ppdu_start_ts);
+		break;
+	}
+	case HAL_RX_PPDU_END_USER_STATS: {
+		struct hal_rx_ppdu_end_user_stats *eu_stats =
+			(struct hal_rx_ppdu_end_user_stats *)tlv_data;
+
+		info0 = __le32_to_cpu(eu_stats->info0);
+		info1 = __le32_to_cpu(eu_stats->info1);
+
+		ppdu_info->tid =
+			ffs(FIELD_GET(HAL_RX_PPDU_END_USER_STATS_INFO6_TID_BITMAP,
+				      __le32_to_cpu(eu_stats->info6))) - 1;
+		ppdu_info->tcp_msdu_count =
+			FIELD_GET(HAL_RX_PPDU_END_USER_STATS_INFO4_TCP_MSDU_CNT,
+				  __le32_to_cpu(eu_stats->info4));
+		ppdu_info->udp_msdu_count =
+			FIELD_GET(HAL_RX_PPDU_END_USER_STATS_INFO4_UDP_MSDU_CNT,
+				  __le32_to_cpu(eu_stats->info4));
+		ppdu_info->other_msdu_count =
+			FIELD_GET(HAL_RX_PPDU_END_USER_STATS_INFO5_OTHER_MSDU_CNT,
+				  __le32_to_cpu(eu_stats->info5));
+		ppdu_info->tcp_ack_msdu_count =
+			FIELD_GET(HAL_RX_PPDU_END_USER_STATS_INFO5_TCP_ACK_MSDU_CNT,
+				  __le32_to_cpu(eu_stats->info5));
+		ppdu_info->preamble_type =
+			FIELD_GET(HAL_RX_PPDU_END_USER_STATS_INFO1_PKT_TYPE, info1);
+		ppdu_info->num_mpdu_fcs_ok =
+			FIELD_GET(HAL_RX_PPDU_END_USER_STATS_INFO1_MPDU_CNT_FCS_OK,
+				  info1);
+		ppdu_info->num_mpdu_fcs_err =
+			FIELD_GET(HAL_RX_PPDU_END_USER_STATS_INFO0_MPDU_CNT_FCS_ERR,
+				  info0);
+		break;
+	}
+	case HAL_PHYRX_HT_SIG: {
+		struct hal_rx_ht_sig_info *ht_sig =
+			(struct hal_rx_ht_sig_info *)tlv_data;
+
+		info0 = __le32_to_cpu(ht_sig->info0);
+		info1 = __le32_to_cpu(ht_sig->info1);
+
+		ppdu_info->mcs = FIELD_GET(HAL_RX_HT_SIG_INFO_INFO0_MCS, info0);
+		ppdu_info->bw = FIELD_GET(HAL_RX_HT_SIG_INFO_INFO0_BW, info0);
+		ppdu_info->is_stbc = FIELD_GET(HAL_RX_HT_SIG_INFO_INFO1_STBC,
+					       info1);
+		ppdu_info->ldpc = FIELD_GET(HAL_RX_HT_SIG_INFO_INFO1_FEC_CODING, info1);
+		ppdu_info->gi = info1 & HAL_RX_HT_SIG_INFO_INFO1_GI;
+
+		switch (ppdu_info->mcs) {
+		case 0 ... 7:
+			ppdu_info->nss = 1;
+			break;
+		case 8 ... 15:
+			ppdu_info->nss = 2;
+			break;
+		case 16 ... 23:
+			ppdu_info->nss = 3;
+			break;
+		case 24 ... 31:
+			ppdu_info->nss = 4;
+			break;
+		}
+
+		if (ppdu_info->nss > 1)
+			ppdu_info->mcs = ppdu_info->mcs % 8;
+
+		ppdu_info->reception_type = HAL_RX_RECEPTION_TYPE_SU;
+		break;
+	}
+	case HAL_PHYRX_L_SIG_B: {
+		struct hal_rx_lsig_b_info *lsigb =
+			(struct hal_rx_lsig_b_info *)tlv_data;
+
+		ppdu_info->rate = FIELD_GET(HAL_RX_LSIG_B_INFO_INFO0_RATE,
+					    __le32_to_cpu(lsigb->info0));
+		ppdu_info->reception_type = HAL_RX_RECEPTION_TYPE_SU;
+		break;
+	}
+	case HAL_PHYRX_L_SIG_A: {
+		struct hal_rx_lsig_a_info *lsiga =
+			(struct hal_rx_lsig_a_info *)tlv_data;
+
+		ppdu_info->rate = FIELD_GET(HAL_RX_LSIG_A_INFO_INFO0_RATE,
+					    __le32_to_cpu(lsiga->info0));
+		ppdu_info->reception_type = HAL_RX_RECEPTION_TYPE_SU;
+		break;
+	}
+	case HAL_PHYRX_VHT_SIG_A: {
+		struct hal_rx_vht_sig_a_info *vht_sig =
+			(struct hal_rx_vht_sig_a_info *)tlv_data;
+		u32 nsts;
+		u32 group_id;
+
+		info0 = __le32_to_cpu(vht_sig->info0);
+		info1 = __le32_to_cpu(vht_sig->info1);
+
+		ppdu_info->ldpc = FIELD_GET(HAL_RX_VHT_SIG_A_INFO_INFO1_SU_MU_CODING,
+					    info0);
+		ppdu_info->mcs = FIELD_GET(HAL_RX_VHT_SIG_A_INFO_INFO1_MCS,
+					   info1);
+		ppdu_info->gi =
+			FIELD_GET(HAL_RX_VHT_SIG_A_INFO_INFO1_GI_SETTING,
+				  info1);
+		ppdu_info->is_stbc = info0 & HAL_RX_VHT_SIG_A_INFO_INFO0_STBC;
+		nsts = FIELD_GET(HAL_RX_VHT_SIG_A_INFO_INFO0_NSTS, info0);
+		if (ppdu_info->is_stbc && nsts > 0)
+			nsts = ((nsts + 1) >> 1) - 1;
+
+		ppdu_info->nss = (nsts & VHT_SIG_SU_NSS_MASK) + 1;
+		ppdu_info->bw = FIELD_GET(HAL_RX_VHT_SIG_A_INFO_INFO0_BW,
+					  info0);
+		ppdu_info->beamformed = info1 &
+					HAL_RX_VHT_SIG_A_INFO_INFO1_BEAMFORMED;
+		group_id = FIELD_GET(HAL_RX_VHT_SIG_A_INFO_INFO0_GROUP_ID,
+				     info0);
+		if (group_id == 0 || group_id == 63)
+			ppdu_info->reception_type = HAL_RX_RECEPTION_TYPE_SU;
+		else
+			ppdu_info->reception_type =
+				HAL_RX_RECEPTION_TYPE_MU_MIMO;
+		break;
+	}
+	case HAL_PHYRX_HE_SIG_A_SU: {
+		struct hal_rx_he_sig_a_su_info *he_sig_a =
+			(struct hal_rx_he_sig_a_su_info *)tlv_data;
+		u32 nsts, cp_ltf, dcm;
+
+		info0 = __le32_to_cpu(he_sig_a->info0);
+		info1 = __le32_to_cpu(he_sig_a->info1);
+
+		ppdu_info->mcs =
+			FIELD_GET(HAL_RX_HE_SIG_A_SU_INFO_INFO0_TRANSMIT_MCS,
+				  info0);
+		ppdu_info->bw =
+			FIELD_GET(HAL_RX_HE_SIG_A_SU_INFO_INFO0_TRANSMIT_BW,
+				  info0);
+		ppdu_info->ldpc = FIELD_GET(HAL_RX_HE_SIG_A_SU_INFO_INFO1_CODING, info0);
+		ppdu_info->is_stbc = info1 &
+				     HAL_RX_HE_SIG_A_SU_INFO_INFO1_STBC;
+		ppdu_info->beamformed = info1 &
+					HAL_RX_HE_SIG_A_SU_INFO_INFO1_TXBF;
+		dcm = info0 & HAL_RX_HE_SIG_A_SU_INFO_INFO0_DCM;
+		cp_ltf = FIELD_GET(HAL_RX_HE_SIG_A_SU_INFO_INFO0_CP_LTF_SIZE,
+				   info0);
+		nsts = FIELD_GET(HAL_RX_HE_SIG_A_SU_INFO_INFO0_NSTS, info0);
+
+		switch (cp_ltf) {
+		case 0:
+		case 1:
+			ppdu_info->gi = HAL_RX_GI_0_8_US;
+			break;
+		case 2:
+			ppdu_info->gi = HAL_RX_GI_1_6_US;
+			break;
+		case 3:
+			if (dcm && ppdu_info->is_stbc)
+				ppdu_info->gi = HAL_RX_GI_0_8_US;
+			else
+				ppdu_info->gi = HAL_RX_GI_3_2_US;
+			break;
+		}
+
+		ppdu_info->nss = nsts + 1;
+		ppdu_info->reception_type = HAL_RX_RECEPTION_TYPE_SU;
+		break;
+	}
+	case HAL_PHYRX_HE_SIG_A_MU_DL: {
+		struct hal_rx_he_sig_a_mu_dl_info *he_sig_a_mu_dl =
+			(struct hal_rx_he_sig_a_mu_dl_info *)tlv_data;
+
+		u32 cp_ltf;
+
+		info0 = __le32_to_cpu(he_sig_a_mu_dl->info0);
+		info1 = __le32_to_cpu(he_sig_a_mu_dl->info1);
+
+		ppdu_info->bw =
+			FIELD_GET(HAL_RX_HE_SIG_A_MU_DL_INFO_INFO0_TRANSMIT_BW,
+				  info0);
+		cp_ltf = FIELD_GET(HAL_RX_HE_SIG_A_MU_DL_INFO_INFO0_CP_LTF_SIZE,
+				   info0);
+
+		switch (cp_ltf) {
+		case 0:
+		case 1:
+			ppdu_info->gi = HAL_RX_GI_0_8_US;
+			break;
+		case 2:
+			ppdu_info->gi = HAL_RX_GI_1_6_US;
+			break;
+		case 3:
+			ppdu_info->gi = HAL_RX_GI_3_2_US;
+			break;
+		}
+
+		ppdu_info->is_stbc = info1 &
+				     HAL_RX_HE_SIG_A_MU_DL_INFO_INFO1_STBC;
+		ppdu_info->reception_type = HAL_RX_RECEPTION_TYPE_MU_MIMO;
+		break;
+	}
+	case HAL_PHYRX_HE_SIG_B1_MU: {
+		/* TODO: Check if resource unit(RU) allocation stats
+		 * are required
+		 */
+		ppdu_info->reception_type = HAL_RX_RECEPTION_TYPE_MU_MIMO;
+		break;
+	}
+	case HAL_PHYRX_HE_SIG_B2_MU: {
+		struct hal_rx_he_sig_b2_mu_info *he_sig_b2_mu =
+			(struct hal_rx_he_sig_b2_mu_info *)tlv_data;
+
+		info0 = __le32_to_cpu(he_sig_b2_mu->info0);
+
+		ppdu_info->mcs =
+			FIELD_GET(HAL_RX_HE_SIG_B2_MU_INFO_INFO0_STA_MCS,
+				  info0);
+		ppdu_info->nss =
+			FIELD_GET(HAL_RX_HE_SIG_B2_MU_INFO_INFO0_STA_NSTS,
+				  info0) + 1;
+		ppdu_info->ldpc = FIELD_GET(HAL_RX_HE_SIG_B2_MU_INFO_INFO0_STA_CODING,
+					    info0);
+		break;
+	}
+	case HAL_PHYRX_HE_SIG_B2_OFDMA: {
+		struct hal_rx_he_sig_b2_ofdma_info *he_sig_b2_ofdma =
+			(struct hal_rx_he_sig_b2_ofdma_info *)tlv_data;
+
+		info0 = __le32_to_cpu(he_sig_b2_ofdma->info0);
+
+		ppdu_info->mcs =
+			FIELD_GET(HAL_RX_HE_SIG_B2_OFDMA_INFO_INFO0_STA_MCS,
+				  info0);
+		ppdu_info->nss =
+			FIELD_GET(HAL_RX_HE_SIG_B2_OFDMA_INFO_INFO0_STA_NSTS,
+				  info0) + 1;
+		ppdu_info->beamformed =
+			info0 &
+			HAL_RX_HE_SIG_B2_OFDMA_INFO_INFO0_STA_TXBF;
+		ppdu_info->ldpc = FIELD_GET(HAL_RX_HE_SIG_B2_OFDMA_INFO_INFO0_STA_CODING,
+					    info0);
+		ppdu_info->reception_type = HAL_RX_RECEPTION_TYPE_MU_OFDMA;
+		break;
+	}
+	case HAL_PHYRX_RSSI_LEGACY: {
+		struct hal_rx_phyrx_rssi_legacy_info *rssi =
+			(struct hal_rx_phyrx_rssi_legacy_info *)tlv_data;
+
+		/* TODO: Please note that the combined rssi will not be accurate
+		 * in MU case. Rssi in MU needs to be retrieved from
+		 * PHYRX_OTHER_RECEIVE_INFO TLV.
+		 */
+		ppdu_info->rssi_comb =
+			FIELD_GET(HAL_RX_PHYRX_RSSI_LEGACY_INFO_INFO1_RSSI_COMB,
+				  __le32_to_cpu(rssi->info0));
+		break;
+	}
+	case HAL_RX_MPDU_START: {
+		struct hal_rx_mpdu_info *mpdu_info =
+			(struct hal_rx_mpdu_info *)tlv_data;
+		u16 peer_id;
+
+		peer_id = FIELD_GET(HAL_RX_MPDU_INFO_INFO0_PEERID,
+				    __le32_to_cpu(mpdu_info->info0));
+		if (peer_id)
+			ppdu_info->peer_id = peer_id;
+		break;
+	}
+	case HAL_RXPCU_PPDU_END_INFO: {
+		struct hal_rx_ppdu_end_duration *ppdu_rx_duration =
+			(struct hal_rx_ppdu_end_duration *)tlv_data;
+		ppdu_info->rx_duration =
+			FIELD_GET(HAL_RX_PPDU_END_DURATION,
+				  __le32_to_cpu(ppdu_rx_duration->info0));
+		break;
+	}
+	case HAL_DUMMY:
+		return HAL_RX_MON_STATUS_BUF_DONE;
+	case HAL_RX_PPDU_END_STATUS_DONE:
+	case 0:
+		return HAL_RX_MON_STATUS_PPDU_DONE;
+	default:
+		break;
+	}
+
+	return HAL_RX_MON_STATUS_PPDU_NOT_DONE;
+}
+
+enum hal_rx_mon_status
+ath11k_hal_rx_parse_mon_status(struct ath11k_base *ab,
+			       struct hal_rx_mon_ppdu_info *ppdu_info,
+			       struct sk_buff *skb)
+{
+	struct hal_tlv_hdr *tlv;
+	enum hal_rx_mon_status hal_status = HAL_RX_MON_STATUS_BUF_DONE;
+	u16 tlv_tag;
+	u16 tlv_len;
+	u8 *ptr = skb->data;
+
+	do {
+		tlv = (struct hal_tlv_hdr *)ptr;
+		tlv_tag = FIELD_GET(HAL_TLV_HDR_TAG, tlv->tl);
+		tlv_len = FIELD_GET(HAL_TLV_HDR_LEN, tlv->tl);
+		ptr += sizeof(*tlv);
+
+		/* The actual length of PPDU_END is the combined length of many PHY
+		 * TLVs that follow. Skip the TLV header and
+		 * rx_rxpcu_classification_overview that follows the header to get to
+		 * next TLV.
+		 */
+		if (tlv_tag == HAL_RX_PPDU_END)
+			tlv_len = sizeof(struct hal_rx_rxpcu_classification_overview);
+
+		hal_status = ath11k_hal_rx_parse_mon_status_tlv(ab, ppdu_info,
+								tlv_tag, ptr);
+		ptr += tlv_len;
+		ptr = PTR_ALIGN(ptr, HAL_TLV_ALIGN);
+
+		if ((ptr - skb->data) >= DP_RX_BUFFER_SIZE)
+			break;
+	} while (hal_status == HAL_RX_MON_STATUS_PPDU_NOT_DONE);
+
+	return hal_status;
+}
+
+void ath11k_hal_rx_reo_ent_buf_paddr_get(void *rx_desc, dma_addr_t *paddr,
+					 u32 *sw_cookie, void **pp_buf_addr,
+					 u32  *msdu_cnt)
+{
+	struct hal_reo_entrance_ring *reo_ent_ring =
+		(struct hal_reo_entrance_ring *)rx_desc;
+	struct ath11k_buffer_addr *buf_addr_info;
+	struct rx_mpdu_desc *rx_mpdu_desc_info_details;
+
+	rx_mpdu_desc_info_details =
+			(struct rx_mpdu_desc *)&reo_ent_ring->rx_mpdu_info;
+
+	*msdu_cnt = FIELD_GET(RX_MPDU_DESC_INFO0_MSDU_COUNT,
+			      rx_mpdu_desc_info_details->info0);
+
+	buf_addr_info = (struct ath11k_buffer_addr *)&reo_ent_ring->buf_addr_info;
+
+	*paddr = (((u64)FIELD_GET(BUFFER_ADDR_INFO1_ADDR,
+				  buf_addr_info->info1)) << 32) |
+			FIELD_GET(BUFFER_ADDR_INFO0_ADDR,
+				  buf_addr_info->info0);
+
+	*sw_cookie = FIELD_GET(BUFFER_ADDR_INFO1_SW_COOKIE,
+			       buf_addr_info->info1);
+
+	*pp_buf_addr = (void *)buf_addr_info;
+}
diff --git a/drivers/net/wireless/ath/ath11k/hal_rx.h b/drivers/net/wireless/ath/ath11k/hal_rx.h
new file mode 100644
index 000000000000..96555e4f4ef3
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/hal_rx.h
@@ -0,0 +1,352 @@
+/* SPDX-License-Identifier: BSD-3-Clause-Clear */
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#ifndef ATH11K_HAL_RX_H
+#define ATH11K_HAL_RX_H
+
+struct hal_rx_mpdu_meta {
+	u32 peer_meta;
+	u16 msdu_cnt;
+	u16 seq_num;
+	bool frag;
+	bool retry;
+	bool ampdu;
+	bool raw;
+};
+
+struct hal_rx_msdu_meta {
+	u32 cookie;
+	u32 msdu_len;
+	bool first;
+	bool last;
+	bool continuation;
+};
+
+struct hal_rx_meta_info {
+	enum hal_reo_dest_ring_push_reason push_reason;
+	struct hal_rx_mpdu_meta mpdu_meta;
+	struct hal_rx_msdu_meta msdu_meta;
+};
+
+struct hal_rx_wbm_rel_info {
+	u32 cookie;
+	enum hal_wbm_rel_src_module err_rel_src;
+	enum hal_reo_dest_ring_push_reason push_reason;
+	u32 err_code;
+	bool first_msdu;
+	bool last_msdu;
+};
+
+#define HAL_INVALID_PEERID 0xffff
+#define VHT_SIG_SU_NSS_MASK 0x7
+
+#define HAL_RX_MAX_MCS 12
+#define HAL_RX_MAX_NSS 8
+
+struct hal_rx_mon_status_tlv_hdr {
+	u32 hdr;
+	u8 value[0];
+};
+
+enum hal_rx_su_mu_coding {
+	HAL_RX_SU_MU_CODING_BCC,
+	HAL_RX_SU_MU_CODING_LDPC,
+	HAL_RX_SU_MU_CODING_MAX,
+};
+
+enum hal_rx_gi {
+	HAL_RX_GI_0_8_US,
+	HAL_RX_GI_0_4_US,
+	HAL_RX_GI_1_6_US,
+	HAL_RX_GI_3_2_US,
+	HAL_RX_GI_MAX,
+};
+
+enum hal_rx_bw {
+	HAL_RX_BW_20MHZ,
+	HAL_RX_BW_40MHZ,
+	HAL_RX_BW_80MHZ,
+	HAL_RX_BW_160MHZ,
+	HAL_RX_BW_MAX,
+};
+
+enum hal_rx_preamble {
+	HAL_RX_PREAMBLE_11A,
+	HAL_RX_PREAMBLE_11B,
+	HAL_RX_PREAMBLE_11N,
+	HAL_RX_PREAMBLE_11AC,
+	HAL_RX_PREAMBLE_11AX,
+	HAL_RX_PREAMBLE_MAX,
+};
+
+enum hal_rx_reception_type {
+	HAL_RX_RECEPTION_TYPE_SU,
+	HAL_RX_RECEPTION_TYPE_MU_MIMO,
+	HAL_RX_RECEPTION_TYPE_MU_OFDMA,
+	HAL_RX_RECEPTION_TYPE_MU_OFDMA_MIMO,
+	HAL_RX_RECEPTION_TYPE_MAX,
+};
+
+#define HAL_TLV_STATUS_PPDU_NOT_DONE            0
+#define HAL_TLV_STATUS_PPDU_DONE                1
+#define HAL_TLV_STATUS_BUF_DONE                 2
+#define HAL_TLV_STATUS_PPDU_NON_STD_DONE        3
+#define HAL_RX_FCS_LEN                          4
+
+enum hal_rx_mon_status {
+	HAL_RX_MON_STATUS_PPDU_NOT_DONE,
+	HAL_RX_MON_STATUS_PPDU_DONE,
+	HAL_RX_MON_STATUS_BUF_DONE,
+};
+
+struct hal_rx_mon_ppdu_info {
+	u32 ppdu_id;
+	u32 ppdu_ts;
+	u32 num_mpdu_fcs_ok;
+	u32 num_mpdu_fcs_err;
+	u32 preamble_type;
+	u16 chan_num;
+	u16 tcp_msdu_count;
+	u16 tcp_ack_msdu_count;
+	u16 udp_msdu_count;
+	u16 other_msdu_count;
+	u16 peer_id;
+	u8 rate;
+	u8 mcs;
+	u8 nss;
+	u8 bw;
+	u8 is_stbc;
+	u8 gi;
+	u8 ldpc;
+	u8 beamformed;
+	u8 rssi_comb;
+	u8 tid;
+	u8 reception_type;
+	u64 rx_duration;
+};
+
+#define HAL_RX_PPDU_START_INFO0_PPDU_ID		GENMASK(15, 0)
+
+struct hal_rx_ppdu_start {
+	__le32 info0;
+	__le32 chan_num;
+	__le32 ppdu_start_ts;
+} __packed;
+
+#define HAL_RX_PPDU_END_USER_STATS_INFO0_MPDU_CNT_FCS_ERR	GENMASK(25, 16)
+
+#define HAL_RX_PPDU_END_USER_STATS_INFO1_MPDU_CNT_FCS_OK	GENMASK(8, 0)
+#define HAL_RX_PPDU_END_USER_STATS_INFO1_FC_VALID		BIT(9)
+#define HAL_RX_PPDU_END_USER_STATS_INFO1_QOS_CTRL_VALID		BIT(10)
+#define HAL_RX_PPDU_END_USER_STATS_INFO1_HT_CTRL_VALID		BIT(11)
+#define HAL_RX_PPDU_END_USER_STATS_INFO1_PKT_TYPE		GENMASK(23, 20)
+
+#define HAL_RX_PPDU_END_USER_STATS_INFO2_AST_INDEX		GENMASK(15, 0)
+#define HAL_RX_PPDU_END_USER_STATS_INFO2_FRAME_CTRL		GENMASK(31, 16)
+
+#define HAL_RX_PPDU_END_USER_STATS_INFO3_QOS_CTRL		GENMASK(31, 16)
+
+#define HAL_RX_PPDU_END_USER_STATS_INFO4_UDP_MSDU_CNT		GENMASK(15, 0)
+#define HAL_RX_PPDU_END_USER_STATS_INFO4_TCP_MSDU_CNT		GENMASK(31, 16)
+
+#define HAL_RX_PPDU_END_USER_STATS_INFO5_OTHER_MSDU_CNT		GENMASK(15, 0)
+#define HAL_RX_PPDU_END_USER_STATS_INFO5_TCP_ACK_MSDU_CNT	GENMASK(31, 16)
+
+#define HAL_RX_PPDU_END_USER_STATS_INFO6_TID_BITMAP		GENMASK(15, 0)
+#define HAL_RX_PPDU_END_USER_STATS_INFO6_TID_EOSP_BITMAP	GENMASK(31, 16)
+
+struct hal_rx_ppdu_end_user_stats {
+	__le32 rsvd0[2];
+	__le32 info0;
+	__le32 info1;
+	__le32 info2;
+	__le32 info3;
+	__le32 ht_ctrl;
+	__le32 rsvd1[2];
+	__le32 info4;
+	__le32 info5;
+	__le32 info6;
+	__le32 rsvd2[11];
+} __packed;
+
+#define HAL_RX_HT_SIG_INFO_INFO0_MCS		GENMASK(6, 0)
+#define HAL_RX_HT_SIG_INFO_INFO0_BW		BIT(7)
+
+#define HAL_RX_HT_SIG_INFO_INFO1_STBC		GENMASK(5, 4)
+#define HAL_RX_HT_SIG_INFO_INFO1_FEC_CODING	BIT(6)
+#define HAL_RX_HT_SIG_INFO_INFO1_GI		BIT(7)
+
+struct hal_rx_ht_sig_info {
+	__le32 info0;
+	__le32 info1;
+} __packed;
+
+#define HAL_RX_LSIG_B_INFO_INFO0_RATE	GENMASK(3, 0)
+#define HAL_RX_LSIG_B_INFO_INFO0_LEN	GENMASK(15, 4)
+
+struct hal_rx_lsig_b_info {
+	__le32 info0;
+} __packed;
+
+#define HAL_RX_LSIG_A_INFO_INFO0_RATE		GENMASK(3, 0)
+#define HAL_RX_LSIG_A_INFO_INFO0_LEN		GENMASK(16, 5)
+#define HAL_RX_LSIG_A_INFO_INFO0_PKT_TYPE	GENMASK(27, 24)
+
+struct hal_rx_lsig_a_info {
+	__le32 info0;
+} __packed;
+
+#define HAL_RX_VHT_SIG_A_INFO_INFO0_BW		GENMASK(1, 0)
+#define HAL_RX_VHT_SIG_A_INFO_INFO0_STBC	BIT(3)
+#define HAL_RX_VHT_SIG_A_INFO_INFO0_GROUP_ID	GENMASK(9, 4)
+#define HAL_RX_VHT_SIG_A_INFO_INFO0_NSTS	GENMASK(21, 10)
+
+#define HAL_RX_VHT_SIG_A_INFO_INFO1_GI_SETTING		GENMASK(1, 0)
+#define HAL_RX_VHT_SIG_A_INFO_INFO1_SU_MU_CODING	BIT(2)
+#define HAL_RX_VHT_SIG_A_INFO_INFO1_MCS			GENMASK(7, 4)
+#define HAL_RX_VHT_SIG_A_INFO_INFO1_BEAMFORMED		BIT(8)
+
+struct hal_rx_vht_sig_a_info {
+	__le32 info0;
+	__le32 info1;
+} __packed;
+
+#define HAL_RX_HE_SIG_A_SU_INFO_INFO0_TRANSMIT_MCS	GENMASK(6, 3)
+#define HAL_RX_HE_SIG_A_SU_INFO_INFO0_DCM		BIT(7)
+#define HAL_RX_HE_SIG_A_SU_INFO_INFO0_TRANSMIT_BW	GENMASK(20, 19)
+#define HAL_RX_HE_SIG_A_SU_INFO_INFO0_CP_LTF_SIZE	GENMASK(22, 21)
+#define HAL_RX_HE_SIG_A_SU_INFO_INFO0_NSTS		GENMASK(25, 23)
+
+#define HAL_RX_HE_SIG_A_SU_INFO_INFO1_CODING		BIT(7)
+#define HAL_RX_HE_SIG_A_SU_INFO_INFO1_STBC		BIT(9)
+#define HAL_RX_HE_SIG_A_SU_INFO_INFO1_TXBF		BIT(10)
+
+struct hal_rx_he_sig_a_su_info {
+	__le32 info0;
+	__le32 info1;
+} __packed;
+
+#define HAL_RX_HE_SIG_A_MU_DL_INFO_INFO0_TRANSMIT_BW	GENMASK(17, 15)
+#define HAL_RX_HE_SIG_A_MU_DL_INFO_INFO0_CP_LTF_SIZE	GENMASK(24, 23)
+
+#define HAL_RX_HE_SIG_A_MU_DL_INFO_INFO1_STBC		BIT(12)
+
+struct hal_rx_he_sig_a_mu_dl_info {
+	__le32 info0;
+	__le32 info1;
+} __packed;
+
+#define HAL_RX_HE_SIG_B1_MU_INFO_INFO0_RU_ALLOCATION	GENMASK(7, 0)
+
+struct hal_rx_he_sig_b1_mu_info {
+	__le32 info0;
+} __packed;
+
+#define HAL_RX_HE_SIG_B2_MU_INFO_INFO0_STA_MCS		GENMASK(18, 15)
+#define HAL_RX_HE_SIG_B2_MU_INFO_INFO0_STA_CODING	BIT(20)
+#define HAL_RX_HE_SIG_B2_MU_INFO_INFO0_STA_NSTS		GENMASK(31, 29)
+
+struct hal_rx_he_sig_b2_mu_info {
+	__le32 info0;
+} __packed;
+
+#define HAL_RX_HE_SIG_B2_OFDMA_INFO_INFO0_STA_NSTS	GENMASK(13, 11)
+#define HAL_RX_HE_SIG_B2_OFDMA_INFO_INFO0_STA_TXBF	BIT(19)
+#define HAL_RX_HE_SIG_B2_OFDMA_INFO_INFO0_STA_MCS	GENMASK(18, 15)
+#define HAL_RX_HE_SIG_B2_OFDMA_INFO_INFO0_STA_DCM	BIT(19)
+#define HAL_RX_HE_SIG_B2_OFDMA_INFO_INFO0_STA_CODING	BIT(20)
+
+struct hal_rx_he_sig_b2_ofdma_info {
+	__le32 info0;
+} __packed;
+
+#define HAL_RX_PHYRX_RSSI_LEGACY_INFO_INFO1_RSSI_COMB	GENMASK(15, 8)
+
+struct hal_rx_phyrx_rssi_legacy_info {
+	__le32 rsvd[35];
+	__le32 info0;
+} __packed;
+
+#define HAL_RX_MPDU_INFO_INFO0_PEERID	GENMASK(31, 16)
+struct hal_rx_mpdu_info {
+	__le32 rsvd0;
+	__le32 info0;
+	__le32 rsvd1[21];
+} __packed;
+
+#define HAL_RX_PPDU_END_DURATION	GENMASK(23, 0)
+struct hal_rx_ppdu_end_duration {
+	__le32 rsvd0[9];
+	__le32 info0;
+	__le32 rsvd1[4];
+} __packed;
+
+struct hal_rx_rxpcu_classification_overview {
+	u32 rsvd0;
+} __packed;
+
+struct hal_rx_msdu_desc_info {
+	u32 msdu_flags;
+	u16 msdu_len; /* 14 bits for length */
+};
+
+#define HAL_RX_NUM_MSDU_DESC 6
+struct hal_rx_msdu_list {
+	struct hal_rx_msdu_desc_info msdu_info[HAL_RX_NUM_MSDU_DESC];
+	u32 sw_cookie[HAL_RX_NUM_MSDU_DESC];
+	u8 rbm[HAL_RX_NUM_MSDU_DESC];
+};
+
+void ath11k_hal_reo_status_queue_stats(struct ath11k_base *ab, u32 *reo_desc,
+				       struct hal_reo_status *status);
+void ath11k_hal_reo_flush_queue_status(struct ath11k_base *ab, u32 *reo_desc,
+				       struct hal_reo_status *status);
+void ath11k_hal_reo_flush_cache_status(struct ath11k_base *ab, u32 *reo_desc,
+				       struct hal_reo_status *status);
+void ath11k_hal_reo_flush_cache_status(struct ath11k_base *ab, u32 *reo_desc,
+				       struct hal_reo_status *status);
+void ath11k_hal_reo_unblk_cache_status(struct ath11k_base *ab, u32 *reo_desc,
+				       struct hal_reo_status *status);
+void ath11k_hal_reo_flush_timeout_list_status(struct ath11k_base *ab,
+					      u32 *reo_desc,
+					      struct hal_reo_status *status);
+void ath11k_hal_reo_desc_thresh_reached_status(struct ath11k_base *ab,
+					       u32 *reo_desc,
+					       struct hal_reo_status *status);
+void ath11k_hal_reo_update_rx_reo_queue_status(struct ath11k_base *ab,
+					       u32 *reo_desc,
+					       struct hal_reo_status *status);
+int ath11k_hal_reo_process_status(u8 *reo_desc, u8 *status);
+void ath11k_hal_rx_msdu_link_info_get(void *link_desc, u32 *num_msdus,
+				      struct hal_rx_msdu_meta *meta,
+				      enum hal_rx_buf_return_buf_manager *rbm);
+void ath11k_hal_rx_msdu_link_desc_set(struct ath11k_base *ab, void *desc,
+				      void *link_desc,
+				      enum hal_wbm_rel_bm_act action);
+void ath11k_hal_rx_buf_addr_info_set(void *desc, dma_addr_t paddr,
+				     u32 cookie, u8 manager);
+void ath11k_hal_rx_buf_addr_info_get(void *desc, dma_addr_t *paddr,
+				     u32 *cookie, u8 *rbm);
+int ath11k_hal_desc_reo_parse_err(struct ath11k_base *ab, u32 *rx_desc,
+				  dma_addr_t *paddr, u32 *desc_bank);
+void ath11k_hal_rx_parse_dst_ring_desc(struct ath11k_base *ab, u32 *rx_desc,
+				       struct hal_rx_meta_info *meta_info);
+int ath11k_hal_wbm_desc_parse_err(struct ath11k_base *ab, void *desc,
+				  struct hal_rx_wbm_rel_info *rel_info);
+void ath11k_hal_rx_reo_ent_paddr_get(struct ath11k_base *ab, void *desc,
+				     dma_addr_t *paddr, u32 *desc_bank);
+void ath11k_hal_rx_reo_ent_buf_paddr_get(void *rx_desc,
+					 dma_addr_t *paddr, u32 *sw_cookie,
+					 void **pp_buf_addr_info,
+					 u32 *msdu_cnt);
+enum hal_rx_mon_status
+ath11k_hal_rx_parse_mon_status(struct ath11k_base *ab,
+			       struct hal_rx_mon_ppdu_info *ppdu_info,
+			       struct sk_buff *skb);
+#define REO_QUEUE_DESC_MAGIC_DEBUG_PATTERN_0 0xDDBEEF
+#define REO_QUEUE_DESC_MAGIC_DEBUG_PATTERN_1 0xADBEEF
+#define REO_QUEUE_DESC_MAGIC_DEBUG_PATTERN_2 0xBDBEEF
+#define REO_QUEUE_DESC_MAGIC_DEBUG_PATTERN_3 0xCDBEEF
+#endif
diff --git a/drivers/net/wireless/ath/ath11k/hal_tx.c b/drivers/net/wireless/ath/ath11k/hal_tx.c
new file mode 100644
index 000000000000..bb27591023fe
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/hal_tx.c
@@ -0,0 +1,226 @@
+// SPDX-License-Identifier: BSD-3-Clause-Clear
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#include "ahb.h"
+#include "hal.h"
+#include "hal_tx.h"
+
+#define DSCP_TID_MAP_TBL_ENTRY_SIZE 64
+
+/* dscp_tid_map - Default DSCP-TID mapping
+ *
+ * DSCP        TID
+ * 000000      0
+ * 001000      1
+ * 010000      2
+ * 011000      3
+ * 100000      4
+ * 101000      5
+ * 110000      6
+ * 111000      7
+ */
+static const u8 dscp_tid_map[DSCP_TID_MAP_TBL_ENTRY_SIZE] = {
+	0, 0, 0, 0, 0, 0, 0, 0,
+	1, 1, 1, 1, 1, 1, 1, 1,
+	2, 2, 2, 2, 2, 2, 2, 2,
+	3, 3, 3, 3, 3, 3, 3, 3,
+	4, 4, 4, 4, 4, 4, 4, 4,
+	5, 5, 5, 5, 5, 5, 5, 5,
+	6, 6, 6, 6, 6, 6, 6, 6,
+	7, 7, 7, 7, 7, 7, 7, 7,
+};
+
+void ath11k_hal_tx_cmd_desc_setup(struct ath11k_base *ab, void *cmd,
+				  struct hal_tx_info *ti)
+{
+	struct hal_tcl_data_cmd *tcl_cmd = (struct hal_tcl_data_cmd *)cmd;
+
+	tcl_cmd->buf_addr_info.info0 =
+		FIELD_PREP(BUFFER_ADDR_INFO0_ADDR, ti->paddr);
+	tcl_cmd->buf_addr_info.info1 =
+		FIELD_PREP(BUFFER_ADDR_INFO1_ADDR,
+			   ((uint64_t)ti->paddr >> HAL_ADDR_MSB_REG_SHIFT));
+	tcl_cmd->buf_addr_info.info1 |=
+		FIELD_PREP(BUFFER_ADDR_INFO1_RET_BUF_MGR,
+			   (ti->ring_id + HAL_RX_BUF_RBM_SW0_BM)) |
+		FIELD_PREP(BUFFER_ADDR_INFO1_SW_COOKIE, ti->desc_id);
+
+	tcl_cmd->info0 =
+		FIELD_PREP(HAL_TCL_DATA_CMD_INFO0_DESC_TYPE, ti->type) |
+		FIELD_PREP(HAL_TCL_DATA_CMD_INFO0_ENCAP_TYPE, ti->encap_type) |
+		FIELD_PREP(HAL_TCL_DATA_CMD_INFO0_ENCRYPT_TYPE,
+			   ti->encrypt_type) |
+		FIELD_PREP(HAL_TCL_DATA_CMD_INFO0_ADDRX_EN,
+			   ti->addr_search_flags) |
+		FIELD_PREP(HAL_TCL_DATA_CMD_INFO0_ADDRY_EN,
+			   ti->addr_search_flags) |
+		FIELD_PREP(HAL_TCL_DATA_CMD_INFO0_CMD_NUM,
+			   ti->meta_data_flags);
+
+	tcl_cmd->info1 = ti->flags0 |
+		FIELD_PREP(HAL_TCL_DATA_CMD_INFO1_DATA_LEN, ti->data_len) |
+		FIELD_PREP(HAL_TCL_DATA_CMD_INFO1_PKT_OFFSET, ti->pkt_offset);
+
+	tcl_cmd->info2 = ti->flags1 |
+		FIELD_PREP(HAL_TCL_DATA_CMD_INFO2_TID, ti->tid) |
+		FIELD_PREP(HAL_TCL_DATA_CMD_INFO2_LMAC_ID, ti->lmac_id);
+
+	tcl_cmd->info3 = FIELD_PREP(HAL_TCL_DATA_CMD_INFO3_DSCP_TID_TABLE_IDX,
+				    ti->dscp_tid_tbl_idx) |
+			 FIELD_PREP(HAL_TCL_DATA_CMD_INFO3_SEARCH_INDEX,
+				    ti->bss_ast_hash);
+}
+
+/* Commit the descriptor to hardware */
+void ath11k_hal_tx_desc_sync(void *tx_desc_cached, void *hw_desc)
+{
+	memcpy(hw_desc + sizeof(struct hal_tlv_hdr), tx_desc_cached,
+	       sizeof(struct hal_tcl_data_cmd));
+}
+
+/* Get the descriptor status from hardware */
+void ath11k_hal_tx_status_desc_sync(void *hw_desc, void *local_desc)
+{
+	memcpy(local_desc, hw_desc, HAL_TX_STATUS_DESC_LEN);
+}
+
+void ath11k_hal_tx_status_parse(struct ath11k_base *ab,
+				struct hal_wbm_release_ring *desc,
+				struct hal_tx_status *ts)
+{
+	ts->buf_rel_source =
+		FIELD_GET(HAL_WBM_RELEASE_INFO0_REL_SRC_MODULE, desc->info0);
+	if (ts->buf_rel_source != HAL_WBM_REL_SRC_MODULE_FW &&
+	    ts->buf_rel_source != HAL_WBM_REL_SRC_MODULE_TQM)
+		return;
+
+	ts->desc_id = FIELD_GET(BUFFER_ADDR_INFO1_SW_COOKIE,
+				desc->buf_addr_info.info1);
+
+	if (ts->buf_rel_source == HAL_WBM_REL_SRC_MODULE_FW)
+		return;
+
+	ts->status = FIELD_GET(HAL_WBM_RELEASE_INFO0_TQM_RELEASE_REASON,
+			       desc->info0);
+	ts->ppdu_id = FIELD_GET(HAL_WBM_RELEASE_INFO1_TQM_STATUS_NUMBER,
+				desc->info1);
+	ts->try_cnt = FIELD_GET(HAL_WBM_RELEASE_INFO1_TRANSMIT_COUNT,
+				desc->info1);
+
+	ts->ack_rssi = FIELD_GET(HAL_WBM_RELEASE_INFO2_ACK_FRAME_RSSI,
+				 desc->info2);
+	if (desc->info2 & HAL_WBM_RELEASE_INFO2_FIRST_MSDU)
+		ts->flags |= HAL_TX_STATUS_FLAGS_FIRST_MSDU;
+
+	if (desc->info2 & HAL_WBM_RELEASE_INFO2_LAST_MSDU)
+		ts->flags |= HAL_TX_STATUS_FLAGS_LAST_MSDU;
+
+	if (desc->info2 & HAL_WBM_RELEASE_INFO2_MSDU_IN_AMSDU)
+		ts->flags |= HAL_TX_STATUS_FLAGS_MSDU_IN_AMSDU;
+
+	ts->peer_id = FIELD_GET(HAL_WBM_RELEASE_INFO3_PEER_ID, desc->info3);
+	ts->tid = FIELD_GET(HAL_WBM_RELEASE_INFO3_TID, desc->info3);
+
+	if (!(desc->rate_stats.info0 & HAL_TX_RATE_STATS_INFO0_VALID))
+		return;
+
+	ts->flags |= HAL_TX_STATUS_FLAGS_RATE_STATS_VALID;
+	ts->tsf = desc->rate_stats.tsf;
+	ts->bw = FIELD_GET(HAL_TX_RATE_STATS_INFO0_BW, desc->rate_stats.info0);
+	ts->pkt_type = FIELD_GET(HAL_TX_RATE_STATS_INFO0_PKT_TYPE,
+				 desc->rate_stats.info0);
+	if (desc->rate_stats.info0 & HAL_TX_RATE_STATS_INFO0_STBC)
+		ts->flags |= HAL_TX_STATUS_FLAGS_RATE_STBC;
+	if (desc->rate_stats.info0 & HAL_TX_RATE_STATS_INFO0_LDPC)
+		ts->flags |= HAL_TX_STATUS_FLAGS_RATE_LDPC;
+	if (desc->rate_stats.info0 & HAL_TX_RATE_STATS_INFO0_OFDMA_TX)
+		ts->flags |= HAL_TX_STATUS_FLAGS_OFDMA;
+
+	ts->sgi = FIELD_GET(HAL_TX_RATE_STATS_INFO0_SGI,
+			    desc->rate_stats.info0);
+	ts->mcs = FIELD_GET(HAL_TX_RATE_STATS_INFO0_MCS,
+			    desc->rate_stats.info0);
+	ts->num_tones_in_ru = FIELD_GET(HAL_TX_RATE_STATS_INFO0_TONES_IN_RU,
+					desc->rate_stats.info0);
+}
+
+void ath11k_hal_tx_set_dscp_tid_map(struct ath11k_base *ab, int id)
+{
+	u32 ctrl_reg_val;
+	u32 addr;
+	u8 hw_map_val[HAL_DSCP_TID_TBL_SIZE];
+	int i;
+	u32 value;
+	int cnt = 0;
+
+	ctrl_reg_val = ath11k_ahb_read32(ab, HAL_SEQ_WCSS_UMAC_TCL_REG +
+					 HAL_TCL1_RING_CMN_CTRL_REG);
+	/* Enable read/write access */
+	ctrl_reg_val |= HAL_TCL1_RING_CMN_CTRL_DSCP_TID_MAP_PROG_EN;
+	ath11k_ahb_write32(ab, HAL_SEQ_WCSS_UMAC_TCL_REG +
+			   HAL_TCL1_RING_CMN_CTRL_REG, ctrl_reg_val);
+
+	addr = HAL_SEQ_WCSS_UMAC_TCL_REG + HAL_TCL1_RING_DSCP_TID_MAP +
+	       (4 * id * (HAL_DSCP_TID_TBL_SIZE / 4));
+
+	/* Configure each DSCP-TID mapping in three bits there by configure
+	 * three bytes in an iteration.
+	 */
+	for (i = 0; i < DSCP_TID_MAP_TBL_ENTRY_SIZE; i += 8) {
+		value = FIELD_PREP(HAL_TCL1_RING_FIELD_DSCP_TID_MAP0,
+				   dscp_tid_map[i]) |
+			FIELD_PREP(HAL_TCL1_RING_FIELD_DSCP_TID_MAP1,
+				   dscp_tid_map[i + 1]) |
+			FIELD_PREP(HAL_TCL1_RING_FIELD_DSCP_TID_MAP2,
+				   dscp_tid_map[i + 2]) |
+			FIELD_PREP(HAL_TCL1_RING_FIELD_DSCP_TID_MAP3,
+				   dscp_tid_map[i + 3]) |
+			FIELD_PREP(HAL_TCL1_RING_FIELD_DSCP_TID_MAP4,
+				   dscp_tid_map[i + 4]) |
+			FIELD_PREP(HAL_TCL1_RING_FIELD_DSCP_TID_MAP5,
+				   dscp_tid_map[i + 5]) |
+			FIELD_PREP(HAL_TCL1_RING_FIELD_DSCP_TID_MAP6,
+				   dscp_tid_map[i + 6]) |
+			FIELD_PREP(HAL_TCL1_RING_FIELD_DSCP_TID_MAP7,
+				   dscp_tid_map[i + 7]);
+		memcpy(&hw_map_val[cnt], (u8 *)&value, 3);
+		cnt += 3;
+	}
+
+	for (i = 0; i < HAL_DSCP_TID_TBL_SIZE; i += 4) {
+		ath11k_ahb_write32(ab, addr, *(u32 *)&hw_map_val[i]);
+		addr += 4;
+	}
+
+	/* Disable read/write access */
+	ctrl_reg_val = ath11k_ahb_read32(ab, HAL_SEQ_WCSS_UMAC_TCL_REG +
+					 HAL_TCL1_RING_CMN_CTRL_REG);
+	ctrl_reg_val &= ~HAL_TCL1_RING_CMN_CTRL_DSCP_TID_MAP_PROG_EN;
+	ath11k_ahb_write32(ab, HAL_SEQ_WCSS_UMAC_TCL_REG +
+			   HAL_TCL1_RING_CMN_CTRL_REG,
+			   ctrl_reg_val);
+}
+
+void ath11k_hal_tx_init_data_ring(struct ath11k_base *ab, struct hal_srng *srng)
+{
+	struct hal_srng_params params;
+	struct hal_tlv_hdr *tlv;
+	int i, entry_size;
+	u8 *desc;
+
+	memset(&params, 0, sizeof(params));
+
+	entry_size = ath11k_hal_srng_get_entrysize(HAL_TCL_DATA);
+	ath11k_hal_srng_get_params(ab, srng, &params);
+	desc = (u8 *)params.ring_base_vaddr;
+
+	for (i = 0; i < params.num_entries; i++) {
+		tlv = (struct hal_tlv_hdr *)desc;
+		tlv->tl = FIELD_PREP(HAL_TLV_HDR_TAG, HAL_TCL_DATA_CMD) |
+			  FIELD_PREP(HAL_TLV_HDR_LEN,
+				     sizeof(struct hal_tcl_data_cmd));
+		desc += entry_size;
+	}
+}
diff --git a/drivers/net/wireless/ath/ath11k/hal_tx.h b/drivers/net/wireless/ath/ath11k/hal_tx.h
new file mode 100644
index 000000000000..f955070bcf1e
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/hal_tx.h
@@ -0,0 +1,80 @@
+/* SPDX-License-Identifier: BSD-3-Clause-Clear */
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#ifndef ATH11K_HAL_TX_H
+#define ATH11K_HAL_TX_H
+
+#include "hal_desc.h"
+
+#define HAL_TX_ADDRX_EN			1
+#define HAL_TX_ADDRY_EN			2
+
+#define HAL_TX_ADDR_SEARCH_INDEX	0
+#define HAL_TX_ADDR_SEARCH_DEFAULT	1
+
+struct hal_tx_info {
+	u16 meta_data_flags; /* %HAL_TCL_DATA_CMD_INFO0_META_ */
+	u8 ring_id;
+	u32 desc_id;
+	enum hal_tcl_desc_type type;
+	enum hal_tcl_encap_type encap_type;
+	dma_addr_t paddr;
+	u32 data_len;
+	u32 pkt_offset;
+	enum hal_encrypt_type encrypt_type;
+	u32 flags0; /* %HAL_TCL_DATA_CMD_INFO1_ */
+	u32 flags1; /* %HAL_TCL_DATA_CMD_INFO2_ */
+	u16 addr_search_flags; /* %HAL_TCL_DATA_CMD_INFO0_ADDR(X/Y)_ */
+	u16 bss_ast_hash;
+	u8 tid;
+	u8 search_type; /* %HAL_TX_ADDR_SEARCH_ */
+	u8 lmac_id;
+	u8 dscp_tid_tbl_idx;
+};
+
+/* TODO: Check if the actual desc macros can be used instead */
+#define HAL_TX_STATUS_FLAGS_FIRST_MSDU		BIT(0)
+#define HAL_TX_STATUS_FLAGS_LAST_MSDU		BIT(1)
+#define HAL_TX_STATUS_FLAGS_MSDU_IN_AMSDU	BIT(2)
+#define HAL_TX_STATUS_FLAGS_RATE_STATS_VALID	BIT(3)
+#define HAL_TX_STATUS_FLAGS_RATE_LDPC		BIT(4)
+#define HAL_TX_STATUS_FLAGS_RATE_STBC		BIT(5)
+#define HAL_TX_STATUS_FLAGS_OFDMA		BIT(6)
+
+#define HAL_TX_STATUS_DESC_LEN		sizeof(struct hal_wbm_release_ring)
+
+/* Tx status parsed from srng desc */
+struct hal_tx_status {
+	enum hal_wbm_rel_src_module buf_rel_source;
+	u32 desc_id;
+	enum hal_wbm_tqm_rel_reason status;
+	u8 ack_rssi;
+	enum hal_tx_rate_stats_bw bw;
+	enum hal_tx_rate_stats_pkt_type pkt_type;
+	enum hal_tx_rate_stats_sgi sgi;
+	u8 mcs;
+	u16 num_tones_in_ru;
+	u32 flags; /* %HAL_TX_STATUS_FLAGS_ */
+	u32 tsf;
+	u32 ppdu_id;
+	u8 try_cnt;
+	u8 tid;
+	u16 peer_id;
+};
+
+void ath11k_hal_tx_cmd_desc_setup(struct ath11k_base *ab, void *cmd,
+				  struct hal_tx_info *ti);
+void ath11k_hal_tx_desc_sync(void *tx_desc_cached, void *hw_desc);
+void ath11k_hal_tx_status_parse(struct ath11k_base *ab,
+				struct hal_wbm_release_ring *desc,
+				struct hal_tx_status *ts);
+void ath11k_hal_tx_status_desc_sync(void *hw_desc, void *local_desc);
+void ath11k_hal_tx_set_dscp_tid_map(struct ath11k_base *ab, int id);
+int ath11k_hal_reo_cmd_send(struct ath11k_base *ab, struct hal_srng *srng,
+			    enum hal_reo_cmd_type type,
+			    struct ath11k_hal_reo_cmd *cmd);
+void ath11k_hal_tx_init_data_ring(struct ath11k_base *ab,
+				  struct hal_srng *srng);
+#endif
diff --git a/drivers/net/wireless/ath/ath11k/htc.c b/drivers/net/wireless/ath/ath11k/htc.c
new file mode 100644
index 000000000000..9d4115f25ceb
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/htc.c
@@ -0,0 +1,773 @@
+// SPDX-License-Identifier: BSD-3-Clause-Clear
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+#include <linux/skbuff.h>
+#include <linux/ctype.h>
+
+#include "ahb.h"
+#include "debug.h"
+
+struct sk_buff *ath11k_htc_alloc_skb(struct ath11k_base *ab, int size)
+{
+	struct sk_buff *skb;
+
+	skb = dev_alloc_skb(size + sizeof(struct ath11k_htc_hdr));
+	if (!skb)
+		return NULL;
+
+	skb_reserve(skb, sizeof(struct ath11k_htc_hdr));
+
+	/* FW/HTC requires 4-byte aligned streams */
+	if (!IS_ALIGNED((unsigned long)skb->data, 4))
+		ath11k_warn(ab, "Unaligned HTC tx skb\n");
+
+	return skb;
+}
+
+static void ath11k_htc_control_tx_complete(struct ath11k_base *ab,
+					   struct sk_buff *skb)
+{
+	kfree_skb(skb);
+}
+
+static struct sk_buff *ath11k_htc_build_tx_ctrl_skb(void *ab)
+{
+	struct sk_buff *skb;
+	struct ath11k_skb_cb *skb_cb;
+
+	skb = dev_alloc_skb(ATH11K_HTC_CONTROL_BUFFER_SIZE);
+	if (!skb)
+		return NULL;
+
+	skb_reserve(skb, sizeof(struct ath11k_htc_hdr));
+	WARN_ON_ONCE(!IS_ALIGNED((unsigned long)skb->data, 4));
+
+	skb_cb = ATH11K_SKB_CB(skb);
+	memset(skb_cb, 0, sizeof(*skb_cb));
+
+	ath11k_dbg(ab, ATH11K_DBG_HTC, "%s: skb %pK\n", __func__, skb);
+	return skb;
+}
+
+static inline void ath11k_htc_restore_tx_skb(struct ath11k_htc *htc,
+					     struct sk_buff *skb)
+{
+	struct ath11k_skb_cb *skb_cb = ATH11K_SKB_CB(skb);
+
+	dma_unmap_single(htc->ab->dev, skb_cb->paddr, skb->len, DMA_TO_DEVICE);
+	skb_pull(skb, sizeof(struct ath11k_htc_hdr));
+}
+
+static void ath11k_htc_prepare_tx_skb(struct ath11k_htc_ep *ep,
+				      struct sk_buff *skb)
+{
+	struct ath11k_htc_hdr *hdr;
+
+	hdr = (struct ath11k_htc_hdr *)skb->data;
+
+	memset(hdr, 0, sizeof(*hdr));
+	hdr->htc_info = FIELD_PREP(HTC_HDR_ENDPOINTID, ep->eid) |
+			FIELD_PREP(HTC_HDR_PAYLOADLEN,
+				   (skb->len - sizeof(*hdr))) |
+			FIELD_PREP(HTC_HDR_FLAGS,
+				   ATH11K_HTC_FLAG_NEED_CREDIT_UPDATE);
+
+	spin_lock_bh(&ep->htc->tx_lock);
+	hdr->ctrl_info = FIELD_PREP(HTC_HDR_CONTROLBYTES1, ep->seq_no++);
+	spin_unlock_bh(&ep->htc->tx_lock);
+}
+
+int ath11k_htc_send(struct ath11k_htc *htc,
+		    enum ath11k_htc_ep_id eid,
+		    struct sk_buff *skb)
+{
+	struct ath11k_htc_ep *ep = &htc->endpoint[eid];
+	struct ath11k_skb_cb *skb_cb = ATH11K_SKB_CB(skb);
+	struct device *dev = htc->ab->dev;
+	struct ath11k_base *ab = htc->ab;
+	int credits = 0;
+	int ret;
+
+	if (eid >= ATH11K_HTC_EP_COUNT) {
+		ath11k_warn(ab, "Invalid endpoint id: %d\n", eid);
+		return -ENOENT;
+	}
+
+	skb_push(skb, sizeof(struct ath11k_htc_hdr));
+
+	if (ep->tx_credit_flow_enabled) {
+		credits = DIV_ROUND_UP(skb->len, htc->target_credit_size);
+		spin_lock_bh(&htc->tx_lock);
+		if (ep->tx_credits < credits) {
+			ath11k_dbg(ab, ATH11K_DBG_HTC,
+				   "htc insufficient credits ep %d required %d available %d\n",
+				   eid, credits, ep->tx_credits);
+			spin_unlock_bh(&htc->tx_lock);
+			ret = -EAGAIN;
+			goto err_pull;
+		}
+		ep->tx_credits -= credits;
+		ath11k_dbg(ab, ATH11K_DBG_HTC,
+			   "htc ep %d consumed %d credits (total %d)\n",
+			   eid, credits, ep->tx_credits);
+		spin_unlock_bh(&htc->tx_lock);
+	}
+
+	ath11k_htc_prepare_tx_skb(ep, skb);
+
+	skb_cb->eid = eid;
+	skb_cb->paddr = dma_map_single(dev, skb->data, skb->len, DMA_TO_DEVICE);
+	ret = dma_mapping_error(dev, skb_cb->paddr);
+	if (ret) {
+		ret = -EIO;
+		goto err_credits;
+	}
+
+	ret = ath11k_ce_send(htc->ab, skb, ep->ul_pipe_id, ep->eid);
+	if (ret)
+		goto err_unmap;
+
+	return 0;
+
+err_unmap:
+	dma_unmap_single(dev, skb_cb->paddr, skb->len, DMA_TO_DEVICE);
+err_credits:
+	if (ep->tx_credit_flow_enabled) {
+		spin_lock_bh(&htc->tx_lock);
+		ep->tx_credits += credits;
+		ath11k_dbg(ab, ATH11K_DBG_HTC,
+			   "htc ep %d reverted %d credits back (total %d)\n",
+			   eid, credits, ep->tx_credits);
+		spin_unlock_bh(&htc->tx_lock);
+
+		if (ep->ep_ops.ep_tx_credits)
+			ep->ep_ops.ep_tx_credits(htc->ab);
+	}
+err_pull:
+	skb_pull(skb, sizeof(struct ath11k_htc_hdr));
+	return ret;
+}
+
+static void
+ath11k_htc_process_credit_report(struct ath11k_htc *htc,
+				 const struct ath11k_htc_credit_report *report,
+				 int len,
+				 enum ath11k_htc_ep_id eid)
+{
+	struct ath11k_base *ab = htc->ab;
+	struct ath11k_htc_ep *ep;
+	int i, n_reports;
+
+	if (len % sizeof(*report))
+		ath11k_warn(ab, "Uneven credit report len %d", len);
+
+	n_reports = len / sizeof(*report);
+
+	spin_lock_bh(&htc->tx_lock);
+	for (i = 0; i < n_reports; i++, report++) {
+		if (report->eid >= ATH11K_HTC_EP_COUNT)
+			break;
+
+		ep = &htc->endpoint[report->eid];
+		ep->tx_credits += report->credits;
+
+		ath11k_dbg(ab, ATH11K_DBG_HTC, "htc ep %d got %d credits (total %d)\n",
+			   report->eid, report->credits, ep->tx_credits);
+
+		if (ep->ep_ops.ep_tx_credits) {
+			spin_unlock_bh(&htc->tx_lock);
+			ep->ep_ops.ep_tx_credits(htc->ab);
+			spin_lock_bh(&htc->tx_lock);
+		}
+	}
+	spin_unlock_bh(&htc->tx_lock);
+}
+
+static int ath11k_htc_process_trailer(struct ath11k_htc *htc,
+				      u8 *buffer,
+				      int length,
+				      enum ath11k_htc_ep_id src_eid)
+{
+	struct ath11k_base *ab = htc->ab;
+	int status = 0;
+	struct ath11k_htc_record *record;
+	size_t len;
+
+	while (length > 0) {
+		record = (struct ath11k_htc_record *)buffer;
+
+		if (length < sizeof(record->hdr)) {
+			status = -EINVAL;
+			break;
+		}
+
+		if (record->hdr.len > length) {
+			/* no room left in buffer for record */
+			ath11k_warn(ab, "Invalid record length: %d\n",
+				    record->hdr.len);
+			status = -EINVAL;
+			break;
+		}
+
+		switch (record->hdr.id) {
+		case ATH11K_HTC_RECORD_CREDITS:
+			len = sizeof(struct ath11k_htc_credit_report);
+			if (record->hdr.len < len) {
+				ath11k_warn(ab, "Credit report too long\n");
+				status = -EINVAL;
+				break;
+			}
+			ath11k_htc_process_credit_report(htc,
+							 record->credit_report,
+							 record->hdr.len,
+							 src_eid);
+			break;
+		default:
+			ath11k_warn(ab, "Unhandled record: id:%d length:%d\n",
+				    record->hdr.id, record->hdr.len);
+			break;
+		}
+
+		if (status)
+			break;
+
+		/* multiple records may be present in a trailer */
+		buffer += sizeof(record->hdr) + record->hdr.len;
+		length -= sizeof(record->hdr) + record->hdr.len;
+	}
+
+	return status;
+}
+
+void ath11k_htc_rx_completion_handler(struct ath11k_base *ab,
+				      struct sk_buff *skb)
+{
+	int status = 0;
+	struct ath11k_htc *htc = &ab->htc;
+	struct ath11k_htc_hdr *hdr;
+	struct ath11k_htc_ep *ep;
+	u16 payload_len;
+	u32 trailer_len = 0;
+	size_t min_len;
+	u8 eid;
+	bool trailer_present;
+
+	hdr = (struct ath11k_htc_hdr *)skb->data;
+	skb_pull(skb, sizeof(*hdr));
+
+	eid = FIELD_GET(HTC_HDR_ENDPOINTID, hdr->htc_info);
+
+	if (eid >= ATH11K_HTC_EP_COUNT) {
+		ath11k_warn(ab, "HTC Rx: invalid eid %d\n", eid);
+		goto out;
+	}
+
+	ep = &htc->endpoint[eid];
+
+	payload_len = FIELD_GET(HTC_HDR_PAYLOADLEN, hdr->htc_info);
+
+	if (payload_len + sizeof(*hdr) > ATH11K_HTC_MAX_LEN) {
+		ath11k_warn(ab, "HTC rx frame too long, len: %zu\n",
+			    payload_len + sizeof(*hdr));
+		goto out;
+	}
+
+	if (skb->len < payload_len) {
+		ath11k_warn(ab, "HTC Rx: insufficient length, got %d, expected %d\n",
+			    skb->len, payload_len);
+		goto out;
+	}
+
+	/* get flags to check for trailer */
+	trailer_present = (FIELD_GET(HTC_HDR_FLAGS, hdr->htc_info)) &
+			  ATH11K_HTC_FLAG_TRAILER_PRESENT;
+
+	if (trailer_present) {
+		u8 *trailer;
+
+		trailer_len = FIELD_GET(HTC_HDR_CONTROLBYTES0, hdr->ctrl_info);
+		min_len = sizeof(struct ath11k_htc_record_hdr);
+
+		if ((trailer_len < min_len) ||
+		    (trailer_len > payload_len)) {
+			ath11k_warn(ab, "Invalid trailer length: %d\n",
+				    trailer_len);
+			goto out;
+		}
+
+		trailer = (u8 *)hdr;
+		trailer += sizeof(*hdr);
+		trailer += payload_len;
+		trailer -= trailer_len;
+		status = ath11k_htc_process_trailer(htc, trailer,
+						    trailer_len, eid);
+		if (status)
+			goto out;
+
+		skb_trim(skb, skb->len - trailer_len);
+	}
+
+	if (trailer_len >= payload_len)
+		/* zero length packet with trailer data, just drop these */
+		goto out;
+
+	if (eid == ATH11K_HTC_EP_0) {
+		struct ath11k_htc_msg *msg = (struct ath11k_htc_msg *)skb->data;
+
+		switch (FIELD_GET(HTC_MSG_MESSAGEID, msg->msg_svc_id)) {
+		case ATH11K_HTC_MSG_READY_ID:
+		case ATH11K_HTC_MSG_CONNECT_SERVICE_RESP_ID:
+			/* handle HTC control message */
+			if (completion_done(&htc->ctl_resp)) {
+				/* this is a fatal error, target should not be
+				 * sending unsolicited messages on the ep 0
+				 */
+				ath11k_warn(ab, "HTC rx ctrl still processing\n");
+				complete(&htc->ctl_resp);
+				goto out;
+			}
+
+			htc->control_resp_len =
+				min_t(int, skb->len,
+				      ATH11K_HTC_MAX_CTRL_MSG_LEN);
+
+			memcpy(htc->control_resp_buffer, skb->data,
+			       htc->control_resp_len);
+
+			complete(&htc->ctl_resp);
+			break;
+		default:
+			ath11k_warn(ab, "ignoring unsolicited htc ep0 event\n");
+			break;
+		}
+		goto out;
+	}
+
+	ath11k_dbg(ab, ATH11K_DBG_HTC, "htc rx completion ep %d skb %pK\n",
+		   eid, skb);
+	ep->ep_ops.ep_rx_complete(ab, skb);
+
+	/* poll tx completion for interrupt disabled CE's */
+	ath11k_ce_poll_send_completed(ab, ep->ul_pipe_id);
+
+	/* skb is now owned by the rx completion handler */
+	skb = NULL;
+out:
+	kfree_skb(skb);
+}
+
+static void ath11k_htc_control_rx_complete(struct ath11k_base *ab,
+					   struct sk_buff *skb)
+{
+	/* This is unexpected. FW is not supposed to send regular rx on this
+	 * endpoint.
+	 */
+	ath11k_warn(ab, "unexpected htc rx\n");
+	kfree_skb(skb);
+}
+
+static const char *htc_service_name(enum ath11k_htc_svc_id id)
+{
+	switch (id) {
+	case ATH11K_HTC_SVC_ID_RESERVED:
+		return "Reserved";
+	case ATH11K_HTC_SVC_ID_RSVD_CTRL:
+		return "Control";
+	case ATH11K_HTC_SVC_ID_WMI_CONTROL:
+		return "WMI";
+	case ATH11K_HTC_SVC_ID_WMI_DATA_BE:
+		return "DATA BE";
+	case ATH11K_HTC_SVC_ID_WMI_DATA_BK:
+		return "DATA BK";
+	case ATH11K_HTC_SVC_ID_WMI_DATA_VI:
+		return "DATA VI";
+	case ATH11K_HTC_SVC_ID_WMI_DATA_VO:
+		return "DATA VO";
+	case ATH11K_HTC_SVC_ID_WMI_CONTROL_MAC1:
+		return "WMI MAC1";
+	case ATH11K_HTC_SVC_ID_WMI_CONTROL_MAC2:
+		return "WMI MAC2";
+	case ATH11K_HTC_SVC_ID_NMI_CONTROL:
+		return "NMI Control";
+	case ATH11K_HTC_SVC_ID_NMI_DATA:
+		return "NMI Data";
+	case ATH11K_HTC_SVC_ID_HTT_DATA_MSG:
+		return "HTT Data";
+	case ATH11K_HTC_SVC_ID_TEST_RAW_STREAMS:
+		return "RAW";
+	case ATH11K_HTC_SVC_ID_IPA_TX:
+		return "IPA TX";
+	case ATH11K_HTC_SVC_ID_PKT_LOG:
+		return "PKT LOG";
+	}
+
+	return "Unknown";
+}
+
+static void ath11k_htc_reset_endpoint_states(struct ath11k_htc *htc)
+{
+	struct ath11k_htc_ep *ep;
+	int i;
+
+	for (i = ATH11K_HTC_EP_0; i < ATH11K_HTC_EP_COUNT; i++) {
+		ep = &htc->endpoint[i];
+		ep->service_id = ATH11K_HTC_SVC_ID_UNUSED;
+		ep->max_ep_message_len = 0;
+		ep->max_tx_queue_depth = 0;
+		ep->eid = i;
+		ep->htc = htc;
+		ep->tx_credit_flow_enabled = true;
+	}
+}
+
+static u8 ath11k_htc_get_credit_allocation(struct ath11k_htc *htc,
+					   u16 service_id)
+{
+	u8 i, allocation = 0;
+
+	for (i = 0; i < ATH11K_HTC_MAX_SERVICE_ALLOC_ENTRIES; i++) {
+		if (htc->service_alloc_table[i].service_id == service_id) {
+			allocation =
+				htc->service_alloc_table[i].credit_allocation;
+		}
+	}
+
+	return allocation;
+}
+
+static int ath11k_htc_setup_target_buffer_assignments(struct ath11k_htc *htc)
+{
+	struct ath11k_htc_svc_tx_credits *serv_entry;
+	u32 svc_id[] = {
+		ATH11K_HTC_SVC_ID_WMI_CONTROL,
+		ATH11K_HTC_SVC_ID_WMI_CONTROL_MAC1,
+		ATH11K_HTC_SVC_ID_WMI_CONTROL_MAC2,
+	};
+	int i, credits;
+
+	credits =  htc->total_transmit_credits;
+	serv_entry = htc->service_alloc_table;
+
+	if ((htc->wmi_ep_count == 0) ||
+	    (htc->wmi_ep_count > ARRAY_SIZE(svc_id)))
+		return -EINVAL;
+
+	/* Divide credits among number of endpoints for WMI */
+	credits = credits / htc->wmi_ep_count;
+	for (i = 0; i < htc->wmi_ep_count; i++) {
+		serv_entry[i].service_id = svc_id[i];
+		serv_entry[i].credit_allocation = credits;
+	}
+
+	return 0;
+}
+
+int ath11k_htc_wait_target(struct ath11k_htc *htc)
+{
+	int i, status = 0;
+	struct ath11k_base *ab = htc->ab;
+	unsigned long time_left;
+	struct ath11k_htc_ready *ready;
+	u16 message_id;
+	u16 credit_count;
+	u16 credit_size;
+
+	time_left = wait_for_completion_timeout(&htc->ctl_resp,
+						ATH11K_HTC_WAIT_TIMEOUT_HZ);
+	if (!time_left) {
+		ath11k_warn(ab, "failed to receive control response completion, polling..\n");
+
+		for (i = 0; i < CE_COUNT; i++)
+			ath11k_ce_per_engine_service(htc->ab, i);
+
+		time_left =
+			wait_for_completion_timeout(&htc->ctl_resp,
+						    ATH11K_HTC_WAIT_TIMEOUT_HZ);
+
+		if (!time_left)
+			status = -ETIMEDOUT;
+	}
+
+	if (status < 0) {
+		ath11k_warn(ab, "ctl_resp never came in (%d)\n", status);
+		return status;
+	}
+
+	if (htc->control_resp_len < sizeof(*ready)) {
+		ath11k_warn(ab, "Invalid HTC ready msg len:%d\n",
+			    htc->control_resp_len);
+		return -ECOMM;
+	}
+
+	ready = (struct ath11k_htc_ready *)htc->control_resp_buffer;
+	message_id   = FIELD_GET(HTC_MSG_MESSAGEID, ready->id_credit_count);
+	credit_count = FIELD_GET(HTC_READY_MSG_CREDITCOUNT,
+				 ready->id_credit_count);
+	credit_size  = FIELD_GET(HTC_READY_MSG_CREDITSIZE, ready->size_ep);
+
+	if (message_id != ATH11K_HTC_MSG_READY_ID) {
+		ath11k_warn(ab, "Invalid HTC ready msg: 0x%x\n", message_id);
+		return -ECOMM;
+	}
+
+	htc->total_transmit_credits = credit_count;
+	htc->target_credit_size = credit_size;
+
+	ath11k_dbg(ab, ATH11K_DBG_HTC,
+		   "Target ready! transmit resources: %d size:%d\n",
+		   htc->total_transmit_credits, htc->target_credit_size);
+
+	if ((htc->total_transmit_credits == 0) ||
+	    (htc->target_credit_size == 0)) {
+		ath11k_warn(ab, "Invalid credit size received\n");
+		return -ECOMM;
+	}
+
+	ath11k_htc_setup_target_buffer_assignments(htc);
+
+	return 0;
+}
+
+int ath11k_htc_connect_service(struct ath11k_htc *htc,
+			       struct ath11k_htc_svc_conn_req *conn_req,
+			       struct ath11k_htc_svc_conn_resp *conn_resp)
+{
+	struct ath11k_base *ab = htc->ab;
+	struct ath11k_htc_conn_svc *req_msg;
+	struct ath11k_htc_conn_svc_resp resp_msg_dummy;
+	struct ath11k_htc_conn_svc_resp *resp_msg = &resp_msg_dummy;
+	enum ath11k_htc_ep_id assigned_eid = ATH11K_HTC_EP_COUNT;
+	struct ath11k_htc_ep *ep;
+	struct sk_buff *skb;
+	unsigned int max_msg_size = 0;
+	int length, status;
+	unsigned long time_left;
+	bool disable_credit_flow_ctrl = false;
+	u16 message_id, service_id, flags = 0;
+	u8 tx_alloc = 0;
+
+	/* special case for HTC pseudo control service */
+	if (conn_req->service_id == ATH11K_HTC_SVC_ID_RSVD_CTRL) {
+		disable_credit_flow_ctrl = true;
+		assigned_eid = ATH11K_HTC_EP_0;
+		max_msg_size = ATH11K_HTC_MAX_CTRL_MSG_LEN;
+		memset(&resp_msg_dummy, 0, sizeof(resp_msg_dummy));
+		goto setup;
+	}
+
+	tx_alloc = ath11k_htc_get_credit_allocation(htc,
+						    conn_req->service_id);
+	if (!tx_alloc)
+		ath11k_dbg(ab, ATH11K_DBG_BOOT,
+			   "boot htc service %s does not allocate target credits\n",
+			   htc_service_name(conn_req->service_id));
+
+	skb = ath11k_htc_build_tx_ctrl_skb(htc->ab);
+	if (!skb) {
+		ath11k_warn(ab, "Failed to allocate HTC packet\n");
+		return -ENOMEM;
+	}
+
+	length = sizeof(*req_msg);
+	skb_put(skb, length);
+	memset(skb->data, 0, length);
+
+	req_msg = (struct ath11k_htc_conn_svc *)skb->data;
+	req_msg->msg_svc_id = FIELD_PREP(HTC_MSG_MESSAGEID,
+					 ATH11K_HTC_MSG_CONNECT_SERVICE_ID);
+
+	flags |= FIELD_PREP(ATH11K_HTC_CONN_FLAGS_RECV_ALLOC, tx_alloc);
+
+	/* Only enable credit flow control for WMI ctrl service */
+	if (!(conn_req->service_id == ATH11K_HTC_SVC_ID_WMI_CONTROL ||
+	      conn_req->service_id == ATH11K_HTC_SVC_ID_WMI_CONTROL_MAC1 ||
+	      conn_req->service_id == ATH11K_HTC_SVC_ID_WMI_CONTROL_MAC2)) {
+		flags |= ATH11K_HTC_CONN_FLAGS_DISABLE_CREDIT_FLOW_CTRL;
+		disable_credit_flow_ctrl = true;
+	}
+
+	req_msg->flags_len = FIELD_PREP(HTC_SVC_MSG_CONNECTIONFLAGS, flags);
+	req_msg->msg_svc_id |= FIELD_PREP(HTC_SVC_MSG_SERVICE_ID,
+					  conn_req->service_id);
+
+	reinit_completion(&htc->ctl_resp);
+
+	status = ath11k_htc_send(htc, ATH11K_HTC_EP_0, skb);
+	if (status) {
+		kfree_skb(skb);
+		return status;
+	}
+
+	/* wait for response */
+	time_left = wait_for_completion_timeout(&htc->ctl_resp,
+						ATH11K_HTC_CONN_SVC_TIMEOUT_HZ);
+	if (!time_left) {
+		ath11k_err(ab, "Service connect timeout\n");
+		return -ETIMEDOUT;
+	}
+
+	/* we controlled the buffer creation, it's aligned */
+	resp_msg = (struct ath11k_htc_conn_svc_resp *)htc->control_resp_buffer;
+	message_id = FIELD_GET(HTC_MSG_MESSAGEID, resp_msg->msg_svc_id);
+	service_id = FIELD_GET(HTC_SVC_RESP_MSG_SERVICEID,
+			       resp_msg->msg_svc_id);
+
+	if ((message_id != ATH11K_HTC_MSG_CONNECT_SERVICE_RESP_ID) ||
+	    (htc->control_resp_len < sizeof(*resp_msg))) {
+		ath11k_err(ab, "Invalid resp message ID 0x%x", message_id);
+		return -EPROTO;
+	}
+
+	ath11k_dbg(ab, ATH11K_DBG_HTC,
+		   "HTC Service %s connect response: status: 0x%lx, assigned ep: 0x%lx\n",
+		   htc_service_name(service_id),
+		   FIELD_GET(HTC_SVC_RESP_MSG_STATUS, resp_msg->flags_len),
+		   FIELD_GET(HTC_SVC_RESP_MSG_ENDPOINTID, resp_msg->flags_len));
+
+	conn_resp->connect_resp_code = FIELD_GET(HTC_SVC_RESP_MSG_STATUS,
+						 resp_msg->flags_len);
+
+	/* check response status */
+	if (conn_resp->connect_resp_code != ATH11K_HTC_CONN_SVC_STATUS_SUCCESS) {
+		ath11k_err(ab, "HTC Service %s connect request failed: 0x%x)\n",
+			   htc_service_name(service_id),
+		       conn_resp->connect_resp_code);
+		return -EPROTO;
+	}
+
+	assigned_eid = (enum ath11k_htc_ep_id)FIELD_GET(
+						HTC_SVC_RESP_MSG_ENDPOINTID,
+						resp_msg->flags_len);
+
+	max_msg_size = FIELD_GET(HTC_SVC_RESP_MSG_MAXMSGSIZE,
+				 resp_msg->flags_len);
+
+setup:
+
+	if (assigned_eid >= ATH11K_HTC_EP_COUNT)
+		return -EPROTO;
+
+	if (max_msg_size == 0)
+		return -EPROTO;
+
+	ep = &htc->endpoint[assigned_eid];
+	ep->eid = assigned_eid;
+
+	if (ep->service_id != ATH11K_HTC_SVC_ID_UNUSED)
+		return -EPROTO;
+
+	/* return assigned endpoint to caller */
+	conn_resp->eid = assigned_eid;
+	conn_resp->max_msg_len = FIELD_GET(HTC_SVC_RESP_MSG_MAXMSGSIZE,
+					   resp_msg->flags_len);
+
+	/* setup the endpoint */
+	ep->service_id = conn_req->service_id;
+	ep->max_tx_queue_depth = conn_req->max_send_queue_depth;
+	ep->max_ep_message_len = FIELD_GET(HTC_SVC_RESP_MSG_MAXMSGSIZE,
+					   resp_msg->flags_len);
+	ep->tx_credits = tx_alloc;
+
+	/* copy all the callbacks */
+	ep->ep_ops = conn_req->ep_ops;
+
+	status = ath11k_ahb_map_service_to_pipe(htc->ab,
+						ep->service_id,
+						&ep->ul_pipe_id,
+						&ep->dl_pipe_id);
+	if (status)
+		return status;
+
+	ath11k_dbg(ab, ATH11K_DBG_BOOT,
+		   "boot htc service '%s' ul pipe %d dl pipe %d eid %d ready\n",
+		   htc_service_name(ep->service_id), ep->ul_pipe_id,
+		   ep->dl_pipe_id, ep->eid);
+
+	if (disable_credit_flow_ctrl && ep->tx_credit_flow_enabled) {
+		ep->tx_credit_flow_enabled = false;
+		ath11k_dbg(ab, ATH11K_DBG_BOOT,
+			   "boot htc service '%s' eid %d TX flow control disabled\n",
+			   htc_service_name(ep->service_id), assigned_eid);
+	}
+
+	return status;
+}
+
+int ath11k_htc_start(struct ath11k_htc *htc)
+{
+	struct sk_buff *skb;
+	int status = 0;
+	struct ath11k_base *ab = htc->ab;
+	struct ath11k_htc_setup_complete_extended *msg;
+
+	skb = ath11k_htc_build_tx_ctrl_skb(htc->ab);
+	if (!skb)
+		return -ENOMEM;
+
+	skb_put(skb, sizeof(*msg));
+	memset(skb->data, 0, skb->len);
+
+	msg = (struct ath11k_htc_setup_complete_extended *)skb->data;
+	msg->msg_id = FIELD_PREP(HTC_MSG_MESSAGEID,
+				 ATH11K_HTC_MSG_SETUP_COMPLETE_EX_ID);
+
+	ath11k_dbg(ab, ATH11K_DBG_HTC, "HTC is using TX credit flow control\n");
+
+	status = ath11k_htc_send(htc, ATH11K_HTC_EP_0, skb);
+	if (status) {
+		kfree_skb(skb);
+		return status;
+	}
+
+	return 0;
+}
+
+int ath11k_htc_init(struct ath11k_base *ab)
+{
+	struct ath11k_htc *htc = &ab->htc;
+	struct ath11k_htc_svc_conn_req conn_req;
+	struct ath11k_htc_svc_conn_resp conn_resp;
+	int ret;
+
+	spin_lock_init(&htc->tx_lock);
+
+	ath11k_htc_reset_endpoint_states(htc);
+
+	htc->ab = ab;
+
+	switch (ab->wmi_sc.preferred_hw_mode) {
+	case WMI_HOST_HW_MODE_SINGLE:
+		htc->wmi_ep_count = 1;
+		break;
+	case WMI_HOST_HW_MODE_DBS:
+	case WMI_HOST_HW_MODE_DBS_OR_SBS:
+		htc->wmi_ep_count = 2;
+		break;
+	case WMI_HOST_HW_MODE_DBS_SBS:
+		htc->wmi_ep_count = 3;
+		break;
+	default:
+		htc->wmi_ep_count = 3;
+		break;
+	}
+
+	/* setup our pseudo HTC control endpoint connection */
+	memset(&conn_req, 0, sizeof(conn_req));
+	memset(&conn_resp, 0, sizeof(conn_resp));
+	conn_req.ep_ops.ep_tx_complete = ath11k_htc_control_tx_complete;
+	conn_req.ep_ops.ep_rx_complete = ath11k_htc_control_rx_complete;
+	conn_req.max_send_queue_depth = ATH11K_NUM_CONTROL_TX_BUFFERS;
+	conn_req.service_id = ATH11K_HTC_SVC_ID_RSVD_CTRL;
+
+	/* connect fake service */
+	ret = ath11k_htc_connect_service(htc, &conn_req, &conn_resp);
+	if (ret) {
+		ath11k_err(ab, "could not connect to htc service (%d)\n", ret);
+		return ret;
+	}
+
+	init_completion(&htc->ctl_resp);
+
+	return 0;
+}
diff --git a/drivers/net/wireless/ath/ath11k/htc.h b/drivers/net/wireless/ath/ath11k/htc.h
new file mode 100644
index 000000000000..f0a3387567ca
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/htc.h
@@ -0,0 +1,313 @@
+/* SPDX-License-Identifier: BSD-3-Clause-Clear */
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#ifndef ATH11K_HTC_H
+#define ATH11K_HTC_H
+
+#include <linux/kernel.h>
+#include <linux/list.h>
+#include <linux/bug.h>
+#include <linux/skbuff.h>
+#include <linux/timer.h>
+
+struct ath11k_base;
+
+#define HTC_HDR_ENDPOINTID                       GENMASK(7, 0)
+#define HTC_HDR_FLAGS                            GENMASK(15, 8)
+#define HTC_HDR_PAYLOADLEN                       GENMASK(31, 16)
+#define HTC_HDR_CONTROLBYTES0                    GENMASK(7, 0)
+#define HTC_HDR_CONTROLBYTES1                    GENMASK(15, 8)
+#define HTC_HDR_RESERVED                         GENMASK(31, 16)
+
+#define HTC_SVC_MSG_SERVICE_ID                   GENMASK(31, 16)
+#define HTC_SVC_MSG_CONNECTIONFLAGS              GENMASK(15, 0)
+#define HTC_SVC_MSG_SERVICEMETALENGTH            GENMASK(23, 16)
+#define HTC_READY_MSG_CREDITCOUNT                GENMASK(31, 16)
+#define HTC_READY_MSG_CREDITSIZE                 GENMASK(15, 0)
+#define HTC_READY_MSG_MAXENDPOINTS               GENMASK(23, 16)
+
+#define HTC_READY_EX_MSG_HTCVERSION              GENMASK(7, 0)
+#define HTC_READY_EX_MSG_MAXMSGSPERHTCBUNDLE     GENMASK(15, 8)
+
+#define HTC_SVC_RESP_MSG_SERVICEID           GENMASK(31, 16)
+#define HTC_SVC_RESP_MSG_STATUS              GENMASK(7, 0)
+#define HTC_SVC_RESP_MSG_ENDPOINTID          GENMASK(15, 8)
+#define HTC_SVC_RESP_MSG_MAXMSGSIZE          GENMASK(31, 16)
+#define HTC_SVC_RESP_MSG_SERVICEMETALENGTH   GENMASK(7, 0)
+
+#define HTC_MSG_MESSAGEID                        GENMASK(15, 0)
+#define HTC_SETUP_COMPLETE_EX_MSG_SETUPFLAGS     GENMASK(31, 0)
+#define HTC_SETUP_COMPLETE_EX_MSG_MAXMSGSPERBUNDLEDRECV      GENMASK(7, 0)
+#define HTC_SETUP_COMPLETE_EX_MSG_RSVD0          GENMASK(15, 8)
+#define HTC_SETUP_COMPLETE_EX_MSG_RSVD1          GENMASK(23, 16)
+#define HTC_SETUP_COMPLETE_EX_MSG_RSVD2          GENMASK(31, 24)
+
+enum ath11k_htc_tx_flags {
+	ATH11K_HTC_FLAG_NEED_CREDIT_UPDATE = 0x01,
+	ATH11K_HTC_FLAG_SEND_BUNDLE        = 0x02
+};
+
+enum ath11k_htc_rx_flags {
+	ATH11K_HTC_FLAG_TRAILER_PRESENT = 0x02,
+	ATH11K_HTC_FLAG_BUNDLE_MASK     = 0xF0
+};
+
+struct ath11k_htc_hdr {
+	u32 htc_info;
+	u32 ctrl_info;
+} __packed __aligned(4);
+
+enum ath11k_htc_msg_id {
+	ATH11K_HTC_MSG_READY_ID                = 1,
+	ATH11K_HTC_MSG_CONNECT_SERVICE_ID      = 2,
+	ATH11K_HTC_MSG_CONNECT_SERVICE_RESP_ID = 3,
+	ATH11K_HTC_MSG_SETUP_COMPLETE_ID       = 4,
+	ATH11K_HTC_MSG_SETUP_COMPLETE_EX_ID    = 5,
+	ATH11K_HTC_MSG_SEND_SUSPEND_COMPLETE   = 6
+};
+
+enum ath11k_htc_version {
+	ATH11K_HTC_VERSION_2P0 = 0x00, /* 2.0 */
+	ATH11K_HTC_VERSION_2P1 = 0x01, /* 2.1 */
+};
+
+#define ATH11K_HTC_CONN_FLAGS_THRESHOLD_LEVEL_MASK GENMASK(1, 0)
+#define ATH11K_HTC_CONN_FLAGS_RECV_ALLOC GENMASK(15, 8)
+
+enum ath11k_htc_conn_flags {
+	ATH11K_HTC_CONN_FLAGS_THRESHOLD_LEVEL_ONE_FOURTH    = 0x0,
+	ATH11K_HTC_CONN_FLAGS_THRESHOLD_LEVEL_ONE_HALF      = 0x1,
+	ATH11K_HTC_CONN_FLAGS_THRESHOLD_LEVEL_THREE_FOURTHS = 0x2,
+	ATH11K_HTC_CONN_FLAGS_THRESHOLD_LEVEL_UNITY         = 0x3,
+	ATH11K_HTC_CONN_FLAGS_REDUCE_CREDIT_DRIBBLE    = 1 << 2,
+	ATH11K_HTC_CONN_FLAGS_DISABLE_CREDIT_FLOW_CTRL = 1 << 3
+};
+
+enum ath11k_htc_conn_svc_status {
+	ATH11K_HTC_CONN_SVC_STATUS_SUCCESS      = 0,
+	ATH11K_HTC_CONN_SVC_STATUS_NOT_FOUND    = 1,
+	ATH11K_HTC_CONN_SVC_STATUS_FAILED       = 2,
+	ATH11K_HTC_CONN_SVC_STATUS_NO_RESOURCES = 3,
+	ATH11K_HTC_CONN_SVC_STATUS_NO_MORE_EP   = 4
+};
+
+struct ath11k_htc_ready {
+	u32 id_credit_count;
+	u32 size_ep;
+} __packed;
+
+struct ath11k_htc_ready_extended {
+	struct ath11k_htc_ready base;
+	u32 ver_bundle;
+} __packed;
+
+struct ath11k_htc_conn_svc {
+	u32 msg_svc_id;
+	u32 flags_len;
+} __packed;
+
+struct ath11k_htc_conn_svc_resp {
+	u32 msg_svc_id;
+	u32 flags_len;
+	u32 svc_meta_pad;
+} __packed;
+
+struct ath11k_htc_setup_complete_extended {
+	u32 msg_id;
+	u32 flags;
+	u32 max_msgs_per_bundled_recv;
+} __packed;
+
+struct ath11k_htc_msg {
+	u32 msg_svc_id;
+	u32 flags_len;
+} __packed __aligned(4);
+
+enum ath11k_htc_record_id {
+	ATH11K_HTC_RECORD_NULL    = 0,
+	ATH11K_HTC_RECORD_CREDITS = 1
+};
+
+struct ath11k_htc_record_hdr {
+	u8 id; /* @enum ath11k_htc_record_id */
+	u8 len;
+	u8 pad0;
+	u8 pad1;
+} __packed;
+
+struct ath11k_htc_credit_report {
+	u8 eid; /* @enum ath11k_htc_ep_id */
+	u8 credits;
+	u8 pad0;
+	u8 pad1;
+} __packed;
+
+struct ath11k_htc_record {
+	struct ath11k_htc_record_hdr hdr;
+	union {
+		struct ath11k_htc_credit_report credit_report[0];
+		u8 pauload[0];
+	};
+} __packed __aligned(4);
+
+/* note: the trailer offset is dynamic depending
+ * on payload length. this is only a struct layout draft
+ */
+struct ath11k_htc_frame {
+	struct ath11k_htc_hdr hdr;
+	union {
+		struct ath11k_htc_msg msg;
+		u8 payload[0];
+	};
+	struct ath11k_htc_record trailer[0];
+} __packed __aligned(4);
+
+enum ath11k_htc_svc_gid {
+	ATH11K_HTC_SVC_GRP_RSVD = 0,
+	ATH11K_HTC_SVC_GRP_WMI = 1,
+	ATH11K_HTC_SVC_GRP_NMI = 2,
+	ATH11K_HTC_SVC_GRP_HTT = 3,
+	ATH11K_HTC_SVC_GRP_CFG = 4,
+	ATH11K_HTC_SVC_GRP_IPA = 5,
+	ATH11K_HTC_SVC_GRP_PKTLOG = 6,
+
+	ATH11K_HTC_SVC_GRP_TEST = 254,
+	ATH11K_HTC_SVC_GRP_LAST = 255,
+};
+
+#define SVC(group, idx) \
+	(int)(((int)(group) << 8) | (int)(idx))
+
+enum ath11k_htc_svc_id {
+	/* NOTE: service ID of 0x0000 is reserved and should never be used */
+	ATH11K_HTC_SVC_ID_RESERVED	= 0x0000,
+	ATH11K_HTC_SVC_ID_UNUSED	= ATH11K_HTC_SVC_ID_RESERVED,
+
+	ATH11K_HTC_SVC_ID_RSVD_CTRL	= SVC(ATH11K_HTC_SVC_GRP_RSVD, 1),
+	ATH11K_HTC_SVC_ID_WMI_CONTROL	= SVC(ATH11K_HTC_SVC_GRP_WMI, 0),
+	ATH11K_HTC_SVC_ID_WMI_DATA_BE	= SVC(ATH11K_HTC_SVC_GRP_WMI, 1),
+	ATH11K_HTC_SVC_ID_WMI_DATA_BK	= SVC(ATH11K_HTC_SVC_GRP_WMI, 2),
+	ATH11K_HTC_SVC_ID_WMI_DATA_VI	= SVC(ATH11K_HTC_SVC_GRP_WMI, 3),
+	ATH11K_HTC_SVC_ID_WMI_DATA_VO	= SVC(ATH11K_HTC_SVC_GRP_WMI, 4),
+	ATH11K_HTC_SVC_ID_WMI_CONTROL_MAC1 = SVC(ATH11K_HTC_SVC_GRP_WMI, 5),
+	ATH11K_HTC_SVC_ID_WMI_CONTROL_MAC2 = SVC(ATH11K_HTC_SVC_GRP_WMI, 6),
+
+	ATH11K_HTC_SVC_ID_NMI_CONTROL	= SVC(ATH11K_HTC_SVC_GRP_NMI, 0),
+	ATH11K_HTC_SVC_ID_NMI_DATA	= SVC(ATH11K_HTC_SVC_GRP_NMI, 1),
+
+	ATH11K_HTC_SVC_ID_HTT_DATA_MSG	= SVC(ATH11K_HTC_SVC_GRP_HTT, 0),
+
+	/* raw stream service (i.e. flash, tcmd, calibration apps) */
+	ATH11K_HTC_SVC_ID_TEST_RAW_STREAMS = SVC(ATH11K_HTC_SVC_GRP_TEST, 0),
+	ATH11K_HTC_SVC_ID_IPA_TX = SVC(ATH11K_HTC_SVC_GRP_IPA, 0),
+	ATH11K_HTC_SVC_ID_PKT_LOG = SVC(ATH11K_HTC_SVC_GRP_PKTLOG, 0),
+};
+
+#undef SVC
+
+enum ath11k_htc_ep_id {
+	ATH11K_HTC_EP_UNUSED = -1,
+	ATH11K_HTC_EP_0 = 0,
+	ATH11K_HTC_EP_1 = 1,
+	ATH11K_HTC_EP_2,
+	ATH11K_HTC_EP_3,
+	ATH11K_HTC_EP_4,
+	ATH11K_HTC_EP_5,
+	ATH11K_HTC_EP_6,
+	ATH11K_HTC_EP_7,
+	ATH11K_HTC_EP_8,
+	ATH11K_HTC_EP_COUNT,
+};
+
+struct ath11k_htc_ops {
+	void (*target_send_suspend_complete)(struct ath11k_base *ar);
+};
+
+struct ath11k_htc_ep_ops {
+	void (*ep_tx_complete)(struct ath11k_base *, struct sk_buff *);
+	void (*ep_rx_complete)(struct ath11k_base *, struct sk_buff *);
+	void (*ep_tx_credits)(struct ath11k_base *);
+};
+
+/* service connection information */
+struct ath11k_htc_svc_conn_req {
+	u16 service_id;
+	struct ath11k_htc_ep_ops ep_ops;
+	int max_send_queue_depth;
+};
+
+/* service connection response information */
+struct ath11k_htc_svc_conn_resp {
+	u8 buffer_len;
+	u8 actual_len;
+	enum ath11k_htc_ep_id eid;
+	unsigned int max_msg_len;
+	u8 connect_resp_code;
+};
+
+#define ATH11K_NUM_CONTROL_TX_BUFFERS 2
+#define ATH11K_HTC_MAX_LEN 4096
+#define ATH11K_HTC_MAX_CTRL_MSG_LEN 256
+#define ATH11K_HTC_WAIT_TIMEOUT_HZ (1 * HZ)
+#define ATH11K_HTC_CONTROL_BUFFER_SIZE (ATH11K_HTC_MAX_CTRL_MSG_LEN + \
+					sizeof(struct ath11k_htc_hdr))
+#define ATH11K_HTC_CONN_SVC_TIMEOUT_HZ (1 * HZ)
+#define ATH11K_HTC_MAX_SERVICE_ALLOC_ENTRIES 8
+
+struct ath11k_htc_ep {
+	struct ath11k_htc *htc;
+	enum ath11k_htc_ep_id eid;
+	enum ath11k_htc_svc_id service_id;
+	struct ath11k_htc_ep_ops ep_ops;
+
+	int max_tx_queue_depth;
+	int max_ep_message_len;
+	u8 ul_pipe_id;
+	u8 dl_pipe_id;
+
+	u8 seq_no; /* for debugging */
+	int tx_credits;
+	bool tx_credit_flow_enabled;
+};
+
+struct ath11k_htc_svc_tx_credits {
+	u16 service_id;
+	u8  credit_allocation;
+};
+
+struct ath11k_htc {
+	struct ath11k_base *ab;
+	struct ath11k_htc_ep endpoint[ATH11K_HTC_EP_COUNT];
+
+	/* protects endpoints */
+	spinlock_t tx_lock;
+
+	struct ath11k_htc_ops htc_ops;
+
+	u8 control_resp_buffer[ATH11K_HTC_MAX_CTRL_MSG_LEN];
+	int control_resp_len;
+
+	struct completion ctl_resp;
+
+	int total_transmit_credits;
+	struct ath11k_htc_svc_tx_credits
+		service_alloc_table[ATH11K_HTC_MAX_SERVICE_ALLOC_ENTRIES];
+	int target_credit_size;
+	u8 wmi_ep_count;
+};
+
+int ath11k_htc_init(struct ath11k_base *ar);
+int ath11k_htc_wait_target(struct ath11k_htc *htc);
+int ath11k_htc_start(struct ath11k_htc *htc);
+int ath11k_htc_connect_service(struct ath11k_htc *htc,
+			       struct ath11k_htc_svc_conn_req  *conn_req,
+			       struct ath11k_htc_svc_conn_resp *conn_resp);
+int ath11k_htc_send(struct ath11k_htc *htc, enum ath11k_htc_ep_id eid,
+		    struct sk_buff *packet);
+struct sk_buff *ath11k_htc_alloc_skb(struct ath11k_base *ar, int size);
+void ath11k_htc_rx_completion_handler(struct ath11k_base *ar,
+				      struct sk_buff *skb);
+
+#endif
diff --git a/drivers/net/wireless/ath/ath11k/hw.h b/drivers/net/wireless/ath/ath11k/hw.h
new file mode 100644
index 000000000000..dd39333ec0ea
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/hw.h
@@ -0,0 +1,127 @@
+/* SPDX-License-Identifier: BSD-3-Clause-Clear */
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#ifndef ATH11K_HW_H
+#define ATH11K_HW_H
+
+/* Target configuration defines */
+
+/* Num VDEVS per radio */
+#define TARGET_NUM_VDEVS	(16 + 1)
+
+#define TARGET_NUM_PEERS_PDEV	(512 + TARGET_NUM_VDEVS)
+
+/* Num of peers for Single Radio mode */
+#define TARGET_NUM_PEERS_SINGLE		(TARGET_NUM_PEERS_PDEV)
+
+/* Num of peers for DBS */
+#define TARGET_NUM_PEERS_DBS		(2 * TARGET_NUM_PEERS_PDEV)
+
+/* Num of peers for DBS_SBS */
+#define TARGET_NUM_PEERS_DBS_SBS	(3 * TARGET_NUM_PEERS_PDEV)
+
+/* Max num of stations (per radio) */
+#define TARGET_NUM_STATIONS	512
+
+#define TARGET_NUM_PEERS(x)	TARGET_NUM_PEERS_##x
+#define TARGET_NUM_PEER_KEYS	2
+#define TARGET_NUM_TIDS(x)	(2 * TARGET_NUM_PEERS(x) + \
+				 4 * TARGET_NUM_VDEVS + 8)
+
+#define TARGET_AST_SKID_LIMIT	16
+#define TARGET_NUM_OFFLD_PEERS	4
+#define TARGET_NUM_OFFLD_REORDER_BUFFS 4
+
+#define TARGET_TX_CHAIN_MASK	(BIT(0) | BIT(1) | BIT(2) | BIT(4))
+#define TARGET_RX_CHAIN_MASK	(BIT(0) | BIT(1) | BIT(2) | BIT(4))
+#define TARGET_RX_TIMEOUT_LO_PRI	100
+#define TARGET_RX_TIMEOUT_HI_PRI	40
+
+#define TARGET_DECAP_MODE_RAW		0
+#define TARGET_DECAP_MODE_NATIVE_WIFI	1
+#define TARGET_DECAP_MODE_ETH		2
+
+#define TARGET_SCAN_MAX_PENDING_REQS	4
+#define TARGET_BMISS_OFFLOAD_MAX_VDEV	3
+#define TARGET_ROAM_OFFLOAD_MAX_VDEV	3
+#define TARGET_ROAM_OFFLOAD_MAX_AP_PROFILES	8
+#define TARGET_GTK_OFFLOAD_MAX_VDEV	3
+#define TARGET_NUM_MCAST_GROUPS		12
+#define TARGET_NUM_MCAST_TABLE_ELEMS	64
+#define TARGET_MCAST2UCAST_MODE		2
+#define TARGET_TX_DBG_LOG_SIZE		1024
+#define TARGET_RX_SKIP_DEFRAG_TIMEOUT_DUP_DETECTION_CHECK 1
+#define TARGET_VOW_CONFIG		0
+#define TARGET_NUM_MSDU_DESC		(2500)
+#define TARGET_MAX_FRAG_ENTRIES		6
+#define TARGET_MAX_BCN_OFFLD		16
+#define TARGET_NUM_WDS_ENTRIES		32
+#define TARGET_DMA_BURST_SIZE		1
+#define TARGET_RX_BATCHMODE		1
+
+#define ATH11K_HW_MAX_QUEUES		4
+
+#define ATH11k_HW_RATECODE_CCK_SHORT_PREAM_MASK  0x4
+
+#define ATH11K_FW_DIR			"ath11k"
+
+/* IPQ8074 definitions */
+#define IPQ8074_FW_DIR			"IPQ8074"
+#define IPQ8074_MAX_BOARD_DATA_SZ	(256 * 1024)
+#define IPQ8074_MAX_CAL_DATA_SZ		IPQ8074_MAX_BOARD_DATA_SZ
+
+#define ATH11K_BOARD_MAGIC		"QCA-ATH11K-BOARD"
+#define ATH11K_BOARD_API2_FILE		"board-2.bin"
+#define ATH11K_DEFAULT_BOARD_FILE	"bdwlan.bin"
+#define ATH11K_DEFAULT_CAL_FILE		"caldata.bin"
+
+enum ath11k_hw_rate_cck {
+	ATH11K_HW_RATE_CCK_LP_11M = 0,
+	ATH11K_HW_RATE_CCK_LP_5_5M,
+	ATH11K_HW_RATE_CCK_LP_2M,
+	ATH11K_HW_RATE_CCK_LP_1M,
+	ATH11K_HW_RATE_CCK_SP_11M,
+	ATH11K_HW_RATE_CCK_SP_5_5M,
+	ATH11K_HW_RATE_CCK_SP_2M,
+};
+
+enum ath11k_hw_rate_ofdm {
+	ATH11K_HW_RATE_OFDM_48M = 0,
+	ATH11K_HW_RATE_OFDM_24M,
+	ATH11K_HW_RATE_OFDM_12M,
+	ATH11K_HW_RATE_OFDM_6M,
+	ATH11K_HW_RATE_OFDM_54M,
+	ATH11K_HW_RATE_OFDM_36M,
+	ATH11K_HW_RATE_OFDM_18M,
+	ATH11K_HW_RATE_OFDM_9M,
+};
+
+struct ath11k_hw_params {
+	const char *name;
+	struct {
+		const char *dir;
+		size_t board_size;
+		size_t cal_size;
+	} fw;
+};
+
+struct ath11k_fw_ie {
+	__le32 id;
+	__le32 len;
+	u8 data[0];
+};
+
+enum ath11k_bd_ie_board_type {
+	ATH11K_BD_IE_BOARD_NAME = 0,
+	ATH11K_BD_IE_BOARD_DATA = 1,
+};
+
+enum ath11k_bd_ie_type {
+	/* contains sub IEs of enum ath11k_bd_ie_board_type */
+	ATH11K_BD_IE_BOARD = 0,
+	ATH11K_BD_IE_BOARD_EXT = 1,
+};
+
+#endif
diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
new file mode 100644
index 000000000000..cb025a4a5785
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -0,0 +1,5427 @@
+// SPDX-License-Identifier: BSD-3-Clause-Clear
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#include <net/mac80211.h>
+#include <linux/etherdevice.h>
+#include "mac.h"
+#include "core.h"
+#include "debug.h"
+#include "wmi.h"
+#include "hw.h"
+#include "dp_tx.h"
+#include "dp_rx.h"
+#include "testmode.h"
+#include "peer.h"
+
+#define CHAN2G(_channel, _freq, _flags) { \
+	.band                   = NL80211_BAND_2GHZ, \
+	.hw_value               = (_channel), \
+	.center_freq            = (_freq), \
+	.flags                  = (_flags), \
+	.max_antenna_gain       = 0, \
+	.max_power              = 30, \
+}
+
+#define CHAN5G(_channel, _freq, _flags) { \
+	.band                   = NL80211_BAND_5GHZ, \
+	.hw_value               = (_channel), \
+	.center_freq            = (_freq), \
+	.flags                  = (_flags), \
+	.max_antenna_gain       = 0, \
+	.max_power              = 30, \
+}
+
+static const struct ieee80211_channel ath11k_2ghz_channels[] = {
+	CHAN2G(1, 2412, 0),
+	CHAN2G(2, 2417, 0),
+	CHAN2G(3, 2422, 0),
+	CHAN2G(4, 2427, 0),
+	CHAN2G(5, 2432, 0),
+	CHAN2G(6, 2437, 0),
+	CHAN2G(7, 2442, 0),
+	CHAN2G(8, 2447, 0),
+	CHAN2G(9, 2452, 0),
+	CHAN2G(10, 2457, 0),
+	CHAN2G(11, 2462, 0),
+	CHAN2G(12, 2467, 0),
+	CHAN2G(13, 2472, 0),
+	CHAN2G(14, 2484, 0),
+};
+
+static const struct ieee80211_channel ath11k_5ghz_channels[] = {
+	CHAN5G(36, 5180, 0),
+	CHAN5G(40, 5200, 0),
+	CHAN5G(44, 5220, 0),
+	CHAN5G(48, 5240, 0),
+	CHAN5G(52, 5260, 0),
+	CHAN5G(56, 5280, 0),
+	CHAN5G(60, 5300, 0),
+	CHAN5G(64, 5320, 0),
+	CHAN5G(100, 5500, 0),
+	CHAN5G(104, 5520, 0),
+	CHAN5G(108, 5540, 0),
+	CHAN5G(112, 5560, 0),
+	CHAN5G(116, 5580, 0),
+	CHAN5G(120, 5600, 0),
+	CHAN5G(124, 5620, 0),
+	CHAN5G(128, 5640, 0),
+	CHAN5G(132, 5660, 0),
+	CHAN5G(136, 5680, 0),
+	CHAN5G(140, 5700, 0),
+	CHAN5G(144, 5720, 0),
+	CHAN5G(149, 5745, 0),
+	CHAN5G(153, 5765, 0),
+	CHAN5G(157, 5785, 0),
+	CHAN5G(161, 5805, 0),
+	CHAN5G(165, 5825, 0),
+	CHAN5G(169, 5845, 0),
+	CHAN5G(173, 5865, 0),
+};
+
+static struct ieee80211_rate ath11k_legacy_rates[] = {
+	{ .bitrate = 10,
+	  .hw_value = ATH11K_HW_RATE_CCK_LP_1M },
+	{ .bitrate = 20,
+	  .hw_value = ATH11K_HW_RATE_CCK_LP_2M,
+	  .hw_value_short = ATH11K_HW_RATE_CCK_SP_2M,
+	  .flags = IEEE80211_RATE_SHORT_PREAMBLE },
+	{ .bitrate = 55,
+	  .hw_value = ATH11K_HW_RATE_CCK_LP_5_5M,
+	  .hw_value_short = ATH11K_HW_RATE_CCK_SP_5_5M,
+	  .flags = IEEE80211_RATE_SHORT_PREAMBLE },
+	{ .bitrate = 110,
+	  .hw_value = ATH11K_HW_RATE_CCK_LP_11M,
+	  .hw_value_short = ATH11K_HW_RATE_CCK_SP_11M,
+	  .flags = IEEE80211_RATE_SHORT_PREAMBLE },
+
+	{ .bitrate = 60, .hw_value = ATH11K_HW_RATE_OFDM_6M },
+	{ .bitrate = 90, .hw_value = ATH11K_HW_RATE_OFDM_9M },
+	{ .bitrate = 120, .hw_value = ATH11K_HW_RATE_OFDM_12M },
+	{ .bitrate = 180, .hw_value = ATH11K_HW_RATE_OFDM_18M },
+	{ .bitrate = 240, .hw_value = ATH11K_HW_RATE_OFDM_24M },
+	{ .bitrate = 360, .hw_value = ATH11K_HW_RATE_OFDM_36M },
+	{ .bitrate = 480, .hw_value = ATH11K_HW_RATE_OFDM_48M },
+	{ .bitrate = 540, .hw_value = ATH11K_HW_RATE_OFDM_54M },
+};
+
+static const int
+ath11k_phymodes[NUM_NL80211_BANDS][ATH11K_CHAN_WIDTH_NUM] = {
+	[NL80211_BAND_2GHZ] = {
+			[NL80211_CHAN_WIDTH_5] = MODE_UNKNOWN,
+			[NL80211_CHAN_WIDTH_10] = MODE_UNKNOWN,
+			[NL80211_CHAN_WIDTH_20_NOHT] = MODE_11AX_HE20_2G,
+			[NL80211_CHAN_WIDTH_20] = MODE_11AX_HE20_2G,
+			[NL80211_CHAN_WIDTH_40] = MODE_11AX_HE40_2G,
+			[NL80211_CHAN_WIDTH_80] = MODE_11AX_HE80_2G,
+			[NL80211_CHAN_WIDTH_80P80] = MODE_UNKNOWN,
+			[NL80211_CHAN_WIDTH_160] = MODE_UNKNOWN,
+	},
+	[NL80211_BAND_5GHZ] = {
+			[NL80211_CHAN_WIDTH_5] = MODE_UNKNOWN,
+			[NL80211_CHAN_WIDTH_10] = MODE_UNKNOWN,
+			[NL80211_CHAN_WIDTH_20_NOHT] = MODE_11AX_HE20,
+			[NL80211_CHAN_WIDTH_20] = MODE_11AX_HE20,
+			[NL80211_CHAN_WIDTH_40] = MODE_11AX_HE40,
+			[NL80211_CHAN_WIDTH_80] = MODE_11AX_HE80,
+			[NL80211_CHAN_WIDTH_160] = MODE_11AX_HE160,
+			[NL80211_CHAN_WIDTH_80P80] = MODE_11AX_HE80_80,
+	},
+};
+
+const struct htt_rx_ring_tlv_filter ath11k_mac_mon_status_filter_default = {
+	.rx_filter = HTT_RX_FILTER_TLV_FLAGS_MPDU_START |
+		     HTT_RX_FILTER_TLV_FLAGS_PPDU_END |
+		     HTT_RX_FILTER_TLV_FLAGS_PPDU_END_STATUS_DONE,
+	.pkt_filter_flags0 = HTT_RX_FP_MGMT_FILTER_FLAGS0,
+	.pkt_filter_flags1 = HTT_RX_FP_MGMT_FILTER_FLAGS1,
+	.pkt_filter_flags2 = HTT_RX_FP_CTRL_FILTER_FLASG2,
+	.pkt_filter_flags3 = HTT_RX_FP_DATA_FILTER_FLASG3 |
+			     HTT_RX_FP_CTRL_FILTER_FLASG3
+};
+
+#define ATH11K_MAC_FIRST_OFDM_RATE_IDX 4
+#define ath11k_g_rates ath11k_legacy_rates
+#define ath11k_g_rates_size (ARRAY_SIZE(ath11k_legacy_rates))
+#define ath11k_a_rates (ath11k_legacy_rates + 4)
+#define ath11k_a_rates_size (ARRAY_SIZE(ath11k_legacy_rates) - 4)
+
+#define ATH11K_MAC_SCAN_TIMEOUT_MSECS 200 /* in msecs */
+
+static const u32 ath11k_smps_map[] = {
+	[WLAN_HT_CAP_SM_PS_STATIC] = WMI_PEER_SMPS_STATIC,
+	[WLAN_HT_CAP_SM_PS_DYNAMIC] = WMI_PEER_SMPS_DYNAMIC,
+	[WLAN_HT_CAP_SM_PS_INVALID] = WMI_PEER_SMPS_PS_NONE,
+	[WLAN_HT_CAP_SM_PS_DISABLED] = WMI_PEER_SMPS_PS_NONE,
+};
+
+int ath11k_mac_hw_ratecode_to_legacy_rate(u8 hw_rc, u8 preamble, u8 *rateidx,
+					  u16 *rate)
+{
+	/* As default, it is OFDM rates */
+	int i = ATH11K_MAC_FIRST_OFDM_RATE_IDX;
+	int max_rates_idx = ath11k_g_rates_size;
+
+	if (preamble == WMI_RATE_PREAMBLE_CCK) {
+		hw_rc &= ~ATH11k_HW_RATECODE_CCK_SHORT_PREAM_MASK;
+		i = 0;
+		max_rates_idx = ATH11K_MAC_FIRST_OFDM_RATE_IDX;
+	}
+
+	while (i < max_rates_idx) {
+		if (hw_rc == ath11k_legacy_rates[i].hw_value) {
+			*rateidx = i;
+			*rate = ath11k_legacy_rates[i].bitrate;
+			return 0;
+		}
+		i++;
+	}
+
+	return -EINVAL;
+}
+
+static int get_num_chains(u32 mask)
+{
+	int num_chains = 0;
+
+	while (mask) {
+		if (mask & BIT(0))
+			num_chains++;
+		mask >>= 1;
+	}
+
+	return num_chains;
+}
+
+u8 ath11k_mac_bitrate_to_idx(const struct ieee80211_supported_band *sband,
+			     u32 bitrate)
+{
+	int i;
+
+	for (i = 0; i < sband->n_bitrates; i++)
+		if (sband->bitrates[i].bitrate == bitrate)
+			return i;
+
+	return 0;
+}
+
+static u32
+ath11k_mac_max_ht_nss(const u8 ht_mcs_mask[IEEE80211_HT_MCS_MASK_LEN])
+{
+	int nss;
+
+	for (nss = IEEE80211_HT_MCS_MASK_LEN - 1; nss >= 0; nss--)
+		if (ht_mcs_mask[nss])
+			return nss + 1;
+
+	return 1;
+}
+
+static u32
+ath11k_mac_max_vht_nss(const u16 vht_mcs_mask[NL80211_VHT_NSS_MAX])
+{
+	int nss;
+
+	for (nss = NL80211_VHT_NSS_MAX - 1; nss >= 0; nss--)
+		if (vht_mcs_mask[nss])
+			return nss + 1;
+
+	return 1;
+}
+
+static u8 ath11k_parse_mpdudensity(u8 mpdudensity)
+{
+/* 802.11n D2.0 defined values for "Minimum MPDU Start Spacing":
+ *   0 for no restriction
+ *   1 for 1/4 us
+ *   2 for 1/2 us
+ *   3 for 1 us
+ *   4 for 2 us
+ *   5 for 4 us
+ *   6 for 8 us
+ *   7 for 16 us
+ */
+	switch (mpdudensity) {
+	case 0:
+		return 0;
+	case 1:
+	case 2:
+	case 3:
+	/* Our lower layer calculations limit our precision to
+	 * 1 microsecond
+	 */
+		return 1;
+	case 4:
+		return 2;
+	case 5:
+		return 4;
+	case 6:
+		return 8;
+	case 7:
+		return 16;
+	default:
+		return 0;
+	}
+}
+
+static int ath11k_mac_vif_chan(struct ieee80211_vif *vif,
+			       struct cfg80211_chan_def *def)
+{
+	struct ieee80211_chanctx_conf *conf;
+
+	rcu_read_lock();
+	conf = rcu_dereference(vif->chanctx_conf);
+	if (!conf) {
+		rcu_read_unlock();
+		return -ENOENT;
+	}
+
+	*def = conf->def;
+	rcu_read_unlock();
+
+	return 0;
+}
+
+static bool ath11k_mac_bitrate_is_cck(int bitrate)
+{
+	switch (bitrate) {
+	case 10:
+	case 20:
+	case 55:
+	case 110:
+		return true;
+	}
+
+	return false;
+}
+
+u8 ath11k_mac_hw_rate_to_idx(const struct ieee80211_supported_band *sband,
+			     u8 hw_rate, bool cck)
+{
+	const struct ieee80211_rate *rate;
+	int i;
+
+	for (i = 0; i < sband->n_bitrates; i++) {
+		rate = &sband->bitrates[i];
+
+		if (ath11k_mac_bitrate_is_cck(rate->bitrate) != cck)
+			continue;
+
+		if (rate->hw_value == hw_rate)
+			return i;
+		else if (rate->flags & IEEE80211_RATE_SHORT_PREAMBLE &&
+			 rate->hw_value_short == hw_rate)
+			return i;
+	}
+
+	return 0;
+}
+
+static u8 ath11k_mac_bitrate_to_rate(int bitrate)
+{
+	return DIV_ROUND_UP(bitrate, 5) |
+	       (ath11k_mac_bitrate_is_cck(bitrate) ? BIT(7) : 0);
+}
+
+static void ath11k_get_arvif_iter(void *data, u8 *mac,
+				  struct ieee80211_vif *vif)
+{
+	struct ath11k_vif_iter *arvif_iter = data;
+	struct ath11k_vif *arvif = (void *)vif->drv_priv;
+
+	if (arvif->vdev_id == arvif_iter->vdev_id)
+		arvif_iter->arvif = arvif;
+}
+
+struct ath11k_vif *ath11k_mac_get_arvif(struct ath11k *ar, u32 vdev_id)
+{
+	struct ath11k_vif_iter arvif_iter;
+	u32 flags;
+
+	memset(&arvif_iter, 0, sizeof(struct ath11k_vif_iter));
+	arvif_iter.vdev_id = vdev_id;
+
+	flags = IEEE80211_IFACE_ITER_RESUME_ALL;
+	ieee80211_iterate_active_interfaces_atomic(ar->hw,
+						   flags,
+						   ath11k_get_arvif_iter,
+						   &arvif_iter);
+	if (!arvif_iter.arvif)
+		return NULL;
+
+	return arvif_iter.arvif;
+}
+
+struct ath11k_vif *ath11k_mac_get_arvif_by_vdev_id(struct ath11k_base *ab,
+						   u32 vdev_id)
+{
+	int i;
+	struct ath11k_pdev *pdev;
+	struct ath11k_vif *arvif;
+
+	for (i = 0; i < ab->num_radios; i++) {
+		pdev = rcu_dereference(ab->pdevs_active[i]);
+		if (pdev && pdev->ar) {
+			arvif = ath11k_mac_get_arvif(pdev->ar, vdev_id);
+			if (arvif)
+				return arvif;
+		}
+	}
+
+	return NULL;
+}
+
+struct ath11k *ath11k_mac_get_ar_by_vdev_id(struct ath11k_base *ab, u32 vdev_id)
+{
+	int i;
+	struct ath11k_pdev *pdev;
+	struct ath11k_vif *arvif;
+
+	for (i = 0; i < ab->num_radios; i++) {
+		pdev = rcu_dereference(ab->pdevs_active[i]);
+		if (pdev && pdev->ar) {
+			arvif = ath11k_mac_get_arvif(pdev->ar, vdev_id);
+			if (arvif)
+				return arvif->ar;
+		}
+	}
+
+	return NULL;
+}
+
+struct ath11k *ath11k_mac_get_ar_by_pdev_id(struct ath11k_base *ab, u32 pdev_id)
+{
+	int i;
+	struct ath11k_pdev *pdev;
+
+	if (WARN_ON(pdev_id > ab->num_radios))
+		return NULL;
+
+	for (i = 0; i < ab->num_radios; i++) {
+		pdev = rcu_dereference(ab->pdevs_active[i]);
+
+		if (pdev && pdev->pdev_id == pdev_id)
+			return (pdev->ar ? pdev->ar : NULL);
+	}
+
+	return NULL;
+}
+
+struct ath11k *ath11k_mac_get_ar_vdev_stop_status(struct ath11k_base *ab,
+						  u32 vdev_id)
+{
+	int i;
+	struct ath11k_pdev *pdev;
+	struct ath11k *ar;
+
+	for (i = 0; i < ab->num_radios; i++) {
+		pdev = rcu_dereference(ab->pdevs_active[i]);
+		if (pdev && pdev->ar) {
+			ar = pdev->ar;
+
+			spin_lock_bh(&ar->data_lock);
+			if (ar->vdev_stop_status.stop_in_progress &&
+			    ar->vdev_stop_status.vdev_id == vdev_id) {
+				ar->vdev_stop_status.stop_in_progress = false;
+				spin_unlock_bh(&ar->data_lock);
+				return ar;
+			}
+			spin_unlock_bh(&ar->data_lock);
+		}
+	}
+	return NULL;
+}
+
+static void ath11k_pdev_caps_update(struct ath11k *ar)
+{
+	struct ath11k_base *ab = ar->ab;
+
+	ar->max_tx_power = ab->target_caps.hw_max_tx_power;
+
+	/* FIXME Set min_tx_power to ab->target_caps.hw_min_tx_power.
+	 * But since the received value in svcrdy is same as hw_max_tx_power,
+	 * we can set ar->min_tx_power to 0 currently until
+	 * this is fixed in firmware
+	 */
+	ar->min_tx_power = 0;
+
+	ar->txpower_limit_2g = ar->max_tx_power;
+	ar->txpower_limit_5g = ar->max_tx_power;
+	ar->txpower_scale = WMI_HOST_TP_SCALE_MAX;
+}
+
+static int ath11k_mac_txpower_recalc(struct ath11k *ar)
+{
+	struct ath11k_pdev *pdev = ar->pdev;
+	struct ath11k_vif *arvif;
+	int ret, txpower = -1;
+	u32 param;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	list_for_each_entry(arvif, &ar->arvifs, list) {
+		if (arvif->txpower <= 0)
+			continue;
+
+		if (txpower == -1)
+			txpower = arvif->txpower;
+		else
+			txpower = min(txpower, arvif->txpower);
+	}
+
+	if (txpower == -1)
+		return 0;
+
+	/* txpwr is set as 2 units per dBm in FW*/
+	txpower = min_t(u32, max_t(u32, ar->min_tx_power, txpower),
+			ar->max_tx_power) * 2;
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "txpower to set in hw %d\n",
+		   txpower / 2);
+
+	if ((pdev->cap.supported_bands & WMI_HOST_WLAN_2G_CAP) &&
+	    ar->txpower_limit_2g != txpower) {
+		param = WMI_PDEV_PARAM_TXPOWER_LIMIT2G;
+		ret = ath11k_wmi_pdev_set_param(ar, param,
+						txpower, ar->pdev->pdev_id);
+		if (ret)
+			goto fail;
+		ar->txpower_limit_2g = txpower;
+	}
+
+	if ((pdev->cap.supported_bands & WMI_HOST_WLAN_5G_CAP) &&
+	    ar->txpower_limit_5g != txpower) {
+		param = WMI_PDEV_PARAM_TXPOWER_LIMIT5G;
+		ret = ath11k_wmi_pdev_set_param(ar, param,
+						txpower, ar->pdev->pdev_id);
+		if (ret)
+			goto fail;
+		ar->txpower_limit_5g = txpower;
+	}
+
+	return 0;
+
+fail:
+	ath11k_warn(ar->ab, "failed to recalc txpower limit %d using pdev param %d: %d\n",
+		    txpower / 2, param, ret);
+	return ret;
+}
+
+static int ath11k_recalc_rtscts_prot(struct ath11k_vif *arvif)
+{
+	struct ath11k *ar = arvif->ar;
+	u32 vdev_param, rts_cts = 0;
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	vdev_param = WMI_VDEV_PARAM_ENABLE_RTSCTS;
+
+	/* Enable RTS/CTS protection for sw retries (when legacy stations
+	 * are in BSS) or by default only for second rate series.
+	 * TODO: Check if we need to enable CTS 2 Self in any case
+	 */
+	rts_cts = WMI_USE_RTS_CTS;
+
+	if (arvif->num_legacy_stations > 0)
+		rts_cts |= WMI_RTSCTS_ACROSS_SW_RETRIES << 4;
+	else
+		rts_cts |= WMI_RTSCTS_FOR_SECOND_RATESERIES << 4;
+
+	/* Need not send duplicate param value to firmware */
+	if (arvif->rtscts_prot_mode == rts_cts)
+		return 0;
+
+	arvif->rtscts_prot_mode = rts_cts;
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "mac vdev %d recalc rts/cts prot %d\n",
+		   arvif->vdev_id, rts_cts);
+
+	ret =  ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id,
+					     vdev_param, rts_cts);
+	if (ret)
+		ath11k_warn(ar->ab, "failed to recalculate rts/cts prot for vdev %d: %d\n",
+			    arvif->vdev_id, ret);
+
+	return ret;
+}
+
+static int ath11k_mac_set_kickout(struct ath11k_vif *arvif)
+{
+	struct ath11k *ar = arvif->ar;
+	u32 param;
+	int ret;
+
+	ret = ath11k_wmi_pdev_set_param(ar, WMI_PDEV_PARAM_STA_KICKOUT_TH,
+					ATH11K_KICKOUT_THRESHOLD,
+					ar->pdev->pdev_id);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to set kickout threshold on vdev %i: %d\n",
+			    arvif->vdev_id, ret);
+		return ret;
+	}
+
+	param = WMI_VDEV_PARAM_AP_KEEPALIVE_MIN_IDLE_INACTIVE_TIME_SECS;
+	ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id, param,
+					    ATH11K_KEEPALIVE_MIN_IDLE);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to set keepalive minimum idle time on vdev %i: %d\n",
+			    arvif->vdev_id, ret);
+		return ret;
+	}
+
+	param = WMI_VDEV_PARAM_AP_KEEPALIVE_MAX_IDLE_INACTIVE_TIME_SECS;
+	ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id, param,
+					    ATH11K_KEEPALIVE_MAX_IDLE);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to set keepalive maximum idle time on vdev %i: %d\n",
+			    arvif->vdev_id, ret);
+		return ret;
+	}
+
+	param = WMI_VDEV_PARAM_AP_KEEPALIVE_MAX_UNRESPONSIVE_TIME_SECS;
+	ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id, param,
+					    ATH11K_KEEPALIVE_MAX_UNRESPONSIVE);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to set keepalive maximum unresponsive time on vdev %i: %d\n",
+			    arvif->vdev_id, ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+void ath11k_mac_peer_cleanup_all(struct ath11k *ar)
+{
+	struct ath11k_peer *peer, *tmp;
+	struct ath11k_base *ab = ar->ab;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	spin_lock_bh(&ab->base_lock);
+	list_for_each_entry_safe(peer, tmp, &ab->peers, list) {
+		ath11k_peer_rx_tid_cleanup(ar, peer);
+		list_del(&peer->list);
+		kfree(peer);
+	}
+	spin_unlock_bh(&ab->base_lock);
+
+	ar->num_peers = 0;
+	ar->num_stations = 0;
+}
+
+static int ath11k_monitor_vdev_up(struct ath11k *ar, int vdev_id)
+{
+	int ret = 0;
+
+	ret = ath11k_wmi_vdev_up(ar, vdev_id, 0, ar->mac_addr);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to put up monitor vdev %i: %d\n",
+			    vdev_id, ret);
+		return ret;
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "mac monitor vdev %i started\n",
+		   vdev_id);
+	return 0;
+}
+
+static int ath11k_mac_op_config(struct ieee80211_hw *hw, u32 changed)
+{
+	struct ath11k *ar = hw->priv;
+	int ret = 0;
+
+	/* mac80211 requires this op to be present and that's why
+	 * there's an empty function, this can be extended when
+	 * required.
+	 */
+
+	mutex_lock(&ar->conf_mutex);
+
+	/* TODO: Handle configuration changes as appropriate */
+
+	mutex_unlock(&ar->conf_mutex);
+
+	return ret;
+}
+
+static int ath11k_mac_setup_bcn_tmpl(struct ath11k_vif *arvif)
+{
+	struct ath11k *ar = arvif->ar;
+	struct ath11k_base *ab = ar->ab;
+	struct ieee80211_hw *hw = ar->hw;
+	struct ieee80211_vif *vif = arvif->vif;
+	struct ieee80211_mutable_offsets offs = {};
+	struct sk_buff *bcn;
+	int ret;
+
+	if (arvif->vdev_type != WMI_VDEV_TYPE_AP)
+		return 0;
+
+	bcn = ieee80211_beacon_get_template(hw, vif, &offs);
+	if (!bcn) {
+		ath11k_warn(ab, "failed to get beacon template from mac80211\n");
+		return -EPERM;
+	}
+
+	ret = ath11k_wmi_bcn_tmpl(ar, arvif->vdev_id, &offs, bcn);
+
+	kfree_skb(bcn);
+
+	if (ret)
+		ath11k_warn(ab, "failed to submit beacon template command: %d\n",
+			    ret);
+
+	return ret;
+}
+
+static void ath11k_control_beaconing(struct ath11k_vif *arvif,
+				     struct ieee80211_bss_conf *info)
+{
+	struct ath11k *ar = arvif->ar;
+	int ret = 0;
+
+	lockdep_assert_held(&arvif->ar->conf_mutex);
+
+	if (!info->enable_beacon) {
+		ret = ath11k_wmi_vdev_down(ar, arvif->vdev_id);
+		if (ret)
+			ath11k_warn(ar->ab, "failed to down vdev_id %i: %d\n",
+				    arvif->vdev_id, ret);
+
+		arvif->is_up = false;
+		return;
+	}
+
+	/* Install the beacon template to the FW */
+	ret = ath11k_mac_setup_bcn_tmpl(arvif);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to update bcn tmpl during vdev up: %d\n",
+			    ret);
+		return;
+	}
+
+	arvif->tx_seq_no = 0x1000;
+
+	arvif->aid = 0;
+
+	ether_addr_copy(arvif->bssid, info->bssid);
+
+	ret = ath11k_wmi_vdev_up(arvif->ar, arvif->vdev_id, arvif->aid,
+				 arvif->bssid);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to bring up vdev %d: %i\n",
+			    arvif->vdev_id, ret);
+		return;
+	}
+
+	arvif->is_up = true;
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "mac vdev %d up\n", arvif->vdev_id);
+}
+
+static void ath11k_peer_assoc_h_basic(struct ath11k *ar,
+				      struct ieee80211_vif *vif,
+				      struct ieee80211_sta *sta,
+				      struct peer_assoc_params *arg)
+{
+	struct ath11k_vif *arvif = (void *)vif->drv_priv;
+	u32 aid;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	if (vif->type == NL80211_IFTYPE_STATION)
+		aid = vif->bss_conf.aid;
+	else
+		aid = sta->aid;
+
+	ether_addr_copy(arg->peer_mac, sta->addr);
+	arg->vdev_id = arvif->vdev_id;
+	arg->peer_associd = aid;
+	arg->auth_flag = true;
+	/* TODO: STA WAR in ath10k for listen interval required? */
+	arg->peer_listen_intval = ar->hw->conf.listen_interval;
+	arg->peer_nss = 1;
+	arg->peer_caps = vif->bss_conf.assoc_capability;
+}
+
+static void ath11k_peer_assoc_h_crypto(struct ath11k *ar,
+				       struct ieee80211_vif *vif,
+				       struct ieee80211_sta *sta,
+				       struct peer_assoc_params *arg)
+{
+	struct ieee80211_bss_conf *info = &vif->bss_conf;
+	struct cfg80211_chan_def def;
+	struct cfg80211_bss *bss;
+	const u8 *rsnie = NULL;
+	const u8 *wpaie = NULL;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	if (WARN_ON(ath11k_mac_vif_chan(vif, &def)))
+		return;
+
+	bss = cfg80211_get_bss(ar->hw->wiphy, def.chan, info->bssid, NULL, 0,
+			       IEEE80211_BSS_TYPE_ANY, IEEE80211_PRIVACY_ANY);
+	if (bss) {
+		const struct cfg80211_bss_ies *ies;
+
+		rcu_read_lock();
+		rsnie = ieee80211_bss_get_ie(bss, WLAN_EID_RSN);
+
+		ies = rcu_dereference(bss->ies);
+
+		wpaie = cfg80211_find_vendor_ie(WLAN_OUI_MICROSOFT,
+						WLAN_OUI_TYPE_MICROSOFT_WPA,
+						ies->data,
+						ies->len);
+		rcu_read_unlock();
+		cfg80211_put_bss(ar->hw->wiphy, bss);
+	}
+
+	/* FIXME: base on RSN IE/WPA IE is a correct idea? */
+	if (rsnie || wpaie) {
+		ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+			   "%s: rsn ie found\n", __func__);
+		arg->need_ptk_4_way = true;
+	}
+
+	if (wpaie) {
+		ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+			   "%s: wpa ie found\n", __func__);
+		arg->need_gtk_2_way = true;
+	}
+
+	if (sta->mfp) {
+		/* TODO: Need to check if FW supports PMF? */
+		arg->is_pmf_enabled = true;
+	}
+
+	/* TODO: safe_mode_enabled (bypass 4-way handshake) flag req? */
+}
+
+static void ath11k_peer_assoc_h_rates(struct ath11k *ar,
+				      struct ieee80211_vif *vif,
+				      struct ieee80211_sta *sta,
+				      struct peer_assoc_params *arg)
+{
+	struct ath11k_vif *arvif = (void *)vif->drv_priv;
+	struct wmi_rate_set_arg *rateset = &arg->peer_legacy_rates;
+	struct cfg80211_chan_def def;
+	const struct ieee80211_supported_band *sband;
+	const struct ieee80211_rate *rates;
+	enum nl80211_band band;
+	u32 ratemask;
+	u8 rate;
+	int i;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	if (WARN_ON(ath11k_mac_vif_chan(vif, &def)))
+		return;
+
+	band = def.chan->band;
+	sband = ar->hw->wiphy->bands[band];
+	ratemask = sta->supp_rates[band];
+	ratemask &= arvif->bitrate_mask.control[band].legacy;
+	rates = sband->bitrates;
+
+	rateset->num_rates = 0;
+
+	for (i = 0; i < 32; i++, ratemask >>= 1, rates++) {
+		if (!(ratemask & 1))
+			continue;
+
+		rate = ath11k_mac_bitrate_to_rate(rates->bitrate);
+		rateset->rates[rateset->num_rates] = rate;
+		rateset->num_rates++;
+	}
+}
+
+static bool
+ath11k_peer_assoc_h_ht_masked(const u8 ht_mcs_mask[IEEE80211_HT_MCS_MASK_LEN])
+{
+	int nss;
+
+	for (nss = 0; nss < IEEE80211_HT_MCS_MASK_LEN; nss++)
+		if (ht_mcs_mask[nss])
+			return false;
+
+	return true;
+}
+
+static bool
+ath11k_peer_assoc_h_vht_masked(const u16 vht_mcs_mask[NL80211_VHT_NSS_MAX])
+{
+	int nss;
+
+	for (nss = 0; nss < NL80211_VHT_NSS_MAX; nss++)
+		if (vht_mcs_mask[nss])
+			return false;
+
+	return true;
+}
+
+static void ath11k_peer_assoc_h_ht(struct ath11k *ar,
+				   struct ieee80211_vif *vif,
+				   struct ieee80211_sta *sta,
+				   struct peer_assoc_params *arg)
+{
+	const struct ieee80211_sta_ht_cap *ht_cap = &sta->ht_cap;
+	struct ath11k_vif *arvif = (void *)vif->drv_priv;
+	struct cfg80211_chan_def def;
+	enum nl80211_band band;
+	const u8 *ht_mcs_mask;
+	int i, n;
+	u8 max_nss;
+	u32 stbc;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	if (WARN_ON(ath11k_mac_vif_chan(vif, &def)))
+		return;
+
+	if (!ht_cap->ht_supported)
+		return;
+
+	band = def.chan->band;
+	ht_mcs_mask = arvif->bitrate_mask.control[band].ht_mcs;
+
+	if (ath11k_peer_assoc_h_ht_masked(ht_mcs_mask))
+		return;
+
+	arg->ht_flag = true;
+
+	arg->peer_max_mpdu = (1 << (IEEE80211_HT_MAX_AMPDU_FACTOR +
+				    ht_cap->ampdu_factor)) - 1;
+
+	arg->peer_mpdu_density =
+		ath11k_parse_mpdudensity(ht_cap->ampdu_density);
+
+	arg->peer_ht_caps = ht_cap->cap;
+	arg->peer_rate_caps |= WMI_HOST_RC_HT_FLAG;
+
+	if (ht_cap->cap & IEEE80211_HT_CAP_LDPC_CODING)
+		arg->ldpc_flag = true;
+
+	if (sta->bandwidth >= IEEE80211_STA_RX_BW_40) {
+		arg->bw_40 = true;
+		arg->peer_rate_caps |= WMI_HOST_RC_CW40_FLAG;
+	}
+
+	if (arvif->bitrate_mask.control[band].gi != NL80211_TXRATE_FORCE_LGI) {
+		if (ht_cap->cap & (IEEE80211_HT_CAP_SGI_20 |
+		    IEEE80211_HT_CAP_SGI_40))
+			arg->peer_rate_caps |= WMI_HOST_RC_SGI_FLAG;
+	}
+
+	if (ht_cap->cap & IEEE80211_HT_CAP_TX_STBC) {
+		arg->peer_rate_caps |= WMI_HOST_RC_TX_STBC_FLAG;
+		arg->stbc_flag = true;
+	}
+
+	if (ht_cap->cap & IEEE80211_HT_CAP_RX_STBC) {
+		stbc = ht_cap->cap & IEEE80211_HT_CAP_RX_STBC;
+		stbc = stbc >> IEEE80211_HT_CAP_RX_STBC_SHIFT;
+		stbc = stbc << WMI_HOST_RC_RX_STBC_FLAG_S;
+		arg->peer_rate_caps |= stbc;
+		arg->stbc_flag = true;
+	}
+
+	if (ht_cap->mcs.rx_mask[1] && ht_cap->mcs.rx_mask[2])
+		arg->peer_rate_caps |= WMI_HOST_RC_TS_FLAG;
+	else if (ht_cap->mcs.rx_mask[1])
+		arg->peer_rate_caps |= WMI_HOST_RC_DS_FLAG;
+
+	for (i = 0, n = 0, max_nss = 0; i < IEEE80211_HT_MCS_MASK_LEN * 8; i++)
+		if ((ht_cap->mcs.rx_mask[i / 8] & BIT(i % 8)) &&
+		    (ht_mcs_mask[i / 8] & BIT(i % 8))) {
+			max_nss = (i / 8) + 1;
+			arg->peer_ht_rates.rates[n++] = i;
+		}
+
+	/* This is a workaround for HT-enabled STAs which break the spec
+	 * and have no HT capabilities RX mask (no HT RX MCS map).
+	 *
+	 * As per spec, in section 20.3.5 Modulation and coding scheme (MCS),
+	 * MCS 0 through 7 are mandatory in 20MHz with 800 ns GI at all STAs.
+	 *
+	 * Firmware asserts if such situation occurs.
+	 */
+	if (n == 0) {
+		arg->peer_ht_rates.num_rates = 8;
+		for (i = 0; i < arg->peer_ht_rates.num_rates; i++)
+			arg->peer_ht_rates.rates[i] = i;
+	} else {
+		arg->peer_ht_rates.num_rates = n;
+		arg->peer_nss = min(sta->rx_nss, max_nss);
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "mac ht peer %pM mcs cnt %d nss %d\n",
+		   arg->peer_mac,
+		   arg->peer_ht_rates.num_rates,
+		   arg->peer_nss);
+}
+
+static int ath11k_mac_get_max_vht_mcs_map(u16 mcs_map, int nss)
+{
+	switch ((mcs_map >> (2 * nss)) & 0x3) {
+	case IEEE80211_VHT_MCS_SUPPORT_0_7: return BIT(8) - 1;
+	case IEEE80211_VHT_MCS_SUPPORT_0_8: return BIT(9) - 1;
+	case IEEE80211_VHT_MCS_SUPPORT_0_9: return BIT(10) - 1;
+	}
+	return 0;
+}
+
+static u16
+ath11k_peer_assoc_h_vht_limit(u16 tx_mcs_set,
+			      const u16 vht_mcs_limit[NL80211_VHT_NSS_MAX])
+{
+	int idx_limit;
+	int nss;
+	u16 mcs_map;
+	u16 mcs;
+
+	for (nss = 0; nss < NL80211_VHT_NSS_MAX; nss++) {
+		mcs_map = ath11k_mac_get_max_vht_mcs_map(tx_mcs_set, nss) &
+			  vht_mcs_limit[nss];
+
+		if (mcs_map)
+			idx_limit = fls(mcs_map) - 1;
+		else
+			idx_limit = -1;
+
+		switch (idx_limit) {
+		case 0: /* fall through */
+		case 1: /* fall through */
+		case 2: /* fall through */
+		case 3: /* fall through */
+		case 4: /* fall through */
+		case 5: /* fall through */
+		case 6: /* fall through */
+		case 7:
+			mcs = IEEE80211_VHT_MCS_SUPPORT_0_7;
+			break;
+		case 8:
+			mcs = IEEE80211_VHT_MCS_SUPPORT_0_8;
+			break;
+		case 9:
+			mcs = IEEE80211_VHT_MCS_SUPPORT_0_9;
+			break;
+		default:
+			WARN_ON(1);
+			/* fall through */
+		case -1:
+			mcs = IEEE80211_VHT_MCS_NOT_SUPPORTED;
+			break;
+		}
+
+		tx_mcs_set &= ~(0x3 << (nss * 2));
+		tx_mcs_set |= mcs << (nss * 2);
+	}
+
+	return tx_mcs_set;
+}
+
+static void ath11k_peer_assoc_h_vht(struct ath11k *ar,
+				    struct ieee80211_vif *vif,
+				    struct ieee80211_sta *sta,
+				    struct peer_assoc_params *arg)
+{
+	const struct ieee80211_sta_vht_cap *vht_cap = &sta->vht_cap;
+	struct ath11k_vif *arvif = (void *)vif->drv_priv;
+	struct cfg80211_chan_def def;
+	enum nl80211_band band;
+	const u16 *vht_mcs_mask;
+	u8 ampdu_factor;
+	u8 max_nss, vht_mcs;
+	int i;
+
+	if (WARN_ON(ath11k_mac_vif_chan(vif, &def)))
+		return;
+
+	if (!vht_cap->vht_supported)
+		return;
+
+	band = def.chan->band;
+	vht_mcs_mask = arvif->bitrate_mask.control[band].vht_mcs;
+
+	if (ath11k_peer_assoc_h_vht_masked(vht_mcs_mask))
+		return;
+
+	arg->vht_flag = true;
+
+	/* TODO: similar flags required? */
+	arg->vht_capable = true;
+
+	if (def.chan->band == NL80211_BAND_2GHZ)
+		arg->vht_ng_flag = true;
+
+	arg->peer_vht_caps = vht_cap->cap;
+
+	ampdu_factor = (vht_cap->cap &
+			IEEE80211_VHT_CAP_MAX_A_MPDU_LENGTH_EXPONENT_MASK) >>
+		       IEEE80211_VHT_CAP_MAX_A_MPDU_LENGTH_EXPONENT_SHIFT;
+
+	/* Workaround: Some Netgear/Linksys 11ac APs set Rx A-MPDU factor to
+	 * zero in VHT IE. Using it would result in degraded throughput.
+	 * arg->peer_max_mpdu at this point contains HT max_mpdu so keep
+	 * it if VHT max_mpdu is smaller.
+	 */
+	arg->peer_max_mpdu = max(arg->peer_max_mpdu,
+				 (1U << (IEEE80211_HT_MAX_AMPDU_FACTOR +
+					ampdu_factor)) - 1);
+
+	if (sta->bandwidth == IEEE80211_STA_RX_BW_80)
+		arg->bw_80 = true;
+
+	if (sta->bandwidth == IEEE80211_STA_RX_BW_160)
+		arg->bw_160 = true;
+
+	/* Calculate peer NSS capability from VHT capabilities if STA
+	 * supports VHT.
+	 */
+	for (i = 0, max_nss = 0, vht_mcs = 0; i < NL80211_VHT_NSS_MAX; i++) {
+		vht_mcs = __le16_to_cpu(vht_cap->vht_mcs.rx_mcs_map) >>
+			  (2 * i) & 3;
+
+		if (vht_mcs != IEEE80211_VHT_MCS_NOT_SUPPORTED &&
+		    vht_mcs_mask[i])
+			max_nss = i + 1;
+	}
+	arg->peer_nss = min(sta->rx_nss, max_nss);
+	arg->rx_max_rate = __le16_to_cpu(vht_cap->vht_mcs.rx_highest);
+	arg->rx_mcs_set = __le16_to_cpu(vht_cap->vht_mcs.rx_mcs_map);
+	arg->tx_max_rate = __le16_to_cpu(vht_cap->vht_mcs.tx_highest);
+	arg->tx_mcs_set = ath11k_peer_assoc_h_vht_limit(
+		__le16_to_cpu(vht_cap->vht_mcs.tx_mcs_map), vht_mcs_mask);
+
+	/* In IPQ8074 platform, VHT mcs rate 10 and 11 is enabled by default.
+	 * VHT mcs rate 10 and 11 is not suppoerted in 11ac standard.
+	 * so explicitly disable the VHT MCS rate 10 and 11 in 11ac mode.
+	 */
+	arg->tx_mcs_set &= ~IEEE80211_VHT_MCS_SUPPORT_0_11_MASK;
+	arg->tx_mcs_set |= IEEE80211_DISABLE_VHT_MCS_SUPPORT_0_11;
+
+	/* TODO:  Check */
+	arg->tx_max_mcs_nss = 0xFF;
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "mac vht peer %pM max_mpdu %d flags 0x%x\n",
+		   sta->addr, arg->peer_max_mpdu, arg->peer_flags);
+
+	/* TODO: rxnss_override */
+}
+
+static void ath11k_peer_assoc_h_he(struct ath11k *ar,
+				   struct ieee80211_vif *vif,
+				   struct ieee80211_sta *sta,
+				   struct peer_assoc_params *arg)
+{
+	/* TODO: Implementation */
+}
+
+static void ath11k_peer_assoc_h_smps(struct ieee80211_sta *sta,
+				     struct peer_assoc_params *arg)
+{
+	const struct ieee80211_sta_ht_cap *ht_cap = &sta->ht_cap;
+	int smps;
+
+	if (!ht_cap->ht_supported)
+		return;
+
+	smps = ht_cap->cap & IEEE80211_HT_CAP_SM_PS;
+	smps >>= IEEE80211_HT_CAP_SM_PS_SHIFT;
+
+	switch (smps) {
+	case WLAN_HT_CAP_SM_PS_STATIC:
+		arg->static_mimops_flag = true;
+		break;
+	case WLAN_HT_CAP_SM_PS_DYNAMIC:
+		arg->dynamic_mimops_flag = true;
+		break;
+	case WLAN_HT_CAP_SM_PS_DISABLED:
+		arg->spatial_mux_flag = true;
+		break;
+	default:
+		break;
+	}
+}
+
+static void ath11k_peer_assoc_h_qos(struct ath11k *ar,
+				    struct ieee80211_vif *vif,
+				    struct ieee80211_sta *sta,
+				    struct peer_assoc_params *arg)
+{
+	struct ath11k_vif *arvif = (void *)vif->drv_priv;
+
+	switch (arvif->vdev_type) {
+	case WMI_VDEV_TYPE_AP:
+		if (sta->wme) {
+			/* TODO: Check WME vs QoS */
+			arg->is_wme_set = true;
+			arg->qos_flag = true;
+		}
+
+		if (sta->wme && sta->uapsd_queues) {
+			/* TODO: Check WME vs QoS */
+			arg->is_wme_set = true;
+			arg->apsd_flag = true;
+			arg->peer_rate_caps |= WMI_HOST_RC_UAPSD_FLAG;
+		}
+		break;
+	case WMI_VDEV_TYPE_STA:
+		if (sta->wme) {
+			arg->is_wme_set = true;
+			arg->qos_flag = true;
+		}
+		break;
+	default:
+		break;
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "mac peer %pM qos %d\n",
+		   sta->addr, arg->qos_flag);
+}
+
+static int ath11k_peer_assoc_qos_ap(struct ath11k *ar,
+				    struct ath11k_vif *arvif,
+				    struct ieee80211_sta *sta)
+{
+	struct ap_ps_params params;
+	u32 max_sp;
+	u32 uapsd;
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	params.vdev_id = arvif->vdev_id;
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "mac uapsd_queues 0x%x max_sp %d\n",
+		   sta->uapsd_queues, sta->max_sp);
+
+	uapsd = 0;
+	if (sta->uapsd_queues & IEEE80211_WMM_IE_STA_QOSINFO_AC_VO)
+		uapsd |= WMI_AP_PS_UAPSD_AC3_DELIVERY_EN |
+			 WMI_AP_PS_UAPSD_AC3_TRIGGER_EN;
+	if (sta->uapsd_queues & IEEE80211_WMM_IE_STA_QOSINFO_AC_VI)
+		uapsd |= WMI_AP_PS_UAPSD_AC2_DELIVERY_EN |
+			 WMI_AP_PS_UAPSD_AC2_TRIGGER_EN;
+	if (sta->uapsd_queues & IEEE80211_WMM_IE_STA_QOSINFO_AC_BK)
+		uapsd |= WMI_AP_PS_UAPSD_AC1_DELIVERY_EN |
+			 WMI_AP_PS_UAPSD_AC1_TRIGGER_EN;
+	if (sta->uapsd_queues & IEEE80211_WMM_IE_STA_QOSINFO_AC_BE)
+		uapsd |= WMI_AP_PS_UAPSD_AC0_DELIVERY_EN |
+			 WMI_AP_PS_UAPSD_AC0_TRIGGER_EN;
+
+	max_sp = 0;
+	if (sta->max_sp < MAX_WMI_AP_PS_PEER_PARAM_MAX_SP)
+		max_sp = sta->max_sp;
+
+	params.param = WMI_AP_PS_PEER_PARAM_UAPSD;
+	params.value = uapsd;
+	ret = ath11k_wmi_send_set_ap_ps_param_cmd(ar, sta->addr, &params);
+	if (ret)
+		goto err;
+
+	params.param = WMI_AP_PS_PEER_PARAM_MAX_SP;
+	params.value = max_sp;
+	ret = ath11k_wmi_send_set_ap_ps_param_cmd(ar, sta->addr, &params);
+	if (ret)
+		goto err;
+
+	/* TODO revisit during testing */
+	params.param = WMI_AP_PS_PEER_PARAM_SIFS_RESP_FRMTYPE;
+	params.value = DISABLE_SIFS_RESPONSE_TRIGGER;
+	ret = ath11k_wmi_send_set_ap_ps_param_cmd(ar, sta->addr, &params);
+	if (ret)
+		goto err;
+
+	params.param = WMI_AP_PS_PEER_PARAM_SIFS_RESP_UAPSD;
+	params.value = DISABLE_SIFS_RESPONSE_TRIGGER;
+	ret = ath11k_wmi_send_set_ap_ps_param_cmd(ar, sta->addr, &params);
+	if (ret)
+		goto err;
+
+	return 0;
+
+err:
+	ath11k_warn(ar->ab, "failed to set ap ps peer param %d for vdev %i: %d\n",
+		    params.param, arvif->vdev_id, ret);
+	return ret;
+}
+
+static bool ath11k_mac_sta_has_ofdm_only(struct ieee80211_sta *sta)
+{
+	return sta->supp_rates[NL80211_BAND_2GHZ] >>
+	       ATH11K_MAC_FIRST_OFDM_RATE_IDX;
+}
+
+static enum wmi_phy_mode ath11k_mac_get_phymode_vht(struct ath11k *ar,
+						    struct ieee80211_sta *sta)
+{
+	if (sta->bandwidth == IEEE80211_STA_RX_BW_160) {
+		switch (sta->vht_cap.cap &
+			IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_MASK) {
+		case IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_160MHZ:
+			return MODE_11AC_VHT160;
+		case IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_160_80PLUS80MHZ:
+			return MODE_11AC_VHT80_80;
+		default:
+			/* not sure if this is a valid case? */
+			return MODE_11AC_VHT160;
+		}
+	}
+
+	if (sta->bandwidth == IEEE80211_STA_RX_BW_80)
+		return MODE_11AC_VHT80;
+
+	if (sta->bandwidth == IEEE80211_STA_RX_BW_40)
+		return MODE_11AC_VHT40;
+
+	if (sta->bandwidth == IEEE80211_STA_RX_BW_20)
+		return MODE_11AC_VHT20;
+
+	return MODE_UNKNOWN;
+}
+
+static void ath11k_peer_assoc_h_phymode(struct ath11k *ar,
+					struct ieee80211_vif *vif,
+					struct ieee80211_sta *sta,
+					struct peer_assoc_params *arg)
+{
+	struct ath11k_vif *arvif = (void *)vif->drv_priv;
+	struct cfg80211_chan_def def;
+	enum nl80211_band band;
+	const u8 *ht_mcs_mask;
+	const u16 *vht_mcs_mask;
+	enum wmi_phy_mode phymode = MODE_UNKNOWN;
+
+	if (WARN_ON(ath11k_mac_vif_chan(vif, &def)))
+		return;
+
+	band = def.chan->band;
+	ht_mcs_mask = arvif->bitrate_mask.control[band].ht_mcs;
+	vht_mcs_mask = arvif->bitrate_mask.control[band].vht_mcs;
+
+	switch (band) {
+	case NL80211_BAND_2GHZ:
+		if (sta->vht_cap.vht_supported &&
+		    !ath11k_peer_assoc_h_vht_masked(vht_mcs_mask)) {
+			if (sta->bandwidth == IEEE80211_STA_RX_BW_40)
+				phymode = MODE_11AC_VHT40;
+			else
+				phymode = MODE_11AC_VHT20;
+		} else if (sta->ht_cap.ht_supported &&
+			   !ath11k_peer_assoc_h_ht_masked(ht_mcs_mask)) {
+			if (sta->bandwidth == IEEE80211_STA_RX_BW_40)
+				phymode = MODE_11NG_HT40;
+			else
+				phymode = MODE_11NG_HT20;
+		} else if (ath11k_mac_sta_has_ofdm_only(sta)) {
+			phymode = MODE_11G;
+		} else {
+			phymode = MODE_11B;
+		}
+		/* TODO: HE */
+
+		break;
+	case NL80211_BAND_5GHZ:
+		/* Check VHT first */
+		if (sta->vht_cap.vht_supported &&
+		    !ath11k_peer_assoc_h_vht_masked(vht_mcs_mask)) {
+			phymode = ath11k_mac_get_phymode_vht(ar, sta);
+		} else if (sta->ht_cap.ht_supported &&
+			   !ath11k_peer_assoc_h_ht_masked(ht_mcs_mask)) {
+			if (sta->bandwidth >= IEEE80211_STA_RX_BW_40)
+				phymode = MODE_11NA_HT40;
+			else
+				phymode = MODE_11NA_HT20;
+		} else {
+			phymode = MODE_11A;
+		}
+		/* TODO: HE Phymode */
+		break;
+	default:
+		break;
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "mac peer %pM phymode %s\n",
+		   sta->addr, ath11k_wmi_phymode_str(phymode));
+
+	arg->peer_phymode = phymode;
+	WARN_ON(phymode == MODE_UNKNOWN);
+}
+
+static void ath11k_peer_assoc_prepare(struct ath11k *ar,
+				      struct ieee80211_vif *vif,
+				      struct ieee80211_sta *sta,
+				      struct peer_assoc_params *arg,
+				      bool reassoc)
+{
+	lockdep_assert_held(&ar->conf_mutex);
+
+	memset(arg, 0, sizeof(*arg));
+
+	reinit_completion(&ar->peer_assoc_done);
+
+	arg->peer_new_assoc = !reassoc;
+	ath11k_peer_assoc_h_basic(ar, vif, sta, arg);
+	ath11k_peer_assoc_h_crypto(ar, vif, sta, arg);
+	ath11k_peer_assoc_h_rates(ar, vif, sta, arg);
+	ath11k_peer_assoc_h_ht(ar, vif, sta, arg);
+	ath11k_peer_assoc_h_vht(ar, vif, sta, arg);
+	ath11k_peer_assoc_h_he(ar, vif, sta, arg);
+	ath11k_peer_assoc_h_qos(ar, vif, sta, arg);
+	ath11k_peer_assoc_h_phymode(ar, vif, sta, arg);
+	ath11k_peer_assoc_h_smps(sta, arg);
+
+	/* TODO: amsdu_disable req? */
+}
+
+static int ath11k_setup_peer_smps(struct ath11k *ar, struct ath11k_vif *arvif,
+				  const u8 *addr,
+				  const struct ieee80211_sta_ht_cap *ht_cap)
+{
+	int smps;
+
+	if (!ht_cap->ht_supported)
+		return 0;
+
+	smps = ht_cap->cap & IEEE80211_HT_CAP_SM_PS;
+	smps >>= IEEE80211_HT_CAP_SM_PS_SHIFT;
+
+	if (smps >= ARRAY_SIZE(ath11k_smps_map))
+		return -EINVAL;
+
+	return ath11k_wmi_set_peer_param(ar, addr, arvif->vdev_id,
+					 WMI_PEER_MIMO_PS_STATE,
+					 ath11k_smps_map[smps]);
+}
+
+static void ath11k_bss_assoc(struct ieee80211_hw *hw,
+			     struct ieee80211_vif *vif,
+			     struct ieee80211_bss_conf *bss_conf)
+{
+	struct ath11k *ar = hw->priv;
+	struct ath11k_vif *arvif = (void *)vif->drv_priv;
+	struct peer_assoc_params peer_arg;
+	struct ieee80211_sta *ap_sta;
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "mac vdev %i assoc bssid %pM aid %d\n",
+		   arvif->vdev_id, arvif->bssid, arvif->aid);
+
+	rcu_read_lock();
+
+	ap_sta = ieee80211_find_sta(vif, bss_conf->bssid);
+	if (!ap_sta) {
+		ath11k_warn(ar->ab, "failed to find station entry for bss %pM vdev %i\n",
+			    bss_conf->bssid, arvif->vdev_id);
+		rcu_read_unlock();
+		return;
+	}
+
+	ath11k_peer_assoc_prepare(ar, vif, ap_sta, &peer_arg, false);
+
+	rcu_read_unlock();
+
+	ret = ath11k_wmi_send_peer_assoc_cmd(ar, &peer_arg);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to run peer assoc for %pM vdev %i: %d\n",
+			    bss_conf->bssid, arvif->vdev_id, ret);
+		return;
+	}
+
+	if (!wait_for_completion_timeout(&ar->peer_assoc_done, 1 * HZ)) {
+		ath11k_warn(ar->ab, "failed to get peer assoc conf event for %pM vdev %i\n",
+			    bss_conf->bssid, arvif->vdev_id);
+		return;
+	}
+
+	ret = ath11k_setup_peer_smps(ar, arvif, bss_conf->bssid,
+				     &ap_sta->ht_cap);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to setup peer SMPS for vdev %d: %d\n",
+			    arvif->vdev_id, ret);
+		return;
+	}
+
+	WARN_ON(arvif->is_up);
+
+	arvif->aid = bss_conf->aid;
+	ether_addr_copy(arvif->bssid, bss_conf->bssid);
+
+	ret = ath11k_wmi_vdev_up(ar, arvif->vdev_id, arvif->aid, arvif->bssid);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to set vdev %d up: %d\n",
+			    arvif->vdev_id, ret);
+		return;
+	}
+
+	arvif->is_up = true;
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_MAC,
+		   "mac vdev %d up (associated) bssid %pM aid %d\n",
+		   arvif->vdev_id, bss_conf->bssid, bss_conf->aid);
+
+	/* Authorize BSS Peer */
+	ret = ath11k_wmi_set_peer_param(ar, arvif->bssid,
+					arvif->vdev_id,
+					WMI_PEER_AUTHORIZE,
+					1);
+	if (ret)
+		ath11k_warn(ar->ab, "Unable to authorize BSS peer: %d\n", ret);
+}
+
+static void ath11k_bss_disassoc(struct ieee80211_hw *hw,
+				struct ieee80211_vif *vif)
+{
+	struct ath11k *ar = hw->priv;
+	struct ath11k_vif *arvif = (void *)vif->drv_priv;
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "mac vdev %i disassoc bssid %pM\n",
+		   arvif->vdev_id, arvif->bssid);
+
+	ret = ath11k_wmi_vdev_down(ar, arvif->vdev_id);
+	if (ret)
+		ath11k_warn(ar->ab, "failed to down vdev %i: %d\n",
+			    arvif->vdev_id, ret);
+
+	arvif->is_up = false;
+
+	/* TODO: cancel connection_loss_work */
+}
+
+static u32 ath11k_mac_get_rate_hw_value(int bitrate)
+{
+	u32 preamble;
+	u16 hw_value;
+	int rate;
+	size_t i;
+
+	if (ath11k_mac_bitrate_is_cck(bitrate))
+		preamble = WMI_RATE_PREAMBLE_CCK;
+	else
+		preamble = WMI_RATE_PREAMBLE_OFDM;
+
+	for (i = 0; i < ARRAY_SIZE(ath11k_legacy_rates); i++) {
+		if (ath11k_legacy_rates[i].bitrate != bitrate)
+			continue;
+
+		hw_value = ath11k_legacy_rates[i].hw_value;
+		rate = ATH11K_HW_RATE_CODE(hw_value, 0, preamble);
+
+		return rate;
+	}
+
+	return -EINVAL;
+}
+
+static void ath11k_recalculate_mgmt_rate(struct ath11k *ar,
+					 struct ieee80211_vif *vif,
+					 struct cfg80211_chan_def *def)
+{
+	struct ath11k_vif *arvif = (void *)vif->drv_priv;
+	const struct ieee80211_supported_band *sband;
+	u8 basic_rate_idx;
+	int hw_rate_code;
+	u32 vdev_param;
+	u16 bitrate;
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	sband = ar->hw->wiphy->bands[def->chan->band];
+	basic_rate_idx = ffs(vif->bss_conf.basic_rates) - 1;
+	bitrate = sband->bitrates[basic_rate_idx].bitrate;
+
+	hw_rate_code = ath11k_mac_get_rate_hw_value(bitrate);
+	if (hw_rate_code < 0) {
+		ath11k_warn(ar->ab, "bitrate not supported %d\n", bitrate);
+		return;
+	}
+
+	vdev_param = WMI_VDEV_PARAM_MGMT_RATE;
+	ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id, vdev_param,
+					    hw_rate_code);
+	if (ret)
+		ath11k_warn(ar->ab, "failed to set mgmt tx rate %d\n", ret);
+
+	vdev_param = WMI_VDEV_PARAM_BEACON_RATE;
+	ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id, vdev_param,
+					    hw_rate_code);
+	if (ret)
+		ath11k_warn(ar->ab, "failed to set beacon tx rate %d\n", ret);
+}
+
+static void ath11k_mac_op_bss_info_changed(struct ieee80211_hw *hw,
+					   struct ieee80211_vif *vif,
+					   struct ieee80211_bss_conf *info,
+					   u32 changed)
+{
+	struct ath11k *ar = hw->priv;
+	struct ath11k_vif *arvif = ath11k_vif_to_arvif(vif);
+	struct cfg80211_chan_def def;
+	u32 param_id, param_value;
+	enum nl80211_band band;
+	u32 vdev_param;
+	int mcast_rate;
+	u32 preamble;
+	u16 hw_value;
+	u16 bitrate;
+	int ret = 0;
+	u8 rateidx;
+	u32 rate;
+
+	mutex_lock(&ar->conf_mutex);
+
+	if (changed & BSS_CHANGED_BEACON_INT) {
+		arvif->beacon_interval = info->beacon_int;
+
+		param_id = WMI_VDEV_PARAM_BEACON_INTERVAL;
+		ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id,
+						    param_id,
+						    arvif->beacon_interval);
+		if (ret)
+			ath11k_warn(ar->ab, "Failed to set beacon interval for VDEV: %d\n",
+				    arvif->vdev_id);
+		else
+			ath11k_dbg(ar->ab, ATH11K_DBG_MAC,
+				   "Beacon interval: %d set for VDEV: %d\n",
+				   arvif->beacon_interval, arvif->vdev_id);
+	}
+
+	if (changed & BSS_CHANGED_BEACON) {
+		param_id = WMI_PDEV_PARAM_BEACON_TX_MODE;
+		param_value = WMI_BEACON_STAGGERED_MODE;
+		ret = ath11k_wmi_pdev_set_param(ar, param_id,
+						param_value, ar->pdev->pdev_id);
+		if (ret)
+			ath11k_warn(ar->ab, "Failed to set beacon mode for VDEV: %d\n",
+				    arvif->vdev_id);
+		else
+			ath11k_dbg(ar->ab, ATH11K_DBG_MAC,
+				   "Set staggered beacon mode for VDEV: %d\n",
+				   arvif->vdev_id);
+
+		ret = ath11k_mac_setup_bcn_tmpl(arvif);
+		if (ret)
+			ath11k_warn(ar->ab, "failed to update bcn template: %d\n",
+				    ret);
+	}
+
+	if (changed & (BSS_CHANGED_BEACON_INFO | BSS_CHANGED_BEACON)) {
+		arvif->dtim_period = info->dtim_period;
+
+		param_id = WMI_VDEV_PARAM_DTIM_PERIOD;
+		ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id,
+						    param_id,
+						    arvif->dtim_period);
+
+		if (ret)
+			ath11k_warn(ar->ab, "Failed to set dtim period for VDEV %d: %i\n",
+				    arvif->vdev_id, ret);
+		else
+			ath11k_dbg(ar->ab, ATH11K_DBG_MAC,
+				   "DTIM period: %d set for VDEV: %d\n",
+				   arvif->dtim_period, arvif->vdev_id);
+	}
+
+	if (changed & BSS_CHANGED_SSID &&
+	    vif->type == NL80211_IFTYPE_AP) {
+		arvif->u.ap.ssid_len = info->ssid_len;
+		if (info->ssid_len)
+			memcpy(arvif->u.ap.ssid, info->ssid, info->ssid_len);
+		arvif->u.ap.hidden_ssid = info->hidden_ssid;
+	}
+
+	if (changed & BSS_CHANGED_BSSID && !is_zero_ether_addr(info->bssid))
+		ether_addr_copy(arvif->bssid, info->bssid);
+
+	if (changed & BSS_CHANGED_BEACON_ENABLED)
+		ath11k_control_beaconing(arvif, info);
+
+	if (changed & BSS_CHANGED_ERP_CTS_PROT) {
+		u32 cts_prot;
+
+		cts_prot = !!(info->use_cts_prot);
+		param_id = WMI_VDEV_PARAM_PROTECTION_MODE;
+
+		if (arvif->is_started) {
+			ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id,
+							    param_id, cts_prot);
+			if (ret)
+				ath11k_warn(ar->ab, "Failed to set CTS prot for VDEV: %d\n",
+					    arvif->vdev_id);
+			else
+				ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "Set CTS prot: %d for VDEV: %d\n",
+					   cts_prot, arvif->vdev_id);
+		} else {
+			ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "defer protection mode setup, vdev is not ready yet\n");
+		}
+	}
+
+	if (changed & BSS_CHANGED_ERP_SLOT) {
+		u32 slottime;
+
+		if (info->use_short_slot)
+			slottime = WMI_VDEV_SLOT_TIME_SHORT; /* 9us */
+
+		else
+			slottime = WMI_VDEV_SLOT_TIME_LONG; /* 20us */
+
+		param_id = WMI_VDEV_PARAM_SLOT_TIME;
+		ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id,
+						    param_id, slottime);
+		if (ret)
+			ath11k_warn(ar->ab, "Failed to set erp slot for VDEV: %d\n",
+				    arvif->vdev_id);
+		else
+			ath11k_dbg(ar->ab, ATH11K_DBG_MAC,
+				   "Set slottime: %d for VDEV: %d\n",
+				   slottime, arvif->vdev_id);
+	}
+
+	if (changed & BSS_CHANGED_ERP_PREAMBLE) {
+		u32 preamble;
+
+		if (info->use_short_preamble)
+			preamble = WMI_VDEV_PREAMBLE_SHORT;
+		else
+			preamble = WMI_VDEV_PREAMBLE_LONG;
+
+		param_id = WMI_VDEV_PARAM_PREAMBLE;
+		ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id,
+						    param_id, preamble);
+		if (ret)
+			ath11k_warn(ar->ab, "Failed to set preamble for VDEV: %d\n",
+				    arvif->vdev_id);
+		else
+			ath11k_dbg(ar->ab, ATH11K_DBG_MAC,
+				   "Set preamble: %d for VDEV: %d\n",
+				   preamble, arvif->vdev_id);
+	}
+
+	if (changed & BSS_CHANGED_ASSOC) {
+		if (info->assoc)
+			ath11k_bss_assoc(hw, vif, info);
+		else
+			ath11k_bss_disassoc(hw, vif);
+	}
+
+	if (changed & BSS_CHANGED_TXPOWER) {
+		ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "mac vdev_id %i txpower %d\n",
+			   arvif->vdev_id, info->txpower);
+
+		arvif->txpower = info->txpower;
+		ath11k_mac_txpower_recalc(ar);
+	}
+
+	if (changed & BSS_CHANGED_MCAST_RATE &&
+	    !ath11k_mac_vif_chan(arvif->vif, &def)) {
+		band = def.chan->band;
+		mcast_rate = vif->bss_conf.mcast_rate[band];
+
+		if (mcast_rate > 0)
+			rateidx = mcast_rate - 1;
+		else
+			rateidx = ffs(vif->bss_conf.basic_rates) - 1;
+
+		if (ar->pdev->cap.supported_bands & WMI_HOST_WLAN_5G_CAP)
+			rateidx += ATH11K_MAC_FIRST_OFDM_RATE_IDX;
+
+		bitrate = ath11k_legacy_rates[rateidx].bitrate;
+		hw_value = ath11k_legacy_rates[rateidx].hw_value;
+
+		if (ath11k_mac_bitrate_is_cck(bitrate))
+			preamble = WMI_RATE_PREAMBLE_CCK;
+		else
+			preamble = WMI_RATE_PREAMBLE_OFDM;
+
+		rate = ATH11K_HW_RATE_CODE(hw_value, 0, preamble);
+
+		ath11k_dbg(ar->ab, ATH11K_DBG_MAC,
+			   "mac vdev %d mcast_rate %x\n",
+			   arvif->vdev_id, rate);
+
+		vdev_param = WMI_VDEV_PARAM_MCAST_DATA_RATE;
+		ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id,
+						    vdev_param, rate);
+		if (ret)
+			ath11k_warn(ar->ab,
+				    "failed to set mcast rate on vdev %i: %d\n",
+				    arvif->vdev_id,  ret);
+
+		vdev_param = WMI_VDEV_PARAM_BCAST_DATA_RATE;
+		ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id,
+						    vdev_param, rate);
+		if (ret)
+			ath11k_warn(ar->ab,
+				    "failed to set bcast rate on vdev %i: %d\n",
+				    arvif->vdev_id,  ret);
+	}
+
+	if (changed & BSS_CHANGED_BASIC_RATES &&
+	    !ath11k_mac_vif_chan(arvif->vif, &def))
+		ath11k_recalculate_mgmt_rate(ar, vif, &def);
+
+	mutex_unlock(&ar->conf_mutex);
+}
+
+void __ath11k_mac_scan_finish(struct ath11k *ar)
+{
+	lockdep_assert_held(&ar->data_lock);
+
+	switch (ar->scan.state) {
+	case ATH11K_SCAN_IDLE:
+		break;
+	case ATH11K_SCAN_RUNNING:
+	case ATH11K_SCAN_ABORTING:
+		if (!ar->scan.is_roc) {
+			struct cfg80211_scan_info info = {
+				.aborted = (ar->scan.state ==
+					    ATH11K_SCAN_ABORTING),
+			};
+
+			ieee80211_scan_completed(ar->hw, &info);
+		} else if (ar->scan.roc_notify) {
+			ieee80211_remain_on_channel_expired(ar->hw);
+		}
+		/* fall through */
+	case ATH11K_SCAN_STARTING:
+		ar->scan.state = ATH11K_SCAN_IDLE;
+		ar->scan_channel = NULL;
+		ar->scan.roc_freq = 0;
+		cancel_delayed_work(&ar->scan.timeout);
+		complete(&ar->scan.completed);
+		break;
+	}
+}
+
+void ath11k_mac_scan_finish(struct ath11k *ar)
+{
+	spin_lock_bh(&ar->data_lock);
+	__ath11k_mac_scan_finish(ar);
+	spin_unlock_bh(&ar->data_lock);
+}
+
+static int ath11k_scan_stop(struct ath11k *ar)
+{
+	struct scan_cancel_param arg = {
+		.req_type = WLAN_SCAN_CANCEL_SINGLE,
+		.scan_id = ATH11K_SCAN_ID,
+	};
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	/* TODO: Fill other STOP Params */
+	arg.pdev_id = ar->pdev->pdev_id;
+
+	ret = ath11k_wmi_send_scan_stop_cmd(ar, &arg);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to stop wmi scan: %d\n", ret);
+		goto out;
+	}
+
+	ret = wait_for_completion_timeout(&ar->scan.completed, 3 * HZ);
+	if (ret == 0) {
+		ath11k_warn(ar->ab,
+			    "failed to receive scan abort comple: timed out\n");
+		ret = -ETIMEDOUT;
+	} else if (ret > 0) {
+		ret = 0;
+	}
+
+out:
+	/* Scan state should be updated upon scan completion but in case
+	 * firmware fails to deliver the event (for whatever reason) it is
+	 * desired to clean up scan state anyway. Firmware may have just
+	 * dropped the scan completion event delivery due to transport pipe
+	 * being overflown with data and/or it can recover on its own before
+	 * next scan request is submitted.
+	 */
+	spin_lock_bh(&ar->data_lock);
+	if (ar->scan.state != ATH11K_SCAN_IDLE)
+		__ath11k_mac_scan_finish(ar);
+	spin_unlock_bh(&ar->data_lock);
+
+	return ret;
+}
+
+static void ath11k_scan_abort(struct ath11k *ar)
+{
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	spin_lock_bh(&ar->data_lock);
+
+	switch (ar->scan.state) {
+	case ATH11K_SCAN_IDLE:
+		/* This can happen if timeout worker kicked in and called
+		 * abortion while scan completion was being processed.
+		 */
+		break;
+	case ATH11K_SCAN_STARTING:
+	case ATH11K_SCAN_ABORTING:
+		ath11k_warn(ar->ab, "refusing scan abortion due to invalid scan state: %d\n",
+			    ar->scan.state);
+		break;
+	case ATH11K_SCAN_RUNNING:
+		ar->scan.state = ATH11K_SCAN_ABORTING;
+		spin_unlock_bh(&ar->data_lock);
+
+		ret = ath11k_scan_stop(ar);
+		if (ret)
+			ath11k_warn(ar->ab, "failed to abort scan: %d\n", ret);
+
+		spin_lock_bh(&ar->data_lock);
+		break;
+	}
+
+	spin_unlock_bh(&ar->data_lock);
+}
+
+static void ath11k_scan_timeout_work(struct work_struct *work)
+{
+	struct ath11k *ar = container_of(work, struct ath11k,
+					 scan.timeout.work);
+
+	mutex_lock(&ar->conf_mutex);
+	ath11k_scan_abort(ar);
+	mutex_unlock(&ar->conf_mutex);
+}
+
+static int ath11k_start_scan(struct ath11k *ar,
+			     struct scan_req_params *arg)
+{
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	ret = ath11k_wmi_send_scan_start_cmd(ar, arg);
+	if (ret)
+		return ret;
+
+	ret = wait_for_completion_timeout(&ar->scan.started, 1 * HZ);
+	if (ret == 0) {
+		ret = ath11k_scan_stop(ar);
+		if (ret)
+			ath11k_warn(ar->ab, "failed to stop scan: %d\n", ret);
+
+		return -ETIMEDOUT;
+	}
+
+	/* If we failed to start the scan, return error code at
+	 * this point.  This is probably due to some issue in the
+	 * firmware, but no need to wedge the driver due to that...
+	 */
+	spin_lock_bh(&ar->data_lock);
+	if (ar->scan.state == ATH11K_SCAN_IDLE) {
+		spin_unlock_bh(&ar->data_lock);
+		return -EINVAL;
+	}
+	spin_unlock_bh(&ar->data_lock);
+
+	return 0;
+}
+
+static int ath11k_mac_op_hw_scan(struct ieee80211_hw *hw,
+				 struct ieee80211_vif *vif,
+				 struct ieee80211_scan_request *hw_req)
+{
+	struct ath11k *ar = hw->priv;
+	struct ath11k_vif *arvif = ath11k_vif_to_arvif(vif);
+	struct cfg80211_scan_request *req = &hw_req->req;
+	struct scan_req_params arg;
+	int ret = 0;
+	int i;
+
+	mutex_lock(&ar->conf_mutex);
+
+	spin_lock_bh(&ar->data_lock);
+	switch (ar->scan.state) {
+	case ATH11K_SCAN_IDLE:
+		reinit_completion(&ar->scan.started);
+		reinit_completion(&ar->scan.completed);
+		ar->scan.state = ATH11K_SCAN_STARTING;
+		ar->scan.is_roc = false;
+		ar->scan.vdev_id = arvif->vdev_id;
+		ret = 0;
+		break;
+	case ATH11K_SCAN_STARTING:
+	case ATH11K_SCAN_RUNNING:
+	case ATH11K_SCAN_ABORTING:
+		ret = -EBUSY;
+		break;
+	}
+	spin_unlock_bh(&ar->data_lock);
+
+	if (ret)
+		goto exit;
+
+	memset(&arg, 0, sizeof(arg));
+	ath11k_wmi_start_scan_init(ar, &arg);
+	arg.vdev_id = arvif->vdev_id;
+	arg.scan_id = ATH11K_SCAN_ID;
+
+	if (req->ie_len) {
+		arg.extraie.len = req->ie_len;
+		arg.extraie.ptr = kzalloc(req->ie_len, GFP_KERNEL);
+		memcpy(arg.extraie.ptr, req->ie, req->ie_len);
+	}
+
+	if (req->n_ssids) {
+		arg.num_ssids = req->n_ssids;
+		for (i = 0; i < arg.num_ssids; i++) {
+			arg.ssid[i].length  = req->ssids[i].ssid_len;
+			memcpy(&arg.ssid[i].ssid, req->ssids[i].ssid,
+			       req->ssids[i].ssid_len);
+		}
+	} else {
+		arg.scan_flags |= WMI_SCAN_FLAG_PASSIVE;
+	}
+
+	if (req->n_channels) {
+		arg.num_chan = req->n_channels;
+		for (i = 0; i < arg.num_chan; i++)
+			arg.chan_list[i] = req->channels[i]->center_freq;
+	}
+
+	ret = ath11k_start_scan(ar, &arg);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to start hw scan: %d\n", ret);
+		spin_lock_bh(&ar->data_lock);
+		ar->scan.state = ATH11K_SCAN_IDLE;
+		spin_unlock_bh(&ar->data_lock);
+	}
+
+	/* Add a 200ms margin to account for event/command processing */
+	ieee80211_queue_delayed_work(ar->hw, &ar->scan.timeout,
+				     msecs_to_jiffies(arg.max_scan_time +
+						      ATH11K_MAC_SCAN_TIMEOUT_MSECS));
+
+exit:
+	if (req->ie_len)
+		kfree(arg.extraie.ptr);
+
+	mutex_unlock(&ar->conf_mutex);
+	return ret;
+}
+
+static void ath11k_mac_op_cancel_hw_scan(struct ieee80211_hw *hw,
+					 struct ieee80211_vif *vif)
+{
+	struct ath11k *ar = hw->priv;
+
+	mutex_lock(&ar->conf_mutex);
+	ath11k_scan_abort(ar);
+	mutex_unlock(&ar->conf_mutex);
+
+	cancel_delayed_work_sync(&ar->scan.timeout);
+}
+
+static int ath11k_install_key(struct ath11k_vif *arvif,
+			      struct ieee80211_key_conf *key,
+			      enum set_key_cmd cmd,
+			      const u8 *macaddr, u32 flags)
+{
+	int ret;
+	struct ath11k *ar = arvif->ar;
+	struct wmi_vdev_install_key_arg arg = {
+		.vdev_id = arvif->vdev_id,
+		.key_idx = key->keyidx,
+		.key_len = key->keylen,
+		.key_data = key->key,
+		.key_flags = flags,
+		.macaddr = macaddr,
+	};
+
+	lockdep_assert_held(&arvif->ar->conf_mutex);
+
+	reinit_completion(&ar->install_key_done);
+
+	if (cmd == DISABLE_KEY) {
+		/* TODO: Check if FW expects  value other than NONE for del */
+		/* arg.key_cipher = WMI_CIPHER_NONE; */
+		arg.key_len = 0;
+		arg.key_data = NULL;
+		goto install;
+	}
+
+	switch (key->cipher) {
+	case WLAN_CIPHER_SUITE_CCMP:
+		arg.key_cipher = WMI_CIPHER_AES_CCM;
+		/* TODO: Re-check if flag is valid */
+		key->flags |= IEEE80211_KEY_FLAG_GENERATE_IV_MGMT;
+		break;
+	case WLAN_CIPHER_SUITE_TKIP:
+		arg.key_cipher = WMI_CIPHER_TKIP;
+		arg.key_txmic_len = 8;
+		arg.key_rxmic_len = 8;
+		break;
+	case WLAN_CIPHER_SUITE_CCMP_256:
+		arg.key_cipher = WMI_CIPHER_AES_CCM;
+		break;
+	case WLAN_CIPHER_SUITE_GCMP:
+	case WLAN_CIPHER_SUITE_GCMP_256:
+		arg.key_cipher = WMI_CIPHER_AES_GCM;
+		break;
+	default:
+		ath11k_warn(ar->ab, "cipher %d is not supported\n", key->cipher);
+		return -EOPNOTSUPP;
+	}
+
+install:
+	ret = ath11k_wmi_vdev_install_key(arvif->ar, &arg);
+	if (ret)
+		return ret;
+
+	if (!wait_for_completion_timeout(&ar->install_key_done, 1 * HZ))
+		return -ETIMEDOUT;
+
+	return ar->install_key_status ? -EINVAL : 0;
+}
+
+static int ath11k_clear_peer_keys(struct ath11k_vif *arvif,
+				  const u8 *addr)
+{
+	struct ath11k *ar = arvif->ar;
+	struct ath11k_base *ab = ar->ab;
+	struct ath11k_peer *peer;
+	int first_errno = 0;
+	int ret;
+	int i;
+	u32 flags = 0;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	spin_lock_bh(&ab->base_lock);
+	peer = ath11k_peer_find(ab, arvif->vdev_id, addr);
+	spin_unlock_bh(&ab->base_lock);
+
+	if (!peer)
+		return -ENOENT;
+
+	for (i = 0; i < ARRAY_SIZE(peer->keys); i++) {
+		if (!peer->keys[i])
+			continue;
+
+		/* key flags are not required to delete the key */
+		ret = ath11k_install_key(arvif, peer->keys[i],
+					 DISABLE_KEY, addr, flags);
+		if (ret < 0 && first_errno == 0)
+			first_errno = ret;
+
+		if (ret < 0)
+			ath11k_warn(ab, "failed to remove peer key %d: %d\n",
+				    i, ret);
+
+		spin_lock_bh(&ab->base_lock);
+		peer->keys[i] = NULL;
+		spin_unlock_bh(&ab->base_lock);
+	}
+
+	return first_errno;
+}
+
+static int ath11k_mac_op_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
+				 struct ieee80211_vif *vif, struct ieee80211_sta *sta,
+				 struct ieee80211_key_conf *key)
+{
+	struct ath11k *ar = hw->priv;
+	struct ath11k_base *ab = ar->ab;
+	struct ath11k_vif *arvif = ath11k_vif_to_arvif(vif);
+	struct ath11k_peer *peer;
+	const u8 *peer_addr;
+	int ret = 0;
+	u32 flags = 0;
+
+	/* BIP needs to be done in software */
+	if (key->cipher == WLAN_CIPHER_SUITE_AES_CMAC ||
+	    key->cipher == WLAN_CIPHER_SUITE_BIP_GMAC_128 ||
+	    key->cipher == WLAN_CIPHER_SUITE_BIP_GMAC_256 ||
+	    key->cipher == WLAN_CIPHER_SUITE_BIP_CMAC_256)
+		return 1;
+
+	if (key->keyidx > WMI_MAX_KEY_INDEX)
+		return -ENOSPC;
+
+	mutex_lock(&ar->conf_mutex);
+
+	if (sta)
+		peer_addr = sta->addr;
+	else if (arvif->vdev_type == WMI_VDEV_TYPE_STA)
+		peer_addr = vif->bss_conf.bssid;
+	else
+		peer_addr = vif->addr;
+
+	key->hw_key_idx = key->keyidx;
+
+	/* the peer should not disappear in mid-way (unless FW goes awry) since
+	 * we already hold conf_mutex. we just make sure its there now.
+	 */
+	spin_lock_bh(&ab->base_lock);
+	peer = ath11k_peer_find(ab, arvif->vdev_id, peer_addr);
+	spin_unlock_bh(&ab->base_lock);
+
+	if (!peer) {
+		if (cmd == SET_KEY) {
+			ath11k_warn(ab, "cannot install key for non-existent peer %pM\n",
+				    peer_addr);
+			ret = -EOPNOTSUPP;
+			goto exit;
+		} else {
+			/* if the peer doesn't exist there is no key to disable
+			 * anymore
+			 */
+			goto exit;
+		}
+	}
+
+	if (key->flags & IEEE80211_KEY_FLAG_PAIRWISE)
+		flags |= WMI_KEY_PAIRWISE;
+	else
+		flags |= WMI_KEY_GROUP;
+
+	ret = ath11k_install_key(arvif, key, cmd, peer_addr, flags);
+	if (ret) {
+		ath11k_warn(ab, "ath11k_install_key failed (%d)\n", ret);
+		goto exit;
+	}
+
+	spin_lock_bh(&ab->base_lock);
+	peer = ath11k_peer_find(ab, arvif->vdev_id, peer_addr);
+	if (peer && cmd == SET_KEY)
+		peer->keys[key->keyidx] = key;
+	else if (peer && cmd == DISABLE_KEY)
+		peer->keys[key->keyidx] = NULL;
+	else if (!peer)
+		/* impossible unless FW goes crazy */
+		ath11k_warn(ab, "peer %pM disappeared!\n", peer_addr);
+	spin_unlock_bh(&ab->base_lock);
+
+exit:
+	mutex_unlock(&ar->conf_mutex);
+	return ret;
+}
+
+static int
+ath11k_mac_bitrate_mask_num_vht_rates(struct ath11k *ar,
+				      enum nl80211_band band,
+				      const struct cfg80211_bitrate_mask *mask)
+{
+	int num_rates = 0;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(mask->control[band].vht_mcs); i++)
+		num_rates += hweight16(mask->control[band].vht_mcs[i]);
+
+	return num_rates;
+}
+
+static int
+ath11k_mac_set_peer_vht_fixed_rate(struct ath11k_vif *arvif,
+				   struct ieee80211_sta *sta,
+				   const struct cfg80211_bitrate_mask *mask,
+				   enum nl80211_band band)
+{
+	struct ath11k *ar = arvif->ar;
+	u8 vht_rate, nss;
+	u32 rate_code;
+	int ret, i;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	nss = 0;
+
+	for (i = 0; i < ARRAY_SIZE(mask->control[band].vht_mcs); i++) {
+		if (hweight16(mask->control[band].vht_mcs[i]) == 1) {
+			nss = i + 1;
+			vht_rate = ffs(mask->control[band].vht_mcs[i]) - 1;
+		}
+	}
+
+	if (!nss) {
+		ath11k_warn(ar->ab, "No single VHT Fixed rate found to set for %pM",
+			    sta->addr);
+		return -EINVAL;
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_MAC,
+		   "Setting Fixed VHT Rate for peer %pM. Device will not switch to any other selected rates",
+		   sta->addr);
+
+	rate_code = ATH11K_HW_RATE_CODE(vht_rate, nss - 1,
+					WMI_RATE_PREAMBLE_VHT);
+	ret = ath11k_wmi_set_peer_param(ar, sta->addr,
+					arvif->vdev_id,
+					WMI_PEER_PARAM_FIXED_RATE,
+					rate_code);
+	if (ret)
+		ath11k_warn(ar->ab,
+			    "failed to update STA %pM Fixed Rate %d: %d\n",
+			     sta->addr, rate_code, ret);
+
+	return ret;
+}
+
+static int ath11k_station_assoc(struct ath11k *ar,
+				struct ieee80211_vif *vif,
+				struct ieee80211_sta *sta,
+				bool reassoc)
+{
+	struct ath11k_vif *arvif = ath11k_vif_to_arvif(vif);
+	struct peer_assoc_params peer_arg;
+	int ret = 0;
+	struct cfg80211_chan_def def;
+	enum nl80211_band band;
+	struct cfg80211_bitrate_mask *mask;
+	u8 num_vht_rates;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	if (WARN_ON(ath11k_mac_vif_chan(vif, &def)))
+		return -EPERM;
+
+	band = def.chan->band;
+	mask = &arvif->bitrate_mask;
+
+	ath11k_peer_assoc_prepare(ar, vif, sta, &peer_arg, reassoc);
+
+	ret = ath11k_wmi_send_peer_assoc_cmd(ar, &peer_arg);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to run peer assoc for STA %pM vdev %i: %d\n",
+			    sta->addr, arvif->vdev_id, ret);
+		return ret;
+	}
+
+	if (!wait_for_completion_timeout(&ar->peer_assoc_done, 1 * HZ)) {
+		ath11k_warn(ar->ab, "failed to get peer assoc conf event for %pM vdev %i\n",
+			    sta->addr, arvif->vdev_id);
+		return -ETIMEDOUT;
+	}
+
+	num_vht_rates = ath11k_mac_bitrate_mask_num_vht_rates(ar, band, mask);
+
+	/* If single VHT rate is configured (by set_bitrate_mask()),
+	 * peer_assoc will disable VHT. This is now enabled by a peer specific
+	 * fixed param.
+	 * Note that all other rates and NSS will be disabled for this peer.
+	 */
+	if (sta->vht_cap.vht_supported && num_vht_rates == 1) {
+		ret = ath11k_mac_set_peer_vht_fixed_rate(arvif, sta, mask,
+							 band);
+		if (ret)
+			return ret;
+	}
+
+	/* Re-assoc is run only to update supported rates for given station. It
+	 * doesn't make much sense to reconfigure the peer completely.
+	 */
+	if (reassoc)
+		return 0;
+
+	ret = ath11k_setup_peer_smps(ar, arvif, sta->addr,
+				     &sta->ht_cap);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to setup peer SMPS for vdev %d: %d\n",
+			    arvif->vdev_id, ret);
+		return ret;
+	}
+
+	if (!sta->wme) {
+		arvif->num_legacy_stations++;
+		ret = ath11k_recalc_rtscts_prot(arvif);
+		if (ret)
+			return ret;
+	}
+
+	if (sta->wme && sta->uapsd_queues) {
+		ret = ath11k_peer_assoc_qos_ap(ar, arvif, sta);
+		if (ret) {
+			ath11k_warn(ar->ab, "failed to set qos params for STA %pM for vdev %i: %d\n",
+				    sta->addr, arvif->vdev_id, ret);
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
+static int ath11k_station_disassoc(struct ath11k *ar,
+				   struct ieee80211_vif *vif,
+				   struct ieee80211_sta *sta)
+{
+	struct ath11k_vif *arvif = (void *)vif->drv_priv;
+	int ret = 0;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	if (!sta->wme) {
+		arvif->num_legacy_stations--;
+		ret = ath11k_recalc_rtscts_prot(arvif);
+		if (ret)
+			return ret;
+	}
+
+	ret = ath11k_clear_peer_keys(arvif, sta->addr);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to clear all peer keys for vdev %i: %d\n",
+			    arvif->vdev_id, ret);
+		return ret;
+	}
+	return 0;
+}
+
+static void ath11k_sta_rc_update_wk(struct work_struct *wk)
+{
+	struct ath11k *ar;
+	struct ath11k_vif *arvif;
+	struct ath11k_sta *arsta;
+	struct ieee80211_sta *sta;
+	struct cfg80211_chan_def def;
+	enum nl80211_band band;
+	const u8 *ht_mcs_mask;
+	const u16 *vht_mcs_mask;
+	u32 changed, bw, nss, smps;
+	int err, num_vht_rates;
+	const struct cfg80211_bitrate_mask *mask;
+	struct peer_assoc_params peer_arg;
+
+	arsta = container_of(wk, struct ath11k_sta, update_wk);
+	sta = container_of((void *)arsta, struct ieee80211_sta, drv_priv);
+	arvif = arsta->arvif;
+	ar = arvif->ar;
+
+	if (WARN_ON(ath11k_mac_vif_chan(arvif->vif, &def)))
+		return;
+
+	band = def.chan->band;
+	ht_mcs_mask = arvif->bitrate_mask.control[band].ht_mcs;
+	vht_mcs_mask = arvif->bitrate_mask.control[band].vht_mcs;
+
+	spin_lock_bh(&ar->data_lock);
+
+	changed = arsta->changed;
+	arsta->changed = 0;
+
+	bw = arsta->bw;
+	nss = arsta->nss;
+	smps = arsta->smps;
+
+	spin_unlock_bh(&ar->data_lock);
+
+	mutex_lock(&ar->conf_mutex);
+
+	nss = max_t(u32, 1, nss);
+	nss = min(nss, max(ath11k_mac_max_ht_nss(ht_mcs_mask),
+			   ath11k_mac_max_vht_nss(vht_mcs_mask)));
+
+	if (changed & IEEE80211_RC_BW_CHANGED) {
+		err = ath11k_wmi_set_peer_param(ar, sta->addr, arvif->vdev_id,
+						WMI_PEER_CHWIDTH, bw);
+		if (err)
+			ath11k_warn(ar->ab, "failed to update STA %pM peer bw %d: %d\n",
+				    sta->addr, bw, err);
+	}
+
+	if (changed & IEEE80211_RC_NSS_CHANGED) {
+		ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "mac update sta %pM nss %d\n",
+			   sta->addr, nss);
+
+		err = ath11k_wmi_set_peer_param(ar, sta->addr, arvif->vdev_id,
+						WMI_PEER_NSS, nss);
+		if (err)
+			ath11k_warn(ar->ab, "failed to update STA %pM nss %d: %d\n",
+				    sta->addr, nss, err);
+	}
+
+	if (changed & IEEE80211_RC_SMPS_CHANGED) {
+		ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "mac update sta %pM smps %d\n",
+			   sta->addr, smps);
+
+		err = ath11k_wmi_set_peer_param(ar, sta->addr, arvif->vdev_id,
+						WMI_PEER_MIMO_PS_STATE, smps);
+		if (err)
+			ath11k_warn(ar->ab, "failed to update STA %pM smps %d: %d\n",
+				    sta->addr, smps, err);
+	}
+
+	if (changed & IEEE80211_RC_SUPP_RATES_CHANGED) {
+		mask = &arvif->bitrate_mask;
+		num_vht_rates = ath11k_mac_bitrate_mask_num_vht_rates(ar, band,
+								      mask);
+
+		/* Peer_assoc_prepare will reject vht rates in
+		 * bitrate_mask if its not available in range format and
+		 * sets vht tx_rateset as unsupported. So multiple VHT MCS
+		 * setting(eg. MCS 4,5,6) per peer is not supported here.
+		 * But, Single rate in VHT mask can be set as per-peer
+		 * fixed rate. But even if any HT rates are configured in
+		 * the bitrate mask, device will not switch to those rates
+		 * when per-peer Fixed rate is set.
+		 * TODO: Check RATEMASK_CMDID to support auto rates selection
+		 * across HT/VHT and for multiple VHT MCS support.
+		 */
+		if (sta->vht_cap.vht_supported && num_vht_rates == 1) {
+			ath11k_mac_set_peer_vht_fixed_rate(arvif, sta, mask,
+							   band);
+		} else {
+			/* If the peer is non-VHT or no fixed VHT rate
+			 * is provided in the new bitrate mask we set the
+			 * other rates using peer_assoc command.
+			 */
+			ath11k_peer_assoc_prepare(ar, arvif->vif, sta,
+						  &peer_arg, true);
+
+			err = ath11k_wmi_send_peer_assoc_cmd(ar, &peer_arg);
+			if (err)
+				ath11k_warn(ar->ab, "failed to run peer assoc for STA %pM vdev %i: %d\n",
+					    sta->addr, arvif->vdev_id, err);
+
+			if (!wait_for_completion_timeout(&ar->peer_assoc_done, 1 * HZ))
+				ath11k_warn(ar->ab, "failed to get peer assoc conf event for %pM vdev %i\n",
+					    sta->addr, arvif->vdev_id);
+		}
+	}
+
+	mutex_unlock(&ar->conf_mutex);
+}
+
+static int ath11k_mac_inc_num_stations(struct ath11k_vif *arvif,
+				       struct ieee80211_sta *sta)
+{
+	struct ath11k *ar = arvif->ar;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	if (arvif->vdev_type == WMI_VDEV_TYPE_STA && !sta->tdls)
+		return 0;
+
+	if (ar->num_stations >= ar->max_num_stations)
+		return -ENOBUFS;
+
+	ar->num_stations++;
+
+	return 0;
+}
+
+static void ath11k_mac_dec_num_stations(struct ath11k_vif *arvif,
+					struct ieee80211_sta *sta)
+{
+	struct ath11k *ar = arvif->ar;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	if (arvif->vdev_type == WMI_VDEV_TYPE_STA && !sta->tdls)
+		return;
+
+	ar->num_stations--;
+}
+
+static int ath11k_mac_op_sta_state(struct ieee80211_hw *hw,
+				   struct ieee80211_vif *vif,
+				   struct ieee80211_sta *sta,
+				   enum ieee80211_sta_state old_state,
+				   enum ieee80211_sta_state new_state)
+{
+	struct ath11k *ar = hw->priv;
+	struct ath11k_vif *arvif = ath11k_vif_to_arvif(vif);
+	struct ath11k_sta *arsta = (struct ath11k_sta *)sta->drv_priv;
+	struct peer_create_params peer_param;
+	int ret = 0;
+
+	/* cancel must be done outside the mutex to avoid deadlock */
+	if ((old_state == IEEE80211_STA_NONE &&
+	     new_state == IEEE80211_STA_NOTEXIST))
+		cancel_work_sync(&arsta->update_wk);
+
+	mutex_lock(&ar->conf_mutex);
+
+	if (old_state == IEEE80211_STA_NOTEXIST &&
+	    new_state == IEEE80211_STA_NONE) {
+		memset(arsta, 0, sizeof(*arsta));
+		arsta->arvif = arvif;
+		INIT_WORK(&arsta->update_wk, ath11k_sta_rc_update_wk);
+
+		ret = ath11k_mac_inc_num_stations(arvif, sta);
+		if (ret) {
+			ath11k_warn(ar->ab, "refusing to associate station: too many connected already (%d)\n",
+				    ar->max_num_stations);
+			goto exit;
+		}
+
+		arsta->rx_stats = kzalloc(sizeof(*arsta->rx_stats), GFP_KERNEL);
+		if (!arsta->rx_stats) {
+			ret = -ENOMEM;
+			goto exit;
+		}
+
+		peer_param.vdev_id = arvif->vdev_id;
+		peer_param.peer_addr = sta->addr;
+		peer_param.peer_type = WMI_PEER_TYPE_DEFAULT;
+		ret = ath11k_peer_create(ar, arvif, sta, &peer_param);
+		if (ret) {
+			ath11k_warn(ar->ab, "Failed to add peer: %pM for VDEV: %d\n",
+				    sta->addr, arvif->vdev_id);
+			ath11k_mac_dec_num_stations(arvif, sta);
+			goto exit;
+		}
+
+		ath11k_info(ar->ab, "Added peer: %pM for VDEV: %d\n",
+			    sta->addr, arvif->vdev_id);
+
+		if (ath11k_debug_is_extd_tx_stats_enabled(ar)) {
+			arsta->tx_stats = kzalloc(sizeof(*arsta->tx_stats),
+						  GFP_KERNEL);
+			if (!arsta->tx_stats) {
+				ret = -ENOMEM;
+				goto exit;
+			}
+		}
+
+		if (ieee80211_vif_is_mesh(vif)) {
+			ret = ath11k_wmi_set_peer_param(ar, sta->addr,
+							arvif->vdev_id,
+							WMI_PEER_USE_4ADDR, 1);
+			if (ret) {
+				ath11k_warn(ar->ab, "failed to STA %pM 4addr capability: %d\n",
+					    sta->addr, ret);
+				goto exit;
+			}
+		}
+
+		ret = ath11k_dp_peer_setup(ar, arvif->vdev_id, sta->addr);
+		if (ret) {
+			ath11k_warn(ar->ab, "failed to setup dp for peer %pM on vdev %i (%d)\n",
+				    sta->addr, arvif->vdev_id, ret);
+			ath11k_peer_delete(ar, arvif->vdev_id, sta->addr);
+			ath11k_mac_dec_num_stations(arvif, sta);
+		}
+	} else if ((old_state == IEEE80211_STA_NONE &&
+		    new_state == IEEE80211_STA_NOTEXIST)) {
+		ath11k_dp_peer_cleanup(ar, arvif->vdev_id, sta->addr);
+
+		ret = ath11k_peer_delete(ar, arvif->vdev_id, sta->addr);
+		if (ret)
+			ath11k_warn(ar->ab, "Failed to delete peer: %pM for VDEV: %d\n",
+				    sta->addr, arvif->vdev_id);
+		else
+			ath11k_info(ar->ab,
+				    "Removed peer: %pM for VDEV: %d\n",
+				    sta->addr, arvif->vdev_id);
+
+		ath11k_mac_dec_num_stations(arvif, sta);
+
+		kfree(arsta->tx_stats);
+		arsta->tx_stats = NULL;
+
+		kfree(arsta->rx_stats);
+		arsta->rx_stats = NULL;
+	} else if (old_state == IEEE80211_STA_AUTH &&
+		   new_state == IEEE80211_STA_ASSOC &&
+		   (vif->type == NL80211_IFTYPE_AP ||
+		    vif->type == NL80211_IFTYPE_MESH_POINT ||
+		    vif->type == NL80211_IFTYPE_ADHOC)) {
+		ret = ath11k_station_assoc(ar, vif, sta, false);
+		if (ret)
+			ath11k_warn(ar->ab, "Failed to associate station: %pM\n",
+				    sta->addr);
+		else
+			ath11k_info(ar->ab,
+				    "Station %pM moved to assoc state\n",
+				    sta->addr);
+	} else if (old_state == IEEE80211_STA_ASSOC &&
+		   new_state == IEEE80211_STA_AUTH &&
+		   (vif->type == NL80211_IFTYPE_AP ||
+		    vif->type == NL80211_IFTYPE_MESH_POINT ||
+		    vif->type == NL80211_IFTYPE_ADHOC)) {
+		ret = ath11k_station_disassoc(ar, vif, sta);
+		if (ret)
+			ath11k_warn(ar->ab, "Failed to disassociate station: %pM\n",
+				    sta->addr);
+		else
+			ath11k_info(ar->ab,
+				    "Station %pM moved to disassociated state\n",
+				    sta->addr);
+	}
+
+exit:
+	mutex_unlock(&ar->conf_mutex);
+	return ret;
+}
+
+static void ath11k_mac_op_sta_rc_update(struct ieee80211_hw *hw,
+					struct ieee80211_vif *vif,
+					struct ieee80211_sta *sta,
+					u32 changed)
+{
+	struct ath11k *ar = hw->priv;
+	struct ath11k_sta *arsta = (struct ath11k_sta *)sta->drv_priv;
+	struct ath11k_vif *arvif = (void *)vif->drv_priv;
+	struct ath11k_peer *peer;
+	u32 bw, smps;
+
+	spin_lock_bh(&ar->ab->base_lock);
+
+	peer = ath11k_peer_find(ar->ab, arvif->vdev_id, sta->addr);
+	if (!peer) {
+		spin_unlock_bh(&ar->ab->base_lock);
+		ath11k_warn(ar->ab, "mac sta rc update failed to find peer %pM on vdev %i\n",
+			    sta->addr, arvif->vdev_id);
+		return;
+	}
+
+	spin_unlock_bh(&ar->ab->base_lock);
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_MAC,
+		   "mac sta rc update for %pM changed %08x bw %d nss %d smps %d\n",
+		   sta->addr, changed, sta->bandwidth, sta->rx_nss,
+		   sta->smps_mode);
+
+	spin_lock_bh(&ar->data_lock);
+
+	if (changed & IEEE80211_RC_BW_CHANGED) {
+		bw = WMI_PEER_CHWIDTH_20MHZ;
+
+		switch (sta->bandwidth) {
+		case IEEE80211_STA_RX_BW_20:
+			bw = WMI_PEER_CHWIDTH_20MHZ;
+			break;
+		case IEEE80211_STA_RX_BW_40:
+			bw = WMI_PEER_CHWIDTH_40MHZ;
+			break;
+		case IEEE80211_STA_RX_BW_80:
+			bw = WMI_PEER_CHWIDTH_80MHZ;
+			break;
+		case IEEE80211_STA_RX_BW_160:
+			bw = WMI_PEER_CHWIDTH_160MHZ;
+			break;
+		default:
+			ath11k_warn(ar->ab, "Invalid bandwidth %d in rc update for %pM\n",
+				    sta->bandwidth, sta->addr);
+			bw = WMI_PEER_CHWIDTH_20MHZ;
+			break;
+		}
+
+		arsta->bw = bw;
+	}
+
+	if (changed & IEEE80211_RC_NSS_CHANGED)
+		arsta->nss = sta->rx_nss;
+
+	if (changed & IEEE80211_RC_SMPS_CHANGED) {
+		smps = WMI_PEER_SMPS_PS_NONE;
+
+		switch (sta->smps_mode) {
+		case IEEE80211_SMPS_AUTOMATIC:
+		case IEEE80211_SMPS_OFF:
+			smps = WMI_PEER_SMPS_PS_NONE;
+			break;
+		case IEEE80211_SMPS_STATIC:
+			smps = WMI_PEER_SMPS_STATIC;
+			break;
+		case IEEE80211_SMPS_DYNAMIC:
+			smps = WMI_PEER_SMPS_DYNAMIC;
+			break;
+		default:
+			ath11k_warn(ar->ab, "Invalid smps %d in sta rc update for %pM\n",
+				    sta->smps_mode, sta->addr);
+			smps = WMI_PEER_SMPS_PS_NONE;
+			break;
+		}
+
+		arsta->smps = smps;
+	}
+
+	arsta->changed |= changed;
+
+	spin_unlock_bh(&ar->data_lock);
+
+	ieee80211_queue_work(hw, &arsta->update_wk);
+}
+
+static int ath11k_conf_tx_uapsd(struct ath11k *ar, struct ieee80211_vif *vif,
+				u16 ac, bool enable)
+{
+	struct ath11k_vif *arvif = ath11k_vif_to_arvif(vif);
+	u32 value = 0;
+	int ret = 0;
+
+	if (arvif->vdev_type != WMI_VDEV_TYPE_STA)
+		return 0;
+
+	switch (ac) {
+	case IEEE80211_AC_VO:
+		value = WMI_STA_PS_UAPSD_AC3_DELIVERY_EN |
+			WMI_STA_PS_UAPSD_AC3_TRIGGER_EN;
+		break;
+	case IEEE80211_AC_VI:
+		value = WMI_STA_PS_UAPSD_AC2_DELIVERY_EN |
+			WMI_STA_PS_UAPSD_AC2_TRIGGER_EN;
+		break;
+	case IEEE80211_AC_BE:
+		value = WMI_STA_PS_UAPSD_AC1_DELIVERY_EN |
+			WMI_STA_PS_UAPSD_AC1_TRIGGER_EN;
+		break;
+	case IEEE80211_AC_BK:
+		value = WMI_STA_PS_UAPSD_AC0_DELIVERY_EN |
+			WMI_STA_PS_UAPSD_AC0_TRIGGER_EN;
+		break;
+	}
+
+	if (enable)
+		arvif->u.sta.uapsd |= value;
+	else
+		arvif->u.sta.uapsd &= ~value;
+
+	ret = ath11k_wmi_set_sta_ps_param(ar, arvif->vdev_id,
+					  WMI_STA_PS_PARAM_UAPSD,
+					  arvif->u.sta.uapsd);
+	if (ret) {
+		ath11k_warn(ar->ab, "could not set uapsd params %d\n", ret);
+		goto exit;
+	}
+
+	if (arvif->u.sta.uapsd)
+		value = WMI_STA_PS_RX_WAKE_POLICY_POLL_UAPSD;
+	else
+		value = WMI_STA_PS_RX_WAKE_POLICY_WAKE;
+
+	ret = ath11k_wmi_set_sta_ps_param(ar, arvif->vdev_id,
+					  WMI_STA_PS_PARAM_RX_WAKE_POLICY,
+					  value);
+	if (ret)
+		ath11k_warn(ar->ab, "could not set rx wake param %d\n", ret);
+
+exit:
+	return ret;
+}
+
+static int ath11k_mac_op_conf_tx(struct ieee80211_hw *hw,
+				 struct ieee80211_vif *vif, u16 ac,
+				 const struct ieee80211_tx_queue_params *params)
+{
+	struct ath11k *ar = hw->priv;
+	struct ath11k_vif *arvif = (void *)vif->drv_priv;
+	struct wmi_wmm_params_arg *p = NULL;
+	int ret;
+
+	mutex_lock(&ar->conf_mutex);
+
+	switch (ac) {
+	case IEEE80211_AC_VO:
+		p = &arvif->wmm_params.ac_vo;
+		break;
+	case IEEE80211_AC_VI:
+		p = &arvif->wmm_params.ac_vi;
+		break;
+	case IEEE80211_AC_BE:
+		p = &arvif->wmm_params.ac_be;
+		break;
+	case IEEE80211_AC_BK:
+		p = &arvif->wmm_params.ac_bk;
+		break;
+	}
+
+	if (WARN_ON(!p)) {
+		ret = -EINVAL;
+		goto exit;
+	}
+
+	p->cwmin = params->cw_min;
+	p->cwmax = params->cw_max;
+	p->aifs = params->aifs;
+
+	/* The channel time duration programmed in the HW is in absolute
+	 * microseconds, while mac80211 gives the txop in units of
+	 * 32 microseconds.
+	 */
+	p->txop = params->txop * 32;
+
+	ret = ath11k_wmi_send_wmm_update_cmd_tlv(ar, arvif->vdev_id,
+						 &arvif->wmm_params);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to set wmm params: %d\n", ret);
+		goto exit;
+	}
+
+	ret = ath11k_conf_tx_uapsd(ar, vif, ac, params->uapsd);
+
+	if (ret)
+		ath11k_warn(ar->ab, "failed to set sta uapsd: %d\n", ret);
+
+exit:
+	mutex_unlock(&ar->conf_mutex);
+	return ret;
+}
+
+static struct ieee80211_sta_ht_cap
+ath11k_create_ht_cap(struct ath11k *ar, u32 ar_ht_cap, u32 rate_cap_rx_chainmask)
+{
+	int i;
+	struct ieee80211_sta_ht_cap ht_cap = {0};
+	u32 ar_vht_cap = ar->pdev->cap.vht_cap;
+
+	if (!(ar_ht_cap & WMI_HT_CAP_ENABLED))
+		return ht_cap;
+
+	ht_cap.ht_supported = 1;
+	ht_cap.ampdu_factor = IEEE80211_HT_MAX_AMPDU_64K;
+	ht_cap.ampdu_density = IEEE80211_HT_MPDU_DENSITY_8;
+	ht_cap.cap |= IEEE80211_HT_CAP_SUP_WIDTH_20_40;
+	ht_cap.cap |= IEEE80211_HT_CAP_DSSSCCK40;
+	ht_cap.cap |= WLAN_HT_CAP_SM_PS_STATIC << IEEE80211_HT_CAP_SM_PS_SHIFT;
+
+	if (ar_ht_cap & WMI_HT_CAP_HT20_SGI)
+		ht_cap.cap |= IEEE80211_HT_CAP_SGI_20;
+
+	if (ar_ht_cap & WMI_HT_CAP_HT40_SGI)
+		ht_cap.cap |= IEEE80211_HT_CAP_SGI_40;
+
+	if (ar_ht_cap & WMI_HT_CAP_DYNAMIC_SMPS) {
+		u32 smps;
+
+		smps   = WLAN_HT_CAP_SM_PS_DYNAMIC;
+		smps <<= IEEE80211_HT_CAP_SM_PS_SHIFT;
+
+		ht_cap.cap |= smps;
+	}
+
+	if (ar_ht_cap & WMI_HT_CAP_TX_STBC)
+		ht_cap.cap |= IEEE80211_HT_CAP_TX_STBC;
+
+	if (ar_ht_cap & WMI_HT_CAP_RX_STBC) {
+		u32 stbc;
+
+		stbc   = ar_ht_cap;
+		stbc  &= WMI_HT_CAP_RX_STBC;
+		stbc >>= WMI_HT_CAP_RX_STBC_MASK_SHIFT;
+		stbc <<= IEEE80211_HT_CAP_RX_STBC_SHIFT;
+		stbc  &= IEEE80211_HT_CAP_RX_STBC;
+
+		ht_cap.cap |= stbc;
+	}
+
+	if (ar_ht_cap & WMI_HT_CAP_RX_LDPC)
+		ht_cap.cap |= IEEE80211_HT_CAP_LDPC_CODING;
+
+	if (ar_ht_cap & WMI_HT_CAP_L_SIG_TXOP_PROT)
+		ht_cap.cap |= IEEE80211_HT_CAP_LSIG_TXOP_PROT;
+
+	if (ar_vht_cap & WMI_VHT_CAP_MAX_MPDU_LEN_MASK)
+		ht_cap.cap |= IEEE80211_HT_CAP_MAX_AMSDU;
+
+	for (i = 0; i < ar->num_rx_chains; i++) {
+		if (rate_cap_rx_chainmask & BIT(i))
+			ht_cap.mcs.rx_mask[i] = 0xFF;
+	}
+
+	ht_cap.mcs.tx_params |= IEEE80211_HT_MCS_TX_DEFINED;
+
+	return ht_cap;
+}
+
+static int ath11k_mac_set_txbf_conf(struct ath11k_vif *arvif)
+{
+	u32 value = 0;
+	struct ath11k *ar = arvif->ar;
+	int nsts;
+	int sound_dim;
+	u32 vht_cap = ar->pdev->cap.vht_cap;
+	u32 vdev_param = WMI_VDEV_PARAM_TXBF;
+
+	if (vht_cap & (IEEE80211_VHT_CAP_SU_BEAMFORMEE_CAPABLE)) {
+		nsts = vht_cap & IEEE80211_VHT_CAP_BEAMFORMEE_STS_MASK;
+		nsts >>= IEEE80211_VHT_CAP_BEAMFORMEE_STS_SHIFT;
+		value |= SM(nsts, WMI_TXBF_STS_CAP_OFFSET);
+	}
+
+	if (vht_cap & (IEEE80211_VHT_CAP_SU_BEAMFORMER_CAPABLE)) {
+		sound_dim = vht_cap &
+			    IEEE80211_VHT_CAP_SOUNDING_DIMENSIONS_MASK;
+		sound_dim >>= IEEE80211_VHT_CAP_SOUNDING_DIMENSIONS_SHIFT;
+		if (sound_dim > (ar->num_tx_chains - 1))
+			sound_dim = ar->num_tx_chains - 1;
+		value |= SM(sound_dim, WMI_BF_SOUND_DIM_OFFSET);
+	}
+
+	if (!value)
+		return 0;
+
+	if (vht_cap & IEEE80211_VHT_CAP_SU_BEAMFORMER_CAPABLE) {
+		value |= WMI_VDEV_PARAM_TXBF_SU_TX_BFER;
+
+		if ((vht_cap & IEEE80211_VHT_CAP_MU_BEAMFORMER_CAPABLE) &&
+		    arvif->vdev_type == WMI_VDEV_TYPE_AP)
+			value |= WMI_VDEV_PARAM_TXBF_MU_TX_BFER;
+	}
+
+	/* TODO: SUBFEE not validated in HK, disable here until validated? */
+
+	if (vht_cap & IEEE80211_VHT_CAP_SU_BEAMFORMEE_CAPABLE) {
+		value |= WMI_VDEV_PARAM_TXBF_SU_TX_BFEE;
+
+		if ((vht_cap & IEEE80211_VHT_CAP_MU_BEAMFORMEE_CAPABLE) &&
+		    arvif->vdev_type == WMI_VDEV_TYPE_STA)
+			value |= WMI_VDEV_PARAM_TXBF_MU_TX_BFEE;
+	}
+
+	return ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id,
+					     vdev_param, value);
+}
+
+static void ath11k_set_vht_txbf_cap(struct ath11k *ar, u32 *vht_cap)
+{
+	bool subfer, subfee;
+	int sound_dim = 0;
+
+	subfer = !!(*vht_cap & (IEEE80211_VHT_CAP_SU_BEAMFORMER_CAPABLE));
+	subfee = !!(*vht_cap & (IEEE80211_VHT_CAP_SU_BEAMFORMEE_CAPABLE));
+
+	if (ar->num_tx_chains < 2) {
+		*vht_cap &= ~(IEEE80211_VHT_CAP_SU_BEAMFORMER_CAPABLE);
+		subfer = false;
+	}
+
+	/* If SU Beaformer is not set, then disable MU Beamformer Capability */
+	if (!subfer)
+		*vht_cap &= ~(IEEE80211_VHT_CAP_MU_BEAMFORMER_CAPABLE);
+
+	/* If SU Beaformee is not set, then disable MU Beamformee Capability */
+	if (!subfee)
+		*vht_cap &= ~(IEEE80211_VHT_CAP_MU_BEAMFORMEE_CAPABLE);
+
+	sound_dim = (*vht_cap & IEEE80211_VHT_CAP_SOUNDING_DIMENSIONS_MASK);
+	sound_dim >>= IEEE80211_VHT_CAP_SOUNDING_DIMENSIONS_SHIFT;
+	*vht_cap &= ~IEEE80211_VHT_CAP_SOUNDING_DIMENSIONS_MASK;
+
+	/* TODO: Need to check invalid STS and Sound_dim values set by FW? */
+
+	/* Enable Sounding Dimension Field only if SU BF is enabled */
+	if (subfer) {
+		if (sound_dim > (ar->num_tx_chains - 1))
+			sound_dim = ar->num_tx_chains - 1;
+
+		sound_dim <<= IEEE80211_VHT_CAP_SOUNDING_DIMENSIONS_SHIFT;
+		sound_dim &=  IEEE80211_VHT_CAP_SOUNDING_DIMENSIONS_MASK;
+		*vht_cap |= sound_dim;
+	}
+
+	/* Use the STS advertised by FW unless SU Beamformee is not supported*/
+	if (!subfee)
+		*vht_cap &= ~(IEEE80211_VHT_CAP_BEAMFORMEE_STS_MASK);
+}
+
+static struct ieee80211_sta_vht_cap
+ath11k_create_vht_cap(struct ath11k *ar, u32 rate_cap_tx_chainmask,
+		      u32 rate_cap_rx_chainmask)
+{
+	struct ieee80211_sta_vht_cap vht_cap = {0};
+	u16 txmcs_map, rxmcs_map;
+	int i;
+
+	vht_cap.vht_supported = 1;
+	vht_cap.cap = ar->pdev->cap.vht_cap;
+
+	ath11k_set_vht_txbf_cap(ar, &vht_cap.cap);
+
+	/* TODO: Enable back VHT160 mode once association issues are fixed */
+	/* Disabling VHT160 and VHT80+80 modes */
+	vht_cap.cap &= ~IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_MASK;
+	vht_cap.cap &= ~IEEE80211_VHT_CAP_SHORT_GI_160;
+
+	rxmcs_map = 0;
+	txmcs_map = 0;
+	for (i = 0; i < 8; i++) {
+		if (i < ar->num_tx_chains && rate_cap_tx_chainmask & BIT(i))
+			txmcs_map |= IEEE80211_VHT_MCS_SUPPORT_0_9 << (i * 2);
+		else
+			txmcs_map |= IEEE80211_VHT_MCS_NOT_SUPPORTED << (i * 2);
+
+		if (i < ar->num_rx_chains && rate_cap_rx_chainmask & BIT(i))
+			rxmcs_map |= IEEE80211_VHT_MCS_SUPPORT_0_9 << (i * 2);
+		else
+			rxmcs_map |= IEEE80211_VHT_MCS_NOT_SUPPORTED << (i * 2);
+	}
+
+	if (rate_cap_tx_chainmask <= 1)
+		vht_cap.cap &= ~IEEE80211_VHT_CAP_TXSTBC;
+
+	vht_cap.vht_mcs.rx_mcs_map = cpu_to_le16(rxmcs_map);
+	vht_cap.vht_mcs.tx_mcs_map = cpu_to_le16(txmcs_map);
+
+	return vht_cap;
+}
+
+static void ath11k_mac_setup_ht_vht_cap(struct ath11k *ar,
+					struct ath11k_pdev_cap *cap,
+					u32 *ht_cap_info)
+{
+	struct ieee80211_supported_band *band;
+	u32 rate_cap_tx_chainmask;
+	u32 rate_cap_rx_chainmask;
+	u32 ht_cap;
+
+	rate_cap_tx_chainmask = ar->cfg_tx_chainmask >> cap->tx_chain_mask_shift;
+	rate_cap_rx_chainmask = ar->cfg_rx_chainmask >> cap->rx_chain_mask_shift;
+
+	if (cap->supported_bands & WMI_HOST_WLAN_2G_CAP) {
+		band = &ar->mac.sbands[NL80211_BAND_2GHZ];
+		ht_cap = cap->band[NL80211_BAND_2GHZ].ht_cap_info;
+		if (ht_cap_info)
+			*ht_cap_info = ht_cap;
+		band->ht_cap = ath11k_create_ht_cap(ar, ht_cap,
+						    rate_cap_rx_chainmask);
+	}
+
+	if (cap->supported_bands & WMI_HOST_WLAN_5G_CAP) {
+		band = &ar->mac.sbands[NL80211_BAND_5GHZ];
+		ht_cap = cap->band[NL80211_BAND_5GHZ].ht_cap_info;
+		if (ht_cap_info)
+			*ht_cap_info = ht_cap;
+		band->ht_cap = ath11k_create_ht_cap(ar, ht_cap,
+						    rate_cap_rx_chainmask);
+		band->vht_cap = ath11k_create_vht_cap(ar, rate_cap_tx_chainmask,
+						      rate_cap_rx_chainmask);
+	}
+}
+
+static int ath11k_check_chain_mask(struct ath11k *ar, u32 ant, bool is_tx_ant)
+{
+	/* TODO: Check the request chainmask against the supported
+	 * chainmask table which is advertised in extented_service_ready event
+	 */
+
+	return 0;
+}
+
+static int __ath11k_set_antenna(struct ath11k *ar, u32 tx_ant, u32 rx_ant)
+{
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	if (ath11k_check_chain_mask(ar, tx_ant, true))
+		return -EINVAL;
+
+	if (ath11k_check_chain_mask(ar, rx_ant, false))
+		return -EINVAL;
+
+	ar->cfg_tx_chainmask = tx_ant;
+	ar->cfg_rx_chainmask = rx_ant;
+
+	if (ar->state != ATH11K_STATE_ON &&
+	    ar->state != ATH11K_STATE_RESTARTED)
+		return 0;
+
+	ret = ath11k_wmi_pdev_set_param(ar, WMI_PDEV_PARAM_TX_CHAIN_MASK,
+					tx_ant, ar->pdev->pdev_id);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to set tx-chainmask: %d, req 0x%x\n",
+			    ret, tx_ant);
+		return ret;
+	}
+
+	ret = ath11k_wmi_pdev_set_param(ar, WMI_PDEV_PARAM_RX_CHAIN_MASK,
+					rx_ant, ar->pdev->pdev_id);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to set rx-chainmask: %d, req 0x%x\n",
+			    ret, rx_ant);
+		return ret;
+	}
+
+	/* Reload HT/VHT capability */
+	ath11k_mac_setup_ht_vht_cap(ar, &ar->pdev->cap, NULL);
+
+	return 0;
+}
+
+int ath11k_mac_tx_mgmt_pending_free(int buf_id, void *skb, void *ctx)
+{
+	struct ath11k *ar = ctx;
+	struct ath11k_base *ab = ar->ab;
+	struct sk_buff *msdu = skb;
+	struct ieee80211_tx_info *info;
+
+	spin_lock_bh(&ar->txmgmt_idr_lock);
+	idr_remove(&ar->txmgmt_idr, buf_id);
+	spin_unlock_bh(&ar->txmgmt_idr_lock);
+	dma_unmap_single(ab->dev, ATH11K_SKB_CB(msdu)->paddr, msdu->len,
+			 DMA_TO_DEVICE);
+
+	info = IEEE80211_SKB_CB(msdu);
+	memset(&info->status, 0, sizeof(info->status));
+
+	ieee80211_free_txskb(ar->hw, msdu);
+
+	return 0;
+}
+
+static int ath11k_mac_vif_txmgmt_idr_remove(int buf_id, void *skb, void *ctx)
+{
+	struct ieee80211_vif *vif = ctx;
+	struct ath11k_skb_cb *skb_cb = ATH11K_SKB_CB((struct sk_buff *)skb);
+	struct sk_buff *msdu = skb;
+	struct ath11k *ar = skb_cb->ar;
+	struct ath11k_base *ab = ar->ab;
+
+	if (skb_cb->vif == vif) {
+		spin_lock_bh(&ar->txmgmt_idr_lock);
+		idr_remove(&ar->txmgmt_idr, buf_id);
+		spin_unlock_bh(&ar->txmgmt_idr_lock);
+		dma_unmap_single(ab->dev, skb_cb->paddr, msdu->len,
+				 DMA_TO_DEVICE);
+	}
+
+	return 0;
+}
+
+static int ath11k_mac_mgmt_tx_wmi(struct ath11k *ar, struct ath11k_vif *arvif,
+				  struct sk_buff *skb)
+{
+	struct ath11k_base *ab = ar->ab;
+	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data;
+	dma_addr_t paddr;
+	int buf_id;
+	int ret;
+
+	spin_lock_bh(&ar->txmgmt_idr_lock);
+	buf_id = idr_alloc(&ar->txmgmt_idr, skb, 0,
+			   ATH11K_TX_MGMT_NUM_PENDING_MAX, GFP_ATOMIC);
+	spin_unlock_bh(&ar->txmgmt_idr_lock);
+	if (buf_id < 0)
+		return -ENOSPC;
+
+	if ((ieee80211_is_action(hdr->frame_control) ||
+	     ieee80211_is_deauth(hdr->frame_control) ||
+	     ieee80211_is_disassoc(hdr->frame_control)) &&
+	     ieee80211_has_protected(hdr->frame_control)) {
+		skb_put(skb, IEEE80211_CCMP_MIC_LEN);
+	}
+
+	paddr = dma_map_single(ab->dev, skb->data, skb->len, DMA_TO_DEVICE);
+	if (dma_mapping_error(ab->dev, paddr)) {
+		ath11k_warn(ab, "failed to DMA map mgmt Tx buffer\n");
+		ret = -EIO;
+		goto err_free_idr;
+	}
+
+	ATH11K_SKB_CB(skb)->paddr = paddr;
+
+	ret = ath11k_wmi_mgmt_send(ar, arvif->vdev_id, buf_id, skb);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to send mgmt frame: %d\n", ret);
+		goto err_unmap_buf;
+	}
+
+	return 0;
+
+err_unmap_buf:
+	dma_unmap_single(ab->dev, ATH11K_SKB_CB(skb)->paddr,
+			 skb->len, DMA_TO_DEVICE);
+err_free_idr:
+	spin_lock_bh(&ar->txmgmt_idr_lock);
+	idr_remove(&ar->txmgmt_idr, buf_id);
+	spin_unlock_bh(&ar->txmgmt_idr_lock);
+
+	return ret;
+}
+
+static void ath11k_mgmt_over_wmi_tx_purge(struct ath11k *ar)
+{
+	struct sk_buff *skb;
+
+	while ((skb = skb_dequeue(&ar->wmi_mgmt_tx_queue)) != NULL)
+		ieee80211_free_txskb(ar->hw, skb);
+}
+
+static void ath11k_mgmt_over_wmi_tx_work(struct work_struct *work)
+{
+	struct ath11k *ar = container_of(work, struct ath11k, wmi_mgmt_tx_work);
+	struct ieee80211_tx_info *info;
+	struct ath11k_vif *arvif;
+	struct sk_buff *skb;
+	int ret;
+
+	while ((skb = skb_dequeue(&ar->wmi_mgmt_tx_queue)) != NULL) {
+		info = IEEE80211_SKB_CB(skb);
+		arvif = ath11k_vif_to_arvif(info->control.vif);
+
+		ret = ath11k_mac_mgmt_tx_wmi(ar, arvif, skb);
+		if (ret) {
+			ath11k_warn(ar->ab, "failed to transmit management frame %d\n",
+				    ret);
+			ieee80211_free_txskb(ar->hw, skb);
+		} else {
+			atomic_inc(&ar->num_pending_mgmt_tx);
+		}
+	}
+}
+
+static int ath11k_mac_mgmt_tx(struct ath11k *ar, struct sk_buff *skb,
+			      bool is_prb_rsp)
+{
+	struct sk_buff_head *q = &ar->wmi_mgmt_tx_queue;
+
+	if (test_bit(ATH11K_FLAG_CRASH_FLUSH, &ar->ab->dev_flags))
+		return -ESHUTDOWN;
+
+	/* Drop probe response packets when the pending management tx
+	 * count has reached a certain threshold, so as to prioritize
+	 * other mgmt packets like auth and assoc to be sent on time
+	 * for establishing successful connections.
+	 */
+	if (is_prb_rsp &&
+	    atomic_read(&ar->num_pending_mgmt_tx) > ATH11K_PRB_RSP_DROP_THRESHOLD) {
+		ath11k_warn(ar->ab,
+			    "dropping probe response as pending queue is almost full\n");
+		return -ENOSPC;
+	}
+
+	if (skb_queue_len(q) == ATH11K_TX_MGMT_NUM_PENDING_MAX) {
+		ath11k_warn(ar->ab, "mgmt tx queue is full\n");
+		return -ENOSPC;
+	}
+
+	skb_queue_tail(q, skb);
+	ieee80211_queue_work(ar->hw, &ar->wmi_mgmt_tx_work);
+
+	return 0;
+}
+
+static void ath11k_mac_op_tx(struct ieee80211_hw *hw,
+			     struct ieee80211_tx_control *control,
+			     struct sk_buff *skb)
+{
+	struct ath11k *ar = hw->priv;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
+	struct ieee80211_vif *vif = info->control.vif;
+	struct ath11k_vif *arvif = ath11k_vif_to_arvif(vif);
+	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data;
+	bool is_prb_rsp;
+	int ret;
+
+	if (ieee80211_is_mgmt(hdr->frame_control)) {
+		is_prb_rsp = ieee80211_is_probe_resp(hdr->frame_control);
+		ret = ath11k_mac_mgmt_tx(ar, skb, is_prb_rsp);
+		if (ret) {
+			ath11k_warn(ar->ab, "failed to queue management frame %d\n",
+				    ret);
+			ieee80211_free_txskb(ar->hw, skb);
+		}
+		return;
+	}
+
+	ret = ath11k_dp_tx(ar, arvif, skb);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to transmit frame %d\n", ret);
+		ieee80211_free_txskb(ar->hw, skb);
+	}
+}
+
+void ath11k_mac_drain_tx(struct ath11k *ar)
+{
+	/* make sure rcu-protected mac80211 tx path itself is drained */
+	synchronize_net();
+
+	cancel_work_sync(&ar->wmi_mgmt_tx_work);
+	ath11k_mgmt_over_wmi_tx_purge(ar);
+}
+
+static int ath11k_mac_config_mon_status_default(struct ath11k *ar, bool enable)
+{
+	struct htt_rx_ring_tlv_filter tlv_filter = {0};
+	u32 ring_id;
+
+	if (enable)
+		tlv_filter = ath11k_mac_mon_status_filter_default;
+
+	ring_id = ar->dp.rx_mon_status_refill_ring.refill_buf_ring.ring_id;
+
+	return ath11k_dp_tx_htt_rx_filter_setup(ar->ab, ring_id, ar->dp.mac_id,
+						HAL_RXDMA_MONITOR_STATUS,
+						DP_RX_BUFFER_SIZE, &tlv_filter);
+}
+
+static int ath11k_mac_op_start(struct ieee80211_hw *hw)
+{
+	struct ath11k *ar = hw->priv;
+	struct ath11k_base *ab = ar->ab;
+	struct ath11k_pdev *pdev = ar->pdev;
+	int ret;
+
+	ath11k_mac_drain_tx(ar);
+	mutex_lock(&ar->conf_mutex);
+
+	switch (ar->state) {
+	case ATH11K_STATE_OFF:
+		ar->state = ATH11K_STATE_ON;
+		break;
+	case ATH11K_STATE_RESTARTING:
+		ar->state = ATH11K_STATE_RESTARTED;
+		break;
+	case ATH11K_STATE_RESTARTED:
+	case ATH11K_STATE_WEDGED:
+	case ATH11K_STATE_ON:
+		WARN_ON(1);
+		ret = -EINVAL;
+		goto err;
+	}
+
+	ret = ath11k_wmi_pdev_set_param(ar, WMI_PDEV_PARAM_PMF_QOS,
+					1, pdev->pdev_id);
+
+	if (ret) {
+		ath11k_err(ar->ab, "failed to enable PMF QOS: (%d\n", ret);
+		goto err;
+	}
+
+	ret = ath11k_wmi_pdev_set_param(ar, WMI_PDEV_PARAM_DYNAMIC_BW, 1,
+					pdev->pdev_id);
+	if (ret) {
+		ath11k_err(ar->ab, "failed to enable dynamic bw: %d\n", ret);
+		goto err;
+	}
+
+	ret = ath11k_wmi_pdev_set_param(ar, WMI_PDEV_PARAM_ARP_AC_OVERRIDE,
+					0, pdev->pdev_id);
+	if (ret) {
+		ath11k_err(ab, "failed to set ac override for ARP: %d\n",
+			   ret);
+		goto err;
+	}
+
+	ret = ath11k_wmi_send_dfs_phyerr_offload_enable_cmd(ar, pdev->pdev_id);
+	if (ret) {
+		ath11k_err(ab, "failed to offload radar detection: %d\n",
+			   ret);
+		goto err;
+	}
+
+	ret = ath11k_dp_tx_htt_h2t_ppdu_stats_req(ar,
+						  HTT_PPDU_STATS_TAG_DEFAULT);
+	if (ret) {
+		ath11k_err(ab, "failed to req ppdu stats: %d\n", ret);
+		goto err;
+	}
+
+	ret = ath11k_wmi_pdev_set_param(ar, WMI_PDEV_PARAM_MESH_MCAST_ENABLE,
+					1, pdev->pdev_id);
+
+	if (ret) {
+		ath11k_err(ar->ab, "failed to enable MESH MCAST ENABLE: (%d\n", ret);
+		goto err;
+	}
+
+	__ath11k_set_antenna(ar, ar->cfg_tx_chainmask, ar->cfg_rx_chainmask);
+
+	/* TODO: Do we need to enable ANI? */
+
+	ath11k_reg_update_chan_list(ar);
+
+	ar->num_started_vdevs = 0;
+	ar->num_created_vdevs = 0;
+	ar->num_peers = 0;
+
+	/* Configure monitor status ring with default rx_filter to get rx status
+	 * such as rssi, rx_duration.
+	 */
+	ret = ath11k_mac_config_mon_status_default(ar, true);
+	if (ret) {
+		ath11k_err(ab, "failed to configure monitor status ring with default rx_filter: (%d)\n",
+			   ret);
+		goto err;
+	}
+
+	mutex_unlock(&ar->conf_mutex);
+
+	rcu_assign_pointer(ab->pdevs_active[ar->pdev_idx],
+			   &ab->pdevs[ar->pdev_idx]);
+
+	return 0;
+
+err:
+	ar->state = ATH11K_STATE_OFF;
+	mutex_unlock(&ar->conf_mutex);
+
+	return ret;
+}
+
+static void ath11k_mac_op_stop(struct ieee80211_hw *hw)
+{
+	struct ath11k *ar = hw->priv;
+	struct htt_ppdu_stats_info *ppdu_stats, *tmp;
+	int ret;
+
+	ath11k_mac_drain_tx(ar);
+
+	mutex_lock(&ar->conf_mutex);
+	ret = ath11k_mac_config_mon_status_default(ar, false);
+	if (ret)
+		ath11k_err(ar->ab, "failed to clear rx_filter for monitor status ring: (%d)\n",
+			   ret);
+
+	clear_bit(ATH11K_CAC_RUNNING, &ar->dev_flags);
+	ar->state = ATH11K_STATE_OFF;
+	mutex_unlock(&ar->conf_mutex);
+
+	cancel_delayed_work_sync(&ar->scan.timeout);
+	cancel_work_sync(&ar->regd_update_work);
+
+	spin_lock_bh(&ar->data_lock);
+	list_for_each_entry_safe(ppdu_stats, tmp, &ar->ppdu_stats_info, list) {
+		list_del(&ppdu_stats->list);
+		kfree(ppdu_stats);
+	}
+	spin_unlock_bh(&ar->data_lock);
+
+	rcu_assign_pointer(ar->ab->pdevs_active[ar->pdev_idx], NULL);
+
+	synchronize_rcu();
+
+	atomic_set(&ar->num_pending_mgmt_tx, 0);
+}
+
+static void
+ath11k_mac_setup_vdev_create_params(struct ath11k_vif *arvif,
+				    struct vdev_create_params *params)
+{
+	struct ath11k *ar = arvif->ar;
+	struct ath11k_pdev *pdev = ar->pdev;
+
+	params->if_id = arvif->vdev_id;
+	params->type = arvif->vdev_type;
+	params->subtype = arvif->vdev_subtype;
+	params->pdev_id = pdev->pdev_id;
+
+	if (pdev->cap.supported_bands & WMI_HOST_WLAN_2G_CAP) {
+		params->chains[NL80211_BAND_2GHZ].tx = ar->num_tx_chains;
+		params->chains[NL80211_BAND_2GHZ].rx = ar->num_rx_chains;
+	}
+	if (pdev->cap.supported_bands & WMI_HOST_WLAN_5G_CAP) {
+		params->chains[NL80211_BAND_5GHZ].tx = ar->num_tx_chains;
+		params->chains[NL80211_BAND_5GHZ].rx = ar->num_rx_chains;
+	}
+}
+
+static u32
+ath11k_mac_prepare_he_mode(struct ath11k_pdev *pdev, u32 viftype)
+{
+	struct ath11k_pdev_cap *pdev_cap = &pdev->cap;
+	struct ath11k_band_cap *cap_band = NULL;
+	u32 *hecap_phy_ptr = NULL;
+	u32 hemode = 0;
+
+	if (pdev->cap.supported_bands & WMI_HOST_WLAN_2G_CAP)
+		cap_band = &pdev_cap->band[NL80211_BAND_2GHZ];
+	else
+		cap_band = &pdev_cap->band[NL80211_BAND_5GHZ];
+
+	hecap_phy_ptr = &cap_band->he_cap_phy_info[0];
+
+	hemode = FIELD_PREP(HE_MODE_SU_TX_BFEE, HE_SU_BFEE_ENABLE) |
+		FIELD_PREP(HE_MODE_SU_TX_BFER, HECAP_PHY_SUBFMR_GET(hecap_phy_ptr)) |
+		FIELD_PREP(HE_MODE_UL_MUMIMO, HECAP_PHY_ULMUMIMO_GET(hecap_phy_ptr));
+
+	/* TODO WDS and other modes */
+	if (viftype == NL80211_IFTYPE_AP) {
+		hemode |= FIELD_PREP(HE_MODE_MU_TX_BFER,
+			   HECAP_PHY_MUBFMR_GET(hecap_phy_ptr)) |
+			FIELD_PREP(HE_MODE_DL_OFDMA, HE_DL_MUOFDMA_ENABLE) |
+			FIELD_PREP(HE_MODE_UL_OFDMA, HE_UL_MUOFDMA_ENABLE);
+	} else {
+		hemode |= FIELD_PREP(HE_MODE_MU_TX_BFEE, HE_MU_BFEE_ENABLE);
+	}
+
+	return hemode;
+}
+
+static int ath11k_set_he_mu_sounding_mode(struct ath11k *ar,
+					  struct ath11k_vif *arvif)
+{
+	u32 param_id, param_value;
+	struct ath11k_base *ab = ar->ab;
+	int ret = 0;
+
+	param_id = WMI_VDEV_PARAM_SET_HEMU_MODE;
+	param_value = ath11k_mac_prepare_he_mode(ar->pdev, arvif->vif->type);
+	ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id,
+					    param_id, param_value);
+	if (ret) {
+		ath11k_warn(ab, "failed to set vdev %d HE MU mode: %d param_value %x\n",
+			    arvif->vdev_id, ret, param_value);
+		return ret;
+	}
+	param_id = WMI_VDEV_PARAM_SET_HE_SOUNDING_MODE;
+	param_value =
+		FIELD_PREP(HE_VHT_SOUNDING_MODE, HE_VHT_SOUNDING_MODE_ENABLE) |
+		FIELD_PREP(HE_TRIG_NONTRIG_SOUNDING_MODE,
+			   HE_TRIG_NONTRIG_SOUNDING_MODE_ENABLE);
+	ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id,
+					    param_id, param_value);
+	if (ret) {
+		ath11k_warn(ab, "failed to set vdev %d HE MU mode: %d\n",
+			    arvif->vdev_id, ret);
+		return ret;
+	}
+	return ret;
+}
+
+static int ath11k_mac_op_add_interface(struct ieee80211_hw *hw,
+				       struct ieee80211_vif *vif)
+{
+	struct ath11k *ar = hw->priv;
+	struct ath11k_base *ab = ar->ab;
+	struct ath11k_vif *arvif = ath11k_vif_to_arvif(vif);
+	struct vdev_create_params vdev_param = {0};
+	struct peer_create_params peer_param;
+	u32 param_id, param_value;
+	u16 nss;
+	int i;
+	int ret;
+	int bit;
+
+	vif->driver_flags |= IEEE80211_VIF_SUPPORTS_UAPSD;
+
+	mutex_lock(&ar->conf_mutex);
+
+	if (vif->type == NL80211_IFTYPE_AP &&
+	    ar->num_peers > (ar->max_num_peers - 1)) {
+		ath11k_warn(ab, "failed to create vdev due to insufficient peer entry resource in firmware\n");
+		ret = -ENOBUFS;
+		goto err;
+	}
+
+	if (ar->num_created_vdevs > (TARGET_NUM_VDEVS - 1)) {
+		ath11k_warn(ab, "failed to create vdev, reached max vdev limit %d\n",
+			    TARGET_NUM_VDEVS);
+		ret = -EBUSY;
+		goto err;
+	}
+
+	memset(arvif, 0, sizeof(*arvif));
+
+	arvif->ar = ar;
+	arvif->vif = vif;
+
+	INIT_LIST_HEAD(&arvif->list);
+
+	/* Should we initialize any worker to handle connection loss indication
+	 * from firmware in sta mode?
+	 */
+
+	for (i = 0; i < ARRAY_SIZE(arvif->bitrate_mask.control); i++) {
+		arvif->bitrate_mask.control[i].legacy = 0xffffffff;
+		memset(arvif->bitrate_mask.control[i].ht_mcs, 0xff,
+		       sizeof(arvif->bitrate_mask.control[i].ht_mcs));
+		memset(arvif->bitrate_mask.control[i].vht_mcs, 0xff,
+		       sizeof(arvif->bitrate_mask.control[i].vht_mcs));
+	}
+
+	bit = __ffs64(ab->free_vdev_map);
+
+	arvif->vdev_id = bit;
+	arvif->vdev_subtype = WMI_VDEV_SUBTYPE_NONE;
+
+	switch (vif->type) {
+	case NL80211_IFTYPE_UNSPECIFIED:
+	case NL80211_IFTYPE_STATION:
+		arvif->vdev_type = WMI_VDEV_TYPE_STA;
+		break;
+	case NL80211_IFTYPE_MESH_POINT:
+		arvif->vdev_subtype = WMI_VDEV_SUBTYPE_MESH_11S;
+		/* fall through */
+	case NL80211_IFTYPE_AP:
+		arvif->vdev_type = WMI_VDEV_TYPE_AP;
+		break;
+	case NL80211_IFTYPE_MONITOR:
+		arvif->vdev_type = WMI_VDEV_TYPE_MONITOR;
+		break;
+	default:
+		WARN_ON(1);
+		break;
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "mac add interface id %d type %d subtype %d map %llx\n",
+		   arvif->vdev_id, arvif->vdev_type, arvif->vdev_subtype,
+		   ab->free_vdev_map);
+
+	vif->cab_queue = arvif->vdev_id % (ATH11K_HW_MAX_QUEUES - 1);
+	for (i = 0; i < ARRAY_SIZE(vif->hw_queue); i++)
+		vif->hw_queue[i] = i % (ATH11K_HW_MAX_QUEUES - 1);
+
+	ath11k_mac_setup_vdev_create_params(arvif, &vdev_param);
+
+	ret = ath11k_wmi_vdev_create(ar, vif->addr, &vdev_param);
+	if (ret) {
+		ath11k_warn(ab, "failed to create WMI vdev %d: %d\n",
+			    arvif->vdev_id, ret);
+		goto err;
+	}
+
+	ar->num_created_vdevs++;
+
+	ab->free_vdev_map &= ~(1LL << arvif->vdev_id);
+	spin_lock_bh(&ar->data_lock);
+	list_add(&arvif->list, &ar->arvifs);
+	spin_unlock_bh(&ar->data_lock);
+
+	param_id = WMI_VDEV_PARAM_TX_ENCAP_TYPE;
+	param_value = ATH11K_HW_TXRX_NATIVE_WIFI;
+	ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id,
+					    param_id, param_value);
+	if (ret) {
+		ath11k_warn(ab, "failed to set vdev %d tx encap mode: %d\n",
+			    arvif->vdev_id, ret);
+		goto err_vdev_del;
+	}
+
+	nss = get_num_chains(ar->cfg_tx_chainmask) ? : 1;
+	ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id,
+					    WMI_VDEV_PARAM_NSS, nss);
+	if (ret) {
+		ath11k_warn(ab, "failed to set vdev %d chainmask 0x%x, nss %d :%d\n",
+			    arvif->vdev_id, ar->cfg_tx_chainmask, nss, ret);
+		goto err_vdev_del;
+	}
+
+	switch (arvif->vdev_type) {
+	case WMI_VDEV_TYPE_AP:
+		peer_param.vdev_id = arvif->vdev_id;
+		peer_param.peer_addr = vif->addr;
+		peer_param.peer_type = WMI_PEER_TYPE_DEFAULT;
+		ret = ath11k_peer_create(ar, arvif, NULL, &peer_param);
+		if (ret) {
+			ath11k_warn(ab, "failed to vdev %d create peer for AP: %d\n",
+				    arvif->vdev_id, ret);
+			goto err_vdev_del;
+		}
+
+		ret = ath11k_mac_set_kickout(arvif);
+		if (ret) {
+			ath11k_warn(ar->ab, "failed to set vdev %i kickout parameters: %d\n",
+				    arvif->vdev_id, ret);
+			goto err_peer_del;
+		}
+		break;
+	case WMI_VDEV_TYPE_STA:
+		param_id = WMI_STA_PS_PARAM_RX_WAKE_POLICY;
+		param_value = WMI_STA_PS_RX_WAKE_POLICY_WAKE;
+		ret = ath11k_wmi_set_sta_ps_param(ar, arvif->vdev_id,
+						  param_id, param_value);
+		if (ret) {
+			ath11k_warn(ar->ab, "failed to set vdev %d RX wake policy: %d\n",
+				    arvif->vdev_id, ret);
+			goto err_peer_del;
+		}
+
+		param_id = WMI_STA_PS_PARAM_TX_WAKE_THRESHOLD;
+		param_value = WMI_STA_PS_TX_WAKE_THRESHOLD_ALWAYS;
+		ret = ath11k_wmi_set_sta_ps_param(ar, arvif->vdev_id,
+						  param_id, param_value);
+		if (ret) {
+			ath11k_warn(ar->ab, "failed to set vdev %d TX wake threshold: %d\n",
+				    arvif->vdev_id, ret);
+			goto err_peer_del;
+		}
+
+		param_id = WMI_STA_PS_PARAM_PSPOLL_COUNT;
+		param_value = WMI_STA_PS_PSPOLL_COUNT_NO_MAX;
+		ret = ath11k_wmi_set_sta_ps_param(ar, arvif->vdev_id,
+						  param_id, param_value);
+		if (ret) {
+			ath11k_warn(ar->ab, "failed to set vdev %d pspoll count: %d\n",
+				    arvif->vdev_id, ret);
+			goto err_peer_del;
+		}
+		break;
+	default:
+		break;
+	}
+
+	arvif->txpower = vif->bss_conf.txpower;
+	ret = ath11k_mac_txpower_recalc(ar);
+	if (ret)
+		goto err_peer_del;
+
+	param_id = WMI_VDEV_PARAM_RTS_THRESHOLD;
+	param_value = ar->hw->wiphy->rts_threshold;
+	ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id,
+					    param_id, param_value);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to set rts threshold for vdev %d: %d\n",
+			    arvif->vdev_id, ret);
+	}
+
+	ret = ath11k_mac_set_txbf_conf(arvif);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to set txbf conf for vdev %d: %d\n",
+			    arvif->vdev_id, ret);
+	}
+
+	ath11k_dp_vdev_tx_attach(ar, arvif);
+
+	mutex_unlock(&ar->conf_mutex);
+
+	return 0;
+
+err_peer_del:
+	if (arvif->vdev_type == WMI_VDEV_TYPE_AP) {
+		ar->num_peers--;
+		ath11k_wmi_send_peer_delete_cmd(ar, vif->addr, arvif->vdev_id);
+	}
+
+err_vdev_del:
+	ath11k_wmi_vdev_delete(ar, arvif->vdev_id);
+	ar->num_created_vdevs--;
+	ab->free_vdev_map |= 1LL << arvif->vdev_id;
+	spin_lock_bh(&ar->data_lock);
+	list_del(&arvif->list);
+	spin_unlock_bh(&ar->data_lock);
+
+err:
+	mutex_unlock(&ar->conf_mutex);
+
+	return ret;
+}
+
+static int ath11k_mac_vif_unref(int buf_id, void *skb, void *ctx)
+{
+	struct ieee80211_vif *vif = (struct ieee80211_vif *)ctx;
+	struct ath11k_skb_cb *skb_cb = ATH11K_SKB_CB((struct sk_buff *)skb);
+
+	if (skb_cb->vif == vif)
+		skb_cb->vif = NULL;
+
+	return 0;
+}
+
+static void ath11k_mac_op_remove_interface(struct ieee80211_hw *hw,
+					   struct ieee80211_vif *vif)
+{
+	struct ath11k *ar = hw->priv;
+	struct ath11k_vif *arvif = ath11k_vif_to_arvif(vif);
+	struct ath11k_base *ab = ar->ab;
+	int ret;
+	int i;
+
+	mutex_lock(&ar->conf_mutex);
+
+	ath11k_dbg(ab, ATH11K_DBG_MAC, "mac remove interface (vdev %d)\n",
+		   arvif->vdev_id);
+
+	ab->free_vdev_map |= 1LL << (arvif->vdev_id);
+	spin_lock_bh(&ar->data_lock);
+	list_del(&arvif->list);
+	spin_unlock_bh(&ar->data_lock);
+
+	if (arvif->vdev_type == WMI_VDEV_TYPE_AP) {
+		ret = ath11k_peer_delete(ar, arvif->vdev_id, vif->addr);
+		if (ret)
+			ath11k_warn(ab, "failed to submit AP self-peer removal on vdev %d: %d\n",
+				    arvif->vdev_id, ret);
+	}
+
+	ret = ath11k_wmi_vdev_delete(ar, arvif->vdev_id);
+	if (ret)
+		ath11k_warn(ab, "failed to delete WMI vdev %d: %d\n",
+			    arvif->vdev_id, ret);
+
+	ar->num_created_vdevs--;
+
+	ath11k_peer_cleanup(ar, arvif->vdev_id);
+
+	idr_for_each(&ar->txmgmt_idr,
+		     ath11k_mac_vif_txmgmt_idr_remove, vif);
+
+	for (i = 0; i < DP_TCL_NUM_RING_MAX; i++) {
+		spin_lock_bh(&ab->dp.tx_ring[i].tx_idr_lock);
+		idr_for_each(&ab->dp.tx_ring[i].txbuf_idr,
+			     ath11k_mac_vif_unref, vif);
+		spin_unlock_bh(&ab->dp.tx_ring[i].tx_idr_lock);
+	}
+
+	/* Recalc txpower for remaining vdev */
+	ath11k_mac_txpower_recalc(ar);
+	clear_bit(ATH11K_FLAG_MONITOR_ENABLED, &ar->monitor_flags);
+
+	/* TODO: recal traffic pause state based on the available vdevs */
+
+	mutex_unlock(&ar->conf_mutex);
+}
+
+/* FIXME: Has to be verified. */
+#define SUPPORTED_FILTERS			\
+	(FIF_ALLMULTI |				\
+	FIF_CONTROL |				\
+	FIF_PSPOLL |				\
+	FIF_OTHER_BSS |				\
+	FIF_BCN_PRBRESP_PROMISC |		\
+	FIF_PROBE_REQ |				\
+	FIF_FCSFAIL)
+
+static void ath11k_mac_op_configure_filter(struct ieee80211_hw *hw,
+					   unsigned int changed_flags,
+					   unsigned int *total_flags,
+					   u64 multicast)
+{
+	struct ath11k *ar = hw->priv;
+	bool reset_flag = false;
+	int ret = 0;
+
+	mutex_lock(&ar->conf_mutex);
+
+	changed_flags &= SUPPORTED_FILTERS;
+	*total_flags &= SUPPORTED_FILTERS;
+	ar->filter_flags = *total_flags;
+
+	/* For monitor mode */
+	reset_flag = !(ar->filter_flags & FIF_BCN_PRBRESP_PROMISC);
+
+	ret = ath11k_dp_tx_htt_monitor_mode_ring_config(ar, reset_flag);
+	if (!ret) {
+		if (!reset_flag)
+			set_bit(ATH11K_FLAG_MONITOR_ENABLED, &ar->monitor_flags);
+		else
+			clear_bit(ATH11K_FLAG_MONITOR_ENABLED, &ar->monitor_flags);
+	} else {
+		ath11k_warn(ar->ab,
+			    "fail to set monitor filter: %d\n", ret);
+	}
+	mutex_unlock(&ar->conf_mutex);
+}
+
+static int ath11k_mac_op_get_antenna(struct ieee80211_hw *hw, u32 *tx_ant, u32 *rx_ant)
+{
+	struct ath11k *ar = hw->priv;
+
+	mutex_lock(&ar->conf_mutex);
+
+	*tx_ant = ar->cfg_tx_chainmask;
+	*rx_ant = ar->cfg_rx_chainmask;
+
+	mutex_unlock(&ar->conf_mutex);
+
+	return 0;
+}
+
+static int ath11k_mac_op_set_antenna(struct ieee80211_hw *hw, u32 tx_ant, u32 rx_ant)
+{
+	struct ath11k *ar = hw->priv;
+	int ret;
+
+	mutex_lock(&ar->conf_mutex);
+	ret = __ath11k_set_antenna(ar, tx_ant, rx_ant);
+	mutex_unlock(&ar->conf_mutex);
+
+	return ret;
+}
+
+static int ath11k_mac_op_ampdu_action(struct ieee80211_hw *hw,
+				      struct ieee80211_vif *vif,
+				      struct ieee80211_ampdu_params *params)
+{
+	struct ath11k *ar = hw->priv;
+	int ret = -EINVAL;
+
+	mutex_lock(&ar->conf_mutex);
+
+	switch (params->action) {
+	case IEEE80211_AMPDU_RX_START:
+		ret = ath11k_dp_rx_ampdu_start(ar, params);
+		break;
+	case IEEE80211_AMPDU_RX_STOP:
+		ret = ath11k_dp_rx_ampdu_stop(ar, params);
+		break;
+	case IEEE80211_AMPDU_TX_START:
+	case IEEE80211_AMPDU_TX_STOP_CONT:
+	case IEEE80211_AMPDU_TX_STOP_FLUSH:
+	case IEEE80211_AMPDU_TX_STOP_FLUSH_CONT:
+	case IEEE80211_AMPDU_TX_OPERATIONAL:
+		/* Tx A-MPDU aggregation offloaded to hw/fw so deny mac80211
+		 * Tx aggregation requests.
+		 */
+		ret = -EOPNOTSUPP;
+		break;
+	}
+
+	mutex_unlock(&ar->conf_mutex);
+
+	return ret;
+}
+
+static int ath11k_mac_op_add_chanctx(struct ieee80211_hw *hw,
+				     struct ieee80211_chanctx_conf *ctx)
+{
+	struct ath11k *ar = hw->priv;
+	struct ath11k_base *ab = ar->ab;
+
+	ath11k_dbg(ab, ATH11K_DBG_MAC,
+		   "mac chanctx add freq %hu width %d ptr %pK\n",
+		   ctx->def.chan->center_freq, ctx->def.width, ctx);
+
+	mutex_lock(&ar->conf_mutex);
+
+	spin_lock_bh(&ar->data_lock);
+	/* TODO: In case of multiple channel context, populate rx_channel from
+	 * Rx PPDU desc information.
+	 */
+	ar->rx_channel = ctx->def.chan;
+	spin_unlock_bh(&ar->data_lock);
+
+	mutex_unlock(&ar->conf_mutex);
+
+	return 0;
+}
+
+static void ath11k_mac_op_remove_chanctx(struct ieee80211_hw *hw,
+					 struct ieee80211_chanctx_conf *ctx)
+{
+	struct ath11k *ar = hw->priv;
+	struct ath11k_base *ab = ar->ab;
+
+	ath11k_dbg(ab, ATH11K_DBG_MAC,
+		   "mac chanctx remove freq %hu width %d ptr %pK\n",
+		   ctx->def.chan->center_freq, ctx->def.width, ctx);
+
+	mutex_lock(&ar->conf_mutex);
+
+	spin_lock_bh(&ar->data_lock);
+	/* TODO: In case of there is one more channel context left, populate
+	 * rx_channel with the channel of that remaining channel context.
+	 */
+	ar->rx_channel = NULL;
+	spin_unlock_bh(&ar->data_lock);
+
+	mutex_unlock(&ar->conf_mutex);
+}
+
+static inline int ath11k_mac_vdev_setup_sync(struct ath11k *ar)
+{
+	lockdep_assert_held(&ar->conf_mutex);
+
+	if (test_bit(ATH11K_FLAG_CRASH_FLUSH, &ar->ab->dev_flags))
+		return -ESHUTDOWN;
+
+	if (!wait_for_completion_timeout(&ar->vdev_setup_done,
+					 ATH11K_VDEV_SETUP_TIMEOUT_HZ))
+		return -ETIMEDOUT;
+
+	return ar->last_wmi_vdev_start_status ? -EINVAL : 0;
+}
+
+static int
+ath11k_mac_vdev_start_restart(struct ath11k_vif *arvif,
+			      const struct cfg80211_chan_def *chandef,
+			      bool restart)
+{
+	struct ath11k *ar = arvif->ar;
+	struct ath11k_base *ab = ar->ab;
+	struct wmi_vdev_start_req_arg arg = {};
+	int he_support = arvif->vif->bss_conf.he_support;
+	int ret = 0;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	reinit_completion(&ar->vdev_setup_done);
+
+	arg.vdev_id = arvif->vdev_id;
+	arg.dtim_period = arvif->dtim_period;
+	arg.bcn_intval = arvif->beacon_interval;
+
+	arg.channel.freq = chandef->chan->center_freq;
+	arg.channel.band_center_freq1 = chandef->center_freq1;
+	arg.channel.band_center_freq2 = chandef->center_freq2;
+	arg.channel.mode =
+		ath11k_phymodes[chandef->chan->band][chandef->width];
+
+	arg.channel.min_power = 0;
+	arg.channel.max_power = chandef->chan->max_power * 2;
+	arg.channel.max_reg_power = chandef->chan->max_reg_power * 2;
+	arg.channel.max_antenna_gain = chandef->chan->max_antenna_gain * 2;
+
+	arg.pref_tx_streams = ar->num_tx_chains;
+	arg.pref_rx_streams = ar->num_rx_chains;
+
+	if (arvif->vdev_type == WMI_VDEV_TYPE_AP) {
+		arg.ssid = arvif->u.ap.ssid;
+		arg.ssid_len = arvif->u.ap.ssid_len;
+		arg.hidden_ssid = arvif->u.ap.hidden_ssid;
+
+		/* For now allow DFS for AP mode */
+		arg.channel.chan_radar =
+			!!(chandef->chan->flags & IEEE80211_CHAN_RADAR);
+
+		arg.channel.passive = arg.channel.chan_radar;
+
+		spin_lock_bh(&ab->base_lock);
+		arg.regdomain = ar->ab->dfs_region;
+		spin_unlock_bh(&ab->base_lock);
+
+		/* TODO: Notify if secondary 80Mhz also needs radar detection */
+		if (he_support) {
+			ret = ath11k_set_he_mu_sounding_mode(ar, arvif);
+			if (ret) {
+				ath11k_warn(ar->ab, "failed to set he mode vdev %i\n",
+					    arg.vdev_id);
+				return ret;
+			}
+		}
+	}
+
+	arg.channel.passive |= !!(chandef->chan->flags & IEEE80211_CHAN_NO_IR);
+
+	ath11k_dbg(ab, ATH11K_DBG_MAC,
+		   "mac vdev %d start center_freq %d phymode %s\n",
+		   arg.vdev_id, arg.channel.freq,
+		   ath11k_wmi_phymode_str(arg.channel.mode));
+
+	ret = ath11k_wmi_vdev_start(ar, &arg, restart);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to %s WMI vdev %i\n",
+			    restart ? "restart" : "start", arg.vdev_id);
+		return ret;
+	}
+
+	ret = ath11k_mac_vdev_setup_sync(ar);
+	if (ret) {
+		ath11k_warn(ab, "failed to synchronize setup for vdev %i %s: %d\n",
+			    arg.vdev_id, restart ? "restart" : "start", ret);
+		return ret;
+	}
+
+	ar->num_started_vdevs++;
+
+	/* Enable CAC Flag in the driver by checking the channel DFS cac time,
+	 * i.e dfs_cac_ms value which will be valid only for radar channels
+	 * and state as NL80211_DFS_USABLE which indicates CAC needs to be
+	 * done before channel usage. This flags is used to drop rx packets.
+	 * during CAC.
+	 */
+	/* TODO Set the flag for other interface types as required */
+	if (arvif->vdev_type == WMI_VDEV_TYPE_AP &&
+	    chandef->chan->dfs_cac_ms &&
+	    chandef->chan->dfs_state == NL80211_DFS_USABLE) {
+		set_bit(ATH11K_CAC_RUNNING, &ar->dev_flags);
+		ath11k_dbg(ab, ATH11K_DBG_MAC,
+			   "CAC Started in chan_freq %d for vdev %d\n",
+			   arg.channel.freq, arg.vdev_id);
+	}
+
+	return 0;
+}
+
+static int ath11k_mac_vdev_stop(struct ath11k_vif *arvif)
+{
+	struct ath11k *ar = arvif->ar;
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	reinit_completion(&ar->vdev_setup_done);
+
+	spin_lock_bh(&ar->data_lock);
+
+	ar->vdev_stop_status.stop_in_progress = true;
+	ar->vdev_stop_status.vdev_id = arvif->vdev_id;
+
+	spin_unlock_bh(&ar->data_lock);
+
+	ret = ath11k_wmi_vdev_stop(ar, arvif->vdev_id);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to stop WMI vdev %i: %d\n",
+			    arvif->vdev_id, ret);
+		goto err;
+	}
+
+	ret = ath11k_mac_vdev_setup_sync(ar);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to synchronize setup for vdev %i: %d\n",
+			    arvif->vdev_id, ret);
+		goto err;
+	}
+
+	WARN_ON(ar->num_started_vdevs == 0);
+
+	ar->num_started_vdevs--;
+
+	if (test_bit(ATH11K_CAC_RUNNING, &ar->dev_flags)) {
+		clear_bit(ATH11K_CAC_RUNNING, &ar->dev_flags);
+		ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "CAC Stopped for vdev %d\n",
+			   arvif->vdev_id);
+	}
+
+	return 0;
+err:
+	spin_lock_bh(&ar->data_lock);
+	ar->vdev_stop_status.stop_in_progress = false;
+	spin_unlock_bh(&ar->data_lock);
+
+	return ret;
+}
+
+static int ath11k_mac_vdev_start(struct ath11k_vif *arvif,
+				 const struct cfg80211_chan_def *chandef)
+{
+	return ath11k_mac_vdev_start_restart(arvif, chandef, false);
+}
+
+static int ath11k_mac_vdev_restart(struct ath11k_vif *arvif,
+				   const struct cfg80211_chan_def *chandef)
+{
+	return ath11k_mac_vdev_start_restart(arvif, chandef, true);
+}
+
+struct ath11k_mac_change_chanctx_arg {
+	struct ieee80211_chanctx_conf *ctx;
+	struct ieee80211_vif_chanctx_switch *vifs;
+	int n_vifs;
+	int next_vif;
+};
+
+static void
+ath11k_mac_change_chanctx_cnt_iter(void *data, u8 *mac,
+				   struct ieee80211_vif *vif)
+{
+	struct ath11k_mac_change_chanctx_arg *arg = data;
+
+	if (rcu_access_pointer(vif->chanctx_conf) != arg->ctx)
+		return;
+
+	arg->n_vifs++;
+}
+
+static void
+ath11k_mac_change_chanctx_fill_iter(void *data, u8 *mac,
+				    struct ieee80211_vif *vif)
+{
+	struct ath11k_mac_change_chanctx_arg *arg = data;
+	struct ieee80211_chanctx_conf *ctx;
+
+	ctx = rcu_access_pointer(vif->chanctx_conf);
+	if (ctx != arg->ctx)
+		return;
+
+	if (WARN_ON(arg->next_vif == arg->n_vifs))
+		return;
+
+	arg->vifs[arg->next_vif].vif = vif;
+	arg->vifs[arg->next_vif].old_ctx = ctx;
+	arg->vifs[arg->next_vif].new_ctx = ctx;
+	arg->next_vif++;
+}
+
+static void
+ath11k_mac_update_vif_chan(struct ath11k *ar,
+			   struct ieee80211_vif_chanctx_switch *vifs,
+			   int n_vifs)
+{
+	struct ath11k_base *ab = ar->ab;
+	struct ath11k_vif *arvif;
+	int ret;
+	int i;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	for (i = 0; i < n_vifs; i++) {
+		arvif = (void *)vifs[i].vif->drv_priv;
+
+		ath11k_dbg(ab, ATH11K_DBG_MAC,
+			   "mac chanctx switch vdev_id %i freq %hu->%hu width %d->%d\n",
+			   arvif->vdev_id,
+			   vifs[i].old_ctx->def.chan->center_freq,
+			   vifs[i].new_ctx->def.chan->center_freq,
+			   vifs[i].old_ctx->def.width,
+			   vifs[i].new_ctx->def.width);
+
+		if (WARN_ON(!arvif->is_started))
+			continue;
+
+		if (WARN_ON(!arvif->is_up))
+			continue;
+
+		ret = ath11k_wmi_vdev_down(ar, arvif->vdev_id);
+		if (ret) {
+			ath11k_warn(ab, "failed to down vdev %d: %d\n",
+				    arvif->vdev_id, ret);
+			continue;
+		}
+	}
+
+	/* All relevant vdevs are downed and associated channel resources
+	 * should be available for the channel switch now.
+	 */
+
+	/* TODO: Update ar->rx_channel */
+
+	for (i = 0; i < n_vifs; i++) {
+		arvif = (void *)vifs[i].vif->drv_priv;
+
+		if (WARN_ON(!arvif->is_started))
+			continue;
+
+		if (WARN_ON(!arvif->is_up))
+			continue;
+
+		ret = ath11k_mac_setup_bcn_tmpl(arvif);
+		if (ret)
+			ath11k_warn(ab, "failed to update bcn tmpl during csa: %d\n",
+				    ret);
+
+		ret = ath11k_mac_vdev_restart(arvif, &vifs[i].new_ctx->def);
+		if (ret) {
+			ath11k_warn(ab, "failed to restart vdev %d: %d\n",
+				    arvif->vdev_id, ret);
+			continue;
+		}
+
+		ret = ath11k_wmi_vdev_up(arvif->ar, arvif->vdev_id, arvif->aid,
+					 arvif->bssid);
+		if (ret) {
+			ath11k_warn(ab, "failed to bring vdev up %d: %d\n",
+				    arvif->vdev_id, ret);
+			continue;
+		}
+	}
+}
+
+static void
+ath11k_mac_update_active_vif_chan(struct ath11k *ar,
+				  struct ieee80211_chanctx_conf *ctx)
+{
+	struct ath11k_mac_change_chanctx_arg arg = { .ctx = ctx };
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	ieee80211_iterate_active_interfaces_atomic(ar->hw,
+						   IEEE80211_IFACE_ITER_NORMAL,
+						   ath11k_mac_change_chanctx_cnt_iter,
+						   &arg);
+	if (arg.n_vifs == 0)
+		return;
+
+	arg.vifs = kcalloc(arg.n_vifs, sizeof(arg.vifs[0]), GFP_KERNEL);
+	if (!arg.vifs)
+		return;
+
+	ieee80211_iterate_active_interfaces_atomic(ar->hw,
+						   IEEE80211_IFACE_ITER_NORMAL,
+						   ath11k_mac_change_chanctx_fill_iter,
+						   &arg);
+
+	ath11k_mac_update_vif_chan(ar, arg.vifs, arg.n_vifs);
+
+	kfree(arg.vifs);
+}
+
+static void ath11k_mac_op_change_chanctx(struct ieee80211_hw *hw,
+					 struct ieee80211_chanctx_conf *ctx,
+					 u32 changed)
+{
+	struct ath11k *ar = hw->priv;
+	struct ath11k_base *ab = ar->ab;
+
+	mutex_lock(&ar->conf_mutex);
+
+	ath11k_dbg(ab, ATH11K_DBG_MAC,
+		   "mac chanctx change freq %hu width %d ptr %pK changed %x\n",
+		   ctx->def.chan->center_freq, ctx->def.width, ctx, changed);
+
+	/* This shouldn't really happen because channel switching should use
+	 * switch_vif_chanctx().
+	 */
+	if (WARN_ON(changed & IEEE80211_CHANCTX_CHANGE_CHANNEL))
+		goto unlock;
+
+	if (changed & IEEE80211_CHANCTX_CHANGE_WIDTH)
+		ath11k_mac_update_active_vif_chan(ar, ctx);
+
+	/* TODO: Recalc radar detection */
+
+unlock:
+	mutex_unlock(&ar->conf_mutex);
+}
+
+static int
+ath11k_mac_op_assign_vif_chanctx(struct ieee80211_hw *hw,
+				 struct ieee80211_vif *vif,
+				 struct ieee80211_chanctx_conf *ctx)
+{
+	struct ath11k *ar = hw->priv;
+	struct ath11k_base *ab = ar->ab;
+	struct ath11k_vif *arvif = (void *)vif->drv_priv;
+	int ret;
+
+	mutex_lock(&ar->conf_mutex);
+
+	ath11k_dbg(ab, ATH11K_DBG_MAC,
+		   "mac chanctx assign ptr %pK vdev_id %i\n",
+		   ctx, arvif->vdev_id);
+
+	if (WARN_ON(arvif->is_started)) {
+		mutex_unlock(&ar->conf_mutex);
+		return -EBUSY;
+	}
+
+	ret = ath11k_mac_vdev_start(arvif, &ctx->def);
+	if (ret) {
+		ath11k_warn(ab, "failed to start vdev %i addr %pM on freq %d: %d\n",
+			    arvif->vdev_id, vif->addr,
+			    ctx->def.chan->center_freq, ret);
+		goto err;
+	}
+	if (arvif->vdev_type == WMI_VDEV_TYPE_MONITOR) {
+		ret = ath11k_monitor_vdev_up(ar, arvif->vdev_id);
+		if (ret)
+			goto err;
+	}
+
+	arvif->is_started = true;
+
+	/* TODO: Setup ps and cts/rts protection */
+
+	mutex_unlock(&ar->conf_mutex);
+
+	return 0;
+
+err:
+	mutex_unlock(&ar->conf_mutex);
+
+	return ret;
+}
+
+static void
+ath11k_mac_op_unassign_vif_chanctx(struct ieee80211_hw *hw,
+				   struct ieee80211_vif *vif,
+				   struct ieee80211_chanctx_conf *ctx)
+{
+	struct ath11k *ar = hw->priv;
+	struct ath11k_base *ab = ar->ab;
+	struct ath11k_vif *arvif = (void *)vif->drv_priv;
+	int ret;
+
+	mutex_lock(&ar->conf_mutex);
+
+	ath11k_dbg(ab, ATH11K_DBG_MAC,
+		   "mac chanctx unassign ptr %pK vdev_id %i\n",
+		   ctx, arvif->vdev_id);
+
+	WARN_ON(!arvif->is_started);
+
+	ret = ath11k_mac_vdev_stop(arvif);
+	if (ret)
+		ath11k_warn(ab, "failed to stop vdev %i: %d\n",
+			    arvif->vdev_id, ret);
+
+	arvif->is_started = false;
+
+	mutex_unlock(&ar->conf_mutex);
+}
+
+static int
+ath11k_mac_op_switch_vif_chanctx(struct ieee80211_hw *hw,
+				 struct ieee80211_vif_chanctx_switch *vifs,
+				 int n_vifs,
+				 enum ieee80211_chanctx_switch_mode mode)
+{
+	struct ath11k *ar = hw->priv;
+
+	mutex_lock(&ar->conf_mutex);
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_MAC,
+		   "mac chanctx switch n_vifs %d mode %d\n",
+		   n_vifs, mode);
+	ath11k_mac_update_vif_chan(ar, vifs, n_vifs);
+
+	mutex_unlock(&ar->conf_mutex);
+
+	return 0;
+}
+
+static int
+ath11k_set_vdev_param_to_all_vifs(struct ath11k *ar, int param, u32 value)
+{
+	struct ath11k_vif *arvif;
+	int ret = 0;
+
+	mutex_lock(&ar->conf_mutex);
+	list_for_each_entry(arvif, &ar->arvifs, list) {
+		ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "setting mac vdev %d param %d value %d\n",
+			   param, arvif->vdev_id, value);
+
+		ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id,
+						    param, value);
+		if (ret) {
+			ath11k_warn(ar->ab, "failed to set param %d for vdev %d: %d\n",
+				    param, arvif->vdev_id, ret);
+			break;
+		}
+	}
+	mutex_unlock(&ar->conf_mutex);
+	return ret;
+}
+
+/* mac80211 stores device specific RTS/Fragmentation threshold value,
+ * this is set interface specific to firmware from ath11k driver
+ */
+static int ath11k_mac_op_set_rts_threshold(struct ieee80211_hw *hw, u32 value)
+{
+	struct ath11k *ar = hw->priv;
+	int param_id = WMI_VDEV_PARAM_RTS_THRESHOLD;
+
+	return ath11k_set_vdev_param_to_all_vifs(ar, param_id, value);
+}
+
+static int ath11k_mac_op_set_frag_threshold(struct ieee80211_hw *hw, u32 value)
+{
+	/* Even though there's a WMI vdev param for fragmentation threshold no
+	 * known firmware actually implements it. Moreover it is not possible to
+	 * rely frame fragmentation to mac80211 because firmware clears the
+	 * "more fragments" bit in frame control making it impossible for remote
+	 * devices to reassemble frames.
+	 *
+	 * Hence implement a dummy callback just to say fragmentation isn't
+	 * supported. This effectively prevents mac80211 from doing frame
+	 * fragmentation in software.
+	 */
+	return -EOPNOTSUPP;
+}
+
+static void ath11k_mac_op_flush(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+				u32 queues, bool drop)
+{
+	struct ath11k *ar = hw->priv;
+	long time_left;
+
+	if (drop)
+		return;
+
+	time_left = wait_event_timeout(ar->dp.tx_empty_waitq,
+				       (atomic_read(&ar->dp.num_tx_pending) == 0),
+				       ATH11K_FLUSH_TIMEOUT);
+	if (time_left == 0)
+		ath11k_warn(ar->ab, "failed to flush transmit queue %ld\n", time_left);
+}
+
+static int
+ath11k_mac_bitrate_mask_num_ht_rates(struct ath11k *ar,
+				     enum nl80211_band band,
+				     const struct cfg80211_bitrate_mask *mask)
+{
+	int num_rates = 0;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(mask->control[band].ht_mcs); i++)
+		num_rates += hweight16(mask->control[band].ht_mcs[i]);
+
+	return num_rates;
+}
+
+static bool
+ath11k_mac_has_single_legacy_rate(struct ath11k *ar,
+				  enum nl80211_band band,
+				  const struct cfg80211_bitrate_mask *mask)
+{
+	int num_rates = 0;
+
+	num_rates = hweight32(mask->control[band].legacy);
+
+	if (ath11k_mac_bitrate_mask_num_ht_rates(ar, band, mask))
+		return false;
+
+	if (ath11k_mac_bitrate_mask_num_vht_rates(ar, band, mask))
+		return false;
+
+	return num_rates == 1;
+}
+
+static bool
+ath11k_mac_bitrate_mask_get_single_nss(struct ath11k *ar,
+				       enum nl80211_band band,
+				       const struct cfg80211_bitrate_mask *mask,
+				       int *nss)
+{
+	struct ieee80211_supported_band *sband = &ar->mac.sbands[band];
+	u16 vht_mcs_map = le16_to_cpu(sband->vht_cap.vht_mcs.tx_mcs_map);
+	u8 ht_nss_mask = 0;
+	u8 vht_nss_mask = 0;
+	int i;
+
+	/* No need to consider legacy here. Basic rates are always present
+	 * in bitrate mask
+	 */
+
+	for (i = 0; i < ARRAY_SIZE(mask->control[band].ht_mcs); i++) {
+		if (mask->control[band].ht_mcs[i] == 0)
+			continue;
+		else if (mask->control[band].ht_mcs[i] ==
+			 sband->ht_cap.mcs.rx_mask[i])
+			ht_nss_mask |= BIT(i);
+		else
+			return false;
+	}
+
+	for (i = 0; i < ARRAY_SIZE(mask->control[band].vht_mcs); i++) {
+		if (mask->control[band].vht_mcs[i] == 0)
+			continue;
+		else if (mask->control[band].vht_mcs[i] ==
+			 ath11k_mac_get_max_vht_mcs_map(vht_mcs_map, i))
+			vht_nss_mask |= BIT(i);
+		else
+			return false;
+	}
+
+	if (ht_nss_mask != vht_nss_mask)
+		return false;
+
+	if (ht_nss_mask == 0)
+		return false;
+
+	if (BIT(fls(ht_nss_mask)) - 1 != ht_nss_mask)
+		return false;
+
+	*nss = fls(ht_nss_mask);
+
+	return true;
+}
+
+static int
+ath11k_mac_get_single_legacy_rate(struct ath11k *ar,
+				  enum nl80211_band band,
+				  const struct cfg80211_bitrate_mask *mask,
+				  u32 *rate, u8 *nss)
+{
+	int rate_idx;
+	u16 bitrate;
+	u8 preamble;
+	u8 hw_rate;
+
+	if (hweight32(mask->control[band].legacy) != 1)
+		return -EINVAL;
+
+	rate_idx = ffs(mask->control[band].legacy) - 1;
+
+	if (band == NL80211_BAND_5GHZ)
+		rate_idx += ATH11K_MAC_FIRST_OFDM_RATE_IDX;
+
+	hw_rate = ath11k_legacy_rates[rate_idx].hw_value;
+	bitrate = ath11k_legacy_rates[rate_idx].bitrate;
+
+	if (ath11k_mac_bitrate_is_cck(bitrate))
+		preamble = WMI_RATE_PREAMBLE_CCK;
+	else
+		preamble = WMI_RATE_PREAMBLE_OFDM;
+
+	*nss = 1;
+	*rate = ATH11K_HW_RATE_CODE(hw_rate, 0, preamble);
+
+	return 0;
+}
+
+static int ath11k_mac_set_fixed_rate_params(struct ath11k_vif *arvif,
+					    u32 rate, u8 nss, u8 sgi, u8 ldpc)
+{
+	struct ath11k *ar = arvif->ar;
+	u32 vdev_param;
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "mac set fixed rate params vdev %i rate 0x%02hhx nss %hhu sgi %hhu\n",
+		   arvif->vdev_id, rate, nss, sgi);
+
+	vdev_param = WMI_VDEV_PARAM_FIXED_RATE;
+	ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id,
+					    vdev_param, rate);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to set fixed rate param 0x%02x: %d\n",
+			    rate, ret);
+		return ret;
+	}
+
+	vdev_param = WMI_VDEV_PARAM_NSS;
+	ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id,
+					    vdev_param, nss);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to set nss param %d: %d\n",
+			    nss, ret);
+		return ret;
+	}
+
+	vdev_param = WMI_VDEV_PARAM_SGI;
+	ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id,
+					    vdev_param, sgi);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to set sgi param %d: %d\n",
+			    sgi, ret);
+		return ret;
+	}
+
+	vdev_param = WMI_VDEV_PARAM_LDPC;
+	ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id,
+					    vdev_param, ldpc);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to set ldpc param %d: %d\n",
+			    ldpc, ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static bool
+ath11k_mac_vht_mcs_range_present(struct ath11k *ar,
+				 enum nl80211_band band,
+				 const struct cfg80211_bitrate_mask *mask)
+{
+	int i;
+	u16 vht_mcs;
+
+	for (i = 0; i < NL80211_VHT_NSS_MAX; i++) {
+		vht_mcs = mask->control[band].vht_mcs[i];
+
+		switch (vht_mcs) {
+		case 0:
+		case BIT(8) - 1:
+		case BIT(9) - 1:
+		case BIT(10) - 1:
+			break;
+		default:
+			return false;
+		}
+	}
+
+	return true;
+}
+
+static void ath11k_mac_set_bitrate_mask_iter(void *data,
+					     struct ieee80211_sta *sta)
+{
+	struct ath11k_vif *arvif = data;
+	struct ath11k_sta *arsta = (struct ath11k_sta *)sta->drv_priv;
+	struct ath11k *ar = arvif->ar;
+
+	spin_lock_bh(&ar->data_lock);
+	arsta->changed |= IEEE80211_RC_SUPP_RATES_CHANGED;
+	spin_unlock_bh(&ar->data_lock);
+
+	ieee80211_queue_work(ar->hw, &arsta->update_wk);
+}
+
+static void ath11k_mac_disable_peer_fixed_rate(void *data,
+					       struct ieee80211_sta *sta)
+{
+	struct ath11k_vif *arvif = data;
+	struct ath11k *ar = arvif->ar;
+	int ret;
+
+	ret = ath11k_wmi_set_peer_param(ar, sta->addr,
+					arvif->vdev_id,
+					WMI_PEER_PARAM_FIXED_RATE,
+					WMI_FIXED_RATE_NONE);
+	if (ret)
+		ath11k_warn(ar->ab,
+			    "failed to disable peer fixed rate for STA %pM ret %d\n",
+			    sta->addr, ret);
+}
+
+static int
+ath11k_mac_op_set_bitrate_mask(struct ieee80211_hw *hw,
+			       struct ieee80211_vif *vif,
+			       const struct cfg80211_bitrate_mask *mask)
+{
+	struct ath11k_vif *arvif = (void *)vif->drv_priv;
+	struct cfg80211_chan_def def;
+	struct ath11k *ar = arvif->ar;
+	enum nl80211_band band;
+	const u8 *ht_mcs_mask;
+	const u16 *vht_mcs_mask;
+	u32 rate;
+	u8 nss;
+	u8 sgi;
+	u8 ldpc;
+	int single_nss;
+	int ret;
+	int num_rates;
+
+	if (ath11k_mac_vif_chan(vif, &def))
+		return -EPERM;
+
+	band = def.chan->band;
+	ht_mcs_mask = mask->control[band].ht_mcs;
+	vht_mcs_mask = mask->control[band].vht_mcs;
+	ldpc = !!(ar->ht_cap_info & WMI_HT_CAP_LDPC);
+
+	sgi = mask->control[band].gi;
+	if (sgi == NL80211_TXRATE_FORCE_LGI)
+		return -EINVAL;
+
+	/* mac80211 doesn't support sending a fixed HT/VHT MCS alone, rather it
+	 * requires passing atleast one of used basic rates along with them.
+	 * Fixed rate setting across different preambles(legacy, HT, VHT) is
+	 * not supported by the FW. Hence use of FIXED_RATE vdev param is not
+	 * suitable for setting single HT/VHT rates.
+	 * But, there could be a single basic rate passed from userspace which
+	 * can be done through the FIXED_RATE param.
+	 */
+	if (ath11k_mac_has_single_legacy_rate(ar, band, mask)) {
+		ret = ath11k_mac_get_single_legacy_rate(ar, band, mask, &rate,
+							&nss);
+		if (ret) {
+			ath11k_warn(ar->ab, "failed to get single legacy rate for vdev %i: %d\n",
+				    arvif->vdev_id, ret);
+			return ret;
+		}
+		ieee80211_iterate_stations_atomic(ar->hw,
+						  ath11k_mac_disable_peer_fixed_rate,
+						  arvif);
+	} else if (ath11k_mac_bitrate_mask_get_single_nss(ar, band, mask,
+							  &single_nss)) {
+		rate = WMI_FIXED_RATE_NONE;
+		nss = single_nss;
+	} else {
+		rate = WMI_FIXED_RATE_NONE;
+		nss = min_t(u32, ar->num_tx_chains,
+			    max(ath11k_mac_max_ht_nss(ht_mcs_mask),
+				ath11k_mac_max_vht_nss(vht_mcs_mask)));
+
+		/* If multiple rates across different preambles are given
+		 * we can reconfigure this info with all peers using PEER_ASSOC
+		 * command with the below exception cases.
+		 * - Single VHT Rate : peer_assoc command accommodates only MCS
+		 * range values i.e 0-7, 0-8, 0-9 for VHT. Though mac80211
+		 * mandates passing basic rates along with HT/VHT rates, FW
+		 * doesn't allow switching from VHT to Legacy. Hence instead of
+		 * setting legacy and VHT rates using RATEMASK_CMD vdev cmd,
+		 * we could set this VHT rate as peer fixed rate param, which
+		 * will override FIXED rate and FW rate control algorithm.
+		 * If single VHT rate is passed along with HT rates, we select
+		 * the VHT rate as fixed rate for vht peers.
+		 * - Multiple VHT Rates : When Multiple VHT rates are given,this
+		 * can be set using RATEMASK CMD which uses FW rate-ctl alg.
+		 * TODO: Setting multiple VHT MCS and replacing peer_assoc with
+		 * RATEMASK_CMDID can cover all use cases of setting rates
+		 * across multiple preambles and rates within same type.
+		 * But requires more validation of the command at this point.
+		 */
+
+		num_rates = ath11k_mac_bitrate_mask_num_vht_rates(ar, band,
+								  mask);
+
+		if (!ath11k_mac_vht_mcs_range_present(ar, band, mask) &&
+		    num_rates > 1) {
+			/* TODO: Handle multiple VHT MCS values setting using
+			 * RATEMASK CMD
+			 */
+			ath11k_warn(ar->ab,
+				    "Setting more than one MCS Value in bitrate mask not supported\n");
+			return -EINVAL;
+		}
+
+		ieee80211_iterate_stations_atomic(ar->hw,
+						  ath11k_mac_disable_peer_fixed_rate,
+						  arvif);
+
+		mutex_lock(&ar->conf_mutex);
+
+		arvif->bitrate_mask = *mask;
+		ieee80211_iterate_stations_atomic(ar->hw,
+						  ath11k_mac_set_bitrate_mask_iter,
+						  arvif);
+
+		mutex_unlock(&ar->conf_mutex);
+	}
+
+	mutex_lock(&ar->conf_mutex);
+
+	ret = ath11k_mac_set_fixed_rate_params(arvif, rate, nss, sgi, ldpc);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to set fixed rate params on vdev %i: %d\n",
+			    arvif->vdev_id, ret);
+	}
+
+	mutex_unlock(&ar->conf_mutex);
+
+	return ret;
+}
+
+static void
+ath11k_mac_op_reconfig_complete(struct ieee80211_hw *hw,
+				enum ieee80211_reconfig_type reconfig_type)
+{
+	struct ath11k *ar = hw->priv;
+
+	if (reconfig_type != IEEE80211_RECONFIG_TYPE_RESTART)
+		return;
+
+	mutex_lock(&ar->conf_mutex);
+
+	if (ar->state == ATH11K_STATE_RESTARTED) {
+		ath11k_warn(ar->ab, "pdev %d successfully recovered\n",
+			    ar->pdev->pdev_id);
+		ar->state = ATH11K_STATE_ON;
+		ieee80211_wake_queues(ar->hw);
+	}
+
+	mutex_unlock(&ar->conf_mutex);
+}
+
+static void
+ath11k_mac_update_bss_chan_survey(struct ath11k *ar,
+				  struct ieee80211_channel *channel)
+{
+	int ret;
+	enum wmi_bss_chan_info_req_type type = WMI_BSS_SURVEY_REQ_TYPE_READ;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	if (!test_bit(WMI_TLV_SERVICE_BSS_CHANNEL_INFO_64, ar->ab->wmi_sc.svc_map) ||
+	    ar->rx_channel != channel)
+		return;
+
+	if (ar->scan.state != ATH11K_SCAN_IDLE) {
+		ath11k_dbg(ar->ab, ATH11K_DBG_MAC,
+			   "ignoring bss chan info req while scanning..\n");
+		return;
+	}
+
+	reinit_completion(&ar->bss_survey_done);
+
+	ret = ath11k_wmi_pdev_bss_chan_info_request(ar, type);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to send pdev bss chan info request\n");
+		return;
+	}
+
+	ret = wait_for_completion_timeout(&ar->bss_survey_done, 3 * HZ);
+	if (ret == 0)
+		ath11k_warn(ar->ab, "bss channel survey timed out\n");
+}
+
+static int ath11k_mac_op_get_survey(struct ieee80211_hw *hw, int idx,
+				    struct survey_info *survey)
+{
+	struct ath11k *ar = hw->priv;
+	struct ieee80211_supported_band *sband;
+	struct survey_info *ar_survey;
+	int ret = 0;
+
+	if (idx >= ATH11K_NUM_CHANS)
+		return -ENOENT;
+
+	ar_survey = &ar->survey[idx];
+
+	mutex_lock(&ar->conf_mutex);
+
+	sband = hw->wiphy->bands[NL80211_BAND_2GHZ];
+	if (sband && idx >= sband->n_channels) {
+		idx -= sband->n_channels;
+		sband = NULL;
+	}
+
+	if (!sband)
+		sband = hw->wiphy->bands[NL80211_BAND_5GHZ];
+
+	if (!sband || idx >= sband->n_channels) {
+		ret = -ENOENT;
+		goto exit;
+	}
+
+	ath11k_mac_update_bss_chan_survey(ar, &sband->channels[idx]);
+
+	spin_lock_bh(&ar->data_lock);
+	memcpy(survey, ar_survey, sizeof(*survey));
+	spin_unlock_bh(&ar->data_lock);
+
+	survey->channel = &sband->channels[idx];
+
+	if (ar->rx_channel == survey->channel)
+		survey->filled |= SURVEY_INFO_IN_USE;
+
+exit:
+	mutex_unlock(&ar->conf_mutex);
+	return ret;
+}
+
+static void ath11k_mac_op_sta_statistics(struct ieee80211_hw *hw,
+					 struct ieee80211_vif *vif,
+					 struct ieee80211_sta *sta,
+					 struct station_info *sinfo)
+{
+	struct ath11k_sta *arsta = (struct ath11k_sta *)sta->drv_priv;
+
+	sinfo->rx_duration = arsta->rx_duration;
+	sinfo->filled |= BIT_ULL(NL80211_STA_INFO_RX_DURATION);
+
+	if (!arsta->txrate.legacy && !arsta->txrate.nss)
+		return;
+
+	if (arsta->txrate.legacy) {
+		sinfo->txrate.legacy = arsta->txrate.legacy;
+	} else {
+		sinfo->txrate.mcs = arsta->txrate.mcs;
+		sinfo->txrate.nss = arsta->txrate.nss;
+		sinfo->txrate.bw = arsta->txrate.bw;
+		sinfo->txrate.he_gi = arsta->txrate.he_gi;
+		sinfo->txrate.he_dcm = arsta->txrate.he_dcm;
+		sinfo->txrate.he_ru_alloc = arsta->txrate.he_ru_alloc;
+	}
+	sinfo->txrate.flags = arsta->txrate.flags;
+	sinfo->filled |= BIT_ULL(NL80211_STA_INFO_TX_BITRATE);
+
+	/* TODO: Use real NF instead of default one. */
+	sinfo->signal = arsta->rssi_comb + ATH11K_DEFAULT_NOISE_FLOOR;
+}
+
+static const struct ieee80211_ops ath11k_ops = {
+	.tx				= ath11k_mac_op_tx,
+	.start                          = ath11k_mac_op_start,
+	.stop                           = ath11k_mac_op_stop,
+	.reconfig_complete              = ath11k_mac_op_reconfig_complete,
+	.add_interface                  = ath11k_mac_op_add_interface,
+	.remove_interface		= ath11k_mac_op_remove_interface,
+	.config                         = ath11k_mac_op_config,
+	.bss_info_changed               = ath11k_mac_op_bss_info_changed,
+	.configure_filter		= ath11k_mac_op_configure_filter,
+	.hw_scan                        = ath11k_mac_op_hw_scan,
+	.cancel_hw_scan                 = ath11k_mac_op_cancel_hw_scan,
+	.set_key                        = ath11k_mac_op_set_key,
+	.sta_state                      = ath11k_mac_op_sta_state,
+	.sta_rc_update			= ath11k_mac_op_sta_rc_update,
+	.conf_tx                        = ath11k_mac_op_conf_tx,
+	.set_antenna			= ath11k_mac_op_set_antenna,
+	.get_antenna			= ath11k_mac_op_get_antenna,
+	.ampdu_action			= ath11k_mac_op_ampdu_action,
+	.add_chanctx			= ath11k_mac_op_add_chanctx,
+	.remove_chanctx			= ath11k_mac_op_remove_chanctx,
+	.change_chanctx			= ath11k_mac_op_change_chanctx,
+	.assign_vif_chanctx		= ath11k_mac_op_assign_vif_chanctx,
+	.unassign_vif_chanctx		= ath11k_mac_op_unassign_vif_chanctx,
+	.switch_vif_chanctx		= ath11k_mac_op_switch_vif_chanctx,
+	.set_rts_threshold		= ath11k_mac_op_set_rts_threshold,
+	.set_frag_threshold		= ath11k_mac_op_set_frag_threshold,
+	.set_bitrate_mask		= ath11k_mac_op_set_bitrate_mask,
+	.get_survey			= ath11k_mac_op_get_survey,
+	.flush				= ath11k_mac_op_flush,
+	.sta_statistics			= ath11k_mac_op_sta_statistics,
+	CFG80211_TESTMODE_CMD(ath11k_tm_cmd)
+#ifdef CONFIG_MAC80211_DEBUGFS
+	.sta_add_debugfs		= ath11k_sta_add_debugfs,
+#endif
+};
+
+static const struct ieee80211_iface_limit ath11k_if_limits[] = {
+	{
+		.max = 1,
+		.types = BIT(NL80211_IFTYPE_STATION),
+	},
+	{
+		.max    = 16,
+		.types  = BIT(NL80211_IFTYPE_AP)
+#ifdef CONFIG_MAC80211_MESH
+			| BIT(NL80211_IFTYPE_MESH_POINT)
+#endif
+	},
+};
+
+static const struct ieee80211_iface_combination ath11k_if_comb[] = {
+	{
+		.limits = ath11k_if_limits,
+		.n_limits = ARRAY_SIZE(ath11k_if_limits),
+		.max_interfaces = 16,
+		.num_different_channels = 1,
+		.beacon_int_infra_match = true,
+		.beacon_int_min_gcd = 100,
+		.radar_detect_widths =	BIT(NL80211_CHAN_WIDTH_20_NOHT) |
+					BIT(NL80211_CHAN_WIDTH_20) |
+					BIT(NL80211_CHAN_WIDTH_40) |
+					BIT(NL80211_CHAN_WIDTH_80),
+	},
+};
+
+static void ath11k_mac_update_ch_list(struct ath11k *ar,
+				      struct ieee80211_supported_band *band,
+				      u32 freq_low, u32 freq_high)
+{
+	int i;
+
+	if (!(freq_low && freq_high))
+		return;
+
+	for (i = 0; i < band->n_channels; i++) {
+		if (band->channels[i].center_freq < freq_low ||
+		    band->channels[i].center_freq > freq_high)
+			band->channels[i].flags |= IEEE80211_CHAN_DISABLED;
+	}
+}
+
+static int ath11k_mac_setup_channels_rates(struct ath11k *ar,
+					   u32 supported_bands)
+{
+	struct ieee80211_supported_band *band;
+	struct ath11k_hal_reg_capabilities_ext *reg_cap;
+	void *channels;
+
+	BUILD_BUG_ON((ARRAY_SIZE(ath11k_2ghz_channels) +
+		      ARRAY_SIZE(ath11k_5ghz_channels)) !=
+		     ATH11K_NUM_CHANS);
+
+	reg_cap = &ar->ab->hal_reg_cap[ar->pdev_idx];
+
+	if (supported_bands & WMI_HOST_WLAN_2G_CAP) {
+		channels = kmemdup(ath11k_2ghz_channels,
+				   sizeof(ath11k_2ghz_channels),
+				   GFP_KERNEL);
+		if (!channels)
+			return -ENOMEM;
+
+		band = &ar->mac.sbands[NL80211_BAND_2GHZ];
+		band->n_channels = ARRAY_SIZE(ath11k_2ghz_channels);
+		band->channels = channels;
+		band->n_bitrates = ath11k_g_rates_size;
+		band->bitrates = ath11k_g_rates;
+		ar->hw->wiphy->bands[NL80211_BAND_2GHZ] = band;
+		ath11k_mac_update_ch_list(ar, band,
+					  reg_cap->low_2ghz_chan,
+					  reg_cap->high_2ghz_chan);
+	}
+
+	if (supported_bands & WMI_HOST_WLAN_5G_CAP) {
+		channels = kmemdup(ath11k_5ghz_channels,
+				   sizeof(ath11k_5ghz_channels),
+				   GFP_KERNEL);
+		if (!channels) {
+			kfree(ar->mac.sbands[NL80211_BAND_2GHZ].channels);
+			return -ENOMEM;
+		}
+
+		band = &ar->mac.sbands[NL80211_BAND_5GHZ];
+		band->n_channels = ARRAY_SIZE(ath11k_5ghz_channels);
+		band->channels = channels;
+		band->n_bitrates = ath11k_a_rates_size;
+		band->bitrates = ath11k_a_rates;
+		ar->hw->wiphy->bands[NL80211_BAND_5GHZ] = band;
+		ath11k_mac_update_ch_list(ar, band,
+					  reg_cap->low_5ghz_chan,
+					  reg_cap->high_5ghz_chan);
+	}
+
+	return 0;
+}
+
+static const u8 ath11k_if_types_ext_capa[] = {
+	[0] = WLAN_EXT_CAPA1_EXT_CHANNEL_SWITCHING,
+	[7] = WLAN_EXT_CAPA8_OPMODE_NOTIF,
+};
+
+static const u8 ath11k_if_types_ext_capa_sta[] = {
+	[0] = WLAN_EXT_CAPA1_EXT_CHANNEL_SWITCHING,
+	[7] = WLAN_EXT_CAPA8_OPMODE_NOTIF,
+	[9] = WLAN_EXT_CAPA10_TWT_REQUESTER_SUPPORT,
+};
+
+static const u8 ath11k_if_types_ext_capa_ap[] = {
+	[0] = WLAN_EXT_CAPA1_EXT_CHANNEL_SWITCHING,
+	[7] = WLAN_EXT_CAPA8_OPMODE_NOTIF,
+	[9] = WLAN_EXT_CAPA10_TWT_RESPONDER_SUPPORT,
+};
+
+static const struct wiphy_iftype_ext_capab ath11k_iftypes_ext_capa[] = {
+	{
+		.extended_capabilities = ath11k_if_types_ext_capa,
+		.extended_capabilities_mask = ath11k_if_types_ext_capa,
+		.extended_capabilities_len = sizeof(ath11k_if_types_ext_capa),
+	}, {
+		.iftype = NL80211_IFTYPE_STATION,
+		.extended_capabilities = ath11k_if_types_ext_capa_sta,
+		.extended_capabilities_mask = ath11k_if_types_ext_capa_sta,
+		.extended_capabilities_len =
+				sizeof(ath11k_if_types_ext_capa_sta),
+	}, {
+		.iftype = NL80211_IFTYPE_AP,
+		.extended_capabilities = ath11k_if_types_ext_capa_ap,
+		.extended_capabilities_mask = ath11k_if_types_ext_capa_ap,
+		.extended_capabilities_len =
+				sizeof(ath11k_if_types_ext_capa_ap),
+	},
+};
+
+static int ath11k_mac_register(struct ath11k *ar)
+{
+	struct ath11k_base *ab = ar->ab;
+	struct ath11k_pdev_cap *cap = &ar->pdev->cap;
+	static const u32 cipher_suites[] = {
+		WLAN_CIPHER_SUITE_TKIP,
+		WLAN_CIPHER_SUITE_CCMP,
+		WLAN_CIPHER_SUITE_AES_CMAC,
+		WLAN_CIPHER_SUITE_BIP_CMAC_256,
+		WLAN_CIPHER_SUITE_BIP_GMAC_128,
+		WLAN_CIPHER_SUITE_BIP_GMAC_256,
+		WLAN_CIPHER_SUITE_GCMP,
+		WLAN_CIPHER_SUITE_GCMP_256,
+		WLAN_CIPHER_SUITE_CCMP_256,
+	};
+	int ret;
+	u32 ht_cap = 0;
+
+	ath11k_pdev_caps_update(ar);
+
+	SET_IEEE80211_PERM_ADDR(ar->hw, ar->mac_addr);
+
+	SET_IEEE80211_DEV(ar->hw, ab->dev);
+
+	ret = ath11k_mac_setup_channels_rates(ar,
+					      cap->supported_bands);
+	if (ret)
+		goto err_free;
+
+	ath11k_mac_setup_ht_vht_cap(ar, cap, &ht_cap);
+
+	ar->hw->wiphy->available_antennas_rx = cap->rx_chain_mask;
+	ar->hw->wiphy->available_antennas_tx = cap->tx_chain_mask;
+
+	ar->hw->wiphy->interface_modes = BIT(NL80211_IFTYPE_STATION) |
+					 BIT(NL80211_IFTYPE_AP) |
+					 BIT(NL80211_IFTYPE_MESH_POINT);
+
+	ieee80211_hw_set(ar->hw, SIGNAL_DBM);
+	ieee80211_hw_set(ar->hw, SUPPORTS_PS);
+	ieee80211_hw_set(ar->hw, SUPPORTS_DYNAMIC_PS);
+	ieee80211_hw_set(ar->hw, MFP_CAPABLE);
+	ieee80211_hw_set(ar->hw, REPORTS_TX_ACK_STATUS);
+	ieee80211_hw_set(ar->hw, HAS_RATE_CONTROL);
+	ieee80211_hw_set(ar->hw, AP_LINK_PS);
+	ieee80211_hw_set(ar->hw, SPECTRUM_MGMT);
+	ieee80211_hw_set(ar->hw, SUPPORT_FAST_XMIT);
+	ieee80211_hw_set(ar->hw, CONNECTION_MONITOR);
+	ieee80211_hw_set(ar->hw, SUPPORTS_PER_STA_GTK);
+	ieee80211_hw_set(ar->hw, WANT_MONITOR_VIF);
+	ieee80211_hw_set(ar->hw, CHANCTX_STA_CSA);
+	ieee80211_hw_set(ar->hw, QUEUE_CONTROL);
+	ieee80211_hw_set(ar->hw, SUPPORTS_TX_FRAG);
+	ieee80211_hw_set(ar->hw, REPORTS_LOW_ACK);
+	if (ht_cap & WMI_HT_CAP_ENABLED) {
+		ieee80211_hw_set(ar->hw, AMPDU_AGGREGATION);
+		ieee80211_hw_set(ar->hw, TX_AMPDU_SETUP_IN_HW);
+		ieee80211_hw_set(ar->hw, SUPPORTS_REORDERING_BUFFER);
+		ieee80211_hw_set(ar->hw, SUPPORTS_AMSDU_IN_AMPDU);
+	}
+
+	ar->hw->wiphy->features |= NL80211_FEATURE_STATIC_SMPS;
+	ar->hw->wiphy->flags |= WIPHY_FLAG_IBSS_RSN;
+
+	/* TODO: Check if HT capability advertised from firmware is different
+	 * for each band for a dual band capable radio. It will be tricky to
+	 * handle it when the ht capability different for each band.
+	 */
+	if (ht_cap & WMI_HT_CAP_DYNAMIC_SMPS)
+		ar->hw->wiphy->features |= NL80211_FEATURE_DYNAMIC_SMPS;
+
+	ar->hw->wiphy->max_scan_ssids = WLAN_SCAN_PARAMS_MAX_SSID;
+	ar->hw->wiphy->max_scan_ie_len = WLAN_SCAN_PARAMS_MAX_IE_LEN;
+
+	ar->hw->max_listen_interval = ATH11K_MAX_HW_LISTEN_INTERVAL;
+
+	ar->hw->wiphy->flags |= WIPHY_FLAG_HAS_REMAIN_ON_CHANNEL;
+	ar->hw->wiphy->flags |= WIPHY_FLAG_HAS_CHANNEL_SWITCH;
+	ar->hw->wiphy->max_remain_on_channel_duration = 5000;
+
+	ar->hw->wiphy->flags |= WIPHY_FLAG_AP_UAPSD;
+	ar->hw->wiphy->features |= NL80211_FEATURE_AP_MODE_CHAN_WIDTH_CHANGE |
+				   NL80211_FEATURE_AP_SCAN;
+
+	ar->max_num_stations = TARGET_NUM_STATIONS;
+	ar->max_num_peers = TARGET_NUM_PEERS_PDEV;
+
+	ar->hw->wiphy->max_ap_assoc_sta = ar->max_num_stations;
+
+	ar->hw->queues = ATH11K_HW_MAX_QUEUES;
+	ar->hw->offchannel_tx_hw_queue = ATH11K_HW_MAX_QUEUES - 1;
+	ar->hw->max_rx_aggregation_subframes = IEEE80211_MAX_AMPDU_BUF;
+
+	ar->hw->vif_data_size = sizeof(struct ath11k_vif);
+	ar->hw->sta_data_size = sizeof(struct ath11k_sta);
+
+	ar->hw->wiphy->iface_combinations = ath11k_if_comb;
+	ar->hw->wiphy->n_iface_combinations = ARRAY_SIZE(ath11k_if_comb);
+
+	wiphy_ext_feature_set(ar->hw->wiphy, NL80211_EXT_FEATURE_CQM_RSSI_LIST);
+
+	ar->hw->wiphy->cipher_suites = cipher_suites;
+	ar->hw->wiphy->n_cipher_suites = ARRAY_SIZE(cipher_suites);
+
+	ar->hw->wiphy->iftype_ext_capab = ath11k_iftypes_ext_capa;
+	ar->hw->wiphy->num_iftype_ext_capab =
+		ARRAY_SIZE(ath11k_iftypes_ext_capa);
+
+	ath11k_reg_init(ar);
+
+	/* advertise HW checksum offload capabilities */
+	ar->hw->netdev_features = NETIF_F_HW_CSUM;
+
+	ret = ieee80211_register_hw(ar->hw);
+	if (ret) {
+		ath11k_err(ar->ab, "ieee80211 registration failed: %d\n", ret);
+		goto err_free;
+	}
+
+	/* Apply the regd received during initialization */
+	ret = ath11k_regd_update(ar, true);
+	if (ret) {
+		ath11k_err(ar->ab, "ath11k regd update failed: %d\n", ret);
+		goto err_free;
+	}
+
+	ret = ath11k_debug_register(ar);
+	if (ret) {
+		ath11k_err(ar->ab, "debugfs registration failed: %d\n", ret);
+		goto err_free;
+	}
+
+	return 0;
+
+err_free:
+	kfree(ar->mac.sbands[NL80211_BAND_2GHZ].channels);
+	kfree(ar->mac.sbands[NL80211_BAND_5GHZ].channels);
+
+	SET_IEEE80211_DEV(ar->hw, NULL);
+	return ret;
+}
+
+void ath11k_mac_unregister(struct ath11k_base *ab)
+{
+	struct ath11k *ar;
+	struct ath11k_pdev *pdev;
+	int i;
+
+	for (i = 0; i < ab->num_radios; i++) {
+		pdev = &ab->pdevs[i];
+		ar = pdev->ar;
+		if (!ar)
+			continue;
+		cancel_work_sync(&ar->regd_update_work);
+
+		ieee80211_unregister_hw(ar->hw);
+
+		idr_for_each(&ar->txmgmt_idr, ath11k_mac_tx_mgmt_pending_free, ar);
+		idr_destroy(&ar->txmgmt_idr);
+
+		kfree(ar->mac.sbands[NL80211_BAND_2GHZ].channels);
+		kfree(ar->mac.sbands[NL80211_BAND_5GHZ].channels);
+
+		SET_IEEE80211_DEV(ar->hw, NULL);
+	}
+}
+
+int ath11k_mac_create(struct ath11k_base *ab)
+{
+	struct ieee80211_hw *hw;
+	struct ath11k *ar;
+	struct ath11k_pdev *pdev;
+	int ret;
+	int i;
+
+	if (test_bit(ATH11K_FLAG_REGISTERED, &ab->dev_flags))
+		return 0;
+
+	for (i = 0; i < ab->num_radios; i++) {
+		pdev = &ab->pdevs[i];
+		hw = ieee80211_alloc_hw(sizeof(struct ath11k), &ath11k_ops);
+		if (!hw) {
+			ath11k_warn(ab, "failed to allocate mac80211 hw device\n");
+			ret = -ENOMEM;
+			goto err_destroy_mac;
+		}
+
+		ar = hw->priv;
+		ar->hw = hw;
+		ar->ab = ab;
+		ar->pdev = pdev;
+		ar->pdev_idx = i;
+		ar->lmac_id = ath11k_core_get_hw_mac_id(ab, i);
+
+		ar->wmi = &ab->wmi_sc.wmi[i];
+		/* FIXME wmi[0] is already initialized during attach,
+		 * Should we do this again?
+		 */
+		ath11k_wmi_pdev_attach(ab, i);
+
+		ar->cfg_tx_chainmask = pdev->cap.tx_chain_mask;
+		ar->cfg_rx_chainmask = pdev->cap.rx_chain_mask;
+		ar->num_tx_chains = get_num_chains(pdev->cap.tx_chain_mask);
+		ar->num_rx_chains = get_num_chains(pdev->cap.rx_chain_mask);
+
+		if (ab->pdevs_macaddr_valid) {
+			ether_addr_copy(ar->mac_addr, pdev->mac_addr);
+		} else {
+			ether_addr_copy(ar->mac_addr, ab->mac_addr);
+			ar->mac_addr[4] += i;
+		}
+
+		pdev->ar = ar;
+		spin_lock_init(&ar->data_lock);
+		INIT_LIST_HEAD(&ar->arvifs);
+		INIT_LIST_HEAD(&ar->ppdu_stats_info);
+		mutex_init(&ar->conf_mutex);
+		init_completion(&ar->vdev_setup_done);
+		init_completion(&ar->peer_assoc_done);
+		init_completion(&ar->install_key_done);
+		init_completion(&ar->bss_survey_done);
+		init_completion(&ar->scan.started);
+		init_completion(&ar->scan.completed);
+		INIT_DELAYED_WORK(&ar->scan.timeout, ath11k_scan_timeout_work);
+		INIT_WORK(&ar->regd_update_work, ath11k_regd_update_work);
+
+		INIT_WORK(&ar->wmi_mgmt_tx_work, ath11k_mgmt_over_wmi_tx_work);
+		skb_queue_head_init(&ar->wmi_mgmt_tx_queue);
+		clear_bit(ATH11K_FLAG_MONITOR_ENABLED, &ar->monitor_flags);
+
+		ret = ath11k_mac_register(ar);
+		if (ret) {
+			ath11k_warn(ab, "failed to register hw device\n");
+			pdev->ar = NULL;
+			ieee80211_free_hw(hw);
+			goto err_destroy_mac;
+		}
+
+		idr_init(&ar->txmgmt_idr);
+		spin_lock_init(&ar->txmgmt_idr_lock);
+	}
+
+	/* Initialize channel counters frequency value in hertz */
+	ab->cc_freq_hz = IPQ8074_CC_FREQ_HERTZ;
+	ab->free_vdev_map = (1LL << (ab->num_radios * TARGET_NUM_VDEVS)) - 1;
+
+	return 0;
+
+err_destroy_mac:
+	ath11k_mac_destroy(ab);
+
+	return ret;
+}
+
+void ath11k_mac_destroy(struct ath11k_base *ab)
+{
+	struct ath11k *ar;
+	struct ath11k_pdev *pdev;
+	int i;
+
+	for (i = 0; i < ab->num_radios; i++) {
+		pdev = &ab->pdevs[i];
+		ar = pdev->ar;
+		if (!ar)
+			continue;
+
+		ieee80211_free_hw(ar->hw);
+		pdev->ar = NULL;
+	}
+}
diff --git a/drivers/net/wireless/ath/ath11k/mac.h b/drivers/net/wireless/ath/ath11k/mac.h
new file mode 100644
index 000000000000..f84af1fc4952
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/mac.h
@@ -0,0 +1,145 @@
+/* SPDX-License-Identifier: BSD-3-Clause-Clear */
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#ifndef ATH11K_MAC_H
+#define ATH11K_MAC_H
+
+#include <net/mac80211.h>
+#include <net/cfg80211.h>
+
+struct ath11k;
+struct ath11k_base;
+
+struct ath11k_generic_iter {
+	struct ath11k *ar;
+	int ret;
+};
+
+/* number of failed packets (20 packets with 16 sw reties each) */
+#define ATH11K_KICKOUT_THRESHOLD		(20 * 16)
+
+/* Use insanely high numbers to make sure that the firmware implementation
+ * won't start, we have the same functionality already in hostapd. Unit
+ * is seconds.
+ */
+#define ATH11K_KEEPALIVE_MIN_IDLE		3747
+#define ATH11K_KEEPALIVE_MAX_IDLE		3895
+#define ATH11K_KEEPALIVE_MAX_UNRESPONSIVE	3900
+
+#define WMI_HOST_RC_DS_FLAG			0x01
+#define WMI_HOST_RC_CW40_FLAG			0x02
+#define WMI_HOST_RC_SGI_FLAG			0x04
+#define WMI_HOST_RC_HT_FLAG			0x08
+#define WMI_HOST_RC_RTSCTS_FLAG			0x10
+#define WMI_HOST_RC_TX_STBC_FLAG		0x20
+#define WMI_HOST_RC_RX_STBC_FLAG		0xC0
+#define WMI_HOST_RC_RX_STBC_FLAG_S		6
+#define WMI_HOST_RC_WEP_TKIP_FLAG		0x100
+#define WMI_HOST_RC_TS_FLAG			0x200
+#define WMI_HOST_RC_UAPSD_FLAG			0x400
+
+#define WMI_HT_CAP_ENABLED			0x0001
+#define WMI_HT_CAP_HT20_SGI			0x0002
+#define WMI_HT_CAP_DYNAMIC_SMPS			0x0004
+#define WMI_HT_CAP_TX_STBC			0x0008
+#define WMI_HT_CAP_TX_STBC_MASK_SHIFT		3
+#define WMI_HT_CAP_RX_STBC			0x0030
+#define WMI_HT_CAP_RX_STBC_MASK_SHIFT		4
+#define WMI_HT_CAP_LDPC				0x0040
+#define WMI_HT_CAP_L_SIG_TXOP_PROT		0x0080
+#define WMI_HT_CAP_MPDU_DENSITY			0x0700
+#define WMI_HT_CAP_MPDU_DENSITY_MASK_SHIFT	8
+#define WMI_HT_CAP_HT40_SGI			0x0800
+#define WMI_HT_CAP_RX_LDPC			0x1000
+#define WMI_HT_CAP_TX_LDPC			0x2000
+#define WMI_HT_CAP_IBF_BFER			0x4000
+
+/* These macros should be used when we wish to advertise STBC support for
+ * only 1SS or 2SS or 3SS.
+ */
+#define WMI_HT_CAP_RX_STBC_1SS			0x0010
+#define WMI_HT_CAP_RX_STBC_2SS			0x0020
+#define WMI_HT_CAP_RX_STBC_3SS			0x0030
+
+#define WMI_HT_CAP_DEFAULT_ALL (WMI_HT_CAP_ENABLED    | \
+				WMI_HT_CAP_HT20_SGI   | \
+				WMI_HT_CAP_HT40_SGI   | \
+				WMI_HT_CAP_TX_STBC    | \
+				WMI_HT_CAP_RX_STBC    | \
+				WMI_HT_CAP_LDPC)
+
+#define WMI_VHT_CAP_MAX_MPDU_LEN_MASK		0x00000003
+#define WMI_VHT_CAP_RX_LDPC			0x00000010
+#define WMI_VHT_CAP_SGI_80MHZ			0x00000020
+#define WMI_VHT_CAP_SGI_160MHZ			0x00000040
+#define WMI_VHT_CAP_TX_STBC			0x00000080
+#define WMI_VHT_CAP_RX_STBC_MASK		0x00000300
+#define WMI_VHT_CAP_RX_STBC_MASK_SHIFT		8
+#define WMI_VHT_CAP_SU_BFER			0x00000800
+#define WMI_VHT_CAP_SU_BFEE			0x00001000
+#define WMI_VHT_CAP_MAX_CS_ANT_MASK		0x0000E000
+#define WMI_VHT_CAP_MAX_CS_ANT_MASK_SHIFT	13
+#define WMI_VHT_CAP_MAX_SND_DIM_MASK		0x00070000
+#define WMI_VHT_CAP_MAX_SND_DIM_MASK_SHIFT	16
+#define WMI_VHT_CAP_MU_BFER			0x00080000
+#define WMI_VHT_CAP_MU_BFEE			0x00100000
+#define WMI_VHT_CAP_MAX_AMPDU_LEN_EXP		0x03800000
+#define WMI_VHT_CAP_MAX_AMPDU_LEN_EXP_SHIT	23
+#define WMI_VHT_CAP_RX_FIXED_ANT		0x10000000
+#define WMI_VHT_CAP_TX_FIXED_ANT		0x20000000
+
+#define WMI_VHT_CAP_MAX_MPDU_LEN_11454		0x00000002
+
+/* These macros should be used when we wish to advertise STBC support for
+ * only 1SS or 2SS or 3SS.
+ */
+#define WMI_VHT_CAP_RX_STBC_1SS			0x00000100
+#define WMI_VHT_CAP_RX_STBC_2SS			0x00000200
+#define WMI_VHT_CAP_RX_STBC_3SS			0x00000300
+
+#define WMI_VHT_CAP_DEFAULT_ALL (WMI_VHT_CAP_MAX_MPDU_LEN_11454  | \
+				 WMI_VHT_CAP_SGI_80MHZ      |       \
+				 WMI_VHT_CAP_TX_STBC        |       \
+				 WMI_VHT_CAP_RX_STBC_MASK   |       \
+				 WMI_VHT_CAP_RX_LDPC        |       \
+				 WMI_VHT_CAP_MAX_AMPDU_LEN_EXP   |  \
+				 WMI_VHT_CAP_RX_FIXED_ANT   |       \
+				 WMI_VHT_CAP_TX_FIXED_ANT)
+
+/* FIXME: should these be in ieee80211.h? */
+#define IEEE80211_VHT_MCS_SUPPORT_0_11_MASK	GENMASK(23, 16)
+#define IEEE80211_DISABLE_VHT_MCS_SUPPORT_0_11	BIT(24)
+
+#define WMI_MAX_SPATIAL_STREAM			3
+
+#define ATH11K_CHAN_WIDTH_NUM			8
+
+extern const struct htt_rx_ring_tlv_filter ath11k_mac_mon_status_filter_default;
+
+int ath11k_mac_create(struct ath11k_base *ab);
+void ath11k_mac_destroy(struct ath11k_base *ab);
+void ath11k_mac_unregister(struct ath11k_base *ab);
+int ath11k_mac_hw_ratecode_to_legacy_rate(u8 hw_rc, u8 preamble, u8 *rateidx,
+					  u16 *rate);
+u8 ath11k_mac_bitrate_to_idx(const struct ieee80211_supported_band *sband,
+			     u32 bitrate);
+u8 ath11k_mac_hw_rate_to_idx(const struct ieee80211_supported_band *sband,
+			     u8 hw_rate, bool cck);
+
+void __ath11k_mac_scan_finish(struct ath11k *ar);
+void ath11k_mac_scan_finish(struct ath11k *ar);
+
+struct ath11k_vif *ath11k_mac_get_arvif(struct ath11k *ar, u32 vdev_id);
+struct ath11k_vif *ath11k_mac_get_arvif_by_vdev_id(struct ath11k_base *ab,
+						   u32 vdev_id);
+struct ath11k *ath11k_mac_get_ar_by_vdev_id(struct ath11k_base *ab, u32 vdev_id);
+struct ath11k *ath11k_mac_get_ar_by_pdev_id(struct ath11k_base *ab, u32 pdev_id);
+struct ath11k *ath11k_mac_get_ar_vdev_stop_status(struct ath11k_base *ab,
+						  u32 vdev_id);
+
+void ath11k_mac_drain_tx(struct ath11k *ar);
+void ath11k_mac_peer_cleanup_all(struct ath11k *ar);
+int ath11k_mac_tx_mgmt_pending_free(int buf_id, void *skb, void *ctx);
+#endif
diff --git a/drivers/net/wireless/ath/ath11k/peer.c b/drivers/net/wireless/ath/ath11k/peer.c
new file mode 100644
index 000000000000..4bf1dfa498b6
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/peer.c
@@ -0,0 +1,236 @@
+// SPDX-License-Identifier: BSD-3-Clause-Clear
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#include "core.h"
+#include "peer.h"
+#include "debug.h"
+
+struct ath11k_peer *ath11k_peer_find(struct ath11k_base *ab, int vdev_id,
+				     const u8 *addr)
+{
+	struct ath11k_peer *peer;
+
+	lockdep_assert_held(&ab->base_lock);
+
+	list_for_each_entry(peer, &ab->peers, list) {
+		if (peer->vdev_id != vdev_id)
+			continue;
+		if (memcmp(peer->addr, addr, ETH_ALEN))
+			continue;
+
+		return peer;
+	}
+
+	return NULL;
+}
+
+struct ath11k_peer *ath11k_peer_find_by_addr(struct ath11k_base *ab,
+					     const u8 *addr)
+{
+	struct ath11k_peer *peer;
+
+	lockdep_assert_held(&ab->base_lock);
+
+	list_for_each_entry(peer, &ab->peers, list) {
+		if (memcmp(peer->addr, addr, ETH_ALEN))
+			continue;
+
+		return peer;
+	}
+
+	return NULL;
+}
+
+struct ath11k_peer *ath11k_peer_find_by_id(struct ath11k_base *ab,
+					   int peer_id)
+{
+	struct ath11k_peer *peer;
+
+	lockdep_assert_held(&ab->base_lock);
+
+	list_for_each_entry(peer, &ab->peers, list)
+		if (peer_id == peer->peer_id)
+			return peer;
+
+	return NULL;
+}
+
+void ath11k_peer_unmap_event(struct ath11k_base *ab, u16 peer_id)
+{
+	struct ath11k_peer *peer;
+
+	spin_lock_bh(&ab->base_lock);
+
+	peer = ath11k_peer_find_by_id(ab, peer_id);
+	if (!peer) {
+		ath11k_warn(ab, "peer-unmap-event: unknown peer id %d\n",
+			    peer_id);
+		goto exit;
+	}
+
+	ath11k_dbg(ab, ATH11K_DBG_DP_HTT, "htt peer unmap vdev %d peer %pM id %d\n",
+		   peer->vdev_id, peer->addr, peer_id);
+
+	list_del(&peer->list);
+	kfree(peer);
+	wake_up(&ab->peer_mapping_wq);
+
+exit:
+	spin_unlock_bh(&ab->base_lock);
+}
+
+void ath11k_peer_map_event(struct ath11k_base *ab, u8 vdev_id, u16 peer_id,
+			   u8 *mac_addr, u16 ast_hash)
+{
+	struct ath11k_peer *peer;
+
+	spin_lock_bh(&ab->base_lock);
+	peer = ath11k_peer_find(ab, vdev_id, mac_addr);
+	if (!peer) {
+		peer = kzalloc(sizeof(*peer), GFP_ATOMIC);
+		if (!peer)
+			goto exit;
+
+		peer->vdev_id = vdev_id;
+		peer->peer_id = peer_id;
+		peer->ast_hash = ast_hash;
+		ether_addr_copy(peer->addr, mac_addr);
+		list_add(&peer->list, &ab->peers);
+		wake_up(&ab->peer_mapping_wq);
+	}
+
+	ath11k_dbg(ab, ATH11K_DBG_DP_HTT, "htt peer map vdev %d peer %pM id %d\n",
+		   vdev_id, mac_addr, peer_id);
+
+exit:
+	spin_unlock_bh(&ab->base_lock);
+}
+
+static int ath11k_wait_for_peer_common(struct ath11k_base *ab, int vdev_id,
+				       const u8 *addr, bool expect_mapped)
+{
+	int ret;
+
+	ret = wait_event_timeout(ab->peer_mapping_wq, ({
+				bool mapped;
+
+				spin_lock_bh(&ab->base_lock);
+				mapped = !!ath11k_peer_find(ab, vdev_id, addr);
+				spin_unlock_bh(&ab->base_lock);
+
+				(mapped == expect_mapped ||
+				 test_bit(ATH11K_FLAG_CRASH_FLUSH, &ab->dev_flags));
+				}), 3 * HZ);
+
+	if (ret <= 0)
+		return -ETIMEDOUT;
+
+	return 0;
+}
+
+void ath11k_peer_cleanup(struct ath11k *ar, u32 vdev_id)
+{
+	struct ath11k_peer *peer, *tmp;
+	struct ath11k_base *ab = ar->ab;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	spin_lock_bh(&ab->base_lock);
+	list_for_each_entry_safe(peer, tmp, &ab->peers, list) {
+		if (peer->vdev_id != vdev_id)
+			continue;
+
+		ath11k_warn(ab, "removing stale peer %pM from vdev_id %d\n",
+			    peer->addr, vdev_id);
+
+		list_del(&peer->list);
+		kfree(peer);
+		ar->num_peers--;
+	}
+
+	spin_unlock_bh(&ab->base_lock);
+}
+
+static int ath11k_wait_for_peer_deleted(struct ath11k *ar, int vdev_id, const u8 *addr)
+{
+	return ath11k_wait_for_peer_common(ar->ab, vdev_id, addr, false);
+}
+
+int ath11k_peer_delete(struct ath11k *ar, u32 vdev_id, u8 *addr)
+{
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	ret = ath11k_wmi_send_peer_delete_cmd(ar, addr, vdev_id);
+	if (ret) {
+		ath11k_warn(ar->ab,
+			    "failed to delete peer vdev_id %d addr %pM ret %d\n",
+			    vdev_id, addr, ret);
+		return ret;
+	}
+
+	ret = ath11k_wait_for_peer_deleted(ar, vdev_id, addr);
+	if (ret)
+		return ret;
+
+	ar->num_peers--;
+
+	return 0;
+}
+
+static int ath11k_wait_for_peer_created(struct ath11k *ar, int vdev_id, const u8 *addr)
+{
+	return ath11k_wait_for_peer_common(ar->ab, vdev_id, addr, true);
+}
+
+int ath11k_peer_create(struct ath11k *ar, struct ath11k_vif *arvif,
+		       struct ieee80211_sta *sta, struct peer_create_params *param)
+{
+	struct ath11k_peer *peer;
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	if (ar->num_peers > (ar->max_num_peers - 1)) {
+		ath11k_warn(ar->ab,
+			    "failed to create peer due to insufficient peer entry resource in firmware\n");
+		return -ENOBUFS;
+	}
+
+	ret = ath11k_wmi_send_peer_create_cmd(ar, param);
+	if (ret) {
+		ath11k_warn(ar->ab,
+			    "failed to send peer create vdev_id %d ret %d\n",
+			    param->vdev_id, ret);
+		return ret;
+	}
+
+	ret = ath11k_wait_for_peer_created(ar, param->vdev_id,
+					   param->peer_addr);
+	if (ret)
+		return ret;
+
+	spin_lock_bh(&ar->ab->base_lock);
+
+	peer = ath11k_peer_find(ar->ab, param->vdev_id, param->peer_addr);
+	if (!peer) {
+		spin_unlock_bh(&ar->ab->base_lock);
+		ath11k_warn(ar->ab, "failed to find peer %pM on vdev %i after creation\n",
+			    param->peer_addr, param->vdev_id);
+		ath11k_wmi_send_peer_delete_cmd(ar, param->peer_addr,
+						param->vdev_id);
+		return -ENOENT;
+	}
+
+	peer->sta = sta;
+	arvif->ast_hash = peer->ast_hash;
+
+	ar->num_peers++;
+
+	spin_unlock_bh(&ar->ab->base_lock);
+
+	return 0;
+}
diff --git a/drivers/net/wireless/ath/ath11k/peer.h b/drivers/net/wireless/ath/ath11k/peer.h
new file mode 100644
index 000000000000..9a40d1f6ccd9
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/peer.h
@@ -0,0 +1,35 @@
+/* SPDX-License-Identifier: BSD-3-Clause-Clear */
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#ifndef ATH11K_PEER_H
+#define ATH11K_PEER_H
+
+struct ath11k_peer {
+	struct list_head list;
+	struct ieee80211_sta *sta;
+	int vdev_id;
+	u8 addr[ETH_ALEN];
+	int peer_id;
+	u16 ast_hash;
+
+	/* protected by ab->data_lock */
+	struct ieee80211_key_conf *keys[WMI_MAX_KEY_INDEX + 1];
+	struct dp_rx_tid rx_tid[IEEE80211_NUM_TIDS + 1];
+};
+
+void ath11k_peer_unmap_event(struct ath11k_base *ab, u16 peer_id);
+void ath11k_peer_map_event(struct ath11k_base *ab, u8 vdev_id, u16 peer_id,
+			   u8 *mac_addr, u16 ast_hash);
+struct ath11k_peer *ath11k_peer_find(struct ath11k_base *ab, int vdev_id,
+				     const u8 *addr);
+struct ath11k_peer *ath11k_peer_find_by_addr(struct ath11k_base *ab,
+					     const u8 *addr);
+struct ath11k_peer *ath11k_peer_find_by_id(struct ath11k_base *ab, int peer_id);
+void ath11k_peer_cleanup(struct ath11k *ar, u32 vdev_id);
+int ath11k_peer_delete(struct ath11k *ar, u32 vdev_id, u8 *addr);
+int ath11k_peer_create(struct ath11k *ar, struct ath11k_vif *arvif,
+		       struct ieee80211_sta *sta, struct peer_create_params *param);
+
+#endif /* _PEER_H_ */
diff --git a/drivers/net/wireless/ath/ath11k/qmi.c b/drivers/net/wireless/ath/ath11k/qmi.c
new file mode 100644
index 000000000000..adf41f28a231
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/qmi.c
@@ -0,0 +1,2435 @@
+// SPDX-License-Identifier: BSD-3-Clause-Clear
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#include "qmi.h"
+#include "core.h"
+#include "debug.h"
+#include <linux/of.h>
+#include <linux/firmware.h>
+
+static struct qmi_elem_info qmi_wlanfw_host_cap_req_msg_v01_ei[] = {
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x10,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_req_msg_v01,
+					   num_clients_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_4_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u32),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x10,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_req_msg_v01,
+					   num_clients),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x11,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_req_msg_v01,
+					   wake_msi_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_4_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u32),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x11,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_req_msg_v01,
+					   wake_msi),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x12,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_req_msg_v01,
+					   gpios_valid),
+	},
+	{
+		.data_type	= QMI_DATA_LEN,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x12,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_req_msg_v01,
+					   gpios_len),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_4_BYTE,
+		.elem_len	= QMI_WLFW_MAX_NUM_GPIO_V01,
+		.elem_size	= sizeof(u32),
+		.array_type	= VAR_LEN_ARRAY,
+		.tlv_type	= 0x12,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_req_msg_v01,
+					   gpios),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x13,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_req_msg_v01,
+					   nm_modem_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x13,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_req_msg_v01,
+					   nm_modem),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x14,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_req_msg_v01,
+					   bdf_support_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x14,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_req_msg_v01,
+					   bdf_support),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x15,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_req_msg_v01,
+					   bdf_cache_support_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x15,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_req_msg_v01,
+					   bdf_cache_support),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x16,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_req_msg_v01,
+					   m3_support_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x16,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_req_msg_v01,
+					   m3_support),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x17,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_req_msg_v01,
+					   m3_cache_support_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x17,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_req_msg_v01,
+					   m3_cache_support),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x18,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_req_msg_v01,
+					   cal_filesys_support_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x18,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_req_msg_v01,
+					   cal_filesys_support),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x19,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_req_msg_v01,
+					   cal_cache_support_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x19,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_req_msg_v01,
+					   cal_cache_support),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x1A,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_req_msg_v01,
+					   cal_done_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x1A,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_req_msg_v01,
+					   cal_done),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x1B,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_req_msg_v01,
+					   mem_bucket_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_4_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u32),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x1B,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_req_msg_v01,
+					   mem_bucket),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x1C,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_req_msg_v01,
+					   mem_cfg_mode_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x1C,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_req_msg_v01,
+					   mem_cfg_mode),
+	},
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= NO_ARRAY,
+		.tlv_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_host_cap_resp_msg_v01_ei[] = {
+	{
+		.data_type	= QMI_STRUCT,
+		.elem_len	= 1,
+		.elem_size	= sizeof(struct qmi_response_type_v01),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x02,
+		.offset		= offsetof(struct qmi_wlanfw_host_cap_resp_msg_v01, resp),
+		.ei_array	= qmi_response_type_v01_ei,
+	},
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= NO_ARRAY,
+		.tlv_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_ind_register_req_msg_v01_ei[] = {
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x10,
+		.offset		= offsetof(struct qmi_wlanfw_ind_register_req_msg_v01,
+					   fw_ready_enable_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x10,
+		.offset		= offsetof(struct qmi_wlanfw_ind_register_req_msg_v01,
+					   fw_ready_enable),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x11,
+		.offset		= offsetof(struct qmi_wlanfw_ind_register_req_msg_v01,
+					   initiate_cal_download_enable_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x11,
+		.offset		= offsetof(struct qmi_wlanfw_ind_register_req_msg_v01,
+					   initiate_cal_download_enable),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x12,
+		.offset		= offsetof(struct qmi_wlanfw_ind_register_req_msg_v01,
+					   initiate_cal_update_enable_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x12,
+		.offset		= offsetof(struct qmi_wlanfw_ind_register_req_msg_v01,
+					   initiate_cal_update_enable),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x13,
+		.offset		= offsetof(struct qmi_wlanfw_ind_register_req_msg_v01,
+					   msa_ready_enable_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x13,
+		.offset		= offsetof(struct qmi_wlanfw_ind_register_req_msg_v01,
+					   msa_ready_enable),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x14,
+		.offset		= offsetof(struct qmi_wlanfw_ind_register_req_msg_v01,
+					   pin_connect_result_enable_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x14,
+		.offset		= offsetof(struct qmi_wlanfw_ind_register_req_msg_v01,
+					   pin_connect_result_enable),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x15,
+		.offset		= offsetof(struct qmi_wlanfw_ind_register_req_msg_v01,
+					   client_id_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_4_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u32),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x15,
+		.offset		= offsetof(struct qmi_wlanfw_ind_register_req_msg_v01,
+					   client_id),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x16,
+		.offset		= offsetof(struct qmi_wlanfw_ind_register_req_msg_v01,
+					   request_mem_enable_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x16,
+		.offset		= offsetof(struct qmi_wlanfw_ind_register_req_msg_v01,
+					   request_mem_enable),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x17,
+		.offset		= offsetof(struct qmi_wlanfw_ind_register_req_msg_v01,
+					   fw_mem_ready_enable_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x17,
+		.offset		= offsetof(struct qmi_wlanfw_ind_register_req_msg_v01,
+					   fw_mem_ready_enable),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x18,
+		.offset		= offsetof(struct qmi_wlanfw_ind_register_req_msg_v01,
+					   fw_init_done_enable_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x18,
+		.offset		= offsetof(struct qmi_wlanfw_ind_register_req_msg_v01,
+					   fw_init_done_enable),
+	},
+
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x19,
+		.offset		= offsetof(struct qmi_wlanfw_ind_register_req_msg_v01,
+					   rejuvenate_enable_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x19,
+		.offset		= offsetof(struct qmi_wlanfw_ind_register_req_msg_v01,
+					   rejuvenate_enable),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x1A,
+		.offset		= offsetof(struct qmi_wlanfw_ind_register_req_msg_v01,
+					   xo_cal_enable_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x1A,
+		.offset		= offsetof(struct qmi_wlanfw_ind_register_req_msg_v01,
+					   xo_cal_enable),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x1B,
+		.offset		= offsetof(struct qmi_wlanfw_ind_register_req_msg_v01,
+					   cal_done_enable_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x1B,
+		.offset		= offsetof(struct qmi_wlanfw_ind_register_req_msg_v01,
+					   cal_done_enable),
+	},
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= NO_ARRAY,
+		.tlv_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_ind_register_resp_msg_v01_ei[] = {
+	{
+		.data_type	= QMI_STRUCT,
+		.elem_len	= 1,
+		.elem_size	= sizeof(struct qmi_response_type_v01),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x02,
+		.offset		= offsetof(struct qmi_wlanfw_ind_register_resp_msg_v01,
+					   resp),
+		.ei_array	= qmi_response_type_v01_ei,
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x10,
+		.offset		= offsetof(struct qmi_wlanfw_ind_register_resp_msg_v01,
+					   fw_status_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_8_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u64),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x10,
+		.offset		= offsetof(struct qmi_wlanfw_ind_register_resp_msg_v01,
+					   fw_status),
+	},
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= NO_ARRAY,
+		.tlv_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_mem_cfg_s_v01_ei[] = {
+	{
+		.data_type	= QMI_UNSIGNED_8_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u64),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_mem_cfg_s_v01, offset),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_4_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u32),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_mem_cfg_s_v01, size),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_mem_cfg_s_v01, secure_flag),
+	},
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= NO_ARRAY,
+		.tlv_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_mem_seg_s_v01_ei[] = {
+	{
+		.data_type	= QMI_UNSIGNED_4_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u32),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_mem_seg_s_v01,
+				  size),
+	},
+	{
+		.data_type	= QMI_SIGNED_4_BYTE_ENUM,
+		.elem_len	= 1,
+		.elem_size	= sizeof(enum qmi_wlanfw_mem_type_enum_v01),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_mem_seg_s_v01, type),
+	},
+	{
+		.data_type	= QMI_DATA_LEN,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_mem_seg_s_v01, mem_cfg_len),
+	},
+	{
+		.data_type	= QMI_STRUCT,
+		.elem_len	= QMI_WLANFW_MAX_NUM_MEM_CFG_V01,
+		.elem_size	= sizeof(struct qmi_wlanfw_mem_cfg_s_v01),
+		.array_type	= VAR_LEN_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_mem_seg_s_v01, mem_cfg),
+		.ei_array	= qmi_wlanfw_mem_cfg_s_v01_ei,
+	},
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= NO_ARRAY,
+		.tlv_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_request_mem_ind_msg_v01_ei[] = {
+	{
+		.data_type	= QMI_DATA_LEN,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x01,
+		.offset		= offsetof(struct qmi_wlanfw_request_mem_ind_msg_v01,
+					   mem_seg_len),
+	},
+	{
+		.data_type	= QMI_STRUCT,
+		.elem_len	= ATH11K_QMI_WLANFW_MAX_NUM_MEM_SEG_V01,
+		.elem_size	= sizeof(struct qmi_wlanfw_mem_seg_s_v01),
+		.array_type	= VAR_LEN_ARRAY,
+		.tlv_type	= 0x01,
+		.offset		= offsetof(struct qmi_wlanfw_request_mem_ind_msg_v01,
+					   mem_seg),
+		.ei_array	= qmi_wlanfw_mem_seg_s_v01_ei,
+	},
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= NO_ARRAY,
+		.tlv_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_mem_seg_resp_s_v01_ei[] = {
+	{
+		.data_type	= QMI_UNSIGNED_8_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u64),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_mem_seg_resp_s_v01, addr),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_4_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u32),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_mem_seg_resp_s_v01, size),
+	},
+	{
+		.data_type	= QMI_SIGNED_4_BYTE_ENUM,
+		.elem_len	= 1,
+		.elem_size	= sizeof(enum qmi_wlanfw_mem_type_enum_v01),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_mem_seg_resp_s_v01, type),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_mem_seg_resp_s_v01, restore),
+	},
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= NO_ARRAY,
+		.tlv_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_respond_mem_req_msg_v01_ei[] = {
+	{
+		.data_type	= QMI_DATA_LEN,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x01,
+		.offset		= offsetof(struct qmi_wlanfw_respond_mem_req_msg_v01,
+					   mem_seg_len),
+	},
+	{
+		.data_type	= QMI_STRUCT,
+		.elem_len	= ATH11K_QMI_WLANFW_MAX_NUM_MEM_SEG_V01,
+		.elem_size	= sizeof(struct qmi_wlanfw_mem_seg_resp_s_v01),
+		.array_type	= VAR_LEN_ARRAY,
+		.tlv_type	= 0x01,
+		.offset		= offsetof(struct qmi_wlanfw_respond_mem_req_msg_v01,
+					   mem_seg),
+		.ei_array	= qmi_wlanfw_mem_seg_resp_s_v01_ei,
+	},
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= NO_ARRAY,
+		.tlv_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_respond_mem_resp_msg_v01_ei[] = {
+	{
+		.data_type	= QMI_STRUCT,
+		.elem_len	= 1,
+		.elem_size	= sizeof(struct qmi_response_type_v01),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x02,
+		.offset		= offsetof(struct qmi_wlanfw_respond_mem_resp_msg_v01,
+					   resp),
+		.ei_array	= qmi_response_type_v01_ei,
+	},
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= NO_ARRAY,
+		.tlv_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_cap_req_msg_v01_ei[] = {
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= NO_ARRAY,
+		.tlv_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_rf_chip_info_s_v01_ei[] = {
+	{
+		.data_type	= QMI_UNSIGNED_4_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u32),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_rf_chip_info_s_v01,
+					   chip_id),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_4_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u32),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_rf_chip_info_s_v01,
+					   chip_family),
+	},
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= NO_ARRAY,
+		.tlv_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_rf_board_info_s_v01_ei[] = {
+	{
+		.data_type	= QMI_UNSIGNED_4_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u32),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_rf_board_info_s_v01,
+					   board_id),
+	},
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= NO_ARRAY,
+		.tlv_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_soc_info_s_v01_ei[] = {
+	{
+		.data_type	= QMI_UNSIGNED_4_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u32),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_soc_info_s_v01, soc_id),
+	},
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= NO_ARRAY,
+		.tlv_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_fw_version_info_s_v01_ei[] = {
+	{
+		.data_type	= QMI_UNSIGNED_4_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u32),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_fw_version_info_s_v01,
+					   fw_version),
+	},
+	{
+		.data_type	= QMI_STRING,
+		.elem_len	= ATH11K_QMI_WLANFW_MAX_TIMESTAMP_LEN_V01 + 1,
+		.elem_size	= sizeof(char),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_fw_version_info_s_v01,
+					   fw_build_timestamp),
+	},
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= NO_ARRAY,
+		.tlv_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_cap_resp_msg_v01_ei[] = {
+	{
+		.data_type	= QMI_STRUCT,
+		.elem_len	= 1,
+		.elem_size	= sizeof(struct qmi_response_type_v01),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x02,
+		.offset		= offsetof(struct qmi_wlanfw_cap_resp_msg_v01, resp),
+		.ei_array	= qmi_response_type_v01_ei,
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x10,
+		.offset		= offsetof(struct qmi_wlanfw_cap_resp_msg_v01,
+					   chip_info_valid),
+	},
+	{
+		.data_type	= QMI_STRUCT,
+		.elem_len	= 1,
+		.elem_size	= sizeof(struct qmi_wlanfw_rf_chip_info_s_v01),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x10,
+		.offset		= offsetof(struct qmi_wlanfw_cap_resp_msg_v01,
+					   chip_info),
+		.ei_array	= qmi_wlanfw_rf_chip_info_s_v01_ei,
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x11,
+		.offset		= offsetof(struct qmi_wlanfw_cap_resp_msg_v01,
+					   board_info_valid),
+	},
+	{
+		.data_type	= QMI_STRUCT,
+		.elem_len	= 1,
+		.elem_size	= sizeof(struct qmi_wlanfw_rf_board_info_s_v01),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x11,
+		.offset		= offsetof(struct qmi_wlanfw_cap_resp_msg_v01,
+					   board_info),
+		.ei_array	= qmi_wlanfw_rf_board_info_s_v01_ei,
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x12,
+		.offset		= offsetof(struct qmi_wlanfw_cap_resp_msg_v01,
+					   soc_info_valid),
+	},
+	{
+		.data_type	= QMI_STRUCT,
+		.elem_len	= 1,
+		.elem_size	= sizeof(struct qmi_wlanfw_soc_info_s_v01),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x12,
+		.offset		= offsetof(struct qmi_wlanfw_cap_resp_msg_v01,
+					   soc_info),
+		.ei_array	= qmi_wlanfw_soc_info_s_v01_ei,
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x13,
+		.offset		= offsetof(struct qmi_wlanfw_cap_resp_msg_v01,
+					   fw_version_info_valid),
+	},
+	{
+		.data_type	= QMI_STRUCT,
+		.elem_len	= 1,
+		.elem_size	= sizeof(struct qmi_wlanfw_fw_version_info_s_v01),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x13,
+		.offset		= offsetof(struct qmi_wlanfw_cap_resp_msg_v01,
+					   fw_version_info),
+		.ei_array	= qmi_wlanfw_fw_version_info_s_v01_ei,
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x14,
+		.offset		= offsetof(struct qmi_wlanfw_cap_resp_msg_v01,
+					   fw_build_id_valid),
+	},
+	{
+		.data_type	= QMI_STRING,
+		.elem_len	= ATH11K_QMI_WLANFW_MAX_BUILD_ID_LEN_V01 + 1,
+		.elem_size	= sizeof(char),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x14,
+		.offset		= offsetof(struct qmi_wlanfw_cap_resp_msg_v01,
+					   fw_build_id),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x15,
+		.offset		= offsetof(struct qmi_wlanfw_cap_resp_msg_v01,
+					   num_macs_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x15,
+		.offset		= offsetof(struct qmi_wlanfw_cap_resp_msg_v01,
+					   num_macs),
+	},
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= NO_ARRAY,
+		.tlv_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_bdf_download_req_msg_v01_ei[] = {
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x01,
+		.offset		= offsetof(struct qmi_wlanfw_bdf_download_req_msg_v01,
+					   valid),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x10,
+		.offset		= offsetof(struct qmi_wlanfw_bdf_download_req_msg_v01,
+					   file_id_valid),
+	},
+	{
+		.data_type	= QMI_SIGNED_4_BYTE_ENUM,
+		.elem_len	= 1,
+		.elem_size	= sizeof(enum qmi_wlanfw_cal_temp_id_enum_v01),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x10,
+		.offset		= offsetof(struct qmi_wlanfw_bdf_download_req_msg_v01,
+					   file_id),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x11,
+		.offset		= offsetof(struct qmi_wlanfw_bdf_download_req_msg_v01,
+					   total_size_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_4_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u32),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x11,
+		.offset		= offsetof(struct qmi_wlanfw_bdf_download_req_msg_v01,
+					   total_size),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x12,
+		.offset		= offsetof(struct qmi_wlanfw_bdf_download_req_msg_v01,
+					   seg_id_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_4_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u32),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x12,
+		.offset		= offsetof(struct qmi_wlanfw_bdf_download_req_msg_v01,
+					   seg_id),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x13,
+		.offset		= offsetof(struct qmi_wlanfw_bdf_download_req_msg_v01,
+					   data_valid),
+	},
+	{
+		.data_type	= QMI_DATA_LEN,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u16),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x13,
+		.offset		= offsetof(struct qmi_wlanfw_bdf_download_req_msg_v01,
+					   data_len),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= QMI_WLANFW_MAX_DATA_SIZE_V01,
+		.elem_size	= sizeof(u8),
+		.array_type	= VAR_LEN_ARRAY,
+		.tlv_type	= 0x13,
+		.offset		= offsetof(struct qmi_wlanfw_bdf_download_req_msg_v01,
+					   data),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x14,
+		.offset		= offsetof(struct qmi_wlanfw_bdf_download_req_msg_v01,
+					   end_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x14,
+		.offset		= offsetof(struct qmi_wlanfw_bdf_download_req_msg_v01,
+					   end),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x15,
+		.offset		= offsetof(struct qmi_wlanfw_bdf_download_req_msg_v01,
+					   bdf_type_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x15,
+		.offset		= offsetof(struct qmi_wlanfw_bdf_download_req_msg_v01,
+					   bdf_type),
+	},
+
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= NO_ARRAY,
+		.tlv_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_bdf_download_resp_msg_v01_ei[] = {
+	{
+		.data_type	= QMI_STRUCT,
+		.elem_len	= 1,
+		.elem_size	= sizeof(struct qmi_response_type_v01),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x02,
+		.offset		= offsetof(struct qmi_wlanfw_bdf_download_resp_msg_v01,
+					   resp),
+		.ei_array	= qmi_response_type_v01_ei,
+	},
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= NO_ARRAY,
+		.tlv_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_m3_info_req_msg_v01_ei[] = {
+	{
+		.data_type	= QMI_UNSIGNED_8_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u64),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x01,
+		.offset		= offsetof(struct qmi_wlanfw_m3_info_req_msg_v01, addr),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_4_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u32),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x02,
+		.offset		= offsetof(struct qmi_wlanfw_m3_info_req_msg_v01, size),
+	},
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= NO_ARRAY,
+		.tlv_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_m3_info_resp_msg_v01_ei[] = {
+	{
+		.data_type	= QMI_STRUCT,
+		.elem_len	= 1,
+		.elem_size	= sizeof(struct qmi_response_type_v01),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x02,
+		.offset		= offsetof(struct qmi_wlanfw_m3_info_resp_msg_v01, resp),
+		.ei_array	= qmi_response_type_v01_ei,
+	},
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= NO_ARRAY,
+		.tlv_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_ce_tgt_pipe_cfg_s_v01_ei[] = {
+	{
+		.data_type	= QMI_UNSIGNED_4_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u32),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_ce_tgt_pipe_cfg_s_v01,
+					   pipe_num),
+	},
+	{
+		.data_type	= QMI_SIGNED_4_BYTE_ENUM,
+		.elem_len	= 1,
+		.elem_size	= sizeof(enum qmi_wlanfw_pipedir_enum_v01),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_ce_tgt_pipe_cfg_s_v01,
+					   pipe_dir),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_4_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u32),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_ce_tgt_pipe_cfg_s_v01,
+					   nentries),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_4_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u32),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_ce_tgt_pipe_cfg_s_v01,
+					   nbytes_max),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_4_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u32),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_ce_tgt_pipe_cfg_s_v01,
+					   flags),
+	},
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= NO_ARRAY,
+		.tlv_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_ce_svc_pipe_cfg_s_v01_ei[] = {
+	{
+		.data_type	= QMI_UNSIGNED_4_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u32),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_ce_svc_pipe_cfg_s_v01,
+					   service_id),
+	},
+	{
+		.data_type	= QMI_SIGNED_4_BYTE_ENUM,
+		.elem_len	= 1,
+		.elem_size	= sizeof(enum qmi_wlanfw_pipedir_enum_v01),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_ce_svc_pipe_cfg_s_v01,
+					   pipe_dir),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_4_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u32),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_ce_svc_pipe_cfg_s_v01,
+					   pipe_num),
+	},
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= NO_ARRAY,
+		.tlv_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_shadow_reg_cfg_s_v01_ei[] = {
+	{
+		.data_type	= QMI_UNSIGNED_2_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u16),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_shadow_reg_cfg_s_v01, id),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_2_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u16),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_shadow_reg_cfg_s_v01,
+					   offset),
+	},
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_shadow_reg_v2_cfg_s_v01_ei[] = {
+	{
+		.data_type	= QMI_UNSIGNED_4_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u32),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0,
+		.offset		= offsetof(struct qmi_wlanfw_shadow_reg_v2_cfg_s_v01,
+					   addr),
+	},
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= NO_ARRAY,
+		.tlv_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_wlan_mode_req_msg_v01_ei[] = {
+	{
+		.data_type	= QMI_UNSIGNED_4_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u32),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x01,
+		.offset		= offsetof(struct qmi_wlanfw_wlan_mode_req_msg_v01,
+					   mode),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x10,
+		.offset		= offsetof(struct qmi_wlanfw_wlan_mode_req_msg_v01,
+					   hw_debug_valid),
+	},
+	{
+		.data_type	= QMI_UNSIGNED_1_BYTE,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x10,
+		.offset		= offsetof(struct qmi_wlanfw_wlan_mode_req_msg_v01,
+					   hw_debug),
+	},
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= NO_ARRAY,
+		.tlv_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_wlan_mode_resp_msg_v01_ei[] = {
+	{
+		.data_type	= QMI_STRUCT,
+		.elem_len	= 1,
+		.elem_size	= sizeof(struct qmi_response_type_v01),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x02,
+		.offset		= offsetof(struct qmi_wlanfw_wlan_mode_resp_msg_v01,
+					   resp),
+		.ei_array	= qmi_response_type_v01_ei,
+	},
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= NO_ARRAY,
+		.tlv_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_wlan_cfg_req_msg_v01_ei[] = {
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x10,
+		.offset		= offsetof(struct qmi_wlanfw_wlan_cfg_req_msg_v01,
+					   host_version_valid),
+	},
+	{
+		.data_type	= QMI_STRING,
+		.elem_len	= QMI_WLANFW_MAX_STR_LEN_V01 + 1,
+		.elem_size	= sizeof(char),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x10,
+		.offset		= offsetof(struct qmi_wlanfw_wlan_cfg_req_msg_v01,
+					   host_version),
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x11,
+		.offset		= offsetof(struct qmi_wlanfw_wlan_cfg_req_msg_v01,
+					   tgt_cfg_valid),
+	},
+	{
+		.data_type	= QMI_DATA_LEN,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x11,
+		.offset		= offsetof(struct qmi_wlanfw_wlan_cfg_req_msg_v01,
+					   tgt_cfg_len),
+	},
+	{
+		.data_type	= QMI_STRUCT,
+		.elem_len	= QMI_WLANFW_MAX_NUM_CE_V01,
+		.elem_size	= sizeof(
+				struct qmi_wlanfw_ce_tgt_pipe_cfg_s_v01),
+		.array_type	= VAR_LEN_ARRAY,
+		.tlv_type	= 0x11,
+		.offset		= offsetof(struct qmi_wlanfw_wlan_cfg_req_msg_v01,
+					   tgt_cfg),
+		.ei_array	= qmi_wlanfw_ce_tgt_pipe_cfg_s_v01_ei,
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x12,
+		.offset		= offsetof(struct qmi_wlanfw_wlan_cfg_req_msg_v01,
+					   svc_cfg_valid),
+	},
+	{
+		.data_type	= QMI_DATA_LEN,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x12,
+		.offset		= offsetof(struct qmi_wlanfw_wlan_cfg_req_msg_v01,
+					   svc_cfg_len),
+	},
+	{
+		.data_type	= QMI_STRUCT,
+		.elem_len	= QMI_WLANFW_MAX_NUM_SVC_V01,
+		.elem_size	= sizeof(struct qmi_wlanfw_ce_svc_pipe_cfg_s_v01),
+		.array_type	= VAR_LEN_ARRAY,
+		.tlv_type	= 0x12,
+		.offset		= offsetof(struct qmi_wlanfw_wlan_cfg_req_msg_v01,
+					   svc_cfg),
+		.ei_array	= qmi_wlanfw_ce_svc_pipe_cfg_s_v01_ei,
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x13,
+		.offset		= offsetof(struct qmi_wlanfw_wlan_cfg_req_msg_v01,
+					   shadow_reg_valid),
+	},
+	{
+		.data_type	= QMI_DATA_LEN,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x13,
+		.offset		= offsetof(struct qmi_wlanfw_wlan_cfg_req_msg_v01,
+					   shadow_reg_len),
+	},
+	{
+		.data_type	= QMI_STRUCT,
+		.elem_len	= QMI_WLANFW_MAX_NUM_SHADOW_REG_V01,
+		.elem_size	= sizeof(struct qmi_wlanfw_shadow_reg_cfg_s_v01),
+		.array_type	= VAR_LEN_ARRAY,
+		.tlv_type	= 0x13,
+		.offset		= offsetof(struct qmi_wlanfw_wlan_cfg_req_msg_v01,
+					   shadow_reg),
+		.ei_array	= qmi_wlanfw_shadow_reg_cfg_s_v01_ei,
+	},
+	{
+		.data_type	= QMI_OPT_FLAG,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x14,
+		.offset		= offsetof(struct qmi_wlanfw_wlan_cfg_req_msg_v01,
+					   shadow_reg_v2_valid),
+	},
+	{
+		.data_type	= QMI_DATA_LEN,
+		.elem_len	= 1,
+		.elem_size	= sizeof(u8),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x14,
+		.offset		= offsetof(struct qmi_wlanfw_wlan_cfg_req_msg_v01,
+					   shadow_reg_v2_len),
+	},
+	{
+		.data_type	= QMI_STRUCT,
+		.elem_len	= QMI_WLANFW_MAX_NUM_SHADOW_REG_V2_V01,
+		.elem_size	= sizeof(struct qmi_wlanfw_shadow_reg_v2_cfg_s_v01),
+		.array_type	= VAR_LEN_ARRAY,
+		.tlv_type	= 0x14,
+		.offset		= offsetof(struct qmi_wlanfw_wlan_cfg_req_msg_v01,
+					   shadow_reg_v2),
+		.ei_array	= qmi_wlanfw_shadow_reg_v2_cfg_s_v01_ei,
+	},
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= NO_ARRAY,
+		.tlv_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_wlan_cfg_resp_msg_v01_ei[] = {
+	{
+		.data_type	= QMI_STRUCT,
+		.elem_len	= 1,
+		.elem_size	= sizeof(struct qmi_response_type_v01),
+		.array_type	= NO_ARRAY,
+		.tlv_type	= 0x02,
+		.offset		= offsetof(struct qmi_wlanfw_wlan_cfg_resp_msg_v01, resp),
+		.ei_array	= qmi_response_type_v01_ei,
+	},
+	{
+		.data_type	= QMI_EOTI,
+		.array_type	= NO_ARRAY,
+		.tlv_type	= QMI_COMMON_TLV_TYPE,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_mem_ready_ind_msg_v01_ei[] = {
+	{
+		.data_type = QMI_EOTI,
+		.array_type = NO_ARRAY,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_fw_ready_ind_msg_v01_ei[] = {
+	{
+		.data_type = QMI_EOTI,
+		.array_type = NO_ARRAY,
+	},
+};
+
+static struct qmi_elem_info qmi_wlanfw_cold_boot_cal_done_ind_msg_v01_ei[] = {
+	{
+		.data_type = QMI_EOTI,
+		.array_type = NO_ARRAY,
+	},
+};
+
+static int ath11k_qmi_host_cap_send(struct ath11k_base *ab)
+{
+	struct qmi_wlanfw_host_cap_req_msg_v01 req;
+	struct qmi_wlanfw_host_cap_resp_msg_v01 resp;
+	struct qmi_txn txn = {};
+	int ret = 0;
+
+	memset(&req, 0, sizeof(req));
+	memset(&resp, 0, sizeof(resp));
+
+	req.num_clients_valid = 1;
+	req.num_clients = 1;
+	req.mem_cfg_mode = ab->qmi.target_mem_mode;
+	req.mem_cfg_mode_valid = 1;
+	req.bdf_support_valid = 1;
+	req.bdf_support = 1;
+
+	req.m3_support_valid = 0;
+	req.m3_support = 0;
+
+	req.m3_cache_support_valid = 0;
+	req.m3_cache_support = 0;
+
+	req.cal_done_valid = 1;
+	req.cal_done = ab->qmi.cal_done;
+
+	ret = qmi_txn_init(&ab->qmi.handle, &txn,
+			   qmi_wlanfw_host_cap_resp_msg_v01_ei, &resp);
+	if (ret < 0)
+		goto out;
+
+	ret = qmi_send_request(&ab->qmi.handle, NULL, &txn,
+			       QMI_WLANFW_HOST_CAP_REQ_V01,
+			       QMI_WLANFW_HOST_CAP_REQ_MSG_V01_MAX_LEN,
+			       qmi_wlanfw_host_cap_req_msg_v01_ei, &req);
+	if (ret < 0) {
+		ath11k_warn(ab, "Failed to send host capability request,err = %d\n", ret);
+		goto out;
+	}
+
+	ret = qmi_txn_wait(&txn, msecs_to_jiffies(ATH11K_QMI_WLANFW_TIMEOUT_MS));
+	if (ret < 0)
+		goto out;
+
+	if (resp.resp.result != QMI_RESULT_SUCCESS_V01) {
+		ath11k_warn(ab, "Host capability request failed, result: %d, err: %d\n",
+			    resp.resp.result, resp.resp.error);
+		ret = -EINVAL;
+		goto out;
+	}
+
+out:
+	return ret;
+}
+
+static int ath11k_qmi_fw_ind_register_send(struct ath11k_base *ab)
+{
+	struct qmi_wlanfw_ind_register_req_msg_v01 *req;
+	struct qmi_wlanfw_ind_register_resp_msg_v01 *resp;
+	struct qmi_handle *handle = &ab->qmi.handle;
+	struct qmi_txn txn;
+	int ret = 0;
+
+	req = kzalloc(sizeof(*req), GFP_KERNEL);
+	if (!req)
+		return -ENOMEM;
+
+	resp = kzalloc(sizeof(*resp), GFP_KERNEL);
+	if (!resp)
+		goto resp_out;
+
+	req->client_id_valid = 1;
+	req->client_id = QMI_WLANFW_CLIENT_ID;
+	req->fw_ready_enable_valid = 1;
+	req->fw_ready_enable = 1;
+	req->request_mem_enable_valid = 1;
+	req->request_mem_enable = 1;
+	req->fw_mem_ready_enable_valid = 1;
+	req->fw_mem_ready_enable = 1;
+	req->cal_done_enable_valid = 1;
+	req->cal_done_enable = 1;
+	req->fw_init_done_enable_valid = 1;
+	req->fw_init_done_enable = 1;
+
+	req->pin_connect_result_enable_valid = 0;
+	req->pin_connect_result_enable = 0;
+
+	ret = qmi_txn_init(handle, &txn,
+			   qmi_wlanfw_ind_register_resp_msg_v01_ei, resp);
+	if (ret < 0)
+		goto out;
+
+	ret = qmi_send_request(&ab->qmi.handle, NULL, &txn,
+			       QMI_WLANFW_IND_REGISTER_REQ_V01,
+			       QMI_WLANFW_IND_REGISTER_REQ_MSG_V01_MAX_LEN,
+			       qmi_wlanfw_ind_register_req_msg_v01_ei, req);
+	if (ret < 0) {
+		ath11k_warn(ab, "Failed to send indication register request, err = %d\n",
+			    ret);
+		goto out;
+	}
+
+	ret = qmi_txn_wait(&txn, msecs_to_jiffies(ATH11K_QMI_WLANFW_TIMEOUT_MS));
+	if (ret < 0) {
+		ath11k_warn(ab, "failed to register fw indication %d\n", ret);
+		goto out;
+	}
+
+	if (resp->resp.result != QMI_RESULT_SUCCESS_V01) {
+		ath11k_warn(ab, "FW Ind register request failed, result: %d, err: %d\n",
+			    resp->resp.result, resp->resp.error);
+		ret = -EINVAL;
+		goto out;
+	}
+
+out:
+	kfree(resp);
+resp_out:
+	kfree(req);
+	return ret;
+}
+
+static int ath11k_qmi_respond_fw_mem_request(struct ath11k_base *ab)
+{
+	struct qmi_wlanfw_respond_mem_req_msg_v01 *req;
+	struct qmi_wlanfw_respond_mem_resp_msg_v01 resp;
+	struct qmi_txn txn = {};
+	int ret = 0, i;
+
+	req = kzalloc(sizeof(*req), GFP_KERNEL);
+	if (!req)
+		return -ENOMEM;
+
+	memset(&resp, 0, sizeof(resp));
+
+	req->mem_seg_len = ab->qmi.mem_seg_count;
+
+	ret = qmi_txn_init(&ab->qmi.handle, &txn,
+			   qmi_wlanfw_respond_mem_resp_msg_v01_ei, &resp);
+	if (ret < 0)
+		goto out;
+
+	for (i = 0; i < req->mem_seg_len ; i++) {
+		req->mem_seg[i].addr = ab->qmi.target_mem[i].paddr;
+		req->mem_seg[i].size = ab->qmi.target_mem[i].size;
+		req->mem_seg[i].type = ab->qmi.target_mem[i].type;
+	}
+
+	ret = qmi_send_request(&ab->qmi.handle, NULL, &txn,
+			       QMI_WLANFW_RESPOND_MEM_REQ_V01,
+			       QMI_WLANFW_RESPOND_MEM_REQ_MSG_V01_MAX_LEN,
+			       qmi_wlanfw_respond_mem_req_msg_v01_ei, req);
+	if (ret < 0) {
+		ath11k_warn(ab, "qmi failed to respond memory request, err = %d\n",
+			    ret);
+		goto out;
+	}
+
+	ret = qmi_txn_wait(&txn, msecs_to_jiffies(ATH11K_QMI_WLANFW_TIMEOUT_MS));
+	if (ret < 0) {
+		ath11k_warn(ab, "qmi failed memory request, err = %d\n", ret);
+		goto out;
+	}
+
+	if (resp.resp.result != QMI_RESULT_SUCCESS_V01) {
+		ath11k_warn(ab, "Respond mem req failed, result: %d, err: %d\n",
+			    resp.resp.result, resp.resp.error);
+		ret = -EINVAL;
+		goto out;
+	}
+out:
+	kfree(req);
+	return ret;
+}
+
+static int ath11k_qmi_alloc_target_mem_chunk(struct ath11k_base *ab)
+{
+	int i, idx;
+
+	for (i = 0, idx = 0; i < ab->qmi.mem_seg_count; i++) {
+		switch (ab->qmi.target_mem[i].type) {
+		case BDF_MEM_REGION_TYPE:
+			ab->qmi.target_mem[idx].paddr = ATH11K_QMI_BDF_ADDRESS;
+			ab->qmi.target_mem[idx].vaddr = ATH11K_QMI_BDF_ADDRESS;
+			ab->qmi.target_mem[idx].size = ab->qmi.target_mem[i].size;
+			ab->qmi.target_mem[idx].type = ab->qmi.target_mem[i].type;
+			idx++;
+			break;
+		case CALDB_MEM_REGION_TYPE:
+			if (ab->qmi.target_mem[i].size > ATH11K_QMI_CALDB_SIZE) {
+				ath11k_warn(ab, "qmi mem size is low to load caldata\n");
+				return -EINVAL;
+			}
+			/* TODO ath11k does not support cold boot calibration */
+			ab->qmi.target_mem[idx].paddr = 0;
+			ab->qmi.target_mem[idx].vaddr = 0;
+			ab->qmi.target_mem[idx].size = ab->qmi.target_mem[i].size;
+			ab->qmi.target_mem[idx].type = ab->qmi.target_mem[i].type;
+			idx++;
+			break;
+		default:
+			ath11k_warn(ab, "qmi ignore invalid mem req type %d\n",
+				    ab->qmi.target_mem[i].type);
+			break;
+		}
+	}
+	ab->qmi.mem_seg_count = idx;
+
+	return 0;
+}
+
+static int ath11k_qmi_request_target_cap(struct ath11k_base *ab)
+{
+	struct qmi_wlanfw_cap_req_msg_v01 req;
+	struct qmi_wlanfw_cap_resp_msg_v01 resp;
+	struct qmi_txn txn = {};
+	int ret = 0;
+
+	memset(&req, 0, sizeof(req));
+	memset(&resp, 0, sizeof(resp));
+
+	ret = qmi_txn_init(&ab->qmi.handle, &txn,
+			   qmi_wlanfw_cap_resp_msg_v01_ei, &resp);
+	if (ret < 0)
+		goto out;
+
+	ret = qmi_send_request(&ab->qmi.handle, NULL, &txn,
+			       QMI_WLANFW_CAP_REQ_V01,
+			       QMI_WLANFW_CAP_REQ_MSG_V01_MAX_LEN,
+			       qmi_wlanfw_cap_req_msg_v01_ei, &req);
+	if (ret < 0) {
+		ath11k_warn(ab, "qmi failed to send target cap request, err = %d\n",
+			    ret);
+		goto out;
+	}
+
+	ret = qmi_txn_wait(&txn, msecs_to_jiffies(ATH11K_QMI_WLANFW_TIMEOUT_MS));
+	if (ret < 0) {
+		ath11k_warn(ab, "qmi failed target cap request %d\n", ret);
+		goto out;
+	}
+
+	if (resp.resp.result != QMI_RESULT_SUCCESS_V01) {
+		ath11k_warn(ab, "qmi targetcap req failed, result: %d, err: %d\n",
+			    resp.resp.result, resp.resp.error);
+		ret = -EINVAL;
+		goto out;
+	}
+
+	if (resp.chip_info_valid) {
+		ab->qmi.target.chip_id = resp.chip_info.chip_id;
+		ab->qmi.target.chip_family = resp.chip_info.chip_family;
+	}
+
+	if (resp.board_info_valid)
+		ab->qmi.target.board_id = resp.board_info.board_id;
+	else
+		ab->qmi.target.board_id = 0xFF;
+
+	if (resp.soc_info_valid)
+		ab->qmi.target.soc_id = resp.soc_info.soc_id;
+
+	if (resp.fw_version_info_valid) {
+		ab->qmi.target.fw_version = resp.fw_version_info.fw_version;
+		strlcpy(ab->qmi.target.fw_build_timestamp,
+			resp.fw_version_info.fw_build_timestamp,
+			sizeof(ab->qmi.target.fw_build_timestamp));
+	}
+
+	if (resp.fw_build_id_valid)
+		strlcpy(ab->qmi.target.fw_build_id, resp.fw_build_id,
+			sizeof(ab->qmi.target.fw_build_id));
+
+	ath11k_info(ab, "qmi target: chip_id: 0x%x, chip_family: 0x%x, board_id: 0x%x, soc_id: 0x%x\n",
+		    ab->qmi.target.chip_id, ab->qmi.target.chip_family,
+		    ab->qmi.target.board_id, ab->qmi.target.soc_id);
+
+	ath11k_info(ab, "qmi fw_version: 0x%x fw_build_timestamp: %s fw_build_id: %s",
+		    ab->qmi.target.fw_version,
+		    ab->qmi.target.fw_build_timestamp,
+		    ab->qmi.target.fw_build_id);
+
+out:
+	return ret;
+}
+
+static int
+ath11k_qmi_prepare_bdf_download(struct ath11k_base *ab, int type,
+				struct qmi_wlanfw_bdf_download_req_msg_v01 *req,
+				void __iomem *bdf_addr)
+{
+	struct device *dev = ab->dev;
+	char filename[ATH11K_QMI_MAX_BDF_FILE_NAME_SIZE];
+	const struct firmware *fw_entry;
+	struct ath11k_board_data bd;
+	u32 fw_size;
+	int ret = 0;
+
+	memset(&bd, 0, sizeof(bd));
+
+	switch (type) {
+	case ATH11K_QMI_FILE_TYPE_BDF_GOLDEN:
+		ret = ath11k_core_fetch_bdf(ab, &bd);
+		if (ret) {
+			ath11k_warn(ab, "qmi failed to load BDF\n");
+			goto out;
+		}
+
+		fw_size = min_t(u32, ab->hw_params.fw.board_size, bd.len);
+		memcpy_toio(bdf_addr, bd.data, fw_size);
+		ath11k_core_free_bdf(ab, &bd);
+		break;
+	case ATH11K_QMI_FILE_TYPE_CALDATA:
+		snprintf(filename, sizeof(filename),
+			 "%s/%s", ab->hw_params.fw.dir, ATH11K_QMI_DEFAULT_CAL_FILE_NAME);
+		ret = request_firmware(&fw_entry, filename, dev);
+		if (ret) {
+			ath11k_warn(ab, "qmi failed to load CAL: %s\n", filename);
+			goto out;
+		}
+
+		fw_size = min_t(u32, ab->hw_params.fw.board_size,
+				fw_entry->size);
+
+		memcpy_toio(bdf_addr + ATH11K_QMI_CALDATA_OFFSET,
+			    fw_entry->data, fw_size);
+		ath11k_info(ab, "qmi downloading BDF: %s, size: %zu\n",
+			    filename, fw_entry->size);
+
+		release_firmware(fw_entry);
+		break;
+	default:
+		ret = -EINVAL;
+		goto out;
+	}
+
+	req->total_size = fw_size;
+
+out:
+	return ret;
+}
+
+static int ath11k_qmi_load_bdf(struct ath11k_base *ab)
+{
+	struct qmi_wlanfw_bdf_download_req_msg_v01 *req;
+	struct qmi_wlanfw_bdf_download_resp_msg_v01 resp;
+	struct qmi_txn txn = {};
+	void __iomem *bdf_addr = NULL;
+	int type, ret;
+
+	req = kzalloc(sizeof(*req), GFP_KERNEL);
+	if (!req)
+		return -ENOMEM;
+	memset(&resp, 0, sizeof(resp));
+
+	bdf_addr = ioremap(ATH11K_QMI_BDF_ADDRESS, ATH11K_QMI_BDF_MAX_SIZE);
+	if (!bdf_addr) {
+		ath11k_warn(ab, "qmi ioremap error for BDF\n");
+		ret = -EIO;
+		goto out;
+	}
+
+	for (type = 0; type < ATH11K_QMI_MAX_FILE_TYPE; type++) {
+		req->valid = 1;
+		req->file_id_valid = 1;
+		req->file_id = ab->qmi.target.board_id;
+		req->total_size_valid = 1;
+		req->seg_id_valid = 1;
+		req->seg_id = type;
+		req->data_valid = 0;
+		req->data_len = ATH11K_QMI_MAX_BDF_FILE_NAME_SIZE;
+		req->bdf_type = 0;
+		req->bdf_type_valid = 0;
+		req->end_valid = 1;
+		req->end = 1;
+
+		ret = ath11k_qmi_prepare_bdf_download(ab, type, req, bdf_addr);
+		if (ret < 0)
+			goto out_qmi_bdf;
+
+		ret = qmi_txn_init(&ab->qmi.handle, &txn,
+				   qmi_wlanfw_bdf_download_resp_msg_v01_ei,
+				   &resp);
+		if (ret < 0)
+			goto out_qmi_bdf;
+
+		ret = qmi_send_request(&ab->qmi.handle, NULL, &txn,
+				       QMI_WLANFW_BDF_DOWNLOAD_REQ_V01,
+				       QMI_WLANFW_BDF_DOWNLOAD_REQ_MSG_V01_MAX_LEN,
+				       qmi_wlanfw_bdf_download_req_msg_v01_ei, req);
+		if (ret < 0) {
+			qmi_txn_cancel(&txn);
+			goto out_qmi_bdf;
+		}
+
+		ret = qmi_txn_wait(&txn, msecs_to_jiffies(ATH11K_QMI_WLANFW_TIMEOUT_MS));
+		if (ret < 0)
+			goto out_qmi_bdf;
+
+		if (resp.resp.result != QMI_RESULT_SUCCESS_V01) {
+			ath11k_warn(ab, "qmi BDF download failed, result: %d, err: %d\n",
+				    resp.resp.result, resp.resp.error);
+			ret = -EINVAL;
+			goto out_qmi_bdf;
+		}
+	}
+	ath11k_info(ab, "qmi BDF downloaded\n");
+
+out_qmi_bdf:
+	iounmap(bdf_addr);
+out:
+	kfree(req);
+	return ret;
+}
+
+static int ath11k_qmi_wlanfw_m3_info_send(struct ath11k_base *ab)
+{
+	struct qmi_wlanfw_m3_info_req_msg_v01 req;
+	struct qmi_wlanfw_m3_info_resp_msg_v01 resp;
+	struct qmi_txn txn = {};
+	int ret = 0;
+
+	memset(&req, 0, sizeof(req));
+	memset(&resp, 0, sizeof(resp));
+	req.addr = 0;
+	req.size = 0;
+
+	ret = qmi_txn_init(&ab->qmi.handle, &txn,
+			   qmi_wlanfw_m3_info_resp_msg_v01_ei, &resp);
+	if (ret < 0)
+		goto out;
+
+	ret = qmi_send_request(&ab->qmi.handle, NULL, &txn,
+			       QMI_WLANFW_M3_INFO_REQ_V01,
+			       QMI_WLANFW_M3_INFO_REQ_MSG_V01_MAX_MSG_LEN,
+			       qmi_wlanfw_m3_info_req_msg_v01_ei, &req);
+	if (ret < 0) {
+		ath11k_warn(ab, "qmi failed to send M3 information request, err = %d\n",
+			    ret);
+		goto out;
+	}
+
+	ret = qmi_txn_wait(&txn, msecs_to_jiffies(ATH11K_QMI_WLANFW_TIMEOUT_MS));
+	if (ret < 0) {
+		ath11k_warn(ab, "qmi failed M3 information request %d\n", ret);
+		goto out;
+	}
+
+	if (resp.resp.result != QMI_RESULT_SUCCESS_V01) {
+		ath11k_warn(ab, "qmi M3 info request failed, result: %d, err: %d\n",
+			    resp.resp.result, resp.resp.error);
+		ret = -EINVAL;
+		goto out;
+	}
+out:
+	return ret;
+}
+
+static int ath11k_qmi_wlanfw_mode_send(struct ath11k_base *ab,
+				       u32 mode)
+{
+	struct qmi_wlanfw_wlan_mode_req_msg_v01 req;
+	struct qmi_wlanfw_wlan_mode_resp_msg_v01 resp;
+	struct qmi_txn txn = {};
+	int ret = 0;
+
+	memset(&req, 0, sizeof(req));
+	memset(&resp, 0, sizeof(resp));
+
+	req.mode = mode;
+	req.hw_debug_valid = 1;
+	req.hw_debug = 0;
+
+	ret = qmi_txn_init(&ab->qmi.handle, &txn,
+			   qmi_wlanfw_wlan_mode_resp_msg_v01_ei, &resp);
+	if (ret < 0)
+		goto out;
+
+	ret = qmi_send_request(&ab->qmi.handle, NULL, &txn,
+			       QMI_WLANFW_WLAN_MODE_REQ_V01,
+			       QMI_WLANFW_WLAN_MODE_REQ_MSG_V01_MAX_LEN,
+			       qmi_wlanfw_wlan_mode_req_msg_v01_ei, &req);
+	if (ret < 0) {
+		ath11k_warn(ab, "qmi failed to send mode request, mode: %d, err = %d\n",
+			    mode, ret);
+		goto out;
+	}
+
+	ret = qmi_txn_wait(&txn, msecs_to_jiffies(ATH11K_QMI_WLANFW_TIMEOUT_MS));
+	if (ret < 0) {
+		if (mode == ATH11K_FIRMWARE_MODE_OFF && ret == -ENETRESET) {
+			ath11k_warn(ab, "WLFW service is dis-connected\n");
+			return 0;
+		}
+		ath11k_warn(ab, "qmi failed set mode request, mode: %d, err = %d\n",
+			    mode, ret);
+		goto out;
+	}
+
+	if (resp.resp.result != QMI_RESULT_SUCCESS_V01) {
+		ath11k_warn(ab, "Mode request failed, mode: %d, result: %d err: %d\n",
+			    mode, resp.resp.result, resp.resp.error);
+		ret = -EINVAL;
+		goto out;
+	}
+
+out:
+	return ret;
+}
+
+static int ath11k_qmi_wlanfw_wlan_cfg_send(struct ath11k_base *ab)
+{
+	struct qmi_wlanfw_wlan_cfg_req_msg_v01 *req;
+	struct qmi_wlanfw_wlan_cfg_resp_msg_v01 resp;
+	struct ce_pipe_config *ce_cfg;
+	struct service_to_pipe *svc_cfg;
+	struct qmi_txn txn = {};
+	int ret = 0;
+
+	ce_cfg	= (struct ce_pipe_config *)ab->qmi.ce_cfg.tgt_ce;
+	svc_cfg	= (struct service_to_pipe *)ab->qmi.ce_cfg.svc_to_ce_map;
+
+	req = kzalloc(sizeof(*req), GFP_KERNEL);
+	if (!req)
+		return -ENOMEM;
+
+	memset(&resp, 0, sizeof(resp));
+
+	req->host_version_valid = 1;
+	strlcpy(req->host_version, ATH11K_HOST_VERSION_STRING,
+		sizeof(req->host_version));
+
+	req->tgt_cfg_valid = 1;
+	/* This is number of CE configs */
+	req->tgt_cfg_len = ((ab->qmi.ce_cfg.tgt_ce_len) /
+				(sizeof(struct ce_pipe_config))) - 1;
+	for (ret = 0; ret <= req->tgt_cfg_len ; ret++) {
+		req->tgt_cfg[ret].pipe_num = ce_cfg[ret].pipenum;
+		req->tgt_cfg[ret].pipe_dir = ce_cfg[ret].pipedir;
+		req->tgt_cfg[ret].nentries = ce_cfg[ret].nentries;
+		req->tgt_cfg[ret].nbytes_max = ce_cfg[ret].nbytes_max;
+		req->tgt_cfg[ret].flags = ce_cfg[ret].flags;
+	}
+
+	req->svc_cfg_valid = 1;
+	/* This is number of Service/CE configs */
+	req->svc_cfg_len = (ab->qmi.ce_cfg.svc_to_ce_map_len) /
+				(sizeof(struct service_to_pipe));
+	for (ret = 0; ret < req->svc_cfg_len; ret++) {
+		req->svc_cfg[ret].service_id = svc_cfg[ret].service_id;
+		req->svc_cfg[ret].pipe_dir = svc_cfg[ret].pipedir;
+		req->svc_cfg[ret].pipe_num = svc_cfg[ret].pipenum;
+	}
+	req->shadow_reg_valid = 0;
+	req->shadow_reg_v2_valid = 0;
+
+	ret = qmi_txn_init(&ab->qmi.handle, &txn,
+			   qmi_wlanfw_wlan_cfg_resp_msg_v01_ei, &resp);
+	if (ret < 0)
+		goto out;
+
+	ret = qmi_send_request(&ab->qmi.handle, NULL, &txn,
+			       QMI_WLANFW_WLAN_CFG_REQ_V01,
+			       QMI_WLANFW_WLAN_CFG_REQ_MSG_V01_MAX_LEN,
+			       qmi_wlanfw_wlan_cfg_req_msg_v01_ei, req);
+	if (ret < 0) {
+		ath11k_warn(ab, "qmi failed to send wlan config request, err = %d\n",
+			    ret);
+		goto out;
+	}
+
+	ret = qmi_txn_wait(&txn, msecs_to_jiffies(ATH11K_QMI_WLANFW_TIMEOUT_MS));
+	if (ret < 0) {
+		ath11k_warn(ab, "qmi failed wlan config request, err = %d\n", ret);
+		goto out;
+	}
+
+	if (resp.resp.result != QMI_RESULT_SUCCESS_V01) {
+		ath11k_warn(ab, "qmi wlan config request failed, result: %d, err: %d\n",
+			    resp.resp.result, resp.resp.error);
+		ret = -EINVAL;
+		goto out;
+	}
+
+out:
+	kfree(req);
+	return ret;
+}
+
+void ath11k_qmi_firmware_stop(struct ath11k_base *ab)
+{
+	int ret;
+
+	ret = ath11k_qmi_wlanfw_mode_send(ab, ATH11K_FIRMWARE_MODE_OFF);
+	if (ret < 0) {
+		ath11k_warn(ab, "qmi failed to send wlan mode off\n");
+		return;
+	}
+}
+
+int ath11k_qmi_firmware_start(struct ath11k_base *ab,
+			      u32 mode)
+{
+	int ret;
+
+	ret = ath11k_qmi_wlanfw_wlan_cfg_send(ab);
+	if (ret < 0) {
+		ath11k_warn(ab, "qmi failed to send wlan cfg:%d\n", ret);
+		return ret;
+	}
+
+	ret = ath11k_qmi_wlanfw_mode_send(ab, mode);
+	if (ret < 0) {
+		ath11k_warn(ab, "qmi failed to send wlan fw mode:%d\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int
+ath11k_qmi_driver_event_post(struct ath11k_qmi *qmi,
+			     enum ath11k_qmi_event_type type,
+			     void *data)
+{
+	struct ath11k_qmi_driver_event *event;
+
+	event = kzalloc(sizeof(*event), GFP_ATOMIC);
+	if (!event)
+		return -ENOMEM;
+
+	event->type = type;
+	event->data = data;
+
+	spin_lock(&qmi->event_lock);
+	list_add_tail(&event->list, &qmi->event_list);
+	spin_unlock(&qmi->event_lock);
+
+	queue_work(qmi->event_wq, &qmi->event_work);
+
+	return 0;
+}
+
+static void ath11k_qmi_event_server_arrive(struct ath11k_qmi *qmi)
+{
+	struct ath11k_base *ab = qmi->ab;
+	int ret;
+
+	ret = ath11k_qmi_fw_ind_register_send(ab);
+	if (ret < 0) {
+		ath11k_warn(ab, "qmi failed to send FW indication QMI:%d\n", ret);
+		return;
+	}
+
+	ret = ath11k_qmi_host_cap_send(ab);
+	if (ret < 0) {
+		ath11k_warn(ab, "qmi failed to send host cap QMI:%d\n", ret);
+		return;
+	}
+}
+
+static void ath11k_qmi_event_mem_request(struct ath11k_qmi *qmi)
+{
+	struct ath11k_base *ab = qmi->ab;
+	int ret;
+
+	ret = ath11k_qmi_respond_fw_mem_request(ab);
+	if (ret < 0) {
+		ath11k_warn(ab, "qmi failed to respond fw mem req:%d\n", ret);
+		return;
+	}
+}
+
+static void ath11k_qmi_event_load_bdf(struct ath11k_qmi *qmi)
+{
+	struct ath11k_base *ab = qmi->ab;
+	int ret;
+
+	ret = ath11k_qmi_request_target_cap(ab);
+	if (ret < 0) {
+		ath11k_warn(ab, "qmi failed to req target capabilities:%d\n", ret);
+		return;
+	}
+
+	ret = ath11k_qmi_load_bdf(ab);
+	if (ret < 0) {
+		ath11k_warn(ab, "qmi failed to load board data file:%d\n", ret);
+		return;
+	}
+
+	ret = ath11k_qmi_wlanfw_m3_info_send(ab);
+	if (ret < 0) {
+		ath11k_warn(ab, "qmi failed to send m3 info req:%d\n", ret);
+		return;
+	}
+}
+
+static void ath11k_qmi_msg_mem_request_cb(struct qmi_handle *qmi_hdl,
+					  struct sockaddr_qrtr *sq,
+					  struct qmi_txn *txn,
+					  const void *data)
+{
+	struct ath11k_qmi *qmi = container_of(qmi_hdl, struct ath11k_qmi, handle);
+	struct ath11k_base *ab = qmi->ab;
+	const struct qmi_wlanfw_request_mem_ind_msg_v01 *msg = data;
+	int i, ret;
+
+	ath11k_dbg(ab, ATH11K_DBG_QMI, "qmi firmware request memory request\n");
+
+	if (msg->mem_seg_len == 0 ||
+	    msg->mem_seg_len > ATH11K_QMI_WLANFW_MAX_NUM_MEM_SEG_V01)
+		ath11k_warn(ab, "Invalid memory segment length: %u\n",
+			    msg->mem_seg_len);
+
+	ab->qmi.mem_seg_count = msg->mem_seg_len;
+
+	for (i = 0; i < qmi->mem_seg_count ; i++) {
+		ab->qmi.target_mem[i].type = msg->mem_seg[i].type;
+		ab->qmi.target_mem[i].size = msg->mem_seg[i].size;
+		ath11k_dbg(ab, ATH11K_DBG_QMI, "qmi mem seg type %d size %d\n",
+			   msg->mem_seg[i].type, msg->mem_seg[i].size);
+	}
+
+	ret = ath11k_qmi_alloc_target_mem_chunk(ab);
+	if (ret < 0) {
+		ath11k_warn(ab, "qmi failed to alloc target memory:%d\n", ret);
+		return;
+	}
+
+	ath11k_qmi_driver_event_post(qmi, ATH11K_QMI_EVENT_REQUEST_MEM, NULL);
+}
+
+static void ath11k_qmi_msg_mem_ready_cb(struct qmi_handle *qmi_hdl,
+					struct sockaddr_qrtr *sq,
+					struct qmi_txn *txn,
+					const void *decoded)
+{
+	struct ath11k_qmi *qmi = container_of(qmi_hdl, struct ath11k_qmi, handle);
+	struct ath11k_base *ab = qmi->ab;
+
+	ath11k_dbg(ab, ATH11K_DBG_QMI, "qmi firmware memory ready indication\n");
+	ath11k_qmi_driver_event_post(qmi, ATH11K_QMI_EVENT_FW_MEM_READY, NULL);
+}
+
+static void ath11k_qmi_msg_fw_ready_cb(struct qmi_handle *qmi_hdl,
+				       struct sockaddr_qrtr *sq,
+				       struct qmi_txn *txn,
+				       const void *decoded)
+{
+	struct ath11k_qmi *qmi = container_of(qmi_hdl, struct ath11k_qmi, handle);
+	struct ath11k_base *ab = qmi->ab;
+
+	ath11k_dbg(ab, ATH11K_DBG_QMI, "qmi firmware ready\n");
+	ath11k_qmi_driver_event_post(qmi, ATH11K_QMI_EVENT_FW_READY, NULL);
+}
+
+static void ath11k_qmi_msg_cold_boot_cal_done_cb(struct qmi_handle *qmi,
+						 struct sockaddr_qrtr *sq,
+						 struct qmi_txn *txn,
+						 const void *decoded)
+{
+}
+
+static const struct qmi_msg_handler ath11k_qmi_msg_handlers[] = {
+	{
+		.type = QMI_INDICATION,
+		.msg_id = QMI_WLFW_REQUEST_MEM_IND_V01,
+		.ei = qmi_wlanfw_request_mem_ind_msg_v01_ei,
+		.decoded_size = sizeof(qmi_wlanfw_request_mem_ind_msg_v01_ei),
+		.fn = ath11k_qmi_msg_mem_request_cb,
+	},
+	{
+		.type = QMI_INDICATION,
+		.msg_id = QMI_WLFW_FW_MEM_READY_IND_V01,
+		.ei = qmi_wlanfw_mem_ready_ind_msg_v01_ei,
+		.decoded_size = sizeof(qmi_wlanfw_mem_ready_ind_msg_v01_ei),
+		.fn = ath11k_qmi_msg_mem_ready_cb,
+	},
+	{
+		.type = QMI_INDICATION,
+		.msg_id = QMI_WLFW_FW_READY_IND_V01,
+		.ei = qmi_wlanfw_fw_ready_ind_msg_v01_ei,
+		.decoded_size = sizeof(qmi_wlanfw_fw_ready_ind_msg_v01_ei),
+		.fn = ath11k_qmi_msg_fw_ready_cb,
+	},
+	{
+		.type = QMI_INDICATION,
+		.msg_id = QMI_WLFW_COLD_BOOT_CAL_DONE_IND_V01,
+		.ei = qmi_wlanfw_cold_boot_cal_done_ind_msg_v01_ei,
+		.decoded_size =
+			sizeof(qmi_wlanfw_cold_boot_cal_done_ind_msg_v01_ei),
+		.fn = ath11k_qmi_msg_cold_boot_cal_done_cb,
+	},
+};
+
+static int ath11k_qmi_ops_new_server(struct qmi_handle *qmi_hdl,
+				     struct qmi_service *service)
+{
+	struct ath11k_qmi *qmi = container_of(qmi_hdl, struct ath11k_qmi, handle);
+	struct ath11k_base *ab = qmi->ab;
+	struct sockaddr_qrtr *sq = &qmi->sq;
+	int ret;
+
+	sq->sq_family = AF_QIPCRTR;
+	sq->sq_node = service->node;
+	sq->sq_port = service->port;
+
+	ret = kernel_connect(qmi_hdl->sock, (struct sockaddr *)sq,
+			     sizeof(*sq), 0);
+	if (ret) {
+		ath11k_warn(ab, "qmi failed to connect to remote service %d\n", ret);
+		return ret;
+	}
+
+	ath11k_dbg(ab, ATH11K_DBG_QMI, "qmi wifi fw qmi service connected\n");
+	ath11k_qmi_driver_event_post(qmi, ATH11K_QMI_EVENT_SERVER_ARRIVE, NULL);
+
+	return 0;
+}
+
+static void ath11k_qmi_ops_del_server(struct qmi_handle *qmi_hdl,
+				      struct qmi_service *service)
+{
+	struct ath11k_qmi *qmi = container_of(qmi_hdl, struct ath11k_qmi, handle);
+	struct ath11k_base *ab = qmi->ab;
+
+	ath11k_dbg(ab, ATH11K_DBG_QMI, "qmi wifi fw del server\n");
+	ath11k_qmi_driver_event_post(qmi, ATH11K_QMI_EVENT_SERVER_EXIT, NULL);
+}
+
+static const struct qmi_ops ath11k_qmi_ops = {
+	.new_server = ath11k_qmi_ops_new_server,
+	.del_server = ath11k_qmi_ops_del_server,
+};
+
+static void ath11k_qmi_driver_event_work(struct work_struct *work)
+{
+	struct ath11k_qmi *qmi = container_of(work, struct ath11k_qmi,
+					      event_work);
+	struct ath11k_qmi_driver_event *event;
+	struct ath11k_base *ab = qmi->ab;
+
+	spin_lock(&qmi->event_lock);
+	while (!list_empty(&qmi->event_list)) {
+		event = list_first_entry(&qmi->event_list,
+					 struct ath11k_qmi_driver_event, list);
+		list_del(&event->list);
+		spin_unlock(&qmi->event_lock);
+
+		if (test_bit(ATH11K_FLAG_UNREGISTERING, &ab->dev_flags))
+			return;
+
+		switch (event->type) {
+		case ATH11K_QMI_EVENT_SERVER_ARRIVE:
+			ath11k_qmi_event_server_arrive(qmi);
+			break;
+		case ATH11K_QMI_EVENT_SERVER_EXIT:
+			set_bit(ATH11K_FLAG_CRASH_FLUSH, &ab->dev_flags);
+			set_bit(ATH11K_FLAG_RECOVERY, &ab->dev_flags);
+			break;
+		case ATH11K_QMI_EVENT_REQUEST_MEM:
+			ath11k_qmi_event_mem_request(qmi);
+			break;
+		case ATH11K_QMI_EVENT_FW_MEM_READY:
+			ath11k_qmi_event_load_bdf(qmi);
+			break;
+		case ATH11K_QMI_EVENT_FW_READY:
+			if (test_bit(ATH11K_FLAG_REGISTERED, &ab->dev_flags)) {
+				queue_work(ab->workqueue, &ab->restart_work);
+				break;
+			}
+
+			ath11k_core_qmi_firmware_ready(ab);
+			ab->qmi.cal_done = 1;
+			set_bit(ATH11K_FLAG_REGISTERED, &ab->dev_flags);
+
+			break;
+		case ATH11K_QMI_EVENT_COLD_BOOT_CAL_DONE:
+			break;
+		default:
+			ath11k_warn(ab, "invalid event type: %d", event->type);
+			break;
+		}
+		kfree(event);
+		spin_lock(&qmi->event_lock);
+	}
+	spin_unlock(&qmi->event_lock);
+}
+
+int ath11k_qmi_init_service(struct ath11k_base *ab)
+{
+	int ret;
+
+	memset(&ab->qmi.target, 0, sizeof(struct target_info));
+	memset(&ab->qmi.target_mem, 0, sizeof(struct target_mem_chunk));
+	ab->qmi.ab = ab;
+
+	ab->qmi.target_mem_mode = ATH11K_QMI_TARGET_MEM_MODE_DEFAULT;
+	ret = qmi_handle_init(&ab->qmi.handle, ATH11K_QMI_RESP_LEN_MAX,
+			      &ath11k_qmi_ops, ath11k_qmi_msg_handlers);
+	if (ret < 0) {
+		ath11k_warn(ab, "failed to initialize qmi handle\n");
+		return ret;
+	}
+
+	ab->qmi.event_wq = alloc_workqueue("ath11k_qmi_driver_event",
+					   WQ_UNBOUND, 1);
+	if (!ab->qmi.event_wq) {
+		ath11k_err(ab, "failed to allocate workqueue\n");
+		return -EFAULT;
+	}
+
+	INIT_LIST_HEAD(&ab->qmi.event_list);
+	spin_lock_init(&ab->qmi.event_lock);
+	INIT_WORK(&ab->qmi.event_work, ath11k_qmi_driver_event_work);
+
+	ret = qmi_add_lookup(&ab->qmi.handle, ATH11K_QMI_WLFW_SERVICE_ID_V01,
+			     ATH11K_QMI_WLFW_SERVICE_VERS_V01,
+			     ATH11K_QMI_WLFW_SERVICE_INS_ID_V01);
+	if (ret < 0) {
+		ath11k_warn(ab, "failed to add qmi lookup\n");
+		return ret;
+	}
+
+	return ret;
+}
+
+void ath11k_qmi_deinit_service(struct ath11k_base *ab)
+{
+	qmi_handle_release(&ab->qmi.handle);
+	cancel_work_sync(&ab->qmi.event_work);
+	destroy_workqueue(ab->qmi.event_wq);
+}
+
diff --git a/drivers/net/wireless/ath/ath11k/qmi.h b/drivers/net/wireless/ath/ath11k/qmi.h
new file mode 100644
index 000000000000..df0685564e21
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/qmi.h
@@ -0,0 +1,445 @@
+/* SPDX-License-Identifier: BSD-3-Clause-Clear */
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#ifndef ATH11K_QMI_H
+#define ATH11K_QMI_H
+
+#include <linux/mutex.h>
+#include <linux/soc/qcom/qmi.h>
+
+#define ATH11K_HOST_VERSION_STRING		"WIN"
+#define ATH11K_QMI_WLANFW_TIMEOUT_MS		5000
+#define ATH11K_QMI_MAX_BDF_FILE_NAME_SIZE	64
+#define ATH11K_QMI_BDF_ADDRESS			0x4B0C0000
+#define ATH11K_QMI_BDF_MAX_SIZE			(256 * 1024)
+#define ATH11K_QMI_CALDATA_OFFSET		(128 * 1024)
+#define ATH11K_QMI_WLANFW_MAX_BUILD_ID_LEN_V01	128
+#define ATH11K_QMI_WLFW_SERVICE_ID_V01		0x45
+#define ATH11K_QMI_WLFW_SERVICE_VERS_V01	0x01
+#define ATH11K_QMI_WLFW_SERVICE_INS_ID_V01	0x02
+#define ATH11K_QMI_WLANFW_MAX_TIMESTAMP_LEN_V01	32
+#define ATH11K_QMI_RESP_LEN_MAX			8192
+#define ATH11K_QMI_WLANFW_MAX_NUM_MEM_SEG_V01	32
+#define ATH11K_QMI_CALDB_SIZE			0x480000
+#define ATH11K_QMI_DEFAULT_CAL_FILE_NAME	"caldata.bin"
+
+#define QMI_WLFW_REQUEST_MEM_IND_V01		0x0035
+#define QMI_WLFW_FW_MEM_READY_IND_V01		0x0037
+#define QMI_WLFW_COLD_BOOT_CAL_DONE_IND_V01	0x0021
+#define QMI_WLFW_FW_READY_IND_V01		0x0038
+
+#define QMI_WLANFW_MAX_DATA_SIZE_V01		6144
+#define ATH11K_FIRMWARE_MODE_OFF		4
+#define ATH11K_QMI_TARGET_MEM_MODE_DEFAULT	0
+
+struct ath11k_base;
+
+enum ath11k_qmi_file_type {
+	ATH11K_QMI_FILE_TYPE_BDF_GOLDEN,
+	ATH11K_QMI_FILE_TYPE_CALDATA,
+	ATH11K_QMI_MAX_FILE_TYPE,
+};
+
+enum ath11k_qmi_event_type {
+	ATH11K_QMI_EVENT_SERVER_ARRIVE,
+	ATH11K_QMI_EVENT_SERVER_EXIT,
+	ATH11K_QMI_EVENT_REQUEST_MEM,
+	ATH11K_QMI_EVENT_FW_MEM_READY,
+	ATH11K_QMI_EVENT_FW_READY,
+	ATH11K_QMI_EVENT_COLD_BOOT_CAL_START,
+	ATH11K_QMI_EVENT_COLD_BOOT_CAL_DONE,
+	ATH11K_QMI_EVENT_REGISTER_DRIVER,
+	ATH11K_QMI_EVENT_UNREGISTER_DRIVER,
+	ATH11K_QMI_EVENT_RECOVERY,
+	ATH11K_QMI_EVENT_FORCE_FW_ASSERT,
+	ATH11K_QMI_EVENT_POWER_UP,
+	ATH11K_QMI_EVENT_POWER_DOWN,
+	ATH11K_QMI_EVENT_MAX,
+};
+
+struct ath11k_qmi_driver_event {
+	struct list_head list;
+	enum ath11k_qmi_event_type type;
+	void *data;
+};
+
+struct ath11k_qmi_ce_cfg {
+	const u8 *tgt_ce;
+	int tgt_ce_len;
+	const u8 *svc_to_ce_map;
+	int svc_to_ce_map_len;
+	const u8 *shadow_reg;
+	int shadow_reg_len;
+	u8 *shadow_reg_v2;
+	int shadow_reg_v2_len;
+};
+
+struct ath11k_qmi_event_msg {
+	struct list_head list;
+	enum ath11k_qmi_event_type type;
+};
+
+struct target_mem_chunk {
+	u32 size;
+	u32 type;
+	dma_addr_t paddr;
+	u32 vaddr;
+};
+
+struct target_info {
+	u32 chip_id;
+	u32 chip_family;
+	u32 board_id;
+	u32 soc_id;
+	u32 fw_version;
+	char fw_build_timestamp[ATH11K_QMI_WLANFW_MAX_TIMESTAMP_LEN_V01 + 1];
+	char fw_build_id[ATH11K_QMI_WLANFW_MAX_BUILD_ID_LEN_V01 + 1];
+};
+
+struct ath11k_qmi {
+	struct ath11k_base *ab;
+	struct qmi_handle handle;
+	struct sockaddr_qrtr sq;
+	struct work_struct event_work;
+	struct workqueue_struct *event_wq;
+	struct list_head event_list;
+	spinlock_t event_lock; /* spinlock for qmi event list */
+	struct ath11k_qmi_ce_cfg ce_cfg;
+	struct target_mem_chunk target_mem[ATH11K_QMI_WLANFW_MAX_NUM_MEM_SEG_V01];
+	u32 mem_seg_count;
+	u32 target_mem_mode;
+	u8 cal_done;
+	struct target_info target;
+};
+
+#define QMI_WLANFW_HOST_CAP_REQ_MSG_V01_MAX_LEN		189
+#define QMI_WLANFW_HOST_CAP_REQ_V01			0x0034
+#define QMI_WLANFW_HOST_CAP_RESP_MSG_V01_MAX_LEN	7
+#define QMI_WLFW_HOST_CAP_RESP_V01			0x0034
+#define QMI_WLFW_MAX_NUM_GPIO_V01			32
+#define QMI_IPQ8074_FW_MEM_MODE				0xFF
+#define HOST_DDR_REGION_TYPE				0x1
+#define BDF_MEM_REGION_TYPE				0x2
+#define CALDB_MEM_REGION_TYPE				0x4
+
+struct qmi_wlanfw_host_cap_req_msg_v01 {
+	u8 num_clients_valid;
+	u32 num_clients;
+	u8 wake_msi_valid;
+	u32 wake_msi;
+	u8 gpios_valid;
+	u32 gpios_len;
+	u32 gpios[QMI_WLFW_MAX_NUM_GPIO_V01];
+	u8 nm_modem_valid;
+	u8 nm_modem;
+	u8 bdf_support_valid;
+	u8 bdf_support;
+	u8 bdf_cache_support_valid;
+	u8 bdf_cache_support;
+	u8 m3_support_valid;
+	u8 m3_support;
+	u8 m3_cache_support_valid;
+	u8 m3_cache_support;
+	u8 cal_filesys_support_valid;
+	u8 cal_filesys_support;
+	u8 cal_cache_support_valid;
+	u8 cal_cache_support;
+	u8 cal_done_valid;
+	u8 cal_done;
+	u8 mem_bucket_valid;
+	u32 mem_bucket;
+	u8 mem_cfg_mode_valid;
+	u8 mem_cfg_mode;
+};
+
+struct qmi_wlanfw_host_cap_resp_msg_v01 {
+	struct qmi_response_type_v01 resp;
+};
+
+#define QMI_WLANFW_IND_REGISTER_REQ_MSG_V01_MAX_LEN		54
+#define QMI_WLANFW_IND_REGISTER_REQ_V01				0x0020
+#define QMI_WLANFW_IND_REGISTER_RESP_MSG_V01_MAX_LEN		18
+#define QMI_WLANFW_IND_REGISTER_RESP_V01			0x0020
+#define QMI_WLANFW_CLIENT_ID					0x4b4e454c
+
+struct qmi_wlanfw_ind_register_req_msg_v01 {
+	u8 fw_ready_enable_valid;
+	u8 fw_ready_enable;
+	u8 initiate_cal_download_enable_valid;
+	u8 initiate_cal_download_enable;
+	u8 initiate_cal_update_enable_valid;
+	u8 initiate_cal_update_enable;
+	u8 msa_ready_enable_valid;
+	u8 msa_ready_enable;
+	u8 pin_connect_result_enable_valid;
+	u8 pin_connect_result_enable;
+	u8 client_id_valid;
+	u32 client_id;
+	u8 request_mem_enable_valid;
+	u8 request_mem_enable;
+	u8 fw_mem_ready_enable_valid;
+	u8 fw_mem_ready_enable;
+	u8 fw_init_done_enable_valid;
+	u8 fw_init_done_enable;
+	u8 rejuvenate_enable_valid;
+	u32 rejuvenate_enable;
+	u8 xo_cal_enable_valid;
+	u8 xo_cal_enable;
+	u8 cal_done_enable_valid;
+	u8 cal_done_enable;
+};
+
+struct qmi_wlanfw_ind_register_resp_msg_v01 {
+	struct qmi_response_type_v01 resp;
+	u8 fw_status_valid;
+	u64 fw_status;
+};
+
+#define QMI_WLANFW_REQUEST_MEM_IND_MSG_V01_MAX_LEN	1124
+#define QMI_WLANFW_RESPOND_MEM_REQ_MSG_V01_MAX_LEN	548
+#define QMI_WLANFW_RESPOND_MEM_RESP_MSG_V01_MAX_LEN	7
+#define QMI_WLANFW_REQUEST_MEM_IND_V01			0x0035
+#define QMI_WLANFW_RESPOND_MEM_REQ_V01			0x0036
+#define QMI_WLANFW_RESPOND_MEM_RESP_V01			0x0036
+#define QMI_WLANFW_MAX_NUM_MEM_CFG_V01			2
+
+struct qmi_wlanfw_mem_cfg_s_v01 {
+	u64 offset;
+	u32 size;
+	u8 secure_flag;
+};
+
+enum qmi_wlanfw_mem_type_enum_v01 {
+	WLANFW_MEM_TYPE_ENUM_MIN_VAL_V01 = INT_MIN,
+	QMI_WLANFW_MEM_TYPE_MSA_V01 = 0,
+	QMI_WLANFW_MEM_TYPE_DDR_V01 = 1,
+	QMI_WLANFW_MEM_BDF_V01 = 2,
+	QMI_WLANFW_MEM_M3_V01 = 3,
+	QMI_WLANFW_MEM_CAL_V01 = 4,
+	QMI_WLANFW_MEM_DPD_V01 = 5,
+	WLANFW_MEM_TYPE_ENUM_MAX_VAL_V01 = INT_MAX,
+};
+
+struct qmi_wlanfw_mem_seg_s_v01 {
+	u32 size;
+	enum qmi_wlanfw_mem_type_enum_v01 type;
+	u32 mem_cfg_len;
+	struct qmi_wlanfw_mem_cfg_s_v01 mem_cfg[QMI_WLANFW_MAX_NUM_MEM_CFG_V01];
+};
+
+struct qmi_wlanfw_request_mem_ind_msg_v01 {
+	u32 mem_seg_len;
+	struct qmi_wlanfw_mem_seg_s_v01 mem_seg[ATH11K_QMI_WLANFW_MAX_NUM_MEM_SEG_V01];
+};
+
+struct qmi_wlanfw_mem_seg_resp_s_v01 {
+	u64 addr;
+	u32 size;
+	enum qmi_wlanfw_mem_type_enum_v01 type;
+	u8 restore;
+};
+
+struct qmi_wlanfw_respond_mem_req_msg_v01 {
+	u32 mem_seg_len;
+	struct qmi_wlanfw_mem_seg_resp_s_v01 mem_seg[ATH11K_QMI_WLANFW_MAX_NUM_MEM_SEG_V01];
+};
+
+struct qmi_wlanfw_respond_mem_resp_msg_v01 {
+	struct qmi_response_type_v01 resp;
+};
+
+struct qmi_wlanfw_fw_mem_ready_ind_msg_v01 {
+	char placeholder;
+};
+
+#define QMI_WLANFW_CAP_REQ_MSG_V01_MAX_LEN	0
+#define QMI_WLANFW_CAP_RESP_MSG_V01_MAX_LEN	207
+#define QMI_WLANFW_CAP_REQ_V01			0x0024
+#define QMI_WLANFW_CAP_RESP_V01			0x0024
+
+enum qmi_wlanfw_pipedir_enum_v01 {
+	QMI_WLFW_PIPEDIR_NONE_V01 = 0,
+	QMI_WLFW_PIPEDIR_IN_V01 = 1,
+	QMI_WLFW_PIPEDIR_OUT_V01 = 2,
+	QMI_WLFW_PIPEDIR_INOUT_V01 = 3,
+};
+
+struct qmi_wlanfw_ce_tgt_pipe_cfg_s_v01 {
+	__le32 pipe_num;
+	__le32 pipe_dir;
+	__le32 nentries;
+	__le32 nbytes_max;
+	__le32 flags;
+};
+
+struct qmi_wlanfw_ce_svc_pipe_cfg_s_v01 {
+	__le32 service_id;
+	__le32 pipe_dir;
+	__le32 pipe_num;
+};
+
+struct qmi_wlanfw_shadow_reg_cfg_s_v01 {
+	u16 id;
+	u16 offset;
+};
+
+struct qmi_wlanfw_shadow_reg_v2_cfg_s_v01 {
+	u32 addr;
+};
+
+struct qmi_wlanfw_memory_region_info_s_v01 {
+	u64 region_addr;
+	u32 size;
+	u8 secure_flag;
+};
+
+struct qmi_wlanfw_rf_chip_info_s_v01 {
+	u32 chip_id;
+	u32 chip_family;
+};
+
+struct qmi_wlanfw_rf_board_info_s_v01 {
+	u32 board_id;
+};
+
+struct qmi_wlanfw_soc_info_s_v01 {
+	u32 soc_id;
+};
+
+struct qmi_wlanfw_fw_version_info_s_v01 {
+	u32 fw_version;
+	char fw_build_timestamp[ATH11K_QMI_WLANFW_MAX_TIMESTAMP_LEN_V01 + 1];
+};
+
+enum qmi_wlanfw_cal_temp_id_enum_v01 {
+	QMI_WLANFW_CAL_TEMP_IDX_0_V01 = 0,
+	QMI_WLANFW_CAL_TEMP_IDX_1_V01 = 1,
+	QMI_WLANFW_CAL_TEMP_IDX_2_V01 = 2,
+	QMI_WLANFW_CAL_TEMP_IDX_3_V01 = 3,
+	QMI_WLANFW_CAL_TEMP_IDX_4_V01 = 4,
+	QMI_WLANFW_CAL_TEMP_ID_MAX_V01 = 0xFF,
+};
+
+struct qmi_wlanfw_cap_resp_msg_v01 {
+	struct qmi_response_type_v01 resp;
+	u8 chip_info_valid;
+	struct qmi_wlanfw_rf_chip_info_s_v01 chip_info;
+	u8 board_info_valid;
+	struct qmi_wlanfw_rf_board_info_s_v01 board_info;
+	u8 soc_info_valid;
+	struct qmi_wlanfw_soc_info_s_v01 soc_info;
+	u8 fw_version_info_valid;
+	struct qmi_wlanfw_fw_version_info_s_v01 fw_version_info;
+	u8 fw_build_id_valid;
+	char fw_build_id[ATH11K_QMI_WLANFW_MAX_BUILD_ID_LEN_V01 + 1];
+	u8 num_macs_valid;
+	u8 num_macs;
+};
+
+struct qmi_wlanfw_cap_req_msg_v01 {
+	char placeholder;
+};
+
+#define QMI_WLANFW_BDF_DOWNLOAD_REQ_MSG_V01_MAX_LEN	6182
+#define QMI_WLANFW_BDF_DOWNLOAD_RESP_MSG_V01_MAX_LEN	7
+#define QMI_WLANFW_BDF_DOWNLOAD_RESP_V01		0x0025
+#define QMI_WLANFW_BDF_DOWNLOAD_REQ_V01			0x0025
+/* TODO: Need to check with MCL and FW team that data can be pointer and
+ * can be last element in structure
+ */
+struct qmi_wlanfw_bdf_download_req_msg_v01 {
+	u8 valid;
+	u8 file_id_valid;
+	enum qmi_wlanfw_cal_temp_id_enum_v01 file_id;
+	u8 total_size_valid;
+	u32 total_size;
+	u8 seg_id_valid;
+	u32 seg_id;
+	u8 data_valid;
+	u32 data_len;
+	u8 data[QMI_WLANFW_MAX_DATA_SIZE_V01];
+	u8 end_valid;
+	u8 end;
+	u8 bdf_type_valid;
+	u8 bdf_type;
+
+};
+
+struct qmi_wlanfw_bdf_download_resp_msg_v01 {
+	struct qmi_response_type_v01 resp;
+};
+
+#define QMI_WLANFW_M3_INFO_REQ_MSG_V01_MAX_MSG_LEN	18
+#define QMI_WLANFW_M3_INFO_RESP_MSG_V01_MAX_MSG_LEN	7
+#define QMI_WLANFW_M3_INFO_RESP_V01		0x003C
+#define QMI_WLANFW_M3_INFO_REQ_V01		0x003C
+
+struct qmi_wlanfw_m3_info_req_msg_v01 {
+	u64 addr;
+	u32 size;
+};
+
+struct qmi_wlanfw_m3_info_resp_msg_v01 {
+	struct qmi_response_type_v01 resp;
+};
+
+#define QMI_WLANFW_WLAN_MODE_REQ_MSG_V01_MAX_LEN	11
+#define QMI_WLANFW_WLAN_MODE_RESP_MSG_V01_MAX_LEN	7
+#define QMI_WLANFW_WLAN_CFG_REQ_MSG_V01_MAX_LEN		803
+#define QMI_WLANFW_WLAN_CFG_RESP_MSG_V01_MAX_LEN	7
+#define QMI_WLANFW_WLAN_MODE_REQ_V01			0x0022
+#define QMI_WLANFW_WLAN_MODE_RESP_V01			0x0022
+#define QMI_WLANFW_WLAN_CFG_REQ_V01			0x0023
+#define QMI_WLANFW_WLAN_CFG_RESP_V01			0x0023
+#define QMI_WLANFW_MAX_STR_LEN_V01			16
+#define QMI_WLANFW_MAX_NUM_CE_V01			12
+#define QMI_WLANFW_MAX_NUM_SVC_V01			24
+#define QMI_WLANFW_MAX_NUM_SHADOW_REG_V01		24
+#define QMI_WLANFW_MAX_NUM_SHADOW_REG_V2_V01		36
+
+struct qmi_wlanfw_wlan_mode_req_msg_v01 {
+	u32 mode;
+	u8 hw_debug_valid;
+	u8 hw_debug;
+};
+
+struct qmi_wlanfw_wlan_mode_resp_msg_v01 {
+	struct qmi_response_type_v01 resp;
+};
+
+struct qmi_wlanfw_wlan_cfg_req_msg_v01 {
+	u8 host_version_valid;
+	char host_version[QMI_WLANFW_MAX_STR_LEN_V01 + 1];
+	u8  tgt_cfg_valid;
+	u32  tgt_cfg_len;
+	struct qmi_wlanfw_ce_tgt_pipe_cfg_s_v01
+			tgt_cfg[QMI_WLANFW_MAX_NUM_CE_V01];
+	u8  svc_cfg_valid;
+	u32 svc_cfg_len;
+	struct qmi_wlanfw_ce_svc_pipe_cfg_s_v01
+			svc_cfg[QMI_WLANFW_MAX_NUM_SVC_V01];
+	u8 shadow_reg_valid;
+	u32 shadow_reg_len;
+	struct qmi_wlanfw_shadow_reg_cfg_s_v01
+		shadow_reg[QMI_WLANFW_MAX_NUM_SHADOW_REG_V01];
+	u8 shadow_reg_v2_valid;
+	u32 shadow_reg_v2_len;
+	struct qmi_wlanfw_shadow_reg_v2_cfg_s_v01
+		shadow_reg_v2[QMI_WLANFW_MAX_NUM_SHADOW_REG_V2_V01];
+};
+
+struct qmi_wlanfw_wlan_cfg_resp_msg_v01 {
+	struct qmi_response_type_v01 resp;
+};
+
+int ath11k_qmi_firmware_start(struct ath11k_base *ab,
+			      u32 mode);
+void ath11k_qmi_firmware_stop(struct ath11k_base *ab);
+void ath11k_qmi_event_work(struct work_struct *work);
+void ath11k_qmi_msg_recv_work(struct work_struct *work);
+void ath11k_qmi_deinit_service(struct ath11k_base *ab);
+int ath11k_qmi_init_service(struct ath11k_base *ab);
+
+#endif
diff --git a/drivers/net/wireless/ath/ath11k/reg.c b/drivers/net/wireless/ath/ath11k/reg.c
new file mode 100644
index 000000000000..810050b033c4
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/reg.c
@@ -0,0 +1,701 @@
+// SPDX-License-Identifier: BSD-3-Clause-Clear
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+#include "core.h"
+#include "debug.h"
+
+/* World regdom to be used in case default regd from fw is unavailable */
+#define ATH11K_2GHZ_CH01_11      REG_RULE(2412 - 10, 2462 + 10, 40, 0, 20, 0)
+#define ATH11K_5GHZ_5150_5350    REG_RULE(5150 - 10, 5350 + 10, 80, 0, 30,\
+					  NL80211_RRF_NO_IR)
+#define ATH11K_5GHZ_5725_5850    REG_RULE(5725 - 10, 5850 + 10, 80, 0, 30,\
+					  NL80211_RRF_NO_IR)
+
+#define ETSI_WEATHER_RADAR_BAND_LOW		5590
+#define ETSI_WEATHER_RADAR_BAND_HIGH		5650
+#define ETSI_WEATHER_RADAR_BAND_CAC_TIMEOUT	600000
+
+static const struct ieee80211_regdomain ath11k_world_regd = {
+	.n_reg_rules = 3,
+	.alpha2 =  "00",
+	.reg_rules = {
+		ATH11K_2GHZ_CH01_11,
+		ATH11K_5GHZ_5150_5350,
+		ATH11K_5GHZ_5725_5850,
+	}
+};
+
+static bool ath11k_regdom_changes(struct ath11k *ar, char *alpha2)
+{
+	const struct ieee80211_regdomain *regd;
+
+	regd = rcu_dereference_rtnl(ar->hw->wiphy->regd);
+	/* This can happen during wiphy registration where the previous
+	 * user request is received before we update the regd received
+	 * from firmware.
+	 */
+	if (!regd)
+		return true;
+
+	return memcmp(regd->alpha2, alpha2, 2) != 0;
+}
+
+static void
+ath11k_reg_notifier(struct wiphy *wiphy, struct regulatory_request *request)
+{
+	struct ieee80211_hw *hw = wiphy_to_ieee80211_hw(wiphy);
+	struct wmi_init_country_params init_country_param;
+	struct ath11k *ar = hw->priv;
+	int ret;
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_REG,
+		   "Regulatory Notification received for %s\n", wiphy_name(wiphy));
+
+	/* Currently supporting only General User Hints. Cell base user
+	 * hints to be handled later.
+	 * Hints from other sources like Core, Beacons are not expected for
+	 * self managed wiphy's
+	 */
+	if (!(request->initiator == NL80211_REGDOM_SET_BY_USER &&
+	      request->user_reg_hint_type == NL80211_USER_REG_HINT_USER)) {
+		ath11k_warn(ar->ab, "Unexpected Regulatory event for this wiphy\n");
+		return;
+	}
+
+	if (!IS_ENABLED(CONFIG_ATH_REG_DYNAMIC_USER_REG_HINTS)) {
+		ath11k_dbg(ar->ab, ATH11K_DBG_REG,
+			   "Country Setting is not allowed\n");
+		return;
+	}
+
+	if (!ath11k_regdom_changes(ar, request->alpha2)) {
+		ath11k_dbg(ar->ab, ATH11K_DBG_REG, "Country is already set\n");
+		return;
+	}
+
+	/* Set the country code to the firmware and wait for
+	 * the WMI_REG_CHAN_LIST_CC EVENT for updating the
+	 * reg info
+	 */
+	init_country_param.flags = ALPHA_IS_SET;
+	memcpy(&init_country_param.cc_info.alpha2, request->alpha2, 2);
+
+	ret = ath11k_wmi_send_init_country_cmd(ar, init_country_param);
+	if (ret)
+		ath11k_warn(ar->ab,
+			    "INIT Country code set to fw failed : %d\n", ret);
+}
+
+int ath11k_reg_update_chan_list(struct ath11k *ar)
+{
+	struct ieee80211_supported_band **bands;
+	struct scan_chan_list_params *params;
+	struct ieee80211_channel *channel;
+	struct ieee80211_hw *hw = ar->hw;
+	struct channel_param *ch;
+	enum nl80211_band band;
+	int num_channels = 0;
+	int params_len;
+	int i, ret;
+
+	bands = hw->wiphy->bands;
+	for (band = 0; band < NUM_NL80211_BANDS; band++) {
+		if (!bands[band])
+			continue;
+
+		for (i = 0; i < bands[band]->n_channels; i++) {
+			if (bands[band]->channels[i].flags &
+			    IEEE80211_CHAN_DISABLED)
+				continue;
+
+			num_channels++;
+		}
+	}
+
+	if (WARN_ON(!num_channels))
+		return -EINVAL;
+
+	params_len = sizeof(struct scan_chan_list_params) +
+			num_channels * sizeof(struct channel_param);
+	params = kzalloc(params_len, GFP_KERNEL);
+
+	if (!params)
+		return -ENOMEM;
+
+	params->pdev_id = ar->pdev->pdev_id;
+	params->nallchans = num_channels;
+
+	ch = params->ch_param;
+
+	for (band = 0; band < NUM_NL80211_BANDS; band++) {
+		if (!bands[band])
+			continue;
+
+		for (i = 0; i < bands[band]->n_channels; i++) {
+			channel = &bands[band]->channels[i];
+
+			if (channel->flags & IEEE80211_CHAN_DISABLED)
+				continue;
+
+			/* TODO: Set to true/false based on some condition? */
+			ch->allow_ht = true;
+			ch->allow_vht = true;
+
+			ch->dfs_set =
+				!!(channel->flags & IEEE80211_CHAN_RADAR);
+			ch->is_chan_passive = !!(channel->flags &
+						IEEE80211_CHAN_NO_IR);
+			ch->is_chan_passive |= ch->dfs_set;
+			ch->mhz = channel->center_freq;
+			ch->cfreq1 = channel->center_freq;
+			ch->minpower = 0;
+			ch->maxpower = channel->max_power * 2;
+			ch->maxregpower = channel->max_reg_power * 2;
+			ch->antennamax = channel->max_antenna_gain * 2;
+
+			/* TODO: Use appropriate phymodes */
+			if (channel->band == NL80211_BAND_2GHZ)
+				ch->phy_mode = MODE_11G;
+			else
+				ch->phy_mode = MODE_11A;
+
+			ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+				   "mac channel [%d/%d] freq %d maxpower %d regpower %d antenna %d mode %d\n",
+				   i, params->nallchans,
+				   ch->mhz, ch->maxpower, ch->maxregpower,
+				   ch->antennamax, ch->phy_mode);
+
+			ch++;
+			/* TODO: use quarrter/half rate, cfreq12, dfs_cfreq2
+			 * set_agile, reg_class_idx
+			 */
+		}
+	}
+
+	ret = ath11k_wmi_send_scan_chan_list_cmd(ar, params);
+	kfree(params);
+
+	return ret;
+}
+
+static void ath11k_copy_regd(struct ieee80211_regdomain *regd_orig,
+			     struct ieee80211_regdomain *regd_copy)
+{
+	u8 i;
+
+	/* The caller should have checked error conditions */
+	memcpy(regd_copy, regd_orig, sizeof(*regd_orig));
+
+	for (i = 0; i < regd_orig->n_reg_rules; i++)
+		memcpy(&regd_copy->reg_rules[i], &regd_orig->reg_rules[i],
+		       sizeof(struct ieee80211_reg_rule));
+}
+
+int ath11k_regd_update(struct ath11k *ar, bool init)
+{
+	struct ieee80211_regdomain *regd, *regd_copy = NULL;
+	int ret, regd_len, pdev_id;
+	struct ath11k_base *ab;
+
+	ab = ar->ab;
+	pdev_id = ar->pdev_idx;
+
+	spin_lock(&ab->base_lock);
+
+	if (init) {
+		/* Apply the regd received during init through
+		 * WMI_REG_CHAN_LIST_CC event. In case of failure to
+		 * receive the regd, initialize with a default world
+		 * regulatory.
+		 */
+		if (ab->default_regd[pdev_id]) {
+			regd = ab->default_regd[pdev_id];
+		} else {
+			ath11k_warn(ab,
+				    "failed to receive default regd during init\n");
+			regd = (struct ieee80211_regdomain *)&ath11k_world_regd;
+		}
+	} else {
+		regd = ab->new_regd[pdev_id];
+	}
+
+	if (!regd) {
+		ret = -EINVAL;
+		spin_unlock(&ab->base_lock);
+		goto err;
+	}
+
+	regd_len = sizeof(*regd) + (regd->n_reg_rules *
+		sizeof(struct ieee80211_reg_rule));
+
+	regd_copy = kzalloc(regd_len, GFP_ATOMIC);
+	if (regd_copy)
+		ath11k_copy_regd(regd, regd_copy);
+
+	spin_unlock(&ab->base_lock);
+
+	if (!regd_copy) {
+		ret = -ENOMEM;
+		goto err;
+	}
+
+	rtnl_lock();
+	ret = regulatory_set_wiphy_regd_sync_rtnl(ar->hw->wiphy, regd_copy);
+	rtnl_unlock();
+
+	kfree(regd_copy);
+
+	if (ret)
+		goto err;
+
+	if (ar->state == ATH11K_STATE_ON) {
+		ret = ath11k_reg_update_chan_list(ar);
+		if (ret)
+			goto err;
+	}
+
+	return 0;
+err:
+	ath11k_warn(ab, "failed to perform regd update : %d\n", ret);
+	return ret;
+}
+
+static enum nl80211_dfs_regions
+ath11k_map_fw_dfs_region(enum ath11k_dfs_region dfs_region)
+{
+	switch (dfs_region) {
+	case ATH11K_DFS_REG_FCC:
+	case ATH11K_DFS_REG_CN:
+		return NL80211_DFS_FCC;
+	case ATH11K_DFS_REG_ETSI:
+	case ATH11K_DFS_REG_KR:
+		return NL80211_DFS_ETSI;
+	case ATH11K_DFS_REG_MKK:
+		return NL80211_DFS_JP;
+	default:
+		return NL80211_DFS_UNSET;
+	}
+}
+
+static u32 ath11k_map_fw_reg_flags(u16 reg_flags)
+{
+	u32 flags = 0;
+
+	if (reg_flags & REGULATORY_CHAN_NO_IR)
+		flags = NL80211_RRF_NO_IR;
+
+	if (reg_flags & REGULATORY_CHAN_RADAR)
+		flags |= NL80211_RRF_DFS;
+
+	if (reg_flags & REGULATORY_CHAN_NO_OFDM)
+		flags |= NL80211_RRF_NO_OFDM;
+
+	if (reg_flags & REGULATORY_CHAN_INDOOR_ONLY)
+		flags |= NL80211_RRF_NO_OUTDOOR;
+
+	if (reg_flags & REGULATORY_CHAN_NO_HT40)
+		flags |= NL80211_RRF_NO_HT40;
+
+	if (reg_flags & REGULATORY_CHAN_NO_80MHZ)
+		flags |= NL80211_RRF_NO_80MHZ;
+
+	if (reg_flags & REGULATORY_CHAN_NO_160MHZ)
+		flags |= NL80211_RRF_NO_160MHZ;
+
+	return flags;
+}
+
+static bool
+ath11k_reg_can_intersect(struct ieee80211_reg_rule *rule1,
+			 struct ieee80211_reg_rule *rule2)
+{
+	u32 start_freq1, end_freq1;
+	u32 start_freq2, end_freq2;
+
+	start_freq1 = rule1->freq_range.start_freq_khz;
+	start_freq2 = rule2->freq_range.start_freq_khz;
+
+	end_freq1 = rule1->freq_range.end_freq_khz;
+	end_freq2 = rule2->freq_range.end_freq_khz;
+
+	if ((start_freq1 >= start_freq2 &&
+	     start_freq1 < end_freq2) ||
+	    (start_freq2 > start_freq1 &&
+	     start_freq2 < end_freq1))
+		return true;
+
+	/* TODO: Should we restrict intersection feasibility
+	 *  based on min bandwidth of the intersected region also,
+	 *  say the intersected rule should have a  min bandwidth
+	 * of 20MHz?
+	 */
+
+	return false;
+}
+
+static void ath11k_reg_intersect_rules(struct ieee80211_reg_rule *rule1,
+				       struct ieee80211_reg_rule *rule2,
+				       struct ieee80211_reg_rule *new_rule)
+{
+	u32 start_freq1, end_freq1;
+	u32 start_freq2, end_freq2;
+	u32 freq_diff, max_bw;
+
+	start_freq1 = rule1->freq_range.start_freq_khz;
+	start_freq2 = rule2->freq_range.start_freq_khz;
+
+	end_freq1 = rule1->freq_range.end_freq_khz;
+	end_freq2 = rule2->freq_range.end_freq_khz;
+
+	new_rule->freq_range.start_freq_khz = max_t(u32, start_freq1,
+						    start_freq2);
+	new_rule->freq_range.end_freq_khz = min_t(u32, end_freq1, end_freq2);
+
+	freq_diff = new_rule->freq_range.end_freq_khz -
+			new_rule->freq_range.start_freq_khz;
+	max_bw = min_t(u32, rule1->freq_range.max_bandwidth_khz,
+		       rule2->freq_range.max_bandwidth_khz);
+	new_rule->freq_range.max_bandwidth_khz = min_t(u32, max_bw, freq_diff);
+
+	new_rule->power_rule.max_antenna_gain =
+		min_t(u32, rule1->power_rule.max_antenna_gain,
+		      rule2->power_rule.max_antenna_gain);
+
+	new_rule->power_rule.max_eirp = min_t(u32, rule1->power_rule.max_eirp,
+					      rule2->power_rule.max_eirp);
+
+	/* Use the flags of both the rules */
+	new_rule->flags = rule1->flags | rule2->flags;
+
+	/* To be safe, lts use the max cac timeout of both rules */
+	new_rule->dfs_cac_ms = max_t(u32, rule1->dfs_cac_ms,
+				     rule2->dfs_cac_ms);
+}
+
+static struct ieee80211_regdomain *
+ath11k_regd_intersect(struct ieee80211_regdomain *default_regd,
+		      struct ieee80211_regdomain *curr_regd)
+{
+	u8 num_old_regd_rules, num_curr_regd_rules, num_new_regd_rules;
+	struct ieee80211_reg_rule *old_rule, *curr_rule, *new_rule;
+	struct ieee80211_regdomain *new_regd = NULL;
+	u8 i, j, k;
+
+	num_old_regd_rules = default_regd->n_reg_rules;
+	num_curr_regd_rules = curr_regd->n_reg_rules;
+	num_new_regd_rules = 0;
+
+	/* Find the number of intersecting rules to allocate new regd memory */
+	for (i = 0; i < num_old_regd_rules; i++) {
+		old_rule = default_regd->reg_rules + i;
+		for (j = 0; j < num_curr_regd_rules; j++) {
+			curr_rule = curr_regd->reg_rules + j;
+
+			if (ath11k_reg_can_intersect(old_rule, curr_rule))
+				num_new_regd_rules++;
+		}
+	}
+
+	if (!num_new_regd_rules)
+		return NULL;
+
+	new_regd = kzalloc(sizeof(*new_regd) + (num_new_regd_rules *
+			sizeof(struct ieee80211_reg_rule)),
+			GFP_ATOMIC);
+
+	if (!new_regd)
+		return NULL;
+
+	/* We set the new country and dfs region directly and only trim
+	 * the freq, power, antenna gain by intersecting with the
+	 * default regdomain. Also MAX of the dfs cac timeout is selected.
+	 */
+	new_regd->n_reg_rules = num_new_regd_rules;
+	memcpy(new_regd->alpha2, curr_regd->alpha2, sizeof(new_regd->alpha2));
+	new_regd->dfs_region = curr_regd->dfs_region;
+	new_rule = new_regd->reg_rules;
+
+	for (i = 0, k = 0; i < num_old_regd_rules; i++) {
+		old_rule = default_regd->reg_rules + i;
+		for (j = 0; j < num_curr_regd_rules; j++) {
+			curr_rule = curr_regd->reg_rules + j;
+
+			if (ath11k_reg_can_intersect(old_rule, curr_rule))
+				ath11k_reg_intersect_rules(old_rule, curr_rule,
+							   (new_rule + k++));
+		}
+	}
+	return new_regd;
+}
+
+static const char *
+ath11k_reg_get_regdom_str(enum nl80211_dfs_regions dfs_region)
+{
+	switch (dfs_region) {
+	case NL80211_DFS_FCC:
+		return "FCC";
+	case NL80211_DFS_ETSI:
+		return "ETSI";
+	case NL80211_DFS_JP:
+		return "JP";
+	default:
+		return "UNSET";
+	}
+}
+
+static u16
+ath11k_reg_adjust_bw(u16 start_freq, u16 end_freq, u16 max_bw)
+{
+	u16 bw;
+
+	bw = end_freq - start_freq;
+	bw = min_t(u16, bw, max_bw);
+
+	if (bw >= 80 && bw < 160)
+		bw = 80;
+	else if (bw >= 40 && bw < 80)
+		bw = 40;
+	else if (bw < 40)
+		bw = 20;
+
+	return bw;
+}
+
+static void
+ath11k_reg_update_rule(struct ieee80211_reg_rule *reg_rule, u32 start_freq,
+		       u32 end_freq, u32 bw, u32 ant_gain, u32 reg_pwr,
+		       u32 reg_flags)
+{
+	reg_rule->freq_range.start_freq_khz = MHZ_TO_KHZ(start_freq);
+	reg_rule->freq_range.end_freq_khz = MHZ_TO_KHZ(end_freq);
+	reg_rule->freq_range.max_bandwidth_khz = MHZ_TO_KHZ(bw);
+	reg_rule->power_rule.max_antenna_gain = DBI_TO_MBI(ant_gain);
+	reg_rule->power_rule.max_eirp = DBM_TO_MBM(reg_pwr);
+	reg_rule->flags = reg_flags;
+}
+
+static void
+ath11k_reg_update_weather_radar_band(struct ath11k_base *ab,
+				     struct ieee80211_regdomain *regd,
+				     struct cur_reg_rule *reg_rule,
+				     u8 *rule_idx, u32 flags, u16 max_bw)
+{
+	u32 end_freq;
+	u16 bw;
+	u8 i;
+
+	i = *rule_idx;
+
+	bw = ath11k_reg_adjust_bw(reg_rule->start_freq,
+				  ETSI_WEATHER_RADAR_BAND_LOW, max_bw);
+
+	ath11k_reg_update_rule(regd->reg_rules + i, reg_rule->start_freq,
+			       ETSI_WEATHER_RADAR_BAND_LOW, bw,
+			       reg_rule->ant_gain, reg_rule->reg_power,
+			       flags);
+
+	ath11k_dbg(ab, ATH11K_DBG_REG,
+		   "\t%d. (%d - %d @ %d) (%d, %d) (%d ms) (FLAGS %d)\n",
+		   i + 1, reg_rule->start_freq, ETSI_WEATHER_RADAR_BAND_LOW,
+		   bw, reg_rule->ant_gain, reg_rule->reg_power,
+		   regd->reg_rules[i].dfs_cac_ms,
+		   flags);
+
+	if (reg_rule->end_freq > ETSI_WEATHER_RADAR_BAND_HIGH)
+		end_freq = ETSI_WEATHER_RADAR_BAND_HIGH;
+	else
+		end_freq = reg_rule->end_freq;
+
+	bw = ath11k_reg_adjust_bw(ETSI_WEATHER_RADAR_BAND_LOW, end_freq,
+				  max_bw);
+
+	i++;
+
+	ath11k_reg_update_rule(regd->reg_rules + i,
+			       ETSI_WEATHER_RADAR_BAND_LOW, end_freq, bw,
+			       reg_rule->ant_gain, reg_rule->reg_power,
+			       flags);
+
+	regd->reg_rules[i].dfs_cac_ms = ETSI_WEATHER_RADAR_BAND_CAC_TIMEOUT;
+
+	ath11k_dbg(ab, ATH11K_DBG_REG,
+		   "\t%d. (%d - %d @ %d) (%d, %d) (%d ms) (FLAGS %d)\n",
+		   i + 1, ETSI_WEATHER_RADAR_BAND_LOW, end_freq,
+		   bw, reg_rule->ant_gain, reg_rule->reg_power,
+		   regd->reg_rules[i].dfs_cac_ms,
+		   flags);
+
+	if (end_freq == reg_rule->end_freq) {
+		regd->n_reg_rules--;
+		*rule_idx = i;
+		return;
+	}
+
+	bw = ath11k_reg_adjust_bw(ETSI_WEATHER_RADAR_BAND_HIGH,
+				  reg_rule->end_freq, max_bw);
+
+	i++;
+
+	ath11k_reg_update_rule(regd->reg_rules + i, ETSI_WEATHER_RADAR_BAND_HIGH,
+			       reg_rule->end_freq, bw,
+			       reg_rule->ant_gain, reg_rule->reg_power,
+			       flags);
+
+	ath11k_dbg(ab, ATH11K_DBG_REG,
+		   "\t%d. (%d - %d @ %d) (%d, %d) (%d ms) (FLAGS %d)\n",
+		   i + 1, ETSI_WEATHER_RADAR_BAND_HIGH, reg_rule->end_freq,
+		   bw, reg_rule->ant_gain, reg_rule->reg_power,
+		   regd->reg_rules[i].dfs_cac_ms,
+		   flags);
+
+	*rule_idx = i;
+}
+
+struct ieee80211_regdomain *
+ath11k_reg_build_regd(struct ath11k_base *ab,
+		      struct cur_regulatory_info *reg_info, bool intersect)
+{
+	struct ieee80211_regdomain *tmp_regd, *default_regd, *new_regd = NULL;
+	struct cur_reg_rule *reg_rule;
+	u8 i = 0, j = 0;
+	u8 num_rules;
+	u16 max_bw;
+	u32 flags;
+	char alpha2[3];
+
+	num_rules = reg_info->num_5g_reg_rules + reg_info->num_2g_reg_rules;
+
+	if (!num_rules)
+		goto ret;
+
+	/* Add max additional rules to accommodate weather radar band */
+	if (reg_info->dfs_region == ATH11K_DFS_REG_ETSI)
+		num_rules += 2;
+
+	tmp_regd =  kzalloc(sizeof(*tmp_regd) +
+			(num_rules * sizeof(struct ieee80211_reg_rule)),
+			GFP_ATOMIC);
+	if (!tmp_regd)
+		goto ret;
+
+	tmp_regd->n_reg_rules = num_rules;
+	memcpy(tmp_regd->alpha2, reg_info->alpha2, REG_ALPHA2_LEN + 1);
+	memcpy(alpha2, reg_info->alpha2, REG_ALPHA2_LEN + 1);
+	alpha2[2] = '\0';
+	tmp_regd->dfs_region = ath11k_map_fw_dfs_region(reg_info->dfs_region);
+
+	ath11k_dbg(ab, ATH11K_DBG_REG,
+		   "\r\nCountry %s, CFG Regdomain %s FW Regdomain %d, num_reg_rules %d\n",
+		   alpha2, ath11k_reg_get_regdom_str(tmp_regd->dfs_region),
+		   reg_info->dfs_region, num_rules);
+	/* Update reg_rules[] below. Firmware is expected to
+	 * send these rules in order(2G rules first and then 5G)
+	 */
+	for (; i < tmp_regd->n_reg_rules; i++) {
+		if (reg_info->num_2g_reg_rules &&
+		    (i < reg_info->num_2g_reg_rules)) {
+			reg_rule = reg_info->reg_rules_2g_ptr + i;
+			max_bw = min_t(u16, reg_rule->max_bw,
+				       reg_info->max_bw_2g);
+			flags = 0;
+		} else if (reg_info->num_5g_reg_rules &&
+			   (j < reg_info->num_5g_reg_rules)) {
+			reg_rule = reg_info->reg_rules_5g_ptr + j++;
+			max_bw = min_t(u16, reg_rule->max_bw,
+				       reg_info->max_bw_5g);
+
+			/* FW doesn't pass NL80211_RRF_AUTO_BW flag for
+			 * BW Auto correction, we can enable this by default
+			 * for all 5G rules here. The regulatory core performs
+			 * BW correction if required and applies flags as
+			 * per other BW rule flags we pass from here
+			 */
+			flags = NL80211_RRF_AUTO_BW;
+		} else {
+			break;
+		}
+
+		flags |= ath11k_map_fw_reg_flags(reg_rule->flags);
+
+		ath11k_reg_update_rule(tmp_regd->reg_rules + i,
+				       reg_rule->start_freq,
+				       reg_rule->end_freq, max_bw,
+				       reg_rule->ant_gain, reg_rule->reg_power,
+				       flags);
+
+		/* Update dfs cac timeout if the dfs domain is ETSI and the
+		 * new rule covers weather radar band.
+		 * Default value of '0' corresponds to 60s timeout, so no
+		 * need to update that for other rules.
+		 */
+		if (flags & NL80211_RRF_DFS &&
+		    reg_info->dfs_region == ATH11K_DFS_REG_ETSI &&
+		    (reg_rule->end_freq > ETSI_WEATHER_RADAR_BAND_LOW &&
+		    reg_rule->start_freq < ETSI_WEATHER_RADAR_BAND_HIGH)){
+			ath11k_reg_update_weather_radar_band(ab, tmp_regd,
+							     reg_rule, &i,
+							     flags, max_bw);
+			continue;
+		}
+
+		ath11k_dbg(ab, ATH11K_DBG_REG,
+			   "\t%d. (%d - %d @ %d) (%d, %d) (%d ms) (FLAGS %d)\n",
+			   i + 1, reg_rule->start_freq, reg_rule->end_freq,
+			   max_bw, reg_rule->ant_gain, reg_rule->reg_power,
+			   tmp_regd->reg_rules[i].dfs_cac_ms,
+			   flags);
+	}
+
+	if (intersect) {
+		default_regd = ab->default_regd[reg_info->phy_id];
+
+		/* Get a new regd by intersecting the received regd with
+		 * our default regd.
+		 */
+		new_regd = ath11k_regd_intersect(default_regd, tmp_regd);
+		kfree(tmp_regd);
+		if (!new_regd) {
+			ath11k_warn(ab, "Unable to create intersected regdomain\n");
+			goto ret;
+		}
+	} else {
+		new_regd = tmp_regd;
+	}
+
+ret:
+	return new_regd;
+}
+
+void ath11k_regd_update_work(struct work_struct *work)
+{
+	struct ath11k *ar = container_of(work, struct ath11k,
+					 regd_update_work);
+	int ret;
+
+	ret = ath11k_regd_update(ar, false);
+	if (ret) {
+		/* Firmware has already moved to the new regd. We need
+		 * to maintain channel consistency across FW, Host driver
+		 * and userspace. Hence as a fallback mechanism we can set
+		 * the prev or default country code to the firmware.
+		 */
+		/* TODO: Implement Fallback Mechanism */
+	}
+}
+
+void ath11k_reg_init(struct ath11k *ar)
+{
+	ar->hw->wiphy->regulatory_flags = REGULATORY_WIPHY_SELF_MANAGED;
+	ar->hw->wiphy->reg_notifier = ath11k_reg_notifier;
+}
+
+void ath11k_reg_free(struct ath11k_base *ab)
+{
+	int i;
+
+	for (i = 0; i < MAX_RADIOS; i++) {
+		kfree(ab->default_regd[i]);
+		kfree(ab->new_regd[i]);
+	}
+}
diff --git a/drivers/net/wireless/ath/ath11k/reg.h b/drivers/net/wireless/ath/ath11k/reg.h
new file mode 100644
index 000000000000..39b7fc943541
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/reg.h
@@ -0,0 +1,35 @@
+/* SPDX-License-Identifier: BSD-3-Clause-Clear */
+/*
+ * Copyright (c) 2019 The Linux Foundation. All rights reserved.
+ */
+
+#ifndef ATH11K_REG_H
+#define ATH11K_REG_H
+
+#include <linux/kernel.h>
+#include <net/regulatory.h>
+
+struct ath11k_base;
+struct ath11k;
+
+/* DFS regdomains supported by Firmware */
+enum ath11k_dfs_region {
+	ATH11K_DFS_REG_UNSET,
+	ATH11K_DFS_REG_FCC,
+	ATH11K_DFS_REG_ETSI,
+	ATH11K_DFS_REG_MKK,
+	ATH11K_DFS_REG_CN,
+	ATH11K_DFS_REG_KR,
+	ATH11K_DFS_REG_UNDEF,
+};
+
+/* ATH11K Regulatory API's */
+void ath11k_reg_init(struct ath11k *ar);
+void ath11k_reg_free(struct ath11k_base *ab);
+void ath11k_regd_update_work(struct work_struct *work);
+struct ieee80211_regdomain *
+ath11k_reg_build_regd(struct ath11k_base *ab,
+		      struct cur_regulatory_info *reg_info, bool intersect);
+int ath11k_regd_update(struct ath11k *ar, bool init);
+int ath11k_reg_update_chan_list(struct ath11k *ar);
+#endif
diff --git a/drivers/net/wireless/ath/ath11k/rx_desc.h b/drivers/net/wireless/ath/ath11k/rx_desc.h
new file mode 100644
index 000000000000..a5aff801f17f
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/rx_desc.h
@@ -0,0 +1,1212 @@
+/* SPDX-License-Identifier: BSD-3-Clause-Clear */
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+#ifndef ATH11K_RX_DESC_H
+#define ATH11K_RX_DESC_H
+
+enum rx_desc_rxpcu_filter {
+	RX_DESC_RXPCU_FILTER_PASS,
+	RX_DESC_RXPCU_FILTER_MONITOR_CLIENT,
+	RX_DESC_RXPCU_FILTER_MONITOR_OTHER,
+};
+
+/* rxpcu_filter_pass
+ *		This MPDU passed the normal frame filter programming of rxpcu.
+ *
+ * rxpcu_filter_monitor_client
+ *		 This MPDU did not pass the regular frame filter and would
+ *		 have been dropped, were it not for the frame fitting into the
+ *		 'monitor_client' category.
+ *
+ * rxpcu_filter_monitor_other
+ *		This MPDU did not pass the regular frame filter and also did
+ *		not pass the rxpcu_monitor_client filter. It would have been
+ *		dropped accept that it did pass the 'monitor_other' category.
+ */
+
+#define RX_DESC_INFO0_RXPCU_MPDU_FITLER	GENMASK(1, 0)
+#define RX_DESC_INFO0_SW_FRAME_GRP_ID	GENMASK(8, 2)
+
+enum rx_desc_sw_frame_grp_id {
+	RX_DESC_SW_FRAME_GRP_ID_NDP_FRAME,
+	RX_DESC_SW_FRAME_GRP_ID_MCAST_DATA,
+	RX_DESC_SW_FRAME_GRP_ID_UCAST_DATA,
+	RX_DESC_SW_FRAME_GRP_ID_NULL_DATA,
+	RX_DESC_SW_FRAME_GRP_ID_MGMT_0000,
+	RX_DESC_SW_FRAME_GRP_ID_MGMT_0001,
+	RX_DESC_SW_FRAME_GRP_ID_MGMT_0010,
+	RX_DESC_SW_FRAME_GRP_ID_MGMT_0011,
+	RX_DESC_SW_FRAME_GRP_ID_MGMT_0100,
+	RX_DESC_SW_FRAME_GRP_ID_MGMT_0101,
+	RX_DESC_SW_FRAME_GRP_ID_MGMT_0110,
+	RX_DESC_SW_FRAME_GRP_ID_MGMT_0111,
+	RX_DESC_SW_FRAME_GRP_ID_MGMT_1000,
+	RX_DESC_SW_FRAME_GRP_ID_MGMT_1001,
+	RX_DESC_SW_FRAME_GRP_ID_MGMT_1010,
+	RX_DESC_SW_FRAME_GRP_ID_MGMT_1011,
+	RX_DESC_SW_FRAME_GRP_ID_MGMT_1100,
+	RX_DESC_SW_FRAME_GRP_ID_MGMT_1101,
+	RX_DESC_SW_FRAME_GRP_ID_MGMT_1110,
+	RX_DESC_SW_FRAME_GRP_ID_MGMT_1111,
+	RX_DESC_SW_FRAME_GRP_ID_CTRL_0000,
+	RX_DESC_SW_FRAME_GRP_ID_CTRL_0001,
+	RX_DESC_SW_FRAME_GRP_ID_CTRL_0010,
+	RX_DESC_SW_FRAME_GRP_ID_CTRL_0011,
+	RX_DESC_SW_FRAME_GRP_ID_CTRL_0100,
+	RX_DESC_SW_FRAME_GRP_ID_CTRL_0101,
+	RX_DESC_SW_FRAME_GRP_ID_CTRL_0110,
+	RX_DESC_SW_FRAME_GRP_ID_CTRL_0111,
+	RX_DESC_SW_FRAME_GRP_ID_CTRL_1000,
+	RX_DESC_SW_FRAME_GRP_ID_CTRL_1001,
+	RX_DESC_SW_FRAME_GRP_ID_CTRL_1010,
+	RX_DESC_SW_FRAME_GRP_ID_CTRL_1011,
+	RX_DESC_SW_FRAME_GRP_ID_CTRL_1100,
+	RX_DESC_SW_FRAME_GRP_ID_CTRL_1101,
+	RX_DESC_SW_FRAME_GRP_ID_CTRL_1110,
+	RX_DESC_SW_FRAME_GRP_ID_CTRL_1111,
+	RX_DESC_SW_FRAME_GRP_ID_UNSUPPORTED,
+	RX_DESC_SW_FRAME_GRP_ID_PHY_ERR,
+};
+
+enum rx_desc_decap_type {
+	RX_DESC_DECAP_TYPE_RAW,
+	RX_DESC_DECAP_TYPE_NATIVE_WIFI,
+	RX_DESC_DECAP_TYPE_ETHERNET2_DIX,
+	RX_DESC_DECAP_TYPE_8023,
+};
+
+enum rx_desc_decrypt_status_code {
+	RX_DESC_DECRYPT_STATUS_CODE_OK,
+	RX_DESC_DECRYPT_STATUS_CODE_UNPROTECTED_FRAME,
+	RX_DESC_DECRYPT_STATUS_CODE_DATA_ERR,
+	RX_DESC_DECRYPT_STATUS_CODE_KEY_INVALID,
+	RX_DESC_DECRYPT_STATUS_CODE_PEER_ENTRY_INVALID,
+	RX_DESC_DECRYPT_STATUS_CODE_OTHER,
+};
+
+#define RX_ATTENTION_INFO1_FIRST_MPDU		BIT(0)
+#define RX_ATTENTION_INFO1_RSVD_1A		BIT(1)
+#define RX_ATTENTION_INFO1_MCAST_BCAST		BIT(2)
+#define RX_ATTENTION_INFO1_AST_IDX_NOT_FOUND	BIT(3)
+#define RX_ATTENTION_INFO1_AST_IDX_TIMEDOUT	BIT(4)
+#define RX_ATTENTION_INFO1_POWER_MGMT		BIT(5)
+#define RX_ATTENTION_INFO1_NON_QOS		BIT(6)
+#define RX_ATTENTION_INFO1_NULL_DATA		BIT(7)
+#define RX_ATTENTION_INFO1_MGMT_TYPE		BIT(8)
+#define RX_ATTENTION_INFO1_CTRL_TYPE		BIT(9)
+#define RX_ATTENTION_INFO1_MORE_DATA		BIT(10)
+#define RX_ATTENTION_INFO1_EOSP			BIT(11)
+#define RX_ATTENTION_INFO1_A_MSDU_ERROR		BIT(12)
+#define RX_ATTENTION_INFO1_FRAGMENT		BIT(13)
+#define RX_ATTENTION_INFO1_ORDER		BIT(14)
+#define RX_ATTENTION_INFO1_CCE_MATCH		BIT(15)
+#define RX_ATTENTION_INFO1_OVERFLOW_ERR		BIT(16)
+#define RX_ATTENTION_INFO1_MSDU_LEN_ERR		BIT(17)
+#define RX_ATTENTION_INFO1_TCP_UDP_CKSUM_FAIL	BIT(18)
+#define RX_ATTENTION_INFO1_IP_CKSUM_FAIL	BIT(19)
+#define RX_ATTENTION_INFO1_SA_IDX_INVALID	BIT(20)
+#define RX_ATTENTION_INFO1_DA_IDX_INVALID	BIT(21)
+#define RX_ATTENTION_INFO1_RSVD_1B		BIT(22)
+#define RX_ATTENTION_INFO1_RX_IN_TX_DECRYPT_BYP	BIT(23)
+#define RX_ATTENTION_INFO1_ENCRYPT_REQUIRED	BIT(24)
+#define RX_ATTENTION_INFO1_DIRECTED		BIT(25)
+#define RX_ATTENTION_INFO1_BUFFER_FRAGMENT	BIT(26)
+#define RX_ATTENTION_INFO1_MPDU_LEN_ERR		BIT(27)
+#define RX_ATTENTION_INFO1_TKIP_MIC_ERR		BIT(28)
+#define RX_ATTENTION_INFO1_DECRYPT_ERR		BIT(29)
+#define RX_ATTENTION_INFO1_UNDECRYPT_FRAME_ERR	BIT(30)
+#define RX_ATTENTION_INFO1_FCS_ERR		BIT(31)
+
+#define RX_ATTENTION_INFO2_FLOW_IDX_TIMEOUT	BIT(0)
+#define RX_ATTENTION_INFO2_FLOW_IDX_INVALID	BIT(1)
+#define RX_ATTENTION_INFO2_WIFI_PARSER_ERR	BIT(2)
+#define RX_ATTENTION_INFO2_AMSDU_PARSER_ERR	BIT(3)
+#define RX_ATTENTION_INFO2_SA_IDX_TIMEOUT	BIT(4)
+#define RX_ATTENTION_INFO2_DA_IDX_TIMEOUT	BIT(5)
+#define RX_ATTENTION_INFO2_MSDU_LIMIT_ERR	BIT(6)
+#define RX_ATTENTION_INFO2_DA_IS_VALID		BIT(7)
+#define RX_ATTENTION_INFO2_DA_IS_MCBC		BIT(8)
+#define RX_ATTENTION_INFO2_SA_IS_VALID		BIT(9)
+#define RX_ATTENTION_INFO2_DCRYPT_STATUS_CODE	GENMASK(12, 10)
+#define RX_ATTENTION_INFO2_RX_BITMAP_NOT_UPDED	BIT(13)
+#define RX_ATTENTION_INFO2_MSDU_DONE		BIT(31)
+
+struct rx_attention {
+	__le16 info0;
+	__le16 phy_ppdu_id;
+	__le32 info1;
+	__le32 info2;
+} __packed;
+
+/* rx_attention
+ *
+ * rxpcu_mpdu_filter_in_category
+ *		Field indicates what the reason was that this mpdu frame
+ *		was allowed to come into the receive path by rxpcu. Values
+ *		are defined in enum %RX_DESC_RXPCU_FILTER_*.
+ *
+ * sw_frame_group_id
+ *		SW processes frames based on certain classifications. Values
+ *		are defined in enum %RX_DESC_SW_FRAME_GRP_ID_*.
+ *
+ * phy_ppdu_id
+ *		A ppdu counter value that PHY increments for every PPDU
+ *		received. The counter value wraps around.
+ *
+ * first_mpdu
+ *		Indicates the first MSDU of the PPDU.  If both first_mpdu
+ *		and last_mpdu are set in the MSDU then this is a not an
+ *		A-MPDU frame but a stand alone MPDU.  Interior MPDU in an
+ *		A-MPDU shall have both first_mpdu and last_mpdu bits set to
+ *		0.  The PPDU start status will only be valid when this bit
+ *		is set.
+ *
+ * mcast_bcast
+ *		Multicast / broadcast indicator.  Only set when the MAC
+ *		address 1 bit 0 is set indicating mcast/bcast and the BSSID
+ *		matches one of the 4 BSSID registers. Only set when
+ *		first_msdu is set.
+ *
+ * ast_index_not_found
+ *		Only valid when first_msdu is set. Indicates no AST matching
+ *		entries within the the max search count.
+ *
+ * ast_index_timeout
+ *		Only valid when first_msdu is set. Indicates an unsuccessful
+ *		search in the address search table due to timeout.
+ *
+ * power_mgmt
+ *		Power management bit set in the 802.11 header.  Only set
+ *		when first_msdu is set.
+ *
+ * non_qos
+ *		Set if packet is not a non-QoS data frame.  Only set when
+ *		first_msdu is set.
+ *
+ * null_data
+ *		Set if frame type indicates either null data or QoS null
+ *		data format.  Only set when first_msdu is set.
+ *
+ * mgmt_type
+ *		Set if packet is a management packet.  Only set when
+ *		first_msdu is set.
+ *
+ * ctrl_type
+ *		Set if packet is a control packet.  Only set when first_msdu
+ *		is set.
+ *
+ * more_data
+ *		Set if more bit in frame control is set.  Only set when
+ *		first_msdu is set.
+ *
+ * eosp
+ *		Set if the EOSP (end of service period) bit in the QoS
+ *		control field is set.  Only set when first_msdu is set.
+ *
+ * a_msdu_error
+ *		Set if number of MSDUs in A-MSDU is above a threshold or if the
+ *		size of the MSDU is invalid. This receive buffer will contain
+ *		all of the remainder of MSDUs in this MPDU w/o decapsulation.
+ *
+ * fragment
+ *		Indicates that this is an 802.11 fragment frame.  This is
+ *		set when either the more_frag bit is set in the frame
+ *		control or the fragment number is not zero.  Only set when
+ *		first_msdu is set.
+ *
+ * order
+ *		Set if the order bit in the frame control is set.  Only set
+ *		when first_msdu is set.
+ *
+ * cce_match
+ *		Indicates that this status has a corresponding MSDU that
+ *		requires FW processing. The OLE will have classification
+ *		ring mask registers which will indicate the ring(s) for
+ *		packets and descriptors which need FW attention.
+ *
+ * overflow_err
+ *		PCU Receive FIFO does not have enough space to store the
+ *		full receive packet.  Enough space is reserved in the
+ *		receive FIFO for the status is written.  This MPDU remaining
+ *		packets in the PPDU will be filtered and no Ack response
+ *		will be transmitted.
+ *
+ * msdu_length_err
+ *		Indicates that the MSDU length from the 802.3 encapsulated
+ *		length field extends beyond the MPDU boundary.
+ *
+ * tcp_udp_chksum_fail
+ *		Indicates that the computed checksum (tcp_udp_chksum) did
+ *		not match the checksum in the TCP/UDP header.
+ *
+ * ip_chksum_fail
+ *		Indicates that the computed checksum did not match the
+ *		checksum in the IP header.
+ *
+ * sa_idx_invalid
+ *		Indicates no matching entry was found in the address search
+ *		table for the source MAC address.
+ *
+ * da_idx_invalid
+ *		Indicates no matching entry was found in the address search
+ *		table for the destination MAC address.
+ *
+ * rx_in_tx_decrypt_byp
+ *		Indicates that RX packet is not decrypted as Crypto is busy
+ *		with TX packet processing.
+ *
+ * encrypt_required
+ *		Indicates that this data type frame is not encrypted even if
+ *		the policy for this MPDU requires encryption as indicated in
+ *		the peer table key type.
+ *
+ * directed
+ *		MPDU is a directed packet which means that the RA matched
+ *		our STA addresses.  In proxySTA it means that the TA matched
+ *		an entry in our address search table with the corresponding
+ *		'no_ack' bit is the address search entry cleared.
+ *
+ * buffer_fragment
+ *		Indicates that at least one of the rx buffers has been
+ *		fragmented.  If set the FW should look at the rx_frag_info
+ *		descriptor described below.
+ *
+ * mpdu_length_err
+ *		Indicates that the MPDU was pre-maturely terminated
+ *		resulting in a truncated MPDU.  Don't trust the MPDU length
+ *		field.
+ *
+ * tkip_mic_err
+ *		Indicates that the MPDU Michael integrity check failed
+ *
+ * decrypt_err
+ *		Indicates that the MPDU decrypt integrity check failed
+ *
+ * fcs_err
+ *		Indicates that the MPDU FCS check failed
+ *
+ * flow_idx_timeout
+ *		Indicates an unsuccessful flow search due to the expiring of
+ *		the search timer.
+ *
+ * flow_idx_invalid
+ *		flow id is not valid.
+ *
+ * amsdu_parser_error
+ *		A-MSDU could not be properly de-agregated.
+ *
+ * sa_idx_timeout
+ *		Indicates an unsuccessful search for the source MAC address
+ *		due to the expiring of the search timer.
+ *
+ * da_idx_timeout
+ *		Indicates an unsuccessful search for the destination MAC
+ *		address due to the expiring of the search timer.
+ *
+ * msdu_limit_error
+ *		Indicates that the MSDU threshold was exceeded and thus
+ *		all the rest of the MSDUs will not be scattered and will not
+ *		be decasulated but will be DMA'ed in RAW format as a single
+ *		MSDU buffer.
+ *
+ * da_is_valid
+ *		Indicates that OLE found a valid DA entry.
+ *
+ * da_is_mcbc
+ *		Field Only valid if da_is_valid is set. Indicates the DA address
+ *		was a Multicast or Broadcast address.
+ *
+ * sa_is_valid
+ *		Indicates that OLE found a valid SA entry.
+ *
+ * decrypt_status_code
+ *		Field provides insight into the decryption performed. Values are
+ *		defined in enum %RX_DESC_DECRYPT_STATUS_CODE*.
+ *
+ * rx_bitmap_not_updated
+ *		Frame is received, but RXPCU could not update the receive bitmap
+ *		due to (temporary) fifo constraints.
+ *
+ * msdu_done
+ *		If set indicates that the RX packet data, RX header data, RX
+ *		PPDU start descriptor, RX MPDU start/end descriptor, RX MSDU
+ *		start/end descriptors and RX Attention descriptor are all
+ *		valid.  This bit must be in the last octet of the
+ *		descriptor.
+ */
+
+#define RX_MPDU_START_INFO0_NDP_FRAME		BIT(9)
+#define RX_MPDU_START_INFO0_PHY_ERR		BIT(10)
+#define RX_MPDU_START_INFO0_PHY_ERR_MPDU_HDR	BIT(11)
+#define RX_MPDU_START_INFO0_PROTO_VER_ERR	BIT(12)
+#define RX_MPDU_START_INFO0_AST_LOOKUP_VALID	BIT(13)
+
+#define RX_MPDU_START_INFO1_MPDU_CTRL_VALID	BIT(0)
+#define RX_MPDU_START_INFO1_MPDU_DUR_VALID	BIT(1)
+#define RX_MPDU_START_INFO1_MAC_ADDR1_VALID	BIT(2)
+#define RX_MPDU_START_INFO1_MAC_ADDR2_VALID	BIT(3)
+#define RX_MPDU_START_INFO1_MAC_ADDR3_VALID	BIT(4)
+#define RX_MPDU_START_INFO1_MAC_ADDR4_VALID	BIT(5)
+#define RX_MPDU_START_INFO1_MPDU_SEQ_CTRL_VALID	BIT(6)
+#define RX_MPDU_START_INFO1_MPDU_QOS_CTRL_VALID	BIT(7)
+#define RX_MPDU_START_INFO1_MPDU_HT_CTRL_VALID	BIT(8)
+#define RX_MPDU_START_INFO1_ENCRYPT_INFO_VALID	BIT(9)
+#define RX_MPDU_START_INFO1_MPDU_FRAG_NUMBER	GENMASK(13, 10)
+#define RX_MPDU_START_INFO1_MORE_FRAG_FLAG	BIT(14)
+#define RX_MPDU_START_INFO1_FROM_DS		BIT(16)
+#define RX_MPDU_START_INFO1_TO_DS		BIT(17)
+#define RX_MPDU_START_INFO1_ENCRYPTED		BIT(18)
+#define RX_MPDU_START_INFO1_MPDU_RETRY		BIT(19)
+#define RX_MPDU_START_INFO1_MPDU_SEQ_NUM	GENMASK(31, 20)
+
+#define RX_MPDU_START_INFO2_EPD_EN		BIT(0)
+#define RX_MPDU_START_INFO2_ALL_FRAME_ENCPD	BIT(1)
+#define RX_MPDU_START_INFO2_ENC_TYPE		GENMASK(5, 2)
+#define RX_MPDU_START_INFO2_VAR_WEP_KEY_WIDTH	GENMASK(7, 6)
+#define RX_MPDU_START_INFO2_MESH_STA		BIT(8)
+#define RX_MPDU_START_INFO2_BSSID_HIT		BIT(9)
+#define RX_MPDU_START_INFO2_BSSID_NUM		GENMASK(13, 10)
+#define RX_MPDU_START_INFO2_TID			GENMASK(17, 14)
+
+#define RX_MPDU_START_INFO3_REO_DEST_IND		GENMASK(4, 0)
+#define RX_MPDU_START_INFO3_FLOW_ID_TOEPLITZ		BIT(7)
+#define RX_MPDU_START_INFO3_PKT_SEL_FP_UCAST_DATA	BIT(8)
+#define RX_MPDU_START_INFO3_PKT_SEL_FP_MCAST_DATA	BIT(9)
+#define RX_MPDU_START_INFO3_PKT_SEL_FP_CTRL_BAR		BIT(10)
+#define RX_MPDU_START_INFO3_RXDMA0_SRC_RING_SEL		GENMASK(12, 11)
+#define RX_MPDU_START_INFO3_RXDMA0_DST_RING_SEL		GENMASK(14, 13)
+
+#define RX_MPDU_START_INFO4_REO_QUEUE_DESC_HI	GENMASK(7, 0)
+#define RX_MPDU_START_INFO4_RECV_QUEUE_NUM	GENMASK(23, 8)
+#define RX_MPDU_START_INFO4_PRE_DELIM_ERR_WARN	BIT(24)
+#define RX_MPDU_START_INFO4_FIRST_DELIM_ERR	BIT(25)
+
+#define RX_MPDU_START_INFO5_KEY_ID		GENMASK(7, 0)
+#define RX_MPDU_START_INFO5_NEW_PEER_ENTRY	BIT(8)
+#define RX_MPDU_START_INFO5_DECRYPT_NEEDED	BIT(9)
+#define RX_MPDU_START_INFO5_DECAP_TYPE		GENMASK(11, 10)
+#define RX_MPDU_START_INFO5_VLAN_TAG_C_PADDING	BIT(12)
+#define RX_MPDU_START_INFO5_VLAN_TAG_S_PADDING	BIT(13)
+#define RX_MPDU_START_INFO5_STRIP_VLAN_TAG_C	BIT(14)
+#define RX_MPDU_START_INFO5_STRIP_VLAN_TAG_S	BIT(15)
+#define RX_MPDU_START_INFO5_PRE_DELIM_COUNT	GENMASK(27, 16)
+#define RX_MPDU_START_INFO5_AMPDU_FLAG		BIT(28)
+#define RX_MPDU_START_INFO5_BAR_FRAME		BIT(29)
+
+#define RX_MPDU_START_INFO6_MPDU_LEN		GENMASK(13, 0)
+#define RX_MPDU_START_INFO6_FIRST_MPDU		BIT(14)
+#define RX_MPDU_START_INFO6_MCAST_BCAST		BIT(15)
+#define RX_MPDU_START_INFO6_AST_IDX_NOT_FOUND	BIT(16)
+#define RX_MPDU_START_INFO6_AST_IDX_TIMEOUT	BIT(17)
+#define RX_MPDU_START_INFO6_POWER_MGMT		BIT(18)
+#define RX_MPDU_START_INFO6_NON_QOS		BIT(19)
+#define RX_MPDU_START_INFO6_NULL_DATA		BIT(20)
+#define RX_MPDU_START_INFO6_MGMT_TYPE		BIT(21)
+#define RX_MPDU_START_INFO6_CTRL_TYPE		BIT(22)
+#define RX_MPDU_START_INFO6_MORE_DATA		BIT(23)
+#define RX_MPDU_START_INFO6_EOSP		BIT(24)
+#define RX_MPDU_START_INFO6_FRAGMENT		BIT(25)
+#define RX_MPDU_START_INFO6_ORDER		BIT(26)
+#define RX_MPDU_START_INFO6_UAPSD_TRIGGER	BIT(27)
+#define RX_MPDU_START_INFO6_ENCRYPT_REQUIRED	BIT(28)
+#define RX_MPDU_START_INFO6_DIRECTED		BIT(29)
+
+#define RX_MPDU_START_RAW_MPDU			BIT(0)
+
+struct rx_mpdu_start {
+	__le16 info0;
+	__le16 phy_ppdu_id;
+	__le16 ast_index;
+	__le16 sw_peer_id;
+	__le32 info1;
+	__le32 info2;
+	__le32 pn[4];
+	__le32 peer_meta_data;
+	__le32 info3;
+	__le32 reo_queue_desc_lo;
+	__le32 info4;
+	__le32 info5;
+	__le32 info6;
+	__le16 frame_ctrl;
+	__le16 duration;
+	u8 addr1[ETH_ALEN];
+	u8 addr2[ETH_ALEN];
+	u8 addr3[ETH_ALEN];
+	__le16 seq_ctrl;
+	u8 addr4[ETH_ALEN];
+	__le16 qos_ctrl;
+	__le32 ht_ctrl;
+	__le32 raw;
+} __packed;
+
+/* rx_mpdu_start
+ *
+ * rxpcu_mpdu_filter_in_category
+ *		Field indicates what the reason was that this mpdu frame
+ *		was allowed to come into the receive path by rxpcu. Values
+ *		are defined in enum %RX_DESC_RXPCU_FILTER_*.
+ *		Note: for ndp frame, if it was expected because the preceding
+ *		NDPA was filter_pass, the setting rxpcu_filter_pass will be
+ *		used. This setting will also be used for every ndp frame in
+ *		case Promiscuous mode is enabled.
+ *
+ * sw_frame_group_id
+ *		SW processes frames based on certain classifications. Values
+ *		are defined in enum %RX_DESC_SW_FRAME_GRP_ID_*.
+ *
+ * ndp_frame
+ *		Indicates that the received frame was an NDP frame.
+ *
+ * phy_err
+ *		Indicates that PHY error was received before MAC received data.
+ *
+ * phy_err_during_mpdu_header
+ *		PHY error was received before MAC received the complete MPDU
+ *		header which was needed for proper decoding.
+ *
+ * protocol_version_err
+ *		RXPCU detected a version error in the frame control field.
+ *
+ * ast_based_lookup_valid
+ *		AST based lookup for this frame has found a valid result.
+ *
+ * phy_ppdu_id
+ *		A ppdu counter value that PHY increments for every PPDU
+ *		received. The counter value wraps around.
+ *
+ * ast_index
+ *		This field indicates the index of the AST entry corresponding
+ *		to this MPDU. It is provided by the GSE module instantiated in
+ *		RXPCU. A value of 0xFFFF indicates an invalid AST index.
+ *
+ * sw_peer_id
+ *		This field indicates a unique peer identifier. It is set equal
+ *		to field 'sw_peer_id' from the AST entry.
+ *
+ * mpdu_frame_control_valid, mpdu_duration_valid, mpdu_qos_control_valid,
+ * mpdu_ht_control_valid, frame_encryption_info_valid
+ *		Indicates that each fields have valid entries.
+ *
+ * mac_addr_adx_valid
+ *		Corresponding mac_addr_adx_{lo/hi} has valid entries.
+ *
+ * from_ds, to_ds
+ *		Valid only when mpdu_frame_control_valid is set. Indicates that
+ *		frame is received from DS and sent to DS.
+ *
+ * encrypted
+ *		Protected bit from the frame control.
+ *
+ * mpdu_retry
+ *		Retry bit from frame control. Only valid when first_msdu is set.
+ *
+ * mpdu_sequence_number
+ *		The sequence number from the 802.11 header.
+ *
+ * epd_en
+ *		If set, use EPD instead of LPD.
+ *
+ * all_frames_shall_be_encrypted
+ *		If set, all frames (data only?) shall be encrypted. If not,
+ *		RX CRYPTO shall set an error flag.
+ *
+ * encrypt_type
+ *		Values are defined in enum %HAL_ENCRYPT_TYPE_.
+ *
+ * mesh_sta
+ *		Indicates a Mesh (11s) STA.
+ *
+ * bssid_hit
+ *		 BSSID of the incoming frame matched one of the 8 BSSID
+ *		 register values.
+ *
+ * bssid_number
+ *		This number indicates which one out of the 8 BSSID register
+ *		values matched the incoming frame.
+ *
+ * tid
+ *		TID field in the QoS control field
+ *
+ * pn
+ *		The PN number.
+ *
+ * peer_meta_data
+ *		Meta data that SW has programmed in the Peer table entry
+ *		of the transmitting STA.
+ *
+ * rx_reo_queue_desc_addr_lo
+ *		Address (lower 32 bits) of the REO queue descriptor.
+ *
+ * rx_reo_queue_desc_addr_hi
+ *		Address (upper 8 bits) of the REO queue descriptor.
+ *
+ * receive_queue_number
+ *		Indicates the MPDU queue ID to which this MPDU link
+ *		descriptor belongs.
+ *
+ * pre_delim_err_warning
+ *		Indicates that a delimiter FCS error was found in between the
+ *		previous MPDU and this MPDU. Note that this is just a warning,
+ *		and does not mean that this MPDU is corrupted in any way. If
+ *		it is, there will be other errors indicated such as FCS or
+ *		decrypt errors.
+ *
+ * first_delim_err
+ *		Indicates that the first delimiter had a FCS failure.
+ *
+ * key_id
+ *		The key ID octet from the IV.
+ *
+ * new_peer_entry
+ *		Set if new RX_PEER_ENTRY TLV follows. If clear, RX_PEER_ENTRY
+ *		doesn't follow so RX DECRYPTION module either uses old peer
+ *		entry or not decrypt.
+ *
+ * decrypt_needed
+ *		When RXPCU sets bit 'ast_index_not_found or ast_index_timeout',
+ *		RXPCU will also ensure that this bit is NOT set. CRYPTO for that
+ *		reason only needs to evaluate this bit and non of the other ones
+ *
+ * decap_type
+ *		Used by the OLE during decapsulation. Values are defined in
+ *		enum %MPDU_START_DECAP_TYPE_*.
+ *
+ * rx_insert_vlan_c_tag_padding
+ * rx_insert_vlan_s_tag_padding
+ *		Insert 4 byte of all zeros as VLAN tag or double VLAN tag if
+ *		the rx payload does not have VLAN.
+ *
+ * strip_vlan_c_tag_decap
+ * strip_vlan_s_tag_decap
+ *		Strip VLAN or double VLAN during decapsulation.
+ *
+ * pre_delim_count
+ *		The number of delimiters before this MPDU. Note that this
+ *		number is cleared at PPDU start. If this MPDU is the first
+ *		received MPDU in the PPDU and this MPDU gets filtered-in,
+ *		this field will indicate the number of delimiters located
+ *		after the last MPDU in the previous PPDU.
+ *
+ *		If this MPDU is located after the first received MPDU in
+ *		an PPDU, this field will indicate the number of delimiters
+ *		located between the previous MPDU and this MPDU.
+ *
+ * ampdu_flag
+ *		Received frame was part of an A-MPDU.
+ *
+ * bar_frame
+ *		Received frame is a BAR frame
+ *
+ * mpdu_length
+ *		MPDU length before decapsulation.
+ *
+ * first_mpdu..directed
+ *		See definition in RX attention descriptor
+ *
+ */
+
+enum rx_msdu_start_pkt_type {
+	RX_MSDU_START_PKT_TYPE_11A,
+	RX_MSDU_START_PKT_TYPE_11B,
+	RX_MSDU_START_PKT_TYPE_11N,
+	RX_MSDU_START_PKT_TYPE_11AC,
+	RX_MSDU_START_PKT_TYPE_11AX,
+};
+
+enum rx_msdu_start_sgi {
+	RX_MSDU_START_SGI_0_8_US,
+	RX_MSDU_START_SGI_0_4_US,
+	RX_MSDU_START_SGI_1_6_US,
+	RX_MSDU_START_SGI_3_2_US,
+};
+
+enum rx_msdu_start_recv_bw {
+	RX_MSDU_START_RECV_BW_20MHZ,
+	RX_MSDU_START_RECV_BW_40MHZ,
+	RX_MSDU_START_RECV_BW_80MHZ,
+	RX_MSDU_START_RECV_BW_160MHZ,
+};
+
+enum rx_msdu_start_reception_type {
+	RX_MSDU_START_RECEPTION_TYPE_SU,
+	RX_MSDU_START_RECEPTION_TYPE_DL_MU_MIMO,
+	RX_MSDU_START_RECEPTION_TYPE_DL_MU_OFDMA,
+	RX_MSDU_START_RECEPTION_TYPE_DL_MU_OFDMA_MIMO,
+	RX_MSDU_START_RECEPTION_TYPE_UL_MU_MIMO,
+	RX_MSDU_START_RECEPTION_TYPE_UL_MU_OFDMA,
+	RX_MSDU_START_RECEPTION_TYPE_UL_MU_OFDMA_MIMO,
+};
+
+#define RX_MSDU_START_INFO1_MSDU_LENGTH		GENMASK(13, 0)
+#define RX_MSDU_START_INFO1_RSVD_1A		BIT(14)
+#define RX_MSDU_START_INFO1_IPSEC_ESP		BIT(15)
+#define RX_MSDU_START_INFO1_L3_OFFSET		GENMASK(22, 16)
+#define RX_MSDU_START_INFO1_IPSEC_AH		BIT(23)
+#define RX_MSDU_START_INFO1_L4_OFFSET		GENMASK(31, 24)
+
+#define RX_MSDU_START_INFO2_MSDU_NUMBER		GENMASK(7, 0)
+#define RX_MSDU_START_INFO2_DECAP_TYPE		GENMASK(9, 8)
+#define RX_MSDU_START_INFO2_IPV4		BIT(10)
+#define RX_MSDU_START_INFO2_IPV6		BIT(11)
+#define RX_MSDU_START_INFO2_TCP			BIT(12)
+#define RX_MSDU_START_INFO2_UDP			BIT(13)
+#define RX_MSDU_START_INFO2_IP_FRAG		BIT(14)
+#define RX_MSDU_START_INFO2_TCP_ONLY_ACK	BIT(15)
+#define RX_MSDU_START_INFO2_DA_IS_BCAST_MCAST	BIT(16)
+#define RX_MSDU_START_INFO2_SELECTED_TOEPLITZ_HASH	GENMASK(18, 17)
+#define RX_MSDU_START_INFO2_IP_FIXED_HDR_VALID		BIT(19)
+#define RX_MSDU_START_INFO2_IP_EXTN_HDR_VALID		BIT(20)
+#define RX_MSDU_START_INFO2_IP_TCP_UDP_HDR_VALID	BIT(21)
+#define RX_MSDU_START_INFO2_MESH_CTRL_PRESENT		BIT(22)
+#define RX_MSDU_START_INFO2_LDPC			BIT(23)
+#define RX_MSDU_START_INFO2_IP4_IP6_NXT_HDR		GENMASK(31, 24)
+#define RX_MSDU_START_INFO2_DECAP_FORMAT		GENMASK(9, 8)
+
+#define RX_MSDU_START_INFO3_USER_RSSI		GENMASK(7, 0)
+#define RX_MSDU_START_INFO3_PKT_TYPE		GENMASK(11, 8)
+#define RX_MSDU_START_INFO3_STBC		BIT(12)
+#define RX_MSDU_START_INFO3_SGI			GENMASK(14, 13)
+#define RX_MSDU_START_INFO3_RATE_MCS		GENMASK(18, 15)
+#define RX_MSDU_START_INFO3_RECV_BW		GENMASK(20, 19)
+#define RX_MSDU_START_INFO3_RECEPTION_TYPE	GENMASK(23, 21)
+#define RX_MSDU_START_INFO3_MIMO_SS_BITMAP	GENMASK(31, 24)
+
+struct rx_msdu_start {
+	__le16 info0;
+	__le16 phy_ppdu_id;
+	__le32 info1;
+	__le32 info2;
+	__le32 toeplitz_hash;
+	__le32 flow_id_toeplitz;
+	__le32 info3;
+	__le32 ppdu_start_timestamp;
+	__le32 phy_meta_data;
+} __packed;
+
+/* rx_msdu_start
+ *
+ * rxpcu_mpdu_filter_in_category
+ *		Field indicates what the reason was that this mpdu frame
+ *		was allowed to come into the receive path by rxpcu. Values
+ *		are defined in enum %RX_DESC_RXPCU_FILTER_*.
+ *
+ * sw_frame_group_id
+ *		SW processes frames based on certain classifications. Values
+ *		are defined in enum %RX_DESC_SW_FRAME_GRP_ID_*.
+ *
+ * phy_ppdu_id
+ *		A ppdu counter value that PHY increments for every PPDU
+ *		received. The counter value wraps around.
+ *
+ * msdu_length
+ *		MSDU length in bytes after decapsulation.
+ *
+ * ipsec_esp
+ *		Set if IPv4/v6 packet is using IPsec ESP.
+ *
+ * l3_offset
+ *		Depending upon mode bit, this field either indicates the
+ *		L3 offset in bytes from the start of the RX_HEADER or the IP
+ *		offset in bytes from the start of the packet after
+ *		decapsulation. The latter is only valid if ipv4_proto or
+ *		ipv6_proto is set.
+ *
+ * ipsec_ah
+ *		Set if IPv4/v6 packet is using IPsec AH
+ *
+ * l4_offset
+ *		Depending upon mode bit, this field either indicates the
+ *		L4 offset nin bytes from the start of RX_HEADER (only valid
+ *		if either ipv4_proto or ipv6_proto is set to 1) or indicates
+ *		the offset in bytes to the start of TCP or UDP header from
+ *		the start of the IP header after decapsulation (Only valid if
+ *		tcp_proto or udp_proto is set). The value 0 indicates that
+ *		the offset is longer than 127 bytes.
+ *
+ * msdu_number
+ *		Indicates the MSDU number within a MPDU.  This value is
+ *		reset to zero at the start of each MPDU.  If the number of
+ *		MSDU exceeds 255 this number will wrap using modulo 256.
+ *
+ * decap_type
+ *		Indicates the format after decapsulation. Values are defined in
+ *		enum %MPDU_START_DECAP_TYPE_*.
+ *
+ * ipv4_proto
+ *		Set if L2 layer indicates IPv4 protocol.
+ *
+ * ipv6_proto
+ *		Set if L2 layer indicates IPv6 protocol.
+ *
+ * tcp_proto
+ *		Set if the ipv4_proto or ipv6_proto are set and the IP protocol
+ *		indicates TCP.
+ *
+ * udp_proto
+ *		Set if the ipv4_proto or ipv6_proto are set and the IP protocol
+ *		indicates UDP.
+ *
+ * ip_frag
+ *		Indicates that either the IP More frag bit is set or IP frag
+ *		number is non-zero.  If set indicates that this is a fragmented
+ *		IP packet.
+ *
+ * tcp_only_ack
+ *		Set if only the TCP Ack bit is set in the TCP flags and if
+ *		the TCP payload is 0.
+ *
+ * da_is_bcast_mcast
+ *		The destination address is broadcast or multicast.
+ *
+ * toeplitz_hash
+ *		Actual chosen Hash.
+ *		0 - Toeplitz hash of 2-tuple (IP source address, IP
+ *		    destination address)
+ *		1 - Toeplitz hash of 4-tuple (IP source	address,
+ *		    IP destination address, L4 (TCP/UDP) source port,
+ *		    L4 (TCP/UDP) destination port)
+ *		2 - Toeplitz of flow_id
+ *		3 - Zero is used
+ *
+ * ip_fixed_header_valid
+ *		Fixed 20-byte IPv4 header or 40-byte IPv6 header parsed
+ *		fully within first 256 bytes of the packet
+ *
+ * ip_extn_header_valid
+ *		IPv6/IPv6 header, including IPv4 options and
+ *		recognizable extension headers parsed fully within first 256
+ *		bytes of the packet
+ *
+ * tcp_udp_header_valid
+ *		Fixed 20-byte TCP (excluding TCP options) or 8-byte UDP
+ *		header parsed fully within first 256 bytes of the packet
+ *
+ * mesh_control_present
+ *		When set, this MSDU includes the 'Mesh Control' field
+ *
+ * ldpc
+ *
+ * ip4_protocol_ip6_next_header
+ *		For IPv4, this is the 8 bit protocol field set). For IPv6 this
+ *		is the 8 bit next_header field.
+ *
+ * toeplitz_hash_2_or_4
+ *		Controlled by RxOLE register - If register bit set to 0,
+ *		Toeplitz hash is computed over 2-tuple IPv4 or IPv6 src/dest
+ *		addresses; otherwise, toeplitz hash is computed over 4-tuple
+ *		IPv4 or IPv6 src/dest addresses and src/dest ports.
+ *
+ * flow_id_toeplitz
+ *		Toeplitz hash of 5-tuple
+ *		{IP source address, IP destination address, IP source port, IP
+ *		destination port, L4 protocol}  in case of non-IPSec.
+ *
+ *		In case of IPSec - Toeplitz hash of 4-tuple
+ *		{IP source address, IP destination address, SPI, L4 protocol}
+ *
+ *		The relevant Toeplitz key registers are provided in RxOLE's
+ *		instance of common parser module. These registers are separate
+ *		from the Toeplitz keys used by ASE/FSE modules inside RxOLE.
+ *		The actual value will be passed on from common parser module
+ *		to RxOLE in one of the WHO_* TLVs.
+ *
+ * user_rssi
+ *		RSSI for this user
+ *
+ * pkt_type
+ *		Values are defined in enum %RX_MSDU_START_PKT_TYPE_*.
+ *
+ * stbc
+ *		When set, use STBC transmission rates.
+ *
+ * sgi
+ *		Field only valid when pkt type is HT, VHT or HE. Values are
+ *		defined in enum %RX_MSDU_START_SGI_*.
+ *
+ * rate_mcs
+ *		MCS Rate used.
+ *
+ * receive_bandwidth
+ *		Full receive Bandwidth. Values are defined in enum
+ *		%RX_MSDU_START_RECV_*.
+ *
+ * reception_type
+ *		Indicates what type of reception this is and defined in enum
+ *		%RX_MSDU_START_RECEPTION_TYPE_*.
+ *
+ * mimo_ss_bitmap
+ *		Field only valid when
+ *		Reception_type is RX_MSDU_START_RECEPTION_TYPE_DL_MU_MIMO or
+ *		RX_MSDU_START_RECEPTION_TYPE_DL_MU_OFDMA_MIMO.
+ *
+ *		Bitmap, with each bit indicating if the related spatial
+ *		stream is used for this STA
+ *
+ *		LSB related to SS 0
+ *
+ *		0 - spatial stream not used for this reception
+ *		1 - spatial stream used for this reception
+ *
+ * ppdu_start_timestamp
+ *		Timestamp that indicates when the PPDU that contained this MPDU
+ *		started on the medium.
+ *
+ * phy_meta_data
+ *		SW programmed Meta data provided by the PHY. Can be used for SW
+ *		to indicate the channel the device is on.
+ */
+
+#define RX_MSDU_END_INFO0_RXPCU_MPDU_FITLER	GENMASK(1, 0)
+#define RX_MSDU_END_INFO0_SW_FRAME_GRP_ID	GENMASK(8, 2)
+
+#define RX_MSDU_END_INFO1_KEY_ID		GENMASK(7, 0)
+#define RX_MSDU_END_INFO1_CCE_SUPER_RULE	GENMASK(13, 8)
+#define RX_MSDU_END_INFO1_CCND_TRUNCATE		BIT(14)
+#define RX_MSDU_END_INFO1_CCND_CCE_DIS		BIT(15)
+#define RX_MSDU_END_INFO1_EXT_WAPI_PN		GENMASK(31, 16)
+
+#define RX_MSDU_END_INFO2_REPORTED_MPDU_LEN	GENMASK(13, 0)
+#define RX_MSDU_END_INFO2_FIRST_MSDU		BIT(14)
+#define RX_MSDU_END_INFO2_LAST_MSDU		BIT(15)
+#define RX_MSDU_END_INFO2_SA_IDX_TIMEOUT	BIT(16)
+#define RX_MSDU_END_INFO2_DA_IDX_TIMEOUT	BIT(17)
+#define RX_MSDU_END_INFO2_MSDU_LIMIT_ERR	BIT(18)
+#define RX_MSDU_END_INFO2_FLOW_IDX_TIMEOUT	BIT(19)
+#define RX_MSDU_END_INFO2_FLOW_IDX_INVALID	BIT(20)
+#define RX_MSDU_END_INFO2_WIFI_PARSER_ERR	BIT(21)
+#define RX_MSDU_END_INFO2_AMSDU_PARSET_ERR	BIT(22)
+#define RX_MSDU_END_INFO2_SA_IS_VALID		BIT(23)
+#define RX_MSDU_END_INFO2_DA_IS_VALID		BIT(24)
+#define RX_MSDU_END_INFO2_DA_IS_MCBC		BIT(25)
+#define RX_MSDU_END_INFO2_L3_HDR_PADDING	GENMASK(27, 26)
+
+#define RX_MSDU_END_INFO3_TCP_FLAG		GENMASK(8, 0)
+#define RX_MSDU_END_INFO3_LRO_ELIGIBLE		BIT(9)
+
+#define RX_MSDU_END_INFO4_DA_OFFSET		GENMASK(5, 0)
+#define RX_MSDU_END_INFO4_SA_OFFSET		GENMASK(11, 6)
+#define RX_MSDU_END_INFO4_DA_OFFSET_VALID	BIT(12)
+#define RX_MSDU_END_INFO4_SA_OFFSET_VALID	BIT(13)
+#define RX_MSDU_END_INFO4_L3_TYPE		GENMASK(31, 16)
+
+#define RX_MSDU_END_INFO5_MSDU_DROP		BIT(0)
+#define RX_MSDU_END_INFO5_REO_DEST_IND		GENMASK(5, 1)
+#define RX_MSDU_END_INFO5_FLOW_IDX		GENMASK(25, 6)
+
+struct rx_msdu_end {
+	__le16 info0;
+	__le16 phy_ppdu_id;
+	__le16 ip_hdr_cksum;
+	__le16 tcp_udp_cksum;
+	__le32 info1;
+	__le32 ext_wapi_pn[2];
+	__le32 info2;
+	__le32 ipv6_options_crc;
+	__le32 tcp_seq_num;
+	__le32 tcp_ack_num;
+	__le16 info3;
+	__le16 window_size;
+	__le32 info4;
+	__le32 rule_indication[2];
+	__le16 sa_idx;
+	__le16 da_idx;
+	__le32 info5;
+	__le32 fse_metadata;
+	__le16 cce_metadata;
+	__le16 sa_sw_peer_id;
+} __packed;
+
+/* rx_msdu_end
+ *
+ * rxpcu_mpdu_filter_in_category
+ *		Field indicates what the reason was that this mpdu frame
+ *		was allowed to come into the receive path by rxpcu. Values
+ *		are defined in enum %RX_DESC_RXPCU_FILTER_*.
+ *
+ * sw_frame_group_id
+ *		SW processes frames based on certain classifications. Values
+ *		are defined in enum %RX_DESC_SW_FRAME_GRP_ID_*.
+ *
+ * phy_ppdu_id
+ *		A ppdu counter value that PHY increments for every PPDU
+ *		received. The counter value wraps around.
+ *
+ * ip_hdr_cksum
+ *		This can include the IP header checksum or the pseudo
+ *		header checksum used by TCP/UDP checksum.
+ *
+ * tcp_udp_chksum
+ *		The value of the computed TCP/UDP checksum.  A mode bit
+ *		selects whether this checksum is the full checksum or the
+ *		partial checksum which does not include the pseudo header.
+ *
+ * key_id
+ *		The key ID octet from the IV. Only valid when first_msdu is set.
+ *
+ * cce_super_rule
+ *		Indicates the super filter rule.
+ *
+ * cce_classify_not_done_truncate
+ *		Classification failed due to truncated frame.
+ *
+ * cce_classify_not_done_cce_dis
+ *		Classification failed due to CCE global disable
+ *
+ * ext_wapi_pn*
+ *		Extension PN (packet number) which is only used by WAPI.
+ *
+ * reported_mpdu_length
+ *		MPDU length before decapsulation. Only valid when first_msdu is
+ *		set. This field is taken directly from the length field of the
+ *		A-MPDU delimiter or the preamble length field for non-A-MPDU
+ *		frames.
+ *
+ * first_msdu
+ *		Indicates the first MSDU of A-MSDU. If both first_msdu and
+ *		last_msdu are set in the MSDU then this is a non-aggregated MSDU
+ *		frame: normal MPDU. Interior MSDU in an A-MSDU shall have both
+ *		first_mpdu and last_mpdu bits set to 0.
+ *
+ * last_msdu
+ *		Indicates the last MSDU of the A-MSDU. MPDU end status is only
+ *		valid when last_msdu is set.
+ *
+ * sa_idx_timeout
+ *		Indicates an unsuccessful MAC source address search due to the
+ *		expiring of the search timer.
+ *
+ * da_idx_timeout
+ *		Indicates an unsuccessful MAC destination address search due to
+ *		the expiring of the search timer.
+ *
+ * msdu_limit_error
+ *		Indicates that the MSDU threshold was exceeded and thus all the
+ *		rest of the MSDUs will not be scattered and will not be
+ *		decapsulated but will be DMA'ed in RAW format as a single MSDU.
+ *
+ * flow_idx_timeout
+ *		Indicates an unsuccessful flow search due to the expiring of
+ *		the search timer.
+ *
+ * flow_idx_invalid
+ *		flow id is not valid.
+ *
+ * amsdu_parser_error
+ *		A-MSDU could not be properly de-agregated.
+ *
+ * sa_is_valid
+ *		Indicates that OLE found a valid SA entry.
+ *
+ * da_is_valid
+ *		Indicates that OLE found a valid DA entry.
+ *
+ * da_is_mcbc
+ *		Field Only valid if da_is_valid is set. Indicates the DA address
+ *		was a Multicast of Broadcast address.
+ *
+ * l3_header_padding
+ *		Number of bytes padded  to make sure that the L3 header will
+ *		always start of a Dword boundary.
+ *
+ * ipv6_options_crc
+ *		32 bit CRC computed out of  IP v6 extension headers.
+ *
+ * tcp_seq_number
+ *		TCP sequence number.
+ *
+ * tcp_ack_number
+ *		TCP acknowledge number.
+ *
+ * tcp_flag
+ *		TCP flags {NS, CWR, ECE, URG, ACK, PSH, RST, SYN, FIN}.
+ *
+ * lro_eligible
+ *		Computed out of TCP and IP fields to indicate that this
+ *		MSDU is eligible for LRO.
+ *
+ * window_size
+ *		TCP receive window size.
+ *
+ * da_offset
+ *		Offset into MSDU buffer for DA.
+ *
+ * sa_offset
+ *		Offset into MSDU buffer for SA.
+ *
+ * da_offset_valid
+ *		da_offset field is valid. This will be set to 0 in case
+ *		of a dynamic A-MSDU when DA is compressed.
+ *
+ * sa_offset_valid
+ *		sa_offset field is valid. This will be set to 0 in case
+ *		of a dynamic A-MSDU when SA is compressed.
+ *
+ * l3_type
+ *		The 16-bit type value indicating the type of L3 later
+ *		extracted from LLC/SNAP, set to zero if SNAP is not
+ *		available.
+ *
+ * rule_indication
+ *		Bitmap indicating which of rules have matched.
+ *
+ * sa_idx
+ *		The offset in the address table which matches MAC source address
+ *
+ * da_idx
+ *		The offset in the address table which matches MAC destination
+ *		address.
+ *
+ * msdu_drop
+ *		REO shall drop this MSDU and not forward it to any other ring.
+ *
+ * reo_destination_indication
+ *		The id of the reo exit ring where the msdu frame shall push
+ *		after (MPDU level) reordering has finished. Values are defined
+ *		in enum %HAL_RX_MSDU_DESC_REO_DEST_IND_.
+ *
+ * flow_idx
+ *		Flow table index.
+ *
+ * fse_metadata
+ *		FSE related meta data.
+ *
+ * cce_metadata
+ *		CCE related meta data.
+ *
+ * sa_sw_peer_id
+ *		sw_peer_id from the address search entry corresponding to the
+ *		source address of the MSDU.
+ */
+
+enum rx_mpdu_end_rxdma_dest_ring {
+	RX_MPDU_END_RXDMA_DEST_RING_RELEASE,
+	RX_MPDU_END_RXDMA_DEST_RING_FW,
+	RX_MPDU_END_RXDMA_DEST_RING_SW,
+	RX_MPDU_END_RXDMA_DEST_RING_REO,
+};
+
+#define RX_MPDU_END_INFO1_UNSUP_KTYPE_SHORT_FRAME	BIT(11)
+#define RX_MPDU_END_INFO1_RX_IN_TX_DECRYPT_BYT		BIT(12)
+#define RX_MPDU_END_INFO1_OVERFLOW_ERR			BIT(13)
+#define RX_MPDU_END_INFO1_MPDU_LEN_ERR			BIT(14)
+#define RX_MPDU_END_INFO1_TKIP_MIC_ERR			BIT(15)
+#define RX_MPDU_END_INFO1_DECRYPT_ERR			BIT(16)
+#define RX_MPDU_END_INFO1_UNENCRYPTED_FRAME_ERR		BIT(17)
+#define RX_MPDU_END_INFO1_PN_FIELDS_VALID		BIT(18)
+#define RX_MPDU_END_INFO1_FCS_ERR			BIT(19)
+#define RX_MPDU_END_INFO1_MSDU_LEN_ERR			BIT(20)
+#define RX_MPDU_END_INFO1_RXDMA0_DEST_RING		GENMASK(22, 21)
+#define RX_MPDU_END_INFO1_RXDMA1_DEST_RING		GENMASK(24, 23)
+#define RX_MPDU_END_INFO1_DECRYPT_STATUS_CODE		GENMASK(27, 25)
+#define RX_MPDU_END_INFO1_RX_BITMAP_NOT_UPD		BIT(28)
+
+struct rx_mpdu_end {
+	__le16 info0;
+	__le16 phy_ppdu_id;
+	__le32 info1;
+} __packed;
+
+/* rx_mpdu_end
+ *
+ * rxpcu_mpdu_filter_in_category
+ *		Field indicates what the reason was that this mpdu frame
+ *		was allowed to come into the receive path by rxpcu. Values
+ *		are defined in enum %RX_DESC_RXPCU_FILTER_*.
+ *
+ * sw_frame_group_id
+ *		SW processes frames based on certain classifications. Values
+ *		are defined in enum %RX_DESC_SW_FRAME_GRP_ID_*.
+ *
+ * phy_ppdu_id
+ *		A ppdu counter value that PHY increments for every PPDU
+ *		received. The counter value wraps around.
+ *
+ * unsup_ktype_short_frame
+ *		This bit will be '1' when WEP or TKIP or WAPI key type is
+ *		received for 11ah short frame. Crypto will bypass the received
+ *		packet without decryption to RxOLE after setting this bit.
+ *
+ * rx_in_tx_decrypt_byp
+ *		Indicates that RX packet is not decrypted as Crypto is
+ *		busy with TX packet processing.
+ *
+ * overflow_err
+ *		RXPCU Receive FIFO ran out of space to receive the full MPDU.
+ *		Therefore this MPDU is terminated early and is thus corrupted.
+ *
+ *		This MPDU will not be ACKed.
+ *
+ *		RXPCU might still be able to correctly receive the following
+ *		MPDUs in the PPDU if enough fifo space became available in time.
+ *
+ * mpdu_length_err
+ *		Set by RXPCU if the expected MPDU length does not correspond
+ *		with the actually received number of bytes in the MPDU.
+ *
+ * tkip_mic_err
+ *		Set by Rx crypto when crypto detected a TKIP MIC error for
+ *		this MPDU.
+ *
+ * decrypt_err
+ *		Set by RX CRYPTO when CRYPTO detected a decrypt error for this
+ *		MPDU or CRYPTO received an encrypted frame, but did not get a
+ *		valid corresponding key id in the peer entry.
+ *
+ * unencrypted_frame_err
+ *		Set by RX CRYPTO when CRYPTO detected an unencrypted frame while
+ *		in the peer entry field 'All_frames_shall_be_encrypted' is set.
+ *
+ * pn_fields_contain_valid_info
+ *		Set by RX CRYPTO to indicate that there is a valid PN field
+ *		present in this MPDU.
+ *
+ * fcs_err
+ *		Set by RXPCU when there is an FCS error detected for this MPDU.
+ *
+ * msdu_length_err
+ *		Set by RXOLE when there is an msdu length error detected
+ *		in at least 1 of the MSDUs embedded within the MPDU.
+ *
+ * rxdma0_destination_ring
+ * rxdma1_destination_ring
+ *		The ring to which RXDMA0/1 shall push the frame, assuming
+ *		no MPDU level errors are detected. In case of MPDU level
+ *		errors, RXDMA0/1 might change the RXDMA0/1 destination. Values
+ *		are defined in %enum RX_MPDU_END_RXDMA_DEST_RING_*.
+ *
+ * decrypt_status_code
+ *		Field provides insight into the decryption performed. Values
+ *		are defined in enum %RX_DESC_DECRYPT_STATUS_CODE_*.
+ *
+ * rx_bitmap_not_updated
+ *		Frame is received, but RXPCU could not update the receive bitmap
+ *		due to (temporary) fifo constraints.
+ */
+
+/* Padding bytes to avoid TLV's spanning across 128 byte boundary */
+#define HAL_RX_DESC_PADDING0_BYTES	4
+#define HAL_RX_DESC_PADDING1_BYTES	16
+
+#define HAL_RX_DESC_HDR_STATUS_LEN	120
+
+struct hal_rx_desc {
+	__le32 msdu_end_tag;
+	struct rx_msdu_end msdu_end;
+	__le32 rx_attn_tag;
+	struct rx_attention attention;
+	__le32 msdu_start_tag;
+	struct rx_msdu_start msdu_start;
+	u8 rx_padding0[HAL_RX_DESC_PADDING0_BYTES];
+	__le32 mpdu_start_tag;
+	struct rx_mpdu_start mpdu_start;
+	__le32 mpdu_end_tag;
+	struct rx_mpdu_end mpdu_end;
+	u8 rx_padding1[HAL_RX_DESC_PADDING1_BYTES];
+	__le32 hdr_status_tag;
+	__le32 phy_ppdu_id;
+	u8 hdr_status[HAL_RX_DESC_HDR_STATUS_LEN];
+	u8 msdu_payload[0];
+} __packed;
+
+#endif /* ATH11K_RX_DESC_H */
diff --git a/drivers/net/wireless/ath/ath11k/testmode.c b/drivers/net/wireless/ath/ath11k/testmode.c
new file mode 100644
index 000000000000..932bee6d1740
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/testmode.c
@@ -0,0 +1,199 @@
+// SPDX-License-Identifier: BSD-3-Clause-Clear
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#include "testmode.h"
+#include <net/netlink.h>
+#include "debug.h"
+#include "wmi.h"
+#include "hw.h"
+#include "core.h"
+#include "testmode_i.h"
+
+static const struct nla_policy ath11k_tm_policy[ATH11K_TM_ATTR_MAX + 1] = {
+	[ATH11K_TM_ATTR_CMD]		= { .type = NLA_U32 },
+	[ATH11K_TM_ATTR_DATA]		= { .type = NLA_BINARY,
+					    .len = ATH11K_TM_DATA_MAX_LEN },
+	[ATH11K_TM_ATTR_WMI_CMDID]	= { .type = NLA_U32 },
+	[ATH11K_TM_ATTR_VERSION_MAJOR]	= { .type = NLA_U32 },
+	[ATH11K_TM_ATTR_VERSION_MINOR]	= { .type = NLA_U32 },
+};
+
+/* Returns true if callee consumes the skb and the skb should be discarded.
+ * Returns false if skb is not used. Does not sleep.
+ */
+bool ath11k_tm_event_wmi(struct ath11k *ar, u32 cmd_id, struct sk_buff *skb)
+{
+	struct sk_buff *nl_skb;
+	bool consumed;
+	int ret;
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_TESTMODE,
+		   "testmode event wmi cmd_id %d skb %pK skb->len %d\n",
+		   cmd_id, skb, skb->len);
+
+	ath11k_dbg_dump(ar->ab, ATH11K_DBG_TESTMODE, NULL, "", skb->data, skb->len);
+
+	spin_lock_bh(&ar->data_lock);
+
+	consumed = true;
+
+	nl_skb = cfg80211_testmode_alloc_event_skb(ar->hw->wiphy,
+						   2 * sizeof(u32) + skb->len,
+						   GFP_ATOMIC);
+	if (!nl_skb) {
+		ath11k_warn(ar->ab,
+			    "failed to allocate skb for testmode wmi event\n");
+		goto out;
+	}
+
+	ret = nla_put_u32(nl_skb, ATH11K_TM_ATTR_CMD, ATH11K_TM_CMD_WMI);
+	if (ret) {
+		ath11k_warn(ar->ab,
+			    "failed to to put testmode wmi event cmd attribute: %d\n",
+			    ret);
+		kfree_skb(nl_skb);
+		goto out;
+	}
+
+	ret = nla_put_u32(nl_skb, ATH11K_TM_ATTR_WMI_CMDID, cmd_id);
+	if (ret) {
+		ath11k_warn(ar->ab,
+			    "failed to to put testmode wmi even cmd_id: %d\n",
+			    ret);
+		kfree_skb(nl_skb);
+		goto out;
+	}
+
+	ret = nla_put(nl_skb, ATH11K_TM_ATTR_DATA, skb->len, skb->data);
+	if (ret) {
+		ath11k_warn(ar->ab,
+			    "failed to copy skb to testmode wmi event: %d\n",
+			    ret);
+		kfree_skb(nl_skb);
+		goto out;
+	}
+
+	cfg80211_testmode_event(nl_skb, GFP_ATOMIC);
+
+out:
+	spin_unlock_bh(&ar->data_lock);
+
+	return consumed;
+}
+
+static int ath11k_tm_cmd_get_version(struct ath11k *ar, struct nlattr *tb[])
+{
+	struct sk_buff *skb;
+	int ret;
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_TESTMODE,
+		   "testmode cmd get version_major %d version_minor %d\n",
+		   ATH11K_TESTMODE_VERSION_MAJOR,
+		   ATH11K_TESTMODE_VERSION_MINOR);
+
+	skb = cfg80211_testmode_alloc_reply_skb(ar->hw->wiphy,
+						nla_total_size(sizeof(u32)));
+	if (!skb)
+		return -ENOMEM;
+
+	ret = nla_put_u32(skb, ATH11K_TM_ATTR_VERSION_MAJOR,
+			  ATH11K_TESTMODE_VERSION_MAJOR);
+	if (ret) {
+		kfree_skb(skb);
+		return ret;
+	}
+
+	ret = nla_put_u32(skb, ATH11K_TM_ATTR_VERSION_MINOR,
+			  ATH11K_TESTMODE_VERSION_MINOR);
+	if (ret) {
+		kfree_skb(skb);
+		return ret;
+	}
+
+	return cfg80211_testmode_reply(skb);
+}
+
+static int ath11k_tm_cmd_wmi(struct ath11k *ar, struct nlattr *tb[])
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct sk_buff *skb;
+	u32 cmd_id, buf_len;
+	int ret;
+	void *buf;
+
+	mutex_lock(&ar->conf_mutex);
+
+	if (ar->state != ATH11K_STATE_ON) {
+		ret = -ENETDOWN;
+		goto out;
+	}
+
+	if (!tb[ATH11K_TM_ATTR_DATA]) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	if (!tb[ATH11K_TM_ATTR_WMI_CMDID]) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	buf = nla_data(tb[ATH11K_TM_ATTR_DATA]);
+	buf_len = nla_len(tb[ATH11K_TM_ATTR_DATA]);
+	cmd_id = nla_get_u32(tb[ATH11K_TM_ATTR_WMI_CMDID]);
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_TESTMODE,
+		   "testmode cmd wmi cmd_id %d buf %pK buf_len %d\n",
+		   cmd_id, buf, buf_len);
+
+	ath11k_dbg_dump(ar->ab, ATH11K_DBG_TESTMODE, NULL, "", buf, buf_len);
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, buf_len);
+	if (!skb) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	memcpy(skb->data, buf, buf_len);
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, cmd_id);
+	if (ret) {
+		dev_kfree_skb(skb);
+		ath11k_warn(ar->ab, "failed to transmit wmi command (testmode): %d\n",
+			    ret);
+		goto out;
+	}
+
+	ret = 0;
+
+out:
+	mutex_unlock(&ar->conf_mutex);
+	return ret;
+}
+
+int ath11k_tm_cmd(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+		  void *data, int len)
+{
+	struct ath11k *ar = hw->priv;
+	struct nlattr *tb[ATH11K_TM_ATTR_MAX + 1];
+	int ret;
+
+	ret = nla_parse(tb, ATH11K_TM_ATTR_MAX, data, len, ath11k_tm_policy,
+			NULL);
+	if (ret)
+		return ret;
+
+	if (!tb[ATH11K_TM_ATTR_CMD])
+		return -EINVAL;
+
+	switch (nla_get_u32(tb[ATH11K_TM_ATTR_CMD])) {
+	case ATH11K_TM_CMD_GET_VERSION:
+		return ath11k_tm_cmd_get_version(ar, tb);
+	case ATH11K_TM_CMD_WMI:
+		return ath11k_tm_cmd_wmi(ar, tb);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
diff --git a/drivers/net/wireless/ath/ath11k/testmode.h b/drivers/net/wireless/ath/ath11k/testmode.h
new file mode 100644
index 000000000000..aaa122ed9069
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/testmode.h
@@ -0,0 +1,29 @@
+/* SPDX-License-Identifier: BSD-3-Clause-Clear */
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#include "core.h"
+
+#ifdef CONFIG_NL80211_TESTMODE
+
+bool ath11k_tm_event_wmi(struct ath11k *ar, u32 cmd_id, struct sk_buff *skb);
+int ath11k_tm_cmd(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+		  void *data, int len);
+
+#else
+
+static inline bool ath11k_tm_event_wmi(struct ath11k *ar, u32 cmd_id,
+				       struct sk_buff *skb)
+{
+	return false;
+}
+
+static inline int ath11k_tm_cmd(struct ieee80211_hw *hw,
+				struct ieee80211_vif *vif,
+				void *data, int len)
+{
+	return 0;
+}
+
+#endif
diff --git a/drivers/net/wireless/ath/ath11k/testmode_i.h b/drivers/net/wireless/ath/ath11k/testmode_i.h
new file mode 100644
index 000000000000..4bae2a9eeea4
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/testmode_i.h
@@ -0,0 +1,50 @@
+/* SPDX-License-Identifier: BSD-3-Clause-Clear */
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+/* "API" level of the ath11k testmode interface. Bump it after every
+ * incompatible interface change.
+ */
+#define ATH11K_TESTMODE_VERSION_MAJOR 1
+
+/* Bump this after every _compatible_ interface change, for example
+ * addition of a new command or an attribute.
+ */
+#define ATH11K_TESTMODE_VERSION_MINOR 0
+
+#define ATH11K_TM_DATA_MAX_LEN		5000
+
+enum ath11k_tm_attr {
+	__ATH11K_TM_ATTR_INVALID		= 0,
+	ATH11K_TM_ATTR_CMD			= 1,
+	ATH11K_TM_ATTR_DATA			= 2,
+	ATH11K_TM_ATTR_WMI_CMDID		= 3,
+	ATH11K_TM_ATTR_VERSION_MAJOR		= 4,
+	ATH11K_TM_ATTR_VERSION_MINOR		= 5,
+	ATH11K_TM_ATTR_WMI_OP_VERSION		= 6,
+
+	/* keep last */
+	__ATH11K_TM_ATTR_AFTER_LAST,
+	ATH11K_TM_ATTR_MAX		= __ATH11K_TM_ATTR_AFTER_LAST - 1,
+};
+
+/* All ath11k testmode interface commands specified in
+ * ATH11K_TM_ATTR_CMD
+ */
+enum ath11k_tm_cmd {
+	/* Returns the supported ath11k testmode interface version in
+	 * ATH11K_TM_ATTR_VERSION. Always guaranteed to work. User space
+	 * uses this to verify it's using the correct version of the
+	 * testmode interface
+	 */
+	ATH11K_TM_CMD_GET_VERSION = 0,
+
+	/* The command used to transmit a WMI command to the firmware and
+	 * the event to receive WMI events from the firmware. Without
+	 * struct wmi_cmd_hdr header, only the WMI payload. Command id is
+	 * provided with ATH11K_TM_ATTR_WMI_CMDID and payload in
+	 * ATH11K_TM_ATTR_DATA.
+	 */
+	ATH11K_TM_CMD_WMI = 1,
+};
diff --git a/drivers/net/wireless/ath/ath11k/trace.c b/drivers/net/wireless/ath/ath11k/trace.c
new file mode 100644
index 000000000000..f0cc49ba0387
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/trace.c
@@ -0,0 +1,9 @@
+// SPDX-License-Identifier: BSD-3-Clause-Clear
+/*
+ * Copyright (c) 2019 The Linux Foundation. All rights reserved.
+ */
+
+#include <linux/module.h>
+
+#define CREATE_TRACE_POINTS
+#include "trace.h"
diff --git a/drivers/net/wireless/ath/ath11k/trace.h b/drivers/net/wireless/ath/ath11k/trace.h
new file mode 100644
index 000000000000..400f759568e1
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/trace.h
@@ -0,0 +1,110 @@
+/* SPDX-License-Identifier: BSD-3-Clause-Clear */
+/*
+ * Copyright (c) 2019 The Linux Foundation. All rights reserved.
+ */
+
+#if !defined(_TRACE_H_) || defined(TRACE_HEADER_MULTI_READ)
+
+#include <linux/tracepoint.h>
+#include "core.h"
+
+#define _TRACE_H_
+
+/* create empty functions when tracing is disabled */
+#if !defined(CONFIG_ATH11K_TRACING)
+#undef TRACE_EVENT
+#define TRACE_EVENT(name, proto, ...) \
+static inline void trace_ ## name(proto) {}
+#endif /* !CONFIG_ATH11K_TRACING || __CHECKER__ */
+
+TRACE_EVENT(ath11k_htt_pktlog,
+	    TP_PROTO(struct ath11k *ar, const void *buf, u16 buf_len),
+
+	TP_ARGS(ar, buf, buf_len),
+
+	TP_STRUCT__entry(
+		__string(device, dev_name(ar->ab->dev))
+		__string(driver, dev_driver_string(ar->ab->dev))
+		__field(u16, buf_len)
+		__dynamic_array(u8, pktlog, buf_len)
+	),
+
+	TP_fast_assign(
+		__assign_str(device, dev_name(ar->ab->dev));
+		__assign_str(driver, dev_driver_string(ar->ab->dev));
+		__entry->buf_len = buf_len;
+		memcpy(__get_dynamic_array(pktlog), buf, buf_len);
+	),
+
+	TP_printk(
+		"%s %s size %hu",
+		__get_str(driver),
+		__get_str(device),
+		__entry->buf_len
+	 )
+);
+
+TRACE_EVENT(ath11k_htt_ppdu_stats,
+	    TP_PROTO(struct ath11k *ar, const void *data, size_t len),
+
+	TP_ARGS(ar, data, len),
+
+	TP_STRUCT__entry(
+		__string(device, dev_name(ar->ab->dev))
+		__string(driver, dev_driver_string(ar->ab->dev))
+		__field(u16, len)
+		__dynamic_array(u8, ppdu, len)
+	),
+
+	TP_fast_assign(
+		__assign_str(device, dev_name(ar->ab->dev));
+		__assign_str(driver, dev_driver_string(ar->ab->dev));
+		__entry->len = len;
+		memcpy(__get_dynamic_array(ppdu), data, len);
+	),
+
+	TP_printk(
+		"%s %s ppdu len %d",
+		__get_str(driver),
+		__get_str(device),
+		__entry->len
+	 )
+);
+
+TRACE_EVENT(ath11k_htt_rxdesc,
+	    TP_PROTO(struct ath11k *ar, const void *data, size_t len),
+
+	TP_ARGS(ar, data, len),
+
+	TP_STRUCT__entry(
+		__string(device, dev_name(ar->ab->dev))
+		__string(driver, dev_driver_string(ar->ab->dev))
+		__field(u16, len)
+		__dynamic_array(u8, rxdesc, len)
+	),
+
+	TP_fast_assign(
+		__assign_str(device, dev_name(ar->ab->dev));
+		__assign_str(driver, dev_driver_string(ar->ab->dev));
+		__entry->len = len;
+		memcpy(__get_dynamic_array(rxdesc), data, len);
+	),
+
+	TP_printk(
+		"%s %s rxdesc len %d",
+		__get_str(driver),
+		__get_str(device),
+		__entry->len
+	 )
+);
+
+#endif /* _TRACE_H_ || TRACE_HEADER_MULTI_READ*/
+
+/* we don't want to use include/trace/events */
+#undef TRACE_INCLUDE_PATH
+#define TRACE_INCLUDE_PATH .
+#undef TRACE_INCLUDE_FILE
+#define TRACE_INCLUDE_FILE trace
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
diff --git a/drivers/net/wireless/ath/ath11k/wmi.c b/drivers/net/wireless/ath/ath11k/wmi.c
new file mode 100644
index 000000000000..3d1d4d3c3f20
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/wmi.c
@@ -0,0 +1,5632 @@
+// SPDX-License-Identifier: BSD-3-Clause-Clear
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+#include <linux/skbuff.h>
+#include <linux/ctype.h>
+#include <net/mac80211.h>
+#include <net/cfg80211.h>
+#include <linux/completion.h>
+#include <linux/if_ether.h>
+#include <linux/types.h>
+#include <linux/pci.h>
+#include <linux/uuid.h>
+#include <linux/time.h>
+#include <linux/of.h>
+#include "core.h"
+#include "debug.h"
+#include "mac.h"
+#include "hw.h"
+#include "peer.h"
+
+struct wmi_tlv_policy {
+	size_t min_len;
+};
+
+struct wmi_tlv_svc_ready_parse {
+	bool wmi_svc_bitmap_done;
+};
+
+struct wmi_tlv_svc_rdy_ext_parse {
+	struct ath11k_service_ext_param param;
+	struct wmi_soc_mac_phy_hw_mode_caps *hw_caps;
+	struct wmi_hw_mode_capabilities *hw_mode_caps;
+	u32 n_hw_mode_caps;
+	u32 tot_phy_id;
+	struct wmi_hw_mode_capabilities pref_hw_mode_caps;
+	struct wmi_mac_phy_capabilities *mac_phy_caps;
+	u32 n_mac_phy_caps;
+	struct wmi_soc_hal_reg_capabilities *soc_hal_reg_caps;
+	struct wmi_hal_reg_capabilities_ext *ext_hal_reg_caps;
+	u32 n_ext_hal_reg_caps;
+	bool hw_mode_done;
+	bool mac_phy_done;
+	bool ext_hal_reg_done;
+};
+
+struct wmi_tlv_rdy_parse {
+	u32 num_extra_mac_addr;
+};
+
+static const struct wmi_tlv_policy wmi_tlv_policies[] = {
+	[WMI_TAG_ARRAY_BYTE]
+		= { .min_len = 0 },
+	[WMI_TAG_ARRAY_UINT32]
+		= { .min_len = 0 },
+	[WMI_TAG_SERVICE_READY_EVENT]
+		= { .min_len = sizeof(struct wmi_service_ready_event) },
+	[WMI_TAG_SERVICE_READY_EXT_EVENT]
+		= { .min_len =  sizeof(struct wmi_service_ready_ext_event) },
+	[WMI_TAG_SOC_MAC_PHY_HW_MODE_CAPS]
+		= { .min_len = sizeof(struct wmi_soc_mac_phy_hw_mode_caps) },
+	[WMI_TAG_SOC_HAL_REG_CAPABILITIES]
+		= { .min_len = sizeof(struct wmi_soc_hal_reg_capabilities) },
+	[WMI_TAG_VDEV_START_RESPONSE_EVENT]
+		= { .min_len = sizeof(struct wmi_vdev_start_resp_event) },
+	[WMI_TAG_PEER_DELETE_RESP_EVENT]
+		= { .min_len = sizeof(struct wmi_peer_delete_resp_event) },
+	[WMI_TAG_OFFLOAD_BCN_TX_STATUS_EVENT]
+		= { .min_len = sizeof(struct wmi_bcn_tx_status_event) },
+	[WMI_TAG_VDEV_STOPPED_EVENT]
+		= { .min_len = sizeof(struct wmi_vdev_stopped_event) },
+	[WMI_TAG_REG_CHAN_LIST_CC_EVENT]
+		= { .min_len = sizeof(struct wmi_reg_chan_list_cc_event) },
+	[WMI_TAG_MGMT_RX_HDR]
+		= { .min_len = sizeof(struct wmi_mgmt_rx_hdr) },
+	[WMI_TAG_MGMT_TX_COMPL_EVENT]
+		= { .min_len = sizeof(struct wmi_mgmt_tx_compl_event) },
+	[WMI_TAG_SCAN_EVENT]
+		= { .min_len = sizeof(struct wmi_scan_event) },
+	[WMI_TAG_PEER_STA_KICKOUT_EVENT]
+		= { .min_len = sizeof(struct wmi_peer_sta_kickout_event) },
+	[WMI_TAG_ROAM_EVENT]
+		= { .min_len = sizeof(struct wmi_roam_event) },
+	[WMI_TAG_CHAN_INFO_EVENT]
+		= { .min_len = sizeof(struct wmi_chan_info_event) },
+	[WMI_TAG_PDEV_BSS_CHAN_INFO_EVENT]
+		= { .min_len = sizeof(struct wmi_pdev_bss_chan_info_event) },
+	[WMI_TAG_VDEV_INSTALL_KEY_COMPLETE_EVENT]
+		= { .min_len = sizeof(struct wmi_vdev_install_key_compl_event) },
+	[WMI_TAG_READY_EVENT]
+		= {.min_len = sizeof(struct wmi_ready_event) },
+	[WMI_TAG_SERVICE_AVAILABLE_EVENT]
+		= {.min_len = sizeof(struct wmi_service_available_event) },
+	[WMI_TAG_PEER_ASSOC_CONF_EVENT]
+		= { .min_len = sizeof(struct wmi_peer_assoc_conf_event) },
+	[WMI_TAG_STATS_EVENT]
+		= { .min_len = sizeof(struct wmi_stats_event) },
+	[WMI_TAG_PDEV_CTL_FAILSAFE_CHECK_EVENT]
+		= { .min_len = sizeof(struct wmi_pdev_ctl_failsafe_chk_event) },
+};
+
+#define PRIMAP(_hw_mode_) \
+	[_hw_mode_] = _hw_mode_##_PRI
+
+static const int ath11k_hw_mode_pri_map[] = {
+	PRIMAP(WMI_HOST_HW_MODE_SINGLE),
+	PRIMAP(WMI_HOST_HW_MODE_DBS),
+	PRIMAP(WMI_HOST_HW_MODE_SBS_PASSIVE),
+	PRIMAP(WMI_HOST_HW_MODE_SBS),
+	PRIMAP(WMI_HOST_HW_MODE_DBS_SBS),
+	PRIMAP(WMI_HOST_HW_MODE_DBS_OR_SBS),
+	/* keep last */
+	PRIMAP(WMI_HOST_HW_MODE_MAX),
+};
+
+static int
+ath11k_wmi_tlv_iter(struct ath11k_base *ab, const void *ptr, size_t len,
+		    int (*iter)(struct ath11k_base *ab, u16 tag, u16 len,
+				const void *ptr, void *data),
+		    void *data)
+{
+	const void *begin = ptr;
+	const struct wmi_tlv *tlv;
+	u16 tlv_tag, tlv_len;
+	int ret;
+
+	while (len > 0) {
+		if (len < sizeof(*tlv)) {
+			ath11k_err(ab, "wmi tlv parse failure at byte %zd (%zu bytes left, %zu expected)\n",
+				   ptr - begin, len, sizeof(*tlv));
+			return -EINVAL;
+		}
+
+		tlv = ptr;
+		tlv_tag = FIELD_GET(WMI_TLV_TAG, tlv->header);
+		tlv_len = FIELD_GET(WMI_TLV_LEN, tlv->header);
+		ptr += sizeof(*tlv);
+		len -= sizeof(*tlv);
+
+		if (tlv_len > len) {
+			ath11k_err(ab, "wmi tlv parse failure of tag %hhu at byte %zd (%zu bytes left, %hhu expected)\n",
+				   tlv_tag, ptr - begin, len, tlv_len);
+			return -EINVAL;
+		}
+
+		if (tlv_tag < ARRAY_SIZE(wmi_tlv_policies) &&
+		    wmi_tlv_policies[tlv_tag].min_len &&
+		    wmi_tlv_policies[tlv_tag].min_len > tlv_len) {
+			ath11k_err(ab, "wmi tlv parse failure of tag %hhu at byte %zd (%hhu bytes is less than min length %zu)\n",
+				   tlv_tag, ptr - begin, tlv_len,
+				   wmi_tlv_policies[tlv_tag].min_len);
+			return -EINVAL;
+		}
+
+		ret = iter(ab, tlv_tag, tlv_len, ptr, data);
+		if (ret)
+			return ret;
+
+		ptr += tlv_len;
+		len -= tlv_len;
+	}
+
+	return 0;
+}
+
+static int ath11k_wmi_tlv_iter_parse(struct ath11k_base *ab, u16 tag, u16 len,
+				     const void *ptr, void *data)
+{
+	const void **tb = data;
+
+	if (tag < WMI_TAG_MAX)
+		tb[tag] = ptr;
+
+	return 0;
+}
+
+static int ath11k_wmi_tlv_parse(struct ath11k_base *ar, const void **tb,
+				const void *ptr, size_t len)
+{
+	return ath11k_wmi_tlv_iter(ar, ptr, len, ath11k_wmi_tlv_iter_parse,
+				   (void *)tb);
+}
+
+static const void **
+ath11k_wmi_tlv_parse_alloc(struct ath11k_base *ab, const void *ptr,
+			   size_t len, gfp_t gfp)
+{
+	const void **tb;
+	int ret;
+
+	tb = kcalloc(WMI_TAG_MAX, sizeof(*tb), gfp);
+	if (!tb)
+		return ERR_PTR(-ENOMEM);
+
+	ret = ath11k_wmi_tlv_parse(ab, tb, ptr, len);
+	if (ret) {
+		kfree(tb);
+		return ERR_PTR(ret);
+	}
+
+	return tb;
+}
+
+static int ath11k_wmi_cmd_send_nowait(struct ath11k_pdev_wmi *wmi, struct sk_buff *skb,
+				      u32 cmd_id)
+{
+	struct ath11k_skb_cb *skb_cb = ATH11K_SKB_CB(skb);
+	struct ath11k_base *ab = wmi->wmi_sc->ab;
+	struct wmi_cmd_hdr *cmd_hdr;
+	int ret;
+	u32 cmd = 0;
+
+	if (skb_push(skb, sizeof(struct wmi_cmd_hdr)) == NULL)
+		return -ENOMEM;
+
+	cmd |= FIELD_PREP(WMI_CMD_HDR_CMD_ID, cmd_id);
+
+	cmd_hdr = (struct wmi_cmd_hdr *)skb->data;
+	cmd_hdr->cmd_id = cmd;
+
+	memset(skb_cb, 0, sizeof(*skb_cb));
+	ret = ath11k_htc_send(&ab->htc, wmi->eid, skb);
+
+	if (ret)
+		goto err_pull;
+
+	return 0;
+
+err_pull:
+	skb_pull(skb, sizeof(struct wmi_cmd_hdr));
+	return ret;
+}
+
+int ath11k_wmi_cmd_send(struct ath11k_pdev_wmi *wmi, struct sk_buff *skb,
+			u32 cmd_id)
+{
+	struct ath11k_wmi_base *wmi_sc = wmi->wmi_sc;
+	int ret = -EOPNOTSUPP;
+
+	might_sleep();
+
+	wait_event_timeout(wmi_sc->tx_credits_wq, ({
+		ret = ath11k_wmi_cmd_send_nowait(wmi, skb, cmd_id);
+
+		if (ret && test_bit(ATH11K_FLAG_CRASH_FLUSH, &wmi_sc->ab->dev_flags))
+			ret = -ESHUTDOWN;
+
+		(ret != -EAGAIN);
+	}), WMI_SEND_TIMEOUT_HZ);
+
+	if (ret == -EAGAIN)
+		ath11k_warn(wmi_sc->ab, "wmi command %d timeout\n", cmd_id);
+
+	return ret;
+}
+
+static int ath11k_pull_svc_ready_ext(struct ath11k_pdev_wmi *wmi_handle,
+				     const void *ptr,
+				     struct ath11k_service_ext_param *param)
+{
+	const struct wmi_service_ready_ext_event *ev = ptr;
+
+	if (!ev)
+		return -EINVAL;
+
+	/* Move this to host based bitmap */
+	param->default_conc_scan_config_bits = ev->default_conc_scan_config_bits;
+	param->default_fw_config_bits =	ev->default_fw_config_bits;
+	param->he_cap_info = ev->he_cap_info;
+	param->mpdu_density = ev->mpdu_density;
+	param->max_bssid_rx_filters = ev->max_bssid_rx_filters;
+	memcpy(&param->ppet, &ev->ppet, sizeof(param->ppet));
+
+	return 0;
+}
+
+static int
+ath11k_pull_mac_phy_cap_svc_ready_ext(struct ath11k_pdev_wmi *wmi_handle,
+				      struct wmi_soc_mac_phy_hw_mode_caps *hw_caps,
+				      struct wmi_hw_mode_capabilities *wmi_hw_mode_caps,
+				      struct wmi_soc_hal_reg_capabilities *hal_reg_caps,
+				      struct wmi_mac_phy_capabilities *wmi_mac_phy_caps,
+				      u8 hw_mode_id, u8 phy_id,
+				      struct ath11k_pdev *pdev)
+{
+	struct wmi_mac_phy_capabilities *mac_phy_caps;
+	struct ath11k_band_cap *cap_band;
+	struct ath11k_pdev_cap *pdev_cap = &pdev->cap;
+	u32 phy_map;
+	u32 hw_idx, phy_idx = 0;
+
+	if (!hw_caps || !wmi_hw_mode_caps || !hal_reg_caps)
+		return -EINVAL;
+
+	for (hw_idx = 0; hw_idx < hw_caps->num_hw_modes; hw_idx++) {
+		if (hw_mode_id == wmi_hw_mode_caps[hw_idx].hw_mode_id)
+			break;
+
+		phy_map = wmi_hw_mode_caps[hw_idx].phy_id_map;
+		while (phy_map) {
+			phy_map >>= 1;
+			phy_idx++;
+		}
+	}
+
+	if (hw_idx == hw_caps->num_hw_modes)
+		return -EINVAL;
+
+	phy_idx += phy_id;
+	if (phy_id >= hal_reg_caps->num_phy)
+		return -EINVAL;
+
+	mac_phy_caps = &wmi_mac_phy_caps[phy_idx];
+
+	pdev->pdev_id = mac_phy_caps->pdev_id;
+	pdev_cap->supported_bands = mac_phy_caps->supported_bands;
+	pdev_cap->ampdu_density = mac_phy_caps->ampdu_density;
+
+	/* Take non-zero tx/rx chainmask. If tx/rx chainmask differs from
+	 * band to band for a single radio, need to see how this should be
+	 * handled.
+	 */
+	if (mac_phy_caps->supported_bands & WMI_HOST_WLAN_2G_CAP) {
+		pdev_cap->tx_chain_mask = mac_phy_caps->tx_chain_mask_2g;
+		pdev_cap->rx_chain_mask = mac_phy_caps->rx_chain_mask_2g;
+	} else if (mac_phy_caps->supported_bands & WMI_HOST_WLAN_5G_CAP) {
+		pdev_cap->vht_cap = mac_phy_caps->vht_cap_info_5g;
+		pdev_cap->vht_mcs = mac_phy_caps->vht_supp_mcs_5g;
+		pdev_cap->he_mcs = mac_phy_caps->he_supp_mcs_5g;
+		pdev_cap->tx_chain_mask = mac_phy_caps->tx_chain_mask_5g;
+		pdev_cap->rx_chain_mask = mac_phy_caps->rx_chain_mask_5g;
+	} else {
+		return -EINVAL;
+	}
+
+	/* tx/rx chainmask reported from fw depends on the actual hw chains used,
+	 * For example, for 4x4 capable macphys, first 4 chains can be used for first
+	 * mac and the remaing 4 chains can be used for the second mac or vice-versa.
+	 * In this case, tx/rx chainmask 0xf will be advertised for first mac and 0xf0
+	 * will be advertised for second mac or vice-versa. Compute the shift value for
+	 * for tx/rx chainmask which will be used to advertise supported ht/vht rates to
+	 * mac80211.
+	 */
+	pdev_cap->tx_chain_mask_shift =
+			find_first_bit((unsigned long *)&pdev_cap->tx_chain_mask, 32);
+	pdev_cap->rx_chain_mask_shift =
+			find_first_bit((unsigned long *)&pdev_cap->rx_chain_mask, 32);
+
+	cap_band = &pdev_cap->band[NL80211_BAND_2GHZ];
+	cap_band->max_bw_supported = mac_phy_caps->max_bw_supported_2g;
+	cap_band->ht_cap_info = mac_phy_caps->ht_cap_info_2g;
+	cap_band->he_cap_info[0] = mac_phy_caps->he_cap_info_2g;
+	cap_band->he_cap_info[1] = mac_phy_caps->he_cap_info_2g_ext;
+	cap_band->he_mcs = mac_phy_caps->he_supp_mcs_2g;
+	memcpy(cap_band->he_cap_phy_info, &mac_phy_caps->he_cap_phy_info_2g,
+	       sizeof(u32) * PSOC_HOST_MAX_PHY_SIZE);
+	memcpy(&cap_band->he_ppet, &mac_phy_caps->he_ppet2g,
+	       sizeof(struct ath11k_ppe_threshold));
+
+	cap_band = &pdev_cap->band[NL80211_BAND_5GHZ];
+	cap_band->max_bw_supported = mac_phy_caps->max_bw_supported_5g;
+	cap_band->ht_cap_info = mac_phy_caps->ht_cap_info_5g;
+	cap_band->he_cap_info[0] = mac_phy_caps->he_cap_info_5g;
+	cap_band->he_cap_info[1] = mac_phy_caps->he_cap_info_5g_ext;
+	cap_band->he_mcs = mac_phy_caps->he_supp_mcs_5g;
+	memcpy(cap_band->he_cap_phy_info, &mac_phy_caps->he_cap_phy_info_5g,
+	       sizeof(u32) * PSOC_HOST_MAX_PHY_SIZE);
+	memcpy(&cap_band->he_ppet, &mac_phy_caps->he_ppet5g,
+	       sizeof(struct ath11k_ppe_threshold));
+
+	return 0;
+}
+
+static int
+ath11k_pull_reg_cap_svc_rdy_ext(struct ath11k_pdev_wmi *wmi_handle,
+				struct wmi_soc_hal_reg_capabilities *reg_caps,
+				struct wmi_hal_reg_capabilities_ext *wmi_ext_reg_cap,
+				u8 phy_idx,
+				struct ath11k_hal_reg_capabilities_ext *param)
+{
+	struct wmi_hal_reg_capabilities_ext *ext_reg_cap;
+
+	if (!reg_caps || !wmi_ext_reg_cap)
+		return -EINVAL;
+
+	if (phy_idx >= reg_caps->num_phy)
+		return -EINVAL;
+
+	ext_reg_cap = &wmi_ext_reg_cap[phy_idx];
+
+	param->phy_id = ext_reg_cap->phy_id;
+	param->eeprom_reg_domain = ext_reg_cap->eeprom_reg_domain;
+	param->eeprom_reg_domain_ext =
+			      ext_reg_cap->eeprom_reg_domain_ext;
+	param->regcap1 = ext_reg_cap->regcap1;
+	param->regcap2 = ext_reg_cap->regcap2;
+	/* check if param->wireless_mode is needed */
+	param->low_2ghz_chan = ext_reg_cap->low_2ghz_chan;
+	param->high_2ghz_chan = ext_reg_cap->high_2ghz_chan;
+	param->low_5ghz_chan = ext_reg_cap->low_5ghz_chan;
+	param->high_5ghz_chan = ext_reg_cap->high_5ghz_chan;
+
+	return 0;
+}
+
+static int ath11k_pull_service_ready_tlv(struct ath11k_base *ab,
+					 const void *evt_buf,
+					 struct ath11k_targ_cap *cap)
+{
+	const struct wmi_service_ready_event *ev = evt_buf;
+
+	if (!ev) {
+		ath11k_err(ab, "%s: failed by NULL param\n",
+			   __func__);
+		return -EINVAL;
+	}
+
+	cap->phy_capability = ev->phy_capability;
+	cap->max_frag_entry = ev->max_frag_entry;
+	cap->num_rf_chains = ev->num_rf_chains;
+	cap->ht_cap_info = ev->ht_cap_info;
+	cap->vht_cap_info = ev->vht_cap_info;
+	cap->vht_supp_mcs = ev->vht_supp_mcs;
+	cap->hw_min_tx_power = ev->hw_min_tx_power;
+	cap->hw_max_tx_power = ev->hw_max_tx_power;
+	cap->sys_cap_info = ev->sys_cap_info;
+	cap->min_pkt_size_enable = ev->min_pkt_size_enable;
+	cap->max_bcn_ie_size = ev->max_bcn_ie_size;
+	cap->max_num_scan_channels = ev->max_num_scan_channels;
+	cap->max_supported_macs = ev->max_supported_macs;
+	cap->wmi_fw_sub_feat_caps = ev->wmi_fw_sub_feat_caps;
+	cap->txrx_chainmask = ev->txrx_chainmask;
+	cap->default_dbs_hw_mode_index = ev->default_dbs_hw_mode_index;
+	cap->num_msdu_desc = ev->num_msdu_desc;
+
+	return 0;
+}
+
+/* Save the wmi_service_bitmap into a linear bitmap. The wmi_services in
+ * wmi_service ready event are advertised in b0-b3 (LSB 4-bits) of each
+ * 4-byte word.
+ */
+static void ath11k_wmi_service_bitmap_copy(struct ath11k_pdev_wmi *wmi,
+					   const u32 *wmi_svc_bm)
+{
+	int i, j;
+
+	for (i = 0, j = 0; i < WMI_SERVICE_BM_SIZE && j < WMI_MAX_SERVICE; i++) {
+		do {
+			if (wmi_svc_bm[i] & BIT(j % WMI_SERVICE_BITS_IN_SIZE32))
+				set_bit(j, wmi->wmi_sc->svc_map);
+		} while (++j % WMI_SERVICE_BITS_IN_SIZE32);
+	}
+}
+
+static int ath11k_wmi_tlv_svc_rdy_parse(struct ath11k_base *ab, u16 tag, u16 len,
+					const void *ptr, void *data)
+{
+	struct wmi_tlv_svc_ready_parse *svc_ready = data;
+	struct ath11k_pdev_wmi *wmi_handle = &ab->wmi_sc.wmi[0];
+	u16 expect_len;
+
+	switch (tag) {
+	case WMI_TAG_SERVICE_READY_EVENT:
+		if (ath11k_pull_service_ready_tlv(ab, ptr, &ab->target_caps))
+			return -EINVAL;
+		break;
+
+	case WMI_TAG_ARRAY_UINT32:
+		if (!svc_ready->wmi_svc_bitmap_done) {
+			expect_len = WMI_SERVICE_BM_SIZE * sizeof(u32);
+			if (len < expect_len) {
+				ath11k_warn(ab, "invalid len %d for the tag 0x%x\n",
+					    len, tag);
+				return -EINVAL;
+			}
+
+			ath11k_wmi_service_bitmap_copy(wmi_handle, ptr);
+
+			svc_ready->wmi_svc_bitmap_done = true;
+		}
+		break;
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+static int ath11k_service_ready_event(struct ath11k_base *ab, struct sk_buff *skb)
+{
+	struct wmi_tlv_svc_ready_parse svc_ready = { };
+	int ret;
+
+	ret = ath11k_wmi_tlv_iter(ab, skb->data, skb->len,
+				  ath11k_wmi_tlv_svc_rdy_parse,
+				  &svc_ready);
+	if (ret) {
+		ath11k_warn(ab, "failed to parse tlv %d\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+struct sk_buff *ath11k_wmi_alloc_skb(struct ath11k_wmi_base *wmi_sc, u32 len)
+{
+	struct sk_buff *skb;
+	struct ath11k_base *ab = wmi_sc->ab;
+	u32 round_len = roundup(len, 4);
+
+	skb = ath11k_htc_alloc_skb(ab, WMI_SKB_HEADROOM + round_len);
+	if (!skb)
+		return NULL;
+
+	skb_reserve(skb, WMI_SKB_HEADROOM);
+	if (!IS_ALIGNED((unsigned long)skb->data, 4))
+		ath11k_warn(ab, "unaligned WMI skb data\n");
+
+	skb_put(skb, round_len);
+	memset(skb->data, 0, round_len);
+
+	return skb;
+}
+
+int ath11k_wmi_mgmt_send(struct ath11k *ar, u32 vdev_id, u32 buf_id,
+			 struct sk_buff *frame)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_mgmt_send_cmd *cmd;
+	struct wmi_tlv *frame_tlv;
+	struct sk_buff *skb;
+	u32 buf_len;
+	int ret, len;
+
+	buf_len = frame->len < WMI_MGMT_SEND_DOWNLD_LEN ?
+		  frame->len : WMI_MGMT_SEND_DOWNLD_LEN;
+
+	len = sizeof(*cmd) + sizeof(*frame_tlv) + roundup(buf_len, 4);
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_mgmt_send_cmd *)skb->data;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_MGMT_TX_SEND_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+	cmd->vdev_id = vdev_id;
+	cmd->desc_id = buf_id;
+	cmd->chanfreq = 0;
+	cmd->paddr_lo = lower_32_bits(ATH11K_SKB_CB(frame)->paddr);
+	cmd->paddr_hi = upper_32_bits(ATH11K_SKB_CB(frame)->paddr);
+	cmd->frame_len = frame->len;
+	cmd->buf_len = buf_len;
+	cmd->tx_params_valid = 0;
+
+	frame_tlv = (struct wmi_tlv *)(skb->data + sizeof(*cmd));
+	frame_tlv->header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_ARRAY_BYTE) |
+			    FIELD_PREP(WMI_TLV_LEN, buf_len);
+
+	memcpy(frame_tlv->value, frame->data, buf_len);
+
+	ath11k_ce_byte_swap(frame_tlv->value, buf_len);
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, WMI_MGMT_TX_SEND_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab,
+			    "failed to submit WMI_MGMT_TX_SEND_CMDID cmd\n");
+		dev_kfree_skb(skb);
+	}
+
+	return ret;
+}
+
+int ath11k_wmi_vdev_create(struct ath11k *ar, u8 *macaddr,
+			   struct vdev_create_params *param)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_vdev_create_cmd *cmd;
+	struct sk_buff *skb;
+	struct wmi_vdev_txrx_streams *txrx_streams;
+	struct wmi_tlv *tlv;
+	int ret, len;
+	void *ptr;
+
+	/* It can be optimized my sending tx/rx chain configuration
+	 * only for supported bands instead of always sending it for
+	 * both the bands.
+	 */
+	len = sizeof(*cmd) + TLV_HDR_SIZE +
+		(WMI_NUM_SUPPORTED_BAND_MAX * sizeof(*txrx_streams));
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_vdev_create_cmd *)skb->data;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_VDEV_CREATE_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+
+	cmd->vdev_id = param->if_id;
+	cmd->vdev_type = param->type;
+	cmd->vdev_subtype = param->subtype;
+	cmd->num_cfg_txrx_streams = WMI_NUM_SUPPORTED_BAND_MAX;
+	cmd->pdev_id = param->pdev_id;
+	ether_addr_copy(cmd->vdev_macaddr.addr, macaddr);
+
+	ptr = skb->data + sizeof(*cmd);
+	len = WMI_NUM_SUPPORTED_BAND_MAX * sizeof(*txrx_streams);
+
+	tlv = ptr;
+	tlv->header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_ARRAY_STRUCT) |
+		      FIELD_PREP(WMI_TLV_LEN, len);
+
+	ptr += TLV_HDR_SIZE;
+	txrx_streams = ptr;
+	len = sizeof(*txrx_streams);
+	txrx_streams->tlv_header =
+		FIELD_PREP(WMI_TLV_TAG, WMI_TAG_VDEV_TXRX_STREAMS) |
+		FIELD_PREP(WMI_TLV_LEN, len - TLV_HDR_SIZE);
+	txrx_streams->band = WMI_TPC_CHAINMASK_CONFIG_BAND_2G;
+	txrx_streams->supported_tx_streams =
+				 param->chains[NL80211_BAND_2GHZ].tx;
+	txrx_streams->supported_rx_streams =
+				 param->chains[NL80211_BAND_2GHZ].rx;
+
+	txrx_streams++;
+	txrx_streams->tlv_header =
+		FIELD_PREP(WMI_TLV_TAG, WMI_TAG_VDEV_TXRX_STREAMS) |
+		FIELD_PREP(WMI_TLV_LEN, len - TLV_HDR_SIZE);
+	txrx_streams->band = WMI_TPC_CHAINMASK_CONFIG_BAND_5G;
+	txrx_streams->supported_tx_streams =
+				 param->chains[NL80211_BAND_5GHZ].tx;
+	txrx_streams->supported_rx_streams =
+				 param->chains[NL80211_BAND_5GHZ].rx;
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, WMI_VDEV_CREATE_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab,
+			    "failed to submit WMI_VDEV_CREATE_CMDID\n");
+		dev_kfree_skb(skb);
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+		   "WMI vdev create: id %d type %d subtype %d macaddr %pM pdevid %d\n",
+		   param->if_id, param->type, param->subtype,
+		   macaddr, param->pdev_id);
+
+	return ret;
+}
+
+int ath11k_wmi_vdev_delete(struct ath11k *ar, u8 vdev_id)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_vdev_delete_cmd *cmd;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_vdev_delete_cmd *)skb->data;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_VDEV_DELETE_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+	cmd->vdev_id = vdev_id;
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, WMI_VDEV_DELETE_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to submit WMI_VDEV_DELETE_CMDID\n");
+		dev_kfree_skb(skb);
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI, "WMI vdev delete id %d\n", vdev_id);
+
+	return ret;
+}
+
+int ath11k_wmi_vdev_stop(struct ath11k *ar, u8 vdev_id)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_vdev_stop_cmd *cmd;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_vdev_stop_cmd *)skb->data;
+
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_VDEV_STOP_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+	cmd->vdev_id = vdev_id;
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, WMI_VDEV_STOP_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to submit WMI_VDEV_STOP cmd\n");
+		dev_kfree_skb(skb);
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI, "WMI vdev stop id 0x%x\n", vdev_id);
+
+	return ret;
+}
+
+int ath11k_wmi_vdev_down(struct ath11k *ar, u8 vdev_id)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_vdev_down_cmd *cmd;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_vdev_down_cmd *)skb->data;
+
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_VDEV_DOWN_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+	cmd->vdev_id = vdev_id;
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, WMI_VDEV_DOWN_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to submit WMI_VDEV_DOWN cmd\n");
+		dev_kfree_skb(skb);
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI, "WMI vdev down id 0x%x\n", vdev_id);
+
+	return ret;
+}
+
+static void ath11k_wmi_put_wmi_channel(struct wmi_channel *chan,
+				       struct wmi_vdev_start_req_arg *arg)
+{
+	memset(chan, 0, sizeof(*chan));
+
+	chan->mhz = arg->channel.freq;
+	chan->band_center_freq1 = arg->channel.band_center_freq1;
+	if (arg->channel.mode == MODE_11AC_VHT80_80)
+		chan->band_center_freq2 = arg->channel.band_center_freq2;
+	else
+		chan->band_center_freq2 = 0;
+
+	chan->info |= FIELD_PREP(WMI_CHAN_INFO_MODE, arg->channel.mode);
+	if (arg->channel.passive)
+		chan->info |= WMI_CHAN_INFO_PASSIVE;
+	if (arg->channel.allow_ibss)
+		chan->info |= WMI_CHAN_INFO_ADHOC_ALLOWED;
+	if (arg->channel.allow_ht)
+		chan->info |= WMI_CHAN_INFO_ALLOW_HT;
+	if (arg->channel.allow_vht)
+		chan->info |= WMI_CHAN_INFO_ALLOW_VHT;
+	if (arg->channel.ht40plus)
+		chan->info |= WMI_CHAN_INFO_HT40_PLUS;
+	if (arg->channel.chan_radar)
+		chan->info |= WMI_CHAN_INFO_DFS;
+	if (arg->channel.freq2_radar)
+		chan->info |= WMI_CHAN_INFO_DFS_FREQ2;
+
+	chan->reg_info_1 = FIELD_PREP(WMI_CHAN_REG_INFO1_MAX_PWR,
+				      arg->channel.max_power) |
+		FIELD_PREP(WMI_CHAN_REG_INFO1_MAX_REG_PWR,
+			   arg->channel.max_reg_power);
+
+	chan->reg_info_2 = FIELD_PREP(WMI_CHAN_REG_INFO2_ANT_MAX,
+				      arg->channel.max_antenna_gain) |
+		FIELD_PREP(WMI_CHAN_REG_INFO2_MAX_TX_PWR,
+			   arg->channel.max_power);
+}
+
+int ath11k_wmi_vdev_start(struct ath11k *ar, struct wmi_vdev_start_req_arg *arg,
+			  bool restart)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_vdev_start_request_cmd *cmd;
+	struct sk_buff *skb;
+	struct wmi_channel *chan;
+	struct wmi_tlv *tlv;
+	void *ptr;
+	int ret, len;
+
+	if (WARN_ON(arg->ssid_len > sizeof(cmd->ssid.ssid)))
+		return -EINVAL;
+
+	len = sizeof(*cmd) + sizeof(*chan) + TLV_HDR_SIZE;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_vdev_start_request_cmd *)skb->data;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG,
+				     WMI_TAG_VDEV_START_REQUEST_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+	cmd->vdev_id = arg->vdev_id;
+	cmd->beacon_interval = arg->bcn_intval;
+	cmd->bcn_tx_rate = arg->bcn_tx_rate;
+	cmd->dtim_period = arg->dtim_period;
+	cmd->num_noa_descriptors = arg->num_noa_descriptors;
+	cmd->preferred_rx_streams = arg->pref_rx_streams;
+	cmd->preferred_tx_streams = arg->pref_tx_streams;
+	cmd->cac_duration_ms = arg->cac_duration_ms;
+	cmd->regdomain = arg->regdomain;
+	cmd->he_ops = arg->he_ops;
+
+	if (!restart) {
+		if (arg->ssid) {
+			cmd->ssid.ssid_len = arg->ssid_len;
+			memcpy(cmd->ssid.ssid, arg->ssid, arg->ssid_len);
+		}
+		if (arg->hidden_ssid)
+			cmd->flags |= WMI_VDEV_START_HIDDEN_SSID;
+		if (arg->pmf_enabled)
+			cmd->flags |= WMI_VDEV_START_PMF_ENABLED;
+	}
+
+	cmd->flags |= WMI_VDEV_START_LDPC_RX_ENABLED;
+
+	ptr = skb->data + sizeof(*cmd);
+	chan = ptr;
+
+	ath11k_wmi_put_wmi_channel(chan, arg);
+
+	chan->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_CHANNEL) |
+			   FIELD_PREP(WMI_TLV_LEN,
+				      sizeof(*chan) - TLV_HDR_SIZE);
+	ptr += sizeof(*chan);
+
+	tlv = ptr;
+	tlv->header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_ARRAY_STRUCT) |
+		      FIELD_PREP(WMI_TLV_LEN, 0);
+
+	/* Note: This is a nested TLV containing:
+	 * [wmi_tlv][wmi_p2p_noa_descriptor][wmi_tlv]..
+	 */
+
+	ptr += sizeof(*tlv);
+
+	if (restart)
+		ret = ath11k_wmi_cmd_send(wmi, skb,
+					  WMI_VDEV_RESTART_REQUEST_CMDID);
+	else
+		ret = ath11k_wmi_cmd_send(wmi, skb,
+					  WMI_VDEV_START_REQUEST_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to submit vdev_%s cmd\n",
+			    restart ? "restart" : "start");
+		dev_kfree_skb(skb);
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI, "vdev %s id 0x%x freq 0x%x mode 0x%x\n",
+		   restart ? "restart" : "start", arg->vdev_id,
+		   arg->channel.freq, arg->channel.mode);
+
+	return ret;
+}
+
+int ath11k_wmi_vdev_up(struct ath11k *ar, u32 vdev_id, u32 aid, const u8 *bssid)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_vdev_up_cmd *cmd;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_vdev_up_cmd *)skb->data;
+
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_VDEV_UP_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+	cmd->vdev_id = vdev_id;
+	cmd->vdev_assoc_id = aid;
+
+	ether_addr_copy(cmd->vdev_bssid.addr, bssid);
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, WMI_VDEV_UP_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to submit WMI_VDEV_UP cmd\n");
+		dev_kfree_skb(skb);
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+		   "WMI mgmt vdev up id 0x%x assoc id %d bssid %pM\n",
+		   vdev_id, aid, bssid);
+
+	return ret;
+}
+
+int ath11k_wmi_send_peer_create_cmd(struct ath11k *ar,
+				    struct peer_create_params *param)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_peer_create_cmd *cmd;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_peer_create_cmd *)skb->data;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_PEER_CREATE_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+
+	ether_addr_copy(cmd->peer_macaddr.addr, param->peer_addr);
+	cmd->peer_type = param->peer_type;
+	cmd->vdev_id = param->vdev_id;
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, WMI_PEER_CREATE_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to submit WMI_PEER_CREATE cmd\n");
+		dev_kfree_skb(skb);
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+		   "WMI peer create vdev_id %d peer_addr %pM\n",
+		   param->vdev_id, param->peer_addr);
+
+	return ret;
+}
+
+int ath11k_wmi_send_peer_delete_cmd(struct ath11k *ar,
+				    const u8 *peer_addr, u8 vdev_id)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_peer_delete_cmd *cmd;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_peer_delete_cmd *)skb->data;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_PEER_DELETE_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+
+	ether_addr_copy(cmd->peer_macaddr.addr, peer_addr);
+	cmd->vdev_id = vdev_id;
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+		   "WMI peer delete vdev_id %d peer_addr %pM\n",
+		   vdev_id,  peer_addr);
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, WMI_PEER_DELETE_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to send WMI_PEER_DELETE cmd\n");
+		dev_kfree_skb(skb);
+	}
+
+	return ret;
+}
+
+int ath11k_wmi_send_pdev_set_regdomain(struct ath11k *ar,
+				       struct pdev_set_regdomain_params *param)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_pdev_set_regdomain_cmd *cmd;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_pdev_set_regdomain_cmd *)skb->data;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG,
+				     WMI_TAG_PDEV_SET_REGDOMAIN_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+
+	cmd->reg_domain = param->current_rd_in_use;
+	cmd->reg_domain_2g = param->current_rd_2g;
+	cmd->reg_domain_5g = param->current_rd_5g;
+	cmd->conformance_test_limit_2g = param->ctl_2g;
+	cmd->conformance_test_limit_5g = param->ctl_5g;
+	cmd->dfs_domain = param->dfs_domain;
+	cmd->pdev_id = param->pdev_id;
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+		   "WMI pdev regd rd %d rd2g %d rd5g %d domain %d pdev id %d\n",
+		   param->current_rd_in_use, param->current_rd_2g,
+		   param->current_rd_5g, param->dfs_domain, param->pdev_id);
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, WMI_PDEV_SET_REGDOMAIN_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab,
+			    "failed to send WMI_PDEV_SET_REGDOMAIN cmd\n");
+		dev_kfree_skb(skb);
+	}
+
+	return ret;
+}
+
+int ath11k_wmi_set_peer_param(struct ath11k *ar, const u8 *peer_addr,
+			      u32 vdev_id, u32 param_id, u32 param_val)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_peer_set_param_cmd *cmd;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_peer_set_param_cmd *)skb->data;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_PEER_SET_PARAM_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+	ether_addr_copy(cmd->peer_macaddr.addr, peer_addr);
+	cmd->vdev_id = vdev_id;
+	cmd->param_id = param_id;
+	cmd->param_value = param_val;
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, WMI_PEER_SET_PARAM_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to send WMI_PEER_SET_PARAM cmd\n");
+		dev_kfree_skb(skb);
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+		   "WMI vdev %d peer 0x%pM set param %d value %d\n",
+		   vdev_id, peer_addr, param_id, param_val);
+
+	return ret;
+}
+
+int ath11k_wmi_send_peer_flush_tids_cmd(struct ath11k *ar,
+					u8 peer_addr[ETH_ALEN],
+					struct peer_flush_params *param)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_peer_flush_tids_cmd *cmd;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_peer_flush_tids_cmd *)skb->data;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_PEER_FLUSH_TIDS_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+
+	ether_addr_copy(cmd->peer_macaddr.addr, peer_addr);
+	cmd->peer_tid_bitmap = param->peer_tid_bitmap;
+	cmd->vdev_id = param->vdev_id;
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, WMI_PEER_FLUSH_TIDS_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab,
+			    "failed to send WMI_PEER_FLUSH_TIDS cmd\n");
+		dev_kfree_skb(skb);
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+		   "WMI peer flush vdev_id %d peer_addr %pM tids %08x\n",
+		   param->vdev_id, peer_addr, param->peer_tid_bitmap);
+
+	return ret;
+}
+
+int ath11k_wmi_peer_rx_reorder_queue_setup(struct ath11k *ar,
+					   int vdev_id, const u8 *addr,
+					   dma_addr_t paddr, u8 tid,
+					   u8 ba_window_size_valid,
+					   u32 ba_window_size)
+{
+	struct wmi_peer_reorder_queue_setup_cmd *cmd;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = ath11k_wmi_alloc_skb(ar->wmi->wmi_sc, sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_peer_reorder_queue_setup_cmd *)skb->data;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG,
+				     WMI_TAG_REORDER_QUEUE_SETUP_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+
+	ether_addr_copy(cmd->peer_macaddr.addr, addr);
+	cmd->vdev_id = vdev_id;
+	cmd->tid = tid;
+	cmd->queue_ptr_lo = lower_32_bits(paddr);
+	cmd->queue_ptr_hi = upper_32_bits(paddr);
+	cmd->queue_no = tid;
+	cmd->ba_window_size_valid = ba_window_size_valid;
+	cmd->ba_window_size = ba_window_size;
+
+	ret = ath11k_wmi_cmd_send(ar->wmi, skb,
+				  WMI_PEER_REORDER_QUEUE_SETUP_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab,
+			    "failed to send WMI_PEER_REORDER_QUEUE_SETUP\n");
+		dev_kfree_skb(skb);
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+		   "wmi rx reorder queue setup addr %pM vdev_id %d tid %d\n",
+		   addr, vdev_id, tid);
+
+	return ret;
+}
+
+int
+ath11k_wmi_rx_reord_queue_remove(struct ath11k *ar,
+				 struct rx_reorder_queue_remove_params *param)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_peer_reorder_queue_remove_cmd *cmd;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_peer_reorder_queue_remove_cmd *)skb->data;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG,
+				     WMI_TAG_REORDER_QUEUE_REMOVE_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+
+	ether_addr_copy(cmd->peer_macaddr.addr, param->peer_macaddr);
+	cmd->vdev_id = param->vdev_id;
+	cmd->tid_mask = param->peer_tid_bitmap;
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+		   "%s: peer_macaddr %pM vdev_id %d, tid_map %d", __func__,
+		   param->peer_macaddr, param->vdev_id, param->peer_tid_bitmap);
+
+	ret = ath11k_wmi_cmd_send(wmi, skb,
+				  WMI_PEER_REORDER_QUEUE_REMOVE_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab,
+			    "failed to send WMI_PEER_REORDER_QUEUE_REMOVE_CMDID");
+		dev_kfree_skb(skb);
+	}
+
+	return ret;
+}
+
+int ath11k_wmi_pdev_set_param(struct ath11k *ar, u32 param_id,
+			      u32 param_value, u8 pdev_id)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_pdev_set_param_cmd *cmd;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_pdev_set_param_cmd *)skb->data;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_PDEV_SET_PARAM_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+	cmd->pdev_id = pdev_id;
+	cmd->param_id = param_id;
+	cmd->param_value = param_value;
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, WMI_PDEV_SET_PARAM_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to send WMI_PDEV_SET_PARAM cmd\n");
+		dev_kfree_skb(skb);
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+		   "WMI pdev set param %d pdev id %d value %d\n",
+		   param_id, pdev_id, param_value);
+
+	return ret;
+}
+
+int ath11k_wmi_pdev_suspend(struct ath11k *ar, u32 suspend_opt,
+			    u32 pdev_id)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_pdev_suspend_cmd *cmd;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_pdev_suspend_cmd *)skb->data;
+
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_PDEV_SUSPEND_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+
+	cmd->suspend_opt = suspend_opt;
+	cmd->pdev_id = pdev_id;
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, WMI_PDEV_SUSPEND_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to send WMI_PDEV_SUSPEND cmd\n");
+		dev_kfree_skb(skb);
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+		   "WMI pdev suspend pdev_id %d\n", pdev_id);
+
+	return ret;
+}
+
+int ath11k_wmi_pdev_resume(struct ath11k *ar, u32 pdev_id)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_pdev_resume_cmd *cmd;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_pdev_resume_cmd *)skb->data;
+
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_PDEV_RESUME_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+	cmd->pdev_id = pdev_id;
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+		   "WMI pdev resume pdev id %d\n", pdev_id);
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, WMI_PDEV_RESUME_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to send WMI_PDEV_RESUME cmd\n");
+		dev_kfree_skb(skb);
+	}
+
+	return ret;
+}
+
+/* TODO FW Support for the cmd is not available yet.
+ * Can be tested once the command and corresponding
+ * event is implemented in FW
+ */
+int ath11k_wmi_pdev_bss_chan_info_request(struct ath11k *ar,
+					  enum wmi_bss_chan_info_req_type type)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_pdev_bss_chan_info_req_cmd *cmd;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_pdev_bss_chan_info_req_cmd *)skb->data;
+
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG,
+				     WMI_TAG_PDEV_BSS_CHAN_INFO_REQUEST) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+	cmd->req_type = type;
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+		   "WMI bss chan info req type %d\n", type);
+
+	ret = ath11k_wmi_cmd_send(wmi, skb,
+				  WMI_PDEV_BSS_CHAN_INFO_REQUEST_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab,
+			    "failed to send WMI_PDEV_BSS_CHAN_INFO_REQUEST cmd\n");
+		dev_kfree_skb(skb);
+	}
+
+	return ret;
+}
+
+int ath11k_wmi_send_set_ap_ps_param_cmd(struct ath11k *ar, u8 *peer_addr,
+					struct ap_ps_params *param)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_ap_ps_peer_cmd *cmd;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_ap_ps_peer_cmd *)skb->data;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_AP_PS_PEER_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+
+	cmd->vdev_id = param->vdev_id;
+	ether_addr_copy(cmd->peer_macaddr.addr, peer_addr);
+	cmd->param = param->param;
+	cmd->value = param->value;
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, WMI_AP_PS_PEER_PARAM_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab,
+			    "failed to send WMI_AP_PS_PEER_PARAM_CMDID\n");
+		dev_kfree_skb(skb);
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+		   "WMI set ap ps vdev id %d peer %pM param %d value %d\n",
+		   param->vdev_id, peer_addr, param->param, param->value);
+
+	return ret;
+}
+
+int ath11k_wmi_set_sta_ps_param(struct ath11k *ar, u32 vdev_id,
+				u32 param, u32 param_value)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_sta_powersave_param_cmd *cmd;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_sta_powersave_param_cmd *)skb->data;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG,
+				     WMI_TAG_STA_POWERSAVE_PARAM_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+
+	cmd->vdev_id = vdev_id;
+	cmd->param = param;
+	cmd->value = param_value;
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+		   "WMI set sta ps vdev_id %d param %d value %d\n",
+		   vdev_id, param, param_value);
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, WMI_STA_POWERSAVE_PARAM_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to send WMI_STA_POWERSAVE_PARAM_CMDID");
+		dev_kfree_skb(skb);
+	}
+
+	return ret;
+}
+
+int ath11k_wmi_force_fw_hang_cmd(struct ath11k *ar, u32 type, u32 delay_time_ms)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_force_fw_hang_cmd *cmd;
+	struct sk_buff *skb;
+	int ret, len;
+
+	len = sizeof(*cmd);
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_force_fw_hang_cmd *)skb->data;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_FORCE_FW_HANG_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, len - TLV_HDR_SIZE);
+
+	cmd->type = type;
+	cmd->delay_time_ms = delay_time_ms;
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, WMI_FORCE_FW_HANG_CMDID);
+
+	if (ret) {
+		ath11k_warn(ar->ab, "Failed to send WMI_FORCE_FW_HANG_CMDID");
+		dev_kfree_skb(skb);
+	}
+	return ret;
+}
+
+int ath11k_wmi_vdev_set_param_cmd(struct ath11k *ar, u32 vdev_id,
+				  u32 param_id, u32 param_value)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_vdev_set_param_cmd *cmd;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_vdev_set_param_cmd *)skb->data;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_VDEV_SET_PARAM_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+
+	cmd->vdev_id = vdev_id;
+	cmd->param_id = param_id;
+	cmd->param_value = param_value;
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, WMI_VDEV_SET_PARAM_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab,
+			    "failed to send WMI_VDEV_SET_PARAM_CMDID\n");
+		dev_kfree_skb(skb);
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+		   "WMI vdev id 0x%x set param %d value %d\n",
+		   vdev_id, param_id, param_value);
+
+	return ret;
+}
+
+int ath11k_wmi_send_stats_request_cmd(struct ath11k *ar,
+				      struct stats_request_params *param)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_request_stats_cmd *cmd;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_request_stats_cmd *)skb->data;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_REQUEST_STATS_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+
+	cmd->stats_id = param->stats_id;
+	cmd->vdev_id = param->vdev_id;
+	cmd->pdev_id = param->pdev_id;
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, WMI_REQUEST_STATS_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to send WMI_REQUEST_STATS cmd\n");
+		dev_kfree_skb(skb);
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+		   "WMI request stats 0x%x vdev id %d pdev id %d\n",
+		   param->stats_id, param->vdev_id, param->pdev_id);
+
+	return ret;
+}
+
+int ath11k_wmi_send_bcn_offload_control_cmd(struct ath11k *ar,
+					    u32 vdev_id, u32 bcn_ctrl_op)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_bcn_offload_ctrl_cmd *cmd;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_bcn_offload_ctrl_cmd *)skb->data;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG,
+				     WMI_TAG_BCN_OFFLOAD_CTRL_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+
+	cmd->vdev_id = vdev_id;
+	cmd->bcn_ctrl_op = bcn_ctrl_op;
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+		   "WMI bcn ctrl offload vdev id %d ctrl_op %d\n",
+		   vdev_id, bcn_ctrl_op);
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, WMI_BCN_OFFLOAD_CTRL_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab,
+			    "failed to send WMI_BCN_OFFLOAD_CTRL_CMDID\n");
+		dev_kfree_skb(skb);
+	}
+
+	return ret;
+}
+
+int ath11k_wmi_bcn_tmpl(struct ath11k *ar, u32 vdev_id,
+			struct ieee80211_mutable_offsets *offs,
+			struct sk_buff *bcn)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_bcn_tmpl_cmd *cmd;
+	struct wmi_bcn_prb_info *bcn_prb_info;
+	struct wmi_tlv *tlv;
+	struct sk_buff *skb;
+	void *ptr;
+	int ret, len;
+	size_t aligned_len = roundup(bcn->len, 4);
+
+	len = sizeof(*cmd) + sizeof(*bcn_prb_info) + TLV_HDR_SIZE + aligned_len;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_bcn_tmpl_cmd *)skb->data;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_BCN_TMPL_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+	cmd->vdev_id = vdev_id;
+	cmd->tim_ie_offset = offs->tim_offset;
+	cmd->csa_switch_count_offset = offs->csa_counter_offs[0];
+	cmd->ext_csa_switch_count_offset = offs->csa_counter_offs[1];
+	cmd->buf_len = bcn->len;
+
+	ptr = skb->data + sizeof(*cmd);
+
+	bcn_prb_info = ptr;
+	len = sizeof(*bcn_prb_info);
+	bcn_prb_info->tlv_header = FIELD_PREP(WMI_TLV_TAG,
+					      WMI_TAG_BCN_PRB_INFO) |
+				   FIELD_PREP(WMI_TLV_LEN, len - TLV_HDR_SIZE);
+	bcn_prb_info->caps = 0;
+	bcn_prb_info->erp = 0;
+
+	ptr += sizeof(*bcn_prb_info);
+
+	tlv = ptr;
+	tlv->header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_ARRAY_BYTE) |
+		      FIELD_PREP(WMI_TLV_LEN, aligned_len);
+	memcpy(tlv->value, bcn->data, bcn->len);
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, WMI_BCN_TMPL_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to send WMI_BCN_TMPL_CMDID\n");
+		dev_kfree_skb(skb);
+	}
+
+	return ret;
+}
+
+int ath11k_wmi_vdev_install_key(struct ath11k *ar,
+				struct wmi_vdev_install_key_arg *arg)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_vdev_install_key_cmd *cmd;
+	struct wmi_tlv *tlv;
+	struct sk_buff *skb;
+	int ret, len;
+	int key_len_aligned = roundup(arg->key_len, sizeof(uint32_t));
+
+	len = sizeof(*cmd) + TLV_HDR_SIZE + key_len_aligned;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_vdev_install_key_cmd *)skb->data;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_VDEV_INSTALL_KEY_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+	cmd->vdev_id = arg->vdev_id;
+	ether_addr_copy(cmd->peer_macaddr.addr, arg->macaddr);
+	cmd->key_idx = arg->key_idx;
+	cmd->key_flags = arg->key_flags;
+	cmd->key_cipher = arg->key_cipher;
+	cmd->key_len = arg->key_len;
+	cmd->key_txmic_len = arg->key_txmic_len;
+	cmd->key_rxmic_len = arg->key_rxmic_len;
+
+	if (arg->key_rsc_counter)
+		memcpy(&cmd->key_rsc_counter, &arg->key_rsc_counter,
+		       sizeof(struct wmi_key_seq_counter));
+
+	tlv = (struct wmi_tlv *)(skb->data + sizeof(*cmd));
+	tlv->header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_ARRAY_BYTE) |
+		      FIELD_PREP(WMI_TLV_LEN, key_len_aligned);
+	memcpy(tlv->value, (u8 *)arg->key_data, key_len_aligned);
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, WMI_VDEV_INSTALL_KEY_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab,
+			    "failed to send WMI_VDEV_INSTALL_KEY cmd\n");
+		dev_kfree_skb(skb);
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+		   "WMI vdev install key idx %d cipher %d len %d\n",
+		   arg->key_idx, arg->key_cipher, arg->key_len);
+
+	return ret;
+}
+
+static inline void
+ath11k_wmi_copy_peer_flags(struct wmi_peer_assoc_complete_cmd *cmd,
+			   struct peer_assoc_params *param)
+{
+	cmd->peer_flags = 0;
+
+	if (param->is_wme_set) {
+		if (param->qos_flag)
+			cmd->peer_flags |= WMI_PEER_QOS;
+		if (param->apsd_flag)
+			cmd->peer_flags |= WMI_PEER_APSD;
+		if (param->ht_flag)
+			cmd->peer_flags |= WMI_PEER_HT;
+		if (param->bw_40)
+			cmd->peer_flags |= WMI_PEER_40MHZ;
+		if (param->bw_80)
+			cmd->peer_flags |= WMI_PEER_80MHZ;
+		if (param->bw_160)
+			cmd->peer_flags |= WMI_PEER_160MHZ;
+
+		/* Typically if STBC is enabled for VHT it should be enabled
+		 * for HT as well
+		 **/
+		if (param->stbc_flag)
+			cmd->peer_flags |= WMI_PEER_STBC;
+
+		/* Typically if LDPC is enabled for VHT it should be enabled
+		 * for HT as well
+		 **/
+		if (param->ldpc_flag)
+			cmd->peer_flags |= WMI_PEER_LDPC;
+
+		if (param->static_mimops_flag)
+			cmd->peer_flags |= WMI_PEER_STATIC_MIMOPS;
+		if (param->dynamic_mimops_flag)
+			cmd->peer_flags |= WMI_PEER_DYN_MIMOPS;
+		if (param->spatial_mux_flag)
+			cmd->peer_flags |= WMI_PEER_SPATIAL_MUX;
+		if (param->vht_flag)
+			cmd->peer_flags |= WMI_PEER_VHT;
+		if (param->he_flag)
+			cmd->peer_flags |= WMI_PEER_HE;
+	}
+
+	/* Suppress authorization for all AUTH modes that need 4-way handshake
+	 * (during re-association).
+	 * Authorization will be done for these modes on key installation.
+	 */
+	if (param->auth_flag)
+		cmd->peer_flags |= WMI_PEER_AUTH;
+	if (param->need_ptk_4_way)
+		cmd->peer_flags |= WMI_PEER_NEED_PTK_4_WAY;
+	else
+		cmd->peer_flags &= ~WMI_PEER_NEED_PTK_4_WAY;
+	if (param->need_gtk_2_way)
+		cmd->peer_flags |= WMI_PEER_NEED_GTK_2_WAY;
+	/* safe mode bypass the 4-way handshake */
+	if (param->safe_mode_enabled)
+		cmd->peer_flags &= ~(WMI_PEER_NEED_PTK_4_WAY |
+				     WMI_PEER_NEED_GTK_2_WAY);
+
+	if (param->is_pmf_enabled)
+		cmd->peer_flags |= WMI_PEER_PMF;
+
+	/* Disable AMSDU for station transmit, if user configures it */
+	/* Disable AMSDU for AP transmit to 11n Stations, if user configures
+	 * it
+	 * if (param->amsdu_disable) Add after FW support
+	 **/
+
+	/* Target asserts if node is marked HT and all MCS is set to 0.
+	 * Mark the node as non-HT if all the mcs rates are disabled through
+	 * iwpriv
+	 **/
+	if (param->peer_ht_rates.num_rates == 0)
+		cmd->peer_flags &= ~WMI_PEER_HT;
+}
+
+int ath11k_wmi_send_peer_assoc_cmd(struct ath11k *ar,
+				   struct peer_assoc_params *param)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_peer_assoc_complete_cmd *cmd;
+	struct wmi_vht_rate_set *mcs;
+	struct wmi_he_rate_set *he_mcs;
+	struct sk_buff *skb;
+	struct wmi_tlv *tlv;
+	void *ptr;
+	u32 peer_legacy_rates_align;
+	u32 peer_ht_rates_align;
+	int i, ret, len;
+
+	peer_legacy_rates_align = roundup(param->peer_legacy_rates.num_rates,
+					  sizeof(u32));
+	peer_ht_rates_align = roundup(param->peer_ht_rates.num_rates,
+				      sizeof(u32));
+
+	len = sizeof(*cmd) +
+	      TLV_HDR_SIZE + (peer_legacy_rates_align * sizeof(u8)) +
+	      TLV_HDR_SIZE + (peer_ht_rates_align * sizeof(u8)) +
+	      sizeof(*mcs) + TLV_HDR_SIZE +
+	      (sizeof(*he_mcs) * param->peer_he_mcs_count);
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	if (!skb)
+		return -ENOMEM;
+
+	ptr = skb->data;
+
+	cmd = ptr;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG,
+				     WMI_TAG_PEER_ASSOC_COMPLETE_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+
+	cmd->vdev_id = param->vdev_id;
+
+	cmd->peer_new_assoc = param->peer_new_assoc;
+	cmd->peer_associd = param->peer_associd;
+
+	ath11k_wmi_copy_peer_flags(cmd, param);
+
+	ether_addr_copy(cmd->peer_macaddr.addr, param->peer_mac);
+
+	cmd->peer_rate_caps = param->peer_rate_caps;
+	cmd->peer_caps = param->peer_caps;
+	cmd->peer_listen_intval = param->peer_listen_intval;
+	cmd->peer_ht_caps = param->peer_ht_caps;
+	cmd->peer_max_mpdu = param->peer_max_mpdu;
+	cmd->peer_mpdu_density = param->peer_mpdu_density;
+	cmd->peer_vht_caps = param->peer_vht_caps;
+	cmd->peer_phymode = param->peer_phymode;
+
+	/* Update 11ax capabilities */
+	cmd->peer_he_cap_info = param->peer_he_cap_macinfo[0];
+	cmd->peer_he_cap_info_ext = param->peer_he_cap_macinfo[1];
+	cmd->peer_he_cap_info_internal = param->peer_he_cap_macinfo_internal;
+	cmd->peer_he_ops = param->peer_he_ops;
+	memcpy(&cmd->peer_he_cap_phy, &param->peer_he_cap_phyinfo,
+	       sizeof(param->peer_he_cap_phyinfo));
+	memcpy(&cmd->peer_ppet, &param->peer_ppet,
+	       sizeof(param->peer_ppet));
+
+	/* Update peer legacy rate information */
+	ptr += sizeof(*cmd);
+
+	tlv = ptr;
+	tlv->header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_ARRAY_BYTE) |
+		      FIELD_PREP(WMI_TLV_LEN, peer_legacy_rates_align);
+
+	ptr += TLV_HDR_SIZE;
+
+	cmd->num_peer_legacy_rates = param->peer_legacy_rates.num_rates;
+	memcpy(ptr, param->peer_legacy_rates.rates,
+	       param->peer_legacy_rates.num_rates);
+
+	/* Update peer HT rate information */
+	ptr += peer_legacy_rates_align;
+
+	tlv = ptr;
+	tlv->header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_ARRAY_BYTE) |
+		      FIELD_PREP(WMI_TLV_LEN, peer_ht_rates_align);
+	ptr += TLV_HDR_SIZE;
+	cmd->num_peer_ht_rates = param->peer_ht_rates.num_rates;
+	memcpy(ptr, param->peer_ht_rates.rates,
+	       param->peer_ht_rates.num_rates);
+
+	/* VHT Rates */
+	ptr += peer_ht_rates_align;
+
+	mcs = ptr;
+
+	mcs->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_VHT_RATE_SET) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*mcs) - TLV_HDR_SIZE);
+
+	cmd->peer_nss = param->peer_nss;
+
+	/* Update bandwidth-NSS mapping */
+	cmd->peer_bw_rxnss_override = 0;
+	cmd->peer_bw_rxnss_override |= param->peer_bw_rxnss_override;
+
+	if (param->vht_capable) {
+		mcs->rx_max_rate = param->rx_max_rate;
+		mcs->rx_mcs_set = param->rx_mcs_set;
+		mcs->tx_max_rate = param->tx_max_rate;
+		mcs->tx_mcs_set = param->tx_mcs_set;
+	}
+
+	/* HE Rates */
+	cmd->peer_he_mcs = param->peer_he_mcs_count;
+
+	ptr += sizeof(*mcs);
+
+	len = param->peer_he_mcs_count * sizeof(*he_mcs);
+
+	tlv = ptr;
+	tlv->header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_ARRAY_STRUCT) |
+		      FIELD_PREP(WMI_TLV_LEN, len);
+	ptr += TLV_HDR_SIZE;
+
+	/* Loop through the HE rate set */
+	for (i = 0; i < param->peer_he_mcs_count; i++) {
+		he_mcs = ptr;
+		he_mcs->tlv_header = FIELD_PREP(WMI_TLV_TAG,
+						WMI_TAG_HE_RATE_SET) |
+				     FIELD_PREP(WMI_TLV_LEN,
+						sizeof(*he_mcs) - TLV_HDR_SIZE);
+
+		he_mcs->rx_mcs_set = param->peer_he_rx_mcs_set[i];
+		he_mcs->tx_mcs_set = param->peer_he_tx_mcs_set[i];
+		ptr += sizeof(*he_mcs);
+	}
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, WMI_PEER_ASSOC_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab,
+			    "failed to send WMI_PEER_ASSOC_CMDID\n");
+		dev_kfree_skb(skb);
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+		   "WMI peer assoc vdev id %d assoc id %d peer mac %pM\n",
+		   param->vdev_id, param->peer_associd, param->peer_mac);
+
+	return ret;
+}
+
+void ath11k_wmi_start_scan_init(struct ath11k *ar,
+				struct scan_req_params *arg)
+{
+	/* setup commonly used values */
+	arg->scan_req_id = 1;
+	arg->scan_priority = WMI_SCAN_PRIORITY_LOW;
+	arg->dwell_time_active = 50;
+	arg->dwell_time_active_2g = 0;
+	arg->dwell_time_passive = 150;
+	arg->min_rest_time = 50;
+	arg->max_rest_time = 500;
+	arg->repeat_probe_time = 0;
+	arg->probe_spacing_time = 0;
+	arg->idle_time = 0;
+	arg->max_scan_time = 20000;
+	arg->probe_delay = 5;
+	arg->notify_scan_events = WMI_SCAN_EVENT_STARTED |
+				  WMI_SCAN_EVENT_COMPLETED |
+				  WMI_SCAN_EVENT_BSS_CHANNEL |
+				  WMI_SCAN_EVENT_FOREIGN_CHAN |
+				  WMI_SCAN_EVENT_DEQUEUED;
+	arg->scan_flags |= WMI_SCAN_CHAN_STAT_EVENT;
+	arg->num_bssid = 1;
+}
+
+static inline void
+ath11k_wmi_copy_scan_event_cntrl_flags(struct wmi_start_scan_cmd *cmd,
+				       struct scan_req_params *param)
+{
+	/* Scan events subscription */
+	if (param->scan_ev_started)
+		cmd->notify_scan_events |=  WMI_SCAN_EVENT_STARTED;
+	if (param->scan_ev_completed)
+		cmd->notify_scan_events |=  WMI_SCAN_EVENT_COMPLETED;
+	if (param->scan_ev_bss_chan)
+		cmd->notify_scan_events |=  WMI_SCAN_EVENT_BSS_CHANNEL;
+	if (param->scan_ev_foreign_chan)
+		cmd->notify_scan_events |=  WMI_SCAN_EVENT_FOREIGN_CHAN;
+	if (param->scan_ev_dequeued)
+		cmd->notify_scan_events |=  WMI_SCAN_EVENT_DEQUEUED;
+	if (param->scan_ev_preempted)
+		cmd->notify_scan_events |=  WMI_SCAN_EVENT_PREEMPTED;
+	if (param->scan_ev_start_failed)
+		cmd->notify_scan_events |=  WMI_SCAN_EVENT_START_FAILED;
+	if (param->scan_ev_restarted)
+		cmd->notify_scan_events |=  WMI_SCAN_EVENT_RESTARTED;
+	if (param->scan_ev_foreign_chn_exit)
+		cmd->notify_scan_events |=  WMI_SCAN_EVENT_FOREIGN_CHAN_EXIT;
+	if (param->scan_ev_suspended)
+		cmd->notify_scan_events |=  WMI_SCAN_EVENT_SUSPENDED;
+	if (param->scan_ev_resumed)
+		cmd->notify_scan_events |=  WMI_SCAN_EVENT_RESUMED;
+
+	/** Set scan control flags */
+	cmd->scan_ctrl_flags = 0;
+	if (param->scan_f_passive)
+		cmd->scan_ctrl_flags |=  WMI_SCAN_FLAG_PASSIVE;
+	if (param->scan_f_strict_passive_pch)
+		cmd->scan_ctrl_flags |=  WMI_SCAN_FLAG_STRICT_PASSIVE_ON_PCHN;
+	if (param->scan_f_promisc_mode)
+		cmd->scan_ctrl_flags |=  WMI_SCAN_FILTER_PROMISCUOS;
+	if (param->scan_f_capture_phy_err)
+		cmd->scan_ctrl_flags |=  WMI_SCAN_CAPTURE_PHY_ERROR;
+	if (param->scan_f_half_rate)
+		cmd->scan_ctrl_flags |=  WMI_SCAN_FLAG_HALF_RATE_SUPPORT;
+	if (param->scan_f_quarter_rate)
+		cmd->scan_ctrl_flags |=  WMI_SCAN_FLAG_QUARTER_RATE_SUPPORT;
+	if (param->scan_f_cck_rates)
+		cmd->scan_ctrl_flags |=  WMI_SCAN_ADD_CCK_RATES;
+	if (param->scan_f_ofdm_rates)
+		cmd->scan_ctrl_flags |=  WMI_SCAN_ADD_OFDM_RATES;
+	if (param->scan_f_chan_stat_evnt)
+		cmd->scan_ctrl_flags |=  WMI_SCAN_CHAN_STAT_EVENT;
+	if (param->scan_f_filter_prb_req)
+		cmd->scan_ctrl_flags |=  WMI_SCAN_FILTER_PROBE_REQ;
+	if (param->scan_f_bcast_probe)
+		cmd->scan_ctrl_flags |=  WMI_SCAN_ADD_BCAST_PROBE_REQ;
+	if (param->scan_f_offchan_mgmt_tx)
+		cmd->scan_ctrl_flags |=  WMI_SCAN_OFFCHAN_MGMT_TX;
+	if (param->scan_f_offchan_data_tx)
+		cmd->scan_ctrl_flags |=  WMI_SCAN_OFFCHAN_DATA_TX;
+	if (param->scan_f_force_active_dfs_chn)
+		cmd->scan_ctrl_flags |=  WMI_SCAN_FLAG_FORCE_ACTIVE_ON_DFS;
+	if (param->scan_f_add_tpc_ie_in_probe)
+		cmd->scan_ctrl_flags |=  WMI_SCAN_ADD_TPC_IE_IN_PROBE_REQ;
+	if (param->scan_f_add_ds_ie_in_probe)
+		cmd->scan_ctrl_flags |=  WMI_SCAN_ADD_DS_IE_IN_PROBE_REQ;
+	if (param->scan_f_add_spoofed_mac_in_probe)
+		cmd->scan_ctrl_flags |=  WMI_SCAN_ADD_SPOOF_MAC_IN_PROBE_REQ;
+	if (param->scan_f_add_rand_seq_in_probe)
+		cmd->scan_ctrl_flags |=  WMI_SCAN_RANDOM_SEQ_NO_IN_PROBE_REQ;
+	if (param->scan_f_en_ie_whitelist_in_probe)
+		cmd->scan_ctrl_flags |=
+			 WMI_SCAN_ENABLE_IE_WHTELIST_IN_PROBE_REQ;
+
+	/* for adaptive scan mode using 3 bits (21 - 23 bits) */
+	WMI_SCAN_SET_DWELL_MODE(cmd->scan_ctrl_flags,
+				param->adaptive_dwell_time_mode);
+}
+
+int ath11k_wmi_send_scan_start_cmd(struct ath11k *ar,
+				   struct scan_req_params *params)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_start_scan_cmd *cmd;
+	struct wmi_ssid *ssid = NULL;
+	struct wmi_mac_addr *bssid;
+	struct sk_buff *skb;
+	struct wmi_tlv *tlv;
+	void *ptr;
+	int i, ret, len;
+	u32 *tmp_ptr;
+	u8 extraie_len_with_pad = 0;
+
+	len = sizeof(*cmd);
+
+	len += TLV_HDR_SIZE;
+	if (params->num_chan)
+		len += params->num_chan * sizeof(u32);
+
+	len += TLV_HDR_SIZE;
+	if (params->num_ssids)
+		len += params->num_ssids * sizeof(*ssid);
+
+	len += TLV_HDR_SIZE;
+	if (params->num_bssid)
+		len += sizeof(*bssid) * params->num_bssid;
+
+	len += TLV_HDR_SIZE;
+	if (params->extraie.len)
+		extraie_len_with_pad =
+			roundup(params->extraie.len, sizeof(u32));
+	len += extraie_len_with_pad;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	if (!skb)
+		return -ENOMEM;
+
+	ptr = skb->data;
+
+	cmd = ptr;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_START_SCAN_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+
+	cmd->scan_id = params->scan_id;
+	cmd->scan_req_id = params->scan_req_id;
+	cmd->vdev_id = params->vdev_id;
+	cmd->scan_priority = params->scan_priority;
+	cmd->notify_scan_events = params->notify_scan_events;
+
+	ath11k_wmi_copy_scan_event_cntrl_flags(cmd, params);
+
+	cmd->dwell_time_active = params->dwell_time_active;
+	cmd->dwell_time_active_2g = params->dwell_time_active_2g;
+	cmd->dwell_time_passive = params->dwell_time_passive;
+	cmd->min_rest_time = params->min_rest_time;
+	cmd->max_rest_time = params->max_rest_time;
+	cmd->repeat_probe_time = params->repeat_probe_time;
+	cmd->probe_spacing_time = params->probe_spacing_time;
+	cmd->idle_time = params->idle_time;
+	cmd->max_scan_time = params->max_scan_time;
+	cmd->probe_delay = params->probe_delay;
+	cmd->burst_duration = params->burst_duration;
+	cmd->num_chan = params->num_chan;
+	cmd->num_bssid = params->num_bssid;
+	cmd->num_ssids = params->num_ssids;
+	cmd->ie_len = params->extraie.len;
+	cmd->n_probes = params->n_probes;
+
+	ptr += sizeof(*cmd);
+
+	len = params->num_chan * sizeof(u32);
+
+	tlv = ptr;
+	tlv->header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_ARRAY_UINT32) |
+		      FIELD_PREP(WMI_TLV_LEN, len);
+	ptr += TLV_HDR_SIZE;
+	tmp_ptr = (u32 *)ptr;
+
+	for (i = 0; i < params->num_chan; ++i)
+		tmp_ptr[i] = params->chan_list[i];
+
+	ptr += len;
+
+	len = params->num_ssids * sizeof(*ssid);
+	tlv = ptr;
+	tlv->header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_ARRAY_FIXED_STRUCT) |
+		      FIELD_PREP(WMI_TLV_LEN, len);
+
+	ptr += TLV_HDR_SIZE;
+
+	if (params->num_ssids) {
+		ssid = ptr;
+		for (i = 0; i < params->num_ssids; ++i) {
+			ssid->ssid_len = params->ssid[i].length;
+			memcpy(ssid->ssid, params->ssid[i].ssid,
+			       params->ssid[i].length);
+			ssid++;
+		}
+	}
+
+	ptr += (params->num_ssids * sizeof(*ssid));
+	len = params->num_bssid * sizeof(*bssid);
+	tlv = ptr;
+	tlv->header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_ARRAY_FIXED_STRUCT) |
+		      FIELD_PREP(WMI_TLV_LEN, len);
+
+	ptr += TLV_HDR_SIZE;
+	bssid = ptr;
+
+	if (params->num_bssid) {
+		for (i = 0; i < params->num_bssid; ++i) {
+			ether_addr_copy(bssid->addr,
+					params->bssid_list[i].addr);
+			bssid++;
+		}
+	}
+
+	ptr += params->num_bssid * sizeof(*bssid);
+
+	len = extraie_len_with_pad;
+	tlv = ptr;
+	tlv->header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_ARRAY_BYTE) |
+		      FIELD_PREP(WMI_TLV_LEN, len);
+	ptr += TLV_HDR_SIZE;
+
+	if (params->extraie.len)
+		memcpy(ptr, params->extraie.ptr,
+		       params->extraie.len);
+
+	ptr += extraie_len_with_pad;
+
+	ret = ath11k_wmi_cmd_send(wmi, skb,
+				  WMI_START_SCAN_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to send WMI_START_SCAN_CMDID\n");
+		dev_kfree_skb(skb);
+	}
+
+	return ret;
+}
+
+int ath11k_wmi_send_scan_stop_cmd(struct ath11k *ar,
+				  struct scan_cancel_param *param)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_stop_scan_cmd *cmd;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_stop_scan_cmd *)skb->data;
+
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_STOP_SCAN_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+
+	cmd->vdev_id = param->vdev_id;
+	cmd->requestor = param->requester;
+	cmd->scan_id = param->scan_id;
+	cmd->pdev_id = param->pdev_id;
+	/* stop the scan with the corresponding scan_id */
+	if (param->req_type == WLAN_SCAN_CANCEL_PDEV_ALL) {
+		/* Cancelling all scans */
+		cmd->req_type =  WMI_SCAN_STOP_ALL;
+	} else if (param->req_type == WLAN_SCAN_CANCEL_VDEV_ALL) {
+		/* Cancelling VAP scans */
+		cmd->req_type =  WMI_SCN_STOP_VAP_ALL;
+	} else if (param->req_type == WLAN_SCAN_CANCEL_SINGLE) {
+		/* Cancelling specific scan */
+		cmd->req_type =  WMI_SCAN_STOP_ONE;
+	} else {
+		ath11k_warn(ar->ab, "invalid scan cancel param %d",
+			    param->req_type);
+		dev_kfree_skb(skb);
+		return -EINVAL;
+	}
+
+	ret = ath11k_wmi_cmd_send(wmi, skb,
+				  WMI_STOP_SCAN_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to send WMI_STOP_SCAN_CMDID\n");
+		dev_kfree_skb(skb);
+	}
+
+	return ret;
+}
+
+int ath11k_wmi_send_scan_chan_list_cmd(struct ath11k *ar,
+				       struct scan_chan_list_params *chan_list)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_scan_chan_list_cmd *cmd;
+	struct sk_buff *skb;
+	struct wmi_channel *chan_info;
+	struct channel_param *tchan_info;
+	struct wmi_tlv *tlv;
+	void *ptr;
+	int i, ret, len;
+	u32 *reg1, *reg2;
+
+	len = sizeof(*cmd) + TLV_HDR_SIZE +
+		 sizeof(*chan_info) * chan_list->nallchans;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_scan_chan_list_cmd *)skb->data;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_SCAN_CHAN_LIST_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+		   "WMI no.of chan = %d len = %d\n", chan_list->nallchans, len);
+	cmd->pdev_id = chan_list->pdev_id;
+	cmd->num_scan_chans = chan_list->nallchans;
+
+	ptr = skb->data + sizeof(*cmd);
+
+	len = sizeof(*chan_info) * chan_list->nallchans;
+	tlv = ptr;
+	tlv->header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_ARRAY_STRUCT) |
+		      FIELD_PREP(WMI_TLV_LEN, len - TLV_HDR_SIZE);
+	ptr += TLV_HDR_SIZE;
+
+	tchan_info = &chan_list->ch_param[0];
+
+	for (i = 0; i < chan_list->nallchans; ++i) {
+		chan_info = ptr;
+		memset(chan_info, 0, sizeof(*chan_info));
+		len = sizeof(*chan_info);
+		chan_info->tlv_header = FIELD_PREP(WMI_TLV_TAG,
+						   WMI_TAG_CHANNEL) |
+					FIELD_PREP(WMI_TLV_LEN,
+						   len - TLV_HDR_SIZE);
+
+		reg1 = &chan_info->reg_info_1;
+		reg2 = &chan_info->reg_info_2;
+		chan_info->mhz = tchan_info->mhz;
+		chan_info->band_center_freq1 = tchan_info->cfreq1;
+		chan_info->band_center_freq2 = tchan_info->cfreq2;
+
+		if (tchan_info->is_chan_passive)
+			chan_info->info |= WMI_CHAN_INFO_PASSIVE;
+		if (tchan_info->allow_vht)
+			chan_info->info |= WMI_CHAN_INFO_ALLOW_VHT;
+		else if (tchan_info->allow_ht)
+			chan_info->info |= WMI_CHAN_INFO_ALLOW_HT;
+		if (tchan_info->half_rate)
+			chan_info->info |= WMI_CHAN_INFO_HALF_RATE;
+		if (tchan_info->quarter_rate)
+			chan_info->info |= WMI_CHAN_INFO_QUARTER_RATE;
+
+		chan_info->info |= FIELD_PREP(WMI_CHAN_INFO_MODE,
+					      tchan_info->phy_mode);
+		*reg1 |= FIELD_PREP(WMI_CHAN_REG_INFO1_MIN_PWR,
+				    tchan_info->minpower);
+		*reg1 |= FIELD_PREP(WMI_CHAN_REG_INFO1_MAX_PWR,
+				    tchan_info->maxpower);
+		*reg1 |= FIELD_PREP(WMI_CHAN_REG_INFO1_MAX_REG_PWR,
+				    tchan_info->maxregpower);
+		*reg1 |= FIELD_PREP(WMI_CHAN_REG_INFO1_REG_CLS,
+				    tchan_info->reg_class_id);
+		*reg2 |= FIELD_PREP(WMI_CHAN_REG_INFO2_ANT_MAX,
+				    tchan_info->antennamax);
+
+		ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+			   "WMI chan scan list chan[%d] = %u\n",
+			   i, chan_info->mhz);
+
+		ptr += sizeof(*chan_info);
+
+		tchan_info++;
+	}
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, WMI_SCAN_CHAN_LIST_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to send WMI_SCAN_CHAN_LIST cmd\n");
+		dev_kfree_skb(skb);
+	}
+
+	return ret;
+}
+
+int ath11k_wmi_send_wmm_update_cmd_tlv(struct ath11k *ar, u32 vdev_id,
+				       struct wmi_wmm_params_all_arg *param)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_vdev_set_wmm_params_cmd *cmd;
+	struct wmi_wmm_params *wmm_param;
+	struct wmi_wmm_params_arg *wmi_wmm_arg;
+	struct sk_buff *skb;
+	int ret, ac;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_vdev_set_wmm_params_cmd *)skb->data;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG,
+				     WMI_TAG_VDEV_SET_WMM_PARAMS_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+
+	cmd->vdev_id = vdev_id;
+	cmd->wmm_param_type = 0;
+
+	for (ac = 0; ac < WME_NUM_AC; ac++) {
+		switch (ac) {
+		case WME_AC_BE:
+			wmi_wmm_arg = &param->ac_be;
+			break;
+		case WME_AC_BK:
+			wmi_wmm_arg = &param->ac_bk;
+			break;
+		case WME_AC_VI:
+			wmi_wmm_arg = &param->ac_vi;
+			break;
+		case WME_AC_VO:
+			wmi_wmm_arg = &param->ac_vo;
+			break;
+		}
+
+		wmm_param = (struct wmi_wmm_params *)&cmd->wmm_params[ac];
+		wmm_param->tlv_header =
+				FIELD_PREP(WMI_TLV_TAG,
+					   WMI_TAG_VDEV_SET_WMM_PARAMS_CMD) |
+				FIELD_PREP(WMI_TLV_LEN,
+					   sizeof(*wmm_param) - TLV_HDR_SIZE);
+
+		wmm_param->aifs = wmi_wmm_arg->aifs;
+		wmm_param->cwmin = wmi_wmm_arg->cwmin;
+		wmm_param->cwmax = wmi_wmm_arg->cwmax;
+		wmm_param->txoplimit = wmi_wmm_arg->txop;
+		wmm_param->acm = wmi_wmm_arg->acm;
+		wmm_param->no_ack = wmi_wmm_arg->no_ack;
+
+		ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+			   "wmi wmm set ac %d aifs %d cwmin %d cwmax %d txop %d acm %d no_ack %d\n",
+			   ac, wmm_param->aifs, wmm_param->cwmin,
+			   wmm_param->cwmax, wmm_param->txoplimit,
+			   wmm_param->acm, wmm_param->no_ack);
+	}
+	ret = ath11k_wmi_cmd_send(wmi, skb,
+				  WMI_VDEV_SET_WMM_PARAMS_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab,
+			    "failed to send WMI_VDEV_SET_WMM_PARAMS_CMDID");
+		dev_kfree_skb(skb);
+	}
+
+	return ret;
+}
+
+int ath11k_wmi_send_dfs_phyerr_offload_enable_cmd(struct ath11k *ar,
+						  u32 pdev_id)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_dfs_phyerr_offload_cmd *cmd;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_dfs_phyerr_offload_cmd *)skb->data;
+	cmd->tlv_header =
+		FIELD_PREP(WMI_TLV_TAG,
+			   WMI_TAG_PDEV_DFS_PHYERR_OFFLOAD_ENABLE_CMD) |
+		FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+
+	cmd->pdev_id = pdev_id;
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+		   "WMI dfs phy err offload enable pdev id %d\n", pdev_id);
+
+	ret = ath11k_wmi_cmd_send(wmi, skb,
+				  WMI_PDEV_DFS_PHYERR_OFFLOAD_ENABLE_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab,
+			    "failed to send WMI_PDEV_DFS_PHYERR_OFFLOAD_ENABLE cmd\n");
+		dev_kfree_skb(skb);
+	}
+
+	return ret;
+}
+
+int ath11k_wmi_pdev_peer_pktlog_filter(struct ath11k *ar, u8 *addr, u8 enable)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_pdev_pktlog_filter_cmd *cmd;
+	struct wmi_pdev_pktlog_filter_info *info;
+	struct sk_buff *skb;
+	struct wmi_tlv *tlv;
+	void *ptr;
+	int ret, len;
+
+	len = sizeof(*cmd) + sizeof(*info) + TLV_HDR_SIZE;
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_pdev_pktlog_filter_cmd *)skb->data;
+
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_PDEV_PEER_PKTLOG_FILTER_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+
+	cmd->pdev_id = ar->pdev->pdev_id;
+	cmd->num_mac = 1;
+	cmd->enable = enable;
+
+	ptr = skb->data + sizeof(*cmd);
+
+	tlv = ptr;
+	tlv->header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_ARRAY_STRUCT) |
+		      FIELD_PREP(WMI_TLV_LEN, 0);
+
+	ptr += TLV_HDR_SIZE;
+	info = ptr;
+
+	ether_addr_copy(info->peer_macaddr.addr, addr);
+	info->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_PDEV_PEER_PKTLOG_FILTER_INFO) |
+			   FIELD_PREP(WMI_TLV_LEN,
+				      sizeof(*info) - TLV_HDR_SIZE);
+
+	ret = ath11k_wmi_cmd_send(wmi, skb,
+				  WMI_PDEV_PKTLOG_FILTER_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to send WMI_PDEV_PKTLOG_ENABLE_CMDID\n");
+		dev_kfree_skb(skb);
+	}
+
+	return ret;
+}
+
+int
+ath11k_wmi_send_init_country_cmd(struct ath11k *ar,
+				 struct wmi_init_country_params init_cc_params)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_init_country_cmd *cmd;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_init_country_cmd *)skb->data;
+	cmd->tlv_header =
+		FIELD_PREP(WMI_TLV_TAG,
+			   WMI_TAG_SET_INIT_COUNTRY_CMD) |
+		FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+
+	cmd->pdev_id = ar->pdev->pdev_id;
+
+	switch (init_cc_params.flags) {
+	case ALPHA_IS_SET:
+		cmd->init_cc_type = WMI_COUNTRY_INFO_TYPE_ALPHA;
+		memcpy((u8 *)&cmd->cc_info.alpha2,
+		       init_cc_params.cc_info.alpha2, 3);
+		break;
+	case CC_IS_SET:
+		cmd->init_cc_type = WMI_COUNTRY_INFO_TYPE_COUNTRY_CODE;
+		cmd->cc_info.country_code = init_cc_params.cc_info.country_code;
+		break;
+	case REGDMN_IS_SET:
+		cmd->init_cc_type = WMI_COUNTRY_INFO_TYPE_REGDOMAIN;
+		cmd->cc_info.regdom_id = init_cc_params.cc_info.regdom_id;
+		break;
+	default:
+		ret = -EINVAL;
+		goto out;
+	}
+
+	ret = ath11k_wmi_cmd_send(wmi, skb,
+				  WMI_SET_INIT_COUNTRY_CMDID);
+
+out:
+	if (ret) {
+		ath11k_warn(ar->ab,
+			    "failed to send WMI_SET_INIT_COUNTRY CMD :%d\n",
+			    ret);
+		dev_kfree_skb(skb);
+	}
+
+	return ret;
+}
+
+int ath11k_wmi_pdev_pktlog_enable(struct ath11k *ar, u32 pktlog_filter)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_pktlog_enable_cmd *cmd;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_pktlog_enable_cmd *)skb->data;
+
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_PDEV_PKTLOG_ENABLE_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+
+	cmd->pdev_id = ar->pdev->pdev_id;
+	cmd->evlist = pktlog_filter;
+	cmd->enable = ATH11K_WMI_PKTLOG_ENABLE_FORCE;
+
+	ret = ath11k_wmi_cmd_send(wmi, skb,
+				  WMI_PDEV_PKTLOG_ENABLE_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to send WMI_PDEV_PKTLOG_ENABLE_CMDID\n");
+		dev_kfree_skb(skb);
+	}
+
+	return ret;
+}
+
+int ath11k_wmi_pdev_pktlog_disable(struct ath11k *ar)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_pktlog_disable_cmd *cmd;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_pktlog_disable_cmd *)skb->data;
+
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_PDEV_PKTLOG_DISABLE_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+
+	cmd->pdev_id = ar->pdev->pdev_id;
+
+	ret = ath11k_wmi_cmd_send(wmi, skb,
+				  WMI_PDEV_PKTLOG_DISABLE_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to send WMI_PDEV_PKTLOG_ENABLE_CMDID\n");
+		dev_kfree_skb(skb);
+	}
+
+	return ret;
+}
+
+static void
+ath11k_fill_band_to_mac_param(struct ath11k_base  *soc,
+			      struct wmi_host_pdev_band_to_mac *band_to_mac)
+{
+	u8 i;
+	struct ath11k_hal_reg_capabilities_ext *hal_reg_cap;
+	struct ath11k_pdev *pdev;
+
+	for (i = 0; i < soc->num_radios; i++) {
+		pdev = &soc->pdevs[i];
+		hal_reg_cap = &soc->hal_reg_cap[i];
+		band_to_mac[i].pdev_id = pdev->pdev_id;
+
+		switch (pdev->cap.supported_bands) {
+		case WMI_HOST_WLAN_2G_5G_CAP:
+			band_to_mac[i].start_freq = hal_reg_cap->low_2ghz_chan;
+			band_to_mac[i].end_freq = hal_reg_cap->high_5ghz_chan;
+			break;
+		case WMI_HOST_WLAN_2G_CAP:
+			band_to_mac[i].start_freq = hal_reg_cap->low_2ghz_chan;
+			band_to_mac[i].end_freq = hal_reg_cap->high_2ghz_chan;
+			break;
+		case WMI_HOST_WLAN_5G_CAP:
+			band_to_mac[i].start_freq = hal_reg_cap->low_5ghz_chan;
+			band_to_mac[i].end_freq = hal_reg_cap->high_5ghz_chan;
+			break;
+		default:
+			break;
+		}
+	}
+}
+
+static void
+ath11k_wmi_copy_resource_config(struct wmi_resource_config *wmi_cfg,
+				struct target_resource_config *tg_cfg)
+{
+	wmi_cfg->num_vdevs = tg_cfg->num_vdevs;
+	wmi_cfg->num_peers = tg_cfg->num_peers;
+	wmi_cfg->num_offload_peers = tg_cfg->num_offload_peers;
+	wmi_cfg->num_offload_reorder_buffs = tg_cfg->num_offload_reorder_buffs;
+	wmi_cfg->num_peer_keys = tg_cfg->num_peer_keys;
+	wmi_cfg->num_tids = tg_cfg->num_tids;
+	wmi_cfg->ast_skid_limit = tg_cfg->ast_skid_limit;
+	wmi_cfg->tx_chain_mask = tg_cfg->tx_chain_mask;
+	wmi_cfg->rx_chain_mask = tg_cfg->rx_chain_mask;
+	wmi_cfg->rx_timeout_pri[0] = tg_cfg->rx_timeout_pri[0];
+	wmi_cfg->rx_timeout_pri[1] = tg_cfg->rx_timeout_pri[1];
+	wmi_cfg->rx_timeout_pri[2] = tg_cfg->rx_timeout_pri[2];
+	wmi_cfg->rx_timeout_pri[3] = tg_cfg->rx_timeout_pri[3];
+	wmi_cfg->rx_decap_mode = tg_cfg->rx_decap_mode;
+	wmi_cfg->scan_max_pending_req = tg_cfg->scan_max_pending_req;
+	wmi_cfg->bmiss_offload_max_vdev = tg_cfg->bmiss_offload_max_vdev;
+	wmi_cfg->roam_offload_max_vdev = tg_cfg->roam_offload_max_vdev;
+	wmi_cfg->roam_offload_max_ap_profiles =
+		tg_cfg->roam_offload_max_ap_profiles;
+	wmi_cfg->num_mcast_groups = tg_cfg->num_mcast_groups;
+	wmi_cfg->num_mcast_table_elems = tg_cfg->num_mcast_table_elems;
+	wmi_cfg->mcast2ucast_mode = tg_cfg->mcast2ucast_mode;
+	wmi_cfg->tx_dbg_log_size = tg_cfg->tx_dbg_log_size;
+	wmi_cfg->num_wds_entries = tg_cfg->num_wds_entries;
+	wmi_cfg->dma_burst_size = tg_cfg->dma_burst_size;
+	wmi_cfg->mac_aggr_delim = tg_cfg->mac_aggr_delim;
+	wmi_cfg->rx_skip_defrag_timeout_dup_detection_check =
+		tg_cfg->rx_skip_defrag_timeout_dup_detection_check;
+	wmi_cfg->vow_config = tg_cfg->vow_config;
+	wmi_cfg->gtk_offload_max_vdev = tg_cfg->gtk_offload_max_vdev;
+	wmi_cfg->num_msdu_desc = tg_cfg->num_msdu_desc;
+	wmi_cfg->max_frag_entries = tg_cfg->max_frag_entries;
+	wmi_cfg->num_tdls_vdevs = tg_cfg->num_tdls_vdevs;
+	wmi_cfg->num_tdls_conn_table_entries =
+		tg_cfg->num_tdls_conn_table_entries;
+	wmi_cfg->beacon_tx_offload_max_vdev =
+		tg_cfg->beacon_tx_offload_max_vdev;
+	wmi_cfg->num_multicast_filter_entries =
+		tg_cfg->num_multicast_filter_entries;
+	wmi_cfg->num_wow_filters = tg_cfg->num_wow_filters;
+	wmi_cfg->num_keep_alive_pattern = tg_cfg->num_keep_alive_pattern;
+	wmi_cfg->keep_alive_pattern_size = tg_cfg->keep_alive_pattern_size;
+	wmi_cfg->max_tdls_concurrent_sleep_sta =
+		tg_cfg->max_tdls_concurrent_sleep_sta;
+	wmi_cfg->max_tdls_concurrent_buffer_sta =
+		tg_cfg->max_tdls_concurrent_buffer_sta;
+	wmi_cfg->wmi_send_separate = tg_cfg->wmi_send_separate;
+	wmi_cfg->num_ocb_vdevs = tg_cfg->num_ocb_vdevs;
+	wmi_cfg->num_ocb_channels = tg_cfg->num_ocb_channels;
+	wmi_cfg->num_ocb_schedules = tg_cfg->num_ocb_schedules;
+	wmi_cfg->bpf_instruction_size = tg_cfg->bpf_instruction_size;
+	wmi_cfg->max_bssid_rx_filters = tg_cfg->max_bssid_rx_filters;
+	wmi_cfg->use_pdev_id = tg_cfg->use_pdev_id;
+	wmi_cfg->flag1 = tg_cfg->atf_config;
+	wmi_cfg->peer_map_unmap_v2_support = tg_cfg->peer_map_unmap_v2_support;
+}
+
+static int ath11k_init_cmd_send(struct ath11k_pdev_wmi *wmi,
+				struct wmi_init_cmd_param *param)
+{
+	struct ath11k_base *ab = wmi->wmi_sc->ab;
+	struct sk_buff *skb;
+	struct wmi_init_cmd *cmd;
+	struct wmi_resource_config *cfg;
+	struct wmi_pdev_set_hw_mode_cmd_param *hw_mode;
+	struct wmi_pdev_band_to_mac *band_to_mac;
+	struct wlan_host_mem_chunk *host_mem_chunks;
+	struct wmi_tlv *tlv;
+	size_t ret, len;
+	void *ptr;
+	u32 hw_mode_len = 0;
+	u16 idx;
+
+	if (param->hw_mode_id != WMI_HOST_HW_MODE_MAX)
+		hw_mode_len = sizeof(*hw_mode) + TLV_HDR_SIZE +
+			      (param->num_band_to_mac * sizeof(*band_to_mac));
+
+	len = sizeof(*cmd) + TLV_HDR_SIZE + sizeof(*cfg) + hw_mode_len +
+	      (sizeof(*host_mem_chunks) * WMI_MAX_MEM_REQS);
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_init_cmd *)skb->data;
+
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_INIT_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+
+	ptr = skb->data + sizeof(*cmd);
+	cfg = ptr;
+
+	ath11k_wmi_copy_resource_config(cfg, param->res_cfg);
+
+	cfg->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_RESOURCE_CONFIG) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cfg) - TLV_HDR_SIZE);
+
+	ptr += sizeof(*cfg);
+	host_mem_chunks = ptr + TLV_HDR_SIZE;
+	len = sizeof(struct wlan_host_mem_chunk);
+
+	for (idx = 0; idx < param->num_mem_chunks; ++idx) {
+		host_mem_chunks[idx].tlv_header =
+				FIELD_PREP(WMI_TLV_TAG,
+					   WMI_TAG_WLAN_HOST_MEMORY_CHUNK) |
+				FIELD_PREP(WMI_TLV_LEN, len);
+
+		host_mem_chunks[idx].ptr = param->mem_chunks[idx].paddr;
+		host_mem_chunks[idx].size = param->mem_chunks[idx].len;
+		host_mem_chunks[idx].req_id = param->mem_chunks[idx].req_id;
+
+		ath11k_dbg(ab, ATH11K_DBG_WMI,
+			   "WMI host mem chunk req_id %d paddr 0x%llx len %d\n",
+			   param->mem_chunks[idx].req_id,
+			   (u64)param->mem_chunks[idx].paddr,
+			   param->mem_chunks[idx].len);
+	}
+	cmd->num_host_mem_chunks = param->num_mem_chunks;
+	len = sizeof(struct wlan_host_mem_chunk) * param->num_mem_chunks;
+
+	/* num_mem_chunks is zero */
+	tlv = ptr;
+	tlv->header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_ARRAY_STRUCT) |
+		      FIELD_PREP(WMI_TLV_LEN, len);
+	ptr += TLV_HDR_SIZE + len;
+
+	if (param->hw_mode_id != WMI_HOST_HW_MODE_MAX) {
+		hw_mode = (struct wmi_pdev_set_hw_mode_cmd_param *)ptr;
+		hw_mode->tlv_header = FIELD_PREP(WMI_TLV_TAG,
+						 WMI_TAG_PDEV_SET_HW_MODE_CMD) |
+				      FIELD_PREP(WMI_TLV_LEN,
+						 sizeof(*hw_mode) - TLV_HDR_SIZE);
+
+		hw_mode->hw_mode_index = param->hw_mode_id;
+		hw_mode->num_band_to_mac = param->num_band_to_mac;
+
+		ptr += sizeof(*hw_mode);
+
+		len = param->num_band_to_mac * sizeof(*band_to_mac);
+		tlv = ptr;
+		tlv->header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_ARRAY_STRUCT) |
+			      FIELD_PREP(WMI_TLV_LEN, len);
+
+		ptr += TLV_HDR_SIZE;
+		len = sizeof(*band_to_mac);
+
+		for (idx = 0; idx < param->num_band_to_mac; idx++) {
+			band_to_mac = (void *)ptr;
+
+			band_to_mac->tlv_header = FIELD_PREP(WMI_TLV_TAG,
+							     WMI_TAG_PDEV_BAND_TO_MAC) |
+						  FIELD_PREP(WMI_TLV_LEN,
+							     len - TLV_HDR_SIZE);
+			band_to_mac->pdev_id = param->band_to_mac[idx].pdev_id;
+			band_to_mac->start_freq =
+				param->band_to_mac[idx].start_freq;
+			band_to_mac->end_freq =
+				param->band_to_mac[idx].end_freq;
+			ptr += sizeof(*band_to_mac);
+		}
+	}
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, WMI_INIT_CMDID);
+	if (ret) {
+		ath11k_warn(ab, "failed to send WMI_INIT_CMDID\n");
+		dev_kfree_skb(skb);
+	}
+
+	return ret;
+}
+
+int ath11k_wmi_wait_for_service_ready(struct ath11k_base *ab)
+{
+	unsigned long time_left;
+
+	time_left = wait_for_completion_timeout(&ab->wmi_sc.service_ready,
+						WMI_SERVICE_READY_TIMEOUT_HZ);
+	if (!time_left)
+		return -ETIMEDOUT;
+
+	return 0;
+}
+
+int ath11k_wmi_wait_for_unified_ready(struct ath11k_base *ab)
+{
+	unsigned long time_left;
+
+	time_left = wait_for_completion_timeout(&ab->wmi_sc.unified_ready,
+						WMI_SERVICE_READY_TIMEOUT_HZ);
+	if (!time_left)
+		return -ETIMEDOUT;
+
+	return 0;
+}
+
+int ath11k_wmi_cmd_init(struct ath11k_base *ab)
+{
+	struct ath11k_wmi_base *wmi_sc = &ab->wmi_sc;
+	struct wmi_init_cmd_param init_param;
+	struct target_resource_config  config;
+
+	memset(&init_param, 0, sizeof(init_param));
+	memset(&config, 0, sizeof(config));
+
+	config.num_vdevs = ab->num_radios * TARGET_NUM_VDEVS;
+
+	if (ab->num_radios == 2) {
+		config.num_peers = TARGET_NUM_PEERS(DBS);
+		config.num_tids = TARGET_NUM_TIDS(DBS);
+	} else if (ab->num_radios == 3) {
+		config.num_peers = TARGET_NUM_PEERS(DBS_SBS);
+		config.num_tids = TARGET_NUM_TIDS(DBS_SBS);
+	} else {
+		/* Control should not reach here */
+		config.num_peers = TARGET_NUM_PEERS(SINGLE);
+		config.num_tids = TARGET_NUM_TIDS(SINGLE);
+	}
+	config.num_offload_peers = TARGET_NUM_OFFLD_PEERS;
+	config.num_offload_reorder_buffs = TARGET_NUM_OFFLD_REORDER_BUFFS;
+	config.num_peer_keys = TARGET_NUM_PEER_KEYS;
+	config.ast_skid_limit = TARGET_AST_SKID_LIMIT;
+	config.tx_chain_mask = (1 << ab->target_caps.num_rf_chains) - 1;
+	config.rx_chain_mask = (1 << ab->target_caps.num_rf_chains) - 1;
+	config.rx_timeout_pri[0] = TARGET_RX_TIMEOUT_LO_PRI;
+	config.rx_timeout_pri[1] = TARGET_RX_TIMEOUT_LO_PRI;
+	config.rx_timeout_pri[2] = TARGET_RX_TIMEOUT_LO_PRI;
+	config.rx_timeout_pri[3] = TARGET_RX_TIMEOUT_HI_PRI;
+	config.rx_decap_mode = TARGET_DECAP_MODE_NATIVE_WIFI;
+	config.scan_max_pending_req = TARGET_SCAN_MAX_PENDING_REQS;
+	config.bmiss_offload_max_vdev = TARGET_BMISS_OFFLOAD_MAX_VDEV;
+	config.roam_offload_max_vdev = TARGET_ROAM_OFFLOAD_MAX_VDEV;
+	config.roam_offload_max_ap_profiles = TARGET_ROAM_OFFLOAD_MAX_AP_PROFILES;
+	config.num_mcast_groups = TARGET_NUM_MCAST_GROUPS;
+	config.num_mcast_table_elems = TARGET_NUM_MCAST_TABLE_ELEMS;
+	config.mcast2ucast_mode = TARGET_MCAST2UCAST_MODE;
+	config.tx_dbg_log_size = TARGET_TX_DBG_LOG_SIZE;
+	config.num_wds_entries = TARGET_NUM_WDS_ENTRIES;
+	config.dma_burst_size = TARGET_DMA_BURST_SIZE;
+	config.rx_skip_defrag_timeout_dup_detection_check =
+		TARGET_RX_SKIP_DEFRAG_TIMEOUT_DUP_DETECTION_CHECK;
+	config.vow_config = TARGET_VOW_CONFIG;
+	config.gtk_offload_max_vdev = TARGET_GTK_OFFLOAD_MAX_VDEV;
+	config.num_msdu_desc = TARGET_NUM_MSDU_DESC;
+	config.beacon_tx_offload_max_vdev = ab->num_radios * TARGET_MAX_BCN_OFFLD;
+	config.rx_batchmode = TARGET_RX_BATCHMODE;
+	config.peer_map_unmap_v2_support = 1;
+
+	memcpy(&wmi_sc->wlan_resource_config, &config, sizeof(config));
+
+	init_param.res_cfg = &wmi_sc->wlan_resource_config;
+	init_param.num_mem_chunks = wmi_sc->num_mem_chunks;
+	init_param.hw_mode_id = wmi_sc->preferred_hw_mode;
+	init_param.mem_chunks = wmi_sc->mem_chunks;
+
+	if (wmi_sc->preferred_hw_mode == WMI_HOST_HW_MODE_SINGLE)
+		init_param.hw_mode_id = WMI_HOST_HW_MODE_MAX;
+
+	init_param.num_band_to_mac = ab->num_radios;
+
+	ath11k_fill_band_to_mac_param(ab, init_param.band_to_mac);
+
+	return ath11k_init_cmd_send(&wmi_sc->wmi[0], &init_param);
+}
+
+static int ath11k_wmi_tlv_hw_mode_caps_parse(struct ath11k_base *soc,
+					     u16 tag, u16 len,
+					     const void *ptr, void *data)
+{
+	struct wmi_tlv_svc_rdy_ext_parse *svc_rdy_ext = data;
+	struct wmi_hw_mode_capabilities *hw_mode_cap;
+	u32 phy_map = 0;
+
+	if (tag != WMI_TAG_HW_MODE_CAPABILITIES)
+		return -EPROTO;
+
+	if (svc_rdy_ext->n_hw_mode_caps >= svc_rdy_ext->param.num_hw_modes)
+		return -ENOBUFS;
+
+	hw_mode_cap = container_of(ptr, struct wmi_hw_mode_capabilities,
+				   hw_mode_id);
+	svc_rdy_ext->n_hw_mode_caps++;
+
+	phy_map = hw_mode_cap->phy_id_map;
+	while (phy_map) {
+		svc_rdy_ext->tot_phy_id++;
+		phy_map = phy_map >> 1;
+	}
+
+	return 0;
+}
+
+static int ath11k_wmi_tlv_hw_mode_caps(struct ath11k_base *soc,
+				       u16 len, const void *ptr, void *data)
+{
+	struct wmi_tlv_svc_rdy_ext_parse *svc_rdy_ext = data;
+	struct wmi_hw_mode_capabilities *hw_mode_caps;
+	enum wmi_host_hw_mode_config_type mode, pref;
+	u32 i;
+	int ret;
+
+	svc_rdy_ext->n_hw_mode_caps = 0;
+	svc_rdy_ext->hw_mode_caps = (struct wmi_hw_mode_capabilities *)ptr;
+
+	ret = ath11k_wmi_tlv_iter(soc, ptr, len,
+				  ath11k_wmi_tlv_hw_mode_caps_parse,
+				  svc_rdy_ext);
+	if (ret) {
+		ath11k_warn(soc, "failed to parse tlv %d\n", ret);
+		return ret;
+	}
+
+	i = 0;
+	while (i < svc_rdy_ext->n_hw_mode_caps) {
+		hw_mode_caps = &svc_rdy_ext->hw_mode_caps[i];
+		mode = hw_mode_caps->hw_mode_id;
+		pref = soc->wmi_sc.preferred_hw_mode;
+
+		if (ath11k_hw_mode_pri_map[mode] < ath11k_hw_mode_pri_map[pref]) {
+			svc_rdy_ext->pref_hw_mode_caps = *hw_mode_caps;
+			soc->wmi_sc.preferred_hw_mode = mode;
+		}
+		i++;
+	}
+
+	if (soc->wmi_sc.preferred_hw_mode == WMI_HOST_HW_MODE_MAX)
+		return -EINVAL;
+
+	return 0;
+}
+
+static int ath11k_wmi_tlv_mac_phy_caps_parse(struct ath11k_base *soc,
+					     u16 tag, u16 len,
+					     const void *ptr, void *data)
+{
+	struct wmi_tlv_svc_rdy_ext_parse *svc_rdy_ext = data;
+
+	if (tag != WMI_TAG_MAC_PHY_CAPABILITIES)
+		return -EPROTO;
+
+	if (svc_rdy_ext->n_mac_phy_caps >= svc_rdy_ext->tot_phy_id)
+		return -ENOBUFS;
+
+	svc_rdy_ext->n_mac_phy_caps++;
+	return 0;
+}
+
+static int ath11k_wmi_tlv_ext_hal_reg_caps_parse(struct ath11k_base *soc,
+						 u16 tag, u16 len,
+						 const void *ptr, void *data)
+{
+	struct wmi_tlv_svc_rdy_ext_parse *svc_rdy_ext = data;
+
+	if (tag != WMI_TAG_HAL_REG_CAPABILITIES_EXT)
+		return -EPROTO;
+
+	if (svc_rdy_ext->n_ext_hal_reg_caps >= svc_rdy_ext->param.num_phy)
+		return -ENOBUFS;
+
+	svc_rdy_ext->n_ext_hal_reg_caps++;
+	return 0;
+}
+
+static int ath11k_wmi_tlv_ext_hal_reg_caps(struct ath11k_base *soc,
+					   u16 len, const void *ptr, void *data)
+{
+	struct ath11k_pdev_wmi *wmi_handle = &soc->wmi_sc.wmi[0];
+	struct wmi_tlv_svc_rdy_ext_parse *svc_rdy_ext = data;
+	struct ath11k_hal_reg_capabilities_ext reg_cap;
+	int ret;
+	u32 i;
+
+	svc_rdy_ext->n_ext_hal_reg_caps = 0;
+	svc_rdy_ext->ext_hal_reg_caps = (struct wmi_hal_reg_capabilities_ext *)ptr;
+	ret = ath11k_wmi_tlv_iter(soc, ptr, len,
+				  ath11k_wmi_tlv_ext_hal_reg_caps_parse,
+				  svc_rdy_ext);
+	if (ret) {
+		ath11k_warn(soc, "failed to parse tlv %d\n", ret);
+		return ret;
+	}
+
+	for (i = 0; i < svc_rdy_ext->param.num_phy; i++) {
+		ret = ath11k_pull_reg_cap_svc_rdy_ext(wmi_handle,
+						      svc_rdy_ext->soc_hal_reg_caps,
+						      svc_rdy_ext->ext_hal_reg_caps, i,
+						      &reg_cap);
+		if (ret) {
+			ath11k_warn(soc, "failed to extract reg cap %d\n", i);
+			return ret;
+		}
+
+		memcpy(&soc->hal_reg_cap[reg_cap.phy_id],
+		       &reg_cap, sizeof(reg_cap));
+	}
+	return 0;
+}
+
+static int ath11k_wmi_tlv_ext_soc_hal_reg_caps_parse(struct ath11k_base *soc,
+						     u16 len, const void *ptr,
+						     void *data)
+{
+	struct ath11k_pdev_wmi *wmi_handle = &soc->wmi_sc.wmi[0];
+	struct wmi_tlv_svc_rdy_ext_parse *svc_rdy_ext = data;
+	u8 hw_mode_id = svc_rdy_ext->pref_hw_mode_caps.hw_mode_id;
+	u32 phy_id_map;
+	int ret;
+
+	svc_rdy_ext->soc_hal_reg_caps = (struct wmi_soc_hal_reg_capabilities *)ptr;
+	svc_rdy_ext->param.num_phy = svc_rdy_ext->soc_hal_reg_caps->num_phy;
+
+	soc->num_radios = 0;
+	phy_id_map = svc_rdy_ext->pref_hw_mode_caps.phy_id_map;
+
+	while (phy_id_map && soc->num_radios < MAX_RADIOS) {
+		ret = ath11k_pull_mac_phy_cap_svc_ready_ext(wmi_handle,
+							    svc_rdy_ext->hw_caps,
+							    svc_rdy_ext->hw_mode_caps,
+							    svc_rdy_ext->soc_hal_reg_caps,
+							    svc_rdy_ext->mac_phy_caps,
+							    hw_mode_id, soc->num_radios,
+							    &soc->pdevs[soc->num_radios]);
+		if (ret) {
+			ath11k_warn(soc, "failed to extract mac caps, idx :%d\n",
+				    soc->num_radios);
+			return ret;
+		}
+
+		soc->num_radios++;
+
+		/* TODO: mac_phy_cap prints */
+		phy_id_map >>= 1;
+	}
+	return 0;
+}
+
+static int ath11k_wmi_tlv_svc_rdy_ext_parse(struct ath11k_base *ab,
+					    u16 tag, u16 len,
+					    const void *ptr, void *data)
+{
+	struct ath11k_pdev_wmi *wmi_handle = &ab->wmi_sc.wmi[0];
+	struct wmi_tlv_svc_rdy_ext_parse *svc_rdy_ext = data;
+	int ret;
+
+	switch (tag) {
+	case WMI_TAG_SERVICE_READY_EXT_EVENT:
+		ret = ath11k_pull_svc_ready_ext(wmi_handle, ptr,
+						&svc_rdy_ext->param);
+		if (ret) {
+			ath11k_warn(ab, "unable to extract ext params\n");
+			return ret;
+		}
+		break;
+
+	case WMI_TAG_SOC_MAC_PHY_HW_MODE_CAPS:
+		svc_rdy_ext->hw_caps = (struct wmi_soc_mac_phy_hw_mode_caps *)ptr;
+		svc_rdy_ext->param.num_hw_modes = svc_rdy_ext->hw_caps->num_hw_modes;
+		break;
+
+	case WMI_TAG_SOC_HAL_REG_CAPABILITIES:
+		ret = ath11k_wmi_tlv_ext_soc_hal_reg_caps_parse(ab, len, ptr,
+								svc_rdy_ext);
+		if (ret)
+			return ret;
+		break;
+
+	case WMI_TAG_ARRAY_STRUCT:
+		if (!svc_rdy_ext->hw_mode_done) {
+			ret = ath11k_wmi_tlv_hw_mode_caps(ab, len, ptr,
+							  svc_rdy_ext);
+			if (ret)
+				return ret;
+
+			svc_rdy_ext->hw_mode_done = true;
+		} else if (!svc_rdy_ext->mac_phy_done) {
+			svc_rdy_ext->n_mac_phy_caps = 0;
+			svc_rdy_ext->mac_phy_caps =
+					(struct wmi_mac_phy_capabilities *)ptr;
+			ret = ath11k_wmi_tlv_iter(ab, ptr, len,
+						  ath11k_wmi_tlv_mac_phy_caps_parse,
+						  svc_rdy_ext);
+			if (ret) {
+				ath11k_warn(ab, "failed to parse tlv %d\n", ret);
+				return ret;
+			}
+
+			svc_rdy_ext->mac_phy_done = true;
+		} else if (!svc_rdy_ext->ext_hal_reg_done) {
+			ret = ath11k_wmi_tlv_ext_hal_reg_caps(ab, len, ptr,
+							      svc_rdy_ext);
+			if (ret)
+				return ret;
+
+			svc_rdy_ext->ext_hal_reg_done = true;
+			complete(&ab->wmi_sc.service_ready);
+		}
+		break;
+
+	default:
+		break;
+	}
+	return 0;
+}
+
+static int ath11k_service_ready_ext_event(struct ath11k_base *ab,
+					  struct sk_buff *skb)
+{
+	struct wmi_tlv_svc_rdy_ext_parse svc_rdy_ext = { };
+	int ret;
+
+	ret = ath11k_wmi_tlv_iter(ab, skb->data, skb->len,
+				  ath11k_wmi_tlv_svc_rdy_ext_parse,
+				  &svc_rdy_ext);
+	if (ret) {
+		ath11k_warn(ab, "failed to parse tlv %d\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int ath11k_pull_vdev_start_resp_tlv(struct ath11k_base *ab, struct sk_buff *skb,
+					   struct wmi_vdev_start_resp_event *vdev_rsp)
+{
+	const void **tb;
+	const struct wmi_vdev_start_resp_event *ev;
+	int ret;
+
+	tb = ath11k_wmi_tlv_parse_alloc(ab, skb->data, skb->len, GFP_ATOMIC);
+	if (IS_ERR(tb)) {
+		ret = PTR_ERR(tb);
+		ath11k_warn(ab, "failed to parse tlv: %d\n", ret);
+		return ret;
+	}
+
+	ev = tb[WMI_TAG_VDEV_START_RESPONSE_EVENT];
+	if (!ev) {
+		ath11k_warn(ab, "failed to fetch vdev start resp ev");
+		kfree(tb);
+		return -EPROTO;
+	}
+
+	memset(vdev_rsp, 0, sizeof(*vdev_rsp));
+
+	vdev_rsp->vdev_id = ev->vdev_id;
+	vdev_rsp->requestor_id = ev->requestor_id;
+	vdev_rsp->resp_type = ev->resp_type;
+	vdev_rsp->status = ev->status;
+	vdev_rsp->chain_mask = ev->chain_mask;
+	vdev_rsp->smps_mode = ev->smps_mode;
+	vdev_rsp->mac_id = ev->mac_id;
+	vdev_rsp->cfgd_tx_streams = ev->cfgd_tx_streams;
+	vdev_rsp->cfgd_rx_streams = ev->cfgd_rx_streams;
+
+	kfree(tb);
+	return 0;
+}
+
+static struct cur_reg_rule
+*create_reg_rules_from_wmi(u32 num_reg_rules,
+			   struct wmi_regulatory_rule_struct *wmi_reg_rule)
+{
+	struct cur_reg_rule *reg_rule_ptr;
+	u32 count;
+
+	reg_rule_ptr =  kzalloc((num_reg_rules * sizeof(*reg_rule_ptr)),
+				GFP_ATOMIC);
+
+	if (!reg_rule_ptr)
+		return NULL;
+
+	for (count = 0; count < num_reg_rules; count++) {
+		reg_rule_ptr[count].start_freq =
+			FIELD_GET(REG_RULE_START_FREQ,
+				  wmi_reg_rule[count].freq_info);
+		reg_rule_ptr[count].end_freq =
+			FIELD_GET(REG_RULE_END_FREQ,
+				  wmi_reg_rule[count].freq_info);
+		reg_rule_ptr[count].max_bw =
+			FIELD_GET(REG_RULE_MAX_BW,
+				  wmi_reg_rule[count].bw_pwr_info);
+		reg_rule_ptr[count].reg_power =
+			FIELD_GET(REG_RULE_REG_PWR,
+				  wmi_reg_rule[count].bw_pwr_info);
+		reg_rule_ptr[count].ant_gain =
+			FIELD_GET(REG_RULE_ANT_GAIN,
+				  wmi_reg_rule[count].bw_pwr_info);
+		reg_rule_ptr[count].flags =
+			FIELD_GET(REG_RULE_FLAGS,
+				  wmi_reg_rule[count].flag_info);
+	}
+
+	return reg_rule_ptr;
+}
+
+static int ath11k_pull_reg_chan_list_update_ev(struct ath11k_base *ab,
+					       struct sk_buff *skb,
+					       struct cur_regulatory_info *reg_info)
+{
+	const void **tb;
+	const struct wmi_reg_chan_list_cc_event *chan_list_event_hdr;
+	struct wmi_regulatory_rule_struct *wmi_reg_rule;
+	u32 num_2g_reg_rules, num_5g_reg_rules;
+	int ret;
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI, "processing regulatory channel list\n");
+
+	tb = ath11k_wmi_tlv_parse_alloc(ab, skb->data, skb->len, GFP_ATOMIC);
+	if (IS_ERR(tb)) {
+		ret = PTR_ERR(tb);
+		ath11k_warn(ab, "failed to parse tlv: %d\n", ret);
+		return ret;
+	}
+
+	chan_list_event_hdr = tb[WMI_TAG_REG_CHAN_LIST_CC_EVENT];
+	if (!chan_list_event_hdr) {
+		ath11k_warn(ab, "failed to fetch reg chan list update ev\n");
+		kfree(tb);
+		return -EPROTO;
+	}
+
+	reg_info->num_2g_reg_rules = chan_list_event_hdr->num_2g_reg_rules;
+	reg_info->num_5g_reg_rules = chan_list_event_hdr->num_5g_reg_rules;
+
+	if (!(reg_info->num_2g_reg_rules + reg_info->num_5g_reg_rules)) {
+		ath11k_warn(ab, "No regulatory rules available in the event info\n");
+		kfree(tb);
+		return -EINVAL;
+	}
+
+	memcpy(reg_info->alpha2, &chan_list_event_hdr->alpha2,
+	       REG_ALPHA2_LEN);
+	reg_info->dfs_region = chan_list_event_hdr->dfs_region;
+	reg_info->phybitmap = chan_list_event_hdr->phybitmap;
+	reg_info->num_phy = chan_list_event_hdr->num_phy;
+	reg_info->phy_id = chan_list_event_hdr->phy_id;
+	reg_info->ctry_code = chan_list_event_hdr->country_id;
+	reg_info->reg_dmn_pair = chan_list_event_hdr->domain_code;
+	if (chan_list_event_hdr->status_code == WMI_REG_SET_CC_STATUS_PASS)
+		reg_info->status_code = REG_SET_CC_STATUS_PASS;
+	else if (chan_list_event_hdr->status_code == WMI_REG_CURRENT_ALPHA2_NOT_FOUND)
+		reg_info->status_code = REG_CURRENT_ALPHA2_NOT_FOUND;
+	else if (chan_list_event_hdr->status_code == WMI_REG_INIT_ALPHA2_NOT_FOUND)
+		reg_info->status_code = REG_INIT_ALPHA2_NOT_FOUND;
+	else if (chan_list_event_hdr->status_code == WMI_REG_SET_CC_CHANGE_NOT_ALLOWED)
+		reg_info->status_code = REG_SET_CC_CHANGE_NOT_ALLOWED;
+	else if (chan_list_event_hdr->status_code == WMI_REG_SET_CC_STATUS_NO_MEMORY)
+		reg_info->status_code = REG_SET_CC_STATUS_NO_MEMORY;
+	else if (chan_list_event_hdr->status_code == WMI_REG_SET_CC_STATUS_FAIL)
+		reg_info->status_code = REG_SET_CC_STATUS_FAIL;
+
+	reg_info->min_bw_2g = chan_list_event_hdr->min_bw_2g;
+	reg_info->max_bw_2g = chan_list_event_hdr->max_bw_2g;
+	reg_info->min_bw_5g = chan_list_event_hdr->min_bw_5g;
+	reg_info->max_bw_5g = chan_list_event_hdr->max_bw_5g;
+
+	num_2g_reg_rules = reg_info->num_2g_reg_rules;
+	num_5g_reg_rules = reg_info->num_5g_reg_rules;
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI,
+		   "%s:cc %s dsf %d BW: min_2g %d max_2g %d min_5g %d max_5g %d",
+		   __func__, reg_info->alpha2, reg_info->dfs_region,
+		   reg_info->min_bw_2g, reg_info->max_bw_2g,
+		   reg_info->min_bw_5g, reg_info->max_bw_5g);
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI,
+		   "%s: num_2g_reg_rules %d num_5g_reg_rules %d", __func__,
+		   num_2g_reg_rules, num_5g_reg_rules);
+
+	wmi_reg_rule =
+		(struct wmi_regulatory_rule_struct *)((u8 *)chan_list_event_hdr
+						+ sizeof(*chan_list_event_hdr)
+						+ sizeof(struct wmi_tlv));
+
+	if (num_2g_reg_rules) {
+		reg_info->reg_rules_2g_ptr = create_reg_rules_from_wmi(num_2g_reg_rules,
+								       wmi_reg_rule);
+		if (!reg_info->reg_rules_2g_ptr) {
+			kfree(tb);
+			ath11k_warn(ab, "Unable to Allocate memory for 2g rules\n");
+			return -ENOMEM;
+		}
+	}
+
+	if (num_5g_reg_rules) {
+		wmi_reg_rule += num_2g_reg_rules;
+		reg_info->reg_rules_5g_ptr = create_reg_rules_from_wmi(num_5g_reg_rules,
+								       wmi_reg_rule);
+		if (!reg_info->reg_rules_5g_ptr) {
+			kfree(tb);
+			ath11k_warn(ab, "Unable to Allocate memory for 5g rules\n");
+			return -ENOMEM;
+		}
+	}
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI, "processed regulatory channel list\n");
+
+	kfree(tb);
+	return 0;
+}
+
+static int ath11k_pull_peer_del_resp_ev(struct ath11k_base *ab, struct sk_buff *skb,
+					struct wmi_peer_delete_resp_event *peer_del_resp)
+{
+	const void **tb;
+	const struct wmi_peer_delete_resp_event *ev;
+	int ret;
+
+	tb = ath11k_wmi_tlv_parse_alloc(ab, skb->data, skb->len, GFP_ATOMIC);
+	if (IS_ERR(tb)) {
+		ret = PTR_ERR(tb);
+		ath11k_warn(ab, "failed to parse tlv: %d\n", ret);
+		return ret;
+	}
+
+	ev = tb[WMI_TAG_PEER_DELETE_RESP_EVENT];
+	if (!ev) {
+		ath11k_warn(ab, "failed to fetch peer delete resp ev");
+		kfree(tb);
+		return -EPROTO;
+	}
+
+	memset(peer_del_resp, 0, sizeof(*peer_del_resp));
+
+	peer_del_resp->vdev_id = ev->vdev_id;
+	ether_addr_copy(peer_del_resp->peer_macaddr.addr,
+			ev->peer_macaddr.addr);
+
+	kfree(tb);
+	return 0;
+}
+
+static int ath11k_pull_bcn_tx_status_ev(struct ath11k_base *ab, void *evt_buf,
+					u32 len, u32 *vdev_id,
+					u32 *tx_status)
+{
+	const void **tb;
+	const struct wmi_bcn_tx_status_event *ev;
+	int ret;
+
+	tb = ath11k_wmi_tlv_parse_alloc(ab, evt_buf, len, GFP_ATOMIC);
+	if (IS_ERR(tb)) {
+		ret = PTR_ERR(tb);
+		ath11k_warn(ab, "failed to parse tlv: %d\n", ret);
+		return ret;
+	}
+
+	ev = tb[WMI_TAG_OFFLOAD_BCN_TX_STATUS_EVENT];
+	if (!ev) {
+		ath11k_warn(ab, "failed to fetch bcn tx status ev");
+		kfree(tb);
+		return -EPROTO;
+	}
+
+	*vdev_id   = ev->vdev_id;
+	*tx_status = ev->tx_status;
+
+	kfree(tb);
+	return 0;
+}
+
+static int ath11k_pull_vdev_stopped_param_tlv(struct ath11k_base *ab, struct sk_buff *skb,
+					      u32 *vdev_id)
+{
+	const void **tb;
+	const struct wmi_vdev_stopped_event *ev;
+	int ret;
+
+	tb = ath11k_wmi_tlv_parse_alloc(ab, skb->data, skb->len, GFP_ATOMIC);
+	if (IS_ERR(tb)) {
+		ret = PTR_ERR(tb);
+		ath11k_warn(ab, "failed to parse tlv: %d\n", ret);
+		return ret;
+	}
+
+	ev = tb[WMI_TAG_VDEV_STOPPED_EVENT];
+	if (!ev) {
+		ath11k_warn(ab, "failed to fetch vdev stop ev");
+		kfree(tb);
+		return -EPROTO;
+	}
+
+	*vdev_id =  ev->vdev_id;
+
+	kfree(tb);
+	return 0;
+}
+
+static int ath11k_pull_mgmt_rx_params_tlv(struct ath11k_base *ab,
+					  struct sk_buff *skb,
+					  struct mgmt_rx_event_params *hdr)
+{
+	const void **tb;
+	const struct wmi_mgmt_rx_hdr *ev;
+	const u8 *frame;
+	int ret;
+
+	tb = ath11k_wmi_tlv_parse_alloc(ab, skb->data, skb->len, GFP_ATOMIC);
+	if (IS_ERR(tb)) {
+		ret = PTR_ERR(tb);
+		ath11k_warn(ab, "failed to parse tlv: %d\n", ret);
+		return ret;
+	}
+
+	ev = tb[WMI_TAG_MGMT_RX_HDR];
+	frame = tb[WMI_TAG_ARRAY_BYTE];
+
+	if (!ev || !frame) {
+		ath11k_warn(ab, "failed to fetch mgmt rx hdr");
+		kfree(tb);
+		return -EPROTO;
+	}
+
+	hdr->pdev_id =  ev->pdev_id;
+	hdr->channel =  ev->channel;
+	hdr->snr =  ev->snr;
+	hdr->rate =  ev->rate;
+	hdr->phy_mode =  ev->phy_mode;
+	hdr->buf_len =  ev->buf_len;
+	hdr->status =  ev->status;
+	hdr->flags =  ev->flags;
+	hdr->rssi =  ev->rssi;
+	hdr->tsf_delta =  ev->tsf_delta;
+	memcpy(hdr->rssi_ctl, ev->rssi_ctl, sizeof(hdr->rssi_ctl));
+
+	if (skb->len < (frame - skb->data) + hdr->buf_len) {
+		ath11k_warn(ab, "invalid length in mgmt rx hdr ev");
+		kfree(tb);
+		return -EPROTO;
+	}
+
+	/* shift the sk_buff to point to `frame` */
+	skb_trim(skb, 0);
+	skb_put(skb, frame - skb->data);
+	skb_pull(skb, frame - skb->data);
+	skb_put(skb, hdr->buf_len);
+
+	ath11k_ce_byte_swap(skb->data, hdr->buf_len);
+
+	kfree(tb);
+	return 0;
+}
+
+static int wmi_process_mgmt_tx_comp(struct ath11k *ar, u32 desc_id,
+				    u32 status)
+{
+	struct sk_buff *msdu;
+	struct ieee80211_tx_info *info;
+	struct ath11k_skb_cb *skb_cb;
+
+	spin_lock_bh(&ar->txmgmt_idr_lock);
+	msdu = idr_find(&ar->txmgmt_idr, desc_id);
+
+	if (!msdu) {
+		ath11k_warn(ar->ab, "received mgmt tx compl for invalid msdu_id: %d\n",
+			    desc_id);
+		spin_unlock_bh(&ar->txmgmt_idr_lock);
+		return -ENOENT;
+	}
+
+	idr_remove(&ar->txmgmt_idr, desc_id);
+	spin_unlock_bh(&ar->txmgmt_idr_lock);
+
+	skb_cb = ATH11K_SKB_CB(msdu);
+	dma_unmap_single(ar->ab->dev, skb_cb->paddr, msdu->len, DMA_TO_DEVICE);
+
+	info = IEEE80211_SKB_CB(msdu);
+	if ((!(info->flags & IEEE80211_TX_CTL_NO_ACK)) && !status)
+		info->flags |= IEEE80211_TX_STAT_ACK;
+
+	ieee80211_tx_status_irqsafe(ar->hw, msdu);
+
+	WARN_ON_ONCE(atomic_read(&ar->num_pending_mgmt_tx) == 0);
+	atomic_dec(&ar->num_pending_mgmt_tx);
+
+	return 0;
+}
+
+static int ath11k_pull_mgmt_tx_compl_param_tlv(struct ath11k_base *ab,
+					       struct sk_buff *skb,
+					       struct wmi_mgmt_tx_compl_event *param)
+{
+	const void **tb;
+	const struct wmi_mgmt_tx_compl_event *ev;
+	int ret;
+
+	tb = ath11k_wmi_tlv_parse_alloc(ab, skb->data, skb->len, GFP_ATOMIC);
+	if (IS_ERR(tb)) {
+		ret = PTR_ERR(tb);
+		ath11k_warn(ab, "failed to parse tlv: %d\n", ret);
+		return ret;
+	}
+
+	ev = tb[WMI_TAG_MGMT_TX_COMPL_EVENT];
+	if (!ev) {
+		ath11k_warn(ab, "failed to fetch mgmt tx compl ev");
+		kfree(tb);
+		return -EPROTO;
+	}
+
+	param->pdev_id = ev->pdev_id;
+	param->desc_id = ev->desc_id;
+	param->status = ev->status;
+
+	kfree(tb);
+	return 0;
+}
+
+static void ath11k_wmi_event_scan_started(struct ath11k *ar)
+{
+	lockdep_assert_held(&ar->data_lock);
+
+	switch (ar->scan.state) {
+	case ATH11K_SCAN_IDLE:
+	case ATH11K_SCAN_RUNNING:
+	case ATH11K_SCAN_ABORTING:
+		ath11k_warn(ar->ab, "received scan started event in an invalid scan state: %s (%d)\n",
+			    ath11k_scan_state_str(ar->scan.state),
+			    ar->scan.state);
+		break;
+	case ATH11K_SCAN_STARTING:
+		ar->scan.state = ATH11K_SCAN_RUNNING;
+		complete(&ar->scan.started);
+		break;
+	}
+}
+
+static void ath11k_wmi_event_scan_start_failed(struct ath11k *ar)
+{
+	lockdep_assert_held(&ar->data_lock);
+
+	switch (ar->scan.state) {
+	case ATH11K_SCAN_IDLE:
+	case ATH11K_SCAN_RUNNING:
+	case ATH11K_SCAN_ABORTING:
+		ath11k_warn(ar->ab, "received scan start failed event in an invalid scan state: %s (%d)\n",
+			    ath11k_scan_state_str(ar->scan.state),
+			    ar->scan.state);
+		break;
+	case ATH11K_SCAN_STARTING:
+		complete(&ar->scan.started);
+		__ath11k_mac_scan_finish(ar);
+		break;
+	}
+}
+
+static void ath11k_wmi_event_scan_completed(struct ath11k *ar)
+{
+	lockdep_assert_held(&ar->data_lock);
+
+	switch (ar->scan.state) {
+	case ATH11K_SCAN_IDLE:
+	case ATH11K_SCAN_STARTING:
+		/* One suspected reason scan can be completed while starting is
+		 * if firmware fails to deliver all scan events to the host,
+		 * e.g. when transport pipe is full. This has been observed
+		 * with spectral scan phyerr events starving wmi transport
+		 * pipe. In such case the "scan completed" event should be (and
+		 * is) ignored by the host as it may be just firmware's scan
+		 * state machine recovering.
+		 */
+		ath11k_warn(ar->ab, "received scan completed event in an invalid scan state: %s (%d)\n",
+			    ath11k_scan_state_str(ar->scan.state),
+			    ar->scan.state);
+		break;
+	case ATH11K_SCAN_RUNNING:
+	case ATH11K_SCAN_ABORTING:
+		__ath11k_mac_scan_finish(ar);
+		break;
+	}
+}
+
+static void ath11k_wmi_event_scan_bss_chan(struct ath11k *ar)
+{
+	lockdep_assert_held(&ar->data_lock);
+
+	switch (ar->scan.state) {
+	case ATH11K_SCAN_IDLE:
+	case ATH11K_SCAN_STARTING:
+		ath11k_warn(ar->ab, "received scan bss chan event in an invalid scan state: %s (%d)\n",
+			    ath11k_scan_state_str(ar->scan.state),
+			    ar->scan.state);
+		break;
+	case ATH11K_SCAN_RUNNING:
+	case ATH11K_SCAN_ABORTING:
+		ar->scan_channel = NULL;
+		break;
+	}
+}
+
+static void ath11k_wmi_event_scan_foreign_chan(struct ath11k *ar, u32 freq)
+{
+	lockdep_assert_held(&ar->data_lock);
+
+	switch (ar->scan.state) {
+	case ATH11K_SCAN_IDLE:
+	case ATH11K_SCAN_STARTING:
+		ath11k_warn(ar->ab, "received scan foreign chan event in an invalid scan state: %s (%d)\n",
+			    ath11k_scan_state_str(ar->scan.state),
+			    ar->scan.state);
+		break;
+	case ATH11K_SCAN_RUNNING:
+	case ATH11K_SCAN_ABORTING:
+		ar->scan_channel = ieee80211_get_channel(ar->hw->wiphy, freq);
+		break;
+	}
+}
+
+static const char *
+ath11k_wmi_event_scan_type_str(enum wmi_scan_event_type type,
+			       enum wmi_scan_completion_reason reason)
+{
+	switch (type) {
+	case WMI_SCAN_EVENT_STARTED:
+		return "started";
+	case WMI_SCAN_EVENT_COMPLETED:
+		switch (reason) {
+		case WMI_SCAN_REASON_COMPLETED:
+			return "completed";
+		case WMI_SCAN_REASON_CANCELLED:
+			return "completed [cancelled]";
+		case WMI_SCAN_REASON_PREEMPTED:
+			return "completed [preempted]";
+		case WMI_SCAN_REASON_TIMEDOUT:
+			return "completed [timedout]";
+		case WMI_SCAN_REASON_INTERNAL_FAILURE:
+			return "completed [internal err]";
+		case WMI_SCAN_REASON_MAX:
+			break;
+		}
+		return "completed [unknown]";
+	case WMI_SCAN_EVENT_BSS_CHANNEL:
+		return "bss channel";
+	case WMI_SCAN_EVENT_FOREIGN_CHAN:
+		return "foreign channel";
+	case WMI_SCAN_EVENT_DEQUEUED:
+		return "dequeued";
+	case WMI_SCAN_EVENT_PREEMPTED:
+		return "preempted";
+	case WMI_SCAN_EVENT_START_FAILED:
+		return "start failed";
+	case WMI_SCAN_EVENT_RESTARTED:
+		return "restarted";
+	case WMI_SCAN_EVENT_FOREIGN_CHAN_EXIT:
+		return "foreign channel exit";
+	default:
+		return "unknown";
+	}
+}
+
+static int ath11k_pull_scan_ev(struct ath11k_base *ab, struct sk_buff *skb,
+			       struct wmi_scan_event *scan_evt_param)
+{
+	const void **tb;
+	const struct wmi_scan_event *ev;
+	int ret;
+
+	tb = ath11k_wmi_tlv_parse_alloc(ab, skb->data, skb->len, GFP_ATOMIC);
+	if (IS_ERR(tb)) {
+		ret = PTR_ERR(tb);
+		ath11k_warn(ab, "failed to parse tlv: %d\n", ret);
+		return ret;
+	}
+
+	ev = tb[WMI_TAG_SCAN_EVENT];
+	if (!ev) {
+		ath11k_warn(ab, "failed to fetch scan ev");
+		kfree(tb);
+		return -EPROTO;
+	}
+
+	scan_evt_param->event_type = ev->event_type;
+	scan_evt_param->reason = ev->reason;
+	scan_evt_param->channel_freq = ev->channel_freq;
+	scan_evt_param->scan_req_id = ev->scan_req_id;
+	scan_evt_param->scan_id = ev->scan_id;
+	scan_evt_param->vdev_id = ev->vdev_id;
+	scan_evt_param->tsf_timestamp = ev->tsf_timestamp;
+
+	kfree(tb);
+	return 0;
+}
+
+static int ath11k_pull_peer_sta_kickout_ev(struct ath11k_base *ab, struct sk_buff *skb,
+					   struct wmi_peer_sta_kickout_arg *arg)
+{
+	const void **tb;
+	const struct wmi_peer_sta_kickout_event *ev;
+	int ret;
+
+	tb = ath11k_wmi_tlv_parse_alloc(ab, skb->data, skb->len, GFP_ATOMIC);
+	if (IS_ERR(tb)) {
+		ret = PTR_ERR(tb);
+		ath11k_warn(ab, "failed to parse tlv: %d\n", ret);
+		return ret;
+	}
+
+	ev = tb[WMI_TAG_PEER_STA_KICKOUT_EVENT];
+	if (!ev) {
+		ath11k_warn(ab, "failed to fetch peer sta kickout ev");
+		kfree(tb);
+		return -EPROTO;
+	}
+
+	arg->mac_addr = ev->peer_macaddr.addr;
+
+	kfree(tb);
+	return 0;
+}
+
+static int ath11k_pull_roam_ev(struct ath11k_base *ab, struct sk_buff *skb,
+			       struct wmi_roam_event *roam_ev)
+{
+	const void **tb;
+	const struct wmi_roam_event *ev;
+	int ret;
+
+	tb = ath11k_wmi_tlv_parse_alloc(ab, skb->data, skb->len, GFP_ATOMIC);
+	if (IS_ERR(tb)) {
+		ret = PTR_ERR(tb);
+		ath11k_warn(ab, "failed to parse tlv: %d\n", ret);
+		return ret;
+	}
+
+	ev = tb[WMI_TAG_ROAM_EVENT];
+	if (!ev) {
+		ath11k_warn(ab, "failed to fetch roam ev");
+		kfree(tb);
+		return -EPROTO;
+	}
+
+	roam_ev->vdev_id = ev->vdev_id;
+	roam_ev->reason = ev->reason;
+	roam_ev->rssi = ev->rssi;
+
+	kfree(tb);
+	return 0;
+}
+
+static int freq_to_idx(struct ath11k *ar, int freq)
+{
+	struct ieee80211_supported_band *sband;
+	int band, ch, idx = 0;
+
+	for (band = NL80211_BAND_2GHZ; band < NUM_NL80211_BANDS; band++) {
+		sband = ar->hw->wiphy->bands[band];
+		if (!sband)
+			continue;
+
+		for (ch = 0; ch < sband->n_channels; ch++, idx++)
+			if (sband->channels[ch].center_freq == freq)
+				goto exit;
+	}
+
+exit:
+	return idx;
+}
+
+static int ath11k_pull_chan_info_ev(struct ath11k_base *ab, u8 *evt_buf,
+				    u32 len, struct wmi_chan_info_event *ch_info_ev)
+{
+	const void **tb;
+	const struct wmi_chan_info_event *ev;
+	int ret;
+
+	tb = ath11k_wmi_tlv_parse_alloc(ab, evt_buf, len, GFP_ATOMIC);
+	if (IS_ERR(tb)) {
+		ret = PTR_ERR(tb);
+		ath11k_warn(ab, "failed to parse tlv: %d\n", ret);
+		return ret;
+	}
+
+	ev = tb[WMI_TAG_CHAN_INFO_EVENT];
+	if (!ev) {
+		ath11k_warn(ab, "failed to fetch chan info ev");
+		kfree(tb);
+		return -EPROTO;
+	}
+
+	ch_info_ev->err_code = ev->err_code;
+	ch_info_ev->freq = ev->freq;
+	ch_info_ev->cmd_flags = ev->cmd_flags;
+	ch_info_ev->noise_floor = ev->noise_floor;
+	ch_info_ev->rx_clear_count = ev->rx_clear_count;
+	ch_info_ev->cycle_count = ev->cycle_count;
+	ch_info_ev->chan_tx_pwr_range = ev->chan_tx_pwr_range;
+	ch_info_ev->chan_tx_pwr_tp = ev->chan_tx_pwr_tp;
+	ch_info_ev->rx_frame_count = ev->rx_frame_count;
+	ch_info_ev->tx_frame_cnt = ev->tx_frame_cnt;
+	ch_info_ev->mac_clk_mhz = ev->mac_clk_mhz;
+	ch_info_ev->vdev_id = ev->vdev_id;
+
+	kfree(tb);
+	return 0;
+}
+
+static int
+ath11k_pull_pdev_bss_chan_info_ev(struct ath11k_base *ab, struct sk_buff *skb,
+				  struct wmi_pdev_bss_chan_info_event *bss_ch_info_ev)
+{
+	const void **tb;
+	const struct wmi_pdev_bss_chan_info_event *ev;
+	int ret;
+
+	tb = ath11k_wmi_tlv_parse_alloc(ab, skb->data, skb->len, GFP_ATOMIC);
+	if (IS_ERR(tb)) {
+		ret = PTR_ERR(tb);
+		ath11k_warn(ab, "failed to parse tlv: %d\n", ret);
+		return ret;
+	}
+
+	ev = tb[WMI_TAG_PDEV_BSS_CHAN_INFO_EVENT];
+	if (!ev) {
+		ath11k_warn(ab, "failed to fetch pdev bss chan info ev");
+		kfree(tb);
+		return -EPROTO;
+	}
+
+	bss_ch_info_ev->pdev_id = ev->pdev_id;
+	bss_ch_info_ev->freq = ev->freq;
+	bss_ch_info_ev->noise_floor = ev->noise_floor;
+	bss_ch_info_ev->rx_clear_count_low = ev->rx_clear_count_low;
+	bss_ch_info_ev->rx_clear_count_high = ev->rx_clear_count_high;
+	bss_ch_info_ev->cycle_count_low = ev->cycle_count_low;
+	bss_ch_info_ev->cycle_count_high = ev->cycle_count_high;
+	bss_ch_info_ev->tx_cycle_count_low = ev->tx_cycle_count_low;
+	bss_ch_info_ev->tx_cycle_count_high = ev->tx_cycle_count_high;
+	bss_ch_info_ev->rx_cycle_count_low = ev->rx_cycle_count_low;
+	bss_ch_info_ev->rx_cycle_count_high = ev->rx_cycle_count_high;
+	bss_ch_info_ev->rx_bss_cycle_count_low = ev->rx_bss_cycle_count_low;
+	bss_ch_info_ev->rx_bss_cycle_count_high = ev->rx_bss_cycle_count_high;
+
+	kfree(tb);
+	return 0;
+}
+
+static int
+ath11k_pull_vdev_install_key_compl_ev(struct ath11k_base *ab, struct sk_buff *skb,
+				      struct wmi_vdev_install_key_complete_arg *arg)
+{
+	const void **tb;
+	const struct wmi_vdev_install_key_compl_event *ev;
+	int ret;
+
+	tb = ath11k_wmi_tlv_parse_alloc(ab, skb->data, skb->len, GFP_ATOMIC);
+	if (IS_ERR(tb)) {
+		ret = PTR_ERR(tb);
+		ath11k_warn(ab, "failed to parse tlv: %d\n", ret);
+		return ret;
+	}
+
+	ev = tb[WMI_TAG_VDEV_INSTALL_KEY_COMPLETE_EVENT];
+	if (!ev) {
+		ath11k_warn(ab, "failed to fetch vdev install key compl ev");
+		kfree(tb);
+		return -EPROTO;
+	}
+
+	arg->vdev_id = ev->vdev_id;
+	arg->macaddr = ev->peer_macaddr.addr;
+	arg->key_idx = ev->key_idx;
+	arg->key_flags = ev->key_flags;
+	arg->status = ev->status;
+
+	kfree(tb);
+	return 0;
+}
+
+static int ath11k_pull_peer_assoc_conf_ev(struct ath11k_base *ab, struct sk_buff *skb,
+					  struct wmi_peer_assoc_conf_arg *peer_assoc_conf)
+{
+	const void **tb;
+	const struct wmi_peer_assoc_conf_event *ev;
+	int ret;
+
+	tb = ath11k_wmi_tlv_parse_alloc(ab, skb->data, skb->len, GFP_ATOMIC);
+	if (IS_ERR(tb)) {
+		ret = PTR_ERR(tb);
+		ath11k_warn(ab, "failed to parse tlv: %d\n", ret);
+		return ret;
+	}
+
+	ev = tb[WMI_TAG_PEER_ASSOC_CONF_EVENT];
+	if (!ev) {
+		ath11k_warn(ab, "failed to fetch peer assoc conf ev");
+		kfree(tb);
+		return -EPROTO;
+	}
+
+	peer_assoc_conf->vdev_id = ev->vdev_id;
+	peer_assoc_conf->macaddr = ev->peer_macaddr.addr;
+
+	kfree(tb);
+	return 0;
+}
+
+static void ath11k_wmi_pull_pdev_stats_base(const struct wmi_pdev_stats_base *src,
+					    struct ath11k_fw_stats_pdev *dst)
+{
+	dst->ch_noise_floor = src->chan_nf;
+	dst->tx_frame_count = src->tx_frame_count;
+	dst->rx_frame_count = src->rx_frame_count;
+	dst->rx_clear_count = src->rx_clear_count;
+	dst->cycle_count = src->cycle_count;
+	dst->phy_err_count = src->phy_err_count;
+	dst->chan_tx_power = src->chan_tx_pwr;
+}
+
+static void
+ath11k_wmi_pull_pdev_stats_tx(const struct wmi_pdev_stats_tx *src,
+			      struct ath11k_fw_stats_pdev *dst)
+{
+	dst->comp_queued = src->comp_queued;
+	dst->comp_delivered = src->comp_delivered;
+	dst->msdu_enqued = src->msdu_enqued;
+	dst->mpdu_enqued = src->mpdu_enqued;
+	dst->wmm_drop = src->wmm_drop;
+	dst->local_enqued = src->local_enqued;
+	dst->local_freed = src->local_freed;
+	dst->hw_queued = src->hw_queued;
+	dst->hw_reaped = src->hw_reaped;
+	dst->underrun = src->underrun;
+	dst->tx_abort = src->tx_abort;
+	dst->mpdus_requed = src->mpdus_requed;
+	dst->tx_ko = src->tx_ko;
+	dst->data_rc = src->data_rc;
+	dst->self_triggers = src->self_triggers;
+	dst->sw_retry_failure = src->sw_retry_failure;
+	dst->illgl_rate_phy_err = src->illgl_rate_phy_err;
+	dst->pdev_cont_xretry = src->pdev_cont_xretry;
+	dst->pdev_tx_timeout = src->pdev_tx_timeout;
+	dst->pdev_resets = src->pdev_resets;
+	dst->stateless_tid_alloc_failure = src->stateless_tid_alloc_failure;
+	dst->phy_underrun = src->phy_underrun;
+	dst->txop_ovf = src->txop_ovf;
+}
+
+static void ath11k_wmi_pull_pdev_stats_rx(const struct wmi_pdev_stats_rx *src,
+					  struct ath11k_fw_stats_pdev *dst)
+{
+	dst->mid_ppdu_route_change = src->mid_ppdu_route_change;
+	dst->status_rcvd = src->status_rcvd;
+	dst->r0_frags = src->r0_frags;
+	dst->r1_frags = src->r1_frags;
+	dst->r2_frags = src->r2_frags;
+	dst->r3_frags = src->r3_frags;
+	dst->htt_msdus = src->htt_msdus;
+	dst->htt_mpdus = src->htt_mpdus;
+	dst->loc_msdus = src->loc_msdus;
+	dst->loc_mpdus = src->loc_mpdus;
+	dst->oversize_amsdu = src->oversize_amsdu;
+	dst->phy_errs = src->phy_errs;
+	dst->phy_err_drop = src->phy_err_drop;
+	dst->mpdu_errs = src->mpdu_errs;
+}
+
+static void
+ath11k_wmi_pull_vdev_stats(const struct wmi_vdev_stats *src,
+			   struct ath11k_fw_stats_vdev *dst)
+{
+	int i;
+
+	dst->vdev_id = src->vdev_id;
+	dst->beacon_snr = src->beacon_snr;
+	dst->data_snr = src->data_snr;
+	dst->num_rx_frames = src->num_rx_frames;
+	dst->num_rts_fail = src->num_rts_fail;
+	dst->num_rts_success = src->num_rts_success;
+	dst->num_rx_err = src->num_rx_err;
+	dst->num_rx_discard = src->num_rx_discard;
+	dst->num_tx_not_acked = src->num_tx_not_acked;
+
+	for (i = 0; i < ARRAY_SIZE(src->num_tx_frames); i++)
+		dst->num_tx_frames[i] = src->num_tx_frames[i];
+
+	for (i = 0; i < ARRAY_SIZE(src->num_tx_frames_retries); i++)
+		dst->num_tx_frames_retries[i] = src->num_tx_frames_retries[i];
+
+	for (i = 0; i < ARRAY_SIZE(src->num_tx_frames_failures); i++)
+		dst->num_tx_frames_failures[i] = src->num_tx_frames_failures[i];
+
+	for (i = 0; i < ARRAY_SIZE(src->tx_rate_history); i++)
+		dst->tx_rate_history[i] = src->tx_rate_history[i];
+
+	for (i = 0; i < ARRAY_SIZE(src->beacon_rssi_history); i++)
+		dst->beacon_rssi_history[i] = src->beacon_rssi_history[i];
+}
+
+static void
+ath11k_wmi_pull_bcn_stats(const struct wmi_bcn_stats *src,
+			  struct ath11k_fw_stats_bcn *dst)
+{
+	dst->vdev_id = src->vdev_id;
+	dst->tx_bcn_succ_cnt = src->tx_bcn_succ_cnt;
+	dst->tx_bcn_outage_cnt = src->tx_bcn_outage_cnt;
+}
+
+int ath11k_wmi_pull_fw_stats(struct ath11k_base *ab, struct sk_buff *skb,
+			     struct ath11k_fw_stats *stats)
+{
+	const void **tb;
+	const struct wmi_stats_event *ev;
+	const void *data;
+	int i, ret;
+	u32 len = skb->len;
+
+	tb = ath11k_wmi_tlv_parse_alloc(ab, skb->data, len, GFP_ATOMIC);
+	if (IS_ERR(tb)) {
+		ret = PTR_ERR(tb);
+		ath11k_warn(ab, "failed to parse tlv: %d\n", ret);
+		return ret;
+	}
+
+	ev = tb[WMI_TAG_STATS_EVENT];
+	data = tb[WMI_TAG_ARRAY_BYTE];
+	if (!ev || !data) {
+		ath11k_warn(ab, "failed to fetch update stats ev");
+		kfree(tb);
+		return -EPROTO;
+	}
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI,
+		   "wmi stats update ev pdev_id %d pdev %i vdev %i bcn %i\n",
+		   ev->pdev_id,
+		   ev->num_pdev_stats, ev->num_vdev_stats,
+		   ev->num_bcn_stats);
+
+	stats->pdev_id = ev->pdev_id;
+	stats->stats_id = 0;
+
+	for (i = 0; i < ev->num_pdev_stats; i++) {
+		const struct wmi_pdev_stats *src;
+		struct ath11k_fw_stats_pdev *dst;
+
+		src = data;
+		if (len < sizeof(*src)) {
+			kfree(tb);
+			return -EPROTO;
+		}
+
+		stats->stats_id = WMI_REQUEST_PDEV_STAT;
+
+		data += sizeof(*src);
+		len -= sizeof(*src);
+
+		dst = kzalloc(sizeof(*dst), GFP_ATOMIC);
+		if (!dst)
+			continue;
+
+		ath11k_wmi_pull_pdev_stats_base(&src->base, dst);
+		ath11k_wmi_pull_pdev_stats_tx(&src->tx, dst);
+		ath11k_wmi_pull_pdev_stats_rx(&src->rx, dst);
+		list_add_tail(&dst->list, &stats->pdevs);
+	}
+
+	for (i = 0; i < ev->num_vdev_stats; i++) {
+		const struct wmi_vdev_stats *src;
+		struct ath11k_fw_stats_vdev *dst;
+
+		src = data;
+		if (len < sizeof(*src)) {
+			kfree(tb);
+			return -EPROTO;
+		}
+
+		stats->stats_id = WMI_REQUEST_VDEV_STAT;
+
+		data += sizeof(*src);
+		len -= sizeof(*src);
+
+		dst = kzalloc(sizeof(*dst), GFP_ATOMIC);
+		if (!dst)
+			continue;
+
+		ath11k_wmi_pull_vdev_stats(src, dst);
+		list_add_tail(&dst->list, &stats->vdevs);
+	}
+
+	for (i = 0; i < ev->num_bcn_stats; i++) {
+		const struct wmi_bcn_stats *src;
+		struct ath11k_fw_stats_bcn *dst;
+
+		src = data;
+		if (len < sizeof(*src)) {
+			kfree(tb);
+			return -EPROTO;
+		}
+
+		stats->stats_id = WMI_REQUEST_BCN_STAT;
+
+		data += sizeof(*src);
+		len -= sizeof(*src);
+
+		dst = kzalloc(sizeof(*dst), GFP_ATOMIC);
+		if (!dst)
+			continue;
+
+		ath11k_wmi_pull_bcn_stats(src, dst);
+		list_add_tail(&dst->list, &stats->bcn);
+	}
+
+	kfree(tb);
+	return 0;
+}
+
+size_t ath11k_wmi_fw_stats_num_vdevs(struct list_head *head)
+{
+	struct ath11k_fw_stats_vdev *i;
+	size_t num = 0;
+
+	list_for_each_entry(i, head, list)
+		++num;
+
+	return num;
+}
+
+static size_t ath11k_wmi_fw_stats_num_bcn(struct list_head *head)
+{
+	struct ath11k_fw_stats_bcn *i;
+	size_t num = 0;
+
+	list_for_each_entry(i, head, list)
+		++num;
+
+	return num;
+}
+
+static void
+ath11k_wmi_fw_pdev_base_stats_fill(const struct ath11k_fw_stats_pdev *pdev,
+				   char *buf, u32 *length)
+{
+	u32 len = *length;
+	u32 buf_len = ATH11K_FW_STATS_BUF_SIZE;
+
+	len += scnprintf(buf + len, buf_len - len, "\n");
+	len += scnprintf(buf + len, buf_len - len, "%30s\n",
+			"ath11k PDEV stats");
+	len += scnprintf(buf + len, buf_len - len, "%30s\n\n",
+			"=================");
+
+	len += scnprintf(buf + len, buf_len - len, "%30s %10d\n",
+			"Channel noise floor", pdev->ch_noise_floor);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10u\n",
+			"Channel TX power", pdev->chan_tx_power);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10u\n",
+			"TX frame count", pdev->tx_frame_count);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10u\n",
+			"RX frame count", pdev->rx_frame_count);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10u\n",
+			"RX clear count", pdev->rx_clear_count);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10u\n",
+			"Cycle count", pdev->cycle_count);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10u\n",
+			"PHY error count", pdev->phy_err_count);
+
+	*length = len;
+}
+
+static void
+ath11k_wmi_fw_pdev_tx_stats_fill(const struct ath11k_fw_stats_pdev *pdev,
+				 char *buf, u32 *length)
+{
+	u32 len = *length;
+	u32 buf_len = ATH11K_FW_STATS_BUF_SIZE;
+
+	len += scnprintf(buf + len, buf_len - len, "\n%30s\n",
+			 "ath11k PDEV TX stats");
+	len += scnprintf(buf + len, buf_len - len, "%30s\n\n",
+			 "====================");
+
+	len += scnprintf(buf + len, buf_len - len, "%30s %10d\n",
+			 "HTT cookies queued", pdev->comp_queued);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10d\n",
+			 "HTT cookies disp.", pdev->comp_delivered);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10d\n",
+			 "MSDU queued", pdev->msdu_enqued);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10d\n",
+			 "MPDU queued", pdev->mpdu_enqued);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10d\n",
+			 "MSDUs dropped", pdev->wmm_drop);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10d\n",
+			 "Local enqued", pdev->local_enqued);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10d\n",
+			 "Local freed", pdev->local_freed);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10d\n",
+			 "HW queued", pdev->hw_queued);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10d\n",
+			 "PPDUs reaped", pdev->hw_reaped);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10d\n",
+			 "Num underruns", pdev->underrun);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10d\n",
+			 "PPDUs cleaned", pdev->tx_abort);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10d\n",
+			 "MPDUs requed", pdev->mpdus_requed);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10u\n",
+			 "Excessive retries", pdev->tx_ko);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10u\n",
+			 "HW rate", pdev->data_rc);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10u\n",
+			 "Sched self triggers", pdev->self_triggers);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10u\n",
+			 "Dropped due to SW retries",
+			 pdev->sw_retry_failure);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10u\n",
+			 "Illegal rate phy errors",
+			 pdev->illgl_rate_phy_err);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10u\n",
+			 "PDEV continuous xretry", pdev->pdev_cont_xretry);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10u\n",
+			 "TX timeout", pdev->pdev_tx_timeout);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10u\n",
+			 "PDEV resets", pdev->pdev_resets);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10u\n",
+			 "Stateless TIDs alloc failures",
+			 pdev->stateless_tid_alloc_failure);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10u\n",
+			 "PHY underrun", pdev->phy_underrun);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10u\n",
+			 "MPDU is more than txop limit", pdev->txop_ovf);
+	*length = len;
+}
+
+static void
+ath11k_wmi_fw_pdev_rx_stats_fill(const struct ath11k_fw_stats_pdev *pdev,
+				 char *buf, u32 *length)
+{
+	u32 len = *length;
+	u32 buf_len = ATH11K_FW_STATS_BUF_SIZE;
+
+	len += scnprintf(buf + len, buf_len - len, "\n%30s\n",
+			 "ath11k PDEV RX stats");
+	len += scnprintf(buf + len, buf_len - len, "%30s\n\n",
+			 "====================");
+
+	len += scnprintf(buf + len, buf_len - len, "%30s %10d\n",
+			 "Mid PPDU route change",
+			 pdev->mid_ppdu_route_change);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10d\n",
+			 "Tot. number of statuses", pdev->status_rcvd);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10d\n",
+			 "Extra frags on rings 0", pdev->r0_frags);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10d\n",
+			 "Extra frags on rings 1", pdev->r1_frags);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10d\n",
+			 "Extra frags on rings 2", pdev->r2_frags);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10d\n",
+			 "Extra frags on rings 3", pdev->r3_frags);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10d\n",
+			 "MSDUs delivered to HTT", pdev->htt_msdus);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10d\n",
+			 "MPDUs delivered to HTT", pdev->htt_mpdus);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10d\n",
+			 "MSDUs delivered to stack", pdev->loc_msdus);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10d\n",
+			 "MPDUs delivered to stack", pdev->loc_mpdus);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10d\n",
+			 "Oversized AMSUs", pdev->oversize_amsdu);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10d\n",
+			 "PHY errors", pdev->phy_errs);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10d\n",
+			 "PHY errors drops", pdev->phy_err_drop);
+	len += scnprintf(buf + len, buf_len - len, "%30s %10d\n",
+			 "MPDU errors (FCS, MIC, ENC)", pdev->mpdu_errs);
+	*length = len;
+}
+
+static void
+ath11k_wmi_fw_vdev_stats_fill(struct ath11k *ar,
+			      const struct ath11k_fw_stats_vdev *vdev,
+			      char *buf, u32 *length)
+{
+	u32 len = *length;
+	u32 buf_len = ATH11K_FW_STATS_BUF_SIZE;
+	struct ath11k_vif *arvif = ath11k_mac_get_arvif(ar, vdev->vdev_id);
+	u8 *vif_macaddr;
+	int i;
+
+	/* VDEV stats has all the active VDEVs of other PDEVs as well,
+	 * ignoring those not part of requested PDEV
+	 */
+	if (!arvif)
+		return;
+
+	vif_macaddr = arvif->vif->addr;
+
+	len += scnprintf(buf + len, buf_len - len, "%30s %u\n",
+			 "VDEV ID", vdev->vdev_id);
+	len += scnprintf(buf + len, buf_len - len, "%30s %pM\n",
+			 "VDEV MAC address", vif_macaddr);
+	len += scnprintf(buf + len, buf_len - len, "%30s %u\n",
+			 "beacon snr", vdev->beacon_snr);
+	len += scnprintf(buf + len, buf_len - len, "%30s %u\n",
+			 "data snr", vdev->data_snr);
+	len += scnprintf(buf + len, buf_len - len, "%30s %u\n",
+			 "num rx frames", vdev->num_rx_frames);
+	len += scnprintf(buf + len, buf_len - len, "%30s %u\n",
+			 "num rts fail", vdev->num_rts_fail);
+	len += scnprintf(buf + len, buf_len - len, "%30s %u\n",
+			 "num rts success", vdev->num_rts_success);
+	len += scnprintf(buf + len, buf_len - len, "%30s %u\n",
+			 "num rx err", vdev->num_rx_err);
+	len += scnprintf(buf + len, buf_len - len, "%30s %u\n",
+			 "num rx discard", vdev->num_rx_discard);
+	len += scnprintf(buf + len, buf_len - len, "%30s %u\n",
+			 "num tx not acked", vdev->num_tx_not_acked);
+
+	for (i = 0 ; i < ARRAY_SIZE(vdev->num_tx_frames); i++)
+		len += scnprintf(buf + len, buf_len - len,
+				"%25s [%02d] %u\n",
+				"num tx frames", i,
+				vdev->num_tx_frames[i]);
+
+	for (i = 0 ; i < ARRAY_SIZE(vdev->num_tx_frames_retries); i++)
+		len += scnprintf(buf + len, buf_len - len,
+				"%25s [%02d] %u\n",
+				"num tx frames retries", i,
+				vdev->num_tx_frames_retries[i]);
+
+	for (i = 0 ; i < ARRAY_SIZE(vdev->num_tx_frames_failures); i++)
+		len += scnprintf(buf + len, buf_len - len,
+				"%25s [%02d] %u\n",
+				"num tx frames failures", i,
+				vdev->num_tx_frames_failures[i]);
+
+	for (i = 0 ; i < ARRAY_SIZE(vdev->tx_rate_history); i++)
+		len += scnprintf(buf + len, buf_len - len,
+				"%25s [%02d] 0x%08x\n",
+				"tx rate history", i,
+				vdev->tx_rate_history[i]);
+
+	for (i = 0 ; i < ARRAY_SIZE(vdev->beacon_rssi_history); i++)
+		len += scnprintf(buf + len, buf_len - len,
+				"%25s [%02d] %u\n",
+				"beacon rssi history", i,
+				vdev->beacon_rssi_history[i]);
+
+	len += scnprintf(buf + len, buf_len - len, "\n");
+	*length = len;
+}
+
+static void
+ath11k_wmi_fw_bcn_stats_fill(struct ath11k *ar,
+			     const struct ath11k_fw_stats_bcn *bcn,
+			     char *buf, u32 *length)
+{
+	u32 len = *length;
+	u32 buf_len = ATH11K_FW_STATS_BUF_SIZE;
+	struct ath11k_vif *arvif = ath11k_mac_get_arvif(ar, bcn->vdev_id);
+	u8 *vdev_macaddr;
+
+	if (!arvif) {
+		ath11k_warn(ar->ab, "invalid vdev id %d in bcn stats",
+			    bcn->vdev_id);
+		return;
+	}
+
+	vdev_macaddr = arvif->vif->addr;
+
+	len += scnprintf(buf + len, buf_len - len, "%30s %u\n",
+			 "VDEV ID", bcn->vdev_id);
+	len += scnprintf(buf + len, buf_len - len, "%30s %pM\n",
+			 "VDEV MAC address", vdev_macaddr);
+	len += scnprintf(buf + len, buf_len - len, "%30s\n\n",
+			 "================");
+	len += scnprintf(buf + len, buf_len - len, "%30s %u\n",
+			 "Num of beacon tx success", bcn->tx_bcn_succ_cnt);
+	len += scnprintf(buf + len, buf_len - len, "%30s %u\n",
+			 "Num of beacon tx failures", bcn->tx_bcn_outage_cnt);
+
+	len += scnprintf(buf + len, buf_len - len, "\n");
+	*length = len;
+}
+
+void ath11k_wmi_fw_stats_fill(struct ath11k *ar,
+			      struct ath11k_fw_stats *fw_stats,
+			      u32 stats_id, char *buf)
+{
+	u32 len = 0;
+	u32 buf_len = ATH11K_FW_STATS_BUF_SIZE;
+	const struct ath11k_fw_stats_pdev *pdev;
+	const struct ath11k_fw_stats_vdev *vdev;
+	const struct ath11k_fw_stats_bcn *bcn;
+	size_t num_bcn;
+
+	spin_lock_bh(&ar->data_lock);
+
+	if (stats_id == WMI_REQUEST_PDEV_STAT) {
+		pdev = list_first_entry_or_null(&fw_stats->pdevs,
+						struct ath11k_fw_stats_pdev, list);
+		if (!pdev) {
+			ath11k_warn(ar->ab, "failed to get pdev stats\n");
+			goto unlock;
+		}
+
+		ath11k_wmi_fw_pdev_base_stats_fill(pdev, buf, &len);
+		ath11k_wmi_fw_pdev_tx_stats_fill(pdev, buf, &len);
+		ath11k_wmi_fw_pdev_rx_stats_fill(pdev, buf, &len);
+	}
+
+	if (stats_id == WMI_REQUEST_VDEV_STAT) {
+		len += scnprintf(buf + len, buf_len - len, "\n");
+		len += scnprintf(buf + len, buf_len - len, "%30s\n",
+				 "ath11k VDEV stats");
+		len += scnprintf(buf + len, buf_len - len, "%30s\n\n",
+				 "=================");
+
+		list_for_each_entry(vdev, &fw_stats->vdevs, list)
+			ath11k_wmi_fw_vdev_stats_fill(ar, vdev, buf, &len);
+	}
+
+	if (stats_id == WMI_REQUEST_BCN_STAT) {
+		num_bcn = ath11k_wmi_fw_stats_num_bcn(&fw_stats->bcn);
+
+		len += scnprintf(buf + len, buf_len - len, "\n");
+		len += scnprintf(buf + len, buf_len - len, "%30s (%zu)\n",
+				 "ath11k Beacon stats", num_bcn);
+		len += scnprintf(buf + len, buf_len - len, "%30s\n\n",
+				 "===================");
+
+		list_for_each_entry(bcn, &fw_stats->bcn, list)
+			ath11k_wmi_fw_bcn_stats_fill(ar, bcn, buf, &len);
+	}
+
+unlock:
+	spin_unlock_bh(&ar->data_lock);
+
+	if (len >= buf_len)
+		buf[len - 1] = 0;
+	else
+		buf[len] = 0;
+}
+
+static void ath11k_wmi_op_ep_tx_credits(struct ath11k_base *ab)
+{
+	/* try to send pending beacons first. they take priority */
+	wake_up(&ab->wmi_sc.tx_credits_wq);
+}
+
+static void ath11k_wmi_htc_tx_complete(struct ath11k_base *ab,
+				       struct sk_buff *skb)
+{
+	dev_kfree_skb(skb);
+}
+
+static bool ath11k_reg_is_world_alpha(char *alpha)
+{
+	return alpha[0] == '0' && alpha[1] == '0';
+}
+
+static int ath11k_reg_chan_list_event(struct ath11k_base *ab, struct sk_buff *skb)
+{
+	struct cur_regulatory_info *reg_info = NULL;
+	struct ieee80211_regdomain *regd = NULL;
+	bool intersect = false;
+	int ret = 0, pdev_idx;
+	struct ath11k *ar;
+
+	reg_info = kzalloc(sizeof(*reg_info), GFP_ATOMIC);
+	if (!reg_info) {
+		ret = -ENOMEM;
+		goto fallback;
+	}
+
+	ret = ath11k_pull_reg_chan_list_update_ev(ab, skb, reg_info);
+	if (ret) {
+		ath11k_warn(ab, "failed to extract regulatory info from received event\n");
+		goto fallback;
+	}
+
+	if (reg_info->status_code != REG_SET_CC_STATUS_PASS) {
+		/* In case of failure to set the requested ctry,
+		 * fw retains the current regd. We print a failure info
+		 * and return from here.
+		 */
+		ath11k_warn(ab, "Failed to set the requested Country regulatory setting\n");
+		goto mem_free;
+	}
+
+	pdev_idx = reg_info->phy_id;
+
+	if (pdev_idx >= ab->num_radios)
+		goto fallback;
+
+	/* Avoid multiple overwrites to default regd, during core
+	 * stop-start after mac registration.
+	 */
+	if (ab->default_regd[pdev_idx] && !ab->new_regd[pdev_idx] &&
+	    !memcmp((char *)ab->default_regd[pdev_idx]->alpha2,
+		    (char *)reg_info->alpha2, 2))
+		return 0;
+
+	/* Intersect new rules with default regd if a new country setting was
+	 * requested, i.e a default regd was already set during initialization
+	 * and the regd coming from this event has a valid country info.
+	 */
+	if (ab->default_regd[pdev_idx] &&
+	    !ath11k_reg_is_world_alpha((char *)
+		ab->default_regd[pdev_idx]->alpha2) &&
+	    !ath11k_reg_is_world_alpha((char *)reg_info->alpha2))
+		intersect = true;
+
+	regd = ath11k_reg_build_regd(ab, reg_info, intersect);
+	if (!regd) {
+		ath11k_warn(ab, "failed to build regd from reg_info\n");
+		goto fallback;
+	}
+
+	spin_lock(&ab->base_lock);
+	if (test_bit(ATH11K_FLAG_REGISTERED, &ab->dev_flags)) {
+		/* Once mac is registered, ar is valid and all CC events from
+		 * fw is considered to be received due to user requests
+		 * currently.
+		 * Free previously built regd before assigning the newly
+		 * generated regd to ar. NULL pointer handling will be
+		 * taken care by kfree itself.
+		 */
+		ar = ab->pdevs[pdev_idx].ar;
+		kfree(ab->new_regd[pdev_idx]);
+		ab->new_regd[pdev_idx] = regd;
+		ieee80211_queue_work(ar->hw, &ar->regd_update_work);
+	} else {
+		/* Multiple events for the same *ar is not expected. But we
+		 * can still clear any previously stored default_regd if we
+		 * are receiving this event for the same radio by mistake.
+		 * NULL pointer handling will be taken care by kfree itself.
+		 */
+		kfree(ab->default_regd[pdev_idx]);
+		/* This regd would be applied during mac registration */
+		ab->default_regd[pdev_idx] = regd;
+	}
+	ab->dfs_region = reg_info->dfs_region;
+	spin_unlock(&ab->base_lock);
+
+	goto mem_free;
+
+fallback:
+	/* Fallback to older reg (by sending previous country setting
+	 * again if fw has succeded and we failed to process here.
+	 * The Regdomain should be uniform across driver and fw. Since the
+	 * FW has processed the command and sent a success status, we expect
+	 * this function to succeed as well. If it doesn't, CTRY needs to be
+	 * reverted at the fw and the old SCAN_CHAN_LIST cmd needs to be sent.
+	 */
+	/* TODO: This is rare, but still should also be handled */
+	WARN_ON(1);
+mem_free:
+	if (reg_info) {
+		kfree(reg_info->reg_rules_2g_ptr);
+		kfree(reg_info->reg_rules_5g_ptr);
+		kfree(reg_info);
+	}
+	return ret;
+}
+
+static int ath11k_wmi_tlv_rdy_parse(struct ath11k_base *ab, u16 tag, u16 len,
+				    const void *ptr, void *data)
+{
+	struct wmi_tlv_rdy_parse *rdy_parse = data;
+	struct wmi_ready_event *fixed_param;
+	struct wmi_mac_addr *addr_list;
+	struct ath11k_pdev *pdev;
+	u32 num_mac_addr;
+	int i;
+
+	switch (tag) {
+	case WMI_TAG_READY_EVENT:
+		fixed_param = (struct wmi_ready_event *)ptr;
+		ab->wlan_init_status = fixed_param->status;
+		rdy_parse->num_extra_mac_addr = fixed_param->num_extra_mac_addr;
+
+		ether_addr_copy(ab->mac_addr, fixed_param->mac_addr.addr);
+		ab->wmi_ready = true;
+		break;
+	case WMI_TAG_ARRAY_FIXED_STRUCT:
+		addr_list = (struct wmi_mac_addr *)ptr;
+		num_mac_addr = rdy_parse->num_extra_mac_addr;
+
+		if (!(ab->num_radios > 1 && num_mac_addr >= ab->num_radios))
+			break;
+
+		for (i = 0; i < ab->num_radios; i++) {
+			pdev = &ab->pdevs[i];
+			ether_addr_copy(pdev->mac_addr, addr_list[i].addr);
+		}
+		ab->pdevs_macaddr_valid = true;
+		break;
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+static int ath11k_ready_event(struct ath11k_base *ab, struct sk_buff *skb)
+{
+	struct wmi_tlv_rdy_parse rdy_parse = { };
+	int ret;
+
+	ret = ath11k_wmi_tlv_iter(ab, skb->data, skb->len,
+				  ath11k_wmi_tlv_rdy_parse, &rdy_parse);
+	if (ret) {
+		ath11k_warn(ab, "failed to parse tlv %d\n", ret);
+		return ret;
+	}
+
+	complete(&ab->wmi_sc.unified_ready);
+	return 0;
+}
+
+static void ath11k_peer_delete_resp_event(struct ath11k_base *ab, struct sk_buff *skb)
+{
+	struct wmi_peer_delete_resp_event peer_del_resp;
+
+	if (ath11k_pull_peer_del_resp_ev(ab, skb, &peer_del_resp) != 0) {
+		ath11k_warn(ab, "failed to extract peer delete resp");
+		return;
+	}
+
+	/* TODO: Do we need to validate whether ath11k_peer_find() return NULL
+	 *	 Why this is needed when there is HTT event for peer delete
+	 */
+}
+
+static inline const char *ath11k_wmi_vdev_resp_print(u32 vdev_resp_status)
+{
+	switch (vdev_resp_status) {
+	case WMI_VDEV_START_RESPONSE_INVALID_VDEVID:
+		return "invalid vdev id";
+	case WMI_VDEV_START_RESPONSE_NOT_SUPPORTED:
+		return "not supported";
+	case WMI_VDEV_START_RESPONSE_DFS_VIOLATION:
+		return "dfs violation";
+	case WMI_VDEV_START_RESPONSE_INVALID_REGDOMAIN:
+		return "invalid regdomain";
+	default:
+		return "unknown";
+	}
+}
+
+static void ath11k_vdev_start_resp_event(struct ath11k_base *ab, struct sk_buff *skb)
+{
+	struct wmi_vdev_start_resp_event vdev_start_resp;
+	struct ath11k *ar;
+	u32 status;
+
+	if (ath11k_pull_vdev_start_resp_tlv(ab, skb, &vdev_start_resp) != 0) {
+		ath11k_warn(ab, "failed to extract vdev start resp");
+		return;
+	}
+
+	rcu_read_lock();
+	ar = ath11k_mac_get_ar_by_vdev_id(ab, vdev_start_resp.vdev_id);
+	if (!ar) {
+		ath11k_warn(ab, "invalid vdev id in vdev start resp ev %d",
+			    vdev_start_resp.vdev_id);
+		rcu_read_unlock();
+		return;
+	}
+
+	ar->last_wmi_vdev_start_status = 0;
+
+	status = vdev_start_resp.status;
+
+	if (WARN_ON_ONCE(status)) {
+		ath11k_warn(ab, "vdev start resp error status %d (%s)\n",
+			    status, ath11k_wmi_vdev_resp_print(status));
+		ar->last_wmi_vdev_start_status = status;
+	}
+
+	complete(&ar->vdev_setup_done);
+
+	rcu_read_unlock();
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI, "vdev start resp for vdev id %d",
+		   vdev_start_resp.vdev_id);
+}
+
+static void ath11k_bcn_tx_status_event(struct ath11k_base *ab, struct sk_buff *skb)
+{
+	u32 vdev_id, tx_status;
+
+	if (ath11k_pull_bcn_tx_status_ev(ab, skb->data, skb->len,
+					 &vdev_id, &tx_status) != 0) {
+		ath11k_warn(ab, "failed to extract bcn tx status");
+		return;
+	}
+}
+
+static void ath11k_vdev_stopped_event(struct ath11k_base *ab, struct sk_buff *skb)
+{
+	struct ath11k *ar;
+	u32 vdev_id = 0;
+
+	if (ath11k_pull_vdev_stopped_param_tlv(ab, skb, &vdev_id) != 0) {
+		ath11k_warn(ab, "failed to extract vdev stopped event");
+		return;
+	}
+
+	rcu_read_lock();
+	ar = ath11k_mac_get_ar_vdev_stop_status(ab, vdev_id);
+	if (!ar) {
+		ath11k_warn(ab, "invalid vdev id in vdev stopped ev %d",
+			    vdev_id);
+		rcu_read_unlock();
+		return;
+	}
+
+	complete(&ar->vdev_setup_done);
+
+	rcu_read_unlock();
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI, "vdev stopped for vdev id %d", vdev_id);
+}
+
+static void ath11k_mgmt_rx_event(struct ath11k_base *ab, struct sk_buff *skb)
+{
+	struct mgmt_rx_event_params rx_ev = {0};
+	struct ath11k *ar;
+	struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(skb);
+	struct ieee80211_hdr *hdr;
+	u16 fc;
+	struct ieee80211_supported_band *sband;
+
+	if (ath11k_pull_mgmt_rx_params_tlv(ab, skb, &rx_ev) != 0) {
+		ath11k_warn(ab, "failed to extract mgmt rx event");
+		dev_kfree_skb(skb);
+		return;
+	}
+
+	memset(status, 0, sizeof(*status));
+
+	ath11k_dbg(ab, ATH11K_DBG_MGMT, "mgmt rx event status %08x\n",
+		   rx_ev.status);
+
+	rcu_read_lock();
+	ar = ath11k_mac_get_ar_by_pdev_id(ab, rx_ev.pdev_id);
+
+	if (!ar) {
+		ath11k_warn(ab, "invalid pdev_id %d in mgmt_rx_event\n",
+			    rx_ev.pdev_id);
+		dev_kfree_skb(skb);
+		goto exit;
+	}
+
+	if ((test_bit(ATH11K_CAC_RUNNING, &ar->dev_flags)) ||
+	    (rx_ev.status & (WMI_RX_STATUS_ERR_DECRYPT |
+	    WMI_RX_STATUS_ERR_KEY_CACHE_MISS | WMI_RX_STATUS_ERR_CRC))) {
+		dev_kfree_skb(skb);
+		goto exit;
+	}
+
+	if (rx_ev.status & WMI_RX_STATUS_ERR_MIC)
+		status->flag |= RX_FLAG_MMIC_ERROR;
+
+	if (rx_ev.channel >= 1 && rx_ev.channel <= 14) {
+		status->band = NL80211_BAND_2GHZ;
+	} else if (rx_ev.channel >= 36 && rx_ev.channel <= ATH11K_MAX_5G_CHAN) {
+		status->band = NL80211_BAND_5GHZ;
+	} else {
+		/* Shouldn't happen unless list of advertised channels to
+		 * mac80211 has been changed.
+		 */
+		WARN_ON_ONCE(1);
+		dev_kfree_skb(skb);
+		goto exit;
+	}
+
+	if (rx_ev.phy_mode == MODE_11B && status->band == NL80211_BAND_5GHZ)
+		ath11k_dbg(ab, ATH11K_DBG_WMI,
+			   "wmi mgmt rx 11b (CCK) on 5GHz\n");
+
+	sband = &ar->mac.sbands[status->band];
+
+	status->freq = ieee80211_channel_to_frequency(rx_ev.channel,
+						      status->band);
+	status->signal = rx_ev.snr + ATH11K_DEFAULT_NOISE_FLOOR;
+	status->rate_idx = ath11k_mac_bitrate_to_idx(sband, rx_ev.rate / 100);
+
+	hdr = (struct ieee80211_hdr *)skb->data;
+	fc = le16_to_cpu(hdr->frame_control);
+
+	/* Firmware is guaranteed to report all essential management frames via
+	 * WMI while it can deliver some extra via HTT. Since there can be
+	 * duplicates split the reporting wrt monitor/sniffing.
+	 */
+	status->flag |= RX_FLAG_SKIP_MONITOR;
+
+	/* In case of PMF, FW delivers decrypted frames with Protected Bit set.
+	 * Don't clear that. Also, FW delivers broadcast management frames
+	 * (ex: group privacy action frames in mesh) as encrypted payload.
+	 */
+	if (ieee80211_has_protected(hdr->frame_control) &&
+	    !is_multicast_ether_addr(ieee80211_get_DA(hdr))) {
+		status->flag |= RX_FLAG_DECRYPTED;
+
+		if (!ieee80211_is_robust_mgmt_frame(skb)) {
+			status->flag |= RX_FLAG_IV_STRIPPED |
+					RX_FLAG_MMIC_STRIPPED;
+			hdr->frame_control = __cpu_to_le16(fc &
+					     ~IEEE80211_FCTL_PROTECTED);
+		}
+	}
+
+	/* TODO: Pending handle beacon implementation
+	 *if (ieee80211_is_beacon(hdr->frame_control))
+	 *	ath11k_mac_handle_beacon(ar, skb);
+	 */
+
+	ath11k_dbg(ab, ATH11K_DBG_MGMT,
+		   "event mgmt rx skb %pK len %d ftype %02x stype %02x\n",
+		   skb, skb->len,
+		   fc & IEEE80211_FCTL_FTYPE, fc & IEEE80211_FCTL_STYPE);
+
+	ath11k_dbg(ab, ATH11K_DBG_MGMT,
+		   "event mgmt rx freq %d band %d snr %d, rate_idx %d\n",
+		   status->freq, status->band, status->signal,
+		   status->rate_idx);
+
+	ieee80211_rx_ni(ar->hw, skb);
+
+exit:
+	rcu_read_unlock();
+}
+
+static void ath11k_mgmt_tx_compl_event(struct ath11k_base *ab, struct sk_buff *skb)
+{
+	struct wmi_mgmt_tx_compl_event tx_compl_param = {0};
+	struct ath11k *ar;
+
+	if (ath11k_pull_mgmt_tx_compl_param_tlv(ab, skb, &tx_compl_param) != 0) {
+		ath11k_warn(ab, "failed to extract mgmt tx compl event");
+		return;
+	}
+
+	rcu_read_lock();
+	ar = ath11k_mac_get_ar_by_pdev_id(ab, tx_compl_param.pdev_id);
+	if (!ar) {
+		ath11k_warn(ab, "invalid pdev id %d in mgmt_tx_compl_event\n",
+			    tx_compl_param.pdev_id);
+		goto exit;
+	}
+
+	wmi_process_mgmt_tx_comp(ar, tx_compl_param.desc_id,
+				 tx_compl_param.status);
+
+	ath11k_dbg(ab, ATH11K_DBG_MGMT,
+		   "mgmt tx compl ev pdev_id %d, desc_id %d, status %d",
+		   tx_compl_param.pdev_id, tx_compl_param.desc_id,
+		   tx_compl_param.status);
+
+exit:
+	rcu_read_unlock();
+}
+
+static struct ath11k *ath11k_get_ar_on_scan_abort(struct ath11k_base *ab,
+						  u32 vdev_id)
+{
+	int i;
+	struct ath11k_pdev *pdev;
+	struct ath11k *ar;
+
+	for (i = 0; i < ab->num_radios; i++) {
+		pdev = rcu_dereference(ab->pdevs_active[i]);
+		if (pdev && pdev->ar) {
+			ar = pdev->ar;
+
+			spin_lock_bh(&ar->data_lock);
+			if (ar->scan.state == ATH11K_SCAN_ABORTING &&
+			    ar->scan.vdev_id == vdev_id) {
+				spin_unlock_bh(&ar->data_lock);
+				return ar;
+			}
+			spin_unlock_bh(&ar->data_lock);
+		}
+	}
+	return NULL;
+}
+
+static void ath11k_scan_event(struct ath11k_base *ab, struct sk_buff *skb)
+{
+	struct ath11k *ar;
+	struct wmi_scan_event scan_ev = {0};
+
+	if (ath11k_pull_scan_ev(ab, skb, &scan_ev) != 0) {
+		ath11k_warn(ab, "failed to extract scan event");
+		return;
+	}
+
+	rcu_read_lock();
+
+	/* In case the scan was cancelled, ex. during interface teardown,
+	 * the interface will not be found in active interfaces.
+	 * Rather, in such scenarios, iterate over the active pdev's to
+	 * search 'ar' if the corresponding 'ar' scan is ABORTING and the
+	 * aborting scan's vdev id matches this event info.
+	 */
+	if (scan_ev.event_type == WMI_SCAN_EVENT_COMPLETED &&
+	    scan_ev.reason == WMI_SCAN_REASON_CANCELLED)
+		ar = ath11k_get_ar_on_scan_abort(ab, scan_ev.vdev_id);
+	else
+		ar = ath11k_mac_get_ar_by_vdev_id(ab, scan_ev.vdev_id);
+
+	if (!ar) {
+		ath11k_warn(ab, "Received scan event for unknown vdev");
+		rcu_read_unlock();
+		return;
+	}
+
+	spin_lock_bh(&ar->data_lock);
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI,
+		   "scan event %s type %d reason %d freq %d req_id %d scan_id %d vdev_id %d state %s (%d)\n",
+		   ath11k_wmi_event_scan_type_str(scan_ev.event_type, scan_ev.reason),
+		   scan_ev.event_type, scan_ev.reason, scan_ev.channel_freq,
+		   scan_ev.scan_req_id, scan_ev.scan_id, scan_ev.vdev_id,
+		   ath11k_scan_state_str(ar->scan.state), ar->scan.state);
+
+	switch (scan_ev.event_type) {
+	case WMI_SCAN_EVENT_STARTED:
+		ath11k_wmi_event_scan_started(ar);
+		break;
+	case WMI_SCAN_EVENT_COMPLETED:
+		ath11k_wmi_event_scan_completed(ar);
+		break;
+	case WMI_SCAN_EVENT_BSS_CHANNEL:
+		ath11k_wmi_event_scan_bss_chan(ar);
+		break;
+	case WMI_SCAN_EVENT_FOREIGN_CHAN:
+		ath11k_wmi_event_scan_foreign_chan(ar, scan_ev.channel_freq);
+		break;
+	case WMI_SCAN_EVENT_START_FAILED:
+		ath11k_warn(ab, "received scan start failure event\n");
+		ath11k_wmi_event_scan_start_failed(ar);
+		break;
+	case WMI_SCAN_EVENT_DEQUEUED:
+	case WMI_SCAN_EVENT_PREEMPTED:
+	case WMI_SCAN_EVENT_RESTARTED:
+	case WMI_SCAN_EVENT_FOREIGN_CHAN_EXIT:
+	default:
+		break;
+	}
+
+	spin_unlock_bh(&ar->data_lock);
+
+	rcu_read_unlock();
+}
+
+static void ath11k_peer_sta_kickout_event(struct ath11k_base *ab, struct sk_buff *skb)
+{
+	struct wmi_peer_sta_kickout_arg arg = {};
+	struct ieee80211_sta *sta;
+	struct ath11k_peer *peer;
+	struct ath11k *ar;
+
+	if (ath11k_pull_peer_sta_kickout_ev(ab, skb, &arg) != 0) {
+		ath11k_warn(ab, "failed to extract peer sta kickout event");
+		return;
+	}
+
+	rcu_read_lock();
+
+	spin_lock_bh(&ab->base_lock);
+
+	peer = ath11k_peer_find_by_addr(ab, arg.mac_addr);
+
+	if (!peer) {
+		ath11k_warn(ab, "peer not found %pM\n",
+			    arg.mac_addr);
+		goto exit;
+	}
+
+	ar = ath11k_mac_get_ar_by_vdev_id(ab, peer->vdev_id);
+	if (!ar) {
+		ath11k_warn(ab, "invalid vdev id in peer sta kickout ev %d",
+			    peer->vdev_id);
+		goto exit;
+	}
+
+	sta = ieee80211_find_sta_by_ifaddr(ar->hw,
+					   arg.mac_addr, NULL);
+	if (!sta) {
+		ath11k_warn(ab, "Spurious quick kickout for STA %pM\n",
+			    arg.mac_addr);
+		goto exit;
+	}
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI, "peer sta kickout event %pM",
+		   arg.mac_addr);
+
+	ieee80211_report_low_ack(sta, 10);
+
+exit:
+	spin_unlock_bh(&ab->base_lock);
+	rcu_read_unlock();
+}
+
+static void ath11k_roam_event(struct ath11k_base *ab, struct sk_buff *skb)
+{
+	struct wmi_roam_event roam_ev = {};
+	struct ath11k *ar;
+
+	if (ath11k_pull_roam_ev(ab, skb, &roam_ev) != 0) {
+		ath11k_warn(ab, "failed to extract roam event");
+		return;
+	}
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI,
+		   "wmi roam event vdev %u reason 0x%08x rssi %d\n",
+		   roam_ev.vdev_id, roam_ev.reason, roam_ev.rssi);
+
+	rcu_read_lock();
+	ar = ath11k_mac_get_ar_by_vdev_id(ab, roam_ev.vdev_id);
+	if (!ar) {
+		ath11k_warn(ab, "invalid vdev id in roam ev %d",
+			    roam_ev.vdev_id);
+		rcu_read_unlock();
+		return;
+	}
+
+	if (roam_ev.reason >= WMI_ROAM_REASON_MAX)
+		ath11k_warn(ab, "ignoring unknown roam event reason %d on vdev %i\n",
+			    roam_ev.reason, roam_ev.vdev_id);
+
+	switch (roam_ev.reason) {
+	case WMI_ROAM_REASON_BEACON_MISS:
+		/* TODO: Pending beacon miss and connection_loss_work
+		 * implementation
+		 * ath11k_mac_handle_beacon_miss(ar, vdev_id);
+		 */
+		break;
+	case WMI_ROAM_REASON_BETTER_AP:
+	case WMI_ROAM_REASON_LOW_RSSI:
+	case WMI_ROAM_REASON_SUITABLE_AP_FOUND:
+	case WMI_ROAM_REASON_HO_FAILED:
+		ath11k_warn(ab, "ignoring not implemented roam event reason %d on vdev %i\n",
+			    roam_ev.reason, roam_ev.vdev_id);
+		break;
+	}
+
+	rcu_read_unlock();
+}
+
+static void ath11k_chan_info_event(struct ath11k_base *ab, struct sk_buff *skb)
+{
+	struct wmi_chan_info_event ch_info_ev = {0};
+	struct ath11k *ar;
+	struct survey_info *survey;
+	int idx;
+	/* HW channel counters frequency value in hertz */
+	u32 cc_freq_hz = ab->cc_freq_hz;
+
+	if (ath11k_pull_chan_info_ev(ab, skb->data, skb->len, &ch_info_ev) != 0) {
+		ath11k_warn(ab, "failed to extract chan info event");
+		return;
+	}
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI,
+		   "chan info vdev_id %d err_code %d freq %d cmd_flags %d noise_floor %d rx_clear_count %d cycle_count %d mac_clk_mhz %d\n",
+		   ch_info_ev.vdev_id, ch_info_ev.err_code, ch_info_ev.freq,
+		   ch_info_ev.cmd_flags, ch_info_ev.noise_floor,
+		   ch_info_ev.rx_clear_count, ch_info_ev.cycle_count,
+		   ch_info_ev.mac_clk_mhz);
+
+	if (ch_info_ev.cmd_flags == WMI_CHAN_INFO_END_RESP) {
+		ath11k_dbg(ab, ATH11K_DBG_WMI, "chan info report completed\n");
+		return;
+	}
+
+	rcu_read_lock();
+	ar = ath11k_mac_get_ar_by_vdev_id(ab, ch_info_ev.vdev_id);
+	if (!ar) {
+		ath11k_warn(ab, "invalid vdev id in chan info ev %d",
+			    ch_info_ev.vdev_id);
+		rcu_read_unlock();
+		return;
+	}
+	spin_lock_bh(&ar->data_lock);
+
+	switch (ar->scan.state) {
+	case ATH11K_SCAN_IDLE:
+	case ATH11K_SCAN_STARTING:
+		ath11k_warn(ab, "received chan info event without a scan request, ignoring\n");
+		goto exit;
+	case ATH11K_SCAN_RUNNING:
+	case ATH11K_SCAN_ABORTING:
+		break;
+	}
+
+	idx = freq_to_idx(ar, ch_info_ev.freq);
+	if (idx >= ARRAY_SIZE(ar->survey)) {
+		ath11k_warn(ab, "chan info: invalid frequency %d (idx %d out of bounds)\n",
+			    ch_info_ev.freq, idx);
+		goto exit;
+	}
+
+	/* If FW provides MAC clock frequency in Mhz, overriding the initialized
+	 * HW channel counters frequency value
+	 */
+	if (ch_info_ev.mac_clk_mhz)
+		cc_freq_hz = (ch_info_ev.mac_clk_mhz * 1000);
+
+	if (ch_info_ev.cmd_flags == WMI_CHAN_INFO_START_RESP) {
+		survey = &ar->survey[idx];
+		memset(survey, 0, sizeof(*survey));
+		survey->noise = ch_info_ev.noise_floor;
+		survey->filled = SURVEY_INFO_NOISE_DBM | SURVEY_INFO_TIME |
+				 SURVEY_INFO_TIME_BUSY;
+		survey->time = div_u64(ch_info_ev.cycle_count, cc_freq_hz);
+		survey->time_busy = div_u64(ch_info_ev.rx_clear_count, cc_freq_hz);
+	}
+exit:
+	spin_unlock_bh(&ar->data_lock);
+	rcu_read_unlock();
+}
+
+static void
+ath11k_pdev_bss_chan_info_event(struct ath11k_base *ab, struct sk_buff *skb)
+{
+	struct wmi_pdev_bss_chan_info_event bss_ch_info_ev = {};
+	struct survey_info *survey;
+	struct ath11k *ar;
+	u32 cc_freq_hz = ab->cc_freq_hz;
+	u64 busy, total, tx, rx, rx_bss;
+	int idx;
+
+	if (ath11k_pull_pdev_bss_chan_info_ev(ab, skb, &bss_ch_info_ev) != 0) {
+		ath11k_warn(ab, "failed to extract pdev bss chan info event");
+		return;
+	}
+
+	busy = (u64)(bss_ch_info_ev.rx_clear_count_high) << 32 |
+			bss_ch_info_ev.rx_clear_count_low;
+
+	total = (u64)(bss_ch_info_ev.cycle_count_high) << 32 |
+			bss_ch_info_ev.cycle_count_low;
+
+	tx = (u64)(bss_ch_info_ev.tx_cycle_count_high) << 32 |
+			bss_ch_info_ev.tx_cycle_count_low;
+
+	rx = (u64)(bss_ch_info_ev.rx_cycle_count_high) << 32 |
+			bss_ch_info_ev.rx_cycle_count_low;
+
+	rx_bss = (u64)(bss_ch_info_ev.rx_bss_cycle_count_high) << 32 |
+			bss_ch_info_ev.rx_bss_cycle_count_low;
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI,
+		   "pdev bss chan info:\n pdev_id: %d freq: %d noise: %d cycle: busy %llu total %llu tx %llu rx %llu rx_bss %llu\n",
+		   bss_ch_info_ev.pdev_id, bss_ch_info_ev.freq,
+		   bss_ch_info_ev.noise_floor, busy, total,
+		   tx, rx, rx_bss);
+
+	rcu_read_lock();
+	ar = ath11k_mac_get_ar_by_pdev_id(ab, bss_ch_info_ev.pdev_id);
+
+	if (!ar) {
+		ath11k_warn(ab, "invalid pdev id %d in bss_chan_info event\n",
+			    bss_ch_info_ev.pdev_id);
+		rcu_read_unlock();
+		return;
+	}
+
+	spin_lock_bh(&ar->data_lock);
+	idx = freq_to_idx(ar, bss_ch_info_ev.freq);
+	if (idx >= ARRAY_SIZE(ar->survey)) {
+		ath11k_warn(ab, "bss chan info: invalid frequency %d (idx %d out of bounds)\n",
+			    bss_ch_info_ev.freq, idx);
+		goto exit;
+	}
+
+	survey = &ar->survey[idx];
+
+	survey->noise     = bss_ch_info_ev.noise_floor;
+	survey->time      = div_u64(total, cc_freq_hz);
+	survey->time_busy = div_u64(busy, cc_freq_hz);
+	survey->time_rx   = div_u64(rx_bss, cc_freq_hz);
+	survey->time_tx   = div_u64(tx, cc_freq_hz);
+	survey->filled   |= (SURVEY_INFO_NOISE_DBM |
+			     SURVEY_INFO_TIME |
+			     SURVEY_INFO_TIME_BUSY |
+			     SURVEY_INFO_TIME_RX |
+			     SURVEY_INFO_TIME_TX);
+exit:
+	spin_unlock_bh(&ar->data_lock);
+	complete(&ar->bss_survey_done);
+
+	rcu_read_unlock();
+}
+
+static void ath11k_vdev_install_key_compl_event(struct ath11k_base *ab,
+						struct sk_buff *skb)
+{
+	struct wmi_vdev_install_key_complete_arg install_key_compl = {0};
+	struct ath11k *ar;
+
+	if (ath11k_pull_vdev_install_key_compl_ev(ab, skb, &install_key_compl) != 0) {
+		ath11k_warn(ab, "failed to extract install key compl event");
+		return;
+	}
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI,
+		   "vdev install key ev idx %d flags %08x macaddr %pM status %d\n",
+		   install_key_compl.key_idx, install_key_compl.key_flags,
+		   install_key_compl.macaddr, install_key_compl.status);
+
+	rcu_read_lock();
+	ar = ath11k_mac_get_ar_by_vdev_id(ab, install_key_compl.vdev_id);
+	if (!ar) {
+		ath11k_warn(ab, "invalid vdev id in install key compl ev %d",
+			    install_key_compl.vdev_id);
+		rcu_read_unlock();
+		return;
+	}
+
+	ar->install_key_status = 0;
+
+	if (install_key_compl.status != WMI_VDEV_INSTALL_KEY_COMPL_STATUS_SUCCESS) {
+		ath11k_warn(ab, "install key failed for %pM status %d\n",
+			    install_key_compl.macaddr, install_key_compl.status);
+		ar->install_key_status = install_key_compl.status;
+	}
+
+	complete(&ar->install_key_done);
+	rcu_read_unlock();
+}
+
+static void ath11k_service_available_event(struct ath11k_base *ab, struct sk_buff *skb)
+{
+	const void **tb;
+	const struct wmi_service_available_event *ev;
+	int ret;
+	int i, j;
+
+	tb = ath11k_wmi_tlv_parse_alloc(ab, skb->data, skb->len, GFP_ATOMIC);
+	if (IS_ERR(tb)) {
+		ret = PTR_ERR(tb);
+		ath11k_warn(ab, "failed to parse tlv: %d\n", ret);
+		return;
+	}
+
+	ev = tb[WMI_TAG_SERVICE_AVAILABLE_EVENT];
+	if (!ev) {
+		ath11k_warn(ab, "failed to fetch svc available ev");
+		kfree(tb);
+		return;
+	}
+
+	/* TODO: Use wmi_service_segment_offset information to get the service
+	 * especially when more services are advertised in multiple sevice
+	 * available events.
+	 */
+	for (i = 0, j = WMI_MAX_SERVICE;
+	     i < WMI_SERVICE_SEGMENT_BM_SIZE32 && j < WMI_MAX_EXT_SERVICE;
+	     i++) {
+		do {
+			if (ev->wmi_service_segment_bitmap[i] &
+			    BIT(j % WMI_AVAIL_SERVICE_BITS_IN_SIZE32))
+				set_bit(j, ab->wmi_sc.svc_map);
+		} while (++j % WMI_AVAIL_SERVICE_BITS_IN_SIZE32);
+	}
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI,
+		   "wmi_ext_service_bitmap 0:0x%x, 1:0x%x, 2:0x%x, 3:0x%x",
+		   ev->wmi_service_segment_bitmap[0], ev->wmi_service_segment_bitmap[1],
+		   ev->wmi_service_segment_bitmap[2], ev->wmi_service_segment_bitmap[3]);
+
+	kfree(tb);
+}
+
+static void ath11k_peer_assoc_conf_event(struct ath11k_base *ab, struct sk_buff *skb)
+{
+	struct wmi_peer_assoc_conf_arg peer_assoc_conf = {0};
+	struct ath11k *ar;
+
+	if (ath11k_pull_peer_assoc_conf_ev(ab, skb, &peer_assoc_conf) != 0) {
+		ath11k_warn(ab, "failed to extract peer assoc conf event");
+		return;
+	}
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI,
+		   "peer assoc conf ev vdev id %d macaddr %pM\n",
+		   peer_assoc_conf.vdev_id, peer_assoc_conf.macaddr);
+
+	ar = ath11k_mac_get_ar_by_vdev_id(ab, peer_assoc_conf.vdev_id);
+
+	if (!ar) {
+		ath11k_warn(ab, "invalid vdev id in peer assoc conf ev %d",
+			    peer_assoc_conf.vdev_id);
+		return;
+	}
+
+	complete(&ar->peer_assoc_done);
+}
+
+static void ath11k_update_stats_event(struct ath11k_base *ab, struct sk_buff *skb)
+{
+	ath11k_debug_fw_stats_process(ab, skb);
+}
+
+/* PDEV_CTL_FAILSAFE_CHECK_EVENT is received from FW when the frequency scanned
+ * is not part of BDF CTL(Conformance test limits) table entries.
+ */
+static void ath11k_pdev_ctl_failsafe_check_event(struct ath11k_base *ab,
+						 struct sk_buff *skb)
+{
+	const void **tb;
+	const struct wmi_pdev_ctl_failsafe_chk_event *ev;
+	int ret;
+
+	tb = ath11k_wmi_tlv_parse_alloc(ab, skb->data, skb->len, GFP_ATOMIC);
+	if (IS_ERR(tb)) {
+		ret = PTR_ERR(tb);
+		ath11k_warn(ab, "failed to parse tlv: %d\n", ret);
+		return;
+	}
+
+	ev = tb[WMI_TAG_PDEV_CTL_FAILSAFE_CHECK_EVENT];
+	if (!ev) {
+		ath11k_warn(ab, "failed to fetch pdev ctl failsafe check ev");
+		kfree(tb);
+		return;
+	}
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI,
+		   "pdev ctl failsafe check ev status %d\n",
+		   ev->ctl_failsafe_status);
+
+	/* If ctl_failsafe_status is set to 1 FW will max out the Transmit power
+	 * to 10 dBm else the CTL power entry in the BDF would be picked up.
+	 */
+	if (ev->ctl_failsafe_status != 0)
+		ath11k_warn(ab, "pdev ctl failsafe failure status %d",
+			    ev->ctl_failsafe_status);
+
+	kfree(tb);
+}
+
+static void
+ath11k_wmi_process_csa_switch_count_event(struct ath11k_base *ab,
+					  const struct wmi_pdev_csa_switch_ev *ev,
+					  const u32 *vdev_ids)
+{
+	int i;
+	struct ath11k_vif *arvif;
+
+	/* Finish CSA once the switch count becomes NULL */
+	if (ev->current_switch_count)
+		return;
+
+	rcu_read_lock();
+	for (i = 0; i < ev->num_vdevs; i++) {
+		arvif = ath11k_mac_get_arvif_by_vdev_id(ab, vdev_ids[i]);
+
+		if (!arvif) {
+			ath11k_warn(ab, "Recvd csa status for unknown vdev %d",
+				    vdev_ids[i]);
+			continue;
+		}
+
+		if (arvif->is_up && arvif->vif->csa_active)
+			ieee80211_csa_finish(arvif->vif);
+	}
+	rcu_read_unlock();
+}
+
+static void
+ath11k_wmi_pdev_csa_switch_count_status_event(struct ath11k_base *ab,
+					      struct sk_buff *skb)
+{
+	const void **tb;
+	const struct wmi_pdev_csa_switch_ev *ev;
+	const u32 *vdev_ids;
+	int ret;
+
+	tb = ath11k_wmi_tlv_parse_alloc(ab, skb->data, skb->len, GFP_ATOMIC);
+	if (IS_ERR(tb)) {
+		ret = PTR_ERR(tb);
+		ath11k_warn(ab, "failed to parse tlv: %d\n", ret);
+		return;
+	}
+
+	ev = tb[WMI_TAG_PDEV_CSA_SWITCH_COUNT_STATUS_EVENT];
+	vdev_ids = tb[WMI_TAG_ARRAY_UINT32];
+
+	if (!ev || !vdev_ids) {
+		ath11k_warn(ab, "failed to fetch pdev csa switch count ev");
+		kfree(tb);
+		return;
+	}
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI,
+		   "pdev csa switch count %d for pdev %d, num_vdevs %d",
+		   ev->current_switch_count, ev->pdev_id,
+		   ev->num_vdevs);
+
+	ath11k_wmi_process_csa_switch_count_event(ab, ev, vdev_ids);
+
+	kfree(tb);
+}
+
+static void
+ath11k_wmi_pdev_dfs_radar_detected_event(struct ath11k_base *ab, struct sk_buff *skb)
+{
+	const void **tb;
+	const struct wmi_pdev_radar_ev *ev;
+	struct ath11k *ar;
+	int ret;
+
+	tb = ath11k_wmi_tlv_parse_alloc(ab, skb->data, skb->len, GFP_ATOMIC);
+	if (IS_ERR(tb)) {
+		ret = PTR_ERR(tb);
+		ath11k_warn(ab, "failed to parse tlv: %d\n", ret);
+		return;
+	}
+
+	ev = tb[WMI_TAG_PDEV_DFS_RADAR_DETECTION_EVENT];
+
+	if (!ev) {
+		ath11k_warn(ab, "failed to fetch pdev dfs radar detected ev");
+		kfree(tb);
+		return;
+	}
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI,
+		   "pdev dfs radar detected on pdev %d, detection mode %d, chan freq %d, chan_width %d, detector id %d, seg id %d, timestamp %d, chirp %d, freq offset %d, sidx %d",
+		   ev->pdev_id, ev->detection_mode, ev->chan_freq, ev->chan_width,
+		   ev->detector_id, ev->segment_id, ev->timestamp, ev->is_chirp,
+		   ev->freq_offset, ev->sidx);
+
+	ar = ath11k_mac_get_ar_by_pdev_id(ab, ev->pdev_id);
+
+	if (!ar) {
+		ath11k_warn(ab, "radar detected in invalid pdev %d\n",
+			    ev->pdev_id);
+		goto exit;
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_REG, "DFS Radar Detected in pdev %d\n",
+		   ev->pdev_id);
+
+	if (ar->dfs_block_radar_events)
+		ath11k_info(ab, "DFS Radar detected, but ignored as requested\n");
+	else
+		ieee80211_radar_detected(ar->hw);
+
+exit:
+	kfree(tb);
+}
+
+static void ath11k_wmi_tlv_op_rx(struct ath11k_base *ab, struct sk_buff *skb)
+{
+	struct wmi_cmd_hdr *cmd_hdr;
+	enum wmi_tlv_event_id id;
+
+	cmd_hdr = (struct wmi_cmd_hdr *)skb->data;
+	id = FIELD_GET(WMI_CMD_HDR_CMD_ID, (cmd_hdr->cmd_id));
+
+	if (skb_pull(skb, sizeof(struct wmi_cmd_hdr)) == NULL)
+		goto out;
+
+	switch (id) {
+		/* Process all the WMI events here */
+	case WMI_SERVICE_READY_EVENTID:
+		ath11k_service_ready_event(ab, skb);
+		break;
+	case WMI_SERVICE_READY_EXT_EVENTID:
+		ath11k_service_ready_ext_event(ab, skb);
+		break;
+	case WMI_REG_CHAN_LIST_CC_EVENTID:
+		ath11k_reg_chan_list_event(ab, skb);
+		break;
+	case WMI_READY_EVENTID:
+		ath11k_ready_event(ab, skb);
+		break;
+	case WMI_PEER_DELETE_RESP_EVENTID:
+		ath11k_peer_delete_resp_event(ab, skb);
+		break;
+	case WMI_VDEV_START_RESP_EVENTID:
+		ath11k_vdev_start_resp_event(ab, skb);
+		break;
+	case WMI_OFFLOAD_BCN_TX_STATUS_EVENTID:
+		ath11k_bcn_tx_status_event(ab, skb);
+		break;
+	case WMI_VDEV_STOPPED_EVENTID:
+		ath11k_vdev_stopped_event(ab, skb);
+		break;
+	case WMI_MGMT_RX_EVENTID:
+		ath11k_mgmt_rx_event(ab, skb);
+		/* mgmt_rx_event() owns the skb now! */
+		return;
+	case WMI_MGMT_TX_COMPLETION_EVENTID:
+		ath11k_mgmt_tx_compl_event(ab, skb);
+		break;
+	case WMI_SCAN_EVENTID:
+		ath11k_scan_event(ab, skb);
+		break;
+	case WMI_PEER_STA_KICKOUT_EVENTID:
+		ath11k_peer_sta_kickout_event(ab, skb);
+		break;
+	case WMI_ROAM_EVENTID:
+		ath11k_roam_event(ab, skb);
+		break;
+	case WMI_CHAN_INFO_EVENTID:
+		ath11k_chan_info_event(ab, skb);
+		break;
+	case WMI_PDEV_BSS_CHAN_INFO_EVENTID:
+		ath11k_pdev_bss_chan_info_event(ab, skb);
+		break;
+	case WMI_VDEV_INSTALL_KEY_COMPLETE_EVENTID:
+		ath11k_vdev_install_key_compl_event(ab, skb);
+		break;
+	case WMI_SERVICE_AVAILABLE_EVENTID:
+		ath11k_service_available_event(ab, skb);
+		break;
+	case WMI_PEER_ASSOC_CONF_EVENTID:
+		ath11k_peer_assoc_conf_event(ab, skb);
+		break;
+	case WMI_UPDATE_STATS_EVENTID:
+		ath11k_update_stats_event(ab, skb);
+		break;
+	case WMI_PDEV_CTL_FAILSAFE_CHECK_EVENTID:
+		ath11k_pdev_ctl_failsafe_check_event(ab, skb);
+		break;
+	case WMI_PDEV_CSA_SWITCH_COUNT_STATUS_EVENTID:
+		ath11k_wmi_pdev_csa_switch_count_status_event(ab, skb);
+		break;
+	/* add Unsupported events here */
+	case WMI_TBTTOFFSET_EXT_UPDATE_EVENTID:
+	case WMI_VDEV_DELETE_RESP_EVENTID:
+		ath11k_dbg(ab, ATH11K_DBG_WMI,
+			   "ignoring unsupported event 0x%x\n", id);
+		break;
+	case WMI_PDEV_DFS_RADAR_DETECTION_EVENTID:
+		ath11k_wmi_pdev_dfs_radar_detected_event(ab, skb);
+		break;
+	/* TODO: Add remaining events */
+	default:
+		ath11k_warn(ab, "Unknown eventid: 0x%x\n", id);
+		break;
+	}
+
+out:
+	dev_kfree_skb(skb);
+}
+
+static int ath11k_connect_pdev_htc_service(struct ath11k_base *ab,
+					   u32 pdev_idx)
+{
+	int status;
+	u32 svc_id[] = { ATH11K_HTC_SVC_ID_WMI_CONTROL,
+			 ATH11K_HTC_SVC_ID_WMI_CONTROL_MAC1,
+			 ATH11K_HTC_SVC_ID_WMI_CONTROL_MAC2 };
+
+	struct ath11k_htc_svc_conn_req conn_req;
+	struct ath11k_htc_svc_conn_resp conn_resp;
+
+	memset(&conn_req, 0, sizeof(conn_req));
+	memset(&conn_resp, 0, sizeof(conn_resp));
+
+	/* these fields are the same for all service endpoints */
+	conn_req.ep_ops.ep_tx_complete = ath11k_wmi_htc_tx_complete;
+	conn_req.ep_ops.ep_rx_complete = ath11k_wmi_tlv_op_rx;
+	conn_req.ep_ops.ep_tx_credits = ath11k_wmi_op_ep_tx_credits;
+
+	/* connect to control service */
+	conn_req.service_id = svc_id[pdev_idx];
+
+	status = ath11k_htc_connect_service(&ab->htc, &conn_req, &conn_resp);
+	if (status) {
+		ath11k_warn(ab, "failed to connect to WMI CONTROL service status: %d\n",
+			    status);
+		return status;
+	}
+
+	ab->wmi_sc.wmi_endpoint_id[pdev_idx] = conn_resp.eid;
+	ab->wmi_sc.wmi[pdev_idx].eid = conn_resp.eid;
+	ab->wmi_sc.max_msg_len[pdev_idx] = conn_resp.max_msg_len;
+
+	return 0;
+}
+
+static int
+ath11k_wmi_send_unit_test_cmd(struct ath11k *ar,
+			      struct wmi_unit_test_cmd ut_cmd,
+			      u32 *test_args)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_unit_test_cmd *cmd;
+	struct sk_buff *skb;
+	struct wmi_tlv *tlv;
+	void *ptr;
+	u32 *ut_cmd_args;
+	int buf_len, arg_len;
+	int ret;
+	int i;
+
+	arg_len = sizeof(u32) * ut_cmd.num_args;
+	buf_len = sizeof(ut_cmd) + arg_len + TLV_HDR_SIZE;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, buf_len);
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_unit_test_cmd *)skb->data;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_UNIT_TEST_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(ut_cmd) - TLV_HDR_SIZE);
+
+	cmd->vdev_id = ut_cmd.vdev_id;
+	cmd->module_id = ut_cmd.module_id;
+	cmd->num_args = ut_cmd.num_args;
+	cmd->diag_token = ut_cmd.diag_token;
+
+	ptr = skb->data + sizeof(ut_cmd);
+
+	tlv = ptr;
+	tlv->header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_ARRAY_UINT32) |
+		      FIELD_PREP(WMI_TLV_LEN, arg_len);
+
+	ptr += TLV_HDR_SIZE;
+
+	ut_cmd_args = ptr;
+	for (i = 0; i < ut_cmd.num_args; i++)
+		ut_cmd_args[i] = test_args[i];
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, WMI_UNIT_TEST_CMDID);
+
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to send WMI_UNIT_TEST CMD :%d\n",
+			    ret);
+		dev_kfree_skb(skb);
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+		   "WMI unit test : module %d vdev %d n_args %d token %d\n",
+		   cmd->module_id, cmd->vdev_id, cmd->num_args,
+		   cmd->diag_token);
+
+	return ret;
+}
+
+int ath11k_wmi_simulate_radar(struct ath11k *ar)
+{
+	struct ath11k_vif *arvif;
+	u32 dfs_args[DFS_MAX_TEST_ARGS];
+	struct wmi_unit_test_cmd wmi_ut;
+	bool arvif_found = false;
+
+	list_for_each_entry(arvif, &ar->arvifs, list) {
+		if (arvif->is_started && arvif->vdev_type == WMI_VDEV_TYPE_AP) {
+			arvif_found = true;
+			break;
+		}
+	}
+
+	if (!arvif_found)
+		return -EINVAL;
+
+	dfs_args[DFS_TEST_CMDID] = 0;
+	dfs_args[DFS_TEST_PDEV_ID] = ar->pdev->pdev_id;
+	/* Currently we could pass segment_id(b0 - b1), chirp(b2)
+	 * freq offset (b3 - b10) to unit test. For simulation
+	 * purpose this can be set to 0 which is valid.
+	 */
+	dfs_args[DFS_TEST_RADAR_PARAM] = 0;
+
+	wmi_ut.vdev_id = arvif->vdev_id;
+	wmi_ut.module_id = DFS_UNIT_TEST_MODULE;
+	wmi_ut.num_args = DFS_MAX_TEST_ARGS;
+	wmi_ut.diag_token = DFS_UNIT_TEST_TOKEN;
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_REG, "Triggering Radar Simulation\n");
+
+	return ath11k_wmi_send_unit_test_cmd(ar, wmi_ut, dfs_args);
+}
+
+int ath11k_wmi_connect(struct ath11k_base *ab)
+{
+	u32 i;
+	u8 wmi_ep_count;
+
+	wmi_ep_count = ab->htc.wmi_ep_count;
+	if (wmi_ep_count > MAX_RADIOS)
+		return -1;
+
+	for (i = 0; i < wmi_ep_count; i++)
+		ath11k_connect_pdev_htc_service(ab, i);
+
+	return 0;
+}
+
+static void ath11k_wmi_pdev_detach(struct ath11k_base *ab, u8 pdev_id)
+{
+	if (WARN_ON(pdev_id >= MAX_RADIOS))
+		return;
+
+	/* TODO: Deinit any pdev specific wmi resource */
+}
+
+int ath11k_wmi_pdev_attach(struct ath11k_base *ab,
+			   u8 pdev_id)
+{
+	struct ath11k_pdev_wmi *wmi_handle;
+
+	if (pdev_id >= MAX_RADIOS)
+		return -EINVAL;
+
+	wmi_handle = &ab->wmi_sc.wmi[pdev_id];
+
+	wmi_handle->wmi_sc = &ab->wmi_sc;
+
+	ab->wmi_sc.ab = ab;
+	/* TODO: Init remaining resource specific to pdev */
+
+	return 0;
+}
+
+int ath11k_wmi_attach(struct ath11k_base *ab)
+{
+	int ret;
+
+	ret = ath11k_wmi_pdev_attach(ab, 0);
+	if (ret)
+		return ret;
+
+	ab->wmi_sc.ab = ab;
+	ab->wmi_sc.preferred_hw_mode = WMI_HOST_HW_MODE_MAX;
+
+	/* TODO: Init remaining wmi soc resources required */
+	init_completion(&ab->wmi_sc.service_ready);
+	init_completion(&ab->wmi_sc.unified_ready);
+
+	return 0;
+}
+
+void ath11k_wmi_detach(struct ath11k_base *ab)
+{
+	int i;
+
+	/* TODO: Deinit wmi resource specific to SOC as required */
+
+	for (i = 0; i < ab->htc.wmi_ep_count; i++)
+		ath11k_wmi_pdev_detach(ab, i);
+}
diff --git a/drivers/net/wireless/ath/ath11k/wmi.h b/drivers/net/wireless/ath/ath11k/wmi.h
new file mode 100644
index 000000000000..00c51cc83d47
--- /dev/null
+++ b/drivers/net/wireless/ath/ath11k/wmi.h
@@ -0,0 +1,4664 @@
+/* SPDX-License-Identifier: BSD-3-Clause-Clear */
+/*
+ * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ */
+
+#ifndef ATH11K_WMI_H
+#define ATH11K_WMI_H
+
+#include <net/mac80211.h>
+#include "htc.h"
+
+struct ath11k_base;
+struct ath11k;
+struct ath11k_fw_stats;
+
+#define PSOC_HOST_MAX_NUM_SS (8)
+
+/* defines to set Packet extension values whic can be 0 us, 8 usec or 16 usec */
+#define MAX_HE_NSS               8
+#define MAX_HE_MODULATION        8
+#define MAX_HE_RU                4
+#define HE_MODULATION_NONE       7
+#define HE_PET_0_USEC            0
+#define HE_PET_8_USEC            1
+#define HE_PET_16_USEC           2
+
+#define WMI_MAX_NUM_SS                    MAX_HE_NSS
+#define WMI_MAX_NUM_RU                    MAX_HE_RU
+
+#define WMI_TLV_CMD(grp_id) (((grp_id) << 12) | 0x1)
+#define WMI_TLV_EV(grp_id) (((grp_id) << 12) | 0x1)
+#define WMI_TLV_CMD_UNSUPPORTED 0
+#define WMI_TLV_PDEV_PARAM_UNSUPPORTED 0
+#define WMI_TLV_VDEV_PARAM_UNSUPPORTED 0
+
+struct wmi_cmd_hdr {
+	u32 cmd_id;
+} __packed;
+
+struct wmi_tlv {
+	u32 header;
+	u8 value[0];
+} __packed;
+
+#define WMI_TLV_LEN	GENMASK(15, 0)
+#define WMI_TLV_TAG	GENMASK(31, 16)
+#define TLV_HDR_SIZE	FIELD_SIZEOF(struct wmi_tlv, header)
+
+#define WMI_CMD_HDR_CMD_ID      GENMASK(23, 0)
+#define WMI_MAX_MEM_REQS        32
+#define ATH11K_MAX_HW_LISTEN_INTERVAL 5
+
+#define WLAN_SCAN_PARAMS_MAX_SSID    16
+#define WLAN_SCAN_PARAMS_MAX_BSSID   4
+#define WLAN_SCAN_PARAMS_MAX_IE_LEN  256
+
+/*
+ * HW mode config type replicated from FW header
+ * @WMI_HOST_HW_MODE_SINGLE: Only one PHY is active.
+ * @WMI_HOST_HW_MODE_DBS: Both PHYs are active in different bands,
+ *                        one in 2G and another in 5G.
+ * @WMI_HOST_HW_MODE_SBS_PASSIVE: Both PHYs are in passive mode (only rx) in
+ *                        same band; no tx allowed.
+ * @WMI_HOST_HW_MODE_SBS: Both PHYs are active in the same band.
+ *                        Support for both PHYs within one band is planned
+ *                        for 5G only(as indicated in WMI_MAC_PHY_CAPABILITIES),
+ *                        but could be extended to other bands in the future.
+ *                        The separation of the band between the two PHYs needs
+ *                        to be communicated separately.
+ * @WMI_HOST_HW_MODE_DBS_SBS: 3 PHYs, with 2 on the same band doing SBS
+ *                           as in WMI_HW_MODE_SBS, and 3rd on the other band
+ * @WMI_HOST_HW_MODE_DBS_OR_SBS: Two PHY with one PHY capabale of both 2G and
+ *                        5G. It can support SBS (5G + 5G) OR DBS (5G + 2G).
+ * @WMI_HOST_HW_MODE_MAX: Max hw_mode_id. Used to indicate invalid mode.
+ */
+enum wmi_host_hw_mode_config_type {
+	WMI_HOST_HW_MODE_SINGLE       = 0,
+	WMI_HOST_HW_MODE_DBS          = 1,
+	WMI_HOST_HW_MODE_SBS_PASSIVE  = 2,
+	WMI_HOST_HW_MODE_SBS          = 3,
+	WMI_HOST_HW_MODE_DBS_SBS      = 4,
+	WMI_HOST_HW_MODE_DBS_OR_SBS   = 5,
+
+	/* keep last */
+	WMI_HOST_HW_MODE_MAX
+};
+
+/* HW mode priority values used to detect the preferred HW mode
+ * on the available modes.
+ */
+enum wmi_host_hw_mode_priority {
+	WMI_HOST_HW_MODE_DBS_SBS_PRI,
+	WMI_HOST_HW_MODE_DBS_PRI,
+	WMI_HOST_HW_MODE_DBS_OR_SBS_PRI,
+	WMI_HOST_HW_MODE_SBS_PRI,
+	WMI_HOST_HW_MODE_SBS_PASSIVE_PRI,
+	WMI_HOST_HW_MODE_SINGLE_PRI,
+
+	/* keep last the lowest priority */
+	WMI_HOST_HW_MODE_MAX_PRI
+};
+
+enum {
+	WMI_HOST_WLAN_2G_CAP	= 0x1,
+	WMI_HOST_WLAN_5G_CAP	= 0x2,
+	WMI_HOST_WLAN_2G_5G_CAP	= 0x3,
+};
+
+/*
+ * wmi command groups.
+ */
+enum wmi_cmd_group {
+	/* 0 to 2 are reserved */
+	WMI_GRP_START = 0x3,
+	WMI_GRP_SCAN = WMI_GRP_START,
+	WMI_GRP_PDEV		= 0x4,
+	WMI_GRP_VDEV           = 0x5,
+	WMI_GRP_PEER           = 0x6,
+	WMI_GRP_MGMT           = 0x7,
+	WMI_GRP_BA_NEG         = 0x8,
+	WMI_GRP_STA_PS         = 0x9,
+	WMI_GRP_DFS            = 0xa,
+	WMI_GRP_ROAM           = 0xb,
+	WMI_GRP_OFL_SCAN       = 0xc,
+	WMI_GRP_P2P            = 0xd,
+	WMI_GRP_AP_PS          = 0xe,
+	WMI_GRP_RATE_CTRL      = 0xf,
+	WMI_GRP_PROFILE        = 0x10,
+	WMI_GRP_SUSPEND        = 0x11,
+	WMI_GRP_BCN_FILTER     = 0x12,
+	WMI_GRP_WOW            = 0x13,
+	WMI_GRP_RTT            = 0x14,
+	WMI_GRP_SPECTRAL       = 0x15,
+	WMI_GRP_STATS          = 0x16,
+	WMI_GRP_ARP_NS_OFL     = 0x17,
+	WMI_GRP_NLO_OFL        = 0x18,
+	WMI_GRP_GTK_OFL        = 0x19,
+	WMI_GRP_CSA_OFL        = 0x1a,
+	WMI_GRP_CHATTER        = 0x1b,
+	WMI_GRP_TID_ADDBA      = 0x1c,
+	WMI_GRP_MISC           = 0x1d,
+	WMI_GRP_GPIO           = 0x1e,
+	WMI_GRP_FWTEST         = 0x1f,
+	WMI_GRP_TDLS           = 0x20,
+	WMI_GRP_RESMGR         = 0x21,
+	WMI_GRP_STA_SMPS       = 0x22,
+	WMI_GRP_WLAN_HB        = 0x23,
+	WMI_GRP_RMC            = 0x24,
+	WMI_GRP_MHF_OFL        = 0x25,
+	WMI_GRP_LOCATION_SCAN  = 0x26,
+	WMI_GRP_OEM            = 0x27,
+	WMI_GRP_NAN            = 0x28,
+	WMI_GRP_COEX           = 0x29,
+	WMI_GRP_OBSS_OFL       = 0x2a,
+	WMI_GRP_LPI            = 0x2b,
+	WMI_GRP_EXTSCAN        = 0x2c,
+	WMI_GRP_DHCP_OFL       = 0x2d,
+	WMI_GRP_IPA            = 0x2e,
+	WMI_GRP_MDNS_OFL       = 0x2f,
+	WMI_GRP_SAP_OFL        = 0x30,
+	WMI_GRP_OCB            = 0x31,
+	WMI_GRP_SOC            = 0x32,
+	WMI_GRP_PKT_FILTER     = 0x33,
+	WMI_GRP_MAWC           = 0x34,
+	WMI_GRP_PMF_OFFLOAD    = 0x35,
+	WMI_GRP_BPF_OFFLOAD    = 0x36,
+	WMI_GRP_NAN_DATA       = 0x37,
+	WMI_GRP_PROTOTYPE      = 0x38,
+	WMI_GRP_MONITOR        = 0x39,
+	WMI_GRP_REGULATORY     = 0x3a,
+	WMI_GRP_HW_DATA_FILTER = 0x3b,
+};
+
+#define WMI_CMD_GRP(grp_id) (((grp_id) << 12) | 0x1)
+#define WMI_EVT_GRP_START_ID(grp_id) (((grp_id) << 12) | 0x1)
+
+#define WMI_CMD_UNSUPPORTED 0
+
+enum wmi_tlv_cmd_id {
+	WMI_INIT_CMDID = 0x1,
+	WMI_START_SCAN_CMDID = WMI_TLV_CMD(WMI_GRP_SCAN),
+	WMI_STOP_SCAN_CMDID,
+	WMI_SCAN_CHAN_LIST_CMDID,
+	WMI_SCAN_SCH_PRIO_TBL_CMDID,
+	WMI_SCAN_UPDATE_REQUEST_CMDID,
+	WMI_SCAN_PROB_REQ_OUI_CMDID,
+	WMI_SCAN_ADAPTIVE_DWELL_CONFIG_CMDID,
+	WMI_PDEV_SET_REGDOMAIN_CMDID = WMI_TLV_CMD(WMI_GRP_PDEV),
+	WMI_PDEV_SET_CHANNEL_CMDID,
+	WMI_PDEV_SET_PARAM_CMDID,
+	WMI_PDEV_PKTLOG_ENABLE_CMDID,
+	WMI_PDEV_PKTLOG_DISABLE_CMDID,
+	WMI_PDEV_SET_WMM_PARAMS_CMDID,
+	WMI_PDEV_SET_HT_CAP_IE_CMDID,
+	WMI_PDEV_SET_VHT_CAP_IE_CMDID,
+	WMI_PDEV_SET_DSCP_TID_MAP_CMDID,
+	WMI_PDEV_SET_QUIET_MODE_CMDID,
+	WMI_PDEV_GREEN_AP_PS_ENABLE_CMDID,
+	WMI_PDEV_GET_TPC_CONFIG_CMDID,
+	WMI_PDEV_SET_BASE_MACADDR_CMDID,
+	WMI_PDEV_DUMP_CMDID,
+	WMI_PDEV_SET_LED_CONFIG_CMDID,
+	WMI_PDEV_GET_TEMPERATURE_CMDID,
+	WMI_PDEV_SET_LED_FLASHING_CMDID,
+	WMI_PDEV_SMART_ANT_ENABLE_CMDID,
+	WMI_PDEV_SMART_ANT_SET_RX_ANTENNA_CMDID,
+	WMI_PDEV_SET_ANTENNA_SWITCH_TABLE_CMDID,
+	WMI_PDEV_SET_CTL_TABLE_CMDID,
+	WMI_PDEV_SET_MIMOGAIN_TABLE_CMDID,
+	WMI_PDEV_FIPS_CMDID,
+	WMI_PDEV_GET_ANI_CCK_CONFIG_CMDID,
+	WMI_PDEV_GET_ANI_OFDM_CONFIG_CMDID,
+	WMI_PDEV_GET_NFCAL_POWER_CMDID,
+	WMI_PDEV_GET_TPC_CMDID,
+	WMI_MIB_STATS_ENABLE_CMDID,
+	WMI_PDEV_SET_PCL_CMDID,
+	WMI_PDEV_SET_HW_MODE_CMDID,
+	WMI_PDEV_SET_MAC_CONFIG_CMDID,
+	WMI_PDEV_SET_ANTENNA_MODE_CMDID,
+	WMI_SET_PERIODIC_CHANNEL_STATS_CONFIG_CMDID,
+	WMI_PDEV_WAL_POWER_DEBUG_CMDID,
+	WMI_PDEV_SET_REORDER_TIMEOUT_VAL_CMDID,
+	WMI_PDEV_SET_WAKEUP_CONFIG_CMDID,
+	WMI_PDEV_GET_ANTDIV_STATUS_CMDID,
+	WMI_PDEV_GET_CHIP_POWER_STATS_CMDID,
+	WMI_PDEV_SET_STATS_THRESHOLD_CMDID,
+	WMI_PDEV_MULTIPLE_VDEV_RESTART_REQUEST_CMDID,
+	WMI_PDEV_UPDATE_PKT_ROUTING_CMDID,
+	WMI_PDEV_CHECK_CAL_VERSION_CMDID,
+	WMI_PDEV_SET_DIVERSITY_GAIN_CMDID,
+	WMI_PDEV_DIV_GET_RSSI_ANTID_CMDID,
+	WMI_PDEV_BSS_CHAN_INFO_REQUEST_CMDID,
+	WMI_PDEV_UPDATE_PMK_CACHE_CMDID,
+	WMI_PDEV_UPDATE_FILS_HLP_PKT_CMDID,
+	WMI_PDEV_UPDATE_CTLTABLE_REQUEST_CMDID,
+	WMI_PDEV_CONFIG_VENDOR_OUI_ACTION_CMDID,
+	WMI_PDEV_SET_AC_TX_QUEUE_OPTIMIZED_CMDID,
+	WMI_PDEV_SET_RX_FILTER_PROMISCUOUS_CMDID,
+	WMI_PDEV_DMA_RING_CFG_REQ_CMDID,
+	WMI_PDEV_HE_TB_ACTION_FRM_CMDID,
+	WMI_PDEV_PKTLOG_FILTER_CMDID,
+	WMI_VDEV_CREATE_CMDID = WMI_TLV_CMD(WMI_GRP_VDEV),
+	WMI_VDEV_DELETE_CMDID,
+	WMI_VDEV_START_REQUEST_CMDID,
+	WMI_VDEV_RESTART_REQUEST_CMDID,
+	WMI_VDEV_UP_CMDID,
+	WMI_VDEV_STOP_CMDID,
+	WMI_VDEV_DOWN_CMDID,
+	WMI_VDEV_SET_PARAM_CMDID,
+	WMI_VDEV_INSTALL_KEY_CMDID,
+	WMI_VDEV_WNM_SLEEPMODE_CMDID,
+	WMI_VDEV_WMM_ADDTS_CMDID,
+	WMI_VDEV_WMM_DELTS_CMDID,
+	WMI_VDEV_SET_WMM_PARAMS_CMDID,
+	WMI_VDEV_SET_GTX_PARAMS_CMDID,
+	WMI_VDEV_IPSEC_NATKEEPALIVE_FILTER_CMDID,
+	WMI_VDEV_PLMREQ_START_CMDID,
+	WMI_VDEV_PLMREQ_STOP_CMDID,
+	WMI_VDEV_TSF_TSTAMP_ACTION_CMDID,
+	WMI_VDEV_SET_IE_CMDID,
+	WMI_VDEV_RATEMASK_CMDID,
+	WMI_VDEV_ATF_REQUEST_CMDID,
+	WMI_VDEV_SET_DSCP_TID_MAP_CMDID,
+	WMI_VDEV_FILTER_NEIGHBOR_RX_PACKETS_CMDID,
+	WMI_VDEV_SET_QUIET_MODE_CMDID,
+	WMI_VDEV_SET_CUSTOM_AGGR_SIZE_CMDID,
+	WMI_VDEV_ENCRYPT_DECRYPT_DATA_REQ_CMDID,
+	WMI_VDEV_ADD_MAC_ADDR_TO_RX_FILTER_CMDID,
+	WMI_PEER_CREATE_CMDID = WMI_TLV_CMD(WMI_GRP_PEER),
+	WMI_PEER_DELETE_CMDID,
+	WMI_PEER_FLUSH_TIDS_CMDID,
+	WMI_PEER_SET_PARAM_CMDID,
+	WMI_PEER_ASSOC_CMDID,
+	WMI_PEER_ADD_WDS_ENTRY_CMDID,
+	WMI_PEER_REMOVE_WDS_ENTRY_CMDID,
+	WMI_PEER_MCAST_GROUP_CMDID,
+	WMI_PEER_INFO_REQ_CMDID,
+	WMI_PEER_GET_ESTIMATED_LINKSPEED_CMDID,
+	WMI_PEER_SET_RATE_REPORT_CONDITION_CMDID,
+	WMI_PEER_UPDATE_WDS_ENTRY_CMDID,
+	WMI_PEER_ADD_PROXY_STA_ENTRY_CMDID,
+	WMI_PEER_SMART_ANT_SET_TX_ANTENNA_CMDID,
+	WMI_PEER_SMART_ANT_SET_TRAIN_INFO_CMDID,
+	WMI_PEER_SMART_ANT_SET_NODE_CONFIG_OPS_CMDID,
+	WMI_PEER_ATF_REQUEST_CMDID,
+	WMI_PEER_BWF_REQUEST_CMDID,
+	WMI_PEER_REORDER_QUEUE_SETUP_CMDID,
+	WMI_PEER_REORDER_QUEUE_REMOVE_CMDID,
+	WMI_PEER_SET_RX_BLOCKSIZE_CMDID,
+	WMI_PEER_ANTDIV_INFO_REQ_CMDID,
+	WMI_PEER_OPER_MODE_CHANGE_EVENTID,
+	WMI_BCN_TX_CMDID = WMI_TLV_CMD(WMI_GRP_MGMT),
+	WMI_PDEV_SEND_BCN_CMDID,
+	WMI_BCN_TMPL_CMDID,
+	WMI_BCN_FILTER_RX_CMDID,
+	WMI_PRB_REQ_FILTER_RX_CMDID,
+	WMI_MGMT_TX_CMDID,
+	WMI_PRB_TMPL_CMDID,
+	WMI_MGMT_TX_SEND_CMDID,
+	WMI_OFFCHAN_DATA_TX_SEND_CMDID,
+	WMI_PDEV_SEND_FD_CMDID,
+	WMI_BCN_OFFLOAD_CTRL_CMDID,
+	WMI_BSS_COLOR_CHANGE_ENABLE_CMDID,
+	WMI_VDEV_BCN_OFFLOAD_QUIET_CONFIG_CMDID,
+	WMI_ADDBA_CLEAR_RESP_CMDID = WMI_TLV_CMD(WMI_GRP_BA_NEG),
+	WMI_ADDBA_SEND_CMDID,
+	WMI_ADDBA_STATUS_CMDID,
+	WMI_DELBA_SEND_CMDID,
+	WMI_ADDBA_SET_RESP_CMDID,
+	WMI_SEND_SINGLEAMSDU_CMDID,
+	WMI_STA_POWERSAVE_MODE_CMDID = WMI_TLV_CMD(WMI_GRP_STA_PS),
+	WMI_STA_POWERSAVE_PARAM_CMDID,
+	WMI_STA_MIMO_PS_MODE_CMDID,
+	WMI_PDEV_DFS_ENABLE_CMDID = WMI_TLV_CMD(WMI_GRP_DFS),
+	WMI_PDEV_DFS_DISABLE_CMDID,
+	WMI_DFS_PHYERR_FILTER_ENA_CMDID,
+	WMI_DFS_PHYERR_FILTER_DIS_CMDID,
+	WMI_PDEV_DFS_PHYERR_OFFLOAD_ENABLE_CMDID,
+	WMI_PDEV_DFS_PHYERR_OFFLOAD_DISABLE_CMDID,
+	WMI_VDEV_ADFS_CH_CFG_CMDID,
+	WMI_VDEV_ADFS_OCAC_ABORT_CMDID,
+	WMI_ROAM_SCAN_MODE = WMI_TLV_CMD(WMI_GRP_ROAM),
+	WMI_ROAM_SCAN_RSSI_THRESHOLD,
+	WMI_ROAM_SCAN_PERIOD,
+	WMI_ROAM_SCAN_RSSI_CHANGE_THRESHOLD,
+	WMI_ROAM_AP_PROFILE,
+	WMI_ROAM_CHAN_LIST,
+	WMI_ROAM_SCAN_CMD,
+	WMI_ROAM_SYNCH_COMPLETE,
+	WMI_ROAM_SET_RIC_REQUEST_CMDID,
+	WMI_ROAM_INVOKE_CMDID,
+	WMI_ROAM_FILTER_CMDID,
+	WMI_ROAM_SUBNET_CHANGE_CONFIG_CMDID,
+	WMI_ROAM_CONFIGURE_MAWC_CMDID,
+	WMI_ROAM_SET_MBO_PARAM_CMDID,
+	WMI_ROAM_PER_CONFIG_CMDID,
+	WMI_OFL_SCAN_ADD_AP_PROFILE = WMI_TLV_CMD(WMI_GRP_OFL_SCAN),
+	WMI_OFL_SCAN_REMOVE_AP_PROFILE,
+	WMI_OFL_SCAN_PERIOD,
+	WMI_P2P_DEV_SET_DEVICE_INFO = WMI_TLV_CMD(WMI_GRP_P2P),
+	WMI_P2P_DEV_SET_DISCOVERABILITY,
+	WMI_P2P_GO_SET_BEACON_IE,
+	WMI_P2P_GO_SET_PROBE_RESP_IE,
+	WMI_P2P_SET_VENDOR_IE_DATA_CMDID,
+	WMI_P2P_DISC_OFFLOAD_CONFIG_CMDID,
+	WMI_P2P_DISC_OFFLOAD_APPIE_CMDID,
+	WMI_P2P_DISC_OFFLOAD_PATTERN_CMDID,
+	WMI_P2P_SET_OPPPS_PARAM_CMDID,
+	WMI_P2P_LISTEN_OFFLOAD_START_CMDID,
+	WMI_P2P_LISTEN_OFFLOAD_STOP_CMDID,
+	WMI_AP_PS_PEER_PARAM_CMDID = WMI_TLV_CMD(WMI_GRP_AP_PS),
+	WMI_AP_PS_PEER_UAPSD_COEX_CMDID,
+	WMI_AP_PS_EGAP_PARAM_CMDID,
+	WMI_PEER_RATE_RETRY_SCHED_CMDID = WMI_TLV_CMD(WMI_GRP_RATE_CTRL),
+	WMI_WLAN_PROFILE_TRIGGER_CMDID = WMI_TLV_CMD(WMI_GRP_PROFILE),
+	WMI_WLAN_PROFILE_SET_HIST_INTVL_CMDID,
+	WMI_WLAN_PROFILE_GET_PROFILE_DATA_CMDID,
+	WMI_WLAN_PROFILE_ENABLE_PROFILE_ID_CMDID,
+	WMI_WLAN_PROFILE_LIST_PROFILE_ID_CMDID,
+	WMI_PDEV_SUSPEND_CMDID = WMI_TLV_CMD(WMI_GRP_SUSPEND),
+	WMI_PDEV_RESUME_CMDID,
+	WMI_ADD_BCN_FILTER_CMDID = WMI_TLV_CMD(WMI_GRP_BCN_FILTER),
+	WMI_RMV_BCN_FILTER_CMDID,
+	WMI_WOW_ADD_WAKE_PATTERN_CMDID = WMI_TLV_CMD(WMI_GRP_WOW),
+	WMI_WOW_DEL_WAKE_PATTERN_CMDID,
+	WMI_WOW_ENABLE_DISABLE_WAKE_EVENT_CMDID,
+	WMI_WOW_ENABLE_CMDID,
+	WMI_WOW_HOSTWAKEUP_FROM_SLEEP_CMDID,
+	WMI_WOW_IOAC_ADD_KEEPALIVE_CMDID,
+	WMI_WOW_IOAC_DEL_KEEPALIVE_CMDID,
+	WMI_WOW_IOAC_ADD_WAKE_PATTERN_CMDID,
+	WMI_WOW_IOAC_DEL_WAKE_PATTERN_CMDID,
+	WMI_D0_WOW_ENABLE_DISABLE_CMDID,
+	WMI_EXTWOW_ENABLE_CMDID,
+	WMI_EXTWOW_SET_APP_TYPE1_PARAMS_CMDID,
+	WMI_EXTWOW_SET_APP_TYPE2_PARAMS_CMDID,
+	WMI_WOW_ENABLE_ICMPV6_NA_FLT_CMDID,
+	WMI_WOW_UDP_SVC_OFLD_CMDID,
+	WMI_WOW_HOSTWAKEUP_GPIO_PIN_PATTERN_CONFIG_CMDID,
+	WMI_WOW_SET_ACTION_WAKE_UP_CMDID,
+	WMI_RTT_MEASREQ_CMDID = WMI_TLV_CMD(WMI_GRP_RTT),
+	WMI_RTT_TSF_CMDID,
+	WMI_VDEV_SPECTRAL_SCAN_CONFIGURE_CMDID = WMI_TLV_CMD(WMI_GRP_SPECTRAL),
+	WMI_VDEV_SPECTRAL_SCAN_ENABLE_CMDID,
+	WMI_REQUEST_STATS_CMDID = WMI_TLV_CMD(WMI_GRP_STATS),
+	WMI_MCC_SCHED_TRAFFIC_STATS_CMDID,
+	WMI_REQUEST_STATS_EXT_CMDID,
+	WMI_REQUEST_LINK_STATS_CMDID,
+	WMI_START_LINK_STATS_CMDID,
+	WMI_CLEAR_LINK_STATS_CMDID,
+	WMI_GET_FW_MEM_DUMP_CMDID,
+	WMI_DEBUG_MESG_FLUSH_CMDID,
+	WMI_DIAG_EVENT_LOG_CONFIG_CMDID,
+	WMI_REQUEST_WLAN_STATS_CMDID,
+	WMI_REQUEST_RCPI_CMDID,
+	WMI_REQUEST_PEER_STATS_INFO_CMDID,
+	WMI_REQUEST_RADIO_CHAN_STATS_CMDID,
+	WMI_SET_ARP_NS_OFFLOAD_CMDID = WMI_TLV_CMD(WMI_GRP_ARP_NS_OFL),
+	WMI_ADD_PROACTIVE_ARP_RSP_PATTERN_CMDID,
+	WMI_DEL_PROACTIVE_ARP_RSP_PATTERN_CMDID,
+	WMI_NETWORK_LIST_OFFLOAD_CONFIG_CMDID = WMI_TLV_CMD(WMI_GRP_NLO_OFL),
+	WMI_APFIND_CMDID,
+	WMI_PASSPOINT_LIST_CONFIG_CMDID,
+	WMI_NLO_CONFIGURE_MAWC_CMDID,
+	WMI_GTK_OFFLOAD_CMDID = WMI_TLV_CMD(WMI_GRP_GTK_OFL),
+	WMI_CSA_OFFLOAD_ENABLE_CMDID = WMI_TLV_CMD(WMI_GRP_CSA_OFL),
+	WMI_CSA_OFFLOAD_CHANSWITCH_CMDID,
+	WMI_CHATTER_SET_MODE_CMDID = WMI_TLV_CMD(WMI_GRP_CHATTER),
+	WMI_CHATTER_ADD_COALESCING_FILTER_CMDID,
+	WMI_CHATTER_DELETE_COALESCING_FILTER_CMDID,
+	WMI_CHATTER_COALESCING_QUERY_CMDID,
+	WMI_PEER_TID_ADDBA_CMDID = WMI_TLV_CMD(WMI_GRP_TID_ADDBA),
+	WMI_PEER_TID_DELBA_CMDID,
+	WMI_STA_DTIM_PS_METHOD_CMDID,
+	WMI_STA_UAPSD_AUTO_TRIG_CMDID,
+	WMI_STA_KEEPALIVE_CMDID,
+	WMI_BA_REQ_SSN_CMDID,
+	WMI_ECHO_CMDID = WMI_TLV_CMD(WMI_GRP_MISC),
+	WMI_PDEV_UTF_CMDID,
+	WMI_DBGLOG_CFG_CMDID,
+	WMI_PDEV_QVIT_CMDID,
+	WMI_PDEV_FTM_INTG_CMDID,
+	WMI_VDEV_SET_KEEPALIVE_CMDID,
+	WMI_VDEV_GET_KEEPALIVE_CMDID,
+	WMI_FORCE_FW_HANG_CMDID,
+	WMI_SET_MCASTBCAST_FILTER_CMDID,
+	WMI_THERMAL_MGMT_CMDID,
+	WMI_HOST_AUTO_SHUTDOWN_CFG_CMDID,
+	WMI_TPC_CHAINMASK_CONFIG_CMDID,
+	WMI_SET_ANTENNA_DIVERSITY_CMDID,
+	WMI_OCB_SET_SCHED_CMDID,
+	WMI_RSSI_BREACH_MONITOR_CONFIG_CMDID,
+	WMI_LRO_CONFIG_CMDID,
+	WMI_TRANSFER_DATA_TO_FLASH_CMDID,
+	WMI_CONFIG_ENHANCED_MCAST_FILTER_CMDID,
+	WMI_VDEV_WISA_CMDID,
+	WMI_DBGLOG_TIME_STAMP_SYNC_CMDID,
+	WMI_SET_MULTIPLE_MCAST_FILTER_CMDID,
+	WMI_READ_DATA_FROM_FLASH_CMDID,
+	WMI_GPIO_CONFIG_CMDID = WMI_TLV_CMD(WMI_GRP_GPIO),
+	WMI_GPIO_OUTPUT_CMDID,
+	WMI_TXBF_CMDID,
+	WMI_FWTEST_VDEV_MCC_SET_TBTT_MODE_CMDID = WMI_TLV_CMD(WMI_GRP_FWTEST),
+	WMI_FWTEST_P2P_SET_NOA_PARAM_CMDID,
+	WMI_UNIT_TEST_CMDID,
+	WMI_FWTEST_CMDID,
+	WMI_QBOOST_CFG_CMDID,
+	WMI_TDLS_SET_STATE_CMDID = WMI_TLV_CMD(WMI_GRP_TDLS),
+	WMI_TDLS_PEER_UPDATE_CMDID,
+	WMI_TDLS_SET_OFFCHAN_MODE_CMDID,
+	WMI_RESMGR_ADAPTIVE_OCS_EN_DIS_CMDID = WMI_TLV_CMD(WMI_GRP_RESMGR),
+	WMI_RESMGR_SET_CHAN_TIME_QUOTA_CMDID,
+	WMI_RESMGR_SET_CHAN_LATENCY_CMDID,
+	WMI_STA_SMPS_FORCE_MODE_CMDID = WMI_TLV_CMD(WMI_GRP_STA_SMPS),
+	WMI_STA_SMPS_PARAM_CMDID,
+	WMI_HB_SET_ENABLE_CMDID = WMI_TLV_CMD(WMI_GRP_WLAN_HB),
+	WMI_HB_SET_TCP_PARAMS_CMDID,
+	WMI_HB_SET_TCP_PKT_FILTER_CMDID,
+	WMI_HB_SET_UDP_PARAMS_CMDID,
+	WMI_HB_SET_UDP_PKT_FILTER_CMDID,
+	WMI_RMC_SET_MODE_CMDID = WMI_TLV_CMD(WMI_GRP_RMC),
+	WMI_RMC_SET_ACTION_PERIOD_CMDID,
+	WMI_RMC_CONFIG_CMDID,
+	WMI_RMC_SET_MANUAL_LEADER_CMDID,
+	WMI_MHF_OFFLOAD_SET_MODE_CMDID = WMI_TLV_CMD(WMI_GRP_MHF_OFL),
+	WMI_MHF_OFFLOAD_PLUMB_ROUTING_TBL_CMDID,
+	WMI_BATCH_SCAN_ENABLE_CMDID = WMI_TLV_CMD(WMI_GRP_LOCATION_SCAN),
+	WMI_BATCH_SCAN_DISABLE_CMDID,
+	WMI_BATCH_SCAN_TRIGGER_RESULT_CMDID,
+	WMI_OEM_REQ_CMDID = WMI_TLV_CMD(WMI_GRP_OEM),
+	WMI_OEM_REQUEST_CMDID,
+	WMI_LPI_OEM_REQ_CMDID,
+	WMI_NAN_CMDID = WMI_TLV_CMD(WMI_GRP_NAN),
+	WMI_MODEM_POWER_STATE_CMDID = WMI_TLV_CMD(WMI_GRP_COEX),
+	WMI_CHAN_AVOID_UPDATE_CMDID,
+	WMI_COEX_CONFIG_CMDID,
+	WMI_CHAN_AVOID_RPT_ALLOW_CMDID,
+	WMI_COEX_GET_ANTENNA_ISOLATION_CMDID,
+	WMI_SAR_LIMITS_CMDID,
+	WMI_OBSS_SCAN_ENABLE_CMDID = WMI_TLV_CMD(WMI_GRP_OBSS_OFL),
+	WMI_OBSS_SCAN_DISABLE_CMDID,
+	WMI_LPI_MGMT_SNOOPING_CONFIG_CMDID = WMI_TLV_CMD(WMI_GRP_LPI),
+	WMI_LPI_START_SCAN_CMDID,
+	WMI_LPI_STOP_SCAN_CMDID,
+	WMI_EXTSCAN_START_CMDID = WMI_TLV_CMD(WMI_GRP_EXTSCAN),
+	WMI_EXTSCAN_STOP_CMDID,
+	WMI_EXTSCAN_CONFIGURE_WLAN_CHANGE_MONITOR_CMDID,
+	WMI_EXTSCAN_CONFIGURE_HOTLIST_MONITOR_CMDID,
+	WMI_EXTSCAN_GET_CACHED_RESULTS_CMDID,
+	WMI_EXTSCAN_GET_WLAN_CHANGE_RESULTS_CMDID,
+	WMI_EXTSCAN_SET_CAPABILITIES_CMDID,
+	WMI_EXTSCAN_GET_CAPABILITIES_CMDID,
+	WMI_EXTSCAN_CONFIGURE_HOTLIST_SSID_MONITOR_CMDID,
+	WMI_EXTSCAN_CONFIGURE_MAWC_CMDID,
+	WMI_SET_DHCP_SERVER_OFFLOAD_CMDID = WMI_TLV_CMD(WMI_GRP_DHCP_OFL),
+	WMI_IPA_OFFLOAD_ENABLE_DISABLE_CMDID = WMI_TLV_CMD(WMI_GRP_IPA),
+	WMI_MDNS_OFFLOAD_ENABLE_CMDID = WMI_TLV_CMD(WMI_GRP_MDNS_OFL),
+	WMI_MDNS_SET_FQDN_CMDID,
+	WMI_MDNS_SET_RESPONSE_CMDID,
+	WMI_MDNS_GET_STATS_CMDID,
+	WMI_SAP_OFL_ENABLE_CMDID = WMI_TLV_CMD(WMI_GRP_SAP_OFL),
+	WMI_SAP_SET_BLACKLIST_PARAM_CMDID,
+	WMI_OCB_SET_CONFIG_CMDID = WMI_TLV_CMD(WMI_GRP_OCB),
+	WMI_OCB_SET_UTC_TIME_CMDID,
+	WMI_OCB_START_TIMING_ADVERT_CMDID,
+	WMI_OCB_STOP_TIMING_ADVERT_CMDID,
+	WMI_OCB_GET_TSF_TIMER_CMDID,
+	WMI_DCC_GET_STATS_CMDID,
+	WMI_DCC_CLEAR_STATS_CMDID,
+	WMI_DCC_UPDATE_NDL_CMDID,
+	WMI_SOC_SET_PCL_CMDID = WMI_TLV_CMD(WMI_GRP_SOC),
+	WMI_SOC_SET_HW_MODE_CMDID,
+	WMI_SOC_SET_DUAL_MAC_CONFIG_CMDID,
+	WMI_SOC_SET_ANTENNA_MODE_CMDID,
+	WMI_PACKET_FILTER_CONFIG_CMDID = WMI_TLV_CMD(WMI_GRP_PKT_FILTER),
+	WMI_PACKET_FILTER_ENABLE_CMDID,
+	WMI_MAWC_SENSOR_REPORT_IND_CMDID = WMI_TLV_CMD(WMI_GRP_MAWC),
+	WMI_PMF_OFFLOAD_SET_SA_QUERY_CMDID = WMI_TLV_CMD(WMI_GRP_PMF_OFFLOAD),
+	WMI_BPF_GET_CAPABILITY_CMDID = WMI_TLV_CMD(WMI_GRP_BPF_OFFLOAD),
+	WMI_BPF_GET_VDEV_STATS_CMDID,
+	WMI_BPF_SET_VDEV_INSTRUCTIONS_CMDID,
+	WMI_BPF_DEL_VDEV_INSTRUCTIONS_CMDID,
+	WMI_BPF_SET_VDEV_ACTIVE_MODE_CMDID,
+	WMI_MNT_FILTER_CMDID = WMI_TLV_CMD(WMI_GRP_MONITOR),
+	WMI_SET_CURRENT_COUNTRY_CMDID = WMI_TLV_CMD(WMI_GRP_REGULATORY),
+	WMI_11D_SCAN_START_CMDID,
+	WMI_11D_SCAN_STOP_CMDID,
+	WMI_SET_INIT_COUNTRY_CMDID,
+	WMI_NDI_GET_CAP_REQ_CMDID = WMI_TLV_CMD(WMI_GRP_PROTOTYPE),
+	WMI_NDP_INITIATOR_REQ_CMDID,
+	WMI_NDP_RESPONDER_REQ_CMDID,
+	WMI_NDP_END_REQ_CMDID,
+	WMI_HW_DATA_FILTER_CMDID = WMI_TLV_CMD(WMI_GRP_HW_DATA_FILTER),
+};
+
+enum wmi_tlv_event_id {
+	WMI_SERVICE_READY_EVENTID = 0x1,
+	WMI_READY_EVENTID,
+	WMI_SERVICE_AVAILABLE_EVENTID,
+	WMI_SCAN_EVENTID = WMI_EVT_GRP_START_ID(WMI_GRP_SCAN),
+	WMI_PDEV_TPC_CONFIG_EVENTID = WMI_TLV_CMD(WMI_GRP_PDEV),
+	WMI_CHAN_INFO_EVENTID,
+	WMI_PHYERR_EVENTID,
+	WMI_PDEV_DUMP_EVENTID,
+	WMI_TX_PAUSE_EVENTID,
+	WMI_DFS_RADAR_EVENTID,
+	WMI_PDEV_L1SS_TRACK_EVENTID,
+	WMI_PDEV_TEMPERATURE_EVENTID,
+	WMI_SERVICE_READY_EXT_EVENTID,
+	WMI_PDEV_FIPS_EVENTID,
+	WMI_PDEV_CHANNEL_HOPPING_EVENTID,
+	WMI_PDEV_ANI_CCK_LEVEL_EVENTID,
+	WMI_PDEV_ANI_OFDM_LEVEL_EVENTID,
+	WMI_PDEV_TPC_EVENTID,
+	WMI_PDEV_NFCAL_POWER_ALL_CHANNELS_EVENTID,
+	WMI_PDEV_SET_HW_MODE_RESP_EVENTID,
+	WMI_PDEV_HW_MODE_TRANSITION_EVENTID,
+	WMI_PDEV_SET_MAC_CONFIG_RESP_EVENTID,
+	WMI_PDEV_ANTDIV_STATUS_EVENTID,
+	WMI_PDEV_CHIP_POWER_STATS_EVENTID,
+	WMI_PDEV_CHIP_POWER_SAVE_FAILURE_DETECTED_EVENTID,
+	WMI_PDEV_CSA_SWITCH_COUNT_STATUS_EVENTID,
+	WMI_PDEV_CHECK_CAL_VERSION_EVENTID,
+	WMI_PDEV_DIV_RSSI_ANTID_EVENTID,
+	WMI_PDEV_BSS_CHAN_INFO_EVENTID,
+	WMI_PDEV_UPDATE_CTLTABLE_EVENTID,
+	WMI_PDEV_DMA_RING_CFG_RSP_EVENTID,
+	WMI_PDEV_DMA_RING_BUF_RELEASE_EVENTID,
+	WMI_PDEV_CTL_FAILSAFE_CHECK_EVENTID,
+	WMI_VDEV_START_RESP_EVENTID = WMI_TLV_CMD(WMI_GRP_VDEV),
+	WMI_VDEV_STOPPED_EVENTID,
+	WMI_VDEV_INSTALL_KEY_COMPLETE_EVENTID,
+	WMI_VDEV_MCC_BCN_INTERVAL_CHANGE_REQ_EVENTID,
+	WMI_VDEV_TSF_REPORT_EVENTID,
+	WMI_VDEV_DELETE_RESP_EVENTID,
+	WMI_VDEV_ENCRYPT_DECRYPT_DATA_RESP_EVENTID,
+	WMI_VDEV_ADD_MAC_ADDR_TO_RX_FILTER_STATUS_EVENTID,
+	WMI_PEER_STA_KICKOUT_EVENTID = WMI_TLV_CMD(WMI_GRP_PEER),
+	WMI_PEER_INFO_EVENTID,
+	WMI_PEER_TX_FAIL_CNT_THR_EVENTID,
+	WMI_PEER_ESTIMATED_LINKSPEED_EVENTID,
+	WMI_PEER_STATE_EVENTID,
+	WMI_PEER_ASSOC_CONF_EVENTID,
+	WMI_PEER_DELETE_RESP_EVENTID,
+	WMI_PEER_RATECODE_LIST_EVENTID,
+	WMI_WDS_PEER_EVENTID,
+	WMI_PEER_STA_PS_STATECHG_EVENTID,
+	WMI_PEER_ANTDIV_INFO_EVENTID,
+	WMI_MGMT_RX_EVENTID = WMI_TLV_CMD(WMI_GRP_MGMT),
+	WMI_HOST_SWBA_EVENTID,
+	WMI_TBTTOFFSET_UPDATE_EVENTID,
+	WMI_OFFLOAD_BCN_TX_STATUS_EVENTID,
+	WMI_OFFLOAD_PROB_RESP_TX_STATUS_EVENTID,
+	WMI_MGMT_TX_COMPLETION_EVENTID,
+	WMI_MGMT_TX_BUNDLE_COMPLETION_EVENTID,
+	WMI_TBTTOFFSET_EXT_UPDATE_EVENTID,
+	WMI_TX_DELBA_COMPLETE_EVENTID = WMI_TLV_CMD(WMI_GRP_BA_NEG),
+	WMI_TX_ADDBA_COMPLETE_EVENTID,
+	WMI_BA_RSP_SSN_EVENTID,
+	WMI_AGGR_STATE_TRIG_EVENTID,
+	WMI_ROAM_EVENTID = WMI_TLV_CMD(WMI_GRP_ROAM),
+	WMI_PROFILE_MATCH,
+	WMI_ROAM_SYNCH_EVENTID,
+	WMI_P2P_DISC_EVENTID = WMI_TLV_CMD(WMI_GRP_P2P),
+	WMI_P2P_NOA_EVENTID,
+	WMI_P2P_LISTEN_OFFLOAD_STOPPED_EVENTID,
+	WMI_AP_PS_EGAP_INFO_EVENTID = WMI_TLV_CMD(WMI_GRP_AP_PS),
+	WMI_PDEV_RESUME_EVENTID = WMI_TLV_CMD(WMI_GRP_SUSPEND),
+	WMI_WOW_WAKEUP_HOST_EVENTID = WMI_TLV_CMD(WMI_GRP_WOW),
+	WMI_D0_WOW_DISABLE_ACK_EVENTID,
+	WMI_WOW_INITIAL_WAKEUP_EVENTID,
+	WMI_RTT_MEASUREMENT_REPORT_EVENTID = WMI_TLV_CMD(WMI_GRP_RTT),
+	WMI_TSF_MEASUREMENT_REPORT_EVENTID,
+	WMI_RTT_ERROR_REPORT_EVENTID,
+	WMI_STATS_EXT_EVENTID = WMI_TLV_CMD(WMI_GRP_STATS),
+	WMI_IFACE_LINK_STATS_EVENTID,
+	WMI_PEER_LINK_STATS_EVENTID,
+	WMI_RADIO_LINK_STATS_EVENTID,
+	WMI_UPDATE_FW_MEM_DUMP_EVENTID,
+	WMI_DIAG_EVENT_LOG_SUPPORTED_EVENTID,
+	WMI_INST_RSSI_STATS_EVENTID,
+	WMI_RADIO_TX_POWER_LEVEL_STATS_EVENTID,
+	WMI_REPORT_STATS_EVENTID,
+	WMI_UPDATE_RCPI_EVENTID,
+	WMI_PEER_STATS_INFO_EVENTID,
+	WMI_RADIO_CHAN_STATS_EVENTID,
+	WMI_NLO_MATCH_EVENTID = WMI_TLV_CMD(WMI_GRP_NLO_OFL),
+	WMI_NLO_SCAN_COMPLETE_EVENTID,
+	WMI_APFIND_EVENTID,
+	WMI_PASSPOINT_MATCH_EVENTID,
+	WMI_GTK_OFFLOAD_STATUS_EVENTID = WMI_TLV_CMD(WMI_GRP_GTK_OFL),
+	WMI_GTK_REKEY_FAIL_EVENTID,
+	WMI_CSA_HANDLING_EVENTID = WMI_TLV_CMD(WMI_GRP_CSA_OFL),
+	WMI_CHATTER_PC_QUERY_EVENTID = WMI_TLV_CMD(WMI_GRP_CHATTER),
+	WMI_PDEV_DFS_RADAR_DETECTION_EVENTID = WMI_TLV_CMD(WMI_GRP_DFS),
+	WMI_VDEV_DFS_CAC_COMPLETE_EVENTID,
+	WMI_VDEV_ADFS_OCAC_COMPLETE_EVENTID,
+	WMI_ECHO_EVENTID = WMI_TLV_CMD(WMI_GRP_MISC),
+	WMI_PDEV_UTF_EVENTID,
+	WMI_DEBUG_MESG_EVENTID,
+	WMI_UPDATE_STATS_EVENTID,
+	WMI_DEBUG_PRINT_EVENTID,
+	WMI_DCS_INTERFERENCE_EVENTID,
+	WMI_PDEV_QVIT_EVENTID,
+	WMI_WLAN_PROFILE_DATA_EVENTID,
+	WMI_PDEV_FTM_INTG_EVENTID,
+	WMI_WLAN_FREQ_AVOID_EVENTID,
+	WMI_VDEV_GET_KEEPALIVE_EVENTID,
+	WMI_THERMAL_MGMT_EVENTID,
+	WMI_DIAG_DATA_CONTAINER_EVENTID,
+	WMI_HOST_AUTO_SHUTDOWN_EVENTID,
+	WMI_UPDATE_WHAL_MIB_STATS_EVENTID,
+	WMI_UPDATE_VDEV_RATE_STATS_EVENTID,
+	WMI_DIAG_EVENTID,
+	WMI_OCB_SET_SCHED_EVENTID,
+	WMI_DEBUG_MESG_FLUSH_COMPLETE_EVENTID,
+	WMI_RSSI_BREACH_EVENTID,
+	WMI_TRANSFER_DATA_TO_FLASH_COMPLETE_EVENTID,
+	WMI_PDEV_UTF_SCPC_EVENTID,
+	WMI_READ_DATA_FROM_FLASH_EVENTID,
+	WMI_REPORT_RX_AGGR_FAILURE_EVENTID,
+	WMI_PKGID_EVENTID,
+	WMI_GPIO_INPUT_EVENTID = WMI_TLV_CMD(WMI_GRP_GPIO),
+	WMI_UPLOADH_EVENTID,
+	WMI_CAPTUREH_EVENTID,
+	WMI_RFKILL_STATE_CHANGE_EVENTID,
+	WMI_TDLS_PEER_EVENTID = WMI_TLV_CMD(WMI_GRP_TDLS),
+	WMI_STA_SMPS_FORCE_MODE_COMPL_EVENTID = WMI_TLV_CMD(WMI_GRP_STA_SMPS),
+	WMI_BATCH_SCAN_ENABLED_EVENTID = WMI_TLV_CMD(WMI_GRP_LOCATION_SCAN),
+	WMI_BATCH_SCAN_RESULT_EVENTID,
+	WMI_OEM_CAPABILITY_EVENTID = WMI_TLV_CMD(WMI_GRP_OEM),
+	WMI_OEM_MEASUREMENT_REPORT_EVENTID,
+	WMI_OEM_ERROR_REPORT_EVENTID,
+	WMI_OEM_RESPONSE_EVENTID,
+	WMI_NAN_EVENTID = WMI_TLV_CMD(WMI_GRP_NAN),
+	WMI_NAN_DISC_IFACE_CREATED_EVENTID,
+	WMI_NAN_DISC_IFACE_DELETED_EVENTID,
+	WMI_NAN_STARTED_CLUSTER_EVENTID,
+	WMI_NAN_JOINED_CLUSTER_EVENTID,
+	WMI_COEX_REPORT_ANTENNA_ISOLATION_EVENTID = WMI_TLV_CMD(WMI_GRP_COEX),
+	WMI_LPI_RESULT_EVENTID = WMI_TLV_CMD(WMI_GRP_LPI),
+	WMI_LPI_STATUS_EVENTID,
+	WMI_LPI_HANDOFF_EVENTID,
+	WMI_EXTSCAN_START_STOP_EVENTID = WMI_TLV_CMD(WMI_GRP_EXTSCAN),
+	WMI_EXTSCAN_OPERATION_EVENTID,
+	WMI_EXTSCAN_TABLE_USAGE_EVENTID,
+	WMI_EXTSCAN_CACHED_RESULTS_EVENTID,
+	WMI_EXTSCAN_WLAN_CHANGE_RESULTS_EVENTID,
+	WMI_EXTSCAN_HOTLIST_MATCH_EVENTID,
+	WMI_EXTSCAN_CAPABILITIES_EVENTID,
+	WMI_EXTSCAN_HOTLIST_SSID_MATCH_EVENTID,
+	WMI_MDNS_STATS_EVENTID = WMI_TLV_CMD(WMI_GRP_MDNS_OFL),
+	WMI_SAP_OFL_ADD_STA_EVENTID = WMI_TLV_CMD(WMI_GRP_SAP_OFL),
+	WMI_SAP_OFL_DEL_STA_EVENTID,
+	WMI_OCB_SET_CONFIG_RESP_EVENTID = WMI_TLV_CMD(WMI_GRP_OCB),
+	WMI_OCB_GET_TSF_TIMER_RESP_EVENTID,
+	WMI_DCC_GET_STATS_RESP_EVENTID,
+	WMI_DCC_UPDATE_NDL_RESP_EVENTID,
+	WMI_DCC_STATS_EVENTID,
+	WMI_SOC_SET_HW_MODE_RESP_EVENTID = WMI_TLV_CMD(WMI_GRP_SOC),
+	WMI_SOC_HW_MODE_TRANSITION_EVENTID,
+	WMI_SOC_SET_DUAL_MAC_CONFIG_RESP_EVENTID,
+	WMI_MAWC_ENABLE_SENSOR_EVENTID = WMI_TLV_CMD(WMI_GRP_MAWC),
+	WMI_BPF_CAPABILIY_INFO_EVENTID = WMI_TLV_CMD(WMI_GRP_BPF_OFFLOAD),
+	WMI_BPF_VDEV_STATS_INFO_EVENTID,
+	WMI_RMC_NEW_LEADER_EVENTID = WMI_TLV_CMD(WMI_GRP_RMC),
+	WMI_REG_CHAN_LIST_CC_EVENTID = WMI_TLV_CMD(WMI_GRP_REGULATORY),
+	WMI_11D_NEW_COUNTRY_EVENTID,
+	WMI_NDI_CAP_RSP_EVENTID = WMI_TLV_CMD(WMI_GRP_PROTOTYPE),
+	WMI_NDP_INITIATOR_RSP_EVENTID,
+	WMI_NDP_RESPONDER_RSP_EVENTID,
+	WMI_NDP_END_RSP_EVENTID,
+	WMI_NDP_INDICATION_EVENTID,
+	WMI_NDP_CONFIRM_EVENTID,
+	WMI_NDP_END_INDICATION_EVENTID,
+};
+
+enum wmi_tlv_pdev_param {
+	WMI_PDEV_PARAM_TX_CHAIN_MASK = 0x1,
+	WMI_PDEV_PARAM_RX_CHAIN_MASK,
+	WMI_PDEV_PARAM_TXPOWER_LIMIT2G,
+	WMI_PDEV_PARAM_TXPOWER_LIMIT5G,
+	WMI_PDEV_PARAM_TXPOWER_SCALE,
+	WMI_PDEV_PARAM_BEACON_GEN_MODE,
+	WMI_PDEV_PARAM_BEACON_TX_MODE,
+	WMI_PDEV_PARAM_RESMGR_OFFCHAN_MODE,
+	WMI_PDEV_PARAM_PROTECTION_MODE,
+	WMI_PDEV_PARAM_DYNAMIC_BW,
+	WMI_PDEV_PARAM_NON_AGG_SW_RETRY_TH,
+	WMI_PDEV_PARAM_AGG_SW_RETRY_TH,
+	WMI_PDEV_PARAM_STA_KICKOUT_TH,
+	WMI_PDEV_PARAM_AC_AGGRSIZE_SCALING,
+	WMI_PDEV_PARAM_LTR_ENABLE,
+	WMI_PDEV_PARAM_LTR_AC_LATENCY_BE,
+	WMI_PDEV_PARAM_LTR_AC_LATENCY_BK,
+	WMI_PDEV_PARAM_LTR_AC_LATENCY_VI,
+	WMI_PDEV_PARAM_LTR_AC_LATENCY_VO,
+	WMI_PDEV_PARAM_LTR_AC_LATENCY_TIMEOUT,
+	WMI_PDEV_PARAM_LTR_SLEEP_OVERRIDE,
+	WMI_PDEV_PARAM_LTR_RX_OVERRIDE,
+	WMI_PDEV_PARAM_LTR_TX_ACTIVITY_TIMEOUT,
+	WMI_PDEV_PARAM_L1SS_ENABLE,
+	WMI_PDEV_PARAM_DSLEEP_ENABLE,
+	WMI_PDEV_PARAM_PCIELP_TXBUF_FLUSH,
+	WMI_PDEV_PARAM_PCIELP_TXBUF_WATERMARK,
+	WMI_PDEV_PARAM_PCIELP_TXBUF_TMO_EN,
+	WMI_PDEV_PARAM_PCIELP_TXBUF_TMO_VALUE,
+	WMI_PDEV_PARAM_PDEV_STATS_UPDATE_PERIOD,
+	WMI_PDEV_PARAM_VDEV_STATS_UPDATE_PERIOD,
+	WMI_PDEV_PARAM_PEER_STATS_UPDATE_PERIOD,
+	WMI_PDEV_PARAM_BCNFLT_STATS_UPDATE_PERIOD,
+	WMI_PDEV_PARAM_PMF_QOS,
+	WMI_PDEV_PARAM_ARP_AC_OVERRIDE,
+	WMI_PDEV_PARAM_DCS,
+	WMI_PDEV_PARAM_ANI_ENABLE,
+	WMI_PDEV_PARAM_ANI_POLL_PERIOD,
+	WMI_PDEV_PARAM_ANI_LISTEN_PERIOD,
+	WMI_PDEV_PARAM_ANI_OFDM_LEVEL,
+	WMI_PDEV_PARAM_ANI_CCK_LEVEL,
+	WMI_PDEV_PARAM_DYNTXCHAIN,
+	WMI_PDEV_PARAM_PROXY_STA,
+	WMI_PDEV_PARAM_IDLE_PS_CONFIG,
+	WMI_PDEV_PARAM_POWER_GATING_SLEEP,
+	WMI_PDEV_PARAM_RFKILL_ENABLE,
+	WMI_PDEV_PARAM_BURST_DUR,
+	WMI_PDEV_PARAM_BURST_ENABLE,
+	WMI_PDEV_PARAM_HW_RFKILL_CONFIG,
+	WMI_PDEV_PARAM_LOW_POWER_RF_ENABLE,
+	WMI_PDEV_PARAM_L1SS_TRACK,
+	WMI_PDEV_PARAM_HYST_EN,
+	WMI_PDEV_PARAM_POWER_COLLAPSE_ENABLE,
+	WMI_PDEV_PARAM_LED_SYS_STATE,
+	WMI_PDEV_PARAM_LED_ENABLE,
+	WMI_PDEV_PARAM_AUDIO_OVER_WLAN_LATENCY,
+	WMI_PDEV_PARAM_AUDIO_OVER_WLAN_ENABLE,
+	WMI_PDEV_PARAM_WHAL_MIB_STATS_UPDATE_ENABLE,
+	WMI_PDEV_PARAM_VDEV_RATE_STATS_UPDATE_PERIOD,
+	WMI_PDEV_PARAM_CTS_CBW,
+	WMI_PDEV_PARAM_WNTS_CONFIG,
+	WMI_PDEV_PARAM_ADAPTIVE_EARLY_RX_ENABLE,
+	WMI_PDEV_PARAM_ADAPTIVE_EARLY_RX_MIN_SLEEP_SLOP,
+	WMI_PDEV_PARAM_ADAPTIVE_EARLY_RX_INC_DEC_STEP,
+	WMI_PDEV_PARAM_EARLY_RX_FIX_SLEEP_SLOP,
+	WMI_PDEV_PARAM_BMISS_BASED_ADAPTIVE_BTO_ENABLE,
+	WMI_PDEV_PARAM_BMISS_BTO_MIN_BCN_TIMEOUT,
+	WMI_PDEV_PARAM_BMISS_BTO_INC_DEC_STEP,
+	WMI_PDEV_PARAM_BTO_FIX_BCN_TIMEOUT,
+	WMI_PDEV_PARAM_CE_BASED_ADAPTIVE_BTO_ENABLE,
+	WMI_PDEV_PARAM_CE_BTO_COMBO_CE_VALUE,
+	WMI_PDEV_PARAM_TX_CHAIN_MASK_2G,
+	WMI_PDEV_PARAM_RX_CHAIN_MASK_2G,
+	WMI_PDEV_PARAM_TX_CHAIN_MASK_5G,
+	WMI_PDEV_PARAM_RX_CHAIN_MASK_5G,
+	WMI_PDEV_PARAM_TX_CHAIN_MASK_CCK,
+	WMI_PDEV_PARAM_TX_CHAIN_MASK_1SS,
+	WMI_PDEV_PARAM_CTS2SELF_FOR_P2P_GO_CONFIG,
+	WMI_PDEV_PARAM_TXPOWER_DECR_DB,
+	WMI_PDEV_PARAM_AGGR_BURST,
+	WMI_PDEV_PARAM_RX_DECAP_MODE,
+	WMI_PDEV_PARAM_FAST_CHANNEL_RESET,
+	WMI_PDEV_PARAM_SMART_ANTENNA_DEFAULT_ANTENNA,
+	WMI_PDEV_PARAM_ANTENNA_GAIN,
+	WMI_PDEV_PARAM_RX_FILTER,
+	WMI_PDEV_SET_MCAST_TO_UCAST_TID,
+	WMI_PDEV_PARAM_PROXY_STA_MODE,
+	WMI_PDEV_PARAM_SET_MCAST2UCAST_MODE,
+	WMI_PDEV_PARAM_SET_MCAST2UCAST_BUFFER,
+	WMI_PDEV_PARAM_REMOVE_MCAST2UCAST_BUFFER,
+	WMI_PDEV_PEER_STA_PS_STATECHG_ENABLE,
+	WMI_PDEV_PARAM_IGMPMLD_AC_OVERRIDE,
+	WMI_PDEV_PARAM_BLOCK_INTERBSS,
+	WMI_PDEV_PARAM_SET_DISABLE_RESET_CMDID,
+	WMI_PDEV_PARAM_SET_MSDU_TTL_CMDID,
+	WMI_PDEV_PARAM_SET_PPDU_DURATION_CMDID,
+	WMI_PDEV_PARAM_TXBF_SOUND_PERIOD_CMDID,
+	WMI_PDEV_PARAM_SET_PROMISC_MODE_CMDID,
+	WMI_PDEV_PARAM_SET_BURST_MODE_CMDID,
+	WMI_PDEV_PARAM_EN_STATS,
+	WMI_PDEV_PARAM_MU_GROUP_POLICY,
+	WMI_PDEV_PARAM_NOISE_DETECTION,
+	WMI_PDEV_PARAM_NOISE_THRESHOLD,
+	WMI_PDEV_PARAM_DPD_ENABLE,
+	WMI_PDEV_PARAM_SET_MCAST_BCAST_ECHO,
+	WMI_PDEV_PARAM_ATF_STRICT_SCH,
+	WMI_PDEV_PARAM_ATF_SCHED_DURATION,
+	WMI_PDEV_PARAM_ANT_PLZN,
+	WMI_PDEV_PARAM_MGMT_RETRY_LIMIT,
+	WMI_PDEV_PARAM_SENSITIVITY_LEVEL,
+	WMI_PDEV_PARAM_SIGNED_TXPOWER_2G,
+	WMI_PDEV_PARAM_SIGNED_TXPOWER_5G,
+	WMI_PDEV_PARAM_ENABLE_PER_TID_AMSDU,
+	WMI_PDEV_PARAM_ENABLE_PER_TID_AMPDU,
+	WMI_PDEV_PARAM_CCA_THRESHOLD,
+	WMI_PDEV_PARAM_RTS_FIXED_RATE,
+	WMI_PDEV_PARAM_PDEV_RESET,
+	WMI_PDEV_PARAM_WAPI_MBSSID_OFFSET,
+	WMI_PDEV_PARAM_ARP_DBG_SRCADDR,
+	WMI_PDEV_PARAM_ARP_DBG_DSTADDR,
+	WMI_PDEV_PARAM_ATF_OBSS_NOISE_SCH,
+	WMI_PDEV_PARAM_ATF_OBSS_NOISE_SCALING_FACTOR,
+	WMI_PDEV_PARAM_CUST_TXPOWER_SCALE,
+	WMI_PDEV_PARAM_ATF_DYNAMIC_ENABLE,
+	WMI_PDEV_PARAM_CTRL_RETRY_LIMIT,
+	WMI_PDEV_PARAM_PROPAGATION_DELAY,
+	WMI_PDEV_PARAM_ENA_ANT_DIV,
+	WMI_PDEV_PARAM_FORCE_CHAIN_ANT,
+	WMI_PDEV_PARAM_ANT_DIV_SELFTEST,
+	WMI_PDEV_PARAM_ANT_DIV_SELFTEST_INTVL,
+	WMI_PDEV_PARAM_STATS_OBSERVATION_PERIOD,
+	WMI_PDEV_PARAM_TX_PPDU_DELAY_BIN_SIZE_MS,
+	WMI_PDEV_PARAM_TX_PPDU_DELAY_ARRAY_LEN,
+	WMI_PDEV_PARAM_TX_MPDU_AGGR_ARRAY_LEN,
+	WMI_PDEV_PARAM_RX_MPDU_AGGR_ARRAY_LEN,
+	WMI_PDEV_PARAM_TX_SCH_DELAY,
+	WMI_PDEV_PARAM_ENABLE_RTS_SIFS_BURSTING,
+	WMI_PDEV_PARAM_MAX_MPDUS_IN_AMPDU,
+	WMI_PDEV_PARAM_PEER_STATS_INFO_ENABLE,
+	WMI_PDEV_PARAM_FAST_PWR_TRANSITION,
+	WMI_PDEV_PARAM_RADIO_CHAN_STATS_ENABLE,
+	WMI_PDEV_PARAM_RADIO_DIAGNOSIS_ENABLE,
+	WMI_PDEV_PARAM_MESH_MCAST_ENABLE,
+};
+
+enum wmi_tlv_vdev_param {
+	WMI_VDEV_PARAM_RTS_THRESHOLD = 0x1,
+	WMI_VDEV_PARAM_FRAGMENTATION_THRESHOLD,
+	WMI_VDEV_PARAM_BEACON_INTERVAL,
+	WMI_VDEV_PARAM_LISTEN_INTERVAL,
+	WMI_VDEV_PARAM_MULTICAST_RATE,
+	WMI_VDEV_PARAM_MGMT_TX_RATE,
+	WMI_VDEV_PARAM_SLOT_TIME,
+	WMI_VDEV_PARAM_PREAMBLE,
+	WMI_VDEV_PARAM_SWBA_TIME,
+	WMI_VDEV_STATS_UPDATE_PERIOD,
+	WMI_VDEV_PWRSAVE_AGEOUT_TIME,
+	WMI_VDEV_HOST_SWBA_INTERVAL,
+	WMI_VDEV_PARAM_DTIM_PERIOD,
+	WMI_VDEV_OC_SCHEDULER_AIR_TIME_LIMIT,
+	WMI_VDEV_PARAM_WDS,
+	WMI_VDEV_PARAM_ATIM_WINDOW,
+	WMI_VDEV_PARAM_BMISS_COUNT_MAX,
+	WMI_VDEV_PARAM_BMISS_FIRST_BCNT,
+	WMI_VDEV_PARAM_BMISS_FINAL_BCNT,
+	WMI_VDEV_PARAM_FEATURE_WMM,
+	WMI_VDEV_PARAM_CHWIDTH,
+	WMI_VDEV_PARAM_CHEXTOFFSET,
+	WMI_VDEV_PARAM_DISABLE_HTPROTECTION,
+	WMI_VDEV_PARAM_STA_QUICKKICKOUT,
+	WMI_VDEV_PARAM_MGMT_RATE,
+	WMI_VDEV_PARAM_PROTECTION_MODE,
+	WMI_VDEV_PARAM_FIXED_RATE,
+	WMI_VDEV_PARAM_SGI,
+	WMI_VDEV_PARAM_LDPC,
+	WMI_VDEV_PARAM_TX_STBC,
+	WMI_VDEV_PARAM_RX_STBC,
+	WMI_VDEV_PARAM_INTRA_BSS_FWD,
+	WMI_VDEV_PARAM_DEF_KEYID,
+	WMI_VDEV_PARAM_NSS,
+	WMI_VDEV_PARAM_BCAST_DATA_RATE,
+	WMI_VDEV_PARAM_MCAST_DATA_RATE,
+	WMI_VDEV_PARAM_MCAST_INDICATE,
+	WMI_VDEV_PARAM_DHCP_INDICATE,
+	WMI_VDEV_PARAM_UNKNOWN_DEST_INDICATE,
+	WMI_VDEV_PARAM_AP_KEEPALIVE_MIN_IDLE_INACTIVE_TIME_SECS,
+	WMI_VDEV_PARAM_AP_KEEPALIVE_MAX_IDLE_INACTIVE_TIME_SECS,
+	WMI_VDEV_PARAM_AP_KEEPALIVE_MAX_UNRESPONSIVE_TIME_SECS,
+	WMI_VDEV_PARAM_AP_ENABLE_NAWDS,
+	WMI_VDEV_PARAM_ENABLE_RTSCTS,
+	WMI_VDEV_PARAM_TXBF,
+	WMI_VDEV_PARAM_PACKET_POWERSAVE,
+	WMI_VDEV_PARAM_DROP_UNENCRY,
+	WMI_VDEV_PARAM_TX_ENCAP_TYPE,
+	WMI_VDEV_PARAM_AP_DETECT_OUT_OF_SYNC_SLEEPING_STA_TIME_SECS,
+	WMI_VDEV_PARAM_EARLY_RX_ADJUST_ENABLE,
+	WMI_VDEV_PARAM_EARLY_RX_TGT_BMISS_NUM,
+	WMI_VDEV_PARAM_EARLY_RX_BMISS_SAMPLE_CYCLE,
+	WMI_VDEV_PARAM_EARLY_RX_SLOP_STEP,
+	WMI_VDEV_PARAM_EARLY_RX_INIT_SLOP,
+	WMI_VDEV_PARAM_EARLY_RX_ADJUST_PAUSE,
+	WMI_VDEV_PARAM_TX_PWRLIMIT,
+	WMI_VDEV_PARAM_SNR_NUM_FOR_CAL,
+	WMI_VDEV_PARAM_ROAM_FW_OFFLOAD,
+	WMI_VDEV_PARAM_ENABLE_RMC,
+	WMI_VDEV_PARAM_IBSS_MAX_BCN_LOST_MS,
+	WMI_VDEV_PARAM_MAX_RATE,
+	WMI_VDEV_PARAM_EARLY_RX_DRIFT_SAMPLE,
+	WMI_VDEV_PARAM_SET_IBSS_TX_FAIL_CNT_THR,
+	WMI_VDEV_PARAM_EBT_RESYNC_TIMEOUT,
+	WMI_VDEV_PARAM_AGGR_TRIG_EVENT_ENABLE,
+	WMI_VDEV_PARAM_IS_IBSS_POWER_SAVE_ALLOWED,
+	WMI_VDEV_PARAM_IS_POWER_COLLAPSE_ALLOWED,
+	WMI_VDEV_PARAM_IS_AWAKE_ON_TXRX_ENABLED,
+	WMI_VDEV_PARAM_INACTIVITY_CNT,
+	WMI_VDEV_PARAM_TXSP_END_INACTIVITY_TIME_MS,
+	WMI_VDEV_PARAM_DTIM_POLICY,
+	WMI_VDEV_PARAM_IBSS_PS_WARMUP_TIME_SECS,
+	WMI_VDEV_PARAM_IBSS_PS_1RX_CHAIN_IN_ATIM_WINDOW_ENABLE,
+	WMI_VDEV_PARAM_RX_LEAK_WINDOW,
+	WMI_VDEV_PARAM_STATS_AVG_FACTOR,
+	WMI_VDEV_PARAM_DISCONNECT_TH,
+	WMI_VDEV_PARAM_RTSCTS_RATE,
+	WMI_VDEV_PARAM_MCC_RTSCTS_PROTECTION_ENABLE,
+	WMI_VDEV_PARAM_MCC_BROADCAST_PROBE_ENABLE,
+	WMI_VDEV_PARAM_TXPOWER_SCALE,
+	WMI_VDEV_PARAM_TXPOWER_SCALE_DECR_DB,
+	WMI_VDEV_PARAM_MCAST2UCAST_SET,
+	WMI_VDEV_PARAM_RC_NUM_RETRIES,
+	WMI_VDEV_PARAM_CABQ_MAXDUR,
+	WMI_VDEV_PARAM_MFPTEST_SET,
+	WMI_VDEV_PARAM_RTS_FIXED_RATE,
+	WMI_VDEV_PARAM_VHT_SGIMASK,
+	WMI_VDEV_PARAM_VHT80_RATEMASK,
+	WMI_VDEV_PARAM_PROXY_STA,
+	WMI_VDEV_PARAM_VIRTUAL_CELL_MODE,
+	WMI_VDEV_PARAM_RX_DECAP_TYPE,
+	WMI_VDEV_PARAM_BW_NSS_RATEMASK,
+	WMI_VDEV_PARAM_SENSOR_AP,
+	WMI_VDEV_PARAM_BEACON_RATE,
+	WMI_VDEV_PARAM_DTIM_ENABLE_CTS,
+	WMI_VDEV_PARAM_STA_KICKOUT,
+	WMI_VDEV_PARAM_CAPABILITIES,
+	WMI_VDEV_PARAM_TSF_INCREMENT,
+	WMI_VDEV_PARAM_AMPDU_PER_AC,
+	WMI_VDEV_PARAM_RX_FILTER,
+	WMI_VDEV_PARAM_MGMT_TX_POWER,
+	WMI_VDEV_PARAM_NON_AGG_SW_RETRY_TH,
+	WMI_VDEV_PARAM_AGG_SW_RETRY_TH,
+	WMI_VDEV_PARAM_DISABLE_DYN_BW_RTS,
+	WMI_VDEV_PARAM_ATF_SSID_SCHED_POLICY,
+	WMI_VDEV_PARAM_HE_DCM,
+	WMI_VDEV_PARAM_HE_RANGE_EXT,
+	WMI_VDEV_PARAM_ENABLE_BCAST_PROBE_RESPONSE,
+	WMI_VDEV_PARAM_FILS_MAX_CHANNEL_GUARD_TIME,
+	WMI_VDEV_PARAM_SET_HE_SOUNDING_MODE = 0x87,
+	WMI_VDEV_PARAM_PROTOTYPE = 0x8000,
+	WMI_VDEV_PARAM_BSS_COLOR,
+	WMI_VDEV_PARAM_SET_HEMU_MODE,
+	WMI_VDEV_PARAM_TX_OFDMA_CPLEN,
+};
+
+enum wmi_tlv_peer_flags {
+	WMI_TLV_PEER_AUTH = 0x00000001,
+	WMI_TLV_PEER_QOS = 0x00000002,
+	WMI_TLV_PEER_NEED_PTK_4_WAY = 0x00000004,
+	WMI_TLV_PEER_NEED_GTK_2_WAY = 0x00000010,
+	WMI_TLV_PEER_APSD = 0x00000800,
+	WMI_TLV_PEER_HT = 0x00001000,
+	WMI_TLV_PEER_40MHZ = 0x00002000,
+	WMI_TLV_PEER_STBC = 0x00008000,
+	WMI_TLV_PEER_LDPC = 0x00010000,
+	WMI_TLV_PEER_DYN_MIMOPS = 0x00020000,
+	WMI_TLV_PEER_STATIC_MIMOPS = 0x00040000,
+	WMI_TLV_PEER_SPATIAL_MUX = 0x00200000,
+	WMI_TLV_PEER_VHT = 0x02000000,
+	WMI_TLV_PEER_80MHZ = 0x04000000,
+	WMI_TLV_PEER_PMF = 0x08000000,
+	WMI_PEER_IS_P2P_CAPABLE = 0x20000000,
+	WMI_PEER_160MHZ         = 0x40000000,
+	WMI_PEER_SAFEMODE_EN    = 0x80000000,
+
+};
+
+/** Enum list of TLV Tags for each parameter structure type. */
+enum wmi_tlv_tag {
+	WMI_TAG_LAST_RESERVED = 15,
+	WMI_TAG_FIRST_ARRAY_ENUM,
+	WMI_TAG_ARRAY_UINT32 = WMI_TAG_FIRST_ARRAY_ENUM,
+	WMI_TAG_ARRAY_BYTE,
+	WMI_TAG_ARRAY_STRUCT,
+	WMI_TAG_ARRAY_FIXED_STRUCT,
+	WMI_TAG_LAST_ARRAY_ENUM = 31,
+	WMI_TAG_SERVICE_READY_EVENT,
+	WMI_TAG_HAL_REG_CAPABILITIES,
+	WMI_TAG_WLAN_HOST_MEM_REQ,
+	WMI_TAG_READY_EVENT,
+	WMI_TAG_SCAN_EVENT,
+	WMI_TAG_PDEV_TPC_CONFIG_EVENT,
+	WMI_TAG_CHAN_INFO_EVENT,
+	WMI_TAG_COMB_PHYERR_RX_HDR,
+	WMI_TAG_VDEV_START_RESPONSE_EVENT,
+	WMI_TAG_VDEV_STOPPED_EVENT,
+	WMI_TAG_VDEV_INSTALL_KEY_COMPLETE_EVENT,
+	WMI_TAG_PEER_STA_KICKOUT_EVENT,
+	WMI_TAG_MGMT_RX_HDR,
+	WMI_TAG_TBTT_OFFSET_EVENT,
+	WMI_TAG_TX_DELBA_COMPLETE_EVENT,
+	WMI_TAG_TX_ADDBA_COMPLETE_EVENT,
+	WMI_TAG_ROAM_EVENT,
+	WMI_TAG_WOW_EVENT_INFO,
+	WMI_TAG_WOW_EVENT_INFO_SECTION_BITMAP,
+	WMI_TAG_RTT_EVENT_HEADER,
+	WMI_TAG_RTT_ERROR_REPORT_EVENT,
+	WMI_TAG_RTT_MEAS_EVENT,
+	WMI_TAG_ECHO_EVENT,
+	WMI_TAG_FTM_INTG_EVENT,
+	WMI_TAG_VDEV_GET_KEEPALIVE_EVENT,
+	WMI_TAG_GPIO_INPUT_EVENT,
+	WMI_TAG_CSA_EVENT,
+	WMI_TAG_GTK_OFFLOAD_STATUS_EVENT,
+	WMI_TAG_IGTK_INFO,
+	WMI_TAG_DCS_INTERFERENCE_EVENT,
+	WMI_TAG_ATH_DCS_CW_INT,
+	WMI_TAG_WLAN_DCS_CW_INT = /* ALIAS */
+		WMI_TAG_ATH_DCS_CW_INT,
+	WMI_TAG_ATH_DCS_WLAN_INT_STAT,
+	WMI_TAG_WLAN_DCS_IM_TGT_STATS_T = /* ALIAS */
+		WMI_TAG_ATH_DCS_WLAN_INT_STAT,
+	WMI_TAG_WLAN_PROFILE_CTX_T,
+	WMI_TAG_WLAN_PROFILE_T,
+	WMI_TAG_PDEV_QVIT_EVENT,
+	WMI_TAG_HOST_SWBA_EVENT,
+	WMI_TAG_TIM_INFO,
+	WMI_TAG_P2P_NOA_INFO,
+	WMI_TAG_STATS_EVENT,
+	WMI_TAG_AVOID_FREQ_RANGES_EVENT,
+	WMI_TAG_AVOID_FREQ_RANGE_DESC,
+	WMI_TAG_GTK_REKEY_FAIL_EVENT,
+	WMI_TAG_INIT_CMD,
+	WMI_TAG_RESOURCE_CONFIG,
+	WMI_TAG_WLAN_HOST_MEMORY_CHUNK,
+	WMI_TAG_START_SCAN_CMD,
+	WMI_TAG_STOP_SCAN_CMD,
+	WMI_TAG_SCAN_CHAN_LIST_CMD,
+	WMI_TAG_CHANNEL,
+	WMI_TAG_PDEV_SET_REGDOMAIN_CMD,
+	WMI_TAG_PDEV_SET_PARAM_CMD,
+	WMI_TAG_PDEV_SET_WMM_PARAMS_CMD,
+	WMI_TAG_WMM_PARAMS,
+	WMI_TAG_PDEV_SET_QUIET_CMD,
+	WMI_TAG_VDEV_CREATE_CMD,
+	WMI_TAG_VDEV_DELETE_CMD,
+	WMI_TAG_VDEV_START_REQUEST_CMD,
+	WMI_TAG_P2P_NOA_DESCRIPTOR,
+	WMI_TAG_P2P_GO_SET_BEACON_IE,
+	WMI_TAG_GTK_OFFLOAD_CMD,
+	WMI_TAG_VDEV_UP_CMD,
+	WMI_TAG_VDEV_STOP_CMD,
+	WMI_TAG_VDEV_DOWN_CMD,
+	WMI_TAG_VDEV_SET_PARAM_CMD,
+	WMI_TAG_VDEV_INSTALL_KEY_CMD,
+	WMI_TAG_PEER_CREATE_CMD,
+	WMI_TAG_PEER_DELETE_CMD,
+	WMI_TAG_PEER_FLUSH_TIDS_CMD,
+	WMI_TAG_PEER_SET_PARAM_CMD,
+	WMI_TAG_PEER_ASSOC_COMPLETE_CMD,
+	WMI_TAG_VHT_RATE_SET,
+	WMI_TAG_BCN_TMPL_CMD,
+	WMI_TAG_PRB_TMPL_CMD,
+	WMI_TAG_BCN_PRB_INFO,
+	WMI_TAG_PEER_TID_ADDBA_CMD,
+	WMI_TAG_PEER_TID_DELBA_CMD,
+	WMI_TAG_STA_POWERSAVE_MODE_CMD,
+	WMI_TAG_STA_POWERSAVE_PARAM_CMD,
+	WMI_TAG_STA_DTIM_PS_METHOD_CMD,
+	WMI_TAG_ROAM_SCAN_MODE,
+	WMI_TAG_ROAM_SCAN_RSSI_THRESHOLD,
+	WMI_TAG_ROAM_SCAN_PERIOD,
+	WMI_TAG_ROAM_SCAN_RSSI_CHANGE_THRESHOLD,
+	WMI_TAG_PDEV_SUSPEND_CMD,
+	WMI_TAG_PDEV_RESUME_CMD,
+	WMI_TAG_ADD_BCN_FILTER_CMD,
+	WMI_TAG_RMV_BCN_FILTER_CMD,
+	WMI_TAG_WOW_ENABLE_CMD,
+	WMI_TAG_WOW_HOSTWAKEUP_FROM_SLEEP_CMD,
+	WMI_TAG_STA_UAPSD_AUTO_TRIG_CMD,
+	WMI_TAG_STA_UAPSD_AUTO_TRIG_PARAM,
+	WMI_TAG_SET_ARP_NS_OFFLOAD_CMD,
+	WMI_TAG_ARP_OFFLOAD_TUPLE,
+	WMI_TAG_NS_OFFLOAD_TUPLE,
+	WMI_TAG_FTM_INTG_CMD,
+	WMI_TAG_STA_KEEPALIVE_CMD,
+	WMI_TAG_STA_KEEPALVE_ARP_RESPONSE,
+	WMI_TAG_P2P_SET_VENDOR_IE_DATA_CMD,
+	WMI_TAG_AP_PS_PEER_CMD,
+	WMI_TAG_PEER_RATE_RETRY_SCHED_CMD,
+	WMI_TAG_WLAN_PROFILE_TRIGGER_CMD,
+	WMI_TAG_WLAN_PROFILE_SET_HIST_INTVL_CMD,
+	WMI_TAG_WLAN_PROFILE_GET_PROF_DATA_CMD,
+	WMI_TAG_WLAN_PROFILE_ENABLE_PROFILE_ID_CMD,
+	WMI_TAG_WOW_DEL_PATTERN_CMD,
+	WMI_TAG_WOW_ADD_DEL_EVT_CMD,
+	WMI_TAG_RTT_MEASREQ_HEAD,
+	WMI_TAG_RTT_MEASREQ_BODY,
+	WMI_TAG_RTT_TSF_CMD,
+	WMI_TAG_VDEV_SPECTRAL_CONFIGURE_CMD,
+	WMI_TAG_VDEV_SPECTRAL_ENABLE_CMD,
+	WMI_TAG_REQUEST_STATS_CMD,
+	WMI_TAG_NLO_CONFIG_CMD,
+	WMI_TAG_NLO_CONFIGURED_PARAMETERS,
+	WMI_TAG_CSA_OFFLOAD_ENABLE_CMD,
+	WMI_TAG_CSA_OFFLOAD_CHANSWITCH_CMD,
+	WMI_TAG_CHATTER_SET_MODE_CMD,
+	WMI_TAG_ECHO_CMD,
+	WMI_TAG_VDEV_SET_KEEPALIVE_CMD,
+	WMI_TAG_VDEV_GET_KEEPALIVE_CMD,
+	WMI_TAG_FORCE_FW_HANG_CMD,
+	WMI_TAG_GPIO_CONFIG_CMD,
+	WMI_TAG_GPIO_OUTPUT_CMD,
+	WMI_TAG_PEER_ADD_WDS_ENTRY_CMD,
+	WMI_TAG_PEER_REMOVE_WDS_ENTRY_CMD,
+	WMI_TAG_BCN_TX_HDR,
+	WMI_TAG_BCN_SEND_FROM_HOST_CMD,
+	WMI_TAG_MGMT_TX_HDR,
+	WMI_TAG_ADDBA_CLEAR_RESP_CMD,
+	WMI_TAG_ADDBA_SEND_CMD,
+	WMI_TAG_DELBA_SEND_CMD,
+	WMI_TAG_ADDBA_SETRESPONSE_CMD,
+	WMI_TAG_SEND_SINGLEAMSDU_CMD,
+	WMI_TAG_PDEV_PKTLOG_ENABLE_CMD,
+	WMI_TAG_PDEV_PKTLOG_DISABLE_CMD,
+	WMI_TAG_PDEV_SET_HT_IE_CMD,
+	WMI_TAG_PDEV_SET_VHT_IE_CMD,
+	WMI_TAG_PDEV_SET_DSCP_TID_MAP_CMD,
+	WMI_TAG_PDEV_GREEN_AP_PS_ENABLE_CMD,
+	WMI_TAG_PDEV_GET_TPC_CONFIG_CMD,
+	WMI_TAG_PDEV_SET_BASE_MACADDR_CMD,
+	WMI_TAG_PEER_MCAST_GROUP_CMD,
+	WMI_TAG_ROAM_AP_PROFILE,
+	WMI_TAG_AP_PROFILE,
+	WMI_TAG_SCAN_SCH_PRIORITY_TABLE_CMD,
+	WMI_TAG_PDEV_DFS_ENABLE_CMD,
+	WMI_TAG_PDEV_DFS_DISABLE_CMD,
+	WMI_TAG_WOW_ADD_PATTERN_CMD,
+	WMI_TAG_WOW_BITMAP_PATTERN_T,
+	WMI_TAG_WOW_IPV4_SYNC_PATTERN_T,
+	WMI_TAG_WOW_IPV6_SYNC_PATTERN_T,
+	WMI_TAG_WOW_MAGIC_PATTERN_CMD,
+	WMI_TAG_SCAN_UPDATE_REQUEST_CMD,
+	WMI_TAG_CHATTER_PKT_COALESCING_FILTER,
+	WMI_TAG_CHATTER_COALESCING_ADD_FILTER_CMD,
+	WMI_TAG_CHATTER_COALESCING_DELETE_FILTER_CMD,
+	WMI_TAG_CHATTER_COALESCING_QUERY_CMD,
+	WMI_TAG_TXBF_CMD,
+	WMI_TAG_DEBUG_LOG_CONFIG_CMD,
+	WMI_TAG_NLO_EVENT,
+	WMI_TAG_CHATTER_QUERY_REPLY_EVENT,
+	WMI_TAG_UPLOAD_H_HDR,
+	WMI_TAG_CAPTURE_H_EVENT_HDR,
+	WMI_TAG_VDEV_WNM_SLEEPMODE_CMD,
+	WMI_TAG_VDEV_IPSEC_NATKEEPALIVE_FILTER_CMD,
+	WMI_TAG_VDEV_WMM_ADDTS_CMD,
+	WMI_TAG_VDEV_WMM_DELTS_CMD,
+	WMI_TAG_VDEV_SET_WMM_PARAMS_CMD,
+	WMI_TAG_TDLS_SET_STATE_CMD,
+	WMI_TAG_TDLS_PEER_UPDATE_CMD,
+	WMI_TAG_TDLS_PEER_EVENT,
+	WMI_TAG_TDLS_PEER_CAPABILITIES,
+	WMI_TAG_VDEV_MCC_SET_TBTT_MODE_CMD,
+	WMI_TAG_ROAM_CHAN_LIST,
+	WMI_TAG_VDEV_MCC_BCN_INTVL_CHANGE_EVENT,
+	WMI_TAG_RESMGR_ADAPTIVE_OCS_ENABLE_DISABLE_CMD,
+	WMI_TAG_RESMGR_SET_CHAN_TIME_QUOTA_CMD,
+	WMI_TAG_RESMGR_SET_CHAN_LATENCY_CMD,
+	WMI_TAG_BA_REQ_SSN_CMD,
+	WMI_TAG_BA_RSP_SSN_EVENT,
+	WMI_TAG_STA_SMPS_FORCE_MODE_CMD,
+	WMI_TAG_SET_MCASTBCAST_FILTER_CMD,
+	WMI_TAG_P2P_SET_OPPPS_CMD,
+	WMI_TAG_P2P_SET_NOA_CMD,
+	WMI_TAG_BA_REQ_SSN_CMD_SUB_STRUCT_PARAM,
+	WMI_TAG_BA_REQ_SSN_EVENT_SUB_STRUCT_PARAM,
+	WMI_TAG_STA_SMPS_PARAM_CMD,
+	WMI_TAG_VDEV_SET_GTX_PARAMS_CMD,
+	WMI_TAG_MCC_SCHED_TRAFFIC_STATS_CMD,
+	WMI_TAG_MCC_SCHED_STA_TRAFFIC_STATS,
+	WMI_TAG_OFFLOAD_BCN_TX_STATUS_EVENT,
+	WMI_TAG_P2P_NOA_EVENT,
+	WMI_TAG_HB_SET_ENABLE_CMD,
+	WMI_TAG_HB_SET_TCP_PARAMS_CMD,
+	WMI_TAG_HB_SET_TCP_PKT_FILTER_CMD,
+	WMI_TAG_HB_SET_UDP_PARAMS_CMD,
+	WMI_TAG_HB_SET_UDP_PKT_FILTER_CMD,
+	WMI_TAG_HB_IND_EVENT,
+	WMI_TAG_TX_PAUSE_EVENT,
+	WMI_TAG_RFKILL_EVENT,
+	WMI_TAG_DFS_RADAR_EVENT,
+	WMI_TAG_DFS_PHYERR_FILTER_ENA_CMD,
+	WMI_TAG_DFS_PHYERR_FILTER_DIS_CMD,
+	WMI_TAG_BATCH_SCAN_RESULT_SCAN_LIST,
+	WMI_TAG_BATCH_SCAN_RESULT_NETWORK_INFO,
+	WMI_TAG_BATCH_SCAN_ENABLE_CMD,
+	WMI_TAG_BATCH_SCAN_DISABLE_CMD,
+	WMI_TAG_BATCH_SCAN_TRIGGER_RESULT_CMD,
+	WMI_TAG_BATCH_SCAN_ENABLED_EVENT,
+	WMI_TAG_BATCH_SCAN_RESULT_EVENT,
+	WMI_TAG_VDEV_PLMREQ_START_CMD,
+	WMI_TAG_VDEV_PLMREQ_STOP_CMD,
+	WMI_TAG_THERMAL_MGMT_CMD,
+	WMI_TAG_THERMAL_MGMT_EVENT,
+	WMI_TAG_PEER_INFO_REQ_CMD,
+	WMI_TAG_PEER_INFO_EVENT,
+	WMI_TAG_PEER_INFO,
+	WMI_TAG_PEER_TX_FAIL_CNT_THR_EVENT,
+	WMI_TAG_RMC_SET_MODE_CMD,
+	WMI_TAG_RMC_SET_ACTION_PERIOD_CMD,
+	WMI_TAG_RMC_CONFIG_CMD,
+	WMI_TAG_MHF_OFFLOAD_SET_MODE_CMD,
+	WMI_TAG_MHF_OFFLOAD_PLUMB_ROUTING_TABLE_CMD,
+	WMI_TAG_ADD_PROACTIVE_ARP_RSP_PATTERN_CMD,
+	WMI_TAG_DEL_PROACTIVE_ARP_RSP_PATTERN_CMD,
+	WMI_TAG_NAN_CMD_PARAM,
+	WMI_TAG_NAN_EVENT_HDR,
+	WMI_TAG_PDEV_L1SS_TRACK_EVENT,
+	WMI_TAG_DIAG_DATA_CONTAINER_EVENT,
+	WMI_TAG_MODEM_POWER_STATE_CMD_PARAM,
+	WMI_TAG_PEER_GET_ESTIMATED_LINKSPEED_CMD,
+	WMI_TAG_PEER_ESTIMATED_LINKSPEED_EVENT,
+	WMI_TAG_AGGR_STATE_TRIG_EVENT,
+	WMI_TAG_MHF_OFFLOAD_ROUTING_TABLE_ENTRY,
+	WMI_TAG_ROAM_SCAN_CMD,
+	WMI_TAG_REQ_STATS_EXT_CMD,
+	WMI_TAG_STATS_EXT_EVENT,
+	WMI_TAG_OBSS_SCAN_ENABLE_CMD,
+	WMI_TAG_OBSS_SCAN_DISABLE_CMD,
+	WMI_TAG_OFFLOAD_PRB_RSP_TX_STATUS_EVENT,
+	WMI_TAG_PDEV_SET_LED_CONFIG_CMD,
+	WMI_TAG_HOST_AUTO_SHUTDOWN_CFG_CMD,
+	WMI_TAG_HOST_AUTO_SHUTDOWN_EVENT,
+	WMI_TAG_UPDATE_WHAL_MIB_STATS_EVENT,
+	WMI_TAG_CHAN_AVOID_UPDATE_CMD_PARAM,
+	WMI_TAG_WOW_IOAC_PKT_PATTERN_T,
+	WMI_TAG_WOW_IOAC_TMR_PATTERN_T,
+	WMI_TAG_WOW_IOAC_ADD_KEEPALIVE_CMD,
+	WMI_TAG_WOW_IOAC_DEL_KEEPALIVE_CMD,
+	WMI_TAG_WOW_IOAC_KEEPALIVE_T,
+	WMI_TAG_WOW_IOAC_ADD_PATTERN_CMD,
+	WMI_TAG_WOW_IOAC_DEL_PATTERN_CMD,
+	WMI_TAG_START_LINK_STATS_CMD,
+	WMI_TAG_CLEAR_LINK_STATS_CMD,
+	WMI_TAG_REQUEST_LINK_STATS_CMD,
+	WMI_TAG_IFACE_LINK_STATS_EVENT,
+	WMI_TAG_RADIO_LINK_STATS_EVENT,
+	WMI_TAG_PEER_STATS_EVENT,
+	WMI_TAG_CHANNEL_STATS,
+	WMI_TAG_RADIO_LINK_STATS,
+	WMI_TAG_RATE_STATS,
+	WMI_TAG_PEER_LINK_STATS,
+	WMI_TAG_WMM_AC_STATS,
+	WMI_TAG_IFACE_LINK_STATS,
+	WMI_TAG_LPI_MGMT_SNOOPING_CONFIG_CMD,
+	WMI_TAG_LPI_START_SCAN_CMD,
+	WMI_TAG_LPI_STOP_SCAN_CMD,
+	WMI_TAG_LPI_RESULT_EVENT,
+	WMI_TAG_PEER_STATE_EVENT,
+	WMI_TAG_EXTSCAN_BUCKET_CMD,
+	WMI_TAG_EXTSCAN_BUCKET_CHANNEL_EVENT,
+	WMI_TAG_EXTSCAN_START_CMD,
+	WMI_TAG_EXTSCAN_STOP_CMD,
+	WMI_TAG_EXTSCAN_CONFIGURE_WLAN_CHANGE_MONITOR_CMD,
+	WMI_TAG_EXTSCAN_WLAN_CHANGE_BSSID_PARAM_CMD,
+	WMI_TAG_EXTSCAN_CONFIGURE_HOTLIST_MONITOR_CMD,
+	WMI_TAG_EXTSCAN_GET_CACHED_RESULTS_CMD,
+	WMI_TAG_EXTSCAN_GET_WLAN_CHANGE_RESULTS_CMD,
+	WMI_TAG_EXTSCAN_SET_CAPABILITIES_CMD,
+	WMI_TAG_EXTSCAN_GET_CAPABILITIES_CMD,
+	WMI_TAG_EXTSCAN_OPERATION_EVENT,
+	WMI_TAG_EXTSCAN_START_STOP_EVENT,
+	WMI_TAG_EXTSCAN_TABLE_USAGE_EVENT,
+	WMI_TAG_EXTSCAN_WLAN_DESCRIPTOR_EVENT,
+	WMI_TAG_EXTSCAN_RSSI_INFO_EVENT,
+	WMI_TAG_EXTSCAN_CACHED_RESULTS_EVENT,
+	WMI_TAG_EXTSCAN_WLAN_CHANGE_RESULTS_EVENT,
+	WMI_TAG_EXTSCAN_WLAN_CHANGE_RESULT_BSSID_EVENT,
+	WMI_TAG_EXTSCAN_HOTLIST_MATCH_EVENT,
+	WMI_TAG_EXTSCAN_CAPABILITIES_EVENT,
+	WMI_TAG_EXTSCAN_CACHE_CAPABILITIES_EVENT,
+	WMI_TAG_EXTSCAN_WLAN_CHANGE_MONITOR_CAPABILITIES_EVENT,
+	WMI_TAG_EXTSCAN_HOTLIST_MONITOR_CAPABILITIES_EVENT,
+	WMI_TAG_D0_WOW_ENABLE_DISABLE_CMD,
+	WMI_TAG_D0_WOW_DISABLE_ACK_EVENT,
+	WMI_TAG_UNIT_TEST_CMD,
+	WMI_TAG_ROAM_OFFLOAD_TLV_PARAM,
+	WMI_TAG_ROAM_11I_OFFLOAD_TLV_PARAM,
+	WMI_TAG_ROAM_11R_OFFLOAD_TLV_PARAM,
+	WMI_TAG_ROAM_ESE_OFFLOAD_TLV_PARAM,
+	WMI_TAG_ROAM_SYNCH_EVENT,
+	WMI_TAG_ROAM_SYNCH_COMPLETE,
+	WMI_TAG_EXTWOW_ENABLE_CMD,
+	WMI_TAG_EXTWOW_SET_APP_TYPE1_PARAMS_CMD,
+	WMI_TAG_EXTWOW_SET_APP_TYPE2_PARAMS_CMD,
+	WMI_TAG_LPI_STATUS_EVENT,
+	WMI_TAG_LPI_HANDOFF_EVENT,
+	WMI_TAG_VDEV_RATE_STATS_EVENT,
+	WMI_TAG_VDEV_RATE_HT_INFO,
+	WMI_TAG_RIC_REQUEST,
+	WMI_TAG_PDEV_GET_TEMPERATURE_CMD,
+	WMI_TAG_PDEV_TEMPERATURE_EVENT,
+	WMI_TAG_SET_DHCP_SERVER_OFFLOAD_CMD,
+	WMI_TAG_TPC_CHAINMASK_CONFIG_CMD,
+	WMI_TAG_RIC_TSPEC,
+	WMI_TAG_TPC_CHAINMASK_CONFIG,
+	WMI_TAG_IPA_OFFLOAD_ENABLE_DISABLE_CMD,
+	WMI_TAG_SCAN_PROB_REQ_OUI_CMD,
+	WMI_TAG_KEY_MATERIAL,
+	WMI_TAG_TDLS_SET_OFFCHAN_MODE_CMD,
+	WMI_TAG_SET_LED_FLASHING_CMD,
+	WMI_TAG_MDNS_OFFLOAD_CMD,
+	WMI_TAG_MDNS_SET_FQDN_CMD,
+	WMI_TAG_MDNS_SET_RESP_CMD,
+	WMI_TAG_MDNS_GET_STATS_CMD,
+	WMI_TAG_MDNS_STATS_EVENT,
+	WMI_TAG_ROAM_INVOKE_CMD,
+	WMI_TAG_PDEV_RESUME_EVENT,
+	WMI_TAG_PDEV_SET_ANTENNA_DIVERSITY_CMD,
+	WMI_TAG_SAP_OFL_ENABLE_CMD,
+	WMI_TAG_SAP_OFL_ADD_STA_EVENT,
+	WMI_TAG_SAP_OFL_DEL_STA_EVENT,
+	WMI_TAG_APFIND_CMD_PARAM,
+	WMI_TAG_APFIND_EVENT_HDR,
+	WMI_TAG_OCB_SET_SCHED_CMD,
+	WMI_TAG_OCB_SET_SCHED_EVENT,
+	WMI_TAG_OCB_SET_CONFIG_CMD,
+	WMI_TAG_OCB_SET_CONFIG_RESP_EVENT,
+	WMI_TAG_OCB_SET_UTC_TIME_CMD,
+	WMI_TAG_OCB_START_TIMING_ADVERT_CMD,
+	WMI_TAG_OCB_STOP_TIMING_ADVERT_CMD,
+	WMI_TAG_OCB_GET_TSF_TIMER_CMD,
+	WMI_TAG_OCB_GET_TSF_TIMER_RESP_EVENT,
+	WMI_TAG_DCC_GET_STATS_CMD,
+	WMI_TAG_DCC_CHANNEL_STATS_REQUEST,
+	WMI_TAG_DCC_GET_STATS_RESP_EVENT,
+	WMI_TAG_DCC_CLEAR_STATS_CMD,
+	WMI_TAG_DCC_UPDATE_NDL_CMD,
+	WMI_TAG_DCC_UPDATE_NDL_RESP_EVENT,
+	WMI_TAG_DCC_STATS_EVENT,
+	WMI_TAG_OCB_CHANNEL,
+	WMI_TAG_OCB_SCHEDULE_ELEMENT,
+	WMI_TAG_DCC_NDL_STATS_PER_CHANNEL,
+	WMI_TAG_DCC_NDL_CHAN,
+	WMI_TAG_QOS_PARAMETER,
+	WMI_TAG_DCC_NDL_ACTIVE_STATE_CONFIG,
+	WMI_TAG_ROAM_SCAN_EXTENDED_THRESHOLD_PARAM,
+	WMI_TAG_ROAM_FILTER,
+	WMI_TAG_PASSPOINT_CONFIG_CMD,
+	WMI_TAG_PASSPOINT_EVENT_HDR,
+	WMI_TAG_EXTSCAN_CONFIGURE_HOTLIST_SSID_MONITOR_CMD,
+	WMI_TAG_EXTSCAN_HOTLIST_SSID_MATCH_EVENT,
+	WMI_TAG_VDEV_TSF_TSTAMP_ACTION_CMD,
+	WMI_TAG_VDEV_TSF_REPORT_EVENT,
+	WMI_TAG_GET_FW_MEM_DUMP,
+	WMI_TAG_UPDATE_FW_MEM_DUMP,
+	WMI_TAG_FW_MEM_DUMP_PARAMS,
+	WMI_TAG_DEBUG_MESG_FLUSH,
+	WMI_TAG_DEBUG_MESG_FLUSH_COMPLETE,
+	WMI_TAG_PEER_SET_RATE_REPORT_CONDITION,
+	WMI_TAG_ROAM_SUBNET_CHANGE_CONFIG,
+	WMI_TAG_VDEV_SET_IE_CMD,
+	WMI_TAG_RSSI_BREACH_MONITOR_CONFIG,
+	WMI_TAG_RSSI_BREACH_EVENT,
+	WMI_TAG_WOW_EVENT_INITIAL_WAKEUP,
+	WMI_TAG_SOC_SET_PCL_CMD,
+	WMI_TAG_SOC_SET_HW_MODE_CMD,
+	WMI_TAG_SOC_SET_HW_MODE_RESPONSE_EVENT,
+	WMI_TAG_SOC_HW_MODE_TRANSITION_EVENT,
+	WMI_TAG_VDEV_TXRX_STREAMS,
+	WMI_TAG_SOC_SET_HW_MODE_RESPONSE_VDEV_MAC_ENTRY,
+	WMI_TAG_SOC_SET_DUAL_MAC_CONFIG_CMD,
+	WMI_TAG_SOC_SET_DUAL_MAC_CONFIG_RESPONSE_EVENT,
+	WMI_TAG_WOW_IOAC_SOCK_PATTERN_T,
+	WMI_TAG_WOW_ENABLE_ICMPV6_NA_FLT_CMD,
+	WMI_TAG_DIAG_EVENT_LOG_CONFIG,
+	WMI_TAG_DIAG_EVENT_LOG_SUPPORTED_EVENT_FIXED_PARAMS,
+	WMI_TAG_PACKET_FILTER_CONFIG,
+	WMI_TAG_PACKET_FILTER_ENABLE,
+	WMI_TAG_SAP_SET_BLACKLIST_PARAM_CMD,
+	WMI_TAG_MGMT_TX_SEND_CMD,
+	WMI_TAG_MGMT_TX_COMPL_EVENT,
+	WMI_TAG_SOC_SET_ANTENNA_MODE_CMD,
+	WMI_TAG_WOW_UDP_SVC_OFLD_CMD,
+	WMI_TAG_LRO_INFO_CMD,
+	WMI_TAG_ROAM_EARLYSTOP_RSSI_THRES_PARAM,
+	WMI_TAG_SERVICE_READY_EXT_EVENT,
+	WMI_TAG_MAWC_SENSOR_REPORT_IND_CMD,
+	WMI_TAG_MAWC_ENABLE_SENSOR_EVENT,
+	WMI_TAG_ROAM_CONFIGURE_MAWC_CMD,
+	WMI_TAG_NLO_CONFIGURE_MAWC_CMD,
+	WMI_TAG_EXTSCAN_CONFIGURE_MAWC_CMD,
+	WMI_TAG_PEER_ASSOC_CONF_EVENT,
+	WMI_TAG_WOW_HOSTWAKEUP_GPIO_PIN_PATTERN_CONFIG_CMD,
+	WMI_TAG_AP_PS_EGAP_PARAM_CMD,
+	WMI_TAG_AP_PS_EGAP_INFO_EVENT,
+	WMI_TAG_PMF_OFFLOAD_SET_SA_QUERY_CMD,
+	WMI_TAG_TRANSFER_DATA_TO_FLASH_CMD,
+	WMI_TAG_TRANSFER_DATA_TO_FLASH_COMPLETE_EVENT,
+	WMI_TAG_SCPC_EVENT,
+	WMI_TAG_AP_PS_EGAP_INFO_CHAINMASK_LIST,
+	WMI_TAG_STA_SMPS_FORCE_MODE_COMPLETE_EVENT,
+	WMI_TAG_BPF_GET_CAPABILITY_CMD,
+	WMI_TAG_BPF_CAPABILITY_INFO_EVT,
+	WMI_TAG_BPF_GET_VDEV_STATS_CMD,
+	WMI_TAG_BPF_VDEV_STATS_INFO_EVT,
+	WMI_TAG_BPF_SET_VDEV_INSTRUCTIONS_CMD,
+	WMI_TAG_BPF_DEL_VDEV_INSTRUCTIONS_CMD,
+	WMI_TAG_VDEV_DELETE_RESP_EVENT,
+	WMI_TAG_PEER_DELETE_RESP_EVENT,
+	WMI_TAG_ROAM_DENSE_THRES_PARAM,
+	WMI_TAG_ENLO_CANDIDATE_SCORE_PARAM,
+	WMI_TAG_PEER_UPDATE_WDS_ENTRY_CMD,
+	WMI_TAG_VDEV_CONFIG_RATEMASK,
+	WMI_TAG_PDEV_FIPS_CMD,
+	WMI_TAG_PDEV_SMART_ANT_ENABLE_CMD,
+	WMI_TAG_PDEV_SMART_ANT_SET_RX_ANTENNA_CMD,
+	WMI_TAG_PEER_SMART_ANT_SET_TX_ANTENNA_CMD,
+	WMI_TAG_PEER_SMART_ANT_SET_TRAIN_ANTENNA_CMD,
+	WMI_TAG_PEER_SMART_ANT_SET_NODE_CONFIG_OPS_CMD,
+	WMI_TAG_PDEV_SET_ANT_SWITCH_TBL_CMD,
+	WMI_TAG_PDEV_SET_CTL_TABLE_CMD,
+	WMI_TAG_PDEV_SET_MIMOGAIN_TABLE_CMD,
+	WMI_TAG_FWTEST_SET_PARAM_CMD,
+	WMI_TAG_PEER_ATF_REQUEST,
+	WMI_TAG_VDEV_ATF_REQUEST,
+	WMI_TAG_PDEV_GET_ANI_CCK_CONFIG_CMD,
+	WMI_TAG_PDEV_GET_ANI_OFDM_CONFIG_CMD,
+	WMI_TAG_INST_RSSI_STATS_RESP,
+	WMI_TAG_MED_UTIL_REPORT_EVENT,
+	WMI_TAG_PEER_STA_PS_STATECHANGE_EVENT,
+	WMI_TAG_WDS_ADDR_EVENT,
+	WMI_TAG_PEER_RATECODE_LIST_EVENT,
+	WMI_TAG_PDEV_NFCAL_POWER_ALL_CHANNELS_EVENT,
+	WMI_TAG_PDEV_TPC_EVENT,
+	WMI_TAG_ANI_OFDM_EVENT,
+	WMI_TAG_ANI_CCK_EVENT,
+	WMI_TAG_PDEV_CHANNEL_HOPPING_EVENT,
+	WMI_TAG_PDEV_FIPS_EVENT,
+	WMI_TAG_ATF_PEER_INFO,
+	WMI_TAG_PDEV_GET_TPC_CMD,
+	WMI_TAG_VDEV_FILTER_NRP_CONFIG_CMD,
+	WMI_TAG_QBOOST_CFG_CMD,
+	WMI_TAG_PDEV_SMART_ANT_GPIO_HANDLE,
+	WMI_TAG_PEER_SMART_ANT_SET_TX_ANTENNA_SERIES,
+	WMI_TAG_PEER_SMART_ANT_SET_TRAIN_ANTENNA_PARAM,
+	WMI_TAG_PDEV_SET_ANT_CTRL_CHAIN,
+	WMI_TAG_PEER_CCK_OFDM_RATE_INFO,
+	WMI_TAG_PEER_MCS_RATE_INFO,
+	WMI_TAG_PDEV_NFCAL_POWER_ALL_CHANNELS_NFDBR,
+	WMI_TAG_PDEV_NFCAL_POWER_ALL_CHANNELS_NFDBM,
+	WMI_TAG_PDEV_NFCAL_POWER_ALL_CHANNELS_FREQNUM,
+	WMI_TAG_MU_REPORT_TOTAL_MU,
+	WMI_TAG_VDEV_SET_DSCP_TID_MAP_CMD,
+	WMI_TAG_ROAM_SET_MBO,
+	WMI_TAG_MIB_STATS_ENABLE_CMD,
+	WMI_TAG_NAN_DISC_IFACE_CREATED_EVENT,
+	WMI_TAG_NAN_DISC_IFACE_DELETED_EVENT,
+	WMI_TAG_NAN_STARTED_CLUSTER_EVENT,
+	WMI_TAG_NAN_JOINED_CLUSTER_EVENT,
+	WMI_TAG_NDI_GET_CAP_REQ,
+	WMI_TAG_NDP_INITIATOR_REQ,
+	WMI_TAG_NDP_RESPONDER_REQ,
+	WMI_TAG_NDP_END_REQ,
+	WMI_TAG_NDI_CAP_RSP_EVENT,
+	WMI_TAG_NDP_INITIATOR_RSP_EVENT,
+	WMI_TAG_NDP_RESPONDER_RSP_EVENT,
+	WMI_TAG_NDP_END_RSP_EVENT,
+	WMI_TAG_NDP_INDICATION_EVENT,
+	WMI_TAG_NDP_CONFIRM_EVENT,
+	WMI_TAG_NDP_END_INDICATION_EVENT,
+	WMI_TAG_VDEV_SET_QUIET_CMD,
+	WMI_TAG_PDEV_SET_PCL_CMD,
+	WMI_TAG_PDEV_SET_HW_MODE_CMD,
+	WMI_TAG_PDEV_SET_MAC_CONFIG_CMD,
+	WMI_TAG_PDEV_SET_ANTENNA_MODE_CMD,
+	WMI_TAG_PDEV_SET_HW_MODE_RESPONSE_EVENT,
+	WMI_TAG_PDEV_HW_MODE_TRANSITION_EVENT,
+	WMI_TAG_PDEV_SET_HW_MODE_RESPONSE_VDEV_MAC_ENTRY,
+	WMI_TAG_PDEV_SET_MAC_CONFIG_RESPONSE_EVENT,
+	WMI_TAG_COEX_CONFIG_CMD,
+	WMI_TAG_CONFIG_ENHANCED_MCAST_FILTER,
+	WMI_TAG_CHAN_AVOID_RPT_ALLOW_CMD,
+	WMI_TAG_SET_PERIODIC_CHANNEL_STATS_CONFIG,
+	WMI_TAG_VDEV_SET_CUSTOM_AGGR_SIZE_CMD,
+	WMI_TAG_PDEV_WAL_POWER_DEBUG_CMD,
+	WMI_TAG_MAC_PHY_CAPABILITIES,
+	WMI_TAG_HW_MODE_CAPABILITIES,
+	WMI_TAG_SOC_MAC_PHY_HW_MODE_CAPS,
+	WMI_TAG_HAL_REG_CAPABILITIES_EXT,
+	WMI_TAG_SOC_HAL_REG_CAPABILITIES,
+	WMI_TAG_VDEV_WISA_CMD,
+	WMI_TAG_TX_POWER_LEVEL_STATS_EVT,
+	WMI_TAG_SCAN_ADAPTIVE_DWELL_PARAMETERS_TLV,
+	WMI_TAG_SCAN_ADAPTIVE_DWELL_CONFIG,
+	WMI_TAG_WOW_SET_ACTION_WAKE_UP_CMD,
+	WMI_TAG_NDP_END_RSP_PER_NDI,
+	WMI_TAG_PEER_BWF_REQUEST,
+	WMI_TAG_BWF_PEER_INFO,
+	WMI_TAG_DBGLOG_TIME_STAMP_SYNC_CMD,
+	WMI_TAG_RMC_SET_LEADER_CMD,
+	WMI_TAG_RMC_MANUAL_LEADER_EVENT,
+	WMI_TAG_PER_CHAIN_RSSI_STATS,
+	WMI_TAG_RSSI_STATS,
+	WMI_TAG_P2P_LO_START_CMD,
+	WMI_TAG_P2P_LO_STOP_CMD,
+	WMI_TAG_P2P_LO_STOPPED_EVENT,
+	WMI_TAG_REORDER_QUEUE_SETUP_CMD,
+	WMI_TAG_REORDER_QUEUE_REMOVE_CMD,
+	WMI_TAG_SET_MULTIPLE_MCAST_FILTER_CMD,
+	WMI_TAG_MGMT_TX_COMPL_BUNDLE_EVENT,
+	WMI_TAG_READ_DATA_FROM_FLASH_CMD,
+	WMI_TAG_READ_DATA_FROM_FLASH_EVENT,
+	WMI_TAG_PDEV_SET_REORDER_TIMEOUT_VAL_CMD,
+	WMI_TAG_PEER_SET_RX_BLOCKSIZE_CMD,
+	WMI_TAG_PDEV_SET_WAKEUP_CONFIG_CMDID,
+	WMI_TAG_TLV_BUF_LEN_PARAM,
+	WMI_TAG_SERVICE_AVAILABLE_EVENT,
+	WMI_TAG_PEER_ANTDIV_INFO_REQ_CMD,
+	WMI_TAG_PEER_ANTDIV_INFO_EVENT,
+	WMI_TAG_PEER_ANTDIV_INFO,
+	WMI_TAG_PDEV_GET_ANTDIV_STATUS_CMD,
+	WMI_TAG_PDEV_ANTDIV_STATUS_EVENT,
+	WMI_TAG_MNT_FILTER_CMD,
+	WMI_TAG_GET_CHIP_POWER_STATS_CMD,
+	WMI_TAG_PDEV_CHIP_POWER_STATS_EVENT,
+	WMI_TAG_COEX_GET_ANTENNA_ISOLATION_CMD,
+	WMI_TAG_COEX_REPORT_ISOLATION_EVENT,
+	WMI_TAG_CHAN_CCA_STATS,
+	WMI_TAG_PEER_SIGNAL_STATS,
+	WMI_TAG_TX_STATS,
+	WMI_TAG_PEER_AC_TX_STATS,
+	WMI_TAG_RX_STATS,
+	WMI_TAG_PEER_AC_RX_STATS,
+	WMI_TAG_REPORT_STATS_EVENT,
+	WMI_TAG_CHAN_CCA_STATS_THRESH,
+	WMI_TAG_PEER_SIGNAL_STATS_THRESH,
+	WMI_TAG_TX_STATS_THRESH,
+	WMI_TAG_RX_STATS_THRESH,
+	WMI_TAG_PDEV_SET_STATS_THRESHOLD_CMD,
+	WMI_TAG_REQUEST_WLAN_STATS_CMD,
+	WMI_TAG_RX_AGGR_FAILURE_EVENT,
+	WMI_TAG_RX_AGGR_FAILURE_INFO,
+	WMI_TAG_VDEV_ENCRYPT_DECRYPT_DATA_REQ_CMD,
+	WMI_TAG_VDEV_ENCRYPT_DECRYPT_DATA_RESP_EVENT,
+	WMI_TAG_PDEV_BAND_TO_MAC,
+	WMI_TAG_TBTT_OFFSET_INFO,
+	WMI_TAG_TBTT_OFFSET_EXT_EVENT,
+	WMI_TAG_SAR_LIMITS_CMD,
+	WMI_TAG_SAR_LIMIT_CMD_ROW,
+	WMI_TAG_PDEV_DFS_PHYERR_OFFLOAD_ENABLE_CMD,
+	WMI_TAG_PDEV_DFS_PHYERR_OFFLOAD_DISABLE_CMD,
+	WMI_TAG_VDEV_ADFS_CH_CFG_CMD,
+	WMI_TAG_VDEV_ADFS_OCAC_ABORT_CMD,
+	WMI_TAG_PDEV_DFS_RADAR_DETECTION_EVENT,
+	WMI_TAG_VDEV_ADFS_OCAC_COMPLETE_EVENT,
+	WMI_TAG_VDEV_DFS_CAC_COMPLETE_EVENT,
+	WMI_TAG_VENDOR_OUI,
+	WMI_TAG_REQUEST_RCPI_CMD,
+	WMI_TAG_UPDATE_RCPI_EVENT,
+	WMI_TAG_REQUEST_PEER_STATS_INFO_CMD,
+	WMI_TAG_PEER_STATS_INFO,
+	WMI_TAG_PEER_STATS_INFO_EVENT,
+	WMI_TAG_PKGID_EVENT,
+	WMI_TAG_CONNECTED_NLO_RSSI_PARAMS,
+	WMI_TAG_SET_CURRENT_COUNTRY_CMD,
+	WMI_TAG_REGULATORY_RULE_STRUCT,
+	WMI_TAG_REG_CHAN_LIST_CC_EVENT,
+	WMI_TAG_11D_SCAN_START_CMD,
+	WMI_TAG_11D_SCAN_STOP_CMD,
+	WMI_TAG_11D_NEW_COUNTRY_EVENT,
+	WMI_TAG_REQUEST_RADIO_CHAN_STATS_CMD,
+	WMI_TAG_RADIO_CHAN_STATS,
+	WMI_TAG_RADIO_CHAN_STATS_EVENT,
+	WMI_TAG_ROAM_PER_CONFIG,
+	WMI_TAG_VDEV_ADD_MAC_ADDR_TO_RX_FILTER_CMD,
+	WMI_TAG_VDEV_ADD_MAC_ADDR_TO_RX_FILTER_STATUS_EVENT,
+	WMI_TAG_BPF_SET_VDEV_ACTIVE_MODE_CMD,
+	WMI_TAG_HW_DATA_FILTER_CMD,
+	WMI_TAG_CONNECTED_NLO_BSS_BAND_RSSI_PREF,
+	WMI_TAG_PEER_OPER_MODE_CHANGE_EVENT,
+	WMI_TAG_CHIP_POWER_SAVE_FAILURE_DETECTED,
+	WMI_TAG_PDEV_MULTIPLE_VDEV_RESTART_REQUEST_CMD,
+	WMI_TAG_PDEV_CSA_SWITCH_COUNT_STATUS_EVENT,
+	WMI_TAG_PDEV_UPDATE_PKT_ROUTING_CMD,
+	WMI_TAG_PDEV_CHECK_CAL_VERSION_CMD,
+	WMI_TAG_PDEV_CHECK_CAL_VERSION_EVENT,
+	WMI_TAG_PDEV_SET_DIVERSITY_GAIN_CMD,
+	WMI_TAG_MAC_PHY_CHAINMASK_COMBO,
+	WMI_TAG_MAC_PHY_CHAINMASK_CAPABILITY,
+	WMI_TAG_VDEV_SET_ARP_STATS_CMD,
+	WMI_TAG_VDEV_GET_ARP_STATS_CMD,
+	WMI_TAG_VDEV_GET_ARP_STATS_EVENT,
+	WMI_TAG_IFACE_OFFLOAD_STATS,
+	WMI_TAG_REQUEST_STATS_CMD_SUB_STRUCT_PARAM,
+	WMI_TAG_RSSI_CTL_EXT,
+	WMI_TAG_SINGLE_PHYERR_EXT_RX_HDR,
+	WMI_TAG_COEX_BT_ACTIVITY_EVENT,
+	WMI_TAG_VDEV_GET_TX_POWER_CMD,
+	WMI_TAG_VDEV_TX_POWER_EVENT,
+	WMI_TAG_OFFCHAN_DATA_TX_COMPL_EVENT,
+	WMI_TAG_OFFCHAN_DATA_TX_SEND_CMD,
+	WMI_TAG_TX_SEND_PARAMS,
+	WMI_TAG_HE_RATE_SET,
+	WMI_TAG_CONGESTION_STATS,
+	WMI_TAG_SET_INIT_COUNTRY_CMD,
+	WMI_TAG_SCAN_DBS_DUTY_CYCLE,
+	WMI_TAG_SCAN_DBS_DUTY_CYCLE_PARAM_TLV,
+	WMI_TAG_PDEV_DIV_GET_RSSI_ANTID,
+	WMI_TAG_THERM_THROT_CONFIG_REQUEST,
+	WMI_TAG_THERM_THROT_LEVEL_CONFIG_INFO,
+	WMI_TAG_THERM_THROT_STATS_EVENT,
+	WMI_TAG_THERM_THROT_LEVEL_STATS_INFO,
+	WMI_TAG_PDEV_DIV_RSSI_ANTID_EVENT,
+	WMI_TAG_OEM_DMA_RING_CAPABILITIES,
+	WMI_TAG_OEM_DMA_RING_CFG_REQ,
+	WMI_TAG_OEM_DMA_RING_CFG_RSP,
+	WMI_TAG_OEM_INDIRECT_DATA,
+	WMI_TAG_OEM_DMA_BUF_RELEASE,
+	WMI_TAG_OEM_DMA_BUF_RELEASE_ENTRY,
+	WMI_TAG_PDEV_BSS_CHAN_INFO_REQUEST,
+	WMI_TAG_PDEV_BSS_CHAN_INFO_EVENT,
+	WMI_TAG_ROAM_LCA_DISALLOW_CONFIG,
+	WMI_TAG_VDEV_LIMIT_OFFCHAN_CMD,
+	WMI_TAG_ROAM_RSSI_REJECTION_OCE_CONFIG,
+	WMI_TAG_UNIT_TEST_EVENT,
+	WMI_TAG_ROAM_FILS_OFFLOAD,
+	WMI_TAG_PDEV_UPDATE_PMK_CACHE_CMD,
+	WMI_TAG_PMK_CACHE,
+	WMI_TAG_PDEV_UPDATE_FILS_HLP_PKT_CMD,
+	WMI_TAG_ROAM_FILS_SYNCH,
+	WMI_TAG_GTK_OFFLOAD_EXTENDED,
+	WMI_TAG_ROAM_BG_SCAN_ROAMING,
+	WMI_TAG_OIC_PING_OFFLOAD_PARAMS_CMD,
+	WMI_TAG_OIC_PING_OFFLOAD_SET_ENABLE_CMD,
+	WMI_TAG_OIC_PING_HANDOFF_EVENT,
+	WMI_TAG_DHCP_LEASE_RENEW_OFFLOAD_CMD,
+	WMI_TAG_DHCP_LEASE_RENEW_EVENT,
+	WMI_TAG_BTM_CONFIG,
+	WMI_TAG_DEBUG_MESG_FW_DATA_STALL,
+	WMI_TAG_WLM_CONFIG_CMD,
+	WMI_TAG_PDEV_UPDATE_CTLTABLE_REQUEST,
+	WMI_TAG_PDEV_UPDATE_CTLTABLE_EVENT,
+	WMI_TAG_ROAM_CND_SCORING_PARAM,
+	WMI_TAG_PDEV_CONFIG_VENDOR_OUI_ACTION,
+	WMI_TAG_VENDOR_OUI_EXT,
+	WMI_TAG_ROAM_SYNCH_FRAME_EVENT,
+	WMI_TAG_FD_SEND_FROM_HOST_CMD,
+	WMI_TAG_ENABLE_FILS_CMD,
+	WMI_TAG_HOST_SWFDA_EVENT,
+	WMI_TAG_BCN_OFFLOAD_CTRL_CMD,
+	WMI_TAG_PDEV_SET_AC_TX_QUEUE_OPTIMIZED_CMD,
+	WMI_TAG_STATS_PERIOD,
+	WMI_TAG_NDL_SCHEDULE_UPDATE,
+	WMI_TAG_PEER_TID_MSDUQ_QDEPTH_THRESH_UPDATE_CMD,
+	WMI_TAG_MSDUQ_QDEPTH_THRESH_UPDATE,
+	WMI_TAG_PDEV_SET_RX_FILTER_PROMISCUOUS_CMD,
+	WMI_TAG_SAR2_RESULT_EVENT,
+	WMI_TAG_SAR_CAPABILITIES,
+	WMI_TAG_SAP_OBSS_DETECTION_CFG_CMD,
+	WMI_TAG_SAP_OBSS_DETECTION_INFO_EVT,
+	WMI_TAG_DMA_RING_CAPABILITIES,
+	WMI_TAG_DMA_RING_CFG_REQ,
+	WMI_TAG_DMA_RING_CFG_RSP,
+	WMI_TAG_DMA_BUF_RELEASE,
+	WMI_TAG_DMA_BUF_RELEASE_ENTRY,
+	WMI_TAG_SAR_GET_LIMITS_CMD,
+	WMI_TAG_SAR_GET_LIMITS_EVENT,
+	WMI_TAG_SAR_GET_LIMITS_EVENT_ROW,
+	WMI_TAG_OFFLOAD_11K_REPORT,
+	WMI_TAG_INVOKE_NEIGHBOR_REPORT,
+	WMI_TAG_NEIGHBOR_REPORT_OFFLOAD,
+	WMI_TAG_VDEV_SET_CONNECTIVITY_CHECK_STATS,
+	WMI_TAG_VDEV_GET_CONNECTIVITY_CHECK_STATS,
+	WMI_TAG_BPF_SET_VDEV_ENABLE_CMD,
+	WMI_TAG_BPF_SET_VDEV_WORK_MEMORY_CMD,
+	WMI_TAG_BPF_GET_VDEV_WORK_MEMORY_CMD,
+	WMI_TAG_BPF_GET_VDEV_WORK_MEMORY_RESP_EVT,
+	WMI_TAG_PDEV_GET_NFCAL_POWER,
+	WMI_TAG_BSS_COLOR_CHANGE_ENABLE,
+	WMI_TAG_OBSS_COLOR_COLLISION_DET_CONFIG,
+	WMI_TAG_OBSS_COLOR_COLLISION_EVT,
+	WMI_TAG_RUNTIME_DPD_RECAL_CMD,
+	WMI_TAG_TWT_ENABLE_CMD,
+	WMI_TAG_TWT_DISABLE_CMD,
+	WMI_TAG_TWT_ADD_DIALOG_CMD,
+	WMI_TAG_TWT_DEL_DIALOG_CMD,
+	WMI_TAG_TWT_PAUSE_DIALOG_CMD,
+	WMI_TAG_TWT_RESUME_DIALOG_CMD,
+	WMI_TAG_TWT_ENABLE_COMPLETE_EVENT,
+	WMI_TAG_TWT_DISABLE_COMPLETE_EVENT,
+	WMI_TAG_TWT_ADD_DIALOG_COMPLETE_EVENT,
+	WMI_TAG_TWT_DEL_DIALOG_COMPLETE_EVENT,
+	WMI_TAG_TWT_PAUSE_DIALOG_COMPLETE_EVENT,
+	WMI_TAG_TWT_RESUME_DIALOG_COMPLETE_EVENT,
+	WMI_TAG_REQUEST_ROAM_SCAN_STATS_CMD,
+	WMI_TAG_ROAM_SCAN_STATS_EVENT,
+	WMI_TAG_PEER_TID_CONFIGURATIONS_CMD,
+	WMI_TAG_VDEV_SET_CUSTOM_SW_RETRY_TH_CMD,
+	WMI_TAG_GET_TPC_POWER_CMD,
+	WMI_TAG_GET_TPC_POWER_EVENT,
+	WMI_TAG_DMA_BUF_RELEASE_SPECTRAL_META_DATA,
+	WMI_TAG_MOTION_DET_CONFIG_PARAMS_CMD,
+	WMI_TAG_MOTION_DET_BASE_LINE_CONFIG_PARAMS_CMD,
+	WMI_TAG_MOTION_DET_START_STOP_CMD,
+	WMI_TAG_MOTION_DET_BASE_LINE_START_STOP_CMD,
+	WMI_TAG_MOTION_DET_EVENT,
+	WMI_TAG_MOTION_DET_BASE_LINE_EVENT,
+	WMI_TAG_NDP_TRANSPORT_IP,
+	WMI_TAG_OBSS_SPATIAL_REUSE_SET_CMD,
+	WMI_TAG_ESP_ESTIMATE_EVENT,
+	WMI_TAG_NAN_HOST_CONFIG,
+	WMI_TAG_SPECTRAL_BIN_SCALING_PARAMS,
+	WMI_TAG_PEER_CFR_CAPTURE_CMD,
+	WMI_TAG_PEER_CHAN_WIDTH_SWITCH_CMD,
+	WMI_TAG_CHAN_WIDTH_PEER_LIST,
+	WMI_TAG_OBSS_SPATIAL_REUSE_SET_DEF_OBSS_THRESH_CMD,
+	WMI_TAG_PDEV_HE_TB_ACTION_FRM_CMD,
+	WMI_TAG_PEER_EXTD2_STATS,
+	WMI_TAG_HPCS_PULSE_START_CMD,
+	WMI_TAG_PDEV_CTL_FAILSAFE_CHECK_EVENT,
+	WMI_TAG_VDEV_CHAINMASK_CONFIG_CMD,
+	WMI_TAG_VDEV_BCN_OFFLOAD_QUIET_CONFIG_CMD,
+	WMI_TAG_NAN_EVENT_INFO,
+	WMI_TAG_NDP_CHANNEL_INFO,
+	WMI_TAG_NDP_CMD,
+	WMI_TAG_NDP_EVENT,
+	/* TODO add all the missing cmds */
+	WMI_TAG_PDEV_PEER_PKTLOG_FILTER_CMD = 0x301,
+	WMI_TAG_PDEV_PEER_PKTLOG_FILTER_INFO,
+	WMI_TAG_MAX
+};
+
+enum wmi_tlv_service {
+	WMI_TLV_SERVICE_BEACON_OFFLOAD = 0,
+	WMI_TLV_SERVICE_SCAN_OFFLOAD = 1,
+	WMI_TLV_SERVICE_ROAM_SCAN_OFFLOAD = 2,
+	WMI_TLV_SERVICE_BCN_MISS_OFFLOAD = 3,
+	WMI_TLV_SERVICE_STA_PWRSAVE = 4,
+	WMI_TLV_SERVICE_STA_ADVANCED_PWRSAVE = 5,
+	WMI_TLV_SERVICE_AP_UAPSD = 6,
+	WMI_TLV_SERVICE_AP_DFS = 7,
+	WMI_TLV_SERVICE_11AC = 8,
+	WMI_TLV_SERVICE_BLOCKACK = 9,
+	WMI_TLV_SERVICE_PHYERR = 10,
+	WMI_TLV_SERVICE_BCN_FILTER = 11,
+	WMI_TLV_SERVICE_RTT = 12,
+	WMI_TLV_SERVICE_WOW = 13,
+	WMI_TLV_SERVICE_RATECTRL_CACHE = 14,
+	WMI_TLV_SERVICE_IRAM_TIDS = 15,
+	WMI_TLV_SERVICE_ARPNS_OFFLOAD = 16,
+	WMI_TLV_SERVICE_NLO = 17,
+	WMI_TLV_SERVICE_GTK_OFFLOAD = 18,
+	WMI_TLV_SERVICE_SCAN_SCH = 19,
+	WMI_TLV_SERVICE_CSA_OFFLOAD = 20,
+	WMI_TLV_SERVICE_CHATTER = 21,
+	WMI_TLV_SERVICE_COEX_FREQAVOID = 22,
+	WMI_TLV_SERVICE_PACKET_POWER_SAVE = 23,
+	WMI_TLV_SERVICE_FORCE_FW_HANG = 24,
+	WMI_TLV_SERVICE_GPIO = 25,
+	WMI_TLV_SERVICE_STA_DTIM_PS_MODULATED_DTIM = 26,
+	WMI_STA_UAPSD_BASIC_AUTO_TRIG = 27,
+	WMI_STA_UAPSD_VAR_AUTO_TRIG = 28,
+	WMI_TLV_SERVICE_STA_KEEP_ALIVE = 29,
+	WMI_TLV_SERVICE_TX_ENCAP = 30,
+	WMI_TLV_SERVICE_AP_PS_DETECT_OUT_OF_SYNC = 31,
+	WMI_TLV_SERVICE_EARLY_RX = 32,
+	WMI_TLV_SERVICE_STA_SMPS = 33,
+	WMI_TLV_SERVICE_FWTEST = 34,
+	WMI_TLV_SERVICE_STA_WMMAC = 35,
+	WMI_TLV_SERVICE_TDLS = 36,
+	WMI_TLV_SERVICE_BURST = 37,
+	WMI_TLV_SERVICE_MCC_BCN_INTERVAL_CHANGE = 38,
+	WMI_TLV_SERVICE_ADAPTIVE_OCS = 39,
+	WMI_TLV_SERVICE_BA_SSN_SUPPORT = 40,
+	WMI_TLV_SERVICE_FILTER_IPSEC_NATKEEPALIVE = 41,
+	WMI_TLV_SERVICE_WLAN_HB = 42,
+	WMI_TLV_SERVICE_LTE_ANT_SHARE_SUPPORT = 43,
+	WMI_TLV_SERVICE_BATCH_SCAN = 44,
+	WMI_TLV_SERVICE_QPOWER = 45,
+	WMI_TLV_SERVICE_PLMREQ = 46,
+	WMI_TLV_SERVICE_THERMAL_MGMT = 47,
+	WMI_TLV_SERVICE_RMC = 48,
+	WMI_TLV_SERVICE_MHF_OFFLOAD = 49,
+	WMI_TLV_SERVICE_COEX_SAR = 50,
+	WMI_TLV_SERVICE_BCN_TXRATE_OVERRIDE = 51,
+	WMI_TLV_SERVICE_NAN = 52,
+	WMI_TLV_SERVICE_L1SS_STAT = 53,
+	WMI_TLV_SERVICE_ESTIMATE_LINKSPEED = 54,
+	WMI_TLV_SERVICE_OBSS_SCAN = 55,
+	WMI_TLV_SERVICE_TDLS_OFFCHAN = 56,
+	WMI_TLV_SERVICE_TDLS_UAPSD_BUFFER_STA = 57,
+	WMI_TLV_SERVICE_TDLS_UAPSD_SLEEP_STA = 58,
+	WMI_TLV_SERVICE_IBSS_PWRSAVE = 59,
+	WMI_TLV_SERVICE_LPASS = 60,
+	WMI_TLV_SERVICE_EXTSCAN = 61,
+	WMI_TLV_SERVICE_D0WOW = 62,
+	WMI_TLV_SERVICE_HSOFFLOAD = 63,
+	WMI_TLV_SERVICE_ROAM_HO_OFFLOAD = 64,
+	WMI_TLV_SERVICE_RX_FULL_REORDER = 65,
+	WMI_TLV_SERVICE_DHCP_OFFLOAD = 66,
+	WMI_TLV_SERVICE_STA_RX_IPA_OFFLOAD_SUPPORT = 67,
+	WMI_TLV_SERVICE_MDNS_OFFLOAD = 68,
+	WMI_TLV_SERVICE_SAP_AUTH_OFFLOAD = 69,
+	WMI_TLV_SERVICE_DUAL_BAND_SIMULTANEOUS_SUPPORT = 70,
+	WMI_TLV_SERVICE_OCB = 71,
+	WMI_TLV_SERVICE_AP_ARPNS_OFFLOAD = 72,
+	WMI_TLV_SERVICE_PER_BAND_CHAINMASK_SUPPORT = 73,
+	WMI_TLV_SERVICE_PACKET_FILTER_OFFLOAD = 74,
+	WMI_TLV_SERVICE_MGMT_TX_HTT = 75,
+	WMI_TLV_SERVICE_MGMT_TX_WMI = 76,
+	WMI_TLV_SERVICE_EXT_MSG = 77,
+	WMI_TLV_SERVICE_MAWC = 78,
+	WMI_TLV_SERVICE_PEER_ASSOC_CONF = 79,
+	WMI_TLV_SERVICE_EGAP = 80,
+	WMI_TLV_SERVICE_STA_PMF_OFFLOAD = 81,
+	WMI_TLV_SERVICE_UNIFIED_WOW_CAPABILITY = 82,
+	WMI_TLV_SERVICE_ENHANCED_PROXY_STA = 83,
+	WMI_TLV_SERVICE_ATF = 84,
+	WMI_TLV_SERVICE_COEX_GPIO = 85,
+	WMI_TLV_SERVICE_AUX_SPECTRAL_INTF = 86,
+	WMI_TLV_SERVICE_AUX_CHAN_LOAD_INTF = 87,
+	WMI_TLV_SERVICE_BSS_CHANNEL_INFO_64 = 88,
+	WMI_TLV_SERVICE_ENTERPRISE_MESH = 89,
+	WMI_TLV_SERVICE_RESTRT_CHNL_SUPPORT = 90,
+	WMI_TLV_SERVICE_BPF_OFFLOAD = 91,
+	WMI_TLV_SERVICE_SYNC_DELETE_CMDS = 92,
+	WMI_TLV_SERVICE_SMART_ANTENNA_SW_SUPPORT = 93,
+	WMI_TLV_SERVICE_SMART_ANTENNA_HW_SUPPORT = 94,
+	WMI_TLV_SERVICE_RATECTRL_LIMIT_MAX_MIN_RATES = 95,
+	WMI_TLV_SERVICE_NAN_DATA = 96,
+	WMI_TLV_SERVICE_NAN_RTT = 97,
+	WMI_TLV_SERVICE_11AX = 98,
+	WMI_TLV_SERVICE_DEPRECATED_REPLACE = 99,
+	WMI_TLV_SERVICE_TDLS_CONN_TRACKER_IN_HOST_MODE = 100,
+	WMI_TLV_SERVICE_ENHANCED_MCAST_FILTER = 101,
+	WMI_TLV_SERVICE_PERIODIC_CHAN_STAT_SUPPORT = 102,
+	WMI_TLV_SERVICE_MESH_11S = 103,
+	WMI_TLV_SERVICE_HALF_RATE_QUARTER_RATE_SUPPORT = 104,
+	WMI_TLV_SERVICE_VDEV_RX_FILTER = 105,
+	WMI_TLV_SERVICE_P2P_LISTEN_OFFLOAD_SUPPORT = 106,
+	WMI_TLV_SERVICE_MARK_FIRST_WAKEUP_PACKET = 107,
+	WMI_TLV_SERVICE_MULTIPLE_MCAST_FILTER_SET = 108,
+	WMI_TLV_SERVICE_HOST_MANAGED_RX_REORDER = 109,
+	WMI_TLV_SERVICE_FLASH_RDWR_SUPPORT = 110,
+	WMI_TLV_SERVICE_WLAN_STATS_REPORT = 111,
+	WMI_TLV_SERVICE_TX_MSDU_ID_NEW_PARTITION_SUPPORT = 112,
+	WMI_TLV_SERVICE_DFS_PHYERR_OFFLOAD = 113,
+	WMI_TLV_SERVICE_RCPI_SUPPORT = 114,
+	WMI_TLV_SERVICE_FW_MEM_DUMP_SUPPORT = 115,
+	WMI_TLV_SERVICE_PEER_STATS_INFO = 116,
+	WMI_TLV_SERVICE_REGULATORY_DB = 117,
+	WMI_TLV_SERVICE_11D_OFFLOAD = 118,
+	WMI_TLV_SERVICE_HW_DATA_FILTERING = 119,
+	WMI_TLV_SERVICE_MULTIPLE_VDEV_RESTART = 120,
+	WMI_TLV_SERVICE_PKT_ROUTING = 121,
+	WMI_TLV_SERVICE_CHECK_CAL_VERSION = 122,
+	WMI_TLV_SERVICE_OFFCHAN_TX_WMI = 123,
+	WMI_TLV_SERVICE_8SS_TX_BFEE  =  124,
+	WMI_TLV_SERVICE_EXTENDED_NSS_SUPPORT = 125,
+	WMI_TLV_SERVICE_ACK_TIMEOUT = 126,
+	WMI_TLV_SERVICE_PDEV_BSS_CHANNEL_INFO_64 = 127,
+
+	WMI_MAX_SERVICE = 128,
+
+	WMI_TLV_SERVICE_CHAN_LOAD_INFO = 128,
+	WMI_TLV_SERVICE_TX_PPDU_INFO_STATS_SUPPORT = 129,
+	WMI_TLV_SERVICE_VDEV_LIMIT_OFFCHAN_SUPPORT = 130,
+	WMI_TLV_SERVICE_FILS_SUPPORT = 131,
+	WMI_TLV_SERVICE_WLAN_OIC_PING_OFFLOAD = 132,
+	WMI_TLV_SERVICE_WLAN_DHCP_RENEW = 133,
+	WMI_TLV_SERVICE_MAWC_SUPPORT = 134,
+	WMI_TLV_SERVICE_VDEV_LATENCY_CONFIG = 135,
+	WMI_TLV_SERVICE_PDEV_UPDATE_CTLTABLE_SUPPORT = 136,
+	WMI_TLV_SERVICE_PKTLOG_SUPPORT_OVER_HTT = 137,
+	WMI_TLV_SERVICE_VDEV_MULTI_GROUP_KEY_SUPPORT = 138,
+	WMI_TLV_SERVICE_SCAN_PHYMODE_SUPPORT = 139,
+	WMI_TLV_SERVICE_THERM_THROT = 140,
+	WMI_TLV_SERVICE_BCN_OFFLOAD_START_STOP_SUPPORT = 141,
+	WMI_TLV_SERVICE_WOW_WAKEUP_BY_TIMER_PATTERN = 142,
+	WMI_TLV_SERVICE_PEER_MAP_UNMAP_V2_SUPPORT = 143,
+	WMI_TLV_SERVICE_OFFCHAN_DATA_TID_SUPPORT = 144,
+	WMI_TLV_SERVICE_RX_PROMISC_ENABLE_SUPPORT = 145,
+	WMI_TLV_SERVICE_SUPPORT_DIRECT_DMA = 146,
+	WMI_TLV_SERVICE_AP_OBSS_DETECTION_OFFLOAD = 147,
+	WMI_TLV_SERVICE_11K_NEIGHBOUR_REPORT_SUPPORT = 148,
+	WMI_TLV_SERVICE_LISTEN_INTERVAL_OFFLOAD_SUPPORT = 149,
+	WMI_TLV_SERVICE_BSS_COLOR_OFFLOAD = 150,
+	WMI_TLV_SERVICE_RUNTIME_DPD_RECAL = 151,
+	WMI_TLV_SERVICE_STA_TWT = 152,
+	WMI_TLV_SERVICE_AP_TWT = 153,
+	WMI_TLV_SERVICE_GMAC_OFFLOAD_SUPPORT = 154,
+	WMI_TLV_SERVICE_SPOOF_MAC_SUPPORT = 155,
+	WMI_TLV_SERVICE_PEER_TID_CONFIGS_SUPPORT = 156,
+	WMI_TLV_SERVICE_VDEV_SWRETRY_PER_AC_CONFIG_SUPPORT = 157,
+	WMI_TLV_SERVICE_DUAL_BEACON_ON_SINGLE_MAC_SCC_SUPPORT = 158,
+	WMI_TLV_SERVICE_DUAL_BEACON_ON_SINGLE_MAC_MCC_SUPPORT = 159,
+	WMI_TLV_SERVICE_MOTION_DET = 160,
+	WMI_TLV_SERVICE_INFRA_MBSSID = 161,
+	WMI_TLV_SERVICE_OBSS_SPATIAL_REUSE = 162,
+	WMI_TLV_SERVICE_VDEV_DIFFERENT_BEACON_INTERVAL_SUPPORT = 163,
+	WMI_TLV_SERVICE_NAN_DBS_SUPPORT = 164,
+	WMI_TLV_SERVICE_NDI_DBS_SUPPORT = 165,
+	WMI_TLV_SERVICE_NAN_SAP_SUPPORT = 166,
+	WMI_TLV_SERVICE_NDI_SAP_SUPPORT = 167,
+	WMI_TLV_SERVICE_CFR_CAPTURE_SUPPORT = 168,
+	WMI_TLV_SERVICE_CFR_CAPTURE_IND_MSG_TYPE_1 = 169,
+	WMI_TLV_SERVICE_ESP_SUPPORT = 170,
+	WMI_TLV_SERVICE_PEER_CHWIDTH_CHANGE = 171,
+	WMI_TLV_SERVICE_WLAN_HPCS_PULSE = 172,
+	WMI_TLV_SERVICE_PER_VDEV_CHAINMASK_CONFIG_SUPPORT = 173,
+	WMI_TLV_SERVICE_TX_DATA_MGMT_ACK_RSSI = 174,
+	WMI_TLV_SERVICE_NAN_DISABLE_SUPPORT = 175,
+	WMI_TLV_SERVICE_HTT_H2T_NO_HTC_HDR_LEN_IN_MSG_LEN = 176,
+
+	WMI_MAX_EXT_SERVICE
+
+};
+
+enum {
+	WMI_SMPS_FORCED_MODE_NONE = 0,
+	WMI_SMPS_FORCED_MODE_DISABLED,
+	WMI_SMPS_FORCED_MODE_STATIC,
+	WMI_SMPS_FORCED_MODE_DYNAMIC
+};
+
+#define WMI_TPC_CHAINMASK_CONFIG_BAND_2G      0
+#define WMI_TPC_CHAINMASK_CONFIG_BAND_5G      1
+#define WMI_NUM_SUPPORTED_BAND_MAX 2
+
+#define WMI_PEER_MIMO_PS_STATE                          0x1
+#define WMI_PEER_AMPDU                                  0x2
+#define WMI_PEER_AUTHORIZE                              0x3
+#define WMI_PEER_CHWIDTH                                0x4
+#define WMI_PEER_NSS                                    0x5
+#define WMI_PEER_USE_4ADDR                              0x6
+#define WMI_PEER_MEMBERSHIP                             0x7
+#define WMI_PEER_USERPOS                                0x8
+#define WMI_PEER_CRIT_PROTO_HINT_ENABLED                0x9
+#define WMI_PEER_TX_FAIL_CNT_THR                        0xA
+#define WMI_PEER_SET_HW_RETRY_CTS2S                     0xB
+#define WMI_PEER_IBSS_ATIM_WINDOW_LENGTH                0xC
+#define WMI_PEER_PHYMODE                                0xD
+#define WMI_PEER_USE_FIXED_PWR                          0xE
+#define WMI_PEER_PARAM_FIXED_RATE                       0xF
+#define WMI_PEER_SET_MU_WHITELIST                       0x10
+#define WMI_PEER_SET_MAX_TX_RATE                        0x11
+#define WMI_PEER_SET_MIN_TX_RATE                        0x12
+#define WMI_PEER_SET_DEFAULT_ROUTING                    0x13
+
+/* slot time long */
+#define WMI_VDEV_SLOT_TIME_LONG         0x1
+/* slot time short */
+#define WMI_VDEV_SLOT_TIME_SHORT        0x2
+/* preablbe long */
+#define WMI_VDEV_PREAMBLE_LONG          0x1
+/* preablbe short */
+#define WMI_VDEV_PREAMBLE_SHORT         0x2
+
+enum wmi_peer_smps_state {
+	WMI_PEER_SMPS_PS_NONE = 0x0,
+	WMI_PEER_SMPS_STATIC  = 0x1,
+	WMI_PEER_SMPS_DYNAMIC = 0x2
+};
+
+enum wmi_peer_chwidth {
+	WMI_PEER_CHWIDTH_20MHZ = 0,
+	WMI_PEER_CHWIDTH_40MHZ = 1,
+	WMI_PEER_CHWIDTH_80MHZ = 2,
+	WMI_PEER_CHWIDTH_160MHZ = 3,
+};
+
+enum wmi_beacon_gen_mode {
+	WMI_BEACON_STAGGERED_MODE = 0,
+	WMI_BEACON_BURST_MODE = 1
+};
+
+struct wmi_host_pdev_band_to_mac {
+	u32 pdev_id;
+	u32 start_freq;
+	u32 end_freq;
+};
+
+struct ath11k_ppe_threshold {
+	u32 numss_m1;
+	u32 ru_bit_mask;
+	u32 ppet16_ppet8_ru3_ru0[PSOC_HOST_MAX_NUM_SS];
+};
+
+struct ath11k_service_ext_param {
+	u32 default_conc_scan_config_bits;
+	u32 default_fw_config_bits;
+	struct ath11k_ppe_threshold ppet;
+	u32 he_cap_info;
+	u32 mpdu_density;
+	u32 max_bssid_rx_filters;
+	u32 num_hw_modes;
+	u32 num_phy;
+};
+
+struct ath11k_hw_mode_caps {
+	u32 hw_mode_id;
+	u32 phy_id_map;
+	u32 hw_mode_config_type;
+};
+
+#define PSOC_HOST_MAX_PHY_SIZE (3)
+#define ATH11K_11B_SUPPORT                 BIT(0)
+#define ATH11K_11G_SUPPORT                 BIT(1)
+#define ATH11K_11A_SUPPORT                 BIT(2)
+#define ATH11K_11N_SUPPORT                 BIT(3)
+#define ATH11K_11AC_SUPPORT                BIT(4)
+#define ATH11K_11AX_SUPPORT                BIT(5)
+
+struct ath11k_hal_reg_capabilities_ext {
+	u32 phy_id;
+	u32 eeprom_reg_domain;
+	u32 eeprom_reg_domain_ext;
+	u32 regcap1;
+	u32 regcap2;
+	u32 wireless_modes;
+	u32 low_2ghz_chan;
+	u32 high_2ghz_chan;
+	u32 low_5ghz_chan;
+	u32 high_5ghz_chan;
+};
+
+#define WMI_HOST_MAX_PDEV 3
+
+struct wlan_host_mem_chunk {
+	u32 tlv_header;
+	u32 req_id;
+	u32 ptr;
+	u32 size;
+} __packed;
+
+struct wmi_host_mem_chunk {
+	void *vaddr;
+	dma_addr_t paddr;
+	u32 len;
+	u32 req_id;
+};
+
+struct wmi_init_cmd_param {
+	u32 tlv_header;
+	struct target_resource_config *res_cfg;
+	u8 num_mem_chunks;
+	struct wmi_host_mem_chunk *mem_chunks;
+	u32 hw_mode_id;
+	u32 num_band_to_mac;
+	struct wmi_host_pdev_band_to_mac band_to_mac[WMI_HOST_MAX_PDEV];
+};
+
+struct wmi_pdev_band_to_mac {
+	u32 tlv_header;
+	u32 pdev_id;
+	u32 start_freq;
+	u32 end_freq;
+} __packed;
+
+struct wmi_pdev_set_hw_mode_cmd_param {
+	u32 tlv_header;
+	u32 pdev_id;
+	u32 hw_mode_index;
+	u32 num_band_to_mac;
+} __packed;
+
+struct wmi_ppe_threshold {
+	u32 numss_m1; /** NSS - 1*/
+	union {
+		u32 ru_count;
+		u32 ru_mask;
+	} __packed;
+	u32 ppet16_ppet8_ru3_ru0[WMI_MAX_NUM_SS];
+} __packed;
+
+#define HW_BD_INFO_SIZE       5
+
+struct wmi_abi_version {
+	u32 abi_version_0;
+	u32 abi_version_1;
+	u32 abi_version_ns_0;
+	u32 abi_version_ns_1;
+	u32 abi_version_ns_2;
+	u32 abi_version_ns_3;
+} __packed;
+
+struct wmi_init_cmd {
+	u32 tlv_header;
+	struct wmi_abi_version host_abi_vers;
+	u32 num_host_mem_chunks;
+} __packed;
+
+struct wmi_resource_config {
+	u32 tlv_header;
+	u32 num_vdevs;
+	u32 num_peers;
+	u32 num_offload_peers;
+	u32 num_offload_reorder_buffs;
+	u32 num_peer_keys;
+	u32 num_tids;
+	u32 ast_skid_limit;
+	u32 tx_chain_mask;
+	u32 rx_chain_mask;
+	u32 rx_timeout_pri[4];
+	u32 rx_decap_mode;
+	u32 scan_max_pending_req;
+	u32 bmiss_offload_max_vdev;
+	u32 roam_offload_max_vdev;
+	u32 roam_offload_max_ap_profiles;
+	u32 num_mcast_groups;
+	u32 num_mcast_table_elems;
+	u32 mcast2ucast_mode;
+	u32 tx_dbg_log_size;
+	u32 num_wds_entries;
+	u32 dma_burst_size;
+	u32 mac_aggr_delim;
+	u32 rx_skip_defrag_timeout_dup_detection_check;
+	u32 vow_config;
+	u32 gtk_offload_max_vdev;
+	u32 num_msdu_desc;
+	u32 max_frag_entries;
+	u32 num_tdls_vdevs;
+	u32 num_tdls_conn_table_entries;
+	u32 beacon_tx_offload_max_vdev;
+	u32 num_multicast_filter_entries;
+	u32 num_wow_filters;
+	u32 num_keep_alive_pattern;
+	u32 keep_alive_pattern_size;
+	u32 max_tdls_concurrent_sleep_sta;
+	u32 max_tdls_concurrent_buffer_sta;
+	u32 wmi_send_separate;
+	u32 num_ocb_vdevs;
+	u32 num_ocb_channels;
+	u32 num_ocb_schedules;
+	u32 flag1;
+	u32 smart_ant_cap;
+	u32 bk_minfree;
+	u32 be_minfree;
+	u32 vi_minfree;
+	u32 vo_minfree;
+	u32 alloc_frag_desc_for_data_pkt;
+	u32 num_ns_ext_tuples_cfg;
+	u32 bpf_instruction_size;
+	u32 max_bssid_rx_filters;
+	u32 use_pdev_id;
+	u32 max_num_dbs_scan_duty_cycle;
+	u32 max_num_group_keys;
+	u32 peer_map_unmap_v2_support;
+} __packed;
+
+struct wmi_service_ready_event {
+	u32 fw_build_vers;
+	struct wmi_abi_version fw_abi_vers;
+	u32 phy_capability;
+	u32 max_frag_entry;
+	u32 num_rf_chains;
+	u32 ht_cap_info;
+	u32 vht_cap_info;
+	u32 vht_supp_mcs;
+	u32 hw_min_tx_power;
+	u32 hw_max_tx_power;
+	u32 sys_cap_info;
+	u32 min_pkt_size_enable;
+	u32 max_bcn_ie_size;
+	u32 num_mem_reqs;
+	u32 max_num_scan_channels;
+	u32 hw_bd_id;
+	u32 hw_bd_info[HW_BD_INFO_SIZE];
+	u32 max_supported_macs;
+	u32 wmi_fw_sub_feat_caps;
+	u32 num_dbs_hw_modes;
+	/* txrx_chainmask
+	 *    [7:0]   - 2G band tx chain mask
+	 *    [15:8]  - 2G band rx chain mask
+	 *    [23:16] - 5G band tx chain mask
+	 *    [31:24] - 5G band rx chain mask
+	 */
+	u32 txrx_chainmask;
+	u32 default_dbs_hw_mode_index;
+	u32 num_msdu_desc;
+} __packed;
+
+#define WMI_SERVICE_BM_SIZE	((WMI_MAX_SERVICE + sizeof(u32) - 1) / sizeof(u32))
+
+#define WMI_SERVICE_SEGMENT_BM_SIZE32 4 /* 4x u32 = 128 bits */
+#define WMI_SERVICE_EXT_BM_SIZE (WMI_SERVICE_SEGMENT_BM_SIZE32 * sizeof(u32))
+#define WMI_AVAIL_SERVICE_BITS_IN_SIZE32 32
+#define WMI_SERVICE_BITS_IN_SIZE32 4
+
+struct wmi_service_ready_ext_event {
+	u32 default_conc_scan_config_bits;
+	u32 default_fw_config_bits;
+	struct wmi_ppe_threshold ppet;
+	u32 he_cap_info;
+	u32 mpdu_density;
+	u32 max_bssid_rx_filters;
+	u32 fw_build_vers_ext;
+	u32 max_nlo_ssids;
+	u32 max_bssid_indicator;
+	u32 he_cap_info_ext;
+} __packed;
+
+struct wmi_soc_mac_phy_hw_mode_caps {
+	u32 num_hw_modes;
+	u32 num_chainmask_tables;
+} __packed;
+
+struct wmi_hw_mode_capabilities {
+	u32 tlv_header;
+	u32 hw_mode_id;
+	u32 phy_id_map;
+	u32 hw_mode_config_type;
+} __packed;
+
+#define WMI_MAX_HECAP_PHY_SIZE                 (3)
+
+struct wmi_mac_phy_capabilities {
+	u32 tlv_header;
+	u32 hw_mode_id;
+	u32 pdev_id;
+	u32 phy_id;
+	u32 supported_flags;
+	u32 supported_bands;
+	u32 ampdu_density;
+	u32 max_bw_supported_2g;
+	u32 ht_cap_info_2g;
+	u32 vht_cap_info_2g;
+	u32 vht_supp_mcs_2g;
+	u32 he_cap_info_2g;
+	u32 he_supp_mcs_2g;
+	u32 tx_chain_mask_2g;
+	u32 rx_chain_mask_2g;
+	u32 max_bw_supported_5g;
+	u32 ht_cap_info_5g;
+	u32 vht_cap_info_5g;
+	u32 vht_supp_mcs_5g;
+	u32 he_cap_info_5g;
+	u32 he_supp_mcs_5g;
+	u32 tx_chain_mask_5g;
+	u32 rx_chain_mask_5g;
+	u32 he_cap_phy_info_2g[WMI_MAX_HECAP_PHY_SIZE];
+	u32 he_cap_phy_info_5g[WMI_MAX_HECAP_PHY_SIZE];
+	struct wmi_ppe_threshold he_ppet2g;
+	struct wmi_ppe_threshold he_ppet5g;
+	u32 chainmask_table_id;
+	u32 lmac_id;
+	u32 he_cap_info_2g_ext;
+	u32 he_cap_info_5g_ext;
+	u32 he_cap_info_internal;
+} __packed;
+
+struct wmi_hal_reg_capabilities_ext {
+	u32 tlv_header;
+	u32 phy_id;
+	u32 eeprom_reg_domain;
+	u32 eeprom_reg_domain_ext;
+	u32 regcap1;
+	u32 regcap2;
+	u32 wireless_modes;
+	u32 low_2ghz_chan;
+	u32 high_2ghz_chan;
+	u32 low_5ghz_chan;
+	u32 high_5ghz_chan;
+} __packed;
+
+struct wmi_soc_hal_reg_capabilities {
+	u32 num_phy;
+} __packed;
+
+/* 2 word representation of MAC addr */
+struct wmi_mac_addr {
+	union {
+		u8 addr[6];
+		struct {
+			u32 word0;
+			u32 word1;
+		} __packed;
+	} __packed;
+} __packed;
+
+struct wmi_ready_event {
+	struct wmi_abi_version fw_abi_vers;
+	struct wmi_mac_addr mac_addr;
+	u32 status;
+	u32 num_dscp_table;
+	u32 num_extra_mac_addr;
+	u32 num_total_peers;
+	u32 num_extra_peers;
+} __packed;
+
+struct wmi_service_available_event {
+	u32 wmi_service_segment_offset;
+	u32 wmi_service_segment_bitmap[WMI_SERVICE_SEGMENT_BM_SIZE32];
+} __packed;
+
+struct ath11k_pdev_wmi {
+	struct ath11k_wmi_base *wmi_sc;
+	enum ath11k_htc_ep_id eid;
+	const struct wmi_peer_flags_map *peer_flags;
+	u32 rx_decap_mode;
+};
+
+struct vdev_create_params {
+	u8 if_id;
+	u32 type;
+	u32 subtype;
+	struct {
+		u8 tx;
+		u8 rx;
+	} chains[NUM_NL80211_BANDS];
+	u32 pdev_id;
+};
+
+struct wmi_vdev_create_cmd {
+	u32 tlv_header;
+	u32 vdev_id;
+	u32 vdev_type;
+	u32 vdev_subtype;
+	struct wmi_mac_addr vdev_macaddr;
+	u32 num_cfg_txrx_streams;
+	u32 pdev_id;
+} __packed;
+
+struct wmi_vdev_txrx_streams {
+	u32 tlv_header;
+	u32 band;
+	u32 supported_tx_streams;
+	u32 supported_rx_streams;
+} __packed;
+
+struct wmi_vdev_delete_cmd {
+	u32 tlv_header;
+	u32 vdev_id;
+} __packed;
+
+struct wmi_vdev_up_cmd {
+	u32 tlv_header;
+	u32 vdev_id;
+	u32 vdev_assoc_id;
+	struct wmi_mac_addr vdev_bssid;
+	struct wmi_mac_addr trans_bssid;
+	u32 profile_idx;
+	u32 profile_num;
+} __packed;
+
+struct wmi_vdev_stop_cmd {
+	u32 tlv_header;
+	u32 vdev_id;
+} __packed;
+
+struct wmi_vdev_down_cmd {
+	u32 tlv_header;
+	u32 vdev_id;
+} __packed;
+
+#define WMI_VDEV_START_HIDDEN_SSID  BIT(0)
+#define WMI_VDEV_START_PMF_ENABLED  BIT(1)
+#define WMI_VDEV_START_LDPC_RX_ENABLED BIT(3)
+
+struct wmi_ssid {
+	u32 ssid_len;
+	u32 ssid[8];
+} __packed;
+
+#define ATH11K_VDEV_SETUP_TIMEOUT_HZ (1 * HZ)
+
+struct wmi_vdev_start_request_cmd {
+	u32 tlv_header;
+	u32 vdev_id;
+	u32 requestor_id;
+	u32 beacon_interval;
+	u32 dtim_period;
+	u32 flags;
+	struct wmi_ssid ssid;
+	u32 bcn_tx_rate;
+	u32 bcn_txpower;
+	u32 num_noa_descriptors;
+	u32 disable_hw_ack;
+	u32 preferred_tx_streams;
+	u32 preferred_rx_streams;
+	u32 he_ops;
+	u32 cac_duration_ms;
+	u32 regdomain;
+} __packed;
+
+#define MGMT_TX_DL_FRM_LEN		     64
+#define WMI_MAC_MAX_SSID_LENGTH              32
+struct mac_ssid {
+	u8 length;
+	u8 mac_ssid[WMI_MAC_MAX_SSID_LENGTH];
+} __packed;
+
+struct wmi_p2p_noa_descriptor {
+	u32 type_count;
+	u32 duration;
+	u32 interval;
+	u32 start_time;
+};
+
+struct channel_param {
+	u8 chan_id;
+	u8 pwr;
+	u32 mhz;
+	u32 half_rate:1,
+	    quarter_rate:1,
+	    dfs_set:1,
+	    dfs_set_cfreq2:1,
+	    is_chan_passive:1,
+	    allow_ht:1,
+	    allow_vht:1,
+	    set_agile:1;
+	u32 phy_mode;
+	u32 cfreq1;
+	u32 cfreq2;
+	char   maxpower;
+	char   minpower;
+	char   maxregpower;
+	u8  antennamax;
+	u8  reg_class_id;
+} __packed;
+
+enum wmi_phy_mode {
+	MODE_11A        = 0,
+	MODE_11G        = 1,   /* 11b/g Mode */
+	MODE_11B        = 2,   /* 11b Mode */
+	MODE_11GONLY    = 3,   /* 11g only Mode */
+	MODE_11NA_HT20   = 4,
+	MODE_11NG_HT20   = 5,
+	MODE_11NA_HT40   = 6,
+	MODE_11NG_HT40   = 7,
+	MODE_11AC_VHT20 = 8,
+	MODE_11AC_VHT40 = 9,
+	MODE_11AC_VHT80 = 10,
+	MODE_11AC_VHT20_2G = 11,
+	MODE_11AC_VHT40_2G = 12,
+	MODE_11AC_VHT80_2G = 13,
+	MODE_11AC_VHT80_80 = 14,
+	MODE_11AC_VHT160 = 15,
+	MODE_11AX_HE20 = 16,
+	MODE_11AX_HE40 = 17,
+	MODE_11AX_HE80 = 18,
+	MODE_11AX_HE80_80 = 19,
+	MODE_11AX_HE160 = 20,
+	MODE_11AX_HE20_2G = 21,
+	MODE_11AX_HE40_2G = 22,
+	MODE_11AX_HE80_2G = 23,
+	MODE_UNKNOWN = 24,
+	MODE_MAX = 24
+};
+
+static inline const char *ath11k_wmi_phymode_str(enum wmi_phy_mode mode)
+{
+	switch (mode) {
+	case MODE_11A:
+		return "11a";
+	case MODE_11G:
+		return "11g";
+	case MODE_11B:
+		return "11b";
+	case MODE_11GONLY:
+		return "11gonly";
+	case MODE_11NA_HT20:
+		return "11na-ht20";
+	case MODE_11NG_HT20:
+		return "11ng-ht20";
+	case MODE_11NA_HT40:
+		return "11na-ht40";
+	case MODE_11NG_HT40:
+		return "11ng-ht40";
+	case MODE_11AC_VHT20:
+		return "11ac-vht20";
+	case MODE_11AC_VHT40:
+		return "11ac-vht40";
+	case MODE_11AC_VHT80:
+		return "11ac-vht80";
+	case MODE_11AC_VHT160:
+		return "11ac-vht160";
+	case MODE_11AC_VHT80_80:
+		return "11ac-vht80+80";
+	case MODE_11AC_VHT20_2G:
+		return "11ac-vht20-2g";
+	case MODE_11AC_VHT40_2G:
+		return "11ac-vht40-2g";
+	case MODE_11AC_VHT80_2G:
+		return "11ac-vht80-2g";
+	case MODE_11AX_HE20:
+		return "11ax-he20";
+	case MODE_11AX_HE40:
+		return "11ax-he40";
+	case MODE_11AX_HE80:
+		return "11ax-he80";
+	case MODE_11AX_HE80_80:
+		return "11ax-he80+80";
+	case MODE_11AX_HE160:
+		return "11ax-he160";
+	case MODE_11AX_HE20_2G:
+		return "11ax-he20-2g";
+	case MODE_11AX_HE40_2G:
+		return "11ax-he40-2g";
+	case MODE_11AX_HE80_2G:
+		return "11ax-he80-2g";
+	case MODE_UNKNOWN:
+		/* skip */
+		break;
+
+		/* no default handler to allow compiler to check that the
+		 * enum is fully handled
+		 */
+	};
+
+	return "<unknown>";
+}
+
+struct wmi_channel_arg {
+	u32 freq;
+	u32 band_center_freq1;
+	u32 band_center_freq2;
+	bool passive;
+	bool allow_ibss;
+	bool allow_ht;
+	bool allow_vht;
+	bool ht40plus;
+	bool chan_radar;
+	bool freq2_radar;
+	bool allow_he;
+	u32 min_power;
+	u32 max_power;
+	u32 max_reg_power;
+	u32 max_antenna_gain;
+	enum wmi_phy_mode mode;
+};
+
+struct wmi_vdev_start_req_arg {
+	u32 vdev_id;
+	struct wmi_channel_arg channel;
+	u32 bcn_intval;
+	u32 dtim_period;
+	u8 *ssid;
+	u32 ssid_len;
+	u32 bcn_tx_rate;
+	u32 bcn_tx_power;
+	bool disable_hw_ack;
+	bool hidden_ssid;
+	bool pmf_enabled;
+	u32 he_ops;
+	u32 cac_duration_ms;
+	u32 regdomain;
+	u32 pref_rx_streams;
+	u32 pref_tx_streams;
+	u32 num_noa_descriptors;
+};
+
+struct peer_create_params {
+	const u8 *peer_addr;
+	u32 peer_type;
+	u32 vdev_id;
+};
+
+struct peer_delete_params {
+	u8 vdev_id;
+};
+
+struct peer_flush_params {
+	u32 peer_tid_bitmap;
+	u8 vdev_id;
+};
+
+struct pdev_set_regdomain_params {
+	u16 current_rd_in_use;
+	u16 current_rd_2g;
+	u16 current_rd_5g;
+	u32 ctl_2g;
+	u32 ctl_5g;
+	u8 dfs_domain;
+	u32 pdev_id;
+};
+
+struct rx_reorder_queue_remove_params {
+	u8 *peer_macaddr;
+	u16 vdev_id;
+	u32 peer_tid_bitmap;
+};
+
+#define WMI_HOST_PDEV_ID_SOC 0xFF
+#define WMI_HOST_PDEV_ID_0   0
+#define WMI_HOST_PDEV_ID_1   1
+#define WMI_HOST_PDEV_ID_2   2
+
+#define WMI_PDEV_ID_SOC         0
+#define WMI_PDEV_ID_1ST         1
+#define WMI_PDEV_ID_2ND         2
+#define WMI_PDEV_ID_3RD         3
+
+/* Freq units in MHz */
+#define REG_RULE_START_FREQ			0x0000ffff
+#define REG_RULE_END_FREQ			0xffff0000
+#define REG_RULE_FLAGS				0x0000ffff
+#define REG_RULE_MAX_BW				0x0000ffff
+#define REG_RULE_REG_PWR			0x00ff0000
+#define REG_RULE_ANT_GAIN			0xff000000
+
+#define WMI_VDEV_PARAM_TXBF_SU_TX_BFEE BIT(0)
+#define WMI_VDEV_PARAM_TXBF_MU_TX_BFEE BIT(1)
+#define WMI_VDEV_PARAM_TXBF_SU_TX_BFER BIT(2)
+#define WMI_VDEV_PARAM_TXBF_MU_TX_BFER BIT(3)
+
+#define HECAP_PHYDWORD_0	0
+#define HECAP_PHYDWORD_1	1
+#define HECAP_PHYDWORD_2	2
+
+#define HECAP_PHY_SU_BFER		BIT(31)
+#define HECAP_PHY_SU_BFEE		BIT(0)
+#define HECAP_PHY_MU_BFER		BIT(1)
+#define HECAP_PHY_UL_MUMIMO		BIT(22)
+#define HECAP_PHY_UL_MUOFDMA		BIT(23)
+
+#define HECAP_PHY_SUBFMR_GET(hecap_phy) \
+	FIELD_GET(HECAP_PHY_SU_BFER, hecap_phy[HECAP_PHYDWORD_0])
+
+#define HECAP_PHY_SUBFME_GET(hecap_phy) \
+	FIELD_GET(HECAP_PHY_SU_BFEE, hecap_phy[HECAP_PHYDWORD_1])
+
+#define HECAP_PHY_MUBFMR_GET(hecap_phy) \
+	FIELD_GET(HECAP_PHY_MU_BFER, hecap_phy[HECAP_PHYDWORD_1])
+
+#define HECAP_PHY_ULMUMIMO_GET(hecap_phy) \
+	FIELD_GET(HECAP_PHY_UL_MUMIMO, hecap_phy[HECAP_PHYDWORD_0])
+
+#define HECAP_PHY_ULOFDMA_GET(hecap_phy) \
+	FIELD_GET(HECAP_PHY_UL_MUOFDMA, hecap_phy[HECAP_PHYDWORD_0])
+
+#define HE_MODE_SU_TX_BFEE	BIT(0)
+#define HE_MODE_SU_TX_BFER	BIT(1)
+#define HE_MODE_MU_TX_BFEE	BIT(2)
+#define HE_MODE_MU_TX_BFER	BIT(3)
+#define HE_MODE_DL_OFDMA	BIT(4)
+#define HE_MODE_UL_OFDMA	BIT(5)
+#define HE_MODE_UL_MUMIMO	BIT(6)
+
+#define HE_DL_MUOFDMA_ENABLE	1
+#define HE_UL_MUOFDMA_ENABLE	1
+#define HE_DL_MUMIMO_ENABLE	1
+#define HE_MU_BFEE_ENABLE	1
+#define HE_SU_BFEE_ENABLE	1
+
+#define HE_VHT_SOUNDING_MODE_ENABLE		1
+#define HE_SU_MU_SOUNDING_MODE_ENABLE		1
+#define HE_TRIG_NONTRIG_SOUNDING_MODE_ENABLE	1
+
+/* HE or VHT Sounding */
+#define HE_VHT_SOUNDING_MODE		BIT(0)
+/* SU or MU Sounding */
+#define HE_SU_MU_SOUNDING_MODE		BIT(2)
+/* Trig or Non-Trig Sounding */
+#define HE_TRIG_NONTRIG_SOUNDING_MODE	BIT(3)
+
+#define WMI_TXBF_STS_CAP_OFFSET_LSB	4
+#define WMI_TXBF_STS_CAP_OFFSET_MASK	0x70
+#define WMI_BF_SOUND_DIM_OFFSET_LSB	8
+#define WMI_BF_SOUND_DIM_OFFSET_MASK	0x700
+
+struct pdev_params {
+	u32 param_id;
+	u32 param_value;
+};
+
+enum wmi_peer_type {
+	WMI_PEER_TYPE_DEFAULT = 0,
+	WMI_PEER_TYPE_BSS = 1,
+	WMI_PEER_TYPE_TDLS = 2,
+};
+
+struct wmi_peer_create_cmd {
+	u32 tlv_header;
+	u32 vdev_id;
+	struct wmi_mac_addr peer_macaddr;
+	u32 peer_type;
+} __packed;
+
+struct wmi_peer_delete_cmd {
+	u32 tlv_header;
+	u32 vdev_id;
+	struct wmi_mac_addr peer_macaddr;
+} __packed;
+
+struct wmi_peer_reorder_queue_setup_cmd {
+	u32 tlv_header;
+	u32 vdev_id;
+	struct wmi_mac_addr peer_macaddr;
+	u32 tid;
+	u32 queue_ptr_lo;
+	u32 queue_ptr_hi;
+	u32 queue_no;
+	u32 ba_window_size_valid;
+	u32 ba_window_size;
+} __packed;
+
+struct wmi_peer_reorder_queue_remove_cmd {
+	u32 tlv_header;
+	u32 vdev_id;
+	struct wmi_mac_addr peer_macaddr;
+	u32 tid_mask;
+} __packed;
+
+struct gpio_config_params {
+	u32 gpio_num;
+	u32 input;
+	u32 pull_type;
+	u32 intr_mode;
+};
+
+enum wmi_gpio_type {
+	WMI_GPIO_PULL_NONE,
+	WMI_GPIO_PULL_UP,
+	WMI_GPIO_PULL_DOWN
+};
+
+enum wmi_gpio_intr_type {
+	WMI_GPIO_INTTYPE_DISABLE,
+	WMI_GPIO_INTTYPE_RISING_EDGE,
+	WMI_GPIO_INTTYPE_FALLING_EDGE,
+	WMI_GPIO_INTTYPE_BOTH_EDGE,
+	WMI_GPIO_INTTYPE_LEVEL_LOW,
+	WMI_GPIO_INTTYPE_LEVEL_HIGH
+};
+
+enum wmi_bss_chan_info_req_type {
+	WMI_BSS_SURVEY_REQ_TYPE_READ = 1,
+	WMI_BSS_SURVEY_REQ_TYPE_READ_CLEAR,
+};
+
+struct wmi_gpio_config_cmd_param {
+	u32 tlv_header;
+	u32 gpio_num;
+	u32 input;
+	u32 pull_type;
+	u32 intr_mode;
+};
+
+struct gpio_output_params {
+	u32 gpio_num;
+	u32 set;
+};
+
+struct wmi_gpio_output_cmd_param {
+	u32 tlv_header;
+	u32 gpio_num;
+	u32 set;
+};
+
+struct set_fwtest_params {
+	u32 arg;
+	u32 value;
+};
+
+struct wmi_fwtest_set_param_cmd_param {
+	u32 tlv_header;
+	u32 param_id;
+	u32 param_value;
+};
+
+struct wmi_pdev_set_param_cmd {
+	u32 tlv_header;
+	u32 pdev_id;
+	u32 param_id;
+	u32 param_value;
+} __packed;
+
+struct wmi_pdev_suspend_cmd {
+	u32 tlv_header;
+	u32 pdev_id;
+	u32 suspend_opt;
+} __packed;
+
+struct wmi_pdev_resume_cmd {
+	u32 tlv_header;
+	u32 pdev_id;
+} __packed;
+
+struct wmi_pdev_bss_chan_info_req_cmd {
+	u32 tlv_header;
+	/* ref wmi_bss_chan_info_req_type */
+	u32 req_type;
+} __packed;
+
+struct wmi_ap_ps_peer_cmd {
+	u32 tlv_header;
+	u32 vdev_id;
+	struct wmi_mac_addr peer_macaddr;
+	u32 param;
+	u32 value;
+} __packed;
+
+struct wmi_sta_powersave_param_cmd {
+	u32 tlv_header;
+	u32 vdev_id;
+	u32 param;
+	u32 value;
+} __packed;
+
+struct wmi_pdev_set_regdomain_cmd {
+	u32 tlv_header;
+	u32 pdev_id;
+	u32 reg_domain;
+	u32 reg_domain_2g;
+	u32 reg_domain_5g;
+	u32 conformance_test_limit_2g;
+	u32 conformance_test_limit_5g;
+	u32 dfs_domain;
+} __packed;
+
+struct wmi_peer_set_param_cmd {
+	u32 tlv_header;
+	u32 vdev_id;
+	struct wmi_mac_addr peer_macaddr;
+	u32 param_id;
+	u32 param_value;
+} __packed;
+
+struct wmi_peer_flush_tids_cmd {
+	u32 tlv_header;
+	u32 vdev_id;
+	struct wmi_mac_addr peer_macaddr;
+	u32 peer_tid_bitmap;
+} __packed;
+
+struct wmi_dfs_phyerr_offload_cmd {
+	u32 tlv_header;
+	u32 pdev_id;
+} __packed;
+
+struct wmi_bcn_offload_ctrl_cmd {
+	u32 tlv_header;
+	u32 vdev_id;
+	u32 bcn_ctrl_op;
+} __packed;
+
+enum scan_priority {
+	SCAN_PRIORITY_VERY_LOW,
+	SCAN_PRIORITY_LOW,
+	SCAN_PRIORITY_MEDIUM,
+	SCAN_PRIORITY_HIGH,
+	SCAN_PRIORITY_VERY_HIGH,
+	SCAN_PRIORITY_COUNT,
+};
+
+enum scan_dwelltime_adaptive_mode {
+	SCAN_DWELL_MODE_DEFAULT = 0,
+	SCAN_DWELL_MODE_CONSERVATIVE = 1,
+	SCAN_DWELL_MODE_MODERATE = 2,
+	SCAN_DWELL_MODE_AGGRESSIVE = 3,
+	SCAN_DWELL_MODE_STATIC = 4
+};
+
+#define WLAN_SCAN_MAX_NUM_SSID          10
+#define WLAN_SCAN_MAX_NUM_BSSID         10
+#define WLAN_SCAN_MAX_NUM_CHANNELS      40
+
+#define WLAN_SSID_MAX_LEN 32
+
+struct element_info {
+	u32 len;
+	u8 *ptr;
+};
+
+struct wlan_ssid {
+	u8 length;
+	u8 ssid[WLAN_SSID_MAX_LEN];
+};
+
+#define WMI_IE_BITMAP_SIZE             8
+
+#define WMI_SCAN_MAX_NUM_SSID                0x0A
+/* prefix used by scan requestor ids on the host */
+#define WMI_HOST_SCAN_REQUESTOR_ID_PREFIX 0xA000
+
+/* prefix used by scan request ids generated on the host */
+/* host cycles through the lower 12 bits to generate ids */
+#define WMI_HOST_SCAN_REQ_ID_PREFIX 0xA000
+
+#define WLAN_SCAN_PARAMS_MAX_SSID    16
+#define WLAN_SCAN_PARAMS_MAX_BSSID   4
+#define WLAN_SCAN_PARAMS_MAX_IE_LEN  256
+
+/* Values lower than this may be refused by some firmware revisions with a scan
+ * completion with a timedout reason.
+ */
+#define WMI_SCAN_CHAN_MIN_TIME_MSEC 40
+
+/* Scan priority numbers must be sequential, starting with 0 */
+enum wmi_scan_priority {
+	WMI_SCAN_PRIORITY_VERY_LOW = 0,
+	WMI_SCAN_PRIORITY_LOW,
+	WMI_SCAN_PRIORITY_MEDIUM,
+	WMI_SCAN_PRIORITY_HIGH,
+	WMI_SCAN_PRIORITY_VERY_HIGH,
+	WMI_SCAN_PRIORITY_COUNT   /* number of priorities supported */
+};
+
+enum wmi_scan_event_type {
+	WMI_SCAN_EVENT_STARTED              = BIT(0),
+	WMI_SCAN_EVENT_COMPLETED            = BIT(1),
+	WMI_SCAN_EVENT_BSS_CHANNEL          = BIT(2),
+	WMI_SCAN_EVENT_FOREIGN_CHAN         = BIT(3),
+	WMI_SCAN_EVENT_DEQUEUED             = BIT(4),
+	/* possibly by high-prio scan */
+	WMI_SCAN_EVENT_PREEMPTED            = BIT(5),
+	WMI_SCAN_EVENT_START_FAILED         = BIT(6),
+	WMI_SCAN_EVENT_RESTARTED            = BIT(7),
+	WMI_SCAN_EVENT_FOREIGN_CHAN_EXIT    = BIT(8),
+	WMI_SCAN_EVENT_SUSPENDED            = BIT(9),
+	WMI_SCAN_EVENT_RESUMED              = BIT(10),
+	WMI_SCAN_EVENT_MAX                  = BIT(15),
+};
+
+enum wmi_scan_completion_reason {
+	WMI_SCAN_REASON_COMPLETED,
+	WMI_SCAN_REASON_CANCELLED,
+	WMI_SCAN_REASON_PREEMPTED,
+	WMI_SCAN_REASON_TIMEDOUT,
+	WMI_SCAN_REASON_INTERNAL_FAILURE,
+	WMI_SCAN_REASON_MAX,
+};
+
+struct  wmi_start_scan_cmd {
+	u32 tlv_header;
+	u32 scan_id;
+	u32 scan_req_id;
+	u32 vdev_id;
+	u32 scan_priority;
+	u32 notify_scan_events;
+	u32 dwell_time_active;
+	u32 dwell_time_passive;
+	u32 min_rest_time;
+	u32 max_rest_time;
+	u32 repeat_probe_time;
+	u32 probe_spacing_time;
+	u32 idle_time;
+	u32 max_scan_time;
+	u32 probe_delay;
+	u32 scan_ctrl_flags;
+	u32 burst_duration;
+	u32 num_chan;
+	u32 num_bssid;
+	u32 num_ssids;
+	u32 ie_len;
+	u32 n_probes;
+	struct wmi_mac_addr mac_addr;
+	struct wmi_mac_addr mac_mask;
+	u32 ie_bitmap[WMI_IE_BITMAP_SIZE];
+	u32 num_vendor_oui;
+	u32 scan_ctrl_flags_ext;
+	u32 dwell_time_active_2g;
+} __packed;
+
+#define WMI_SCAN_FLAG_PASSIVE        0x1
+#define WMI_SCAN_ADD_BCAST_PROBE_REQ 0x2
+#define WMI_SCAN_ADD_CCK_RATES       0x4
+#define WMI_SCAN_ADD_OFDM_RATES      0x8
+#define WMI_SCAN_CHAN_STAT_EVENT     0x10
+#define WMI_SCAN_FILTER_PROBE_REQ    0x20
+#define WMI_SCAN_BYPASS_DFS_CHN      0x40
+#define WMI_SCAN_CONTINUE_ON_ERROR   0x80
+#define WMI_SCAN_FILTER_PROMISCUOS   0x100
+#define WMI_SCAN_FLAG_FORCE_ACTIVE_ON_DFS 0x200
+#define WMI_SCAN_ADD_TPC_IE_IN_PROBE_REQ  0x400
+#define WMI_SCAN_ADD_DS_IE_IN_PROBE_REQ   0x800
+#define WMI_SCAN_ADD_SPOOF_MAC_IN_PROBE_REQ   0x1000
+#define WMI_SCAN_OFFCHAN_MGMT_TX    0x2000
+#define WMI_SCAN_OFFCHAN_DATA_TX    0x4000
+#define WMI_SCAN_CAPTURE_PHY_ERROR  0x8000
+#define WMI_SCAN_FLAG_STRICT_PASSIVE_ON_PCHN 0x10000
+#define WMI_SCAN_FLAG_HALF_RATE_SUPPORT      0x20000
+#define WMI_SCAN_FLAG_QUARTER_RATE_SUPPORT   0x40000
+#define WMI_SCAN_RANDOM_SEQ_NO_IN_PROBE_REQ 0x80000
+#define WMI_SCAN_ENABLE_IE_WHTELIST_IN_PROBE_REQ 0x100000
+
+#define WMI_SCAN_DWELL_MODE_MASK 0x00E00000
+#define WMI_SCAN_DWELL_MODE_SHIFT        21
+
+enum {
+	WMI_SCAN_DWELL_MODE_DEFAULT      = 0,
+	WMI_SCAN_DWELL_MODE_CONSERVATIVE = 1,
+	WMI_SCAN_DWELL_MODE_MODERATE     = 2,
+	WMI_SCAN_DWELL_MODE_AGGRESSIVE   = 3,
+	WMI_SCAN_DWELL_MODE_STATIC       = 4,
+};
+
+#define WMI_SCAN_SET_DWELL_MODE(flag, mode) \
+	((flag) |= (((mode) << WMI_SCAN_DWELL_MODE_SHIFT) & \
+		    WMI_SCAN_DWELL_MODE_MASK))
+
+struct scan_req_params {
+	u32 scan_id;
+	u32 scan_req_id;
+	u32 vdev_id;
+	u32 pdev_id;
+	enum scan_priority scan_priority;
+	union {
+		struct {
+			u32 scan_ev_started:1,
+			    scan_ev_completed:1,
+			    scan_ev_bss_chan:1,
+			    scan_ev_foreign_chan:1,
+			    scan_ev_dequeued:1,
+			    scan_ev_preempted:1,
+			    scan_ev_start_failed:1,
+			    scan_ev_restarted:1,
+			    scan_ev_foreign_chn_exit:1,
+			    scan_ev_invalid:1,
+			    scan_ev_gpio_timeout:1,
+			    scan_ev_suspended:1,
+			    scan_ev_resumed:1;
+		};
+		u32 scan_events;
+	};
+	u32 dwell_time_active;
+	u32 dwell_time_active_2g;
+	u32 dwell_time_passive;
+	u32 min_rest_time;
+	u32 max_rest_time;
+	u32 repeat_probe_time;
+	u32 probe_spacing_time;
+	u32 idle_time;
+	u32 max_scan_time;
+	u32 probe_delay;
+	union {
+		struct {
+			u32 scan_f_passive:1,
+			    scan_f_bcast_probe:1,
+			    scan_f_cck_rates:1,
+			    scan_f_ofdm_rates:1,
+			    scan_f_chan_stat_evnt:1,
+			    scan_f_filter_prb_req:1,
+			    scan_f_bypass_dfs_chn:1,
+			    scan_f_continue_on_err:1,
+			    scan_f_offchan_mgmt_tx:1,
+			    scan_f_offchan_data_tx:1,
+			    scan_f_promisc_mode:1,
+			    scan_f_capture_phy_err:1,
+			    scan_f_strict_passive_pch:1,
+			    scan_f_half_rate:1,
+			    scan_f_quarter_rate:1,
+			    scan_f_force_active_dfs_chn:1,
+			    scan_f_add_tpc_ie_in_probe:1,
+			    scan_f_add_ds_ie_in_probe:1,
+			    scan_f_add_spoofed_mac_in_probe:1,
+			    scan_f_add_rand_seq_in_probe:1,
+			    scan_f_en_ie_whitelist_in_probe:1,
+			    scan_f_forced:1,
+			    scan_f_2ghz:1,
+			    scan_f_5ghz:1,
+			    scan_f_80mhz:1;
+		};
+		u32 scan_flags;
+	};
+	enum scan_dwelltime_adaptive_mode adaptive_dwell_time_mode;
+	u32 burst_duration;
+	u32 num_chan;
+	u32 num_bssid;
+	u32 num_ssids;
+	u32 n_probes;
+	u32 chan_list[WLAN_SCAN_MAX_NUM_CHANNELS];
+	u32 notify_scan_events;
+	struct wlan_ssid ssid[WLAN_SCAN_MAX_NUM_SSID];
+	struct wmi_mac_addr bssid_list[WLAN_SCAN_MAX_NUM_BSSID];
+	struct element_info extraie;
+	struct element_info htcap;
+	struct element_info vhtcap;
+};
+
+struct wmi_ssid_arg {
+	int len;
+	const u8 *ssid;
+};
+
+struct wmi_bssid_arg {
+	const u8 *bssid;
+};
+
+struct wmi_start_scan_arg {
+	u32 scan_id;
+	u32 scan_req_id;
+	u32 vdev_id;
+	u32 scan_priority;
+	u32 notify_scan_events;
+	u32 dwell_time_active;
+	u32 dwell_time_passive;
+	u32 min_rest_time;
+	u32 max_rest_time;
+	u32 repeat_probe_time;
+	u32 probe_spacing_time;
+	u32 idle_time;
+	u32 max_scan_time;
+	u32 probe_delay;
+	u32 scan_ctrl_flags;
+
+	u32 ie_len;
+	u32 n_channels;
+	u32 n_ssids;
+	u32 n_bssids;
+
+	u8 ie[WLAN_SCAN_PARAMS_MAX_IE_LEN];
+	u32 channels[64];
+	struct wmi_ssid_arg ssids[WLAN_SCAN_PARAMS_MAX_SSID];
+	struct wmi_bssid_arg bssids[WLAN_SCAN_PARAMS_MAX_BSSID];
+};
+
+#define WMI_SCAN_STOP_ONE       0x00000000
+#define WMI_SCN_STOP_VAP_ALL    0x01000000
+#define WMI_SCAN_STOP_ALL       0x04000000
+
+/* Prefix 0xA000 indicates that the scan request
+ * is trigger by HOST
+ */
+#define ATH11K_SCAN_ID          0xA000
+
+enum scan_cancel_req_type {
+	WLAN_SCAN_CANCEL_SINGLE = 1,
+	WLAN_SCAN_CANCEL_VDEV_ALL,
+	WLAN_SCAN_CANCEL_PDEV_ALL,
+};
+
+struct scan_cancel_param {
+	u32 requester;
+	u32 scan_id;
+	enum scan_cancel_req_type req_type;
+	u32 vdev_id;
+	u32 pdev_id;
+};
+
+struct  wmi_bcn_send_from_host_cmd {
+	u32 tlv_header;
+	u32 vdev_id;
+	u32 data_len;
+	union {
+		u32 frag_ptr;
+		u32 frag_ptr_lo;
+	};
+	u32 frame_ctrl;
+	u32 dtim_flag;
+	u32 bcn_antenna;
+	u32 frag_ptr_hi;
+};
+
+#define WMI_CHAN_INFO_MODE		GENMASK(5, 0)
+#define WMI_CHAN_INFO_HT40_PLUS		BIT(6)
+#define WMI_CHAN_INFO_PASSIVE		BIT(7)
+#define WMI_CHAN_INFO_ADHOC_ALLOWED	BIT(8)
+#define WMI_CHAN_INFO_AP_DISABLED	BIT(9)
+#define WMI_CHAN_INFO_DFS		BIT(10)
+#define WMI_CHAN_INFO_ALLOW_HT		BIT(11)
+#define WMI_CHAN_INFO_ALLOW_VHT		BIT(12)
+#define WMI_CHAN_INFO_CHAN_CHANGE_CAUSE_CSA	BIT(13)
+#define WMI_CHAN_INFO_HALF_RATE		BIT(14)
+#define WMI_CHAN_INFO_QUARTER_RATE	BIT(15)
+#define WMI_CHAN_INFO_DFS_FREQ2		BIT(16)
+#define WMI_CHAN_INFO_ALLOW_HE		BIT(17)
+
+#define WMI_CHAN_REG_INFO1_MIN_PWR	GENMASK(7, 0)
+#define WMI_CHAN_REG_INFO1_MAX_PWR	GENMASK(15, 8)
+#define WMI_CHAN_REG_INFO1_MAX_REG_PWR	GENMASK(23, 16)
+#define WMI_CHAN_REG_INFO1_REG_CLS	GENMASK(31, 24)
+
+#define WMI_CHAN_REG_INFO2_ANT_MAX	GENMASK(7, 0)
+#define WMI_CHAN_REG_INFO2_MAX_TX_PWR	GENMASK(15, 8)
+
+struct wmi_channel {
+	u32 tlv_header;
+	u32 mhz;
+	u32 band_center_freq1;
+	u32 band_center_freq2;
+	u32 info;
+	u32 reg_info_1;
+	u32 reg_info_2;
+} __packed;
+
+struct wmi_mgmt_params {
+	void *tx_frame;
+	u16 frm_len;
+	u8 vdev_id;
+	u16 chanfreq;
+	void *pdata;
+	u16 desc_id;
+	u8 *macaddr;
+	void *qdf_ctx;
+};
+
+enum wmi_sta_ps_mode {
+	WMI_STA_PS_MODE_DISABLED = 0,
+	WMI_STA_PS_MODE_ENABLED = 1,
+};
+
+#define WMI_SMPS_MASK_LOWER_16BITS 0xFF
+#define WMI_SMPS_MASK_UPPER_3BITS 0x7
+#define WMI_SMPS_PARAM_VALUE_SHIFT 29
+
+#define ATH11K_WMI_FW_HANG_ASSERT_TYPE 1
+#define ATH11K_WMI_FW_HANG_DELAY 0
+
+/* type, 0:unused 1: ASSERT 2: not respond detect command
+ * delay_time_ms, the simulate will delay time
+ */
+
+struct wmi_force_fw_hang_cmd {
+	u32 tlv_header;
+	u32 type;
+	u32 delay_time_ms;
+};
+
+struct wmi_vdev_set_param_cmd {
+	u32 tlv_header;
+	u32 vdev_id;
+	u32 param_id;
+	u32 param_value;
+} __packed;
+
+enum wmi_stats_id {
+	WMI_REQUEST_PEER_STAT			= BIT(0),
+	WMI_REQUEST_AP_STAT			= BIT(1),
+	WMI_REQUEST_PDEV_STAT			= BIT(2),
+	WMI_REQUEST_VDEV_STAT			= BIT(3),
+	WMI_REQUEST_BCNFLT_STAT			= BIT(4),
+	WMI_REQUEST_VDEV_RATE_STAT		= BIT(5),
+	WMI_REQUEST_INST_STAT			= BIT(6),
+	WMI_REQUEST_MIB_STAT			= BIT(7),
+	WMI_REQUEST_RSSI_PER_CHAIN_STAT		= BIT(8),
+	WMI_REQUEST_CONGESTION_STAT		= BIT(9),
+	WMI_REQUEST_PEER_EXTD_STAT		= BIT(10),
+	WMI_REQUEST_BCN_STAT			= BIT(11),
+	WMI_REQUEST_BCN_STAT_RESET		= BIT(12),
+	WMI_REQUEST_PEER_EXTD2_STAT		= BIT(13),
+};
+
+struct wmi_request_stats_cmd {
+	u32 tlv_header;
+	enum wmi_stats_id stats_id;
+	u32 vdev_id;
+	struct wmi_mac_addr peer_macaddr;
+	u32 pdev_id;
+} __packed;
+
+#define WMI_BEACON_TX_BUFFER_SIZE	512
+
+struct wmi_bcn_tmpl_cmd {
+	u32 tlv_header;
+	u32 vdev_id;
+	u32 tim_ie_offset;
+	u32 buf_len;
+	u32 csa_switch_count_offset;
+	u32 ext_csa_switch_count_offset;
+	u32 csa_event_bitmap;
+	u32 mbssid_ie_offset;
+	u32 esp_ie_offset;
+} __packed;
+
+struct wmi_key_seq_counter {
+	u32 key_seq_counter_l;
+	u32 key_seq_counter_h;
+} __packed;
+
+struct wmi_vdev_install_key_cmd {
+	u32 tlv_header;
+	u32 vdev_id;
+	struct wmi_mac_addr peer_macaddr;
+	u32 key_idx;
+	u32 key_flags;
+	u32 key_cipher;
+	struct wmi_key_seq_counter key_rsc_counter;
+	struct wmi_key_seq_counter key_global_rsc_counter;
+	struct wmi_key_seq_counter key_tsc_counter;
+	u8 wpi_key_rsc_counter[16];
+	u8 wpi_key_tsc_counter[16];
+	u32 key_len;
+	u32 key_txmic_len;
+	u32 key_rxmic_len;
+	u32 is_group_key_id_valid;
+	u32 group_key_id;
+
+	/* Followed by key_data containing key followed by
+	 * tx mic and then rx mic
+	 */
+} __packed;
+
+struct wmi_vdev_install_key_arg {
+	u32 vdev_id;
+	const u8 *macaddr;
+	u32 key_idx;
+	u32 key_flags;
+	u32 key_cipher;
+	u32 key_len;
+	u32 key_txmic_len;
+	u32 key_rxmic_len;
+	u64 key_rsc_counter;
+	const void *key_data;
+};
+
+#define WMI_MAX_SUPPORTED_RATES			128
+#define WMI_HOST_MAX_HECAP_PHY_SIZE		3
+#define WMI_HOST_MAX_HE_RATE_SET		1
+
+struct wmi_rate_set_arg {
+	u32 num_rates;
+	u8 rates[WMI_MAX_SUPPORTED_RATES];
+};
+
+struct peer_assoc_params {
+	struct wmi_mac_addr peer_macaddr;
+	u32 vdev_id;
+	u32 peer_new_assoc;
+	u32 peer_associd;
+	u32 peer_flags;
+	u32 peer_caps;
+	u32 peer_listen_intval;
+	u32 peer_ht_caps;
+	u32 peer_max_mpdu;
+	u32 peer_mpdu_density;
+	u32 peer_rate_caps;
+	u32 peer_nss;
+	u32 peer_vht_caps;
+	u32 peer_phymode;
+	u32 peer_ht_info[2];
+	struct wmi_rate_set_arg peer_legacy_rates;
+	struct wmi_rate_set_arg peer_ht_rates;
+	u32 rx_max_rate;
+	u32 rx_mcs_set;
+	u32 tx_max_rate;
+	u32 tx_mcs_set;
+	u8 vht_capable;
+	u32 tx_max_mcs_nss;
+	u32 peer_bw_rxnss_override;
+	bool is_pmf_enabled;
+	bool is_wme_set;
+	bool qos_flag;
+	bool apsd_flag;
+	bool ht_flag;
+	bool bw_40;
+	bool bw_80;
+	bool bw_160;
+	bool stbc_flag;
+	bool ldpc_flag;
+	bool static_mimops_flag;
+	bool dynamic_mimops_flag;
+	bool spatial_mux_flag;
+	bool vht_flag;
+	bool vht_ng_flag;
+	bool need_ptk_4_way;
+	bool need_gtk_2_way;
+	bool auth_flag;
+	bool safe_mode_enabled;
+	bool amsdu_disable;
+	/* Use common structure */
+	u8 peer_mac[ETH_ALEN];
+
+	bool he_flag;
+	u32 peer_he_cap_macinfo[2];
+	u32 peer_he_cap_macinfo_internal;
+	u32 peer_he_ops;
+	u32 peer_he_cap_phyinfo[WMI_HOST_MAX_HECAP_PHY_SIZE];
+	u32 peer_he_mcs_count;
+	u32 peer_he_rx_mcs_set[WMI_HOST_MAX_HE_RATE_SET];
+	u32 peer_he_tx_mcs_set[WMI_HOST_MAX_HE_RATE_SET];
+	struct ath11k_ppe_threshold peer_ppet;
+};
+
+struct  wmi_peer_assoc_complete_cmd {
+	u32 tlv_header;
+	struct wmi_mac_addr peer_macaddr;
+	u32 vdev_id;
+	u32 peer_new_assoc;
+	u32 peer_associd;
+	u32 peer_flags;
+	u32 peer_caps;
+	u32 peer_listen_intval;
+	u32 peer_ht_caps;
+	u32 peer_max_mpdu;
+	u32 peer_mpdu_density;
+	u32 peer_rate_caps;
+	u32 peer_nss;
+	u32 peer_vht_caps;
+	u32 peer_phymode;
+	u32 peer_ht_info[2];
+	u32 num_peer_legacy_rates;
+	u32 num_peer_ht_rates;
+	u32 peer_bw_rxnss_override;
+	struct  wmi_ppe_threshold peer_ppet;
+	u32 peer_he_cap_info;
+	u32 peer_he_ops;
+	u32 peer_he_cap_phy[WMI_MAX_HECAP_PHY_SIZE];
+	u32 peer_he_mcs;
+	u32 peer_he_cap_info_ext;
+	u32 peer_he_cap_info_internal;
+} __packed;
+
+struct wmi_stop_scan_cmd {
+	u32 tlv_header;
+	u32 requestor;
+	u32 scan_id;
+	u32 req_type;
+	u32 vdev_id;
+	u32 pdev_id;
+};
+
+struct scan_chan_list_params {
+	u32 pdev_id;
+	u16 nallchans;
+	struct channel_param ch_param[1];
+};
+
+struct wmi_scan_chan_list_cmd {
+	u32 tlv_header;
+	u32 num_scan_chans;
+	u32 flags;
+	u32 pdev_id;
+} __packed;
+
+#define WMI_MGMT_SEND_DOWNLD_LEN	64
+
+#define WMI_TX_PARAMS_DWORD0_POWER		GENMASK(7, 0)
+#define WMI_TX_PARAMS_DWORD0_MCS_MASK		GENMASK(19, 8)
+#define WMI_TX_PARAMS_DWORD0_NSS_MASK		GENMASK(27, 20)
+#define WMI_TX_PARAMS_DWORD0_RETRY_LIMIT	GENMASK(31, 28)
+
+#define WMI_TX_PARAMS_DWORD1_CHAIN_MASK		GENMASK(7, 0)
+#define WMI_TX_PARAMS_DWORD1_BW_MASK		GENMASK(14, 8)
+#define WMI_TX_PARAMS_DWORD1_PREAMBLE_TYPE	GENMASK(19, 15)
+#define WMI_TX_PARAMS_DWORD1_FRAME_TYPE		BIT(20)
+#define WMI_TX_PARAMS_DWORD1_RSVD		GENMASK(31, 21)
+
+struct wmi_mgmt_send_params {
+	u32 tlv_header;
+	u32 tx_params_dword0;
+	u32 tx_params_dword1;
+};
+
+struct wmi_mgmt_send_cmd {
+	u32 tlv_header;
+	u32 vdev_id;
+	u32 desc_id;
+	u32 chanfreq;
+	u32 paddr_lo;
+	u32 paddr_hi;
+	u32 frame_len;
+	u32 buf_len;
+	u32 tx_params_valid;
+
+	/* This TLV is followed by struct wmi_mgmt_frame */
+
+	/* Followed by struct wmi_mgmt_send_params */
+} __packed;
+
+struct wmi_sta_powersave_mode_cmd {
+	u32 tlv_header;
+	u32 vdev_id;
+	u32 sta_ps_mode;
+};
+
+struct wmi_sta_smps_force_mode_cmd {
+	u32 tlv_header;
+	u32 vdev_id;
+	u32 forced_mode;
+};
+
+struct wmi_sta_smps_param_cmd {
+	u32 tlv_header;
+	u32 vdev_id;
+	u32 param;
+	u32 value;
+};
+
+struct wmi_bcn_prb_info {
+	u32 tlv_header;
+	u32 caps;
+	u32 erp;
+} __packed;
+
+enum {
+	WMI_PDEV_SUSPEND,
+	WMI_PDEV_SUSPEND_AND_DISABLE_INTR,
+};
+
+struct green_ap_ps_params {
+	u32 value;
+};
+
+struct wmi_pdev_green_ap_ps_enable_cmd_param {
+	u32 tlv_header;
+	u32 pdev_id;
+	u32 enable;
+};
+
+struct ap_ps_params {
+	u32 vdev_id;
+	u32 param;
+	u32 value;
+};
+
+struct vdev_set_params {
+	u32 if_id;
+	u32 param_id;
+	u32 param_value;
+};
+
+struct stats_request_params {
+	u32 stats_id;
+	u32 vdev_id;
+	u32 pdev_id;
+};
+
+enum set_init_cc_type {
+	WMI_COUNTRY_INFO_TYPE_ALPHA,
+	WMI_COUNTRY_INFO_TYPE_COUNTRY_CODE,
+	WMI_COUNTRY_INFO_TYPE_REGDOMAIN,
+};
+
+enum set_init_cc_flags {
+	INVALID_CC,
+	CC_IS_SET,
+	REGDMN_IS_SET,
+	ALPHA_IS_SET,
+};
+
+struct wmi_init_country_params {
+	union {
+		u16 country_code;
+		u16 regdom_id;
+		u8 alpha2[3];
+	} cc_info;
+	enum set_init_cc_flags flags;
+};
+
+struct wmi_init_country_cmd {
+	u32 tlv_header;
+	u32 pdev_id;
+	u32 init_cc_type;
+	union {
+		u32 country_code;
+		u32 regdom_id;
+		u32 alpha2;
+	} cc_info;
+} __packed;
+
+struct wmi_pdev_pktlog_filter_info {
+	u32 tlv_header;
+	struct wmi_mac_addr peer_macaddr;
+} __packed;
+
+struct wmi_pdev_pktlog_filter_cmd {
+	u32 tlv_header;
+	u32 pdev_id;
+	u32 enable;
+	u32 filter_type;
+	u32 num_mac;
+} __packed;
+
+enum ath11k_wmi_pktlog_enable {
+	ATH11K_WMI_PKTLOG_ENABLE_AUTO  = 0,
+	ATH11K_WMI_PKTLOG_ENABLE_FORCE = 1,
+};
+
+struct wmi_pktlog_enable_cmd {
+	u32 tlv_header;
+	u32 pdev_id;
+	u32 evlist; /* WMI_PKTLOG_EVENT */
+	u32 enable;
+} __packed;
+
+struct wmi_pktlog_disable_cmd {
+	u32 tlv_header;
+	u32 pdev_id;
+} __packed;
+
+#define DFS_PHYERR_UNIT_TEST_CMD 0
+#define DFS_UNIT_TEST_MODULE	0x2b
+#define DFS_UNIT_TEST_TOKEN	0xAA
+
+enum dfs_test_args_idx {
+	DFS_TEST_CMDID = 0,
+	DFS_TEST_PDEV_ID,
+	DFS_TEST_RADAR_PARAM,
+	DFS_MAX_TEST_ARGS,
+};
+
+struct wmi_dfs_unit_test_arg {
+	u32 cmd_id;
+	u32 pdev_id;
+	u32 radar_param;
+};
+
+struct wmi_unit_test_cmd {
+	u32 tlv_header;
+	u32 vdev_id;
+	u32 module_id;
+	u32 num_args;
+	u32 diag_token;
+	/* Followed by test args*/
+} __packed;
+
+#define MAX_SUPPORTED_RATES 128
+
+#define WMI_PEER_AUTH		0x00000001
+#define WMI_PEER_QOS		0x00000002
+#define WMI_PEER_NEED_PTK_4_WAY	0x00000004
+#define WMI_PEER_NEED_GTK_2_WAY	0x00000010
+#define WMI_PEER_HE		0x00000400
+#define WMI_PEER_APSD		0x00000800
+#define WMI_PEER_HT		0x00001000
+#define WMI_PEER_40MHZ		0x00002000
+#define WMI_PEER_STBC		0x00008000
+#define WMI_PEER_LDPC		0x00010000
+#define WMI_PEER_DYN_MIMOPS	0x00020000
+#define WMI_PEER_STATIC_MIMOPS	0x00040000
+#define WMI_PEER_SPATIAL_MUX	0x00200000
+#define WMI_PEER_VHT		0x02000000
+#define WMI_PEER_80MHZ		0x04000000
+#define WMI_PEER_PMF		0x08000000
+/* TODO: Place holder for WLAN_PEER_F_PS_PRESEND_REQUIRED = 0x10000000.
+ * Need to be cleaned up
+ */
+#define WMI_PEER_IS_P2P_CAPABLE	0x20000000
+#define WMI_PEER_160MHZ		0x40000000
+#define WMI_PEER_SAFEMODE_EN	0x80000000
+
+struct beacon_tmpl_params {
+	u8 vdev_id;
+	u32 tim_ie_offset;
+	u32 tmpl_len;
+	u32 tmpl_len_aligned;
+	u32 csa_switch_count_offset;
+	u32 ext_csa_switch_count_offset;
+	u8 *frm;
+};
+
+struct wmi_rate_set {
+	u32 num_rates;
+	u32 rates[(MAX_SUPPORTED_RATES / 4) + 1];
+};
+
+struct wmi_vht_rate_set {
+	u32 tlv_header;
+	u32 rx_max_rate;
+	u32 rx_mcs_set;
+	u32 tx_max_rate;
+	u32 tx_mcs_set;
+	u32 tx_max_mcs_nss;
+} __packed;
+
+struct wmi_he_rate_set {
+	u32 tlv_header;
+	u32 rx_mcs_set;
+	u32 tx_mcs_set;
+} __packed;
+
+#define MAX_REG_RULES 10
+#define REG_ALPHA2_LEN 2
+
+enum wmi_start_event_param {
+	WMI_VDEV_START_RESP_EVENT = 0,
+	WMI_VDEV_RESTART_RESP_EVENT,
+};
+
+struct wmi_vdev_start_resp_event {
+	u32 vdev_id;
+	u32 requestor_id;
+	enum wmi_start_event_param resp_type;
+	u32 status;
+	u32 chain_mask;
+	u32 smps_mode;
+	union {
+		u32 mac_id;
+		u32 pdev_id;
+	};
+	u32 cfgd_tx_streams;
+	u32 cfgd_rx_streams;
+} __packed;
+
+/* VDEV start response status codes */
+enum wmi_vdev_start_resp_status_code {
+	WMI_VDEV_START_RESPONSE_STATUS_SUCCESS = 0,
+	WMI_VDEV_START_RESPONSE_INVALID_VDEVID = 1,
+	WMI_VDEV_START_RESPONSE_NOT_SUPPORTED = 2,
+	WMI_VDEV_START_RESPONSE_DFS_VIOLATION = 3,
+	WMI_VDEV_START_RESPONSE_INVALID_REGDOMAIN = 4,
+};
+
+;
+enum cc_setting_code {
+	REG_SET_CC_STATUS_PASS = 0,
+	REG_CURRENT_ALPHA2_NOT_FOUND = 1,
+	REG_INIT_ALPHA2_NOT_FOUND = 2,
+	REG_SET_CC_CHANGE_NOT_ALLOWED = 3,
+	REG_SET_CC_STATUS_NO_MEMORY = 4,
+	REG_SET_CC_STATUS_FAIL = 5,
+};
+
+/* Regaulatory Rule Flags Passed by FW */
+#define REGULATORY_CHAN_DISABLED     BIT(0)
+#define REGULATORY_CHAN_NO_IR        BIT(1)
+#define REGULATORY_CHAN_RADAR        BIT(3)
+#define REGULATORY_CHAN_NO_OFDM      BIT(6)
+#define REGULATORY_CHAN_INDOOR_ONLY  BIT(9)
+
+#define REGULATORY_CHAN_NO_HT40      BIT(4)
+#define REGULATORY_CHAN_NO_80MHZ     BIT(7)
+#define REGULATORY_CHAN_NO_160MHZ    BIT(8)
+#define REGULATORY_CHAN_NO_20MHZ     BIT(11)
+#define REGULATORY_CHAN_NO_10MHZ     BIT(12)
+
+enum {
+	WMI_REG_SET_CC_STATUS_PASS = 0,
+	WMI_REG_CURRENT_ALPHA2_NOT_FOUND = 1,
+	WMI_REG_INIT_ALPHA2_NOT_FOUND = 2,
+	WMI_REG_SET_CC_CHANGE_NOT_ALLOWED = 3,
+	WMI_REG_SET_CC_STATUS_NO_MEMORY = 4,
+	WMI_REG_SET_CC_STATUS_FAIL = 5,
+};
+
+struct cur_reg_rule {
+	u16 start_freq;
+	u16 end_freq;
+	u16 max_bw;
+	u8 reg_power;
+	u8 ant_gain;
+	u16 flags;
+};
+
+struct cur_regulatory_info {
+	enum cc_setting_code status_code;
+	u8 num_phy;
+	u8 phy_id;
+	u16 reg_dmn_pair;
+	u16 ctry_code;
+	u8 alpha2[REG_ALPHA2_LEN + 1];
+	u32 dfs_region;
+	u32 phybitmap;
+	u32 min_bw_2g;
+	u32 max_bw_2g;
+	u32 min_bw_5g;
+	u32 max_bw_5g;
+	u32 num_2g_reg_rules;
+	u32 num_5g_reg_rules;
+	struct cur_reg_rule *reg_rules_2g_ptr;
+	struct cur_reg_rule *reg_rules_5g_ptr;
+};
+
+struct wmi_reg_chan_list_cc_event {
+	u32 status_code;
+	u32 phy_id;
+	u32 alpha2;
+	u32 num_phy;
+	u32 country_id;
+	u32 domain_code;
+	u32 dfs_region;
+	u32 phybitmap;
+	u32 min_bw_2g;
+	u32 max_bw_2g;
+	u32 min_bw_5g;
+	u32 max_bw_5g;
+	u32 num_2g_reg_rules;
+	u32 num_5g_reg_rules;
+} __packed;
+
+struct wmi_regulatory_rule_struct {
+	u32  tlv_header;
+	u32  freq_info;
+	u32  bw_pwr_info;
+	u32  flag_info;
+};
+
+struct wmi_peer_delete_resp_event {
+	u32 vdev_id;
+	struct wmi_mac_addr peer_macaddr;
+} __packed;
+
+struct wmi_bcn_tx_status_event {
+	u32 vdev_id;
+	u32 tx_status;
+} __packed;
+
+struct wmi_vdev_stopped_event {
+	u32 vdev_id;
+} __packed;
+
+struct wmi_pdev_bss_chan_info_event {
+	u32 pdev_id;
+	u32 freq;	/* Units in MHz */
+	u32 noise_floor;	/* units are dBm */
+	/* rx clear - how often the channel was unused */
+	u32 rx_clear_count_low;
+	u32 rx_clear_count_high;
+	/* cycle count - elapsed time during measured period, in clock ticks */
+	u32 cycle_count_low;
+	u32 cycle_count_high;
+	/* tx cycle count - elapsed time spent in tx, in clock ticks */
+	u32 tx_cycle_count_low;
+	u32 tx_cycle_count_high;
+	/* rx cycle count - elapsed time spent in rx, in clock ticks */
+	u32 rx_cycle_count_low;
+	u32 rx_cycle_count_high;
+	/*rx_cycle cnt for my bss in 64bits format */
+	u32 rx_bss_cycle_count_low;
+	u32 rx_bss_cycle_count_high;
+} __packed;
+
+#define WMI_VDEV_INSTALL_KEY_COMPL_STATUS_SUCCESS 0
+
+struct wmi_vdev_install_key_compl_event {
+	u32 vdev_id;
+	struct wmi_mac_addr peer_macaddr;
+	u32 key_idx;
+	u32 key_flags;
+	u32 status;
+} __packed;
+
+struct wmi_vdev_install_key_complete_arg {
+	u32 vdev_id;
+	const u8 *macaddr;
+	u32 key_idx;
+	u32 key_flags;
+	u32 status;
+};
+
+struct wmi_peer_assoc_conf_event {
+	u32 vdev_id;
+	struct wmi_mac_addr peer_macaddr;
+} __packed;
+
+struct wmi_peer_assoc_conf_arg {
+	u32 vdev_id;
+	const u8 *macaddr;
+};
+
+/*
+ * PDEV statistics
+ */
+struct wmi_pdev_stats_base {
+	s32 chan_nf;
+	u32 tx_frame_count; /* Cycles spent transmitting frames */
+	u32 rx_frame_count; /* Cycles spent receiving frames */
+	u32 rx_clear_count; /* Total channel busy time, evidently */
+	u32 cycle_count; /* Total on-channel time */
+	u32 phy_err_count;
+	u32 chan_tx_pwr;
+} __packed;
+
+struct wmi_pdev_stats_extra {
+	u32 ack_rx_bad;
+	u32 rts_bad;
+	u32 rts_good;
+	u32 fcs_bad;
+	u32 no_beacons;
+	u32 mib_int_count;
+} __packed;
+
+struct wmi_pdev_stats_tx {
+	/* Num HTT cookies queued to dispatch list */
+	s32 comp_queued;
+
+	/* Num HTT cookies dispatched */
+	s32 comp_delivered;
+
+	/* Num MSDU queued to WAL */
+	s32 msdu_enqued;
+
+	/* Num MPDU queue to WAL */
+	s32 mpdu_enqued;
+
+	/* Num MSDUs dropped by WMM limit */
+	s32 wmm_drop;
+
+	/* Num Local frames queued */
+	s32 local_enqued;
+
+	/* Num Local frames done */
+	s32 local_freed;
+
+	/* Num queued to HW */
+	s32 hw_queued;
+
+	/* Num PPDU reaped from HW */
+	s32 hw_reaped;
+
+	/* Num underruns */
+	s32 underrun;
+
+	/* Num PPDUs cleaned up in TX abort */
+	s32 tx_abort;
+
+	/* Num MPDUs requed by SW */
+	s32 mpdus_requed;
+
+	/* excessive retries */
+	u32 tx_ko;
+
+	/* data hw rate code */
+	u32 data_rc;
+
+	/* Scheduler self triggers */
+	u32 self_triggers;
+
+	/* frames dropped due to excessive sw retries */
+	u32 sw_retry_failure;
+
+	/* illegal rate phy errors  */
+	u32 illgl_rate_phy_err;
+
+	/* wal pdev continuous xretry */
+	u32 pdev_cont_xretry;
+
+	/* wal pdev tx timeouts */
+	u32 pdev_tx_timeout;
+
+	/* wal pdev resets  */
+	u32 pdev_resets;
+
+	/* frames dropped due to non-availability of stateless TIDs */
+	u32 stateless_tid_alloc_failure;
+
+	/* PhY/BB underrun */
+	u32 phy_underrun;
+
+	/* MPDU is more than txop limit */
+	u32 txop_ovf;
+} __packed;
+
+struct wmi_pdev_stats_rx {
+	/* Cnts any change in ring routing mid-ppdu */
+	s32 mid_ppdu_route_change;
+
+	/* Total number of statuses processed */
+	s32 status_rcvd;
+
+	/* Extra frags on rings 0-3 */
+	s32 r0_frags;
+	s32 r1_frags;
+	s32 r2_frags;
+	s32 r3_frags;
+
+	/* MSDUs / MPDUs delivered to HTT */
+	s32 htt_msdus;
+	s32 htt_mpdus;
+
+	/* MSDUs / MPDUs delivered to local stack */
+	s32 loc_msdus;
+	s32 loc_mpdus;
+
+	/* AMSDUs that have more MSDUs than the status ring size */
+	s32 oversize_amsdu;
+
+	/* Number of PHY errors */
+	s32 phy_errs;
+
+	/* Number of PHY errors drops */
+	s32 phy_err_drop;
+
+	/* Number of mpdu errors - FCS, MIC, ENC etc. */
+	s32 mpdu_errs;
+} __packed;
+
+struct wmi_pdev_stats {
+	struct wmi_pdev_stats_base base;
+	struct wmi_pdev_stats_tx tx;
+	struct wmi_pdev_stats_rx rx;
+} __packed;
+
+#define WLAN_MAX_AC 4
+#define MAX_TX_RATE_VALUES 10
+#define MAX_TX_RATE_VALUES 10
+
+struct wmi_vdev_stats {
+	u32 vdev_id;
+	u32 beacon_snr;
+	u32 data_snr;
+	u32 num_tx_frames[WLAN_MAX_AC];
+	u32 num_rx_frames;
+	u32 num_tx_frames_retries[WLAN_MAX_AC];
+	u32 num_tx_frames_failures[WLAN_MAX_AC];
+	u32 num_rts_fail;
+	u32 num_rts_success;
+	u32 num_rx_err;
+	u32 num_rx_discard;
+	u32 num_tx_not_acked;
+	u32 tx_rate_history[MAX_TX_RATE_VALUES];
+	u32 beacon_rssi_history[MAX_TX_RATE_VALUES];
+} __packed;
+
+struct wmi_bcn_stats {
+	u32 vdev_id;
+	u32 tx_bcn_succ_cnt;
+	u32 tx_bcn_outage_cnt;
+} __packed;
+
+struct wmi_stats_event {
+	u32 stats_id;
+	u32 num_pdev_stats;
+	u32 num_vdev_stats;
+	u32 num_peer_stats;
+	u32 num_bcnflt_stats;
+	u32 num_chan_stats;
+	u32 num_mib_stats;
+	u32 pdev_id;
+	u32 num_bcn_stats;
+	u32 num_peer_extd_stats;
+	u32 num_peer_extd2_stats;
+} __packed;
+
+struct wmi_pdev_ctl_failsafe_chk_event {
+	u32 pdev_id;
+	u32 ctl_failsafe_status;
+} __packed;
+
+struct wmi_pdev_csa_switch_ev {
+	u32 pdev_id;
+	u32 current_switch_count;
+	u32 num_vdevs;
+} __packed;
+
+struct wmi_pdev_radar_ev {
+	u32 pdev_id;
+	u32 detection_mode;
+	u32 chan_freq;
+	u32 chan_width;
+	u32 detector_id;
+	u32 segment_id;
+	u32 timestamp;
+	u32 is_chirp;
+	s32 freq_offset;
+	s32 sidx;
+} __packed;
+
+#define WMI_RX_STATUS_OK			0x00
+#define WMI_RX_STATUS_ERR_CRC			0x01
+#define WMI_RX_STATUS_ERR_DECRYPT		0x08
+#define WMI_RX_STATUS_ERR_MIC			0x10
+#define WMI_RX_STATUS_ERR_KEY_CACHE_MISS	0x20
+
+#define WLAN_MGMT_TXRX_HOST_MAX_ANTENNA 4
+
+struct mgmt_rx_event_params {
+	u32 channel;
+	u32 snr;
+	u8 rssi_ctl[WLAN_MGMT_TXRX_HOST_MAX_ANTENNA];
+	u32 rate;
+	enum wmi_phy_mode phy_mode;
+	u32 buf_len;
+	int status;
+	u32 flags;
+	int rssi;
+	u32 tsf_delta;
+	u8 pdev_id;
+};
+
+#define ATH_MAX_ANTENNA 4
+
+struct wmi_mgmt_rx_hdr {
+	u32 channel;
+	u32 snr;
+	u32 rate;
+	u32 phy_mode;
+	u32 buf_len;
+	u32 status;
+	u32 rssi_ctl[ATH_MAX_ANTENNA];
+	u32 flags;
+	int rssi;
+	u32 tsf_delta;
+	u32 rx_tsf_l32;
+	u32 rx_tsf_u32;
+	u32 pdev_id;
+} __packed;
+
+#define MAX_ANTENNA_EIGHT 8
+
+struct wmi_rssi_ctl_ext {
+	u32 tlv_header;
+	u32 rssi_ctl_ext[MAX_ANTENNA_EIGHT - ATH_MAX_ANTENNA];
+};
+
+struct wmi_mgmt_tx_compl_event {
+	u32 desc_id;
+	u32 status;
+	u32 pdev_id;
+} __packed;
+
+struct wmi_scan_event {
+	u32 event_type; /* %WMI_SCAN_EVENT_ */
+	u32 reason; /* %WMI_SCAN_REASON_ */
+	u32 channel_freq; /* only valid for WMI_SCAN_EVENT_FOREIGN_CHANNEL */
+	u32 scan_req_id;
+	u32 scan_id;
+	u32 vdev_id;
+	/* TSF Timestamp when the scan event (%WMI_SCAN_EVENT_) is completed
+	 * In case of AP it is TSF of the AP vdev
+	 * In case of STA connected state, this is the TSF of the AP
+	 * In case of STA not connected, it will be the free running HW timer
+	 */
+	u32 tsf_timestamp;
+} __packed;
+
+struct wmi_peer_sta_kickout_arg {
+	const u8 *mac_addr;
+};
+
+struct wmi_peer_sta_kickout_event {
+	struct wmi_mac_addr peer_macaddr;
+} __packed;
+
+enum wmi_roam_reason {
+	WMI_ROAM_REASON_BETTER_AP = 1,
+	WMI_ROAM_REASON_BEACON_MISS = 2,
+	WMI_ROAM_REASON_LOW_RSSI = 3,
+	WMI_ROAM_REASON_SUITABLE_AP_FOUND = 4,
+	WMI_ROAM_REASON_HO_FAILED = 5,
+
+	/* keep last */
+	WMI_ROAM_REASON_MAX,
+};
+
+struct wmi_roam_event {
+	u32 vdev_id;
+	u32 reason;
+	u32 rssi;
+} __packed;
+
+#define WMI_CHAN_INFO_START_RESP 0
+#define WMI_CHAN_INFO_END_RESP 1
+
+struct wmi_chan_info_event {
+	u32 err_code;
+	u32 freq;
+	u32 cmd_flags;
+	u32 noise_floor;
+	u32 rx_clear_count;
+	u32 cycle_count;
+	u32 chan_tx_pwr_range;
+	u32 chan_tx_pwr_tp;
+	u32 rx_frame_count;
+	u32 my_bss_rx_cycle_count;
+	u32 rx_11b_mode_data_duration;
+	u32 tx_frame_cnt;
+	u32 mac_clk_mhz;
+	u32 vdev_id;
+} __packed;
+
+struct ath11k_targ_cap {
+	u32 phy_capability;
+	u32 max_frag_entry;
+	u32 num_rf_chains;
+	u32 ht_cap_info;
+	u32 vht_cap_info;
+	u32 vht_supp_mcs;
+	u32 hw_min_tx_power;
+	u32 hw_max_tx_power;
+	u32 sys_cap_info;
+	u32 min_pkt_size_enable;
+	u32 max_bcn_ie_size;
+	u32 max_num_scan_channels;
+	u32 max_supported_macs;
+	u32 wmi_fw_sub_feat_caps;
+	u32 txrx_chainmask;
+	u32 default_dbs_hw_mode_index;
+	u32 num_msdu_desc;
+};
+
+enum wmi_vdev_type {
+	WMI_VDEV_TYPE_AP      = 1,
+	WMI_VDEV_TYPE_STA     = 2,
+	WMI_VDEV_TYPE_IBSS    = 3,
+	WMI_VDEV_TYPE_MONITOR = 4,
+};
+
+enum wmi_vdev_subtype {
+	WMI_VDEV_SUBTYPE_NONE,
+	WMI_VDEV_SUBTYPE_P2P_DEVICE,
+	WMI_VDEV_SUBTYPE_P2P_CLIENT,
+	WMI_VDEV_SUBTYPE_P2P_GO,
+	WMI_VDEV_SUBTYPE_PROXY_STA,
+	WMI_VDEV_SUBTYPE_MESH_NON_11S,
+	WMI_VDEV_SUBTYPE_MESH_11S,
+};
+
+enum wmi_sta_powersave_param {
+	WMI_STA_PS_PARAM_RX_WAKE_POLICY = 0,
+	WMI_STA_PS_PARAM_TX_WAKE_THRESHOLD = 1,
+	WMI_STA_PS_PARAM_PSPOLL_COUNT = 2,
+	WMI_STA_PS_PARAM_INACTIVITY_TIME = 3,
+	WMI_STA_PS_PARAM_UAPSD = 4,
+};
+
+#define WMI_UAPSD_AC_TYPE_DELI 0
+#define WMI_UAPSD_AC_TYPE_TRIG 1
+
+#define WMI_UAPSD_AC_BIT_MASK(ac, type) \
+	((type ==  WMI_UAPSD_AC_TYPE_DELI) ? \
+	 (1 << (ac << 1)) : (1 << ((ac << 1) + 1)))
+
+enum wmi_sta_ps_param_uapsd {
+	WMI_STA_PS_UAPSD_AC0_DELIVERY_EN = (1 << 0),
+	WMI_STA_PS_UAPSD_AC0_TRIGGER_EN  = (1 << 1),
+	WMI_STA_PS_UAPSD_AC1_DELIVERY_EN = (1 << 2),
+	WMI_STA_PS_UAPSD_AC1_TRIGGER_EN  = (1 << 3),
+	WMI_STA_PS_UAPSD_AC2_DELIVERY_EN = (1 << 4),
+	WMI_STA_PS_UAPSD_AC2_TRIGGER_EN  = (1 << 5),
+	WMI_STA_PS_UAPSD_AC3_DELIVERY_EN = (1 << 6),
+	WMI_STA_PS_UAPSD_AC3_TRIGGER_EN  = (1 << 7),
+};
+
+#define WMI_STA_UAPSD_MAX_INTERVAL_MSEC UINT_MAX
+
+struct wmi_sta_uapsd_auto_trig_param {
+	u32 wmm_ac;
+	u32 user_priority;
+	u32 service_interval;
+	u32 suspend_interval;
+	u32 delay_interval;
+};
+
+struct wmi_sta_uapsd_auto_trig_cmd_fixed_param {
+	u32 vdev_id;
+	struct wmi_mac_addr peer_macaddr;
+	u32 num_ac;
+};
+
+struct wmi_sta_uapsd_auto_trig_arg {
+	u32 wmm_ac;
+	u32 user_priority;
+	u32 service_interval;
+	u32 suspend_interval;
+	u32 delay_interval;
+};
+
+enum wmi_sta_ps_param_tx_wake_threshold {
+	WMI_STA_PS_TX_WAKE_THRESHOLD_NEVER = 0,
+	WMI_STA_PS_TX_WAKE_THRESHOLD_ALWAYS = 1,
+
+	/* Values greater than one indicate that many TX attempts per beacon
+	 * interval before the STA will wake up
+	 */
+};
+
+/* The maximum number of PS-Poll frames the FW will send in response to
+ * traffic advertised in TIM before waking up (by sending a null frame with PS
+ * = 0). Value 0 has a special meaning: there is no maximum count and the FW
+ * will send as many PS-Poll as are necessary to retrieve buffered BU. This
+ * parameter is used when the RX wake policy is
+ * WMI_STA_PS_RX_WAKE_POLICY_POLL_UAPSD and ignored when the RX wake
+ * policy is WMI_STA_PS_RX_WAKE_POLICY_WAKE.
+ */
+enum wmi_sta_ps_param_pspoll_count {
+	WMI_STA_PS_PSPOLL_COUNT_NO_MAX = 0,
+	/* Values greater than 0 indicate the maximum numer of PS-Poll frames
+	 * FW will send before waking up.
+	 */
+};
+
+/* U-APSD configuration of peer station from (re)assoc request and TSPECs */
+enum wmi_ap_ps_param_uapsd {
+	WMI_AP_PS_UAPSD_AC0_DELIVERY_EN = (1 << 0),
+	WMI_AP_PS_UAPSD_AC0_TRIGGER_EN  = (1 << 1),
+	WMI_AP_PS_UAPSD_AC1_DELIVERY_EN = (1 << 2),
+	WMI_AP_PS_UAPSD_AC1_TRIGGER_EN  = (1 << 3),
+	WMI_AP_PS_UAPSD_AC2_DELIVERY_EN = (1 << 4),
+	WMI_AP_PS_UAPSD_AC2_TRIGGER_EN  = (1 << 5),
+	WMI_AP_PS_UAPSD_AC3_DELIVERY_EN = (1 << 6),
+	WMI_AP_PS_UAPSD_AC3_TRIGGER_EN  = (1 << 7),
+};
+
+/* U-APSD maximum service period of peer station */
+enum wmi_ap_ps_peer_param_max_sp {
+	WMI_AP_PS_PEER_PARAM_MAX_SP_UNLIMITED = 0,
+	WMI_AP_PS_PEER_PARAM_MAX_SP_2 = 1,
+	WMI_AP_PS_PEER_PARAM_MAX_SP_4 = 2,
+	WMI_AP_PS_PEER_PARAM_MAX_SP_6 = 3,
+	MAX_WMI_AP_PS_PEER_PARAM_MAX_SP,
+};
+
+enum wmi_ap_ps_peer_param {
+	/** Set uapsd configuration for a given peer.
+	 *
+	 * This include the delivery and trigger enabled state for each AC.
+	 * The host MLME needs to set this based on AP capability and stations
+	 * request Set in the association request  received from the station.
+	 *
+	 * Lower 8 bits of the value specify the UAPSD configuration.
+	 *
+	 * (see enum wmi_ap_ps_param_uapsd)
+	 * The default value is 0.
+	 */
+	WMI_AP_PS_PEER_PARAM_UAPSD = 0,
+
+	/**
+	 * Set the service period for a UAPSD capable station
+	 *
+	 * The service period from wme ie in the (re)assoc request frame.
+	 *
+	 * (see enum wmi_ap_ps_peer_param_max_sp)
+	 */
+	WMI_AP_PS_PEER_PARAM_MAX_SP = 1,
+
+	/** Time in seconds for aging out buffered frames
+	 * for STA in power save
+	 */
+	WMI_AP_PS_PEER_PARAM_AGEOUT_TIME = 2,
+
+	/** Specify frame types that are considered SIFS
+	 * RESP trigger frame
+	 */
+	WMI_AP_PS_PEER_PARAM_SIFS_RESP_FRMTYPE = 3,
+
+	/** Specifies the trigger state of TID.
+	 * Valid only for UAPSD frame type
+	 */
+	WMI_AP_PS_PEER_PARAM_SIFS_RESP_UAPSD = 4,
+
+	/* Specifies the WNM sleep state of a STA */
+	WMI_AP_PS_PEER_PARAM_WNM_SLEEP = 5,
+};
+
+#define DISABLE_SIFS_RESPONSE_TRIGGER 0
+
+#define WMI_MAX_KEY_INDEX   3
+#define WMI_MAX_KEY_LEN     32
+
+#define WMI_KEY_PAIRWISE 0x00
+#define WMI_KEY_GROUP    0x01
+
+#define WMI_CIPHER_NONE     0x0 /* clear key */
+#define WMI_CIPHER_WEP      0x1
+#define WMI_CIPHER_TKIP     0x2
+#define WMI_CIPHER_AES_OCB  0x3
+#define WMI_CIPHER_AES_CCM  0x4
+#define WMI_CIPHER_WAPI     0x5
+#define WMI_CIPHER_CKIP     0x6
+#define WMI_CIPHER_AES_CMAC 0x7
+#define WMI_CIPHER_ANY      0x8
+#define WMI_CIPHER_AES_GCM  0x9
+#define WMI_CIPHER_AES_GMAC 0xa
+
+/* Value to disable fixed rate setting */
+#define WMI_FIXED_RATE_NONE	(0xffff)
+
+#define ATH11K_RC_VERSION_OFFSET	28
+#define ATH11K_RC_PREAMBLE_OFFSET	8
+#define ATH11K_RC_NSS_OFFSET		5
+
+#define ATH11K_HW_RATE_CODE(rate, nss, preamble)	\
+	((1 << ATH11K_RC_VERSION_OFFSET) |		\
+	 ((nss) << ATH11K_RC_NSS_OFFSET) |		\
+	 ((preamble) << ATH11K_RC_PREAMBLE_OFFSET) |	\
+	 (rate))
+
+/* Preamble types to be used with VDEV fixed rate configuration */
+enum wmi_rate_preamble {
+	WMI_RATE_PREAMBLE_OFDM,
+	WMI_RATE_PREAMBLE_CCK,
+	WMI_RATE_PREAMBLE_HT,
+	WMI_RATE_PREAMBLE_VHT,
+	WMI_RATE_PREAMBLE_HE,
+};
+
+/**
+ * enum wmi_rtscts_prot_mode - Enable/Disable RTS/CTS and CTS2Self Protection.
+ * @WMI_RTS_CTS_DISABLED : RTS/CTS protection is disabled.
+ * @WMI_USE_RTS_CTS : RTS/CTS Enabled.
+ * @WMI_USE_CTS2SELF : CTS to self protection Enabled.
+ */
+enum wmi_rtscts_prot_mode {
+	WMI_RTS_CTS_DISABLED = 0,
+	WMI_USE_RTS_CTS = 1,
+	WMI_USE_CTS2SELF = 2,
+};
+
+/**
+ * enum wmi_rtscts_profile - Selection of RTS CTS profile along with enabling
+ *                           protection mode.
+ * @WMI_RTSCTS_FOR_NO_RATESERIES - Neither of rate-series should use RTS-CTS
+ * @WMI_RTSCTS_FOR_SECOND_RATESERIES - Only second rate-series will use RTS-CTS
+ * @WMI_RTSCTS_ACROSS_SW_RETRIES - Only the second rate-series will use RTS-CTS,
+ *                                 but if there's a sw retry, both the rate
+ *                                 series will use RTS-CTS.
+ * @WMI_RTSCTS_ERP - RTS/CTS used for ERP protection for every PPDU.
+ * @WMI_RTSCTS_FOR_ALL_RATESERIES - Enable RTS-CTS for all rate series.
+ */
+enum wmi_rtscts_profile {
+	WMI_RTSCTS_FOR_NO_RATESERIES = 0,
+	WMI_RTSCTS_FOR_SECOND_RATESERIES = 1,
+	WMI_RTSCTS_ACROSS_SW_RETRIES = 2,
+	WMI_RTSCTS_ERP = 3,
+	WMI_RTSCTS_FOR_ALL_RATESERIES = 4,
+};
+
+struct ath11k_hal_reg_cap {
+	u32 eeprom_rd;
+	u32 eeprom_rd_ext;
+	u32 regcap1;
+	u32 regcap2;
+	u32 wireless_modes;
+	u32 low_2ghz_chan;
+	u32 high_2ghz_chan;
+	u32 low_5ghz_chan;
+	u32 high_5ghz_chan;
+};
+
+struct ath11k_mem_chunk {
+	void *vaddr;
+	dma_addr_t paddr;
+	u32 len;
+	u32 req_id;
+};
+
+#define WMI_SKB_HEADROOM sizeof(struct wmi_cmd_hdr)
+
+enum wmi_sta_ps_param_rx_wake_policy {
+	WMI_STA_PS_RX_WAKE_POLICY_WAKE = 0,
+	WMI_STA_PS_RX_WAKE_POLICY_POLL_UAPSD = 1,
+};
+
+enum ath11k_hw_txrx_mode {
+	ATH11K_HW_TXRX_RAW = 0,
+	ATH11K_HW_TXRX_NATIVE_WIFI = 1,
+	ATH11K_HW_TXRX_ETHERNET = 2,
+};
+
+struct wmi_wmm_params {
+	u32 tlv_header;
+	u32 cwmin;
+	u32 cwmax;
+	u32 aifs;
+	u32 txoplimit;
+	u32 acm;
+	u32 no_ack;
+} __packed;
+
+struct wmi_wmm_params_arg {
+	u8 acm;
+	u8 aifs;
+	u8 cwmin;
+	u8 cwmax;
+	u16 txop;
+	u8 no_ack;
+};
+
+struct wmi_vdev_set_wmm_params_cmd {
+	u32 tlv_header;
+	u32 vdev_id;
+	struct wmi_wmm_params wmm_params[4];
+	u32 wmm_param_type;
+} __packed;
+
+struct wmi_wmm_params_all_arg {
+	struct wmi_wmm_params_arg ac_be;
+	struct wmi_wmm_params_arg ac_bk;
+	struct wmi_wmm_params_arg ac_vi;
+	struct wmi_wmm_params_arg ac_vo;
+};
+
+struct target_resource_config {
+	u32 num_vdevs;
+	u32 num_peers;
+	u32 num_active_peers;
+	u32 num_offload_peers;
+	u32 num_offload_reorder_buffs;
+	u32 num_peer_keys;
+	u32 num_tids;
+	u32 ast_skid_limit;
+	u32 tx_chain_mask;
+	u32 rx_chain_mask;
+	u32 rx_timeout_pri[4];
+	u32 rx_decap_mode;
+	u32 scan_max_pending_req;
+	u32 bmiss_offload_max_vdev;
+	u32 roam_offload_max_vdev;
+	u32 roam_offload_max_ap_profiles;
+	u32 num_mcast_groups;
+	u32 num_mcast_table_elems;
+	u32 mcast2ucast_mode;
+	u32 tx_dbg_log_size;
+	u32 num_wds_entries;
+	u32 dma_burst_size;
+	u32 mac_aggr_delim;
+	u32 rx_skip_defrag_timeout_dup_detection_check;
+	u32 vow_config;
+	u32 gtk_offload_max_vdev;
+	u32 num_msdu_desc;
+	u32 max_frag_entries;
+	u32 max_peer_ext_stats;
+	u32 smart_ant_cap;
+	u32 bk_minfree;
+	u32 be_minfree;
+	u32 vi_minfree;
+	u32 vo_minfree;
+	u32 rx_batchmode;
+	u32 tt_support;
+	u32 atf_config;
+	u32 iphdr_pad_config;
+	u32 qwrap_config:16,
+	    alloc_frag_desc_for_data_pkt:16;
+	u32 num_tdls_vdevs;
+	u32 num_tdls_conn_table_entries;
+	u32 beacon_tx_offload_max_vdev;
+	u32 num_multicast_filter_entries;
+	u32 num_wow_filters;
+	u32 num_keep_alive_pattern;
+	u32 keep_alive_pattern_size;
+	u32 max_tdls_concurrent_sleep_sta;
+	u32 max_tdls_concurrent_buffer_sta;
+	u32 wmi_send_separate;
+	u32 num_ocb_vdevs;
+	u32 num_ocb_channels;
+	u32 num_ocb_schedules;
+	u32 num_ns_ext_tuples_cfg;
+	u32 bpf_instruction_size;
+	u32 max_bssid_rx_filters;
+	u32 use_pdev_id;
+	u32 peer_map_unmap_v2_support;
+};
+
+#define WMI_MAX_MEM_REQS 32
+
+#define MAX_RADIOS 3
+
+#define WMI_SERVICE_READY_TIMEOUT_HZ (5 * HZ)
+#define WMI_SEND_TIMEOUT_HZ (3 * HZ)
+
+struct ath11k_wmi_base {
+	struct ath11k_base *ab;
+	struct ath11k_pdev_wmi wmi[MAX_RADIOS];
+	enum ath11k_htc_ep_id wmi_endpoint_id[MAX_RADIOS];
+	u32 max_msg_len[MAX_RADIOS];
+
+	struct completion service_ready;
+	struct completion unified_ready;
+	DECLARE_BITMAP(svc_map, WMI_MAX_EXT_SERVICE);
+	wait_queue_head_t tx_credits_wq;
+	const struct wmi_peer_flags_map *peer_flags;
+	u32 num_mem_chunks;
+	u32 rx_decap_mode;
+	struct wmi_host_mem_chunk mem_chunks[WMI_MAX_MEM_REQS];
+
+	enum wmi_host_hw_mode_config_type preferred_hw_mode;
+	struct target_resource_config  wlan_resource_config;
+
+	struct ath11k_targ_cap *targ_cap;
+};
+
+int ath11k_wmi_cmd_send(struct ath11k_pdev_wmi *wmi, struct sk_buff *skb,
+			u32 cmd_id);
+struct sk_buff *ath11k_wmi_alloc_skb(struct ath11k_wmi_base *wmi_sc, u32 len);
+int ath11k_wmi_mgmt_send(struct ath11k *ar, u32 vdev_id, u32 buf_id,
+			 struct sk_buff *frame);
+int ath11k_wmi_bcn_tmpl(struct ath11k *ar, u32 vdev_id,
+			struct ieee80211_mutable_offsets *offs,
+			struct sk_buff *bcn);
+int ath11k_wmi_vdev_down(struct ath11k *ar, u8 vdev_id);
+int ath11k_wmi_vdev_up(struct ath11k *ar, u32 vdev_id, u32 aid,
+		       const u8 *bssid);
+int ath11k_wmi_vdev_stop(struct ath11k *ar, u8 vdev_id);
+int ath11k_wmi_vdev_start(struct ath11k *ar, struct wmi_vdev_start_req_arg *arg,
+			  bool restart);
+int ath11k_wmi_set_peer_param(struct ath11k *ar, const u8 *peer_addr,
+			      u32 vdev_id, u32 param_id, u32 param_val);
+int ath11k_wmi_pdev_set_param(struct ath11k *ar, u32 param_id,
+			      u32 param_value, u8 pdev_id);
+int ath11k_wmi_wait_for_unified_ready(struct ath11k_base *ab);
+int ath11k_wmi_cmd_init(struct ath11k_base *ab);
+int ath11k_wmi_wait_for_service_ready(struct ath11k_base *ab);
+int ath11k_wmi_connect(struct ath11k_base *ab);
+int ath11k_wmi_pdev_attach(struct ath11k_base *ab,
+			   u8 pdev_id);
+int ath11k_wmi_attach(struct ath11k_base *ab);
+void ath11k_wmi_detach(struct ath11k_base *ab);
+int ath11k_wmi_vdev_create(struct ath11k *ar, u8 *macaddr,
+			   struct vdev_create_params *param);
+int ath11k_wmi_peer_rx_reorder_queue_setup(struct ath11k *ar, int vdev_id,
+					   const u8 *addr, dma_addr_t paddr,
+					   u8 tid, u8 ba_window_size_valid,
+					   u32 ba_window_size);
+int ath11k_wmi_send_peer_create_cmd(struct ath11k *ar,
+				    struct peer_create_params *param);
+int ath11k_wmi_vdev_set_param_cmd(struct ath11k *ar, u32 vdev_id,
+				  u32 param_id, u32 param_value);
+
+int ath11k_wmi_set_sta_ps_param(struct ath11k *ar, u32 vdev_id,
+				u32 param, u32 param_value);
+int ath11k_wmi_force_fw_hang_cmd(struct ath11k *ar, u32 type, u32 delay_time_ms);
+int ath11k_wmi_send_peer_delete_cmd(struct ath11k *ar,
+				    const u8 *peer_addr, u8 vdev_id);
+int ath11k_wmi_vdev_delete(struct ath11k *ar, u8 vdev_id);
+void ath11k_wmi_start_scan_init(struct ath11k *ar, struct scan_req_params *arg);
+int ath11k_wmi_send_scan_start_cmd(struct ath11k *ar,
+				   struct scan_req_params *params);
+int ath11k_wmi_send_scan_stop_cmd(struct ath11k *ar,
+				  struct scan_cancel_param *param);
+int ath11k_wmi_send_wmm_update_cmd_tlv(struct ath11k *ar, u32 vdev_id,
+				       struct wmi_wmm_params_all_arg *param);
+int ath11k_wmi_pdev_suspend(struct ath11k *ar, u32 suspend_opt,
+			    u32 pdev_id);
+int ath11k_wmi_pdev_resume(struct ath11k *ar, u32 pdev_id);
+
+int ath11k_wmi_send_peer_assoc_cmd(struct ath11k *ar,
+				   struct peer_assoc_params *param);
+int ath11k_wmi_vdev_install_key(struct ath11k *ar,
+				struct wmi_vdev_install_key_arg *arg);
+int ath11k_wmi_pdev_bss_chan_info_request(struct ath11k *ar,
+					  enum wmi_bss_chan_info_req_type type);
+int ath11k_wmi_send_stats_request_cmd(struct ath11k *ar,
+				      struct stats_request_params *param);
+int ath11k_wmi_send_peer_flush_tids_cmd(struct ath11k *ar,
+					u8 peer_addr[ETH_ALEN],
+					struct peer_flush_params *param);
+int ath11k_wmi_send_set_ap_ps_param_cmd(struct ath11k *ar, u8 *peer_addr,
+					struct ap_ps_params *param);
+int ath11k_wmi_send_scan_chan_list_cmd(struct ath11k *ar,
+				       struct scan_chan_list_params *chan_list);
+int ath11k_wmi_send_dfs_phyerr_offload_enable_cmd(struct ath11k *ar,
+						  u32 pdev_id);
+int ath11k_wmi_send_bcn_offload_control_cmd(struct ath11k *ar,
+					    u32 vdev_id, u32 bcn_ctrl_op);
+int
+ath11k_wmi_send_init_country_cmd(struct ath11k *ar,
+				 struct wmi_init_country_params init_cc_param);
+int ath11k_wmi_pdev_pktlog_enable(struct ath11k *ar, u32 pktlog_filter);
+int ath11k_wmi_pdev_pktlog_disable(struct ath11k *ar);
+int ath11k_wmi_pdev_peer_pktlog_filter(struct ath11k *ar, u8 *addr, u8 enable);
+int
+ath11k_wmi_rx_reord_queue_remove(struct ath11k *ar,
+				 struct rx_reorder_queue_remove_params *param);
+int ath11k_wmi_send_pdev_set_regdomain(struct ath11k *ar,
+				       struct pdev_set_regdomain_params *param);
+int ath11k_wmi_pull_fw_stats(struct ath11k_base *ab, struct sk_buff *skb,
+			     struct ath11k_fw_stats *stats);
+size_t ath11k_wmi_fw_stats_num_peers(struct list_head *head);
+size_t ath11k_wmi_fw_stats_num_peers_extd(struct list_head *head);
+size_t ath11k_wmi_fw_stats_num_vdevs(struct list_head *head);
+void ath11k_wmi_fw_stats_fill(struct ath11k *ar,
+			      struct ath11k_fw_stats *fw_stats, u32 stats_id,
+			      char *buf);
+int ath11k_wmi_simulate_radar(struct ath11k *ar);
+#endif
-- 
cgit v1.2.3-59-g8ed1b


From 02fdd85da2ce9434dab7d5172284336046afeb17 Mon Sep 17 00:00:00 2001
From: Kalle Valo <kvalo@codeaurora.org>
Date: Sat, 23 Nov 2019 10:01:16 +0200
Subject: MAINTAINERS: add ath11k

Add an entry for ath11k to the MAINTAINERS file.

Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 MAINTAINERS | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index c0024b296158..ff08668b1f30 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -13458,6 +13458,13 @@ T:	git git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/ath.git
 S:	Supported
 F:	drivers/net/wireless/ath/ath10k/
 
+QUALCOMM ATHEROS ATH11K WIRELESS DRIVER
+M:	Kalle Valo <kvalo@codeaurora.org>
+L:	ath11k@lists.infradead.org
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/ath.git
+S:	Supported
+F:	drivers/net/wireless/ath/ath11k/
+
 QUALCOMM ATHEROS ATH9K WIRELESS DRIVER
 M:	QCA ath9k Development <ath9k-devel@qca.qualcomm.com>
 L:	linux-wireless@vger.kernel.org
-- 
cgit v1.2.3-59-g8ed1b


From 34682110abc50ffea7e002b0c2fd7ea9e0000ccc Mon Sep 17 00:00:00 2001
From: Max Chou <max.chou@realtek.com>
Date: Wed, 27 Nov 2019 11:01:07 +0800
Subject: Bluetooth: btusb: Edit the logical value for Realtek Bluetooth reset

It should be pull low and pull high on the physical line for the Realtek
Bluetooth reset. gpiod_set_value_cansleep() takes ACTIVE_LOW status for
the logical value settings, so the original commit should be corrected.

Signed-off-by: Max Chou <max.chou@realtek.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 drivers/bluetooth/btusb.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 70e385987d41..82fb2e7b2892 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -552,9 +552,9 @@ static void btusb_rtl_cmd_timeout(struct hci_dev *hdev)
 	}
 
 	bt_dev_err(hdev, "Reset Realtek device via gpio");
-	gpiod_set_value_cansleep(reset_gpio, 0);
-	msleep(200);
 	gpiod_set_value_cansleep(reset_gpio, 1);
+	msleep(200);
+	gpiod_set_value_cansleep(reset_gpio, 0);
 }
 
 static inline void btusb_free_frags(struct btusb_data *data)
-- 
cgit v1.2.3-59-g8ed1b


From 5d6f391073d5c1c903ac12be72c66b96b2ae93f4 Mon Sep 17 00:00:00 2001
From: Abhishek Pandit-Subedi <abhishekpandit@chromium.org>
Date: Tue, 26 Nov 2019 08:17:29 +0100
Subject: Bluetooth: hci_bcm: Disallow set_baudrate for BCM4354

Without updating the patchram, the BCM4354 does not support a higher
operating speed. The normal bcm_setup follows the correct order
(init_speed, patchram and then oper_speed) but the serdev driver will
set the operating speed before calling the hu->setup function. Thus,
for the BCM4354, don't set the operating speed before patchram.

Signed-off-by: Abhishek Pandit-Subedi <abhishekpandit@chromium.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
---
 drivers/bluetooth/hci_bcm.c | 31 +++++++++++++++++++++++++++++--
 1 file changed, 29 insertions(+), 2 deletions(-)

diff --git a/drivers/bluetooth/hci_bcm.c b/drivers/bluetooth/hci_bcm.c
index d2a6a4afdbbb..0200eff12bc9 100644
--- a/drivers/bluetooth/hci_bcm.c
+++ b/drivers/bluetooth/hci_bcm.c
@@ -47,6 +47,14 @@
 
 #define BCM_NUM_SUPPLIES 2
 
+/**
+ * struct bcm_device_data - device specific data
+ * @no_early_set_baudrate: Disallow set baudrate before driver setup()
+ */
+struct bcm_device_data {
+	bool	no_early_set_baudrate;
+};
+
 /**
  * struct bcm_device - device driver resources
  * @serdev_hu: HCI UART controller struct
@@ -79,6 +87,7 @@
  * @hu: pointer to HCI UART controller struct,
  *	used to disable flow control during runtime suspend and system sleep
  * @is_suspended: whether flow control is currently disabled
+ * @no_early_set_baudrate: don't set_baudrate before setup()
  */
 struct bcm_device {
 	/* Must be the first member, hci_serdev.c expects this. */
@@ -112,6 +121,7 @@ struct bcm_device {
 	struct hci_uart		*hu;
 	bool			is_suspended;
 #endif
+	bool			no_early_set_baudrate;
 };
 
 /* generic bcm uart resources */
@@ -447,7 +457,13 @@ out:
 	if (bcm->dev) {
 		hci_uart_set_flow_control(hu, true);
 		hu->init_speed = bcm->dev->init_speed;
-		hu->oper_speed = bcm->dev->oper_speed;
+
+		/* If oper_speed is set, ldisc/serdev will set the baudrate
+		 * before calling setup()
+		 */
+		if (!bcm->dev->no_early_set_baudrate)
+			hu->oper_speed = bcm->dev->oper_speed;
+
 		err = bcm_gpio_set_power(bcm->dev, true);
 		hci_uart_set_flow_control(hu, false);
 		if (err)
@@ -565,6 +581,8 @@ static int bcm_setup(struct hci_uart *hu)
 	/* Operational speed if any */
 	if (hu->oper_speed)
 		speed = hu->oper_speed;
+	else if (bcm->dev && bcm->dev->oper_speed)
+		speed = bcm->dev->oper_speed;
 	else if (hu->proto->oper_speed)
 		speed = hu->proto->oper_speed;
 	else
@@ -1374,6 +1392,7 @@ static struct platform_driver bcm_driver = {
 static int bcm_serdev_probe(struct serdev_device *serdev)
 {
 	struct bcm_device *bcmdev;
+	const struct bcm_device_data *data;
 	int err;
 
 	bcmdev = devm_kzalloc(&serdev->dev, sizeof(*bcmdev), GFP_KERNEL);
@@ -1408,6 +1427,10 @@ static int bcm_serdev_probe(struct serdev_device *serdev)
 	if (err)
 		dev_err(&serdev->dev, "Failed to power down\n");
 
+	data = device_get_match_data(bcmdev->dev);
+	if (data)
+		bcmdev->no_early_set_baudrate = data->no_early_set_baudrate;
+
 	return hci_uart_register_device(&bcmdev->serdev_hu, &bcm_proto);
 }
 
@@ -1419,12 +1442,16 @@ static void bcm_serdev_remove(struct serdev_device *serdev)
 }
 
 #ifdef CONFIG_OF
+static struct bcm_device_data bcm4354_device_data = {
+	.no_early_set_baudrate = true,
+};
+
 static const struct of_device_id bcm_bluetooth_of_match[] = {
 	{ .compatible = "brcm,bcm20702a1" },
 	{ .compatible = "brcm,bcm4345c5" },
 	{ .compatible = "brcm,bcm4330-bt" },
 	{ .compatible = "brcm,bcm43438-bt" },
-	{ .compatible = "brcm,bcm43540-bt" },
+	{ .compatible = "brcm,bcm43540-bt", .data = &bcm4354_device_data },
 	{ .compatible = "brcm,bcm4335a0" },
 	{ },
 };
-- 
cgit v1.2.3-59-g8ed1b


From 528379902337102b0264fe5343eafb3d6c59fa45 Mon Sep 17 00:00:00 2001
From: Abhishek Pandit-Subedi <abhishekpandit@chromium.org>
Date: Tue, 26 Nov 2019 08:17:30 +0100
Subject: Bluetooth: btbcm: Support pcm configuration

Add BCM vendor specific command to configure PCM parameters. The new
vendor opcode allows us to set the sco routing, the pcm interface rate,
and a few other pcm specific options (frame sync, sync mode, and clock
mode). See broadcom-bluetooth.txt in Documentation for more information
about valid values for those settings.

Here is an example trace where this opcode was used to configure
a BCM4354:

        < HCI Command: Vendor (0x3f|0x001c) plen 5
                01 02 00 01 01
        > HCI Event: Command Complete (0x0e) plen 4
        Vendor (0x3f|0x001c) ncmd 1
                Status: Success (0x00)

We can read back the values as well with ocf 0x001d to confirm the
values that were set:
        $ hcitool cmd 0x3f 0x001d
        < HCI Command: ogf 0x3f, ocf 0x001d, plen 0
        > HCI Event: 0x0e plen 9
        01 1D FC 00 01 02 00 01 01

Signed-off-by: Abhishek Pandit-Subedi <abhishekpandit@chromium.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
---
 drivers/bluetooth/btbcm.c | 46 ++++++++++++++++++++++++++++++++++++++++++++++
 drivers/bluetooth/btbcm.h | 16 ++++++++++++++++
 2 files changed, 62 insertions(+)

diff --git a/drivers/bluetooth/btbcm.c b/drivers/bluetooth/btbcm.c
index 8e05706fe5d9..0795a49edfae 100644
--- a/drivers/bluetooth/btbcm.c
+++ b/drivers/bluetooth/btbcm.c
@@ -107,6 +107,52 @@ int btbcm_set_bdaddr(struct hci_dev *hdev, const bdaddr_t *bdaddr)
 }
 EXPORT_SYMBOL_GPL(btbcm_set_bdaddr);
 
+int btbcm_read_pcm_int_params(struct hci_dev *hdev,
+			      struct bcm_set_pcm_int_params *params)
+{
+	struct sk_buff *skb;
+	int err = 0;
+
+	skb = __hci_cmd_sync(hdev, 0xfc1d, 0, NULL, HCI_INIT_TIMEOUT);
+	if (IS_ERR(skb)) {
+		err = PTR_ERR(skb);
+		bt_dev_err(hdev, "BCM: Read PCM int params failed (%d)", err);
+		return err;
+	}
+
+	if (skb->len != 6 || skb->data[0]) {
+		bt_dev_err(hdev, "BCM: Read PCM int params length mismatch");
+		kfree_skb(skb);
+		return -EIO;
+	}
+
+	if (params)
+		memcpy(params, skb->data + 1, 5);
+
+	kfree_skb(skb);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(btbcm_read_pcm_int_params);
+
+int btbcm_write_pcm_int_params(struct hci_dev *hdev,
+			       const struct bcm_set_pcm_int_params *params)
+{
+	struct sk_buff *skb;
+	int err;
+
+	skb = __hci_cmd_sync(hdev, 0xfc1c, 5, params, HCI_INIT_TIMEOUT);
+	if (IS_ERR(skb)) {
+		err = PTR_ERR(skb);
+		bt_dev_err(hdev, "BCM: Write PCM int params failed (%d)", err);
+		return err;
+	}
+	kfree_skb(skb);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(btbcm_write_pcm_int_params);
+
 int btbcm_patchram(struct hci_dev *hdev, const struct firmware *fw)
 {
 	const struct hci_command_hdr *cmd;
diff --git a/drivers/bluetooth/btbcm.h b/drivers/bluetooth/btbcm.h
index d204be8a84bf..3c7dd0765837 100644
--- a/drivers/bluetooth/btbcm.h
+++ b/drivers/bluetooth/btbcm.h
@@ -54,6 +54,10 @@ struct bcm_set_pcm_format_params {
 int btbcm_check_bdaddr(struct hci_dev *hdev);
 int btbcm_set_bdaddr(struct hci_dev *hdev, const bdaddr_t *bdaddr);
 int btbcm_patchram(struct hci_dev *hdev, const struct firmware *fw);
+int btbcm_read_pcm_int_params(struct hci_dev *hdev,
+			      struct bcm_set_pcm_int_params *params);
+int btbcm_write_pcm_int_params(struct hci_dev *hdev,
+			       const struct bcm_set_pcm_int_params *params);
 
 int btbcm_setup_patchram(struct hci_dev *hdev);
 int btbcm_setup_apple(struct hci_dev *hdev);
@@ -74,6 +78,18 @@ static inline int btbcm_set_bdaddr(struct hci_dev *hdev, const bdaddr_t *bdaddr)
 	return -EOPNOTSUPP;
 }
 
+int btbcm_read_pcm_int_params(struct hci_dev *hdev,
+			      struct bcm_set_pcm_int_params *params)
+{
+	return -EOPNOTSUPP;
+}
+
+int btbcm_write_pcm_int_params(struct hci_dev *hdev,
+			       const struct bcm_set_pcm_int_params *params)
+{
+	return -EOPNOTSUPP;
+}
+
 static inline int btbcm_patchram(struct hci_dev *hdev, const struct firmware *fw)
 {
 	return -EOPNOTSUPP;
-- 
cgit v1.2.3-59-g8ed1b


From 6fc0e19036d6198b9287bb74c0c9df7c17ec8c2e Mon Sep 17 00:00:00 2001
From: Abhishek Pandit-Subedi <abhishekpandit@chromium.org>
Date: Tue, 26 Nov 2019 08:17:31 +0100
Subject: dt-bindings: net: bluetooth: update broadcom-bluetooth

Add documentation for brcm,bt-pcm-int-params vendor specific
configuration of the SCO PCM settings.

Signed-off-by: Abhishek Pandit-Subedi <abhishekpandit@chromium.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
---
 Documentation/devicetree/bindings/net/broadcom-bluetooth.txt | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/broadcom-bluetooth.txt b/Documentation/devicetree/bindings/net/broadcom-bluetooth.txt
index f16b99571af1..b02a53275c98 100644
--- a/Documentation/devicetree/bindings/net/broadcom-bluetooth.txt
+++ b/Documentation/devicetree/bindings/net/broadcom-bluetooth.txt
@@ -30,6 +30,12 @@ Optional properties:
    - "lpo": external low power 32.768 kHz clock
  - vbat-supply: phandle to regulator supply for VBAT
  - vddio-supply: phandle to regulator supply for VDDIO
+ - brcm,bt-pcm-int-params: configure PCM parameters via a 5-byte array
+    - sco-routing: 0 = PCM, 1 = Transport, 2 = Codec, 3 = I2S
+    - pcm-interface-rate: 128KBps, 256KBps, 512KBps, 1024KBps, 2048KBps
+    - pcm-frame-type: short, long
+    - pcm-sync-mode: slave, master
+    - pcm-clock-mode: slave, master
 
 
 Example:
@@ -41,5 +47,6 @@ Example:
        bluetooth {
                compatible = "brcm,bcm43438-bt";
                max-speed = <921600>;
+               brcm,bt-pcm-int-params = [1 2 0 1 1];
        };
 };
-- 
cgit v1.2.3-59-g8ed1b


From eb762b94111b646b4f116ebfdbfcadbad14e12b3 Mon Sep 17 00:00:00 2001
From: Abhishek Pandit-Subedi <abhishekpandit@chromium.org>
Date: Tue, 26 Nov 2019 08:17:32 +0100
Subject: Bluetooth: hci_bcm: Support pcm params in dts

BCM chips may require configuration of PCM to operate correctly and
there is a vendor specific HCI command to do this. Add support in the
hci_bcm driver to parse this from devicetree and configure the chip.

Signed-off-by: Abhishek Pandit-Subedi <abhishekpandit@chromium.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
---
 drivers/bluetooth/hci_bcm.c | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

diff --git a/drivers/bluetooth/hci_bcm.c b/drivers/bluetooth/hci_bcm.c
index 0200eff12bc9..f8f5c593a05c 100644
--- a/drivers/bluetooth/hci_bcm.c
+++ b/drivers/bluetooth/hci_bcm.c
@@ -122,6 +122,7 @@ struct bcm_device {
 	bool			is_suspended;
 #endif
 	bool			no_early_set_baudrate;
+	u8			pcm_int_params[5];
 };
 
 /* generic bcm uart resources */
@@ -594,6 +595,16 @@ static int bcm_setup(struct hci_uart *hu)
 			host_set_baudrate(hu, speed);
 	}
 
+	/* PCM parameters if provided */
+	if (bcm->dev && bcm->dev->pcm_int_params[0] != 0xff) {
+		struct bcm_set_pcm_int_params params;
+
+		btbcm_read_pcm_int_params(hu->hdev, &params);
+
+		memcpy(&params, bcm->dev->pcm_int_params, 5);
+		btbcm_write_pcm_int_params(hu->hdev, &params);
+	}
+
 finalize:
 	release_firmware(fw);
 
@@ -1131,6 +1142,8 @@ static int bcm_acpi_probe(struct bcm_device *dev)
 static int bcm_of_probe(struct bcm_device *bdev)
 {
 	device_property_read_u32(bdev->dev, "max-speed", &bdev->oper_speed);
+	device_property_read_u8_array(bdev->dev, "brcm,bt-pcm-int-params",
+				      bdev->pcm_int_params, 5);
 	return 0;
 }
 
@@ -1146,6 +1159,9 @@ static int bcm_probe(struct platform_device *pdev)
 	dev->dev = &pdev->dev;
 	dev->irq = platform_get_irq(pdev, 0);
 
+	/* Initialize routing field to an unused value */
+	dev->pcm_int_params[0] = 0xff;
+
 	if (has_acpi_companion(&pdev->dev)) {
 		ret = bcm_acpi_probe(dev);
 		if (ret)
@@ -1406,6 +1422,9 @@ static int bcm_serdev_probe(struct serdev_device *serdev)
 	bcmdev->serdev_hu.serdev = serdev;
 	serdev_device_set_drvdata(serdev, bcmdev);
 
+	/* Initialize routing field to an unused value */
+	bcmdev->pcm_int_params[0] = 0xff;
+
 	if (has_acpi_companion(&serdev->dev))
 		err = bcm_acpi_probe(bcmdev);
 	else
-- 
cgit v1.2.3-59-g8ed1b


From e4eb7b5c335ccfdb5c2a9a2004aca7cb81e0d577 Mon Sep 17 00:00:00 2001
From: John Crispin <john@phrozen.org>
Date: Mon, 25 Nov 2019 16:36:18 +0000
Subject: ath11k: add RX stats support for radiotap

mac80211 expects the definition of what HE rate info is available inside a
struct prepended to the skb.

Signed-off-by: Shashidhar Lakkavalli <slakkavalli@datto.com>
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/dp_rx.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c b/drivers/net/wireless/ath/ath11k/dp_rx.c
index 9491a479fab5..c498c4920e11 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
@@ -2038,11 +2038,21 @@ static char *ath11k_print_get_tid(struct ieee80211_hdr *hdr, char *out,
 static void ath11k_dp_rx_deliver_msdu(struct ath11k *ar, struct napi_struct *napi,
 				      struct sk_buff *msdu)
 {
+	static const struct ieee80211_radiotap_he known = {
+		.data1 = cpu_to_le16(IEEE80211_RADIOTAP_HE_DATA1_DATA_MCS_KNOWN),
+		.data2 = cpu_to_le16(IEEE80211_RADIOTAP_HE_DATA2_GI_KNOWN),
+	};
 	struct ieee80211_rx_status *status;
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)msdu->data;
+	struct ieee80211_radiotap_he *he = NULL;
 	char tid[32];
 
 	status = IEEE80211_SKB_RXCB(msdu);
+	if (status->encoding == RX_ENC_HE) {
+		he = skb_push(msdu, sizeof(known));
+		memcpy(he, &known, sizeof(known));
+		status->flag |= RX_FLAG_RADIOTAP_HE;
+	}
 
 	ath11k_dbg(ar->ab, ATH11K_DBG_DATA,
 		   "rx skb %pK len %u peer %pM %s %s sn %u %s%s%s%s%s%s%s %srate_idx %u vht_nss %u freq %u band %u flag 0x%x fcs-err %i mic-err %i amsdu-more %i\n",
-- 
cgit v1.2.3-59-g8ed1b


From 9cfbae4632faf26ab43f65f3beaf59a00dc699f6 Mon Sep 17 00:00:00 2001
From: John Crispin <john@phrozen.org>
Date: Mon, 25 Nov 2019 16:36:19 +0000
Subject: ath11k: ignore event 0x6017

Everytime a new peer gets associated, we see the following message in the log
- ath11k c000000.wifi1: Unknown eventid: 0x6017
Ignore this event for now. We probably need to handle the event properly when
we add OMI support.

Signed-off-by: Shashidhar Lakkavalli <slakkavalli@datto.com>
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/wmi.c |  1 +
 drivers/net/wireless/ath/ath11k/wmi.h | 13 ++++++++++++-
 2 files changed, 13 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath11k/wmi.c b/drivers/net/wireless/ath/ath11k/wmi.c
index 3d1d4d3c3f20..3ec5dd47cdc8 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.c
+++ b/drivers/net/wireless/ath/ath11k/wmi.c
@@ -5416,6 +5416,7 @@ static void ath11k_wmi_tlv_op_rx(struct ath11k_base *ab, struct sk_buff *skb)
 	/* add Unsupported events here */
 	case WMI_TBTTOFFSET_EXT_UPDATE_EVENTID:
 	case WMI_VDEV_DELETE_RESP_EVENTID:
+	case WMI_PEER_OPER_MODE_CHANGE_EVENTID:
 		ath11k_dbg(ab, ATH11K_DBG_WMI,
 			   "ignoring unsupported event 0x%x\n", id);
 		break;
diff --git a/drivers/net/wireless/ath/ath11k/wmi.h b/drivers/net/wireless/ath/ath11k/wmi.h
index 00c51cc83d47..1110da29cfcd 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.h
+++ b/drivers/net/wireless/ath/ath11k/wmi.h
@@ -288,7 +288,6 @@ enum wmi_tlv_cmd_id {
 	WMI_PEER_REORDER_QUEUE_REMOVE_CMDID,
 	WMI_PEER_SET_RX_BLOCKSIZE_CMDID,
 	WMI_PEER_ANTDIV_INFO_REQ_CMDID,
-	WMI_PEER_OPER_MODE_CHANGE_EVENTID,
 	WMI_BCN_TX_CMDID = WMI_TLV_CMD(WMI_GRP_MGMT),
 	WMI_PDEV_SEND_BCN_CMDID,
 	WMI_BCN_TMPL_CMDID,
@@ -586,6 +585,18 @@ enum wmi_tlv_event_id {
 	WMI_WDS_PEER_EVENTID,
 	WMI_PEER_STA_PS_STATECHG_EVENTID,
 	WMI_PEER_ANTDIV_INFO_EVENTID,
+	WMI_PEER_RESERVED0_EVENTID,
+	WMI_PEER_RESERVED1_EVENTID,
+	WMI_PEER_RESERVED2_EVENTID,
+	WMI_PEER_RESERVED3_EVENTID,
+	WMI_PEER_RESERVED4_EVENTID,
+	WMI_PEER_RESERVED5_EVENTID,
+	WMI_PEER_RESERVED6_EVENTID,
+	WMI_PEER_RESERVED7_EVENTID,
+	WMI_PEER_RESERVED8_EVENTID,
+	WMI_PEER_RESERVED9_EVENTID,
+	WMI_PEER_RESERVED10_EVENTID,
+	WMI_PEER_OPER_MODE_CHANGE_EVENTID,
 	WMI_MGMT_RX_EVENTID = WMI_TLV_CMD(WMI_GRP_MGMT),
 	WMI_HOST_SWBA_EVENTID,
 	WMI_TBTTOFFSET_UPDATE_EVENTID,
-- 
cgit v1.2.3-59-g8ed1b


From 2dab7d221e88c83f88af5996d811dea62512dff2 Mon Sep 17 00:00:00 2001
From: John Crispin <john@phrozen.org>
Date: Mon, 25 Nov 2019 16:36:21 +0000
Subject: ath11k: convert message from info to dbg

We can regularly see the following message.
- "ath11k c000000.wifi1: failed to find the peer with peer_id 4"
This happens when the FW starts sending stats for the peer whilst the
peer is not fully associated. Convert this info message to a debug one.

Signed-off-by: Shashidhar Lakkavalli <slakkavalli@datto.com>
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/dp_rx.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c b/drivers/net/wireless/ath/ath11k/dp_rx.c
index c498c4920e11..d5963509419b 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
@@ -2621,8 +2621,9 @@ int ath11k_dp_rx_process_mon_status(struct ath11k_base *ab, int mac_id,
 		peer = ath11k_peer_find_by_id(ab, ppdu_info.peer_id);
 
 		if (!peer || !peer->sta) {
-			ath11k_warn(ab, "failed to find the peer with peer_id %d\n",
-				    ppdu_info.peer_id);
+			ath11k_dbg(ab, ATH11K_DBG_DATA,
+				   "failed to find the peer with peer_id %d\n",
+				   ppdu_info.peer_id);
 			spin_unlock_bh(&ab->base_lock);
 			rcu_read_unlock();
 			dev_kfree_skb_any(skb);
-- 
cgit v1.2.3-59-g8ed1b


From 39e81c6a2907eff835247d61c0f467327ab6d6c0 Mon Sep 17 00:00:00 2001
From: Tamizh chelvam <tamizhr@codeaurora.org>
Date: Mon, 25 Nov 2019 16:36:23 +0000
Subject: ath11k: fix missed bw conversion in tx completion

TX rate stats for the retried packets for a station comes through
tx completion events. Assigning hw reported bandwidth information
directly to station's txrate bandwidth will cause below warning.
Fix this warning by converting the hw reported bandwidth to
mac80211 base bandwidth.

[ 134.758190] PC is at cfg80211_calculate_bitrate+0x1bc/0x214 [cfg80211]
[ 134.765730] LR is at cfg80211_calculate_bitrate+0x1bc/0x214 [cfg80211]
[ 134.875014] [<ffffffbffca8d708>] cfg80211_calculate_bitrate+0x1bc/0x214 [cfg80211]
[ 134.877192] [<ffffffbffcaa9704>] nl80211_put_sta_rate+0x54/0xf24 [cfg80211]
[ 134.884829] [<ffffffbffcaa9d48>] nl80211_put_sta_rate+0x698/0xf24 [cfg80211]
[ 134.891687] [<ffffffbffcaaa490>] nl80211_put_sta_rate+0xde0/0xf24 [cfg80211]
[ 134.898975] [<ffffffc0004de748>] genl_lock_dumpit+0x30/0x4c
[ 134.905998] [<ffffffc0004dc264>] netlink_dump+0xf4/0x248
[ 134.911291] [<ffffffc0004dc910>] __netlink_dump_start+0xe0/0x174
[ 134.916850] [<ffffffc0004df114>] genl_family_rcv_msg+0x130/0x2c0

Signed-off-by: Tamizh chelvam <tamizhr@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/debugfs_sta.c |  2 +-
 drivers/net/wireless/ath/ath11k/dp_rx.c       | 30 ++++-----------------------
 drivers/net/wireless/ath/ath11k/mac.c         | 22 ++++++++++++++++++++
 drivers/net/wireless/ath/ath11k/mac.h         |  1 +
 4 files changed, 28 insertions(+), 27 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/debugfs_sta.c b/drivers/net/wireless/ath/ath11k/debugfs_sta.c
index b392117eff3c..3c5f931e22a9 100644
--- a/drivers/net/wireless/ath/ath11k/debugfs_sta.c
+++ b/drivers/net/wireless/ath/ath11k/debugfs_sta.c
@@ -192,7 +192,7 @@ void ath11k_update_per_peer_stats_from_txcompl(struct ath11k *ar,
 	}
 
 	arsta->txrate.nss = arsta->last_txrate.nss;
-	arsta->txrate.bw = ts->bw;
+	arsta->txrate.bw = ath11k_mac_bw_to_mac80211_bw(ts->bw);
 
 	ath11k_accumulate_per_peer_tx_stats(arsta, peer_stats, rate_idx);
 	spin_unlock_bh(&ab->base_lock);
diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c b/drivers/net/wireless/ath/ath11k/dp_rx.c
index d5963509419b..0ff0ff81adeb 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
@@ -987,28 +987,6 @@ int ath11k_dp_htt_tlv_iter(struct ath11k_base *ab, const void *ptr, size_t len,
 	return 0;
 }
 
-static u8 ath11k_bw_to_mac80211_bw(u8 bw)
-{
-	u8 ret = 0;
-
-	switch (bw) {
-	case ATH11K_BW_20:
-		ret = RATE_INFO_BW_20;
-		break;
-	case ATH11K_BW_40:
-		ret = RATE_INFO_BW_40;
-		break;
-	case ATH11K_BW_80:
-		ret = RATE_INFO_BW_80;
-		break;
-	case ATH11K_BW_160:
-		ret = RATE_INFO_BW_160;
-		break;
-	}
-
-	return ret;
-}
-
 static u32 ath11k_bw_to_mac80211_bwflags(u8 bw)
 {
 	u32 bwflags = 0;
@@ -1157,7 +1135,7 @@ ath11k_update_per_peer_tx_stats(struct ath11k *ar,
 	}
 
 	arsta->txrate.nss = nss;
-	arsta->txrate.bw = ath11k_bw_to_mac80211_bw(bw);
+	arsta->txrate.bw = ath11k_mac_bw_to_mac80211_bw(bw);
 	arsta->tx_info.status.rates[0].flags |= ath11k_bw_to_mac80211_bwflags(bw);
 
 	memcpy(&arsta->last_txrate, &arsta->txrate, sizeof(struct rate_info));
@@ -1934,7 +1912,7 @@ static void ath11k_dp_rx_h_rate(struct ath11k *ar, struct hal_rx_desc *rx_desc,
 		rx_status->rate_idx = rate_mcs + (8 * (nss - 1));
 		if (sgi)
 			rx_status->enc_flags |= RX_ENC_FLAG_SHORT_GI;
-		rx_status->bw = ath11k_bw_to_mac80211_bw(bw);
+		rx_status->bw = ath11k_mac_bw_to_mac80211_bw(bw);
 		break;
 	case RX_MSDU_START_PKT_TYPE_11AC:
 		rx_status->encoding = RX_ENC_VHT;
@@ -1948,7 +1926,7 @@ static void ath11k_dp_rx_h_rate(struct ath11k *ar, struct hal_rx_desc *rx_desc,
 		rx_status->nss = nss;
 		if (sgi)
 			rx_status->enc_flags |= RX_ENC_FLAG_SHORT_GI;
-		rx_status->bw = ath11k_bw_to_mac80211_bw(bw);
+		rx_status->bw = ath11k_mac_bw_to_mac80211_bw(bw);
 		break;
 	case RX_MSDU_START_PKT_TYPE_11AX:
 		rx_status->rate_idx = rate_mcs;
@@ -1960,7 +1938,7 @@ static void ath11k_dp_rx_h_rate(struct ath11k *ar, struct hal_rx_desc *rx_desc,
 		}
 		rx_status->encoding = RX_ENC_HE;
 		rx_status->nss = nss;
-		rx_status->bw = ath11k_bw_to_mac80211_bw(bw);
+		rx_status->bw = ath11k_mac_bw_to_mac80211_bw(bw);
 		break;
 	}
 }
diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index cb025a4a5785..939f5880bce6 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -156,6 +156,28 @@ static const u32 ath11k_smps_map[] = {
 	[WLAN_HT_CAP_SM_PS_DISABLED] = WMI_PEER_SMPS_PS_NONE,
 };
 
+u8 ath11k_mac_bw_to_mac80211_bw(u8 bw)
+{
+	u8 ret = 0;
+
+	switch (bw) {
+	case ATH11K_BW_20:
+		ret = RATE_INFO_BW_20;
+		break;
+	case ATH11K_BW_40:
+		ret = RATE_INFO_BW_40;
+		break;
+	case ATH11K_BW_80:
+		ret = RATE_INFO_BW_80;
+		break;
+	case ATH11K_BW_160:
+		ret = RATE_INFO_BW_160;
+		break;
+	}
+
+	return ret;
+}
+
 int ath11k_mac_hw_ratecode_to_legacy_rate(u8 hw_rc, u8 preamble, u8 *rateidx,
 					  u16 *rate)
 {
diff --git a/drivers/net/wireless/ath/ath11k/mac.h b/drivers/net/wireless/ath/ath11k/mac.h
index f84af1fc4952..8c37573ae5dc 100644
--- a/drivers/net/wireless/ath/ath11k/mac.h
+++ b/drivers/net/wireless/ath/ath11k/mac.h
@@ -142,4 +142,5 @@ struct ath11k *ath11k_mac_get_ar_vdev_stop_status(struct ath11k_base *ab,
 void ath11k_mac_drain_tx(struct ath11k *ar);
 void ath11k_mac_peer_cleanup_all(struct ath11k *ar);
 int ath11k_mac_tx_mgmt_pending_free(int buf_id, void *skb, void *ctx);
+u8 ath11k_mac_bw_to_mac80211_bw(u8 bw);
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From 33782a3c379ede6ed443873eae69a0989f5a65ba Mon Sep 17 00:00:00 2001
From: Tamizh chelvam <tamizhr@codeaurora.org>
Date: Mon, 25 Nov 2019 16:36:25 +0000
Subject: ath11k: Remove dead code while handling amsdu packets

Remove unexecuted code while handling amsdu packets.
The same logic is done before calling ath11k_dp_rx_msdu_coalesce

Signed-off-by: Tamizh chelvam <tamizhr@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/dp_rx.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c b/drivers/net/wireless/ath/ath11k/dp_rx.c
index 0ff0ff81adeb..7aade0314e61 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
@@ -1349,12 +1349,6 @@ static int ath11k_dp_rx_msdu_coalesce(struct ath11k *ar,
 	int rem_len;
 	int buf_len;
 
-	if (!rxcb->is_continuation) {
-		skb_put(first, HAL_RX_DESC_SIZE + l3pad_bytes + msdu_len);
-		skb_pull(first, HAL_RX_DESC_SIZE + l3pad_bytes);
-		return 0;
-	}
-
 	if (WARN_ON_ONCE(msdu_len <= (DP_RX_BUFFER_SIZE -
 			 (HAL_RX_DESC_SIZE + l3pad_bytes)))) {
 		skb_put(first, HAL_RX_DESC_SIZE + l3pad_bytes + msdu_len);
-- 
cgit v1.2.3-59-g8ed1b


From 9f056ed8ee01ad6898db49707cdc70ce923be3d0 Mon Sep 17 00:00:00 2001
From: John Crispin <john@phrozen.org>
Date: Mon, 25 Nov 2019 16:36:27 +0000
Subject: ath11k: add HE support

Add basic HE support to the driver. The sband_iftype data is generated from
the capabilities read from the FW.

Signed-off-by: Shashidhar Lakkavalli <slakkavalli@datto.com>
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/core.h |   2 +
 drivers/net/wireless/ath/ath11k/mac.c  | 265 ++++++++++++++++++++++++++++++++-
 drivers/net/wireless/ath/ath11k/reg.c  |   1 +
 drivers/net/wireless/ath/ath11k/wmi.c  |   6 +-
 drivers/net/wireless/ath/ath11k/wmi.h  |   6 +-
 5 files changed, 270 insertions(+), 10 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/core.h b/drivers/net/wireless/ath/ath11k/core.h
index 95447d1be104..065cb9dc880b 100644
--- a/drivers/net/wireless/ath/ath11k/core.h
+++ b/drivers/net/wireless/ath/ath11k/core.h
@@ -431,6 +431,8 @@ struct ath11k {
 
 	struct {
 		struct ieee80211_supported_band sbands[NUM_NL80211_BANDS];
+		struct ieee80211_sband_iftype_data
+			iftype[NUM_NL80211_BANDS][NUM_NL80211_IFTYPES];
 	} mac;
 	unsigned long dev_flags;
 	unsigned int filter_flags;
diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index 939f5880bce6..79c22d86e0fe 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -1140,7 +1140,86 @@ static void ath11k_peer_assoc_h_he(struct ath11k *ar,
 				   struct ieee80211_sta *sta,
 				   struct peer_assoc_params *arg)
 {
-	/* TODO: Implementation */
+	const struct ieee80211_sta_he_cap *he_cap = &sta->he_cap;
+	u16 v;
+
+	if (!he_cap->has_he)
+		return;
+
+	arg->he_flag = true;
+
+	memcpy(&arg->peer_he_cap_macinfo, he_cap->he_cap_elem.mac_cap_info,
+	       sizeof(arg->peer_he_cap_macinfo));
+	memcpy(&arg->peer_he_cap_phyinfo, he_cap->he_cap_elem.phy_cap_info,
+	       sizeof(arg->peer_he_cap_phyinfo));
+	memcpy(&arg->peer_he_ops, &vif->bss_conf.he_operation,
+	       sizeof(arg->peer_he_ops));
+
+	/* the top most byte is used to indicate BSS color info */
+	arg->peer_he_ops &= 0xffffff;
+
+	if (he_cap->he_cap_elem.phy_cap_info[6] &
+	    IEEE80211_HE_PHY_CAP6_PPE_THRESHOLD_PRESENT) {
+		int bit = 7;
+		int nss, ru;
+
+		arg->peer_ppet.numss_m1 = he_cap->ppe_thres[0] &
+					  IEEE80211_PPE_THRES_NSS_MASK;
+		arg->peer_ppet.ru_bit_mask =
+			(he_cap->ppe_thres[0] &
+			 IEEE80211_PPE_THRES_RU_INDEX_BITMASK_MASK) >>
+			IEEE80211_PPE_THRES_RU_INDEX_BITMASK_POS;
+
+		for (nss = 0; nss <= arg->peer_ppet.numss_m1; nss++) {
+			for (ru = 0; ru < 4; ru++) {
+				u32 val = 0;
+				int i;
+
+				if ((arg->peer_ppet.ru_bit_mask & BIT(ru)) == 0)
+					continue;
+				for (i = 0; i < 6; i++) {
+					val >>= 1;
+					val |= ((he_cap->ppe_thres[bit / 8] >>
+						 (bit % 8)) & 0x1) << 5;
+					bit++;
+				}
+				arg->peer_ppet.ppet16_ppet8_ru3_ru0[nss] |=
+								val << (ru * 6);
+			}
+		}
+	}
+
+	switch (sta->bandwidth) {
+	case IEEE80211_STA_RX_BW_160:
+		if (he_cap->he_cap_elem.phy_cap_info[0] &
+		    IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_80PLUS80_MHZ_IN_5G) {
+			v = le16_to_cpu(he_cap->he_mcs_nss_supp.rx_mcs_80p80);
+			arg->peer_he_rx_mcs_set[WMI_HECAP_TXRX_MCS_NSS_IDX_80_80] = v;
+
+			v = le16_to_cpu(he_cap->he_mcs_nss_supp.tx_mcs_80p80);
+			arg->peer_he_tx_mcs_set[WMI_HECAP_TXRX_MCS_NSS_IDX_80_80] = v;
+
+			arg->peer_he_mcs_count++;
+		}
+		v = le16_to_cpu(he_cap->he_mcs_nss_supp.rx_mcs_160);
+		arg->peer_he_rx_mcs_set[WMI_HECAP_TXRX_MCS_NSS_IDX_160] = v;
+
+		v = le16_to_cpu(he_cap->he_mcs_nss_supp.tx_mcs_160);
+		arg->peer_he_tx_mcs_set[WMI_HECAP_TXRX_MCS_NSS_IDX_160] = v;
+
+		arg->peer_he_mcs_count++;
+		/* fall through */
+
+	default:
+		v = le16_to_cpu(he_cap->he_mcs_nss_supp.rx_mcs_80);
+		arg->peer_he_rx_mcs_set[WMI_HECAP_TXRX_MCS_NSS_IDX_80] = v;
+
+		v = le16_to_cpu(he_cap->he_mcs_nss_supp.tx_mcs_80);
+		arg->peer_he_tx_mcs_set[WMI_HECAP_TXRX_MCS_NSS_IDX_80] = v;
+
+		arg->peer_he_mcs_count++;
+		break;
+	}
 }
 
 static void ath11k_peer_assoc_h_smps(struct ieee80211_sta *sta,
@@ -1307,6 +1386,32 @@ static enum wmi_phy_mode ath11k_mac_get_phymode_vht(struct ath11k *ar,
 	return MODE_UNKNOWN;
 }
 
+static enum wmi_phy_mode ath11k_mac_get_phymode_he(struct ath11k *ar,
+						   struct ieee80211_sta *sta)
+{
+	if (sta->bandwidth == IEEE80211_STA_RX_BW_160) {
+		if (sta->he_cap.he_cap_elem.phy_cap_info[0] &
+		     IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G)
+			return MODE_11AX_HE160;
+		else if (sta->he_cap.he_cap_elem.phy_cap_info[0] &
+		     IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_80PLUS80_MHZ_IN_5G)
+			return MODE_11AX_HE80_80;
+		/* not sure if this is a valid case? */
+		return MODE_11AX_HE160;
+	}
+
+	if (sta->bandwidth == IEEE80211_STA_RX_BW_80)
+		return MODE_11AX_HE80;
+
+	if (sta->bandwidth == IEEE80211_STA_RX_BW_40)
+		return MODE_11AX_HE40;
+
+	if (sta->bandwidth == IEEE80211_STA_RX_BW_20)
+		return MODE_11AX_HE20;
+
+	return MODE_UNKNOWN;
+}
+
 static void ath11k_peer_assoc_h_phymode(struct ath11k *ar,
 					struct ieee80211_vif *vif,
 					struct ieee80211_sta *sta,
@@ -1328,7 +1433,14 @@ static void ath11k_peer_assoc_h_phymode(struct ath11k *ar,
 
 	switch (band) {
 	case NL80211_BAND_2GHZ:
-		if (sta->vht_cap.vht_supported &&
+		if (sta->he_cap.has_he) {
+			if (sta->bandwidth == IEEE80211_STA_RX_BW_80)
+				phymode = MODE_11AX_HE80_2G;
+			else if (sta->bandwidth == IEEE80211_STA_RX_BW_40)
+				phymode = MODE_11AX_HE40_2G;
+			else
+				phymode = MODE_11AX_HE20_2G;
+		} else if (sta->vht_cap.vht_supported &&
 		    !ath11k_peer_assoc_h_vht_masked(vht_mcs_mask)) {
 			if (sta->bandwidth == IEEE80211_STA_RX_BW_40)
 				phymode = MODE_11AC_VHT40;
@@ -1345,12 +1457,12 @@ static void ath11k_peer_assoc_h_phymode(struct ath11k *ar,
 		} else {
 			phymode = MODE_11B;
 		}
-		/* TODO: HE */
-
 		break;
 	case NL80211_BAND_5GHZ:
-		/* Check VHT first */
-		if (sta->vht_cap.vht_supported &&
+		/* Check HE first */
+		if (sta->he_cap.has_he) {
+			phymode = ath11k_mac_get_phymode_he(ar, sta);
+		} else if (sta->vht_cap.vht_supported &&
 		    !ath11k_peer_assoc_h_vht_masked(vht_mcs_mask)) {
 			phymode = ath11k_mac_get_phymode_vht(ar, sta);
 		} else if (sta->ht_cap.ht_supported &&
@@ -1362,7 +1474,6 @@ static void ath11k_peer_assoc_h_phymode(struct ath11k *ar,
 		} else {
 			phymode = MODE_11A;
 		}
-		/* TODO: HE Phymode */
 		break;
 	default:
 		break;
@@ -3126,6 +3237,142 @@ static int ath11k_check_chain_mask(struct ath11k *ar, u32 ant, bool is_tx_ant)
 	return 0;
 }
 
+static void ath11k_gen_ppe_thresh(struct ath11k_ppe_threshold *fw_ppet,
+				  u8 *he_ppet)
+{
+	int nss, ru;
+	u8 bit = 7;
+
+	he_ppet[0] = fw_ppet->numss_m1 & IEEE80211_PPE_THRES_NSS_MASK;
+	he_ppet[0] |= (fw_ppet->ru_bit_mask <<
+		       IEEE80211_PPE_THRES_RU_INDEX_BITMASK_POS) &
+		      IEEE80211_PPE_THRES_RU_INDEX_BITMASK_MASK;
+	for (nss = 0; nss <= fw_ppet->numss_m1; nss++) {
+		for (ru = 0; ru < 4; ru++) {
+			u8 val;
+			int i;
+
+			if ((fw_ppet->ru_bit_mask & BIT(ru)) == 0)
+				continue;
+			val = (fw_ppet->ppet16_ppet8_ru3_ru0[nss] >> (ru * 6)) &
+			       0x3f;
+			val = ((val >> 3) & 0x7) | ((val & 0x7) << 3);
+			for (i = 5; i >= 0; i--) {
+				he_ppet[bit / 8] |=
+					((val >> i) & 0x1) << ((bit % 8));
+				bit++;
+			}
+		}
+	}
+}
+
+static int ath11k_mac_copy_he_cap(struct ath11k *ar,
+				  struct ath11k_pdev_cap *cap,
+				  struct ieee80211_sband_iftype_data *data,
+				  int band)
+{
+	int i, idx = 0;
+
+	for (i = 0; i < NUM_NL80211_IFTYPES; i++) {
+		struct ieee80211_sta_he_cap *he_cap = &data[idx].he_cap;
+		struct ath11k_band_cap *band_cap = &cap->band[band];
+		struct ieee80211_he_cap_elem *he_cap_elem =
+				&he_cap->he_cap_elem;
+
+		switch (i) {
+		case NL80211_IFTYPE_STATION:
+		case NL80211_IFTYPE_AP:
+			break;
+
+		default:
+			continue;
+		}
+
+		data[idx].types_mask = BIT(i);
+		he_cap->has_he = true;
+		memcpy(he_cap_elem->mac_cap_info, band_cap->he_cap_info,
+		       sizeof(he_cap_elem->mac_cap_info));
+		memcpy(he_cap_elem->phy_cap_info, band_cap->he_cap_phy_info,
+		       sizeof(he_cap_elem->phy_cap_info));
+
+		he_cap_elem->mac_cap_info[1] |=
+			IEEE80211_HE_MAC_CAP1_TF_MAC_PAD_DUR_MASK;
+		he_cap_elem->phy_cap_info[4] &=
+			~IEEE80211_HE_PHY_CAP4_BEAMFORMEE_MAX_STS_UNDER_80MHZ_MASK;
+		he_cap_elem->phy_cap_info[4] &=
+			~IEEE80211_HE_PHY_CAP4_BEAMFORMEE_MAX_STS_ABOVE_80MHZ_MASK;
+		he_cap_elem->phy_cap_info[4] |= (ar->num_tx_chains - 1) << 2;
+
+		he_cap_elem->phy_cap_info[5] &=
+			~IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_UNDER_80MHZ_MASK;
+		he_cap_elem->phy_cap_info[5] &=
+			~IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_ABOVE_80MHZ_MASK;
+		he_cap_elem->phy_cap_info[5] |= ar->num_tx_chains - 1;
+
+		switch (i) {
+		case NL80211_IFTYPE_AP:
+			he_cap_elem->phy_cap_info[9] |=
+				IEEE80211_HE_PHY_CAP9_RX_1024_QAM_LESS_THAN_242_TONE_RU;
+			break;
+		case NL80211_IFTYPE_STATION:
+			he_cap_elem->mac_cap_info[0] &=
+				~IEEE80211_HE_MAC_CAP0_TWT_RES;
+			he_cap_elem->mac_cap_info[0] |=
+				IEEE80211_HE_MAC_CAP0_TWT_REQ;
+			he_cap_elem->phy_cap_info[9] |=
+				IEEE80211_HE_PHY_CAP9_TX_1024_QAM_LESS_THAN_242_TONE_RU;
+			break;
+		}
+
+		he_cap->he_mcs_nss_supp.rx_mcs_80 =
+			cpu_to_le16(band_cap->he_mcs & 0xffff);
+		he_cap->he_mcs_nss_supp.tx_mcs_80 =
+			cpu_to_le16(band_cap->he_mcs & 0xffff);
+		he_cap->he_mcs_nss_supp.rx_mcs_160 =
+			cpu_to_le16((band_cap->he_mcs >> 16) & 0xffff);
+		he_cap->he_mcs_nss_supp.tx_mcs_160 =
+			cpu_to_le16((band_cap->he_mcs >> 16) & 0xffff);
+		he_cap->he_mcs_nss_supp.rx_mcs_80p80 =
+			cpu_to_le16((band_cap->he_mcs >> 16) & 0xffff);
+		he_cap->he_mcs_nss_supp.tx_mcs_80p80 =
+			cpu_to_le16((band_cap->he_mcs >> 16) & 0xffff);
+
+		memset(he_cap->ppe_thres, 0, sizeof(he_cap->ppe_thres));
+		if (he_cap_elem->phy_cap_info[6] &
+		    IEEE80211_HE_PHY_CAP6_PPE_THRESHOLD_PRESENT)
+			ath11k_gen_ppe_thresh(&band_cap->he_ppet,
+					      he_cap->ppe_thres);
+		idx++;
+	}
+
+	return idx;
+}
+
+static void ath11k_mac_setup_he_cap(struct ath11k *ar,
+				    struct ath11k_pdev_cap *cap)
+{
+	struct ieee80211_supported_band *band = NULL;
+	int count = 0;
+
+	if (cap->supported_bands & WMI_HOST_WLAN_2G_CAP) {
+		count = ath11k_mac_copy_he_cap(ar, cap,
+					       ar->mac.iftype[NL80211_BAND_2GHZ],
+					       NL80211_BAND_2GHZ);
+		band = &ar->mac.sbands[NL80211_BAND_2GHZ];
+		band->iftype_data = ar->mac.iftype[NL80211_BAND_2GHZ];
+	}
+
+	if (cap->supported_bands & WMI_HOST_WLAN_5G_CAP) {
+		count = ath11k_mac_copy_he_cap(ar, cap,
+					       ar->mac.iftype[NL80211_BAND_5GHZ],
+					       NL80211_BAND_5GHZ);
+		band = &ar->mac.sbands[NL80211_BAND_5GHZ];
+		band->iftype_data = ar->mac.iftype[NL80211_BAND_5GHZ];
+	}
+
+	band->n_iftype_data = count;
+}
+
 static int __ath11k_set_antenna(struct ath11k *ar, u32 tx_ant, u32 rx_ant)
 {
 	int ret;
@@ -3161,8 +3408,9 @@ static int __ath11k_set_antenna(struct ath11k *ar, u32 tx_ant, u32 rx_ant)
 		return ret;
 	}
 
-	/* Reload HT/VHT capability */
+	/* Reload HT/VHT/HE capability */
 	ath11k_mac_setup_ht_vht_cap(ar, &ar->pdev->cap, NULL);
+	ath11k_mac_setup_he_cap(ar, &ar->pdev->cap);
 
 	return 0;
 }
@@ -5207,6 +5455,7 @@ static int ath11k_mac_register(struct ath11k *ar)
 		goto err_free;
 
 	ath11k_mac_setup_ht_vht_cap(ar, cap, &ht_cap);
+	ath11k_mac_setup_he_cap(ar, cap);
 
 	ar->hw->wiphy->available_antennas_rx = cap->rx_chain_mask;
 	ar->hw->wiphy->available_antennas_tx = cap->tx_chain_mask;
diff --git a/drivers/net/wireless/ath/ath11k/reg.c b/drivers/net/wireless/ath/ath11k/reg.c
index 810050b033c4..453aa9c06969 100644
--- a/drivers/net/wireless/ath/ath11k/reg.c
+++ b/drivers/net/wireless/ath/ath11k/reg.c
@@ -141,6 +141,7 @@ int ath11k_reg_update_chan_list(struct ath11k *ar)
 			/* TODO: Set to true/false based on some condition? */
 			ch->allow_ht = true;
 			ch->allow_vht = true;
+			ch->allow_he = true;
 
 			ch->dfs_set =
 				!!(channel->flags & IEEE80211_CHAN_RADAR);
diff --git a/drivers/net/wireless/ath/ath11k/wmi.c b/drivers/net/wireless/ath/ath11k/wmi.c
index 3ec5dd47cdc8..d077ea30f3c1 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.c
+++ b/drivers/net/wireless/ath/ath11k/wmi.c
@@ -753,6 +753,8 @@ static void ath11k_wmi_put_wmi_channel(struct wmi_channel *chan,
 		chan->info |= WMI_CHAN_INFO_ALLOW_HT;
 	if (arg->channel.allow_vht)
 		chan->info |= WMI_CHAN_INFO_ALLOW_VHT;
+	if (arg->channel.allow_he)
+		chan->info |= WMI_CHAN_INFO_ALLOW_HE;
 	if (arg->channel.ht40plus)
 		chan->info |= WMI_CHAN_INFO_HT40_PLUS;
 	if (arg->channel.chan_radar)
@@ -2150,7 +2152,9 @@ int ath11k_wmi_send_scan_chan_list_cmd(struct ath11k *ar,
 
 		if (tchan_info->is_chan_passive)
 			chan_info->info |= WMI_CHAN_INFO_PASSIVE;
-		if (tchan_info->allow_vht)
+		if (tchan_info->allow_he)
+			chan_info->info |= WMI_CHAN_INFO_ALLOW_HE;
+		else if (tchan_info->allow_vht)
 			chan_info->info |= WMI_CHAN_INFO_ALLOW_VHT;
 		else if (tchan_info->allow_ht)
 			chan_info->info |= WMI_CHAN_INFO_ALLOW_HT;
diff --git a/drivers/net/wireless/ath/ath11k/wmi.h b/drivers/net/wireless/ath/ath11k/wmi.h
index 1110da29cfcd..fd1ffa593a41 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.h
+++ b/drivers/net/wireless/ath/ath11k/wmi.h
@@ -2445,6 +2445,7 @@ struct channel_param {
 	    is_chan_passive:1,
 	    allow_ht:1,
 	    allow_vht:1,
+	    allow_he:1,
 	    set_agile:1;
 	u32 phy_mode;
 	u32 cfreq1;
@@ -3334,7 +3335,10 @@ struct wmi_vdev_install_key_arg {
 
 #define WMI_MAX_SUPPORTED_RATES			128
 #define WMI_HOST_MAX_HECAP_PHY_SIZE		3
-#define WMI_HOST_MAX_HE_RATE_SET		1
+#define WMI_HOST_MAX_HE_RATE_SET		3
+#define WMI_HECAP_TXRX_MCS_NSS_IDX_80		0
+#define WMI_HECAP_TXRX_MCS_NSS_IDX_160		1
+#define WMI_HECAP_TXRX_MCS_NSS_IDX_80_80	2
 
 struct wmi_rate_set_arg {
 	u32 num_rates;
-- 
cgit v1.2.3-59-g8ed1b


From 6d293d447670da6325cc9c8fb809878d1930c234 Mon Sep 17 00:00:00 2001
From: John Crispin <john@phrozen.org>
Date: Mon, 25 Nov 2019 16:36:28 +0000
Subject: ath11k: add TWT support

Add target wait time wmi calls to the driver. En/disable the support
from when the bss_config changes. We ignore the cmd completion events.

Signed-off-by: Shashidhar Lakkavalli <slakkavalli@datto.com>
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/mac.c | 12 +++++
 drivers/net/wireless/ath/ath11k/wmi.c | 91 +++++++++++++++++++++++++++++++++++
 drivers/net/wireless/ath/ath11k/wmi.h | 70 +++++++++++++++++++++++++++
 3 files changed, 173 insertions(+)

diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index 79c22d86e0fe..70945d121fba 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -1189,6 +1189,11 @@ static void ath11k_peer_assoc_h_he(struct ath11k *ar,
 		}
 	}
 
+	if (he_cap->he_cap_elem.mac_cap_info[0] & IEEE80211_HE_MAC_CAP0_TWT_RES)
+		arg->twt_responder = true;
+	if (he_cap->he_cap_elem.mac_cap_info[0] & IEEE80211_HE_MAC_CAP0_TWT_REQ)
+		arg->twt_requester = true;
+
 	switch (sta->bandwidth) {
 	case IEEE80211_STA_RX_BW_160:
 		if (he_cap->he_cap_elem.phy_cap_info[0] &
@@ -1904,6 +1909,13 @@ static void ath11k_mac_op_bss_info_changed(struct ieee80211_hw *hw,
 	    !ath11k_mac_vif_chan(arvif->vif, &def))
 		ath11k_recalculate_mgmt_rate(ar, vif, &def);
 
+	if (changed & BSS_CHANGED_TWT) {
+		if (info->twt_requester || info->twt_responder)
+			ath11k_wmi_send_twt_enable_cmd(ar, ar->pdev_idx);
+		else
+			ath11k_wmi_send_twt_disable_cmd(ar, ar->pdev_idx);
+	}
+
 	mutex_unlock(&ar->conf_mutex);
 }
 
diff --git a/drivers/net/wireless/ath/ath11k/wmi.c b/drivers/net/wireless/ath/ath11k/wmi.c
index d077ea30f3c1..7aa66e8eecb6 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.c
+++ b/drivers/net/wireless/ath/ath11k/wmi.c
@@ -1622,6 +1622,10 @@ ath11k_wmi_copy_peer_flags(struct wmi_peer_assoc_complete_cmd *cmd,
 			cmd->peer_flags |= WMI_PEER_VHT;
 		if (param->he_flag)
 			cmd->peer_flags |= WMI_PEER_HE;
+		if (param->twt_requester)
+			cmd->peer_flags |= WMI_PEER_TWT_REQ;
+		if (param->twt_responder)
+			cmd->peer_flags |= WMI_PEER_TWT_RESP;
 	}
 
 	/* Suppress authorization for all AUTH modes that need 4-way handshake
@@ -2457,6 +2461,86 @@ int ath11k_wmi_pdev_pktlog_disable(struct ath11k *ar)
 	return ret;
 }
 
+int
+ath11k_wmi_send_twt_enable_cmd(struct ath11k *ar, u32 pdev_id)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct ath11k_base *ab = wmi->wmi_sc->ab;
+	struct wmi_twt_enable_params_cmd *cmd;
+	struct sk_buff *skb;
+	int ret, len;
+
+	len = sizeof(*cmd);
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (void *)skb->data;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_TWT_ENABLE_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, len - TLV_HDR_SIZE);
+	cmd->pdev_id = pdev_id;
+	cmd->sta_cong_timer_ms = ATH11K_TWT_DEF_STA_CONG_TIMER_MS;
+	cmd->default_slot_size = ATH11K_TWT_DEF_DEFAULT_SLOT_SIZE;
+	cmd->congestion_thresh_setup = ATH11K_TWT_DEF_CONGESTION_THRESH_SETUP;
+	cmd->congestion_thresh_teardown =
+		ATH11K_TWT_DEF_CONGESTION_THRESH_TEARDOWN;
+	cmd->congestion_thresh_critical =
+		ATH11K_TWT_DEF_CONGESTION_THRESH_CRITICAL;
+	cmd->interference_thresh_teardown =
+		ATH11K_TWT_DEF_INTERFERENCE_THRESH_TEARDOWN;
+	cmd->interference_thresh_setup =
+		ATH11K_TWT_DEF_INTERFERENCE_THRESH_SETUP;
+	cmd->min_no_sta_setup = ATH11K_TWT_DEF_MIN_NO_STA_SETUP;
+	cmd->min_no_sta_teardown = ATH11K_TWT_DEF_MIN_NO_STA_TEARDOWN;
+	cmd->no_of_bcast_mcast_slots = ATH11K_TWT_DEF_NO_OF_BCAST_MCAST_SLOTS;
+	cmd->min_no_twt_slots = ATH11K_TWT_DEF_MIN_NO_TWT_SLOTS;
+	cmd->max_no_sta_twt = ATH11K_TWT_DEF_MAX_NO_STA_TWT;
+	cmd->mode_check_interval = ATH11K_TWT_DEF_MODE_CHECK_INTERVAL;
+	cmd->add_sta_slot_interval = ATH11K_TWT_DEF_ADD_STA_SLOT_INTERVAL;
+	cmd->remove_sta_slot_interval =
+		ATH11K_TWT_DEF_REMOVE_STA_SLOT_INTERVAL;
+	/* TODO add MBSSID support */
+	cmd->mbss_support = 0;
+
+	ret = ath11k_wmi_cmd_send(wmi, skb,
+				  WMI_TWT_ENABLE_CMDID);
+	if (ret) {
+		ath11k_warn(ab, "Failed to send WMI_TWT_ENABLE_CMDID");
+		dev_kfree_skb(skb);
+	}
+	return ret;
+}
+
+int
+ath11k_wmi_send_twt_disable_cmd(struct ath11k *ar, u32 pdev_id)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct ath11k_base *ab = wmi->wmi_sc->ab;
+	struct wmi_twt_disable_params_cmd *cmd;
+	struct sk_buff *skb;
+	int ret, len;
+
+	len = sizeof(*cmd);
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (void *)skb->data;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_TWT_DISABLE_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, len - TLV_HDR_SIZE);
+	cmd->pdev_id = pdev_id;
+
+	ret = ath11k_wmi_cmd_send(wmi, skb,
+				  WMI_TWT_DISABLE_CMDID);
+	if (ret) {
+		ath11k_warn(ab, "Failed to send WMI_TWT_DIeABLE_CMDID");
+		dev_kfree_skb(skb);
+	}
+	return ret;
+}
+
 static void
 ath11k_fill_band_to_mac_param(struct ath11k_base  *soc,
 			      struct wmi_host_pdev_band_to_mac *band_to_mac)
@@ -2548,6 +2632,9 @@ ath11k_wmi_copy_resource_config(struct wmi_resource_config *wmi_cfg,
 	wmi_cfg->use_pdev_id = tg_cfg->use_pdev_id;
 	wmi_cfg->flag1 = tg_cfg->atf_config;
 	wmi_cfg->peer_map_unmap_v2_support = tg_cfg->peer_map_unmap_v2_support;
+	wmi_cfg->sched_params = tg_cfg->sched_params;
+	wmi_cfg->twt_ap_pdev_count = tg_cfg->twt_ap_pdev_count;
+	wmi_cfg->twt_ap_sta_count = tg_cfg->twt_ap_sta_count;
 }
 
 static int ath11k_init_cmd_send(struct ath11k_pdev_wmi *wmi,
@@ -2739,6 +2826,8 @@ int ath11k_wmi_cmd_init(struct ath11k_base *ab)
 	config.beacon_tx_offload_max_vdev = ab->num_radios * TARGET_MAX_BCN_OFFLD;
 	config.rx_batchmode = TARGET_RX_BATCHMODE;
 	config.peer_map_unmap_v2_support = 1;
+	config.twt_ap_pdev_count = 2;
+	config.twt_ap_sta_count = 1000;
 
 	memcpy(&wmi_sc->wlan_resource_config, &config, sizeof(config));
 
@@ -5421,6 +5510,8 @@ static void ath11k_wmi_tlv_op_rx(struct ath11k_base *ab, struct sk_buff *skb)
 	case WMI_TBTTOFFSET_EXT_UPDATE_EVENTID:
 	case WMI_VDEV_DELETE_RESP_EVENTID:
 	case WMI_PEER_OPER_MODE_CHANGE_EVENTID:
+	case WMI_TWT_ENABLE_EVENTID:
+	case WMI_TWT_DISABLE_EVENTID:
 		ath11k_dbg(ab, ATH11K_DBG_WMI,
 			   "ignoring unsupported event 0x%x\n", id);
 		break;
diff --git a/drivers/net/wireless/ath/ath11k/wmi.h b/drivers/net/wireless/ath/ath11k/wmi.h
index fd1ffa593a41..b5c53f2ac5cc 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.h
+++ b/drivers/net/wireless/ath/ath11k/wmi.h
@@ -169,6 +169,9 @@ enum wmi_cmd_group {
 	WMI_GRP_MONITOR        = 0x39,
 	WMI_GRP_REGULATORY     = 0x3a,
 	WMI_GRP_HW_DATA_FILTER = 0x3b,
+	WMI_GRP_WLM            = 0x3c,
+	WMI_GRP_11K_OFFLOAD    = 0x3d,
+	WMI_GRP_TWT            = 0x3e,
 };
 
 #define WMI_CMD_GRP(grp_id) (((grp_id) << 12) | 0x1)
@@ -530,6 +533,12 @@ enum wmi_tlv_cmd_id {
 	WMI_NDP_RESPONDER_REQ_CMDID,
 	WMI_NDP_END_REQ_CMDID,
 	WMI_HW_DATA_FILTER_CMDID = WMI_TLV_CMD(WMI_GRP_HW_DATA_FILTER),
+	WMI_TWT_ENABLE_CMDID = WMI_TLV_CMD(WMI_GRP_TWT),
+	WMI_TWT_DISABLE_CMDID,
+	WMI_TWT_ADD_DIALOG_CMDID,
+	WMI_TWT_DEL_DIALOG_CMDID,
+	WMI_TWT_PAUSE_DIALOG_CMDID,
+	WMI_TWT_RESUME_DIALOG_CMDID,
 };
 
 enum wmi_tlv_event_id {
@@ -724,6 +733,13 @@ enum wmi_tlv_event_id {
 	WMI_NDP_INDICATION_EVENTID,
 	WMI_NDP_CONFIRM_EVENTID,
 	WMI_NDP_END_INDICATION_EVENTID,
+
+	WMI_TWT_ENABLE_EVENTID = WMI_TLV_CMD(WMI_GRP_TWT),
+	WMI_TWT_DISABLE_EVENTID,
+	WMI_TWT_ADD_DIALOG_EVENTID,
+	WMI_TWT_DEL_DIALOG_EVENTID,
+	WMI_TWT_PAUSE_DIALOG_EVENTID,
+	WMI_TWT_RESUME_DIALOG_EVENTID,
 };
 
 enum wmi_tlv_pdev_param {
@@ -2183,6 +2199,9 @@ struct wmi_resource_config {
 	u32 max_num_dbs_scan_duty_cycle;
 	u32 max_num_group_keys;
 	u32 peer_map_unmap_v2_support;
+	u32 sched_params;
+	u32 twt_ap_pdev_count;
+	u32 twt_ap_sta_count;
 } __packed;
 
 struct wmi_service_ready_event {
@@ -3401,6 +3420,8 @@ struct peer_assoc_params {
 	u32 peer_he_mcs_count;
 	u32 peer_he_rx_mcs_set[WMI_HOST_MAX_HE_RATE_SET];
 	u32 peer_he_tx_mcs_set[WMI_HOST_MAX_HE_RATE_SET];
+	bool twt_responder;
+	bool twt_requester;
 	struct ath11k_ppe_threshold peer_ppet;
 };
 
@@ -3652,6 +3673,8 @@ struct wmi_unit_test_cmd {
 #define WMI_PEER_DYN_MIMOPS	0x00020000
 #define WMI_PEER_STATIC_MIMOPS	0x00040000
 #define WMI_PEER_SPATIAL_MUX	0x00200000
+#define WMI_PEER_TWT_REQ	0x00400000
+#define WMI_PEER_TWT_RESP	0x00800000
 #define WMI_PEER_VHT		0x02000000
 #define WMI_PEER_80MHZ		0x04000000
 #define WMI_PEER_PMF		0x08000000
@@ -4494,6 +4517,48 @@ struct wmi_wmm_params_all_arg {
 	struct wmi_wmm_params_arg ac_vo;
 };
 
+#define ATH11K_TWT_DEF_STA_CONG_TIMER_MS		5000
+#define ATH11K_TWT_DEF_DEFAULT_SLOT_SIZE		10
+#define ATH11K_TWT_DEF_CONGESTION_THRESH_SETUP		50
+#define ATH11K_TWT_DEF_CONGESTION_THRESH_TEARDOWN	20
+#define ATH11K_TWT_DEF_CONGESTION_THRESH_CRITICAL	100
+#define ATH11K_TWT_DEF_INTERFERENCE_THRESH_TEARDOWN	80
+#define ATH11K_TWT_DEF_INTERFERENCE_THRESH_SETUP	50
+#define ATH11K_TWT_DEF_MIN_NO_STA_SETUP			10
+#define ATH11K_TWT_DEF_MIN_NO_STA_TEARDOWN		2
+#define ATH11K_TWT_DEF_NO_OF_BCAST_MCAST_SLOTS		2
+#define ATH11K_TWT_DEF_MIN_NO_TWT_SLOTS			2
+#define ATH11K_TWT_DEF_MAX_NO_STA_TWT			500
+#define ATH11K_TWT_DEF_MODE_CHECK_INTERVAL		10000
+#define ATH11K_TWT_DEF_ADD_STA_SLOT_INTERVAL		1000
+#define ATH11K_TWT_DEF_REMOVE_STA_SLOT_INTERVAL		5000
+
+struct wmi_twt_enable_params_cmd {
+	u32 tlv_header;
+	u32 pdev_id;
+	u32 sta_cong_timer_ms;
+	u32 mbss_support;
+	u32 default_slot_size;
+	u32 congestion_thresh_setup;
+	u32 congestion_thresh_teardown;
+	u32 congestion_thresh_critical;
+	u32 interference_thresh_teardown;
+	u32 interference_thresh_setup;
+	u32 min_no_sta_setup;
+	u32 min_no_sta_teardown;
+	u32 no_of_bcast_mcast_slots;
+	u32 min_no_twt_slots;
+	u32 max_no_sta_twt;
+	u32 mode_check_interval;
+	u32 add_sta_slot_interval;
+	u32 remove_sta_slot_interval;
+};
+
+struct wmi_twt_disable_params_cmd {
+	u32 tlv_header;
+	u32 pdev_id;
+};
+
 struct target_resource_config {
 	u32 num_vdevs;
 	u32 num_peers;
@@ -4553,6 +4618,9 @@ struct target_resource_config {
 	u32 max_bssid_rx_filters;
 	u32 use_pdev_id;
 	u32 peer_map_unmap_v2_support;
+	u32 sched_params;
+	u32 twt_ap_pdev_count;
+	u32 twt_ap_sta_count;
 };
 
 #define WMI_MAX_MEM_REQS 32
@@ -4676,4 +4744,6 @@ void ath11k_wmi_fw_stats_fill(struct ath11k *ar,
 			      struct ath11k_fw_stats *fw_stats, u32 stats_id,
 			      char *buf);
 int ath11k_wmi_simulate_radar(struct ath11k *ar);
+int ath11k_wmi_send_twt_enable_cmd(struct ath11k *ar, u32 pdev_id);
+int ath11k_wmi_send_twt_disable_cmd(struct ath11k *ar, u32 pdev_id);
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From 3f8be640077a33ef3bd717ade8b50bd0ef815ba9 Mon Sep 17 00:00:00 2001
From: John Crispin <john@phrozen.org>
Date: Mon, 25 Nov 2019 16:36:29 +0000
Subject: ath11k: add spatial reuse support

Trigger the WMI call en/disabling OBSS PD when the bss config changes or we
assoc to an AP that broadcasts the IE.

Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/mac.c | 10 ++++++++++
 drivers/net/wireless/ath/ath11k/wmi.c | 35 +++++++++++++++++++++++++++++++++++
 drivers/net/wireless/ath/ath11k/wmi.h | 16 ++++++++++++++++
 3 files changed, 61 insertions(+)

diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index 70945d121fba..c25817c51d87 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -1612,6 +1612,12 @@ static void ath11k_bss_assoc(struct ieee80211_hw *hw,
 					1);
 	if (ret)
 		ath11k_warn(ar->ab, "Unable to authorize BSS peer: %d\n", ret);
+
+	ret = ath11k_wmi_send_obss_spr_cmd(ar, arvif->vdev_id,
+					   &bss_conf->he_obss_pd);
+	if (ret)
+		ath11k_warn(ar->ab, "failed to set vdev %i OBSS PD parameters: %d\n",
+			    arvif->vdev_id, ret);
 }
 
 static void ath11k_bss_disassoc(struct ieee80211_hw *hw,
@@ -1916,6 +1922,10 @@ static void ath11k_mac_op_bss_info_changed(struct ieee80211_hw *hw,
 			ath11k_wmi_send_twt_disable_cmd(ar, ar->pdev_idx);
 	}
 
+	if (changed & BSS_CHANGED_HE_OBSS_PD)
+		ath11k_wmi_send_obss_spr_cmd(ar, arvif->vdev_id,
+					     &info->he_obss_pd);
+
 	mutex_unlock(&ar->conf_mutex);
 }
 
diff --git a/drivers/net/wireless/ath/ath11k/wmi.c b/drivers/net/wireless/ath/ath11k/wmi.c
index 7aa66e8eecb6..2c3c973f5f7c 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.c
+++ b/drivers/net/wireless/ath/ath11k/wmi.c
@@ -2541,6 +2541,41 @@ ath11k_wmi_send_twt_disable_cmd(struct ath11k *ar, u32 pdev_id)
 	return ret;
 }
 
+int
+ath11k_wmi_send_obss_spr_cmd(struct ath11k *ar, u32 vdev_id,
+			     struct ieee80211_he_obss_pd *he_obss_pd)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct ath11k_base *ab = wmi->wmi_sc->ab;
+	struct wmi_obss_spatial_reuse_params_cmd *cmd;
+	struct sk_buff *skb;
+	int ret, len;
+
+	len = sizeof(*cmd);
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (void *)skb->data;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG,
+				     WMI_TAG_OBSS_SPATIAL_REUSE_SET_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, len - TLV_HDR_SIZE);
+	cmd->vdev_id = vdev_id;
+	cmd->enable = he_obss_pd->enable;
+	cmd->obss_min = he_obss_pd->min_offset;
+	cmd->obss_max = he_obss_pd->max_offset;
+
+	ret = ath11k_wmi_cmd_send(wmi, skb,
+				  WMI_PDEV_OBSS_PD_SPATIAL_REUSE_CMDID);
+	if (ret) {
+		ath11k_warn(ab,
+			    "Failed to send WMI_PDEV_OBSS_PD_SPATIAL_REUSE_CMDID");
+		dev_kfree_skb(skb);
+	}
+	return ret;
+}
+
 static void
 ath11k_fill_band_to_mac_param(struct ath11k_base  *soc,
 			      struct wmi_host_pdev_band_to_mac *band_to_mac)
diff --git a/drivers/net/wireless/ath/ath11k/wmi.h b/drivers/net/wireless/ath/ath11k/wmi.h
index b5c53f2ac5cc..7d5690c65279 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.h
+++ b/drivers/net/wireless/ath/ath11k/wmi.h
@@ -172,6 +172,8 @@ enum wmi_cmd_group {
 	WMI_GRP_WLM            = 0x3c,
 	WMI_GRP_11K_OFFLOAD    = 0x3d,
 	WMI_GRP_TWT            = 0x3e,
+	WMI_GRP_MOTION_DET     = 0x3f,
+	WMI_GRP_SPATIAL_REUSE  = 0x40,
 };
 
 #define WMI_CMD_GRP(grp_id) (((grp_id) << 12) | 0x1)
@@ -539,6 +541,9 @@ enum wmi_tlv_cmd_id {
 	WMI_TWT_DEL_DIALOG_CMDID,
 	WMI_TWT_PAUSE_DIALOG_CMDID,
 	WMI_TWT_RESUME_DIALOG_CMDID,
+	WMI_PDEV_OBSS_PD_SPATIAL_REUSE_CMDID =
+				WMI_TLV_CMD(WMI_GRP_SPATIAL_REUSE),
+	WMI_PDEV_OBSS_PD_SPATIAL_REUSE_SET_DEF_OBSS_THRESH_CMDID,
 };
 
 enum wmi_tlv_event_id {
@@ -4559,6 +4564,15 @@ struct wmi_twt_disable_params_cmd {
 	u32 pdev_id;
 };
 
+struct wmi_obss_spatial_reuse_params_cmd {
+	u32 tlv_header;
+	u32 pdev_id;
+	u32 enable;
+	s32 obss_min;
+	s32 obss_max;
+	u32 vdev_id;
+};
+
 struct target_resource_config {
 	u32 num_vdevs;
 	u32 num_peers;
@@ -4746,4 +4760,6 @@ void ath11k_wmi_fw_stats_fill(struct ath11k *ar,
 int ath11k_wmi_simulate_radar(struct ath11k *ar);
 int ath11k_wmi_send_twt_enable_cmd(struct ath11k *ar, u32 pdev_id);
 int ath11k_wmi_send_twt_disable_cmd(struct ath11k *ar, u32 pdev_id);
+int ath11k_wmi_send_obss_spr_cmd(struct ath11k *ar, u32 vdev_id,
+				 struct ieee80211_he_obss_pd *he_obss_pd);
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From ef39ac1b0d9f523b66757dfc080ecdbed688d669 Mon Sep 17 00:00:00 2001
From: Govind Singh <govinds@codeaurora.org>
Date: Mon, 25 Nov 2019 13:10:01 +0000
Subject: ath10k: move non-fatal warn logs to dbg level

During driver load below warn logs are printed in the console if
firmware doesn't support some optional HTC services, ex:pktlog.
It is likely some older fw version may not support PKTLOG HTC
service as legacy fw uses HTC DATA service  for pktlog.
Move this log to debug level to remove un-necessary warn message
on console.

htc.c:803:  ath10k_warn(ar, "unsupported HTC service id: %d\n",
htc.c:881:  ath10k_warn(ar, "unsupported HTC service id: %d\n",

Signed-off-by: Govind Singh <govinds@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/htc.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/htc.c b/drivers/net/wireless/ath/ath10k/htc.c
index 1d4d1a1992fe..0bbd1415d85c 100644
--- a/drivers/net/wireless/ath/ath10k/htc.c
+++ b/drivers/net/wireless/ath/ath10k/htc.c
@@ -800,8 +800,8 @@ setup:
 						&ep->ul_pipe_id,
 						&ep->dl_pipe_id);
 	if (status) {
-		ath10k_warn(ar, "unsupported HTC service id: %d\n",
-			    ep->service_id);
+		ath10k_dbg(ar, ATH10K_DBG_BOOT, "unsupported HTC service id: %d\n",
+			   ep->service_id);
 		return status;
 	}
 
@@ -878,8 +878,8 @@ static bool ath10k_htc_pktlog_svc_supported(struct ath10k *ar)
 						&ul_pipe_id,
 						&dl_pipe_id);
 	if (status) {
-		ath10k_warn(ar, "unsupported HTC service id: %d\n",
-			    ATH10K_HTC_SVC_ID_HTT_LOG_MSG);
+		ath10k_dbg(ar, ATH10K_DBG_BOOT, "unsupported HTC pktlog service id: %d\n",
+			   ATH10K_HTC_SVC_ID_HTT_LOG_MSG);
 
 		return false;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 3cd9d35ee5f7a23203443f1a0d3c344f034c0dc7 Mon Sep 17 00:00:00 2001
From: Abhishek Pandit-Subedi <abhishekpandit@chromium.org>
Date: Wed, 27 Nov 2019 14:45:09 -0800
Subject: dt-bindings: net: bluetooth: Minor fix in broadcom-bluetooth

The example for brcm,bt-pcm-int-params should be a bytestring and all
values need to be two hex characters.

Signed-off-by: Abhishek Pandit-Subedi <abhishekpandit@chromium.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 Documentation/devicetree/bindings/net/broadcom-bluetooth.txt | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/net/broadcom-bluetooth.txt b/Documentation/devicetree/bindings/net/broadcom-bluetooth.txt
index b02a53275c98..b5eadee4a9a7 100644
--- a/Documentation/devicetree/bindings/net/broadcom-bluetooth.txt
+++ b/Documentation/devicetree/bindings/net/broadcom-bluetooth.txt
@@ -47,6 +47,6 @@ Example:
        bluetooth {
                compatible = "brcm,bcm43438-bt";
                max-speed = <921600>;
-               brcm,bt-pcm-int-params = [1 2 0 1 1];
+               brcm,bt-pcm-int-params = [01 02 00 01 01];
        };
 };
-- 
cgit v1.2.3-59-g8ed1b


From 22cc6b7a1dbb58da4afc539d9b7d470b23a25eea Mon Sep 17 00:00:00 2001
From: Johan Hovold <johan@kernel.org>
Date: Thu, 28 Nov 2019 19:24:27 +0100
Subject: Bluetooth: btusb: fix non-atomic allocation in completion handler

USB completion handlers are called in atomic context and must
specifically not allocate memory using GFP_KERNEL.

Fixes: a1c49c434e15 ("Bluetooth: btusb: Add protocol support for MediaTek MT7668U USB devices")
Cc: stable <stable@vger.kernel.org>     # 5.3
Cc: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 drivers/bluetooth/btusb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 82fb2e7b2892..9b587e662b48 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -2602,7 +2602,7 @@ static void btusb_mtk_wmt_recv(struct urb *urb)
 		 * and being processed the events from there then.
 		 */
 		if (test_bit(BTUSB_TX_WAIT_VND_EVT, &data->flags)) {
-			data->evt_skb = skb_clone(skb, GFP_KERNEL);
+			data->evt_skb = skb_clone(skb, GFP_ATOMIC);
 			if (!data->evt_skb)
 				goto err_out;
 		}
-- 
cgit v1.2.3-59-g8ed1b


From 86d4def890fdc25cee378c346dc31ca2ee0ef713 Mon Sep 17 00:00:00 2001
From: Ganesh Sesetti <gseset@codeaurora.org>
Date: Wed, 27 Nov 2019 18:29:45 +0200
Subject: ath11k: Fix htt stats sounding info and pdev cca stats

The Previous configuartion of htt stats sounding info and pdev cca stats
are invalid due to that getting time out error.

Changing htt stats sounding info value from 0xFF to 0x00 and htt pdev cca
stats from 0x10 to 0x00

Signed-off-by: Ganesh Sesetti <gseset@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/dp.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp.h b/drivers/net/wireless/ath/ath11k/dp.h
index 4db71173db62..eb0f3df131c5 100644
--- a/drivers/net/wireless/ath/ath11k/dp.h
+++ b/drivers/net/wireless/ath/ath11k/dp.h
@@ -1365,8 +1365,8 @@ struct htt_ext_stats_cfg_cmd {
 #define HTT_STAT_DEFAULT_CFG0_ALL_CMDQS 0xffff
 #define HTT_STAT_DEFAULT_CFG0_ALL_RINGS 0xffff
 #define HTT_STAT_DEFAULT_CFG0_ACTIVE_PEERS 0xff
-#define HTT_STAT_DEFAULT_CFG0_CCA_CUMULATIVE 0x10
-#define HTT_STAT_DEFAULT_CFG0_ACTIVE_VDEVS 0xff
+#define HTT_STAT_DEFAULT_CFG0_CCA_CUMULATIVE 0x00
+#define HTT_STAT_DEFAULT_CFG0_ACTIVE_VDEVS 0x00
 
 /* HTT_DBG_EXT_STATS_PEER_INFO
  * PARAMS:
-- 
cgit v1.2.3-59-g8ed1b


From 5e97128759b49b32740e1486007f239f2d7a9b8a Mon Sep 17 00:00:00 2001
From: Karthikeyan Periyasamy <periyasa@codeaurora.org>
Date: Wed, 27 Nov 2019 18:29:47 +0200
Subject: ath11k: fix resource leak in ath11k_mac_sta_state

Handled the error case with proper resource cleanup and
moved the handling into a separate function from
ath11k_mac_sta_state.

Signed-off-by: Karthikeyan Periyasamy <periyasa@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/mac.c | 144 ++++++++++++++++++++--------------
 1 file changed, 86 insertions(+), 58 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index c25817c51d87..6f82fdbbd358 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -2682,6 +2682,87 @@ static void ath11k_mac_dec_num_stations(struct ath11k_vif *arvif,
 	ar->num_stations--;
 }
 
+static int ath11k_mac_station_add(struct ath11k *ar,
+				  struct ieee80211_vif *vif,
+				  struct ieee80211_sta *sta)
+{
+	struct ath11k_base *ab = ar->ab;
+	struct ath11k_vif *arvif = ath11k_vif_to_arvif(vif);
+	struct ath11k_sta *arsta = (struct ath11k_sta *)sta->drv_priv;
+	struct peer_create_params peer_param;
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	ret = ath11k_mac_inc_num_stations(arvif, sta);
+	if (ret) {
+		ath11k_warn(ab, "refusing to associate station: too many connected already (%d)\n",
+			    ar->max_num_stations);
+		goto exit;
+	}
+
+	arsta->rx_stats = kzalloc(sizeof(*arsta->rx_stats), GFP_KERNEL);
+	if (!arsta->rx_stats) {
+		ret = -ENOMEM;
+		goto dec_num_station;
+	}
+
+	peer_param.vdev_id = arvif->vdev_id;
+	peer_param.peer_addr = sta->addr;
+	peer_param.peer_type = WMI_PEER_TYPE_DEFAULT;
+
+	ret = ath11k_peer_create(ar, arvif, sta, &peer_param);
+	if (ret) {
+		ath11k_warn(ab, "Failed to add peer: %pM for VDEV: %d\n",
+			    sta->addr, arvif->vdev_id);
+		goto free_rx_stats;
+	}
+
+	ath11k_dbg(ab, ATH11K_DBG_MAC, "Added peer: %pM for VDEV: %d\n",
+		   sta->addr, arvif->vdev_id);
+
+	if (ath11k_debug_is_extd_tx_stats_enabled(ar)) {
+		arsta->tx_stats = kzalloc(sizeof(*arsta->tx_stats), GFP_KERNEL);
+		if (!arsta->tx_stats) {
+			ret = -ENOMEM;
+			goto free_peer;
+		}
+	}
+
+	if (ieee80211_vif_is_mesh(vif)) {
+		ret = ath11k_wmi_set_peer_param(ar, sta->addr,
+						arvif->vdev_id,
+						WMI_PEER_USE_4ADDR, 1);
+		if (ret) {
+			ath11k_warn(ab, "failed to STA %pM 4addr capability: %d\n",
+				    sta->addr, ret);
+			goto free_tx_stats;
+		}
+	}
+
+	ret = ath11k_dp_peer_setup(ar, arvif->vdev_id, sta->addr);
+	if (ret) {
+		ath11k_warn(ab, "failed to setup dp for peer %pM on vdev %i (%d)\n",
+			    sta->addr, arvif->vdev_id, ret);
+		goto free_tx_stats;
+	}
+
+	return 0;
+
+free_tx_stats:
+	kfree(arsta->tx_stats);
+	arsta->tx_stats = NULL;
+free_peer:
+	ath11k_peer_delete(ar, arvif->vdev_id, sta->addr);
+free_rx_stats:
+	kfree(arsta->rx_stats);
+	arsta->rx_stats = NULL;
+dec_num_station:
+	ath11k_mac_dec_num_stations(arvif, sta);
+exit:
+	return ret;
+}
+
 static int ath11k_mac_op_sta_state(struct ieee80211_hw *hw,
 				   struct ieee80211_vif *vif,
 				   struct ieee80211_sta *sta,
@@ -2691,7 +2772,6 @@ static int ath11k_mac_op_sta_state(struct ieee80211_hw *hw,
 	struct ath11k *ar = hw->priv;
 	struct ath11k_vif *arvif = ath11k_vif_to_arvif(vif);
 	struct ath11k_sta *arsta = (struct ath11k_sta *)sta->drv_priv;
-	struct peer_create_params peer_param;
 	int ret = 0;
 
 	/* cancel must be done outside the mutex to avoid deadlock */
@@ -2707,60 +2787,10 @@ static int ath11k_mac_op_sta_state(struct ieee80211_hw *hw,
 		arsta->arvif = arvif;
 		INIT_WORK(&arsta->update_wk, ath11k_sta_rc_update_wk);
 
-		ret = ath11k_mac_inc_num_stations(arvif, sta);
-		if (ret) {
-			ath11k_warn(ar->ab, "refusing to associate station: too many connected already (%d)\n",
-				    ar->max_num_stations);
-			goto exit;
-		}
-
-		arsta->rx_stats = kzalloc(sizeof(*arsta->rx_stats), GFP_KERNEL);
-		if (!arsta->rx_stats) {
-			ret = -ENOMEM;
-			goto exit;
-		}
-
-		peer_param.vdev_id = arvif->vdev_id;
-		peer_param.peer_addr = sta->addr;
-		peer_param.peer_type = WMI_PEER_TYPE_DEFAULT;
-		ret = ath11k_peer_create(ar, arvif, sta, &peer_param);
-		if (ret) {
-			ath11k_warn(ar->ab, "Failed to add peer: %pM for VDEV: %d\n",
+		ret = ath11k_mac_station_add(ar, vif, sta);
+		if (ret)
+			ath11k_warn(ar->ab, "Failed to add station: %pM for VDEV: %d\n",
 				    sta->addr, arvif->vdev_id);
-			ath11k_mac_dec_num_stations(arvif, sta);
-			goto exit;
-		}
-
-		ath11k_info(ar->ab, "Added peer: %pM for VDEV: %d\n",
-			    sta->addr, arvif->vdev_id);
-
-		if (ath11k_debug_is_extd_tx_stats_enabled(ar)) {
-			arsta->tx_stats = kzalloc(sizeof(*arsta->tx_stats),
-						  GFP_KERNEL);
-			if (!arsta->tx_stats) {
-				ret = -ENOMEM;
-				goto exit;
-			}
-		}
-
-		if (ieee80211_vif_is_mesh(vif)) {
-			ret = ath11k_wmi_set_peer_param(ar, sta->addr,
-							arvif->vdev_id,
-							WMI_PEER_USE_4ADDR, 1);
-			if (ret) {
-				ath11k_warn(ar->ab, "failed to STA %pM 4addr capability: %d\n",
-					    sta->addr, ret);
-				goto exit;
-			}
-		}
-
-		ret = ath11k_dp_peer_setup(ar, arvif->vdev_id, sta->addr);
-		if (ret) {
-			ath11k_warn(ar->ab, "failed to setup dp for peer %pM on vdev %i (%d)\n",
-				    sta->addr, arvif->vdev_id, ret);
-			ath11k_peer_delete(ar, arvif->vdev_id, sta->addr);
-			ath11k_mac_dec_num_stations(arvif, sta);
-		}
 	} else if ((old_state == IEEE80211_STA_NONE &&
 		    new_state == IEEE80211_STA_NOTEXIST)) {
 		ath11k_dp_peer_cleanup(ar, arvif->vdev_id, sta->addr);
@@ -2770,9 +2800,8 @@ static int ath11k_mac_op_sta_state(struct ieee80211_hw *hw,
 			ath11k_warn(ar->ab, "Failed to delete peer: %pM for VDEV: %d\n",
 				    sta->addr, arvif->vdev_id);
 		else
-			ath11k_info(ar->ab,
-				    "Removed peer: %pM for VDEV: %d\n",
-				    sta->addr, arvif->vdev_id);
+			ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "Removed peer: %pM for VDEV: %d\n",
+				   sta->addr, arvif->vdev_id);
 
 		ath11k_mac_dec_num_stations(arvif, sta);
 
@@ -2809,7 +2838,6 @@ static int ath11k_mac_op_sta_state(struct ieee80211_hw *hw,
 				    sta->addr);
 	}
 
-exit:
 	mutex_unlock(&ar->conf_mutex);
 	return ret;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 5b90fc760db5a969ed26d70f8e62c91915f012bd Mon Sep 17 00:00:00 2001
From: Anilkumar Kolli <akolli@codeaurora.org>
Date: Wed, 27 Nov 2019 18:29:50 +0200
Subject: ath11k: fix wmi service ready ext tlv parsing

The current ath11k driver failed to parse
wmi_tlv_svc_rdy_ext_parse if there is change in
wmi_mac_phy_capabilities length with below error.

ath11k c000000.wifi1: failed to extract mac caps, idx :0
ath11k c000000.wifi1: failed to parse tlv -22

This is needed to get firmware version
WLAN.HK.2.0.0.1-00240-QCAHKSWPL_SILICONZ-1 working.

Signed-off-by: Anilkumar Kolli <akolli@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/wmi.c | 14 +++++++++++---
 drivers/net/wireless/ath/ath11k/wmi.h |  1 -
 2 files changed, 11 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/wmi.c b/drivers/net/wireless/ath/ath11k/wmi.c
index 2c3c973f5f7c..ccebf7ce420a 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.c
+++ b/drivers/net/wireless/ath/ath11k/wmi.c
@@ -310,7 +310,7 @@ ath11k_pull_mac_phy_cap_svc_ready_ext(struct ath11k_pdev_wmi *wmi_handle,
 	if (phy_id >= hal_reg_caps->num_phy)
 		return -EINVAL;
 
-	mac_phy_caps = &wmi_mac_phy_caps[phy_idx];
+	mac_phy_caps = wmi_mac_phy_caps + phy_idx;
 
 	pdev->pdev_id = mac_phy_caps->pdev_id;
 	pdev_cap->supported_bands = mac_phy_caps->supported_bands;
@@ -2959,6 +2959,15 @@ static int ath11k_wmi_tlv_mac_phy_caps_parse(struct ath11k_base *soc,
 	if (svc_rdy_ext->n_mac_phy_caps >= svc_rdy_ext->tot_phy_id)
 		return -ENOBUFS;
 
+	len = min_t(u16, len, sizeof(struct wmi_mac_phy_capabilities));
+	if (!svc_rdy_ext->n_mac_phy_caps) {
+		svc_rdy_ext->mac_phy_caps = kzalloc((svc_rdy_ext->tot_phy_id) * len,
+						    GFP_ATOMIC);
+		if (!svc_rdy_ext->mac_phy_caps)
+			return -ENOMEM;
+	}
+
+	memcpy(svc_rdy_ext->mac_phy_caps + svc_rdy_ext->n_mac_phy_caps, ptr, len);
 	svc_rdy_ext->n_mac_phy_caps++;
 	return 0;
 }
@@ -3092,8 +3101,6 @@ static int ath11k_wmi_tlv_svc_rdy_ext_parse(struct ath11k_base *ab,
 			svc_rdy_ext->hw_mode_done = true;
 		} else if (!svc_rdy_ext->mac_phy_done) {
 			svc_rdy_ext->n_mac_phy_caps = 0;
-			svc_rdy_ext->mac_phy_caps =
-					(struct wmi_mac_phy_capabilities *)ptr;
 			ret = ath11k_wmi_tlv_iter(ab, ptr, len,
 						  ath11k_wmi_tlv_mac_phy_caps_parse,
 						  svc_rdy_ext);
@@ -3134,6 +3141,7 @@ static int ath11k_service_ready_ext_event(struct ath11k_base *ab,
 		return ret;
 	}
 
+	kfree(svc_rdy_ext.mac_phy_caps);
 	return 0;
 }
 
diff --git a/drivers/net/wireless/ath/ath11k/wmi.h b/drivers/net/wireless/ath/ath11k/wmi.h
index 7d5690c65279..6a0468094694 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.h
+++ b/drivers/net/wireless/ath/ath11k/wmi.h
@@ -2276,7 +2276,6 @@ struct wmi_hw_mode_capabilities {
 #define WMI_MAX_HECAP_PHY_SIZE                 (3)
 
 struct wmi_mac_phy_capabilities {
-	u32 tlv_header;
 	u32 hw_mode_id;
 	u32 pdev_id;
 	u32 phy_id;
-- 
cgit v1.2.3-59-g8ed1b


From 0f37fbf43c3fc39a6c49c5a27846df937d217356 Mon Sep 17 00:00:00 2001
From: Anilkumar Kolli <akolli@codeaurora.org>
Date: Wed, 27 Nov 2019 18:29:53 +0200
Subject: ath11k: update tcl cmd descriptor parameters for STA mode

It is observed that ath11k STA mode UL throughput is low.
This is due to packets delivered to FW from TCL instead of TQM.
TCL AST search fail causes packet delivered to FW, fix this by
properly configuring the TCL address search type and ast_hash.
STA UL throughput is improved 10times with 11AC AP.

Signed-off-by: Anilkumar Kolli <akolli@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/dp.c       | 9 ++++-----
 drivers/net/wireless/ath/ath11k/dp_rx.c    | 2 +-
 drivers/net/wireless/ath/ath11k/hal_desc.h | 5 ++---
 drivers/net/wireless/ath/ath11k/hal_tx.c   | 6 +++---
 drivers/net/wireless/ath/ath11k/hal_tx.h   | 4 ++--
 5 files changed, 12 insertions(+), 14 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp.c b/drivers/net/wireless/ath/ath11k/dp.c
index aa2b28189cda..ff510e821a29 100644
--- a/drivers/net/wireless/ath/ath11k/dp.c
+++ b/drivers/net/wireless/ath/ath11k/dp.c
@@ -758,13 +758,12 @@ int ath11k_dp_htt_connect(struct ath11k_dp *dp)
 
 static void ath11k_dp_update_vdev_search(struct ath11k_vif *arvif)
 {
-	/* Enable AddrY (SA based search) for STA mode. All other modes it
-	 * is going to be AddrX (DA based search). For STA mode, set search
-	 * type based on AST value.
-	 */
+	 /* For STA mode, enable address search index,
+	  * tcl uses ast_hash value in the descriptor.
+	  */
 	switch (arvif->vdev_type) {
 	case WMI_VDEV_TYPE_STA:
-		arvif->hal_addr_search_flags = HAL_TX_ADDRY_EN;
+		arvif->hal_addr_search_flags = HAL_TX_ADDRX_EN;
 		arvif->search_type = HAL_TX_ADDR_SEARCH_INDEX;
 		break;
 	case WMI_VDEV_TYPE_AP:
diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c b/drivers/net/wireless/ath/ath11k/dp_rx.c
index 7aade0314e61..396658be6f26 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
@@ -1312,7 +1312,7 @@ void ath11k_dp_htt_htc_t2h_msg_handler(struct ath11k_base *ab,
 		ath11k_dp_get_mac_addr(resp->peer_map_ev.mac_addr_l32,
 				       peer_mac_h16, mac_addr);
 		ast_hash = FIELD_GET(HTT_T2H_PEER_MAP_INFO2_AST_HASH_VAL,
-				     resp->peer_map_ev.info1);
+				     resp->peer_map_ev.info2);
 		ath11k_peer_map_event(ab, vdev_id, peer_id, mac_addr, ast_hash);
 		break;
 	case HTT_T2H_MSG_TYPE_PEER_UNMAP:
diff --git a/drivers/net/wireless/ath/ath11k/hal_desc.h b/drivers/net/wireless/ath/ath11k/hal_desc.h
index 35e191b77271..5e200380cca4 100644
--- a/drivers/net/wireless/ath/ath11k/hal_desc.h
+++ b/drivers/net/wireless/ath/ath11k/hal_desc.h
@@ -934,9 +934,8 @@ struct hal_reo_flush_cache {
 #define HAL_TCL_DATA_CMD_INFO0_ENCRYPT_TYPE	GENMASK(7, 4)
 #define HAL_TCL_DATA_CMD_INFO0_SRC_BUF_SWAP	BIT(8)
 #define HAL_TCL_DATA_CMD_INFO0_LNK_META_SWAP	BIT(9)
-#define HAL_TCL_DATA_CMD_INFO0_SEARCH_TYPE	BIT(12)
-#define HAL_TCL_DATA_CMD_INFO0_ADDRX_EN		BIT(14)
-#define HAL_TCL_DATA_CMD_INFO0_ADDRY_EN		BIT(15)
+#define HAL_TCL_DATA_CMD_INFO0_SEARCH_TYPE	GENMASK(13, 12)
+#define HAL_TCL_DATA_CMD_INFO0_ADDR_EN		GENMASK(15, 14)
 #define HAL_TCL_DATA_CMD_INFO0_CMD_NUM		GENMASK(31, 16)
 
 #define HAL_TCL_DATA_CMD_INFO1_DATA_LEN		GENMASK(15, 0)
diff --git a/drivers/net/wireless/ath/ath11k/hal_tx.c b/drivers/net/wireless/ath/ath11k/hal_tx.c
index bb27591023fe..cbe549798762 100644
--- a/drivers/net/wireless/ath/ath11k/hal_tx.c
+++ b/drivers/net/wireless/ath/ath11k/hal_tx.c
@@ -52,9 +52,9 @@ void ath11k_hal_tx_cmd_desc_setup(struct ath11k_base *ab, void *cmd,
 		FIELD_PREP(HAL_TCL_DATA_CMD_INFO0_ENCAP_TYPE, ti->encap_type) |
 		FIELD_PREP(HAL_TCL_DATA_CMD_INFO0_ENCRYPT_TYPE,
 			   ti->encrypt_type) |
-		FIELD_PREP(HAL_TCL_DATA_CMD_INFO0_ADDRX_EN,
-			   ti->addr_search_flags) |
-		FIELD_PREP(HAL_TCL_DATA_CMD_INFO0_ADDRY_EN,
+		FIELD_PREP(HAL_TCL_DATA_CMD_INFO0_SEARCH_TYPE,
+			   ti->search_type) |
+		FIELD_PREP(HAL_TCL_DATA_CMD_INFO0_ADDR_EN,
 			   ti->addr_search_flags) |
 		FIELD_PREP(HAL_TCL_DATA_CMD_INFO0_CMD_NUM,
 			   ti->meta_data_flags);
diff --git a/drivers/net/wireless/ath/ath11k/hal_tx.h b/drivers/net/wireless/ath/ath11k/hal_tx.h
index f955070bcf1e..5217eaf9da50 100644
--- a/drivers/net/wireless/ath/ath11k/hal_tx.h
+++ b/drivers/net/wireless/ath/ath11k/hal_tx.h
@@ -11,8 +11,8 @@
 #define HAL_TX_ADDRX_EN			1
 #define HAL_TX_ADDRY_EN			2
 
-#define HAL_TX_ADDR_SEARCH_INDEX	0
-#define HAL_TX_ADDR_SEARCH_DEFAULT	1
+#define HAL_TX_ADDR_SEARCH_DEFAULT	0
+#define HAL_TX_ADDR_SEARCH_INDEX	1
 
 struct hal_tx_info {
 	u16 meta_data_flags; /* %HAL_TCL_DATA_CMD_INFO0_META_ */
-- 
cgit v1.2.3-59-g8ed1b


From 293cb5839729b186f951a82289aa5e0257c6d1b8 Mon Sep 17 00:00:00 2001
From: John Crispin <john@phrozen.org>
Date: Wed, 27 Nov 2019 18:29:56 +0200
Subject: ath11k: optimize RX path latency

This patch drops ath11k_hal_rx_parse_dst_ring_desc(). This function was
creating a huge amount of load, which lead to a signifcant latency delay
when processing data in the RX path.

Pegging the processing on a specific core and running perf --top we get
the following output when running HE80 at a fixed bandwidth of 1gbit.

with patch
    19.19%  [ath11k]       [k] ath11k_dp_process_rx
     5.02%  [ath11k]       [k] ath11k_dp_rx_tid_del_func
     4.39%  [kernel]       [k] v7_dma_inv_range
     4.15%  [kernel]       [k] __slab_alloc.constprop.1
     4.03%  [kernel]       [k] dev_gro_receive
     3.86%  [kernel]       [k] tcp_gro_receive
     3.07%  [ip_tables]    [k] ipt_do_table
     2.96%  [kernel]       [k] dma_cache_maint_page

without patch
    21.64%  [ath11k]       [k] ath11k_hal_rx_parse_dst_ring_desc
    10.80%  [ath11k]       [k] ath11k_dp_process_rx
     3.77%  [kernel]       [k] v7_dma_inv_range
     3.48%  [kernel]       [k] dev_gro_receive
     3.32%  [ath11k]       [k] ath11k_dp_rx_tid_del_func
     3.17%  [mac80211]     [k] ieee80211_rx_napi
     2.70%  [kernel]       [k] dma_cache_maint_page
     2.65%  [mac80211]     [k] ieee80211_sta_ps_transition

When removing the the bandwidth limit and rerunning the test we see an
overall throughput improvement of 3-400mbit when running 4x4 HE80.

Signed-off-by: Shashidhar Lakkavalli <slakkavalli@datto.com>
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/dp_rx.c  | 47 ++++++++++++++++--------------
 drivers/net/wireless/ath/ath11k/hal_rx.c | 49 +++-----------------------------
 drivers/net/wireless/ath/ath11k/hal_rx.h | 28 +-----------------
 3 files changed, 31 insertions(+), 93 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c b/drivers/net/wireless/ath/ath11k/dp_rx.c
index 396658be6f26..321a2bb657e8 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
@@ -2124,7 +2124,6 @@ int ath11k_dp_process_rx(struct ath11k_base *ab, int mac_id,
 	struct ieee80211_rx_status *rx_status = &dp->rx_status;
 	struct dp_rxdma_ring *rx_ring = &dp->rx_refill_buf_ring;
 	struct hal_srng *srng;
-	struct hal_rx_meta_info meta_info;
 	struct sk_buff *msdu;
 	struct sk_buff_head msdu_list;
 	struct sk_buff_head amsdu_list;
@@ -2160,11 +2159,14 @@ int ath11k_dp_process_rx(struct ath11k_base *ab, int mac_id,
 
 try_again:
 	while ((rx_desc = ath11k_hal_srng_dst_get_next_entry(ab, srng))) {
-		memset(&meta_info, 0, sizeof(meta_info));
-		ath11k_hal_rx_parse_dst_ring_desc(ab, rx_desc, &meta_info);
+		struct hal_reo_dest_ring *desc = (struct hal_reo_dest_ring *)rx_desc;
+		enum hal_reo_dest_ring_push_reason push_reason;
+		u32 cookie;
 
+		cookie = FIELD_GET(BUFFER_ADDR_INFO1_SW_COOKIE,
+				   desc->buf_addr_info.info1);
 		buf_id = FIELD_GET(DP_RXDMA_BUF_COOKIE_BUF_ID,
-				   meta_info.msdu_meta.cookie);
+				   cookie);
 		spin_lock_bh(&rx_ring->idr_lock);
 		msdu = idr_find(&rx_ring->bufs_idr, buf_id);
 		if (!msdu) {
@@ -2184,7 +2186,9 @@ try_again:
 
 		num_buffs_reaped++;
 
-		if (meta_info.push_reason !=
+		push_reason = FIELD_GET(HAL_REO_DEST_RING_INFO0_PUSH_REASON,
+					desc->info0);
+		if (push_reason !=
 		    HAL_REO_DEST_RING_PUSH_REASON_ROUTING_INSTRUCTION) {
 			/* TODO: Check if the msdu can be sent up for processing */
 			dev_kfree_skb_any(msdu);
@@ -2192,9 +2196,12 @@ try_again:
 			continue;
 		}
 
-		rxcb->is_first_msdu = meta_info.msdu_meta.first;
-		rxcb->is_last_msdu = meta_info.msdu_meta.last;
-		rxcb->is_continuation = meta_info.msdu_meta.continuation;
+		rxcb->is_first_msdu = !!(desc->rx_msdu_info.info0 &
+					 RX_MSDU_DESC_INFO0_FIRST_MSDU_IN_MPDU);
+		rxcb->is_last_msdu = !!(desc->rx_msdu_info.info0 &
+					RX_MSDU_DESC_INFO0_LAST_MSDU_IN_MPDU);
+		rxcb->is_continuation = !!(desc->rx_msdu_info.info0 &
+					   RX_MSDU_DESC_INFO0_MSDU_CONTINUATION);
 		rxcb->mac_id = mac_id;
 		__skb_queue_tail(&msdu_list, msdu);
 
@@ -2770,12 +2777,11 @@ exit:
 int ath11k_dp_process_rx_err(struct ath11k_base *ab, struct napi_struct *napi,
 			     int budget)
 {
-	struct hal_rx_msdu_meta meta[HAL_NUM_RX_MSDUS_PER_LINK_DESC];
+	u32 msdu_cookies[HAL_NUM_RX_MSDUS_PER_LINK_DESC];
 	struct dp_link_desc_bank *link_desc_banks;
 	enum hal_rx_buf_return_buf_manager rbm;
 	int tot_n_bufs_reaped, quota, ret, i;
 	int n_bufs_reaped[MAX_RADIOS] = {0};
-	struct hal_rx_meta_info meta_info;
 	struct dp_rxdma_ring *rx_ring;
 	struct dp_srng *reo_except;
 	u32 desc_bank, num_msdus;
@@ -2803,6 +2809,8 @@ int ath11k_dp_process_rx_err(struct ath11k_base *ab, struct napi_struct *napi,
 
 	while (budget &&
 	       (desc = ath11k_hal_srng_dst_get_next_entry(ab, srng))) {
+		struct hal_reo_dest_ring *reo_desc = (struct hal_reo_dest_ring *)desc;
+
 		ab->soc_stats.err_ring_pkts++;
 		ret = ath11k_hal_desc_reo_parse_err(ab, desc, &paddr,
 						    &desc_bank);
@@ -2813,7 +2821,7 @@ int ath11k_dp_process_rx_err(struct ath11k_base *ab, struct napi_struct *napi,
 		}
 		link_desc_va = link_desc_banks[desc_bank].vaddr +
 			       (paddr - link_desc_banks[desc_bank].paddr);
-		ath11k_hal_rx_msdu_link_info_get(link_desc_va, &num_msdus, meta,
+		ath11k_hal_rx_msdu_link_info_get(link_desc_va, &num_msdus, msdu_cookies,
 						 &rbm);
 		if (rbm != HAL_RX_BUF_RBM_WBM_IDLE_DESC_LIST &&
 		    rbm != HAL_RX_BUF_RBM_SW3_BM) {
@@ -2824,10 +2832,7 @@ int ath11k_dp_process_rx_err(struct ath11k_base *ab, struct napi_struct *napi,
 			continue;
 		}
 
-		memset(&meta_info, 0, sizeof(meta_info));
-		ath11k_hal_rx_parse_dst_ring_desc(ab, desc, &meta_info);
-
-		is_frag = meta_info.mpdu_meta.frag;
+		is_frag = !!(reo_desc->rx_mpdu_info.info0 & RX_MPDU_DESC_INFO0_FRAG_FLAG);
 
 		/* Return the link desc back to wbm idle list */
 		ath11k_dp_rx_link_desc_return(ab, desc,
@@ -2835,10 +2840,10 @@ int ath11k_dp_process_rx_err(struct ath11k_base *ab, struct napi_struct *napi,
 
 		for (i = 0; i < num_msdus; i++) {
 			buf_id = FIELD_GET(DP_RXDMA_BUF_COOKIE_BUF_ID,
-					   meta[i].cookie);
+					   msdu_cookies[i]);
 
 			mac_id = FIELD_GET(DP_RXDMA_BUF_COOKIE_PDEV_ID,
-					   meta[i].cookie);
+					   msdu_cookies[i]);
 
 			ar = ab->pdevs[mac_id].ar;
 
@@ -3192,7 +3197,7 @@ int ath11k_dp_process_rxdma_err(struct ath11k_base *ab, int mac_id, int budget)
 	struct dp_rxdma_ring *rx_ring = &ar->dp.rx_refill_buf_ring;
 	struct dp_link_desc_bank *link_desc_banks = ab->dp.link_desc_banks;
 	struct hal_srng *srng;
-	struct hal_rx_msdu_meta meta[HAL_NUM_RX_MSDUS_PER_LINK_DESC];
+	u32 msdu_cookies[HAL_NUM_RX_MSDUS_PER_LINK_DESC];
 	enum hal_rx_buf_return_buf_manager rbm;
 	enum hal_reo_entr_rxdma_ecode rxdma_err_code;
 	struct ath11k_skb_rxcb *rxcb;
@@ -3226,12 +3231,12 @@ int ath11k_dp_process_rxdma_err(struct ath11k_base *ab, int mac_id, int budget)
 
 		link_desc_va = link_desc_banks[desc_bank].vaddr +
 			       (paddr - link_desc_banks[desc_bank].paddr);
-		ath11k_hal_rx_msdu_link_info_get(link_desc_va, &num_msdus, meta,
-						 &rbm);
+		ath11k_hal_rx_msdu_link_info_get(link_desc_va, &num_msdus,
+						 msdu_cookies, &rbm);
 
 		for (i = 0; i < num_msdus; i++) {
 			buf_id = FIELD_GET(DP_RXDMA_BUF_COOKIE_BUF_ID,
-					   meta[i].cookie);
+					   msdu_cookies[i]);
 
 			spin_lock_bh(&rx_ring->idr_lock);
 			skb = idr_find(&rx_ring->bufs_idr, buf_id);
diff --git a/drivers/net/wireless/ath/ath11k/hal_rx.c b/drivers/net/wireless/ath/ath11k/hal_rx.c
index 2de4b388db20..6168a4eb0c0b 100644
--- a/drivers/net/wireless/ath/ath11k/hal_rx.c
+++ b/drivers/net/wireless/ath/ath11k/hal_rx.c
@@ -290,7 +290,7 @@ void ath11k_hal_rx_buf_addr_info_get(void *desc, dma_addr_t *paddr,
 }
 
 void ath11k_hal_rx_msdu_link_info_get(void *link_desc, u32 *num_msdus,
-				      struct hal_rx_msdu_meta *meta,
+				      u32 *msdu_cookies,
 				      enum hal_rx_buf_return_buf_manager *rbm)
 {
 	struct hal_rx_msdu_link *link = (struct hal_rx_msdu_link *)link_desc;
@@ -311,17 +311,9 @@ void ath11k_hal_rx_msdu_link_info_get(void *link_desc, u32 *num_msdus,
 			*num_msdus = i;
 			break;
 		}
-		meta->msdu_len = FIELD_GET(RX_MSDU_DESC_INFO0_MSDU_LENGTH,
-					   msdu->rx_msdu_info.info0);
-		meta->first = !!(msdu->rx_msdu_info.info0 &
-				 RX_MSDU_DESC_INFO0_FIRST_MSDU_IN_MPDU);
-		meta->last = !!(msdu->rx_msdu_info.info0 &
-				RX_MSDU_DESC_INFO0_LAST_MSDU_IN_MPDU);
-		meta->continuation = !!(msdu->rx_msdu_info.info0 &
-					RX_MSDU_DESC_INFO0_MSDU_CONTINUATION);
-		meta->cookie = FIELD_GET(BUFFER_ADDR_INFO1_SW_COOKIE,
-					 msdu->buf_addr_info.info1);
-		meta++;
+		*msdu_cookies = FIELD_GET(BUFFER_ADDR_INFO1_SW_COOKIE,
+					  msdu->buf_addr_info.info1);
+		msdu_cookies++;
 	}
 }
 
@@ -356,39 +348,6 @@ int ath11k_hal_desc_reo_parse_err(struct ath11k_base *ab, u32 *rx_desc,
 	return 0;
 }
 
-void ath11k_hal_rx_parse_dst_ring_desc(struct ath11k_base *ab, u32 *rx_desc,
-				       struct hal_rx_meta_info *meta_info)
-{
-	struct hal_reo_dest_ring *desc = (struct hal_reo_dest_ring *)rx_desc;
-	struct rx_mpdu_desc *mpdu = &desc->rx_mpdu_info;
-	struct rx_msdu_desc *msdu = &desc->rx_msdu_info;
-	struct hal_rx_mpdu_meta *meta_mpdu = &meta_info->mpdu_meta;
-	struct hal_rx_msdu_meta *meta_msdu = &meta_info->msdu_meta;
-
-	meta_info->push_reason = FIELD_GET(HAL_REO_DEST_RING_INFO0_PUSH_REASON,
-					   desc->info0);
-
-	meta_mpdu->msdu_cnt = FIELD_GET(RX_MPDU_DESC_INFO0_MSDU_COUNT,
-					mpdu->info0);
-	meta_mpdu->seq_num = FIELD_GET(RX_MPDU_DESC_INFO0_SEQ_NUM, mpdu->info0);
-	meta_mpdu->frag = !!(mpdu->info0 & RX_MPDU_DESC_INFO0_FRAG_FLAG);
-	meta_mpdu->retry = !!(mpdu->info0 & RX_MPDU_DESC_INFO0_MPDU_RETRY);
-	meta_mpdu->ampdu = !!(mpdu->info0 & RX_MPDU_DESC_INFO0_AMPDU_FLAG);
-	meta_mpdu->raw = !!(mpdu->info0 & RX_MPDU_DESC_INFO0_RAW_MPDU);
-	meta_mpdu->peer_meta = mpdu->meta_data;
-
-	meta_msdu->cookie = FIELD_GET(BUFFER_ADDR_INFO1_SW_COOKIE,
-				      desc->buf_addr_info.info1);
-	meta_msdu->msdu_len = FIELD_GET(RX_MSDU_DESC_INFO0_MSDU_LENGTH,
-					msdu->info0);
-	meta_msdu->first =
-		!!(msdu->info0 & RX_MSDU_DESC_INFO0_FIRST_MSDU_IN_MPDU);
-	meta_msdu->last =
-		!!(msdu->info0 & RX_MSDU_DESC_INFO0_LAST_MSDU_IN_MPDU);
-	meta_msdu->continuation =
-		!!(msdu->info0 & RX_MSDU_DESC_INFO0_MSDU_CONTINUATION);
-}
-
 int ath11k_hal_wbm_desc_parse_err(struct ath11k_base *ab, void *desc,
 				  struct hal_rx_wbm_rel_info *rel_info)
 {
diff --git a/drivers/net/wireless/ath/ath11k/hal_rx.h b/drivers/net/wireless/ath/ath11k/hal_rx.h
index 96555e4f4ef3..2271bf56183b 100644
--- a/drivers/net/wireless/ath/ath11k/hal_rx.h
+++ b/drivers/net/wireless/ath/ath11k/hal_rx.h
@@ -6,30 +6,6 @@
 #ifndef ATH11K_HAL_RX_H
 #define ATH11K_HAL_RX_H
 
-struct hal_rx_mpdu_meta {
-	u32 peer_meta;
-	u16 msdu_cnt;
-	u16 seq_num;
-	bool frag;
-	bool retry;
-	bool ampdu;
-	bool raw;
-};
-
-struct hal_rx_msdu_meta {
-	u32 cookie;
-	u32 msdu_len;
-	bool first;
-	bool last;
-	bool continuation;
-};
-
-struct hal_rx_meta_info {
-	enum hal_reo_dest_ring_push_reason push_reason;
-	struct hal_rx_mpdu_meta mpdu_meta;
-	struct hal_rx_msdu_meta msdu_meta;
-};
-
 struct hal_rx_wbm_rel_info {
 	u32 cookie;
 	enum hal_wbm_rel_src_module err_rel_src;
@@ -320,7 +296,7 @@ void ath11k_hal_reo_update_rx_reo_queue_status(struct ath11k_base *ab,
 					       struct hal_reo_status *status);
 int ath11k_hal_reo_process_status(u8 *reo_desc, u8 *status);
 void ath11k_hal_rx_msdu_link_info_get(void *link_desc, u32 *num_msdus,
-				      struct hal_rx_msdu_meta *meta,
+				      u32 *msdu_cookies,
 				      enum hal_rx_buf_return_buf_manager *rbm);
 void ath11k_hal_rx_msdu_link_desc_set(struct ath11k_base *ab, void *desc,
 				      void *link_desc,
@@ -331,8 +307,6 @@ void ath11k_hal_rx_buf_addr_info_get(void *desc, dma_addr_t *paddr,
 				     u32 *cookie, u8 *rbm);
 int ath11k_hal_desc_reo_parse_err(struct ath11k_base *ab, u32 *rx_desc,
 				  dma_addr_t *paddr, u32 *desc_bank);
-void ath11k_hal_rx_parse_dst_ring_desc(struct ath11k_base *ab, u32 *rx_desc,
-				       struct hal_rx_meta_info *meta_info);
 int ath11k_hal_wbm_desc_parse_err(struct ath11k_base *ab, void *desc,
 				  struct hal_rx_wbm_rel_info *rel_info);
 void ath11k_hal_rx_reo_ent_paddr_get(struct ath11k_base *ab, void *desc,
-- 
cgit v1.2.3-59-g8ed1b


From f1d34a01ed547793c223eb3df4849bdb19ff1d22 Mon Sep 17 00:00:00 2001
From: Karthikeyan Periyasamy <periyasa@codeaurora.org>
Date: Wed, 27 Nov 2019 18:29:58 +0200
Subject: ath11k: avoid WMM param truncation

In conf_tx() mac operation callback, we are truncating the tx
params cw_min and cw_max due to lower data type cast. so modified
the data type of cwmin and cwmax to avoid the trucation issue.

Signed-off-by: Karthikeyan Periyasamy <periyasa@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/wmi.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/wmi.h b/drivers/net/wireless/ath/ath11k/wmi.h
index 6a0468094694..4a518d406bc5 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.h
+++ b/drivers/net/wireless/ath/ath11k/wmi.h
@@ -4501,8 +4501,8 @@ struct wmi_wmm_params {
 struct wmi_wmm_params_arg {
 	u8 acm;
 	u8 aifs;
-	u8 cwmin;
-	u8 cwmax;
+	u16 cwmin;
+	u16 cwmax;
 	u16 txop;
 	u8 no_ack;
 };
-- 
cgit v1.2.3-59-g8ed1b


From d12ac6c47a32b720d77181b8c2915be43c5be2d0 Mon Sep 17 00:00:00 2001
From: Pradeep Kumar Chitrapu <pradeepc@codeaurora.org>
Date: Wed, 27 Nov 2019 18:30:01 +0200
Subject: ath11k: remove unused tx ring counters

remove unused counter to avoid taking locks inorder to optimize
cpu load.

Using Flamegraph, cpu usage of ath11k_dp_tx() observed to be decreased
from 5.58% to 3.74% with iperf traffic running with 80MHz bandwidth ap
mode.

Signed-off-by: Pradeep Kumar Chitrapu <pradeepc@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/dp.h    | 1 -
 drivers/net/wireless/ath/ath11k/dp_tx.c | 7 -------
 2 files changed, 8 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp.h b/drivers/net/wireless/ath/ath11k/dp.h
index eb0f3df131c5..db216f055c42 100644
--- a/drivers/net/wireless/ath/ath11k/dp.h
+++ b/drivers/net/wireless/ath/ath11k/dp.h
@@ -63,7 +63,6 @@ struct dp_tx_ring {
 	struct dp_srng tcl_data_ring;
 	struct dp_srng tcl_comp_ring;
 	struct idr txbuf_idr;
-	u32 num_tx_pending;
 	/* Protects txbuf_idr and num_pending */
 	spinlock_t tx_idr_lock;
 	DECLARE_KFIFO_PTR(tx_status_fifo, struct hal_wbm_release_ring);
diff --git a/drivers/net/wireless/ath/ath11k/dp_tx.c b/drivers/net/wireless/ath/ath11k/dp_tx.c
index 51a02b8e66cc..a8b9557c2346 100644
--- a/drivers/net/wireless/ath/ath11k/dp_tx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_tx.c
@@ -196,10 +196,6 @@ int ath11k_dp_tx(struct ath11k *ar, struct ath11k_vif *arvif,
 
 	spin_unlock_bh(&tcl_ring->lock);
 
-	spin_lock_bh(&tx_ring->tx_idr_lock);
-	tx_ring->num_tx_pending++;
-	spin_unlock_bh(&tx_ring->tx_idr_lock);
-
 	atomic_inc(&ar->dp.num_tx_pending);
 
 	return 0;
@@ -236,7 +232,6 @@ static void ath11k_dp_tx_free_txbuf(struct ath11k_base *ab, u8 mac_id,
 	skb_cb = ATH11K_SKB_CB(msdu);
 
 	idr_remove(&tx_ring->txbuf_idr, msdu_id);
-	tx_ring->num_tx_pending--;
 	spin_unlock_bh(&tx_ring->tx_idr_lock);
 
 	dma_unmap_single(ab->dev, skb_cb->paddr, msdu->len, DMA_TO_DEVICE);
@@ -272,7 +267,6 @@ ath11k_dp_tx_htt_tx_complete_buf(struct ath11k_base *ab,
 	ar = skb_cb->ar;
 
 	idr_remove(&tx_ring->txbuf_idr, ts->msdu_id);
-	tx_ring->num_tx_pending--;
 	spin_unlock_bh(&tx_ring->tx_idr_lock);
 
 	if (atomic_dec_and_test(&ar->dp.num_tx_pending))
@@ -495,7 +489,6 @@ void ath11k_dp_tx_completion_handler(struct ath11k_base *ab, int ring_id)
 			continue;
 		}
 		idr_remove(&tx_ring->txbuf_idr, msdu_id);
-		tx_ring->num_tx_pending--;
 		spin_unlock_bh(&tx_ring->tx_idr_lock);
 
 		ar = ab->pdevs[mac_id].ar;
-- 
cgit v1.2.3-59-g8ed1b


From d2f510fa0103a789b7fd310dbf14e0effa2f1fdf Mon Sep 17 00:00:00 2001
From: Sriram R <srirrama@codeaurora.org>
Date: Wed, 27 Nov 2019 18:30:04 +0200
Subject: ath11k: Fix skb_panic observed during msdu coalescing

skb_panic is hit during msdu coalescing whenever
enough tailroom is not allocated based on the remaining
msdu length which is spread across in different rx buffers.

Compute the extra length for resizing the skb based on
the total msdu length and the msdu length of the first buffer.

Signed-off-by: Sriram R <srirrama@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/dp_rx.c | 22 ++++++++++++++--------
 1 file changed, 14 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c b/drivers/net/wireless/ath/ath11k/dp_rx.c
index 321a2bb657e8..acad74658e64 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
@@ -1344,15 +1344,22 @@ static int ath11k_dp_rx_msdu_coalesce(struct ath11k *ar,
 {
 	struct sk_buff *skb;
 	struct ath11k_skb_rxcb *rxcb = ATH11K_SKB_RXCB(first);
+	int buf_first_hdr_len, buf_first_len;
 	struct hal_rx_desc *ldesc;
 	int space_extra;
 	int rem_len;
 	int buf_len;
 
-	if (WARN_ON_ONCE(msdu_len <= (DP_RX_BUFFER_SIZE -
-			 (HAL_RX_DESC_SIZE + l3pad_bytes)))) {
-		skb_put(first, HAL_RX_DESC_SIZE + l3pad_bytes + msdu_len);
-		skb_pull(first, HAL_RX_DESC_SIZE + l3pad_bytes);
+	/* As the msdu is spread across multiple rx buffers,
+	 * find the offset to the start of msdu for computing
+	 * the length of the msdu in the first buffer.
+	 */
+	buf_first_hdr_len = HAL_RX_DESC_SIZE + l3pad_bytes;
+	buf_first_len = DP_RX_BUFFER_SIZE - buf_first_hdr_len;
+
+	if (WARN_ON_ONCE(msdu_len <= buf_first_len)) {
+		skb_put(first, buf_first_hdr_len + msdu_len);
+		skb_pull(first, buf_first_hdr_len);
 		return 0;
 	}
 
@@ -1365,9 +1372,9 @@ static int ath11k_dp_rx_msdu_coalesce(struct ath11k *ar,
 	 * in the first buf is of length DP_RX_BUFFER_SIZE - HAL_RX_DESC_SIZE.
 	 */
 	skb_put(first, DP_RX_BUFFER_SIZE);
-	skb_pull(first, HAL_RX_DESC_SIZE + l3pad_bytes);
+	skb_pull(first, buf_first_hdr_len);
 
-	space_extra = msdu_len - (DP_RX_BUFFER_SIZE + skb_tailroom(first));
+	space_extra = msdu_len - (buf_first_len + skb_tailroom(first));
 	if (space_extra > 0 &&
 	    (pskb_expand_head(first, 0, space_extra, GFP_ATOMIC) < 0)) {
 		/* Free up all buffers of the MSDU */
@@ -1387,8 +1394,7 @@ static int ath11k_dp_rx_msdu_coalesce(struct ath11k *ar,
 	 */
 	ath11k_dp_rx_desc_end_tlv_copy(rxcb->rx_desc, ldesc);
 
-	rem_len = msdu_len -
-		  (DP_RX_BUFFER_SIZE - HAL_RX_DESC_SIZE - l3pad_bytes);
+	rem_len = msdu_len - buf_first_len;
 	while ((skb = __skb_dequeue(msdu_list)) != NULL && rem_len > 0) {
 		rxcb = ATH11K_SKB_RXCB(skb);
 		if (rxcb->is_continuation)
-- 
cgit v1.2.3-59-g8ed1b


From 79c647a3c59ddc1a1acf68e82e555201c10d3c18 Mon Sep 17 00:00:00 2001
From: Vikas Patel <vikpatel@codeaurora.org>
Date: Wed, 27 Nov 2019 18:30:07 +0200
Subject: ath11k: Fixing TLV length issue in peer pktlog WMI command

TLV length was 0 for TLV tag 'WMI_TAG_ARRAY_STRUCT' causing
Q6 to crash when trying to configure pktlog filter via debugfs.

Signed-off-by: Vikas Patel <vikpatel@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/wmi.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath11k/wmi.c b/drivers/net/wireless/ath/ath11k/wmi.c
index ccebf7ce420a..aae6e76330da 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.c
+++ b/drivers/net/wireless/ath/ath11k/wmi.c
@@ -2329,7 +2329,7 @@ int ath11k_wmi_pdev_peer_pktlog_filter(struct ath11k *ar, u8 *addr, u8 enable)
 
 	tlv = ptr;
 	tlv->header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_ARRAY_STRUCT) |
-		      FIELD_PREP(WMI_TLV_LEN, 0);
+		      FIELD_PREP(WMI_TLV_LEN, sizeof(*info));
 
 	ptr += TLV_HDR_SIZE;
 	info = ptr;
-- 
cgit v1.2.3-59-g8ed1b


From 051cefa4466744cb669d7eef9399cbf850b2a6d4 Mon Sep 17 00:00:00 2001
From: Vasanthakumar Thiagarajan <vthiagar@codeaurora.org>
Date: Wed, 27 Nov 2019 18:30:09 +0200
Subject: ath11k: Fix target crash due to WBM_IDLE_LINK ring desc shortage

Make sure the number of WBM_IDLE_LINK ring descriptors is power
of 2. This increases the number of descriptors to 32k from the
current ~18k to fix the target assert because of the shortage in
the descriptors in WBM_IDLE_LINK ring. Remove unnecessary
power of 2 calculation in ath11k_dp_link_desc_setup() as it is
not required after this change.

Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/dp.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp.c b/drivers/net/wireless/ath/ath11k/dp.c
index ff510e821a29..72c21cf6a352 100644
--- a/drivers/net/wireless/ath/ath11k/dp.c
+++ b/drivers/net/wireless/ath/ath11k/dp.c
@@ -478,6 +478,9 @@ static int ath11k_wbm_idle_ring_setup(struct ath11k_base *ab, u32 *n_link_desc)
 	*n_link_desc = n_mpdu_link_desc + n_mpdu_queue_desc +
 		      n_tx_msdu_link_desc + n_rx_msdu_link_desc;
 
+	if (*n_link_desc & (*n_link_desc - 1))
+		*n_link_desc = 1 << fls(*n_link_desc);
+
 	ret = ath11k_dp_srng_setup(ab, &dp->wbm_idle_ring,
 				   HAL_WBM_IDLE_LINK, 0, 0, *n_link_desc);
 	if (ret) {
@@ -499,9 +502,6 @@ int ath11k_dp_link_desc_setup(struct ath11k_base *ab,
 	u32 *desc;
 	int i, ret;
 
-	if (n_link_desc & (n_link_desc - 1))
-		n_link_desc = 1 << fls(n_link_desc);
-
 	tot_mem_sz = n_link_desc * HAL_LINK_DESC_SIZE;
 	tot_mem_sz += HAL_LINK_DESC_ALIGN;
 
-- 
cgit v1.2.3-59-g8ed1b


From 7cbf4c96d7159e4abe762f4bafa9911fc1f7d339 Mon Sep 17 00:00:00 2001
From: Wen Gong <wgong@codeaurora.org>
Date: Fri, 8 Nov 2019 17:19:14 +0800
Subject: ath10k: enable firmware log by default for sdio

On SDIO chips the firmware log does not impact performance. To make it
easier to debug firmware problems keep it enabled on the firmware.

Tested with QCA6174 SDIO with firmware WLAN.RMH.4.4.1-00029.

Signed-off-by: Wen Gong <wgong@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/core.c | 12 ------------
 1 file changed, 12 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c
index 0d345852c04a..6da377317d1f 100644
--- a/drivers/net/wireless/ath/ath10k/core.c
+++ b/drivers/net/wireless/ath/ath10k/core.c
@@ -715,18 +715,6 @@ static int ath10k_init_sdio(struct ath10k *ar, enum ath10k_firmware_mode mode)
 	if (ret)
 		return ret;
 
-	/* Explicitly set fwlog prints to zero as target may turn it on
-	 * based on scratch registers.
-	 */
-	ret = ath10k_bmi_read32(ar, hi_option_flag, &param);
-	if (ret)
-		return ret;
-
-	param |= HI_OPTION_DISABLE_DBGLOG;
-	ret = ath10k_bmi_write32(ar, hi_option_flag, param);
-	if (ret)
-		return ret;
-
 	return 0;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 376a30c7c9a05382d2cad09e9ff9b2998d0d53c3 Mon Sep 17 00:00:00 2001
From: Wen Gong <wgong@codeaurora.org>
Date: Wed, 27 Nov 2019 02:59:29 +0000
Subject: ath10k: set max mtu to 1500 for sdio chip

For sdio chip, the max credit size in firmware is 1556, the 1556
include payload, ieee80211 header, htt header, htc header. So it
need to set the max mtu to 1500 to forbidden TX packet which exceed
1500 form application.

Tested with QCA6174 SDIO with firmware
WLAN.RMH.4.4.1-00017-QCARMSWP-1.

Signed-off-by: Wen Gong <wgong@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/sdio.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/wireless/ath/ath10k/sdio.c b/drivers/net/wireless/ath/ath10k/sdio.c
index 3e63adfe3e9c..2e155bba2aa3 100644
--- a/drivers/net/wireless/ath/ath10k/sdio.c
+++ b/drivers/net/wireless/ath/ath10k/sdio.c
@@ -2137,6 +2137,8 @@ static int ath10k_sdio_probe(struct sdio_func *func,
 	bus_params.chip_id = 0;
 	bus_params.hl_msdu_ids = true;
 
+	ar->hw->max_mtu = ETH_DATA_LEN;
+
 	ret = ath10k_core_register(ar, &bus_params);
 	if (ret) {
 		ath10k_err(ar, "failed to register driver core: %d\n", ret);
-- 
cgit v1.2.3-59-g8ed1b


From 382e51c139ef9d0b8e65ff5c249771f864b411fd Mon Sep 17 00:00:00 2001
From: Wen Gong <wgong@codeaurora.org>
Date: Wed, 27 Nov 2019 03:19:24 +0000
Subject: ath10k: set WMI_PEER_AUTHORIZE after a firmware crash

After the firmware crashes ath10k recovers via ieee80211_reconfig(),
which eventually leads to firmware configuration and including the
encryption keys. However, because there is no new auth/assoc and
4-way-handshake, and firmware set the authorize flag after
4-way-handshake, so the authorize flag in firmware is not set in
firmware without 4-way-handshake. This will lead to a failure of data
transmission after recovery done when using encrypted connections like
WPA-PSK. Set authorize flag after installing keys to firmware will fix
the issue.

This was noticed by testing firmware crashing using simulate_fw_crash
debugfs file.

Tested with QCA6174 SDIO with firmware WLAN.RMH.4.4.1-00007-QCARMSWP-1.

Signed-off-by: Wen Gong <wgong@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/mac.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index 6c6eb9341e6d..767c7bf16975 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -6329,6 +6329,9 @@ static int ath10k_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
 	if (sta && sta->tdls)
 		ath10k_wmi_peer_set_param(ar, arvif->vdev_id, sta->addr,
 					  ar->wmi.peer_param->authorize, 1);
+	else if (sta && cmd == SET_KEY && (key->flags & IEEE80211_KEY_FLAG_PAIRWISE))
+		ath10k_wmi_peer_set_param(ar, arvif->vdev_id, peer_addr,
+					  ar->wmi.peer_param->authorize, 1);
 
 exit:
 	mutex_unlock(&ar->conf_mutex);
-- 
cgit v1.2.3-59-g8ed1b


From 5d2467299a17a2c13c2a1a608b72ec032f559914 Mon Sep 17 00:00:00 2001
From: Wen Gong <wgong@codeaurora.org>
Date: Wed, 27 Nov 2019 03:33:14 +0000
Subject: ath10k: change log level for mpdu status of sdio chip

Change log level from warn to dbg level of mpdu status
of sdio chip.

Tested with QCA6174 SDIO with firmware
WLAN.RMH.4.4.1-00007-QCARMSWP-1.

Signed-off-by: Wen Gong <wgong@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/htt_rx.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c
index 9faecfa5f9d7..bbcfa728917d 100644
--- a/drivers/net/wireless/ath/ath10k/htt_rx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_rx.c
@@ -2203,8 +2203,8 @@ static bool ath10k_htt_rx_proc_rx_ind_hl(struct ath10k_htt *htt,
 	    HTT_RX_IND_MPDU_STATUS_OK &&
 	    mpdu_ranges->mpdu_range_status !=
 	    HTT_RX_IND_MPDU_STATUS_TKIP_MIC_ERR) {
-		ath10k_warn(ar, "MPDU range status: %d\n",
-			    mpdu_ranges->mpdu_range_status);
+		ath10k_dbg(ar, ATH10K_DBG_HTT, "htt mpdu_range_status %d\n",
+			   mpdu_ranges->mpdu_range_status);
 		goto err;
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From 7321095ca39c8bb4da5e86f7a3fb8c22b47e5063 Mon Sep 17 00:00:00 2001
From: Wen Gong <wgong@codeaurora.org>
Date: Wed, 27 Nov 2019 06:04:13 +0000
Subject: ath10k: enable wow feature for sdio chip

sdio does not support wow, this patch is to enable it. When system enter
sleep state, if wowlan is enabled, then sdio chip will keep power if
platform support keep power, after resume, it will not need to re-load
firmware again.

Tested with QCA6174 SDIO with firmware
WLAN.RMH.4.4.1-00029.

Signed-off-by: Wen Gong <wgong@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/sdio.c | 23 +++++++++++++++++++++--
 1 file changed, 21 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/sdio.c b/drivers/net/wireless/ath/ath10k/sdio.c
index 2e155bba2aa3..b8e249909699 100644
--- a/drivers/net/wireless/ath/ath10k/sdio.c
+++ b/drivers/net/wireless/ath/ath10k/sdio.c
@@ -1859,7 +1859,7 @@ static void ath10k_sdio_hif_stop(struct ath10k *ar)
 
 static int ath10k_sdio_hif_suspend(struct ath10k *ar)
 {
-	return -EOPNOTSUPP;
+	return 0;
 }
 
 static int ath10k_sdio_hif_resume(struct ath10k *ar)
@@ -2009,7 +2009,26 @@ static const struct ath10k_hif_ops ath10k_sdio_hif_ops = {
  */
 static int ath10k_sdio_pm_suspend(struct device *device)
 {
-	return 0;
+	struct sdio_func *func = dev_to_sdio_func(device);
+	struct ath10k_sdio *ar_sdio = sdio_get_drvdata(func);
+	struct ath10k *ar = ar_sdio->ar;
+	mmc_pm_flag_t pm_flag, pm_caps;
+	int ret;
+
+	if (!device_may_wakeup(ar->dev))
+		return 0;
+
+	pm_flag = MMC_PM_KEEP_POWER;
+
+	ret = sdio_set_host_pm_flags(func, pm_flag);
+	if (ret) {
+		pm_caps = sdio_get_host_pm_caps(func);
+		ath10k_warn(ar, "failed to set sdio host pm flags (0x%x, 0x%x): %d\n",
+			    pm_flag, pm_caps, ret);
+		return ret;
+	}
+
+	return ret;
 }
 
 static int ath10k_sdio_pm_resume(struct device *device)
-- 
cgit v1.2.3-59-g8ed1b


From e8da3986570eff40e02bebcffd62fd3ccdf485db Mon Sep 17 00:00:00 2001
From: Anilkumar Kolli <akolli@codeaurora.org>
Date: Wed, 27 Nov 2019 14:08:45 +0000
Subject: ath11k: tracing: fix ath11k tracing

Add missing tracing subsystem define.

Signed-off-by: Anilkumar Kolli <akolli@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/trace.h | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/wireless/ath/ath11k/trace.h b/drivers/net/wireless/ath/ath11k/trace.h
index 400f759568e1..8700a622be7b 100644
--- a/drivers/net/wireless/ath/ath11k/trace.h
+++ b/drivers/net/wireless/ath/ath11k/trace.h
@@ -17,6 +17,9 @@
 static inline void trace_ ## name(proto) {}
 #endif /* !CONFIG_ATH11K_TRACING || __CHECKER__ */
 
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM ath11k
+
 TRACE_EVENT(ath11k_htt_pktlog,
 	    TP_PROTO(struct ath11k *ar, const void *buf, u16 buf_len),
 
-- 
cgit v1.2.3-59-g8ed1b


From d6af906d8307a804a1af488564def0264ea57282 Mon Sep 17 00:00:00 2001
From: Anilkumar Kolli <akolli@codeaurora.org>
Date: Wed, 27 Nov 2019 14:08:47 +0000
Subject: ath11k: qmi clean up ce and HTC service config update

Copy CE and htc service configs for all pipes.

Signed-off-by: Anilkumar Kolli <akolli@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/ahb.c | 9 ++++-----
 drivers/net/wireless/ath/ath11k/qmi.h | 4 ++--
 2 files changed, 6 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/ahb.c b/drivers/net/wireless/ath/ath11k/ahb.c
index f80173b8afc6..e7e3e64c07aa 100644
--- a/drivers/net/wireless/ath/ath11k/ahb.c
+++ b/drivers/net/wireless/ath/ath11k/ahb.c
@@ -635,11 +635,10 @@ static void ath11k_ahb_init_qmi_ce_config(struct ath11k_base *ab)
 {
 	struct ath11k_qmi_ce_cfg *cfg = &ab->qmi.ce_cfg;
 
-	cfg->tgt_ce = (u8 *)target_ce_config_wlan;
-	cfg->tgt_ce_len = sizeof(target_ce_config_wlan);
-
-	cfg->svc_to_ce_map = (u8 *)target_service_to_ce_map_wlan;
-	cfg->svc_to_ce_map_len = sizeof(target_service_to_ce_map_wlan);
+	cfg->tgt_ce_len = ARRAY_SIZE(target_ce_config_wlan) - 1;
+	cfg->tgt_ce = target_ce_config_wlan;
+	cfg->svc_to_ce_map_len = ARRAY_SIZE(target_service_to_ce_map_wlan);
+	cfg->svc_to_ce_map = target_service_to_ce_map_wlan;
 }
 
 static void ath11k_ahb_free_ext_irq(struct ath11k_base *ab)
diff --git a/drivers/net/wireless/ath/ath11k/qmi.h b/drivers/net/wireless/ath/ath11k/qmi.h
index df0685564e21..3f7db642d869 100644
--- a/drivers/net/wireless/ath/ath11k/qmi.h
+++ b/drivers/net/wireless/ath/ath11k/qmi.h
@@ -66,9 +66,9 @@ struct ath11k_qmi_driver_event {
 };
 
 struct ath11k_qmi_ce_cfg {
-	const u8 *tgt_ce;
+	const struct ce_pipe_config *tgt_ce;
 	int tgt_ce_len;
-	const u8 *svc_to_ce_map;
+	const struct service_to_pipe *svc_to_ce_map;
 	int svc_to_ce_map_len;
 	const u8 *shadow_reg;
 	int shadow_reg_len;
-- 
cgit v1.2.3-59-g8ed1b


From 0c408515cd3b6fb980ec31213c0e55bbbe74d975 Mon Sep 17 00:00:00 2001
From: Anilkumar Kolli <akolli@codeaurora.org>
Date: Wed, 27 Nov 2019 14:08:50 +0000
Subject: ath11k: qmi clean up in ath11k_qmi_wlanfw_wlan_cfg_send()

Use pipe_id id instead of ret in for loop.

Signed-off-by: Anilkumar Kolli <akolli@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/ce.h  |  9 +++++++--
 drivers/net/wireless/ath/ath11k/qmi.c | 28 +++++++++++++---------------
 2 files changed, 20 insertions(+), 17 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/ce.h b/drivers/net/wireless/ath/ath11k/ce.h
index 6174eac84410..e355dfda48bf 100644
--- a/drivers/net/wireless/ath/ath11k/ce.h
+++ b/drivers/net/wireless/ath/ath11k/ce.h
@@ -57,7 +57,12 @@ void ath11k_ce_byte_swap(void *mem, u32 len);
 
 struct ath11k_base;
 
-/* Establish a mapping between a service/direction and a pipe. */
+/*
+ * Establish a mapping between a service/direction and a pipe.
+ * Configuration information for a Copy Engine pipe and services.
+ * Passed from Host to Target through QMI message and must be in
+ * little endian format.
+ */
 struct service_to_pipe {
 	__le32 service_id;
 	__le32 pipedir;
@@ -66,7 +71,7 @@ struct service_to_pipe {
 
 /*
  * Configuration information for a Copy Engine pipe.
- * Passed from Host to Target during startup (one per CE).
+ * Passed from Host to Target through QMI message during startup (one per CE).
  *
  * NOTE: Structure is shared between Host software and Target firmware!
  */
diff --git a/drivers/net/wireless/ath/ath11k/qmi.c b/drivers/net/wireless/ath/ath11k/qmi.c
index adf41f28a231..2377895a58ec 100644
--- a/drivers/net/wireless/ath/ath11k/qmi.c
+++ b/drivers/net/wireless/ath/ath11k/qmi.c
@@ -2016,7 +2016,7 @@ static int ath11k_qmi_wlanfw_wlan_cfg_send(struct ath11k_base *ab)
 	struct ce_pipe_config *ce_cfg;
 	struct service_to_pipe *svc_cfg;
 	struct qmi_txn txn = {};
-	int ret = 0;
+	int ret = 0, pipe_num;
 
 	ce_cfg	= (struct ce_pipe_config *)ab->qmi.ce_cfg.tgt_ce;
 	svc_cfg	= (struct service_to_pipe *)ab->qmi.ce_cfg.svc_to_ce_map;
@@ -2033,24 +2033,22 @@ static int ath11k_qmi_wlanfw_wlan_cfg_send(struct ath11k_base *ab)
 
 	req->tgt_cfg_valid = 1;
 	/* This is number of CE configs */
-	req->tgt_cfg_len = ((ab->qmi.ce_cfg.tgt_ce_len) /
-				(sizeof(struct ce_pipe_config))) - 1;
-	for (ret = 0; ret <= req->tgt_cfg_len ; ret++) {
-		req->tgt_cfg[ret].pipe_num = ce_cfg[ret].pipenum;
-		req->tgt_cfg[ret].pipe_dir = ce_cfg[ret].pipedir;
-		req->tgt_cfg[ret].nentries = ce_cfg[ret].nentries;
-		req->tgt_cfg[ret].nbytes_max = ce_cfg[ret].nbytes_max;
-		req->tgt_cfg[ret].flags = ce_cfg[ret].flags;
+	req->tgt_cfg_len = ab->qmi.ce_cfg.tgt_ce_len;
+	for (pipe_num = 0; pipe_num <= req->tgt_cfg_len ; pipe_num++) {
+		req->tgt_cfg[pipe_num].pipe_num = ce_cfg[pipe_num].pipenum;
+		req->tgt_cfg[pipe_num].pipe_dir = ce_cfg[pipe_num].pipedir;
+		req->tgt_cfg[pipe_num].nentries = ce_cfg[pipe_num].nentries;
+		req->tgt_cfg[pipe_num].nbytes_max = ce_cfg[pipe_num].nbytes_max;
+		req->tgt_cfg[pipe_num].flags = ce_cfg[pipe_num].flags;
 	}
 
 	req->svc_cfg_valid = 1;
 	/* This is number of Service/CE configs */
-	req->svc_cfg_len = (ab->qmi.ce_cfg.svc_to_ce_map_len) /
-				(sizeof(struct service_to_pipe));
-	for (ret = 0; ret < req->svc_cfg_len; ret++) {
-		req->svc_cfg[ret].service_id = svc_cfg[ret].service_id;
-		req->svc_cfg[ret].pipe_dir = svc_cfg[ret].pipedir;
-		req->svc_cfg[ret].pipe_num = svc_cfg[ret].pipenum;
+	req->svc_cfg_len = ab->qmi.ce_cfg.svc_to_ce_map_len;
+	for (pipe_num = 0; pipe_num < req->svc_cfg_len; pipe_num++) {
+		req->svc_cfg[pipe_num].service_id = svc_cfg[pipe_num].service_id;
+		req->svc_cfg[pipe_num].pipe_dir = svc_cfg[pipe_num].pipedir;
+		req->svc_cfg[pipe_num].pipe_num = svc_cfg[pipe_num].pipenum;
 	}
 	req->shadow_reg_valid = 0;
 	req->shadow_reg_v2_valid = 0;
-- 
cgit v1.2.3-59-g8ed1b


From d0f390eae07284939b10fe34edd45d4c59226783 Mon Sep 17 00:00:00 2001
From: Anilkumar Kolli <akolli@codeaurora.org>
Date: Wed, 27 Nov 2019 14:08:50 +0000
Subject: ath11k: pktlog: fix sending/using the pdev id

Fixes sending the pdev id(0,1,2 for mac0, mac1, mac2)
to FW in wmi cmd pktlog enable/disable.

Signed-off-by: Anilkumar Kolli <akolli@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/dp_rx.c | 8 +++++---
 drivers/net/wireless/ath/ath11k/wmi.c   | 6 +++---
 2 files changed, 8 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c b/drivers/net/wireless/ath/ath11k/dp_rx.c
index acad74658e64..67efa247bf65 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
@@ -1265,7 +1265,6 @@ static void ath11k_htt_pktlog(struct ath11k_base *ab, struct sk_buff *skb)
 	u8 pdev_id;
 
 	len = FIELD_GET(HTT_T2H_PPDU_STATS_INFO_PAYLOAD_SIZE, data->hdr);
-
 	if (len > ATH11K_HTT_PKTLOG_MAX_SIZE) {
 		ath11k_warn(ab, "htt pktlog buffer size %d, expected < %d\n",
 			    len,
@@ -1274,8 +1273,11 @@ static void ath11k_htt_pktlog(struct ath11k_base *ab, struct sk_buff *skb)
 	}
 
 	pdev_id = FIELD_GET(HTT_T2H_PPDU_STATS_INFO_PDEV_ID, data->hdr);
-	pdev_id = DP_HW2SW_MACID(pdev_id);
-	ar = ab->pdevs[pdev_id].ar;
+	ar = ath11k_mac_get_ar_by_pdev_id(ab, pdev_id);
+	if (!ar) {
+		ath11k_warn(ab, "invalid pdev id %d on htt pktlog\n", pdev_id);
+		return;
+	}
 
 	trace_ath11k_htt_pktlog(ar, data->payload, len);
 }
diff --git a/drivers/net/wireless/ath/ath11k/wmi.c b/drivers/net/wireless/ath/ath11k/wmi.c
index aae6e76330da..f349bf2c1ee8 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.c
+++ b/drivers/net/wireless/ath/ath11k/wmi.c
@@ -2321,7 +2321,7 @@ int ath11k_wmi_pdev_peer_pktlog_filter(struct ath11k *ar, u8 *addr, u8 enable)
 	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_PDEV_PEER_PKTLOG_FILTER_CMD) |
 			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
 
-	cmd->pdev_id = ar->pdev->pdev_id;
+	cmd->pdev_id = DP_HW2SW_MACID(ar->pdev->pdev_id);
 	cmd->num_mac = 1;
 	cmd->enable = enable;
 
@@ -2419,7 +2419,7 @@ int ath11k_wmi_pdev_pktlog_enable(struct ath11k *ar, u32 pktlog_filter)
 	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_PDEV_PKTLOG_ENABLE_CMD) |
 			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
 
-	cmd->pdev_id = ar->pdev->pdev_id;
+	cmd->pdev_id = DP_HW2SW_MACID(ar->pdev->pdev_id);
 	cmd->evlist = pktlog_filter;
 	cmd->enable = ATH11K_WMI_PKTLOG_ENABLE_FORCE;
 
@@ -2449,7 +2449,7 @@ int ath11k_wmi_pdev_pktlog_disable(struct ath11k *ar)
 	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_PDEV_PKTLOG_DISABLE_CMD) |
 			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
 
-	cmd->pdev_id = ar->pdev->pdev_id;
+	cmd->pdev_id = DP_HW2SW_MACID(ar->pdev->pdev_id);
 
 	ret = ath11k_wmi_cmd_send(wmi, skb,
 				  WMI_PDEV_PKTLOG_DISABLE_CMDID);
-- 
cgit v1.2.3-59-g8ed1b


From f425078b449f90793c73423e4bbc44da6aad48d6 Mon Sep 17 00:00:00 2001
From: Karthikeyan Periyasamy <periyasa@codeaurora.org>
Date: Wed, 27 Nov 2019 14:08:52 +0000
Subject: ath11k: avoid burst time conversion logic

WMI_VDEV_SET_WMM_PARAMS commmand expects the txoplimit param in the units of
32 microseconds. convert the txop unit from 32 microseconds to absolute
microseconds leads to the higher burst values which is incorrect. so no need
to convert the txop unit from 32 microseconds to absolute microseconds.

Signed-off-by: Karthikeyan Periyasamy <periyasa@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/mac.c | 7 +------
 1 file changed, 1 insertion(+), 6 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index 6f82fdbbd358..412c258143ca 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -3023,12 +3023,7 @@ static int ath11k_mac_op_conf_tx(struct ieee80211_hw *hw,
 	p->cwmin = params->cw_min;
 	p->cwmax = params->cw_max;
 	p->aifs = params->aifs;
-
-	/* The channel time duration programmed in the HW is in absolute
-	 * microseconds, while mac80211 gives the txop in units of
-	 * 32 microseconds.
-	 */
-	p->txop = params->txop * 32;
+	p->txop = params->txop;
 
 	ret = ath11k_wmi_send_wmm_update_cmd_tlv(ar, arvif->vdev_id,
 						 &arvif->wmm_params);
-- 
cgit v1.2.3-59-g8ed1b


From 30679ec409189de89f55552a623c264091d72b66 Mon Sep 17 00:00:00 2001
From: Karthikeyan Periyasamy <periyasa@codeaurora.org>
Date: Wed, 27 Nov 2019 14:08:53 +0000
Subject: ath11k: avoid use_after_free in ath11k_dp_rx_msdu_coalesce API

Accessing already stored first msdu data after the skb expand trigger
use_after_free, since first msdu got deleted. so do the descriptor copy
operation before the skb expand operation.

Signed-off-by: Karthikeyan Periyasamy <periyasa@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/dp_rx.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c b/drivers/net/wireless/ath/ath11k/dp_rx.c
index 67efa247bf65..f87bd327b082 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
@@ -1376,6 +1376,11 @@ static int ath11k_dp_rx_msdu_coalesce(struct ath11k *ar,
 	skb_put(first, DP_RX_BUFFER_SIZE);
 	skb_pull(first, buf_first_hdr_len);
 
+	/* When an MSDU spread over multiple buffers attention, MSDU_END and
+	 * MPDU_END tlvs are valid only in the last buffer. Copy those tlvs.
+	 */
+	ath11k_dp_rx_desc_end_tlv_copy(rxcb->rx_desc, ldesc);
+
 	space_extra = msdu_len - (buf_first_len + skb_tailroom(first));
 	if (space_extra > 0 &&
 	    (pskb_expand_head(first, 0, space_extra, GFP_ATOMIC) < 0)) {
@@ -1391,11 +1396,6 @@ static int ath11k_dp_rx_msdu_coalesce(struct ath11k *ar,
 		return -ENOMEM;
 	}
 
-	/* When an MSDU spread over multiple buffers attention, MSDU_END and
-	 * MPDU_END tlvs are valid only in the last buffer. Copy those tlvs.
-	 */
-	ath11k_dp_rx_desc_end_tlv_copy(rxcb->rx_desc, ldesc);
-
 	rem_len = msdu_len - buf_first_len;
 	while ((skb = __skb_dequeue(msdu_list)) != NULL && rem_len > 0) {
 		rxcb = ATH11K_SKB_RXCB(skb);
-- 
cgit v1.2.3-59-g8ed1b


From fe201947f8bd4aed649a69e5faf0934652327804 Mon Sep 17 00:00:00 2001
From: Venkateswara Naralasetty <vnaralas@codeaurora.org>
Date: Wed, 27 Nov 2019 14:08:57 +0000
Subject: ath11k: update bawindow size in delba process

Currenly in delba process calling ath11k_peer_rx_tid_delete() updates
reo with desc invalid and add tid queue to the flush list. If station
send data traffic without addba req and before tid flush, hw gives
those packets as invalid desc reo error. Since we are dropping these
invalid desc packets results in traffic stall.

This patch fix this issue by updating the reo queue with bawindow size 1
instead of tid removal in delba process.

Signed-off-by: Venkateswara Naralasetty <vnaralas@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/dp_rx.c | 28 +++++++++++++++++++---------
 1 file changed, 19 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c b/drivers/net/wireless/ath/ath11k/dp_rx.c
index f87bd327b082..8c21925a522a 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
@@ -641,7 +641,8 @@ void ath11k_peer_rx_tid_cleanup(struct ath11k *ar, struct ath11k_peer *peer)
 static int ath11k_peer_rx_tid_reo_update(struct ath11k *ar,
 					 struct ath11k_peer *peer,
 					 struct dp_rx_tid *rx_tid,
-					 u32 ba_win_sz, u16 ssn)
+					 u32 ba_win_sz, u16 ssn,
+					 bool update_ssn)
 {
 	struct ath11k_hal_reo_cmd cmd = {0};
 	int ret;
@@ -649,10 +650,13 @@ static int ath11k_peer_rx_tid_reo_update(struct ath11k *ar,
 	cmd.addr_lo = lower_32_bits(rx_tid->paddr);
 	cmd.addr_hi = upper_32_bits(rx_tid->paddr);
 	cmd.flag = HAL_REO_CMD_FLG_NEED_STATUS;
-	cmd.upd0 = HAL_REO_CMD_UPD0_BA_WINDOW_SIZE |
-		   HAL_REO_CMD_UPD0_SSN;
+	cmd.upd0 = HAL_REO_CMD_UPD0_BA_WINDOW_SIZE;
 	cmd.ba_window_size = ba_win_sz;
-	cmd.upd2 = FIELD_PREP(HAL_REO_CMD_UPD2_SSN, ssn);
+
+	if (update_ssn) {
+		cmd.upd0 |= HAL_REO_CMD_UPD0_SSN;
+		cmd.upd2 = FIELD_PREP(HAL_REO_CMD_UPD2_SSN, ssn);
+	}
 
 	ret = ath11k_dp_tx_send_reo_cmd(ar->ab, rx_tid,
 					HAL_REO_CMD_UPDATE_RX_QUEUE, &cmd,
@@ -722,7 +726,7 @@ int ath11k_peer_rx_tid_setup(struct ath11k *ar, const u8 *peer_mac, int vdev_id,
 	if (rx_tid->active) {
 		paddr = rx_tid->paddr;
 		ret = ath11k_peer_rx_tid_reo_update(ar, peer, rx_tid,
-						    ba_win_sz, ssn);
+						    ba_win_sz, ssn, true);
 		spin_unlock_bh(&ab->base_lock);
 		if (ret) {
 			ath11k_warn(ab, "failed to update reo for rx tid %d\n", tid);
@@ -832,12 +836,18 @@ int ath11k_dp_rx_ampdu_stop(struct ath11k *ar,
 	paddr = peer->rx_tid[params->tid].paddr;
 	active = peer->rx_tid[params->tid].active;
 
-	ath11k_peer_rx_tid_delete(ar, peer, params->tid);
+	if (!active) {
+		spin_unlock_bh(&ab->base_lock);
+		return 0;
+	}
 
+	ret = ath11k_peer_rx_tid_reo_update(ar, peer, peer->rx_tid, 1, 0, false);
 	spin_unlock_bh(&ab->base_lock);
-
-	if (!active)
-		return 0;
+	if (ret) {
+		ath11k_warn(ab, "failed to update reo for rx tid %d: %d\n",
+			    params->tid, ret);
+		return ret;
+	}
 
 	ret = ath11k_wmi_peer_rx_reorder_queue_setup(ar, vdev_id,
 						     params->sta->addr, paddr,
-- 
cgit v1.2.3-59-g8ed1b


From 64f1d7e94daaaf53208e9505448406b5c8fbbd8a Mon Sep 17 00:00:00 2001
From: Maharaja Kennadyrajan <mkenna@codeaurora.org>
Date: Wed, 27 Nov 2019 14:08:57 +0000
Subject: ath11k: add support for controlling tx power to a station

This patch will add the support to control the transmit power
for traffic to a station associated with the AP.

Underlying firmware will enforce that the maximum tx power will
be based on the regulatory requirements. If the user given
transmit power is greater than the allowed tx power in the given
channel, then the firmware will use the maximum tx power in the
same channel.

Max and Min tx power values will depends on number of tx chain
masks. The allowed tx power range values are from 6 to 23.

When 0 is sent to the firmware as tx power, it will revert to
the default tx power for the station.

Signed-off-by: Maharaja Kennadyrajan <mkenna@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/debug.h |  3 +++
 drivers/net/wireless/ath/ath11k/mac.c   | 37 +++++++++++++++++++++++++++++++++
 2 files changed, 40 insertions(+)

diff --git a/drivers/net/wireless/ath/ath11k/debug.h b/drivers/net/wireless/ath/ath11k/debug.h
index aef33f83c9b1..a317a7bdb9a2 100644
--- a/drivers/net/wireless/ath/ath11k/debug.h
+++ b/drivers/net/wireless/ath/ath11k/debug.h
@@ -9,6 +9,9 @@
 #include "hal_tx.h"
 #include "trace.h"
 
+#define ATH11K_TX_POWER_MAX_VAL	70
+#define ATH11K_TX_POWER_MIN_VAL	0
+
 enum ath11k_debug_mask {
 	ATH11K_DBG_AHB		= 0x00000001,
 	ATH11K_DBG_WMI		= 0x00000002,
diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index 412c258143ca..5231c4fa0e38 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -2842,6 +2842,41 @@ static int ath11k_mac_op_sta_state(struct ieee80211_hw *hw,
 	return ret;
 }
 
+static int ath11k_mac_op_sta_set_txpwr(struct ieee80211_hw *hw,
+				       struct ieee80211_vif *vif,
+				       struct ieee80211_sta *sta)
+{
+	struct ath11k *ar = hw->priv;
+	struct ath11k_vif *arvif = (void *)vif->drv_priv;
+	int ret = 0;
+	s16 txpwr;
+
+	if (sta->txpwr.type == NL80211_TX_POWER_AUTOMATIC) {
+		txpwr = 0;
+	} else {
+		txpwr = sta->txpwr.power;
+		if (!txpwr)
+			return -EINVAL;
+	}
+
+	if (txpwr > ATH11K_TX_POWER_MAX_VAL || txpwr < ATH11K_TX_POWER_MIN_VAL)
+		return -EINVAL;
+
+	mutex_lock(&ar->conf_mutex);
+
+	ret = ath11k_wmi_set_peer_param(ar, sta->addr, arvif->vdev_id,
+					WMI_PEER_USE_FIXED_PWR, txpwr);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to set tx power for station ret: %d\n",
+			    ret);
+		goto out;
+	}
+
+out:
+	mutex_unlock(&ar->conf_mutex);
+	return ret;
+}
+
 static void ath11k_mac_op_sta_rc_update(struct ieee80211_hw *hw,
 					struct ieee80211_vif *vif,
 					struct ieee80211_sta *sta,
@@ -5311,6 +5346,7 @@ static const struct ieee80211_ops ath11k_ops = {
 	.cancel_hw_scan                 = ath11k_mac_op_cancel_hw_scan,
 	.set_key                        = ath11k_mac_op_set_key,
 	.sta_state                      = ath11k_mac_op_sta_state,
+	.sta_set_txpwr			= ath11k_mac_op_sta_set_txpwr,
 	.sta_rc_update			= ath11k_mac_op_sta_rc_update,
 	.conf_tx                        = ath11k_mac_op_conf_tx,
 	.set_antenna			= ath11k_mac_op_set_antenna,
@@ -5571,6 +5607,7 @@ static int ath11k_mac_register(struct ath11k *ar)
 	ar->hw->wiphy->n_iface_combinations = ARRAY_SIZE(ath11k_if_comb);
 
 	wiphy_ext_feature_set(ar->hw->wiphy, NL80211_EXT_FEATURE_CQM_RSSI_LIST);
+	wiphy_ext_feature_set(ar->hw->wiphy, NL80211_EXT_FEATURE_STA_TX_PWR);
 
 	ar->hw->wiphy->cipher_suites = cipher_suites;
 	ar->hw->wiphy->n_cipher_suites = ARRAY_SIZE(cipher_suites);
-- 
cgit v1.2.3-59-g8ed1b


From ba47923974fb67bf612b7a7c07c5884e95c5518b Mon Sep 17 00:00:00 2001
From: Govindaraj Saminathan <gsamin@codeaurora.org>
Date: Wed, 27 Nov 2019 14:08:58 +0000
Subject: ath11k: unlock mutex during failure in qmi fw ready

qmi firmware ready event start to initialize the core modules and
the sequence executed with mutex lock. In case of any failure
mutex should be unlocked otherwise it will hang during the recovery.

Signed-off-by: Govindaraj Saminathan <gsamin@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/core.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wireless/ath/ath11k/core.c b/drivers/net/wireless/ath/ath11k/core.c
index b0c71b5879b1..37d4ecb458db 100644
--- a/drivers/net/wireless/ath/ath11k/core.c
+++ b/drivers/net/wireless/ath/ath11k/core.c
@@ -539,6 +539,7 @@ err_core_stop:
 	ath11k_core_stop(ab);
 err_dp_free:
 	ath11k_dp_free(ab);
+	mutex_unlock(&ab->core_lock);
 	return ret;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 1cb616a3b497d6f8b8160a3fee97e822cdb38f39 Mon Sep 17 00:00:00 2001
From: Sriram R <srirrama@codeaurora.org>
Date: Wed, 27 Nov 2019 14:09:00 +0000
Subject: ath11k: add necessary peer assoc params in wmi dbg

Add necessary peer assoc params in WMI debug message
while sending the peer assoc command to firmware
to aid in debugging.

Signed-off-by: Sriram R <srirrama@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/wmi.c | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/wmi.c b/drivers/net/wireless/ath/ath11k/wmi.c
index f349bf2c1ee8..b05642617b78 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.c
+++ b/drivers/net/wireless/ath/ath11k/wmi.c
@@ -1804,8 +1804,17 @@ int ath11k_wmi_send_peer_assoc_cmd(struct ath11k *ar,
 	}
 
 	ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
-		   "WMI peer assoc vdev id %d assoc id %d peer mac %pM\n",
-		   param->vdev_id, param->peer_associd, param->peer_mac);
+		   "wmi peer assoc vdev id %d assoc id %d peer mac %pM peer_flags %x rate_caps %x peer_caps %x listen_intval %d ht_caps %x max_mpdu %d nss %d phymode %d peer_mpdu_density %d vht_caps %x he cap_info %x he ops %x he cap_info_ext %x he phy %x %x %x peer_bw_rxnss_override %x\n",
+		   cmd->vdev_id, cmd->peer_associd, param->peer_mac,
+		   cmd->peer_flags, cmd->peer_rate_caps, cmd->peer_caps,
+		   cmd->peer_listen_intval, cmd->peer_ht_caps,
+		   cmd->peer_max_mpdu, cmd->peer_nss, cmd->peer_phymode,
+		   cmd->peer_mpdu_density,
+		   cmd->peer_vht_caps, cmd->peer_he_cap_info,
+		   cmd->peer_he_ops, cmd->peer_he_cap_info_ext,
+		   cmd->peer_he_cap_phy[0], cmd->peer_he_cap_phy[1],
+		   cmd->peer_he_cap_phy[2],
+		   cmd->peer_bw_rxnss_override);
 
 	return ret;
 }
-- 
cgit v1.2.3-59-g8ed1b


From a3c5195a97afdfff1f9ae0569c3a4cd98ebd5696 Mon Sep 17 00:00:00 2001
From: Sriram R <srirrama@codeaurora.org>
Date: Thu, 28 Nov 2019 08:21:45 +0000
Subject: ath11k: Update tx and rx chain count properly on drv_set_antenna

Set the number of tx and rx chains properly on drv_set_antenna().
This will ensure the related ht/vht/he caps are properly recalculated
based on the tx/rx chains set.

Signed-off-by: Sriram R <srirrama@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/mac.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index 5231c4fa0e38..5ee7c2a1c611 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -3480,6 +3480,8 @@ static int __ath11k_set_antenna(struct ath11k *ar, u32 tx_ant, u32 rx_ant)
 		return ret;
 	}
 
+	ar->num_tx_chains = get_num_chains(tx_ant);
+
 	ret = ath11k_wmi_pdev_set_param(ar, WMI_PDEV_PARAM_RX_CHAIN_MASK,
 					rx_ant, ar->pdev->pdev_id);
 	if (ret) {
@@ -3488,6 +3490,8 @@ static int __ath11k_set_antenna(struct ath11k *ar, u32 tx_ant, u32 rx_ant)
 		return ret;
 	}
 
+	ar->num_rx_chains = get_num_chains(rx_ant);
+
 	/* Reload HT/VHT/HE capability */
 	ath11k_mac_setup_ht_vht_cap(ar, &ar->pdev->cap, NULL);
 	ath11k_mac_setup_he_cap(ar, &ar->pdev->cap);
-- 
cgit v1.2.3-59-g8ed1b


From c000e56ee6b0b84e1ff33f6ee540e3b7be263ffa Mon Sep 17 00:00:00 2001
From: Venkateswara Naralasetty <vnaralas@codeaurora.org>
Date: Thu, 28 Nov 2019 08:21:47 +0000
Subject: ath11k: Advertise MPDU start spacing as no restriction

Adverise MPDU start spacing as no restriction in ht capabilities,
Since IPQ8074 hw support all sorts of mpdu start spcing. With this
observed minor uplink performance improvement in lower data frame
size case with Veriwave clients.

Signed-off-by: Venkateswara Naralasetty <vnaralas@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/mac.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index 5ee7c2a1c611..b048057b398f 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -3089,7 +3089,7 @@ ath11k_create_ht_cap(struct ath11k *ar, u32 ar_ht_cap, u32 rate_cap_rx_chainmask
 
 	ht_cap.ht_supported = 1;
 	ht_cap.ampdu_factor = IEEE80211_HT_MAX_AMPDU_64K;
-	ht_cap.ampdu_density = IEEE80211_HT_MPDU_DENSITY_8;
+	ht_cap.ampdu_density = IEEE80211_HT_MPDU_DENSITY_NONE;
 	ht_cap.cap |= IEEE80211_HT_CAP_SUP_WIDTH_20_40;
 	ht_cap.cap |= IEEE80211_HT_CAP_DSSSCCK40;
 	ht_cap.cap |= WLAN_HT_CAP_SM_PS_STATIC << IEEE80211_HT_CAP_SM_PS_SHIFT;
-- 
cgit v1.2.3-59-g8ed1b


From 5e02bc7354e1339ad801c7a2e31eb7f7bbeb7b22 Mon Sep 17 00:00:00 2001
From: Miles Hu <milehu@codeaurora.org>
Date: Thu, 28 Nov 2019 08:21:49 +0000
Subject: ath11k: fix memory leak in monitor mode

remove tail check to avoid last amsdu leak.
recycle skb in ppdu id wrap around case

Signed-off-by: Miles Hu <milehu@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/dp_rx.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c b/drivers/net/wireless/ath/ath11k/dp_rx.c
index 8c21925a522a..85e999c7a238 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
@@ -3735,8 +3735,15 @@ ath11k_dp_rx_mon_mpdu_pop(struct ath11k *ar,
 
 				if (ath11k_dp_rx_mon_comp_ppduid(msdu_ppdu_id,
 								 ppdu_id,
-								 &rx_bufs_used))
+								 &rx_bufs_used)) {
+					if (rx_bufs_used) {
+						drop_mpdu = true;
+						dev_kfree_skb_any(msdu);
+						msdu = NULL;
+						goto next_msdu;
+					}
 					return rx_bufs_used;
+				}
 				pmon->mon_last_linkdesc_paddr = paddr;
 				is_first_msdu = false;
 			}
@@ -3938,7 +3945,7 @@ static int ath11k_dp_rx_mon_deliver(struct ath11k *ar, u32 mac_id,
 
 		ath11k_dp_rx_deliver_msdu(ar, napi, mon_skb);
 		mon_skb = skb_next;
-	} while (mon_skb && (mon_skb != tail_msdu));
+	} while (mon_skb);
 	rxs->flag = 0;
 
 	return 0;
-- 
cgit v1.2.3-59-g8ed1b


From 28dee8ef7645cdc3e19814ac909c7eff781b66a5 Mon Sep 17 00:00:00 2001
From: Manikanta Pubbisetty <mpubbise@codeaurora.org>
Date: Thu, 28 Nov 2019 08:21:50 +0000
Subject: ath11k: fix vht guard interval mapping

Guard interval value which comes from VHT_SIG_A TLV has a mapping
where value 0 corresponds to LGI, 1 and 3 corresponds to SGI.

Value 3 which is SGI(0.4us) in VHT was incorrectly mapped to a GI
of 3.2us(only applicable in HE) resulting in incorrect rx GI stats.
Fixing the mapping.

Signed-off-by: Manikanta Pubbisetty <mpubbise@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/hal_rx.c | 16 +++++++++++++---
 drivers/net/wireless/ath/ath11k/hal_rx.h |  6 ++++++
 2 files changed, 19 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/hal_rx.c b/drivers/net/wireless/ath/ath11k/hal_rx.c
index 6168a4eb0c0b..9e0f8064e427 100644
--- a/drivers/net/wireless/ath/ath11k/hal_rx.c
+++ b/drivers/net/wireless/ath/ath11k/hal_rx.c
@@ -920,6 +920,7 @@ ath11k_hal_rx_parse_mon_status_tlv(struct ath11k_base *ab,
 			(struct hal_rx_vht_sig_a_info *)tlv_data;
 		u32 nsts;
 		u32 group_id;
+		u8 gi_setting;
 
 		info0 = __le32_to_cpu(vht_sig->info0);
 		info1 = __le32_to_cpu(vht_sig->info1);
@@ -928,9 +929,18 @@ ath11k_hal_rx_parse_mon_status_tlv(struct ath11k_base *ab,
 					    info0);
 		ppdu_info->mcs = FIELD_GET(HAL_RX_VHT_SIG_A_INFO_INFO1_MCS,
 					   info1);
-		ppdu_info->gi =
-			FIELD_GET(HAL_RX_VHT_SIG_A_INFO_INFO1_GI_SETTING,
-				  info1);
+		gi_setting = FIELD_GET(HAL_RX_VHT_SIG_A_INFO_INFO1_GI_SETTING,
+				       info1);
+		switch (gi_setting) {
+		case HAL_RX_VHT_SIG_A_NORMAL_GI:
+			ppdu_info->gi = HAL_RX_GI_0_8_US;
+			break;
+		case HAL_RX_VHT_SIG_A_SHORT_GI:
+		case HAL_RX_VHT_SIG_A_SHORT_GI_AMBIGUITY:
+			ppdu_info->gi = HAL_RX_GI_0_4_US;
+			break;
+		}
+
 		ppdu_info->is_stbc = info0 & HAL_RX_VHT_SIG_A_INFO_INFO0_STBC;
 		nsts = FIELD_GET(HAL_RX_VHT_SIG_A_INFO_INFO0_NSTS, info0);
 		if (ppdu_info->is_stbc && nsts > 0)
diff --git a/drivers/net/wireless/ath/ath11k/hal_rx.h b/drivers/net/wireless/ath/ath11k/hal_rx.h
index 2271bf56183b..bb022c781c48 100644
--- a/drivers/net/wireless/ath/ath11k/hal_rx.h
+++ b/drivers/net/wireless/ath/ath11k/hal_rx.h
@@ -189,6 +189,12 @@ struct hal_rx_vht_sig_a_info {
 	__le32 info1;
 } __packed;
 
+enum hal_rx_vht_sig_a_gi_setting {
+	HAL_RX_VHT_SIG_A_NORMAL_GI = 0,
+	HAL_RX_VHT_SIG_A_SHORT_GI = 1,
+	HAL_RX_VHT_SIG_A_SHORT_GI_AMBIGUITY = 3,
+};
+
 #define HAL_RX_HE_SIG_A_SU_INFO_INFO0_TRANSMIT_MCS	GENMASK(6, 3)
 #define HAL_RX_HE_SIG_A_SU_INFO_INFO0_DCM		BIT(7)
 #define HAL_RX_HE_SIG_A_SU_INFO_INFO0_TRANSMIT_BW	GENMASK(20, 19)
-- 
cgit v1.2.3-59-g8ed1b


From a9e945eadf1759fd2255e952a0b57ac7bf04b4b0 Mon Sep 17 00:00:00 2001
From: Venkateswara Naralasetty <vnaralas@codeaurora.org>
Date: Thu, 28 Nov 2019 08:21:51 +0000
Subject: ath11k: update tx duration in station info

Update tx duration in station info form PPDU stats
so that users can dump tx duration of the station.

Signed-off-by: Venkateswara Naralasetty <vnaralas@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/core.h  | 1 +
 drivers/net/wireless/ath/ath11k/dp_rx.c | 2 +-
 drivers/net/wireless/ath/ath11k/mac.c   | 3 +++
 3 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath11k/core.h b/drivers/net/wireless/ath/ath11k/core.h
index 065cb9dc880b..06482a5c102d 100644
--- a/drivers/net/wireless/ath/ath11k/core.h
+++ b/drivers/net/wireless/ath/ath11k/core.h
@@ -335,6 +335,7 @@ struct ath11k_sta {
 	struct rate_info txrate;
 	struct rate_info last_txrate;
 	u64 rx_duration;
+	u64 tx_duration;
 	u8 rssi_comb;
 	struct ath11k_htt_tx_stats *tx_stats;
 	struct ath11k_rx_peer_stats *rx_stats;
diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c b/drivers/net/wireless/ath/ath11k/dp_rx.c
index 85e999c7a238..8f69de594ada 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
@@ -1147,7 +1147,7 @@ ath11k_update_per_peer_tx_stats(struct ath11k *ar,
 	arsta->txrate.nss = nss;
 	arsta->txrate.bw = ath11k_mac_bw_to_mac80211_bw(bw);
 	arsta->tx_info.status.rates[0].flags |= ath11k_bw_to_mac80211_bwflags(bw);
-
+	arsta->tx_duration += tx_duration;
 	memcpy(&arsta->last_txrate, &arsta->txrate, sizeof(struct rate_info));
 
 	if (succ_pkts) {
diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index b048057b398f..624534d4c33a 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -5316,6 +5316,9 @@ static void ath11k_mac_op_sta_statistics(struct ieee80211_hw *hw,
 	sinfo->rx_duration = arsta->rx_duration;
 	sinfo->filled |= BIT_ULL(NL80211_STA_INFO_RX_DURATION);
 
+	sinfo->tx_duration = arsta->tx_duration;
+	sinfo->filled |= BIT_ULL(NL80211_STA_INFO_TX_DURATION);
+
 	if (!arsta->txrate.legacy && !arsta->txrate.nss)
 		return;
 
-- 
cgit v1.2.3-59-g8ed1b


From b9269a078ae609f1315726f1ac4c7d3d8425384f Mon Sep 17 00:00:00 2001
From: Venkateswara Naralasetty <vnaralas@codeaurora.org>
Date: Thu, 28 Nov 2019 08:21:54 +0000
Subject: ath11k: Skip update peer stats for management packets

Currently HTT_PPDU_STATS_TAG_USR_COMPLTN_ACK_BA_STATUS tag of PPDU stats
doesn't have valid success bytes info of management frames.
So skip update peer stats for management packets.

Signed-off-by: Venkateswara Naralasetty <vnaralas@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/dp.h    |  4 +++-
 drivers/net/wireless/ath/ath11k/dp_rx.c | 33 ++++++++++++++++++++-------------
 2 files changed, 23 insertions(+), 14 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp.h b/drivers/net/wireless/ath/ath11k/dp.h
index db216f055c42..82ddd7017722 100644
--- a/drivers/net/wireless/ath/ath11k/dp.h
+++ b/drivers/net/wireless/ath/ath11k/dp.h
@@ -1176,7 +1176,9 @@ struct htt_ppdu_stats_usr_cmpltn_cmn {
 
 #define HTT_PPDU_STATS_ACK_BA_INFO_NUM_MPDU_M	GENMASK(8, 0)
 #define HTT_PPDU_STATS_ACK_BA_INFO_NUM_MSDU_M	GENMASK(24, 9)
-#define HTT_PPDU_STATS_ACK_BA_INFO_TID_NUM	GENMASK(3, 0)
+#define HTT_PPDU_STATS_ACK_BA_INFO_TID_NUM	GENMASK(31, 25)
+
+#define HTT_PPDU_STATS_NON_QOS_TID	16
 
 struct htt_ppdu_stats_usr_cmpltn_ack_ba_status {
 	u32 ppdu_id;
diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c b/drivers/net/wireless/ath/ath11k/dp_rx.c
index 8f69de594ada..a380efb9caa6 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
@@ -1034,6 +1034,7 @@ ath11k_update_per_peer_tx_stats(struct ath11k *ar,
 	u32 succ_bytes = 0;
 	u16 rate = 0, succ_pkts = 0;
 	u32 tx_duration = 0;
+	u8 tid = HTT_PPDU_STATS_NON_QOS_TID;
 	bool is_ampdu = false;
 
 	if (!usr_stats)
@@ -1051,6 +1052,8 @@ ath11k_update_per_peer_tx_stats(struct ath11k *ar,
 		succ_bytes = usr_stats->ack_ba.success_bytes;
 		succ_pkts = FIELD_GET(HTT_PPDU_STATS_ACK_BA_INFO_NUM_MSDU_M,
 				      usr_stats->ack_ba.info);
+		tid = FIELD_GET(HTT_PPDU_STATS_ACK_BA_INFO_TID_NUM,
+				usr_stats->ack_ba.info);
 	}
 
 	if (common->fes_duration_us)
@@ -1156,19 +1159,23 @@ ath11k_update_per_peer_tx_stats(struct ath11k *ar,
 		ieee80211_tx_rate_update(ar->hw, sta, &arsta->tx_info);
 	}
 
-	memset(peer_stats, 0, sizeof(*peer_stats));
-
-	peer_stats->succ_pkts = succ_pkts;
-	peer_stats->succ_bytes = succ_bytes;
-	peer_stats->is_ampdu = is_ampdu;
-	peer_stats->duration = tx_duration;
-	peer_stats->ba_fails =
-		HTT_USR_CMPLTN_LONG_RETRY(usr_stats->cmpltn_cmn.flags) +
-		HTT_USR_CMPLTN_SHORT_RETRY(usr_stats->cmpltn_cmn.flags);
-
-	if (ath11k_debug_is_extd_tx_stats_enabled(ar))
-		ath11k_accumulate_per_peer_tx_stats(arsta,
-						    peer_stats, rate_idx);
+	/* PPDU stats reported for mgmt packet doesn't have valid tx bytes.
+	 * So skip peer stats update for mgmt packets.
+	 */
+	if (tid < HTT_PPDU_STATS_NON_QOS_TID) {
+		memset(peer_stats, 0, sizeof(*peer_stats));
+		peer_stats->succ_pkts = succ_pkts;
+		peer_stats->succ_bytes = succ_bytes;
+		peer_stats->is_ampdu = is_ampdu;
+		peer_stats->duration = tx_duration;
+		peer_stats->ba_fails =
+			HTT_USR_CMPLTN_LONG_RETRY(usr_stats->cmpltn_cmn.flags) +
+			HTT_USR_CMPLTN_SHORT_RETRY(usr_stats->cmpltn_cmn.flags);
+
+		if (ath11k_debug_is_extd_tx_stats_enabled(ar))
+			ath11k_accumulate_per_peer_tx_stats(arsta,
+							    peer_stats, rate_idx);
+	}
 
 	spin_unlock_bh(&ab->base_lock);
 	rcu_read_unlock();
-- 
cgit v1.2.3-59-g8ed1b


From 0366f42640a4100198edcf03bf7625ee1c46f42b Mon Sep 17 00:00:00 2001
From: Vasanthakumar Thiagarajan <vthiagar@codeaurora.org>
Date: Thu, 28 Nov 2019 08:21:57 +0000
Subject: ath11k: Move mac80211 hw allocation before wmi_init command

This is to prepare REO ring setup before sending wmi_init command.
Firmware expects all the required REO rings to be setup while processing
wmi_init command. But as per the current initialization sequence, REO ring
configurations are done only after wmi_init command is sent.
Also refactoring ath11k_mac_create() into ath11k_mac_alloc() and
ath11k_mac_register() to it mac80211 hw structure available before sending
wmi_init command.

Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/core.c | 27 ++++++----
 drivers/net/wireless/ath/ath11k/mac.c  | 97 +++++++++++++++++++++-------------
 drivers/net/wireless/ath/ath11k/mac.h  |  3 +-
 3 files changed, 81 insertions(+), 46 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/core.c b/drivers/net/wireless/ath/ath11k/core.c
index 37d4ecb458db..c9c9e4b5ef6b 100644
--- a/drivers/net/wireless/ath/ath11k/core.c
+++ b/drivers/net/wireless/ath/ath11k/core.c
@@ -378,23 +378,22 @@ static int ath11k_core_pdev_create(struct ath11k_base *ab)
 		return ret;
 	}
 
-	ret = ath11k_mac_create(ab);
+	ret = ath11k_mac_register(ab);
 	if (ret) {
-		ath11k_err(ab, "failed to create new hw device with mac80211 :%d\n",
-			   ret);
+		ath11k_err(ab, "failed register the radio with mac80211: %d\n", ret);
 		goto err_pdev_debug;
 	}
 
 	ret = ath11k_dp_pdev_alloc(ab);
 	if (ret) {
 		ath11k_err(ab, "failed to attach DP pdev: %d\n", ret);
-		goto err_mac_destroy;
+		goto err_mac_unregister;
 	}
 
 	return 0;
 
-err_mac_destroy:
-	ath11k_mac_destroy(ab);
+err_mac_unregister:
+	ath11k_mac_unregister(ab);
 
 err_pdev_debug:
 	ath11k_debug_pdev_destroy(ab);
@@ -470,28 +469,37 @@ static int ath11k_core_start(struct ath11k_base *ab,
 		goto err_hif_stop;
 	}
 
+	ret = ath11k_mac_allocate(ab);
+	if (ret) {
+		ath11k_err(ab, "failed to create new hw device with mac80211 :%d\n",
+			   ret);
+		goto err_hif_stop;
+	}
+
 	ret = ath11k_wmi_cmd_init(ab);
 	if (ret) {
 		ath11k_err(ab, "failed to send wmi init cmd: %d\n", ret);
-		goto err_hif_stop;
+		goto err_mac_destroy;
 	}
 
 	ret = ath11k_wmi_wait_for_unified_ready(ab);
 	if (ret) {
 		ath11k_err(ab, "failed to receive wmi unified ready event: %d\n",
 			   ret);
-		goto err_hif_stop;
+		goto err_mac_destroy;
 	}
 
 	ret = ath11k_dp_tx_htt_h2t_ver_req_msg(ab);
 	if (ret) {
 		ath11k_err(ab, "failed to send htt version request message: %d\n",
 			   ret);
-		goto err_hif_stop;
+		goto err_mac_destroy;
 	}
 
 	return 0;
 
+err_mac_destroy:
+	ath11k_mac_destroy(ab);
 err_hif_stop:
 	ath11k_ahb_stop(ab);
 err_wmi_detach:
@@ -537,6 +545,7 @@ int ath11k_core_qmi_firmware_ready(struct ath11k_base *ab)
 
 err_core_stop:
 	ath11k_core_stop(ab);
+	ath11k_mac_destroy(ab);
 err_dp_free:
 	ath11k_dp_free(ab);
 	mutex_unlock(&ab->core_lock);
diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index 624534d4c33a..e4dfce6b63f0 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -5513,7 +5513,38 @@ static const struct wiphy_iftype_ext_capab ath11k_iftypes_ext_capa[] = {
 	},
 };
 
-static int ath11k_mac_register(struct ath11k *ar)
+static void __ath11k_mac_unregister(struct ath11k *ar)
+{
+	cancel_work_sync(&ar->regd_update_work);
+
+	ieee80211_unregister_hw(ar->hw);
+
+	idr_for_each(&ar->txmgmt_idr, ath11k_mac_tx_mgmt_pending_free, ar);
+	idr_destroy(&ar->txmgmt_idr);
+
+	kfree(ar->mac.sbands[NL80211_BAND_2GHZ].channels);
+	kfree(ar->mac.sbands[NL80211_BAND_5GHZ].channels);
+
+	SET_IEEE80211_DEV(ar->hw, NULL);
+}
+
+void ath11k_mac_unregister(struct ath11k_base *ab)
+{
+	struct ath11k *ar;
+	struct ath11k_pdev *pdev;
+	int i;
+
+	for (i = 0; i < ab->num_radios; i++) {
+		pdev = &ab->pdevs[i];
+		ar = pdev->ar;
+		if (!ar)
+			continue;
+
+		__ath11k_mac_unregister(ar);
+	}
+}
+
+static int __ath11k_mac_register(struct ath11k *ar)
 {
 	struct ath11k_base *ab = ar->ab;
 	struct ath11k_pdev_cap *cap = &ar->pdev->cap;
@@ -5657,32 +5688,48 @@ err_free:
 	return ret;
 }
 
-void ath11k_mac_unregister(struct ath11k_base *ab)
+int ath11k_mac_register(struct ath11k_base *ab)
 {
 	struct ath11k *ar;
 	struct ath11k_pdev *pdev;
 	int i;
+	int ret;
 
 	for (i = 0; i < ab->num_radios; i++) {
 		pdev = &ab->pdevs[i];
 		ar = pdev->ar;
-		if (!ar)
-			continue;
-		cancel_work_sync(&ar->regd_update_work);
+		if (ab->pdevs_macaddr_valid) {
+			ether_addr_copy(ar->mac_addr, pdev->mac_addr);
+		} else {
+			ether_addr_copy(ar->mac_addr, ab->mac_addr);
+			ar->mac_addr[4] += i;
+		}
 
-		ieee80211_unregister_hw(ar->hw);
+		ret = __ath11k_mac_register(ar);
+		if (ret)
+			goto err_cleanup;
 
-		idr_for_each(&ar->txmgmt_idr, ath11k_mac_tx_mgmt_pending_free, ar);
-		idr_destroy(&ar->txmgmt_idr);
+		idr_init(&ar->txmgmt_idr);
+		spin_lock_init(&ar->txmgmt_idr_lock);
+	}
 
-		kfree(ar->mac.sbands[NL80211_BAND_2GHZ].channels);
-		kfree(ar->mac.sbands[NL80211_BAND_5GHZ].channels);
+	/* Initialize channel counters frequency value in hertz */
+	ab->cc_freq_hz = IPQ8074_CC_FREQ_HERTZ;
+	ab->free_vdev_map = (1LL << (ab->num_radios * TARGET_NUM_VDEVS)) - 1;
+
+	return 0;
 
-		SET_IEEE80211_DEV(ar->hw, NULL);
+err_cleanup:
+	for (i = i - 1; i >= 0; i--) {
+		pdev = &ab->pdevs[i];
+		ar = pdev->ar;
+		__ath11k_mac_unregister(ar);
 	}
+
+	return ret;
 }
 
-int ath11k_mac_create(struct ath11k_base *ab)
+int ath11k_mac_allocate(struct ath11k_base *ab)
 {
 	struct ieee80211_hw *hw;
 	struct ath11k *ar;
@@ -5699,7 +5746,7 @@ int ath11k_mac_create(struct ath11k_base *ab)
 		if (!hw) {
 			ath11k_warn(ab, "failed to allocate mac80211 hw device\n");
 			ret = -ENOMEM;
-			goto err_destroy_mac;
+			goto err_free_mac;
 		}
 
 		ar = hw->priv;
@@ -5720,13 +5767,6 @@ int ath11k_mac_create(struct ath11k_base *ab)
 		ar->num_tx_chains = get_num_chains(pdev->cap.tx_chain_mask);
 		ar->num_rx_chains = get_num_chains(pdev->cap.rx_chain_mask);
 
-		if (ab->pdevs_macaddr_valid) {
-			ether_addr_copy(ar->mac_addr, pdev->mac_addr);
-		} else {
-			ether_addr_copy(ar->mac_addr, ab->mac_addr);
-			ar->mac_addr[4] += i;
-		}
-
 		pdev->ar = ar;
 		spin_lock_init(&ar->data_lock);
 		INIT_LIST_HEAD(&ar->arvifs);
@@ -5744,26 +5784,11 @@ int ath11k_mac_create(struct ath11k_base *ab)
 		INIT_WORK(&ar->wmi_mgmt_tx_work, ath11k_mgmt_over_wmi_tx_work);
 		skb_queue_head_init(&ar->wmi_mgmt_tx_queue);
 		clear_bit(ATH11K_FLAG_MONITOR_ENABLED, &ar->monitor_flags);
-
-		ret = ath11k_mac_register(ar);
-		if (ret) {
-			ath11k_warn(ab, "failed to register hw device\n");
-			pdev->ar = NULL;
-			ieee80211_free_hw(hw);
-			goto err_destroy_mac;
-		}
-
-		idr_init(&ar->txmgmt_idr);
-		spin_lock_init(&ar->txmgmt_idr_lock);
 	}
 
-	/* Initialize channel counters frequency value in hertz */
-	ab->cc_freq_hz = IPQ8074_CC_FREQ_HERTZ;
-	ab->free_vdev_map = (1LL << (ab->num_radios * TARGET_NUM_VDEVS)) - 1;
-
 	return 0;
 
-err_destroy_mac:
+err_free_mac:
 	ath11k_mac_destroy(ab);
 
 	return ret;
diff --git a/drivers/net/wireless/ath/ath11k/mac.h b/drivers/net/wireless/ath/ath11k/mac.h
index 8c37573ae5dc..f286531cdd78 100644
--- a/drivers/net/wireless/ath/ath11k/mac.h
+++ b/drivers/net/wireless/ath/ath11k/mac.h
@@ -118,9 +118,10 @@ struct ath11k_generic_iter {
 
 extern const struct htt_rx_ring_tlv_filter ath11k_mac_mon_status_filter_default;
 
-int ath11k_mac_create(struct ath11k_base *ab);
 void ath11k_mac_destroy(struct ath11k_base *ab);
 void ath11k_mac_unregister(struct ath11k_base *ab);
+int ath11k_mac_register(struct ath11k_base *ab);
+int ath11k_mac_allocate(struct ath11k_base *ab);
 int ath11k_mac_hw_ratecode_to_legacy_rate(u8 hw_rc, u8 preamble, u8 *rateidx,
 					  u16 *rate);
 u8 ath11k_mac_bitrate_to_idx(const struct ieee80211_supported_band *sband,
-- 
cgit v1.2.3-59-g8ed1b


From 9c57d7e3b4882c29bbcf3e555a1f4163d6e55646 Mon Sep 17 00:00:00 2001
From: Vasanthakumar Thiagarajan <vthiagar@codeaurora.org>
Date: Thu, 28 Nov 2019 08:21:58 +0000
Subject: ath11k: Setup REO destination ring before sending wmi_init command

Firmware expects all the required REO destination rings setup
while processing wmi_init command. Not doing this causes connected
stations getting disconnected and not able to connect back.

Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/core.c  | 19 +++++++++++--
 drivers/net/wireless/ath/ath11k/dp.c    | 10 +++++--
 drivers/net/wireless/ath/ath11k/dp.h    |  1 +
 drivers/net/wireless/ath/ath11k/dp_rx.c | 49 +++++++++++++++++++++++++++------
 drivers/net/wireless/ath/ath11k/dp_rx.h |  2 ++
 5 files changed, 67 insertions(+), 14 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/core.c b/drivers/net/wireless/ath/ath11k/core.c
index c9c9e4b5ef6b..5cdc9b2aee51 100644
--- a/drivers/net/wireless/ath/ath11k/core.c
+++ b/drivers/net/wireless/ath/ath11k/core.c
@@ -10,6 +10,7 @@
 #include "ahb.h"
 #include "core.h"
 #include "dp_tx.h"
+#include "dp_rx.h"
 #include "debug.h"
 
 unsigned int ath11k_debug_mask;
@@ -325,6 +326,7 @@ static void ath11k_core_stop(struct ath11k_base *ab)
 		ath11k_qmi_firmware_stop(ab);
 	ath11k_ahb_stop(ab);
 	ath11k_wmi_detach(ab);
+	ath11k_dp_pdev_reo_cleanup(ab);
 
 	/* De-Init of components as needed */
 }
@@ -476,28 +478,38 @@ static int ath11k_core_start(struct ath11k_base *ab,
 		goto err_hif_stop;
 	}
 
+	ath11k_dp_pdev_pre_alloc(ab);
+
+	ret = ath11k_dp_pdev_reo_setup(ab);
+	if (ret) {
+		ath11k_err(ab, "failed to initialize reo destination rings: %d\n", ret);
+		goto err_mac_destroy;
+	}
+
 	ret = ath11k_wmi_cmd_init(ab);
 	if (ret) {
 		ath11k_err(ab, "failed to send wmi init cmd: %d\n", ret);
-		goto err_mac_destroy;
+		goto err_reo_cleanup;
 	}
 
 	ret = ath11k_wmi_wait_for_unified_ready(ab);
 	if (ret) {
 		ath11k_err(ab, "failed to receive wmi unified ready event: %d\n",
 			   ret);
-		goto err_mac_destroy;
+		goto err_reo_cleanup;
 	}
 
 	ret = ath11k_dp_tx_htt_h2t_ver_req_msg(ab);
 	if (ret) {
 		ath11k_err(ab, "failed to send htt version request message: %d\n",
 			   ret);
-		goto err_mac_destroy;
+		goto err_reo_cleanup;
 	}
 
 	return 0;
 
+err_reo_cleanup:
+	ath11k_dp_pdev_reo_cleanup(ab);
 err_mac_destroy:
 	ath11k_mac_destroy(ab);
 err_hif_stop:
@@ -561,6 +573,7 @@ static int ath11k_core_reconfigure_on_crash(struct ath11k_base *ab)
 	ath11k_dp_pdev_free(ab);
 	ath11k_ahb_stop(ab);
 	ath11k_wmi_detach(ab);
+	ath11k_dp_pdev_reo_cleanup(ab);
 	mutex_unlock(&ab->core_lock);
 
 	ath11k_dp_free(ab);
diff --git a/drivers/net/wireless/ath/ath11k/dp.c b/drivers/net/wireless/ath/ath11k/dp.c
index 72c21cf6a352..b966a16a930f 100644
--- a/drivers/net/wireless/ath/ath11k/dp.c
+++ b/drivers/net/wireless/ath/ath11k/dp.c
@@ -684,11 +684,10 @@ void ath11k_dp_pdev_free(struct ath11k_base *ab)
 	}
 }
 
-int ath11k_dp_pdev_alloc(struct ath11k_base *ab)
+void ath11k_dp_pdev_pre_alloc(struct ath11k_base *ab)
 {
 	struct ath11k *ar;
 	struct ath11k_pdev_dp *dp;
-	int ret;
 	int i;
 
 	for (i = 0; i <  ab->num_radios; i++) {
@@ -704,6 +703,13 @@ int ath11k_dp_pdev_alloc(struct ath11k_base *ab)
 		idr_init(&dp->rxdma_mon_buf_ring.bufs_idr);
 		spin_lock_init(&dp->rxdma_mon_buf_ring.idr_lock);
 	}
+}
+
+int ath11k_dp_pdev_alloc(struct ath11k_base *ab)
+{
+	struct ath11k *ar;
+	int ret;
+	int i;
 
 	/* TODO:Per-pdev rx ring unlike tx ring which is mapped to different AC's */
 	for (i = 0; i < ab->num_radios; i++) {
diff --git a/drivers/net/wireless/ath/ath11k/dp.h b/drivers/net/wireless/ath/ath11k/dp.h
index 82ddd7017722..f7e53509ae07 100644
--- a/drivers/net/wireless/ath/ath11k/dp.h
+++ b/drivers/net/wireless/ath/ath11k/dp.h
@@ -1507,6 +1507,7 @@ void ath11k_dp_vdev_tx_attach(struct ath11k *ar, struct ath11k_vif *arvif);
 void ath11k_dp_free(struct ath11k_base *ab);
 int ath11k_dp_alloc(struct ath11k_base *ab);
 int ath11k_dp_pdev_alloc(struct ath11k_base *ab);
+void ath11k_dp_pdev_pre_alloc(struct ath11k_base *ab);
 void ath11k_dp_pdev_free(struct ath11k_base *ab);
 int ath11k_dp_tx_htt_srng_setup(struct ath11k_base *ab, u32 ring_id,
 				int mac_id, enum hal_ring_type ring_type);
diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c b/drivers/net/wireless/ath/ath11k/dp_rx.c
index a380efb9caa6..1b3b65c0038c 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
@@ -395,12 +395,51 @@ static void ath11k_dp_rx_pdev_srng_free(struct ath11k *ar)
 	struct ath11k_pdev_dp *dp = &ar->dp;
 
 	ath11k_dp_srng_cleanup(ar->ab, &dp->rx_refill_buf_ring.refill_buf_ring);
-	ath11k_dp_srng_cleanup(ar->ab, &dp->reo_dst_ring);
 	ath11k_dp_srng_cleanup(ar->ab, &dp->rxdma_err_dst_ring);
 	ath11k_dp_srng_cleanup(ar->ab, &dp->rx_mon_status_refill_ring.refill_buf_ring);
 	ath11k_dp_srng_cleanup(ar->ab, &dp->rxdma_mon_buf_ring.refill_buf_ring);
 }
 
+void ath11k_dp_pdev_reo_cleanup(struct ath11k_base *ab)
+{
+	struct ath11k_pdev_dp *dp;
+	struct ath11k *ar;
+	int i;
+
+	for (i = 0; i < ab->num_radios; i++) {
+		ar = ab->pdevs[i].ar;
+		dp = &ar->dp;
+		ath11k_dp_srng_cleanup(ab, &dp->reo_dst_ring);
+	}
+}
+
+int ath11k_dp_pdev_reo_setup(struct ath11k_base *ab)
+{
+	struct ath11k *ar;
+	struct ath11k_pdev_dp *dp;
+	int ret;
+	int i;
+
+	for (i = 0; i < ab->num_radios; i++) {
+		ar = ab->pdevs[i].ar;
+		dp = &ar->dp;
+		ret = ath11k_dp_srng_setup(ab, &dp->reo_dst_ring, HAL_REO_DST,
+					   dp->mac_id, dp->mac_id,
+					   DP_REO_DST_RING_SIZE);
+		if (ret) {
+			ath11k_warn(ar->ab, "failed to setup reo_dst_ring\n");
+			goto err_reo_cleanup;
+		}
+	}
+
+	return 0;
+
+err_reo_cleanup:
+	ath11k_dp_pdev_reo_cleanup(ab);
+
+	return ret;
+}
+
 static int ath11k_dp_rx_pdev_srng_alloc(struct ath11k *ar)
 {
 	struct ath11k_pdev_dp *dp = &ar->dp;
@@ -416,14 +455,6 @@ static int ath11k_dp_rx_pdev_srng_alloc(struct ath11k *ar)
 		return ret;
 	}
 
-	ret = ath11k_dp_srng_setup(ar->ab, &dp->reo_dst_ring, HAL_REO_DST,
-				   dp->mac_id, dp->mac_id,
-				   DP_REO_DST_RING_SIZE);
-	if (ret) {
-		ath11k_warn(ar->ab, "failed to setup reo_dst_ring\n");
-		return ret;
-	}
-
 	ret = ath11k_dp_srng_setup(ar->ab, &dp->rxdma_err_dst_ring,
 				   HAL_RXDMA_DST, 0, dp->mac_id,
 				   DP_RXDMA_ERR_DST_RING_SIZE);
diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.h b/drivers/net/wireless/ath/ath11k/dp_rx.h
index 026820fda568..eec5deaa59ad 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.h
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.h
@@ -48,6 +48,8 @@ int ath11k_peer_rx_tid_setup(struct ath11k *ar, const u8 *peer_mac, int vdev_id,
 			     u8 tid, u32 ba_win_sz, u16 ssn);
 void ath11k_dp_htt_htc_t2h_msg_handler(struct ath11k_base *ab,
 				       struct sk_buff *skb);
+int ath11k_dp_pdev_reo_setup(struct ath11k_base *ab);
+void ath11k_dp_pdev_reo_cleanup(struct ath11k_base *ab);
 int ath11k_dp_rx_pdev_alloc(struct ath11k_base *ab, int pdev_idx);
 void ath11k_dp_rx_pdev_free(struct ath11k_base *ab, int pdev_idx);
 void ath11k_dp_reo_cmd_list_cleanup(struct ath11k_base *ab);
-- 
cgit v1.2.3-59-g8ed1b


From fcaf49d0f2dce08208dfb545571ea02c8da6d97a Mon Sep 17 00:00:00 2001
From: John Crispin <john@phrozen.org>
Date: Thu, 28 Nov 2019 08:22:00 +0000
Subject: ath11k: fix indentation in ath11k_mac_prepare_he_mode()

Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/mac.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index e4dfce6b63f0..978d8768d68a 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -3883,15 +3883,15 @@ ath11k_mac_prepare_he_mode(struct ath11k_pdev *pdev, u32 viftype)
 	hecap_phy_ptr = &cap_band->he_cap_phy_info[0];
 
 	hemode = FIELD_PREP(HE_MODE_SU_TX_BFEE, HE_SU_BFEE_ENABLE) |
-		FIELD_PREP(HE_MODE_SU_TX_BFER, HECAP_PHY_SUBFMR_GET(hecap_phy_ptr)) |
-		FIELD_PREP(HE_MODE_UL_MUMIMO, HECAP_PHY_ULMUMIMO_GET(hecap_phy_ptr));
+		 FIELD_PREP(HE_MODE_SU_TX_BFER, HECAP_PHY_SUBFMR_GET(hecap_phy_ptr)) |
+		 FIELD_PREP(HE_MODE_UL_MUMIMO, HECAP_PHY_ULMUMIMO_GET(hecap_phy_ptr));
 
 	/* TODO WDS and other modes */
 	if (viftype == NL80211_IFTYPE_AP) {
 		hemode |= FIELD_PREP(HE_MODE_MU_TX_BFER,
-			   HECAP_PHY_MUBFMR_GET(hecap_phy_ptr)) |
-			FIELD_PREP(HE_MODE_DL_OFDMA, HE_DL_MUOFDMA_ENABLE) |
-			FIELD_PREP(HE_MODE_UL_OFDMA, HE_UL_MUOFDMA_ENABLE);
+			  HECAP_PHY_MUBFMR_GET(hecap_phy_ptr)) |
+			  FIELD_PREP(HE_MODE_DL_OFDMA, HE_DL_MUOFDMA_ENABLE) |
+			  FIELD_PREP(HE_MODE_UL_OFDMA, HE_UL_MUOFDMA_ENABLE);
 	} else {
 		hemode |= FIELD_PREP(HE_MODE_MU_TX_BFEE, HE_MU_BFEE_ENABLE);
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 3168c19d7eb17a0108a3b60ad8e8c1b18ea05c63 Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.king@canonical.com>
Date: Fri, 29 Nov 2019 17:36:35 +0000
Subject: Bluetooth: btusb: fix memory leak on fw

Currently the error return path when the call to btusb_mtk_hci_wmt_sync
fails does not free fw.  Fix this by returning via the error_release_fw
label that performs the free'ing.

Addresses-Coverity: ("Resource leak")
Fixes: a1c49c434e15 ("Bluetooth: btusb: Add protocol support for MediaTek MT7668U USB devices")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 drivers/bluetooth/btusb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 9b587e662b48..d9cd0677d41c 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -2867,7 +2867,7 @@ static int btusb_mtk_setup_firmware(struct hci_dev *hdev, const char *fwname)
 	err = btusb_mtk_hci_wmt_sync(hdev, &wmt_params);
 	if (err < 0) {
 		bt_dev_err(hdev, "Failed to send wmt rst (%d)", err);
-		return err;
+		goto err_release_fw;
 	}
 
 	/* Wait a few moments for firmware activation done */
-- 
cgit v1.2.3-59-g8ed1b


From cfee8793a74dc3afabb08fc9ed5bbe2045709dbb Mon Sep 17 00:00:00 2001
From: Wen Gong <wgong@codeaurora.org>
Date: Fri, 29 Nov 2019 09:59:24 +0200
Subject: ath10k: enable napi on RX path for sdio

For tcp RX, the quantity of tcp acks to remote is 1/2 of the quantity
of tcp data from remote, then it will have many small length packets
on TX path of sdio bus, then it reduce the RX packets's bandwidth of
tcp.

This patch enable napi on RX path, then the RX packet of tcp will not
feed to tcp stack immeditely from mac80211 since GRO is enabled by
default, it will feed to tcp stack after napi complete, if rx bundle
is enabled, then it will feed to tcp stack one time for each bundle
of RX. For example, RX bundle size is 32, then tcp stack will receive
one large length packet, its length is neary 1500*32, then tcp stack
will send a tcp ack for this large packet, this will reduce the tcp
acks ratio from 1/2 to 1/32. This results in significant performance
improvement for tcp RX.

Tcp rx throughout is 240Mbps without this patch, and it arrive 390Mbps
with this patch. The cpu usage has no obvious difference with and
without NAPI.

call stack for each RX packet on GRO path:
(skb length is about 1500 bytes)
  skb_gro_receive ([kernel.kallsyms])
  tcp4_gro_receive ([kernel.kallsyms])
  inet_gro_receive ([kernel.kallsyms])
  dev_gro_receive ([kernel.kallsyms])
  napi_gro_receive ([kernel.kallsyms])
  ieee80211_deliver_skb ([mac80211])
  ieee80211_rx_handlers ([mac80211])
  ieee80211_prepare_and_rx_handle ([mac80211])
  ieee80211_rx_napi ([mac80211])
  ath10k_htt_rx_proc_rx_ind_hl ([ath10k_core])
  ath10k_htt_rx_pktlog_completion_handler ([ath10k_core])
  ath10k_sdio_napi_poll ([ath10k_sdio])
  net_rx_action ([kernel.kallsyms])
  softirqentry_text_start ([kernel.kallsyms])
  do_softirq ([kernel.kallsyms])

call stack for napi complete and send tcp ack from tcp stack:
(skb length is about 1500*32 bytes)
 _tcp_ack_snd_check ([kernel.kallsyms])
 tcp_v4_do_rcv ([kernel.kallsyms])
 tcp_v4_rcv ([kernel.kallsyms])
 local_deliver_finish ([kernel.kallsyms])
 ip_local_deliver ([kernel.kallsyms])
 ip_rcv_finish ([kernel.kallsyms])
 ip_rcv ([kernel.kallsyms])
 netif_receive_skb_core ([kernel.kallsyms])
 netif_receive_skb_one_core([kernel.kallsyms])
 netif_receive_skb ([kernel.kallsyms])
 netif_receive_skb_internal ([kernel.kallsyms])
 napi_gro_complete ([kernel.kallsyms])
 napi_gro_flush ([kernel.kallsyms])
 napi_complete_done ([kernel.kallsyms])
 ath10k_sdio_napi_poll ([ath10k_sdio])
 net_rx_action ([kernel.kallsyms])
 __softirqentry_text_start ([kernel.kallsyms])
 do_softirq ([kernel.kallsyms])

Tested with QCA6174 SDIO with firmware
WLAN.RMH.4.4.1-00017-QCARMSWP-1.

Signed-off-by: Wen Gong <wgong@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/core.c   |  2 ++
 drivers/net/wireless/ath/ath10k/htt.h    |  3 ++
 drivers/net/wireless/ath/ath10k/htt_rx.c | 48 ++++++++++++++++++++++++++------
 drivers/net/wireless/ath/ath10k/sdio.c   | 28 +++++++++++++++++++
 4 files changed, 73 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c
index 6da377317d1f..5ec16ce19b69 100644
--- a/drivers/net/wireless/ath/ath10k/core.c
+++ b/drivers/net/wireless/ath/ath10k/core.c
@@ -3220,6 +3220,8 @@ struct ath10k *ath10k_core_create(size_t priv_size, struct device *dev,
 	init_waitqueue_head(&ar->htt.empty_tx_wq);
 	init_waitqueue_head(&ar->wmi.tx_credits_wq);
 
+	skb_queue_head_init(&ar->htt.rx_indication_head);
+
 	init_completion(&ar->offchan_tx_completed);
 	INIT_WORK(&ar->offchan_tx_work, ath10k_offchan_tx_work);
 	skb_queue_head_init(&ar->offchan_tx_queue);
diff --git a/drivers/net/wireless/ath/ath10k/htt.h b/drivers/net/wireless/ath/ath10k/htt.h
index 30c080094af1..4a12564fc30e 100644
--- a/drivers/net/wireless/ath/ath10k/htt.h
+++ b/drivers/net/wireless/ath/ath10k/htt.h
@@ -1869,6 +1869,8 @@ struct ath10k_htt {
 	struct ath10k *ar;
 	enum ath10k_htc_ep_id eid;
 
+	struct sk_buff_head rx_indication_head;
+
 	u8 target_version_major;
 	u8 target_version_minor;
 	struct completion target_version_received;
@@ -2283,6 +2285,7 @@ int ath10k_htt_mgmt_tx(struct ath10k_htt *htt, struct sk_buff *msdu);
 void ath10k_htt_rx_pktlog_completion_handler(struct ath10k *ar,
 					     struct sk_buff *skb);
 int ath10k_htt_txrx_compl_task(struct ath10k *ar, int budget);
+int ath10k_htt_rx_hl_indication(struct ath10k *ar, int budget);
 void ath10k_htt_set_tx_ops(struct ath10k_htt *htt);
 void ath10k_htt_set_rx_ops(struct ath10k_htt *htt);
 #endif
diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c
index bbcfa728917d..7fb15180c685 100644
--- a/drivers/net/wireless/ath/ath10k/htt_rx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_rx.c
@@ -2359,7 +2359,10 @@ static bool ath10k_htt_rx_proc_rx_ind_hl(struct ath10k_htt *htt,
 		memcpy(skb->data + offset, &qos_ctrl, IEEE80211_QOS_CTL_LEN);
 	}
 
-	ieee80211_rx_ni(ar->hw, skb);
+	if (ar->napi.dev)
+		ieee80211_rx_napi(ar->hw, NULL, skb, &ar->napi);
+	else
+		ieee80211_rx_ni(ar->hw, skb);
 
 	/* We have delivered the skb to the upper layers (mac80211) so we
 	 * must not free it.
@@ -3760,14 +3763,12 @@ bool ath10k_htt_t2h_msg_handler(struct ath10k *ar, struct sk_buff *skb)
 		break;
 	}
 	case HTT_T2H_MSG_TYPE_RX_IND:
-		if (ar->bus_param.dev_type == ATH10K_DEV_TYPE_HL)
-			return ath10k_htt_rx_proc_rx_ind_hl(htt,
-							    &resp->rx_ind_hl,
-							    skb,
-							    HTT_RX_PN_CHECK,
-							    HTT_RX_NON_TKIP_MIC);
-		else
+		if (ar->bus_param.dev_type != ATH10K_DEV_TYPE_HL) {
 			ath10k_htt_rx_proc_rx_ind_ll(htt, &resp->rx_ind);
+		} else {
+			skb_queue_tail(&htt->rx_indication_head, skb);
+			return false;
+		}
 		break;
 	case HTT_T2H_MSG_TYPE_PEER_MAP: {
 		struct htt_peer_map_event ev = {
@@ -3957,6 +3958,37 @@ static int ath10k_htt_rx_deliver_msdu(struct ath10k *ar, int quota, int budget)
 	return quota;
 }
 
+int ath10k_htt_rx_hl_indication(struct ath10k *ar, int budget)
+{
+	struct htt_resp *resp;
+	struct ath10k_htt *htt = &ar->htt;
+	struct sk_buff *skb;
+	bool release;
+	int quota;
+
+	for (quota = 0; quota < budget; quota++) {
+		skb = skb_dequeue(&htt->rx_indication_head);
+		if (!skb)
+			break;
+
+		resp = (struct htt_resp *)skb->data;
+
+		release = ath10k_htt_rx_proc_rx_ind_hl(htt,
+						       &resp->rx_ind_hl,
+						       skb,
+						       HTT_RX_PN_CHECK,
+						       HTT_RX_NON_TKIP_MIC);
+
+		if (release)
+			dev_kfree_skb_any(skb);
+
+		ath10k_dbg(ar, ATH10K_DBG_HTT, "rx indication poll pending count:%d\n",
+			   skb_queue_len(&htt->rx_indication_head));
+	}
+	return quota;
+}
+EXPORT_SYMBOL(ath10k_htt_rx_hl_indication);
+
 int ath10k_htt_txrx_compl_task(struct ath10k *ar, int budget)
 {
 	struct ath10k_htt *htt = &ar->htt;
diff --git a/drivers/net/wireless/ath/ath10k/sdio.c b/drivers/net/wireless/ath/ath10k/sdio.c
index b8e249909699..e36deda23644 100644
--- a/drivers/net/wireless/ath/ath10k/sdio.c
+++ b/drivers/net/wireless/ath/ath10k/sdio.c
@@ -1339,6 +1339,9 @@ static void ath10k_rx_indication_async_work(struct work_struct *work)
 		ep = &ar->htc.endpoint[cb->eid];
 		ep->ep_ops.ep_rx_complete(ar, skb);
 	}
+
+	if (test_bit(ATH10K_FLAG_CORE_REGISTERED, &ar->dev_flags))
+		napi_schedule(&ar->napi);
 }
 
 static void ath10k_sdio_write_async_work(struct work_struct *work)
@@ -1729,6 +1732,8 @@ static int ath10k_sdio_hif_start(struct ath10k *ar)
 	struct ath10k_sdio *ar_sdio = ath10k_sdio_priv(ar);
 	int ret;
 
+	napi_enable(&ar->napi);
+
 	/* Sleep 20 ms before HIF interrupts are disabled.
 	 * This will give target plenty of time to process the BMI done
 	 * request before interrupts are disabled.
@@ -1853,6 +1858,9 @@ static void ath10k_sdio_hif_stop(struct ath10k *ar)
 	}
 
 	spin_unlock_bh(&ar_sdio->wr_async_lock);
+
+	napi_synchronize(&ar->napi);
+	napi_disable(&ar->napi);
 }
 
 #ifdef CONFIG_PM
@@ -2047,6 +2055,20 @@ static SIMPLE_DEV_PM_OPS(ath10k_sdio_pm_ops, ath10k_sdio_pm_suspend,
 
 #endif /* CONFIG_PM_SLEEP */
 
+static int ath10k_sdio_napi_poll(struct napi_struct *ctx, int budget)
+{
+	struct ath10k *ar = container_of(ctx, struct ath10k, napi);
+	int done;
+
+	done = ath10k_htt_rx_hl_indication(ar, budget);
+	ath10k_dbg(ar, ATH10K_DBG_SDIO, "napi poll: done: %d, budget:%d\n", done, budget);
+
+	if (done < budget)
+		napi_complete_done(ctx, done);
+
+	return done;
+}
+
 static int ath10k_sdio_probe(struct sdio_func *func,
 			     const struct sdio_device_id *id)
 {
@@ -2072,6 +2094,9 @@ static int ath10k_sdio_probe(struct sdio_func *func,
 		return -ENOMEM;
 	}
 
+	netif_napi_add(&ar->napi_dev, &ar->napi, ath10k_sdio_napi_poll,
+		       ATH10K_NAPI_BUDGET);
+
 	ath10k_dbg(ar, ATH10K_DBG_BOOT,
 		   "sdio new func %d vendor 0x%x device 0x%x block 0x%x/0x%x\n",
 		   func->num, func->vendor, func->device,
@@ -2184,6 +2209,9 @@ static void ath10k_sdio_remove(struct sdio_func *func)
 		   func->num, func->vendor, func->device);
 
 	ath10k_core_unregister(ar);
+
+	netif_napi_del(&ar->napi);
+
 	ath10k_core_destroy(ar);
 
 	flush_workqueue(ar_sdio->workqueue);
-- 
cgit v1.2.3-59-g8ed1b


From 4a99124516307e0fc4ed7517c0af796091a159d5 Mon Sep 17 00:00:00 2001
From: Wen Gong <wgong@codeaurora.org>
Date: Fri, 29 Nov 2019 09:59:59 +0200
Subject: ath10k: change bundle count for max rx bundle for sdio

For max bundle size 32, the bundle mask is not same with 8/16.
Change it to match the max bundle size of htc. Otherwise it
will not match with firmware, for example, when bundle count
is 17, then flags of ath10k_htc_hdr is 0x4, if without this
patch, it will be considered as non-bundled packet because it
does not have mask 0xF0, then trigger error message later:
payload length 56747 exceeds max htc length: 4088.

htc->max_msgs_per_htc_bundle is the min value of
HTC_HOST_MAX_MSG_PER_RX_BUNDLE and
msg->ready_ext.max_msgs_per_htc_bundle of ath10k_htc_wait_target,
it will be sent to firmware later in ath10k_htc_start, then
firmware will use it as the final max rx bundle count, in
WLAN.RMH.4.4.1-00029, msg->ready_ext.max_msgs_per_htc_bundle
is 32, it is same with HTC_HOST_MAX_MSG_PER_RX_BUNDLE, so the
final max rx bundle count will be set to 32 in firmware.

This patch only effect sdio chips.

Tested with QCA6174 SDIO with firmware WLAN.RMH.4.4.1-00029.

Signed-off-by: Wen Gong <wgong@codeaurora.org>
Fixes: 224776520ead69e ("ath10k: change max RX bundle size from 8 to 32 for sdio")
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/htc.c  |  2 +-
 drivers/net/wireless/ath/ath10k/htc.h  | 10 ++++++----
 drivers/net/wireless/ath/ath10k/sdio.c | 10 ++++++----
 3 files changed, 13 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/htc.c b/drivers/net/wireless/ath/ath10k/htc.c
index 0bbd1415d85c..2248d6c022f4 100644
--- a/drivers/net/wireless/ath/ath10k/htc.c
+++ b/drivers/net/wireless/ath/ath10k/htc.c
@@ -270,7 +270,7 @@ ath10k_htc_process_lookahead_bundle(struct ath10k_htc *htc,
 	struct ath10k *ar = htc->ar;
 	int bundle_cnt = len / sizeof(*report);
 
-	if (!bundle_cnt || (bundle_cnt > HTC_HOST_MAX_MSG_PER_RX_BUNDLE)) {
+	if (!bundle_cnt || (bundle_cnt > htc->max_msgs_per_htc_bundle)) {
 		ath10k_warn(ar, "Invalid lookahead bundle count: %d\n",
 			    bundle_cnt);
 		return -EINVAL;
diff --git a/drivers/net/wireless/ath/ath10k/htc.h b/drivers/net/wireless/ath/ath10k/htc.h
index 16102175e07a..065c82d9d689 100644
--- a/drivers/net/wireless/ath/ath10k/htc.h
+++ b/drivers/net/wireless/ath/ath10k/htc.h
@@ -58,13 +58,15 @@ enum ath10k_htc_rx_flags {
 #define ATH10K_HTC_BUNDLE_EXTRA_MASK GENMASK(3, 2)
 #define ATH10K_HTC_BUNDLE_EXTRA_SHIFT 4
 
-static inline unsigned int ath10k_htc_get_bundle_count(u8 flags)
+static inline unsigned int ath10k_htc_get_bundle_count(u8 max_msgs, u8 flags)
 {
-	unsigned int count, extra_count;
+	unsigned int count, extra_count = 0;
 
 	count = FIELD_GET(ATH10K_HTC_FLAG_BUNDLE_MASK, flags);
-	extra_count = FIELD_GET(ATH10K_HTC_BUNDLE_EXTRA_MASK, flags) <<
-		ATH10K_HTC_BUNDLE_EXTRA_SHIFT;
+
+	if (max_msgs > 16)
+		extra_count = FIELD_GET(ATH10K_HTC_BUNDLE_EXTRA_MASK, flags) <<
+			ATH10K_HTC_BUNDLE_EXTRA_SHIFT;
 
 	return count + extra_count;
 }
diff --git a/drivers/net/wireless/ath/ath10k/sdio.c b/drivers/net/wireless/ath/ath10k/sdio.c
index e36deda23644..6fdf71b8b676 100644
--- a/drivers/net/wireless/ath/ath10k/sdio.c
+++ b/drivers/net/wireless/ath/ath10k/sdio.c
@@ -500,14 +500,15 @@ static int ath10k_sdio_mbox_alloc_bundle(struct ath10k *ar,
 					 size_t *bndl_cnt)
 {
 	int ret, i;
+	u8 max_msgs = ar->htc.max_msgs_per_htc_bundle;
 
-	*bndl_cnt = ath10k_htc_get_bundle_count(htc_hdr->flags);
+	*bndl_cnt = ath10k_htc_get_bundle_count(max_msgs, htc_hdr->flags);
 
-	if (*bndl_cnt > HTC_HOST_MAX_MSG_PER_RX_BUNDLE) {
+	if (*bndl_cnt > max_msgs) {
 		ath10k_warn(ar,
 			    "HTC bundle length %u exceeds maximum %u\n",
 			    le16_to_cpu(htc_hdr->len),
-			    HTC_HOST_MAX_MSG_PER_RX_BUNDLE);
+			    max_msgs);
 		return -ENOMEM;
 	}
 
@@ -570,7 +571,8 @@ static int ath10k_sdio_mbox_rx_alloc(struct ath10k *ar,
 			goto err;
 		}
 
-		if (htc_hdr->flags & ATH10K_HTC_FLAG_BUNDLE_MASK) {
+		if (ath10k_htc_get_bundle_count(
+			ar->htc.max_msgs_per_htc_bundle, htc_hdr->flags)) {
 			/* HTC header indicates that every packet to follow
 			 * has the same padded length so that it can be
 			 * optimally fetched as a full bundle.
-- 
cgit v1.2.3-59-g8ed1b


From 2dc016599cfa9672a147528ca26d70c3654a5423 Mon Sep 17 00:00:00 2001
From: Wen Gong <wgong@codeaurora.org>
Date: Fri, 29 Nov 2019 07:34:09 +0000
Subject: ath: add support for special 0x0 regulatory domain

Some sdio chips of rome QCA6174's regulatory domain code of EEPROM is
empty, then ath_is_world_regd will return false for this case, and
it will lead function __ath_reg_dyn_country not work, thus the regdomain
will not update for NL80211_REGDOM_SET_BY_COUNTRY_IE type, it result
ath10k set the same regdomain/reg_5ghz_ctl/reg_2ghz_ctl to firmware,
then the tx power will not changed with different regdomain's AP. The
regulatory domain code of EEPROM of some QCA6174 PCIE chip is 0x6c, it
means world wide regdomain, for this chip, it does not have the issue.

For empty reulatory domain code chip, set it to world regulatory domain
in functio ath_regd_sanitize, then it will fix the issue.

Tested with QCA6174 SDIO with firmware
WLAN.RMH.4.4.1-00029.

Signed-off-by: Wen Gong <wgong@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/regd.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/ath/regd.c b/drivers/net/wireless/ath/regd.c
index 20f4f8ea9f89..bee9110b91f3 100644
--- a/drivers/net/wireless/ath/regd.c
+++ b/drivers/net/wireless/ath/regd.c
@@ -666,14 +666,14 @@ ath_regd_init_wiphy(struct ath_regulatory *reg,
 
 /*
  * Some users have reported their EEPROM programmed with
- * 0x8000 set, this is not a supported regulatory domain
- * but since we have more than one user with it we need
- * a solution for them. We default to 0x64, which is the
- * default Atheros world regulatory domain.
+ * 0x8000 or 0x0 set, this is not a supported regulatory
+ * domain but since we have more than one user with it we
+ * need a solution for them. We default to 0x64, which is
+ * the default Atheros world regulatory domain.
  */
 static void ath_regd_sanitize(struct ath_regulatory *reg)
 {
-	if (reg->current_rd != COUNTRY_ERD_FLAG)
+	if (reg->current_rd != COUNTRY_ERD_FLAG && reg->current_rd != 0)
 		return;
 	printk(KERN_DEBUG "ath: EEPROM regdomain sanitized\n");
 	reg->current_rd = 0x64;
-- 
cgit v1.2.3-59-g8ed1b


From 7ecacafc240638148567742cca41aa7144b4fe1e Mon Sep 17 00:00:00 2001
From: Kai-Heng Feng <kai.heng.feng@canonical.com>
Date: Thu, 5 Dec 2019 17:07:01 +0800
Subject: Bluetooth: btusb: Disable runtime suspend on Realtek devices

After commit 9e45524a0111 ("Bluetooth: btusb: Fix suspend issue for
Realtek devices") both WiFi and Bluetooth stop working after reboot:
[   34.322617] usb 1-8: reset full-speed USB device number 3 using xhci_hcd
[   34.450401] usb 1-8: device descriptor read/64, error -71
[   34.694375] usb 1-8: device descriptor read/64, error -71
...
[   44.599111] rtw_pci 0000:02:00.0: failed to poll offset=0x5 mask=0x3 value=0x0
[   44.599113] rtw_pci 0000:02:00.0: mac power on failed
[   44.599114] rtw_pci 0000:02:00.0: failed to power on mac
[   44.599114] rtw_pci 0000:02:00.0: leave idle state failed
[   44.599492] rtw_pci 0000:02:00.0: failed to leave ips state
[   44.599493] rtw_pci 0000:02:00.0: failed to leave idle state

That commit removed USB_QUIRK_RESET_RESUME, which not only resets the USB
device after resume, it also prevents the device from being runtime
suspended by USB core. My experiment shows if the Realtek btusb device
ever runtime suspends once, the entire wireless module becomes useless
after reboot.

So let's explicitly disable runtime suspend on Realtek btusb device for
now.

Fixes: 9e45524a0111 ("Bluetooth: btusb: Fix suspend issue for Realtek devices")
Signed-off-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 drivers/bluetooth/btusb.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index d9cd0677d41c..0eaeca0a64fb 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -3832,6 +3832,10 @@ static int btusb_probe(struct usb_interface *intf,
 		 * (DEVICE_REMOTE_WAKEUP)
 		 */
 		set_bit(BTUSB_WAKEUP_DISABLE, &data->flags);
+
+		err = usb_autopm_get_interface(intf);
+		if (err < 0)
+			goto out_free_dev;
 	}
 
 	if (id->driver_info & BTUSB_AMP) {
-- 
cgit v1.2.3-59-g8ed1b


From 2b1aa5a4c654cc665eb5711f49a2855ea6ebe0db Mon Sep 17 00:00:00 2001
From: René Treffer <treffer@measite.de>
Date: Tue, 26 Nov 2019 10:27:38 +0800
Subject: batman-adv: ELP - use wifi tx bitrate as fallback throughput
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Some wifi drivers (e.g. ath10k) provide per-station rx/tx values but no
estimated throughput. Setting a better estimate than the default 1 MBit
makes these devices work well with B.A.T.M.A.N. V.

Signed-off-by: René Treffer <treffer@measite.de>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Simon Wunderlich <sw@simonwunderlich.de>
---
 net/batman-adv/bat_v_elp.c | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

diff --git a/net/batman-adv/bat_v_elp.c b/net/batman-adv/bat_v_elp.c
index 2614a9caee00..6536e8cc53a0 100644
--- a/net/batman-adv/bat_v_elp.c
+++ b/net/batman-adv/bat_v_elp.c
@@ -107,10 +107,17 @@ static u32 batadv_v_elp_get_throughput(struct batadv_hardif_neigh_node *neigh)
 		}
 		if (ret)
 			goto default_throughput;
-		if (!(sinfo.filled & BIT(NL80211_STA_INFO_EXPECTED_THROUGHPUT)))
-			goto default_throughput;
 
-		return sinfo.expected_throughput / 100;
+		if (sinfo.filled & BIT(NL80211_STA_INFO_EXPECTED_THROUGHPUT))
+			return sinfo.expected_throughput / 100;
+
+		/* try to estimate the expected throughput based on reported tx
+		 * rates
+		 */
+		if (sinfo.filled & BIT(NL80211_STA_INFO_TX_BITRATE))
+			return cfg80211_calculate_bitrate(&sinfo.txrate) / 3;
+
+		goto default_throughput;
 	}
 
 	/* if not a wifi interface, check if this device provides data via
-- 
cgit v1.2.3-59-g8ed1b


From 61a292860de6ab66c9b6aa1b244ccd4590ddc8ec Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Thu, 28 Nov 2019 11:26:06 +0100
Subject: batman-adv: Annotate bitwise integer pointer casts

The sparse commit 6002ded74587 ("add a flag to warn on casts to/from
bitwise pointers") introduced a check for non-direct casts from/to
restricted datatypes (when -Wbitwise-pointer is enabled).

This triggered various warnings in batman-adv when some (already big
endian) buffer content was casted to/from the corresponding big endian
integer data types. But these were correct and can therefore be marked with
__force to signalize sparse an intended cast from/to a bitwise type.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Simon Wunderlich <sw@simonwunderlich.de>
---
 net/batman-adv/bridge_loop_avoidance.c | 2 +-
 net/batman-adv/distributed-arp-table.c | 8 +++++---
 2 files changed, 6 insertions(+), 4 deletions(-)

diff --git a/net/batman-adv/bridge_loop_avoidance.c b/net/batman-adv/bridge_loop_avoidance.c
index 663a53b6d36e..0eff33580f95 100644
--- a/net/batman-adv/bridge_loop_avoidance.c
+++ b/net/batman-adv/bridge_loop_avoidance.c
@@ -844,7 +844,7 @@ static bool batadv_handle_announce(struct batadv_priv *bat_priv, u8 *an_addr,
 
 	/* handle as ANNOUNCE frame */
 	backbone_gw->lasttime = jiffies;
-	crc = ntohs(*((__be16 *)(&an_addr[4])));
+	crc = ntohs(*((__force __be16 *)(&an_addr[4])));
 
 	batadv_dbg(BATADV_DBG_BLA, bat_priv,
 		   "%s(): ANNOUNCE vid %d (sent by %pM)... CRC = %#.4x\n",
diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c
index b0af3a11d406..5004e38fe792 100644
--- a/net/batman-adv/distributed-arp-table.c
+++ b/net/batman-adv/distributed-arp-table.c
@@ -246,7 +246,7 @@ static u8 *batadv_arp_hw_src(struct sk_buff *skb, int hdr_size)
  */
 static __be32 batadv_arp_ip_src(struct sk_buff *skb, int hdr_size)
 {
-	return *(__be32 *)(batadv_arp_hw_src(skb, hdr_size) + ETH_ALEN);
+	return *(__force __be32 *)(batadv_arp_hw_src(skb, hdr_size) + ETH_ALEN);
 }
 
 /**
@@ -270,7 +270,9 @@ static u8 *batadv_arp_hw_dst(struct sk_buff *skb, int hdr_size)
  */
 static __be32 batadv_arp_ip_dst(struct sk_buff *skb, int hdr_size)
 {
-	return *(__be32 *)(batadv_arp_hw_src(skb, hdr_size) + ETH_ALEN * 2 + 4);
+	u8 *dst = batadv_arp_hw_src(skb, hdr_size) + ETH_ALEN * 2 + 4;
+
+	return *(__force __be32 *)dst;
 }
 
 /**
@@ -287,7 +289,7 @@ static u32 batadv_hash_dat(const void *data, u32 size)
 	const unsigned char *key;
 	u32 i;
 
-	key = (const unsigned char *)&dat->ip;
+	key = (__force const unsigned char *)&dat->ip;
 	for (i = 0; i < sizeof(dat->ip); i++) {
 		hash += key[i];
 		hash += (hash << 10);
-- 
cgit v1.2.3-59-g8ed1b


From e7096c131e5161fa3b8e52a650d7719d2857adfd Mon Sep 17 00:00:00 2001
From: "Jason A. Donenfeld" <Jason@zx2c4.com>
Date: Mon, 9 Dec 2019 00:27:34 +0100
Subject: net: WireGuard secure network tunnel

WireGuard is a layer 3 secure networking tunnel made specifically for
the kernel, that aims to be much simpler and easier to audit than IPsec.
Extensive documentation and description of the protocol and
considerations, along with formal proofs of the cryptography, are
available at:

  * https://www.wireguard.com/
  * https://www.wireguard.com/papers/wireguard.pdf

This commit implements WireGuard as a simple network device driver,
accessible in the usual RTNL way used by virtual network drivers. It
makes use of the udp_tunnel APIs, GRO, GSO, NAPI, and the usual set of
networking subsystem APIs. It has a somewhat novel multicore queueing
system designed for maximum throughput and minimal latency of encryption
operations, but it is implemented modestly using workqueues and NAPI.
Configuration is done via generic Netlink, and following a review from
the Netlink maintainer a year ago, several high profile userspace tools
have already implemented the API.

This commit also comes with several different tests, both in-kernel
tests and out-of-kernel tests based on network namespaces, taking profit
of the fact that sockets used by WireGuard intentionally stay in the
namespace the WireGuard interface was originally created, exactly like
the semantics of userspace tun devices. See wireguard.com/netns/ for
pictures and examples.

The source code is fairly short, but rather than combining everything
into a single file, WireGuard is developed as cleanly separable files,
making auditing and comprehension easier. Things are laid out as
follows:

  * noise.[ch], cookie.[ch], messages.h: These implement the bulk of the
    cryptographic aspects of the protocol, and are mostly data-only in
    nature, taking in buffers of bytes and spitting out buffers of
    bytes. They also handle reference counting for their various shared
    pieces of data, like keys and key lists.

  * ratelimiter.[ch]: Used as an integral part of cookie.[ch] for
    ratelimiting certain types of cryptographic operations in accordance
    with particular WireGuard semantics.

  * allowedips.[ch], peerlookup.[ch]: The main lookup structures of
    WireGuard, the former being trie-like with particular semantics, an
    integral part of the design of the protocol, and the latter just
    being nice helper functions around the various hashtables we use.

  * device.[ch]: Implementation of functions for the netdevice and for
    rtnl, responsible for maintaining the life of a given interface and
    wiring it up to the rest of WireGuard.

  * peer.[ch]: Each interface has a list of peers, with helper functions
    available here for creation, destruction, and reference counting.

  * socket.[ch]: Implementation of functions related to udp_socket and
    the general set of kernel socket APIs, for sending and receiving
    ciphertext UDP packets, and taking care of WireGuard-specific sticky
    socket routing semantics for the automatic roaming.

  * netlink.[ch]: Userspace API entry point for configuring WireGuard
    peers and devices. The API has been implemented by several userspace
    tools and network management utility, and the WireGuard project
    distributes the basic wg(8) tool.

  * queueing.[ch]: Shared function on the rx and tx path for handling
    the various queues used in the multicore algorithms.

  * send.c: Handles encrypting outgoing packets in parallel on
    multiple cores, before sending them in order on a single core, via
    workqueues and ring buffers. Also handles sending handshake and cookie
    messages as part of the protocol, in parallel.

  * receive.c: Handles decrypting incoming packets in parallel on
    multiple cores, before passing them off in order to be ingested via
    the rest of the networking subsystem with GRO via the typical NAPI
    poll function. Also handles receiving handshake and cookie messages
    as part of the protocol, in parallel.

  * timers.[ch]: Uses the timer wheel to implement protocol particular
    event timeouts, and gives a set of very simple event-driven entry
    point functions for callers.

  * main.c, version.h: Initialization and deinitialization of the module.

  * selftest/*.h: Runtime unit tests for some of the most security
    sensitive functions.

  * tools/testing/selftests/wireguard/netns.sh: Aforementioned testing
    script using network namespaces.

This commit aims to be as self-contained as possible, implementing
WireGuard as a standalone module not needing much special handling or
coordination from the network subsystem. I expect for future
optimizations to the network stack to positively improve WireGuard, and
vice-versa, but for the time being, this exists as intentionally
standalone.

We introduce a menu option for CONFIG_WIREGUARD, as well as providing a
verbose debug log and self-tests via CONFIG_WIREGUARD_DEBUG.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Cc: David Miller <davem@davemloft.net>
Cc: Greg KH <gregkh@linuxfoundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: linux-crypto@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: netdev@vger.kernel.org
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 MAINTAINERS                                  |   8 +
 drivers/net/Kconfig                          |  41 ++
 drivers/net/Makefile                         |   1 +
 drivers/net/wireguard/Makefile               |  18 +
 drivers/net/wireguard/allowedips.c           | 381 ++++++++++++
 drivers/net/wireguard/allowedips.h           |  59 ++
 drivers/net/wireguard/cookie.c               | 236 ++++++++
 drivers/net/wireguard/cookie.h               |  59 ++
 drivers/net/wireguard/device.c               | 458 +++++++++++++++
 drivers/net/wireguard/device.h               |  73 +++
 drivers/net/wireguard/main.c                 |  64 +++
 drivers/net/wireguard/messages.h             | 128 +++++
 drivers/net/wireguard/netlink.c              | 642 +++++++++++++++++++++
 drivers/net/wireguard/netlink.h              |  12 +
 drivers/net/wireguard/noise.c                | 828 +++++++++++++++++++++++++++
 drivers/net/wireguard/noise.h                | 137 +++++
 drivers/net/wireguard/peer.c                 | 240 ++++++++
 drivers/net/wireguard/peer.h                 |  83 +++
 drivers/net/wireguard/peerlookup.c           | 221 +++++++
 drivers/net/wireguard/peerlookup.h           |  64 +++
 drivers/net/wireguard/queueing.c             |  53 ++
 drivers/net/wireguard/queueing.h             | 197 +++++++
 drivers/net/wireguard/ratelimiter.c          | 223 ++++++++
 drivers/net/wireguard/ratelimiter.h          |  19 +
 drivers/net/wireguard/receive.c              | 595 +++++++++++++++++++
 drivers/net/wireguard/selftest/allowedips.c  | 683 ++++++++++++++++++++++
 drivers/net/wireguard/selftest/counter.c     | 104 ++++
 drivers/net/wireguard/selftest/ratelimiter.c | 226 ++++++++
 drivers/net/wireguard/send.c                 | 413 +++++++++++++
 drivers/net/wireguard/socket.c               | 437 ++++++++++++++
 drivers/net/wireguard/socket.h               |  44 ++
 drivers/net/wireguard/timers.c               | 243 ++++++++
 drivers/net/wireguard/timers.h               |  31 +
 drivers/net/wireguard/version.h              |   1 +
 include/uapi/linux/wireguard.h               | 196 +++++++
 tools/testing/selftests/wireguard/netns.sh   | 537 +++++++++++++++++
 36 files changed, 7755 insertions(+)
 create mode 100644 drivers/net/wireguard/Makefile
 create mode 100644 drivers/net/wireguard/allowedips.c
 create mode 100644 drivers/net/wireguard/allowedips.h
 create mode 100644 drivers/net/wireguard/cookie.c
 create mode 100644 drivers/net/wireguard/cookie.h
 create mode 100644 drivers/net/wireguard/device.c
 create mode 100644 drivers/net/wireguard/device.h
 create mode 100644 drivers/net/wireguard/main.c
 create mode 100644 drivers/net/wireguard/messages.h
 create mode 100644 drivers/net/wireguard/netlink.c
 create mode 100644 drivers/net/wireguard/netlink.h
 create mode 100644 drivers/net/wireguard/noise.c
 create mode 100644 drivers/net/wireguard/noise.h
 create mode 100644 drivers/net/wireguard/peer.c
 create mode 100644 drivers/net/wireguard/peer.h
 create mode 100644 drivers/net/wireguard/peerlookup.c
 create mode 100644 drivers/net/wireguard/peerlookup.h
 create mode 100644 drivers/net/wireguard/queueing.c
 create mode 100644 drivers/net/wireguard/queueing.h
 create mode 100644 drivers/net/wireguard/ratelimiter.c
 create mode 100644 drivers/net/wireguard/ratelimiter.h
 create mode 100644 drivers/net/wireguard/receive.c
 create mode 100644 drivers/net/wireguard/selftest/allowedips.c
 create mode 100644 drivers/net/wireguard/selftest/counter.c
 create mode 100644 drivers/net/wireguard/selftest/ratelimiter.c
 create mode 100644 drivers/net/wireguard/send.c
 create mode 100644 drivers/net/wireguard/socket.c
 create mode 100644 drivers/net/wireguard/socket.h
 create mode 100644 drivers/net/wireguard/timers.c
 create mode 100644 drivers/net/wireguard/timers.h
 create mode 100644 drivers/net/wireguard/version.h
 create mode 100644 include/uapi/linux/wireguard.h
 create mode 100755 tools/testing/selftests/wireguard/netns.sh

diff --git a/MAINTAINERS b/MAINTAINERS
index bd5847e802de..ce7fd2e7aa1a 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -17850,6 +17850,14 @@ L:	linux-gpio@vger.kernel.org
 S:	Maintained
 F:	drivers/gpio/gpio-ws16c48.c
 
+WIREGUARD SECURE NETWORK TUNNEL
+M:	Jason A. Donenfeld <Jason@zx2c4.com>
+S:	Maintained
+F:	drivers/net/wireguard/
+F:	tools/testing/selftests/wireguard/
+L:	wireguard@lists.zx2c4.com
+L:	netdev@vger.kernel.org
+
 WISTRON LAPTOP BUTTON DRIVER
 M:	Miloslav Trmac <mitr@volny.cz>
 S:	Maintained
diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index d02f12a5254e..ffe8d4e2b206 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -71,6 +71,47 @@ config DUMMY
 	  To compile this driver as a module, choose M here: the module
 	  will be called dummy.
 
+config WIREGUARD
+	tristate "WireGuard secure network tunnel"
+	depends on NET && INET
+	depends on IPV6 || !IPV6
+	select NET_UDP_TUNNEL
+	select DST_CACHE
+	select CRYPTO
+	select CRYPTO_LIB_CURVE25519
+	select CRYPTO_LIB_CHACHA20POLY1305
+	select CRYPTO_LIB_BLAKE2S
+	select CRYPTO_CHACHA20_X86_64 if X86 && 64BIT
+	select CRYPTO_POLY1305_X86_64 if X86 && 64BIT
+	select CRYPTO_BLAKE2S_X86 if X86 && 64BIT
+	select CRYPTO_CURVE25519_X86 if X86 && 64BIT
+	select CRYPTO_CHACHA20_NEON if (ARM || ARM64) && KERNEL_MODE_NEON
+	select CRYPTO_POLY1305_NEON if ARM64 && KERNEL_MODE_NEON
+	select CRYPTO_POLY1305_ARM if ARM
+	select CRYPTO_CURVE25519_NEON if ARM && KERNEL_MODE_NEON
+	select CRYPTO_CHACHA_MIPS if CPU_MIPS32_R2
+	select CRYPTO_POLY1305_MIPS if CPU_MIPS32 || (CPU_MIPS64 && 64BIT)
+	help
+	  WireGuard is a secure, fast, and easy to use replacement for IPSec
+	  that uses modern cryptography and clever networking tricks. It's
+	  designed to be fairly general purpose and abstract enough to fit most
+	  use cases, while at the same time remaining extremely simple to
+	  configure. See www.wireguard.com for more info.
+
+	  It's safe to say Y or M here, as the driver is very lightweight and
+	  is only in use when an administrator chooses to add an interface.
+
+config WIREGUARD_DEBUG
+	bool "Debugging checks and verbose messages"
+	depends on WIREGUARD
+	help
+	  This will write log messages for handshake and other events
+	  that occur for a WireGuard interface. It will also perform some
+	  extra validation checks and unit tests at various points. This is
+	  only useful for debugging.
+
+	  Say N here unless you know what you're doing.
+
 config EQUALIZER
 	tristate "EQL (serial line load balancing) support"
 	---help---
diff --git a/drivers/net/Makefile b/drivers/net/Makefile
index 0d3ba056cda3..953b7c12f0b0 100644
--- a/drivers/net/Makefile
+++ b/drivers/net/Makefile
@@ -10,6 +10,7 @@ obj-$(CONFIG_BONDING) += bonding/
 obj-$(CONFIG_IPVLAN) += ipvlan/
 obj-$(CONFIG_IPVTAP) += ipvlan/
 obj-$(CONFIG_DUMMY) += dummy.o
+obj-$(CONFIG_WIREGUARD) += wireguard/
 obj-$(CONFIG_EQUALIZER) += eql.o
 obj-$(CONFIG_IFB) += ifb.o
 obj-$(CONFIG_MACSEC) += macsec.o
diff --git a/drivers/net/wireguard/Makefile b/drivers/net/wireguard/Makefile
new file mode 100644
index 000000000000..fc52b2cb500b
--- /dev/null
+++ b/drivers/net/wireguard/Makefile
@@ -0,0 +1,18 @@
+ccflags-y := -O3
+ccflags-y += -D'pr_fmt(fmt)=KBUILD_MODNAME ": " fmt'
+ccflags-$(CONFIG_WIREGUARD_DEBUG) += -DDEBUG
+wireguard-y := main.o
+wireguard-y += noise.o
+wireguard-y += device.o
+wireguard-y += peer.o
+wireguard-y += timers.o
+wireguard-y += queueing.o
+wireguard-y += send.o
+wireguard-y += receive.o
+wireguard-y += socket.o
+wireguard-y += peerlookup.o
+wireguard-y += allowedips.o
+wireguard-y += ratelimiter.o
+wireguard-y += cookie.o
+wireguard-y += netlink.o
+obj-$(CONFIG_WIREGUARD) := wireguard.o
diff --git a/drivers/net/wireguard/allowedips.c b/drivers/net/wireguard/allowedips.c
new file mode 100644
index 000000000000..72667d5399c3
--- /dev/null
+++ b/drivers/net/wireguard/allowedips.c
@@ -0,0 +1,381 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#include "allowedips.h"
+#include "peer.h"
+
+static void swap_endian(u8 *dst, const u8 *src, u8 bits)
+{
+	if (bits == 32) {
+		*(u32 *)dst = be32_to_cpu(*(const __be32 *)src);
+	} else if (bits == 128) {
+		((u64 *)dst)[0] = be64_to_cpu(((const __be64 *)src)[0]);
+		((u64 *)dst)[1] = be64_to_cpu(((const __be64 *)src)[1]);
+	}
+}
+
+static void copy_and_assign_cidr(struct allowedips_node *node, const u8 *src,
+				 u8 cidr, u8 bits)
+{
+	node->cidr = cidr;
+	node->bit_at_a = cidr / 8U;
+#ifdef __LITTLE_ENDIAN
+	node->bit_at_a ^= (bits / 8U - 1U) % 8U;
+#endif
+	node->bit_at_b = 7U - (cidr % 8U);
+	node->bitlen = bits;
+	memcpy(node->bits, src, bits / 8U);
+}
+#define CHOOSE_NODE(parent, key) \
+	parent->bit[(key[parent->bit_at_a] >> parent->bit_at_b) & 1]
+
+static void node_free_rcu(struct rcu_head *rcu)
+{
+	kfree(container_of(rcu, struct allowedips_node, rcu));
+}
+
+static void push_rcu(struct allowedips_node **stack,
+		     struct allowedips_node __rcu *p, unsigned int *len)
+{
+	if (rcu_access_pointer(p)) {
+		WARN_ON(IS_ENABLED(DEBUG) && *len >= 128);
+		stack[(*len)++] = rcu_dereference_raw(p);
+	}
+}
+
+static void root_free_rcu(struct rcu_head *rcu)
+{
+	struct allowedips_node *node, *stack[128] = {
+		container_of(rcu, struct allowedips_node, rcu) };
+	unsigned int len = 1;
+
+	while (len > 0 && (node = stack[--len])) {
+		push_rcu(stack, node->bit[0], &len);
+		push_rcu(stack, node->bit[1], &len);
+		kfree(node);
+	}
+}
+
+static void root_remove_peer_lists(struct allowedips_node *root)
+{
+	struct allowedips_node *node, *stack[128] = { root };
+	unsigned int len = 1;
+
+	while (len > 0 && (node = stack[--len])) {
+		push_rcu(stack, node->bit[0], &len);
+		push_rcu(stack, node->bit[1], &len);
+		if (rcu_access_pointer(node->peer))
+			list_del(&node->peer_list);
+	}
+}
+
+static void walk_remove_by_peer(struct allowedips_node __rcu **top,
+				struct wg_peer *peer, struct mutex *lock)
+{
+#define REF(p) rcu_access_pointer(p)
+#define DEREF(p) rcu_dereference_protected(*(p), lockdep_is_held(lock))
+#define PUSH(p) ({                                                             \
+		WARN_ON(IS_ENABLED(DEBUG) && len >= 128);                      \
+		stack[len++] = p;                                              \
+	})
+
+	struct allowedips_node __rcu **stack[128], **nptr;
+	struct allowedips_node *node, *prev;
+	unsigned int len;
+
+	if (unlikely(!peer || !REF(*top)))
+		return;
+
+	for (prev = NULL, len = 0, PUSH(top); len > 0; prev = node) {
+		nptr = stack[len - 1];
+		node = DEREF(nptr);
+		if (!node) {
+			--len;
+			continue;
+		}
+		if (!prev || REF(prev->bit[0]) == node ||
+		    REF(prev->bit[1]) == node) {
+			if (REF(node->bit[0]))
+				PUSH(&node->bit[0]);
+			else if (REF(node->bit[1]))
+				PUSH(&node->bit[1]);
+		} else if (REF(node->bit[0]) == prev) {
+			if (REF(node->bit[1]))
+				PUSH(&node->bit[1]);
+		} else {
+			if (rcu_dereference_protected(node->peer,
+				lockdep_is_held(lock)) == peer) {
+				RCU_INIT_POINTER(node->peer, NULL);
+				list_del_init(&node->peer_list);
+				if (!node->bit[0] || !node->bit[1]) {
+					rcu_assign_pointer(*nptr, DEREF(
+					       &node->bit[!REF(node->bit[0])]));
+					call_rcu(&node->rcu, node_free_rcu);
+					node = DEREF(nptr);
+				}
+			}
+			--len;
+		}
+	}
+
+#undef REF
+#undef DEREF
+#undef PUSH
+}
+
+static unsigned int fls128(u64 a, u64 b)
+{
+	return a ? fls64(a) + 64U : fls64(b);
+}
+
+static u8 common_bits(const struct allowedips_node *node, const u8 *key,
+		      u8 bits)
+{
+	if (bits == 32)
+		return 32U - fls(*(const u32 *)node->bits ^ *(const u32 *)key);
+	else if (bits == 128)
+		return 128U - fls128(
+			*(const u64 *)&node->bits[0] ^ *(const u64 *)&key[0],
+			*(const u64 *)&node->bits[8] ^ *(const u64 *)&key[8]);
+	return 0;
+}
+
+static bool prefix_matches(const struct allowedips_node *node, const u8 *key,
+			   u8 bits)
+{
+	/* This could be much faster if it actually just compared the common
+	 * bits properly, by precomputing a mask bswap(~0 << (32 - cidr)), and
+	 * the rest, but it turns out that common_bits is already super fast on
+	 * modern processors, even taking into account the unfortunate bswap.
+	 * So, we just inline it like this instead.
+	 */
+	return common_bits(node, key, bits) >= node->cidr;
+}
+
+static struct allowedips_node *find_node(struct allowedips_node *trie, u8 bits,
+					 const u8 *key)
+{
+	struct allowedips_node *node = trie, *found = NULL;
+
+	while (node && prefix_matches(node, key, bits)) {
+		if (rcu_access_pointer(node->peer))
+			found = node;
+		if (node->cidr == bits)
+			break;
+		node = rcu_dereference_bh(CHOOSE_NODE(node, key));
+	}
+	return found;
+}
+
+/* Returns a strong reference to a peer */
+static struct wg_peer *lookup(struct allowedips_node __rcu *root, u8 bits,
+			      const void *be_ip)
+{
+	/* Aligned so it can be passed to fls/fls64 */
+	u8 ip[16] __aligned(__alignof(u64));
+	struct allowedips_node *node;
+	struct wg_peer *peer = NULL;
+
+	swap_endian(ip, be_ip, bits);
+
+	rcu_read_lock_bh();
+retry:
+	node = find_node(rcu_dereference_bh(root), bits, ip);
+	if (node) {
+		peer = wg_peer_get_maybe_zero(rcu_dereference_bh(node->peer));
+		if (!peer)
+			goto retry;
+	}
+	rcu_read_unlock_bh();
+	return peer;
+}
+
+static bool node_placement(struct allowedips_node __rcu *trie, const u8 *key,
+			   u8 cidr, u8 bits, struct allowedips_node **rnode,
+			   struct mutex *lock)
+{
+	struct allowedips_node *node = rcu_dereference_protected(trie,
+						lockdep_is_held(lock));
+	struct allowedips_node *parent = NULL;
+	bool exact = false;
+
+	while (node && node->cidr <= cidr && prefix_matches(node, key, bits)) {
+		parent = node;
+		if (parent->cidr == cidr) {
+			exact = true;
+			break;
+		}
+		node = rcu_dereference_protected(CHOOSE_NODE(parent, key),
+						 lockdep_is_held(lock));
+	}
+	*rnode = parent;
+	return exact;
+}
+
+static int add(struct allowedips_node __rcu **trie, u8 bits, const u8 *key,
+	       u8 cidr, struct wg_peer *peer, struct mutex *lock)
+{
+	struct allowedips_node *node, *parent, *down, *newnode;
+
+	if (unlikely(cidr > bits || !peer))
+		return -EINVAL;
+
+	if (!rcu_access_pointer(*trie)) {
+		node = kzalloc(sizeof(*node), GFP_KERNEL);
+		if (unlikely(!node))
+			return -ENOMEM;
+		RCU_INIT_POINTER(node->peer, peer);
+		list_add_tail(&node->peer_list, &peer->allowedips_list);
+		copy_and_assign_cidr(node, key, cidr, bits);
+		rcu_assign_pointer(*trie, node);
+		return 0;
+	}
+	if (node_placement(*trie, key, cidr, bits, &node, lock)) {
+		rcu_assign_pointer(node->peer, peer);
+		list_move_tail(&node->peer_list, &peer->allowedips_list);
+		return 0;
+	}
+
+	newnode = kzalloc(sizeof(*newnode), GFP_KERNEL);
+	if (unlikely(!newnode))
+		return -ENOMEM;
+	RCU_INIT_POINTER(newnode->peer, peer);
+	list_add_tail(&newnode->peer_list, &peer->allowedips_list);
+	copy_and_assign_cidr(newnode, key, cidr, bits);
+
+	if (!node) {
+		down = rcu_dereference_protected(*trie, lockdep_is_held(lock));
+	} else {
+		down = rcu_dereference_protected(CHOOSE_NODE(node, key),
+						 lockdep_is_held(lock));
+		if (!down) {
+			rcu_assign_pointer(CHOOSE_NODE(node, key), newnode);
+			return 0;
+		}
+	}
+	cidr = min(cidr, common_bits(down, key, bits));
+	parent = node;
+
+	if (newnode->cidr == cidr) {
+		rcu_assign_pointer(CHOOSE_NODE(newnode, down->bits), down);
+		if (!parent)
+			rcu_assign_pointer(*trie, newnode);
+		else
+			rcu_assign_pointer(CHOOSE_NODE(parent, newnode->bits),
+					   newnode);
+	} else {
+		node = kzalloc(sizeof(*node), GFP_KERNEL);
+		if (unlikely(!node)) {
+			kfree(newnode);
+			return -ENOMEM;
+		}
+		INIT_LIST_HEAD(&node->peer_list);
+		copy_and_assign_cidr(node, newnode->bits, cidr, bits);
+
+		rcu_assign_pointer(CHOOSE_NODE(node, down->bits), down);
+		rcu_assign_pointer(CHOOSE_NODE(node, newnode->bits), newnode);
+		if (!parent)
+			rcu_assign_pointer(*trie, node);
+		else
+			rcu_assign_pointer(CHOOSE_NODE(parent, node->bits),
+					   node);
+	}
+	return 0;
+}
+
+void wg_allowedips_init(struct allowedips *table)
+{
+	table->root4 = table->root6 = NULL;
+	table->seq = 1;
+}
+
+void wg_allowedips_free(struct allowedips *table, struct mutex *lock)
+{
+	struct allowedips_node __rcu *old4 = table->root4, *old6 = table->root6;
+
+	++table->seq;
+	RCU_INIT_POINTER(table->root4, NULL);
+	RCU_INIT_POINTER(table->root6, NULL);
+	if (rcu_access_pointer(old4)) {
+		struct allowedips_node *node = rcu_dereference_protected(old4,
+							lockdep_is_held(lock));
+
+		root_remove_peer_lists(node);
+		call_rcu(&node->rcu, root_free_rcu);
+	}
+	if (rcu_access_pointer(old6)) {
+		struct allowedips_node *node = rcu_dereference_protected(old6,
+							lockdep_is_held(lock));
+
+		root_remove_peer_lists(node);
+		call_rcu(&node->rcu, root_free_rcu);
+	}
+}
+
+int wg_allowedips_insert_v4(struct allowedips *table, const struct in_addr *ip,
+			    u8 cidr, struct wg_peer *peer, struct mutex *lock)
+{
+	/* Aligned so it can be passed to fls */
+	u8 key[4] __aligned(__alignof(u32));
+
+	++table->seq;
+	swap_endian(key, (const u8 *)ip, 32);
+	return add(&table->root4, 32, key, cidr, peer, lock);
+}
+
+int wg_allowedips_insert_v6(struct allowedips *table, const struct in6_addr *ip,
+			    u8 cidr, struct wg_peer *peer, struct mutex *lock)
+{
+	/* Aligned so it can be passed to fls64 */
+	u8 key[16] __aligned(__alignof(u64));
+
+	++table->seq;
+	swap_endian(key, (const u8 *)ip, 128);
+	return add(&table->root6, 128, key, cidr, peer, lock);
+}
+
+void wg_allowedips_remove_by_peer(struct allowedips *table,
+				  struct wg_peer *peer, struct mutex *lock)
+{
+	++table->seq;
+	walk_remove_by_peer(&table->root4, peer, lock);
+	walk_remove_by_peer(&table->root6, peer, lock);
+}
+
+int wg_allowedips_read_node(struct allowedips_node *node, u8 ip[16], u8 *cidr)
+{
+	const unsigned int cidr_bytes = DIV_ROUND_UP(node->cidr, 8U);
+	swap_endian(ip, node->bits, node->bitlen);
+	memset(ip + cidr_bytes, 0, node->bitlen / 8U - cidr_bytes);
+	if (node->cidr)
+		ip[cidr_bytes - 1U] &= ~0U << (-node->cidr % 8U);
+
+	*cidr = node->cidr;
+	return node->bitlen == 32 ? AF_INET : AF_INET6;
+}
+
+/* Returns a strong reference to a peer */
+struct wg_peer *wg_allowedips_lookup_dst(struct allowedips *table,
+					 struct sk_buff *skb)
+{
+	if (skb->protocol == htons(ETH_P_IP))
+		return lookup(table->root4, 32, &ip_hdr(skb)->daddr);
+	else if (skb->protocol == htons(ETH_P_IPV6))
+		return lookup(table->root6, 128, &ipv6_hdr(skb)->daddr);
+	return NULL;
+}
+
+/* Returns a strong reference to a peer */
+struct wg_peer *wg_allowedips_lookup_src(struct allowedips *table,
+					 struct sk_buff *skb)
+{
+	if (skb->protocol == htons(ETH_P_IP))
+		return lookup(table->root4, 32, &ip_hdr(skb)->saddr);
+	else if (skb->protocol == htons(ETH_P_IPV6))
+		return lookup(table->root6, 128, &ipv6_hdr(skb)->saddr);
+	return NULL;
+}
+
+#include "selftest/allowedips.c"
diff --git a/drivers/net/wireguard/allowedips.h b/drivers/net/wireguard/allowedips.h
new file mode 100644
index 000000000000..e5c83cafcef4
--- /dev/null
+++ b/drivers/net/wireguard/allowedips.h
@@ -0,0 +1,59 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#ifndef _WG_ALLOWEDIPS_H
+#define _WG_ALLOWEDIPS_H
+
+#include <linux/mutex.h>
+#include <linux/ip.h>
+#include <linux/ipv6.h>
+
+struct wg_peer;
+
+struct allowedips_node {
+	struct wg_peer __rcu *peer;
+	struct allowedips_node __rcu *bit[2];
+	/* While it may seem scandalous that we waste space for v4,
+	 * we're alloc'ing to the nearest power of 2 anyway, so this
+	 * doesn't actually make a difference.
+	 */
+	u8 bits[16] __aligned(__alignof(u64));
+	u8 cidr, bit_at_a, bit_at_b, bitlen;
+
+	/* Keep rarely used list at bottom to be beyond cache line. */
+	union {
+		struct list_head peer_list;
+		struct rcu_head rcu;
+	};
+};
+
+struct allowedips {
+	struct allowedips_node __rcu *root4;
+	struct allowedips_node __rcu *root6;
+	u64 seq;
+};
+
+void wg_allowedips_init(struct allowedips *table);
+void wg_allowedips_free(struct allowedips *table, struct mutex *mutex);
+int wg_allowedips_insert_v4(struct allowedips *table, const struct in_addr *ip,
+			    u8 cidr, struct wg_peer *peer, struct mutex *lock);
+int wg_allowedips_insert_v6(struct allowedips *table, const struct in6_addr *ip,
+			    u8 cidr, struct wg_peer *peer, struct mutex *lock);
+void wg_allowedips_remove_by_peer(struct allowedips *table,
+				  struct wg_peer *peer, struct mutex *lock);
+/* The ip input pointer should be __aligned(__alignof(u64))) */
+int wg_allowedips_read_node(struct allowedips_node *node, u8 ip[16], u8 *cidr);
+
+/* These return a strong reference to a peer: */
+struct wg_peer *wg_allowedips_lookup_dst(struct allowedips *table,
+					 struct sk_buff *skb);
+struct wg_peer *wg_allowedips_lookup_src(struct allowedips *table,
+					 struct sk_buff *skb);
+
+#ifdef DEBUG
+bool wg_allowedips_selftest(void);
+#endif
+
+#endif /* _WG_ALLOWEDIPS_H */
diff --git a/drivers/net/wireguard/cookie.c b/drivers/net/wireguard/cookie.c
new file mode 100644
index 000000000000..4956f0499c19
--- /dev/null
+++ b/drivers/net/wireguard/cookie.c
@@ -0,0 +1,236 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#include "cookie.h"
+#include "peer.h"
+#include "device.h"
+#include "messages.h"
+#include "ratelimiter.h"
+#include "timers.h"
+
+#include <crypto/blake2s.h>
+#include <crypto/chacha20poly1305.h>
+
+#include <net/ipv6.h>
+#include <crypto/algapi.h>
+
+void wg_cookie_checker_init(struct cookie_checker *checker,
+			    struct wg_device *wg)
+{
+	init_rwsem(&checker->secret_lock);
+	checker->secret_birthdate = ktime_get_coarse_boottime_ns();
+	get_random_bytes(checker->secret, NOISE_HASH_LEN);
+	checker->device = wg;
+}
+
+enum { COOKIE_KEY_LABEL_LEN = 8 };
+static const u8 mac1_key_label[COOKIE_KEY_LABEL_LEN] = "mac1----";
+static const u8 cookie_key_label[COOKIE_KEY_LABEL_LEN] = "cookie--";
+
+static void precompute_key(u8 key[NOISE_SYMMETRIC_KEY_LEN],
+			   const u8 pubkey[NOISE_PUBLIC_KEY_LEN],
+			   const u8 label[COOKIE_KEY_LABEL_LEN])
+{
+	struct blake2s_state blake;
+
+	blake2s_init(&blake, NOISE_SYMMETRIC_KEY_LEN);
+	blake2s_update(&blake, label, COOKIE_KEY_LABEL_LEN);
+	blake2s_update(&blake, pubkey, NOISE_PUBLIC_KEY_LEN);
+	blake2s_final(&blake, key);
+}
+
+/* Must hold peer->handshake.static_identity->lock */
+void wg_cookie_checker_precompute_device_keys(struct cookie_checker *checker)
+{
+	if (likely(checker->device->static_identity.has_identity)) {
+		precompute_key(checker->cookie_encryption_key,
+			       checker->device->static_identity.static_public,
+			       cookie_key_label);
+		precompute_key(checker->message_mac1_key,
+			       checker->device->static_identity.static_public,
+			       mac1_key_label);
+	} else {
+		memset(checker->cookie_encryption_key, 0,
+		       NOISE_SYMMETRIC_KEY_LEN);
+		memset(checker->message_mac1_key, 0, NOISE_SYMMETRIC_KEY_LEN);
+	}
+}
+
+void wg_cookie_checker_precompute_peer_keys(struct wg_peer *peer)
+{
+	precompute_key(peer->latest_cookie.cookie_decryption_key,
+		       peer->handshake.remote_static, cookie_key_label);
+	precompute_key(peer->latest_cookie.message_mac1_key,
+		       peer->handshake.remote_static, mac1_key_label);
+}
+
+void wg_cookie_init(struct cookie *cookie)
+{
+	memset(cookie, 0, sizeof(*cookie));
+	init_rwsem(&cookie->lock);
+}
+
+static void compute_mac1(u8 mac1[COOKIE_LEN], const void *message, size_t len,
+			 const u8 key[NOISE_SYMMETRIC_KEY_LEN])
+{
+	len = len - sizeof(struct message_macs) +
+	      offsetof(struct message_macs, mac1);
+	blake2s(mac1, message, key, COOKIE_LEN, len, NOISE_SYMMETRIC_KEY_LEN);
+}
+
+static void compute_mac2(u8 mac2[COOKIE_LEN], const void *message, size_t len,
+			 const u8 cookie[COOKIE_LEN])
+{
+	len = len - sizeof(struct message_macs) +
+	      offsetof(struct message_macs, mac2);
+	blake2s(mac2, message, cookie, COOKIE_LEN, len, COOKIE_LEN);
+}
+
+static void make_cookie(u8 cookie[COOKIE_LEN], struct sk_buff *skb,
+			struct cookie_checker *checker)
+{
+	struct blake2s_state state;
+
+	if (wg_birthdate_has_expired(checker->secret_birthdate,
+				     COOKIE_SECRET_MAX_AGE)) {
+		down_write(&checker->secret_lock);
+		checker->secret_birthdate = ktime_get_coarse_boottime_ns();
+		get_random_bytes(checker->secret, NOISE_HASH_LEN);
+		up_write(&checker->secret_lock);
+	}
+
+	down_read(&checker->secret_lock);
+
+	blake2s_init_key(&state, COOKIE_LEN, checker->secret, NOISE_HASH_LEN);
+	if (skb->protocol == htons(ETH_P_IP))
+		blake2s_update(&state, (u8 *)&ip_hdr(skb)->saddr,
+			       sizeof(struct in_addr));
+	else if (skb->protocol == htons(ETH_P_IPV6))
+		blake2s_update(&state, (u8 *)&ipv6_hdr(skb)->saddr,
+			       sizeof(struct in6_addr));
+	blake2s_update(&state, (u8 *)&udp_hdr(skb)->source, sizeof(__be16));
+	blake2s_final(&state, cookie);
+
+	up_read(&checker->secret_lock);
+}
+
+enum cookie_mac_state wg_cookie_validate_packet(struct cookie_checker *checker,
+						struct sk_buff *skb,
+						bool check_cookie)
+{
+	struct message_macs *macs = (struct message_macs *)
+		(skb->data + skb->len - sizeof(*macs));
+	enum cookie_mac_state ret;
+	u8 computed_mac[COOKIE_LEN];
+	u8 cookie[COOKIE_LEN];
+
+	ret = INVALID_MAC;
+	compute_mac1(computed_mac, skb->data, skb->len,
+		     checker->message_mac1_key);
+	if (crypto_memneq(computed_mac, macs->mac1, COOKIE_LEN))
+		goto out;
+
+	ret = VALID_MAC_BUT_NO_COOKIE;
+
+	if (!check_cookie)
+		goto out;
+
+	make_cookie(cookie, skb, checker);
+
+	compute_mac2(computed_mac, skb->data, skb->len, cookie);
+	if (crypto_memneq(computed_mac, macs->mac2, COOKIE_LEN))
+		goto out;
+
+	ret = VALID_MAC_WITH_COOKIE_BUT_RATELIMITED;
+	if (!wg_ratelimiter_allow(skb, dev_net(checker->device->dev)))
+		goto out;
+
+	ret = VALID_MAC_WITH_COOKIE;
+
+out:
+	return ret;
+}
+
+void wg_cookie_add_mac_to_packet(void *message, size_t len,
+				 struct wg_peer *peer)
+{
+	struct message_macs *macs = (struct message_macs *)
+		((u8 *)message + len - sizeof(*macs));
+
+	down_write(&peer->latest_cookie.lock);
+	compute_mac1(macs->mac1, message, len,
+		     peer->latest_cookie.message_mac1_key);
+	memcpy(peer->latest_cookie.last_mac1_sent, macs->mac1, COOKIE_LEN);
+	peer->latest_cookie.have_sent_mac1 = true;
+	up_write(&peer->latest_cookie.lock);
+
+	down_read(&peer->latest_cookie.lock);
+	if (peer->latest_cookie.is_valid &&
+	    !wg_birthdate_has_expired(peer->latest_cookie.birthdate,
+				COOKIE_SECRET_MAX_AGE - COOKIE_SECRET_LATENCY))
+		compute_mac2(macs->mac2, message, len,
+			     peer->latest_cookie.cookie);
+	else
+		memset(macs->mac2, 0, COOKIE_LEN);
+	up_read(&peer->latest_cookie.lock);
+}
+
+void wg_cookie_message_create(struct message_handshake_cookie *dst,
+			      struct sk_buff *skb, __le32 index,
+			      struct cookie_checker *checker)
+{
+	struct message_macs *macs = (struct message_macs *)
+		((u8 *)skb->data + skb->len - sizeof(*macs));
+	u8 cookie[COOKIE_LEN];
+
+	dst->header.type = cpu_to_le32(MESSAGE_HANDSHAKE_COOKIE);
+	dst->receiver_index = index;
+	get_random_bytes_wait(dst->nonce, COOKIE_NONCE_LEN);
+
+	make_cookie(cookie, skb, checker);
+	xchacha20poly1305_encrypt(dst->encrypted_cookie, cookie, COOKIE_LEN,
+				  macs->mac1, COOKIE_LEN, dst->nonce,
+				  checker->cookie_encryption_key);
+}
+
+void wg_cookie_message_consume(struct message_handshake_cookie *src,
+			       struct wg_device *wg)
+{
+	struct wg_peer *peer = NULL;
+	u8 cookie[COOKIE_LEN];
+	bool ret;
+
+	if (unlikely(!wg_index_hashtable_lookup(wg->index_hashtable,
+						INDEX_HASHTABLE_HANDSHAKE |
+						INDEX_HASHTABLE_KEYPAIR,
+						src->receiver_index, &peer)))
+		return;
+
+	down_read(&peer->latest_cookie.lock);
+	if (unlikely(!peer->latest_cookie.have_sent_mac1)) {
+		up_read(&peer->latest_cookie.lock);
+		goto out;
+	}
+	ret = xchacha20poly1305_decrypt(
+		cookie, src->encrypted_cookie, sizeof(src->encrypted_cookie),
+		peer->latest_cookie.last_mac1_sent, COOKIE_LEN, src->nonce,
+		peer->latest_cookie.cookie_decryption_key);
+	up_read(&peer->latest_cookie.lock);
+
+	if (ret) {
+		down_write(&peer->latest_cookie.lock);
+		memcpy(peer->latest_cookie.cookie, cookie, COOKIE_LEN);
+		peer->latest_cookie.birthdate = ktime_get_coarse_boottime_ns();
+		peer->latest_cookie.is_valid = true;
+		peer->latest_cookie.have_sent_mac1 = false;
+		up_write(&peer->latest_cookie.lock);
+	} else {
+		net_dbg_ratelimited("%s: Could not decrypt invalid cookie response\n",
+				    wg->dev->name);
+	}
+
+out:
+	wg_peer_put(peer);
+}
diff --git a/drivers/net/wireguard/cookie.h b/drivers/net/wireguard/cookie.h
new file mode 100644
index 000000000000..c4bd61ca03f2
--- /dev/null
+++ b/drivers/net/wireguard/cookie.h
@@ -0,0 +1,59 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#ifndef _WG_COOKIE_H
+#define _WG_COOKIE_H
+
+#include "messages.h"
+#include <linux/rwsem.h>
+
+struct wg_peer;
+
+struct cookie_checker {
+	u8 secret[NOISE_HASH_LEN];
+	u8 cookie_encryption_key[NOISE_SYMMETRIC_KEY_LEN];
+	u8 message_mac1_key[NOISE_SYMMETRIC_KEY_LEN];
+	u64 secret_birthdate;
+	struct rw_semaphore secret_lock;
+	struct wg_device *device;
+};
+
+struct cookie {
+	u64 birthdate;
+	bool is_valid;
+	u8 cookie[COOKIE_LEN];
+	bool have_sent_mac1;
+	u8 last_mac1_sent[COOKIE_LEN];
+	u8 cookie_decryption_key[NOISE_SYMMETRIC_KEY_LEN];
+	u8 message_mac1_key[NOISE_SYMMETRIC_KEY_LEN];
+	struct rw_semaphore lock;
+};
+
+enum cookie_mac_state {
+	INVALID_MAC,
+	VALID_MAC_BUT_NO_COOKIE,
+	VALID_MAC_WITH_COOKIE_BUT_RATELIMITED,
+	VALID_MAC_WITH_COOKIE
+};
+
+void wg_cookie_checker_init(struct cookie_checker *checker,
+			    struct wg_device *wg);
+void wg_cookie_checker_precompute_device_keys(struct cookie_checker *checker);
+void wg_cookie_checker_precompute_peer_keys(struct wg_peer *peer);
+void wg_cookie_init(struct cookie *cookie);
+
+enum cookie_mac_state wg_cookie_validate_packet(struct cookie_checker *checker,
+						struct sk_buff *skb,
+						bool check_cookie);
+void wg_cookie_add_mac_to_packet(void *message, size_t len,
+				 struct wg_peer *peer);
+
+void wg_cookie_message_create(struct message_handshake_cookie *src,
+			      struct sk_buff *skb, __le32 index,
+			      struct cookie_checker *checker);
+void wg_cookie_message_consume(struct message_handshake_cookie *src,
+			       struct wg_device *wg);
+
+#endif /* _WG_COOKIE_H */
diff --git a/drivers/net/wireguard/device.c b/drivers/net/wireguard/device.c
new file mode 100644
index 000000000000..16b19824b9ad
--- /dev/null
+++ b/drivers/net/wireguard/device.c
@@ -0,0 +1,458 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#include "queueing.h"
+#include "socket.h"
+#include "timers.h"
+#include "device.h"
+#include "ratelimiter.h"
+#include "peer.h"
+#include "messages.h"
+
+#include <linux/module.h>
+#include <linux/rtnetlink.h>
+#include <linux/inet.h>
+#include <linux/netdevice.h>
+#include <linux/inetdevice.h>
+#include <linux/if_arp.h>
+#include <linux/icmp.h>
+#include <linux/suspend.h>
+#include <net/icmp.h>
+#include <net/rtnetlink.h>
+#include <net/ip_tunnels.h>
+#include <net/addrconf.h>
+
+static LIST_HEAD(device_list);
+
+static int wg_open(struct net_device *dev)
+{
+	struct in_device *dev_v4 = __in_dev_get_rtnl(dev);
+	struct inet6_dev *dev_v6 = __in6_dev_get(dev);
+	struct wg_device *wg = netdev_priv(dev);
+	struct wg_peer *peer;
+	int ret;
+
+	if (dev_v4) {
+		/* At some point we might put this check near the ip_rt_send_
+		 * redirect call of ip_forward in net/ipv4/ip_forward.c, similar
+		 * to the current secpath check.
+		 */
+		IN_DEV_CONF_SET(dev_v4, SEND_REDIRECTS, false);
+		IPV4_DEVCONF_ALL(dev_net(dev), SEND_REDIRECTS) = false;
+	}
+	if (dev_v6)
+		dev_v6->cnf.addr_gen_mode = IN6_ADDR_GEN_MODE_NONE;
+
+	ret = wg_socket_init(wg, wg->incoming_port);
+	if (ret < 0)
+		return ret;
+	mutex_lock(&wg->device_update_lock);
+	list_for_each_entry(peer, &wg->peer_list, peer_list) {
+		wg_packet_send_staged_packets(peer);
+		if (peer->persistent_keepalive_interval)
+			wg_packet_send_keepalive(peer);
+	}
+	mutex_unlock(&wg->device_update_lock);
+	return 0;
+}
+
+#ifdef CONFIG_PM_SLEEP
+static int wg_pm_notification(struct notifier_block *nb, unsigned long action,
+			      void *data)
+{
+	struct wg_device *wg;
+	struct wg_peer *peer;
+
+	/* If the machine is constantly suspending and resuming, as part of
+	 * its normal operation rather than as a somewhat rare event, then we
+	 * don't actually want to clear keys.
+	 */
+	if (IS_ENABLED(CONFIG_PM_AUTOSLEEP) || IS_ENABLED(CONFIG_ANDROID))
+		return 0;
+
+	if (action != PM_HIBERNATION_PREPARE && action != PM_SUSPEND_PREPARE)
+		return 0;
+
+	rtnl_lock();
+	list_for_each_entry(wg, &device_list, device_list) {
+		mutex_lock(&wg->device_update_lock);
+		list_for_each_entry(peer, &wg->peer_list, peer_list) {
+			del_timer(&peer->timer_zero_key_material);
+			wg_noise_handshake_clear(&peer->handshake);
+			wg_noise_keypairs_clear(&peer->keypairs);
+		}
+		mutex_unlock(&wg->device_update_lock);
+	}
+	rtnl_unlock();
+	rcu_barrier();
+	return 0;
+}
+
+static struct notifier_block pm_notifier = { .notifier_call = wg_pm_notification };
+#endif
+
+static int wg_stop(struct net_device *dev)
+{
+	struct wg_device *wg = netdev_priv(dev);
+	struct wg_peer *peer;
+
+	mutex_lock(&wg->device_update_lock);
+	list_for_each_entry(peer, &wg->peer_list, peer_list) {
+		wg_packet_purge_staged_packets(peer);
+		wg_timers_stop(peer);
+		wg_noise_handshake_clear(&peer->handshake);
+		wg_noise_keypairs_clear(&peer->keypairs);
+		wg_noise_reset_last_sent_handshake(&peer->last_sent_handshake);
+	}
+	mutex_unlock(&wg->device_update_lock);
+	skb_queue_purge(&wg->incoming_handshakes);
+	wg_socket_reinit(wg, NULL, NULL);
+	return 0;
+}
+
+static netdev_tx_t wg_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+	struct wg_device *wg = netdev_priv(dev);
+	struct sk_buff_head packets;
+	struct wg_peer *peer;
+	struct sk_buff *next;
+	sa_family_t family;
+	u32 mtu;
+	int ret;
+
+	if (unlikely(wg_skb_examine_untrusted_ip_hdr(skb) != skb->protocol)) {
+		ret = -EPROTONOSUPPORT;
+		net_dbg_ratelimited("%s: Invalid IP packet\n", dev->name);
+		goto err;
+	}
+
+	peer = wg_allowedips_lookup_dst(&wg->peer_allowedips, skb);
+	if (unlikely(!peer)) {
+		ret = -ENOKEY;
+		if (skb->protocol == htons(ETH_P_IP))
+			net_dbg_ratelimited("%s: No peer has allowed IPs matching %pI4\n",
+					    dev->name, &ip_hdr(skb)->daddr);
+		else if (skb->protocol == htons(ETH_P_IPV6))
+			net_dbg_ratelimited("%s: No peer has allowed IPs matching %pI6\n",
+					    dev->name, &ipv6_hdr(skb)->daddr);
+		goto err;
+	}
+
+	family = READ_ONCE(peer->endpoint.addr.sa_family);
+	if (unlikely(family != AF_INET && family != AF_INET6)) {
+		ret = -EDESTADDRREQ;
+		net_dbg_ratelimited("%s: No valid endpoint has been configured or discovered for peer %llu\n",
+				    dev->name, peer->internal_id);
+		goto err_peer;
+	}
+
+	mtu = skb_dst(skb) ? dst_mtu(skb_dst(skb)) : dev->mtu;
+
+	__skb_queue_head_init(&packets);
+	if (!skb_is_gso(skb)) {
+		skb_mark_not_on_list(skb);
+	} else {
+		struct sk_buff *segs = skb_gso_segment(skb, 0);
+
+		if (unlikely(IS_ERR(segs))) {
+			ret = PTR_ERR(segs);
+			goto err_peer;
+		}
+		dev_kfree_skb(skb);
+		skb = segs;
+	}
+
+	skb_list_walk_safe(skb, skb, next) {
+		skb_mark_not_on_list(skb);
+
+		skb = skb_share_check(skb, GFP_ATOMIC);
+		if (unlikely(!skb))
+			continue;
+
+		/* We only need to keep the original dst around for icmp,
+		 * so at this point we're in a position to drop it.
+		 */
+		skb_dst_drop(skb);
+
+		PACKET_CB(skb)->mtu = mtu;
+
+		__skb_queue_tail(&packets, skb);
+	}
+
+	spin_lock_bh(&peer->staged_packet_queue.lock);
+	/* If the queue is getting too big, we start removing the oldest packets
+	 * until it's small again. We do this before adding the new packet, so
+	 * we don't remove GSO segments that are in excess.
+	 */
+	while (skb_queue_len(&peer->staged_packet_queue) > MAX_STAGED_PACKETS) {
+		dev_kfree_skb(__skb_dequeue(&peer->staged_packet_queue));
+		++dev->stats.tx_dropped;
+	}
+	skb_queue_splice_tail(&packets, &peer->staged_packet_queue);
+	spin_unlock_bh(&peer->staged_packet_queue.lock);
+
+	wg_packet_send_staged_packets(peer);
+
+	wg_peer_put(peer);
+	return NETDEV_TX_OK;
+
+err_peer:
+	wg_peer_put(peer);
+err:
+	++dev->stats.tx_errors;
+	if (skb->protocol == htons(ETH_P_IP))
+		icmp_send(skb, ICMP_DEST_UNREACH, ICMP_HOST_UNREACH, 0);
+	else if (skb->protocol == htons(ETH_P_IPV6))
+		icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_ADDR_UNREACH, 0);
+	kfree_skb(skb);
+	return ret;
+}
+
+static const struct net_device_ops netdev_ops = {
+	.ndo_open		= wg_open,
+	.ndo_stop		= wg_stop,
+	.ndo_start_xmit		= wg_xmit,
+	.ndo_get_stats64	= ip_tunnel_get_stats64
+};
+
+static void wg_destruct(struct net_device *dev)
+{
+	struct wg_device *wg = netdev_priv(dev);
+
+	rtnl_lock();
+	list_del(&wg->device_list);
+	rtnl_unlock();
+	mutex_lock(&wg->device_update_lock);
+	wg->incoming_port = 0;
+	wg_socket_reinit(wg, NULL, NULL);
+	/* The final references are cleared in the below calls to destroy_workqueue. */
+	wg_peer_remove_all(wg);
+	destroy_workqueue(wg->handshake_receive_wq);
+	destroy_workqueue(wg->handshake_send_wq);
+	destroy_workqueue(wg->packet_crypt_wq);
+	wg_packet_queue_free(&wg->decrypt_queue, true);
+	wg_packet_queue_free(&wg->encrypt_queue, true);
+	rcu_barrier(); /* Wait for all the peers to be actually freed. */
+	wg_ratelimiter_uninit();
+	memzero_explicit(&wg->static_identity, sizeof(wg->static_identity));
+	skb_queue_purge(&wg->incoming_handshakes);
+	free_percpu(dev->tstats);
+	free_percpu(wg->incoming_handshakes_worker);
+	if (wg->have_creating_net_ref)
+		put_net(wg->creating_net);
+	kvfree(wg->index_hashtable);
+	kvfree(wg->peer_hashtable);
+	mutex_unlock(&wg->device_update_lock);
+
+	pr_debug("%s: Interface deleted\n", dev->name);
+	free_netdev(dev);
+}
+
+static const struct device_type device_type = { .name = KBUILD_MODNAME };
+
+static void wg_setup(struct net_device *dev)
+{
+	struct wg_device *wg = netdev_priv(dev);
+	enum { WG_NETDEV_FEATURES = NETIF_F_HW_CSUM | NETIF_F_RXCSUM |
+				    NETIF_F_SG | NETIF_F_GSO |
+				    NETIF_F_GSO_SOFTWARE | NETIF_F_HIGHDMA };
+
+	dev->netdev_ops = &netdev_ops;
+	dev->hard_header_len = 0;
+	dev->addr_len = 0;
+	dev->needed_headroom = DATA_PACKET_HEAD_ROOM;
+	dev->needed_tailroom = noise_encrypted_len(MESSAGE_PADDING_MULTIPLE);
+	dev->type = ARPHRD_NONE;
+	dev->flags = IFF_POINTOPOINT | IFF_NOARP;
+	dev->priv_flags |= IFF_NO_QUEUE;
+	dev->features |= NETIF_F_LLTX;
+	dev->features |= WG_NETDEV_FEATURES;
+	dev->hw_features |= WG_NETDEV_FEATURES;
+	dev->hw_enc_features |= WG_NETDEV_FEATURES;
+	dev->mtu = ETH_DATA_LEN - MESSAGE_MINIMUM_LENGTH -
+		   sizeof(struct udphdr) -
+		   max(sizeof(struct ipv6hdr), sizeof(struct iphdr));
+
+	SET_NETDEV_DEVTYPE(dev, &device_type);
+
+	/* We need to keep the dst around in case of icmp replies. */
+	netif_keep_dst(dev);
+
+	memset(wg, 0, sizeof(*wg));
+	wg->dev = dev;
+}
+
+static int wg_newlink(struct net *src_net, struct net_device *dev,
+		      struct nlattr *tb[], struct nlattr *data[],
+		      struct netlink_ext_ack *extack)
+{
+	struct wg_device *wg = netdev_priv(dev);
+	int ret = -ENOMEM;
+
+	wg->creating_net = src_net;
+	init_rwsem(&wg->static_identity.lock);
+	mutex_init(&wg->socket_update_lock);
+	mutex_init(&wg->device_update_lock);
+	skb_queue_head_init(&wg->incoming_handshakes);
+	wg_allowedips_init(&wg->peer_allowedips);
+	wg_cookie_checker_init(&wg->cookie_checker, wg);
+	INIT_LIST_HEAD(&wg->peer_list);
+	wg->device_update_gen = 1;
+
+	wg->peer_hashtable = wg_pubkey_hashtable_alloc();
+	if (!wg->peer_hashtable)
+		return ret;
+
+	wg->index_hashtable = wg_index_hashtable_alloc();
+	if (!wg->index_hashtable)
+		goto err_free_peer_hashtable;
+
+	dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats);
+	if (!dev->tstats)
+		goto err_free_index_hashtable;
+
+	wg->incoming_handshakes_worker =
+		wg_packet_percpu_multicore_worker_alloc(
+				wg_packet_handshake_receive_worker, wg);
+	if (!wg->incoming_handshakes_worker)
+		goto err_free_tstats;
+
+	wg->handshake_receive_wq = alloc_workqueue("wg-kex-%s",
+			WQ_CPU_INTENSIVE | WQ_FREEZABLE, 0, dev->name);
+	if (!wg->handshake_receive_wq)
+		goto err_free_incoming_handshakes;
+
+	wg->handshake_send_wq = alloc_workqueue("wg-kex-%s",
+			WQ_UNBOUND | WQ_FREEZABLE, 0, dev->name);
+	if (!wg->handshake_send_wq)
+		goto err_destroy_handshake_receive;
+
+	wg->packet_crypt_wq = alloc_workqueue("wg-crypt-%s",
+			WQ_CPU_INTENSIVE | WQ_MEM_RECLAIM, 0, dev->name);
+	if (!wg->packet_crypt_wq)
+		goto err_destroy_handshake_send;
+
+	ret = wg_packet_queue_init(&wg->encrypt_queue, wg_packet_encrypt_worker,
+				   true, MAX_QUEUED_PACKETS);
+	if (ret < 0)
+		goto err_destroy_packet_crypt;
+
+	ret = wg_packet_queue_init(&wg->decrypt_queue, wg_packet_decrypt_worker,
+				   true, MAX_QUEUED_PACKETS);
+	if (ret < 0)
+		goto err_free_encrypt_queue;
+
+	ret = wg_ratelimiter_init();
+	if (ret < 0)
+		goto err_free_decrypt_queue;
+
+	ret = register_netdevice(dev);
+	if (ret < 0)
+		goto err_uninit_ratelimiter;
+
+	list_add(&wg->device_list, &device_list);
+
+	/* We wait until the end to assign priv_destructor, so that
+	 * register_netdevice doesn't call it for us if it fails.
+	 */
+	dev->priv_destructor = wg_destruct;
+
+	pr_debug("%s: Interface created\n", dev->name);
+	return ret;
+
+err_uninit_ratelimiter:
+	wg_ratelimiter_uninit();
+err_free_decrypt_queue:
+	wg_packet_queue_free(&wg->decrypt_queue, true);
+err_free_encrypt_queue:
+	wg_packet_queue_free(&wg->encrypt_queue, true);
+err_destroy_packet_crypt:
+	destroy_workqueue(wg->packet_crypt_wq);
+err_destroy_handshake_send:
+	destroy_workqueue(wg->handshake_send_wq);
+err_destroy_handshake_receive:
+	destroy_workqueue(wg->handshake_receive_wq);
+err_free_incoming_handshakes:
+	free_percpu(wg->incoming_handshakes_worker);
+err_free_tstats:
+	free_percpu(dev->tstats);
+err_free_index_hashtable:
+	kvfree(wg->index_hashtable);
+err_free_peer_hashtable:
+	kvfree(wg->peer_hashtable);
+	return ret;
+}
+
+static struct rtnl_link_ops link_ops __read_mostly = {
+	.kind			= KBUILD_MODNAME,
+	.priv_size		= sizeof(struct wg_device),
+	.setup			= wg_setup,
+	.newlink		= wg_newlink,
+};
+
+static int wg_netdevice_notification(struct notifier_block *nb,
+				     unsigned long action, void *data)
+{
+	struct net_device *dev = ((struct netdev_notifier_info *)data)->dev;
+	struct wg_device *wg = netdev_priv(dev);
+
+	ASSERT_RTNL();
+
+	if (action != NETDEV_REGISTER || dev->netdev_ops != &netdev_ops)
+		return 0;
+
+	if (dev_net(dev) == wg->creating_net && wg->have_creating_net_ref) {
+		put_net(wg->creating_net);
+		wg->have_creating_net_ref = false;
+	} else if (dev_net(dev) != wg->creating_net &&
+		   !wg->have_creating_net_ref) {
+		wg->have_creating_net_ref = true;
+		get_net(wg->creating_net);
+	}
+	return 0;
+}
+
+static struct notifier_block netdevice_notifier = {
+	.notifier_call = wg_netdevice_notification
+};
+
+int __init wg_device_init(void)
+{
+	int ret;
+
+#ifdef CONFIG_PM_SLEEP
+	ret = register_pm_notifier(&pm_notifier);
+	if (ret)
+		return ret;
+#endif
+
+	ret = register_netdevice_notifier(&netdevice_notifier);
+	if (ret)
+		goto error_pm;
+
+	ret = rtnl_link_register(&link_ops);
+	if (ret)
+		goto error_netdevice;
+
+	return 0;
+
+error_netdevice:
+	unregister_netdevice_notifier(&netdevice_notifier);
+error_pm:
+#ifdef CONFIG_PM_SLEEP
+	unregister_pm_notifier(&pm_notifier);
+#endif
+	return ret;
+}
+
+void wg_device_uninit(void)
+{
+	rtnl_link_unregister(&link_ops);
+	unregister_netdevice_notifier(&netdevice_notifier);
+#ifdef CONFIG_PM_SLEEP
+	unregister_pm_notifier(&pm_notifier);
+#endif
+	rcu_barrier();
+}
diff --git a/drivers/net/wireguard/device.h b/drivers/net/wireguard/device.h
new file mode 100644
index 000000000000..c91f3051c5c7
--- /dev/null
+++ b/drivers/net/wireguard/device.h
@@ -0,0 +1,73 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#ifndef _WG_DEVICE_H
+#define _WG_DEVICE_H
+
+#include "noise.h"
+#include "allowedips.h"
+#include "peerlookup.h"
+#include "cookie.h"
+
+#include <linux/types.h>
+#include <linux/netdevice.h>
+#include <linux/workqueue.h>
+#include <linux/mutex.h>
+#include <linux/net.h>
+#include <linux/ptr_ring.h>
+
+struct wg_device;
+
+struct multicore_worker {
+	void *ptr;
+	struct work_struct work;
+};
+
+struct crypt_queue {
+	struct ptr_ring ring;
+	union {
+		struct {
+			struct multicore_worker __percpu *worker;
+			int last_cpu;
+		};
+		struct work_struct work;
+	};
+};
+
+struct wg_device {
+	struct net_device *dev;
+	struct crypt_queue encrypt_queue, decrypt_queue;
+	struct sock __rcu *sock4, *sock6;
+	struct net *creating_net;
+	struct noise_static_identity static_identity;
+	struct workqueue_struct *handshake_receive_wq, *handshake_send_wq;
+	struct workqueue_struct *packet_crypt_wq;
+	struct sk_buff_head incoming_handshakes;
+	int incoming_handshake_cpu;
+	struct multicore_worker __percpu *incoming_handshakes_worker;
+	struct cookie_checker cookie_checker;
+	struct pubkey_hashtable *peer_hashtable;
+	struct index_hashtable *index_hashtable;
+	struct allowedips peer_allowedips;
+	struct mutex device_update_lock, socket_update_lock;
+	struct list_head device_list, peer_list;
+	unsigned int num_peers, device_update_gen;
+	u32 fwmark;
+	u16 incoming_port;
+	bool have_creating_net_ref;
+};
+
+int wg_device_init(void);
+void wg_device_uninit(void);
+
+/* Later after the dust settles, this can be moved into include/linux/skbuff.h,
+ * where virtually all code that deals with GSO segs can benefit, around ~30
+ * drivers as of writing.
+ */
+#define skb_list_walk_safe(first, skb, next)                                   \
+	for (skb = first, next = skb->next; skb;                               \
+	     skb = next, next = skb ? skb->next : NULL)
+
+#endif /* _WG_DEVICE_H */
diff --git a/drivers/net/wireguard/main.c b/drivers/net/wireguard/main.c
new file mode 100644
index 000000000000..10c0a40f6a9e
--- /dev/null
+++ b/drivers/net/wireguard/main.c
@@ -0,0 +1,64 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#include "version.h"
+#include "device.h"
+#include "noise.h"
+#include "queueing.h"
+#include "ratelimiter.h"
+#include "netlink.h"
+
+#include <uapi/linux/wireguard.h>
+
+#include <linux/version.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/genetlink.h>
+#include <net/rtnetlink.h>
+
+static int __init mod_init(void)
+{
+	int ret;
+
+#ifdef DEBUG
+	if (!wg_allowedips_selftest() || !wg_packet_counter_selftest() ||
+	    !wg_ratelimiter_selftest())
+		return -ENOTRECOVERABLE;
+#endif
+	wg_noise_init();
+
+	ret = wg_device_init();
+	if (ret < 0)
+		goto err_device;
+
+	ret = wg_genetlink_init();
+	if (ret < 0)
+		goto err_netlink;
+
+	pr_info("WireGuard " WIREGUARD_VERSION " loaded. See www.wireguard.com for information.\n");
+	pr_info("Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.\n");
+
+	return 0;
+
+err_netlink:
+	wg_device_uninit();
+err_device:
+	return ret;
+}
+
+static void __exit mod_exit(void)
+{
+	wg_genetlink_uninit();
+	wg_device_uninit();
+}
+
+module_init(mod_init);
+module_exit(mod_exit);
+MODULE_LICENSE("GPL v2");
+MODULE_DESCRIPTION("WireGuard secure network tunnel");
+MODULE_AUTHOR("Jason A. Donenfeld <Jason@zx2c4.com>");
+MODULE_VERSION(WIREGUARD_VERSION);
+MODULE_ALIAS_RTNL_LINK(KBUILD_MODNAME);
+MODULE_ALIAS_GENL_FAMILY(WG_GENL_NAME);
diff --git a/drivers/net/wireguard/messages.h b/drivers/net/wireguard/messages.h
new file mode 100644
index 000000000000..b8a7b9ce32ba
--- /dev/null
+++ b/drivers/net/wireguard/messages.h
@@ -0,0 +1,128 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#ifndef _WG_MESSAGES_H
+#define _WG_MESSAGES_H
+
+#include <crypto/curve25519.h>
+#include <crypto/chacha20poly1305.h>
+#include <crypto/blake2s.h>
+
+#include <linux/kernel.h>
+#include <linux/param.h>
+#include <linux/skbuff.h>
+
+enum noise_lengths {
+	NOISE_PUBLIC_KEY_LEN = CURVE25519_KEY_SIZE,
+	NOISE_SYMMETRIC_KEY_LEN = CHACHA20POLY1305_KEY_SIZE,
+	NOISE_TIMESTAMP_LEN = sizeof(u64) + sizeof(u32),
+	NOISE_AUTHTAG_LEN = CHACHA20POLY1305_AUTHTAG_SIZE,
+	NOISE_HASH_LEN = BLAKE2S_HASH_SIZE
+};
+
+#define noise_encrypted_len(plain_len) ((plain_len) + NOISE_AUTHTAG_LEN)
+
+enum cookie_values {
+	COOKIE_SECRET_MAX_AGE = 2 * 60,
+	COOKIE_SECRET_LATENCY = 5,
+	COOKIE_NONCE_LEN = XCHACHA20POLY1305_NONCE_SIZE,
+	COOKIE_LEN = 16
+};
+
+enum counter_values {
+	COUNTER_BITS_TOTAL = 2048,
+	COUNTER_REDUNDANT_BITS = BITS_PER_LONG,
+	COUNTER_WINDOW_SIZE = COUNTER_BITS_TOTAL - COUNTER_REDUNDANT_BITS
+};
+
+enum limits {
+	REKEY_AFTER_MESSAGES = 1ULL << 60,
+	REJECT_AFTER_MESSAGES = U64_MAX - COUNTER_WINDOW_SIZE - 1,
+	REKEY_TIMEOUT = 5,
+	REKEY_TIMEOUT_JITTER_MAX_JIFFIES = HZ / 3,
+	REKEY_AFTER_TIME = 120,
+	REJECT_AFTER_TIME = 180,
+	INITIATIONS_PER_SECOND = 50,
+	MAX_PEERS_PER_DEVICE = 1U << 20,
+	KEEPALIVE_TIMEOUT = 10,
+	MAX_TIMER_HANDSHAKES = 90 / REKEY_TIMEOUT,
+	MAX_QUEUED_INCOMING_HANDSHAKES = 4096, /* TODO: replace this with DQL */
+	MAX_STAGED_PACKETS = 128,
+	MAX_QUEUED_PACKETS = 1024 /* TODO: replace this with DQL */
+};
+
+enum message_type {
+	MESSAGE_INVALID = 0,
+	MESSAGE_HANDSHAKE_INITIATION = 1,
+	MESSAGE_HANDSHAKE_RESPONSE = 2,
+	MESSAGE_HANDSHAKE_COOKIE = 3,
+	MESSAGE_DATA = 4
+};
+
+struct message_header {
+	/* The actual layout of this that we want is:
+	 * u8 type
+	 * u8 reserved_zero[3]
+	 *
+	 * But it turns out that by encoding this as little endian,
+	 * we achieve the same thing, and it makes checking faster.
+	 */
+	__le32 type;
+};
+
+struct message_macs {
+	u8 mac1[COOKIE_LEN];
+	u8 mac2[COOKIE_LEN];
+};
+
+struct message_handshake_initiation {
+	struct message_header header;
+	__le32 sender_index;
+	u8 unencrypted_ephemeral[NOISE_PUBLIC_KEY_LEN];
+	u8 encrypted_static[noise_encrypted_len(NOISE_PUBLIC_KEY_LEN)];
+	u8 encrypted_timestamp[noise_encrypted_len(NOISE_TIMESTAMP_LEN)];
+	struct message_macs macs;
+};
+
+struct message_handshake_response {
+	struct message_header header;
+	__le32 sender_index;
+	__le32 receiver_index;
+	u8 unencrypted_ephemeral[NOISE_PUBLIC_KEY_LEN];
+	u8 encrypted_nothing[noise_encrypted_len(0)];
+	struct message_macs macs;
+};
+
+struct message_handshake_cookie {
+	struct message_header header;
+	__le32 receiver_index;
+	u8 nonce[COOKIE_NONCE_LEN];
+	u8 encrypted_cookie[noise_encrypted_len(COOKIE_LEN)];
+};
+
+struct message_data {
+	struct message_header header;
+	__le32 key_idx;
+	__le64 counter;
+	u8 encrypted_data[];
+};
+
+#define message_data_len(plain_len) \
+	(noise_encrypted_len(plain_len) + sizeof(struct message_data))
+
+enum message_alignments {
+	MESSAGE_PADDING_MULTIPLE = 16,
+	MESSAGE_MINIMUM_LENGTH = message_data_len(0)
+};
+
+#define SKB_HEADER_LEN                                       \
+	(max(sizeof(struct iphdr), sizeof(struct ipv6hdr)) + \
+	 sizeof(struct udphdr) + NET_SKB_PAD)
+#define DATA_PACKET_HEAD_ROOM \
+	ALIGN(sizeof(struct message_data) + SKB_HEADER_LEN, 4)
+
+enum { HANDSHAKE_DSCP = 0x88 /* AF41, plus 00 ECN */ };
+
+#endif /* _WG_MESSAGES_H */
diff --git a/drivers/net/wireguard/netlink.c b/drivers/net/wireguard/netlink.c
new file mode 100644
index 000000000000..0fdbd1c45977
--- /dev/null
+++ b/drivers/net/wireguard/netlink.c
@@ -0,0 +1,642 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#include "netlink.h"
+#include "device.h"
+#include "peer.h"
+#include "socket.h"
+#include "queueing.h"
+#include "messages.h"
+
+#include <uapi/linux/wireguard.h>
+
+#include <linux/if.h>
+#include <net/genetlink.h>
+#include <net/sock.h>
+#include <crypto/algapi.h>
+
+static struct genl_family genl_family;
+
+static const struct nla_policy device_policy[WGDEVICE_A_MAX + 1] = {
+	[WGDEVICE_A_IFINDEX]		= { .type = NLA_U32 },
+	[WGDEVICE_A_IFNAME]		= { .type = NLA_NUL_STRING, .len = IFNAMSIZ - 1 },
+	[WGDEVICE_A_PRIVATE_KEY]	= { .type = NLA_EXACT_LEN, .len = NOISE_PUBLIC_KEY_LEN },
+	[WGDEVICE_A_PUBLIC_KEY]		= { .type = NLA_EXACT_LEN, .len = NOISE_PUBLIC_KEY_LEN },
+	[WGDEVICE_A_FLAGS]		= { .type = NLA_U32 },
+	[WGDEVICE_A_LISTEN_PORT]	= { .type = NLA_U16 },
+	[WGDEVICE_A_FWMARK]		= { .type = NLA_U32 },
+	[WGDEVICE_A_PEERS]		= { .type = NLA_NESTED }
+};
+
+static const struct nla_policy peer_policy[WGPEER_A_MAX + 1] = {
+	[WGPEER_A_PUBLIC_KEY]				= { .type = NLA_EXACT_LEN, .len = NOISE_PUBLIC_KEY_LEN },
+	[WGPEER_A_PRESHARED_KEY]			= { .type = NLA_EXACT_LEN, .len = NOISE_SYMMETRIC_KEY_LEN },
+	[WGPEER_A_FLAGS]				= { .type = NLA_U32 },
+	[WGPEER_A_ENDPOINT]				= { .type = NLA_MIN_LEN, .len = sizeof(struct sockaddr) },
+	[WGPEER_A_PERSISTENT_KEEPALIVE_INTERVAL]	= { .type = NLA_U16 },
+	[WGPEER_A_LAST_HANDSHAKE_TIME]			= { .type = NLA_EXACT_LEN, .len = sizeof(struct __kernel_timespec) },
+	[WGPEER_A_RX_BYTES]				= { .type = NLA_U64 },
+	[WGPEER_A_TX_BYTES]				= { .type = NLA_U64 },
+	[WGPEER_A_ALLOWEDIPS]				= { .type = NLA_NESTED },
+	[WGPEER_A_PROTOCOL_VERSION]			= { .type = NLA_U32 }
+};
+
+static const struct nla_policy allowedip_policy[WGALLOWEDIP_A_MAX + 1] = {
+	[WGALLOWEDIP_A_FAMILY]		= { .type = NLA_U16 },
+	[WGALLOWEDIP_A_IPADDR]		= { .type = NLA_MIN_LEN, .len = sizeof(struct in_addr) },
+	[WGALLOWEDIP_A_CIDR_MASK]	= { .type = NLA_U8 }
+};
+
+static struct wg_device *lookup_interface(struct nlattr **attrs,
+					  struct sk_buff *skb)
+{
+	struct net_device *dev = NULL;
+
+	if (!attrs[WGDEVICE_A_IFINDEX] == !attrs[WGDEVICE_A_IFNAME])
+		return ERR_PTR(-EBADR);
+	if (attrs[WGDEVICE_A_IFINDEX])
+		dev = dev_get_by_index(sock_net(skb->sk),
+				       nla_get_u32(attrs[WGDEVICE_A_IFINDEX]));
+	else if (attrs[WGDEVICE_A_IFNAME])
+		dev = dev_get_by_name(sock_net(skb->sk),
+				      nla_data(attrs[WGDEVICE_A_IFNAME]));
+	if (!dev)
+		return ERR_PTR(-ENODEV);
+	if (!dev->rtnl_link_ops || !dev->rtnl_link_ops->kind ||
+	    strcmp(dev->rtnl_link_ops->kind, KBUILD_MODNAME)) {
+		dev_put(dev);
+		return ERR_PTR(-EOPNOTSUPP);
+	}
+	return netdev_priv(dev);
+}
+
+static int get_allowedips(struct sk_buff *skb, const u8 *ip, u8 cidr,
+			  int family)
+{
+	struct nlattr *allowedip_nest;
+
+	allowedip_nest = nla_nest_start(skb, 0);
+	if (!allowedip_nest)
+		return -EMSGSIZE;
+
+	if (nla_put_u8(skb, WGALLOWEDIP_A_CIDR_MASK, cidr) ||
+	    nla_put_u16(skb, WGALLOWEDIP_A_FAMILY, family) ||
+	    nla_put(skb, WGALLOWEDIP_A_IPADDR, family == AF_INET6 ?
+		    sizeof(struct in6_addr) : sizeof(struct in_addr), ip)) {
+		nla_nest_cancel(skb, allowedip_nest);
+		return -EMSGSIZE;
+	}
+
+	nla_nest_end(skb, allowedip_nest);
+	return 0;
+}
+
+struct dump_ctx {
+	struct wg_device *wg;
+	struct wg_peer *next_peer;
+	u64 allowedips_seq;
+	struct allowedips_node *next_allowedip;
+};
+
+#define DUMP_CTX(cb) ((struct dump_ctx *)(cb)->args)
+
+static int
+get_peer(struct wg_peer *peer, struct sk_buff *skb, struct dump_ctx *ctx)
+{
+
+	struct nlattr *allowedips_nest, *peer_nest = nla_nest_start(skb, 0);
+	struct allowedips_node *allowedips_node = ctx->next_allowedip;
+	bool fail;
+
+	if (!peer_nest)
+		return -EMSGSIZE;
+
+	down_read(&peer->handshake.lock);
+	fail = nla_put(skb, WGPEER_A_PUBLIC_KEY, NOISE_PUBLIC_KEY_LEN,
+		       peer->handshake.remote_static);
+	up_read(&peer->handshake.lock);
+	if (fail)
+		goto err;
+
+	if (!allowedips_node) {
+		const struct __kernel_timespec last_handshake = {
+			.tv_sec = peer->walltime_last_handshake.tv_sec,
+			.tv_nsec = peer->walltime_last_handshake.tv_nsec
+		};
+
+		down_read(&peer->handshake.lock);
+		fail = nla_put(skb, WGPEER_A_PRESHARED_KEY,
+			       NOISE_SYMMETRIC_KEY_LEN,
+			       peer->handshake.preshared_key);
+		up_read(&peer->handshake.lock);
+		if (fail)
+			goto err;
+
+		if (nla_put(skb, WGPEER_A_LAST_HANDSHAKE_TIME,
+			    sizeof(last_handshake), &last_handshake) ||
+		    nla_put_u16(skb, WGPEER_A_PERSISTENT_KEEPALIVE_INTERVAL,
+				peer->persistent_keepalive_interval) ||
+		    nla_put_u64_64bit(skb, WGPEER_A_TX_BYTES, peer->tx_bytes,
+				      WGPEER_A_UNSPEC) ||
+		    nla_put_u64_64bit(skb, WGPEER_A_RX_BYTES, peer->rx_bytes,
+				      WGPEER_A_UNSPEC) ||
+		    nla_put_u32(skb, WGPEER_A_PROTOCOL_VERSION, 1))
+			goto err;
+
+		read_lock_bh(&peer->endpoint_lock);
+		if (peer->endpoint.addr.sa_family == AF_INET)
+			fail = nla_put(skb, WGPEER_A_ENDPOINT,
+				       sizeof(peer->endpoint.addr4),
+				       &peer->endpoint.addr4);
+		else if (peer->endpoint.addr.sa_family == AF_INET6)
+			fail = nla_put(skb, WGPEER_A_ENDPOINT,
+				       sizeof(peer->endpoint.addr6),
+				       &peer->endpoint.addr6);
+		read_unlock_bh(&peer->endpoint_lock);
+		if (fail)
+			goto err;
+		allowedips_node =
+			list_first_entry_or_null(&peer->allowedips_list,
+					struct allowedips_node, peer_list);
+	}
+	if (!allowedips_node)
+		goto no_allowedips;
+	if (!ctx->allowedips_seq)
+		ctx->allowedips_seq = peer->device->peer_allowedips.seq;
+	else if (ctx->allowedips_seq != peer->device->peer_allowedips.seq)
+		goto no_allowedips;
+
+	allowedips_nest = nla_nest_start(skb, WGPEER_A_ALLOWEDIPS);
+	if (!allowedips_nest)
+		goto err;
+
+	list_for_each_entry_from(allowedips_node, &peer->allowedips_list,
+				 peer_list) {
+		u8 cidr, ip[16] __aligned(__alignof(u64));
+		int family;
+
+		family = wg_allowedips_read_node(allowedips_node, ip, &cidr);
+		if (get_allowedips(skb, ip, cidr, family)) {
+			nla_nest_end(skb, allowedips_nest);
+			nla_nest_end(skb, peer_nest);
+			ctx->next_allowedip = allowedips_node;
+			return -EMSGSIZE;
+		}
+	}
+	nla_nest_end(skb, allowedips_nest);
+no_allowedips:
+	nla_nest_end(skb, peer_nest);
+	ctx->next_allowedip = NULL;
+	ctx->allowedips_seq = 0;
+	return 0;
+err:
+	nla_nest_cancel(skb, peer_nest);
+	return -EMSGSIZE;
+}
+
+static int wg_get_device_start(struct netlink_callback *cb)
+{
+	struct wg_device *wg;
+
+	wg = lookup_interface(genl_dumpit_info(cb)->attrs, cb->skb);
+	if (IS_ERR(wg))
+		return PTR_ERR(wg);
+	DUMP_CTX(cb)->wg = wg;
+	return 0;
+}
+
+static int wg_get_device_dump(struct sk_buff *skb, struct netlink_callback *cb)
+{
+	struct wg_peer *peer, *next_peer_cursor;
+	struct dump_ctx *ctx = DUMP_CTX(cb);
+	struct wg_device *wg = ctx->wg;
+	struct nlattr *peers_nest;
+	int ret = -EMSGSIZE;
+	bool done = true;
+	void *hdr;
+
+	rtnl_lock();
+	mutex_lock(&wg->device_update_lock);
+	cb->seq = wg->device_update_gen;
+	next_peer_cursor = ctx->next_peer;
+
+	hdr = genlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq,
+			  &genl_family, NLM_F_MULTI, WG_CMD_GET_DEVICE);
+	if (!hdr)
+		goto out;
+	genl_dump_check_consistent(cb, hdr);
+
+	if (!ctx->next_peer) {
+		if (nla_put_u16(skb, WGDEVICE_A_LISTEN_PORT,
+				wg->incoming_port) ||
+		    nla_put_u32(skb, WGDEVICE_A_FWMARK, wg->fwmark) ||
+		    nla_put_u32(skb, WGDEVICE_A_IFINDEX, wg->dev->ifindex) ||
+		    nla_put_string(skb, WGDEVICE_A_IFNAME, wg->dev->name))
+			goto out;
+
+		down_read(&wg->static_identity.lock);
+		if (wg->static_identity.has_identity) {
+			if (nla_put(skb, WGDEVICE_A_PRIVATE_KEY,
+				    NOISE_PUBLIC_KEY_LEN,
+				    wg->static_identity.static_private) ||
+			    nla_put(skb, WGDEVICE_A_PUBLIC_KEY,
+				    NOISE_PUBLIC_KEY_LEN,
+				    wg->static_identity.static_public)) {
+				up_read(&wg->static_identity.lock);
+				goto out;
+			}
+		}
+		up_read(&wg->static_identity.lock);
+	}
+
+	peers_nest = nla_nest_start(skb, WGDEVICE_A_PEERS);
+	if (!peers_nest)
+		goto out;
+	ret = 0;
+	/* If the last cursor was removed via list_del_init in peer_remove, then
+	 * we just treat this the same as there being no more peers left. The
+	 * reason is that seq_nr should indicate to userspace that this isn't a
+	 * coherent dump anyway, so they'll try again.
+	 */
+	if (list_empty(&wg->peer_list) ||
+	    (ctx->next_peer && list_empty(&ctx->next_peer->peer_list))) {
+		nla_nest_cancel(skb, peers_nest);
+		goto out;
+	}
+	lockdep_assert_held(&wg->device_update_lock);
+	peer = list_prepare_entry(ctx->next_peer, &wg->peer_list, peer_list);
+	list_for_each_entry_continue(peer, &wg->peer_list, peer_list) {
+		if (get_peer(peer, skb, ctx)) {
+			done = false;
+			break;
+		}
+		next_peer_cursor = peer;
+	}
+	nla_nest_end(skb, peers_nest);
+
+out:
+	if (!ret && !done && next_peer_cursor)
+		wg_peer_get(next_peer_cursor);
+	wg_peer_put(ctx->next_peer);
+	mutex_unlock(&wg->device_update_lock);
+	rtnl_unlock();
+
+	if (ret) {
+		genlmsg_cancel(skb, hdr);
+		return ret;
+	}
+	genlmsg_end(skb, hdr);
+	if (done) {
+		ctx->next_peer = NULL;
+		return 0;
+	}
+	ctx->next_peer = next_peer_cursor;
+	return skb->len;
+
+	/* At this point, we can't really deal ourselves with safely zeroing out
+	 * the private key material after usage. This will need an additional API
+	 * in the kernel for marking skbs as zero_on_free.
+	 */
+}
+
+static int wg_get_device_done(struct netlink_callback *cb)
+{
+	struct dump_ctx *ctx = DUMP_CTX(cb);
+
+	if (ctx->wg)
+		dev_put(ctx->wg->dev);
+	wg_peer_put(ctx->next_peer);
+	return 0;
+}
+
+static int set_port(struct wg_device *wg, u16 port)
+{
+	struct wg_peer *peer;
+
+	if (wg->incoming_port == port)
+		return 0;
+	list_for_each_entry(peer, &wg->peer_list, peer_list)
+		wg_socket_clear_peer_endpoint_src(peer);
+	if (!netif_running(wg->dev)) {
+		wg->incoming_port = port;
+		return 0;
+	}
+	return wg_socket_init(wg, port);
+}
+
+static int set_allowedip(struct wg_peer *peer, struct nlattr **attrs)
+{
+	int ret = -EINVAL;
+	u16 family;
+	u8 cidr;
+
+	if (!attrs[WGALLOWEDIP_A_FAMILY] || !attrs[WGALLOWEDIP_A_IPADDR] ||
+	    !attrs[WGALLOWEDIP_A_CIDR_MASK])
+		return ret;
+	family = nla_get_u16(attrs[WGALLOWEDIP_A_FAMILY]);
+	cidr = nla_get_u8(attrs[WGALLOWEDIP_A_CIDR_MASK]);
+
+	if (family == AF_INET && cidr <= 32 &&
+	    nla_len(attrs[WGALLOWEDIP_A_IPADDR]) == sizeof(struct in_addr))
+		ret = wg_allowedips_insert_v4(
+			&peer->device->peer_allowedips,
+			nla_data(attrs[WGALLOWEDIP_A_IPADDR]), cidr, peer,
+			&peer->device->device_update_lock);
+	else if (family == AF_INET6 && cidr <= 128 &&
+		 nla_len(attrs[WGALLOWEDIP_A_IPADDR]) == sizeof(struct in6_addr))
+		ret = wg_allowedips_insert_v6(
+			&peer->device->peer_allowedips,
+			nla_data(attrs[WGALLOWEDIP_A_IPADDR]), cidr, peer,
+			&peer->device->device_update_lock);
+
+	return ret;
+}
+
+static int set_peer(struct wg_device *wg, struct nlattr **attrs)
+{
+	u8 *public_key = NULL, *preshared_key = NULL;
+	struct wg_peer *peer = NULL;
+	u32 flags = 0;
+	int ret;
+
+	ret = -EINVAL;
+	if (attrs[WGPEER_A_PUBLIC_KEY] &&
+	    nla_len(attrs[WGPEER_A_PUBLIC_KEY]) == NOISE_PUBLIC_KEY_LEN)
+		public_key = nla_data(attrs[WGPEER_A_PUBLIC_KEY]);
+	else
+		goto out;
+	if (attrs[WGPEER_A_PRESHARED_KEY] &&
+	    nla_len(attrs[WGPEER_A_PRESHARED_KEY]) == NOISE_SYMMETRIC_KEY_LEN)
+		preshared_key = nla_data(attrs[WGPEER_A_PRESHARED_KEY]);
+
+	if (attrs[WGPEER_A_FLAGS])
+		flags = nla_get_u32(attrs[WGPEER_A_FLAGS]);
+	ret = -EOPNOTSUPP;
+	if (flags & ~__WGPEER_F_ALL)
+		goto out;
+
+	ret = -EPFNOSUPPORT;
+	if (attrs[WGPEER_A_PROTOCOL_VERSION]) {
+		if (nla_get_u32(attrs[WGPEER_A_PROTOCOL_VERSION]) != 1)
+			goto out;
+	}
+
+	peer = wg_pubkey_hashtable_lookup(wg->peer_hashtable,
+					  nla_data(attrs[WGPEER_A_PUBLIC_KEY]));
+	ret = 0;
+	if (!peer) { /* Peer doesn't exist yet. Add a new one. */
+		if (flags & (WGPEER_F_REMOVE_ME | WGPEER_F_UPDATE_ONLY))
+			goto out;
+
+		/* The peer is new, so there aren't allowed IPs to remove. */
+		flags &= ~WGPEER_F_REPLACE_ALLOWEDIPS;
+
+		down_read(&wg->static_identity.lock);
+		if (wg->static_identity.has_identity &&
+		    !memcmp(nla_data(attrs[WGPEER_A_PUBLIC_KEY]),
+			    wg->static_identity.static_public,
+			    NOISE_PUBLIC_KEY_LEN)) {
+			/* We silently ignore peers that have the same public
+			 * key as the device. The reason we do it silently is
+			 * that we'd like for people to be able to reuse the
+			 * same set of API calls across peers.
+			 */
+			up_read(&wg->static_identity.lock);
+			ret = 0;
+			goto out;
+		}
+		up_read(&wg->static_identity.lock);
+
+		peer = wg_peer_create(wg, public_key, preshared_key);
+		if (IS_ERR(peer)) {
+			/* Similar to the above, if the key is invalid, we skip
+			 * it without fanfare, so that services don't need to
+			 * worry about doing key validation themselves.
+			 */
+			ret = PTR_ERR(peer) == -EKEYREJECTED ? 0 : PTR_ERR(peer);
+			peer = NULL;
+			goto out;
+		}
+		/* Take additional reference, as though we've just been
+		 * looked up.
+		 */
+		wg_peer_get(peer);
+	}
+
+	if (flags & WGPEER_F_REMOVE_ME) {
+		wg_peer_remove(peer);
+		goto out;
+	}
+
+	if (preshared_key) {
+		down_write(&peer->handshake.lock);
+		memcpy(&peer->handshake.preshared_key, preshared_key,
+		       NOISE_SYMMETRIC_KEY_LEN);
+		up_write(&peer->handshake.lock);
+	}
+
+	if (attrs[WGPEER_A_ENDPOINT]) {
+		struct sockaddr *addr = nla_data(attrs[WGPEER_A_ENDPOINT]);
+		size_t len = nla_len(attrs[WGPEER_A_ENDPOINT]);
+
+		if ((len == sizeof(struct sockaddr_in) &&
+		     addr->sa_family == AF_INET) ||
+		    (len == sizeof(struct sockaddr_in6) &&
+		     addr->sa_family == AF_INET6)) {
+			struct endpoint endpoint = { { { 0 } } };
+
+			memcpy(&endpoint.addr, addr, len);
+			wg_socket_set_peer_endpoint(peer, &endpoint);
+		}
+	}
+
+	if (flags & WGPEER_F_REPLACE_ALLOWEDIPS)
+		wg_allowedips_remove_by_peer(&wg->peer_allowedips, peer,
+					     &wg->device_update_lock);
+
+	if (attrs[WGPEER_A_ALLOWEDIPS]) {
+		struct nlattr *attr, *allowedip[WGALLOWEDIP_A_MAX + 1];
+		int rem;
+
+		nla_for_each_nested(attr, attrs[WGPEER_A_ALLOWEDIPS], rem) {
+			ret = nla_parse_nested(allowedip, WGALLOWEDIP_A_MAX,
+					       attr, allowedip_policy, NULL);
+			if (ret < 0)
+				goto out;
+			ret = set_allowedip(peer, allowedip);
+			if (ret < 0)
+				goto out;
+		}
+	}
+
+	if (attrs[WGPEER_A_PERSISTENT_KEEPALIVE_INTERVAL]) {
+		const u16 persistent_keepalive_interval = nla_get_u16(
+				attrs[WGPEER_A_PERSISTENT_KEEPALIVE_INTERVAL]);
+		const bool send_keepalive =
+			!peer->persistent_keepalive_interval &&
+			persistent_keepalive_interval &&
+			netif_running(wg->dev);
+
+		peer->persistent_keepalive_interval = persistent_keepalive_interval;
+		if (send_keepalive)
+			wg_packet_send_keepalive(peer);
+	}
+
+	if (netif_running(wg->dev))
+		wg_packet_send_staged_packets(peer);
+
+out:
+	wg_peer_put(peer);
+	if (attrs[WGPEER_A_PRESHARED_KEY])
+		memzero_explicit(nla_data(attrs[WGPEER_A_PRESHARED_KEY]),
+				 nla_len(attrs[WGPEER_A_PRESHARED_KEY]));
+	return ret;
+}
+
+static int wg_set_device(struct sk_buff *skb, struct genl_info *info)
+{
+	struct wg_device *wg = lookup_interface(info->attrs, skb);
+	u32 flags = 0;
+	int ret;
+
+	if (IS_ERR(wg)) {
+		ret = PTR_ERR(wg);
+		goto out_nodev;
+	}
+
+	rtnl_lock();
+	mutex_lock(&wg->device_update_lock);
+
+	if (info->attrs[WGDEVICE_A_FLAGS])
+		flags = nla_get_u32(info->attrs[WGDEVICE_A_FLAGS]);
+	ret = -EOPNOTSUPP;
+	if (flags & ~__WGDEVICE_F_ALL)
+		goto out;
+
+	ret = -EPERM;
+	if ((info->attrs[WGDEVICE_A_LISTEN_PORT] ||
+	     info->attrs[WGDEVICE_A_FWMARK]) &&
+	    !ns_capable(wg->creating_net->user_ns, CAP_NET_ADMIN))
+		goto out;
+
+	++wg->device_update_gen;
+
+	if (info->attrs[WGDEVICE_A_FWMARK]) {
+		struct wg_peer *peer;
+
+		wg->fwmark = nla_get_u32(info->attrs[WGDEVICE_A_FWMARK]);
+		list_for_each_entry(peer, &wg->peer_list, peer_list)
+			wg_socket_clear_peer_endpoint_src(peer);
+	}
+
+	if (info->attrs[WGDEVICE_A_LISTEN_PORT]) {
+		ret = set_port(wg,
+			nla_get_u16(info->attrs[WGDEVICE_A_LISTEN_PORT]));
+		if (ret)
+			goto out;
+	}
+
+	if (flags & WGDEVICE_F_REPLACE_PEERS)
+		wg_peer_remove_all(wg);
+
+	if (info->attrs[WGDEVICE_A_PRIVATE_KEY] &&
+	    nla_len(info->attrs[WGDEVICE_A_PRIVATE_KEY]) ==
+		    NOISE_PUBLIC_KEY_LEN) {
+		u8 *private_key = nla_data(info->attrs[WGDEVICE_A_PRIVATE_KEY]);
+		u8 public_key[NOISE_PUBLIC_KEY_LEN];
+		struct wg_peer *peer, *temp;
+
+		if (!crypto_memneq(wg->static_identity.static_private,
+				   private_key, NOISE_PUBLIC_KEY_LEN))
+			goto skip_set_private_key;
+
+		/* We remove before setting, to prevent race, which means doing
+		 * two 25519-genpub ops.
+		 */
+		if (curve25519_generate_public(public_key, private_key)) {
+			peer = wg_pubkey_hashtable_lookup(wg->peer_hashtable,
+							  public_key);
+			if (peer) {
+				wg_peer_put(peer);
+				wg_peer_remove(peer);
+			}
+		}
+
+		down_write(&wg->static_identity.lock);
+		wg_noise_set_static_identity_private_key(&wg->static_identity,
+							 private_key);
+		list_for_each_entry_safe(peer, temp, &wg->peer_list,
+					 peer_list) {
+			if (wg_noise_precompute_static_static(peer))
+				wg_noise_expire_current_peer_keypairs(peer);
+			else
+				wg_peer_remove(peer);
+		}
+		wg_cookie_checker_precompute_device_keys(&wg->cookie_checker);
+		up_write(&wg->static_identity.lock);
+	}
+skip_set_private_key:
+
+	if (info->attrs[WGDEVICE_A_PEERS]) {
+		struct nlattr *attr, *peer[WGPEER_A_MAX + 1];
+		int rem;
+
+		nla_for_each_nested(attr, info->attrs[WGDEVICE_A_PEERS], rem) {
+			ret = nla_parse_nested(peer, WGPEER_A_MAX, attr,
+					       peer_policy, NULL);
+			if (ret < 0)
+				goto out;
+			ret = set_peer(wg, peer);
+			if (ret < 0)
+				goto out;
+		}
+	}
+	ret = 0;
+
+out:
+	mutex_unlock(&wg->device_update_lock);
+	rtnl_unlock();
+	dev_put(wg->dev);
+out_nodev:
+	if (info->attrs[WGDEVICE_A_PRIVATE_KEY])
+		memzero_explicit(nla_data(info->attrs[WGDEVICE_A_PRIVATE_KEY]),
+				 nla_len(info->attrs[WGDEVICE_A_PRIVATE_KEY]));
+	return ret;
+}
+
+static const struct genl_ops genl_ops[] = {
+	{
+		.cmd = WG_CMD_GET_DEVICE,
+		.start = wg_get_device_start,
+		.dumpit = wg_get_device_dump,
+		.done = wg_get_device_done,
+		.flags = GENL_UNS_ADMIN_PERM
+	}, {
+		.cmd = WG_CMD_SET_DEVICE,
+		.doit = wg_set_device,
+		.flags = GENL_UNS_ADMIN_PERM
+	}
+};
+
+static struct genl_family genl_family __ro_after_init = {
+	.ops = genl_ops,
+	.n_ops = ARRAY_SIZE(genl_ops),
+	.name = WG_GENL_NAME,
+	.version = WG_GENL_VERSION,
+	.maxattr = WGDEVICE_A_MAX,
+	.module = THIS_MODULE,
+	.policy = device_policy,
+	.netnsok = true
+};
+
+int __init wg_genetlink_init(void)
+{
+	return genl_register_family(&genl_family);
+}
+
+void __exit wg_genetlink_uninit(void)
+{
+	genl_unregister_family(&genl_family);
+}
diff --git a/drivers/net/wireguard/netlink.h b/drivers/net/wireguard/netlink.h
new file mode 100644
index 000000000000..15100d92e2e3
--- /dev/null
+++ b/drivers/net/wireguard/netlink.h
@@ -0,0 +1,12 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#ifndef _WG_NETLINK_H
+#define _WG_NETLINK_H
+
+int wg_genetlink_init(void);
+void wg_genetlink_uninit(void);
+
+#endif /* _WG_NETLINK_H */
diff --git a/drivers/net/wireguard/noise.c b/drivers/net/wireguard/noise.c
new file mode 100644
index 000000000000..d71c8db68a8c
--- /dev/null
+++ b/drivers/net/wireguard/noise.c
@@ -0,0 +1,828 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#include "noise.h"
+#include "device.h"
+#include "peer.h"
+#include "messages.h"
+#include "queueing.h"
+#include "peerlookup.h"
+
+#include <linux/rcupdate.h>
+#include <linux/slab.h>
+#include <linux/bitmap.h>
+#include <linux/scatterlist.h>
+#include <linux/highmem.h>
+#include <crypto/algapi.h>
+
+/* This implements Noise_IKpsk2:
+ *
+ * <- s
+ * ******
+ * -> e, es, s, ss, {t}
+ * <- e, ee, se, psk, {}
+ */
+
+static const u8 handshake_name[37] = "Noise_IKpsk2_25519_ChaChaPoly_BLAKE2s";
+static const u8 identifier_name[34] = "WireGuard v1 zx2c4 Jason@zx2c4.com";
+static u8 handshake_init_hash[NOISE_HASH_LEN] __ro_after_init;
+static u8 handshake_init_chaining_key[NOISE_HASH_LEN] __ro_after_init;
+static atomic64_t keypair_counter = ATOMIC64_INIT(0);
+
+void __init wg_noise_init(void)
+{
+	struct blake2s_state blake;
+
+	blake2s(handshake_init_chaining_key, handshake_name, NULL,
+		NOISE_HASH_LEN, sizeof(handshake_name), 0);
+	blake2s_init(&blake, NOISE_HASH_LEN);
+	blake2s_update(&blake, handshake_init_chaining_key, NOISE_HASH_LEN);
+	blake2s_update(&blake, identifier_name, sizeof(identifier_name));
+	blake2s_final(&blake, handshake_init_hash);
+}
+
+/* Must hold peer->handshake.static_identity->lock */
+bool wg_noise_precompute_static_static(struct wg_peer *peer)
+{
+	bool ret = true;
+
+	down_write(&peer->handshake.lock);
+	if (peer->handshake.static_identity->has_identity)
+		ret = curve25519(
+			peer->handshake.precomputed_static_static,
+			peer->handshake.static_identity->static_private,
+			peer->handshake.remote_static);
+	else
+		memset(peer->handshake.precomputed_static_static, 0,
+		       NOISE_PUBLIC_KEY_LEN);
+	up_write(&peer->handshake.lock);
+	return ret;
+}
+
+bool wg_noise_handshake_init(struct noise_handshake *handshake,
+			   struct noise_static_identity *static_identity,
+			   const u8 peer_public_key[NOISE_PUBLIC_KEY_LEN],
+			   const u8 peer_preshared_key[NOISE_SYMMETRIC_KEY_LEN],
+			   struct wg_peer *peer)
+{
+	memset(handshake, 0, sizeof(*handshake));
+	init_rwsem(&handshake->lock);
+	handshake->entry.type = INDEX_HASHTABLE_HANDSHAKE;
+	handshake->entry.peer = peer;
+	memcpy(handshake->remote_static, peer_public_key, NOISE_PUBLIC_KEY_LEN);
+	if (peer_preshared_key)
+		memcpy(handshake->preshared_key, peer_preshared_key,
+		       NOISE_SYMMETRIC_KEY_LEN);
+	handshake->static_identity = static_identity;
+	handshake->state = HANDSHAKE_ZEROED;
+	return wg_noise_precompute_static_static(peer);
+}
+
+static void handshake_zero(struct noise_handshake *handshake)
+{
+	memset(&handshake->ephemeral_private, 0, NOISE_PUBLIC_KEY_LEN);
+	memset(&handshake->remote_ephemeral, 0, NOISE_PUBLIC_KEY_LEN);
+	memset(&handshake->hash, 0, NOISE_HASH_LEN);
+	memset(&handshake->chaining_key, 0, NOISE_HASH_LEN);
+	handshake->remote_index = 0;
+	handshake->state = HANDSHAKE_ZEROED;
+}
+
+void wg_noise_handshake_clear(struct noise_handshake *handshake)
+{
+	wg_index_hashtable_remove(
+			handshake->entry.peer->device->index_hashtable,
+			&handshake->entry);
+	down_write(&handshake->lock);
+	handshake_zero(handshake);
+	up_write(&handshake->lock);
+	wg_index_hashtable_remove(
+			handshake->entry.peer->device->index_hashtable,
+			&handshake->entry);
+}
+
+static struct noise_keypair *keypair_create(struct wg_peer *peer)
+{
+	struct noise_keypair *keypair = kzalloc(sizeof(*keypair), GFP_KERNEL);
+
+	if (unlikely(!keypair))
+		return NULL;
+	keypair->internal_id = atomic64_inc_return(&keypair_counter);
+	keypair->entry.type = INDEX_HASHTABLE_KEYPAIR;
+	keypair->entry.peer = peer;
+	kref_init(&keypair->refcount);
+	return keypair;
+}
+
+static void keypair_free_rcu(struct rcu_head *rcu)
+{
+	kzfree(container_of(rcu, struct noise_keypair, rcu));
+}
+
+static void keypair_free_kref(struct kref *kref)
+{
+	struct noise_keypair *keypair =
+		container_of(kref, struct noise_keypair, refcount);
+
+	net_dbg_ratelimited("%s: Keypair %llu destroyed for peer %llu\n",
+			    keypair->entry.peer->device->dev->name,
+			    keypair->internal_id,
+			    keypair->entry.peer->internal_id);
+	wg_index_hashtable_remove(keypair->entry.peer->device->index_hashtable,
+				  &keypair->entry);
+	call_rcu(&keypair->rcu, keypair_free_rcu);
+}
+
+void wg_noise_keypair_put(struct noise_keypair *keypair, bool unreference_now)
+{
+	if (unlikely(!keypair))
+		return;
+	if (unlikely(unreference_now))
+		wg_index_hashtable_remove(
+			keypair->entry.peer->device->index_hashtable,
+			&keypair->entry);
+	kref_put(&keypair->refcount, keypair_free_kref);
+}
+
+struct noise_keypair *wg_noise_keypair_get(struct noise_keypair *keypair)
+{
+	RCU_LOCKDEP_WARN(!rcu_read_lock_bh_held(),
+		"Taking noise keypair reference without holding the RCU BH read lock");
+	if (unlikely(!keypair || !kref_get_unless_zero(&keypair->refcount)))
+		return NULL;
+	return keypair;
+}
+
+void wg_noise_keypairs_clear(struct noise_keypairs *keypairs)
+{
+	struct noise_keypair *old;
+
+	spin_lock_bh(&keypairs->keypair_update_lock);
+
+	/* We zero the next_keypair before zeroing the others, so that
+	 * wg_noise_received_with_keypair returns early before subsequent ones
+	 * are zeroed.
+	 */
+	old = rcu_dereference_protected(keypairs->next_keypair,
+		lockdep_is_held(&keypairs->keypair_update_lock));
+	RCU_INIT_POINTER(keypairs->next_keypair, NULL);
+	wg_noise_keypair_put(old, true);
+
+	old = rcu_dereference_protected(keypairs->previous_keypair,
+		lockdep_is_held(&keypairs->keypair_update_lock));
+	RCU_INIT_POINTER(keypairs->previous_keypair, NULL);
+	wg_noise_keypair_put(old, true);
+
+	old = rcu_dereference_protected(keypairs->current_keypair,
+		lockdep_is_held(&keypairs->keypair_update_lock));
+	RCU_INIT_POINTER(keypairs->current_keypair, NULL);
+	wg_noise_keypair_put(old, true);
+
+	spin_unlock_bh(&keypairs->keypair_update_lock);
+}
+
+void wg_noise_expire_current_peer_keypairs(struct wg_peer *peer)
+{
+	struct noise_keypair *keypair;
+
+	wg_noise_handshake_clear(&peer->handshake);
+	wg_noise_reset_last_sent_handshake(&peer->last_sent_handshake);
+
+	spin_lock_bh(&peer->keypairs.keypair_update_lock);
+	keypair = rcu_dereference_protected(peer->keypairs.next_keypair,
+			lockdep_is_held(&peer->keypairs.keypair_update_lock));
+	if (keypair)
+		keypair->sending.is_valid = false;
+	keypair = rcu_dereference_protected(peer->keypairs.current_keypair,
+			lockdep_is_held(&peer->keypairs.keypair_update_lock));
+	if (keypair)
+		keypair->sending.is_valid = false;
+	spin_unlock_bh(&peer->keypairs.keypair_update_lock);
+}
+
+static void add_new_keypair(struct noise_keypairs *keypairs,
+			    struct noise_keypair *new_keypair)
+{
+	struct noise_keypair *previous_keypair, *next_keypair, *current_keypair;
+
+	spin_lock_bh(&keypairs->keypair_update_lock);
+	previous_keypair = rcu_dereference_protected(keypairs->previous_keypair,
+		lockdep_is_held(&keypairs->keypair_update_lock));
+	next_keypair = rcu_dereference_protected(keypairs->next_keypair,
+		lockdep_is_held(&keypairs->keypair_update_lock));
+	current_keypair = rcu_dereference_protected(keypairs->current_keypair,
+		lockdep_is_held(&keypairs->keypair_update_lock));
+	if (new_keypair->i_am_the_initiator) {
+		/* If we're the initiator, it means we've sent a handshake, and
+		 * received a confirmation response, which means this new
+		 * keypair can now be used.
+		 */
+		if (next_keypair) {
+			/* If there already was a next keypair pending, we
+			 * demote it to be the previous keypair, and free the
+			 * existing current. Note that this means KCI can result
+			 * in this transition. It would perhaps be more sound to
+			 * always just get rid of the unused next keypair
+			 * instead of putting it in the previous slot, but this
+			 * might be a bit less robust. Something to think about
+			 * for the future.
+			 */
+			RCU_INIT_POINTER(keypairs->next_keypair, NULL);
+			rcu_assign_pointer(keypairs->previous_keypair,
+					   next_keypair);
+			wg_noise_keypair_put(current_keypair, true);
+		} else /* If there wasn't an existing next keypair, we replace
+			* the previous with the current one.
+			*/
+			rcu_assign_pointer(keypairs->previous_keypair,
+					   current_keypair);
+		/* At this point we can get rid of the old previous keypair, and
+		 * set up the new keypair.
+		 */
+		wg_noise_keypair_put(previous_keypair, true);
+		rcu_assign_pointer(keypairs->current_keypair, new_keypair);
+	} else {
+		/* If we're the responder, it means we can't use the new keypair
+		 * until we receive confirmation via the first data packet, so
+		 * we get rid of the existing previous one, the possibly
+		 * existing next one, and slide in the new next one.
+		 */
+		rcu_assign_pointer(keypairs->next_keypair, new_keypair);
+		wg_noise_keypair_put(next_keypair, true);
+		RCU_INIT_POINTER(keypairs->previous_keypair, NULL);
+		wg_noise_keypair_put(previous_keypair, true);
+	}
+	spin_unlock_bh(&keypairs->keypair_update_lock);
+}
+
+bool wg_noise_received_with_keypair(struct noise_keypairs *keypairs,
+				    struct noise_keypair *received_keypair)
+{
+	struct noise_keypair *old_keypair;
+	bool key_is_new;
+
+	/* We first check without taking the spinlock. */
+	key_is_new = received_keypair ==
+		     rcu_access_pointer(keypairs->next_keypair);
+	if (likely(!key_is_new))
+		return false;
+
+	spin_lock_bh(&keypairs->keypair_update_lock);
+	/* After locking, we double check that things didn't change from
+	 * beneath us.
+	 */
+	if (unlikely(received_keypair !=
+		    rcu_dereference_protected(keypairs->next_keypair,
+			    lockdep_is_held(&keypairs->keypair_update_lock)))) {
+		spin_unlock_bh(&keypairs->keypair_update_lock);
+		return false;
+	}
+
+	/* When we've finally received the confirmation, we slide the next
+	 * into the current, the current into the previous, and get rid of
+	 * the old previous.
+	 */
+	old_keypair = rcu_dereference_protected(keypairs->previous_keypair,
+		lockdep_is_held(&keypairs->keypair_update_lock));
+	rcu_assign_pointer(keypairs->previous_keypair,
+		rcu_dereference_protected(keypairs->current_keypair,
+			lockdep_is_held(&keypairs->keypair_update_lock)));
+	wg_noise_keypair_put(old_keypair, true);
+	rcu_assign_pointer(keypairs->current_keypair, received_keypair);
+	RCU_INIT_POINTER(keypairs->next_keypair, NULL);
+
+	spin_unlock_bh(&keypairs->keypair_update_lock);
+	return true;
+}
+
+/* Must hold static_identity->lock */
+void wg_noise_set_static_identity_private_key(
+	struct noise_static_identity *static_identity,
+	const u8 private_key[NOISE_PUBLIC_KEY_LEN])
+{
+	memcpy(static_identity->static_private, private_key,
+	       NOISE_PUBLIC_KEY_LEN);
+	curve25519_clamp_secret(static_identity->static_private);
+	static_identity->has_identity = curve25519_generate_public(
+		static_identity->static_public, private_key);
+}
+
+/* This is Hugo Krawczyk's HKDF:
+ *  - https://eprint.iacr.org/2010/264.pdf
+ *  - https://tools.ietf.org/html/rfc5869
+ */
+static void kdf(u8 *first_dst, u8 *second_dst, u8 *third_dst, const u8 *data,
+		size_t first_len, size_t second_len, size_t third_len,
+		size_t data_len, const u8 chaining_key[NOISE_HASH_LEN])
+{
+	u8 output[BLAKE2S_HASH_SIZE + 1];
+	u8 secret[BLAKE2S_HASH_SIZE];
+
+	WARN_ON(IS_ENABLED(DEBUG) &&
+		(first_len > BLAKE2S_HASH_SIZE ||
+		 second_len > BLAKE2S_HASH_SIZE ||
+		 third_len > BLAKE2S_HASH_SIZE ||
+		 ((second_len || second_dst || third_len || third_dst) &&
+		  (!first_len || !first_dst)) ||
+		 ((third_len || third_dst) && (!second_len || !second_dst))));
+
+	/* Extract entropy from data into secret */
+	blake2s256_hmac(secret, data, chaining_key, data_len, NOISE_HASH_LEN);
+
+	if (!first_dst || !first_len)
+		goto out;
+
+	/* Expand first key: key = secret, data = 0x1 */
+	output[0] = 1;
+	blake2s256_hmac(output, output, secret, 1, BLAKE2S_HASH_SIZE);
+	memcpy(first_dst, output, first_len);
+
+	if (!second_dst || !second_len)
+		goto out;
+
+	/* Expand second key: key = secret, data = first-key || 0x2 */
+	output[BLAKE2S_HASH_SIZE] = 2;
+	blake2s256_hmac(output, output, secret, BLAKE2S_HASH_SIZE + 1,
+			BLAKE2S_HASH_SIZE);
+	memcpy(second_dst, output, second_len);
+
+	if (!third_dst || !third_len)
+		goto out;
+
+	/* Expand third key: key = secret, data = second-key || 0x3 */
+	output[BLAKE2S_HASH_SIZE] = 3;
+	blake2s256_hmac(output, output, secret, BLAKE2S_HASH_SIZE + 1,
+			BLAKE2S_HASH_SIZE);
+	memcpy(third_dst, output, third_len);
+
+out:
+	/* Clear sensitive data from stack */
+	memzero_explicit(secret, BLAKE2S_HASH_SIZE);
+	memzero_explicit(output, BLAKE2S_HASH_SIZE + 1);
+}
+
+static void symmetric_key_init(struct noise_symmetric_key *key)
+{
+	spin_lock_init(&key->counter.receive.lock);
+	atomic64_set(&key->counter.counter, 0);
+	memset(key->counter.receive.backtrack, 0,
+	       sizeof(key->counter.receive.backtrack));
+	key->birthdate = ktime_get_coarse_boottime_ns();
+	key->is_valid = true;
+}
+
+static void derive_keys(struct noise_symmetric_key *first_dst,
+			struct noise_symmetric_key *second_dst,
+			const u8 chaining_key[NOISE_HASH_LEN])
+{
+	kdf(first_dst->key, second_dst->key, NULL, NULL,
+	    NOISE_SYMMETRIC_KEY_LEN, NOISE_SYMMETRIC_KEY_LEN, 0, 0,
+	    chaining_key);
+	symmetric_key_init(first_dst);
+	symmetric_key_init(second_dst);
+}
+
+static bool __must_check mix_dh(u8 chaining_key[NOISE_HASH_LEN],
+				u8 key[NOISE_SYMMETRIC_KEY_LEN],
+				const u8 private[NOISE_PUBLIC_KEY_LEN],
+				const u8 public[NOISE_PUBLIC_KEY_LEN])
+{
+	u8 dh_calculation[NOISE_PUBLIC_KEY_LEN];
+
+	if (unlikely(!curve25519(dh_calculation, private, public)))
+		return false;
+	kdf(chaining_key, key, NULL, dh_calculation, NOISE_HASH_LEN,
+	    NOISE_SYMMETRIC_KEY_LEN, 0, NOISE_PUBLIC_KEY_LEN, chaining_key);
+	memzero_explicit(dh_calculation, NOISE_PUBLIC_KEY_LEN);
+	return true;
+}
+
+static void mix_hash(u8 hash[NOISE_HASH_LEN], const u8 *src, size_t src_len)
+{
+	struct blake2s_state blake;
+
+	blake2s_init(&blake, NOISE_HASH_LEN);
+	blake2s_update(&blake, hash, NOISE_HASH_LEN);
+	blake2s_update(&blake, src, src_len);
+	blake2s_final(&blake, hash);
+}
+
+static void mix_psk(u8 chaining_key[NOISE_HASH_LEN], u8 hash[NOISE_HASH_LEN],
+		    u8 key[NOISE_SYMMETRIC_KEY_LEN],
+		    const u8 psk[NOISE_SYMMETRIC_KEY_LEN])
+{
+	u8 temp_hash[NOISE_HASH_LEN];
+
+	kdf(chaining_key, temp_hash, key, psk, NOISE_HASH_LEN, NOISE_HASH_LEN,
+	    NOISE_SYMMETRIC_KEY_LEN, NOISE_SYMMETRIC_KEY_LEN, chaining_key);
+	mix_hash(hash, temp_hash, NOISE_HASH_LEN);
+	memzero_explicit(temp_hash, NOISE_HASH_LEN);
+}
+
+static void handshake_init(u8 chaining_key[NOISE_HASH_LEN],
+			   u8 hash[NOISE_HASH_LEN],
+			   const u8 remote_static[NOISE_PUBLIC_KEY_LEN])
+{
+	memcpy(hash, handshake_init_hash, NOISE_HASH_LEN);
+	memcpy(chaining_key, handshake_init_chaining_key, NOISE_HASH_LEN);
+	mix_hash(hash, remote_static, NOISE_PUBLIC_KEY_LEN);
+}
+
+static void message_encrypt(u8 *dst_ciphertext, const u8 *src_plaintext,
+			    size_t src_len, u8 key[NOISE_SYMMETRIC_KEY_LEN],
+			    u8 hash[NOISE_HASH_LEN])
+{
+	chacha20poly1305_encrypt(dst_ciphertext, src_plaintext, src_len, hash,
+				 NOISE_HASH_LEN,
+				 0 /* Always zero for Noise_IK */, key);
+	mix_hash(hash, dst_ciphertext, noise_encrypted_len(src_len));
+}
+
+static bool message_decrypt(u8 *dst_plaintext, const u8 *src_ciphertext,
+			    size_t src_len, u8 key[NOISE_SYMMETRIC_KEY_LEN],
+			    u8 hash[NOISE_HASH_LEN])
+{
+	if (!chacha20poly1305_decrypt(dst_plaintext, src_ciphertext, src_len,
+				      hash, NOISE_HASH_LEN,
+				      0 /* Always zero for Noise_IK */, key))
+		return false;
+	mix_hash(hash, src_ciphertext, src_len);
+	return true;
+}
+
+static void message_ephemeral(u8 ephemeral_dst[NOISE_PUBLIC_KEY_LEN],
+			      const u8 ephemeral_src[NOISE_PUBLIC_KEY_LEN],
+			      u8 chaining_key[NOISE_HASH_LEN],
+			      u8 hash[NOISE_HASH_LEN])
+{
+	if (ephemeral_dst != ephemeral_src)
+		memcpy(ephemeral_dst, ephemeral_src, NOISE_PUBLIC_KEY_LEN);
+	mix_hash(hash, ephemeral_src, NOISE_PUBLIC_KEY_LEN);
+	kdf(chaining_key, NULL, NULL, ephemeral_src, NOISE_HASH_LEN, 0, 0,
+	    NOISE_PUBLIC_KEY_LEN, chaining_key);
+}
+
+static void tai64n_now(u8 output[NOISE_TIMESTAMP_LEN])
+{
+	struct timespec64 now;
+
+	ktime_get_real_ts64(&now);
+
+	/* In order to prevent some sort of infoleak from precise timers, we
+	 * round down the nanoseconds part to the closest rounded-down power of
+	 * two to the maximum initiations per second allowed anyway by the
+	 * implementation.
+	 */
+	now.tv_nsec = ALIGN_DOWN(now.tv_nsec,
+		rounddown_pow_of_two(NSEC_PER_SEC / INITIATIONS_PER_SECOND));
+
+	/* https://cr.yp.to/libtai/tai64.html */
+	*(__be64 *)output = cpu_to_be64(0x400000000000000aULL + now.tv_sec);
+	*(__be32 *)(output + sizeof(__be64)) = cpu_to_be32(now.tv_nsec);
+}
+
+bool
+wg_noise_handshake_create_initiation(struct message_handshake_initiation *dst,
+				     struct noise_handshake *handshake)
+{
+	u8 timestamp[NOISE_TIMESTAMP_LEN];
+	u8 key[NOISE_SYMMETRIC_KEY_LEN];
+	bool ret = false;
+
+	/* We need to wait for crng _before_ taking any locks, since
+	 * curve25519_generate_secret uses get_random_bytes_wait.
+	 */
+	wait_for_random_bytes();
+
+	down_read(&handshake->static_identity->lock);
+	down_write(&handshake->lock);
+
+	if (unlikely(!handshake->static_identity->has_identity))
+		goto out;
+
+	dst->header.type = cpu_to_le32(MESSAGE_HANDSHAKE_INITIATION);
+
+	handshake_init(handshake->chaining_key, handshake->hash,
+		       handshake->remote_static);
+
+	/* e */
+	curve25519_generate_secret(handshake->ephemeral_private);
+	if (!curve25519_generate_public(dst->unencrypted_ephemeral,
+					handshake->ephemeral_private))
+		goto out;
+	message_ephemeral(dst->unencrypted_ephemeral,
+			  dst->unencrypted_ephemeral, handshake->chaining_key,
+			  handshake->hash);
+
+	/* es */
+	if (!mix_dh(handshake->chaining_key, key, handshake->ephemeral_private,
+		    handshake->remote_static))
+		goto out;
+
+	/* s */
+	message_encrypt(dst->encrypted_static,
+			handshake->static_identity->static_public,
+			NOISE_PUBLIC_KEY_LEN, key, handshake->hash);
+
+	/* ss */
+	kdf(handshake->chaining_key, key, NULL,
+	    handshake->precomputed_static_static, NOISE_HASH_LEN,
+	    NOISE_SYMMETRIC_KEY_LEN, 0, NOISE_PUBLIC_KEY_LEN,
+	    handshake->chaining_key);
+
+	/* {t} */
+	tai64n_now(timestamp);
+	message_encrypt(dst->encrypted_timestamp, timestamp,
+			NOISE_TIMESTAMP_LEN, key, handshake->hash);
+
+	dst->sender_index = wg_index_hashtable_insert(
+		handshake->entry.peer->device->index_hashtable,
+		&handshake->entry);
+
+	handshake->state = HANDSHAKE_CREATED_INITIATION;
+	ret = true;
+
+out:
+	up_write(&handshake->lock);
+	up_read(&handshake->static_identity->lock);
+	memzero_explicit(key, NOISE_SYMMETRIC_KEY_LEN);
+	return ret;
+}
+
+struct wg_peer *
+wg_noise_handshake_consume_initiation(struct message_handshake_initiation *src,
+				      struct wg_device *wg)
+{
+	struct wg_peer *peer = NULL, *ret_peer = NULL;
+	struct noise_handshake *handshake;
+	bool replay_attack, flood_attack;
+	u8 key[NOISE_SYMMETRIC_KEY_LEN];
+	u8 chaining_key[NOISE_HASH_LEN];
+	u8 hash[NOISE_HASH_LEN];
+	u8 s[NOISE_PUBLIC_KEY_LEN];
+	u8 e[NOISE_PUBLIC_KEY_LEN];
+	u8 t[NOISE_TIMESTAMP_LEN];
+	u64 initiation_consumption;
+
+	down_read(&wg->static_identity.lock);
+	if (unlikely(!wg->static_identity.has_identity))
+		goto out;
+
+	handshake_init(chaining_key, hash, wg->static_identity.static_public);
+
+	/* e */
+	message_ephemeral(e, src->unencrypted_ephemeral, chaining_key, hash);
+
+	/* es */
+	if (!mix_dh(chaining_key, key, wg->static_identity.static_private, e))
+		goto out;
+
+	/* s */
+	if (!message_decrypt(s, src->encrypted_static,
+			     sizeof(src->encrypted_static), key, hash))
+		goto out;
+
+	/* Lookup which peer we're actually talking to */
+	peer = wg_pubkey_hashtable_lookup(wg->peer_hashtable, s);
+	if (!peer)
+		goto out;
+	handshake = &peer->handshake;
+
+	/* ss */
+	kdf(chaining_key, key, NULL, handshake->precomputed_static_static,
+	    NOISE_HASH_LEN, NOISE_SYMMETRIC_KEY_LEN, 0, NOISE_PUBLIC_KEY_LEN,
+	    chaining_key);
+
+	/* {t} */
+	if (!message_decrypt(t, src->encrypted_timestamp,
+			     sizeof(src->encrypted_timestamp), key, hash))
+		goto out;
+
+	down_read(&handshake->lock);
+	replay_attack = memcmp(t, handshake->latest_timestamp,
+			       NOISE_TIMESTAMP_LEN) <= 0;
+	flood_attack = (s64)handshake->last_initiation_consumption +
+			       NSEC_PER_SEC / INITIATIONS_PER_SECOND >
+		       (s64)ktime_get_coarse_boottime_ns();
+	up_read(&handshake->lock);
+	if (replay_attack || flood_attack)
+		goto out;
+
+	/* Success! Copy everything to peer */
+	down_write(&handshake->lock);
+	memcpy(handshake->remote_ephemeral, e, NOISE_PUBLIC_KEY_LEN);
+	if (memcmp(t, handshake->latest_timestamp, NOISE_TIMESTAMP_LEN) > 0)
+		memcpy(handshake->latest_timestamp, t, NOISE_TIMESTAMP_LEN);
+	memcpy(handshake->hash, hash, NOISE_HASH_LEN);
+	memcpy(handshake->chaining_key, chaining_key, NOISE_HASH_LEN);
+	handshake->remote_index = src->sender_index;
+	if ((s64)(handshake->last_initiation_consumption -
+	    (initiation_consumption = ktime_get_coarse_boottime_ns())) < 0)
+		handshake->last_initiation_consumption = initiation_consumption;
+	handshake->state = HANDSHAKE_CONSUMED_INITIATION;
+	up_write(&handshake->lock);
+	ret_peer = peer;
+
+out:
+	memzero_explicit(key, NOISE_SYMMETRIC_KEY_LEN);
+	memzero_explicit(hash, NOISE_HASH_LEN);
+	memzero_explicit(chaining_key, NOISE_HASH_LEN);
+	up_read(&wg->static_identity.lock);
+	if (!ret_peer)
+		wg_peer_put(peer);
+	return ret_peer;
+}
+
+bool wg_noise_handshake_create_response(struct message_handshake_response *dst,
+					struct noise_handshake *handshake)
+{
+	u8 key[NOISE_SYMMETRIC_KEY_LEN];
+	bool ret = false;
+
+	/* We need to wait for crng _before_ taking any locks, since
+	 * curve25519_generate_secret uses get_random_bytes_wait.
+	 */
+	wait_for_random_bytes();
+
+	down_read(&handshake->static_identity->lock);
+	down_write(&handshake->lock);
+
+	if (handshake->state != HANDSHAKE_CONSUMED_INITIATION)
+		goto out;
+
+	dst->header.type = cpu_to_le32(MESSAGE_HANDSHAKE_RESPONSE);
+	dst->receiver_index = handshake->remote_index;
+
+	/* e */
+	curve25519_generate_secret(handshake->ephemeral_private);
+	if (!curve25519_generate_public(dst->unencrypted_ephemeral,
+					handshake->ephemeral_private))
+		goto out;
+	message_ephemeral(dst->unencrypted_ephemeral,
+			  dst->unencrypted_ephemeral, handshake->chaining_key,
+			  handshake->hash);
+
+	/* ee */
+	if (!mix_dh(handshake->chaining_key, NULL, handshake->ephemeral_private,
+		    handshake->remote_ephemeral))
+		goto out;
+
+	/* se */
+	if (!mix_dh(handshake->chaining_key, NULL, handshake->ephemeral_private,
+		    handshake->remote_static))
+		goto out;
+
+	/* psk */
+	mix_psk(handshake->chaining_key, handshake->hash, key,
+		handshake->preshared_key);
+
+	/* {} */
+	message_encrypt(dst->encrypted_nothing, NULL, 0, key, handshake->hash);
+
+	dst->sender_index = wg_index_hashtable_insert(
+		handshake->entry.peer->device->index_hashtable,
+		&handshake->entry);
+
+	handshake->state = HANDSHAKE_CREATED_RESPONSE;
+	ret = true;
+
+out:
+	up_write(&handshake->lock);
+	up_read(&handshake->static_identity->lock);
+	memzero_explicit(key, NOISE_SYMMETRIC_KEY_LEN);
+	return ret;
+}
+
+struct wg_peer *
+wg_noise_handshake_consume_response(struct message_handshake_response *src,
+				    struct wg_device *wg)
+{
+	enum noise_handshake_state state = HANDSHAKE_ZEROED;
+	struct wg_peer *peer = NULL, *ret_peer = NULL;
+	struct noise_handshake *handshake;
+	u8 key[NOISE_SYMMETRIC_KEY_LEN];
+	u8 hash[NOISE_HASH_LEN];
+	u8 chaining_key[NOISE_HASH_LEN];
+	u8 e[NOISE_PUBLIC_KEY_LEN];
+	u8 ephemeral_private[NOISE_PUBLIC_KEY_LEN];
+	u8 static_private[NOISE_PUBLIC_KEY_LEN];
+
+	down_read(&wg->static_identity.lock);
+
+	if (unlikely(!wg->static_identity.has_identity))
+		goto out;
+
+	handshake = (struct noise_handshake *)wg_index_hashtable_lookup(
+		wg->index_hashtable, INDEX_HASHTABLE_HANDSHAKE,
+		src->receiver_index, &peer);
+	if (unlikely(!handshake))
+		goto out;
+
+	down_read(&handshake->lock);
+	state = handshake->state;
+	memcpy(hash, handshake->hash, NOISE_HASH_LEN);
+	memcpy(chaining_key, handshake->chaining_key, NOISE_HASH_LEN);
+	memcpy(ephemeral_private, handshake->ephemeral_private,
+	       NOISE_PUBLIC_KEY_LEN);
+	up_read(&handshake->lock);
+
+	if (state != HANDSHAKE_CREATED_INITIATION)
+		goto fail;
+
+	/* e */
+	message_ephemeral(e, src->unencrypted_ephemeral, chaining_key, hash);
+
+	/* ee */
+	if (!mix_dh(chaining_key, NULL, ephemeral_private, e))
+		goto fail;
+
+	/* se */
+	if (!mix_dh(chaining_key, NULL, wg->static_identity.static_private, e))
+		goto fail;
+
+	/* psk */
+	mix_psk(chaining_key, hash, key, handshake->preshared_key);
+
+	/* {} */
+	if (!message_decrypt(NULL, src->encrypted_nothing,
+			     sizeof(src->encrypted_nothing), key, hash))
+		goto fail;
+
+	/* Success! Copy everything to peer */
+	down_write(&handshake->lock);
+	/* It's important to check that the state is still the same, while we
+	 * have an exclusive lock.
+	 */
+	if (handshake->state != state) {
+		up_write(&handshake->lock);
+		goto fail;
+	}
+	memcpy(handshake->remote_ephemeral, e, NOISE_PUBLIC_KEY_LEN);
+	memcpy(handshake->hash, hash, NOISE_HASH_LEN);
+	memcpy(handshake->chaining_key, chaining_key, NOISE_HASH_LEN);
+	handshake->remote_index = src->sender_index;
+	handshake->state = HANDSHAKE_CONSUMED_RESPONSE;
+	up_write(&handshake->lock);
+	ret_peer = peer;
+	goto out;
+
+fail:
+	wg_peer_put(peer);
+out:
+	memzero_explicit(key, NOISE_SYMMETRIC_KEY_LEN);
+	memzero_explicit(hash, NOISE_HASH_LEN);
+	memzero_explicit(chaining_key, NOISE_HASH_LEN);
+	memzero_explicit(ephemeral_private, NOISE_PUBLIC_KEY_LEN);
+	memzero_explicit(static_private, NOISE_PUBLIC_KEY_LEN);
+	up_read(&wg->static_identity.lock);
+	return ret_peer;
+}
+
+bool wg_noise_handshake_begin_session(struct noise_handshake *handshake,
+				      struct noise_keypairs *keypairs)
+{
+	struct noise_keypair *new_keypair;
+	bool ret = false;
+
+	down_write(&handshake->lock);
+	if (handshake->state != HANDSHAKE_CREATED_RESPONSE &&
+	    handshake->state != HANDSHAKE_CONSUMED_RESPONSE)
+		goto out;
+
+	new_keypair = keypair_create(handshake->entry.peer);
+	if (!new_keypair)
+		goto out;
+	new_keypair->i_am_the_initiator = handshake->state ==
+					  HANDSHAKE_CONSUMED_RESPONSE;
+	new_keypair->remote_index = handshake->remote_index;
+
+	if (new_keypair->i_am_the_initiator)
+		derive_keys(&new_keypair->sending, &new_keypair->receiving,
+			    handshake->chaining_key);
+	else
+		derive_keys(&new_keypair->receiving, &new_keypair->sending,
+			    handshake->chaining_key);
+
+	handshake_zero(handshake);
+	rcu_read_lock_bh();
+	if (likely(!READ_ONCE(container_of(handshake, struct wg_peer,
+					   handshake)->is_dead))) {
+		add_new_keypair(keypairs, new_keypair);
+		net_dbg_ratelimited("%s: Keypair %llu created for peer %llu\n",
+				    handshake->entry.peer->device->dev->name,
+				    new_keypair->internal_id,
+				    handshake->entry.peer->internal_id);
+		ret = wg_index_hashtable_replace(
+			handshake->entry.peer->device->index_hashtable,
+			&handshake->entry, &new_keypair->entry);
+	} else {
+		kzfree(new_keypair);
+	}
+	rcu_read_unlock_bh();
+
+out:
+	up_write(&handshake->lock);
+	return ret;
+}
diff --git a/drivers/net/wireguard/noise.h b/drivers/net/wireguard/noise.h
new file mode 100644
index 000000000000..138a07bb817c
--- /dev/null
+++ b/drivers/net/wireguard/noise.h
@@ -0,0 +1,137 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+#ifndef _WG_NOISE_H
+#define _WG_NOISE_H
+
+#include "messages.h"
+#include "peerlookup.h"
+
+#include <linux/types.h>
+#include <linux/spinlock.h>
+#include <linux/atomic.h>
+#include <linux/rwsem.h>
+#include <linux/mutex.h>
+#include <linux/kref.h>
+
+union noise_counter {
+	struct {
+		u64 counter;
+		unsigned long backtrack[COUNTER_BITS_TOTAL / BITS_PER_LONG];
+		spinlock_t lock;
+	} receive;
+	atomic64_t counter;
+};
+
+struct noise_symmetric_key {
+	u8 key[NOISE_SYMMETRIC_KEY_LEN];
+	union noise_counter counter;
+	u64 birthdate;
+	bool is_valid;
+};
+
+struct noise_keypair {
+	struct index_hashtable_entry entry;
+	struct noise_symmetric_key sending;
+	struct noise_symmetric_key receiving;
+	__le32 remote_index;
+	bool i_am_the_initiator;
+	struct kref refcount;
+	struct rcu_head rcu;
+	u64 internal_id;
+};
+
+struct noise_keypairs {
+	struct noise_keypair __rcu *current_keypair;
+	struct noise_keypair __rcu *previous_keypair;
+	struct noise_keypair __rcu *next_keypair;
+	spinlock_t keypair_update_lock;
+};
+
+struct noise_static_identity {
+	u8 static_public[NOISE_PUBLIC_KEY_LEN];
+	u8 static_private[NOISE_PUBLIC_KEY_LEN];
+	struct rw_semaphore lock;
+	bool has_identity;
+};
+
+enum noise_handshake_state {
+	HANDSHAKE_ZEROED,
+	HANDSHAKE_CREATED_INITIATION,
+	HANDSHAKE_CONSUMED_INITIATION,
+	HANDSHAKE_CREATED_RESPONSE,
+	HANDSHAKE_CONSUMED_RESPONSE
+};
+
+struct noise_handshake {
+	struct index_hashtable_entry entry;
+
+	enum noise_handshake_state state;
+	u64 last_initiation_consumption;
+
+	struct noise_static_identity *static_identity;
+
+	u8 ephemeral_private[NOISE_PUBLIC_KEY_LEN];
+	u8 remote_static[NOISE_PUBLIC_KEY_LEN];
+	u8 remote_ephemeral[NOISE_PUBLIC_KEY_LEN];
+	u8 precomputed_static_static[NOISE_PUBLIC_KEY_LEN];
+
+	u8 preshared_key[NOISE_SYMMETRIC_KEY_LEN];
+
+	u8 hash[NOISE_HASH_LEN];
+	u8 chaining_key[NOISE_HASH_LEN];
+
+	u8 latest_timestamp[NOISE_TIMESTAMP_LEN];
+	__le32 remote_index;
+
+	/* Protects all members except the immutable (after noise_handshake_
+	 * init): remote_static, precomputed_static_static, static_identity.
+	 */
+	struct rw_semaphore lock;
+};
+
+struct wg_device;
+
+void wg_noise_init(void);
+bool wg_noise_handshake_init(struct noise_handshake *handshake,
+			   struct noise_static_identity *static_identity,
+			   const u8 peer_public_key[NOISE_PUBLIC_KEY_LEN],
+			   const u8 peer_preshared_key[NOISE_SYMMETRIC_KEY_LEN],
+			   struct wg_peer *peer);
+void wg_noise_handshake_clear(struct noise_handshake *handshake);
+static inline void wg_noise_reset_last_sent_handshake(atomic64_t *handshake_ns)
+{
+	atomic64_set(handshake_ns, ktime_get_coarse_boottime_ns() -
+				       (u64)(REKEY_TIMEOUT + 1) * NSEC_PER_SEC);
+}
+
+void wg_noise_keypair_put(struct noise_keypair *keypair, bool unreference_now);
+struct noise_keypair *wg_noise_keypair_get(struct noise_keypair *keypair);
+void wg_noise_keypairs_clear(struct noise_keypairs *keypairs);
+bool wg_noise_received_with_keypair(struct noise_keypairs *keypairs,
+				    struct noise_keypair *received_keypair);
+void wg_noise_expire_current_peer_keypairs(struct wg_peer *peer);
+
+void wg_noise_set_static_identity_private_key(
+	struct noise_static_identity *static_identity,
+	const u8 private_key[NOISE_PUBLIC_KEY_LEN]);
+bool wg_noise_precompute_static_static(struct wg_peer *peer);
+
+bool
+wg_noise_handshake_create_initiation(struct message_handshake_initiation *dst,
+				     struct noise_handshake *handshake);
+struct wg_peer *
+wg_noise_handshake_consume_initiation(struct message_handshake_initiation *src,
+				      struct wg_device *wg);
+
+bool wg_noise_handshake_create_response(struct message_handshake_response *dst,
+					struct noise_handshake *handshake);
+struct wg_peer *
+wg_noise_handshake_consume_response(struct message_handshake_response *src,
+				    struct wg_device *wg);
+
+bool wg_noise_handshake_begin_session(struct noise_handshake *handshake,
+				      struct noise_keypairs *keypairs);
+
+#endif /* _WG_NOISE_H */
diff --git a/drivers/net/wireguard/peer.c b/drivers/net/wireguard/peer.c
new file mode 100644
index 000000000000..071eedf33f5a
--- /dev/null
+++ b/drivers/net/wireguard/peer.c
@@ -0,0 +1,240 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#include "peer.h"
+#include "device.h"
+#include "queueing.h"
+#include "timers.h"
+#include "peerlookup.h"
+#include "noise.h"
+
+#include <linux/kref.h>
+#include <linux/lockdep.h>
+#include <linux/rcupdate.h>
+#include <linux/list.h>
+
+static atomic64_t peer_counter = ATOMIC64_INIT(0);
+
+struct wg_peer *wg_peer_create(struct wg_device *wg,
+			       const u8 public_key[NOISE_PUBLIC_KEY_LEN],
+			       const u8 preshared_key[NOISE_SYMMETRIC_KEY_LEN])
+{
+	struct wg_peer *peer;
+	int ret = -ENOMEM;
+
+	lockdep_assert_held(&wg->device_update_lock);
+
+	if (wg->num_peers >= MAX_PEERS_PER_DEVICE)
+		return ERR_PTR(ret);
+
+	peer = kzalloc(sizeof(*peer), GFP_KERNEL);
+	if (unlikely(!peer))
+		return ERR_PTR(ret);
+	peer->device = wg;
+
+	if (!wg_noise_handshake_init(&peer->handshake, &wg->static_identity,
+				     public_key, preshared_key, peer)) {
+		ret = -EKEYREJECTED;
+		goto err_1;
+	}
+	if (dst_cache_init(&peer->endpoint_cache, GFP_KERNEL))
+		goto err_1;
+	if (wg_packet_queue_init(&peer->tx_queue, wg_packet_tx_worker, false,
+				 MAX_QUEUED_PACKETS))
+		goto err_2;
+	if (wg_packet_queue_init(&peer->rx_queue, NULL, false,
+				 MAX_QUEUED_PACKETS))
+		goto err_3;
+
+	peer->internal_id = atomic64_inc_return(&peer_counter);
+	peer->serial_work_cpu = nr_cpumask_bits;
+	wg_cookie_init(&peer->latest_cookie);
+	wg_timers_init(peer);
+	wg_cookie_checker_precompute_peer_keys(peer);
+	spin_lock_init(&peer->keypairs.keypair_update_lock);
+	INIT_WORK(&peer->transmit_handshake_work,
+		  wg_packet_handshake_send_worker);
+	rwlock_init(&peer->endpoint_lock);
+	kref_init(&peer->refcount);
+	skb_queue_head_init(&peer->staged_packet_queue);
+	wg_noise_reset_last_sent_handshake(&peer->last_sent_handshake);
+	set_bit(NAPI_STATE_NO_BUSY_POLL, &peer->napi.state);
+	netif_napi_add(wg->dev, &peer->napi, wg_packet_rx_poll,
+		       NAPI_POLL_WEIGHT);
+	napi_enable(&peer->napi);
+	list_add_tail(&peer->peer_list, &wg->peer_list);
+	INIT_LIST_HEAD(&peer->allowedips_list);
+	wg_pubkey_hashtable_add(wg->peer_hashtable, peer);
+	++wg->num_peers;
+	pr_debug("%s: Peer %llu created\n", wg->dev->name, peer->internal_id);
+	return peer;
+
+err_3:
+	wg_packet_queue_free(&peer->tx_queue, false);
+err_2:
+	dst_cache_destroy(&peer->endpoint_cache);
+err_1:
+	kfree(peer);
+	return ERR_PTR(ret);
+}
+
+struct wg_peer *wg_peer_get_maybe_zero(struct wg_peer *peer)
+{
+	RCU_LOCKDEP_WARN(!rcu_read_lock_bh_held(),
+			 "Taking peer reference without holding the RCU read lock");
+	if (unlikely(!peer || !kref_get_unless_zero(&peer->refcount)))
+		return NULL;
+	return peer;
+}
+
+static void peer_make_dead(struct wg_peer *peer)
+{
+	/* Remove from configuration-time lookup structures. */
+	list_del_init(&peer->peer_list);
+	wg_allowedips_remove_by_peer(&peer->device->peer_allowedips, peer,
+				     &peer->device->device_update_lock);
+	wg_pubkey_hashtable_remove(peer->device->peer_hashtable, peer);
+
+	/* Mark as dead, so that we don't allow jumping contexts after. */
+	WRITE_ONCE(peer->is_dead, true);
+
+	/* The caller must now synchronize_rcu() for this to take effect. */
+}
+
+static void peer_remove_after_dead(struct wg_peer *peer)
+{
+	WARN_ON(!peer->is_dead);
+
+	/* No more keypairs can be created for this peer, since is_dead protects
+	 * add_new_keypair, so we can now destroy existing ones.
+	 */
+	wg_noise_keypairs_clear(&peer->keypairs);
+
+	/* Destroy all ongoing timers that were in-flight at the beginning of
+	 * this function.
+	 */
+	wg_timers_stop(peer);
+
+	/* The transition between packet encryption/decryption queues isn't
+	 * guarded by is_dead, but each reference's life is strictly bounded by
+	 * two generations: once for parallel crypto and once for serial
+	 * ingestion, so we can simply flush twice, and be sure that we no
+	 * longer have references inside these queues.
+	 */
+
+	/* a) For encrypt/decrypt. */
+	flush_workqueue(peer->device->packet_crypt_wq);
+	/* b.1) For send (but not receive, since that's napi). */
+	flush_workqueue(peer->device->packet_crypt_wq);
+	/* b.2.1) For receive (but not send, since that's wq). */
+	napi_disable(&peer->napi);
+	/* b.2.1) It's now safe to remove the napi struct, which must be done
+	 * here from process context.
+	 */
+	netif_napi_del(&peer->napi);
+
+	/* Ensure any workstructs we own (like transmit_handshake_work or
+	 * clear_peer_work) no longer are in use.
+	 */
+	flush_workqueue(peer->device->handshake_send_wq);
+
+	/* After the above flushes, a peer might still be active in a few
+	 * different contexts: 1) from xmit(), before hitting is_dead and
+	 * returning, 2) from wg_packet_consume_data(), before hitting is_dead
+	 * and returning, 3) from wg_receive_handshake_packet() after a point
+	 * where it has processed an incoming handshake packet, but where
+	 * all calls to pass it off to timers fails because of is_dead. We won't
+	 * have new references in (1) eventually, because we're removed from
+	 * allowedips; we won't have new references in (2) eventually, because
+	 * wg_index_hashtable_lookup will always return NULL, since we removed
+	 * all existing keypairs and no more can be created; we won't have new
+	 * references in (3) eventually, because we're removed from the pubkey
+	 * hash table, which allows for a maximum of one handshake response,
+	 * via the still-uncleared index hashtable entry, but not more than one,
+	 * and in wg_cookie_message_consume, the lookup eventually gets a peer
+	 * with a refcount of zero, so no new reference is taken.
+	 */
+
+	--peer->device->num_peers;
+	wg_peer_put(peer);
+}
+
+/* We have a separate "remove" function make sure that all active places where
+ * a peer is currently operating will eventually come to an end and not pass
+ * their reference onto another context.
+ */
+void wg_peer_remove(struct wg_peer *peer)
+{
+	if (unlikely(!peer))
+		return;
+	lockdep_assert_held(&peer->device->device_update_lock);
+
+	peer_make_dead(peer);
+	synchronize_rcu();
+	peer_remove_after_dead(peer);
+}
+
+void wg_peer_remove_all(struct wg_device *wg)
+{
+	struct wg_peer *peer, *temp;
+	LIST_HEAD(dead_peers);
+
+	lockdep_assert_held(&wg->device_update_lock);
+
+	/* Avoid having to traverse individually for each one. */
+	wg_allowedips_free(&wg->peer_allowedips, &wg->device_update_lock);
+
+	list_for_each_entry_safe(peer, temp, &wg->peer_list, peer_list) {
+		peer_make_dead(peer);
+		list_add_tail(&peer->peer_list, &dead_peers);
+	}
+	synchronize_rcu();
+	list_for_each_entry_safe(peer, temp, &dead_peers, peer_list)
+		peer_remove_after_dead(peer);
+}
+
+static void rcu_release(struct rcu_head *rcu)
+{
+	struct wg_peer *peer = container_of(rcu, struct wg_peer, rcu);
+
+	dst_cache_destroy(&peer->endpoint_cache);
+	wg_packet_queue_free(&peer->rx_queue, false);
+	wg_packet_queue_free(&peer->tx_queue, false);
+
+	/* The final zeroing takes care of clearing any remaining handshake key
+	 * material and other potentially sensitive information.
+	 */
+	kzfree(peer);
+}
+
+static void kref_release(struct kref *refcount)
+{
+	struct wg_peer *peer = container_of(refcount, struct wg_peer, refcount);
+
+	pr_debug("%s: Peer %llu (%pISpfsc) destroyed\n",
+		 peer->device->dev->name, peer->internal_id,
+		 &peer->endpoint.addr);
+
+	/* Remove ourself from dynamic runtime lookup structures, now that the
+	 * last reference is gone.
+	 */
+	wg_index_hashtable_remove(peer->device->index_hashtable,
+				  &peer->handshake.entry);
+
+	/* Remove any lingering packets that didn't have a chance to be
+	 * transmitted.
+	 */
+	wg_packet_purge_staged_packets(peer);
+
+	/* Free the memory used. */
+	call_rcu(&peer->rcu, rcu_release);
+}
+
+void wg_peer_put(struct wg_peer *peer)
+{
+	if (unlikely(!peer))
+		return;
+	kref_put(&peer->refcount, kref_release);
+}
diff --git a/drivers/net/wireguard/peer.h b/drivers/net/wireguard/peer.h
new file mode 100644
index 000000000000..23af40922997
--- /dev/null
+++ b/drivers/net/wireguard/peer.h
@@ -0,0 +1,83 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#ifndef _WG_PEER_H
+#define _WG_PEER_H
+
+#include "device.h"
+#include "noise.h"
+#include "cookie.h"
+
+#include <linux/types.h>
+#include <linux/netfilter.h>
+#include <linux/spinlock.h>
+#include <linux/kref.h>
+#include <net/dst_cache.h>
+
+struct wg_device;
+
+struct endpoint {
+	union {
+		struct sockaddr addr;
+		struct sockaddr_in addr4;
+		struct sockaddr_in6 addr6;
+	};
+	union {
+		struct {
+			struct in_addr src4;
+			/* Essentially the same as addr6->scope_id */
+			int src_if4;
+		};
+		struct in6_addr src6;
+	};
+};
+
+struct wg_peer {
+	struct wg_device *device;
+	struct crypt_queue tx_queue, rx_queue;
+	struct sk_buff_head staged_packet_queue;
+	int serial_work_cpu;
+	struct noise_keypairs keypairs;
+	struct endpoint endpoint;
+	struct dst_cache endpoint_cache;
+	rwlock_t endpoint_lock;
+	struct noise_handshake handshake;
+	atomic64_t last_sent_handshake;
+	struct work_struct transmit_handshake_work, clear_peer_work;
+	struct cookie latest_cookie;
+	struct hlist_node pubkey_hash;
+	u64 rx_bytes, tx_bytes;
+	struct timer_list timer_retransmit_handshake, timer_send_keepalive;
+	struct timer_list timer_new_handshake, timer_zero_key_material;
+	struct timer_list timer_persistent_keepalive;
+	unsigned int timer_handshake_attempts;
+	u16 persistent_keepalive_interval;
+	bool timer_need_another_keepalive;
+	bool sent_lastminute_handshake;
+	struct timespec64 walltime_last_handshake;
+	struct kref refcount;
+	struct rcu_head rcu;
+	struct list_head peer_list;
+	struct list_head allowedips_list;
+	u64 internal_id;
+	struct napi_struct napi;
+	bool is_dead;
+};
+
+struct wg_peer *wg_peer_create(struct wg_device *wg,
+			       const u8 public_key[NOISE_PUBLIC_KEY_LEN],
+			       const u8 preshared_key[NOISE_SYMMETRIC_KEY_LEN]);
+
+struct wg_peer *__must_check wg_peer_get_maybe_zero(struct wg_peer *peer);
+static inline struct wg_peer *wg_peer_get(struct wg_peer *peer)
+{
+	kref_get(&peer->refcount);
+	return peer;
+}
+void wg_peer_put(struct wg_peer *peer);
+void wg_peer_remove(struct wg_peer *peer);
+void wg_peer_remove_all(struct wg_device *wg);
+
+#endif /* _WG_PEER_H */
diff --git a/drivers/net/wireguard/peerlookup.c b/drivers/net/wireguard/peerlookup.c
new file mode 100644
index 000000000000..e4deb331476b
--- /dev/null
+++ b/drivers/net/wireguard/peerlookup.c
@@ -0,0 +1,221 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#include "peerlookup.h"
+#include "peer.h"
+#include "noise.h"
+
+static struct hlist_head *pubkey_bucket(struct pubkey_hashtable *table,
+					const u8 pubkey[NOISE_PUBLIC_KEY_LEN])
+{
+	/* siphash gives us a secure 64bit number based on a random key. Since
+	 * the bits are uniformly distributed, we can then mask off to get the
+	 * bits we need.
+	 */
+	const u64 hash = siphash(pubkey, NOISE_PUBLIC_KEY_LEN, &table->key);
+
+	return &table->hashtable[hash & (HASH_SIZE(table->hashtable) - 1)];
+}
+
+struct pubkey_hashtable *wg_pubkey_hashtable_alloc(void)
+{
+	struct pubkey_hashtable *table = kvmalloc(sizeof(*table), GFP_KERNEL);
+
+	if (!table)
+		return NULL;
+
+	get_random_bytes(&table->key, sizeof(table->key));
+	hash_init(table->hashtable);
+	mutex_init(&table->lock);
+	return table;
+}
+
+void wg_pubkey_hashtable_add(struct pubkey_hashtable *table,
+			     struct wg_peer *peer)
+{
+	mutex_lock(&table->lock);
+	hlist_add_head_rcu(&peer->pubkey_hash,
+			   pubkey_bucket(table, peer->handshake.remote_static));
+	mutex_unlock(&table->lock);
+}
+
+void wg_pubkey_hashtable_remove(struct pubkey_hashtable *table,
+				struct wg_peer *peer)
+{
+	mutex_lock(&table->lock);
+	hlist_del_init_rcu(&peer->pubkey_hash);
+	mutex_unlock(&table->lock);
+}
+
+/* Returns a strong reference to a peer */
+struct wg_peer *
+wg_pubkey_hashtable_lookup(struct pubkey_hashtable *table,
+			   const u8 pubkey[NOISE_PUBLIC_KEY_LEN])
+{
+	struct wg_peer *iter_peer, *peer = NULL;
+
+	rcu_read_lock_bh();
+	hlist_for_each_entry_rcu_bh(iter_peer, pubkey_bucket(table, pubkey),
+				    pubkey_hash) {
+		if (!memcmp(pubkey, iter_peer->handshake.remote_static,
+			    NOISE_PUBLIC_KEY_LEN)) {
+			peer = iter_peer;
+			break;
+		}
+	}
+	peer = wg_peer_get_maybe_zero(peer);
+	rcu_read_unlock_bh();
+	return peer;
+}
+
+static struct hlist_head *index_bucket(struct index_hashtable *table,
+				       const __le32 index)
+{
+	/* Since the indices are random and thus all bits are uniformly
+	 * distributed, we can find its bucket simply by masking.
+	 */
+	return &table->hashtable[(__force u32)index &
+				 (HASH_SIZE(table->hashtable) - 1)];
+}
+
+struct index_hashtable *wg_index_hashtable_alloc(void)
+{
+	struct index_hashtable *table = kvmalloc(sizeof(*table), GFP_KERNEL);
+
+	if (!table)
+		return NULL;
+
+	hash_init(table->hashtable);
+	spin_lock_init(&table->lock);
+	return table;
+}
+
+/* At the moment, we limit ourselves to 2^20 total peers, which generally might
+ * amount to 2^20*3 items in this hashtable. The algorithm below works by
+ * picking a random number and testing it. We can see that these limits mean we
+ * usually succeed pretty quickly:
+ *
+ * >>> def calculation(tries, size):
+ * ...     return (size / 2**32)**(tries - 1) *  (1 - (size / 2**32))
+ * ...
+ * >>> calculation(1, 2**20 * 3)
+ * 0.999267578125
+ * >>> calculation(2, 2**20 * 3)
+ * 0.0007318854331970215
+ * >>> calculation(3, 2**20 * 3)
+ * 5.360489012673497e-07
+ * >>> calculation(4, 2**20 * 3)
+ * 3.9261394135792216e-10
+ *
+ * At the moment, we don't do any masking, so this algorithm isn't exactly
+ * constant time in either the random guessing or in the hash list lookup. We
+ * could require a minimum of 3 tries, which would successfully mask the
+ * guessing. this would not, however, help with the growing hash lengths, which
+ * is another thing to consider moving forward.
+ */
+
+__le32 wg_index_hashtable_insert(struct index_hashtable *table,
+				 struct index_hashtable_entry *entry)
+{
+	struct index_hashtable_entry *existing_entry;
+
+	spin_lock_bh(&table->lock);
+	hlist_del_init_rcu(&entry->index_hash);
+	spin_unlock_bh(&table->lock);
+
+	rcu_read_lock_bh();
+
+search_unused_slot:
+	/* First we try to find an unused slot, randomly, while unlocked. */
+	entry->index = (__force __le32)get_random_u32();
+	hlist_for_each_entry_rcu_bh(existing_entry,
+				    index_bucket(table, entry->index),
+				    index_hash) {
+		if (existing_entry->index == entry->index)
+			/* If it's already in use, we continue searching. */
+			goto search_unused_slot;
+	}
+
+	/* Once we've found an unused slot, we lock it, and then double-check
+	 * that nobody else stole it from us.
+	 */
+	spin_lock_bh(&table->lock);
+	hlist_for_each_entry_rcu_bh(existing_entry,
+				    index_bucket(table, entry->index),
+				    index_hash) {
+		if (existing_entry->index == entry->index) {
+			spin_unlock_bh(&table->lock);
+			/* If it was stolen, we start over. */
+			goto search_unused_slot;
+		}
+	}
+	/* Otherwise, we know we have it exclusively (since we're locked),
+	 * so we insert.
+	 */
+	hlist_add_head_rcu(&entry->index_hash,
+			   index_bucket(table, entry->index));
+	spin_unlock_bh(&table->lock);
+
+	rcu_read_unlock_bh();
+
+	return entry->index;
+}
+
+bool wg_index_hashtable_replace(struct index_hashtable *table,
+				struct index_hashtable_entry *old,
+				struct index_hashtable_entry *new)
+{
+	if (unlikely(hlist_unhashed(&old->index_hash)))
+		return false;
+	spin_lock_bh(&table->lock);
+	new->index = old->index;
+	hlist_replace_rcu(&old->index_hash, &new->index_hash);
+
+	/* Calling init here NULLs out index_hash, and in fact after this
+	 * function returns, it's theoretically possible for this to get
+	 * reinserted elsewhere. That means the RCU lookup below might either
+	 * terminate early or jump between buckets, in which case the packet
+	 * simply gets dropped, which isn't terrible.
+	 */
+	INIT_HLIST_NODE(&old->index_hash);
+	spin_unlock_bh(&table->lock);
+	return true;
+}
+
+void wg_index_hashtable_remove(struct index_hashtable *table,
+			       struct index_hashtable_entry *entry)
+{
+	spin_lock_bh(&table->lock);
+	hlist_del_init_rcu(&entry->index_hash);
+	spin_unlock_bh(&table->lock);
+}
+
+/* Returns a strong reference to a entry->peer */
+struct index_hashtable_entry *
+wg_index_hashtable_lookup(struct index_hashtable *table,
+			  const enum index_hashtable_type type_mask,
+			  const __le32 index, struct wg_peer **peer)
+{
+	struct index_hashtable_entry *iter_entry, *entry = NULL;
+
+	rcu_read_lock_bh();
+	hlist_for_each_entry_rcu_bh(iter_entry, index_bucket(table, index),
+				    index_hash) {
+		if (iter_entry->index == index) {
+			if (likely(iter_entry->type & type_mask))
+				entry = iter_entry;
+			break;
+		}
+	}
+	if (likely(entry)) {
+		entry->peer = wg_peer_get_maybe_zero(entry->peer);
+		if (likely(entry->peer))
+			*peer = entry->peer;
+		else
+			entry = NULL;
+	}
+	rcu_read_unlock_bh();
+	return entry;
+}
diff --git a/drivers/net/wireguard/peerlookup.h b/drivers/net/wireguard/peerlookup.h
new file mode 100644
index 000000000000..ced811797680
--- /dev/null
+++ b/drivers/net/wireguard/peerlookup.h
@@ -0,0 +1,64 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#ifndef _WG_PEERLOOKUP_H
+#define _WG_PEERLOOKUP_H
+
+#include "messages.h"
+
+#include <linux/hashtable.h>
+#include <linux/mutex.h>
+#include <linux/siphash.h>
+
+struct wg_peer;
+
+struct pubkey_hashtable {
+	/* TODO: move to rhashtable */
+	DECLARE_HASHTABLE(hashtable, 11);
+	siphash_key_t key;
+	struct mutex lock;
+};
+
+struct pubkey_hashtable *wg_pubkey_hashtable_alloc(void);
+void wg_pubkey_hashtable_add(struct pubkey_hashtable *table,
+			     struct wg_peer *peer);
+void wg_pubkey_hashtable_remove(struct pubkey_hashtable *table,
+				struct wg_peer *peer);
+struct wg_peer *
+wg_pubkey_hashtable_lookup(struct pubkey_hashtable *table,
+			   const u8 pubkey[NOISE_PUBLIC_KEY_LEN]);
+
+struct index_hashtable {
+	/* TODO: move to rhashtable */
+	DECLARE_HASHTABLE(hashtable, 13);
+	spinlock_t lock;
+};
+
+enum index_hashtable_type {
+	INDEX_HASHTABLE_HANDSHAKE = 1U << 0,
+	INDEX_HASHTABLE_KEYPAIR = 1U << 1
+};
+
+struct index_hashtable_entry {
+	struct wg_peer *peer;
+	struct hlist_node index_hash;
+	enum index_hashtable_type type;
+	__le32 index;
+};
+
+struct index_hashtable *wg_index_hashtable_alloc(void);
+__le32 wg_index_hashtable_insert(struct index_hashtable *table,
+				 struct index_hashtable_entry *entry);
+bool wg_index_hashtable_replace(struct index_hashtable *table,
+				struct index_hashtable_entry *old,
+				struct index_hashtable_entry *new);
+void wg_index_hashtable_remove(struct index_hashtable *table,
+			       struct index_hashtable_entry *entry);
+struct index_hashtable_entry *
+wg_index_hashtable_lookup(struct index_hashtable *table,
+			  const enum index_hashtable_type type_mask,
+			  const __le32 index, struct wg_peer **peer);
+
+#endif /* _WG_PEERLOOKUP_H */
diff --git a/drivers/net/wireguard/queueing.c b/drivers/net/wireguard/queueing.c
new file mode 100644
index 000000000000..5c964fcb994e
--- /dev/null
+++ b/drivers/net/wireguard/queueing.c
@@ -0,0 +1,53 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#include "queueing.h"
+
+struct multicore_worker __percpu *
+wg_packet_percpu_multicore_worker_alloc(work_func_t function, void *ptr)
+{
+	int cpu;
+	struct multicore_worker __percpu *worker =
+		alloc_percpu(struct multicore_worker);
+
+	if (!worker)
+		return NULL;
+
+	for_each_possible_cpu(cpu) {
+		per_cpu_ptr(worker, cpu)->ptr = ptr;
+		INIT_WORK(&per_cpu_ptr(worker, cpu)->work, function);
+	}
+	return worker;
+}
+
+int wg_packet_queue_init(struct crypt_queue *queue, work_func_t function,
+			 bool multicore, unsigned int len)
+{
+	int ret;
+
+	memset(queue, 0, sizeof(*queue));
+	ret = ptr_ring_init(&queue->ring, len, GFP_KERNEL);
+	if (ret)
+		return ret;
+	if (function) {
+		if (multicore) {
+			queue->worker = wg_packet_percpu_multicore_worker_alloc(
+				function, queue);
+			if (!queue->worker)
+				return -ENOMEM;
+		} else {
+			INIT_WORK(&queue->work, function);
+		}
+	}
+	return 0;
+}
+
+void wg_packet_queue_free(struct crypt_queue *queue, bool multicore)
+{
+	if (multicore)
+		free_percpu(queue->worker);
+	WARN_ON(!__ptr_ring_empty(&queue->ring));
+	ptr_ring_cleanup(&queue->ring, NULL);
+}
diff --git a/drivers/net/wireguard/queueing.h b/drivers/net/wireguard/queueing.h
new file mode 100644
index 000000000000..e49a464238fd
--- /dev/null
+++ b/drivers/net/wireguard/queueing.h
@@ -0,0 +1,197 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#ifndef _WG_QUEUEING_H
+#define _WG_QUEUEING_H
+
+#include "peer.h"
+#include <linux/types.h>
+#include <linux/skbuff.h>
+#include <linux/ip.h>
+#include <linux/ipv6.h>
+
+struct wg_device;
+struct wg_peer;
+struct multicore_worker;
+struct crypt_queue;
+struct sk_buff;
+
+/* queueing.c APIs: */
+int wg_packet_queue_init(struct crypt_queue *queue, work_func_t function,
+			 bool multicore, unsigned int len);
+void wg_packet_queue_free(struct crypt_queue *queue, bool multicore);
+struct multicore_worker __percpu *
+wg_packet_percpu_multicore_worker_alloc(work_func_t function, void *ptr);
+
+/* receive.c APIs: */
+void wg_packet_receive(struct wg_device *wg, struct sk_buff *skb);
+void wg_packet_handshake_receive_worker(struct work_struct *work);
+/* NAPI poll function: */
+int wg_packet_rx_poll(struct napi_struct *napi, int budget);
+/* Workqueue worker: */
+void wg_packet_decrypt_worker(struct work_struct *work);
+
+/* send.c APIs: */
+void wg_packet_send_queued_handshake_initiation(struct wg_peer *peer,
+						bool is_retry);
+void wg_packet_send_handshake_response(struct wg_peer *peer);
+void wg_packet_send_handshake_cookie(struct wg_device *wg,
+				     struct sk_buff *initiating_skb,
+				     __le32 sender_index);
+void wg_packet_send_keepalive(struct wg_peer *peer);
+void wg_packet_purge_staged_packets(struct wg_peer *peer);
+void wg_packet_send_staged_packets(struct wg_peer *peer);
+/* Workqueue workers: */
+void wg_packet_handshake_send_worker(struct work_struct *work);
+void wg_packet_tx_worker(struct work_struct *work);
+void wg_packet_encrypt_worker(struct work_struct *work);
+
+enum packet_state {
+	PACKET_STATE_UNCRYPTED,
+	PACKET_STATE_CRYPTED,
+	PACKET_STATE_DEAD
+};
+
+struct packet_cb {
+	u64 nonce;
+	struct noise_keypair *keypair;
+	atomic_t state;
+	u32 mtu;
+	u8 ds;
+};
+
+#define PACKET_CB(skb) ((struct packet_cb *)((skb)->cb))
+#define PACKET_PEER(skb) (PACKET_CB(skb)->keypair->entry.peer)
+
+/* Returns either the correct skb->protocol value, or 0 if invalid. */
+static inline __be16 wg_skb_examine_untrusted_ip_hdr(struct sk_buff *skb)
+{
+	if (skb_network_header(skb) >= skb->head &&
+	    (skb_network_header(skb) + sizeof(struct iphdr)) <=
+		    skb_tail_pointer(skb) &&
+	    ip_hdr(skb)->version == 4)
+		return htons(ETH_P_IP);
+	if (skb_network_header(skb) >= skb->head &&
+	    (skb_network_header(skb) + sizeof(struct ipv6hdr)) <=
+		    skb_tail_pointer(skb) &&
+	    ipv6_hdr(skb)->version == 6)
+		return htons(ETH_P_IPV6);
+	return 0;
+}
+
+static inline void wg_reset_packet(struct sk_buff *skb)
+{
+	const int pfmemalloc = skb->pfmemalloc;
+
+	skb_scrub_packet(skb, true);
+	memset(&skb->headers_start, 0,
+	       offsetof(struct sk_buff, headers_end) -
+		       offsetof(struct sk_buff, headers_start));
+	skb->pfmemalloc = pfmemalloc;
+	skb->queue_mapping = 0;
+	skb->nohdr = 0;
+	skb->peeked = 0;
+	skb->mac_len = 0;
+	skb->dev = NULL;
+#ifdef CONFIG_NET_SCHED
+	skb->tc_index = 0;
+	skb_reset_tc(skb);
+#endif
+	skb->hdr_len = skb_headroom(skb);
+	skb_reset_mac_header(skb);
+	skb_reset_network_header(skb);
+	skb_reset_transport_header(skb);
+	skb_probe_transport_header(skb);
+	skb_reset_inner_headers(skb);
+}
+
+static inline int wg_cpumask_choose_online(int *stored_cpu, unsigned int id)
+{
+	unsigned int cpu = *stored_cpu, cpu_index, i;
+
+	if (unlikely(cpu == nr_cpumask_bits ||
+		     !cpumask_test_cpu(cpu, cpu_online_mask))) {
+		cpu_index = id % cpumask_weight(cpu_online_mask);
+		cpu = cpumask_first(cpu_online_mask);
+		for (i = 0; i < cpu_index; ++i)
+			cpu = cpumask_next(cpu, cpu_online_mask);
+		*stored_cpu = cpu;
+	}
+	return cpu;
+}
+
+/* This function is racy, in the sense that next is unlocked, so it could return
+ * the same CPU twice. A race-free version of this would be to instead store an
+ * atomic sequence number, do an increment-and-return, and then iterate through
+ * every possible CPU until we get to that index -- choose_cpu. However that's
+ * a bit slower, and it doesn't seem like this potential race actually
+ * introduces any performance loss, so we live with it.
+ */
+static inline int wg_cpumask_next_online(int *next)
+{
+	int cpu = *next;
+
+	while (unlikely(!cpumask_test_cpu(cpu, cpu_online_mask)))
+		cpu = cpumask_next(cpu, cpu_online_mask) % nr_cpumask_bits;
+	*next = cpumask_next(cpu, cpu_online_mask) % nr_cpumask_bits;
+	return cpu;
+}
+
+static inline int wg_queue_enqueue_per_device_and_peer(
+	struct crypt_queue *device_queue, struct crypt_queue *peer_queue,
+	struct sk_buff *skb, struct workqueue_struct *wq, int *next_cpu)
+{
+	int cpu;
+
+	atomic_set_release(&PACKET_CB(skb)->state, PACKET_STATE_UNCRYPTED);
+	/* We first queue this up for the peer ingestion, but the consumer
+	 * will wait for the state to change to CRYPTED or DEAD before.
+	 */
+	if (unlikely(ptr_ring_produce_bh(&peer_queue->ring, skb)))
+		return -ENOSPC;
+	/* Then we queue it up in the device queue, which consumes the
+	 * packet as soon as it can.
+	 */
+	cpu = wg_cpumask_next_online(next_cpu);
+	if (unlikely(ptr_ring_produce_bh(&device_queue->ring, skb)))
+		return -EPIPE;
+	queue_work_on(cpu, wq, &per_cpu_ptr(device_queue->worker, cpu)->work);
+	return 0;
+}
+
+static inline void wg_queue_enqueue_per_peer(struct crypt_queue *queue,
+					     struct sk_buff *skb,
+					     enum packet_state state)
+{
+	/* We take a reference, because as soon as we call atomic_set, the
+	 * peer can be freed from below us.
+	 */
+	struct wg_peer *peer = wg_peer_get(PACKET_PEER(skb));
+
+	atomic_set_release(&PACKET_CB(skb)->state, state);
+	queue_work_on(wg_cpumask_choose_online(&peer->serial_work_cpu,
+					       peer->internal_id),
+		      peer->device->packet_crypt_wq, &queue->work);
+	wg_peer_put(peer);
+}
+
+static inline void wg_queue_enqueue_per_peer_napi(struct sk_buff *skb,
+						  enum packet_state state)
+{
+	/* We take a reference, because as soon as we call atomic_set, the
+	 * peer can be freed from below us.
+	 */
+	struct wg_peer *peer = wg_peer_get(PACKET_PEER(skb));
+
+	atomic_set_release(&PACKET_CB(skb)->state, state);
+	napi_schedule(&peer->napi);
+	wg_peer_put(peer);
+}
+
+#ifdef DEBUG
+bool wg_packet_counter_selftest(void);
+#endif
+
+#endif /* _WG_QUEUEING_H */
diff --git a/drivers/net/wireguard/ratelimiter.c b/drivers/net/wireguard/ratelimiter.c
new file mode 100644
index 000000000000..3fedd1d21f5e
--- /dev/null
+++ b/drivers/net/wireguard/ratelimiter.c
@@ -0,0 +1,223 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#include "ratelimiter.h"
+#include <linux/siphash.h>
+#include <linux/mm.h>
+#include <linux/slab.h>
+#include <net/ip.h>
+
+static struct kmem_cache *entry_cache;
+static hsiphash_key_t key;
+static spinlock_t table_lock = __SPIN_LOCK_UNLOCKED("ratelimiter_table_lock");
+static DEFINE_MUTEX(init_lock);
+static u64 init_refcnt; /* Protected by init_lock, hence not atomic. */
+static atomic_t total_entries = ATOMIC_INIT(0);
+static unsigned int max_entries, table_size;
+static void wg_ratelimiter_gc_entries(struct work_struct *);
+static DECLARE_DEFERRABLE_WORK(gc_work, wg_ratelimiter_gc_entries);
+static struct hlist_head *table_v4;
+#if IS_ENABLED(CONFIG_IPV6)
+static struct hlist_head *table_v6;
+#endif
+
+struct ratelimiter_entry {
+	u64 last_time_ns, tokens, ip;
+	void *net;
+	spinlock_t lock;
+	struct hlist_node hash;
+	struct rcu_head rcu;
+};
+
+enum {
+	PACKETS_PER_SECOND = 20,
+	PACKETS_BURSTABLE = 5,
+	PACKET_COST = NSEC_PER_SEC / PACKETS_PER_SECOND,
+	TOKEN_MAX = PACKET_COST * PACKETS_BURSTABLE
+};
+
+static void entry_free(struct rcu_head *rcu)
+{
+	kmem_cache_free(entry_cache,
+			container_of(rcu, struct ratelimiter_entry, rcu));
+	atomic_dec(&total_entries);
+}
+
+static void entry_uninit(struct ratelimiter_entry *entry)
+{
+	hlist_del_rcu(&entry->hash);
+	call_rcu(&entry->rcu, entry_free);
+}
+
+/* Calling this function with a NULL work uninits all entries. */
+static void wg_ratelimiter_gc_entries(struct work_struct *work)
+{
+	const u64 now = ktime_get_coarse_boottime_ns();
+	struct ratelimiter_entry *entry;
+	struct hlist_node *temp;
+	unsigned int i;
+
+	for (i = 0; i < table_size; ++i) {
+		spin_lock(&table_lock);
+		hlist_for_each_entry_safe(entry, temp, &table_v4[i], hash) {
+			if (unlikely(!work) ||
+			    now - entry->last_time_ns > NSEC_PER_SEC)
+				entry_uninit(entry);
+		}
+#if IS_ENABLED(CONFIG_IPV6)
+		hlist_for_each_entry_safe(entry, temp, &table_v6[i], hash) {
+			if (unlikely(!work) ||
+			    now - entry->last_time_ns > NSEC_PER_SEC)
+				entry_uninit(entry);
+		}
+#endif
+		spin_unlock(&table_lock);
+		if (likely(work))
+			cond_resched();
+	}
+	if (likely(work))
+		queue_delayed_work(system_power_efficient_wq, &gc_work, HZ);
+}
+
+bool wg_ratelimiter_allow(struct sk_buff *skb, struct net *net)
+{
+	/* We only take the bottom half of the net pointer, so that we can hash
+	 * 3 words in the end. This way, siphash's len param fits into the final
+	 * u32, and we don't incur an extra round.
+	 */
+	const u32 net_word = (unsigned long)net;
+	struct ratelimiter_entry *entry;
+	struct hlist_head *bucket;
+	u64 ip;
+
+	if (skb->protocol == htons(ETH_P_IP)) {
+		ip = (u64 __force)ip_hdr(skb)->saddr;
+		bucket = &table_v4[hsiphash_2u32(net_word, ip, &key) &
+				   (table_size - 1)];
+	}
+#if IS_ENABLED(CONFIG_IPV6)
+	else if (skb->protocol == htons(ETH_P_IPV6)) {
+		/* Only use 64 bits, so as to ratelimit the whole /64. */
+		memcpy(&ip, &ipv6_hdr(skb)->saddr, sizeof(ip));
+		bucket = &table_v6[hsiphash_3u32(net_word, ip >> 32, ip, &key) &
+				   (table_size - 1)];
+	}
+#endif
+	else
+		return false;
+	rcu_read_lock();
+	hlist_for_each_entry_rcu(entry, bucket, hash) {
+		if (entry->net == net && entry->ip == ip) {
+			u64 now, tokens;
+			bool ret;
+			/* Quasi-inspired by nft_limit.c, but this is actually a
+			 * slightly different algorithm. Namely, we incorporate
+			 * the burst as part of the maximum tokens, rather than
+			 * as part of the rate.
+			 */
+			spin_lock(&entry->lock);
+			now = ktime_get_coarse_boottime_ns();
+			tokens = min_t(u64, TOKEN_MAX,
+				       entry->tokens + now -
+					       entry->last_time_ns);
+			entry->last_time_ns = now;
+			ret = tokens >= PACKET_COST;
+			entry->tokens = ret ? tokens - PACKET_COST : tokens;
+			spin_unlock(&entry->lock);
+			rcu_read_unlock();
+			return ret;
+		}
+	}
+	rcu_read_unlock();
+
+	if (atomic_inc_return(&total_entries) > max_entries)
+		goto err_oom;
+
+	entry = kmem_cache_alloc(entry_cache, GFP_KERNEL);
+	if (unlikely(!entry))
+		goto err_oom;
+
+	entry->net = net;
+	entry->ip = ip;
+	INIT_HLIST_NODE(&entry->hash);
+	spin_lock_init(&entry->lock);
+	entry->last_time_ns = ktime_get_coarse_boottime_ns();
+	entry->tokens = TOKEN_MAX - PACKET_COST;
+	spin_lock(&table_lock);
+	hlist_add_head_rcu(&entry->hash, bucket);
+	spin_unlock(&table_lock);
+	return true;
+
+err_oom:
+	atomic_dec(&total_entries);
+	return false;
+}
+
+int wg_ratelimiter_init(void)
+{
+	mutex_lock(&init_lock);
+	if (++init_refcnt != 1)
+		goto out;
+
+	entry_cache = KMEM_CACHE(ratelimiter_entry, 0);
+	if (!entry_cache)
+		goto err;
+
+	/* xt_hashlimit.c uses a slightly different algorithm for ratelimiting,
+	 * but what it shares in common is that it uses a massive hashtable. So,
+	 * we borrow their wisdom about good table sizes on different systems
+	 * dependent on RAM. This calculation here comes from there.
+	 */
+	table_size = (totalram_pages() > (1U << 30) / PAGE_SIZE) ? 8192 :
+		max_t(unsigned long, 16, roundup_pow_of_two(
+			(totalram_pages() << PAGE_SHIFT) /
+			(1U << 14) / sizeof(struct hlist_head)));
+	max_entries = table_size * 8;
+
+	table_v4 = kvzalloc(table_size * sizeof(*table_v4), GFP_KERNEL);
+	if (unlikely(!table_v4))
+		goto err_kmemcache;
+
+#if IS_ENABLED(CONFIG_IPV6)
+	table_v6 = kvzalloc(table_size * sizeof(*table_v6), GFP_KERNEL);
+	if (unlikely(!table_v6)) {
+		kvfree(table_v4);
+		goto err_kmemcache;
+	}
+#endif
+
+	queue_delayed_work(system_power_efficient_wq, &gc_work, HZ);
+	get_random_bytes(&key, sizeof(key));
+out:
+	mutex_unlock(&init_lock);
+	return 0;
+
+err_kmemcache:
+	kmem_cache_destroy(entry_cache);
+err:
+	--init_refcnt;
+	mutex_unlock(&init_lock);
+	return -ENOMEM;
+}
+
+void wg_ratelimiter_uninit(void)
+{
+	mutex_lock(&init_lock);
+	if (!init_refcnt || --init_refcnt)
+		goto out;
+
+	cancel_delayed_work_sync(&gc_work);
+	wg_ratelimiter_gc_entries(NULL);
+	rcu_barrier();
+	kvfree(table_v4);
+#if IS_ENABLED(CONFIG_IPV6)
+	kvfree(table_v6);
+#endif
+	kmem_cache_destroy(entry_cache);
+out:
+	mutex_unlock(&init_lock);
+}
+
+#include "selftest/ratelimiter.c"
diff --git a/drivers/net/wireguard/ratelimiter.h b/drivers/net/wireguard/ratelimiter.h
new file mode 100644
index 000000000000..83067f71ea99
--- /dev/null
+++ b/drivers/net/wireguard/ratelimiter.h
@@ -0,0 +1,19 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#ifndef _WG_RATELIMITER_H
+#define _WG_RATELIMITER_H
+
+#include <linux/skbuff.h>
+
+int wg_ratelimiter_init(void);
+void wg_ratelimiter_uninit(void);
+bool wg_ratelimiter_allow(struct sk_buff *skb, struct net *net);
+
+#ifdef DEBUG
+bool wg_ratelimiter_selftest(void);
+#endif
+
+#endif /* _WG_RATELIMITER_H */
diff --git a/drivers/net/wireguard/receive.c b/drivers/net/wireguard/receive.c
new file mode 100644
index 000000000000..7e675f541491
--- /dev/null
+++ b/drivers/net/wireguard/receive.c
@@ -0,0 +1,595 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#include "queueing.h"
+#include "device.h"
+#include "peer.h"
+#include "timers.h"
+#include "messages.h"
+#include "cookie.h"
+#include "socket.h"
+
+#include <linux/ip.h>
+#include <linux/ipv6.h>
+#include <linux/udp.h>
+#include <net/ip_tunnels.h>
+
+/* Must be called with bh disabled. */
+static void update_rx_stats(struct wg_peer *peer, size_t len)
+{
+	struct pcpu_sw_netstats *tstats =
+		get_cpu_ptr(peer->device->dev->tstats);
+
+	u64_stats_update_begin(&tstats->syncp);
+	++tstats->rx_packets;
+	tstats->rx_bytes += len;
+	peer->rx_bytes += len;
+	u64_stats_update_end(&tstats->syncp);
+	put_cpu_ptr(tstats);
+}
+
+#define SKB_TYPE_LE32(skb) (((struct message_header *)(skb)->data)->type)
+
+static size_t validate_header_len(struct sk_buff *skb)
+{
+	if (unlikely(skb->len < sizeof(struct message_header)))
+		return 0;
+	if (SKB_TYPE_LE32(skb) == cpu_to_le32(MESSAGE_DATA) &&
+	    skb->len >= MESSAGE_MINIMUM_LENGTH)
+		return sizeof(struct message_data);
+	if (SKB_TYPE_LE32(skb) == cpu_to_le32(MESSAGE_HANDSHAKE_INITIATION) &&
+	    skb->len == sizeof(struct message_handshake_initiation))
+		return sizeof(struct message_handshake_initiation);
+	if (SKB_TYPE_LE32(skb) == cpu_to_le32(MESSAGE_HANDSHAKE_RESPONSE) &&
+	    skb->len == sizeof(struct message_handshake_response))
+		return sizeof(struct message_handshake_response);
+	if (SKB_TYPE_LE32(skb) == cpu_to_le32(MESSAGE_HANDSHAKE_COOKIE) &&
+	    skb->len == sizeof(struct message_handshake_cookie))
+		return sizeof(struct message_handshake_cookie);
+	return 0;
+}
+
+static int prepare_skb_header(struct sk_buff *skb, struct wg_device *wg)
+{
+	size_t data_offset, data_len, header_len;
+	struct udphdr *udp;
+
+	if (unlikely(wg_skb_examine_untrusted_ip_hdr(skb) != skb->protocol ||
+		     skb_transport_header(skb) < skb->head ||
+		     (skb_transport_header(skb) + sizeof(struct udphdr)) >
+			     skb_tail_pointer(skb)))
+		return -EINVAL; /* Bogus IP header */
+	udp = udp_hdr(skb);
+	data_offset = (u8 *)udp - skb->data;
+	if (unlikely(data_offset > U16_MAX ||
+		     data_offset + sizeof(struct udphdr) > skb->len))
+		/* Packet has offset at impossible location or isn't big enough
+		 * to have UDP fields.
+		 */
+		return -EINVAL;
+	data_len = ntohs(udp->len);
+	if (unlikely(data_len < sizeof(struct udphdr) ||
+		     data_len > skb->len - data_offset))
+		/* UDP packet is reporting too small of a size or lying about
+		 * its size.
+		 */
+		return -EINVAL;
+	data_len -= sizeof(struct udphdr);
+	data_offset = (u8 *)udp + sizeof(struct udphdr) - skb->data;
+	if (unlikely(!pskb_may_pull(skb,
+				data_offset + sizeof(struct message_header)) ||
+		     pskb_trim(skb, data_len + data_offset) < 0))
+		return -EINVAL;
+	skb_pull(skb, data_offset);
+	if (unlikely(skb->len != data_len))
+		/* Final len does not agree with calculated len */
+		return -EINVAL;
+	header_len = validate_header_len(skb);
+	if (unlikely(!header_len))
+		return -EINVAL;
+	__skb_push(skb, data_offset);
+	if (unlikely(!pskb_may_pull(skb, data_offset + header_len)))
+		return -EINVAL;
+	__skb_pull(skb, data_offset);
+	return 0;
+}
+
+static void wg_receive_handshake_packet(struct wg_device *wg,
+					struct sk_buff *skb)
+{
+	enum cookie_mac_state mac_state;
+	struct wg_peer *peer = NULL;
+	/* This is global, so that our load calculation applies to the whole
+	 * system. We don't care about races with it at all.
+	 */
+	static u64 last_under_load;
+	bool packet_needs_cookie;
+	bool under_load;
+
+	if (SKB_TYPE_LE32(skb) == cpu_to_le32(MESSAGE_HANDSHAKE_COOKIE)) {
+		net_dbg_skb_ratelimited("%s: Receiving cookie response from %pISpfsc\n",
+					wg->dev->name, skb);
+		wg_cookie_message_consume(
+			(struct message_handshake_cookie *)skb->data, wg);
+		return;
+	}
+
+	under_load = skb_queue_len(&wg->incoming_handshakes) >=
+		     MAX_QUEUED_INCOMING_HANDSHAKES / 8;
+	if (under_load)
+		last_under_load = ktime_get_coarse_boottime_ns();
+	else if (last_under_load)
+		under_load = !wg_birthdate_has_expired(last_under_load, 1);
+	mac_state = wg_cookie_validate_packet(&wg->cookie_checker, skb,
+					      under_load);
+	if ((under_load && mac_state == VALID_MAC_WITH_COOKIE) ||
+	    (!under_load && mac_state == VALID_MAC_BUT_NO_COOKIE)) {
+		packet_needs_cookie = false;
+	} else if (under_load && mac_state == VALID_MAC_BUT_NO_COOKIE) {
+		packet_needs_cookie = true;
+	} else {
+		net_dbg_skb_ratelimited("%s: Invalid MAC of handshake, dropping packet from %pISpfsc\n",
+					wg->dev->name, skb);
+		return;
+	}
+
+	switch (SKB_TYPE_LE32(skb)) {
+	case cpu_to_le32(MESSAGE_HANDSHAKE_INITIATION): {
+		struct message_handshake_initiation *message =
+			(struct message_handshake_initiation *)skb->data;
+
+		if (packet_needs_cookie) {
+			wg_packet_send_handshake_cookie(wg, skb,
+							message->sender_index);
+			return;
+		}
+		peer = wg_noise_handshake_consume_initiation(message, wg);
+		if (unlikely(!peer)) {
+			net_dbg_skb_ratelimited("%s: Invalid handshake initiation from %pISpfsc\n",
+						wg->dev->name, skb);
+			return;
+		}
+		wg_socket_set_peer_endpoint_from_skb(peer, skb);
+		net_dbg_ratelimited("%s: Receiving handshake initiation from peer %llu (%pISpfsc)\n",
+				    wg->dev->name, peer->internal_id,
+				    &peer->endpoint.addr);
+		wg_packet_send_handshake_response(peer);
+		break;
+	}
+	case cpu_to_le32(MESSAGE_HANDSHAKE_RESPONSE): {
+		struct message_handshake_response *message =
+			(struct message_handshake_response *)skb->data;
+
+		if (packet_needs_cookie) {
+			wg_packet_send_handshake_cookie(wg, skb,
+							message->sender_index);
+			return;
+		}
+		peer = wg_noise_handshake_consume_response(message, wg);
+		if (unlikely(!peer)) {
+			net_dbg_skb_ratelimited("%s: Invalid handshake response from %pISpfsc\n",
+						wg->dev->name, skb);
+			return;
+		}
+		wg_socket_set_peer_endpoint_from_skb(peer, skb);
+		net_dbg_ratelimited("%s: Receiving handshake response from peer %llu (%pISpfsc)\n",
+				    wg->dev->name, peer->internal_id,
+				    &peer->endpoint.addr);
+		if (wg_noise_handshake_begin_session(&peer->handshake,
+						     &peer->keypairs)) {
+			wg_timers_session_derived(peer);
+			wg_timers_handshake_complete(peer);
+			/* Calling this function will either send any existing
+			 * packets in the queue and not send a keepalive, which
+			 * is the best case, Or, if there's nothing in the
+			 * queue, it will send a keepalive, in order to give
+			 * immediate confirmation of the session.
+			 */
+			wg_packet_send_keepalive(peer);
+		}
+		break;
+	}
+	}
+
+	if (unlikely(!peer)) {
+		WARN(1, "Somehow a wrong type of packet wound up in the handshake queue!\n");
+		return;
+	}
+
+	local_bh_disable();
+	update_rx_stats(peer, skb->len);
+	local_bh_enable();
+
+	wg_timers_any_authenticated_packet_received(peer);
+	wg_timers_any_authenticated_packet_traversal(peer);
+	wg_peer_put(peer);
+}
+
+void wg_packet_handshake_receive_worker(struct work_struct *work)
+{
+	struct wg_device *wg = container_of(work, struct multicore_worker,
+					    work)->ptr;
+	struct sk_buff *skb;
+
+	while ((skb = skb_dequeue(&wg->incoming_handshakes)) != NULL) {
+		wg_receive_handshake_packet(wg, skb);
+		dev_kfree_skb(skb);
+		cond_resched();
+	}
+}
+
+static void keep_key_fresh(struct wg_peer *peer)
+{
+	struct noise_keypair *keypair;
+	bool send = false;
+
+	if (peer->sent_lastminute_handshake)
+		return;
+
+	rcu_read_lock_bh();
+	keypair = rcu_dereference_bh(peer->keypairs.current_keypair);
+	if (likely(keypair && READ_ONCE(keypair->sending.is_valid)) &&
+	    keypair->i_am_the_initiator &&
+	    unlikely(wg_birthdate_has_expired(keypair->sending.birthdate,
+			REJECT_AFTER_TIME - KEEPALIVE_TIMEOUT - REKEY_TIMEOUT)))
+		send = true;
+	rcu_read_unlock_bh();
+
+	if (send) {
+		peer->sent_lastminute_handshake = true;
+		wg_packet_send_queued_handshake_initiation(peer, false);
+	}
+}
+
+static bool decrypt_packet(struct sk_buff *skb, struct noise_symmetric_key *key)
+{
+	struct scatterlist sg[MAX_SKB_FRAGS + 8];
+	struct sk_buff *trailer;
+	unsigned int offset;
+	int num_frags;
+
+	if (unlikely(!key))
+		return false;
+
+	if (unlikely(!READ_ONCE(key->is_valid) ||
+		  wg_birthdate_has_expired(key->birthdate, REJECT_AFTER_TIME) ||
+		  key->counter.receive.counter >= REJECT_AFTER_MESSAGES)) {
+		WRITE_ONCE(key->is_valid, false);
+		return false;
+	}
+
+	PACKET_CB(skb)->nonce =
+		le64_to_cpu(((struct message_data *)skb->data)->counter);
+
+	/* We ensure that the network header is part of the packet before we
+	 * call skb_cow_data, so that there's no chance that data is removed
+	 * from the skb, so that later we can extract the original endpoint.
+	 */
+	offset = skb->data - skb_network_header(skb);
+	skb_push(skb, offset);
+	num_frags = skb_cow_data(skb, 0, &trailer);
+	offset += sizeof(struct message_data);
+	skb_pull(skb, offset);
+	if (unlikely(num_frags < 0 || num_frags > ARRAY_SIZE(sg)))
+		return false;
+
+	sg_init_table(sg, num_frags);
+	if (skb_to_sgvec(skb, sg, 0, skb->len) <= 0)
+		return false;
+
+	if (!chacha20poly1305_decrypt_sg_inplace(sg, skb->len, NULL, 0,
+					         PACKET_CB(skb)->nonce,
+						 key->key))
+		return false;
+
+	/* Another ugly situation of pushing and pulling the header so as to
+	 * keep endpoint information intact.
+	 */
+	skb_push(skb, offset);
+	if (pskb_trim(skb, skb->len - noise_encrypted_len(0)))
+		return false;
+	skb_pull(skb, offset);
+
+	return true;
+}
+
+/* This is RFC6479, a replay detection bitmap algorithm that avoids bitshifts */
+static bool counter_validate(union noise_counter *counter, u64 their_counter)
+{
+	unsigned long index, index_current, top, i;
+	bool ret = false;
+
+	spin_lock_bh(&counter->receive.lock);
+
+	if (unlikely(counter->receive.counter >= REJECT_AFTER_MESSAGES + 1 ||
+		     their_counter >= REJECT_AFTER_MESSAGES))
+		goto out;
+
+	++their_counter;
+
+	if (unlikely((COUNTER_WINDOW_SIZE + their_counter) <
+		     counter->receive.counter))
+		goto out;
+
+	index = their_counter >> ilog2(BITS_PER_LONG);
+
+	if (likely(their_counter > counter->receive.counter)) {
+		index_current = counter->receive.counter >> ilog2(BITS_PER_LONG);
+		top = min_t(unsigned long, index - index_current,
+			    COUNTER_BITS_TOTAL / BITS_PER_LONG);
+		for (i = 1; i <= top; ++i)
+			counter->receive.backtrack[(i + index_current) &
+				((COUNTER_BITS_TOTAL / BITS_PER_LONG) - 1)] = 0;
+		counter->receive.counter = their_counter;
+	}
+
+	index &= (COUNTER_BITS_TOTAL / BITS_PER_LONG) - 1;
+	ret = !test_and_set_bit(their_counter & (BITS_PER_LONG - 1),
+				&counter->receive.backtrack[index]);
+
+out:
+	spin_unlock_bh(&counter->receive.lock);
+	return ret;
+}
+
+#include "selftest/counter.c"
+
+static void wg_packet_consume_data_done(struct wg_peer *peer,
+					struct sk_buff *skb,
+					struct endpoint *endpoint)
+{
+	struct net_device *dev = peer->device->dev;
+	unsigned int len, len_before_trim;
+	struct wg_peer *routed_peer;
+
+	wg_socket_set_peer_endpoint(peer, endpoint);
+
+	if (unlikely(wg_noise_received_with_keypair(&peer->keypairs,
+						    PACKET_CB(skb)->keypair))) {
+		wg_timers_handshake_complete(peer);
+		wg_packet_send_staged_packets(peer);
+	}
+
+	keep_key_fresh(peer);
+
+	wg_timers_any_authenticated_packet_received(peer);
+	wg_timers_any_authenticated_packet_traversal(peer);
+
+	/* A packet with length 0 is a keepalive packet */
+	if (unlikely(!skb->len)) {
+		update_rx_stats(peer, message_data_len(0));
+		net_dbg_ratelimited("%s: Receiving keepalive packet from peer %llu (%pISpfsc)\n",
+				    dev->name, peer->internal_id,
+				    &peer->endpoint.addr);
+		goto packet_processed;
+	}
+
+	wg_timers_data_received(peer);
+
+	if (unlikely(skb_network_header(skb) < skb->head))
+		goto dishonest_packet_size;
+	if (unlikely(!(pskb_network_may_pull(skb, sizeof(struct iphdr)) &&
+		       (ip_hdr(skb)->version == 4 ||
+			(ip_hdr(skb)->version == 6 &&
+			 pskb_network_may_pull(skb, sizeof(struct ipv6hdr)))))))
+		goto dishonest_packet_type;
+
+	skb->dev = dev;
+	/* We've already verified the Poly1305 auth tag, which means this packet
+	 * was not modified in transit. We can therefore tell the networking
+	 * stack that all checksums of every layer of encapsulation have already
+	 * been checked "by the hardware" and therefore is unneccessary to check
+	 * again in software.
+	 */
+	skb->ip_summed = CHECKSUM_UNNECESSARY;
+	skb->csum_level = ~0; /* All levels */
+	skb->protocol = wg_skb_examine_untrusted_ip_hdr(skb);
+	if (skb->protocol == htons(ETH_P_IP)) {
+		len = ntohs(ip_hdr(skb)->tot_len);
+		if (unlikely(len < sizeof(struct iphdr)))
+			goto dishonest_packet_size;
+		if (INET_ECN_is_ce(PACKET_CB(skb)->ds))
+			IP_ECN_set_ce(ip_hdr(skb));
+	} else if (skb->protocol == htons(ETH_P_IPV6)) {
+		len = ntohs(ipv6_hdr(skb)->payload_len) +
+		      sizeof(struct ipv6hdr);
+		if (INET_ECN_is_ce(PACKET_CB(skb)->ds))
+			IP6_ECN_set_ce(skb, ipv6_hdr(skb));
+	} else {
+		goto dishonest_packet_type;
+	}
+
+	if (unlikely(len > skb->len))
+		goto dishonest_packet_size;
+	len_before_trim = skb->len;
+	if (unlikely(pskb_trim(skb, len)))
+		goto packet_processed;
+
+	routed_peer = wg_allowedips_lookup_src(&peer->device->peer_allowedips,
+					       skb);
+	wg_peer_put(routed_peer); /* We don't need the extra reference. */
+
+	if (unlikely(routed_peer != peer))
+		goto dishonest_packet_peer;
+
+	if (unlikely(napi_gro_receive(&peer->napi, skb) == GRO_DROP)) {
+		++dev->stats.rx_dropped;
+		net_dbg_ratelimited("%s: Failed to give packet to userspace from peer %llu (%pISpfsc)\n",
+				    dev->name, peer->internal_id,
+				    &peer->endpoint.addr);
+	} else {
+		update_rx_stats(peer, message_data_len(len_before_trim));
+	}
+	return;
+
+dishonest_packet_peer:
+	net_dbg_skb_ratelimited("%s: Packet has unallowed src IP (%pISc) from peer %llu (%pISpfsc)\n",
+				dev->name, skb, peer->internal_id,
+				&peer->endpoint.addr);
+	++dev->stats.rx_errors;
+	++dev->stats.rx_frame_errors;
+	goto packet_processed;
+dishonest_packet_type:
+	net_dbg_ratelimited("%s: Packet is neither ipv4 nor ipv6 from peer %llu (%pISpfsc)\n",
+			    dev->name, peer->internal_id, &peer->endpoint.addr);
+	++dev->stats.rx_errors;
+	++dev->stats.rx_frame_errors;
+	goto packet_processed;
+dishonest_packet_size:
+	net_dbg_ratelimited("%s: Packet has incorrect size from peer %llu (%pISpfsc)\n",
+			    dev->name, peer->internal_id, &peer->endpoint.addr);
+	++dev->stats.rx_errors;
+	++dev->stats.rx_length_errors;
+	goto packet_processed;
+packet_processed:
+	dev_kfree_skb(skb);
+}
+
+int wg_packet_rx_poll(struct napi_struct *napi, int budget)
+{
+	struct wg_peer *peer = container_of(napi, struct wg_peer, napi);
+	struct crypt_queue *queue = &peer->rx_queue;
+	struct noise_keypair *keypair;
+	struct endpoint endpoint;
+	enum packet_state state;
+	struct sk_buff *skb;
+	int work_done = 0;
+	bool free;
+
+	if (unlikely(budget <= 0))
+		return 0;
+
+	while ((skb = __ptr_ring_peek(&queue->ring)) != NULL &&
+	       (state = atomic_read_acquire(&PACKET_CB(skb)->state)) !=
+		       PACKET_STATE_UNCRYPTED) {
+		__ptr_ring_discard_one(&queue->ring);
+		peer = PACKET_PEER(skb);
+		keypair = PACKET_CB(skb)->keypair;
+		free = true;
+
+		if (unlikely(state != PACKET_STATE_CRYPTED))
+			goto next;
+
+		if (unlikely(!counter_validate(&keypair->receiving.counter,
+					       PACKET_CB(skb)->nonce))) {
+			net_dbg_ratelimited("%s: Packet has invalid nonce %llu (max %llu)\n",
+					    peer->device->dev->name,
+					    PACKET_CB(skb)->nonce,
+					    keypair->receiving.counter.receive.counter);
+			goto next;
+		}
+
+		if (unlikely(wg_socket_endpoint_from_skb(&endpoint, skb)))
+			goto next;
+
+		wg_reset_packet(skb);
+		wg_packet_consume_data_done(peer, skb, &endpoint);
+		free = false;
+
+next:
+		wg_noise_keypair_put(keypair, false);
+		wg_peer_put(peer);
+		if (unlikely(free))
+			dev_kfree_skb(skb);
+
+		if (++work_done >= budget)
+			break;
+	}
+
+	if (work_done < budget)
+		napi_complete_done(napi, work_done);
+
+	return work_done;
+}
+
+void wg_packet_decrypt_worker(struct work_struct *work)
+{
+	struct crypt_queue *queue = container_of(work, struct multicore_worker,
+						 work)->ptr;
+	struct sk_buff *skb;
+
+	while ((skb = ptr_ring_consume_bh(&queue->ring)) != NULL) {
+		enum packet_state state = likely(decrypt_packet(skb,
+				&PACKET_CB(skb)->keypair->receiving)) ?
+				PACKET_STATE_CRYPTED : PACKET_STATE_DEAD;
+		wg_queue_enqueue_per_peer_napi(skb, state);
+	}
+}
+
+static void wg_packet_consume_data(struct wg_device *wg, struct sk_buff *skb)
+{
+	__le32 idx = ((struct message_data *)skb->data)->key_idx;
+	struct wg_peer *peer = NULL;
+	int ret;
+
+	rcu_read_lock_bh();
+	PACKET_CB(skb)->keypair =
+		(struct noise_keypair *)wg_index_hashtable_lookup(
+			wg->index_hashtable, INDEX_HASHTABLE_KEYPAIR, idx,
+			&peer);
+	if (unlikely(!wg_noise_keypair_get(PACKET_CB(skb)->keypair)))
+		goto err_keypair;
+
+	if (unlikely(READ_ONCE(peer->is_dead)))
+		goto err;
+
+	ret = wg_queue_enqueue_per_device_and_peer(&wg->decrypt_queue,
+						   &peer->rx_queue, skb,
+						   wg->packet_crypt_wq,
+						   &wg->decrypt_queue.last_cpu);
+	if (unlikely(ret == -EPIPE))
+		wg_queue_enqueue_per_peer_napi(skb, PACKET_STATE_DEAD);
+	if (likely(!ret || ret == -EPIPE)) {
+		rcu_read_unlock_bh();
+		return;
+	}
+err:
+	wg_noise_keypair_put(PACKET_CB(skb)->keypair, false);
+err_keypair:
+	rcu_read_unlock_bh();
+	wg_peer_put(peer);
+	dev_kfree_skb(skb);
+}
+
+void wg_packet_receive(struct wg_device *wg, struct sk_buff *skb)
+{
+	if (unlikely(prepare_skb_header(skb, wg) < 0))
+		goto err;
+	switch (SKB_TYPE_LE32(skb)) {
+	case cpu_to_le32(MESSAGE_HANDSHAKE_INITIATION):
+	case cpu_to_le32(MESSAGE_HANDSHAKE_RESPONSE):
+	case cpu_to_le32(MESSAGE_HANDSHAKE_COOKIE): {
+		int cpu;
+
+		if (skb_queue_len(&wg->incoming_handshakes) >
+			    MAX_QUEUED_INCOMING_HANDSHAKES ||
+		    unlikely(!rng_is_initialized())) {
+			net_dbg_skb_ratelimited("%s: Dropping handshake packet from %pISpfsc\n",
+						wg->dev->name, skb);
+			goto err;
+		}
+		skb_queue_tail(&wg->incoming_handshakes, skb);
+		/* Queues up a call to packet_process_queued_handshake_
+		 * packets(skb):
+		 */
+		cpu = wg_cpumask_next_online(&wg->incoming_handshake_cpu);
+		queue_work_on(cpu, wg->handshake_receive_wq,
+			&per_cpu_ptr(wg->incoming_handshakes_worker, cpu)->work);
+		break;
+	}
+	case cpu_to_le32(MESSAGE_DATA):
+		PACKET_CB(skb)->ds = ip_tunnel_get_dsfield(ip_hdr(skb), skb);
+		wg_packet_consume_data(wg, skb);
+		break;
+	default:
+		net_dbg_skb_ratelimited("%s: Invalid packet from %pISpfsc\n",
+					wg->dev->name, skb);
+		goto err;
+	}
+	return;
+
+err:
+	dev_kfree_skb(skb);
+}
diff --git a/drivers/net/wireguard/selftest/allowedips.c b/drivers/net/wireguard/selftest/allowedips.c
new file mode 100644
index 000000000000..846db14cb046
--- /dev/null
+++ b/drivers/net/wireguard/selftest/allowedips.c
@@ -0,0 +1,683 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ *
+ * This contains some basic static unit tests for the allowedips data structure.
+ * It also has two additional modes that are disabled and meant to be used by
+ * folks directly playing with this file. If you define the macro
+ * DEBUG_PRINT_TRIE_GRAPHVIZ to be 1, then every time there's a full tree in
+ * memory, it will be printed out as KERN_DEBUG in a format that can be passed
+ * to graphviz (the dot command) to visualize it. If you define the macro
+ * DEBUG_RANDOM_TRIE to be 1, then there will be an extremely costly set of
+ * randomized tests done against a trivial implementation, which may take
+ * upwards of a half-hour to complete. There's no set of users who should be
+ * enabling these, and the only developers that should go anywhere near these
+ * nobs are the ones who are reading this comment.
+ */
+
+#ifdef DEBUG
+
+#include <linux/siphash.h>
+
+static __init void swap_endian_and_apply_cidr(u8 *dst, const u8 *src, u8 bits,
+					      u8 cidr)
+{
+	swap_endian(dst, src, bits);
+	memset(dst + (cidr + 7) / 8, 0, bits / 8 - (cidr + 7) / 8);
+	if (cidr)
+		dst[(cidr + 7) / 8 - 1] &= ~0U << ((8 - (cidr % 8)) % 8);
+}
+
+static __init void print_node(struct allowedips_node *node, u8 bits)
+{
+	char *fmt_connection = KERN_DEBUG "\t\"%p/%d\" -> \"%p/%d\";\n";
+	char *fmt_declaration = KERN_DEBUG
+		"\t\"%p/%d\"[style=%s, color=\"#%06x\"];\n";
+	char *style = "dotted";
+	u8 ip1[16], ip2[16];
+	u32 color = 0;
+
+	if (bits == 32) {
+		fmt_connection = KERN_DEBUG "\t\"%pI4/%d\" -> \"%pI4/%d\";\n";
+		fmt_declaration = KERN_DEBUG
+			"\t\"%pI4/%d\"[style=%s, color=\"#%06x\"];\n";
+	} else if (bits == 128) {
+		fmt_connection = KERN_DEBUG "\t\"%pI6/%d\" -> \"%pI6/%d\";\n";
+		fmt_declaration = KERN_DEBUG
+			"\t\"%pI6/%d\"[style=%s, color=\"#%06x\"];\n";
+	}
+	if (node->peer) {
+		hsiphash_key_t key = { { 0 } };
+
+		memcpy(&key, &node->peer, sizeof(node->peer));
+		color = hsiphash_1u32(0xdeadbeef, &key) % 200 << 16 |
+			hsiphash_1u32(0xbabecafe, &key) % 200 << 8 |
+			hsiphash_1u32(0xabad1dea, &key) % 200;
+		style = "bold";
+	}
+	swap_endian_and_apply_cidr(ip1, node->bits, bits, node->cidr);
+	printk(fmt_declaration, ip1, node->cidr, style, color);
+	if (node->bit[0]) {
+		swap_endian_and_apply_cidr(ip2,
+				rcu_dereference_raw(node->bit[0])->bits, bits,
+				node->cidr);
+		printk(fmt_connection, ip1, node->cidr, ip2,
+		       rcu_dereference_raw(node->bit[0])->cidr);
+		print_node(rcu_dereference_raw(node->bit[0]), bits);
+	}
+	if (node->bit[1]) {
+		swap_endian_and_apply_cidr(ip2,
+				rcu_dereference_raw(node->bit[1])->bits,
+				bits, node->cidr);
+		printk(fmt_connection, ip1, node->cidr, ip2,
+		       rcu_dereference_raw(node->bit[1])->cidr);
+		print_node(rcu_dereference_raw(node->bit[1]), bits);
+	}
+}
+
+static __init void print_tree(struct allowedips_node __rcu *top, u8 bits)
+{
+	printk(KERN_DEBUG "digraph trie {\n");
+	print_node(rcu_dereference_raw(top), bits);
+	printk(KERN_DEBUG "}\n");
+}
+
+enum {
+	NUM_PEERS = 2000,
+	NUM_RAND_ROUTES = 400,
+	NUM_MUTATED_ROUTES = 100,
+	NUM_QUERIES = NUM_RAND_ROUTES * NUM_MUTATED_ROUTES * 30
+};
+
+struct horrible_allowedips {
+	struct hlist_head head;
+};
+
+struct horrible_allowedips_node {
+	struct hlist_node table;
+	union nf_inet_addr ip;
+	union nf_inet_addr mask;
+	u8 ip_version;
+	void *value;
+};
+
+static __init void horrible_allowedips_init(struct horrible_allowedips *table)
+{
+	INIT_HLIST_HEAD(&table->head);
+}
+
+static __init void horrible_allowedips_free(struct horrible_allowedips *table)
+{
+	struct horrible_allowedips_node *node;
+	struct hlist_node *h;
+
+	hlist_for_each_entry_safe(node, h, &table->head, table) {
+		hlist_del(&node->table);
+		kfree(node);
+	}
+}
+
+static __init inline union nf_inet_addr horrible_cidr_to_mask(u8 cidr)
+{
+	union nf_inet_addr mask;
+
+	memset(&mask, 0x00, 128 / 8);
+	memset(&mask, 0xff, cidr / 8);
+	if (cidr % 32)
+		mask.all[cidr / 32] = (__force u32)htonl(
+			(0xFFFFFFFFUL << (32 - (cidr % 32))) & 0xFFFFFFFFUL);
+	return mask;
+}
+
+static __init inline u8 horrible_mask_to_cidr(union nf_inet_addr subnet)
+{
+	return hweight32(subnet.all[0]) + hweight32(subnet.all[1]) +
+	       hweight32(subnet.all[2]) + hweight32(subnet.all[3]);
+}
+
+static __init inline void
+horrible_mask_self(struct horrible_allowedips_node *node)
+{
+	if (node->ip_version == 4) {
+		node->ip.ip &= node->mask.ip;
+	} else if (node->ip_version == 6) {
+		node->ip.ip6[0] &= node->mask.ip6[0];
+		node->ip.ip6[1] &= node->mask.ip6[1];
+		node->ip.ip6[2] &= node->mask.ip6[2];
+		node->ip.ip6[3] &= node->mask.ip6[3];
+	}
+}
+
+static __init inline bool
+horrible_match_v4(const struct horrible_allowedips_node *node,
+		  struct in_addr *ip)
+{
+	return (ip->s_addr & node->mask.ip) == node->ip.ip;
+}
+
+static __init inline bool
+horrible_match_v6(const struct horrible_allowedips_node *node,
+		  struct in6_addr *ip)
+{
+	return (ip->in6_u.u6_addr32[0] & node->mask.ip6[0]) ==
+		       node->ip.ip6[0] &&
+	       (ip->in6_u.u6_addr32[1] & node->mask.ip6[1]) ==
+		       node->ip.ip6[1] &&
+	       (ip->in6_u.u6_addr32[2] & node->mask.ip6[2]) ==
+		       node->ip.ip6[2] &&
+	       (ip->in6_u.u6_addr32[3] & node->mask.ip6[3]) == node->ip.ip6[3];
+}
+
+static __init void
+horrible_insert_ordered(struct horrible_allowedips *table,
+			struct horrible_allowedips_node *node)
+{
+	struct horrible_allowedips_node *other = NULL, *where = NULL;
+	u8 my_cidr = horrible_mask_to_cidr(node->mask);
+
+	hlist_for_each_entry(other, &table->head, table) {
+		if (!memcmp(&other->mask, &node->mask,
+			    sizeof(union nf_inet_addr)) &&
+		    !memcmp(&other->ip, &node->ip,
+			    sizeof(union nf_inet_addr)) &&
+		    other->ip_version == node->ip_version) {
+			other->value = node->value;
+			kfree(node);
+			return;
+		}
+		where = other;
+		if (horrible_mask_to_cidr(other->mask) <= my_cidr)
+			break;
+	}
+	if (!other && !where)
+		hlist_add_head(&node->table, &table->head);
+	else if (!other)
+		hlist_add_behind(&node->table, &where->table);
+	else
+		hlist_add_before(&node->table, &where->table);
+}
+
+static __init int
+horrible_allowedips_insert_v4(struct horrible_allowedips *table,
+			      struct in_addr *ip, u8 cidr, void *value)
+{
+	struct horrible_allowedips_node *node = kzalloc(sizeof(*node),
+							GFP_KERNEL);
+
+	if (unlikely(!node))
+		return -ENOMEM;
+	node->ip.in = *ip;
+	node->mask = horrible_cidr_to_mask(cidr);
+	node->ip_version = 4;
+	node->value = value;
+	horrible_mask_self(node);
+	horrible_insert_ordered(table, node);
+	return 0;
+}
+
+static __init int
+horrible_allowedips_insert_v6(struct horrible_allowedips *table,
+			      struct in6_addr *ip, u8 cidr, void *value)
+{
+	struct horrible_allowedips_node *node = kzalloc(sizeof(*node),
+							GFP_KERNEL);
+
+	if (unlikely(!node))
+		return -ENOMEM;
+	node->ip.in6 = *ip;
+	node->mask = horrible_cidr_to_mask(cidr);
+	node->ip_version = 6;
+	node->value = value;
+	horrible_mask_self(node);
+	horrible_insert_ordered(table, node);
+	return 0;
+}
+
+static __init void *
+horrible_allowedips_lookup_v4(struct horrible_allowedips *table,
+			      struct in_addr *ip)
+{
+	struct horrible_allowedips_node *node;
+	void *ret = NULL;
+
+	hlist_for_each_entry(node, &table->head, table) {
+		if (node->ip_version != 4)
+			continue;
+		if (horrible_match_v4(node, ip)) {
+			ret = node->value;
+			break;
+		}
+	}
+	return ret;
+}
+
+static __init void *
+horrible_allowedips_lookup_v6(struct horrible_allowedips *table,
+			      struct in6_addr *ip)
+{
+	struct horrible_allowedips_node *node;
+	void *ret = NULL;
+
+	hlist_for_each_entry(node, &table->head, table) {
+		if (node->ip_version != 6)
+			continue;
+		if (horrible_match_v6(node, ip)) {
+			ret = node->value;
+			break;
+		}
+	}
+	return ret;
+}
+
+static __init bool randomized_test(void)
+{
+	unsigned int i, j, k, mutate_amount, cidr;
+	u8 ip[16], mutate_mask[16], mutated[16];
+	struct wg_peer **peers, *peer;
+	struct horrible_allowedips h;
+	DEFINE_MUTEX(mutex);
+	struct allowedips t;
+	bool ret = false;
+
+	mutex_init(&mutex);
+
+	wg_allowedips_init(&t);
+	horrible_allowedips_init(&h);
+
+	peers = kcalloc(NUM_PEERS, sizeof(*peers), GFP_KERNEL);
+	if (unlikely(!peers)) {
+		pr_err("allowedips random self-test malloc: FAIL\n");
+		goto free;
+	}
+	for (i = 0; i < NUM_PEERS; ++i) {
+		peers[i] = kzalloc(sizeof(*peers[i]), GFP_KERNEL);
+		if (unlikely(!peers[i])) {
+			pr_err("allowedips random self-test malloc: FAIL\n");
+			goto free;
+		}
+		kref_init(&peers[i]->refcount);
+	}
+
+	mutex_lock(&mutex);
+
+	for (i = 0; i < NUM_RAND_ROUTES; ++i) {
+		prandom_bytes(ip, 4);
+		cidr = prandom_u32_max(32) + 1;
+		peer = peers[prandom_u32_max(NUM_PEERS)];
+		if (wg_allowedips_insert_v4(&t, (struct in_addr *)ip, cidr,
+					    peer, &mutex) < 0) {
+			pr_err("allowedips random self-test malloc: FAIL\n");
+			goto free_locked;
+		}
+		if (horrible_allowedips_insert_v4(&h, (struct in_addr *)ip,
+						  cidr, peer) < 0) {
+			pr_err("allowedips random self-test malloc: FAIL\n");
+			goto free_locked;
+		}
+		for (j = 0; j < NUM_MUTATED_ROUTES; ++j) {
+			memcpy(mutated, ip, 4);
+			prandom_bytes(mutate_mask, 4);
+			mutate_amount = prandom_u32_max(32);
+			for (k = 0; k < mutate_amount / 8; ++k)
+				mutate_mask[k] = 0xff;
+			mutate_mask[k] = 0xff
+					 << ((8 - (mutate_amount % 8)) % 8);
+			for (; k < 4; ++k)
+				mutate_mask[k] = 0;
+			for (k = 0; k < 4; ++k)
+				mutated[k] = (mutated[k] & mutate_mask[k]) |
+					     (~mutate_mask[k] &
+					      prandom_u32_max(256));
+			cidr = prandom_u32_max(32) + 1;
+			peer = peers[prandom_u32_max(NUM_PEERS)];
+			if (wg_allowedips_insert_v4(&t,
+						    (struct in_addr *)mutated,
+						    cidr, peer, &mutex) < 0) {
+				pr_err("allowedips random malloc: FAIL\n");
+				goto free_locked;
+			}
+			if (horrible_allowedips_insert_v4(&h,
+				(struct in_addr *)mutated, cidr, peer)) {
+				pr_err("allowedips random self-test malloc: FAIL\n");
+				goto free_locked;
+			}
+		}
+	}
+
+	for (i = 0; i < NUM_RAND_ROUTES; ++i) {
+		prandom_bytes(ip, 16);
+		cidr = prandom_u32_max(128) + 1;
+		peer = peers[prandom_u32_max(NUM_PEERS)];
+		if (wg_allowedips_insert_v6(&t, (struct in6_addr *)ip, cidr,
+					    peer, &mutex) < 0) {
+			pr_err("allowedips random self-test malloc: FAIL\n");
+			goto free_locked;
+		}
+		if (horrible_allowedips_insert_v6(&h, (struct in6_addr *)ip,
+						  cidr, peer) < 0) {
+			pr_err("allowedips random self-test malloc: FAIL\n");
+			goto free_locked;
+		}
+		for (j = 0; j < NUM_MUTATED_ROUTES; ++j) {
+			memcpy(mutated, ip, 16);
+			prandom_bytes(mutate_mask, 16);
+			mutate_amount = prandom_u32_max(128);
+			for (k = 0; k < mutate_amount / 8; ++k)
+				mutate_mask[k] = 0xff;
+			mutate_mask[k] = 0xff
+					 << ((8 - (mutate_amount % 8)) % 8);
+			for (; k < 4; ++k)
+				mutate_mask[k] = 0;
+			for (k = 0; k < 4; ++k)
+				mutated[k] = (mutated[k] & mutate_mask[k]) |
+					     (~mutate_mask[k] &
+					      prandom_u32_max(256));
+			cidr = prandom_u32_max(128) + 1;
+			peer = peers[prandom_u32_max(NUM_PEERS)];
+			if (wg_allowedips_insert_v6(&t,
+						    (struct in6_addr *)mutated,
+						    cidr, peer, &mutex) < 0) {
+				pr_err("allowedips random self-test malloc: FAIL\n");
+				goto free_locked;
+			}
+			if (horrible_allowedips_insert_v6(
+				    &h, (struct in6_addr *)mutated, cidr,
+				    peer)) {
+				pr_err("allowedips random self-test malloc: FAIL\n");
+				goto free_locked;
+			}
+		}
+	}
+
+	mutex_unlock(&mutex);
+
+	if (IS_ENABLED(DEBUG_PRINT_TRIE_GRAPHVIZ)) {
+		print_tree(t.root4, 32);
+		print_tree(t.root6, 128);
+	}
+
+	for (i = 0; i < NUM_QUERIES; ++i) {
+		prandom_bytes(ip, 4);
+		if (lookup(t.root4, 32, ip) !=
+		    horrible_allowedips_lookup_v4(&h, (struct in_addr *)ip)) {
+			pr_err("allowedips random self-test: FAIL\n");
+			goto free;
+		}
+	}
+
+	for (i = 0; i < NUM_QUERIES; ++i) {
+		prandom_bytes(ip, 16);
+		if (lookup(t.root6, 128, ip) !=
+		    horrible_allowedips_lookup_v6(&h, (struct in6_addr *)ip)) {
+			pr_err("allowedips random self-test: FAIL\n");
+			goto free;
+		}
+	}
+	ret = true;
+
+free:
+	mutex_lock(&mutex);
+free_locked:
+	wg_allowedips_free(&t, &mutex);
+	mutex_unlock(&mutex);
+	horrible_allowedips_free(&h);
+	if (peers) {
+		for (i = 0; i < NUM_PEERS; ++i)
+			kfree(peers[i]);
+	}
+	kfree(peers);
+	return ret;
+}
+
+static __init inline struct in_addr *ip4(u8 a, u8 b, u8 c, u8 d)
+{
+	static struct in_addr ip;
+	u8 *split = (u8 *)&ip;
+
+	split[0] = a;
+	split[1] = b;
+	split[2] = c;
+	split[3] = d;
+	return &ip;
+}
+
+static __init inline struct in6_addr *ip6(u32 a, u32 b, u32 c, u32 d)
+{
+	static struct in6_addr ip;
+	__be32 *split = (__be32 *)&ip;
+
+	split[0] = cpu_to_be32(a);
+	split[1] = cpu_to_be32(b);
+	split[2] = cpu_to_be32(c);
+	split[3] = cpu_to_be32(d);
+	return &ip;
+}
+
+static __init struct wg_peer *init_peer(void)
+{
+	struct wg_peer *peer = kzalloc(sizeof(*peer), GFP_KERNEL);
+
+	if (!peer)
+		return NULL;
+	kref_init(&peer->refcount);
+	INIT_LIST_HEAD(&peer->allowedips_list);
+	return peer;
+}
+
+#define insert(version, mem, ipa, ipb, ipc, ipd, cidr)                       \
+	wg_allowedips_insert_v##version(&t, ip##version(ipa, ipb, ipc, ipd), \
+					cidr, mem, &mutex)
+
+#define maybe_fail() do {                                               \
+		++i;                                                    \
+		if (!_s) {                                              \
+			pr_info("allowedips self-test %zu: FAIL\n", i); \
+			success = false;                                \
+		}                                                       \
+	} while (0)
+
+#define test(version, mem, ipa, ipb, ipc, ipd) do {                          \
+		bool _s = lookup(t.root##version, (version) == 4 ? 32 : 128, \
+				 ip##version(ipa, ipb, ipc, ipd)) == (mem);  \
+		maybe_fail();                                                \
+	} while (0)
+
+#define test_negative(version, mem, ipa, ipb, ipc, ipd) do {                 \
+		bool _s = lookup(t.root##version, (version) == 4 ? 32 : 128, \
+				 ip##version(ipa, ipb, ipc, ipd)) != (mem);  \
+		maybe_fail();                                                \
+	} while (0)
+
+#define test_boolean(cond) do {   \
+		bool _s = (cond); \
+		maybe_fail();     \
+	} while (0)
+
+bool __init wg_allowedips_selftest(void)
+{
+	bool found_a = false, found_b = false, found_c = false, found_d = false,
+	     found_e = false, found_other = false;
+	struct wg_peer *a = init_peer(), *b = init_peer(), *c = init_peer(),
+		       *d = init_peer(), *e = init_peer(), *f = init_peer(),
+		       *g = init_peer(), *h = init_peer();
+	struct allowedips_node *iter_node;
+	bool success = false;
+	struct allowedips t;
+	DEFINE_MUTEX(mutex);
+	struct in6_addr ip;
+	size_t i = 0, count = 0;
+	__be64 part;
+
+	mutex_init(&mutex);
+	mutex_lock(&mutex);
+	wg_allowedips_init(&t);
+
+	if (!a || !b || !c || !d || !e || !f || !g || !h) {
+		pr_err("allowedips self-test malloc: FAIL\n");
+		goto free;
+	}
+
+	insert(4, a, 192, 168, 4, 0, 24);
+	insert(4, b, 192, 168, 4, 4, 32);
+	insert(4, c, 192, 168, 0, 0, 16);
+	insert(4, d, 192, 95, 5, 64, 27);
+	/* replaces previous entry, and maskself is required */
+	insert(4, c, 192, 95, 5, 65, 27);
+	insert(6, d, 0x26075300, 0x60006b00, 0, 0xc05f0543, 128);
+	insert(6, c, 0x26075300, 0x60006b00, 0, 0, 64);
+	insert(4, e, 0, 0, 0, 0, 0);
+	insert(6, e, 0, 0, 0, 0, 0);
+	/* replaces previous entry */
+	insert(6, f, 0, 0, 0, 0, 0);
+	insert(6, g, 0x24046800, 0, 0, 0, 32);
+	/* maskself is required */
+	insert(6, h, 0x24046800, 0x40040800, 0xdeadbeef, 0xdeadbeef, 64);
+	insert(6, a, 0x24046800, 0x40040800, 0xdeadbeef, 0xdeadbeef, 128);
+	insert(6, c, 0x24446800, 0x40e40800, 0xdeaebeef, 0xdefbeef, 128);
+	insert(6, b, 0x24446800, 0xf0e40800, 0xeeaebeef, 0, 98);
+	insert(4, g, 64, 15, 112, 0, 20);
+	/* maskself is required */
+	insert(4, h, 64, 15, 123, 211, 25);
+	insert(4, a, 10, 0, 0, 0, 25);
+	insert(4, b, 10, 0, 0, 128, 25);
+	insert(4, a, 10, 1, 0, 0, 30);
+	insert(4, b, 10, 1, 0, 4, 30);
+	insert(4, c, 10, 1, 0, 8, 29);
+	insert(4, d, 10, 1, 0, 16, 29);
+
+	if (IS_ENABLED(DEBUG_PRINT_TRIE_GRAPHVIZ)) {
+		print_tree(t.root4, 32);
+		print_tree(t.root6, 128);
+	}
+
+	success = true;
+
+	test(4, a, 192, 168, 4, 20);
+	test(4, a, 192, 168, 4, 0);
+	test(4, b, 192, 168, 4, 4);
+	test(4, c, 192, 168, 200, 182);
+	test(4, c, 192, 95, 5, 68);
+	test(4, e, 192, 95, 5, 96);
+	test(6, d, 0x26075300, 0x60006b00, 0, 0xc05f0543);
+	test(6, c, 0x26075300, 0x60006b00, 0, 0xc02e01ee);
+	test(6, f, 0x26075300, 0x60006b01, 0, 0);
+	test(6, g, 0x24046800, 0x40040806, 0, 0x1006);
+	test(6, g, 0x24046800, 0x40040806, 0x1234, 0x5678);
+	test(6, f, 0x240467ff, 0x40040806, 0x1234, 0x5678);
+	test(6, f, 0x24046801, 0x40040806, 0x1234, 0x5678);
+	test(6, h, 0x24046800, 0x40040800, 0x1234, 0x5678);
+	test(6, h, 0x24046800, 0x40040800, 0, 0);
+	test(6, h, 0x24046800, 0x40040800, 0x10101010, 0x10101010);
+	test(6, a, 0x24046800, 0x40040800, 0xdeadbeef, 0xdeadbeef);
+	test(4, g, 64, 15, 116, 26);
+	test(4, g, 64, 15, 127, 3);
+	test(4, g, 64, 15, 123, 1);
+	test(4, h, 64, 15, 123, 128);
+	test(4, h, 64, 15, 123, 129);
+	test(4, a, 10, 0, 0, 52);
+	test(4, b, 10, 0, 0, 220);
+	test(4, a, 10, 1, 0, 2);
+	test(4, b, 10, 1, 0, 6);
+	test(4, c, 10, 1, 0, 10);
+	test(4, d, 10, 1, 0, 20);
+
+	insert(4, a, 1, 0, 0, 0, 32);
+	insert(4, a, 64, 0, 0, 0, 32);
+	insert(4, a, 128, 0, 0, 0, 32);
+	insert(4, a, 192, 0, 0, 0, 32);
+	insert(4, a, 255, 0, 0, 0, 32);
+	wg_allowedips_remove_by_peer(&t, a, &mutex);
+	test_negative(4, a, 1, 0, 0, 0);
+	test_negative(4, a, 64, 0, 0, 0);
+	test_negative(4, a, 128, 0, 0, 0);
+	test_negative(4, a, 192, 0, 0, 0);
+	test_negative(4, a, 255, 0, 0, 0);
+
+	wg_allowedips_free(&t, &mutex);
+	wg_allowedips_init(&t);
+	insert(4, a, 192, 168, 0, 0, 16);
+	insert(4, a, 192, 168, 0, 0, 24);
+	wg_allowedips_remove_by_peer(&t, a, &mutex);
+	test_negative(4, a, 192, 168, 0, 1);
+
+	/* These will hit the WARN_ON(len >= 128) in free_node if something
+	 * goes wrong.
+	 */
+	for (i = 0; i < 128; ++i) {
+		part = cpu_to_be64(~(1LLU << (i % 64)));
+		memset(&ip, 0xff, 16);
+		memcpy((u8 *)&ip + (i < 64) * 8, &part, 8);
+		wg_allowedips_insert_v6(&t, &ip, 128, a, &mutex);
+	}
+
+	wg_allowedips_free(&t, &mutex);
+
+	wg_allowedips_init(&t);
+	insert(4, a, 192, 95, 5, 93, 27);
+	insert(6, a, 0x26075300, 0x60006b00, 0, 0xc05f0543, 128);
+	insert(4, a, 10, 1, 0, 20, 29);
+	insert(6, a, 0x26075300, 0x6d8a6bf8, 0xdab1f1df, 0xc05f1523, 83);
+	insert(6, a, 0x26075300, 0x6d8a6bf8, 0xdab1f1df, 0xc05f1523, 21);
+	list_for_each_entry(iter_node, &a->allowedips_list, peer_list) {
+		u8 cidr, ip[16] __aligned(__alignof(u64));
+		int family = wg_allowedips_read_node(iter_node, ip, &cidr);
+
+		count++;
+
+		if (cidr == 27 && family == AF_INET &&
+		    !memcmp(ip, ip4(192, 95, 5, 64), sizeof(struct in_addr)))
+			found_a = true;
+		else if (cidr == 128 && family == AF_INET6 &&
+			 !memcmp(ip, ip6(0x26075300, 0x60006b00, 0, 0xc05f0543),
+				 sizeof(struct in6_addr)))
+			found_b = true;
+		else if (cidr == 29 && family == AF_INET &&
+			 !memcmp(ip, ip4(10, 1, 0, 16), sizeof(struct in_addr)))
+			found_c = true;
+		else if (cidr == 83 && family == AF_INET6 &&
+			 !memcmp(ip, ip6(0x26075300, 0x6d8a6bf8, 0xdab1e000, 0),
+				 sizeof(struct in6_addr)))
+			found_d = true;
+		else if (cidr == 21 && family == AF_INET6 &&
+			 !memcmp(ip, ip6(0x26075000, 0, 0, 0),
+				 sizeof(struct in6_addr)))
+			found_e = true;
+		else
+			found_other = true;
+	}
+	test_boolean(count == 5);
+	test_boolean(found_a);
+	test_boolean(found_b);
+	test_boolean(found_c);
+	test_boolean(found_d);
+	test_boolean(found_e);
+	test_boolean(!found_other);
+
+	if (IS_ENABLED(DEBUG_RANDOM_TRIE) && success)
+		success = randomized_test();
+
+	if (success)
+		pr_info("allowedips self-tests: pass\n");
+
+free:
+	wg_allowedips_free(&t, &mutex);
+	kfree(a);
+	kfree(b);
+	kfree(c);
+	kfree(d);
+	kfree(e);
+	kfree(f);
+	kfree(g);
+	kfree(h);
+	mutex_unlock(&mutex);
+
+	return success;
+}
+
+#undef test_negative
+#undef test
+#undef remove
+#undef insert
+#undef init_peer
+
+#endif
diff --git a/drivers/net/wireguard/selftest/counter.c b/drivers/net/wireguard/selftest/counter.c
new file mode 100644
index 000000000000..f4fbb9072ed7
--- /dev/null
+++ b/drivers/net/wireguard/selftest/counter.c
@@ -0,0 +1,104 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#ifdef DEBUG
+bool __init wg_packet_counter_selftest(void)
+{
+	unsigned int test_num = 0, i;
+	union noise_counter counter;
+	bool success = true;
+
+#define T_INIT do {                                               \
+		memset(&counter, 0, sizeof(union noise_counter)); \
+		spin_lock_init(&counter.receive.lock);            \
+	} while (0)
+#define T_LIM (COUNTER_WINDOW_SIZE + 1)
+#define T(n, v) do {                                                  \
+		++test_num;                                           \
+		if (counter_validate(&counter, n) != (v)) {           \
+			pr_err("nonce counter self-test %u: FAIL\n",  \
+			       test_num);                             \
+			success = false;                              \
+		}                                                     \
+	} while (0)
+
+	T_INIT;
+	/*  1 */ T(0, true);
+	/*  2 */ T(1, true);
+	/*  3 */ T(1, false);
+	/*  4 */ T(9, true);
+	/*  5 */ T(8, true);
+	/*  6 */ T(7, true);
+	/*  7 */ T(7, false);
+	/*  8 */ T(T_LIM, true);
+	/*  9 */ T(T_LIM - 1, true);
+	/* 10 */ T(T_LIM - 1, false);
+	/* 11 */ T(T_LIM - 2, true);
+	/* 12 */ T(2, true);
+	/* 13 */ T(2, false);
+	/* 14 */ T(T_LIM + 16, true);
+	/* 15 */ T(3, false);
+	/* 16 */ T(T_LIM + 16, false);
+	/* 17 */ T(T_LIM * 4, true);
+	/* 18 */ T(T_LIM * 4 - (T_LIM - 1), true);
+	/* 19 */ T(10, false);
+	/* 20 */ T(T_LIM * 4 - T_LIM, false);
+	/* 21 */ T(T_LIM * 4 - (T_LIM + 1), false);
+	/* 22 */ T(T_LIM * 4 - (T_LIM - 2), true);
+	/* 23 */ T(T_LIM * 4 + 1 - T_LIM, false);
+	/* 24 */ T(0, false);
+	/* 25 */ T(REJECT_AFTER_MESSAGES, false);
+	/* 26 */ T(REJECT_AFTER_MESSAGES - 1, true);
+	/* 27 */ T(REJECT_AFTER_MESSAGES, false);
+	/* 28 */ T(REJECT_AFTER_MESSAGES - 1, false);
+	/* 29 */ T(REJECT_AFTER_MESSAGES - 2, true);
+	/* 30 */ T(REJECT_AFTER_MESSAGES + 1, false);
+	/* 31 */ T(REJECT_AFTER_MESSAGES + 2, false);
+	/* 32 */ T(REJECT_AFTER_MESSAGES - 2, false);
+	/* 33 */ T(REJECT_AFTER_MESSAGES - 3, true);
+	/* 34 */ T(0, false);
+
+	T_INIT;
+	for (i = 1; i <= COUNTER_WINDOW_SIZE; ++i)
+		T(i, true);
+	T(0, true);
+	T(0, false);
+
+	T_INIT;
+	for (i = 2; i <= COUNTER_WINDOW_SIZE + 1; ++i)
+		T(i, true);
+	T(1, true);
+	T(0, false);
+
+	T_INIT;
+	for (i = COUNTER_WINDOW_SIZE + 1; i-- > 0;)
+		T(i, true);
+
+	T_INIT;
+	for (i = COUNTER_WINDOW_SIZE + 2; i-- > 1;)
+		T(i, true);
+	T(0, false);
+
+	T_INIT;
+	for (i = COUNTER_WINDOW_SIZE + 1; i-- > 1;)
+		T(i, true);
+	T(COUNTER_WINDOW_SIZE + 1, true);
+	T(0, false);
+
+	T_INIT;
+	for (i = COUNTER_WINDOW_SIZE + 1; i-- > 1;)
+		T(i, true);
+	T(0, true);
+	T(COUNTER_WINDOW_SIZE + 1, true);
+
+#undef T
+#undef T_LIM
+#undef T_INIT
+
+	if (success)
+		pr_info("nonce counter self-tests: pass\n");
+	return success;
+}
+#endif
diff --git a/drivers/net/wireguard/selftest/ratelimiter.c b/drivers/net/wireguard/selftest/ratelimiter.c
new file mode 100644
index 000000000000..bcd6462e4540
--- /dev/null
+++ b/drivers/net/wireguard/selftest/ratelimiter.c
@@ -0,0 +1,226 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#ifdef DEBUG
+
+#include <linux/jiffies.h>
+
+static const struct {
+	bool result;
+	unsigned int msec_to_sleep_before;
+} expected_results[] __initconst = {
+	[0 ... PACKETS_BURSTABLE - 1] = { true, 0 },
+	[PACKETS_BURSTABLE] = { false, 0 },
+	[PACKETS_BURSTABLE + 1] = { true, MSEC_PER_SEC / PACKETS_PER_SECOND },
+	[PACKETS_BURSTABLE + 2] = { false, 0 },
+	[PACKETS_BURSTABLE + 3] = { true, (MSEC_PER_SEC / PACKETS_PER_SECOND) * 2 },
+	[PACKETS_BURSTABLE + 4] = { true, 0 },
+	[PACKETS_BURSTABLE + 5] = { false, 0 }
+};
+
+static __init unsigned int maximum_jiffies_at_index(int index)
+{
+	unsigned int total_msecs = 2 * MSEC_PER_SEC / PACKETS_PER_SECOND / 3;
+	int i;
+
+	for (i = 0; i <= index; ++i)
+		total_msecs += expected_results[i].msec_to_sleep_before;
+	return msecs_to_jiffies(total_msecs);
+}
+
+static __init int timings_test(struct sk_buff *skb4, struct iphdr *hdr4,
+			       struct sk_buff *skb6, struct ipv6hdr *hdr6,
+			       int *test)
+{
+	unsigned long loop_start_time;
+	int i;
+
+	wg_ratelimiter_gc_entries(NULL);
+	rcu_barrier();
+	loop_start_time = jiffies;
+
+	for (i = 0; i < ARRAY_SIZE(expected_results); ++i) {
+		if (expected_results[i].msec_to_sleep_before)
+			msleep(expected_results[i].msec_to_sleep_before);
+
+		if (time_is_before_jiffies(loop_start_time +
+					   maximum_jiffies_at_index(i)))
+			return -ETIMEDOUT;
+		if (wg_ratelimiter_allow(skb4, &init_net) !=
+					expected_results[i].result)
+			return -EXFULL;
+		++(*test);
+
+		hdr4->saddr = htonl(ntohl(hdr4->saddr) + i + 1);
+		if (time_is_before_jiffies(loop_start_time +
+					   maximum_jiffies_at_index(i)))
+			return -ETIMEDOUT;
+		if (!wg_ratelimiter_allow(skb4, &init_net))
+			return -EXFULL;
+		++(*test);
+
+		hdr4->saddr = htonl(ntohl(hdr4->saddr) - i - 1);
+
+#if IS_ENABLED(CONFIG_IPV6)
+		hdr6->saddr.in6_u.u6_addr32[2] = htonl(i);
+		hdr6->saddr.in6_u.u6_addr32[3] = htonl(i);
+		if (time_is_before_jiffies(loop_start_time +
+					   maximum_jiffies_at_index(i)))
+			return -ETIMEDOUT;
+		if (wg_ratelimiter_allow(skb6, &init_net) !=
+					expected_results[i].result)
+			return -EXFULL;
+		++(*test);
+
+		hdr6->saddr.in6_u.u6_addr32[0] =
+			htonl(ntohl(hdr6->saddr.in6_u.u6_addr32[0]) + i + 1);
+		if (time_is_before_jiffies(loop_start_time +
+					   maximum_jiffies_at_index(i)))
+			return -ETIMEDOUT;
+		if (!wg_ratelimiter_allow(skb6, &init_net))
+			return -EXFULL;
+		++(*test);
+
+		hdr6->saddr.in6_u.u6_addr32[0] =
+			htonl(ntohl(hdr6->saddr.in6_u.u6_addr32[0]) - i - 1);
+
+		if (time_is_before_jiffies(loop_start_time +
+					   maximum_jiffies_at_index(i)))
+			return -ETIMEDOUT;
+#endif
+	}
+	return 0;
+}
+
+static __init int capacity_test(struct sk_buff *skb4, struct iphdr *hdr4,
+				int *test)
+{
+	int i;
+
+	wg_ratelimiter_gc_entries(NULL);
+	rcu_barrier();
+
+	if (atomic_read(&total_entries))
+		return -EXFULL;
+	++(*test);
+
+	for (i = 0; i <= max_entries; ++i) {
+		hdr4->saddr = htonl(i);
+		if (wg_ratelimiter_allow(skb4, &init_net) != (i != max_entries))
+			return -EXFULL;
+		++(*test);
+	}
+	return 0;
+}
+
+bool __init wg_ratelimiter_selftest(void)
+{
+	enum { TRIALS_BEFORE_GIVING_UP = 5000 };
+	bool success = false;
+	int test = 0, trials;
+	struct sk_buff *skb4, *skb6;
+	struct iphdr *hdr4;
+	struct ipv6hdr *hdr6;
+
+	if (IS_ENABLED(CONFIG_KASAN) || IS_ENABLED(CONFIG_UBSAN))
+		return true;
+
+	BUILD_BUG_ON(MSEC_PER_SEC % PACKETS_PER_SECOND != 0);
+
+	if (wg_ratelimiter_init())
+		goto out;
+	++test;
+	if (wg_ratelimiter_init()) {
+		wg_ratelimiter_uninit();
+		goto out;
+	}
+	++test;
+	if (wg_ratelimiter_init()) {
+		wg_ratelimiter_uninit();
+		wg_ratelimiter_uninit();
+		goto out;
+	}
+	++test;
+
+	skb4 = alloc_skb(sizeof(struct iphdr), GFP_KERNEL);
+	if (unlikely(!skb4))
+		goto err_nofree;
+	skb4->protocol = htons(ETH_P_IP);
+	hdr4 = (struct iphdr *)skb_put(skb4, sizeof(*hdr4));
+	hdr4->saddr = htonl(8182);
+	skb_reset_network_header(skb4);
+	++test;
+
+#if IS_ENABLED(CONFIG_IPV6)
+	skb6 = alloc_skb(sizeof(struct ipv6hdr), GFP_KERNEL);
+	if (unlikely(!skb6)) {
+		kfree_skb(skb4);
+		goto err_nofree;
+	}
+	skb6->protocol = htons(ETH_P_IPV6);
+	hdr6 = (struct ipv6hdr *)skb_put(skb6, sizeof(*hdr6));
+	hdr6->saddr.in6_u.u6_addr32[0] = htonl(1212);
+	hdr6->saddr.in6_u.u6_addr32[1] = htonl(289188);
+	skb_reset_network_header(skb6);
+	++test;
+#endif
+
+	for (trials = TRIALS_BEFORE_GIVING_UP;;) {
+		int test_count = 0, ret;
+
+		ret = timings_test(skb4, hdr4, skb6, hdr6, &test_count);
+		if (ret == -ETIMEDOUT) {
+			if (!trials--) {
+				test += test_count;
+				goto err;
+			}
+			msleep(500);
+			continue;
+		} else if (ret < 0) {
+			test += test_count;
+			goto err;
+		} else {
+			test += test_count;
+			break;
+		}
+	}
+
+	for (trials = TRIALS_BEFORE_GIVING_UP;;) {
+		int test_count = 0;
+
+		if (capacity_test(skb4, hdr4, &test_count) < 0) {
+			if (!trials--) {
+				test += test_count;
+				goto err;
+			}
+			msleep(50);
+			continue;
+		}
+		test += test_count;
+		break;
+	}
+
+	success = true;
+
+err:
+	kfree_skb(skb4);
+#if IS_ENABLED(CONFIG_IPV6)
+	kfree_skb(skb6);
+#endif
+err_nofree:
+	wg_ratelimiter_uninit();
+	wg_ratelimiter_uninit();
+	wg_ratelimiter_uninit();
+	/* Uninit one extra time to check underflow detection. */
+	wg_ratelimiter_uninit();
+out:
+	if (success)
+		pr_info("ratelimiter self-tests: pass\n");
+	else
+		pr_err("ratelimiter self-test %d: FAIL\n", test);
+
+	return success;
+}
+#endif
diff --git a/drivers/net/wireguard/send.c b/drivers/net/wireguard/send.c
new file mode 100644
index 000000000000..c13260563446
--- /dev/null
+++ b/drivers/net/wireguard/send.c
@@ -0,0 +1,413 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#include "queueing.h"
+#include "timers.h"
+#include "device.h"
+#include "peer.h"
+#include "socket.h"
+#include "messages.h"
+#include "cookie.h"
+
+#include <linux/uio.h>
+#include <linux/inetdevice.h>
+#include <linux/socket.h>
+#include <net/ip_tunnels.h>
+#include <net/udp.h>
+#include <net/sock.h>
+
+static void wg_packet_send_handshake_initiation(struct wg_peer *peer)
+{
+	struct message_handshake_initiation packet;
+
+	if (!wg_birthdate_has_expired(atomic64_read(&peer->last_sent_handshake),
+				      REKEY_TIMEOUT))
+		return; /* This function is rate limited. */
+
+	atomic64_set(&peer->last_sent_handshake, ktime_get_coarse_boottime_ns());
+	net_dbg_ratelimited("%s: Sending handshake initiation to peer %llu (%pISpfsc)\n",
+			    peer->device->dev->name, peer->internal_id,
+			    &peer->endpoint.addr);
+
+	if (wg_noise_handshake_create_initiation(&packet, &peer->handshake)) {
+		wg_cookie_add_mac_to_packet(&packet, sizeof(packet), peer);
+		wg_timers_any_authenticated_packet_traversal(peer);
+		wg_timers_any_authenticated_packet_sent(peer);
+		atomic64_set(&peer->last_sent_handshake,
+			     ktime_get_coarse_boottime_ns());
+		wg_socket_send_buffer_to_peer(peer, &packet, sizeof(packet),
+					      HANDSHAKE_DSCP);
+		wg_timers_handshake_initiated(peer);
+	}
+}
+
+void wg_packet_handshake_send_worker(struct work_struct *work)
+{
+	struct wg_peer *peer = container_of(work, struct wg_peer,
+					    transmit_handshake_work);
+
+	wg_packet_send_handshake_initiation(peer);
+	wg_peer_put(peer);
+}
+
+void wg_packet_send_queued_handshake_initiation(struct wg_peer *peer,
+						bool is_retry)
+{
+	if (!is_retry)
+		peer->timer_handshake_attempts = 0;
+
+	rcu_read_lock_bh();
+	/* We check last_sent_handshake here in addition to the actual function
+	 * we're queueing up, so that we don't queue things if not strictly
+	 * necessary:
+	 */
+	if (!wg_birthdate_has_expired(atomic64_read(&peer->last_sent_handshake),
+				      REKEY_TIMEOUT) ||
+			unlikely(READ_ONCE(peer->is_dead)))
+		goto out;
+
+	wg_peer_get(peer);
+	/* Queues up calling packet_send_queued_handshakes(peer), where we do a
+	 * peer_put(peer) after:
+	 */
+	if (!queue_work(peer->device->handshake_send_wq,
+			&peer->transmit_handshake_work))
+		/* If the work was already queued, we want to drop the
+		 * extra reference:
+		 */
+		wg_peer_put(peer);
+out:
+	rcu_read_unlock_bh();
+}
+
+void wg_packet_send_handshake_response(struct wg_peer *peer)
+{
+	struct message_handshake_response packet;
+
+	atomic64_set(&peer->last_sent_handshake, ktime_get_coarse_boottime_ns());
+	net_dbg_ratelimited("%s: Sending handshake response to peer %llu (%pISpfsc)\n",
+			    peer->device->dev->name, peer->internal_id,
+			    &peer->endpoint.addr);
+
+	if (wg_noise_handshake_create_response(&packet, &peer->handshake)) {
+		wg_cookie_add_mac_to_packet(&packet, sizeof(packet), peer);
+		if (wg_noise_handshake_begin_session(&peer->handshake,
+						     &peer->keypairs)) {
+			wg_timers_session_derived(peer);
+			wg_timers_any_authenticated_packet_traversal(peer);
+			wg_timers_any_authenticated_packet_sent(peer);
+			atomic64_set(&peer->last_sent_handshake,
+				     ktime_get_coarse_boottime_ns());
+			wg_socket_send_buffer_to_peer(peer, &packet,
+						      sizeof(packet),
+						      HANDSHAKE_DSCP);
+		}
+	}
+}
+
+void wg_packet_send_handshake_cookie(struct wg_device *wg,
+				     struct sk_buff *initiating_skb,
+				     __le32 sender_index)
+{
+	struct message_handshake_cookie packet;
+
+	net_dbg_skb_ratelimited("%s: Sending cookie response for denied handshake message for %pISpfsc\n",
+				wg->dev->name, initiating_skb);
+	wg_cookie_message_create(&packet, initiating_skb, sender_index,
+				 &wg->cookie_checker);
+	wg_socket_send_buffer_as_reply_to_skb(wg, initiating_skb, &packet,
+					      sizeof(packet));
+}
+
+static void keep_key_fresh(struct wg_peer *peer)
+{
+	struct noise_keypair *keypair;
+	bool send = false;
+
+	rcu_read_lock_bh();
+	keypair = rcu_dereference_bh(peer->keypairs.current_keypair);
+	if (likely(keypair && READ_ONCE(keypair->sending.is_valid)) &&
+	    (unlikely(atomic64_read(&keypair->sending.counter.counter) >
+		      REKEY_AFTER_MESSAGES) ||
+	     (keypair->i_am_the_initiator &&
+	      unlikely(wg_birthdate_has_expired(keypair->sending.birthdate,
+						REKEY_AFTER_TIME)))))
+		send = true;
+	rcu_read_unlock_bh();
+
+	if (send)
+		wg_packet_send_queued_handshake_initiation(peer, false);
+}
+
+static unsigned int calculate_skb_padding(struct sk_buff *skb)
+{
+	/* We do this modulo business with the MTU, just in case the networking
+	 * layer gives us a packet that's bigger than the MTU. In that case, we
+	 * wouldn't want the final subtraction to overflow in the case of the
+	 * padded_size being clamped.
+	 */
+	unsigned int last_unit = skb->len % PACKET_CB(skb)->mtu;
+	unsigned int padded_size = ALIGN(last_unit, MESSAGE_PADDING_MULTIPLE);
+
+	if (padded_size > PACKET_CB(skb)->mtu)
+		padded_size = PACKET_CB(skb)->mtu;
+	return padded_size - last_unit;
+}
+
+static bool encrypt_packet(struct sk_buff *skb, struct noise_keypair *keypair)
+{
+	unsigned int padding_len, plaintext_len, trailer_len;
+	struct scatterlist sg[MAX_SKB_FRAGS + 8];
+	struct message_data *header;
+	struct sk_buff *trailer;
+	int num_frags;
+
+	/* Calculate lengths. */
+	padding_len = calculate_skb_padding(skb);
+	trailer_len = padding_len + noise_encrypted_len(0);
+	plaintext_len = skb->len + padding_len;
+
+	/* Expand data section to have room for padding and auth tag. */
+	num_frags = skb_cow_data(skb, trailer_len, &trailer);
+	if (unlikely(num_frags < 0 || num_frags > ARRAY_SIZE(sg)))
+		return false;
+
+	/* Set the padding to zeros, and make sure it and the auth tag are part
+	 * of the skb.
+	 */
+	memset(skb_tail_pointer(trailer), 0, padding_len);
+
+	/* Expand head section to have room for our header and the network
+	 * stack's headers.
+	 */
+	if (unlikely(skb_cow_head(skb, DATA_PACKET_HEAD_ROOM) < 0))
+		return false;
+
+	/* Finalize checksum calculation for the inner packet, if required. */
+	if (unlikely(skb->ip_summed == CHECKSUM_PARTIAL &&
+		     skb_checksum_help(skb)))
+		return false;
+
+	/* Only after checksumming can we safely add on the padding at the end
+	 * and the header.
+	 */
+	skb_set_inner_network_header(skb, 0);
+	header = (struct message_data *)skb_push(skb, sizeof(*header));
+	header->header.type = cpu_to_le32(MESSAGE_DATA);
+	header->key_idx = keypair->remote_index;
+	header->counter = cpu_to_le64(PACKET_CB(skb)->nonce);
+	pskb_put(skb, trailer, trailer_len);
+
+	/* Now we can encrypt the scattergather segments */
+	sg_init_table(sg, num_frags);
+	if (skb_to_sgvec(skb, sg, sizeof(struct message_data),
+			 noise_encrypted_len(plaintext_len)) <= 0)
+		return false;
+	return chacha20poly1305_encrypt_sg_inplace(sg, plaintext_len, NULL, 0,
+						   PACKET_CB(skb)->nonce,
+						   keypair->sending.key);
+}
+
+void wg_packet_send_keepalive(struct wg_peer *peer)
+{
+	struct sk_buff *skb;
+
+	if (skb_queue_empty(&peer->staged_packet_queue)) {
+		skb = alloc_skb(DATA_PACKET_HEAD_ROOM + MESSAGE_MINIMUM_LENGTH,
+				GFP_ATOMIC);
+		if (unlikely(!skb))
+			return;
+		skb_reserve(skb, DATA_PACKET_HEAD_ROOM);
+		skb->dev = peer->device->dev;
+		PACKET_CB(skb)->mtu = skb->dev->mtu;
+		skb_queue_tail(&peer->staged_packet_queue, skb);
+		net_dbg_ratelimited("%s: Sending keepalive packet to peer %llu (%pISpfsc)\n",
+				    peer->device->dev->name, peer->internal_id,
+				    &peer->endpoint.addr);
+	}
+
+	wg_packet_send_staged_packets(peer);
+}
+
+static void wg_packet_create_data_done(struct sk_buff *first,
+				       struct wg_peer *peer)
+{
+	struct sk_buff *skb, *next;
+	bool is_keepalive, data_sent = false;
+
+	wg_timers_any_authenticated_packet_traversal(peer);
+	wg_timers_any_authenticated_packet_sent(peer);
+	skb_list_walk_safe(first, skb, next) {
+		is_keepalive = skb->len == message_data_len(0);
+		if (likely(!wg_socket_send_skb_to_peer(peer, skb,
+				PACKET_CB(skb)->ds) && !is_keepalive))
+			data_sent = true;
+	}
+
+	if (likely(data_sent))
+		wg_timers_data_sent(peer);
+
+	keep_key_fresh(peer);
+}
+
+void wg_packet_tx_worker(struct work_struct *work)
+{
+	struct crypt_queue *queue = container_of(work, struct crypt_queue,
+						 work);
+	struct noise_keypair *keypair;
+	enum packet_state state;
+	struct sk_buff *first;
+	struct wg_peer *peer;
+
+	while ((first = __ptr_ring_peek(&queue->ring)) != NULL &&
+	       (state = atomic_read_acquire(&PACKET_CB(first)->state)) !=
+		       PACKET_STATE_UNCRYPTED) {
+		__ptr_ring_discard_one(&queue->ring);
+		peer = PACKET_PEER(first);
+		keypair = PACKET_CB(first)->keypair;
+
+		if (likely(state == PACKET_STATE_CRYPTED))
+			wg_packet_create_data_done(first, peer);
+		else
+			kfree_skb_list(first);
+
+		wg_noise_keypair_put(keypair, false);
+		wg_peer_put(peer);
+	}
+}
+
+void wg_packet_encrypt_worker(struct work_struct *work)
+{
+	struct crypt_queue *queue = container_of(work, struct multicore_worker,
+						 work)->ptr;
+	struct sk_buff *first, *skb, *next;
+
+	while ((first = ptr_ring_consume_bh(&queue->ring)) != NULL) {
+		enum packet_state state = PACKET_STATE_CRYPTED;
+
+		skb_list_walk_safe(first, skb, next) {
+			if (likely(encrypt_packet(skb,
+					PACKET_CB(first)->keypair))) {
+				wg_reset_packet(skb);
+			} else {
+				state = PACKET_STATE_DEAD;
+				break;
+			}
+		}
+		wg_queue_enqueue_per_peer(&PACKET_PEER(first)->tx_queue, first,
+					  state);
+
+	}
+}
+
+static void wg_packet_create_data(struct sk_buff *first)
+{
+	struct wg_peer *peer = PACKET_PEER(first);
+	struct wg_device *wg = peer->device;
+	int ret = -EINVAL;
+
+	rcu_read_lock_bh();
+	if (unlikely(READ_ONCE(peer->is_dead)))
+		goto err;
+
+	ret = wg_queue_enqueue_per_device_and_peer(&wg->encrypt_queue,
+						   &peer->tx_queue, first,
+						   wg->packet_crypt_wq,
+						   &wg->encrypt_queue.last_cpu);
+	if (unlikely(ret == -EPIPE))
+		wg_queue_enqueue_per_peer(&peer->tx_queue, first,
+					  PACKET_STATE_DEAD);
+err:
+	rcu_read_unlock_bh();
+	if (likely(!ret || ret == -EPIPE))
+		return;
+	wg_noise_keypair_put(PACKET_CB(first)->keypair, false);
+	wg_peer_put(peer);
+	kfree_skb_list(first);
+}
+
+void wg_packet_purge_staged_packets(struct wg_peer *peer)
+{
+	spin_lock_bh(&peer->staged_packet_queue.lock);
+	peer->device->dev->stats.tx_dropped += peer->staged_packet_queue.qlen;
+	__skb_queue_purge(&peer->staged_packet_queue);
+	spin_unlock_bh(&peer->staged_packet_queue.lock);
+}
+
+void wg_packet_send_staged_packets(struct wg_peer *peer)
+{
+	struct noise_symmetric_key *key;
+	struct noise_keypair *keypair;
+	struct sk_buff_head packets;
+	struct sk_buff *skb;
+
+	/* Steal the current queue into our local one. */
+	__skb_queue_head_init(&packets);
+	spin_lock_bh(&peer->staged_packet_queue.lock);
+	skb_queue_splice_init(&peer->staged_packet_queue, &packets);
+	spin_unlock_bh(&peer->staged_packet_queue.lock);
+	if (unlikely(skb_queue_empty(&packets)))
+		return;
+
+	/* First we make sure we have a valid reference to a valid key. */
+	rcu_read_lock_bh();
+	keypair = wg_noise_keypair_get(
+		rcu_dereference_bh(peer->keypairs.current_keypair));
+	rcu_read_unlock_bh();
+	if (unlikely(!keypair))
+		goto out_nokey;
+	key = &keypair->sending;
+	if (unlikely(!READ_ONCE(key->is_valid)))
+		goto out_nokey;
+	if (unlikely(wg_birthdate_has_expired(key->birthdate,
+					      REJECT_AFTER_TIME)))
+		goto out_invalid;
+
+	/* After we know we have a somewhat valid key, we now try to assign
+	 * nonces to all of the packets in the queue. If we can't assign nonces
+	 * for all of them, we just consider it a failure and wait for the next
+	 * handshake.
+	 */
+	skb_queue_walk(&packets, skb) {
+		/* 0 for no outer TOS: no leak. TODO: at some later point, we
+		 * might consider using flowi->tos as outer instead.
+		 */
+		PACKET_CB(skb)->ds = ip_tunnel_ecn_encap(0, ip_hdr(skb), skb);
+		PACKET_CB(skb)->nonce =
+				atomic64_inc_return(&key->counter.counter) - 1;
+		if (unlikely(PACKET_CB(skb)->nonce >= REJECT_AFTER_MESSAGES))
+			goto out_invalid;
+	}
+
+	packets.prev->next = NULL;
+	wg_peer_get(keypair->entry.peer);
+	PACKET_CB(packets.next)->keypair = keypair;
+	wg_packet_create_data(packets.next);
+	return;
+
+out_invalid:
+	WRITE_ONCE(key->is_valid, false);
+out_nokey:
+	wg_noise_keypair_put(keypair, false);
+
+	/* We orphan the packets if we're waiting on a handshake, so that they
+	 * don't block a socket's pool.
+	 */
+	skb_queue_walk(&packets, skb)
+		skb_orphan(skb);
+	/* Then we put them back on the top of the queue. We're not too
+	 * concerned about accidentally getting things a little out of order if
+	 * packets are being added really fast, because this queue is for before
+	 * packets can even be sent and it's small anyway.
+	 */
+	spin_lock_bh(&peer->staged_packet_queue.lock);
+	skb_queue_splice(&packets, &peer->staged_packet_queue);
+	spin_unlock_bh(&peer->staged_packet_queue.lock);
+
+	/* If we're exiting because there's something wrong with the key, it
+	 * means we should initiate a new handshake.
+	 */
+	wg_packet_send_queued_handshake_initiation(peer, false);
+}
diff --git a/drivers/net/wireguard/socket.c b/drivers/net/wireguard/socket.c
new file mode 100644
index 000000000000..c46256d0d81c
--- /dev/null
+++ b/drivers/net/wireguard/socket.c
@@ -0,0 +1,437 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#include "device.h"
+#include "peer.h"
+#include "socket.h"
+#include "queueing.h"
+#include "messages.h"
+
+#include <linux/ctype.h>
+#include <linux/net.h>
+#include <linux/if_vlan.h>
+#include <linux/if_ether.h>
+#include <linux/inetdevice.h>
+#include <net/udp_tunnel.h>
+#include <net/ipv6.h>
+
+static int send4(struct wg_device *wg, struct sk_buff *skb,
+		 struct endpoint *endpoint, u8 ds, struct dst_cache *cache)
+{
+	struct flowi4 fl = {
+		.saddr = endpoint->src4.s_addr,
+		.daddr = endpoint->addr4.sin_addr.s_addr,
+		.fl4_dport = endpoint->addr4.sin_port,
+		.flowi4_mark = wg->fwmark,
+		.flowi4_proto = IPPROTO_UDP
+	};
+	struct rtable *rt = NULL;
+	struct sock *sock;
+	int ret = 0;
+
+	skb_mark_not_on_list(skb);
+	skb->dev = wg->dev;
+	skb->mark = wg->fwmark;
+
+	rcu_read_lock_bh();
+	sock = rcu_dereference_bh(wg->sock4);
+
+	if (unlikely(!sock)) {
+		ret = -ENONET;
+		goto err;
+	}
+
+	fl.fl4_sport = inet_sk(sock)->inet_sport;
+
+	if (cache)
+		rt = dst_cache_get_ip4(cache, &fl.saddr);
+
+	if (!rt) {
+		security_sk_classify_flow(sock, flowi4_to_flowi(&fl));
+		if (unlikely(!inet_confirm_addr(sock_net(sock), NULL, 0,
+						fl.saddr, RT_SCOPE_HOST))) {
+			endpoint->src4.s_addr = 0;
+			*(__force __be32 *)&endpoint->src_if4 = 0;
+			fl.saddr = 0;
+			if (cache)
+				dst_cache_reset(cache);
+		}
+		rt = ip_route_output_flow(sock_net(sock), &fl, sock);
+		if (unlikely(endpoint->src_if4 && ((IS_ERR(rt) &&
+			     PTR_ERR(rt) == -EINVAL) || (!IS_ERR(rt) &&
+			     rt->dst.dev->ifindex != endpoint->src_if4)))) {
+			endpoint->src4.s_addr = 0;
+			*(__force __be32 *)&endpoint->src_if4 = 0;
+			fl.saddr = 0;
+			if (cache)
+				dst_cache_reset(cache);
+			if (!IS_ERR(rt))
+				ip_rt_put(rt);
+			rt = ip_route_output_flow(sock_net(sock), &fl, sock);
+		}
+		if (unlikely(IS_ERR(rt))) {
+			ret = PTR_ERR(rt);
+			net_dbg_ratelimited("%s: No route to %pISpfsc, error %d\n",
+					    wg->dev->name, &endpoint->addr, ret);
+			goto err;
+		} else if (unlikely(rt->dst.dev == skb->dev)) {
+			ip_rt_put(rt);
+			ret = -ELOOP;
+			net_dbg_ratelimited("%s: Avoiding routing loop to %pISpfsc\n",
+					    wg->dev->name, &endpoint->addr);
+			goto err;
+		}
+		if (cache)
+			dst_cache_set_ip4(cache, &rt->dst, fl.saddr);
+	}
+
+	skb->ignore_df = 1;
+	udp_tunnel_xmit_skb(rt, sock, skb, fl.saddr, fl.daddr, ds,
+			    ip4_dst_hoplimit(&rt->dst), 0, fl.fl4_sport,
+			    fl.fl4_dport, false, false);
+	goto out;
+
+err:
+	kfree_skb(skb);
+out:
+	rcu_read_unlock_bh();
+	return ret;
+}
+
+static int send6(struct wg_device *wg, struct sk_buff *skb,
+		 struct endpoint *endpoint, u8 ds, struct dst_cache *cache)
+{
+#if IS_ENABLED(CONFIG_IPV6)
+	struct flowi6 fl = {
+		.saddr = endpoint->src6,
+		.daddr = endpoint->addr6.sin6_addr,
+		.fl6_dport = endpoint->addr6.sin6_port,
+		.flowi6_mark = wg->fwmark,
+		.flowi6_oif = endpoint->addr6.sin6_scope_id,
+		.flowi6_proto = IPPROTO_UDP
+		/* TODO: addr->sin6_flowinfo */
+	};
+	struct dst_entry *dst = NULL;
+	struct sock *sock;
+	int ret = 0;
+
+	skb_mark_not_on_list(skb);
+	skb->dev = wg->dev;
+	skb->mark = wg->fwmark;
+
+	rcu_read_lock_bh();
+	sock = rcu_dereference_bh(wg->sock6);
+
+	if (unlikely(!sock)) {
+		ret = -ENONET;
+		goto err;
+	}
+
+	fl.fl6_sport = inet_sk(sock)->inet_sport;
+
+	if (cache)
+		dst = dst_cache_get_ip6(cache, &fl.saddr);
+
+	if (!dst) {
+		security_sk_classify_flow(sock, flowi6_to_flowi(&fl));
+		if (unlikely(!ipv6_addr_any(&fl.saddr) &&
+			     !ipv6_chk_addr(sock_net(sock), &fl.saddr, NULL, 0))) {
+			endpoint->src6 = fl.saddr = in6addr_any;
+			if (cache)
+				dst_cache_reset(cache);
+		}
+		dst = ipv6_stub->ipv6_dst_lookup_flow(sock_net(sock), sock, &fl,
+						      NULL);
+		if (unlikely(IS_ERR(dst))) {
+			ret = PTR_ERR(dst);
+			net_dbg_ratelimited("%s: No route to %pISpfsc, error %d\n",
+					    wg->dev->name, &endpoint->addr, ret);
+			goto err;
+		} else if (unlikely(dst->dev == skb->dev)) {
+			dst_release(dst);
+			ret = -ELOOP;
+			net_dbg_ratelimited("%s: Avoiding routing loop to %pISpfsc\n",
+					    wg->dev->name, &endpoint->addr);
+			goto err;
+		}
+		if (cache)
+			dst_cache_set_ip6(cache, dst, &fl.saddr);
+	}
+
+	skb->ignore_df = 1;
+	udp_tunnel6_xmit_skb(dst, sock, skb, skb->dev, &fl.saddr, &fl.daddr, ds,
+			     ip6_dst_hoplimit(dst), 0, fl.fl6_sport,
+			     fl.fl6_dport, false);
+	goto out;
+
+err:
+	kfree_skb(skb);
+out:
+	rcu_read_unlock_bh();
+	return ret;
+#else
+	return -EAFNOSUPPORT;
+#endif
+}
+
+int wg_socket_send_skb_to_peer(struct wg_peer *peer, struct sk_buff *skb, u8 ds)
+{
+	size_t skb_len = skb->len;
+	int ret = -EAFNOSUPPORT;
+
+	read_lock_bh(&peer->endpoint_lock);
+	if (peer->endpoint.addr.sa_family == AF_INET)
+		ret = send4(peer->device, skb, &peer->endpoint, ds,
+			    &peer->endpoint_cache);
+	else if (peer->endpoint.addr.sa_family == AF_INET6)
+		ret = send6(peer->device, skb, &peer->endpoint, ds,
+			    &peer->endpoint_cache);
+	else
+		dev_kfree_skb(skb);
+	if (likely(!ret))
+		peer->tx_bytes += skb_len;
+	read_unlock_bh(&peer->endpoint_lock);
+
+	return ret;
+}
+
+int wg_socket_send_buffer_to_peer(struct wg_peer *peer, void *buffer,
+				  size_t len, u8 ds)
+{
+	struct sk_buff *skb = alloc_skb(len + SKB_HEADER_LEN, GFP_ATOMIC);
+
+	if (unlikely(!skb))
+		return -ENOMEM;
+
+	skb_reserve(skb, SKB_HEADER_LEN);
+	skb_set_inner_network_header(skb, 0);
+	skb_put_data(skb, buffer, len);
+	return wg_socket_send_skb_to_peer(peer, skb, ds);
+}
+
+int wg_socket_send_buffer_as_reply_to_skb(struct wg_device *wg,
+					  struct sk_buff *in_skb, void *buffer,
+					  size_t len)
+{
+	int ret = 0;
+	struct sk_buff *skb;
+	struct endpoint endpoint;
+
+	if (unlikely(!in_skb))
+		return -EINVAL;
+	ret = wg_socket_endpoint_from_skb(&endpoint, in_skb);
+	if (unlikely(ret < 0))
+		return ret;
+
+	skb = alloc_skb(len + SKB_HEADER_LEN, GFP_ATOMIC);
+	if (unlikely(!skb))
+		return -ENOMEM;
+	skb_reserve(skb, SKB_HEADER_LEN);
+	skb_set_inner_network_header(skb, 0);
+	skb_put_data(skb, buffer, len);
+
+	if (endpoint.addr.sa_family == AF_INET)
+		ret = send4(wg, skb, &endpoint, 0, NULL);
+	else if (endpoint.addr.sa_family == AF_INET6)
+		ret = send6(wg, skb, &endpoint, 0, NULL);
+	/* No other possibilities if the endpoint is valid, which it is,
+	 * as we checked above.
+	 */
+
+	return ret;
+}
+
+int wg_socket_endpoint_from_skb(struct endpoint *endpoint,
+				const struct sk_buff *skb)
+{
+	memset(endpoint, 0, sizeof(*endpoint));
+	if (skb->protocol == htons(ETH_P_IP)) {
+		endpoint->addr4.sin_family = AF_INET;
+		endpoint->addr4.sin_port = udp_hdr(skb)->source;
+		endpoint->addr4.sin_addr.s_addr = ip_hdr(skb)->saddr;
+		endpoint->src4.s_addr = ip_hdr(skb)->daddr;
+		endpoint->src_if4 = skb->skb_iif;
+	} else if (skb->protocol == htons(ETH_P_IPV6)) {
+		endpoint->addr6.sin6_family = AF_INET6;
+		endpoint->addr6.sin6_port = udp_hdr(skb)->source;
+		endpoint->addr6.sin6_addr = ipv6_hdr(skb)->saddr;
+		endpoint->addr6.sin6_scope_id = ipv6_iface_scope_id(
+			&ipv6_hdr(skb)->saddr, skb->skb_iif);
+		endpoint->src6 = ipv6_hdr(skb)->daddr;
+	} else {
+		return -EINVAL;
+	}
+	return 0;
+}
+
+static bool endpoint_eq(const struct endpoint *a, const struct endpoint *b)
+{
+	return (a->addr.sa_family == AF_INET && b->addr.sa_family == AF_INET &&
+		a->addr4.sin_port == b->addr4.sin_port &&
+		a->addr4.sin_addr.s_addr == b->addr4.sin_addr.s_addr &&
+		a->src4.s_addr == b->src4.s_addr && a->src_if4 == b->src_if4) ||
+	       (a->addr.sa_family == AF_INET6 &&
+		b->addr.sa_family == AF_INET6 &&
+		a->addr6.sin6_port == b->addr6.sin6_port &&
+		ipv6_addr_equal(&a->addr6.sin6_addr, &b->addr6.sin6_addr) &&
+		a->addr6.sin6_scope_id == b->addr6.sin6_scope_id &&
+		ipv6_addr_equal(&a->src6, &b->src6)) ||
+	       unlikely(!a->addr.sa_family && !b->addr.sa_family);
+}
+
+void wg_socket_set_peer_endpoint(struct wg_peer *peer,
+				 const struct endpoint *endpoint)
+{
+	/* First we check unlocked, in order to optimize, since it's pretty rare
+	 * that an endpoint will change. If we happen to be mid-write, and two
+	 * CPUs wind up writing the same thing or something slightly different,
+	 * it doesn't really matter much either.
+	 */
+	if (endpoint_eq(endpoint, &peer->endpoint))
+		return;
+	write_lock_bh(&peer->endpoint_lock);
+	if (endpoint->addr.sa_family == AF_INET) {
+		peer->endpoint.addr4 = endpoint->addr4;
+		peer->endpoint.src4 = endpoint->src4;
+		peer->endpoint.src_if4 = endpoint->src_if4;
+	} else if (endpoint->addr.sa_family == AF_INET6) {
+		peer->endpoint.addr6 = endpoint->addr6;
+		peer->endpoint.src6 = endpoint->src6;
+	} else {
+		goto out;
+	}
+	dst_cache_reset(&peer->endpoint_cache);
+out:
+	write_unlock_bh(&peer->endpoint_lock);
+}
+
+void wg_socket_set_peer_endpoint_from_skb(struct wg_peer *peer,
+					  const struct sk_buff *skb)
+{
+	struct endpoint endpoint;
+
+	if (!wg_socket_endpoint_from_skb(&endpoint, skb))
+		wg_socket_set_peer_endpoint(peer, &endpoint);
+}
+
+void wg_socket_clear_peer_endpoint_src(struct wg_peer *peer)
+{
+	write_lock_bh(&peer->endpoint_lock);
+	memset(&peer->endpoint.src6, 0, sizeof(peer->endpoint.src6));
+	dst_cache_reset(&peer->endpoint_cache);
+	write_unlock_bh(&peer->endpoint_lock);
+}
+
+static int wg_receive(struct sock *sk, struct sk_buff *skb)
+{
+	struct wg_device *wg;
+
+	if (unlikely(!sk))
+		goto err;
+	wg = sk->sk_user_data;
+	if (unlikely(!wg))
+		goto err;
+	wg_packet_receive(wg, skb);
+	return 0;
+
+err:
+	kfree_skb(skb);
+	return 0;
+}
+
+static void sock_free(struct sock *sock)
+{
+	if (unlikely(!sock))
+		return;
+	sk_clear_memalloc(sock);
+	udp_tunnel_sock_release(sock->sk_socket);
+}
+
+static void set_sock_opts(struct socket *sock)
+{
+	sock->sk->sk_allocation = GFP_ATOMIC;
+	sock->sk->sk_sndbuf = INT_MAX;
+	sk_set_memalloc(sock->sk);
+}
+
+int wg_socket_init(struct wg_device *wg, u16 port)
+{
+	int ret;
+	struct udp_tunnel_sock_cfg cfg = {
+		.sk_user_data = wg,
+		.encap_type = 1,
+		.encap_rcv = wg_receive
+	};
+	struct socket *new4 = NULL, *new6 = NULL;
+	struct udp_port_cfg port4 = {
+		.family = AF_INET,
+		.local_ip.s_addr = htonl(INADDR_ANY),
+		.local_udp_port = htons(port),
+		.use_udp_checksums = true
+	};
+#if IS_ENABLED(CONFIG_IPV6)
+	int retries = 0;
+	struct udp_port_cfg port6 = {
+		.family = AF_INET6,
+		.local_ip6 = IN6ADDR_ANY_INIT,
+		.use_udp6_tx_checksums = true,
+		.use_udp6_rx_checksums = true,
+		.ipv6_v6only = true
+	};
+#endif
+
+#if IS_ENABLED(CONFIG_IPV6)
+retry:
+#endif
+
+	ret = udp_sock_create(wg->creating_net, &port4, &new4);
+	if (ret < 0) {
+		pr_err("%s: Could not create IPv4 socket\n", wg->dev->name);
+		return ret;
+	}
+	set_sock_opts(new4);
+	setup_udp_tunnel_sock(wg->creating_net, new4, &cfg);
+
+#if IS_ENABLED(CONFIG_IPV6)
+	if (ipv6_mod_enabled()) {
+		port6.local_udp_port = inet_sk(new4->sk)->inet_sport;
+		ret = udp_sock_create(wg->creating_net, &port6, &new6);
+		if (ret < 0) {
+			udp_tunnel_sock_release(new4);
+			if (ret == -EADDRINUSE && !port && retries++ < 100)
+				goto retry;
+			pr_err("%s: Could not create IPv6 socket\n",
+			       wg->dev->name);
+			return ret;
+		}
+		set_sock_opts(new6);
+		setup_udp_tunnel_sock(wg->creating_net, new6, &cfg);
+	}
+#endif
+
+	wg_socket_reinit(wg, new4->sk, new6 ? new6->sk : NULL);
+	return 0;
+}
+
+void wg_socket_reinit(struct wg_device *wg, struct sock *new4,
+		      struct sock *new6)
+{
+	struct sock *old4, *old6;
+
+	mutex_lock(&wg->socket_update_lock);
+	old4 = rcu_dereference_protected(wg->sock4,
+				lockdep_is_held(&wg->socket_update_lock));
+	old6 = rcu_dereference_protected(wg->sock6,
+				lockdep_is_held(&wg->socket_update_lock));
+	rcu_assign_pointer(wg->sock4, new4);
+	rcu_assign_pointer(wg->sock6, new6);
+	if (new4)
+		wg->incoming_port = ntohs(inet_sk(new4)->inet_sport);
+	mutex_unlock(&wg->socket_update_lock);
+	synchronize_rcu();
+	synchronize_net();
+	sock_free(old4);
+	sock_free(old6);
+}
diff --git a/drivers/net/wireguard/socket.h b/drivers/net/wireguard/socket.h
new file mode 100644
index 000000000000..bab5848efbcd
--- /dev/null
+++ b/drivers/net/wireguard/socket.h
@@ -0,0 +1,44 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#ifndef _WG_SOCKET_H
+#define _WG_SOCKET_H
+
+#include <linux/netdevice.h>
+#include <linux/udp.h>
+#include <linux/if_vlan.h>
+#include <linux/if_ether.h>
+
+int wg_socket_init(struct wg_device *wg, u16 port);
+void wg_socket_reinit(struct wg_device *wg, struct sock *new4,
+		      struct sock *new6);
+int wg_socket_send_buffer_to_peer(struct wg_peer *peer, void *data,
+				  size_t len, u8 ds);
+int wg_socket_send_skb_to_peer(struct wg_peer *peer, struct sk_buff *skb,
+			       u8 ds);
+int wg_socket_send_buffer_as_reply_to_skb(struct wg_device *wg,
+					  struct sk_buff *in_skb,
+					  void *out_buffer, size_t len);
+
+int wg_socket_endpoint_from_skb(struct endpoint *endpoint,
+				const struct sk_buff *skb);
+void wg_socket_set_peer_endpoint(struct wg_peer *peer,
+				 const struct endpoint *endpoint);
+void wg_socket_set_peer_endpoint_from_skb(struct wg_peer *peer,
+					  const struct sk_buff *skb);
+void wg_socket_clear_peer_endpoint_src(struct wg_peer *peer);
+
+#if defined(CONFIG_DYNAMIC_DEBUG) || defined(DEBUG)
+#define net_dbg_skb_ratelimited(fmt, dev, skb, ...) do {                       \
+		struct endpoint __endpoint;                                    \
+		wg_socket_endpoint_from_skb(&__endpoint, skb);                 \
+		net_dbg_ratelimited(fmt, dev, &__endpoint.addr,                \
+				    ##__VA_ARGS__);                            \
+	} while (0)
+#else
+#define net_dbg_skb_ratelimited(fmt, skb, ...)
+#endif
+
+#endif /* _WG_SOCKET_H */
diff --git a/drivers/net/wireguard/timers.c b/drivers/net/wireguard/timers.c
new file mode 100644
index 000000000000..d54d32ac9bc4
--- /dev/null
+++ b/drivers/net/wireguard/timers.c
@@ -0,0 +1,243 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#include "timers.h"
+#include "device.h"
+#include "peer.h"
+#include "queueing.h"
+#include "socket.h"
+
+/*
+ * - Timer for retransmitting the handshake if we don't hear back after
+ * `REKEY_TIMEOUT + jitter` ms.
+ *
+ * - Timer for sending empty packet if we have received a packet but after have
+ * not sent one for `KEEPALIVE_TIMEOUT` ms.
+ *
+ * - Timer for initiating new handshake if we have sent a packet but after have
+ * not received one (even empty) for `(KEEPALIVE_TIMEOUT + REKEY_TIMEOUT) +
+ * jitter` ms.
+ *
+ * - Timer for zeroing out all ephemeral keys after `(REJECT_AFTER_TIME * 3)` ms
+ * if no new keys have been received.
+ *
+ * - Timer for, if enabled, sending an empty authenticated packet every user-
+ * specified seconds.
+ */
+
+static inline void mod_peer_timer(struct wg_peer *peer,
+				  struct timer_list *timer,
+				  unsigned long expires)
+{
+	rcu_read_lock_bh();
+	if (likely(netif_running(peer->device->dev) &&
+		   !READ_ONCE(peer->is_dead)))
+		mod_timer(timer, expires);
+	rcu_read_unlock_bh();
+}
+
+static void wg_expired_retransmit_handshake(struct timer_list *timer)
+{
+	struct wg_peer *peer = from_timer(peer, timer,
+					  timer_retransmit_handshake);
+
+	if (peer->timer_handshake_attempts > MAX_TIMER_HANDSHAKES) {
+		pr_debug("%s: Handshake for peer %llu (%pISpfsc) did not complete after %d attempts, giving up\n",
+			 peer->device->dev->name, peer->internal_id,
+			 &peer->endpoint.addr, MAX_TIMER_HANDSHAKES + 2);
+
+		del_timer(&peer->timer_send_keepalive);
+		/* We drop all packets without a keypair and don't try again,
+		 * if we try unsuccessfully for too long to make a handshake.
+		 */
+		wg_packet_purge_staged_packets(peer);
+
+		/* We set a timer for destroying any residue that might be left
+		 * of a partial exchange.
+		 */
+		if (!timer_pending(&peer->timer_zero_key_material))
+			mod_peer_timer(peer, &peer->timer_zero_key_material,
+				       jiffies + REJECT_AFTER_TIME * 3 * HZ);
+	} else {
+		++peer->timer_handshake_attempts;
+		pr_debug("%s: Handshake for peer %llu (%pISpfsc) did not complete after %d seconds, retrying (try %d)\n",
+			 peer->device->dev->name, peer->internal_id,
+			 &peer->endpoint.addr, REKEY_TIMEOUT,
+			 peer->timer_handshake_attempts + 1);
+
+		/* We clear the endpoint address src address, in case this is
+		 * the cause of trouble.
+		 */
+		wg_socket_clear_peer_endpoint_src(peer);
+
+		wg_packet_send_queued_handshake_initiation(peer, true);
+	}
+}
+
+static void wg_expired_send_keepalive(struct timer_list *timer)
+{
+	struct wg_peer *peer = from_timer(peer, timer, timer_send_keepalive);
+
+	wg_packet_send_keepalive(peer);
+	if (peer->timer_need_another_keepalive) {
+		peer->timer_need_another_keepalive = false;
+		mod_peer_timer(peer, &peer->timer_send_keepalive,
+			       jiffies + KEEPALIVE_TIMEOUT * HZ);
+	}
+}
+
+static void wg_expired_new_handshake(struct timer_list *timer)
+{
+	struct wg_peer *peer = from_timer(peer, timer, timer_new_handshake);
+
+	pr_debug("%s: Retrying handshake with peer %llu (%pISpfsc) because we stopped hearing back after %d seconds\n",
+		 peer->device->dev->name, peer->internal_id,
+		 &peer->endpoint.addr, KEEPALIVE_TIMEOUT + REKEY_TIMEOUT);
+	/* We clear the endpoint address src address, in case this is the cause
+	 * of trouble.
+	 */
+	wg_socket_clear_peer_endpoint_src(peer);
+	wg_packet_send_queued_handshake_initiation(peer, false);
+}
+
+static void wg_expired_zero_key_material(struct timer_list *timer)
+{
+	struct wg_peer *peer = from_timer(peer, timer, timer_zero_key_material);
+
+	rcu_read_lock_bh();
+	if (!READ_ONCE(peer->is_dead)) {
+		wg_peer_get(peer);
+		if (!queue_work(peer->device->handshake_send_wq,
+				&peer->clear_peer_work))
+			/* If the work was already on the queue, we want to drop
+			 * the extra reference.
+			 */
+			wg_peer_put(peer);
+	}
+	rcu_read_unlock_bh();
+}
+
+static void wg_queued_expired_zero_key_material(struct work_struct *work)
+{
+	struct wg_peer *peer = container_of(work, struct wg_peer,
+					    clear_peer_work);
+
+	pr_debug("%s: Zeroing out all keys for peer %llu (%pISpfsc), since we haven't received a new one in %d seconds\n",
+		 peer->device->dev->name, peer->internal_id,
+		 &peer->endpoint.addr, REJECT_AFTER_TIME * 3);
+	wg_noise_handshake_clear(&peer->handshake);
+	wg_noise_keypairs_clear(&peer->keypairs);
+	wg_peer_put(peer);
+}
+
+static void wg_expired_send_persistent_keepalive(struct timer_list *timer)
+{
+	struct wg_peer *peer = from_timer(peer, timer,
+					  timer_persistent_keepalive);
+
+	if (likely(peer->persistent_keepalive_interval))
+		wg_packet_send_keepalive(peer);
+}
+
+/* Should be called after an authenticated data packet is sent. */
+void wg_timers_data_sent(struct wg_peer *peer)
+{
+	if (!timer_pending(&peer->timer_new_handshake))
+		mod_peer_timer(peer, &peer->timer_new_handshake,
+			jiffies + (KEEPALIVE_TIMEOUT + REKEY_TIMEOUT) * HZ +
+			prandom_u32_max(REKEY_TIMEOUT_JITTER_MAX_JIFFIES));
+}
+
+/* Should be called after an authenticated data packet is received. */
+void wg_timers_data_received(struct wg_peer *peer)
+{
+	if (likely(netif_running(peer->device->dev))) {
+		if (!timer_pending(&peer->timer_send_keepalive))
+			mod_peer_timer(peer, &peer->timer_send_keepalive,
+				       jiffies + KEEPALIVE_TIMEOUT * HZ);
+		else
+			peer->timer_need_another_keepalive = true;
+	}
+}
+
+/* Should be called after any type of authenticated packet is sent, whether
+ * keepalive, data, or handshake.
+ */
+void wg_timers_any_authenticated_packet_sent(struct wg_peer *peer)
+{
+	del_timer(&peer->timer_send_keepalive);
+}
+
+/* Should be called after any type of authenticated packet is received, whether
+ * keepalive, data, or handshake.
+ */
+void wg_timers_any_authenticated_packet_received(struct wg_peer *peer)
+{
+	del_timer(&peer->timer_new_handshake);
+}
+
+/* Should be called after a handshake initiation message is sent. */
+void wg_timers_handshake_initiated(struct wg_peer *peer)
+{
+	mod_peer_timer(peer, &peer->timer_retransmit_handshake,
+		       jiffies + REKEY_TIMEOUT * HZ +
+		       prandom_u32_max(REKEY_TIMEOUT_JITTER_MAX_JIFFIES));
+}
+
+/* Should be called after a handshake response message is received and processed
+ * or when getting key confirmation via the first data message.
+ */
+void wg_timers_handshake_complete(struct wg_peer *peer)
+{
+	del_timer(&peer->timer_retransmit_handshake);
+	peer->timer_handshake_attempts = 0;
+	peer->sent_lastminute_handshake = false;
+	ktime_get_real_ts64(&peer->walltime_last_handshake);
+}
+
+/* Should be called after an ephemeral key is created, which is before sending a
+ * handshake response or after receiving a handshake response.
+ */
+void wg_timers_session_derived(struct wg_peer *peer)
+{
+	mod_peer_timer(peer, &peer->timer_zero_key_material,
+		       jiffies + REJECT_AFTER_TIME * 3 * HZ);
+}
+
+/* Should be called before a packet with authentication, whether
+ * keepalive, data, or handshakem is sent, or after one is received.
+ */
+void wg_timers_any_authenticated_packet_traversal(struct wg_peer *peer)
+{
+	if (peer->persistent_keepalive_interval)
+		mod_peer_timer(peer, &peer->timer_persistent_keepalive,
+			jiffies + peer->persistent_keepalive_interval * HZ);
+}
+
+void wg_timers_init(struct wg_peer *peer)
+{
+	timer_setup(&peer->timer_retransmit_handshake,
+		    wg_expired_retransmit_handshake, 0);
+	timer_setup(&peer->timer_send_keepalive, wg_expired_send_keepalive, 0);
+	timer_setup(&peer->timer_new_handshake, wg_expired_new_handshake, 0);
+	timer_setup(&peer->timer_zero_key_material,
+		    wg_expired_zero_key_material, 0);
+	timer_setup(&peer->timer_persistent_keepalive,
+		    wg_expired_send_persistent_keepalive, 0);
+	INIT_WORK(&peer->clear_peer_work, wg_queued_expired_zero_key_material);
+	peer->timer_handshake_attempts = 0;
+	peer->sent_lastminute_handshake = false;
+	peer->timer_need_another_keepalive = false;
+}
+
+void wg_timers_stop(struct wg_peer *peer)
+{
+	del_timer_sync(&peer->timer_retransmit_handshake);
+	del_timer_sync(&peer->timer_send_keepalive);
+	del_timer_sync(&peer->timer_new_handshake);
+	del_timer_sync(&peer->timer_zero_key_material);
+	del_timer_sync(&peer->timer_persistent_keepalive);
+	flush_work(&peer->clear_peer_work);
+}
diff --git a/drivers/net/wireguard/timers.h b/drivers/net/wireguard/timers.h
new file mode 100644
index 000000000000..f0653dcb1326
--- /dev/null
+++ b/drivers/net/wireguard/timers.h
@@ -0,0 +1,31 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#ifndef _WG_TIMERS_H
+#define _WG_TIMERS_H
+
+#include <linux/ktime.h>
+
+struct wg_peer;
+
+void wg_timers_init(struct wg_peer *peer);
+void wg_timers_stop(struct wg_peer *peer);
+void wg_timers_data_sent(struct wg_peer *peer);
+void wg_timers_data_received(struct wg_peer *peer);
+void wg_timers_any_authenticated_packet_sent(struct wg_peer *peer);
+void wg_timers_any_authenticated_packet_received(struct wg_peer *peer);
+void wg_timers_handshake_initiated(struct wg_peer *peer);
+void wg_timers_handshake_complete(struct wg_peer *peer);
+void wg_timers_session_derived(struct wg_peer *peer);
+void wg_timers_any_authenticated_packet_traversal(struct wg_peer *peer);
+
+static inline bool wg_birthdate_has_expired(u64 birthday_nanoseconds,
+					    u64 expiration_seconds)
+{
+	return (s64)(birthday_nanoseconds + expiration_seconds * NSEC_PER_SEC)
+		<= (s64)ktime_get_coarse_boottime_ns();
+}
+
+#endif /* _WG_TIMERS_H */
diff --git a/drivers/net/wireguard/version.h b/drivers/net/wireguard/version.h
new file mode 100644
index 000000000000..a1a269a11634
--- /dev/null
+++ b/drivers/net/wireguard/version.h
@@ -0,0 +1 @@
+#define WIREGUARD_VERSION "1.0.0"
diff --git a/include/uapi/linux/wireguard.h b/include/uapi/linux/wireguard.h
new file mode 100644
index 000000000000..dd8a47c4ad11
--- /dev/null
+++ b/include/uapi/linux/wireguard.h
@@ -0,0 +1,196 @@
+/* SPDX-License-Identifier: (GPL-2.0 WITH Linux-syscall-note) OR MIT */
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ *
+ * Documentation
+ * =============
+ *
+ * The below enums and macros are for interfacing with WireGuard, using generic
+ * netlink, with family WG_GENL_NAME and version WG_GENL_VERSION. It defines two
+ * methods: get and set. Note that while they share many common attributes,
+ * these two functions actually accept a slightly different set of inputs and
+ * outputs.
+ *
+ * WG_CMD_GET_DEVICE
+ * -----------------
+ *
+ * May only be called via NLM_F_REQUEST | NLM_F_DUMP. The command should contain
+ * one but not both of:
+ *
+ *    WGDEVICE_A_IFINDEX: NLA_U32
+ *    WGDEVICE_A_IFNAME: NLA_NUL_STRING, maxlen IFNAMESIZ - 1
+ *
+ * The kernel will then return several messages (NLM_F_MULTI) containing the
+ * following tree of nested items:
+ *
+ *    WGDEVICE_A_IFINDEX: NLA_U32
+ *    WGDEVICE_A_IFNAME: NLA_NUL_STRING, maxlen IFNAMESIZ - 1
+ *    WGDEVICE_A_PRIVATE_KEY: NLA_EXACT_LEN, len WG_KEY_LEN
+ *    WGDEVICE_A_PUBLIC_KEY: NLA_EXACT_LEN, len WG_KEY_LEN
+ *    WGDEVICE_A_LISTEN_PORT: NLA_U16
+ *    WGDEVICE_A_FWMARK: NLA_U32
+ *    WGDEVICE_A_PEERS: NLA_NESTED
+ *        0: NLA_NESTED
+ *            WGPEER_A_PUBLIC_KEY: NLA_EXACT_LEN, len WG_KEY_LEN
+ *            WGPEER_A_PRESHARED_KEY: NLA_EXACT_LEN, len WG_KEY_LEN
+ *            WGPEER_A_ENDPOINT: NLA_MIN_LEN(struct sockaddr), struct sockaddr_in or struct sockaddr_in6
+ *            WGPEER_A_PERSISTENT_KEEPALIVE_INTERVAL: NLA_U16
+ *            WGPEER_A_LAST_HANDSHAKE_TIME: NLA_EXACT_LEN, struct __kernel_timespec
+ *            WGPEER_A_RX_BYTES: NLA_U64
+ *            WGPEER_A_TX_BYTES: NLA_U64
+ *            WGPEER_A_ALLOWEDIPS: NLA_NESTED
+ *                0: NLA_NESTED
+ *                    WGALLOWEDIP_A_FAMILY: NLA_U16
+ *                    WGALLOWEDIP_A_IPADDR: NLA_MIN_LEN(struct in_addr), struct in_addr or struct in6_addr
+ *                    WGALLOWEDIP_A_CIDR_MASK: NLA_U8
+ *                0: NLA_NESTED
+ *                    ...
+ *                0: NLA_NESTED
+ *                    ...
+ *                ...
+ *            WGPEER_A_PROTOCOL_VERSION: NLA_U32
+ *        0: NLA_NESTED
+ *            ...
+ *        ...
+ *
+ * It is possible that all of the allowed IPs of a single peer will not
+ * fit within a single netlink message. In that case, the same peer will
+ * be written in the following message, except it will only contain
+ * WGPEER_A_PUBLIC_KEY and WGPEER_A_ALLOWEDIPS. This may occur several
+ * times in a row for the same peer. It is then up to the receiver to
+ * coalesce adjacent peers. Likewise, it is possible that all peers will
+ * not fit within a single message. So, subsequent peers will be sent
+ * in following messages, except those will only contain WGDEVICE_A_IFNAME
+ * and WGDEVICE_A_PEERS. It is then up to the receiver to coalesce these
+ * messages to form the complete list of peers.
+ *
+ * Since this is an NLA_F_DUMP command, the final message will always be
+ * NLMSG_DONE, even if an error occurs. However, this NLMSG_DONE message
+ * contains an integer error code. It is either zero or a negative error
+ * code corresponding to the errno.
+ *
+ * WG_CMD_SET_DEVICE
+ * -----------------
+ *
+ * May only be called via NLM_F_REQUEST. The command should contain the
+ * following tree of nested items, containing one but not both of
+ * WGDEVICE_A_IFINDEX and WGDEVICE_A_IFNAME:
+ *
+ *    WGDEVICE_A_IFINDEX: NLA_U32
+ *    WGDEVICE_A_IFNAME: NLA_NUL_STRING, maxlen IFNAMESIZ - 1
+ *    WGDEVICE_A_FLAGS: NLA_U32, 0 or WGDEVICE_F_REPLACE_PEERS if all current
+ *                      peers should be removed prior to adding the list below.
+ *    WGDEVICE_A_PRIVATE_KEY: len WG_KEY_LEN, all zeros to remove
+ *    WGDEVICE_A_LISTEN_PORT: NLA_U16, 0 to choose randomly
+ *    WGDEVICE_A_FWMARK: NLA_U32, 0 to disable
+ *    WGDEVICE_A_PEERS: NLA_NESTED
+ *        0: NLA_NESTED
+ *            WGPEER_A_PUBLIC_KEY: len WG_KEY_LEN
+ *            WGPEER_A_FLAGS: NLA_U32, 0 and/or WGPEER_F_REMOVE_ME if the
+ *                            specified peer should not exist at the end of the
+ *                            operation, rather than added/updated and/or
+ *                            WGPEER_F_REPLACE_ALLOWEDIPS if all current allowed
+ *                            IPs of this peer should be removed prior to adding
+ *                            the list below and/or WGPEER_F_UPDATE_ONLY if the
+ *                            peer should only be set if it already exists.
+ *            WGPEER_A_PRESHARED_KEY: len WG_KEY_LEN, all zeros to remove
+ *            WGPEER_A_ENDPOINT: struct sockaddr_in or struct sockaddr_in6
+ *            WGPEER_A_PERSISTENT_KEEPALIVE_INTERVAL: NLA_U16, 0 to disable
+ *            WGPEER_A_ALLOWEDIPS: NLA_NESTED
+ *                0: NLA_NESTED
+ *                    WGALLOWEDIP_A_FAMILY: NLA_U16
+ *                    WGALLOWEDIP_A_IPADDR: struct in_addr or struct in6_addr
+ *                    WGALLOWEDIP_A_CIDR_MASK: NLA_U8
+ *                0: NLA_NESTED
+ *                    ...
+ *                0: NLA_NESTED
+ *                    ...
+ *                ...
+ *            WGPEER_A_PROTOCOL_VERSION: NLA_U32, should not be set or used at
+ *                                       all by most users of this API, as the
+ *                                       most recent protocol will be used when
+ *                                       this is unset. Otherwise, must be set
+ *                                       to 1.
+ *        0: NLA_NESTED
+ *            ...
+ *        ...
+ *
+ * It is possible that the amount of configuration data exceeds that of
+ * the maximum message length accepted by the kernel. In that case, several
+ * messages should be sent one after another, with each successive one
+ * filling in information not contained in the prior. Note that if
+ * WGDEVICE_F_REPLACE_PEERS is specified in the first message, it probably
+ * should not be specified in fragments that come after, so that the list
+ * of peers is only cleared the first time but appened after. Likewise for
+ * peers, if WGPEER_F_REPLACE_ALLOWEDIPS is specified in the first message
+ * of a peer, it likely should not be specified in subsequent fragments.
+ *
+ * If an error occurs, NLMSG_ERROR will reply containing an errno.
+ */
+
+#ifndef _WG_UAPI_WIREGUARD_H
+#define _WG_UAPI_WIREGUARD_H
+
+#define WG_GENL_NAME "wireguard"
+#define WG_GENL_VERSION 1
+
+#define WG_KEY_LEN 32
+
+enum wg_cmd {
+	WG_CMD_GET_DEVICE,
+	WG_CMD_SET_DEVICE,
+	__WG_CMD_MAX
+};
+#define WG_CMD_MAX (__WG_CMD_MAX - 1)
+
+enum wgdevice_flag {
+	WGDEVICE_F_REPLACE_PEERS = 1U << 0,
+	__WGDEVICE_F_ALL = WGDEVICE_F_REPLACE_PEERS
+};
+enum wgdevice_attribute {
+	WGDEVICE_A_UNSPEC,
+	WGDEVICE_A_IFINDEX,
+	WGDEVICE_A_IFNAME,
+	WGDEVICE_A_PRIVATE_KEY,
+	WGDEVICE_A_PUBLIC_KEY,
+	WGDEVICE_A_FLAGS,
+	WGDEVICE_A_LISTEN_PORT,
+	WGDEVICE_A_FWMARK,
+	WGDEVICE_A_PEERS,
+	__WGDEVICE_A_LAST
+};
+#define WGDEVICE_A_MAX (__WGDEVICE_A_LAST - 1)
+
+enum wgpeer_flag {
+	WGPEER_F_REMOVE_ME = 1U << 0,
+	WGPEER_F_REPLACE_ALLOWEDIPS = 1U << 1,
+	WGPEER_F_UPDATE_ONLY = 1U << 2,
+	__WGPEER_F_ALL = WGPEER_F_REMOVE_ME | WGPEER_F_REPLACE_ALLOWEDIPS |
+			 WGPEER_F_UPDATE_ONLY
+};
+enum wgpeer_attribute {
+	WGPEER_A_UNSPEC,
+	WGPEER_A_PUBLIC_KEY,
+	WGPEER_A_PRESHARED_KEY,
+	WGPEER_A_FLAGS,
+	WGPEER_A_ENDPOINT,
+	WGPEER_A_PERSISTENT_KEEPALIVE_INTERVAL,
+	WGPEER_A_LAST_HANDSHAKE_TIME,
+	WGPEER_A_RX_BYTES,
+	WGPEER_A_TX_BYTES,
+	WGPEER_A_ALLOWEDIPS,
+	WGPEER_A_PROTOCOL_VERSION,
+	__WGPEER_A_LAST
+};
+#define WGPEER_A_MAX (__WGPEER_A_LAST - 1)
+
+enum wgallowedip_attribute {
+	WGALLOWEDIP_A_UNSPEC,
+	WGALLOWEDIP_A_FAMILY,
+	WGALLOWEDIP_A_IPADDR,
+	WGALLOWEDIP_A_CIDR_MASK,
+	__WGALLOWEDIP_A_LAST
+};
+#define WGALLOWEDIP_A_MAX (__WGALLOWEDIP_A_LAST - 1)
+
+#endif /* _WG_UAPI_WIREGUARD_H */
diff --git a/tools/testing/selftests/wireguard/netns.sh b/tools/testing/selftests/wireguard/netns.sh
new file mode 100755
index 000000000000..e7310d9390f7
--- /dev/null
+++ b/tools/testing/selftests/wireguard/netns.sh
@@ -0,0 +1,537 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+#
+# Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+#
+# This script tests the below topology:
+#
+# ┌─────────────────────┐   ┌──────────────────────────────────┐   ┌─────────────────────┐
+# │   $ns1 namespace    │   │          $ns0 namespace          │   │   $ns2 namespace    │
+# │                     │   │                                  │   │                     │
+# │┌────────┐           │   │            ┌────────┐            │   │           ┌────────┐│
+# ││  wg0   │───────────┼───┼────────────│   lo   │────────────┼───┼───────────│  wg0   ││
+# │├────────┴──────────┐│   │    ┌───────┴────────┴────────┐   │   │┌──────────┴────────┤│
+# ││192.168.241.1/24   ││   │    │(ns1)         (ns2)      │   │   ││192.168.241.2/24   ││
+# ││fd00::1/24         ││   │    │127.0.0.1:1   127.0.0.1:2│   │   ││fd00::2/24         ││
+# │└───────────────────┘│   │    │[::]:1        [::]:2     │   │   │└───────────────────┘│
+# └─────────────────────┘   │    └─────────────────────────┘   │   └─────────────────────┘
+#                           └──────────────────────────────────┘
+#
+# After the topology is prepared we run a series of TCP/UDP iperf3 tests between the
+# wireguard peers in $ns1 and $ns2. Note that $ns0 is the endpoint for the wg0
+# interfaces in $ns1 and $ns2. See https://www.wireguard.com/netns/ for further
+# details on how this is accomplished.
+set -e
+
+exec 3>&1
+export WG_HIDE_KEYS=never
+netns0="wg-test-$$-0"
+netns1="wg-test-$$-1"
+netns2="wg-test-$$-2"
+pretty() { echo -e "\x1b[32m\x1b[1m[+] ${1:+NS$1: }${2}\x1b[0m" >&3; }
+pp() { pretty "" "$*"; "$@"; }
+maybe_exec() { if [[ $BASHPID -eq $$ ]]; then "$@"; else exec "$@"; fi; }
+n0() { pretty 0 "$*"; maybe_exec ip netns exec $netns0 "$@"; }
+n1() { pretty 1 "$*"; maybe_exec ip netns exec $netns1 "$@"; }
+n2() { pretty 2 "$*"; maybe_exec ip netns exec $netns2 "$@"; }
+ip0() { pretty 0 "ip $*"; ip -n $netns0 "$@"; }
+ip1() { pretty 1 "ip $*"; ip -n $netns1 "$@"; }
+ip2() { pretty 2 "ip $*"; ip -n $netns2 "$@"; }
+sleep() { read -t "$1" -N 0 || true; }
+waitiperf() { pretty "${1//*-}" "wait for iperf:5201"; while [[ $(ss -N "$1" -tlp 'sport = 5201') != *iperf3* ]]; do sleep 0.1; done; }
+waitncatudp() { pretty "${1//*-}" "wait for udp:1111"; while [[ $(ss -N "$1" -ulp 'sport = 1111') != *ncat* ]]; do sleep 0.1; done; }
+waitncattcp() { pretty "${1//*-}" "wait for tcp:1111"; while [[ $(ss -N "$1" -tlp 'sport = 1111') != *ncat* ]]; do sleep 0.1; done; }
+waitiface() { pretty "${1//*-}" "wait for $2 to come up"; ip netns exec "$1" bash -c "while [[ \$(< \"/sys/class/net/$2/operstate\") != up ]]; do read -t .1 -N 0 || true; done;"; }
+
+cleanup() {
+	set +e
+	exec 2>/dev/null
+	printf "$orig_message_cost" > /proc/sys/net/core/message_cost
+	ip0 link del dev wg0
+	ip1 link del dev wg0
+	ip2 link del dev wg0
+	local to_kill="$(ip netns pids $netns0) $(ip netns pids $netns1) $(ip netns pids $netns2)"
+	[[ -n $to_kill ]] && kill $to_kill
+	pp ip netns del $netns1
+	pp ip netns del $netns2
+	pp ip netns del $netns0
+	exit
+}
+
+orig_message_cost="$(< /proc/sys/net/core/message_cost)"
+trap cleanup EXIT
+printf 0 > /proc/sys/net/core/message_cost
+
+ip netns del $netns0 2>/dev/null || true
+ip netns del $netns1 2>/dev/null || true
+ip netns del $netns2 2>/dev/null || true
+pp ip netns add $netns0
+pp ip netns add $netns1
+pp ip netns add $netns2
+ip0 link set up dev lo
+
+ip0 link add dev wg0 type wireguard
+ip0 link set wg0 netns $netns1
+ip0 link add dev wg0 type wireguard
+ip0 link set wg0 netns $netns2
+key1="$(pp wg genkey)"
+key2="$(pp wg genkey)"
+key3="$(pp wg genkey)"
+pub1="$(pp wg pubkey <<<"$key1")"
+pub2="$(pp wg pubkey <<<"$key2")"
+pub3="$(pp wg pubkey <<<"$key3")"
+psk="$(pp wg genpsk)"
+[[ -n $key1 && -n $key2 && -n $psk ]]
+
+configure_peers() {
+	ip1 addr add 192.168.241.1/24 dev wg0
+	ip1 addr add fd00::1/24 dev wg0
+
+	ip2 addr add 192.168.241.2/24 dev wg0
+	ip2 addr add fd00::2/24 dev wg0
+
+	n1 wg set wg0 \
+		private-key <(echo "$key1") \
+		listen-port 1 \
+		peer "$pub2" \
+			preshared-key <(echo "$psk") \
+			allowed-ips 192.168.241.2/32,fd00::2/128
+	n2 wg set wg0 \
+		private-key <(echo "$key2") \
+		listen-port 2 \
+		peer "$pub1" \
+			preshared-key <(echo "$psk") \
+			allowed-ips 192.168.241.1/32,fd00::1/128
+
+	ip1 link set up dev wg0
+	ip2 link set up dev wg0
+}
+configure_peers
+
+tests() {
+	# Ping over IPv4
+	n2 ping -c 10 -f -W 1 192.168.241.1
+	n1 ping -c 10 -f -W 1 192.168.241.2
+
+	# Ping over IPv6
+	n2 ping6 -c 10 -f -W 1 fd00::1
+	n1 ping6 -c 10 -f -W 1 fd00::2
+
+	# TCP over IPv4
+	n2 iperf3 -s -1 -B 192.168.241.2 &
+	waitiperf $netns2
+	n1 iperf3 -Z -t 3 -c 192.168.241.2
+
+	# TCP over IPv6
+	n1 iperf3 -s -1 -B fd00::1 &
+	waitiperf $netns1
+	n2 iperf3 -Z -t 3 -c fd00::1
+
+	# UDP over IPv4
+	n1 iperf3 -s -1 -B 192.168.241.1 &
+	waitiperf $netns1
+	n2 iperf3 -Z -t 3 -b 0 -u -c 192.168.241.1
+
+	# UDP over IPv6
+	n2 iperf3 -s -1 -B fd00::2 &
+	waitiperf $netns2
+	n1 iperf3 -Z -t 3 -b 0 -u -c fd00::2
+}
+
+[[ $(ip1 link show dev wg0) =~ mtu\ ([0-9]+) ]] && orig_mtu="${BASH_REMATCH[1]}"
+big_mtu=$(( 34816 - 1500 + $orig_mtu ))
+
+# Test using IPv4 as outer transport
+n1 wg set wg0 peer "$pub2" endpoint 127.0.0.1:2
+n2 wg set wg0 peer "$pub1" endpoint 127.0.0.1:1
+# Before calling tests, we first make sure that the stats counters and timestamper are working
+n2 ping -c 10 -f -W 1 192.168.241.1
+{ read _; read _; read _; read rx_bytes _; read _; read tx_bytes _; } < <(ip2 -stats link show dev wg0)
+(( rx_bytes == 1372 && (tx_bytes == 1428 || tx_bytes == 1460) ))
+{ read _; read _; read _; read rx_bytes _; read _; read tx_bytes _; } < <(ip1 -stats link show dev wg0)
+(( tx_bytes == 1372 && (rx_bytes == 1428 || rx_bytes == 1460) ))
+read _ rx_bytes tx_bytes < <(n2 wg show wg0 transfer)
+(( rx_bytes == 1372 && (tx_bytes == 1428 || tx_bytes == 1460) ))
+read _ rx_bytes tx_bytes < <(n1 wg show wg0 transfer)
+(( tx_bytes == 1372 && (rx_bytes == 1428 || rx_bytes == 1460) ))
+read _ timestamp < <(n1 wg show wg0 latest-handshakes)
+(( timestamp != 0 ))
+
+tests
+ip1 link set wg0 mtu $big_mtu
+ip2 link set wg0 mtu $big_mtu
+tests
+
+ip1 link set wg0 mtu $orig_mtu
+ip2 link set wg0 mtu $orig_mtu
+
+# Test using IPv6 as outer transport
+n1 wg set wg0 peer "$pub2" endpoint [::1]:2
+n2 wg set wg0 peer "$pub1" endpoint [::1]:1
+tests
+ip1 link set wg0 mtu $big_mtu
+ip2 link set wg0 mtu $big_mtu
+tests
+
+# Test that route MTUs work with the padding
+ip1 link set wg0 mtu 1300
+ip2 link set wg0 mtu 1300
+n1 wg set wg0 peer "$pub2" endpoint 127.0.0.1:2
+n2 wg set wg0 peer "$pub1" endpoint 127.0.0.1:1
+n0 iptables -A INPUT -m length --length 1360 -j DROP
+n1 ip route add 192.168.241.2/32 dev wg0 mtu 1299
+n2 ip route add 192.168.241.1/32 dev wg0 mtu 1299
+n2 ping -c 1 -W 1 -s 1269 192.168.241.1
+n2 ip route delete 192.168.241.1/32 dev wg0 mtu 1299
+n1 ip route delete 192.168.241.2/32 dev wg0 mtu 1299
+n0 iptables -F INPUT
+
+ip1 link set wg0 mtu $orig_mtu
+ip2 link set wg0 mtu $orig_mtu
+
+# Test using IPv4 that roaming works
+ip0 -4 addr del 127.0.0.1/8 dev lo
+ip0 -4 addr add 127.212.121.99/8 dev lo
+n1 wg set wg0 listen-port 9999
+n1 wg set wg0 peer "$pub2" endpoint 127.0.0.1:2
+n1 ping6 -W 1 -c 1 fd00::2
+[[ $(n2 wg show wg0 endpoints) == "$pub1	127.212.121.99:9999" ]]
+
+# Test using IPv6 that roaming works
+n1 wg set wg0 listen-port 9998
+n1 wg set wg0 peer "$pub2" endpoint [::1]:2
+n1 ping -W 1 -c 1 192.168.241.2
+[[ $(n2 wg show wg0 endpoints) == "$pub1	[::1]:9998" ]]
+
+# Test that crypto-RP filter works
+n1 wg set wg0 peer "$pub2" allowed-ips 192.168.241.0/24
+exec 4< <(n1 ncat -l -u -p 1111)
+ncat_pid=$!
+waitncatudp $netns1
+n2 ncat -u 192.168.241.1 1111 <<<"X"
+read -r -N 1 -t 1 out <&4 && [[ $out == "X" ]]
+kill $ncat_pid
+more_specific_key="$(pp wg genkey | pp wg pubkey)"
+n1 wg set wg0 peer "$more_specific_key" allowed-ips 192.168.241.2/32
+n2 wg set wg0 listen-port 9997
+exec 4< <(n1 ncat -l -u -p 1111)
+ncat_pid=$!
+waitncatudp $netns1
+n2 ncat -u 192.168.241.1 1111 <<<"X"
+! read -r -N 1 -t 1 out <&4 || false
+kill $ncat_pid
+n1 wg set wg0 peer "$more_specific_key" remove
+[[ $(n1 wg show wg0 endpoints) == "$pub2	[::1]:9997" ]]
+
+# Test that we can change private keys keys and immediately handshake
+n1 wg set wg0 private-key <(echo "$key1") peer "$pub2" preshared-key <(echo "$psk") allowed-ips 192.168.241.2/32 endpoint 127.0.0.1:2
+n2 wg set wg0 private-key <(echo "$key2") listen-port 2 peer "$pub1" preshared-key <(echo "$psk") allowed-ips 192.168.241.1/32
+n1 ping -W 1 -c 1 192.168.241.2
+n1 wg set wg0 private-key <(echo "$key3")
+n2 wg set wg0 peer "$pub3" preshared-key <(echo "$psk") allowed-ips 192.168.241.1/32 peer "$pub1" remove
+n1 ping -W 1 -c 1 192.168.241.2
+
+ip1 link del wg0
+ip2 link del wg0
+
+# Test using NAT. We now change the topology to this:
+# ┌────────────────────────────────────────┐    ┌────────────────────────────────────────────────┐     ┌────────────────────────────────────────┐
+# │             $ns1 namespace             │    │                 $ns0 namespace                 │     │             $ns2 namespace             │
+# │                                        │    │                                                │     │                                        │
+# │  ┌─────┐             ┌─────┐           │    │    ┌──────┐              ┌──────┐              │     │  ┌─────┐            ┌─────┐            │
+# │  │ wg0 │─────────────│vethc│───────────┼────┼────│vethrc│              │vethrs│──────────────┼─────┼──│veths│────────────│ wg0 │            │
+# │  ├─────┴──────────┐  ├─────┴──────────┐│    │    ├──────┴─────────┐    ├──────┴────────────┐ │     │  ├─────┴──────────┐ ├─────┴──────────┐ │
+# │  │192.168.241.1/24│  │192.168.1.100/24││    │    │192.168.1.1/24  │    │10.0.0.1/24        │ │     │  │10.0.0.100/24   │ │192.168.241.2/24│ │
+# │  │fd00::1/24      │  │                ││    │    │                │    │SNAT:192.168.1.0/24│ │     │  │                │ │fd00::2/24      │ │
+# │  └────────────────┘  └────────────────┘│    │    └────────────────┘    └───────────────────┘ │     │  └────────────────┘ └────────────────┘ │
+# └────────────────────────────────────────┘    └────────────────────────────────────────────────┘     └────────────────────────────────────────┘
+
+ip1 link add dev wg0 type wireguard
+ip2 link add dev wg0 type wireguard
+configure_peers
+
+ip0 link add vethrc type veth peer name vethc
+ip0 link add vethrs type veth peer name veths
+ip0 link set vethc netns $netns1
+ip0 link set veths netns $netns2
+ip0 link set vethrc up
+ip0 link set vethrs up
+ip0 addr add 192.168.1.1/24 dev vethrc
+ip0 addr add 10.0.0.1/24 dev vethrs
+ip1 addr add 192.168.1.100/24 dev vethc
+ip1 link set vethc up
+ip1 route add default via 192.168.1.1
+ip2 addr add 10.0.0.100/24 dev veths
+ip2 link set veths up
+waitiface $netns0 vethrc
+waitiface $netns0 vethrs
+waitiface $netns1 vethc
+waitiface $netns2 veths
+
+n0 bash -c 'printf 1 > /proc/sys/net/ipv4/ip_forward'
+n0 bash -c 'printf 2 > /proc/sys/net/netfilter/nf_conntrack_udp_timeout'
+n0 bash -c 'printf 2 > /proc/sys/net/netfilter/nf_conntrack_udp_timeout_stream'
+n0 iptables -t nat -A POSTROUTING -s 192.168.1.0/24 -d 10.0.0.0/24 -j SNAT --to 10.0.0.1
+
+n1 wg set wg0 peer "$pub2" endpoint 10.0.0.100:2 persistent-keepalive 1
+n1 ping -W 1 -c 1 192.168.241.2
+n2 ping -W 1 -c 1 192.168.241.1
+[[ $(n2 wg show wg0 endpoints) == "$pub1	10.0.0.1:1" ]]
+# Demonstrate n2 can still send packets to n1, since persistent-keepalive will prevent connection tracking entry from expiring (to see entries: `n0 conntrack -L`).
+pp sleep 3
+n2 ping -W 1 -c 1 192.168.241.1
+n1 wg set wg0 peer "$pub2" persistent-keepalive 0
+
+# Do a wg-quick(8)-style policy routing for the default route, making sure vethc has a v6 address to tease out bugs.
+ip1 -6 addr add fc00::9/96 dev vethc
+ip1 -6 route add default via fc00::1
+ip2 -4 addr add 192.168.99.7/32 dev wg0
+ip2 -6 addr add abab::1111/128 dev wg0
+n1 wg set wg0 fwmark 51820 peer "$pub2" allowed-ips 192.168.99.7,abab::1111
+ip1 -6 route add default dev wg0 table 51820
+ip1 -6 rule add not fwmark 51820 table 51820
+ip1 -6 rule add table main suppress_prefixlength 0
+ip1 -4 route add default dev wg0 table 51820
+ip1 -4 rule add not fwmark 51820 table 51820
+ip1 -4 rule add table main suppress_prefixlength 0
+# suppress_prefixlength only got added in 3.12, and we want to support 3.10+.
+if [[ $(ip1 -4 rule show all) == *suppress_prefixlength* ]]; then
+	# Flood the pings instead of sending just one, to trigger routing table reference counting bugs.
+	n1 ping -W 1 -c 100 -f 192.168.99.7
+	n1 ping -W 1 -c 100 -f abab::1111
+fi
+
+n0 iptables -t nat -F
+ip0 link del vethrc
+ip0 link del vethrs
+ip1 link del wg0
+ip2 link del wg0
+
+# Test that saddr routing is sticky but not too sticky, changing to this topology:
+# ┌────────────────────────────────────────┐    ┌────────────────────────────────────────┐
+# │             $ns1 namespace             │    │             $ns2 namespace             │
+# │                                        │    │                                        │
+# │  ┌─────┐             ┌─────┐           │    │  ┌─────┐            ┌─────┐            │
+# │  │ wg0 │─────────────│veth1│───────────┼────┼──│veth2│────────────│ wg0 │            │
+# │  ├─────┴──────────┐  ├─────┴──────────┐│    │  ├─────┴──────────┐ ├─────┴──────────┐ │
+# │  │192.168.241.1/24│  │10.0.0.1/24     ││    │  │10.0.0.2/24     │ │192.168.241.2/24│ │
+# │  │fd00::1/24      │  │fd00:aa::1/96   ││    │  │fd00:aa::2/96   │ │fd00::2/24      │ │
+# │  └────────────────┘  └────────────────┘│    │  └────────────────┘ └────────────────┘ │
+# └────────────────────────────────────────┘    └────────────────────────────────────────┘
+
+ip1 link add dev wg0 type wireguard
+ip2 link add dev wg0 type wireguard
+configure_peers
+ip1 link add veth1 type veth peer name veth2
+ip1 link set veth2 netns $netns2
+n1 bash -c 'printf 0 > /proc/sys/net/ipv6/conf/all/accept_dad'
+n2 bash -c 'printf 0 > /proc/sys/net/ipv6/conf/all/accept_dad'
+n1 bash -c 'printf 0 > /proc/sys/net/ipv6/conf/veth1/accept_dad'
+n2 bash -c 'printf 0 > /proc/sys/net/ipv6/conf/veth2/accept_dad'
+n1 bash -c 'printf 1 > /proc/sys/net/ipv4/conf/veth1/promote_secondaries'
+
+# First we check that we aren't overly sticky and can fall over to new IPs when old ones are removed
+ip1 addr add 10.0.0.1/24 dev veth1
+ip1 addr add fd00:aa::1/96 dev veth1
+ip2 addr add 10.0.0.2/24 dev veth2
+ip2 addr add fd00:aa::2/96 dev veth2
+ip1 link set veth1 up
+ip2 link set veth2 up
+waitiface $netns1 veth1
+waitiface $netns2 veth2
+n1 wg set wg0 peer "$pub2" endpoint 10.0.0.2:2
+n1 ping -W 1 -c 1 192.168.241.2
+ip1 addr add 10.0.0.10/24 dev veth1
+ip1 addr del 10.0.0.1/24 dev veth1
+n1 ping -W 1 -c 1 192.168.241.2
+n1 wg set wg0 peer "$pub2" endpoint [fd00:aa::2]:2
+n1 ping -W 1 -c 1 192.168.241.2
+ip1 addr add fd00:aa::10/96 dev veth1
+ip1 addr del fd00:aa::1/96 dev veth1
+n1 ping -W 1 -c 1 192.168.241.2
+
+# Now we show that we can successfully do reply to sender routing
+ip1 link set veth1 down
+ip2 link set veth2 down
+ip1 addr flush dev veth1
+ip2 addr flush dev veth2
+ip1 addr add 10.0.0.1/24 dev veth1
+ip1 addr add 10.0.0.2/24 dev veth1
+ip1 addr add fd00:aa::1/96 dev veth1
+ip1 addr add fd00:aa::2/96 dev veth1
+ip2 addr add 10.0.0.3/24 dev veth2
+ip2 addr add fd00:aa::3/96 dev veth2
+ip1 link set veth1 up
+ip2 link set veth2 up
+waitiface $netns1 veth1
+waitiface $netns2 veth2
+n2 wg set wg0 peer "$pub1" endpoint 10.0.0.1:1
+n2 ping -W 1 -c 1 192.168.241.1
+[[ $(n2 wg show wg0 endpoints) == "$pub1	10.0.0.1:1" ]]
+n2 wg set wg0 peer "$pub1" endpoint [fd00:aa::1]:1
+n2 ping -W 1 -c 1 192.168.241.1
+[[ $(n2 wg show wg0 endpoints) == "$pub1	[fd00:aa::1]:1" ]]
+n2 wg set wg0 peer "$pub1" endpoint 10.0.0.2:1
+n2 ping -W 1 -c 1 192.168.241.1
+[[ $(n2 wg show wg0 endpoints) == "$pub1	10.0.0.2:1" ]]
+n2 wg set wg0 peer "$pub1" endpoint [fd00:aa::2]:1
+n2 ping -W 1 -c 1 192.168.241.1
+[[ $(n2 wg show wg0 endpoints) == "$pub1	[fd00:aa::2]:1" ]]
+
+# What happens if the inbound destination address belongs to a different interface as the default route?
+ip1 link add dummy0 type dummy
+ip1 addr add 10.50.0.1/24 dev dummy0
+ip1 link set dummy0 up
+ip2 route add 10.50.0.0/24 dev veth2
+n2 wg set wg0 peer "$pub1" endpoint 10.50.0.1:1
+n2 ping -W 1 -c 1 192.168.241.1
+[[ $(n2 wg show wg0 endpoints) == "$pub1	10.50.0.1:1" ]]
+
+ip1 link del dummy0
+ip1 addr flush dev veth1
+ip2 addr flush dev veth2
+ip1 route flush dev veth1
+ip2 route flush dev veth2
+
+# Now we see what happens if another interface route takes precedence over an ongoing one
+ip1 link add veth3 type veth peer name veth4
+ip1 link set veth4 netns $netns2
+ip1 addr add 10.0.0.1/24 dev veth1
+ip2 addr add 10.0.0.2/24 dev veth2
+ip1 addr add 10.0.0.3/24 dev veth3
+ip1 link set veth1 up
+ip2 link set veth2 up
+ip1 link set veth3 up
+ip2 link set veth4 up
+waitiface $netns1 veth1
+waitiface $netns2 veth2
+waitiface $netns1 veth3
+waitiface $netns2 veth4
+ip1 route flush dev veth1
+ip1 route flush dev veth3
+ip1 route add 10.0.0.0/24 dev veth1 src 10.0.0.1 metric 2
+n1 wg set wg0 peer "$pub2" endpoint 10.0.0.2:2
+n1 ping -W 1 -c 1 192.168.241.2
+[[ $(n2 wg show wg0 endpoints) == "$pub1	10.0.0.1:1" ]]
+ip1 route add 10.0.0.0/24 dev veth3 src 10.0.0.3 metric 1
+n1 bash -c 'printf 0 > /proc/sys/net/ipv4/conf/veth1/rp_filter'
+n2 bash -c 'printf 0 > /proc/sys/net/ipv4/conf/veth4/rp_filter'
+n1 bash -c 'printf 0 > /proc/sys/net/ipv4/conf/all/rp_filter'
+n2 bash -c 'printf 0 > /proc/sys/net/ipv4/conf/all/rp_filter'
+n1 ping -W 1 -c 1 192.168.241.2
+[[ $(n2 wg show wg0 endpoints) == "$pub1	10.0.0.3:1" ]]
+
+ip1 link del veth1
+ip1 link del veth3
+ip1 link del wg0
+ip2 link del wg0
+
+# We test that Netlink/IPC is working properly by doing things that usually cause split responses
+ip0 link add dev wg0 type wireguard
+config=( "[Interface]" "PrivateKey=$(wg genkey)" "[Peer]" "PublicKey=$(wg genkey)" )
+for a in {1..255}; do
+	for b in {0..255}; do
+		config+=( "AllowedIPs=$a.$b.0.0/16,$a::$b/128" )
+	done
+done
+n0 wg setconf wg0 <(printf '%s\n' "${config[@]}")
+i=0
+for ip in $(n0 wg show wg0 allowed-ips); do
+	((++i))
+done
+((i == 255*256*2+1))
+ip0 link del wg0
+ip0 link add dev wg0 type wireguard
+config=( "[Interface]" "PrivateKey=$(wg genkey)" )
+for a in {1..40}; do
+	config+=( "[Peer]" "PublicKey=$(wg genkey)" )
+	for b in {1..52}; do
+		config+=( "AllowedIPs=$a.$b.0.0/16" )
+	done
+done
+n0 wg setconf wg0 <(printf '%s\n' "${config[@]}")
+i=0
+while read -r line; do
+	j=0
+	for ip in $line; do
+		((++j))
+	done
+	((j == 53))
+	((++i))
+done < <(n0 wg show wg0 allowed-ips)
+((i == 40))
+ip0 link del wg0
+ip0 link add wg0 type wireguard
+config=( )
+for i in {1..29}; do
+	config+=( "[Peer]" "PublicKey=$(wg genkey)" )
+done
+config+=( "[Peer]" "PublicKey=$(wg genkey)" "AllowedIPs=255.2.3.4/32,abcd::255/128" )
+n0 wg setconf wg0 <(printf '%s\n' "${config[@]}")
+n0 wg showconf wg0 > /dev/null
+ip0 link del wg0
+
+allowedips=( )
+for i in {1..197}; do
+        allowedips+=( abcd::$i )
+done
+saved_ifs="$IFS"
+IFS=,
+allowedips="${allowedips[*]}"
+IFS="$saved_ifs"
+ip0 link add wg0 type wireguard
+n0 wg set wg0 peer "$pub1"
+n0 wg set wg0 peer "$pub2" allowed-ips "$allowedips"
+{
+	read -r pub allowedips
+	[[ $pub == "$pub1" && $allowedips == "(none)" ]]
+	read -r pub allowedips
+	[[ $pub == "$pub2" ]]
+	i=0
+	for _ in $allowedips; do
+		((++i))
+	done
+	((i == 197))
+} < <(n0 wg show wg0 allowed-ips)
+ip0 link del wg0
+
+! n0 wg show doesnotexist || false
+
+ip0 link add wg0 type wireguard
+n0 wg set wg0 private-key <(echo "$key1") peer "$pub2" preshared-key <(echo "$psk")
+[[ $(n0 wg show wg0 private-key) == "$key1" ]]
+[[ $(n0 wg show wg0 preshared-keys) == "$pub2	$psk" ]]
+n0 wg set wg0 private-key /dev/null peer "$pub2" preshared-key /dev/null
+[[ $(n0 wg show wg0 private-key) == "(none)" ]]
+[[ $(n0 wg show wg0 preshared-keys) == "$pub2	(none)" ]]
+n0 wg set wg0 peer "$pub2"
+n0 wg set wg0 private-key <(echo "$key2")
+[[ $(n0 wg show wg0 public-key) == "$pub2" ]]
+[[ -z $(n0 wg show wg0 peers) ]]
+n0 wg set wg0 peer "$pub2"
+[[ -z $(n0 wg show wg0 peers) ]]
+n0 wg set wg0 private-key <(echo "$key1")
+n0 wg set wg0 peer "$pub2"
+[[ $(n0 wg show wg0 peers) == "$pub2" ]]
+n0 wg set wg0 private-key <(echo "/${key1:1}")
+[[ $(n0 wg show wg0 private-key) == "+${key1:1}" ]]
+n0 wg set wg0 peer "$pub2" allowed-ips 0.0.0.0/0,10.0.0.0/8,100.0.0.0/10,172.16.0.0/12,192.168.0.0/16
+n0 wg set wg0 peer "$pub2" allowed-ips 0.0.0.0/0
+n0 wg set wg0 peer "$pub2" allowed-ips ::/0,1700::/111,5000::/4,e000::/37,9000::/75
+n0 wg set wg0 peer "$pub2" allowed-ips ::/0
+ip0 link del wg0
+
+declare -A objects
+while read -t 0.1 -r line 2>/dev/null || [[ $? -ne 142 ]]; do
+	[[ $line =~ .*(wg[0-9]+:\ [A-Z][a-z]+\ [0-9]+)\ .*(created|destroyed).* ]] || continue
+	objects["${BASH_REMATCH[1]}"]+="${BASH_REMATCH[2]}"
+done < /dev/kmsg
+alldeleted=1
+for object in "${!objects[@]}"; do
+	if [[ ${objects["$object"]} != *createddestroyed ]]; then
+		echo "Error: $object: merely ${objects["$object"]}" >&3
+		alldeleted=0
+	fi
+done
+[[ $alldeleted -eq 1 ]]
+pretty "" "Objects that were created were also destroyed."
-- 
cgit v1.2.3-59-g8ed1b


From b50b0580d27bc45a0637aefc8bac4d31aa85771a Mon Sep 17 00:00:00 2001
From: Sabrina Dubroca <sd@queasysnail.net>
Date: Mon, 25 Nov 2019 14:48:57 +0100
Subject: net: add queue argument to __skb_wait_for_more_packets and
 __skb_{,try_}recv_datagram

This will be used by ESP over TCP to handle the queue of IKE messages.

Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
---
 include/linux/skbuff.h | 11 ++++++++---
 net/core/datagram.c    | 27 +++++++++++++++++----------
 net/ipv4/udp.c         |  3 ++-
 net/unix/af_unix.c     |  7 ++++---
 4 files changed, 31 insertions(+), 17 deletions(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index e9133bcf0544..49a10f9cc538 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -3459,7 +3459,8 @@ static inline void skb_frag_list_init(struct sk_buff *skb)
 	for (iter = skb_shinfo(skb)->frag_list; iter; iter = iter->next)
 
 
-int __skb_wait_for_more_packets(struct sock *sk, int *err, long *timeo_p,
+int __skb_wait_for_more_packets(struct sock *sk, struct sk_buff_head *queue,
+				int *err, long *timeo_p,
 				const struct sk_buff *skb);
 struct sk_buff *__skb_try_recv_from_queue(struct sock *sk,
 					  struct sk_buff_head *queue,
@@ -3468,12 +3469,16 @@ struct sk_buff *__skb_try_recv_from_queue(struct sock *sk,
 							   struct sk_buff *skb),
 					  int *off, int *err,
 					  struct sk_buff **last);
-struct sk_buff *__skb_try_recv_datagram(struct sock *sk, unsigned flags,
+struct sk_buff *__skb_try_recv_datagram(struct sock *sk,
+					struct sk_buff_head *queue,
+					unsigned int flags,
 					void (*destructor)(struct sock *sk,
 							   struct sk_buff *skb),
 					int *off, int *err,
 					struct sk_buff **last);
-struct sk_buff *__skb_recv_datagram(struct sock *sk, unsigned flags,
+struct sk_buff *__skb_recv_datagram(struct sock *sk,
+				    struct sk_buff_head *sk_queue,
+				    unsigned int flags,
 				    void (*destructor)(struct sock *sk,
 						       struct sk_buff *skb),
 				    int *off, int *err);
diff --git a/net/core/datagram.c b/net/core/datagram.c
index da3c24ed129c..a78e7f864c1e 100644
--- a/net/core/datagram.c
+++ b/net/core/datagram.c
@@ -84,7 +84,8 @@ static int receiver_wake_function(wait_queue_entry_t *wait, unsigned int mode, i
 /*
  * Wait for the last received packet to be different from skb
  */
-int __skb_wait_for_more_packets(struct sock *sk, int *err, long *timeo_p,
+int __skb_wait_for_more_packets(struct sock *sk, struct sk_buff_head *queue,
+				int *err, long *timeo_p,
 				const struct sk_buff *skb)
 {
 	int error;
@@ -97,7 +98,7 @@ int __skb_wait_for_more_packets(struct sock *sk, int *err, long *timeo_p,
 	if (error)
 		goto out_err;
 
-	if (READ_ONCE(sk->sk_receive_queue.prev) != skb)
+	if (READ_ONCE(queue->prev) != skb)
 		goto out;
 
 	/* Socket shut down? */
@@ -209,6 +210,7 @@ struct sk_buff *__skb_try_recv_from_queue(struct sock *sk,
 /**
  *	__skb_try_recv_datagram - Receive a datagram skbuff
  *	@sk: socket
+ *	@queue: socket queue from which to receive
  *	@flags: MSG\_ flags
  *	@destructor: invoked under the receive lock on successful dequeue
  *	@off: an offset in bytes to peek skb from. Returns an offset
@@ -241,13 +243,14 @@ struct sk_buff *__skb_try_recv_from_queue(struct sock *sk,
  *	quite explicitly by POSIX 1003.1g, don't change them without having
  *	the standard around please.
  */
-struct sk_buff *__skb_try_recv_datagram(struct sock *sk, unsigned int flags,
+struct sk_buff *__skb_try_recv_datagram(struct sock *sk,
+					struct sk_buff_head *queue,
+					unsigned int flags,
 					void (*destructor)(struct sock *sk,
 							   struct sk_buff *skb),
 					int *off, int *err,
 					struct sk_buff **last)
 {
-	struct sk_buff_head *queue = &sk->sk_receive_queue;
 	struct sk_buff *skb;
 	unsigned long cpu_flags;
 	/*
@@ -278,7 +281,7 @@ struct sk_buff *__skb_try_recv_datagram(struct sock *sk, unsigned int flags,
 			break;
 
 		sk_busy_loop(sk, flags & MSG_DONTWAIT);
-	} while (READ_ONCE(sk->sk_receive_queue.prev) != *last);
+	} while (READ_ONCE(queue->prev) != *last);
 
 	error = -EAGAIN;
 
@@ -288,7 +291,9 @@ no_packet:
 }
 EXPORT_SYMBOL(__skb_try_recv_datagram);
 
-struct sk_buff *__skb_recv_datagram(struct sock *sk, unsigned int flags,
+struct sk_buff *__skb_recv_datagram(struct sock *sk,
+				    struct sk_buff_head *sk_queue,
+				    unsigned int flags,
 				    void (*destructor)(struct sock *sk,
 						       struct sk_buff *skb),
 				    int *off, int *err)
@@ -299,15 +304,16 @@ struct sk_buff *__skb_recv_datagram(struct sock *sk, unsigned int flags,
 	timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT);
 
 	do {
-		skb = __skb_try_recv_datagram(sk, flags, destructor, off, err,
-					      &last);
+		skb = __skb_try_recv_datagram(sk, sk_queue, flags, destructor,
+					      off, err, &last);
 		if (skb)
 			return skb;
 
 		if (*err != -EAGAIN)
 			break;
 	} while (timeo &&
-		!__skb_wait_for_more_packets(sk, err, &timeo, last));
+		 !__skb_wait_for_more_packets(sk, sk_queue, err,
+					      &timeo, last));
 
 	return NULL;
 }
@@ -318,7 +324,8 @@ struct sk_buff *skb_recv_datagram(struct sock *sk, unsigned int flags,
 {
 	int off = 0;
 
-	return __skb_recv_datagram(sk, flags | (noblock ? MSG_DONTWAIT : 0),
+	return __skb_recv_datagram(sk, &sk->sk_receive_queue,
+				   flags | (noblock ? MSG_DONTWAIT : 0),
 				   NULL, &off, err);
 }
 EXPORT_SYMBOL(skb_recv_datagram);
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 4da5758cc718..e5738d1217a1 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1708,7 +1708,8 @@ busy_check:
 
 		/* sk_queue is empty, reader_queue may contain peeked packets */
 	} while (timeo &&
-		 !__skb_wait_for_more_packets(sk, &error, &timeo,
+		 !__skb_wait_for_more_packets(sk, &sk->sk_receive_queue,
+					      &error, &timeo,
 					      (struct sk_buff *)sk_queue));
 
 	*err = error;
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index 7cfdce10de36..a7f707fc4cac 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -2058,8 +2058,8 @@ static int unix_dgram_recvmsg(struct socket *sock, struct msghdr *msg,
 		mutex_lock(&u->iolock);
 
 		skip = sk_peek_offset(sk, flags);
-		skb = __skb_try_recv_datagram(sk, flags, NULL, &skip, &err,
-					      &last);
+		skb = __skb_try_recv_datagram(sk, &sk->sk_receive_queue, flags,
+					      NULL, &skip, &err, &last);
 		if (skb)
 			break;
 
@@ -2068,7 +2068,8 @@ static int unix_dgram_recvmsg(struct socket *sock, struct msghdr *msg,
 		if (err != -EAGAIN)
 			break;
 	} while (timeo &&
-		 !__skb_wait_for_more_packets(sk, &err, &timeo, last));
+		 !__skb_wait_for_more_packets(sk, &sk->sk_receive_queue,
+					      &err, &timeo, last));
 
 	if (!skb) { /* implies iolock unlocked */
 		unix_state_lock(sk);
-- 
cgit v1.2.3-59-g8ed1b


From 7b3801927e52f8621de311277f7fc727635019e7 Mon Sep 17 00:00:00 2001
From: Sabrina Dubroca <sd@queasysnail.net>
Date: Mon, 25 Nov 2019 14:48:58 +0100
Subject: xfrm: introduce xfrm_trans_queue_net

This will be used by TCP encapsulation to write packets to the encap
socket without holding the user socket's lock. Without this reinjection,
we're already holding the lock of the user socket, and then try to lock
the encap socket as well when we enqueue the encrypted packet.

While at it, add a BUILD_BUG_ON like we usually do for skb->cb, since
it's missing for struct xfrm_trans_cb.

Co-developed-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
---
 include/net/xfrm.h    |  3 +++
 net/xfrm/xfrm_input.c | 21 +++++++++++++++++----
 2 files changed, 20 insertions(+), 4 deletions(-)

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index dda3c025452e..56ff86621bb4 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -1547,6 +1547,9 @@ int __xfrm_init_state(struct xfrm_state *x, bool init_replay, bool offload);
 int xfrm_init_state(struct xfrm_state *x);
 int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type);
 int xfrm_input_resume(struct sk_buff *skb, int nexthdr);
+int xfrm_trans_queue_net(struct net *net, struct sk_buff *skb,
+			 int (*finish)(struct net *, struct sock *,
+				       struct sk_buff *));
 int xfrm_trans_queue(struct sk_buff *skb,
 		     int (*finish)(struct net *, struct sock *,
 				   struct sk_buff *));
diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c
index 2c86a2fc3915..aa35f23c4912 100644
--- a/net/xfrm/xfrm_input.c
+++ b/net/xfrm/xfrm_input.c
@@ -36,6 +36,7 @@ struct xfrm_trans_cb {
 #endif
 	} header;
 	int (*finish)(struct net *net, struct sock *sk, struct sk_buff *skb);
+	struct net *net;
 };
 
 #define XFRM_TRANS_SKB_CB(__skb) ((struct xfrm_trans_cb *)&((__skb)->cb[0]))
@@ -766,12 +767,13 @@ static void xfrm_trans_reinject(unsigned long data)
 	skb_queue_splice_init(&trans->queue, &queue);
 
 	while ((skb = __skb_dequeue(&queue)))
-		XFRM_TRANS_SKB_CB(skb)->finish(dev_net(skb->dev), NULL, skb);
+		XFRM_TRANS_SKB_CB(skb)->finish(XFRM_TRANS_SKB_CB(skb)->net,
+					       NULL, skb);
 }
 
-int xfrm_trans_queue(struct sk_buff *skb,
-		     int (*finish)(struct net *, struct sock *,
-				   struct sk_buff *))
+int xfrm_trans_queue_net(struct net *net, struct sk_buff *skb,
+			 int (*finish)(struct net *, struct sock *,
+				       struct sk_buff *))
 {
 	struct xfrm_trans_tasklet *trans;
 
@@ -780,11 +782,22 @@ int xfrm_trans_queue(struct sk_buff *skb,
 	if (skb_queue_len(&trans->queue) >= netdev_max_backlog)
 		return -ENOBUFS;
 
+	BUILD_BUG_ON(sizeof(struct xfrm_trans_cb) > sizeof(skb->cb));
+
 	XFRM_TRANS_SKB_CB(skb)->finish = finish;
+	XFRM_TRANS_SKB_CB(skb)->net = net;
 	__skb_queue_tail(&trans->queue, skb);
 	tasklet_schedule(&trans->tasklet);
 	return 0;
 }
+EXPORT_SYMBOL(xfrm_trans_queue_net);
+
+int xfrm_trans_queue(struct sk_buff *skb,
+		     int (*finish)(struct net *, struct sock *,
+				   struct sk_buff *))
+{
+	return xfrm_trans_queue_net(dev_net(skb->dev), skb, finish);
+}
 EXPORT_SYMBOL(xfrm_trans_queue);
 
 void __init xfrm_input_init(void)
-- 
cgit v1.2.3-59-g8ed1b


From cac3c71604cf4eada8df00b1e66892636399cda5 Mon Sep 17 00:00:00 2001
From: Sabrina Dubroca <sd@queasysnail.net>
Date: Mon, 25 Nov 2019 14:48:59 +0100
Subject: xfrm: add route lookup to xfrm4_rcv_encap

At this point, with TCP encapsulation, the dst may be gone, but
xfrm_input needs one.

Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
---
 net/ipv4/xfrm4_protocol.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/net/ipv4/xfrm4_protocol.c b/net/ipv4/xfrm4_protocol.c
index 8a4285712808..ea595c8549c7 100644
--- a/net/ipv4/xfrm4_protocol.c
+++ b/net/ipv4/xfrm4_protocol.c
@@ -72,6 +72,14 @@ int xfrm4_rcv_encap(struct sk_buff *skb, int nexthdr, __be32 spi,
 	if (!head)
 		goto out;
 
+	if (!skb_dst(skb)) {
+		const struct iphdr *iph = ip_hdr(skb);
+
+		if (ip_route_input_noref(skb, iph->daddr, iph->saddr,
+					 iph->tos, skb->dev))
+			goto drop;
+	}
+
 	for_each_protocol_rcu(*head, handler)
 		if ((ret = handler->input_handler(skb, nexthdr, spi, encap_type)) != -EINVAL)
 			return ret;
@@ -79,6 +87,7 @@ int xfrm4_rcv_encap(struct sk_buff *skb, int nexthdr, __be32 spi,
 out:
 	icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0);
 
+drop:
 	kfree_skb(skb);
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 25f6802b4c18817c82cd581d38ce155ad6412176 Mon Sep 17 00:00:00 2001
From: Sabrina Dubroca <sd@queasysnail.net>
Date: Mon, 25 Nov 2019 14:49:00 +0100
Subject: esp4: prepare esp_input_done2 for non-UDP encapsulation

For espintcp encapsulation, we will need to get the source port from the
TCP header instead of UDP. Introduce a variable to hold the port.

Co-developed-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
---
 net/ipv4/esp4.c | 16 ++++++++++++++--
 1 file changed, 14 insertions(+), 2 deletions(-)

diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c
index 5c967764041f..c5d826642229 100644
--- a/net/ipv4/esp4.c
+++ b/net/ipv4/esp4.c
@@ -601,6 +601,18 @@ int esp_input_done2(struct sk_buff *skb, int err)
 	if (x->encap) {
 		struct xfrm_encap_tmpl *encap = x->encap;
 		struct udphdr *uh = (void *)(skb_network_header(skb) + ihl);
+		__be16 source;
+
+		switch (x->encap->encap_type) {
+		case UDP_ENCAP_ESPINUDP:
+		case UDP_ENCAP_ESPINUDP_NON_IKE:
+			source = uh->source;
+			break;
+		default:
+			WARN_ON_ONCE(1);
+			err = -EINVAL;
+			goto out;
+		}
 
 		/*
 		 * 1) if the NAT-T peer's IP or port changed then
@@ -609,11 +621,11 @@ int esp_input_done2(struct sk_buff *skb, int err)
 		 *    SRC ports.
 		 */
 		if (iph->saddr != x->props.saddr.a4 ||
-		    uh->source != encap->encap_sport) {
+		    source != encap->encap_sport) {
 			xfrm_address_t ipaddr;
 
 			ipaddr.a4 = iph->saddr;
-			km_new_mapping(x, &ipaddr, uh->source);
+			km_new_mapping(x, &ipaddr, source);
 
 			/* XXX: perhaps add an extra
 			 * policy check here, to see
-- 
cgit v1.2.3-59-g8ed1b


From eecd227a9a3479038ba2a2f579b3ce9edb364b80 Mon Sep 17 00:00:00 2001
From: Sabrina Dubroca <sd@queasysnail.net>
Date: Mon, 25 Nov 2019 14:49:01 +0100
Subject: esp4: split esp_output_udp_encap and introduce esp_output_encap

Co-developed-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
---
 net/ipv4/esp4.c | 57 +++++++++++++++++++++++++++++++++++++--------------------
 1 file changed, 37 insertions(+), 20 deletions(-)

diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c
index c5d826642229..033c61d27148 100644
--- a/net/ipv4/esp4.c
+++ b/net/ipv4/esp4.c
@@ -225,45 +225,62 @@ static void esp_output_fill_trailer(u8 *tail, int tfclen, int plen, __u8 proto)
 	tail[plen - 1] = proto;
 }
 
-static int esp_output_udp_encap(struct xfrm_state *x, struct sk_buff *skb, struct esp_info *esp)
+static struct ip_esp_hdr *esp_output_udp_encap(struct sk_buff *skb,
+					       int encap_type,
+					       struct esp_info *esp,
+					       __be16 sport,
+					       __be16 dport)
 {
-	int encap_type;
 	struct udphdr *uh;
 	__be32 *udpdata32;
-	__be16 sport, dport;
-	struct xfrm_encap_tmpl *encap = x->encap;
-	struct ip_esp_hdr *esph = esp->esph;
 	unsigned int len;
 
-	spin_lock_bh(&x->lock);
-	sport = encap->encap_sport;
-	dport = encap->encap_dport;
-	encap_type = encap->encap_type;
-	spin_unlock_bh(&x->lock);
-
 	len = skb->len + esp->tailen - skb_transport_offset(skb);
 	if (len + sizeof(struct iphdr) >= IP_MAX_MTU)
-		return -EMSGSIZE;
+		return ERR_PTR(-EMSGSIZE);
 
-	uh = (struct udphdr *)esph;
+	uh = (struct udphdr *)esp->esph;
 	uh->source = sport;
 	uh->dest = dport;
 	uh->len = htons(len);
 	uh->check = 0;
 
+	*skb_mac_header(skb) = IPPROTO_UDP;
+
+	if (encap_type == UDP_ENCAP_ESPINUDP_NON_IKE) {
+		udpdata32 = (__be32 *)(uh + 1);
+		udpdata32[0] = udpdata32[1] = 0;
+		return (struct ip_esp_hdr *)(udpdata32 + 2);
+	}
+
+	return (struct ip_esp_hdr *)(uh + 1);
+}
+
+static int esp_output_encap(struct xfrm_state *x, struct sk_buff *skb,
+			    struct esp_info *esp)
+{
+	struct xfrm_encap_tmpl *encap = x->encap;
+	struct ip_esp_hdr *esph;
+	__be16 sport, dport;
+	int encap_type;
+
+	spin_lock_bh(&x->lock);
+	sport = encap->encap_sport;
+	dport = encap->encap_dport;
+	encap_type = encap->encap_type;
+	spin_unlock_bh(&x->lock);
+
 	switch (encap_type) {
 	default:
 	case UDP_ENCAP_ESPINUDP:
-		esph = (struct ip_esp_hdr *)(uh + 1);
-		break;
 	case UDP_ENCAP_ESPINUDP_NON_IKE:
-		udpdata32 = (__be32 *)(uh + 1);
-		udpdata32[0] = udpdata32[1] = 0;
-		esph = (struct ip_esp_hdr *)(udpdata32 + 2);
+		esph = esp_output_udp_encap(skb, encap_type, esp, sport, dport);
 		break;
 	}
 
-	*skb_mac_header(skb) = IPPROTO_UDP;
+	if (IS_ERR(esph))
+		return PTR_ERR(esph);
+
 	esp->esph = esph;
 
 	return 0;
@@ -281,7 +298,7 @@ int esp_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info *
 
 	/* this is non-NULL only with UDP Encapsulation */
 	if (x->encap) {
-		int err = esp_output_udp_encap(x, skb, esp);
+		int err = esp_output_encap(x, skb, esp);
 
 		if (err < 0)
 			return err;
-- 
cgit v1.2.3-59-g8ed1b


From e27cca96cd68fa2c6814c90f9a1cfd36bb68c593 Mon Sep 17 00:00:00 2001
From: Sabrina Dubroca <sd@queasysnail.net>
Date: Mon, 25 Nov 2019 14:49:02 +0100
Subject: xfrm: add espintcp (RFC 8229)

TCP encapsulation of IKE and IPsec messages (RFC 8229) is implemented
as a TCP ULP, overriding in particular the sendmsg and recvmsg
operations. A Stream Parser is used to extract messages out of the TCP
stream using the first 2 bytes as length marker. Received IKE messages
are put on "ike_queue", waiting to be dequeued by the custom recvmsg
implementation. Received ESP messages are sent to XFRM, like with UDP
encapsulation.

Some of this code is taken from the original submission by Herbert
Xu. Currently, only IPv4 is supported, like for UDP encapsulation.

Co-developed-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
---
 include/net/espintcp.h   |  39 ++++
 include/net/xfrm.h       |   1 +
 include/uapi/linux/udp.h |   1 +
 net/ipv4/Kconfig         |  11 +
 net/ipv4/esp4.c          | 191 +++++++++++++++++-
 net/xfrm/Makefile        |   1 +
 net/xfrm/espintcp.c      | 509 +++++++++++++++++++++++++++++++++++++++++++++++
 net/xfrm/xfrm_policy.c   |   7 +
 net/xfrm/xfrm_state.c    |   3 +
 9 files changed, 760 insertions(+), 3 deletions(-)
 create mode 100644 include/net/espintcp.h
 create mode 100644 net/xfrm/espintcp.c

diff --git a/include/net/espintcp.h b/include/net/espintcp.h
new file mode 100644
index 000000000000..dd7026a00066
--- /dev/null
+++ b/include/net/espintcp.h
@@ -0,0 +1,39 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _NET_ESPINTCP_H
+#define _NET_ESPINTCP_H
+
+#include <net/strparser.h>
+#include <linux/skmsg.h>
+
+void __init espintcp_init(void);
+
+int espintcp_push_skb(struct sock *sk, struct sk_buff *skb);
+int espintcp_queue_out(struct sock *sk, struct sk_buff *skb);
+bool tcp_is_ulp_esp(struct sock *sk);
+
+struct espintcp_msg {
+	struct sk_buff *skb;
+	struct sk_msg skmsg;
+	int offset;
+	int len;
+};
+
+struct espintcp_ctx {
+	struct strparser strp;
+	struct sk_buff_head ike_queue;
+	struct sk_buff_head out_queue;
+	struct espintcp_msg partial;
+	void (*saved_data_ready)(struct sock *sk);
+	void (*saved_write_space)(struct sock *sk);
+	struct work_struct work;
+	bool tx_running;
+};
+
+static inline struct espintcp_ctx *espintcp_getctx(const struct sock *sk)
+{
+	struct inet_connection_sock *icsk = inet_csk(sk);
+
+	/* RCU is only needed for diag */
+	return (__force void *)icsk->icsk_ulp_data;
+}
+#endif
diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 56ff86621bb4..8f71c111e65a 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -193,6 +193,7 @@ struct xfrm_state {
 
 	/* Data for encapsulator */
 	struct xfrm_encap_tmpl	*encap;
+	struct sock __rcu	*encap_sk;
 
 	/* Data for care-of address */
 	xfrm_address_t	*coaddr;
diff --git a/include/uapi/linux/udp.h b/include/uapi/linux/udp.h
index 30baccb6c9c4..4828794efcf8 100644
--- a/include/uapi/linux/udp.h
+++ b/include/uapi/linux/udp.h
@@ -42,5 +42,6 @@ struct udphdr {
 #define UDP_ENCAP_GTP0		4 /* GSM TS 09.60 */
 #define UDP_ENCAP_GTP1U		5 /* 3GPP TS 29.060 */
 #define UDP_ENCAP_RXRPC		6
+#define TCP_ENCAP_ESPINTCP	7 /* Yikes, this is really xfrm encap types. */
 
 #endif /* _UAPI_LINUX_UDP_H */
diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig
index fc816b187170..f96bd489b362 100644
--- a/net/ipv4/Kconfig
+++ b/net/ipv4/Kconfig
@@ -378,6 +378,17 @@ config INET_ESP_OFFLOAD
 
 	  If unsure, say N.
 
+config INET_ESPINTCP
+	bool "IP: ESP in TCP encapsulation (RFC 8229)"
+	depends on XFRM && INET_ESP
+	select STREAM_PARSER
+	select NET_SOCK_MSG
+	help
+	  Support for RFC 8229 encapsulation of ESP and IKE over
+	  TCP/IPv4 sockets.
+
+	  If unsure, say N.
+
 config INET_IPCOMP
 	tristate "IP: IPComp transformation"
 	select INET_XFRM_TUNNEL
diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c
index 033c61d27148..103c7d599a3c 100644
--- a/net/ipv4/esp4.c
+++ b/net/ipv4/esp4.c
@@ -18,6 +18,8 @@
 #include <net/icmp.h>
 #include <net/protocol.h>
 #include <net/udp.h>
+#include <net/tcp.h>
+#include <net/espintcp.h>
 
 #include <linux/highmem.h>
 
@@ -117,6 +119,132 @@ static void esp_ssg_unref(struct xfrm_state *x, void *tmp)
 			put_page(sg_page(sg));
 }
 
+#ifdef CONFIG_INET_ESPINTCP
+struct esp_tcp_sk {
+	struct sock *sk;
+	struct rcu_head rcu;
+};
+
+static void esp_free_tcp_sk(struct rcu_head *head)
+{
+	struct esp_tcp_sk *esk = container_of(head, struct esp_tcp_sk, rcu);
+
+	sock_put(esk->sk);
+	kfree(esk);
+}
+
+static struct sock *esp_find_tcp_sk(struct xfrm_state *x)
+{
+	struct xfrm_encap_tmpl *encap = x->encap;
+	struct esp_tcp_sk *esk;
+	__be16 sport, dport;
+	struct sock *nsk;
+	struct sock *sk;
+
+	sk = rcu_dereference(x->encap_sk);
+	if (sk && sk->sk_state == TCP_ESTABLISHED)
+		return sk;
+
+	spin_lock_bh(&x->lock);
+	sport = encap->encap_sport;
+	dport = encap->encap_dport;
+	nsk = rcu_dereference_protected(x->encap_sk,
+					lockdep_is_held(&x->lock));
+	if (sk && sk == nsk) {
+		esk = kmalloc(sizeof(*esk), GFP_ATOMIC);
+		if (!esk) {
+			spin_unlock_bh(&x->lock);
+			return ERR_PTR(-ENOMEM);
+		}
+		RCU_INIT_POINTER(x->encap_sk, NULL);
+		esk->sk = sk;
+		call_rcu(&esk->rcu, esp_free_tcp_sk);
+	}
+	spin_unlock_bh(&x->lock);
+
+	sk = inet_lookup_established(xs_net(x), &tcp_hashinfo, x->id.daddr.a4,
+				     dport, x->props.saddr.a4, sport, 0);
+	if (!sk)
+		return ERR_PTR(-ENOENT);
+
+	if (!tcp_is_ulp_esp(sk)) {
+		sock_put(sk);
+		return ERR_PTR(-EINVAL);
+	}
+
+	spin_lock_bh(&x->lock);
+	nsk = rcu_dereference_protected(x->encap_sk,
+					lockdep_is_held(&x->lock));
+	if (encap->encap_sport != sport ||
+	    encap->encap_dport != dport) {
+		sock_put(sk);
+		sk = nsk ?: ERR_PTR(-EREMCHG);
+	} else if (sk == nsk) {
+		sock_put(sk);
+	} else {
+		rcu_assign_pointer(x->encap_sk, sk);
+	}
+	spin_unlock_bh(&x->lock);
+
+	return sk;
+}
+
+static int esp_output_tcp_finish(struct xfrm_state *x, struct sk_buff *skb)
+{
+	struct sock *sk;
+	int err;
+
+	rcu_read_lock();
+
+	sk = esp_find_tcp_sk(x);
+	err = PTR_ERR_OR_ZERO(sk);
+	if (err)
+		goto out;
+
+	bh_lock_sock(sk);
+	if (sock_owned_by_user(sk))
+		err = espintcp_queue_out(sk, skb);
+	else
+		err = espintcp_push_skb(sk, skb);
+	bh_unlock_sock(sk);
+
+out:
+	rcu_read_unlock();
+	return err;
+}
+
+static int esp_output_tcp_encap_cb(struct net *net, struct sock *sk,
+				   struct sk_buff *skb)
+{
+	struct dst_entry *dst = skb_dst(skb);
+	struct xfrm_state *x = dst->xfrm;
+
+	return esp_output_tcp_finish(x, skb);
+}
+
+static int esp_output_tail_tcp(struct xfrm_state *x, struct sk_buff *skb)
+{
+	int err;
+
+	local_bh_disable();
+	err = xfrm_trans_queue_net(xs_net(x), skb, esp_output_tcp_encap_cb);
+	local_bh_enable();
+
+	/* EINPROGRESS just happens to do the right thing.  It
+	 * actually means that the skb has been consumed and
+	 * isn't coming back.
+	 */
+	return err ?: -EINPROGRESS;
+}
+#else
+static int esp_output_tail_tcp(struct xfrm_state *x, struct sk_buff *skb)
+{
+	kfree_skb(skb);
+
+	return -EOPNOTSUPP;
+}
+#endif
+
 static void esp_output_done(struct crypto_async_request *base, int err)
 {
 	struct sk_buff *skb = base->data;
@@ -147,7 +275,11 @@ static void esp_output_done(struct crypto_async_request *base, int err)
 		secpath_reset(skb);
 		xfrm_dev_resume(skb);
 	} else {
-		xfrm_output_resume(skb, err);
+		if (!err &&
+		    x->encap && x->encap->encap_type == TCP_ENCAP_ESPINTCP)
+			esp_output_tail_tcp(x, skb);
+		else
+			xfrm_output_resume(skb, err);
 	}
 }
 
@@ -236,7 +368,7 @@ static struct ip_esp_hdr *esp_output_udp_encap(struct sk_buff *skb,
 	unsigned int len;
 
 	len = skb->len + esp->tailen - skb_transport_offset(skb);
-	if (len + sizeof(struct iphdr) >= IP_MAX_MTU)
+	if (len + sizeof(struct iphdr) > IP_MAX_MTU)
 		return ERR_PTR(-EMSGSIZE);
 
 	uh = (struct udphdr *)esp->esph;
@@ -256,6 +388,41 @@ static struct ip_esp_hdr *esp_output_udp_encap(struct sk_buff *skb,
 	return (struct ip_esp_hdr *)(uh + 1);
 }
 
+#ifdef CONFIG_INET_ESPINTCP
+static struct ip_esp_hdr *esp_output_tcp_encap(struct xfrm_state *x,
+						    struct sk_buff *skb,
+						    struct esp_info *esp)
+{
+	__be16 *lenp = (void *)esp->esph;
+	struct ip_esp_hdr *esph;
+	unsigned int len;
+	struct sock *sk;
+
+	len = skb->len + esp->tailen - skb_transport_offset(skb);
+	if (len > IP_MAX_MTU)
+		return ERR_PTR(-EMSGSIZE);
+
+	rcu_read_lock();
+	sk = esp_find_tcp_sk(x);
+	rcu_read_unlock();
+
+	if (IS_ERR(sk))
+		return ERR_CAST(sk);
+
+	*lenp = htons(len);
+	esph = (struct ip_esp_hdr *)(lenp + 1);
+
+	return esph;
+}
+#else
+static struct ip_esp_hdr *esp_output_tcp_encap(struct xfrm_state *x,
+						    struct sk_buff *skb,
+						    struct esp_info *esp)
+{
+	return ERR_PTR(-EOPNOTSUPP);
+}
+#endif
+
 static int esp_output_encap(struct xfrm_state *x, struct sk_buff *skb,
 			    struct esp_info *esp)
 {
@@ -276,6 +443,9 @@ static int esp_output_encap(struct xfrm_state *x, struct sk_buff *skb,
 	case UDP_ENCAP_ESPINUDP_NON_IKE:
 		esph = esp_output_udp_encap(skb, encap_type, esp, sport, dport);
 		break;
+	case TCP_ENCAP_ESPINTCP:
+		esph = esp_output_tcp_encap(x, skb, esp);
+		break;
 	}
 
 	if (IS_ERR(esph))
@@ -296,7 +466,7 @@ int esp_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info *
 	struct sk_buff *trailer;
 	int tailen = esp->tailen;
 
-	/* this is non-NULL only with UDP Encapsulation */
+	/* this is non-NULL only with TCP/UDP Encapsulation */
 	if (x->encap) {
 		int err = esp_output_encap(x, skb, esp);
 
@@ -491,6 +661,9 @@ int esp_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info *
 	if (sg != dsg)
 		esp_ssg_unref(x, tmp);
 
+	if (!err && x->encap && x->encap->encap_type == TCP_ENCAP_ESPINTCP)
+		err = esp_output_tail_tcp(x, skb);
+
 error_free:
 	kfree(tmp);
 error:
@@ -617,10 +790,14 @@ int esp_input_done2(struct sk_buff *skb, int err)
 
 	if (x->encap) {
 		struct xfrm_encap_tmpl *encap = x->encap;
+		struct tcphdr *th = (void *)(skb_network_header(skb) + ihl);
 		struct udphdr *uh = (void *)(skb_network_header(skb) + ihl);
 		__be16 source;
 
 		switch (x->encap->encap_type) {
+		case TCP_ENCAP_ESPINTCP:
+			source = th->source;
+			break;
 		case UDP_ENCAP_ESPINUDP:
 		case UDP_ENCAP_ESPINUDP_NON_IKE:
 			source = uh->source;
@@ -1017,6 +1194,14 @@ static int esp_init_state(struct xfrm_state *x)
 		case UDP_ENCAP_ESPINUDP_NON_IKE:
 			x->props.header_len += sizeof(struct udphdr) + 2 * sizeof(u32);
 			break;
+#ifdef CONFIG_INET_ESPINTCP
+		case TCP_ENCAP_ESPINTCP:
+			/* only the length field, TCP encap is done by
+			 * the socket
+			 */
+			x->props.header_len += 2;
+			break;
+#endif
 		}
 	}
 
diff --git a/net/xfrm/Makefile b/net/xfrm/Makefile
index fbc4552d17b8..212a4fcb4a88 100644
--- a/net/xfrm/Makefile
+++ b/net/xfrm/Makefile
@@ -11,3 +11,4 @@ obj-$(CONFIG_XFRM_ALGO) += xfrm_algo.o
 obj-$(CONFIG_XFRM_USER) += xfrm_user.o
 obj-$(CONFIG_XFRM_IPCOMP) += xfrm_ipcomp.o
 obj-$(CONFIG_XFRM_INTERFACE) += xfrm_interface.o
+obj-$(CONFIG_INET_ESPINTCP) += espintcp.o
diff --git a/net/xfrm/espintcp.c b/net/xfrm/espintcp.c
new file mode 100644
index 000000000000..f15d6a564b0e
--- /dev/null
+++ b/net/xfrm/espintcp.c
@@ -0,0 +1,509 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <net/tcp.h>
+#include <net/strparser.h>
+#include <net/xfrm.h>
+#include <net/esp.h>
+#include <net/espintcp.h>
+#include <linux/skmsg.h>
+#include <net/inet_common.h>
+
+static void handle_nonesp(struct espintcp_ctx *ctx, struct sk_buff *skb,
+			  struct sock *sk)
+{
+	if (atomic_read(&sk->sk_rmem_alloc) >= sk->sk_rcvbuf ||
+	    !sk_rmem_schedule(sk, skb, skb->truesize)) {
+		kfree_skb(skb);
+		return;
+	}
+
+	skb_set_owner_r(skb, sk);
+
+	memset(skb->cb, 0, sizeof(skb->cb));
+	skb_queue_tail(&ctx->ike_queue, skb);
+	ctx->saved_data_ready(sk);
+}
+
+static void handle_esp(struct sk_buff *skb, struct sock *sk)
+{
+	skb_reset_transport_header(skb);
+	memset(skb->cb, 0, sizeof(skb->cb));
+
+	rcu_read_lock();
+	skb->dev = dev_get_by_index_rcu(sock_net(sk), skb->skb_iif);
+	local_bh_disable();
+	xfrm4_rcv_encap(skb, IPPROTO_ESP, 0, TCP_ENCAP_ESPINTCP);
+	local_bh_enable();
+	rcu_read_unlock();
+}
+
+static void espintcp_rcv(struct strparser *strp, struct sk_buff *skb)
+{
+	struct espintcp_ctx *ctx = container_of(strp, struct espintcp_ctx,
+						strp);
+	struct strp_msg *rxm = strp_msg(skb);
+	u32 nonesp_marker;
+	int err;
+
+	err = skb_copy_bits(skb, rxm->offset + 2, &nonesp_marker,
+			    sizeof(nonesp_marker));
+	if (err < 0) {
+		kfree_skb(skb);
+		return;
+	}
+
+	/* remove header, leave non-ESP marker/SPI */
+	if (!__pskb_pull(skb, rxm->offset + 2)) {
+		kfree_skb(skb);
+		return;
+	}
+
+	if (pskb_trim(skb, rxm->full_len - 2) != 0) {
+		kfree_skb(skb);
+		return;
+	}
+
+	if (nonesp_marker == 0)
+		handle_nonesp(ctx, skb, strp->sk);
+	else
+		handle_esp(skb, strp->sk);
+}
+
+static int espintcp_parse(struct strparser *strp, struct sk_buff *skb)
+{
+	struct strp_msg *rxm = strp_msg(skb);
+	__be16 blen;
+	u16 len;
+	int err;
+
+	if (skb->len < rxm->offset + 2)
+		return 0;
+
+	err = skb_copy_bits(skb, rxm->offset, &blen, sizeof(blen));
+	if (err < 0)
+		return err;
+
+	len = be16_to_cpu(blen);
+	if (len < 6)
+		return -EINVAL;
+
+	return len;
+}
+
+static int espintcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
+			    int nonblock, int flags, int *addr_len)
+{
+	struct espintcp_ctx *ctx = espintcp_getctx(sk);
+	struct sk_buff *skb;
+	int err = 0;
+	int copied;
+	int off = 0;
+
+	flags |= nonblock ? MSG_DONTWAIT : 0;
+
+	skb = __skb_recv_datagram(sk, &ctx->ike_queue, flags, NULL, &off, &err);
+	if (!skb)
+		return err;
+
+	copied = len;
+	if (copied > skb->len)
+		copied = skb->len;
+	else if (copied < skb->len)
+		msg->msg_flags |= MSG_TRUNC;
+
+	err = skb_copy_datagram_msg(skb, 0, msg, copied);
+	if (unlikely(err)) {
+		kfree_skb(skb);
+		return err;
+	}
+
+	if (flags & MSG_TRUNC)
+		copied = skb->len;
+	kfree_skb(skb);
+	return copied;
+}
+
+int espintcp_queue_out(struct sock *sk, struct sk_buff *skb)
+{
+	struct espintcp_ctx *ctx = espintcp_getctx(sk);
+
+	if (skb_queue_len(&ctx->out_queue) >= netdev_max_backlog)
+		return -ENOBUFS;
+
+	__skb_queue_tail(&ctx->out_queue, skb);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(espintcp_queue_out);
+
+/* espintcp length field is 2B and length includes the length field's size */
+#define MAX_ESPINTCP_MSG (((1 << 16) - 1) - 2)
+
+static int espintcp_sendskb_locked(struct sock *sk, struct espintcp_msg *emsg,
+				   int flags)
+{
+	do {
+		int ret;
+
+		ret = skb_send_sock_locked(sk, emsg->skb,
+					   emsg->offset, emsg->len);
+		if (ret < 0)
+			return ret;
+
+		emsg->len -= ret;
+		emsg->offset += ret;
+	} while (emsg->len > 0);
+
+	kfree_skb(emsg->skb);
+	memset(emsg, 0, sizeof(*emsg));
+
+	return 0;
+}
+
+static int espintcp_sendskmsg_locked(struct sock *sk,
+				     struct espintcp_msg *emsg, int flags)
+{
+	struct sk_msg *skmsg = &emsg->skmsg;
+	struct scatterlist *sg;
+	int done = 0;
+	int ret;
+
+	flags |= MSG_SENDPAGE_NOTLAST;
+	sg = &skmsg->sg.data[skmsg->sg.start];
+	do {
+		size_t size = sg->length - emsg->offset;
+		int offset = sg->offset + emsg->offset;
+		struct page *p;
+
+		emsg->offset = 0;
+
+		if (sg_is_last(sg))
+			flags &= ~MSG_SENDPAGE_NOTLAST;
+
+		p = sg_page(sg);
+retry:
+		ret = do_tcp_sendpages(sk, p, offset, size, flags);
+		if (ret < 0) {
+			emsg->offset = offset - sg->offset;
+			skmsg->sg.start += done;
+			return ret;
+		}
+
+		if (ret != size) {
+			offset += ret;
+			size -= ret;
+			goto retry;
+		}
+
+		done++;
+		put_page(p);
+		sk_mem_uncharge(sk, sg->length);
+		sg = sg_next(sg);
+	} while (sg);
+
+	memset(emsg, 0, sizeof(*emsg));
+
+	return 0;
+}
+
+static int espintcp_push_msgs(struct sock *sk)
+{
+	struct espintcp_ctx *ctx = espintcp_getctx(sk);
+	struct espintcp_msg *emsg = &ctx->partial;
+	int err;
+
+	if (!emsg->len)
+		return 0;
+
+	if (ctx->tx_running)
+		return -EAGAIN;
+	ctx->tx_running = 1;
+
+	if (emsg->skb)
+		err = espintcp_sendskb_locked(sk, emsg, 0);
+	else
+		err = espintcp_sendskmsg_locked(sk, emsg, 0);
+	if (err == -EAGAIN) {
+		ctx->tx_running = 0;
+		return 0;
+	}
+	if (!err)
+		memset(emsg, 0, sizeof(*emsg));
+
+	ctx->tx_running = 0;
+
+	return err;
+}
+
+int espintcp_push_skb(struct sock *sk, struct sk_buff *skb)
+{
+	struct espintcp_ctx *ctx = espintcp_getctx(sk);
+	struct espintcp_msg *emsg = &ctx->partial;
+	unsigned int len;
+	int offset;
+
+	if (sk->sk_state != TCP_ESTABLISHED) {
+		kfree_skb(skb);
+		return -ECONNRESET;
+	}
+
+	offset = skb_transport_offset(skb);
+	len = skb->len - offset;
+
+	espintcp_push_msgs(sk);
+
+	if (emsg->len) {
+		kfree_skb(skb);
+		return -ENOBUFS;
+	}
+
+	skb_set_owner_w(skb, sk);
+
+	emsg->offset = offset;
+	emsg->len = len;
+	emsg->skb = skb;
+
+	espintcp_push_msgs(sk);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(espintcp_push_skb);
+
+static int espintcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t size)
+{
+	long timeo = sock_sndtimeo(sk, msg->msg_flags & MSG_DONTWAIT);
+	struct espintcp_ctx *ctx = espintcp_getctx(sk);
+	struct espintcp_msg *emsg = &ctx->partial;
+	struct iov_iter pfx_iter;
+	struct kvec pfx_iov = {};
+	size_t msglen = size + 2;
+	char buf[2] = {0};
+	int err, end;
+
+	if (msg->msg_flags)
+		return -EOPNOTSUPP;
+
+	if (size > MAX_ESPINTCP_MSG)
+		return -EMSGSIZE;
+
+	if (msg->msg_controllen)
+		return -EOPNOTSUPP;
+
+	lock_sock(sk);
+
+	err = espintcp_push_msgs(sk);
+	if (err < 0) {
+		err = -ENOBUFS;
+		goto unlock;
+	}
+
+	sk_msg_init(&emsg->skmsg);
+	while (1) {
+		/* only -ENOMEM is possible since we don't coalesce */
+		err = sk_msg_alloc(sk, &emsg->skmsg, msglen, 0);
+		if (!err)
+			break;
+
+		err = sk_stream_wait_memory(sk, &timeo);
+		if (err)
+			goto fail;
+	}
+
+	*((__be16 *)buf) = cpu_to_be16(msglen);
+	pfx_iov.iov_base = buf;
+	pfx_iov.iov_len = sizeof(buf);
+	iov_iter_kvec(&pfx_iter, WRITE, &pfx_iov, 1, pfx_iov.iov_len);
+
+	err = sk_msg_memcopy_from_iter(sk, &pfx_iter, &emsg->skmsg,
+				       pfx_iov.iov_len);
+	if (err < 0)
+		goto fail;
+
+	err = sk_msg_memcopy_from_iter(sk, &msg->msg_iter, &emsg->skmsg, size);
+	if (err < 0)
+		goto fail;
+
+	end = emsg->skmsg.sg.end;
+	emsg->len = size;
+	sk_msg_iter_var_prev(end);
+	sg_mark_end(sk_msg_elem(&emsg->skmsg, end));
+
+	tcp_rate_check_app_limited(sk);
+
+	err = espintcp_push_msgs(sk);
+	/* this message could be partially sent, keep it */
+	if (err < 0)
+		goto unlock;
+	release_sock(sk);
+
+	return size;
+
+fail:
+	sk_msg_free(sk, &emsg->skmsg);
+	memset(emsg, 0, sizeof(*emsg));
+unlock:
+	release_sock(sk);
+	return err;
+}
+
+static struct proto espintcp_prot __ro_after_init;
+static struct proto_ops espintcp_ops __ro_after_init;
+
+static void espintcp_data_ready(struct sock *sk)
+{
+	struct espintcp_ctx *ctx = espintcp_getctx(sk);
+
+	strp_data_ready(&ctx->strp);
+}
+
+static void espintcp_tx_work(struct work_struct *work)
+{
+	struct espintcp_ctx *ctx = container_of(work,
+						struct espintcp_ctx, work);
+	struct sock *sk = ctx->strp.sk;
+
+	lock_sock(sk);
+	if (!ctx->tx_running)
+		espintcp_push_msgs(sk);
+	release_sock(sk);
+}
+
+static void espintcp_write_space(struct sock *sk)
+{
+	struct espintcp_ctx *ctx = espintcp_getctx(sk);
+
+	schedule_work(&ctx->work);
+	ctx->saved_write_space(sk);
+}
+
+static void espintcp_destruct(struct sock *sk)
+{
+	struct espintcp_ctx *ctx = espintcp_getctx(sk);
+
+	kfree(ctx);
+}
+
+bool tcp_is_ulp_esp(struct sock *sk)
+{
+	return sk->sk_prot == &espintcp_prot;
+}
+EXPORT_SYMBOL_GPL(tcp_is_ulp_esp);
+
+static int espintcp_init_sk(struct sock *sk)
+{
+	struct inet_connection_sock *icsk = inet_csk(sk);
+	struct strp_callbacks cb = {
+		.rcv_msg = espintcp_rcv,
+		.parse_msg = espintcp_parse,
+	};
+	struct espintcp_ctx *ctx;
+	int err;
+
+	/* sockmap is not compatible with espintcp */
+	if (sk->sk_user_data)
+		return -EBUSY;
+
+	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
+	if (!ctx)
+		return -ENOMEM;
+
+	err = strp_init(&ctx->strp, sk, &cb);
+	if (err)
+		goto free;
+
+	__sk_dst_reset(sk);
+
+	strp_check_rcv(&ctx->strp);
+	skb_queue_head_init(&ctx->ike_queue);
+	skb_queue_head_init(&ctx->out_queue);
+	sk->sk_prot = &espintcp_prot;
+	sk->sk_socket->ops = &espintcp_ops;
+	ctx->saved_data_ready = sk->sk_data_ready;
+	ctx->saved_write_space = sk->sk_write_space;
+	sk->sk_data_ready = espintcp_data_ready;
+	sk->sk_write_space = espintcp_write_space;
+	sk->sk_destruct = espintcp_destruct;
+	rcu_assign_pointer(icsk->icsk_ulp_data, ctx);
+	INIT_WORK(&ctx->work, espintcp_tx_work);
+
+	/* avoid using task_frag */
+	sk->sk_allocation = GFP_ATOMIC;
+
+	return 0;
+
+free:
+	kfree(ctx);
+	return err;
+}
+
+static void espintcp_release(struct sock *sk)
+{
+	struct espintcp_ctx *ctx = espintcp_getctx(sk);
+	struct sk_buff_head queue;
+	struct sk_buff *skb;
+
+	__skb_queue_head_init(&queue);
+	skb_queue_splice_init(&ctx->out_queue, &queue);
+
+	while ((skb = __skb_dequeue(&queue)))
+		espintcp_push_skb(sk, skb);
+
+	tcp_release_cb(sk);
+}
+
+static void espintcp_close(struct sock *sk, long timeout)
+{
+	struct espintcp_ctx *ctx = espintcp_getctx(sk);
+	struct espintcp_msg *emsg = &ctx->partial;
+
+	strp_stop(&ctx->strp);
+
+	sk->sk_prot = &tcp_prot;
+	barrier();
+
+	cancel_work_sync(&ctx->work);
+	strp_done(&ctx->strp);
+
+	skb_queue_purge(&ctx->out_queue);
+	skb_queue_purge(&ctx->ike_queue);
+
+	if (emsg->len) {
+		if (emsg->skb)
+			kfree_skb(emsg->skb);
+		else
+			sk_msg_free(sk, &emsg->skmsg);
+	}
+
+	tcp_close(sk, timeout);
+}
+
+static __poll_t espintcp_poll(struct file *file, struct socket *sock,
+			      poll_table *wait)
+{
+	__poll_t mask = datagram_poll(file, sock, wait);
+	struct sock *sk = sock->sk;
+	struct espintcp_ctx *ctx = espintcp_getctx(sk);
+
+	if (!skb_queue_empty(&ctx->ike_queue))
+		mask |= EPOLLIN | EPOLLRDNORM;
+
+	return mask;
+}
+
+static struct tcp_ulp_ops espintcp_ulp __read_mostly = {
+	.name = "espintcp",
+	.owner = THIS_MODULE,
+	.init = espintcp_init_sk,
+};
+
+void __init espintcp_init(void)
+{
+	memcpy(&espintcp_prot, &tcp_prot, sizeof(tcp_prot));
+	memcpy(&espintcp_ops, &inet_stream_ops, sizeof(inet_stream_ops));
+	espintcp_prot.sendmsg = espintcp_sendmsg;
+	espintcp_prot.recvmsg = espintcp_recvmsg;
+	espintcp_prot.close = espintcp_close;
+	espintcp_prot.release_cb = espintcp_release;
+	espintcp_ops.poll = espintcp_poll;
+
+	tcp_register_ulp(&espintcp_ulp);
+}
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index f2d1e573ea55..297d1eb79e5c 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -39,6 +39,9 @@
 #ifdef CONFIG_XFRM_STATISTICS
 #include <net/snmp.h>
 #endif
+#ifdef CONFIG_INET_ESPINTCP
+#include <net/espintcp.h>
+#endif
 
 #include "xfrm_hash.h"
 
@@ -4157,6 +4160,10 @@ void __init xfrm_init(void)
 	seqcount_init(&xfrm_policy_hash_generation);
 	xfrm_input_init();
 
+#ifdef CONFIG_INET_ESPINTCP
+	espintcp_init();
+#endif
+
 	RCU_INIT_POINTER(xfrm_if_cb, NULL);
 	synchronize_rcu();
 }
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index f3423562d933..170d6e7f31d3 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -670,6 +670,9 @@ int __xfrm_state_delete(struct xfrm_state *x)
 		net->xfrm.state_num--;
 		spin_unlock(&net->xfrm.xfrm_state_lock);
 
+		if (x->encap_sk)
+			sock_put(rcu_dereference_raw(x->encap_sk));
+
 		xfrm_dev_state_delete(x);
 
 		/* All xfrm_state objects are created by xfrm_state_alloc.
-- 
cgit v1.2.3-59-g8ed1b


From 718eae277e62a26e5862eb72a830b5e0fe37b04a Mon Sep 17 00:00:00 2001
From: Mao Wenan <maowenan@huawei.com>
Date: Mon, 9 Dec 2019 21:08:45 +0800
Subject: NFC: port100: Convert cpu_to_le16(le16_to_cpu(E1) + E2) to use
 le16_add_cpu().

Convert cpu_to_le16(le16_to_cpu(frame->datalen) + len) to
use le16_add_cpu(), which is more concise and does the same thing.

Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Mao Wenan <maowenan@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/nfc/port100.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/nfc/port100.c b/drivers/nfc/port100.c
index 604dba4f18af..8e4d355dc3ae 100644
--- a/drivers/nfc/port100.c
+++ b/drivers/nfc/port100.c
@@ -565,7 +565,7 @@ static void port100_tx_update_payload_len(void *_frame, int len)
 {
 	struct port100_frame *frame = _frame;
 
-	frame->datalen = cpu_to_le16(le16_to_cpu(frame->datalen) + len);
+	le16_add_cpu(&frame->datalen, len);
 }
 
 static bool port100_rx_frame_is_valid(void *_frame)
-- 
cgit v1.2.3-59-g8ed1b


From e70ac628289766bc2c81a0db161368b69da774fd Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.king@canonical.com>
Date: Mon, 9 Dec 2019 13:16:07 +0000
Subject: qed: remove redundant assignments to rc

The variable rc is assigned with a value that is never read and
it is re-assigned a new value later on.  The assignment is redundant
and can be removed.  Clean up multiple occurrances of this pattern.

Addresses-Coverity: ("Unused value")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/qlogic/qed/qed_sp_commands.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c b/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c
index 7e0b795230b2..900bc603e30a 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c
@@ -331,8 +331,8 @@ int qed_sp_pf_start(struct qed_hwfn *p_hwfn,
 	u8 sb_index = p_hwfn->p_eq->eq_sb_index;
 	struct qed_spq_entry *p_ent = NULL;
 	struct qed_sp_init_data init_data;
-	int rc = -EINVAL;
 	u8 page_cnt, i;
+	int rc;
 
 	/* update initial eq producer */
 	qed_eq_prod_update(p_hwfn,
@@ -447,7 +447,7 @@ int qed_sp_pf_update(struct qed_hwfn *p_hwfn)
 {
 	struct qed_spq_entry *p_ent = NULL;
 	struct qed_sp_init_data init_data;
-	int rc = -EINVAL;
+	int rc;
 
 	/* Get SPQ entry */
 	memset(&init_data, 0, sizeof(init_data));
@@ -471,7 +471,7 @@ int qed_sp_pf_update_ufp(struct qed_hwfn *p_hwfn)
 {
 	struct qed_spq_entry *p_ent = NULL;
 	struct qed_sp_init_data init_data;
-	int rc = -EOPNOTSUPP;
+	int rc;
 
 	if (p_hwfn->ufp_info.pri_type == QED_UFP_PRI_UNKNOWN) {
 		DP_INFO(p_hwfn, "Invalid priority type %d\n",
@@ -509,7 +509,7 @@ int qed_sp_pf_update_tunn_cfg(struct qed_hwfn *p_hwfn,
 {
 	struct qed_spq_entry *p_ent = NULL;
 	struct qed_sp_init_data init_data;
-	int rc = -EINVAL;
+	int rc;
 
 	if (IS_VF(p_hwfn->cdev))
 		return qed_vf_pf_tunnel_param_update(p_hwfn, p_tunn);
@@ -546,7 +546,7 @@ int qed_sp_pf_stop(struct qed_hwfn *p_hwfn)
 {
 	struct qed_spq_entry *p_ent = NULL;
 	struct qed_sp_init_data init_data;
-	int rc = -EINVAL;
+	int rc;
 
 	/* Get SPQ entry */
 	memset(&init_data, 0, sizeof(init_data));
-- 
cgit v1.2.3-59-g8ed1b


From 26c97a2d82c41938c3053d9ec27cdbcd8e34fb8b Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Mon, 9 Dec 2019 13:40:23 +0000
Subject: net: sfp: avoid tx-fault with Nokia GPON module

The Nokia GPON module can hold tx-fault active while it is initialising
which can take up to 60s. Avoid this causing the module to be declared
faulty after the SFP MSA defined non-cooled module timeout.

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/sfp.c | 42 ++++++++++++++++++++++++++++++------------
 1 file changed, 30 insertions(+), 12 deletions(-)

diff --git a/drivers/net/phy/sfp.c b/drivers/net/phy/sfp.c
index c0b9a8e4e65a..27360d1840b2 100644
--- a/drivers/net/phy/sfp.c
+++ b/drivers/net/phy/sfp.c
@@ -155,10 +155,20 @@ static const enum gpiod_flags gpio_flags[] = {
 	GPIOD_ASIS,
 };
 
-#define T_WAIT		msecs_to_jiffies(50)
-#define T_INIT_JIFFIES	msecs_to_jiffies(300)
-#define T_RESET_US	10
-#define T_FAULT_RECOVER	msecs_to_jiffies(1000)
+/* t_start_up (SFF-8431) or t_init (SFF-8472) is the time required for a
+ * non-cooled module to initialise its laser safety circuitry. We wait
+ * an initial T_WAIT period before we check the tx fault to give any PHY
+ * on board (for a copper SFP) time to initialise.
+ */
+#define T_WAIT			msecs_to_jiffies(50)
+#define T_START_UP		msecs_to_jiffies(300)
+#define T_START_UP_BAD_GPON	msecs_to_jiffies(60000)
+
+/* t_reset is the time required to assert the TX_DISABLE signal to reset
+ * an indicated TX_FAULT.
+ */
+#define T_RESET_US		10
+#define T_FAULT_RECOVER		msecs_to_jiffies(1000)
 
 /* SFP module presence detection is poor: the three MOD DEF signals are
  * the same length on the PCB, which means it's possible for MOD DEF 0 to
@@ -218,6 +228,7 @@ struct sfp {
 
 	struct sfp_eeprom_id id;
 	unsigned int module_power_mW;
+	unsigned int module_t_start_up;
 
 #if IS_ENABLED(CONFIG_HWMON)
 	struct sfp_diag diag;
@@ -1655,6 +1666,12 @@ static int sfp_sm_mod_probe(struct sfp *sfp, bool report)
 	if (ret < 0)
 		return ret;
 
+	if (!memcmp(id.base.vendor_name, "ALCATELLUCENT   ", 16) &&
+	    !memcmp(id.base.vendor_pn, "3FE46541AA      ", 16))
+		sfp->module_t_start_up = T_START_UP_BAD_GPON;
+	else
+		sfp->module_t_start_up = T_START_UP;
+
 	return 0;
 }
 
@@ -1855,11 +1872,12 @@ static void sfp_sm_main(struct sfp *sfp, unsigned int event)
 			break;
 
 		if (sfp->state & SFP_F_TX_FAULT) {
-			/* Wait t_init before indicating that the link is up,
-			 * provided the current state indicates no TX_FAULT. If
-			 * TX_FAULT clears before this time, that's fine too.
+			/* Wait up to t_init (SFF-8472) or t_start_up (SFF-8431)
+			 * from the TX_DISABLE deassertion for the module to
+			 * initialise, which is indicated by TX_FAULT
+			 * deasserting.
 			 */
-			timeout = T_INIT_JIFFIES;
+			timeout = sfp->module_t_start_up;
 			if (timeout > T_WAIT)
 				timeout -= T_WAIT;
 			else
@@ -1876,8 +1894,8 @@ static void sfp_sm_main(struct sfp *sfp, unsigned int event)
 
 	case SFP_S_INIT:
 		if (event == SFP_E_TIMEOUT && sfp->state & SFP_F_TX_FAULT) {
-			/* TX_FAULT is still asserted after t_init, so assume
-			 * there is a fault.
+			/* TX_FAULT is still asserted after t_init or
+			 * or t_start_up, so assume there is a fault.
 			 */
 			sfp_sm_fault(sfp, SFP_S_INIT_TX_FAULT,
 				     sfp->sm_retries == 5);
@@ -1896,7 +1914,7 @@ static void sfp_sm_main(struct sfp *sfp, unsigned int event)
 	case SFP_S_INIT_TX_FAULT:
 		if (event == SFP_E_TIMEOUT) {
 			sfp_module_tx_fault_reset(sfp);
-			sfp_sm_next(sfp, SFP_S_INIT, T_INIT_JIFFIES);
+			sfp_sm_next(sfp, SFP_S_INIT, sfp->module_t_start_up);
 		}
 		break;
 
@@ -1920,7 +1938,7 @@ static void sfp_sm_main(struct sfp *sfp, unsigned int event)
 	case SFP_S_TX_FAULT:
 		if (event == SFP_E_TIMEOUT) {
 			sfp_module_tx_fault_reset(sfp);
-			sfp_sm_next(sfp, SFP_S_REINIT, T_INIT_JIFFIES);
+			sfp_sm_next(sfp, SFP_S_REINIT, sfp->module_t_start_up);
 		}
 		break;
 
-- 
cgit v1.2.3-59-g8ed1b


From 4e7696d90b51a1a73ce0e8174f3aff58b914619c Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Mon, 9 Dec 2019 13:45:18 +0800
Subject: sctp: get netns from asoc and ep base

Commit 312434617cb1 ("sctp: cache netns in sctp_ep_common") set netns
in asoc and ep base since they're created, and it will never change.
It's a better way to get netns from asoc and ep base, comparing to
calling sock_net().

This patch is to replace them.

v1->v2:
  - no change.

Suggested-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sctp/associola.c         | 10 +++++-----
 net/sctp/chunk.c             |  2 +-
 net/sctp/endpointola.c       |  6 +++---
 net/sctp/input.c             |  5 ++---
 net/sctp/output.c            |  2 +-
 net/sctp/outqueue.c          |  6 +++---
 net/sctp/sm_make_chunk.c     |  7 +++----
 net/sctp/sm_sideeffect.c     | 16 ++++++----------
 net/sctp/sm_statefuns.c      |  2 +-
 net/sctp/socket.c            | 12 +++++-------
 net/sctp/stream.c            |  3 +--
 net/sctp/stream_interleave.c | 23 ++++++++++-------------
 net/sctp/transport.c         |  2 +-
 net/sctp/ulpqueue.c          | 15 +++++++--------
 14 files changed, 49 insertions(+), 62 deletions(-)

diff --git a/net/sctp/associola.c b/net/sctp/associola.c
index bbd5004a5d09..437079a4883d 100644
--- a/net/sctp/associola.c
+++ b/net/sctp/associola.c
@@ -584,7 +584,6 @@ struct sctp_transport *sctp_assoc_add_peer(struct sctp_association *asoc,
 					   const gfp_t gfp,
 					   const int peer_state)
 {
-	struct net *net = sock_net(asoc->base.sk);
 	struct sctp_transport *peer;
 	struct sctp_sock *sp;
 	unsigned short port;
@@ -614,7 +613,7 @@ struct sctp_transport *sctp_assoc_add_peer(struct sctp_association *asoc,
 		return peer;
 	}
 
-	peer = sctp_transport_new(net, addr, gfp);
+	peer = sctp_transport_new(asoc->base.net, addr, gfp);
 	if (!peer)
 		return NULL;
 
@@ -974,7 +973,7 @@ static void sctp_assoc_bh_rcv(struct work_struct *work)
 	struct sctp_association *asoc =
 		container_of(work, struct sctp_association,
 			     base.inqueue.immediate);
-	struct net *net = sock_net(asoc->base.sk);
+	struct net *net = asoc->base.net;
 	union sctp_subtype subtype;
 	struct sctp_endpoint *ep;
 	struct sctp_chunk *chunk;
@@ -1442,7 +1441,8 @@ void sctp_assoc_sync_pmtu(struct sctp_association *asoc)
 /* Should we send a SACK to update our peer? */
 static inline bool sctp_peer_needs_update(struct sctp_association *asoc)
 {
-	struct net *net = sock_net(asoc->base.sk);
+	struct net *net = asoc->base.net;
+
 	switch (asoc->state) {
 	case SCTP_STATE_ESTABLISHED:
 	case SCTP_STATE_SHUTDOWN_PENDING:
@@ -1576,7 +1576,7 @@ int sctp_assoc_set_bind_addr_from_ep(struct sctp_association *asoc,
 	if (asoc->peer.ipv6_address)
 		flags |= SCTP_ADDR6_PEERSUPP;
 
-	return sctp_bind_addr_copy(sock_net(asoc->base.sk),
+	return sctp_bind_addr_copy(asoc->base.net,
 				   &asoc->base.bind_addr,
 				   &asoc->ep->base.bind_addr,
 				   scope, gfp, flags);
diff --git a/net/sctp/chunk.c b/net/sctp/chunk.c
index cc3ce5d80b08..ab6a997e222f 100644
--- a/net/sctp/chunk.c
+++ b/net/sctp/chunk.c
@@ -225,7 +225,7 @@ struct sctp_datamsg *sctp_datamsg_from_user(struct sctp_association *asoc,
 	if (msg_len >= first_len) {
 		msg->can_delay = 0;
 		if (msg_len > first_len)
-			SCTP_INC_STATS(sock_net(asoc->base.sk),
+			SCTP_INC_STATS(asoc->base.net,
 				       SCTP_MIB_FRAGUSRMSGS);
 	} else {
 		/* Which may be the only one... */
diff --git a/net/sctp/endpointola.c b/net/sctp/endpointola.c
index 3ccab7440c9e..48c9c2c7602f 100644
--- a/net/sctp/endpointola.c
+++ b/net/sctp/endpointola.c
@@ -244,7 +244,7 @@ struct sctp_endpoint *sctp_endpoint_is_match(struct sctp_endpoint *ep,
 	struct sctp_endpoint *retval = NULL;
 
 	if ((htons(ep->base.bind_addr.port) == laddr->v4.sin_port) &&
-	    net_eq(sock_net(ep->base.sk), net)) {
+	    net_eq(ep->base.net, net)) {
 		if (sctp_bind_addr_match(&ep->base.bind_addr, laddr,
 					 sctp_sk(ep->base.sk)))
 			retval = ep;
@@ -292,8 +292,8 @@ bool sctp_endpoint_is_peeled_off(struct sctp_endpoint *ep,
 				 const union sctp_addr *paddr)
 {
 	struct sctp_sockaddr_entry *addr;
+	struct net *net = ep->base.net;
 	struct sctp_bind_addr *bp;
-	struct net *net = sock_net(ep->base.sk);
 
 	bp = &ep->base.bind_addr;
 	/* This function is called with the socket lock held,
@@ -384,7 +384,7 @@ normal:
 		if (asoc && sctp_chunk_is_data(chunk))
 			asoc->peer.last_data_from = chunk->transport;
 		else {
-			SCTP_INC_STATS(sock_net(ep->base.sk), SCTP_MIB_INCTRLCHUNKS);
+			SCTP_INC_STATS(ep->base.net, SCTP_MIB_INCTRLCHUNKS);
 			if (asoc)
 				asoc->stats.ictrlchunks++;
 		}
diff --git a/net/sctp/input.c b/net/sctp/input.c
index 4d2bcfc9d7f8..efaaefc3bb1c 100644
--- a/net/sctp/input.c
+++ b/net/sctp/input.c
@@ -937,7 +937,7 @@ int sctp_hash_transport(struct sctp_transport *t)
 	if (t->asoc->temp)
 		return 0;
 
-	arg.net   = sock_net(t->asoc->base.sk);
+	arg.net   = t->asoc->base.net;
 	arg.paddr = &t->ipaddr;
 	arg.lport = htons(t->asoc->base.bind_addr.port);
 
@@ -1004,12 +1004,11 @@ struct sctp_transport *sctp_epaddr_lookup_transport(
 				const struct sctp_endpoint *ep,
 				const union sctp_addr *paddr)
 {
-	struct net *net = sock_net(ep->base.sk);
 	struct rhlist_head *tmp, *list;
 	struct sctp_transport *t;
 	struct sctp_hash_cmp_arg arg = {
 		.paddr = paddr,
-		.net   = net,
+		.net   = ep->base.net,
 		.lport = htons(ep->base.bind_addr.port),
 	};
 
diff --git a/net/sctp/output.c b/net/sctp/output.c
index dbda7e7927fd..1441eaf460bb 100644
--- a/net/sctp/output.c
+++ b/net/sctp/output.c
@@ -282,7 +282,7 @@ static enum sctp_xmit sctp_packet_bundle_sack(struct sctp_packet *pkt,
 					sctp_chunk_free(sack);
 					goto out;
 				}
-				SCTP_INC_STATS(sock_net(asoc->base.sk),
+				SCTP_INC_STATS(asoc->base.net,
 					       SCTP_MIB_OUTCTRLCHUNKS);
 				asoc->stats.octrlchunks++;
 				asoc->peer.sack_needed = 0;
diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c
index 0dab62b67b9a..a031d1134c60 100644
--- a/net/sctp/outqueue.c
+++ b/net/sctp/outqueue.c
@@ -279,7 +279,7 @@ void sctp_outq_free(struct sctp_outq *q)
 /* Put a new chunk in an sctp_outq.  */
 void sctp_outq_tail(struct sctp_outq *q, struct sctp_chunk *chunk, gfp_t gfp)
 {
-	struct net *net = sock_net(q->asoc->base.sk);
+	struct net *net = q->asoc->base.net;
 
 	pr_debug("%s: outq:%p, chunk:%p[%s]\n", __func__, q, chunk,
 		 chunk && chunk->chunk_hdr ?
@@ -533,7 +533,7 @@ void sctp_retransmit_mark(struct sctp_outq *q,
 void sctp_retransmit(struct sctp_outq *q, struct sctp_transport *transport,
 		     enum sctp_retransmit_reason reason)
 {
-	struct net *net = sock_net(q->asoc->base.sk);
+	struct net *net = q->asoc->base.net;
 
 	switch (reason) {
 	case SCTP_RTXR_T3_RTX:
@@ -1884,6 +1884,6 @@ void sctp_generate_fwdtsn(struct sctp_outq *q, __u32 ctsn)
 
 	if (ftsn_chunk) {
 		list_add_tail(&ftsn_chunk->list, &q->control_chunk_list);
-		SCTP_INC_STATS(sock_net(asoc->base.sk), SCTP_MIB_OUTCTRLCHUNKS);
+		SCTP_INC_STATS(asoc->base.net, SCTP_MIB_OUTCTRLCHUNKS);
 	}
 }
diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c
index 48d63956a68c..09050c1d5517 100644
--- a/net/sctp/sm_make_chunk.c
+++ b/net/sctp/sm_make_chunk.c
@@ -2307,7 +2307,6 @@ int sctp_process_init(struct sctp_association *asoc, struct sctp_chunk *chunk,
 		      const union sctp_addr *peer_addr,
 		      struct sctp_init_chunk *peer_init, gfp_t gfp)
 {
-	struct net *net = sock_net(asoc->base.sk);
 	struct sctp_transport *transport;
 	struct list_head *pos, *temp;
 	union sctp_params param;
@@ -2363,8 +2362,8 @@ int sctp_process_init(struct sctp_association *asoc, struct sctp_chunk *chunk,
 	 * also give us an option to silently ignore the packet, which
 	 * is what we'll do here.
 	 */
-	if (!net->sctp.addip_noauth &&
-	     (asoc->peer.asconf_capable && !asoc->peer.auth_capable)) {
+	if (!asoc->base.net->sctp.addip_noauth &&
+	    (asoc->peer.asconf_capable && !asoc->peer.auth_capable)) {
 		asoc->peer.addip_disabled_mask |= (SCTP_PARAM_ADD_IP |
 						  SCTP_PARAM_DEL_IP |
 						  SCTP_PARAM_SET_PRIMARY);
@@ -2491,9 +2490,9 @@ static int sctp_process_param(struct sctp_association *asoc,
 			      const union sctp_addr *peer_addr,
 			      gfp_t gfp)
 {
-	struct net *net = sock_net(asoc->base.sk);
 	struct sctp_endpoint *ep = asoc->ep;
 	union sctp_addr_param *addr_param;
+	struct net *net = asoc->base.net;
 	struct sctp_transport *t;
 	enum sctp_scope scope;
 	union sctp_addr addr;
diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c
index acd737d4c0e0..ce82699d0dca 100644
--- a/net/sctp/sm_sideeffect.c
+++ b/net/sctp/sm_sideeffect.c
@@ -516,8 +516,6 @@ static void sctp_do_8_2_transport_strike(struct sctp_cmd_seq *commands,
 					 struct sctp_transport *transport,
 					 int is_hb)
 {
-	struct net *net = sock_net(asoc->base.sk);
-
 	/* The check for association's overall error counter exceeding the
 	 * threshold is done in the state function.
 	 */
@@ -544,10 +542,10 @@ static void sctp_do_8_2_transport_strike(struct sctp_cmd_seq *commands,
 	 * is SCTP_ACTIVE, then mark this transport as Partially Failed,
 	 * see SCTP Quick Failover Draft, section 5.1
 	 */
-	if (net->sctp.pf_enable &&
-	   (transport->state == SCTP_ACTIVE) &&
-	   (transport->error_count < transport->pathmaxrxt) &&
-	   (transport->error_count > transport->pf_retrans)) {
+	if (asoc->base.net->sctp.pf_enable &&
+	    transport->state == SCTP_ACTIVE &&
+	    transport->error_count < transport->pathmaxrxt &&
+	    transport->error_count > transport->pf_retrans) {
 
 		sctp_assoc_control_transport(asoc, transport,
 					     SCTP_TRANSPORT_PF,
@@ -798,10 +796,8 @@ static int sctp_cmd_process_sack(struct sctp_cmd_seq *cmds,
 	int err = 0;
 
 	if (sctp_outq_sack(&asoc->outqueue, chunk)) {
-		struct net *net = sock_net(asoc->base.sk);
-
 		/* There are no more TSNs awaiting SACK.  */
-		err = sctp_do_sm(net, SCTP_EVENT_T_OTHER,
+		err = sctp_do_sm(asoc->base.net, SCTP_EVENT_T_OTHER,
 				 SCTP_ST_OTHER(SCTP_EVENT_NO_PENDING_TSN),
 				 asoc->state, asoc->ep, asoc, NULL,
 				 GFP_ATOMIC);
@@ -834,7 +830,7 @@ static void sctp_cmd_assoc_update(struct sctp_cmd_seq *cmds,
 				  struct sctp_association *asoc,
 				  struct sctp_association *new)
 {
-	struct net *net = sock_net(asoc->base.sk);
+	struct net *net = asoc->base.net;
 	struct sctp_chunk *abort;
 
 	if (!sctp_assoc_update(asoc, new))
diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c
index 4ab8208a2dd4..42558fa3f3e4 100644
--- a/net/sctp/sm_statefuns.c
+++ b/net/sctp/sm_statefuns.c
@@ -1320,7 +1320,7 @@ static int sctp_sf_check_restart_addrs(const struct sctp_association *new_asoc,
 				       struct sctp_chunk *init,
 				       struct sctp_cmd_seq *commands)
 {
-	struct net *net = sock_net(new_asoc->base.sk);
+	struct net *net = new_asoc->base.net;
 	struct sctp_transport *new_addr;
 	int ret = 1;
 
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index 0b485952a71c..1b56fc440606 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -436,8 +436,7 @@ static int sctp_do_bind(struct sock *sk, union sctp_addr *addr, int len)
 static int sctp_send_asconf(struct sctp_association *asoc,
 			    struct sctp_chunk *chunk)
 {
-	struct net 	*net = sock_net(asoc->base.sk);
-	int		retval = 0;
+	int retval = 0;
 
 	/* If there is an outstanding ASCONF chunk, queue it for later
 	 * transmission.
@@ -449,7 +448,7 @@ static int sctp_send_asconf(struct sctp_association *asoc,
 
 	/* Hold the chunk until an ASCONF_ACK is received. */
 	sctp_chunk_hold(chunk);
-	retval = sctp_primitive_ASCONF(net, asoc, chunk);
+	retval = sctp_primitive_ASCONF(asoc->base.net, asoc, chunk);
 	if (retval)
 		sctp_chunk_free(chunk);
 	else
@@ -2428,9 +2427,8 @@ static int sctp_apply_peer_addr_params(struct sctp_paddrparams *params,
 	int error;
 
 	if (params->spp_flags & SPP_HB_DEMAND && trans) {
-		struct net *net = sock_net(trans->asoc->base.sk);
-
-		error = sctp_primitive_REQUESTHEARTBEAT(net, trans->asoc, trans);
+		error = sctp_primitive_REQUESTHEARTBEAT(trans->asoc->base.net,
+							trans->asoc, trans);
 		if (error)
 			return error;
 	}
@@ -5364,7 +5362,7 @@ struct sctp_transport *sctp_transport_get_next(struct net *net,
 		if (!sctp_transport_hold(t))
 			continue;
 
-		if (net_eq(sock_net(t->asoc->base.sk), net) &&
+		if (net_eq(t->asoc->base.net, net) &&
 		    t->asoc->peer.primary_path == t)
 			break;
 
diff --git a/net/sctp/stream.c b/net/sctp/stream.c
index e83cdaa2ab76..df60b5ef24cb 100644
--- a/net/sctp/stream.c
+++ b/net/sctp/stream.c
@@ -218,10 +218,9 @@ void sctp_stream_update(struct sctp_stream *stream, struct sctp_stream *new)
 static int sctp_send_reconf(struct sctp_association *asoc,
 			    struct sctp_chunk *chunk)
 {
-	struct net *net = sock_net(asoc->base.sk);
 	int retval = 0;
 
-	retval = sctp_primitive_RECONF(net, asoc, chunk);
+	retval = sctp_primitive_RECONF(asoc->base.net, asoc, chunk);
 	if (retval)
 		sctp_chunk_free(chunk);
 
diff --git a/net/sctp/stream_interleave.c b/net/sctp/stream_interleave.c
index 40c40be23fcb..6b13f737ebf2 100644
--- a/net/sctp/stream_interleave.c
+++ b/net/sctp/stream_interleave.c
@@ -241,9 +241,8 @@ out:
 	if (!first_frag)
 		return NULL;
 
-	retval = sctp_make_reassembled_event(sock_net(ulpq->asoc->base.sk),
-					     &ulpq->reasm, first_frag,
-					     last_frag);
+	retval = sctp_make_reassembled_event(ulpq->asoc->base.net, &ulpq->reasm,
+					     first_frag, last_frag);
 	if (retval) {
 		sin->fsn = next_fsn;
 		if (is_last) {
@@ -326,7 +325,7 @@ static struct sctp_ulpevent *sctp_intl_retrieve_reassembled(
 
 	pd_point = sctp_sk(asoc->base.sk)->pd_point;
 	if (pd_point && pd_point <= pd_len) {
-		retval = sctp_make_reassembled_event(sock_net(asoc->base.sk),
+		retval = sctp_make_reassembled_event(asoc->base.net,
 						     &ulpq->reasm,
 						     pd_first, pd_last);
 		if (retval) {
@@ -337,8 +336,7 @@ static struct sctp_ulpevent *sctp_intl_retrieve_reassembled(
 	goto out;
 
 found:
-	retval = sctp_make_reassembled_event(sock_net(asoc->base.sk),
-					     &ulpq->reasm,
+	retval = sctp_make_reassembled_event(asoc->base.net, &ulpq->reasm,
 					     first_frag, pos);
 	if (retval)
 		retval->msg_flags |= MSG_EOR;
@@ -630,7 +628,7 @@ out:
 	if (!first_frag)
 		return NULL;
 
-	retval = sctp_make_reassembled_event(sock_net(ulpq->asoc->base.sk),
+	retval = sctp_make_reassembled_event(ulpq->asoc->base.net,
 					     &ulpq->reasm_uo, first_frag,
 					     last_frag);
 	if (retval) {
@@ -716,7 +714,7 @@ static struct sctp_ulpevent *sctp_intl_retrieve_reassembled_uo(
 
 	pd_point = sctp_sk(asoc->base.sk)->pd_point;
 	if (pd_point && pd_point <= pd_len) {
-		retval = sctp_make_reassembled_event(sock_net(asoc->base.sk),
+		retval = sctp_make_reassembled_event(asoc->base.net,
 						     &ulpq->reasm_uo,
 						     pd_first, pd_last);
 		if (retval) {
@@ -727,8 +725,7 @@ static struct sctp_ulpevent *sctp_intl_retrieve_reassembled_uo(
 	goto out;
 
 found:
-	retval = sctp_make_reassembled_event(sock_net(asoc->base.sk),
-					     &ulpq->reasm_uo,
+	retval = sctp_make_reassembled_event(asoc->base.net, &ulpq->reasm_uo,
 					     first_frag, pos);
 	if (retval)
 		retval->msg_flags |= MSG_EOR;
@@ -814,7 +811,7 @@ static struct sctp_ulpevent *sctp_intl_retrieve_first_uo(struct sctp_ulpq *ulpq)
 		return NULL;
 
 out:
-	retval = sctp_make_reassembled_event(sock_net(ulpq->asoc->base.sk),
+	retval = sctp_make_reassembled_event(ulpq->asoc->base.net,
 					     &ulpq->reasm_uo, first_frag,
 					     last_frag);
 	if (retval) {
@@ -921,7 +918,7 @@ static struct sctp_ulpevent *sctp_intl_retrieve_first(struct sctp_ulpq *ulpq)
 		return NULL;
 
 out:
-	retval = sctp_make_reassembled_event(sock_net(ulpq->asoc->base.sk),
+	retval = sctp_make_reassembled_event(ulpq->asoc->base.net,
 					     &ulpq->reasm, first_frag,
 					     last_frag);
 	if (retval) {
@@ -1159,7 +1156,7 @@ static void sctp_generate_iftsn(struct sctp_outq *q, __u32 ctsn)
 
 	if (ftsn_chunk) {
 		list_add_tail(&ftsn_chunk->list, &q->control_chunk_list);
-		SCTP_INC_STATS(sock_net(asoc->base.sk), SCTP_MIB_OUTCTRLCHUNKS);
+		SCTP_INC_STATS(asoc->base.net, SCTP_MIB_OUTCTRLCHUNKS);
 	}
 }
 
diff --git a/net/sctp/transport.c b/net/sctp/transport.c
index 7235a6032671..f4de064598f8 100644
--- a/net/sctp/transport.c
+++ b/net/sctp/transport.c
@@ -334,7 +334,7 @@ void sctp_transport_update_rto(struct sctp_transport *tp, __u32 rtt)
 		pr_debug("%s: rto_pending not set on transport %p!\n", __func__, tp);
 
 	if (tp->rttvar || tp->srtt) {
-		struct net *net = sock_net(tp->asoc->base.sk);
+		struct net *net = tp->asoc->base.net;
 		/* 6.3.1 C3) When a new RTT measurement R' is made, set
 		 * RTTVAR <- (1 - RTO.Beta) * RTTVAR + RTO.Beta * |SRTT - R'|
 		 * SRTT <- (1 - RTO.Alpha) * SRTT + RTO.Alpha * R'
diff --git a/net/sctp/ulpqueue.c b/net/sctp/ulpqueue.c
index b6536b7f14c0..1c6c640607c5 100644
--- a/net/sctp/ulpqueue.c
+++ b/net/sctp/ulpqueue.c
@@ -486,10 +486,9 @@ static struct sctp_ulpevent *sctp_ulpq_retrieve_reassembled(struct sctp_ulpq *ul
 		cevent = sctp_skb2event(pd_first);
 		pd_point = sctp_sk(asoc->base.sk)->pd_point;
 		if (pd_point && pd_point <= pd_len) {
-			retval = sctp_make_reassembled_event(sock_net(asoc->base.sk),
+			retval = sctp_make_reassembled_event(asoc->base.net,
 							     &ulpq->reasm,
-							     pd_first,
-							     pd_last);
+							     pd_first, pd_last);
 			if (retval)
 				sctp_ulpq_set_pd(ulpq);
 		}
@@ -497,7 +496,7 @@ static struct sctp_ulpevent *sctp_ulpq_retrieve_reassembled(struct sctp_ulpq *ul
 done:
 	return retval;
 found:
-	retval = sctp_make_reassembled_event(sock_net(ulpq->asoc->base.sk),
+	retval = sctp_make_reassembled_event(ulpq->asoc->base.net,
 					     &ulpq->reasm, first_frag, pos);
 	if (retval)
 		retval->msg_flags |= MSG_EOR;
@@ -563,8 +562,8 @@ static struct sctp_ulpevent *sctp_ulpq_retrieve_partial(struct sctp_ulpq *ulpq)
 	 * further.
 	 */
 done:
-	retval = sctp_make_reassembled_event(sock_net(ulpq->asoc->base.sk),
-					&ulpq->reasm, first_frag, last_frag);
+	retval = sctp_make_reassembled_event(ulpq->asoc->base.net, &ulpq->reasm,
+					     first_frag, last_frag);
 	if (retval && is_last)
 		retval->msg_flags |= MSG_EOR;
 
@@ -664,8 +663,8 @@ static struct sctp_ulpevent *sctp_ulpq_retrieve_first(struct sctp_ulpq *ulpq)
 	 * further.
 	 */
 done:
-	retval = sctp_make_reassembled_event(sock_net(ulpq->asoc->base.sk),
-					&ulpq->reasm, first_frag, last_frag);
+	retval = sctp_make_reassembled_event(ulpq->asoc->base.net, &ulpq->reasm,
+					     first_frag, last_frag);
 	return retval;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 65e6d90168f3593df0ae598502bcbf20d78ff0fb Mon Sep 17 00:00:00 2001
From: "Kevin(Yudong) Yang" <yyd@google.com>
Date: Mon, 9 Dec 2019 14:19:59 -0500
Subject: net-tcp: Disable TCP ssthresh metrics cache by default

This patch introduces a sysctl knob "net.ipv4.tcp_no_ssthresh_metrics_save"
that disables TCP ssthresh metrics cache by default. Other parts of TCP
metrics cache, e.g. rtt, cwnd, remain unchanged.

As modern networks becoming more and more dynamic, TCP metrics cache
today often causes more harm than benefits. For example, the same IP
address is often shared by different subscribers behind NAT in residential
networks. Even if the IP address is not shared by different users,
caching the slow-start threshold of a previous short flow using loss-based
congestion control (e.g. cubic) often causes the future longer flows of
the same network path to exit slow-start prematurely with abysmal
throughput.

Caching ssthresh is very risky and can lead to terrible performance.
Therefore it makes sense to make disabling ssthresh caching by
default and opt-in for specific networks by the administrators.
This practice also has worked well for several years of deployment with
CUBIC congestion control at Google.

Acked-by: Eric Dumazet <edumazet@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Acked-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Kevin(Yudong) Yang <yyd@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/ip-sysctl.txt |  4 ++++
 include/net/netns/ipv4.h               |  1 +
 net/ipv4/sysctl_net_ipv4.c             |  9 +++++++++
 net/ipv4/tcp_ipv4.c                    |  1 +
 net/ipv4/tcp_metrics.c                 | 13 +++++++++----
 5 files changed, 24 insertions(+), 4 deletions(-)

diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt
index fd26788e8c96..90b8f14bc41a 100644
--- a/Documentation/networking/ip-sysctl.txt
+++ b/Documentation/networking/ip-sysctl.txt
@@ -479,6 +479,10 @@ tcp_no_metrics_save - BOOLEAN
 	degradation.  If set, TCP will not cache metrics on closing
 	connections.
 
+tcp_no_ssthresh_metrics_save - BOOLEAN
+	Controls whether TCP saves ssthresh metrics in the route cache.
+	Default is 1, which disables ssthresh metrics.
+
 tcp_orphan_retries - INTEGER
 	This value influences the timeout of a locally closed TCP connection,
 	when RTO retransmissions remain unacknowledged.
diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index c0c0791b1912..08b98414d94e 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -154,6 +154,7 @@ struct netns_ipv4 {
 	int sysctl_tcp_adv_win_scale;
 	int sysctl_tcp_frto;
 	int sysctl_tcp_nometrics_save;
+	int sysctl_tcp_no_ssthresh_metrics_save;
 	int sysctl_tcp_moderate_rcvbuf;
 	int sysctl_tcp_tso_win_divisor;
 	int sysctl_tcp_workaround_signed_windows;
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index fcb2cd167f64..9684af02e0a5 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -1192,6 +1192,15 @@ static struct ctl_table ipv4_net_table[] = {
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec,
 	},
+	{
+		.procname	= "tcp_no_ssthresh_metrics_save",
+		.data		= &init_net.ipv4.sysctl_tcp_no_ssthresh_metrics_save,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= SYSCTL_ZERO,
+		.extra2		= SYSCTL_ONE,
+	},
 	{
 		.procname	= "tcp_moderate_rcvbuf",
 		.data		= &init_net.ipv4.sysctl_tcp_moderate_rcvbuf,
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 92282f98dc82..26637fce324d 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2674,6 +2674,7 @@ static int __net_init tcp_sk_init(struct net *net)
 	net->ipv4.sysctl_tcp_fin_timeout = TCP_FIN_TIMEOUT;
 	net->ipv4.sysctl_tcp_notsent_lowat = UINT_MAX;
 	net->ipv4.sysctl_tcp_tw_reuse = 2;
+	net->ipv4.sysctl_tcp_no_ssthresh_metrics_save = 1;
 
 	cnt = tcp_hashinfo.ehash_mask + 1;
 	net->ipv4.tcp_death_row.sysctl_max_tw_buckets = cnt / 2;
diff --git a/net/ipv4/tcp_metrics.c b/net/ipv4/tcp_metrics.c
index c4848e7a0aad..279db8822439 100644
--- a/net/ipv4/tcp_metrics.c
+++ b/net/ipv4/tcp_metrics.c
@@ -385,7 +385,8 @@ void tcp_update_metrics(struct sock *sk)
 
 	if (tcp_in_initial_slowstart(tp)) {
 		/* Slow start still did not finish. */
-		if (!tcp_metric_locked(tm, TCP_METRIC_SSTHRESH)) {
+		if (!net->ipv4.sysctl_tcp_no_ssthresh_metrics_save &&
+		    !tcp_metric_locked(tm, TCP_METRIC_SSTHRESH)) {
 			val = tcp_metric_get(tm, TCP_METRIC_SSTHRESH);
 			if (val && (tp->snd_cwnd >> 1) > val)
 				tcp_metric_set(tm, TCP_METRIC_SSTHRESH,
@@ -400,7 +401,8 @@ void tcp_update_metrics(struct sock *sk)
 	} else if (!tcp_in_slow_start(tp) &&
 		   icsk->icsk_ca_state == TCP_CA_Open) {
 		/* Cong. avoidance phase, cwnd is reliable. */
-		if (!tcp_metric_locked(tm, TCP_METRIC_SSTHRESH))
+		if (!net->ipv4.sysctl_tcp_no_ssthresh_metrics_save &&
+		    !tcp_metric_locked(tm, TCP_METRIC_SSTHRESH))
 			tcp_metric_set(tm, TCP_METRIC_SSTHRESH,
 				       max(tp->snd_cwnd >> 1, tp->snd_ssthresh));
 		if (!tcp_metric_locked(tm, TCP_METRIC_CWND)) {
@@ -416,7 +418,8 @@ void tcp_update_metrics(struct sock *sk)
 			tcp_metric_set(tm, TCP_METRIC_CWND,
 				       (val + tp->snd_ssthresh) >> 1);
 		}
-		if (!tcp_metric_locked(tm, TCP_METRIC_SSTHRESH)) {
+		if (!net->ipv4.sysctl_tcp_no_ssthresh_metrics_save &&
+		    !tcp_metric_locked(tm, TCP_METRIC_SSTHRESH)) {
 			val = tcp_metric_get(tm, TCP_METRIC_SSTHRESH);
 			if (val && tp->snd_ssthresh > val)
 				tcp_metric_set(tm, TCP_METRIC_SSTHRESH,
@@ -441,6 +444,7 @@ void tcp_init_metrics(struct sock *sk)
 {
 	struct dst_entry *dst = __sk_dst_get(sk);
 	struct tcp_sock *tp = tcp_sk(sk);
+	struct net *net = sock_net(sk);
 	struct tcp_metrics_block *tm;
 	u32 val, crtt = 0; /* cached RTT scaled by 8 */
 
@@ -458,7 +462,8 @@ void tcp_init_metrics(struct sock *sk)
 	if (tcp_metric_locked(tm, TCP_METRIC_CWND))
 		tp->snd_cwnd_clamp = tcp_metric_get(tm, TCP_METRIC_CWND);
 
-	val = tcp_metric_get(tm, TCP_METRIC_SSTHRESH);
+	val = net->ipv4.sysctl_tcp_no_ssthresh_metrics_save ?
+	      0 : tcp_metric_get(tm, TCP_METRIC_SSTHRESH);
 	if (val) {
 		tp->snd_ssthresh = val;
 		if (tp->snd_ssthresh > tp->snd_cwnd_clamp)
-- 
cgit v1.2.3-59-g8ed1b


From 96ae38af9d2992073f553d14d3a85f3e4ad56250 Mon Sep 17 00:00:00 2001
From: Dan Murphy <dmurphy@ti.com>
Date: Mon, 9 Dec 2019 14:10:24 -0600
Subject: dt-bindings: dp83867: Convert fifo-depth to common fifo-depth and
 make optional

Convert the ti,fifo-depth from a TI specific property to the common
tx-fifo-depth property.  Also add support for the rx-fifo-depth.

These are optional properties for this device and if these are not
available then the fifo depths are set to device default values.

Signed-off-by: Dan Murphy <dmurphy@ti.com>
Reported-by: Adrian Bunk <bunk@kernel.org>
CC: Rob Herring <robh@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/devicetree/bindings/net/ti,dp83867.txt | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/Documentation/devicetree/bindings/net/ti,dp83867.txt b/Documentation/devicetree/bindings/net/ti,dp83867.txt
index 388ff48f53ae..44e2a4fab29e 100644
--- a/Documentation/devicetree/bindings/net/ti,dp83867.txt
+++ b/Documentation/devicetree/bindings/net/ti,dp83867.txt
@@ -8,8 +8,6 @@ Required properties:
 	- ti,tx-internal-delay - RGMII Transmit Clock Delay - see dt-bindings/net/ti-dp83867.h
 		for applicable values. Required only if interface type is
 		PHY_INTERFACE_MODE_RGMII_ID or PHY_INTERFACE_MODE_RGMII_TXID
-	- ti,fifo-depth - Transmitt FIFO depth- see dt-bindings/net/ti-dp83867.h
-		for applicable values
 
 Note: If the interface type is PHY_INTERFACE_MODE_RGMII the TX/RX clock delays
       will be left at their default values, as set by the PHY's pin strapping.
@@ -42,6 +40,14 @@ Optional property:
 				    Some MACs work with differential SGMII clock.
 				    See data manual for details.
 
+	- ti,fifo-depth - Transmitt FIFO depth- see dt-bindings/net/ti-dp83867.h
+		for applicable values (deprecated)
+
+	-tx-fifo-depth - As defined in the ethernet-controller.yaml.  Values for
+			 the depth can be found in dt-bindings/net/ti-dp83867.h
+	-rx-fifo-depth - As defined in the ethernet-controller.yaml.  Values for
+			 the depth can be found in dt-bindings/net/ti-dp83867.h
+
 Note: ti,min-output-impedance and ti,max-output-impedance are mutually
       exclusive. When both properties are present ti,max-output-impedance
       takes precedence.
@@ -55,7 +61,7 @@ Example:
 		reg = <0>;
 		ti,rx-internal-delay = <DP83867_RGMIIDCTL_2_25_NS>;
 		ti,tx-internal-delay = <DP83867_RGMIIDCTL_2_75_NS>;
-		ti,fifo-depth = <DP83867_PHYCR_FIFO_DEPTH_4_B_NIB>;
+		tx-fifo-depth = <DP83867_PHYCR_FIFO_DEPTH_4_B_NIB>;
 	};
 
 Datasheet can be found:
-- 
cgit v1.2.3-59-g8ed1b


From e02d18161e9707d9ca6b04da8f4cf9e1c345d0b8 Mon Sep 17 00:00:00 2001
From: Dan Murphy <dmurphy@ti.com>
Date: Mon, 9 Dec 2019 14:10:25 -0600
Subject: net: phy: dp83867: Add rx-fifo-depth and tx-fifo-depth

This code changes the TI specific ti,fifo-depth to the common
tx-fifo-depth property.  The tx depth is applicable for both RGMII and
SGMII modes of operation.

rx-fifo-depth was added as well but this is only applicable for SGMII
mode.

So in summary
if RGMII mode write tx fifo depth only
if SGMII mode write both rx and tx fifo depths

If the property is not populated in the device tree then set the value
to the default values.

Signed-off-by: Dan Murphy <dmurphy@ti.com>
Reported-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/dp83867.c | 62 +++++++++++++++++++++++++++++++++++++----------
 1 file changed, 49 insertions(+), 13 deletions(-)

diff --git a/drivers/net/phy/dp83867.c b/drivers/net/phy/dp83867.c
index 9cd9dcee4eb2..adda0d0eab80 100644
--- a/drivers/net/phy/dp83867.c
+++ b/drivers/net/phy/dp83867.c
@@ -93,9 +93,11 @@
 #define DP83867_STRAP_STS2_CLK_SKEW_NONE	BIT(2)
 
 /* PHY CTRL bits */
-#define DP83867_PHYCR_FIFO_DEPTH_SHIFT		14
+#define DP83867_PHYCR_TX_FIFO_DEPTH_SHIFT	14
+#define DP83867_PHYCR_RX_FIFO_DEPTH_SHIFT	12
 #define DP83867_PHYCR_FIFO_DEPTH_MAX		0x03
-#define DP83867_PHYCR_FIFO_DEPTH_MASK		GENMASK(15, 14)
+#define DP83867_PHYCR_TX_FIFO_DEPTH_MASK	GENMASK(15, 14)
+#define DP83867_PHYCR_RX_FIFO_DEPTH_MASK	GENMASK(13, 12)
 #define DP83867_PHYCR_RESERVED_MASK		BIT(11)
 
 /* RGMIIDCTL bits */
@@ -131,7 +133,8 @@ enum {
 struct dp83867_private {
 	u32 rx_id_delay;
 	u32 tx_id_delay;
-	u32 fifo_depth;
+	u32 tx_fifo_depth;
+	u32 rx_fifo_depth;
 	int io_impedance;
 	int port_mirroring;
 	bool rxctrl_strap_quirk;
@@ -408,18 +411,32 @@ static int dp83867_of_init(struct phy_device *phydev)
 		dp83867->port_mirroring = DP83867_PORT_MIRROING_DIS;
 
 	ret = of_property_read_u32(of_node, "ti,fifo-depth",
-				   &dp83867->fifo_depth);
+				   &dp83867->tx_fifo_depth);
 	if (ret) {
-		phydev_err(phydev,
-			   "ti,fifo-depth property is required\n");
-		return ret;
+		ret = of_property_read_u32(of_node, "tx-fifo-depth",
+					   &dp83867->tx_fifo_depth);
+		if (ret)
+			dp83867->tx_fifo_depth =
+					DP83867_PHYCR_FIFO_DEPTH_4_B_NIB;
 	}
-	if (dp83867->fifo_depth > DP83867_PHYCR_FIFO_DEPTH_MAX) {
-		phydev_err(phydev,
-			   "ti,fifo-depth value %u out of range\n",
-			   dp83867->fifo_depth);
+
+	if (dp83867->tx_fifo_depth > DP83867_PHYCR_FIFO_DEPTH_MAX) {
+		phydev_err(phydev, "tx-fifo-depth value %u out of range\n",
+			   dp83867->tx_fifo_depth);
+		return -EINVAL;
+	}
+
+	ret = of_property_read_u32(of_node, "rx-fifo-depth",
+				   &dp83867->rx_fifo_depth);
+	if (ret)
+		dp83867->rx_fifo_depth = DP83867_PHYCR_FIFO_DEPTH_4_B_NIB;
+
+	if (dp83867->rx_fifo_depth > DP83867_PHYCR_FIFO_DEPTH_MAX) {
+		phydev_err(phydev, "rx-fifo-depth value %u out of range\n",
+			   dp83867->rx_fifo_depth);
 		return -EINVAL;
 	}
+
 	return 0;
 }
 #else
@@ -458,12 +475,31 @@ static int dp83867_config_init(struct phy_device *phydev)
 		phy_clear_bits_mmd(phydev, DP83867_DEVADDR, DP83867_CFG4,
 				   BIT(7));
 
+	if (phy_interface_is_rgmii(phydev) ||
+	    phydev->interface == PHY_INTERFACE_MODE_SGMII) {
+		val = phy_read(phydev, MII_DP83867_PHYCTRL);
+		if (val < 0)
+			return val;
+
+		val &= ~DP83867_PHYCR_TX_FIFO_DEPTH_MASK;
+		val |= (dp83867->tx_fifo_depth <<
+			DP83867_PHYCR_TX_FIFO_DEPTH_SHIFT);
+
+		if (phydev->interface == PHY_INTERFACE_MODE_SGMII) {
+			val &= ~DP83867_PHYCR_RX_FIFO_DEPTH_MASK;
+			val |= (dp83867->rx_fifo_depth <<
+				DP83867_PHYCR_RX_FIFO_DEPTH_SHIFT);
+		}
+
+		ret = phy_write(phydev, MII_DP83867_PHYCTRL, val);
+		if (ret)
+			return ret;
+	}
+
 	if (phy_interface_is_rgmii(phydev)) {
 		val = phy_read(phydev, MII_DP83867_PHYCTRL);
 		if (val < 0)
 			return val;
-		val &= ~DP83867_PHYCR_FIFO_DEPTH_MASK;
-		val |= (dp83867->fifo_depth << DP83867_PHYCR_FIFO_DEPTH_SHIFT);
 
 		/* The code below checks if "port mirroring" N/A MODE4 has been
 		 * enabled during power on bootstrap.
-- 
cgit v1.2.3-59-g8ed1b


From fe06bf3d83ef0d92f35a24e03297172e92ce9ce3 Mon Sep 17 00:00:00 2001
From: Nathan Chancellor <natechancellor@gmail.com>
Date: Mon, 9 Dec 2019 14:16:23 -0700
Subject: net: tulip: Adjust indentation in {dmfe, uli526x}_init_module

Clang warns:

../drivers/net/ethernet/dec/tulip/uli526x.c:1812:3: warning: misleading
indentation; statement is not part of the previous 'if'
[-Wmisleading-indentation]
        switch (mode) {
        ^
../drivers/net/ethernet/dec/tulip/uli526x.c:1809:2: note: previous
statement is here
        if (cr6set)
        ^
1 warning generated.

../drivers/net/ethernet/dec/tulip/dmfe.c:2217:3: warning: misleading
indentation; statement is not part of the previous 'if'
[-Wmisleading-indentation]
        switch(mode) {
        ^
../drivers/net/ethernet/dec/tulip/dmfe.c:2214:2: note: previous
statement is here
        if (cr6set)
        ^
1 warning generated.

This warning occurs because there is a space before the tab on these
lines. Remove them so that the indentation is consistent with the Linux
kernel coding style and clang no longer warns.

While we are here, adjust the default block in dmfe_init_module to have
a proper break between the label and assignment and add a space between
the switch and opening parentheses to avoid a checkpatch warning.

Fixes: e1c3e5014040 ("[PATCH] initialisation cleanup for ULI526x-net-driver")
Link: https://github.com/ClangBuiltLinux/linux/issues/795
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/dec/tulip/dmfe.c    | 7 ++++---
 drivers/net/ethernet/dec/tulip/uli526x.c | 4 ++--
 2 files changed, 6 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/dec/tulip/dmfe.c b/drivers/net/ethernet/dec/tulip/dmfe.c
index 0efdbd1a4a6f..32d470d4122a 100644
--- a/drivers/net/ethernet/dec/tulip/dmfe.c
+++ b/drivers/net/ethernet/dec/tulip/dmfe.c
@@ -2214,15 +2214,16 @@ static int __init dmfe_init_module(void)
 	if (cr6set)
 		dmfe_cr6_user_set = cr6set;
 
- 	switch(mode) {
-   	case DMFE_10MHF:
+	switch (mode) {
+	case DMFE_10MHF:
 	case DMFE_100MHF:
 	case DMFE_10MFD:
 	case DMFE_100MFD:
 	case DMFE_1M_HPNA:
 		dmfe_media_mode = mode;
 		break;
-	default:dmfe_media_mode = DMFE_AUTO;
+	default:
+		dmfe_media_mode = DMFE_AUTO;
 		break;
 	}
 
diff --git a/drivers/net/ethernet/dec/tulip/uli526x.c b/drivers/net/ethernet/dec/tulip/uli526x.c
index b1f30b194300..117ffe08800d 100644
--- a/drivers/net/ethernet/dec/tulip/uli526x.c
+++ b/drivers/net/ethernet/dec/tulip/uli526x.c
@@ -1809,8 +1809,8 @@ static int __init uli526x_init_module(void)
 	if (cr6set)
 		uli526x_cr6_user_set = cr6set;
 
- 	switch (mode) {
-   	case ULI526X_10MHF:
+	switch (mode) {
+	case ULI526X_10MHF:
 	case ULI526X_100MHF:
 	case ULI526X_10MFD:
 	case ULI526X_100MFD:
-- 
cgit v1.2.3-59-g8ed1b


From 5c61e223004b3b5c3f1dd25718e979bc17a3b12d Mon Sep 17 00:00:00 2001
From: Nathan Chancellor <natechancellor@gmail.com>
Date: Mon, 9 Dec 2019 14:50:27 -0700
Subject: net: smc911x: Adjust indentation in smc911x_phy_configure

Clang warns:

../drivers/net/ethernet/smsc/smc911x.c:939:3: warning: misleading
indentation; statement is not part of the previous 'if'
[-Wmisleading-indentation]
         if (!lp->ctl_rfduplx)
         ^
../drivers/net/ethernet/smsc/smc911x.c:936:2: note: previous statement
is here
        if (lp->ctl_rspeed != 100)
        ^
1 warning generated.

This warning occurs because there is a space after the tab on this line.
Remove it so that the indentation is consistent with the Linux kernel
coding style and clang no longer warns.

Fixes: 0a0c72c9118c ("[PATCH] RE: [PATCH 1/1] net driver: Add support for SMSC LAN911x line of ethernet chips")
Link: https://github.com/ClangBuiltLinux/linux/issues/796
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/smsc/smc911x.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/smsc/smc911x.c b/drivers/net/ethernet/smsc/smc911x.c
index 8d88e4083456..7b65e79d6ae9 100644
--- a/drivers/net/ethernet/smsc/smc911x.c
+++ b/drivers/net/ethernet/smsc/smc911x.c
@@ -936,7 +936,7 @@ static void smc911x_phy_configure(struct work_struct *work)
 	if (lp->ctl_rspeed != 100)
 		my_ad_caps &= ~(ADVERTISE_100BASE4|ADVERTISE_100FULL|ADVERTISE_100HALF);
 
-	 if (!lp->ctl_rfduplx)
+	if (!lp->ctl_rfduplx)
 		my_ad_caps &= ~(ADVERTISE_100FULL|ADVERTISE_10FULL);
 
 	/* Update our Auto-Neg Advertisement Register */
-- 
cgit v1.2.3-59-g8ed1b


From 08cbc75f96029d3092664213a844a5e25523aa35 Mon Sep 17 00:00:00 2001
From: Nathan Chancellor <natechancellor@gmail.com>
Date: Mon, 9 Dec 2019 15:38:59 -0700
Subject: ppp: Adjust indentation into ppp_async_input

Clang warns:

../drivers/net/ppp/ppp_async.c:877:6: warning: misleading indentation;
statement is not part of the previous 'if' [-Wmisleading-indentation]
                                ap->rpkt = skb;
                                ^
../drivers/net/ppp/ppp_async.c:875:5: note: previous statement is here
                                if (!skb)
                                ^
1 warning generated.

This warning occurs because there is a space before the tab on this
line. Clean up this entire block's indentation so that it is consistent
with the Linux kernel coding style and clang no longer warns.

Fixes: 6722e78c9005 ("[PPP]: handle misaligned accesses")
Link: https://github.com/ClangBuiltLinux/linux/issues/800
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ppp/ppp_async.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ppp/ppp_async.c b/drivers/net/ppp/ppp_async.c
index a7b9cf3269bf..29a0917a81e6 100644
--- a/drivers/net/ppp/ppp_async.c
+++ b/drivers/net/ppp/ppp_async.c
@@ -874,15 +874,15 @@ ppp_async_input(struct asyncppp *ap, const unsigned char *buf,
 				skb = dev_alloc_skb(ap->mru + PPP_HDRLEN + 2);
 				if (!skb)
 					goto nomem;
- 				ap->rpkt = skb;
- 			}
- 			if (skb->len == 0) {
- 				/* Try to get the payload 4-byte aligned.
- 				 * This should match the
- 				 * PPP_ALLSTATIONS/PPP_UI/compressed tests in
- 				 * process_input_packet, but we do not have
- 				 * enough chars here to test buf[1] and buf[2].
- 				 */
+				ap->rpkt = skb;
+			}
+			if (skb->len == 0) {
+				/* Try to get the payload 4-byte aligned.
+				 * This should match the
+				 * PPP_ALLSTATIONS/PPP_UI/compressed tests in
+				 * process_input_packet, but we do not have
+				 * enough chars here to test buf[1] and buf[2].
+				 */
 				if (buf[0] != PPP_ALLSTATIONS)
 					skb_reserve(skb, 2 + (buf[0] & 1));
 			}
-- 
cgit v1.2.3-59-g8ed1b


From 02288248b051ae9a9438278148f44bdfb0a4068b Mon Sep 17 00:00:00 2001
From: Jon Maloy <jon.maloy@ericsson.com>
Date: Tue, 10 Dec 2019 00:52:44 +0100
Subject: tipc: eliminate gap indicator from ACK messages

When we increase the link send window we sometimes observe the
following scenario:

1) A packet #N arrives out of order far ahead of a sequence of older
   packets which are still under way. The packet is added to the
   deferred queue.
2) The missing packets arrive in sequence, and for each 16th of them
   an ACK is sent back to the receiver, as it should be.
3) When building those ACK messages, it is checked if there is a gap
   between the link's 'rcv_nxt' and the first packet in the deferred
   queue. This is always the case until packet number #N-1 arrives, and
   a 'gap' indicator is added, effectively turning them into NACK
   messages.
4) When those NACKs arrive at the sender, all the requested
   retransmissions are done, since it is a first-time request.

This sometimes leads to a huge amount of redundant retransmissions,
causing a drop in max throughput. This problem gets worse when we
in a later commit introduce variable window congestion control,
since it drops the link back to 'fast recovery' much more often
than necessary.

We now fix this by not sending any 'gap' indicator in regular ACK
messages. We already have a mechanism for sending explicit NACKs
in place, and this is sufficient to keep up the packet flow.

Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/tipc/link.c | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/net/tipc/link.c b/net/tipc/link.c
index 24d4d10756d3..6d864461f83c 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -1521,7 +1521,8 @@ static int tipc_link_build_nack_msg(struct tipc_link *l,
 				    struct sk_buff_head *xmitq)
 {
 	u32 def_cnt = ++l->stats.deferred_recv;
-	u32 defq_len = skb_queue_len(&l->deferdq);
+	struct sk_buff_head *dfq = &l->deferdq;
+	u32 defq_len = skb_queue_len(dfq);
 	int match1, match2;
 
 	if (link_is_bc_rcvlink(l)) {
@@ -1532,8 +1533,12 @@ static int tipc_link_build_nack_msg(struct tipc_link *l,
 		return 0;
 	}
 
-	if (defq_len >= 3 && !((defq_len - 3) % 16))
-		tipc_link_build_proto_msg(l, STATE_MSG, 0, 0, 0, 0, 0, xmitq);
+	if (defq_len >= 3 && !((defq_len - 3) % 16)) {
+		u16 rcvgap = buf_seqno(skb_peek(dfq)) - l->rcv_nxt;
+
+		tipc_link_build_proto_msg(l, STATE_MSG, 0, 0,
+					  rcvgap, 0, 0, xmitq);
+	}
 	return 0;
 }
 
@@ -1631,7 +1636,7 @@ static void tipc_link_build_proto_msg(struct tipc_link *l, int mtyp, bool probe,
 	if (!tipc_link_is_up(l) && (mtyp == STATE_MSG))
 		return;
 
-	if (!skb_queue_empty(dfq))
+	if ((probe || probe_reply) && !skb_queue_empty(dfq))
 		rcvgap = buf_seqno(skb_peek(dfq)) - l->rcv_nxt;
 
 	skb = tipc_msg_create(LINK_PROTOCOL, mtyp, INT_H_SIZE,
@@ -2079,7 +2084,6 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb,
 		if (rcvgap || reply)
 			tipc_link_build_proto_msg(l, STATE_MSG, 0, reply,
 						  rcvgap, 0, 0, xmitq);
-
 		rc |= tipc_link_advance_transmq(l, ack, gap, ga, xmitq);
 
 		/* If NACK, retransmit will now start at right position */
-- 
cgit v1.2.3-59-g8ed1b


From d3b09995ab930df225929b4153b7187f1bb8a396 Mon Sep 17 00:00:00 2001
From: Jon Maloy <jon.maloy@ericsson.com>
Date: Tue, 10 Dec 2019 00:52:45 +0100
Subject: tipc: eliminate more unnecessary nacks and retransmissions

When we increase the link tranmsit window we often observe the following
scenario:

1) A STATE message bypasses a sequence of traffic packets and arrives
   far ahead of those to the receiver. STATE messages contain a
   'peers_nxt_snt' field to indicate which was the last packet sent
   from the peer. This mechanism is intended as a last resort for the
   receiver to detect missing packets, e.g., during very low traffic
   when there is no packet flow to help early loss detection.
3) The receiving link compares the 'peer_nxt_snt' field to its own
   'rcv_nxt', finds that there is a gap, and immediately sends a
   NACK message back to the peer.
4) When this NACKs arrives at the sender, all the requested
   retransmissions are performed, since it is a first-time request.

Just like in the scenario described in the previous commit this leads
to many redundant retransmissions, with decreased throughput as a
consequence.

We fix this by adding two more conditions before we send a NACK in
this sitution. First, the deferred queue must be empty, so we cannot
assume that the potential packet loss has already been detected by
other means. Second, we check the 'peers_snd_nxt' field only in probe/
probe_reply messages, thus turning this into a true mechanism of last
resort as it was really meant to be.

Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/tipc/link.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/net/tipc/link.c b/net/tipc/link.c
index 6d864461f83c..3528181fd0f3 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -2079,8 +2079,12 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb,
 			     &l->mon_state, l->bearer_id);
 
 		/* Send NACK if peer has sent pkts we haven't received yet */
-		if (more(peers_snd_nxt, rcv_nxt) && !tipc_link_is_synching(l))
+		if ((reply || msg_is_keepalive(hdr)) &&
+		    more(peers_snd_nxt, rcv_nxt) &&
+		    !tipc_link_is_synching(l) &&
+		    skb_queue_empty(&l->deferdq))
 			rcvgap = peers_snd_nxt - l->rcv_nxt;
+
 		if (rcvgap || reply)
 			tipc_link_build_proto_msg(l, STATE_MSG, 0, reply,
 						  rcvgap, 0, 0, xmitq);
-- 
cgit v1.2.3-59-g8ed1b


From 16ad3f4022bb53c7541a0bf0410b32d0231ebef9 Mon Sep 17 00:00:00 2001
From: Jon Maloy <jon.maloy@ericsson.com>
Date: Tue, 10 Dec 2019 00:52:46 +0100
Subject: tipc: introduce variable window congestion control

We introduce a simple variable window congestion control for links.
The algorithm is inspired by the Reno algorithm, covering both 'slow
start', 'congestion avoidance', and 'fast recovery' modes.

- We introduce hard lower and upper window limits per link, still
  different and configurable per bearer type.

- We introduce a 'slow start theshold' variable, initially set to
  the maximum window size.

- We let a link start at the minimum congestion window, i.e. in slow
  start mode, and then let is grow rapidly (+1 per rceived ACK) until
  it reaches the slow start threshold and enters congestion avoidance
  mode.

- In congestion avoidance mode we increment the congestion window for
  each window-size number of acked packets, up to a possible maximum
  equal to the configured maximum window.

- For each non-duplicate NACK received, we drop back to fast recovery
  mode, by setting the both the slow start threshold to and the
  congestion window to (current_congestion_window / 2).

- If the timeout handler finds that the transmit queue has not moved
  since the previous timeout, it drops the link back to slow start
  and forces a probe containing the last sent sequence number to the
  sent to the peer, so that this can discover the stale situation.

This change does in reality have effect only on unicast ethernet
transport, as we have seen that there is no room whatsoever for
increasing the window max size for the UDP bearer.
For now, we also choose to keep the limits for the broadcast link
unchanged and equal.

This algorithm seems to give a 50-100% throughput improvement for
messages larger than MTU.

Suggested-by: Xin Long <lucien.xin@gmail.com>
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/tipc/bcast.c     |  11 ++--
 net/tipc/bearer.c    |  11 ++--
 net/tipc/bearer.h    |   6 +-
 net/tipc/eth_media.c |   3 +-
 net/tipc/ib_media.c  |   5 +-
 net/tipc/link.c      | 175 +++++++++++++++++++++++++++++++++++----------------
 net/tipc/link.h      |   9 +--
 net/tipc/node.c      |  16 ++---
 net/tipc/udp_media.c |   3 +-
 9 files changed, 160 insertions(+), 79 deletions(-)

diff --git a/net/tipc/bcast.c b/net/tipc/bcast.c
index 55aeba681cf4..42e01e9cf893 100644
--- a/net/tipc/bcast.c
+++ b/net/tipc/bcast.c
@@ -562,18 +562,18 @@ int tipc_bclink_reset_stats(struct net *net)
 	return 0;
 }
 
-static int tipc_bc_link_set_queue_limits(struct net *net, u32 limit)
+static int tipc_bc_link_set_queue_limits(struct net *net, u32 max_win)
 {
 	struct tipc_link *l = tipc_bc_sndlink(net);
 
 	if (!l)
 		return -ENOPROTOOPT;
-	if (limit < BCLINK_WIN_MIN)
-		limit = BCLINK_WIN_MIN;
-	if (limit > TIPC_MAX_LINK_WIN)
+	if (max_win < BCLINK_WIN_MIN)
+		max_win = BCLINK_WIN_MIN;
+	if (max_win > TIPC_MAX_LINK_WIN)
 		return -EINVAL;
 	tipc_bcast_lock(net);
-	tipc_link_set_queue_limits(l, limit);
+	tipc_link_set_queue_limits(l, BCLINK_WIN_MIN, max_win);
 	tipc_bcast_unlock(net);
 	return 0;
 }
@@ -683,6 +683,7 @@ int tipc_bcast_init(struct net *net)
 	if (!tipc_link_bc_create(net, 0, 0,
 				 FB_MTU,
 				 BCLINK_WIN_DEFAULT,
+				 BCLINK_WIN_DEFAULT,
 				 0,
 				 &bb->inputq,
 				 NULL,
diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c
index d7ec26bd739d..34ca7b789eba 100644
--- a/net/tipc/bearer.c
+++ b/net/tipc/bearer.c
@@ -311,7 +311,8 @@ static int tipc_enable_bearer(struct net *net, const char *name,
 
 	b->identity = bearer_id;
 	b->tolerance = m->tolerance;
-	b->window = m->window;
+	b->min_win = m->min_win;
+	b->max_win = m->max_win;
 	b->domain = disc_domain;
 	b->net_plane = bearer_id + 'A';
 	b->priority = prio;
@@ -796,7 +797,7 @@ static int __tipc_nl_add_bearer(struct tipc_nl_msg *msg,
 		goto prop_msg_full;
 	if (nla_put_u32(msg->skb, TIPC_NLA_PROP_TOL, bearer->tolerance))
 		goto prop_msg_full;
-	if (nla_put_u32(msg->skb, TIPC_NLA_PROP_WIN, bearer->window))
+	if (nla_put_u32(msg->skb, TIPC_NLA_PROP_WIN, bearer->max_win))
 		goto prop_msg_full;
 	if (bearer->media->type_id == TIPC_MEDIA_TYPE_UDP)
 		if (nla_put_u32(msg->skb, TIPC_NLA_PROP_MTU, bearer->mtu))
@@ -1088,7 +1089,7 @@ int __tipc_nl_bearer_set(struct sk_buff *skb, struct genl_info *info)
 		if (props[TIPC_NLA_PROP_PRIO])
 			b->priority = nla_get_u32(props[TIPC_NLA_PROP_PRIO]);
 		if (props[TIPC_NLA_PROP_WIN])
-			b->window = nla_get_u32(props[TIPC_NLA_PROP_WIN]);
+			b->max_win = nla_get_u32(props[TIPC_NLA_PROP_WIN]);
 		if (props[TIPC_NLA_PROP_MTU]) {
 			if (b->media->type_id != TIPC_MEDIA_TYPE_UDP)
 				return -EINVAL;
@@ -1142,7 +1143,7 @@ static int __tipc_nl_add_media(struct tipc_nl_msg *msg,
 		goto prop_msg_full;
 	if (nla_put_u32(msg->skb, TIPC_NLA_PROP_TOL, media->tolerance))
 		goto prop_msg_full;
-	if (nla_put_u32(msg->skb, TIPC_NLA_PROP_WIN, media->window))
+	if (nla_put_u32(msg->skb, TIPC_NLA_PROP_WIN, media->max_win))
 		goto prop_msg_full;
 	if (media->type_id == TIPC_MEDIA_TYPE_UDP)
 		if (nla_put_u32(msg->skb, TIPC_NLA_PROP_MTU, media->mtu))
@@ -1275,7 +1276,7 @@ int __tipc_nl_media_set(struct sk_buff *skb, struct genl_info *info)
 		if (props[TIPC_NLA_PROP_PRIO])
 			m->priority = nla_get_u32(props[TIPC_NLA_PROP_PRIO]);
 		if (props[TIPC_NLA_PROP_WIN])
-			m->window = nla_get_u32(props[TIPC_NLA_PROP_WIN]);
+			m->max_win = nla_get_u32(props[TIPC_NLA_PROP_WIN]);
 		if (props[TIPC_NLA_PROP_MTU]) {
 			if (m->type_id != TIPC_MEDIA_TYPE_UDP)
 				return -EINVAL;
diff --git a/net/tipc/bearer.h b/net/tipc/bearer.h
index d0c79cc6c0c2..bc0023119da2 100644
--- a/net/tipc/bearer.h
+++ b/net/tipc/bearer.h
@@ -119,7 +119,8 @@ struct tipc_media {
 			char *raw);
 	u32 priority;
 	u32 tolerance;
-	u32 window;
+	u32 min_win;
+	u32 max_win;
 	u32 mtu;
 	u32 type_id;
 	u32 hwaddr_len;
@@ -158,7 +159,8 @@ struct tipc_bearer {
 	struct packet_type pt;
 	struct rcu_head rcu;
 	u32 priority;
-	u32 window;
+	u32 min_win;
+	u32 max_win;
 	u32 tolerance;
 	u32 domain;
 	u32 identity;
diff --git a/net/tipc/eth_media.c b/net/tipc/eth_media.c
index f69a2fde9f4a..8b0bb600602d 100644
--- a/net/tipc/eth_media.c
+++ b/net/tipc/eth_media.c
@@ -92,7 +92,8 @@ struct tipc_media eth_media_info = {
 	.raw2addr	= tipc_eth_raw2addr,
 	.priority	= TIPC_DEF_LINK_PRI,
 	.tolerance	= TIPC_DEF_LINK_TOL,
-	.window		= TIPC_DEF_LINK_WIN,
+	.min_win	= TIPC_DEF_LINK_WIN,
+	.max_win	= TIPC_MAX_LINK_WIN,
 	.type_id	= TIPC_MEDIA_TYPE_ETH,
 	.hwaddr_len	= ETH_ALEN,
 	.name		= "eth"
diff --git a/net/tipc/ib_media.c b/net/tipc/ib_media.c
index e8c16718e3fa..7aa9ff88458d 100644
--- a/net/tipc/ib_media.c
+++ b/net/tipc/ib_media.c
@@ -42,6 +42,8 @@
 #include "core.h"
 #include "bearer.h"
 
+#define TIPC_MAX_IB_LINK_WIN 500
+
 /* convert InfiniBand address (media address format) media address to string */
 static int tipc_ib_addr2str(struct tipc_media_addr *a, char *str_buf,
 			    int str_size)
@@ -94,7 +96,8 @@ struct tipc_media ib_media_info = {
 	.raw2addr	= tipc_ib_raw2addr,
 	.priority	= TIPC_DEF_LINK_PRI,
 	.tolerance	= TIPC_DEF_LINK_TOL,
-	.window		= TIPC_DEF_LINK_WIN,
+	.min_win	= TIPC_DEF_LINK_WIN,
+	.max_win	= TIPC_MAX_IB_LINK_WIN,
 	.type_id	= TIPC_MEDIA_TYPE_IB,
 	.hwaddr_len	= INFINIBAND_ALEN,
 	.name		= "ib"
diff --git a/net/tipc/link.c b/net/tipc/link.c
index 3528181fd0f3..94dd48cd70a3 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -164,7 +164,6 @@ struct tipc_link {
 		struct sk_buff *target_bskb;
 	} backlog[5];
 	u16 snd_nxt;
-	u16 window;
 
 	/* Reception */
 	u16 rcv_nxt;
@@ -175,6 +174,12 @@ struct tipc_link {
 
 	/* Congestion handling */
 	struct sk_buff_head wakeupq;
+	u16 window;
+	u16 min_win;
+	u16 ssthresh;
+	u16 max_win;
+	u16 cong_acks;
+	u16 checkpoint;
 
 	/* Fragmentation/reassembly */
 	struct sk_buff *reasm_buf;
@@ -244,12 +249,13 @@ static int tipc_link_build_nack_msg(struct tipc_link *l,
 				    struct sk_buff_head *xmitq);
 static void tipc_link_build_bc_init_msg(struct tipc_link *l,
 					struct sk_buff_head *xmitq);
-static bool tipc_link_release_pkts(struct tipc_link *l, u16 to);
+static int tipc_link_release_pkts(struct tipc_link *l, u16 to);
 static u16 tipc_build_gap_ack_blks(struct tipc_link *l, void *data);
 static int tipc_link_advance_transmq(struct tipc_link *l, u16 acked, u16 gap,
 				     struct tipc_gap_ack_blks *ga,
 				     struct sk_buff_head *xmitq);
-
+static void tipc_link_update_cwin(struct tipc_link *l, int released,
+				  bool retransmitted);
 /*
  *  Simple non-static link routines (i.e. referenced outside this file)
  */
@@ -308,9 +314,14 @@ u32 tipc_link_id(struct tipc_link *l)
 	return l->peer_bearer_id << 16 | l->bearer_id;
 }
 
-int tipc_link_window(struct tipc_link *l)
+int tipc_link_min_win(struct tipc_link *l)
 {
-	return l->window;
+	return l->min_win;
+}
+
+int tipc_link_max_win(struct tipc_link *l)
+{
+	return l->max_win;
 }
 
 int tipc_link_prio(struct tipc_link *l)
@@ -436,7 +447,8 @@ u32 tipc_link_state(struct tipc_link *l)
  * @net_plane: network plane (A,B,c..) this link belongs to
  * @mtu: mtu to be advertised by link
  * @priority: priority to be used by link
- * @window: send window to be used by link
+ * @min_win: minimal send window to be used by link
+ * @max_win: maximal send window to be used by link
  * @session: session to be used by link
  * @ownnode: identity of own node
  * @peer: node id of peer node
@@ -451,7 +463,7 @@ u32 tipc_link_state(struct tipc_link *l)
  */
 bool tipc_link_create(struct net *net, char *if_name, int bearer_id,
 		      int tolerance, char net_plane, u32 mtu, int priority,
-		      int window, u32 session, u32 self,
+		      u32 min_win, u32 max_win, u32 session, u32 self,
 		      u32 peer, u8 *peer_id, u16 peer_caps,
 		      struct tipc_link *bc_sndlink,
 		      struct tipc_link *bc_rcvlink,
@@ -495,7 +507,7 @@ bool tipc_link_create(struct net *net, char *if_name, int bearer_id,
 	l->advertised_mtu = mtu;
 	l->mtu = mtu;
 	l->priority = priority;
-	tipc_link_set_queue_limits(l, window);
+	tipc_link_set_queue_limits(l, min_win, max_win);
 	l->ackers = 1;
 	l->bc_sndlink = bc_sndlink;
 	l->bc_rcvlink = bc_rcvlink;
@@ -523,7 +535,7 @@ bool tipc_link_create(struct net *net, char *if_name, int bearer_id,
  * Returns true if link was created, otherwise false
  */
 bool tipc_link_bc_create(struct net *net, u32 ownnode, u32 peer,
-			 int mtu, int window, u16 peer_caps,
+			 int mtu, u32 min_win, u32 max_win, u16 peer_caps,
 			 struct sk_buff_head *inputq,
 			 struct sk_buff_head *namedq,
 			 struct tipc_link *bc_sndlink,
@@ -531,9 +543,9 @@ bool tipc_link_bc_create(struct net *net, u32 ownnode, u32 peer,
 {
 	struct tipc_link *l;
 
-	if (!tipc_link_create(net, "", MAX_BEARERS, 0, 'Z', mtu, 0, window,
-			      0, ownnode, peer, NULL, peer_caps, bc_sndlink,
-			      NULL, inputq, namedq, link))
+	if (!tipc_link_create(net, "", MAX_BEARERS, 0, 'Z', mtu, 0, min_win,
+			      max_win, 0, ownnode, peer, NULL, peer_caps,
+			      bc_sndlink, NULL, inputq, namedq, link))
 		return false;
 
 	l = *link;
@@ -772,6 +784,8 @@ bool tipc_link_too_silent(struct tipc_link *l)
 	return (l->silent_intv_cnt + 2 > l->abort_limit);
 }
 
+static int tipc_link_bc_retrans(struct tipc_link *l, struct tipc_link *r,
+				u16 from, u16 to, struct sk_buff_head *xmitq);
 /* tipc_link_timeout - perform periodic task as instructed from node timeout
  */
 int tipc_link_timeout(struct tipc_link *l, struct sk_buff_head *xmitq)
@@ -804,6 +818,11 @@ int tipc_link_timeout(struct tipc_link *l, struct sk_buff_head *xmitq)
 		probe |= l->silent_intv_cnt;
 		if (probe || mstate->monitoring)
 			l->silent_intv_cnt++;
+		if (l->snd_nxt == l->checkpoint) {
+			tipc_link_update_cwin(l, 0, 0);
+			probe = true;
+		}
+		l->checkpoint = l->snd_nxt;
 		break;
 	case LINK_RESET:
 		setup = l->rst_cnt++ <= 4;
@@ -959,7 +978,7 @@ int tipc_link_xmit(struct tipc_link *l, struct sk_buff_head *list,
 	int pkt_cnt = skb_queue_len(list);
 	int imp = msg_importance(hdr);
 	unsigned int mss = tipc_link_mss(l);
-	unsigned int maxwin = l->window;
+	unsigned int cwin = l->window;
 	unsigned int mtu = l->mtu;
 	bool new_bundle;
 	int rc = 0;
@@ -988,7 +1007,7 @@ int tipc_link_xmit(struct tipc_link *l, struct sk_buff_head *list,
 
 	/* Prepare each packet for sending, and add to relevant queue: */
 	while ((skb = __skb_dequeue(list))) {
-		if (likely(skb_queue_len(transmq) < maxwin)) {
+		if (likely(skb_queue_len(transmq) < cwin)) {
 			hdr = buf_msg(skb);
 			msg_set_seqno(hdr, seqno);
 			msg_set_ack(hdr, ack);
@@ -1035,17 +1054,61 @@ int tipc_link_xmit(struct tipc_link *l, struct sk_buff_head *list,
 	return rc;
 }
 
+static void tipc_link_update_cwin(struct tipc_link *l, int released,
+				  bool retransmitted)
+{
+	int bklog_len = skb_queue_len(&l->backlogq);
+	struct sk_buff_head *txq = &l->transmq;
+	int txq_len = skb_queue_len(txq);
+	u16 cwin = l->window;
+
+	/* Enter fast recovery */
+	if (unlikely(retransmitted)) {
+		l->ssthresh = max_t(u16, l->window / 2, 300);
+		l->window = l->ssthresh;
+		return;
+	}
+	/* Enter slow start */
+	if (unlikely(!released)) {
+		l->ssthresh = max_t(u16, l->window / 2, 300);
+		l->window = l->min_win;
+		return;
+	}
+	/* Don't increase window if no pressure on the transmit queue */
+	if (txq_len + bklog_len < cwin)
+		return;
+
+	/* Don't increase window if there are holes the transmit queue */
+	if (txq_len && l->snd_nxt - buf_seqno(skb_peek(txq)) != txq_len)
+		return;
+
+	l->cong_acks += released;
+
+	/* Slow start  */
+	if (cwin <= l->ssthresh) {
+		l->window = min_t(u16, cwin + released, l->max_win);
+		return;
+	}
+	/* Congestion avoidance */
+	if (l->cong_acks < cwin)
+		return;
+	l->window = min_t(u16, ++cwin, l->max_win);
+	l->cong_acks = 0;
+}
+
 static void tipc_link_advance_backlog(struct tipc_link *l,
 				      struct sk_buff_head *xmitq)
 {
+	u16 bc_ack = l->bc_rcvlink->rcv_nxt - 1;
+	struct sk_buff_head *txq = &l->transmq;
 	struct sk_buff *skb, *_skb;
-	struct tipc_msg *hdr;
-	u16 seqno = l->snd_nxt;
 	u16 ack = l->rcv_nxt - 1;
-	u16 bc_ack = l->bc_rcvlink->rcv_nxt - 1;
+	u16 seqno = l->snd_nxt;
+	struct tipc_msg *hdr;
+	u16 cwin = l->window;
 	u32 imp;
 
-	while (skb_queue_len(&l->transmq) < l->window) {
+	while (skb_queue_len(txq) < cwin) {
 		skb = skb_peek(&l->backlogq);
 		if (!skb)
 			break;
@@ -1141,6 +1204,7 @@ static int tipc_link_bc_retrans(struct tipc_link *l, struct tipc_link *r,
 	struct sk_buff *_skb, *skb = skb_peek(&l->transmq);
 	u16 bc_ack = l->bc_rcvlink->rcv_nxt - 1;
 	u16 ack = l->rcv_nxt - 1;
+	int retransmitted = 0;
 	struct tipc_msg *hdr;
 	int rc = 0;
 
@@ -1160,7 +1224,6 @@ static int tipc_link_bc_retrans(struct tipc_link *l, struct tipc_link *r,
 			continue;
 		if (more(msg_seqno(hdr), to))
 			break;
-
 		if (time_before(jiffies, TIPC_SKB_CB(skb)->nxt_retr))
 			continue;
 		TIPC_SKB_CB(skb)->nxt_retr = TIPC_BC_RETR_LIM;
@@ -1173,11 +1236,12 @@ static int tipc_link_bc_retrans(struct tipc_link *l, struct tipc_link *r,
 		_skb->priority = TC_PRIO_CONTROL;
 		__skb_queue_tail(xmitq, _skb);
 		l->stats.retransmitted++;
-
+		retransmitted++;
 		/* Increase actual retrans counter & mark first time */
 		if (!TIPC_SKB_CB(skb)->retr_cnt++)
 			TIPC_SKB_CB(skb)->retr_stamp = jiffies;
 	}
+	tipc_link_update_cwin(l, 0, retransmitted);
 	return 0;
 }
 
@@ -1338,9 +1402,9 @@ static int tipc_link_tnl_rcv(struct tipc_link *l, struct sk_buff *skb,
 	return rc;
 }
 
-static bool tipc_link_release_pkts(struct tipc_link *l, u16 acked)
+static int tipc_link_release_pkts(struct tipc_link *l, u16 acked)
 {
-	bool released = false;
+	int released = 0;
 	struct sk_buff *skb, *tmp;
 
 	skb_queue_walk_safe(&l->transmq, skb, tmp) {
@@ -1348,7 +1412,7 @@ static bool tipc_link_release_pkts(struct tipc_link *l, u16 acked)
 			break;
 		__skb_unlink(skb, &l->transmq);
 		kfree_skb(skb);
-		released = true;
+		released++;
 	}
 	return released;
 }
@@ -1417,8 +1481,10 @@ static int tipc_link_advance_transmq(struct tipc_link *l, u16 acked, u16 gap,
 	struct sk_buff *skb, *_skb, *tmp;
 	struct tipc_msg *hdr;
 	u16 bc_ack = l->bc_rcvlink->rcv_nxt - 1;
+	bool retransmitted = false;
 	u16 ack = l->rcv_nxt - 1;
 	bool passed = false;
+	u16 released = 0;
 	u16 seqno, n = 0;
 	int rc = 0;
 
@@ -1430,6 +1496,7 @@ next_gap_ack:
 			/* release skb */
 			__skb_unlink(skb, &l->transmq);
 			kfree_skb(skb);
+			released++;
 		} else if (less_eq(seqno, acked + gap)) {
 			/* First, check if repeated retrans failures occurs? */
 			if (!passed && link_retransmit_failure(l, l, &rc))
@@ -1449,7 +1516,7 @@ next_gap_ack:
 			_skb->priority = TC_PRIO_CONTROL;
 			__skb_queue_tail(xmitq, _skb);
 			l->stats.retransmitted++;
-
+			retransmitted = true;
 			/* Increase actual retrans counter & mark first time */
 			if (!TIPC_SKB_CB(skb)->retr_cnt++)
 				TIPC_SKB_CB(skb)->retr_stamp = jiffies;
@@ -1463,7 +1530,10 @@ next_gap_ack:
 			goto next_gap_ack;
 		}
 	}
-
+	if (released || retransmitted)
+		tipc_link_update_cwin(l, released, retransmitted);
+	if (released)
+		tipc_link_advance_backlog(l, xmitq);
 	return 0;
 }
 
@@ -1487,7 +1557,6 @@ int tipc_link_build_state_msg(struct tipc_link *l, struct sk_buff_head *xmitq)
 		l->snd_nxt = l->rcv_nxt;
 		return TIPC_LINK_SND_STATE;
 	}
-
 	/* Unicast ACK */
 	l->rcv_unacked = 0;
 	l->stats.sent_acks++;
@@ -1553,6 +1622,7 @@ int tipc_link_rcv(struct tipc_link *l, struct sk_buff *skb,
 	struct sk_buff_head *defq = &l->deferdq;
 	struct tipc_msg *hdr = buf_msg(skb);
 	u16 seqno, rcv_nxt, win_lim;
+	int released = 0;
 	int rc = 0;
 
 	/* Verify and update link state */
@@ -1571,21 +1641,17 @@ int tipc_link_rcv(struct tipc_link *l, struct sk_buff *skb,
 		if (unlikely(!link_is_up(l))) {
 			if (l->state == LINK_ESTABLISHING)
 				rc = TIPC_LINK_UP_EVT;
-			goto drop;
+			kfree_skb(skb);
+			break;
 		}
 
 		/* Drop if outside receive window */
 		if (unlikely(less(seqno, rcv_nxt) || more(seqno, win_lim))) {
 			l->stats.duplicates++;
-			goto drop;
-		}
-
-		/* Forward queues and wake up waiting users */
-		if (likely(tipc_link_release_pkts(l, msg_ack(hdr)))) {
-			tipc_link_advance_backlog(l, xmitq);
-			if (unlikely(!skb_queue_empty(&l->wakeupq)))
-				link_prepare_wakeup(l);
+			kfree_skb(skb);
+			break;
 		}
+		released += tipc_link_release_pkts(l, msg_ack(hdr));
 
 		/* Defer delivery if sequence gap */
 		if (unlikely(seqno != rcv_nxt)) {
@@ -1608,9 +1674,13 @@ int tipc_link_rcv(struct tipc_link *l, struct sk_buff *skb,
 			break;
 	} while ((skb = __tipc_skb_dequeue(defq, l->rcv_nxt)));
 
-	return rc;
-drop:
-	kfree_skb(skb);
+	/* Forward queues and wake up waiting users */
+	if (released) {
+		tipc_link_update_cwin(l, released, 0);
+		tipc_link_advance_backlog(l, xmitq);
+		if (unlikely(!skb_queue_empty(&l->wakeupq)))
+			link_prepare_wakeup(l);
+	}
 	return rc;
 }
 
@@ -2084,17 +2154,13 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb,
 		    !tipc_link_is_synching(l) &&
 		    skb_queue_empty(&l->deferdq))
 			rcvgap = peers_snd_nxt - l->rcv_nxt;
-
 		if (rcvgap || reply)
 			tipc_link_build_proto_msg(l, STATE_MSG, 0, reply,
 						  rcvgap, 0, 0, xmitq);
-		rc |= tipc_link_advance_transmq(l, ack, gap, ga, xmitq);
 
-		/* If NACK, retransmit will now start at right position */
+		rc |= tipc_link_advance_transmq(l, ack, gap, ga, xmitq);
 		if (gap)
 			l->stats.recv_nacks++;
-
-		tipc_link_advance_backlog(l, xmitq);
 		if (unlikely(!skb_queue_empty(&l->wakeupq)))
 			link_prepare_wakeup(l);
 	}
@@ -2313,15 +2379,18 @@ int tipc_link_bc_nack_rcv(struct tipc_link *l, struct sk_buff *skb,
 	return 0;
 }
 
-void tipc_link_set_queue_limits(struct tipc_link *l, u32 win)
+void tipc_link_set_queue_limits(struct tipc_link *l, u32 min_win, u32 max_win)
 {
 	int max_bulk = TIPC_MAX_PUBL / (l->mtu / ITEM_SIZE);
 
-	l->window = win;
-	l->backlog[TIPC_LOW_IMPORTANCE].limit      = max_t(u16, 50, win);
-	l->backlog[TIPC_MEDIUM_IMPORTANCE].limit   = max_t(u16, 100, win * 2);
-	l->backlog[TIPC_HIGH_IMPORTANCE].limit     = max_t(u16, 150, win * 3);
-	l->backlog[TIPC_CRITICAL_IMPORTANCE].limit = max_t(u16, 200, win * 4);
+	l->min_win = min_win;
+	l->ssthresh = max_win;
+	l->max_win = max_win;
+	l->window = min_win;
+	l->backlog[TIPC_LOW_IMPORTANCE].limit      = min_win * 2;
+	l->backlog[TIPC_MEDIUM_IMPORTANCE].limit   = min_win * 4;
+	l->backlog[TIPC_HIGH_IMPORTANCE].limit     = min_win * 6;
+	l->backlog[TIPC_CRITICAL_IMPORTANCE].limit = min_win * 8;
 	l->backlog[TIPC_SYSTEM_IMPORTANCE].limit   = max_bulk;
 }
 
@@ -2374,10 +2443,10 @@ int tipc_nl_parse_link_prop(struct nlattr *prop, struct nlattr *props[])
 	}
 
 	if (props[TIPC_NLA_PROP_WIN]) {
-		u32 win;
+		u32 max_win;
 
-		win = nla_get_u32(props[TIPC_NLA_PROP_WIN]);
-		if ((win < TIPC_MIN_LINK_WIN) || (win > TIPC_MAX_LINK_WIN))
+		max_win = nla_get_u32(props[TIPC_NLA_PROP_WIN]);
+		if (max_win < TIPC_DEF_LINK_WIN || max_win > TIPC_MAX_LINK_WIN)
 			return -EINVAL;
 	}
 
@@ -2613,7 +2682,7 @@ int tipc_nl_add_bc_link(struct net *net, struct tipc_nl_msg *msg)
 	prop = nla_nest_start_noflag(msg->skb, TIPC_NLA_LINK_PROP);
 	if (!prop)
 		goto attr_msg_full;
-	if (nla_put_u32(msg->skb, TIPC_NLA_PROP_WIN, bcl->window))
+	if (nla_put_u32(msg->skb, TIPC_NLA_PROP_WIN, bcl->max_win))
 		goto prop_msg_full;
 	if (nla_put_u32(msg->skb, TIPC_NLA_PROP_BROADCAST, bc_mode))
 		goto prop_msg_full;
diff --git a/net/tipc/link.h b/net/tipc/link.h
index c09e9d49d0a3..d3c1c3fc1659 100644
--- a/net/tipc/link.h
+++ b/net/tipc/link.h
@@ -73,7 +73,7 @@ enum {
 
 bool tipc_link_create(struct net *net, char *if_name, int bearer_id,
 		      int tolerance, char net_plane, u32 mtu, int priority,
-		      int window, u32 session, u32 ownnode,
+		      u32 min_win, u32 max_win, u32 session, u32 ownnode,
 		      u32 peer, u8 *peer_id, u16 peer_caps,
 		      struct tipc_link *bc_sndlink,
 		      struct tipc_link *bc_rcvlink,
@@ -81,7 +81,7 @@ bool tipc_link_create(struct net *net, char *if_name, int bearer_id,
 		      struct sk_buff_head *namedq,
 		      struct tipc_link **link);
 bool tipc_link_bc_create(struct net *net, u32 ownnode, u32 peer,
-			 int mtu, int window, u16 peer_caps,
+			 int mtu, u32 min_win, u32 max_win, u16 peer_caps,
 			 struct sk_buff_head *inputq,
 			 struct sk_buff_head *namedq,
 			 struct tipc_link *bc_sndlink,
@@ -115,7 +115,8 @@ char *tipc_link_name_ext(struct tipc_link *l, char *buf);
 u32 tipc_link_state(struct tipc_link *l);
 char tipc_link_plane(struct tipc_link *l);
 int tipc_link_prio(struct tipc_link *l);
-int tipc_link_window(struct tipc_link *l);
+int tipc_link_min_win(struct tipc_link *l);
+int tipc_link_max_win(struct tipc_link *l);
 void tipc_link_update_caps(struct tipc_link *l, u16 capabilities);
 bool tipc_link_validate_msg(struct tipc_link *l, struct tipc_msg *hdr);
 unsigned long tipc_link_tolerance(struct tipc_link *l);
@@ -124,7 +125,7 @@ void tipc_link_set_tolerance(struct tipc_link *l, u32 tol,
 void tipc_link_set_prio(struct tipc_link *l, u32 prio,
 			struct sk_buff_head *xmitq);
 void tipc_link_set_abort_limit(struct tipc_link *l, u32 limit);
-void tipc_link_set_queue_limits(struct tipc_link *l, u32 window);
+void tipc_link_set_queue_limits(struct tipc_link *l, u32 min_win, u32 max_win);
 int __tipc_nl_add_link(struct net *net, struct tipc_nl_msg *msg,
 		       struct tipc_link *link, int nlflags);
 int tipc_nl_parse_link_prop(struct nlattr *prop, struct nlattr *props[]);
diff --git a/net/tipc/node.c b/net/tipc/node.c
index ab04e00cb95b..99b28b69fc17 100644
--- a/net/tipc/node.c
+++ b/net/tipc/node.c
@@ -1139,7 +1139,8 @@ void tipc_node_check_dest(struct net *net, u32 addr,
 		snd_l = tipc_bc_sndlink(net);
 		if (!tipc_link_bc_create(net, tipc_own_addr(net),
 					 addr, U16_MAX,
-					 tipc_link_window(snd_l),
+					 tipc_link_min_win(snd_l),
+					 tipc_link_max_win(snd_l),
 					 n->capabilities,
 					 &n->bc_entry.inputq1,
 					 &n->bc_entry.namedq, snd_l,
@@ -1233,7 +1234,7 @@ void tipc_node_check_dest(struct net *net, u32 addr,
 		get_random_bytes(&session, sizeof(u16));
 		if (!tipc_link_create(net, if_name, b->identity, b->tolerance,
 				      b->net_plane, b->mtu, b->priority,
-				      b->window, session,
+				      b->min_win, b->max_win, session,
 				      tipc_own_addr(net), addr, peer_id,
 				      n->capabilities,
 				      tipc_bc_sndlink(n->net), n->bc_entry.link,
@@ -2360,8 +2361,7 @@ int tipc_nl_node_set_link(struct sk_buff *skb, struct genl_info *info)
 	if (attrs[TIPC_NLA_LINK_PROP]) {
 		struct nlattr *props[TIPC_NLA_PROP_MAX + 1];
 
-		err = tipc_nl_parse_link_prop(attrs[TIPC_NLA_LINK_PROP],
-					      props);
+		err = tipc_nl_parse_link_prop(attrs[TIPC_NLA_LINK_PROP], props);
 		if (err) {
 			res = err;
 			goto out;
@@ -2380,10 +2380,12 @@ int tipc_nl_node_set_link(struct sk_buff *skb, struct genl_info *info)
 			tipc_link_set_prio(link, prio, &xmitq);
 		}
 		if (props[TIPC_NLA_PROP_WIN]) {
-			u32 win;
+			u32 max_win;
 
-			win = nla_get_u32(props[TIPC_NLA_PROP_WIN]);
-			tipc_link_set_queue_limits(link, win);
+			max_win = nla_get_u32(props[TIPC_NLA_PROP_WIN]);
+			tipc_link_set_queue_limits(link,
+						   tipc_link_min_win(link),
+						   max_win);
 		}
 	}
 
diff --git a/net/tipc/udp_media.c b/net/tipc/udp_media.c
index ed113735c019..d6620ad53546 100644
--- a/net/tipc/udp_media.c
+++ b/net/tipc/udp_media.c
@@ -828,7 +828,8 @@ struct tipc_media udp_media_info = {
 	.msg2addr	= tipc_udp_msg2addr,
 	.priority	= TIPC_DEF_LINK_PRI,
 	.tolerance	= TIPC_DEF_LINK_TOL,
-	.window		= TIPC_DEF_LINK_WIN,
+	.min_win	= TIPC_DEF_LINK_WIN,
+	.max_win	= TIPC_DEF_LINK_WIN,
 	.mtu		= TIPC_DEF_LINK_UDP_MTU,
 	.type_id	= TIPC_MEDIA_TYPE_UDP,
 	.hwaddr_len	= 0,
-- 
cgit v1.2.3-59-g8ed1b


From 4caefbce06d138306ef4e9f635e5b02b62d56911 Mon Sep 17 00:00:00 2001
From: Michael Walle <michael@walle.cc>
Date: Tue, 10 Dec 2019 01:15:37 +0100
Subject: enetc: add software timestamping

Provide a software TX timestamp and add it to the ethtool query
interface.

skb_tx_timestamp() is also needed if one would like to use PHY
timestamping.

Signed-off-by: Michael Walle <michael@walle.cc>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/freescale/enetc/enetc.c         | 2 ++
 drivers/net/ethernet/freescale/enetc/enetc_ethtool.c | 1 +
 2 files changed, 3 insertions(+)

diff --git a/drivers/net/ethernet/freescale/enetc/enetc.c b/drivers/net/ethernet/freescale/enetc/enetc.c
index 17739906c966..2ee4a2cd4780 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc.c
+++ b/drivers/net/ethernet/freescale/enetc/enetc.c
@@ -227,6 +227,8 @@ static int enetc_map_tx_buffs(struct enetc_bdr *tx_ring, struct sk_buff *skb,
 	enetc_bdr_idx_inc(tx_ring, &i);
 	tx_ring->next_to_use = i;
 
+	skb_tx_timestamp(skb);
+
 	/* let H/W know BD ring has been updated */
 	enetc_wr_reg(tx_ring->tpir, i); /* includes wmb() */
 
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c
index 880a8ed8bb47..301ee0dde02d 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c
+++ b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c
@@ -579,6 +579,7 @@ static int enetc_get_ts_info(struct net_device *ndev,
 			   (1 << HWTSTAMP_FILTER_ALL);
 #else
 	info->so_timestamping = SOF_TIMESTAMPING_RX_SOFTWARE |
+				SOF_TIMESTAMPING_TX_SOFTWARE |
 				SOF_TIMESTAMPING_SOFTWARE;
 #endif
 	return 0;
-- 
cgit v1.2.3-59-g8ed1b


From 5000b28b0b1a34144b39376318cafb8c2a0f79fd Mon Sep 17 00:00:00 2001
From: Kuniyuki Iwashima <kuni1840@gmail.com>
Date: Tue, 10 Dec 2019 02:41:48 +0000
Subject: tcp: Cleanup duplicate initialization of sk->sk_state.

When a TCP socket is created, sk->sk_state is initialized twice as
TCP_CLOSE in sock_init_data() and tcp_init_sock(). The tcp_init_sock() is
always called after the sock_init_data(), so it is not necessary to update
sk->sk_state in the tcp_init_sock().

Before v2.1.8, the code of the two functions was in the inet_create(). In
the patch of v2.1.8, the tcp_v4/v6_init_sock() were added and the code of
initialization of sk->state was duplicated.

Signed-off-by: Kuniyuki Iwashima <kuni1840@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/tcp.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 8a39ee794891..09e2cae92956 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -443,8 +443,6 @@ void tcp_init_sock(struct sock *sk)
 	tp->tsoffset = 0;
 	tp->rack.reo_wnd_steps = 1;
 
-	sk->sk_state = TCP_CLOSE;
-
 	sk->sk_write_space = sk_stream_write_space;
 	sock_set_flag(sk, SOCK_USE_WRITE_QUEUE);
 
-- 
cgit v1.2.3-59-g8ed1b


From ca866ee825911cb70e64ad5b33ac73d4ab924c1f Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <jakub.kicinski@netronome.com>
Date: Mon, 9 Dec 2019 21:04:01 -0800
Subject: nfp: add support for TLV device stats

Device stats are currently hard coded in the PCI BAR0 layout.
Add a ability to read them from the TLV area instead.
Names for the stats are maintained by the driver, and their
meaning documented. This allows us to more easily add and
remove device stats.

Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../networking/device_drivers/netronome/nfp.rst    | 116 +++++++++++++++++++++
 drivers/net/ethernet/netronome/nfp/nfp_net_ctrl.c  |   9 ++
 drivers/net/ethernet/netronome/nfp/nfp_net_ctrl.h  |  16 +++
 .../net/ethernet/netronome/nfp/nfp_net_ethtool.c   | 108 +++++++++++++++++--
 4 files changed, 242 insertions(+), 7 deletions(-)

diff --git a/Documentation/networking/device_drivers/netronome/nfp.rst b/Documentation/networking/device_drivers/netronome/nfp.rst
index 6c08ac8b5147..ada611fb427c 100644
--- a/Documentation/networking/device_drivers/netronome/nfp.rst
+++ b/Documentation/networking/device_drivers/netronome/nfp.rst
@@ -131,3 +131,119 @@ abi_drv_reset
 abi_drv_load_ifc
     Defines a list of PF devices allowed to load FW on the device.
     This variable is not currently user configurable.
+
+Statistics
+==========
+
+Following device statistics are available through the ``ethtool -S`` interface:
+
+.. flat-table:: NFP device statistics
+   :header-rows: 1
+   :widths: 3 1 11
+
+   * - Name
+     - ID
+     - Meaning
+
+   * - dev_rx_discards
+     - 1
+     - Packet can be discarded on the RX path for one of the following reasons:
+
+        * The NIC is not in promisc mode, and the destination MAC address
+          doesn't match the interfaces' MAC address.
+        * The received packet is larger than the max buffer size on the host.
+          I.e. it exceeds the Layer 3 MRU.
+        * There is no freelist descriptor available on the host for the packet.
+          It is likely that the NIC couldn't cache one in time.
+        * A BPF program discarded the packet.
+        * The datapath drop action was executed.
+        * The MAC discarded the packet due to lack of ingress buffer space
+          on the NIC.
+
+   * - dev_rx_errors
+     - 2
+     - A packet can be counted (and dropped) as RX error for the following
+       reasons:
+
+       * A problem with the VEB lookup (only when SR-IOV is used).
+       * A physical layer problem that causes Ethernet errors, like FCS or
+         alignment errors. The cause is usually faulty cables or SFPs.
+
+   * - dev_rx_bytes
+     - 3
+     - Total number of bytes received.
+
+   * - dev_rx_uc_bytes
+     - 4
+     - Unicast bytes received.
+
+   * - dev_rx_mc_bytes
+     - 5
+     - Multicast bytes received.
+
+   * - dev_rx_bc_bytes
+     - 6
+     - Broadcast bytes received.
+
+   * - dev_rx_pkts
+     - 7
+     - Total number of packets received.
+
+   * - dev_rx_mc_pkts
+     - 8
+     - Multicast packets received.
+
+   * - dev_rx_bc_pkts
+     - 9
+     - Broadcast packets received.
+
+   * - dev_tx_discards
+     - 10
+     - A packet can be discarded in the TX direction if the MAC is
+       being flow controlled and the NIC runs out of TX queue space.
+
+   * - dev_tx_errors
+     - 11
+     - A packet can be counted as TX error (and dropped) for one for the
+       following reasons:
+
+       * The packet is an LSO segment, but the Layer 3 or Layer 4 offset
+         could not be determined. Therefore LSO could not continue.
+       * An invalid packet descriptor was received over PCIe.
+       * The packet Layer 3 length exceeds the device MTU.
+       * An error on the MAC/physical layer. Usually due to faulty cables or
+         SFPs.
+       * A CTM buffer could not be allocated.
+       * The packet offset was incorrect and could not be fixed by the NIC.
+
+   * - dev_tx_bytes
+     - 12
+     - Total number of bytes transmitted.
+
+   * - dev_tx_uc_bytes
+     - 13
+     - Unicast bytes transmitted.
+
+   * - dev_tx_mc_bytes
+     - 14
+     - Multicast bytes transmitted.
+
+   * - dev_tx_bc_bytes
+     - 15
+     - Broadcast bytes transmitted.
+
+   * - dev_tx_pkts
+     - 16
+     - Total number of packets transmitted.
+
+   * - dev_tx_mc_pkts
+     - 17
+     - Multicast packets transmitted.
+
+   * - dev_tx_bc_pkts
+     - 18
+     - Broadcast packets transmitted.
+
+Note that statistics unknown to the driver will be displayed as
+``dev_unknown_stat$ID``, where ``$ID`` refers to the second column
+above.
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_ctrl.c b/drivers/net/ethernet/netronome/nfp/nfp_net_ctrl.c
index d835c14b7257..45756648a85c 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_ctrl.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_ctrl.c
@@ -114,6 +114,15 @@ int nfp_net_tlv_caps_parse(struct device *dev, u8 __iomem *ctrl_mem,
 			caps->crypto_ops = readl(data);
 			caps->crypto_enable_off = data - ctrl_mem + 16;
 			break;
+		case NFP_NET_CFG_TLV_TYPE_VNIC_STATS:
+			if ((data - ctrl_mem) % 8) {
+				dev_warn(dev, "VNIC STATS TLV misaligned, ignoring offset:%u len:%u\n",
+					 offset, length);
+				break;
+			}
+			caps->vnic_stats_off = data - ctrl_mem;
+			caps->vnic_stats_cnt = length / 10;
+			break;
 		default:
 			if (!FIELD_GET(NFP_NET_CFG_TLV_HEADER_REQUIRED, hdr))
 				break;
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_ctrl.h b/drivers/net/ethernet/netronome/nfp/nfp_net_ctrl.h
index ee6b24e4eacd..c38cc36a2a70 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_ctrl.h
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_ctrl.h
@@ -479,6 +479,17 @@
  * 8 words, bitmaps of supported and enabled crypto operations.
  * First 16B (4 words) contains a bitmap of supported crypto operations,
  * and next 16B contain the enabled operations.
+ *
+ * %NFP_NET_CFG_TLV_TYPE_VNIC_STATS:
+ * Variable, per-vNIC statistics, data should be 8B aligned (FW should insert
+ * zero-length RESERVED TLV to pad).
+ * TLV data has two sections.  First is an array of statistics' IDs (2B each).
+ * Second 8B statistics themselves.  Statistics are 8B aligned, meaning there
+ * may be a padding between sections.
+ * Number of statistics can be determined as floor(tlv.length / (2 + 8)).
+ * This TLV overwrites %NFP_NET_CFG_STATS_* values (statistics in this TLV
+ * duplicate the old ones, so driver should be careful not to unnecessarily
+ * render both).
  */
 #define NFP_NET_CFG_TLV_TYPE_UNKNOWN		0
 #define NFP_NET_CFG_TLV_TYPE_RESERVED		1
@@ -490,6 +501,7 @@
 #define NFP_NET_CFG_TLV_TYPE_REPR_CAP		7
 #define NFP_NET_CFG_TLV_TYPE_MBOX_CMSG_TYPES	10
 #define NFP_NET_CFG_TLV_TYPE_CRYPTO_OPS		11 /* see crypto/fw.h */
+#define NFP_NET_CFG_TLV_TYPE_VNIC_STATS		12
 
 struct device;
 
@@ -502,6 +514,8 @@ struct device;
  * @mbox_cmsg_types:	cmsgs which can be passed through the mailbox
  * @crypto_ops:		supported crypto operations
  * @crypto_enable_off:	offset of crypto ops enable region
+ * @vnic_stats_off:	offset of vNIC stats area
+ * @vnic_stats_cnt:	number of vNIC stats
  */
 struct nfp_net_tlv_caps {
 	u32 me_freq_mhz;
@@ -511,6 +525,8 @@ struct nfp_net_tlv_caps {
 	u32 mbox_cmsg_types;
 	u32 crypto_ops;
 	unsigned int crypto_enable_off;
+	unsigned int vnic_stats_off;
+	unsigned int vnic_stats_cnt;
 };
 
 int nfp_net_tlv_caps_parse(struct device *dev, u8 __iomem *ctrl_mem,
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
index 1b840ee47339..b386a221c599 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
@@ -148,6 +148,28 @@ static const struct nfp_et_stat nfp_mac_et_stats[] = {
 	{ "tx_pause_frames_class7",	NFP_MAC_STATS_TX_PAUSE_FRAMES_CLASS7, },
 };
 
+static const char nfp_tlv_stat_names[][ETH_GSTRING_LEN] = {
+	[1]	= "dev_rx_discards",
+	[2]	= "dev_rx_errors",
+	[3]	= "dev_rx_bytes",
+	[4]	= "dev_rx_uc_bytes",
+	[5]	= "dev_rx_mc_bytes",
+	[6]	= "dev_rx_bc_bytes",
+	[7]	= "dev_rx_pkts",
+	[8]	= "dev_rx_mc_pkts",
+	[9]	= "dev_rx_bc_pkts",
+
+	[10]	= "dev_tx_discards",
+	[11]	= "dev_tx_errors",
+	[12]	= "dev_tx_bytes",
+	[13]	= "dev_tx_uc_bytes",
+	[14]	= "dev_tx_mc_bytes",
+	[15]	= "dev_tx_bc_bytes",
+	[16]	= "dev_tx_pkts",
+	[17]	= "dev_tx_mc_pkts",
+	[18]	= "dev_tx_bc_pkts",
+};
+
 #define NN_ET_GLOBAL_STATS_LEN ARRAY_SIZE(nfp_net_et_stats)
 #define NN_ET_SWITCH_STATS_LEN 9
 #define NN_RVEC_GATHER_STATS	13
@@ -560,6 +582,65 @@ nfp_vnic_get_hw_stats(u64 *data, u8 __iomem *mem, unsigned int num_vecs)
 	return data;
 }
 
+static unsigned int nfp_vnic_get_tlv_stats_count(struct nfp_net *nn)
+{
+	return nn->tlv_caps.vnic_stats_cnt + nn->max_r_vecs * 4;
+}
+
+static u8 *nfp_vnic_get_tlv_stats_strings(struct nfp_net *nn, u8 *data)
+{
+	unsigned int i, id;
+	u8 __iomem *mem;
+	u64 id_word = 0;
+
+	mem = nn->dp.ctrl_bar + nn->tlv_caps.vnic_stats_off;
+	for (i = 0; i < nn->tlv_caps.vnic_stats_cnt; i++) {
+		if (!(i % 4))
+			id_word = readq(mem + i * 2);
+
+		id = (u16)id_word;
+		id_word >>= 16;
+
+		if (id < ARRAY_SIZE(nfp_tlv_stat_names) &&
+		    nfp_tlv_stat_names[id][0]) {
+			memcpy(data, nfp_tlv_stat_names[id], ETH_GSTRING_LEN);
+			data += ETH_GSTRING_LEN;
+		} else {
+			data = nfp_pr_et(data, "dev_unknown_stat%u", id);
+		}
+	}
+
+	for (i = 0; i < nn->max_r_vecs; i++) {
+		data = nfp_pr_et(data, "rxq_%u_pkts", i);
+		data = nfp_pr_et(data, "rxq_%u_bytes", i);
+		data = nfp_pr_et(data, "txq_%u_pkts", i);
+		data = nfp_pr_et(data, "txq_%u_bytes", i);
+	}
+
+	return data;
+}
+
+static u64 *nfp_vnic_get_tlv_stats(struct nfp_net *nn, u64 *data)
+{
+	u8 __iomem *mem;
+	unsigned int i;
+
+	mem = nn->dp.ctrl_bar + nn->tlv_caps.vnic_stats_off;
+	mem += roundup(2 * nn->tlv_caps.vnic_stats_cnt, 8);
+	for (i = 0; i < nn->tlv_caps.vnic_stats_cnt; i++)
+		*data++ = readq(mem + i * 8);
+
+	mem = nn->dp.ctrl_bar;
+	for (i = 0; i < nn->max_r_vecs; i++) {
+		*data++ = readq(mem + NFP_NET_CFG_RXR_STATS(i));
+		*data++ = readq(mem + NFP_NET_CFG_RXR_STATS(i) + 8);
+		*data++ = readq(mem + NFP_NET_CFG_TXR_STATS(i));
+		*data++ = readq(mem + NFP_NET_CFG_TXR_STATS(i) + 8);
+	}
+
+	return data;
+}
+
 static unsigned int nfp_mac_get_stats_count(struct net_device *netdev)
 {
 	struct nfp_port *port;
@@ -609,8 +690,12 @@ static void nfp_net_get_strings(struct net_device *netdev,
 	switch (stringset) {
 	case ETH_SS_STATS:
 		data = nfp_vnic_get_sw_stats_strings(netdev, data);
-		data = nfp_vnic_get_hw_stats_strings(data, nn->max_r_vecs,
-						     false);
+		if (!nn->tlv_caps.vnic_stats_off)
+			data = nfp_vnic_get_hw_stats_strings(data,
+							     nn->max_r_vecs,
+							     false);
+		else
+			data = nfp_vnic_get_tlv_stats_strings(nn, data);
 		data = nfp_mac_get_stats_strings(netdev, data);
 		data = nfp_app_port_get_stats_strings(nn->port, data);
 		break;
@@ -624,7 +709,11 @@ nfp_net_get_stats(struct net_device *netdev, struct ethtool_stats *stats,
 	struct nfp_net *nn = netdev_priv(netdev);
 
 	data = nfp_vnic_get_sw_stats(netdev, data);
-	data = nfp_vnic_get_hw_stats(data, nn->dp.ctrl_bar, nn->max_r_vecs);
+	if (!nn->tlv_caps.vnic_stats_off)
+		data = nfp_vnic_get_hw_stats(data, nn->dp.ctrl_bar,
+					     nn->max_r_vecs);
+	else
+		data = nfp_vnic_get_tlv_stats(nn, data);
 	data = nfp_mac_get_stats(netdev, data);
 	data = nfp_app_port_get_stats(nn->port, data);
 }
@@ -632,13 +721,18 @@ nfp_net_get_stats(struct net_device *netdev, struct ethtool_stats *stats,
 static int nfp_net_get_sset_count(struct net_device *netdev, int sset)
 {
 	struct nfp_net *nn = netdev_priv(netdev);
+	unsigned int cnt;
 
 	switch (sset) {
 	case ETH_SS_STATS:
-		return nfp_vnic_get_sw_stats_count(netdev) +
-		       nfp_vnic_get_hw_stats_count(nn->max_r_vecs) +
-		       nfp_mac_get_stats_count(netdev) +
-		       nfp_app_port_get_stats_count(nn->port);
+		cnt = nfp_vnic_get_sw_stats_count(netdev);
+		if (!nn->tlv_caps.vnic_stats_off)
+			cnt += nfp_vnic_get_hw_stats_count(nn->max_r_vecs);
+		else
+			cnt += nfp_vnic_get_tlv_stats_count(nn);
+		cnt += nfp_mac_get_stats_count(netdev);
+		cnt += nfp_app_port_get_stats_count(nn->port);
+		return cnt;
 	default:
 		return -EOPNOTSUPP;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 6525b5ef65fdaf8a782449fb5d585195b573c2c1 Mon Sep 17 00:00:00 2001
From: Chen Wandun <chenwandun@huawei.com>
Date: Tue, 10 Dec 2019 18:24:50 +0800
Subject: enetc: remove variable 'tc_max_sized_frame' set but not used

Fixes gcc '-Wunused-but-set-variable' warning:

drivers/net/ethernet/freescale/enetc/enetc_qos.c: In function enetc_setup_tc_cbs:
drivers/net/ethernet/freescale/enetc/enetc_qos.c:195:6: warning: variable tc_max_sized_frame set but not used [-Wunused-but-set-variable]

Fixes: c431047c4efe ("enetc: add support Credit Based Shaper(CBS) for hardware offload")
Signed-off-by: Chen Wandun <chenwandun@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/freescale/enetc/enetc_qos.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/freescale/enetc/enetc_qos.c b/drivers/net/ethernet/freescale/enetc/enetc_qos.c
index 2e99438cb1bf..9190ffc9f6b2 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc_qos.c
+++ b/drivers/net/ethernet/freescale/enetc/enetc_qos.c
@@ -192,7 +192,6 @@ int enetc_setup_tc_cbs(struct net_device *ndev, void *type_data)
 	u32 hi_credit_bit, hi_credit_reg;
 	u32 max_interference_size;
 	u32 port_frame_max_size;
-	u32 tc_max_sized_frame;
 	u8 tc = cbs->queue;
 	u8 prio_top, prio_next;
 	int bw_sum = 0;
@@ -250,7 +249,7 @@ int enetc_setup_tc_cbs(struct net_device *ndev, void *type_data)
 		return -EINVAL;
 	}
 
-	tc_max_sized_frame = enetc_port_rd(&si->hw, ENETC_PTCMSDUR(tc));
+	enetc_port_rd(&si->hw, ENETC_PTCMSDUR(tc));
 
 	/* For top prio TC, the max_interfrence_size is maxSizedFrame.
 	 *
-- 
cgit v1.2.3-59-g8ed1b


From c21939998802b48e7afd0c0568193f6e4e4954f8 Mon Sep 17 00:00:00 2001
From: Shahjada Abul Husain <shahjada@chelsio.com>
Date: Tue, 10 Dec 2019 16:25:33 +0530
Subject: cxgb4: add support for high priority filters

T6 has a separate region known as high priority filter region
that allows classifying packets going through ULD path. So,
query firmware for HPFILTER resources and enable the high
priority offload filter support when it is available.

Signed-off-by: Shahjada Abul Husain <shahjada@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/chelsio/cxgb4/cxgb4.h         |  11 ++
 drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c |   3 +
 drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c  | 220 ++++++++++++++-------
 drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c    |  53 +++--
 .../net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c   |   7 +-
 .../net/ethernet/chelsio/cxgb4/cxgb4_tc_matchall.c |   2 +
 drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_u32.c  |  18 +-
 drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h     |   5 +
 drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h      |   1 +
 9 files changed, 234 insertions(+), 86 deletions(-)

diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
index a70ac2097892..eda8e1269551 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
@@ -56,6 +56,7 @@
 #include <asm/io.h>
 #include "t4_chip_type.h"
 #include "cxgb4_uld.h"
+#include "t4fw_api.h"
 
 #define CH_WARN(adap, fmt, ...) dev_warn(adap->pdev_dev, fmt, ## __VA_ARGS__)
 extern struct list_head adapter_list;
@@ -68,6 +69,16 @@ extern struct mutex uld_mutex;
 #define ETHTXQ_STOP_THRES \
 	(1 + DIV_ROUND_UP((3 * MAX_SKB_FRAGS) / 2 + (MAX_SKB_FRAGS & 1), 8))
 
+#define FW_PARAM_DEV(param) \
+	(FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) | \
+	 FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_##param))
+
+#define FW_PARAM_PFVF(param) \
+	(FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_PFVF) | \
+	 FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_PFVF_##param) |  \
+	 FW_PARAMS_PARAM_Y_V(0) | \
+	 FW_PARAMS_PARAM_Z_V(0))
+
 enum {
 	MAX_NPORTS	= 4,     /* max # of ports */
 	SERNUM_LEN	= 24,    /* Serial # length */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c
index 93868dca186a..b6a79536e143 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c
@@ -3240,6 +3240,9 @@ static int tid_info_show(struct seq_file *seq, void *v)
 		seq_printf(seq, "SFTID range: %u..%u in use: %u\n",
 			   t->sftid_base, t->sftid_base + t->nsftids - 2,
 			   t->sftids_in_use);
+	if (t->nhpftids)
+		seq_printf(seq, "HPFTID range: %u..%u\n", t->hpftid_base,
+			   t->hpftid_base + t->nhpftids - 1);
 	if (t->ntids)
 		seq_printf(seq, "HW TID usage: %u IP users, %u IPv6 users\n",
 			   t4_read_reg(adap, LE_DB_ACT_CNT_IPV4_A),
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c
index 1d39fca11810..6a5b0346dad9 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c
@@ -507,6 +507,24 @@ static int cxgb4_set_ftid(struct tid_info *t, int fidx, int family,
 	return 0;
 }
 
+static int cxgb4_set_hpftid(struct tid_info *t, int fidx, int family)
+{
+	spin_lock_bh(&t->ftid_lock);
+
+	if (test_bit(fidx, t->hpftid_bmap)) {
+		spin_unlock_bh(&t->ftid_lock);
+		return -EBUSY;
+	}
+
+	if (family == PF_INET)
+		__set_bit(fidx, t->hpftid_bmap);
+	else
+		bitmap_allocate_region(t->hpftid_bmap, fidx, 1);
+
+	spin_unlock_bh(&t->ftid_lock);
+	return 0;
+}
+
 static void cxgb4_clear_ftid(struct tid_info *t, int fidx, int family,
 			     unsigned int chip_ver)
 {
@@ -522,33 +540,58 @@ static void cxgb4_clear_ftid(struct tid_info *t, int fidx, int family,
 	spin_unlock_bh(&t->ftid_lock);
 }
 
+static void cxgb4_clear_hpftid(struct tid_info *t, int fidx, int family)
+{
+	spin_lock_bh(&t->ftid_lock);
+
+	if (family == PF_INET)
+		__clear_bit(fidx, t->hpftid_bmap);
+	else
+		bitmap_release_region(t->hpftid_bmap, fidx, 1);
+
+	spin_unlock_bh(&t->ftid_lock);
+}
+
 bool cxgb4_filter_prio_in_range(struct net_device *dev, u32 idx, u32 prio)
 {
+	struct filter_entry *prev_fe, *next_fe, *tab;
 	struct adapter *adap = netdev2adap(dev);
-	struct filter_entry *prev_fe, *next_fe;
+	u32 prev_ftid, next_ftid, max_tid;
 	struct tid_info *t = &adap->tids;
-	u32 prev_ftid, next_ftid;
+	unsigned long *bmap;
 	bool valid = true;
 
+	if (idx < t->nhpftids) {
+		bmap = t->hpftid_bmap;
+		tab = t->hpftid_tab;
+		max_tid = t->nhpftids;
+	} else {
+		idx -= t->nhpftids;
+		bmap = t->ftid_bmap;
+		tab = t->ftid_tab;
+		max_tid = t->nftids;
+	}
+
 	/* Only insert the rule if both of the following conditions
 	 * are met:
 	 * 1. The immediate previous rule has priority <= @prio.
 	 * 2. The immediate next rule has priority >= @prio.
 	 */
 	spin_lock_bh(&t->ftid_lock);
+
 	/* Don't insert if there's a rule already present at @idx. */
-	if (test_bit(idx, t->ftid_bmap)) {
+	if (test_bit(idx, bmap)) {
 		valid = false;
 		goto out_unlock;
 	}
 
-	next_ftid = find_next_bit(t->ftid_bmap, t->nftids, idx);
-	if (next_ftid >= t->nftids)
+	next_ftid = find_next_bit(bmap, max_tid, idx);
+	if (next_ftid >= max_tid)
 		next_ftid = idx;
 
-	next_fe = &adap->tids.ftid_tab[next_ftid];
+	next_fe = &tab[next_ftid];
 
-	prev_ftid = find_last_bit(t->ftid_bmap, idx);
+	prev_ftid = find_last_bit(bmap, idx);
 	if (prev_ftid >= idx)
 		prev_ftid = idx;
 
@@ -558,13 +601,13 @@ bool cxgb4_filter_prio_in_range(struct net_device *dev, u32 idx, u32 prio)
 	 * accordingly.
 	 */
 	if (CHELSIO_CHIP_VERSION(adap->params.chip) < CHELSIO_T6) {
-		prev_fe = &adap->tids.ftid_tab[prev_ftid & ~0x3];
+		prev_fe = &tab[prev_ftid & ~0x3];
 		if (!prev_fe->fs.type)
-			prev_fe = &adap->tids.ftid_tab[prev_ftid];
+			prev_fe = &tab[prev_ftid];
 	} else {
-		prev_fe = &adap->tids.ftid_tab[prev_ftid & ~0x1];
+		prev_fe = &tab[prev_ftid & ~0x1];
 		if (!prev_fe->fs.type)
-			prev_fe = &adap->tids.ftid_tab[prev_ftid];
+			prev_fe = &tab[prev_ftid];
 	}
 
 	if ((prev_fe->valid && prio < prev_fe->fs.tc_prio) ||
@@ -579,11 +622,16 @@ out_unlock:
 /* Delete the filter at a specified index. */
 static int del_filter_wr(struct adapter *adapter, int fidx)
 {
-	struct filter_entry *f = &adapter->tids.ftid_tab[fidx];
 	struct fw_filter_wr *fwr;
+	struct filter_entry *f;
 	struct sk_buff *skb;
 	unsigned int len;
 
+	if (fidx < adapter->tids.nhpftids)
+		f = &adapter->tids.hpftid_tab[fidx];
+	else
+		f = &adapter->tids.ftid_tab[fidx - adapter->tids.nhpftids];
+
 	len = sizeof(*fwr);
 
 	skb = alloc_skb(len, GFP_KERNEL);
@@ -609,10 +657,15 @@ static int del_filter_wr(struct adapter *adapter, int fidx)
  */
 int set_filter_wr(struct adapter *adapter, int fidx)
 {
-	struct filter_entry *f = &adapter->tids.ftid_tab[fidx];
 	struct fw_filter2_wr *fwr;
+	struct filter_entry *f;
 	struct sk_buff *skb;
 
+	if (fidx < adapter->tids.nhpftids)
+		f = &adapter->tids.hpftid_tab[fidx];
+	else
+		f = &adapter->tids.ftid_tab[fidx - adapter->tids.nhpftids];
+
 	skb = alloc_skb(sizeof(*fwr), GFP_KERNEL);
 	if (!skb)
 		return -ENOMEM;
@@ -811,12 +864,22 @@ void clear_all_filters(struct adapter *adapter)
 	struct net_device *dev = adapter->port[0];
 	unsigned int i;
 
+	if (adapter->tids.hpftid_tab) {
+		struct filter_entry *f = &adapter->tids.hpftid_tab[0];
+
+		for (i = 0; i < adapter->tids.nhpftids; i++, f++)
+			if (f->valid || f->pending)
+				cxgb4_del_filter(dev, i, &f->fs);
+	}
+
 	if (adapter->tids.ftid_tab) {
 		struct filter_entry *f = &adapter->tids.ftid_tab[0];
 		unsigned int max_ftid = adapter->tids.nftids +
-					adapter->tids.nsftids;
+					adapter->tids.nsftids +
+					adapter->tids.nhpftids;
+
 		/* Clear all TCAM filters */
-		for (i = 0; i < max_ftid; i++, f++)
+		for (i = adapter->tids.nhpftids; i < max_ftid; i++, f++)
 			if (f->valid || f->pending)
 				cxgb4_del_filter(dev, i, &f->fs);
 	}
@@ -1319,17 +1382,17 @@ out_err:
  * filter specification in order to facilitate signaling completion of the
  * operation.
  */
-int __cxgb4_set_filter(struct net_device *dev, int filter_id,
+int __cxgb4_set_filter(struct net_device *dev, int ftid,
 		       struct ch_filter_specification *fs,
 		       struct filter_ctx *ctx)
 {
 	struct adapter *adapter = netdev2adap(dev);
-	unsigned int chip_ver = CHELSIO_CHIP_VERSION(adapter->params.chip);
-	unsigned int max_fidx, fidx;
-	struct filter_entry *f;
+	unsigned int max_fidx, fidx, chip_ver;
+	int iq, ret, filter_id = ftid;
+	struct filter_entry *f, *tab;
 	u32 iconf;
-	int iq, ret;
 
+	chip_ver = CHELSIO_CHIP_VERSION(adapter->params.chip);
 	if (fs->hash) {
 		if (is_hashfilter(adapter))
 			return cxgb4_set_hash_filter(dev, fs, ctx);
@@ -1338,7 +1401,7 @@ int __cxgb4_set_filter(struct net_device *dev, int filter_id,
 		return -EINVAL;
 	}
 
-	max_fidx = adapter->tids.nftids;
+	max_fidx = adapter->tids.nftids + adapter->tids.nhpftids;
 	if (filter_id != (max_fidx + adapter->tids.nsftids - 1) &&
 	    filter_id >= max_fidx)
 		return -E2BIG;
@@ -1353,6 +1416,13 @@ int __cxgb4_set_filter(struct net_device *dev, int filter_id,
 	if (iq < 0)
 		return iq;
 
+	if (fs->prio) {
+		tab = &adapter->tids.hpftid_tab[0];
+	} else {
+		tab = &adapter->tids.ftid_tab[0];
+		filter_id = ftid - adapter->tids.nhpftids;
+	}
+
 	/* IPv6 filters occupy four slots and must be aligned on
 	 * four-slot boundaries.  IPv4 filters only occupy a single
 	 * slot and have no alignment requirements but writing a new
@@ -1373,9 +1443,8 @@ int __cxgb4_set_filter(struct net_device *dev, int filter_id,
 		else
 			fidx = filter_id & ~0x1;
 
-		if (fidx != filter_id &&
-		    adapter->tids.ftid_tab[fidx].fs.type) {
-			f = &adapter->tids.ftid_tab[fidx];
+		if (fidx != filter_id && tab[fidx].fs.type) {
+			f = &tab[fidx];
 			if (f->valid) {
 				dev_err(adapter->pdev_dev,
 					"Invalid location. IPv6 requires 4 slots and is occupying slots %u to %u\n",
@@ -1399,7 +1468,7 @@ int __cxgb4_set_filter(struct net_device *dev, int filter_id,
 			 */
 			for (fidx = filter_id + 1; fidx < filter_id + 4;
 			     fidx++) {
-				f = &adapter->tids.ftid_tab[fidx];
+				f = &tab[fidx];
 				if (f->valid) {
 					dev_err(adapter->pdev_dev,
 						"Invalid location.  IPv6 requires 4 slots and an IPv4 filter exists at %u\n",
@@ -1415,7 +1484,7 @@ int __cxgb4_set_filter(struct net_device *dev, int filter_id,
 				return -EINVAL;
 			/* Check overlapping IPv4 filter slot */
 			fidx = filter_id + 1;
-			f = &adapter->tids.ftid_tab[fidx];
+			f = &tab[fidx];
 			if (f->valid) {
 				pr_err("%s: IPv6 filter requires 2 indices. IPv4 filter already present at %d. Please remove IPv4 filter first.\n",
 				       __func__, fidx);
@@ -1427,36 +1496,35 @@ int __cxgb4_set_filter(struct net_device *dev, int filter_id,
 	/* Check to make sure that provided filter index is not
 	 * already in use by someone else
 	 */
-	f = &adapter->tids.ftid_tab[filter_id];
+	f = &tab[filter_id];
 	if (f->valid)
 		return -EBUSY;
 
-	fidx = filter_id + adapter->tids.ftid_base;
-	ret = cxgb4_set_ftid(&adapter->tids, filter_id,
-			     fs->type ? PF_INET6 : PF_INET,
-			     chip_ver);
+	if (fs->prio) {
+		fidx = filter_id + adapter->tids.hpftid_base;
+		ret = cxgb4_set_hpftid(&adapter->tids, filter_id,
+				       fs->type ? PF_INET6 : PF_INET);
+	} else {
+		fidx = filter_id + adapter->tids.ftid_base;
+		ret = cxgb4_set_ftid(&adapter->tids, filter_id,
+				     fs->type ? PF_INET6 : PF_INET,
+				     chip_ver);
+	}
+
 	if (ret)
 		return ret;
 
 	/* Check t  make sure the filter requested is writable ... */
 	ret = writable_filter(f);
-	if (ret) {
-		/* Clear the bits we have set above */
-		cxgb4_clear_ftid(&adapter->tids, filter_id,
-				 fs->type ? PF_INET6 : PF_INET,
-				 chip_ver);
-		return ret;
-	}
+	if (ret)
+		goto free_tid;
 
 	if (is_t6(adapter->params.chip) && fs->type &&
 	    ipv6_addr_type((const struct in6_addr *)fs->val.lip) !=
 	    IPV6_ADDR_ANY) {
 		ret = cxgb4_clip_get(dev, (const u32 *)&fs->val.lip, 1);
-		if (ret) {
-			cxgb4_clear_ftid(&adapter->tids, filter_id, PF_INET6,
-					 chip_ver);
-			return ret;
-		}
+		if (ret)
+			goto free_tid;
 	}
 
 	/* Convert the filter specification into our internal format.
@@ -1487,7 +1555,7 @@ int __cxgb4_set_filter(struct net_device *dev, int filter_id,
 						      f->fs.mask.vni,
 						      0, 1, 1);
 			if (ret < 0)
-				goto free_clip;
+				goto free_tid;
 
 			f->fs.val.ovlan = ret;
 			f->fs.mask.ovlan = 0x1ff;
@@ -1501,21 +1569,22 @@ int __cxgb4_set_filter(struct net_device *dev, int filter_id,
 	 */
 	f->ctx = ctx;
 	f->tid = fidx; /* Save the actual tid */
-	ret = set_filter_wr(adapter, filter_id);
-	if (ret) {
+	ret = set_filter_wr(adapter, ftid);
+	if (ret)
+		goto free_tid;
+
+	return ret;
+
+free_tid:
+	if (f->fs.prio)
+		cxgb4_clear_hpftid(&adapter->tids, filter_id,
+				   fs->type ? PF_INET6 : PF_INET);
+	else
 		cxgb4_clear_ftid(&adapter->tids, filter_id,
 				 fs->type ? PF_INET6 : PF_INET,
 				 chip_ver);
-		clear_filter(adapter, f);
-	}
 
-	return ret;
-
-free_clip:
-	if (is_t6(adapter->params.chip) && f->fs.type)
-		cxgb4_clip_release(f->dev, (const u32 *)&f->fs.val.lip, 1);
-	cxgb4_clear_ftid(&adapter->tids, filter_id,
-			 fs->type ? PF_INET6 : PF_INET, chip_ver);
+	clear_filter(adapter, f);
 	return ret;
 }
 
@@ -1590,11 +1659,11 @@ int __cxgb4_del_filter(struct net_device *dev, int filter_id,
 		       struct filter_ctx *ctx)
 {
 	struct adapter *adapter = netdev2adap(dev);
-	unsigned int chip_ver = CHELSIO_CHIP_VERSION(adapter->params.chip);
+	unsigned int max_fidx, chip_ver;
 	struct filter_entry *f;
-	unsigned int max_fidx;
 	int ret;
 
+	chip_ver = CHELSIO_CHIP_VERSION(adapter->params.chip);
 	if (fs && fs->hash) {
 		if (is_hashfilter(adapter))
 			return cxgb4_del_hash_filter(dev, filter_id, ctx);
@@ -1603,21 +1672,31 @@ int __cxgb4_del_filter(struct net_device *dev, int filter_id,
 		return -EINVAL;
 	}
 
-	max_fidx = adapter->tids.nftids;
+	max_fidx = adapter->tids.nftids + adapter->tids.nhpftids;
 	if (filter_id != (max_fidx + adapter->tids.nsftids - 1) &&
 	    filter_id >= max_fidx)
 		return -E2BIG;
 
-	f = &adapter->tids.ftid_tab[filter_id];
+	if (filter_id < adapter->tids.nhpftids)
+		f = &adapter->tids.hpftid_tab[filter_id];
+	else
+		f = &adapter->tids.ftid_tab[filter_id - adapter->tids.nhpftids];
+
 	ret = writable_filter(f);
 	if (ret)
 		return ret;
 
 	if (f->valid) {
 		f->ctx = ctx;
-		cxgb4_clear_ftid(&adapter->tids, filter_id,
-				 f->fs.type ? PF_INET6 : PF_INET,
-				 chip_ver);
+		if (f->fs.prio)
+			cxgb4_clear_hpftid(&adapter->tids,
+					   f->tid - adapter->tids.hpftid_base,
+					   f->fs.type ? PF_INET6 : PF_INET);
+		else
+			cxgb4_clear_ftid(&adapter->tids,
+					 f->tid - adapter->tids.ftid_base,
+					 f->fs.type ? PF_INET6 : PF_INET,
+					 chip_ver);
 		return del_filter_wr(adapter, filter_id);
 	}
 
@@ -1842,11 +1921,18 @@ void filter_rpl(struct adapter *adap, const struct cpl_set_tcb_rpl *rpl)
 	max_fidx = adap->tids.nftids + adap->tids.nsftids;
 	/* Get the corresponding filter entry for this tid */
 	if (adap->tids.ftid_tab) {
-		/* Check this in normal filter region */
-		idx = tid - adap->tids.ftid_base;
-		if (idx >= max_fidx)
-			return;
-		f = &adap->tids.ftid_tab[idx];
+		idx = tid - adap->tids.hpftid_base;
+		if (idx < adap->tids.nhpftids) {
+			f = &adap->tids.hpftid_tab[idx];
+		} else {
+			/* Check this in normal filter region */
+			idx = tid - adap->tids.ftid_base;
+			if (idx >= max_fidx)
+				return;
+			f = &adap->tids.ftid_tab[idx];
+			idx += adap->tids.nhpftids;
+		}
+
 		if (f->tid != tid)
 			return;
 	}
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
index 12ff69b3ba91..be750f2de23c 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
@@ -804,6 +804,26 @@ static int setup_ppod_edram(struct adapter *adap)
 	return 0;
 }
 
+static void adap_config_hpfilter(struct adapter *adapter)
+{
+	u32 param, val = 0;
+	int ret;
+
+	/* Enable HP filter region. Older fw will fail this request and
+	 * it is fine.
+	 */
+	param = FW_PARAM_DEV(HPFILTER_REGION_SUPPORT);
+	ret = t4_set_params(adapter, adapter->mbox, adapter->pf, 0,
+			    1, &param, &val);
+
+	/* An error means FW doesn't know about HP filter support,
+	 * it's not a problem, don't return an error.
+	 */
+	if (ret < 0)
+		dev_err(adapter->pdev_dev,
+			"HP filter region isn't supported by FW\n");
+}
+
 /**
  *	cxgb4_write_rss - write the RSS table for a given port
  *	@pi: the port
@@ -1518,6 +1538,7 @@ static int tid_init(struct tid_info *t)
 	struct adapter *adap = container_of(t, struct adapter, tids);
 	unsigned int max_ftids = t->nftids + t->nsftids;
 	unsigned int natids = t->natids;
+	unsigned int hpftid_bmap_size;
 	unsigned int eotid_bmap_size;
 	unsigned int stid_bmap_size;
 	unsigned int ftid_bmap_size;
@@ -1525,12 +1546,15 @@ static int tid_init(struct tid_info *t)
 
 	stid_bmap_size = BITS_TO_LONGS(t->nstids + t->nsftids);
 	ftid_bmap_size = BITS_TO_LONGS(t->nftids);
+	hpftid_bmap_size = BITS_TO_LONGS(t->nhpftids);
 	eotid_bmap_size = BITS_TO_LONGS(t->neotids);
 	size = t->ntids * sizeof(*t->tid_tab) +
 	       natids * sizeof(*t->atid_tab) +
 	       t->nstids * sizeof(*t->stid_tab) +
 	       t->nsftids * sizeof(*t->stid_tab) +
 	       stid_bmap_size * sizeof(long) +
+	       t->nhpftids * sizeof(*t->hpftid_tab) +
+	       hpftid_bmap_size * sizeof(long) +
 	       max_ftids * sizeof(*t->ftid_tab) +
 	       ftid_bmap_size * sizeof(long) +
 	       t->neotids * sizeof(*t->eotid_tab) +
@@ -1543,7 +1567,9 @@ static int tid_init(struct tid_info *t)
 	t->atid_tab = (union aopen_entry *)&t->tid_tab[t->ntids];
 	t->stid_tab = (struct serv_entry *)&t->atid_tab[natids];
 	t->stid_bmap = (unsigned long *)&t->stid_tab[t->nstids + t->nsftids];
-	t->ftid_tab = (struct filter_entry *)&t->stid_bmap[stid_bmap_size];
+	t->hpftid_tab = (struct filter_entry *)&t->stid_bmap[stid_bmap_size];
+	t->hpftid_bmap = (unsigned long *)&t->hpftid_tab[t->nhpftids];
+	t->ftid_tab = (struct filter_entry *)&t->hpftid_bmap[hpftid_bmap_size];
 	t->ftid_bmap = (unsigned long *)&t->ftid_tab[max_ftids];
 	t->eotid_tab = (struct eotid_entry *)&t->ftid_bmap[ftid_bmap_size];
 	t->eotid_bmap = (unsigned long *)&t->eotid_tab[t->neotids];
@@ -1578,6 +1604,8 @@ static int tid_init(struct tid_info *t)
 			bitmap_zero(t->eotid_bmap, t->neotids);
 	}
 
+	if (t->nhpftids)
+		bitmap_zero(t->hpftid_bmap, t->nhpftids);
 	bitmap_zero(t->ftid_bmap, t->nftids);
 	return 0;
 }
@@ -4351,6 +4379,7 @@ static int adap_init0_config(struct adapter *adapter, int reset)
 			"HMA configuration failed with error %d\n", ret);
 
 	if (is_t6(adapter->params.chip)) {
+		adap_config_hpfilter(adapter);
 		ret = setup_ppod_edram(adapter);
 		if (!ret)
 			dev_info(adapter->pdev_dev, "Successfully enabled "
@@ -4660,16 +4689,6 @@ static int adap_init0(struct adapter *adap, int vpd_skip)
 	/*
 	 * Grab some of our basic fundamental operating parameters.
 	 */
-#define FW_PARAM_DEV(param) \
-	(FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) | \
-	FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_##param))
-
-#define FW_PARAM_PFVF(param) \
-	FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_PFVF) | \
-	FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_PFVF_##param)|  \
-	FW_PARAMS_PARAM_Y_V(0) | \
-	FW_PARAMS_PARAM_Z_V(0)
-
 	params[0] = FW_PARAM_PFVF(EQ_START);
 	params[1] = FW_PARAM_PFVF(L2T_START);
 	params[2] = FW_PARAM_PFVF(L2T_END);
@@ -4687,6 +4706,16 @@ static int adap_init0(struct adapter *adap, int vpd_skip)
 	adap->sge.ingr_start = val[5];
 
 	if (CHELSIO_CHIP_VERSION(adap->params.chip) > CHELSIO_T5) {
+		params[0] = FW_PARAM_PFVF(HPFILTER_START);
+		params[1] = FW_PARAM_PFVF(HPFILTER_END);
+		ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 2,
+				      params, val);
+		if (ret < 0)
+			goto bye;
+
+		adap->tids.hpftid_base = val[0];
+		adap->tids.nhpftids = val[1] - val[0] + 1;
+
 		/* Read the raw mps entries. In T6, the last 2 tcam entries
 		 * are reserved for raw mac addresses (rawf = 2, one per port).
 		 */
@@ -5050,8 +5079,6 @@ static int adap_init0(struct adapter *adap, int vpd_skip)
 		}
 		adap->params.crypto = ntohs(caps_cmd.cryptocaps);
 	}
-#undef FW_PARAM_PFVF
-#undef FW_PARAM_DEV
 
 	/* The MTU/MSS Table is initialized by now, so load their values.  If
 	 * we're initializing the adapter, then we'll make any modifications
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c
index 0fa80bef575d..706b71e1b6c1 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c
@@ -672,10 +672,13 @@ int cxgb4_tc_flower_replace(struct net_device *dev,
 		 * 0 to driver. However, the hardware TCAM index
 		 * starts from 0. Hence, the -1 here.
 		 */
-		if (cls->common.prio <= adap->tids.nftids)
+		if (cls->common.prio <= adap->tids.nftids) {
 			fidx = cls->common.prio - 1;
-		else
+			if (fidx < adap->tids.nhpftids)
+				fs->prio = 1;
+		} else {
 			fidx = cxgb4_get_free_ftid(dev, inet_family);
+		}
 
 		/* Only insert FLOWER rule if its priority doesn't
 		 * conflict with existing rules in the LETCAM.
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_matchall.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_matchall.c
index 102b370fbd3e..53e8db8b2546 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_matchall.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_matchall.c
@@ -156,6 +156,8 @@ static int cxgb4_matchall_alloc_filter(struct net_device *dev,
 	fs = &tc_port_matchall->ingress.fs;
 	memset(fs, 0, sizeof(*fs));
 
+	if (fidx < adap->tids.nhpftids)
+		fs->prio = 1;
 	fs->tc_prio = cls->common.prio;
 	fs->tc_cookie = cls->cookie;
 	fs->hitcnts = 1;
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_u32.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_u32.c
index 133f8623ba86..269b8d9e25e0 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_u32.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_u32.c
@@ -176,7 +176,7 @@ int cxgb4_config_knode(struct net_device *dev, struct tc_cls_u32_offload *cls)
 	/* Only insert U32 rule if its priority doesn't conflict with
 	 * existing rules in the LETCAM.
 	 */
-	if (filter_id >= adapter->tids.nftids ||
+	if (filter_id >= adapter->tids.nftids + adapter->tids.nhpftids ||
 	    !cxgb4_filter_prio_in_range(dev, filter_id, cls->common.prio)) {
 		NL_SET_ERR_MSG_MOD(extack,
 				   "No free LETCAM index available");
@@ -199,6 +199,8 @@ int cxgb4_config_knode(struct net_device *dev, struct tc_cls_u32_offload *cls)
 
 	memset(&fs, 0, sizeof(fs));
 
+	if (filter_id < adapter->tids.nhpftids)
+		fs.prio = 1;
 	fs.tc_prio = cls->common.prio;
 	fs.tc_cookie = cls->knode.handle;
 
@@ -355,6 +357,7 @@ int cxgb4_delete_knode(struct net_device *dev, struct tc_cls_u32_offload *cls)
 	unsigned int filter_id, max_tids, i, j;
 	struct cxgb4_link *link = NULL;
 	struct cxgb4_tc_u32_table *t;
+	struct filter_entry *f;
 	u32 handle, uhtid;
 	int ret;
 
@@ -363,8 +366,15 @@ int cxgb4_delete_knode(struct net_device *dev, struct tc_cls_u32_offload *cls)
 
 	/* Fetch the location to delete the filter. */
 	filter_id = TC_U32_NODE(cls->knode.handle) - 1;
-	if (filter_id >= adapter->tids.nftids ||
-	    cls->knode.handle != adapter->tids.ftid_tab[filter_id].fs.tc_cookie)
+	if (filter_id >= adapter->tids.nftids + adapter->tids.nhpftids)
+		return -ERANGE;
+
+	if (filter_id < adapter->tids.nhpftids)
+		f = &adapter->tids.hpftid_tab[filter_id];
+	else
+		f = &adapter->tids.ftid_tab[filter_id - adapter->tids.nhpftids];
+
+	if (cls->knode.handle != f->fs.tc_cookie)
 		return -ERANGE;
 
 	t = adapter->tc_u32;
@@ -445,7 +455,7 @@ void cxgb4_cleanup_tc_u32(struct adapter *adap)
 
 struct cxgb4_tc_u32_table *cxgb4_init_tc_u32(struct adapter *adap)
 {
-	unsigned int max_tids = adap->tids.nftids;
+	unsigned int max_tids = adap->tids.nftids + adap->tids.nhpftids;
 	struct cxgb4_tc_u32_table *t;
 	unsigned int i;
 
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h
index 861b25d28ed6..e69de9a296ae 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h
@@ -111,6 +111,11 @@ struct tid_info {
 	unsigned int natids;
 	unsigned int atid_base;
 
+	struct filter_entry *hpftid_tab;
+	unsigned long *hpftid_bmap;
+	unsigned int nhpftids;
+	unsigned int hpftid_base;
+
 	struct filter_entry *ftid_tab;
 	unsigned long *ftid_bmap;
 	unsigned int nftids;
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h b/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h
index ac4fb43bdec6..accad1101ad1 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h
@@ -1321,6 +1321,7 @@ enum fw_params_param_dev {
 	FW_PARAMS_PARAM_DEV_RDMA_WRITE_WITH_IMM = 0x21,
 	FW_PARAMS_PARAM_DEV_PPOD_EDRAM  = 0x23,
 	FW_PARAMS_PARAM_DEV_RI_WRITE_CMPL_WR    = 0x24,
+	FW_PARAMS_PARAM_DEV_HPFILTER_REGION_SUPPORT = 0x26,
 	FW_PARAMS_PARAM_DEV_OPAQUE_VIID_SMT_EXTN = 0x27,
 	FW_PARAMS_PARAM_DEV_HASHFILTER_WITH_OFLD = 0x28,
 	FW_PARAMS_PARAM_DEV_DBQ_TIMER	= 0x29,
-- 
cgit v1.2.3-59-g8ed1b


From c453312857ba41129db3558f5428405bbbb8f1a4 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Mon, 9 Dec 2019 11:17:30 +0000
Subject: ARM: net: bpf: Improve prologue code sequence

Improve the prologue code sequence to be able to take advantage of
64-bit stores, changing the code from:

  push    {r4, r5, r6, r7, r8, r9, fp, lr}
  mov     fp, sp
  sub     ip, sp, #80     ; 0x50
  sub     sp, sp, #600    ; 0x258
  str     ip, [fp, #-100] ; 0xffffff9c
  mov     r6, #0
  str     r6, [fp, #-96]  ; 0xffffffa0
  mov     r4, #0
  mov     r3, r4
  mov     r2, r0
  str     r4, [fp, #-104] ; 0xffffff98
  str     r4, [fp, #-108] ; 0xffffff94

to the tighter:

  push    {r4, r5, r6, r7, r8, r9, fp, lr}
  mov     fp, sp
  mov     r3, #0
  sub     r2, sp, #80     ; 0x50
  sub     sp, sp, #600    ; 0x258
  strd    r2, [fp, #-100] ; 0xffffff9c
  mov     r2, #0
  strd    r2, [fp, #-108] ; 0xffffff94
  mov     r2, r0

resulting in a saving of three instructions.

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/E1ieH2g-0004ih-Rb@rmk-PC.armlinux.org.uk
---
 arch/arm/net/bpf_jit_32.c | 30 +++++++++++++-----------------
 1 file changed, 13 insertions(+), 17 deletions(-)

diff --git a/arch/arm/net/bpf_jit_32.c b/arch/arm/net/bpf_jit_32.c
index 97dc386e3cb8..cc29869d12a3 100644
--- a/arch/arm/net/bpf_jit_32.c
+++ b/arch/arm/net/bpf_jit_32.c
@@ -1260,12 +1260,9 @@ static inline void emit_push_r64(const s8 src[], struct jit_ctx *ctx)
 
 static void build_prologue(struct jit_ctx *ctx)
 {
-	const s8 r0 = bpf2a32[BPF_REG_0][1];
-	const s8 r2 = bpf2a32[BPF_REG_1][1];
-	const s8 r3 = bpf2a32[BPF_REG_1][0];
-	const s8 r4 = bpf2a32[BPF_REG_6][1];
-	const s8 fplo = bpf2a32[BPF_REG_FP][1];
-	const s8 fphi = bpf2a32[BPF_REG_FP][0];
+	const s8 arm_r0 = bpf2a32[BPF_REG_0][1];
+	const s8 *bpf_r1 = bpf2a32[BPF_REG_1];
+	const s8 *bpf_fp = bpf2a32[BPF_REG_FP];
 	const s8 *tcc = bpf2a32[TCALL_CNT];
 
 	/* Save callee saved registers. */
@@ -1278,8 +1275,10 @@ static void build_prologue(struct jit_ctx *ctx)
 	emit(ARM_PUSH(CALLEE_PUSH_MASK), ctx);
 	emit(ARM_MOV_R(ARM_FP, ARM_SP), ctx);
 #endif
-	/* Save frame pointer for later */
-	emit(ARM_SUB_I(ARM_IP, ARM_SP, SCRATCH_SIZE), ctx);
+	/* mov r3, #0 */
+	/* sub r2, sp, #SCRATCH_SIZE */
+	emit(ARM_MOV_I(bpf_r1[0], 0), ctx);
+	emit(ARM_SUB_I(bpf_r1[1], ARM_SP, SCRATCH_SIZE), ctx);
 
 	ctx->stack_size = imm8m(STACK_SIZE);
 
@@ -1287,18 +1286,15 @@ static void build_prologue(struct jit_ctx *ctx)
 	emit(ARM_SUB_I(ARM_SP, ARM_SP, ctx->stack_size), ctx);
 
 	/* Set up BPF prog stack base register */
-	emit_a32_mov_r(fplo, ARM_IP, ctx);
-	emit_a32_mov_i(fphi, 0, ctx);
+	emit_a32_mov_r64(true, bpf_fp, bpf_r1, ctx);
 
-	/* mov r4, 0 */
-	emit(ARM_MOV_I(r4, 0), ctx);
+	/* Initialize Tail Count */
+	emit(ARM_MOV_I(bpf_r1[1], 0), ctx);
+	emit_a32_mov_r64(true, tcc, bpf_r1, ctx);
 
 	/* Move BPF_CTX to BPF_R1 */
-	emit(ARM_MOV_R(r3, r4), ctx);
-	emit(ARM_MOV_R(r2, r0), ctx);
-	/* Initialize Tail Count */
-	emit(ARM_STR_I(r4, ARM_FP, EBPF_SCRATCH_TO_ARM_FP(tcc[0])), ctx);
-	emit(ARM_STR_I(r4, ARM_FP, EBPF_SCRATCH_TO_ARM_FP(tcc[1])), ctx);
+	emit(ARM_MOV_R(bpf_r1[1], arm_r0), ctx);
+
 	/* end of prologue */
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 09c4708d3cf42b2b2211b1aee82e7baf446a24e8 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Mon, 9 Dec 2019 14:40:22 -0800
Subject: libbpf: Bump libpf current version to v0.0.7

New development cycles starts, bump to v0.0.7 proactively.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20191209224022.3544519-1-andriin@fb.com
---
 tools/lib/bpf/libbpf.map | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
index 8ddc2c40e482..495df575f87f 100644
--- a/tools/lib/bpf/libbpf.map
+++ b/tools/lib/bpf/libbpf.map
@@ -208,3 +208,6 @@ LIBBPF_0.0.6 {
 		btf__find_by_name_kind;
 		libbpf_find_vmlinux_btf_id;
 } LIBBPF_0.0.5;
+
+LIBBPF_0.0.7 {
+} LIBBPF_0.0.6;
-- 
cgit v1.2.3-59-g8ed1b


From b590cb5f802dc20c72f507f7fbe6737222d0afba Mon Sep 17 00:00:00 2001
From: Stanislav Fomichev <sdf@google.com>
Date: Tue, 10 Dec 2019 11:19:33 -0800
Subject: bpf: Switch to offsetofend in BPF_PROG_TEST_RUN

Switch existing pattern of "offsetof(..., member) + FIELD_SIZEOF(...,
member)' to "offsetofend(..., member)" which does exactly what
we need without all the copy-paste.

Suggested-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/20191210191933.105321-1-sdf@google.com
---
 net/bpf/test_run.c | 12 ++++--------
 1 file changed, 4 insertions(+), 8 deletions(-)

diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c
index 915c2d6f7fb9..85c8cbbada92 100644
--- a/net/bpf/test_run.c
+++ b/net/bpf/test_run.c
@@ -252,22 +252,19 @@ static int convert___skb_to_skb(struct sk_buff *skb, struct __sk_buff *__skb)
 
 	/* priority is allowed */
 
-	if (!range_is_zero(__skb, offsetof(struct __sk_buff, priority) +
-			   FIELD_SIZEOF(struct __sk_buff, priority),
+	if (!range_is_zero(__skb, offsetofend(struct __sk_buff, priority),
 			   offsetof(struct __sk_buff, cb)))
 		return -EINVAL;
 
 	/* cb is allowed */
 
-	if (!range_is_zero(__skb, offsetof(struct __sk_buff, cb) +
-			   FIELD_SIZEOF(struct __sk_buff, cb),
+	if (!range_is_zero(__skb, offsetofend(struct __sk_buff, cb),
 			   offsetof(struct __sk_buff, tstamp)))
 		return -EINVAL;
 
 	/* tstamp is allowed */
 
-	if (!range_is_zero(__skb, offsetof(struct __sk_buff, tstamp) +
-			   FIELD_SIZEOF(struct __sk_buff, tstamp),
+	if (!range_is_zero(__skb, offsetofend(struct __sk_buff, tstamp),
 			   sizeof(struct __sk_buff)))
 		return -EINVAL;
 
@@ -437,8 +434,7 @@ static int verify_user_bpf_flow_keys(struct bpf_flow_keys *ctx)
 
 	/* flags is allowed */
 
-	if (!range_is_zero(ctx, offsetof(struct bpf_flow_keys, flags) +
-			   FIELD_SIZEOF(struct bpf_flow_keys, flags),
+	if (!range_is_zero(ctx, offsetofend(struct bpf_flow_keys, flags),
 			   sizeof(struct bpf_flow_keys)))
 		return -EINVAL;
 
-- 
cgit v1.2.3-59-g8ed1b


From bae141f54be83b06652c1d47e50e4e75ed4e9c7e Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Fri, 6 Dec 2019 22:49:34 +0100
Subject: bpf: Emit audit messages upon successful prog load and unload

Allow for audit messages to be emitted upon BPF program load and
unload for having a timeline of events. The load itself is in
syscall context, so additional info about the process initiating
the BPF prog creation can be logged and later directly correlated
to the unload event.

The only info really needed from BPF side is the globally unique
prog ID where then audit user space tooling can query / dump all
info needed about the specific BPF program right upon load event
and enrich the record, thus these changes needed here can be kept
small and non-intrusive to the core.

Raw example output:

  # auditctl -D
  # auditctl -a always,exit -F arch=x86_64 -S bpf
  # ausearch --start recent -m 1334
  ...
  ----
  time->Wed Nov 27 16:04:13 2019
  type=PROCTITLE msg=audit(1574867053.120:84664): proctitle="./bpf"
  type=SYSCALL msg=audit(1574867053.120:84664): arch=c000003e syscall=321   \
    success=yes exit=3 a0=5 a1=7ffea484fbe0 a2=70 a3=0 items=0 ppid=7477    \
    pid=12698 auid=1001 uid=1001 gid=1001 euid=1001 suid=1001 fsuid=1001    \
    egid=1001 sgid=1001 fsgid=1001 tty=pts2 ses=4 comm="bpf"                \
    exe="/home/jolsa/auditd/audit-testsuite/tests/bpf/bpf"                  \
    subj=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023 key=(null)
  type=UNKNOWN[1334] msg=audit(1574867053.120:84664): prog-id=76 op=LOAD
  ----
  time->Wed Nov 27 16:04:13 2019
  type=UNKNOWN[1334] msg=audit(1574867053.120:84665): prog-id=76 op=UNLOAD
  ...

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Co-developed-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Acked-by: Paul Moore <paul@paul-moore.com>
Link: https://lore.kernel.org/bpf/20191206214934.11319-1-jolsa@kernel.org
---
 include/uapi/linux/audit.h |  1 +
 kernel/bpf/syscall.c       | 33 +++++++++++++++++++++++++++++++++
 2 files changed, 34 insertions(+)

diff --git a/include/uapi/linux/audit.h b/include/uapi/linux/audit.h
index 3ad935527177..a534d71e689a 100644
--- a/include/uapi/linux/audit.h
+++ b/include/uapi/linux/audit.h
@@ -116,6 +116,7 @@
 #define AUDIT_FANOTIFY		1331	/* Fanotify access decision */
 #define AUDIT_TIME_INJOFFSET	1332	/* Timekeeping offset injected */
 #define AUDIT_TIME_ADJNTPVAL	1333	/* NTP value adjustment */
+#define AUDIT_BPF		1334	/* BPF subsystem */
 
 #define AUDIT_AVC		1400	/* SE Linux avc denial or grant */
 #define AUDIT_SELINUX_ERR	1401	/* Internal SE Linux Errors */
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index e3461ec59570..66b90eaf99fe 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -23,6 +23,7 @@
 #include <linux/timekeeping.h>
 #include <linux/ctype.h>
 #include <linux/nospec.h>
+#include <linux/audit.h>
 #include <uapi/linux/btf.h>
 
 #define IS_FD_ARRAY(map) ((map)->map_type == BPF_MAP_TYPE_PERF_EVENT_ARRAY || \
@@ -1306,6 +1307,36 @@ static int find_prog_type(enum bpf_prog_type type, struct bpf_prog *prog)
 	return 0;
 }
 
+enum bpf_audit {
+	BPF_AUDIT_LOAD,
+	BPF_AUDIT_UNLOAD,
+	BPF_AUDIT_MAX,
+};
+
+static const char * const bpf_audit_str[BPF_AUDIT_MAX] = {
+	[BPF_AUDIT_LOAD]   = "LOAD",
+	[BPF_AUDIT_UNLOAD] = "UNLOAD",
+};
+
+static void bpf_audit_prog(const struct bpf_prog *prog, unsigned int op)
+{
+	struct audit_context *ctx = NULL;
+	struct audit_buffer *ab;
+
+	if (WARN_ON_ONCE(op >= BPF_AUDIT_MAX))
+		return;
+	if (audit_enabled == AUDIT_OFF)
+		return;
+	if (op == BPF_AUDIT_LOAD)
+		ctx = audit_context();
+	ab = audit_log_start(ctx, GFP_ATOMIC, AUDIT_BPF);
+	if (unlikely(!ab))
+		return;
+	audit_log_format(ab, "prog-id=%u op=%s",
+			 prog->aux->id, bpf_audit_str[op]);
+	audit_log_end(ab);
+}
+
 int __bpf_prog_charge(struct user_struct *user, u32 pages)
 {
 	unsigned long memlock_limit = rlimit(RLIMIT_MEMLOCK) >> PAGE_SHIFT;
@@ -1421,6 +1452,7 @@ static void __bpf_prog_put(struct bpf_prog *prog, bool do_idr_lock)
 {
 	if (atomic64_dec_and_test(&prog->aux->refcnt)) {
 		perf_event_bpf_event(prog, PERF_BPF_EVENT_PROG_UNLOAD, 0);
+		bpf_audit_prog(prog, BPF_AUDIT_UNLOAD);
 		/* bpf_prog_free_id() must be called first */
 		bpf_prog_free_id(prog, do_idr_lock);
 		__bpf_prog_put_noref(prog, true);
@@ -1830,6 +1862,7 @@ static int bpf_prog_load(union bpf_attr *attr, union bpf_attr __user *uattr)
 	 */
 	bpf_prog_kallsyms_add(prog);
 	perf_event_bpf_event(prog, PERF_BPF_EVENT_PROG_LOAD, 0);
+	bpf_audit_prog(prog, BPF_AUDIT_LOAD);
 
 	err = bpf_prog_new_fd(prog);
 	if (err < 0)
-- 
cgit v1.2.3-59-g8ed1b


From fa2de660a8d001fdce6209534b60b85b87cda84f Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Wed, 11 Dec 2019 10:55:54 +0000
Subject: net: sfp: remove incomplete 100BASE-FX and 100BASE-LX support

The 100BASE-FX and 100BASE-LX support assumes a PHY is present; this
is probably an incorrect assumption. In any case, sfp_parse_support()
will fail such a module. Let's stop pretending we support these
modules.

Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/sfp-bus.c |  4 +---
 drivers/net/phy/sfp.c     | 13 +------------
 2 files changed, 2 insertions(+), 15 deletions(-)

diff --git a/drivers/net/phy/sfp-bus.c b/drivers/net/phy/sfp-bus.c
index 5a72093ab6e7..02ab07624c89 100644
--- a/drivers/net/phy/sfp-bus.c
+++ b/drivers/net/phy/sfp-bus.c
@@ -342,9 +342,7 @@ phy_interface_t sfp_select_interface(struct sfp_bus *bus,
 	if (phylink_test(link_modes, 2500baseX_Full))
 		return PHY_INTERFACE_MODE_2500BASEX;
 
-	if (id->base.e1000_base_t ||
-	    id->base.e100_base_lx ||
-	    id->base.e100_base_fx)
+	if (id->base.e1000_base_t)
 		return PHY_INTERFACE_MODE_SGMII;
 
 	if (phylink_test(link_modes, 1000baseX_Full))
diff --git a/drivers/net/phy/sfp.c b/drivers/net/phy/sfp.c
index 27360d1840b2..ae6a52a19458 100644
--- a/drivers/net/phy/sfp.c
+++ b/drivers/net/phy/sfp.c
@@ -1489,18 +1489,7 @@ static void sfp_sm_fault(struct sfp *sfp, unsigned int next_state, bool warn)
 
 static void sfp_sm_probe_for_phy(struct sfp *sfp)
 {
-	/* Setting the serdes link mode is guesswork: there's no
-	 * field in the EEPROM which indicates what mode should
-	 * be used.
-	 *
-	 * If it's a gigabit-only fiber module, it probably does
-	 * not have a PHY, so switch to 802.3z negotiation mode.
-	 * Otherwise, switch to SGMII mode (which is required to
-	 * support non-gigabit speeds) and probe for a PHY.
-	 */
-	if (sfp->id.base.e1000_base_t ||
-	    sfp->id.base.e100_base_lx ||
-	    sfp->id.base.e100_base_fx)
+	if (sfp->id.base.e1000_base_t)
 		sfp_sm_probe_phy(sfp);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From a4516c7053b96fed98a0439a9226983b5275474b Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Wed, 11 Dec 2019 10:55:59 +0000
Subject: net: sfp: derive interface mode from ethtool link modes

We don't need the EEPROM ID to derive the phy interface mode as we can
derive it merely from the ethtool link modes.  Remove the EEPROM ID
argument to sfp_select_interface().

Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/marvell10g.c |  2 +-
 drivers/net/phy/phylink.c    |  2 +-
 drivers/net/phy/sfp-bus.c    | 11 ++++-------
 include/linux/sfp.h          |  2 --
 4 files changed, 6 insertions(+), 11 deletions(-)

diff --git a/drivers/net/phy/marvell10g.c b/drivers/net/phy/marvell10g.c
index 1bf13017d288..512f27b0b5cd 100644
--- a/drivers/net/phy/marvell10g.c
+++ b/drivers/net/phy/marvell10g.c
@@ -214,7 +214,7 @@ static int mv3310_sfp_insert(void *upstream, const struct sfp_eeprom_id *id)
 	phy_interface_t iface;
 
 	sfp_parse_support(phydev->sfp_bus, id, support);
-	iface = sfp_select_interface(phydev->sfp_bus, id, support);
+	iface = sfp_select_interface(phydev->sfp_bus, support);
 
 	if (iface != PHY_INTERFACE_MODE_10GKR) {
 		dev_err(&phydev->mdio.dev, "incompatible SFP module inserted\n");
diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index 9a616d6bc4eb..b3e4d9bf8113 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -1714,7 +1714,7 @@ static int phylink_sfp_module_insert(void *upstream,
 
 	linkmode_copy(support1, support);
 
-	iface = sfp_select_interface(pl->sfp_bus, id, config.advertising);
+	iface = sfp_select_interface(pl->sfp_bus, config.advertising);
 	if (iface == PHY_INTERFACE_MODE_NA) {
 		phylink_err(pl,
 			    "selection of interface failed, advertisement %*pb\n",
diff --git a/drivers/net/phy/sfp-bus.c b/drivers/net/phy/sfp-bus.c
index 02ab07624c89..1561962fda30 100644
--- a/drivers/net/phy/sfp-bus.c
+++ b/drivers/net/phy/sfp-bus.c
@@ -320,16 +320,12 @@ EXPORT_SYMBOL_GPL(sfp_parse_support);
 /**
  * sfp_select_interface() - Select appropriate phy_interface_t mode
  * @bus: a pointer to the &struct sfp_bus structure for the sfp module
- * @id: a pointer to the module's &struct sfp_eeprom_id
  * @link_modes: ethtool link modes mask
  *
- * Derive the phy_interface_t mode for the information found in the
- * module's identifying EEPROM and the link modes mask. There is no
- * standard or defined way to derive this information, so we decide
- * based upon the link mode mask.
+ * Derive the phy_interface_t mode for the SFP module from the link
+ * modes mask.
  */
 phy_interface_t sfp_select_interface(struct sfp_bus *bus,
-				     const struct sfp_eeprom_id *id,
 				     unsigned long *link_modes)
 {
 	if (phylink_test(link_modes, 10000baseCR_Full) ||
@@ -342,7 +338,8 @@ phy_interface_t sfp_select_interface(struct sfp_bus *bus,
 	if (phylink_test(link_modes, 2500baseX_Full))
 		return PHY_INTERFACE_MODE_2500BASEX;
 
-	if (id->base.e1000_base_t)
+	if (phylink_test(link_modes, 1000baseT_Half) ||
+	    phylink_test(link_modes, 1000baseT_Full))
 		return PHY_INTERFACE_MODE_SGMII;
 
 	if (phylink_test(link_modes, 1000baseX_Full))
diff --git a/include/linux/sfp.h b/include/linux/sfp.h
index 487fd9412d10..8d7b98c214d7 100644
--- a/include/linux/sfp.h
+++ b/include/linux/sfp.h
@@ -504,7 +504,6 @@ int sfp_parse_port(struct sfp_bus *bus, const struct sfp_eeprom_id *id,
 void sfp_parse_support(struct sfp_bus *bus, const struct sfp_eeprom_id *id,
 		       unsigned long *support);
 phy_interface_t sfp_select_interface(struct sfp_bus *bus,
-				     const struct sfp_eeprom_id *id,
 				     unsigned long *link_modes);
 
 int sfp_get_module_info(struct sfp_bus *bus, struct ethtool_modinfo *modinfo);
@@ -532,7 +531,6 @@ static inline void sfp_parse_support(struct sfp_bus *bus,
 }
 
 static inline phy_interface_t sfp_select_interface(struct sfp_bus *bus,
-						   const struct sfp_eeprom_id *id,
 						   unsigned long *link_modes)
 {
 	return PHY_INTERFACE_MODE_NA;
-- 
cgit v1.2.3-59-g8ed1b


From 0fbd26a9fb6875b98fcfff523831fec47bc5e9a2 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Wed, 11 Dec 2019 10:56:04 +0000
Subject: net: sfp: add more extended compliance codes

SFF-8024 is used to define various constants re-used in several SFF
SFP-related specifications.  Split these constants from the enum, and
rename them to indicate that they're defined by SFF-8024.

Add and use updated SFF-8024 extended compliance code definitions for
10GBASE-T, 5GBASE-T and 2.5GBASE-T modules.

Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/sfp-bus.c | 60 ++++++++++++++++++++--------------
 drivers/net/phy/sfp.c     |  4 +--
 include/linux/sfp.h       | 82 +++++++++++++++++++++++++++++++----------------
 3 files changed, 93 insertions(+), 53 deletions(-)

diff --git a/drivers/net/phy/sfp-bus.c b/drivers/net/phy/sfp-bus.c
index 1561962fda30..c6627f1e5d68 100644
--- a/drivers/net/phy/sfp-bus.c
+++ b/drivers/net/phy/sfp-bus.c
@@ -124,35 +124,35 @@ int sfp_parse_port(struct sfp_bus *bus, const struct sfp_eeprom_id *id,
 
 	/* port is the physical connector, set this from the connector field. */
 	switch (id->base.connector) {
-	case SFP_CONNECTOR_SC:
-	case SFP_CONNECTOR_FIBERJACK:
-	case SFP_CONNECTOR_LC:
-	case SFP_CONNECTOR_MT_RJ:
-	case SFP_CONNECTOR_MU:
-	case SFP_CONNECTOR_OPTICAL_PIGTAIL:
+	case SFF8024_CONNECTOR_SC:
+	case SFF8024_CONNECTOR_FIBERJACK:
+	case SFF8024_CONNECTOR_LC:
+	case SFF8024_CONNECTOR_MT_RJ:
+	case SFF8024_CONNECTOR_MU:
+	case SFF8024_CONNECTOR_OPTICAL_PIGTAIL:
+	case SFF8024_CONNECTOR_MPO_1X12:
+	case SFF8024_CONNECTOR_MPO_2X16:
 		port = PORT_FIBRE;
 		break;
 
-	case SFP_CONNECTOR_RJ45:
+	case SFF8024_CONNECTOR_RJ45:
 		port = PORT_TP;
 		break;
 
-	case SFP_CONNECTOR_COPPER_PIGTAIL:
+	case SFF8024_CONNECTOR_COPPER_PIGTAIL:
 		port = PORT_DA;
 		break;
 
-	case SFP_CONNECTOR_UNSPEC:
+	case SFF8024_CONNECTOR_UNSPEC:
 		if (id->base.e1000_base_t) {
 			port = PORT_TP;
 			break;
 		}
 		/* fallthrough */
-	case SFP_CONNECTOR_SG: /* guess */
-	case SFP_CONNECTOR_MPO_1X12:
-	case SFP_CONNECTOR_MPO_2X16:
-	case SFP_CONNECTOR_HSSDC_II:
-	case SFP_CONNECTOR_NOSEPARATE:
-	case SFP_CONNECTOR_MXC_2X16:
+	case SFF8024_CONNECTOR_SG: /* guess */
+	case SFF8024_CONNECTOR_HSSDC_II:
+	case SFF8024_CONNECTOR_NOSEPARATE:
+	case SFF8024_CONNECTOR_MXC_2X16:
 		port = PORT_OTHER;
 		break;
 	default:
@@ -261,22 +261,33 @@ void sfp_parse_support(struct sfp_bus *bus, const struct sfp_eeprom_id *id,
 	}
 
 	switch (id->base.extended_cc) {
-	case 0x00: /* Unspecified */
+	case SFF8024_ECC_UNSPEC:
 		break;
-	case 0x02: /* 100Gbase-SR4 or 25Gbase-SR */
+	case SFF8024_ECC_100GBASE_SR4_25GBASE_SR:
 		phylink_set(modes, 100000baseSR4_Full);
 		phylink_set(modes, 25000baseSR_Full);
 		break;
-	case 0x03: /* 100Gbase-LR4 or 25Gbase-LR */
-	case 0x04: /* 100Gbase-ER4 or 25Gbase-ER */
+	case SFF8024_ECC_100GBASE_LR4_25GBASE_LR:
+	case SFF8024_ECC_100GBASE_ER4_25GBASE_ER:
 		phylink_set(modes, 100000baseLR4_ER4_Full);
 		break;
-	case 0x0b: /* 100Gbase-CR4 or 25Gbase-CR CA-L */
-	case 0x0c: /* 25Gbase-CR CA-S */
-	case 0x0d: /* 25Gbase-CR CA-N */
+	case SFF8024_ECC_100GBASE_CR4:
 		phylink_set(modes, 100000baseCR4_Full);
+		/* fallthrough */
+	case SFF8024_ECC_25GBASE_CR_S:
+	case SFF8024_ECC_25GBASE_CR_N:
 		phylink_set(modes, 25000baseCR_Full);
 		break;
+	case SFF8024_ECC_10GBASE_T_SFI:
+	case SFF8024_ECC_10GBASE_T_SR:
+		phylink_set(modes, 10000baseT_Full);
+		break;
+	case SFF8024_ECC_5GBASE_T:
+		phylink_set(modes, 5000baseT_Full);
+		break;
+	case SFF8024_ECC_2_5GBASE_T:
+		phylink_set(modes, 2500baseT_Full);
+		break;
 	default:
 		dev_warn(bus->sfp_dev,
 			 "Unknown/unsupported extended compliance code: 0x%02x\n",
@@ -301,7 +312,7 @@ void sfp_parse_support(struct sfp_bus *bus, const struct sfp_eeprom_id *id,
 	 */
 	if (bitmap_empty(modes, __ETHTOOL_LINK_MODE_MASK_NBITS)) {
 		/* If the encoding and bit rate allows 1000baseX */
-		if (id->base.encoding == SFP_ENCODING_8B10B && br_nom &&
+		if (id->base.encoding == SFF8024_ENCODING_8B10B && br_nom &&
 		    br_min <= 1300 && br_max >= 1200)
 			phylink_set(modes, 1000baseX_Full);
 	}
@@ -332,7 +343,8 @@ phy_interface_t sfp_select_interface(struct sfp_bus *bus,
 	    phylink_test(link_modes, 10000baseSR_Full) ||
 	    phylink_test(link_modes, 10000baseLR_Full) ||
 	    phylink_test(link_modes, 10000baseLRM_Full) ||
-	    phylink_test(link_modes, 10000baseER_Full))
+	    phylink_test(link_modes, 10000baseER_Full) ||
+	    phylink_test(link_modes, 10000baseT_Full))
 		return PHY_INTERFACE_MODE_10GKR;
 
 	if (phylink_test(link_modes, 2500baseX_Full))
diff --git a/drivers/net/phy/sfp.c b/drivers/net/phy/sfp.c
index ae6a52a19458..ad3808307dba 100644
--- a/drivers/net/phy/sfp.c
+++ b/drivers/net/phy/sfp.c
@@ -242,7 +242,7 @@ struct sfp {
 
 static bool sff_module_supported(const struct sfp_eeprom_id *id)
 {
-	return id->base.phys_id == SFP_PHYS_ID_SFF &&
+	return id->base.phys_id == SFF8024_ID_SFF_8472 &&
 	       id->base.phys_ext_id == SFP_PHYS_EXT_ID_SFP;
 }
 
@@ -253,7 +253,7 @@ static const struct sff_data sff_data = {
 
 static bool sfp_module_supported(const struct sfp_eeprom_id *id)
 {
-	return id->base.phys_id == SFP_PHYS_ID_SFP &&
+	return id->base.phys_id == SFF8024_ID_SFP &&
 	       id->base.phys_ext_id == SFP_PHYS_EXT_ID_SFP;
 }
 
diff --git a/include/linux/sfp.h b/include/linux/sfp.h
index 8d7b98c214d7..373d8b67ea86 100644
--- a/include/linux/sfp.h
+++ b/include/linux/sfp.h
@@ -275,6 +275,61 @@ struct sfp_diag {
 	__be16 cal_v_offset;
 } __packed;
 
+/* SFF8024 defined constants */
+enum {
+	SFF8024_ID_UNK			= 0x00,
+	SFF8024_ID_SFF_8472		= 0x02,
+	SFF8024_ID_SFP			= 0x03,
+	SFF8024_ID_DWDM_SFP		= 0x0b,
+	SFF8024_ID_QSFP_8438		= 0x0c,
+	SFF8024_ID_QSFP_8436_8636	= 0x0d,
+	SFF8024_ID_QSFP28_8636		= 0x11,
+
+	SFF8024_ENCODING_UNSPEC		= 0x00,
+	SFF8024_ENCODING_8B10B		= 0x01,
+	SFF8024_ENCODING_4B5B		= 0x02,
+	SFF8024_ENCODING_NRZ		= 0x03,
+	SFF8024_ENCODING_8472_MANCHESTER= 0x04,
+	SFF8024_ENCODING_8472_SONET	= 0x05,
+	SFF8024_ENCODING_8472_64B66B	= 0x06,
+	SFF8024_ENCODING_8436_MANCHESTER= 0x06,
+	SFF8024_ENCODING_8436_SONET	= 0x04,
+	SFF8024_ENCODING_8436_64B66B	= 0x05,
+	SFF8024_ENCODING_256B257B	= 0x07,
+	SFF8024_ENCODING_PAM4		= 0x08,
+
+	SFF8024_CONNECTOR_UNSPEC	= 0x00,
+	/* codes 01-05 not supportable on SFP, but some modules have single SC */
+	SFF8024_CONNECTOR_SC		= 0x01,
+	SFF8024_CONNECTOR_FIBERJACK	= 0x06,
+	SFF8024_CONNECTOR_LC		= 0x07,
+	SFF8024_CONNECTOR_MT_RJ		= 0x08,
+	SFF8024_CONNECTOR_MU		= 0x09,
+	SFF8024_CONNECTOR_SG		= 0x0a,
+	SFF8024_CONNECTOR_OPTICAL_PIGTAIL= 0x0b,
+	SFF8024_CONNECTOR_MPO_1X12	= 0x0c,
+	SFF8024_CONNECTOR_MPO_2X16	= 0x0d,
+	SFF8024_CONNECTOR_HSSDC_II	= 0x20,
+	SFF8024_CONNECTOR_COPPER_PIGTAIL= 0x21,
+	SFF8024_CONNECTOR_RJ45		= 0x22,
+	SFF8024_CONNECTOR_NOSEPARATE	= 0x23,
+	SFF8024_CONNECTOR_MXC_2X16	= 0x24,
+
+	SFF8024_ECC_UNSPEC		= 0x00,
+	SFF8024_ECC_100G_25GAUI_C2M_AOC	= 0x01,
+	SFF8024_ECC_100GBASE_SR4_25GBASE_SR = 0x02,
+	SFF8024_ECC_100GBASE_LR4_25GBASE_LR = 0x03,
+	SFF8024_ECC_100GBASE_ER4_25GBASE_ER = 0x04,
+	SFF8024_ECC_100GBASE_SR10	= 0x05,
+	SFF8024_ECC_100GBASE_CR4	= 0x0b,
+	SFF8024_ECC_25GBASE_CR_S	= 0x0c,
+	SFF8024_ECC_25GBASE_CR_N	= 0x0d,
+	SFF8024_ECC_10GBASE_T_SFI	= 0x16,
+	SFF8024_ECC_10GBASE_T_SR	= 0x1c,
+	SFF8024_ECC_5GBASE_T		= 0x1d,
+	SFF8024_ECC_2_5GBASE_T		= 0x1e,
+};
+
 /* SFP EEPROM registers */
 enum {
 	SFP_PHYS_ID			= 0x00,
@@ -309,34 +364,7 @@ enum {
 	SFP_SFF8472_COMPLIANCE		= 0x5e,
 	SFP_CC_EXT			= 0x5f,
 
-	SFP_PHYS_ID_SFF			= 0x02,
-	SFP_PHYS_ID_SFP			= 0x03,
 	SFP_PHYS_EXT_ID_SFP		= 0x04,
-	SFP_CONNECTOR_UNSPEC		= 0x00,
-	/* codes 01-05 not supportable on SFP, but some modules have single SC */
-	SFP_CONNECTOR_SC		= 0x01,
-	SFP_CONNECTOR_FIBERJACK		= 0x06,
-	SFP_CONNECTOR_LC		= 0x07,
-	SFP_CONNECTOR_MT_RJ		= 0x08,
-	SFP_CONNECTOR_MU		= 0x09,
-	SFP_CONNECTOR_SG		= 0x0a,
-	SFP_CONNECTOR_OPTICAL_PIGTAIL	= 0x0b,
-	SFP_CONNECTOR_MPO_1X12		= 0x0c,
-	SFP_CONNECTOR_MPO_2X16		= 0x0d,
-	SFP_CONNECTOR_HSSDC_II		= 0x20,
-	SFP_CONNECTOR_COPPER_PIGTAIL	= 0x21,
-	SFP_CONNECTOR_RJ45		= 0x22,
-	SFP_CONNECTOR_NOSEPARATE	= 0x23,
-	SFP_CONNECTOR_MXC_2X16		= 0x24,
-	SFP_ENCODING_UNSPEC		= 0x00,
-	SFP_ENCODING_8B10B		= 0x01,
-	SFP_ENCODING_4B5B		= 0x02,
-	SFP_ENCODING_NRZ		= 0x03,
-	SFP_ENCODING_8472_MANCHESTER	= 0x04,
-	SFP_ENCODING_8472_SONET		= 0x05,
-	SFP_ENCODING_8472_64B66B	= 0x06,
-	SFP_ENCODING_256B257B		= 0x07,
-	SFP_ENCODING_PAM4		= 0x08,
 	SFP_OPTIONS_HIGH_POWER_LEVEL	= BIT(13),
 	SFP_OPTIONS_PAGING_A2		= BIT(12),
 	SFP_OPTIONS_RETIMER		= BIT(11),
-- 
cgit v1.2.3-59-g8ed1b


From 74c551ca5a0edcc9cf66a3b73fd95b9a8615bfd0 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Wed, 11 Dec 2019 10:56:09 +0000
Subject: net: sfp: add module start/stop upstream notifications

When dealing with some copper modules, we can't positively know the
module capabilities are until we have probed the PHY. Without the full
capabilities, we may end up failing a module that we could otherwise
drive with a restricted set of capabilities.

An example of this would be a module with a NBASE-T PHY plugged into
a host that supports phy interface modes 2500BASE-X and SGMII. The
PHY supports 10GBASE-R, 5000BASE-X, 2500BASE-X, SGMII interface modes,
which means a subset of the capabilities are compatible with the host.

However, reading the module EEPROM leads us to believe that the module
only supports ethtool link mode 10GBASE-T, which is incompatible with
the host - and thus results in the module being rejected.

This patch adds an extra notification which are triggered after the
SFP module's PHY probe, and a corresponding notification just before
the PHY is removed.

Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/sfp-bus.c | 21 +++++++++++++++++++++
 drivers/net/phy/sfp.c     |  8 ++++++++
 drivers/net/phy/sfp.h     |  2 ++
 include/linux/sfp.h       |  4 ++++
 4 files changed, 35 insertions(+)

diff --git a/drivers/net/phy/sfp-bus.c b/drivers/net/phy/sfp-bus.c
index c6627f1e5d68..eabc9e3f0a9e 100644
--- a/drivers/net/phy/sfp-bus.c
+++ b/drivers/net/phy/sfp-bus.c
@@ -712,6 +712,27 @@ void sfp_module_remove(struct sfp_bus *bus)
 }
 EXPORT_SYMBOL_GPL(sfp_module_remove);
 
+int sfp_module_start(struct sfp_bus *bus)
+{
+	const struct sfp_upstream_ops *ops = sfp_get_upstream_ops(bus);
+	int ret = 0;
+
+	if (ops && ops->module_start)
+		ret = ops->module_start(bus->upstream);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(sfp_module_start);
+
+void sfp_module_stop(struct sfp_bus *bus)
+{
+	const struct sfp_upstream_ops *ops = sfp_get_upstream_ops(bus);
+
+	if (ops && ops->module_stop)
+		ops->module_stop(bus->upstream);
+}
+EXPORT_SYMBOL_GPL(sfp_module_stop);
+
 static void sfp_socket_clear(struct sfp_bus *bus)
 {
 	bus->sfp_dev = NULL;
diff --git a/drivers/net/phy/sfp.c b/drivers/net/phy/sfp.c
index ad3808307dba..23f30dac0f17 100644
--- a/drivers/net/phy/sfp.c
+++ b/drivers/net/phy/sfp.c
@@ -59,6 +59,7 @@ enum {
 	SFP_DEV_UP,
 
 	SFP_S_DOWN = 0,
+	SFP_S_FAIL,
 	SFP_S_WAIT,
 	SFP_S_INIT,
 	SFP_S_INIT_TX_FAULT,
@@ -122,6 +123,7 @@ static const char *event_to_str(unsigned short event)
 
 static const char * const sm_state_strings[] = {
 	[SFP_S_DOWN] = "down",
+	[SFP_S_FAIL] = "fail",
 	[SFP_S_WAIT] = "wait",
 	[SFP_S_INIT] = "init",
 	[SFP_S_INIT_TX_FAULT] = "init_tx_fault",
@@ -1826,6 +1828,8 @@ static void sfp_sm_main(struct sfp *sfp, unsigned int event)
 		if (sfp->sm_state == SFP_S_LINK_UP &&
 		    sfp->sm_dev_state == SFP_DEV_UP)
 			sfp_sm_link_down(sfp);
+		if (sfp->sm_state > SFP_S_INIT)
+			sfp_module_stop(sfp->sfp_bus);
 		if (sfp->mod_phy)
 			sfp_sm_phy_detach(sfp);
 		sfp_module_tx_disable(sfp);
@@ -1893,6 +1897,10 @@ static void sfp_sm_main(struct sfp *sfp, unsigned int event)
 			 * clear.  Probe for the PHY and check the LOS state.
 			 */
 			sfp_sm_probe_for_phy(sfp);
+			if (sfp_module_start(sfp->sfp_bus)) {
+				sfp_sm_next(sfp, SFP_S_FAIL, 0);
+				break;
+			}
 			sfp_sm_link_check_los(sfp);
 
 			/* Reset the fault retry count */
diff --git a/drivers/net/phy/sfp.h b/drivers/net/phy/sfp.h
index 64f54b0bbd8c..b83f70526270 100644
--- a/drivers/net/phy/sfp.h
+++ b/drivers/net/phy/sfp.h
@@ -22,6 +22,8 @@ void sfp_link_up(struct sfp_bus *bus);
 void sfp_link_down(struct sfp_bus *bus);
 int sfp_module_insert(struct sfp_bus *bus, const struct sfp_eeprom_id *id);
 void sfp_module_remove(struct sfp_bus *bus);
+int sfp_module_start(struct sfp_bus *bus);
+void sfp_module_stop(struct sfp_bus *bus);
 int sfp_link_configure(struct sfp_bus *bus, const struct sfp_eeprom_id *id);
 struct sfp_bus *sfp_register_socket(struct device *dev, struct sfp *sfp,
 				    const struct sfp_socket_ops *ops);
diff --git a/include/linux/sfp.h b/include/linux/sfp.h
index 373d8b67ea86..66a56396e8e3 100644
--- a/include/linux/sfp.h
+++ b/include/linux/sfp.h
@@ -507,6 +507,8 @@ struct sfp_bus;
  * @module_insert: called after a module has been detected to determine
  *   whether the module is supported for the upstream device.
  * @module_remove: called after the module has been removed.
+ * @module_start: called after the PHY probe step
+ * @module_stop: called before the PHY is removed
  * @link_down: called when the link is non-operational for whatever
  *   reason.
  * @link_up: called when the link is operational.
@@ -520,6 +522,8 @@ struct sfp_upstream_ops {
 	void (*detach)(void *priv, struct sfp_bus *bus);
 	int (*module_insert)(void *priv, const struct sfp_eeprom_id *id);
 	void (*module_remove)(void *priv);
+	int (*module_start)(void *priv);
+	void (*module_stop)(void *priv);
 	void (*link_down)(void *priv);
 	void (*link_up)(void *priv);
 	int (*connect_phy)(void *priv, struct phy_device *);
-- 
cgit v1.2.3-59-g8ed1b


From 4882057ad434e8a6434fce20dc7b7315d40a9d4a Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Wed, 11 Dec 2019 10:56:14 +0000
Subject: net: sfp: move phy_start()/phy_stop() to phylink

Move phy_start() and phy_stop() into the module_start and module_stop
notifications in phylink, rather than having them in the SFP code.
This gives phylink responsibility for controlling the PHY, rather
than having SFP start and stop the PHY state machine.

Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/phylink.c | 22 ++++++++++++++++++++++
 drivers/net/phy/sfp.c     |  2 --
 2 files changed, 22 insertions(+), 2 deletions(-)

diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index b3e4d9bf8113..a89219ac4490 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -1767,6 +1767,26 @@ static int phylink_sfp_module_insert(void *upstream,
 	return ret;
 }
 
+static int phylink_sfp_module_start(void *upstream)
+{
+	struct phylink *pl = upstream;
+
+	/* If this SFP module has a PHY, start the PHY now. */
+	if (pl->phydev)
+		phy_start(pl->phydev);
+
+	return 0;
+}
+
+static void phylink_sfp_module_stop(void *upstream)
+{
+	struct phylink *pl = upstream;
+
+	/* If this SFP module has a PHY, stop it. */
+	if (pl->phydev)
+		phy_stop(pl->phydev);
+}
+
 static void phylink_sfp_link_down(void *upstream)
 {
 	struct phylink *pl = upstream;
@@ -1802,6 +1822,8 @@ static const struct sfp_upstream_ops sfp_phylink_ops = {
 	.attach = phylink_sfp_attach,
 	.detach = phylink_sfp_detach,
 	.module_insert = phylink_sfp_module_insert,
+	.module_start = phylink_sfp_module_start,
+	.module_stop = phylink_sfp_module_stop,
 	.link_up = phylink_sfp_link_up,
 	.link_down = phylink_sfp_link_down,
 	.connect_phy = phylink_sfp_connect_phy,
diff --git a/drivers/net/phy/sfp.c b/drivers/net/phy/sfp.c
index 23f30dac0f17..d7d2c797c89c 100644
--- a/drivers/net/phy/sfp.c
+++ b/drivers/net/phy/sfp.c
@@ -1396,7 +1396,6 @@ static void sfp_sm_mod_next(struct sfp *sfp, unsigned int state,
 
 static void sfp_sm_phy_detach(struct sfp *sfp)
 {
-	phy_stop(sfp->mod_phy);
 	sfp_remove_phy(sfp->sfp_bus);
 	phy_device_remove(sfp->mod_phy);
 	phy_device_free(sfp->mod_phy);
@@ -1427,7 +1426,6 @@ static void sfp_sm_probe_phy(struct sfp *sfp)
 	}
 
 	sfp->mod_phy = phy;
-	phy_start(phy);
 }
 
 static void sfp_sm_link_up(struct sfp *sfp)
-- 
cgit v1.2.3-59-g8ed1b


From 6912b712cd75aaa1c8d90b291ec7d2504f17f263 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Wed, 11 Dec 2019 10:56:20 +0000
Subject: net: mdio-i2c: add support for Clause 45 accesses

Some SFP+ modules have PHYs on them just like SFP modules do, except
they are Clause 45 PHYs.  The I2C protocol used to access them is
modified slightly in order to send the device address and 16-bit
register index.

Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/mdio-i2c.c | 28 ++++++++++++++++++++--------
 1 file changed, 20 insertions(+), 8 deletions(-)

diff --git a/drivers/net/phy/mdio-i2c.c b/drivers/net/phy/mdio-i2c.c
index 0dce67672548..0746e2cc39ae 100644
--- a/drivers/net/phy/mdio-i2c.c
+++ b/drivers/net/phy/mdio-i2c.c
@@ -33,17 +33,24 @@ static int i2c_mii_read(struct mii_bus *bus, int phy_id, int reg)
 {
 	struct i2c_adapter *i2c = bus->priv;
 	struct i2c_msg msgs[2];
-	u8 data[2], dev_addr = reg;
+	u8 addr[3], data[2], *p;
 	int bus_addr, ret;
 
 	if (!i2c_mii_valid_phy_id(phy_id))
 		return 0xffff;
 
+	p = addr;
+	if (reg & MII_ADDR_C45) {
+		*p++ = 0x20 | ((reg >> 16) & 31);
+		*p++ = reg >> 8;
+	}
+	*p++ = reg;
+
 	bus_addr = i2c_mii_phy_addr(phy_id);
 	msgs[0].addr = bus_addr;
 	msgs[0].flags = 0;
-	msgs[0].len = 1;
-	msgs[0].buf = &dev_addr;
+	msgs[0].len = p - addr;
+	msgs[0].buf = addr;
 	msgs[1].addr = bus_addr;
 	msgs[1].flags = I2C_M_RD;
 	msgs[1].len = sizeof(data);
@@ -61,18 +68,23 @@ static int i2c_mii_write(struct mii_bus *bus, int phy_id, int reg, u16 val)
 	struct i2c_adapter *i2c = bus->priv;
 	struct i2c_msg msg;
 	int ret;
-	u8 data[3];
+	u8 data[5], *p;
 
 	if (!i2c_mii_valid_phy_id(phy_id))
 		return 0;
 
-	data[0] = reg;
-	data[1] = val >> 8;
-	data[2] = val;
+	p = data;
+	if (reg & MII_ADDR_C45) {
+		*p++ = (reg >> 16) & 31;
+		*p++ = reg >> 8;
+	}
+	*p++ = reg;
+	*p++ = val >> 8;
+	*p++ = val;
 
 	msg.addr = i2c_mii_phy_addr(phy_id);
 	msg.flags = 0;
-	msg.len = 3;
+	msg.len = p - data;
 	msg.buf = data;
 
 	ret = i2c_transfer(i2c, &msg, 1);
-- 
cgit v1.2.3-59-g8ed1b


From 938d44c2b502ba421a785683653e24b685f5d637 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Wed, 11 Dec 2019 10:56:25 +0000
Subject: net: phylink: re-split __phylink_connect_phy()

In order to support Clause 45 PHYs on SFP+ modules, which have an
indeterminant phy interface mode, we need to be able to call
phylink_bringup_phy() with a different interface mode to that used when
binding the PHY. Reduce __phylink_connect_phy() to an attach operation,
and move the call to phylink_bringup_phy() to its call sites.

Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/phylink.c | 39 ++++++++++++++++++++++++---------------
 1 file changed, 24 insertions(+), 15 deletions(-)

diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index a89219ac4490..e5c2a56e3711 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -764,11 +764,9 @@ static int phylink_bringup_phy(struct phylink *pl, struct phy_device *phy)
 	return 0;
 }
 
-static int __phylink_connect_phy(struct phylink *pl, struct phy_device *phy,
-		phy_interface_t interface)
+static int phylink_attach_phy(struct phylink *pl, struct phy_device *phy,
+			      phy_interface_t interface)
 {
-	int ret;
-
 	if (WARN_ON(pl->link_an_mode == MLO_AN_FIXED ||
 		    (pl->link_an_mode == MLO_AN_INBAND &&
 		     phy_interface_mode_is_8023z(interface))))
@@ -777,15 +775,7 @@ static int __phylink_connect_phy(struct phylink *pl, struct phy_device *phy,
 	if (pl->phydev)
 		return -EBUSY;
 
-	ret = phy_attach_direct(pl->netdev, phy, 0, interface);
-	if (ret)
-		return ret;
-
-	ret = phylink_bringup_phy(pl, phy);
-	if (ret)
-		phy_detach(phy);
-
-	return ret;
+	return phy_attach_direct(pl->netdev, phy, 0, interface);
 }
 
 /**
@@ -805,13 +795,23 @@ static int __phylink_connect_phy(struct phylink *pl, struct phy_device *phy,
  */
 int phylink_connect_phy(struct phylink *pl, struct phy_device *phy)
 {
+	int ret;
+
 	/* Use PHY device/driver interface */
 	if (pl->link_interface == PHY_INTERFACE_MODE_NA) {
 		pl->link_interface = phy->interface;
 		pl->link_config.interface = pl->link_interface;
 	}
 
-	return __phylink_connect_phy(pl, phy, pl->link_interface);
+	ret = phylink_attach_phy(pl, phy, pl->link_interface);
+	if (ret < 0)
+		return ret;
+
+	ret = phylink_bringup_phy(pl, phy);
+	if (ret)
+		phy_detach(phy);
+
+	return ret;
 }
 EXPORT_SYMBOL_GPL(phylink_connect_phy);
 
@@ -1809,8 +1809,17 @@ static void phylink_sfp_link_up(void *upstream)
 static int phylink_sfp_connect_phy(void *upstream, struct phy_device *phy)
 {
 	struct phylink *pl = upstream;
+	int ret;
 
-	return __phylink_connect_phy(upstream, phy, pl->link_config.interface);
+	ret = phylink_attach_phy(pl, phy, pl->link_config.interface);
+	if (ret < 0)
+		return ret;
+
+	ret = phylink_bringup_phy(pl, phy);
+	if (ret)
+		phy_detach(phy);
+
+	return ret;
 }
 
 static void phylink_sfp_disconnect_phy(void *upstream)
-- 
cgit v1.2.3-59-g8ed1b


From e45d1f5288b81e77eb87288ae2d0a0d00e0989d3 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Wed, 11 Dec 2019 10:56:30 +0000
Subject: net: phylink: support Clause 45 PHYs on SFP+ modules

Some SFP+ modules have Clause 45 PHYs embedded on them, which need a
little more handling in order to ensure that they are correctly setup,
as they switch the PHY link mode according to the negotiated speed.

With Clause 22 PHYs, we assumed that they would operate in SGMII mode,
but this assumption is now false.  Adapt phylink to support Clause 45
PHYs on SFP+ modules.

Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/phylink.c | 21 ++++++++++++++++-----
 1 file changed, 16 insertions(+), 5 deletions(-)

diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index e5c2a56e3711..aeb8a56a1323 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -711,7 +711,8 @@ static void phylink_phy_change(struct phy_device *phydev, bool up,
 		    phy_duplex_to_str(phydev->duplex));
 }
 
-static int phylink_bringup_phy(struct phylink *pl, struct phy_device *phy)
+static int phylink_bringup_phy(struct phylink *pl, struct phy_device *phy,
+			       phy_interface_t interface)
 {
 	struct phylink_link_state config;
 	__ETHTOOL_DECLARE_LINK_MODE_MASK(supported);
@@ -729,7 +730,7 @@ static int phylink_bringup_phy(struct phylink *pl, struct phy_device *phy)
 	memset(&config, 0, sizeof(config));
 	linkmode_copy(supported, phy->supported);
 	linkmode_copy(config.advertising, phy->advertising);
-	config.interface = pl->link_config.interface;
+	config.interface = interface;
 
 	ret = phylink_validate(pl, supported, &config);
 	if (ret)
@@ -745,6 +746,7 @@ static int phylink_bringup_phy(struct phylink *pl, struct phy_device *phy)
 	mutex_lock(&phy->lock);
 	mutex_lock(&pl->state_mutex);
 	pl->phydev = phy;
+	pl->phy_state.interface = interface;
 	linkmode_copy(pl->supported, supported);
 	linkmode_copy(pl->link_config.advertising, config.advertising);
 
@@ -807,7 +809,7 @@ int phylink_connect_phy(struct phylink *pl, struct phy_device *phy)
 	if (ret < 0)
 		return ret;
 
-	ret = phylink_bringup_phy(pl, phy);
+	ret = phylink_bringup_phy(pl, phy, pl->link_config.interface);
 	if (ret)
 		phy_detach(phy);
 
@@ -860,7 +862,7 @@ int phylink_of_phy_connect(struct phylink *pl, struct device_node *dn,
 	if (!phy_dev)
 		return -ENODEV;
 
-	ret = phylink_bringup_phy(pl, phy_dev);
+	ret = phylink_bringup_phy(pl, phy_dev, pl->link_config.interface);
 	if (ret)
 		phy_detach(phy_dev);
 
@@ -1809,13 +1811,22 @@ static void phylink_sfp_link_up(void *upstream)
 static int phylink_sfp_connect_phy(void *upstream, struct phy_device *phy)
 {
 	struct phylink *pl = upstream;
+	phy_interface_t interface = pl->link_config.interface;
 	int ret;
 
 	ret = phylink_attach_phy(pl, phy, pl->link_config.interface);
 	if (ret < 0)
 		return ret;
 
-	ret = phylink_bringup_phy(pl, phy);
+	/* Clause 45 PHYs switch their Serdes lane between several different
+	 * modes, normally 10GBASE-R, SGMII. Some use 2500BASE-X for 2.5G
+	 * speeds.  We really need to know which interface modes the PHY and
+	 * MAC supports to properly work out which linkmodes can be supported.
+	 */
+	if (phy->is_c45)
+		interface = PHY_INTERFACE_MODE_NA;
+
+	ret = phylink_bringup_phy(pl, phy, interface);
 	if (ret)
 		phy_detach(phy);
 
-- 
cgit v1.2.3-59-g8ed1b


From 24cf0e693bb50a61e3281009a329f7553a4fbd96 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Wed, 11 Dec 2019 10:56:35 +0000
Subject: net: phylink: split link_an_mode configured and current settings

Split link_an_mode between the configured setting and the current
operating setting.  This is an important distinction to make when we
need to configure PHY mode for a plugged SFP+ module that does not
use in-band signalling.

Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/phylink.c | 59 +++++++++++++++++++++++++----------------------
 1 file changed, 31 insertions(+), 28 deletions(-)

diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index aeb8a56a1323..bce530fc956d 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -48,7 +48,8 @@ struct phylink {
 	unsigned long phylink_disable_state; /* bitmask of disables */
 	struct phy_device *phydev;
 	phy_interface_t link_interface;	/* PHY_INTERFACE_xxx */
-	u8 link_an_mode;		/* MLO_AN_xxx */
+	u8 cfg_link_an_mode;		/* MLO_AN_xxx */
+	u8 cur_link_an_mode;
 	u8 link_port;			/* The current non-phy ethtool port */
 	__ETHTOOL_DECLARE_LINK_MODE_MASK(supported);
 
@@ -256,12 +257,12 @@ static int phylink_parse_mode(struct phylink *pl, struct fwnode_handle *fwnode)
 
 	dn = fwnode_get_named_child_node(fwnode, "fixed-link");
 	if (dn || fwnode_property_present(fwnode, "fixed-link"))
-		pl->link_an_mode = MLO_AN_FIXED;
+		pl->cfg_link_an_mode = MLO_AN_FIXED;
 	fwnode_handle_put(dn);
 
 	if (fwnode_property_read_string(fwnode, "managed", &managed) == 0 &&
 	    strcmp(managed, "in-band-status") == 0) {
-		if (pl->link_an_mode == MLO_AN_FIXED) {
+		if (pl->cfg_link_an_mode == MLO_AN_FIXED) {
 			phylink_err(pl,
 				    "can't use both fixed-link and in-band-status\n");
 			return -EINVAL;
@@ -273,7 +274,7 @@ static int phylink_parse_mode(struct phylink *pl, struct fwnode_handle *fwnode)
 		phylink_set(pl->supported, Asym_Pause);
 		phylink_set(pl->supported, Pause);
 		pl->link_config.an_enabled = true;
-		pl->link_an_mode = MLO_AN_INBAND;
+		pl->cfg_link_an_mode = MLO_AN_INBAND;
 
 		switch (pl->link_config.interface) {
 		case PHY_INTERFACE_MODE_SGMII:
@@ -333,14 +334,14 @@ static void phylink_mac_config(struct phylink *pl,
 {
 	phylink_dbg(pl,
 		    "%s: mode=%s/%s/%s/%s adv=%*pb pause=%02x link=%u an=%u\n",
-		    __func__, phylink_an_mode_str(pl->link_an_mode),
+		    __func__, phylink_an_mode_str(pl->cur_link_an_mode),
 		    phy_modes(state->interface),
 		    phy_speed_to_str(state->speed),
 		    phy_duplex_to_str(state->duplex),
 		    __ETHTOOL_LINK_MODE_MASK_NBITS, state->advertising,
 		    state->pause, state->link, state->an_enabled);
 
-	pl->ops->mac_config(pl->config, pl->link_an_mode, state);
+	pl->ops->mac_config(pl->config, pl->cur_link_an_mode, state);
 }
 
 static void phylink_mac_config_up(struct phylink *pl,
@@ -441,7 +442,7 @@ static void phylink_mac_link_up(struct phylink *pl,
 	struct net_device *ndev = pl->netdev;
 
 	pl->cur_interface = link_state.interface;
-	pl->ops->mac_link_up(pl->config, pl->link_an_mode,
+	pl->ops->mac_link_up(pl->config, pl->cur_link_an_mode,
 			     pl->phy_state.interface,
 			     pl->phydev);
 
@@ -461,7 +462,7 @@ static void phylink_mac_link_down(struct phylink *pl)
 
 	if (ndev)
 		netif_carrier_off(ndev);
-	pl->ops->mac_link_down(pl->config, pl->link_an_mode,
+	pl->ops->mac_link_down(pl->config, pl->cur_link_an_mode,
 			       pl->cur_interface);
 	phylink_info(pl, "Link is Down\n");
 }
@@ -480,7 +481,7 @@ static void phylink_resolve(struct work_struct *w)
 	} else if (pl->mac_link_dropped) {
 		link_state.link = false;
 	} else {
-		switch (pl->link_an_mode) {
+		switch (pl->cur_link_an_mode) {
 		case MLO_AN_PHY:
 			link_state = pl->phy_state;
 			phylink_resolve_flow(pl, &link_state);
@@ -648,7 +649,7 @@ struct phylink *phylink_create(struct phylink_config *config,
 		return ERR_PTR(ret);
 	}
 
-	if (pl->link_an_mode == MLO_AN_FIXED) {
+	if (pl->cfg_link_an_mode == MLO_AN_FIXED) {
 		ret = phylink_parse_fixedlink(pl, fwnode);
 		if (ret < 0) {
 			kfree(pl);
@@ -656,6 +657,8 @@ struct phylink *phylink_create(struct phylink_config *config,
 		}
 	}
 
+	pl->cur_link_an_mode = pl->cfg_link_an_mode;
+
 	ret = phylink_register_sfp(pl, fwnode);
 	if (ret < 0) {
 		kfree(pl);
@@ -769,8 +772,8 @@ static int phylink_bringup_phy(struct phylink *pl, struct phy_device *phy,
 static int phylink_attach_phy(struct phylink *pl, struct phy_device *phy,
 			      phy_interface_t interface)
 {
-	if (WARN_ON(pl->link_an_mode == MLO_AN_FIXED ||
-		    (pl->link_an_mode == MLO_AN_INBAND &&
+	if (WARN_ON(pl->cfg_link_an_mode == MLO_AN_FIXED ||
+		    (pl->cfg_link_an_mode == MLO_AN_INBAND &&
 		     phy_interface_mode_is_8023z(interface))))
 		return -EINVAL;
 
@@ -837,8 +840,8 @@ int phylink_of_phy_connect(struct phylink *pl, struct device_node *dn,
 	int ret;
 
 	/* Fixed links and 802.3z are handled without needing a PHY */
-	if (pl->link_an_mode == MLO_AN_FIXED ||
-	    (pl->link_an_mode == MLO_AN_INBAND &&
+	if (pl->cfg_link_an_mode == MLO_AN_FIXED ||
+	    (pl->cfg_link_an_mode == MLO_AN_INBAND &&
 	     phy_interface_mode_is_8023z(pl->link_interface)))
 		return 0;
 
@@ -849,7 +852,7 @@ int phylink_of_phy_connect(struct phylink *pl, struct device_node *dn,
 		phy_node = of_parse_phandle(dn, "phy-device", 0);
 
 	if (!phy_node) {
-		if (pl->link_an_mode == MLO_AN_PHY)
+		if (pl->cfg_link_an_mode == MLO_AN_PHY)
 			return -ENODEV;
 		return 0;
 	}
@@ -912,7 +915,7 @@ int phylink_fixed_state_cb(struct phylink *pl,
 	/* It does not make sense to let the link be overriden unless we use
 	 * MLO_AN_FIXED
 	 */
-	if (pl->link_an_mode != MLO_AN_FIXED)
+	if (pl->cfg_link_an_mode != MLO_AN_FIXED)
 		return -EINVAL;
 
 	mutex_lock(&pl->state_mutex);
@@ -962,7 +965,7 @@ void phylink_start(struct phylink *pl)
 	ASSERT_RTNL();
 
 	phylink_info(pl, "configuring for %s/%s link mode\n",
-		     phylink_an_mode_str(pl->link_an_mode),
+		     phylink_an_mode_str(pl->cur_link_an_mode),
 		     phy_modes(pl->link_config.interface));
 
 	/* Always set the carrier off */
@@ -985,7 +988,7 @@ void phylink_start(struct phylink *pl)
 	clear_bit(PHYLINK_DISABLE_STOPPED, &pl->phylink_disable_state);
 	phylink_run_resolve(pl);
 
-	if (pl->link_an_mode == MLO_AN_FIXED && pl->link_gpio) {
+	if (pl->cfg_link_an_mode == MLO_AN_FIXED && pl->link_gpio) {
 		int irq = gpiod_to_irq(pl->link_gpio);
 
 		if (irq > 0) {
@@ -1000,7 +1003,7 @@ void phylink_start(struct phylink *pl)
 		if (irq <= 0)
 			mod_timer(&pl->link_poll, jiffies + HZ);
 	}
-	if (pl->link_an_mode == MLO_AN_FIXED && pl->get_fixed_state)
+	if (pl->cfg_link_an_mode == MLO_AN_FIXED && pl->get_fixed_state)
 		mod_timer(&pl->link_poll, jiffies + HZ);
 	if (pl->phydev)
 		phy_start(pl->phydev);
@@ -1127,7 +1130,7 @@ int phylink_ethtool_ksettings_get(struct phylink *pl,
 
 	linkmode_copy(kset->link_modes.supported, pl->supported);
 
-	switch (pl->link_an_mode) {
+	switch (pl->cur_link_an_mode) {
 	case MLO_AN_FIXED:
 		/* We are using fixed settings. Report these as the
 		 * current link settings - and note that these also
@@ -1199,7 +1202,7 @@ int phylink_ethtool_ksettings_set(struct phylink *pl,
 		/* If we have a fixed link (as specified by firmware), refuse
 		 * to change link parameters.
 		 */
-		if (pl->link_an_mode == MLO_AN_FIXED &&
+		if (pl->cur_link_an_mode == MLO_AN_FIXED &&
 		    (s->speed != pl->link_config.speed ||
 		     s->duplex != pl->link_config.duplex))
 			return -EINVAL;
@@ -1211,7 +1214,7 @@ int phylink_ethtool_ksettings_set(struct phylink *pl,
 		__clear_bit(ETHTOOL_LINK_MODE_Autoneg_BIT, config.advertising);
 	} else {
 		/* If we have a fixed link, refuse to enable autonegotiation */
-		if (pl->link_an_mode == MLO_AN_FIXED)
+		if (pl->cur_link_an_mode == MLO_AN_FIXED)
 			return -EINVAL;
 
 		config.speed = SPEED_UNKNOWN;
@@ -1253,7 +1256,7 @@ int phylink_ethtool_ksettings_set(struct phylink *pl,
 	 * configuration. For a fixed link, this isn't able to change any
 	 * parameters, which just leaves inband mode.
 	 */
-	if (pl->link_an_mode == MLO_AN_INBAND &&
+	if (pl->cur_link_an_mode == MLO_AN_INBAND &&
 	    !test_bit(PHYLINK_DISABLE_STOPPED, &pl->phylink_disable_state)) {
 		phylink_mac_config(pl, &pl->link_config);
 		phylink_mac_an_restart(pl);
@@ -1343,7 +1346,7 @@ int phylink_ethtool_set_pauseparam(struct phylink *pl,
 				   pause->tx_pause);
 	} else if (!test_bit(PHYLINK_DISABLE_STOPPED,
 			     &pl->phylink_disable_state)) {
-		switch (pl->link_an_mode) {
+		switch (pl->cur_link_an_mode) {
 		case MLO_AN_FIXED:
 			/* Should we allow fixed links to change against the config? */
 			phylink_resolve_flow(pl, config);
@@ -1550,7 +1553,7 @@ static int phylink_mii_read(struct phylink *pl, unsigned int phy_id,
 	struct phylink_link_state state;
 	int val = 0xffff;
 
-	switch (pl->link_an_mode) {
+	switch (pl->cur_link_an_mode) {
 	case MLO_AN_FIXED:
 		if (phy_id == 0) {
 			phylink_get_fixed_state(pl, &state);
@@ -1575,7 +1578,7 @@ static int phylink_mii_read(struct phylink *pl, unsigned int phy_id,
 static int phylink_mii_write(struct phylink *pl, unsigned int phy_id,
 			     unsigned int reg, unsigned int val)
 {
-	switch (pl->link_an_mode) {
+	switch (pl->cur_link_an_mode) {
 	case MLO_AN_FIXED:
 		break;
 
@@ -1748,10 +1751,10 @@ static int phylink_sfp_module_insert(void *upstream,
 		linkmode_copy(pl->link_config.advertising, config.advertising);
 	}
 
-	if (pl->link_an_mode != MLO_AN_INBAND ||
+	if (pl->cur_link_an_mode != MLO_AN_INBAND ||
 	    pl->link_config.interface != config.interface) {
 		pl->link_config.interface = config.interface;
-		pl->link_an_mode = MLO_AN_INBAND;
+		pl->cur_link_an_mode = MLO_AN_INBAND;
 
 		changed = true;
 
-- 
cgit v1.2.3-59-g8ed1b


From c0de2f47202941d936f0482ccd5b010b642a0008 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Wed, 11 Dec 2019 10:56:40 +0000
Subject: net: phylink: split phylink_sfp_module_insert()

Split out the configuration step from phylink_sfp_module_insert() so
we can re-use this later.

Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/phylink.c | 47 ++++++++++++++++++++++++++++-------------------
 1 file changed, 28 insertions(+), 19 deletions(-)

diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index bce530fc956d..571521d928f2 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -1684,25 +1684,21 @@ static void phylink_sfp_detach(void *upstream, struct sfp_bus *bus)
 	pl->netdev->sfp_bus = NULL;
 }
 
-static int phylink_sfp_module_insert(void *upstream,
-				     const struct sfp_eeprom_id *id)
+static int phylink_sfp_config(struct phylink *pl, u8 mode, u8 port,
+			      const unsigned long *supported,
+			      const unsigned long *advertising)
 {
-	struct phylink *pl = upstream;
-	__ETHTOOL_DECLARE_LINK_MODE_MASK(support) = { 0, };
 	__ETHTOOL_DECLARE_LINK_MODE_MASK(support1);
+	__ETHTOOL_DECLARE_LINK_MODE_MASK(support);
 	struct phylink_link_state config;
 	phy_interface_t iface;
-	int ret = 0;
 	bool changed;
-	u8 port;
-
-	ASSERT_RTNL();
+	int ret;
 
-	sfp_parse_support(pl->sfp_bus, id, support);
-	port = sfp_parse_port(pl->sfp_bus, id, support);
+	linkmode_copy(support, supported);
 
 	memset(&config, 0, sizeof(config));
-	linkmode_copy(config.advertising, support);
+	linkmode_copy(config.advertising, advertising);
 	config.interface = PHY_INTERFACE_MODE_NA;
 	config.speed = SPEED_UNKNOWN;
 	config.duplex = DUPLEX_UNKNOWN;
@@ -1717,8 +1713,6 @@ static int phylink_sfp_module_insert(void *upstream,
 		return ret;
 	}
 
-	linkmode_copy(support1, support);
-
 	iface = sfp_select_interface(pl->sfp_bus, config.advertising);
 	if (iface == PHY_INTERFACE_MODE_NA) {
 		phylink_err(pl,
@@ -1728,18 +1722,18 @@ static int phylink_sfp_module_insert(void *upstream,
 	}
 
 	config.interface = iface;
+	linkmode_copy(support1, support);
 	ret = phylink_validate(pl, support1, &config);
 	if (ret) {
 		phylink_err(pl, "validation of %s/%s with support %*pb failed: %d\n",
-			    phylink_an_mode_str(MLO_AN_INBAND),
+			    phylink_an_mode_str(mode),
 			    phy_modes(config.interface),
 			    __ETHTOOL_LINK_MODE_MASK_NBITS, support, ret);
 		return ret;
 	}
 
 	phylink_dbg(pl, "requesting link mode %s/%s with support %*pb\n",
-		    phylink_an_mode_str(MLO_AN_INBAND),
-		    phy_modes(config.interface),
+		    phylink_an_mode_str(mode), phy_modes(config.interface),
 		    __ETHTOOL_LINK_MODE_MASK_NBITS, support);
 
 	if (phy_interface_mode_is_8023z(iface) && pl->phydev)
@@ -1751,15 +1745,15 @@ static int phylink_sfp_module_insert(void *upstream,
 		linkmode_copy(pl->link_config.advertising, config.advertising);
 	}
 
-	if (pl->cur_link_an_mode != MLO_AN_INBAND ||
+	if (pl->cur_link_an_mode != mode ||
 	    pl->link_config.interface != config.interface) {
 		pl->link_config.interface = config.interface;
-		pl->cur_link_an_mode = MLO_AN_INBAND;
+		pl->cur_link_an_mode = mode;
 
 		changed = true;
 
 		phylink_info(pl, "switched to %s/%s link mode\n",
-			     phylink_an_mode_str(MLO_AN_INBAND),
+			     phylink_an_mode_str(mode),
 			     phy_modes(config.interface));
 	}
 
@@ -1772,6 +1766,21 @@ static int phylink_sfp_module_insert(void *upstream,
 	return ret;
 }
 
+static int phylink_sfp_module_insert(void *upstream,
+				     const struct sfp_eeprom_id *id)
+{
+	struct phylink *pl = upstream;
+	__ETHTOOL_DECLARE_LINK_MODE_MASK(support) = { 0, };
+	u8 port;
+
+	ASSERT_RTNL();
+
+	sfp_parse_support(pl->sfp_bus, id, support);
+	port = sfp_parse_port(pl->sfp_bus, id, support);
+
+	return phylink_sfp_config(pl, MLO_AN_INBAND, port, support, support);
+}
+
 static int phylink_sfp_module_start(void *upstream)
 {
 	struct phylink *pl = upstream;
-- 
cgit v1.2.3-59-g8ed1b


From 52c956003a9d5bcae1f445f9dfd42b624adb6e87 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Wed, 11 Dec 2019 10:56:45 +0000
Subject: net: phylink: delay MAC configuration for copper SFP modules

Knowing whether we need to delay the MAC configuration because a module
may have a PHY is useful to phylink to allow NBASE-T modules to work on
systems supporting no more than 2.5G speeds.

This commit allows us to delay such configuration until after the PHY
has been probed by recording the parsed capabilities, and if the module
may have a PHY, doing no more until the module_start() notification is
called.  At that point, we either have a PHY, or we don't.

We move the PHY-based setup a little later, and use the PHYs support
capabilities rather than the EEPROM parsed capabilities to determine
whether we can support the PHY.

Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/phylink.c | 53 ++++++++++++++++++++++++++++++++++++++---------
 drivers/net/phy/sfp-bus.c | 28 +++++++++++++++++++++++++
 include/linux/sfp.h       |  7 +++++++
 3 files changed, 78 insertions(+), 10 deletions(-)

diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index 571521d928f2..01f958508f80 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -72,6 +72,9 @@ struct phylink {
 	bool mac_link_dropped;
 
 	struct sfp_bus *sfp_bus;
+	bool sfp_may_have_phy;
+	__ETHTOOL_DECLARE_LINK_MODE_MASK(sfp_support);
+	u8 sfp_port;
 };
 
 #define phylink_printk(level, pl, fmt, ...) \
@@ -1684,7 +1687,7 @@ static void phylink_sfp_detach(void *upstream, struct sfp_bus *bus)
 	pl->netdev->sfp_bus = NULL;
 }
 
-static int phylink_sfp_config(struct phylink *pl, u8 mode, u8 port,
+static int phylink_sfp_config(struct phylink *pl, u8 mode,
 			      const unsigned long *supported,
 			      const unsigned long *advertising)
 {
@@ -1757,7 +1760,7 @@ static int phylink_sfp_config(struct phylink *pl, u8 mode, u8 port,
 			     phy_modes(config.interface));
 	}
 
-	pl->link_port = port;
+	pl->link_port = pl->sfp_port;
 
 	if (changed && !test_bit(PHYLINK_DISABLE_STOPPED,
 				 &pl->phylink_disable_state))
@@ -1770,15 +1773,20 @@ static int phylink_sfp_module_insert(void *upstream,
 				     const struct sfp_eeprom_id *id)
 {
 	struct phylink *pl = upstream;
-	__ETHTOOL_DECLARE_LINK_MODE_MASK(support) = { 0, };
-	u8 port;
+	unsigned long *support = pl->sfp_support;
 
 	ASSERT_RTNL();
 
+	linkmode_zero(support);
 	sfp_parse_support(pl->sfp_bus, id, support);
-	port = sfp_parse_port(pl->sfp_bus, id, support);
+	pl->sfp_port = sfp_parse_port(pl->sfp_bus, id, support);
 
-	return phylink_sfp_config(pl, MLO_AN_INBAND, port, support, support);
+	/* If this module may have a PHY connecting later, defer until later */
+	pl->sfp_may_have_phy = sfp_may_have_phy(pl->sfp_bus, id);
+	if (pl->sfp_may_have_phy)
+		return 0;
+
+	return phylink_sfp_config(pl, MLO_AN_INBAND, support, support);
 }
 
 static int phylink_sfp_module_start(void *upstream)
@@ -1786,10 +1794,19 @@ static int phylink_sfp_module_start(void *upstream)
 	struct phylink *pl = upstream;
 
 	/* If this SFP module has a PHY, start the PHY now. */
-	if (pl->phydev)
+	if (pl->phydev) {
 		phy_start(pl->phydev);
+		return 0;
+	}
 
-	return 0;
+	/* If the module may have a PHY but we didn't detect one we
+	 * need to configure the MAC here.
+	 */
+	if (!pl->sfp_may_have_phy)
+		return 0;
+
+	return phylink_sfp_config(pl, MLO_AN_INBAND,
+				  pl->sfp_support, pl->sfp_support);
 }
 
 static void phylink_sfp_module_stop(void *upstream)
@@ -1823,10 +1840,26 @@ static void phylink_sfp_link_up(void *upstream)
 static int phylink_sfp_connect_phy(void *upstream, struct phy_device *phy)
 {
 	struct phylink *pl = upstream;
-	phy_interface_t interface = pl->link_config.interface;
+	phy_interface_t interface;
 	int ret;
 
-	ret = phylink_attach_phy(pl, phy, pl->link_config.interface);
+	/*
+	 * This is the new way of dealing with flow control for PHYs,
+	 * as described by Timur Tabi in commit 529ed1275263 ("net: phy:
+	 * phy drivers should not set SUPPORTED_[Asym_]Pause") except
+	 * using our validate call to the MAC, we rely upon the MAC
+	 * clearing the bits from both supported and advertising fields.
+	 */
+	phy_support_asym_pause(phy);
+
+	/* Do the initial configuration */
+	ret = phylink_sfp_config(pl, MLO_AN_INBAND, phy->supported,
+				 phy->advertising);
+	if (ret < 0)
+		return ret;
+
+	interface = pl->link_config.interface;
+	ret = phylink_attach_phy(pl, phy, interface);
 	if (ret < 0)
 		return ret;
 
diff --git a/drivers/net/phy/sfp-bus.c b/drivers/net/phy/sfp-bus.c
index eabc9e3f0a9e..06e6429b8b71 100644
--- a/drivers/net/phy/sfp-bus.c
+++ b/drivers/net/phy/sfp-bus.c
@@ -103,6 +103,7 @@ static const struct sfp_quirk *sfp_lookup_quirk(const struct sfp_eeprom_id *id)
 
 	return NULL;
 }
+
 /**
  * sfp_parse_port() - Parse the EEPROM base ID, setting the port type
  * @bus: a pointer to the &struct sfp_bus structure for the sfp module
@@ -178,6 +179,33 @@ int sfp_parse_port(struct sfp_bus *bus, const struct sfp_eeprom_id *id,
 }
 EXPORT_SYMBOL_GPL(sfp_parse_port);
 
+/**
+ * sfp_may_have_phy() - indicate whether the module may have a PHY
+ * @bus: a pointer to the &struct sfp_bus structure for the sfp module
+ * @id: a pointer to the module's &struct sfp_eeprom_id
+ *
+ * Parse the EEPROM identification given in @id, and return whether
+ * this module may have a PHY.
+ */
+bool sfp_may_have_phy(struct sfp_bus *bus, const struct sfp_eeprom_id *id)
+{
+	if (id->base.e1000_base_t)
+		return true;
+
+	if (id->base.phys_id != SFF8024_ID_DWDM_SFP) {
+		switch (id->base.extended_cc) {
+		case SFF8024_ECC_10GBASE_T_SFI:
+		case SFF8024_ECC_10GBASE_T_SR:
+		case SFF8024_ECC_5GBASE_T:
+		case SFF8024_ECC_2_5GBASE_T:
+			return true;
+		}
+	}
+
+	return false;
+}
+EXPORT_SYMBOL_GPL(sfp_may_have_phy);
+
 /**
  * sfp_parse_support() - Parse the eeprom id for supported link modes
  * @bus: a pointer to the &struct sfp_bus structure for the sfp module
diff --git a/include/linux/sfp.h b/include/linux/sfp.h
index 66a56396e8e3..38893e4dd0f0 100644
--- a/include/linux/sfp.h
+++ b/include/linux/sfp.h
@@ -533,6 +533,7 @@ struct sfp_upstream_ops {
 #if IS_ENABLED(CONFIG_SFP)
 int sfp_parse_port(struct sfp_bus *bus, const struct sfp_eeprom_id *id,
 		   unsigned long *support);
+bool sfp_may_have_phy(struct sfp_bus *bus, const struct sfp_eeprom_id *id);
 void sfp_parse_support(struct sfp_bus *bus, const struct sfp_eeprom_id *id,
 		       unsigned long *support);
 phy_interface_t sfp_select_interface(struct sfp_bus *bus,
@@ -556,6 +557,12 @@ static inline int sfp_parse_port(struct sfp_bus *bus,
 	return PORT_OTHER;
 }
 
+static inline bool sfp_may_have_phy(struct sfp_bus *bus,
+				    const struct sfp_eeprom_id *id)
+{
+	return false;
+}
+
 static inline void sfp_parse_support(struct sfp_bus *bus,
 				     const struct sfp_eeprom_id *id,
 				     unsigned long *support)
-- 
cgit v1.2.3-59-g8ed1b


From 7adb5b2126bc013f0964ddaefad6ad1b132e86c3 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Wed, 11 Dec 2019 10:56:50 +0000
Subject: net: phylink: make Broadcom BCM84881 based SFPs work

The Broadcom BCM84881 does not appear to send the SGMII control word
when operating in SGMII mode, which causes network adapters to fail
to link with the PHY, or decide to operate at fixed 1G speed, even if
the PHY negotiated 100M.

Work around this by detecting the Broadcom BCM84881 and switch to phy
mode rather than inband mode.

Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/phylink.c | 18 ++++++++++++++++--
 1 file changed, 16 insertions(+), 2 deletions(-)

diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index 01f958508f80..47614f975344 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -1837,10 +1837,20 @@ static void phylink_sfp_link_up(void *upstream)
 	phylink_run_resolve(pl);
 }
 
+/* The Broadcom BCM84881 in the Methode DM7052 is unable to provide a SGMII
+ * or 802.3z control word, so inband will not work.
+ */
+static bool phylink_phy_no_inband(struct phy_device *phy)
+{
+	return phy->is_c45 &&
+		(phy->c45_ids.device_ids[1] & 0xfffffff0) == 0xae025150;
+}
+
 static int phylink_sfp_connect_phy(void *upstream, struct phy_device *phy)
 {
 	struct phylink *pl = upstream;
 	phy_interface_t interface;
+	u8 mode;
 	int ret;
 
 	/*
@@ -1852,9 +1862,13 @@ static int phylink_sfp_connect_phy(void *upstream, struct phy_device *phy)
 	 */
 	phy_support_asym_pause(phy);
 
+	if (phylink_phy_no_inband(phy))
+		mode = MLO_AN_PHY;
+	else
+		mode = MLO_AN_INBAND;
+
 	/* Do the initial configuration */
-	ret = phylink_sfp_config(pl, MLO_AN_INBAND, phy->supported,
-				 phy->advertising);
+	ret = phylink_sfp_config(pl, mode, phy->supported, phy->advertising);
 	if (ret < 0)
 		return ret;
 
-- 
cgit v1.2.3-59-g8ed1b


From 75f4d8d10e016f7428c268424483a927ee7a78bb Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Wed, 11 Dec 2019 10:56:56 +0000
Subject: net: phy: add Broadcom BCM84881 PHY driver

Add a rudimentary Clause 45 driver for the BCM84881 PHY, found on
Methode DM7052 SFPs.

Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/Kconfig    |   6 +
 drivers/net/phy/Makefile   |   1 +
 drivers/net/phy/bcm84881.c | 269 +++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 276 insertions(+)
 create mode 100644 drivers/net/phy/bcm84881.c

diff --git a/drivers/net/phy/Kconfig b/drivers/net/phy/Kconfig
index 5848219005d7..2e016271e126 100644
--- a/drivers/net/phy/Kconfig
+++ b/drivers/net/phy/Kconfig
@@ -324,6 +324,12 @@ config BROADCOM_PHY
 	  Currently supports the BCM5411, BCM5421, BCM5461, BCM54616S, BCM5464,
 	  BCM5481, BCM54810 and BCM5482 PHYs.
 
+config BCM84881_PHY
+	bool "Broadcom BCM84881 PHY"
+	depends on PHYLIB=y
+	---help---
+	  Support the Broadcom BCM84881 PHY.
+
 config CICADA_PHY
 	tristate "Cicada PHYs"
 	---help---
diff --git a/drivers/net/phy/Makefile b/drivers/net/phy/Makefile
index b433ec3bf9a6..d846b4dc1c68 100644
--- a/drivers/net/phy/Makefile
+++ b/drivers/net/phy/Makefile
@@ -62,6 +62,7 @@ obj-$(CONFIG_BCM87XX_PHY)	+= bcm87xx.o
 obj-$(CONFIG_BCM_CYGNUS_PHY)	+= bcm-cygnus.o
 obj-$(CONFIG_BCM_NET_PHYLIB)	+= bcm-phy-lib.o
 obj-$(CONFIG_BROADCOM_PHY)	+= broadcom.o
+obj-$(CONFIG_BCM84881_PHY)	+= bcm84881.o
 obj-$(CONFIG_CICADA_PHY)	+= cicada.o
 obj-$(CONFIG_CORTINA_PHY)	+= cortina.o
 obj-$(CONFIG_DAVICOM_PHY)	+= davicom.o
diff --git a/drivers/net/phy/bcm84881.c b/drivers/net/phy/bcm84881.c
new file mode 100644
index 000000000000..db59911b9b3c
--- /dev/null
+++ b/drivers/net/phy/bcm84881.c
@@ -0,0 +1,269 @@
+// SPDX-License-Identifier: GPL-2.0
+// Broadcom BCM84881 NBASE-T PHY driver, as found on a SFP+ module.
+// Copyright (C) 2019 Russell King, Deep Blue Solutions Ltd.
+//
+// Like the Marvell 88x3310, the Broadcom 84881 changes its host-side
+// interface according to the operating speed between 10GBASE-R,
+// 2500BASE-X and SGMII (but unlike the 88x3310, without the control
+// word).
+//
+// This driver only supports those aspects of the PHY that I'm able to
+// observe and test with the SFP+ module, which is an incomplete subset
+// of what this PHY is able to support. For example, I only assume it
+// supports a single lane Serdes connection, but it may be that the PHY
+// is able to support more than that.
+#include <linux/delay.h>
+#include <linux/module.h>
+#include <linux/phy.h>
+
+enum {
+	MDIO_AN_C22 = 0xffe0,
+};
+
+static int bcm84881_wait_init(struct phy_device *phydev)
+{
+	unsigned int tries = 20;
+	int ret, val;
+
+	do {
+		val = phy_read_mmd(phydev, MDIO_MMD_PMAPMD, MDIO_CTRL1);
+		if (val < 0) {
+			ret = val;
+			break;
+		}
+		if (!(val & MDIO_CTRL1_RESET)) {
+			ret = 0;
+			break;
+		}
+		if (!--tries) {
+			ret = -ETIMEDOUT;
+			break;
+		}
+		msleep(100);
+	} while (1);
+
+	if (ret)
+		phydev_err(phydev, "%s failed: %d\n", __func__, ret);
+
+	return ret;
+}
+
+static int bcm84881_config_init(struct phy_device *phydev)
+{
+	switch (phydev->interface) {
+	case PHY_INTERFACE_MODE_SGMII:
+	case PHY_INTERFACE_MODE_2500BASEX:
+	case PHY_INTERFACE_MODE_10GKR:
+		break;
+	default:
+		return -ENODEV;
+	}
+	return 0;
+}
+
+static int bcm84881_probe(struct phy_device *phydev)
+{
+	/* This driver requires PMAPMD and AN blocks */
+	const u32 mmd_mask = MDIO_DEVS_PMAPMD | MDIO_DEVS_AN;
+
+	if (!phydev->is_c45 ||
+	    (phydev->c45_ids.devices_in_package & mmd_mask) != mmd_mask)
+		return -ENODEV;
+
+	return 0;
+}
+
+static int bcm84881_get_features(struct phy_device *phydev)
+{
+	int ret;
+
+	ret = genphy_c45_pma_read_abilities(phydev);
+	if (ret)
+		return ret;
+
+	/* Although the PHY sets bit 1.11.8, it does not support 10M modes */
+	linkmode_clear_bit(ETHTOOL_LINK_MODE_10baseT_Half_BIT,
+			   phydev->supported);
+	linkmode_clear_bit(ETHTOOL_LINK_MODE_10baseT_Full_BIT,
+			   phydev->supported);
+
+	return 0;
+}
+
+static int bcm84881_config_aneg(struct phy_device *phydev)
+{
+	bool changed = false;
+	u32 adv;
+	int ret;
+
+	/* Wait for the PHY to finish initialising, otherwise our
+	 * advertisement may be overwritten.
+	 */
+	ret = bcm84881_wait_init(phydev);
+	if (ret)
+		return ret;
+
+	/* We don't support manual MDI control */
+	phydev->mdix_ctrl = ETH_TP_MDI_AUTO;
+
+	/* disabled autoneg doesn't seem to work with this PHY */
+	if (phydev->autoneg == AUTONEG_DISABLE)
+		return -EINVAL;
+
+	ret = genphy_c45_an_config_aneg(phydev);
+	if (ret < 0)
+		return ret;
+	if (ret > 0)
+		changed = true;
+
+	adv = linkmode_adv_to_mii_ctrl1000_t(phydev->advertising);
+	ret = phy_modify_mmd_changed(phydev, MDIO_MMD_AN,
+				     MDIO_AN_C22 + MII_CTRL1000,
+				     ADVERTISE_1000FULL | ADVERTISE_1000HALF,
+				     adv);
+	if (ret < 0)
+		return ret;
+	if (ret > 0)
+		changed = true;
+
+	return genphy_c45_check_and_restart_aneg(phydev, changed);
+}
+
+static int bcm84881_aneg_done(struct phy_device *phydev)
+{
+	int bmsr, val;
+
+	val = phy_read_mmd(phydev, MDIO_MMD_AN, MDIO_STAT1);
+	if (val < 0)
+		return val;
+
+	bmsr = phy_read_mmd(phydev, MDIO_MMD_AN, MDIO_AN_C22 + MII_BMSR);
+	if (bmsr < 0)
+		return val;
+
+	return !!(val & MDIO_AN_STAT1_COMPLETE) &&
+	       !!(bmsr & BMSR_ANEGCOMPLETE);
+}
+
+static int bcm84881_read_status(struct phy_device *phydev)
+{
+	unsigned int mode;
+	int bmsr, val;
+
+	val = phy_read_mmd(phydev, MDIO_MMD_AN, MDIO_CTRL1);
+	if (val < 0)
+		return val;
+
+	if (val & MDIO_AN_CTRL1_RESTART) {
+		phydev->link = 0;
+		return 0;
+	}
+
+	val = phy_read_mmd(phydev, MDIO_MMD_AN, MDIO_STAT1);
+	if (val < 0)
+		return val;
+
+	bmsr = phy_read_mmd(phydev, MDIO_MMD_AN, MDIO_AN_C22 + MII_BMSR);
+	if (bmsr < 0)
+		return val;
+
+	phydev->autoneg_complete = !!(val & MDIO_AN_STAT1_COMPLETE) &&
+				   !!(bmsr & BMSR_ANEGCOMPLETE);
+	phydev->link = !!(val & MDIO_STAT1_LSTATUS) &&
+		       !!(bmsr & BMSR_LSTATUS);
+	if (phydev->autoneg == AUTONEG_ENABLE && !phydev->autoneg_complete)
+		phydev->link = false;
+
+	if (!phydev->link)
+		return 0;
+
+	linkmode_zero(phydev->lp_advertising);
+	phydev->speed = SPEED_UNKNOWN;
+	phydev->duplex = DUPLEX_UNKNOWN;
+	phydev->pause = 0;
+	phydev->asym_pause = 0;
+	phydev->mdix = 0;
+
+	if (phydev->autoneg_complete) {
+		val = genphy_c45_read_lpa(phydev);
+		if (val < 0)
+			return val;
+
+		val = phy_read_mmd(phydev, MDIO_MMD_AN,
+				   MDIO_AN_C22 + MII_STAT1000);
+		if (val < 0)
+			return val;
+
+		mii_stat1000_mod_linkmode_lpa_t(phydev->lp_advertising, val);
+
+		if (phydev->autoneg == AUTONEG_ENABLE)
+			phy_resolve_aneg_linkmode(phydev);
+	}
+
+	if (phydev->autoneg == AUTONEG_DISABLE) {
+		/* disabled autoneg doesn't seem to work, so force the link
+		 * down.
+		 */
+		phydev->link = 0;
+		return 0;
+	}
+
+	/* Set the host link mode - we set the phy interface mode and
+	 * the speed according to this register so that downshift works.
+	 * We leave the duplex setting as per the resolution from the
+	 * above.
+	 */
+	val = phy_read_mmd(phydev, MDIO_MMD_VEND1, 0x4011);
+	mode = (val & 0x1e) >> 1;
+	if (mode == 1 || mode == 2)
+		phydev->interface = PHY_INTERFACE_MODE_SGMII;
+	else if (mode == 3)
+		phydev->interface = PHY_INTERFACE_MODE_10GKR;
+	else if (mode == 4)
+		phydev->interface = PHY_INTERFACE_MODE_2500BASEX;
+	switch (mode & 7) {
+	case 1:
+		phydev->speed = SPEED_100;
+		break;
+	case 2:
+		phydev->speed = SPEED_1000;
+		break;
+	case 3:
+		phydev->speed = SPEED_10000;
+		break;
+	case 4:
+		phydev->speed = SPEED_2500;
+		break;
+	case 5:
+		phydev->speed = SPEED_5000;
+		break;
+	}
+
+	return genphy_c45_read_mdix(phydev);
+}
+
+static struct phy_driver bcm84881_drivers[] = {
+	{
+		.phy_id		= 0xae025150,
+		.phy_id_mask	= 0xfffffff0,
+		.name		= "Broadcom BCM84881",
+		.config_init	= bcm84881_config_init,
+		.probe		= bcm84881_probe,
+		.get_features	= bcm84881_get_features,
+		.config_aneg	= bcm84881_config_aneg,
+		.aneg_done	= bcm84881_aneg_done,
+		.read_status	= bcm84881_read_status,
+	},
+};
+
+module_phy_driver(bcm84881_drivers);
+
+/* FIXME: module auto-loading for Clause 45 PHYs seems non-functional */
+static struct mdio_device_id __maybe_unused bcm84881_tbl[] = {
+	{ 0xae025150, 0xfffffff0 },
+	{ },
+};
+MODULE_AUTHOR("Russell King");
+MODULE_DESCRIPTION("Broadcom BCM84881 PHY driver");
+MODULE_DEVICE_TABLE(mdio, bcm84881_tbl);
+MODULE_LICENSE("GPL");
-- 
cgit v1.2.3-59-g8ed1b


From 9a484621e88ac8386f131596ce8d76668daf7996 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Wed, 11 Dec 2019 10:57:01 +0000
Subject: net: sfp: add support for Clause 45 PHYs

Some SFP+ modules have a Clause 45 PHY onboard, which is accessible via
the normal I2C address.  Detect 10G BASE-T PHYs which may have an
accessible PHY and probe for it.

Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/sfp.c | 44 ++++++++++++++++++++++++++++++++++++++++----
 1 file changed, 40 insertions(+), 4 deletions(-)

diff --git a/drivers/net/phy/sfp.c b/drivers/net/phy/sfp.c
index d7d2c797c89c..bfe268028154 100644
--- a/drivers/net/phy/sfp.c
+++ b/drivers/net/phy/sfp.c
@@ -1402,12 +1402,12 @@ static void sfp_sm_phy_detach(struct sfp *sfp)
 	sfp->mod_phy = NULL;
 }
 
-static void sfp_sm_probe_phy(struct sfp *sfp)
+static void sfp_sm_probe_phy(struct sfp *sfp, bool is_c45)
 {
 	struct phy_device *phy;
 	int err;
 
-	phy = mdiobus_scan(sfp->i2c_mii, SFP_PHY_ADDR);
+	phy = get_phy_device(sfp->i2c_mii, SFP_PHY_ADDR, is_c45);
 	if (phy == ERR_PTR(-ENODEV)) {
 		dev_info(sfp->dev, "no PHY detected\n");
 		return;
@@ -1417,6 +1417,13 @@ static void sfp_sm_probe_phy(struct sfp *sfp)
 		return;
 	}
 
+	err = phy_device_register(phy);
+	if (err) {
+		phy_device_free(phy);
+		dev_err(sfp->dev, "phy_device_register failed: %d\n", err);
+		return;
+	}
+
 	err = sfp_add_phy(sfp->sfp_bus, phy);
 	if (err) {
 		phy_device_remove(phy);
@@ -1487,10 +1494,32 @@ static void sfp_sm_fault(struct sfp *sfp, unsigned int next_state, bool warn)
 	}
 }
 
+/* Probe a SFP for a PHY device if the module supports copper - the PHY
+ * normally sits at I2C bus address 0x56, and may either be a clause 22
+ * or clause 45 PHY.
+ *
+ * Clause 22 copper SFP modules normally operate in Cisco SGMII mode with
+ * negotiation enabled, but some may be in 1000base-X - which is for the
+ * PHY driver to determine.
+ *
+ * Clause 45 copper SFP+ modules (10G) appear to switch their interface
+ * mode according to the negotiated line speed.
+ */
 static void sfp_sm_probe_for_phy(struct sfp *sfp)
 {
-	if (sfp->id.base.e1000_base_t)
-		sfp_sm_probe_phy(sfp);
+	switch (sfp->id.base.extended_cc) {
+	case SFF8024_ECC_10GBASE_T_SFI:
+	case SFF8024_ECC_10GBASE_T_SR:
+	case SFF8024_ECC_5GBASE_T:
+	case SFF8024_ECC_2_5GBASE_T:
+		sfp_sm_probe_phy(sfp, true);
+		break;
+
+	default:
+		if (sfp->id.base.e1000_base_t)
+			sfp_sm_probe_phy(sfp, false);
+		break;
+	}
 }
 
 static int sfp_module_parse_power(struct sfp *sfp)
@@ -1550,6 +1579,13 @@ static int sfp_sm_mod_hpower(struct sfp *sfp, bool enable)
 		return -EAGAIN;
 	}
 
+	/* DM7052 reports as a high power module, responds to reads (with
+	 * all bytes 0xff) at 0x51 but does not accept writes.  In any case,
+	 * if the bit is already set, we're already in high power mode.
+	 */
+	if (!!(val & BIT(0)) == enable)
+		return 0;
+
 	if (enable)
 		val |= BIT(0);
 	else
-- 
cgit v1.2.3-59-g8ed1b


From 65ef2d5c41d93c2fc28bfa6ac2b60232d1d0c677 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Mon, 9 Dec 2019 14:15:55 +0000
Subject: net: sfp: use a definition for the fault recovery attempts

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/sfp.c | 14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)

diff --git a/drivers/net/phy/sfp.c b/drivers/net/phy/sfp.c
index bfe268028154..b1c564b79e3e 100644
--- a/drivers/net/phy/sfp.c
+++ b/drivers/net/phy/sfp.c
@@ -172,6 +172,14 @@ static const enum gpiod_flags gpio_flags[] = {
 #define T_RESET_US		10
 #define T_FAULT_RECOVER		msecs_to_jiffies(1000)
 
+/* N_FAULT_INIT is the number of recovery attempts at module initialisation
+ * time. If the TX_FAULT signal is not deasserted after this number of
+ * attempts at clearing it, we decide that the module is faulty.
+ * N_FAULT is the same but after the module has initialised.
+ */
+#define N_FAULT_INIT		5
+#define N_FAULT			5
+
 /* SFP module presence detection is poor: the three MOD DEF signals are
  * the same length on the PCB, which means it's possible for MOD DEF 0 to
  * connect before the I2C bus on MOD DEF 1/2.
@@ -1885,7 +1893,7 @@ static void sfp_sm_main(struct sfp *sfp, unsigned int event)
 		sfp_module_tx_enable(sfp);
 
 		/* Initialise the fault clearance retries */
-		sfp->sm_retries = 5;
+		sfp->sm_retries = N_FAULT_INIT;
 
 		/* We need to check the TX_FAULT state, which is not defined
 		 * while TX_DISABLE is asserted. The earliest we want to do
@@ -1925,7 +1933,7 @@ static void sfp_sm_main(struct sfp *sfp, unsigned int event)
 			 * or t_start_up, so assume there is a fault.
 			 */
 			sfp_sm_fault(sfp, SFP_S_INIT_TX_FAULT,
-				     sfp->sm_retries == 5);
+				     sfp->sm_retries == N_FAULT_INIT);
 		} else if (event == SFP_E_TIMEOUT || event == SFP_E_TX_CLEAR) {
 	init_done:	/* TX_FAULT deasserted or we timed out with TX_FAULT
 			 * clear.  Probe for the PHY and check the LOS state.
@@ -1938,7 +1946,7 @@ static void sfp_sm_main(struct sfp *sfp, unsigned int event)
 			sfp_sm_link_check_los(sfp);
 
 			/* Reset the fault retry count */
-			sfp->sm_retries = 5;
+			sfp->sm_retries = N_FAULT;
 		}
 		break;
 
-- 
cgit v1.2.3-59-g8ed1b


From 281e4eab1abeb179610de0884a4bd65bf2d4e61e Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Mon, 9 Dec 2019 14:16:00 +0000
Subject: net: sfp: rename sm_retries

Rename sm_retries as sm_fault_retries, as this is what this member is
tracking.

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/sfp.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/phy/sfp.c b/drivers/net/phy/sfp.c
index b1c564b79e3e..a67f089f2106 100644
--- a/drivers/net/phy/sfp.c
+++ b/drivers/net/phy/sfp.c
@@ -234,7 +234,7 @@ struct sfp {
 	unsigned char sm_mod_tries;
 	unsigned char sm_dev_state;
 	unsigned short sm_state;
-	unsigned int sm_retries;
+	unsigned char sm_fault_retries;
 
 	struct sfp_eeprom_id id;
 	unsigned int module_power_mW;
@@ -1490,7 +1490,7 @@ static bool sfp_los_event_inactive(struct sfp *sfp, unsigned int event)
 
 static void sfp_sm_fault(struct sfp *sfp, unsigned int next_state, bool warn)
 {
-	if (sfp->sm_retries && !--sfp->sm_retries) {
+	if (sfp->sm_fault_retries && !--sfp->sm_fault_retries) {
 		dev_err(sfp->dev,
 			"module persistently indicates fault, disabling\n");
 		sfp_sm_next(sfp, SFP_S_TX_DISABLE, 0);
@@ -1893,7 +1893,7 @@ static void sfp_sm_main(struct sfp *sfp, unsigned int event)
 		sfp_module_tx_enable(sfp);
 
 		/* Initialise the fault clearance retries */
-		sfp->sm_retries = N_FAULT_INIT;
+		sfp->sm_fault_retries = N_FAULT_INIT;
 
 		/* We need to check the TX_FAULT state, which is not defined
 		 * while TX_DISABLE is asserted. The earliest we want to do
@@ -1933,7 +1933,7 @@ static void sfp_sm_main(struct sfp *sfp, unsigned int event)
 			 * or t_start_up, so assume there is a fault.
 			 */
 			sfp_sm_fault(sfp, SFP_S_INIT_TX_FAULT,
-				     sfp->sm_retries == N_FAULT_INIT);
+				     sfp->sm_fault_retries == N_FAULT_INIT);
 		} else if (event == SFP_E_TIMEOUT || event == SFP_E_TX_CLEAR) {
 	init_done:	/* TX_FAULT deasserted or we timed out with TX_FAULT
 			 * clear.  Probe for the PHY and check the LOS state.
@@ -1946,7 +1946,7 @@ static void sfp_sm_main(struct sfp *sfp, unsigned int event)
 			sfp_sm_link_check_los(sfp);
 
 			/* Reset the fault retry count */
-			sfp->sm_retries = N_FAULT;
+			sfp->sm_fault_retries = N_FAULT;
 		}
 		break;
 
-- 
cgit v1.2.3-59-g8ed1b


From 256e43cb8c696e0754514686d92f4e27b4985ad5 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Mon, 9 Dec 2019 14:16:05 +0000
Subject: net: sfp: error handling for phy probe

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/sfp.c | 26 +++++++++++++++++---------
 1 file changed, 17 insertions(+), 9 deletions(-)

diff --git a/drivers/net/phy/sfp.c b/drivers/net/phy/sfp.c
index a67f089f2106..76fa95e54542 100644
--- a/drivers/net/phy/sfp.c
+++ b/drivers/net/phy/sfp.c
@@ -1410,7 +1410,7 @@ static void sfp_sm_phy_detach(struct sfp *sfp)
 	sfp->mod_phy = NULL;
 }
 
-static void sfp_sm_probe_phy(struct sfp *sfp, bool is_c45)
+static int sfp_sm_probe_phy(struct sfp *sfp, bool is_c45)
 {
 	struct phy_device *phy;
 	int err;
@@ -1418,18 +1418,18 @@ static void sfp_sm_probe_phy(struct sfp *sfp, bool is_c45)
 	phy = get_phy_device(sfp->i2c_mii, SFP_PHY_ADDR, is_c45);
 	if (phy == ERR_PTR(-ENODEV)) {
 		dev_info(sfp->dev, "no PHY detected\n");
-		return;
+		return 0;
 	}
 	if (IS_ERR(phy)) {
 		dev_err(sfp->dev, "mdiobus scan returned %ld\n", PTR_ERR(phy));
-		return;
+		return PTR_ERR(phy);
 	}
 
 	err = phy_device_register(phy);
 	if (err) {
 		phy_device_free(phy);
 		dev_err(sfp->dev, "phy_device_register failed: %d\n", err);
-		return;
+		return err;
 	}
 
 	err = sfp_add_phy(sfp->sfp_bus, phy);
@@ -1437,10 +1437,12 @@ static void sfp_sm_probe_phy(struct sfp *sfp, bool is_c45)
 		phy_device_remove(phy);
 		phy_device_free(phy);
 		dev_err(sfp->dev, "sfp_add_phy failed: %d\n", err);
-		return;
+		return err;
 	}
 
 	sfp->mod_phy = phy;
+
+	return 0;
 }
 
 static void sfp_sm_link_up(struct sfp *sfp)
@@ -1513,21 +1515,24 @@ static void sfp_sm_fault(struct sfp *sfp, unsigned int next_state, bool warn)
  * Clause 45 copper SFP+ modules (10G) appear to switch their interface
  * mode according to the negotiated line speed.
  */
-static void sfp_sm_probe_for_phy(struct sfp *sfp)
+static int sfp_sm_probe_for_phy(struct sfp *sfp)
 {
+	int err = 0;
+
 	switch (sfp->id.base.extended_cc) {
 	case SFF8024_ECC_10GBASE_T_SFI:
 	case SFF8024_ECC_10GBASE_T_SR:
 	case SFF8024_ECC_5GBASE_T:
 	case SFF8024_ECC_2_5GBASE_T:
-		sfp_sm_probe_phy(sfp, true);
+		err = sfp_sm_probe_phy(sfp, true);
 		break;
 
 	default:
 		if (sfp->id.base.e1000_base_t)
-			sfp_sm_probe_phy(sfp, false);
+			err = sfp_sm_probe_phy(sfp, false);
 		break;
 	}
+	return err;
 }
 
 static int sfp_module_parse_power(struct sfp *sfp)
@@ -1938,7 +1943,10 @@ static void sfp_sm_main(struct sfp *sfp, unsigned int event)
 	init_done:	/* TX_FAULT deasserted or we timed out with TX_FAULT
 			 * clear.  Probe for the PHY and check the LOS state.
 			 */
-			sfp_sm_probe_for_phy(sfp);
+			if (sfp_sm_probe_for_phy(sfp)) {
+				sfp_sm_next(sfp, SFP_S_FAIL, 0);
+				break;
+			}
 			if (sfp_module_start(sfp->sfp_bus)) {
 				sfp_sm_next(sfp, SFP_S_FAIL, 0);
 				break;
-- 
cgit v1.2.3-59-g8ed1b


From 1cb89a14c80a87abe1f5bd47b1e7b3171b968f8c Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Mon, 9 Dec 2019 14:16:11 +0000
Subject: net: sfp: re-attempt probing for phy

Some 1000BASE-T PHY modules take a while for the PHY to wake up.
Retry the probe a number of times before deciding that the module has
no PHY.

Tested with:
 Sourcephotonics SPGBTXCNFC - PHY takes less than 50ms to respond.
 Champion One 1000SFPT - PHY takes about 200ms to respond.
 Mikrotik S-RJ01 - no PHY

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/sfp.c | 59 ++++++++++++++++++++++++++++++++++++---------------
 1 file changed, 42 insertions(+), 17 deletions(-)

diff --git a/drivers/net/phy/sfp.c b/drivers/net/phy/sfp.c
index 76fa95e54542..e54aef921038 100644
--- a/drivers/net/phy/sfp.c
+++ b/drivers/net/phy/sfp.c
@@ -62,6 +62,7 @@ enum {
 	SFP_S_FAIL,
 	SFP_S_WAIT,
 	SFP_S_INIT,
+	SFP_S_INIT_PHY,
 	SFP_S_INIT_TX_FAULT,
 	SFP_S_WAIT_LOS,
 	SFP_S_LINK_UP,
@@ -126,6 +127,7 @@ static const char * const sm_state_strings[] = {
 	[SFP_S_FAIL] = "fail",
 	[SFP_S_WAIT] = "wait",
 	[SFP_S_INIT] = "init",
+	[SFP_S_INIT_PHY] = "init_phy",
 	[SFP_S_INIT_TX_FAULT] = "init_tx_fault",
 	[SFP_S_WAIT_LOS] = "wait_los",
 	[SFP_S_LINK_UP] = "link_up",
@@ -180,6 +182,12 @@ static const enum gpiod_flags gpio_flags[] = {
 #define N_FAULT_INIT		5
 #define N_FAULT			5
 
+/* T_PHY_RETRY is the time interval between attempts to probe the PHY.
+ * R_PHY_RETRY is the number of attempts.
+ */
+#define T_PHY_RETRY		msecs_to_jiffies(50)
+#define R_PHY_RETRY		12
+
 /* SFP module presence detection is poor: the three MOD DEF signals are
  * the same length on the PCB, which means it's possible for MOD DEF 0 to
  * connect before the I2C bus on MOD DEF 1/2.
@@ -235,6 +243,7 @@ struct sfp {
 	unsigned char sm_dev_state;
 	unsigned short sm_state;
 	unsigned char sm_fault_retries;
+	unsigned char sm_phy_retries;
 
 	struct sfp_eeprom_id id;
 	unsigned int module_power_mW;
@@ -1416,10 +1425,8 @@ static int sfp_sm_probe_phy(struct sfp *sfp, bool is_c45)
 	int err;
 
 	phy = get_phy_device(sfp->i2c_mii, SFP_PHY_ADDR, is_c45);
-	if (phy == ERR_PTR(-ENODEV)) {
-		dev_info(sfp->dev, "no PHY detected\n");
-		return 0;
-	}
+	if (phy == ERR_PTR(-ENODEV))
+		return PTR_ERR(phy);
 	if (IS_ERR(phy)) {
 		dev_err(sfp->dev, "mdiobus scan returned %ld\n", PTR_ERR(phy));
 		return PTR_ERR(phy);
@@ -1867,6 +1874,7 @@ static void sfp_sm_module(struct sfp *sfp, unsigned int event)
 static void sfp_sm_main(struct sfp *sfp, unsigned int event)
 {
 	unsigned long timeout;
+	int ret;
 
 	/* Some events are global */
 	if (sfp->sm_state != SFP_S_DOWN &&
@@ -1940,22 +1948,39 @@ static void sfp_sm_main(struct sfp *sfp, unsigned int event)
 			sfp_sm_fault(sfp, SFP_S_INIT_TX_FAULT,
 				     sfp->sm_fault_retries == N_FAULT_INIT);
 		} else if (event == SFP_E_TIMEOUT || event == SFP_E_TX_CLEAR) {
-	init_done:	/* TX_FAULT deasserted or we timed out with TX_FAULT
-			 * clear.  Probe for the PHY and check the LOS state.
-			 */
-			if (sfp_sm_probe_for_phy(sfp)) {
-				sfp_sm_next(sfp, SFP_S_FAIL, 0);
-				break;
-			}
-			if (sfp_module_start(sfp->sfp_bus)) {
-				sfp_sm_next(sfp, SFP_S_FAIL, 0);
+	init_done:
+			sfp->sm_phy_retries = R_PHY_RETRY;
+			goto phy_probe;
+		}
+		break;
+
+	case SFP_S_INIT_PHY:
+		if (event != SFP_E_TIMEOUT)
+			break;
+	phy_probe:
+		/* TX_FAULT deasserted or we timed out with TX_FAULT
+		 * clear.  Probe for the PHY and check the LOS state.
+		 */
+		ret = sfp_sm_probe_for_phy(sfp);
+		if (ret == -ENODEV) {
+			if (--sfp->sm_phy_retries) {
+				sfp_sm_next(sfp, SFP_S_INIT_PHY, T_PHY_RETRY);
 				break;
+			} else {
+				dev_info(sfp->dev, "no PHY detected\n");
 			}
-			sfp_sm_link_check_los(sfp);
-
-			/* Reset the fault retry count */
-			sfp->sm_fault_retries = N_FAULT;
+		} else if (ret) {
+			sfp_sm_next(sfp, SFP_S_FAIL, 0);
+			break;
 		}
+		if (sfp_module_start(sfp->sfp_bus)) {
+			sfp_sm_next(sfp, SFP_S_FAIL, 0);
+			break;
+		}
+		sfp_sm_link_check_los(sfp);
+
+		/* Reset the fault retry count */
+		sfp->sm_fault_retries = N_FAULT;
 		break;
 
 	case SFP_S_INIT_TX_FAULT:
-- 
cgit v1.2.3-59-g8ed1b


From c5144fcbf25015b850e97a06a24b1ddfbc5579e8 Mon Sep 17 00:00:00 2001
From: Stefano Garzarella <sgarzare@redhat.com>
Date: Tue, 10 Dec 2019 11:43:02 +0100
Subject: vsock/virtio_transport_common: remove unused virtio header includes

We can remove virtio header includes, because virtio_transport_common
doesn't use virtio API, but provides common functions to interface
virtio/vhost transports with the af_vsock core, and to handle
the protocol.

Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/vmw_vsock/virtio_transport_common.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c
index e5ea29c6bca7..0e20b0f6eb65 100644
--- a/net/vmw_vsock/virtio_transport_common.c
+++ b/net/vmw_vsock/virtio_transport_common.c
@@ -11,9 +11,6 @@
 #include <linux/sched/signal.h>
 #include <linux/ctype.h>
 #include <linux/list.h>
-#include <linux/virtio.h>
-#include <linux/virtio_ids.h>
-#include <linux/virtio_config.h>
 #include <linux/virtio_vsock.h>
 #include <uapi/linux/vsockmon.h>
 
-- 
cgit v1.2.3-59-g8ed1b


From ef343b35d46667668a099655fca4a5b2e43a5dfe Mon Sep 17 00:00:00 2001
From: Stefano Garzarella <sgarzare@redhat.com>
Date: Tue, 10 Dec 2019 11:43:03 +0100
Subject: vsock: add VMADDR_CID_LOCAL definition

The VMADDR_CID_RESERVED (1) was used by VMCI, but now it is not
used anymore, so we can reuse it for local communication
(loopback) adding the new well-know CID: VMADDR_CID_LOCAL.

Cc: Jorgen Hansen <jhansen@vmware.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Jorgen Hansen <jhansen@vmware.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/uapi/linux/vm_sockets.h | 8 +++++---
 net/vmw_vsock/vmci_transport.c  | 2 +-
 2 files changed, 6 insertions(+), 4 deletions(-)

diff --git a/include/uapi/linux/vm_sockets.h b/include/uapi/linux/vm_sockets.h
index 68d57c5e99bc..fd0ed7221645 100644
--- a/include/uapi/linux/vm_sockets.h
+++ b/include/uapi/linux/vm_sockets.h
@@ -99,11 +99,13 @@
 
 #define VMADDR_CID_HYPERVISOR 0
 
-/* This CID is specific to VMCI and can be considered reserved (even VMCI
- * doesn't use it anymore, it's a legacy value from an older release).
+/* Use this as the destination CID in an address when referring to the
+ * local communication (loopback).
+ * (This was VMADDR_CID_RESERVED, but even VMCI doesn't use it anymore,
+ * it was a legacy value from an older release).
  */
 
-#define VMADDR_CID_RESERVED 1
+#define VMADDR_CID_LOCAL 1
 
 /* Use this as the destination CID in an address when referring to the host
  * (any process other than the hypervisor).  VMCI relies on it being 2, but
diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c
index 644d32e43d23..4b8b1150a738 100644
--- a/net/vmw_vsock/vmci_transport.c
+++ b/net/vmw_vsock/vmci_transport.c
@@ -648,7 +648,7 @@ static int vmci_transport_recv_dgram_cb(void *data, struct vmci_datagram *dg)
 static bool vmci_transport_stream_allow(u32 cid, u32 port)
 {
 	static const u32 non_socket_contexts[] = {
-		VMADDR_CID_RESERVED,
+		VMADDR_CID_LOCAL,
 	};
 	int i;
 
-- 
cgit v1.2.3-59-g8ed1b


From 0e12190578d081d4fe54d41635ec6e5a6eb0d01a Mon Sep 17 00:00:00 2001
From: Stefano Garzarella <sgarzare@redhat.com>
Date: Tue, 10 Dec 2019 11:43:04 +0100
Subject: vsock: add local transport support in the vsock core

This patch allows to register a transport able to handle
local communication (loopback).

Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/af_vsock.h   |  2 ++
 net/vmw_vsock/af_vsock.c | 17 ++++++++++++++++-
 2 files changed, 18 insertions(+), 1 deletion(-)

diff --git a/include/net/af_vsock.h b/include/net/af_vsock.h
index 4206dc6d813f..b1c717286993 100644
--- a/include/net/af_vsock.h
+++ b/include/net/af_vsock.h
@@ -98,6 +98,8 @@ struct vsock_transport_send_notify_data {
 #define VSOCK_TRANSPORT_F_G2H		0x00000002
 /* Transport provides DGRAM communication */
 #define VSOCK_TRANSPORT_F_DGRAM		0x00000004
+/* Transport provides local (loopback) communication */
+#define VSOCK_TRANSPORT_F_LOCAL		0x00000008
 
 struct vsock_transport {
 	struct module *module;
diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c
index 74db4cd637a7..3da0749a0c97 100644
--- a/net/vmw_vsock/af_vsock.c
+++ b/net/vmw_vsock/af_vsock.c
@@ -136,6 +136,8 @@ static const struct vsock_transport *transport_h2g;
 static const struct vsock_transport *transport_g2h;
 /* Transport used for DGRAM communication */
 static const struct vsock_transport *transport_dgram;
+/* Transport used for local communication */
+static const struct vsock_transport *transport_local;
 static DEFINE_MUTEX(vsock_register_mutex);
 
 /**** UTILS ****/
@@ -2137,7 +2139,7 @@ EXPORT_SYMBOL_GPL(vsock_core_get_transport);
 
 int vsock_core_register(const struct vsock_transport *t, int features)
 {
-	const struct vsock_transport *t_h2g, *t_g2h, *t_dgram;
+	const struct vsock_transport *t_h2g, *t_g2h, *t_dgram, *t_local;
 	int err = mutex_lock_interruptible(&vsock_register_mutex);
 
 	if (err)
@@ -2146,6 +2148,7 @@ int vsock_core_register(const struct vsock_transport *t, int features)
 	t_h2g = transport_h2g;
 	t_g2h = transport_g2h;
 	t_dgram = transport_dgram;
+	t_local = transport_local;
 
 	if (features & VSOCK_TRANSPORT_F_H2G) {
 		if (t_h2g) {
@@ -2171,9 +2174,18 @@ int vsock_core_register(const struct vsock_transport *t, int features)
 		t_dgram = t;
 	}
 
+	if (features & VSOCK_TRANSPORT_F_LOCAL) {
+		if (t_local) {
+			err = -EBUSY;
+			goto err_busy;
+		}
+		t_local = t;
+	}
+
 	transport_h2g = t_h2g;
 	transport_g2h = t_g2h;
 	transport_dgram = t_dgram;
+	transport_local = t_local;
 
 err_busy:
 	mutex_unlock(&vsock_register_mutex);
@@ -2194,6 +2206,9 @@ void vsock_core_unregister(const struct vsock_transport *t)
 	if (transport_dgram == t)
 		transport_dgram = NULL;
 
+	if (transport_local == t)
+		transport_local = NULL;
+
 	mutex_unlock(&vsock_register_mutex);
 }
 EXPORT_SYMBOL_GPL(vsock_core_unregister);
-- 
cgit v1.2.3-59-g8ed1b


From 077263fba10011d7e1d17dc7691ff20c15c56081 Mon Sep 17 00:00:00 2001
From: Stefano Garzarella <sgarzare@redhat.com>
Date: Tue, 10 Dec 2019 11:43:05 +0100
Subject: vsock: add vsock_loopback transport

This patch adds a new vsock_loopback transport to handle local
communication.
This transport is based on the loopback implementation of
virtio_transport, so it uses the virtio_transport_common APIs
to interface with the vsock core.

Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 MAINTAINERS                    |   1 +
 net/vmw_vsock/Kconfig          |  12 +++
 net/vmw_vsock/Makefile         |   1 +
 net/vmw_vsock/vsock_loopback.c | 180 +++++++++++++++++++++++++++++++++++++++++
 4 files changed, 194 insertions(+)
 create mode 100644 net/vmw_vsock/vsock_loopback.c

diff --git a/MAINTAINERS b/MAINTAINERS
index ce7fd2e7aa1a..a28c77ee6b0d 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -17480,6 +17480,7 @@ F:	net/vmw_vsock/diag.c
 F:	net/vmw_vsock/af_vsock_tap.c
 F:	net/vmw_vsock/virtio_transport_common.c
 F:	net/vmw_vsock/virtio_transport.c
+F:	net/vmw_vsock/vsock_loopback.c
 F:	drivers/net/vsockmon.c
 F:	drivers/vhost/vsock.c
 F:	tools/testing/vsock/
diff --git a/net/vmw_vsock/Kconfig b/net/vmw_vsock/Kconfig
index 8abcb815af2d..56356d2980c8 100644
--- a/net/vmw_vsock/Kconfig
+++ b/net/vmw_vsock/Kconfig
@@ -26,6 +26,18 @@ config VSOCKETS_DIAG
 
 	  Enable this module so userspace applications can query open sockets.
 
+config VSOCKETS_LOOPBACK
+	tristate "Virtual Sockets loopback transport"
+	depends on VSOCKETS
+	default y
+	select VIRTIO_VSOCKETS_COMMON
+	help
+	  This module implements a loopback transport for Virtual Sockets,
+	  using vmw_vsock_virtio_transport_common.
+
+	  To compile this driver as a module, choose M here: the module
+	  will be called vsock_loopback. If unsure, say N.
+
 config VMWARE_VMCI_VSOCKETS
 	tristate "VMware VMCI transport for Virtual Sockets"
 	depends on VSOCKETS && VMWARE_VMCI
diff --git a/net/vmw_vsock/Makefile b/net/vmw_vsock/Makefile
index 7c6f9a0b67b0..6a943ec95c4a 100644
--- a/net/vmw_vsock/Makefile
+++ b/net/vmw_vsock/Makefile
@@ -5,6 +5,7 @@ obj-$(CONFIG_VMWARE_VMCI_VSOCKETS) += vmw_vsock_vmci_transport.o
 obj-$(CONFIG_VIRTIO_VSOCKETS) += vmw_vsock_virtio_transport.o
 obj-$(CONFIG_VIRTIO_VSOCKETS_COMMON) += vmw_vsock_virtio_transport_common.o
 obj-$(CONFIG_HYPERV_VSOCKETS) += hv_sock.o
+obj-$(CONFIG_VSOCKETS_LOOPBACK) += vsock_loopback.o
 
 vsock-y += af_vsock.o af_vsock_tap.o vsock_addr.o
 
diff --git a/net/vmw_vsock/vsock_loopback.c b/net/vmw_vsock/vsock_loopback.c
new file mode 100644
index 000000000000..a45f7ffca8c5
--- /dev/null
+++ b/net/vmw_vsock/vsock_loopback.c
@@ -0,0 +1,180 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* loopback transport for vsock using virtio_transport_common APIs
+ *
+ * Copyright (C) 2013-2019 Red Hat, Inc.
+ * Authors: Asias He <asias@redhat.com>
+ *          Stefan Hajnoczi <stefanha@redhat.com>
+ *          Stefano Garzarella <sgarzare@redhat.com>
+ *
+ */
+#include <linux/spinlock.h>
+#include <linux/module.h>
+#include <linux/list.h>
+#include <linux/virtio_vsock.h>
+
+struct vsock_loopback {
+	struct workqueue_struct *workqueue;
+
+	spinlock_t pkt_list_lock; /* protects pkt_list */
+	struct list_head pkt_list;
+	struct work_struct pkt_work;
+};
+
+static struct vsock_loopback the_vsock_loopback;
+
+static u32 vsock_loopback_get_local_cid(void)
+{
+	return VMADDR_CID_LOCAL;
+}
+
+static int vsock_loopback_send_pkt(struct virtio_vsock_pkt *pkt)
+{
+	struct vsock_loopback *vsock = &the_vsock_loopback;
+	int len = pkt->len;
+
+	spin_lock_bh(&vsock->pkt_list_lock);
+	list_add_tail(&pkt->list, &vsock->pkt_list);
+	spin_unlock_bh(&vsock->pkt_list_lock);
+
+	queue_work(vsock->workqueue, &vsock->pkt_work);
+
+	return len;
+}
+
+static int vsock_loopback_cancel_pkt(struct vsock_sock *vsk)
+{
+	struct vsock_loopback *vsock = &the_vsock_loopback;
+	struct virtio_vsock_pkt *pkt, *n;
+	LIST_HEAD(freeme);
+
+	spin_lock_bh(&vsock->pkt_list_lock);
+	list_for_each_entry_safe(pkt, n, &vsock->pkt_list, list) {
+		if (pkt->vsk != vsk)
+			continue;
+		list_move(&pkt->list, &freeme);
+	}
+	spin_unlock_bh(&vsock->pkt_list_lock);
+
+	list_for_each_entry_safe(pkt, n, &freeme, list) {
+		list_del(&pkt->list);
+		virtio_transport_free_pkt(pkt);
+	}
+
+	return 0;
+}
+
+static struct virtio_transport loopback_transport = {
+	.transport = {
+		.module                   = THIS_MODULE,
+
+		.get_local_cid            = vsock_loopback_get_local_cid,
+
+		.init                     = virtio_transport_do_socket_init,
+		.destruct                 = virtio_transport_destruct,
+		.release                  = virtio_transport_release,
+		.connect                  = virtio_transport_connect,
+		.shutdown                 = virtio_transport_shutdown,
+		.cancel_pkt               = vsock_loopback_cancel_pkt,
+
+		.dgram_bind               = virtio_transport_dgram_bind,
+		.dgram_dequeue            = virtio_transport_dgram_dequeue,
+		.dgram_enqueue            = virtio_transport_dgram_enqueue,
+		.dgram_allow              = virtio_transport_dgram_allow,
+
+		.stream_dequeue           = virtio_transport_stream_dequeue,
+		.stream_enqueue           = virtio_transport_stream_enqueue,
+		.stream_has_data          = virtio_transport_stream_has_data,
+		.stream_has_space         = virtio_transport_stream_has_space,
+		.stream_rcvhiwat          = virtio_transport_stream_rcvhiwat,
+		.stream_is_active         = virtio_transport_stream_is_active,
+		.stream_allow             = virtio_transport_stream_allow,
+
+		.notify_poll_in           = virtio_transport_notify_poll_in,
+		.notify_poll_out          = virtio_transport_notify_poll_out,
+		.notify_recv_init         = virtio_transport_notify_recv_init,
+		.notify_recv_pre_block    = virtio_transport_notify_recv_pre_block,
+		.notify_recv_pre_dequeue  = virtio_transport_notify_recv_pre_dequeue,
+		.notify_recv_post_dequeue = virtio_transport_notify_recv_post_dequeue,
+		.notify_send_init         = virtio_transport_notify_send_init,
+		.notify_send_pre_block    = virtio_transport_notify_send_pre_block,
+		.notify_send_pre_enqueue  = virtio_transport_notify_send_pre_enqueue,
+		.notify_send_post_enqueue = virtio_transport_notify_send_post_enqueue,
+		.notify_buffer_size       = virtio_transport_notify_buffer_size,
+	},
+
+	.send_pkt = vsock_loopback_send_pkt,
+};
+
+static void vsock_loopback_work(struct work_struct *work)
+{
+	struct vsock_loopback *vsock =
+		container_of(work, struct vsock_loopback, pkt_work);
+	LIST_HEAD(pkts);
+
+	spin_lock_bh(&vsock->pkt_list_lock);
+	list_splice_init(&vsock->pkt_list, &pkts);
+	spin_unlock_bh(&vsock->pkt_list_lock);
+
+	while (!list_empty(&pkts)) {
+		struct virtio_vsock_pkt *pkt;
+
+		pkt = list_first_entry(&pkts, struct virtio_vsock_pkt, list);
+		list_del_init(&pkt->list);
+
+		virtio_transport_deliver_tap_pkt(pkt);
+		virtio_transport_recv_pkt(&loopback_transport, pkt);
+	}
+}
+
+static int __init vsock_loopback_init(void)
+{
+	struct vsock_loopback *vsock = &the_vsock_loopback;
+	int ret;
+
+	vsock->workqueue = alloc_workqueue("vsock-loopback", 0, 0);
+	if (!vsock->workqueue)
+		return -ENOMEM;
+
+	spin_lock_init(&vsock->pkt_list_lock);
+	INIT_LIST_HEAD(&vsock->pkt_list);
+	INIT_WORK(&vsock->pkt_work, vsock_loopback_work);
+
+	ret = vsock_core_register(&loopback_transport.transport,
+				  VSOCK_TRANSPORT_F_LOCAL);
+	if (ret)
+		goto out_wq;
+
+	return 0;
+
+out_wq:
+	destroy_workqueue(vsock->workqueue);
+	return ret;
+}
+
+static void __exit vsock_loopback_exit(void)
+{
+	struct vsock_loopback *vsock = &the_vsock_loopback;
+	struct virtio_vsock_pkt *pkt;
+
+	vsock_core_unregister(&loopback_transport.transport);
+
+	flush_work(&vsock->pkt_work);
+
+	spin_lock_bh(&vsock->pkt_list_lock);
+	while (!list_empty(&vsock->pkt_list)) {
+		pkt = list_first_entry(&vsock->pkt_list,
+				       struct virtio_vsock_pkt, list);
+		list_del(&pkt->list);
+		virtio_transport_free_pkt(pkt);
+	}
+	spin_unlock_bh(&vsock->pkt_list_lock);
+
+	destroy_workqueue(vsock->workqueue);
+}
+
+module_init(vsock_loopback_init);
+module_exit(vsock_loopback_exit);
+MODULE_LICENSE("GPL v2");
+MODULE_AUTHOR("Stefano Garzarella <sgarzare@redhat.com>");
+MODULE_DESCRIPTION("loopback transport for vsock");
+MODULE_ALIAS_NETPROTO(PF_VSOCK);
-- 
cgit v1.2.3-59-g8ed1b


From 408624af4c89989117bb2c6517bd50b7708a2fcd Mon Sep 17 00:00:00 2001
From: Stefano Garzarella <sgarzare@redhat.com>
Date: Tue, 10 Dec 2019 11:43:06 +0100
Subject: vsock: use local transport when it is loaded

Now that we have a transport that can handle the local communication,
we can use it when it is loaded.

A socket will use the local transport (loopback) when the remote
CID is:
- equal to VMADDR_CID_LOCAL
- or equal to transport_g2h->get_local_cid(), if transport_g2h
  is loaded (this allows us to keep the same behavior implemented
  by virtio and vmci transports)
- or equal to VMADDR_CID_HOST, if transport_g2h is not loaded

Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/vmw_vsock/af_vsock.c | 28 +++++++++++++++++++++++-----
 1 file changed, 23 insertions(+), 5 deletions(-)

diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c
index 3da0749a0c97..9c5b2a91baad 100644
--- a/net/vmw_vsock/af_vsock.c
+++ b/net/vmw_vsock/af_vsock.c
@@ -388,6 +388,21 @@ void vsock_enqueue_accept(struct sock *listener, struct sock *connected)
 }
 EXPORT_SYMBOL_GPL(vsock_enqueue_accept);
 
+static bool vsock_use_local_transport(unsigned int remote_cid)
+{
+	if (!transport_local)
+		return false;
+
+	if (remote_cid == VMADDR_CID_LOCAL)
+		return true;
+
+	if (transport_g2h) {
+		return remote_cid == transport_g2h->get_local_cid();
+	} else {
+		return remote_cid == VMADDR_CID_HOST;
+	}
+}
+
 static void vsock_deassign_transport(struct vsock_sock *vsk)
 {
 	if (!vsk->transport)
@@ -404,9 +419,9 @@ static void vsock_deassign_transport(struct vsock_sock *vsk)
  * (e.g. during the connect() or when a connection request on a listener
  * socket is received).
  * The vsk->remote_addr is used to decide which transport to use:
+ *  - remote CID == VMADDR_CID_LOCAL or g2h->local_cid or VMADDR_CID_HOST if
+ *    g2h is not loaded, will use local transport;
  *  - remote CID <= VMADDR_CID_HOST will use guest->host transport;
- *  - remote CID == local_cid (guest->host transport) will use guest->host
- *    transport for loopback (host->guest transports don't support loopback);
  *  - remote CID > VMADDR_CID_HOST will use host->guest transport;
  */
 int vsock_assign_transport(struct vsock_sock *vsk, struct vsock_sock *psk)
@@ -421,9 +436,9 @@ int vsock_assign_transport(struct vsock_sock *vsk, struct vsock_sock *psk)
 		new_transport = transport_dgram;
 		break;
 	case SOCK_STREAM:
-		if (remote_cid <= VMADDR_CID_HOST ||
-		    (transport_g2h &&
-		     remote_cid == transport_g2h->get_local_cid()))
+		if (vsock_use_local_transport(remote_cid))
+			new_transport = transport_local;
+		else if (remote_cid <= VMADDR_CID_HOST)
 			new_transport = transport_g2h;
 		else
 			new_transport = transport_h2g;
@@ -466,6 +481,9 @@ bool vsock_find_cid(unsigned int cid)
 	if (transport_h2g && cid == VMADDR_CID_HOST)
 		return true;
 
+	if (transport_local && cid == VMADDR_CID_LOCAL)
+		return true;
+
 	return false;
 }
 EXPORT_SYMBOL_GPL(vsock_find_cid);
-- 
cgit v1.2.3-59-g8ed1b


From bf5432b1de1fcfd1d389111c350b88dd238860ba Mon Sep 17 00:00:00 2001
From: Stefano Garzarella <sgarzare@redhat.com>
Date: Tue, 10 Dec 2019 11:43:07 +0100
Subject: vsock/virtio: remove loopback handling

We can remove the loopback handling from virtio_transport,
because now the vsock core is able to handle local communication
using the new vsock_loopback device.

Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/vmw_vsock/virtio_transport.c | 61 ++--------------------------------------
 1 file changed, 2 insertions(+), 59 deletions(-)

diff --git a/net/vmw_vsock/virtio_transport.c b/net/vmw_vsock/virtio_transport.c
index 1458c5c8b64d..dfbaf6bd8b1c 100644
--- a/net/vmw_vsock/virtio_transport.c
+++ b/net/vmw_vsock/virtio_transport.c
@@ -44,10 +44,6 @@ struct virtio_vsock {
 	spinlock_t send_pkt_list_lock;
 	struct list_head send_pkt_list;
 
-	struct work_struct loopback_work;
-	spinlock_t loopback_list_lock; /* protects loopback_list */
-	struct list_head loopback_list;
-
 	atomic_t queued_replies;
 
 	/* The following fields are protected by rx_lock.  vqs[VSOCK_VQ_RX]
@@ -86,20 +82,6 @@ out_rcu:
 	return ret;
 }
 
-static int virtio_transport_send_pkt_loopback(struct virtio_vsock *vsock,
-					      struct virtio_vsock_pkt *pkt)
-{
-	int len = pkt->len;
-
-	spin_lock_bh(&vsock->loopback_list_lock);
-	list_add_tail(&pkt->list, &vsock->loopback_list);
-	spin_unlock_bh(&vsock->loopback_list_lock);
-
-	queue_work(virtio_vsock_workqueue, &vsock->loopback_work);
-
-	return len;
-}
-
 static void
 virtio_transport_send_pkt_work(struct work_struct *work)
 {
@@ -194,7 +176,8 @@ virtio_transport_send_pkt(struct virtio_vsock_pkt *pkt)
 	}
 
 	if (le64_to_cpu(pkt->hdr.dst_cid) == vsock->guest_cid) {
-		len = virtio_transport_send_pkt_loopback(vsock, pkt);
+		virtio_transport_free_pkt(pkt);
+		len = -ENODEV;
 		goto out_rcu;
 	}
 
@@ -502,33 +485,6 @@ static struct virtio_transport virtio_transport = {
 	.send_pkt = virtio_transport_send_pkt,
 };
 
-static void virtio_transport_loopback_work(struct work_struct *work)
-{
-	struct virtio_vsock *vsock =
-		container_of(work, struct virtio_vsock, loopback_work);
-	LIST_HEAD(pkts);
-
-	spin_lock_bh(&vsock->loopback_list_lock);
-	list_splice_init(&vsock->loopback_list, &pkts);
-	spin_unlock_bh(&vsock->loopback_list_lock);
-
-	mutex_lock(&vsock->rx_lock);
-
-	if (!vsock->rx_run)
-		goto out;
-
-	while (!list_empty(&pkts)) {
-		struct virtio_vsock_pkt *pkt;
-
-		pkt = list_first_entry(&pkts, struct virtio_vsock_pkt, list);
-		list_del_init(&pkt->list);
-
-		virtio_transport_recv_pkt(&virtio_transport, pkt);
-	}
-out:
-	mutex_unlock(&vsock->rx_lock);
-}
-
 static void virtio_transport_rx_work(struct work_struct *work)
 {
 	struct virtio_vsock *vsock =
@@ -633,13 +589,10 @@ static int virtio_vsock_probe(struct virtio_device *vdev)
 	mutex_init(&vsock->event_lock);
 	spin_lock_init(&vsock->send_pkt_list_lock);
 	INIT_LIST_HEAD(&vsock->send_pkt_list);
-	spin_lock_init(&vsock->loopback_list_lock);
-	INIT_LIST_HEAD(&vsock->loopback_list);
 	INIT_WORK(&vsock->rx_work, virtio_transport_rx_work);
 	INIT_WORK(&vsock->tx_work, virtio_transport_tx_work);
 	INIT_WORK(&vsock->event_work, virtio_transport_event_work);
 	INIT_WORK(&vsock->send_pkt_work, virtio_transport_send_pkt_work);
-	INIT_WORK(&vsock->loopback_work, virtio_transport_loopback_work);
 
 	mutex_lock(&vsock->tx_lock);
 	vsock->tx_run = true;
@@ -720,22 +673,12 @@ static void virtio_vsock_remove(struct virtio_device *vdev)
 	}
 	spin_unlock_bh(&vsock->send_pkt_list_lock);
 
-	spin_lock_bh(&vsock->loopback_list_lock);
-	while (!list_empty(&vsock->loopback_list)) {
-		pkt = list_first_entry(&vsock->loopback_list,
-				       struct virtio_vsock_pkt, list);
-		list_del(&pkt->list);
-		virtio_transport_free_pkt(pkt);
-	}
-	spin_unlock_bh(&vsock->loopback_list_lock);
-
 	/* Delete virtqueues and flush outstanding callbacks if any */
 	vdev->config->del_vqs(vdev);
 
 	/* Other works can be queued before 'config->del_vqs()', so we flush
 	 * all works before to free the vsock object to avoid use after free.
 	 */
-	flush_work(&vsock->loopback_work);
 	flush_work(&vsock->rx_work);
 	flush_work(&vsock->tx_work);
 	flush_work(&vsock->event_work);
-- 
cgit v1.2.3-59-g8ed1b


From 81c22041d9f19df07b9cba95e3cd02e0f41bc1e1 Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Mon, 9 Dec 2019 16:08:03 +0100
Subject: bpf, x86, arm64: Enable jit by default when not built as always-on

After Spectre 2 fix via 290af86629b2 ("bpf: introduce BPF_JIT_ALWAYS_ON
config") most major distros use BPF_JIT_ALWAYS_ON configuration these days
which compiles out the BPF interpreter entirely and always enables the
JIT. Also given recent fix in e1608f3fa857 ("bpf: Avoid setting bpf insns
pages read-only when prog is jited"), we additionally avoid fragmenting
the direct map for the BPF insns pages sitting in the general data heap
since they are not used during execution. Latter is only needed when run
through the interpreter.

Since both x86 and arm64 JITs have seen a lot of exposure over the years,
are generally most up to date and maintained, there is more downside in
!BPF_JIT_ALWAYS_ON configurations to have the interpreter enabled by default
rather than the JIT. Add a ARCH_WANT_DEFAULT_BPF_JIT config which archs can
use to set the bpf_jit_{enable,kallsyms} to 1. Back in the days the
bpf_jit_kallsyms knob was set to 0 by default since major distros still
had /proc/kallsyms addresses exposed to unprivileged user space which is
not the case anymore. Hence both knobs are set via BPF_JIT_DEFAULT_ON which
is set to 'y' in case of BPF_JIT_ALWAYS_ON or ARCH_WANT_DEFAULT_BPF_JIT.

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Will Deacon <will@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/f78ad24795c2966efcc2ee19025fa3459f622185.1575903816.git.daniel@iogearbox.net
---
 arch/arm64/Kconfig | 1 +
 arch/x86/Kconfig   | 1 +
 init/Kconfig       | 7 +++++++
 kernel/bpf/core.c  | 4 ++--
 4 files changed, 11 insertions(+), 2 deletions(-)

diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index b1b4476ddb83..29d03459de20 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -69,6 +69,7 @@ config ARM64
 	select ARCH_SUPPORTS_INT128 if CC_HAS_INT128 && (GCC_VERSION >= 50000 || CC_IS_CLANG)
 	select ARCH_SUPPORTS_NUMA_BALANCING
 	select ARCH_WANT_COMPAT_IPC_PARSE_VERSION if COMPAT
+	select ARCH_WANT_DEFAULT_BPF_JIT
 	select ARCH_WANT_DEFAULT_TOPDOWN_MMAP_LAYOUT
 	select ARCH_WANT_FRAME_POINTERS
 	select ARCH_WANT_HUGE_PMD_SHARE if ARM64_4K_PAGES || (ARM64_16K_PAGES && !ARM64_VA_BITS_36)
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 5e8949953660..1f6a0388a65f 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -93,6 +93,7 @@ config X86
 	select ARCH_USE_QUEUED_RWLOCKS
 	select ARCH_USE_QUEUED_SPINLOCKS
 	select ARCH_WANT_BATCHED_UNMAP_TLB_FLUSH
+	select ARCH_WANT_DEFAULT_BPF_JIT	if X86_64
 	select ARCH_WANTS_DYNAMIC_TASK_STRUCT
 	select ARCH_WANT_HUGE_PMD_SHARE
 	select ARCH_WANTS_THP_SWAP		if X86_64
diff --git a/init/Kconfig b/init/Kconfig
index a34064a031a5..890aaa62efde 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -1604,6 +1604,9 @@ config BPF_SYSCALL
 	  Enable the bpf() system call that allows to manipulate eBPF
 	  programs and maps via file descriptors.
 
+config ARCH_WANT_DEFAULT_BPF_JIT
+	bool
+
 config BPF_JIT_ALWAYS_ON
 	bool "Permanently enable BPF JIT and remove BPF interpreter"
 	depends on BPF_SYSCALL && HAVE_EBPF_JIT && BPF_JIT
@@ -1611,6 +1614,10 @@ config BPF_JIT_ALWAYS_ON
 	  Enables BPF JIT and removes BPF interpreter to avoid
 	  speculative execution of BPF instructions by the interpreter
 
+config BPF_JIT_DEFAULT_ON
+	def_bool ARCH_WANT_DEFAULT_BPF_JIT || BPF_JIT_ALWAYS_ON
+	depends on HAVE_EBPF_JIT && BPF_JIT
+
 config USERFAULTFD
 	bool "Enable userfaultfd() system call"
 	depends on MMU
diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c
index 49e32acad7d8..2ff01a716128 100644
--- a/kernel/bpf/core.c
+++ b/kernel/bpf/core.c
@@ -520,9 +520,9 @@ void bpf_prog_kallsyms_del_all(struct bpf_prog *fp)
 
 #ifdef CONFIG_BPF_JIT
 /* All BPF JIT sysctl knobs here. */
-int bpf_jit_enable   __read_mostly = IS_BUILTIN(CONFIG_BPF_JIT_ALWAYS_ON);
+int bpf_jit_enable   __read_mostly = IS_BUILTIN(CONFIG_BPF_JIT_DEFAULT_ON);
+int bpf_jit_kallsyms __read_mostly = IS_BUILTIN(CONFIG_BPF_JIT_DEFAULT_ON);
 int bpf_jit_harden   __read_mostly;
-int bpf_jit_kallsyms __read_mostly;
 long bpf_jit_limit   __read_mostly;
 
 static __always_inline void
-- 
cgit v1.2.3-59-g8ed1b


From 679152d3a32e305c213f83160c328c37566ae8bc Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Thu, 12 Dec 2019 09:19:18 -0800
Subject: libbpf: Fix printf compilation warnings on ppc64le arch

On ppc64le __u64 and __s64 are defined as long int and unsigned long int,
respectively. This causes compiler to emit warning when %lld/%llu are used to
printf 64-bit numbers. Fix this by casting to size_t/ssize_t with %zu and %zd
format specifiers, respectively.

v1->v2:
- use size_t/ssize_t instead of custom typedefs (Martin).

Fixes: 1f8e2bcb2cd5 ("libbpf: Refactor relocation handling")
Fixes: abd29c931459 ("libbpf: allow specifying map definitions using BTF")
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20191212171918.638010-1-andriin@fb.com
---
 tools/lib/bpf/libbpf.c | 37 +++++++++++++++++++------------------
 1 file changed, 19 insertions(+), 18 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 3f09772192f1..920d4e06a5f9 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -1242,15 +1242,15 @@ static int bpf_object__init_user_btf_map(struct bpf_object *obj,
 			}
 			sz = btf__resolve_size(obj->btf, t->type);
 			if (sz < 0) {
-				pr_warn("map '%s': can't determine key size for type [%u]: %lld.\n",
-					map_name, t->type, sz);
+				pr_warn("map '%s': can't determine key size for type [%u]: %zd.\n",
+					map_name, t->type, (ssize_t)sz);
 				return sz;
 			}
-			pr_debug("map '%s': found key [%u], sz = %lld.\n",
-				 map_name, t->type, sz);
+			pr_debug("map '%s': found key [%u], sz = %zd.\n",
+				 map_name, t->type, (ssize_t)sz);
 			if (map->def.key_size && map->def.key_size != sz) {
-				pr_warn("map '%s': conflicting key size %u != %lld.\n",
-					map_name, map->def.key_size, sz);
+				pr_warn("map '%s': conflicting key size %u != %zd.\n",
+					map_name, map->def.key_size, (ssize_t)sz);
 				return -EINVAL;
 			}
 			map->def.key_size = sz;
@@ -1285,15 +1285,15 @@ static int bpf_object__init_user_btf_map(struct bpf_object *obj,
 			}
 			sz = btf__resolve_size(obj->btf, t->type);
 			if (sz < 0) {
-				pr_warn("map '%s': can't determine value size for type [%u]: %lld.\n",
-					map_name, t->type, sz);
+				pr_warn("map '%s': can't determine value size for type [%u]: %zd.\n",
+					map_name, t->type, (ssize_t)sz);
 				return sz;
 			}
-			pr_debug("map '%s': found value [%u], sz = %lld.\n",
-				 map_name, t->type, sz);
+			pr_debug("map '%s': found value [%u], sz = %zd.\n",
+				 map_name, t->type, (ssize_t)sz);
 			if (map->def.value_size && map->def.value_size != sz) {
-				pr_warn("map '%s': conflicting value size %u != %lld.\n",
-					map_name, map->def.value_size, sz);
+				pr_warn("map '%s': conflicting value size %u != %zd.\n",
+					map_name, map->def.value_size, (ssize_t)sz);
 				return -EINVAL;
 			}
 			map->def.value_size = sz;
@@ -1817,7 +1817,8 @@ static int bpf_program__record_reloc(struct bpf_program *prog,
 			return -LIBBPF_ERRNO__RELOC;
 		}
 		if (sym->st_value % 8) {
-			pr_warn("bad call relo offset: %llu\n", (__u64)sym->st_value);
+			pr_warn("bad call relo offset: %zu\n",
+				(size_t)sym->st_value);
 			return -LIBBPF_ERRNO__RELOC;
 		}
 		reloc_desc->type = RELO_CALL;
@@ -1859,8 +1860,8 @@ static int bpf_program__record_reloc(struct bpf_program *prog,
 			break;
 		}
 		if (map_idx >= nr_maps) {
-			pr_warn("map relo failed to find map for sec %u, off %llu\n",
-				shdr_idx, (__u64)sym->st_value);
+			pr_warn("map relo failed to find map for sec %u, off %zu\n",
+				shdr_idx, (size_t)sym->st_value);
 			return -LIBBPF_ERRNO__RELOC;
 		}
 		reloc_desc->type = RELO_LD64;
@@ -1941,9 +1942,9 @@ bpf_program__collect_reloc(struct bpf_program *prog, GElf_Shdr *shdr,
 		name = elf_strptr(obj->efile.elf, obj->efile.strtabidx,
 				  sym.st_name) ? : "<?>";
 
-		pr_debug("relo for shdr %u, symb %llu, value %llu, type %d, bind %d, name %d (\'%s\'), insn %u\n",
-			 (__u32)sym.st_shndx, (__u64)GELF_R_SYM(rel.r_info),
-			 (__u64)sym.st_value, GELF_ST_TYPE(sym.st_info),
+		pr_debug("relo for shdr %u, symb %zu, value %zu, type %d, bind %d, name %d (\'%s\'), insn %u\n",
+			 (__u32)sym.st_shndx, (size_t)GELF_R_SYM(rel.r_info),
+			 (size_t)sym.st_value, GELF_ST_TYPE(sym.st_info),
 			 GELF_ST_BIND(sym.st_info), sym.st_name, name,
 			 insn_idx);
 
-- 
cgit v1.2.3-59-g8ed1b


From b4653342b1514cb11f25b727c689451aff02996d Mon Sep 17 00:00:00 2001
From: Kirill Tkhai <ktkhai@virtuozzo.com>
Date: Mon, 9 Dec 2019 13:03:40 +0300
Subject: net: Allow to show socket-specific information in
 /proc/[pid]/fdinfo/[fd]

This adds .show_fdinfo to socket_file_ops, so protocols will be able
to print their specific data in fdinfo.

Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/net.h |  1 +
 net/socket.c        | 12 ++++++++++++
 2 files changed, 13 insertions(+)

diff --git a/include/linux/net.h b/include/linux/net.h
index 9cafb5f353a9..6451425e828f 100644
--- a/include/linux/net.h
+++ b/include/linux/net.h
@@ -171,6 +171,7 @@ struct proto_ops {
 	int		(*compat_getsockopt)(struct socket *sock, int level,
 				      int optname, char __user *optval, int __user *optlen);
 #endif
+	void		(*show_fdinfo)(struct seq_file *m, struct socket *sock);
 	int		(*sendmsg)   (struct socket *sock, struct msghdr *m,
 				      size_t total_len);
 	/* Notes for implementing recvmsg:
diff --git a/net/socket.c b/net/socket.c
index 4d38d49d6ad9..532f123e0459 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -128,6 +128,7 @@ static ssize_t sock_sendpage(struct file *file, struct page *page,
 static ssize_t sock_splice_read(struct file *file, loff_t *ppos,
 				struct pipe_inode_info *pipe, size_t len,
 				unsigned int flags);
+static void sock_show_fdinfo(struct seq_file *m, struct file *f);
 
 /*
  *	Socket files have a set of 'special' operations as well as the generic file ones. These don't appear
@@ -150,6 +151,9 @@ static const struct file_operations socket_file_ops = {
 	.sendpage =	sock_sendpage,
 	.splice_write = generic_splice_sendpage,
 	.splice_read =	sock_splice_read,
+#ifdef CONFIG_PROC_FS
+	.show_fdinfo =	sock_show_fdinfo,
+#endif
 };
 
 /*
@@ -993,6 +997,14 @@ static ssize_t sock_write_iter(struct kiocb *iocb, struct iov_iter *from)
 	return res;
 }
 
+static void sock_show_fdinfo(struct seq_file *m, struct file *f)
+{
+	struct socket *sock = f->private_data;
+
+	if (sock->ops->show_fdinfo)
+		sock->ops->show_fdinfo(m, sock);
+}
+
 /*
  * Atomic setting of ioctl hooks to avoid race
  * with module unload.
-- 
cgit v1.2.3-59-g8ed1b


From 3c32da19a858fb1ae8a76bf899160be49f338506 Mon Sep 17 00:00:00 2001
From: Kirill Tkhai <ktkhai@virtuozzo.com>
Date: Mon, 9 Dec 2019 13:03:46 +0300
Subject: unix: Show number of pending scm files of receive queue in fdinfo

Unix sockets like a block box. You never know what is stored there:
there may be a file descriptor holding a mount or a block device,
or there may be whole universes with namespaces, sockets with receive
queues full of sockets etc.

The patch adds a little debug and accounts number of files (not recursive),
which is in receive queue of a unix socket. Sometimes this is useful
to determine, that socket should be investigated or which task should
be killed to put reference counter on a resourse.

v2: Pass correct argument to lockdep

Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/af_unix.h |  5 +++++
 net/unix/af_unix.c    | 56 ++++++++++++++++++++++++++++++++++++++++++++++-----
 2 files changed, 56 insertions(+), 5 deletions(-)

diff --git a/include/net/af_unix.h b/include/net/af_unix.h
index 3426d6dacc45..17e10fba2152 100644
--- a/include/net/af_unix.h
+++ b/include/net/af_unix.h
@@ -41,6 +41,10 @@ struct unix_skb_parms {
 	u32			consumed;
 } __randomize_layout;
 
+struct scm_stat {
+	u32 nr_fds;
+};
+
 #define UNIXCB(skb)	(*(struct unix_skb_parms *)&((skb)->cb))
 
 #define unix_state_lock(s)	spin_lock(&unix_sk(s)->lock)
@@ -65,6 +69,7 @@ struct unix_sock {
 #define UNIX_GC_MAYBE_CYCLE	1
 	struct socket_wq	peer_wq;
 	wait_queue_entry_t	peer_wake;
+	struct scm_stat		scm_stat;
 };
 
 static inline struct unix_sock *unix_sk(const struct sock *sk)
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index 7cfdce10de36..050eed3b30dc 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -676,6 +676,16 @@ static int unix_set_peek_off(struct sock *sk, int val)
 	return 0;
 }
 
+static void unix_show_fdinfo(struct seq_file *m, struct socket *sock)
+{
+	struct sock *sk = sock->sk;
+	struct unix_sock *u;
+
+	if (sk) {
+		u = unix_sk(sock->sk);
+		seq_printf(m, "scm_fds: %u\n", READ_ONCE(u->scm_stat.nr_fds));
+	}
+}
 
 static const struct proto_ops unix_stream_ops = {
 	.family =	PF_UNIX,
@@ -701,6 +711,7 @@ static const struct proto_ops unix_stream_ops = {
 	.sendpage =	unix_stream_sendpage,
 	.splice_read =	unix_stream_splice_read,
 	.set_peek_off =	unix_set_peek_off,
+	.show_fdinfo =	unix_show_fdinfo,
 };
 
 static const struct proto_ops unix_dgram_ops = {
@@ -726,6 +737,7 @@ static const struct proto_ops unix_dgram_ops = {
 	.mmap =		sock_no_mmap,
 	.sendpage =	sock_no_sendpage,
 	.set_peek_off =	unix_set_peek_off,
+	.show_fdinfo =	unix_show_fdinfo,
 };
 
 static const struct proto_ops unix_seqpacket_ops = {
@@ -751,6 +763,7 @@ static const struct proto_ops unix_seqpacket_ops = {
 	.mmap =		sock_no_mmap,
 	.sendpage =	sock_no_sendpage,
 	.set_peek_off =	unix_set_peek_off,
+	.show_fdinfo =	unix_show_fdinfo,
 };
 
 static struct proto unix_proto = {
@@ -788,6 +801,7 @@ static struct sock *unix_create1(struct net *net, struct socket *sock, int kern)
 	mutex_init(&u->bindlock); /* single task binding lock */
 	init_waitqueue_head(&u->peer_wait);
 	init_waitqueue_func_entry(&u->peer_wake, unix_dgram_peer_wake_relay);
+	memset(&u->scm_stat, 0, sizeof(struct scm_stat));
 	unix_insert_socket(unix_sockets_unbound(sk), sk);
 out:
 	if (sk == NULL)
@@ -1572,6 +1586,28 @@ static bool unix_skb_scm_eq(struct sk_buff *skb,
 	       unix_secdata_eq(scm, skb);
 }
 
+static void scm_stat_add(struct sock *sk, struct sk_buff *skb)
+{
+	struct scm_fp_list *fp = UNIXCB(skb).fp;
+	struct unix_sock *u = unix_sk(sk);
+
+	lockdep_assert_held(&sk->sk_receive_queue.lock);
+
+	if (unlikely(fp && fp->count))
+		u->scm_stat.nr_fds += fp->count;
+}
+
+static void scm_stat_del(struct sock *sk, struct sk_buff *skb)
+{
+	struct scm_fp_list *fp = UNIXCB(skb).fp;
+	struct unix_sock *u = unix_sk(sk);
+
+	lockdep_assert_held(&sk->sk_receive_queue.lock);
+
+	if (unlikely(fp && fp->count))
+		u->scm_stat.nr_fds -= fp->count;
+}
+
 /*
  *	Send AF_UNIX data.
  */
@@ -1757,7 +1793,10 @@ restart_locked:
 	if (sock_flag(other, SOCK_RCVTSTAMP))
 		__net_timestamp(skb);
 	maybe_add_creds(skb, sock, other);
-	skb_queue_tail(&other->sk_receive_queue, skb);
+	spin_lock(&other->sk_receive_queue.lock);
+	scm_stat_add(other, skb);
+	__skb_queue_tail(&other->sk_receive_queue, skb);
+	spin_unlock(&other->sk_receive_queue.lock);
 	unix_state_unlock(other);
 	other->sk_data_ready(other);
 	sock_put(other);
@@ -1859,7 +1898,10 @@ static int unix_stream_sendmsg(struct socket *sock, struct msghdr *msg,
 			goto pipe_err_free;
 
 		maybe_add_creds(skb, sock, other);
-		skb_queue_tail(&other->sk_receive_queue, skb);
+		spin_lock(&other->sk_receive_queue.lock);
+		scm_stat_add(other, skb);
+		__skb_queue_tail(&other->sk_receive_queue, skb);
+		spin_unlock(&other->sk_receive_queue.lock);
 		unix_state_unlock(other);
 		other->sk_data_ready(other);
 		sent += size;
@@ -2058,8 +2100,8 @@ static int unix_dgram_recvmsg(struct socket *sock, struct msghdr *msg,
 		mutex_lock(&u->iolock);
 
 		skip = sk_peek_offset(sk, flags);
-		skb = __skb_try_recv_datagram(sk, flags, NULL, &skip, &err,
-					      &last);
+		skb = __skb_try_recv_datagram(sk, flags, scm_stat_del,
+					      &skip, &err, &last);
 		if (skb)
 			break;
 
@@ -2353,8 +2395,12 @@ unlock:
 
 			sk_peek_offset_bwd(sk, chunk);
 
-			if (UNIXCB(skb).fp)
+			if (UNIXCB(skb).fp) {
+				spin_lock(&sk->sk_receive_queue.lock);
+				scm_stat_del(sk, skb);
+				spin_unlock(&sk->sk_receive_queue.lock);
 				unix_detach_fds(&scm, skb);
+			}
 
 			if (unix_skb_len(skb))
 				break;
-- 
cgit v1.2.3-59-g8ed1b


From f74877a5457d34d604dba6dbbb13c4c05bac8b93 Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Wed, 11 Dec 2019 10:58:14 +0100
Subject: rtnetlink: provide permanent hardware address in RTM_NEWLINK

Permanent hardware address of a network device was traditionally provided
via ethtool ioctl interface but as Jiri Pirko pointed out in a review of
ethtool netlink interface, rtnetlink is much more suitable for it so let's
add it to the RTM_NEWLINK message.

Add IFLA_PERM_ADDRESS attribute to RTM_NEWLINK messages unless the
permanent address is all zeros (i.e. device driver did not fill it). As
permanent address is not modifiable, reject userspace requests containing
IFLA_PERM_ADDRESS attribute.

Note: we already provide permanent hardware address for bond slaves;
unfortunately we cannot drop that attribute for backward compatibility
reasons.

v5 -> v6: only add the attribute if permanent address is not zero

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/uapi/linux/if_link.h | 1 +
 net/core/rtnetlink.c         | 5 +++++
 2 files changed, 6 insertions(+)

diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h
index 8aec8769d944..1d69f637c5d6 100644
--- a/include/uapi/linux/if_link.h
+++ b/include/uapi/linux/if_link.h
@@ -169,6 +169,7 @@ enum {
 	IFLA_MAX_MTU,
 	IFLA_PROP_LIST,
 	IFLA_ALT_IFNAME, /* Alternative ifname */
+	IFLA_PERM_ADDRESS,
 	__IFLA_MAX
 };
 
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 02916f43bf63..20bc406f3871 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -1041,6 +1041,7 @@ static noinline size_t if_nlmsg_size(const struct net_device *dev,
 	       + nla_total_size(4)  /* IFLA_MIN_MTU */
 	       + nla_total_size(4)  /* IFLA_MAX_MTU */
 	       + rtnl_prop_list_size(dev)
+	       + nla_total_size(MAX_ADDR_LEN) /* IFLA_PERM_ADDRESS */
 	       + 0;
 }
 
@@ -1757,6 +1758,9 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb,
 	    nla_put_s32(skb, IFLA_NEW_IFINDEX, new_ifindex) < 0)
 		goto nla_put_failure;
 
+	if (memchr_inv(dev->perm_addr, '\0', dev->addr_len) &&
+	    nla_put(skb, IFLA_PERM_ADDRESS, dev->addr_len, dev->perm_addr))
+		goto nla_put_failure;
 
 	rcu_read_lock();
 	if (rtnl_fill_link_af(skb, dev, ext_filter_mask))
@@ -1822,6 +1826,7 @@ static const struct nla_policy ifla_policy[IFLA_MAX+1] = {
 	[IFLA_PROP_LIST]	= { .type = NLA_NESTED },
 	[IFLA_ALT_IFNAME]	= { .type = NLA_STRING,
 				    .len = ALTIFNAMSIZ - 1 },
+	[IFLA_PERM_ADDRESS]	= { .type = NLA_REJECT },
 };
 
 static const struct nla_policy ifla_info_policy[IFLA_INFO_MAX+1] = {
-- 
cgit v1.2.3-59-g8ed1b


From 32d5109a9d864aea3981f0b5ea736eee4e11b42a Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Wed, 11 Dec 2019 10:58:19 +0100
Subject: netlink: rename nl80211_validate_nested() to nla_validate_nested()

Function nl80211_validate_nested() is not specific to nl80211, it's
a counterpart to nla_validate_nested_deprecated() with strict validation.
For consistency with other validation and parse functions, rename it to
nla_validate_nested().

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netlink.h  | 8 ++++----
 net/wireless/nl80211.c | 3 +--
 2 files changed, 5 insertions(+), 6 deletions(-)

diff --git a/include/net/netlink.h b/include/net/netlink.h
index b140c8f1be22..56c365dc6dc7 100644
--- a/include/net/netlink.h
+++ b/include/net/netlink.h
@@ -1735,7 +1735,7 @@ static inline void nla_nest_cancel(struct sk_buff *skb, struct nlattr *start)
 }
 
 /**
- * nla_validate_nested - Validate a stream of nested attributes
+ * __nla_validate_nested - Validate a stream of nested attributes
  * @start: container attribute
  * @maxtype: maximum attribute type to be expected
  * @policy: validation policy
@@ -1758,9 +1758,9 @@ static inline int __nla_validate_nested(const struct nlattr *start, int maxtype,
 }
 
 static inline int
-nl80211_validate_nested(const struct nlattr *start, int maxtype,
-			const struct nla_policy *policy,
-			struct netlink_ext_ack *extack)
+nla_validate_nested(const struct nlattr *start, int maxtype,
+		    const struct nla_policy *policy,
+		    struct netlink_ext_ack *extack)
 {
 	return __nla_validate_nested(start, maxtype, policy,
 				     NL_VALIDATE_STRICT, extack);
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index da5262b2298b..fa3526592c51 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -12900,8 +12900,7 @@ static int nl80211_vendor_check_policy(const struct wiphy_vendor_command *vcmd,
 		return -EINVAL;
 	}
 
-	return nl80211_validate_nested(attr, vcmd->maxattr, vcmd->policy,
-				       extack);
+	return nla_validate_nested(attr, vcmd->maxattr, vcmd->policy, extack);
 }
 
 static int nl80211_vendor_cmd(struct sk_buff *skb, struct genl_info *info)
-- 
cgit v1.2.3-59-g8ed1b


From 9ce48e5a09ea63a7867647af68caa0605d99757d Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Wed, 11 Dec 2019 10:58:24 +0100
Subject: ethtool: move to its own directory

The ethtool netlink interface is going to be split into multiple files so
that it will be more convenient to put all of them in a separate directory
net/ethtool. Start by moving current ethtool.c with ioctl interface into
this directory and renaming it to ioctl.c.

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/Makefile         |    2 +-
 net/core/Makefile    |    2 +-
 net/core/ethtool.c   | 3116 --------------------------------------------------
 net/ethtool/Makefile |    3 +
 net/ethtool/ioctl.c  | 3116 ++++++++++++++++++++++++++++++++++++++++++++++++++
 5 files changed, 3121 insertions(+), 3118 deletions(-)
 delete mode 100644 net/core/ethtool.c
 create mode 100644 net/ethtool/Makefile
 create mode 100644 net/ethtool/ioctl.c

diff --git a/net/Makefile b/net/Makefile
index 449fc0b221f8..848303d98d3d 100644
--- a/net/Makefile
+++ b/net/Makefile
@@ -13,7 +13,7 @@ obj-$(CONFIG_NET)		+= $(tmp-y)
 
 # LLC has to be linked before the files in net/802/
 obj-$(CONFIG_LLC)		+= llc/
-obj-$(CONFIG_NET)		+= ethernet/ 802/ sched/ netlink/ bpf/
+obj-$(CONFIG_NET)		+= ethernet/ 802/ sched/ netlink/ bpf/ ethtool/
 obj-$(CONFIG_NETFILTER)		+= netfilter/
 obj-$(CONFIG_INET)		+= ipv4/
 obj-$(CONFIG_TLS)		+= tls/
diff --git a/net/core/Makefile b/net/core/Makefile
index a104dc8faafc..3e2c378e5f31 100644
--- a/net/core/Makefile
+++ b/net/core/Makefile
@@ -8,7 +8,7 @@ obj-y := sock.o request_sock.o skbuff.o datagram.o stream.o scm.o \
 
 obj-$(CONFIG_SYSCTL) += sysctl_net_core.o
 
-obj-y		     += dev.o ethtool.o dev_addr_lists.o dst.o netevent.o \
+obj-y		     += dev.o dev_addr_lists.o dst.o netevent.o \
 			neighbour.o rtnetlink.o utils.o link_watch.o filter.o \
 			sock_diag.o dev_ioctl.o tso.o sock_reuseport.o \
 			fib_notifier.o xdp.o flow_offload.o
diff --git a/net/core/ethtool.c b/net/core/ethtool.c
deleted file mode 100644
index cd9bc67381b2..000000000000
--- a/net/core/ethtool.c
+++ /dev/null
@@ -1,3116 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0-or-later
-/*
- * net/core/ethtool.c - Ethtool ioctl handler
- * Copyright (c) 2003 Matthew Wilcox <matthew@wil.cx>
- *
- * This file is where we call all the ethtool_ops commands to get
- * the information ethtool needs.
- */
-
-#include <linux/module.h>
-#include <linux/types.h>
-#include <linux/capability.h>
-#include <linux/errno.h>
-#include <linux/ethtool.h>
-#include <linux/netdevice.h>
-#include <linux/net_tstamp.h>
-#include <linux/phy.h>
-#include <linux/bitops.h>
-#include <linux/uaccess.h>
-#include <linux/vmalloc.h>
-#include <linux/sfp.h>
-#include <linux/slab.h>
-#include <linux/rtnetlink.h>
-#include <linux/sched/signal.h>
-#include <linux/net.h>
-#include <net/devlink.h>
-#include <net/xdp_sock.h>
-#include <net/flow_offload.h>
-
-/*
- * Some useful ethtool_ops methods that're device independent.
- * If we find that all drivers want to do the same thing here,
- * we can turn these into dev_() function calls.
- */
-
-u32 ethtool_op_get_link(struct net_device *dev)
-{
-	return netif_carrier_ok(dev) ? 1 : 0;
-}
-EXPORT_SYMBOL(ethtool_op_get_link);
-
-int ethtool_op_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info)
-{
-	info->so_timestamping =
-		SOF_TIMESTAMPING_TX_SOFTWARE |
-		SOF_TIMESTAMPING_RX_SOFTWARE |
-		SOF_TIMESTAMPING_SOFTWARE;
-	info->phc_index = -1;
-	return 0;
-}
-EXPORT_SYMBOL(ethtool_op_get_ts_info);
-
-/* Handlers for each ethtool command */
-
-#define ETHTOOL_DEV_FEATURE_WORDS	((NETDEV_FEATURE_COUNT + 31) / 32)
-
-static const char netdev_features_strings[NETDEV_FEATURE_COUNT][ETH_GSTRING_LEN] = {
-	[NETIF_F_SG_BIT] =               "tx-scatter-gather",
-	[NETIF_F_IP_CSUM_BIT] =          "tx-checksum-ipv4",
-	[NETIF_F_HW_CSUM_BIT] =          "tx-checksum-ip-generic",
-	[NETIF_F_IPV6_CSUM_BIT] =        "tx-checksum-ipv6",
-	[NETIF_F_HIGHDMA_BIT] =          "highdma",
-	[NETIF_F_FRAGLIST_BIT] =         "tx-scatter-gather-fraglist",
-	[NETIF_F_HW_VLAN_CTAG_TX_BIT] =  "tx-vlan-hw-insert",
-
-	[NETIF_F_HW_VLAN_CTAG_RX_BIT] =  "rx-vlan-hw-parse",
-	[NETIF_F_HW_VLAN_CTAG_FILTER_BIT] = "rx-vlan-filter",
-	[NETIF_F_HW_VLAN_STAG_TX_BIT] =  "tx-vlan-stag-hw-insert",
-	[NETIF_F_HW_VLAN_STAG_RX_BIT] =  "rx-vlan-stag-hw-parse",
-	[NETIF_F_HW_VLAN_STAG_FILTER_BIT] = "rx-vlan-stag-filter",
-	[NETIF_F_VLAN_CHALLENGED_BIT] =  "vlan-challenged",
-	[NETIF_F_GSO_BIT] =              "tx-generic-segmentation",
-	[NETIF_F_LLTX_BIT] =             "tx-lockless",
-	[NETIF_F_NETNS_LOCAL_BIT] =      "netns-local",
-	[NETIF_F_GRO_BIT] =              "rx-gro",
-	[NETIF_F_GRO_HW_BIT] =           "rx-gro-hw",
-	[NETIF_F_LRO_BIT] =              "rx-lro",
-
-	[NETIF_F_TSO_BIT] =              "tx-tcp-segmentation",
-	[NETIF_F_GSO_ROBUST_BIT] =       "tx-gso-robust",
-	[NETIF_F_TSO_ECN_BIT] =          "tx-tcp-ecn-segmentation",
-	[NETIF_F_TSO_MANGLEID_BIT] =	 "tx-tcp-mangleid-segmentation",
-	[NETIF_F_TSO6_BIT] =             "tx-tcp6-segmentation",
-	[NETIF_F_FSO_BIT] =              "tx-fcoe-segmentation",
-	[NETIF_F_GSO_GRE_BIT] =		 "tx-gre-segmentation",
-	[NETIF_F_GSO_GRE_CSUM_BIT] =	 "tx-gre-csum-segmentation",
-	[NETIF_F_GSO_IPXIP4_BIT] =	 "tx-ipxip4-segmentation",
-	[NETIF_F_GSO_IPXIP6_BIT] =	 "tx-ipxip6-segmentation",
-	[NETIF_F_GSO_UDP_TUNNEL_BIT] =	 "tx-udp_tnl-segmentation",
-	[NETIF_F_GSO_UDP_TUNNEL_CSUM_BIT] = "tx-udp_tnl-csum-segmentation",
-	[NETIF_F_GSO_PARTIAL_BIT] =	 "tx-gso-partial",
-	[NETIF_F_GSO_SCTP_BIT] =	 "tx-sctp-segmentation",
-	[NETIF_F_GSO_ESP_BIT] =		 "tx-esp-segmentation",
-	[NETIF_F_GSO_UDP_L4_BIT] =	 "tx-udp-segmentation",
-
-	[NETIF_F_FCOE_CRC_BIT] =         "tx-checksum-fcoe-crc",
-	[NETIF_F_SCTP_CRC_BIT] =        "tx-checksum-sctp",
-	[NETIF_F_FCOE_MTU_BIT] =         "fcoe-mtu",
-	[NETIF_F_NTUPLE_BIT] =           "rx-ntuple-filter",
-	[NETIF_F_RXHASH_BIT] =           "rx-hashing",
-	[NETIF_F_RXCSUM_BIT] =           "rx-checksum",
-	[NETIF_F_NOCACHE_COPY_BIT] =     "tx-nocache-copy",
-	[NETIF_F_LOOPBACK_BIT] =         "loopback",
-	[NETIF_F_RXFCS_BIT] =            "rx-fcs",
-	[NETIF_F_RXALL_BIT] =            "rx-all",
-	[NETIF_F_HW_L2FW_DOFFLOAD_BIT] = "l2-fwd-offload",
-	[NETIF_F_HW_TC_BIT] =		 "hw-tc-offload",
-	[NETIF_F_HW_ESP_BIT] =		 "esp-hw-offload",
-	[NETIF_F_HW_ESP_TX_CSUM_BIT] =	 "esp-tx-csum-hw-offload",
-	[NETIF_F_RX_UDP_TUNNEL_PORT_BIT] =	 "rx-udp_tunnel-port-offload",
-	[NETIF_F_HW_TLS_RECORD_BIT] =	"tls-hw-record",
-	[NETIF_F_HW_TLS_TX_BIT] =	 "tls-hw-tx-offload",
-	[NETIF_F_HW_TLS_RX_BIT] =	 "tls-hw-rx-offload",
-};
-
-static const char
-rss_hash_func_strings[ETH_RSS_HASH_FUNCS_COUNT][ETH_GSTRING_LEN] = {
-	[ETH_RSS_HASH_TOP_BIT] =	"toeplitz",
-	[ETH_RSS_HASH_XOR_BIT] =	"xor",
-	[ETH_RSS_HASH_CRC32_BIT] =	"crc32",
-};
-
-static const char
-tunable_strings[__ETHTOOL_TUNABLE_COUNT][ETH_GSTRING_LEN] = {
-	[ETHTOOL_ID_UNSPEC]     = "Unspec",
-	[ETHTOOL_RX_COPYBREAK]	= "rx-copybreak",
-	[ETHTOOL_TX_COPYBREAK]	= "tx-copybreak",
-	[ETHTOOL_PFC_PREVENTION_TOUT] = "pfc-prevention-tout",
-};
-
-static const char
-phy_tunable_strings[__ETHTOOL_PHY_TUNABLE_COUNT][ETH_GSTRING_LEN] = {
-	[ETHTOOL_ID_UNSPEC]     = "Unspec",
-	[ETHTOOL_PHY_DOWNSHIFT]	= "phy-downshift",
-	[ETHTOOL_PHY_FAST_LINK_DOWN] = "phy-fast-link-down",
-	[ETHTOOL_PHY_EDPD]	= "phy-energy-detect-power-down",
-};
-
-static int ethtool_get_features(struct net_device *dev, void __user *useraddr)
-{
-	struct ethtool_gfeatures cmd = {
-		.cmd = ETHTOOL_GFEATURES,
-		.size = ETHTOOL_DEV_FEATURE_WORDS,
-	};
-	struct ethtool_get_features_block features[ETHTOOL_DEV_FEATURE_WORDS];
-	u32 __user *sizeaddr;
-	u32 copy_size;
-	int i;
-
-	/* in case feature bits run out again */
-	BUILD_BUG_ON(ETHTOOL_DEV_FEATURE_WORDS * sizeof(u32) > sizeof(netdev_features_t));
-
-	for (i = 0; i < ETHTOOL_DEV_FEATURE_WORDS; ++i) {
-		features[i].available = (u32)(dev->hw_features >> (32 * i));
-		features[i].requested = (u32)(dev->wanted_features >> (32 * i));
-		features[i].active = (u32)(dev->features >> (32 * i));
-		features[i].never_changed =
-			(u32)(NETIF_F_NEVER_CHANGE >> (32 * i));
-	}
-
-	sizeaddr = useraddr + offsetof(struct ethtool_gfeatures, size);
-	if (get_user(copy_size, sizeaddr))
-		return -EFAULT;
-
-	if (copy_size > ETHTOOL_DEV_FEATURE_WORDS)
-		copy_size = ETHTOOL_DEV_FEATURE_WORDS;
-
-	if (copy_to_user(useraddr, &cmd, sizeof(cmd)))
-		return -EFAULT;
-	useraddr += sizeof(cmd);
-	if (copy_to_user(useraddr, features, copy_size * sizeof(*features)))
-		return -EFAULT;
-
-	return 0;
-}
-
-static int ethtool_set_features(struct net_device *dev, void __user *useraddr)
-{
-	struct ethtool_sfeatures cmd;
-	struct ethtool_set_features_block features[ETHTOOL_DEV_FEATURE_WORDS];
-	netdev_features_t wanted = 0, valid = 0;
-	int i, ret = 0;
-
-	if (copy_from_user(&cmd, useraddr, sizeof(cmd)))
-		return -EFAULT;
-	useraddr += sizeof(cmd);
-
-	if (cmd.size != ETHTOOL_DEV_FEATURE_WORDS)
-		return -EINVAL;
-
-	if (copy_from_user(features, useraddr, sizeof(features)))
-		return -EFAULT;
-
-	for (i = 0; i < ETHTOOL_DEV_FEATURE_WORDS; ++i) {
-		valid |= (netdev_features_t)features[i].valid << (32 * i);
-		wanted |= (netdev_features_t)features[i].requested << (32 * i);
-	}
-
-	if (valid & ~NETIF_F_ETHTOOL_BITS)
-		return -EINVAL;
-
-	if (valid & ~dev->hw_features) {
-		valid &= dev->hw_features;
-		ret |= ETHTOOL_F_UNSUPPORTED;
-	}
-
-	dev->wanted_features &= ~valid;
-	dev->wanted_features |= wanted & valid;
-	__netdev_update_features(dev);
-
-	if ((dev->wanted_features ^ dev->features) & valid)
-		ret |= ETHTOOL_F_WISH;
-
-	return ret;
-}
-
-static int __ethtool_get_sset_count(struct net_device *dev, int sset)
-{
-	const struct ethtool_ops *ops = dev->ethtool_ops;
-
-	if (sset == ETH_SS_FEATURES)
-		return ARRAY_SIZE(netdev_features_strings);
-
-	if (sset == ETH_SS_RSS_HASH_FUNCS)
-		return ARRAY_SIZE(rss_hash_func_strings);
-
-	if (sset == ETH_SS_TUNABLES)
-		return ARRAY_SIZE(tunable_strings);
-
-	if (sset == ETH_SS_PHY_TUNABLES)
-		return ARRAY_SIZE(phy_tunable_strings);
-
-	if (sset == ETH_SS_PHY_STATS && dev->phydev &&
-	    !ops->get_ethtool_phy_stats)
-		return phy_ethtool_get_sset_count(dev->phydev);
-
-	if (ops->get_sset_count && ops->get_strings)
-		return ops->get_sset_count(dev, sset);
-	else
-		return -EOPNOTSUPP;
-}
-
-static void __ethtool_get_strings(struct net_device *dev,
-	u32 stringset, u8 *data)
-{
-	const struct ethtool_ops *ops = dev->ethtool_ops;
-
-	if (stringset == ETH_SS_FEATURES)
-		memcpy(data, netdev_features_strings,
-			sizeof(netdev_features_strings));
-	else if (stringset == ETH_SS_RSS_HASH_FUNCS)
-		memcpy(data, rss_hash_func_strings,
-		       sizeof(rss_hash_func_strings));
-	else if (stringset == ETH_SS_TUNABLES)
-		memcpy(data, tunable_strings, sizeof(tunable_strings));
-	else if (stringset == ETH_SS_PHY_TUNABLES)
-		memcpy(data, phy_tunable_strings, sizeof(phy_tunable_strings));
-	else if (stringset == ETH_SS_PHY_STATS && dev->phydev &&
-		 !ops->get_ethtool_phy_stats)
-		phy_ethtool_get_strings(dev->phydev, data);
-	else
-		/* ops->get_strings is valid because checked earlier */
-		ops->get_strings(dev, stringset, data);
-}
-
-static netdev_features_t ethtool_get_feature_mask(u32 eth_cmd)
-{
-	/* feature masks of legacy discrete ethtool ops */
-
-	switch (eth_cmd) {
-	case ETHTOOL_GTXCSUM:
-	case ETHTOOL_STXCSUM:
-		return NETIF_F_CSUM_MASK | NETIF_F_SCTP_CRC;
-	case ETHTOOL_GRXCSUM:
-	case ETHTOOL_SRXCSUM:
-		return NETIF_F_RXCSUM;
-	case ETHTOOL_GSG:
-	case ETHTOOL_SSG:
-		return NETIF_F_SG;
-	case ETHTOOL_GTSO:
-	case ETHTOOL_STSO:
-		return NETIF_F_ALL_TSO;
-	case ETHTOOL_GGSO:
-	case ETHTOOL_SGSO:
-		return NETIF_F_GSO;
-	case ETHTOOL_GGRO:
-	case ETHTOOL_SGRO:
-		return NETIF_F_GRO;
-	default:
-		BUG();
-	}
-}
-
-static int ethtool_get_one_feature(struct net_device *dev,
-	char __user *useraddr, u32 ethcmd)
-{
-	netdev_features_t mask = ethtool_get_feature_mask(ethcmd);
-	struct ethtool_value edata = {
-		.cmd = ethcmd,
-		.data = !!(dev->features & mask),
-	};
-
-	if (copy_to_user(useraddr, &edata, sizeof(edata)))
-		return -EFAULT;
-	return 0;
-}
-
-static int ethtool_set_one_feature(struct net_device *dev,
-	void __user *useraddr, u32 ethcmd)
-{
-	struct ethtool_value edata;
-	netdev_features_t mask;
-
-	if (copy_from_user(&edata, useraddr, sizeof(edata)))
-		return -EFAULT;
-
-	mask = ethtool_get_feature_mask(ethcmd);
-	mask &= dev->hw_features;
-	if (!mask)
-		return -EOPNOTSUPP;
-
-	if (edata.data)
-		dev->wanted_features |= mask;
-	else
-		dev->wanted_features &= ~mask;
-
-	__netdev_update_features(dev);
-
-	return 0;
-}
-
-#define ETH_ALL_FLAGS    (ETH_FLAG_LRO | ETH_FLAG_RXVLAN | ETH_FLAG_TXVLAN | \
-			  ETH_FLAG_NTUPLE | ETH_FLAG_RXHASH)
-#define ETH_ALL_FEATURES (NETIF_F_LRO | NETIF_F_HW_VLAN_CTAG_RX | \
-			  NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_NTUPLE | \
-			  NETIF_F_RXHASH)
-
-static u32 __ethtool_get_flags(struct net_device *dev)
-{
-	u32 flags = 0;
-
-	if (dev->features & NETIF_F_LRO)
-		flags |= ETH_FLAG_LRO;
-	if (dev->features & NETIF_F_HW_VLAN_CTAG_RX)
-		flags |= ETH_FLAG_RXVLAN;
-	if (dev->features & NETIF_F_HW_VLAN_CTAG_TX)
-		flags |= ETH_FLAG_TXVLAN;
-	if (dev->features & NETIF_F_NTUPLE)
-		flags |= ETH_FLAG_NTUPLE;
-	if (dev->features & NETIF_F_RXHASH)
-		flags |= ETH_FLAG_RXHASH;
-
-	return flags;
-}
-
-static int __ethtool_set_flags(struct net_device *dev, u32 data)
-{
-	netdev_features_t features = 0, changed;
-
-	if (data & ~ETH_ALL_FLAGS)
-		return -EINVAL;
-
-	if (data & ETH_FLAG_LRO)
-		features |= NETIF_F_LRO;
-	if (data & ETH_FLAG_RXVLAN)
-		features |= NETIF_F_HW_VLAN_CTAG_RX;
-	if (data & ETH_FLAG_TXVLAN)
-		features |= NETIF_F_HW_VLAN_CTAG_TX;
-	if (data & ETH_FLAG_NTUPLE)
-		features |= NETIF_F_NTUPLE;
-	if (data & ETH_FLAG_RXHASH)
-		features |= NETIF_F_RXHASH;
-
-	/* allow changing only bits set in hw_features */
-	changed = (features ^ dev->features) & ETH_ALL_FEATURES;
-	if (changed & ~dev->hw_features)
-		return (changed & dev->hw_features) ? -EINVAL : -EOPNOTSUPP;
-
-	dev->wanted_features =
-		(dev->wanted_features & ~changed) | (features & changed);
-
-	__netdev_update_features(dev);
-
-	return 0;
-}
-
-/* Given two link masks, AND them together and save the result in dst. */
-void ethtool_intersect_link_masks(struct ethtool_link_ksettings *dst,
-				  struct ethtool_link_ksettings *src)
-{
-	unsigned int size = BITS_TO_LONGS(__ETHTOOL_LINK_MODE_MASK_NBITS);
-	unsigned int idx = 0;
-
-	for (; idx < size; idx++) {
-		dst->link_modes.supported[idx] &=
-			src->link_modes.supported[idx];
-		dst->link_modes.advertising[idx] &=
-			src->link_modes.advertising[idx];
-	}
-}
-EXPORT_SYMBOL(ethtool_intersect_link_masks);
-
-void ethtool_convert_legacy_u32_to_link_mode(unsigned long *dst,
-					     u32 legacy_u32)
-{
-	bitmap_zero(dst, __ETHTOOL_LINK_MODE_MASK_NBITS);
-	dst[0] = legacy_u32;
-}
-EXPORT_SYMBOL(ethtool_convert_legacy_u32_to_link_mode);
-
-/* return false if src had higher bits set. lower bits always updated. */
-bool ethtool_convert_link_mode_to_legacy_u32(u32 *legacy_u32,
-					     const unsigned long *src)
-{
-	bool retval = true;
-
-	/* TODO: following test will soon always be true */
-	if (__ETHTOOL_LINK_MODE_MASK_NBITS > 32) {
-		__ETHTOOL_DECLARE_LINK_MODE_MASK(ext);
-
-		bitmap_zero(ext, __ETHTOOL_LINK_MODE_MASK_NBITS);
-		bitmap_fill(ext, 32);
-		bitmap_complement(ext, ext, __ETHTOOL_LINK_MODE_MASK_NBITS);
-		if (bitmap_intersects(ext, src,
-				      __ETHTOOL_LINK_MODE_MASK_NBITS)) {
-			/* src mask goes beyond bit 31 */
-			retval = false;
-		}
-	}
-	*legacy_u32 = src[0];
-	return retval;
-}
-EXPORT_SYMBOL(ethtool_convert_link_mode_to_legacy_u32);
-
-/* return false if legacy contained non-0 deprecated fields
- * maxtxpkt/maxrxpkt. rest of ksettings always updated
- */
-static bool
-convert_legacy_settings_to_link_ksettings(
-	struct ethtool_link_ksettings *link_ksettings,
-	const struct ethtool_cmd *legacy_settings)
-{
-	bool retval = true;
-
-	memset(link_ksettings, 0, sizeof(*link_ksettings));
-
-	/* This is used to tell users that driver is still using these
-	 * deprecated legacy fields, and they should not use
-	 * %ETHTOOL_GLINKSETTINGS/%ETHTOOL_SLINKSETTINGS
-	 */
-	if (legacy_settings->maxtxpkt ||
-	    legacy_settings->maxrxpkt)
-		retval = false;
-
-	ethtool_convert_legacy_u32_to_link_mode(
-		link_ksettings->link_modes.supported,
-		legacy_settings->supported);
-	ethtool_convert_legacy_u32_to_link_mode(
-		link_ksettings->link_modes.advertising,
-		legacy_settings->advertising);
-	ethtool_convert_legacy_u32_to_link_mode(
-		link_ksettings->link_modes.lp_advertising,
-		legacy_settings->lp_advertising);
-	link_ksettings->base.speed
-		= ethtool_cmd_speed(legacy_settings);
-	link_ksettings->base.duplex
-		= legacy_settings->duplex;
-	link_ksettings->base.port
-		= legacy_settings->port;
-	link_ksettings->base.phy_address
-		= legacy_settings->phy_address;
-	link_ksettings->base.autoneg
-		= legacy_settings->autoneg;
-	link_ksettings->base.mdio_support
-		= legacy_settings->mdio_support;
-	link_ksettings->base.eth_tp_mdix
-		= legacy_settings->eth_tp_mdix;
-	link_ksettings->base.eth_tp_mdix_ctrl
-		= legacy_settings->eth_tp_mdix_ctrl;
-	return retval;
-}
-
-/* return false if ksettings link modes had higher bits
- * set. legacy_settings always updated (best effort)
- */
-static bool
-convert_link_ksettings_to_legacy_settings(
-	struct ethtool_cmd *legacy_settings,
-	const struct ethtool_link_ksettings *link_ksettings)
-{
-	bool retval = true;
-
-	memset(legacy_settings, 0, sizeof(*legacy_settings));
-	/* this also clears the deprecated fields in legacy structure:
-	 * __u8		transceiver;
-	 * __u32	maxtxpkt;
-	 * __u32	maxrxpkt;
-	 */
-
-	retval &= ethtool_convert_link_mode_to_legacy_u32(
-		&legacy_settings->supported,
-		link_ksettings->link_modes.supported);
-	retval &= ethtool_convert_link_mode_to_legacy_u32(
-		&legacy_settings->advertising,
-		link_ksettings->link_modes.advertising);
-	retval &= ethtool_convert_link_mode_to_legacy_u32(
-		&legacy_settings->lp_advertising,
-		link_ksettings->link_modes.lp_advertising);
-	ethtool_cmd_speed_set(legacy_settings, link_ksettings->base.speed);
-	legacy_settings->duplex
-		= link_ksettings->base.duplex;
-	legacy_settings->port
-		= link_ksettings->base.port;
-	legacy_settings->phy_address
-		= link_ksettings->base.phy_address;
-	legacy_settings->autoneg
-		= link_ksettings->base.autoneg;
-	legacy_settings->mdio_support
-		= link_ksettings->base.mdio_support;
-	legacy_settings->eth_tp_mdix
-		= link_ksettings->base.eth_tp_mdix;
-	legacy_settings->eth_tp_mdix_ctrl
-		= link_ksettings->base.eth_tp_mdix_ctrl;
-	legacy_settings->transceiver
-		= link_ksettings->base.transceiver;
-	return retval;
-}
-
-/* number of 32-bit words to store the user's link mode bitmaps */
-#define __ETHTOOL_LINK_MODE_MASK_NU32			\
-	DIV_ROUND_UP(__ETHTOOL_LINK_MODE_MASK_NBITS, 32)
-
-/* layout of the struct passed from/to userland */
-struct ethtool_link_usettings {
-	struct ethtool_link_settings base;
-	struct {
-		__u32 supported[__ETHTOOL_LINK_MODE_MASK_NU32];
-		__u32 advertising[__ETHTOOL_LINK_MODE_MASK_NU32];
-		__u32 lp_advertising[__ETHTOOL_LINK_MODE_MASK_NU32];
-	} link_modes;
-};
-
-/* Internal kernel helper to query a device ethtool_link_settings. */
-int __ethtool_get_link_ksettings(struct net_device *dev,
-				 struct ethtool_link_ksettings *link_ksettings)
-{
-	ASSERT_RTNL();
-
-	if (!dev->ethtool_ops->get_link_ksettings)
-		return -EOPNOTSUPP;
-
-	memset(link_ksettings, 0, sizeof(*link_ksettings));
-	return dev->ethtool_ops->get_link_ksettings(dev, link_ksettings);
-}
-EXPORT_SYMBOL(__ethtool_get_link_ksettings);
-
-/* convert ethtool_link_usettings in user space to a kernel internal
- * ethtool_link_ksettings. return 0 on success, errno on error.
- */
-static int load_link_ksettings_from_user(struct ethtool_link_ksettings *to,
-					 const void __user *from)
-{
-	struct ethtool_link_usettings link_usettings;
-
-	if (copy_from_user(&link_usettings, from, sizeof(link_usettings)))
-		return -EFAULT;
-
-	memcpy(&to->base, &link_usettings.base, sizeof(to->base));
-	bitmap_from_arr32(to->link_modes.supported,
-			  link_usettings.link_modes.supported,
-			  __ETHTOOL_LINK_MODE_MASK_NBITS);
-	bitmap_from_arr32(to->link_modes.advertising,
-			  link_usettings.link_modes.advertising,
-			  __ETHTOOL_LINK_MODE_MASK_NBITS);
-	bitmap_from_arr32(to->link_modes.lp_advertising,
-			  link_usettings.link_modes.lp_advertising,
-			  __ETHTOOL_LINK_MODE_MASK_NBITS);
-
-	return 0;
-}
-
-/* convert a kernel internal ethtool_link_ksettings to
- * ethtool_link_usettings in user space. return 0 on success, errno on
- * error.
- */
-static int
-store_link_ksettings_for_user(void __user *to,
-			      const struct ethtool_link_ksettings *from)
-{
-	struct ethtool_link_usettings link_usettings;
-
-	memcpy(&link_usettings.base, &from->base, sizeof(link_usettings));
-	bitmap_to_arr32(link_usettings.link_modes.supported,
-			from->link_modes.supported,
-			__ETHTOOL_LINK_MODE_MASK_NBITS);
-	bitmap_to_arr32(link_usettings.link_modes.advertising,
-			from->link_modes.advertising,
-			__ETHTOOL_LINK_MODE_MASK_NBITS);
-	bitmap_to_arr32(link_usettings.link_modes.lp_advertising,
-			from->link_modes.lp_advertising,
-			__ETHTOOL_LINK_MODE_MASK_NBITS);
-
-	if (copy_to_user(to, &link_usettings, sizeof(link_usettings)))
-		return -EFAULT;
-
-	return 0;
-}
-
-/* Query device for its ethtool_link_settings. */
-static int ethtool_get_link_ksettings(struct net_device *dev,
-				      void __user *useraddr)
-{
-	int err = 0;
-	struct ethtool_link_ksettings link_ksettings;
-
-	ASSERT_RTNL();
-	if (!dev->ethtool_ops->get_link_ksettings)
-		return -EOPNOTSUPP;
-
-	/* handle bitmap nbits handshake */
-	if (copy_from_user(&link_ksettings.base, useraddr,
-			   sizeof(link_ksettings.base)))
-		return -EFAULT;
-
-	if (__ETHTOOL_LINK_MODE_MASK_NU32
-	    != link_ksettings.base.link_mode_masks_nwords) {
-		/* wrong link mode nbits requested */
-		memset(&link_ksettings, 0, sizeof(link_ksettings));
-		link_ksettings.base.cmd = ETHTOOL_GLINKSETTINGS;
-		/* send back number of words required as negative val */
-		compiletime_assert(__ETHTOOL_LINK_MODE_MASK_NU32 <= S8_MAX,
-				   "need too many bits for link modes!");
-		link_ksettings.base.link_mode_masks_nwords
-			= -((s8)__ETHTOOL_LINK_MODE_MASK_NU32);
-
-		/* copy the base fields back to user, not the link
-		 * mode bitmaps
-		 */
-		if (copy_to_user(useraddr, &link_ksettings.base,
-				 sizeof(link_ksettings.base)))
-			return -EFAULT;
-
-		return 0;
-	}
-
-	/* handshake successful: user/kernel agree on
-	 * link_mode_masks_nwords
-	 */
-
-	memset(&link_ksettings, 0, sizeof(link_ksettings));
-	err = dev->ethtool_ops->get_link_ksettings(dev, &link_ksettings);
-	if (err < 0)
-		return err;
-
-	/* make sure we tell the right values to user */
-	link_ksettings.base.cmd = ETHTOOL_GLINKSETTINGS;
-	link_ksettings.base.link_mode_masks_nwords
-		= __ETHTOOL_LINK_MODE_MASK_NU32;
-
-	return store_link_ksettings_for_user(useraddr, &link_ksettings);
-}
-
-/* Update device ethtool_link_settings. */
-static int ethtool_set_link_ksettings(struct net_device *dev,
-				      void __user *useraddr)
-{
-	int err;
-	struct ethtool_link_ksettings link_ksettings;
-
-	ASSERT_RTNL();
-
-	if (!dev->ethtool_ops->set_link_ksettings)
-		return -EOPNOTSUPP;
-
-	/* make sure nbits field has expected value */
-	if (copy_from_user(&link_ksettings.base, useraddr,
-			   sizeof(link_ksettings.base)))
-		return -EFAULT;
-
-	if (__ETHTOOL_LINK_MODE_MASK_NU32
-	    != link_ksettings.base.link_mode_masks_nwords)
-		return -EINVAL;
-
-	/* copy the whole structure, now that we know it has expected
-	 * format
-	 */
-	err = load_link_ksettings_from_user(&link_ksettings, useraddr);
-	if (err)
-		return err;
-
-	/* re-check nwords field, just in case */
-	if (__ETHTOOL_LINK_MODE_MASK_NU32
-	    != link_ksettings.base.link_mode_masks_nwords)
-		return -EINVAL;
-
-	return dev->ethtool_ops->set_link_ksettings(dev, &link_ksettings);
-}
-
-/* Query device for its ethtool_cmd settings.
- *
- * Backward compatibility note: for compatibility with legacy ethtool, this is
- * now implemented via get_link_ksettings. When driver reports higher link mode
- * bits, a kernel warning is logged once (with name of 1st driver/device) to
- * recommend user to upgrade ethtool, but the command is successful (only the
- * lower link mode bits reported back to user). Deprecated fields from
- * ethtool_cmd (transceiver/maxrxpkt/maxtxpkt) are always set to zero.
- */
-static int ethtool_get_settings(struct net_device *dev, void __user *useraddr)
-{
-	struct ethtool_link_ksettings link_ksettings;
-	struct ethtool_cmd cmd;
-	int err;
-
-	ASSERT_RTNL();
-	if (!dev->ethtool_ops->get_link_ksettings)
-		return -EOPNOTSUPP;
-
-	memset(&link_ksettings, 0, sizeof(link_ksettings));
-	err = dev->ethtool_ops->get_link_ksettings(dev, &link_ksettings);
-	if (err < 0)
-		return err;
-	convert_link_ksettings_to_legacy_settings(&cmd, &link_ksettings);
-
-	/* send a sensible cmd tag back to user */
-	cmd.cmd = ETHTOOL_GSET;
-
-	if (copy_to_user(useraddr, &cmd, sizeof(cmd)))
-		return -EFAULT;
-
-	return 0;
-}
-
-/* Update device link settings with given ethtool_cmd.
- *
- * Backward compatibility note: for compatibility with legacy ethtool, this is
- * now always implemented via set_link_settings. When user's request updates
- * deprecated ethtool_cmd fields (transceiver/maxrxpkt/maxtxpkt), a kernel
- * warning is logged once (with name of 1st driver/device) to recommend user to
- * upgrade ethtool, and the request is rejected.
- */
-static int ethtool_set_settings(struct net_device *dev, void __user *useraddr)
-{
-	struct ethtool_link_ksettings link_ksettings;
-	struct ethtool_cmd cmd;
-
-	ASSERT_RTNL();
-
-	if (copy_from_user(&cmd, useraddr, sizeof(cmd)))
-		return -EFAULT;
-	if (!dev->ethtool_ops->set_link_ksettings)
-		return -EOPNOTSUPP;
-
-	if (!convert_legacy_settings_to_link_ksettings(&link_ksettings, &cmd))
-		return -EINVAL;
-	link_ksettings.base.link_mode_masks_nwords =
-		__ETHTOOL_LINK_MODE_MASK_NU32;
-	return dev->ethtool_ops->set_link_ksettings(dev, &link_ksettings);
-}
-
-static noinline_for_stack int ethtool_get_drvinfo(struct net_device *dev,
-						  void __user *useraddr)
-{
-	struct ethtool_drvinfo info;
-	const struct ethtool_ops *ops = dev->ethtool_ops;
-
-	memset(&info, 0, sizeof(info));
-	info.cmd = ETHTOOL_GDRVINFO;
-	if (ops->get_drvinfo) {
-		ops->get_drvinfo(dev, &info);
-	} else if (dev->dev.parent && dev->dev.parent->driver) {
-		strlcpy(info.bus_info, dev_name(dev->dev.parent),
-			sizeof(info.bus_info));
-		strlcpy(info.driver, dev->dev.parent->driver->name,
-			sizeof(info.driver));
-	} else {
-		return -EOPNOTSUPP;
-	}
-
-	/*
-	 * this method of obtaining string set info is deprecated;
-	 * Use ETHTOOL_GSSET_INFO instead.
-	 */
-	if (ops->get_sset_count) {
-		int rc;
-
-		rc = ops->get_sset_count(dev, ETH_SS_TEST);
-		if (rc >= 0)
-			info.testinfo_len = rc;
-		rc = ops->get_sset_count(dev, ETH_SS_STATS);
-		if (rc >= 0)
-			info.n_stats = rc;
-		rc = ops->get_sset_count(dev, ETH_SS_PRIV_FLAGS);
-		if (rc >= 0)
-			info.n_priv_flags = rc;
-	}
-	if (ops->get_regs_len) {
-		int ret = ops->get_regs_len(dev);
-
-		if (ret > 0)
-			info.regdump_len = ret;
-	}
-
-	if (ops->get_eeprom_len)
-		info.eedump_len = ops->get_eeprom_len(dev);
-
-	if (!info.fw_version[0])
-		devlink_compat_running_version(dev, info.fw_version,
-					       sizeof(info.fw_version));
-
-	if (copy_to_user(useraddr, &info, sizeof(info)))
-		return -EFAULT;
-	return 0;
-}
-
-static noinline_for_stack int ethtool_get_sset_info(struct net_device *dev,
-						    void __user *useraddr)
-{
-	struct ethtool_sset_info info;
-	u64 sset_mask;
-	int i, idx = 0, n_bits = 0, ret, rc;
-	u32 *info_buf = NULL;
-
-	if (copy_from_user(&info, useraddr, sizeof(info)))
-		return -EFAULT;
-
-	/* store copy of mask, because we zero struct later on */
-	sset_mask = info.sset_mask;
-	if (!sset_mask)
-		return 0;
-
-	/* calculate size of return buffer */
-	n_bits = hweight64(sset_mask);
-
-	memset(&info, 0, sizeof(info));
-	info.cmd = ETHTOOL_GSSET_INFO;
-
-	info_buf = kcalloc(n_bits, sizeof(u32), GFP_USER);
-	if (!info_buf)
-		return -ENOMEM;
-
-	/*
-	 * fill return buffer based on input bitmask and successful
-	 * get_sset_count return
-	 */
-	for (i = 0; i < 64; i++) {
-		if (!(sset_mask & (1ULL << i)))
-			continue;
-
-		rc = __ethtool_get_sset_count(dev, i);
-		if (rc >= 0) {
-			info.sset_mask |= (1ULL << i);
-			info_buf[idx++] = rc;
-		}
-	}
-
-	ret = -EFAULT;
-	if (copy_to_user(useraddr, &info, sizeof(info)))
-		goto out;
-
-	useraddr += offsetof(struct ethtool_sset_info, data);
-	if (copy_to_user(useraddr, info_buf, idx * sizeof(u32)))
-		goto out;
-
-	ret = 0;
-
-out:
-	kfree(info_buf);
-	return ret;
-}
-
-static noinline_for_stack int ethtool_set_rxnfc(struct net_device *dev,
-						u32 cmd, void __user *useraddr)
-{
-	struct ethtool_rxnfc info;
-	size_t info_size = sizeof(info);
-	int rc;
-
-	if (!dev->ethtool_ops->set_rxnfc)
-		return -EOPNOTSUPP;
-
-	/* struct ethtool_rxnfc was originally defined for
-	 * ETHTOOL_{G,S}RXFH with only the cmd, flow_type and data
-	 * members.  User-space might still be using that
-	 * definition. */
-	if (cmd == ETHTOOL_SRXFH)
-		info_size = (offsetof(struct ethtool_rxnfc, data) +
-			     sizeof(info.data));
-
-	if (copy_from_user(&info, useraddr, info_size))
-		return -EFAULT;
-
-	rc = dev->ethtool_ops->set_rxnfc(dev, &info);
-	if (rc)
-		return rc;
-
-	if (cmd == ETHTOOL_SRXCLSRLINS &&
-	    copy_to_user(useraddr, &info, info_size))
-		return -EFAULT;
-
-	return 0;
-}
-
-static noinline_for_stack int ethtool_get_rxnfc(struct net_device *dev,
-						u32 cmd, void __user *useraddr)
-{
-	struct ethtool_rxnfc info;
-	size_t info_size = sizeof(info);
-	const struct ethtool_ops *ops = dev->ethtool_ops;
-	int ret;
-	void *rule_buf = NULL;
-
-	if (!ops->get_rxnfc)
-		return -EOPNOTSUPP;
-
-	/* struct ethtool_rxnfc was originally defined for
-	 * ETHTOOL_{G,S}RXFH with only the cmd, flow_type and data
-	 * members.  User-space might still be using that
-	 * definition. */
-	if (cmd == ETHTOOL_GRXFH)
-		info_size = (offsetof(struct ethtool_rxnfc, data) +
-			     sizeof(info.data));
-
-	if (copy_from_user(&info, useraddr, info_size))
-		return -EFAULT;
-
-	/* If FLOW_RSS was requested then user-space must be using the
-	 * new definition, as FLOW_RSS is newer.
-	 */
-	if (cmd == ETHTOOL_GRXFH && info.flow_type & FLOW_RSS) {
-		info_size = sizeof(info);
-		if (copy_from_user(&info, useraddr, info_size))
-			return -EFAULT;
-		/* Since malicious users may modify the original data,
-		 * we need to check whether FLOW_RSS is still requested.
-		 */
-		if (!(info.flow_type & FLOW_RSS))
-			return -EINVAL;
-	}
-
-	if (info.cmd != cmd)
-		return -EINVAL;
-
-	if (info.cmd == ETHTOOL_GRXCLSRLALL) {
-		if (info.rule_cnt > 0) {
-			if (info.rule_cnt <= KMALLOC_MAX_SIZE / sizeof(u32))
-				rule_buf = kcalloc(info.rule_cnt, sizeof(u32),
-						   GFP_USER);
-			if (!rule_buf)
-				return -ENOMEM;
-		}
-	}
-
-	ret = ops->get_rxnfc(dev, &info, rule_buf);
-	if (ret < 0)
-		goto err_out;
-
-	ret = -EFAULT;
-	if (copy_to_user(useraddr, &info, info_size))
-		goto err_out;
-
-	if (rule_buf) {
-		useraddr += offsetof(struct ethtool_rxnfc, rule_locs);
-		if (copy_to_user(useraddr, rule_buf,
-				 info.rule_cnt * sizeof(u32)))
-			goto err_out;
-	}
-	ret = 0;
-
-err_out:
-	kfree(rule_buf);
-
-	return ret;
-}
-
-static int ethtool_copy_validate_indir(u32 *indir, void __user *useraddr,
-					struct ethtool_rxnfc *rx_rings,
-					u32 size)
-{
-	int i;
-
-	if (copy_from_user(indir, useraddr, size * sizeof(indir[0])))
-		return -EFAULT;
-
-	/* Validate ring indices */
-	for (i = 0; i < size; i++)
-		if (indir[i] >= rx_rings->data)
-			return -EINVAL;
-
-	return 0;
-}
-
-u8 netdev_rss_key[NETDEV_RSS_KEY_LEN] __read_mostly;
-
-void netdev_rss_key_fill(void *buffer, size_t len)
-{
-	BUG_ON(len > sizeof(netdev_rss_key));
-	net_get_random_once(netdev_rss_key, sizeof(netdev_rss_key));
-	memcpy(buffer, netdev_rss_key, len);
-}
-EXPORT_SYMBOL(netdev_rss_key_fill);
-
-static int ethtool_get_max_rxfh_channel(struct net_device *dev, u32 *max)
-{
-	u32 dev_size, current_max = 0;
-	u32 *indir;
-	int ret;
-
-	if (!dev->ethtool_ops->get_rxfh_indir_size ||
-	    !dev->ethtool_ops->get_rxfh)
-		return -EOPNOTSUPP;
-	dev_size = dev->ethtool_ops->get_rxfh_indir_size(dev);
-	if (dev_size == 0)
-		return -EOPNOTSUPP;
-
-	indir = kcalloc(dev_size, sizeof(indir[0]), GFP_USER);
-	if (!indir)
-		return -ENOMEM;
-
-	ret = dev->ethtool_ops->get_rxfh(dev, indir, NULL, NULL);
-	if (ret)
-		goto out;
-
-	while (dev_size--)
-		current_max = max(current_max, indir[dev_size]);
-
-	*max = current_max;
-
-out:
-	kfree(indir);
-	return ret;
-}
-
-static noinline_for_stack int ethtool_get_rxfh_indir(struct net_device *dev,
-						     void __user *useraddr)
-{
-	u32 user_size, dev_size;
-	u32 *indir;
-	int ret;
-
-	if (!dev->ethtool_ops->get_rxfh_indir_size ||
-	    !dev->ethtool_ops->get_rxfh)
-		return -EOPNOTSUPP;
-	dev_size = dev->ethtool_ops->get_rxfh_indir_size(dev);
-	if (dev_size == 0)
-		return -EOPNOTSUPP;
-
-	if (copy_from_user(&user_size,
-			   useraddr + offsetof(struct ethtool_rxfh_indir, size),
-			   sizeof(user_size)))
-		return -EFAULT;
-
-	if (copy_to_user(useraddr + offsetof(struct ethtool_rxfh_indir, size),
-			 &dev_size, sizeof(dev_size)))
-		return -EFAULT;
-
-	/* If the user buffer size is 0, this is just a query for the
-	 * device table size.  Otherwise, if it's smaller than the
-	 * device table size it's an error.
-	 */
-	if (user_size < dev_size)
-		return user_size == 0 ? 0 : -EINVAL;
-
-	indir = kcalloc(dev_size, sizeof(indir[0]), GFP_USER);
-	if (!indir)
-		return -ENOMEM;
-
-	ret = dev->ethtool_ops->get_rxfh(dev, indir, NULL, NULL);
-	if (ret)
-		goto out;
-
-	if (copy_to_user(useraddr +
-			 offsetof(struct ethtool_rxfh_indir, ring_index[0]),
-			 indir, dev_size * sizeof(indir[0])))
-		ret = -EFAULT;
-
-out:
-	kfree(indir);
-	return ret;
-}
-
-static noinline_for_stack int ethtool_set_rxfh_indir(struct net_device *dev,
-						     void __user *useraddr)
-{
-	struct ethtool_rxnfc rx_rings;
-	u32 user_size, dev_size, i;
-	u32 *indir;
-	const struct ethtool_ops *ops = dev->ethtool_ops;
-	int ret;
-	u32 ringidx_offset = offsetof(struct ethtool_rxfh_indir, ring_index[0]);
-
-	if (!ops->get_rxfh_indir_size || !ops->set_rxfh ||
-	    !ops->get_rxnfc)
-		return -EOPNOTSUPP;
-
-	dev_size = ops->get_rxfh_indir_size(dev);
-	if (dev_size == 0)
-		return -EOPNOTSUPP;
-
-	if (copy_from_user(&user_size,
-			   useraddr + offsetof(struct ethtool_rxfh_indir, size),
-			   sizeof(user_size)))
-		return -EFAULT;
-
-	if (user_size != 0 && user_size != dev_size)
-		return -EINVAL;
-
-	indir = kcalloc(dev_size, sizeof(indir[0]), GFP_USER);
-	if (!indir)
-		return -ENOMEM;
-
-	rx_rings.cmd = ETHTOOL_GRXRINGS;
-	ret = ops->get_rxnfc(dev, &rx_rings, NULL);
-	if (ret)
-		goto out;
-
-	if (user_size == 0) {
-		for (i = 0; i < dev_size; i++)
-			indir[i] = ethtool_rxfh_indir_default(i, rx_rings.data);
-	} else {
-		ret = ethtool_copy_validate_indir(indir,
-						  useraddr + ringidx_offset,
-						  &rx_rings,
-						  dev_size);
-		if (ret)
-			goto out;
-	}
-
-	ret = ops->set_rxfh(dev, indir, NULL, ETH_RSS_HASH_NO_CHANGE);
-	if (ret)
-		goto out;
-
-	/* indicate whether rxfh was set to default */
-	if (user_size == 0)
-		dev->priv_flags &= ~IFF_RXFH_CONFIGURED;
-	else
-		dev->priv_flags |= IFF_RXFH_CONFIGURED;
-
-out:
-	kfree(indir);
-	return ret;
-}
-
-static noinline_for_stack int ethtool_get_rxfh(struct net_device *dev,
-					       void __user *useraddr)
-{
-	int ret;
-	const struct ethtool_ops *ops = dev->ethtool_ops;
-	u32 user_indir_size, user_key_size;
-	u32 dev_indir_size = 0, dev_key_size = 0;
-	struct ethtool_rxfh rxfh;
-	u32 total_size;
-	u32 indir_bytes;
-	u32 *indir = NULL;
-	u8 dev_hfunc = 0;
-	u8 *hkey = NULL;
-	u8 *rss_config;
-
-	if (!ops->get_rxfh)
-		return -EOPNOTSUPP;
-
-	if (ops->get_rxfh_indir_size)
-		dev_indir_size = ops->get_rxfh_indir_size(dev);
-	if (ops->get_rxfh_key_size)
-		dev_key_size = ops->get_rxfh_key_size(dev);
-
-	if (copy_from_user(&rxfh, useraddr, sizeof(rxfh)))
-		return -EFAULT;
-	user_indir_size = rxfh.indir_size;
-	user_key_size = rxfh.key_size;
-
-	/* Check that reserved fields are 0 for now */
-	if (rxfh.rsvd8[0] || rxfh.rsvd8[1] || rxfh.rsvd8[2] || rxfh.rsvd32)
-		return -EINVAL;
-	/* Most drivers don't handle rss_context, check it's 0 as well */
-	if (rxfh.rss_context && !ops->get_rxfh_context)
-		return -EOPNOTSUPP;
-
-	rxfh.indir_size = dev_indir_size;
-	rxfh.key_size = dev_key_size;
-	if (copy_to_user(useraddr, &rxfh, sizeof(rxfh)))
-		return -EFAULT;
-
-	if ((user_indir_size && (user_indir_size != dev_indir_size)) ||
-	    (user_key_size && (user_key_size != dev_key_size)))
-		return -EINVAL;
-
-	indir_bytes = user_indir_size * sizeof(indir[0]);
-	total_size = indir_bytes + user_key_size;
-	rss_config = kzalloc(total_size, GFP_USER);
-	if (!rss_config)
-		return -ENOMEM;
-
-	if (user_indir_size)
-		indir = (u32 *)rss_config;
-
-	if (user_key_size)
-		hkey = rss_config + indir_bytes;
-
-	if (rxfh.rss_context)
-		ret = dev->ethtool_ops->get_rxfh_context(dev, indir, hkey,
-							 &dev_hfunc,
-							 rxfh.rss_context);
-	else
-		ret = dev->ethtool_ops->get_rxfh(dev, indir, hkey, &dev_hfunc);
-	if (ret)
-		goto out;
-
-	if (copy_to_user(useraddr + offsetof(struct ethtool_rxfh, hfunc),
-			 &dev_hfunc, sizeof(rxfh.hfunc))) {
-		ret = -EFAULT;
-	} else if (copy_to_user(useraddr +
-			      offsetof(struct ethtool_rxfh, rss_config[0]),
-			      rss_config, total_size)) {
-		ret = -EFAULT;
-	}
-out:
-	kfree(rss_config);
-
-	return ret;
-}
-
-static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev,
-					       void __user *useraddr)
-{
-	int ret;
-	const struct ethtool_ops *ops = dev->ethtool_ops;
-	struct ethtool_rxnfc rx_rings;
-	struct ethtool_rxfh rxfh;
-	u32 dev_indir_size = 0, dev_key_size = 0, i;
-	u32 *indir = NULL, indir_bytes = 0;
-	u8 *hkey = NULL;
-	u8 *rss_config;
-	u32 rss_cfg_offset = offsetof(struct ethtool_rxfh, rss_config[0]);
-	bool delete = false;
-
-	if (!ops->get_rxnfc || !ops->set_rxfh)
-		return -EOPNOTSUPP;
-
-	if (ops->get_rxfh_indir_size)
-		dev_indir_size = ops->get_rxfh_indir_size(dev);
-	if (ops->get_rxfh_key_size)
-		dev_key_size = ops->get_rxfh_key_size(dev);
-
-	if (copy_from_user(&rxfh, useraddr, sizeof(rxfh)))
-		return -EFAULT;
-
-	/* Check that reserved fields are 0 for now */
-	if (rxfh.rsvd8[0] || rxfh.rsvd8[1] || rxfh.rsvd8[2] || rxfh.rsvd32)
-		return -EINVAL;
-	/* Most drivers don't handle rss_context, check it's 0 as well */
-	if (rxfh.rss_context && !ops->set_rxfh_context)
-		return -EOPNOTSUPP;
-
-	/* If either indir, hash key or function is valid, proceed further.
-	 * Must request at least one change: indir size, hash key or function.
-	 */
-	if ((rxfh.indir_size &&
-	     rxfh.indir_size != ETH_RXFH_INDIR_NO_CHANGE &&
-	     rxfh.indir_size != dev_indir_size) ||
-	    (rxfh.key_size && (rxfh.key_size != dev_key_size)) ||
-	    (rxfh.indir_size == ETH_RXFH_INDIR_NO_CHANGE &&
-	     rxfh.key_size == 0 && rxfh.hfunc == ETH_RSS_HASH_NO_CHANGE))
-		return -EINVAL;
-
-	if (rxfh.indir_size != ETH_RXFH_INDIR_NO_CHANGE)
-		indir_bytes = dev_indir_size * sizeof(indir[0]);
-
-	rss_config = kzalloc(indir_bytes + rxfh.key_size, GFP_USER);
-	if (!rss_config)
-		return -ENOMEM;
-
-	rx_rings.cmd = ETHTOOL_GRXRINGS;
-	ret = ops->get_rxnfc(dev, &rx_rings, NULL);
-	if (ret)
-		goto out;
-
-	/* rxfh.indir_size == 0 means reset the indir table to default (master
-	 * context) or delete the context (other RSS contexts).
-	 * rxfh.indir_size == ETH_RXFH_INDIR_NO_CHANGE means leave it unchanged.
-	 */
-	if (rxfh.indir_size &&
-	    rxfh.indir_size != ETH_RXFH_INDIR_NO_CHANGE) {
-		indir = (u32 *)rss_config;
-		ret = ethtool_copy_validate_indir(indir,
-						  useraddr + rss_cfg_offset,
-						  &rx_rings,
-						  rxfh.indir_size);
-		if (ret)
-			goto out;
-	} else if (rxfh.indir_size == 0) {
-		if (rxfh.rss_context == 0) {
-			indir = (u32 *)rss_config;
-			for (i = 0; i < dev_indir_size; i++)
-				indir[i] = ethtool_rxfh_indir_default(i, rx_rings.data);
-		} else {
-			delete = true;
-		}
-	}
-
-	if (rxfh.key_size) {
-		hkey = rss_config + indir_bytes;
-		if (copy_from_user(hkey,
-				   useraddr + rss_cfg_offset + indir_bytes,
-				   rxfh.key_size)) {
-			ret = -EFAULT;
-			goto out;
-		}
-	}
-
-	if (rxfh.rss_context)
-		ret = ops->set_rxfh_context(dev, indir, hkey, rxfh.hfunc,
-					    &rxfh.rss_context, delete);
-	else
-		ret = ops->set_rxfh(dev, indir, hkey, rxfh.hfunc);
-	if (ret)
-		goto out;
-
-	if (copy_to_user(useraddr + offsetof(struct ethtool_rxfh, rss_context),
-			 &rxfh.rss_context, sizeof(rxfh.rss_context)))
-		ret = -EFAULT;
-
-	if (!rxfh.rss_context) {
-		/* indicate whether rxfh was set to default */
-		if (rxfh.indir_size == 0)
-			dev->priv_flags &= ~IFF_RXFH_CONFIGURED;
-		else if (rxfh.indir_size != ETH_RXFH_INDIR_NO_CHANGE)
-			dev->priv_flags |= IFF_RXFH_CONFIGURED;
-	}
-
-out:
-	kfree(rss_config);
-	return ret;
-}
-
-static int ethtool_get_regs(struct net_device *dev, char __user *useraddr)
-{
-	struct ethtool_regs regs;
-	const struct ethtool_ops *ops = dev->ethtool_ops;
-	void *regbuf;
-	int reglen, ret;
-
-	if (!ops->get_regs || !ops->get_regs_len)
-		return -EOPNOTSUPP;
-
-	if (copy_from_user(&regs, useraddr, sizeof(regs)))
-		return -EFAULT;
-
-	reglen = ops->get_regs_len(dev);
-	if (reglen <= 0)
-		return reglen;
-
-	if (regs.len > reglen)
-		regs.len = reglen;
-
-	regbuf = vzalloc(reglen);
-	if (!regbuf)
-		return -ENOMEM;
-
-	if (regs.len < reglen)
-		reglen = regs.len;
-
-	ops->get_regs(dev, &regs, regbuf);
-
-	ret = -EFAULT;
-	if (copy_to_user(useraddr, &regs, sizeof(regs)))
-		goto out;
-	useraddr += offsetof(struct ethtool_regs, data);
-	if (copy_to_user(useraddr, regbuf, reglen))
-		goto out;
-	ret = 0;
-
- out:
-	vfree(regbuf);
-	return ret;
-}
-
-static int ethtool_reset(struct net_device *dev, char __user *useraddr)
-{
-	struct ethtool_value reset;
-	int ret;
-
-	if (!dev->ethtool_ops->reset)
-		return -EOPNOTSUPP;
-
-	if (copy_from_user(&reset, useraddr, sizeof(reset)))
-		return -EFAULT;
-
-	ret = dev->ethtool_ops->reset(dev, &reset.data);
-	if (ret)
-		return ret;
-
-	if (copy_to_user(useraddr, &reset, sizeof(reset)))
-		return -EFAULT;
-	return 0;
-}
-
-static int ethtool_get_wol(struct net_device *dev, char __user *useraddr)
-{
-	struct ethtool_wolinfo wol;
-
-	if (!dev->ethtool_ops->get_wol)
-		return -EOPNOTSUPP;
-
-	memset(&wol, 0, sizeof(struct ethtool_wolinfo));
-	wol.cmd = ETHTOOL_GWOL;
-	dev->ethtool_ops->get_wol(dev, &wol);
-
-	if (copy_to_user(useraddr, &wol, sizeof(wol)))
-		return -EFAULT;
-	return 0;
-}
-
-static int ethtool_set_wol(struct net_device *dev, char __user *useraddr)
-{
-	struct ethtool_wolinfo wol;
-	int ret;
-
-	if (!dev->ethtool_ops->set_wol)
-		return -EOPNOTSUPP;
-
-	if (copy_from_user(&wol, useraddr, sizeof(wol)))
-		return -EFAULT;
-
-	ret = dev->ethtool_ops->set_wol(dev, &wol);
-	if (ret)
-		return ret;
-
-	dev->wol_enabled = !!wol.wolopts;
-
-	return 0;
-}
-
-static int ethtool_get_eee(struct net_device *dev, char __user *useraddr)
-{
-	struct ethtool_eee edata;
-	int rc;
-
-	if (!dev->ethtool_ops->get_eee)
-		return -EOPNOTSUPP;
-
-	memset(&edata, 0, sizeof(struct ethtool_eee));
-	edata.cmd = ETHTOOL_GEEE;
-	rc = dev->ethtool_ops->get_eee(dev, &edata);
-
-	if (rc)
-		return rc;
-
-	if (copy_to_user(useraddr, &edata, sizeof(edata)))
-		return -EFAULT;
-
-	return 0;
-}
-
-static int ethtool_set_eee(struct net_device *dev, char __user *useraddr)
-{
-	struct ethtool_eee edata;
-
-	if (!dev->ethtool_ops->set_eee)
-		return -EOPNOTSUPP;
-
-	if (copy_from_user(&edata, useraddr, sizeof(edata)))
-		return -EFAULT;
-
-	return dev->ethtool_ops->set_eee(dev, &edata);
-}
-
-static int ethtool_nway_reset(struct net_device *dev)
-{
-	if (!dev->ethtool_ops->nway_reset)
-		return -EOPNOTSUPP;
-
-	return dev->ethtool_ops->nway_reset(dev);
-}
-
-static int ethtool_get_link(struct net_device *dev, char __user *useraddr)
-{
-	struct ethtool_value edata = { .cmd = ETHTOOL_GLINK };
-
-	if (!dev->ethtool_ops->get_link)
-		return -EOPNOTSUPP;
-
-	edata.data = netif_running(dev) && dev->ethtool_ops->get_link(dev);
-
-	if (copy_to_user(useraddr, &edata, sizeof(edata)))
-		return -EFAULT;
-	return 0;
-}
-
-static int ethtool_get_any_eeprom(struct net_device *dev, void __user *useraddr,
-				  int (*getter)(struct net_device *,
-						struct ethtool_eeprom *, u8 *),
-				  u32 total_len)
-{
-	struct ethtool_eeprom eeprom;
-	void __user *userbuf = useraddr + sizeof(eeprom);
-	u32 bytes_remaining;
-	u8 *data;
-	int ret = 0;
-
-	if (copy_from_user(&eeprom, useraddr, sizeof(eeprom)))
-		return -EFAULT;
-
-	/* Check for wrap and zero */
-	if (eeprom.offset + eeprom.len <= eeprom.offset)
-		return -EINVAL;
-
-	/* Check for exceeding total eeprom len */
-	if (eeprom.offset + eeprom.len > total_len)
-		return -EINVAL;
-
-	data = kmalloc(PAGE_SIZE, GFP_USER);
-	if (!data)
-		return -ENOMEM;
-
-	bytes_remaining = eeprom.len;
-	while (bytes_remaining > 0) {
-		eeprom.len = min(bytes_remaining, (u32)PAGE_SIZE);
-
-		ret = getter(dev, &eeprom, data);
-		if (ret)
-			break;
-		if (copy_to_user(userbuf, data, eeprom.len)) {
-			ret = -EFAULT;
-			break;
-		}
-		userbuf += eeprom.len;
-		eeprom.offset += eeprom.len;
-		bytes_remaining -= eeprom.len;
-	}
-
-	eeprom.len = userbuf - (useraddr + sizeof(eeprom));
-	eeprom.offset -= eeprom.len;
-	if (copy_to_user(useraddr, &eeprom, sizeof(eeprom)))
-		ret = -EFAULT;
-
-	kfree(data);
-	return ret;
-}
-
-static int ethtool_get_eeprom(struct net_device *dev, void __user *useraddr)
-{
-	const struct ethtool_ops *ops = dev->ethtool_ops;
-
-	if (!ops->get_eeprom || !ops->get_eeprom_len ||
-	    !ops->get_eeprom_len(dev))
-		return -EOPNOTSUPP;
-
-	return ethtool_get_any_eeprom(dev, useraddr, ops->get_eeprom,
-				      ops->get_eeprom_len(dev));
-}
-
-static int ethtool_set_eeprom(struct net_device *dev, void __user *useraddr)
-{
-	struct ethtool_eeprom eeprom;
-	const struct ethtool_ops *ops = dev->ethtool_ops;
-	void __user *userbuf = useraddr + sizeof(eeprom);
-	u32 bytes_remaining;
-	u8 *data;
-	int ret = 0;
-
-	if (!ops->set_eeprom || !ops->get_eeprom_len ||
-	    !ops->get_eeprom_len(dev))
-		return -EOPNOTSUPP;
-
-	if (copy_from_user(&eeprom, useraddr, sizeof(eeprom)))
-		return -EFAULT;
-
-	/* Check for wrap and zero */
-	if (eeprom.offset + eeprom.len <= eeprom.offset)
-		return -EINVAL;
-
-	/* Check for exceeding total eeprom len */
-	if (eeprom.offset + eeprom.len > ops->get_eeprom_len(dev))
-		return -EINVAL;
-
-	data = kmalloc(PAGE_SIZE, GFP_USER);
-	if (!data)
-		return -ENOMEM;
-
-	bytes_remaining = eeprom.len;
-	while (bytes_remaining > 0) {
-		eeprom.len = min(bytes_remaining, (u32)PAGE_SIZE);
-
-		if (copy_from_user(data, userbuf, eeprom.len)) {
-			ret = -EFAULT;
-			break;
-		}
-		ret = ops->set_eeprom(dev, &eeprom, data);
-		if (ret)
-			break;
-		userbuf += eeprom.len;
-		eeprom.offset += eeprom.len;
-		bytes_remaining -= eeprom.len;
-	}
-
-	kfree(data);
-	return ret;
-}
-
-static noinline_for_stack int ethtool_get_coalesce(struct net_device *dev,
-						   void __user *useraddr)
-{
-	struct ethtool_coalesce coalesce = { .cmd = ETHTOOL_GCOALESCE };
-
-	if (!dev->ethtool_ops->get_coalesce)
-		return -EOPNOTSUPP;
-
-	dev->ethtool_ops->get_coalesce(dev, &coalesce);
-
-	if (copy_to_user(useraddr, &coalesce, sizeof(coalesce)))
-		return -EFAULT;
-	return 0;
-}
-
-static noinline_for_stack int ethtool_set_coalesce(struct net_device *dev,
-						   void __user *useraddr)
-{
-	struct ethtool_coalesce coalesce;
-
-	if (!dev->ethtool_ops->set_coalesce)
-		return -EOPNOTSUPP;
-
-	if (copy_from_user(&coalesce, useraddr, sizeof(coalesce)))
-		return -EFAULT;
-
-	return dev->ethtool_ops->set_coalesce(dev, &coalesce);
-}
-
-static int ethtool_get_ringparam(struct net_device *dev, void __user *useraddr)
-{
-	struct ethtool_ringparam ringparam = { .cmd = ETHTOOL_GRINGPARAM };
-
-	if (!dev->ethtool_ops->get_ringparam)
-		return -EOPNOTSUPP;
-
-	dev->ethtool_ops->get_ringparam(dev, &ringparam);
-
-	if (copy_to_user(useraddr, &ringparam, sizeof(ringparam)))
-		return -EFAULT;
-	return 0;
-}
-
-static int ethtool_set_ringparam(struct net_device *dev, void __user *useraddr)
-{
-	struct ethtool_ringparam ringparam, max = { .cmd = ETHTOOL_GRINGPARAM };
-
-	if (!dev->ethtool_ops->set_ringparam || !dev->ethtool_ops->get_ringparam)
-		return -EOPNOTSUPP;
-
-	if (copy_from_user(&ringparam, useraddr, sizeof(ringparam)))
-		return -EFAULT;
-
-	dev->ethtool_ops->get_ringparam(dev, &max);
-
-	/* ensure new ring parameters are within the maximums */
-	if (ringparam.rx_pending > max.rx_max_pending ||
-	    ringparam.rx_mini_pending > max.rx_mini_max_pending ||
-	    ringparam.rx_jumbo_pending > max.rx_jumbo_max_pending ||
-	    ringparam.tx_pending > max.tx_max_pending)
-		return -EINVAL;
-
-	return dev->ethtool_ops->set_ringparam(dev, &ringparam);
-}
-
-static noinline_for_stack int ethtool_get_channels(struct net_device *dev,
-						   void __user *useraddr)
-{
-	struct ethtool_channels channels = { .cmd = ETHTOOL_GCHANNELS };
-
-	if (!dev->ethtool_ops->get_channels)
-		return -EOPNOTSUPP;
-
-	dev->ethtool_ops->get_channels(dev, &channels);
-
-	if (copy_to_user(useraddr, &channels, sizeof(channels)))
-		return -EFAULT;
-	return 0;
-}
-
-static noinline_for_stack int ethtool_set_channels(struct net_device *dev,
-						   void __user *useraddr)
-{
-	struct ethtool_channels channels, curr = { .cmd = ETHTOOL_GCHANNELS };
-	u16 from_channel, to_channel;
-	u32 max_rx_in_use = 0;
-	unsigned int i;
-
-	if (!dev->ethtool_ops->set_channels || !dev->ethtool_ops->get_channels)
-		return -EOPNOTSUPP;
-
-	if (copy_from_user(&channels, useraddr, sizeof(channels)))
-		return -EFAULT;
-
-	dev->ethtool_ops->get_channels(dev, &curr);
-
-	/* ensure new counts are within the maximums */
-	if (channels.rx_count > curr.max_rx ||
-	    channels.tx_count > curr.max_tx ||
-	    channels.combined_count > curr.max_combined ||
-	    channels.other_count > curr.max_other)
-		return -EINVAL;
-
-	/* ensure the new Rx count fits within the configured Rx flow
-	 * indirection table settings */
-	if (netif_is_rxfh_configured(dev) &&
-	    !ethtool_get_max_rxfh_channel(dev, &max_rx_in_use) &&
-	    (channels.combined_count + channels.rx_count) <= max_rx_in_use)
-	    return -EINVAL;
-
-	/* Disabling channels, query zero-copy AF_XDP sockets */
-	from_channel = channels.combined_count +
-		min(channels.rx_count, channels.tx_count);
-	to_channel = curr.combined_count + max(curr.rx_count, curr.tx_count);
-	for (i = from_channel; i < to_channel; i++)
-		if (xdp_get_umem_from_qid(dev, i))
-			return -EINVAL;
-
-	return dev->ethtool_ops->set_channels(dev, &channels);
-}
-
-static int ethtool_get_pauseparam(struct net_device *dev, void __user *useraddr)
-{
-	struct ethtool_pauseparam pauseparam = { .cmd = ETHTOOL_GPAUSEPARAM };
-
-	if (!dev->ethtool_ops->get_pauseparam)
-		return -EOPNOTSUPP;
-
-	dev->ethtool_ops->get_pauseparam(dev, &pauseparam);
-
-	if (copy_to_user(useraddr, &pauseparam, sizeof(pauseparam)))
-		return -EFAULT;
-	return 0;
-}
-
-static int ethtool_set_pauseparam(struct net_device *dev, void __user *useraddr)
-{
-	struct ethtool_pauseparam pauseparam;
-
-	if (!dev->ethtool_ops->set_pauseparam)
-		return -EOPNOTSUPP;
-
-	if (copy_from_user(&pauseparam, useraddr, sizeof(pauseparam)))
-		return -EFAULT;
-
-	return dev->ethtool_ops->set_pauseparam(dev, &pauseparam);
-}
-
-static int ethtool_self_test(struct net_device *dev, char __user *useraddr)
-{
-	struct ethtool_test test;
-	const struct ethtool_ops *ops = dev->ethtool_ops;
-	u64 *data;
-	int ret, test_len;
-
-	if (!ops->self_test || !ops->get_sset_count)
-		return -EOPNOTSUPP;
-
-	test_len = ops->get_sset_count(dev, ETH_SS_TEST);
-	if (test_len < 0)
-		return test_len;
-	WARN_ON(test_len == 0);
-
-	if (copy_from_user(&test, useraddr, sizeof(test)))
-		return -EFAULT;
-
-	test.len = test_len;
-	data = kmalloc_array(test_len, sizeof(u64), GFP_USER);
-	if (!data)
-		return -ENOMEM;
-
-	ops->self_test(dev, &test, data);
-
-	ret = -EFAULT;
-	if (copy_to_user(useraddr, &test, sizeof(test)))
-		goto out;
-	useraddr += sizeof(test);
-	if (copy_to_user(useraddr, data, test.len * sizeof(u64)))
-		goto out;
-	ret = 0;
-
- out:
-	kfree(data);
-	return ret;
-}
-
-static int ethtool_get_strings(struct net_device *dev, void __user *useraddr)
-{
-	struct ethtool_gstrings gstrings;
-	u8 *data;
-	int ret;
-
-	if (copy_from_user(&gstrings, useraddr, sizeof(gstrings)))
-		return -EFAULT;
-
-	ret = __ethtool_get_sset_count(dev, gstrings.string_set);
-	if (ret < 0)
-		return ret;
-	if (ret > S32_MAX / ETH_GSTRING_LEN)
-		return -ENOMEM;
-	WARN_ON_ONCE(!ret);
-
-	gstrings.len = ret;
-
-	if (gstrings.len) {
-		data = vzalloc(array_size(gstrings.len, ETH_GSTRING_LEN));
-		if (!data)
-			return -ENOMEM;
-
-		__ethtool_get_strings(dev, gstrings.string_set, data);
-	} else {
-		data = NULL;
-	}
-
-	ret = -EFAULT;
-	if (copy_to_user(useraddr, &gstrings, sizeof(gstrings)))
-		goto out;
-	useraddr += sizeof(gstrings);
-	if (gstrings.len &&
-	    copy_to_user(useraddr, data, gstrings.len * ETH_GSTRING_LEN))
-		goto out;
-	ret = 0;
-
-out:
-	vfree(data);
-	return ret;
-}
-
-static int ethtool_phys_id(struct net_device *dev, void __user *useraddr)
-{
-	struct ethtool_value id;
-	static bool busy;
-	const struct ethtool_ops *ops = dev->ethtool_ops;
-	int rc;
-
-	if (!ops->set_phys_id)
-		return -EOPNOTSUPP;
-
-	if (busy)
-		return -EBUSY;
-
-	if (copy_from_user(&id, useraddr, sizeof(id)))
-		return -EFAULT;
-
-	rc = ops->set_phys_id(dev, ETHTOOL_ID_ACTIVE);
-	if (rc < 0)
-		return rc;
-
-	/* Drop the RTNL lock while waiting, but prevent reentry or
-	 * removal of the device.
-	 */
-	busy = true;
-	dev_hold(dev);
-	rtnl_unlock();
-
-	if (rc == 0) {
-		/* Driver will handle this itself */
-		schedule_timeout_interruptible(
-			id.data ? (id.data * HZ) : MAX_SCHEDULE_TIMEOUT);
-	} else {
-		/* Driver expects to be called at twice the frequency in rc */
-		int n = rc * 2, i, interval = HZ / n;
-
-		/* Count down seconds */
-		do {
-			/* Count down iterations per second */
-			i = n;
-			do {
-				rtnl_lock();
-				rc = ops->set_phys_id(dev,
-				    (i & 1) ? ETHTOOL_ID_OFF : ETHTOOL_ID_ON);
-				rtnl_unlock();
-				if (rc)
-					break;
-				schedule_timeout_interruptible(interval);
-			} while (!signal_pending(current) && --i != 0);
-		} while (!signal_pending(current) &&
-			 (id.data == 0 || --id.data != 0));
-	}
-
-	rtnl_lock();
-	dev_put(dev);
-	busy = false;
-
-	(void) ops->set_phys_id(dev, ETHTOOL_ID_INACTIVE);
-	return rc;
-}
-
-static int ethtool_get_stats(struct net_device *dev, void __user *useraddr)
-{
-	struct ethtool_stats stats;
-	const struct ethtool_ops *ops = dev->ethtool_ops;
-	u64 *data;
-	int ret, n_stats;
-
-	if (!ops->get_ethtool_stats || !ops->get_sset_count)
-		return -EOPNOTSUPP;
-
-	n_stats = ops->get_sset_count(dev, ETH_SS_STATS);
-	if (n_stats < 0)
-		return n_stats;
-	if (n_stats > S32_MAX / sizeof(u64))
-		return -ENOMEM;
-	WARN_ON_ONCE(!n_stats);
-	if (copy_from_user(&stats, useraddr, sizeof(stats)))
-		return -EFAULT;
-
-	stats.n_stats = n_stats;
-
-	if (n_stats) {
-		data = vzalloc(array_size(n_stats, sizeof(u64)));
-		if (!data)
-			return -ENOMEM;
-		ops->get_ethtool_stats(dev, &stats, data);
-	} else {
-		data = NULL;
-	}
-
-	ret = -EFAULT;
-	if (copy_to_user(useraddr, &stats, sizeof(stats)))
-		goto out;
-	useraddr += sizeof(stats);
-	if (n_stats && copy_to_user(useraddr, data, n_stats * sizeof(u64)))
-		goto out;
-	ret = 0;
-
- out:
-	vfree(data);
-	return ret;
-}
-
-static int ethtool_get_phy_stats(struct net_device *dev, void __user *useraddr)
-{
-	const struct ethtool_ops *ops = dev->ethtool_ops;
-	struct phy_device *phydev = dev->phydev;
-	struct ethtool_stats stats;
-	u64 *data;
-	int ret, n_stats;
-
-	if (!phydev && (!ops->get_ethtool_phy_stats || !ops->get_sset_count))
-		return -EOPNOTSUPP;
-
-	if (dev->phydev && !ops->get_ethtool_phy_stats)
-		n_stats = phy_ethtool_get_sset_count(dev->phydev);
-	else
-		n_stats = ops->get_sset_count(dev, ETH_SS_PHY_STATS);
-	if (n_stats < 0)
-		return n_stats;
-	if (n_stats > S32_MAX / sizeof(u64))
-		return -ENOMEM;
-	WARN_ON_ONCE(!n_stats);
-
-	if (copy_from_user(&stats, useraddr, sizeof(stats)))
-		return -EFAULT;
-
-	stats.n_stats = n_stats;
-
-	if (n_stats) {
-		data = vzalloc(array_size(n_stats, sizeof(u64)));
-		if (!data)
-			return -ENOMEM;
-
-		if (dev->phydev && !ops->get_ethtool_phy_stats) {
-			ret = phy_ethtool_get_stats(dev->phydev, &stats, data);
-			if (ret < 0)
-				goto out;
-		} else {
-			ops->get_ethtool_phy_stats(dev, &stats, data);
-		}
-	} else {
-		data = NULL;
-	}
-
-	ret = -EFAULT;
-	if (copy_to_user(useraddr, &stats, sizeof(stats)))
-		goto out;
-	useraddr += sizeof(stats);
-	if (n_stats && copy_to_user(useraddr, data, n_stats * sizeof(u64)))
-		goto out;
-	ret = 0;
-
- out:
-	vfree(data);
-	return ret;
-}
-
-static int ethtool_get_perm_addr(struct net_device *dev, void __user *useraddr)
-{
-	struct ethtool_perm_addr epaddr;
-
-	if (copy_from_user(&epaddr, useraddr, sizeof(epaddr)))
-		return -EFAULT;
-
-	if (epaddr.size < dev->addr_len)
-		return -ETOOSMALL;
-	epaddr.size = dev->addr_len;
-
-	if (copy_to_user(useraddr, &epaddr, sizeof(epaddr)))
-		return -EFAULT;
-	useraddr += sizeof(epaddr);
-	if (copy_to_user(useraddr, dev->perm_addr, epaddr.size))
-		return -EFAULT;
-	return 0;
-}
-
-static int ethtool_get_value(struct net_device *dev, char __user *useraddr,
-			     u32 cmd, u32 (*actor)(struct net_device *))
-{
-	struct ethtool_value edata = { .cmd = cmd };
-
-	if (!actor)
-		return -EOPNOTSUPP;
-
-	edata.data = actor(dev);
-
-	if (copy_to_user(useraddr, &edata, sizeof(edata)))
-		return -EFAULT;
-	return 0;
-}
-
-static int ethtool_set_value_void(struct net_device *dev, char __user *useraddr,
-			     void (*actor)(struct net_device *, u32))
-{
-	struct ethtool_value edata;
-
-	if (!actor)
-		return -EOPNOTSUPP;
-
-	if (copy_from_user(&edata, useraddr, sizeof(edata)))
-		return -EFAULT;
-
-	actor(dev, edata.data);
-	return 0;
-}
-
-static int ethtool_set_value(struct net_device *dev, char __user *useraddr,
-			     int (*actor)(struct net_device *, u32))
-{
-	struct ethtool_value edata;
-
-	if (!actor)
-		return -EOPNOTSUPP;
-
-	if (copy_from_user(&edata, useraddr, sizeof(edata)))
-		return -EFAULT;
-
-	return actor(dev, edata.data);
-}
-
-static noinline_for_stack int ethtool_flash_device(struct net_device *dev,
-						   char __user *useraddr)
-{
-	struct ethtool_flash efl;
-
-	if (copy_from_user(&efl, useraddr, sizeof(efl)))
-		return -EFAULT;
-	efl.data[ETHTOOL_FLASH_MAX_FILENAME - 1] = 0;
-
-	if (!dev->ethtool_ops->flash_device)
-		return devlink_compat_flash_update(dev, efl.data);
-
-	return dev->ethtool_ops->flash_device(dev, &efl);
-}
-
-static int ethtool_set_dump(struct net_device *dev,
-			void __user *useraddr)
-{
-	struct ethtool_dump dump;
-
-	if (!dev->ethtool_ops->set_dump)
-		return -EOPNOTSUPP;
-
-	if (copy_from_user(&dump, useraddr, sizeof(dump)))
-		return -EFAULT;
-
-	return dev->ethtool_ops->set_dump(dev, &dump);
-}
-
-static int ethtool_get_dump_flag(struct net_device *dev,
-				void __user *useraddr)
-{
-	int ret;
-	struct ethtool_dump dump;
-	const struct ethtool_ops *ops = dev->ethtool_ops;
-
-	if (!ops->get_dump_flag)
-		return -EOPNOTSUPP;
-
-	if (copy_from_user(&dump, useraddr, sizeof(dump)))
-		return -EFAULT;
-
-	ret = ops->get_dump_flag(dev, &dump);
-	if (ret)
-		return ret;
-
-	if (copy_to_user(useraddr, &dump, sizeof(dump)))
-		return -EFAULT;
-	return 0;
-}
-
-static int ethtool_get_dump_data(struct net_device *dev,
-				void __user *useraddr)
-{
-	int ret;
-	__u32 len;
-	struct ethtool_dump dump, tmp;
-	const struct ethtool_ops *ops = dev->ethtool_ops;
-	void *data = NULL;
-
-	if (!ops->get_dump_data || !ops->get_dump_flag)
-		return -EOPNOTSUPP;
-
-	if (copy_from_user(&dump, useraddr, sizeof(dump)))
-		return -EFAULT;
-
-	memset(&tmp, 0, sizeof(tmp));
-	tmp.cmd = ETHTOOL_GET_DUMP_FLAG;
-	ret = ops->get_dump_flag(dev, &tmp);
-	if (ret)
-		return ret;
-
-	len = min(tmp.len, dump.len);
-	if (!len)
-		return -EFAULT;
-
-	/* Don't ever let the driver think there's more space available
-	 * than it requested with .get_dump_flag().
-	 */
-	dump.len = len;
-
-	/* Always allocate enough space to hold the whole thing so that the
-	 * driver does not need to check the length and bother with partial
-	 * dumping.
-	 */
-	data = vzalloc(tmp.len);
-	if (!data)
-		return -ENOMEM;
-	ret = ops->get_dump_data(dev, &dump, data);
-	if (ret)
-		goto out;
-
-	/* There are two sane possibilities:
-	 * 1. The driver's .get_dump_data() does not touch dump.len.
-	 * 2. Or it may set dump.len to how much it really writes, which
-	 *    should be tmp.len (or len if it can do a partial dump).
-	 * In any case respond to userspace with the actual length of data
-	 * it's receiving.
-	 */
-	WARN_ON(dump.len != len && dump.len != tmp.len);
-	dump.len = len;
-
-	if (copy_to_user(useraddr, &dump, sizeof(dump))) {
-		ret = -EFAULT;
-		goto out;
-	}
-	useraddr += offsetof(struct ethtool_dump, data);
-	if (copy_to_user(useraddr, data, len))
-		ret = -EFAULT;
-out:
-	vfree(data);
-	return ret;
-}
-
-static int ethtool_get_ts_info(struct net_device *dev, void __user *useraddr)
-{
-	int err = 0;
-	struct ethtool_ts_info info;
-	const struct ethtool_ops *ops = dev->ethtool_ops;
-	struct phy_device *phydev = dev->phydev;
-
-	memset(&info, 0, sizeof(info));
-	info.cmd = ETHTOOL_GET_TS_INFO;
-
-	if (phydev && phydev->drv && phydev->drv->ts_info) {
-		err = phydev->drv->ts_info(phydev, &info);
-	} else if (ops->get_ts_info) {
-		err = ops->get_ts_info(dev, &info);
-	} else {
-		info.so_timestamping =
-			SOF_TIMESTAMPING_RX_SOFTWARE |
-			SOF_TIMESTAMPING_SOFTWARE;
-		info.phc_index = -1;
-	}
-
-	if (err)
-		return err;
-
-	if (copy_to_user(useraddr, &info, sizeof(info)))
-		err = -EFAULT;
-
-	return err;
-}
-
-static int __ethtool_get_module_info(struct net_device *dev,
-				     struct ethtool_modinfo *modinfo)
-{
-	const struct ethtool_ops *ops = dev->ethtool_ops;
-	struct phy_device *phydev = dev->phydev;
-
-	if (dev->sfp_bus)
-		return sfp_get_module_info(dev->sfp_bus, modinfo);
-
-	if (phydev && phydev->drv && phydev->drv->module_info)
-		return phydev->drv->module_info(phydev, modinfo);
-
-	if (ops->get_module_info)
-		return ops->get_module_info(dev, modinfo);
-
-	return -EOPNOTSUPP;
-}
-
-static int ethtool_get_module_info(struct net_device *dev,
-				   void __user *useraddr)
-{
-	int ret;
-	struct ethtool_modinfo modinfo;
-
-	if (copy_from_user(&modinfo, useraddr, sizeof(modinfo)))
-		return -EFAULT;
-
-	ret = __ethtool_get_module_info(dev, &modinfo);
-	if (ret)
-		return ret;
-
-	if (copy_to_user(useraddr, &modinfo, sizeof(modinfo)))
-		return -EFAULT;
-
-	return 0;
-}
-
-static int __ethtool_get_module_eeprom(struct net_device *dev,
-				       struct ethtool_eeprom *ee, u8 *data)
-{
-	const struct ethtool_ops *ops = dev->ethtool_ops;
-	struct phy_device *phydev = dev->phydev;
-
-	if (dev->sfp_bus)
-		return sfp_get_module_eeprom(dev->sfp_bus, ee, data);
-
-	if (phydev && phydev->drv && phydev->drv->module_eeprom)
-		return phydev->drv->module_eeprom(phydev, ee, data);
-
-	if (ops->get_module_eeprom)
-		return ops->get_module_eeprom(dev, ee, data);
-
-	return -EOPNOTSUPP;
-}
-
-static int ethtool_get_module_eeprom(struct net_device *dev,
-				     void __user *useraddr)
-{
-	int ret;
-	struct ethtool_modinfo modinfo;
-
-	ret = __ethtool_get_module_info(dev, &modinfo);
-	if (ret)
-		return ret;
-
-	return ethtool_get_any_eeprom(dev, useraddr,
-				      __ethtool_get_module_eeprom,
-				      modinfo.eeprom_len);
-}
-
-static int ethtool_tunable_valid(const struct ethtool_tunable *tuna)
-{
-	switch (tuna->id) {
-	case ETHTOOL_RX_COPYBREAK:
-	case ETHTOOL_TX_COPYBREAK:
-		if (tuna->len != sizeof(u32) ||
-		    tuna->type_id != ETHTOOL_TUNABLE_U32)
-			return -EINVAL;
-		break;
-	case ETHTOOL_PFC_PREVENTION_TOUT:
-		if (tuna->len != sizeof(u16) ||
-		    tuna->type_id != ETHTOOL_TUNABLE_U16)
-			return -EINVAL;
-		break;
-	default:
-		return -EINVAL;
-	}
-
-	return 0;
-}
-
-static int ethtool_get_tunable(struct net_device *dev, void __user *useraddr)
-{
-	int ret;
-	struct ethtool_tunable tuna;
-	const struct ethtool_ops *ops = dev->ethtool_ops;
-	void *data;
-
-	if (!ops->get_tunable)
-		return -EOPNOTSUPP;
-	if (copy_from_user(&tuna, useraddr, sizeof(tuna)))
-		return -EFAULT;
-	ret = ethtool_tunable_valid(&tuna);
-	if (ret)
-		return ret;
-	data = kmalloc(tuna.len, GFP_USER);
-	if (!data)
-		return -ENOMEM;
-	ret = ops->get_tunable(dev, &tuna, data);
-	if (ret)
-		goto out;
-	useraddr += sizeof(tuna);
-	ret = -EFAULT;
-	if (copy_to_user(useraddr, data, tuna.len))
-		goto out;
-	ret = 0;
-
-out:
-	kfree(data);
-	return ret;
-}
-
-static int ethtool_set_tunable(struct net_device *dev, void __user *useraddr)
-{
-	int ret;
-	struct ethtool_tunable tuna;
-	const struct ethtool_ops *ops = dev->ethtool_ops;
-	void *data;
-
-	if (!ops->set_tunable)
-		return -EOPNOTSUPP;
-	if (copy_from_user(&tuna, useraddr, sizeof(tuna)))
-		return -EFAULT;
-	ret = ethtool_tunable_valid(&tuna);
-	if (ret)
-		return ret;
-	useraddr += sizeof(tuna);
-	data = memdup_user(useraddr, tuna.len);
-	if (IS_ERR(data))
-		return PTR_ERR(data);
-	ret = ops->set_tunable(dev, &tuna, data);
-
-	kfree(data);
-	return ret;
-}
-
-static noinline_for_stack int
-ethtool_get_per_queue_coalesce(struct net_device *dev,
-			       void __user *useraddr,
-			       struct ethtool_per_queue_op *per_queue_opt)
-{
-	u32 bit;
-	int ret;
-	DECLARE_BITMAP(queue_mask, MAX_NUM_QUEUE);
-
-	if (!dev->ethtool_ops->get_per_queue_coalesce)
-		return -EOPNOTSUPP;
-
-	useraddr += sizeof(*per_queue_opt);
-
-	bitmap_from_arr32(queue_mask, per_queue_opt->queue_mask,
-			  MAX_NUM_QUEUE);
-
-	for_each_set_bit(bit, queue_mask, MAX_NUM_QUEUE) {
-		struct ethtool_coalesce coalesce = { .cmd = ETHTOOL_GCOALESCE };
-
-		ret = dev->ethtool_ops->get_per_queue_coalesce(dev, bit, &coalesce);
-		if (ret != 0)
-			return ret;
-		if (copy_to_user(useraddr, &coalesce, sizeof(coalesce)))
-			return -EFAULT;
-		useraddr += sizeof(coalesce);
-	}
-
-	return 0;
-}
-
-static noinline_for_stack int
-ethtool_set_per_queue_coalesce(struct net_device *dev,
-			       void __user *useraddr,
-			       struct ethtool_per_queue_op *per_queue_opt)
-{
-	u32 bit;
-	int i, ret = 0;
-	int n_queue;
-	struct ethtool_coalesce *backup = NULL, *tmp = NULL;
-	DECLARE_BITMAP(queue_mask, MAX_NUM_QUEUE);
-
-	if ((!dev->ethtool_ops->set_per_queue_coalesce) ||
-	    (!dev->ethtool_ops->get_per_queue_coalesce))
-		return -EOPNOTSUPP;
-
-	useraddr += sizeof(*per_queue_opt);
-
-	bitmap_from_arr32(queue_mask, per_queue_opt->queue_mask, MAX_NUM_QUEUE);
-	n_queue = bitmap_weight(queue_mask, MAX_NUM_QUEUE);
-	tmp = backup = kmalloc_array(n_queue, sizeof(*backup), GFP_KERNEL);
-	if (!backup)
-		return -ENOMEM;
-
-	for_each_set_bit(bit, queue_mask, MAX_NUM_QUEUE) {
-		struct ethtool_coalesce coalesce;
-
-		ret = dev->ethtool_ops->get_per_queue_coalesce(dev, bit, tmp);
-		if (ret != 0)
-			goto roll_back;
-
-		tmp++;
-
-		if (copy_from_user(&coalesce, useraddr, sizeof(coalesce))) {
-			ret = -EFAULT;
-			goto roll_back;
-		}
-
-		ret = dev->ethtool_ops->set_per_queue_coalesce(dev, bit, &coalesce);
-		if (ret != 0)
-			goto roll_back;
-
-		useraddr += sizeof(coalesce);
-	}
-
-roll_back:
-	if (ret != 0) {
-		tmp = backup;
-		for_each_set_bit(i, queue_mask, bit) {
-			dev->ethtool_ops->set_per_queue_coalesce(dev, i, tmp);
-			tmp++;
-		}
-	}
-	kfree(backup);
-
-	return ret;
-}
-
-static int noinline_for_stack ethtool_set_per_queue(struct net_device *dev,
-				 void __user *useraddr, u32 sub_cmd)
-{
-	struct ethtool_per_queue_op per_queue_opt;
-
-	if (copy_from_user(&per_queue_opt, useraddr, sizeof(per_queue_opt)))
-		return -EFAULT;
-
-	if (per_queue_opt.sub_command != sub_cmd)
-		return -EINVAL;
-
-	switch (per_queue_opt.sub_command) {
-	case ETHTOOL_GCOALESCE:
-		return ethtool_get_per_queue_coalesce(dev, useraddr, &per_queue_opt);
-	case ETHTOOL_SCOALESCE:
-		return ethtool_set_per_queue_coalesce(dev, useraddr, &per_queue_opt);
-	default:
-		return -EOPNOTSUPP;
-	};
-}
-
-static int ethtool_phy_tunable_valid(const struct ethtool_tunable *tuna)
-{
-	switch (tuna->id) {
-	case ETHTOOL_PHY_DOWNSHIFT:
-	case ETHTOOL_PHY_FAST_LINK_DOWN:
-		if (tuna->len != sizeof(u8) ||
-		    tuna->type_id != ETHTOOL_TUNABLE_U8)
-			return -EINVAL;
-		break;
-	case ETHTOOL_PHY_EDPD:
-		if (tuna->len != sizeof(u16) ||
-		    tuna->type_id != ETHTOOL_TUNABLE_U16)
-			return -EINVAL;
-		break;
-	default:
-		return -EINVAL;
-	}
-
-	return 0;
-}
-
-static int get_phy_tunable(struct net_device *dev, void __user *useraddr)
-{
-	int ret;
-	struct ethtool_tunable tuna;
-	struct phy_device *phydev = dev->phydev;
-	void *data;
-
-	if (!(phydev && phydev->drv && phydev->drv->get_tunable))
-		return -EOPNOTSUPP;
-
-	if (copy_from_user(&tuna, useraddr, sizeof(tuna)))
-		return -EFAULT;
-	ret = ethtool_phy_tunable_valid(&tuna);
-	if (ret)
-		return ret;
-	data = kmalloc(tuna.len, GFP_USER);
-	if (!data)
-		return -ENOMEM;
-	mutex_lock(&phydev->lock);
-	ret = phydev->drv->get_tunable(phydev, &tuna, data);
-	mutex_unlock(&phydev->lock);
-	if (ret)
-		goto out;
-	useraddr += sizeof(tuna);
-	ret = -EFAULT;
-	if (copy_to_user(useraddr, data, tuna.len))
-		goto out;
-	ret = 0;
-
-out:
-	kfree(data);
-	return ret;
-}
-
-static int set_phy_tunable(struct net_device *dev, void __user *useraddr)
-{
-	int ret;
-	struct ethtool_tunable tuna;
-	struct phy_device *phydev = dev->phydev;
-	void *data;
-
-	if (!(phydev && phydev->drv && phydev->drv->set_tunable))
-		return -EOPNOTSUPP;
-	if (copy_from_user(&tuna, useraddr, sizeof(tuna)))
-		return -EFAULT;
-	ret = ethtool_phy_tunable_valid(&tuna);
-	if (ret)
-		return ret;
-	useraddr += sizeof(tuna);
-	data = memdup_user(useraddr, tuna.len);
-	if (IS_ERR(data))
-		return PTR_ERR(data);
-	mutex_lock(&phydev->lock);
-	ret = phydev->drv->set_tunable(phydev, &tuna, data);
-	mutex_unlock(&phydev->lock);
-
-	kfree(data);
-	return ret;
-}
-
-static int ethtool_get_fecparam(struct net_device *dev, void __user *useraddr)
-{
-	struct ethtool_fecparam fecparam = { .cmd = ETHTOOL_GFECPARAM };
-	int rc;
-
-	if (!dev->ethtool_ops->get_fecparam)
-		return -EOPNOTSUPP;
-
-	rc = dev->ethtool_ops->get_fecparam(dev, &fecparam);
-	if (rc)
-		return rc;
-
-	if (copy_to_user(useraddr, &fecparam, sizeof(fecparam)))
-		return -EFAULT;
-	return 0;
-}
-
-static int ethtool_set_fecparam(struct net_device *dev, void __user *useraddr)
-{
-	struct ethtool_fecparam fecparam;
-
-	if (!dev->ethtool_ops->set_fecparam)
-		return -EOPNOTSUPP;
-
-	if (copy_from_user(&fecparam, useraddr, sizeof(fecparam)))
-		return -EFAULT;
-
-	return dev->ethtool_ops->set_fecparam(dev, &fecparam);
-}
-
-/* The main entry point in this file.  Called from net/core/dev_ioctl.c */
-
-int dev_ethtool(struct net *net, struct ifreq *ifr)
-{
-	struct net_device *dev = __dev_get_by_name(net, ifr->ifr_name);
-	void __user *useraddr = ifr->ifr_data;
-	u32 ethcmd, sub_cmd;
-	int rc;
-	netdev_features_t old_features;
-
-	if (!dev || !netif_device_present(dev))
-		return -ENODEV;
-
-	if (copy_from_user(&ethcmd, useraddr, sizeof(ethcmd)))
-		return -EFAULT;
-
-	if (ethcmd == ETHTOOL_PERQUEUE) {
-		if (copy_from_user(&sub_cmd, useraddr + sizeof(ethcmd), sizeof(sub_cmd)))
-			return -EFAULT;
-	} else {
-		sub_cmd = ethcmd;
-	}
-	/* Allow some commands to be done by anyone */
-	switch (sub_cmd) {
-	case ETHTOOL_GSET:
-	case ETHTOOL_GDRVINFO:
-	case ETHTOOL_GMSGLVL:
-	case ETHTOOL_GLINK:
-	case ETHTOOL_GCOALESCE:
-	case ETHTOOL_GRINGPARAM:
-	case ETHTOOL_GPAUSEPARAM:
-	case ETHTOOL_GRXCSUM:
-	case ETHTOOL_GTXCSUM:
-	case ETHTOOL_GSG:
-	case ETHTOOL_GSSET_INFO:
-	case ETHTOOL_GSTRINGS:
-	case ETHTOOL_GSTATS:
-	case ETHTOOL_GPHYSTATS:
-	case ETHTOOL_GTSO:
-	case ETHTOOL_GPERMADDR:
-	case ETHTOOL_GUFO:
-	case ETHTOOL_GGSO:
-	case ETHTOOL_GGRO:
-	case ETHTOOL_GFLAGS:
-	case ETHTOOL_GPFLAGS:
-	case ETHTOOL_GRXFH:
-	case ETHTOOL_GRXRINGS:
-	case ETHTOOL_GRXCLSRLCNT:
-	case ETHTOOL_GRXCLSRULE:
-	case ETHTOOL_GRXCLSRLALL:
-	case ETHTOOL_GRXFHINDIR:
-	case ETHTOOL_GRSSH:
-	case ETHTOOL_GFEATURES:
-	case ETHTOOL_GCHANNELS:
-	case ETHTOOL_GET_TS_INFO:
-	case ETHTOOL_GEEE:
-	case ETHTOOL_GTUNABLE:
-	case ETHTOOL_PHY_GTUNABLE:
-	case ETHTOOL_GLINKSETTINGS:
-	case ETHTOOL_GFECPARAM:
-		break;
-	default:
-		if (!ns_capable(net->user_ns, CAP_NET_ADMIN))
-			return -EPERM;
-	}
-
-	if (dev->ethtool_ops->begin) {
-		rc = dev->ethtool_ops->begin(dev);
-		if (rc  < 0)
-			return rc;
-	}
-	old_features = dev->features;
-
-	switch (ethcmd) {
-	case ETHTOOL_GSET:
-		rc = ethtool_get_settings(dev, useraddr);
-		break;
-	case ETHTOOL_SSET:
-		rc = ethtool_set_settings(dev, useraddr);
-		break;
-	case ETHTOOL_GDRVINFO:
-		rc = ethtool_get_drvinfo(dev, useraddr);
-		break;
-	case ETHTOOL_GREGS:
-		rc = ethtool_get_regs(dev, useraddr);
-		break;
-	case ETHTOOL_GWOL:
-		rc = ethtool_get_wol(dev, useraddr);
-		break;
-	case ETHTOOL_SWOL:
-		rc = ethtool_set_wol(dev, useraddr);
-		break;
-	case ETHTOOL_GMSGLVL:
-		rc = ethtool_get_value(dev, useraddr, ethcmd,
-				       dev->ethtool_ops->get_msglevel);
-		break;
-	case ETHTOOL_SMSGLVL:
-		rc = ethtool_set_value_void(dev, useraddr,
-				       dev->ethtool_ops->set_msglevel);
-		break;
-	case ETHTOOL_GEEE:
-		rc = ethtool_get_eee(dev, useraddr);
-		break;
-	case ETHTOOL_SEEE:
-		rc = ethtool_set_eee(dev, useraddr);
-		break;
-	case ETHTOOL_NWAY_RST:
-		rc = ethtool_nway_reset(dev);
-		break;
-	case ETHTOOL_GLINK:
-		rc = ethtool_get_link(dev, useraddr);
-		break;
-	case ETHTOOL_GEEPROM:
-		rc = ethtool_get_eeprom(dev, useraddr);
-		break;
-	case ETHTOOL_SEEPROM:
-		rc = ethtool_set_eeprom(dev, useraddr);
-		break;
-	case ETHTOOL_GCOALESCE:
-		rc = ethtool_get_coalesce(dev, useraddr);
-		break;
-	case ETHTOOL_SCOALESCE:
-		rc = ethtool_set_coalesce(dev, useraddr);
-		break;
-	case ETHTOOL_GRINGPARAM:
-		rc = ethtool_get_ringparam(dev, useraddr);
-		break;
-	case ETHTOOL_SRINGPARAM:
-		rc = ethtool_set_ringparam(dev, useraddr);
-		break;
-	case ETHTOOL_GPAUSEPARAM:
-		rc = ethtool_get_pauseparam(dev, useraddr);
-		break;
-	case ETHTOOL_SPAUSEPARAM:
-		rc = ethtool_set_pauseparam(dev, useraddr);
-		break;
-	case ETHTOOL_TEST:
-		rc = ethtool_self_test(dev, useraddr);
-		break;
-	case ETHTOOL_GSTRINGS:
-		rc = ethtool_get_strings(dev, useraddr);
-		break;
-	case ETHTOOL_PHYS_ID:
-		rc = ethtool_phys_id(dev, useraddr);
-		break;
-	case ETHTOOL_GSTATS:
-		rc = ethtool_get_stats(dev, useraddr);
-		break;
-	case ETHTOOL_GPERMADDR:
-		rc = ethtool_get_perm_addr(dev, useraddr);
-		break;
-	case ETHTOOL_GFLAGS:
-		rc = ethtool_get_value(dev, useraddr, ethcmd,
-					__ethtool_get_flags);
-		break;
-	case ETHTOOL_SFLAGS:
-		rc = ethtool_set_value(dev, useraddr, __ethtool_set_flags);
-		break;
-	case ETHTOOL_GPFLAGS:
-		rc = ethtool_get_value(dev, useraddr, ethcmd,
-				       dev->ethtool_ops->get_priv_flags);
-		break;
-	case ETHTOOL_SPFLAGS:
-		rc = ethtool_set_value(dev, useraddr,
-				       dev->ethtool_ops->set_priv_flags);
-		break;
-	case ETHTOOL_GRXFH:
-	case ETHTOOL_GRXRINGS:
-	case ETHTOOL_GRXCLSRLCNT:
-	case ETHTOOL_GRXCLSRULE:
-	case ETHTOOL_GRXCLSRLALL:
-		rc = ethtool_get_rxnfc(dev, ethcmd, useraddr);
-		break;
-	case ETHTOOL_SRXFH:
-	case ETHTOOL_SRXCLSRLDEL:
-	case ETHTOOL_SRXCLSRLINS:
-		rc = ethtool_set_rxnfc(dev, ethcmd, useraddr);
-		break;
-	case ETHTOOL_FLASHDEV:
-		rc = ethtool_flash_device(dev, useraddr);
-		break;
-	case ETHTOOL_RESET:
-		rc = ethtool_reset(dev, useraddr);
-		break;
-	case ETHTOOL_GSSET_INFO:
-		rc = ethtool_get_sset_info(dev, useraddr);
-		break;
-	case ETHTOOL_GRXFHINDIR:
-		rc = ethtool_get_rxfh_indir(dev, useraddr);
-		break;
-	case ETHTOOL_SRXFHINDIR:
-		rc = ethtool_set_rxfh_indir(dev, useraddr);
-		break;
-	case ETHTOOL_GRSSH:
-		rc = ethtool_get_rxfh(dev, useraddr);
-		break;
-	case ETHTOOL_SRSSH:
-		rc = ethtool_set_rxfh(dev, useraddr);
-		break;
-	case ETHTOOL_GFEATURES:
-		rc = ethtool_get_features(dev, useraddr);
-		break;
-	case ETHTOOL_SFEATURES:
-		rc = ethtool_set_features(dev, useraddr);
-		break;
-	case ETHTOOL_GTXCSUM:
-	case ETHTOOL_GRXCSUM:
-	case ETHTOOL_GSG:
-	case ETHTOOL_GTSO:
-	case ETHTOOL_GGSO:
-	case ETHTOOL_GGRO:
-		rc = ethtool_get_one_feature(dev, useraddr, ethcmd);
-		break;
-	case ETHTOOL_STXCSUM:
-	case ETHTOOL_SRXCSUM:
-	case ETHTOOL_SSG:
-	case ETHTOOL_STSO:
-	case ETHTOOL_SGSO:
-	case ETHTOOL_SGRO:
-		rc = ethtool_set_one_feature(dev, useraddr, ethcmd);
-		break;
-	case ETHTOOL_GCHANNELS:
-		rc = ethtool_get_channels(dev, useraddr);
-		break;
-	case ETHTOOL_SCHANNELS:
-		rc = ethtool_set_channels(dev, useraddr);
-		break;
-	case ETHTOOL_SET_DUMP:
-		rc = ethtool_set_dump(dev, useraddr);
-		break;
-	case ETHTOOL_GET_DUMP_FLAG:
-		rc = ethtool_get_dump_flag(dev, useraddr);
-		break;
-	case ETHTOOL_GET_DUMP_DATA:
-		rc = ethtool_get_dump_data(dev, useraddr);
-		break;
-	case ETHTOOL_GET_TS_INFO:
-		rc = ethtool_get_ts_info(dev, useraddr);
-		break;
-	case ETHTOOL_GMODULEINFO:
-		rc = ethtool_get_module_info(dev, useraddr);
-		break;
-	case ETHTOOL_GMODULEEEPROM:
-		rc = ethtool_get_module_eeprom(dev, useraddr);
-		break;
-	case ETHTOOL_GTUNABLE:
-		rc = ethtool_get_tunable(dev, useraddr);
-		break;
-	case ETHTOOL_STUNABLE:
-		rc = ethtool_set_tunable(dev, useraddr);
-		break;
-	case ETHTOOL_GPHYSTATS:
-		rc = ethtool_get_phy_stats(dev, useraddr);
-		break;
-	case ETHTOOL_PERQUEUE:
-		rc = ethtool_set_per_queue(dev, useraddr, sub_cmd);
-		break;
-	case ETHTOOL_GLINKSETTINGS:
-		rc = ethtool_get_link_ksettings(dev, useraddr);
-		break;
-	case ETHTOOL_SLINKSETTINGS:
-		rc = ethtool_set_link_ksettings(dev, useraddr);
-		break;
-	case ETHTOOL_PHY_GTUNABLE:
-		rc = get_phy_tunable(dev, useraddr);
-		break;
-	case ETHTOOL_PHY_STUNABLE:
-		rc = set_phy_tunable(dev, useraddr);
-		break;
-	case ETHTOOL_GFECPARAM:
-		rc = ethtool_get_fecparam(dev, useraddr);
-		break;
-	case ETHTOOL_SFECPARAM:
-		rc = ethtool_set_fecparam(dev, useraddr);
-		break;
-	default:
-		rc = -EOPNOTSUPP;
-	}
-
-	if (dev->ethtool_ops->complete)
-		dev->ethtool_ops->complete(dev);
-
-	if (old_features != dev->features)
-		netdev_features_change(dev);
-
-	return rc;
-}
-
-struct ethtool_rx_flow_key {
-	struct flow_dissector_key_basic			basic;
-	union {
-		struct flow_dissector_key_ipv4_addrs	ipv4;
-		struct flow_dissector_key_ipv6_addrs	ipv6;
-	};
-	struct flow_dissector_key_ports			tp;
-	struct flow_dissector_key_ip			ip;
-	struct flow_dissector_key_vlan			vlan;
-	struct flow_dissector_key_eth_addrs		eth_addrs;
-} __aligned(BITS_PER_LONG / 8); /* Ensure that we can do comparisons as longs. */
-
-struct ethtool_rx_flow_match {
-	struct flow_dissector		dissector;
-	struct ethtool_rx_flow_key	key;
-	struct ethtool_rx_flow_key	mask;
-};
-
-struct ethtool_rx_flow_rule *
-ethtool_rx_flow_rule_create(const struct ethtool_rx_flow_spec_input *input)
-{
-	const struct ethtool_rx_flow_spec *fs = input->fs;
-	static struct in6_addr zero_addr = {};
-	struct ethtool_rx_flow_match *match;
-	struct ethtool_rx_flow_rule *flow;
-	struct flow_action_entry *act;
-
-	flow = kzalloc(sizeof(struct ethtool_rx_flow_rule) +
-		       sizeof(struct ethtool_rx_flow_match), GFP_KERNEL);
-	if (!flow)
-		return ERR_PTR(-ENOMEM);
-
-	/* ethtool_rx supports only one single action per rule. */
-	flow->rule = flow_rule_alloc(1);
-	if (!flow->rule) {
-		kfree(flow);
-		return ERR_PTR(-ENOMEM);
-	}
-
-	match = (struct ethtool_rx_flow_match *)flow->priv;
-	flow->rule->match.dissector	= &match->dissector;
-	flow->rule->match.mask		= &match->mask;
-	flow->rule->match.key		= &match->key;
-
-	match->mask.basic.n_proto = htons(0xffff);
-
-	switch (fs->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT | FLOW_RSS)) {
-	case ETHER_FLOW: {
-		const struct ethhdr *ether_spec, *ether_m_spec;
-
-		ether_spec = &fs->h_u.ether_spec;
-		ether_m_spec = &fs->m_u.ether_spec;
-
-		if (!is_zero_ether_addr(ether_m_spec->h_source)) {
-			ether_addr_copy(match->key.eth_addrs.src,
-					ether_spec->h_source);
-			ether_addr_copy(match->mask.eth_addrs.src,
-					ether_m_spec->h_source);
-		}
-		if (!is_zero_ether_addr(ether_m_spec->h_dest)) {
-			ether_addr_copy(match->key.eth_addrs.dst,
-					ether_spec->h_dest);
-			ether_addr_copy(match->mask.eth_addrs.dst,
-					ether_m_spec->h_dest);
-		}
-		if (ether_m_spec->h_proto) {
-			match->key.basic.n_proto = ether_spec->h_proto;
-			match->mask.basic.n_proto = ether_m_spec->h_proto;
-		}
-		}
-		break;
-	case TCP_V4_FLOW:
-	case UDP_V4_FLOW: {
-		const struct ethtool_tcpip4_spec *v4_spec, *v4_m_spec;
-
-		match->key.basic.n_proto = htons(ETH_P_IP);
-
-		v4_spec = &fs->h_u.tcp_ip4_spec;
-		v4_m_spec = &fs->m_u.tcp_ip4_spec;
-
-		if (v4_m_spec->ip4src) {
-			match->key.ipv4.src = v4_spec->ip4src;
-			match->mask.ipv4.src = v4_m_spec->ip4src;
-		}
-		if (v4_m_spec->ip4dst) {
-			match->key.ipv4.dst = v4_spec->ip4dst;
-			match->mask.ipv4.dst = v4_m_spec->ip4dst;
-		}
-		if (v4_m_spec->ip4src ||
-		    v4_m_spec->ip4dst) {
-			match->dissector.used_keys |=
-				BIT(FLOW_DISSECTOR_KEY_IPV4_ADDRS);
-			match->dissector.offset[FLOW_DISSECTOR_KEY_IPV4_ADDRS] =
-				offsetof(struct ethtool_rx_flow_key, ipv4);
-		}
-		if (v4_m_spec->psrc) {
-			match->key.tp.src = v4_spec->psrc;
-			match->mask.tp.src = v4_m_spec->psrc;
-		}
-		if (v4_m_spec->pdst) {
-			match->key.tp.dst = v4_spec->pdst;
-			match->mask.tp.dst = v4_m_spec->pdst;
-		}
-		if (v4_m_spec->psrc ||
-		    v4_m_spec->pdst) {
-			match->dissector.used_keys |=
-				BIT(FLOW_DISSECTOR_KEY_PORTS);
-			match->dissector.offset[FLOW_DISSECTOR_KEY_PORTS] =
-				offsetof(struct ethtool_rx_flow_key, tp);
-		}
-		if (v4_m_spec->tos) {
-			match->key.ip.tos = v4_spec->tos;
-			match->mask.ip.tos = v4_m_spec->tos;
-			match->dissector.used_keys |=
-				BIT(FLOW_DISSECTOR_KEY_IP);
-			match->dissector.offset[FLOW_DISSECTOR_KEY_IP] =
-				offsetof(struct ethtool_rx_flow_key, ip);
-		}
-		}
-		break;
-	case TCP_V6_FLOW:
-	case UDP_V6_FLOW: {
-		const struct ethtool_tcpip6_spec *v6_spec, *v6_m_spec;
-
-		match->key.basic.n_proto = htons(ETH_P_IPV6);
-
-		v6_spec = &fs->h_u.tcp_ip6_spec;
-		v6_m_spec = &fs->m_u.tcp_ip6_spec;
-		if (memcmp(v6_m_spec->ip6src, &zero_addr, sizeof(zero_addr))) {
-			memcpy(&match->key.ipv6.src, v6_spec->ip6src,
-			       sizeof(match->key.ipv6.src));
-			memcpy(&match->mask.ipv6.src, v6_m_spec->ip6src,
-			       sizeof(match->mask.ipv6.src));
-		}
-		if (memcmp(v6_m_spec->ip6dst, &zero_addr, sizeof(zero_addr))) {
-			memcpy(&match->key.ipv6.dst, v6_spec->ip6dst,
-			       sizeof(match->key.ipv6.dst));
-			memcpy(&match->mask.ipv6.dst, v6_m_spec->ip6dst,
-			       sizeof(match->mask.ipv6.dst));
-		}
-		if (memcmp(v6_m_spec->ip6src, &zero_addr, sizeof(zero_addr)) ||
-		    memcmp(v6_m_spec->ip6src, &zero_addr, sizeof(zero_addr))) {
-			match->dissector.used_keys |=
-				BIT(FLOW_DISSECTOR_KEY_IPV6_ADDRS);
-			match->dissector.offset[FLOW_DISSECTOR_KEY_IPV6_ADDRS] =
-				offsetof(struct ethtool_rx_flow_key, ipv6);
-		}
-		if (v6_m_spec->psrc) {
-			match->key.tp.src = v6_spec->psrc;
-			match->mask.tp.src = v6_m_spec->psrc;
-		}
-		if (v6_m_spec->pdst) {
-			match->key.tp.dst = v6_spec->pdst;
-			match->mask.tp.dst = v6_m_spec->pdst;
-		}
-		if (v6_m_spec->psrc ||
-		    v6_m_spec->pdst) {
-			match->dissector.used_keys |=
-				BIT(FLOW_DISSECTOR_KEY_PORTS);
-			match->dissector.offset[FLOW_DISSECTOR_KEY_PORTS] =
-				offsetof(struct ethtool_rx_flow_key, tp);
-		}
-		if (v6_m_spec->tclass) {
-			match->key.ip.tos = v6_spec->tclass;
-			match->mask.ip.tos = v6_m_spec->tclass;
-			match->dissector.used_keys |=
-				BIT(FLOW_DISSECTOR_KEY_IP);
-			match->dissector.offset[FLOW_DISSECTOR_KEY_IP] =
-				offsetof(struct ethtool_rx_flow_key, ip);
-		}
-		}
-		break;
-	default:
-		ethtool_rx_flow_rule_destroy(flow);
-		return ERR_PTR(-EINVAL);
-	}
-
-	switch (fs->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT | FLOW_RSS)) {
-	case TCP_V4_FLOW:
-	case TCP_V6_FLOW:
-		match->key.basic.ip_proto = IPPROTO_TCP;
-		break;
-	case UDP_V4_FLOW:
-	case UDP_V6_FLOW:
-		match->key.basic.ip_proto = IPPROTO_UDP;
-		break;
-	}
-	match->mask.basic.ip_proto = 0xff;
-
-	match->dissector.used_keys |= BIT(FLOW_DISSECTOR_KEY_BASIC);
-	match->dissector.offset[FLOW_DISSECTOR_KEY_BASIC] =
-		offsetof(struct ethtool_rx_flow_key, basic);
-
-	if (fs->flow_type & FLOW_EXT) {
-		const struct ethtool_flow_ext *ext_h_spec = &fs->h_ext;
-		const struct ethtool_flow_ext *ext_m_spec = &fs->m_ext;
-
-		if (ext_m_spec->vlan_etype) {
-			match->key.vlan.vlan_tpid = ext_h_spec->vlan_etype;
-			match->mask.vlan.vlan_tpid = ext_m_spec->vlan_etype;
-		}
-
-		if (ext_m_spec->vlan_tci) {
-			match->key.vlan.vlan_id =
-				ntohs(ext_h_spec->vlan_tci) & 0x0fff;
-			match->mask.vlan.vlan_id =
-				ntohs(ext_m_spec->vlan_tci) & 0x0fff;
-
-			match->key.vlan.vlan_dei =
-				!!(ext_h_spec->vlan_tci & htons(0x1000));
-			match->mask.vlan.vlan_dei =
-				!!(ext_m_spec->vlan_tci & htons(0x1000));
-
-			match->key.vlan.vlan_priority =
-				(ntohs(ext_h_spec->vlan_tci) & 0xe000) >> 13;
-			match->mask.vlan.vlan_priority =
-				(ntohs(ext_m_spec->vlan_tci) & 0xe000) >> 13;
-		}
-
-		if (ext_m_spec->vlan_etype ||
-		    ext_m_spec->vlan_tci) {
-			match->dissector.used_keys |=
-				BIT(FLOW_DISSECTOR_KEY_VLAN);
-			match->dissector.offset[FLOW_DISSECTOR_KEY_VLAN] =
-				offsetof(struct ethtool_rx_flow_key, vlan);
-		}
-	}
-	if (fs->flow_type & FLOW_MAC_EXT) {
-		const struct ethtool_flow_ext *ext_h_spec = &fs->h_ext;
-		const struct ethtool_flow_ext *ext_m_spec = &fs->m_ext;
-
-		memcpy(match->key.eth_addrs.dst, ext_h_spec->h_dest,
-		       ETH_ALEN);
-		memcpy(match->mask.eth_addrs.dst, ext_m_spec->h_dest,
-		       ETH_ALEN);
-
-		match->dissector.used_keys |=
-			BIT(FLOW_DISSECTOR_KEY_ETH_ADDRS);
-		match->dissector.offset[FLOW_DISSECTOR_KEY_ETH_ADDRS] =
-			offsetof(struct ethtool_rx_flow_key, eth_addrs);
-	}
-
-	act = &flow->rule->action.entries[0];
-	switch (fs->ring_cookie) {
-	case RX_CLS_FLOW_DISC:
-		act->id = FLOW_ACTION_DROP;
-		break;
-	case RX_CLS_FLOW_WAKE:
-		act->id = FLOW_ACTION_WAKE;
-		break;
-	default:
-		act->id = FLOW_ACTION_QUEUE;
-		if (fs->flow_type & FLOW_RSS)
-			act->queue.ctx = input->rss_ctx;
-
-		act->queue.vf = ethtool_get_flow_spec_ring_vf(fs->ring_cookie);
-		act->queue.index = ethtool_get_flow_spec_ring(fs->ring_cookie);
-		break;
-	}
-
-	return flow;
-}
-EXPORT_SYMBOL(ethtool_rx_flow_rule_create);
-
-void ethtool_rx_flow_rule_destroy(struct ethtool_rx_flow_rule *flow)
-{
-	kfree(flow->rule);
-	kfree(flow);
-}
-EXPORT_SYMBOL(ethtool_rx_flow_rule_destroy);
diff --git a/net/ethtool/Makefile b/net/ethtool/Makefile
new file mode 100644
index 000000000000..7e5c9eb85c90
--- /dev/null
+++ b/net/ethtool/Makefile
@@ -0,0 +1,3 @@
+# SPDX-License-Identifier: GPL-2.0-only
+
+obj-y		+= ioctl.o
diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
new file mode 100644
index 000000000000..cd9bc67381b2
--- /dev/null
+++ b/net/ethtool/ioctl.c
@@ -0,0 +1,3116 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * net/core/ethtool.c - Ethtool ioctl handler
+ * Copyright (c) 2003 Matthew Wilcox <matthew@wil.cx>
+ *
+ * This file is where we call all the ethtool_ops commands to get
+ * the information ethtool needs.
+ */
+
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/capability.h>
+#include <linux/errno.h>
+#include <linux/ethtool.h>
+#include <linux/netdevice.h>
+#include <linux/net_tstamp.h>
+#include <linux/phy.h>
+#include <linux/bitops.h>
+#include <linux/uaccess.h>
+#include <linux/vmalloc.h>
+#include <linux/sfp.h>
+#include <linux/slab.h>
+#include <linux/rtnetlink.h>
+#include <linux/sched/signal.h>
+#include <linux/net.h>
+#include <net/devlink.h>
+#include <net/xdp_sock.h>
+#include <net/flow_offload.h>
+
+/*
+ * Some useful ethtool_ops methods that're device independent.
+ * If we find that all drivers want to do the same thing here,
+ * we can turn these into dev_() function calls.
+ */
+
+u32 ethtool_op_get_link(struct net_device *dev)
+{
+	return netif_carrier_ok(dev) ? 1 : 0;
+}
+EXPORT_SYMBOL(ethtool_op_get_link);
+
+int ethtool_op_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info)
+{
+	info->so_timestamping =
+		SOF_TIMESTAMPING_TX_SOFTWARE |
+		SOF_TIMESTAMPING_RX_SOFTWARE |
+		SOF_TIMESTAMPING_SOFTWARE;
+	info->phc_index = -1;
+	return 0;
+}
+EXPORT_SYMBOL(ethtool_op_get_ts_info);
+
+/* Handlers for each ethtool command */
+
+#define ETHTOOL_DEV_FEATURE_WORDS	((NETDEV_FEATURE_COUNT + 31) / 32)
+
+static const char netdev_features_strings[NETDEV_FEATURE_COUNT][ETH_GSTRING_LEN] = {
+	[NETIF_F_SG_BIT] =               "tx-scatter-gather",
+	[NETIF_F_IP_CSUM_BIT] =          "tx-checksum-ipv4",
+	[NETIF_F_HW_CSUM_BIT] =          "tx-checksum-ip-generic",
+	[NETIF_F_IPV6_CSUM_BIT] =        "tx-checksum-ipv6",
+	[NETIF_F_HIGHDMA_BIT] =          "highdma",
+	[NETIF_F_FRAGLIST_BIT] =         "tx-scatter-gather-fraglist",
+	[NETIF_F_HW_VLAN_CTAG_TX_BIT] =  "tx-vlan-hw-insert",
+
+	[NETIF_F_HW_VLAN_CTAG_RX_BIT] =  "rx-vlan-hw-parse",
+	[NETIF_F_HW_VLAN_CTAG_FILTER_BIT] = "rx-vlan-filter",
+	[NETIF_F_HW_VLAN_STAG_TX_BIT] =  "tx-vlan-stag-hw-insert",
+	[NETIF_F_HW_VLAN_STAG_RX_BIT] =  "rx-vlan-stag-hw-parse",
+	[NETIF_F_HW_VLAN_STAG_FILTER_BIT] = "rx-vlan-stag-filter",
+	[NETIF_F_VLAN_CHALLENGED_BIT] =  "vlan-challenged",
+	[NETIF_F_GSO_BIT] =              "tx-generic-segmentation",
+	[NETIF_F_LLTX_BIT] =             "tx-lockless",
+	[NETIF_F_NETNS_LOCAL_BIT] =      "netns-local",
+	[NETIF_F_GRO_BIT] =              "rx-gro",
+	[NETIF_F_GRO_HW_BIT] =           "rx-gro-hw",
+	[NETIF_F_LRO_BIT] =              "rx-lro",
+
+	[NETIF_F_TSO_BIT] =              "tx-tcp-segmentation",
+	[NETIF_F_GSO_ROBUST_BIT] =       "tx-gso-robust",
+	[NETIF_F_TSO_ECN_BIT] =          "tx-tcp-ecn-segmentation",
+	[NETIF_F_TSO_MANGLEID_BIT] =	 "tx-tcp-mangleid-segmentation",
+	[NETIF_F_TSO6_BIT] =             "tx-tcp6-segmentation",
+	[NETIF_F_FSO_BIT] =              "tx-fcoe-segmentation",
+	[NETIF_F_GSO_GRE_BIT] =		 "tx-gre-segmentation",
+	[NETIF_F_GSO_GRE_CSUM_BIT] =	 "tx-gre-csum-segmentation",
+	[NETIF_F_GSO_IPXIP4_BIT] =	 "tx-ipxip4-segmentation",
+	[NETIF_F_GSO_IPXIP6_BIT] =	 "tx-ipxip6-segmentation",
+	[NETIF_F_GSO_UDP_TUNNEL_BIT] =	 "tx-udp_tnl-segmentation",
+	[NETIF_F_GSO_UDP_TUNNEL_CSUM_BIT] = "tx-udp_tnl-csum-segmentation",
+	[NETIF_F_GSO_PARTIAL_BIT] =	 "tx-gso-partial",
+	[NETIF_F_GSO_SCTP_BIT] =	 "tx-sctp-segmentation",
+	[NETIF_F_GSO_ESP_BIT] =		 "tx-esp-segmentation",
+	[NETIF_F_GSO_UDP_L4_BIT] =	 "tx-udp-segmentation",
+
+	[NETIF_F_FCOE_CRC_BIT] =         "tx-checksum-fcoe-crc",
+	[NETIF_F_SCTP_CRC_BIT] =        "tx-checksum-sctp",
+	[NETIF_F_FCOE_MTU_BIT] =         "fcoe-mtu",
+	[NETIF_F_NTUPLE_BIT] =           "rx-ntuple-filter",
+	[NETIF_F_RXHASH_BIT] =           "rx-hashing",
+	[NETIF_F_RXCSUM_BIT] =           "rx-checksum",
+	[NETIF_F_NOCACHE_COPY_BIT] =     "tx-nocache-copy",
+	[NETIF_F_LOOPBACK_BIT] =         "loopback",
+	[NETIF_F_RXFCS_BIT] =            "rx-fcs",
+	[NETIF_F_RXALL_BIT] =            "rx-all",
+	[NETIF_F_HW_L2FW_DOFFLOAD_BIT] = "l2-fwd-offload",
+	[NETIF_F_HW_TC_BIT] =		 "hw-tc-offload",
+	[NETIF_F_HW_ESP_BIT] =		 "esp-hw-offload",
+	[NETIF_F_HW_ESP_TX_CSUM_BIT] =	 "esp-tx-csum-hw-offload",
+	[NETIF_F_RX_UDP_TUNNEL_PORT_BIT] =	 "rx-udp_tunnel-port-offload",
+	[NETIF_F_HW_TLS_RECORD_BIT] =	"tls-hw-record",
+	[NETIF_F_HW_TLS_TX_BIT] =	 "tls-hw-tx-offload",
+	[NETIF_F_HW_TLS_RX_BIT] =	 "tls-hw-rx-offload",
+};
+
+static const char
+rss_hash_func_strings[ETH_RSS_HASH_FUNCS_COUNT][ETH_GSTRING_LEN] = {
+	[ETH_RSS_HASH_TOP_BIT] =	"toeplitz",
+	[ETH_RSS_HASH_XOR_BIT] =	"xor",
+	[ETH_RSS_HASH_CRC32_BIT] =	"crc32",
+};
+
+static const char
+tunable_strings[__ETHTOOL_TUNABLE_COUNT][ETH_GSTRING_LEN] = {
+	[ETHTOOL_ID_UNSPEC]     = "Unspec",
+	[ETHTOOL_RX_COPYBREAK]	= "rx-copybreak",
+	[ETHTOOL_TX_COPYBREAK]	= "tx-copybreak",
+	[ETHTOOL_PFC_PREVENTION_TOUT] = "pfc-prevention-tout",
+};
+
+static const char
+phy_tunable_strings[__ETHTOOL_PHY_TUNABLE_COUNT][ETH_GSTRING_LEN] = {
+	[ETHTOOL_ID_UNSPEC]     = "Unspec",
+	[ETHTOOL_PHY_DOWNSHIFT]	= "phy-downshift",
+	[ETHTOOL_PHY_FAST_LINK_DOWN] = "phy-fast-link-down",
+	[ETHTOOL_PHY_EDPD]	= "phy-energy-detect-power-down",
+};
+
+static int ethtool_get_features(struct net_device *dev, void __user *useraddr)
+{
+	struct ethtool_gfeatures cmd = {
+		.cmd = ETHTOOL_GFEATURES,
+		.size = ETHTOOL_DEV_FEATURE_WORDS,
+	};
+	struct ethtool_get_features_block features[ETHTOOL_DEV_FEATURE_WORDS];
+	u32 __user *sizeaddr;
+	u32 copy_size;
+	int i;
+
+	/* in case feature bits run out again */
+	BUILD_BUG_ON(ETHTOOL_DEV_FEATURE_WORDS * sizeof(u32) > sizeof(netdev_features_t));
+
+	for (i = 0; i < ETHTOOL_DEV_FEATURE_WORDS; ++i) {
+		features[i].available = (u32)(dev->hw_features >> (32 * i));
+		features[i].requested = (u32)(dev->wanted_features >> (32 * i));
+		features[i].active = (u32)(dev->features >> (32 * i));
+		features[i].never_changed =
+			(u32)(NETIF_F_NEVER_CHANGE >> (32 * i));
+	}
+
+	sizeaddr = useraddr + offsetof(struct ethtool_gfeatures, size);
+	if (get_user(copy_size, sizeaddr))
+		return -EFAULT;
+
+	if (copy_size > ETHTOOL_DEV_FEATURE_WORDS)
+		copy_size = ETHTOOL_DEV_FEATURE_WORDS;
+
+	if (copy_to_user(useraddr, &cmd, sizeof(cmd)))
+		return -EFAULT;
+	useraddr += sizeof(cmd);
+	if (copy_to_user(useraddr, features, copy_size * sizeof(*features)))
+		return -EFAULT;
+
+	return 0;
+}
+
+static int ethtool_set_features(struct net_device *dev, void __user *useraddr)
+{
+	struct ethtool_sfeatures cmd;
+	struct ethtool_set_features_block features[ETHTOOL_DEV_FEATURE_WORDS];
+	netdev_features_t wanted = 0, valid = 0;
+	int i, ret = 0;
+
+	if (copy_from_user(&cmd, useraddr, sizeof(cmd)))
+		return -EFAULT;
+	useraddr += sizeof(cmd);
+
+	if (cmd.size != ETHTOOL_DEV_FEATURE_WORDS)
+		return -EINVAL;
+
+	if (copy_from_user(features, useraddr, sizeof(features)))
+		return -EFAULT;
+
+	for (i = 0; i < ETHTOOL_DEV_FEATURE_WORDS; ++i) {
+		valid |= (netdev_features_t)features[i].valid << (32 * i);
+		wanted |= (netdev_features_t)features[i].requested << (32 * i);
+	}
+
+	if (valid & ~NETIF_F_ETHTOOL_BITS)
+		return -EINVAL;
+
+	if (valid & ~dev->hw_features) {
+		valid &= dev->hw_features;
+		ret |= ETHTOOL_F_UNSUPPORTED;
+	}
+
+	dev->wanted_features &= ~valid;
+	dev->wanted_features |= wanted & valid;
+	__netdev_update_features(dev);
+
+	if ((dev->wanted_features ^ dev->features) & valid)
+		ret |= ETHTOOL_F_WISH;
+
+	return ret;
+}
+
+static int __ethtool_get_sset_count(struct net_device *dev, int sset)
+{
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+
+	if (sset == ETH_SS_FEATURES)
+		return ARRAY_SIZE(netdev_features_strings);
+
+	if (sset == ETH_SS_RSS_HASH_FUNCS)
+		return ARRAY_SIZE(rss_hash_func_strings);
+
+	if (sset == ETH_SS_TUNABLES)
+		return ARRAY_SIZE(tunable_strings);
+
+	if (sset == ETH_SS_PHY_TUNABLES)
+		return ARRAY_SIZE(phy_tunable_strings);
+
+	if (sset == ETH_SS_PHY_STATS && dev->phydev &&
+	    !ops->get_ethtool_phy_stats)
+		return phy_ethtool_get_sset_count(dev->phydev);
+
+	if (ops->get_sset_count && ops->get_strings)
+		return ops->get_sset_count(dev, sset);
+	else
+		return -EOPNOTSUPP;
+}
+
+static void __ethtool_get_strings(struct net_device *dev,
+	u32 stringset, u8 *data)
+{
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+
+	if (stringset == ETH_SS_FEATURES)
+		memcpy(data, netdev_features_strings,
+			sizeof(netdev_features_strings));
+	else if (stringset == ETH_SS_RSS_HASH_FUNCS)
+		memcpy(data, rss_hash_func_strings,
+		       sizeof(rss_hash_func_strings));
+	else if (stringset == ETH_SS_TUNABLES)
+		memcpy(data, tunable_strings, sizeof(tunable_strings));
+	else if (stringset == ETH_SS_PHY_TUNABLES)
+		memcpy(data, phy_tunable_strings, sizeof(phy_tunable_strings));
+	else if (stringset == ETH_SS_PHY_STATS && dev->phydev &&
+		 !ops->get_ethtool_phy_stats)
+		phy_ethtool_get_strings(dev->phydev, data);
+	else
+		/* ops->get_strings is valid because checked earlier */
+		ops->get_strings(dev, stringset, data);
+}
+
+static netdev_features_t ethtool_get_feature_mask(u32 eth_cmd)
+{
+	/* feature masks of legacy discrete ethtool ops */
+
+	switch (eth_cmd) {
+	case ETHTOOL_GTXCSUM:
+	case ETHTOOL_STXCSUM:
+		return NETIF_F_CSUM_MASK | NETIF_F_SCTP_CRC;
+	case ETHTOOL_GRXCSUM:
+	case ETHTOOL_SRXCSUM:
+		return NETIF_F_RXCSUM;
+	case ETHTOOL_GSG:
+	case ETHTOOL_SSG:
+		return NETIF_F_SG;
+	case ETHTOOL_GTSO:
+	case ETHTOOL_STSO:
+		return NETIF_F_ALL_TSO;
+	case ETHTOOL_GGSO:
+	case ETHTOOL_SGSO:
+		return NETIF_F_GSO;
+	case ETHTOOL_GGRO:
+	case ETHTOOL_SGRO:
+		return NETIF_F_GRO;
+	default:
+		BUG();
+	}
+}
+
+static int ethtool_get_one_feature(struct net_device *dev,
+	char __user *useraddr, u32 ethcmd)
+{
+	netdev_features_t mask = ethtool_get_feature_mask(ethcmd);
+	struct ethtool_value edata = {
+		.cmd = ethcmd,
+		.data = !!(dev->features & mask),
+	};
+
+	if (copy_to_user(useraddr, &edata, sizeof(edata)))
+		return -EFAULT;
+	return 0;
+}
+
+static int ethtool_set_one_feature(struct net_device *dev,
+	void __user *useraddr, u32 ethcmd)
+{
+	struct ethtool_value edata;
+	netdev_features_t mask;
+
+	if (copy_from_user(&edata, useraddr, sizeof(edata)))
+		return -EFAULT;
+
+	mask = ethtool_get_feature_mask(ethcmd);
+	mask &= dev->hw_features;
+	if (!mask)
+		return -EOPNOTSUPP;
+
+	if (edata.data)
+		dev->wanted_features |= mask;
+	else
+		dev->wanted_features &= ~mask;
+
+	__netdev_update_features(dev);
+
+	return 0;
+}
+
+#define ETH_ALL_FLAGS    (ETH_FLAG_LRO | ETH_FLAG_RXVLAN | ETH_FLAG_TXVLAN | \
+			  ETH_FLAG_NTUPLE | ETH_FLAG_RXHASH)
+#define ETH_ALL_FEATURES (NETIF_F_LRO | NETIF_F_HW_VLAN_CTAG_RX | \
+			  NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_NTUPLE | \
+			  NETIF_F_RXHASH)
+
+static u32 __ethtool_get_flags(struct net_device *dev)
+{
+	u32 flags = 0;
+
+	if (dev->features & NETIF_F_LRO)
+		flags |= ETH_FLAG_LRO;
+	if (dev->features & NETIF_F_HW_VLAN_CTAG_RX)
+		flags |= ETH_FLAG_RXVLAN;
+	if (dev->features & NETIF_F_HW_VLAN_CTAG_TX)
+		flags |= ETH_FLAG_TXVLAN;
+	if (dev->features & NETIF_F_NTUPLE)
+		flags |= ETH_FLAG_NTUPLE;
+	if (dev->features & NETIF_F_RXHASH)
+		flags |= ETH_FLAG_RXHASH;
+
+	return flags;
+}
+
+static int __ethtool_set_flags(struct net_device *dev, u32 data)
+{
+	netdev_features_t features = 0, changed;
+
+	if (data & ~ETH_ALL_FLAGS)
+		return -EINVAL;
+
+	if (data & ETH_FLAG_LRO)
+		features |= NETIF_F_LRO;
+	if (data & ETH_FLAG_RXVLAN)
+		features |= NETIF_F_HW_VLAN_CTAG_RX;
+	if (data & ETH_FLAG_TXVLAN)
+		features |= NETIF_F_HW_VLAN_CTAG_TX;
+	if (data & ETH_FLAG_NTUPLE)
+		features |= NETIF_F_NTUPLE;
+	if (data & ETH_FLAG_RXHASH)
+		features |= NETIF_F_RXHASH;
+
+	/* allow changing only bits set in hw_features */
+	changed = (features ^ dev->features) & ETH_ALL_FEATURES;
+	if (changed & ~dev->hw_features)
+		return (changed & dev->hw_features) ? -EINVAL : -EOPNOTSUPP;
+
+	dev->wanted_features =
+		(dev->wanted_features & ~changed) | (features & changed);
+
+	__netdev_update_features(dev);
+
+	return 0;
+}
+
+/* Given two link masks, AND them together and save the result in dst. */
+void ethtool_intersect_link_masks(struct ethtool_link_ksettings *dst,
+				  struct ethtool_link_ksettings *src)
+{
+	unsigned int size = BITS_TO_LONGS(__ETHTOOL_LINK_MODE_MASK_NBITS);
+	unsigned int idx = 0;
+
+	for (; idx < size; idx++) {
+		dst->link_modes.supported[idx] &=
+			src->link_modes.supported[idx];
+		dst->link_modes.advertising[idx] &=
+			src->link_modes.advertising[idx];
+	}
+}
+EXPORT_SYMBOL(ethtool_intersect_link_masks);
+
+void ethtool_convert_legacy_u32_to_link_mode(unsigned long *dst,
+					     u32 legacy_u32)
+{
+	bitmap_zero(dst, __ETHTOOL_LINK_MODE_MASK_NBITS);
+	dst[0] = legacy_u32;
+}
+EXPORT_SYMBOL(ethtool_convert_legacy_u32_to_link_mode);
+
+/* return false if src had higher bits set. lower bits always updated. */
+bool ethtool_convert_link_mode_to_legacy_u32(u32 *legacy_u32,
+					     const unsigned long *src)
+{
+	bool retval = true;
+
+	/* TODO: following test will soon always be true */
+	if (__ETHTOOL_LINK_MODE_MASK_NBITS > 32) {
+		__ETHTOOL_DECLARE_LINK_MODE_MASK(ext);
+
+		bitmap_zero(ext, __ETHTOOL_LINK_MODE_MASK_NBITS);
+		bitmap_fill(ext, 32);
+		bitmap_complement(ext, ext, __ETHTOOL_LINK_MODE_MASK_NBITS);
+		if (bitmap_intersects(ext, src,
+				      __ETHTOOL_LINK_MODE_MASK_NBITS)) {
+			/* src mask goes beyond bit 31 */
+			retval = false;
+		}
+	}
+	*legacy_u32 = src[0];
+	return retval;
+}
+EXPORT_SYMBOL(ethtool_convert_link_mode_to_legacy_u32);
+
+/* return false if legacy contained non-0 deprecated fields
+ * maxtxpkt/maxrxpkt. rest of ksettings always updated
+ */
+static bool
+convert_legacy_settings_to_link_ksettings(
+	struct ethtool_link_ksettings *link_ksettings,
+	const struct ethtool_cmd *legacy_settings)
+{
+	bool retval = true;
+
+	memset(link_ksettings, 0, sizeof(*link_ksettings));
+
+	/* This is used to tell users that driver is still using these
+	 * deprecated legacy fields, and they should not use
+	 * %ETHTOOL_GLINKSETTINGS/%ETHTOOL_SLINKSETTINGS
+	 */
+	if (legacy_settings->maxtxpkt ||
+	    legacy_settings->maxrxpkt)
+		retval = false;
+
+	ethtool_convert_legacy_u32_to_link_mode(
+		link_ksettings->link_modes.supported,
+		legacy_settings->supported);
+	ethtool_convert_legacy_u32_to_link_mode(
+		link_ksettings->link_modes.advertising,
+		legacy_settings->advertising);
+	ethtool_convert_legacy_u32_to_link_mode(
+		link_ksettings->link_modes.lp_advertising,
+		legacy_settings->lp_advertising);
+	link_ksettings->base.speed
+		= ethtool_cmd_speed(legacy_settings);
+	link_ksettings->base.duplex
+		= legacy_settings->duplex;
+	link_ksettings->base.port
+		= legacy_settings->port;
+	link_ksettings->base.phy_address
+		= legacy_settings->phy_address;
+	link_ksettings->base.autoneg
+		= legacy_settings->autoneg;
+	link_ksettings->base.mdio_support
+		= legacy_settings->mdio_support;
+	link_ksettings->base.eth_tp_mdix
+		= legacy_settings->eth_tp_mdix;
+	link_ksettings->base.eth_tp_mdix_ctrl
+		= legacy_settings->eth_tp_mdix_ctrl;
+	return retval;
+}
+
+/* return false if ksettings link modes had higher bits
+ * set. legacy_settings always updated (best effort)
+ */
+static bool
+convert_link_ksettings_to_legacy_settings(
+	struct ethtool_cmd *legacy_settings,
+	const struct ethtool_link_ksettings *link_ksettings)
+{
+	bool retval = true;
+
+	memset(legacy_settings, 0, sizeof(*legacy_settings));
+	/* this also clears the deprecated fields in legacy structure:
+	 * __u8		transceiver;
+	 * __u32	maxtxpkt;
+	 * __u32	maxrxpkt;
+	 */
+
+	retval &= ethtool_convert_link_mode_to_legacy_u32(
+		&legacy_settings->supported,
+		link_ksettings->link_modes.supported);
+	retval &= ethtool_convert_link_mode_to_legacy_u32(
+		&legacy_settings->advertising,
+		link_ksettings->link_modes.advertising);
+	retval &= ethtool_convert_link_mode_to_legacy_u32(
+		&legacy_settings->lp_advertising,
+		link_ksettings->link_modes.lp_advertising);
+	ethtool_cmd_speed_set(legacy_settings, link_ksettings->base.speed);
+	legacy_settings->duplex
+		= link_ksettings->base.duplex;
+	legacy_settings->port
+		= link_ksettings->base.port;
+	legacy_settings->phy_address
+		= link_ksettings->base.phy_address;
+	legacy_settings->autoneg
+		= link_ksettings->base.autoneg;
+	legacy_settings->mdio_support
+		= link_ksettings->base.mdio_support;
+	legacy_settings->eth_tp_mdix
+		= link_ksettings->base.eth_tp_mdix;
+	legacy_settings->eth_tp_mdix_ctrl
+		= link_ksettings->base.eth_tp_mdix_ctrl;
+	legacy_settings->transceiver
+		= link_ksettings->base.transceiver;
+	return retval;
+}
+
+/* number of 32-bit words to store the user's link mode bitmaps */
+#define __ETHTOOL_LINK_MODE_MASK_NU32			\
+	DIV_ROUND_UP(__ETHTOOL_LINK_MODE_MASK_NBITS, 32)
+
+/* layout of the struct passed from/to userland */
+struct ethtool_link_usettings {
+	struct ethtool_link_settings base;
+	struct {
+		__u32 supported[__ETHTOOL_LINK_MODE_MASK_NU32];
+		__u32 advertising[__ETHTOOL_LINK_MODE_MASK_NU32];
+		__u32 lp_advertising[__ETHTOOL_LINK_MODE_MASK_NU32];
+	} link_modes;
+};
+
+/* Internal kernel helper to query a device ethtool_link_settings. */
+int __ethtool_get_link_ksettings(struct net_device *dev,
+				 struct ethtool_link_ksettings *link_ksettings)
+{
+	ASSERT_RTNL();
+
+	if (!dev->ethtool_ops->get_link_ksettings)
+		return -EOPNOTSUPP;
+
+	memset(link_ksettings, 0, sizeof(*link_ksettings));
+	return dev->ethtool_ops->get_link_ksettings(dev, link_ksettings);
+}
+EXPORT_SYMBOL(__ethtool_get_link_ksettings);
+
+/* convert ethtool_link_usettings in user space to a kernel internal
+ * ethtool_link_ksettings. return 0 on success, errno on error.
+ */
+static int load_link_ksettings_from_user(struct ethtool_link_ksettings *to,
+					 const void __user *from)
+{
+	struct ethtool_link_usettings link_usettings;
+
+	if (copy_from_user(&link_usettings, from, sizeof(link_usettings)))
+		return -EFAULT;
+
+	memcpy(&to->base, &link_usettings.base, sizeof(to->base));
+	bitmap_from_arr32(to->link_modes.supported,
+			  link_usettings.link_modes.supported,
+			  __ETHTOOL_LINK_MODE_MASK_NBITS);
+	bitmap_from_arr32(to->link_modes.advertising,
+			  link_usettings.link_modes.advertising,
+			  __ETHTOOL_LINK_MODE_MASK_NBITS);
+	bitmap_from_arr32(to->link_modes.lp_advertising,
+			  link_usettings.link_modes.lp_advertising,
+			  __ETHTOOL_LINK_MODE_MASK_NBITS);
+
+	return 0;
+}
+
+/* convert a kernel internal ethtool_link_ksettings to
+ * ethtool_link_usettings in user space. return 0 on success, errno on
+ * error.
+ */
+static int
+store_link_ksettings_for_user(void __user *to,
+			      const struct ethtool_link_ksettings *from)
+{
+	struct ethtool_link_usettings link_usettings;
+
+	memcpy(&link_usettings.base, &from->base, sizeof(link_usettings));
+	bitmap_to_arr32(link_usettings.link_modes.supported,
+			from->link_modes.supported,
+			__ETHTOOL_LINK_MODE_MASK_NBITS);
+	bitmap_to_arr32(link_usettings.link_modes.advertising,
+			from->link_modes.advertising,
+			__ETHTOOL_LINK_MODE_MASK_NBITS);
+	bitmap_to_arr32(link_usettings.link_modes.lp_advertising,
+			from->link_modes.lp_advertising,
+			__ETHTOOL_LINK_MODE_MASK_NBITS);
+
+	if (copy_to_user(to, &link_usettings, sizeof(link_usettings)))
+		return -EFAULT;
+
+	return 0;
+}
+
+/* Query device for its ethtool_link_settings. */
+static int ethtool_get_link_ksettings(struct net_device *dev,
+				      void __user *useraddr)
+{
+	int err = 0;
+	struct ethtool_link_ksettings link_ksettings;
+
+	ASSERT_RTNL();
+	if (!dev->ethtool_ops->get_link_ksettings)
+		return -EOPNOTSUPP;
+
+	/* handle bitmap nbits handshake */
+	if (copy_from_user(&link_ksettings.base, useraddr,
+			   sizeof(link_ksettings.base)))
+		return -EFAULT;
+
+	if (__ETHTOOL_LINK_MODE_MASK_NU32
+	    != link_ksettings.base.link_mode_masks_nwords) {
+		/* wrong link mode nbits requested */
+		memset(&link_ksettings, 0, sizeof(link_ksettings));
+		link_ksettings.base.cmd = ETHTOOL_GLINKSETTINGS;
+		/* send back number of words required as negative val */
+		compiletime_assert(__ETHTOOL_LINK_MODE_MASK_NU32 <= S8_MAX,
+				   "need too many bits for link modes!");
+		link_ksettings.base.link_mode_masks_nwords
+			= -((s8)__ETHTOOL_LINK_MODE_MASK_NU32);
+
+		/* copy the base fields back to user, not the link
+		 * mode bitmaps
+		 */
+		if (copy_to_user(useraddr, &link_ksettings.base,
+				 sizeof(link_ksettings.base)))
+			return -EFAULT;
+
+		return 0;
+	}
+
+	/* handshake successful: user/kernel agree on
+	 * link_mode_masks_nwords
+	 */
+
+	memset(&link_ksettings, 0, sizeof(link_ksettings));
+	err = dev->ethtool_ops->get_link_ksettings(dev, &link_ksettings);
+	if (err < 0)
+		return err;
+
+	/* make sure we tell the right values to user */
+	link_ksettings.base.cmd = ETHTOOL_GLINKSETTINGS;
+	link_ksettings.base.link_mode_masks_nwords
+		= __ETHTOOL_LINK_MODE_MASK_NU32;
+
+	return store_link_ksettings_for_user(useraddr, &link_ksettings);
+}
+
+/* Update device ethtool_link_settings. */
+static int ethtool_set_link_ksettings(struct net_device *dev,
+				      void __user *useraddr)
+{
+	int err;
+	struct ethtool_link_ksettings link_ksettings;
+
+	ASSERT_RTNL();
+
+	if (!dev->ethtool_ops->set_link_ksettings)
+		return -EOPNOTSUPP;
+
+	/* make sure nbits field has expected value */
+	if (copy_from_user(&link_ksettings.base, useraddr,
+			   sizeof(link_ksettings.base)))
+		return -EFAULT;
+
+	if (__ETHTOOL_LINK_MODE_MASK_NU32
+	    != link_ksettings.base.link_mode_masks_nwords)
+		return -EINVAL;
+
+	/* copy the whole structure, now that we know it has expected
+	 * format
+	 */
+	err = load_link_ksettings_from_user(&link_ksettings, useraddr);
+	if (err)
+		return err;
+
+	/* re-check nwords field, just in case */
+	if (__ETHTOOL_LINK_MODE_MASK_NU32
+	    != link_ksettings.base.link_mode_masks_nwords)
+		return -EINVAL;
+
+	return dev->ethtool_ops->set_link_ksettings(dev, &link_ksettings);
+}
+
+/* Query device for its ethtool_cmd settings.
+ *
+ * Backward compatibility note: for compatibility with legacy ethtool, this is
+ * now implemented via get_link_ksettings. When driver reports higher link mode
+ * bits, a kernel warning is logged once (with name of 1st driver/device) to
+ * recommend user to upgrade ethtool, but the command is successful (only the
+ * lower link mode bits reported back to user). Deprecated fields from
+ * ethtool_cmd (transceiver/maxrxpkt/maxtxpkt) are always set to zero.
+ */
+static int ethtool_get_settings(struct net_device *dev, void __user *useraddr)
+{
+	struct ethtool_link_ksettings link_ksettings;
+	struct ethtool_cmd cmd;
+	int err;
+
+	ASSERT_RTNL();
+	if (!dev->ethtool_ops->get_link_ksettings)
+		return -EOPNOTSUPP;
+
+	memset(&link_ksettings, 0, sizeof(link_ksettings));
+	err = dev->ethtool_ops->get_link_ksettings(dev, &link_ksettings);
+	if (err < 0)
+		return err;
+	convert_link_ksettings_to_legacy_settings(&cmd, &link_ksettings);
+
+	/* send a sensible cmd tag back to user */
+	cmd.cmd = ETHTOOL_GSET;
+
+	if (copy_to_user(useraddr, &cmd, sizeof(cmd)))
+		return -EFAULT;
+
+	return 0;
+}
+
+/* Update device link settings with given ethtool_cmd.
+ *
+ * Backward compatibility note: for compatibility with legacy ethtool, this is
+ * now always implemented via set_link_settings. When user's request updates
+ * deprecated ethtool_cmd fields (transceiver/maxrxpkt/maxtxpkt), a kernel
+ * warning is logged once (with name of 1st driver/device) to recommend user to
+ * upgrade ethtool, and the request is rejected.
+ */
+static int ethtool_set_settings(struct net_device *dev, void __user *useraddr)
+{
+	struct ethtool_link_ksettings link_ksettings;
+	struct ethtool_cmd cmd;
+
+	ASSERT_RTNL();
+
+	if (copy_from_user(&cmd, useraddr, sizeof(cmd)))
+		return -EFAULT;
+	if (!dev->ethtool_ops->set_link_ksettings)
+		return -EOPNOTSUPP;
+
+	if (!convert_legacy_settings_to_link_ksettings(&link_ksettings, &cmd))
+		return -EINVAL;
+	link_ksettings.base.link_mode_masks_nwords =
+		__ETHTOOL_LINK_MODE_MASK_NU32;
+	return dev->ethtool_ops->set_link_ksettings(dev, &link_ksettings);
+}
+
+static noinline_for_stack int ethtool_get_drvinfo(struct net_device *dev,
+						  void __user *useraddr)
+{
+	struct ethtool_drvinfo info;
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+
+	memset(&info, 0, sizeof(info));
+	info.cmd = ETHTOOL_GDRVINFO;
+	if (ops->get_drvinfo) {
+		ops->get_drvinfo(dev, &info);
+	} else if (dev->dev.parent && dev->dev.parent->driver) {
+		strlcpy(info.bus_info, dev_name(dev->dev.parent),
+			sizeof(info.bus_info));
+		strlcpy(info.driver, dev->dev.parent->driver->name,
+			sizeof(info.driver));
+	} else {
+		return -EOPNOTSUPP;
+	}
+
+	/*
+	 * this method of obtaining string set info is deprecated;
+	 * Use ETHTOOL_GSSET_INFO instead.
+	 */
+	if (ops->get_sset_count) {
+		int rc;
+
+		rc = ops->get_sset_count(dev, ETH_SS_TEST);
+		if (rc >= 0)
+			info.testinfo_len = rc;
+		rc = ops->get_sset_count(dev, ETH_SS_STATS);
+		if (rc >= 0)
+			info.n_stats = rc;
+		rc = ops->get_sset_count(dev, ETH_SS_PRIV_FLAGS);
+		if (rc >= 0)
+			info.n_priv_flags = rc;
+	}
+	if (ops->get_regs_len) {
+		int ret = ops->get_regs_len(dev);
+
+		if (ret > 0)
+			info.regdump_len = ret;
+	}
+
+	if (ops->get_eeprom_len)
+		info.eedump_len = ops->get_eeprom_len(dev);
+
+	if (!info.fw_version[0])
+		devlink_compat_running_version(dev, info.fw_version,
+					       sizeof(info.fw_version));
+
+	if (copy_to_user(useraddr, &info, sizeof(info)))
+		return -EFAULT;
+	return 0;
+}
+
+static noinline_for_stack int ethtool_get_sset_info(struct net_device *dev,
+						    void __user *useraddr)
+{
+	struct ethtool_sset_info info;
+	u64 sset_mask;
+	int i, idx = 0, n_bits = 0, ret, rc;
+	u32 *info_buf = NULL;
+
+	if (copy_from_user(&info, useraddr, sizeof(info)))
+		return -EFAULT;
+
+	/* store copy of mask, because we zero struct later on */
+	sset_mask = info.sset_mask;
+	if (!sset_mask)
+		return 0;
+
+	/* calculate size of return buffer */
+	n_bits = hweight64(sset_mask);
+
+	memset(&info, 0, sizeof(info));
+	info.cmd = ETHTOOL_GSSET_INFO;
+
+	info_buf = kcalloc(n_bits, sizeof(u32), GFP_USER);
+	if (!info_buf)
+		return -ENOMEM;
+
+	/*
+	 * fill return buffer based on input bitmask and successful
+	 * get_sset_count return
+	 */
+	for (i = 0; i < 64; i++) {
+		if (!(sset_mask & (1ULL << i)))
+			continue;
+
+		rc = __ethtool_get_sset_count(dev, i);
+		if (rc >= 0) {
+			info.sset_mask |= (1ULL << i);
+			info_buf[idx++] = rc;
+		}
+	}
+
+	ret = -EFAULT;
+	if (copy_to_user(useraddr, &info, sizeof(info)))
+		goto out;
+
+	useraddr += offsetof(struct ethtool_sset_info, data);
+	if (copy_to_user(useraddr, info_buf, idx * sizeof(u32)))
+		goto out;
+
+	ret = 0;
+
+out:
+	kfree(info_buf);
+	return ret;
+}
+
+static noinline_for_stack int ethtool_set_rxnfc(struct net_device *dev,
+						u32 cmd, void __user *useraddr)
+{
+	struct ethtool_rxnfc info;
+	size_t info_size = sizeof(info);
+	int rc;
+
+	if (!dev->ethtool_ops->set_rxnfc)
+		return -EOPNOTSUPP;
+
+	/* struct ethtool_rxnfc was originally defined for
+	 * ETHTOOL_{G,S}RXFH with only the cmd, flow_type and data
+	 * members.  User-space might still be using that
+	 * definition. */
+	if (cmd == ETHTOOL_SRXFH)
+		info_size = (offsetof(struct ethtool_rxnfc, data) +
+			     sizeof(info.data));
+
+	if (copy_from_user(&info, useraddr, info_size))
+		return -EFAULT;
+
+	rc = dev->ethtool_ops->set_rxnfc(dev, &info);
+	if (rc)
+		return rc;
+
+	if (cmd == ETHTOOL_SRXCLSRLINS &&
+	    copy_to_user(useraddr, &info, info_size))
+		return -EFAULT;
+
+	return 0;
+}
+
+static noinline_for_stack int ethtool_get_rxnfc(struct net_device *dev,
+						u32 cmd, void __user *useraddr)
+{
+	struct ethtool_rxnfc info;
+	size_t info_size = sizeof(info);
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+	int ret;
+	void *rule_buf = NULL;
+
+	if (!ops->get_rxnfc)
+		return -EOPNOTSUPP;
+
+	/* struct ethtool_rxnfc was originally defined for
+	 * ETHTOOL_{G,S}RXFH with only the cmd, flow_type and data
+	 * members.  User-space might still be using that
+	 * definition. */
+	if (cmd == ETHTOOL_GRXFH)
+		info_size = (offsetof(struct ethtool_rxnfc, data) +
+			     sizeof(info.data));
+
+	if (copy_from_user(&info, useraddr, info_size))
+		return -EFAULT;
+
+	/* If FLOW_RSS was requested then user-space must be using the
+	 * new definition, as FLOW_RSS is newer.
+	 */
+	if (cmd == ETHTOOL_GRXFH && info.flow_type & FLOW_RSS) {
+		info_size = sizeof(info);
+		if (copy_from_user(&info, useraddr, info_size))
+			return -EFAULT;
+		/* Since malicious users may modify the original data,
+		 * we need to check whether FLOW_RSS is still requested.
+		 */
+		if (!(info.flow_type & FLOW_RSS))
+			return -EINVAL;
+	}
+
+	if (info.cmd != cmd)
+		return -EINVAL;
+
+	if (info.cmd == ETHTOOL_GRXCLSRLALL) {
+		if (info.rule_cnt > 0) {
+			if (info.rule_cnt <= KMALLOC_MAX_SIZE / sizeof(u32))
+				rule_buf = kcalloc(info.rule_cnt, sizeof(u32),
+						   GFP_USER);
+			if (!rule_buf)
+				return -ENOMEM;
+		}
+	}
+
+	ret = ops->get_rxnfc(dev, &info, rule_buf);
+	if (ret < 0)
+		goto err_out;
+
+	ret = -EFAULT;
+	if (copy_to_user(useraddr, &info, info_size))
+		goto err_out;
+
+	if (rule_buf) {
+		useraddr += offsetof(struct ethtool_rxnfc, rule_locs);
+		if (copy_to_user(useraddr, rule_buf,
+				 info.rule_cnt * sizeof(u32)))
+			goto err_out;
+	}
+	ret = 0;
+
+err_out:
+	kfree(rule_buf);
+
+	return ret;
+}
+
+static int ethtool_copy_validate_indir(u32 *indir, void __user *useraddr,
+					struct ethtool_rxnfc *rx_rings,
+					u32 size)
+{
+	int i;
+
+	if (copy_from_user(indir, useraddr, size * sizeof(indir[0])))
+		return -EFAULT;
+
+	/* Validate ring indices */
+	for (i = 0; i < size; i++)
+		if (indir[i] >= rx_rings->data)
+			return -EINVAL;
+
+	return 0;
+}
+
+u8 netdev_rss_key[NETDEV_RSS_KEY_LEN] __read_mostly;
+
+void netdev_rss_key_fill(void *buffer, size_t len)
+{
+	BUG_ON(len > sizeof(netdev_rss_key));
+	net_get_random_once(netdev_rss_key, sizeof(netdev_rss_key));
+	memcpy(buffer, netdev_rss_key, len);
+}
+EXPORT_SYMBOL(netdev_rss_key_fill);
+
+static int ethtool_get_max_rxfh_channel(struct net_device *dev, u32 *max)
+{
+	u32 dev_size, current_max = 0;
+	u32 *indir;
+	int ret;
+
+	if (!dev->ethtool_ops->get_rxfh_indir_size ||
+	    !dev->ethtool_ops->get_rxfh)
+		return -EOPNOTSUPP;
+	dev_size = dev->ethtool_ops->get_rxfh_indir_size(dev);
+	if (dev_size == 0)
+		return -EOPNOTSUPP;
+
+	indir = kcalloc(dev_size, sizeof(indir[0]), GFP_USER);
+	if (!indir)
+		return -ENOMEM;
+
+	ret = dev->ethtool_ops->get_rxfh(dev, indir, NULL, NULL);
+	if (ret)
+		goto out;
+
+	while (dev_size--)
+		current_max = max(current_max, indir[dev_size]);
+
+	*max = current_max;
+
+out:
+	kfree(indir);
+	return ret;
+}
+
+static noinline_for_stack int ethtool_get_rxfh_indir(struct net_device *dev,
+						     void __user *useraddr)
+{
+	u32 user_size, dev_size;
+	u32 *indir;
+	int ret;
+
+	if (!dev->ethtool_ops->get_rxfh_indir_size ||
+	    !dev->ethtool_ops->get_rxfh)
+		return -EOPNOTSUPP;
+	dev_size = dev->ethtool_ops->get_rxfh_indir_size(dev);
+	if (dev_size == 0)
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&user_size,
+			   useraddr + offsetof(struct ethtool_rxfh_indir, size),
+			   sizeof(user_size)))
+		return -EFAULT;
+
+	if (copy_to_user(useraddr + offsetof(struct ethtool_rxfh_indir, size),
+			 &dev_size, sizeof(dev_size)))
+		return -EFAULT;
+
+	/* If the user buffer size is 0, this is just a query for the
+	 * device table size.  Otherwise, if it's smaller than the
+	 * device table size it's an error.
+	 */
+	if (user_size < dev_size)
+		return user_size == 0 ? 0 : -EINVAL;
+
+	indir = kcalloc(dev_size, sizeof(indir[0]), GFP_USER);
+	if (!indir)
+		return -ENOMEM;
+
+	ret = dev->ethtool_ops->get_rxfh(dev, indir, NULL, NULL);
+	if (ret)
+		goto out;
+
+	if (copy_to_user(useraddr +
+			 offsetof(struct ethtool_rxfh_indir, ring_index[0]),
+			 indir, dev_size * sizeof(indir[0])))
+		ret = -EFAULT;
+
+out:
+	kfree(indir);
+	return ret;
+}
+
+static noinline_for_stack int ethtool_set_rxfh_indir(struct net_device *dev,
+						     void __user *useraddr)
+{
+	struct ethtool_rxnfc rx_rings;
+	u32 user_size, dev_size, i;
+	u32 *indir;
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+	int ret;
+	u32 ringidx_offset = offsetof(struct ethtool_rxfh_indir, ring_index[0]);
+
+	if (!ops->get_rxfh_indir_size || !ops->set_rxfh ||
+	    !ops->get_rxnfc)
+		return -EOPNOTSUPP;
+
+	dev_size = ops->get_rxfh_indir_size(dev);
+	if (dev_size == 0)
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&user_size,
+			   useraddr + offsetof(struct ethtool_rxfh_indir, size),
+			   sizeof(user_size)))
+		return -EFAULT;
+
+	if (user_size != 0 && user_size != dev_size)
+		return -EINVAL;
+
+	indir = kcalloc(dev_size, sizeof(indir[0]), GFP_USER);
+	if (!indir)
+		return -ENOMEM;
+
+	rx_rings.cmd = ETHTOOL_GRXRINGS;
+	ret = ops->get_rxnfc(dev, &rx_rings, NULL);
+	if (ret)
+		goto out;
+
+	if (user_size == 0) {
+		for (i = 0; i < dev_size; i++)
+			indir[i] = ethtool_rxfh_indir_default(i, rx_rings.data);
+	} else {
+		ret = ethtool_copy_validate_indir(indir,
+						  useraddr + ringidx_offset,
+						  &rx_rings,
+						  dev_size);
+		if (ret)
+			goto out;
+	}
+
+	ret = ops->set_rxfh(dev, indir, NULL, ETH_RSS_HASH_NO_CHANGE);
+	if (ret)
+		goto out;
+
+	/* indicate whether rxfh was set to default */
+	if (user_size == 0)
+		dev->priv_flags &= ~IFF_RXFH_CONFIGURED;
+	else
+		dev->priv_flags |= IFF_RXFH_CONFIGURED;
+
+out:
+	kfree(indir);
+	return ret;
+}
+
+static noinline_for_stack int ethtool_get_rxfh(struct net_device *dev,
+					       void __user *useraddr)
+{
+	int ret;
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+	u32 user_indir_size, user_key_size;
+	u32 dev_indir_size = 0, dev_key_size = 0;
+	struct ethtool_rxfh rxfh;
+	u32 total_size;
+	u32 indir_bytes;
+	u32 *indir = NULL;
+	u8 dev_hfunc = 0;
+	u8 *hkey = NULL;
+	u8 *rss_config;
+
+	if (!ops->get_rxfh)
+		return -EOPNOTSUPP;
+
+	if (ops->get_rxfh_indir_size)
+		dev_indir_size = ops->get_rxfh_indir_size(dev);
+	if (ops->get_rxfh_key_size)
+		dev_key_size = ops->get_rxfh_key_size(dev);
+
+	if (copy_from_user(&rxfh, useraddr, sizeof(rxfh)))
+		return -EFAULT;
+	user_indir_size = rxfh.indir_size;
+	user_key_size = rxfh.key_size;
+
+	/* Check that reserved fields are 0 for now */
+	if (rxfh.rsvd8[0] || rxfh.rsvd8[1] || rxfh.rsvd8[2] || rxfh.rsvd32)
+		return -EINVAL;
+	/* Most drivers don't handle rss_context, check it's 0 as well */
+	if (rxfh.rss_context && !ops->get_rxfh_context)
+		return -EOPNOTSUPP;
+
+	rxfh.indir_size = dev_indir_size;
+	rxfh.key_size = dev_key_size;
+	if (copy_to_user(useraddr, &rxfh, sizeof(rxfh)))
+		return -EFAULT;
+
+	if ((user_indir_size && (user_indir_size != dev_indir_size)) ||
+	    (user_key_size && (user_key_size != dev_key_size)))
+		return -EINVAL;
+
+	indir_bytes = user_indir_size * sizeof(indir[0]);
+	total_size = indir_bytes + user_key_size;
+	rss_config = kzalloc(total_size, GFP_USER);
+	if (!rss_config)
+		return -ENOMEM;
+
+	if (user_indir_size)
+		indir = (u32 *)rss_config;
+
+	if (user_key_size)
+		hkey = rss_config + indir_bytes;
+
+	if (rxfh.rss_context)
+		ret = dev->ethtool_ops->get_rxfh_context(dev, indir, hkey,
+							 &dev_hfunc,
+							 rxfh.rss_context);
+	else
+		ret = dev->ethtool_ops->get_rxfh(dev, indir, hkey, &dev_hfunc);
+	if (ret)
+		goto out;
+
+	if (copy_to_user(useraddr + offsetof(struct ethtool_rxfh, hfunc),
+			 &dev_hfunc, sizeof(rxfh.hfunc))) {
+		ret = -EFAULT;
+	} else if (copy_to_user(useraddr +
+			      offsetof(struct ethtool_rxfh, rss_config[0]),
+			      rss_config, total_size)) {
+		ret = -EFAULT;
+	}
+out:
+	kfree(rss_config);
+
+	return ret;
+}
+
+static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev,
+					       void __user *useraddr)
+{
+	int ret;
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+	struct ethtool_rxnfc rx_rings;
+	struct ethtool_rxfh rxfh;
+	u32 dev_indir_size = 0, dev_key_size = 0, i;
+	u32 *indir = NULL, indir_bytes = 0;
+	u8 *hkey = NULL;
+	u8 *rss_config;
+	u32 rss_cfg_offset = offsetof(struct ethtool_rxfh, rss_config[0]);
+	bool delete = false;
+
+	if (!ops->get_rxnfc || !ops->set_rxfh)
+		return -EOPNOTSUPP;
+
+	if (ops->get_rxfh_indir_size)
+		dev_indir_size = ops->get_rxfh_indir_size(dev);
+	if (ops->get_rxfh_key_size)
+		dev_key_size = ops->get_rxfh_key_size(dev);
+
+	if (copy_from_user(&rxfh, useraddr, sizeof(rxfh)))
+		return -EFAULT;
+
+	/* Check that reserved fields are 0 for now */
+	if (rxfh.rsvd8[0] || rxfh.rsvd8[1] || rxfh.rsvd8[2] || rxfh.rsvd32)
+		return -EINVAL;
+	/* Most drivers don't handle rss_context, check it's 0 as well */
+	if (rxfh.rss_context && !ops->set_rxfh_context)
+		return -EOPNOTSUPP;
+
+	/* If either indir, hash key or function is valid, proceed further.
+	 * Must request at least one change: indir size, hash key or function.
+	 */
+	if ((rxfh.indir_size &&
+	     rxfh.indir_size != ETH_RXFH_INDIR_NO_CHANGE &&
+	     rxfh.indir_size != dev_indir_size) ||
+	    (rxfh.key_size && (rxfh.key_size != dev_key_size)) ||
+	    (rxfh.indir_size == ETH_RXFH_INDIR_NO_CHANGE &&
+	     rxfh.key_size == 0 && rxfh.hfunc == ETH_RSS_HASH_NO_CHANGE))
+		return -EINVAL;
+
+	if (rxfh.indir_size != ETH_RXFH_INDIR_NO_CHANGE)
+		indir_bytes = dev_indir_size * sizeof(indir[0]);
+
+	rss_config = kzalloc(indir_bytes + rxfh.key_size, GFP_USER);
+	if (!rss_config)
+		return -ENOMEM;
+
+	rx_rings.cmd = ETHTOOL_GRXRINGS;
+	ret = ops->get_rxnfc(dev, &rx_rings, NULL);
+	if (ret)
+		goto out;
+
+	/* rxfh.indir_size == 0 means reset the indir table to default (master
+	 * context) or delete the context (other RSS contexts).
+	 * rxfh.indir_size == ETH_RXFH_INDIR_NO_CHANGE means leave it unchanged.
+	 */
+	if (rxfh.indir_size &&
+	    rxfh.indir_size != ETH_RXFH_INDIR_NO_CHANGE) {
+		indir = (u32 *)rss_config;
+		ret = ethtool_copy_validate_indir(indir,
+						  useraddr + rss_cfg_offset,
+						  &rx_rings,
+						  rxfh.indir_size);
+		if (ret)
+			goto out;
+	} else if (rxfh.indir_size == 0) {
+		if (rxfh.rss_context == 0) {
+			indir = (u32 *)rss_config;
+			for (i = 0; i < dev_indir_size; i++)
+				indir[i] = ethtool_rxfh_indir_default(i, rx_rings.data);
+		} else {
+			delete = true;
+		}
+	}
+
+	if (rxfh.key_size) {
+		hkey = rss_config + indir_bytes;
+		if (copy_from_user(hkey,
+				   useraddr + rss_cfg_offset + indir_bytes,
+				   rxfh.key_size)) {
+			ret = -EFAULT;
+			goto out;
+		}
+	}
+
+	if (rxfh.rss_context)
+		ret = ops->set_rxfh_context(dev, indir, hkey, rxfh.hfunc,
+					    &rxfh.rss_context, delete);
+	else
+		ret = ops->set_rxfh(dev, indir, hkey, rxfh.hfunc);
+	if (ret)
+		goto out;
+
+	if (copy_to_user(useraddr + offsetof(struct ethtool_rxfh, rss_context),
+			 &rxfh.rss_context, sizeof(rxfh.rss_context)))
+		ret = -EFAULT;
+
+	if (!rxfh.rss_context) {
+		/* indicate whether rxfh was set to default */
+		if (rxfh.indir_size == 0)
+			dev->priv_flags &= ~IFF_RXFH_CONFIGURED;
+		else if (rxfh.indir_size != ETH_RXFH_INDIR_NO_CHANGE)
+			dev->priv_flags |= IFF_RXFH_CONFIGURED;
+	}
+
+out:
+	kfree(rss_config);
+	return ret;
+}
+
+static int ethtool_get_regs(struct net_device *dev, char __user *useraddr)
+{
+	struct ethtool_regs regs;
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+	void *regbuf;
+	int reglen, ret;
+
+	if (!ops->get_regs || !ops->get_regs_len)
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&regs, useraddr, sizeof(regs)))
+		return -EFAULT;
+
+	reglen = ops->get_regs_len(dev);
+	if (reglen <= 0)
+		return reglen;
+
+	if (regs.len > reglen)
+		regs.len = reglen;
+
+	regbuf = vzalloc(reglen);
+	if (!regbuf)
+		return -ENOMEM;
+
+	if (regs.len < reglen)
+		reglen = regs.len;
+
+	ops->get_regs(dev, &regs, regbuf);
+
+	ret = -EFAULT;
+	if (copy_to_user(useraddr, &regs, sizeof(regs)))
+		goto out;
+	useraddr += offsetof(struct ethtool_regs, data);
+	if (copy_to_user(useraddr, regbuf, reglen))
+		goto out;
+	ret = 0;
+
+ out:
+	vfree(regbuf);
+	return ret;
+}
+
+static int ethtool_reset(struct net_device *dev, char __user *useraddr)
+{
+	struct ethtool_value reset;
+	int ret;
+
+	if (!dev->ethtool_ops->reset)
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&reset, useraddr, sizeof(reset)))
+		return -EFAULT;
+
+	ret = dev->ethtool_ops->reset(dev, &reset.data);
+	if (ret)
+		return ret;
+
+	if (copy_to_user(useraddr, &reset, sizeof(reset)))
+		return -EFAULT;
+	return 0;
+}
+
+static int ethtool_get_wol(struct net_device *dev, char __user *useraddr)
+{
+	struct ethtool_wolinfo wol;
+
+	if (!dev->ethtool_ops->get_wol)
+		return -EOPNOTSUPP;
+
+	memset(&wol, 0, sizeof(struct ethtool_wolinfo));
+	wol.cmd = ETHTOOL_GWOL;
+	dev->ethtool_ops->get_wol(dev, &wol);
+
+	if (copy_to_user(useraddr, &wol, sizeof(wol)))
+		return -EFAULT;
+	return 0;
+}
+
+static int ethtool_set_wol(struct net_device *dev, char __user *useraddr)
+{
+	struct ethtool_wolinfo wol;
+	int ret;
+
+	if (!dev->ethtool_ops->set_wol)
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&wol, useraddr, sizeof(wol)))
+		return -EFAULT;
+
+	ret = dev->ethtool_ops->set_wol(dev, &wol);
+	if (ret)
+		return ret;
+
+	dev->wol_enabled = !!wol.wolopts;
+
+	return 0;
+}
+
+static int ethtool_get_eee(struct net_device *dev, char __user *useraddr)
+{
+	struct ethtool_eee edata;
+	int rc;
+
+	if (!dev->ethtool_ops->get_eee)
+		return -EOPNOTSUPP;
+
+	memset(&edata, 0, sizeof(struct ethtool_eee));
+	edata.cmd = ETHTOOL_GEEE;
+	rc = dev->ethtool_ops->get_eee(dev, &edata);
+
+	if (rc)
+		return rc;
+
+	if (copy_to_user(useraddr, &edata, sizeof(edata)))
+		return -EFAULT;
+
+	return 0;
+}
+
+static int ethtool_set_eee(struct net_device *dev, char __user *useraddr)
+{
+	struct ethtool_eee edata;
+
+	if (!dev->ethtool_ops->set_eee)
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&edata, useraddr, sizeof(edata)))
+		return -EFAULT;
+
+	return dev->ethtool_ops->set_eee(dev, &edata);
+}
+
+static int ethtool_nway_reset(struct net_device *dev)
+{
+	if (!dev->ethtool_ops->nway_reset)
+		return -EOPNOTSUPP;
+
+	return dev->ethtool_ops->nway_reset(dev);
+}
+
+static int ethtool_get_link(struct net_device *dev, char __user *useraddr)
+{
+	struct ethtool_value edata = { .cmd = ETHTOOL_GLINK };
+
+	if (!dev->ethtool_ops->get_link)
+		return -EOPNOTSUPP;
+
+	edata.data = netif_running(dev) && dev->ethtool_ops->get_link(dev);
+
+	if (copy_to_user(useraddr, &edata, sizeof(edata)))
+		return -EFAULT;
+	return 0;
+}
+
+static int ethtool_get_any_eeprom(struct net_device *dev, void __user *useraddr,
+				  int (*getter)(struct net_device *,
+						struct ethtool_eeprom *, u8 *),
+				  u32 total_len)
+{
+	struct ethtool_eeprom eeprom;
+	void __user *userbuf = useraddr + sizeof(eeprom);
+	u32 bytes_remaining;
+	u8 *data;
+	int ret = 0;
+
+	if (copy_from_user(&eeprom, useraddr, sizeof(eeprom)))
+		return -EFAULT;
+
+	/* Check for wrap and zero */
+	if (eeprom.offset + eeprom.len <= eeprom.offset)
+		return -EINVAL;
+
+	/* Check for exceeding total eeprom len */
+	if (eeprom.offset + eeprom.len > total_len)
+		return -EINVAL;
+
+	data = kmalloc(PAGE_SIZE, GFP_USER);
+	if (!data)
+		return -ENOMEM;
+
+	bytes_remaining = eeprom.len;
+	while (bytes_remaining > 0) {
+		eeprom.len = min(bytes_remaining, (u32)PAGE_SIZE);
+
+		ret = getter(dev, &eeprom, data);
+		if (ret)
+			break;
+		if (copy_to_user(userbuf, data, eeprom.len)) {
+			ret = -EFAULT;
+			break;
+		}
+		userbuf += eeprom.len;
+		eeprom.offset += eeprom.len;
+		bytes_remaining -= eeprom.len;
+	}
+
+	eeprom.len = userbuf - (useraddr + sizeof(eeprom));
+	eeprom.offset -= eeprom.len;
+	if (copy_to_user(useraddr, &eeprom, sizeof(eeprom)))
+		ret = -EFAULT;
+
+	kfree(data);
+	return ret;
+}
+
+static int ethtool_get_eeprom(struct net_device *dev, void __user *useraddr)
+{
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+
+	if (!ops->get_eeprom || !ops->get_eeprom_len ||
+	    !ops->get_eeprom_len(dev))
+		return -EOPNOTSUPP;
+
+	return ethtool_get_any_eeprom(dev, useraddr, ops->get_eeprom,
+				      ops->get_eeprom_len(dev));
+}
+
+static int ethtool_set_eeprom(struct net_device *dev, void __user *useraddr)
+{
+	struct ethtool_eeprom eeprom;
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+	void __user *userbuf = useraddr + sizeof(eeprom);
+	u32 bytes_remaining;
+	u8 *data;
+	int ret = 0;
+
+	if (!ops->set_eeprom || !ops->get_eeprom_len ||
+	    !ops->get_eeprom_len(dev))
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&eeprom, useraddr, sizeof(eeprom)))
+		return -EFAULT;
+
+	/* Check for wrap and zero */
+	if (eeprom.offset + eeprom.len <= eeprom.offset)
+		return -EINVAL;
+
+	/* Check for exceeding total eeprom len */
+	if (eeprom.offset + eeprom.len > ops->get_eeprom_len(dev))
+		return -EINVAL;
+
+	data = kmalloc(PAGE_SIZE, GFP_USER);
+	if (!data)
+		return -ENOMEM;
+
+	bytes_remaining = eeprom.len;
+	while (bytes_remaining > 0) {
+		eeprom.len = min(bytes_remaining, (u32)PAGE_SIZE);
+
+		if (copy_from_user(data, userbuf, eeprom.len)) {
+			ret = -EFAULT;
+			break;
+		}
+		ret = ops->set_eeprom(dev, &eeprom, data);
+		if (ret)
+			break;
+		userbuf += eeprom.len;
+		eeprom.offset += eeprom.len;
+		bytes_remaining -= eeprom.len;
+	}
+
+	kfree(data);
+	return ret;
+}
+
+static noinline_for_stack int ethtool_get_coalesce(struct net_device *dev,
+						   void __user *useraddr)
+{
+	struct ethtool_coalesce coalesce = { .cmd = ETHTOOL_GCOALESCE };
+
+	if (!dev->ethtool_ops->get_coalesce)
+		return -EOPNOTSUPP;
+
+	dev->ethtool_ops->get_coalesce(dev, &coalesce);
+
+	if (copy_to_user(useraddr, &coalesce, sizeof(coalesce)))
+		return -EFAULT;
+	return 0;
+}
+
+static noinline_for_stack int ethtool_set_coalesce(struct net_device *dev,
+						   void __user *useraddr)
+{
+	struct ethtool_coalesce coalesce;
+
+	if (!dev->ethtool_ops->set_coalesce)
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&coalesce, useraddr, sizeof(coalesce)))
+		return -EFAULT;
+
+	return dev->ethtool_ops->set_coalesce(dev, &coalesce);
+}
+
+static int ethtool_get_ringparam(struct net_device *dev, void __user *useraddr)
+{
+	struct ethtool_ringparam ringparam = { .cmd = ETHTOOL_GRINGPARAM };
+
+	if (!dev->ethtool_ops->get_ringparam)
+		return -EOPNOTSUPP;
+
+	dev->ethtool_ops->get_ringparam(dev, &ringparam);
+
+	if (copy_to_user(useraddr, &ringparam, sizeof(ringparam)))
+		return -EFAULT;
+	return 0;
+}
+
+static int ethtool_set_ringparam(struct net_device *dev, void __user *useraddr)
+{
+	struct ethtool_ringparam ringparam, max = { .cmd = ETHTOOL_GRINGPARAM };
+
+	if (!dev->ethtool_ops->set_ringparam || !dev->ethtool_ops->get_ringparam)
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&ringparam, useraddr, sizeof(ringparam)))
+		return -EFAULT;
+
+	dev->ethtool_ops->get_ringparam(dev, &max);
+
+	/* ensure new ring parameters are within the maximums */
+	if (ringparam.rx_pending > max.rx_max_pending ||
+	    ringparam.rx_mini_pending > max.rx_mini_max_pending ||
+	    ringparam.rx_jumbo_pending > max.rx_jumbo_max_pending ||
+	    ringparam.tx_pending > max.tx_max_pending)
+		return -EINVAL;
+
+	return dev->ethtool_ops->set_ringparam(dev, &ringparam);
+}
+
+static noinline_for_stack int ethtool_get_channels(struct net_device *dev,
+						   void __user *useraddr)
+{
+	struct ethtool_channels channels = { .cmd = ETHTOOL_GCHANNELS };
+
+	if (!dev->ethtool_ops->get_channels)
+		return -EOPNOTSUPP;
+
+	dev->ethtool_ops->get_channels(dev, &channels);
+
+	if (copy_to_user(useraddr, &channels, sizeof(channels)))
+		return -EFAULT;
+	return 0;
+}
+
+static noinline_for_stack int ethtool_set_channels(struct net_device *dev,
+						   void __user *useraddr)
+{
+	struct ethtool_channels channels, curr = { .cmd = ETHTOOL_GCHANNELS };
+	u16 from_channel, to_channel;
+	u32 max_rx_in_use = 0;
+	unsigned int i;
+
+	if (!dev->ethtool_ops->set_channels || !dev->ethtool_ops->get_channels)
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&channels, useraddr, sizeof(channels)))
+		return -EFAULT;
+
+	dev->ethtool_ops->get_channels(dev, &curr);
+
+	/* ensure new counts are within the maximums */
+	if (channels.rx_count > curr.max_rx ||
+	    channels.tx_count > curr.max_tx ||
+	    channels.combined_count > curr.max_combined ||
+	    channels.other_count > curr.max_other)
+		return -EINVAL;
+
+	/* ensure the new Rx count fits within the configured Rx flow
+	 * indirection table settings */
+	if (netif_is_rxfh_configured(dev) &&
+	    !ethtool_get_max_rxfh_channel(dev, &max_rx_in_use) &&
+	    (channels.combined_count + channels.rx_count) <= max_rx_in_use)
+	    return -EINVAL;
+
+	/* Disabling channels, query zero-copy AF_XDP sockets */
+	from_channel = channels.combined_count +
+		min(channels.rx_count, channels.tx_count);
+	to_channel = curr.combined_count + max(curr.rx_count, curr.tx_count);
+	for (i = from_channel; i < to_channel; i++)
+		if (xdp_get_umem_from_qid(dev, i))
+			return -EINVAL;
+
+	return dev->ethtool_ops->set_channels(dev, &channels);
+}
+
+static int ethtool_get_pauseparam(struct net_device *dev, void __user *useraddr)
+{
+	struct ethtool_pauseparam pauseparam = { .cmd = ETHTOOL_GPAUSEPARAM };
+
+	if (!dev->ethtool_ops->get_pauseparam)
+		return -EOPNOTSUPP;
+
+	dev->ethtool_ops->get_pauseparam(dev, &pauseparam);
+
+	if (copy_to_user(useraddr, &pauseparam, sizeof(pauseparam)))
+		return -EFAULT;
+	return 0;
+}
+
+static int ethtool_set_pauseparam(struct net_device *dev, void __user *useraddr)
+{
+	struct ethtool_pauseparam pauseparam;
+
+	if (!dev->ethtool_ops->set_pauseparam)
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&pauseparam, useraddr, sizeof(pauseparam)))
+		return -EFAULT;
+
+	return dev->ethtool_ops->set_pauseparam(dev, &pauseparam);
+}
+
+static int ethtool_self_test(struct net_device *dev, char __user *useraddr)
+{
+	struct ethtool_test test;
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+	u64 *data;
+	int ret, test_len;
+
+	if (!ops->self_test || !ops->get_sset_count)
+		return -EOPNOTSUPP;
+
+	test_len = ops->get_sset_count(dev, ETH_SS_TEST);
+	if (test_len < 0)
+		return test_len;
+	WARN_ON(test_len == 0);
+
+	if (copy_from_user(&test, useraddr, sizeof(test)))
+		return -EFAULT;
+
+	test.len = test_len;
+	data = kmalloc_array(test_len, sizeof(u64), GFP_USER);
+	if (!data)
+		return -ENOMEM;
+
+	ops->self_test(dev, &test, data);
+
+	ret = -EFAULT;
+	if (copy_to_user(useraddr, &test, sizeof(test)))
+		goto out;
+	useraddr += sizeof(test);
+	if (copy_to_user(useraddr, data, test.len * sizeof(u64)))
+		goto out;
+	ret = 0;
+
+ out:
+	kfree(data);
+	return ret;
+}
+
+static int ethtool_get_strings(struct net_device *dev, void __user *useraddr)
+{
+	struct ethtool_gstrings gstrings;
+	u8 *data;
+	int ret;
+
+	if (copy_from_user(&gstrings, useraddr, sizeof(gstrings)))
+		return -EFAULT;
+
+	ret = __ethtool_get_sset_count(dev, gstrings.string_set);
+	if (ret < 0)
+		return ret;
+	if (ret > S32_MAX / ETH_GSTRING_LEN)
+		return -ENOMEM;
+	WARN_ON_ONCE(!ret);
+
+	gstrings.len = ret;
+
+	if (gstrings.len) {
+		data = vzalloc(array_size(gstrings.len, ETH_GSTRING_LEN));
+		if (!data)
+			return -ENOMEM;
+
+		__ethtool_get_strings(dev, gstrings.string_set, data);
+	} else {
+		data = NULL;
+	}
+
+	ret = -EFAULT;
+	if (copy_to_user(useraddr, &gstrings, sizeof(gstrings)))
+		goto out;
+	useraddr += sizeof(gstrings);
+	if (gstrings.len &&
+	    copy_to_user(useraddr, data, gstrings.len * ETH_GSTRING_LEN))
+		goto out;
+	ret = 0;
+
+out:
+	vfree(data);
+	return ret;
+}
+
+static int ethtool_phys_id(struct net_device *dev, void __user *useraddr)
+{
+	struct ethtool_value id;
+	static bool busy;
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+	int rc;
+
+	if (!ops->set_phys_id)
+		return -EOPNOTSUPP;
+
+	if (busy)
+		return -EBUSY;
+
+	if (copy_from_user(&id, useraddr, sizeof(id)))
+		return -EFAULT;
+
+	rc = ops->set_phys_id(dev, ETHTOOL_ID_ACTIVE);
+	if (rc < 0)
+		return rc;
+
+	/* Drop the RTNL lock while waiting, but prevent reentry or
+	 * removal of the device.
+	 */
+	busy = true;
+	dev_hold(dev);
+	rtnl_unlock();
+
+	if (rc == 0) {
+		/* Driver will handle this itself */
+		schedule_timeout_interruptible(
+			id.data ? (id.data * HZ) : MAX_SCHEDULE_TIMEOUT);
+	} else {
+		/* Driver expects to be called at twice the frequency in rc */
+		int n = rc * 2, i, interval = HZ / n;
+
+		/* Count down seconds */
+		do {
+			/* Count down iterations per second */
+			i = n;
+			do {
+				rtnl_lock();
+				rc = ops->set_phys_id(dev,
+				    (i & 1) ? ETHTOOL_ID_OFF : ETHTOOL_ID_ON);
+				rtnl_unlock();
+				if (rc)
+					break;
+				schedule_timeout_interruptible(interval);
+			} while (!signal_pending(current) && --i != 0);
+		} while (!signal_pending(current) &&
+			 (id.data == 0 || --id.data != 0));
+	}
+
+	rtnl_lock();
+	dev_put(dev);
+	busy = false;
+
+	(void) ops->set_phys_id(dev, ETHTOOL_ID_INACTIVE);
+	return rc;
+}
+
+static int ethtool_get_stats(struct net_device *dev, void __user *useraddr)
+{
+	struct ethtool_stats stats;
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+	u64 *data;
+	int ret, n_stats;
+
+	if (!ops->get_ethtool_stats || !ops->get_sset_count)
+		return -EOPNOTSUPP;
+
+	n_stats = ops->get_sset_count(dev, ETH_SS_STATS);
+	if (n_stats < 0)
+		return n_stats;
+	if (n_stats > S32_MAX / sizeof(u64))
+		return -ENOMEM;
+	WARN_ON_ONCE(!n_stats);
+	if (copy_from_user(&stats, useraddr, sizeof(stats)))
+		return -EFAULT;
+
+	stats.n_stats = n_stats;
+
+	if (n_stats) {
+		data = vzalloc(array_size(n_stats, sizeof(u64)));
+		if (!data)
+			return -ENOMEM;
+		ops->get_ethtool_stats(dev, &stats, data);
+	} else {
+		data = NULL;
+	}
+
+	ret = -EFAULT;
+	if (copy_to_user(useraddr, &stats, sizeof(stats)))
+		goto out;
+	useraddr += sizeof(stats);
+	if (n_stats && copy_to_user(useraddr, data, n_stats * sizeof(u64)))
+		goto out;
+	ret = 0;
+
+ out:
+	vfree(data);
+	return ret;
+}
+
+static int ethtool_get_phy_stats(struct net_device *dev, void __user *useraddr)
+{
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+	struct phy_device *phydev = dev->phydev;
+	struct ethtool_stats stats;
+	u64 *data;
+	int ret, n_stats;
+
+	if (!phydev && (!ops->get_ethtool_phy_stats || !ops->get_sset_count))
+		return -EOPNOTSUPP;
+
+	if (dev->phydev && !ops->get_ethtool_phy_stats)
+		n_stats = phy_ethtool_get_sset_count(dev->phydev);
+	else
+		n_stats = ops->get_sset_count(dev, ETH_SS_PHY_STATS);
+	if (n_stats < 0)
+		return n_stats;
+	if (n_stats > S32_MAX / sizeof(u64))
+		return -ENOMEM;
+	WARN_ON_ONCE(!n_stats);
+
+	if (copy_from_user(&stats, useraddr, sizeof(stats)))
+		return -EFAULT;
+
+	stats.n_stats = n_stats;
+
+	if (n_stats) {
+		data = vzalloc(array_size(n_stats, sizeof(u64)));
+		if (!data)
+			return -ENOMEM;
+
+		if (dev->phydev && !ops->get_ethtool_phy_stats) {
+			ret = phy_ethtool_get_stats(dev->phydev, &stats, data);
+			if (ret < 0)
+				goto out;
+		} else {
+			ops->get_ethtool_phy_stats(dev, &stats, data);
+		}
+	} else {
+		data = NULL;
+	}
+
+	ret = -EFAULT;
+	if (copy_to_user(useraddr, &stats, sizeof(stats)))
+		goto out;
+	useraddr += sizeof(stats);
+	if (n_stats && copy_to_user(useraddr, data, n_stats * sizeof(u64)))
+		goto out;
+	ret = 0;
+
+ out:
+	vfree(data);
+	return ret;
+}
+
+static int ethtool_get_perm_addr(struct net_device *dev, void __user *useraddr)
+{
+	struct ethtool_perm_addr epaddr;
+
+	if (copy_from_user(&epaddr, useraddr, sizeof(epaddr)))
+		return -EFAULT;
+
+	if (epaddr.size < dev->addr_len)
+		return -ETOOSMALL;
+	epaddr.size = dev->addr_len;
+
+	if (copy_to_user(useraddr, &epaddr, sizeof(epaddr)))
+		return -EFAULT;
+	useraddr += sizeof(epaddr);
+	if (copy_to_user(useraddr, dev->perm_addr, epaddr.size))
+		return -EFAULT;
+	return 0;
+}
+
+static int ethtool_get_value(struct net_device *dev, char __user *useraddr,
+			     u32 cmd, u32 (*actor)(struct net_device *))
+{
+	struct ethtool_value edata = { .cmd = cmd };
+
+	if (!actor)
+		return -EOPNOTSUPP;
+
+	edata.data = actor(dev);
+
+	if (copy_to_user(useraddr, &edata, sizeof(edata)))
+		return -EFAULT;
+	return 0;
+}
+
+static int ethtool_set_value_void(struct net_device *dev, char __user *useraddr,
+			     void (*actor)(struct net_device *, u32))
+{
+	struct ethtool_value edata;
+
+	if (!actor)
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&edata, useraddr, sizeof(edata)))
+		return -EFAULT;
+
+	actor(dev, edata.data);
+	return 0;
+}
+
+static int ethtool_set_value(struct net_device *dev, char __user *useraddr,
+			     int (*actor)(struct net_device *, u32))
+{
+	struct ethtool_value edata;
+
+	if (!actor)
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&edata, useraddr, sizeof(edata)))
+		return -EFAULT;
+
+	return actor(dev, edata.data);
+}
+
+static noinline_for_stack int ethtool_flash_device(struct net_device *dev,
+						   char __user *useraddr)
+{
+	struct ethtool_flash efl;
+
+	if (copy_from_user(&efl, useraddr, sizeof(efl)))
+		return -EFAULT;
+	efl.data[ETHTOOL_FLASH_MAX_FILENAME - 1] = 0;
+
+	if (!dev->ethtool_ops->flash_device)
+		return devlink_compat_flash_update(dev, efl.data);
+
+	return dev->ethtool_ops->flash_device(dev, &efl);
+}
+
+static int ethtool_set_dump(struct net_device *dev,
+			void __user *useraddr)
+{
+	struct ethtool_dump dump;
+
+	if (!dev->ethtool_ops->set_dump)
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&dump, useraddr, sizeof(dump)))
+		return -EFAULT;
+
+	return dev->ethtool_ops->set_dump(dev, &dump);
+}
+
+static int ethtool_get_dump_flag(struct net_device *dev,
+				void __user *useraddr)
+{
+	int ret;
+	struct ethtool_dump dump;
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+
+	if (!ops->get_dump_flag)
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&dump, useraddr, sizeof(dump)))
+		return -EFAULT;
+
+	ret = ops->get_dump_flag(dev, &dump);
+	if (ret)
+		return ret;
+
+	if (copy_to_user(useraddr, &dump, sizeof(dump)))
+		return -EFAULT;
+	return 0;
+}
+
+static int ethtool_get_dump_data(struct net_device *dev,
+				void __user *useraddr)
+{
+	int ret;
+	__u32 len;
+	struct ethtool_dump dump, tmp;
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+	void *data = NULL;
+
+	if (!ops->get_dump_data || !ops->get_dump_flag)
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&dump, useraddr, sizeof(dump)))
+		return -EFAULT;
+
+	memset(&tmp, 0, sizeof(tmp));
+	tmp.cmd = ETHTOOL_GET_DUMP_FLAG;
+	ret = ops->get_dump_flag(dev, &tmp);
+	if (ret)
+		return ret;
+
+	len = min(tmp.len, dump.len);
+	if (!len)
+		return -EFAULT;
+
+	/* Don't ever let the driver think there's more space available
+	 * than it requested with .get_dump_flag().
+	 */
+	dump.len = len;
+
+	/* Always allocate enough space to hold the whole thing so that the
+	 * driver does not need to check the length and bother with partial
+	 * dumping.
+	 */
+	data = vzalloc(tmp.len);
+	if (!data)
+		return -ENOMEM;
+	ret = ops->get_dump_data(dev, &dump, data);
+	if (ret)
+		goto out;
+
+	/* There are two sane possibilities:
+	 * 1. The driver's .get_dump_data() does not touch dump.len.
+	 * 2. Or it may set dump.len to how much it really writes, which
+	 *    should be tmp.len (or len if it can do a partial dump).
+	 * In any case respond to userspace with the actual length of data
+	 * it's receiving.
+	 */
+	WARN_ON(dump.len != len && dump.len != tmp.len);
+	dump.len = len;
+
+	if (copy_to_user(useraddr, &dump, sizeof(dump))) {
+		ret = -EFAULT;
+		goto out;
+	}
+	useraddr += offsetof(struct ethtool_dump, data);
+	if (copy_to_user(useraddr, data, len))
+		ret = -EFAULT;
+out:
+	vfree(data);
+	return ret;
+}
+
+static int ethtool_get_ts_info(struct net_device *dev, void __user *useraddr)
+{
+	int err = 0;
+	struct ethtool_ts_info info;
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+	struct phy_device *phydev = dev->phydev;
+
+	memset(&info, 0, sizeof(info));
+	info.cmd = ETHTOOL_GET_TS_INFO;
+
+	if (phydev && phydev->drv && phydev->drv->ts_info) {
+		err = phydev->drv->ts_info(phydev, &info);
+	} else if (ops->get_ts_info) {
+		err = ops->get_ts_info(dev, &info);
+	} else {
+		info.so_timestamping =
+			SOF_TIMESTAMPING_RX_SOFTWARE |
+			SOF_TIMESTAMPING_SOFTWARE;
+		info.phc_index = -1;
+	}
+
+	if (err)
+		return err;
+
+	if (copy_to_user(useraddr, &info, sizeof(info)))
+		err = -EFAULT;
+
+	return err;
+}
+
+static int __ethtool_get_module_info(struct net_device *dev,
+				     struct ethtool_modinfo *modinfo)
+{
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+	struct phy_device *phydev = dev->phydev;
+
+	if (dev->sfp_bus)
+		return sfp_get_module_info(dev->sfp_bus, modinfo);
+
+	if (phydev && phydev->drv && phydev->drv->module_info)
+		return phydev->drv->module_info(phydev, modinfo);
+
+	if (ops->get_module_info)
+		return ops->get_module_info(dev, modinfo);
+
+	return -EOPNOTSUPP;
+}
+
+static int ethtool_get_module_info(struct net_device *dev,
+				   void __user *useraddr)
+{
+	int ret;
+	struct ethtool_modinfo modinfo;
+
+	if (copy_from_user(&modinfo, useraddr, sizeof(modinfo)))
+		return -EFAULT;
+
+	ret = __ethtool_get_module_info(dev, &modinfo);
+	if (ret)
+		return ret;
+
+	if (copy_to_user(useraddr, &modinfo, sizeof(modinfo)))
+		return -EFAULT;
+
+	return 0;
+}
+
+static int __ethtool_get_module_eeprom(struct net_device *dev,
+				       struct ethtool_eeprom *ee, u8 *data)
+{
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+	struct phy_device *phydev = dev->phydev;
+
+	if (dev->sfp_bus)
+		return sfp_get_module_eeprom(dev->sfp_bus, ee, data);
+
+	if (phydev && phydev->drv && phydev->drv->module_eeprom)
+		return phydev->drv->module_eeprom(phydev, ee, data);
+
+	if (ops->get_module_eeprom)
+		return ops->get_module_eeprom(dev, ee, data);
+
+	return -EOPNOTSUPP;
+}
+
+static int ethtool_get_module_eeprom(struct net_device *dev,
+				     void __user *useraddr)
+{
+	int ret;
+	struct ethtool_modinfo modinfo;
+
+	ret = __ethtool_get_module_info(dev, &modinfo);
+	if (ret)
+		return ret;
+
+	return ethtool_get_any_eeprom(dev, useraddr,
+				      __ethtool_get_module_eeprom,
+				      modinfo.eeprom_len);
+}
+
+static int ethtool_tunable_valid(const struct ethtool_tunable *tuna)
+{
+	switch (tuna->id) {
+	case ETHTOOL_RX_COPYBREAK:
+	case ETHTOOL_TX_COPYBREAK:
+		if (tuna->len != sizeof(u32) ||
+		    tuna->type_id != ETHTOOL_TUNABLE_U32)
+			return -EINVAL;
+		break;
+	case ETHTOOL_PFC_PREVENTION_TOUT:
+		if (tuna->len != sizeof(u16) ||
+		    tuna->type_id != ETHTOOL_TUNABLE_U16)
+			return -EINVAL;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int ethtool_get_tunable(struct net_device *dev, void __user *useraddr)
+{
+	int ret;
+	struct ethtool_tunable tuna;
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+	void *data;
+
+	if (!ops->get_tunable)
+		return -EOPNOTSUPP;
+	if (copy_from_user(&tuna, useraddr, sizeof(tuna)))
+		return -EFAULT;
+	ret = ethtool_tunable_valid(&tuna);
+	if (ret)
+		return ret;
+	data = kmalloc(tuna.len, GFP_USER);
+	if (!data)
+		return -ENOMEM;
+	ret = ops->get_tunable(dev, &tuna, data);
+	if (ret)
+		goto out;
+	useraddr += sizeof(tuna);
+	ret = -EFAULT;
+	if (copy_to_user(useraddr, data, tuna.len))
+		goto out;
+	ret = 0;
+
+out:
+	kfree(data);
+	return ret;
+}
+
+static int ethtool_set_tunable(struct net_device *dev, void __user *useraddr)
+{
+	int ret;
+	struct ethtool_tunable tuna;
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+	void *data;
+
+	if (!ops->set_tunable)
+		return -EOPNOTSUPP;
+	if (copy_from_user(&tuna, useraddr, sizeof(tuna)))
+		return -EFAULT;
+	ret = ethtool_tunable_valid(&tuna);
+	if (ret)
+		return ret;
+	useraddr += sizeof(tuna);
+	data = memdup_user(useraddr, tuna.len);
+	if (IS_ERR(data))
+		return PTR_ERR(data);
+	ret = ops->set_tunable(dev, &tuna, data);
+
+	kfree(data);
+	return ret;
+}
+
+static noinline_for_stack int
+ethtool_get_per_queue_coalesce(struct net_device *dev,
+			       void __user *useraddr,
+			       struct ethtool_per_queue_op *per_queue_opt)
+{
+	u32 bit;
+	int ret;
+	DECLARE_BITMAP(queue_mask, MAX_NUM_QUEUE);
+
+	if (!dev->ethtool_ops->get_per_queue_coalesce)
+		return -EOPNOTSUPP;
+
+	useraddr += sizeof(*per_queue_opt);
+
+	bitmap_from_arr32(queue_mask, per_queue_opt->queue_mask,
+			  MAX_NUM_QUEUE);
+
+	for_each_set_bit(bit, queue_mask, MAX_NUM_QUEUE) {
+		struct ethtool_coalesce coalesce = { .cmd = ETHTOOL_GCOALESCE };
+
+		ret = dev->ethtool_ops->get_per_queue_coalesce(dev, bit, &coalesce);
+		if (ret != 0)
+			return ret;
+		if (copy_to_user(useraddr, &coalesce, sizeof(coalesce)))
+			return -EFAULT;
+		useraddr += sizeof(coalesce);
+	}
+
+	return 0;
+}
+
+static noinline_for_stack int
+ethtool_set_per_queue_coalesce(struct net_device *dev,
+			       void __user *useraddr,
+			       struct ethtool_per_queue_op *per_queue_opt)
+{
+	u32 bit;
+	int i, ret = 0;
+	int n_queue;
+	struct ethtool_coalesce *backup = NULL, *tmp = NULL;
+	DECLARE_BITMAP(queue_mask, MAX_NUM_QUEUE);
+
+	if ((!dev->ethtool_ops->set_per_queue_coalesce) ||
+	    (!dev->ethtool_ops->get_per_queue_coalesce))
+		return -EOPNOTSUPP;
+
+	useraddr += sizeof(*per_queue_opt);
+
+	bitmap_from_arr32(queue_mask, per_queue_opt->queue_mask, MAX_NUM_QUEUE);
+	n_queue = bitmap_weight(queue_mask, MAX_NUM_QUEUE);
+	tmp = backup = kmalloc_array(n_queue, sizeof(*backup), GFP_KERNEL);
+	if (!backup)
+		return -ENOMEM;
+
+	for_each_set_bit(bit, queue_mask, MAX_NUM_QUEUE) {
+		struct ethtool_coalesce coalesce;
+
+		ret = dev->ethtool_ops->get_per_queue_coalesce(dev, bit, tmp);
+		if (ret != 0)
+			goto roll_back;
+
+		tmp++;
+
+		if (copy_from_user(&coalesce, useraddr, sizeof(coalesce))) {
+			ret = -EFAULT;
+			goto roll_back;
+		}
+
+		ret = dev->ethtool_ops->set_per_queue_coalesce(dev, bit, &coalesce);
+		if (ret != 0)
+			goto roll_back;
+
+		useraddr += sizeof(coalesce);
+	}
+
+roll_back:
+	if (ret != 0) {
+		tmp = backup;
+		for_each_set_bit(i, queue_mask, bit) {
+			dev->ethtool_ops->set_per_queue_coalesce(dev, i, tmp);
+			tmp++;
+		}
+	}
+	kfree(backup);
+
+	return ret;
+}
+
+static int noinline_for_stack ethtool_set_per_queue(struct net_device *dev,
+				 void __user *useraddr, u32 sub_cmd)
+{
+	struct ethtool_per_queue_op per_queue_opt;
+
+	if (copy_from_user(&per_queue_opt, useraddr, sizeof(per_queue_opt)))
+		return -EFAULT;
+
+	if (per_queue_opt.sub_command != sub_cmd)
+		return -EINVAL;
+
+	switch (per_queue_opt.sub_command) {
+	case ETHTOOL_GCOALESCE:
+		return ethtool_get_per_queue_coalesce(dev, useraddr, &per_queue_opt);
+	case ETHTOOL_SCOALESCE:
+		return ethtool_set_per_queue_coalesce(dev, useraddr, &per_queue_opt);
+	default:
+		return -EOPNOTSUPP;
+	};
+}
+
+static int ethtool_phy_tunable_valid(const struct ethtool_tunable *tuna)
+{
+	switch (tuna->id) {
+	case ETHTOOL_PHY_DOWNSHIFT:
+	case ETHTOOL_PHY_FAST_LINK_DOWN:
+		if (tuna->len != sizeof(u8) ||
+		    tuna->type_id != ETHTOOL_TUNABLE_U8)
+			return -EINVAL;
+		break;
+	case ETHTOOL_PHY_EDPD:
+		if (tuna->len != sizeof(u16) ||
+		    tuna->type_id != ETHTOOL_TUNABLE_U16)
+			return -EINVAL;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int get_phy_tunable(struct net_device *dev, void __user *useraddr)
+{
+	int ret;
+	struct ethtool_tunable tuna;
+	struct phy_device *phydev = dev->phydev;
+	void *data;
+
+	if (!(phydev && phydev->drv && phydev->drv->get_tunable))
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&tuna, useraddr, sizeof(tuna)))
+		return -EFAULT;
+	ret = ethtool_phy_tunable_valid(&tuna);
+	if (ret)
+		return ret;
+	data = kmalloc(tuna.len, GFP_USER);
+	if (!data)
+		return -ENOMEM;
+	mutex_lock(&phydev->lock);
+	ret = phydev->drv->get_tunable(phydev, &tuna, data);
+	mutex_unlock(&phydev->lock);
+	if (ret)
+		goto out;
+	useraddr += sizeof(tuna);
+	ret = -EFAULT;
+	if (copy_to_user(useraddr, data, tuna.len))
+		goto out;
+	ret = 0;
+
+out:
+	kfree(data);
+	return ret;
+}
+
+static int set_phy_tunable(struct net_device *dev, void __user *useraddr)
+{
+	int ret;
+	struct ethtool_tunable tuna;
+	struct phy_device *phydev = dev->phydev;
+	void *data;
+
+	if (!(phydev && phydev->drv && phydev->drv->set_tunable))
+		return -EOPNOTSUPP;
+	if (copy_from_user(&tuna, useraddr, sizeof(tuna)))
+		return -EFAULT;
+	ret = ethtool_phy_tunable_valid(&tuna);
+	if (ret)
+		return ret;
+	useraddr += sizeof(tuna);
+	data = memdup_user(useraddr, tuna.len);
+	if (IS_ERR(data))
+		return PTR_ERR(data);
+	mutex_lock(&phydev->lock);
+	ret = phydev->drv->set_tunable(phydev, &tuna, data);
+	mutex_unlock(&phydev->lock);
+
+	kfree(data);
+	return ret;
+}
+
+static int ethtool_get_fecparam(struct net_device *dev, void __user *useraddr)
+{
+	struct ethtool_fecparam fecparam = { .cmd = ETHTOOL_GFECPARAM };
+	int rc;
+
+	if (!dev->ethtool_ops->get_fecparam)
+		return -EOPNOTSUPP;
+
+	rc = dev->ethtool_ops->get_fecparam(dev, &fecparam);
+	if (rc)
+		return rc;
+
+	if (copy_to_user(useraddr, &fecparam, sizeof(fecparam)))
+		return -EFAULT;
+	return 0;
+}
+
+static int ethtool_set_fecparam(struct net_device *dev, void __user *useraddr)
+{
+	struct ethtool_fecparam fecparam;
+
+	if (!dev->ethtool_ops->set_fecparam)
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&fecparam, useraddr, sizeof(fecparam)))
+		return -EFAULT;
+
+	return dev->ethtool_ops->set_fecparam(dev, &fecparam);
+}
+
+/* The main entry point in this file.  Called from net/core/dev_ioctl.c */
+
+int dev_ethtool(struct net *net, struct ifreq *ifr)
+{
+	struct net_device *dev = __dev_get_by_name(net, ifr->ifr_name);
+	void __user *useraddr = ifr->ifr_data;
+	u32 ethcmd, sub_cmd;
+	int rc;
+	netdev_features_t old_features;
+
+	if (!dev || !netif_device_present(dev))
+		return -ENODEV;
+
+	if (copy_from_user(&ethcmd, useraddr, sizeof(ethcmd)))
+		return -EFAULT;
+
+	if (ethcmd == ETHTOOL_PERQUEUE) {
+		if (copy_from_user(&sub_cmd, useraddr + sizeof(ethcmd), sizeof(sub_cmd)))
+			return -EFAULT;
+	} else {
+		sub_cmd = ethcmd;
+	}
+	/* Allow some commands to be done by anyone */
+	switch (sub_cmd) {
+	case ETHTOOL_GSET:
+	case ETHTOOL_GDRVINFO:
+	case ETHTOOL_GMSGLVL:
+	case ETHTOOL_GLINK:
+	case ETHTOOL_GCOALESCE:
+	case ETHTOOL_GRINGPARAM:
+	case ETHTOOL_GPAUSEPARAM:
+	case ETHTOOL_GRXCSUM:
+	case ETHTOOL_GTXCSUM:
+	case ETHTOOL_GSG:
+	case ETHTOOL_GSSET_INFO:
+	case ETHTOOL_GSTRINGS:
+	case ETHTOOL_GSTATS:
+	case ETHTOOL_GPHYSTATS:
+	case ETHTOOL_GTSO:
+	case ETHTOOL_GPERMADDR:
+	case ETHTOOL_GUFO:
+	case ETHTOOL_GGSO:
+	case ETHTOOL_GGRO:
+	case ETHTOOL_GFLAGS:
+	case ETHTOOL_GPFLAGS:
+	case ETHTOOL_GRXFH:
+	case ETHTOOL_GRXRINGS:
+	case ETHTOOL_GRXCLSRLCNT:
+	case ETHTOOL_GRXCLSRULE:
+	case ETHTOOL_GRXCLSRLALL:
+	case ETHTOOL_GRXFHINDIR:
+	case ETHTOOL_GRSSH:
+	case ETHTOOL_GFEATURES:
+	case ETHTOOL_GCHANNELS:
+	case ETHTOOL_GET_TS_INFO:
+	case ETHTOOL_GEEE:
+	case ETHTOOL_GTUNABLE:
+	case ETHTOOL_PHY_GTUNABLE:
+	case ETHTOOL_GLINKSETTINGS:
+	case ETHTOOL_GFECPARAM:
+		break;
+	default:
+		if (!ns_capable(net->user_ns, CAP_NET_ADMIN))
+			return -EPERM;
+	}
+
+	if (dev->ethtool_ops->begin) {
+		rc = dev->ethtool_ops->begin(dev);
+		if (rc  < 0)
+			return rc;
+	}
+	old_features = dev->features;
+
+	switch (ethcmd) {
+	case ETHTOOL_GSET:
+		rc = ethtool_get_settings(dev, useraddr);
+		break;
+	case ETHTOOL_SSET:
+		rc = ethtool_set_settings(dev, useraddr);
+		break;
+	case ETHTOOL_GDRVINFO:
+		rc = ethtool_get_drvinfo(dev, useraddr);
+		break;
+	case ETHTOOL_GREGS:
+		rc = ethtool_get_regs(dev, useraddr);
+		break;
+	case ETHTOOL_GWOL:
+		rc = ethtool_get_wol(dev, useraddr);
+		break;
+	case ETHTOOL_SWOL:
+		rc = ethtool_set_wol(dev, useraddr);
+		break;
+	case ETHTOOL_GMSGLVL:
+		rc = ethtool_get_value(dev, useraddr, ethcmd,
+				       dev->ethtool_ops->get_msglevel);
+		break;
+	case ETHTOOL_SMSGLVL:
+		rc = ethtool_set_value_void(dev, useraddr,
+				       dev->ethtool_ops->set_msglevel);
+		break;
+	case ETHTOOL_GEEE:
+		rc = ethtool_get_eee(dev, useraddr);
+		break;
+	case ETHTOOL_SEEE:
+		rc = ethtool_set_eee(dev, useraddr);
+		break;
+	case ETHTOOL_NWAY_RST:
+		rc = ethtool_nway_reset(dev);
+		break;
+	case ETHTOOL_GLINK:
+		rc = ethtool_get_link(dev, useraddr);
+		break;
+	case ETHTOOL_GEEPROM:
+		rc = ethtool_get_eeprom(dev, useraddr);
+		break;
+	case ETHTOOL_SEEPROM:
+		rc = ethtool_set_eeprom(dev, useraddr);
+		break;
+	case ETHTOOL_GCOALESCE:
+		rc = ethtool_get_coalesce(dev, useraddr);
+		break;
+	case ETHTOOL_SCOALESCE:
+		rc = ethtool_set_coalesce(dev, useraddr);
+		break;
+	case ETHTOOL_GRINGPARAM:
+		rc = ethtool_get_ringparam(dev, useraddr);
+		break;
+	case ETHTOOL_SRINGPARAM:
+		rc = ethtool_set_ringparam(dev, useraddr);
+		break;
+	case ETHTOOL_GPAUSEPARAM:
+		rc = ethtool_get_pauseparam(dev, useraddr);
+		break;
+	case ETHTOOL_SPAUSEPARAM:
+		rc = ethtool_set_pauseparam(dev, useraddr);
+		break;
+	case ETHTOOL_TEST:
+		rc = ethtool_self_test(dev, useraddr);
+		break;
+	case ETHTOOL_GSTRINGS:
+		rc = ethtool_get_strings(dev, useraddr);
+		break;
+	case ETHTOOL_PHYS_ID:
+		rc = ethtool_phys_id(dev, useraddr);
+		break;
+	case ETHTOOL_GSTATS:
+		rc = ethtool_get_stats(dev, useraddr);
+		break;
+	case ETHTOOL_GPERMADDR:
+		rc = ethtool_get_perm_addr(dev, useraddr);
+		break;
+	case ETHTOOL_GFLAGS:
+		rc = ethtool_get_value(dev, useraddr, ethcmd,
+					__ethtool_get_flags);
+		break;
+	case ETHTOOL_SFLAGS:
+		rc = ethtool_set_value(dev, useraddr, __ethtool_set_flags);
+		break;
+	case ETHTOOL_GPFLAGS:
+		rc = ethtool_get_value(dev, useraddr, ethcmd,
+				       dev->ethtool_ops->get_priv_flags);
+		break;
+	case ETHTOOL_SPFLAGS:
+		rc = ethtool_set_value(dev, useraddr,
+				       dev->ethtool_ops->set_priv_flags);
+		break;
+	case ETHTOOL_GRXFH:
+	case ETHTOOL_GRXRINGS:
+	case ETHTOOL_GRXCLSRLCNT:
+	case ETHTOOL_GRXCLSRULE:
+	case ETHTOOL_GRXCLSRLALL:
+		rc = ethtool_get_rxnfc(dev, ethcmd, useraddr);
+		break;
+	case ETHTOOL_SRXFH:
+	case ETHTOOL_SRXCLSRLDEL:
+	case ETHTOOL_SRXCLSRLINS:
+		rc = ethtool_set_rxnfc(dev, ethcmd, useraddr);
+		break;
+	case ETHTOOL_FLASHDEV:
+		rc = ethtool_flash_device(dev, useraddr);
+		break;
+	case ETHTOOL_RESET:
+		rc = ethtool_reset(dev, useraddr);
+		break;
+	case ETHTOOL_GSSET_INFO:
+		rc = ethtool_get_sset_info(dev, useraddr);
+		break;
+	case ETHTOOL_GRXFHINDIR:
+		rc = ethtool_get_rxfh_indir(dev, useraddr);
+		break;
+	case ETHTOOL_SRXFHINDIR:
+		rc = ethtool_set_rxfh_indir(dev, useraddr);
+		break;
+	case ETHTOOL_GRSSH:
+		rc = ethtool_get_rxfh(dev, useraddr);
+		break;
+	case ETHTOOL_SRSSH:
+		rc = ethtool_set_rxfh(dev, useraddr);
+		break;
+	case ETHTOOL_GFEATURES:
+		rc = ethtool_get_features(dev, useraddr);
+		break;
+	case ETHTOOL_SFEATURES:
+		rc = ethtool_set_features(dev, useraddr);
+		break;
+	case ETHTOOL_GTXCSUM:
+	case ETHTOOL_GRXCSUM:
+	case ETHTOOL_GSG:
+	case ETHTOOL_GTSO:
+	case ETHTOOL_GGSO:
+	case ETHTOOL_GGRO:
+		rc = ethtool_get_one_feature(dev, useraddr, ethcmd);
+		break;
+	case ETHTOOL_STXCSUM:
+	case ETHTOOL_SRXCSUM:
+	case ETHTOOL_SSG:
+	case ETHTOOL_STSO:
+	case ETHTOOL_SGSO:
+	case ETHTOOL_SGRO:
+		rc = ethtool_set_one_feature(dev, useraddr, ethcmd);
+		break;
+	case ETHTOOL_GCHANNELS:
+		rc = ethtool_get_channels(dev, useraddr);
+		break;
+	case ETHTOOL_SCHANNELS:
+		rc = ethtool_set_channels(dev, useraddr);
+		break;
+	case ETHTOOL_SET_DUMP:
+		rc = ethtool_set_dump(dev, useraddr);
+		break;
+	case ETHTOOL_GET_DUMP_FLAG:
+		rc = ethtool_get_dump_flag(dev, useraddr);
+		break;
+	case ETHTOOL_GET_DUMP_DATA:
+		rc = ethtool_get_dump_data(dev, useraddr);
+		break;
+	case ETHTOOL_GET_TS_INFO:
+		rc = ethtool_get_ts_info(dev, useraddr);
+		break;
+	case ETHTOOL_GMODULEINFO:
+		rc = ethtool_get_module_info(dev, useraddr);
+		break;
+	case ETHTOOL_GMODULEEEPROM:
+		rc = ethtool_get_module_eeprom(dev, useraddr);
+		break;
+	case ETHTOOL_GTUNABLE:
+		rc = ethtool_get_tunable(dev, useraddr);
+		break;
+	case ETHTOOL_STUNABLE:
+		rc = ethtool_set_tunable(dev, useraddr);
+		break;
+	case ETHTOOL_GPHYSTATS:
+		rc = ethtool_get_phy_stats(dev, useraddr);
+		break;
+	case ETHTOOL_PERQUEUE:
+		rc = ethtool_set_per_queue(dev, useraddr, sub_cmd);
+		break;
+	case ETHTOOL_GLINKSETTINGS:
+		rc = ethtool_get_link_ksettings(dev, useraddr);
+		break;
+	case ETHTOOL_SLINKSETTINGS:
+		rc = ethtool_set_link_ksettings(dev, useraddr);
+		break;
+	case ETHTOOL_PHY_GTUNABLE:
+		rc = get_phy_tunable(dev, useraddr);
+		break;
+	case ETHTOOL_PHY_STUNABLE:
+		rc = set_phy_tunable(dev, useraddr);
+		break;
+	case ETHTOOL_GFECPARAM:
+		rc = ethtool_get_fecparam(dev, useraddr);
+		break;
+	case ETHTOOL_SFECPARAM:
+		rc = ethtool_set_fecparam(dev, useraddr);
+		break;
+	default:
+		rc = -EOPNOTSUPP;
+	}
+
+	if (dev->ethtool_ops->complete)
+		dev->ethtool_ops->complete(dev);
+
+	if (old_features != dev->features)
+		netdev_features_change(dev);
+
+	return rc;
+}
+
+struct ethtool_rx_flow_key {
+	struct flow_dissector_key_basic			basic;
+	union {
+		struct flow_dissector_key_ipv4_addrs	ipv4;
+		struct flow_dissector_key_ipv6_addrs	ipv6;
+	};
+	struct flow_dissector_key_ports			tp;
+	struct flow_dissector_key_ip			ip;
+	struct flow_dissector_key_vlan			vlan;
+	struct flow_dissector_key_eth_addrs		eth_addrs;
+} __aligned(BITS_PER_LONG / 8); /* Ensure that we can do comparisons as longs. */
+
+struct ethtool_rx_flow_match {
+	struct flow_dissector		dissector;
+	struct ethtool_rx_flow_key	key;
+	struct ethtool_rx_flow_key	mask;
+};
+
+struct ethtool_rx_flow_rule *
+ethtool_rx_flow_rule_create(const struct ethtool_rx_flow_spec_input *input)
+{
+	const struct ethtool_rx_flow_spec *fs = input->fs;
+	static struct in6_addr zero_addr = {};
+	struct ethtool_rx_flow_match *match;
+	struct ethtool_rx_flow_rule *flow;
+	struct flow_action_entry *act;
+
+	flow = kzalloc(sizeof(struct ethtool_rx_flow_rule) +
+		       sizeof(struct ethtool_rx_flow_match), GFP_KERNEL);
+	if (!flow)
+		return ERR_PTR(-ENOMEM);
+
+	/* ethtool_rx supports only one single action per rule. */
+	flow->rule = flow_rule_alloc(1);
+	if (!flow->rule) {
+		kfree(flow);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	match = (struct ethtool_rx_flow_match *)flow->priv;
+	flow->rule->match.dissector	= &match->dissector;
+	flow->rule->match.mask		= &match->mask;
+	flow->rule->match.key		= &match->key;
+
+	match->mask.basic.n_proto = htons(0xffff);
+
+	switch (fs->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT | FLOW_RSS)) {
+	case ETHER_FLOW: {
+		const struct ethhdr *ether_spec, *ether_m_spec;
+
+		ether_spec = &fs->h_u.ether_spec;
+		ether_m_spec = &fs->m_u.ether_spec;
+
+		if (!is_zero_ether_addr(ether_m_spec->h_source)) {
+			ether_addr_copy(match->key.eth_addrs.src,
+					ether_spec->h_source);
+			ether_addr_copy(match->mask.eth_addrs.src,
+					ether_m_spec->h_source);
+		}
+		if (!is_zero_ether_addr(ether_m_spec->h_dest)) {
+			ether_addr_copy(match->key.eth_addrs.dst,
+					ether_spec->h_dest);
+			ether_addr_copy(match->mask.eth_addrs.dst,
+					ether_m_spec->h_dest);
+		}
+		if (ether_m_spec->h_proto) {
+			match->key.basic.n_proto = ether_spec->h_proto;
+			match->mask.basic.n_proto = ether_m_spec->h_proto;
+		}
+		}
+		break;
+	case TCP_V4_FLOW:
+	case UDP_V4_FLOW: {
+		const struct ethtool_tcpip4_spec *v4_spec, *v4_m_spec;
+
+		match->key.basic.n_proto = htons(ETH_P_IP);
+
+		v4_spec = &fs->h_u.tcp_ip4_spec;
+		v4_m_spec = &fs->m_u.tcp_ip4_spec;
+
+		if (v4_m_spec->ip4src) {
+			match->key.ipv4.src = v4_spec->ip4src;
+			match->mask.ipv4.src = v4_m_spec->ip4src;
+		}
+		if (v4_m_spec->ip4dst) {
+			match->key.ipv4.dst = v4_spec->ip4dst;
+			match->mask.ipv4.dst = v4_m_spec->ip4dst;
+		}
+		if (v4_m_spec->ip4src ||
+		    v4_m_spec->ip4dst) {
+			match->dissector.used_keys |=
+				BIT(FLOW_DISSECTOR_KEY_IPV4_ADDRS);
+			match->dissector.offset[FLOW_DISSECTOR_KEY_IPV4_ADDRS] =
+				offsetof(struct ethtool_rx_flow_key, ipv4);
+		}
+		if (v4_m_spec->psrc) {
+			match->key.tp.src = v4_spec->psrc;
+			match->mask.tp.src = v4_m_spec->psrc;
+		}
+		if (v4_m_spec->pdst) {
+			match->key.tp.dst = v4_spec->pdst;
+			match->mask.tp.dst = v4_m_spec->pdst;
+		}
+		if (v4_m_spec->psrc ||
+		    v4_m_spec->pdst) {
+			match->dissector.used_keys |=
+				BIT(FLOW_DISSECTOR_KEY_PORTS);
+			match->dissector.offset[FLOW_DISSECTOR_KEY_PORTS] =
+				offsetof(struct ethtool_rx_flow_key, tp);
+		}
+		if (v4_m_spec->tos) {
+			match->key.ip.tos = v4_spec->tos;
+			match->mask.ip.tos = v4_m_spec->tos;
+			match->dissector.used_keys |=
+				BIT(FLOW_DISSECTOR_KEY_IP);
+			match->dissector.offset[FLOW_DISSECTOR_KEY_IP] =
+				offsetof(struct ethtool_rx_flow_key, ip);
+		}
+		}
+		break;
+	case TCP_V6_FLOW:
+	case UDP_V6_FLOW: {
+		const struct ethtool_tcpip6_spec *v6_spec, *v6_m_spec;
+
+		match->key.basic.n_proto = htons(ETH_P_IPV6);
+
+		v6_spec = &fs->h_u.tcp_ip6_spec;
+		v6_m_spec = &fs->m_u.tcp_ip6_spec;
+		if (memcmp(v6_m_spec->ip6src, &zero_addr, sizeof(zero_addr))) {
+			memcpy(&match->key.ipv6.src, v6_spec->ip6src,
+			       sizeof(match->key.ipv6.src));
+			memcpy(&match->mask.ipv6.src, v6_m_spec->ip6src,
+			       sizeof(match->mask.ipv6.src));
+		}
+		if (memcmp(v6_m_spec->ip6dst, &zero_addr, sizeof(zero_addr))) {
+			memcpy(&match->key.ipv6.dst, v6_spec->ip6dst,
+			       sizeof(match->key.ipv6.dst));
+			memcpy(&match->mask.ipv6.dst, v6_m_spec->ip6dst,
+			       sizeof(match->mask.ipv6.dst));
+		}
+		if (memcmp(v6_m_spec->ip6src, &zero_addr, sizeof(zero_addr)) ||
+		    memcmp(v6_m_spec->ip6src, &zero_addr, sizeof(zero_addr))) {
+			match->dissector.used_keys |=
+				BIT(FLOW_DISSECTOR_KEY_IPV6_ADDRS);
+			match->dissector.offset[FLOW_DISSECTOR_KEY_IPV6_ADDRS] =
+				offsetof(struct ethtool_rx_flow_key, ipv6);
+		}
+		if (v6_m_spec->psrc) {
+			match->key.tp.src = v6_spec->psrc;
+			match->mask.tp.src = v6_m_spec->psrc;
+		}
+		if (v6_m_spec->pdst) {
+			match->key.tp.dst = v6_spec->pdst;
+			match->mask.tp.dst = v6_m_spec->pdst;
+		}
+		if (v6_m_spec->psrc ||
+		    v6_m_spec->pdst) {
+			match->dissector.used_keys |=
+				BIT(FLOW_DISSECTOR_KEY_PORTS);
+			match->dissector.offset[FLOW_DISSECTOR_KEY_PORTS] =
+				offsetof(struct ethtool_rx_flow_key, tp);
+		}
+		if (v6_m_spec->tclass) {
+			match->key.ip.tos = v6_spec->tclass;
+			match->mask.ip.tos = v6_m_spec->tclass;
+			match->dissector.used_keys |=
+				BIT(FLOW_DISSECTOR_KEY_IP);
+			match->dissector.offset[FLOW_DISSECTOR_KEY_IP] =
+				offsetof(struct ethtool_rx_flow_key, ip);
+		}
+		}
+		break;
+	default:
+		ethtool_rx_flow_rule_destroy(flow);
+		return ERR_PTR(-EINVAL);
+	}
+
+	switch (fs->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT | FLOW_RSS)) {
+	case TCP_V4_FLOW:
+	case TCP_V6_FLOW:
+		match->key.basic.ip_proto = IPPROTO_TCP;
+		break;
+	case UDP_V4_FLOW:
+	case UDP_V6_FLOW:
+		match->key.basic.ip_proto = IPPROTO_UDP;
+		break;
+	}
+	match->mask.basic.ip_proto = 0xff;
+
+	match->dissector.used_keys |= BIT(FLOW_DISSECTOR_KEY_BASIC);
+	match->dissector.offset[FLOW_DISSECTOR_KEY_BASIC] =
+		offsetof(struct ethtool_rx_flow_key, basic);
+
+	if (fs->flow_type & FLOW_EXT) {
+		const struct ethtool_flow_ext *ext_h_spec = &fs->h_ext;
+		const struct ethtool_flow_ext *ext_m_spec = &fs->m_ext;
+
+		if (ext_m_spec->vlan_etype) {
+			match->key.vlan.vlan_tpid = ext_h_spec->vlan_etype;
+			match->mask.vlan.vlan_tpid = ext_m_spec->vlan_etype;
+		}
+
+		if (ext_m_spec->vlan_tci) {
+			match->key.vlan.vlan_id =
+				ntohs(ext_h_spec->vlan_tci) & 0x0fff;
+			match->mask.vlan.vlan_id =
+				ntohs(ext_m_spec->vlan_tci) & 0x0fff;
+
+			match->key.vlan.vlan_dei =
+				!!(ext_h_spec->vlan_tci & htons(0x1000));
+			match->mask.vlan.vlan_dei =
+				!!(ext_m_spec->vlan_tci & htons(0x1000));
+
+			match->key.vlan.vlan_priority =
+				(ntohs(ext_h_spec->vlan_tci) & 0xe000) >> 13;
+			match->mask.vlan.vlan_priority =
+				(ntohs(ext_m_spec->vlan_tci) & 0xe000) >> 13;
+		}
+
+		if (ext_m_spec->vlan_etype ||
+		    ext_m_spec->vlan_tci) {
+			match->dissector.used_keys |=
+				BIT(FLOW_DISSECTOR_KEY_VLAN);
+			match->dissector.offset[FLOW_DISSECTOR_KEY_VLAN] =
+				offsetof(struct ethtool_rx_flow_key, vlan);
+		}
+	}
+	if (fs->flow_type & FLOW_MAC_EXT) {
+		const struct ethtool_flow_ext *ext_h_spec = &fs->h_ext;
+		const struct ethtool_flow_ext *ext_m_spec = &fs->m_ext;
+
+		memcpy(match->key.eth_addrs.dst, ext_h_spec->h_dest,
+		       ETH_ALEN);
+		memcpy(match->mask.eth_addrs.dst, ext_m_spec->h_dest,
+		       ETH_ALEN);
+
+		match->dissector.used_keys |=
+			BIT(FLOW_DISSECTOR_KEY_ETH_ADDRS);
+		match->dissector.offset[FLOW_DISSECTOR_KEY_ETH_ADDRS] =
+			offsetof(struct ethtool_rx_flow_key, eth_addrs);
+	}
+
+	act = &flow->rule->action.entries[0];
+	switch (fs->ring_cookie) {
+	case RX_CLS_FLOW_DISC:
+		act->id = FLOW_ACTION_DROP;
+		break;
+	case RX_CLS_FLOW_WAKE:
+		act->id = FLOW_ACTION_WAKE;
+		break;
+	default:
+		act->id = FLOW_ACTION_QUEUE;
+		if (fs->flow_type & FLOW_RSS)
+			act->queue.ctx = input->rss_ctx;
+
+		act->queue.vf = ethtool_get_flow_spec_ring_vf(fs->ring_cookie);
+		act->queue.index = ethtool_get_flow_spec_ring(fs->ring_cookie);
+		break;
+	}
+
+	return flow;
+}
+EXPORT_SYMBOL(ethtool_rx_flow_rule_create);
+
+void ethtool_rx_flow_rule_destroy(struct ethtool_rx_flow_rule *flow)
+{
+	kfree(flow->rule);
+	kfree(flow);
+}
+EXPORT_SYMBOL(ethtool_rx_flow_rule_destroy);
-- 
cgit v1.2.3-59-g8ed1b


From d44e13108b6da4e1c4778bd654a470208e02ef37 Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Wed, 11 Dec 2019 10:58:29 +0100
Subject: ethtool: move string arrays into common file

Introduce file net/ethtool/common.c for code shared by ioctl and netlink
ethtool interface. Move name tables of features, RSS hash functions,
tunables and PHY tunables into this file.

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ethtool/Makefile |  2 +-
 net/ethtool/common.c | 85 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 net/ethtool/common.h | 17 +++++++++++
 net/ethtool/ioctl.c  | 84 ++-------------------------------------------------
 4 files changed, 105 insertions(+), 83 deletions(-)
 create mode 100644 net/ethtool/common.c
 create mode 100644 net/ethtool/common.h

diff --git a/net/ethtool/Makefile b/net/ethtool/Makefile
index 7e5c9eb85c90..f68387618973 100644
--- a/net/ethtool/Makefile
+++ b/net/ethtool/Makefile
@@ -1,3 +1,3 @@
 # SPDX-License-Identifier: GPL-2.0-only
 
-obj-y		+= ioctl.o
+obj-y		+= ioctl.o common.o
diff --git a/net/ethtool/common.c b/net/ethtool/common.c
new file mode 100644
index 000000000000..8b5e11e7e0a6
--- /dev/null
+++ b/net/ethtool/common.c
@@ -0,0 +1,85 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include "common.h"
+
+const char netdev_features_strings[NETDEV_FEATURE_COUNT][ETH_GSTRING_LEN] = {
+	[NETIF_F_SG_BIT] =               "tx-scatter-gather",
+	[NETIF_F_IP_CSUM_BIT] =          "tx-checksum-ipv4",
+	[NETIF_F_HW_CSUM_BIT] =          "tx-checksum-ip-generic",
+	[NETIF_F_IPV6_CSUM_BIT] =        "tx-checksum-ipv6",
+	[NETIF_F_HIGHDMA_BIT] =          "highdma",
+	[NETIF_F_FRAGLIST_BIT] =         "tx-scatter-gather-fraglist",
+	[NETIF_F_HW_VLAN_CTAG_TX_BIT] =  "tx-vlan-hw-insert",
+
+	[NETIF_F_HW_VLAN_CTAG_RX_BIT] =  "rx-vlan-hw-parse",
+	[NETIF_F_HW_VLAN_CTAG_FILTER_BIT] = "rx-vlan-filter",
+	[NETIF_F_HW_VLAN_STAG_TX_BIT] =  "tx-vlan-stag-hw-insert",
+	[NETIF_F_HW_VLAN_STAG_RX_BIT] =  "rx-vlan-stag-hw-parse",
+	[NETIF_F_HW_VLAN_STAG_FILTER_BIT] = "rx-vlan-stag-filter",
+	[NETIF_F_VLAN_CHALLENGED_BIT] =  "vlan-challenged",
+	[NETIF_F_GSO_BIT] =              "tx-generic-segmentation",
+	[NETIF_F_LLTX_BIT] =             "tx-lockless",
+	[NETIF_F_NETNS_LOCAL_BIT] =      "netns-local",
+	[NETIF_F_GRO_BIT] =              "rx-gro",
+	[NETIF_F_GRO_HW_BIT] =           "rx-gro-hw",
+	[NETIF_F_LRO_BIT] =              "rx-lro",
+
+	[NETIF_F_TSO_BIT] =              "tx-tcp-segmentation",
+	[NETIF_F_GSO_ROBUST_BIT] =       "tx-gso-robust",
+	[NETIF_F_TSO_ECN_BIT] =          "tx-tcp-ecn-segmentation",
+	[NETIF_F_TSO_MANGLEID_BIT] =	 "tx-tcp-mangleid-segmentation",
+	[NETIF_F_TSO6_BIT] =             "tx-tcp6-segmentation",
+	[NETIF_F_FSO_BIT] =              "tx-fcoe-segmentation",
+	[NETIF_F_GSO_GRE_BIT] =		 "tx-gre-segmentation",
+	[NETIF_F_GSO_GRE_CSUM_BIT] =	 "tx-gre-csum-segmentation",
+	[NETIF_F_GSO_IPXIP4_BIT] =	 "tx-ipxip4-segmentation",
+	[NETIF_F_GSO_IPXIP6_BIT] =	 "tx-ipxip6-segmentation",
+	[NETIF_F_GSO_UDP_TUNNEL_BIT] =	 "tx-udp_tnl-segmentation",
+	[NETIF_F_GSO_UDP_TUNNEL_CSUM_BIT] = "tx-udp_tnl-csum-segmentation",
+	[NETIF_F_GSO_PARTIAL_BIT] =	 "tx-gso-partial",
+	[NETIF_F_GSO_SCTP_BIT] =	 "tx-sctp-segmentation",
+	[NETIF_F_GSO_ESP_BIT] =		 "tx-esp-segmentation",
+	[NETIF_F_GSO_UDP_L4_BIT] =	 "tx-udp-segmentation",
+
+	[NETIF_F_FCOE_CRC_BIT] =         "tx-checksum-fcoe-crc",
+	[NETIF_F_SCTP_CRC_BIT] =        "tx-checksum-sctp",
+	[NETIF_F_FCOE_MTU_BIT] =         "fcoe-mtu",
+	[NETIF_F_NTUPLE_BIT] =           "rx-ntuple-filter",
+	[NETIF_F_RXHASH_BIT] =           "rx-hashing",
+	[NETIF_F_RXCSUM_BIT] =           "rx-checksum",
+	[NETIF_F_NOCACHE_COPY_BIT] =     "tx-nocache-copy",
+	[NETIF_F_LOOPBACK_BIT] =         "loopback",
+	[NETIF_F_RXFCS_BIT] =            "rx-fcs",
+	[NETIF_F_RXALL_BIT] =            "rx-all",
+	[NETIF_F_HW_L2FW_DOFFLOAD_BIT] = "l2-fwd-offload",
+	[NETIF_F_HW_TC_BIT] =		 "hw-tc-offload",
+	[NETIF_F_HW_ESP_BIT] =		 "esp-hw-offload",
+	[NETIF_F_HW_ESP_TX_CSUM_BIT] =	 "esp-tx-csum-hw-offload",
+	[NETIF_F_RX_UDP_TUNNEL_PORT_BIT] =	 "rx-udp_tunnel-port-offload",
+	[NETIF_F_HW_TLS_RECORD_BIT] =	"tls-hw-record",
+	[NETIF_F_HW_TLS_TX_BIT] =	 "tls-hw-tx-offload",
+	[NETIF_F_HW_TLS_RX_BIT] =	 "tls-hw-rx-offload",
+};
+
+const char
+rss_hash_func_strings[ETH_RSS_HASH_FUNCS_COUNT][ETH_GSTRING_LEN] = {
+	[ETH_RSS_HASH_TOP_BIT] =	"toeplitz",
+	[ETH_RSS_HASH_XOR_BIT] =	"xor",
+	[ETH_RSS_HASH_CRC32_BIT] =	"crc32",
+};
+
+const char
+tunable_strings[__ETHTOOL_TUNABLE_COUNT][ETH_GSTRING_LEN] = {
+	[ETHTOOL_ID_UNSPEC]     = "Unspec",
+	[ETHTOOL_RX_COPYBREAK]	= "rx-copybreak",
+	[ETHTOOL_TX_COPYBREAK]	= "tx-copybreak",
+	[ETHTOOL_PFC_PREVENTION_TOUT] = "pfc-prevention-tout",
+};
+
+const char
+phy_tunable_strings[__ETHTOOL_PHY_TUNABLE_COUNT][ETH_GSTRING_LEN] = {
+	[ETHTOOL_ID_UNSPEC]     = "Unspec",
+	[ETHTOOL_PHY_DOWNSHIFT]	= "phy-downshift",
+	[ETHTOOL_PHY_FAST_LINK_DOWN] = "phy-fast-link-down",
+	[ETHTOOL_PHY_EDPD]	= "phy-energy-detect-power-down",
+};
diff --git a/net/ethtool/common.h b/net/ethtool/common.h
new file mode 100644
index 000000000000..336566430be4
--- /dev/null
+++ b/net/ethtool/common.h
@@ -0,0 +1,17 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+
+#ifndef _ETHTOOL_COMMON_H
+#define _ETHTOOL_COMMON_H
+
+#include <linux/ethtool.h>
+
+extern const char
+netdev_features_strings[NETDEV_FEATURE_COUNT][ETH_GSTRING_LEN];
+extern const char
+rss_hash_func_strings[ETH_RSS_HASH_FUNCS_COUNT][ETH_GSTRING_LEN];
+extern const char
+tunable_strings[__ETHTOOL_TUNABLE_COUNT][ETH_GSTRING_LEN];
+extern const char
+phy_tunable_strings[__ETHTOOL_PHY_TUNABLE_COUNT][ETH_GSTRING_LEN];
+
+#endif /* _ETHTOOL_COMMON_H */
diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
index cd9bc67381b2..b262db5a1d91 100644
--- a/net/ethtool/ioctl.c
+++ b/net/ethtool/ioctl.c
@@ -27,6 +27,8 @@
 #include <net/xdp_sock.h>
 #include <net/flow_offload.h>
 
+#include "common.h"
+
 /*
  * Some useful ethtool_ops methods that're device independent.
  * If we find that all drivers want to do the same thing here,
@@ -54,88 +56,6 @@ EXPORT_SYMBOL(ethtool_op_get_ts_info);
 
 #define ETHTOOL_DEV_FEATURE_WORDS	((NETDEV_FEATURE_COUNT + 31) / 32)
 
-static const char netdev_features_strings[NETDEV_FEATURE_COUNT][ETH_GSTRING_LEN] = {
-	[NETIF_F_SG_BIT] =               "tx-scatter-gather",
-	[NETIF_F_IP_CSUM_BIT] =          "tx-checksum-ipv4",
-	[NETIF_F_HW_CSUM_BIT] =          "tx-checksum-ip-generic",
-	[NETIF_F_IPV6_CSUM_BIT] =        "tx-checksum-ipv6",
-	[NETIF_F_HIGHDMA_BIT] =          "highdma",
-	[NETIF_F_FRAGLIST_BIT] =         "tx-scatter-gather-fraglist",
-	[NETIF_F_HW_VLAN_CTAG_TX_BIT] =  "tx-vlan-hw-insert",
-
-	[NETIF_F_HW_VLAN_CTAG_RX_BIT] =  "rx-vlan-hw-parse",
-	[NETIF_F_HW_VLAN_CTAG_FILTER_BIT] = "rx-vlan-filter",
-	[NETIF_F_HW_VLAN_STAG_TX_BIT] =  "tx-vlan-stag-hw-insert",
-	[NETIF_F_HW_VLAN_STAG_RX_BIT] =  "rx-vlan-stag-hw-parse",
-	[NETIF_F_HW_VLAN_STAG_FILTER_BIT] = "rx-vlan-stag-filter",
-	[NETIF_F_VLAN_CHALLENGED_BIT] =  "vlan-challenged",
-	[NETIF_F_GSO_BIT] =              "tx-generic-segmentation",
-	[NETIF_F_LLTX_BIT] =             "tx-lockless",
-	[NETIF_F_NETNS_LOCAL_BIT] =      "netns-local",
-	[NETIF_F_GRO_BIT] =              "rx-gro",
-	[NETIF_F_GRO_HW_BIT] =           "rx-gro-hw",
-	[NETIF_F_LRO_BIT] =              "rx-lro",
-
-	[NETIF_F_TSO_BIT] =              "tx-tcp-segmentation",
-	[NETIF_F_GSO_ROBUST_BIT] =       "tx-gso-robust",
-	[NETIF_F_TSO_ECN_BIT] =          "tx-tcp-ecn-segmentation",
-	[NETIF_F_TSO_MANGLEID_BIT] =	 "tx-tcp-mangleid-segmentation",
-	[NETIF_F_TSO6_BIT] =             "tx-tcp6-segmentation",
-	[NETIF_F_FSO_BIT] =              "tx-fcoe-segmentation",
-	[NETIF_F_GSO_GRE_BIT] =		 "tx-gre-segmentation",
-	[NETIF_F_GSO_GRE_CSUM_BIT] =	 "tx-gre-csum-segmentation",
-	[NETIF_F_GSO_IPXIP4_BIT] =	 "tx-ipxip4-segmentation",
-	[NETIF_F_GSO_IPXIP6_BIT] =	 "tx-ipxip6-segmentation",
-	[NETIF_F_GSO_UDP_TUNNEL_BIT] =	 "tx-udp_tnl-segmentation",
-	[NETIF_F_GSO_UDP_TUNNEL_CSUM_BIT] = "tx-udp_tnl-csum-segmentation",
-	[NETIF_F_GSO_PARTIAL_BIT] =	 "tx-gso-partial",
-	[NETIF_F_GSO_SCTP_BIT] =	 "tx-sctp-segmentation",
-	[NETIF_F_GSO_ESP_BIT] =		 "tx-esp-segmentation",
-	[NETIF_F_GSO_UDP_L4_BIT] =	 "tx-udp-segmentation",
-
-	[NETIF_F_FCOE_CRC_BIT] =         "tx-checksum-fcoe-crc",
-	[NETIF_F_SCTP_CRC_BIT] =        "tx-checksum-sctp",
-	[NETIF_F_FCOE_MTU_BIT] =         "fcoe-mtu",
-	[NETIF_F_NTUPLE_BIT] =           "rx-ntuple-filter",
-	[NETIF_F_RXHASH_BIT] =           "rx-hashing",
-	[NETIF_F_RXCSUM_BIT] =           "rx-checksum",
-	[NETIF_F_NOCACHE_COPY_BIT] =     "tx-nocache-copy",
-	[NETIF_F_LOOPBACK_BIT] =         "loopback",
-	[NETIF_F_RXFCS_BIT] =            "rx-fcs",
-	[NETIF_F_RXALL_BIT] =            "rx-all",
-	[NETIF_F_HW_L2FW_DOFFLOAD_BIT] = "l2-fwd-offload",
-	[NETIF_F_HW_TC_BIT] =		 "hw-tc-offload",
-	[NETIF_F_HW_ESP_BIT] =		 "esp-hw-offload",
-	[NETIF_F_HW_ESP_TX_CSUM_BIT] =	 "esp-tx-csum-hw-offload",
-	[NETIF_F_RX_UDP_TUNNEL_PORT_BIT] =	 "rx-udp_tunnel-port-offload",
-	[NETIF_F_HW_TLS_RECORD_BIT] =	"tls-hw-record",
-	[NETIF_F_HW_TLS_TX_BIT] =	 "tls-hw-tx-offload",
-	[NETIF_F_HW_TLS_RX_BIT] =	 "tls-hw-rx-offload",
-};
-
-static const char
-rss_hash_func_strings[ETH_RSS_HASH_FUNCS_COUNT][ETH_GSTRING_LEN] = {
-	[ETH_RSS_HASH_TOP_BIT] =	"toeplitz",
-	[ETH_RSS_HASH_XOR_BIT] =	"xor",
-	[ETH_RSS_HASH_CRC32_BIT] =	"crc32",
-};
-
-static const char
-tunable_strings[__ETHTOOL_TUNABLE_COUNT][ETH_GSTRING_LEN] = {
-	[ETHTOOL_ID_UNSPEC]     = "Unspec",
-	[ETHTOOL_RX_COPYBREAK]	= "rx-copybreak",
-	[ETHTOOL_TX_COPYBREAK]	= "tx-copybreak",
-	[ETHTOOL_PFC_PREVENTION_TOUT] = "pfc-prevention-tout",
-};
-
-static const char
-phy_tunable_strings[__ETHTOOL_PHY_TUNABLE_COUNT][ETH_GSTRING_LEN] = {
-	[ETHTOOL_ID_UNSPEC]     = "Unspec",
-	[ETHTOOL_PHY_DOWNSHIFT]	= "phy-downshift",
-	[ETHTOOL_PHY_FAST_LINK_DOWN] = "phy-fast-link-down",
-	[ETHTOOL_PHY_EDPD]	= "phy-energy-detect-power-down",
-};
-
 static int ethtool_get_features(struct net_device *dev, void __user *useraddr)
 {
 	struct ethtool_gfeatures cmd = {
-- 
cgit v1.2.3-59-g8ed1b


From 428c122f5f6b54f40bd51c47495104b534b5a57c Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Wed, 11 Dec 2019 10:58:34 +0100
Subject: ethtool: provide link mode names as a string set

Unlike e.g. netdev features, the ethtool ioctl interface requires link mode
table to be in sync between kernel and userspace for userspace to be able
to display and set all link modes supported by kernel. The way arbitrary
length bitsets are implemented in netlink interface, this will be no longer
needed.

To allow userspace to access all link modes running kernel supports, add
table of ethernet link mode names and make it available as a string set to
userspace GET_STRSET requests. Add build time check to make sure names
are defined for all modes declared in enum ethtool_link_mode_bit_indices.

Once the string set is available, make it also accessible via ioctl.

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/uapi/linux/ethtool.h |  2 ++
 net/ethtool/common.c         | 86 ++++++++++++++++++++++++++++++++++++++++++++
 net/ethtool/common.h         |  5 +++
 net/ethtool/ioctl.c          |  6 ++++
 4 files changed, 99 insertions(+)

diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h
index d4591792f0b4..f44155840b07 100644
--- a/include/uapi/linux/ethtool.h
+++ b/include/uapi/linux/ethtool.h
@@ -593,6 +593,7 @@ struct ethtool_pauseparam {
  * @ETH_SS_RSS_HASH_FUNCS: RSS hush function names
  * @ETH_SS_PHY_STATS: Statistic names, for use with %ETHTOOL_GPHYSTATS
  * @ETH_SS_PHY_TUNABLES: PHY tunable names
+ * @ETH_SS_LINK_MODES: link mode names
  */
 enum ethtool_stringset {
 	ETH_SS_TEST		= 0,
@@ -604,6 +605,7 @@ enum ethtool_stringset {
 	ETH_SS_TUNABLES,
 	ETH_SS_PHY_STATS,
 	ETH_SS_PHY_TUNABLES,
+	ETH_SS_LINK_MODES,
 };
 
 /**
diff --git a/net/ethtool/common.c b/net/ethtool/common.c
index 8b5e11e7e0a6..0a8728565356 100644
--- a/net/ethtool/common.c
+++ b/net/ethtool/common.c
@@ -83,3 +83,89 @@ phy_tunable_strings[__ETHTOOL_PHY_TUNABLE_COUNT][ETH_GSTRING_LEN] = {
 	[ETHTOOL_PHY_FAST_LINK_DOWN] = "phy-fast-link-down",
 	[ETHTOOL_PHY_EDPD]	= "phy-energy-detect-power-down",
 };
+
+#define __LINK_MODE_NAME(speed, type, duplex) \
+	#speed "base" #type "/" #duplex
+#define __DEFINE_LINK_MODE_NAME(speed, type, duplex) \
+	[ETHTOOL_LINK_MODE(speed, type, duplex)] = \
+	__LINK_MODE_NAME(speed, type, duplex)
+#define __DEFINE_SPECIAL_MODE_NAME(_mode, _name) \
+	[ETHTOOL_LINK_MODE_ ## _mode ## _BIT] = _name
+
+const char link_mode_names[][ETH_GSTRING_LEN] = {
+	__DEFINE_LINK_MODE_NAME(10, T, Half),
+	__DEFINE_LINK_MODE_NAME(10, T, Full),
+	__DEFINE_LINK_MODE_NAME(100, T, Half),
+	__DEFINE_LINK_MODE_NAME(100, T, Full),
+	__DEFINE_LINK_MODE_NAME(1000, T, Half),
+	__DEFINE_LINK_MODE_NAME(1000, T, Full),
+	__DEFINE_SPECIAL_MODE_NAME(Autoneg, "Autoneg"),
+	__DEFINE_SPECIAL_MODE_NAME(TP, "TP"),
+	__DEFINE_SPECIAL_MODE_NAME(AUI, "AUI"),
+	__DEFINE_SPECIAL_MODE_NAME(MII, "MII"),
+	__DEFINE_SPECIAL_MODE_NAME(FIBRE, "FIBRE"),
+	__DEFINE_SPECIAL_MODE_NAME(BNC, "BNC"),
+	__DEFINE_LINK_MODE_NAME(10000, T, Full),
+	__DEFINE_SPECIAL_MODE_NAME(Pause, "Pause"),
+	__DEFINE_SPECIAL_MODE_NAME(Asym_Pause, "Asym_Pause"),
+	__DEFINE_LINK_MODE_NAME(2500, X, Full),
+	__DEFINE_SPECIAL_MODE_NAME(Backplane, "Backplane"),
+	__DEFINE_LINK_MODE_NAME(1000, KX, Full),
+	__DEFINE_LINK_MODE_NAME(10000, KX4, Full),
+	__DEFINE_LINK_MODE_NAME(10000, KR, Full),
+	__DEFINE_SPECIAL_MODE_NAME(10000baseR_FEC, "10000baseR_FEC"),
+	__DEFINE_LINK_MODE_NAME(20000, MLD2, Full),
+	__DEFINE_LINK_MODE_NAME(20000, KR2, Full),
+	__DEFINE_LINK_MODE_NAME(40000, KR4, Full),
+	__DEFINE_LINK_MODE_NAME(40000, CR4, Full),
+	__DEFINE_LINK_MODE_NAME(40000, SR4, Full),
+	__DEFINE_LINK_MODE_NAME(40000, LR4, Full),
+	__DEFINE_LINK_MODE_NAME(56000, KR4, Full),
+	__DEFINE_LINK_MODE_NAME(56000, CR4, Full),
+	__DEFINE_LINK_MODE_NAME(56000, SR4, Full),
+	__DEFINE_LINK_MODE_NAME(56000, LR4, Full),
+	__DEFINE_LINK_MODE_NAME(25000, CR, Full),
+	__DEFINE_LINK_MODE_NAME(25000, KR, Full),
+	__DEFINE_LINK_MODE_NAME(25000, SR, Full),
+	__DEFINE_LINK_MODE_NAME(50000, CR2, Full),
+	__DEFINE_LINK_MODE_NAME(50000, KR2, Full),
+	__DEFINE_LINK_MODE_NAME(100000, KR4, Full),
+	__DEFINE_LINK_MODE_NAME(100000, SR4, Full),
+	__DEFINE_LINK_MODE_NAME(100000, CR4, Full),
+	__DEFINE_LINK_MODE_NAME(100000, LR4_ER4, Full),
+	__DEFINE_LINK_MODE_NAME(50000, SR2, Full),
+	__DEFINE_LINK_MODE_NAME(1000, X, Full),
+	__DEFINE_LINK_MODE_NAME(10000, CR, Full),
+	__DEFINE_LINK_MODE_NAME(10000, SR, Full),
+	__DEFINE_LINK_MODE_NAME(10000, LR, Full),
+	__DEFINE_LINK_MODE_NAME(10000, LRM, Full),
+	__DEFINE_LINK_MODE_NAME(10000, ER, Full),
+	__DEFINE_LINK_MODE_NAME(2500, T, Full),
+	__DEFINE_LINK_MODE_NAME(5000, T, Full),
+	__DEFINE_SPECIAL_MODE_NAME(FEC_NONE, "None"),
+	__DEFINE_SPECIAL_MODE_NAME(FEC_RS, "RS"),
+	__DEFINE_SPECIAL_MODE_NAME(FEC_BASER, "BASER"),
+	__DEFINE_LINK_MODE_NAME(50000, KR, Full),
+	__DEFINE_LINK_MODE_NAME(50000, SR, Full),
+	__DEFINE_LINK_MODE_NAME(50000, CR, Full),
+	__DEFINE_LINK_MODE_NAME(50000, LR_ER_FR, Full),
+	__DEFINE_LINK_MODE_NAME(50000, DR, Full),
+	__DEFINE_LINK_MODE_NAME(100000, KR2, Full),
+	__DEFINE_LINK_MODE_NAME(100000, SR2, Full),
+	__DEFINE_LINK_MODE_NAME(100000, CR2, Full),
+	__DEFINE_LINK_MODE_NAME(100000, LR2_ER2_FR2, Full),
+	__DEFINE_LINK_MODE_NAME(100000, DR2, Full),
+	__DEFINE_LINK_MODE_NAME(200000, KR4, Full),
+	__DEFINE_LINK_MODE_NAME(200000, SR4, Full),
+	__DEFINE_LINK_MODE_NAME(200000, LR4_ER4_FR4, Full),
+	__DEFINE_LINK_MODE_NAME(200000, DR4, Full),
+	__DEFINE_LINK_MODE_NAME(200000, CR4, Full),
+	__DEFINE_LINK_MODE_NAME(100, T1, Full),
+	__DEFINE_LINK_MODE_NAME(1000, T1, Full),
+	__DEFINE_LINK_MODE_NAME(400000, KR8, Full),
+	__DEFINE_LINK_MODE_NAME(400000, SR8, Full),
+	__DEFINE_LINK_MODE_NAME(400000, LR8_ER8_FR8, Full),
+	__DEFINE_LINK_MODE_NAME(400000, DR8, Full),
+	__DEFINE_LINK_MODE_NAME(400000, CR8, Full),
+};
+static_assert(ARRAY_SIZE(link_mode_names) == __ETHTOOL_LINK_MODE_MASK_NBITS);
diff --git a/net/ethtool/common.h b/net/ethtool/common.h
index 336566430be4..bbb788908cb1 100644
--- a/net/ethtool/common.h
+++ b/net/ethtool/common.h
@@ -5,6 +5,10 @@
 
 #include <linux/ethtool.h>
 
+/* compose link mode index from speed, type and duplex */
+#define ETHTOOL_LINK_MODE(speed, type, duplex) \
+	ETHTOOL_LINK_MODE_ ## speed ## base ## type ## _ ## duplex ## _BIT
+
 extern const char
 netdev_features_strings[NETDEV_FEATURE_COUNT][ETH_GSTRING_LEN];
 extern const char
@@ -13,5 +17,6 @@ extern const char
 tunable_strings[__ETHTOOL_TUNABLE_COUNT][ETH_GSTRING_LEN];
 extern const char
 phy_tunable_strings[__ETHTOOL_PHY_TUNABLE_COUNT][ETH_GSTRING_LEN];
+extern const char link_mode_names[][ETH_GSTRING_LEN];
 
 #endif /* _ETHTOOL_COMMON_H */
diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
index b262db5a1d91..aed2c2cf1623 100644
--- a/net/ethtool/ioctl.c
+++ b/net/ethtool/ioctl.c
@@ -154,6 +154,9 @@ static int __ethtool_get_sset_count(struct net_device *dev, int sset)
 	    !ops->get_ethtool_phy_stats)
 		return phy_ethtool_get_sset_count(dev->phydev);
 
+	if (sset == ETH_SS_LINK_MODES)
+		return __ETHTOOL_LINK_MODE_MASK_NBITS;
+
 	if (ops->get_sset_count && ops->get_strings)
 		return ops->get_sset_count(dev, sset);
 	else
@@ -178,6 +181,9 @@ static void __ethtool_get_strings(struct net_device *dev,
 	else if (stringset == ETH_SS_PHY_STATS && dev->phydev &&
 		 !ops->get_ethtool_phy_stats)
 		phy_ethtool_get_strings(dev->phydev, data);
+	else if (stringset == ETH_SS_LINK_MODES)
+		memcpy(data, link_mode_names,
+		       __ETHTOOL_LINK_MODE_MASK_NBITS * ETH_GSTRING_LEN);
 	else
 		/* ops->get_strings is valid because checked earlier */
 		ops->get_strings(dev, stringset, data);
-- 
cgit v1.2.3-59-g8ed1b


From 838c93dc5449e5d6378bae117b0a65a122cf7361 Mon Sep 17 00:00:00 2001
From: Sameeh Jubran <sameehj@amazon.com>
Date: Tue, 10 Dec 2019 15:12:12 +0200
Subject: net: ena: implement XDP drop support

This commit implements the basic functionality of drop/pass logic in the
ena driver.

Signed-off-by: Sameeh Jubran <sameehj@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/amazon/ena/ena_netdev.c | 213 +++++++++++++++++++++++++--
 drivers/net/ethernet/amazon/ena/ena_netdev.h |  31 ++++
 2 files changed, 233 insertions(+), 11 deletions(-)

diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c
index d46a912002ff..81fe378295da 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.c
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c
@@ -36,7 +36,6 @@
 #include <linux/cpu_rmap.h>
 #endif /* CONFIG_RFS_ACCEL */
 #include <linux/ethtool.h>
-#include <linux/if_vlan.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/numa.h>
@@ -47,6 +46,7 @@
 #include <net/ip.h>
 
 #include "ena_netdev.h"
+#include <linux/bpf_trace.h>
 #include "ena_pci_id_tbl.h"
 
 static char version[] = DEVICE_NAME " v" DRV_MODULE_VERSION "\n";
@@ -77,6 +77,8 @@ static int ena_rss_init_default(struct ena_adapter *adapter);
 static void check_for_admin_com_state(struct ena_adapter *adapter);
 static void ena_destroy_device(struct ena_adapter *adapter, bool graceful);
 static int ena_restore_device(struct ena_adapter *adapter);
+static void ena_down(struct ena_adapter *adapter);
+static int ena_up(struct ena_adapter *adapter);
 
 static void ena_tx_timeout(struct net_device *dev)
 {
@@ -123,6 +125,115 @@ static int ena_change_mtu(struct net_device *dev, int new_mtu)
 	return ret;
 }
 
+static int ena_xdp_execute(struct ena_ring *rx_ring, struct xdp_buff *xdp)
+{
+	struct bpf_prog *xdp_prog;
+	u32 verdict = XDP_PASS;
+
+	rcu_read_lock();
+	xdp_prog = READ_ONCE(rx_ring->xdp_bpf_prog);
+
+	if (!xdp_prog)
+		goto out;
+
+	verdict = bpf_prog_run_xdp(xdp_prog, xdp);
+
+	if (unlikely(verdict == XDP_ABORTED))
+		trace_xdp_exception(rx_ring->netdev, xdp_prog, verdict);
+	else if (unlikely(verdict >= XDP_TX))
+		bpf_warn_invalid_xdp_action(verdict);
+out:
+	rcu_read_unlock();
+	return verdict;
+}
+
+void ena_xdp_exchange_program_rx_in_range(struct ena_adapter *adapter,
+					  struct bpf_prog *prog,
+					  int first,
+					  int count)
+{
+	struct ena_ring *rx_ring;
+	int i = 0;
+
+	for (i = first; i < count; i++) {
+		rx_ring = &adapter->rx_ring[i];
+		xchg(&rx_ring->xdp_bpf_prog, prog);
+		if (prog)
+			rx_ring->rx_headroom = XDP_PACKET_HEADROOM;
+		else
+			rx_ring->rx_headroom = 0;
+	}
+}
+
+void ena_xdp_exchange_program(struct ena_adapter *adapter,
+			      struct bpf_prog *prog)
+{
+	struct bpf_prog *old_bpf_prog = xchg(&adapter->xdp_bpf_prog, prog);
+
+	ena_xdp_exchange_program_rx_in_range(adapter,
+					     prog,
+					     0,
+					     adapter->num_io_queues);
+
+	if (old_bpf_prog)
+		bpf_prog_put(old_bpf_prog);
+}
+
+static int ena_xdp_set(struct net_device *netdev, struct netdev_bpf *bpf)
+{
+	struct ena_adapter *adapter = netdev_priv(netdev);
+	struct bpf_prog *prog = bpf->prog;
+	int rc, prev_mtu;
+	bool is_up;
+
+	is_up = test_bit(ENA_FLAG_DEV_UP, &adapter->flags);
+
+	if (ena_xdp_allowed(adapter)) {
+		if (is_up)
+			ena_down(adapter);
+
+		ena_xdp_exchange_program(adapter, prog);
+
+		prev_mtu = netdev->max_mtu;
+		netdev->max_mtu = prog ? ENA_XDP_MAX_MTU : adapter->max_mtu;
+		if (is_up) {
+			rc = ena_up(adapter);
+			if (rc)
+				return rc;
+		}
+		netif_info(adapter, drv, adapter->netdev, "xdp program set, changging the max_mtu from %d to %d",
+			   prev_mtu, netdev->max_mtu);
+
+	} else {
+		netif_err(adapter, drv, adapter->netdev, "Failed to set xdp program, the current MTU (%d) is larger than the maximum allowed MTU (%lu) while xdp is on",
+			  netdev->mtu, ENA_XDP_MAX_MTU);
+		NL_SET_ERR_MSG_MOD(bpf->extack, "Failed to set xdp program, the current MTU is larger than the maximum allowed MTU. Check the dmesg for more info");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+/* This is the main xdp callback, it's used by the kernel to set/unset the xdp
+ * program as well as to query the current xdp program id.
+ */
+static int ena_xdp(struct net_device *netdev, struct netdev_bpf *bpf)
+{
+	struct ena_adapter *adapter = netdev_priv(netdev);
+
+	switch (bpf->command) {
+	case XDP_SETUP_PROG:
+		return ena_xdp_set(netdev, bpf);
+	case XDP_QUERY_PROG:
+		bpf->prog_id = adapter->xdp_bpf_prog ?
+			adapter->xdp_bpf_prog->aux->id : 0;
+		break;
+	default:
+		return -EINVAL;
+	}
+	return 0;
+}
+
 static int ena_init_rx_cpu_rmap(struct ena_adapter *adapter)
 {
 #ifdef CONFIG_RFS_ACCEL
@@ -417,6 +528,9 @@ static void ena_free_rx_resources(struct ena_adapter *adapter,
 
 	vfree(rx_ring->free_ids);
 	rx_ring->free_ids = NULL;
+
+	xdp_rxq_info_unreg_mem_model(&rx_ring->xdp_rxq);
+	xdp_rxq_info_unreg(&rx_ring->xdp_rxq);
 }
 
 /* ena_setup_all_rx_resources - allocate I/O Rx queues resources for all queues
@@ -495,7 +609,7 @@ static int ena_alloc_rx_page(struct ena_ring *rx_ring,
 	rx_info->page = page;
 	rx_info->page_offset = 0;
 	ena_buf = &rx_info->ena_buf;
-	ena_buf->paddr = dma;
+	ena_buf->paddr = dma + rx_ring->rx_headroom;
 	ena_buf->len = ENA_PAGE_SIZE;
 
 	return 0;
@@ -1037,6 +1151,33 @@ static void ena_set_rx_hash(struct ena_ring *rx_ring,
 	}
 }
 
+int ena_xdp_handle_buff(struct ena_ring *rx_ring, struct xdp_buff *xdp)
+{
+	struct ena_rx_buffer *rx_info;
+	int ret;
+
+	rx_info = &rx_ring->rx_buffer_info[rx_ring->ena_bufs[0].req_id];
+	xdp->data = page_address(rx_info->page) +
+		rx_info->page_offset + rx_ring->rx_headroom;
+	xdp_set_data_meta_invalid(xdp);
+	xdp->data_hard_start = page_address(rx_info->page);
+	xdp->data_end = xdp->data + rx_ring->ena_bufs[0].len;
+	/* If for some reason we received a bigger packet than
+	 * we expect, then we simply drop it
+	 */
+	if (unlikely(rx_ring->ena_bufs[0].len > ENA_XDP_MAX_MTU))
+		return XDP_DROP;
+
+	ret = ena_xdp_execute(rx_ring, xdp);
+
+	/* The xdp program might expand the headers */
+	if (ret == XDP_PASS) {
+		rx_info->page_offset = xdp->data - xdp->data_hard_start;
+		rx_ring->ena_bufs[0].len = xdp->data_end - xdp->data;
+	}
+
+	return ret;
+}
 /* ena_clean_rx_irq - Cleanup RX irq
  * @rx_ring: RX ring to clean
  * @napi: napi handler
@@ -1052,19 +1193,23 @@ static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi,
 
 	struct ena_com_rx_ctx ena_rx_ctx;
 	struct ena_adapter *adapter;
+	int rx_copybreak_pkt = 0;
+	int refill_threshold;
 	struct sk_buff *skb;
 	int refill_required;
-	int refill_threshold;
-	int rc = 0;
+	struct xdp_buff xdp;
 	int total_len = 0;
-	int rx_copybreak_pkt = 0;
+	int xdp_verdict;
+	int rc = 0;
 	int i;
 
 	netif_dbg(rx_ring->adapter, rx_status, rx_ring->netdev,
 		  "%s qid %d\n", __func__, rx_ring->qid);
 	res_budget = budget;
-
+	xdp.rxq = &rx_ring->xdp_rxq;
 	do {
+		xdp_verdict = XDP_PASS;
+		skb = NULL;
 		ena_rx_ctx.ena_bufs = rx_ring->ena_bufs;
 		ena_rx_ctx.max_bufs = rx_ring->sgl_size;
 		ena_rx_ctx.descs = 0;
@@ -1082,11 +1227,16 @@ static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi,
 			  rx_ring->qid, ena_rx_ctx.descs, ena_rx_ctx.l3_proto,
 			  ena_rx_ctx.l4_proto, ena_rx_ctx.hash);
 
+		if (ena_xdp_present_ring(rx_ring))
+			xdp_verdict = ena_xdp_handle_buff(rx_ring, &xdp);
+
 		/* allocate skb and fill it */
-		skb = ena_rx_skb(rx_ring, rx_ring->ena_bufs, ena_rx_ctx.descs,
-				 &next_to_clean);
+		if (xdp_verdict == XDP_PASS)
+			skb = ena_rx_skb(rx_ring,
+					 rx_ring->ena_bufs,
+					 ena_rx_ctx.descs,
+					 &next_to_clean);
 
-		/* exit if we failed to retrieve a buffer */
 		if (unlikely(!skb)) {
 			for (i = 0; i < ena_rx_ctx.descs; i++) {
 				rx_ring->free_ids[next_to_clean] =
@@ -1095,6 +1245,8 @@ static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi,
 					ENA_RX_RING_IDX_NEXT(next_to_clean,
 							     rx_ring->ring_size);
 			}
+			if (xdp_verdict == XDP_DROP)
+				continue;
 			break;
 		}
 
@@ -1727,12 +1879,34 @@ static int ena_create_io_rx_queue(struct ena_adapter *adapter, int qid)
 		netif_err(adapter, ifup, adapter->netdev,
 			  "Failed to get RX queue handlers. RX queue num %d rc: %d\n",
 			  qid, rc);
-		ena_com_destroy_io_queue(ena_dev, ena_qid);
-		return rc;
+		goto err;
 	}
 
 	ena_com_update_numa_node(rx_ring->ena_com_io_cq, ctx.numa_node);
 
+	rc = xdp_rxq_info_reg(&rx_ring->xdp_rxq, adapter->netdev, qid);
+
+	if (rc) {
+		netif_err(adapter, ifup, adapter->netdev,
+			  "Failed to register xdp rx queue info. RX queue num %d rc: %d\n",
+			  qid, rc);
+		goto err;
+	}
+
+	rc = xdp_rxq_info_reg_mem_model(&rx_ring->xdp_rxq, MEM_TYPE_PAGE_SHARED,
+					NULL);
+
+	if (rc) {
+		netif_err(adapter, ifup, adapter->netdev,
+			  "Failed to register xdp rx queue info memory model. RX queue num %d rc: %d\n",
+			  qid, rc);
+		xdp_rxq_info_unreg(&rx_ring->xdp_rxq);
+		goto err;
+	}
+
+	return rc;
+err:
+	ena_com_destroy_io_queue(ena_dev, ena_qid);
 	return rc;
 }
 
@@ -2056,11 +2230,27 @@ int ena_update_queue_sizes(struct ena_adapter *adapter,
 int ena_update_queue_count(struct ena_adapter *adapter, u32 new_channel_count)
 {
 	struct ena_com_dev *ena_dev = adapter->ena_dev;
+	int prev_channel_count;
 	bool dev_was_up;
 
 	dev_was_up = test_bit(ENA_FLAG_DEV_UP, &adapter->flags);
 	ena_close(adapter->netdev);
+	prev_channel_count = adapter->num_io_queues;
 	adapter->num_io_queues = new_channel_count;
+
+	if (ena_xdp_present(adapter) && ena_xdp_allowed(adapter)) {
+		if (prev_channel_count > new_channel_count)
+			ena_xdp_exchange_program_rx_in_range(adapter,
+							     NULL,
+							     new_channel_count,
+							     prev_channel_count);
+		else
+			ena_xdp_exchange_program_rx_in_range(adapter,
+							     adapter->xdp_bpf_prog,
+							     prev_channel_count,
+							     new_channel_count);
+	}
+
 	/* We need to destroy the rss table so that the indirection
 	 * table will be reinitialized by ena_up()
 	 */
@@ -2568,6 +2758,7 @@ static const struct net_device_ops ena_netdev_ops = {
 	.ndo_change_mtu		= ena_change_mtu,
 	.ndo_set_mac_address	= NULL,
 	.ndo_validate_addr	= eth_validate_addr,
+	.ndo_bpf		= ena_xdp,
 };
 
 static int ena_device_validate_params(struct ena_adapter *adapter,
diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.h b/drivers/net/ethernet/amazon/ena/ena_netdev.h
index bffd778f2ce3..c9016014b86d 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.h
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.h
@@ -36,6 +36,7 @@
 #include <linux/bitops.h>
 #include <linux/dim.h>
 #include <linux/etherdevice.h>
+#include <linux/if_vlan.h>
 #include <linux/inetdevice.h>
 #include <linux/interrupt.h>
 #include <linux/netdevice.h>
@@ -142,6 +143,15 @@
 
 #define ENA_MMIO_DISABLE_REG_READ	BIT(0)
 
+/* The max MTU size is configured to be the ethernet frame size without
+ * the overhead of the ethernet header, which can have a VLAN header, and
+ * a frame check sequence (FCS).
+ * The buffer size we share with the device is defined to be ENA_PAGE_SIZE
+ */
+
+#define ENA_XDP_MAX_MTU (ENA_PAGE_SIZE - ETH_HLEN - ETH_FCS_LEN - \
+				VLAN_HLEN - XDP_PACKET_HEADROOM)
+
 struct ena_irq {
 	irq_handler_t handler;
 	void *data;
@@ -258,10 +268,13 @@ struct ena_ring {
 	struct ena_adapter *adapter;
 	struct ena_com_io_cq *ena_com_io_cq;
 	struct ena_com_io_sq *ena_com_io_sq;
+	struct bpf_prog *xdp_bpf_prog;
+	struct xdp_rxq_info xdp_rxq;
 
 	u16 next_to_use;
 	u16 next_to_clean;
 	u16 rx_copybreak;
+	u16 rx_headroom;
 	u16 qid;
 	u16 mtu;
 	u16 sgl_size;
@@ -379,6 +392,8 @@ struct ena_adapter {
 	u32 last_monitored_tx_qid;
 
 	enum ena_regs_reset_reason_types reset_reason;
+
+	struct bpf_prog *xdp_bpf_prog;
 };
 
 void ena_set_ethtool_ops(struct net_device *netdev);
@@ -390,8 +405,24 @@ void ena_dump_stats_to_buf(struct ena_adapter *adapter, u8 *buf);
 int ena_update_queue_sizes(struct ena_adapter *adapter,
 			   u32 new_tx_size,
 			   u32 new_rx_size);
+
 int ena_update_queue_count(struct ena_adapter *adapter, u32 new_channel_count);
 
 int ena_get_sset_count(struct net_device *netdev, int sset);
 
+static inline bool ena_xdp_present(struct ena_adapter *adapter)
+{
+	return !!adapter->xdp_bpf_prog;
+}
+
+static inline bool ena_xdp_present_ring(struct ena_ring *ring)
+{
+	return !!ring->xdp_bpf_prog;
+}
+
+static inline bool ena_xdp_allowed(struct ena_adapter *adapter)
+{
+	return adapter->netdev->mtu <= ENA_XDP_MAX_MTU;
+}
+
 #endif /* !(ENA_H) */
-- 
cgit v1.2.3-59-g8ed1b


From 548c4940b9f1f527f81509468dd60b61418880b6 Mon Sep 17 00:00:00 2001
From: Sameeh Jubran <sameehj@amazon.com>
Date: Tue, 10 Dec 2019 15:12:13 +0200
Subject: net: ena: Implement XDP_TX action

This commit implements the XDP_TX action in the ena driver. We allocate
separate tx queues for the XDP_TX. We currently allow xdp only when
there is enough queues to allocate for xdp.

Signed-off-by: Sameeh Jubran <sameehj@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/amazon/ena/ena_ethtool.c |   4 +-
 drivers/net/ethernet/amazon/ena/ena_netdev.c  | 838 ++++++++++++++++++++------
 drivers/net/ethernet/amazon/ena/ena_netdev.h  |  45 +-
 3 files changed, 701 insertions(+), 186 deletions(-)

diff --git a/drivers/net/ethernet/amazon/ena/ena_ethtool.c b/drivers/net/ethernet/amazon/ena/ena_ethtool.c
index a3250dcf7d53..745fffd422aa 100644
--- a/drivers/net/ethernet/amazon/ena/ena_ethtool.c
+++ b/drivers/net/ethernet/amazon/ena/ena_ethtool.c
@@ -744,7 +744,9 @@ static int ena_set_channels(struct net_device *netdev,
 	struct ena_adapter *adapter = netdev_priv(netdev);
 	u32 count = channels->combined_count;
 	/* The check for max value is already done in ethtool */
-	if (count < ENA_MIN_NUM_IO_QUEUES)
+	if (count < ENA_MIN_NUM_IO_QUEUES ||
+	    (ena_xdp_present(adapter) &&
+	    !ena_xdp_legal_queue_count(adapter, channels->combined_count)))
 		return -EINVAL;
 
 	return ena_update_queue_count(adapter, count);
diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c
index 81fe378295da..55eeaab7956c 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.c
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c
@@ -77,8 +77,36 @@ static int ena_rss_init_default(struct ena_adapter *adapter);
 static void check_for_admin_com_state(struct ena_adapter *adapter);
 static void ena_destroy_device(struct ena_adapter *adapter, bool graceful);
 static int ena_restore_device(struct ena_adapter *adapter);
-static void ena_down(struct ena_adapter *adapter);
+
+static void ena_init_io_rings(struct ena_adapter *adapter,
+			      int first_index, int count);
+static void ena_init_napi_in_range(struct ena_adapter *adapter, int first_index,
+				   int count);
+static void ena_del_napi_in_range(struct ena_adapter *adapter, int first_index,
+				  int count);
+static int ena_setup_tx_resources(struct ena_adapter *adapter, int qid);
+static int ena_setup_tx_resources_in_range(struct ena_adapter *adapter,
+					   int first_index,
+					   int count);
+static int ena_create_io_tx_queue(struct ena_adapter *adapter, int qid);
+static void ena_free_tx_resources(struct ena_adapter *adapter, int qid);
+static int ena_clean_xdp_irq(struct ena_ring *xdp_ring, u32 budget);
+static void ena_destroy_all_tx_queues(struct ena_adapter *adapter);
+static void ena_free_all_io_tx_resources(struct ena_adapter *adapter);
+static void ena_napi_disable_in_range(struct ena_adapter *adapter,
+				      int first_index, int count);
+static void ena_napi_enable_in_range(struct ena_adapter *adapter,
+				     int first_index, int count);
 static int ena_up(struct ena_adapter *adapter);
+static void ena_down(struct ena_adapter *adapter);
+static void ena_unmask_interrupt(struct ena_ring *tx_ring,
+				 struct ena_ring *rx_ring);
+static void ena_update_ring_numa_node(struct ena_ring *tx_ring,
+				      struct ena_ring *rx_ring);
+static void ena_unmap_tx_buff(struct ena_ring *tx_ring,
+			      struct ena_tx_buffer *tx_info);
+static int ena_create_io_tx_queues_in_range(struct ena_adapter *adapter,
+					    int first_index, int count);
 
 static void ena_tx_timeout(struct net_device *dev)
 {
@@ -125,7 +153,220 @@ static int ena_change_mtu(struct net_device *dev, int new_mtu)
 	return ret;
 }
 
-static int ena_xdp_execute(struct ena_ring *rx_ring, struct xdp_buff *xdp)
+static int ena_xmit_common(struct net_device *dev,
+			   struct ena_ring *ring,
+			   struct ena_tx_buffer *tx_info,
+			   struct ena_com_tx_ctx *ena_tx_ctx,
+			   u16 next_to_use,
+			   u32 bytes)
+{
+	struct ena_adapter *adapter = netdev_priv(dev);
+	int rc, nb_hw_desc;
+
+	if (unlikely(ena_com_is_doorbell_needed(ring->ena_com_io_sq,
+						ena_tx_ctx))) {
+		netif_dbg(adapter, tx_queued, dev,
+			  "llq tx max burst size of queue %d achieved, writing doorbell to send burst\n",
+			  ring->qid);
+		ena_com_write_sq_doorbell(ring->ena_com_io_sq);
+	}
+
+	/* prepare the packet's descriptors to dma engine */
+	rc = ena_com_prepare_tx(ring->ena_com_io_sq, ena_tx_ctx,
+				&nb_hw_desc);
+
+	/* In case there isn't enough space in the queue for the packet,
+	 * we simply drop it. All other failure reasons of
+	 * ena_com_prepare_tx() are fatal and therefore require a device reset.
+	 */
+	if (unlikely(rc)) {
+		netif_err(adapter, tx_queued, dev,
+			  "failed to prepare tx bufs\n");
+		u64_stats_update_begin(&ring->syncp);
+		ring->tx_stats.prepare_ctx_err++;
+		u64_stats_update_end(&ring->syncp);
+		if (rc != -ENOMEM) {
+			adapter->reset_reason =
+				ENA_REGS_RESET_DRIVER_INVALID_STATE;
+			set_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags);
+		}
+		return rc;
+	}
+
+	u64_stats_update_begin(&ring->syncp);
+	ring->tx_stats.cnt++;
+	ring->tx_stats.bytes += bytes;
+	u64_stats_update_end(&ring->syncp);
+
+	tx_info->tx_descs = nb_hw_desc;
+	tx_info->last_jiffies = jiffies;
+	tx_info->print_once = 0;
+
+	ring->next_to_use = ENA_TX_RING_IDX_NEXT(next_to_use,
+						 ring->ring_size);
+	return 0;
+}
+
+/* This is the XDP napi callback. XDP queues use a separate napi callback
+ * than Rx/Tx queues.
+ */
+static int ena_xdp_io_poll(struct napi_struct *napi, int budget)
+{
+	struct ena_napi *ena_napi = container_of(napi, struct ena_napi, napi);
+	u32 xdp_work_done, xdp_budget;
+	struct ena_ring *xdp_ring;
+	int napi_comp_call = 0;
+	int ret;
+
+	xdp_ring = ena_napi->xdp_ring;
+
+	xdp_budget = budget;
+
+	if (!test_bit(ENA_FLAG_DEV_UP, &xdp_ring->adapter->flags) ||
+	    test_bit(ENA_FLAG_TRIGGER_RESET, &xdp_ring->adapter->flags)) {
+		napi_complete_done(napi, 0);
+		return 0;
+	}
+
+	xdp_work_done = ena_clean_xdp_irq(xdp_ring, xdp_budget);
+
+	/* If the device is about to reset or down, avoid unmask
+	 * the interrupt and return 0 so NAPI won't reschedule
+	 */
+	if (unlikely(!test_bit(ENA_FLAG_DEV_UP, &xdp_ring->adapter->flags))) {
+		napi_complete_done(napi, 0);
+		ret = 0;
+	} else if (xdp_budget > xdp_work_done) {
+		napi_comp_call = 1;
+		if (napi_complete_done(napi, xdp_work_done))
+			ena_unmask_interrupt(xdp_ring, NULL);
+		ena_update_ring_numa_node(xdp_ring, NULL);
+		ret = xdp_work_done;
+	} else {
+		ret = xdp_budget;
+	}
+
+	u64_stats_update_begin(&xdp_ring->syncp);
+	xdp_ring->tx_stats.napi_comp += napi_comp_call;
+	xdp_ring->tx_stats.tx_poll++;
+	u64_stats_update_end(&xdp_ring->syncp);
+
+	return ret;
+}
+
+static int ena_xdp_tx_map_buff(struct ena_ring *xdp_ring,
+			       struct ena_tx_buffer *tx_info,
+			       struct xdp_buff *xdp,
+			       void **push_hdr,
+			       u32 *push_len)
+{
+	struct ena_adapter *adapter = xdp_ring->adapter;
+	struct ena_com_buf *ena_buf;
+	dma_addr_t dma = 0;
+	u32 size;
+
+	tx_info->xdpf = convert_to_xdp_frame(xdp);
+	size = tx_info->xdpf->len;
+	ena_buf = tx_info->bufs;
+
+	/* llq push buffer */
+	*push_len = min_t(u32, size, xdp_ring->tx_max_header_size);
+	*push_hdr = tx_info->xdpf->data;
+
+	if (size - *push_len > 0) {
+		dma = dma_map_single(xdp_ring->dev,
+				     *push_hdr + *push_len,
+				     size - *push_len,
+				     DMA_TO_DEVICE);
+		if (unlikely(dma_mapping_error(xdp_ring->dev, dma)))
+			goto error_report_dma_error;
+
+		tx_info->map_linear_data = 1;
+		tx_info->num_of_bufs = 1;
+	}
+
+	ena_buf->paddr = dma;
+	ena_buf->len = size;
+
+	return 0;
+
+error_report_dma_error:
+	u64_stats_update_begin(&xdp_ring->syncp);
+	xdp_ring->tx_stats.dma_mapping_err++;
+	u64_stats_update_end(&xdp_ring->syncp);
+	netdev_warn(adapter->netdev, "failed to map xdp buff\n");
+
+	xdp_return_frame_rx_napi(tx_info->xdpf);
+	tx_info->xdpf = NULL;
+	tx_info->num_of_bufs = 0;
+
+	return -EINVAL;
+}
+
+static int ena_xdp_xmit_buff(struct net_device *dev,
+			     struct xdp_buff *xdp,
+			     int qid,
+			     struct ena_rx_buffer *rx_info)
+{
+	struct ena_adapter *adapter = netdev_priv(dev);
+	struct ena_com_tx_ctx ena_tx_ctx = {0};
+	struct ena_tx_buffer *tx_info;
+	struct ena_ring *xdp_ring;
+	struct ena_ring *rx_ring;
+	u16 next_to_use, req_id;
+	int rc;
+	void *push_hdr;
+	u32 push_len;
+
+	xdp_ring = &adapter->tx_ring[qid];
+	next_to_use = xdp_ring->next_to_use;
+	req_id = xdp_ring->free_ids[next_to_use];
+	tx_info = &xdp_ring->tx_buffer_info[req_id];
+	tx_info->num_of_bufs = 0;
+	rx_ring = &xdp_ring->adapter->rx_ring[qid -
+		  xdp_ring->adapter->xdp_first_ring];
+	page_ref_inc(rx_info->page);
+	tx_info->xdp_rx_page = rx_info->page;
+
+	rc = ena_xdp_tx_map_buff(xdp_ring, tx_info, xdp, &push_hdr, &push_len);
+	if (unlikely(rc))
+		goto error_drop_packet;
+
+	ena_tx_ctx.ena_bufs = tx_info->bufs;
+	ena_tx_ctx.push_header = push_hdr;
+	ena_tx_ctx.num_bufs = tx_info->num_of_bufs;
+	ena_tx_ctx.req_id = req_id;
+	ena_tx_ctx.header_len = push_len;
+
+	rc = ena_xmit_common(dev,
+			     xdp_ring,
+			     tx_info,
+			     &ena_tx_ctx,
+			     next_to_use,
+			     xdp->data_end - xdp->data);
+	if (rc)
+		goto error_unmap_dma;
+	/* trigger the dma engine. ena_com_write_sq_doorbell()
+	 * has a mb
+	 */
+	ena_com_write_sq_doorbell(xdp_ring->ena_com_io_sq);
+	u64_stats_update_begin(&xdp_ring->syncp);
+	xdp_ring->tx_stats.doorbells++;
+	u64_stats_update_end(&xdp_ring->syncp);
+
+	return NETDEV_TX_OK;
+
+error_unmap_dma:
+	ena_unmap_tx_buff(xdp_ring, tx_info);
+	tx_info->xdpf = NULL;
+error_drop_packet:
+
+	return NETDEV_TX_OK;
+}
+
+static int ena_xdp_execute(struct ena_ring *rx_ring,
+			   struct xdp_buff *xdp,
+			   struct ena_rx_buffer *rx_info)
 {
 	struct bpf_prog *xdp_prog;
 	u32 verdict = XDP_PASS;
@@ -138,15 +379,89 @@ static int ena_xdp_execute(struct ena_ring *rx_ring, struct xdp_buff *xdp)
 
 	verdict = bpf_prog_run_xdp(xdp_prog, xdp);
 
-	if (unlikely(verdict == XDP_ABORTED))
+	if (verdict == XDP_TX)
+		ena_xdp_xmit_buff(rx_ring->netdev,
+				  xdp,
+				  rx_ring->qid + rx_ring->adapter->num_io_queues,
+				  rx_info);
+	else if (unlikely(verdict == XDP_ABORTED))
 		trace_xdp_exception(rx_ring->netdev, xdp_prog, verdict);
-	else if (unlikely(verdict >= XDP_TX))
+	else if (unlikely(verdict > XDP_TX))
 		bpf_warn_invalid_xdp_action(verdict);
 out:
 	rcu_read_unlock();
 	return verdict;
 }
 
+static void ena_init_all_xdp_queues(struct ena_adapter *adapter)
+{
+	adapter->xdp_first_ring = adapter->num_io_queues;
+	adapter->xdp_num_queues = adapter->num_io_queues;
+
+	ena_init_io_rings(adapter,
+			  adapter->xdp_first_ring,
+			  adapter->xdp_num_queues);
+}
+
+static int ena_setup_and_create_all_xdp_queues(struct ena_adapter *adapter)
+{
+	int rc = 0;
+
+	rc = ena_setup_tx_resources_in_range(adapter, adapter->xdp_first_ring,
+					     adapter->xdp_num_queues);
+	if (rc)
+		goto setup_err;
+
+	rc = ena_create_io_tx_queues_in_range(adapter,
+					      adapter->xdp_first_ring,
+					      adapter->xdp_num_queues);
+	if (rc)
+		goto create_err;
+
+	return 0;
+
+create_err:
+	ena_free_all_io_tx_resources(adapter);
+setup_err:
+	return rc;
+}
+
+/* Provides a way for both kernel and bpf-prog to know
+ * more about the RX-queue a given XDP frame arrived on.
+ */
+static int ena_xdp_register_rxq_info(struct ena_ring *rx_ring)
+{
+	int rc;
+
+	rc = xdp_rxq_info_reg(&rx_ring->xdp_rxq, rx_ring->netdev, rx_ring->qid);
+
+	if (rc) {
+		netif_err(rx_ring->adapter, ifup, rx_ring->netdev,
+			  "Failed to register xdp rx queue info. RX queue num %d rc: %d\n",
+			  rx_ring->qid, rc);
+		goto err;
+	}
+
+	rc = xdp_rxq_info_reg_mem_model(&rx_ring->xdp_rxq, MEM_TYPE_PAGE_SHARED,
+					NULL);
+
+	if (rc) {
+		netif_err(rx_ring->adapter, ifup, rx_ring->netdev,
+			  "Failed to register xdp rx queue info memory model. RX queue num %d rc: %d\n",
+			  rx_ring->qid, rc);
+		xdp_rxq_info_unreg(&rx_ring->xdp_rxq);
+	}
+
+err:
+	return rc;
+}
+
+static void ena_xdp_unregister_rxq_info(struct ena_ring *rx_ring)
+{
+	xdp_rxq_info_unreg_mem_model(&rx_ring->xdp_rxq);
+	xdp_rxq_info_unreg(&rx_ring->xdp_rxq);
+}
+
 void ena_xdp_exchange_program_rx_in_range(struct ena_adapter *adapter,
 					  struct bpf_prog *prog,
 					  int first,
@@ -158,10 +473,13 @@ void ena_xdp_exchange_program_rx_in_range(struct ena_adapter *adapter,
 	for (i = first; i < count; i++) {
 		rx_ring = &adapter->rx_ring[i];
 		xchg(&rx_ring->xdp_bpf_prog, prog);
-		if (prog)
+		if (prog) {
+			ena_xdp_register_rxq_info(rx_ring);
 			rx_ring->rx_headroom = XDP_PACKET_HEADROOM;
-		else
+		} else {
+			ena_xdp_unregister_rxq_info(rx_ring);
 			rx_ring->rx_headroom = 0;
+		}
 	}
 }
 
@@ -179,35 +497,80 @@ void ena_xdp_exchange_program(struct ena_adapter *adapter,
 		bpf_prog_put(old_bpf_prog);
 }
 
+static int ena_destroy_and_free_all_xdp_queues(struct ena_adapter *adapter)
+{
+	bool was_up;
+	int rc;
+
+	was_up = test_bit(ENA_FLAG_DEV_UP, &adapter->flags);
+
+	if (was_up)
+		ena_down(adapter);
+
+	adapter->xdp_first_ring = 0;
+	adapter->xdp_num_queues = 0;
+	ena_xdp_exchange_program(adapter, NULL);
+	if (was_up) {
+		rc = ena_up(adapter);
+		if (rc)
+			return rc;
+	}
+	return 0;
+}
+
 static int ena_xdp_set(struct net_device *netdev, struct netdev_bpf *bpf)
 {
 	struct ena_adapter *adapter = netdev_priv(netdev);
 	struct bpf_prog *prog = bpf->prog;
+	struct bpf_prog *old_bpf_prog;
 	int rc, prev_mtu;
 	bool is_up;
 
 	is_up = test_bit(ENA_FLAG_DEV_UP, &adapter->flags);
+	rc = ena_xdp_allowed(adapter);
+	if (rc == ENA_XDP_ALLOWED) {
+		old_bpf_prog = adapter->xdp_bpf_prog;
+		if (prog) {
+			if (!is_up) {
+				ena_init_all_xdp_queues(adapter);
+			} else if (!old_bpf_prog) {
+				ena_down(adapter);
+				ena_init_all_xdp_queues(adapter);
+			}
+			ena_xdp_exchange_program(adapter, prog);
 
-	if (ena_xdp_allowed(adapter)) {
-		if (is_up)
-			ena_down(adapter);
-
-		ena_xdp_exchange_program(adapter, prog);
-
-		prev_mtu = netdev->max_mtu;
-		netdev->max_mtu = prog ? ENA_XDP_MAX_MTU : adapter->max_mtu;
-		if (is_up) {
-			rc = ena_up(adapter);
+			if (is_up && !old_bpf_prog) {
+				rc = ena_up(adapter);
+				if (rc)
+					return rc;
+			}
+		} else if (old_bpf_prog) {
+			rc = ena_destroy_and_free_all_xdp_queues(adapter);
 			if (rc)
 				return rc;
 		}
-		netif_info(adapter, drv, adapter->netdev, "xdp program set, changging the max_mtu from %d to %d",
-			   prev_mtu, netdev->max_mtu);
 
-	} else {
-		netif_err(adapter, drv, adapter->netdev, "Failed to set xdp program, the current MTU (%d) is larger than the maximum allowed MTU (%lu) while xdp is on",
+		prev_mtu = netdev->max_mtu;
+		netdev->max_mtu = prog ? ENA_XDP_MAX_MTU : adapter->max_mtu;
+
+		if (!old_bpf_prog)
+			netif_info(adapter, drv, adapter->netdev,
+				   "xdp program set, changing the max_mtu from %d to %d",
+				   prev_mtu, netdev->max_mtu);
+
+	} else if (rc == ENA_XDP_CURRENT_MTU_TOO_LARGE) {
+		netif_err(adapter, drv, adapter->netdev,
+			  "Failed to set xdp program, the current MTU (%d) is larger than the maximum allowed MTU (%lu) while xdp is on",
 			  netdev->mtu, ENA_XDP_MAX_MTU);
-		NL_SET_ERR_MSG_MOD(bpf->extack, "Failed to set xdp program, the current MTU is larger than the maximum allowed MTU. Check the dmesg for more info");
+		NL_SET_ERR_MSG_MOD(bpf->extack,
+				   "Failed to set xdp program, the current MTU is larger than the maximum allowed MTU. Check the dmesg for more info");
+		return -EINVAL;
+	} else if (rc == ENA_XDP_NO_ENOUGH_QUEUES) {
+		netif_err(adapter, drv, adapter->netdev,
+			  "Failed to set xdp program, the Rx/Tx channel count should be at most half of the maximum allowed channel count. The current queue count (%d), the maximal queue count (%d)\n",
+			  adapter->num_io_queues, adapter->max_num_io_queues);
+		NL_SET_ERR_MSG_MOD(bpf->extack,
+				   "Failed to set xdp program, there is no enough space for allocating XDP queues, Check the dmesg for more info");
 		return -EINVAL;
 	}
 
@@ -275,7 +638,8 @@ static void ena_init_io_rings_common(struct ena_adapter *adapter,
 	u64_stats_init(&ring->syncp);
 }
 
-static void ena_init_io_rings(struct ena_adapter *adapter)
+static void ena_init_io_rings(struct ena_adapter *adapter,
+			      int first_index, int count)
 {
 	struct ena_com_dev *ena_dev;
 	struct ena_ring *txr, *rxr;
@@ -283,13 +647,12 @@ static void ena_init_io_rings(struct ena_adapter *adapter)
 
 	ena_dev = adapter->ena_dev;
 
-	for (i = 0; i < adapter->num_io_queues; i++) {
+	for (i = first_index; i < first_index + count; i++) {
 		txr = &adapter->tx_ring[i];
 		rxr = &adapter->rx_ring[i];
 
-		/* TX/RX common ring state */
+		/* TX common ring state */
 		ena_init_io_rings_common(adapter, txr, i);
-		ena_init_io_rings_common(adapter, rxr, i);
 
 		/* TX specific ring state */
 		txr->ring_size = adapter->requested_tx_ring_size;
@@ -299,14 +662,20 @@ static void ena_init_io_rings(struct ena_adapter *adapter)
 		txr->smoothed_interval =
 			ena_com_get_nonadaptive_moderation_interval_tx(ena_dev);
 
-		/* RX specific ring state */
-		rxr->ring_size = adapter->requested_rx_ring_size;
-		rxr->rx_copybreak = adapter->rx_copybreak;
-		rxr->sgl_size = adapter->max_rx_sgl_size;
-		rxr->smoothed_interval =
-			ena_com_get_nonadaptive_moderation_interval_rx(ena_dev);
-		rxr->empty_rx_queue = 0;
-		adapter->ena_napi[i].dim.mode = DIM_CQ_PERIOD_MODE_START_FROM_EQE;
+		/* Don't init RX queues for xdp queues */
+		if (!ENA_IS_XDP_INDEX(adapter, i)) {
+			/* RX common ring state */
+			ena_init_io_rings_common(adapter, rxr, i);
+
+			/* RX specific ring state */
+			rxr->ring_size = adapter->requested_rx_ring_size;
+			rxr->rx_copybreak = adapter->rx_copybreak;
+			rxr->sgl_size = adapter->max_rx_sgl_size;
+			rxr->smoothed_interval =
+				ena_com_get_nonadaptive_moderation_interval_rx(ena_dev);
+			rxr->empty_rx_queue = 0;
+			adapter->ena_napi[i].dim.mode = DIM_CQ_PERIOD_MODE_START_FROM_EQE;
+		}
 	}
 }
 
@@ -396,16 +765,13 @@ static void ena_free_tx_resources(struct ena_adapter *adapter, int qid)
 	tx_ring->push_buf_intermediate_buf = NULL;
 }
 
-/* ena_setup_all_tx_resources - allocate I/O Tx queues resources for All queues
- * @adapter: private structure
- *
- * Return 0 on success, negative on failure
- */
-static int ena_setup_all_tx_resources(struct ena_adapter *adapter)
+static int ena_setup_tx_resources_in_range(struct ena_adapter *adapter,
+					   int first_index,
+					   int count)
 {
 	int i, rc = 0;
 
-	for (i = 0; i < adapter->num_io_queues; i++) {
+	for (i = first_index; i < first_index + count; i++) {
 		rc = ena_setup_tx_resources(adapter, i);
 		if (rc)
 			goto err_setup_tx;
@@ -419,11 +785,20 @@ err_setup_tx:
 		  "Tx queue %d: allocation failed\n", i);
 
 	/* rewind the index freeing the rings as we go */
-	while (i--)
+	while (first_index < i--)
 		ena_free_tx_resources(adapter, i);
 	return rc;
 }
 
+static void ena_free_all_io_tx_resources_in_range(struct ena_adapter *adapter,
+						  int first_index, int count)
+{
+	int i;
+
+	for (i = first_index; i < first_index + count; i++)
+		ena_free_tx_resources(adapter, i);
+}
+
 /* ena_free_all_io_tx_resources - Free I/O Tx Resources for All Queues
  * @adapter: board private structure
  *
@@ -431,10 +806,10 @@ err_setup_tx:
  */
 static void ena_free_all_io_tx_resources(struct ena_adapter *adapter)
 {
-	int i;
-
-	for (i = 0; i < adapter->num_io_queues; i++)
-		ena_free_tx_resources(adapter, i);
+	ena_free_all_io_tx_resources_in_range(adapter,
+					      0,
+					      adapter->xdp_num_queues +
+					      adapter->num_io_queues);
 }
 
 static int validate_rx_req_id(struct ena_ring *rx_ring, u16 req_id)
@@ -528,9 +903,6 @@ static void ena_free_rx_resources(struct ena_adapter *adapter,
 
 	vfree(rx_ring->free_ids);
 	rx_ring->free_ids = NULL;
-
-	xdp_rxq_info_unreg_mem_model(&rx_ring->xdp_rxq);
-	xdp_rxq_info_unreg(&rx_ring->xdp_rxq);
 }
 
 /* ena_setup_all_rx_resources - allocate I/O Rx queues resources for all queues
@@ -610,7 +982,7 @@ static int ena_alloc_rx_page(struct ena_ring *rx_ring,
 	rx_info->page_offset = 0;
 	ena_buf = &rx_info->ena_buf;
 	ena_buf->paddr = dma + rx_ring->rx_headroom;
-	ena_buf->len = ENA_PAGE_SIZE;
+	ena_buf->len = ENA_PAGE_SIZE - rx_ring->rx_headroom;
 
 	return 0;
 }
@@ -627,7 +999,9 @@ static void ena_free_rx_page(struct ena_ring *rx_ring,
 		return;
 	}
 
-	dma_unmap_page(rx_ring->dev, ena_buf->paddr, ENA_PAGE_SIZE,
+	dma_unmap_page(rx_ring->dev,
+		       ena_buf->paddr - rx_ring->rx_headroom,
+		       ENA_PAGE_SIZE,
 		       DMA_FROM_DEVICE);
 
 	__free_page(page);
@@ -734,8 +1108,8 @@ static void ena_free_all_rx_bufs(struct ena_adapter *adapter)
 		ena_free_rx_bufs(adapter, i);
 }
 
-static void ena_unmap_tx_skb(struct ena_ring *tx_ring,
-				    struct ena_tx_buffer *tx_info)
+static void ena_unmap_tx_buff(struct ena_ring *tx_ring,
+			      struct ena_tx_buffer *tx_info)
 {
 	struct ena_com_buf *ena_buf;
 	u32 cnt;
@@ -789,7 +1163,7 @@ static void ena_free_tx_bufs(struct ena_ring *tx_ring)
 				   tx_ring->qid, i);
 		}
 
-		ena_unmap_tx_skb(tx_ring, tx_info);
+		ena_unmap_tx_buff(tx_ring, tx_info);
 
 		dev_kfree_skb_any(tx_info->skb);
 	}
@@ -802,7 +1176,7 @@ static void ena_free_all_tx_bufs(struct ena_adapter *adapter)
 	struct ena_ring *tx_ring;
 	int i;
 
-	for (i = 0; i < adapter->num_io_queues; i++) {
+	for (i = 0; i < adapter->num_io_queues + adapter->xdp_num_queues; i++) {
 		tx_ring = &adapter->tx_ring[i];
 		ena_free_tx_bufs(tx_ring);
 	}
@@ -813,7 +1187,7 @@ static void ena_destroy_all_tx_queues(struct ena_adapter *adapter)
 	u16 ena_qid;
 	int i;
 
-	for (i = 0; i < adapter->num_io_queues; i++) {
+	for (i = 0; i < adapter->num_io_queues + adapter->xdp_num_queues; i++) {
 		ena_qid = ENA_IO_TXQ_IDX(i);
 		ena_com_destroy_io_queue(adapter->ena_dev, ena_qid);
 	}
@@ -837,6 +1211,32 @@ static void ena_destroy_all_io_queues(struct ena_adapter *adapter)
 	ena_destroy_all_rx_queues(adapter);
 }
 
+static int handle_invalid_req_id(struct ena_ring *ring, u16 req_id,
+				 struct ena_tx_buffer *tx_info, bool is_xdp)
+{
+	if (tx_info)
+		netif_err(ring->adapter,
+			  tx_done,
+			  ring->netdev,
+			  "tx_info doesn't have valid %s",
+			   is_xdp ? "xdp frame" : "skb");
+	else
+		netif_err(ring->adapter,
+			  tx_done,
+			  ring->netdev,
+			  "Invalid req_id: %hu\n",
+			  req_id);
+
+	u64_stats_update_begin(&ring->syncp);
+	ring->tx_stats.bad_req_id++;
+	u64_stats_update_end(&ring->syncp);
+
+	/* Trigger device reset */
+	ring->adapter->reset_reason = ENA_REGS_RESET_INV_TX_REQ_ID;
+	set_bit(ENA_FLAG_TRIGGER_RESET, &ring->adapter->flags);
+	return -EFAULT;
+}
+
 static int validate_tx_req_id(struct ena_ring *tx_ring, u16 req_id)
 {
 	struct ena_tx_buffer *tx_info = NULL;
@@ -847,21 +1247,20 @@ static int validate_tx_req_id(struct ena_ring *tx_ring, u16 req_id)
 			return 0;
 	}
 
-	if (tx_info)
-		netif_err(tx_ring->adapter, tx_done, tx_ring->netdev,
-			  "tx_info doesn't have valid skb\n");
-	else
-		netif_err(tx_ring->adapter, tx_done, tx_ring->netdev,
-			  "Invalid req_id: %hu\n", req_id);
+	return handle_invalid_req_id(tx_ring, req_id, tx_info, false);
+}
 
-	u64_stats_update_begin(&tx_ring->syncp);
-	tx_ring->tx_stats.bad_req_id++;
-	u64_stats_update_end(&tx_ring->syncp);
+static int validate_xdp_req_id(struct ena_ring *xdp_ring, u16 req_id)
+{
+	struct ena_tx_buffer *tx_info = NULL;
 
-	/* Trigger device reset */
-	tx_ring->adapter->reset_reason = ENA_REGS_RESET_INV_TX_REQ_ID;
-	set_bit(ENA_FLAG_TRIGGER_RESET, &tx_ring->adapter->flags);
-	return -EFAULT;
+	if (likely(req_id < xdp_ring->ring_size)) {
+		tx_info = &xdp_ring->tx_buffer_info[req_id];
+		if (likely(tx_info->xdpf))
+			return 0;
+	}
+
+	return handle_invalid_req_id(xdp_ring, req_id, tx_info, true);
 }
 
 static int ena_clean_tx_irq(struct ena_ring *tx_ring, u32 budget)
@@ -900,7 +1299,7 @@ static int ena_clean_tx_irq(struct ena_ring *tx_ring, u32 budget)
 		tx_info->skb = NULL;
 		tx_info->last_jiffies = 0;
 
-		ena_unmap_tx_skb(tx_ring, tx_info);
+		ena_unmap_tx_buff(tx_ring, tx_info);
 
 		netif_dbg(tx_ring->adapter, tx_done, tx_ring->netdev,
 			  "tx_poll: q %d skb %p completed\n", tx_ring->qid,
@@ -1168,7 +1567,7 @@ int ena_xdp_handle_buff(struct ena_ring *rx_ring, struct xdp_buff *xdp)
 	if (unlikely(rx_ring->ena_bufs[0].len > ENA_XDP_MAX_MTU))
 		return XDP_DROP;
 
-	ret = ena_xdp_execute(rx_ring, xdp);
+	ret = ena_xdp_execute(rx_ring, xdp, rx_info);
 
 	/* The xdp program might expand the headers */
 	if (ret == XDP_PASS) {
@@ -1189,10 +1588,9 @@ static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi,
 			    u32 budget)
 {
 	u16 next_to_clean = rx_ring->next_to_clean;
-	u32 res_budget, work_done;
-
 	struct ena_com_rx_ctx ena_rx_ctx;
 	struct ena_adapter *adapter;
+	u32 res_budget, work_done;
 	int rx_copybreak_pkt = 0;
 	int refill_threshold;
 	struct sk_buff *skb;
@@ -1207,6 +1605,7 @@ static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi,
 		  "%s qid %d\n", __func__, rx_ring->qid);
 	res_budget = budget;
 	xdp.rxq = &rx_ring->xdp_rxq;
+
 	do {
 		xdp_verdict = XDP_PASS;
 		skb = NULL;
@@ -1238,6 +1637,11 @@ static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi,
 					 &next_to_clean);
 
 		if (unlikely(!skb)) {
+			if (xdp_verdict == XDP_TX) {
+				ena_free_rx_page(rx_ring,
+						 &rx_ring->rx_buffer_info[rx_ring->ena_bufs[0].req_id]);
+				res_budget--;
+			}
 			for (i = 0; i < ena_rx_ctx.descs; i++) {
 				rx_ring->free_ids[next_to_clean] =
 					rx_ring->ena_bufs[i].req_id;
@@ -1245,7 +1649,7 @@ static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi,
 					ENA_RX_RING_IDX_NEXT(next_to_clean,
 							     rx_ring->ring_size);
 			}
-			if (xdp_verdict == XDP_DROP)
+			if (xdp_verdict == XDP_TX || xdp_verdict == XDP_DROP)
 				continue;
 			break;
 		}
@@ -1340,9 +1744,14 @@ static void ena_unmask_interrupt(struct ena_ring *tx_ring,
 					struct ena_ring *rx_ring)
 {
 	struct ena_eth_io_intr_reg intr_reg;
-	u32 rx_interval = ena_com_get_adaptive_moderation_enabled(rx_ring->ena_dev) ?
-		rx_ring->smoothed_interval :
-		ena_com_get_nonadaptive_moderation_interval_rx(rx_ring->ena_dev);
+	u32 rx_interval = 0;
+	/* Rx ring can be NULL when for XDP tx queues which don't have an
+	 * accompanying rx_ring pair.
+	 */
+	if (rx_ring)
+		rx_interval = ena_com_get_adaptive_moderation_enabled(rx_ring->ena_dev) ?
+			rx_ring->smoothed_interval :
+			ena_com_get_nonadaptive_moderation_interval_rx(rx_ring->ena_dev);
 
 	/* Update intr register: rx intr delay,
 	 * tx intr delay and interrupt unmask
@@ -1355,8 +1764,9 @@ static void ena_unmask_interrupt(struct ena_ring *tx_ring,
 	/* It is a shared MSI-X.
 	 * Tx and Rx CQ have pointer to it.
 	 * So we use one of them to reach the intr reg
+	 * The Tx ring is used because the rx_ring is NULL for XDP queues
 	 */
-	ena_com_unmask_intr(rx_ring->ena_com_io_cq, &intr_reg);
+	ena_com_unmask_intr(tx_ring->ena_com_io_cq, &intr_reg);
 }
 
 static void ena_update_ring_numa_node(struct ena_ring *tx_ring,
@@ -1374,22 +1784,82 @@ static void ena_update_ring_numa_node(struct ena_ring *tx_ring,
 
 	if (numa_node != NUMA_NO_NODE) {
 		ena_com_update_numa_node(tx_ring->ena_com_io_cq, numa_node);
-		ena_com_update_numa_node(rx_ring->ena_com_io_cq, numa_node);
+		if (rx_ring)
+			ena_com_update_numa_node(rx_ring->ena_com_io_cq,
+						 numa_node);
 	}
 
 	tx_ring->cpu = cpu;
-	rx_ring->cpu = cpu;
+	if (rx_ring)
+		rx_ring->cpu = cpu;
 
 	return;
 out:
 	put_cpu();
 }
 
+static int ena_clean_xdp_irq(struct ena_ring *xdp_ring, u32 budget)
+{
+	u32 total_done = 0;
+	u16 next_to_clean;
+	u32 tx_bytes = 0;
+	int tx_pkts = 0;
+	u16 req_id;
+	int rc;
+
+	if (unlikely(!xdp_ring))
+		return 0;
+	next_to_clean = xdp_ring->next_to_clean;
+
+	while (tx_pkts < budget) {
+		struct ena_tx_buffer *tx_info;
+		struct xdp_frame *xdpf;
+
+		rc = ena_com_tx_comp_req_id_get(xdp_ring->ena_com_io_cq,
+						&req_id);
+		if (rc)
+			break;
+
+		rc = validate_xdp_req_id(xdp_ring, req_id);
+		if (rc)
+			break;
+
+		tx_info = &xdp_ring->tx_buffer_info[req_id];
+		xdpf = tx_info->xdpf;
+
+		tx_info->xdpf = NULL;
+		tx_info->last_jiffies = 0;
+		ena_unmap_tx_buff(xdp_ring, tx_info);
+
+		netif_dbg(xdp_ring->adapter, tx_done, xdp_ring->netdev,
+			  "tx_poll: q %d skb %p completed\n", xdp_ring->qid,
+			  xdpf);
+
+		tx_bytes += xdpf->len;
+		tx_pkts++;
+		total_done += tx_info->tx_descs;
+
+		__free_page(tx_info->xdp_rx_page);
+		xdp_ring->free_ids[next_to_clean] = req_id;
+		next_to_clean = ENA_TX_RING_IDX_NEXT(next_to_clean,
+						     xdp_ring->ring_size);
+	}
+
+	xdp_ring->next_to_clean = next_to_clean;
+	ena_com_comp_ack(xdp_ring->ena_com_io_sq, total_done);
+	ena_com_update_dev_comp_head(xdp_ring->ena_com_io_cq);
+
+	netif_dbg(xdp_ring->adapter, tx_done, xdp_ring->netdev,
+		  "tx_poll: q %d done. total pkts: %d\n",
+		  xdp_ring->qid, tx_pkts);
+
+	return tx_pkts;
+}
+
 static int ena_io_poll(struct napi_struct *napi, int budget)
 {
 	struct ena_napi *ena_napi = container_of(napi, struct ena_napi, napi);
 	struct ena_ring *tx_ring, *rx_ring;
-
 	u32 tx_work_done;
 	u32 rx_work_done;
 	int tx_budget;
@@ -1546,10 +2016,12 @@ static void ena_setup_io_intr(struct ena_adapter *adapter)
 {
 	struct net_device *netdev;
 	int irq_idx, i, cpu;
+	int io_queue_count;
 
 	netdev = adapter->netdev;
+	io_queue_count = adapter->num_io_queues + adapter->xdp_num_queues;
 
-	for (i = 0; i < adapter->num_io_queues; i++) {
+	for (i = 0; i < io_queue_count; i++) {
 		irq_idx = ENA_IO_IRQ_IDX(i);
 		cpu = i % num_online_cpus();
 
@@ -1677,45 +2149,64 @@ static void ena_disable_io_intr_sync(struct ena_adapter *adapter)
 		synchronize_irq(adapter->irq_tbl[i].vector);
 }
 
-static void ena_del_napi(struct ena_adapter *adapter)
+static void ena_del_napi_in_range(struct ena_adapter *adapter,
+				  int first_index,
+				  int count)
 {
 	int i;
 
-	for (i = 0; i < adapter->num_io_queues; i++)
-		netif_napi_del(&adapter->ena_napi[i].napi);
+	for (i = first_index; i < first_index + count; i++) {
+		/* Check if napi was initialized before */
+		if (!ENA_IS_XDP_INDEX(adapter, i) ||
+		    adapter->ena_napi[i].xdp_ring)
+			netif_napi_del(&adapter->ena_napi[i].napi);
+		else
+			WARN_ON(ENA_IS_XDP_INDEX(adapter, i) &&
+				adapter->ena_napi[i].xdp_ring);
+	}
 }
 
-static void ena_init_napi(struct ena_adapter *adapter)
+static void ena_init_napi_in_range(struct ena_adapter *adapter,
+				   int first_index, int count)
 {
-	struct ena_napi *napi;
+	struct ena_napi *napi = {0};
 	int i;
 
-	for (i = 0; i < adapter->num_io_queues; i++) {
+	for (i = first_index; i < first_index + count; i++) {
 		napi = &adapter->ena_napi[i];
 
 		netif_napi_add(adapter->netdev,
 			       &adapter->ena_napi[i].napi,
-			       ena_io_poll,
+			       ENA_IS_XDP_INDEX(adapter, i) ? ena_xdp_io_poll : ena_io_poll,
 			       ENA_NAPI_BUDGET);
-		napi->rx_ring = &adapter->rx_ring[i];
-		napi->tx_ring = &adapter->tx_ring[i];
+
+		if (!ENA_IS_XDP_INDEX(adapter, i)) {
+			napi->rx_ring = &adapter->rx_ring[i];
+			napi->tx_ring = &adapter->tx_ring[i];
+		} else {
+			napi->xdp_ring = &adapter->tx_ring[i];
+		}
 		napi->qid = i;
 	}
 }
 
-static void ena_napi_disable_all(struct ena_adapter *adapter)
+static void ena_napi_disable_in_range(struct ena_adapter *adapter,
+				      int first_index,
+				      int count)
 {
 	int i;
 
-	for (i = 0; i < adapter->num_io_queues; i++)
+	for (i = first_index; i < first_index + count; i++)
 		napi_disable(&adapter->ena_napi[i].napi);
 }
 
-static void ena_napi_enable_all(struct ena_adapter *adapter)
+static void ena_napi_enable_in_range(struct ena_adapter *adapter,
+				     int first_index,
+				     int count)
 {
 	int i;
 
-	for (i = 0; i < adapter->num_io_queues; i++)
+	for (i = first_index; i < first_index + count; i++)
 		napi_enable(&adapter->ena_napi[i].napi);
 }
 
@@ -1730,7 +2221,7 @@ static int ena_rss_configure(struct ena_adapter *adapter)
 		rc = ena_rss_init_default(adapter);
 		if (rc && (rc != -EOPNOTSUPP)) {
 			netif_err(adapter, ifup, adapter->netdev,
-				  "Failed to init RSS rc: %d\n", rc);
+					"Failed to init RSS rc: %d\n", rc);
 			return rc;
 		}
 	}
@@ -1768,7 +2259,9 @@ static int ena_up_complete(struct ena_adapter *adapter)
 	/* enable transmits */
 	netif_tx_start_all_queues(adapter->netdev);
 
-	ena_napi_enable_all(adapter);
+	ena_napi_enable_in_range(adapter,
+				 0,
+				 adapter->xdp_num_queues + adapter->num_io_queues);
 
 	return 0;
 }
@@ -1801,7 +2294,7 @@ static int ena_create_io_tx_queue(struct ena_adapter *adapter, int qid)
 	if (rc) {
 		netif_err(adapter, ifup, adapter->netdev,
 			  "Failed to create I/O TX queue num %d rc: %d\n",
-			  qid, rc);
+			   qid, rc);
 		return rc;
 	}
 
@@ -1820,12 +2313,13 @@ static int ena_create_io_tx_queue(struct ena_adapter *adapter, int qid)
 	return rc;
 }
 
-static int ena_create_all_io_tx_queues(struct ena_adapter *adapter)
+static int ena_create_io_tx_queues_in_range(struct ena_adapter *adapter,
+					    int first_index, int count)
 {
 	struct ena_com_dev *ena_dev = adapter->ena_dev;
 	int rc, i;
 
-	for (i = 0; i < adapter->num_io_queues; i++) {
+	for (i = first_index; i < first_index + count; i++) {
 		rc = ena_create_io_tx_queue(adapter, i);
 		if (rc)
 			goto create_err;
@@ -1834,7 +2328,7 @@ static int ena_create_all_io_tx_queues(struct ena_adapter *adapter)
 	return 0;
 
 create_err:
-	while (i--)
+	while (i-- > first_index)
 		ena_com_destroy_io_queue(ena_dev, ENA_IO_TXQ_IDX(i));
 
 	return rc;
@@ -1884,26 +2378,6 @@ static int ena_create_io_rx_queue(struct ena_adapter *adapter, int qid)
 
 	ena_com_update_numa_node(rx_ring->ena_com_io_cq, ctx.numa_node);
 
-	rc = xdp_rxq_info_reg(&rx_ring->xdp_rxq, adapter->netdev, qid);
-
-	if (rc) {
-		netif_err(adapter, ifup, adapter->netdev,
-			  "Failed to register xdp rx queue info. RX queue num %d rc: %d\n",
-			  qid, rc);
-		goto err;
-	}
-
-	rc = xdp_rxq_info_reg_mem_model(&rx_ring->xdp_rxq, MEM_TYPE_PAGE_SHARED,
-					NULL);
-
-	if (rc) {
-		netif_err(adapter, ifup, adapter->netdev,
-			  "Failed to register xdp rx queue info memory model. RX queue num %d rc: %d\n",
-			  qid, rc);
-		xdp_rxq_info_unreg(&rx_ring->xdp_rxq);
-		goto err;
-	}
-
 	return rc;
 err:
 	ena_com_destroy_io_queue(ena_dev, ena_qid);
@@ -1934,7 +2408,8 @@ create_err:
 }
 
 static void set_io_rings_size(struct ena_adapter *adapter,
-				     int new_tx_size, int new_rx_size)
+			      int new_tx_size,
+			      int new_rx_size)
 {
 	int i;
 
@@ -1968,14 +2443,24 @@ static int create_queues_with_size_backoff(struct ena_adapter *adapter)
 	 * ones due to past queue allocation failures.
 	 */
 	set_io_rings_size(adapter, adapter->requested_tx_ring_size,
-			  adapter->requested_rx_ring_size);
+			adapter->requested_rx_ring_size);
 
 	while (1) {
-		rc = ena_setup_all_tx_resources(adapter);
+		if (ena_xdp_present(adapter)) {
+			rc = ena_setup_and_create_all_xdp_queues(adapter);
+
+			if (rc)
+				goto err_setup_tx;
+		}
+		rc = ena_setup_tx_resources_in_range(adapter,
+						     0,
+						     adapter->num_io_queues);
 		if (rc)
 			goto err_setup_tx;
 
-		rc = ena_create_all_io_tx_queues(adapter);
+		rc = ena_create_io_tx_queues_in_range(adapter,
+						      0,
+						      adapter->num_io_queues);
 		if (rc)
 			goto err_create_tx_queues;
 
@@ -1999,7 +2484,7 @@ err_setup_tx:
 		if (rc != -ENOMEM) {
 			netif_err(adapter, ifup, adapter->netdev,
 				  "Queue creation failed with error code %d\n",
-				  rc);
+				   rc);
 			return rc;
 		}
 
@@ -2022,7 +2507,7 @@ err_setup_tx:
 			new_rx_ring_size = cur_rx_ring_size / 2;
 
 		if (new_tx_ring_size < ENA_MIN_RING_SIZE ||
-		    new_rx_ring_size < ENA_MIN_RING_SIZE) {
+				new_rx_ring_size < ENA_MIN_RING_SIZE) {
 			netif_err(adapter, ifup, adapter->netdev,
 				  "Queue creation failed with the smallest possible queue size of %d for both queues. Not retrying with smaller queues\n",
 				  ENA_MIN_RING_SIZE);
@@ -2041,10 +2526,11 @@ err_setup_tx:
 
 static int ena_up(struct ena_adapter *adapter)
 {
-	int rc, i;
+	int io_queue_count, rc, i;
 
 	netdev_dbg(adapter->netdev, "%s\n", __func__);
 
+	io_queue_count = adapter->num_io_queues + adapter->xdp_num_queues;
 	ena_setup_io_intr(adapter);
 
 	/* napi poll functions should be initialized before running
@@ -2052,7 +2538,7 @@ static int ena_up(struct ena_adapter *adapter)
 	 * interrupt, causing the ISR to fire immediately while the poll
 	 * function wasn't set yet, causing a null dereference
 	 */
-	ena_init_napi(adapter);
+	ena_init_napi_in_range(adapter, 0, io_queue_count);
 
 	rc = ena_request_io_irq(adapter);
 	if (rc)
@@ -2083,7 +2569,7 @@ static int ena_up(struct ena_adapter *adapter)
 	/* schedule napi in case we had pending packets
 	 * from the last time we disable napi
 	 */
-	for (i = 0; i < adapter->num_io_queues; i++)
+	for (i = 0; i < io_queue_count; i++)
 		napi_schedule(&adapter->ena_napi[i].napi);
 
 	return rc;
@@ -2096,13 +2582,15 @@ err_up:
 err_create_queues_with_backoff:
 	ena_free_io_irq(adapter);
 err_req_irq:
-	ena_del_napi(adapter);
+	ena_del_napi_in_range(adapter, 0, io_queue_count);
 
 	return rc;
 }
 
 static void ena_down(struct ena_adapter *adapter)
 {
+	int io_queue_count = adapter->num_io_queues + adapter->xdp_num_queues;
+
 	netif_info(adapter, ifdown, adapter->netdev, "%s\n", __func__);
 
 	clear_bit(ENA_FLAG_DEV_UP, &adapter->flags);
@@ -2115,7 +2603,7 @@ static void ena_down(struct ena_adapter *adapter)
 	netif_tx_disable(adapter->netdev);
 
 	/* After this point the napi handler won't enable the tx queue */
-	ena_napi_disable_all(adapter);
+	ena_napi_disable_in_range(adapter, 0, io_queue_count);
 
 	/* After destroy the queue there won't be any new interrupts */
 
@@ -2133,7 +2621,7 @@ static void ena_down(struct ena_adapter *adapter)
 
 	ena_disable_io_intr_sync(adapter);
 	ena_free_io_irq(adapter);
-	ena_del_napi(adapter);
+	ena_del_napi_in_range(adapter, 0, io_queue_count);
 
 	ena_free_all_tx_bufs(adapter);
 	ena_free_all_rx_bufs(adapter);
@@ -2223,7 +2711,10 @@ int ena_update_queue_sizes(struct ena_adapter *adapter,
 	ena_close(adapter->netdev);
 	adapter->requested_tx_ring_size = new_tx_size;
 	adapter->requested_rx_ring_size = new_rx_size;
-	ena_init_io_rings(adapter);
+	ena_init_io_rings(adapter,
+			  0,
+			  adapter->xdp_num_queues +
+			  adapter->num_io_queues);
 	return dev_was_up ? ena_up(adapter) : 0;
 }
 
@@ -2237,8 +2728,10 @@ int ena_update_queue_count(struct ena_adapter *adapter, u32 new_channel_count)
 	ena_close(adapter->netdev);
 	prev_channel_count = adapter->num_io_queues;
 	adapter->num_io_queues = new_channel_count;
-
-	if (ena_xdp_present(adapter) && ena_xdp_allowed(adapter)) {
+	if (ena_xdp_present(adapter) &&
+	    ena_xdp_allowed(adapter) == ENA_XDP_ALLOWED) {
+		adapter->xdp_first_ring = new_channel_count;
+		adapter->xdp_num_queues = new_channel_count;
 		if (prev_channel_count > new_channel_count)
 			ena_xdp_exchange_program_rx_in_range(adapter,
 							     NULL,
@@ -2255,7 +2748,10 @@ int ena_update_queue_count(struct ena_adapter *adapter, u32 new_channel_count)
 	 * table will be reinitialized by ena_up()
 	 */
 	ena_com_rss_destroy(ena_dev);
-	ena_init_io_rings(adapter);
+	ena_init_io_rings(adapter,
+			  0,
+			  adapter->xdp_num_queues +
+			  adapter->num_io_queues);
 	return dev_was_up ? ena_open(adapter->netdev) : 0;
 }
 
@@ -2439,7 +2935,7 @@ error_report_dma_error:
 	tx_info->skb = NULL;
 
 	tx_info->num_of_bufs += i;
-	ena_unmap_tx_skb(tx_ring, tx_info);
+	ena_unmap_tx_buff(tx_ring, tx_info);
 
 	return -EINVAL;
 }
@@ -2454,7 +2950,7 @@ static netdev_tx_t ena_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	struct netdev_queue *txq;
 	void *push_hdr;
 	u16 next_to_use, req_id, header_len;
-	int qid, rc, nb_hw_desc;
+	int qid, rc;
 
 	netif_dbg(adapter, tx_queued, dev, "%s skb %p\n", __func__, skb);
 	/*  Determine which tx ring we will be placed on */
@@ -2489,50 +2985,17 @@ static netdev_tx_t ena_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	/* set flags and meta data */
 	ena_tx_csum(&ena_tx_ctx, skb);
 
-	if (unlikely(ena_com_is_doorbell_needed(tx_ring->ena_com_io_sq, &ena_tx_ctx))) {
-		netif_dbg(adapter, tx_queued, dev,
-			  "llq tx max burst size of queue %d achieved, writing doorbell to send burst\n",
-			  qid);
-		ena_com_write_sq_doorbell(tx_ring->ena_com_io_sq);
-	}
-
-	/* prepare the packet's descriptors to dma engine */
-	rc = ena_com_prepare_tx(tx_ring->ena_com_io_sq, &ena_tx_ctx,
-				&nb_hw_desc);
-
-	/* ena_com_prepare_tx() can't fail due to overflow of tx queue,
-	 * since the number of free descriptors in the queue is checked
-	 * after sending the previous packet. In case there isn't enough
-	 * space in the queue for the next packet, it is stopped
-	 * until there is again enough available space in the queue.
-	 * All other failure reasons of ena_com_prepare_tx() are fatal
-	 * and therefore require a device reset.
-	 */
-	if (unlikely(rc)) {
-		netif_err(adapter, tx_queued, dev,
-			  "failed to prepare tx bufs\n");
-		u64_stats_update_begin(&tx_ring->syncp);
-		tx_ring->tx_stats.prepare_ctx_err++;
-		u64_stats_update_end(&tx_ring->syncp);
-		adapter->reset_reason = ENA_REGS_RESET_DRIVER_INVALID_STATE;
-		set_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags);
+	rc = ena_xmit_common(dev,
+			     tx_ring,
+			     tx_info,
+			     &ena_tx_ctx,
+			     next_to_use,
+			     skb->len);
+	if (rc)
 		goto error_unmap_dma;
-	}
 
 	netdev_tx_sent_queue(txq, skb->len);
 
-	u64_stats_update_begin(&tx_ring->syncp);
-	tx_ring->tx_stats.cnt++;
-	tx_ring->tx_stats.bytes += skb->len;
-	u64_stats_update_end(&tx_ring->syncp);
-
-	tx_info->tx_descs = nb_hw_desc;
-	tx_info->last_jiffies = jiffies;
-	tx_info->print_once = 0;
-
-	tx_ring->next_to_use = ENA_TX_RING_IDX_NEXT(next_to_use,
-		tx_ring->ring_size);
-
 	/* stop the queue when no more space available, the packet can have up
 	 * to sgl_size + 2. one for the meta descriptor and one for header
 	 * (if the header is larger than tx_max_header_size).
@@ -2579,7 +3042,7 @@ static netdev_tx_t ena_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	return NETDEV_TX_OK;
 
 error_unmap_dma:
-	ena_unmap_tx_skb(tx_ring, tx_info);
+	ena_unmap_tx_buff(tx_ring, tx_info);
 	tx_info->skb = NULL;
 
 error_drop_packet:
@@ -3138,7 +3601,9 @@ static void check_for_missing_completions(struct ena_adapter *adapter)
 	struct ena_ring *tx_ring;
 	struct ena_ring *rx_ring;
 	int i, budget, rc;
+	int io_queue_count;
 
+	io_queue_count = adapter->xdp_num_queues + adapter->num_io_queues;
 	/* Make sure the driver doesn't turn the device in other process */
 	smp_rmb();
 
@@ -3153,7 +3618,7 @@ static void check_for_missing_completions(struct ena_adapter *adapter)
 
 	budget = ENA_MONITORED_TX_QUEUES;
 
-	for (i = adapter->last_monitored_tx_qid; i < adapter->num_io_queues; i++) {
+	for (i = adapter->last_monitored_tx_qid; i < io_queue_count; i++) {
 		tx_ring = &adapter->tx_ring[i];
 		rx_ring = &adapter->rx_ring[i];
 
@@ -3161,7 +3626,8 @@ static void check_for_missing_completions(struct ena_adapter *adapter)
 		if (unlikely(rc))
 			return;
 
-		rc = check_for_rx_interrupt_queue(adapter, rx_ring);
+		rc =  !ENA_IS_XDP_INDEX(adapter, i) ?
+			check_for_rx_interrupt_queue(adapter, rx_ring) : 0;
 		if (unlikely(rc))
 			return;
 
@@ -3170,7 +3636,7 @@ static void check_for_missing_completions(struct ena_adapter *adapter)
 			break;
 	}
 
-	adapter->last_monitored_tx_qid = i % adapter->num_io_queues;
+	adapter->last_monitored_tx_qid = i % io_queue_count;
 }
 
 /* trigger napi schedule after 2 consecutive detections */
@@ -3747,6 +4213,9 @@ static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	adapter->num_io_queues = max_num_io_queues;
 	adapter->max_num_io_queues = max_num_io_queues;
 
+	adapter->xdp_first_ring = 0;
+	adapter->xdp_num_queues = 0;
+
 	adapter->last_monitored_tx_qid = 0;
 
 	adapter->rx_copybreak = ENA_DEFAULT_RX_COPYBREAK;
@@ -3760,7 +4229,10 @@ static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 			"Failed to query interrupt moderation feature\n");
 		goto err_netdev_destroy;
 	}
-	ena_init_io_rings(adapter);
+	ena_init_io_rings(adapter,
+			  0,
+			  adapter->xdp_num_queues +
+			  adapter->num_io_queues);
 
 	netdev->netdev_ops = &ena_netdev_ops;
 	netdev->watchdog_timeo = TX_TIMEOUT;
diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.h b/drivers/net/ethernet/amazon/ena/ena_netdev.h
index c9016014b86d..9bfb275b2514 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.h
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.h
@@ -152,6 +152,9 @@
 #define ENA_XDP_MAX_MTU (ENA_PAGE_SIZE - ETH_HLEN - ETH_FCS_LEN - \
 				VLAN_HLEN - XDP_PACKET_HEADROOM)
 
+#define ENA_IS_XDP_INDEX(adapter, index) (((index) >= (adapter)->xdp_first_ring) && \
+	((index) < (adapter)->xdp_first_ring + (adapter)->xdp_num_queues))
+
 struct ena_irq {
 	irq_handler_t handler;
 	void *data;
@@ -165,6 +168,7 @@ struct ena_napi {
 	struct napi_struct napi ____cacheline_aligned;
 	struct ena_ring *tx_ring;
 	struct ena_ring *rx_ring;
+	struct ena_ring *xdp_ring;
 	u32 qid;
 	struct dim dim;
 };
@@ -190,6 +194,17 @@ struct ena_tx_buffer {
 	/* num of buffers used by this skb */
 	u32 num_of_bufs;
 
+	/* XDP buffer structure which is used for sending packets in
+	 * the xdp queues
+	 */
+	struct xdp_frame *xdpf;
+	/* The rx page for the rx buffer that was received in rx and
+	 * re transmitted on xdp tx queues as a result of XDP_TX action.
+	 * We need to free the page once we finished cleaning the buffer in
+	 * clean_xdp_irq()
+	 */
+	struct page *xdp_rx_page;
+
 	/* Indicate if bufs[0] map the linear data of the skb. */
 	u8 map_linear_data;
 
@@ -394,6 +409,8 @@ struct ena_adapter {
 	enum ena_regs_reset_reason_types reset_reason;
 
 	struct bpf_prog *xdp_bpf_prog;
+	u32 xdp_first_ring;
+	u32 xdp_num_queues;
 };
 
 void ena_set_ethtool_ops(struct net_device *netdev);
@@ -410,6 +427,17 @@ int ena_update_queue_count(struct ena_adapter *adapter, u32 new_channel_count);
 
 int ena_get_sset_count(struct net_device *netdev, int sset);
 
+enum ena_xdp_errors_t {
+	ENA_XDP_ALLOWED = 0,
+	ENA_XDP_CURRENT_MTU_TOO_LARGE,
+	ENA_XDP_NO_ENOUGH_QUEUES,
+};
+
+static inline bool ena_xdp_queues_present(struct ena_adapter *adapter)
+{
+	return adapter->xdp_first_ring != 0;
+}
+
 static inline bool ena_xdp_present(struct ena_adapter *adapter)
 {
 	return !!adapter->xdp_bpf_prog;
@@ -420,9 +448,22 @@ static inline bool ena_xdp_present_ring(struct ena_ring *ring)
 	return !!ring->xdp_bpf_prog;
 }
 
-static inline bool ena_xdp_allowed(struct ena_adapter *adapter)
+static inline int ena_xdp_legal_queue_count(struct ena_adapter *adapter,
+					    u32 queues)
 {
-	return adapter->netdev->mtu <= ENA_XDP_MAX_MTU;
+	return 2 * queues <= adapter->max_num_io_queues;
+}
+
+static inline enum ena_xdp_errors_t ena_xdp_allowed(struct ena_adapter *adapter)
+{
+	enum ena_xdp_errors_t rc = ENA_XDP_ALLOWED;
+
+	if (adapter->netdev->mtu > ENA_XDP_MAX_MTU)
+		rc = ENA_XDP_CURRENT_MTU_TOO_LARGE;
+	else if (!ena_xdp_legal_queue_count(adapter, adapter->num_io_queues))
+		rc = ENA_XDP_NO_ENOUGH_QUEUES;
+
+	return rc;
 }
 
 #endif /* !(ENA_H) */
-- 
cgit v1.2.3-59-g8ed1b


From 913b0bfd6c859aaf1735e1c8c69731e1d5359d31 Mon Sep 17 00:00:00 2001
From: Sameeh Jubran <sameehj@amazon.com>
Date: Tue, 10 Dec 2019 15:12:14 +0200
Subject: net: ena: Add first_interrupt field to napi struct

The first_interrupt field is accessed in ena_intr_msix_io() upon
receiving an interrupt.The rx_ring and tx_ring fields of napi can
be NULL when receiving interrupt for xdp queues. This patch fixes
the issue by moving the field to the ena_napi struct.

Signed-off-by: Sameeh Jubran <sameehj@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/amazon/ena/ena_netdev.c | 7 +++++--
 drivers/net/ethernet/amazon/ena/ena_netdev.h | 1 +
 2 files changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c
index 55eeaab7956c..d161537e24e2 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.c
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c
@@ -219,6 +219,7 @@ static int ena_xdp_io_poll(struct napi_struct *napi, int budget)
 	int ret;
 
 	xdp_ring = ena_napi->xdp_ring;
+	xdp_ring->first_interrupt = ena_napi->first_interrupt;
 
 	xdp_budget = budget;
 
@@ -1869,6 +1870,9 @@ static int ena_io_poll(struct napi_struct *napi, int budget)
 	tx_ring = ena_napi->tx_ring;
 	rx_ring = ena_napi->rx_ring;
 
+	tx_ring->first_interrupt = ena_napi->first_interrupt;
+	rx_ring->first_interrupt = ena_napi->first_interrupt;
+
 	tx_budget = tx_ring->ring_size / ENA_TX_POLL_BUDGET_DIVIDER;
 
 	if (!test_bit(ENA_FLAG_DEV_UP, &tx_ring->adapter->flags) ||
@@ -1940,8 +1944,7 @@ static irqreturn_t ena_intr_msix_io(int irq, void *data)
 {
 	struct ena_napi *ena_napi = data;
 
-	ena_napi->tx_ring->first_interrupt = true;
-	ena_napi->rx_ring->first_interrupt = true;
+	ena_napi->first_interrupt = true;
 
 	napi_schedule_irqoff(&ena_napi->napi);
 
diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.h b/drivers/net/ethernet/amazon/ena/ena_netdev.h
index 9bfb275b2514..094324fd0edc 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.h
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.h
@@ -169,6 +169,7 @@ struct ena_napi {
 	struct ena_ring *tx_ring;
 	struct ena_ring *rx_ring;
 	struct ena_ring *xdp_ring;
+	bool first_interrupt;
 	u32 qid;
 	struct dim dim;
 };
-- 
cgit v1.2.3-59-g8ed1b


From 0290bd291cc0e0488e35e66bf39efcd7d9d9122b Mon Sep 17 00:00:00 2001
From: "Michael S. Tsirkin" <mst@redhat.com>
Date: Tue, 10 Dec 2019 09:23:51 -0500
Subject: netdev: pass the stuck queue to the timeout handler

This allows incrementing the correct timeout statistic without any mess.
Down the road, devices can learn to reset just the specific queue.

The patch was generated with the following script:

use strict;
use warnings;

our $^I = '.bak';

my @work = (
["arch/m68k/emu/nfeth.c", "nfeth_tx_timeout"],
["arch/um/drivers/net_kern.c", "uml_net_tx_timeout"],
["arch/um/drivers/vector_kern.c", "vector_net_tx_timeout"],
["arch/xtensa/platforms/iss/network.c", "iss_net_tx_timeout"],
["drivers/char/pcmcia/synclink_cs.c", "hdlcdev_tx_timeout"],
["drivers/infiniband/ulp/ipoib/ipoib_main.c", "ipoib_timeout"],
["drivers/infiniband/ulp/ipoib/ipoib_main.c", "ipoib_timeout"],
["drivers/message/fusion/mptlan.c", "mpt_lan_tx_timeout"],
["drivers/misc/sgi-xp/xpnet.c", "xpnet_dev_tx_timeout"],
["drivers/net/appletalk/cops.c", "cops_timeout"],
["drivers/net/arcnet/arcdevice.h", "arcnet_timeout"],
["drivers/net/arcnet/arcnet.c", "arcnet_timeout"],
["drivers/net/arcnet/com20020.c", "arcnet_timeout"],
["drivers/net/ethernet/3com/3c509.c", "el3_tx_timeout"],
["drivers/net/ethernet/3com/3c515.c", "corkscrew_timeout"],
["drivers/net/ethernet/3com/3c574_cs.c", "el3_tx_timeout"],
["drivers/net/ethernet/3com/3c589_cs.c", "el3_tx_timeout"],
["drivers/net/ethernet/3com/3c59x.c", "vortex_tx_timeout"],
["drivers/net/ethernet/3com/3c59x.c", "vortex_tx_timeout"],
["drivers/net/ethernet/3com/typhoon.c", "typhoon_tx_timeout"],
["drivers/net/ethernet/8390/8390.h", "ei_tx_timeout"],
["drivers/net/ethernet/8390/8390.h", "eip_tx_timeout"],
["drivers/net/ethernet/8390/8390.c", "ei_tx_timeout"],
["drivers/net/ethernet/8390/8390p.c", "eip_tx_timeout"],
["drivers/net/ethernet/8390/ax88796.c", "ax_ei_tx_timeout"],
["drivers/net/ethernet/8390/axnet_cs.c", "axnet_tx_timeout"],
["drivers/net/ethernet/8390/etherh.c", "__ei_tx_timeout"],
["drivers/net/ethernet/8390/hydra.c", "__ei_tx_timeout"],
["drivers/net/ethernet/8390/mac8390.c", "__ei_tx_timeout"],
["drivers/net/ethernet/8390/mcf8390.c", "__ei_tx_timeout"],
["drivers/net/ethernet/8390/lib8390.c", "__ei_tx_timeout"],
["drivers/net/ethernet/8390/ne2k-pci.c", "ei_tx_timeout"],
["drivers/net/ethernet/8390/pcnet_cs.c", "ei_tx_timeout"],
["drivers/net/ethernet/8390/smc-ultra.c", "ei_tx_timeout"],
["drivers/net/ethernet/8390/wd.c", "ei_tx_timeout"],
["drivers/net/ethernet/8390/zorro8390.c", "__ei_tx_timeout"],
["drivers/net/ethernet/adaptec/starfire.c", "tx_timeout"],
["drivers/net/ethernet/agere/et131x.c", "et131x_tx_timeout"],
["drivers/net/ethernet/allwinner/sun4i-emac.c", "emac_timeout"],
["drivers/net/ethernet/alteon/acenic.c", "ace_watchdog"],
["drivers/net/ethernet/amazon/ena/ena_netdev.c", "ena_tx_timeout"],
["drivers/net/ethernet/amd/7990.h", "lance_tx_timeout"],
["drivers/net/ethernet/amd/7990.c", "lance_tx_timeout"],
["drivers/net/ethernet/amd/a2065.c", "lance_tx_timeout"],
["drivers/net/ethernet/amd/am79c961a.c", "am79c961_timeout"],
["drivers/net/ethernet/amd/amd8111e.c", "amd8111e_tx_timeout"],
["drivers/net/ethernet/amd/ariadne.c", "ariadne_tx_timeout"],
["drivers/net/ethernet/amd/atarilance.c", "lance_tx_timeout"],
["drivers/net/ethernet/amd/au1000_eth.c", "au1000_tx_timeout"],
["drivers/net/ethernet/amd/declance.c", "lance_tx_timeout"],
["drivers/net/ethernet/amd/lance.c", "lance_tx_timeout"],
["drivers/net/ethernet/amd/mvme147.c", "lance_tx_timeout"],
["drivers/net/ethernet/amd/ni65.c", "ni65_timeout"],
["drivers/net/ethernet/amd/nmclan_cs.c", "mace_tx_timeout"],
["drivers/net/ethernet/amd/pcnet32.c", "pcnet32_tx_timeout"],
["drivers/net/ethernet/amd/sunlance.c", "lance_tx_timeout"],
["drivers/net/ethernet/amd/xgbe/xgbe-drv.c", "xgbe_tx_timeout"],
["drivers/net/ethernet/apm/xgene-v2/main.c", "xge_timeout"],
["drivers/net/ethernet/apm/xgene/xgene_enet_main.c", "xgene_enet_timeout"],
["drivers/net/ethernet/apple/macmace.c", "mace_tx_timeout"],
["drivers/net/ethernet/atheros/ag71xx.c", "ag71xx_tx_timeout"],
["drivers/net/ethernet/atheros/alx/main.c", "alx_tx_timeout"],
["drivers/net/ethernet/atheros/atl1c/atl1c_main.c", "atl1c_tx_timeout"],
["drivers/net/ethernet/atheros/atl1e/atl1e_main.c", "atl1e_tx_timeout"],
["drivers/net/ethernet/atheros/atlx/atl.c", "atlx_tx_timeout"],
["drivers/net/ethernet/atheros/atlx/atl1.c", "atlx_tx_timeout"],
["drivers/net/ethernet/atheros/atlx/atl2.c", "atl2_tx_timeout"],
["drivers/net/ethernet/broadcom/b44.c", "b44_tx_timeout"],
["drivers/net/ethernet/broadcom/bcmsysport.c", "bcm_sysport_tx_timeout"],
["drivers/net/ethernet/broadcom/bnx2.c", "bnx2_tx_timeout"],
["drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h", "bnx2x_tx_timeout"],
["drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c", "bnx2x_tx_timeout"],
["drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c", "bnx2x_tx_timeout"],
["drivers/net/ethernet/broadcom/bnxt/bnxt.c", "bnxt_tx_timeout"],
["drivers/net/ethernet/broadcom/genet/bcmgenet.c", "bcmgenet_timeout"],
["drivers/net/ethernet/broadcom/sb1250-mac.c", "sbmac_tx_timeout"],
["drivers/net/ethernet/broadcom/tg3.c", "tg3_tx_timeout"],
["drivers/net/ethernet/calxeda/xgmac.c", "xgmac_tx_timeout"],
["drivers/net/ethernet/cavium/liquidio/lio_main.c", "liquidio_tx_timeout"],
["drivers/net/ethernet/cavium/liquidio/lio_vf_main.c", "liquidio_tx_timeout"],
["drivers/net/ethernet/cavium/liquidio/lio_vf_rep.c", "lio_vf_rep_tx_timeout"],
["drivers/net/ethernet/cavium/thunder/nicvf_main.c", "nicvf_tx_timeout"],
["drivers/net/ethernet/cirrus/cs89x0.c", "net_timeout"],
["drivers/net/ethernet/cisco/enic/enic_main.c", "enic_tx_timeout"],
["drivers/net/ethernet/cisco/enic/enic_main.c", "enic_tx_timeout"],
["drivers/net/ethernet/cortina/gemini.c", "gmac_tx_timeout"],
["drivers/net/ethernet/davicom/dm9000.c", "dm9000_timeout"],
["drivers/net/ethernet/dec/tulip/de2104x.c", "de_tx_timeout"],
["drivers/net/ethernet/dec/tulip/tulip_core.c", "tulip_tx_timeout"],
["drivers/net/ethernet/dec/tulip/winbond-840.c", "tx_timeout"],
["drivers/net/ethernet/dlink/dl2k.c", "rio_tx_timeout"],
["drivers/net/ethernet/dlink/sundance.c", "tx_timeout"],
["drivers/net/ethernet/emulex/benet/be_main.c", "be_tx_timeout"],
["drivers/net/ethernet/ethoc.c", "ethoc_tx_timeout"],
["drivers/net/ethernet/faraday/ftgmac100.c", "ftgmac100_tx_timeout"],
["drivers/net/ethernet/fealnx.c", "fealnx_tx_timeout"],
["drivers/net/ethernet/freescale/dpaa/dpaa_eth.c", "dpaa_tx_timeout"],
["drivers/net/ethernet/freescale/fec_main.c", "fec_timeout"],
["drivers/net/ethernet/freescale/fec_mpc52xx.c", "mpc52xx_fec_tx_timeout"],
["drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c", "fs_timeout"],
["drivers/net/ethernet/freescale/gianfar.c", "gfar_timeout"],
["drivers/net/ethernet/freescale/ucc_geth.c", "ucc_geth_timeout"],
["drivers/net/ethernet/fujitsu/fmvj18x_cs.c", "fjn_tx_timeout"],
["drivers/net/ethernet/google/gve/gve_main.c", "gve_tx_timeout"],
["drivers/net/ethernet/hisilicon/hip04_eth.c", "hip04_timeout"],
["drivers/net/ethernet/hisilicon/hix5hd2_gmac.c", "hix5hd2_net_timeout"],
["drivers/net/ethernet/hisilicon/hns/hns_enet.c", "hns_nic_net_timeout"],
["drivers/net/ethernet/hisilicon/hns3/hns3_enet.c", "hns3_nic_net_timeout"],
["drivers/net/ethernet/huawei/hinic/hinic_main.c", "hinic_tx_timeout"],
["drivers/net/ethernet/i825xx/82596.c", "i596_tx_timeout"],
["drivers/net/ethernet/i825xx/ether1.c", "ether1_timeout"],
["drivers/net/ethernet/i825xx/lib82596.c", "i596_tx_timeout"],
["drivers/net/ethernet/i825xx/sun3_82586.c", "sun3_82586_timeout"],
["drivers/net/ethernet/ibm/ehea/ehea_main.c", "ehea_tx_watchdog"],
["drivers/net/ethernet/ibm/emac/core.c", "emac_tx_timeout"],
["drivers/net/ethernet/ibm/emac/core.c", "emac_tx_timeout"],
["drivers/net/ethernet/ibm/ibmvnic.c", "ibmvnic_tx_timeout"],
["drivers/net/ethernet/intel/e100.c", "e100_tx_timeout"],
["drivers/net/ethernet/intel/e1000/e1000_main.c", "e1000_tx_timeout"],
["drivers/net/ethernet/intel/e1000e/netdev.c", "e1000_tx_timeout"],
["drivers/net/ethernet/intel/fm10k/fm10k_netdev.c", "fm10k_tx_timeout"],
["drivers/net/ethernet/intel/i40e/i40e_main.c", "i40e_tx_timeout"],
["drivers/net/ethernet/intel/iavf/iavf_main.c", "iavf_tx_timeout"],
["drivers/net/ethernet/intel/ice/ice_main.c", "ice_tx_timeout"],
["drivers/net/ethernet/intel/ice/ice_main.c", "ice_tx_timeout"],
["drivers/net/ethernet/intel/igb/igb_main.c", "igb_tx_timeout"],
["drivers/net/ethernet/intel/igbvf/netdev.c", "igbvf_tx_timeout"],
["drivers/net/ethernet/intel/ixgb/ixgb_main.c", "ixgb_tx_timeout"],
["drivers/net/ethernet/intel/ixgbe/ixgbe_debugfs.c", "adapter->netdev->netdev_ops->ndo_tx_timeout(adapter->netdev);"],
["drivers/net/ethernet/intel/ixgbe/ixgbe_main.c", "ixgbe_tx_timeout"],
["drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c", "ixgbevf_tx_timeout"],
["drivers/net/ethernet/jme.c", "jme_tx_timeout"],
["drivers/net/ethernet/korina.c", "korina_tx_timeout"],
["drivers/net/ethernet/lantiq_etop.c", "ltq_etop_tx_timeout"],
["drivers/net/ethernet/marvell/mv643xx_eth.c", "mv643xx_eth_tx_timeout"],
["drivers/net/ethernet/marvell/pxa168_eth.c", "pxa168_eth_tx_timeout"],
["drivers/net/ethernet/marvell/skge.c", "skge_tx_timeout"],
["drivers/net/ethernet/marvell/sky2.c", "sky2_tx_timeout"],
["drivers/net/ethernet/marvell/sky2.c", "sky2_tx_timeout"],
["drivers/net/ethernet/mediatek/mtk_eth_soc.c", "mtk_tx_timeout"],
["drivers/net/ethernet/mellanox/mlx4/en_netdev.c", "mlx4_en_tx_timeout"],
["drivers/net/ethernet/mellanox/mlx4/en_netdev.c", "mlx4_en_tx_timeout"],
["drivers/net/ethernet/mellanox/mlx5/core/en_main.c", "mlx5e_tx_timeout"],
["drivers/net/ethernet/micrel/ks8842.c", "ks8842_tx_timeout"],
["drivers/net/ethernet/micrel/ksz884x.c", "netdev_tx_timeout"],
["drivers/net/ethernet/microchip/enc28j60.c", "enc28j60_tx_timeout"],
["drivers/net/ethernet/microchip/encx24j600.c", "encx24j600_tx_timeout"],
["drivers/net/ethernet/natsemi/sonic.h", "sonic_tx_timeout"],
["drivers/net/ethernet/natsemi/sonic.c", "sonic_tx_timeout"],
["drivers/net/ethernet/natsemi/jazzsonic.c", "sonic_tx_timeout"],
["drivers/net/ethernet/natsemi/macsonic.c", "sonic_tx_timeout"],
["drivers/net/ethernet/natsemi/natsemi.c", "ns_tx_timeout"],
["drivers/net/ethernet/natsemi/ns83820.c", "ns83820_tx_timeout"],
["drivers/net/ethernet/natsemi/xtsonic.c", "sonic_tx_timeout"],
["drivers/net/ethernet/neterion/s2io.h", "s2io_tx_watchdog"],
["drivers/net/ethernet/neterion/s2io.c", "s2io_tx_watchdog"],
["drivers/net/ethernet/neterion/vxge/vxge-main.c", "vxge_tx_watchdog"],
["drivers/net/ethernet/netronome/nfp/nfp_net_common.c", "nfp_net_tx_timeout"],
["drivers/net/ethernet/nvidia/forcedeth.c", "nv_tx_timeout"],
["drivers/net/ethernet/nvidia/forcedeth.c", "nv_tx_timeout"],
["drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c", "pch_gbe_tx_timeout"],
["drivers/net/ethernet/packetengines/hamachi.c", "hamachi_tx_timeout"],
["drivers/net/ethernet/packetengines/yellowfin.c", "yellowfin_tx_timeout"],
["drivers/net/ethernet/pensando/ionic/ionic_lif.c", "ionic_tx_timeout"],
["drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c", "netxen_tx_timeout"],
["drivers/net/ethernet/qlogic/qla3xxx.c", "ql3xxx_tx_timeout"],
["drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c", "qlcnic_tx_timeout"],
["drivers/net/ethernet/qualcomm/emac/emac.c", "emac_tx_timeout"],
["drivers/net/ethernet/qualcomm/qca_spi.c", "qcaspi_netdev_tx_timeout"],
["drivers/net/ethernet/qualcomm/qca_uart.c", "qcauart_netdev_tx_timeout"],
["drivers/net/ethernet/rdc/r6040.c", "r6040_tx_timeout"],
["drivers/net/ethernet/realtek/8139cp.c", "cp_tx_timeout"],
["drivers/net/ethernet/realtek/8139too.c", "rtl8139_tx_timeout"],
["drivers/net/ethernet/realtek/atp.c", "tx_timeout"],
["drivers/net/ethernet/realtek/r8169_main.c", "rtl8169_tx_timeout"],
["drivers/net/ethernet/renesas/ravb_main.c", "ravb_tx_timeout"],
["drivers/net/ethernet/renesas/sh_eth.c", "sh_eth_tx_timeout"],
["drivers/net/ethernet/renesas/sh_eth.c", "sh_eth_tx_timeout"],
["drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c", "sxgbe_tx_timeout"],
["drivers/net/ethernet/seeq/ether3.c", "ether3_timeout"],
["drivers/net/ethernet/seeq/sgiseeq.c", "timeout"],
["drivers/net/ethernet/sfc/efx.c", "efx_watchdog"],
["drivers/net/ethernet/sfc/falcon/efx.c", "ef4_watchdog"],
["drivers/net/ethernet/sgi/ioc3-eth.c", "ioc3_timeout"],
["drivers/net/ethernet/sgi/meth.c", "meth_tx_timeout"],
["drivers/net/ethernet/silan/sc92031.c", "sc92031_tx_timeout"],
["drivers/net/ethernet/sis/sis190.c", "sis190_tx_timeout"],
["drivers/net/ethernet/sis/sis900.c", "sis900_tx_timeout"],
["drivers/net/ethernet/smsc/epic100.c", "epic_tx_timeout"],
["drivers/net/ethernet/smsc/smc911x.c", "smc911x_timeout"],
["drivers/net/ethernet/smsc/smc9194.c", "smc_timeout"],
["drivers/net/ethernet/smsc/smc91c92_cs.c", "smc_tx_timeout"],
["drivers/net/ethernet/smsc/smc91x.c", "smc_timeout"],
["drivers/net/ethernet/stmicro/stmmac/stmmac_main.c", "stmmac_tx_timeout"],
["drivers/net/ethernet/sun/cassini.c", "cas_tx_timeout"],
["drivers/net/ethernet/sun/ldmvsw.c", "sunvnet_tx_timeout_common"],
["drivers/net/ethernet/sun/niu.c", "niu_tx_timeout"],
["drivers/net/ethernet/sun/sunbmac.c", "bigmac_tx_timeout"],
["drivers/net/ethernet/sun/sungem.c", "gem_tx_timeout"],
["drivers/net/ethernet/sun/sunhme.c", "happy_meal_tx_timeout"],
["drivers/net/ethernet/sun/sunqe.c", "qe_tx_timeout"],
["drivers/net/ethernet/sun/sunvnet.c", "sunvnet_tx_timeout_common"],
["drivers/net/ethernet/sun/sunvnet_common.c", "sunvnet_tx_timeout_common"],
["drivers/net/ethernet/sun/sunvnet_common.h", "sunvnet_tx_timeout_common"],
["drivers/net/ethernet/synopsys/dwc-xlgmac-net.c", "xlgmac_tx_timeout"],
["drivers/net/ethernet/ti/cpmac.c", "cpmac_tx_timeout"],
["drivers/net/ethernet/ti/cpsw.c", "cpsw_ndo_tx_timeout"],
["drivers/net/ethernet/ti/cpsw_priv.c", "cpsw_ndo_tx_timeout"],
["drivers/net/ethernet/ti/cpsw_priv.h", "cpsw_ndo_tx_timeout"],
["drivers/net/ethernet/ti/davinci_emac.c", "emac_dev_tx_timeout"],
["drivers/net/ethernet/ti/netcp_core.c", "netcp_ndo_tx_timeout"],
["drivers/net/ethernet/ti/tlan.c", "tlan_tx_timeout"],
["drivers/net/ethernet/toshiba/ps3_gelic_net.h", "gelic_net_tx_timeout"],
["drivers/net/ethernet/toshiba/ps3_gelic_net.c", "gelic_net_tx_timeout"],
["drivers/net/ethernet/toshiba/ps3_gelic_wireless.c", "gelic_net_tx_timeout"],
["drivers/net/ethernet/toshiba/spider_net.c", "spider_net_tx_timeout"],
["drivers/net/ethernet/toshiba/tc35815.c", "tc35815_tx_timeout"],
["drivers/net/ethernet/via/via-rhine.c", "rhine_tx_timeout"],
["drivers/net/ethernet/wiznet/w5100.c", "w5100_tx_timeout"],
["drivers/net/ethernet/wiznet/w5300.c", "w5300_tx_timeout"],
["drivers/net/ethernet/xilinx/xilinx_emaclite.c", "xemaclite_tx_timeout"],
["drivers/net/ethernet/xircom/xirc2ps_cs.c", "xirc_tx_timeout"],
["drivers/net/fjes/fjes_main.c", "fjes_tx_retry"],
["drivers/net/slip/slip.c", "sl_tx_timeout"],
["include/linux/usb/usbnet.h", "usbnet_tx_timeout"],
["drivers/net/usb/aqc111.c", "usbnet_tx_timeout"],
["drivers/net/usb/asix_devices.c", "usbnet_tx_timeout"],
["drivers/net/usb/asix_devices.c", "usbnet_tx_timeout"],
["drivers/net/usb/asix_devices.c", "usbnet_tx_timeout"],
["drivers/net/usb/ax88172a.c", "usbnet_tx_timeout"],
["drivers/net/usb/ax88179_178a.c", "usbnet_tx_timeout"],
["drivers/net/usb/catc.c", "catc_tx_timeout"],
["drivers/net/usb/cdc_mbim.c", "usbnet_tx_timeout"],
["drivers/net/usb/cdc_ncm.c", "usbnet_tx_timeout"],
["drivers/net/usb/dm9601.c", "usbnet_tx_timeout"],
["drivers/net/usb/hso.c", "hso_net_tx_timeout"],
["drivers/net/usb/int51x1.c", "usbnet_tx_timeout"],
["drivers/net/usb/ipheth.c", "ipheth_tx_timeout"],
["drivers/net/usb/kaweth.c", "kaweth_tx_timeout"],
["drivers/net/usb/lan78xx.c", "lan78xx_tx_timeout"],
["drivers/net/usb/mcs7830.c", "usbnet_tx_timeout"],
["drivers/net/usb/pegasus.c", "pegasus_tx_timeout"],
["drivers/net/usb/qmi_wwan.c", "usbnet_tx_timeout"],
["drivers/net/usb/r8152.c", "rtl8152_tx_timeout"],
["drivers/net/usb/rndis_host.c", "usbnet_tx_timeout"],
["drivers/net/usb/rtl8150.c", "rtl8150_tx_timeout"],
["drivers/net/usb/sierra_net.c", "usbnet_tx_timeout"],
["drivers/net/usb/smsc75xx.c", "usbnet_tx_timeout"],
["drivers/net/usb/smsc95xx.c", "usbnet_tx_timeout"],
["drivers/net/usb/sr9700.c", "usbnet_tx_timeout"],
["drivers/net/usb/sr9800.c", "usbnet_tx_timeout"],
["drivers/net/usb/usbnet.c", "usbnet_tx_timeout"],
["drivers/net/vmxnet3/vmxnet3_drv.c", "vmxnet3_tx_timeout"],
["drivers/net/wan/cosa.c", "cosa_net_timeout"],
["drivers/net/wan/farsync.c", "fst_tx_timeout"],
["drivers/net/wan/fsl_ucc_hdlc.c", "uhdlc_tx_timeout"],
["drivers/net/wan/lmc/lmc_main.c", "lmc_driver_timeout"],
["drivers/net/wan/x25_asy.c", "x25_asy_timeout"],
["drivers/net/wimax/i2400m/netdev.c", "i2400m_tx_timeout"],
["drivers/net/wireless/intel/ipw2x00/ipw2100.c", "ipw2100_tx_timeout"],
["drivers/net/wireless/intersil/hostap/hostap_main.c", "prism2_tx_timeout"],
["drivers/net/wireless/intersil/hostap/hostap_main.c", "prism2_tx_timeout"],
["drivers/net/wireless/intersil/hostap/hostap_main.c", "prism2_tx_timeout"],
["drivers/net/wireless/intersil/orinoco/main.c", "orinoco_tx_timeout"],
["drivers/net/wireless/intersil/orinoco/orinoco_usb.c", "orinoco_tx_timeout"],
["drivers/net/wireless/intersil/orinoco/orinoco.h", "orinoco_tx_timeout"],
["drivers/net/wireless/intersil/prism54/islpci_dev.c", "islpci_eth_tx_timeout"],
["drivers/net/wireless/intersil/prism54/islpci_eth.c", "islpci_eth_tx_timeout"],
["drivers/net/wireless/intersil/prism54/islpci_eth.h", "islpci_eth_tx_timeout"],
["drivers/net/wireless/marvell/mwifiex/main.c", "mwifiex_tx_timeout"],
["drivers/net/wireless/quantenna/qtnfmac/core.c", "qtnf_netdev_tx_timeout"],
["drivers/net/wireless/quantenna/qtnfmac/core.h", "qtnf_netdev_tx_timeout"],
["drivers/net/wireless/rndis_wlan.c", "usbnet_tx_timeout"],
["drivers/net/wireless/wl3501_cs.c", "wl3501_tx_timeout"],
["drivers/net/wireless/zydas/zd1201.c", "zd1201_tx_timeout"],
["drivers/s390/net/qeth_core.h", "qeth_tx_timeout"],
["drivers/s390/net/qeth_core_main.c", "qeth_tx_timeout"],
["drivers/s390/net/qeth_l2_main.c", "qeth_tx_timeout"],
["drivers/s390/net/qeth_l2_main.c", "qeth_tx_timeout"],
["drivers/s390/net/qeth_l3_main.c", "qeth_tx_timeout"],
["drivers/s390/net/qeth_l3_main.c", "qeth_tx_timeout"],
["drivers/staging/ks7010/ks_wlan_net.c", "ks_wlan_tx_timeout"],
["drivers/staging/qlge/qlge_main.c", "qlge_tx_timeout"],
["drivers/staging/rtl8192e/rtl8192e/rtl_core.c", "_rtl92e_tx_timeout"],
["drivers/staging/rtl8192u/r8192U_core.c", "tx_timeout"],
["drivers/staging/unisys/visornic/visornic_main.c", "visornic_xmit_timeout"],
["drivers/staging/wlan-ng/p80211netdev.c", "p80211knetdev_tx_timeout"],
["drivers/tty/n_gsm.c", "gsm_mux_net_tx_timeout"],
["drivers/tty/synclink.c", "hdlcdev_tx_timeout"],
["drivers/tty/synclink_gt.c", "hdlcdev_tx_timeout"],
["drivers/tty/synclinkmp.c", "hdlcdev_tx_timeout"],
["net/atm/lec.c", "lec_tx_timeout"],
["net/bluetooth/bnep/netdev.c", "bnep_net_timeout"]
);

for my $p (@work) {
	my @pair = @$p;
	my $file = $pair[0];
	my $func = $pair[1];
	print STDERR $file , ": ", $func,"\n";
	our @ARGV = ($file);
	while (<ARGV>) {
		if (m/($func\s*\(struct\s+net_device\s+\*[A-Za-z_]?[A-Za-z-0-9_]*)(\))/) {
			print STDERR "found $1+$2 in $file\n";
		}
		if (s/($func\s*\(struct\s+net_device\s+\*[A-Za-z_]?[A-Za-z-0-9_]*)(\))/$1, unsigned int txqueue$2/) {
			print STDERR "$func found in $file\n";
		}
		print;
	}
}

where the list of files and functions is simply from:

git grep ndo_tx_timeout, with manual addition of headers
in the rare cases where the function is from a header,
then manually changing the few places which actually
call ndo_tx_timeout.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Heiner Kallweit <hkallweit1@gmail.com>
Acked-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Acked-by: Shannon Nelson <snelson@pensando.io>
Reviewed-by: Martin Habets <mhabets@solarflare.com>

changes from v9:
	fixup a forward declaration
changes from v9:
	more leftovers from v3 change
changes from v8:
        fix up a missing direct call to timeout
        rebased on net-next
changes from v7:
	fixup leftovers from v3 change
changes from v6:
	fix typo in rtl driver
changes from v5:
	add missing files (allow any net device argument name)
changes from v4:
	add a missing driver header
changes from v3:
        change queue # to unsigned
Changes from v2:
        added headers
Changes from v1:
        Fix errors found by kbuild:
        generalize the pattern a bit, to pick up
        a couple of instances missed by the previous
        version.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 arch/m68k/emu/nfeth.c                                 | 2 +-
 arch/um/drivers/net_kern.c                            | 2 +-
 arch/um/drivers/vector_kern.c                         | 2 +-
 arch/xtensa/platforms/iss/network.c                   | 2 +-
 drivers/char/pcmcia/synclink_cs.c                     | 2 +-
 drivers/infiniband/ulp/ipoib/ipoib_main.c             | 2 +-
 drivers/message/fusion/mptlan.c                       | 2 +-
 drivers/misc/sgi-xp/xpnet.c                           | 2 +-
 drivers/net/appletalk/cops.c                          | 4 ++--
 drivers/net/arcnet/arcdevice.h                        | 2 +-
 drivers/net/arcnet/arcnet.c                           | 2 +-
 drivers/net/ethernet/3com/3c509.c                     | 4 ++--
 drivers/net/ethernet/3com/3c515.c                     | 4 ++--
 drivers/net/ethernet/3com/3c574_cs.c                  | 4 ++--
 drivers/net/ethernet/3com/3c589_cs.c                  | 4 ++--
 drivers/net/ethernet/3com/3c59x.c                     | 4 ++--
 drivers/net/ethernet/3com/typhoon.c                   | 2 +-
 drivers/net/ethernet/8390/8390.c                      | 4 ++--
 drivers/net/ethernet/8390/8390.h                      | 4 ++--
 drivers/net/ethernet/8390/8390p.c                     | 4 ++--
 drivers/net/ethernet/8390/axnet_cs.c                  | 4 ++--
 drivers/net/ethernet/8390/lib8390.c                   | 2 +-
 drivers/net/ethernet/adaptec/starfire.c               | 4 ++--
 drivers/net/ethernet/agere/et131x.c                   | 2 +-
 drivers/net/ethernet/allwinner/sun4i-emac.c           | 2 +-
 drivers/net/ethernet/alteon/acenic.c                  | 4 ++--
 drivers/net/ethernet/amazon/ena/ena_netdev.c          | 2 +-
 drivers/net/ethernet/amd/7990.c                       | 2 +-
 drivers/net/ethernet/amd/7990.h                       | 2 +-
 drivers/net/ethernet/amd/a2065.c                      | 2 +-
 drivers/net/ethernet/amd/am79c961a.c                  | 2 +-
 drivers/net/ethernet/amd/amd8111e.c                   | 2 +-
 drivers/net/ethernet/amd/ariadne.c                    | 2 +-
 drivers/net/ethernet/amd/atarilance.c                 | 4 ++--
 drivers/net/ethernet/amd/au1000_eth.c                 | 2 +-
 drivers/net/ethernet/amd/declance.c                   | 2 +-
 drivers/net/ethernet/amd/lance.c                      | 4 ++--
 drivers/net/ethernet/amd/ni65.c                       | 4 ++--
 drivers/net/ethernet/amd/nmclan_cs.c                  | 4 ++--
 drivers/net/ethernet/amd/pcnet32.c                    | 4 ++--
 drivers/net/ethernet/amd/sunlance.c                   | 2 +-
 drivers/net/ethernet/amd/xgbe/xgbe-drv.c              | 2 +-
 drivers/net/ethernet/apm/xgene-v2/main.c              | 2 +-
 drivers/net/ethernet/apm/xgene/xgene_enet_main.c      | 2 +-
 drivers/net/ethernet/apple/macmace.c                  | 4 ++--
 drivers/net/ethernet/atheros/ag71xx.c                 | 2 +-
 drivers/net/ethernet/atheros/alx/main.c               | 2 +-
 drivers/net/ethernet/atheros/atl1c/atl1c_main.c       | 2 +-
 drivers/net/ethernet/atheros/atl1e/atl1e_main.c       | 2 +-
 drivers/net/ethernet/atheros/atlx/atl2.c              | 2 +-
 drivers/net/ethernet/atheros/atlx/atlx.c              | 2 +-
 drivers/net/ethernet/broadcom/b44.c                   | 2 +-
 drivers/net/ethernet/broadcom/bcmsysport.c            | 2 +-
 drivers/net/ethernet/broadcom/bnx2.c                  | 2 +-
 drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c       | 2 +-
 drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h       | 2 +-
 drivers/net/ethernet/broadcom/bnxt/bnxt.c             | 2 +-
 drivers/net/ethernet/broadcom/genet/bcmgenet.c        | 2 +-
 drivers/net/ethernet/broadcom/sb1250-mac.c            | 4 ++--
 drivers/net/ethernet/broadcom/tg3.c                   | 2 +-
 drivers/net/ethernet/calxeda/xgmac.c                  | 2 +-
 drivers/net/ethernet/cavium/liquidio/lio_main.c       | 2 +-
 drivers/net/ethernet/cavium/liquidio/lio_vf_main.c    | 2 +-
 drivers/net/ethernet/cavium/liquidio/lio_vf_rep.c     | 4 ++--
 drivers/net/ethernet/cavium/thunder/nicvf_main.c      | 2 +-
 drivers/net/ethernet/cirrus/cs89x0.c                  | 2 +-
 drivers/net/ethernet/cisco/enic/enic_main.c           | 2 +-
 drivers/net/ethernet/cortina/gemini.c                 | 2 +-
 drivers/net/ethernet/davicom/dm9000.c                 | 2 +-
 drivers/net/ethernet/dec/tulip/de2104x.c              | 2 +-
 drivers/net/ethernet/dec/tulip/tulip_core.c           | 4 ++--
 drivers/net/ethernet/dec/tulip/winbond-840.c          | 4 ++--
 drivers/net/ethernet/dlink/dl2k.c                     | 4 ++--
 drivers/net/ethernet/dlink/sundance.c                 | 4 ++--
 drivers/net/ethernet/emulex/benet/be_main.c           | 2 +-
 drivers/net/ethernet/ethoc.c                          | 2 +-
 drivers/net/ethernet/faraday/ftgmac100.c              | 2 +-
 drivers/net/ethernet/fealnx.c                         | 4 ++--
 drivers/net/ethernet/freescale/dpaa/dpaa_eth.c        | 2 +-
 drivers/net/ethernet/freescale/fec_main.c             | 2 +-
 drivers/net/ethernet/freescale/fec_mpc52xx.c          | 2 +-
 drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c | 2 +-
 drivers/net/ethernet/freescale/gianfar.c              | 2 +-
 drivers/net/ethernet/freescale/ucc_geth.c             | 2 +-
 drivers/net/ethernet/fujitsu/fmvj18x_cs.c             | 4 ++--
 drivers/net/ethernet/google/gve/gve_main.c            | 2 +-
 drivers/net/ethernet/hisilicon/hip04_eth.c            | 2 +-
 drivers/net/ethernet/hisilicon/hix5hd2_gmac.c         | 2 +-
 drivers/net/ethernet/hisilicon/hns/hns_enet.c         | 2 +-
 drivers/net/ethernet/hisilicon/hns3/hns3_enet.c       | 2 +-
 drivers/net/ethernet/huawei/hinic/hinic_main.c        | 2 +-
 drivers/net/ethernet/i825xx/82596.c                   | 4 ++--
 drivers/net/ethernet/i825xx/ether1.c                  | 4 ++--
 drivers/net/ethernet/i825xx/lib82596.c                | 4 ++--
 drivers/net/ethernet/i825xx/sun3_82586.c              | 4 ++--
 drivers/net/ethernet/ibm/ehea/ehea_main.c             | 2 +-
 drivers/net/ethernet/ibm/emac/core.c                  | 2 +-
 drivers/net/ethernet/ibm/ibmvnic.c                    | 2 +-
 drivers/net/ethernet/intel/e100.c                     | 2 +-
 drivers/net/ethernet/intel/e1000/e1000_main.c         | 4 ++--
 drivers/net/ethernet/intel/e1000e/netdev.c            | 2 +-
 drivers/net/ethernet/intel/fm10k/fm10k_netdev.c       | 2 +-
 drivers/net/ethernet/intel/i40e/i40e_main.c           | 2 +-
 drivers/net/ethernet/intel/iavf/iavf_main.c           | 2 +-
 drivers/net/ethernet/intel/ice/ice_main.c             | 2 +-
 drivers/net/ethernet/intel/igb/igb_main.c             | 4 ++--
 drivers/net/ethernet/intel/igbvf/netdev.c             | 2 +-
 drivers/net/ethernet/intel/ixgb/ixgb_main.c           | 4 ++--
 drivers/net/ethernet/intel/ixgbe/ixgbe_debugfs.c      | 4 +++-
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c         | 2 +-
 drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c     | 2 +-
 drivers/net/ethernet/jme.c                            | 2 +-
 drivers/net/ethernet/korina.c                         | 2 +-
 drivers/net/ethernet/lantiq_etop.c                    | 2 +-
 drivers/net/ethernet/marvell/mv643xx_eth.c            | 2 +-
 drivers/net/ethernet/marvell/pxa168_eth.c             | 2 +-
 drivers/net/ethernet/marvell/skge.c                   | 2 +-
 drivers/net/ethernet/marvell/sky2.c                   | 2 +-
 drivers/net/ethernet/mediatek/mtk_eth_soc.c           | 2 +-
 drivers/net/ethernet/mellanox/mlx4/en_netdev.c        | 2 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c     | 2 +-
 drivers/net/ethernet/micrel/ks8842.c                  | 2 +-
 drivers/net/ethernet/micrel/ksz884x.c                 | 2 +-
 drivers/net/ethernet/microchip/enc28j60.c             | 2 +-
 drivers/net/ethernet/microchip/encx24j600.c           | 2 +-
 drivers/net/ethernet/natsemi/natsemi.c                | 4 ++--
 drivers/net/ethernet/natsemi/ns83820.c                | 4 ++--
 drivers/net/ethernet/natsemi/sonic.c                  | 2 +-
 drivers/net/ethernet/natsemi/sonic.h                  | 2 +-
 drivers/net/ethernet/neterion/s2io.c                  | 2 +-
 drivers/net/ethernet/neterion/s2io.h                  | 2 +-
 drivers/net/ethernet/neterion/vxge/vxge-main.c        | 2 +-
 drivers/net/ethernet/netronome/nfp/nfp_net_common.c   | 2 +-
 drivers/net/ethernet/nvidia/forcedeth.c               | 2 +-
 drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c  | 2 +-
 drivers/net/ethernet/packetengines/hamachi.c          | 4 ++--
 drivers/net/ethernet/packetengines/yellowfin.c        | 4 ++--
 drivers/net/ethernet/pensando/ionic/ionic_lif.c       | 2 +-
 drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c  | 4 ++--
 drivers/net/ethernet/qlogic/qla3xxx.c                 | 2 +-
 drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c      | 4 ++--
 drivers/net/ethernet/qualcomm/emac/emac.c             | 2 +-
 drivers/net/ethernet/qualcomm/qca_spi.c               | 2 +-
 drivers/net/ethernet/qualcomm/qca_uart.c              | 2 +-
 drivers/net/ethernet/rdc/r6040.c                      | 2 +-
 drivers/net/ethernet/realtek/8139cp.c                 | 2 +-
 drivers/net/ethernet/realtek/8139too.c                | 4 ++--
 drivers/net/ethernet/realtek/atp.c                    | 4 ++--
 drivers/net/ethernet/realtek/r8169_main.c             | 2 +-
 drivers/net/ethernet/renesas/ravb_main.c              | 2 +-
 drivers/net/ethernet/renesas/sh_eth.c                 | 2 +-
 drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c       | 2 +-
 drivers/net/ethernet/seeq/ether3.c                    | 4 ++--
 drivers/net/ethernet/seeq/sgiseeq.c                   | 2 +-
 drivers/net/ethernet/sfc/efx.c                        | 2 +-
 drivers/net/ethernet/sfc/falcon/efx.c                 | 2 +-
 drivers/net/ethernet/sgi/ioc3-eth.c                   | 4 ++--
 drivers/net/ethernet/sgi/meth.c                       | 4 ++--
 drivers/net/ethernet/silan/sc92031.c                  | 2 +-
 drivers/net/ethernet/sis/sis190.c                     | 2 +-
 drivers/net/ethernet/sis/sis900.c                     | 4 ++--
 drivers/net/ethernet/smsc/epic100.c                   | 4 ++--
 drivers/net/ethernet/smsc/smc911x.c                   | 2 +-
 drivers/net/ethernet/smsc/smc9194.c                   | 4 ++--
 drivers/net/ethernet/smsc/smc91c92_cs.c               | 4 ++--
 drivers/net/ethernet/smsc/smc91x.c                    | 2 +-
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c     | 2 +-
 drivers/net/ethernet/sun/cassini.c                    | 2 +-
 drivers/net/ethernet/sun/niu.c                        | 2 +-
 drivers/net/ethernet/sun/sunbmac.c                    | 2 +-
 drivers/net/ethernet/sun/sungem.c                     | 2 +-
 drivers/net/ethernet/sun/sunhme.c                     | 2 +-
 drivers/net/ethernet/sun/sunqe.c                      | 2 +-
 drivers/net/ethernet/sun/sunvnet_common.c             | 2 +-
 drivers/net/ethernet/sun/sunvnet_common.h             | 2 +-
 drivers/net/ethernet/synopsys/dwc-xlgmac-net.c        | 2 +-
 drivers/net/ethernet/ti/cpmac.c                       | 2 +-
 drivers/net/ethernet/ti/cpsw_priv.c                   | 2 +-
 drivers/net/ethernet/ti/cpsw_priv.h                   | 2 +-
 drivers/net/ethernet/ti/davinci_emac.c                | 2 +-
 drivers/net/ethernet/ti/netcp_core.c                  | 2 +-
 drivers/net/ethernet/ti/tlan.c                        | 6 +++---
 drivers/net/ethernet/toshiba/ps3_gelic_net.c          | 2 +-
 drivers/net/ethernet/toshiba/ps3_gelic_net.h          | 2 +-
 drivers/net/ethernet/toshiba/spider_net.c             | 2 +-
 drivers/net/ethernet/toshiba/tc35815.c                | 4 ++--
 drivers/net/ethernet/via/via-rhine.c                  | 4 ++--
 drivers/net/ethernet/wiznet/w5100.c                   | 2 +-
 drivers/net/ethernet/wiznet/w5300.c                   | 2 +-
 drivers/net/ethernet/xilinx/xilinx_emaclite.c         | 2 +-
 drivers/net/ethernet/xircom/xirc2ps_cs.c              | 4 ++--
 drivers/net/fjes/fjes_main.c                          | 4 ++--
 drivers/net/slip/slip.c                               | 2 +-
 drivers/net/usb/catc.c                                | 2 +-
 drivers/net/usb/hso.c                                 | 2 +-
 drivers/net/usb/ipheth.c                              | 2 +-
 drivers/net/usb/kaweth.c                              | 2 +-
 drivers/net/usb/lan78xx.c                             | 2 +-
 drivers/net/usb/pegasus.c                             | 2 +-
 drivers/net/usb/r8152.c                               | 2 +-
 drivers/net/usb/rtl8150.c                             | 2 +-
 drivers/net/usb/usbnet.c                              | 2 +-
 drivers/net/vmxnet3/vmxnet3_drv.c                     | 2 +-
 drivers/net/wan/cosa.c                                | 4 ++--
 drivers/net/wan/farsync.c                             | 2 +-
 drivers/net/wan/fsl_ucc_hdlc.c                        | 2 +-
 drivers/net/wan/lmc/lmc_main.c                        | 4 ++--
 drivers/net/wan/x25_asy.c                             | 2 +-
 drivers/net/wimax/i2400m/netdev.c                     | 2 +-
 drivers/net/wireless/intel/ipw2x00/ipw2100.c          | 2 +-
 drivers/net/wireless/intersil/hostap/hostap_main.c    | 2 +-
 drivers/net/wireless/intersil/orinoco/main.c          | 2 +-
 drivers/net/wireless/intersil/orinoco/orinoco.h       | 2 +-
 drivers/net/wireless/intersil/prism54/islpci_eth.c    | 2 +-
 drivers/net/wireless/intersil/prism54/islpci_eth.h    | 2 +-
 drivers/net/wireless/marvell/mwifiex/main.c           | 2 +-
 drivers/net/wireless/quantenna/qtnfmac/core.c         | 2 +-
 drivers/net/wireless/wl3501_cs.c                      | 2 +-
 drivers/net/wireless/zydas/zd1201.c                   | 2 +-
 drivers/s390/net/qeth_core.h                          | 2 +-
 drivers/s390/net/qeth_core_main.c                     | 2 +-
 drivers/staging/ks7010/ks_wlan_net.c                  | 4 ++--
 drivers/staging/qlge/qlge_main.c                      | 2 +-
 drivers/staging/rtl8192e/rtl8192e/rtl_core.c          | 2 +-
 drivers/staging/rtl8192u/r8192U_core.c                | 2 +-
 drivers/staging/unisys/visornic/visornic_main.c       | 2 +-
 drivers/staging/wlan-ng/p80211netdev.c                | 4 ++--
 drivers/tty/n_gsm.c                                   | 2 +-
 drivers/tty/synclink.c                                | 2 +-
 drivers/tty/synclink_gt.c                             | 2 +-
 drivers/tty/synclinkmp.c                              | 2 +-
 include/linux/netdevice.h                             | 5 +++--
 include/linux/usb/usbnet.h                            | 2 +-
 net/atm/lec.c                                         | 2 +-
 net/bluetooth/bnep/netdev.c                           | 2 +-
 net/sched/sch_generic.c                               | 2 +-
 236 files changed, 298 insertions(+), 295 deletions(-)

diff --git a/arch/m68k/emu/nfeth.c b/arch/m68k/emu/nfeth.c
index a4ebd2445eda..d2875e32abfc 100644
--- a/arch/m68k/emu/nfeth.c
+++ b/arch/m68k/emu/nfeth.c
@@ -167,7 +167,7 @@ static int nfeth_xmit(struct sk_buff *skb, struct net_device *dev)
 	return 0;
 }
 
-static void nfeth_tx_timeout(struct net_device *dev)
+static void nfeth_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	dev->stats.tx_errors++;
 	netif_wake_queue(dev);
diff --git a/arch/um/drivers/net_kern.c b/arch/um/drivers/net_kern.c
index 327b728f7244..35ebeebfc1a8 100644
--- a/arch/um/drivers/net_kern.c
+++ b/arch/um/drivers/net_kern.c
@@ -247,7 +247,7 @@ static void uml_net_set_multicast_list(struct net_device *dev)
 	return;
 }
 
-static void uml_net_tx_timeout(struct net_device *dev)
+static void uml_net_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	netif_trans_update(dev);
 	netif_wake_queue(dev);
diff --git a/arch/um/drivers/vector_kern.c b/arch/um/drivers/vector_kern.c
index 92617e16829e..0ff86391f77d 100644
--- a/arch/um/drivers/vector_kern.c
+++ b/arch/um/drivers/vector_kern.c
@@ -1332,7 +1332,7 @@ static void vector_net_set_multicast_list(struct net_device *dev)
 	return;
 }
 
-static void vector_net_tx_timeout(struct net_device *dev)
+static void vector_net_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct vector_private *vp = netdev_priv(dev);
 
diff --git a/arch/xtensa/platforms/iss/network.c b/arch/xtensa/platforms/iss/network.c
index fa9f3893b002..4986226a5ab2 100644
--- a/arch/xtensa/platforms/iss/network.c
+++ b/arch/xtensa/platforms/iss/network.c
@@ -455,7 +455,7 @@ static void iss_net_set_multicast_list(struct net_device *dev)
 {
 }
 
-static void iss_net_tx_timeout(struct net_device *dev)
+static void iss_net_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 }
 
diff --git a/drivers/char/pcmcia/synclink_cs.c b/drivers/char/pcmcia/synclink_cs.c
index 82f9a6a814ae..e342daa73d1b 100644
--- a/drivers/char/pcmcia/synclink_cs.c
+++ b/drivers/char/pcmcia/synclink_cs.c
@@ -4169,7 +4169,7 @@ static int hdlcdev_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
  *
  * dev  pointer to network device structure
  */
-static void hdlcdev_tx_timeout(struct net_device *dev)
+static void hdlcdev_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	MGSLPC_INFO *info = dev_to_port(dev);
 	unsigned long flags;
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c
index e5f438ab716c..4a0d3a9e72e1 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_main.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c
@@ -1182,7 +1182,7 @@ unref:
 	return NETDEV_TX_OK;
 }
 
-static void ipoib_timeout(struct net_device *dev)
+static void ipoib_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct ipoib_dev_priv *priv = ipoib_priv(dev);
 
diff --git a/drivers/message/fusion/mptlan.c b/drivers/message/fusion/mptlan.c
index ebc00d47abf5..7d3784aa20e5 100644
--- a/drivers/message/fusion/mptlan.c
+++ b/drivers/message/fusion/mptlan.c
@@ -552,7 +552,7 @@ mpt_lan_close(struct net_device *dev)
 /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/
 /* Tx timeout handler. */
 static void
-mpt_lan_tx_timeout(struct net_device *dev)
+mpt_lan_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct mpt_lan_priv *priv = netdev_priv(dev);
 	MPT_ADAPTER *mpt_dev = priv->mpt_dev;
diff --git a/drivers/misc/sgi-xp/xpnet.c b/drivers/misc/sgi-xp/xpnet.c
index f7d610a22347..ada94e6a3c91 100644
--- a/drivers/misc/sgi-xp/xpnet.c
+++ b/drivers/misc/sgi-xp/xpnet.c
@@ -496,7 +496,7 @@ xpnet_dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
  * Deal with transmit timeouts coming from the network layer.
  */
 static void
-xpnet_dev_tx_timeout(struct net_device *dev)
+xpnet_dev_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	dev->stats.tx_errors++;
 }
diff --git a/drivers/net/appletalk/cops.c b/drivers/net/appletalk/cops.c
index b3c63d2f16aa..18428e104445 100644
--- a/drivers/net/appletalk/cops.c
+++ b/drivers/net/appletalk/cops.c
@@ -189,7 +189,7 @@ static int  cops_nodeid (struct net_device *dev, int nodeid);
 
 static irqreturn_t cops_interrupt (int irq, void *dev_id);
 static void cops_poll(struct timer_list *t);
-static void cops_timeout(struct net_device *dev);
+static void cops_timeout(struct net_device *dev, unsigned int txqueue);
 static void cops_rx (struct net_device *dev);
 static netdev_tx_t  cops_send_packet (struct sk_buff *skb,
 					    struct net_device *dev);
@@ -844,7 +844,7 @@ static void cops_rx(struct net_device *dev)
         netif_rx(skb);
 }
 
-static void cops_timeout(struct net_device *dev)
+static void cops_timeout(struct net_device *dev, unsigned int txqueue)
 {
         struct cops_local *lp = netdev_priv(dev);
         int ioaddr = dev->base_addr;
diff --git a/drivers/net/arcnet/arcdevice.h b/drivers/net/arcnet/arcdevice.h
index b0f5bc07aef5..22a49c6d7ae6 100644
--- a/drivers/net/arcnet/arcdevice.h
+++ b/drivers/net/arcnet/arcdevice.h
@@ -356,7 +356,7 @@ int arcnet_open(struct net_device *dev);
 int arcnet_close(struct net_device *dev);
 netdev_tx_t arcnet_send_packet(struct sk_buff *skb,
 			       struct net_device *dev);
-void arcnet_timeout(struct net_device *dev);
+void arcnet_timeout(struct net_device *dev, unsigned int txqueue);
 
 /* I/O equivalents */
 
diff --git a/drivers/net/arcnet/arcnet.c b/drivers/net/arcnet/arcnet.c
index 553776cc1d29..e04efc0a5c97 100644
--- a/drivers/net/arcnet/arcnet.c
+++ b/drivers/net/arcnet/arcnet.c
@@ -763,7 +763,7 @@ static int go_tx(struct net_device *dev)
 }
 
 /* Called by the kernel when transmit times out */
-void arcnet_timeout(struct net_device *dev)
+void arcnet_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	unsigned long flags;
 	struct arcnet_local *lp = netdev_priv(dev);
diff --git a/drivers/net/ethernet/3com/3c509.c b/drivers/net/ethernet/3com/3c509.c
index 3da97996bdf3..8cafd06ff0c4 100644
--- a/drivers/net/ethernet/3com/3c509.c
+++ b/drivers/net/ethernet/3com/3c509.c
@@ -196,7 +196,7 @@ static struct net_device_stats *el3_get_stats(struct net_device *dev);
 static int el3_rx(struct net_device *dev);
 static int el3_close(struct net_device *dev);
 static void set_multicast_list(struct net_device *dev);
-static void el3_tx_timeout (struct net_device *dev);
+static void el3_tx_timeout (struct net_device *dev, unsigned int txqueue);
 static void el3_down(struct net_device *dev);
 static void el3_up(struct net_device *dev);
 static const struct ethtool_ops ethtool_ops;
@@ -689,7 +689,7 @@ el3_open(struct net_device *dev)
 }
 
 static void
-el3_tx_timeout (struct net_device *dev)
+el3_tx_timeout (struct net_device *dev, unsigned int txqueue)
 {
 	int ioaddr = dev->base_addr;
 
diff --git a/drivers/net/ethernet/3com/3c515.c b/drivers/net/ethernet/3com/3c515.c
index b15752267c8d..1e233e2f0a5a 100644
--- a/drivers/net/ethernet/3com/3c515.c
+++ b/drivers/net/ethernet/3com/3c515.c
@@ -371,7 +371,7 @@ static void corkscrew_timer(struct timer_list *t);
 static netdev_tx_t corkscrew_start_xmit(struct sk_buff *skb,
 					struct net_device *dev);
 static int corkscrew_rx(struct net_device *dev);
-static void corkscrew_timeout(struct net_device *dev);
+static void corkscrew_timeout(struct net_device *dev, unsigned int txqueue);
 static int boomerang_rx(struct net_device *dev);
 static irqreturn_t corkscrew_interrupt(int irq, void *dev_id);
 static int corkscrew_close(struct net_device *dev);
@@ -961,7 +961,7 @@ static void corkscrew_timer(struct timer_list *t)
 #endif				/* AUTOMEDIA */
 }
 
-static void corkscrew_timeout(struct net_device *dev)
+static void corkscrew_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	int i;
 	struct corkscrew_private *vp = netdev_priv(dev);
diff --git a/drivers/net/ethernet/3com/3c574_cs.c b/drivers/net/ethernet/3com/3c574_cs.c
index 3044a6f35f04..ef1c3151fbb2 100644
--- a/drivers/net/ethernet/3com/3c574_cs.c
+++ b/drivers/net/ethernet/3com/3c574_cs.c
@@ -234,7 +234,7 @@ static void update_stats(struct net_device *dev);
 static struct net_device_stats *el3_get_stats(struct net_device *dev);
 static int el3_rx(struct net_device *dev, int worklimit);
 static int el3_close(struct net_device *dev);
-static void el3_tx_timeout(struct net_device *dev);
+static void el3_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static int el3_ioctl(struct net_device *dev, struct ifreq *rq, int cmd);
 static void set_rx_mode(struct net_device *dev);
 static void set_multicast_list(struct net_device *dev);
@@ -690,7 +690,7 @@ static int el3_open(struct net_device *dev)
 	return 0;
 }
 
-static void el3_tx_timeout(struct net_device *dev)
+static void el3_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	unsigned int ioaddr = dev->base_addr;
 	
diff --git a/drivers/net/ethernet/3com/3c589_cs.c b/drivers/net/ethernet/3com/3c589_cs.c
index 2b2695311bda..d47cde6c5f08 100644
--- a/drivers/net/ethernet/3com/3c589_cs.c
+++ b/drivers/net/ethernet/3com/3c589_cs.c
@@ -173,7 +173,7 @@ static void update_stats(struct net_device *dev);
 static struct net_device_stats *el3_get_stats(struct net_device *dev);
 static int el3_rx(struct net_device *dev);
 static int el3_close(struct net_device *dev);
-static void el3_tx_timeout(struct net_device *dev);
+static void el3_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static void set_rx_mode(struct net_device *dev);
 static void set_multicast_list(struct net_device *dev);
 static const struct ethtool_ops netdev_ethtool_ops;
@@ -526,7 +526,7 @@ static int el3_open(struct net_device *dev)
 	return 0;
 }
 
-static void el3_tx_timeout(struct net_device *dev)
+static void el3_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	unsigned int ioaddr = dev->base_addr;
 
diff --git a/drivers/net/ethernet/3com/3c59x.c b/drivers/net/ethernet/3com/3c59x.c
index 8785c2ff3825..fc046797c0ea 100644
--- a/drivers/net/ethernet/3com/3c59x.c
+++ b/drivers/net/ethernet/3com/3c59x.c
@@ -776,7 +776,7 @@ static void set_rx_mode(struct net_device *dev);
 #ifdef CONFIG_PCI
 static int vortex_ioctl(struct net_device *dev, struct ifreq *rq, int cmd);
 #endif
-static void vortex_tx_timeout(struct net_device *dev);
+static void vortex_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static void acpi_set_WOL(struct net_device *dev);
 static const struct ethtool_ops vortex_ethtool_ops;
 static void set_8021q_mode(struct net_device *dev, int enable);
@@ -1877,7 +1877,7 @@ leave_media_alone:
 		iowrite16(FakeIntr, ioaddr + EL3_CMD);
 }
 
-static void vortex_tx_timeout(struct net_device *dev)
+static void vortex_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct vortex_private *vp = netdev_priv(dev);
 	void __iomem *ioaddr = vp->ioaddr;
diff --git a/drivers/net/ethernet/3com/typhoon.c b/drivers/net/ethernet/3com/typhoon.c
index be823c186517..14fce6658106 100644
--- a/drivers/net/ethernet/3com/typhoon.c
+++ b/drivers/net/ethernet/3com/typhoon.c
@@ -2013,7 +2013,7 @@ typhoon_stop_runtime(struct typhoon *tp, int wait_type)
 }
 
 static void
-typhoon_tx_timeout(struct net_device *dev)
+typhoon_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct typhoon *tp = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/8390/8390.c b/drivers/net/ethernet/8390/8390.c
index 78f3e532c600..0e0aa4016858 100644
--- a/drivers/net/ethernet/8390/8390.c
+++ b/drivers/net/ethernet/8390/8390.c
@@ -36,9 +36,9 @@ void ei_set_multicast_list(struct net_device *dev)
 }
 EXPORT_SYMBOL(ei_set_multicast_list);
 
-void ei_tx_timeout(struct net_device *dev)
+void ei_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
-	__ei_tx_timeout(dev);
+	__ei_tx_timeout(dev, txqueue);
 }
 EXPORT_SYMBOL(ei_tx_timeout);
 
diff --git a/drivers/net/ethernet/8390/8390.h b/drivers/net/ethernet/8390/8390.h
index 3e2f2c2e7b58..529c728f334a 100644
--- a/drivers/net/ethernet/8390/8390.h
+++ b/drivers/net/ethernet/8390/8390.h
@@ -32,7 +32,7 @@ void NS8390_init(struct net_device *dev, int startp);
 int ei_open(struct net_device *dev);
 int ei_close(struct net_device *dev);
 irqreturn_t ei_interrupt(int irq, void *dev_id);
-void ei_tx_timeout(struct net_device *dev);
+void ei_tx_timeout(struct net_device *dev, unsigned int txqueue);
 netdev_tx_t ei_start_xmit(struct sk_buff *skb, struct net_device *dev);
 void ei_set_multicast_list(struct net_device *dev);
 struct net_device_stats *ei_get_stats(struct net_device *dev);
@@ -50,7 +50,7 @@ void NS8390p_init(struct net_device *dev, int startp);
 int eip_open(struct net_device *dev);
 int eip_close(struct net_device *dev);
 irqreturn_t eip_interrupt(int irq, void *dev_id);
-void eip_tx_timeout(struct net_device *dev);
+void eip_tx_timeout(struct net_device *dev, unsigned int txqueue);
 netdev_tx_t eip_start_xmit(struct sk_buff *skb, struct net_device *dev);
 void eip_set_multicast_list(struct net_device *dev);
 struct net_device_stats *eip_get_stats(struct net_device *dev);
diff --git a/drivers/net/ethernet/8390/8390p.c b/drivers/net/ethernet/8390/8390p.c
index 6cf36992a2c6..6834742057b3 100644
--- a/drivers/net/ethernet/8390/8390p.c
+++ b/drivers/net/ethernet/8390/8390p.c
@@ -41,9 +41,9 @@ void eip_set_multicast_list(struct net_device *dev)
 }
 EXPORT_SYMBOL(eip_set_multicast_list);
 
-void eip_tx_timeout(struct net_device *dev)
+void eip_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
-	__ei_tx_timeout(dev);
+	__ei_tx_timeout(dev, txqueue);
 }
 EXPORT_SYMBOL(eip_tx_timeout);
 
diff --git a/drivers/net/ethernet/8390/axnet_cs.c b/drivers/net/ethernet/8390/axnet_cs.c
index 0b6bbf63f7ca..aeae7966a082 100644
--- a/drivers/net/ethernet/8390/axnet_cs.c
+++ b/drivers/net/ethernet/8390/axnet_cs.c
@@ -83,7 +83,7 @@ static netdev_tx_t axnet_start_xmit(struct sk_buff *skb,
 					  struct net_device *dev);
 static struct net_device_stats *get_stats(struct net_device *dev);
 static void set_multicast_list(struct net_device *dev);
-static void axnet_tx_timeout(struct net_device *dev);
+static void axnet_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static irqreturn_t ei_irq_wrapper(int irq, void *dev_id);
 static void ei_watchdog(struct timer_list *t);
 static void axnet_reset_8390(struct net_device *dev);
@@ -903,7 +903,7 @@ static int ax_close(struct net_device *dev)
  * completed (or failed) - i.e. never posted a Tx related interrupt.
  */
 
-static void axnet_tx_timeout(struct net_device *dev)
+static void axnet_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	long e8390_base = dev->base_addr;
 	struct ei_device *ei_local = netdev_priv(dev);
diff --git a/drivers/net/ethernet/8390/lib8390.c b/drivers/net/ethernet/8390/lib8390.c
index c9c55c9eab9f..babc92e2692e 100644
--- a/drivers/net/ethernet/8390/lib8390.c
+++ b/drivers/net/ethernet/8390/lib8390.c
@@ -251,7 +251,7 @@ static int __ei_close(struct net_device *dev)
  * completed (or failed) - i.e. never posted a Tx related interrupt.
  */
 
-static void __ei_tx_timeout(struct net_device *dev)
+static void __ei_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	unsigned long e8390_base = dev->base_addr;
 	struct ei_device *ei_local = netdev_priv(dev);
diff --git a/drivers/net/ethernet/adaptec/starfire.c b/drivers/net/ethernet/adaptec/starfire.c
index 816540e6beac..165d18405b0c 100644
--- a/drivers/net/ethernet/adaptec/starfire.c
+++ b/drivers/net/ethernet/adaptec/starfire.c
@@ -576,7 +576,7 @@ static int	mdio_read(struct net_device *dev, int phy_id, int location);
 static void	mdio_write(struct net_device *dev, int phy_id, int location, int value);
 static int	netdev_open(struct net_device *dev);
 static void	check_duplex(struct net_device *dev);
-static void	tx_timeout(struct net_device *dev);
+static void	tx_timeout(struct net_device *dev, unsigned int txqueue);
 static void	init_ring(struct net_device *dev);
 static netdev_tx_t start_tx(struct sk_buff *skb, struct net_device *dev);
 static irqreturn_t intr_handler(int irq, void *dev_instance);
@@ -1105,7 +1105,7 @@ static void check_duplex(struct net_device *dev)
 }
 
 
-static void tx_timeout(struct net_device *dev)
+static void tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct netdev_private *np = netdev_priv(dev);
 	void __iomem *ioaddr = np->base;
diff --git a/drivers/net/ethernet/agere/et131x.c b/drivers/net/ethernet/agere/et131x.c
index 174344c450af..3c51d8c502ed 100644
--- a/drivers/net/ethernet/agere/et131x.c
+++ b/drivers/net/ethernet/agere/et131x.c
@@ -3811,7 +3811,7 @@ drop_err:
  * specified by the 'tx_timeo" element in the net_device structure (see
  * et131x_alloc_device() to see how this value is set).
  */
-static void et131x_tx_timeout(struct net_device *netdev)
+static void et131x_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct et131x_adapter *adapter = netdev_priv(netdev);
 	struct tx_ring *tx_ring = &adapter->tx_ring;
diff --git a/drivers/net/ethernet/allwinner/sun4i-emac.c b/drivers/net/ethernet/allwinner/sun4i-emac.c
index 0537df06a9b5..5ea806423e4c 100644
--- a/drivers/net/ethernet/allwinner/sun4i-emac.c
+++ b/drivers/net/ethernet/allwinner/sun4i-emac.c
@@ -407,7 +407,7 @@ static void emac_init_device(struct net_device *dev)
 }
 
 /* Our watchdog timed out. Called by the networking layer */
-static void emac_timeout(struct net_device *dev)
+static void emac_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct emac_board_info *db = netdev_priv(dev);
 	unsigned long flags;
diff --git a/drivers/net/ethernet/alteon/acenic.c b/drivers/net/ethernet/alteon/acenic.c
index 46b4207d3266..f366faf88eee 100644
--- a/drivers/net/ethernet/alteon/acenic.c
+++ b/drivers/net/ethernet/alteon/acenic.c
@@ -437,7 +437,7 @@ static const struct ethtool_ops ace_ethtool_ops = {
 	.set_link_ksettings = ace_set_link_ksettings,
 };
 
-static void ace_watchdog(struct net_device *dev);
+static void ace_watchdog(struct net_device *dev, unsigned int txqueue);
 
 static const struct net_device_ops ace_netdev_ops = {
 	.ndo_open		= ace_open,
@@ -1542,7 +1542,7 @@ static void ace_set_rxtx_parms(struct net_device *dev, int jumbo)
 }
 
 
-static void ace_watchdog(struct net_device *data)
+static void ace_watchdog(struct net_device *data, unsigned int txqueue)
 {
 	struct net_device *dev = data;
 	struct ace_private *ap = netdev_priv(dev);
diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c
index d161537e24e2..26954fde4766 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.c
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c
@@ -108,7 +108,7 @@ static void ena_unmap_tx_buff(struct ena_ring *tx_ring,
 static int ena_create_io_tx_queues_in_range(struct ena_adapter *adapter,
 					    int first_index, int count);
 
-static void ena_tx_timeout(struct net_device *dev)
+static void ena_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct ena_adapter *adapter = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/amd/7990.c b/drivers/net/ethernet/amd/7990.c
index ab30761003da..cf3562e82ca9 100644
--- a/drivers/net/ethernet/amd/7990.c
+++ b/drivers/net/ethernet/amd/7990.c
@@ -527,7 +527,7 @@ int lance_close(struct net_device *dev)
 }
 EXPORT_SYMBOL_GPL(lance_close);
 
-void lance_tx_timeout(struct net_device *dev)
+void lance_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	printk("lance_tx_timeout\n");
 	lance_reset(dev);
diff --git a/drivers/net/ethernet/amd/7990.h b/drivers/net/ethernet/amd/7990.h
index 741cdc392c6b..8266b3c1fefc 100644
--- a/drivers/net/ethernet/amd/7990.h
+++ b/drivers/net/ethernet/amd/7990.h
@@ -243,7 +243,7 @@ int lance_open(struct net_device *dev);
 int lance_close(struct net_device *dev);
 int lance_start_xmit(struct sk_buff *skb, struct net_device *dev);
 void lance_set_multicast(struct net_device *dev);
-void lance_tx_timeout(struct net_device *dev);
+void lance_tx_timeout(struct net_device *dev, unsigned int txqueue);
 #ifdef CONFIG_NET_POLL_CONTROLLER
 void lance_poll(struct net_device *dev);
 #endif
diff --git a/drivers/net/ethernet/amd/a2065.c b/drivers/net/ethernet/amd/a2065.c
index 212fe72a190b..a3faf4feb204 100644
--- a/drivers/net/ethernet/amd/a2065.c
+++ b/drivers/net/ethernet/amd/a2065.c
@@ -522,7 +522,7 @@ static inline int lance_reset(struct net_device *dev)
 	return status;
 }
 
-static void lance_tx_timeout(struct net_device *dev)
+static void lance_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct lance_private *lp = netdev_priv(dev);
 	volatile struct lance_regs *ll = lp->ll;
diff --git a/drivers/net/ethernet/amd/am79c961a.c b/drivers/net/ethernet/amd/am79c961a.c
index 0842da492a64..1c53408f5d47 100644
--- a/drivers/net/ethernet/amd/am79c961a.c
+++ b/drivers/net/ethernet/amd/am79c961a.c
@@ -422,7 +422,7 @@ static void am79c961_setmulticastlist (struct net_device *dev)
 	spin_unlock_irqrestore(&priv->chip_lock, flags);
 }
 
-static void am79c961_timeout(struct net_device *dev)
+static void am79c961_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	printk(KERN_WARNING "%s: transmit timed out, network cable problem?\n",
 		dev->name);
diff --git a/drivers/net/ethernet/amd/amd8111e.c b/drivers/net/ethernet/amd/amd8111e.c
index 573e88fc8ede..0f3b743425e8 100644
--- a/drivers/net/ethernet/amd/amd8111e.c
+++ b/drivers/net/ethernet/amd/amd8111e.c
@@ -1569,7 +1569,7 @@ static int amd8111e_enable_link_change(struct amd8111e_priv *lp)
  * failed or the interface is locked up. This function will reinitialize
  * the hardware.
  */
-static void amd8111e_tx_timeout(struct net_device *dev)
+static void amd8111e_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct amd8111e_priv *lp = netdev_priv(dev);
 	int err;
diff --git a/drivers/net/ethernet/amd/ariadne.c b/drivers/net/ethernet/amd/ariadne.c
index 4b6a5cb85dd2..5e0f645f5bde 100644
--- a/drivers/net/ethernet/amd/ariadne.c
+++ b/drivers/net/ethernet/amd/ariadne.c
@@ -530,7 +530,7 @@ static inline void ariadne_reset(struct net_device *dev)
 	netif_start_queue(dev);
 }
 
-static void ariadne_tx_timeout(struct net_device *dev)
+static void ariadne_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	volatile struct Am79C960 *lance = (struct Am79C960 *)dev->base_addr;
 
diff --git a/drivers/net/ethernet/amd/atarilance.c b/drivers/net/ethernet/amd/atarilance.c
index d3d44e07afbc..4e36122609a3 100644
--- a/drivers/net/ethernet/amd/atarilance.c
+++ b/drivers/net/ethernet/amd/atarilance.c
@@ -346,7 +346,7 @@ static int lance_rx( struct net_device *dev );
 static int lance_close( struct net_device *dev );
 static void set_multicast_list( struct net_device *dev );
 static int lance_set_mac_address( struct net_device *dev, void *addr );
-static void lance_tx_timeout (struct net_device *dev);
+static void lance_tx_timeout (struct net_device *dev, unsigned int txqueue);
 
 /************************* End of Prototypes **************************/
 
@@ -727,7 +727,7 @@ static void lance_init_ring( struct net_device *dev )
 /* XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX */
 
 
-static void lance_tx_timeout (struct net_device *dev)
+static void lance_tx_timeout (struct net_device *dev, unsigned int txqueue)
 {
 	struct lance_private *lp = netdev_priv(dev);
 	struct lance_ioreg	 *IO = lp->iobase;
diff --git a/drivers/net/ethernet/amd/au1000_eth.c b/drivers/net/ethernet/amd/au1000_eth.c
index 1793950f0582..d832c9f4d306 100644
--- a/drivers/net/ethernet/amd/au1000_eth.c
+++ b/drivers/net/ethernet/amd/au1000_eth.c
@@ -1014,7 +1014,7 @@ static netdev_tx_t au1000_tx(struct sk_buff *skb, struct net_device *dev)
  * The Tx ring has been full longer than the watchdog timeout
  * value. The transmitter must be hung?
  */
-static void au1000_tx_timeout(struct net_device *dev)
+static void au1000_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	netdev_err(dev, "au1000_tx_timeout: dev=%p\n", dev);
 	au1000_reset_mac(dev);
diff --git a/drivers/net/ethernet/amd/declance.c b/drivers/net/ethernet/amd/declance.c
index dac4a2fcad6a..6592a2db9efb 100644
--- a/drivers/net/ethernet/amd/declance.c
+++ b/drivers/net/ethernet/amd/declance.c
@@ -884,7 +884,7 @@ static inline int lance_reset(struct net_device *dev)
 	return status;
 }
 
-static void lance_tx_timeout(struct net_device *dev)
+static void lance_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct lance_private *lp = netdev_priv(dev);
 	volatile struct lance_regs *ll = lp->ll;
diff --git a/drivers/net/ethernet/amd/lance.c b/drivers/net/ethernet/amd/lance.c
index f90b454b1642..aff44241988c 100644
--- a/drivers/net/ethernet/amd/lance.c
+++ b/drivers/net/ethernet/amd/lance.c
@@ -306,7 +306,7 @@ static irqreturn_t lance_interrupt(int irq, void *dev_id);
 static int lance_close(struct net_device *dev);
 static struct net_device_stats *lance_get_stats(struct net_device *dev);
 static void set_multicast_list(struct net_device *dev);
-static void lance_tx_timeout (struct net_device *dev);
+static void lance_tx_timeout (struct net_device *dev, unsigned int txqueue);
 
 
 
@@ -913,7 +913,7 @@ lance_restart(struct net_device *dev, unsigned int csr0_bits, int must_reinit)
 }
 
 
-static void lance_tx_timeout (struct net_device *dev)
+static void lance_tx_timeout (struct net_device *dev, unsigned int txqueue)
 {
 	struct lance_private *lp = (struct lance_private *) dev->ml_priv;
 	int ioaddr = dev->base_addr;
diff --git a/drivers/net/ethernet/amd/ni65.c b/drivers/net/ethernet/amd/ni65.c
index c6c2a54c1121..c38edf6f03a3 100644
--- a/drivers/net/ethernet/amd/ni65.c
+++ b/drivers/net/ethernet/amd/ni65.c
@@ -254,7 +254,7 @@ static int  ni65_lance_reinit(struct net_device *dev);
 static void ni65_init_lance(struct priv *p,unsigned char*,int,int);
 static netdev_tx_t ni65_send_packet(struct sk_buff *skb,
 				    struct net_device *dev);
-static void  ni65_timeout(struct net_device *dev);
+static void  ni65_timeout(struct net_device *dev, unsigned int txqueue);
 static int  ni65_close(struct net_device *dev);
 static int  ni65_alloc_buffer(struct net_device *dev);
 static void ni65_free_buffer(struct priv *p);
@@ -1133,7 +1133,7 @@ static void ni65_recv_intr(struct net_device *dev,int csr0)
  * kick xmitter ..
  */
 
-static void ni65_timeout(struct net_device *dev)
+static void ni65_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	int i;
 	struct priv *p = dev->ml_priv;
diff --git a/drivers/net/ethernet/amd/nmclan_cs.c b/drivers/net/ethernet/amd/nmclan_cs.c
index 9c152d85840d..023aecf6ab30 100644
--- a/drivers/net/ethernet/amd/nmclan_cs.c
+++ b/drivers/net/ethernet/amd/nmclan_cs.c
@@ -407,7 +407,7 @@ static int mace_open(struct net_device *dev);
 static int mace_close(struct net_device *dev);
 static netdev_tx_t mace_start_xmit(struct sk_buff *skb,
 					 struct net_device *dev);
-static void mace_tx_timeout(struct net_device *dev);
+static void mace_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static irqreturn_t mace_interrupt(int irq, void *dev_id);
 static struct net_device_stats *mace_get_stats(struct net_device *dev);
 static int mace_rx(struct net_device *dev, unsigned char RxCnt);
@@ -837,7 +837,7 @@ mace_start_xmit
 	failed, put skb back into a list."
 ---------------------------------------------------------------------------- */
 
-static void mace_tx_timeout(struct net_device *dev)
+static void mace_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
   mace_private *lp = netdev_priv(dev);
   struct pcmcia_device *link = lp->p_dev;
diff --git a/drivers/net/ethernet/amd/pcnet32.c b/drivers/net/ethernet/amd/pcnet32.c
index f5ad12c10934..dc7d88227e76 100644
--- a/drivers/net/ethernet/amd/pcnet32.c
+++ b/drivers/net/ethernet/amd/pcnet32.c
@@ -314,7 +314,7 @@ static int pcnet32_open(struct net_device *);
 static int pcnet32_init_ring(struct net_device *);
 static netdev_tx_t pcnet32_start_xmit(struct sk_buff *,
 				      struct net_device *);
-static void pcnet32_tx_timeout(struct net_device *dev);
+static void pcnet32_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static irqreturn_t pcnet32_interrupt(int, void *);
 static int pcnet32_close(struct net_device *);
 static struct net_device_stats *pcnet32_get_stats(struct net_device *);
@@ -2455,7 +2455,7 @@ static void pcnet32_restart(struct net_device *dev, unsigned int csr0_bits)
 	lp->a->write_csr(ioaddr, CSR0, csr0_bits);
 }
 
-static void pcnet32_tx_timeout(struct net_device *dev)
+static void pcnet32_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct pcnet32_private *lp = netdev_priv(dev);
 	unsigned long ioaddr = dev->base_addr, flags;
diff --git a/drivers/net/ethernet/amd/sunlance.c b/drivers/net/ethernet/amd/sunlance.c
index ebcbf8ca4829..b00e00881253 100644
--- a/drivers/net/ethernet/amd/sunlance.c
+++ b/drivers/net/ethernet/amd/sunlance.c
@@ -1097,7 +1097,7 @@ static void lance_piozero(void __iomem *dest, int len)
 		sbus_writeb(0, piobuf);
 }
 
-static void lance_tx_timeout(struct net_device *dev)
+static void lance_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct lance_private *lp = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
index 98f8f2033154..b71f9b04a51e 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
@@ -2152,7 +2152,7 @@ static int xgbe_change_mtu(struct net_device *netdev, int mtu)
 	return 0;
 }
 
-static void xgbe_tx_timeout(struct net_device *netdev)
+static void xgbe_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct xgbe_prv_data *pdata = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/apm/xgene-v2/main.c b/drivers/net/ethernet/apm/xgene-v2/main.c
index 02b4f3af02b5..c48f60996761 100644
--- a/drivers/net/ethernet/apm/xgene-v2/main.c
+++ b/drivers/net/ethernet/apm/xgene-v2/main.c
@@ -575,7 +575,7 @@ static void xge_free_pending_skb(struct net_device *ndev)
 	}
 }
 
-static void xge_timeout(struct net_device *ndev)
+static void xge_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	struct xge_pdata *pdata = netdev_priv(ndev);
 
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c
index d8612131c55e..e284b6753725 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c
@@ -859,7 +859,7 @@ static int xgene_enet_napi(struct napi_struct *napi, const int budget)
 	return processed;
 }
 
-static void xgene_enet_timeout(struct net_device *ndev)
+static void xgene_enet_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	struct xgene_enet_pdata *pdata = netdev_priv(ndev);
 	struct netdev_queue *txq;
diff --git a/drivers/net/ethernet/apple/macmace.c b/drivers/net/ethernet/apple/macmace.c
index 8d03578d5e8c..95d3061c61be 100644
--- a/drivers/net/ethernet/apple/macmace.c
+++ b/drivers/net/ethernet/apple/macmace.c
@@ -91,7 +91,7 @@ static int mace_set_address(struct net_device *dev, void *addr);
 static void mace_reset(struct net_device *dev);
 static irqreturn_t mace_interrupt(int irq, void *dev_id);
 static irqreturn_t mace_dma_intr(int irq, void *dev_id);
-static void mace_tx_timeout(struct net_device *dev);
+static void mace_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static void __mace_set_address(struct net_device *dev, void *addr);
 
 /*
@@ -600,7 +600,7 @@ static irqreturn_t mace_interrupt(int irq, void *dev_id)
 	return IRQ_HANDLED;
 }
 
-static void mace_tx_timeout(struct net_device *dev)
+static void mace_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct mace_data *mp = netdev_priv(dev);
 	volatile struct mace *mb = mp->mace;
diff --git a/drivers/net/ethernet/atheros/ag71xx.c b/drivers/net/ethernet/atheros/ag71xx.c
index 8f5021091eee..ad8b0e3fcd2c 100644
--- a/drivers/net/ethernet/atheros/ag71xx.c
+++ b/drivers/net/ethernet/atheros/ag71xx.c
@@ -1409,7 +1409,7 @@ static void ag71xx_oom_timer_handler(struct timer_list *t)
 	napi_schedule(&ag->napi);
 }
 
-static void ag71xx_tx_timeout(struct net_device *ndev)
+static void ag71xx_tx_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	struct ag71xx *ag = netdev_priv(ndev);
 
diff --git a/drivers/net/ethernet/atheros/alx/main.c b/drivers/net/ethernet/atheros/alx/main.c
index d4bbcdfd691a..1dcbc486eca9 100644
--- a/drivers/net/ethernet/atheros/alx/main.c
+++ b/drivers/net/ethernet/atheros/alx/main.c
@@ -1553,7 +1553,7 @@ static netdev_tx_t alx_start_xmit(struct sk_buff *skb,
 	return alx_start_xmit_ring(skb, alx_tx_queue_mapping(alx, skb));
 }
 
-static void alx_tx_timeout(struct net_device *dev)
+static void alx_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct alx_priv *alx = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/atheros/atl1c/atl1c_main.c b/drivers/net/ethernet/atheros/atl1c/atl1c_main.c
index 2b239ecea05f..4c0b1f8551dd 100644
--- a/drivers/net/ethernet/atheros/atl1c/atl1c_main.c
+++ b/drivers/net/ethernet/atheros/atl1c/atl1c_main.c
@@ -350,7 +350,7 @@ static void atl1c_del_timer(struct atl1c_adapter *adapter)
  * atl1c_tx_timeout - Respond to a Tx Hang
  * @netdev: network interface device structure
  */
-static void atl1c_tx_timeout(struct net_device *netdev)
+static void atl1c_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct atl1c_adapter *adapter = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/atheros/atl1e/atl1e_main.c b/drivers/net/ethernet/atheros/atl1e/atl1e_main.c
index 4f7b65825c15..e0d89942d537 100644
--- a/drivers/net/ethernet/atheros/atl1e/atl1e_main.c
+++ b/drivers/net/ethernet/atheros/atl1e/atl1e_main.c
@@ -251,7 +251,7 @@ static void atl1e_cancel_work(struct atl1e_adapter *adapter)
  * atl1e_tx_timeout - Respond to a Tx Hang
  * @netdev: network interface device structure
  */
-static void atl1e_tx_timeout(struct net_device *netdev)
+static void atl1e_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct atl1e_adapter *adapter = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/atheros/atlx/atl2.c b/drivers/net/ethernet/atheros/atlx/atl2.c
index 3aba38322717..b81a4e0c5b57 100644
--- a/drivers/net/ethernet/atheros/atlx/atl2.c
+++ b/drivers/net/ethernet/atheros/atlx/atl2.c
@@ -1001,7 +1001,7 @@ static int atl2_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
  * atl2_tx_timeout - Respond to a Tx Hang
  * @netdev: network interface device structure
  */
-static void atl2_tx_timeout(struct net_device *netdev)
+static void atl2_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct atl2_adapter *adapter = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/atheros/atlx/atlx.c b/drivers/net/ethernet/atheros/atlx/atlx.c
index 505a22c703f7..0941d07d0833 100644
--- a/drivers/net/ethernet/atheros/atlx/atlx.c
+++ b/drivers/net/ethernet/atheros/atlx/atlx.c
@@ -183,7 +183,7 @@ static void atlx_clear_phy_int(struct atlx_adapter *adapter)
  * atlx_tx_timeout - Respond to a Tx Hang
  * @netdev: network interface device structure
  */
-static void atlx_tx_timeout(struct net_device *netdev)
+static void atlx_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct atlx_adapter *adapter = netdev_priv(netdev);
 	/* Do the reset outside of interrupt context */
diff --git a/drivers/net/ethernet/broadcom/b44.c b/drivers/net/ethernet/broadcom/b44.c
index 035dbb1b2c98..5b3464c3e8d1 100644
--- a/drivers/net/ethernet/broadcom/b44.c
+++ b/drivers/net/ethernet/broadcom/b44.c
@@ -948,7 +948,7 @@ irq_ack:
 	return IRQ_RETVAL(handled);
 }
 
-static void b44_tx_timeout(struct net_device *dev)
+static void b44_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct b44 *bp = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/broadcom/bcmsysport.c b/drivers/net/ethernet/broadcom/bcmsysport.c
index 825af709708e..8e3152779a61 100644
--- a/drivers/net/ethernet/broadcom/bcmsysport.c
+++ b/drivers/net/ethernet/broadcom/bcmsysport.c
@@ -1354,7 +1354,7 @@ out:
 	return ret;
 }
 
-static void bcm_sysport_tx_timeout(struct net_device *dev)
+static void bcm_sysport_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	netdev_warn(dev, "transmit timeout!\n");
 
diff --git a/drivers/net/ethernet/broadcom/bnx2.c b/drivers/net/ethernet/broadcom/bnx2.c
index fbc196b480b6..dbb7874607ca 100644
--- a/drivers/net/ethernet/broadcom/bnx2.c
+++ b/drivers/net/ethernet/broadcom/bnx2.c
@@ -6575,7 +6575,7 @@ bnx2_dump_state(struct bnx2 *bp)
 }
 
 static void
-bnx2_tx_timeout(struct net_device *dev)
+bnx2_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct bnx2 *bp = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
index 5e037a305b83..ee9e9290f112 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
@@ -4970,7 +4970,7 @@ int bnx2x_set_features(struct net_device *dev, netdev_features_t features)
 	return 0;
 }
 
-void bnx2x_tx_timeout(struct net_device *dev)
+void bnx2x_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct bnx2x *bp = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
index 8b08cb18e363..e35f48bfdc85 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
@@ -617,7 +617,7 @@ int bnx2x_set_features(struct net_device *dev, netdev_features_t features);
  *
  * @dev:	net device
  */
-void bnx2x_tx_timeout(struct net_device *dev);
+void bnx2x_tx_timeout(struct net_device *dev, unsigned int txqueue);
 
 /** bnx2x_get_c2s_mapping - read inner-to-outer vlan configuration
  * c2s_map should have BNX2X_MAX_PRIORITY entries.
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 85983f0e3134..4e34841906c7 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -9976,7 +9976,7 @@ static void bnxt_reset_task(struct bnxt *bp, bool silent)
 	}
 }
 
-static void bnxt_tx_timeout(struct net_device *dev)
+static void bnxt_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct bnxt *bp = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
index 120fa05a39ff..32f1245a69e2 100644
--- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c
+++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
@@ -3055,7 +3055,7 @@ static void bcmgenet_dump_tx_queue(struct bcmgenet_tx_ring *ring)
 		  ring->cb_ptr, ring->end_ptr);
 }
 
-static void bcmgenet_timeout(struct net_device *dev)
+static void bcmgenet_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct bcmgenet_priv *priv = netdev_priv(dev);
 	u32 int0_enable = 0;
diff --git a/drivers/net/ethernet/broadcom/sb1250-mac.c b/drivers/net/ethernet/broadcom/sb1250-mac.c
index 1604ad32e920..80ff52527233 100644
--- a/drivers/net/ethernet/broadcom/sb1250-mac.c
+++ b/drivers/net/ethernet/broadcom/sb1250-mac.c
@@ -294,7 +294,7 @@ static int sbmac_set_duplex(struct sbmac_softc *s, enum sbmac_duplex duplex,
 			    enum sbmac_fc fc);
 
 static int sbmac_open(struct net_device *dev);
-static void sbmac_tx_timeout (struct net_device *dev);
+static void sbmac_tx_timeout (struct net_device *dev, unsigned int txqueue);
 static void sbmac_set_rx_mode(struct net_device *dev);
 static int sbmac_mii_ioctl(struct net_device *dev, struct ifreq *rq, int cmd);
 static int sbmac_close(struct net_device *dev);
@@ -2419,7 +2419,7 @@ static void sbmac_mii_poll(struct net_device *dev)
 }
 
 
-static void sbmac_tx_timeout (struct net_device *dev)
+static void sbmac_tx_timeout (struct net_device *dev, unsigned int txqueue)
 {
 	struct sbmac_softc *sc = netdev_priv(dev);
 	unsigned long flags;
diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c
index ca3aa1250dd1..460b4992914a 100644
--- a/drivers/net/ethernet/broadcom/tg3.c
+++ b/drivers/net/ethernet/broadcom/tg3.c
@@ -7645,7 +7645,7 @@ static void tg3_poll_controller(struct net_device *dev)
 }
 #endif
 
-static void tg3_tx_timeout(struct net_device *dev)
+static void tg3_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct tg3 *tp = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/calxeda/xgmac.c b/drivers/net/ethernet/calxeda/xgmac.c
index af04a2c81adb..05a3d067c3fc 100644
--- a/drivers/net/ethernet/calxeda/xgmac.c
+++ b/drivers/net/ethernet/calxeda/xgmac.c
@@ -1251,7 +1251,7 @@ static int xgmac_poll(struct napi_struct *napi, int budget)
  *   netdev structure and arrange for the device to be reset to a sane state
  *   in order to transmit a new packet.
  */
-static void xgmac_tx_timeout(struct net_device *dev)
+static void xgmac_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct xgmac_priv *priv = netdev_priv(dev);
 	schedule_work(&priv->tx_timeout_work);
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c
index 7f3b2e3b0868..eab05b5534ea 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_main.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c
@@ -2562,7 +2562,7 @@ lio_xmit_failed:
 /** \brief Network device Tx timeout
  * @param netdev    pointer to network device
  */
-static void liquidio_tx_timeout(struct net_device *netdev)
+static void liquidio_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct lio *lio;
 
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c b/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c
index 370d76822ee0..7a77544a54f5 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c
@@ -1628,7 +1628,7 @@ lio_xmit_failed:
 /** \brief Network device Tx timeout
  * @param netdev    pointer to network device
  */
-static void liquidio_tx_timeout(struct net_device *netdev)
+static void liquidio_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct lio *lio;
 
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_vf_rep.c b/drivers/net/ethernet/cavium/liquidio/lio_vf_rep.c
index f3f2e71431ac..600de587d7a9 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_vf_rep.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_vf_rep.c
@@ -31,7 +31,7 @@ static int lio_vf_rep_open(struct net_device *ndev);
 static int lio_vf_rep_stop(struct net_device *ndev);
 static netdev_tx_t lio_vf_rep_pkt_xmit(struct sk_buff *skb,
 				       struct net_device *ndev);
-static void lio_vf_rep_tx_timeout(struct net_device *netdev);
+static void lio_vf_rep_tx_timeout(struct net_device *netdev, unsigned int txqueue);
 static int lio_vf_rep_phys_port_name(struct net_device *dev,
 				     char *buf, size_t len);
 static void lio_vf_rep_get_stats64(struct net_device *dev,
@@ -172,7 +172,7 @@ lio_vf_rep_stop(struct net_device *ndev)
 }
 
 static void
-lio_vf_rep_tx_timeout(struct net_device *ndev)
+lio_vf_rep_tx_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	netif_trans_update(ndev);
 
diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_main.c b/drivers/net/ethernet/cavium/thunder/nicvf_main.c
index f28409279ea4..016957285f99 100644
--- a/drivers/net/ethernet/cavium/thunder/nicvf_main.c
+++ b/drivers/net/ethernet/cavium/thunder/nicvf_main.c
@@ -1741,7 +1741,7 @@ static void nicvf_get_stats64(struct net_device *netdev,
 
 }
 
-static void nicvf_tx_timeout(struct net_device *dev)
+static void nicvf_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct nicvf *nic = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/cirrus/cs89x0.c b/drivers/net/ethernet/cirrus/cs89x0.c
index c9aebcde403a..33ace3307059 100644
--- a/drivers/net/ethernet/cirrus/cs89x0.c
+++ b/drivers/net/ethernet/cirrus/cs89x0.c
@@ -1128,7 +1128,7 @@ net_get_stats(struct net_device *dev)
 	return &dev->stats;
 }
 
-static void net_timeout(struct net_device *dev)
+static void net_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	/* If we get here, some higher level has decided we are broken.
 	   There should really be a "kick me" function call instead. */
diff --git a/drivers/net/ethernet/cisco/enic/enic_main.c b/drivers/net/ethernet/cisco/enic/enic_main.c
index acb2856936d2..bbd7b3175f09 100644
--- a/drivers/net/ethernet/cisco/enic/enic_main.c
+++ b/drivers/net/ethernet/cisco/enic/enic_main.c
@@ -1095,7 +1095,7 @@ static void enic_set_rx_mode(struct net_device *netdev)
 }
 
 /* netif_tx_lock held, BHs disabled */
-static void enic_tx_timeout(struct net_device *netdev)
+static void enic_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct enic *enic = netdev_priv(netdev);
 	schedule_work(&enic->tx_hang_reset);
diff --git a/drivers/net/ethernet/cortina/gemini.c b/drivers/net/ethernet/cortina/gemini.c
index a8f4c69252ff..de0b6e066eef 100644
--- a/drivers/net/ethernet/cortina/gemini.c
+++ b/drivers/net/ethernet/cortina/gemini.c
@@ -1296,7 +1296,7 @@ out_drop:
 	return NETDEV_TX_OK;
 }
 
-static void gmac_tx_timeout(struct net_device *netdev)
+static void gmac_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	netdev_err(netdev, "Tx timeout\n");
 	gmac_dump_dma_state(netdev);
diff --git a/drivers/net/ethernet/davicom/dm9000.c b/drivers/net/ethernet/davicom/dm9000.c
index cce90b5925d9..1ea3372775e6 100644
--- a/drivers/net/ethernet/davicom/dm9000.c
+++ b/drivers/net/ethernet/davicom/dm9000.c
@@ -964,7 +964,7 @@ dm9000_init_dm9000(struct net_device *dev)
 }
 
 /* Our watchdog timed out. Called by the networking layer */
-static void dm9000_timeout(struct net_device *dev)
+static void dm9000_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct board_info *db = netdev_priv(dev);
 	u8 reg_save;
diff --git a/drivers/net/ethernet/dec/tulip/de2104x.c b/drivers/net/ethernet/dec/tulip/de2104x.c
index f1a2da15dd0a..fd3c2abf74b5 100644
--- a/drivers/net/ethernet/dec/tulip/de2104x.c
+++ b/drivers/net/ethernet/dec/tulip/de2104x.c
@@ -1436,7 +1436,7 @@ static int de_close (struct net_device *dev)
 	return 0;
 }
 
-static void de_tx_timeout (struct net_device *dev)
+static void de_tx_timeout (struct net_device *dev, unsigned int txqueue)
 {
 	struct de_private *de = netdev_priv(dev);
 	const int irq = de->pdev->irq;
diff --git a/drivers/net/ethernet/dec/tulip/tulip_core.c b/drivers/net/ethernet/dec/tulip/tulip_core.c
index 3e3e08698876..9e9d9eee29d9 100644
--- a/drivers/net/ethernet/dec/tulip/tulip_core.c
+++ b/drivers/net/ethernet/dec/tulip/tulip_core.c
@@ -255,7 +255,7 @@ MODULE_DEVICE_TABLE(pci, tulip_pci_tbl);
 const char tulip_media_cap[32] =
 {0,0,0,16,  3,19,16,24,  27,4,7,5, 0,20,23,20,  28,31,0,0, };
 
-static void tulip_tx_timeout(struct net_device *dev);
+static void tulip_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static void tulip_init_ring(struct net_device *dev);
 static void tulip_free_ring(struct net_device *dev);
 static netdev_tx_t tulip_start_xmit(struct sk_buff *skb,
@@ -534,7 +534,7 @@ free_ring:
 }
 
 
-static void tulip_tx_timeout(struct net_device *dev)
+static void tulip_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct tulip_private *tp = netdev_priv(dev);
 	void __iomem *ioaddr = tp->base_addr;
diff --git a/drivers/net/ethernet/dec/tulip/winbond-840.c b/drivers/net/ethernet/dec/tulip/winbond-840.c
index 70cb2d689c2c..7f136488e67c 100644
--- a/drivers/net/ethernet/dec/tulip/winbond-840.c
+++ b/drivers/net/ethernet/dec/tulip/winbond-840.c
@@ -331,7 +331,7 @@ static void netdev_timer(struct timer_list *t);
 static void init_rxtx_rings(struct net_device *dev);
 static void free_rxtx_rings(struct netdev_private *np);
 static void init_registers(struct net_device *dev);
-static void tx_timeout(struct net_device *dev);
+static void tx_timeout(struct net_device *dev, unsigned int txqueue);
 static int alloc_ringdesc(struct net_device *dev);
 static void free_ringdesc(struct netdev_private *np);
 static netdev_tx_t start_tx(struct sk_buff *skb, struct net_device *dev);
@@ -921,7 +921,7 @@ static void init_registers(struct net_device *dev)
 	iowrite32(0, ioaddr + RxStartDemand);
 }
 
-static void tx_timeout(struct net_device *dev)
+static void tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct netdev_private *np = netdev_priv(dev);
 	void __iomem *ioaddr = np->base_addr;
diff --git a/drivers/net/ethernet/dlink/dl2k.c b/drivers/net/ethernet/dlink/dl2k.c
index 55e720d2ea0c..26c5da032b1e 100644
--- a/drivers/net/ethernet/dlink/dl2k.c
+++ b/drivers/net/ethernet/dlink/dl2k.c
@@ -66,7 +66,7 @@ static const int multicast_filter_limit = 0x40;
 
 static int rio_open (struct net_device *dev);
 static void rio_timer (struct timer_list *t);
-static void rio_tx_timeout (struct net_device *dev);
+static void rio_tx_timeout (struct net_device *dev, unsigned int txqueue);
 static netdev_tx_t start_xmit (struct sk_buff *skb, struct net_device *dev);
 static irqreturn_t rio_interrupt (int irq, void *dev_instance);
 static void rio_free_tx (struct net_device *dev, int irq);
@@ -696,7 +696,7 @@ rio_timer (struct timer_list *t)
 }
 
 static void
-rio_tx_timeout (struct net_device *dev)
+rio_tx_timeout (struct net_device *dev, unsigned int txqueue)
 {
 	struct netdev_private *np = netdev_priv(dev);
 	void __iomem *ioaddr = np->ioaddr;
diff --git a/drivers/net/ethernet/dlink/sundance.c b/drivers/net/ethernet/dlink/sundance.c
index 4a37a69764ce..b91387c456ba 100644
--- a/drivers/net/ethernet/dlink/sundance.c
+++ b/drivers/net/ethernet/dlink/sundance.c
@@ -432,7 +432,7 @@ static int  mdio_wait_link(struct net_device *dev, int wait);
 static int  netdev_open(struct net_device *dev);
 static void check_duplex(struct net_device *dev);
 static void netdev_timer(struct timer_list *t);
-static void tx_timeout(struct net_device *dev);
+static void tx_timeout(struct net_device *dev, unsigned int txqueue);
 static void init_ring(struct net_device *dev);
 static netdev_tx_t start_tx(struct sk_buff *skb, struct net_device *dev);
 static int reset_tx (struct net_device *dev);
@@ -969,7 +969,7 @@ static void netdev_timer(struct timer_list *t)
 	add_timer(&np->timer);
 }
 
-static void tx_timeout(struct net_device *dev)
+static void tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct netdev_private *np = netdev_priv(dev);
 	void __iomem *ioaddr = np->base;
diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index 39eb7d525043..56f59db6ebf2 100644
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -1417,7 +1417,7 @@ drop:
 	return NETDEV_TX_OK;
 }
 
-static void be_tx_timeout(struct net_device *netdev)
+static void be_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
 	struct device *dev = &adapter->pdev->dev;
diff --git a/drivers/net/ethernet/ethoc.c b/drivers/net/ethernet/ethoc.c
index ea4f17f5cce7..66406da16b60 100644
--- a/drivers/net/ethernet/ethoc.c
+++ b/drivers/net/ethernet/ethoc.c
@@ -869,7 +869,7 @@ static int ethoc_change_mtu(struct net_device *dev, int new_mtu)
 	return -ENOSYS;
 }
 
-static void ethoc_tx_timeout(struct net_device *dev)
+static void ethoc_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct ethoc *priv = netdev_priv(dev);
 	u32 pending = ethoc_read(priv, INT_SOURCE);
diff --git a/drivers/net/ethernet/faraday/ftgmac100.c b/drivers/net/ethernet/faraday/ftgmac100.c
index 8ed85037f021..48b3b72fe02e 100644
--- a/drivers/net/ethernet/faraday/ftgmac100.c
+++ b/drivers/net/ethernet/faraday/ftgmac100.c
@@ -1545,7 +1545,7 @@ static int ftgmac100_do_ioctl(struct net_device *netdev, struct ifreq *ifr, int
 	return phy_mii_ioctl(netdev->phydev, ifr, cmd);
 }
 
-static void ftgmac100_tx_timeout(struct net_device *netdev)
+static void ftgmac100_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct ftgmac100 *priv = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/fealnx.c b/drivers/net/ethernet/fealnx.c
index c24fd56a2c71..84f10970299a 100644
--- a/drivers/net/ethernet/fealnx.c
+++ b/drivers/net/ethernet/fealnx.c
@@ -428,7 +428,7 @@ static void getlinktype(struct net_device *dev);
 static void getlinkstatus(struct net_device *dev);
 static void netdev_timer(struct timer_list *t);
 static void reset_timer(struct timer_list *t);
-static void fealnx_tx_timeout(struct net_device *dev);
+static void fealnx_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static void init_ring(struct net_device *dev);
 static netdev_tx_t start_tx(struct sk_buff *skb, struct net_device *dev);
 static irqreturn_t intr_handler(int irq, void *dev_instance);
@@ -1191,7 +1191,7 @@ static void reset_timer(struct timer_list *t)
 }
 
 
-static void fealnx_tx_timeout(struct net_device *dev)
+static void fealnx_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct netdev_private *np = netdev_priv(dev);
 	void __iomem *ioaddr = np->mem;
diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
index 6a9d12dad5d9..a60fc3cfc06e 100644
--- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
+++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
@@ -288,7 +288,7 @@ static int dpaa_stop(struct net_device *net_dev)
 	return err;
 }
 
-static void dpaa_tx_timeout(struct net_device *net_dev)
+static void dpaa_tx_timeout(struct net_device *net_dev, unsigned int txqueue)
 {
 	struct dpaa_percpu_priv *percpu_priv;
 	const struct dpaa_priv	*priv;
diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c
index 05c1899f6628..798fed37be46 100644
--- a/drivers/net/ethernet/freescale/fec_main.c
+++ b/drivers/net/ethernet/freescale/fec_main.c
@@ -1141,7 +1141,7 @@ fec_stop(struct net_device *ndev)
 
 
 static void
-fec_timeout(struct net_device *ndev)
+fec_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
 
diff --git a/drivers/net/ethernet/freescale/fec_mpc52xx.c b/drivers/net/ethernet/freescale/fec_mpc52xx.c
index 30cdb246d020..de5278485062 100644
--- a/drivers/net/ethernet/freescale/fec_mpc52xx.c
+++ b/drivers/net/ethernet/freescale/fec_mpc52xx.c
@@ -84,7 +84,7 @@ static int debug = -1;	/* the above default */
 module_param(debug, int, 0);
 MODULE_PARM_DESC(debug, "debugging messages level");
 
-static void mpc52xx_fec_tx_timeout(struct net_device *dev)
+static void mpc52xx_fec_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct mpc52xx_fec_priv *priv = netdev_priv(dev);
 	unsigned long flags;
diff --git a/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c b/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c
index 3981c06f082f..80903cd58468 100644
--- a/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c
+++ b/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c
@@ -641,7 +641,7 @@ static void fs_timeout_work(struct work_struct *work)
 		netif_wake_queue(dev);
 }
 
-static void fs_timeout(struct net_device *dev)
+static void fs_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct fs_enet_private *fep = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net/ethernet/freescale/gianfar.c
index 72868a28b621..b636d83a7ee9 100644
--- a/drivers/net/ethernet/freescale/gianfar.c
+++ b/drivers/net/ethernet/freescale/gianfar.c
@@ -2093,7 +2093,7 @@ static void gfar_reset_task(struct work_struct *work)
 	reset_gfar(priv->ndev);
 }
 
-static void gfar_timeout(struct net_device *dev)
+static void gfar_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct gfar_private *priv = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/freescale/ucc_geth.c b/drivers/net/ethernet/freescale/ucc_geth.c
index f839fa94ebdd..0d101c00286f 100644
--- a/drivers/net/ethernet/freescale/ucc_geth.c
+++ b/drivers/net/ethernet/freescale/ucc_geth.c
@@ -3545,7 +3545,7 @@ static void ucc_geth_timeout_work(struct work_struct *work)
  * ucc_geth_timeout gets called when a packet has not been
  * transmitted after a set amount of time.
  */
-static void ucc_geth_timeout(struct net_device *dev)
+static void ucc_geth_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct ucc_geth_private *ugeth = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/fujitsu/fmvj18x_cs.c b/drivers/net/ethernet/fujitsu/fmvj18x_cs.c
index 1eca0fdb9933..a7b7a4aace79 100644
--- a/drivers/net/ethernet/fujitsu/fmvj18x_cs.c
+++ b/drivers/net/ethernet/fujitsu/fmvj18x_cs.c
@@ -93,7 +93,7 @@ static irqreturn_t fjn_interrupt(int irq, void *dev_id);
 static void fjn_rx(struct net_device *dev);
 static void fjn_reset(struct net_device *dev);
 static void set_rx_mode(struct net_device *dev);
-static void fjn_tx_timeout(struct net_device *dev);
+static void fjn_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static const struct ethtool_ops netdev_ethtool_ops;
 
 /*
@@ -774,7 +774,7 @@ static irqreturn_t fjn_interrupt(int dummy, void *dev_id)
 
 /*====================================================================*/
 
-static void fjn_tx_timeout(struct net_device *dev)
+static void fjn_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
     struct local_info *lp = netdev_priv(dev);
     unsigned int ioaddr = dev->base_addr;
diff --git a/drivers/net/ethernet/google/gve/gve_main.c b/drivers/net/ethernet/google/gve/gve_main.c
index 9b7a8db9860f..e032563ceefd 100644
--- a/drivers/net/ethernet/google/gve/gve_main.c
+++ b/drivers/net/ethernet/google/gve/gve_main.c
@@ -845,7 +845,7 @@ static void gve_turnup(struct gve_priv *priv)
 	gve_set_napi_enabled(priv);
 }
 
-static void gve_tx_timeout(struct net_device *dev)
+static void gve_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct gve_priv *priv = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/hisilicon/hip04_eth.c b/drivers/net/ethernet/hisilicon/hip04_eth.c
index 3e9b6d543c77..dc8dd5fc1559 100644
--- a/drivers/net/ethernet/hisilicon/hip04_eth.c
+++ b/drivers/net/ethernet/hisilicon/hip04_eth.c
@@ -779,7 +779,7 @@ static int hip04_mac_stop(struct net_device *ndev)
 	return 0;
 }
 
-static void hip04_timeout(struct net_device *ndev)
+static void hip04_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	struct hip04_priv *priv = netdev_priv(ndev);
 
diff --git a/drivers/net/ethernet/hisilicon/hix5hd2_gmac.c b/drivers/net/ethernet/hisilicon/hix5hd2_gmac.c
index 247de9105d10..4fb776920a93 100644
--- a/drivers/net/ethernet/hisilicon/hix5hd2_gmac.c
+++ b/drivers/net/ethernet/hisilicon/hix5hd2_gmac.c
@@ -893,7 +893,7 @@ static void hix5hd2_tx_timeout_task(struct work_struct *work)
 	hix5hd2_net_open(priv->netdev);
 }
 
-static void hix5hd2_net_timeout(struct net_device *dev)
+static void hix5hd2_net_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct hix5hd2_priv *priv = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_enet.c b/drivers/net/ethernet/hisilicon/hns/hns_enet.c
index 14ab20491fd0..e45553ec114a 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_enet.c
+++ b/drivers/net/ethernet/hisilicon/hns/hns_enet.c
@@ -1485,7 +1485,7 @@ static int hns_nic_net_stop(struct net_device *ndev)
 
 static void hns_tx_timeout_reset(struct hns_nic_priv *priv);
 #define HNS_TX_TIMEO_LIMIT (40 * HZ)
-static void hns_nic_net_timeout(struct net_device *ndev)
+static void hns_nic_net_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	struct hns_nic_priv *priv = netdev_priv(ndev);
 
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
index 69545dd6c938..b113b895dab7 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
@@ -1869,7 +1869,7 @@ static bool hns3_get_tx_timeo_queue_info(struct net_device *ndev)
 	return true;
 }
 
-static void hns3_nic_net_timeout(struct net_device *ndev)
+static void hns3_nic_net_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	struct hns3_nic_priv *priv = netdev_priv(ndev);
 	struct hnae3_handle *h = priv->ae_handle;
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_main.c b/drivers/net/ethernet/huawei/hinic/hinic_main.c
index 2411ad270c98..02a14f5e7fe3 100644
--- a/drivers/net/ethernet/huawei/hinic/hinic_main.c
+++ b/drivers/net/ethernet/huawei/hinic/hinic_main.c
@@ -766,7 +766,7 @@ static void hinic_set_rx_mode(struct net_device *netdev)
 	queue_work(nic_dev->workq, &rx_mode_work->work);
 }
 
-static void hinic_tx_timeout(struct net_device *netdev)
+static void hinic_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct hinic_dev *nic_dev = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/i825xx/82596.c b/drivers/net/ethernet/i825xx/82596.c
index 92929750f832..bef676d93339 100644
--- a/drivers/net/ethernet/i825xx/82596.c
+++ b/drivers/net/ethernet/i825xx/82596.c
@@ -363,7 +363,7 @@ static netdev_tx_t i596_start_xmit(struct sk_buff *skb, struct net_device *dev);
 static irqreturn_t i596_interrupt(int irq, void *dev_id);
 static int i596_close(struct net_device *dev);
 static void i596_add_cmd(struct net_device *dev, struct i596_cmd *cmd);
-static void i596_tx_timeout (struct net_device *dev);
+static void i596_tx_timeout (struct net_device *dev, unsigned int txqueue);
 static void print_eth(unsigned char *buf, char *str);
 static void set_multicast_list(struct net_device *dev);
 
@@ -1019,7 +1019,7 @@ err_irq_dev:
 	return res;
 }
 
-static void i596_tx_timeout (struct net_device *dev)
+static void i596_tx_timeout (struct net_device *dev, unsigned int txqueue)
 {
 	struct i596_private *lp = dev->ml_priv;
 	int ioaddr = dev->base_addr;
diff --git a/drivers/net/ethernet/i825xx/ether1.c b/drivers/net/ethernet/i825xx/ether1.c
index bb3b8adbe4f0..a0bfb509e002 100644
--- a/drivers/net/ethernet/i825xx/ether1.c
+++ b/drivers/net/ethernet/i825xx/ether1.c
@@ -66,7 +66,7 @@ static netdev_tx_t ether1_sendpacket(struct sk_buff *skb,
 static irqreturn_t ether1_interrupt(int irq, void *dev_id);
 static int ether1_close(struct net_device *dev);
 static void ether1_setmulticastlist(struct net_device *dev);
-static void ether1_timeout(struct net_device *dev);
+static void ether1_timeout(struct net_device *dev, unsigned int txqueue);
 
 /* ------------------------------------------------------------------------- */
 
@@ -650,7 +650,7 @@ ether1_open (struct net_device *dev)
 }
 
 static void
-ether1_timeout(struct net_device *dev)
+ether1_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	printk(KERN_WARNING "%s: transmit timeout, network cable problem?\n",
 		dev->name);
diff --git a/drivers/net/ethernet/i825xx/lib82596.c b/drivers/net/ethernet/i825xx/lib82596.c
index f9742af7f142..b03757e169e4 100644
--- a/drivers/net/ethernet/i825xx/lib82596.c
+++ b/drivers/net/ethernet/i825xx/lib82596.c
@@ -351,7 +351,7 @@ static netdev_tx_t i596_start_xmit(struct sk_buff *skb, struct net_device *dev);
 static irqreturn_t i596_interrupt(int irq, void *dev_id);
 static int i596_close(struct net_device *dev);
 static void i596_add_cmd(struct net_device *dev, struct i596_cmd *cmd);
-static void i596_tx_timeout (struct net_device *dev);
+static void i596_tx_timeout (struct net_device *dev, unsigned int txqueue);
 static void print_eth(unsigned char *buf, char *str);
 static void set_multicast_list(struct net_device *dev);
 static inline void ca(struct net_device *dev);
@@ -936,7 +936,7 @@ out_remove_rx_bufs:
 	return -EAGAIN;
 }
 
-static void i596_tx_timeout (struct net_device *dev)
+static void i596_tx_timeout (struct net_device *dev, unsigned int txqueue)
 {
 	struct i596_private *lp = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/i825xx/sun3_82586.c b/drivers/net/ethernet/i825xx/sun3_82586.c
index 1a86184d44c0..4564ee02c95f 100644
--- a/drivers/net/ethernet/i825xx/sun3_82586.c
+++ b/drivers/net/ethernet/i825xx/sun3_82586.c
@@ -125,7 +125,7 @@ static netdev_tx_t     sun3_82586_send_packet(struct sk_buff *,
 					      struct net_device *);
 static struct  net_device_stats *sun3_82586_get_stats(struct net_device *dev);
 static void    set_multicast_list(struct net_device *dev);
-static void    sun3_82586_timeout(struct net_device *dev);
+static void    sun3_82586_timeout(struct net_device *dev, unsigned int txqueue);
 #if 0
 static void    sun3_82586_dump(struct net_device *,void *);
 #endif
@@ -965,7 +965,7 @@ static void startrecv586(struct net_device *dev)
 	WAIT_4_SCB_CMD_RUC();	/* wait for accept cmd. (no timeout!!) */
 }
 
-static void sun3_82586_timeout(struct net_device *dev)
+static void sun3_82586_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct priv *p = netdev_priv(dev);
 #ifndef NO_NOPCOMMANDS
diff --git a/drivers/net/ethernet/ibm/ehea/ehea_main.c b/drivers/net/ethernet/ibm/ehea/ehea_main.c
index 13e30eba5349..0273fb7a9d01 100644
--- a/drivers/net/ethernet/ibm/ehea/ehea_main.c
+++ b/drivers/net/ethernet/ibm/ehea/ehea_main.c
@@ -2786,7 +2786,7 @@ out:
 	return;
 }
 
-static void ehea_tx_watchdog(struct net_device *dev)
+static void ehea_tx_watchdog(struct net_device *dev, unsigned int txqueue)
 {
 	struct ehea_port *port = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/ibm/emac/core.c b/drivers/net/ethernet/ibm/emac/core.c
index 2e40425d8a34..b7fc17756c51 100644
--- a/drivers/net/ethernet/ibm/emac/core.c
+++ b/drivers/net/ethernet/ibm/emac/core.c
@@ -776,7 +776,7 @@ static void emac_reset_work(struct work_struct *work)
 	mutex_unlock(&dev->link_lock);
 }
 
-static void emac_tx_timeout(struct net_device *ndev)
+static void emac_tx_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	struct emac_instance *dev = netdev_priv(ndev);
 
diff --git a/drivers/net/ethernet/ibm/ibmvnic.c b/drivers/net/ethernet/ibm/ibmvnic.c
index c90080781924..94b9d8913b66 100644
--- a/drivers/net/ethernet/ibm/ibmvnic.c
+++ b/drivers/net/ethernet/ibm/ibmvnic.c
@@ -2282,7 +2282,7 @@ err:
 	return -ret;
 }
 
-static void ibmvnic_tx_timeout(struct net_device *dev)
+static void ibmvnic_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct ibmvnic_adapter *adapter = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/intel/e100.c b/drivers/net/ethernet/intel/e100.c
index a65d5a9ba7db..1b8d015ebfb0 100644
--- a/drivers/net/ethernet/intel/e100.c
+++ b/drivers/net/ethernet/intel/e100.c
@@ -2316,7 +2316,7 @@ static void e100_down(struct nic *nic)
 	e100_rx_clean_list(nic);
 }
 
-static void e100_tx_timeout(struct net_device *netdev)
+static void e100_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct nic *nic = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/intel/e1000/e1000_main.c b/drivers/net/ethernet/intel/e1000/e1000_main.c
index aca97b084003..2bced34c19ba 100644
--- a/drivers/net/ethernet/intel/e1000/e1000_main.c
+++ b/drivers/net/ethernet/intel/e1000/e1000_main.c
@@ -134,7 +134,7 @@ static int e1000_mii_ioctl(struct net_device *netdev, struct ifreq *ifr,
 			   int cmd);
 static void e1000_enter_82542_rst(struct e1000_adapter *adapter);
 static void e1000_leave_82542_rst(struct e1000_adapter *adapter);
-static void e1000_tx_timeout(struct net_device *dev);
+static void e1000_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static void e1000_reset_task(struct work_struct *work);
 static void e1000_smartspeed(struct e1000_adapter *adapter);
 static int e1000_82547_fifo_workaround(struct e1000_adapter *adapter,
@@ -3488,7 +3488,7 @@ exit:
  * e1000_tx_timeout - Respond to a Tx Hang
  * @netdev: network interface device structure
  **/
-static void e1000_tx_timeout(struct net_device *netdev)
+static void e1000_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c
index fe7997c18a10..4c220600ea9a 100644
--- a/drivers/net/ethernet/intel/e1000e/netdev.c
+++ b/drivers/net/ethernet/intel/e1000e/netdev.c
@@ -5929,7 +5929,7 @@ static netdev_tx_t e1000_xmit_frame(struct sk_buff *skb,
  * e1000_tx_timeout - Respond to a Tx Hang
  * @netdev: network interface device structure
  **/
-static void e1000_tx_timeout(struct net_device *netdev)
+static void e1000_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c b/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c
index 68baee04dc58..ba2566e2123d 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c
@@ -697,7 +697,7 @@ static netdev_tx_t fm10k_xmit_frame(struct sk_buff *skb, struct net_device *dev)
  * fm10k_tx_timeout - Respond to a Tx Hang
  * @netdev: network interface device structure
  **/
-static void fm10k_tx_timeout(struct net_device *netdev)
+static void fm10k_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct fm10k_intfc *interface = netdev_priv(netdev);
 	bool real_tx_hang = false;
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 1ccabeafa44c..4c9ac6c80eb8 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -301,7 +301,7 @@ void i40e_service_event_schedule(struct i40e_pf *pf)
  * device is munged, not just the one netdev port, so go for the full
  * reset.
  **/
-static void i40e_tx_timeout(struct net_device *netdev)
+static void i40e_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
 	struct i40e_vsi *vsi = np->vsi;
diff --git a/drivers/net/ethernet/intel/iavf/iavf_main.c b/drivers/net/ethernet/intel/iavf/iavf_main.c
index 821987da5698..0a8824871618 100644
--- a/drivers/net/ethernet/intel/iavf/iavf_main.c
+++ b/drivers/net/ethernet/intel/iavf/iavf_main.c
@@ -159,7 +159,7 @@ void iavf_schedule_reset(struct iavf_adapter *adapter)
  * iavf_tx_timeout - Respond to a Tx Hang
  * @netdev: network interface device structure
  **/
-static void iavf_tx_timeout(struct net_device *netdev)
+static void iavf_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct iavf_adapter *adapter = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 69bff085acf7..4d5220c9c721 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -5060,7 +5060,7 @@ ice_bridge_setlink(struct net_device *dev, struct nlmsghdr *nlh,
  * ice_tx_timeout - Respond to a Tx Hang
  * @netdev: network interface device structure
  */
-static void ice_tx_timeout(struct net_device *netdev)
+static void ice_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct ice_netdev_priv *np = netdev_priv(netdev);
 	struct ice_ring *tx_ring = NULL;
diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 98346eb064d5..d11e64a58ed1 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -146,7 +146,7 @@ static int igb_poll(struct napi_struct *, int);
 static bool igb_clean_tx_irq(struct igb_q_vector *, int);
 static int igb_clean_rx_irq(struct igb_q_vector *, int);
 static int igb_ioctl(struct net_device *, struct ifreq *, int cmd);
-static void igb_tx_timeout(struct net_device *);
+static void igb_tx_timeout(struct net_device *, unsigned int txqueue);
 static void igb_reset_task(struct work_struct *);
 static void igb_vlan_mode(struct net_device *netdev,
 			  netdev_features_t features);
@@ -6184,7 +6184,7 @@ static netdev_tx_t igb_xmit_frame(struct sk_buff *skb,
  *  igb_tx_timeout - Respond to a Tx Hang
  *  @netdev: network interface device structure
  **/
-static void igb_tx_timeout(struct net_device *netdev)
+static void igb_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct igb_adapter *adapter = netdev_priv(netdev);
 	struct e1000_hw *hw = &adapter->hw;
diff --git a/drivers/net/ethernet/intel/igbvf/netdev.c b/drivers/net/ethernet/intel/igbvf/netdev.c
index 6003dc3ff5fd..5b1800c3ba82 100644
--- a/drivers/net/ethernet/intel/igbvf/netdev.c
+++ b/drivers/net/ethernet/intel/igbvf/netdev.c
@@ -2375,7 +2375,7 @@ static netdev_tx_t igbvf_xmit_frame(struct sk_buff *skb,
  * igbvf_tx_timeout - Respond to a Tx Hang
  * @netdev: network interface device structure
  **/
-static void igbvf_tx_timeout(struct net_device *netdev)
+static void igbvf_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct igbvf_adapter *adapter = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/intel/ixgb/ixgb_main.c b/drivers/net/ethernet/intel/ixgb/ixgb_main.c
index 3d8c051dd327..b64e91ea3465 100644
--- a/drivers/net/ethernet/intel/ixgb/ixgb_main.c
+++ b/drivers/net/ethernet/intel/ixgb/ixgb_main.c
@@ -70,7 +70,7 @@ static int ixgb_clean(struct napi_struct *, int);
 static bool ixgb_clean_rx_irq(struct ixgb_adapter *, int *, int);
 static void ixgb_alloc_rx_buffers(struct ixgb_adapter *, int);
 
-static void ixgb_tx_timeout(struct net_device *dev);
+static void ixgb_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static void ixgb_tx_timeout_task(struct work_struct *work);
 
 static void ixgb_vlan_strip_enable(struct ixgb_adapter *adapter);
@@ -1538,7 +1538,7 @@ ixgb_xmit_frame(struct sk_buff *skb, struct net_device *netdev)
  **/
 
 static void
-ixgb_tx_timeout(struct net_device *netdev)
+ixgb_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct ixgb_adapter *adapter = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_debugfs.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_debugfs.c
index 171cdc552961..5b1cf49df3d3 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_debugfs.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_debugfs.c
@@ -166,7 +166,9 @@ static ssize_t ixgbe_dbg_netdev_ops_write(struct file *filp,
 	ixgbe_dbg_netdev_ops_buf[len] = '\0';
 
 	if (strncmp(ixgbe_dbg_netdev_ops_buf, "tx_timeout", 10) == 0) {
-		adapter->netdev->netdev_ops->ndo_tx_timeout(adapter->netdev);
+		/* TX Queue number below is wrong, but ixgbe does not use it */
+		adapter->netdev->netdev_ops->ndo_tx_timeout(adapter->netdev,
+							    UINT_MAX);
 		e_dev_info("tx_timeout called\n");
 	} else {
 		e_dev_info("Unknown command: %s\n", ixgbe_dbg_netdev_ops_buf);
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 25c097cd8100..8129ea2e94a8 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -6158,7 +6158,7 @@ static void ixgbe_set_eee_capable(struct ixgbe_adapter *adapter)
  * ixgbe_tx_timeout - Respond to a Tx Hang
  * @netdev: network interface device structure
  **/
-static void ixgbe_tx_timeout(struct net_device *netdev)
+static void ixgbe_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c
index 076f2da36f27..fa286694ac2c 100644
--- a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c
+++ b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c
@@ -250,7 +250,7 @@ static void ixgbevf_tx_timeout_reset(struct ixgbevf_adapter *adapter)
  * ixgbevf_tx_timeout - Respond to a Tx Hang
  * @netdev: network interface device structure
  **/
-static void ixgbevf_tx_timeout(struct net_device *netdev)
+static void ixgbevf_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct ixgbevf_adapter *adapter = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/jme.c b/drivers/net/ethernet/jme.c
index 25aa400e2e3c..2e4975572e9f 100644
--- a/drivers/net/ethernet/jme.c
+++ b/drivers/net/ethernet/jme.c
@@ -2337,7 +2337,7 @@ jme_change_mtu(struct net_device *netdev, int new_mtu)
 }
 
 static void
-jme_tx_timeout(struct net_device *netdev)
+jme_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct jme_adapter *jme = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/korina.c b/drivers/net/ethernet/korina.c
index ae195f8adff5..f98d9d627c71 100644
--- a/drivers/net/ethernet/korina.c
+++ b/drivers/net/ethernet/korina.c
@@ -917,7 +917,7 @@ static void korina_restart_task(struct work_struct *work)
 	enable_irq(lp->rx_irq);
 }
 
-static void korina_tx_timeout(struct net_device *dev)
+static void korina_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct korina_private *lp = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/lantiq_etop.c b/drivers/net/ethernet/lantiq_etop.c
index 6e73ffe6f928..028e3e6222e9 100644
--- a/drivers/net/ethernet/lantiq_etop.c
+++ b/drivers/net/ethernet/lantiq_etop.c
@@ -594,7 +594,7 @@ err_hw:
 }
 
 static void
-ltq_etop_tx_timeout(struct net_device *dev)
+ltq_etop_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	int err;
 
diff --git a/drivers/net/ethernet/marvell/mv643xx_eth.c b/drivers/net/ethernet/marvell/mv643xx_eth.c
index d5b644131cff..c43820597218 100644
--- a/drivers/net/ethernet/marvell/mv643xx_eth.c
+++ b/drivers/net/ethernet/marvell/mv643xx_eth.c
@@ -2590,7 +2590,7 @@ static void tx_timeout_task(struct work_struct *ugly)
 	}
 }
 
-static void mv643xx_eth_tx_timeout(struct net_device *dev)
+static void mv643xx_eth_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/marvell/pxa168_eth.c b/drivers/net/ethernet/marvell/pxa168_eth.c
index 3fb7ee3d4d13..1a6877902dd6 100644
--- a/drivers/net/ethernet/marvell/pxa168_eth.c
+++ b/drivers/net/ethernet/marvell/pxa168_eth.c
@@ -742,7 +742,7 @@ txq_reclaim_end:
 	return released;
 }
 
-static void pxa168_eth_tx_timeout(struct net_device *dev)
+static void pxa168_eth_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct pxa168_eth_private *pep = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/marvell/skge.c b/drivers/net/ethernet/marvell/skge.c
index 095f6c71b4fa..8ca15958e752 100644
--- a/drivers/net/ethernet/marvell/skge.c
+++ b/drivers/net/ethernet/marvell/skge.c
@@ -2884,7 +2884,7 @@ static void skge_tx_clean(struct net_device *dev)
 	skge->tx_ring.to_clean = e;
 }
 
-static void skge_tx_timeout(struct net_device *dev)
+static void skge_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct skge_port *skge = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/marvell/sky2.c b/drivers/net/ethernet/marvell/sky2.c
index 5f56ee83e3b1..acd1cba987fb 100644
--- a/drivers/net/ethernet/marvell/sky2.c
+++ b/drivers/net/ethernet/marvell/sky2.c
@@ -2358,7 +2358,7 @@ static void sky2_qlink_intr(struct sky2_hw *hw)
 /* Transmit timeout is only called if we are running, carrier is up
  * and tx queue is full (stopped).
  */
-static void sky2_tx_timeout(struct net_device *dev)
+static void sky2_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct sky2_port *sky2 = netdev_priv(dev);
 	struct sky2_hw *hw = sky2->hw;
diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
index 527ad2aadcca..8c6cfd15481c 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
@@ -2081,7 +2081,7 @@ static void mtk_dma_free(struct mtk_eth *eth)
 	kfree(eth->scratch_head);
 }
 
-static void mtk_tx_timeout(struct net_device *dev)
+static void mtk_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct mtk_mac *mac = netdev_priv(dev);
 	struct mtk_eth *eth = mac->hw;
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
index 7af75b63245f..71c083960a87 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
@@ -1363,7 +1363,7 @@ static void mlx4_en_delete_rss_steer_rules(struct mlx4_en_priv *priv)
 	}
 }
 
-static void mlx4_en_tx_timeout(struct net_device *dev)
+static void mlx4_en_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct mlx4_en_priv *priv = netdev_priv(dev);
 	struct mlx4_en_dev *mdev = priv->mdev;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 4980e80a5e85..68f1c8cb302b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -4332,7 +4332,7 @@ unlock:
 	rtnl_unlock();
 }
 
-static void mlx5e_tx_timeout(struct net_device *dev)
+static void mlx5e_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct mlx5e_priv *priv = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/micrel/ks8842.c b/drivers/net/ethernet/micrel/ks8842.c
index da329ca115cc..f3f6dfe3eddc 100644
--- a/drivers/net/ethernet/micrel/ks8842.c
+++ b/drivers/net/ethernet/micrel/ks8842.c
@@ -1103,7 +1103,7 @@ static void ks8842_tx_timeout_work(struct work_struct *work)
 		__ks8842_start_new_rx_dma(netdev);
 }
 
-static void ks8842_tx_timeout(struct net_device *netdev)
+static void ks8842_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct ks8842_adapter *adapter = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/micrel/ksz884x.c b/drivers/net/ethernet/micrel/ksz884x.c
index e102e1560ac7..d1444ba36e10 100644
--- a/drivers/net/ethernet/micrel/ksz884x.c
+++ b/drivers/net/ethernet/micrel/ksz884x.c
@@ -4896,7 +4896,7 @@ unlock:
  * triggered to free up resources so that the transmit routine can continue
  * sending out packets.  The hardware is reset to correct the problem.
  */
-static void netdev_tx_timeout(struct net_device *dev)
+static void netdev_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	static unsigned long last_reset;
 
diff --git a/drivers/net/ethernet/microchip/enc28j60.c b/drivers/net/ethernet/microchip/enc28j60.c
index 0567e4f387a5..09cdc2f2e7ff 100644
--- a/drivers/net/ethernet/microchip/enc28j60.c
+++ b/drivers/net/ethernet/microchip/enc28j60.c
@@ -1325,7 +1325,7 @@ static irqreturn_t enc28j60_irq(int irq, void *dev_id)
 	return IRQ_HANDLED;
 }
 
-static void enc28j60_tx_timeout(struct net_device *ndev)
+static void enc28j60_tx_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	struct enc28j60_net *priv = netdev_priv(ndev);
 
diff --git a/drivers/net/ethernet/microchip/encx24j600.c b/drivers/net/ethernet/microchip/encx24j600.c
index 52c41d11f565..39925e4bf2ec 100644
--- a/drivers/net/ethernet/microchip/encx24j600.c
+++ b/drivers/net/ethernet/microchip/encx24j600.c
@@ -892,7 +892,7 @@ static netdev_tx_t encx24j600_tx(struct sk_buff *skb, struct net_device *dev)
 }
 
 /* Deal with a transmit timeout */
-static void encx24j600_tx_timeout(struct net_device *dev)
+static void encx24j600_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct encx24j600_priv *priv = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/natsemi/natsemi.c b/drivers/net/ethernet/natsemi/natsemi.c
index 1a2634cbbb69..d21d706b83a7 100644
--- a/drivers/net/ethernet/natsemi/natsemi.c
+++ b/drivers/net/ethernet/natsemi/natsemi.c
@@ -612,7 +612,7 @@ static void undo_cable_magic(struct net_device *dev);
 static void check_link(struct net_device *dev);
 static void netdev_timer(struct timer_list *t);
 static void dump_ring(struct net_device *dev);
-static void ns_tx_timeout(struct net_device *dev);
+static void ns_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static int alloc_ring(struct net_device *dev);
 static void refill_rx(struct net_device *dev);
 static void init_ring(struct net_device *dev);
@@ -1881,7 +1881,7 @@ static void dump_ring(struct net_device *dev)
 	}
 }
 
-static void ns_tx_timeout(struct net_device *dev)
+static void ns_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct netdev_private *np = netdev_priv(dev);
 	void __iomem * ioaddr = ns_ioaddr(dev);
diff --git a/drivers/net/ethernet/natsemi/ns83820.c b/drivers/net/ethernet/natsemi/ns83820.c
index 6af9a7eee114..be5f62f06785 100644
--- a/drivers/net/ethernet/natsemi/ns83820.c
+++ b/drivers/net/ethernet/natsemi/ns83820.c
@@ -1549,7 +1549,7 @@ static int ns83820_stop(struct net_device *ndev)
 	return 0;
 }
 
-static void ns83820_tx_timeout(struct net_device *ndev)
+static void ns83820_tx_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	struct ns83820 *dev = PRIV(ndev);
         u32 tx_done_idx;
@@ -1603,7 +1603,7 @@ static void ns83820_tx_watch(struct timer_list *t)
 			ndev->name,
 			dev->tx_done_idx, dev->tx_free_idx,
 			atomic_read(&dev->nr_tx_skbs));
-		ns83820_tx_timeout(ndev);
+		ns83820_tx_timeout(ndev, UINT_MAX);
 	}
 
 	mod_timer(&dev->tx_watchdog, jiffies + 2*HZ);
diff --git a/drivers/net/ethernet/natsemi/sonic.c b/drivers/net/ethernet/natsemi/sonic.c
index b339125b2f09..fdebc8598b22 100644
--- a/drivers/net/ethernet/natsemi/sonic.c
+++ b/drivers/net/ethernet/natsemi/sonic.c
@@ -161,7 +161,7 @@ static int sonic_close(struct net_device *dev)
 	return 0;
 }
 
-static void sonic_tx_timeout(struct net_device *dev)
+static void sonic_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct sonic_local *lp = netdev_priv(dev);
 	int i;
diff --git a/drivers/net/ethernet/natsemi/sonic.h b/drivers/net/ethernet/natsemi/sonic.h
index 2b27f7049acb..f1544481aac1 100644
--- a/drivers/net/ethernet/natsemi/sonic.h
+++ b/drivers/net/ethernet/natsemi/sonic.h
@@ -336,7 +336,7 @@ static int sonic_close(struct net_device *dev);
 static struct net_device_stats *sonic_get_stats(struct net_device *dev);
 static void sonic_multicast_list(struct net_device *dev);
 static int sonic_init(struct net_device *dev);
-static void sonic_tx_timeout(struct net_device *dev);
+static void sonic_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static void sonic_msg_init(struct net_device *dev);
 
 /* Internal inlines for reading/writing DMA buffers.  Note that bus
diff --git a/drivers/net/ethernet/neterion/s2io.c b/drivers/net/ethernet/neterion/s2io.c
index e0b2bf327905..0ec6b8e8b549 100644
--- a/drivers/net/ethernet/neterion/s2io.c
+++ b/drivers/net/ethernet/neterion/s2io.c
@@ -7238,7 +7238,7 @@ out_unlock:
  *  void
  */
 
-static void s2io_tx_watchdog(struct net_device *dev)
+static void s2io_tx_watchdog(struct net_device *dev, unsigned int txqueue)
 {
 	struct s2io_nic *sp = netdev_priv(dev);
 	struct swStat *swstats = &sp->mac_control.stats_info->sw_stat;
diff --git a/drivers/net/ethernet/neterion/s2io.h b/drivers/net/ethernet/neterion/s2io.h
index 0a921f30f98f..6fa3159a977f 100644
--- a/drivers/net/ethernet/neterion/s2io.h
+++ b/drivers/net/ethernet/neterion/s2io.h
@@ -1065,7 +1065,7 @@ static void s2io_txpic_intr_handle(struct s2io_nic *sp);
 static void tx_intr_handler(struct fifo_info *fifo_data);
 static void s2io_handle_errors(void * dev_id);
 
-static void s2io_tx_watchdog(struct net_device *dev);
+static void s2io_tx_watchdog(struct net_device *dev, unsigned int txqueue);
 static void s2io_set_multicast(struct net_device *dev);
 static int rx_osm_handler(struct ring_info *ring_data, struct RxD_t * rxdp);
 static void s2io_link(struct s2io_nic * sp, int link);
diff --git a/drivers/net/ethernet/neterion/vxge/vxge-main.c b/drivers/net/ethernet/neterion/vxge/vxge-main.c
index 1d334f2e0a56..9b63574b6202 100644
--- a/drivers/net/ethernet/neterion/vxge/vxge-main.c
+++ b/drivers/net/ethernet/neterion/vxge/vxge-main.c
@@ -3273,7 +3273,7 @@ static int vxge_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
  * This function is triggered if the Tx Queue is stopped
  * for a pre-defined amount of time when the Interface is still up.
  */
-static void vxge_tx_watchdog(struct net_device *dev)
+static void vxge_tx_watchdog(struct net_device *dev, unsigned int txqueue)
 {
 	struct vxgedev *vdev;
 
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
index bcdcd6de7dea..bd305fc6ed5a 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
@@ -1321,7 +1321,7 @@ nfp_net_tx_ring_reset(struct nfp_net_dp *dp, struct nfp_net_tx_ring *tx_ring)
 	netdev_tx_reset_queue(nd_q);
 }
 
-static void nfp_net_tx_timeout(struct net_device *netdev)
+static void nfp_net_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct nfp_net *nn = netdev_priv(netdev);
 	int i;
diff --git a/drivers/net/ethernet/nvidia/forcedeth.c b/drivers/net/ethernet/nvidia/forcedeth.c
index 6b54cb3b681d..2fc10a36afa4 100644
--- a/drivers/net/ethernet/nvidia/forcedeth.c
+++ b/drivers/net/ethernet/nvidia/forcedeth.c
@@ -2739,7 +2739,7 @@ static int nv_tx_done_optimized(struct net_device *dev, int limit)
  * nv_tx_timeout: dev->tx_timeout function
  * Called with netif_tx_lock held.
  */
-static void nv_tx_timeout(struct net_device *dev)
+static void nv_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct fe_priv *np = netdev_priv(dev);
 	u8 __iomem *base = get_hwbase(dev);
diff --git a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c
index 18e6d87c607b..73ec195fbc30 100644
--- a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c
+++ b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c
@@ -2271,7 +2271,7 @@ static int pch_gbe_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
  * pch_gbe_tx_timeout - Respond to a Tx Hang
  * @netdev:   Network interface device structure
  */
-static void pch_gbe_tx_timeout(struct net_device *netdev)
+static void pch_gbe_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct pch_gbe_adapter *adapter = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/packetengines/hamachi.c b/drivers/net/ethernet/packetengines/hamachi.c
index eee883a2aa8d..70816d2e2990 100644
--- a/drivers/net/ethernet/packetengines/hamachi.c
+++ b/drivers/net/ethernet/packetengines/hamachi.c
@@ -548,7 +548,7 @@ static void mdio_write(struct net_device *dev, int phy_id, int location, int val
 static int hamachi_open(struct net_device *dev);
 static int netdev_ioctl(struct net_device *dev, struct ifreq *rq, int cmd);
 static void hamachi_timer(struct timer_list *t);
-static void hamachi_tx_timeout(struct net_device *dev);
+static void hamachi_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static void hamachi_init_ring(struct net_device *dev);
 static netdev_tx_t hamachi_start_xmit(struct sk_buff *skb,
 				      struct net_device *dev);
@@ -1042,7 +1042,7 @@ static void hamachi_timer(struct timer_list *t)
 	add_timer(&hmp->timer);
 }
 
-static void hamachi_tx_timeout(struct net_device *dev)
+static void hamachi_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	int i;
 	struct hamachi_private *hmp = netdev_priv(dev);
diff --git a/drivers/net/ethernet/packetengines/yellowfin.c b/drivers/net/ethernet/packetengines/yellowfin.c
index 5113ee647090..520779f05e1a 100644
--- a/drivers/net/ethernet/packetengines/yellowfin.c
+++ b/drivers/net/ethernet/packetengines/yellowfin.c
@@ -344,7 +344,7 @@ static void mdio_write(void __iomem *ioaddr, int phy_id, int location, int value
 static int netdev_ioctl(struct net_device *dev, struct ifreq *rq, int cmd);
 static int yellowfin_open(struct net_device *dev);
 static void yellowfin_timer(struct timer_list *t);
-static void yellowfin_tx_timeout(struct net_device *dev);
+static void yellowfin_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static int yellowfin_init_ring(struct net_device *dev);
 static netdev_tx_t yellowfin_start_xmit(struct sk_buff *skb,
 					struct net_device *dev);
@@ -677,7 +677,7 @@ static void yellowfin_timer(struct timer_list *t)
 	add_timer(&yp->timer);
 }
 
-static void yellowfin_tx_timeout(struct net_device *dev)
+static void yellowfin_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct yellowfin_private *yp = netdev_priv(dev);
 	void __iomem *ioaddr = yp->base;
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.c b/drivers/net/ethernet/pensando/ionic/ionic_lif.c
index ef8258713369..a76108a9c7db 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_lif.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.c
@@ -1285,7 +1285,7 @@ static void ionic_tx_timeout_work(struct work_struct *ws)
 	rtnl_unlock();
 }
 
-static void ionic_tx_timeout(struct net_device *netdev)
+static void ionic_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct ionic_lif *lif = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c b/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c
index c692a41e4548..8067ea04d455 100644
--- a/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c
+++ b/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c
@@ -49,7 +49,7 @@ static int netxen_nic_open(struct net_device *netdev);
 static int netxen_nic_close(struct net_device *netdev);
 static netdev_tx_t netxen_nic_xmit_frame(struct sk_buff *,
 					       struct net_device *);
-static void netxen_tx_timeout(struct net_device *netdev);
+static void netxen_tx_timeout(struct net_device *netdev, unsigned int txqueue);
 static void netxen_tx_timeout_task(struct work_struct *work);
 static void netxen_fw_poll_work(struct work_struct *work);
 static void netxen_schedule_work(struct netxen_adapter *adapter,
@@ -2222,7 +2222,7 @@ static void netxen_nic_handle_phy_intr(struct netxen_adapter *adapter)
 	netxen_advert_link_change(adapter, linkup);
 }
 
-static void netxen_tx_timeout(struct net_device *netdev)
+static void netxen_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct netxen_adapter *adapter = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/qlogic/qla3xxx.c b/drivers/net/ethernet/qlogic/qla3xxx.c
index b4b8ba00ee01..bb864765c761 100644
--- a/drivers/net/ethernet/qlogic/qla3xxx.c
+++ b/drivers/net/ethernet/qlogic/qla3xxx.c
@@ -3602,7 +3602,7 @@ static int ql3xxx_set_mac_address(struct net_device *ndev, void *p)
 	return 0;
 }
 
-static void ql3xxx_tx_timeout(struct net_device *ndev)
+static void ql3xxx_tx_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	struct ql3_adapter *qdev = netdev_priv(ndev);
 
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
index c07438db30ba..9dd6cb36f366 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
@@ -56,7 +56,7 @@ static int qlcnic_probe(struct pci_dev *pdev, const struct pci_device_id *ent);
 static void qlcnic_remove(struct pci_dev *pdev);
 static int qlcnic_open(struct net_device *netdev);
 static int qlcnic_close(struct net_device *netdev);
-static void qlcnic_tx_timeout(struct net_device *netdev);
+static void qlcnic_tx_timeout(struct net_device *netdev, unsigned int txqueue);
 static void qlcnic_attach_work(struct work_struct *work);
 static void qlcnic_fwinit_work(struct work_struct *work);
 
@@ -3068,7 +3068,7 @@ static void qlcnic_dump_rings(struct qlcnic_adapter *adapter)
 
 }
 
-static void qlcnic_tx_timeout(struct net_device *netdev)
+static void qlcnic_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct qlcnic_adapter *adapter = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/qualcomm/emac/emac.c b/drivers/net/ethernet/qualcomm/emac/emac.c
index 98f92268cbaa..522fad4cb2cd 100644
--- a/drivers/net/ethernet/qualcomm/emac/emac.c
+++ b/drivers/net/ethernet/qualcomm/emac/emac.c
@@ -282,7 +282,7 @@ static int emac_close(struct net_device *netdev)
 }
 
 /* Respond to a TX hang */
-static void emac_tx_timeout(struct net_device *netdev)
+static void emac_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct emac_adapter *adpt = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/qualcomm/qca_spi.c b/drivers/net/ethernet/qualcomm/qca_spi.c
index baac016f3ec0..5a3b65a6eb4f 100644
--- a/drivers/net/ethernet/qualcomm/qca_spi.c
+++ b/drivers/net/ethernet/qualcomm/qca_spi.c
@@ -785,7 +785,7 @@ qcaspi_netdev_xmit(struct sk_buff *skb, struct net_device *dev)
 }
 
 static void
-qcaspi_netdev_tx_timeout(struct net_device *dev)
+qcaspi_netdev_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct qcaspi *qca = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/qualcomm/qca_uart.c b/drivers/net/ethernet/qualcomm/qca_uart.c
index 0981068504fa..375a844cd27c 100644
--- a/drivers/net/ethernet/qualcomm/qca_uart.c
+++ b/drivers/net/ethernet/qualcomm/qca_uart.c
@@ -248,7 +248,7 @@ out:
 	return NETDEV_TX_OK;
 }
 
-static void qcauart_netdev_tx_timeout(struct net_device *dev)
+static void qcauart_netdev_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct qcauart *qca = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/rdc/r6040.c b/drivers/net/ethernet/rdc/r6040.c
index 274e5b4bc4ac..c23cb61bbd30 100644
--- a/drivers/net/ethernet/rdc/r6040.c
+++ b/drivers/net/ethernet/rdc/r6040.c
@@ -410,7 +410,7 @@ static void r6040_init_mac_regs(struct net_device *dev)
 	iowrite16(TM2TX, ioaddr + MTPR);
 }
 
-static void r6040_tx_timeout(struct net_device *dev)
+static void r6040_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct r6040_private *priv = netdev_priv(dev);
 	void __iomem *ioaddr = priv->base;
diff --git a/drivers/net/ethernet/realtek/8139cp.c b/drivers/net/ethernet/realtek/8139cp.c
index 4f910c4f67b0..60d342f82fb3 100644
--- a/drivers/net/ethernet/realtek/8139cp.c
+++ b/drivers/net/ethernet/realtek/8139cp.c
@@ -1235,7 +1235,7 @@ static int cp_close (struct net_device *dev)
 	return 0;
 }
 
-static void cp_tx_timeout(struct net_device *dev)
+static void cp_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct cp_private *cp = netdev_priv(dev);
 	unsigned long flags;
diff --git a/drivers/net/ethernet/realtek/8139too.c b/drivers/net/ethernet/realtek/8139too.c
index 55d01266e615..5caeb8368eab 100644
--- a/drivers/net/ethernet/realtek/8139too.c
+++ b/drivers/net/ethernet/realtek/8139too.c
@@ -642,7 +642,7 @@ static int mdio_read (struct net_device *dev, int phy_id, int location);
 static void mdio_write (struct net_device *dev, int phy_id, int location,
 			int val);
 static void rtl8139_start_thread(struct rtl8139_private *tp);
-static void rtl8139_tx_timeout (struct net_device *dev);
+static void rtl8139_tx_timeout (struct net_device *dev, unsigned int txqueue);
 static void rtl8139_init_ring (struct net_device *dev);
 static netdev_tx_t rtl8139_start_xmit (struct sk_buff *skb,
 				       struct net_device *dev);
@@ -1700,7 +1700,7 @@ static void rtl8139_tx_timeout_task (struct work_struct *work)
 	spin_unlock_bh(&tp->rx_lock);
 }
 
-static void rtl8139_tx_timeout (struct net_device *dev)
+static void rtl8139_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct rtl8139_private *tp = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/realtek/atp.c b/drivers/net/ethernet/realtek/atp.c
index 58e0ca9093d3..9e3b35c97e63 100644
--- a/drivers/net/ethernet/realtek/atp.c
+++ b/drivers/net/ethernet/realtek/atp.c
@@ -204,7 +204,7 @@ static void net_rx(struct net_device *dev);
 static void read_block(long ioaddr, int length, unsigned char *buffer, int data_mode);
 static int net_close(struct net_device *dev);
 static void set_rx_mode(struct net_device *dev);
-static void tx_timeout(struct net_device *dev);
+static void tx_timeout(struct net_device *dev, unsigned int txqueue);
 
 
 /* A list of all installed ATP devices, for removing the driver module. */
@@ -533,7 +533,7 @@ static void write_packet(long ioaddr, int length, unsigned char *packet, int pad
     outb(Ctrl_HNibWrite | Ctrl_SelData | Ctrl_IRQEN, ioaddr + PAR_CONTROL);
 }
 
-static void tx_timeout(struct net_device *dev)
+static void tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	long ioaddr = dev->base_addr;
 
diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 67a4d5d45e3a..2cfaa2270319 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -5435,7 +5435,7 @@ static void rtl_reset_work(struct rtl8169_private *tp)
 	netif_wake_queue(dev);
 }
 
-static void rtl8169_tx_timeout(struct net_device *dev)
+static void rtl8169_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct rtl8169_private *tp = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c
index 4b13a184bfc7..067ad25553b9 100644
--- a/drivers/net/ethernet/renesas/ravb_main.c
+++ b/drivers/net/ethernet/renesas/ravb_main.c
@@ -1425,7 +1425,7 @@ out_napi_off:
 }
 
 /* Timeout function for Ethernet AVB */
-static void ravb_tx_timeout(struct net_device *ndev)
+static void ravb_tx_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	struct ravb_private *priv = netdev_priv(ndev);
 
diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c
index e19b49c4013e..cdd8ab2eb910 100644
--- a/drivers/net/ethernet/renesas/sh_eth.c
+++ b/drivers/net/ethernet/renesas/sh_eth.c
@@ -2478,7 +2478,7 @@ out_napi_off:
 }
 
 /* Timeout function */
-static void sh_eth_tx_timeout(struct net_device *ndev)
+static void sh_eth_tx_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	struct sh_eth_private *mdp = netdev_priv(ndev);
 	struct sh_eth_rxdesc *rxdesc;
diff --git a/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c b/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c
index c56fcbb37066..cd6e0de48248 100644
--- a/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c
+++ b/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c
@@ -1572,7 +1572,7 @@ static int sxgbe_poll(struct napi_struct *napi, int budget)
  *   netdev structure and arrange for the device to be reset to a sane state
  *   in order to transmit a new packet.
  */
-static void sxgbe_tx_timeout(struct net_device *dev)
+static void sxgbe_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct sxgbe_priv_data *priv = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/seeq/ether3.c b/drivers/net/ethernet/seeq/ether3.c
index 632a7c85964d..128ee7cda1ed 100644
--- a/drivers/net/ethernet/seeq/ether3.c
+++ b/drivers/net/ethernet/seeq/ether3.c
@@ -79,7 +79,7 @@ static netdev_tx_t	ether3_sendpacket(struct sk_buff *skb,
 static irqreturn_t ether3_interrupt (int irq, void *dev_id);
 static int	ether3_close (struct net_device *dev);
 static void	ether3_setmulticastlist (struct net_device *dev);
-static void	ether3_timeout(struct net_device *dev);
+static void	ether3_timeout(struct net_device *dev, unsigned int txqueue);
 
 #define BUS_16		2
 #define BUS_8		1
@@ -450,7 +450,7 @@ static void ether3_setmulticastlist(struct net_device *dev)
 	ether3_outw(priv(dev)->regs.config1 | CFG1_LOCBUFMEM, REG_CONFIG1);
 }
 
-static void ether3_timeout(struct net_device *dev)
+static void ether3_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	unsigned long flags;
 
diff --git a/drivers/net/ethernet/seeq/sgiseeq.c b/drivers/net/ethernet/seeq/sgiseeq.c
index 276c7cae7cee..8507ff242014 100644
--- a/drivers/net/ethernet/seeq/sgiseeq.c
+++ b/drivers/net/ethernet/seeq/sgiseeq.c
@@ -645,7 +645,7 @@ sgiseeq_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	return NETDEV_TX_OK;
 }
 
-static void timeout(struct net_device *dev)
+static void timeout(struct net_device *dev, unsigned int txqueue)
 {
 	printk(KERN_NOTICE "%s: transmit timed out, resetting\n", dev->name);
 	sgiseeq_reset(dev);
diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index 992c773620ec..f358709fab67 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -2395,7 +2395,7 @@ static void efx_net_stats(struct net_device *net_dev,
 }
 
 /* Context: netif_tx_lock held, BHs disabled. */
-static void efx_watchdog(struct net_device *net_dev)
+static void efx_watchdog(struct net_device *net_dev, unsigned int txqueue)
 {
 	struct efx_nic *efx = netdev_priv(net_dev);
 
diff --git a/drivers/net/ethernet/sfc/falcon/efx.c b/drivers/net/ethernet/sfc/falcon/efx.c
index eecc348b1c32..bee4cd9d7135 100644
--- a/drivers/net/ethernet/sfc/falcon/efx.c
+++ b/drivers/net/ethernet/sfc/falcon/efx.c
@@ -2108,7 +2108,7 @@ static void ef4_net_stats(struct net_device *net_dev,
 }
 
 /* Context: netif_tx_lock held, BHs disabled. */
-static void ef4_watchdog(struct net_device *net_dev)
+static void ef4_watchdog(struct net_device *net_dev, unsigned int txqueue)
 {
 	struct ef4_nic *efx = netdev_priv(net_dev);
 
diff --git a/drivers/net/ethernet/sgi/ioc3-eth.c b/drivers/net/ethernet/sgi/ioc3-eth.c
index d242906ae233..06637b03deed 100644
--- a/drivers/net/ethernet/sgi/ioc3-eth.c
+++ b/drivers/net/ethernet/sgi/ioc3-eth.c
@@ -114,7 +114,7 @@ struct ioc3_private {
 static int ioc3_ioctl(struct net_device *dev, struct ifreq *rq, int cmd);
 static void ioc3_set_multicast_list(struct net_device *dev);
 static netdev_tx_t ioc3_start_xmit(struct sk_buff *skb, struct net_device *dev);
-static void ioc3_timeout(struct net_device *dev);
+static void ioc3_timeout(struct net_device *dev, unsigned int txqueue);
 static inline unsigned int ioc3_hash(const unsigned char *addr);
 static void ioc3_start(struct ioc3_private *ip);
 static inline void ioc3_stop(struct ioc3_private *ip);
@@ -1479,7 +1479,7 @@ drop_packet:
 	return NETDEV_TX_OK;
 }
 
-static void ioc3_timeout(struct net_device *dev)
+static void ioc3_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct ioc3_private *ip = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/sgi/meth.c b/drivers/net/ethernet/sgi/meth.c
index 539bc5db989c..0c396ecd3389 100644
--- a/drivers/net/ethernet/sgi/meth.c
+++ b/drivers/net/ethernet/sgi/meth.c
@@ -90,7 +90,7 @@ struct meth_private {
 	spinlock_t meth_lock;
 };
 
-static void meth_tx_timeout(struct net_device *dev);
+static void meth_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static irqreturn_t meth_interrupt(int irq, void *dev_id);
 
 /* global, initialized in ip32-setup.c */
@@ -727,7 +727,7 @@ static netdev_tx_t meth_tx(struct sk_buff *skb, struct net_device *dev)
 /*
  * Deal with a transmit timeout.
  */
-static void meth_tx_timeout(struct net_device *dev)
+static void meth_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct meth_private *priv = netdev_priv(dev);
 	unsigned long flags;
diff --git a/drivers/net/ethernet/silan/sc92031.c b/drivers/net/ethernet/silan/sc92031.c
index c7641a236eb8..cb043eb1bdc1 100644
--- a/drivers/net/ethernet/silan/sc92031.c
+++ b/drivers/net/ethernet/silan/sc92031.c
@@ -1078,7 +1078,7 @@ static void sc92031_set_multicast_list(struct net_device *dev)
 	spin_unlock_bh(&priv->lock);
 }
 
-static void sc92031_tx_timeout(struct net_device *dev)
+static void sc92031_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct sc92031_priv *priv = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/sis/sis190.c b/drivers/net/ethernet/sis/sis190.c
index 5b351beb78cb..5a4b6e3ab38f 100644
--- a/drivers/net/ethernet/sis/sis190.c
+++ b/drivers/net/ethernet/sis/sis190.c
@@ -1538,7 +1538,7 @@ err_out_0:
 	goto out;
 }
 
-static void sis190_tx_timeout(struct net_device *dev)
+static void sis190_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct sis190_private *tp = netdev_priv(dev);
 	void __iomem *ioaddr = tp->mmio_addr;
diff --git a/drivers/net/ethernet/sis/sis900.c b/drivers/net/ethernet/sis/sis900.c
index 85eaccbbbac1..81ed7589e33c 100644
--- a/drivers/net/ethernet/sis/sis900.c
+++ b/drivers/net/ethernet/sis/sis900.c
@@ -222,7 +222,7 @@ static int mdio_read(struct net_device *net_dev, int phy_id, int location);
 static void mdio_write(struct net_device *net_dev, int phy_id, int location, int val);
 static void sis900_timer(struct timer_list *t);
 static void sis900_check_mode (struct net_device *net_dev, struct mii_phy *mii_phy);
-static void sis900_tx_timeout(struct net_device *net_dev);
+static void sis900_tx_timeout(struct net_device *net_dev, unsigned int txqueue);
 static void sis900_init_tx_ring(struct net_device *net_dev);
 static void sis900_init_rx_ring(struct net_device *net_dev);
 static netdev_tx_t sis900_start_xmit(struct sk_buff *skb,
@@ -1537,7 +1537,7 @@ static void sis900_read_mode(struct net_device *net_dev, int *speed, int *duplex
  *	disable interrupts and do some tasks
  */
 
-static void sis900_tx_timeout(struct net_device *net_dev)
+static void sis900_tx_timeout(struct net_device *net_dev, unsigned int txqueue)
 {
 	struct sis900_private *sis_priv = netdev_priv(net_dev);
 	void __iomem *ioaddr = sis_priv->ioaddr;
diff --git a/drivers/net/ethernet/smsc/epic100.c b/drivers/net/ethernet/smsc/epic100.c
index be47d864f8b9..912760e8514c 100644
--- a/drivers/net/ethernet/smsc/epic100.c
+++ b/drivers/net/ethernet/smsc/epic100.c
@@ -291,7 +291,7 @@ static int mdio_read(struct net_device *dev, int phy_id, int location);
 static void mdio_write(struct net_device *dev, int phy_id, int loc, int val);
 static void epic_restart(struct net_device *dev);
 static void epic_timer(struct timer_list *t);
-static void epic_tx_timeout(struct net_device *dev);
+static void epic_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static void epic_init_ring(struct net_device *dev);
 static netdev_tx_t epic_start_xmit(struct sk_buff *skb,
 				   struct net_device *dev);
@@ -861,7 +861,7 @@ static void epic_timer(struct timer_list *t)
 	add_timer(&ep->timer);
 }
 
-static void epic_tx_timeout(struct net_device *dev)
+static void epic_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct epic_private *ep = netdev_priv(dev);
 	void __iomem *ioaddr = ep->ioaddr;
diff --git a/drivers/net/ethernet/smsc/smc911x.c b/drivers/net/ethernet/smsc/smc911x.c
index 7b65e79d6ae9..186c0bddbe5f 100644
--- a/drivers/net/ethernet/smsc/smc911x.c
+++ b/drivers/net/ethernet/smsc/smc911x.c
@@ -1245,7 +1245,7 @@ static void smc911x_poll_controller(struct net_device *dev)
 #endif
 
 /* Our watchdog timed out. Called by the networking layer */
-static void smc911x_timeout(struct net_device *dev)
+static void smc911x_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct smc911x_local *lp = netdev_priv(dev);
 	int status, mask;
diff --git a/drivers/net/ethernet/smsc/smc9194.c b/drivers/net/ethernet/smsc/smc9194.c
index d3bb2ba51f40..4b2330deed47 100644
--- a/drivers/net/ethernet/smsc/smc9194.c
+++ b/drivers/net/ethernet/smsc/smc9194.c
@@ -216,7 +216,7 @@ static int smc_open(struct net_device *dev);
 /*
  . Our watchdog timed out. Called by the networking layer
 */
-static void smc_timeout(struct net_device *dev);
+static void smc_timeout(struct net_device *dev, unsigned int txqueue);
 
 /*
  . This is called by the kernel in response to 'ifconfig ethX down'.  It
@@ -1094,7 +1094,7 @@ static int smc_open(struct net_device *dev)
  .--------------------------------------------------------
 */
 
-static void smc_timeout(struct net_device *dev)
+static void smc_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	/* If we get here, some higher level has decided we are broken.
 	   There should really be a "kick me" function call instead. */
diff --git a/drivers/net/ethernet/smsc/smc91c92_cs.c b/drivers/net/ethernet/smsc/smc91c92_cs.c
index a55f430f6a7b..f2a50eb3c1e0 100644
--- a/drivers/net/ethernet/smsc/smc91c92_cs.c
+++ b/drivers/net/ethernet/smsc/smc91c92_cs.c
@@ -271,7 +271,7 @@ static void smc91c92_release(struct pcmcia_device *link);
 static int smc_open(struct net_device *dev);
 static int smc_close(struct net_device *dev);
 static int smc_ioctl(struct net_device *dev, struct ifreq *rq, int cmd);
-static void smc_tx_timeout(struct net_device *dev);
+static void smc_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static netdev_tx_t smc_start_xmit(struct sk_buff *skb,
 					struct net_device *dev);
 static irqreturn_t smc_interrupt(int irq, void *dev_id);
@@ -1178,7 +1178,7 @@ static void smc_hardware_send_packet(struct net_device * dev)
 
 /*====================================================================*/
 
-static void smc_tx_timeout(struct net_device *dev)
+static void smc_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
     struct smc_private *smc = netdev_priv(dev);
     unsigned int ioaddr = dev->base_addr;
diff --git a/drivers/net/ethernet/smsc/smc91x.c b/drivers/net/ethernet/smsc/smc91x.c
index 3a6761131f4c..90410f9d3b1a 100644
--- a/drivers/net/ethernet/smsc/smc91x.c
+++ b/drivers/net/ethernet/smsc/smc91x.c
@@ -1321,7 +1321,7 @@ static void smc_poll_controller(struct net_device *dev)
 #endif
 
 /* Our watchdog timed out. Called by the networking layer */
-static void smc_timeout(struct net_device *dev)
+static void smc_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct smc_local *lp = netdev_priv(dev);
 	void __iomem *ioaddr = lp->base;
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index bbc65bd332a8..da80866d0371 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -3792,7 +3792,7 @@ static int stmmac_napi_poll_tx(struct napi_struct *napi, int budget)
  *   netdev structure and arrange for the device to be reset to a sane state
  *   in order to transmit a new packet.
  */
-static void stmmac_tx_timeout(struct net_device *dev)
+static void stmmac_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct stmmac_priv *priv = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/sun/cassini.c b/drivers/net/ethernet/sun/cassini.c
index c91876f8c536..6ec9163e232c 100644
--- a/drivers/net/ethernet/sun/cassini.c
+++ b/drivers/net/ethernet/sun/cassini.c
@@ -2666,7 +2666,7 @@ static void cas_netpoll(struct net_device *dev)
 }
 #endif
 
-static void cas_tx_timeout(struct net_device *dev)
+static void cas_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct cas *cp = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/sun/niu.c b/drivers/net/ethernet/sun/niu.c
index f5fd1f3c07cc..9a5004f674c7 100644
--- a/drivers/net/ethernet/sun/niu.c
+++ b/drivers/net/ethernet/sun/niu.c
@@ -6517,7 +6517,7 @@ static void niu_reset_task(struct work_struct *work)
 	spin_unlock_irqrestore(&np->lock, flags);
 }
 
-static void niu_tx_timeout(struct net_device *dev)
+static void niu_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct niu *np = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/sun/sunbmac.c b/drivers/net/ethernet/sun/sunbmac.c
index e9b757b03b56..c5add0b45eed 100644
--- a/drivers/net/ethernet/sun/sunbmac.c
+++ b/drivers/net/ethernet/sun/sunbmac.c
@@ -941,7 +941,7 @@ static int bigmac_close(struct net_device *dev)
 	return 0;
 }
 
-static void bigmac_tx_timeout(struct net_device *dev)
+static void bigmac_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct bigmac *bp = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/sun/sungem.c b/drivers/net/ethernet/sun/sungem.c
index 3e7631160384..8358064fbd48 100644
--- a/drivers/net/ethernet/sun/sungem.c
+++ b/drivers/net/ethernet/sun/sungem.c
@@ -970,7 +970,7 @@ static void gem_poll_controller(struct net_device *dev)
 }
 #endif
 
-static void gem_tx_timeout(struct net_device *dev)
+static void gem_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct gem *gp = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/sun/sunhme.c b/drivers/net/ethernet/sun/sunhme.c
index d007dfeba5c3..f0fe7bb2a750 100644
--- a/drivers/net/ethernet/sun/sunhme.c
+++ b/drivers/net/ethernet/sun/sunhme.c
@@ -2246,7 +2246,7 @@ static int happy_meal_close(struct net_device *dev)
 #define SXD(x)
 #endif
 
-static void happy_meal_tx_timeout(struct net_device *dev)
+static void happy_meal_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct happy_meal *hp = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/sun/sunqe.c b/drivers/net/ethernet/sun/sunqe.c
index 1468fa0a54e9..2102b95ec347 100644
--- a/drivers/net/ethernet/sun/sunqe.c
+++ b/drivers/net/ethernet/sun/sunqe.c
@@ -544,7 +544,7 @@ static void qe_tx_reclaim(struct sunqe *qep)
 	qep->tx_old = elem;
 }
 
-static void qe_tx_timeout(struct net_device *dev)
+static void qe_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct sunqe *qep = netdev_priv(dev);
 	int tx_full;
diff --git a/drivers/net/ethernet/sun/sunvnet_common.c b/drivers/net/ethernet/sun/sunvnet_common.c
index 8b94d9ad9e2b..a601a306f9a5 100644
--- a/drivers/net/ethernet/sun/sunvnet_common.c
+++ b/drivers/net/ethernet/sun/sunvnet_common.c
@@ -1539,7 +1539,7 @@ out_dropped:
 }
 EXPORT_SYMBOL_GPL(sunvnet_start_xmit_common);
 
-void sunvnet_tx_timeout_common(struct net_device *dev)
+void sunvnet_tx_timeout_common(struct net_device *dev, unsigned int txqueue)
 {
 	/* XXX Implement me XXX */
 }
diff --git a/drivers/net/ethernet/sun/sunvnet_common.h b/drivers/net/ethernet/sun/sunvnet_common.h
index 2b808d2482d6..5416a3cb9e7d 100644
--- a/drivers/net/ethernet/sun/sunvnet_common.h
+++ b/drivers/net/ethernet/sun/sunvnet_common.h
@@ -135,7 +135,7 @@ int sunvnet_open_common(struct net_device *dev);
 int sunvnet_close_common(struct net_device *dev);
 void sunvnet_set_rx_mode_common(struct net_device *dev, struct vnet *vp);
 int sunvnet_set_mac_addr_common(struct net_device *dev, void *p);
-void sunvnet_tx_timeout_common(struct net_device *dev);
+void sunvnet_tx_timeout_common(struct net_device *dev, unsigned int txqueue);
 netdev_tx_t
 sunvnet_start_xmit_common(struct sk_buff *skb, struct net_device *dev,
 			  struct vnet_port *(*vnet_tx_port)
diff --git a/drivers/net/ethernet/synopsys/dwc-xlgmac-net.c b/drivers/net/ethernet/synopsys/dwc-xlgmac-net.c
index a1f5a1e61040..07046a2370b3 100644
--- a/drivers/net/ethernet/synopsys/dwc-xlgmac-net.c
+++ b/drivers/net/ethernet/synopsys/dwc-xlgmac-net.c
@@ -689,7 +689,7 @@ static int xlgmac_close(struct net_device *netdev)
 	return 0;
 }
 
-static void xlgmac_tx_timeout(struct net_device *netdev)
+static void xlgmac_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct xlgmac_pdata *pdata = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/ti/cpmac.c b/drivers/net/ethernet/ti/cpmac.c
index 3a655a4dc10e..5e1b8292cd3f 100644
--- a/drivers/net/ethernet/ti/cpmac.c
+++ b/drivers/net/ethernet/ti/cpmac.c
@@ -797,7 +797,7 @@ static irqreturn_t cpmac_irq(int irq, void *dev_id)
 	return IRQ_HANDLED;
 }
 
-static void cpmac_tx_timeout(struct net_device *dev)
+static void cpmac_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct cpmac_priv *priv = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/ti/cpsw_priv.c b/drivers/net/ethernet/ti/cpsw_priv.c
index 707d5eb480ce..97a058ca60ac 100644
--- a/drivers/net/ethernet/ti/cpsw_priv.c
+++ b/drivers/net/ethernet/ti/cpsw_priv.c
@@ -272,7 +272,7 @@ void soft_reset(const char *module, void __iomem *reg)
 	WARN(readl_relaxed(reg) & 1, "failed to soft-reset %s\n", module);
 }
 
-void cpsw_ndo_tx_timeout(struct net_device *ndev)
+void cpsw_ndo_tx_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	struct cpsw_priv *priv = netdev_priv(ndev);
 	struct cpsw_common *cpsw = priv->cpsw;
diff --git a/drivers/net/ethernet/ti/cpsw_priv.h b/drivers/net/ethernet/ti/cpsw_priv.h
index bc726356a72c..b8d7b924ee3d 100644
--- a/drivers/net/ethernet/ti/cpsw_priv.h
+++ b/drivers/net/ethernet/ti/cpsw_priv.h
@@ -449,7 +449,7 @@ int cpsw_rx_poll(struct napi_struct *napi_rx, int budget);
 void cpsw_rx_vlan_encap(struct sk_buff *skb);
 void soft_reset(const char *module, void __iomem *reg);
 void cpsw_set_slave_mac(struct cpsw_slave *slave, struct cpsw_priv *priv);
-void cpsw_ndo_tx_timeout(struct net_device *ndev);
+void cpsw_ndo_tx_timeout(struct net_device *ndev, unsigned int txqueue);
 int cpsw_need_resplit(struct cpsw_common *cpsw);
 int cpsw_ndo_ioctl(struct net_device *dev, struct ifreq *req, int cmd);
 int cpsw_ndo_set_tx_maxrate(struct net_device *ndev, int queue, u32 rate);
diff --git a/drivers/net/ethernet/ti/davinci_emac.c b/drivers/net/ethernet/ti/davinci_emac.c
index ae27be85e363..75d4e16c692b 100644
--- a/drivers/net/ethernet/ti/davinci_emac.c
+++ b/drivers/net/ethernet/ti/davinci_emac.c
@@ -983,7 +983,7 @@ fail_tx:
  * error and re-initialize the TX channel for hardware operation
  *
  */
-static void emac_dev_tx_timeout(struct net_device *ndev)
+static void emac_dev_tx_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	struct emac_priv *priv = netdev_priv(ndev);
 	struct device *emac_dev = &ndev->dev;
diff --git a/drivers/net/ethernet/ti/netcp_core.c b/drivers/net/ethernet/ti/netcp_core.c
index 1b2702f74455..432645e86495 100644
--- a/drivers/net/ethernet/ti/netcp_core.c
+++ b/drivers/net/ethernet/ti/netcp_core.c
@@ -1811,7 +1811,7 @@ out:
 	return (ret == 0) ? 0 : err;
 }
 
-static void netcp_ndo_tx_timeout(struct net_device *ndev)
+static void netcp_ndo_tx_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	struct netcp_intf *netcp = netdev_priv(ndev);
 	unsigned int descs = knav_pool_count(netcp->tx_pool);
diff --git a/drivers/net/ethernet/ti/tlan.c b/drivers/net/ethernet/ti/tlan.c
index 78f0f2d59e22..ad465202980a 100644
--- a/drivers/net/ethernet/ti/tlan.c
+++ b/drivers/net/ethernet/ti/tlan.c
@@ -161,7 +161,7 @@ static void	tlan_set_multicast_list(struct net_device *);
 static int	tlan_ioctl(struct net_device *dev, struct ifreq *rq, int cmd);
 static int      tlan_probe1(struct pci_dev *pdev, long ioaddr,
 			    int irq, int rev, const struct pci_device_id *ent);
-static void	tlan_tx_timeout(struct net_device *dev);
+static void	tlan_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static void	tlan_tx_timeout_work(struct work_struct *work);
 static int	tlan_init_one(struct pci_dev *pdev,
 			      const struct pci_device_id *ent);
@@ -997,7 +997,7 @@ static int tlan_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
  *
  **************************************************************/
 
-static void tlan_tx_timeout(struct net_device *dev)
+static void tlan_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 
 	TLAN_DBG(TLAN_DEBUG_GNRL, "%s: Transmit timed out.\n", dev->name);
@@ -1028,7 +1028,7 @@ static void tlan_tx_timeout_work(struct work_struct *work)
 	struct tlan_priv	*priv =
 		container_of(work, struct tlan_priv, tlan_tqueue);
 
-	tlan_tx_timeout(priv->dev);
+	tlan_tx_timeout(priv->dev, UINT_MAX);
 }
 
 
diff --git a/drivers/net/ethernet/toshiba/ps3_gelic_net.c b/drivers/net/ethernet/toshiba/ps3_gelic_net.c
index 9d9f8acb7ee3..070dd6fa9401 100644
--- a/drivers/net/ethernet/toshiba/ps3_gelic_net.c
+++ b/drivers/net/ethernet/toshiba/ps3_gelic_net.c
@@ -1405,7 +1405,7 @@ out:
  *
  * called, if tx hangs. Schedules a task that resets the interface
  */
-void gelic_net_tx_timeout(struct net_device *netdev)
+void gelic_net_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct gelic_card *card;
 
diff --git a/drivers/net/ethernet/toshiba/ps3_gelic_net.h b/drivers/net/ethernet/toshiba/ps3_gelic_net.h
index 051033580f0a..805903dbddcc 100644
--- a/drivers/net/ethernet/toshiba/ps3_gelic_net.h
+++ b/drivers/net/ethernet/toshiba/ps3_gelic_net.h
@@ -359,7 +359,7 @@ int gelic_net_open(struct net_device *netdev);
 int gelic_net_stop(struct net_device *netdev);
 netdev_tx_t gelic_net_xmit(struct sk_buff *skb, struct net_device *netdev);
 void gelic_net_set_multi(struct net_device *netdev);
-void gelic_net_tx_timeout(struct net_device *netdev);
+void gelic_net_tx_timeout(struct net_device *netdev, unsigned int txqueue);
 int gelic_net_setup_netdev(struct net_device *netdev, struct gelic_card *card);
 
 /* shared ethtool ops */
diff --git a/drivers/net/ethernet/toshiba/spider_net.c b/drivers/net/ethernet/toshiba/spider_net.c
index 538e70810d3d..6576271642c1 100644
--- a/drivers/net/ethernet/toshiba/spider_net.c
+++ b/drivers/net/ethernet/toshiba/spider_net.c
@@ -2180,7 +2180,7 @@ out:
  * called, if tx hangs. Schedules a task that resets the interface
  */
 static void
-spider_net_tx_timeout(struct net_device *netdev)
+spider_net_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct spider_net_card *card;
 
diff --git a/drivers/net/ethernet/toshiba/tc35815.c b/drivers/net/ethernet/toshiba/tc35815.c
index 12466a72cefc..708de826200e 100644
--- a/drivers/net/ethernet/toshiba/tc35815.c
+++ b/drivers/net/ethernet/toshiba/tc35815.c
@@ -483,7 +483,7 @@ static void	tc35815_txdone(struct net_device *dev);
 static int	tc35815_close(struct net_device *dev);
 static struct	net_device_stats *tc35815_get_stats(struct net_device *dev);
 static void	tc35815_set_multicast_list(struct net_device *dev);
-static void	tc35815_tx_timeout(struct net_device *dev);
+static void	tc35815_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static int	tc35815_ioctl(struct net_device *dev, struct ifreq *rq, int cmd);
 #ifdef CONFIG_NET_POLL_CONTROLLER
 static void	tc35815_poll_controller(struct net_device *dev);
@@ -1189,7 +1189,7 @@ static void tc35815_schedule_restart(struct net_device *dev)
 	spin_unlock_irqrestore(&lp->lock, flags);
 }
 
-static void tc35815_tx_timeout(struct net_device *dev)
+static void tc35815_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct tc35815_regs __iomem *tr =
 		(struct tc35815_regs __iomem *)dev->base_addr;
diff --git a/drivers/net/ethernet/via/via-rhine.c b/drivers/net/ethernet/via/via-rhine.c
index ed12dbd156f0..803247d51fe9 100644
--- a/drivers/net/ethernet/via/via-rhine.c
+++ b/drivers/net/ethernet/via/via-rhine.c
@@ -506,7 +506,7 @@ static void mdio_write(struct net_device *dev, int phy_id, int location, int val
 static int  rhine_open(struct net_device *dev);
 static void rhine_reset_task(struct work_struct *work);
 static void rhine_slow_event_task(struct work_struct *work);
-static void rhine_tx_timeout(struct net_device *dev);
+static void rhine_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static netdev_tx_t rhine_start_tx(struct sk_buff *skb,
 				  struct net_device *dev);
 static irqreturn_t rhine_interrupt(int irq, void *dev_instance);
@@ -1761,7 +1761,7 @@ out_unlock:
 	mutex_unlock(&rp->task_lock);
 }
 
-static void rhine_tx_timeout(struct net_device *dev)
+static void rhine_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct rhine_private *rp = netdev_priv(dev);
 	void __iomem *ioaddr = rp->base;
diff --git a/drivers/net/ethernet/wiznet/w5100.c b/drivers/net/ethernet/wiznet/w5100.c
index bede1ff289c5..c0d181a7f83a 100644
--- a/drivers/net/ethernet/wiznet/w5100.c
+++ b/drivers/net/ethernet/wiznet/w5100.c
@@ -790,7 +790,7 @@ static void w5100_restart_work(struct work_struct *work)
 	w5100_restart(priv->ndev);
 }
 
-static void w5100_tx_timeout(struct net_device *ndev)
+static void w5100_tx_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	struct w5100_priv *priv = netdev_priv(ndev);
 
diff --git a/drivers/net/ethernet/wiznet/w5300.c b/drivers/net/ethernet/wiznet/w5300.c
index 6ba2747779ce..46aae30c4636 100644
--- a/drivers/net/ethernet/wiznet/w5300.c
+++ b/drivers/net/ethernet/wiznet/w5300.c
@@ -341,7 +341,7 @@ static void w5300_get_regs(struct net_device *ndev,
 	}
 }
 
-static void w5300_tx_timeout(struct net_device *ndev)
+static void w5300_tx_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	struct w5300_priv *priv = netdev_priv(ndev);
 
diff --git a/drivers/net/ethernet/xilinx/xilinx_emaclite.c b/drivers/net/ethernet/xilinx/xilinx_emaclite.c
index 0de52e70abcc..0c26f5bcc523 100644
--- a/drivers/net/ethernet/xilinx/xilinx_emaclite.c
+++ b/drivers/net/ethernet/xilinx/xilinx_emaclite.c
@@ -521,7 +521,7 @@ static int xemaclite_set_mac_address(struct net_device *dev, void *address)
  *
  * This function is called when Tx time out occurs for Emaclite device.
  */
-static void xemaclite_tx_timeout(struct net_device *dev)
+static void xemaclite_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct net_local *lp = netdev_priv(dev);
 	unsigned long flags;
diff --git a/drivers/net/ethernet/xircom/xirc2ps_cs.c b/drivers/net/ethernet/xircom/xirc2ps_cs.c
index fd5288ff53b5..480ab7251515 100644
--- a/drivers/net/ethernet/xircom/xirc2ps_cs.c
+++ b/drivers/net/ethernet/xircom/xirc2ps_cs.c
@@ -288,7 +288,7 @@ struct local_info {
  */
 static netdev_tx_t do_start_xmit(struct sk_buff *skb,
 				       struct net_device *dev);
-static void xirc_tx_timeout(struct net_device *dev);
+static void xirc_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static void xirc2ps_tx_timeout_task(struct work_struct *work);
 static void set_addresses(struct net_device *dev);
 static void set_multicast_list(struct net_device *dev);
@@ -1203,7 +1203,7 @@ xirc2ps_tx_timeout_task(struct work_struct *work)
 }
 
 static void
-xirc_tx_timeout(struct net_device *dev)
+xirc_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
     struct local_info *lp = netdev_priv(dev);
     dev->stats.tx_errors++;
diff --git a/drivers/net/fjes/fjes_main.c b/drivers/net/fjes/fjes_main.c
index b517c1af9de0..309a74da2ec3 100644
--- a/drivers/net/fjes/fjes_main.c
+++ b/drivers/net/fjes/fjes_main.c
@@ -48,7 +48,7 @@ static void fjes_get_stats64(struct net_device *, struct rtnl_link_stats64 *);
 static int fjes_change_mtu(struct net_device *, int);
 static int fjes_vlan_rx_add_vid(struct net_device *, __be16 proto, u16);
 static int fjes_vlan_rx_kill_vid(struct net_device *, __be16 proto, u16);
-static void fjes_tx_retry(struct net_device *);
+static void fjes_tx_retry(struct net_device *, unsigned int txqueue);
 
 static int fjes_acpi_add(struct acpi_device *);
 static int fjes_acpi_remove(struct acpi_device *);
@@ -792,7 +792,7 @@ fjes_xmit_frame(struct sk_buff *skb, struct net_device *netdev)
 	return ret;
 }
 
-static void fjes_tx_retry(struct net_device *netdev)
+static void fjes_tx_retry(struct net_device *netdev, unsigned int txqueue)
 {
 	struct netdev_queue *queue = netdev_get_tx_queue(netdev, 0);
 
diff --git a/drivers/net/slip/slip.c b/drivers/net/slip/slip.c
index 2a91c192659f..317d3a8df316 100644
--- a/drivers/net/slip/slip.c
+++ b/drivers/net/slip/slip.c
@@ -457,7 +457,7 @@ static void slip_write_wakeup(struct tty_struct *tty)
 	schedule_work(&sl->tx_work);
 }
 
-static void sl_tx_timeout(struct net_device *dev)
+static void sl_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct slip *sl = netdev_priv(dev);
 
diff --git a/drivers/net/usb/catc.c b/drivers/net/usb/catc.c
index 1e58702c737f..d387bc7ac1b6 100644
--- a/drivers/net/usb/catc.c
+++ b/drivers/net/usb/catc.c
@@ -447,7 +447,7 @@ static netdev_tx_t catc_start_xmit(struct sk_buff *skb,
 	return NETDEV_TX_OK;
 }
 
-static void catc_tx_timeout(struct net_device *netdev)
+static void catc_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct catc *catc = netdev_priv(netdev);
 
diff --git a/drivers/net/usb/hso.c b/drivers/net/usb/hso.c
index ca827802f291..417e42c9fd03 100644
--- a/drivers/net/usb/hso.c
+++ b/drivers/net/usb/hso.c
@@ -820,7 +820,7 @@ static const struct ethtool_ops ops = {
 };
 
 /* called when a packet did not ack after watchdogtimeout */
-static void hso_net_tx_timeout(struct net_device *net)
+static void hso_net_tx_timeout(struct net_device *net, unsigned int txqueue)
 {
 	struct hso_net *odev = netdev_priv(net);
 
diff --git a/drivers/net/usb/ipheth.c b/drivers/net/usb/ipheth.c
index 8c01fbf68a89..c792d65dd7b4 100644
--- a/drivers/net/usb/ipheth.c
+++ b/drivers/net/usb/ipheth.c
@@ -400,7 +400,7 @@ static int ipheth_tx(struct sk_buff *skb, struct net_device *net)
 	return NETDEV_TX_OK;
 }
 
-static void ipheth_tx_timeout(struct net_device *net)
+static void ipheth_tx_timeout(struct net_device *net, unsigned int txqueue)
 {
 	struct ipheth_device *dev = netdev_priv(net);
 
diff --git a/drivers/net/usb/kaweth.c b/drivers/net/usb/kaweth.c
index 8e210ba4a313..ed01dc964c99 100644
--- a/drivers/net/usb/kaweth.c
+++ b/drivers/net/usb/kaweth.c
@@ -894,7 +894,7 @@ static void kaweth_async_set_rx_mode(struct kaweth_device *kaweth)
 /****************************************************************
  *     kaweth_tx_timeout
  ****************************************************************/
-static void kaweth_tx_timeout(struct net_device *net)
+static void kaweth_tx_timeout(struct net_device *net, unsigned int txqueue)
 {
 	struct kaweth_device *kaweth = netdev_priv(net);
 
diff --git a/drivers/net/usb/lan78xx.c b/drivers/net/usb/lan78xx.c
index cf1f3f0a4b9b..0c8b9363366b 100644
--- a/drivers/net/usb/lan78xx.c
+++ b/drivers/net/usb/lan78xx.c
@@ -3662,7 +3662,7 @@ static void lan78xx_disconnect(struct usb_interface *intf)
 	usb_put_dev(udev);
 }
 
-static void lan78xx_tx_timeout(struct net_device *net)
+static void lan78xx_tx_timeout(struct net_device *net, unsigned int txqueue)
 {
 	struct lan78xx_net *dev = netdev_priv(net);
 
diff --git a/drivers/net/usb/pegasus.c b/drivers/net/usb/pegasus.c
index f7d117d80cfb..8783e2ab3ec0 100644
--- a/drivers/net/usb/pegasus.c
+++ b/drivers/net/usb/pegasus.c
@@ -693,7 +693,7 @@ static void intr_callback(struct urb *urb)
 			  "can't resubmit interrupt urb, %d\n", res);
 }
 
-static void pegasus_tx_timeout(struct net_device *net)
+static void pegasus_tx_timeout(struct net_device *net, unsigned int txqueue)
 {
 	pegasus_t *pegasus = netdev_priv(net);
 	netif_warn(pegasus, timer, net, "tx timeout\n");
diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c
index c5ebf35d2488..9ec1da429514 100644
--- a/drivers/net/usb/r8152.c
+++ b/drivers/net/usb/r8152.c
@@ -2507,7 +2507,7 @@ static void rtl_drop_queued_tx(struct r8152 *tp)
 	}
 }
 
-static void rtl8152_tx_timeout(struct net_device *netdev)
+static void rtl8152_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct r8152 *tp = netdev_priv(netdev);
 
diff --git a/drivers/net/usb/rtl8150.c b/drivers/net/usb/rtl8150.c
index 13e51ccf0214..e7c630d37589 100644
--- a/drivers/net/usb/rtl8150.c
+++ b/drivers/net/usb/rtl8150.c
@@ -655,7 +655,7 @@ static void disable_net_traffic(rtl8150_t * dev)
 	set_registers(dev, CR, 1, &cr);
 }
 
-static void rtl8150_tx_timeout(struct net_device *netdev)
+static void rtl8150_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	rtl8150_t *dev = netdev_priv(netdev);
 	dev_warn(&netdev->dev, "Tx timeout.\n");
diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c
index 30e511c2c8d0..bc88923db369 100644
--- a/drivers/net/usb/usbnet.c
+++ b/drivers/net/usb/usbnet.c
@@ -1293,7 +1293,7 @@ static void tx_complete (struct urb *urb)
 
 /*-------------------------------------------------------------------------*/
 
-void usbnet_tx_timeout (struct net_device *net)
+void usbnet_tx_timeout (struct net_device *net, unsigned int txqueue)
 {
 	struct usbnet		*dev = netdev_priv(net);
 
diff --git a/drivers/net/vmxnet3/vmxnet3_drv.c b/drivers/net/vmxnet3/vmxnet3_drv.c
index 216acf37ca7c..18f152fa0068 100644
--- a/drivers/net/vmxnet3/vmxnet3_drv.c
+++ b/drivers/net/vmxnet3/vmxnet3_drv.c
@@ -3198,7 +3198,7 @@ vmxnet3_free_intr_resources(struct vmxnet3_adapter *adapter)
 
 
 static void
-vmxnet3_tx_timeout(struct net_device *netdev)
+vmxnet3_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct vmxnet3_adapter *adapter = netdev_priv(netdev);
 	adapter->tx_timeout_count++;
diff --git a/drivers/net/wan/cosa.c b/drivers/net/wan/cosa.c
index af539151d663..5d6532ad6b78 100644
--- a/drivers/net/wan/cosa.c
+++ b/drivers/net/wan/cosa.c
@@ -268,7 +268,7 @@ static int cosa_net_attach(struct net_device *dev, unsigned short encoding,
 			   unsigned short parity);
 static int cosa_net_open(struct net_device *d);
 static int cosa_net_close(struct net_device *d);
-static void cosa_net_timeout(struct net_device *d);
+static void cosa_net_timeout(struct net_device *d, unsigned int txqueue);
 static netdev_tx_t cosa_net_tx(struct sk_buff *skb, struct net_device *d);
 static char *cosa_net_setup_rx(struct channel_data *channel, int size);
 static int cosa_net_rx_done(struct channel_data *channel);
@@ -670,7 +670,7 @@ static netdev_tx_t cosa_net_tx(struct sk_buff *skb,
 	return NETDEV_TX_OK;
 }
 
-static void cosa_net_timeout(struct net_device *dev)
+static void cosa_net_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct channel_data *chan = dev_to_chan(dev);
 
diff --git a/drivers/net/wan/farsync.c b/drivers/net/wan/farsync.c
index 1901ec7948d8..7916efce7188 100644
--- a/drivers/net/wan/farsync.c
+++ b/drivers/net/wan/farsync.c
@@ -2239,7 +2239,7 @@ fst_attach(struct net_device *dev, unsigned short encoding, unsigned short parit
 }
 
 static void
-fst_tx_timeout(struct net_device *dev)
+fst_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct fst_port_info *port;
 	struct fst_card_info *card;
diff --git a/drivers/net/wan/fsl_ucc_hdlc.c b/drivers/net/wan/fsl_ucc_hdlc.c
index ca0f3be2b6bf..308384756e6f 100644
--- a/drivers/net/wan/fsl_ucc_hdlc.c
+++ b/drivers/net/wan/fsl_ucc_hdlc.c
@@ -1039,7 +1039,7 @@ static const struct dev_pm_ops uhdlc_pm_ops = {
 #define HDLC_PM_OPS NULL
 
 #endif
-static void uhdlc_tx_timeout(struct net_device *ndev)
+static void uhdlc_tx_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	netdev_err(ndev, "%s\n", __func__);
 }
diff --git a/drivers/net/wan/lmc/lmc_main.c b/drivers/net/wan/lmc/lmc_main.c
index 0e6a51525d91..a20f467ca48a 100644
--- a/drivers/net/wan/lmc/lmc_main.c
+++ b/drivers/net/wan/lmc/lmc_main.c
@@ -99,7 +99,7 @@ static int lmc_ifdown(struct net_device * const);
 static void lmc_watchdog(struct timer_list *t);
 static void lmc_reset(lmc_softc_t * const sc);
 static void lmc_dec_reset(lmc_softc_t * const sc);
-static void lmc_driver_timeout(struct net_device *dev);
+static void lmc_driver_timeout(struct net_device *dev, unsigned int txqueue);
 
 /*
  * linux reserves 16 device specific IOCTLs.  We call them
@@ -2044,7 +2044,7 @@ static void lmc_initcsrs(lmc_softc_t * const sc, lmc_csrptr_t csr_base, /*fold00
     lmc_trace(sc->lmc_device, "lmc_initcsrs out");
 }
 
-static void lmc_driver_timeout(struct net_device *dev)
+static void lmc_driver_timeout(struct net_device *dev, unsigned int txqueue)
 {
     lmc_softc_t *sc = dev_to_sc(dev);
     u32 csr6;
diff --git a/drivers/net/wan/x25_asy.c b/drivers/net/wan/x25_asy.c
index 914be5847386..69773d228ec1 100644
--- a/drivers/net/wan/x25_asy.c
+++ b/drivers/net/wan/x25_asy.c
@@ -276,7 +276,7 @@ static void x25_asy_write_wakeup(struct tty_struct *tty)
 	sl->xhead += actual;
 }
 
-static void x25_asy_timeout(struct net_device *dev)
+static void x25_asy_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct x25_asy *sl = netdev_priv(dev);
 
diff --git a/drivers/net/wimax/i2400m/netdev.c b/drivers/net/wimax/i2400m/netdev.c
index a5db3c06b646..a7fcbceb6e6b 100644
--- a/drivers/net/wimax/i2400m/netdev.c
+++ b/drivers/net/wimax/i2400m/netdev.c
@@ -380,7 +380,7 @@ drop:
 
 
 static
-void i2400m_tx_timeout(struct net_device *net_dev)
+void i2400m_tx_timeout(struct net_device *net_dev, unsigned int txqueue)
 {
 	/*
 	 * We might want to kick the device
diff --git a/drivers/net/wireless/intel/ipw2x00/ipw2100.c b/drivers/net/wireless/intel/ipw2x00/ipw2100.c
index c4c83ab60cbc..25d7fd6e54bf 100644
--- a/drivers/net/wireless/intel/ipw2x00/ipw2100.c
+++ b/drivers/net/wireless/intel/ipw2x00/ipw2100.c
@@ -5833,7 +5833,7 @@ static int ipw2100_close(struct net_device *dev)
 /*
  * TODO:  Fix this function... its just wrong
  */
-static void ipw2100_tx_timeout(struct net_device *dev)
+static void ipw2100_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct ipw2100_priv *priv = libipw_priv(dev);
 
diff --git a/drivers/net/wireless/intersil/hostap/hostap_main.c b/drivers/net/wireless/intersil/hostap/hostap_main.c
index 05466281afb6..de97b3304115 100644
--- a/drivers/net/wireless/intersil/hostap/hostap_main.c
+++ b/drivers/net/wireless/intersil/hostap/hostap_main.c
@@ -761,7 +761,7 @@ static void hostap_set_multicast_list(struct net_device *dev)
 }
 
 
-static void prism2_tx_timeout(struct net_device *dev)
+static void prism2_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct hostap_interface *iface;
 	local_info_t *local;
diff --git a/drivers/net/wireless/intersil/orinoco/main.c b/drivers/net/wireless/intersil/orinoco/main.c
index 28dac36d7c4c..00264a14e52c 100644
--- a/drivers/net/wireless/intersil/orinoco/main.c
+++ b/drivers/net/wireless/intersil/orinoco/main.c
@@ -647,7 +647,7 @@ static void __orinoco_ev_txexc(struct net_device *dev, struct hermes *hw)
 	netif_wake_queue(dev);
 }
 
-void orinoco_tx_timeout(struct net_device *dev)
+void orinoco_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct orinoco_private *priv = ndev_priv(dev);
 	struct net_device_stats *stats = &dev->stats;
diff --git a/drivers/net/wireless/intersil/orinoco/orinoco.h b/drivers/net/wireless/intersil/orinoco/orinoco.h
index 430862a6a24b..cdd026af100b 100644
--- a/drivers/net/wireless/intersil/orinoco/orinoco.h
+++ b/drivers/net/wireless/intersil/orinoco/orinoco.h
@@ -207,7 +207,7 @@ int orinoco_open(struct net_device *dev);
 int orinoco_stop(struct net_device *dev);
 void orinoco_set_multicast_list(struct net_device *dev);
 int orinoco_change_mtu(struct net_device *dev, int new_mtu);
-void orinoco_tx_timeout(struct net_device *dev);
+void orinoco_tx_timeout(struct net_device *dev, unsigned int txqueue);
 
 /********************************************************************/
 /* Locking and synchronization functions                            */
diff --git a/drivers/net/wireless/intersil/prism54/islpci_eth.c b/drivers/net/wireless/intersil/prism54/islpci_eth.c
index 2b8fb07d07e7..8d680250a281 100644
--- a/drivers/net/wireless/intersil/prism54/islpci_eth.c
+++ b/drivers/net/wireless/intersil/prism54/islpci_eth.c
@@ -473,7 +473,7 @@ islpci_do_reset_and_wake(struct work_struct *work)
 }
 
 void
-islpci_eth_tx_timeout(struct net_device *ndev)
+islpci_eth_tx_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	islpci_private *priv = netdev_priv(ndev);
 
diff --git a/drivers/net/wireless/intersil/prism54/islpci_eth.h b/drivers/net/wireless/intersil/prism54/islpci_eth.h
index 61f4b43c6054..e433ccdc526b 100644
--- a/drivers/net/wireless/intersil/prism54/islpci_eth.h
+++ b/drivers/net/wireless/intersil/prism54/islpci_eth.h
@@ -53,7 +53,7 @@ struct avs_80211_1_header {
 void islpci_eth_cleanup_transmit(islpci_private *, isl38xx_control_block *);
 netdev_tx_t islpci_eth_transmit(struct sk_buff *, struct net_device *);
 int islpci_eth_receive(islpci_private *);
-void islpci_eth_tx_timeout(struct net_device *);
+void islpci_eth_tx_timeout(struct net_device *, unsigned int txqueue);
 void islpci_do_reset_and_wake(struct work_struct *);
 
 #endif				/* _ISL_GEN_H */
diff --git a/drivers/net/wireless/marvell/mwifiex/main.c b/drivers/net/wireless/marvell/mwifiex/main.c
index d14e55e3c9da..7d94695e7961 100644
--- a/drivers/net/wireless/marvell/mwifiex/main.c
+++ b/drivers/net/wireless/marvell/mwifiex/main.c
@@ -1020,7 +1020,7 @@ static void mwifiex_set_multicast_list(struct net_device *dev)
  * CFG802.11 network device handler for transmission timeout.
  */
 static void
-mwifiex_tx_timeout(struct net_device *dev)
+mwifiex_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct mwifiex_private *priv = mwifiex_netdev_get_priv(dev);
 
diff --git a/drivers/net/wireless/quantenna/qtnfmac/core.c b/drivers/net/wireless/quantenna/qtnfmac/core.c
index 5fb598389487..648dfc38bd70 100644
--- a/drivers/net/wireless/quantenna/qtnfmac/core.c
+++ b/drivers/net/wireless/quantenna/qtnfmac/core.c
@@ -156,7 +156,7 @@ static void qtnf_netdev_get_stats64(struct net_device *ndev,
 
 /* Netdev handler for transmission timeout.
  */
-static void qtnf_netdev_tx_timeout(struct net_device *ndev)
+static void qtnf_netdev_tx_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	struct qtnf_vif *vif = qtnf_netdev_get_priv(ndev);
 	struct qtnf_wmac *mac;
diff --git a/drivers/net/wireless/wl3501_cs.c b/drivers/net/wireless/wl3501_cs.c
index 007bf6803293..686161db8706 100644
--- a/drivers/net/wireless/wl3501_cs.c
+++ b/drivers/net/wireless/wl3501_cs.c
@@ -1285,7 +1285,7 @@ out:
 	return rc;
 }
 
-static void wl3501_tx_timeout(struct net_device *dev)
+static void wl3501_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct net_device_stats *stats = &dev->stats;
 	int rc;
diff --git a/drivers/net/wireless/zydas/zd1201.c b/drivers/net/wireless/zydas/zd1201.c
index 0db7362bedb4..41641fc2be74 100644
--- a/drivers/net/wireless/zydas/zd1201.c
+++ b/drivers/net/wireless/zydas/zd1201.c
@@ -830,7 +830,7 @@ static netdev_tx_t zd1201_hard_start_xmit(struct sk_buff *skb,
 	return NETDEV_TX_OK;
 }
 
-static void zd1201_tx_timeout(struct net_device *dev)
+static void zd1201_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct zd1201 *zd = netdev_priv(dev);
 
diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h
index 871d44746f5c..a23875a4fa2b 100644
--- a/drivers/s390/net/qeth_core.h
+++ b/drivers/s390/net/qeth_core.h
@@ -1076,7 +1076,7 @@ void qeth_clear_working_pool_list(struct qeth_card *);
 void qeth_drain_output_queues(struct qeth_card *card);
 void qeth_setadp_promisc_mode(struct qeth_card *card, bool enable);
 int qeth_setadpparms_change_macaddr(struct qeth_card *);
-void qeth_tx_timeout(struct net_device *);
+void qeth_tx_timeout(struct net_device *, unsigned int txqueue);
 void qeth_prepare_ipa_cmd(struct qeth_card *card, struct qeth_cmd_buffer *iob,
 			  u16 cmd_length);
 int qeth_query_switch_attributes(struct qeth_card *card,
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index b9a2349e4b90..ce7ff1abbef3 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -4325,7 +4325,7 @@ int qeth_set_access_ctrl_online(struct qeth_card *card, int fallback)
 	return rc;
 }
 
-void qeth_tx_timeout(struct net_device *dev)
+void qeth_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct qeth_card *card;
 
diff --git a/drivers/staging/ks7010/ks_wlan_net.c b/drivers/staging/ks7010/ks_wlan_net.c
index 3cffc8be6656..211dd4a11cac 100644
--- a/drivers/staging/ks7010/ks_wlan_net.c
+++ b/drivers/staging/ks7010/ks_wlan_net.c
@@ -45,7 +45,7 @@ struct wep_key {
  *	function prototypes
  */
 static int ks_wlan_open(struct net_device *dev);
-static void ks_wlan_tx_timeout(struct net_device *dev);
+static void ks_wlan_tx_timeout(struct net_device *dev, unsigned int txqueue);
 static int ks_wlan_start_xmit(struct sk_buff *skb, struct net_device *dev);
 static int ks_wlan_close(struct net_device *dev);
 static void ks_wlan_set_rx_mode(struct net_device *dev);
@@ -2498,7 +2498,7 @@ int ks_wlan_set_mac_address(struct net_device *dev, void *addr)
 }
 
 static
-void ks_wlan_tx_timeout(struct net_device *dev)
+void ks_wlan_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct ks_wlan_private *priv = netdev_priv(dev);
 
diff --git a/drivers/staging/qlge/qlge_main.c b/drivers/staging/qlge/qlge_main.c
index 6ad4515311f7..24d20f000435 100644
--- a/drivers/staging/qlge/qlge_main.c
+++ b/drivers/staging/qlge/qlge_main.c
@@ -4274,7 +4274,7 @@ static int qlge_set_mac_address(struct net_device *ndev, void *p)
 	return status;
 }
 
-static void qlge_tx_timeout(struct net_device *ndev)
+static void qlge_tx_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	struct ql_adapter *qdev = netdev_priv(ndev);
 	ql_queue_asic_error(qdev);
diff --git a/drivers/staging/rtl8192e/rtl8192e/rtl_core.c b/drivers/staging/rtl8192e/rtl8192e/rtl_core.c
index dace81a7d1ba..a51d627284d1 100644
--- a/drivers/staging/rtl8192e/rtl8192e/rtl_core.c
+++ b/drivers/staging/rtl8192e/rtl8192e/rtl_core.c
@@ -267,7 +267,7 @@ static short _rtl92e_check_nic_enough_desc(struct net_device *dev, int prio)
 	return 0;
 }
 
-static void _rtl92e_tx_timeout(struct net_device *dev)
+static void _rtl92e_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct r8192_priv *priv = rtllib_priv(dev);
 
diff --git a/drivers/staging/rtl8192u/r8192U_core.c b/drivers/staging/rtl8192u/r8192U_core.c
index 7e2cabd16e88..482382a887f8 100644
--- a/drivers/staging/rtl8192u/r8192U_core.c
+++ b/drivers/staging/rtl8192u/r8192U_core.c
@@ -640,7 +640,7 @@ short check_nic_enough_desc(struct net_device *dev, int queue_index)
 	return (used < MAX_TX_URB);
 }
 
-static void tx_timeout(struct net_device *dev)
+static void tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct r8192_priv *priv = ieee80211_priv(dev);
 
diff --git a/drivers/staging/unisys/visornic/visornic_main.c b/drivers/staging/unisys/visornic/visornic_main.c
index 1d1440d43002..0433536930a9 100644
--- a/drivers/staging/unisys/visornic/visornic_main.c
+++ b/drivers/staging/unisys/visornic/visornic_main.c
@@ -1078,7 +1078,7 @@ out_save_flags:
  * Queue the work and return. Make sure we have not already been informed that
  * the IO Partition is gone; if so, we will have already timed-out the xmits.
  */
-static void visornic_xmit_timeout(struct net_device *netdev)
+static void visornic_xmit_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct visornic_devdata *devdata = netdev_priv(netdev);
 	unsigned long flags;
diff --git a/drivers/staging/wlan-ng/p80211netdev.c b/drivers/staging/wlan-ng/p80211netdev.c
index a70fb84f38f1..b809c0015c0c 100644
--- a/drivers/staging/wlan-ng/p80211netdev.c
+++ b/drivers/staging/wlan-ng/p80211netdev.c
@@ -101,7 +101,7 @@ static void p80211knetdev_set_multicast_list(struct net_device *dev);
 static int p80211knetdev_do_ioctl(struct net_device *dev, struct ifreq *ifr,
 				  int cmd);
 static int p80211knetdev_set_mac_address(struct net_device *dev, void *addr);
-static void p80211knetdev_tx_timeout(struct net_device *netdev);
+static void p80211knetdev_tx_timeout(struct net_device *netdev, unsigned int txqueue);
 static int p80211_rx_typedrop(struct wlandevice *wlandev, u16 fc);
 
 int wlan_watchdog = 5000;
@@ -1074,7 +1074,7 @@ static int p80211_rx_typedrop(struct wlandevice *wlandev, u16 fc)
 	return drop;
 }
 
-static void p80211knetdev_tx_timeout(struct net_device *netdev)
+static void p80211knetdev_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct wlandevice *wlandev = netdev->ml_priv;
 
diff --git a/drivers/tty/n_gsm.c b/drivers/tty/n_gsm.c
index 36a3eb4ad4c5..f1c90fa2978e 100644
--- a/drivers/tty/n_gsm.c
+++ b/drivers/tty/n_gsm.c
@@ -2704,7 +2704,7 @@ static netdev_tx_t gsm_mux_net_start_xmit(struct sk_buff *skb,
 }
 
 /* called when a packet did not ack after watchdogtimeout */
-static void gsm_mux_net_tx_timeout(struct net_device *net)
+static void gsm_mux_net_tx_timeout(struct net_device *net, unsigned int txqueue)
 {
 	/* Tell syslog we are hosed. */
 	dev_dbg(&net->dev, "Tx timed out.\n");
diff --git a/drivers/tty/synclink.c b/drivers/tty/synclink.c
index 84f26e43b229..61dc6b4a43d0 100644
--- a/drivers/tty/synclink.c
+++ b/drivers/tty/synclink.c
@@ -7837,7 +7837,7 @@ static int hdlcdev_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
  *
  * dev  pointer to network device structure
  */
-static void hdlcdev_tx_timeout(struct net_device *dev)
+static void hdlcdev_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct mgsl_struct *info = dev_to_port(dev);
 	unsigned long flags;
diff --git a/drivers/tty/synclink_gt.c b/drivers/tty/synclink_gt.c
index e8a9047de451..5d59e2369c8a 100644
--- a/drivers/tty/synclink_gt.c
+++ b/drivers/tty/synclink_gt.c
@@ -1682,7 +1682,7 @@ static int hdlcdev_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
  *
  * dev  pointer to network device structure
  */
-static void hdlcdev_tx_timeout(struct net_device *dev)
+static void hdlcdev_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct slgt_info *info = dev_to_port(dev);
 	unsigned long flags;
diff --git a/drivers/tty/synclinkmp.c b/drivers/tty/synclinkmp.c
index fcb91bf7a15b..33181fa6eb18 100644
--- a/drivers/tty/synclinkmp.c
+++ b/drivers/tty/synclinkmp.c
@@ -1807,7 +1807,7 @@ static int hdlcdev_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
  *
  * dev  pointer to network device structure
  */
-static void hdlcdev_tx_timeout(struct net_device *dev)
+static void hdlcdev_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	SLMP_INFO *info = dev_to_port(dev);
 	unsigned long flags;
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 9ef20389622d..30745068fb39 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1014,7 +1014,7 @@ int netdev_name_node_alt_destroy(struct net_device *dev, const char *name);
  *	Called when a user wants to change the Maximum Transfer Unit
  *	of a device.
  *
- * void (*ndo_tx_timeout)(struct net_device *dev);
+ * void (*ndo_tx_timeout)(struct net_device *dev, unsigned int txqueue);
  *	Callback used when the transmitter has not made any progress
  *	for dev->watchdog ticks.
  *
@@ -1281,7 +1281,8 @@ struct net_device_ops {
 						  int new_mtu);
 	int			(*ndo_neigh_setup)(struct net_device *dev,
 						   struct neigh_parms *);
-	void			(*ndo_tx_timeout) (struct net_device *dev);
+	void			(*ndo_tx_timeout) (struct net_device *dev,
+						   unsigned int txqueue);
 
 	void			(*ndo_get_stats64)(struct net_device *dev,
 						   struct rtnl_link_stats64 *storage);
diff --git a/include/linux/usb/usbnet.h b/include/linux/usb/usbnet.h
index d8860f2d0976..b0bff3083278 100644
--- a/include/linux/usb/usbnet.h
+++ b/include/linux/usb/usbnet.h
@@ -253,7 +253,7 @@ extern int usbnet_open(struct net_device *net);
 extern int usbnet_stop(struct net_device *net);
 extern netdev_tx_t usbnet_start_xmit(struct sk_buff *skb,
 				     struct net_device *net);
-extern void usbnet_tx_timeout(struct net_device *net);
+extern void usbnet_tx_timeout(struct net_device *net, unsigned int txqueue);
 extern int usbnet_change_mtu(struct net_device *net, int new_mtu);
 
 extern int usbnet_get_endpoints(struct usbnet *, struct usb_interface *);
diff --git a/net/atm/lec.c b/net/atm/lec.c
index 5a77c235a212..b57368e70aab 100644
--- a/net/atm/lec.c
+++ b/net/atm/lec.c
@@ -194,7 +194,7 @@ lec_send(struct atm_vcc *vcc, struct sk_buff *skb)
 	dev->stats.tx_bytes += skb->len;
 }
 
-static void lec_tx_timeout(struct net_device *dev)
+static void lec_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	pr_info("%s\n", dev->name);
 	netif_trans_update(dev);
diff --git a/net/bluetooth/bnep/netdev.c b/net/bluetooth/bnep/netdev.c
index 1d4d7d415730..cc1cff63194f 100644
--- a/net/bluetooth/bnep/netdev.c
+++ b/net/bluetooth/bnep/netdev.c
@@ -112,7 +112,7 @@ static int bnep_net_set_mac_addr(struct net_device *dev, void *arg)
 	return 0;
 }
 
-static void bnep_net_timeout(struct net_device *dev)
+static void bnep_net_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	BT_DBG("net_timeout");
 	netif_wake_queue(dev);
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 5ab696efca95..6c9595f1048a 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -441,7 +441,7 @@ static void dev_watchdog(struct timer_list *t)
 				trace_net_dev_xmit_timeout(dev, i);
 				WARN_ONCE(1, KERN_INFO "NETDEV WATCHDOG: %s (%s): transmit queue %u timed out\n",
 				       dev->name, netdev_drivername(dev), i);
-				dev->netdev_ops->ndo_tx_timeout(dev);
+				dev->netdev_ops->ndo_tx_timeout(dev, i);
 			}
 			if (!mod_timer(&dev->watchdog_timer,
 				       round_jiffies(jiffies +
-- 
cgit v1.2.3-59-g8ed1b


From 77656c277cbfe99eaad7217de29164794cacdc63 Mon Sep 17 00:00:00 2001
From: "Michael S. Tsirkin" <mst@redhat.com>
Date: Tue, 10 Dec 2019 09:23:55 -0500
Subject: mlx4: use new txqueue timeout argument

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx4/en_netdev.c | 14 ++++----------
 1 file changed, 4 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
index 71c083960a87..43dcbd8214c6 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
@@ -1367,20 +1367,14 @@ static void mlx4_en_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct mlx4_en_priv *priv = netdev_priv(dev);
 	struct mlx4_en_dev *mdev = priv->mdev;
-	int i;
+	struct mlx4_en_tx_ring *tx_ring = priv->tx_ring[TX][txqueue];
 
 	if (netif_msg_timer(priv))
 		en_warn(priv, "Tx timeout called on port:%d\n", priv->port);
 
-	for (i = 0; i < priv->tx_ring_num[TX]; i++) {
-		struct mlx4_en_tx_ring *tx_ring = priv->tx_ring[TX][i];
-
-		if (!netif_tx_queue_stopped(netdev_get_tx_queue(dev, i)))
-			continue;
-		en_warn(priv, "TX timeout on queue: %d, QP: 0x%x, CQ: 0x%x, Cons: 0x%x, Prod: 0x%x\n",
-			i, tx_ring->qpn, tx_ring->sp_cqn,
-			tx_ring->cons, tx_ring->prod);
-	}
+	en_warn(priv, "TX timeout on queue: %d, QP: 0x%x, CQ: 0x%x, Cons: 0x%x, Prod: 0x%x\n",
+		txqueue, tx_ring->qpn, tx_ring->sp_cqn,
+		tx_ring->cons, tx_ring->prod);
 
 	priv->port_stats.tx_timeout++;
 	en_dbg(DRV, priv, "Scheduling watchdog\n");
-- 
cgit v1.2.3-59-g8ed1b


From d8968eda7fc433e553c3a9b482014d8493554729 Mon Sep 17 00:00:00 2001
From: "Michael S. Tsirkin" <mst@redhat.com>
Date: Tue, 10 Dec 2019 09:26:00 -0500
Subject: netronome: use the new txqueue timeout argument

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/netronome/nfp/nfp_net_common.c | 8 +-------
 1 file changed, 1 insertion(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
index bd305fc6ed5a..d4eeb3b3cf35 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
@@ -1324,14 +1324,8 @@ nfp_net_tx_ring_reset(struct nfp_net_dp *dp, struct nfp_net_tx_ring *tx_ring)
 static void nfp_net_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct nfp_net *nn = netdev_priv(netdev);
-	int i;
 
-	for (i = 0; i < nn->dp.netdev->real_num_tx_queues; i++) {
-		if (!netif_tx_queue_stopped(netdev_get_tx_queue(netdev, i)))
-			continue;
-		nn_warn(nn, "TX timeout on ring: %d\n", i);
-	}
-	nn_warn(nn, "TX watchdog timeout\n");
+	nn_warn(nn, "TX watchdog timeout on ring: %u\n", txqueue);
 }
 
 /* Receive processing
-- 
cgit v1.2.3-59-g8ed1b


From 67d69ccdf389f3b7fabb00e2d81473a915b3b64d Mon Sep 17 00:00:00 2001
From: Jakub Sitnicki <jakub@cloudflare.com>
Date: Thu, 12 Dec 2019 11:22:50 +0100
Subject: libbpf: Recognize SK_REUSEPORT programs from section name

Allow loading BPF object files that contain SK_REUSEPORT programs without
having to manually set the program type before load if the the section name
is set to "sk_reuseport".

Makes user-space code needed to load SK_REUSEPORT BPF program more concise.

Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191212102259.418536-2-jakub@cloudflare.com
---
 tools/lib/bpf/libbpf.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 920d4e06a5f9..b99c0a9b7756 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -4983,6 +4983,7 @@ static const struct {
 	enum bpf_attach_type attach_type;
 } section_names[] = {
 	BPF_PROG_SEC("socket",			BPF_PROG_TYPE_SOCKET_FILTER),
+	BPF_PROG_SEC("sk_reuseport",		BPF_PROG_TYPE_SK_REUSEPORT),
 	BPF_PROG_SEC("kprobe/",			BPF_PROG_TYPE_KPROBE),
 	BPF_PROG_SEC("uprobe/",			BPF_PROG_TYPE_KPROBE),
 	BPF_PROG_SEC("kretprobe/",		BPF_PROG_TYPE_KPROBE),
-- 
cgit v1.2.3-59-g8ed1b


From 1fbcef929d7d71321745e61e8c20b0a02bd38cf1 Mon Sep 17 00:00:00 2001
From: Jakub Sitnicki <jakub@cloudflare.com>
Date: Thu, 12 Dec 2019 11:22:51 +0100
Subject: selftests/bpf: Let libbpf determine program type from section name

Now that libbpf can recognize SK_REUSEPORT programs, we no longer have to
pass a prog_type hint before loading the object file.

Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191212102259.418536-3-jakub@cloudflare.com
---
 .../testing/selftests/bpf/progs/test_select_reuseport_kern.c |  2 +-
 tools/testing/selftests/bpf/test_select_reuseport.c          | 12 +++---------
 2 files changed, 4 insertions(+), 10 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/test_select_reuseport_kern.c b/tools/testing/selftests/bpf/progs/test_select_reuseport_kern.c
index ea7d84f01235..b1f09f5bb1cf 100644
--- a/tools/testing/selftests/bpf/progs/test_select_reuseport_kern.c
+++ b/tools/testing/selftests/bpf/progs/test_select_reuseport_kern.c
@@ -62,7 +62,7 @@ struct {
 	goto done;				\
 })
 
-SEC("select_by_skb_data")
+SEC("sk_reuseport")
 int _select_by_skb_data(struct sk_reuseport_md *reuse_md)
 {
 	__u32 linum, index = 0, flags = 0, index_zero = 0;
diff --git a/tools/testing/selftests/bpf/test_select_reuseport.c b/tools/testing/selftests/bpf/test_select_reuseport.c
index 7566c13eb51a..1e3cfe1cb28a 100644
--- a/tools/testing/selftests/bpf/test_select_reuseport.c
+++ b/tools/testing/selftests/bpf/test_select_reuseport.c
@@ -87,19 +87,11 @@ static void prepare_bpf_obj(void)
 	struct bpf_program *prog;
 	struct bpf_map *map;
 	int err;
-	struct bpf_object_open_attr attr = {
-		.file = "test_select_reuseport_kern.o",
-		.prog_type = BPF_PROG_TYPE_SK_REUSEPORT,
-	};
 
-	obj = bpf_object__open_xattr(&attr);
+	obj = bpf_object__open("test_select_reuseport_kern.o");
 	CHECK(IS_ERR_OR_NULL(obj), "open test_select_reuseport_kern.o",
 	      "obj:%p PTR_ERR(obj):%ld\n", obj, PTR_ERR(obj));
 
-	prog = bpf_program__next(NULL, obj);
-	CHECK(!prog, "get first bpf_program", "!prog\n");
-	bpf_program__set_type(prog, attr.prog_type);
-
 	map = bpf_object__find_map_by_name(obj, "outer_map");
 	CHECK(!map, "find outer_map", "!map\n");
 	err = bpf_map__reuse_fd(map, outer_map);
@@ -108,6 +100,8 @@ static void prepare_bpf_obj(void)
 	err = bpf_object__load(obj);
 	CHECK(err, "load bpf_object", "err:%d\n", err);
 
+	prog = bpf_program__next(NULL, obj);
+	CHECK(!prog, "get first bpf_program", "!prog\n");
 	select_by_skb_data_prog = bpf_program__fd(prog);
 	CHECK(select_by_skb_data_prog == -1, "get prog fd",
 	      "select_by_skb_data_prog:%d\n", select_by_skb_data_prog);
-- 
cgit v1.2.3-59-g8ed1b


From 11f80355d4d27c7c798fc9890b4056c5d98af992 Mon Sep 17 00:00:00 2001
From: Jakub Sitnicki <jakub@cloudflare.com>
Date: Thu, 12 Dec 2019 11:22:52 +0100
Subject: selftests/bpf: Use sa_family_t everywhere in reuseport tests

Update the only function that is not using sa_family_t in this source file.

Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191212102259.418536-4-jakub@cloudflare.com
---
 tools/testing/selftests/bpf/test_select_reuseport.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/tools/testing/selftests/bpf/test_select_reuseport.c b/tools/testing/selftests/bpf/test_select_reuseport.c
index 1e3cfe1cb28a..a295a087a026 100644
--- a/tools/testing/selftests/bpf/test_select_reuseport.c
+++ b/tools/testing/selftests/bpf/test_select_reuseport.c
@@ -643,7 +643,7 @@ static void prepare_sk_fds(int type, sa_family_t family, bool inany)
 	}
 }
 
-static void setup_per_test(int type, unsigned short family, bool inany)
+static void setup_per_test(int type, sa_family_t family, bool inany)
 {
 	int ovr = -1, err;
 
@@ -680,12 +680,12 @@ static void test_all(void)
 	const int types[] = { SOCK_STREAM, SOCK_DGRAM, SOCK_STREAM };
 	const char * const type_strings[] = { "TCP", "UDP", "TCP" };
 	const char * const family_strings[] = { "IPv6", "IPv4" };
-	const unsigned short families[] = { AF_INET6, AF_INET };
+	const sa_family_t families[] = { AF_INET6, AF_INET };
 	const bool bind_inany[] = { false, false, true };
 	int t, f, err;
 
 	for (f = 0; f < ARRAY_SIZE(families); f++) {
-		unsigned short family = families[f];
+		sa_family_t family = families[f];
 
 		for (t = 0; t < ARRAY_SIZE(types); t++) {
 			bool inany = bind_inany[t];
-- 
cgit v1.2.3-59-g8ed1b


From a9ce4cf4e47c55ab40641b92551bc9e1de4bfa61 Mon Sep 17 00:00:00 2001
From: Jakub Sitnicki <jakub@cloudflare.com>
Date: Thu, 12 Dec 2019 11:22:53 +0100
Subject: selftests/bpf: Add helpers for getting socket family & type name

Having string arrays to map socket family & type to a name prevents us from
unrolling the test runner loop in the subsequent patch. Introduce helpers
that do the same thing.

Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191212102259.418536-5-jakub@cloudflare.com
---
 .../testing/selftests/bpf/test_select_reuseport.c  | 28 +++++++++++++++++++---
 1 file changed, 25 insertions(+), 3 deletions(-)

diff --git a/tools/testing/selftests/bpf/test_select_reuseport.c b/tools/testing/selftests/bpf/test_select_reuseport.c
index a295a087a026..ef98a7de6704 100644
--- a/tools/testing/selftests/bpf/test_select_reuseport.c
+++ b/tools/testing/selftests/bpf/test_select_reuseport.c
@@ -674,12 +674,34 @@ static void cleanup(void)
 	bpf_object__close(obj);
 }
 
+static const char *family_str(sa_family_t family)
+{
+	switch (family) {
+	case AF_INET:
+		return "IPv4";
+	case AF_INET6:
+		return "IPv6";
+	default:
+		return "unknown";
+	}
+}
+
+static const char *sotype_str(int sotype)
+{
+	switch (sotype) {
+	case SOCK_STREAM:
+		return "TCP";
+	case SOCK_DGRAM:
+		return "UDP";
+	default:
+		return "unknown";
+	}
+}
+
 static void test_all(void)
 {
 	/* Extra SOCK_STREAM to test bind_inany==true */
 	const int types[] = { SOCK_STREAM, SOCK_DGRAM, SOCK_STREAM };
-	const char * const type_strings[] = { "TCP", "UDP", "TCP" };
-	const char * const family_strings[] = { "IPv6", "IPv4" };
 	const sa_family_t families[] = { AF_INET6, AF_INET };
 	const bool bind_inany[] = { false, false, true };
 	int t, f, err;
@@ -692,7 +714,7 @@ static void test_all(void)
 			int type = types[t];
 
 			printf("######## %s/%s %s ########\n",
-			       family_strings[f], type_strings[t],
+			       family_str(family), sotype_str(type),
 				inany ? " INANY  " : "LOOPBACK");
 
 			setup_per_test(type, family, inany);
-- 
cgit v1.2.3-59-g8ed1b


From 9936338258308aa008387dc4f9fd3f5fec6c046a Mon Sep 17 00:00:00 2001
From: Jakub Sitnicki <jakub@cloudflare.com>
Date: Thu, 12 Dec 2019 11:22:54 +0100
Subject: selftests/bpf: Unroll the main loop in reuseport test

Prepare for iterating over individual tests without introducing another
nested loop in the main test function.

Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191212102259.418536-6-jakub@cloudflare.com
---
 .../testing/selftests/bpf/test_select_reuseport.c  | 73 ++++++++++++----------
 1 file changed, 41 insertions(+), 32 deletions(-)

diff --git a/tools/testing/selftests/bpf/test_select_reuseport.c b/tools/testing/selftests/bpf/test_select_reuseport.c
index ef98a7de6704..63ce2e75e758 100644
--- a/tools/testing/selftests/bpf/test_select_reuseport.c
+++ b/tools/testing/selftests/bpf/test_select_reuseport.c
@@ -698,47 +698,56 @@ static const char *sotype_str(int sotype)
 	}
 }
 
-static void test_all(void)
+static void test_config(int type, sa_family_t family, bool inany)
 {
-	/* Extra SOCK_STREAM to test bind_inany==true */
-	const int types[] = { SOCK_STREAM, SOCK_DGRAM, SOCK_STREAM };
-	const sa_family_t families[] = { AF_INET6, AF_INET };
-	const bool bind_inany[] = { false, false, true };
-	int t, f, err;
+	int err;
 
-	for (f = 0; f < ARRAY_SIZE(families); f++) {
-		sa_family_t family = families[f];
+	printf("######## %s/%s %s ########\n",
+	       family_str(family), sotype_str(type),
+	       inany ? " INANY  " : "LOOPBACK");
 
-		for (t = 0; t < ARRAY_SIZE(types); t++) {
-			bool inany = bind_inany[t];
-			int type = types[t];
+	setup_per_test(type, family, inany);
 
-			printf("######## %s/%s %s ########\n",
-			       family_str(family), sotype_str(type),
-				inany ? " INANY  " : "LOOPBACK");
+	test_err_inner_map(type, family);
 
-			setup_per_test(type, family, inany);
+	/* Install reuseport_array to the outer_map */
+	err = bpf_map_update_elem(outer_map, &index_zero,
+				  &reuseport_array, BPF_ANY);
+	CHECK(err == -1, "update_elem(outer_map)",
+	      "err:%d errno:%d\n", err, errno);
 
-			test_err_inner_map(type, family);
+	test_err_skb_data(type, family);
+	test_err_sk_select_port(type, family);
+	test_pass(type, family);
+	test_syncookie(type, family);
+	test_pass_on_err(type, family);
+	/* Must be the last test */
+	test_detach_bpf(type, family);
 
-			/* Install reuseport_array to the outer_map */
-			err = bpf_map_update_elem(outer_map, &index_zero,
-						  &reuseport_array, BPF_ANY);
-			CHECK(err == -1, "update_elem(outer_map)",
-			      "err:%d errno:%d\n", err, errno);
+	cleanup_per_test();
+	printf("\n");
+}
 
-			test_err_skb_data(type, family);
-			test_err_sk_select_port(type, family);
-			test_pass(type, family);
-			test_syncookie(type, family);
-			test_pass_on_err(type, family);
-			/* Must be the last test */
-			test_detach_bpf(type, family);
+#define BIND_INANY true
 
-			cleanup_per_test();
-			printf("\n");
-		}
-	}
+static void test_all(void)
+{
+	const struct config {
+		int sotype;
+		sa_family_t family;
+		bool inany;
+	} configs[] = {
+		{ SOCK_STREAM, AF_INET },
+		{ SOCK_STREAM, AF_INET, BIND_INANY },
+		{ SOCK_STREAM, AF_INET6 },
+		{ SOCK_STREAM, AF_INET6, BIND_INANY },
+		{ SOCK_DGRAM, AF_INET },
+		{ SOCK_DGRAM, AF_INET6 },
+	};
+	const struct config *c;
+
+	for (c = configs; c < configs + ARRAY_SIZE(configs); c++)
+		test_config(c->sotype, c->family, c->inany);
 }
 
 int main(int argc, const char **argv)
-- 
cgit v1.2.3-59-g8ed1b


From ce7cb5f3921cdf0f65ab877764265c6c0be34c3c Mon Sep 17 00:00:00 2001
From: Jakub Sitnicki <jakub@cloudflare.com>
Date: Thu, 12 Dec 2019 11:22:55 +0100
Subject: selftests/bpf: Run reuseport tests in a loop

Prepare for switching reuseport tests to test_progs framework. Loop over
the tests and perform setup/cleanup for each test separately, remembering
that with test_progs we can select tests to run.

Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191212102259.418536-7-jakub@cloudflare.com
---
 .../testing/selftests/bpf/test_select_reuseport.c  | 55 +++++++++++++---------
 1 file changed, 34 insertions(+), 21 deletions(-)

diff --git a/tools/testing/selftests/bpf/test_select_reuseport.c b/tools/testing/selftests/bpf/test_select_reuseport.c
index 63ce2e75e758..cfff958da570 100644
--- a/tools/testing/selftests/bpf/test_select_reuseport.c
+++ b/tools/testing/selftests/bpf/test_select_reuseport.c
@@ -643,7 +643,8 @@ static void prepare_sk_fds(int type, sa_family_t family, bool inany)
 	}
 }
 
-static void setup_per_test(int type, sa_family_t family, bool inany)
+static void setup_per_test(int type, sa_family_t family, bool inany,
+			   bool no_inner_map)
 {
 	int ovr = -1, err;
 
@@ -652,9 +653,18 @@ static void setup_per_test(int type, sa_family_t family, bool inany)
 				  BPF_ANY);
 	CHECK(err == -1, "update_elem(tmp_index_ovr_map, 0, -1)",
 	      "err:%d errno:%d\n", err, errno);
+
+	/* Install reuseport_array to outer_map? */
+	if (no_inner_map)
+		return;
+
+	err = bpf_map_update_elem(outer_map, &index_zero, &reuseport_array,
+				  BPF_ANY);
+	CHECK(err == -1, "update_elem(outer_map, 0, reuseport_array)",
+	      "err:%d errno:%d\n", err, errno);
 }
 
-static void cleanup_per_test(void)
+static void cleanup_per_test(bool no_inner_map)
 {
 	int i, err;
 
@@ -662,6 +672,10 @@ static void cleanup_per_test(void)
 		close(sk_fds[i]);
 	close(epfd);
 
+	/* Delete reuseport_array from outer_map? */
+	if (no_inner_map)
+		return;
+
 	err = bpf_map_delete_elem(outer_map, &index_zero);
 	CHECK(err == -1, "delete_elem(outer_map)",
 	      "err:%d errno:%d\n", err, errno);
@@ -700,31 +714,30 @@ static const char *sotype_str(int sotype)
 
 static void test_config(int type, sa_family_t family, bool inany)
 {
-	int err;
+	const struct test {
+		void (*fn)(int sotype, sa_family_t family);
+		bool no_inner_map;
+	} tests[] = {
+		{ test_err_inner_map, true /* no_inner_map */ },
+		{ test_err_skb_data },
+		{ test_err_sk_select_port },
+		{ test_pass },
+		{ test_syncookie },
+		{ test_pass_on_err },
+		{ test_detach_bpf },
+	};
+	const struct test *t;
 
 	printf("######## %s/%s %s ########\n",
 	       family_str(family), sotype_str(type),
 	       inany ? " INANY  " : "LOOPBACK");
 
-	setup_per_test(type, family, inany);
-
-	test_err_inner_map(type, family);
-
-	/* Install reuseport_array to the outer_map */
-	err = bpf_map_update_elem(outer_map, &index_zero,
-				  &reuseport_array, BPF_ANY);
-	CHECK(err == -1, "update_elem(outer_map)",
-	      "err:%d errno:%d\n", err, errno);
-
-	test_err_skb_data(type, family);
-	test_err_sk_select_port(type, family);
-	test_pass(type, family);
-	test_syncookie(type, family);
-	test_pass_on_err(type, family);
-	/* Must be the last test */
-	test_detach_bpf(type, family);
+	for (t = tests; t < tests + ARRAY_SIZE(tests); t++) {
+		setup_per_test(type, family, inany, t->no_inner_map);
+		t->fn(type, family);
+		cleanup_per_test(t->no_inner_map);
+	}
 
-	cleanup_per_test();
 	printf("\n");
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 9af6c84435d0252e532c7f8bb70f1ed64a22ae2a Mon Sep 17 00:00:00 2001
From: Jakub Sitnicki <jakub@cloudflare.com>
Date: Thu, 12 Dec 2019 11:22:56 +0100
Subject: selftests/bpf: Propagate errors during setup for reuseport tests

Prepare for switching reuseport tests to test_progs framework, where we
don't have the luxury to terminate the process on failure.

Modify setup helpers to signal failure via the return value with the help
of a macro similar to the one currently in use by the tests.

Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191212102259.418536-8-jakub@cloudflare.com
---
 .../testing/selftests/bpf/test_select_reuseport.c  | 136 +++++++++++++--------
 1 file changed, 85 insertions(+), 51 deletions(-)

diff --git a/tools/testing/selftests/bpf/test_select_reuseport.c b/tools/testing/selftests/bpf/test_select_reuseport.c
index cfff958da570..cc35816b7b34 100644
--- a/tools/testing/selftests/bpf/test_select_reuseport.c
+++ b/tools/testing/selftests/bpf/test_select_reuseport.c
@@ -32,7 +32,7 @@
 static int result_map, tmp_index_ovr_map, linum_map, data_check_map;
 static enum result expected_results[NR_RESULTS];
 static int sk_fds[REUSEPORT_ARRAY_SIZE];
-static int reuseport_array, outer_map;
+static int reuseport_array = -1, outer_map = -1;
 static int select_by_skb_data_prog;
 static int saved_tcp_syncookie;
 static struct bpf_object *obj;
@@ -55,7 +55,16 @@ static union sa46 {
 	}								\
 })
 
-static void create_maps(void)
+#define RET_ERR(condition, tag, format...) ({				\
+	int __ret = !!(condition);					\
+	if (__ret) {							\
+		printf("%s(%d):FAIL:%s ", __func__, __LINE__, tag);	\
+		printf(format);						\
+		return -1;						\
+	}								\
+})
+
+static int create_maps(void)
 {
 	struct bpf_create_map_attr attr = {};
 
@@ -67,8 +76,8 @@ static void create_maps(void)
 	attr.max_entries = REUSEPORT_ARRAY_SIZE;
 
 	reuseport_array = bpf_create_map_xattr(&attr);
-	CHECK(reuseport_array == -1, "creating reuseport_array",
-	      "reuseport_array:%d errno:%d\n", reuseport_array, errno);
+	RET_ERR(reuseport_array == -1, "creating reuseport_array",
+		"reuseport_array:%d errno:%d\n", reuseport_array, errno);
 
 	/* Creating outer_map */
 	attr.name = "outer_map";
@@ -78,57 +87,61 @@ static void create_maps(void)
 	attr.max_entries = 1;
 	attr.inner_map_fd = reuseport_array;
 	outer_map = bpf_create_map_xattr(&attr);
-	CHECK(outer_map == -1, "creating outer_map",
-	      "outer_map:%d errno:%d\n", outer_map, errno);
+	RET_ERR(outer_map == -1, "creating outer_map",
+		"outer_map:%d errno:%d\n", outer_map, errno);
+
+	return 0;
 }
 
-static void prepare_bpf_obj(void)
+static int prepare_bpf_obj(void)
 {
 	struct bpf_program *prog;
 	struct bpf_map *map;
 	int err;
 
 	obj = bpf_object__open("test_select_reuseport_kern.o");
-	CHECK(IS_ERR_OR_NULL(obj), "open test_select_reuseport_kern.o",
-	      "obj:%p PTR_ERR(obj):%ld\n", obj, PTR_ERR(obj));
+	RET_ERR(IS_ERR_OR_NULL(obj), "open test_select_reuseport_kern.o",
+		"obj:%p PTR_ERR(obj):%ld\n", obj, PTR_ERR(obj));
 
 	map = bpf_object__find_map_by_name(obj, "outer_map");
-	CHECK(!map, "find outer_map", "!map\n");
+	RET_ERR(!map, "find outer_map", "!map\n");
 	err = bpf_map__reuse_fd(map, outer_map);
-	CHECK(err, "reuse outer_map", "err:%d\n", err);
+	RET_ERR(err, "reuse outer_map", "err:%d\n", err);
 
 	err = bpf_object__load(obj);
-	CHECK(err, "load bpf_object", "err:%d\n", err);
+	RET_ERR(err, "load bpf_object", "err:%d\n", err);
 
 	prog = bpf_program__next(NULL, obj);
-	CHECK(!prog, "get first bpf_program", "!prog\n");
+	RET_ERR(!prog, "get first bpf_program", "!prog\n");
 	select_by_skb_data_prog = bpf_program__fd(prog);
-	CHECK(select_by_skb_data_prog == -1, "get prog fd",
-	      "select_by_skb_data_prog:%d\n", select_by_skb_data_prog);
+	RET_ERR(select_by_skb_data_prog == -1, "get prog fd",
+		"select_by_skb_data_prog:%d\n", select_by_skb_data_prog);
 
 	map = bpf_object__find_map_by_name(obj, "result_map");
-	CHECK(!map, "find result_map", "!map\n");
+	RET_ERR(!map, "find result_map", "!map\n");
 	result_map = bpf_map__fd(map);
-	CHECK(result_map == -1, "get result_map fd",
-	      "result_map:%d\n", result_map);
+	RET_ERR(result_map == -1, "get result_map fd",
+		"result_map:%d\n", result_map);
 
 	map = bpf_object__find_map_by_name(obj, "tmp_index_ovr_map");
-	CHECK(!map, "find tmp_index_ovr_map", "!map\n");
+	RET_ERR(!map, "find tmp_index_ovr_map\n", "!map");
 	tmp_index_ovr_map = bpf_map__fd(map);
-	CHECK(tmp_index_ovr_map == -1, "get tmp_index_ovr_map fd",
-	      "tmp_index_ovr_map:%d\n", tmp_index_ovr_map);
+	RET_ERR(tmp_index_ovr_map == -1, "get tmp_index_ovr_map fd",
+		"tmp_index_ovr_map:%d\n", tmp_index_ovr_map);
 
 	map = bpf_object__find_map_by_name(obj, "linum_map");
-	CHECK(!map, "find linum_map", "!map\n");
+	RET_ERR(!map, "find linum_map", "!map\n");
 	linum_map = bpf_map__fd(map);
-	CHECK(linum_map == -1, "get linum_map fd",
-	      "linum_map:%d\n", linum_map);
+	RET_ERR(linum_map == -1, "get linum_map fd",
+		"linum_map:%d\n", linum_map);
 
 	map = bpf_object__find_map_by_name(obj, "data_check_map");
-	CHECK(!map, "find data_check_map", "!map\n");
+	RET_ERR(!map, "find data_check_map", "!map\n");
 	data_check_map = bpf_map__fd(map);
-	CHECK(data_check_map == -1, "get data_check_map fd",
-	      "data_check_map:%d\n", data_check_map);
+	RET_ERR(data_check_map == -1, "get data_check_map fd",
+		"data_check_map:%d\n", data_check_map);
+
+	return 0;
 }
 
 static void sa46_init_loopback(union sa46 *sa, sa_family_t family)
@@ -157,31 +170,34 @@ static int read_int_sysctl(const char *sysctl)
 	int fd, ret;
 
 	fd = open(sysctl, 0);
-	CHECK(fd == -1, "open(sysctl)", "sysctl:%s fd:%d errno:%d\n",
-	      sysctl, fd, errno);
+	RET_ERR(fd == -1, "open(sysctl)",
+		"sysctl:%s fd:%d errno:%d\n", sysctl, fd, errno);
 
 	ret = read(fd, buf, sizeof(buf));
-	CHECK(ret <= 0, "read(sysctl)", "sysctl:%s ret:%d errno:%d\n",
-	      sysctl, ret, errno);
-	close(fd);
+	RET_ERR(ret <= 0, "read(sysctl)",
+		"sysctl:%s ret:%d errno:%d\n", sysctl, ret, errno);
 
+	close(fd);
 	return atoi(buf);
 }
 
-static void write_int_sysctl(const char *sysctl, int v)
+static int write_int_sysctl(const char *sysctl, int v)
 {
 	int fd, ret, size;
 	char buf[16];
 
 	fd = open(sysctl, O_RDWR);
-	CHECK(fd == -1, "open(sysctl)", "sysctl:%s fd:%d errno:%d\n",
-	      sysctl, fd, errno);
+	RET_ERR(fd == -1, "open(sysctl)",
+		"sysctl:%s fd:%d errno:%d\n", sysctl, fd, errno);
 
 	size = snprintf(buf, sizeof(buf), "%d", v);
 	ret = write(fd, buf, size);
-	CHECK(ret != size, "write(sysctl)",
-	      "sysctl:%s ret:%d size:%d errno:%d\n", sysctl, ret, size, errno);
+	RET_ERR(ret != size, "write(sysctl)",
+		"sysctl:%s ret:%d size:%d errno:%d\n",
+		sysctl, ret, size, errno);
+
 	close(fd);
+	return 0;
 }
 
 static void restore_sysctls(void)
@@ -190,22 +206,25 @@ static void restore_sysctls(void)
 	write_int_sysctl(TCP_SYNCOOKIE_SYSCTL, saved_tcp_syncookie);
 }
 
-static void enable_fastopen(void)
+static int enable_fastopen(void)
 {
 	int fo;
 
 	fo = read_int_sysctl(TCP_FO_SYSCTL);
-	write_int_sysctl(TCP_FO_SYSCTL, fo | 7);
+	if (fo < 0)
+		return -1;
+
+	return write_int_sysctl(TCP_FO_SYSCTL, fo | 7);
 }
 
-static void enable_syncookie(void)
+static int enable_syncookie(void)
 {
-	write_int_sysctl(TCP_SYNCOOKIE_SYSCTL, 2);
+	return write_int_sysctl(TCP_SYNCOOKIE_SYSCTL, 2);
 }
 
-static void disable_syncookie(void)
+static int disable_syncookie(void)
 {
-	write_int_sysctl(TCP_SYNCOOKIE_SYSCTL, 0);
+	return write_int_sysctl(TCP_SYNCOOKIE_SYSCTL, 0);
 }
 
 static __u32 get_linum(void)
@@ -683,9 +702,12 @@ static void cleanup_per_test(bool no_inner_map)
 
 static void cleanup(void)
 {
-	close(outer_map);
-	close(reuseport_array);
-	bpf_object__close(obj);
+	if (outer_map != -1)
+		close(outer_map);
+	if (reuseport_array != -1)
+		close(reuseport_array);
+	if (obj)
+		bpf_object__close(obj);
 }
 
 static const char *family_str(sa_family_t family)
@@ -765,16 +787,28 @@ static void test_all(void)
 
 int main(int argc, const char **argv)
 {
-	create_maps();
-	prepare_bpf_obj();
+	int ret = EXIT_FAILURE;
+
+	if (create_maps())
+		goto out;
+	if (prepare_bpf_obj())
+		goto out;
+
 	saved_tcp_fo = read_int_sysctl(TCP_FO_SYSCTL);
 	saved_tcp_syncookie = read_int_sysctl(TCP_SYNCOOKIE_SYSCTL);
-	enable_fastopen();
-	disable_syncookie();
+	if (saved_tcp_syncookie < 0 || saved_tcp_syncookie < 0)
+		goto out;
 	atexit(restore_sysctls);
 
+	if (enable_fastopen())
+		goto out;
+	if (disable_syncookie())
+		goto out;
+
 	test_all();
 
+	ret = EXIT_SUCCESS;
+out:
 	cleanup();
-	return 0;
+	return ret;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 250a91d48ae7616672a34b58388620a554a00180 Mon Sep 17 00:00:00 2001
From: Jakub Sitnicki <jakub@cloudflare.com>
Date: Thu, 12 Dec 2019 11:22:57 +0100
Subject: selftests/bpf: Pull up printing the test name into test runner

Again, prepare for switching reuseport tests to test_progs framework.
test_progs framework will print the subtest name for us if we set it.

Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191212102259.418536-9-jakub@cloudflare.com
---
 .../testing/selftests/bpf/test_select_reuseport.c  | 27 ++++++++++------------
 1 file changed, 12 insertions(+), 15 deletions(-)

diff --git a/tools/testing/selftests/bpf/test_select_reuseport.c b/tools/testing/selftests/bpf/test_select_reuseport.c
index cc35816b7b34..0d5687feb689 100644
--- a/tools/testing/selftests/bpf/test_select_reuseport.c
+++ b/tools/testing/selftests/bpf/test_select_reuseport.c
@@ -441,7 +441,6 @@ static void test_err_inner_map(int type, sa_family_t family)
 		.pass_on_failure = 0,
 	};
 
-	printf("%s: ", __func__);
 	expected_results[DROP_ERR_INNER_MAP]++;
 	do_test(type, family, &cmd, DROP_ERR_INNER_MAP);
 	printf("OK\n");
@@ -449,7 +448,6 @@ static void test_err_inner_map(int type, sa_family_t family)
 
 static void test_err_skb_data(int type, sa_family_t family)
 {
-	printf("%s: ", __func__);
 	expected_results[DROP_ERR_SKB_DATA]++;
 	do_test(type, family, NULL, DROP_ERR_SKB_DATA);
 	printf("OK\n");
@@ -462,7 +460,6 @@ static void test_err_sk_select_port(int type, sa_family_t family)
 		.pass_on_failure = 0,
 	};
 
-	printf("%s: ", __func__);
 	expected_results[DROP_ERR_SK_SELECT_REUSEPORT]++;
 	do_test(type, family, &cmd, DROP_ERR_SK_SELECT_REUSEPORT);
 	printf("OK\n");
@@ -473,7 +470,6 @@ static void test_pass(int type, sa_family_t family)
 	struct cmd cmd;
 	int i;
 
-	printf("%s: ", __func__);
 	cmd.pass_on_failure = 0;
 	for (i = 0; i < REUSEPORT_ARRAY_SIZE; i++) {
 		expected_results[PASS]++;
@@ -494,7 +490,6 @@ static void test_syncookie(int type, sa_family_t family)
 	if (type != SOCK_STREAM)
 		return;
 
-	printf("%s: ", __func__);
 	/*
 	 * +1 for TCP-SYN and
 	 * +1 for the TCP-ACK (ack the syncookie)
@@ -530,7 +525,6 @@ static void test_pass_on_err(int type, sa_family_t family)
 		.pass_on_failure = 1,
 	};
 
-	printf("%s: ", __func__);
 	expected_results[PASS_ERR_SK_SELECT_REUSEPORT] += 1;
 	do_test(type, family, &cmd, PASS_ERR_SK_SELECT_REUSEPORT);
 	printf("OK\n");
@@ -545,7 +539,6 @@ static void test_detach_bpf(int type, sa_family_t family)
 	struct cmd cmd = {};
 	int optvalue = 0;
 
-	printf("%s: ", __func__);
 	err = setsockopt(sk_fds[0], SOL_SOCKET, SO_DETACH_REUSEPORT_BPF,
 			 &optvalue, sizeof(optvalue));
 	CHECK(err == -1, "setsockopt(SO_DETACH_REUSEPORT_BPF)",
@@ -584,7 +577,7 @@ static void test_detach_bpf(int type, sa_family_t family)
 	printf("OK\n");
 	close(cli_fd);
 #else
-	printf("%s: SKIP\n", __func__);
+	printf("SKIP\n");
 #endif
 }
 
@@ -734,19 +727,22 @@ static const char *sotype_str(int sotype)
 	}
 }
 
+#define TEST_INIT(fn, ...) { fn, #fn, __VA_ARGS__ }
+
 static void test_config(int type, sa_family_t family, bool inany)
 {
 	const struct test {
 		void (*fn)(int sotype, sa_family_t family);
+		const char *name;
 		bool no_inner_map;
 	} tests[] = {
-		{ test_err_inner_map, true /* no_inner_map */ },
-		{ test_err_skb_data },
-		{ test_err_sk_select_port },
-		{ test_pass },
-		{ test_syncookie },
-		{ test_pass_on_err },
-		{ test_detach_bpf },
+		TEST_INIT(test_err_inner_map, true /* no_inner_map */),
+		TEST_INIT(test_err_skb_data),
+		TEST_INIT(test_err_sk_select_port),
+		TEST_INIT(test_pass),
+		TEST_INIT(test_syncookie),
+		TEST_INIT(test_pass_on_err),
+		TEST_INIT(test_detach_bpf),
 	};
 	const struct test *t;
 
@@ -756,6 +752,7 @@ static void test_config(int type, sa_family_t family, bool inany)
 
 	for (t = tests; t < tests + ARRAY_SIZE(tests); t++) {
 		setup_per_test(type, family, inany, t->no_inner_map);
+		printf("%s: ", t->name);
 		t->fn(type, family);
 		cleanup_per_test(t->no_inner_map);
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 415bb4e125b68a3aaccb8381baaa4a06dd1dcb27 Mon Sep 17 00:00:00 2001
From: Jakub Sitnicki <jakub@cloudflare.com>
Date: Thu, 12 Dec 2019 11:22:58 +0100
Subject: selftests/bpf: Move reuseport tests under prog_tests/

Do a pure move the show the actual work needed to adapt the tests in
subsequent patch at the cost of breaking test_progs build for the moment.

Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191212102259.418536-10-jakub@cloudflare.com
---
 tools/testing/selftests/bpf/Makefile               |   2 +-
 .../selftests/bpf/prog_tests/select_reuseport.c    | 811 +++++++++++++++++++++
 .../testing/selftests/bpf/test_select_reuseport.c  | 811 ---------------------
 3 files changed, 812 insertions(+), 812 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/select_reuseport.c
 delete mode 100644 tools/testing/selftests/bpf/test_select_reuseport.c

diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index e0fe01d9ec33..90de7d75b5c6 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -28,7 +28,7 @@ LDLIBS += -lcap -lelf -lrt -lpthread
 TEST_GEN_PROGS = test_verifier test_tag test_maps test_lru_map test_lpm_map test_progs \
 	test_align test_verifier_log test_dev_cgroup test_tcpbpf_user \
 	test_sock test_btf test_sockmap get_cgroup_id_user test_socket_cookie \
-	test_cgroup_storage test_select_reuseport \
+	test_cgroup_storage \
 	test_netcnt test_tcpnotify_user test_sock_fields test_sysctl test_hashmap \
 	test_cgroup_attach test_progs-no_alu32
 
diff --git a/tools/testing/selftests/bpf/prog_tests/select_reuseport.c b/tools/testing/selftests/bpf/prog_tests/select_reuseport.c
new file mode 100644
index 000000000000..0d5687feb689
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/select_reuseport.c
@@ -0,0 +1,811 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2018 Facebook */
+
+#include <stdlib.h>
+#include <unistd.h>
+#include <stdbool.h>
+#include <string.h>
+#include <errno.h>
+#include <assert.h>
+#include <fcntl.h>
+#include <linux/bpf.h>
+#include <linux/err.h>
+#include <linux/types.h>
+#include <linux/if_ether.h>
+#include <sys/types.h>
+#include <sys/epoll.h>
+#include <sys/socket.h>
+#include <netinet/in.h>
+#include <bpf/bpf.h>
+#include <bpf/libbpf.h>
+#include "bpf_rlimit.h"
+#include "bpf_util.h"
+#include "test_select_reuseport_common.h"
+
+#define MIN_TCPHDR_LEN 20
+#define UDPHDR_LEN 8
+
+#define TCP_SYNCOOKIE_SYSCTL "/proc/sys/net/ipv4/tcp_syncookies"
+#define TCP_FO_SYSCTL "/proc/sys/net/ipv4/tcp_fastopen"
+#define REUSEPORT_ARRAY_SIZE 32
+
+static int result_map, tmp_index_ovr_map, linum_map, data_check_map;
+static enum result expected_results[NR_RESULTS];
+static int sk_fds[REUSEPORT_ARRAY_SIZE];
+static int reuseport_array = -1, outer_map = -1;
+static int select_by_skb_data_prog;
+static int saved_tcp_syncookie;
+static struct bpf_object *obj;
+static int saved_tcp_fo;
+static __u32 index_zero;
+static int epfd;
+
+static union sa46 {
+	struct sockaddr_in6 v6;
+	struct sockaddr_in v4;
+	sa_family_t family;
+} srv_sa;
+
+#define CHECK(condition, tag, format...) ({				\
+	int __ret = !!(condition);					\
+	if (__ret) {							\
+		printf("%s(%d):FAIL:%s ", __func__, __LINE__, tag);	\
+		printf(format);						\
+		exit(-1);						\
+	}								\
+})
+
+#define RET_ERR(condition, tag, format...) ({				\
+	int __ret = !!(condition);					\
+	if (__ret) {							\
+		printf("%s(%d):FAIL:%s ", __func__, __LINE__, tag);	\
+		printf(format);						\
+		return -1;						\
+	}								\
+})
+
+static int create_maps(void)
+{
+	struct bpf_create_map_attr attr = {};
+
+	/* Creating reuseport_array */
+	attr.name = "reuseport_array";
+	attr.map_type = BPF_MAP_TYPE_REUSEPORT_SOCKARRAY;
+	attr.key_size = sizeof(__u32);
+	attr.value_size = sizeof(__u32);
+	attr.max_entries = REUSEPORT_ARRAY_SIZE;
+
+	reuseport_array = bpf_create_map_xattr(&attr);
+	RET_ERR(reuseport_array == -1, "creating reuseport_array",
+		"reuseport_array:%d errno:%d\n", reuseport_array, errno);
+
+	/* Creating outer_map */
+	attr.name = "outer_map";
+	attr.map_type = BPF_MAP_TYPE_ARRAY_OF_MAPS;
+	attr.key_size = sizeof(__u32);
+	attr.value_size = sizeof(__u32);
+	attr.max_entries = 1;
+	attr.inner_map_fd = reuseport_array;
+	outer_map = bpf_create_map_xattr(&attr);
+	RET_ERR(outer_map == -1, "creating outer_map",
+		"outer_map:%d errno:%d\n", outer_map, errno);
+
+	return 0;
+}
+
+static int prepare_bpf_obj(void)
+{
+	struct bpf_program *prog;
+	struct bpf_map *map;
+	int err;
+
+	obj = bpf_object__open("test_select_reuseport_kern.o");
+	RET_ERR(IS_ERR_OR_NULL(obj), "open test_select_reuseport_kern.o",
+		"obj:%p PTR_ERR(obj):%ld\n", obj, PTR_ERR(obj));
+
+	map = bpf_object__find_map_by_name(obj, "outer_map");
+	RET_ERR(!map, "find outer_map", "!map\n");
+	err = bpf_map__reuse_fd(map, outer_map);
+	RET_ERR(err, "reuse outer_map", "err:%d\n", err);
+
+	err = bpf_object__load(obj);
+	RET_ERR(err, "load bpf_object", "err:%d\n", err);
+
+	prog = bpf_program__next(NULL, obj);
+	RET_ERR(!prog, "get first bpf_program", "!prog\n");
+	select_by_skb_data_prog = bpf_program__fd(prog);
+	RET_ERR(select_by_skb_data_prog == -1, "get prog fd",
+		"select_by_skb_data_prog:%d\n", select_by_skb_data_prog);
+
+	map = bpf_object__find_map_by_name(obj, "result_map");
+	RET_ERR(!map, "find result_map", "!map\n");
+	result_map = bpf_map__fd(map);
+	RET_ERR(result_map == -1, "get result_map fd",
+		"result_map:%d\n", result_map);
+
+	map = bpf_object__find_map_by_name(obj, "tmp_index_ovr_map");
+	RET_ERR(!map, "find tmp_index_ovr_map\n", "!map");
+	tmp_index_ovr_map = bpf_map__fd(map);
+	RET_ERR(tmp_index_ovr_map == -1, "get tmp_index_ovr_map fd",
+		"tmp_index_ovr_map:%d\n", tmp_index_ovr_map);
+
+	map = bpf_object__find_map_by_name(obj, "linum_map");
+	RET_ERR(!map, "find linum_map", "!map\n");
+	linum_map = bpf_map__fd(map);
+	RET_ERR(linum_map == -1, "get linum_map fd",
+		"linum_map:%d\n", linum_map);
+
+	map = bpf_object__find_map_by_name(obj, "data_check_map");
+	RET_ERR(!map, "find data_check_map", "!map\n");
+	data_check_map = bpf_map__fd(map);
+	RET_ERR(data_check_map == -1, "get data_check_map fd",
+		"data_check_map:%d\n", data_check_map);
+
+	return 0;
+}
+
+static void sa46_init_loopback(union sa46 *sa, sa_family_t family)
+{
+	memset(sa, 0, sizeof(*sa));
+	sa->family = family;
+	if (sa->family == AF_INET6)
+		sa->v6.sin6_addr = in6addr_loopback;
+	else
+		sa->v4.sin_addr.s_addr = htonl(INADDR_LOOPBACK);
+}
+
+static void sa46_init_inany(union sa46 *sa, sa_family_t family)
+{
+	memset(sa, 0, sizeof(*sa));
+	sa->family = family;
+	if (sa->family == AF_INET6)
+		sa->v6.sin6_addr = in6addr_any;
+	else
+		sa->v4.sin_addr.s_addr = INADDR_ANY;
+}
+
+static int read_int_sysctl(const char *sysctl)
+{
+	char buf[16];
+	int fd, ret;
+
+	fd = open(sysctl, 0);
+	RET_ERR(fd == -1, "open(sysctl)",
+		"sysctl:%s fd:%d errno:%d\n", sysctl, fd, errno);
+
+	ret = read(fd, buf, sizeof(buf));
+	RET_ERR(ret <= 0, "read(sysctl)",
+		"sysctl:%s ret:%d errno:%d\n", sysctl, ret, errno);
+
+	close(fd);
+	return atoi(buf);
+}
+
+static int write_int_sysctl(const char *sysctl, int v)
+{
+	int fd, ret, size;
+	char buf[16];
+
+	fd = open(sysctl, O_RDWR);
+	RET_ERR(fd == -1, "open(sysctl)",
+		"sysctl:%s fd:%d errno:%d\n", sysctl, fd, errno);
+
+	size = snprintf(buf, sizeof(buf), "%d", v);
+	ret = write(fd, buf, size);
+	RET_ERR(ret != size, "write(sysctl)",
+		"sysctl:%s ret:%d size:%d errno:%d\n",
+		sysctl, ret, size, errno);
+
+	close(fd);
+	return 0;
+}
+
+static void restore_sysctls(void)
+{
+	write_int_sysctl(TCP_FO_SYSCTL, saved_tcp_fo);
+	write_int_sysctl(TCP_SYNCOOKIE_SYSCTL, saved_tcp_syncookie);
+}
+
+static int enable_fastopen(void)
+{
+	int fo;
+
+	fo = read_int_sysctl(TCP_FO_SYSCTL);
+	if (fo < 0)
+		return -1;
+
+	return write_int_sysctl(TCP_FO_SYSCTL, fo | 7);
+}
+
+static int enable_syncookie(void)
+{
+	return write_int_sysctl(TCP_SYNCOOKIE_SYSCTL, 2);
+}
+
+static int disable_syncookie(void)
+{
+	return write_int_sysctl(TCP_SYNCOOKIE_SYSCTL, 0);
+}
+
+static __u32 get_linum(void)
+{
+	__u32 linum;
+	int err;
+
+	err = bpf_map_lookup_elem(linum_map, &index_zero, &linum);
+	CHECK(err == -1, "lookup_elem(linum_map)", "err:%d errno:%d\n",
+	      err, errno);
+
+	return linum;
+}
+
+static void check_data(int type, sa_family_t family, const struct cmd *cmd,
+		       int cli_fd)
+{
+	struct data_check expected = {}, result;
+	union sa46 cli_sa;
+	socklen_t addrlen;
+	int err;
+
+	addrlen = sizeof(cli_sa);
+	err = getsockname(cli_fd, (struct sockaddr *)&cli_sa,
+			  &addrlen);
+	CHECK(err == -1, "getsockname(cli_fd)", "err:%d errno:%d\n",
+	      err, errno);
+
+	err = bpf_map_lookup_elem(data_check_map, &index_zero, &result);
+	CHECK(err == -1, "lookup_elem(data_check_map)", "err:%d errno:%d\n",
+	      err, errno);
+
+	if (type == SOCK_STREAM) {
+		expected.len = MIN_TCPHDR_LEN;
+		expected.ip_protocol = IPPROTO_TCP;
+	} else {
+		expected.len = UDPHDR_LEN;
+		expected.ip_protocol = IPPROTO_UDP;
+	}
+
+	if (family == AF_INET6) {
+		expected.eth_protocol = htons(ETH_P_IPV6);
+		expected.bind_inany = !srv_sa.v6.sin6_addr.s6_addr32[3] &&
+			!srv_sa.v6.sin6_addr.s6_addr32[2] &&
+			!srv_sa.v6.sin6_addr.s6_addr32[1] &&
+			!srv_sa.v6.sin6_addr.s6_addr32[0];
+
+		memcpy(&expected.skb_addrs[0], cli_sa.v6.sin6_addr.s6_addr32,
+		       sizeof(cli_sa.v6.sin6_addr));
+		memcpy(&expected.skb_addrs[4], &in6addr_loopback,
+		       sizeof(in6addr_loopback));
+		expected.skb_ports[0] = cli_sa.v6.sin6_port;
+		expected.skb_ports[1] = srv_sa.v6.sin6_port;
+	} else {
+		expected.eth_protocol = htons(ETH_P_IP);
+		expected.bind_inany = !srv_sa.v4.sin_addr.s_addr;
+
+		expected.skb_addrs[0] = cli_sa.v4.sin_addr.s_addr;
+		expected.skb_addrs[1] = htonl(INADDR_LOOPBACK);
+		expected.skb_ports[0] = cli_sa.v4.sin_port;
+		expected.skb_ports[1] = srv_sa.v4.sin_port;
+	}
+
+	if (memcmp(&result, &expected, offsetof(struct data_check,
+						equal_check_end))) {
+		printf("unexpected data_check\n");
+		printf("  result: (0x%x, %u, %u)\n",
+		       result.eth_protocol, result.ip_protocol,
+		       result.bind_inany);
+		printf("expected: (0x%x, %u, %u)\n",
+		       expected.eth_protocol, expected.ip_protocol,
+		       expected.bind_inany);
+		CHECK(1, "data_check result != expected",
+		      "bpf_prog_linum:%u\n", get_linum());
+	}
+
+	CHECK(!result.hash, "data_check result.hash empty",
+	      "result.hash:%u", result.hash);
+
+	expected.len += cmd ? sizeof(*cmd) : 0;
+	if (type == SOCK_STREAM)
+		CHECK(expected.len > result.len, "expected.len > result.len",
+		      "expected.len:%u result.len:%u bpf_prog_linum:%u\n",
+		      expected.len, result.len, get_linum());
+	else
+		CHECK(expected.len != result.len, "expected.len != result.len",
+		      "expected.len:%u result.len:%u bpf_prog_linum:%u\n",
+		      expected.len, result.len, get_linum());
+}
+
+static void check_results(void)
+{
+	__u32 results[NR_RESULTS];
+	__u32 i, broken = 0;
+	int err;
+
+	for (i = 0; i < NR_RESULTS; i++) {
+		err = bpf_map_lookup_elem(result_map, &i, &results[i]);
+		CHECK(err == -1, "lookup_elem(result_map)",
+		      "i:%u err:%d errno:%d\n", i, err, errno);
+	}
+
+	for (i = 0; i < NR_RESULTS; i++) {
+		if (results[i] != expected_results[i]) {
+			broken = i;
+			break;
+		}
+	}
+
+	if (i == NR_RESULTS)
+		return;
+
+	printf("unexpected result\n");
+	printf(" result: [");
+	printf("%u", results[0]);
+	for (i = 1; i < NR_RESULTS; i++)
+		printf(", %u", results[i]);
+	printf("]\n");
+
+	printf("expected: [");
+	printf("%u", expected_results[0]);
+	for (i = 1; i < NR_RESULTS; i++)
+		printf(", %u", expected_results[i]);
+	printf("]\n");
+
+	CHECK(expected_results[broken] != results[broken],
+	      "unexpected result",
+	      "expected_results[%u] != results[%u] bpf_prog_linum:%u\n",
+	      broken, broken, get_linum());
+}
+
+static int send_data(int type, sa_family_t family, void *data, size_t len,
+		     enum result expected)
+{
+	union sa46 cli_sa;
+	int fd, err;
+
+	fd = socket(family, type, 0);
+	CHECK(fd == -1, "socket()", "fd:%d errno:%d\n", fd, errno);
+
+	sa46_init_loopback(&cli_sa, family);
+	err = bind(fd, (struct sockaddr *)&cli_sa, sizeof(cli_sa));
+	CHECK(fd == -1, "bind(cli_sa)", "err:%d errno:%d\n", err, errno);
+
+	err = sendto(fd, data, len, MSG_FASTOPEN, (struct sockaddr *)&srv_sa,
+		     sizeof(srv_sa));
+	CHECK(err != len && expected >= PASS,
+	      "sendto()", "family:%u err:%d errno:%d expected:%d\n",
+	      family, err, errno, expected);
+
+	return fd;
+}
+
+static void do_test(int type, sa_family_t family, struct cmd *cmd,
+		    enum result expected)
+{
+	int nev, srv_fd, cli_fd;
+	struct epoll_event ev;
+	struct cmd rcv_cmd;
+	ssize_t nread;
+
+	cli_fd = send_data(type, family, cmd, cmd ? sizeof(*cmd) : 0,
+			   expected);
+	nev = epoll_wait(epfd, &ev, 1, expected >= PASS ? 5 : 0);
+	CHECK((nev <= 0 && expected >= PASS) ||
+	      (nev > 0 && expected < PASS),
+	      "nev <> expected",
+	      "nev:%d expected:%d type:%d family:%d data:(%d, %d)\n",
+	      nev, expected, type, family,
+	      cmd ? cmd->reuseport_index : -1,
+	      cmd ? cmd->pass_on_failure : -1);
+	check_results();
+	check_data(type, family, cmd, cli_fd);
+
+	if (expected < PASS)
+		return;
+
+	CHECK(expected != PASS_ERR_SK_SELECT_REUSEPORT &&
+	      cmd->reuseport_index != ev.data.u32,
+	      "check cmd->reuseport_index",
+	      "cmd:(%u, %u) ev.data.u32:%u\n",
+	      cmd->pass_on_failure, cmd->reuseport_index, ev.data.u32);
+
+	srv_fd = sk_fds[ev.data.u32];
+	if (type == SOCK_STREAM) {
+		int new_fd = accept(srv_fd, NULL, 0);
+
+		CHECK(new_fd == -1, "accept(srv_fd)",
+		      "ev.data.u32:%u new_fd:%d errno:%d\n",
+		      ev.data.u32, new_fd, errno);
+
+		nread = recv(new_fd, &rcv_cmd, sizeof(rcv_cmd), MSG_DONTWAIT);
+		CHECK(nread != sizeof(rcv_cmd),
+		      "recv(new_fd)",
+		      "ev.data.u32:%u nread:%zd sizeof(rcv_cmd):%zu errno:%d\n",
+		      ev.data.u32, nread, sizeof(rcv_cmd), errno);
+
+		close(new_fd);
+	} else {
+		nread = recv(srv_fd, &rcv_cmd, sizeof(rcv_cmd), MSG_DONTWAIT);
+		CHECK(nread != sizeof(rcv_cmd),
+		      "recv(sk_fds)",
+		      "ev.data.u32:%u nread:%zd sizeof(rcv_cmd):%zu errno:%d\n",
+		      ev.data.u32, nread, sizeof(rcv_cmd), errno);
+	}
+
+	close(cli_fd);
+}
+
+static void test_err_inner_map(int type, sa_family_t family)
+{
+	struct cmd cmd = {
+		.reuseport_index = 0,
+		.pass_on_failure = 0,
+	};
+
+	expected_results[DROP_ERR_INNER_MAP]++;
+	do_test(type, family, &cmd, DROP_ERR_INNER_MAP);
+	printf("OK\n");
+}
+
+static void test_err_skb_data(int type, sa_family_t family)
+{
+	expected_results[DROP_ERR_SKB_DATA]++;
+	do_test(type, family, NULL, DROP_ERR_SKB_DATA);
+	printf("OK\n");
+}
+
+static void test_err_sk_select_port(int type, sa_family_t family)
+{
+	struct cmd cmd = {
+		.reuseport_index = REUSEPORT_ARRAY_SIZE,
+		.pass_on_failure = 0,
+	};
+
+	expected_results[DROP_ERR_SK_SELECT_REUSEPORT]++;
+	do_test(type, family, &cmd, DROP_ERR_SK_SELECT_REUSEPORT);
+	printf("OK\n");
+}
+
+static void test_pass(int type, sa_family_t family)
+{
+	struct cmd cmd;
+	int i;
+
+	cmd.pass_on_failure = 0;
+	for (i = 0; i < REUSEPORT_ARRAY_SIZE; i++) {
+		expected_results[PASS]++;
+		cmd.reuseport_index = i;
+		do_test(type, family, &cmd, PASS);
+	}
+	printf("OK\n");
+}
+
+static void test_syncookie(int type, sa_family_t family)
+{
+	int err, tmp_index = 1;
+	struct cmd cmd = {
+		.reuseport_index = 0,
+		.pass_on_failure = 0,
+	};
+
+	if (type != SOCK_STREAM)
+		return;
+
+	/*
+	 * +1 for TCP-SYN and
+	 * +1 for the TCP-ACK (ack the syncookie)
+	 */
+	expected_results[PASS] += 2;
+	enable_syncookie();
+	/*
+	 * Simulate TCP-SYN and TCP-ACK are handled by two different sk:
+	 * TCP-SYN: select sk_fds[tmp_index = 1] tmp_index is from the
+	 *          tmp_index_ovr_map
+	 * TCP-ACK: select sk_fds[reuseport_index = 0] reuseport_index
+	 *          is from the cmd.reuseport_index
+	 */
+	err = bpf_map_update_elem(tmp_index_ovr_map, &index_zero,
+				  &tmp_index, BPF_ANY);
+	CHECK(err == -1, "update_elem(tmp_index_ovr_map, 0, 1)",
+	      "err:%d errno:%d\n", err, errno);
+	do_test(type, family, &cmd, PASS);
+	err = bpf_map_lookup_elem(tmp_index_ovr_map, &index_zero,
+				  &tmp_index);
+	CHECK(err == -1 || tmp_index != -1,
+	      "lookup_elem(tmp_index_ovr_map)",
+	      "err:%d errno:%d tmp_index:%d\n",
+	      err, errno, tmp_index);
+	disable_syncookie();
+	printf("OK\n");
+}
+
+static void test_pass_on_err(int type, sa_family_t family)
+{
+	struct cmd cmd = {
+		.reuseport_index = REUSEPORT_ARRAY_SIZE,
+		.pass_on_failure = 1,
+	};
+
+	expected_results[PASS_ERR_SK_SELECT_REUSEPORT] += 1;
+	do_test(type, family, &cmd, PASS_ERR_SK_SELECT_REUSEPORT);
+	printf("OK\n");
+}
+
+static void test_detach_bpf(int type, sa_family_t family)
+{
+#ifdef SO_DETACH_REUSEPORT_BPF
+	__u32 nr_run_before = 0, nr_run_after = 0, tmp, i;
+	struct epoll_event ev;
+	int cli_fd, err, nev;
+	struct cmd cmd = {};
+	int optvalue = 0;
+
+	err = setsockopt(sk_fds[0], SOL_SOCKET, SO_DETACH_REUSEPORT_BPF,
+			 &optvalue, sizeof(optvalue));
+	CHECK(err == -1, "setsockopt(SO_DETACH_REUSEPORT_BPF)",
+	      "err:%d errno:%d\n", err, errno);
+
+	err = setsockopt(sk_fds[1], SOL_SOCKET, SO_DETACH_REUSEPORT_BPF,
+			 &optvalue, sizeof(optvalue));
+	CHECK(err == 0 || errno != ENOENT, "setsockopt(SO_DETACH_REUSEPORT_BPF)",
+	      "err:%d errno:%d\n", err, errno);
+
+	for (i = 0; i < NR_RESULTS; i++) {
+		err = bpf_map_lookup_elem(result_map, &i, &tmp);
+		CHECK(err == -1, "lookup_elem(result_map)",
+		      "i:%u err:%d errno:%d\n", i, err, errno);
+		nr_run_before += tmp;
+	}
+
+	cli_fd = send_data(type, family, &cmd, sizeof(cmd), PASS);
+	nev = epoll_wait(epfd, &ev, 1, 5);
+	CHECK(nev <= 0, "nev <= 0",
+	      "nev:%d expected:1 type:%d family:%d data:(0, 0)\n",
+	      nev,  type, family);
+
+	for (i = 0; i < NR_RESULTS; i++) {
+		err = bpf_map_lookup_elem(result_map, &i, &tmp);
+		CHECK(err == -1, "lookup_elem(result_map)",
+		      "i:%u err:%d errno:%d\n", i, err, errno);
+		nr_run_after += tmp;
+	}
+
+	CHECK(nr_run_before != nr_run_after,
+	      "nr_run_before != nr_run_after",
+	      "nr_run_before:%u nr_run_after:%u\n",
+	      nr_run_before, nr_run_after);
+
+	printf("OK\n");
+	close(cli_fd);
+#else
+	printf("SKIP\n");
+#endif
+}
+
+static void prepare_sk_fds(int type, sa_family_t family, bool inany)
+{
+	const int first = REUSEPORT_ARRAY_SIZE - 1;
+	int i, err, optval = 1;
+	struct epoll_event ev;
+	socklen_t addrlen;
+
+	if (inany)
+		sa46_init_inany(&srv_sa, family);
+	else
+		sa46_init_loopback(&srv_sa, family);
+	addrlen = sizeof(srv_sa);
+
+	/*
+	 * The sk_fds[] is filled from the back such that the order
+	 * is exactly opposite to the (struct sock_reuseport *)reuse->socks[].
+	 */
+	for (i = first; i >= 0; i--) {
+		sk_fds[i] = socket(family, type, 0);
+		CHECK(sk_fds[i] == -1, "socket()", "sk_fds[%d]:%d errno:%d\n",
+		      i, sk_fds[i], errno);
+		err = setsockopt(sk_fds[i], SOL_SOCKET, SO_REUSEPORT,
+				 &optval, sizeof(optval));
+		CHECK(err == -1, "setsockopt(SO_REUSEPORT)",
+		      "sk_fds[%d] err:%d errno:%d\n",
+		      i, err, errno);
+
+		if (i == first) {
+			err = setsockopt(sk_fds[i], SOL_SOCKET,
+					 SO_ATTACH_REUSEPORT_EBPF,
+					 &select_by_skb_data_prog,
+					 sizeof(select_by_skb_data_prog));
+			CHECK(err == -1, "setsockopt(SO_ATTACH_REUEPORT_EBPF)",
+			      "err:%d errno:%d\n", err, errno);
+		}
+
+		err = bind(sk_fds[i], (struct sockaddr *)&srv_sa, addrlen);
+		CHECK(err == -1, "bind()", "sk_fds[%d] err:%d errno:%d\n",
+		      i, err, errno);
+
+		if (type == SOCK_STREAM) {
+			err = listen(sk_fds[i], 10);
+			CHECK(err == -1, "listen()",
+			      "sk_fds[%d] err:%d errno:%d\n",
+			      i, err, errno);
+		}
+
+		err = bpf_map_update_elem(reuseport_array, &i, &sk_fds[i],
+					  BPF_NOEXIST);
+		CHECK(err == -1, "update_elem(reuseport_array)",
+		      "sk_fds[%d] err:%d errno:%d\n", i, err, errno);
+
+		if (i == first) {
+			socklen_t addrlen = sizeof(srv_sa);
+
+			err = getsockname(sk_fds[i], (struct sockaddr *)&srv_sa,
+					  &addrlen);
+			CHECK(err == -1, "getsockname()",
+			      "sk_fds[%d] err:%d errno:%d\n", i, err, errno);
+		}
+	}
+
+	epfd = epoll_create(1);
+	CHECK(epfd == -1, "epoll_create(1)",
+	      "epfd:%d errno:%d\n", epfd, errno);
+
+	ev.events = EPOLLIN;
+	for (i = 0; i < REUSEPORT_ARRAY_SIZE; i++) {
+		ev.data.u32 = i;
+		err = epoll_ctl(epfd, EPOLL_CTL_ADD, sk_fds[i], &ev);
+		CHECK(err, "epoll_ctl(EPOLL_CTL_ADD)", "sk_fds[%d]\n", i);
+	}
+}
+
+static void setup_per_test(int type, sa_family_t family, bool inany,
+			   bool no_inner_map)
+{
+	int ovr = -1, err;
+
+	prepare_sk_fds(type, family, inany);
+	err = bpf_map_update_elem(tmp_index_ovr_map, &index_zero, &ovr,
+				  BPF_ANY);
+	CHECK(err == -1, "update_elem(tmp_index_ovr_map, 0, -1)",
+	      "err:%d errno:%d\n", err, errno);
+
+	/* Install reuseport_array to outer_map? */
+	if (no_inner_map)
+		return;
+
+	err = bpf_map_update_elem(outer_map, &index_zero, &reuseport_array,
+				  BPF_ANY);
+	CHECK(err == -1, "update_elem(outer_map, 0, reuseport_array)",
+	      "err:%d errno:%d\n", err, errno);
+}
+
+static void cleanup_per_test(bool no_inner_map)
+{
+	int i, err;
+
+	for (i = 0; i < REUSEPORT_ARRAY_SIZE; i++)
+		close(sk_fds[i]);
+	close(epfd);
+
+	/* Delete reuseport_array from outer_map? */
+	if (no_inner_map)
+		return;
+
+	err = bpf_map_delete_elem(outer_map, &index_zero);
+	CHECK(err == -1, "delete_elem(outer_map)",
+	      "err:%d errno:%d\n", err, errno);
+}
+
+static void cleanup(void)
+{
+	if (outer_map != -1)
+		close(outer_map);
+	if (reuseport_array != -1)
+		close(reuseport_array);
+	if (obj)
+		bpf_object__close(obj);
+}
+
+static const char *family_str(sa_family_t family)
+{
+	switch (family) {
+	case AF_INET:
+		return "IPv4";
+	case AF_INET6:
+		return "IPv6";
+	default:
+		return "unknown";
+	}
+}
+
+static const char *sotype_str(int sotype)
+{
+	switch (sotype) {
+	case SOCK_STREAM:
+		return "TCP";
+	case SOCK_DGRAM:
+		return "UDP";
+	default:
+		return "unknown";
+	}
+}
+
+#define TEST_INIT(fn, ...) { fn, #fn, __VA_ARGS__ }
+
+static void test_config(int type, sa_family_t family, bool inany)
+{
+	const struct test {
+		void (*fn)(int sotype, sa_family_t family);
+		const char *name;
+		bool no_inner_map;
+	} tests[] = {
+		TEST_INIT(test_err_inner_map, true /* no_inner_map */),
+		TEST_INIT(test_err_skb_data),
+		TEST_INIT(test_err_sk_select_port),
+		TEST_INIT(test_pass),
+		TEST_INIT(test_syncookie),
+		TEST_INIT(test_pass_on_err),
+		TEST_INIT(test_detach_bpf),
+	};
+	const struct test *t;
+
+	printf("######## %s/%s %s ########\n",
+	       family_str(family), sotype_str(type),
+	       inany ? " INANY  " : "LOOPBACK");
+
+	for (t = tests; t < tests + ARRAY_SIZE(tests); t++) {
+		setup_per_test(type, family, inany, t->no_inner_map);
+		printf("%s: ", t->name);
+		t->fn(type, family);
+		cleanup_per_test(t->no_inner_map);
+	}
+
+	printf("\n");
+}
+
+#define BIND_INANY true
+
+static void test_all(void)
+{
+	const struct config {
+		int sotype;
+		sa_family_t family;
+		bool inany;
+	} configs[] = {
+		{ SOCK_STREAM, AF_INET },
+		{ SOCK_STREAM, AF_INET, BIND_INANY },
+		{ SOCK_STREAM, AF_INET6 },
+		{ SOCK_STREAM, AF_INET6, BIND_INANY },
+		{ SOCK_DGRAM, AF_INET },
+		{ SOCK_DGRAM, AF_INET6 },
+	};
+	const struct config *c;
+
+	for (c = configs; c < configs + ARRAY_SIZE(configs); c++)
+		test_config(c->sotype, c->family, c->inany);
+}
+
+int main(int argc, const char **argv)
+{
+	int ret = EXIT_FAILURE;
+
+	if (create_maps())
+		goto out;
+	if (prepare_bpf_obj())
+		goto out;
+
+	saved_tcp_fo = read_int_sysctl(TCP_FO_SYSCTL);
+	saved_tcp_syncookie = read_int_sysctl(TCP_SYNCOOKIE_SYSCTL);
+	if (saved_tcp_syncookie < 0 || saved_tcp_syncookie < 0)
+		goto out;
+	atexit(restore_sysctls);
+
+	if (enable_fastopen())
+		goto out;
+	if (disable_syncookie())
+		goto out;
+
+	test_all();
+
+	ret = EXIT_SUCCESS;
+out:
+	cleanup();
+	return ret;
+}
diff --git a/tools/testing/selftests/bpf/test_select_reuseport.c b/tools/testing/selftests/bpf/test_select_reuseport.c
deleted file mode 100644
index 0d5687feb689..000000000000
--- a/tools/testing/selftests/bpf/test_select_reuseport.c
+++ /dev/null
@@ -1,811 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-/* Copyright (c) 2018 Facebook */
-
-#include <stdlib.h>
-#include <unistd.h>
-#include <stdbool.h>
-#include <string.h>
-#include <errno.h>
-#include <assert.h>
-#include <fcntl.h>
-#include <linux/bpf.h>
-#include <linux/err.h>
-#include <linux/types.h>
-#include <linux/if_ether.h>
-#include <sys/types.h>
-#include <sys/epoll.h>
-#include <sys/socket.h>
-#include <netinet/in.h>
-#include <bpf/bpf.h>
-#include <bpf/libbpf.h>
-#include "bpf_rlimit.h"
-#include "bpf_util.h"
-#include "test_select_reuseport_common.h"
-
-#define MIN_TCPHDR_LEN 20
-#define UDPHDR_LEN 8
-
-#define TCP_SYNCOOKIE_SYSCTL "/proc/sys/net/ipv4/tcp_syncookies"
-#define TCP_FO_SYSCTL "/proc/sys/net/ipv4/tcp_fastopen"
-#define REUSEPORT_ARRAY_SIZE 32
-
-static int result_map, tmp_index_ovr_map, linum_map, data_check_map;
-static enum result expected_results[NR_RESULTS];
-static int sk_fds[REUSEPORT_ARRAY_SIZE];
-static int reuseport_array = -1, outer_map = -1;
-static int select_by_skb_data_prog;
-static int saved_tcp_syncookie;
-static struct bpf_object *obj;
-static int saved_tcp_fo;
-static __u32 index_zero;
-static int epfd;
-
-static union sa46 {
-	struct sockaddr_in6 v6;
-	struct sockaddr_in v4;
-	sa_family_t family;
-} srv_sa;
-
-#define CHECK(condition, tag, format...) ({				\
-	int __ret = !!(condition);					\
-	if (__ret) {							\
-		printf("%s(%d):FAIL:%s ", __func__, __LINE__, tag);	\
-		printf(format);						\
-		exit(-1);						\
-	}								\
-})
-
-#define RET_ERR(condition, tag, format...) ({				\
-	int __ret = !!(condition);					\
-	if (__ret) {							\
-		printf("%s(%d):FAIL:%s ", __func__, __LINE__, tag);	\
-		printf(format);						\
-		return -1;						\
-	}								\
-})
-
-static int create_maps(void)
-{
-	struct bpf_create_map_attr attr = {};
-
-	/* Creating reuseport_array */
-	attr.name = "reuseport_array";
-	attr.map_type = BPF_MAP_TYPE_REUSEPORT_SOCKARRAY;
-	attr.key_size = sizeof(__u32);
-	attr.value_size = sizeof(__u32);
-	attr.max_entries = REUSEPORT_ARRAY_SIZE;
-
-	reuseport_array = bpf_create_map_xattr(&attr);
-	RET_ERR(reuseport_array == -1, "creating reuseport_array",
-		"reuseport_array:%d errno:%d\n", reuseport_array, errno);
-
-	/* Creating outer_map */
-	attr.name = "outer_map";
-	attr.map_type = BPF_MAP_TYPE_ARRAY_OF_MAPS;
-	attr.key_size = sizeof(__u32);
-	attr.value_size = sizeof(__u32);
-	attr.max_entries = 1;
-	attr.inner_map_fd = reuseport_array;
-	outer_map = bpf_create_map_xattr(&attr);
-	RET_ERR(outer_map == -1, "creating outer_map",
-		"outer_map:%d errno:%d\n", outer_map, errno);
-
-	return 0;
-}
-
-static int prepare_bpf_obj(void)
-{
-	struct bpf_program *prog;
-	struct bpf_map *map;
-	int err;
-
-	obj = bpf_object__open("test_select_reuseport_kern.o");
-	RET_ERR(IS_ERR_OR_NULL(obj), "open test_select_reuseport_kern.o",
-		"obj:%p PTR_ERR(obj):%ld\n", obj, PTR_ERR(obj));
-
-	map = bpf_object__find_map_by_name(obj, "outer_map");
-	RET_ERR(!map, "find outer_map", "!map\n");
-	err = bpf_map__reuse_fd(map, outer_map);
-	RET_ERR(err, "reuse outer_map", "err:%d\n", err);
-
-	err = bpf_object__load(obj);
-	RET_ERR(err, "load bpf_object", "err:%d\n", err);
-
-	prog = bpf_program__next(NULL, obj);
-	RET_ERR(!prog, "get first bpf_program", "!prog\n");
-	select_by_skb_data_prog = bpf_program__fd(prog);
-	RET_ERR(select_by_skb_data_prog == -1, "get prog fd",
-		"select_by_skb_data_prog:%d\n", select_by_skb_data_prog);
-
-	map = bpf_object__find_map_by_name(obj, "result_map");
-	RET_ERR(!map, "find result_map", "!map\n");
-	result_map = bpf_map__fd(map);
-	RET_ERR(result_map == -1, "get result_map fd",
-		"result_map:%d\n", result_map);
-
-	map = bpf_object__find_map_by_name(obj, "tmp_index_ovr_map");
-	RET_ERR(!map, "find tmp_index_ovr_map\n", "!map");
-	tmp_index_ovr_map = bpf_map__fd(map);
-	RET_ERR(tmp_index_ovr_map == -1, "get tmp_index_ovr_map fd",
-		"tmp_index_ovr_map:%d\n", tmp_index_ovr_map);
-
-	map = bpf_object__find_map_by_name(obj, "linum_map");
-	RET_ERR(!map, "find linum_map", "!map\n");
-	linum_map = bpf_map__fd(map);
-	RET_ERR(linum_map == -1, "get linum_map fd",
-		"linum_map:%d\n", linum_map);
-
-	map = bpf_object__find_map_by_name(obj, "data_check_map");
-	RET_ERR(!map, "find data_check_map", "!map\n");
-	data_check_map = bpf_map__fd(map);
-	RET_ERR(data_check_map == -1, "get data_check_map fd",
-		"data_check_map:%d\n", data_check_map);
-
-	return 0;
-}
-
-static void sa46_init_loopback(union sa46 *sa, sa_family_t family)
-{
-	memset(sa, 0, sizeof(*sa));
-	sa->family = family;
-	if (sa->family == AF_INET6)
-		sa->v6.sin6_addr = in6addr_loopback;
-	else
-		sa->v4.sin_addr.s_addr = htonl(INADDR_LOOPBACK);
-}
-
-static void sa46_init_inany(union sa46 *sa, sa_family_t family)
-{
-	memset(sa, 0, sizeof(*sa));
-	sa->family = family;
-	if (sa->family == AF_INET6)
-		sa->v6.sin6_addr = in6addr_any;
-	else
-		sa->v4.sin_addr.s_addr = INADDR_ANY;
-}
-
-static int read_int_sysctl(const char *sysctl)
-{
-	char buf[16];
-	int fd, ret;
-
-	fd = open(sysctl, 0);
-	RET_ERR(fd == -1, "open(sysctl)",
-		"sysctl:%s fd:%d errno:%d\n", sysctl, fd, errno);
-
-	ret = read(fd, buf, sizeof(buf));
-	RET_ERR(ret <= 0, "read(sysctl)",
-		"sysctl:%s ret:%d errno:%d\n", sysctl, ret, errno);
-
-	close(fd);
-	return atoi(buf);
-}
-
-static int write_int_sysctl(const char *sysctl, int v)
-{
-	int fd, ret, size;
-	char buf[16];
-
-	fd = open(sysctl, O_RDWR);
-	RET_ERR(fd == -1, "open(sysctl)",
-		"sysctl:%s fd:%d errno:%d\n", sysctl, fd, errno);
-
-	size = snprintf(buf, sizeof(buf), "%d", v);
-	ret = write(fd, buf, size);
-	RET_ERR(ret != size, "write(sysctl)",
-		"sysctl:%s ret:%d size:%d errno:%d\n",
-		sysctl, ret, size, errno);
-
-	close(fd);
-	return 0;
-}
-
-static void restore_sysctls(void)
-{
-	write_int_sysctl(TCP_FO_SYSCTL, saved_tcp_fo);
-	write_int_sysctl(TCP_SYNCOOKIE_SYSCTL, saved_tcp_syncookie);
-}
-
-static int enable_fastopen(void)
-{
-	int fo;
-
-	fo = read_int_sysctl(TCP_FO_SYSCTL);
-	if (fo < 0)
-		return -1;
-
-	return write_int_sysctl(TCP_FO_SYSCTL, fo | 7);
-}
-
-static int enable_syncookie(void)
-{
-	return write_int_sysctl(TCP_SYNCOOKIE_SYSCTL, 2);
-}
-
-static int disable_syncookie(void)
-{
-	return write_int_sysctl(TCP_SYNCOOKIE_SYSCTL, 0);
-}
-
-static __u32 get_linum(void)
-{
-	__u32 linum;
-	int err;
-
-	err = bpf_map_lookup_elem(linum_map, &index_zero, &linum);
-	CHECK(err == -1, "lookup_elem(linum_map)", "err:%d errno:%d\n",
-	      err, errno);
-
-	return linum;
-}
-
-static void check_data(int type, sa_family_t family, const struct cmd *cmd,
-		       int cli_fd)
-{
-	struct data_check expected = {}, result;
-	union sa46 cli_sa;
-	socklen_t addrlen;
-	int err;
-
-	addrlen = sizeof(cli_sa);
-	err = getsockname(cli_fd, (struct sockaddr *)&cli_sa,
-			  &addrlen);
-	CHECK(err == -1, "getsockname(cli_fd)", "err:%d errno:%d\n",
-	      err, errno);
-
-	err = bpf_map_lookup_elem(data_check_map, &index_zero, &result);
-	CHECK(err == -1, "lookup_elem(data_check_map)", "err:%d errno:%d\n",
-	      err, errno);
-
-	if (type == SOCK_STREAM) {
-		expected.len = MIN_TCPHDR_LEN;
-		expected.ip_protocol = IPPROTO_TCP;
-	} else {
-		expected.len = UDPHDR_LEN;
-		expected.ip_protocol = IPPROTO_UDP;
-	}
-
-	if (family == AF_INET6) {
-		expected.eth_protocol = htons(ETH_P_IPV6);
-		expected.bind_inany = !srv_sa.v6.sin6_addr.s6_addr32[3] &&
-			!srv_sa.v6.sin6_addr.s6_addr32[2] &&
-			!srv_sa.v6.sin6_addr.s6_addr32[1] &&
-			!srv_sa.v6.sin6_addr.s6_addr32[0];
-
-		memcpy(&expected.skb_addrs[0], cli_sa.v6.sin6_addr.s6_addr32,
-		       sizeof(cli_sa.v6.sin6_addr));
-		memcpy(&expected.skb_addrs[4], &in6addr_loopback,
-		       sizeof(in6addr_loopback));
-		expected.skb_ports[0] = cli_sa.v6.sin6_port;
-		expected.skb_ports[1] = srv_sa.v6.sin6_port;
-	} else {
-		expected.eth_protocol = htons(ETH_P_IP);
-		expected.bind_inany = !srv_sa.v4.sin_addr.s_addr;
-
-		expected.skb_addrs[0] = cli_sa.v4.sin_addr.s_addr;
-		expected.skb_addrs[1] = htonl(INADDR_LOOPBACK);
-		expected.skb_ports[0] = cli_sa.v4.sin_port;
-		expected.skb_ports[1] = srv_sa.v4.sin_port;
-	}
-
-	if (memcmp(&result, &expected, offsetof(struct data_check,
-						equal_check_end))) {
-		printf("unexpected data_check\n");
-		printf("  result: (0x%x, %u, %u)\n",
-		       result.eth_protocol, result.ip_protocol,
-		       result.bind_inany);
-		printf("expected: (0x%x, %u, %u)\n",
-		       expected.eth_protocol, expected.ip_protocol,
-		       expected.bind_inany);
-		CHECK(1, "data_check result != expected",
-		      "bpf_prog_linum:%u\n", get_linum());
-	}
-
-	CHECK(!result.hash, "data_check result.hash empty",
-	      "result.hash:%u", result.hash);
-
-	expected.len += cmd ? sizeof(*cmd) : 0;
-	if (type == SOCK_STREAM)
-		CHECK(expected.len > result.len, "expected.len > result.len",
-		      "expected.len:%u result.len:%u bpf_prog_linum:%u\n",
-		      expected.len, result.len, get_linum());
-	else
-		CHECK(expected.len != result.len, "expected.len != result.len",
-		      "expected.len:%u result.len:%u bpf_prog_linum:%u\n",
-		      expected.len, result.len, get_linum());
-}
-
-static void check_results(void)
-{
-	__u32 results[NR_RESULTS];
-	__u32 i, broken = 0;
-	int err;
-
-	for (i = 0; i < NR_RESULTS; i++) {
-		err = bpf_map_lookup_elem(result_map, &i, &results[i]);
-		CHECK(err == -1, "lookup_elem(result_map)",
-		      "i:%u err:%d errno:%d\n", i, err, errno);
-	}
-
-	for (i = 0; i < NR_RESULTS; i++) {
-		if (results[i] != expected_results[i]) {
-			broken = i;
-			break;
-		}
-	}
-
-	if (i == NR_RESULTS)
-		return;
-
-	printf("unexpected result\n");
-	printf(" result: [");
-	printf("%u", results[0]);
-	for (i = 1; i < NR_RESULTS; i++)
-		printf(", %u", results[i]);
-	printf("]\n");
-
-	printf("expected: [");
-	printf("%u", expected_results[0]);
-	for (i = 1; i < NR_RESULTS; i++)
-		printf(", %u", expected_results[i]);
-	printf("]\n");
-
-	CHECK(expected_results[broken] != results[broken],
-	      "unexpected result",
-	      "expected_results[%u] != results[%u] bpf_prog_linum:%u\n",
-	      broken, broken, get_linum());
-}
-
-static int send_data(int type, sa_family_t family, void *data, size_t len,
-		     enum result expected)
-{
-	union sa46 cli_sa;
-	int fd, err;
-
-	fd = socket(family, type, 0);
-	CHECK(fd == -1, "socket()", "fd:%d errno:%d\n", fd, errno);
-
-	sa46_init_loopback(&cli_sa, family);
-	err = bind(fd, (struct sockaddr *)&cli_sa, sizeof(cli_sa));
-	CHECK(fd == -1, "bind(cli_sa)", "err:%d errno:%d\n", err, errno);
-
-	err = sendto(fd, data, len, MSG_FASTOPEN, (struct sockaddr *)&srv_sa,
-		     sizeof(srv_sa));
-	CHECK(err != len && expected >= PASS,
-	      "sendto()", "family:%u err:%d errno:%d expected:%d\n",
-	      family, err, errno, expected);
-
-	return fd;
-}
-
-static void do_test(int type, sa_family_t family, struct cmd *cmd,
-		    enum result expected)
-{
-	int nev, srv_fd, cli_fd;
-	struct epoll_event ev;
-	struct cmd rcv_cmd;
-	ssize_t nread;
-
-	cli_fd = send_data(type, family, cmd, cmd ? sizeof(*cmd) : 0,
-			   expected);
-	nev = epoll_wait(epfd, &ev, 1, expected >= PASS ? 5 : 0);
-	CHECK((nev <= 0 && expected >= PASS) ||
-	      (nev > 0 && expected < PASS),
-	      "nev <> expected",
-	      "nev:%d expected:%d type:%d family:%d data:(%d, %d)\n",
-	      nev, expected, type, family,
-	      cmd ? cmd->reuseport_index : -1,
-	      cmd ? cmd->pass_on_failure : -1);
-	check_results();
-	check_data(type, family, cmd, cli_fd);
-
-	if (expected < PASS)
-		return;
-
-	CHECK(expected != PASS_ERR_SK_SELECT_REUSEPORT &&
-	      cmd->reuseport_index != ev.data.u32,
-	      "check cmd->reuseport_index",
-	      "cmd:(%u, %u) ev.data.u32:%u\n",
-	      cmd->pass_on_failure, cmd->reuseport_index, ev.data.u32);
-
-	srv_fd = sk_fds[ev.data.u32];
-	if (type == SOCK_STREAM) {
-		int new_fd = accept(srv_fd, NULL, 0);
-
-		CHECK(new_fd == -1, "accept(srv_fd)",
-		      "ev.data.u32:%u new_fd:%d errno:%d\n",
-		      ev.data.u32, new_fd, errno);
-
-		nread = recv(new_fd, &rcv_cmd, sizeof(rcv_cmd), MSG_DONTWAIT);
-		CHECK(nread != sizeof(rcv_cmd),
-		      "recv(new_fd)",
-		      "ev.data.u32:%u nread:%zd sizeof(rcv_cmd):%zu errno:%d\n",
-		      ev.data.u32, nread, sizeof(rcv_cmd), errno);
-
-		close(new_fd);
-	} else {
-		nread = recv(srv_fd, &rcv_cmd, sizeof(rcv_cmd), MSG_DONTWAIT);
-		CHECK(nread != sizeof(rcv_cmd),
-		      "recv(sk_fds)",
-		      "ev.data.u32:%u nread:%zd sizeof(rcv_cmd):%zu errno:%d\n",
-		      ev.data.u32, nread, sizeof(rcv_cmd), errno);
-	}
-
-	close(cli_fd);
-}
-
-static void test_err_inner_map(int type, sa_family_t family)
-{
-	struct cmd cmd = {
-		.reuseport_index = 0,
-		.pass_on_failure = 0,
-	};
-
-	expected_results[DROP_ERR_INNER_MAP]++;
-	do_test(type, family, &cmd, DROP_ERR_INNER_MAP);
-	printf("OK\n");
-}
-
-static void test_err_skb_data(int type, sa_family_t family)
-{
-	expected_results[DROP_ERR_SKB_DATA]++;
-	do_test(type, family, NULL, DROP_ERR_SKB_DATA);
-	printf("OK\n");
-}
-
-static void test_err_sk_select_port(int type, sa_family_t family)
-{
-	struct cmd cmd = {
-		.reuseport_index = REUSEPORT_ARRAY_SIZE,
-		.pass_on_failure = 0,
-	};
-
-	expected_results[DROP_ERR_SK_SELECT_REUSEPORT]++;
-	do_test(type, family, &cmd, DROP_ERR_SK_SELECT_REUSEPORT);
-	printf("OK\n");
-}
-
-static void test_pass(int type, sa_family_t family)
-{
-	struct cmd cmd;
-	int i;
-
-	cmd.pass_on_failure = 0;
-	for (i = 0; i < REUSEPORT_ARRAY_SIZE; i++) {
-		expected_results[PASS]++;
-		cmd.reuseport_index = i;
-		do_test(type, family, &cmd, PASS);
-	}
-	printf("OK\n");
-}
-
-static void test_syncookie(int type, sa_family_t family)
-{
-	int err, tmp_index = 1;
-	struct cmd cmd = {
-		.reuseport_index = 0,
-		.pass_on_failure = 0,
-	};
-
-	if (type != SOCK_STREAM)
-		return;
-
-	/*
-	 * +1 for TCP-SYN and
-	 * +1 for the TCP-ACK (ack the syncookie)
-	 */
-	expected_results[PASS] += 2;
-	enable_syncookie();
-	/*
-	 * Simulate TCP-SYN and TCP-ACK are handled by two different sk:
-	 * TCP-SYN: select sk_fds[tmp_index = 1] tmp_index is from the
-	 *          tmp_index_ovr_map
-	 * TCP-ACK: select sk_fds[reuseport_index = 0] reuseport_index
-	 *          is from the cmd.reuseport_index
-	 */
-	err = bpf_map_update_elem(tmp_index_ovr_map, &index_zero,
-				  &tmp_index, BPF_ANY);
-	CHECK(err == -1, "update_elem(tmp_index_ovr_map, 0, 1)",
-	      "err:%d errno:%d\n", err, errno);
-	do_test(type, family, &cmd, PASS);
-	err = bpf_map_lookup_elem(tmp_index_ovr_map, &index_zero,
-				  &tmp_index);
-	CHECK(err == -1 || tmp_index != -1,
-	      "lookup_elem(tmp_index_ovr_map)",
-	      "err:%d errno:%d tmp_index:%d\n",
-	      err, errno, tmp_index);
-	disable_syncookie();
-	printf("OK\n");
-}
-
-static void test_pass_on_err(int type, sa_family_t family)
-{
-	struct cmd cmd = {
-		.reuseport_index = REUSEPORT_ARRAY_SIZE,
-		.pass_on_failure = 1,
-	};
-
-	expected_results[PASS_ERR_SK_SELECT_REUSEPORT] += 1;
-	do_test(type, family, &cmd, PASS_ERR_SK_SELECT_REUSEPORT);
-	printf("OK\n");
-}
-
-static void test_detach_bpf(int type, sa_family_t family)
-{
-#ifdef SO_DETACH_REUSEPORT_BPF
-	__u32 nr_run_before = 0, nr_run_after = 0, tmp, i;
-	struct epoll_event ev;
-	int cli_fd, err, nev;
-	struct cmd cmd = {};
-	int optvalue = 0;
-
-	err = setsockopt(sk_fds[0], SOL_SOCKET, SO_DETACH_REUSEPORT_BPF,
-			 &optvalue, sizeof(optvalue));
-	CHECK(err == -1, "setsockopt(SO_DETACH_REUSEPORT_BPF)",
-	      "err:%d errno:%d\n", err, errno);
-
-	err = setsockopt(sk_fds[1], SOL_SOCKET, SO_DETACH_REUSEPORT_BPF,
-			 &optvalue, sizeof(optvalue));
-	CHECK(err == 0 || errno != ENOENT, "setsockopt(SO_DETACH_REUSEPORT_BPF)",
-	      "err:%d errno:%d\n", err, errno);
-
-	for (i = 0; i < NR_RESULTS; i++) {
-		err = bpf_map_lookup_elem(result_map, &i, &tmp);
-		CHECK(err == -1, "lookup_elem(result_map)",
-		      "i:%u err:%d errno:%d\n", i, err, errno);
-		nr_run_before += tmp;
-	}
-
-	cli_fd = send_data(type, family, &cmd, sizeof(cmd), PASS);
-	nev = epoll_wait(epfd, &ev, 1, 5);
-	CHECK(nev <= 0, "nev <= 0",
-	      "nev:%d expected:1 type:%d family:%d data:(0, 0)\n",
-	      nev,  type, family);
-
-	for (i = 0; i < NR_RESULTS; i++) {
-		err = bpf_map_lookup_elem(result_map, &i, &tmp);
-		CHECK(err == -1, "lookup_elem(result_map)",
-		      "i:%u err:%d errno:%d\n", i, err, errno);
-		nr_run_after += tmp;
-	}
-
-	CHECK(nr_run_before != nr_run_after,
-	      "nr_run_before != nr_run_after",
-	      "nr_run_before:%u nr_run_after:%u\n",
-	      nr_run_before, nr_run_after);
-
-	printf("OK\n");
-	close(cli_fd);
-#else
-	printf("SKIP\n");
-#endif
-}
-
-static void prepare_sk_fds(int type, sa_family_t family, bool inany)
-{
-	const int first = REUSEPORT_ARRAY_SIZE - 1;
-	int i, err, optval = 1;
-	struct epoll_event ev;
-	socklen_t addrlen;
-
-	if (inany)
-		sa46_init_inany(&srv_sa, family);
-	else
-		sa46_init_loopback(&srv_sa, family);
-	addrlen = sizeof(srv_sa);
-
-	/*
-	 * The sk_fds[] is filled from the back such that the order
-	 * is exactly opposite to the (struct sock_reuseport *)reuse->socks[].
-	 */
-	for (i = first; i >= 0; i--) {
-		sk_fds[i] = socket(family, type, 0);
-		CHECK(sk_fds[i] == -1, "socket()", "sk_fds[%d]:%d errno:%d\n",
-		      i, sk_fds[i], errno);
-		err = setsockopt(sk_fds[i], SOL_SOCKET, SO_REUSEPORT,
-				 &optval, sizeof(optval));
-		CHECK(err == -1, "setsockopt(SO_REUSEPORT)",
-		      "sk_fds[%d] err:%d errno:%d\n",
-		      i, err, errno);
-
-		if (i == first) {
-			err = setsockopt(sk_fds[i], SOL_SOCKET,
-					 SO_ATTACH_REUSEPORT_EBPF,
-					 &select_by_skb_data_prog,
-					 sizeof(select_by_skb_data_prog));
-			CHECK(err == -1, "setsockopt(SO_ATTACH_REUEPORT_EBPF)",
-			      "err:%d errno:%d\n", err, errno);
-		}
-
-		err = bind(sk_fds[i], (struct sockaddr *)&srv_sa, addrlen);
-		CHECK(err == -1, "bind()", "sk_fds[%d] err:%d errno:%d\n",
-		      i, err, errno);
-
-		if (type == SOCK_STREAM) {
-			err = listen(sk_fds[i], 10);
-			CHECK(err == -1, "listen()",
-			      "sk_fds[%d] err:%d errno:%d\n",
-			      i, err, errno);
-		}
-
-		err = bpf_map_update_elem(reuseport_array, &i, &sk_fds[i],
-					  BPF_NOEXIST);
-		CHECK(err == -1, "update_elem(reuseport_array)",
-		      "sk_fds[%d] err:%d errno:%d\n", i, err, errno);
-
-		if (i == first) {
-			socklen_t addrlen = sizeof(srv_sa);
-
-			err = getsockname(sk_fds[i], (struct sockaddr *)&srv_sa,
-					  &addrlen);
-			CHECK(err == -1, "getsockname()",
-			      "sk_fds[%d] err:%d errno:%d\n", i, err, errno);
-		}
-	}
-
-	epfd = epoll_create(1);
-	CHECK(epfd == -1, "epoll_create(1)",
-	      "epfd:%d errno:%d\n", epfd, errno);
-
-	ev.events = EPOLLIN;
-	for (i = 0; i < REUSEPORT_ARRAY_SIZE; i++) {
-		ev.data.u32 = i;
-		err = epoll_ctl(epfd, EPOLL_CTL_ADD, sk_fds[i], &ev);
-		CHECK(err, "epoll_ctl(EPOLL_CTL_ADD)", "sk_fds[%d]\n", i);
-	}
-}
-
-static void setup_per_test(int type, sa_family_t family, bool inany,
-			   bool no_inner_map)
-{
-	int ovr = -1, err;
-
-	prepare_sk_fds(type, family, inany);
-	err = bpf_map_update_elem(tmp_index_ovr_map, &index_zero, &ovr,
-				  BPF_ANY);
-	CHECK(err == -1, "update_elem(tmp_index_ovr_map, 0, -1)",
-	      "err:%d errno:%d\n", err, errno);
-
-	/* Install reuseport_array to outer_map? */
-	if (no_inner_map)
-		return;
-
-	err = bpf_map_update_elem(outer_map, &index_zero, &reuseport_array,
-				  BPF_ANY);
-	CHECK(err == -1, "update_elem(outer_map, 0, reuseport_array)",
-	      "err:%d errno:%d\n", err, errno);
-}
-
-static void cleanup_per_test(bool no_inner_map)
-{
-	int i, err;
-
-	for (i = 0; i < REUSEPORT_ARRAY_SIZE; i++)
-		close(sk_fds[i]);
-	close(epfd);
-
-	/* Delete reuseport_array from outer_map? */
-	if (no_inner_map)
-		return;
-
-	err = bpf_map_delete_elem(outer_map, &index_zero);
-	CHECK(err == -1, "delete_elem(outer_map)",
-	      "err:%d errno:%d\n", err, errno);
-}
-
-static void cleanup(void)
-{
-	if (outer_map != -1)
-		close(outer_map);
-	if (reuseport_array != -1)
-		close(reuseport_array);
-	if (obj)
-		bpf_object__close(obj);
-}
-
-static const char *family_str(sa_family_t family)
-{
-	switch (family) {
-	case AF_INET:
-		return "IPv4";
-	case AF_INET6:
-		return "IPv6";
-	default:
-		return "unknown";
-	}
-}
-
-static const char *sotype_str(int sotype)
-{
-	switch (sotype) {
-	case SOCK_STREAM:
-		return "TCP";
-	case SOCK_DGRAM:
-		return "UDP";
-	default:
-		return "unknown";
-	}
-}
-
-#define TEST_INIT(fn, ...) { fn, #fn, __VA_ARGS__ }
-
-static void test_config(int type, sa_family_t family, bool inany)
-{
-	const struct test {
-		void (*fn)(int sotype, sa_family_t family);
-		const char *name;
-		bool no_inner_map;
-	} tests[] = {
-		TEST_INIT(test_err_inner_map, true /* no_inner_map */),
-		TEST_INIT(test_err_skb_data),
-		TEST_INIT(test_err_sk_select_port),
-		TEST_INIT(test_pass),
-		TEST_INIT(test_syncookie),
-		TEST_INIT(test_pass_on_err),
-		TEST_INIT(test_detach_bpf),
-	};
-	const struct test *t;
-
-	printf("######## %s/%s %s ########\n",
-	       family_str(family), sotype_str(type),
-	       inany ? " INANY  " : "LOOPBACK");
-
-	for (t = tests; t < tests + ARRAY_SIZE(tests); t++) {
-		setup_per_test(type, family, inany, t->no_inner_map);
-		printf("%s: ", t->name);
-		t->fn(type, family);
-		cleanup_per_test(t->no_inner_map);
-	}
-
-	printf("\n");
-}
-
-#define BIND_INANY true
-
-static void test_all(void)
-{
-	const struct config {
-		int sotype;
-		sa_family_t family;
-		bool inany;
-	} configs[] = {
-		{ SOCK_STREAM, AF_INET },
-		{ SOCK_STREAM, AF_INET, BIND_INANY },
-		{ SOCK_STREAM, AF_INET6 },
-		{ SOCK_STREAM, AF_INET6, BIND_INANY },
-		{ SOCK_DGRAM, AF_INET },
-		{ SOCK_DGRAM, AF_INET6 },
-	};
-	const struct config *c;
-
-	for (c = configs; c < configs + ARRAY_SIZE(configs); c++)
-		test_config(c->sotype, c->family, c->inany);
-}
-
-int main(int argc, const char **argv)
-{
-	int ret = EXIT_FAILURE;
-
-	if (create_maps())
-		goto out;
-	if (prepare_bpf_obj())
-		goto out;
-
-	saved_tcp_fo = read_int_sysctl(TCP_FO_SYSCTL);
-	saved_tcp_syncookie = read_int_sysctl(TCP_SYNCOOKIE_SYSCTL);
-	if (saved_tcp_syncookie < 0 || saved_tcp_syncookie < 0)
-		goto out;
-	atexit(restore_sysctls);
-
-	if (enable_fastopen())
-		goto out;
-	if (disable_syncookie())
-		goto out;
-
-	test_all();
-
-	ret = EXIT_SUCCESS;
-out:
-	cleanup();
-	return ret;
-}
-- 
cgit v1.2.3-59-g8ed1b


From 7ee0d4e97b889c0478af9c1a6e5af658b181423f Mon Sep 17 00:00:00 2001
From: Jakub Sitnicki <jakub@cloudflare.com>
Date: Thu, 12 Dec 2019 11:22:59 +0100
Subject: selftests/bpf: Switch reuseport tests for test_progs framework

The tests were originally written in abort-on-error style. With the switch
to test_progs we can no longer do that. So at the risk of not cleaning up
some resource on failure, we now return to the caller on error.

That said, failure inside one test should not affect others because we run
setup/cleanup before/after every test.

Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191212102259.418536-11-jakub@cloudflare.com
---
 .../selftests/bpf/prog_tests/select_reuseport.c    | 267 ++++++++++-----------
 1 file changed, 131 insertions(+), 136 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/select_reuseport.c b/tools/testing/selftests/bpf/prog_tests/select_reuseport.c
index 0d5687feb689..2c37ae7dc214 100644
--- a/tools/testing/selftests/bpf/prog_tests/select_reuseport.c
+++ b/tools/testing/selftests/bpf/prog_tests/select_reuseport.c
@@ -20,8 +20,11 @@
 #include <bpf/libbpf.h>
 #include "bpf_rlimit.h"
 #include "bpf_util.h"
+
+#include "test_progs.h"
 #include "test_select_reuseport_common.h"
 
+#define MAX_TEST_NAME 80
 #define MIN_TCPHDR_LEN 20
 #define UDPHDR_LEN 8
 
@@ -34,9 +37,9 @@ static enum result expected_results[NR_RESULTS];
 static int sk_fds[REUSEPORT_ARRAY_SIZE];
 static int reuseport_array = -1, outer_map = -1;
 static int select_by_skb_data_prog;
-static int saved_tcp_syncookie;
+static int saved_tcp_syncookie = -1;
 static struct bpf_object *obj;
-static int saved_tcp_fo;
+static int saved_tcp_fo = -1;
 static __u32 index_zero;
 static int epfd;
 
@@ -46,20 +49,16 @@ static union sa46 {
 	sa_family_t family;
 } srv_sa;
 
-#define CHECK(condition, tag, format...) ({				\
-	int __ret = !!(condition);					\
-	if (__ret) {							\
-		printf("%s(%d):FAIL:%s ", __func__, __LINE__, tag);	\
-		printf(format);						\
-		exit(-1);						\
+#define RET_IF(condition, tag, format...) ({				\
+	if (CHECK_FAIL(condition)) {					\
+		printf(tag " " format);					\
+		return;							\
 	}								\
 })
 
 #define RET_ERR(condition, tag, format...) ({				\
-	int __ret = !!(condition);					\
-	if (__ret) {							\
-		printf("%s(%d):FAIL:%s ", __func__, __LINE__, tag);	\
-		printf(format);						\
+	if (CHECK_FAIL(condition)) {					\
+		printf(tag " " format);					\
 		return -1;						\
 	}								\
 })
@@ -202,8 +201,10 @@ static int write_int_sysctl(const char *sysctl, int v)
 
 static void restore_sysctls(void)
 {
-	write_int_sysctl(TCP_FO_SYSCTL, saved_tcp_fo);
-	write_int_sysctl(TCP_SYNCOOKIE_SYSCTL, saved_tcp_syncookie);
+	if (saved_tcp_fo != -1)
+		write_int_sysctl(TCP_FO_SYSCTL, saved_tcp_fo);
+	if (saved_tcp_syncookie != -1)
+		write_int_sysctl(TCP_SYNCOOKIE_SYSCTL, saved_tcp_syncookie);
 }
 
 static int enable_fastopen(void)
@@ -227,14 +228,14 @@ static int disable_syncookie(void)
 	return write_int_sysctl(TCP_SYNCOOKIE_SYSCTL, 0);
 }
 
-static __u32 get_linum(void)
+static long get_linum(void)
 {
 	__u32 linum;
 	int err;
 
 	err = bpf_map_lookup_elem(linum_map, &index_zero, &linum);
-	CHECK(err == -1, "lookup_elem(linum_map)", "err:%d errno:%d\n",
-	      err, errno);
+	RET_ERR(err == -1, "lookup_elem(linum_map)", "err:%d errno:%d\n",
+		err, errno);
 
 	return linum;
 }
@@ -250,12 +251,12 @@ static void check_data(int type, sa_family_t family, const struct cmd *cmd,
 	addrlen = sizeof(cli_sa);
 	err = getsockname(cli_fd, (struct sockaddr *)&cli_sa,
 			  &addrlen);
-	CHECK(err == -1, "getsockname(cli_fd)", "err:%d errno:%d\n",
-	      err, errno);
+	RET_IF(err == -1, "getsockname(cli_fd)", "err:%d errno:%d\n",
+	       err, errno);
 
 	err = bpf_map_lookup_elem(data_check_map, &index_zero, &result);
-	CHECK(err == -1, "lookup_elem(data_check_map)", "err:%d errno:%d\n",
-	      err, errno);
+	RET_IF(err == -1, "lookup_elem(data_check_map)", "err:%d errno:%d\n",
+	       err, errno);
 
 	if (type == SOCK_STREAM) {
 		expected.len = MIN_TCPHDR_LEN;
@@ -297,22 +298,22 @@ static void check_data(int type, sa_family_t family, const struct cmd *cmd,
 		printf("expected: (0x%x, %u, %u)\n",
 		       expected.eth_protocol, expected.ip_protocol,
 		       expected.bind_inany);
-		CHECK(1, "data_check result != expected",
-		      "bpf_prog_linum:%u\n", get_linum());
+		RET_IF(1, "data_check result != expected",
+		       "bpf_prog_linum:%ld\n", get_linum());
 	}
 
-	CHECK(!result.hash, "data_check result.hash empty",
-	      "result.hash:%u", result.hash);
+	RET_IF(!result.hash, "data_check result.hash empty",
+	       "result.hash:%u", result.hash);
 
 	expected.len += cmd ? sizeof(*cmd) : 0;
 	if (type == SOCK_STREAM)
-		CHECK(expected.len > result.len, "expected.len > result.len",
-		      "expected.len:%u result.len:%u bpf_prog_linum:%u\n",
-		      expected.len, result.len, get_linum());
+		RET_IF(expected.len > result.len, "expected.len > result.len",
+		       "expected.len:%u result.len:%u bpf_prog_linum:%ld\n",
+		       expected.len, result.len, get_linum());
 	else
-		CHECK(expected.len != result.len, "expected.len != result.len",
-		      "expected.len:%u result.len:%u bpf_prog_linum:%u\n",
-		      expected.len, result.len, get_linum());
+		RET_IF(expected.len != result.len, "expected.len != result.len",
+		       "expected.len:%u result.len:%u bpf_prog_linum:%ld\n",
+		       expected.len, result.len, get_linum());
 }
 
 static void check_results(void)
@@ -323,8 +324,8 @@ static void check_results(void)
 
 	for (i = 0; i < NR_RESULTS; i++) {
 		err = bpf_map_lookup_elem(result_map, &i, &results[i]);
-		CHECK(err == -1, "lookup_elem(result_map)",
-		      "i:%u err:%d errno:%d\n", i, err, errno);
+		RET_IF(err == -1, "lookup_elem(result_map)",
+		       "i:%u err:%d errno:%d\n", i, err, errno);
 	}
 
 	for (i = 0; i < NR_RESULTS; i++) {
@@ -350,10 +351,10 @@ static void check_results(void)
 		printf(", %u", expected_results[i]);
 	printf("]\n");
 
-	CHECK(expected_results[broken] != results[broken],
-	      "unexpected result",
-	      "expected_results[%u] != results[%u] bpf_prog_linum:%u\n",
-	      broken, broken, get_linum());
+	RET_IF(expected_results[broken] != results[broken],
+	       "unexpected result",
+	       "expected_results[%u] != results[%u] bpf_prog_linum:%ld\n",
+	       broken, broken, get_linum());
 }
 
 static int send_data(int type, sa_family_t family, void *data, size_t len,
@@ -363,17 +364,17 @@ static int send_data(int type, sa_family_t family, void *data, size_t len,
 	int fd, err;
 
 	fd = socket(family, type, 0);
-	CHECK(fd == -1, "socket()", "fd:%d errno:%d\n", fd, errno);
+	RET_ERR(fd == -1, "socket()", "fd:%d errno:%d\n", fd, errno);
 
 	sa46_init_loopback(&cli_sa, family);
 	err = bind(fd, (struct sockaddr *)&cli_sa, sizeof(cli_sa));
-	CHECK(fd == -1, "bind(cli_sa)", "err:%d errno:%d\n", err, errno);
+	RET_ERR(fd == -1, "bind(cli_sa)", "err:%d errno:%d\n", err, errno);
 
 	err = sendto(fd, data, len, MSG_FASTOPEN, (struct sockaddr *)&srv_sa,
 		     sizeof(srv_sa));
-	CHECK(err != len && expected >= PASS,
-	      "sendto()", "family:%u err:%d errno:%d expected:%d\n",
-	      family, err, errno, expected);
+	RET_ERR(err != len && expected >= PASS,
+		"sendto()", "family:%u err:%d errno:%d expected:%d\n",
+		family, err, errno, expected);
 
 	return fd;
 }
@@ -388,47 +389,49 @@ static void do_test(int type, sa_family_t family, struct cmd *cmd,
 
 	cli_fd = send_data(type, family, cmd, cmd ? sizeof(*cmd) : 0,
 			   expected);
+	if (cli_fd < 0)
+		return;
 	nev = epoll_wait(epfd, &ev, 1, expected >= PASS ? 5 : 0);
-	CHECK((nev <= 0 && expected >= PASS) ||
-	      (nev > 0 && expected < PASS),
-	      "nev <> expected",
-	      "nev:%d expected:%d type:%d family:%d data:(%d, %d)\n",
-	      nev, expected, type, family,
-	      cmd ? cmd->reuseport_index : -1,
-	      cmd ? cmd->pass_on_failure : -1);
+	RET_IF((nev <= 0 && expected >= PASS) ||
+	       (nev > 0 && expected < PASS),
+	       "nev <> expected",
+	       "nev:%d expected:%d type:%d family:%d data:(%d, %d)\n",
+	       nev, expected, type, family,
+	       cmd ? cmd->reuseport_index : -1,
+	       cmd ? cmd->pass_on_failure : -1);
 	check_results();
 	check_data(type, family, cmd, cli_fd);
 
 	if (expected < PASS)
 		return;
 
-	CHECK(expected != PASS_ERR_SK_SELECT_REUSEPORT &&
-	      cmd->reuseport_index != ev.data.u32,
-	      "check cmd->reuseport_index",
-	      "cmd:(%u, %u) ev.data.u32:%u\n",
-	      cmd->pass_on_failure, cmd->reuseport_index, ev.data.u32);
+	RET_IF(expected != PASS_ERR_SK_SELECT_REUSEPORT &&
+	       cmd->reuseport_index != ev.data.u32,
+	       "check cmd->reuseport_index",
+	       "cmd:(%u, %u) ev.data.u32:%u\n",
+	       cmd->pass_on_failure, cmd->reuseport_index, ev.data.u32);
 
 	srv_fd = sk_fds[ev.data.u32];
 	if (type == SOCK_STREAM) {
 		int new_fd = accept(srv_fd, NULL, 0);
 
-		CHECK(new_fd == -1, "accept(srv_fd)",
-		      "ev.data.u32:%u new_fd:%d errno:%d\n",
-		      ev.data.u32, new_fd, errno);
+		RET_IF(new_fd == -1, "accept(srv_fd)",
+		       "ev.data.u32:%u new_fd:%d errno:%d\n",
+		       ev.data.u32, new_fd, errno);
 
 		nread = recv(new_fd, &rcv_cmd, sizeof(rcv_cmd), MSG_DONTWAIT);
-		CHECK(nread != sizeof(rcv_cmd),
-		      "recv(new_fd)",
-		      "ev.data.u32:%u nread:%zd sizeof(rcv_cmd):%zu errno:%d\n",
-		      ev.data.u32, nread, sizeof(rcv_cmd), errno);
+		RET_IF(nread != sizeof(rcv_cmd),
+		       "recv(new_fd)",
+		       "ev.data.u32:%u nread:%zd sizeof(rcv_cmd):%zu errno:%d\n",
+		       ev.data.u32, nread, sizeof(rcv_cmd), errno);
 
 		close(new_fd);
 	} else {
 		nread = recv(srv_fd, &rcv_cmd, sizeof(rcv_cmd), MSG_DONTWAIT);
-		CHECK(nread != sizeof(rcv_cmd),
-		      "recv(sk_fds)",
-		      "ev.data.u32:%u nread:%zd sizeof(rcv_cmd):%zu errno:%d\n",
-		      ev.data.u32, nread, sizeof(rcv_cmd), errno);
+		RET_IF(nread != sizeof(rcv_cmd),
+		       "recv(sk_fds)",
+		       "ev.data.u32:%u nread:%zd sizeof(rcv_cmd):%zu errno:%d\n",
+		       ev.data.u32, nread, sizeof(rcv_cmd), errno);
 	}
 
 	close(cli_fd);
@@ -443,14 +446,12 @@ static void test_err_inner_map(int type, sa_family_t family)
 
 	expected_results[DROP_ERR_INNER_MAP]++;
 	do_test(type, family, &cmd, DROP_ERR_INNER_MAP);
-	printf("OK\n");
 }
 
 static void test_err_skb_data(int type, sa_family_t family)
 {
 	expected_results[DROP_ERR_SKB_DATA]++;
 	do_test(type, family, NULL, DROP_ERR_SKB_DATA);
-	printf("OK\n");
 }
 
 static void test_err_sk_select_port(int type, sa_family_t family)
@@ -462,7 +463,6 @@ static void test_err_sk_select_port(int type, sa_family_t family)
 
 	expected_results[DROP_ERR_SK_SELECT_REUSEPORT]++;
 	do_test(type, family, &cmd, DROP_ERR_SK_SELECT_REUSEPORT);
-	printf("OK\n");
 }
 
 static void test_pass(int type, sa_family_t family)
@@ -476,7 +476,6 @@ static void test_pass(int type, sa_family_t family)
 		cmd.reuseport_index = i;
 		do_test(type, family, &cmd, PASS);
 	}
-	printf("OK\n");
 }
 
 static void test_syncookie(int type, sa_family_t family)
@@ -505,17 +504,16 @@ static void test_syncookie(int type, sa_family_t family)
 	 */
 	err = bpf_map_update_elem(tmp_index_ovr_map, &index_zero,
 				  &tmp_index, BPF_ANY);
-	CHECK(err == -1, "update_elem(tmp_index_ovr_map, 0, 1)",
-	      "err:%d errno:%d\n", err, errno);
+	RET_IF(err == -1, "update_elem(tmp_index_ovr_map, 0, 1)",
+	       "err:%d errno:%d\n", err, errno);
 	do_test(type, family, &cmd, PASS);
 	err = bpf_map_lookup_elem(tmp_index_ovr_map, &index_zero,
 				  &tmp_index);
-	CHECK(err == -1 || tmp_index != -1,
-	      "lookup_elem(tmp_index_ovr_map)",
-	      "err:%d errno:%d tmp_index:%d\n",
-	      err, errno, tmp_index);
+	RET_IF(err == -1 || tmp_index != -1,
+	       "lookup_elem(tmp_index_ovr_map)",
+	       "err:%d errno:%d tmp_index:%d\n",
+	       err, errno, tmp_index);
 	disable_syncookie();
-	printf("OK\n");
 }
 
 static void test_pass_on_err(int type, sa_family_t family)
@@ -527,7 +525,6 @@ static void test_pass_on_err(int type, sa_family_t family)
 
 	expected_results[PASS_ERR_SK_SELECT_REUSEPORT] += 1;
 	do_test(type, family, &cmd, PASS_ERR_SK_SELECT_REUSEPORT);
-	printf("OK\n");
 }
 
 static void test_detach_bpf(int type, sa_family_t family)
@@ -541,43 +538,45 @@ static void test_detach_bpf(int type, sa_family_t family)
 
 	err = setsockopt(sk_fds[0], SOL_SOCKET, SO_DETACH_REUSEPORT_BPF,
 			 &optvalue, sizeof(optvalue));
-	CHECK(err == -1, "setsockopt(SO_DETACH_REUSEPORT_BPF)",
-	      "err:%d errno:%d\n", err, errno);
+	RET_IF(err == -1, "setsockopt(SO_DETACH_REUSEPORT_BPF)",
+	       "err:%d errno:%d\n", err, errno);
 
 	err = setsockopt(sk_fds[1], SOL_SOCKET, SO_DETACH_REUSEPORT_BPF,
 			 &optvalue, sizeof(optvalue));
-	CHECK(err == 0 || errno != ENOENT, "setsockopt(SO_DETACH_REUSEPORT_BPF)",
-	      "err:%d errno:%d\n", err, errno);
+	RET_IF(err == 0 || errno != ENOENT,
+	       "setsockopt(SO_DETACH_REUSEPORT_BPF)",
+	       "err:%d errno:%d\n", err, errno);
 
 	for (i = 0; i < NR_RESULTS; i++) {
 		err = bpf_map_lookup_elem(result_map, &i, &tmp);
-		CHECK(err == -1, "lookup_elem(result_map)",
-		      "i:%u err:%d errno:%d\n", i, err, errno);
+		RET_IF(err == -1, "lookup_elem(result_map)",
+		       "i:%u err:%d errno:%d\n", i, err, errno);
 		nr_run_before += tmp;
 	}
 
 	cli_fd = send_data(type, family, &cmd, sizeof(cmd), PASS);
+	if (cli_fd < 0)
+		return;
 	nev = epoll_wait(epfd, &ev, 1, 5);
-	CHECK(nev <= 0, "nev <= 0",
-	      "nev:%d expected:1 type:%d family:%d data:(0, 0)\n",
-	      nev,  type, family);
+	RET_IF(nev <= 0, "nev <= 0",
+	       "nev:%d expected:1 type:%d family:%d data:(0, 0)\n",
+	       nev,  type, family);
 
 	for (i = 0; i < NR_RESULTS; i++) {
 		err = bpf_map_lookup_elem(result_map, &i, &tmp);
-		CHECK(err == -1, "lookup_elem(result_map)",
-		      "i:%u err:%d errno:%d\n", i, err, errno);
+		RET_IF(err == -1, "lookup_elem(result_map)",
+		       "i:%u err:%d errno:%d\n", i, err, errno);
 		nr_run_after += tmp;
 	}
 
-	CHECK(nr_run_before != nr_run_after,
-	      "nr_run_before != nr_run_after",
-	      "nr_run_before:%u nr_run_after:%u\n",
-	      nr_run_before, nr_run_after);
+	RET_IF(nr_run_before != nr_run_after,
+	       "nr_run_before != nr_run_after",
+	       "nr_run_before:%u nr_run_after:%u\n",
+	       nr_run_before, nr_run_after);
 
-	printf("OK\n");
 	close(cli_fd);
 #else
-	printf("SKIP\n");
+	test__skip();
 #endif
 }
 
@@ -600,58 +599,58 @@ static void prepare_sk_fds(int type, sa_family_t family, bool inany)
 	 */
 	for (i = first; i >= 0; i--) {
 		sk_fds[i] = socket(family, type, 0);
-		CHECK(sk_fds[i] == -1, "socket()", "sk_fds[%d]:%d errno:%d\n",
-		      i, sk_fds[i], errno);
+		RET_IF(sk_fds[i] == -1, "socket()", "sk_fds[%d]:%d errno:%d\n",
+		       i, sk_fds[i], errno);
 		err = setsockopt(sk_fds[i], SOL_SOCKET, SO_REUSEPORT,
 				 &optval, sizeof(optval));
-		CHECK(err == -1, "setsockopt(SO_REUSEPORT)",
-		      "sk_fds[%d] err:%d errno:%d\n",
-		      i, err, errno);
+		RET_IF(err == -1, "setsockopt(SO_REUSEPORT)",
+		       "sk_fds[%d] err:%d errno:%d\n",
+		       i, err, errno);
 
 		if (i == first) {
 			err = setsockopt(sk_fds[i], SOL_SOCKET,
 					 SO_ATTACH_REUSEPORT_EBPF,
 					 &select_by_skb_data_prog,
 					 sizeof(select_by_skb_data_prog));
-			CHECK(err == -1, "setsockopt(SO_ATTACH_REUEPORT_EBPF)",
-			      "err:%d errno:%d\n", err, errno);
+			RET_IF(err == -1, "setsockopt(SO_ATTACH_REUEPORT_EBPF)",
+			       "err:%d errno:%d\n", err, errno);
 		}
 
 		err = bind(sk_fds[i], (struct sockaddr *)&srv_sa, addrlen);
-		CHECK(err == -1, "bind()", "sk_fds[%d] err:%d errno:%d\n",
-		      i, err, errno);
+		RET_IF(err == -1, "bind()", "sk_fds[%d] err:%d errno:%d\n",
+		       i, err, errno);
 
 		if (type == SOCK_STREAM) {
 			err = listen(sk_fds[i], 10);
-			CHECK(err == -1, "listen()",
-			      "sk_fds[%d] err:%d errno:%d\n",
-			      i, err, errno);
+			RET_IF(err == -1, "listen()",
+			       "sk_fds[%d] err:%d errno:%d\n",
+			       i, err, errno);
 		}
 
 		err = bpf_map_update_elem(reuseport_array, &i, &sk_fds[i],
 					  BPF_NOEXIST);
-		CHECK(err == -1, "update_elem(reuseport_array)",
-		      "sk_fds[%d] err:%d errno:%d\n", i, err, errno);
+		RET_IF(err == -1, "update_elem(reuseport_array)",
+		       "sk_fds[%d] err:%d errno:%d\n", i, err, errno);
 
 		if (i == first) {
 			socklen_t addrlen = sizeof(srv_sa);
 
 			err = getsockname(sk_fds[i], (struct sockaddr *)&srv_sa,
 					  &addrlen);
-			CHECK(err == -1, "getsockname()",
-			      "sk_fds[%d] err:%d errno:%d\n", i, err, errno);
+			RET_IF(err == -1, "getsockname()",
+			       "sk_fds[%d] err:%d errno:%d\n", i, err, errno);
 		}
 	}
 
 	epfd = epoll_create(1);
-	CHECK(epfd == -1, "epoll_create(1)",
-	      "epfd:%d errno:%d\n", epfd, errno);
+	RET_IF(epfd == -1, "epoll_create(1)",
+	       "epfd:%d errno:%d\n", epfd, errno);
 
 	ev.events = EPOLLIN;
 	for (i = 0; i < REUSEPORT_ARRAY_SIZE; i++) {
 		ev.data.u32 = i;
 		err = epoll_ctl(epfd, EPOLL_CTL_ADD, sk_fds[i], &ev);
-		CHECK(err, "epoll_ctl(EPOLL_CTL_ADD)", "sk_fds[%d]\n", i);
+		RET_IF(err, "epoll_ctl(EPOLL_CTL_ADD)", "sk_fds[%d]\n", i);
 	}
 }
 
@@ -663,8 +662,8 @@ static void setup_per_test(int type, sa_family_t family, bool inany,
 	prepare_sk_fds(type, family, inany);
 	err = bpf_map_update_elem(tmp_index_ovr_map, &index_zero, &ovr,
 				  BPF_ANY);
-	CHECK(err == -1, "update_elem(tmp_index_ovr_map, 0, -1)",
-	      "err:%d errno:%d\n", err, errno);
+	RET_IF(err == -1, "update_elem(tmp_index_ovr_map, 0, -1)",
+	       "err:%d errno:%d\n", err, errno);
 
 	/* Install reuseport_array to outer_map? */
 	if (no_inner_map)
@@ -672,8 +671,8 @@ static void setup_per_test(int type, sa_family_t family, bool inany,
 
 	err = bpf_map_update_elem(outer_map, &index_zero, &reuseport_array,
 				  BPF_ANY);
-	CHECK(err == -1, "update_elem(outer_map, 0, reuseport_array)",
-	      "err:%d errno:%d\n", err, errno);
+	RET_IF(err == -1, "update_elem(outer_map, 0, reuseport_array)",
+	       "err:%d errno:%d\n", err, errno);
 }
 
 static void cleanup_per_test(bool no_inner_map)
@@ -689,8 +688,8 @@ static void cleanup_per_test(bool no_inner_map)
 		return;
 
 	err = bpf_map_delete_elem(outer_map, &index_zero);
-	CHECK(err == -1, "delete_elem(outer_map)",
-	      "err:%d errno:%d\n", err, errno);
+	RET_IF(err == -1, "delete_elem(outer_map)",
+	       "err:%d errno:%d\n", err, errno);
 }
 
 static void cleanup(void)
@@ -729,7 +728,7 @@ static const char *sotype_str(int sotype)
 
 #define TEST_INIT(fn, ...) { fn, #fn, __VA_ARGS__ }
 
-static void test_config(int type, sa_family_t family, bool inany)
+static void test_config(int sotype, sa_family_t family, bool inany)
 {
 	const struct test {
 		void (*fn)(int sotype, sa_family_t family);
@@ -744,20 +743,21 @@ static void test_config(int type, sa_family_t family, bool inany)
 		TEST_INIT(test_pass_on_err),
 		TEST_INIT(test_detach_bpf),
 	};
+	char s[MAX_TEST_NAME];
 	const struct test *t;
 
-	printf("######## %s/%s %s ########\n",
-	       family_str(family), sotype_str(type),
-	       inany ? " INANY  " : "LOOPBACK");
-
 	for (t = tests; t < tests + ARRAY_SIZE(tests); t++) {
-		setup_per_test(type, family, inany, t->no_inner_map);
-		printf("%s: ", t->name);
-		t->fn(type, family);
+		snprintf(s, sizeof(s), "%s/%s %s %s",
+			 family_str(family), sotype_str(sotype),
+			 inany ? "INANY" : "LOOPBACK", t->name);
+
+		if (!test__start_subtest(s))
+			continue;
+
+		setup_per_test(sotype, family, inany, t->no_inner_map);
+		t->fn(sotype, family);
 		cleanup_per_test(t->no_inner_map);
 	}
-
-	printf("\n");
 }
 
 #define BIND_INANY true
@@ -782,10 +782,8 @@ static void test_all(void)
 		test_config(c->sotype, c->family, c->inany);
 }
 
-int main(int argc, const char **argv)
+void test_select_reuseport(void)
 {
-	int ret = EXIT_FAILURE;
-
 	if (create_maps())
 		goto out;
 	if (prepare_bpf_obj())
@@ -795,7 +793,6 @@ int main(int argc, const char **argv)
 	saved_tcp_syncookie = read_int_sysctl(TCP_SYNCOOKIE_SYSCTL);
 	if (saved_tcp_syncookie < 0 || saved_tcp_syncookie < 0)
 		goto out;
-	atexit(restore_sysctls);
 
 	if (enable_fastopen())
 		goto out;
@@ -803,9 +800,7 @@ int main(int argc, const char **argv)
 		goto out;
 
 	test_all();
-
-	ret = EXIT_SUCCESS;
 out:
 	cleanup();
-	return ret;
+	restore_sysctls();
 }
-- 
cgit v1.2.3-59-g8ed1b


From 6803ee25f0ead1e836808acb14396bb9a9849113 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Wed, 11 Dec 2019 17:35:48 -0800
Subject: libbpf: Extract and generalize CPU mask parsing logic

This logic is re-used for parsing a set of online CPUs. Having it as an
isolated piece of code working with input string makes it conveninent to test
this logic as well. While refactoring, also improve the robustness of original
implementation.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191212013548.1690564-1-andriin@fb.com
---
 tools/lib/bpf/libbpf.c          | 125 +++++++++++++++++++++++++++-------------
 tools/lib/bpf/libbpf_internal.h |   2 +
 2 files changed, 86 insertions(+), 41 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index b99c0a9b7756..98455e8aa6db 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -6523,61 +6523,104 @@ void bpf_program__bpil_offs_to_addr(struct bpf_prog_info_linear *info_linear)
 	}
 }
 
-int libbpf_num_possible_cpus(void)
+int parse_cpu_mask_str(const char *s, bool **mask, int *mask_sz)
 {
-	static const char *fcpu = "/sys/devices/system/cpu/possible";
-	int len = 0, n = 0, il = 0, ir = 0;
-	unsigned int start = 0, end = 0;
-	int tmp_cpus = 0;
-	static int cpus;
-	char buf[128];
-	int error = 0;
-	int fd = -1;
+	int err = 0, n, len, start, end = -1;
+	bool *tmp;
 
-	tmp_cpus = READ_ONCE(cpus);
-	if (tmp_cpus > 0)
-		return tmp_cpus;
+	*mask = NULL;
+	*mask_sz = 0;
+
+	/* Each sub string separated by ',' has format \d+-\d+ or \d+ */
+	while (*s) {
+		if (*s == ',' || *s == '\n') {
+			s++;
+			continue;
+		}
+		n = sscanf(s, "%d%n-%d%n", &start, &len, &end, &len);
+		if (n <= 0 || n > 2) {
+			pr_warn("Failed to get CPU range %s: %d\n", s, n);
+			err = -EINVAL;
+			goto cleanup;
+		} else if (n == 1) {
+			end = start;
+		}
+		if (start < 0 || start > end) {
+			pr_warn("Invalid CPU range [%d,%d] in %s\n",
+				start, end, s);
+			err = -EINVAL;
+			goto cleanup;
+		}
+		tmp = realloc(*mask, end + 1);
+		if (!tmp) {
+			err = -ENOMEM;
+			goto cleanup;
+		}
+		*mask = tmp;
+		memset(tmp + *mask_sz, 0, start - *mask_sz);
+		memset(tmp + start, 1, end - start + 1);
+		*mask_sz = end + 1;
+		s += len;
+	}
+	if (!*mask_sz) {
+		pr_warn("Empty CPU range\n");
+		return -EINVAL;
+	}
+	return 0;
+cleanup:
+	free(*mask);
+	*mask = NULL;
+	return err;
+}
+
+int parse_cpu_mask_file(const char *fcpu, bool **mask, int *mask_sz)
+{
+	int fd, err = 0, len;
+	char buf[128];
 
 	fd = open(fcpu, O_RDONLY);
 	if (fd < 0) {
-		error = errno;
-		pr_warn("Failed to open file %s: %s\n", fcpu, strerror(error));
-		return -error;
+		err = -errno;
+		pr_warn("Failed to open cpu mask file %s: %d\n", fcpu, err);
+		return err;
 	}
 	len = read(fd, buf, sizeof(buf));
 	close(fd);
 	if (len <= 0) {
-		error = len ? errno : EINVAL;
-		pr_warn("Failed to read # of possible cpus from %s: %s\n",
-			fcpu, strerror(error));
-		return -error;
+		err = len ? -errno : -EINVAL;
+		pr_warn("Failed to read cpu mask from %s: %d\n", fcpu, err);
+		return err;
 	}
-	if (len == sizeof(buf)) {
-		pr_warn("File %s size overflow\n", fcpu);
-		return -EOVERFLOW;
+	if (len >= sizeof(buf)) {
+		pr_warn("CPU mask is too big in file %s\n", fcpu);
+		return -E2BIG;
 	}
 	buf[len] = '\0';
 
-	for (ir = 0, tmp_cpus = 0; ir <= len; ir++) {
-		/* Each sub string separated by ',' has format \d+-\d+ or \d+ */
-		if (buf[ir] == ',' || buf[ir] == '\0') {
-			buf[ir] = '\0';
-			n = sscanf(&buf[il], "%u-%u", &start, &end);
-			if (n <= 0) {
-				pr_warn("Failed to get # CPUs from %s\n",
-					&buf[il]);
-				return -EINVAL;
-			} else if (n == 1) {
-				end = start;
-			}
-			tmp_cpus += end - start + 1;
-			il = ir + 1;
-		}
-	}
-	if (tmp_cpus <= 0) {
-		pr_warn("Invalid #CPUs %d from %s\n", tmp_cpus, fcpu);
-		return -EINVAL;
+	return parse_cpu_mask_str(buf, mask, mask_sz);
+}
+
+int libbpf_num_possible_cpus(void)
+{
+	static const char *fcpu = "/sys/devices/system/cpu/possible";
+	static int cpus;
+	int err, n, i, tmp_cpus;
+	bool *mask;
+
+	tmp_cpus = READ_ONCE(cpus);
+	if (tmp_cpus > 0)
+		return tmp_cpus;
+
+	err = parse_cpu_mask_file(fcpu, &mask, &n);
+	if (err)
+		return err;
+
+	tmp_cpus = 0;
+	for (i = 0; i < n; i++) {
+		if (mask[i])
+			tmp_cpus++;
 	}
+	free(mask);
 
 	WRITE_ONCE(cpus, tmp_cpus);
 	return tmp_cpus;
diff --git a/tools/lib/bpf/libbpf_internal.h b/tools/lib/bpf/libbpf_internal.h
index 97ac17a64a58..f4f10715db40 100644
--- a/tools/lib/bpf/libbpf_internal.h
+++ b/tools/lib/bpf/libbpf_internal.h
@@ -95,6 +95,8 @@ static inline bool libbpf_validate_opts(const char *opts,
 #define OPTS_GET(opts, field, fallback_value) \
 	(OPTS_HAS(opts, field) ? (opts)->field : fallback_value)
 
+int parse_cpu_mask_str(const char *s, bool **mask, int *mask_sz);
+int parse_cpu_mask_file(const char *fcpu, bool **mask, int *mask_sz);
 int libbpf__load_raw_btf(const char *raw_types, size_t types_len,
 			 const char *str_sec, size_t str_len);
 
-- 
cgit v1.2.3-59-g8ed1b


From 65bc4c4063ae36223375a920b7d116e54d7def34 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Wed, 11 Dec 2019 17:35:58 -0800
Subject: selftests/bpf: Add CPU mask parsing tests

Add a bunch of test validating CPU mask parsing logic and error handling.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191212013559.1690898-1-andriin@fb.com
---
 tools/testing/selftests/bpf/prog_tests/cpu_mask.c | 78 +++++++++++++++++++++++
 1 file changed, 78 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/cpu_mask.c

diff --git a/tools/testing/selftests/bpf/prog_tests/cpu_mask.c b/tools/testing/selftests/bpf/prog_tests/cpu_mask.c
new file mode 100644
index 000000000000..1fa1bdbaffa9
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/cpu_mask.c
@@ -0,0 +1,78 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <test_progs.h>
+#include <bpf/btf.h>
+#include "libbpf_internal.h"
+
+static int duration = 0;
+
+static void validate_mask(int case_nr, const char *exp, bool *mask, int n)
+{
+	int i;
+
+	for (i = 0; exp[i]; i++) {
+		if (exp[i] == '1') {
+			if (CHECK(i + 1 > n, "mask_short",
+				  "case #%d: mask too short, got n=%d, need at least %d\n",
+				  case_nr, n, i + 1))
+				return;
+			CHECK(!mask[i], "cpu_not_set",
+			      "case #%d: mask differs, expected cpu#%d SET\n",
+			      case_nr, i);
+		} else {
+			CHECK(i < n && mask[i], "cpu_set",
+			      "case #%d: mask differs, expected cpu#%d UNSET\n",
+			      case_nr, i);
+		}
+	}
+	CHECK(i < n, "mask_long",
+	      "case #%d: mask too long, got n=%d, expected at most %d\n",
+	      case_nr, n, i);
+}
+
+static struct {
+	const char *cpu_mask;
+	const char *expect;
+	bool fails;
+} test_cases[] = {
+	{ "0\n", "1", false },
+	{ "0,2\n", "101", false },
+	{ "0-2\n", "111", false },
+	{ "0-2,3-4\n", "11111", false },
+	{ "0", "1", false },
+	{ "0-2", "111", false },
+	{ "0,2", "101", false },
+	{ "0,1-3", "1111", false },
+	{ "0,1,2,3", "1111", false },
+	{ "0,2-3,5", "101101", false },
+	{ "3-3", "0001", false },
+	{ "2-4,6,9-10", "00111010011", false },
+	/* failure cases */
+	{ "", "", true },
+	{ "0-", "", true },
+	{ "0 ", "", true },
+	{ "0_1", "", true },
+	{ "1-0", "", true },
+	{ "-1", "", true },
+};
+
+void test_cpu_mask()
+{
+	int i, err, n;
+	bool *mask;
+
+	for (i = 0; i < ARRAY_SIZE(test_cases); i++) {
+		mask = NULL;
+		err = parse_cpu_mask_str(test_cases[i].cpu_mask, &mask, &n);
+		if (test_cases[i].fails) {
+			CHECK(!err, "should_fail",
+			      "case #%d: parsing should fail!\n", i + 1);
+		} else {
+			if (CHECK(err, "parse_err",
+				  "case #%d: cpu mask parsing failed: %d\n",
+				  i + 1, err))
+				continue;
+			validate_mask(i + 1, test_cases[i].expect, mask, n);
+		}
+		free(mask);
+	}
+}
-- 
cgit v1.2.3-59-g8ed1b


From 783b8f01f5942a786998f5577bd9ff3992f22a1a Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Wed, 11 Dec 2019 17:36:09 -0800
Subject: libbpf: Don't attach perf_buffer to offline/missing CPUs

It's quite common on some systems to have more CPUs enlisted as "possible",
than there are (and could ever be) present/online CPUs. In such cases,
perf_buffer creationg will fail due to inability to create perf event on
missing CPU with error like this:

libbpf: failed to open perf buffer event on cpu #16: No such device

This patch fixes the logic of perf_buffer__new() to ignore CPUs that are
missing or currently offline. In rare cases where user explicitly listed
specific CPUs to connect to, behavior is unchanged: libbpf will try to open
perf event buffer on specified CPU(s) anyways.

Fixes: fb84b8224655 ("libbpf: add perf buffer API")
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191212013609.1691168-1-andriin@fb.com
---
 tools/lib/bpf/libbpf.c | 32 +++++++++++++++++++++++++-------
 1 file changed, 25 insertions(+), 7 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 98455e8aa6db..27d5f7ecba32 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -5946,7 +5946,7 @@ struct perf_buffer {
 	size_t mmap_size;
 	struct perf_cpu_buf **cpu_bufs;
 	struct epoll_event *events;
-	int cpu_cnt;
+	int cpu_cnt; /* number of allocated CPU buffers */
 	int epoll_fd; /* perf event FD */
 	int map_fd; /* BPF_MAP_TYPE_PERF_EVENT_ARRAY BPF map FD */
 };
@@ -6080,11 +6080,13 @@ perf_buffer__new_raw(int map_fd, size_t page_cnt,
 static struct perf_buffer *__perf_buffer__new(int map_fd, size_t page_cnt,
 					      struct perf_buffer_params *p)
 {
+	const char *online_cpus_file = "/sys/devices/system/cpu/online";
 	struct bpf_map_info map = {};
 	char msg[STRERR_BUFSIZE];
 	struct perf_buffer *pb;
+	bool *online = NULL;
 	__u32 map_info_len;
-	int err, i;
+	int err, i, j, n;
 
 	if (page_cnt & (page_cnt - 1)) {
 		pr_warn("page count should be power of two, but is %zu\n",
@@ -6153,20 +6155,32 @@ static struct perf_buffer *__perf_buffer__new(int map_fd, size_t page_cnt,
 		goto error;
 	}
 
-	for (i = 0; i < pb->cpu_cnt; i++) {
+	err = parse_cpu_mask_file(online_cpus_file, &online, &n);
+	if (err) {
+		pr_warn("failed to get online CPU mask: %d\n", err);
+		goto error;
+	}
+
+	for (i = 0, j = 0; i < pb->cpu_cnt; i++) {
 		struct perf_cpu_buf *cpu_buf;
 		int cpu, map_key;
 
 		cpu = p->cpu_cnt > 0 ? p->cpus[i] : i;
 		map_key = p->cpu_cnt > 0 ? p->map_keys[i] : i;
 
+		/* in case user didn't explicitly requested particular CPUs to
+		 * be attached to, skip offline/not present CPUs
+		 */
+		if (p->cpu_cnt <= 0 && (cpu >= n || !online[cpu]))
+			continue;
+
 		cpu_buf = perf_buffer__open_cpu_buf(pb, p->attr, cpu, map_key);
 		if (IS_ERR(cpu_buf)) {
 			err = PTR_ERR(cpu_buf);
 			goto error;
 		}
 
-		pb->cpu_bufs[i] = cpu_buf;
+		pb->cpu_bufs[j] = cpu_buf;
 
 		err = bpf_map_update_elem(pb->map_fd, &map_key,
 					  &cpu_buf->fd, 0);
@@ -6178,21 +6192,25 @@ static struct perf_buffer *__perf_buffer__new(int map_fd, size_t page_cnt,
 			goto error;
 		}
 
-		pb->events[i].events = EPOLLIN;
-		pb->events[i].data.ptr = cpu_buf;
+		pb->events[j].events = EPOLLIN;
+		pb->events[j].data.ptr = cpu_buf;
 		if (epoll_ctl(pb->epoll_fd, EPOLL_CTL_ADD, cpu_buf->fd,
-			      &pb->events[i]) < 0) {
+			      &pb->events[j]) < 0) {
 			err = -errno;
 			pr_warn("failed to epoll_ctl cpu #%d perf FD %d: %s\n",
 				cpu, cpu_buf->fd,
 				libbpf_strerror_r(err, msg, sizeof(msg)));
 			goto error;
 		}
+		j++;
 	}
+	pb->cpu_cnt = j;
+	free(online);
 
 	return pb;
 
 error:
+	free(online);
 	if (pb)
 		perf_buffer__free(pb);
 	return ERR_PTR(err);
-- 
cgit v1.2.3-59-g8ed1b


From 91cbdf740a476cf2c744169bf407de2e3ac1f3cf Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Wed, 11 Dec 2019 17:36:20 -0800
Subject: selftests/bpf: Fix perf_buffer test on systems w/ offline CPUs

Fix up perf_buffer.c selftest to take into account offline/missing CPUs.

Fixes: ee5cf82ce04a ("selftests/bpf: test perf buffer API")
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191212013621.1691858-1-andriin@fb.com
---
 .../testing/selftests/bpf/prog_tests/perf_buffer.c | 29 ++++++++++++++++++----
 1 file changed, 24 insertions(+), 5 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/perf_buffer.c b/tools/testing/selftests/bpf/prog_tests/perf_buffer.c
index 3003fddc0613..cf6c87936c69 100644
--- a/tools/testing/selftests/bpf/prog_tests/perf_buffer.c
+++ b/tools/testing/selftests/bpf/prog_tests/perf_buffer.c
@@ -4,6 +4,7 @@
 #include <sched.h>
 #include <sys/socket.h>
 #include <test_progs.h>
+#include "libbpf_internal.h"
 
 static void on_sample(void *ctx, int cpu, void *data, __u32 size)
 {
@@ -19,7 +20,7 @@ static void on_sample(void *ctx, int cpu, void *data, __u32 size)
 
 void test_perf_buffer(void)
 {
-	int err, prog_fd, nr_cpus, i, duration = 0;
+	int err, prog_fd, on_len, nr_on_cpus = 0,  nr_cpus, i, duration = 0;
 	const char *prog_name = "kprobe/sys_nanosleep";
 	const char *file = "./test_perf_buffer.o";
 	struct perf_buffer_opts pb_opts = {};
@@ -29,15 +30,27 @@ void test_perf_buffer(void)
 	struct bpf_object *obj;
 	struct perf_buffer *pb;
 	struct bpf_link *link;
+	bool *online;
 
 	nr_cpus = libbpf_num_possible_cpus();
 	if (CHECK(nr_cpus < 0, "nr_cpus", "err %d\n", nr_cpus))
 		return;
 
+	err = parse_cpu_mask_file("/sys/devices/system/cpu/online",
+				  &online, &on_len);
+	if (CHECK(err, "nr_on_cpus", "err %d\n", err))
+		return;
+
+	for (i = 0; i < on_len; i++)
+		if (online[i])
+			nr_on_cpus++;
+
 	/* load program */
 	err = bpf_prog_load(file, BPF_PROG_TYPE_KPROBE, &obj, &prog_fd);
-	if (CHECK(err, "obj_load", "err %d errno %d\n", err, errno))
-		return;
+	if (CHECK(err, "obj_load", "err %d errno %d\n", err, errno)) {
+		obj = NULL;
+		goto out_close;
+	}
 
 	prog = bpf_object__find_program_by_title(obj, prog_name);
 	if (CHECK(!prog, "find_probe", "prog '%s' not found\n", prog_name))
@@ -64,6 +77,11 @@ void test_perf_buffer(void)
 	/* trigger kprobe on every CPU */
 	CPU_ZERO(&cpu_seen);
 	for (i = 0; i < nr_cpus; i++) {
+		if (i >= on_len || !online[i]) {
+			printf("skipping offline CPU #%d\n", i);
+			continue;
+		}
+
 		CPU_ZERO(&cpu_set);
 		CPU_SET(i, &cpu_set);
 
@@ -81,8 +99,8 @@ void test_perf_buffer(void)
 	if (CHECK(err < 0, "perf_buffer__poll", "err %d\n", err))
 		goto out_free_pb;
 
-	if (CHECK(CPU_COUNT(&cpu_seen) != nr_cpus, "seen_cpu_cnt",
-		  "expect %d, seen %d\n", nr_cpus, CPU_COUNT(&cpu_seen)))
+	if (CHECK(CPU_COUNT(&cpu_seen) != nr_on_cpus, "seen_cpu_cnt",
+		  "expect %d, seen %d\n", nr_on_cpus, CPU_COUNT(&cpu_seen)))
 		goto out_free_pb;
 
 out_free_pb:
@@ -91,4 +109,5 @@ out_detach:
 	bpf_link__destroy(link);
 out_close:
 	bpf_object__close(obj);
+	free(online);
 }
-- 
cgit v1.2.3-59-g8ed1b


From 98e8627efcada18ac043a77b9101b4b4c768090b Mon Sep 17 00:00:00 2001
From: Björn Töpel <bjorn.topel@intel.com>
Date: Fri, 13 Dec 2019 18:51:07 +0100
Subject: bpf: Move trampoline JIT image allocation to a function
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Refactor the image allocation in the BPF trampoline code into a
separate function, so it can be shared with the BPF dispatcher in
upcoming commits.

Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191213175112.30208-2-bjorn.topel@gmail.com
---
 include/linux/bpf.h     |  1 +
 kernel/bpf/trampoline.c | 24 +++++++++++++++++-------
 2 files changed, 18 insertions(+), 7 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 35903f148be5..5d744828b399 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -475,6 +475,7 @@ struct bpf_trampoline *bpf_trampoline_lookup(u64 key);
 int bpf_trampoline_link_prog(struct bpf_prog *prog);
 int bpf_trampoline_unlink_prog(struct bpf_prog *prog);
 void bpf_trampoline_put(struct bpf_trampoline *tr);
+void *bpf_jit_alloc_exec_page(void);
 #else
 static inline struct bpf_trampoline *bpf_trampoline_lookup(u64 key)
 {
diff --git a/kernel/bpf/trampoline.c b/kernel/bpf/trampoline.c
index 7e89f1f49d77..5ee301ddbd00 100644
--- a/kernel/bpf/trampoline.c
+++ b/kernel/bpf/trampoline.c
@@ -13,6 +13,22 @@ static struct hlist_head trampoline_table[TRAMPOLINE_TABLE_SIZE];
 /* serializes access to trampoline_table */
 static DEFINE_MUTEX(trampoline_mutex);
 
+void *bpf_jit_alloc_exec_page(void)
+{
+	void *image;
+
+	image = bpf_jit_alloc_exec(PAGE_SIZE);
+	if (!image)
+		return NULL;
+
+	set_vm_flush_reset_perms(image);
+	/* Keep image as writeable. The alternative is to keep flipping ro/rw
+	 * everytime new program is attached or detached.
+	 */
+	set_memory_x((long)image, 1);
+	return image;
+}
+
 struct bpf_trampoline *bpf_trampoline_lookup(u64 key)
 {
 	struct bpf_trampoline *tr;
@@ -33,7 +49,7 @@ struct bpf_trampoline *bpf_trampoline_lookup(u64 key)
 		goto out;
 
 	/* is_root was checked earlier. No need for bpf_jit_charge_modmem() */
-	image = bpf_jit_alloc_exec(PAGE_SIZE);
+	image = bpf_jit_alloc_exec_page();
 	if (!image) {
 		kfree(tr);
 		tr = NULL;
@@ -47,12 +63,6 @@ struct bpf_trampoline *bpf_trampoline_lookup(u64 key)
 	mutex_init(&tr->mutex);
 	for (i = 0; i < BPF_TRAMP_MAX; i++)
 		INIT_HLIST_HEAD(&tr->progs_hlist[i]);
-
-	set_vm_flush_reset_perms(image);
-	/* Keep image as writeable. The alternative is to keep flipping ro/rw
-	 * everytime new program is attached or detached.
-	 */
-	set_memory_x((long)image, 1);
 	tr->image = image;
 out:
 	mutex_unlock(&trampoline_mutex);
-- 
cgit v1.2.3-59-g8ed1b


From 75ccbef6369e94ecac696a152a998a978d41376b Mon Sep 17 00:00:00 2001
From: Björn Töpel <bjorn.topel@intel.com>
Date: Fri, 13 Dec 2019 18:51:08 +0100
Subject: bpf: Introduce BPF dispatcher
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The BPF dispatcher is a multi-way branch code generator, mainly
targeted for XDP programs. When an XDP program is executed via the
bpf_prog_run_xdp(), it is invoked via an indirect call. The indirect
call has a substantial performance impact, when retpolines are
enabled. The dispatcher transform indirect calls to direct calls, and
therefore avoids the retpoline. The dispatcher is generated using the
BPF JIT, and relies on text poking provided by bpf_arch_text_poke().

The dispatcher hijacks a trampoline function it via the __fentry__ nop
of the trampoline. One dispatcher instance currently supports up to 64
dispatch points. A user creates a dispatcher with its corresponding
trampoline with the DEFINE_BPF_DISPATCHER macro.

Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191213175112.30208-3-bjorn.topel@gmail.com
---
 arch/x86/net/bpf_jit_comp.c | 122 ++++++++++++++++++++++++++++++++++
 include/linux/bpf.h         |  56 ++++++++++++++++
 kernel/bpf/Makefile         |   1 +
 kernel/bpf/dispatcher.c     | 158 ++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 337 insertions(+)
 create mode 100644 kernel/bpf/dispatcher.c

diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c
index b8be18427277..3ce7ad41bd6f 100644
--- a/arch/x86/net/bpf_jit_comp.c
+++ b/arch/x86/net/bpf_jit_comp.c
@@ -10,10 +10,12 @@
 #include <linux/if_vlan.h>
 #include <linux/bpf.h>
 #include <linux/memory.h>
+#include <linux/sort.h>
 #include <asm/extable.h>
 #include <asm/set_memory.h>
 #include <asm/nospec-branch.h>
 #include <asm/text-patching.h>
+#include <asm/asm-prototypes.h>
 
 static u8 *emit_code(u8 *ptr, u32 bytes, unsigned int len)
 {
@@ -1530,6 +1532,126 @@ int arch_prepare_bpf_trampoline(void *image, struct btf_func_model *m, u32 flags
 	return 0;
 }
 
+static int emit_cond_near_jump(u8 **pprog, void *func, void *ip, u8 jmp_cond)
+{
+	u8 *prog = *pprog;
+	int cnt = 0;
+	s64 offset;
+
+	offset = func - (ip + 2 + 4);
+	if (!is_simm32(offset)) {
+		pr_err("Target %p is out of range\n", func);
+		return -EINVAL;
+	}
+	EMIT2_off32(0x0F, jmp_cond + 0x10, offset);
+	*pprog = prog;
+	return 0;
+}
+
+static int emit_fallback_jump(u8 **pprog)
+{
+	u8 *prog = *pprog;
+	int err = 0;
+
+#ifdef CONFIG_RETPOLINE
+	/* Note that this assumes the the compiler uses external
+	 * thunks for indirect calls. Both clang and GCC use the same
+	 * naming convention for external thunks.
+	 */
+	err = emit_jump(&prog, __x86_indirect_thunk_rdx, prog);
+#else
+	int cnt = 0;
+
+	EMIT2(0xFF, 0xE2);	/* jmp rdx */
+#endif
+	*pprog = prog;
+	return err;
+}
+
+static int emit_bpf_dispatcher(u8 **pprog, int a, int b, s64 *progs)
+{
+	int pivot, err, jg_bytes = 1, cnt = 0;
+	u8 *jg_reloc, *prog = *pprog;
+	s64 jg_offset;
+
+	if (a == b) {
+		/* Leaf node of recursion, i.e. not a range of indices
+		 * anymore.
+		 */
+		EMIT1(add_1mod(0x48, BPF_REG_3));	/* cmp rdx,func */
+		if (!is_simm32(progs[a]))
+			return -1;
+		EMIT2_off32(0x81, add_1reg(0xF8, BPF_REG_3),
+			    progs[a]);
+		err = emit_cond_near_jump(&prog,	/* je func */
+					  (void *)progs[a], prog,
+					  X86_JE);
+		if (err)
+			return err;
+
+		err = emit_fallback_jump(&prog);	/* jmp thunk/indirect */
+		if (err)
+			return err;
+
+		*pprog = prog;
+		return 0;
+	}
+
+	/* Not a leaf node, so we pivot, and recursively descend into
+	 * the lower and upper ranges.
+	 */
+	pivot = (b - a) / 2;
+	EMIT1(add_1mod(0x48, BPF_REG_3));		/* cmp rdx,func */
+	if (!is_simm32(progs[a + pivot]))
+		return -1;
+	EMIT2_off32(0x81, add_1reg(0xF8, BPF_REG_3), progs[a + pivot]);
+
+	if (pivot > 2) {				/* jg upper_part */
+		/* Require near jump. */
+		jg_bytes = 4;
+		EMIT2_off32(0x0F, X86_JG + 0x10, 0);
+	} else {
+		EMIT2(X86_JG, 0);
+	}
+	jg_reloc = prog;
+
+	err = emit_bpf_dispatcher(&prog, a, a + pivot,	/* emit lower_part */
+				  progs);
+	if (err)
+		return err;
+
+	jg_offset = prog - jg_reloc;
+	emit_code(jg_reloc - jg_bytes, jg_offset, jg_bytes);
+
+	err = emit_bpf_dispatcher(&prog, a + pivot + 1,	/* emit upper_part */
+				  b, progs);
+	if (err)
+		return err;
+
+	*pprog = prog;
+	return 0;
+}
+
+static int cmp_ips(const void *a, const void *b)
+{
+	const s64 *ipa = a;
+	const s64 *ipb = b;
+
+	if (*ipa > *ipb)
+		return 1;
+	if (*ipa < *ipb)
+		return -1;
+	return 0;
+}
+
+int arch_prepare_bpf_dispatcher(void *image, s64 *funcs, int num_funcs)
+{
+	u8 *prog = image;
+
+	sort(funcs, num_funcs, sizeof(funcs[0]), cmp_ips, NULL);
+	return emit_bpf_dispatcher(&prog, 0, num_funcs - 1, funcs);
+}
+
 struct x64_jit_data {
 	struct bpf_binary_header *header;
 	int *addrs;
diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 5d744828b399..53ae4a50abe4 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -470,12 +470,61 @@ struct bpf_trampoline {
 	void *image;
 	u64 selector;
 };
+
+#define BPF_DISPATCHER_MAX 64 /* Fits in 2048B */
+
+struct bpf_dispatcher_prog {
+	struct bpf_prog *prog;
+	refcount_t users;
+};
+
+struct bpf_dispatcher {
+	/* dispatcher mutex */
+	struct mutex mutex;
+	void *func;
+	struct bpf_dispatcher_prog progs[BPF_DISPATCHER_MAX];
+	int num_progs;
+	void *image;
+	u32 image_off;
+};
+
 #ifdef CONFIG_BPF_JIT
 struct bpf_trampoline *bpf_trampoline_lookup(u64 key);
 int bpf_trampoline_link_prog(struct bpf_prog *prog);
 int bpf_trampoline_unlink_prog(struct bpf_prog *prog);
 void bpf_trampoline_put(struct bpf_trampoline *tr);
 void *bpf_jit_alloc_exec_page(void);
+#define BPF_DISPATCHER_INIT(name) {			\
+	.mutex = __MUTEX_INITIALIZER(name.mutex),	\
+	.func = &name##func,				\
+	.progs = {},					\
+	.num_progs = 0,					\
+	.image = NULL,					\
+	.image_off = 0					\
+}
+
+#define DEFINE_BPF_DISPATCHER(name)					\
+	noinline unsigned int name##func(				\
+		const void *ctx,					\
+		const struct bpf_insn *insnsi,				\
+		unsigned int (*bpf_func)(const void *,			\
+					 const struct bpf_insn *))	\
+	{								\
+		return bpf_func(ctx, insnsi);				\
+	}								\
+	EXPORT_SYMBOL(name##func);			\
+	struct bpf_dispatcher name = BPF_DISPATCHER_INIT(name);
+#define DECLARE_BPF_DISPATCHER(name)					\
+	unsigned int name##func(					\
+		const void *ctx,					\
+		const struct bpf_insn *insnsi,				\
+		unsigned int (*bpf_func)(const void *,			\
+					 const struct bpf_insn *));	\
+	extern struct bpf_dispatcher name;
+#define BPF_DISPATCHER_FUNC(name) name##func
+#define BPF_DISPATCHER_PTR(name) (&name)
+void bpf_dispatcher_change_prog(struct bpf_dispatcher *d, struct bpf_prog *from,
+				struct bpf_prog *to);
 #else
 static inline struct bpf_trampoline *bpf_trampoline_lookup(u64 key)
 {
@@ -490,6 +539,13 @@ static inline int bpf_trampoline_unlink_prog(struct bpf_prog *prog)
 	return -ENOTSUPP;
 }
 static inline void bpf_trampoline_put(struct bpf_trampoline *tr) {}
+#define DEFINE_BPF_DISPATCHER(name)
+#define DECLARE_BPF_DISPATCHER(name)
+#define BPF_DISPATCHER_FUNC(name) bpf_dispatcher_nopfunc
+#define BPF_DISPATCHER_PTR(name) NULL
+static inline void bpf_dispatcher_change_prog(struct bpf_dispatcher *d,
+					      struct bpf_prog *from,
+					      struct bpf_prog *to) {}
 #endif
 
 struct bpf_func_info_aux {
diff --git a/kernel/bpf/Makefile b/kernel/bpf/Makefile
index 3f671bf617e8..d4f330351f87 100644
--- a/kernel/bpf/Makefile
+++ b/kernel/bpf/Makefile
@@ -8,6 +8,7 @@ obj-$(CONFIG_BPF_SYSCALL) += local_storage.o queue_stack_maps.o
 obj-$(CONFIG_BPF_SYSCALL) += disasm.o
 obj-$(CONFIG_BPF_JIT) += trampoline.o
 obj-$(CONFIG_BPF_SYSCALL) += btf.o
+obj-$(CONFIG_BPF_JIT) += dispatcher.o
 ifeq ($(CONFIG_NET),y)
 obj-$(CONFIG_BPF_SYSCALL) += devmap.o
 obj-$(CONFIG_BPF_SYSCALL) += cpumap.o
diff --git a/kernel/bpf/dispatcher.c b/kernel/bpf/dispatcher.c
new file mode 100644
index 000000000000..204ee61a3904
--- /dev/null
+++ b/kernel/bpf/dispatcher.c
@@ -0,0 +1,158 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright(c) 2019 Intel Corporation. */
+
+#include <linux/hash.h>
+#include <linux/bpf.h>
+#include <linux/filter.h>
+
+/* The BPF dispatcher is a multiway branch code generator. The
+ * dispatcher is a mechanism to avoid the performance penalty of an
+ * indirect call, which is expensive when retpolines are enabled. A
+ * dispatch client registers a BPF program into the dispatcher, and if
+ * there is available room in the dispatcher a direct call to the BPF
+ * program will be generated. All calls to the BPF programs called via
+ * the dispatcher will then be a direct call, instead of an
+ * indirect. The dispatcher hijacks a trampoline function it via the
+ * __fentry__ of the trampoline. The trampoline function has the
+ * following signature:
+ *
+ * unsigned int trampoline(const void *ctx, const struct bpf_insn *insnsi,
+ *                         unsigned int (*bpf_func)(const void *,
+ *                                                  const struct bpf_insn *));
+ */
+
+static struct bpf_dispatcher_prog *bpf_dispatcher_find_prog(
+	struct bpf_dispatcher *d, struct bpf_prog *prog)
+{
+	int i;
+
+	for (i = 0; i < BPF_DISPATCHER_MAX; i++) {
+		if (prog == d->progs[i].prog)
+			return &d->progs[i];
+	}
+	return NULL;
+}
+
+static struct bpf_dispatcher_prog *bpf_dispatcher_find_free(
+	struct bpf_dispatcher *d)
+{
+	return bpf_dispatcher_find_prog(d, NULL);
+}
+
+static bool bpf_dispatcher_add_prog(struct bpf_dispatcher *d,
+				    struct bpf_prog *prog)
+{
+	struct bpf_dispatcher_prog *entry;
+
+	if (!prog)
+		return false;
+
+	entry = bpf_dispatcher_find_prog(d, prog);
+	if (entry) {
+		refcount_inc(&entry->users);
+		return false;
+	}
+
+	entry = bpf_dispatcher_find_free(d);
+	if (!entry)
+		return false;
+
+	bpf_prog_inc(prog);
+	entry->prog = prog;
+	refcount_set(&entry->users, 1);
+	d->num_progs++;
+	return true;
+}
+
+static bool bpf_dispatcher_remove_prog(struct bpf_dispatcher *d,
+				       struct bpf_prog *prog)
+{
+	struct bpf_dispatcher_prog *entry;
+
+	if (!prog)
+		return false;
+
+	entry = bpf_dispatcher_find_prog(d, prog);
+	if (!entry)
+		return false;
+
+	if (refcount_dec_and_test(&entry->users)) {
+		entry->prog = NULL;
+		bpf_prog_put(prog);
+		d->num_progs--;
+		return true;
+	}
+	return false;
+}
+
+int __weak arch_prepare_bpf_dispatcher(void *image, s64 *funcs, int num_funcs)
+{
+	return -ENOTSUPP;
+}
+
+static int bpf_dispatcher_prepare(struct bpf_dispatcher *d, void *image)
+{
+	s64 ips[BPF_DISPATCHER_MAX] = {}, *ipsp = &ips[0];
+	int i;
+
+	for (i = 0; i < BPF_DISPATCHER_MAX; i++) {
+		if (d->progs[i].prog)
+			*ipsp++ = (s64)(uintptr_t)d->progs[i].prog->bpf_func;
+	}
+	return arch_prepare_bpf_dispatcher(image, &ips[0], d->num_progs);
+}
+
+static void bpf_dispatcher_update(struct bpf_dispatcher *d, int prev_num_progs)
+{
+	void *old, *new;
+	u32 noff;
+	int err;
+
+	if (!prev_num_progs) {
+		old = NULL;
+		noff = 0;
+	} else {
+		old = d->image + d->image_off;
+		noff = d->image_off ^ (PAGE_SIZE / 2);
+	}
+
+	new = d->num_progs ? d->image + noff : NULL;
+	if (new) {
+		if (bpf_dispatcher_prepare(d, new))
+			return;
+	}
+
+	err = bpf_arch_text_poke(d->func, BPF_MOD_JUMP, old, new);
+	if (err || !new)
+		return;
+
+	d->image_off = noff;
+}
+
+void bpf_dispatcher_change_prog(struct bpf_dispatcher *d, struct bpf_prog *from,
+				struct bpf_prog *to)
+{
+	bool changed = false;
+	int prev_num_progs;
+
+	if (from == to)
+		return;
+
+	mutex_lock(&d->mutex);
+	if (!d->image) {
+		d->image = bpf_jit_alloc_exec_page();
+		if (!d->image)
+			goto out;
+	}
+
+	prev_num_progs = d->num_progs;
+	changed |= bpf_dispatcher_remove_prog(d, from);
+	changed |= bpf_dispatcher_add_prog(d, to);
+
+	if (!changed)
+		goto out;
+
+	bpf_dispatcher_update(d, prev_num_progs);
+out:
+	mutex_unlock(&d->mutex);
+}
-- 
cgit v1.2.3-59-g8ed1b


From 7e6897f95935973c3253fd756135b5ea58043dc8 Mon Sep 17 00:00:00 2001
From: Björn Töpel <bjorn.topel@intel.com>
Date: Fri, 13 Dec 2019 18:51:09 +0100
Subject: bpf, xdp: Start using the BPF dispatcher for XDP
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This commit adds a BPF dispatcher for XDP. The dispatcher is updated
from the XDP control-path, dev_xdp_install(), and used when an XDP
program is run via bpf_prog_run_xdp().

Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191213175112.30208-4-bjorn.topel@gmail.com
---
 include/linux/bpf.h    | 15 +++++++++++++++
 include/linux/filter.h | 40 ++++++++++++++++++++++++----------------
 kernel/bpf/syscall.c   | 26 ++++++++++++++++++--------
 net/core/dev.c         | 19 ++++++++++++++++++-
 net/core/filter.c      |  8 ++++++++
 5 files changed, 83 insertions(+), 25 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 53ae4a50abe4..5970989b99d1 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -488,6 +488,14 @@ struct bpf_dispatcher {
 	u32 image_off;
 };
 
+static __always_inline unsigned int bpf_dispatcher_nopfunc(
+	const void *ctx,
+	const struct bpf_insn *insnsi,
+	unsigned int (*bpf_func)(const void *,
+				 const struct bpf_insn *))
+{
+	return bpf_func(ctx, insnsi);
+}
 #ifdef CONFIG_BPF_JIT
 struct bpf_trampoline *bpf_trampoline_lookup(u64 key);
 int bpf_trampoline_link_prog(struct bpf_prog *prog);
@@ -997,6 +1005,8 @@ int btf_distill_func_proto(struct bpf_verifier_log *log,
 
 int btf_check_func_arg_match(struct bpf_verifier_env *env, int subprog);
 
+struct bpf_prog *bpf_prog_by_id(u32 id);
+
 #else /* !CONFIG_BPF_SYSCALL */
 static inline struct bpf_prog *bpf_prog_get(u32 ufd)
 {
@@ -1128,6 +1138,11 @@ static inline int bpf_prog_test_run_flow_dissector(struct bpf_prog *prog,
 static inline void bpf_map_put(struct bpf_map *map)
 {
 }
+
+static inline struct bpf_prog *bpf_prog_by_id(u32 id)
+{
+	return ERR_PTR(-ENOTSUPP);
+}
 #endif /* CONFIG_BPF_SYSCALL */
 
 static inline struct bpf_prog *bpf_prog_get_type(u32 ufd,
diff --git a/include/linux/filter.h b/include/linux/filter.h
index a141cb07e76a..37ac7025031d 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h
@@ -559,23 +559,26 @@ struct sk_filter {
 
 DECLARE_STATIC_KEY_FALSE(bpf_stats_enabled_key);
 
-#define BPF_PROG_RUN(prog, ctx)	({				\
-	u32 ret;						\
-	cant_sleep();						\
-	if (static_branch_unlikely(&bpf_stats_enabled_key)) {	\
-		struct bpf_prog_stats *stats;			\
-		u64 start = sched_clock();			\
-		ret = (*(prog)->bpf_func)(ctx, (prog)->insnsi);	\
-		stats = this_cpu_ptr(prog->aux->stats);		\
-		u64_stats_update_begin(&stats->syncp);		\
-		stats->cnt++;					\
-		stats->nsecs += sched_clock() - start;		\
-		u64_stats_update_end(&stats->syncp);		\
-	} else {						\
-		ret = (*(prog)->bpf_func)(ctx, (prog)->insnsi);	\
-	}							\
+#define __BPF_PROG_RUN(prog, ctx, dfunc)	({			\
+	u32 ret;							\
+	cant_sleep();							\
+	if (static_branch_unlikely(&bpf_stats_enabled_key)) {		\
+		struct bpf_prog_stats *stats;				\
+		u64 start = sched_clock();				\
+		ret = dfunc(ctx, (prog)->insnsi, (prog)->bpf_func);	\
+		stats = this_cpu_ptr(prog->aux->stats);			\
+		u64_stats_update_begin(&stats->syncp);			\
+		stats->cnt++;						\
+		stats->nsecs += sched_clock() - start;			\
+		u64_stats_update_end(&stats->syncp);			\
+	} else {							\
+		ret = dfunc(ctx, (prog)->insnsi, (prog)->bpf_func);	\
+	}								\
 	ret; })
 
+#define BPF_PROG_RUN(prog, ctx) __BPF_PROG_RUN(prog, ctx,		\
+					       bpf_dispatcher_nopfunc)
+
 #define BPF_SKB_CB_LEN QDISC_CB_PRIV_LEN
 
 struct bpf_skb_data_end {
@@ -699,6 +702,8 @@ static inline u32 bpf_prog_run_clear_cb(const struct bpf_prog *prog,
 	return res;
 }
 
+DECLARE_BPF_DISPATCHER(bpf_dispatcher_xdp)
+
 static __always_inline u32 bpf_prog_run_xdp(const struct bpf_prog *prog,
 					    struct xdp_buff *xdp)
 {
@@ -708,9 +713,12 @@ static __always_inline u32 bpf_prog_run_xdp(const struct bpf_prog *prog,
 	 * already takes rcu_read_lock() when fetching the program, so
 	 * it's not necessary here anymore.
 	 */
-	return BPF_PROG_RUN(prog, xdp);
+	return __BPF_PROG_RUN(prog, xdp,
+			      BPF_DISPATCHER_FUNC(bpf_dispatcher_xdp));
 }
 
+void bpf_prog_change_xdp(struct bpf_prog *prev_prog, struct bpf_prog *prog);
+
 static inline u32 bpf_prog_insn_size(const struct bpf_prog *prog)
 {
 	return prog->len * sizeof(struct bpf_insn);
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 66b90eaf99fe..b08c362f4e02 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -2338,17 +2338,12 @@ static int bpf_obj_get_next_id(const union bpf_attr *attr,
 
 #define BPF_PROG_GET_FD_BY_ID_LAST_FIELD prog_id
 
-static int bpf_prog_get_fd_by_id(const union bpf_attr *attr)
+struct bpf_prog *bpf_prog_by_id(u32 id)
 {
 	struct bpf_prog *prog;
-	u32 id = attr->prog_id;
-	int fd;
-
-	if (CHECK_ATTR(BPF_PROG_GET_FD_BY_ID))
-		return -EINVAL;
 
-	if (!capable(CAP_SYS_ADMIN))
-		return -EPERM;
+	if (!id)
+		return ERR_PTR(-ENOENT);
 
 	spin_lock_bh(&prog_idr_lock);
 	prog = idr_find(&prog_idr, id);
@@ -2357,7 +2352,22 @@ static int bpf_prog_get_fd_by_id(const union bpf_attr *attr)
 	else
 		prog = ERR_PTR(-ENOENT);
 	spin_unlock_bh(&prog_idr_lock);
+	return prog;
+}
+
+static int bpf_prog_get_fd_by_id(const union bpf_attr *attr)
+{
+	struct bpf_prog *prog;
+	u32 id = attr->prog_id;
+	int fd;
+
+	if (CHECK_ATTR(BPF_PROG_GET_FD_BY_ID))
+		return -EINVAL;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
 
+	prog = bpf_prog_by_id(id);
 	if (IS_ERR(prog))
 		return PTR_ERR(prog);
 
diff --git a/net/core/dev.c b/net/core/dev.c
index 2c277b8aba38..255d3cf35360 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -8542,7 +8542,17 @@ static int dev_xdp_install(struct net_device *dev, bpf_op_t bpf_op,
 			   struct netlink_ext_ack *extack, u32 flags,
 			   struct bpf_prog *prog)
 {
+	bool non_hw = !(flags & XDP_FLAGS_HW_MODE);
+	struct bpf_prog *prev_prog = NULL;
 	struct netdev_bpf xdp;
+	int err;
+
+	if (non_hw) {
+		prev_prog = bpf_prog_by_id(__dev_xdp_query(dev, bpf_op,
+							   XDP_QUERY_PROG));
+		if (IS_ERR(prev_prog))
+			prev_prog = NULL;
+	}
 
 	memset(&xdp, 0, sizeof(xdp));
 	if (flags & XDP_FLAGS_HW_MODE)
@@ -8553,7 +8563,14 @@ static int dev_xdp_install(struct net_device *dev, bpf_op_t bpf_op,
 	xdp.flags = flags;
 	xdp.prog = prog;
 
-	return bpf_op(dev, &xdp);
+	err = bpf_op(dev, &xdp);
+	if (!err && non_hw)
+		bpf_prog_change_xdp(prev_prog, prog);
+
+	if (prev_prog)
+		bpf_prog_put(prev_prog);
+
+	return err;
 }
 
 static void dev_xdp_uninstall(struct net_device *dev)
diff --git a/net/core/filter.c b/net/core/filter.c
index f1e703eed3d2..a411f7835dee 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -8940,3 +8940,11 @@ const struct bpf_verifier_ops sk_reuseport_verifier_ops = {
 const struct bpf_prog_ops sk_reuseport_prog_ops = {
 };
 #endif /* CONFIG_INET */
+
+DEFINE_BPF_DISPATCHER(bpf_dispatcher_xdp)
+
+void bpf_prog_change_xdp(struct bpf_prog *prev_prog, struct bpf_prog *prog)
+{
+	bpf_dispatcher_change_prog(BPF_DISPATCHER_PTR(bpf_dispatcher_xdp),
+				   prev_prog, prog);
+}
-- 
cgit v1.2.3-59-g8ed1b


From f23c4b3924d2e9382820ee677b68d42d5dd7b08b Mon Sep 17 00:00:00 2001
From: Björn Töpel <bjorn.topel@intel.com>
Date: Fri, 13 Dec 2019 18:51:10 +0100
Subject: bpf: Start using the BPF dispatcher in BPF_TEST_RUN
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

In order to properly exercise the BPF dispatcher, this commit adds BPF
dispatcher usage to BPF_TEST_RUN when executing XDP programs.

Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191213175112.30208-5-bjorn.topel@gmail.com
---
 net/bpf/test_run.c | 15 ++++++++++-----
 1 file changed, 10 insertions(+), 5 deletions(-)

diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c
index 85c8cbbada92..5016c538d3ce 100644
--- a/net/bpf/test_run.c
+++ b/net/bpf/test_run.c
@@ -15,7 +15,7 @@
 #include <trace/events/bpf_test_run.h>
 
 static int bpf_test_run(struct bpf_prog *prog, void *ctx, u32 repeat,
-			u32 *retval, u32 *time)
+			u32 *retval, u32 *time, bool xdp)
 {
 	struct bpf_cgroup_storage *storage[MAX_BPF_CGROUP_STORAGE_TYPE] = { NULL };
 	enum bpf_cgroup_storage_type stype;
@@ -41,7 +41,11 @@ static int bpf_test_run(struct bpf_prog *prog, void *ctx, u32 repeat,
 	time_start = ktime_get_ns();
 	for (i = 0; i < repeat; i++) {
 		bpf_cgroup_storage_set(storage);
-		*retval = BPF_PROG_RUN(prog, ctx);
+
+		if (xdp)
+			*retval = bpf_prog_run_xdp(prog, ctx);
+		else
+			*retval = BPF_PROG_RUN(prog, ctx);
 
 		if (signal_pending(current)) {
 			ret = -EINTR;
@@ -356,7 +360,7 @@ int bpf_prog_test_run_skb(struct bpf_prog *prog, const union bpf_attr *kattr,
 	ret = convert___skb_to_skb(skb, ctx);
 	if (ret)
 		goto out;
-	ret = bpf_test_run(prog, skb, repeat, &retval, &duration);
+	ret = bpf_test_run(prog, skb, repeat, &retval, &duration, false);
 	if (ret)
 		goto out;
 	if (!is_l2) {
@@ -413,8 +417,8 @@ int bpf_prog_test_run_xdp(struct bpf_prog *prog, const union bpf_attr *kattr,
 
 	rxqueue = __netif_get_rx_queue(current->nsproxy->net_ns->loopback_dev, 0);
 	xdp.rxq = &rxqueue->xdp_rxq;
-
-	ret = bpf_test_run(prog, &xdp, repeat, &retval, &duration);
+	bpf_prog_change_xdp(NULL, prog);
+	ret = bpf_test_run(prog, &xdp, repeat, &retval, &duration, true);
 	if (ret)
 		goto out;
 	if (xdp.data != data + XDP_PACKET_HEADROOM + NET_IP_ALIGN ||
@@ -422,6 +426,7 @@ int bpf_prog_test_run_xdp(struct bpf_prog *prog, const union bpf_attr *kattr,
 		size = xdp.data_end - xdp.data;
 	ret = bpf_test_finish(kattr, uattr, xdp.data, size, retval, duration);
 out:
+	bpf_prog_change_xdp(prog, NULL);
 	kfree(data);
 	return ret;
 }
-- 
cgit v1.2.3-59-g8ed1b


From e754f5a6e36b63d6732f52adcbe4c447fd66896f Mon Sep 17 00:00:00 2001
From: Björn Töpel <bjorn.topel@intel.com>
Date: Fri, 13 Dec 2019 18:51:11 +0100
Subject: selftests: bpf: Add xdp_perf test
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The xdp_perf is a dummy XDP test, only used to measure the the cost of
jumping into a naive XDP program one million times.

To build and run the program:
  $ cd tools/testing/selftests/bpf
  $ make
  $ ./test_progs -v -t xdp_perf

Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191213175112.30208-6-bjorn.topel@gmail.com
---
 tools/testing/selftests/bpf/prog_tests/xdp_perf.c | 25 +++++++++++++++++++++++
 1 file changed, 25 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/xdp_perf.c

diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_perf.c b/tools/testing/selftests/bpf/prog_tests/xdp_perf.c
new file mode 100644
index 000000000000..7185bee16fe4
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/xdp_perf.c
@@ -0,0 +1,25 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <test_progs.h>
+
+void test_xdp_perf(void)
+{
+	const char *file = "./xdp_dummy.o";
+	__u32 duration, retval, size;
+	struct bpf_object *obj;
+	char in[128], out[128];
+	int err, prog_fd;
+
+	err = bpf_prog_load(file, BPF_PROG_TYPE_XDP, &obj, &prog_fd);
+	if (CHECK_FAIL(err))
+		return;
+
+	err = bpf_prog_test_run(prog_fd, 1000000, &in[0], 128,
+				out, &size, &retval, &duration);
+
+	CHECK(err || retval != XDP_PASS || size != 128,
+	      "xdp-perf",
+	      "err %d errno %d retval %d size %d\n",
+	      err, errno, retval, size);
+
+	bpf_object__close(obj);
+}
-- 
cgit v1.2.3-59-g8ed1b


From 116eb788f57c9c35c40b29cfaa2607020de99a84 Mon Sep 17 00:00:00 2001
From: Björn Töpel <bjorn.topel@intel.com>
Date: Fri, 13 Dec 2019 18:51:12 +0100
Subject: bpf, x86: Align dispatcher branch targets to 16B
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

>From Intel 64 and IA-32 Architectures Optimization Reference Manual,
3.4.1.4 Code Alignment, Assembly/Compiler Coding Rule 11: All branch
targets should be 16-byte aligned.

This commits aligns branch targets according to the Intel manual.

The nops used to align branch targets make the dispatcher larger, and
therefore the number of supported dispatch points/programs are
descreased from 64 to 48.

Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191213175112.30208-7-bjorn.topel@gmail.com
---
 arch/x86/net/bpf_jit_comp.c | 30 +++++++++++++++++++++++++++++-
 include/linux/bpf.h         |  2 +-
 2 files changed, 30 insertions(+), 2 deletions(-)

diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c
index 3ce7ad41bd6f..4c8a2d1f8470 100644
--- a/arch/x86/net/bpf_jit_comp.c
+++ b/arch/x86/net/bpf_jit_comp.c
@@ -1548,6 +1548,26 @@ static int emit_cond_near_jump(u8 **pprog, void *func, void *ip, u8 jmp_cond)
 	return 0;
 }
 
+static void emit_nops(u8 **pprog, unsigned int len)
+{
+	unsigned int i, noplen;
+	u8 *prog = *pprog;
+	int cnt = 0;
+
+	while (len > 0) {
+		noplen = len;
+
+		if (noplen > ASM_NOP_MAX)
+			noplen = ASM_NOP_MAX;
+
+		for (i = 0; i < noplen; i++)
+			EMIT1(ideal_nops[noplen][i]);
+		len -= noplen;
+	}
+
+	*pprog = prog;
+}
+
 static int emit_fallback_jump(u8 **pprog)
 {
 	u8 *prog = *pprog;
@@ -1570,8 +1590,8 @@ static int emit_fallback_jump(u8 **pprog)
 
 static int emit_bpf_dispatcher(u8 **pprog, int a, int b, s64 *progs)
 {
+	u8 *jg_reloc, *jg_target, *prog = *pprog;
 	int pivot, err, jg_bytes = 1, cnt = 0;
-	u8 *jg_reloc, *prog = *pprog;
 	s64 jg_offset;
 
 	if (a == b) {
@@ -1620,6 +1640,14 @@ static int emit_bpf_dispatcher(u8 **pprog, int a, int b, s64 *progs)
 	if (err)
 		return err;
 
+	/* From Intel 64 and IA-32 Architectures Optimization
+	 * Reference Manual, 3.4.1.4 Code Alignment, Assembly/Compiler
+	 * Coding Rule 11: All branch targets should be 16-byte
+	 * aligned.
+	 */
+	jg_target = PTR_ALIGN(prog, 16);
+	if (jg_target != prog)
+		emit_nops(&prog, jg_target - prog);
 	jg_offset = prog - jg_reloc;
 	emit_code(jg_reloc - jg_bytes, jg_offset, jg_bytes);
 
diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 5970989b99d1..d467983e61bb 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -471,7 +471,7 @@ struct bpf_trampoline {
 	u64 selector;
 };
 
-#define BPF_DISPATCHER_MAX 64 /* Fits in 2048B */
+#define BPF_DISPATCHER_MAX 48 /* Fits in 2048B */
 
 struct bpf_dispatcher_prog {
 	struct bpf_prog *prog;
-- 
cgit v1.2.3-59-g8ed1b


From 850a88cc4096fe1df407452ba2e4d28cf5b3eee9 Mon Sep 17 00:00:00 2001
From: Stanislav Fomichev <sdf@google.com>
Date: Fri, 13 Dec 2019 14:30:27 -0800
Subject: bpf: Expose __sk_buff wire_len/gso_segs to BPF_PROG_TEST_RUN

wire_len should not be less than real len and is capped by GSO_MAX_SIZE.
gso_segs is capped by GSO_MAX_SEGS.

v2:
* set wire_len to skb->len when passed wire_len is 0 (Alexei Starovoitov)

Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Cc: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20191213223028.161282-1-sdf@google.com
---
 net/bpf/test_run.c | 19 ++++++++++++++++++-
 1 file changed, 18 insertions(+), 1 deletion(-)

diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c
index 5016c538d3ce..93a9c87787e0 100644
--- a/net/bpf/test_run.c
+++ b/net/bpf/test_run.c
@@ -267,8 +267,10 @@ static int convert___skb_to_skb(struct sk_buff *skb, struct __sk_buff *__skb)
 		return -EINVAL;
 
 	/* tstamp is allowed */
+	/* wire_len is allowed */
+	/* gso_segs is allowed */
 
-	if (!range_is_zero(__skb, offsetofend(struct __sk_buff, tstamp),
+	if (!range_is_zero(__skb, offsetofend(struct __sk_buff, gso_segs),
 			   sizeof(struct __sk_buff)))
 		return -EINVAL;
 
@@ -276,6 +278,19 @@ static int convert___skb_to_skb(struct sk_buff *skb, struct __sk_buff *__skb)
 	skb->tstamp = __skb->tstamp;
 	memcpy(&cb->data, __skb->cb, QDISC_CB_PRIV_LEN);
 
+	if (__skb->wire_len == 0) {
+		cb->pkt_len = skb->len;
+	} else {
+		if (__skb->wire_len < skb->len ||
+		    __skb->wire_len > GSO_MAX_SIZE)
+			return -EINVAL;
+		cb->pkt_len = __skb->wire_len;
+	}
+
+	if (__skb->gso_segs > GSO_MAX_SEGS)
+		return -EINVAL;
+	skb_shinfo(skb)->gso_segs = __skb->gso_segs;
+
 	return 0;
 }
 
@@ -289,6 +304,8 @@ static void convert_skb_to___skb(struct sk_buff *skb, struct __sk_buff *__skb)
 	__skb->priority = skb->priority;
 	__skb->tstamp = skb->tstamp;
 	memcpy(__skb->cb, &cb->data, QDISC_CB_PRIV_LEN);
+	__skb->wire_len = cb->pkt_len;
+	__skb->gso_segs = skb_shinfo(skb)->gso_segs;
 }
 
 int bpf_prog_test_run_skb(struct bpf_prog *prog, const union bpf_attr *kattr,
-- 
cgit v1.2.3-59-g8ed1b


From a06bf42f5a95ff462401d59d0c08cf4620213647 Mon Sep 17 00:00:00 2001
From: Stanislav Fomichev <sdf@google.com>
Date: Fri, 13 Dec 2019 14:30:28 -0800
Subject: selftests/bpf: Test wire_len/gso_segs in BPF_PROG_TEST_RUN

Make sure we can pass arbitrary data in wire_len/gso_segs.

Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20191213223028.161282-2-sdf@google.com
---
 tools/testing/selftests/bpf/prog_tests/skb_ctx.c | 2 ++
 tools/testing/selftests/bpf/progs/test_skb_ctx.c | 5 +++++
 2 files changed, 7 insertions(+)

diff --git a/tools/testing/selftests/bpf/prog_tests/skb_ctx.c b/tools/testing/selftests/bpf/prog_tests/skb_ctx.c
index a2eb8db8dafb..edf5e8c7d400 100644
--- a/tools/testing/selftests/bpf/prog_tests/skb_ctx.c
+++ b/tools/testing/selftests/bpf/prog_tests/skb_ctx.c
@@ -11,6 +11,8 @@ void test_skb_ctx(void)
 		.cb[4] = 5,
 		.priority = 6,
 		.tstamp = 7,
+		.wire_len = 100,
+		.gso_segs = 8,
 	};
 	struct bpf_prog_test_run_attr tattr = {
 		.data_in = &pkt_v4,
diff --git a/tools/testing/selftests/bpf/progs/test_skb_ctx.c b/tools/testing/selftests/bpf/progs/test_skb_ctx.c
index 2a9f4c736ebc..534fbf9a7344 100644
--- a/tools/testing/selftests/bpf/progs/test_skb_ctx.c
+++ b/tools/testing/selftests/bpf/progs/test_skb_ctx.c
@@ -18,5 +18,10 @@ int process(struct __sk_buff *skb)
 	skb->priority++;
 	skb->tstamp++;
 
+	if (skb->wire_len != 100)
+		return 1;
+	if (skb->gso_segs != 8)
+		return 1;
+
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 1d1997db870f4058676439ef7014390ba9e24eb2 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <jakub.kicinski@netronome.com>
Date: Tue, 10 Dec 2019 10:20:32 -0800
Subject: Revert "nfp: abm: fix memory leak in nfp_abm_u32_knode_replace"

This reverts commit 78beef629fd9 ("nfp: abm: fix memory leak in
nfp_abm_u32_knode_replace").

The quoted commit does not fix anything and resulted in a bogus
CVE-2019-19076.

If match is NULL then it is known there is no matching entry in
list, hence, calling nfp_abm_u32_knode_delete() is pointless.

Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: John Hurley <john.hurley@netronome.com>
---
 drivers/net/ethernet/netronome/nfp/abm/cls.c | 14 ++++----------
 1 file changed, 4 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/netronome/nfp/abm/cls.c b/drivers/net/ethernet/netronome/nfp/abm/cls.c
index 9f8a1f69c0c4..23ebddfb9532 100644
--- a/drivers/net/ethernet/netronome/nfp/abm/cls.c
+++ b/drivers/net/ethernet/netronome/nfp/abm/cls.c
@@ -176,10 +176,8 @@ nfp_abm_u32_knode_replace(struct nfp_abm_link *alink,
 	u8 mask, val;
 	int err;
 
-	if (!nfp_abm_u32_check_knode(alink->abm, knode, proto, extack)) {
-		err = -EOPNOTSUPP;
+	if (!nfp_abm_u32_check_knode(alink->abm, knode, proto, extack))
 		goto err_delete;
-	}
 
 	tos_off = proto == htons(ETH_P_IP) ? 16 : 20;
 
@@ -200,18 +198,14 @@ nfp_abm_u32_knode_replace(struct nfp_abm_link *alink,
 		if ((iter->val & cmask) == (val & cmask) &&
 		    iter->band != knode->res->classid) {
 			NL_SET_ERR_MSG_MOD(extack, "conflict with already offloaded filter");
-			err = -EOPNOTSUPP;
 			goto err_delete;
 		}
 	}
 
 	if (!match) {
 		match = kzalloc(sizeof(*match), GFP_KERNEL);
-		if (!match) {
-			err = -ENOMEM;
-			goto err_delete;
-		}
-
+		if (!match)
+			return -ENOMEM;
 		list_add(&match->list, &alink->dscp_map);
 	}
 	match->handle = knode->handle;
@@ -227,7 +221,7 @@ nfp_abm_u32_knode_replace(struct nfp_abm_link *alink,
 
 err_delete:
 	nfp_abm_u32_knode_delete(alink, knode);
-	return err;
+	return -EOPNOTSUPP;
 }
 
 static int nfp_abm_setup_tc_block_cb(enum tc_setup_type type,
-- 
cgit v1.2.3-59-g8ed1b


From 826f66b30c2e3d4df533e4224cb8c734afa0a17c Mon Sep 17 00:00:00 2001
From: Andy Roulin <aroulin@cumulusnetworks.com>
Date: Wed, 11 Dec 2019 14:30:58 -0800
Subject: bonding: move 802.3ad port state flags to uapi

The bond slave actor/partner operating state is exported as
bitfield to userspace, which lacks a way to interpret it, e.g.,
iproute2 only prints the state as a number:

ad_actor_oper_port_state 15

For userspace to interpret the bitfield, the bitfield definitions
should be part of the uapi. The bitfield itself is defined in the
802.3ad standard.

This commit moves the 802.3ad bitfield definitions to uapi.

Related iproute2 patches, soon to be posted upstream, use the new uapi
headers to pretty-print bond slave state, e.g., with ip -d link show

ad_actor_oper_port_state_str <active,short_timeout,aggregating,in_sync>

Signed-off-by: Andy Roulin <aroulin@cumulusnetworks.com>
Acked-by: Roopa Prabhu <roopa@cumulusnetworks.com>
Acked-by: Jay Vosburgh <jay.vosburgh@canonical.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
 drivers/net/bonding/bond_3ad.c  | 10 ----------
 include/uapi/linux/if_bonding.h | 10 ++++++++++
 2 files changed, 10 insertions(+), 10 deletions(-)

diff --git a/drivers/net/bonding/bond_3ad.c b/drivers/net/bonding/bond_3ad.c
index e3b25f310936..34bfe99641a3 100644
--- a/drivers/net/bonding/bond_3ad.c
+++ b/drivers/net/bonding/bond_3ad.c
@@ -31,16 +31,6 @@
 #define AD_CHURN_DETECTION_TIME    60
 #define AD_AGGREGATE_WAIT_TIME     2
 
-/* Port state definitions (43.4.2.2 in the 802.3ad standard) */
-#define AD_STATE_LACP_ACTIVITY   0x1
-#define AD_STATE_LACP_TIMEOUT    0x2
-#define AD_STATE_AGGREGATION     0x4
-#define AD_STATE_SYNCHRONIZATION 0x8
-#define AD_STATE_COLLECTING      0x10
-#define AD_STATE_DISTRIBUTING    0x20
-#define AD_STATE_DEFAULTED       0x40
-#define AD_STATE_EXPIRED         0x80
-
 /* Port Variables definitions used by the State Machines (43.4.7 in the
  * 802.3ad standard)
  */
diff --git a/include/uapi/linux/if_bonding.h b/include/uapi/linux/if_bonding.h
index 790585f0e61b..6829213a54c5 100644
--- a/include/uapi/linux/if_bonding.h
+++ b/include/uapi/linux/if_bonding.h
@@ -95,6 +95,16 @@
 #define BOND_XMIT_POLICY_ENCAP23	3 /* encapsulated layer 2+3 */
 #define BOND_XMIT_POLICY_ENCAP34	4 /* encapsulated layer 3+4 */
 
+/* 802.3ad port state definitions (43.4.2.2 in the 802.3ad standard) */
+#define AD_STATE_LACP_ACTIVITY   0x1
+#define AD_STATE_LACP_TIMEOUT    0x2
+#define AD_STATE_AGGREGATION     0x4
+#define AD_STATE_SYNCHRONIZATION 0x8
+#define AD_STATE_COLLECTING      0x10
+#define AD_STATE_DISTRIBUTING    0x20
+#define AD_STATE_DEFAULTED       0x40
+#define AD_STATE_EXPIRED         0x80
+
 typedef struct ifbond {
 	__s32 bond_mode;
 	__s32 num_slaves;
-- 
cgit v1.2.3-59-g8ed1b


From 3a5f494d84187db5a86de6aeceec4ae6b58380cd Mon Sep 17 00:00:00 2001
From: zhengbin <zhengbin13@huawei.com>
Date: Sat, 14 Dec 2019 18:17:24 +0800
Subject: net: phy: dp83869: Remove unneeded semicolon

Fixes coccicheck warning:

drivers/net/phy/dp83869.c:337:2-3: Unneeded semicolon

Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: zhengbin <zhengbin13@huawei.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
 drivers/net/phy/dp83869.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/phy/dp83869.c b/drivers/net/phy/dp83869.c
index 93021904c5e4..7996a4aea8d2 100644
--- a/drivers/net/phy/dp83869.c
+++ b/drivers/net/phy/dp83869.c
@@ -334,7 +334,7 @@ static int dp83869_configure_mode(struct phy_device *phydev,
 		break;
 	default:
 		return -EINVAL;
-	};
+	}
 
 	return ret;
 }
-- 
cgit v1.2.3-59-g8ed1b


From ea6a547669b37453f2b1a5d85188d75b3613dfaa Mon Sep 17 00:00:00 2001
From: Willem de Bruijn <willemb@google.com>
Date: Thu, 12 Dec 2019 11:36:46 -0500
Subject: selftests/net: make so_txtime more robust to timer variance

The SO_TXTIME test depends on accurate timers. In some virtualized
environments the test has been reported to be flaky. This is easily
reproduced by disabling kvm acceleration in Qemu.

Allow greater variance in a run and retry to further reduce flakiness.

Observed errors are one of two kinds: either the packet arrives too
early or late at recv(), or it was dropped in the qdisc itself and the
recv() call times out.

In the latter case, the qdisc queues a notification to the error
queue of the send socket. Also explicitly report this cause.

Link: https://lore.kernel.org/netdev/CA+FuTSdYOnJCsGuj43xwV1jxvYsaoa_LzHQF9qMyhrkLrivxKw@mail.gmail.com
Reported-by: Naresh Kamboju <naresh.kamboju@linaro.org>
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
 tools/testing/selftests/net/so_txtime.c  | 84 ++++++++++++++++++++++++++++++--
 tools/testing/selftests/net/so_txtime.sh |  9 +++-
 2 files changed, 88 insertions(+), 5 deletions(-)

diff --git a/tools/testing/selftests/net/so_txtime.c b/tools/testing/selftests/net/so_txtime.c
index 34df4c8882af..383bac05ac32 100644
--- a/tools/testing/selftests/net/so_txtime.c
+++ b/tools/testing/selftests/net/so_txtime.c
@@ -12,7 +12,11 @@
 #include <arpa/inet.h>
 #include <error.h>
 #include <errno.h>
+#include <inttypes.h>
 #include <linux/net_tstamp.h>
+#include <linux/errqueue.h>
+#include <linux/ipv6.h>
+#include <linux/tcp.h>
 #include <stdbool.h>
 #include <stdlib.h>
 #include <stdio.h>
@@ -28,7 +32,7 @@ static int	cfg_clockid	= CLOCK_TAI;
 static bool	cfg_do_ipv4;
 static bool	cfg_do_ipv6;
 static uint16_t	cfg_port	= 8000;
-static int	cfg_variance_us	= 2000;
+static int	cfg_variance_us	= 4000;
 
 static uint64_t glob_tstart;
 
@@ -43,6 +47,9 @@ static struct timed_send cfg_in[MAX_NUM_PKT];
 static struct timed_send cfg_out[MAX_NUM_PKT];
 static int cfg_num_pkt;
 
+static int cfg_errq_level;
+static int cfg_errq_type;
+
 static uint64_t gettime_ns(void)
 {
 	struct timespec ts;
@@ -90,13 +97,15 @@ static void do_send_one(int fdt, struct timed_send *ts)
 
 }
 
-static void do_recv_one(int fdr, struct timed_send *ts)
+static bool do_recv_one(int fdr, struct timed_send *ts)
 {
 	int64_t tstop, texpect;
 	char rbuf[2];
 	int ret;
 
 	ret = recv(fdr, rbuf, sizeof(rbuf), 0);
+	if (ret == -1 && errno == EAGAIN)
+		return true;
 	if (ret == -1)
 		error(1, errno, "read");
 	if (ret != 1)
@@ -113,6 +122,8 @@ static void do_recv_one(int fdr, struct timed_send *ts)
 
 	if (labs(tstop - texpect) > cfg_variance_us)
 		error(1, 0, "exceeds variance (%d us)", cfg_variance_us);
+
+	return false;
 }
 
 static void do_recv_verify_empty(int fdr)
@@ -125,12 +136,70 @@ static void do_recv_verify_empty(int fdr)
 		error(1, 0, "recv: not empty as expected (%d, %d)", ret, errno);
 }
 
+static void do_recv_errqueue_timeout(int fdt)
+{
+	char control[CMSG_SPACE(sizeof(struct sock_extended_err)) +
+		     CMSG_SPACE(sizeof(struct sockaddr_in6))] = {0};
+	char data[sizeof(struct ipv6hdr) +
+		  sizeof(struct tcphdr) + 1];
+	struct sock_extended_err *err;
+	struct msghdr msg = {0};
+	struct iovec iov = {0};
+	struct cmsghdr *cm;
+	int64_t tstamp = 0;
+	int ret;
+
+	iov.iov_base = data;
+	iov.iov_len = sizeof(data);
+
+	msg.msg_iov = &iov;
+	msg.msg_iovlen = 1;
+
+	msg.msg_control = control;
+	msg.msg_controllen = sizeof(control);
+
+	while (1) {
+		ret = recvmsg(fdt, &msg, MSG_ERRQUEUE);
+		if (ret == -1 && errno == EAGAIN)
+			break;
+		if (ret == -1)
+			error(1, errno, "errqueue");
+		if (msg.msg_flags != MSG_ERRQUEUE)
+			error(1, 0, "errqueue: flags 0x%x\n", msg.msg_flags);
+
+		cm = CMSG_FIRSTHDR(&msg);
+		if (cm->cmsg_level != cfg_errq_level ||
+		    cm->cmsg_type != cfg_errq_type)
+			error(1, 0, "errqueue: type 0x%x.0x%x\n",
+				    cm->cmsg_level, cm->cmsg_type);
+
+		err = (struct sock_extended_err *)CMSG_DATA(cm);
+		if (err->ee_origin != SO_EE_ORIGIN_TXTIME)
+			error(1, 0, "errqueue: origin 0x%x\n", err->ee_origin);
+		if (err->ee_code != ECANCELED)
+			error(1, 0, "errqueue: code 0x%x\n", err->ee_code);
+
+		tstamp = ((int64_t) err->ee_data) << 32 | err->ee_info;
+		tstamp -= (int64_t) glob_tstart;
+		tstamp /= 1000 * 1000;
+		fprintf(stderr, "send: pkt %c at %" PRId64 "ms dropped\n",
+				data[ret - 1], tstamp);
+
+		msg.msg_flags = 0;
+		msg.msg_controllen = sizeof(control);
+	}
+
+	error(1, 0, "recv: timeout");
+}
+
 static void setsockopt_txtime(int fd)
 {
 	struct sock_txtime so_txtime_val = { .clockid = cfg_clockid };
 	struct sock_txtime so_txtime_val_read = { 0 };
 	socklen_t vallen = sizeof(so_txtime_val);
 
+	so_txtime_val.flags = SOF_TXTIME_REPORT_ERRORS;
+
 	if (setsockopt(fd, SOL_SOCKET, SO_TXTIME,
 		       &so_txtime_val, sizeof(so_txtime_val)))
 		error(1, errno, "setsockopt txtime");
@@ -194,7 +263,8 @@ static void do_test(struct sockaddr *addr, socklen_t alen)
 	for (i = 0; i < cfg_num_pkt; i++)
 		do_send_one(fdt, &cfg_in[i]);
 	for (i = 0; i < cfg_num_pkt; i++)
-		do_recv_one(fdr, &cfg_out[i]);
+		if (do_recv_one(fdr, &cfg_out[i]))
+			do_recv_errqueue_timeout(fdt);
 
 	do_recv_verify_empty(fdr);
 
@@ -280,6 +350,10 @@ int main(int argc, char **argv)
 		addr6.sin6_family = AF_INET6;
 		addr6.sin6_port = htons(cfg_port);
 		addr6.sin6_addr = in6addr_loopback;
+
+		cfg_errq_level = SOL_IPV6;
+		cfg_errq_type = IPV6_RECVERR;
+
 		do_test((void *)&addr6, sizeof(addr6));
 	}
 
@@ -289,6 +363,10 @@ int main(int argc, char **argv)
 		addr4.sin_family = AF_INET;
 		addr4.sin_port = htons(cfg_port);
 		addr4.sin_addr.s_addr = htonl(INADDR_LOOPBACK);
+
+		cfg_errq_level = SOL_IP;
+		cfg_errq_type = IP_RECVERR;
+
 		do_test((void *)&addr4, sizeof(addr4));
 	}
 
diff --git a/tools/testing/selftests/net/so_txtime.sh b/tools/testing/selftests/net/so_txtime.sh
index 5aa519328a5b..3f7800eaecb1 100755
--- a/tools/testing/selftests/net/so_txtime.sh
+++ b/tools/testing/selftests/net/so_txtime.sh
@@ -5,7 +5,12 @@
 
 # Run in network namespace
 if [[ $# -eq 0 ]]; then
-	./in_netns.sh $0 __subprocess
+	if ! ./in_netns.sh $0 __subprocess; then
+		# test is time sensitive, can be flaky
+		echo "test failed: retry once"
+		./in_netns.sh $0 __subprocess
+	fi
+
 	exit $?
 fi
 
@@ -18,7 +23,7 @@ tc qdisc add dev lo root fq
 ./so_txtime -4 -6 -c mono a,10,b,20 a,10,b,20
 ./so_txtime -4 -6 -c mono a,20,b,10 b,20,a,20
 
-if tc qdisc replace dev lo root etf clockid CLOCK_TAI delta 200000; then
+if tc qdisc replace dev lo root etf clockid CLOCK_TAI delta 400000; then
 	! ./so_txtime -4 -6 -c tai a,-1 a,-1
 	! ./so_txtime -4 -6 -c tai a,0 a,0
 	./so_txtime -4 -6 -c tai a,10 a,10
-- 
cgit v1.2.3-59-g8ed1b


From de1799667b002331609e8ee6b924ccfd51968d99 Mon Sep 17 00:00:00 2001
From: Vivien Didelot <vivien.didelot@gmail.com>
Date: Wed, 11 Dec 2019 20:07:10 -0500
Subject: net: bridge: add STP xstats

This adds rx_bpdu, tx_bpdu, rx_tcn, tx_tcn, transition_blk,
transition_fwd xstats counters to the bridge ports copied over via
netlink, providing useful information for STP.

Signed-off-by: Vivien Didelot <vivien.didelot@gmail.com>
Acked-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
 include/uapi/linux/if_bridge.h | 10 ++++++++++
 net/bridge/br_netlink.c        | 13 +++++++++++++
 net/bridge/br_private.h        |  2 ++
 net/bridge/br_stp.c            | 15 +++++++++++++++
 net/bridge/br_stp_bpdu.c       |  4 ++++
 5 files changed, 44 insertions(+)

diff --git a/include/uapi/linux/if_bridge.h b/include/uapi/linux/if_bridge.h
index 1b3c2b643a02..4a58e3d7de46 100644
--- a/include/uapi/linux/if_bridge.h
+++ b/include/uapi/linux/if_bridge.h
@@ -156,6 +156,15 @@ struct bridge_vlan_xstats {
 	__u32 pad2;
 };
 
+struct bridge_stp_xstats {
+	__u64 transition_blk;
+	__u64 transition_fwd;
+	__u64 rx_bpdu;
+	__u64 tx_bpdu;
+	__u64 rx_tcn;
+	__u64 tx_tcn;
+};
+
 /* Bridge multicast database attributes
  * [MDBA_MDB] = {
  *     [MDBA_MDB_ENTRY] = {
@@ -262,6 +271,7 @@ enum {
 	BRIDGE_XSTATS_VLAN,
 	BRIDGE_XSTATS_MCAST,
 	BRIDGE_XSTATS_PAD,
+	BRIDGE_XSTATS_STP,
 	__BRIDGE_XSTATS_MAX
 };
 #define BRIDGE_XSTATS_MAX (__BRIDGE_XSTATS_MAX - 1)
diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c
index a0a54482aabc..60136575aea4 100644
--- a/net/bridge/br_netlink.c
+++ b/net/bridge/br_netlink.c
@@ -1607,6 +1607,19 @@ static int br_fill_linkxstats(struct sk_buff *skb,
 		br_multicast_get_stats(br, p, nla_data(nla));
 	}
 #endif
+
+	if (p) {
+		nla = nla_reserve_64bit(skb, BRIDGE_XSTATS_STP,
+					sizeof(p->stp_xstats),
+					BRIDGE_XSTATS_PAD);
+		if (!nla)
+			goto nla_put_failure;
+
+		spin_lock_bh(&br->lock);
+		memcpy(nla_data(nla), &p->stp_xstats, sizeof(p->stp_xstats));
+		spin_unlock_bh(&br->lock);
+	}
+
 	nla_nest_end(skb, nest);
 	*prividx = 0;
 
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index 36b0367ca1e0..f540f3bdf294 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -283,6 +283,8 @@ struct net_bridge_port {
 #endif
 	u16				group_fwd_mask;
 	u16				backup_redirected_cnt;
+
+	struct bridge_stp_xstats	stp_xstats;
 };
 
 #define kobj_to_brport(obj)	container_of(obj, struct net_bridge_port, kobj)
diff --git a/net/bridge/br_stp.c b/net/bridge/br_stp.c
index 1f1410f8d312..6856a6d9282b 100644
--- a/net/bridge/br_stp.c
+++ b/net/bridge/br_stp.c
@@ -45,6 +45,17 @@ void br_set_state(struct net_bridge_port *p, unsigned int state)
 		br_info(p->br, "port %u(%s) entered %s state\n",
 				(unsigned int) p->port_no, p->dev->name,
 				br_port_state_names[p->state]);
+
+	if (p->br->stp_enabled == BR_KERNEL_STP) {
+		switch (p->state) {
+		case BR_STATE_BLOCKING:
+			p->stp_xstats.transition_blk++;
+			break;
+		case BR_STATE_FORWARDING:
+			p->stp_xstats.transition_fwd++;
+			break;
+		}
+	}
 }
 
 /* called under bridge lock */
@@ -484,6 +495,8 @@ void br_received_config_bpdu(struct net_bridge_port *p,
 	struct net_bridge *br;
 	int was_root;
 
+	p->stp_xstats.rx_bpdu++;
+
 	br = p->br;
 	was_root = br_is_root_bridge(br);
 
@@ -517,6 +530,8 @@ void br_received_config_bpdu(struct net_bridge_port *p,
 /* called under bridge lock */
 void br_received_tcn_bpdu(struct net_bridge_port *p)
 {
+	p->stp_xstats.rx_tcn++;
+
 	if (br_is_designated_port(p)) {
 		br_info(p->br, "port %u(%s) received tcn bpdu\n",
 			(unsigned int) p->port_no, p->dev->name);
diff --git a/net/bridge/br_stp_bpdu.c b/net/bridge/br_stp_bpdu.c
index 7796dd9d42d7..0e4572f31330 100644
--- a/net/bridge/br_stp_bpdu.c
+++ b/net/bridge/br_stp_bpdu.c
@@ -118,6 +118,8 @@ void br_send_config_bpdu(struct net_bridge_port *p, struct br_config_bpdu *bpdu)
 	br_set_ticks(buf+33, bpdu->forward_delay);
 
 	br_send_bpdu(p, buf, 35);
+
+	p->stp_xstats.tx_bpdu++;
 }
 
 /* called under bridge lock */
@@ -133,6 +135,8 @@ void br_send_tcn_bpdu(struct net_bridge_port *p)
 	buf[2] = 0;
 	buf[3] = BPDU_TYPE_TCN;
 	br_send_bpdu(p, buf, 4);
+
+	p->stp_xstats.tx_tcn++;
 }
 
 /*
-- 
cgit v1.2.3-59-g8ed1b


From f5058a27dac700952d5551bd7c5546f52b0468e4 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Thu, 12 Dec 2019 17:16:12 +0000
Subject: net: phylink: propagate phy_attach_direct() return code

of_phy_attach() hides the return value of phy_attach_direct(), forcing
us to return a "generic" ENODEV error code that is indistinguishable
from the lack-of-phy-property case.

Switch to using of_phy_find_device() to find the PHY device, and then
propagating any phy_attach_direct() error back to the caller.

Link: https://lore.kernel.org/lkml/20191210113829.GT25745@shell.armlinux.org.uk
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
 drivers/net/phy/phylink.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index 47614f975344..1e0e32c466ee 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -860,14 +860,17 @@ int phylink_of_phy_connect(struct phylink *pl, struct device_node *dn,
 		return 0;
 	}
 
-	phy_dev = of_phy_attach(pl->netdev, phy_node, flags,
-				pl->link_interface);
+	phy_dev = of_phy_find_device(phy_node);
 	/* We're done with the phy_node handle */
 	of_node_put(phy_node);
-
 	if (!phy_dev)
 		return -ENODEV;
 
+	ret = phy_attach_direct(pl->netdev, phy_dev, flags,
+				pl->link_interface);
+	if (ret)
+		return ret;
+
 	ret = phylink_bringup_phy(pl, phy_dev, pl->link_config.interface);
 	if (ret)
 		phy_detach(phy_dev);
-- 
cgit v1.2.3-59-g8ed1b


From ec2025095cf6acda3233a4301809596938b47da5 Mon Sep 17 00:00:00 2001
From: Paul Chaignon <paul.chaignon@orange.com>
Date: Fri, 13 Dec 2019 20:10:04 +0100
Subject: bpftool: Match several programs with same tag

When several BPF programs have the same tag, bpftool matches only the
first (in ID order).  This patch changes that behavior such that dump and
show commands return all matched programs.  Commands that require a single
program (e.g., pin and attach) will error out if given a tag that matches
several.  bpftool prog dump will also error out if file or visual are
given and several programs have the given tag.

In the case of the dump command, a program header is added before each
dump only if the tag matches several programs; this patch doesn't change
the output if a single program matches.  The output when several
programs match thus looks as follows.

$ ./bpftool prog dump xlated tag 6deef7357e7b4530
3: cgroup_skb  tag 6deef7357e7b4530  gpl
   0: (bf) r6 = r1
   [...]
   7: (95) exit

4: cgroup_skb  tag 6deef7357e7b4530  gpl
   0: (bf) r6 = r1
   [...]
   7: (95) exit

Signed-off-by: Paul Chaignon <paul.chaignon@orange.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/fb1fe943202659a69cd21dd5b907c205af1e1e22.1576263640.git.paul.chaignon@gmail.com
---
 tools/bpf/bpftool/Documentation/bpftool-prog.rst |  16 +-
 tools/bpf/bpftool/prog.c                         | 372 ++++++++++++++++-------
 2 files changed, 268 insertions(+), 120 deletions(-)

diff --git a/tools/bpf/bpftool/Documentation/bpftool-prog.rst b/tools/bpf/bpftool/Documentation/bpftool-prog.rst
index 7a374b3c851d..d377d0cb7923 100644
--- a/tools/bpf/bpftool/Documentation/bpftool-prog.rst
+++ b/tools/bpf/bpftool/Documentation/bpftool-prog.rst
@@ -53,8 +53,10 @@ DESCRIPTION
 ===========
 	**bpftool prog { show | list }** [*PROG*]
 		  Show information about loaded programs.  If *PROG* is
-		  specified show information only about given program, otherwise
-		  list all programs currently loaded on the system.
+		  specified show information only about given programs,
+		  otherwise list all programs currently loaded on the system.
+		  In case of **tag**, *PROG* may match several programs which
+		  will all be shown.
 
 		  Output will start with program ID followed by program type and
 		  zero or more named attributes (depending on kernel version).
@@ -68,11 +70,15 @@ DESCRIPTION
 		  performed via the **kernel.bpf_stats_enabled** sysctl knob.
 
 	**bpftool prog dump xlated** *PROG* [{ **file** *FILE* | **opcodes** | **visual** | **linum** }]
-		  Dump eBPF instructions of the program from the kernel. By
+		  Dump eBPF instructions of the programs from the kernel. By
 		  default, eBPF will be disassembled and printed to standard
 		  output in human-readable format. In this case, **opcodes**
 		  controls if raw opcodes should be printed as well.
 
+		  In case of **tag**, *PROG* may match several programs which
+		  will all be dumped.  However, if **file** or **visual** is
+		  specified, *PROG* must match a single program.
+
 		  If **file** is specified, the binary image will instead be
 		  written to *FILE*.
 
@@ -80,15 +86,17 @@ DESCRIPTION
 		  built instead, and eBPF instructions will be presented with
 		  CFG in DOT format, on standard output.
 
-		  If the prog has line_info available, the source line will
+		  If the programs have line_info available, the source line will
 		  be displayed by default.  If **linum** is specified,
 		  the filename, line number and line column will also be
 		  displayed on top of the source line.
 
 	**bpftool prog dump jited**  *PROG* [{ **file** *FILE* | **opcodes** | **linum** }]
 		  Dump jited image (host machine code) of the program.
+
 		  If *FILE* is specified image will be written to a file,
 		  otherwise it will be disassembled and printed to stdout.
+		  *PROG* must match a single program when **file** is specified.
 
 		  **opcodes** controls if raw opcodes will be printed.
 
diff --git a/tools/bpf/bpftool/prog.c b/tools/bpf/bpftool/prog.c
index 4535c863d2cd..37948c47aabd 100644
--- a/tools/bpf/bpftool/prog.c
+++ b/tools/bpf/bpftool/prog.c
@@ -25,6 +25,11 @@
 #include "main.h"
 #include "xlated_dumper.h"
 
+enum dump_mode {
+	DUMP_JITED,
+	DUMP_XLATED,
+};
+
 static const char * const attach_type_strings[] = {
 	[BPF_SK_SKB_STREAM_PARSER] = "stream_parser",
 	[BPF_SK_SKB_STREAM_VERDICT] = "stream_verdict",
@@ -77,11 +82,12 @@ static void print_boot_time(__u64 nsecs, char *buf, unsigned int size)
 		strftime(buf, size, "%FT%T%z", &load_tm);
 }
 
-static int prog_fd_by_tag(unsigned char *tag)
+static int prog_fd_by_tag(unsigned char *tag, int **fds)
 {
 	unsigned int id = 0;
+	int fd, nb_fds = 0;
+	void *tmp;
 	int err;
-	int fd;
 
 	while (true) {
 		struct bpf_prog_info info = {};
@@ -89,36 +95,53 @@ static int prog_fd_by_tag(unsigned char *tag)
 
 		err = bpf_prog_get_next_id(id, &id);
 		if (err) {
-			p_err("%s", strerror(errno));
-			return -1;
+			if (errno != ENOENT) {
+				p_err("%s", strerror(errno));
+				goto err_close_fds;
+			}
+			return nb_fds;
 		}
 
 		fd = bpf_prog_get_fd_by_id(id);
 		if (fd < 0) {
 			p_err("can't get prog by id (%u): %s",
 			      id, strerror(errno));
-			return -1;
+			goto err_close_fds;
 		}
 
 		err = bpf_obj_get_info_by_fd(fd, &info, &len);
 		if (err) {
 			p_err("can't get prog info (%u): %s",
 			      id, strerror(errno));
-			close(fd);
-			return -1;
+			goto err_close_fd;
 		}
 
-		if (!memcmp(tag, info.tag, BPF_TAG_SIZE))
-			return fd;
+		if (memcmp(tag, info.tag, BPF_TAG_SIZE)) {
+			close(fd);
+			continue;
+		}
 
-		close(fd);
+		if (nb_fds > 0) {
+			tmp = realloc(*fds, (nb_fds + 1) * sizeof(int));
+			if (!tmp) {
+				p_err("failed to realloc");
+				goto err_close_fd;
+			}
+			*fds = tmp;
+		}
+		(*fds)[nb_fds++] = fd;
 	}
+
+err_close_fd:
+	close(fd);
+err_close_fds:
+	while (--nb_fds >= 0)
+		close((*fds)[nb_fds]);
+	return -1;
 }
 
-int prog_parse_fd(int *argc, char ***argv)
+static int prog_parse_fds(int *argc, char ***argv, int **fds)
 {
-	int fd;
-
 	if (is_prefix(**argv, "id")) {
 		unsigned int id;
 		char *endptr;
@@ -132,10 +155,12 @@ int prog_parse_fd(int *argc, char ***argv)
 		}
 		NEXT_ARGP();
 
-		fd = bpf_prog_get_fd_by_id(id);
-		if (fd < 0)
+		(*fds)[0] = bpf_prog_get_fd_by_id(id);
+		if ((*fds)[0] < 0) {
 			p_err("get by id (%u): %s", id, strerror(errno));
-		return fd;
+			return -1;
+		}
+		return 1;
 	} else if (is_prefix(**argv, "tag")) {
 		unsigned char tag[BPF_TAG_SIZE];
 
@@ -149,7 +174,7 @@ int prog_parse_fd(int *argc, char ***argv)
 		}
 		NEXT_ARGP();
 
-		return prog_fd_by_tag(tag);
+		return prog_fd_by_tag(tag, fds);
 	} else if (is_prefix(**argv, "pinned")) {
 		char *path;
 
@@ -158,13 +183,43 @@ int prog_parse_fd(int *argc, char ***argv)
 		path = **argv;
 		NEXT_ARGP();
 
-		return open_obj_pinned_any(path, BPF_OBJ_PROG);
+		(*fds)[0] = open_obj_pinned_any(path, BPF_OBJ_PROG);
+		if ((*fds)[0] < 0)
+			return -1;
+		return 1;
 	}
 
 	p_err("expected 'id', 'tag' or 'pinned', got: '%s'?", **argv);
 	return -1;
 }
 
+int prog_parse_fd(int *argc, char ***argv)
+{
+	int *fds = NULL;
+	int nb_fds, fd;
+
+	fds = malloc(sizeof(int));
+	if (!fds) {
+		p_err("mem alloc failed");
+		return -1;
+	}
+	nb_fds = prog_parse_fds(argc, argv, &fds);
+	if (nb_fds != 1) {
+		if (nb_fds > 1) {
+			p_err("several programs match this handle");
+			while (nb_fds--)
+				close(fds[nb_fds]);
+		}
+		fd = -1;
+		goto exit_free;
+	}
+
+	fd = fds[0];
+exit_free:
+	free(fds);
+	return fd;
+}
+
 static void show_prog_maps(int fd, u32 num_maps)
 {
 	struct bpf_prog_info info = {};
@@ -194,11 +249,8 @@ static void show_prog_maps(int fd, u32 num_maps)
 	}
 }
 
-static void print_prog_json(struct bpf_prog_info *info, int fd)
+static void print_prog_header_json(struct bpf_prog_info *info)
 {
-	char *memlock;
-
-	jsonw_start_object(json_wtr);
 	jsonw_uint_field(json_wtr, "id", info->id);
 	if (info->type < ARRAY_SIZE(prog_type_name))
 		jsonw_string_field(json_wtr, "type",
@@ -219,7 +271,14 @@ static void print_prog_json(struct bpf_prog_info *info, int fd)
 		jsonw_uint_field(json_wtr, "run_time_ns", info->run_time_ns);
 		jsonw_uint_field(json_wtr, "run_cnt", info->run_cnt);
 	}
+}
+
+static void print_prog_json(struct bpf_prog_info *info, int fd)
+{
+	char *memlock;
 
+	jsonw_start_object(json_wtr);
+	print_prog_header_json(info);
 	print_dev_json(info->ifindex, info->netns_dev, info->netns_ino);
 
 	if (info->load_time) {
@@ -268,10 +327,8 @@ static void print_prog_json(struct bpf_prog_info *info, int fd)
 	jsonw_end_object(json_wtr);
 }
 
-static void print_prog_plain(struct bpf_prog_info *info, int fd)
+static void print_prog_header_plain(struct bpf_prog_info *info)
 {
-	char *memlock;
-
 	printf("%u: ", info->id);
 	if (info->type < ARRAY_SIZE(prog_type_name))
 		printf("%s  ", prog_type_name[info->type]);
@@ -289,6 +346,13 @@ static void print_prog_plain(struct bpf_prog_info *info, int fd)
 		printf(" run_time_ns %lld run_cnt %lld",
 		       info->run_time_ns, info->run_cnt);
 	printf("\n");
+}
+
+static void print_prog_plain(struct bpf_prog_info *info, int fd)
+{
+	char *memlock;
+
+	print_prog_header_plain(info);
 
 	if (info->load_time) {
 		char buf[32];
@@ -349,6 +413,40 @@ static int show_prog(int fd)
 	return 0;
 }
 
+static int do_show_subset(int argc, char **argv)
+{
+	int *fds = NULL;
+	int nb_fds, i;
+	int err = -1;
+
+	fds = malloc(sizeof(int));
+	if (!fds) {
+		p_err("mem alloc failed");
+		return -1;
+	}
+	nb_fds = prog_parse_fds(&argc, &argv, &fds);
+	if (nb_fds < 1)
+		goto exit_free;
+
+	if (json_output && nb_fds > 1)
+		jsonw_start_array(json_wtr);	/* root array */
+	for (i = 0; i < nb_fds; i++) {
+		err = show_prog(fds[i]);
+		if (err) {
+			for (; i < nb_fds; i++)
+				close(fds[i]);
+			break;
+		}
+		close(fds[i]);
+	}
+	if (json_output && nb_fds > 1)
+		jsonw_end_array(json_wtr);	/* root array */
+
+exit_free:
+	free(fds);
+	return err;
+}
+
 static int do_show(int argc, char **argv)
 {
 	__u32 id = 0;
@@ -358,15 +456,8 @@ static int do_show(int argc, char **argv)
 	if (show_pinned)
 		build_pinned_obj_table(&prog_table, BPF_OBJ_PROG);
 
-	if (argc == 2) {
-		fd = prog_parse_fd(&argc, &argv);
-		if (fd < 0)
-			return -1;
-
-		err = show_prog(fd);
-		close(fd);
-		return err;
-	}
+	if (argc == 2)
+		return do_show_subset(argc, argv);
 
 	if (argc)
 		return BAD_ARG();
@@ -408,101 +499,32 @@ static int do_show(int argc, char **argv)
 	return err;
 }
 
-static int do_dump(int argc, char **argv)
+static int
+prog_dump(struct bpf_prog_info *info, enum dump_mode mode,
+	  char *filepath, bool opcodes, bool visual, bool linum)
 {
-	struct bpf_prog_info_linear *info_linear;
 	struct bpf_prog_linfo *prog_linfo = NULL;
-	enum {DUMP_JITED, DUMP_XLATED} mode;
 	const char *disasm_opt = NULL;
-	struct bpf_prog_info *info;
 	struct dump_data dd = {};
 	void *func_info = NULL;
 	struct btf *btf = NULL;
-	char *filepath = NULL;
-	bool opcodes = false;
-	bool visual = false;
 	char func_sig[1024];
 	unsigned char *buf;
-	bool linum = false;
 	__u32 member_len;
-	__u64 arrays;
 	ssize_t n;
 	int fd;
 
-	if (is_prefix(*argv, "jited")) {
-		if (disasm_init())
-			return -1;
-		mode = DUMP_JITED;
-	} else if (is_prefix(*argv, "xlated")) {
-		mode = DUMP_XLATED;
-	} else {
-		p_err("expected 'xlated' or 'jited', got: %s", *argv);
-		return -1;
-	}
-	NEXT_ARG();
-
-	if (argc < 2)
-		usage();
-
-	fd = prog_parse_fd(&argc, &argv);
-	if (fd < 0)
-		return -1;
-
-	if (is_prefix(*argv, "file")) {
-		NEXT_ARG();
-		if (!argc) {
-			p_err("expected file path");
-			return -1;
-		}
-
-		filepath = *argv;
-		NEXT_ARG();
-	} else if (is_prefix(*argv, "opcodes")) {
-		opcodes = true;
-		NEXT_ARG();
-	} else if (is_prefix(*argv, "visual")) {
-		visual = true;
-		NEXT_ARG();
-	} else if (is_prefix(*argv, "linum")) {
-		linum = true;
-		NEXT_ARG();
-	}
-
-	if (argc) {
-		usage();
-		return -1;
-	}
-
-	if (mode == DUMP_JITED)
-		arrays = 1UL << BPF_PROG_INFO_JITED_INSNS;
-	else
-		arrays = 1UL << BPF_PROG_INFO_XLATED_INSNS;
-
-	arrays |= 1UL << BPF_PROG_INFO_JITED_KSYMS;
-	arrays |= 1UL << BPF_PROG_INFO_JITED_FUNC_LENS;
-	arrays |= 1UL << BPF_PROG_INFO_FUNC_INFO;
-	arrays |= 1UL << BPF_PROG_INFO_LINE_INFO;
-	arrays |= 1UL << BPF_PROG_INFO_JITED_LINE_INFO;
-
-	info_linear = bpf_program__get_prog_info_linear(fd, arrays);
-	close(fd);
-	if (IS_ERR_OR_NULL(info_linear)) {
-		p_err("can't get prog info: %s", strerror(errno));
-		return -1;
-	}
-
-	info = &info_linear->info;
 	if (mode == DUMP_JITED) {
 		if (info->jited_prog_len == 0) {
 			p_info("no instructions returned");
-			goto err_free;
+			return -1;
 		}
 		buf = (unsigned char *)(info->jited_prog_insns);
 		member_len = info->jited_prog_len;
 	} else {	/* DUMP_XLATED */
 		if (info->xlated_prog_len == 0) {
 			p_err("error retrieving insn dump: kernel.kptr_restrict set?");
-			goto err_free;
+			return -1;
 		}
 		buf = (unsigned char *)info->xlated_prog_insns;
 		member_len = info->xlated_prog_len;
@@ -510,7 +532,7 @@ static int do_dump(int argc, char **argv)
 
 	if (info->btf_id && btf__get_from_id(info->btf_id, &btf)) {
 		p_err("failed to get btf");
-		goto err_free;
+		return -1;
 	}
 
 	func_info = (void *)info->func_info;
@@ -526,7 +548,7 @@ static int do_dump(int argc, char **argv)
 		if (fd < 0) {
 			p_err("can't open file %s: %s", filepath,
 			      strerror(errno));
-			goto err_free;
+			return -1;
 		}
 
 		n = write(fd, buf, member_len);
@@ -534,7 +556,7 @@ static int do_dump(int argc, char **argv)
 		if (n != member_len) {
 			p_err("error writing output file: %s",
 			      n < 0 ? strerror(errno) : "short write");
-			goto err_free;
+			return -1;
 		}
 
 		if (json_output)
@@ -548,7 +570,7 @@ static int do_dump(int argc, char **argv)
 						     info->netns_ino,
 						     &disasm_opt);
 			if (!name)
-				goto err_free;
+				return -1;
 		}
 
 		if (info->nr_jited_func_lens && info->jited_func_lens) {
@@ -643,12 +665,130 @@ static int do_dump(int argc, char **argv)
 		kernel_syms_destroy(&dd);
 	}
 
-	free(info_linear);
 	return 0;
+}
 
-err_free:
-	free(info_linear);
-	return -1;
+static int do_dump(int argc, char **argv)
+{
+	struct bpf_prog_info_linear *info_linear;
+	char *filepath = NULL;
+	bool opcodes = false;
+	bool visual = false;
+	enum dump_mode mode;
+	bool linum = false;
+	int *fds = NULL;
+	int nb_fds, i = 0;
+	int err = -1;
+	__u64 arrays;
+
+	if (is_prefix(*argv, "jited")) {
+		if (disasm_init())
+			return -1;
+		mode = DUMP_JITED;
+	} else if (is_prefix(*argv, "xlated")) {
+		mode = DUMP_XLATED;
+	} else {
+		p_err("expected 'xlated' or 'jited', got: %s", *argv);
+		return -1;
+	}
+	NEXT_ARG();
+
+	if (argc < 2)
+		usage();
+
+	fds = malloc(sizeof(int));
+	if (!fds) {
+		p_err("mem alloc failed");
+		return -1;
+	}
+	nb_fds = prog_parse_fds(&argc, &argv, &fds);
+	if (nb_fds < 1)
+		goto exit_free;
+
+	if (is_prefix(*argv, "file")) {
+		NEXT_ARG();
+		if (!argc) {
+			p_err("expected file path");
+			goto exit_close;
+		}
+		if (nb_fds > 1) {
+			p_err("several programs matched");
+			goto exit_close;
+		}
+
+		filepath = *argv;
+		NEXT_ARG();
+	} else if (is_prefix(*argv, "opcodes")) {
+		opcodes = true;
+		NEXT_ARG();
+	} else if (is_prefix(*argv, "visual")) {
+		if (nb_fds > 1) {
+			p_err("several programs matched");
+			goto exit_close;
+		}
+
+		visual = true;
+		NEXT_ARG();
+	} else if (is_prefix(*argv, "linum")) {
+		linum = true;
+		NEXT_ARG();
+	}
+
+	if (argc) {
+		usage();
+		goto exit_close;
+	}
+
+	if (mode == DUMP_JITED)
+		arrays = 1UL << BPF_PROG_INFO_JITED_INSNS;
+	else
+		arrays = 1UL << BPF_PROG_INFO_XLATED_INSNS;
+
+	arrays |= 1UL << BPF_PROG_INFO_JITED_KSYMS;
+	arrays |= 1UL << BPF_PROG_INFO_JITED_FUNC_LENS;
+	arrays |= 1UL << BPF_PROG_INFO_FUNC_INFO;
+	arrays |= 1UL << BPF_PROG_INFO_LINE_INFO;
+	arrays |= 1UL << BPF_PROG_INFO_JITED_LINE_INFO;
+
+	if (json_output && nb_fds > 1)
+		jsonw_start_array(json_wtr);	/* root array */
+	for (i = 0; i < nb_fds; i++) {
+		info_linear = bpf_program__get_prog_info_linear(fds[i], arrays);
+		if (IS_ERR_OR_NULL(info_linear)) {
+			p_err("can't get prog info: %s", strerror(errno));
+			break;
+		}
+
+		if (json_output && nb_fds > 1) {
+			jsonw_start_object(json_wtr);	/* prog object */
+			print_prog_header_json(&info_linear->info);
+			jsonw_name(json_wtr, "insns");
+		} else if (nb_fds > 1) {
+			print_prog_header_plain(&info_linear->info);
+		}
+
+		err = prog_dump(&info_linear->info, mode, filepath, opcodes,
+				visual, linum);
+
+		if (json_output && nb_fds > 1)
+			jsonw_end_object(json_wtr);	/* prog object */
+		else if (i != nb_fds - 1 && nb_fds > 1)
+			printf("\n");
+
+		free(info_linear);
+		if (err)
+			break;
+		close(fds[i]);
+	}
+	if (json_output && nb_fds > 1)
+		jsonw_end_array(json_wtr);	/* root array */
+
+exit_close:
+	for (; i < nb_fds; i++)
+		close(fds[i]);
+exit_free:
+	free(fds);
+	return err;
 }
 
 static int do_pin(int argc, char **argv)
-- 
cgit v1.2.3-59-g8ed1b


From a7d22ca2a483d6c69c0791954447464297315ffa Mon Sep 17 00:00:00 2001
From: Paul Chaignon <paul.chaignon@orange.com>
Date: Fri, 13 Dec 2019 20:10:17 +0100
Subject: bpftool: Match programs by name

When working with frequently modified BPF programs, both the ID and the
tag may change.  bpftool currently doesn't provide a "stable" way to match
such programs.

This patch implements lookup by name for programs.  The show and dump
commands will return all programs with the given name, whereas other
commands will error out if several programs have the same name.

Signed-off-by: Paul Chaignon <paul.chaignon@orange.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com>
Link: https://lore.kernel.org/bpf/b5fc1a5dcfaeb5f16fc80295cdaa606dd2d91534.1576263640.git.paul.chaignon@gmail.com
---
 tools/bpf/bpftool/Documentation/bpftool-map.rst  |  2 +-
 tools/bpf/bpftool/Documentation/bpftool-prog.rst | 12 ++++++------
 tools/bpf/bpftool/bash-completion/bpftool        | 22 ++++++++++++++++------
 tools/bpf/bpftool/main.h                         |  2 +-
 tools/bpf/bpftool/prog.c                         | 22 ++++++++++++++++++----
 5 files changed, 42 insertions(+), 18 deletions(-)

diff --git a/tools/bpf/bpftool/Documentation/bpftool-map.rst b/tools/bpf/bpftool/Documentation/bpftool-map.rst
index 1c0f7146aab0..90d848b5b7d3 100644
--- a/tools/bpf/bpftool/Documentation/bpftool-map.rst
+++ b/tools/bpf/bpftool/Documentation/bpftool-map.rst
@@ -41,7 +41,7 @@ MAP COMMANDS
 |
 |	*MAP* := { **id** *MAP_ID* | **pinned** *FILE* }
 |	*DATA* := { [**hex**] *BYTES* }
-|	*PROG* := { **id** *PROG_ID* | **pinned** *FILE* | **tag** *PROG_TAG* }
+|	*PROG* := { **id** *PROG_ID* | **pinned** *FILE* | **tag** *PROG_TAG* | **name** *PROG_NAME* }
 |	*VALUE* := { *DATA* | *MAP* | *PROG* }
 |	*UPDATE_FLAGS* := { **any** | **exist** | **noexist** }
 |	*TYPE* := { **hash** | **array** | **prog_array** | **perf_event_array** | **percpu_hash**
diff --git a/tools/bpf/bpftool/Documentation/bpftool-prog.rst b/tools/bpf/bpftool/Documentation/bpftool-prog.rst
index d377d0cb7923..64ddf8a4c518 100644
--- a/tools/bpf/bpftool/Documentation/bpftool-prog.rst
+++ b/tools/bpf/bpftool/Documentation/bpftool-prog.rst
@@ -33,7 +33,7 @@ PROG COMMANDS
 |	**bpftool** **prog help**
 |
 |	*MAP* := { **id** *MAP_ID* | **pinned** *FILE* }
-|	*PROG* := { **id** *PROG_ID* | **pinned** *FILE* | **tag** *PROG_TAG* }
+|	*PROG* := { **id** *PROG_ID* | **pinned** *FILE* | **tag** *PROG_TAG* | **name** *PROG_NAME* }
 |	*TYPE* := {
 |		**socket** | **kprobe** | **kretprobe** | **classifier** | **action** |
 |		**tracepoint** | **raw_tracepoint** | **xdp** | **perf_event** | **cgroup/skb** |
@@ -55,8 +55,8 @@ DESCRIPTION
 		  Show information about loaded programs.  If *PROG* is
 		  specified show information only about given programs,
 		  otherwise list all programs currently loaded on the system.
-		  In case of **tag**, *PROG* may match several programs which
-		  will all be shown.
+		  In case of **tag** or **name**, *PROG* may match several
+		  programs which will all be shown.
 
 		  Output will start with program ID followed by program type and
 		  zero or more named attributes (depending on kernel version).
@@ -75,9 +75,9 @@ DESCRIPTION
 		  output in human-readable format. In this case, **opcodes**
 		  controls if raw opcodes should be printed as well.
 
-		  In case of **tag**, *PROG* may match several programs which
-		  will all be dumped.  However, if **file** or **visual** is
-		  specified, *PROG* must match a single program.
+		  In case of **tag** or **name**, *PROG* may match several
+		  programs which will all be dumped.  However, if **file** or
+		  **visual** is specified, *PROG* must match a single program.
 
 		  If **file** is specified, the binary image will instead be
 		  written to *FILE*.
diff --git a/tools/bpf/bpftool/bash-completion/bpftool b/tools/bpf/bpftool/bash-completion/bpftool
index 70493a6da206..05b5be4a6ef9 100644
--- a/tools/bpf/bpftool/bash-completion/bpftool
+++ b/tools/bpf/bpftool/bash-completion/bpftool
@@ -71,6 +71,12 @@ _bpftool_get_prog_tags()
         command sed -n 's/.*"tag": "\(.*\)",$/\1/p' )" -- "$cur" ) )
 }
 
+_bpftool_get_prog_names()
+{
+    COMPREPLY+=( $( compgen -W "$( bpftool -jp prog 2>&1 | \
+        command sed -n 's/.*"name": "\(.*\)",$/\1/p' )" -- "$cur" ) )
+}
+
 _bpftool_get_btf_ids()
 {
     COMPREPLY+=( $( compgen -W "$( bpftool -jp btf 2>&1 | \
@@ -201,6 +207,10 @@ _bpftool()
             _bpftool_get_prog_tags
             return 0
             ;;
+        name)
+            _bpftool_get_prog_names
+            return 0
+            ;;
         dev)
             _sysfs_get_netdevs
             return 0
@@ -263,7 +273,7 @@ _bpftool()
                     ;;
             esac
 
-            local PROG_TYPE='id pinned tag'
+            local PROG_TYPE='id pinned tag name'
             local MAP_TYPE='id pinned'
             case $command in
                 show|list)
@@ -559,7 +569,7 @@ _bpftool()
                                     return 0
                                     ;;
                                 prog_array)
-                                    local PROG_TYPE='id pinned tag'
+                                    local PROG_TYPE='id pinned tag name'
                                     COMPREPLY+=( $( compgen -W "$PROG_TYPE" \
                                         -- "$cur" ) )
                                     return 0
@@ -644,7 +654,7 @@ _bpftool()
             esac
             ;;
         btf)
-            local PROG_TYPE='id pinned tag'
+            local PROG_TYPE='id pinned tag name'
             local MAP_TYPE='id pinned'
             case $command in
                 dump)
@@ -735,7 +745,7 @@ _bpftool()
                         connect6 sendmsg4 sendmsg6 recvmsg4 recvmsg6 sysctl \
                         getsockopt setsockopt'
                     local ATTACH_FLAGS='multi override'
-                    local PROG_TYPE='id pinned tag'
+                    local PROG_TYPE='id pinned tag name'
                     case $prev in
                         $command)
                             _filedir
@@ -760,7 +770,7 @@ _bpftool()
                             elif [[ "$command" == "attach" ]]; then
                                 # We have an attach type on the command line,
                                 # but it is not the previous word, or
-                                # "id|pinned|tag" (we already checked for
+                                # "id|pinned|tag|name" (we already checked for
                                 # that). This should only leave the case when
                                 # we need attach flags for "attach" commamnd.
                                 _bpftool_one_of_list "$ATTACH_FLAGS"
@@ -786,7 +796,7 @@ _bpftool()
             esac
             ;;
         net)
-            local PROG_TYPE='id pinned tag'
+            local PROG_TYPE='id pinned tag name'
             local ATTACH_TYPES='xdp xdpgeneric xdpdrv xdpoffload'
             case $command in
                 show|list)
diff --git a/tools/bpf/bpftool/main.h b/tools/bpf/bpftool/main.h
index 2899095f8254..a7ead7bb9447 100644
--- a/tools/bpf/bpftool/main.h
+++ b/tools/bpf/bpftool/main.h
@@ -42,7 +42,7 @@
 #define BPF_TAG_FMT	"%02hhx%02hhx%02hhx%02hhx%02hhx%02hhx%02hhx%02hhx"
 
 #define HELP_SPEC_PROGRAM						\
-	"PROG := { id PROG_ID | pinned FILE | tag PROG_TAG }"
+	"PROG := { id PROG_ID | pinned FILE | tag PROG_TAG | name PROG_NAME }"
 #define HELP_SPEC_OPTIONS						\
 	"OPTIONS := { {-j|--json} [{-p|--pretty}] | {-f|--bpffs} |\n"	\
 	"\t            {-m|--mapcompat} | {-n|--nomount} }"
diff --git a/tools/bpf/bpftool/prog.c b/tools/bpf/bpftool/prog.c
index 37948c47aabd..2221bae037f1 100644
--- a/tools/bpf/bpftool/prog.c
+++ b/tools/bpf/bpftool/prog.c
@@ -82,7 +82,7 @@ static void print_boot_time(__u64 nsecs, char *buf, unsigned int size)
 		strftime(buf, size, "%FT%T%z", &load_tm);
 }
 
-static int prog_fd_by_tag(unsigned char *tag, int **fds)
+static int prog_fd_by_nametag(void *nametag, int **fds, bool tag)
 {
 	unsigned int id = 0;
 	int fd, nb_fds = 0;
@@ -116,7 +116,8 @@ static int prog_fd_by_tag(unsigned char *tag, int **fds)
 			goto err_close_fd;
 		}
 
-		if (memcmp(tag, info.tag, BPF_TAG_SIZE)) {
+		if ((tag && memcmp(nametag, info.tag, BPF_TAG_SIZE)) ||
+		    (!tag && strncmp(nametag, info.name, BPF_OBJ_NAME_LEN))) {
 			close(fd);
 			continue;
 		}
@@ -174,7 +175,20 @@ static int prog_parse_fds(int *argc, char ***argv, int **fds)
 		}
 		NEXT_ARGP();
 
-		return prog_fd_by_tag(tag, fds);
+		return prog_fd_by_nametag(tag, fds, true);
+	} else if (is_prefix(**argv, "name")) {
+		char *name;
+
+		NEXT_ARGP();
+
+		name = **argv;
+		if (strlen(name) > BPF_OBJ_NAME_LEN - 1) {
+			p_err("can't parse name");
+			return -1;
+		}
+		NEXT_ARGP();
+
+		return prog_fd_by_nametag(name, fds, false);
 	} else if (is_prefix(**argv, "pinned")) {
 		char *path;
 
@@ -189,7 +203,7 @@ static int prog_parse_fds(int *argc, char ***argv, int **fds)
 		return 1;
 	}
 
-	p_err("expected 'id', 'tag' or 'pinned', got: '%s'?", **argv);
+	p_err("expected 'id', 'tag', 'name' or 'pinned', got: '%s'?", **argv);
 	return -1;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 99f9863a0c45f4e87cb99593015090fdc9f44398 Mon Sep 17 00:00:00 2001
From: Paul Chaignon <paul.chaignon@orange.com>
Date: Fri, 13 Dec 2019 20:10:37 +0100
Subject: bpftool: Match maps by name

This patch implements lookup by name for maps and changes the behavior of
lookups by tag to be consistent with prog subcommands.  Similarly to
program subcommands, the show and dump commands will return all maps with
the given name (or tag), whereas other commands will error out if several
maps have the same name (resp. tag).

When a map has BTF info, it is dumped in JSON with available BTF info.
This patch requires that all matched maps have BTF info before switching
the output format to JSON.

Signed-off-by: Paul Chaignon <paul.chaignon@orange.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/8de1c9f273860b3ea1680502928f4da2336b853e.1576263640.git.paul.chaignon@gmail.com
---
 tools/bpf/bpftool/Documentation/bpftool-map.rst |  10 +-
 tools/bpf/bpftool/bash-completion/bpftool       | 131 +++++++-
 tools/bpf/bpftool/main.h                        |   2 +-
 tools/bpf/bpftool/map.c                         | 384 +++++++++++++++++++-----
 4 files changed, 439 insertions(+), 88 deletions(-)

diff --git a/tools/bpf/bpftool/Documentation/bpftool-map.rst b/tools/bpf/bpftool/Documentation/bpftool-map.rst
index 90d848b5b7d3..cdeae8ae90ba 100644
--- a/tools/bpf/bpftool/Documentation/bpftool-map.rst
+++ b/tools/bpf/bpftool/Documentation/bpftool-map.rst
@@ -39,7 +39,7 @@ MAP COMMANDS
 |	**bpftool** **map freeze**     *MAP*
 |	**bpftool** **map help**
 |
-|	*MAP* := { **id** *MAP_ID* | **pinned** *FILE* }
+|	*MAP* := { **id** *MAP_ID* | **pinned** *FILE* | **name** *MAP_NAME* }
 |	*DATA* := { [**hex**] *BYTES* }
 |	*PROG* := { **id** *PROG_ID* | **pinned** *FILE* | **tag** *PROG_TAG* | **name** *PROG_NAME* }
 |	*VALUE* := { *DATA* | *MAP* | *PROG* }
@@ -55,8 +55,9 @@ DESCRIPTION
 ===========
 	**bpftool map { show | list }**   [*MAP*]
 		  Show information about loaded maps.  If *MAP* is specified
-		  show information only about given map, otherwise list all
-		  maps currently loaded on the system.
+		  show information only about given maps, otherwise list all
+		  maps currently loaded on the system.  In case of **name**,
+		  *MAP* may match several maps which will all be shown.
 
 		  Output will start with map ID followed by map type and
 		  zero or more named attributes (depending on kernel version).
@@ -66,7 +67,8 @@ DESCRIPTION
 		  as *FILE*.
 
 	**bpftool map dump**    *MAP*
-		  Dump all entries in a given *MAP*.
+		  Dump all entries in a given *MAP*.  In case of **name**,
+		  *MAP* may match several maps which will all be dumped.
 
 	**bpftool map update**  *MAP* [**key** *DATA*] [**value** *VALUE*] [*UPDATE_FLAGS*]
 		  Update map entry for a given *KEY*.
diff --git a/tools/bpf/bpftool/bash-completion/bpftool b/tools/bpf/bpftool/bash-completion/bpftool
index 05b5be4a6ef9..21c676a1eeb1 100644
--- a/tools/bpf/bpftool/bash-completion/bpftool
+++ b/tools/bpf/bpftool/bash-completion/bpftool
@@ -59,6 +59,21 @@ _bpftool_get_map_ids_for_type()
         command sed -n 's/.*"id": \(.*\),$/\1/p' )" -- "$cur" ) )
 }
 
+_bpftool_get_map_names()
+{
+    COMPREPLY+=( $( compgen -W "$( bpftool -jp map  2>&1 | \
+        command sed -n 's/.*"name": \(.*\),$/\1/p' )" -- "$cur" ) )
+}
+
+# Takes map type and adds matching map names to the list of suggestions.
+_bpftool_get_map_names_for_type()
+{
+    local type="$1"
+    COMPREPLY+=( $( compgen -W "$( bpftool -jp map  2>&1 | \
+        command grep -C2 "$type" | \
+        command sed -n 's/.*"name": \(.*\),$/\1/p' )" -- "$cur" ) )
+}
+
 _bpftool_get_prog_ids()
 {
     COMPREPLY+=( $( compgen -W "$( bpftool -jp prog 2>&1 | \
@@ -186,6 +201,52 @@ _bpftool_map_update_get_id()
     esac
 }
 
+_bpftool_map_update_get_name()
+{
+    local command="$1"
+
+    # Is it the map to update, or a map to insert into the map to update?
+    # Search for "value" keyword.
+    local idx value
+    for (( idx=7; idx < ${#words[@]}-1; idx++ )); do
+        if [[ ${words[idx]} == "value" ]]; then
+            value=1
+            break
+        fi
+    done
+    if [[ $value -eq 0 ]]; then
+        case "$command" in
+            push)
+                _bpftool_get_map_names_for_type stack
+                ;;
+            enqueue)
+                _bpftool_get_map_names_for_type queue
+                ;;
+            *)
+                _bpftool_get_map_names
+                ;;
+        esac
+        return 0
+    fi
+
+    # Name to complete is for a value. It can be either prog name or map name. This
+    # depends on the type of the map to update.
+    local type=$(_bpftool_map_guess_map_type)
+    case $type in
+        array_of_maps|hash_of_maps)
+            _bpftool_get_map_names
+            return 0
+            ;;
+        prog_array)
+            _bpftool_get_prog_names
+            return 0
+            ;;
+        *)
+            return 0
+            ;;
+    esac
+}
+
 _bpftool()
 {
     local cur prev words objword
@@ -207,10 +268,6 @@ _bpftool()
             _bpftool_get_prog_tags
             return 0
             ;;
-        name)
-            _bpftool_get_prog_names
-            return 0
-            ;;
         dev)
             _sysfs_get_netdevs
             return 0
@@ -261,7 +318,8 @@ _bpftool()
     # Completion depends on object and command in use
     case $object in
         prog)
-            # Complete id, only for subcommands that use prog (but no map) ids
+            # Complete id and name, only for subcommands that use prog (but no
+            # map) ids/names.
             case $command in
                 show|list|dump|pin)
                     case $prev in
@@ -269,12 +327,16 @@ _bpftool()
                             _bpftool_get_prog_ids
                             return 0
                             ;;
+                        name)
+                            _bpftool_get_prog_names
+                            return 0
+                            ;;
                     esac
                     ;;
             esac
 
             local PROG_TYPE='id pinned tag name'
-            local MAP_TYPE='id pinned'
+            local MAP_TYPE='id pinned name'
             case $command in
                 show|list)
                     [[ $prev != "$command" ]] && return 0
@@ -325,6 +387,9 @@ _bpftool()
                                 id)
                                     _bpftool_get_prog_ids
                                     ;;
+                                name)
+                                    _bpftool_get_map_names
+                                    ;;
                                 pinned)
                                     _filedir
                                     ;;
@@ -345,6 +410,9 @@ _bpftool()
                                 id)
                                     _bpftool_get_map_ids
                                     ;;
+                                name)
+                                    _bpftool_get_map_names
+                                    ;;
                                 pinned)
                                     _filedir
                                     ;;
@@ -409,6 +477,10 @@ _bpftool()
                             _bpftool_get_map_ids
                             return 0
                             ;;
+                        name)
+                            _bpftool_get_map_names
+                            return 0
+                            ;;
                         pinned|pinmaps)
                             _filedir
                             return 0
@@ -457,7 +529,7 @@ _bpftool()
             esac
             ;;
         map)
-            local MAP_TYPE='id pinned'
+            local MAP_TYPE='id pinned name'
             case $command in
                 show|list|dump|peek|pop|dequeue|freeze)
                     case $prev in
@@ -483,6 +555,24 @@ _bpftool()
                             esac
                             return 0
                             ;;
+                        name)
+                            case "$command" in
+                                peek)
+                                    _bpftool_get_map_names_for_type stack
+                                    _bpftool_get_map_names_for_type queue
+                                    ;;
+                                pop)
+                                    _bpftool_get_map_names_for_type stack
+                                    ;;
+                                dequeue)
+                                    _bpftool_get_map_names_for_type queue
+                                    ;;
+                                *)
+                                    _bpftool_get_map_names
+                                    ;;
+                            esac
+                            return 0
+                            ;;
                         *)
                             return 0
                             ;;
@@ -530,6 +620,10 @@ _bpftool()
                             _bpftool_get_map_ids
                             return 0
                             ;;
+                        name)
+                            _bpftool_get_map_names
+                            return 0
+                            ;;
                         key)
                             COMPREPLY+=( $( compgen -W 'hex' -- "$cur" ) )
                             ;;
@@ -555,6 +649,10 @@ _bpftool()
                             _bpftool_map_update_get_id $command
                             return 0
                             ;;
+                        name)
+                            _bpftool_map_update_get_name $command
+                            return 0
+                            ;;
                         key)
                             COMPREPLY+=( $( compgen -W 'hex' -- "$cur" ) )
                             ;;
@@ -563,7 +661,7 @@ _bpftool()
                             # map, depending on the type of the map to update.
                             case "$(_bpftool_map_guess_map_type)" in
                                 array_of_maps|hash_of_maps)
-                                    local MAP_TYPE='id pinned'
+                                    local MAP_TYPE='id pinned name'
                                     COMPREPLY+=( $( compgen -W "$MAP_TYPE" \
                                         -- "$cur" ) )
                                     return 0
@@ -631,6 +729,10 @@ _bpftool()
                             _bpftool_get_map_ids_for_type perf_event_array
                             return 0
                             ;;
+                        name)
+                            _bpftool_get_map_names_for_type perf_event_array
+                            return 0
+                            ;;
                         cpu)
                             return 0
                             ;;
@@ -655,7 +757,7 @@ _bpftool()
             ;;
         btf)
             local PROG_TYPE='id pinned tag name'
-            local MAP_TYPE='id pinned'
+            local MAP_TYPE='id pinned name'
             case $command in
                 dump)
                     case $prev in
@@ -686,6 +788,17 @@ _bpftool()
                             esac
                             return 0
                             ;;
+                        name)
+                            case $pprev in
+                                prog)
+                                    _bpftool_get_prog_names
+                                    ;;
+                                map)
+                                    _bpftool_get_map_names
+                                    ;;
+                            esac
+                            return 0
+                            ;;
                         format)
                             COMPREPLY=( $( compgen -W "c raw" -- "$cur" ) )
                             ;;
diff --git a/tools/bpf/bpftool/main.h b/tools/bpf/bpftool/main.h
index a7ead7bb9447..81890f4c8cca 100644
--- a/tools/bpf/bpftool/main.h
+++ b/tools/bpf/bpftool/main.h
@@ -47,7 +47,7 @@
 	"OPTIONS := { {-j|--json} [{-p|--pretty}] | {-f|--bpffs} |\n"	\
 	"\t            {-m|--mapcompat} | {-n|--nomount} }"
 #define HELP_SPEC_MAP							\
-	"MAP := { id MAP_ID | pinned FILE }"
+	"MAP := { id MAP_ID | pinned FILE | name MAP_NAME }"
 
 static const char * const prog_type_name[] = {
 	[BPF_PROG_TYPE_UNSPEC]			= "unspec",
diff --git a/tools/bpf/bpftool/map.c b/tools/bpf/bpftool/map.c
index de61d73b9030..c01f76fa6876 100644
--- a/tools/bpf/bpftool/map.c
+++ b/tools/bpf/bpftool/map.c
@@ -91,10 +91,66 @@ static void *alloc_value(struct bpf_map_info *info)
 		return malloc(info->value_size);
 }
 
-int map_parse_fd(int *argc, char ***argv)
+static int map_fd_by_name(char *name, int **fds)
 {
-	int fd;
+	unsigned int id = 0;
+	int fd, nb_fds = 0;
+	void *tmp;
+	int err;
+
+	while (true) {
+		struct bpf_map_info info = {};
+		__u32 len = sizeof(info);
+
+		err = bpf_map_get_next_id(id, &id);
+		if (err) {
+			if (errno != ENOENT) {
+				p_err("%s", strerror(errno));
+				goto err_close_fds;
+			}
+			return nb_fds;
+		}
+
+		fd = bpf_map_get_fd_by_id(id);
+		if (fd < 0) {
+			p_err("can't get map by id (%u): %s",
+			      id, strerror(errno));
+			goto err_close_fds;
+		}
+
+		err = bpf_obj_get_info_by_fd(fd, &info, &len);
+		if (err) {
+			p_err("can't get map info (%u): %s",
+			      id, strerror(errno));
+			goto err_close_fd;
+		}
+
+		if (strncmp(name, info.name, BPF_OBJ_NAME_LEN)) {
+			close(fd);
+			continue;
+		}
+
+		if (nb_fds > 0) {
+			tmp = realloc(*fds, (nb_fds + 1) * sizeof(int));
+			if (!tmp) {
+				p_err("failed to realloc");
+				goto err_close_fd;
+			}
+			*fds = tmp;
+		}
+		(*fds)[nb_fds++] = fd;
+	}
+
+err_close_fd:
+	close(fd);
+err_close_fds:
+	while (--nb_fds >= 0)
+		close((*fds)[nb_fds]);
+	return -1;
+}
 
+static int map_parse_fds(int *argc, char ***argv, int **fds)
+{
 	if (is_prefix(**argv, "id")) {
 		unsigned int id;
 		char *endptr;
@@ -108,10 +164,25 @@ int map_parse_fd(int *argc, char ***argv)
 		}
 		NEXT_ARGP();
 
-		fd = bpf_map_get_fd_by_id(id);
-		if (fd < 0)
+		(*fds)[0] = bpf_map_get_fd_by_id(id);
+		if ((*fds)[0] < 0) {
 			p_err("get map by id (%u): %s", id, strerror(errno));
-		return fd;
+			return -1;
+		}
+		return 1;
+	} else if (is_prefix(**argv, "name")) {
+		char *name;
+
+		NEXT_ARGP();
+
+		name = **argv;
+		if (strlen(name) > BPF_OBJ_NAME_LEN - 1) {
+			p_err("can't parse name");
+			return -1;
+		}
+		NEXT_ARGP();
+
+		return map_fd_by_name(name, fds);
 	} else if (is_prefix(**argv, "pinned")) {
 		char *path;
 
@@ -120,13 +191,43 @@ int map_parse_fd(int *argc, char ***argv)
 		path = **argv;
 		NEXT_ARGP();
 
-		return open_obj_pinned_any(path, BPF_OBJ_MAP);
+		(*fds)[0] = open_obj_pinned_any(path, BPF_OBJ_MAP);
+		if ((*fds)[0] < 0)
+			return -1;
+		return 1;
 	}
 
-	p_err("expected 'id' or 'pinned', got: '%s'?", **argv);
+	p_err("expected 'id', 'name' or 'pinned', got: '%s'?", **argv);
 	return -1;
 }
 
+int map_parse_fd(int *argc, char ***argv)
+{
+	int *fds = NULL;
+	int nb_fds, fd;
+
+	fds = malloc(sizeof(int));
+	if (!fds) {
+		p_err("mem alloc failed");
+		return -1;
+	}
+	nb_fds = map_parse_fds(argc, argv, &fds);
+	if (nb_fds != 1) {
+		if (nb_fds > 1) {
+			p_err("several maps match this handle");
+			while (nb_fds--)
+				close(fds[nb_fds]);
+		}
+		fd = -1;
+		goto exit_free;
+	}
+
+	fd = fds[0];
+exit_free:
+	free(fds);
+	return fd;
+}
+
 int map_parse_fd_and_info(int *argc, char ***argv, void *info, __u32 *info_len)
 {
 	int err;
@@ -479,6 +580,21 @@ static int parse_elem(char **argv, struct bpf_map_info *info,
 	return -1;
 }
 
+static void show_map_header_json(struct bpf_map_info *info, json_writer_t *wtr)
+{
+	jsonw_uint_field(wtr, "id", info->id);
+	if (info->type < ARRAY_SIZE(map_type_name))
+		jsonw_string_field(wtr, "type", map_type_name[info->type]);
+	else
+		jsonw_uint_field(wtr, "type", info->type);
+
+	if (*info->name)
+		jsonw_string_field(wtr, "name", info->name);
+
+	jsonw_name(wtr, "flags");
+	jsonw_printf(wtr, "%d", info->map_flags);
+}
+
 static int show_map_close_json(int fd, struct bpf_map_info *info)
 {
 	char *memlock, *frozen_str;
@@ -489,18 +605,7 @@ static int show_map_close_json(int fd, struct bpf_map_info *info)
 
 	jsonw_start_object(json_wtr);
 
-	jsonw_uint_field(json_wtr, "id", info->id);
-	if (info->type < ARRAY_SIZE(map_type_name))
-		jsonw_string_field(json_wtr, "type",
-				   map_type_name[info->type]);
-	else
-		jsonw_uint_field(json_wtr, "type", info->type);
-
-	if (*info->name)
-		jsonw_string_field(json_wtr, "name", info->name);
-
-	jsonw_name(json_wtr, "flags");
-	jsonw_printf(json_wtr, "%d", info->map_flags);
+	show_map_header_json(info, json_wtr);
 
 	print_dev_json(info->ifindex, info->netns_dev, info->netns_ino);
 
@@ -561,14 +666,8 @@ static int show_map_close_json(int fd, struct bpf_map_info *info)
 	return 0;
 }
 
-static int show_map_close_plain(int fd, struct bpf_map_info *info)
+static void show_map_header_plain(struct bpf_map_info *info)
 {
-	char *memlock, *frozen_str;
-	int frozen = 0;
-
-	memlock = get_fdinfo(fd, "memlock");
-	frozen_str = get_fdinfo(fd, "frozen");
-
 	printf("%u: ", info->id);
 	if (info->type < ARRAY_SIZE(map_type_name))
 		printf("%s  ", map_type_name[info->type]);
@@ -581,6 +680,17 @@ static int show_map_close_plain(int fd, struct bpf_map_info *info)
 	printf("flags 0x%x", info->map_flags);
 	print_dev_plain(info->ifindex, info->netns_dev, info->netns_ino);
 	printf("\n");
+}
+
+static int show_map_close_plain(int fd, struct bpf_map_info *info)
+{
+	char *memlock, *frozen_str;
+	int frozen = 0;
+
+	memlock = get_fdinfo(fd, "memlock");
+	frozen_str = get_fdinfo(fd, "frozen");
+
+	show_map_header_plain(info);
 	printf("\tkey %uB  value %uB  max_entries %u",
 	       info->key_size, info->value_size, info->max_entries);
 
@@ -642,6 +752,50 @@ static int show_map_close_plain(int fd, struct bpf_map_info *info)
 	return 0;
 }
 
+static int do_show_subset(int argc, char **argv)
+{
+	struct bpf_map_info info = {};
+	__u32 len = sizeof(info);
+	int *fds = NULL;
+	int nb_fds, i;
+	int err = -1;
+
+	fds = malloc(sizeof(int));
+	if (!fds) {
+		p_err("mem alloc failed");
+		return -1;
+	}
+	nb_fds = map_parse_fds(&argc, &argv, &fds);
+	if (nb_fds < 1)
+		goto exit_free;
+
+	if (json_output && nb_fds > 1)
+		jsonw_start_array(json_wtr);	/* root array */
+	for (i = 0; i < nb_fds; i++) {
+		err = bpf_obj_get_info_by_fd(fds[i], &info, &len);
+		if (err) {
+			p_err("can't get map info: %s",
+			      strerror(errno));
+			for (; i < nb_fds; i++)
+				close(fds[i]);
+			break;
+		}
+
+		if (json_output)
+			show_map_close_json(fds[i], &info);
+		else
+			show_map_close_plain(fds[i], &info);
+
+		close(fds[i]);
+	}
+	if (json_output && nb_fds > 1)
+		jsonw_end_array(json_wtr);	/* root array */
+
+exit_free:
+	free(fds);
+	return err;
+}
+
 static int do_show(int argc, char **argv)
 {
 	struct bpf_map_info info = {};
@@ -653,16 +807,8 @@ static int do_show(int argc, char **argv)
 	if (show_pinned)
 		build_pinned_obj_table(&map_table, BPF_OBJ_MAP);
 
-	if (argc == 2) {
-		fd = map_parse_fd_and_info(&argc, &argv, &info, &len);
-		if (fd < 0)
-			return -1;
-
-		if (json_output)
-			return show_map_close_json(fd, &info);
-		else
-			return show_map_close_plain(fd, &info);
-	}
+	if (argc == 2)
+		return do_show_subset(argc, argv);
 
 	if (argc)
 		return BAD_ARG();
@@ -765,26 +911,49 @@ static int dump_map_elem(int fd, void *key, void *value,
 	return 0;
 }
 
-static int do_dump(int argc, char **argv)
+static int maps_have_btf(int *fds, int nb_fds)
 {
 	struct bpf_map_info info = {};
-	void *key, *value, *prev_key;
-	unsigned int num_elems = 0;
 	__u32 len = sizeof(info);
-	json_writer_t *btf_wtr;
 	struct btf *btf = NULL;
-	int err;
-	int fd;
+	int err, i;
 
-	if (argc != 2)
-		usage();
+	for (i = 0; i < nb_fds; i++) {
+		err = bpf_obj_get_info_by_fd(fds[i], &info, &len);
+		if (err) {
+			p_err("can't get map info: %s", strerror(errno));
+			goto err_close;
+		}
 
-	fd = map_parse_fd_and_info(&argc, &argv, &info, &len);
-	if (fd < 0)
-		return -1;
+		err = btf__get_from_id(info.btf_id, &btf);
+		if (err) {
+			p_err("failed to get btf");
+			goto err_close;
+		}
 
-	key = malloc(info.key_size);
-	value = alloc_value(&info);
+		if (!btf)
+			return 0;
+	}
+
+	return 1;
+
+err_close:
+	for (; i < nb_fds; i++)
+		close(fds[i]);
+	return -1;
+}
+
+static int
+map_dump(int fd, struct bpf_map_info *info, json_writer_t *wtr,
+	 bool enable_btf, bool show_header)
+{
+	void *key, *value, *prev_key;
+	unsigned int num_elems = 0;
+	struct btf *btf = NULL;
+	int err;
+
+	key = malloc(info->key_size);
+	value = alloc_value(info);
 	if (!key || !value) {
 		p_err("mem alloc failed");
 		err = -1;
@@ -793,30 +962,32 @@ static int do_dump(int argc, char **argv)
 
 	prev_key = NULL;
 
-	err = btf__get_from_id(info.btf_id, &btf);
-	if (err) {
-		p_err("failed to get btf");
-		goto exit_free;
+	if (enable_btf) {
+		err = btf__get_from_id(info->btf_id, &btf);
+		if (err || !btf) {
+			/* enable_btf is true only if we've already checked
+			 * that all maps have BTF information.
+			 */
+			p_err("failed to get btf");
+			goto exit_free;
+		}
 	}
 
-	if (json_output)
-		jsonw_start_array(json_wtr);
-	else
-		if (btf) {
-			btf_wtr = get_btf_writer();
-			if (!btf_wtr) {
-				p_info("failed to create json writer for btf. falling back to plain output");
-				btf__free(btf);
-				btf = NULL;
-			} else {
-				jsonw_start_array(btf_wtr);
-			}
+	if (wtr) {
+		if (show_header) {
+			jsonw_start_object(wtr);	/* map object */
+			show_map_header_json(info, wtr);
+			jsonw_name(wtr, "elements");
 		}
+		jsonw_start_array(wtr);		/* elements */
+	} else if (show_header) {
+		show_map_header_plain(info);
+	}
 
-	if (info.type == BPF_MAP_TYPE_REUSEPORT_SOCKARRAY &&
-	    info.value_size != 8)
+	if (info->type == BPF_MAP_TYPE_REUSEPORT_SOCKARRAY &&
+	    info->value_size != 8)
 		p_info("Warning: cannot read values from %s map with value_size != 8",
-		       map_type_name[info.type]);
+		       map_type_name[info->type]);
 	while (true) {
 		err = bpf_map_get_next_key(fd, prev_key, key);
 		if (err) {
@@ -824,15 +995,14 @@ static int do_dump(int argc, char **argv)
 				err = 0;
 			break;
 		}
-		num_elems += dump_map_elem(fd, key, value, &info, btf, btf_wtr);
+		num_elems += dump_map_elem(fd, key, value, info, btf, wtr);
 		prev_key = key;
 	}
 
-	if (json_output)
-		jsonw_end_array(json_wtr);
-	else if (btf) {
-		jsonw_end_array(btf_wtr);
-		jsonw_destroy(&btf_wtr);
+	if (wtr) {
+		jsonw_end_array(wtr);	/* elements */
+		if (show_header)
+			jsonw_end_object(wtr);	/* map object */
 	} else {
 		printf("Found %u element%s\n", num_elems,
 		       num_elems != 1 ? "s" : "");
@@ -847,6 +1017,72 @@ exit_free:
 	return err;
 }
 
+static int do_dump(int argc, char **argv)
+{
+	json_writer_t *wtr = NULL, *btf_wtr = NULL;
+	struct bpf_map_info info = {};
+	int nb_fds, i = 0, btf = 0;
+	__u32 len = sizeof(info);
+	int *fds = NULL;
+	int err = -1;
+
+	if (argc != 2)
+		usage();
+
+	fds = malloc(sizeof(int));
+	if (!fds) {
+		p_err("mem alloc failed");
+		return -1;
+	}
+	nb_fds = map_parse_fds(&argc, &argv, &fds);
+	if (nb_fds < 1)
+		goto exit_free;
+
+	if (json_output) {
+		wtr = json_wtr;
+	} else {
+		btf = maps_have_btf(fds, nb_fds);
+		if (btf < 0)
+			goto exit_close;
+		if (btf) {
+			btf_wtr = get_btf_writer();
+			if (btf_wtr) {
+				wtr = btf_wtr;
+			} else {
+				p_info("failed to create json writer for btf. falling back to plain output");
+				btf = 0;
+			}
+		}
+	}
+
+	if (wtr && nb_fds > 1)
+		jsonw_start_array(wtr);	/* root array */
+	for (i = 0; i < nb_fds; i++) {
+		if (bpf_obj_get_info_by_fd(fds[i], &info, &len)) {
+			p_err("can't get map info: %s", strerror(errno));
+			break;
+		}
+		err = map_dump(fds[i], &info, wtr, btf, nb_fds > 1);
+		if (!wtr && i != nb_fds - 1)
+			printf("\n");
+
+		if (err)
+			break;
+		close(fds[i]);
+	}
+	if (wtr && nb_fds > 1)
+		jsonw_end_array(wtr);	/* root array */
+
+	if (btf)
+		jsonw_destroy(&btf_wtr);
+exit_close:
+	for (; i < nb_fds; i++)
+		close(fds[i]);
+exit_free:
+	free(fds);
+	return err;
+}
+
 static int alloc_key_value(struct bpf_map_info *info, void **key, void **value)
 {
 	*key = NULL;
-- 
cgit v1.2.3-59-g8ed1b


From aa915931ac3e53ccf371308e6750da510e3591dd Mon Sep 17 00:00:00 2001
From: Thadeu Lima de Souza Cascardo <cascardo@canonical.com>
Date: Fri, 13 Dec 2019 07:11:14 -0300
Subject: libbpf: Fix readelf output parsing for Fedora

Fedora binutils has been patched to show "other info" for a symbol at the
end of the line. This was done in order to support unmaintained scripts
that would break with the extra info. [1]

[1] https://src.fedoraproject.org/rpms/binutils/c/b8265c46f7ddae23a792ee8306fbaaeacba83bf8

This in turn has been done to fix the build of ruby, because of checksec.
[2] Thanks Michael Ellerman for the pointer.

[2] https://bugzilla.redhat.com/show_bug.cgi?id=1479302

As libbpf Makefile is not unmaintained, we can simply deal with either
output format, by just removing the "other info" field, as it always comes
inside brackets.

Fixes: 3464afdf11f9 (libbpf: Fix readelf output parsing on powerpc with recent binutils)
Reported-by: Justin Forbes <jmforbes@linuxtx.org>
Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@canonical.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Cc: Aurelien Jarno <aurelien@aurel32.net>
Link: https://lore.kernel.org/bpf/20191213101114.GA3986@calabresa
---
 tools/lib/bpf/Makefile | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/tools/lib/bpf/Makefile b/tools/lib/bpf/Makefile
index defae23a0169..23ae06c43d08 100644
--- a/tools/lib/bpf/Makefile
+++ b/tools/lib/bpf/Makefile
@@ -147,6 +147,7 @@ TAGS_PROG := $(if $(shell which etags 2>/dev/null),etags,ctags)
 
 GLOBAL_SYM_COUNT = $(shell readelf -s --wide $(BPF_IN_SHARED) | \
 			   cut -d "@" -f1 | sed 's/_v[0-9]_[0-9]_[0-9].*//' | \
+			   sed 's/\[.*\]//' | \
 			   awk '/GLOBAL/ && /DEFAULT/ && !/UND/ {print $$NF}' | \
 			   sort -u | wc -l)
 VERSIONED_SYM_COUNT = $(shell readelf -s --wide $(OUTPUT)libbpf.so | \
@@ -213,6 +214,7 @@ check_abi: $(OUTPUT)libbpf.so
 		     "versioned in $(VERSION_SCRIPT)." >&2;		 \
 		readelf -s --wide $(BPF_IN_SHARED) |			 \
 		    cut -d "@" -f1 | sed 's/_v[0-9]_[0-9]_[0-9].*//' |	 \
+		    sed 's/\[.*\]//' |					 \
 		    awk '/GLOBAL/ && /DEFAULT/ && !/UND/ {print $$NF}'|  \
 		    sort -u > $(OUTPUT)libbpf_global_syms.tmp;		 \
 		readelf -s --wide $(OUTPUT)libbpf.so |			 \
-- 
cgit v1.2.3-59-g8ed1b


From 29115cef85781057e78f0c967979edd40b16c209 Mon Sep 17 00:00:00 2001
From: Ursula Braun <ubraun@linux.ibm.com>
Date: Thu, 12 Dec 2019 22:35:41 +0100
Subject: net/smc: shorten lgr_cnt initialization

Save a line of code by making use of ATOMIC_INIT() for lgr_cnt.

Suggested-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Ursula Braun <ubraun@linux.ibm.com>
Signed-off-by: Karsten Graul <kgraul@linux.ibm.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
 net/smc/smc_core.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/net/smc/smc_core.c b/net/smc/smc_core.c
index bb92c7c6214c..c16b8d3e200b 100644
--- a/net/smc/smc_core.c
+++ b/net/smc/smc_core.c
@@ -41,7 +41,7 @@ static struct smc_lgr_list smc_lgr_list = {	/* established link groups */
 	.num = 0,
 };
 
-static atomic_t lgr_cnt;		/* number of existing link groups */
+static atomic_t lgr_cnt = ATOMIC_INIT(0); /* number of existing link groups */
 static DECLARE_WAIT_QUEUE_HEAD(lgrs_deleted);
 
 static void smc_buf_free(struct smc_link_group *lgr, bool is_rmb,
@@ -1297,7 +1297,6 @@ static struct notifier_block smc_reboot_notifier = {
 
 int __init smc_core_init(void)
 {
-	atomic_set(&lgr_cnt, 0);
 	return register_reboot_notifier(&smc_reboot_notifier);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 0e627190563e8be6fc9c2cc2a8a3c14bd961754c Mon Sep 17 00:00:00 2001
From: Arjun Roy <arjunroy@google.com>
Date: Sun, 15 Dec 2019 11:54:51 -0800
Subject: tcp: Set rcv zerocopy hint correctly if skb last frag is < PAGE_SIZE

At present, if the last frag of paged data in a skb has < PAGE_SIZE
data, we compute the recv_skip_hint as being equal to the size of that
frag and the entire next skb.

Instead, just return the runt frag size as the hint.

recv_skip_hint is used by the application to skip over
bytes that can not be mmaped, so returning a too big
chunk is pessimistic and forces more bytes to be copied.

Signed-off-by: Arjun Roy <arjunroy@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
 net/ipv4/tcp.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 09e2cae92956..93fe77c5b02d 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -1778,6 +1778,8 @@ static int tcp_zerocopy_receive(struct sock *sk,
 	while (length + PAGE_SIZE <= zc->length) {
 		if (zc->recv_skip_hint < PAGE_SIZE) {
 			if (skb) {
+				if (zc->recv_skip_hint > 0)
+					break;
 				skb = skb->next;
 				offset = seq - TCP_SKB_CB(skb)->seq;
 			} else {
-- 
cgit v1.2.3-59-g8ed1b


From 0d13bfce023ac7cef4d0a50b83750254ce31c479 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Fri, 13 Dec 2019 17:43:25 -0800
Subject: libbpf: Don't require root for bpf_object__open()

Reorganize bpf_object__open and bpf_object__load steps such that
bpf_object__open doesn't need root access. This was previously done for
feature probing and BTF sanitization. This doesn't have to happen on open,
though, so move all those steps into the load phase.

This is important, because it makes it possible for tools like bpftool, to
just open BPF object file and inspect their contents: programs, maps, BTF,
etc. For such operations it is prohibitive to require root access. On the
other hand, there is a lot of custom libbpf logic in those steps, so its best
avoided for tools to reimplement all that on their own.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20191214014341.3442258-2-andriin@fb.com
---
 tools/lib/bpf/libbpf.c | 83 +++++++++++++++++++++++++-------------------------
 1 file changed, 41 insertions(+), 42 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 27d5f7ecba32..dc993112b40b 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -101,13 +101,6 @@ void libbpf_print(enum libbpf_print_level level, const char *format, ...)
 
 #define STRERR_BUFSIZE  128
 
-#define CHECK_ERR(action, err, out) do {	\
-	err = action;			\
-	if (err)			\
-		goto out;		\
-} while (0)
-
-
 /* Copied from tools/perf/util/util.h */
 #ifndef zfree
 # define zfree(ptr) ({ free(*ptr); *ptr = NULL; })
@@ -864,8 +857,7 @@ bpf_object__init_internal_map(struct bpf_object *obj, enum libbpf_map_type type,
 	def->value_size = data->d_size;
 	def->max_entries = 1;
 	def->map_flags = type == LIBBPF_MAP_RODATA ? BPF_F_RDONLY_PROG : 0;
-	if (obj->caps.array_mmap)
-		def->map_flags |= BPF_F_MMAPABLE;
+	def->map_flags |= BPF_F_MMAPABLE;
 
 	pr_debug("map '%s' (global data): at sec_idx %d, offset %zu, flags %x.\n",
 		 map_name, map->sec_idx, map->sec_offset, def->map_flags);
@@ -888,8 +880,6 @@ static int bpf_object__init_global_data_maps(struct bpf_object *obj)
 {
 	int err;
 
-	if (!obj->caps.global_data)
-		return 0;
 	/*
 	 * Populate obj->maps with libbpf internal maps.
 	 */
@@ -1393,10 +1383,11 @@ static int bpf_object__init_user_btf_maps(struct bpf_object *obj, bool strict,
 	return 0;
 }
 
-static int bpf_object__init_maps(struct bpf_object *obj, bool relaxed_maps,
-				 const char *pin_root_path)
+static int bpf_object__init_maps(struct bpf_object *obj,
+				 struct bpf_object_open_opts *opts)
 {
-	bool strict = !relaxed_maps;
+	const char *pin_root_path = OPTS_GET(opts, pin_root_path, NULL);
+	bool strict = !OPTS_GET(opts, relaxed_maps, false);
 	int err;
 
 	err = bpf_object__init_user_maps(obj, strict);
@@ -1592,8 +1583,7 @@ static int bpf_object__sanitize_and_load_btf(struct bpf_object *obj)
 	return 0;
 }
 
-static int bpf_object__elf_collect(struct bpf_object *obj, bool relaxed_maps,
-				   const char *pin_root_path)
+static int bpf_object__elf_collect(struct bpf_object *obj)
 {
 	Elf *elf = obj->efile.elf;
 	GElf_Ehdr *ep = &obj->efile.ehdr;
@@ -1728,14 +1718,7 @@ static int bpf_object__elf_collect(struct bpf_object *obj, bool relaxed_maps,
 		pr_warn("Corrupted ELF file: index of strtab invalid\n");
 		return -LIBBPF_ERRNO__FORMAT;
 	}
-	err = bpf_object__init_btf(obj, btf_data, btf_ext_data);
-	if (!err)
-		err = bpf_object__init_maps(obj, relaxed_maps, pin_root_path);
-	if (!err)
-		err = bpf_object__sanitize_and_load_btf(obj);
-	if (!err)
-		err = bpf_object__init_prog_names(obj);
-	return err;
+	return bpf_object__init_btf(obj, btf_data, btf_ext_data);
 }
 
 static struct bpf_program *
@@ -1876,11 +1859,6 @@ static int bpf_program__record_reloc(struct bpf_program *prog,
 		pr_warn("bad data relo against section %u\n", shdr_idx);
 		return -LIBBPF_ERRNO__RELOC;
 	}
-	if (!obj->caps.global_data) {
-		pr_warn("relocation: kernel does not support global \'%s\' variable access in insns[%d]\n",
-			name, insn_idx);
-		return -LIBBPF_ERRNO__RELOC;
-	}
 	for (map_idx = 0; map_idx < nr_maps; map_idx++) {
 		map = &obj->maps[map_idx];
 		if (map->libbpf_type != type)
@@ -3918,12 +3896,10 @@ static struct bpf_object *
 __bpf_object__open(const char *path, const void *obj_buf, size_t obj_buf_sz,
 		   struct bpf_object_open_opts *opts)
 {
-	const char *pin_root_path;
 	struct bpf_program *prog;
 	struct bpf_object *obj;
 	const char *obj_name;
 	char tmp_name[64];
-	bool relaxed_maps;
 	__u32 attach_prog_fd;
 	int err;
 
@@ -3953,16 +3929,16 @@ __bpf_object__open(const char *path, const void *obj_buf, size_t obj_buf_sz,
 		return obj;
 
 	obj->relaxed_core_relocs = OPTS_GET(opts, relaxed_core_relocs, false);
-	relaxed_maps = OPTS_GET(opts, relaxed_maps, false);
-	pin_root_path = OPTS_GET(opts, pin_root_path, NULL);
 	attach_prog_fd = OPTS_GET(opts, attach_prog_fd, 0);
 
-	CHECK_ERR(bpf_object__elf_init(obj), err, out);
-	CHECK_ERR(bpf_object__check_endianness(obj), err, out);
-	CHECK_ERR(bpf_object__probe_caps(obj), err, out);
-	CHECK_ERR(bpf_object__elf_collect(obj, relaxed_maps, pin_root_path),
-		  err, out);
-	CHECK_ERR(bpf_object__collect_reloc(obj), err, out);
+	err = bpf_object__elf_init(obj);
+	err = err ? : bpf_object__check_endianness(obj);
+	err = err ? : bpf_object__elf_collect(obj);
+	err = err ? : bpf_object__init_maps(obj, opts);
+	err = err ? : bpf_object__init_prog_names(obj);
+	err = err ? : bpf_object__collect_reloc(obj);
+	if (err)
+		goto out;
 	bpf_object__elf_finish(obj);
 
 	bpf_object__for_each_program(prog, obj) {
@@ -4080,6 +4056,24 @@ int bpf_object__unload(struct bpf_object *obj)
 	return 0;
 }
 
+static int bpf_object__sanitize_maps(struct bpf_object *obj)
+{
+	struct bpf_map *m;
+
+	bpf_object__for_each_map(m, obj) {
+		if (!bpf_map__is_internal(m))
+			continue;
+		if (!obj->caps.global_data) {
+			pr_warn("kernel doesn't support global data\n");
+			return -ENOTSUP;
+		}
+		if (!obj->caps.array_mmap)
+			m->def.map_flags ^= BPF_F_MMAPABLE;
+	}
+
+	return 0;
+}
+
 int bpf_object__load_xattr(struct bpf_object_load_attr *attr)
 {
 	struct bpf_object *obj;
@@ -4098,9 +4092,14 @@ int bpf_object__load_xattr(struct bpf_object_load_attr *attr)
 
 	obj->loaded = true;
 
-	CHECK_ERR(bpf_object__create_maps(obj), err, out);
-	CHECK_ERR(bpf_object__relocate(obj, attr->target_btf_path), err, out);
-	CHECK_ERR(bpf_object__load_progs(obj, attr->log_level), err, out);
+	err = bpf_object__probe_caps(obj);
+	err = err ? : bpf_object__sanitize_and_load_btf(obj);
+	err = err ? : bpf_object__sanitize_maps(obj);
+	err = err ? : bpf_object__create_maps(obj);
+	err = err ? : bpf_object__relocate(obj, attr->target_btf_path);
+	err = err ? : bpf_object__load_progs(obj, attr->log_level);
+	if (err)
+		goto out;
 
 	return 0;
 out:
-- 
cgit v1.2.3-59-g8ed1b


From d7a18ea7e8b612669acd0131fd075e5c735c1ce5 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Fri, 13 Dec 2019 17:43:26 -0800
Subject: libbpf: Add generic bpf_program__attach()

Generalize BPF program attaching and allow libbpf to auto-detect type (and
extra parameters, where applicable) and attach supported BPF program types
based on program sections. Currently this is supported for:
- kprobe/kretprobe;
- tracepoint;
- raw tracepoint;
- tracing programs (typed raw TP/fentry/fexit).

More types support can be trivially added within this framework.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20191214014341.3442258-3-andriin@fb.com
---
 tools/lib/bpf/libbpf.c                             | 181 +++++++++++++++++----
 tools/lib/bpf/libbpf.h                             |   2 +
 tools/lib/bpf/libbpf.map                           |   2 +
 .../testing/selftests/bpf/prog_tests/probe_user.c  |   6 +-
 4 files changed, 153 insertions(+), 38 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index dc993112b40b..61b8cdf78332 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -4972,7 +4972,28 @@ void bpf_program__set_expected_attach_type(struct bpf_program *prog,
  */
 #define BPF_APROG_COMPAT(string, ptype) BPF_PROG_SEC(string, ptype)
 
-static const struct {
+#define SEC_DEF(sec_pfx, ptype, ...) {					    \
+	.sec = sec_pfx,							    \
+	.len = sizeof(sec_pfx) - 1,					    \
+	.prog_type = BPF_PROG_TYPE_##ptype,				    \
+	__VA_ARGS__							    \
+}
+
+struct bpf_sec_def;
+
+typedef struct bpf_link *(*attach_fn_t)(const struct bpf_sec_def *sec,
+					struct bpf_program *prog);
+
+static struct bpf_link *attach_kprobe(const struct bpf_sec_def *sec,
+				      struct bpf_program *prog);
+static struct bpf_link *attach_tp(const struct bpf_sec_def *sec,
+				  struct bpf_program *prog);
+static struct bpf_link *attach_raw_tp(const struct bpf_sec_def *sec,
+				      struct bpf_program *prog);
+static struct bpf_link *attach_trace(const struct bpf_sec_def *sec,
+				     struct bpf_program *prog);
+
+struct bpf_sec_def {
 	const char *sec;
 	size_t len;
 	enum bpf_prog_type prog_type;
@@ -4980,25 +5001,40 @@ static const struct {
 	bool is_attachable;
 	bool is_attach_btf;
 	enum bpf_attach_type attach_type;
-} section_names[] = {
+	attach_fn_t attach_fn;
+};
+
+static const struct bpf_sec_def section_defs[] = {
 	BPF_PROG_SEC("socket",			BPF_PROG_TYPE_SOCKET_FILTER),
 	BPF_PROG_SEC("sk_reuseport",		BPF_PROG_TYPE_SK_REUSEPORT),
-	BPF_PROG_SEC("kprobe/",			BPF_PROG_TYPE_KPROBE),
+	SEC_DEF("kprobe/", KPROBE,
+		.attach_fn = attach_kprobe),
 	BPF_PROG_SEC("uprobe/",			BPF_PROG_TYPE_KPROBE),
-	BPF_PROG_SEC("kretprobe/",		BPF_PROG_TYPE_KPROBE),
+	SEC_DEF("kretprobe/", KPROBE,
+		.attach_fn = attach_kprobe),
 	BPF_PROG_SEC("uretprobe/",		BPF_PROG_TYPE_KPROBE),
 	BPF_PROG_SEC("classifier",		BPF_PROG_TYPE_SCHED_CLS),
 	BPF_PROG_SEC("action",			BPF_PROG_TYPE_SCHED_ACT),
-	BPF_PROG_SEC("tracepoint/",		BPF_PROG_TYPE_TRACEPOINT),
-	BPF_PROG_SEC("tp/",			BPF_PROG_TYPE_TRACEPOINT),
-	BPF_PROG_SEC("raw_tracepoint/",		BPF_PROG_TYPE_RAW_TRACEPOINT),
-	BPF_PROG_SEC("raw_tp/",			BPF_PROG_TYPE_RAW_TRACEPOINT),
-	BPF_PROG_BTF("tp_btf/",			BPF_PROG_TYPE_TRACING,
-						BPF_TRACE_RAW_TP),
-	BPF_PROG_BTF("fentry/",			BPF_PROG_TYPE_TRACING,
-						BPF_TRACE_FENTRY),
-	BPF_PROG_BTF("fexit/",			BPF_PROG_TYPE_TRACING,
-						BPF_TRACE_FEXIT),
+	SEC_DEF("tracepoint/", TRACEPOINT,
+		.attach_fn = attach_tp),
+	SEC_DEF("tp/", TRACEPOINT,
+		.attach_fn = attach_tp),
+	SEC_DEF("raw_tracepoint/", RAW_TRACEPOINT,
+		.attach_fn = attach_raw_tp),
+	SEC_DEF("raw_tp/", RAW_TRACEPOINT,
+		.attach_fn = attach_raw_tp),
+	SEC_DEF("tp_btf/", TRACING,
+		.expected_attach_type = BPF_TRACE_RAW_TP,
+		.is_attach_btf = true,
+		.attach_fn = attach_trace),
+	SEC_DEF("fentry/", TRACING,
+		.expected_attach_type = BPF_TRACE_FENTRY,
+		.is_attach_btf = true,
+		.attach_fn = attach_trace),
+	SEC_DEF("fexit/", TRACING,
+		.expected_attach_type = BPF_TRACE_FEXIT,
+		.is_attach_btf = true,
+		.attach_fn = attach_trace),
 	BPF_PROG_SEC("xdp",			BPF_PROG_TYPE_XDP),
 	BPF_PROG_SEC("perf_event",		BPF_PROG_TYPE_PERF_EVENT),
 	BPF_PROG_SEC("lwt_in",			BPF_PROG_TYPE_LWT_IN),
@@ -5060,12 +5096,26 @@ static const struct {
 #undef BPF_APROG_SEC
 #undef BPF_EAPROG_SEC
 #undef BPF_APROG_COMPAT
+#undef SEC_DEF
 
 #define MAX_TYPE_NAME_SIZE 32
 
+static const struct bpf_sec_def *find_sec_def(const char *sec_name)
+{
+	int i, n = ARRAY_SIZE(section_defs);
+
+	for (i = 0; i < n; i++) {
+		if (strncmp(sec_name,
+			    section_defs[i].sec, section_defs[i].len))
+			continue;
+		return &section_defs[i];
+	}
+	return NULL;
+}
+
 static char *libbpf_get_type_names(bool attach_type)
 {
-	int i, len = ARRAY_SIZE(section_names) * MAX_TYPE_NAME_SIZE;
+	int i, len = ARRAY_SIZE(section_defs) * MAX_TYPE_NAME_SIZE;
 	char *buf;
 
 	buf = malloc(len);
@@ -5074,16 +5124,16 @@ static char *libbpf_get_type_names(bool attach_type)
 
 	buf[0] = '\0';
 	/* Forge string buf with all available names */
-	for (i = 0; i < ARRAY_SIZE(section_names); i++) {
-		if (attach_type && !section_names[i].is_attachable)
+	for (i = 0; i < ARRAY_SIZE(section_defs); i++) {
+		if (attach_type && !section_defs[i].is_attachable)
 			continue;
 
-		if (strlen(buf) + strlen(section_names[i].sec) + 2 > len) {
+		if (strlen(buf) + strlen(section_defs[i].sec) + 2 > len) {
 			free(buf);
 			return NULL;
 		}
 		strcat(buf, " ");
-		strcat(buf, section_names[i].sec);
+		strcat(buf, section_defs[i].sec);
 	}
 
 	return buf;
@@ -5092,19 +5142,19 @@ static char *libbpf_get_type_names(bool attach_type)
 int libbpf_prog_type_by_name(const char *name, enum bpf_prog_type *prog_type,
 			     enum bpf_attach_type *expected_attach_type)
 {
+	const struct bpf_sec_def *sec_def;
 	char *type_names;
-	int i;
 
 	if (!name)
 		return -EINVAL;
 
-	for (i = 0; i < ARRAY_SIZE(section_names); i++) {
-		if (strncmp(name, section_names[i].sec, section_names[i].len))
-			continue;
-		*prog_type = section_names[i].prog_type;
-		*expected_attach_type = section_names[i].expected_attach_type;
+	sec_def = find_sec_def(name);
+	if (sec_def) {
+		*prog_type = sec_def->prog_type;
+		*expected_attach_type = sec_def->expected_attach_type;
 		return 0;
 	}
+
 	pr_warn("failed to guess program type from ELF section '%s'\n", name);
 	type_names = libbpf_get_type_names(false);
 	if (type_names != NULL) {
@@ -5187,16 +5237,16 @@ static int libbpf_find_attach_btf_id(const char *name,
 	if (!name)
 		return -EINVAL;
 
-	for (i = 0; i < ARRAY_SIZE(section_names); i++) {
-		if (!section_names[i].is_attach_btf)
+	for (i = 0; i < ARRAY_SIZE(section_defs); i++) {
+		if (!section_defs[i].is_attach_btf)
 			continue;
-		if (strncmp(name, section_names[i].sec, section_names[i].len))
+		if (strncmp(name, section_defs[i].sec, section_defs[i].len))
 			continue;
 		if (attach_prog_fd)
-			err = libbpf_find_prog_btf_id(name + section_names[i].len,
+			err = libbpf_find_prog_btf_id(name + section_defs[i].len,
 						      attach_prog_fd);
 		else
-			err = libbpf_find_vmlinux_btf_id(name + section_names[i].len,
+			err = libbpf_find_vmlinux_btf_id(name + section_defs[i].len,
 							 attach_type);
 		if (err <= 0)
 			pr_warn("%s is not found in vmlinux BTF\n", name);
@@ -5215,12 +5265,12 @@ int libbpf_attach_type_by_name(const char *name,
 	if (!name)
 		return -EINVAL;
 
-	for (i = 0; i < ARRAY_SIZE(section_names); i++) {
-		if (strncmp(name, section_names[i].sec, section_names[i].len))
+	for (i = 0; i < ARRAY_SIZE(section_defs); i++) {
+		if (strncmp(name, section_defs[i].sec, section_defs[i].len))
 			continue;
-		if (!section_names[i].is_attachable)
+		if (!section_defs[i].is_attachable)
 			return -EINVAL;
-		*attach_type = section_names[i].attach_type;
+		*attach_type = section_defs[i].attach_type;
 		return 0;
 	}
 	pr_warn("failed to guess attach type based on ELF section name '%s'\n", name);
@@ -5680,6 +5730,18 @@ struct bpf_link *bpf_program__attach_kprobe(struct bpf_program *prog,
 	return link;
 }
 
+static struct bpf_link *attach_kprobe(const struct bpf_sec_def *sec,
+				      struct bpf_program *prog)
+{
+	const char *func_name;
+	bool retprobe;
+
+	func_name = bpf_program__title(prog, false) + sec->len;
+	retprobe = strcmp(sec->sec, "kretprobe/") == 0;
+
+	return bpf_program__attach_kprobe(prog, retprobe, func_name);
+}
+
 struct bpf_link *bpf_program__attach_uprobe(struct bpf_program *prog,
 					    bool retprobe, pid_t pid,
 					    const char *binary_path,
@@ -5792,6 +5854,32 @@ struct bpf_link *bpf_program__attach_tracepoint(struct bpf_program *prog,
 	return link;
 }
 
+static struct bpf_link *attach_tp(const struct bpf_sec_def *sec,
+				  struct bpf_program *prog)
+{
+	char *sec_name, *tp_cat, *tp_name;
+	struct bpf_link *link;
+
+	sec_name = strdup(bpf_program__title(prog, false));
+	if (!sec_name)
+		return ERR_PTR(-ENOMEM);
+
+	/* extract "tp/<category>/<name>" */
+	tp_cat = sec_name + sec->len;
+	tp_name = strchr(tp_cat, '/');
+	if (!tp_name) {
+		link = ERR_PTR(-EINVAL);
+		goto out;
+	}
+	*tp_name = '\0';
+	tp_name++;
+
+	link = bpf_program__attach_tracepoint(prog, tp_cat, tp_name);
+out:
+	free(sec_name);
+	return link;
+}
+
 static int bpf_link__destroy_fd(struct bpf_link *link)
 {
 	struct bpf_link_fd *l = (void *)link;
@@ -5831,6 +5919,14 @@ struct bpf_link *bpf_program__attach_raw_tracepoint(struct bpf_program *prog,
 	return (struct bpf_link *)link;
 }
 
+static struct bpf_link *attach_raw_tp(const struct bpf_sec_def *sec,
+				      struct bpf_program *prog)
+{
+	const char *tp_name = bpf_program__title(prog, false) + sec->len;
+
+	return bpf_program__attach_raw_tracepoint(prog, tp_name);
+}
+
 struct bpf_link *bpf_program__attach_trace(struct bpf_program *prog)
 {
 	char errmsg[STRERR_BUFSIZE];
@@ -5862,6 +5958,23 @@ struct bpf_link *bpf_program__attach_trace(struct bpf_program *prog)
 	return (struct bpf_link *)link;
 }
 
+static struct bpf_link *attach_trace(const struct bpf_sec_def *sec,
+				     struct bpf_program *prog)
+{
+	return bpf_program__attach_trace(prog);
+}
+
+struct bpf_link *bpf_program__attach(struct bpf_program *prog)
+{
+	const struct bpf_sec_def *sec_def;
+
+	sec_def = find_sec_def(bpf_program__title(prog, false));
+	if (!sec_def || !sec_def->attach_fn)
+		return ERR_PTR(-ESRCH);
+
+	return sec_def->attach_fn(sec_def, prog);
+}
+
 enum bpf_perf_event_ret
 bpf_perf_event_read_simple(void *mmap_mem, size_t mmap_size, size_t page_size,
 			   void **copy_mem, size_t *copy_size,
diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index 0dbf4bfba0c4..804f445c9957 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -237,6 +237,8 @@ struct bpf_link;
 
 LIBBPF_API int bpf_link__destroy(struct bpf_link *link);
 
+LIBBPF_API struct bpf_link *
+bpf_program__attach(struct bpf_program *prog);
 LIBBPF_API struct bpf_link *
 bpf_program__attach_perf_event(struct bpf_program *prog, int pfd);
 LIBBPF_API struct bpf_link *
diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
index 495df575f87f..757a88f64b5a 100644
--- a/tools/lib/bpf/libbpf.map
+++ b/tools/lib/bpf/libbpf.map
@@ -210,4 +210,6 @@ LIBBPF_0.0.6 {
 } LIBBPF_0.0.5;
 
 LIBBPF_0.0.7 {
+	global:
+		bpf_program__attach;
 } LIBBPF_0.0.6;
diff --git a/tools/testing/selftests/bpf/prog_tests/probe_user.c b/tools/testing/selftests/bpf/prog_tests/probe_user.c
index 8a3187dec048..7aecfd9e87d1 100644
--- a/tools/testing/selftests/bpf/prog_tests/probe_user.c
+++ b/tools/testing/selftests/bpf/prog_tests/probe_user.c
@@ -3,8 +3,7 @@
 
 void test_probe_user(void)
 {
-#define kprobe_name "__sys_connect"
-	const char *prog_name = "kprobe/" kprobe_name;
+	const char *prog_name = "kprobe/__sys_connect";
 	const char *obj_file = "./test_probe_user.o";
 	DECLARE_LIBBPF_OPTS(bpf_object_open_opts, opts, );
 	int err, results_map_fd, sock_fd, duration = 0;
@@ -33,8 +32,7 @@ void test_probe_user(void)
 		  "err %d\n", results_map_fd))
 		goto cleanup;
 
-	kprobe_link = bpf_program__attach_kprobe(kprobe_prog, false,
-						 kprobe_name);
+	kprobe_link = bpf_program__attach(kprobe_prog);
 	if (CHECK(IS_ERR(kprobe_link), "attach_kprobe",
 		  "err %ld\n", PTR_ERR(kprobe_link))) {
 		kprobe_link = NULL;
-- 
cgit v1.2.3-59-g8ed1b


From 612d05be250aa8804d3baba7a12445a267a580d3 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Fri, 13 Dec 2019 17:43:27 -0800
Subject: libbpf: Move non-public APIs from libbpf.h to libbpf_internal.h

Few libbpf APIs are not public but currently exposed through libbpf.h to be
used by bpftool. Move them to libbpf_internal.h, where intent of being
non-stable and non-public is much more obvious.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20191214014341.3442258-4-andriin@fb.com
---
 tools/bpf/bpftool/net.c         |  1 +
 tools/lib/bpf/libbpf.h          | 17 -----------------
 tools/lib/bpf/libbpf_internal.h | 17 +++++++++++++++++
 3 files changed, 18 insertions(+), 17 deletions(-)

diff --git a/tools/bpf/bpftool/net.c b/tools/bpf/bpftool/net.c
index 4f52d3151616..d93bee298e54 100644
--- a/tools/bpf/bpftool/net.c
+++ b/tools/bpf/bpftool/net.c
@@ -18,6 +18,7 @@
 
 #include <bpf.h>
 #include <nlattr.h>
+#include "libbpf_internal.h"
 #include "main.h"
 #include "netlink_dumper.h"
 
diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index 804f445c9957..2698fbcb0c79 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -126,11 +126,6 @@ bpf_object__open_buffer(const void *obj_buf, size_t obj_buf_sz,
 LIBBPF_API struct bpf_object *
 bpf_object__open_xattr(struct bpf_object_open_attr *attr);
 
-int bpf_object__section_size(const struct bpf_object *obj, const char *name,
-			     __u32 *size);
-int bpf_object__variable_offset(const struct bpf_object *obj, const char *name,
-				__u32 *off);
-
 enum libbpf_pin_type {
 	LIBBPF_PIN_NONE,
 	/* PIN_BY_NAME: pin maps by name (in /sys/fs/bpf by default) */
@@ -514,18 +509,6 @@ bpf_perf_event_read_simple(void *mmap_mem, size_t mmap_size, size_t page_size,
 			   void **copy_mem, size_t *copy_size,
 			   bpf_perf_event_print_t fn, void *private_data);
 
-struct nlattr;
-typedef int (*libbpf_dump_nlmsg_t)(void *cookie, void *msg, struct nlattr **tb);
-int libbpf_netlink_open(unsigned int *nl_pid);
-int libbpf_nl_get_link(int sock, unsigned int nl_pid,
-		       libbpf_dump_nlmsg_t dump_link_nlmsg, void *cookie);
-int libbpf_nl_get_class(int sock, unsigned int nl_pid, int ifindex,
-			libbpf_dump_nlmsg_t dump_class_nlmsg, void *cookie);
-int libbpf_nl_get_qdisc(int sock, unsigned int nl_pid, int ifindex,
-			libbpf_dump_nlmsg_t dump_qdisc_nlmsg, void *cookie);
-int libbpf_nl_get_filter(int sock, unsigned int nl_pid, int ifindex, int handle,
-			 libbpf_dump_nlmsg_t dump_filter_nlmsg, void *cookie);
-
 struct bpf_prog_linfo;
 struct bpf_prog_info;
 
diff --git a/tools/lib/bpf/libbpf_internal.h b/tools/lib/bpf/libbpf_internal.h
index f4f10715db40..7d71621bb7e8 100644
--- a/tools/lib/bpf/libbpf_internal.h
+++ b/tools/lib/bpf/libbpf_internal.h
@@ -100,6 +100,23 @@ int parse_cpu_mask_file(const char *fcpu, bool **mask, int *mask_sz);
 int libbpf__load_raw_btf(const char *raw_types, size_t types_len,
 			 const char *str_sec, size_t str_len);
 
+int bpf_object__section_size(const struct bpf_object *obj, const char *name,
+			     __u32 *size);
+int bpf_object__variable_offset(const struct bpf_object *obj, const char *name,
+				__u32 *off);
+
+struct nlattr;
+typedef int (*libbpf_dump_nlmsg_t)(void *cookie, void *msg, struct nlattr **tb);
+int libbpf_netlink_open(unsigned int *nl_pid);
+int libbpf_nl_get_link(int sock, unsigned int nl_pid,
+		       libbpf_dump_nlmsg_t dump_link_nlmsg, void *cookie);
+int libbpf_nl_get_class(int sock, unsigned int nl_pid, int ifindex,
+			libbpf_dump_nlmsg_t dump_class_nlmsg, void *cookie);
+int libbpf_nl_get_qdisc(int sock, unsigned int nl_pid, int ifindex,
+			libbpf_dump_nlmsg_t dump_qdisc_nlmsg, void *cookie);
+int libbpf_nl_get_filter(int sock, unsigned int nl_pid, int ifindex, int handle,
+			 libbpf_dump_nlmsg_t dump_filter_nlmsg, void *cookie);
+
 struct btf_ext_info {
 	/*
 	 * info points to the individual info section (e.g. func_info and
-- 
cgit v1.2.3-59-g8ed1b


From 917f6b7b07a46e53fa73e112d23c97d1b201a877 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Fri, 13 Dec 2019 17:43:28 -0800
Subject: libbpf: Add BPF_EMBED_OBJ macro for embedding BPF .o files

Add a convenience macro BPF_EMBED_OBJ, which allows to embed other files
(typically used to embed BPF .o files) into a hosting userspace programs. To
C program it is exposed as struct bpf_embed_data, containing a pointer to
raw data and its size in bytes.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20191214014341.3442258-5-andriin@fb.com
---
 tools/lib/bpf/libbpf.h                             | 35 ++++++++++++++++++++++
 .../selftests/bpf/prog_tests/attach_probe.c        | 23 ++------------
 2 files changed, 38 insertions(+), 20 deletions(-)

diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index 2698fbcb0c79..fa803dde1f46 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -615,6 +615,41 @@ bpf_program__bpil_offs_to_addr(struct bpf_prog_info_linear *info_linear);
  */
 LIBBPF_API int libbpf_num_possible_cpus(void);
 
+struct bpf_embed_data {
+	void *data;
+	size_t size;
+};
+
+#define BPF_EMBED_OBJ_DECLARE(NAME)					\
+extern struct bpf_embed_data NAME##_embed;				\
+extern char NAME##_data[];						\
+extern char NAME##_data_end[];
+
+#define __BPF_EMBED_OBJ(NAME, PATH, SZ, ASM_TYPE)			\
+asm (									\
+"	.pushsection \".rodata\", \"a\", @progbits		\n"	\
+"	.global "#NAME"_data					\n"	\
+#NAME"_data:							\n"	\
+"	.incbin \"" PATH "\"					\n"	\
+"	.global "#NAME"_data_end				\n"	\
+#NAME"_data_end:						\n"	\
+"	.global "#NAME"_embed					\n"	\
+"	.type "#NAME"_embed, @object				\n"	\
+"	.size "#NAME"_size, "#SZ"				\n"	\
+"	.align 8,						\n"	\
+#NAME"_embed:							\n"	\
+"	"ASM_TYPE" "#NAME"_data					\n"	\
+"	"ASM_TYPE" "#NAME"_data_end - "#NAME"_data 		\n"	\
+"	.popsection						\n"	\
+);									\
+BPF_EMBED_OBJ_DECLARE(NAME)
+
+#if __SIZEOF_POINTER__ == 4
+#define BPF_EMBED_OBJ(NAME, PATH) __BPF_EMBED_OBJ(NAME, PATH, 8, ".long")
+#else
+#define BPF_EMBED_OBJ(NAME, PATH) __BPF_EMBED_OBJ(NAME, PATH, 16, ".quad")
+#endif
+
 #ifdef __cplusplus
 } /* extern "C" */
 #endif
diff --git a/tools/testing/selftests/bpf/prog_tests/attach_probe.c b/tools/testing/selftests/bpf/prog_tests/attach_probe.c
index a83111a32d4a..b2e7c1424b07 100644
--- a/tools/testing/selftests/bpf/prog_tests/attach_probe.c
+++ b/tools/testing/selftests/bpf/prog_tests/attach_probe.c
@@ -1,24 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <test_progs.h>
 
-#define EMBED_FILE(NAME, PATH)						    \
-asm (									    \
-"      .pushsection \".rodata\", \"a\", @progbits              \n"	    \
-"      .global "#NAME"_data                                    \n"	    \
-#NAME"_data:                                                   \n"	    \
-"      .incbin \"" PATH "\"                                    \n"	    \
-#NAME"_data_end:                                               \n"	    \
-"      .global "#NAME"_size                                    \n"	    \
-"      .type "#NAME"_size, @object                             \n"	    \
-"      .size "#NAME"_size, 4                                   \n"	    \
-"      .align 4,                                               \n"	    \
-#NAME"_size:                                                   \n"	    \
-"      .int "#NAME"_data_end - "#NAME"_data                    \n"	    \
-"      .popsection                                             \n"	    \
-);									    \
-extern char NAME##_data[];						    \
-extern int NAME##_size;
-
 ssize_t get_base_addr() {
 	size_t start;
 	char buf[256];
@@ -39,7 +21,7 @@ ssize_t get_base_addr() {
 	return -EINVAL;
 }
 
-EMBED_FILE(probe, "test_attach_probe.o");
+BPF_EMBED_OBJ(probe, "test_attach_probe.o");
 
 void test_attach_probe(void)
 {
@@ -73,7 +55,8 @@ void test_attach_probe(void)
 	uprobe_offset = (size_t)&get_base_addr - base_addr;
 
 	/* open object */
-	obj = bpf_object__open_mem(probe_data, probe_size, &open_opts);
+	obj = bpf_object__open_mem(probe_embed.data, probe_embed.size,
+				   &open_opts);
 	if (CHECK(IS_ERR(obj), "obj_open_mem", "err %ld\n", PTR_ERR(obj)))
 		return;
 
-- 
cgit v1.2.3-59-g8ed1b


From 544402d4b49332a4a9b2b8fff20f9d9f5ef86559 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Fri, 13 Dec 2019 17:43:29 -0800
Subject: libbpf: Extract common user-facing helpers

LIBBPF_API and DECLARE_LIBBPF_OPTS are needed in many public libbpf API
headers. Extract them into libbpf_common.h to avoid unnecessary
interdependency between btf.h, libbpf.h, and bpf.h or code duplication.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191214014341.3442258-6-andriin@fb.com
---
 tools/lib/bpf/bpf.h           |  6 ++----
 tools/lib/bpf/btf.h           |  6 ++----
 tools/lib/bpf/libbpf.h        | 28 ++--------------------------
 tools/lib/bpf/libbpf_common.h | 38 ++++++++++++++++++++++++++++++++++++++
 4 files changed, 44 insertions(+), 34 deletions(-)
 create mode 100644 tools/lib/bpf/libbpf_common.h

diff --git a/tools/lib/bpf/bpf.h b/tools/lib/bpf/bpf.h
index 3c791fa8e68e..269807ce9ef5 100644
--- a/tools/lib/bpf/bpf.h
+++ b/tools/lib/bpf/bpf.h
@@ -28,14 +28,12 @@
 #include <stddef.h>
 #include <stdint.h>
 
+#include "libbpf_common.h"
+
 #ifdef __cplusplus
 extern "C" {
 #endif
 
-#ifndef LIBBPF_API
-#define LIBBPF_API __attribute__((visibility("default")))
-#endif
-
 struct bpf_create_map_attr {
 	const char *name;
 	enum bpf_map_type map_type;
diff --git a/tools/lib/bpf/btf.h b/tools/lib/bpf/btf.h
index d9ac73a02cde..5fc23b988deb 100644
--- a/tools/lib/bpf/btf.h
+++ b/tools/lib/bpf/btf.h
@@ -8,14 +8,12 @@
 #include <linux/btf.h>
 #include <linux/types.h>
 
+#include "libbpf_common.h"
+
 #ifdef __cplusplus
 extern "C" {
 #endif
 
-#ifndef LIBBPF_API
-#define LIBBPF_API __attribute__((visibility("default")))
-#endif
-
 #define BTF_ELF_SEC ".BTF"
 #define BTF_EXT_ELF_SEC ".BTF.ext"
 #define MAPS_ELF_SEC ".maps"
diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index fa803dde1f46..49e6fa01024b 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -17,14 +17,12 @@
 #include <sys/types.h>  // for size_t
 #include <linux/bpf.h>
 
+#include "libbpf_common.h"
+
 #ifdef __cplusplus
 extern "C" {
 #endif
 
-#ifndef LIBBPF_API
-#define LIBBPF_API __attribute__((visibility("default")))
-#endif
-
 enum libbpf_errno {
 	__LIBBPF_ERRNO__START = 4000,
 
@@ -67,28 +65,6 @@ struct bpf_object_open_attr {
 	enum bpf_prog_type prog_type;
 };
 
-/* Helper macro to declare and initialize libbpf options struct
- *
- * This dance with uninitialized declaration, followed by memset to zero,
- * followed by assignment using compound literal syntax is done to preserve
- * ability to use a nice struct field initialization syntax and **hopefully**
- * have all the padding bytes initialized to zero. It's not guaranteed though,
- * when copying literal, that compiler won't copy garbage in literal's padding
- * bytes, but that's the best way I've found and it seems to work in practice.
- *
- * Macro declares opts struct of given type and name, zero-initializes,
- * including any extra padding, it with memset() and then assigns initial
- * values provided by users in struct initializer-syntax as varargs.
- */
-#define DECLARE_LIBBPF_OPTS(TYPE, NAME, ...)				    \
-	struct TYPE NAME = ({ 						    \
-		memset(&NAME, 0, sizeof(struct TYPE));			    \
-		(struct TYPE) {						    \
-			.sz = sizeof(struct TYPE),			    \
-			__VA_ARGS__					    \
-		};							    \
-	})
-
 struct bpf_object_open_opts {
 	/* size of this struct, for forward/backward compatiblity */
 	size_t sz;
diff --git a/tools/lib/bpf/libbpf_common.h b/tools/lib/bpf/libbpf_common.h
new file mode 100644
index 000000000000..4fb833840961
--- /dev/null
+++ b/tools/lib/bpf/libbpf_common.h
@@ -0,0 +1,38 @@
+/* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */
+
+/*
+ * Common user-facing libbpf helpers.
+ *
+ * Copyright (c) 2019 Facebook
+ */
+
+#ifndef __LIBBPF_LIBBPF_COMMON_H
+#define __LIBBPF_LIBBPF_COMMON_H
+
+#ifndef LIBBPF_API
+#define LIBBPF_API __attribute__((visibility("default")))
+#endif
+
+/* Helper macro to declare and initialize libbpf options struct
+ *
+ * This dance with uninitialized declaration, followed by memset to zero,
+ * followed by assignment using compound literal syntax is done to preserve
+ * ability to use a nice struct field initialization syntax and **hopefully**
+ * have all the padding bytes initialized to zero. It's not guaranteed though,
+ * when copying literal, that compiler won't copy garbage in literal's padding
+ * bytes, but that's the best way I've found and it seems to work in practice.
+ *
+ * Macro declares opts struct of given type and name, zero-initializes,
+ * including any extra padding, it with memset() and then assigns initial
+ * values provided by users in struct initializer-syntax as varargs.
+ */
+#define DECLARE_LIBBPF_OPTS(TYPE, NAME, ...)				    \
+	struct TYPE NAME = ({ 						    \
+		memset(&NAME, 0, sizeof(struct TYPE));			    \
+		(struct TYPE) {						    \
+			.sz = sizeof(struct TYPE),			    \
+			__VA_ARGS__					    \
+		};							    \
+	})
+
+#endif /* __LIBBPF_LIBBPF_COMMON_H */
-- 
cgit v1.2.3-59-g8ed1b


From 3d208f4ca111a614903f49d5a77b93ddc6de294e Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Fri, 13 Dec 2019 17:43:30 -0800
Subject: libbpf: Expose btf__align_of() API

Expose BTF API that calculates type alignment requirements.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191214014341.3442258-7-andriin@fb.com
---
 tools/lib/bpf/btf.c      | 39 +++++++++++++++++++++++++++++++++++++++
 tools/lib/bpf/btf.h      |  1 +
 tools/lib/bpf/btf_dump.c | 47 ++++++-----------------------------------------
 tools/lib/bpf/libbpf.map |  1 +
 4 files changed, 47 insertions(+), 41 deletions(-)

diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
index 88efa2bb7137..84fe82f27bef 100644
--- a/tools/lib/bpf/btf.c
+++ b/tools/lib/bpf/btf.c
@@ -278,6 +278,45 @@ done:
 	return nelems * size;
 }
 
+int btf__align_of(const struct btf *btf, __u32 id)
+{
+	const struct btf_type *t = btf__type_by_id(btf, id);
+	__u16 kind = btf_kind(t);
+
+	switch (kind) {
+	case BTF_KIND_INT:
+	case BTF_KIND_ENUM:
+		return min(sizeof(void *), t->size);
+	case BTF_KIND_PTR:
+		return sizeof(void *);
+	case BTF_KIND_TYPEDEF:
+	case BTF_KIND_VOLATILE:
+	case BTF_KIND_CONST:
+	case BTF_KIND_RESTRICT:
+		return btf__align_of(btf, t->type);
+	case BTF_KIND_ARRAY:
+		return btf__align_of(btf, btf_array(t)->type);
+	case BTF_KIND_STRUCT:
+	case BTF_KIND_UNION: {
+		const struct btf_member *m = btf_members(t);
+		__u16 vlen = btf_vlen(t);
+		int i, align = 1, t;
+
+		for (i = 0; i < vlen; i++, m++) {
+			t = btf__align_of(btf, m->type);
+			if (t <= 0)
+				return t;
+			align = max(align, t);
+		}
+
+		return align;
+	}
+	default:
+		pr_warn("unsupported BTF_KIND:%u\n", btf_kind(t));
+		return 0;
+	}
+}
+
 int btf__resolve_type(const struct btf *btf, __u32 type_id)
 {
 	const struct btf_type *t;
diff --git a/tools/lib/bpf/btf.h b/tools/lib/bpf/btf.h
index 5fc23b988deb..a114c8ef4f08 100644
--- a/tools/lib/bpf/btf.h
+++ b/tools/lib/bpf/btf.h
@@ -77,6 +77,7 @@ LIBBPF_API const struct btf_type *btf__type_by_id(const struct btf *btf,
 						  __u32 id);
 LIBBPF_API __s64 btf__resolve_size(const struct btf *btf, __u32 type_id);
 LIBBPF_API int btf__resolve_type(const struct btf *btf, __u32 type_id);
+LIBBPF_API int btf__align_of(const struct btf *btf, __u32 id);
 LIBBPF_API int btf__fd(const struct btf *btf);
 LIBBPF_API const void *btf__get_raw_data(const struct btf *btf, __u32 *size);
 LIBBPF_API const char *btf__name_by_offset(const struct btf *btf, __u32 offset);
diff --git a/tools/lib/bpf/btf_dump.c b/tools/lib/bpf/btf_dump.c
index cb126d8fcf75..53393026d085 100644
--- a/tools/lib/bpf/btf_dump.c
+++ b/tools/lib/bpf/btf_dump.c
@@ -752,41 +752,6 @@ static void btf_dump_emit_type(struct btf_dump *d, __u32 id, __u32 cont_id)
 	}
 }
 
-static int btf_align_of(const struct btf *btf, __u32 id)
-{
-	const struct btf_type *t = btf__type_by_id(btf, id);
-	__u16 kind = btf_kind(t);
-
-	switch (kind) {
-	case BTF_KIND_INT:
-	case BTF_KIND_ENUM:
-		return min(sizeof(void *), t->size);
-	case BTF_KIND_PTR:
-		return sizeof(void *);
-	case BTF_KIND_TYPEDEF:
-	case BTF_KIND_VOLATILE:
-	case BTF_KIND_CONST:
-	case BTF_KIND_RESTRICT:
-		return btf_align_of(btf, t->type);
-	case BTF_KIND_ARRAY:
-		return btf_align_of(btf, btf_array(t)->type);
-	case BTF_KIND_STRUCT:
-	case BTF_KIND_UNION: {
-		const struct btf_member *m = btf_members(t);
-		__u16 vlen = btf_vlen(t);
-		int i, align = 1;
-
-		for (i = 0; i < vlen; i++, m++)
-			align = max(align, btf_align_of(btf, m->type));
-
-		return align;
-	}
-	default:
-		pr_warn("unsupported BTF_KIND:%u\n", btf_kind(t));
-		return 1;
-	}
-}
-
 static bool btf_is_struct_packed(const struct btf *btf, __u32 id,
 				 const struct btf_type *t)
 {
@@ -794,18 +759,18 @@ static bool btf_is_struct_packed(const struct btf *btf, __u32 id,
 	int align, i, bit_sz;
 	__u16 vlen;
 
-	align = btf_align_of(btf, id);
+	align = btf__align_of(btf, id);
 	/* size of a non-packed struct has to be a multiple of its alignment*/
-	if (t->size % align)
+	if (align && t->size % align)
 		return true;
 
 	m = btf_members(t);
 	vlen = btf_vlen(t);
 	/* all non-bitfield fields have to be naturally aligned */
 	for (i = 0; i < vlen; i++, m++) {
-		align = btf_align_of(btf, m->type);
+		align = btf__align_of(btf, m->type);
 		bit_sz = btf_member_bitfield_size(t, i);
-		if (bit_sz == 0 && m->offset % (8 * align) != 0)
+		if (align && bit_sz == 0 && m->offset % (8 * align) != 0)
 			return true;
 	}
 
@@ -889,7 +854,7 @@ static void btf_dump_emit_struct_def(struct btf_dump *d,
 		fname = btf_name_of(d, m->name_off);
 		m_sz = btf_member_bitfield_size(t, i);
 		m_off = btf_member_bit_offset(t, i);
-		align = packed ? 1 : btf_align_of(d->btf, m->type);
+		align = packed ? 1 : btf__align_of(d->btf, m->type);
 
 		btf_dump_emit_bit_padding(d, off, m_off, m_sz, align, lvl + 1);
 		btf_dump_printf(d, "\n%s", pfx(lvl + 1));
@@ -907,7 +872,7 @@ static void btf_dump_emit_struct_def(struct btf_dump *d,
 
 	/* pad at the end, if necessary */
 	if (is_struct) {
-		align = packed ? 1 : btf_align_of(d->btf, id);
+		align = packed ? 1 : btf__align_of(d->btf, id);
 		btf_dump_emit_bit_padding(d, off, t->size * 8, 0, align,
 					  lvl + 1);
 	}
diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
index 757a88f64b5a..e7fcca36f186 100644
--- a/tools/lib/bpf/libbpf.map
+++ b/tools/lib/bpf/libbpf.map
@@ -212,4 +212,5 @@ LIBBPF_0.0.6 {
 LIBBPF_0.0.7 {
 	global:
 		bpf_program__attach;
+		btf__align_of;
 } LIBBPF_0.0.6;
-- 
cgit v1.2.3-59-g8ed1b


From 9f81654eebe8de7e0db15534816d8f6c84b2e1e5 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Fri, 13 Dec 2019 17:43:31 -0800
Subject: libbpf: Expose BTF-to-C type declaration emitting API

Expose API that allows to emit type declaration and field/variable definition
(if optional field name is specified) in valid C syntax for any provided BTF
type. This is going to be used by bpftool when emitting data section layout as
a struct. As part of making this API useful in a stand-alone fashion, move
initialization of some of the internal btf_dump state to earlier phase.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20191214014341.3442258-8-andriin@fb.com
---
 tools/lib/bpf/btf.h      | 22 ++++++++++++++++
 tools/lib/bpf/btf_dump.c | 68 ++++++++++++++++++++++++++++++------------------
 tools/lib/bpf/libbpf.map |  1 +
 3 files changed, 65 insertions(+), 26 deletions(-)

diff --git a/tools/lib/bpf/btf.h b/tools/lib/bpf/btf.h
index a114c8ef4f08..8d73f7f5551f 100644
--- a/tools/lib/bpf/btf.h
+++ b/tools/lib/bpf/btf.h
@@ -126,6 +126,28 @@ LIBBPF_API void btf_dump__free(struct btf_dump *d);
 
 LIBBPF_API int btf_dump__dump_type(struct btf_dump *d, __u32 id);
 
+struct btf_dump_emit_type_decl_opts {
+	/* size of this struct, for forward/backward compatiblity */
+	size_t sz;
+	/* optional field name for type declaration, e.g.:
+	 * - struct my_struct <FNAME>
+	 * - void (*<FNAME>)(int)
+	 * - char (*<FNAME>)[123]
+	 */
+	const char *field_name;
+	/* extra indentation level (in number of tabs) to emit for multi-line
+	 * type declarations (e.g., anonymous struct); applies for lines
+	 * starting from the second one (first line is assumed to have
+	 * necessary indentation already
+	 */
+	int indent_level;
+};
+#define btf_dump_emit_type_decl_opts__last_field indent_level
+
+LIBBPF_API int
+btf_dump__emit_type_decl(struct btf_dump *d, __u32 id,
+			 const struct btf_dump_emit_type_decl_opts *opts);
+
 /*
  * A set of helpers for easier BTF types handling
  */
diff --git a/tools/lib/bpf/btf_dump.c b/tools/lib/bpf/btf_dump.c
index 53393026d085..e95f7710f210 100644
--- a/tools/lib/bpf/btf_dump.c
+++ b/tools/lib/bpf/btf_dump.c
@@ -116,6 +116,8 @@ static void btf_dump_printf(const struct btf_dump *d, const char *fmt, ...)
 	va_end(args);
 }
 
+static int btf_dump_mark_referenced(struct btf_dump *d);
+
 struct btf_dump *btf_dump__new(const struct btf *btf,
 			       const struct btf_ext *btf_ext,
 			       const struct btf_dump_opts *opts,
@@ -137,18 +139,39 @@ struct btf_dump *btf_dump__new(const struct btf *btf,
 	if (IS_ERR(d->type_names)) {
 		err = PTR_ERR(d->type_names);
 		d->type_names = NULL;
-		btf_dump__free(d);
-		return ERR_PTR(err);
 	}
 	d->ident_names = hashmap__new(str_hash_fn, str_equal_fn, NULL);
 	if (IS_ERR(d->ident_names)) {
 		err = PTR_ERR(d->ident_names);
 		d->ident_names = NULL;
-		btf_dump__free(d);
-		return ERR_PTR(err);
+		goto err;
+	}
+	d->type_states = calloc(1 + btf__get_nr_types(d->btf),
+				sizeof(d->type_states[0]));
+	if (!d->type_states) {
+		err = -ENOMEM;
+		goto err;
+	}
+	d->cached_names = calloc(1 + btf__get_nr_types(d->btf),
+				 sizeof(d->cached_names[0]));
+	if (!d->cached_names) {
+		err = -ENOMEM;
+		goto err;
 	}
 
+	/* VOID is special */
+	d->type_states[0].order_state = ORDERED;
+	d->type_states[0].emit_state = EMITTED;
+
+	/* eagerly determine referenced types for anon enums */
+	err = btf_dump_mark_referenced(d);
+	if (err)
+		goto err;
+
 	return d;
+err:
+	btf_dump__free(d);
+	return ERR_PTR(err);
 }
 
 void btf_dump__free(struct btf_dump *d)
@@ -175,7 +198,6 @@ void btf_dump__free(struct btf_dump *d)
 	free(d);
 }
 
-static int btf_dump_mark_referenced(struct btf_dump *d);
 static int btf_dump_order_type(struct btf_dump *d, __u32 id, bool through_ptr);
 static void btf_dump_emit_type(struct btf_dump *d, __u32 id, __u32 cont_id);
 
@@ -202,27 +224,6 @@ int btf_dump__dump_type(struct btf_dump *d, __u32 id)
 	if (id > btf__get_nr_types(d->btf))
 		return -EINVAL;
 
-	/* type states are lazily allocated, as they might not be needed */
-	if (!d->type_states) {
-		d->type_states = calloc(1 + btf__get_nr_types(d->btf),
-					sizeof(d->type_states[0]));
-		if (!d->type_states)
-			return -ENOMEM;
-		d->cached_names = calloc(1 + btf__get_nr_types(d->btf),
-					 sizeof(d->cached_names[0]));
-		if (!d->cached_names)
-			return -ENOMEM;
-
-		/* VOID is special */
-		d->type_states[0].order_state = ORDERED;
-		d->type_states[0].emit_state = EMITTED;
-
-		/* eagerly determine referenced types for anon enums */
-		err = btf_dump_mark_referenced(d);
-		if (err)
-			return err;
-	}
-
 	d->emit_queue_cnt = 0;
 	err = btf_dump_order_type(d, id, false);
 	if (err < 0)
@@ -1016,6 +1017,21 @@ static int btf_dump_push_decl_stack_id(struct btf_dump *d, __u32 id)
  * of a stack frame. Some care is required to "pop" stack frames after
  * processing type declaration chain.
  */
+int btf_dump__emit_type_decl(struct btf_dump *d, __u32 id,
+			     const struct btf_dump_emit_type_decl_opts *opts)
+{
+	const char *fname;
+	int lvl;
+
+	if (!OPTS_VALID(opts, btf_dump_emit_type_decl_opts))
+		return -EINVAL;
+
+	fname = OPTS_GET(opts, field_name, NULL);
+	lvl = OPTS_GET(opts, indent_level, 0);
+	btf_dump_emit_type_decl(d, id, fname, lvl);
+	return 0;
+}
+
 static void btf_dump_emit_type_decl(struct btf_dump *d, __u32 id,
 				    const char *fname, int lvl)
 {
diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
index e7fcca36f186..990c7c0e2d9f 100644
--- a/tools/lib/bpf/libbpf.map
+++ b/tools/lib/bpf/libbpf.map
@@ -211,6 +211,7 @@ LIBBPF_0.0.6 {
 
 LIBBPF_0.0.7 {
 	global:
+		btf_dump__emit_type_decl;
 		bpf_program__attach;
 		btf__align_of;
 } LIBBPF_0.0.6;
-- 
cgit v1.2.3-59-g8ed1b


From 01af3bf06755dc5cda7050fe5d898998e5775e63 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Fri, 13 Dec 2019 17:43:32 -0800
Subject: libbpf: Expose BPF program's function name

Add APIs to get BPF program function name, as opposed to bpf_program__title(),
which returns BPF program function's section name. Function name has a benefit
of being a valid C identifier and uniquely identifies a specific BPF program,
while section name can be duplicated across multiple independent BPF programs.

Add also bpf_object__find_program_by_name(), similar to
bpf_object__find_program_by_title(), to facilitate looking up BPF programs by
their C function names.

Convert one of selftests to new API for look up.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20191214014341.3442258-9-andriin@fb.com
---
 tools/lib/bpf/libbpf.c                             | 28 ++++++++++++++++++----
 tools/lib/bpf/libbpf.h                             |  9 +++++--
 tools/lib/bpf/libbpf.map                           |  2 ++
 .../testing/selftests/bpf/prog_tests/rdonly_maps.c | 11 ++++-----
 4 files changed, 36 insertions(+), 14 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 61b8cdf78332..e7a6b57d849c 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -209,8 +209,8 @@ static const char * const libbpf_type_to_btf_name[] = {
 };
 
 struct bpf_map {
-	int fd;
 	char *name;
+	int fd;
 	int sec_idx;
 	size_t sec_offset;
 	int map_ifindex;
@@ -1384,7 +1384,7 @@ static int bpf_object__init_user_btf_maps(struct bpf_object *obj, bool strict,
 }
 
 static int bpf_object__init_maps(struct bpf_object *obj,
-				 struct bpf_object_open_opts *opts)
+				 const struct bpf_object_open_opts *opts)
 {
 	const char *pin_root_path = OPTS_GET(opts, pin_root_path, NULL);
 	bool strict = !OPTS_GET(opts, relaxed_maps, false);
@@ -1748,6 +1748,19 @@ bpf_object__find_program_by_title(const struct bpf_object *obj,
 	return NULL;
 }
 
+struct bpf_program *
+bpf_object__find_program_by_name(const struct bpf_object *obj,
+				 const char *name)
+{
+	struct bpf_program *prog;
+
+	bpf_object__for_each_program(prog, obj) {
+		if (!strcmp(prog->name, name))
+			return prog;
+	}
+	return NULL;
+}
+
 static bool bpf_object__shndx_is_data(const struct bpf_object *obj,
 				      int shndx)
 {
@@ -3894,7 +3907,7 @@ static int libbpf_find_attach_btf_id(const char *name,
 				     __u32 attach_prog_fd);
 static struct bpf_object *
 __bpf_object__open(const char *path, const void *obj_buf, size_t obj_buf_sz,
-		   struct bpf_object_open_opts *opts)
+		   const struct bpf_object_open_opts *opts)
 {
 	struct bpf_program *prog;
 	struct bpf_object *obj;
@@ -4003,7 +4016,7 @@ struct bpf_object *bpf_object__open(const char *path)
 }
 
 struct bpf_object *
-bpf_object__open_file(const char *path, struct bpf_object_open_opts *opts)
+bpf_object__open_file(const char *path, const struct bpf_object_open_opts *opts)
 {
 	if (!path)
 		return ERR_PTR(-EINVAL);
@@ -4015,7 +4028,7 @@ bpf_object__open_file(const char *path, struct bpf_object_open_opts *opts)
 
 struct bpf_object *
 bpf_object__open_mem(const void *obj_buf, size_t obj_buf_sz,
-		     struct bpf_object_open_opts *opts)
+		     const struct bpf_object_open_opts *opts)
 {
 	if (!obj_buf || obj_buf_sz == 0)
 		return ERR_PTR(-EINVAL);
@@ -4820,6 +4833,11 @@ void bpf_program__set_ifindex(struct bpf_program *prog, __u32 ifindex)
 	prog->prog_ifindex = ifindex;
 }
 
+const char *bpf_program__name(const struct bpf_program *prog)
+{
+	return prog->name;
+}
+
 const char *bpf_program__title(const struct bpf_program *prog, bool needs_copy)
 {
 	const char *title;
diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index 49e6fa01024b..f37bd4a3e14b 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -90,10 +90,10 @@ struct bpf_object_open_opts {
 
 LIBBPF_API struct bpf_object *bpf_object__open(const char *path);
 LIBBPF_API struct bpf_object *
-bpf_object__open_file(const char *path, struct bpf_object_open_opts *opts);
+bpf_object__open_file(const char *path, const struct bpf_object_open_opts *opts);
 LIBBPF_API struct bpf_object *
 bpf_object__open_mem(const void *obj_buf, size_t obj_buf_sz,
-		     struct bpf_object_open_opts *opts);
+		     const struct bpf_object_open_opts *opts);
 
 /* deprecated bpf_object__open variants */
 LIBBPF_API struct bpf_object *
@@ -132,6 +132,7 @@ struct bpf_object_load_attr {
 LIBBPF_API int bpf_object__load(struct bpf_object *obj);
 LIBBPF_API int bpf_object__load_xattr(struct bpf_object_load_attr *attr);
 LIBBPF_API int bpf_object__unload(struct bpf_object *obj);
+
 LIBBPF_API const char *bpf_object__name(const struct bpf_object *obj);
 LIBBPF_API unsigned int bpf_object__kversion(const struct bpf_object *obj);
 
@@ -142,6 +143,9 @@ LIBBPF_API int bpf_object__btf_fd(const struct bpf_object *obj);
 LIBBPF_API struct bpf_program *
 bpf_object__find_program_by_title(const struct bpf_object *obj,
 				  const char *title);
+LIBBPF_API struct bpf_program *
+bpf_object__find_program_by_name(const struct bpf_object *obj,
+				 const char *name);
 
 LIBBPF_API struct bpf_object *bpf_object__next(struct bpf_object *prev);
 #define bpf_object__for_each_safe(pos, tmp)			\
@@ -185,6 +189,7 @@ LIBBPF_API void *bpf_program__priv(const struct bpf_program *prog);
 LIBBPF_API void bpf_program__set_ifindex(struct bpf_program *prog,
 					 __u32 ifindex);
 
+LIBBPF_API const char *bpf_program__name(const struct bpf_program *prog);
 LIBBPF_API const char *bpf_program__title(const struct bpf_program *prog,
 					  bool needs_copy);
 
diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
index 990c7c0e2d9f..5a7630748eeb 100644
--- a/tools/lib/bpf/libbpf.map
+++ b/tools/lib/bpf/libbpf.map
@@ -212,6 +212,8 @@ LIBBPF_0.0.6 {
 LIBBPF_0.0.7 {
 	global:
 		btf_dump__emit_type_decl;
+		bpf_object__find_program_by_name;
 		bpf_program__attach;
+		bpf_program__name;
 		btf__align_of;
 } LIBBPF_0.0.6;
diff --git a/tools/testing/selftests/bpf/prog_tests/rdonly_maps.c b/tools/testing/selftests/bpf/prog_tests/rdonly_maps.c
index d90acc13d1ec..563e12120e77 100644
--- a/tools/testing/selftests/bpf/prog_tests/rdonly_maps.c
+++ b/tools/testing/selftests/bpf/prog_tests/rdonly_maps.c
@@ -16,14 +16,11 @@ struct rdonly_map_subtest {
 
 void test_rdonly_maps(void)
 {
-	const char *prog_name_skip_loop = "raw_tracepoint/sys_enter:skip_loop";
-	const char *prog_name_part_loop = "raw_tracepoint/sys_enter:part_loop";
-	const char *prog_name_full_loop = "raw_tracepoint/sys_enter:full_loop";
 	const char *file = "test_rdonly_maps.o";
 	struct rdonly_map_subtest subtests[] = {
-		{ "skip loop", prog_name_skip_loop, 0, 0 },
-		{ "part loop", prog_name_part_loop, 3, 2 + 3 + 4 },
-		{ "full loop", prog_name_full_loop, 4, 2 + 3 + 4 + 5 },
+		{ "skip loop", "skip_loop", 0, 0 },
+		{ "part loop", "part_loop", 3, 2 + 3 + 4 },
+		{ "full loop", "full_loop", 4, 2 + 3 + 4 + 5 },
 	};
 	int i, err, zero = 0, duration = 0;
 	struct bpf_link *link = NULL;
@@ -50,7 +47,7 @@ void test_rdonly_maps(void)
 		if (!test__start_subtest(t->subtest_name))
 			continue;
 
-		prog = bpf_object__find_program_by_title(obj, t->prog_name);
+		prog = bpf_object__find_program_by_name(obj, t->prog_name);
 		if (CHECK(!prog, "find_prog", "prog '%s' not found\n",
 			  t->prog_name))
 			goto cleanup;
-- 
cgit v1.2.3-59-g8ed1b


From eba9c5f498a11a8558e099d14692c24f2fea27e3 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Fri, 13 Dec 2019 17:43:33 -0800
Subject: libbpf: Refactor global data map initialization

Refactor global data map initialization to use anonymous mmap()-ed memory
instead of malloc()-ed one. This allows to do a transparent re-mmap()-ing of
already existing memory address to point to BPF map's memory after
bpf_object__load() step (done in follow up patch). This choreographed setup
allows to have a nice and unsurprising way to pre-initialize read-only (and
r/w as well) maps by user and after BPF map creation keep working with
mmap()-ed contents of this map. All in a way that doesn't require user code to
update any pointers: the illusion of working with memory contents is preserved
before and after actual BPF map instantiation.

Selftests and runqslower example demonstrate this feature in follow up patches.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20191214014341.3442258-10-andriin@fb.com
---
 tools/lib/bpf/libbpf.c | 97 ++++++++++++++++++++++++++++++--------------------
 1 file changed, 59 insertions(+), 38 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index e7a6b57d849c..06dfa36ed0bb 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -221,16 +221,12 @@ struct bpf_map {
 	void *priv;
 	bpf_map_clear_priv_t clear_priv;
 	enum libbpf_map_type libbpf_type;
+	void *mmaped;
 	char *pin_path;
 	bool pinned;
 	bool reused;
 };
 
-struct bpf_secdata {
-	void *rodata;
-	void *data;
-};
-
 static LIST_HEAD(bpf_objects_list);
 
 struct bpf_object {
@@ -243,7 +239,6 @@ struct bpf_object {
 	struct bpf_map *maps;
 	size_t nr_maps;
 	size_t maps_cap;
-	struct bpf_secdata sections;
 
 	bool loaded;
 	bool has_pseudo_calls;
@@ -828,13 +823,24 @@ static struct bpf_map *bpf_object__add_map(struct bpf_object *obj)
 	return &obj->maps[obj->nr_maps++];
 }
 
+static size_t bpf_map_mmap_sz(const struct bpf_map *map)
+{
+	long page_sz = sysconf(_SC_PAGE_SIZE);
+	size_t map_sz;
+
+	map_sz = roundup(map->def.value_size, 8) * map->def.max_entries;
+	map_sz = roundup(map_sz, page_sz);
+	return map_sz;
+}
+
 static int
 bpf_object__init_internal_map(struct bpf_object *obj, enum libbpf_map_type type,
-			      int sec_idx, Elf_Data *data, void **data_buff)
+			      int sec_idx, void *data, size_t data_sz)
 {
 	char map_name[BPF_OBJ_NAME_LEN];
 	struct bpf_map_def *def;
 	struct bpf_map *map;
+	int err;
 
 	map = bpf_object__add_map(obj);
 	if (IS_ERR(map))
@@ -854,7 +860,7 @@ bpf_object__init_internal_map(struct bpf_object *obj, enum libbpf_map_type type,
 	def = &map->def;
 	def->type = BPF_MAP_TYPE_ARRAY;
 	def->key_size = sizeof(int);
-	def->value_size = data->d_size;
+	def->value_size = data_sz;
 	def->max_entries = 1;
 	def->map_flags = type == LIBBPF_MAP_RODATA ? BPF_F_RDONLY_PROG : 0;
 	def->map_flags |= BPF_F_MMAPABLE;
@@ -862,16 +868,20 @@ bpf_object__init_internal_map(struct bpf_object *obj, enum libbpf_map_type type,
 	pr_debug("map '%s' (global data): at sec_idx %d, offset %zu, flags %x.\n",
 		 map_name, map->sec_idx, map->sec_offset, def->map_flags);
 
-	if (data_buff) {
-		*data_buff = malloc(data->d_size);
-		if (!*data_buff) {
-			zfree(&map->name);
-			pr_warn("failed to alloc map content buffer\n");
-			return -ENOMEM;
-		}
-		memcpy(*data_buff, data->d_buf, data->d_size);
+	map->mmaped = mmap(NULL, bpf_map_mmap_sz(map), PROT_READ | PROT_WRITE,
+			   MAP_SHARED | MAP_ANONYMOUS, -1, 0);
+	if (map->mmaped == MAP_FAILED) {
+		err = -errno;
+		map->mmaped = NULL;
+		pr_warn("failed to alloc map '%s' content buffer: %d\n",
+			map->name, err);
+		zfree(&map->name);
+		return err;
 	}
 
+	if (type != LIBBPF_MAP_BSS)
+		memcpy(map->mmaped, data, data_sz);
+
 	pr_debug("map %td is \"%s\"\n", map - obj->maps, map->name);
 	return 0;
 }
@@ -886,23 +896,24 @@ static int bpf_object__init_global_data_maps(struct bpf_object *obj)
 	if (obj->efile.data_shndx >= 0) {
 		err = bpf_object__init_internal_map(obj, LIBBPF_MAP_DATA,
 						    obj->efile.data_shndx,
-						    obj->efile.data,
-						    &obj->sections.data);
+						    obj->efile.data->d_buf,
+						    obj->efile.data->d_size);
 		if (err)
 			return err;
 	}
 	if (obj->efile.rodata_shndx >= 0) {
 		err = bpf_object__init_internal_map(obj, LIBBPF_MAP_RODATA,
 						    obj->efile.rodata_shndx,
-						    obj->efile.rodata,
-						    &obj->sections.rodata);
+						    obj->efile.rodata->d_buf,
+						    obj->efile.rodata->d_size);
 		if (err)
 			return err;
 	}
 	if (obj->efile.bss_shndx >= 0) {
 		err = bpf_object__init_internal_map(obj, LIBBPF_MAP_BSS,
 						    obj->efile.bss_shndx,
-						    obj->efile.bss, NULL);
+						    NULL,
+						    obj->efile.bss->d_size);
 		if (err)
 			return err;
 	}
@@ -2292,27 +2303,32 @@ bpf_object__populate_internal_map(struct bpf_object *obj, struct bpf_map *map)
 {
 	char *cp, errmsg[STRERR_BUFSIZE];
 	int err, zero = 0;
-	__u8 *data;
 
 	/* Nothing to do here since kernel already zero-initializes .bss map. */
 	if (map->libbpf_type == LIBBPF_MAP_BSS)
 		return 0;
 
-	data = map->libbpf_type == LIBBPF_MAP_DATA ?
-	       obj->sections.data : obj->sections.rodata;
+	err = bpf_map_update_elem(map->fd, &zero, map->mmaped, 0);
+	if (err) {
+		err = -errno;
+		cp = libbpf_strerror_r(err, errmsg, sizeof(errmsg));
+		pr_warn("Error setting initial map(%s) contents: %s\n",
+			map->name, cp);
+		return err;
+	}
 
-	err = bpf_map_update_elem(map->fd, &zero, data, 0);
 	/* Freeze .rodata map as read-only from syscall side. */
-	if (!err && map->libbpf_type == LIBBPF_MAP_RODATA) {
+	if (map->libbpf_type == LIBBPF_MAP_RODATA) {
 		err = bpf_map_freeze(map->fd);
 		if (err) {
-			cp = libbpf_strerror_r(errno, errmsg, sizeof(errmsg));
+			err = -errno;
+			cp = libbpf_strerror_r(err, errmsg, sizeof(errmsg));
 			pr_warn("Error freezing map(%s) as read-only: %s\n",
 				map->name, cp);
-			err = 0;
+			return err;
 		}
 	}
-	return err;
+	return 0;
 }
 
 static int
@@ -4683,17 +4699,22 @@ void bpf_object__close(struct bpf_object *obj)
 	btf_ext__free(obj->btf_ext);
 
 	for (i = 0; i < obj->nr_maps; i++) {
-		zfree(&obj->maps[i].name);
-		zfree(&obj->maps[i].pin_path);
-		if (obj->maps[i].clear_priv)
-			obj->maps[i].clear_priv(&obj->maps[i],
-						obj->maps[i].priv);
-		obj->maps[i].priv = NULL;
-		obj->maps[i].clear_priv = NULL;
+		struct bpf_map *map = &obj->maps[i];
+
+		if (map->clear_priv)
+			map->clear_priv(map, map->priv);
+		map->priv = NULL;
+		map->clear_priv = NULL;
+
+		if (map->mmaped) {
+			munmap(map->mmaped, bpf_map_mmap_sz(map));
+			map->mmaped = NULL;
+		}
+
+		zfree(&map->name);
+		zfree(&map->pin_path);
 	}
 
-	zfree(&obj->sections.rodata);
-	zfree(&obj->sections.data);
 	zfree(&obj->maps);
 	obj->nr_maps = 0;
 
-- 
cgit v1.2.3-59-g8ed1b


From 13acb508ae203075134327551a6705e6e8f23d48 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Fri, 13 Dec 2019 17:43:34 -0800
Subject: libbpf: Postpone BTF ID finding for TRACING programs to load phase

Move BTF ID determination for BPF_PROG_TYPE_TRACING programs to a load phase.
Performing it at open step is inconvenient, because it prevents BPF skeleton
generation on older host kernel, which doesn't contain BTF_KIND_FUNCs
information in vmlinux BTF. This is a common set up, though, when, e.g.,
selftests are compiled on older host kernel, but the test program itself is
executed in qemu VM with bleeding edge kernel. Having this BTF searching
performed at load time allows to successfully use bpf_object__open() for
codegen and inspection of BPF object file.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20191214014341.3442258-11-andriin@fb.com
---
 tools/lib/bpf/libbpf.c | 37 ++++++++++++++++++-------------------
 1 file changed, 18 insertions(+), 19 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 06dfa36ed0bb..3488ac4f7015 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -3815,11 +3815,22 @@ out:
 	return ret;
 }
 
-int
-bpf_program__load(struct bpf_program *prog,
-		  char *license, __u32 kern_version)
+static int libbpf_find_attach_btf_id(const char *name,
+				     enum bpf_attach_type attach_type,
+				     __u32 attach_prog_fd);
+
+int bpf_program__load(struct bpf_program *prog, char *license, __u32 kern_ver)
 {
-	int err = 0, fd, i;
+	int err = 0, fd, i, btf_id;
+
+	if (prog->type == BPF_PROG_TYPE_TRACING) {
+		btf_id = libbpf_find_attach_btf_id(prog->section_name,
+						   prog->expected_attach_type,
+						   prog->attach_prog_fd);
+		if (btf_id <= 0)
+			return btf_id;
+		prog->attach_btf_id = btf_id;
+	}
 
 	if (prog->instances.nr < 0 || !prog->instances.fds) {
 		if (prog->preprocessor) {
@@ -3843,7 +3854,7 @@ bpf_program__load(struct bpf_program *prog,
 				prog->section_name, prog->instances.nr);
 		}
 		err = load_program(prog, prog->insns, prog->insns_cnt,
-				   license, kern_version, &fd);
+				   license, kern_ver, &fd);
 		if (!err)
 			prog->instances.fds[0] = fd;
 		goto out;
@@ -3872,9 +3883,7 @@ bpf_program__load(struct bpf_program *prog,
 		}
 
 		err = load_program(prog, result.new_insn_ptr,
-				   result.new_insn_cnt,
-				   license, kern_version, &fd);
-
+				   result.new_insn_cnt, license, kern_ver, &fd);
 		if (err) {
 			pr_warn("Loading the %dth instance of program '%s' failed\n",
 				i, prog->section_name);
@@ -3918,9 +3927,6 @@ bpf_object__load_progs(struct bpf_object *obj, int log_level)
 	return 0;
 }
 
-static int libbpf_find_attach_btf_id(const char *name,
-				     enum bpf_attach_type attach_type,
-				     __u32 attach_prog_fd);
 static struct bpf_object *
 __bpf_object__open(const char *path, const void *obj_buf, size_t obj_buf_sz,
 		   const struct bpf_object_open_opts *opts)
@@ -3984,15 +3990,8 @@ __bpf_object__open(const char *path, const void *obj_buf, size_t obj_buf_sz,
 
 		bpf_program__set_type(prog, prog_type);
 		bpf_program__set_expected_attach_type(prog, attach_type);
-		if (prog_type == BPF_PROG_TYPE_TRACING) {
-			err = libbpf_find_attach_btf_id(prog->section_name,
-							attach_type,
-							attach_prog_fd);
-			if (err <= 0)
-				goto out;
-			prog->attach_btf_id = err;
+		if (prog_type == BPF_PROG_TYPE_TRACING)
 			prog->attach_prog_fd = attach_prog_fd;
-		}
 	}
 
 	return obj;
-- 
cgit v1.2.3-59-g8ed1b


From 3f51935314b8c0e0f45f28bed2e7a023b2c7627c Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Fri, 13 Dec 2019 17:43:35 -0800
Subject: libbpf: Reduce log level of supported section names dump

It's quite spammy. And now that bpf_object__open() is trying to determine
program type from its section name, we are getting these verbose messages all
the time. Reduce their log level to DEBUG.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20191214014341.3442258-12-andriin@fb.com
---
 tools/lib/bpf/libbpf.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 3488ac4f7015..b6bd6c47c919 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -5196,7 +5196,7 @@ int libbpf_prog_type_by_name(const char *name, enum bpf_prog_type *prog_type,
 	pr_warn("failed to guess program type from ELF section '%s'\n", name);
 	type_names = libbpf_get_type_names(false);
 	if (type_names != NULL) {
-		pr_info("supported section(type) names are:%s\n", type_names);
+		pr_debug("supported section(type) names are:%s\n", type_names);
 		free(type_names);
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From d66562fba1ce66975bd61b0786fb8b1810f33caa Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Fri, 13 Dec 2019 17:43:36 -0800
Subject: libbpf: Add BPF object skeleton support

Add new set of APIs, allowing to open/load/attach BPF object through BPF
object skeleton, generated by bpftool for a specific BPF object file. All the
xxx_skeleton() APIs wrap up corresponding bpf_object_xxx() APIs, but
additionally also automate map/program lookups by name, global data
initialization and mmap()-ing, etc.  All this greatly improves and simplifies
userspace usability of working with BPF programs. See follow up patches for
examples.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20191214014341.3442258-13-andriin@fb.com
---
 tools/lib/bpf/libbpf.c   | 162 +++++++++++++++++++++++++++++++++++++++++++++++
 tools/lib/bpf/libbpf.h   |  38 +++++++++++
 tools/lib/bpf/libbpf.map |   5 ++
 3 files changed, 205 insertions(+)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index b6bd6c47c919..a1a902fa6e0c 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -6793,3 +6793,165 @@ int libbpf_num_possible_cpus(void)
 	WRITE_ONCE(cpus, tmp_cpus);
 	return tmp_cpus;
 }
+
+int bpf_object__open_skeleton(struct bpf_object_skeleton *s,
+			      const struct bpf_object_open_opts *opts)
+{
+	DECLARE_LIBBPF_OPTS(bpf_object_open_opts, skel_opts,
+		.object_name = s->name,
+	);
+	struct bpf_object *obj;
+	int i;
+
+	/* Attempt to preserve opts->object_name, unless overriden by user
+	 * explicitly. Overwriting object name for skeletons is discouraged,
+	 * as it breaks global data maps, because they contain object name
+	 * prefix as their own map name prefix. When skeleton is generated,
+	 * bpftool is making an assumption that this name will stay the same.
+	 */
+	if (opts) {
+		memcpy(&skel_opts, opts, sizeof(*opts));
+		if (!opts->object_name)
+			skel_opts.object_name = s->name;
+	}
+
+	obj = bpf_object__open_mem(s->data, s->data_sz, &skel_opts);
+	if (IS_ERR(obj)) {
+		pr_warn("failed to initialize skeleton BPF object '%s': %ld\n",
+			s->name, PTR_ERR(obj));
+		return PTR_ERR(obj);
+	}
+
+	*s->obj = obj;
+
+	for (i = 0; i < s->map_cnt; i++) {
+		struct bpf_map **map = s->maps[i].map;
+		const char *name = s->maps[i].name;
+		void **mmaped = s->maps[i].mmaped;
+
+		*map = bpf_object__find_map_by_name(obj, name);
+		if (!*map) {
+			pr_warn("failed to find skeleton map '%s'\n", name);
+			return -ESRCH;
+		}
+
+		if (mmaped)
+			*mmaped = (*map)->mmaped;
+	}
+
+	for (i = 0; i < s->prog_cnt; i++) {
+		struct bpf_program **prog = s->progs[i].prog;
+		const char *name = s->progs[i].name;
+
+		*prog = bpf_object__find_program_by_name(obj, name);
+		if (!*prog) {
+			pr_warn("failed to find skeleton program '%s'\n", name);
+			return -ESRCH;
+		}
+	}
+
+	return 0;
+}
+
+int bpf_object__load_skeleton(struct bpf_object_skeleton *s)
+{
+	int i, err;
+
+	err = bpf_object__load(*s->obj);
+	if (err) {
+		pr_warn("failed to load BPF skeleton '%s': %d\n", s->name, err);
+		return err;
+	}
+
+	for (i = 0; i < s->map_cnt; i++) {
+		struct bpf_map *map = *s->maps[i].map;
+		size_t mmap_sz = bpf_map_mmap_sz(map);
+		int prot, map_fd = bpf_map__fd(map);
+		void **mmaped = s->maps[i].mmaped;
+		void *remapped;
+
+		if (!mmaped)
+			continue;
+
+		if (!(map->def.map_flags & BPF_F_MMAPABLE)) {
+			*mmaped = NULL;
+			continue;
+		}
+
+		if (map->def.map_flags & BPF_F_RDONLY_PROG)
+			prot = PROT_READ;
+		else
+			prot = PROT_READ | PROT_WRITE;
+
+		/* Remap anonymous mmap()-ed "map initialization image" as
+		 * a BPF map-backed mmap()-ed memory, but preserving the same
+		 * memory address. This will cause kernel to change process'
+		 * page table to point to a different piece of kernel memory,
+		 * but from userspace point of view memory address (and its
+		 * contents, being identical at this point) will stay the
+		 * same. This mapping will be released by bpf_object__close()
+		 * as per normal clean up procedure, so we don't need to worry
+		 * about it from skeleton's clean up perspective.
+		 */
+		remapped = mmap(*mmaped, mmap_sz, prot, MAP_SHARED | MAP_FIXED,
+				map_fd, 0);
+		if (remapped == MAP_FAILED) {
+			err = -errno;
+			*mmaped = NULL;
+			pr_warn("failed to re-mmap() map '%s': %d\n",
+				 bpf_map__name(map), err);
+			return err;
+		}
+	}
+
+	return 0;
+}
+
+int bpf_object__attach_skeleton(struct bpf_object_skeleton *s)
+{
+	int i;
+
+	for (i = 0; i < s->prog_cnt; i++) {
+		struct bpf_program *prog = *s->progs[i].prog;
+		struct bpf_link **link = s->progs[i].link;
+		const struct bpf_sec_def *sec_def;
+		const char *sec_name = bpf_program__title(prog, false);
+
+		sec_def = find_sec_def(sec_name);
+		if (!sec_def || !sec_def->attach_fn)
+			continue;
+
+		*link = sec_def->attach_fn(sec_def, prog);
+		if (IS_ERR(*link)) {
+			pr_warn("failed to auto-attach program '%s': %ld\n",
+				bpf_program__name(prog), PTR_ERR(*link));
+			return PTR_ERR(*link);
+		}
+	}
+
+	return 0;
+}
+
+void bpf_object__detach_skeleton(struct bpf_object_skeleton *s)
+{
+	int i;
+
+	for (i = 0; i < s->prog_cnt; i++) {
+		struct bpf_link **link = s->progs[i].link;
+
+		if (!IS_ERR_OR_NULL(*link))
+			bpf_link__destroy(*link);
+		*link = NULL;
+	}
+}
+
+void bpf_object__destroy_skeleton(struct bpf_object_skeleton *s)
+{
+	if (s->progs)
+		bpf_object__detach_skeleton(s);
+	if (s->obj)
+		bpf_object__close(*s->obj);
+	free(s->maps);
+	free(s->progs);
+	free(s);
+}
diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index f37bd4a3e14b..623191e71415 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -631,6 +631,44 @@ BPF_EMBED_OBJ_DECLARE(NAME)
 #define BPF_EMBED_OBJ(NAME, PATH) __BPF_EMBED_OBJ(NAME, PATH, 16, ".quad")
 #endif
 
+struct bpf_map_skeleton {
+	const char *name;
+	struct bpf_map **map;
+	void **mmaped;
+};
+
+struct bpf_prog_skeleton {
+	const char *name;
+	struct bpf_program **prog;
+	struct bpf_link **link;
+};
+
+struct bpf_object_skeleton {
+	size_t sz; /* size of this struct, for forward/backward compatibility */
+
+	const char *name;
+	void *data;
+	size_t data_sz;
+
+	struct bpf_object **obj;
+
+	int map_cnt;
+	int map_skel_sz; /* sizeof(struct bpf_skeleton_map) */
+	struct bpf_map_skeleton *maps;
+
+	int prog_cnt;
+	int prog_skel_sz; /* sizeof(struct bpf_skeleton_prog) */
+	struct bpf_prog_skeleton *progs;
+};
+
+LIBBPF_API int
+bpf_object__open_skeleton(struct bpf_object_skeleton *s,
+			  const struct bpf_object_open_opts *opts);
+LIBBPF_API int bpf_object__load_skeleton(struct bpf_object_skeleton *s);
+LIBBPF_API int bpf_object__attach_skeleton(struct bpf_object_skeleton *s);
+LIBBPF_API void bpf_object__detach_skeleton(struct bpf_object_skeleton *s);
+LIBBPF_API void bpf_object__destroy_skeleton(struct bpf_object_skeleton *s);
+
 #ifdef __cplusplus
 } /* extern "C" */
 #endif
diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
index 5a7630748eeb..1376d992a703 100644
--- a/tools/lib/bpf/libbpf.map
+++ b/tools/lib/bpf/libbpf.map
@@ -213,6 +213,11 @@ LIBBPF_0.0.7 {
 	global:
 		btf_dump__emit_type_decl;
 		bpf_object__find_program_by_name;
+		bpf_object__attach_skeleton;
+		bpf_object__destroy_skeleton;
+		bpf_object__detach_skeleton;
+		bpf_object__load_skeleton;
+		bpf_object__open_skeleton;
 		bpf_program__attach;
 		bpf_program__name;
 		btf__align_of;
-- 
cgit v1.2.3-59-g8ed1b


From 985ead416df39d6fe8e89580cc1db6aa273e0175 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Fri, 13 Dec 2019 17:43:37 -0800
Subject: bpftool: Add skeleton codegen command

Add `bpftool gen skeleton` command, which takes in compiled BPF .o object file
and dumps a BPF skeleton struct and related code to work with that skeleton.
Skeleton itself is tailored to a specific structure of provided BPF object
file, containing accessors (just plain struct fields) for every map and
program, as well as dedicated space for bpf_links. If BPF program is using
global variables, corresponding structure definitions of compatible memory
layout are emitted as well, making it possible to initialize and subsequently
read/update global variables values using simple and clear C syntax for
accessing fields. This skeleton majorly improves usability of
opening/loading/attaching of BPF object, as well as interacting with it
throughout the lifetime of loaded BPF object.

Generated skeleton struct has the following structure:

struct <object-name> {
	/* used by libbpf's skeleton API */
	struct bpf_object_skeleton *skeleton;
	/* bpf_object for libbpf APIs */
	struct bpf_object *obj;
	struct {
		/* for every defined map in BPF object: */
		struct bpf_map *<map-name>;
	} maps;
	struct {
		/* for every program in BPF object: */
		struct bpf_program *<program-name>;
	} progs;
	struct {
		/* for every program in BPF object: */
		struct bpf_link *<program-name>;
	} links;
	/* for every present global data section: */
	struct <object-name>__<one of bss, data, or rodata> {
		/* memory layout of corresponding data section,
		 * with every defined variable represented as a struct field
		 * with exactly the same type, but without const/volatile
		 * modifiers, e.g.:
		 */
		 int *my_var_1;
		 ...
	} *<one of bss, data, or rodata>;
};

This provides great usability improvements:
- no need to look up maps and programs by name, instead just
  my_obj->maps.my_map or my_obj->progs.my_prog would give necessary
  bpf_map/bpf_program pointers, which user can pass to existing libbpf APIs;
- pre-defined places for bpf_links, which will be automatically populated for
  program types that libbpf knows how to attach automatically (currently
  tracepoints, kprobe/kretprobe, raw tracepoint and tracing programs). On
  tearing down skeleton, all active bpf_links will be destroyed (meaning BPF
  programs will be detached, if they are attached). For cases in which libbpf
  doesn't know how to auto-attach BPF program, user can manually create link
  after loading skeleton and they will be auto-detached on skeleton
  destruction:

	my_obj->links.my_fancy_prog = bpf_program__attach_cgroup_whatever(
		my_obj->progs.my_fancy_prog, <whatever extra param);

- it's extremely easy and convenient to work with global data from userspace
  now. Both for read-only and read/write variables, it's possible to
  pre-initialize them before skeleton is loaded:

	skel = my_obj__open(raw_embed_data);
	my_obj->rodata->my_var = 123;
	my_obj__load(skel); /* 123 will be initialization value for my_var */

  After load, if kernel supports mmap() for BPF arrays, user can still read
  (and write for .bss and .data) variables values, but at that point it will
  be directly mmap()-ed to BPF array, backing global variables. This allows to
  seamlessly exchange data with BPF side. From userspace program's POV, all
  the pointers and memory contents stay the same, but mapped kernel memory
  changes to point to created map.
  If kernel doesn't yet support mmap() for BPF arrays, it's still possible to
  use those data section structs to pre-initialize .bss, .data, and .rodata,
  but after load their pointers will be reset to NULL, allowing user code to
  gracefully handle this condition, if necessary.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20191214014341.3442258-14-andriin@fb.com
---
 tools/bpf/bpftool/gen.c  | 551 +++++++++++++++++++++++++++++++++++++++++++++++
 tools/bpf/bpftool/main.c |   3 +-
 tools/bpf/bpftool/main.h |   1 +
 3 files changed, 554 insertions(+), 1 deletion(-)
 create mode 100644 tools/bpf/bpftool/gen.c

diff --git a/tools/bpf/bpftool/gen.c b/tools/bpf/bpftool/gen.c
new file mode 100644
index 000000000000..7379dae35dca
--- /dev/null
+++ b/tools/bpf/bpftool/gen.c
@@ -0,0 +1,551 @@
+// SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+/* Copyright (C) 2019 Facebook */
+
+#ifndef _GNU_SOURCE
+#define _GNU_SOURCE
+#endif
+#include <ctype.h>
+#include <errno.h>
+#include <fcntl.h>
+#include <linux/err.h>
+#include <stdbool.h>
+#include <stdio.h>
+#include <string.h>
+#include <unistd.h>
+#include <bpf.h>
+#include <libbpf.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <unistd.h>
+
+#include "btf.h"
+#include "libbpf_internal.h"
+#include "json_writer.h"
+#include "main.h"
+
+
+#define MAX_OBJ_NAME_LEN 64
+
+static void sanitize_identifier(char *name)
+{
+	int i;
+
+	for (i = 0; name[i]; i++)
+		if (!isalnum(name[i]) && name[i] != '_')
+			name[i] = '_';
+}
+
+static bool str_has_suffix(const char *str, const char *suffix)
+{
+	size_t i, n1 = strlen(str), n2 = strlen(suffix);
+
+	if (n1 < n2)
+		return false;
+
+	for (i = 0; i < n2; i++) {
+		if (str[n1 - i - 1] != suffix[n2 - i - 1])
+			return false;
+	}
+
+	return true;
+}
+
+static void get_obj_name(char *name, const char *file)
+{
+	/* Using basename() GNU version which doesn't modify arg. */
+	strncpy(name, basename(file), MAX_OBJ_NAME_LEN - 1);
+	name[MAX_OBJ_NAME_LEN - 1] = '\0';
+	if (str_has_suffix(name, ".o"))
+		name[strlen(name) - 2] = '\0';
+	sanitize_identifier(name);
+}
+
+static void get_header_guard(char *guard, const char *obj_name)
+{
+	int i;
+
+	sprintf(guard, "__%s_SKEL_H__", obj_name);
+	for (i = 0; guard[i]; i++)
+		guard[i] = toupper(guard[i]);
+}
+
+static const char *get_map_ident(const struct bpf_map *map)
+{
+	const char *name = bpf_map__name(map);
+
+	if (!bpf_map__is_internal(map))
+		return name;
+
+	if (str_has_suffix(name, ".data"))
+		return "data";
+	else if (str_has_suffix(name, ".rodata"))
+		return "rodata";
+	else if (str_has_suffix(name, ".bss"))
+		return "bss";
+	else
+		return NULL;
+}
+
+static void codegen_btf_dump_printf(void *ct, const char *fmt, va_list args)
+{
+	vprintf(fmt, args);
+}
+
+static int codegen_datasec_def(struct bpf_object *obj,
+			       struct btf *btf,
+			       struct btf_dump *d,
+			       const struct btf_type *sec,
+			       const char *obj_name)
+{
+	const char *sec_name = btf__name_by_offset(btf, sec->name_off);
+	const struct btf_var_secinfo *sec_var = btf_var_secinfos(sec);
+	int i, err, off = 0, pad_cnt = 0, vlen = btf_vlen(sec);
+	const char *sec_ident;
+	char var_ident[256];
+
+	if (strcmp(sec_name, ".data") == 0)
+		sec_ident = "data";
+	else if (strcmp(sec_name, ".bss") == 0)
+		sec_ident = "bss";
+	else if (strcmp(sec_name, ".rodata") == 0)
+		sec_ident = "rodata";
+	else
+		return 0;
+
+	printf("	struct %s__%s {\n", obj_name, sec_ident);
+	for (i = 0; i < vlen; i++, sec_var++) {
+		const struct btf_type *var = btf__type_by_id(btf, sec_var->type);
+		const char *var_name = btf__name_by_offset(btf, var->name_off);
+		DECLARE_LIBBPF_OPTS(btf_dump_emit_type_decl_opts, opts,
+			.field_name = var_ident,
+			.indent_level = 2,
+		);
+		int need_off = sec_var->offset, align_off, align;
+		__u32 var_type_id = var->type;
+		const struct btf_type *t;
+
+		t = btf__type_by_id(btf, var_type_id);
+		while (btf_is_mod(t)) {
+			var_type_id = t->type;
+			t = btf__type_by_id(btf, var_type_id);
+		}
+
+		if (off > need_off) {
+			p_err("Something is wrong for %s's variable #%d: need offset %d, already at %d.\n",
+			      sec_name, i, need_off, off);
+			return -EINVAL;
+		}
+
+		align = btf__align_of(btf, var->type);
+		if (align <= 0) {
+			p_err("Failed to determine alignment of variable '%s': %d",
+			      var_name, align);
+			return -EINVAL;
+		}
+
+		align_off = (off + align - 1) / align * align;
+		if (align_off != need_off) {
+			printf("\t\tchar __pad%d[%d];\n",
+			       pad_cnt, need_off - off);
+			pad_cnt++;
+		}
+
+		/* sanitize variable name, e.g., for static vars inside
+		 * a function, it's name is '<function name>.<variable name>',
+		 * which we'll turn into a '<function name>_<variable name>'
+		 */
+		var_ident[0] = '\0';
+		strncat(var_ident, var_name, sizeof(var_ident) - 1);
+		sanitize_identifier(var_ident);
+
+		printf("\t\t");
+		err = btf_dump__emit_type_decl(d, var_type_id, &opts);
+		if (err)
+			return err;
+		printf(";\n");
+
+		off = sec_var->offset + sec_var->size;
+	}
+	printf("	} *%s;\n", sec_ident);
+	return 0;
+}
+
+static int codegen_datasecs(struct bpf_object *obj, const char *obj_name)
+{
+	struct btf *btf = bpf_object__btf(obj);
+	int n = btf__get_nr_types(btf);
+	struct btf_dump *d;
+	int i, err = 0;
+
+	d = btf_dump__new(btf, NULL, NULL, codegen_btf_dump_printf);
+	if (IS_ERR(d))
+		return PTR_ERR(d);
+
+	for (i = 1; i <= n; i++) {
+		const struct btf_type *t = btf__type_by_id(btf, i);
+
+		if (!btf_is_datasec(t))
+			continue;
+
+		err = codegen_datasec_def(obj, btf, d, t, obj_name);
+		if (err)
+			goto out;
+	}
+out:
+	btf_dump__free(d);
+	return err;
+}
+
+static int codegen(const char *template, ...)
+{
+	const char *src, *end;
+	int skip_tabs = 0, n;
+	char *s, *dst;
+	va_list args;
+	char c;
+
+	n = strlen(template);
+	s = malloc(n + 1);
+	if (!s)
+		return -ENOMEM;
+	src = template;
+	dst = s;
+
+	/* find out "baseline" indentation to skip */
+	while ((c = *src++)) {
+		if (c == '\t') {
+			skip_tabs++;
+		} else if (c == '\n') {
+			break;
+		} else {
+			p_err("unrecognized character at pos %td in template '%s'",
+			      src - template - 1, template);
+			return -EINVAL;
+		}
+	}
+
+	while (*src) {
+		/* skip baseline indentation tabs */
+		for (n = skip_tabs; n > 0; n--, src++) {
+			if (*src != '\t') {
+				p_err("not enough tabs at pos %td in template '%s'",
+				      src - template - 1, template);
+				return -EINVAL;
+			}
+		}
+		/* trim trailing whitespace */
+		end = strchrnul(src, '\n');
+		for (n = end - src; n > 0 && isspace(src[n - 1]); n--)
+			;
+		memcpy(dst, src, n);
+		dst += n;
+		if (*end)
+			*dst++ = '\n';
+		src = *end ? end + 1 : end;
+	}
+	*dst++ = '\0';
+
+	/* print out using adjusted template */
+	va_start(args, template);
+	n = vprintf(s, args);
+	va_end(args);
+
+	free(s);
+	return n;
+}
+
+static int do_skeleton(int argc, char **argv)
+{
+	char header_guard[MAX_OBJ_NAME_LEN + sizeof("__SKEL_H__")];
+	size_t i, map_cnt = 0, prog_cnt = 0;
+	char obj_name[MAX_OBJ_NAME_LEN];
+	const char *file, *ident;
+	struct bpf_program *prog;
+	struct bpf_object *obj;
+	struct bpf_map *map;
+	struct btf *btf;
+	int err = -1;
+
+	if (!REQ_ARGS(1)) {
+		usage();
+		return -1;
+	}
+	file = GET_ARG();
+
+	if (argc) {
+		p_err("extra unknown arguments");
+		return -1;
+	}
+
+	obj = bpf_object__open_file(file, NULL);
+	if (IS_ERR(obj)) {
+		p_err("failed to open BPF object file: %ld", PTR_ERR(obj));
+		return -1;
+	}
+
+	get_obj_name(obj_name, file);
+	get_header_guard(header_guard, obj_name);
+
+	bpf_object__for_each_map(map, obj) {
+		ident = get_map_ident(map);
+		if (!ident) {
+			p_err("ignoring unrecognized internal map '%s'...",
+			      bpf_map__name(map));
+			continue;
+		}
+		map_cnt++;
+	}
+	bpf_object__for_each_program(prog, obj) {
+		prog_cnt++;
+	}
+
+	codegen("\
+		\n\
+		/* THIS FILE IS AUTOGENERATED! */			    \n\
+		#ifndef %2$s						    \n\
+		#define %2$s						    \n\
+									    \n\
+		#include <stdlib.h>					    \n\
+		#include <libbpf.h>					    \n\
+									    \n\
+		struct %1$s {						    \n\
+			struct bpf_object_skeleton *skeleton;		    \n\
+			struct bpf_object *obj;				    \n\
+		",
+		obj_name, header_guard
+	);
+
+	if (map_cnt) {
+		printf("\tstruct {\n");
+		bpf_object__for_each_map(map, obj) {
+			ident = get_map_ident(map);
+			if (!ident)
+				continue;
+			printf("\t\tstruct bpf_map *%s;\n", ident);
+		}
+		printf("\t} maps;\n");
+	}
+
+	if (prog_cnt) {
+		printf("\tstruct {\n");
+		bpf_object__for_each_program(prog, obj) {
+			printf("\t\tstruct bpf_program *%s;\n",
+			       bpf_program__name(prog));
+		}
+		printf("\t} progs;\n");
+		printf("\tstruct {\n");
+		bpf_object__for_each_program(prog, obj) {
+			printf("\t\tstruct bpf_link *%s;\n",
+			       bpf_program__name(prog));
+		}
+		printf("\t} links;\n");
+	}
+
+	btf = bpf_object__btf(obj);
+	if (btf) {
+		err = codegen_datasecs(obj, obj_name);
+		if (err)
+			goto out;
+	}
+
+	codegen("\
+		\n\
+		};							    \n\
+									    \n\
+		static inline struct bpf_object_skeleton *		    \n\
+		%1$s__create_skeleton(struct %1$s *obj, struct bpf_embed_data *embed)\n\
+		{							    \n\
+			struct bpf_object_skeleton *s;			    \n\
+									    \n\
+			s = calloc(1, sizeof(*s));			    \n\
+			if (!s)						    \n\
+				return NULL;				    \n\
+									    \n\
+			s->sz = sizeof(*s);				    \n\
+			s->name = \"%1$s\";				    \n\
+			s->data = embed->data;				    \n\
+			s->data_sz = embed->size;			    \n\
+			s->obj = &obj->obj;				    \n\
+		",
+		obj_name
+	);
+	if (map_cnt) {
+		codegen("\
+			\n\
+									    \n\
+				/* maps */				    \n\
+				s->map_cnt = %zu;			    \n\
+				s->map_skel_sz = sizeof(*s->maps);	    \n\
+				s->maps = calloc(s->map_cnt, s->map_skel_sz);\n\
+				if (!s->maps)				    \n\
+					goto err;			    \n\
+			",
+			map_cnt
+		);
+		i = 0;
+		bpf_object__for_each_map(map, obj) {
+			const char *ident = get_map_ident(map);
+
+			if (!ident)
+				continue;
+
+			codegen("\
+				\n\
+									    \n\
+					s->maps[%zu].name = \"%s\";	    \n\
+					s->maps[%zu].map = &obj->maps.%s;   \n\
+				",
+				i, bpf_map__name(map), i, ident);
+			/* memory-mapped internal maps */
+			if (bpf_map__is_internal(map) &&
+			    (bpf_map__def(map)->map_flags & BPF_F_MMAPABLE)) {
+				printf("\ts->maps[%zu].mmaped = (void **)&obj->%s;\n",
+				       i, ident);
+			}
+			i++;
+		}
+	}
+	if (prog_cnt) {
+		codegen("\
+			\n\
+									    \n\
+				/* programs */				    \n\
+				s->prog_cnt = %zu;			    \n\
+				s->prog_skel_sz = sizeof(*s->progs);	    \n\
+				s->progs = calloc(s->prog_cnt, s->prog_skel_sz);\n\
+				if (!s->progs)				    \n\
+					goto err;			    \n\
+			",
+			prog_cnt
+		);
+		i = 0;
+		bpf_object__for_each_program(prog, obj) {
+			codegen("\
+				\n\
+									    \n\
+					s->progs[%1$zu].name = \"%2$s\";    \n\
+					s->progs[%1$zu].prog = &obj->progs.%2$s;\n\
+					s->progs[%1$zu].link = &obj->links.%2$s;\n\
+				",
+				i, bpf_program__name(prog));
+			i++;
+		}
+	}
+	codegen("\
+		\n\
+									    \n\
+			return s;					    \n\
+		err:							    \n\
+			bpf_object__destroy_skeleton(s);		    \n\
+			return NULL;					    \n\
+		}							    \n\
+									    \n\
+		static void						    \n\
+		%1$s__destroy(struct %1$s *obj)				    \n\
+		{							    \n\
+			if (!obj)					    \n\
+				return;					    \n\
+			if (obj->skeleton)				    \n\
+				bpf_object__destroy_skeleton(obj->skeleton);\n\
+			free(obj);					    \n\
+		}							    \n\
+									    \n\
+		static inline struct %1$s *				    \n\
+		%1$s__open_opts(struct bpf_embed_data *embed, const struct bpf_object_open_opts *opts)\n\
+		{							    \n\
+			struct %1$s *obj;				    \n\
+									    \n\
+			obj = calloc(1, sizeof(*obj));			    \n\
+			if (!obj)					    \n\
+				return NULL;				    \n\
+									    \n\
+			obj->skeleton = %1$s__create_skeleton(obj, embed);  \n\
+			if (!obj->skeleton)				    \n\
+				goto err;				    \n\
+									    \n\
+			if (bpf_object__open_skeleton(obj->skeleton, opts)) \n\
+				goto err;				    \n\
+									    \n\
+			return obj;					    \n\
+		err:							    \n\
+			%1$s__destroy(obj);				    \n\
+			return NULL;					    \n\
+		}							    \n\
+									    \n\
+		static inline struct %1$s *				    \n\
+		%1$s__open(struct bpf_embed_data *embed)		    \n\
+		{							    \n\
+			return %1$s__open_opts(embed, NULL);		    \n\
+		}							    \n\
+									    \n\
+		static inline int					    \n\
+		%1$s__load(struct %1$s *obj)				    \n\
+		{							    \n\
+			return bpf_object__load_skeleton(obj->skeleton);    \n\
+		}							    \n\
+									    \n\
+		static inline struct %1$s *				    \n\
+		%1$s__open_and_load(struct bpf_embed_data *embed)	    \n\
+		{							    \n\
+			struct %1$s *obj;				    \n\
+									    \n\
+			obj = %1$s__open(embed);			    \n\
+			if (!obj)					    \n\
+				return NULL;				    \n\
+			if (%1$s__load(obj)) {				    \n\
+				%1$s__destroy(obj);			    \n\
+				return NULL;				    \n\
+			}						    \n\
+			return obj;					    \n\
+		}							    \n\
+									    \n\
+		static inline int					    \n\
+		%1$s__attach(struct %1$s *obj)				    \n\
+		{							    \n\
+			return bpf_object__attach_skeleton(obj->skeleton);  \n\
+		}							    \n\
+									    \n\
+		static inline void					    \n\
+		%1$s__detach(struct %1$s *obj)				    \n\
+		{							    \n\
+			return bpf_object__detach_skeleton(obj->skeleton);  \n\
+		}							    \n\
+									    \n\
+		#endif /* %2$s */					    \n\
+		",
+		obj_name, header_guard
+	);
+	err = 0;
+out:
+	bpf_object__close(obj);
+	return err;
+}
+
+static int do_help(int argc, char **argv)
+{
+	if (json_output) {
+		jsonw_null(json_wtr);
+		return 0;
+	}
+
+	fprintf(stderr,
+		"Usage: %1$s gen skeleton FILE\n"
+		"       %1$s gen help\n"
+		"\n"
+		"       " HELP_SPEC_OPTIONS "\n"
+		"",
+		bin_name);
+
+	return 0;
+}
+
+static const struct cmd cmds[] = {
+	{ "skeleton",	do_skeleton },
+	{ "help",	do_help },
+	{ 0 }
+};
+
+int do_gen(int argc, char **argv)
+{
+	return cmd_select(cmds, argc, argv, do_help);
+}
diff --git a/tools/bpf/bpftool/main.c b/tools/bpf/bpftool/main.c
index 4764581ff9ea..1fe91c558508 100644
--- a/tools/bpf/bpftool/main.c
+++ b/tools/bpf/bpftool/main.c
@@ -58,7 +58,7 @@ static int do_help(int argc, char **argv)
 		"       %s batch file FILE\n"
 		"       %s version\n"
 		"\n"
-		"       OBJECT := { prog | map | cgroup | perf | net | feature | btf }\n"
+		"       OBJECT := { prog | map | cgroup | perf | net | feature | btf | gen }\n"
 		"       " HELP_SPEC_OPTIONS "\n"
 		"",
 		bin_name, bin_name, bin_name);
@@ -227,6 +227,7 @@ static const struct cmd cmds[] = {
 	{ "net",	do_net },
 	{ "feature",	do_feature },
 	{ "btf",	do_btf },
+	{ "gen",	do_gen },
 	{ "version",	do_version },
 	{ 0 }
 };
diff --git a/tools/bpf/bpftool/main.h b/tools/bpf/bpftool/main.h
index 81890f4c8cca..4e75b58d3989 100644
--- a/tools/bpf/bpftool/main.h
+++ b/tools/bpf/bpftool/main.h
@@ -155,6 +155,7 @@ int do_net(int argc, char **arg);
 int do_tracelog(int argc, char **arg);
 int do_feature(int argc, char **argv);
 int do_btf(int argc, char **argv);
+int do_gen(int argc, char **argv);
 
 int parse_u32_arg(int *argc, char ***argv, __u32 *val, const char *what);
 int prog_parse_fd(int *argc, char ***argv);
-- 
cgit v1.2.3-59-g8ed1b


From f3c926a4df2cddf6230c3f56b1f43e439552cdad Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Fri, 13 Dec 2019 17:43:38 -0800
Subject: selftests/bpf: Add BPF skeletons selftests and convert attach_probe.c

Add BPF skeleton generation to selftest/bpf's Makefile. Convert attach_probe.c
to use skeleton.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20191214014341.3442258-15-andriin@fb.com
---
 tools/testing/selftests/bpf/.gitignore             |   2 +
 tools/testing/selftests/bpf/Makefile               |  36 ++++--
 .../selftests/bpf/prog_tests/attach_probe.c        | 135 +++++----------------
 .../selftests/bpf/progs/test_attach_probe.c        |  34 ++----
 4 files changed, 74 insertions(+), 133 deletions(-)

diff --git a/tools/testing/selftests/bpf/.gitignore b/tools/testing/selftests/bpf/.gitignore
index 419652458da4..ce5af95ede42 100644
--- a/tools/testing/selftests/bpf/.gitignore
+++ b/tools/testing/selftests/bpf/.gitignore
@@ -38,5 +38,7 @@ test_hashmap
 test_btf_dump
 xdping
 test_cpp
+*.skel.h
 /no_alu32
 /bpf_gcc
+/tools
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index 90de7d75b5c6..f70c8e735120 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -3,10 +3,12 @@ include ../../../../scripts/Kbuild.include
 include ../../../scripts/Makefile.arch
 
 CURDIR := $(abspath .)
-LIBDIR := $(abspath ../../../lib)
+TOOLSDIR := $(abspath ../../..)
+LIBDIR := $(TOOLSDIR)/lib
 BPFDIR := $(LIBDIR)/bpf
-TOOLSDIR := $(abspath ../../../include)
-APIDIR := $(TOOLSDIR)/uapi
+TOOLSINCDIR := $(TOOLSDIR)/include
+BPFTOOLDIR := $(TOOLSDIR)/bpf/bpftool
+APIDIR := $(TOOLSINCDIR)/uapi
 GENDIR := $(abspath ../../../../include/generated)
 GENHDR := $(GENDIR)/autoconf.h
 
@@ -19,7 +21,7 @@ LLC		?= llc
 LLVM_OBJCOPY	?= llvm-objcopy
 BPF_GCC		?= $(shell command -v bpf-gcc;)
 CFLAGS += -g -Wall -O2 $(GENFLAGS) -I$(APIDIR) -I$(LIBDIR) -I$(BPFDIR)	\
-	  -I$(GENDIR) -I$(TOOLSDIR) -I$(CURDIR)				\
+	  -I$(GENDIR) -I$(TOOLSINCDIR) -I$(CURDIR)			\
 	  -Dbpf_prog_load=bpf_prog_test_load				\
 	  -Dbpf_load_program=bpf_test_load_program
 LDLIBS += -lcap -lelf -lrt -lpthread
@@ -117,6 +119,12 @@ $(OUTPUT)/test_cgroup_attach: cgroup_helpers.c
 # force a rebuild of BPFOBJ when its dependencies are updated
 force:
 
+DEFAULT_BPFTOOL := $(OUTPUT)/tools/usr/local/sbin/bpftool
+BPFTOOL ?= $(DEFAULT_BPFTOOL)
+
+$(DEFAULT_BPFTOOL): force
+	$(MAKE) -C $(BPFTOOLDIR) DESTDIR=$(OUTPUT)/tools install
+
 $(BPFOBJ): force
 	$(MAKE) -C $(BPFDIR) OUTPUT=$(OUTPUT)/
 
@@ -180,6 +188,8 @@ define GCC_BPF_BUILD_RULE
 	$(BPF_GCC) $3 $4 -O2 -c $1 -o $2
 endef
 
+SKEL_BLACKLIST := btf__% test_pinning_invalid.c
+
 # Set up extra TRUNNER_XXX "temporary" variables in the environment (relies on
 # $eval()) and pass control to DEFINE_TEST_RUNNER_RULES.
 # Parameters:
@@ -195,8 +205,11 @@ TRUNNER_EXTRA_OBJS := $$(patsubst %.c,$$(TRUNNER_OUTPUT)/%.o,		\
 				 $$(filter %.c,$(TRUNNER_EXTRA_SOURCES)))
 TRUNNER_EXTRA_HDRS := $$(filter %.h,$(TRUNNER_EXTRA_SOURCES))
 TRUNNER_TESTS_HDR := $(TRUNNER_TESTS_DIR)/tests.h
-TRUNNER_BPF_OBJS := $$(patsubst %.c,$$(TRUNNER_OUTPUT)/%.o,		\
-				$$(notdir $$(wildcard $(TRUNNER_BPF_PROGS_DIR)/*.c)))
+TRUNNER_BPF_SRCS := $$(notdir $$(wildcard $(TRUNNER_BPF_PROGS_DIR)/*.c))
+TRUNNER_BPF_OBJS := $$(patsubst %.c,$$(TRUNNER_OUTPUT)/%.o, $$(TRUNNER_BPF_SRCS))
+TRUNNER_BPF_SKELS := $$(patsubst %.c,$$(TRUNNER_OUTPUT)/%.skel.h,	\
+				 $$(filter-out $(SKEL_BLACKLIST),	\
+					       $$(TRUNNER_BPF_SRCS)))
 
 # Evaluate rules now with extra TRUNNER_XXX variables above already defined
 $$(eval $$(call DEFINE_TEST_RUNNER_RULES,$1,$2))
@@ -226,6 +239,11 @@ $(TRUNNER_BPF_OBJS): $(TRUNNER_OUTPUT)/%.o:				\
 	$$(call $(TRUNNER_BPF_BUILD_RULE),$$<,$$@,			\
 					  $(TRUNNER_BPF_CFLAGS),	\
 					  $(TRUNNER_BPF_LDFLAGS))
+
+$(TRUNNER_BPF_SKELS): $(TRUNNER_OUTPUT)/%.skel.h:			\
+		      $(TRUNNER_OUTPUT)/%.o				\
+		      | $(BPFTOOL) $(TRUNNER_OUTPUT)
+	$$(BPFTOOL) gen skeleton $$< > $$@
 endif
 
 # ensure we set up tests.h header generation rule just once
@@ -245,6 +263,7 @@ $(TRUNNER_TEST_OBJS): $(TRUNNER_OUTPUT)/%.test.o:			\
 		      $(TRUNNER_TESTS_DIR)/%.c				\
 		      $(TRUNNER_EXTRA_HDRS)				\
 		      $(TRUNNER_BPF_OBJS)				\
+		      $(TRUNNER_BPF_SKELS)				\
 		      $$(BPFOBJ) | $(TRUNNER_OUTPUT)
 	cd $$(@D) && $$(CC) $$(CFLAGS) -c $(CURDIR)/$$< $$(LDLIBS) -o $$(@F)
 
@@ -255,9 +274,9 @@ $(TRUNNER_EXTRA_OBJS): $(TRUNNER_OUTPUT)/%.o:				\
 		       $$(BPFOBJ) | $(TRUNNER_OUTPUT)
 	$$(CC) $$(CFLAGS) -c $$< $$(LDLIBS) -o $$@
 
+# only copy extra resources if in flavored build
 $(TRUNNER_BINARY)-extras: $(TRUNNER_EXTRA_FILES) | $(TRUNNER_OUTPUT)
 ifneq ($2,)
-	# only copy extra resources if in flavored build
 	cp -a $$^ $(TRUNNER_OUTPUT)/
 endif
 
@@ -323,4 +342,5 @@ $(OUTPUT)/test_cpp: test_cpp.cpp $(BPFOBJ)
 
 EXTRA_CLEAN := $(TEST_CUSTOM_PROGS)					\
 	prog_tests/tests.h map_tests/tests.h verifier/tests.h		\
-	feature $(OUTPUT)/*.o $(OUTPUT)/no_alu32 $(OUTPUT)/bpf_gcc
+	feature $(OUTPUT)/*.o $(OUTPUT)/no_alu32 $(OUTPUT)/bpf_gcc	\
+	tools *.skel.h
diff --git a/tools/testing/selftests/bpf/prog_tests/attach_probe.c b/tools/testing/selftests/bpf/prog_tests/attach_probe.c
index b2e7c1424b07..60da1d08daa0 100644
--- a/tools/testing/selftests/bpf/prog_tests/attach_probe.c
+++ b/tools/testing/selftests/bpf/prog_tests/attach_probe.c
@@ -1,5 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <test_progs.h>
+#include "test_attach_probe.skel.h"
 
 ssize_t get_base_addr() {
 	size_t start;
@@ -25,26 +26,10 @@ BPF_EMBED_OBJ(probe, "test_attach_probe.o");
 
 void test_attach_probe(void)
 {
-	const char *kprobe_name = "kprobe/sys_nanosleep";
-	const char *kretprobe_name = "kretprobe/sys_nanosleep";
-	const char *uprobe_name = "uprobe/trigger_func";
-	const char *uretprobe_name = "uretprobe/trigger_func";
-	const int kprobe_idx = 0, kretprobe_idx = 1;
-	const int uprobe_idx = 2, uretprobe_idx = 3;
-	const char *obj_name = "attach_probe";
-	DECLARE_LIBBPF_OPTS(bpf_object_open_opts, open_opts,
-		.object_name = obj_name,
-		.relaxed_maps = true,
-	);
-	struct bpf_program *kprobe_prog, *kretprobe_prog;
-	struct bpf_program *uprobe_prog, *uretprobe_prog;
-	struct bpf_object *obj;
-	int err, duration = 0, res;
-	struct bpf_link *kprobe_link = NULL;
-	struct bpf_link *kretprobe_link = NULL;
-	struct bpf_link *uprobe_link = NULL;
-	struct bpf_link *uretprobe_link = NULL;
-	int results_map_fd;
+	int duration = 0;
+	struct bpf_link *kprobe_link, *kretprobe_link;
+	struct bpf_link *uprobe_link, *uretprobe_link;
+	struct test_attach_probe* skel;
 	size_t uprobe_offset;
 	ssize_t base_addr;
 
@@ -54,124 +39,68 @@ void test_attach_probe(void)
 		return;
 	uprobe_offset = (size_t)&get_base_addr - base_addr;
 
-	/* open object */
-	obj = bpf_object__open_mem(probe_embed.data, probe_embed.size,
-				   &open_opts);
-	if (CHECK(IS_ERR(obj), "obj_open_mem", "err %ld\n", PTR_ERR(obj)))
+	skel = test_attach_probe__open_and_load(&probe_embed);
+	if (CHECK(!skel, "skel_open", "failed to open skeleton\n"))
 		return;
-
-	if (CHECK(strcmp(bpf_object__name(obj), obj_name), "obj_name",
-		  "wrong obj name '%s', expected '%s'\n",
-		  bpf_object__name(obj), obj_name))
-		goto cleanup;
-
-	kprobe_prog = bpf_object__find_program_by_title(obj, kprobe_name);
-	if (CHECK(!kprobe_prog, "find_probe",
-		  "prog '%s' not found\n", kprobe_name))
-		goto cleanup;
-	kretprobe_prog = bpf_object__find_program_by_title(obj, kretprobe_name);
-	if (CHECK(!kretprobe_prog, "find_probe",
-		  "prog '%s' not found\n", kretprobe_name))
-		goto cleanup;
-	uprobe_prog = bpf_object__find_program_by_title(obj, uprobe_name);
-	if (CHECK(!uprobe_prog, "find_probe",
-		  "prog '%s' not found\n", uprobe_name))
-		goto cleanup;
-	uretprobe_prog = bpf_object__find_program_by_title(obj, uretprobe_name);
-	if (CHECK(!uretprobe_prog, "find_probe",
-		  "prog '%s' not found\n", uretprobe_name))
+	if (CHECK(!skel->bss, "check_bss", ".bss wasn't mmap()-ed\n"))
 		goto cleanup;
 
-	/* create maps && load programs */
-	err = bpf_object__load(obj);
-	if (CHECK(err, "obj_load", "err %d\n", err))
-		goto cleanup;
-
-	/* load maps */
-	results_map_fd = bpf_find_map(__func__, obj, "results_map");
-	if (CHECK(results_map_fd < 0, "find_results_map",
-		  "err %d\n", results_map_fd))
-		goto cleanup;
-
-	kprobe_link = bpf_program__attach_kprobe(kprobe_prog,
+	kprobe_link = bpf_program__attach_kprobe(skel->progs.handle_kprobe,
 						 false /* retprobe */,
 						 SYS_NANOSLEEP_KPROBE_NAME);
 	if (CHECK(IS_ERR(kprobe_link), "attach_kprobe",
-		  "err %ld\n", PTR_ERR(kprobe_link))) {
-		kprobe_link = NULL;
+		  "err %ld\n", PTR_ERR(kprobe_link)))
 		goto cleanup;
-	}
-	kretprobe_link = bpf_program__attach_kprobe(kretprobe_prog,
+	skel->links.handle_kprobe = kprobe_link;
+
+	kretprobe_link = bpf_program__attach_kprobe(skel->progs.handle_kretprobe,
 						    true /* retprobe */,
 						    SYS_NANOSLEEP_KPROBE_NAME);
 	if (CHECK(IS_ERR(kretprobe_link), "attach_kretprobe",
-		  "err %ld\n", PTR_ERR(kretprobe_link))) {
-		kretprobe_link = NULL;
+		  "err %ld\n", PTR_ERR(kretprobe_link)))
 		goto cleanup;
-	}
-	uprobe_link = bpf_program__attach_uprobe(uprobe_prog,
+	skel->links.handle_kretprobe = kretprobe_link;
+
+	uprobe_link = bpf_program__attach_uprobe(skel->progs.handle_uprobe,
 						 false /* retprobe */,
 						 0 /* self pid */,
 						 "/proc/self/exe",
 						 uprobe_offset);
 	if (CHECK(IS_ERR(uprobe_link), "attach_uprobe",
-		  "err %ld\n", PTR_ERR(uprobe_link))) {
-		uprobe_link = NULL;
+		  "err %ld\n", PTR_ERR(uprobe_link)))
 		goto cleanup;
-	}
-	uretprobe_link = bpf_program__attach_uprobe(uretprobe_prog,
+	skel->links.handle_uprobe = uprobe_link;
+
+	uretprobe_link = bpf_program__attach_uprobe(skel->progs.handle_uretprobe,
 						    true /* retprobe */,
 						    -1 /* any pid */,
 						    "/proc/self/exe",
 						    uprobe_offset);
 	if (CHECK(IS_ERR(uretprobe_link), "attach_uretprobe",
-		  "err %ld\n", PTR_ERR(uretprobe_link))) {
-		uretprobe_link = NULL;
+		  "err %ld\n", PTR_ERR(uretprobe_link)))
 		goto cleanup;
-	}
+	skel->links.handle_uretprobe = uretprobe_link;
 
 	/* trigger & validate kprobe && kretprobe */
 	usleep(1);
 
-	err = bpf_map_lookup_elem(results_map_fd, &kprobe_idx, &res);
-	if (CHECK(err, "get_kprobe_res",
-		  "failed to get kprobe res: %d\n", err))
+	if (CHECK(skel->bss->kprobe_res != 1, "check_kprobe_res",
+		  "wrong kprobe res: %d\n", skel->bss->kprobe_res))
 		goto cleanup;
-	if (CHECK(res != kprobe_idx + 1, "check_kprobe_res",
-		  "wrong kprobe res: %d\n", res))
-		goto cleanup;
-
-	err = bpf_map_lookup_elem(results_map_fd, &kretprobe_idx, &res);
-	if (CHECK(err, "get_kretprobe_res",
-		  "failed to get kretprobe res: %d\n", err))
-		goto cleanup;
-	if (CHECK(res != kretprobe_idx + 1, "check_kretprobe_res",
-		  "wrong kretprobe res: %d\n", res))
+	if (CHECK(skel->bss->kretprobe_res != 2, "check_kretprobe_res",
+		  "wrong kretprobe res: %d\n", skel->bss->kretprobe_res))
 		goto cleanup;
 
 	/* trigger & validate uprobe & uretprobe */
 	get_base_addr();
 
-	err = bpf_map_lookup_elem(results_map_fd, &uprobe_idx, &res);
-	if (CHECK(err, "get_uprobe_res",
-		  "failed to get uprobe res: %d\n", err))
-		goto cleanup;
-	if (CHECK(res != uprobe_idx + 1, "check_uprobe_res",
-		  "wrong uprobe res: %d\n", res))
-		goto cleanup;
-
-	err = bpf_map_lookup_elem(results_map_fd, &uretprobe_idx, &res);
-	if (CHECK(err, "get_uretprobe_res",
-		  "failed to get uretprobe res: %d\n", err))
+	if (CHECK(skel->bss->uprobe_res != 3, "check_uprobe_res",
+		  "wrong uprobe res: %d\n", skel->bss->uprobe_res))
 		goto cleanup;
-	if (CHECK(res != uretprobe_idx + 1, "check_uretprobe_res",
-		  "wrong uretprobe res: %d\n", res))
+	if (CHECK(skel->bss->uretprobe_res != 4, "check_uretprobe_res",
+		  "wrong uretprobe res: %d\n", skel->bss->uretprobe_res))
 		goto cleanup;
 
 cleanup:
-	bpf_link__destroy(kprobe_link);
-	bpf_link__destroy(kretprobe_link);
-	bpf_link__destroy(uprobe_link);
-	bpf_link__destroy(uretprobe_link);
-	bpf_object__close(obj);
+	test_attach_probe__destroy(skel);
 }
diff --git a/tools/testing/selftests/bpf/progs/test_attach_probe.c b/tools/testing/selftests/bpf/progs/test_attach_probe.c
index 534621e38906..221b69700625 100644
--- a/tools/testing/selftests/bpf/progs/test_attach_probe.c
+++ b/tools/testing/selftests/bpf/progs/test_attach_probe.c
@@ -5,46 +5,36 @@
 #include <linux/bpf.h>
 #include "bpf_helpers.h"
 
-struct {
-	__uint(type, BPF_MAP_TYPE_ARRAY);
-	__uint(max_entries, 4);
-	__type(key, int);
-	__type(value, int);
-} results_map SEC(".maps");
+int kprobe_res = 0;
+int kretprobe_res = 0;
+int uprobe_res = 0;
+int uretprobe_res = 0;
 
 SEC("kprobe/sys_nanosleep")
-int handle_sys_nanosleep_entry(struct pt_regs *ctx)
+int handle_kprobe(struct pt_regs *ctx)
 {
-	const int key = 0, value = 1;
-
-	bpf_map_update_elem(&results_map, &key, &value, 0);
+	kprobe_res = 1;
 	return 0;
 }
 
 SEC("kretprobe/sys_nanosleep")
-int handle_sys_getpid_return(struct pt_regs *ctx)
+int handle_kretprobe(struct pt_regs *ctx)
 {
-	const int key = 1, value = 2;
-
-	bpf_map_update_elem(&results_map, &key, &value, 0);
+	kretprobe_res = 2;
 	return 0;
 }
 
 SEC("uprobe/trigger_func")
-int handle_uprobe_entry(struct pt_regs *ctx)
+int handle_uprobe(struct pt_regs *ctx)
 {
-	const int key = 2, value = 3;
-
-	bpf_map_update_elem(&results_map, &key, &value, 0);
+	uprobe_res = 3;
 	return 0;
 }
 
 SEC("uretprobe/trigger_func")
-int handle_uprobe_return(struct pt_regs *ctx)
+int handle_uretprobe(struct pt_regs *ctx)
 {
-	const int key = 3, value = 4;
-
-	bpf_map_update_elem(&results_map, &key, &value, 0);
+	uretprobe_res = 4;
 	return 0;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From dde53c1b763b5038545efa5d812758ce589654e1 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Fri, 13 Dec 2019 17:43:39 -0800
Subject: selftests/bpf: Convert few more selftest to skeletons

Convert few more selftests to use generated BPF skeletons as a demonstration
on how to use it.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20191214014341.3442258-16-andriin@fb.com
---
 .../selftests/bpf/prog_tests/fentry_fexit.c        | 105 ++++++++-------------
 .../testing/selftests/bpf/prog_tests/fentry_test.c |  72 ++++++--------
 tools/testing/selftests/bpf/prog_tests/mmap.c      |  58 +++++-------
 .../selftests/bpf/prog_tests/stacktrace_build_id.c |  79 ++++++----------
 .../bpf/prog_tests/stacktrace_build_id_nmi.c       |  84 +++++++----------
 5 files changed, 149 insertions(+), 249 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/fentry_fexit.c b/tools/testing/selftests/bpf/prog_tests/fentry_fexit.c
index 40bcff2cc274..110fcf053fd0 100644
--- a/tools/testing/selftests/bpf/prog_tests/fentry_fexit.c
+++ b/tools/testing/selftests/bpf/prog_tests/fentry_fexit.c
@@ -1,90 +1,59 @@
 // SPDX-License-Identifier: GPL-2.0
 /* Copyright (c) 2019 Facebook */
 #include <test_progs.h>
+#include "test_pkt_access.skel.h"
+#include "fentry_test.skel.h"
+#include "fexit_test.skel.h"
+
+BPF_EMBED_OBJ(pkt_access, "test_pkt_access.o");
+BPF_EMBED_OBJ(fentry, "fentry_test.o");
+BPF_EMBED_OBJ(fexit, "fexit_test.o");
 
 void test_fentry_fexit(void)
 {
-	struct bpf_prog_load_attr attr_fentry = {
-		.file = "./fentry_test.o",
-	};
-	struct bpf_prog_load_attr attr_fexit = {
-		.file = "./fexit_test.o",
-	};
-
-	struct bpf_object *obj_fentry = NULL, *obj_fexit = NULL, *pkt_obj;
-	struct bpf_map *data_map_fentry, *data_map_fexit;
-	char fentry_name[] = "fentry/bpf_fentry_testX";
-	char fexit_name[] = "fexit/bpf_fentry_testX";
-	int err, pkt_fd, kfree_skb_fd, i;
-	struct bpf_link *link[12] = {};
-	struct bpf_program *prog[12];
-	__u32 duration, retval;
-	const int zero = 0;
-	u64 result[12];
-
-	err = bpf_prog_load("./test_pkt_access.o", BPF_PROG_TYPE_SCHED_CLS,
-			    &pkt_obj, &pkt_fd);
-	if (CHECK(err, "prog_load sched cls", "err %d errno %d\n", err, errno))
+	struct test_pkt_access *pkt_skel = NULL;
+	struct fentry_test *fentry_skel = NULL;
+	struct fexit_test *fexit_skel = NULL;
+	__u64 *fentry_res, *fexit_res;
+	__u32 duration = 0, retval;
+	int err, pkt_fd, i;
+
+	pkt_skel = test_pkt_access__open_and_load(&pkt_access_embed);
+	if (CHECK(!pkt_skel, "pkt_skel_load", "pkt_access skeleton failed\n"))
 		return;
-	err = bpf_prog_load_xattr(&attr_fentry, &obj_fentry, &kfree_skb_fd);
-	if (CHECK(err, "prog_load fail", "err %d errno %d\n", err, errno))
+	fentry_skel = fentry_test__open_and_load(&fentry_embed);
+	if (CHECK(!fentry_skel, "fentry_skel_load", "fentry skeleton failed\n"))
 		goto close_prog;
-	err = bpf_prog_load_xattr(&attr_fexit, &obj_fexit, &kfree_skb_fd);
-	if (CHECK(err, "prog_load fail", "err %d errno %d\n", err, errno))
+	fexit_skel = fexit_test__open_and_load(&fexit_embed);
+	if (CHECK(!fexit_skel, "fexit_skel_load", "fexit skeleton failed\n"))
 		goto close_prog;
 
-	for (i = 0; i < 6; i++) {
-		fentry_name[sizeof(fentry_name) - 2] = '1' + i;
-		prog[i] = bpf_object__find_program_by_title(obj_fentry, fentry_name);
-		if (CHECK(!prog[i], "find_prog", "prog %s not found\n", fentry_name))
-			goto close_prog;
-		link[i] = bpf_program__attach_trace(prog[i]);
-		if (CHECK(IS_ERR(link[i]), "attach_trace", "failed to link\n"))
-			goto close_prog;
-	}
-	data_map_fentry = bpf_object__find_map_by_name(obj_fentry, "fentry_t.bss");
-	if (CHECK(!data_map_fentry, "find_data_map", "data map not found\n"))
+	err = fentry_test__attach(fentry_skel);
+	if (CHECK(err, "fentry_attach", "fentry attach failed: %d\n", err))
 		goto close_prog;
-
-	for (i = 6; i < 12; i++) {
-		fexit_name[sizeof(fexit_name) - 2] = '1' + i - 6;
-		prog[i] = bpf_object__find_program_by_title(obj_fexit, fexit_name);
-		if (CHECK(!prog[i], "find_prog", "prog %s not found\n", fexit_name))
-			goto close_prog;
-		link[i] = bpf_program__attach_trace(prog[i]);
-		if (CHECK(IS_ERR(link[i]), "attach_trace", "failed to link\n"))
-			goto close_prog;
-	}
-	data_map_fexit = bpf_object__find_map_by_name(obj_fexit, "fexit_te.bss");
-	if (CHECK(!data_map_fexit, "find_data_map", "data map not found\n"))
+	err = fexit_test__attach(fexit_skel);
+	if (CHECK(err, "fexit_attach", "fexit attach failed: %d\n", err))
 		goto close_prog;
 
+	pkt_fd = bpf_program__fd(pkt_skel->progs.test_pkt_access);
 	err = bpf_prog_test_run(pkt_fd, 1, &pkt_v6, sizeof(pkt_v6),
 				NULL, NULL, &retval, &duration);
 	CHECK(err || retval, "ipv6",
 	      "err %d errno %d retval %d duration %d\n",
 	      err, errno, retval, duration);
 
-	err = bpf_map_lookup_elem(bpf_map__fd(data_map_fentry), &zero, &result);
-	if (CHECK(err, "get_result",
-		  "failed to get output data: %d\n", err))
-		goto close_prog;
-
-	err = bpf_map_lookup_elem(bpf_map__fd(data_map_fexit), &zero, result + 6);
-	if (CHECK(err, "get_result",
-		  "failed to get output data: %d\n", err))
-		goto close_prog;
-
-	for (i = 0; i < 12; i++)
-		if (CHECK(result[i] != 1, "result", "bpf_fentry_test%d failed err %ld\n",
-			  i % 6 + 1, result[i]))
-			goto close_prog;
+	fentry_res = (__u64 *)fentry_skel->bss;
+	fexit_res = (__u64 *)fexit_skel->bss;
+	printf("%lld\n", fentry_skel->bss->test1_result);
+	for (i = 0; i < 6; i++) {
+		CHECK(fentry_res[i] != 1, "result",
+		      "fentry_test%d failed err %lld\n", i + 1, fentry_res[i]);
+		CHECK(fexit_res[i] != 1, "result",
+		      "fexit_test%d failed err %lld\n", i + 1, fexit_res[i]);
+	}
 
 close_prog:
-	for (i = 0; i < 12; i++)
-		if (!IS_ERR_OR_NULL(link[i]))
-			bpf_link__destroy(link[i]);
-	bpf_object__close(obj_fentry);
-	bpf_object__close(obj_fexit);
-	bpf_object__close(pkt_obj);
+	test_pkt_access__destroy(pkt_skel);
+	fentry_test__destroy(fentry_skel);
+	fexit_test__destroy(fexit_skel);
 }
diff --git a/tools/testing/selftests/bpf/prog_tests/fentry_test.c b/tools/testing/selftests/bpf/prog_tests/fentry_test.c
index 9fb103193878..46a4afdf507a 100644
--- a/tools/testing/selftests/bpf/prog_tests/fentry_test.c
+++ b/tools/testing/selftests/bpf/prog_tests/fentry_test.c
@@ -1,64 +1,46 @@
 // SPDX-License-Identifier: GPL-2.0
 /* Copyright (c) 2019 Facebook */
 #include <test_progs.h>
+#include "test_pkt_access.skel.h"
+#include "fentry_test.skel.h"
+
+BPF_EMBED_OBJ_DECLARE(pkt_access);
+BPF_EMBED_OBJ_DECLARE(fentry);
 
 void test_fentry_test(void)
 {
-	struct bpf_prog_load_attr attr = {
-		.file = "./fentry_test.o",
-	};
-
-	char prog_name[] = "fentry/bpf_fentry_testX";
-	struct bpf_object *obj = NULL, *pkt_obj;
-	int err, pkt_fd, kfree_skb_fd, i;
-	struct bpf_link *link[6] = {};
-	struct bpf_program *prog[6];
+	struct test_pkt_access *pkt_skel = NULL;
+	struct fentry_test *fentry_skel = NULL;
+	int err, pkt_fd, i;
 	__u32 duration, retval;
-	struct bpf_map *data_map;
-	const int zero = 0;
-	u64 result[6];
+	__u64 *result;
 
-	err = bpf_prog_load("./test_pkt_access.o", BPF_PROG_TYPE_SCHED_CLS,
-			    &pkt_obj, &pkt_fd);
-	if (CHECK(err, "prog_load sched cls", "err %d errno %d\n", err, errno))
+	pkt_skel = test_pkt_access__open_and_load(&pkt_access_embed);
+	if (CHECK(!pkt_skel, "pkt_skel_load", "pkt_access skeleton failed\n"))
 		return;
-	err = bpf_prog_load_xattr(&attr, &obj, &kfree_skb_fd);
-	if (CHECK(err, "prog_load fail", "err %d errno %d\n", err, errno))
-		goto close_prog;
+	fentry_skel = fentry_test__open_and_load(&fentry_embed);
+	if (CHECK(!fentry_skel, "fentry_skel_load", "fentry skeleton failed\n"))
+		goto cleanup;
 
-	for (i = 0; i < 6; i++) {
-		prog_name[sizeof(prog_name) - 2] = '1' + i;
-		prog[i] = bpf_object__find_program_by_title(obj, prog_name);
-		if (CHECK(!prog[i], "find_prog", "prog %s not found\n", prog_name))
-			goto close_prog;
-		link[i] = bpf_program__attach_trace(prog[i]);
-		if (CHECK(IS_ERR(link[i]), "attach_trace", "failed to link\n"))
-			goto close_prog;
-	}
-	data_map = bpf_object__find_map_by_name(obj, "fentry_t.bss");
-	if (CHECK(!data_map, "find_data_map", "data map not found\n"))
-		goto close_prog;
+	err = fentry_test__attach(fentry_skel);
+	if (CHECK(err, "fentry_attach", "fentry attach failed: %d\n", err))
+		goto cleanup;
 
+	pkt_fd = bpf_program__fd(pkt_skel->progs.test_pkt_access);
 	err = bpf_prog_test_run(pkt_fd, 1, &pkt_v6, sizeof(pkt_v6),
 				NULL, NULL, &retval, &duration);
 	CHECK(err || retval, "ipv6",
 	      "err %d errno %d retval %d duration %d\n",
 	      err, errno, retval, duration);
 
-	err = bpf_map_lookup_elem(bpf_map__fd(data_map), &zero, &result);
-	if (CHECK(err, "get_result",
-		  "failed to get output data: %d\n", err))
-		goto close_prog;
-
-	for (i = 0; i < 6; i++)
-		if (CHECK(result[i] != 1, "result", "bpf_fentry_test%d failed err %ld\n",
-			  i + 1, result[i]))
-			goto close_prog;
+	result = (__u64 *)fentry_skel->bss;
+	for (i = 0; i < 6; i++) {
+		if (CHECK(result[i] != 1, "result",
+			  "fentry_test%d failed err %lld\n", i + 1, result[i]))
+			goto cleanup;
+	}
 
-close_prog:
-	for (i = 0; i < 6; i++)
-		if (!IS_ERR_OR_NULL(link[i]))
-			bpf_link__destroy(link[i]);
-	bpf_object__close(obj);
-	bpf_object__close(pkt_obj);
+cleanup:
+	fentry_test__destroy(fentry_skel);
+	test_pkt_access__destroy(pkt_skel);
 }
diff --git a/tools/testing/selftests/bpf/prog_tests/mmap.c b/tools/testing/selftests/bpf/prog_tests/mmap.c
index 051a6d48762c..95a44d37ccea 100644
--- a/tools/testing/selftests/bpf/prog_tests/mmap.c
+++ b/tools/testing/selftests/bpf/prog_tests/mmap.c
@@ -1,59 +1,41 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <test_progs.h>
 #include <sys/mman.h>
+#include "test_mmap.skel.h"
 
 struct map_data {
 	__u64 val[512 * 4];
 };
 
-struct bss_data {
-	__u64 in_val;
-	__u64 out_val;
-};
-
 static size_t roundup_page(size_t sz)
 {
 	long page_size = sysconf(_SC_PAGE_SIZE);
 	return (sz + page_size - 1) / page_size * page_size;
 }
 
+BPF_EMBED_OBJ(test_mmap, "test_mmap.o");
+
 void test_mmap(void)
 {
-	const char *file = "test_mmap.o";
-	const char *probe_name = "raw_tracepoint/sys_enter";
-	const char *tp_name = "sys_enter";
-	const size_t bss_sz = roundup_page(sizeof(struct bss_data));
+	const size_t bss_sz = roundup_page(sizeof(struct test_mmap__bss));
 	const size_t map_sz = roundup_page(sizeof(struct map_data));
 	const int zero = 0, one = 1, two = 2, far = 1500;
 	const long page_size = sysconf(_SC_PAGE_SIZE);
 	int err, duration = 0, i, data_map_fd;
-	struct bpf_program *prog;
-	struct bpf_object *obj;
-	struct bpf_link *link = NULL;
 	struct bpf_map *data_map, *bss_map;
 	void *bss_mmaped = NULL, *map_mmaped = NULL, *tmp1, *tmp2;
-	volatile struct bss_data *bss_data;
-	volatile struct map_data *map_data;
+	struct test_mmap__bss *bss_data;
+	struct map_data *map_data;
+	struct test_mmap *skel;
 	__u64 val = 0;
 
-	obj = bpf_object__open_file("test_mmap.o", NULL);
-	if (CHECK(IS_ERR(obj), "obj_open", "failed to open '%s': %ld\n",
-		  file, PTR_ERR(obj)))
+
+	skel = test_mmap__open_and_load(&test_mmap_embed);
+	if (CHECK(!skel, "skel_open_and_load", "skeleton open/load failed\n"))
 		return;
-	prog = bpf_object__find_program_by_title(obj, probe_name);
-	if (CHECK(!prog, "find_probe", "prog '%s' not found\n", probe_name))
-		goto cleanup;
-	err = bpf_object__load(obj);
-	if (CHECK(err, "obj_load", "failed to load prog '%s': %d\n",
-		  probe_name, err))
-		goto cleanup;
 
-	bss_map = bpf_object__find_map_by_name(obj, "test_mma.bss");
-	if (CHECK(!bss_map, "find_bss_map", ".bss map not found\n"))
-		goto cleanup;
-	data_map = bpf_object__find_map_by_name(obj, "data_map");
-	if (CHECK(!data_map, "find_data_map", "data_map map not found\n"))
-		goto cleanup;
+	bss_map = skel->maps.bss;
+	data_map = skel->maps.data_map;
 	data_map_fd = bpf_map__fd(data_map);
 
 	bss_mmaped = mmap(NULL, bss_sz, PROT_READ | PROT_WRITE, MAP_SHARED,
@@ -77,13 +59,15 @@ void test_mmap(void)
 
 	CHECK_FAIL(bss_data->in_val);
 	CHECK_FAIL(bss_data->out_val);
+	CHECK_FAIL(skel->bss->in_val);
+	CHECK_FAIL(skel->bss->out_val);
 	CHECK_FAIL(map_data->val[0]);
 	CHECK_FAIL(map_data->val[1]);
 	CHECK_FAIL(map_data->val[2]);
 	CHECK_FAIL(map_data->val[far]);
 
-	link = bpf_program__attach_raw_tracepoint(prog, tp_name);
-	if (CHECK(IS_ERR(link), "attach_raw_tp", "err %ld\n", PTR_ERR(link)))
+	err = test_mmap__attach(skel);
+	if (CHECK(err, "attach_raw_tp", "err %d\n", err))
 		goto cleanup;
 
 	bss_data->in_val = 123;
@@ -94,6 +78,8 @@ void test_mmap(void)
 
 	CHECK_FAIL(bss_data->in_val != 123);
 	CHECK_FAIL(bss_data->out_val != 123);
+	CHECK_FAIL(skel->bss->in_val != 123);
+	CHECK_FAIL(skel->bss->out_val != 123);
 	CHECK_FAIL(map_data->val[0] != 111);
 	CHECK_FAIL(map_data->val[1] != 222);
 	CHECK_FAIL(map_data->val[2] != 123);
@@ -160,6 +146,8 @@ void test_mmap(void)
 	usleep(1);
 	CHECK_FAIL(bss_data->in_val != 321);
 	CHECK_FAIL(bss_data->out_val != 321);
+	CHECK_FAIL(skel->bss->in_val != 321);
+	CHECK_FAIL(skel->bss->out_val != 321);
 	CHECK_FAIL(map_data->val[0] != 111);
 	CHECK_FAIL(map_data->val[1] != 222);
 	CHECK_FAIL(map_data->val[2] != 321);
@@ -203,6 +191,8 @@ void test_mmap(void)
 	map_data = tmp2;
 	CHECK_FAIL(bss_data->in_val != 321);
 	CHECK_FAIL(bss_data->out_val != 321);
+	CHECK_FAIL(skel->bss->in_val != 321);
+	CHECK_FAIL(skel->bss->out_val != 321);
 	CHECK_FAIL(map_data->val[0] != 111);
 	CHECK_FAIL(map_data->val[1] != 222);
 	CHECK_FAIL(map_data->val[2] != 321);
@@ -214,7 +204,5 @@ cleanup:
 		CHECK_FAIL(munmap(bss_mmaped, bss_sz));
 	if (map_mmaped)
 		CHECK_FAIL(munmap(map_mmaped, map_sz));
-	if (!IS_ERR_OR_NULL(link))
-		bpf_link__destroy(link);
-	bpf_object__close(obj);
+	test_mmap__destroy(skel);
 }
diff --git a/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id.c b/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id.c
index d841dced971f..4af8b8253f25 100644
--- a/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id.c
+++ b/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id.c
@@ -1,16 +1,16 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <test_progs.h>
+#include "test_stacktrace_build_id.skel.h"
+
+BPF_EMBED_OBJ(stacktrace_build_id, "test_stacktrace_build_id.o");
 
 void test_stacktrace_build_id(void)
 {
+
 	int control_map_fd, stackid_hmap_fd, stackmap_fd, stack_amap_fd;
-	const char *prog_name = "tracepoint/random/urandom_read";
-	const char *file = "./test_stacktrace_build_id.o";
-	int err, prog_fd, stack_trace_len;
+	struct test_stacktrace_build_id *skel;
+	int err, stack_trace_len;
 	__u32 key, previous_key, val, duration = 0;
-	struct bpf_program *prog;
-	struct bpf_object *obj;
-	struct bpf_link *link = NULL;
 	char buf[256];
 	int i, j;
 	struct bpf_stack_build_id id_offs[PERF_MAX_STACK_DEPTH];
@@ -18,43 +18,24 @@ void test_stacktrace_build_id(void)
 	int retry = 1;
 
 retry:
-	err = bpf_prog_load(file, BPF_PROG_TYPE_TRACEPOINT, &obj, &prog_fd);
-	if (CHECK(err, "prog_load", "err %d errno %d\n", err, errno))
+	skel = test_stacktrace_build_id__open_and_load(&stacktrace_build_id_embed);
+	if (CHECK(!skel, "skel_open_and_load", "skeleton open/load failed\n"))
 		return;
 
-	prog = bpf_object__find_program_by_title(obj, prog_name);
-	if (CHECK(!prog, "find_prog", "prog '%s' not found\n", prog_name))
-		goto close_prog;
-
-	link = bpf_program__attach_tracepoint(prog, "random", "urandom_read");
-	if (CHECK(IS_ERR(link), "attach_tp", "err %ld\n", PTR_ERR(link)))
-		goto close_prog;
+	err = test_stacktrace_build_id__attach(skel);
+	if (CHECK(err, "attach_tp", "err %d\n", err))
+		goto cleanup;
 
 	/* find map fds */
-	control_map_fd = bpf_find_map(__func__, obj, "control_map");
-	if (CHECK(control_map_fd < 0, "bpf_find_map control_map",
-		  "err %d errno %d\n", err, errno))
-		goto disable_pmu;
-
-	stackid_hmap_fd = bpf_find_map(__func__, obj, "stackid_hmap");
-	if (CHECK(stackid_hmap_fd < 0, "bpf_find_map stackid_hmap",
-		  "err %d errno %d\n", err, errno))
-		goto disable_pmu;
-
-	stackmap_fd = bpf_find_map(__func__, obj, "stackmap");
-	if (CHECK(stackmap_fd < 0, "bpf_find_map stackmap", "err %d errno %d\n",
-		  err, errno))
-		goto disable_pmu;
-
-	stack_amap_fd = bpf_find_map(__func__, obj, "stack_amap");
-	if (CHECK(stack_amap_fd < 0, "bpf_find_map stack_amap",
-		  "err %d errno %d\n", err, errno))
-		goto disable_pmu;
+	control_map_fd = bpf_map__fd(skel->maps.control_map);
+	stackid_hmap_fd = bpf_map__fd(skel->maps.stackid_hmap);
+	stackmap_fd = bpf_map__fd(skel->maps.stackmap);
+	stack_amap_fd = bpf_map__fd(skel->maps.stack_amap);
 
 	if (CHECK_FAIL(system("dd if=/dev/urandom of=/dev/zero count=4 2> /dev/null")))
-		goto disable_pmu;
+		goto cleanup;
 	if (CHECK_FAIL(system("./urandom_read")))
-		goto disable_pmu;
+		goto cleanup;
 	/* disable stack trace collection */
 	key = 0;
 	val = 1;
@@ -66,23 +47,23 @@ retry:
 	err = compare_map_keys(stackid_hmap_fd, stackmap_fd);
 	if (CHECK(err, "compare_map_keys stackid_hmap vs. stackmap",
 		  "err %d errno %d\n", err, errno))
-		goto disable_pmu;
+		goto cleanup;
 
 	err = compare_map_keys(stackmap_fd, stackid_hmap_fd);
 	if (CHECK(err, "compare_map_keys stackmap vs. stackid_hmap",
 		  "err %d errno %d\n", err, errno))
-		goto disable_pmu;
+		goto cleanup;
 
 	err = extract_build_id(buf, 256);
 
 	if (CHECK(err, "get build_id with readelf",
 		  "err %d errno %d\n", err, errno))
-		goto disable_pmu;
+		goto cleanup;
 
 	err = bpf_map_get_next_key(stackmap_fd, NULL, &key);
 	if (CHECK(err, "get_next_key from stackmap",
 		  "err %d, errno %d\n", err, errno))
-		goto disable_pmu;
+		goto cleanup;
 
 	do {
 		char build_id[64];
@@ -90,7 +71,7 @@ retry:
 		err = bpf_map_lookup_elem(stackmap_fd, &key, id_offs);
 		if (CHECK(err, "lookup_elem from stackmap",
 			  "err %d, errno %d\n", err, errno))
-			goto disable_pmu;
+			goto cleanup;
 		for (i = 0; i < PERF_MAX_STACK_DEPTH; ++i)
 			if (id_offs[i].status == BPF_STACK_BUILD_ID_VALID &&
 			    id_offs[i].offset != 0) {
@@ -108,8 +89,7 @@ retry:
 	 * try it one more time.
 	 */
 	if (build_id_matches < 1 && retry--) {
-		bpf_link__destroy(link);
-		bpf_object__close(obj);
+		test_stacktrace_build_id__destroy(skel);
 		printf("%s:WARN:Didn't find expected build ID from the map, retrying\n",
 		       __func__);
 		goto retry;
@@ -117,17 +97,14 @@ retry:
 
 	if (CHECK(build_id_matches < 1, "build id match",
 		  "Didn't find expected build ID from the map\n"))
-		goto disable_pmu;
+		goto cleanup;
 
-	stack_trace_len = PERF_MAX_STACK_DEPTH
-		* sizeof(struct bpf_stack_build_id);
+	stack_trace_len = PERF_MAX_STACK_DEPTH *
+			  sizeof(struct bpf_stack_build_id);
 	err = compare_stack_ips(stackmap_fd, stack_amap_fd, stack_trace_len);
 	CHECK(err, "compare_stack_ips stackmap vs. stack_amap",
 	      "err %d errno %d\n", err, errno);
 
-disable_pmu:
-	bpf_link__destroy(link);
-
-close_prog:
-	bpf_object__close(obj);
+cleanup:
+	test_stacktrace_build_id__destroy(skel);
 }
diff --git a/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id_nmi.c b/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id_nmi.c
index f62aa0eb959b..32fb03881a7b 100644
--- a/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id_nmi.c
+++ b/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id_nmi.c
@@ -1,5 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <test_progs.h>
+#include "test_stacktrace_build_id.skel.h"
 
 static __u64 read_perf_max_sample_freq(void)
 {
@@ -14,21 +15,19 @@ static __u64 read_perf_max_sample_freq(void)
 	return sample_freq;
 }
 
+BPF_EMBED_OBJ_DECLARE(stacktrace_build_id);
+
 void test_stacktrace_build_id_nmi(void)
 {
-	int control_map_fd, stackid_hmap_fd, stackmap_fd, stack_amap_fd;
-	const char *prog_name = "tracepoint/random/urandom_read";
-	const char *file = "./test_stacktrace_build_id.o";
-	int err, pmu_fd, prog_fd;
+	int control_map_fd, stackid_hmap_fd, stackmap_fd;
+	struct test_stacktrace_build_id *skel;
+	int err, pmu_fd;
 	struct perf_event_attr attr = {
 		.freq = 1,
 		.type = PERF_TYPE_HARDWARE,
 		.config = PERF_COUNT_HW_CPU_CYCLES,
 	};
 	__u32 key, previous_key, val, duration = 0;
-	struct bpf_program *prog;
-	struct bpf_object *obj;
-	struct bpf_link *link;
 	char buf[256];
 	int i, j;
 	struct bpf_stack_build_id id_offs[PERF_MAX_STACK_DEPTH];
@@ -38,13 +37,16 @@ void test_stacktrace_build_id_nmi(void)
 	attr.sample_freq = read_perf_max_sample_freq();
 
 retry:
-	err = bpf_prog_load(file, BPF_PROG_TYPE_PERF_EVENT, &obj, &prog_fd);
-	if (CHECK(err, "prog_load", "err %d errno %d\n", err, errno))
+	skel = test_stacktrace_build_id__open(&stacktrace_build_id_embed);
+	if (CHECK(!skel, "skel_open", "skeleton open failed\n"))
 		return;
 
-	prog = bpf_object__find_program_by_title(obj, prog_name);
-	if (CHECK(!prog, "find_prog", "prog '%s' not found\n", prog_name))
-		goto close_prog;
+	/* override program type */
+	bpf_program__set_perf_event(skel->progs.oncpu);
+
+	err = test_stacktrace_build_id__load(skel);
+	if (CHECK(err, "skel_load", "skeleton load failed: %d\n", err))
+		goto cleanup;
 
 	pmu_fd = syscall(__NR_perf_event_open, &attr, -1 /* pid */,
 			 0 /* cpu 0 */, -1 /* group id */,
@@ -52,40 +54,25 @@ retry:
 	if (CHECK(pmu_fd < 0, "perf_event_open",
 		  "err %d errno %d. Does the test host support PERF_COUNT_HW_CPU_CYCLES?\n",
 		  pmu_fd, errno))
-		goto close_prog;
+		goto cleanup;
 
-	link = bpf_program__attach_perf_event(prog, pmu_fd);
-	if (CHECK(IS_ERR(link), "attach_perf_event",
-		  "err %ld\n", PTR_ERR(link))) {
+	skel->links.oncpu = bpf_program__attach_perf_event(skel->progs.oncpu,
+							   pmu_fd);
+	if (CHECK(IS_ERR(skel->links.oncpu), "attach_perf_event",
+		  "err %ld\n", PTR_ERR(skel->links.oncpu))) {
 		close(pmu_fd);
-		goto close_prog;
+		goto cleanup;
 	}
 
 	/* find map fds */
-	control_map_fd = bpf_find_map(__func__, obj, "control_map");
-	if (CHECK(control_map_fd < 0, "bpf_find_map control_map",
-		  "err %d errno %d\n", err, errno))
-		goto disable_pmu;
-
-	stackid_hmap_fd = bpf_find_map(__func__, obj, "stackid_hmap");
-	if (CHECK(stackid_hmap_fd < 0, "bpf_find_map stackid_hmap",
-		  "err %d errno %d\n", err, errno))
-		goto disable_pmu;
-
-	stackmap_fd = bpf_find_map(__func__, obj, "stackmap");
-	if (CHECK(stackmap_fd < 0, "bpf_find_map stackmap", "err %d errno %d\n",
-		  err, errno))
-		goto disable_pmu;
-
-	stack_amap_fd = bpf_find_map(__func__, obj, "stack_amap");
-	if (CHECK(stack_amap_fd < 0, "bpf_find_map stack_amap",
-		  "err %d errno %d\n", err, errno))
-		goto disable_pmu;
+	control_map_fd = bpf_map__fd(skel->maps.control_map);
+	stackid_hmap_fd = bpf_map__fd(skel->maps.stackid_hmap);
+	stackmap_fd = bpf_map__fd(skel->maps.stackmap);
 
 	if (CHECK_FAIL(system("dd if=/dev/urandom of=/dev/zero count=4 2> /dev/null")))
-		goto disable_pmu;
+		goto cleanup;
 	if (CHECK_FAIL(system("taskset 0x1 ./urandom_read 100000")))
-		goto disable_pmu;
+		goto cleanup;
 	/* disable stack trace collection */
 	key = 0;
 	val = 1;
@@ -97,23 +84,23 @@ retry:
 	err = compare_map_keys(stackid_hmap_fd, stackmap_fd);
 	if (CHECK(err, "compare_map_keys stackid_hmap vs. stackmap",
 		  "err %d errno %d\n", err, errno))
-		goto disable_pmu;
+		goto cleanup;
 
 	err = compare_map_keys(stackmap_fd, stackid_hmap_fd);
 	if (CHECK(err, "compare_map_keys stackmap vs. stackid_hmap",
 		  "err %d errno %d\n", err, errno))
-		goto disable_pmu;
+		goto cleanup;
 
 	err = extract_build_id(buf, 256);
 
 	if (CHECK(err, "get build_id with readelf",
 		  "err %d errno %d\n", err, errno))
-		goto disable_pmu;
+		goto cleanup;
 
 	err = bpf_map_get_next_key(stackmap_fd, NULL, &key);
 	if (CHECK(err, "get_next_key from stackmap",
 		  "err %d, errno %d\n", err, errno))
-		goto disable_pmu;
+		goto cleanup;
 
 	do {
 		char build_id[64];
@@ -121,7 +108,7 @@ retry:
 		err = bpf_map_lookup_elem(stackmap_fd, &key, id_offs);
 		if (CHECK(err, "lookup_elem from stackmap",
 			  "err %d, errno %d\n", err, errno))
-			goto disable_pmu;
+			goto cleanup;
 		for (i = 0; i < PERF_MAX_STACK_DEPTH; ++i)
 			if (id_offs[i].status == BPF_STACK_BUILD_ID_VALID &&
 			    id_offs[i].offset != 0) {
@@ -139,8 +126,7 @@ retry:
 	 * try it one more time.
 	 */
 	if (build_id_matches < 1 && retry--) {
-		bpf_link__destroy(link);
-		bpf_object__close(obj);
+		test_stacktrace_build_id__destroy(skel);
 		printf("%s:WARN:Didn't find expected build ID from the map, retrying\n",
 		       __func__);
 		goto retry;
@@ -148,7 +134,7 @@ retry:
 
 	if (CHECK(build_id_matches < 1, "build id match",
 		  "Didn't find expected build ID from the map\n"))
-		goto disable_pmu;
+		goto cleanup;
 
 	/*
 	 * We intentionally skip compare_stack_ips(). This is because we
@@ -157,8 +143,6 @@ retry:
 	 * BPF_STACK_BUILD_ID_IP;
 	 */
 
-disable_pmu:
-	bpf_link__destroy(link);
-close_prog:
-	bpf_object__close(obj);
+cleanup:
+	test_stacktrace_build_id__destroy(skel);
 }
-- 
cgit v1.2.3-59-g8ed1b


From 197448eaac1ab330fb485769bccb62346ba1f458 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Fri, 13 Dec 2019 17:43:40 -0800
Subject: selftests/bpf: Add test validating data section to struct convertion
 layout

Add a simple selftests validating datasection-to-struct layour dumping. Global
variables are constructed in such a way as to cause both natural and
artificial padding (through custom alignment requirement).

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20191214014341.3442258-17-andriin@fb.com
---
 tools/testing/selftests/bpf/prog_tests/skeleton.c | 51 +++++++++++++++++++++++
 tools/testing/selftests/bpf/progs/test_skeleton.c | 37 ++++++++++++++++
 2 files changed, 88 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/skeleton.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_skeleton.c

diff --git a/tools/testing/selftests/bpf/prog_tests/skeleton.c b/tools/testing/selftests/bpf/prog_tests/skeleton.c
new file mode 100644
index 000000000000..79f8d13e6740
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/skeleton.c
@@ -0,0 +1,51 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2019 Facebook */
+
+#include <test_progs.h>
+
+struct s {
+	int a;
+	long long b;
+} __attribute__((packed));
+
+#include "test_skeleton.skel.h"
+
+BPF_EMBED_OBJ(skeleton, "test_skeleton.o");
+
+void test_skeleton(void)
+{
+	int duration = 0, err;
+	struct test_skeleton* skel;
+	struct test_skeleton__bss *bss;
+
+	skel = test_skeleton__open_and_load(&skeleton_embed);
+	if (CHECK(!skel, "skel_open", "failed to open skeleton\n"))
+		return;
+
+	bss = skel->bss;
+	bss->in1 = 1;
+	bss->in2 = 2;
+	bss->in3 = 3;
+	bss->in4 = 4;
+	bss->in5.a = 5;
+	bss->in5.b = 6;
+
+	err = test_skeleton__attach(skel);
+	if (CHECK(err, "skel_attach", "skeleton attach failed: %d\n", err))
+		goto cleanup;
+
+	/* trigger tracepoint */
+	usleep(1);
+
+	CHECK(bss->out1 != 1, "res1", "got %d != exp %d\n", bss->out1, 1);
+	CHECK(bss->out2 != 2, "res2", "got %lld != exp %d\n", bss->out2, 2);
+	CHECK(bss->out3 != 3, "res3", "got %d != exp %d\n", (int)bss->out3, 3);
+	CHECK(bss->out4 != 4, "res4", "got %lld != exp %d\n", bss->out4, 4);
+	CHECK(bss->handler_out5.a != 5, "res5", "got %d != exp %d\n",
+	      bss->handler_out5.a, 5);
+	CHECK(bss->handler_out5.b != 6, "res6", "got %lld != exp %d\n",
+	      bss->handler_out5.b, 6);
+
+cleanup:
+	test_skeleton__destroy(skel);
+}
diff --git a/tools/testing/selftests/bpf/progs/test_skeleton.c b/tools/testing/selftests/bpf/progs/test_skeleton.c
new file mode 100644
index 000000000000..c0013d2e16f2
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_skeleton.c
@@ -0,0 +1,37 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2019 Facebook */
+
+#include <linux/bpf.h>
+#include "bpf_helpers.h"
+
+struct s {
+	int a;
+	long long b;
+} __attribute__((packed));
+
+int in1 = 0;
+long long in2 = 0;
+char in3 = '\0';
+long long in4 __attribute__((aligned(64))) = 0;
+struct s in5 = {};
+
+long long out2 = 0;
+char out3 = 0;
+long long out4 = 0;
+int out1 = 0;
+
+
+SEC("raw_tp/sys_enter")
+int handler(const void *ctx)
+{
+	static volatile struct s out5;
+
+	out1 = in1;
+	out2 = in2;
+	out3 = in3;
+	out4 = in4;
+	out5 = in5;
+	return 0;
+}
+
+char _license[] SEC("license") = "GPL";
-- 
cgit v1.2.3-59-g8ed1b


From d9c00c3b1639a3c8f46663cc042a3768d222021f Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Fri, 13 Dec 2019 17:43:41 -0800
Subject: bpftool: Add `gen skeleton` BASH completions

Add BASH completions for gen sub-command.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Cc: Quentin Monnet <quentin.monnet@netronome.com>
Link: https://lore.kernel.org/bpf/20191214014341.3442258-18-andriin@fb.com
---
 tools/bpf/bpftool/bash-completion/bpftool | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/tools/bpf/bpftool/bash-completion/bpftool b/tools/bpf/bpftool/bash-completion/bpftool
index 21c676a1eeb1..754d8395e451 100644
--- a/tools/bpf/bpftool/bash-completion/bpftool
+++ b/tools/bpf/bpftool/bash-completion/bpftool
@@ -839,6 +839,17 @@ _bpftool()
                     ;;
             esac
             ;;
+        gen)
+            case $command in
+                skeleton)
+                    _filedir
+		    ;;
+                *)
+                    [[ $prev == $object ]] && \
+                        COMPREPLY=( $( compgen -W 'skeleton help' -- "$cur" ) )
+                    ;;
+            esac
+            ;;
         cgroup)
             case $command in
                 show|list|tree)
-- 
cgit v1.2.3-59-g8ed1b


From ac9d1389631a4bee0df47e50d6bee8b94230759d Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Fri, 13 Dec 2019 17:47:07 -0800
Subject: libbpf: Extract internal map names into constants

Instead of duplicating string literals, keep them in one place and consistent.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191214014710.3449601-2-andriin@fb.com
---
 tools/lib/bpf/libbpf.c | 23 ++++++++++++++---------
 1 file changed, 14 insertions(+), 9 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index a1a902fa6e0c..f6c135869701 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -195,6 +195,10 @@ struct bpf_program {
 	__u32 prog_flags;
 };
 
+#define DATA_SEC ".data"
+#define BSS_SEC ".bss"
+#define RODATA_SEC ".rodata"
+
 enum libbpf_map_type {
 	LIBBPF_MAP_UNSPEC,
 	LIBBPF_MAP_DATA,
@@ -203,9 +207,9 @@ enum libbpf_map_type {
 };
 
 static const char * const libbpf_type_to_btf_name[] = {
-	[LIBBPF_MAP_DATA]	= ".data",
-	[LIBBPF_MAP_BSS]	= ".bss",
-	[LIBBPF_MAP_RODATA]	= ".rodata",
+	[LIBBPF_MAP_DATA]	= DATA_SEC,
+	[LIBBPF_MAP_BSS]	= BSS_SEC,
+	[LIBBPF_MAP_RODATA]	= RODATA_SEC,
 };
 
 struct bpf_map {
@@ -736,13 +740,13 @@ int bpf_object__section_size(const struct bpf_object *obj, const char *name,
 	*size = 0;
 	if (!name) {
 		return -EINVAL;
-	} else if (!strcmp(name, ".data")) {
+	} else if (!strcmp(name, DATA_SEC)) {
 		if (obj->efile.data)
 			*size = obj->efile.data->d_size;
-	} else if (!strcmp(name, ".bss")) {
+	} else if (!strcmp(name, BSS_SEC)) {
 		if (obj->efile.bss)
 			*size = obj->efile.bss->d_size;
-	} else if (!strcmp(name, ".rodata")) {
+	} else if (!strcmp(name, RODATA_SEC)) {
 		if (obj->efile.rodata)
 			*size = obj->efile.rodata->d_size;
 	} else {
@@ -1684,10 +1688,10 @@ static int bpf_object__elf_collect(struct bpf_object *obj)
 						name, obj->path, cp);
 					return err;
 				}
-			} else if (strcmp(name, ".data") == 0) {
+			} else if (strcmp(name, DATA_SEC) == 0) {
 				obj->efile.data = data;
 				obj->efile.data_shndx = idx;
-			} else if (strcmp(name, ".rodata") == 0) {
+			} else if (strcmp(name, RODATA_SEC) == 0) {
 				obj->efile.rodata = data;
 				obj->efile.rodata_shndx = idx;
 			} else {
@@ -1717,7 +1721,8 @@ static int bpf_object__elf_collect(struct bpf_object *obj)
 
 			obj->efile.reloc_sects[nr_sects].shdr = sh;
 			obj->efile.reloc_sects[nr_sects].data = data;
-		} else if (sh.sh_type == SHT_NOBITS && strcmp(name, ".bss") == 0) {
+		} else if (sh.sh_type == SHT_NOBITS &&
+			   strcmp(name, BSS_SEC) == 0) {
 			obj->efile.bss = data;
 			obj->efile.bss_shndx = idx;
 		} else {
-- 
cgit v1.2.3-59-g8ed1b


From 166750bc1dd256b2184b22588fb9fe6d3fbb93ae Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Fri, 13 Dec 2019 17:47:08 -0800
Subject: libbpf: Support libbpf-provided extern variables

Add support for extern variables, provided to BPF program by libbpf. Currently
the following extern variables are supported:
  - LINUX_KERNEL_VERSION; version of a kernel in which BPF program is
    executing, follows KERNEL_VERSION() macro convention, can be 4- and 8-byte
    long;
  - CONFIG_xxx values; a set of values of actual kernel config. Tristate,
    boolean, strings, and integer values are supported.

Set of possible values is determined by declared type of extern variable.
Supported types of variables are:
- Tristate values. Are represented as `enum libbpf_tristate`. Accepted values
  are **strictly** 'y', 'n', or 'm', which are represented as TRI_YES, TRI_NO,
  or TRI_MODULE, respectively.
- Boolean values. Are represented as bool (_Bool) types. Accepted values are
  'y' and 'n' only, turning into true/false values, respectively.
- Single-character values. Can be used both as a substritute for
  bool/tristate, or as a small-range integer:
  - 'y'/'n'/'m' are represented as is, as characters 'y', 'n', or 'm';
  - integers in a range [-128, 127] or [0, 255] (depending on signedness of
    char in target architecture) are recognized and represented with
    respective values of char type.
- Strings. String values are declared as fixed-length char arrays. String of
  up to that length will be accepted and put in first N bytes of char array,
  with the rest of bytes zeroed out. If config string value is longer than
  space alloted, it will be truncated and warning message emitted. Char array
  is always zero terminated. String literals in config have to be enclosed in
  double quotes, just like C-style string literals.
- Integers. 8-, 16-, 32-, and 64-bit integers are supported, both signed and
  unsigned variants. Libbpf enforces parsed config value to be in the
  supported range of corresponding integer type. Integers values in config can
  be:
  - decimal integers, with optional + and - signs;
  - hexadecimal integers, prefixed with 0x or 0X;
  - octal integers, starting with 0.

Config file itself is searched in /boot/config-$(uname -r) location with
fallback to /proc/config.gz, unless config path is specified explicitly
through bpf_object_open_opts' kernel_config_path option. Both gzipped and
plain text formats are supported. Libbpf adds explicit dependency on zlib
because of this, but this shouldn't be a problem, given libelf already depends
on zlib.

All detected extern variables, are put into a separate .extern internal map.
It, similarly to .rodata map, is marked as read-only from BPF program side, as
well as is frozen on load. This allows BPF verifier to track extern values as
constants and perform enhanced branch prediction and dead code elimination.
This can be relied upon for doing kernel version/feature detection and using
potentially unsupported field relocations or BPF helpers in a CO-RE-based BPF
program, while still having a single version of BPF program running on old and
new kernels. Selftests are validating this explicitly for unexisting BPF
helper.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191214014710.3449601-3-andriin@fb.com
---
 include/uapi/linux/btf.h             |   3 +-
 tools/include/uapi/linux/btf.h       |   7 +-
 tools/lib/bpf/Makefile               |  15 +-
 tools/lib/bpf/bpf_helpers.h          |   9 +
 tools/lib/bpf/btf.c                  |   9 +-
 tools/lib/bpf/libbpf.c               | 738 ++++++++++++++++++++++++++++++++---
 tools/lib/bpf/libbpf.h               |  12 +-
 tools/testing/selftests/bpf/Makefile |   2 +-
 8 files changed, 729 insertions(+), 66 deletions(-)

diff --git a/include/uapi/linux/btf.h b/include/uapi/linux/btf.h
index c02dec97e1ce..1a2898c482ee 100644
--- a/include/uapi/linux/btf.h
+++ b/include/uapi/linux/btf.h
@@ -142,7 +142,8 @@ struct btf_param {
 
 enum {
 	BTF_VAR_STATIC = 0,
-	BTF_VAR_GLOBAL_ALLOCATED,
+	BTF_VAR_GLOBAL_ALLOCATED = 1,
+	BTF_VAR_GLOBAL_EXTERN = 2,
 };
 
 /* BTF_KIND_VAR is followed by a single "struct btf_var" to describe
diff --git a/tools/include/uapi/linux/btf.h b/tools/include/uapi/linux/btf.h
index 63ae4a39e58b..1a2898c482ee 100644
--- a/tools/include/uapi/linux/btf.h
+++ b/tools/include/uapi/linux/btf.h
@@ -22,9 +22,9 @@ struct btf_header {
 };
 
 /* Max # of type identifier */
-#define BTF_MAX_TYPE	0x0000ffff
+#define BTF_MAX_TYPE	0x000fffff
 /* Max offset into the string section */
-#define BTF_MAX_NAME_OFFSET	0x0000ffff
+#define BTF_MAX_NAME_OFFSET	0x00ffffff
 /* Max # of struct/union/enum members or func args */
 #define BTF_MAX_VLEN	0xffff
 
@@ -142,7 +142,8 @@ struct btf_param {
 
 enum {
 	BTF_VAR_STATIC = 0,
-	BTF_VAR_GLOBAL_ALLOCATED,
+	BTF_VAR_GLOBAL_ALLOCATED = 1,
+	BTF_VAR_GLOBAL_EXTERN = 2,
 };
 
 /* BTF_KIND_VAR is followed by a single "struct btf_var" to describe
diff --git a/tools/lib/bpf/Makefile b/tools/lib/bpf/Makefile
index 23ae06c43d08..a3718cb275f2 100644
--- a/tools/lib/bpf/Makefile
+++ b/tools/lib/bpf/Makefile
@@ -56,8 +56,8 @@ ifndef VERBOSE
 endif
 
 FEATURE_USER = .libbpf
-FEATURE_TESTS = libelf libelf-mmap bpf reallocarray
-FEATURE_DISPLAY = libelf bpf
+FEATURE_TESTS = libelf libelf-mmap zlib bpf reallocarray
+FEATURE_DISPLAY = libelf zlib bpf
 
 INCLUDES = -I. -I$(srctree)/tools/include -I$(srctree)/tools/arch/$(ARCH)/include/uapi -I$(srctree)/tools/include/uapi
 FEATURE_CHECK_CFLAGS-bpf = $(INCLUDES)
@@ -160,7 +160,7 @@ all: fixdep
 
 all_cmd: $(CMD_TARGETS) check
 
-$(BPF_IN_SHARED): force elfdep bpfdep bpf_helper_defs.h
+$(BPF_IN_SHARED): force elfdep zdep bpfdep bpf_helper_defs.h
 	@(test -f ../../include/uapi/linux/bpf.h -a -f ../../../include/uapi/linux/bpf.h && ( \
 	(diff -B ../../include/uapi/linux/bpf.h ../../../include/uapi/linux/bpf.h >/dev/null) || \
 	echo "Warning: Kernel ABI header at 'tools/include/uapi/linux/bpf.h' differs from latest version at 'include/uapi/linux/bpf.h'" >&2 )) || true
@@ -178,7 +178,7 @@ $(BPF_IN_SHARED): force elfdep bpfdep bpf_helper_defs.h
 	echo "Warning: Kernel ABI header at 'tools/include/uapi/linux/if_xdp.h' differs from latest version at 'include/uapi/linux/if_xdp.h'" >&2 )) || true
 	$(Q)$(MAKE) $(build)=libbpf OUTPUT=$(SHARED_OBJDIR) CFLAGS="$(CFLAGS) $(SHLIB_FLAGS)"
 
-$(BPF_IN_STATIC): force elfdep bpfdep bpf_helper_defs.h
+$(BPF_IN_STATIC): force elfdep zdep bpfdep bpf_helper_defs.h
 	$(Q)$(MAKE) $(build)=libbpf OUTPUT=$(STATIC_OBJDIR)
 
 bpf_helper_defs.h: $(srctree)/tools/include/uapi/linux/bpf.h
@@ -190,7 +190,7 @@ $(OUTPUT)libbpf.so: $(OUTPUT)libbpf.so.$(LIBBPF_VERSION)
 $(OUTPUT)libbpf.so.$(LIBBPF_VERSION): $(BPF_IN_SHARED)
 	$(QUIET_LINK)$(CC) $(LDFLAGS) \
 		--shared -Wl,-soname,libbpf.so.$(LIBBPF_MAJOR_VERSION) \
-		-Wl,--version-script=$(VERSION_SCRIPT) $^ -lelf -o $@
+		-Wl,--version-script=$(VERSION_SCRIPT) $^ -lelf -lz -o $@
 	@ln -sf $(@F) $(OUTPUT)libbpf.so
 	@ln -sf $(@F) $(OUTPUT)libbpf.so.$(LIBBPF_MAJOR_VERSION)
 
@@ -279,12 +279,15 @@ clean:
 
 
 
-PHONY += force elfdep bpfdep cscope tags
+PHONY += force elfdep zdep bpfdep cscope tags
 force:
 
 elfdep:
 	@if [ "$(feature-libelf)" != "1" ]; then echo "No libelf found"; exit 1 ; fi
 
+zdep:
+	@if [ "$(feature-zlib)" != "1" ]; then echo "No zlib found"; exit 1 ; fi
+
 bpfdep:
 	@if [ "$(feature-bpf)" != "1" ]; then echo "BPF API too old"; exit 1 ; fi
 
diff --git a/tools/lib/bpf/bpf_helpers.h b/tools/lib/bpf/bpf_helpers.h
index 0c7d28292898..aa46700075e1 100644
--- a/tools/lib/bpf/bpf_helpers.h
+++ b/tools/lib/bpf/bpf_helpers.h
@@ -25,6 +25,9 @@
 #ifndef __always_inline
 #define __always_inline __attribute__((always_inline))
 #endif
+#ifndef __weak
+#define __weak __attribute__((weak))
+#endif
 
 /*
  * Helper structure used by eBPF C program
@@ -44,4 +47,10 @@ enum libbpf_pin_type {
 	LIBBPF_PIN_BY_NAME,
 };
 
+enum libbpf_tristate {
+	TRI_NO = 0,
+	TRI_YES = 1,
+	TRI_MODULE = 2,
+};
+
 #endif
diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
index 84fe82f27bef..520021939d81 100644
--- a/tools/lib/bpf/btf.c
+++ b/tools/lib/bpf/btf.c
@@ -578,6 +578,12 @@ static int btf_fixup_datasec(struct bpf_object *obj, struct btf *btf,
 		return -ENOENT;
 	}
 
+	/* .extern datasec size and var offsets were set correctly during
+	 * extern collection step, so just skip straight to sorting variables
+	 */
+	if (t->size)
+		goto sort_vars;
+
 	ret = bpf_object__section_size(obj, name, &size);
 	if (ret || !size || (t->size && t->size != size)) {
 		pr_debug("Invalid size for section %s: %u bytes\n", name, size);
@@ -614,7 +620,8 @@ static int btf_fixup_datasec(struct bpf_object *obj, struct btf *btf,
 		vsi->offset = off;
 	}
 
-	qsort(t + 1, vars, sizeof(*vsi), compare_vsi_off);
+sort_vars:
+	qsort(btf_var_secinfos(t), vars, sizeof(*vsi), compare_vsi_off);
 	return 0;
 }
 
diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index f6c135869701..26144706ba0b 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -44,6 +44,7 @@
 #include <tools/libc_compat.h>
 #include <libelf.h>
 #include <gelf.h>
+#include <zlib.h>
 
 #include "libbpf.h"
 #include "bpf.h"
@@ -139,6 +140,20 @@ struct bpf_capabilities {
 	__u32 array_mmap:1;
 };
 
+enum reloc_type {
+	RELO_LD64,
+	RELO_CALL,
+	RELO_DATA,
+	RELO_EXTERN,
+};
+
+struct reloc_desc {
+	enum reloc_type type;
+	int insn_idx;
+	int map_idx;
+	int sym_off;
+};
+
 /*
  * bpf_prog should be a better name but it has been used in
  * linux/filter.h.
@@ -157,16 +172,7 @@ struct bpf_program {
 	size_t insns_cnt, main_prog_cnt;
 	enum bpf_prog_type type;
 
-	struct reloc_desc {
-		enum {
-			RELO_LD64,
-			RELO_CALL,
-			RELO_DATA,
-		} type;
-		int insn_idx;
-		int map_idx;
-		int sym_off;
-	} *reloc_desc;
+	struct reloc_desc *reloc_desc;
 	int nr_reloc;
 	int log_level;
 
@@ -198,18 +204,21 @@ struct bpf_program {
 #define DATA_SEC ".data"
 #define BSS_SEC ".bss"
 #define RODATA_SEC ".rodata"
+#define EXTERN_SEC ".extern"
 
 enum libbpf_map_type {
 	LIBBPF_MAP_UNSPEC,
 	LIBBPF_MAP_DATA,
 	LIBBPF_MAP_BSS,
 	LIBBPF_MAP_RODATA,
+	LIBBPF_MAP_EXTERN,
 };
 
 static const char * const libbpf_type_to_btf_name[] = {
 	[LIBBPF_MAP_DATA]	= DATA_SEC,
 	[LIBBPF_MAP_BSS]	= BSS_SEC,
 	[LIBBPF_MAP_RODATA]	= RODATA_SEC,
+	[LIBBPF_MAP_EXTERN]	= EXTERN_SEC,
 };
 
 struct bpf_map {
@@ -231,6 +240,28 @@ struct bpf_map {
 	bool reused;
 };
 
+enum extern_type {
+	EXT_UNKNOWN,
+	EXT_CHAR,
+	EXT_BOOL,
+	EXT_INT,
+	EXT_TRISTATE,
+	EXT_CHAR_ARR,
+};
+
+struct extern_desc {
+	const char *name;
+	int sym_idx;
+	int btf_id;
+	enum extern_type type;
+	int sz;
+	int align;
+	int data_off;
+	bool is_signed;
+	bool is_weak;
+	bool is_set;
+};
+
 static LIST_HEAD(bpf_objects_list);
 
 struct bpf_object {
@@ -244,6 +275,11 @@ struct bpf_object {
 	size_t nr_maps;
 	size_t maps_cap;
 
+	char *kconfig_path;
+	struct extern_desc *externs;
+	int nr_extern;
+	int extern_map_idx;
+
 	bool loaded;
 	bool has_pseudo_calls;
 	bool relaxed_core_relocs;
@@ -271,6 +307,7 @@ struct bpf_object {
 		int maps_shndx;
 		int btf_maps_shndx;
 		int text_shndx;
+		int symbols_shndx;
 		int data_shndx;
 		int rodata_shndx;
 		int bss_shndx;
@@ -542,6 +579,7 @@ static struct bpf_object *bpf_object__new(const char *path,
 	obj->efile.data_shndx = -1;
 	obj->efile.rodata_shndx = -1;
 	obj->efile.bss_shndx = -1;
+	obj->extern_map_idx = -1;
 
 	obj->kern_version = get_kernel_version();
 	obj->loaded = false;
@@ -866,7 +904,8 @@ bpf_object__init_internal_map(struct bpf_object *obj, enum libbpf_map_type type,
 	def->key_size = sizeof(int);
 	def->value_size = data_sz;
 	def->max_entries = 1;
-	def->map_flags = type == LIBBPF_MAP_RODATA ? BPF_F_RDONLY_PROG : 0;
+	def->map_flags = type == LIBBPF_MAP_RODATA || type == LIBBPF_MAP_EXTERN
+			 ? BPF_F_RDONLY_PROG : 0;
 	def->map_flags |= BPF_F_MMAPABLE;
 
 	pr_debug("map '%s' (global data): at sec_idx %d, offset %zu, flags %x.\n",
@@ -883,7 +922,7 @@ bpf_object__init_internal_map(struct bpf_object *obj, enum libbpf_map_type type,
 		return err;
 	}
 
-	if (type != LIBBPF_MAP_BSS)
+	if (data)
 		memcpy(map->mmaped, data, data_sz);
 
 	pr_debug("map %td is \"%s\"\n", map - obj->maps, map->name);
@@ -924,6 +963,271 @@ static int bpf_object__init_global_data_maps(struct bpf_object *obj)
 	return 0;
 }
 
+
+static struct extern_desc *find_extern_by_name(const struct bpf_object *obj,
+					       const void *name)
+{
+	int i;
+
+	for (i = 0; i < obj->nr_extern; i++) {
+		if (strcmp(obj->externs[i].name, name) == 0)
+			return &obj->externs[i];
+	}
+	return NULL;
+}
+
+static int set_ext_value_tri(struct extern_desc *ext, void *ext_val,
+			     char value)
+{
+	switch (ext->type) {
+	case EXT_BOOL:
+		if (value == 'm') {
+			pr_warn("extern %s=%c should be tristate or char\n",
+				ext->name, value);
+			return -EINVAL;
+		}
+		*(bool *)ext_val = value == 'y' ? true : false;
+		break;
+	case EXT_TRISTATE:
+		if (value == 'y')
+			*(enum libbpf_tristate *)ext_val = TRI_YES;
+		else if (value == 'm')
+			*(enum libbpf_tristate *)ext_val = TRI_MODULE;
+		else /* value == 'n' */
+			*(enum libbpf_tristate *)ext_val = TRI_NO;
+		break;
+	case EXT_CHAR:
+		*(char *)ext_val = value;
+		break;
+	case EXT_UNKNOWN:
+	case EXT_INT:
+	case EXT_CHAR_ARR:
+	default:
+		pr_warn("extern %s=%c should be bool, tristate, or char\n",
+			ext->name, value);
+		return -EINVAL;
+	}
+	ext->is_set = true;
+	return 0;
+}
+
+static int set_ext_value_str(struct extern_desc *ext, char *ext_val,
+			     const char *value)
+{
+	size_t len;
+
+	if (ext->type != EXT_CHAR_ARR) {
+		pr_warn("extern %s=%s should char array\n", ext->name, value);
+		return -EINVAL;
+	}
+
+	len = strlen(value);
+	if (value[len - 1] != '"') {
+		pr_warn("extern '%s': invalid string config '%s'\n",
+			ext->name, value);
+		return -EINVAL;
+	}
+
+	/* strip quotes */
+	len -= 2;
+	if (len >= ext->sz) {
+		pr_warn("extern '%s': long string config %s of (%zu bytes) truncated to %d bytes\n",
+			ext->name, value, len, ext->sz - 1);
+		len = ext->sz - 1;
+	}
+	memcpy(ext_val, value + 1, len);
+	ext_val[len] = '\0';
+	ext->is_set = true;
+	return 0;
+}
+
+static int parse_u64(const char *value, __u64 *res)
+{
+	char *value_end;
+	int err;
+
+	errno = 0;
+	*res = strtoull(value, &value_end, 0);
+	if (errno) {
+		err = -errno;
+		pr_warn("failed to parse '%s' as integer: %d\n", value, err);
+		return err;
+	}
+	if (*value_end) {
+		pr_warn("failed to parse '%s' as integer completely\n", value);
+		return -EINVAL;
+	}
+	return 0;
+}
+
+static bool is_ext_value_in_range(const struct extern_desc *ext, __u64 v)
+{
+	int bit_sz = ext->sz * 8;
+
+	if (ext->sz == 8)
+		return true;
+
+	/* Validate that value stored in u64 fits in integer of `ext->sz`
+	 * bytes size without any loss of information. If the target integer
+	 * is signed, we rely on the following limits of integer type of
+	 * Y bits and subsequent transformation:
+	 *
+	 *     -2^(Y-1) <= X           <= 2^(Y-1) - 1
+	 *            0 <= X + 2^(Y-1) <= 2^Y - 1
+	 *            0 <= X + 2^(Y-1) <  2^Y
+	 *
+	 *  For unsigned target integer, check that all the (64 - Y) bits are
+	 *  zero.
+	 */
+	if (ext->is_signed)
+		return v + (1ULL << (bit_sz - 1)) < (1ULL << bit_sz);
+	else
+		return (v >> bit_sz) == 0;
+}
+
+static int set_ext_value_num(struct extern_desc *ext, void *ext_val,
+			     __u64 value)
+{
+	if (ext->type != EXT_INT && ext->type != EXT_CHAR) {
+		pr_warn("extern %s=%llu should be integer\n",
+			ext->name, value);
+		return -EINVAL;
+	}
+	if (!is_ext_value_in_range(ext, value)) {
+		pr_warn("extern %s=%llu value doesn't fit in %d bytes\n",
+			ext->name, value, ext->sz);
+		return -ERANGE;
+	}
+	switch (ext->sz) {
+		case 1: *(__u8 *)ext_val = value; break;
+		case 2: *(__u16 *)ext_val = value; break;
+		case 4: *(__u32 *)ext_val = value; break;
+		case 8: *(__u64 *)ext_val = value; break;
+		default:
+			return -EINVAL;
+	}
+	ext->is_set = true;
+	return 0;
+}
+
+static int bpf_object__read_kernel_config(struct bpf_object *obj,
+					  const char *config_path,
+					  void *data)
+{
+	char buf[PATH_MAX], *sep, *value;
+	struct extern_desc *ext;
+	int len, err = 0;
+	void *ext_val;
+	__u64 num;
+	gzFile file;
+
+	if (config_path) {
+		file = gzopen(config_path, "r");
+	} else {
+		struct utsname uts;
+
+		uname(&uts);
+		len = snprintf(buf, PATH_MAX, "/boot/config-%s", uts.release);
+		if (len < 0)
+			return -EINVAL;
+		else if (len >= PATH_MAX)
+			return -ENAMETOOLONG;
+		/* gzopen also accepts uncompressed files. */
+		file = gzopen(buf, "r");
+		if (!file)
+			file = gzopen("/proc/config.gz", "r");
+	}
+	if (!file) {
+		pr_warn("failed to read kernel config at '%s'\n", config_path);
+		return -ENOENT;
+	}
+
+	while (gzgets(file, buf, sizeof(buf))) {
+		if (strncmp(buf, "CONFIG_", 7))
+			continue;
+
+		sep = strchr(buf, '=');
+		if (!sep) {
+			err = -EINVAL;
+			pr_warn("failed to parse '%s': no separator\n", buf);
+			goto out;
+		}
+		/* Trim ending '\n' */
+		len = strlen(buf);
+		if (buf[len - 1] == '\n')
+			buf[len - 1] = '\0';
+		/* Split on '=' and ensure that a value is present. */
+		*sep = '\0';
+		if (!sep[1]) {
+			err = -EINVAL;
+			*sep = '=';
+			pr_warn("failed to parse '%s': no value\n", buf);
+			goto out;
+		}
+
+		ext = find_extern_by_name(obj, buf);
+		if (!ext)
+			continue;
+		if (ext->is_set) {
+			err = -EINVAL;
+			pr_warn("re-defining extern '%s' not allowed\n", buf);
+			goto out;
+		}
+
+		ext_val = data + ext->data_off;
+		value = sep + 1;
+
+		switch (*value) {
+		case 'y': case 'n': case 'm':
+			err = set_ext_value_tri(ext, ext_val, *value);
+			break;
+		case '"':
+			err = set_ext_value_str(ext, ext_val, value);
+			break;
+		default:
+			/* assume integer */
+			err = parse_u64(value, &num);
+			if (err) {
+				pr_warn("extern %s=%s should be integer\n",
+					ext->name, value);
+				goto out;
+			}
+			err = set_ext_value_num(ext, ext_val, num);
+			break;
+		}
+		if (err)
+			goto out;
+		pr_debug("extern %s=%s\n", ext->name, value);
+	}
+
+out:
+	gzclose(file);
+	return err;
+}
+
+static int bpf_object__init_extern_map(struct bpf_object *obj)
+{
+	struct extern_desc *last_ext;
+	size_t map_sz;
+	int err;
+
+	if (obj->nr_extern == 0)
+		return 0;
+
+	last_ext = &obj->externs[obj->nr_extern - 1];
+	map_sz = last_ext->data_off + last_ext->sz;
+
+	err = bpf_object__init_internal_map(obj, LIBBPF_MAP_EXTERN,
+					    obj->efile.symbols_shndx,
+					    NULL, map_sz);
+	if (err)
+		return err;
+
+	obj->extern_map_idx = obj->nr_maps - 1;
+
+	return 0;
+}
+
 static int bpf_object__init_user_maps(struct bpf_object *obj, bool strict)
 {
 	Elf_Data *symbols = obj->efile.symbols;
@@ -1401,19 +1705,17 @@ static int bpf_object__init_user_btf_maps(struct bpf_object *obj, bool strict,
 static int bpf_object__init_maps(struct bpf_object *obj,
 				 const struct bpf_object_open_opts *opts)
 {
-	const char *pin_root_path = OPTS_GET(opts, pin_root_path, NULL);
-	bool strict = !OPTS_GET(opts, relaxed_maps, false);
+	const char *pin_root_path;
+	bool strict;
 	int err;
 
-	err = bpf_object__init_user_maps(obj, strict);
-	if (err)
-		return err;
+	strict = !OPTS_GET(opts, relaxed_maps, false);
+	pin_root_path = OPTS_GET(opts, pin_root_path, NULL);
 
-	err = bpf_object__init_user_btf_maps(obj, strict, pin_root_path);
-	if (err)
-		return err;
-
-	err = bpf_object__init_global_data_maps(obj);
+	err = bpf_object__init_user_maps(obj, strict);
+	err = err ?: bpf_object__init_user_btf_maps(obj, strict, pin_root_path);
+	err = err ?: bpf_object__init_global_data_maps(obj);
+	err = err ?: bpf_object__init_extern_map(obj);
 	if (err)
 		return err;
 
@@ -1532,11 +1834,6 @@ static int bpf_object__init_btf(struct bpf_object *obj,
 				BTF_ELF_SEC, err);
 			goto out;
 		}
-		err = btf__finalize_data(obj, obj->btf);
-		if (err) {
-			pr_warn("Error finalizing %s: %d.\n", BTF_ELF_SEC, err);
-			goto out;
-		}
 	}
 	if (btf_ext_data) {
 		if (!obj->btf) {
@@ -1570,6 +1867,30 @@ out:
 	return 0;
 }
 
+static int bpf_object__finalize_btf(struct bpf_object *obj)
+{
+	int err;
+
+	if (!obj->btf)
+		return 0;
+
+	err = btf__finalize_data(obj, obj->btf);
+	if (!err)
+		return 0;
+
+	pr_warn("Error finalizing %s: %d.\n", BTF_ELF_SEC, err);
+	btf__free(obj->btf);
+	obj->btf = NULL;
+	btf_ext__free(obj->btf_ext);
+	obj->btf_ext = NULL;
+
+	if (bpf_object__is_btf_mandatory(obj)) {
+		pr_warn("BTF is required, but is missing or corrupted.\n");
+		return -ENOENT;
+	}
+	return 0;
+}
+
 static int bpf_object__sanitize_and_load_btf(struct bpf_object *obj)
 {
 	int err = 0;
@@ -1670,6 +1991,7 @@ static int bpf_object__elf_collect(struct bpf_object *obj)
 				return -LIBBPF_ERRNO__FORMAT;
 			}
 			obj->efile.symbols = data;
+			obj->efile.symbols_shndx = idx;
 			obj->efile.strtabidx = sh.sh_link;
 		} else if (sh.sh_type == SHT_PROGBITS && data->d_size > 0) {
 			if (sh.sh_flags & SHF_EXECINSTR) {
@@ -1737,6 +2059,216 @@ static int bpf_object__elf_collect(struct bpf_object *obj)
 	return bpf_object__init_btf(obj, btf_data, btf_ext_data);
 }
 
+static bool sym_is_extern(const GElf_Sym *sym)
+{
+	int bind = GELF_ST_BIND(sym->st_info);
+	/* externs are symbols w/ type=NOTYPE, bind=GLOBAL|WEAK, section=UND */
+	return sym->st_shndx == SHN_UNDEF &&
+	       (bind == STB_GLOBAL || bind == STB_WEAK) &&
+	       GELF_ST_TYPE(sym->st_info) == STT_NOTYPE;
+}
+
+static int find_extern_btf_id(const struct btf *btf, const char *ext_name)
+{
+	const struct btf_type *t;
+	const char *var_name;
+	int i, n;
+
+	if (!btf)
+		return -ESRCH;
+
+	n = btf__get_nr_types(btf);
+	for (i = 1; i <= n; i++) {
+		t = btf__type_by_id(btf, i);
+
+		if (!btf_is_var(t))
+			continue;
+
+		var_name = btf__name_by_offset(btf, t->name_off);
+		if (strcmp(var_name, ext_name))
+			continue;
+
+		if (btf_var(t)->linkage != BTF_VAR_GLOBAL_EXTERN)
+			return -EINVAL;
+
+		return i;
+	}
+
+	return -ENOENT;
+}
+
+static enum extern_type find_extern_type(const struct btf *btf, int id,
+					 bool *is_signed)
+{
+	const struct btf_type *t;
+	const char *name;
+
+	t = skip_mods_and_typedefs(btf, id, NULL);
+	name = btf__name_by_offset(btf, t->name_off);
+
+	if (is_signed)
+		*is_signed = false;
+	switch (btf_kind(t)) {
+	case BTF_KIND_INT: {
+		int enc = btf_int_encoding(t);
+
+		if (enc & BTF_INT_BOOL)
+			return t->size == 1 ? EXT_BOOL : EXT_UNKNOWN;
+		if (is_signed)
+			*is_signed = enc & BTF_INT_SIGNED;
+		if (t->size == 1)
+			return EXT_CHAR;
+		if (t->size < 1 || t->size > 8 || (t->size & (t->size - 1)))
+			return EXT_UNKNOWN;
+		return EXT_INT;
+	}
+	case BTF_KIND_ENUM:
+		if (t->size != 4)
+			return EXT_UNKNOWN;
+		if (strcmp(name, "libbpf_tristate"))
+			return EXT_UNKNOWN;
+		return EXT_TRISTATE;
+	case BTF_KIND_ARRAY:
+		if (btf_array(t)->nelems == 0)
+			return EXT_UNKNOWN;
+		if (find_extern_type(btf, btf_array(t)->type, NULL) != EXT_CHAR)
+			return EXT_UNKNOWN;
+		return EXT_CHAR_ARR;
+	default:
+		return EXT_UNKNOWN;
+	}
+}
+
+static int cmp_externs(const void *_a, const void *_b)
+{
+	const struct extern_desc *a = _a;
+	const struct extern_desc *b = _b;
+
+	/* descending order by alignment requirements */
+	if (a->align != b->align)
+		return a->align > b->align ? -1 : 1;
+	/* ascending order by size, within same alignment class */
+	if (a->sz != b->sz)
+		return a->sz < b->sz ? -1 : 1;
+	/* resolve ties by name */
+	return strcmp(a->name, b->name);
+}
+
+static int bpf_object__collect_externs(struct bpf_object *obj)
+{
+	const struct btf_type *t;
+	struct extern_desc *ext;
+	int i, n, off, btf_id;
+	struct btf_type *sec;
+	const char *ext_name;
+	Elf_Scn *scn;
+	GElf_Shdr sh;
+
+	if (!obj->efile.symbols)
+		return 0;
+
+	scn = elf_getscn(obj->efile.elf, obj->efile.symbols_shndx);
+	if (!scn)
+		return -LIBBPF_ERRNO__FORMAT;
+	if (gelf_getshdr(scn, &sh) != &sh)
+		return -LIBBPF_ERRNO__FORMAT;
+	n = sh.sh_size / sh.sh_entsize;
+
+	pr_debug("looking for externs among %d symbols...\n", n);
+	for (i = 0; i < n; i++) {
+		GElf_Sym sym;
+
+		if (!gelf_getsym(obj->efile.symbols, i, &sym))
+			return -LIBBPF_ERRNO__FORMAT;
+		if (!sym_is_extern(&sym))
+			continue;
+		ext_name = elf_strptr(obj->efile.elf, obj->efile.strtabidx,
+				      sym.st_name);
+		if (!ext_name || !ext_name[0])
+			continue;
+
+		ext = obj->externs;
+		ext = reallocarray(ext, obj->nr_extern + 1, sizeof(*ext));
+		if (!ext)
+			return -ENOMEM;
+		obj->externs = ext;
+		ext = &ext[obj->nr_extern];
+		memset(ext, 0, sizeof(*ext));
+		obj->nr_extern++;
+
+		ext->btf_id = find_extern_btf_id(obj->btf, ext_name);
+		if (ext->btf_id <= 0) {
+			pr_warn("failed to find BTF for extern '%s': %d\n",
+				ext_name, ext->btf_id);
+			return ext->btf_id;
+		}
+		t = btf__type_by_id(obj->btf, ext->btf_id);
+		ext->name = btf__name_by_offset(obj->btf, t->name_off);
+		ext->sym_idx = i;
+		ext->is_weak = GELF_ST_BIND(sym.st_info) == STB_WEAK;
+		ext->sz = btf__resolve_size(obj->btf, t->type);
+		if (ext->sz <= 0) {
+			pr_warn("failed to resolve size of extern '%s': %d\n",
+				ext_name, ext->sz);
+			return ext->sz;
+		}
+		ext->align = btf__align_of(obj->btf, t->type);
+		if (ext->align <= 0) {
+			pr_warn("failed to determine alignment of extern '%s': %d\n",
+				ext_name, ext->align);
+			return -EINVAL;
+		}
+		ext->type = find_extern_type(obj->btf, t->type,
+					     &ext->is_signed);
+		if (ext->type == EXT_UNKNOWN) {
+			pr_warn("extern '%s' type is unsupported\n", ext_name);
+			return -ENOTSUP;
+		}
+	}
+	pr_debug("collected %d externs total\n", obj->nr_extern);
+
+	if (!obj->nr_extern)
+		return 0;
+
+	/* sort externs by (alignment, size, name) and calculate their offsets
+	 * within a map */
+	qsort(obj->externs, obj->nr_extern, sizeof(*ext), cmp_externs);
+	off = 0;
+	for (i = 0; i < obj->nr_extern; i++) {
+		ext = &obj->externs[i];
+		ext->data_off = roundup(off, ext->align);
+		off = ext->data_off + ext->sz;
+		pr_debug("extern #%d: symbol %d, off %u, name %s\n",
+			 i, ext->sym_idx, ext->data_off, ext->name);
+	}
+
+	btf_id = btf__find_by_name(obj->btf, EXTERN_SEC);
+	if (btf_id <= 0) {
+		pr_warn("no BTF info found for '%s' datasec\n", EXTERN_SEC);
+		return -ESRCH;
+	}
+
+	sec = (struct btf_type *)btf__type_by_id(obj->btf, btf_id);
+	sec->size = off;
+	n = btf_vlen(sec);
+	for (i = 0; i < n; i++) {
+		struct btf_var_secinfo *vs = btf_var_secinfos(sec) + i;
+
+		t = btf__type_by_id(obj->btf, vs->type);
+		ext_name = btf__name_by_offset(obj->btf, t->name_off);
+		ext = find_extern_by_name(obj, ext_name);
+		if (!ext) {
+			pr_warn("failed to find extern definition for BTF var '%s'\n",
+				ext_name);
+			return -ESRCH;
+		}
+		vs->offset = ext->data_off;
+		btf_var(t)->linkage = BTF_VAR_GLOBAL_ALLOCATED;
+	}
+
+	return 0;
+}
+
 static struct bpf_program *
 bpf_object__find_prog_by_idx(struct bpf_object *obj, int idx)
 {
@@ -1801,6 +2333,8 @@ bpf_object__section_to_libbpf_map_type(const struct bpf_object *obj, int shndx)
 		return LIBBPF_MAP_BSS;
 	else if (shndx == obj->efile.rodata_shndx)
 		return LIBBPF_MAP_RODATA;
+	else if (shndx == obj->efile.symbols_shndx)
+		return LIBBPF_MAP_EXTERN;
 	else
 		return LIBBPF_MAP_UNSPEC;
 }
@@ -1845,6 +2379,30 @@ static int bpf_program__record_reloc(struct bpf_program *prog,
 			insn_idx, insn->code);
 		return -LIBBPF_ERRNO__RELOC;
 	}
+
+	if (sym_is_extern(sym)) {
+		int sym_idx = GELF_R_SYM(rel->r_info);
+		int i, n = obj->nr_extern;
+		struct extern_desc *ext;
+
+		for (i = 0; i < n; i++) {
+			ext = &obj->externs[i];
+			if (ext->sym_idx == sym_idx)
+				break;
+		}
+		if (i >= n) {
+			pr_warn("extern relo failed to find extern for sym %d\n",
+				sym_idx);
+			return -LIBBPF_ERRNO__RELOC;
+		}
+		pr_debug("found extern #%d '%s' (sym %d, off %u) for insn %u\n",
+			 i, ext->name, ext->sym_idx, ext->data_off, insn_idx);
+		reloc_desc->type = RELO_EXTERN;
+		reloc_desc->insn_idx = insn_idx;
+		reloc_desc->sym_off = ext->data_off;
+		return 0;
+	}
+
 	if (!shdr_idx || shdr_idx >= SHN_LORESERVE) {
 		pr_warn("invalid relo for \'%s\' in special section 0x%x; forgot to initialize global var?..\n",
 			name, shdr_idx);
@@ -2306,11 +2864,12 @@ bpf_object__reuse_map(struct bpf_map *map)
 static int
 bpf_object__populate_internal_map(struct bpf_object *obj, struct bpf_map *map)
 {
+	enum libbpf_map_type map_type = map->libbpf_type;
 	char *cp, errmsg[STRERR_BUFSIZE];
 	int err, zero = 0;
 
-	/* Nothing to do here since kernel already zero-initializes .bss map. */
-	if (map->libbpf_type == LIBBPF_MAP_BSS)
+	/* kernel already zero-initializes .bss map. */
+	if (map_type == LIBBPF_MAP_BSS)
 		return 0;
 
 	err = bpf_map_update_elem(map->fd, &zero, map->mmaped, 0);
@@ -2322,8 +2881,8 @@ bpf_object__populate_internal_map(struct bpf_object *obj, struct bpf_map *map)
 		return err;
 	}
 
-	/* Freeze .rodata map as read-only from syscall side. */
-	if (map->libbpf_type == LIBBPF_MAP_RODATA) {
+	/* Freeze .rodata and .extern map as read-only from syscall side. */
+	if (map_type == LIBBPF_MAP_RODATA || map_type == LIBBPF_MAP_EXTERN) {
 		err = bpf_map_freeze(map->fd);
 		if (err) {
 			err = -errno;
@@ -3572,9 +4131,6 @@ bpf_program__reloc_text(struct bpf_program *prog, struct bpf_object *obj,
 	size_t new_cnt;
 	int err;
 
-	if (relo->type != RELO_CALL)
-		return -LIBBPF_ERRNO__RELOC;
-
 	if (prog->idx == obj->efile.text_shndx) {
 		pr_warn("relo in .text insn %d into off %d (insn #%d)\n",
 			relo->insn_idx, relo->sym_off, relo->sym_off / 8);
@@ -3636,27 +4192,37 @@ bpf_program__relocate(struct bpf_program *prog, struct bpf_object *obj)
 
 	for (i = 0; i < prog->nr_reloc; i++) {
 		struct reloc_desc *relo = &prog->reloc_desc[i];
+		struct bpf_insn *insn = &prog->insns[relo->insn_idx];
 
-		if (relo->type == RELO_LD64 || relo->type == RELO_DATA) {
-			struct bpf_insn *insn = &prog->insns[relo->insn_idx];
-
-			if (relo->insn_idx + 1 >= (int)prog->insns_cnt) {
-				pr_warn("relocation out of range: '%s'\n",
-					prog->section_name);
-				return -LIBBPF_ERRNO__RELOC;
-			}
+		if (relo->insn_idx + 1 >= (int)prog->insns_cnt) {
+			pr_warn("relocation out of range: '%s'\n",
+				prog->section_name);
+			return -LIBBPF_ERRNO__RELOC;
+		}
 
-			if (relo->type != RELO_DATA) {
-				insn[0].src_reg = BPF_PSEUDO_MAP_FD;
-			} else {
-				insn[0].src_reg = BPF_PSEUDO_MAP_VALUE;
-				insn[1].imm = insn[0].imm + relo->sym_off;
-			}
+		switch (relo->type) {
+		case RELO_LD64:
+			insn[0].src_reg = BPF_PSEUDO_MAP_FD;
+			insn[0].imm = obj->maps[relo->map_idx].fd;
+			break;
+		case RELO_DATA:
+			insn[0].src_reg = BPF_PSEUDO_MAP_VALUE;
+			insn[1].imm = insn[0].imm + relo->sym_off;
 			insn[0].imm = obj->maps[relo->map_idx].fd;
-		} else if (relo->type == RELO_CALL) {
+			break;
+		case RELO_EXTERN:
+			insn[0].src_reg = BPF_PSEUDO_MAP_VALUE;
+			insn[0].imm = obj->maps[obj->extern_map_idx].fd;
+			insn[1].imm = relo->sym_off;
+			break;
+		case RELO_CALL:
 			err = bpf_program__reloc_text(prog, obj, relo);
 			if (err)
 				return err;
+			break;
+		default:
+			pr_warn("relo #%d: bad relo type %d\n", i, relo->type);
+			return -EINVAL;
 		}
 	}
 
@@ -3936,11 +4502,10 @@ static struct bpf_object *
 __bpf_object__open(const char *path, const void *obj_buf, size_t obj_buf_sz,
 		   const struct bpf_object_open_opts *opts)
 {
+	const char *obj_name, *kconfig_path;
 	struct bpf_program *prog;
 	struct bpf_object *obj;
-	const char *obj_name;
 	char tmp_name[64];
-	__u32 attach_prog_fd;
 	int err;
 
 	if (elf_version(EV_CURRENT) == EV_NONE) {
@@ -3969,11 +4534,18 @@ __bpf_object__open(const char *path, const void *obj_buf, size_t obj_buf_sz,
 		return obj;
 
 	obj->relaxed_core_relocs = OPTS_GET(opts, relaxed_core_relocs, false);
-	attach_prog_fd = OPTS_GET(opts, attach_prog_fd, 0);
+	kconfig_path = OPTS_GET(opts, kconfig_path, NULL);
+	if (kconfig_path) {
+		obj->kconfig_path = strdup(kconfig_path);
+		if (!obj->kconfig_path)
+			return ERR_PTR(-ENOMEM);
+	}
 
 	err = bpf_object__elf_init(obj);
 	err = err ? : bpf_object__check_endianness(obj);
 	err = err ? : bpf_object__elf_collect(obj);
+	err = err ? : bpf_object__collect_externs(obj);
+	err = err ? : bpf_object__finalize_btf(obj);
 	err = err ? : bpf_object__init_maps(obj, opts);
 	err = err ? : bpf_object__init_prog_names(obj);
 	err = err ? : bpf_object__collect_reloc(obj);
@@ -3996,7 +4568,7 @@ __bpf_object__open(const char *path, const void *obj_buf, size_t obj_buf_sz,
 		bpf_program__set_type(prog, prog_type);
 		bpf_program__set_expected_attach_type(prog, attach_type);
 		if (prog_type == BPF_PROG_TYPE_TRACING)
-			prog->attach_prog_fd = attach_prog_fd;
+			prog->attach_prog_fd = OPTS_GET(opts, attach_prog_fd, 0);
 	}
 
 	return obj;
@@ -4107,6 +4679,61 @@ static int bpf_object__sanitize_maps(struct bpf_object *obj)
 	return 0;
 }
 
+static int bpf_object__resolve_externs(struct bpf_object *obj,
+				       const char *config_path)
+{
+	bool need_config = false;
+	struct extern_desc *ext;
+	int err, i;
+	void *data;
+
+	if (obj->nr_extern == 0)
+		return 0;
+
+	data = obj->maps[obj->extern_map_idx].mmaped;
+
+	for (i = 0; i < obj->nr_extern; i++) {
+		ext = &obj->externs[i];
+
+		if (strcmp(ext->name, "LINUX_KERNEL_VERSION") == 0) {
+			void *ext_val = data + ext->data_off;
+			__u32 kver = get_kernel_version();
+
+			if (!kver) {
+				pr_warn("failed to get kernel version\n");
+				return -EINVAL;
+			}
+			err = set_ext_value_num(ext, ext_val, kver);
+			if (err)
+				return err;
+			pr_debug("extern %s=0x%x\n", ext->name, kver);
+		} else if (strncmp(ext->name, "CONFIG_", 7) == 0) {
+			need_config = true;
+		} else {
+			pr_warn("unrecognized extern '%s'\n", ext->name);
+			return -EINVAL;
+		}
+	}
+	if (need_config) {
+		err = bpf_object__read_kernel_config(obj, config_path, data);
+		if (err)
+			return -EINVAL;
+	}
+	for (i = 0; i < obj->nr_extern; i++) {
+		ext = &obj->externs[i];
+
+		if (!ext->is_set && !ext->is_weak) {
+			pr_warn("extern %s (strong) not resolved\n", ext->name);
+			return -ESRCH;
+		} else if (!ext->is_set) {
+			pr_debug("extern %s (weak) not resolved, defaulting to zero\n",
+				 ext->name);
+		}
+	}
+
+	return 0;
+}
+
 int bpf_object__load_xattr(struct bpf_object_load_attr *attr)
 {
 	struct bpf_object *obj;
@@ -4126,6 +4753,7 @@ int bpf_object__load_xattr(struct bpf_object_load_attr *attr)
 	obj->loaded = true;
 
 	err = bpf_object__probe_caps(obj);
+	err = err ? : bpf_object__resolve_externs(obj, obj->kconfig_path);
 	err = err ? : bpf_object__sanitize_and_load_btf(obj);
 	err = err ? : bpf_object__sanitize_maps(obj);
 	err = err ? : bpf_object__create_maps(obj);
@@ -4719,6 +5347,10 @@ void bpf_object__close(struct bpf_object *obj)
 		zfree(&map->pin_path);
 	}
 
+	zfree(&obj->kconfig_path);
+	zfree(&obj->externs);
+	obj->nr_extern = 0;
+
 	zfree(&obj->maps);
 	obj->nr_maps = 0;
 
diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index 623191e71415..6340823871e2 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -85,8 +85,12 @@ struct bpf_object_open_opts {
 	 */
 	const char *pin_root_path;
 	__u32 attach_prog_fd;
+	/* kernel config file path override (for CONFIG_ externs); can point
+	 * to either uncompressed text file or .gz file
+	 */
+	const char *kconfig_path;
 };
-#define bpf_object_open_opts__last_field attach_prog_fd
+#define bpf_object_open_opts__last_field kconfig_path
 
 LIBBPF_API struct bpf_object *bpf_object__open(const char *path);
 LIBBPF_API struct bpf_object *
@@ -669,6 +673,12 @@ LIBBPF_API int bpf_object__attach_skeleton(struct bpf_object_skeleton *s);
 LIBBPF_API void bpf_object__detach_skeleton(struct bpf_object_skeleton *s);
 LIBBPF_API void bpf_object__destroy_skeleton(struct bpf_object_skeleton *s);
 
+enum libbpf_tristate {
+	TRI_NO = 0,
+	TRI_YES = 1,
+	TRI_MODULE = 2,
+};
+
 #ifdef __cplusplus
 } /* extern "C" */
 #endif
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index f70c8e735120..ba90621617a8 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -24,7 +24,7 @@ CFLAGS += -g -Wall -O2 $(GENFLAGS) -I$(APIDIR) -I$(LIBDIR) -I$(BPFDIR)	\
 	  -I$(GENDIR) -I$(TOOLSINCDIR) -I$(CURDIR)			\
 	  -Dbpf_prog_load=bpf_prog_test_load				\
 	  -Dbpf_load_program=bpf_test_load_program
-LDLIBS += -lcap -lelf -lrt -lpthread
+LDLIBS += -lcap -lelf -lz -lrt -lpthread
 
 # Order correspond to 'make run_tests' order
 TEST_GEN_PROGS = test_verifier test_tag test_maps test_lru_map test_lpm_map test_progs \
-- 
cgit v1.2.3-59-g8ed1b


From 2ad97d473db57ab866f0756806bb94515f7f2551 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Fri, 13 Dec 2019 17:47:09 -0800
Subject: bpftool: Generate externs datasec in BPF skeleton

Add support for generation of mmap()-ed read-only view of libbpf-provided
extern variables. As externs are not supposed to be provided by user code
(that's what .data, .bss, and .rodata is for), don't mmap() it initially. Only
after skeleton load is performed, map .extern contents as read-only memory.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191214014710.3449601-4-andriin@fb.com
---
 tools/bpf/bpftool/gen.c |  4 ++++
 tools/lib/bpf/libbpf.c  | 10 +++++-----
 2 files changed, 9 insertions(+), 5 deletions(-)

diff --git a/tools/bpf/bpftool/gen.c b/tools/bpf/bpftool/gen.c
index 7379dae35dca..a07c80429c7a 100644
--- a/tools/bpf/bpftool/gen.c
+++ b/tools/bpf/bpftool/gen.c
@@ -82,6 +82,8 @@ static const char *get_map_ident(const struct bpf_map *map)
 		return "rodata";
 	else if (str_has_suffix(name, ".bss"))
 		return "bss";
+	else if (str_has_suffix(name, ".extern"))
+		return "externs"; /* extern is a C keyword */
 	else
 		return NULL;
 }
@@ -109,6 +111,8 @@ static int codegen_datasec_def(struct bpf_object *obj,
 		sec_ident = "bss";
 	else if (strcmp(sec_name, ".rodata") == 0)
 		sec_ident = "rodata";
+	else if (strcmp(sec_name, ".extern") == 0)
+		sec_ident = "externs"; /* extern is a C keyword */
 	else
 		return 0;
 
diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 26144706ba0b..c9f90bfd2f0d 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -7472,7 +7472,8 @@ int bpf_object__open_skeleton(struct bpf_object_skeleton *s,
 			return -ESRCH;
 		}
 
-		if (mmaped)
+		/* externs shouldn't be pre-setup from user code */
+		if (mmaped && (*map)->libbpf_type != LIBBPF_MAP_EXTERN)
 			*mmaped = (*map)->mmaped;
 	}
 
@@ -7505,7 +7506,6 @@ int bpf_object__load_skeleton(struct bpf_object_skeleton *s)
 		size_t mmap_sz = bpf_map_mmap_sz(map);
 		int prot, map_fd = bpf_map__fd(map);
 		void **mmaped = s->maps[i].mmaped;
-		void *remapped;
 
 		if (!mmaped)
 			continue;
@@ -7530,9 +7530,9 @@ int bpf_object__load_skeleton(struct bpf_object_skeleton *s)
 		 * as per normal clean up procedure, so we don't need to worry
 		 * about it from skeleton's clean up perspective.
 		 */
-		remapped = mmap(*mmaped, mmap_sz, prot, MAP_SHARED | MAP_FIXED,
-				map_fd, 0);
-		if (remapped == MAP_FAILED) {
+		*mmaped = mmap(map->mmaped, mmap_sz, prot,
+				MAP_SHARED | MAP_FIXED, map_fd, 0);
+		if (*mmaped == MAP_FAILED) {
 			err = -errno;
 			*mmaped = NULL;
 			pr_warn("failed to re-mmap() map '%s': %d\n",
-- 
cgit v1.2.3-59-g8ed1b


From 330a73a7b6ca93a415de1b7da68d7a0698fe4937 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Fri, 13 Dec 2019 17:47:10 -0800
Subject: selftests/bpf: Add tests for libbpf-provided externs

Add a set of tests validating libbpf-provided extern variables. One crucial
feature that's tested is dead code elimination together with using invalid BPF
helper. CONFIG_MISSING is not supposed to exist and should always be specified
by libbpf as zero, which allows BPF verifier to correctly do branch pruning
and not fail validation, when invalid BPF helper is called from dead if branch.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191214014710.3449601-5-andriin@fb.com
---
 .../testing/selftests/bpf/prog_tests/core_extern.c | 195 +++++++++++++++++++++
 tools/testing/selftests/bpf/prog_tests/skeleton.c  |  18 +-
 .../testing/selftests/bpf/progs/test_core_extern.c |  62 +++++++
 tools/testing/selftests/bpf/progs/test_skeleton.c  |   9 +
 4 files changed, 283 insertions(+), 1 deletion(-)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/core_extern.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_core_extern.c

diff --git a/tools/testing/selftests/bpf/prog_tests/core_extern.c b/tools/testing/selftests/bpf/prog_tests/core_extern.c
new file mode 100644
index 000000000000..30a7972e9012
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/core_extern.c
@@ -0,0 +1,195 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2019 Facebook */
+
+#include <test_progs.h>
+#include <sys/mman.h>
+#include <sys/utsname.h>
+#include <linux/version.h>
+#include "test_core_extern.skel.h"
+
+static uint32_t get_kernel_version(void)
+{
+	uint32_t major, minor, patch;
+	struct utsname info;
+
+	uname(&info);
+	if (sscanf(info.release, "%u.%u.%u", &major, &minor, &patch) != 3)
+		return 0;
+	return KERNEL_VERSION(major, minor, patch);
+}
+
+#define CFG "CONFIG_BPF_SYSCALL=n\n"
+
+static struct test_case {
+	const char *name;
+	const char *cfg;
+	const char *cfg_path;
+	bool fails;
+	struct test_core_extern__data data;
+} test_cases[] = {
+	{ .name = "default search path", .cfg_path = NULL,
+	  .data = { .bpf_syscall = true } },
+	{ .name = "/proc/config.gz", .cfg_path = "/proc/config.gz",
+	  .data = { .bpf_syscall = true } },
+	{ .name = "missing config", .fails = true,
+	  .cfg_path = "/proc/invalid-config.gz" },
+	{
+		.name = "custom values",
+		.cfg = "CONFIG_BPF_SYSCALL=y\n"
+		       "CONFIG_TRISTATE=m\n"
+		       "CONFIG_BOOL=y\n"
+		       "CONFIG_CHAR=100\n"
+		       "CONFIG_USHORT=30000\n"
+		       "CONFIG_INT=123456\n"
+		       "CONFIG_ULONG=0xDEADBEEFC0DE\n"
+		       "CONFIG_STR=\"abracad\"\n"
+		       "CONFIG_MISSING=0",
+		.data = {
+			.bpf_syscall = true,
+			.tristate_val = TRI_MODULE,
+			.bool_val = true,
+			.char_val = 100,
+			.ushort_val = 30000,
+			.int_val = 123456,
+			.ulong_val = 0xDEADBEEFC0DE,
+			.str_val = "abracad",
+		},
+	},
+	/* TRISTATE */
+	{ .name = "tristate (y)", .cfg = CFG"CONFIG_TRISTATE=y\n",
+	  .data = { .tristate_val = TRI_YES } },
+	{ .name = "tristate (n)", .cfg = CFG"CONFIG_TRISTATE=n\n",
+	  .data = { .tristate_val = TRI_NO } },
+	{ .name = "tristate (m)", .cfg = CFG"CONFIG_TRISTATE=m\n",
+	  .data = { .tristate_val = TRI_MODULE } },
+	{ .name = "tristate (int)", .fails = 1, .cfg = CFG"CONFIG_TRISTATE=1" },
+	{ .name = "tristate (bad)", .fails = 1, .cfg = CFG"CONFIG_TRISTATE=M" },
+	/* BOOL */
+	{ .name = "bool (y)", .cfg = CFG"CONFIG_BOOL=y\n",
+	  .data = { .bool_val = true } },
+	{ .name = "bool (n)", .cfg = CFG"CONFIG_BOOL=n\n",
+	  .data = { .bool_val = false } },
+	{ .name = "bool (tristate)", .fails = 1, .cfg = CFG"CONFIG_BOOL=m" },
+	{ .name = "bool (int)", .fails = 1, .cfg = CFG"CONFIG_BOOL=1" },
+	/* CHAR */
+	{ .name = "char (tristate)", .cfg = CFG"CONFIG_CHAR=m\n",
+	  .data = { .char_val = 'm' } },
+	{ .name = "char (bad)", .fails = 1, .cfg = CFG"CONFIG_CHAR=q\n" },
+	{ .name = "char (empty)", .fails = 1, .cfg = CFG"CONFIG_CHAR=\n" },
+	{ .name = "char (str)", .fails = 1, .cfg = CFG"CONFIG_CHAR=\"y\"\n" },
+	/* STRING */
+	{ .name = "str (empty)", .cfg = CFG"CONFIG_STR=\"\"\n",
+	  .data = { .str_val = "\0\0\0\0\0\0\0" } },
+	{ .name = "str (padded)", .cfg = CFG"CONFIG_STR=\"abra\"\n",
+	  .data = { .str_val = "abra\0\0\0" } },
+	{ .name = "str (too long)", .cfg = CFG"CONFIG_STR=\"abracada\"\n",
+	  .data = { .str_val = "abracad" } },
+	{ .name = "str (no value)", .fails = 1, .cfg = CFG"CONFIG_STR=\n" },
+	{ .name = "str (bad value)", .fails = 1, .cfg = CFG"CONFIG_STR=bla\n" },
+	/* INTEGERS */
+	{
+		.name = "integer forms",
+		.cfg = CFG
+		       "CONFIG_CHAR=0xA\n"
+		       "CONFIG_USHORT=0462\n"
+		       "CONFIG_INT=-100\n"
+		       "CONFIG_ULONG=+1000000000000",
+		.data = {
+			.char_val = 0xA,
+			.ushort_val = 0462,
+			.int_val = -100,
+			.ulong_val = 1000000000000,
+		},
+	},
+	{ .name = "int (bad)", .fails = 1, .cfg = CFG"CONFIG_INT=abc" },
+	{ .name = "int (str)", .fails = 1, .cfg = CFG"CONFIG_INT=\"abc\"" },
+	{ .name = "int (empty)", .fails = 1, .cfg = CFG"CONFIG_INT=" },
+	{ .name = "int (mixed)", .fails = 1, .cfg = CFG"CONFIG_INT=123abc" },
+	{ .name = "int (max)", .cfg = CFG"CONFIG_INT=2147483647",
+	  .data = { .int_val = 2147483647 } },
+	{ .name = "int (min)", .cfg = CFG"CONFIG_INT=-2147483648",
+	  .data = { .int_val = -2147483648 } },
+	{ .name = "int (max+1)", .fails = 1, .cfg = CFG"CONFIG_INT=2147483648" },
+	{ .name = "int (min-1)", .fails = 1, .cfg = CFG"CONFIG_INT=-2147483649" },
+	{ .name = "ushort (max)", .cfg = CFG"CONFIG_USHORT=65535",
+	  .data = { .ushort_val = 65535 } },
+	{ .name = "ushort (min)", .cfg = CFG"CONFIG_USHORT=0",
+	  .data = { .ushort_val = 0 } },
+	{ .name = "ushort (max+1)", .fails = 1, .cfg = CFG"CONFIG_USHORT=65536" },
+	{ .name = "ushort (min-1)", .fails = 1, .cfg = CFG"CONFIG_USHORT=-1" },
+	{ .name = "u64 (max)", .cfg = CFG"CONFIG_ULONG=0xffffffffffffffff",
+	  .data = { .ulong_val = 0xffffffffffffffff } },
+	{ .name = "u64 (min)", .cfg = CFG"CONFIG_ULONG=0",
+	  .data = { .ulong_val = 0 } },
+	{ .name = "u64 (max+1)", .fails = 1, .cfg = CFG"CONFIG_ULONG=0x10000000000000000" },
+};
+
+BPF_EMBED_OBJ(core_extern, "test_core_extern.o");
+
+void test_core_extern(void)
+{
+	const uint32_t kern_ver = get_kernel_version();
+	int err, duration = 0, i, j;
+	struct test_core_extern *skel = NULL;
+	uint64_t *got, *exp;
+	int n = sizeof(*skel->data) / sizeof(uint64_t);
+
+	for (i = 0; i < ARRAY_SIZE(test_cases); i++) {
+		char tmp_cfg_path[] = "/tmp/test_core_extern_cfg.XXXXXX";
+		struct test_case *t = &test_cases[i];
+		DECLARE_LIBBPF_OPTS(bpf_object_open_opts, opts,
+			.kconfig_path = t->cfg_path,
+		);
+
+		if (!test__start_subtest(t->name))
+			continue;
+
+		if (t->cfg) {
+			size_t n = strlen(t->cfg) + 1;
+			int fd = mkstemp(tmp_cfg_path);
+			int written;
+
+			if (CHECK(fd < 0, "mkstemp", "errno: %d\n", errno))
+				continue;
+			printf("using '%s' as config file\n", tmp_cfg_path);
+			written = write(fd, t->cfg, n);
+			close(fd);
+			if (CHECK_FAIL(written != n))
+				goto cleanup;
+			opts.kconfig_path = tmp_cfg_path;
+		}
+
+		skel = test_core_extern__open_opts(&core_extern_embed, &opts);
+		if (CHECK(!skel, "skel_open", "skeleton open failed\n"))
+			goto cleanup;
+		err = test_core_extern__load(skel);
+		if (t->fails) {
+			CHECK(!err, "skel_load",
+			      "shouldn't succeed open/load of skeleton\n");
+			goto cleanup;
+		} else if (CHECK(err, "skel_load",
+				 "failed to open/load skeleton\n")) {
+			goto cleanup;
+		}
+		err = test_core_extern__attach(skel);
+		if (CHECK(err, "attach_raw_tp", "failed attach: %d\n", err))
+			goto cleanup;
+
+		usleep(1);
+
+		t->data.kern_ver = kern_ver;
+		t->data.missing_val = 0xDEADC0DE;
+		got = (uint64_t *)skel->data;
+		exp = (uint64_t *)&t->data;
+		for (j = 0; j < n; j++) {
+			CHECK(got[j] != exp[j], "check_res",
+			      "result #%d: expected %lx, but got %lx\n",
+			       j, exp[j], got[j]);
+		}
+cleanup:
+		if (t->cfg)
+			unlink(tmp_cfg_path);
+		test_core_extern__destroy(skel);
+		skel = NULL;
+	}
+}
diff --git a/tools/testing/selftests/bpf/prog_tests/skeleton.c b/tools/testing/selftests/bpf/prog_tests/skeleton.c
index 79f8d13e6740..151cdad3ad0d 100644
--- a/tools/testing/selftests/bpf/prog_tests/skeleton.c
+++ b/tools/testing/selftests/bpf/prog_tests/skeleton.c
@@ -17,11 +17,21 @@ void test_skeleton(void)
 	int duration = 0, err;
 	struct test_skeleton* skel;
 	struct test_skeleton__bss *bss;
+	struct test_skeleton__externs *exts;
 
-	skel = test_skeleton__open_and_load(&skeleton_embed);
+	skel = test_skeleton__open(&skeleton_embed);
 	if (CHECK(!skel, "skel_open", "failed to open skeleton\n"))
 		return;
 
+	printf("EXTERNS BEFORE: %p\n", skel->externs);
+	if (CHECK(skel->externs, "skel_externs", "externs are mmaped()!\n"))
+		goto cleanup;
+
+	err = test_skeleton__load(skel);
+	if (CHECK(err, "skel_load", "failed to load skeleton: %d\n", err))
+		goto cleanup;
+	printf("EXTERNS AFTER: %p\n", skel->externs);
+
 	bss = skel->bss;
 	bss->in1 = 1;
 	bss->in2 = 2;
@@ -29,6 +39,7 @@ void test_skeleton(void)
 	bss->in4 = 4;
 	bss->in5.a = 5;
 	bss->in5.b = 6;
+	exts = skel->externs;
 
 	err = test_skeleton__attach(skel);
 	if (CHECK(err, "skel_attach", "skeleton attach failed: %d\n", err))
@@ -46,6 +57,11 @@ void test_skeleton(void)
 	CHECK(bss->handler_out5.b != 6, "res6", "got %lld != exp %d\n",
 	      bss->handler_out5.b, 6);
 
+	CHECK(bss->bpf_syscall != exts->CONFIG_BPF_SYSCALL, "ext1",
+	      "got %d != exp %d\n", bss->bpf_syscall, exts->CONFIG_BPF_SYSCALL);
+	CHECK(bss->kern_ver != exts->LINUX_KERNEL_VERSION, "ext2",
+	      "got %d != exp %d\n", bss->kern_ver, exts->LINUX_KERNEL_VERSION);
+
 cleanup:
 	test_skeleton__destroy(skel);
 }
diff --git a/tools/testing/selftests/bpf/progs/test_core_extern.c b/tools/testing/selftests/bpf/progs/test_core_extern.c
new file mode 100644
index 000000000000..e12f09f9e881
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_core_extern.c
@@ -0,0 +1,62 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2019 Facebook */
+
+#include <stdint.h>
+#include <stdbool.h>
+#include <linux/ptrace.h>
+#include <linux/bpf.h>
+#include "bpf_helpers.h"
+
+/* non-existing BPF helper, to test dead code elimination */
+static int (*bpf_missing_helper)(const void *arg1, int arg2) = (void *) 999;
+
+extern int LINUX_KERNEL_VERSION;
+extern bool CONFIG_BPF_SYSCALL; /* strong */
+extern enum libbpf_tristate CONFIG_TRISTATE __weak;
+extern bool CONFIG_BOOL __weak;
+extern char CONFIG_CHAR __weak;
+extern uint16_t CONFIG_USHORT __weak;
+extern int CONFIG_INT __weak;
+extern uint64_t CONFIG_ULONG __weak;
+extern const char CONFIG_STR[8] __weak;
+extern uint64_t CONFIG_MISSING __weak;
+
+uint64_t kern_ver = -1;
+uint64_t bpf_syscall = -1;
+uint64_t tristate_val = -1;
+uint64_t bool_val = -1;
+uint64_t char_val = -1;
+uint64_t ushort_val = -1;
+uint64_t int_val = -1;
+uint64_t ulong_val = -1;
+char str_val[8] = {-1, -1, -1, -1, -1, -1, -1, -1};
+uint64_t missing_val = -1;
+
+SEC("raw_tp/sys_enter")
+int handle_sys_enter(struct pt_regs *ctx)
+{
+	int i;
+
+	kern_ver = LINUX_KERNEL_VERSION;
+	bpf_syscall = CONFIG_BPF_SYSCALL;
+	tristate_val = CONFIG_TRISTATE;
+	bool_val = CONFIG_BOOL;
+	char_val = CONFIG_CHAR;
+	ushort_val = CONFIG_USHORT;
+	int_val = CONFIG_INT;
+	ulong_val = CONFIG_ULONG;
+
+	for (i = 0; i < sizeof(CONFIG_STR); i++) {
+		str_val[i] = CONFIG_STR[i];
+	}
+
+	if (CONFIG_MISSING)
+		/* invalid, but dead code - never executed */
+		missing_val = bpf_missing_helper(ctx, 123);
+	else
+		missing_val = 0xDEADC0DE;
+
+	return 0;
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/test_skeleton.c b/tools/testing/selftests/bpf/progs/test_skeleton.c
index c0013d2e16f2..9caa44758ea2 100644
--- a/tools/testing/selftests/bpf/progs/test_skeleton.c
+++ b/tools/testing/selftests/bpf/progs/test_skeleton.c
@@ -1,6 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 /* Copyright (c) 2019 Facebook */
 
+#include <stdbool.h>
 #include <linux/bpf.h>
 #include "bpf_helpers.h"
 
@@ -20,6 +21,10 @@ char out3 = 0;
 long long out4 = 0;
 int out1 = 0;
 
+extern bool CONFIG_BPF_SYSCALL;
+extern int LINUX_KERNEL_VERSION;
+bool bpf_syscall = 0;
+int kern_ver = 0;
 
 SEC("raw_tp/sys_enter")
 int handler(const void *ctx)
@@ -31,6 +36,10 @@ int handler(const void *ctx)
 	out3 = in3;
 	out4 = in4;
 	out5 = in5;
+
+	bpf_syscall = CONFIG_BPF_SYSCALL;
+	kern_ver = LINUX_KERNEL_VERSION;
+
 	return 0;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 1b484b301cec404e8a329f05edf848fc0a5875c0 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Sat, 14 Dec 2019 23:08:43 -0800
Subject: libbpf: Support flexible arrays in CO-RE

Some data stuctures in kernel are defined with either zero-sized array or
flexible (dimensionless) array at the end of a struct. Actual data of such
array follows in memory immediately after the end of that struct, forming its
variable-sized "body" of elements. Support such access pattern in CO-RE
relocation handling.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191215070844.1014385-2-andriin@fb.com
---
 tools/lib/bpf/libbpf.c | 34 +++++++++++++++++++++++++++++-----
 1 file changed, 29 insertions(+), 5 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index c9f90bfd2f0d..a2cc7313763a 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -3108,6 +3108,21 @@ static bool str_is_empty(const char *s)
 	return !s || !s[0];
 }
 
+static bool is_flex_arr(const struct btf *btf,
+			const struct bpf_core_accessor *acc,
+			const struct btf_array *arr)
+{
+	const struct btf_type *t;
+
+	/* not a flexible array, if not inside a struct or has non-zero size */
+	if (!acc->name || arr->nelems > 0)
+		return false;
+
+	/* has to be the last member of enclosing struct */
+	t = btf__type_by_id(btf, acc->type_id);
+	return acc->idx == btf_vlen(t) - 1;
+}
+
 /*
  * Turn bpf_field_reloc into a low- and high-level spec representation,
  * validating correctness along the way, as well as calculating resulting
@@ -3145,6 +3160,7 @@ static int bpf_core_spec_parse(const struct btf *btf,
 			       struct bpf_core_spec *spec)
 {
 	int access_idx, parsed_len, i;
+	struct bpf_core_accessor *acc;
 	const struct btf_type *t;
 	const char *name;
 	__u32 id;
@@ -3192,6 +3208,7 @@ static int bpf_core_spec_parse(const struct btf *btf,
 			return -EINVAL;
 
 		access_idx = spec->raw_spec[i];
+		acc = &spec->spec[spec->len];
 
 		if (btf_is_composite(t)) {
 			const struct btf_member *m;
@@ -3209,18 +3226,23 @@ static int bpf_core_spec_parse(const struct btf *btf,
 				if (str_is_empty(name))
 					return -EINVAL;
 
-				spec->spec[spec->len].type_id = id;
-				spec->spec[spec->len].idx = access_idx;
-				spec->spec[spec->len].name = name;
+				acc->type_id = id;
+				acc->idx = access_idx;
+				acc->name = name;
 				spec->len++;
 			}
 
 			id = m->type;
 		} else if (btf_is_array(t)) {
 			const struct btf_array *a = btf_array(t);
+			bool flex;
 
 			t = skip_mods_and_typedefs(btf, a->type, &id);
-			if (!t || access_idx >= a->nelems)
+			if (!t)
+				return -EINVAL;
+
+			flex = is_flex_arr(btf, acc - 1, a);
+			if (!flex && access_idx >= a->nelems)
 				return -EINVAL;
 
 			spec->spec[spec->len].type_id = id;
@@ -3525,12 +3547,14 @@ static int bpf_core_spec_match(struct bpf_core_spec *local_spec,
 			 */
 			if (i > 0) {
 				const struct btf_array *a;
+				bool flex;
 
 				if (!btf_is_array(targ_type))
 					return 0;
 
 				a = btf_array(targ_type);
-				if (local_acc->idx >= a->nelems)
+				flex = is_flex_arr(targ_btf, targ_acc - 1, a);
+				if (!flex && local_acc->idx >= a->nelems)
 					return 0;
 				if (!skip_mods_and_typedefs(targ_btf, a->type,
 							    &targ_id))
-- 
cgit v1.2.3-59-g8ed1b


From 5f2eeceffb92a0d799b141df7af3d1ac77337dc4 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Sat, 14 Dec 2019 23:08:44 -0800
Subject: selftests/bpf: Add flexible array relocation tests

Add few tests validation CO-RE relocation handling of flexible array accesses.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191215070844.1014385-3-andriin@fb.com
---
 .../testing/selftests/bpf/prog_tests/core_reloc.c  |  4 +++
 .../btf__core_reloc_arrays___equiv_zero_sz_arr.c   |  3 ++
 .../btf__core_reloc_arrays___err_bad_zero_sz_arr.c |  3 ++
 .../bpf/progs/btf__core_reloc_arrays___fixed_arr.c |  3 ++
 .../testing/selftests/bpf/progs/core_reloc_types.h | 39 ++++++++++++++++++++++
 .../selftests/bpf/progs/test_core_reloc_arrays.c   |  8 ++---
 6 files changed, 56 insertions(+), 4 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___equiv_zero_sz_arr.c
 create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_bad_zero_sz_arr.c
 create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___fixed_arr.c

diff --git a/tools/testing/selftests/bpf/prog_tests/core_reloc.c b/tools/testing/selftests/bpf/prog_tests/core_reloc.c
index 05fe85281ff7..31e177adbdf1 100644
--- a/tools/testing/selftests/bpf/prog_tests/core_reloc.c
+++ b/tools/testing/selftests/bpf/prog_tests/core_reloc.c
@@ -74,6 +74,7 @@
 		.b123 = 2,						\
 		.c1c  = 3,						\
 		.d00d = 4,						\
+		.f10c = 0,						\
 	},								\
 	.output_len = sizeof(struct core_reloc_arrays_output)		\
 }
@@ -308,12 +309,15 @@ static struct core_reloc_test_case test_cases[] = {
 	ARRAYS_CASE(arrays),
 	ARRAYS_CASE(arrays___diff_arr_dim),
 	ARRAYS_CASE(arrays___diff_arr_val_sz),
+	ARRAYS_CASE(arrays___equiv_zero_sz_arr),
+	ARRAYS_CASE(arrays___fixed_arr),
 
 	ARRAYS_ERR_CASE(arrays___err_too_small),
 	ARRAYS_ERR_CASE(arrays___err_too_shallow),
 	ARRAYS_ERR_CASE(arrays___err_non_array),
 	ARRAYS_ERR_CASE(arrays___err_wrong_val_type1),
 	ARRAYS_ERR_CASE(arrays___err_wrong_val_type2),
+	ARRAYS_ERR_CASE(arrays___err_bad_zero_sz_arr),
 
 	/* enum/ptr/int handling scenarios */
 	PRIMITIVES_CASE(primitives),
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___equiv_zero_sz_arr.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___equiv_zero_sz_arr.c
new file mode 100644
index 000000000000..65eac371b061
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___equiv_zero_sz_arr.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_arrays___equiv_zero_sz_arr x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_bad_zero_sz_arr.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_bad_zero_sz_arr.c
new file mode 100644
index 000000000000..ecda2b545ac2
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_bad_zero_sz_arr.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_arrays___err_bad_zero_sz_arr x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___fixed_arr.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___fixed_arr.c
new file mode 100644
index 000000000000..fe1d01232c22
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___fixed_arr.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_arrays___fixed_arr x) {}
diff --git a/tools/testing/selftests/bpf/progs/core_reloc_types.h b/tools/testing/selftests/bpf/progs/core_reloc_types.h
index 9311489e14b2..6d598cfbdb3e 100644
--- a/tools/testing/selftests/bpf/progs/core_reloc_types.h
+++ b/tools/testing/selftests/bpf/progs/core_reloc_types.h
@@ -327,6 +327,7 @@ struct core_reloc_arrays_output {
 	char b123;
 	int c1c;
 	int d00d;
+	int f10c;
 };
 
 struct core_reloc_arrays_substruct {
@@ -339,6 +340,7 @@ struct core_reloc_arrays {
 	char b[2][3][4];
 	struct core_reloc_arrays_substruct c[3];
 	struct core_reloc_arrays_substruct d[1][2];
+	struct core_reloc_arrays_substruct f[][2];
 };
 
 /* bigger array dimensions */
@@ -347,6 +349,7 @@ struct core_reloc_arrays___diff_arr_dim {
 	char b[3][4][5];
 	struct core_reloc_arrays_substruct c[4];
 	struct core_reloc_arrays_substruct d[2][3];
+	struct core_reloc_arrays_substruct f[1][3];
 };
 
 /* different size of array's value (struct) */
@@ -363,6 +366,29 @@ struct core_reloc_arrays___diff_arr_val_sz {
 		int d;
 		int __padding2;
 	} d[1][2];
+	struct {
+		int __padding1;
+		int c;
+		int __padding2;
+	} f[][2];
+};
+
+struct core_reloc_arrays___equiv_zero_sz_arr {
+	int a[5];
+	char b[2][3][4];
+	struct core_reloc_arrays_substruct c[3];
+	struct core_reloc_arrays_substruct d[1][2];
+	/* equivalent to flexible array */
+	struct core_reloc_arrays_substruct f[0][2];
+};
+
+struct core_reloc_arrays___fixed_arr {
+	int a[5];
+	char b[2][3][4];
+	struct core_reloc_arrays_substruct c[3];
+	struct core_reloc_arrays_substruct d[1][2];
+	/* not a flexible array anymore, but within access bounds */
+	struct core_reloc_arrays_substruct f[1][2];
 };
 
 struct core_reloc_arrays___err_too_small {
@@ -370,6 +396,7 @@ struct core_reloc_arrays___err_too_small {
 	char b[2][3][4];
 	struct core_reloc_arrays_substruct c[3];
 	struct core_reloc_arrays_substruct d[1][2];
+	struct core_reloc_arrays_substruct f[][2];
 };
 
 struct core_reloc_arrays___err_too_shallow {
@@ -377,6 +404,7 @@ struct core_reloc_arrays___err_too_shallow {
 	char b[2][3]; /* this one lacks one dimension */
 	struct core_reloc_arrays_substruct c[3];
 	struct core_reloc_arrays_substruct d[1][2];
+	struct core_reloc_arrays_substruct f[][2];
 };
 
 struct core_reloc_arrays___err_non_array {
@@ -384,6 +412,7 @@ struct core_reloc_arrays___err_non_array {
 	char b[2][3][4];
 	struct core_reloc_arrays_substruct c[3];
 	struct core_reloc_arrays_substruct d[1][2];
+	struct core_reloc_arrays_substruct f[][2];
 };
 
 struct core_reloc_arrays___err_wrong_val_type {
@@ -391,6 +420,16 @@ struct core_reloc_arrays___err_wrong_val_type {
 	char b[2][3][4];
 	int c[3]; /* value is not a struct */
 	struct core_reloc_arrays_substruct d[1][2];
+	struct core_reloc_arrays_substruct f[][2];
+};
+
+struct core_reloc_arrays___err_bad_zero_sz_arr {
+	/* zero-sized array, but not at the end */
+	struct core_reloc_arrays_substruct f[0][2];
+	int a[5];
+	char b[2][3][4];
+	struct core_reloc_arrays_substruct c[3];
+	struct core_reloc_arrays_substruct d[1][2];
 };
 
 /*
diff --git a/tools/testing/selftests/bpf/progs/test_core_reloc_arrays.c b/tools/testing/selftests/bpf/progs/test_core_reloc_arrays.c
index 89951b684282..053b86f6b53f 100644
--- a/tools/testing/selftests/bpf/progs/test_core_reloc_arrays.c
+++ b/tools/testing/selftests/bpf/progs/test_core_reloc_arrays.c
@@ -18,6 +18,7 @@ struct core_reloc_arrays_output {
 	char b123;
 	int c1c;
 	int d00d;
+	int f01c;
 };
 
 struct core_reloc_arrays_substruct {
@@ -30,6 +31,7 @@ struct core_reloc_arrays {
 	char b[2][3][4];
 	struct core_reloc_arrays_substruct c[3];
 	struct core_reloc_arrays_substruct d[1][2];
+	struct core_reloc_arrays_substruct f[][2];
 };
 
 #define CORE_READ(dst, src) bpf_core_read(dst, sizeof(*(dst)), src)
@@ -40,18 +42,16 @@ int test_core_arrays(void *ctx)
 	struct core_reloc_arrays *in = (void *)&data.in;
 	struct core_reloc_arrays_output *out = (void *)&data.out;
 
-	/* in->a[2] */
 	if (CORE_READ(&out->a2, &in->a[2]))
 		return 1;
-	/* in->b[1][2][3] */
 	if (CORE_READ(&out->b123, &in->b[1][2][3]))
 		return 1;
-	/* in->c[1].c */
 	if (CORE_READ(&out->c1c, &in->c[1].c))
 		return 1;
-	/* in->d[0][0].d */
 	if (CORE_READ(&out->d00d, &in->d[0][0].d))
 		return 1;
+	if (CORE_READ(&out->f01c, &in->f[0][1].c))
+		return 1;
 
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From a79ac2d1036a824abba982c33e938b717d1b659f Mon Sep 17 00:00:00 2001
From: Prashant Bhole <prashantbhole.linux@gmail.com>
Date: Mon, 16 Dec 2019 17:27:38 +0900
Subject: libbpf: Fix build by renaming variables
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

In btf__align_of() variable name 't' is shadowed by inner block
declaration of another variable with same name. Patch renames
variables in order to fix it.

  CC       sharedobjs/btf.o
btf.c: In function ‘btf__align_of’:
btf.c:303:21: error: declaration of ‘t’ shadows a previous local [-Werror=shadow]
  303 |   int i, align = 1, t;
      |                     ^
btf.c:283:25: note: shadowed declaration is here
  283 |  const struct btf_type *t = btf__type_by_id(btf, id);
      |

Fixes: 3d208f4ca111 ("libbpf: Expose btf__align_of() API")
Signed-off-by: Prashant Bhole <prashantbhole.linux@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/bpf/20191216082738.28421-1-prashantbhole.linux@gmail.com
---
 tools/lib/bpf/btf.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
index 520021939d81..5f04f56e1eb6 100644
--- a/tools/lib/bpf/btf.c
+++ b/tools/lib/bpf/btf.c
@@ -300,16 +300,16 @@ int btf__align_of(const struct btf *btf, __u32 id)
 	case BTF_KIND_UNION: {
 		const struct btf_member *m = btf_members(t);
 		__u16 vlen = btf_vlen(t);
-		int i, align = 1, t;
+		int i, max_align = 1, align;
 
 		for (i = 0; i < vlen; i++, m++) {
-			t = btf__align_of(btf, m->type);
-			if (t <= 0)
-				return t;
-			align = max(align, t);
+			align = btf__align_of(btf, m->type);
+			if (align <= 0)
+				return align;
+			max_align = max(max_align, align);
 		}
 
-		return align;
+		return max_align;
 	}
 	default:
 		pr_warn("unsupported BTF_KIND:%u\n", btf_kind(t));
-- 
cgit v1.2.3-59-g8ed1b


From 159ecc002b5f55f9c58de533551a2ee552861185 Mon Sep 17 00:00:00 2001
From: Paul Chaignon <paul.chaignon@orange.com>
Date: Mon, 16 Dec 2019 12:27:33 +0100
Subject: bpftool: Fix compilation warning on shadowed variable

The ident variable has already been declared at the top of the function
and doesn't need to be re-declared.

Fixes: 985ead416df39 ("bpftool: Add skeleton codegen command")
Signed-off-by: Paul Chaignon <paul.chaignon@orange.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191216112733.GA28366@Omicron
---
 tools/bpf/bpftool/gen.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/bpf/bpftool/gen.c b/tools/bpf/bpftool/gen.c
index a07c80429c7a..f70088b4c19b 100644
--- a/tools/bpf/bpftool/gen.c
+++ b/tools/bpf/bpftool/gen.c
@@ -388,7 +388,7 @@ static int do_skeleton(int argc, char **argv)
 		);
 		i = 0;
 		bpf_object__for_each_map(map, obj) {
-			const char *ident = get_map_ident(map);
+			ident = get_map_ident(map);
 
 			if (!ident)
 				continue;
-- 
cgit v1.2.3-59-g8ed1b


From 5984dc6cb5aa6cce342a44f01f984fde09ed05b1 Mon Sep 17 00:00:00 2001
From: Prashant Bhole <prashantbhole.linux@gmail.com>
Date: Mon, 16 Dec 2019 16:16:19 +0900
Subject: samples/bpf: Reintroduce missed build targets

Add xdp_redirect and per_socket_stats_example in build targets.
They got removed from build targets in Makefile reorganization.

Fixes: 1d97c6c2511f ("samples/bpf: Base target programs rules on Makefile.target")
Signed-off-by: Prashant Bhole <prashantbhole.linux@gmail.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191216071619.25479-1-prashantbhole.linux@gmail.com
---
 samples/bpf/Makefile | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile
index 1fc42ad8ff49..8003d2823fa8 100644
--- a/samples/bpf/Makefile
+++ b/samples/bpf/Makefile
@@ -38,6 +38,8 @@ tprogs-y += tc_l2_redirect
 tprogs-y += lwt_len_hist
 tprogs-y += xdp_tx_iptunnel
 tprogs-y += test_map_in_map
+tprogs-y += per_socket_stats_example
+tprogs-y += xdp_redirect
 tprogs-y += xdp_redirect_map
 tprogs-y += xdp_redirect_cpu
 tprogs-y += xdp_monitor
-- 
cgit v1.2.3-59-g8ed1b


From 5615ed472dc04e38a1affca69059f17c1178f770 Mon Sep 17 00:00:00 2001
From: Toke Høiland-Jørgensen <toke@redhat.com>
Date: Mon, 16 Dec 2019 11:24:05 +0100
Subject: samples/bpf: Add missing -lz to TPROGS_LDLIBS
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Since libbpf now links against zlib, this needs to be included in the
linker invocation for the userspace programs in samples/bpf that link
statically against libbpf.

Fixes: 166750bc1dd2 ("libbpf: Support libbpf-provided extern variables")
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Tested-by: Björn Töpel <bjorn.topel@intel.com>
Link: https://lore.kernel.org/bpf/20191216102405.353834-1-toke@redhat.com
---
 samples/bpf/Makefile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile
index 8003d2823fa8..227077ae66ed 100644
--- a/samples/bpf/Makefile
+++ b/samples/bpf/Makefile
@@ -198,7 +198,7 @@ endif
 
 TPROGCFLAGS_bpf_load.o += -Wno-unused-variable
 
-TPROGS_LDLIBS			+= $(LIBBPF) -lelf
+TPROGS_LDLIBS			+= $(LIBBPF) -lelf -lz
 TPROGLDLIBS_tracex4		+= -lrt
 TPROGLDLIBS_trace_output	+= -lrt
 TPROGLDLIBS_map_perf_test	+= -lrt
-- 
cgit v1.2.3-59-g8ed1b


From 450278977acbf494a20367c22fbb38729772d1fc Mon Sep 17 00:00:00 2001
From: Toke Høiland-Jørgensen <toke@redhat.com>
Date: Mon, 16 Dec 2019 11:38:19 +0100
Subject: samples/bpf: Set -fno-stack-protector when building BPF programs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

It seems Clang can in some cases turn on stack protection by default, which
doesn't work with BPF. This was reported once before[0], but it seems the
flag to explicitly turn off the stack protector wasn't added to the
Makefile, so do that now.

The symptom of this is compile errors like the following:

error: <unknown>:0:0: in function bpf_prog1 i32 (%struct.__sk_buff*): A call to built-in function '__stack_chk_fail' is not supported.

[0] https://www.spinics.net/lists/netdev/msg556400.html

Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191216103819.359535-1-toke@redhat.com
---
 samples/bpf/Makefile | 1 +
 1 file changed, 1 insertion(+)

diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile
index 227077ae66ed..5b89c0370f33 100644
--- a/samples/bpf/Makefile
+++ b/samples/bpf/Makefile
@@ -236,6 +236,7 @@ BTF_LLVM_PROBE := $(shell echo "int main() { return 0; }" | \
 			  readelf -S ./llvm_btf_verify.o | grep BTF; \
 			  /bin/rm -f ./llvm_btf_verify.o)
 
+BPF_EXTRA_CFLAGS += -fno-stack-protector
 ifneq ($(BTF_LLVM_PROBE),)
 	BPF_EXTRA_CFLAGS += -g
 else
-- 
cgit v1.2.3-59-g8ed1b


From d50ecc46d18fa19ccf06e0c4d2ee8a050c665e3d Mon Sep 17 00:00:00 2001
From: Toke Høiland-Jørgensen <toke@redhat.com>
Date: Mon, 16 Dec 2019 12:07:42 +0100
Subject: samples/bpf: Attach XDP programs in driver mode by default
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When attaching XDP programs, userspace can set flags to request the attach
mode (generic/SKB mode, driver mode or hw offloaded mode). If no such flags
are requested, the kernel will attempt to attach in driver mode, and then
silently fall back to SKB mode if this fails.

The silent fallback is a major source of user confusion, as users will try
to load a program on a device without XDP support, and instead of an error
they will get the silent fallback behaviour, not notice, and then wonder
why performance is not what they were expecting.

In an attempt to combat this, let's switch all the samples to default to
explicitly requesting driver-mode attach. As part of this, ensure that all
the userspace utilities have a switch to enable SKB mode. For those that
have a switch to request driver mode, keep it but turn it into a no-op.

Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: David Ahern <dsahern@gmail.com>
Link: https://lore.kernel.org/bpf/20191216110742.364456-1-toke@redhat.com
---
 samples/bpf/xdp1_user.c             |  5 ++++-
 samples/bpf/xdp_adjust_tail_user.c  |  5 ++++-
 samples/bpf/xdp_fwd_user.c          | 17 ++++++++++++++---
 samples/bpf/xdp_redirect_cpu_user.c |  4 ++++
 samples/bpf/xdp_redirect_map_user.c |  5 ++++-
 samples/bpf/xdp_redirect_user.c     |  5 ++++-
 samples/bpf/xdp_router_ipv4_user.c  |  3 +++
 samples/bpf/xdp_rxq_info_user.c     |  4 ++++
 samples/bpf/xdp_sample_pkts_user.c  | 12 +++++++++---
 samples/bpf/xdp_tx_iptunnel_user.c  |  5 ++++-
 samples/bpf/xdpsock_user.c          |  5 ++++-
 11 files changed, 58 insertions(+), 12 deletions(-)

diff --git a/samples/bpf/xdp1_user.c b/samples/bpf/xdp1_user.c
index 3e553eed95a7..38a8852cb57f 100644
--- a/samples/bpf/xdp1_user.c
+++ b/samples/bpf/xdp1_user.c
@@ -98,7 +98,7 @@ int main(int argc, char **argv)
 			xdp_flags |= XDP_FLAGS_SKB_MODE;
 			break;
 		case 'N':
-			xdp_flags |= XDP_FLAGS_DRV_MODE;
+			/* default, set below */
 			break;
 		case 'F':
 			xdp_flags &= ~XDP_FLAGS_UPDATE_IF_NOEXIST;
@@ -109,6 +109,9 @@ int main(int argc, char **argv)
 		}
 	}
 
+	if (!(xdp_flags & XDP_FLAGS_SKB_MODE))
+		xdp_flags |= XDP_FLAGS_DRV_MODE;
+
 	if (optind == argc) {
 		usage(basename(argv[0]));
 		return 1;
diff --git a/samples/bpf/xdp_adjust_tail_user.c b/samples/bpf/xdp_adjust_tail_user.c
index d86e9ad0356b..008789eb6ada 100644
--- a/samples/bpf/xdp_adjust_tail_user.c
+++ b/samples/bpf/xdp_adjust_tail_user.c
@@ -120,7 +120,7 @@ int main(int argc, char **argv)
 			xdp_flags |= XDP_FLAGS_SKB_MODE;
 			break;
 		case 'N':
-			xdp_flags |= XDP_FLAGS_DRV_MODE;
+			/* default, set below */
 			break;
 		case 'F':
 			xdp_flags &= ~XDP_FLAGS_UPDATE_IF_NOEXIST;
@@ -132,6 +132,9 @@ int main(int argc, char **argv)
 		opt_flags[opt] = 0;
 	}
 
+	if (!(xdp_flags & XDP_FLAGS_SKB_MODE))
+		xdp_flags |= XDP_FLAGS_DRV_MODE;
+
 	for (i = 0; i < strlen(optstr); i++) {
 		if (opt_flags[(unsigned int)optstr[i]]) {
 			fprintf(stderr, "Missing argument -%c\n", optstr[i]);
diff --git a/samples/bpf/xdp_fwd_user.c b/samples/bpf/xdp_fwd_user.c
index 97ff1dad7669..c30f9acfdb84 100644
--- a/samples/bpf/xdp_fwd_user.c
+++ b/samples/bpf/xdp_fwd_user.c
@@ -27,11 +27,13 @@
 #include "libbpf.h"
 #include <bpf/bpf.h>
 
+static __u32 xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
+
 static int do_attach(int idx, int prog_fd, int map_fd, const char *name)
 {
 	int err;
 
-	err = bpf_set_link_xdp_fd(idx, prog_fd, 0);
+	err = bpf_set_link_xdp_fd(idx, prog_fd, xdp_flags);
 	if (err < 0) {
 		printf("ERROR: failed to attach program to %s\n", name);
 		return err;
@@ -49,7 +51,7 @@ static int do_detach(int idx, const char *name)
 {
 	int err;
 
-	err = bpf_set_link_xdp_fd(idx, -1, 0);
+	err = bpf_set_link_xdp_fd(idx, -1, xdp_flags);
 	if (err < 0)
 		printf("ERROR: failed to detach program from %s\n", name);
 
@@ -83,11 +85,17 @@ int main(int argc, char **argv)
 	int attach = 1;
 	int ret = 0;
 
-	while ((opt = getopt(argc, argv, ":dD")) != -1) {
+	while ((opt = getopt(argc, argv, ":dDSF")) != -1) {
 		switch (opt) {
 		case 'd':
 			attach = 0;
 			break;
+		case 'S':
+			xdp_flags |= XDP_FLAGS_SKB_MODE;
+			break;
+		case 'F':
+			xdp_flags &= ~XDP_FLAGS_UPDATE_IF_NOEXIST;
+			break;
 		case 'D':
 			prog_name = "xdp_fwd_direct";
 			break;
@@ -97,6 +105,9 @@ int main(int argc, char **argv)
 		}
 	}
 
+	if (!(xdp_flags & XDP_FLAGS_SKB_MODE))
+		xdp_flags |= XDP_FLAGS_DRV_MODE;
+
 	if (optind == argc) {
 		usage(basename(argv[0]));
 		return 1;
diff --git a/samples/bpf/xdp_redirect_cpu_user.c b/samples/bpf/xdp_redirect_cpu_user.c
index 0da6e9e7132e..72af628529b5 100644
--- a/samples/bpf/xdp_redirect_cpu_user.c
+++ b/samples/bpf/xdp_redirect_cpu_user.c
@@ -728,6 +728,10 @@ int main(int argc, char **argv)
 			return EXIT_FAIL_OPTION;
 		}
 	}
+
+	if (!(xdp_flags & XDP_FLAGS_SKB_MODE))
+		xdp_flags |= XDP_FLAGS_DRV_MODE;
+
 	/* Required option */
 	if (ifindex == -1) {
 		fprintf(stderr, "ERR: required option --dev missing\n");
diff --git a/samples/bpf/xdp_redirect_map_user.c b/samples/bpf/xdp_redirect_map_user.c
index f70ee33907fd..cc840661faab 100644
--- a/samples/bpf/xdp_redirect_map_user.c
+++ b/samples/bpf/xdp_redirect_map_user.c
@@ -116,7 +116,7 @@ int main(int argc, char **argv)
 			xdp_flags |= XDP_FLAGS_SKB_MODE;
 			break;
 		case 'N':
-			xdp_flags |= XDP_FLAGS_DRV_MODE;
+			/* default, set below */
 			break;
 		case 'F':
 			xdp_flags &= ~XDP_FLAGS_UPDATE_IF_NOEXIST;
@@ -127,6 +127,9 @@ int main(int argc, char **argv)
 		}
 	}
 
+	if (!(xdp_flags & XDP_FLAGS_SKB_MODE))
+		xdp_flags |= XDP_FLAGS_DRV_MODE;
+
 	if (optind == argc) {
 		printf("usage: %s <IFNAME|IFINDEX>_IN <IFNAME|IFINDEX>_OUT\n", argv[0]);
 		return 1;
diff --git a/samples/bpf/xdp_redirect_user.c b/samples/bpf/xdp_redirect_user.c
index 5440cd620607..71dff8e3382a 100644
--- a/samples/bpf/xdp_redirect_user.c
+++ b/samples/bpf/xdp_redirect_user.c
@@ -117,7 +117,7 @@ int main(int argc, char **argv)
 			xdp_flags |= XDP_FLAGS_SKB_MODE;
 			break;
 		case 'N':
-			xdp_flags |= XDP_FLAGS_DRV_MODE;
+			/* default, set below */
 			break;
 		case 'F':
 			xdp_flags &= ~XDP_FLAGS_UPDATE_IF_NOEXIST;
@@ -128,6 +128,9 @@ int main(int argc, char **argv)
 		}
 	}
 
+	if (!(xdp_flags & XDP_FLAGS_SKB_MODE))
+		xdp_flags |= XDP_FLAGS_DRV_MODE;
+
 	if (optind == argc) {
 		printf("usage: %s <IFNAME|IFINDEX>_IN <IFNAME|IFINDEX>_OUT\n", argv[0]);
 		return 1;
diff --git a/samples/bpf/xdp_router_ipv4_user.c b/samples/bpf/xdp_router_ipv4_user.c
index 1469b66ebad1..fef286c5add2 100644
--- a/samples/bpf/xdp_router_ipv4_user.c
+++ b/samples/bpf/xdp_router_ipv4_user.c
@@ -662,6 +662,9 @@ int main(int ac, char **argv)
 		}
 	}
 
+	if (!(flags & XDP_FLAGS_SKB_MODE))
+		flags |= XDP_FLAGS_DRV_MODE;
+
 	if (optind == ac) {
 		usage(basename(argv[0]));
 		return 1;
diff --git a/samples/bpf/xdp_rxq_info_user.c b/samples/bpf/xdp_rxq_info_user.c
index 8fc3ad01de72..fc4983fd6959 100644
--- a/samples/bpf/xdp_rxq_info_user.c
+++ b/samples/bpf/xdp_rxq_info_user.c
@@ -551,6 +551,10 @@ int main(int argc, char **argv)
 			return EXIT_FAIL_OPTION;
 		}
 	}
+
+	if (!(xdp_flags & XDP_FLAGS_SKB_MODE))
+		xdp_flags |= XDP_FLAGS_DRV_MODE;
+
 	/* Required option */
 	if (ifindex == -1) {
 		fprintf(stderr, "ERR: required option --dev missing\n");
diff --git a/samples/bpf/xdp_sample_pkts_user.c b/samples/bpf/xdp_sample_pkts_user.c
index a5760e8bf2c4..8c1af1b7372d 100644
--- a/samples/bpf/xdp_sample_pkts_user.c
+++ b/samples/bpf/xdp_sample_pkts_user.c
@@ -52,13 +52,13 @@ static int do_detach(int idx, const char *name)
 	__u32 curr_prog_id = 0;
 	int err = 0;
 
-	err = bpf_get_link_xdp_id(idx, &curr_prog_id, 0);
+	err = bpf_get_link_xdp_id(idx, &curr_prog_id, xdp_flags);
 	if (err) {
 		printf("bpf_get_link_xdp_id failed\n");
 		return err;
 	}
 	if (prog_id == curr_prog_id) {
-		err = bpf_set_link_xdp_fd(idx, -1, 0);
+		err = bpf_set_link_xdp_fd(idx, -1, xdp_flags);
 		if (err < 0)
 			printf("ERROR: failed to detach prog from %s\n", name);
 	} else if (!curr_prog_id) {
@@ -115,7 +115,7 @@ int main(int argc, char **argv)
 		.prog_type	= BPF_PROG_TYPE_XDP,
 	};
 	struct perf_buffer_opts pb_opts = {};
-	const char *optstr = "F";
+	const char *optstr = "FS";
 	int prog_fd, map_fd, opt;
 	struct bpf_object *obj;
 	struct bpf_map *map;
@@ -127,12 +127,18 @@ int main(int argc, char **argv)
 		case 'F':
 			xdp_flags &= ~XDP_FLAGS_UPDATE_IF_NOEXIST;
 			break;
+		case 'S':
+			xdp_flags |= XDP_FLAGS_SKB_MODE;
+			break;
 		default:
 			usage(basename(argv[0]));
 			return 1;
 		}
 	}
 
+	if (!(xdp_flags & XDP_FLAGS_SKB_MODE))
+		xdp_flags |= XDP_FLAGS_DRV_MODE;
+
 	if (optind == argc) {
 		usage(basename(argv[0]));
 		return 1;
diff --git a/samples/bpf/xdp_tx_iptunnel_user.c b/samples/bpf/xdp_tx_iptunnel_user.c
index 2fe4c7f5ffe5..5f33b5530032 100644
--- a/samples/bpf/xdp_tx_iptunnel_user.c
+++ b/samples/bpf/xdp_tx_iptunnel_user.c
@@ -231,7 +231,7 @@ int main(int argc, char **argv)
 			xdp_flags |= XDP_FLAGS_SKB_MODE;
 			break;
 		case 'N':
-			xdp_flags |= XDP_FLAGS_DRV_MODE;
+			/* default, set below */
 			break;
 		case 'F':
 			xdp_flags &= ~XDP_FLAGS_UPDATE_IF_NOEXIST;
@@ -243,6 +243,9 @@ int main(int argc, char **argv)
 		opt_flags[opt] = 0;
 	}
 
+	if (!(xdp_flags & XDP_FLAGS_SKB_MODE))
+		xdp_flags |= XDP_FLAGS_DRV_MODE;
+
 	for (i = 0; i < strlen(optstr); i++) {
 		if (opt_flags[(unsigned int)optstr[i]]) {
 			fprintf(stderr, "Missing argument -%c\n", optstr[i]);
diff --git a/samples/bpf/xdpsock_user.c b/samples/bpf/xdpsock_user.c
index a15480010828..e7829e5baaff 100644
--- a/samples/bpf/xdpsock_user.c
+++ b/samples/bpf/xdpsock_user.c
@@ -440,7 +440,7 @@ static void parse_command_line(int argc, char **argv)
 			opt_xdp_bind_flags |= XDP_COPY;
 			break;
 		case 'N':
-			opt_xdp_flags |= XDP_FLAGS_DRV_MODE;
+			/* default, set below */
 			break;
 		case 'n':
 			opt_interval = atoi(optarg);
@@ -474,6 +474,9 @@ static void parse_command_line(int argc, char **argv)
 		}
 	}
 
+	if (!(opt_xdp_flags & XDP_FLAGS_SKB_MODE))
+		opt_xdp_flags |= XDP_FLAGS_DRV_MODE;
+
 	opt_ifindex = if_nametoindex(opt_if);
 	if (!opt_ifindex) {
 		fprintf(stderr, "ERROR: interface \"%s\" does not exist\n",
-- 
cgit v1.2.3-59-g8ed1b


From dc3a2d254782dea8e23a55f46129105ecfe787fb Mon Sep 17 00:00:00 2001
From: Toke Høiland-Jørgensen <toke@redhat.com>
Date: Mon, 16 Dec 2019 19:12:04 +0100
Subject: libbpf: Print hint about ulimit when getting permission denied error
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Probably the single most common error newcomers to XDP are stumped by is
the 'permission denied' error they get when trying to load their program
and 'ulimit -l' is set too low. For examples, see [0], [1].

Since the error code is UAPI, we can't change that. Instead, this patch
adds a few heuristics in libbpf and outputs an additional hint if they are
met: If an EPERM is returned on map create or program load, and geteuid()
shows we are root, and the current RLIMIT_MEMLOCK is not infinity, we
output a hint about raising 'ulimit -l' as an additional log line.

[0] https://marc.info/?l=xdp-newbies&m=157043612505624&w=2
[1] https://github.com/xdp-project/xdp-tutorial/issues/86

Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20191216181204.724953-1-toke@redhat.com
---
 tools/lib/bpf/libbpf.c | 29 +++++++++++++++++++++++++++++
 1 file changed, 29 insertions(+)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index a2cc7313763a..3fe42d6b0c2f 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -41,6 +41,7 @@
 #include <sys/types.h>
 #include <sys/vfs.h>
 #include <sys/utsname.h>
+#include <sys/resource.h>
 #include <tools/libc_compat.h>
 #include <libelf.h>
 #include <gelf.h>
@@ -100,6 +101,32 @@ void libbpf_print(enum libbpf_print_level level, const char *format, ...)
 	va_end(args);
 }
 
+static void pr_perm_msg(int err)
+{
+	struct rlimit limit;
+	char buf[100];
+
+	if (err != -EPERM || geteuid() != 0)
+		return;
+
+	err = getrlimit(RLIMIT_MEMLOCK, &limit);
+	if (err)
+		return;
+
+	if (limit.rlim_cur == RLIM_INFINITY)
+		return;
+
+	if (limit.rlim_cur < 1024)
+		snprintf(buf, sizeof(buf), "%lu bytes", limit.rlim_cur);
+	else if (limit.rlim_cur < 1024*1024)
+		snprintf(buf, sizeof(buf), "%.1f KiB", (double)limit.rlim_cur / 1024);
+	else
+		snprintf(buf, sizeof(buf), "%.1f MiB", (double)limit.rlim_cur / (1024*1024));
+
+	pr_warn("permission error while running as root; try raising 'ulimit -l'? current value: %s\n",
+		buf);
+}
+
 #define STRERR_BUFSIZE  128
 
 /* Copied from tools/perf/util/util.h */
@@ -2983,6 +3010,7 @@ err_out:
 			cp = libbpf_strerror_r(err, errmsg, sizeof(errmsg));
 			pr_warn("failed to create map (name: '%s'): %s(%d)\n",
 				map->name, cp, err);
+			pr_perm_msg(err);
 			for (j = 0; j < i; j++)
 				zclose(obj->maps[j].fd);
 			return err;
@@ -4381,6 +4409,7 @@ retry_load:
 	ret = -errno;
 	cp = libbpf_strerror_r(errno, errmsg, sizeof(errmsg));
 	pr_warn("load bpf program failed: %s\n", cp);
+	pr_perm_msg(ret);
 
 	if (log_buf && log_buf[0] != '\0') {
 		ret = -LIBBPF_ERRNO__VERIFY;
-- 
cgit v1.2.3-59-g8ed1b


From dbd8f6bae6f4a98aa517aeda6e5276ed4a20f571 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Mon, 16 Dec 2019 10:38:29 -0800
Subject: libbpf: Add zlib as a dependency in pkg-config template

List zlib as another dependency of libbpf in pkg-config template.
Verified it is correctly resolved to proper -lz flag:

$ make DESTDIR=/tmp/libbpf-install install
$ pkg-config --libs /tmp/libbpf-install/usr/local/lib64/pkgconfig/libbpf.pc
-L/usr/local/lib64 -lbpf
$ pkg-config --libs --static /tmp/libbpf-install/usr/local/lib64/pkgconfig/libbpf.pc
-L/usr/local/lib64 -lbpf -lelf -lz

Fixes: 166750bc1dd2 ("libbpf: Support libbpf-provided extern variables")
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com>
Cc: Luca Boccassi <bluca@debian.org>
Link: https://lore.kernel.org/bpf/20191216183830.3972964-1-andriin@fb.com
---
 tools/lib/bpf/libbpf.pc.template | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/lib/bpf/libbpf.pc.template b/tools/lib/bpf/libbpf.pc.template
index ac17fcef2108..b45ed534bdfb 100644
--- a/tools/lib/bpf/libbpf.pc.template
+++ b/tools/lib/bpf/libbpf.pc.template
@@ -8,5 +8,5 @@ Name: libbpf
 Description: BPF library
 Version: @VERSION@
 Libs: -L${libdir} -lbpf
-Requires.private: libelf
+Requires.private: libelf zlib
 Cflags: -I${includedir}
-- 
cgit v1.2.3-59-g8ed1b


From 10bc877c762339b8501a5b397137cf2f47cde575 Mon Sep 17 00:00:00 2001
From: Yangbo Lu <yangbo.lu@nxp.com>
Date: Thu, 12 Dec 2019 18:08:04 +0800
Subject: ptp_qoriq: check valid status before reading extts fifo

For PTP timer supporting external trigger timestamp FIFO,
there is a valid bit in TMR_STAT register indicating the
timestamp is available. For PTP timer without FIFO, there
is not TMR_STAT register.
This patch is to check the valid bit for the FIFO before
reading timestamp, and to avoid operating TMR_STAT register
for PTP timer without the FIFO.

Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/ptp/ptp_qoriq.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/drivers/ptp/ptp_qoriq.c b/drivers/ptp/ptp_qoriq.c
index a577218d1ab7..a3062cd4d174 100644
--- a/drivers/ptp/ptp_qoriq.c
+++ b/drivers/ptp/ptp_qoriq.c
@@ -81,7 +81,7 @@ static int extts_clean_up(struct ptp_qoriq *ptp_qoriq, int index,
 	struct ptp_clock_event event;
 	void __iomem *reg_etts_l;
 	void __iomem *reg_etts_h;
-	u32 valid, stat, lo, hi;
+	u32 valid, lo, hi;
 
 	switch (index) {
 	case 0:
@@ -101,6 +101,10 @@ static int extts_clean_up(struct ptp_qoriq *ptp_qoriq, int index,
 	event.type = PTP_CLOCK_EXTTS;
 	event.index = index;
 
+	if (ptp_qoriq->extts_fifo_support)
+		if (!(ptp_qoriq->read(&regs->ctrl_regs->tmr_stat) & valid))
+			return 0;
+
 	do {
 		lo = ptp_qoriq->read(reg_etts_l);
 		hi = ptp_qoriq->read(reg_etts_h);
@@ -111,8 +115,9 @@ static int extts_clean_up(struct ptp_qoriq *ptp_qoriq, int index,
 			ptp_clock_event(ptp_qoriq->clock, &event);
 		}
 
-		stat = ptp_qoriq->read(&regs->ctrl_regs->tmr_stat);
-	} while (ptp_qoriq->extts_fifo_support && (stat & valid));
+		if (!ptp_qoriq->extts_fifo_support)
+			break;
+	} while (ptp_qoriq->read(&regs->ctrl_regs->tmr_stat) & valid);
 
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 9429439f59cd3b82a3e2732ead5363578de97a84 Mon Sep 17 00:00:00 2001
From: Yangbo Lu <yangbo.lu@nxp.com>
Date: Thu, 12 Dec 2019 18:08:05 +0800
Subject: ptp_qoriq: export extts_clean_up() function

Export extts_clean_up() function so that dpaa2-ptp
driver is able to reuse it.

Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/ptp/ptp_qoriq.c       | 4 ++--
 include/linux/fsl/ptp_qoriq.h | 1 +
 2 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/ptp/ptp_qoriq.c b/drivers/ptp/ptp_qoriq.c
index a3062cd4d174..b27c46ebfc8f 100644
--- a/drivers/ptp/ptp_qoriq.c
+++ b/drivers/ptp/ptp_qoriq.c
@@ -74,8 +74,7 @@ static void set_fipers(struct ptp_qoriq *ptp_qoriq)
 	ptp_qoriq->write(&regs->fiper_regs->tmr_fiper2, ptp_qoriq->tmr_fiper2);
 }
 
-static int extts_clean_up(struct ptp_qoriq *ptp_qoriq, int index,
-			  bool update_event)
+int extts_clean_up(struct ptp_qoriq *ptp_qoriq, int index, bool update_event)
 {
 	struct ptp_qoriq_registers *regs = &ptp_qoriq->regs;
 	struct ptp_clock_event event;
@@ -121,6 +120,7 @@ static int extts_clean_up(struct ptp_qoriq *ptp_qoriq, int index,
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(extts_clean_up);
 
 /*
  * Interrupt service routine
diff --git a/include/linux/fsl/ptp_qoriq.h b/include/linux/fsl/ptp_qoriq.h
index 992bf9fa1729..b0b743563f43 100644
--- a/include/linux/fsl/ptp_qoriq.h
+++ b/include/linux/fsl/ptp_qoriq.h
@@ -192,6 +192,7 @@ int ptp_qoriq_settime(struct ptp_clock_info *ptp,
 		      const struct timespec64 *ts);
 int ptp_qoriq_enable(struct ptp_clock_info *ptp,
 		     struct ptp_clock_request *rq, int on);
+int extts_clean_up(struct ptp_qoriq *ptp_qoriq, int index, bool update_event);
 #ifdef CONFIG_DEBUG_FS
 void ptp_qoriq_create_debugfs(struct ptp_qoriq *ptp_qoriq);
 void ptp_qoriq_remove_debugfs(struct ptp_qoriq *ptp_qoriq);
-- 
cgit v1.2.3-59-g8ed1b


From 17568c62cd2c9a8d75eb5da44b7df093819b6ebe Mon Sep 17 00:00:00 2001
From: Yangbo Lu <yangbo.lu@nxp.com>
Date: Thu, 12 Dec 2019 18:08:06 +0800
Subject: dpaa2-ptp: add external trigger event support

This patch is to add external trigger event support
for dpaa2-ptp driver with v2 dprtc_set_irq_mask() API.

Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/freescale/dpaa2/dpaa2-ptp.c | 20 ++++++++++++++++++++
 drivers/net/ethernet/freescale/dpaa2/dprtc-cmd.h |  4 +++-
 drivers/net/ethernet/freescale/dpaa2/dprtc.h     |  2 ++
 3 files changed, 25 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-ptp.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-ptp.c
index a9503aea527f..a0061e929908 100644
--- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-ptp.c
+++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-ptp.c
@@ -27,6 +27,20 @@ static int dpaa2_ptp_enable(struct ptp_clock_info *ptp,
 	mc_dev = to_fsl_mc_device(dev);
 
 	switch (rq->type) {
+	case PTP_CLK_REQ_EXTTS:
+		switch (rq->extts.index) {
+		case 0:
+			bit = DPRTC_EVENT_ETS1;
+			break;
+		case 1:
+			bit = DPRTC_EVENT_ETS2;
+			break;
+		default:
+			return -EINVAL;
+		}
+		if (on)
+			extts_clean_up(ptp_qoriq, rq->extts.index, false);
+		break;
 	case PTP_CLK_REQ_PPS:
 		bit = DPRTC_EVENT_PPS;
 		break;
@@ -96,6 +110,12 @@ static irqreturn_t dpaa2_ptp_irq_handler_thread(int irq, void *priv)
 		ptp_clock_event(ptp_qoriq->clock, &event);
 	}
 
+	if (status & DPRTC_EVENT_ETS1)
+		extts_clean_up(ptp_qoriq, 0, true);
+
+	if (status & DPRTC_EVENT_ETS2)
+		extts_clean_up(ptp_qoriq, 1, true);
+
 	err = dprtc_clear_irq_status(mc_dev->mc_io, 0, mc_dev->mc_handle,
 				     DPRTC_IRQ_INDEX, status);
 	if (unlikely(err)) {
diff --git a/drivers/net/ethernet/freescale/dpaa2/dprtc-cmd.h b/drivers/net/ethernet/freescale/dpaa2/dprtc-cmd.h
index 4ac05bfef338..96ffeb948f08 100644
--- a/drivers/net/ethernet/freescale/dpaa2/dprtc-cmd.h
+++ b/drivers/net/ethernet/freescale/dpaa2/dprtc-cmd.h
@@ -9,9 +9,11 @@
 
 /* Command versioning */
 #define DPRTC_CMD_BASE_VERSION		1
+#define DPRTC_CMD_VERSION_2		2
 #define DPRTC_CMD_ID_OFFSET		4
 
 #define DPRTC_CMD(id)	(((id) << DPRTC_CMD_ID_OFFSET) | DPRTC_CMD_BASE_VERSION)
+#define DPRTC_CMD_V2(id) (((id) << DPRTC_CMD_ID_OFFSET) | DPRTC_CMD_VERSION_2)
 
 /* Command IDs */
 #define DPRTC_CMDID_CLOSE			DPRTC_CMD(0x800)
@@ -19,7 +21,7 @@
 
 #define DPRTC_CMDID_SET_IRQ_ENABLE		DPRTC_CMD(0x012)
 #define DPRTC_CMDID_GET_IRQ_ENABLE		DPRTC_CMD(0x013)
-#define DPRTC_CMDID_SET_IRQ_MASK		DPRTC_CMD(0x014)
+#define DPRTC_CMDID_SET_IRQ_MASK		DPRTC_CMD_V2(0x014)
 #define DPRTC_CMDID_GET_IRQ_MASK		DPRTC_CMD(0x015)
 #define DPRTC_CMDID_GET_IRQ_STATUS		DPRTC_CMD(0x016)
 #define DPRTC_CMDID_CLEAR_IRQ_STATUS		DPRTC_CMD(0x017)
diff --git a/drivers/net/ethernet/freescale/dpaa2/dprtc.h b/drivers/net/ethernet/freescale/dpaa2/dprtc.h
index 311c184e1aef..05c413719e55 100644
--- a/drivers/net/ethernet/freescale/dpaa2/dprtc.h
+++ b/drivers/net/ethernet/freescale/dpaa2/dprtc.h
@@ -20,6 +20,8 @@ struct fsl_mc_io;
 #define DPRTC_IRQ_INDEX		0
 
 #define DPRTC_EVENT_PPS		0x08000000
+#define DPRTC_EVENT_ETS1	0x00800000
+#define DPRTC_EVENT_ETS2	0x00400000
 
 int dprtc_open(struct fsl_mc_io *mc_io,
 	       u32 cmd_flags,
-- 
cgit v1.2.3-59-g8ed1b


From f325937735498afb054a0195291bbf68d0b60be5 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Fri, 13 Dec 2019 16:53:37 +0100
Subject: r8169: check that Realtek PHY driver module is loaded

Some users complained about problems with r8169 and it turned out that
the generic PHY driver was used instead instead of the dedicated one.
In all cases reason was that r8169.ko was in initramfs, but realtek.ko
not. Manually adding realtek.ko to initramfs fixed the issues.
Root cause seems to be that tools like dracut and genkernel don't
consider softdeps. Add a check for loaded Realtek PHY driver module
and provide the user with a hint if it's not loaded.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 2cfaa2270319..a887b685d070 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -6825,6 +6825,15 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	int chipset, region;
 	int jumbo_max, rc;
 
+	/* Some tools for creating an initramfs don't consider softdeps, then
+	 * r8169.ko may be in initramfs, but realtek.ko not. Then the generic
+	 * PHY driver is used that doesn't work with most chip versions.
+	 */
+	if (!driver_find("RTL8201CP Ethernet", &mdio_bus_type)) {
+		dev_err(&pdev->dev, "realtek.ko not loaded, maybe it needs to be added to initramfs?\n");
+		return -ENOENT;
+	}
+
 	dev = devm_alloc_etherdev(&pdev->dev, sizeof (*tp));
 	if (!dev)
 		return -ENOMEM;
-- 
cgit v1.2.3-59-g8ed1b


From 1f1c1d7c89ee538f3e36b43098e95973f8fa37db Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Fri, 13 Dec 2019 21:24:27 +0100
Subject: ipv6: Annotate bitwise IPv6 dsfield pointer cast

The sparse commit 6002ded74587 ("add a flag to warn on casts to/from
bitwise pointers") introduced a check for non-direct casts from/to
restricted datatypes (when -Wbitwise-pointer is enabled).

This triggered a warning in ipv6_get_dsfield() because sparse doesn't know
that the buffer already points to some data in the correct bitwise integer
format. This was already fixed in ipv6_change_dsfield() by the __force
attribute and can be fixed here the same way.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/dsfield.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/net/dsfield.h b/include/net/dsfield.h
index 1a245ee10c95..a59a57ffc546 100644
--- a/include/net/dsfield.h
+++ b/include/net/dsfield.h
@@ -21,7 +21,7 @@ static inline __u8 ipv4_get_dsfield(const struct iphdr *iph)
 
 static inline __u8 ipv6_get_dsfield(const struct ipv6hdr *ipv6h)
 {
-	return ntohs(*(const __be16 *)ipv6h) >> 4;
+	return ntohs(*(__force const __be16 *)ipv6h) >> 4;
 }
 
 
-- 
cgit v1.2.3-59-g8ed1b


From 54e1f08bddbe63a3c0ae44f65df2c8b895003ef4 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Fri, 13 Dec 2019 21:24:28 +0100
Subject: ipv6: Annotate ipv6_addr_is_* bitwise pointer casts

The sparse commit 6002ded74587 ("add a flag to warn on casts to/from
bitwise pointers") introduced a check for non-direct casts from/to
restricted datatypes (when -Wbitwise-pointer is enabled).

This triggered a warning in the 64 bit optimized ipv6_addr_is_*() functions
because sparse doesn't know that the buffer already points to some data in
the correct bitwise integer format. But these were correct and can
therefore be marked with __force to signalize sparse an intended cast to a
specific bitwise type.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/addrconf.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index 1bab88184d3c..a088349dd94f 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -437,7 +437,7 @@ static inline void addrconf_addr_solict_mult(const struct in6_addr *addr,
 static inline bool ipv6_addr_is_ll_all_nodes(const struct in6_addr *addr)
 {
 #if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) && BITS_PER_LONG == 64
-	__be64 *p = (__be64 *)addr;
+	__be64 *p = (__force __be64 *)addr;
 	return ((p[0] ^ cpu_to_be64(0xff02000000000000UL)) | (p[1] ^ cpu_to_be64(1))) == 0UL;
 #else
 	return ((addr->s6_addr32[0] ^ htonl(0xff020000)) |
@@ -449,7 +449,7 @@ static inline bool ipv6_addr_is_ll_all_nodes(const struct in6_addr *addr)
 static inline bool ipv6_addr_is_ll_all_routers(const struct in6_addr *addr)
 {
 #if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) && BITS_PER_LONG == 64
-	__be64 *p = (__be64 *)addr;
+	__be64 *p = (__force __be64 *)addr;
 	return ((p[0] ^ cpu_to_be64(0xff02000000000000UL)) | (p[1] ^ cpu_to_be64(2))) == 0UL;
 #else
 	return ((addr->s6_addr32[0] ^ htonl(0xff020000)) |
@@ -466,7 +466,7 @@ static inline bool ipv6_addr_is_isatap(const struct in6_addr *addr)
 static inline bool ipv6_addr_is_solict_mult(const struct in6_addr *addr)
 {
 #if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) && BITS_PER_LONG == 64
-	__be64 *p = (__be64 *)addr;
+	__be64 *p = (__force __be64 *)addr;
 	return ((p[0] ^ cpu_to_be64(0xff02000000000000UL)) |
 		((p[1] ^ cpu_to_be64(0x00000001ff000000UL)) &
 		 cpu_to_be64(0xffffffffff000000UL))) == 0UL;
@@ -481,7 +481,7 @@ static inline bool ipv6_addr_is_solict_mult(const struct in6_addr *addr)
 static inline bool ipv6_addr_is_all_snoopers(const struct in6_addr *addr)
 {
 #if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) && BITS_PER_LONG == 64
-	__be64 *p = (__be64 *)addr;
+	__be64 *p = (__force __be64 *)addr;
 
 	return ((p[0] ^ cpu_to_be64(0xff02000000000000UL)) |
 		(p[1] ^ cpu_to_be64(0x6a))) == 0UL;
-- 
cgit v1.2.3-59-g8ed1b


From b3c3fe8e9961c3c76f8c6c6d5e50336cb6b16b48 Mon Sep 17 00:00:00 2001
From: Yunsheng Lin <linyunsheng@huawei.com>
Date: Sat, 14 Dec 2019 10:06:37 +0800
Subject: net: hns3: schedule hclgevf_service by using delayed workqueue

Currently, a timer is defined to schedule hclgevf_service per
second. To simplify the code, this patch uses the delayed work
instead of timer to schedule hclgevf_serive.

Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c  | 39 ++++++++--------------
 .../ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h  |  3 +-
 2 files changed, 15 insertions(+), 27 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
index 25d78a5aaa34..e48f8fe585bd 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
@@ -1786,8 +1786,11 @@ void hclgevf_mbx_task_schedule(struct hclgevf_dev *hdev)
 static void hclgevf_task_schedule(struct hclgevf_dev *hdev)
 {
 	if (!test_bit(HCLGEVF_STATE_DOWN, &hdev->state)  &&
-	    !test_and_set_bit(HCLGEVF_STATE_SERVICE_SCHED, &hdev->state))
-		schedule_work(&hdev->service_task);
+	    !test_and_set_bit(HCLGEVF_STATE_SERVICE_SCHED, &hdev->state)) {
+		mod_delayed_work(system_wq, &hdev->service_task,
+				 round_jiffies_relative(HZ));
+		hdev->stats_timer++;
+	}
 }
 
 static void hclgevf_deferred_task_schedule(struct hclgevf_dev *hdev)
@@ -1800,17 +1803,6 @@ static void hclgevf_deferred_task_schedule(struct hclgevf_dev *hdev)
 		hclgevf_reset_task_schedule(hdev);
 }
 
-static void hclgevf_service_timer(struct timer_list *t)
-{
-	struct hclgevf_dev *hdev = from_timer(hdev, t, service_timer);
-
-	mod_timer(&hdev->service_timer, jiffies +
-		  HCLGEVF_GENERAL_TASK_INTERVAL * HZ);
-
-	hdev->stats_timer++;
-	hclgevf_task_schedule(hdev);
-}
-
 static void hclgevf_reset_service_task(struct work_struct *work)
 {
 #define	HCLGEVF_MAX_RESET_ATTEMPTS_CNT	3
@@ -1933,7 +1925,7 @@ static void hclgevf_service_task(struct work_struct *work)
 	struct hnae3_handle *handle;
 	struct hclgevf_dev *hdev;
 
-	hdev = container_of(work, struct hclgevf_dev, service_task);
+	hdev = container_of(work, struct hclgevf_dev, service_task.work);
 	handle = &hdev->nic;
 
 	if (hdev->stats_timer >= HCLGEVF_STATS_TIMER_INTERVAL) {
@@ -1953,6 +1945,8 @@ static void hclgevf_service_task(struct work_struct *work)
 	hclgevf_deferred_task_schedule(hdev);
 
 	clear_bit(HCLGEVF_STATE_SERVICE_SCHED, &hdev->state);
+
+	hclgevf_task_schedule(hdev);
 }
 
 static void hclgevf_clear_event_cause(struct hclgevf_dev *hdev, u32 regclr)
@@ -2194,10 +2188,10 @@ static void hclgevf_set_timer_task(struct hnae3_handle *handle, bool enable)
 	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
 
 	if (enable) {
-		mod_timer(&hdev->service_timer, jiffies + HZ);
+		hclgevf_task_schedule(hdev);
 	} else {
-		del_timer_sync(&hdev->service_timer);
-		cancel_work_sync(&hdev->service_task);
+		set_bit(HCLGEVF_STATE_DOWN, &hdev->state);
+		cancel_delayed_work_sync(&hdev->service_task);
 		clear_bit(HCLGEVF_STATE_SERVICE_SCHED, &hdev->state);
 	}
 }
@@ -2279,10 +2273,7 @@ static void hclgevf_state_init(struct hclgevf_dev *hdev)
 	clear_bit(HCLGEVF_STATE_MBX_SERVICE_SCHED, &hdev->state);
 	clear_bit(HCLGEVF_STATE_MBX_HANDLING, &hdev->state);
 
-	/* setup tasks for service timer */
-	timer_setup(&hdev->service_timer, hclgevf_service_timer, 0);
-
-	INIT_WORK(&hdev->service_task, hclgevf_service_task);
+	INIT_DELAYED_WORK(&hdev->service_task, hclgevf_service_task);
 	clear_bit(HCLGEVF_STATE_SERVICE_SCHED, &hdev->state);
 
 	INIT_WORK(&hdev->rst_service_task, hclgevf_reset_service_task);
@@ -2302,10 +2293,8 @@ static void hclgevf_state_uninit(struct hclgevf_dev *hdev)
 		del_timer_sync(&hdev->keep_alive_timer);
 	if (hdev->keep_alive_task.func)
 		cancel_work_sync(&hdev->keep_alive_task);
-	if (hdev->service_timer.function)
-		del_timer_sync(&hdev->service_timer);
-	if (hdev->service_task.func)
-		cancel_work_sync(&hdev->service_task);
+	if (hdev->service_task.work.func)
+		cancel_delayed_work_sync(&hdev->service_task);
 	if (hdev->mbx_service_task.func)
 		cancel_work_sync(&hdev->mbx_service_task);
 	if (hdev->rst_service_task.func)
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h
index 2f4c81bf4169..991a924da5ed 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h
@@ -283,9 +283,8 @@ struct hclgevf_dev {
 	struct hclgevf_mbx_resp_status mbx_resp; /* mailbox response */
 	struct hclgevf_mbx_arq_ring arq; /* mailbox async rx queue */
 
-	struct timer_list service_timer;
 	struct timer_list keep_alive_timer;
-	struct work_struct service_task;
+	struct delayed_work service_task;
 	struct work_struct keep_alive_task;
 	struct work_struct rst_service_task;
 	struct work_struct mbx_service_task;
-- 
cgit v1.2.3-59-g8ed1b


From 1c6dfe6fc6f77f3d124e63c2c2c3f9bb73f8d92c Mon Sep 17 00:00:00 2001
From: Yunsheng Lin <linyunsheng@huawei.com>
Date: Sat, 14 Dec 2019 10:06:38 +0800
Subject: net: hns3: remove mailbox and reset work in hclge_main

There are three work (mbx_service_task, service_task,
rst_service_task) in the HNS3 driver, mbx_service_task is for
handling mailbox work, service_task is for periodic management
issue and rst_service_task is for reset related issue, which can
be handled in a single work.

This patch removes the mbx_service_task and rst_service_task
work, and moves the related handling to the service_task work
in order to remove concurrency between the three work and to
improve efficiency.

BTW, since stats_timer in struct hclge_hw_stats is not needed
anymore, so removes the definition of struct hclge_hw_stats,
and moves mac_stats into struct hclge_dev.

Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c |  10 ++
 .../ethernet/hisilicon/hns3/hns3pf/hclge_main.c    | 163 +++++++++++++--------
 .../ethernet/hisilicon/hns3/hns3pf/hclge_main.h    |  17 +--
 .../net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c |   1 -
 4 files changed, 118 insertions(+), 73 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c
index 112df34b3869..92a99e4fdda5 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c
@@ -976,6 +976,14 @@ void hclge_dbg_dump_rst_info(struct hclge_dev *hdev)
 	dev_info(&hdev->pdev->dev, "hdev state: 0x%lx\n", hdev->state);
 }
 
+static void hclge_dbg_dump_serv_info(struct hclge_dev *hdev)
+{
+	dev_info(&hdev->pdev->dev, "last_serv_processed: %lu\n",
+		 hdev->last_serv_processed);
+	dev_info(&hdev->pdev->dev, "last_serv_cnt: %lu\n",
+		 hdev->serv_processed_cnt);
+}
+
 static void hclge_dbg_get_m7_stats_info(struct hclge_dev *hdev)
 {
 	struct hclge_desc *desc_src, *desc_tmp;
@@ -1227,6 +1235,8 @@ int hclge_dbg_run_cmd(struct hnae3_handle *handle, const char *cmd_buf)
 		hclge_dbg_dump_reg_cmd(hdev, &cmd_buf[sizeof(DUMP_REG)]);
 	} else if (strncmp(cmd_buf, "dump reset info", 15) == 0) {
 		hclge_dbg_dump_rst_info(hdev);
+	} else if (strncmp(cmd_buf, "dump serv info", 14) == 0) {
+		hclge_dbg_dump_serv_info(hdev);
 	} else if (strncmp(cmd_buf, "dump m7 info", 12) == 0) {
 		hclge_dbg_get_m7_stats_info(hdev);
 	} else if (strncmp(cmd_buf, "dump ncl_config", 15) == 0) {
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
index d862e9ba27e1..a0177e319550 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
@@ -416,7 +416,7 @@ static int hclge_mac_update_stats_defective(struct hclge_dev *hdev)
 {
 #define HCLGE_MAC_CMD_NUM 21
 
-	u64 *data = (u64 *)(&hdev->hw_stats.mac_stats);
+	u64 *data = (u64 *)(&hdev->mac_stats);
 	struct hclge_desc desc[HCLGE_MAC_CMD_NUM];
 	__le64 *desc_data;
 	int i, k, n;
@@ -453,7 +453,7 @@ static int hclge_mac_update_stats_defective(struct hclge_dev *hdev)
 
 static int hclge_mac_update_stats_complete(struct hclge_dev *hdev, u32 desc_num)
 {
-	u64 *data = (u64 *)(&hdev->hw_stats.mac_stats);
+	u64 *data = (u64 *)(&hdev->mac_stats);
 	struct hclge_desc *desc;
 	__le64 *desc_data;
 	u16 i, k, n;
@@ -802,7 +802,7 @@ static void hclge_get_stats(struct hnae3_handle *handle, u64 *data)
 	struct hclge_dev *hdev = vport->back;
 	u64 *p;
 
-	p = hclge_comm_get_stats(&hdev->hw_stats.mac_stats, g_mac_stats_string,
+	p = hclge_comm_get_stats(&hdev->mac_stats, g_mac_stats_string,
 				 ARRAY_SIZE(g_mac_stats_string), data);
 	p = hclge_tqps_get_stats(handle, p);
 }
@@ -815,8 +815,8 @@ static void hclge_get_mac_stat(struct hnae3_handle *handle,
 
 	hclge_update_stats(handle, NULL);
 
-	mac_stats->tx_pause_cnt = hdev->hw_stats.mac_stats.mac_tx_mac_pause_num;
-	mac_stats->rx_pause_cnt = hdev->hw_stats.mac_stats.mac_rx_mac_pause_num;
+	mac_stats->tx_pause_cnt = hdev->mac_stats.mac_tx_mac_pause_num;
+	mac_stats->rx_pause_cnt = hdev->mac_stats.mac_rx_mac_pause_num;
 }
 
 static int hclge_parse_func_status(struct hclge_dev *hdev,
@@ -2665,31 +2665,26 @@ static int hclge_mac_init(struct hclge_dev *hdev)
 
 static void hclge_mbx_task_schedule(struct hclge_dev *hdev)
 {
-	if (!test_bit(HCLGE_STATE_CMD_DISABLE, &hdev->state) &&
+	if (!test_bit(HCLGE_STATE_REMOVING, &hdev->state) &&
 	    !test_and_set_bit(HCLGE_STATE_MBX_SERVICE_SCHED, &hdev->state))
-		queue_work_on(cpumask_first(&hdev->affinity_mask), system_wq,
-			      &hdev->mbx_service_task);
+		mod_delayed_work_on(cpumask_first(&hdev->affinity_mask),
+				    system_wq, &hdev->service_task, 0);
 }
 
 static void hclge_reset_task_schedule(struct hclge_dev *hdev)
 {
 	if (!test_bit(HCLGE_STATE_REMOVING, &hdev->state) &&
 	    !test_and_set_bit(HCLGE_STATE_RST_SERVICE_SCHED, &hdev->state))
-		queue_work_on(cpumask_first(&hdev->affinity_mask), system_wq,
-			      &hdev->rst_service_task);
+		mod_delayed_work_on(cpumask_first(&hdev->affinity_mask),
+				    system_wq, &hdev->service_task, 0);
 }
 
 void hclge_task_schedule(struct hclge_dev *hdev, unsigned long delay_time)
 {
-	if (!test_bit(HCLGE_STATE_DOWN, &hdev->state) &&
-	    !test_bit(HCLGE_STATE_REMOVING, &hdev->state) &&
-	    !test_and_set_bit(HCLGE_STATE_SERVICE_SCHED, &hdev->state)) {
-		hdev->hw_stats.stats_timer++;
-		hdev->fd_arfs_expire_timer++;
+	if (!test_bit(HCLGE_STATE_REMOVING, &hdev->state))
 		mod_delayed_work_on(cpumask_first(&hdev->affinity_mask),
 				    system_wq, &hdev->service_task,
 				    delay_time);
-	}
 }
 
 static int hclge_get_mac_link_status(struct hclge_dev *hdev)
@@ -2748,6 +2743,10 @@ static void hclge_update_link_status(struct hclge_dev *hdev)
 
 	if (!client)
 		return;
+
+	if (test_and_set_bit(HCLGE_STATE_LINK_UPDATING, &hdev->state))
+		return;
+
 	state = hclge_get_mac_phy_link(hdev);
 	if (state != hdev->hw.mac.link) {
 		for (i = 0; i < hdev->num_vmdq_vport + 1; i++) {
@@ -2761,6 +2760,8 @@ static void hclge_update_link_status(struct hclge_dev *hdev)
 		}
 		hdev->hw.mac.link = state;
 	}
+
+	clear_bit(HCLGE_STATE_LINK_UPDATING, &hdev->state);
 }
 
 static void hclge_update_port_capability(struct hclge_mac *mac)
@@ -3352,6 +3353,18 @@ static int hclge_set_all_vf_rst(struct hclge_dev *hdev, bool reset)
 	return 0;
 }
 
+static void hclge_mailbox_service_task(struct hclge_dev *hdev)
+{
+	if (!test_and_clear_bit(HCLGE_STATE_MBX_SERVICE_SCHED, &hdev->state) ||
+	    test_bit(HCLGE_STATE_CMD_DISABLE, &hdev->state) ||
+	    test_and_set_bit(HCLGE_STATE_MBX_HANDLING, &hdev->state))
+		return;
+
+	hclge_mbx_handler(hdev);
+
+	clear_bit(HCLGE_STATE_MBX_HANDLING, &hdev->state);
+}
+
 static int hclge_func_reset_sync_vf(struct hclge_dev *hdev)
 {
 	struct hclge_pf_rst_sync_cmd *req;
@@ -3363,6 +3376,9 @@ static int hclge_func_reset_sync_vf(struct hclge_dev *hdev)
 	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_QUERY_VF_RST_RDY, true);
 
 	do {
+		/* vf need to down netdev by mbx during PF or FLR reset */
+		hclge_mailbox_service_task(hdev);
+
 		ret = hclge_cmd_send(&hdev->hw, &desc, 1);
 		/* for compatible with old firmware, wait
 		 * 100 ms for VF to stop IO
@@ -3939,36 +3955,19 @@ static void hclge_reset_subtask(struct hclge_dev *hdev)
 	hdev->reset_type = HNAE3_NONE_RESET;
 }
 
-static void hclge_reset_service_task(struct work_struct *work)
+static void hclge_reset_service_task(struct hclge_dev *hdev)
 {
-	struct hclge_dev *hdev =
-		container_of(work, struct hclge_dev, rst_service_task);
+	if (!test_and_clear_bit(HCLGE_STATE_RST_SERVICE_SCHED, &hdev->state))
+		return;
 
 	if (test_and_set_bit(HCLGE_STATE_RST_HANDLING, &hdev->state))
 		return;
 
-	clear_bit(HCLGE_STATE_RST_SERVICE_SCHED, &hdev->state);
-
 	hclge_reset_subtask(hdev);
 
 	clear_bit(HCLGE_STATE_RST_HANDLING, &hdev->state);
 }
 
-static void hclge_mailbox_service_task(struct work_struct *work)
-{
-	struct hclge_dev *hdev =
-		container_of(work, struct hclge_dev, mbx_service_task);
-
-	if (test_and_set_bit(HCLGE_STATE_MBX_HANDLING, &hdev->state))
-		return;
-
-	clear_bit(HCLGE_STATE_MBX_SERVICE_SCHED, &hdev->state);
-
-	hclge_mbx_handler(hdev);
-
-	clear_bit(HCLGE_STATE_MBX_HANDLING, &hdev->state);
-}
-
 static void hclge_update_vport_alive(struct hclge_dev *hdev)
 {
 	int i;
@@ -3986,29 +3985,62 @@ static void hclge_update_vport_alive(struct hclge_dev *hdev)
 	}
 }
 
-static void hclge_service_task(struct work_struct *work)
+static void hclge_periodic_service_task(struct hclge_dev *hdev)
 {
-	struct hclge_dev *hdev =
-		container_of(work, struct hclge_dev, service_task.work);
+	unsigned long delta = round_jiffies_relative(HZ);
 
-	clear_bit(HCLGE_STATE_SERVICE_SCHED, &hdev->state);
+	/* Always handle the link updating to make sure link state is
+	 * updated when it is triggered by mbx.
+	 */
+	hclge_update_link_status(hdev);
 
-	if (hdev->hw_stats.stats_timer >= HCLGE_STATS_TIMER_INTERVAL) {
-		hclge_update_stats_for_all(hdev);
-		hdev->hw_stats.stats_timer = 0;
+	if (time_is_after_jiffies(hdev->last_serv_processed + HZ)) {
+		delta = jiffies - hdev->last_serv_processed;
+
+		if (delta < round_jiffies_relative(HZ)) {
+			delta = round_jiffies_relative(HZ) - delta;
+			goto out;
+		}
 	}
 
-	hclge_update_port_info(hdev);
-	hclge_update_link_status(hdev);
+	hdev->serv_processed_cnt++;
 	hclge_update_vport_alive(hdev);
+
+	if (test_bit(HCLGE_STATE_DOWN, &hdev->state)) {
+		hdev->last_serv_processed = jiffies;
+		goto out;
+	}
+
+	if (!(hdev->serv_processed_cnt % HCLGE_STATS_TIMER_INTERVAL))
+		hclge_update_stats_for_all(hdev);
+
+	hclge_update_port_info(hdev);
 	hclge_sync_vlan_filter(hdev);
 
-	if (hdev->fd_arfs_expire_timer >= HCLGE_FD_ARFS_EXPIRE_TIMER_INTERVAL) {
+	if (!(hdev->serv_processed_cnt % HCLGE_ARFS_EXPIRE_INTERVAL))
 		hclge_rfs_filter_expire(hdev);
-		hdev->fd_arfs_expire_timer = 0;
-	}
 
-	hclge_task_schedule(hdev, round_jiffies_relative(HZ));
+	hdev->last_serv_processed = jiffies;
+
+out:
+	hclge_task_schedule(hdev, delta);
+}
+
+static void hclge_service_task(struct work_struct *work)
+{
+	struct hclge_dev *hdev =
+		container_of(work, struct hclge_dev, service_task.work);
+
+	hclge_reset_service_task(hdev);
+	hclge_mailbox_service_task(hdev);
+	hclge_periodic_service_task(hdev);
+
+	/* Handle reset and mbx again in case periodical task delays the
+	 * handling by calling hclge_task_schedule() in
+	 * hclge_periodic_service_task().
+	 */
+	hclge_reset_service_task(hdev);
+	hclge_mailbox_service_task(hdev);
 }
 
 struct hclge_vport *hclge_get_vport(struct hnae3_handle *handle)
@@ -6734,6 +6766,19 @@ static void hclge_reset_tqp_stats(struct hnae3_handle *handle)
 	}
 }
 
+static void hclge_flush_link_update(struct hclge_dev *hdev)
+{
+#define HCLGE_FLUSH_LINK_TIMEOUT	100000
+
+	unsigned long last = hdev->serv_processed_cnt;
+	int i = 0;
+
+	while (test_bit(HCLGE_STATE_LINK_UPDATING, &hdev->state) &&
+	       i++ < HCLGE_FLUSH_LINK_TIMEOUT &&
+	       last == hdev->serv_processed_cnt)
+		usleep_range(1, 1);
+}
+
 static void hclge_set_timer_task(struct hnae3_handle *handle, bool enable)
 {
 	struct hclge_vport *vport = hclge_get_vport(handle);
@@ -6742,12 +6787,12 @@ static void hclge_set_timer_task(struct hnae3_handle *handle, bool enable)
 	if (enable) {
 		hclge_task_schedule(hdev, round_jiffies_relative(HZ));
 	} else {
-		/* Set the DOWN flag here to disable the service to be
-		 * scheduled again
-		 */
+		/* Set the DOWN flag here to disable link updating */
 		set_bit(HCLGE_STATE_DOWN, &hdev->state);
-		cancel_delayed_work_sync(&hdev->service_task);
-		clear_bit(HCLGE_STATE_SERVICE_SCHED, &hdev->state);
+
+		/* flush memory to make sure DOWN is seen by service task */
+		smp_mb__before_atomic();
+		hclge_flush_link_update(hdev);
 	}
 }
 
@@ -9269,10 +9314,6 @@ static void hclge_state_uninit(struct hclge_dev *hdev)
 		del_timer_sync(&hdev->reset_timer);
 	if (hdev->service_task.work.func)
 		cancel_delayed_work_sync(&hdev->service_task);
-	if (hdev->rst_service_task.func)
-		cancel_work_sync(&hdev->rst_service_task);
-	if (hdev->mbx_service_task.func)
-		cancel_work_sync(&hdev->mbx_service_task);
 }
 
 static void hclge_flr_prepare(struct hnae3_ae_dev *ae_dev)
@@ -9477,8 +9518,6 @@ static int hclge_init_ae_dev(struct hnae3_ae_dev *ae_dev)
 
 	timer_setup(&hdev->reset_timer, hclge_reset_timer, 0);
 	INIT_DELAYED_WORK(&hdev->service_task, hclge_service_task);
-	INIT_WORK(&hdev->rst_service_task, hclge_reset_service_task);
-	INIT_WORK(&hdev->mbx_service_task, hclge_mailbox_service_task);
 
 	/* Setup affinity after service timer setup because add_timer_on
 	 * is called in affinity notify.
@@ -9512,6 +9551,8 @@ static int hclge_init_ae_dev(struct hnae3_ae_dev *ae_dev)
 	dev_info(&hdev->pdev->dev, "%s driver initialization finished.\n",
 		 HCLGE_DRIVER_NAME);
 
+	hclge_task_schedule(hdev, round_jiffies_relative(HZ));
+
 	return 0;
 
 err_mdiobus_unreg:
@@ -9534,7 +9575,7 @@ out:
 
 static void hclge_stats_clear(struct hclge_dev *hdev)
 {
-	memset(&hdev->hw_stats, 0, sizeof(hdev->hw_stats));
+	memset(&hdev->mac_stats, 0, sizeof(hdev->mac_stats));
 }
 
 static int hclge_set_mac_spoofchk(struct hclge_dev *hdev, int vf, bool enable)
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h
index ebb4c6e9aed3..ad40cf6be4f1 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h
@@ -208,13 +208,13 @@ enum HCLGE_DEV_STATE {
 	HCLGE_STATE_NIC_REGISTERED,
 	HCLGE_STATE_ROCE_REGISTERED,
 	HCLGE_STATE_SERVICE_INITED,
-	HCLGE_STATE_SERVICE_SCHED,
 	HCLGE_STATE_RST_SERVICE_SCHED,
 	HCLGE_STATE_RST_HANDLING,
 	HCLGE_STATE_MBX_SERVICE_SCHED,
 	HCLGE_STATE_MBX_HANDLING,
 	HCLGE_STATE_STATISTICS_UPDATING,
 	HCLGE_STATE_CMD_DISABLE,
+	HCLGE_STATE_LINK_UPDATING,
 	HCLGE_STATE_MAX
 };
 
@@ -454,11 +454,7 @@ struct hclge_mac_stats {
 	u64 mac_rx_ctrl_pkt_num;
 };
 
-#define HCLGE_STATS_TIMER_INTERVAL	(60 * 5)
-struct hclge_hw_stats {
-	struct hclge_mac_stats      mac_stats;
-	u32 stats_timer;
-};
+#define HCLGE_STATS_TIMER_INTERVAL	300UL
 
 struct hclge_vlan_type_cfg {
 	u16 rx_ot_fst_vlan_type;
@@ -549,7 +545,7 @@ struct key_info {
 
 /* assigned by firmware, the real filter number for each pf may be less */
 #define MAX_FD_FILTER_NUM	4096
-#define HCLGE_FD_ARFS_EXPIRE_TIMER_INTERVAL	5
+#define HCLGE_ARFS_EXPIRE_INTERVAL	5UL
 
 enum HCLGE_FD_ACTIVE_RULE_TYPE {
 	HCLGE_FD_RULE_NONE,
@@ -712,7 +708,7 @@ struct hclge_dev {
 	struct hnae3_ae_dev *ae_dev;
 	struct hclge_hw hw;
 	struct hclge_misc_vector misc_vector;
-	struct hclge_hw_stats hw_stats;
+	struct hclge_mac_stats mac_stats;
 	unsigned long state;
 	unsigned long flr_state;
 	unsigned long last_reset_time;
@@ -774,8 +770,6 @@ struct hclge_dev {
 	unsigned long service_timer_previous;
 	struct timer_list reset_timer;
 	struct delayed_work service_task;
-	struct work_struct rst_service_task;
-	struct work_struct mbx_service_task;
 
 	bool cur_promisc;
 	int num_alloc_vfs;	/* Actual number of VFs allocated */
@@ -811,7 +805,8 @@ struct hclge_dev {
 	struct hlist_head fd_rule_list;
 	spinlock_t fd_rule_lock; /* protect fd_rule_list and fd_bmap */
 	u16 hclge_fd_rule_num;
-	u16 fd_arfs_expire_timer;
+	unsigned long serv_processed_cnt;
+	unsigned long last_serv_processed;
 	unsigned long fd_bmap[BITS_TO_LONGS(MAX_FD_FILTER_NUM)];
 	enum HCLGE_FD_ACTIVE_RULE_TYPE fd_active_type;
 	u8 fd_en;
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c
index 0b433ebe6a2d..a79d1512119d 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c
@@ -635,7 +635,6 @@ static void hclge_handle_link_change_event(struct hclge_dev *hdev,
 #define LINK_STATUS_OFFSET	1
 #define LINK_FAIL_CODE_OFFSET	2
 
-	clear_bit(HCLGE_STATE_SERVICE_SCHED, &hdev->state);
 	hclge_task_schedule(hdev, 0);
 
 	if (!req->msg[LINK_STATUS_OFFSET])
-- 
cgit v1.2.3-59-g8ed1b


From ff200099d271cb14e4415449559da7cc6a38f4a1 Mon Sep 17 00:00:00 2001
From: Yunsheng Lin <linyunsheng@huawei.com>
Date: Sat, 14 Dec 2019 10:06:39 +0800
Subject: net: hns3: remove unnecessary work in hclgevf_main

There are four work (mbx_service_task, service_task,
rst_service_task and keep_alive_task)in the hclgevf module,
mbx_service_task is for handling mailbox issue, service_task is
for periodic management issue and rst_service_task is for reset
related issue, keep_alive_task is used to keepalive between PF
and VF, which can be done in a single work.

This patch removes the mbx_service_task, rst_service_task and
keep_alive_task, and moves the related handling to the
service_task work in order to remove concurrency between the four
work and to improve efficiency.

Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c  | 175 ++++++++++-----------
 .../ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h  |   9 +-
 2 files changed, 88 insertions(+), 96 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
index e48f8fe585bd..004dbaffee2d 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
@@ -440,6 +440,9 @@ void hclgevf_update_link_status(struct hclgevf_dev *hdev, int link_state)
 	struct hnae3_client *rclient;
 	struct hnae3_client *client;
 
+	if (test_and_set_bit(HCLGEVF_STATE_LINK_UPDATING, &hdev->state))
+		return;
+
 	client = handle->client;
 	rclient = hdev->roce_client;
 
@@ -452,6 +455,8 @@ void hclgevf_update_link_status(struct hclgevf_dev *hdev, int link_state)
 			rclient->ops->link_status_change(rhandle, !!link_state);
 		hdev->hw.mac.link = link_state;
 	}
+
+	clear_bit(HCLGEVF_STATE_LINK_UPDATING, &hdev->state);
 }
 
 static void hclgevf_update_link_mode(struct hclgevf_dev *hdev)
@@ -1767,54 +1772,38 @@ static void hclgevf_get_misc_vector(struct hclgevf_dev *hdev)
 
 void hclgevf_reset_task_schedule(struct hclgevf_dev *hdev)
 {
-	if (!test_bit(HCLGEVF_STATE_RST_SERVICE_SCHED, &hdev->state) &&
-	    !test_bit(HCLGEVF_STATE_REMOVING, &hdev->state)) {
-		set_bit(HCLGEVF_STATE_RST_SERVICE_SCHED, &hdev->state);
-		schedule_work(&hdev->rst_service_task);
-	}
+	if (!test_bit(HCLGEVF_STATE_REMOVING, &hdev->state) &&
+	    !test_and_set_bit(HCLGEVF_STATE_RST_SERVICE_SCHED,
+			      &hdev->state))
+		mod_delayed_work(system_wq, &hdev->service_task, 0);
 }
 
 void hclgevf_mbx_task_schedule(struct hclgevf_dev *hdev)
 {
-	if (!test_bit(HCLGEVF_STATE_MBX_SERVICE_SCHED, &hdev->state) &&
-	    !test_bit(HCLGEVF_STATE_MBX_HANDLING, &hdev->state)) {
-		set_bit(HCLGEVF_STATE_MBX_SERVICE_SCHED, &hdev->state);
-		schedule_work(&hdev->mbx_service_task);
-	}
+	if (!test_bit(HCLGEVF_STATE_REMOVING, &hdev->state) &&
+	    !test_and_set_bit(HCLGEVF_STATE_MBX_SERVICE_SCHED,
+			      &hdev->state))
+		mod_delayed_work(system_wq, &hdev->service_task, 0);
 }
 
-static void hclgevf_task_schedule(struct hclgevf_dev *hdev)
+static void hclgevf_task_schedule(struct hclgevf_dev *hdev,
+				  unsigned long delay)
 {
-	if (!test_bit(HCLGEVF_STATE_DOWN, &hdev->state)  &&
-	    !test_and_set_bit(HCLGEVF_STATE_SERVICE_SCHED, &hdev->state)) {
-		mod_delayed_work(system_wq, &hdev->service_task,
-				 round_jiffies_relative(HZ));
-		hdev->stats_timer++;
-	}
+	if (!test_bit(HCLGEVF_STATE_REMOVING, &hdev->state))
+		mod_delayed_work(system_wq, &hdev->service_task, delay);
 }
 
-static void hclgevf_deferred_task_schedule(struct hclgevf_dev *hdev)
-{
-	/* if we have any pending mailbox event then schedule the mbx task */
-	if (hdev->mbx_event_pending)
-		hclgevf_mbx_task_schedule(hdev);
-
-	if (test_bit(HCLGEVF_RESET_PENDING, &hdev->reset_state))
-		hclgevf_reset_task_schedule(hdev);
-}
-
-static void hclgevf_reset_service_task(struct work_struct *work)
+static void hclgevf_reset_service_task(struct hclgevf_dev *hdev)
 {
 #define	HCLGEVF_MAX_RESET_ATTEMPTS_CNT	3
 
-	struct hclgevf_dev *hdev =
-		container_of(work, struct hclgevf_dev, rst_service_task);
 	int ret;
 
-	if (test_and_set_bit(HCLGEVF_STATE_RST_HANDLING, &hdev->state))
+	if (!test_and_clear_bit(HCLGEVF_STATE_RST_SERVICE_SCHED, &hdev->state))
 		return;
 
-	clear_bit(HCLGEVF_STATE_RST_SERVICE_SCHED, &hdev->state);
+	if (test_and_set_bit(HCLGEVF_STATE_RST_HANDLING, &hdev->state))
+		return;
 
 	if (test_and_clear_bit(HCLGEVF_RESET_PENDING,
 			       &hdev->reset_state)) {
@@ -1877,39 +1866,24 @@ static void hclgevf_reset_service_task(struct work_struct *work)
 	clear_bit(HCLGEVF_STATE_RST_HANDLING, &hdev->state);
 }
 
-static void hclgevf_mailbox_service_task(struct work_struct *work)
+static void hclgevf_mailbox_service_task(struct hclgevf_dev *hdev)
 {
-	struct hclgevf_dev *hdev;
-
-	hdev = container_of(work, struct hclgevf_dev, mbx_service_task);
+	if (!test_and_clear_bit(HCLGEVF_STATE_MBX_SERVICE_SCHED, &hdev->state))
+		return;
 
 	if (test_and_set_bit(HCLGEVF_STATE_MBX_HANDLING, &hdev->state))
 		return;
 
-	clear_bit(HCLGEVF_STATE_MBX_SERVICE_SCHED, &hdev->state);
-
 	hclgevf_mbx_async_handler(hdev);
 
 	clear_bit(HCLGEVF_STATE_MBX_HANDLING, &hdev->state);
 }
 
-static void hclgevf_keep_alive_timer(struct timer_list *t)
-{
-	struct hclgevf_dev *hdev = from_timer(hdev, t, keep_alive_timer);
-
-	schedule_work(&hdev->keep_alive_task);
-	mod_timer(&hdev->keep_alive_timer, jiffies +
-		  HCLGEVF_KEEP_ALIVE_TASK_INTERVAL * HZ);
-}
-
-static void hclgevf_keep_alive_task(struct work_struct *work)
+static void hclgevf_keep_alive(struct hclgevf_dev *hdev)
 {
-	struct hclgevf_dev *hdev;
 	u8 respmsg;
 	int ret;
 
-	hdev = container_of(work, struct hclgevf_dev, keep_alive_task);
-
 	if (test_bit(HCLGEVF_STATE_CMD_DISABLE, &hdev->state))
 		return;
 
@@ -1920,19 +1894,32 @@ static void hclgevf_keep_alive_task(struct work_struct *work)
 			"VF sends keep alive cmd failed(=%d)\n", ret);
 }
 
-static void hclgevf_service_task(struct work_struct *work)
+static void hclgevf_periodic_service_task(struct hclgevf_dev *hdev)
 {
-	struct hnae3_handle *handle;
-	struct hclgevf_dev *hdev;
+	unsigned long delta = round_jiffies_relative(HZ);
+	struct hnae3_handle *handle = &hdev->nic;
 
-	hdev = container_of(work, struct hclgevf_dev, service_task.work);
-	handle = &hdev->nic;
+	if (time_is_after_jiffies(hdev->last_serv_processed + HZ)) {
+		delta = jiffies - hdev->last_serv_processed;
 
-	if (hdev->stats_timer >= HCLGEVF_STATS_TIMER_INTERVAL) {
-		hclgevf_tqps_update_stats(handle);
-		hdev->stats_timer = 0;
+		if (delta < round_jiffies_relative(HZ)) {
+			delta = round_jiffies_relative(HZ) - delta;
+			goto out;
+		}
+	}
+
+	hdev->serv_processed_cnt++;
+	if (!(hdev->serv_processed_cnt % HCLGEVF_KEEP_ALIVE_TASK_INTERVAL))
+		hclgevf_keep_alive(hdev);
+
+	if (test_bit(HCLGEVF_STATE_DOWN, &hdev->state)) {
+		hdev->last_serv_processed = jiffies;
+		goto out;
 	}
 
+	if (!(hdev->serv_processed_cnt % HCLGEVF_STATS_TIMER_INTERVAL))
+		hclgevf_tqps_update_stats(handle);
+
 	/* request the link status from the PF. PF would be able to tell VF
 	 * about such updates in future so we might remove this later
 	 */
@@ -1942,11 +1929,27 @@ static void hclgevf_service_task(struct work_struct *work)
 
 	hclgevf_sync_vlan_filter(hdev);
 
-	hclgevf_deferred_task_schedule(hdev);
+	hdev->last_serv_processed = jiffies;
 
-	clear_bit(HCLGEVF_STATE_SERVICE_SCHED, &hdev->state);
+out:
+	hclgevf_task_schedule(hdev, delta);
+}
+
+static void hclgevf_service_task(struct work_struct *work)
+{
+	struct hclgevf_dev *hdev = container_of(work, struct hclgevf_dev,
+						service_task.work);
 
-	hclgevf_task_schedule(hdev);
+	hclgevf_reset_service_task(hdev);
+	hclgevf_mailbox_service_task(hdev);
+	hclgevf_periodic_service_task(hdev);
+
+	/* Handle reset and mbx again in case periodical task delays the
+	 * handling by calling hclgevf_task_schedule() in
+	 * hclgevf_periodic_service_task()
+	 */
+	hclgevf_reset_service_task(hdev);
+	hclgevf_mailbox_service_task(hdev);
 }
 
 static void hclgevf_clear_event_cause(struct hclgevf_dev *hdev, u32 regclr)
@@ -2183,16 +2186,31 @@ static int hclgevf_init_vlan_config(struct hclgevf_dev *hdev)
 				       false);
 }
 
+static void hclgevf_flush_link_update(struct hclgevf_dev *hdev)
+{
+#define HCLGEVF_FLUSH_LINK_TIMEOUT	100000
+
+	unsigned long last = hdev->serv_processed_cnt;
+	int i = 0;
+
+	while (test_bit(HCLGEVF_STATE_LINK_UPDATING, &hdev->state) &&
+	       i++ < HCLGEVF_FLUSH_LINK_TIMEOUT &&
+	       last == hdev->serv_processed_cnt)
+		usleep_range(1, 1);
+}
+
 static void hclgevf_set_timer_task(struct hnae3_handle *handle, bool enable)
 {
 	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
 
 	if (enable) {
-		hclgevf_task_schedule(hdev);
+		hclgevf_task_schedule(hdev, 0);
 	} else {
 		set_bit(HCLGEVF_STATE_DOWN, &hdev->state);
-		cancel_delayed_work_sync(&hdev->service_task);
-		clear_bit(HCLGEVF_STATE_SERVICE_SCHED, &hdev->state);
+
+		/* flush memory to make sure DOWN is seen by service task */
+		smp_mb__before_atomic();
+		hclgevf_flush_link_update(hdev);
 	}
 }
 
@@ -2239,16 +2257,12 @@ static int hclgevf_set_alive(struct hnae3_handle *handle, bool alive)
 
 static int hclgevf_client_start(struct hnae3_handle *handle)
 {
-	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
 	int ret;
 
 	ret = hclgevf_set_alive(handle, true);
 	if (ret)
 		return ret;
 
-	mod_timer(&hdev->keep_alive_timer, jiffies +
-		  HCLGEVF_KEEP_ALIVE_TASK_INTERVAL * HZ);
-
 	return 0;
 }
 
@@ -2261,22 +2275,14 @@ static void hclgevf_client_stop(struct hnae3_handle *handle)
 	if (ret)
 		dev_warn(&hdev->pdev->dev,
 			 "%s failed %d\n", __func__, ret);
-
-	del_timer_sync(&hdev->keep_alive_timer);
-	cancel_work_sync(&hdev->keep_alive_task);
 }
 
 static void hclgevf_state_init(struct hclgevf_dev *hdev)
 {
-	/* setup tasks for the MBX */
-	INIT_WORK(&hdev->mbx_service_task, hclgevf_mailbox_service_task);
 	clear_bit(HCLGEVF_STATE_MBX_SERVICE_SCHED, &hdev->state);
 	clear_bit(HCLGEVF_STATE_MBX_HANDLING, &hdev->state);
 
 	INIT_DELAYED_WORK(&hdev->service_task, hclgevf_service_task);
-	clear_bit(HCLGEVF_STATE_SERVICE_SCHED, &hdev->state);
-
-	INIT_WORK(&hdev->rst_service_task, hclgevf_reset_service_task);
 
 	mutex_init(&hdev->mbx_resp.mbx_mutex);
 
@@ -2289,16 +2295,8 @@ static void hclgevf_state_uninit(struct hclgevf_dev *hdev)
 	set_bit(HCLGEVF_STATE_DOWN, &hdev->state);
 	set_bit(HCLGEVF_STATE_REMOVING, &hdev->state);
 
-	if (hdev->keep_alive_timer.function)
-		del_timer_sync(&hdev->keep_alive_timer);
-	if (hdev->keep_alive_task.func)
-		cancel_work_sync(&hdev->keep_alive_task);
 	if (hdev->service_task.work.func)
 		cancel_delayed_work_sync(&hdev->service_task);
-	if (hdev->mbx_service_task.func)
-		cancel_work_sync(&hdev->mbx_service_task);
-	if (hdev->rst_service_task.func)
-		cancel_work_sync(&hdev->rst_service_task);
 
 	mutex_destroy(&hdev->mbx_resp.mbx_mutex);
 }
@@ -2796,6 +2794,8 @@ static int hclgevf_init_hdev(struct hclgevf_dev *hdev)
 	dev_info(&hdev->pdev->dev, "finished initializing %s driver\n",
 		 HCLGEVF_DRIVER_NAME);
 
+	hclgevf_task_schedule(hdev, round_jiffies_relative(HZ));
+
 	return 0;
 
 err_config:
@@ -2827,7 +2827,6 @@ static void hclgevf_uninit_hdev(struct hclgevf_dev *hdev)
 static int hclgevf_init_ae_dev(struct hnae3_ae_dev *ae_dev)
 {
 	struct pci_dev *pdev = ae_dev->pdev;
-	struct hclgevf_dev *hdev;
 	int ret;
 
 	ret = hclgevf_alloc_hdev(ae_dev);
@@ -2842,10 +2841,6 @@ static int hclgevf_init_ae_dev(struct hnae3_ae_dev *ae_dev)
 		return ret;
 	}
 
-	hdev = ae_dev->priv;
-	timer_setup(&hdev->keep_alive_timer, hclgevf_keep_alive_timer, 0);
-	INIT_WORK(&hdev->keep_alive_task, hclgevf_keep_alive_task);
-
 	return 0;
 }
 
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h
index 991a924da5ed..450e5878293f 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h
@@ -142,12 +142,12 @@ enum hclgevf_states {
 	HCLGEVF_STATE_REMOVING,
 	HCLGEVF_STATE_NIC_REGISTERED,
 	/* task states */
-	HCLGEVF_STATE_SERVICE_SCHED,
 	HCLGEVF_STATE_RST_SERVICE_SCHED,
 	HCLGEVF_STATE_RST_HANDLING,
 	HCLGEVF_STATE_MBX_SERVICE_SCHED,
 	HCLGEVF_STATE_MBX_HANDLING,
 	HCLGEVF_STATE_CMD_DISABLE,
+	HCLGEVF_STATE_LINK_UPDATING,
 };
 
 struct hclgevf_mac {
@@ -283,11 +283,7 @@ struct hclgevf_dev {
 	struct hclgevf_mbx_resp_status mbx_resp; /* mailbox response */
 	struct hclgevf_mbx_arq_ring arq; /* mailbox async rx queue */
 
-	struct timer_list keep_alive_timer;
 	struct delayed_work service_task;
-	struct work_struct keep_alive_task;
-	struct work_struct rst_service_task;
-	struct work_struct mbx_service_task;
 
 	struct hclgevf_tqp *htqp;
 
@@ -297,7 +293,8 @@ struct hclgevf_dev {
 	struct hnae3_client *nic_client;
 	struct hnae3_client *roce_client;
 	u32 flag;
-	u32 stats_timer;
+	unsigned long serv_processed_cnt;
+	unsigned long last_serv_processed;
 };
 
 static inline bool hclgevf_is_reset_pending(struct hclgevf_dev *hdev)
-- 
cgit v1.2.3-59-g8ed1b


From 0ea68902256e6cc705c94d1844dd8cf0805fd3e9 Mon Sep 17 00:00:00 2001
From: Yunsheng Lin <linyunsheng@huawei.com>
Date: Sat, 14 Dec 2019 10:06:40 +0800
Subject: net: hns3: allocate WQ with WQ_MEM_RECLAIM flag

The hns3 driver may be used in memory reclaim path when it
is the low level transport of a network file system, so it
needs to guarantee forward progress even under memory pressure.

This patch allocates a private WQ with WQ_MEM_RECLAIM set for
both hclge_main and hclgevf_main modules.

Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c   | 15 ++++++++++++---
 drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c | 15 ++++++++++++---
 2 files changed, 24 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
index a0177e319550..5129b4a55f9b 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
@@ -72,6 +72,8 @@ static int hclge_set_default_loopback(struct hclge_dev *hdev);
 
 static struct hnae3_ae_algo ae_algo;
 
+static struct workqueue_struct *hclge_wq;
+
 static const struct pci_device_id ae_algo_pci_tbl[] = {
 	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_GE), 0},
 	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_25GE), 0},
@@ -2668,7 +2670,7 @@ static void hclge_mbx_task_schedule(struct hclge_dev *hdev)
 	if (!test_bit(HCLGE_STATE_REMOVING, &hdev->state) &&
 	    !test_and_set_bit(HCLGE_STATE_MBX_SERVICE_SCHED, &hdev->state))
 		mod_delayed_work_on(cpumask_first(&hdev->affinity_mask),
-				    system_wq, &hdev->service_task, 0);
+				    hclge_wq, &hdev->service_task, 0);
 }
 
 static void hclge_reset_task_schedule(struct hclge_dev *hdev)
@@ -2676,14 +2678,14 @@ static void hclge_reset_task_schedule(struct hclge_dev *hdev)
 	if (!test_bit(HCLGE_STATE_REMOVING, &hdev->state) &&
 	    !test_and_set_bit(HCLGE_STATE_RST_SERVICE_SCHED, &hdev->state))
 		mod_delayed_work_on(cpumask_first(&hdev->affinity_mask),
-				    system_wq, &hdev->service_task, 0);
+				    hclge_wq, &hdev->service_task, 0);
 }
 
 void hclge_task_schedule(struct hclge_dev *hdev, unsigned long delay_time)
 {
 	if (!test_bit(HCLGE_STATE_REMOVING, &hdev->state))
 		mod_delayed_work_on(cpumask_first(&hdev->affinity_mask),
-				    system_wq, &hdev->service_task,
+				    hclge_wq, &hdev->service_task,
 				    delay_time);
 }
 
@@ -10652,6 +10654,12 @@ static int hclge_init(void)
 {
 	pr_info("%s is initializing\n", HCLGE_NAME);
 
+	hclge_wq = alloc_workqueue("%s", WQ_MEM_RECLAIM, 0, HCLGE_NAME);
+	if (!hclge_wq) {
+		pr_err("%s: failed to create workqueue\n", HCLGE_NAME);
+		return -ENOMEM;
+	}
+
 	hnae3_register_ae_algo(&ae_algo);
 
 	return 0;
@@ -10660,6 +10668,7 @@ static int hclge_init(void)
 static void hclge_exit(void)
 {
 	hnae3_unregister_ae_algo(&ae_algo);
+	destroy_workqueue(hclge_wq);
 }
 module_init(hclge_init);
 module_exit(hclge_exit);
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
index 004dbaffee2d..b56c19afc3b9 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
@@ -16,6 +16,8 @@
 static int hclgevf_reset_hdev(struct hclgevf_dev *hdev);
 static struct hnae3_ae_algo ae_algovf;
 
+static struct workqueue_struct *hclgevf_wq;
+
 static const struct pci_device_id ae_algovf_pci_tbl[] = {
 	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_100G_VF), 0},
 	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_100G_RDMA_DCB_PFC_VF), 0},
@@ -1775,7 +1777,7 @@ void hclgevf_reset_task_schedule(struct hclgevf_dev *hdev)
 	if (!test_bit(HCLGEVF_STATE_REMOVING, &hdev->state) &&
 	    !test_and_set_bit(HCLGEVF_STATE_RST_SERVICE_SCHED,
 			      &hdev->state))
-		mod_delayed_work(system_wq, &hdev->service_task, 0);
+		mod_delayed_work(hclgevf_wq, &hdev->service_task, 0);
 }
 
 void hclgevf_mbx_task_schedule(struct hclgevf_dev *hdev)
@@ -1783,14 +1785,14 @@ void hclgevf_mbx_task_schedule(struct hclgevf_dev *hdev)
 	if (!test_bit(HCLGEVF_STATE_REMOVING, &hdev->state) &&
 	    !test_and_set_bit(HCLGEVF_STATE_MBX_SERVICE_SCHED,
 			      &hdev->state))
-		mod_delayed_work(system_wq, &hdev->service_task, 0);
+		mod_delayed_work(hclgevf_wq, &hdev->service_task, 0);
 }
 
 static void hclgevf_task_schedule(struct hclgevf_dev *hdev,
 				  unsigned long delay)
 {
 	if (!test_bit(HCLGEVF_STATE_REMOVING, &hdev->state))
-		mod_delayed_work(system_wq, &hdev->service_task, delay);
+		mod_delayed_work(hclgevf_wq, &hdev->service_task, delay);
 }
 
 static void hclgevf_reset_service_task(struct hclgevf_dev *hdev)
@@ -3197,6 +3199,12 @@ static int hclgevf_init(void)
 {
 	pr_info("%s is initializing\n", HCLGEVF_NAME);
 
+	hclgevf_wq = alloc_workqueue("%s", WQ_MEM_RECLAIM, 0, HCLGEVF_NAME);
+	if (!hclgevf_wq) {
+		pr_err("%s: failed to create workqueue\n", HCLGEVF_NAME);
+		return -ENOMEM;
+	}
+
 	hnae3_register_ae_algo(&ae_algovf);
 
 	return 0;
@@ -3205,6 +3213,7 @@ static int hclgevf_init(void)
 static void hclgevf_exit(void)
 {
 	hnae3_unregister_ae_algo(&ae_algovf);
+	destroy_workqueue(hclgevf_wq);
 }
 module_init(hclgevf_init);
 module_exit(hclgevf_exit);
-- 
cgit v1.2.3-59-g8ed1b


From d5432455b1a869d92d5643de0b4875bbcfbcefed Mon Sep 17 00:00:00 2001
From: Guojia Liao <liaoguojia@huawei.com>
Date: Sat, 14 Dec 2019 10:06:41 +0800
Subject: net: hns3: do not schedule the periodic task when reset fail

service_task will be scheduled  per second to do some periodic
jobs. When reset fails, it means this device is not available
now, so the periodic jobs do not need to be handled.

This patch adds flag HCLGE_STATE_RST_FAIL/HCLGEVF_STATE_RST_FAIL
to indicate that reset fails, and checks this flag before
schedule periodic task.

Signed-off-by: Guojia Liao <liaoguojia@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c   | 7 ++++++-
 drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h   | 1 +
 drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c | 6 +++++-
 drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h | 1 +
 4 files changed, 13 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
index 5129b4a55f9b..4e7a07877e5d 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
@@ -2683,7 +2683,8 @@ static void hclge_reset_task_schedule(struct hclge_dev *hdev)
 
 void hclge_task_schedule(struct hclge_dev *hdev, unsigned long delay_time)
 {
-	if (!test_bit(HCLGE_STATE_REMOVING, &hdev->state))
+	if (!test_bit(HCLGE_STATE_REMOVING, &hdev->state) &&
+	    !test_bit(HCLGE_STATE_RST_FAIL, &hdev->state))
 		mod_delayed_work_on(cpumask_first(&hdev->affinity_mask),
 				    hclge_wq, &hdev->service_task,
 				    delay_time);
@@ -3690,6 +3691,8 @@ static bool hclge_reset_err_handle(struct hclge_dev *hdev)
 
 	hclge_dbg_dump_rst_info(hdev);
 
+	set_bit(HCLGE_STATE_RST_FAIL, &hdev->state);
+
 	return false;
 }
 
@@ -3843,6 +3846,7 @@ static void hclge_reset(struct hclge_dev *hdev)
 	hdev->rst_stats.reset_fail_cnt = 0;
 	hdev->rst_stats.reset_done_cnt++;
 	ae_dev->reset_type = HNAE3_NONE_RESET;
+	clear_bit(HCLGE_STATE_RST_FAIL, &hdev->state);
 
 	/* if default_reset_request has a higher level reset request,
 	 * it should be handled as soon as possible. since some errors
@@ -9303,6 +9307,7 @@ static void hclge_state_init(struct hclge_dev *hdev)
 	set_bit(HCLGE_STATE_DOWN, &hdev->state);
 	clear_bit(HCLGE_STATE_RST_SERVICE_SCHED, &hdev->state);
 	clear_bit(HCLGE_STATE_RST_HANDLING, &hdev->state);
+	clear_bit(HCLGE_STATE_RST_FAIL, &hdev->state);
 	clear_bit(HCLGE_STATE_MBX_SERVICE_SCHED, &hdev->state);
 	clear_bit(HCLGE_STATE_MBX_HANDLING, &hdev->state);
 }
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h
index ad40cf6be4f1..3a913976b820 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h
@@ -215,6 +215,7 @@ enum HCLGE_DEV_STATE {
 	HCLGE_STATE_STATISTICS_UPDATING,
 	HCLGE_STATE_CMD_DISABLE,
 	HCLGE_STATE_LINK_UPDATING,
+	HCLGE_STATE_RST_FAIL,
 	HCLGE_STATE_MAX
 };
 
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
index b56c19afc3b9..c33b8027f801 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
@@ -1598,6 +1598,7 @@ static void hclgevf_reset_err_handle(struct hclgevf_dev *hdev)
 		set_bit(HCLGEVF_RESET_PENDING, &hdev->reset_state);
 		hclgevf_reset_task_schedule(hdev);
 	} else {
+		set_bit(HCLGEVF_STATE_RST_FAIL, &hdev->state);
 		hclgevf_dump_rst_info(hdev);
 	}
 }
@@ -1659,6 +1660,7 @@ static int hclgevf_reset(struct hclgevf_dev *hdev)
 	ae_dev->reset_type = HNAE3_NONE_RESET;
 	hdev->rst_stats.rst_done_cnt++;
 	hdev->rst_stats.rst_fail_cnt = 0;
+	clear_bit(HCLGEVF_STATE_RST_FAIL, &hdev->state);
 
 	return ret;
 err_reset_lock:
@@ -1791,7 +1793,8 @@ void hclgevf_mbx_task_schedule(struct hclgevf_dev *hdev)
 static void hclgevf_task_schedule(struct hclgevf_dev *hdev,
 				  unsigned long delay)
 {
-	if (!test_bit(HCLGEVF_STATE_REMOVING, &hdev->state))
+	if (!test_bit(HCLGEVF_STATE_REMOVING, &hdev->state) &&
+	    !test_bit(HCLGEVF_STATE_RST_FAIL, &hdev->state))
 		mod_delayed_work(hclgevf_wq, &hdev->service_task, delay);
 }
 
@@ -2283,6 +2286,7 @@ static void hclgevf_state_init(struct hclgevf_dev *hdev)
 {
 	clear_bit(HCLGEVF_STATE_MBX_SERVICE_SCHED, &hdev->state);
 	clear_bit(HCLGEVF_STATE_MBX_HANDLING, &hdev->state);
+	clear_bit(HCLGEVF_STATE_RST_FAIL, &hdev->state);
 
 	INIT_DELAYED_WORK(&hdev->service_task, hclgevf_service_task);
 
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h
index 450e5878293f..003114f6db6c 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h
@@ -148,6 +148,7 @@ enum hclgevf_states {
 	HCLGEVF_STATE_MBX_HANDLING,
 	HCLGEVF_STATE_CMD_DISABLE,
 	HCLGEVF_STATE_LINK_UPDATING,
+	HCLGEVF_STATE_RST_FAIL,
 };
 
 struct hclgevf_mac {
-- 
cgit v1.2.3-59-g8ed1b


From ba8381236fd0fa0ab4e8383fb752c843a31dd019 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Sat, 14 Dec 2019 17:53:06 +0200
Subject: net: fib_notifier: Add temporary events to the FIB notification chain

Subsequent patches are going to simplify the IPv4 route offload API,
which will only use two events - replace and delete.

Introduce a temporary version of these two events in order to make the
conversion easier to review.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/fib_notifier.h | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/include/net/fib_notifier.h b/include/net/fib_notifier.h
index 6d59221ff05a..b3c54325caec 100644
--- a/include/net/fib_notifier.h
+++ b/include/net/fib_notifier.h
@@ -23,6 +23,8 @@ enum fib_event_type {
 	FIB_EVENT_NH_DEL,
 	FIB_EVENT_VIF_ADD,
 	FIB_EVENT_VIF_DEL,
+	FIB_EVENT_ENTRY_REPLACE_TMP,
+	FIB_EVENT_ENTRY_DEL_TMP,
 };
 
 struct fib_notifier_ops {
-- 
cgit v1.2.3-59-g8ed1b


From a6c76c17df021b141b0d306828c9fe4ba2d2717c Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Sat, 14 Dec 2019 17:53:07 +0200
Subject: ipv4: Notify route after insertion to the routing table

Currently, a new route is notified in the FIB notification chain before
it is inserted to the FIB alias list.

Subsequent patches will use the placement of the new route in the
ordered FIB alias list in order to determine if the route should be
notified or not.

As a preparatory step, change the order so that the route is first
inserted into the FIB alias list and only then notified.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/fib_trie.c | 29 ++++++++++++++---------------
 1 file changed, 14 insertions(+), 15 deletions(-)

diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index b9df9c09b84e..9264d6628e9f 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -1063,9 +1063,6 @@ noleaf:
 	return -ENOMEM;
 }
 
-/* fib notifier for ADD is sent before calling fib_insert_alias with
- * the expectation that the only possible failure ENOMEM
- */
 static int fib_insert_alias(struct trie *t, struct key_vector *tp,
 			    struct key_vector *l, struct fib_alias *new,
 			    struct fib_alias *fa, t_key key)
@@ -1118,6 +1115,9 @@ static bool fib_valid_key_len(u32 key, u8 plen, struct netlink_ext_ack *extack)
 	return true;
 }
 
+static void fib_remove_alias(struct trie *t, struct key_vector *tp,
+			     struct key_vector *l, struct fib_alias *old);
+
 /* Caller must hold RTNL. */
 int fib_table_insert(struct net *net, struct fib_table *tb,
 		     struct fib_config *cfg, struct netlink_ext_ack *extack)
@@ -1269,14 +1269,19 @@ int fib_table_insert(struct net *net, struct fib_table *tb,
 	new_fa->tb_id = tb->tb_id;
 	new_fa->fa_default = -1;
 
-	err = call_fib_entry_notifiers(net, event, key, plen, new_fa, extack);
+	/* Insert new entry to the list. */
+	err = fib_insert_alias(t, tp, l, new_fa, fa, key);
 	if (err)
 		goto out_free_new_fa;
 
-	/* Insert new entry to the list. */
-	err = fib_insert_alias(t, tp, l, new_fa, fa, key);
+	/* The alias was already inserted, so the node must exist. */
+	l = l ? l : fib_find_node(t, &tp, key);
+	if (WARN_ON_ONCE(!l))
+		goto out_free_new_fa;
+
+	err = call_fib_entry_notifiers(net, event, key, plen, new_fa, extack);
 	if (err)
-		goto out_fib_notif;
+		goto out_remove_new_fa;
 
 	if (!plen)
 		tb->tb_num_default++;
@@ -1287,14 +1292,8 @@ int fib_table_insert(struct net *net, struct fib_table *tb,
 succeeded:
 	return 0;
 
-out_fib_notif:
-	/* notifier was sent that entry would be added to trie, but
-	 * the add failed and need to recover. Only failure for
-	 * fib_insert_alias is ENOMEM.
-	 */
-	NL_SET_ERR_MSG(extack, "Failed to insert route into trie");
-	call_fib_entry_notifiers(net, FIB_EVENT_ENTRY_DEL, key,
-				 plen, new_fa, NULL);
+out_remove_new_fa:
+	fib_remove_alias(t, tp, l, new_fa);
 out_free_new_fa:
 	kmem_cache_free(fn_alias_kmem, new_fa);
 out:
-- 
cgit v1.2.3-59-g8ed1b


From b5fc0430dcbb36294b29a25cddb3bf7e92fe2307 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Sat, 14 Dec 2019 17:53:08 +0200
Subject: ipv4: Extend FIB alias find function

Extend the function with another argument, 'find_first'. When set, the
function returns the first FIB alias with the matching {prefix, prefix
length, table ID}. The TOS and priority parameters are ignored. Current
callers are converted to pass 'false' in order to maintain existing
behavior.

This will be used by subsequent patches in the series.

v2:
* New patch

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Suggested-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/fib_trie.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index 9264d6628e9f..4c22e768a5b5 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -980,9 +980,12 @@ static struct key_vector *fib_find_node(struct trie *t,
 
 /* Return the first fib alias matching TOS with
  * priority less than or equal to PRIO.
+ * If 'find_first' is set, return the first matching
+ * fib alias, regardless of TOS and priority.
  */
 static struct fib_alias *fib_find_alias(struct hlist_head *fah, u8 slen,
-					u8 tos, u32 prio, u32 tb_id)
+					u8 tos, u32 prio, u32 tb_id,
+					bool find_first)
 {
 	struct fib_alias *fa;
 
@@ -998,6 +1001,8 @@ static struct fib_alias *fib_find_alias(struct hlist_head *fah, u8 slen,
 			continue;
 		if (fa->tb_id != tb_id)
 			break;
+		if (find_first)
+			return fa;
 		if (fa->fa_tos > tos)
 			continue;
 		if (fa->fa_info->fib_priority >= prio || fa->fa_tos < tos)
@@ -1149,7 +1154,7 @@ int fib_table_insert(struct net *net, struct fib_table *tb,
 
 	l = fib_find_node(t, &tp, key);
 	fa = l ? fib_find_alias(&l->leaf, slen, tos, fi->fib_priority,
-				tb->tb_id) : NULL;
+				tb->tb_id, false) : NULL;
 
 	/* Now fa, if non-NULL, points to the first fib alias
 	 * with the same keys [prefix,tos,priority], if such key already
@@ -1565,7 +1570,7 @@ int fib_table_delete(struct net *net, struct fib_table *tb,
 	if (!l)
 		return -ESRCH;
 
-	fa = fib_find_alias(&l->leaf, slen, tos, 0, tb->tb_id);
+	fa = fib_find_alias(&l->leaf, slen, tos, 0, tb->tb_id, false);
 	if (!fa)
 		return -ESRCH;
 
-- 
cgit v1.2.3-59-g8ed1b


From ee3936d658821d26de9039c64c90750ba240989d Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Sat, 14 Dec 2019 17:53:09 +0200
Subject: ipv4: Notify route if replacing currently offloaded one

When replacing a route, its replacement should only be notified in case
the replaced route is of any interest to listeners. In other words, if
the replaced route is currently used in the data path, which means it is
the first route in the FIB alias list with the given {prefix, prefix
length, table ID}.

v2:
* Convert to use fib_find_alias() instead of fib_find_first_alias()

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/fib_trie.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index 4c22e768a5b5..4c80ac0344f4 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -1222,6 +1222,17 @@ int fib_table_insert(struct net *net, struct fib_table *tb,
 			new_fa->tb_id = tb->tb_id;
 			new_fa->fa_default = -1;
 
+			if (fib_find_alias(&l->leaf, fa->fa_slen, 0, 0,
+					   tb->tb_id, true) == fa) {
+				enum fib_event_type fib_event;
+
+				fib_event = FIB_EVENT_ENTRY_REPLACE_TMP;
+				err = call_fib_entry_notifiers(net, fib_event,
+							       key, plen,
+							       new_fa, extack);
+				if (err)
+					goto out_free_new_fa;
+			}
 			err = call_fib_entry_notifiers(net,
 						       FIB_EVENT_ENTRY_REPLACE,
 						       key, plen, new_fa,
-- 
cgit v1.2.3-59-g8ed1b


From a8674f753e36f566d6c1d992ab85323d784281d9 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Sat, 14 Dec 2019 17:53:10 +0200
Subject: ipv4: Notify newly added route if should be offloaded

When a route is added, it should only be notified in case it is the
first route in the FIB alias list with the given {prefix, prefix length,
table ID}. Otherwise, it is not used in the data path and should not be
considered by switch drivers.

v2:
* Convert to use fib_find_alias() instead of fib_find_first_alias()

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/fib_trie.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index 4c80ac0344f4..f56945d00d7a 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -1295,6 +1295,16 @@ int fib_table_insert(struct net *net, struct fib_table *tb,
 	if (WARN_ON_ONCE(!l))
 		goto out_free_new_fa;
 
+	if (fib_find_alias(&l->leaf, new_fa->fa_slen, 0, 0, tb->tb_id, true) ==
+	    new_fa) {
+		enum fib_event_type fib_event;
+
+		fib_event = FIB_EVENT_ENTRY_REPLACE_TMP;
+		err = call_fib_entry_notifiers(net, fib_event, key, plen,
+					       new_fa, extack);
+		if (err)
+			goto out_remove_new_fa;
+	}
 	err = call_fib_entry_notifiers(net, event, key, plen, new_fa, extack);
 	if (err)
 		goto out_remove_new_fa;
-- 
cgit v1.2.3-59-g8ed1b


From f613b6e2ffe13e9acba0d330a058b87aa0e19a2a Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Sat, 14 Dec 2019 17:53:11 +0200
Subject: ipv4: Handle route deletion notification

When a route is deleted we potentially need to promote the next route in
the FIB alias list (e.g., with an higher metric). In case we find such a
route, a replace notification is emitted. Otherwise, a delete
notification for the deleted route.

v2:
* Convert to use fib_find_alias() instead of fib_find_first_alias()

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/fib_trie.c | 31 +++++++++++++++++++++++++++++++
 1 file changed, 31 insertions(+)

diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index f56945d00d7a..012aca433a91 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -1570,6 +1570,36 @@ static void fib_remove_alias(struct trie *t, struct key_vector *tp,
 	node_pull_suffix(tp, fa->fa_slen);
 }
 
+static void fib_notify_alias_delete(struct net *net, u32 key,
+				    struct hlist_head *fah,
+				    struct fib_alias *fa_to_delete,
+				    struct netlink_ext_ack *extack)
+{
+	struct fib_alias *fa_next, *fa_to_notify;
+	u32 tb_id = fa_to_delete->tb_id;
+	u8 slen = fa_to_delete->fa_slen;
+	enum fib_event_type fib_event;
+
+	/* Do not notify if we do not care about the route. */
+	if (fib_find_alias(fah, slen, 0, 0, tb_id, true) != fa_to_delete)
+		return;
+
+	/* Determine if the route should be replaced by the next route in the
+	 * list.
+	 */
+	fa_next = hlist_entry_safe(fa_to_delete->fa_list.next,
+				   struct fib_alias, fa_list);
+	if (fa_next && fa_next->fa_slen == slen && fa_next->tb_id == tb_id) {
+		fib_event = FIB_EVENT_ENTRY_REPLACE_TMP;
+		fa_to_notify = fa_next;
+	} else {
+		fib_event = FIB_EVENT_ENTRY_DEL_TMP;
+		fa_to_notify = fa_to_delete;
+	}
+	call_fib_entry_notifiers(net, fib_event, key, KEYLENGTH - slen,
+				 fa_to_notify, extack);
+}
+
 /* Caller must hold RTNL. */
 int fib_table_delete(struct net *net, struct fib_table *tb,
 		     struct fib_config *cfg, struct netlink_ext_ack *extack)
@@ -1623,6 +1653,7 @@ int fib_table_delete(struct net *net, struct fib_table *tb,
 	if (!fa_to_delete)
 		return -ESRCH;
 
+	fib_notify_alias_delete(net, key, &l->leaf, fa_to_delete, extack);
 	call_fib_entry_notifiers(net, FIB_EVENT_ENTRY_DEL, key, plen,
 				 fa_to_delete, extack);
 	rtmsg_fib(RTM_DELROUTE, htonl(key), fa_to_delete, plen, tb->tb_id,
-- 
cgit v1.2.3-59-g8ed1b


From 525bc345fcbc6f0d58977452999dd18a15dc0e9b Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Sat, 14 Dec 2019 17:53:12 +0200
Subject: ipv4: Handle route deletion notification during flush

In a similar fashion to previous patch, when a route is deleted as part
of table flushing, promote the next route in the list, if exists.
Otherwise, simply emit a delete notification.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/fib_trie.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index 012aca433a91..c23be49ca51c 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -1979,6 +1979,8 @@ int fib_table_flush(struct net *net, struct fib_table *tb, bool flush_all)
 				continue;
 			}
 
+			fib_notify_alias_delete(net, n->key, &n->leaf, fa,
+						NULL);
 			call_fib_entry_notifiers(net, FIB_EVENT_ENTRY_DEL,
 						 n->key,
 						 KEYLENGTH - fa->fa_slen, fa,
-- 
cgit v1.2.3-59-g8ed1b


From 20d1565203e68cd32d5db3e293e7323173ad7a0d Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Sat, 14 Dec 2019 17:53:13 +0200
Subject: ipv4: Only Replay routes of interest to new listeners

When a new listener is registered to the FIB notification chain it
receives a dump of all the available routes in the system. Instead, make
sure to only replay the IPv4 routes that are actually used in the data
path and are of any interest to the new listener.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/fib_trie.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index c23be49ca51c..3f2ff97618ec 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -2080,6 +2080,7 @@ static int fib_leaf_notify(struct key_vector *l, struct fib_table *tb,
 			   struct netlink_ext_ack *extack)
 {
 	struct fib_alias *fa;
+	int last_slen = -1;
 	int err;
 
 	hlist_for_each_entry_rcu(fa, &l->leaf, fa_list) {
@@ -2099,6 +2100,16 @@ static int fib_leaf_notify(struct key_vector *l, struct fib_table *tb,
 					      fa, extack);
 		if (err)
 			return err;
+
+		if (fa->fa_slen == last_slen)
+			continue;
+
+		last_slen = fa->fa_slen;
+		err = call_fib_entry_notifier(nb, FIB_EVENT_ENTRY_REPLACE_TMP,
+					      l->key, KEYLENGTH - fa->fa_slen,
+					      fa, extack);
+		if (err)
+			return err;
 	}
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From b6a1d871d37a8a1beffb5432631ca77ca3712c84 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Sat, 14 Dec 2019 17:53:14 +0200
Subject: mlxsw: spectrum_router: Start using new IPv4 route notifications

With the new notifications mlxsw does not need to handle identical
routes itself, as this is taken care of by the core IPv4 code.

Instead, mlxsw only needs to take care of inserting and removing routes
from the device.

Convert mlxsw to use the new IPv4 route notifications and simplify the
code.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../net/ethernet/mellanox/mlxsw/spectrum_router.c  | 141 +++------------------
 1 file changed, 20 insertions(+), 121 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
index 30bfe3880faf..396b27b9cdb4 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
@@ -3845,7 +3845,7 @@ static void mlxsw_sp_nexthop4_event(struct mlxsw_sp *mlxsw_sp,
 
 	key.fib_nh = fib_nh;
 	nh = mlxsw_sp_nexthop_lookup(mlxsw_sp, key);
-	if (WARN_ON_ONCE(!nh))
+	if (!nh)
 		return;
 
 	switch (event) {
@@ -4780,95 +4780,6 @@ static void mlxsw_sp_fib_node_put(struct mlxsw_sp *mlxsw_sp,
 	mlxsw_sp_vr_put(mlxsw_sp, vr);
 }
 
-static struct mlxsw_sp_fib4_entry *
-mlxsw_sp_fib4_node_entry_find(const struct mlxsw_sp_fib_node *fib_node,
-			      const struct mlxsw_sp_fib4_entry *new4_entry)
-{
-	struct mlxsw_sp_fib4_entry *fib4_entry;
-
-	list_for_each_entry(fib4_entry, &fib_node->entry_list, common.list) {
-		if (fib4_entry->tb_id > new4_entry->tb_id)
-			continue;
-		if (fib4_entry->tb_id != new4_entry->tb_id)
-			break;
-		if (fib4_entry->tos > new4_entry->tos)
-			continue;
-		if (fib4_entry->prio >= new4_entry->prio ||
-		    fib4_entry->tos < new4_entry->tos)
-			return fib4_entry;
-	}
-
-	return NULL;
-}
-
-static int
-mlxsw_sp_fib4_node_list_append(struct mlxsw_sp_fib4_entry *fib4_entry,
-			       struct mlxsw_sp_fib4_entry *new4_entry)
-{
-	struct mlxsw_sp_fib_node *fib_node;
-
-	if (WARN_ON(!fib4_entry))
-		return -EINVAL;
-
-	fib_node = fib4_entry->common.fib_node;
-	list_for_each_entry_from(fib4_entry, &fib_node->entry_list,
-				 common.list) {
-		if (fib4_entry->tb_id != new4_entry->tb_id ||
-		    fib4_entry->tos != new4_entry->tos ||
-		    fib4_entry->prio != new4_entry->prio)
-			break;
-	}
-
-	list_add_tail(&new4_entry->common.list, &fib4_entry->common.list);
-	return 0;
-}
-
-static int
-mlxsw_sp_fib4_node_list_insert(struct mlxsw_sp_fib4_entry *new4_entry,
-			       bool replace, bool append)
-{
-	struct mlxsw_sp_fib_node *fib_node = new4_entry->common.fib_node;
-	struct mlxsw_sp_fib4_entry *fib4_entry;
-
-	fib4_entry = mlxsw_sp_fib4_node_entry_find(fib_node, new4_entry);
-
-	if (append)
-		return mlxsw_sp_fib4_node_list_append(fib4_entry, new4_entry);
-	if (replace && WARN_ON(!fib4_entry))
-		return -EINVAL;
-
-	/* Insert new entry before replaced one, so that we can later
-	 * remove the second.
-	 */
-	if (fib4_entry) {
-		list_add_tail(&new4_entry->common.list,
-			      &fib4_entry->common.list);
-	} else {
-		struct mlxsw_sp_fib4_entry *last;
-
-		list_for_each_entry(last, &fib_node->entry_list, common.list) {
-			if (new4_entry->tb_id > last->tb_id)
-				break;
-			fib4_entry = last;
-		}
-
-		if (fib4_entry)
-			list_add(&new4_entry->common.list,
-				 &fib4_entry->common.list);
-		else
-			list_add(&new4_entry->common.list,
-				 &fib_node->entry_list);
-	}
-
-	return 0;
-}
-
-static void
-mlxsw_sp_fib4_node_list_remove(struct mlxsw_sp_fib4_entry *fib4_entry)
-{
-	list_del(&fib4_entry->common.list);
-}
-
 static int mlxsw_sp_fib_node_entry_add(struct mlxsw_sp *mlxsw_sp,
 				       struct mlxsw_sp_fib_entry *fib_entry)
 {
@@ -4912,14 +4823,12 @@ static void mlxsw_sp_fib_node_entry_del(struct mlxsw_sp *mlxsw_sp,
 }
 
 static int mlxsw_sp_fib4_node_entry_link(struct mlxsw_sp *mlxsw_sp,
-					 struct mlxsw_sp_fib4_entry *fib4_entry,
-					 bool replace, bool append)
+					 struct mlxsw_sp_fib4_entry *fib4_entry)
 {
+	struct mlxsw_sp_fib_node *fib_node = fib4_entry->common.fib_node;
 	int err;
 
-	err = mlxsw_sp_fib4_node_list_insert(fib4_entry, replace, append);
-	if (err)
-		return err;
+	list_add(&fib4_entry->common.list, &fib_node->entry_list);
 
 	err = mlxsw_sp_fib_node_entry_add(mlxsw_sp, &fib4_entry->common);
 	if (err)
@@ -4928,7 +4837,7 @@ static int mlxsw_sp_fib4_node_entry_link(struct mlxsw_sp *mlxsw_sp,
 	return 0;
 
 err_fib_node_entry_add:
-	mlxsw_sp_fib4_node_list_remove(fib4_entry);
+	list_del(&fib4_entry->common.list);
 	return err;
 }
 
@@ -4937,20 +4846,19 @@ mlxsw_sp_fib4_node_entry_unlink(struct mlxsw_sp *mlxsw_sp,
 				struct mlxsw_sp_fib4_entry *fib4_entry)
 {
 	mlxsw_sp_fib_node_entry_del(mlxsw_sp, &fib4_entry->common);
-	mlxsw_sp_fib4_node_list_remove(fib4_entry);
+	list_del(&fib4_entry->common.list);
 
 	if (fib4_entry->common.type == MLXSW_SP_FIB_ENTRY_TYPE_IPIP_DECAP)
 		mlxsw_sp_fib_entry_decap_fini(mlxsw_sp, &fib4_entry->common);
 }
 
 static void mlxsw_sp_fib4_entry_replace(struct mlxsw_sp *mlxsw_sp,
-					struct mlxsw_sp_fib4_entry *fib4_entry,
-					bool replace)
+					struct mlxsw_sp_fib4_entry *fib4_entry)
 {
 	struct mlxsw_sp_fib_node *fib_node = fib4_entry->common.fib_node;
 	struct mlxsw_sp_fib4_entry *replaced;
 
-	if (!replace)
+	if (list_is_singular(&fib_node->entry_list))
 		return;
 
 	/* We inserted the new entry before replaced one */
@@ -4962,9 +4870,8 @@ static void mlxsw_sp_fib4_entry_replace(struct mlxsw_sp *mlxsw_sp,
 }
 
 static int
-mlxsw_sp_router_fib4_add(struct mlxsw_sp *mlxsw_sp,
-			 const struct fib_entry_notifier_info *fen_info,
-			 bool replace, bool append)
+mlxsw_sp_router_fib4_replace(struct mlxsw_sp *mlxsw_sp,
+			     const struct fib_entry_notifier_info *fen_info)
 {
 	struct mlxsw_sp_fib4_entry *fib4_entry;
 	struct mlxsw_sp_fib_node *fib_node;
@@ -4989,14 +4896,13 @@ mlxsw_sp_router_fib4_add(struct mlxsw_sp *mlxsw_sp,
 		goto err_fib4_entry_create;
 	}
 
-	err = mlxsw_sp_fib4_node_entry_link(mlxsw_sp, fib4_entry, replace,
-					    append);
+	err = mlxsw_sp_fib4_node_entry_link(mlxsw_sp, fib4_entry);
 	if (err) {
 		dev_warn(mlxsw_sp->bus_info->dev, "Failed to link FIB entry to node\n");
 		goto err_fib4_node_entry_link;
 	}
 
-	mlxsw_sp_fib4_entry_replace(mlxsw_sp, fib4_entry, replace);
+	mlxsw_sp_fib4_entry_replace(mlxsw_sp, fib4_entry);
 
 	return 0;
 
@@ -6094,7 +6000,6 @@ static void mlxsw_sp_router_fib4_event_work(struct work_struct *work)
 	struct mlxsw_sp_fib_event_work *fib_work =
 		container_of(work, struct mlxsw_sp_fib_event_work, work);
 	struct mlxsw_sp *mlxsw_sp = fib_work->mlxsw_sp;
-	bool replace, append;
 	int err;
 
 	/* Protect internal structures from changes */
@@ -6102,18 +6007,14 @@ static void mlxsw_sp_router_fib4_event_work(struct work_struct *work)
 	mlxsw_sp_span_respin(mlxsw_sp);
 
 	switch (fib_work->event) {
-	case FIB_EVENT_ENTRY_REPLACE: /* fall through */
-	case FIB_EVENT_ENTRY_APPEND: /* fall through */
-	case FIB_EVENT_ENTRY_ADD:
-		replace = fib_work->event == FIB_EVENT_ENTRY_REPLACE;
-		append = fib_work->event == FIB_EVENT_ENTRY_APPEND;
-		err = mlxsw_sp_router_fib4_add(mlxsw_sp, &fib_work->fen_info,
-					       replace, append);
+	case FIB_EVENT_ENTRY_REPLACE_TMP:
+		err = mlxsw_sp_router_fib4_replace(mlxsw_sp,
+						   &fib_work->fen_info);
 		if (err)
 			mlxsw_sp_router_fib_abort(mlxsw_sp);
 		fib_info_put(fib_work->fen_info.fi);
 		break;
-	case FIB_EVENT_ENTRY_DEL:
+	case FIB_EVENT_ENTRY_DEL_TMP:
 		mlxsw_sp_router_fib4_del(mlxsw_sp, &fib_work->fen_info);
 		fib_info_put(fib_work->fen_info.fi);
 		break;
@@ -6210,10 +6111,8 @@ static void mlxsw_sp_router_fib4_event(struct mlxsw_sp_fib_event_work *fib_work,
 	struct fib_nh_notifier_info *fnh_info;
 
 	switch (fib_work->event) {
-	case FIB_EVENT_ENTRY_REPLACE: /* fall through */
-	case FIB_EVENT_ENTRY_APPEND: /* fall through */
-	case FIB_EVENT_ENTRY_ADD: /* fall through */
-	case FIB_EVENT_ENTRY_DEL:
+	case FIB_EVENT_ENTRY_REPLACE_TMP: /* fall through */
+	case FIB_EVENT_ENTRY_DEL_TMP:
 		fen_info = container_of(info, struct fib_entry_notifier_info,
 					info);
 		fib_work->fen_info = *fen_info;
@@ -6343,9 +6242,9 @@ static int mlxsw_sp_router_fib_event(struct notifier_block *nb,
 		err = mlxsw_sp_router_fib_rule_event(event, info,
 						     router->mlxsw_sp);
 		return notifier_from_errno(err);
-	case FIB_EVENT_ENTRY_ADD:
+	case FIB_EVENT_ENTRY_ADD: /* fall through */
 	case FIB_EVENT_ENTRY_REPLACE: /* fall through */
-	case FIB_EVENT_ENTRY_APPEND:  /* fall through */
+	case FIB_EVENT_ENTRY_REPLACE_TMP:
 		if (router->aborted) {
 			NL_SET_ERR_MSG_MOD(info->extack, "FIB offload was aborted. Not configuring route");
 			return notifier_from_errno(-EINVAL);
-- 
cgit v1.2.3-59-g8ed1b


From 446f739104f4da6207230363848ec2c89dfd858d Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Sat, 14 Dec 2019 17:53:15 +0200
Subject: ipv4: Remove old route notifications and convert listeners

Unlike mlxsw, the other listeners to the FIB notification chain do not
require any special modifications as they never considered multiple
identical routes.

This patch removes the old route notifications and converts all the
listeners to use the new replace / delete notifications.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c   |  4 ---
 .../net/ethernet/mellanox/mlxsw/spectrum_router.c  | 11 +++----
 drivers/net/ethernet/rocker/rocker_main.c          |  4 +--
 drivers/net/netdevsim/fib.c                        |  4 +--
 include/net/fib_notifier.h                         |  2 --
 net/ipv4/fib_trie.c                                | 38 ++++------------------
 6 files changed, 16 insertions(+), 47 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c b/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c
index b70afa310ad2..416676c35b1f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c
@@ -200,8 +200,6 @@ static void mlx5_lag_fib_update(struct work_struct *work)
 	rtnl_lock();
 	switch (fib_work->event) {
 	case FIB_EVENT_ENTRY_REPLACE: /* fall through */
-	case FIB_EVENT_ENTRY_APPEND: /* fall through */
-	case FIB_EVENT_ENTRY_ADD: /* fall through */
 	case FIB_EVENT_ENTRY_DEL:
 		mlx5_lag_fib_route_event(ldev, fib_work->event,
 					 fib_work->fen_info.fi);
@@ -259,8 +257,6 @@ static int mlx5_lag_fib_event(struct notifier_block *nb,
 
 	switch (event) {
 	case FIB_EVENT_ENTRY_REPLACE: /* fall through */
-	case FIB_EVENT_ENTRY_APPEND: /* fall through */
-	case FIB_EVENT_ENTRY_ADD: /* fall through */
 	case FIB_EVENT_ENTRY_DEL:
 		fen_info = container_of(info, struct fib_entry_notifier_info,
 					info);
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
index 396b27b9cdb4..bba1c8215d06 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
@@ -6007,14 +6007,14 @@ static void mlxsw_sp_router_fib4_event_work(struct work_struct *work)
 	mlxsw_sp_span_respin(mlxsw_sp);
 
 	switch (fib_work->event) {
-	case FIB_EVENT_ENTRY_REPLACE_TMP:
+	case FIB_EVENT_ENTRY_REPLACE:
 		err = mlxsw_sp_router_fib4_replace(mlxsw_sp,
 						   &fib_work->fen_info);
 		if (err)
 			mlxsw_sp_router_fib_abort(mlxsw_sp);
 		fib_info_put(fib_work->fen_info.fi);
 		break;
-	case FIB_EVENT_ENTRY_DEL_TMP:
+	case FIB_EVENT_ENTRY_DEL:
 		mlxsw_sp_router_fib4_del(mlxsw_sp, &fib_work->fen_info);
 		fib_info_put(fib_work->fen_info.fi);
 		break;
@@ -6111,8 +6111,8 @@ static void mlxsw_sp_router_fib4_event(struct mlxsw_sp_fib_event_work *fib_work,
 	struct fib_nh_notifier_info *fnh_info;
 
 	switch (fib_work->event) {
-	case FIB_EVENT_ENTRY_REPLACE_TMP: /* fall through */
-	case FIB_EVENT_ENTRY_DEL_TMP:
+	case FIB_EVENT_ENTRY_REPLACE: /* fall through */
+	case FIB_EVENT_ENTRY_DEL:
 		fen_info = container_of(info, struct fib_entry_notifier_info,
 					info);
 		fib_work->fen_info = *fen_info;
@@ -6243,8 +6243,7 @@ static int mlxsw_sp_router_fib_event(struct notifier_block *nb,
 						     router->mlxsw_sp);
 		return notifier_from_errno(err);
 	case FIB_EVENT_ENTRY_ADD: /* fall through */
-	case FIB_EVENT_ENTRY_REPLACE: /* fall through */
-	case FIB_EVENT_ENTRY_REPLACE_TMP:
+	case FIB_EVENT_ENTRY_REPLACE:
 		if (router->aborted) {
 			NL_SET_ERR_MSG_MOD(info->extack, "FIB offload was aborted. Not configuring route");
 			return notifier_from_errno(-EINVAL);
diff --git a/drivers/net/ethernet/rocker/rocker_main.c b/drivers/net/ethernet/rocker/rocker_main.c
index bc4f951315da..7585cd2270ba 100644
--- a/drivers/net/ethernet/rocker/rocker_main.c
+++ b/drivers/net/ethernet/rocker/rocker_main.c
@@ -2159,7 +2159,7 @@ static void rocker_router_fib_event_work(struct work_struct *work)
 	/* Protect internal structures from changes */
 	rtnl_lock();
 	switch (fib_work->event) {
-	case FIB_EVENT_ENTRY_ADD:
+	case FIB_EVENT_ENTRY_REPLACE:
 		err = rocker_world_fib4_add(rocker, &fib_work->fen_info);
 		if (err)
 			rocker_world_fib4_abort(rocker);
@@ -2201,7 +2201,7 @@ static int rocker_router_fib_event(struct notifier_block *nb,
 	fib_work->event = event;
 
 	switch (event) {
-	case FIB_EVENT_ENTRY_ADD: /* fall through */
+	case FIB_EVENT_ENTRY_REPLACE: /* fall through */
 	case FIB_EVENT_ENTRY_DEL:
 		if (info->family == AF_INET) {
 			struct fib_entry_notifier_info *fen_info = ptr;
diff --git a/drivers/net/netdevsim/fib.c b/drivers/net/netdevsim/fib.c
index 13540dee7364..4e02a4231fcb 100644
--- a/drivers/net/netdevsim/fib.c
+++ b/drivers/net/netdevsim/fib.c
@@ -177,10 +177,10 @@ static int nsim_fib_event_nb(struct notifier_block *nb, unsigned long event,
 					  event == FIB_EVENT_RULE_ADD);
 		break;
 
+	case FIB_EVENT_ENTRY_REPLACE:  /* fall through */
 	case FIB_EVENT_ENTRY_ADD:  /* fall through */
 	case FIB_EVENT_ENTRY_DEL:
-		err = nsim_fib_event(data, info,
-				     event == FIB_EVENT_ENTRY_ADD);
+		err = nsim_fib_event(data, info, event != FIB_EVENT_ENTRY_DEL);
 		break;
 	}
 
diff --git a/include/net/fib_notifier.h b/include/net/fib_notifier.h
index b3c54325caec..6d59221ff05a 100644
--- a/include/net/fib_notifier.h
+++ b/include/net/fib_notifier.h
@@ -23,8 +23,6 @@ enum fib_event_type {
 	FIB_EVENT_NH_DEL,
 	FIB_EVENT_VIF_ADD,
 	FIB_EVENT_VIF_DEL,
-	FIB_EVENT_ENTRY_REPLACE_TMP,
-	FIB_EVENT_ENTRY_DEL_TMP,
 };
 
 struct fib_notifier_ops {
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index 3f2ff97618ec..b92a42433a7d 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -1127,7 +1127,6 @@ static void fib_remove_alias(struct trie *t, struct key_vector *tp,
 int fib_table_insert(struct net *net, struct fib_table *tb,
 		     struct fib_config *cfg, struct netlink_ext_ack *extack)
 {
-	enum fib_event_type event = FIB_EVENT_ENTRY_ADD;
 	struct trie *t = (struct trie *)tb->tb_data;
 	struct fib_alias *fa, *new_fa;
 	struct key_vector *l, *tp;
@@ -1226,19 +1225,13 @@ int fib_table_insert(struct net *net, struct fib_table *tb,
 					   tb->tb_id, true) == fa) {
 				enum fib_event_type fib_event;
 
-				fib_event = FIB_EVENT_ENTRY_REPLACE_TMP;
+				fib_event = FIB_EVENT_ENTRY_REPLACE;
 				err = call_fib_entry_notifiers(net, fib_event,
 							       key, plen,
 							       new_fa, extack);
 				if (err)
 					goto out_free_new_fa;
 			}
-			err = call_fib_entry_notifiers(net,
-						       FIB_EVENT_ENTRY_REPLACE,
-						       key, plen, new_fa,
-						       extack);
-			if (err)
-				goto out_free_new_fa;
 
 			rtmsg_fib(RTM_NEWROUTE, htonl(key), new_fa, plen,
 				  tb->tb_id, &cfg->fc_nlinfo, nlflags);
@@ -1260,12 +1253,10 @@ int fib_table_insert(struct net *net, struct fib_table *tb,
 		if (fa_match)
 			goto out;
 
-		if (cfg->fc_nlflags & NLM_F_APPEND) {
-			event = FIB_EVENT_ENTRY_APPEND;
+		if (cfg->fc_nlflags & NLM_F_APPEND)
 			nlflags |= NLM_F_APPEND;
-		} else {
+		else
 			fa = fa_first;
-		}
 	}
 	err = -ENOENT;
 	if (!(cfg->fc_nlflags & NLM_F_CREATE))
@@ -1299,15 +1290,12 @@ int fib_table_insert(struct net *net, struct fib_table *tb,
 	    new_fa) {
 		enum fib_event_type fib_event;
 
-		fib_event = FIB_EVENT_ENTRY_REPLACE_TMP;
+		fib_event = FIB_EVENT_ENTRY_REPLACE;
 		err = call_fib_entry_notifiers(net, fib_event, key, plen,
 					       new_fa, extack);
 		if (err)
 			goto out_remove_new_fa;
 	}
-	err = call_fib_entry_notifiers(net, event, key, plen, new_fa, extack);
-	if (err)
-		goto out_remove_new_fa;
 
 	if (!plen)
 		tb->tb_num_default++;
@@ -1590,10 +1578,10 @@ static void fib_notify_alias_delete(struct net *net, u32 key,
 	fa_next = hlist_entry_safe(fa_to_delete->fa_list.next,
 				   struct fib_alias, fa_list);
 	if (fa_next && fa_next->fa_slen == slen && fa_next->tb_id == tb_id) {
-		fib_event = FIB_EVENT_ENTRY_REPLACE_TMP;
+		fib_event = FIB_EVENT_ENTRY_REPLACE;
 		fa_to_notify = fa_next;
 	} else {
-		fib_event = FIB_EVENT_ENTRY_DEL_TMP;
+		fib_event = FIB_EVENT_ENTRY_DEL;
 		fa_to_notify = fa_to_delete;
 	}
 	call_fib_entry_notifiers(net, fib_event, key, KEYLENGTH - slen,
@@ -1654,8 +1642,6 @@ int fib_table_delete(struct net *net, struct fib_table *tb,
 		return -ESRCH;
 
 	fib_notify_alias_delete(net, key, &l->leaf, fa_to_delete, extack);
-	call_fib_entry_notifiers(net, FIB_EVENT_ENTRY_DEL, key, plen,
-				 fa_to_delete, extack);
 	rtmsg_fib(RTM_DELROUTE, htonl(key), fa_to_delete, plen, tb->tb_id,
 		  &cfg->fc_nlinfo, 0);
 
@@ -1981,10 +1967,6 @@ int fib_table_flush(struct net *net, struct fib_table *tb, bool flush_all)
 
 			fib_notify_alias_delete(net, n->key, &n->leaf, fa,
 						NULL);
-			call_fib_entry_notifiers(net, FIB_EVENT_ENTRY_DEL,
-						 n->key,
-						 KEYLENGTH - fa->fa_slen, fa,
-						 NULL);
 			hlist_del_rcu(&fa->fa_list);
 			fib_release_info(fa->fa_info);
 			alias_free_mem_rcu(fa);
@@ -2095,17 +2077,11 @@ static int fib_leaf_notify(struct key_vector *l, struct fib_table *tb,
 		if (tb->tb_id != fa->tb_id)
 			continue;
 
-		err = call_fib_entry_notifier(nb, FIB_EVENT_ENTRY_ADD, l->key,
-					      KEYLENGTH - fa->fa_slen,
-					      fa, extack);
-		if (err)
-			return err;
-
 		if (fa->fa_slen == last_slen)
 			continue;
 
 		last_slen = fa->fa_slen;
-		err = call_fib_entry_notifier(nb, FIB_EVENT_ENTRY_REPLACE_TMP,
+		err = call_fib_entry_notifier(nb, FIB_EVENT_ENTRY_REPLACE,
 					      l->key, KEYLENGTH - fa->fa_slen,
 					      fa, extack);
 		if (err)
-- 
cgit v1.2.3-59-g8ed1b


From bbd20c939c8aa3f27fa30e86691af250bf92973a Mon Sep 17 00:00:00 2001
From: Aditya Pakki <pakki001@umn.edu>
Date: Sun, 15 Dec 2019 10:14:51 -0600
Subject: fore200e: Fix incorrect checks of NULL pointer dereference

In fore200e_send and fore200e_close, the pointers from the arguments
are dereferenced in the variable declaration block and then checked
for NULL. The patch fixes these issues by avoiding NULL pointer
dereferences.

Signed-off-by: Aditya Pakki <pakki001@umn.edu>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/atm/fore200e.c | 25 ++++++++++++++++++-------
 1 file changed, 18 insertions(+), 7 deletions(-)

diff --git a/drivers/atm/fore200e.c b/drivers/atm/fore200e.c
index f1a500205313..8fbd36eb8941 100644
--- a/drivers/atm/fore200e.c
+++ b/drivers/atm/fore200e.c
@@ -1414,12 +1414,14 @@ fore200e_open(struct atm_vcc *vcc)
 static void
 fore200e_close(struct atm_vcc* vcc)
 {
-    struct fore200e*        fore200e = FORE200E_DEV(vcc->dev);
     struct fore200e_vcc*    fore200e_vcc;
+    struct fore200e*        fore200e;
     struct fore200e_vc_map* vc_map;
     unsigned long           flags;
 
     ASSERT(vcc);
+    fore200e = FORE200E_DEV(vcc->dev);
+
     ASSERT((vcc->vpi >= 0) && (vcc->vpi < 1<<FORE200E_VPI_BITS));
     ASSERT((vcc->vci >= 0) && (vcc->vci < 1<<FORE200E_VCI_BITS));
 
@@ -1464,10 +1466,10 @@ fore200e_close(struct atm_vcc* vcc)
 static int
 fore200e_send(struct atm_vcc *vcc, struct sk_buff *skb)
 {
-    struct fore200e*        fore200e     = FORE200E_DEV(vcc->dev);
-    struct fore200e_vcc*    fore200e_vcc = FORE200E_VCC(vcc);
+    struct fore200e*        fore200e;
+    struct fore200e_vcc*    fore200e_vcc;
     struct fore200e_vc_map* vc_map;
-    struct host_txq*        txq          = &fore200e->host_txq;
+    struct host_txq*        txq;
     struct host_txq_entry*  entry;
     struct tpd*             tpd;
     struct tpd_haddr        tpd_haddr;
@@ -1480,9 +1482,18 @@ fore200e_send(struct atm_vcc *vcc, struct sk_buff *skb)
     unsigned char*          data;
     unsigned long           flags;
 
-    ASSERT(vcc);
-    ASSERT(fore200e);
-    ASSERT(fore200e_vcc);
+    if (!vcc)
+        return -EINVAL;
+
+    fore200e = FORE200E_DEV(vcc->dev);
+    fore200e_vcc = FORE200E_VCC(vcc);
+
+    if (!fore200e)
+        return -EINVAL;
+
+    txq = &fore200e->host_txq;
+    if (!fore200e_vcc)
+        return -EINVAL;
 
     if (!test_bit(ATM_VF_READY, &vcc->flags)) {
 	DPRINTK(1, "VC %d.%d.%d not ready for tx\n", vcc->itf, vcc->vpi, vcc->vpi);
-- 
cgit v1.2.3-59-g8ed1b


From c5dea815834c7d2e9fc633785455bc428b7a1956 Mon Sep 17 00:00:00 2001
From: Aditya Pakki <pakki001@umn.edu>
Date: Sun, 15 Dec 2019 11:51:30 -0600
Subject: net: caif: replace BUG_ON with recovery code

In caif_xmit, there is a crash if the ptr dev is NULL. However, by
returning the error to the callers, the error can be handled. The
patch fixes this issue.

Signed-off-by: Aditya Pakki <pakki001@umn.edu>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/caif/caif_serial.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/caif/caif_serial.c b/drivers/net/caif/caif_serial.c
index bd40b114d6cd..d737ceb61203 100644
--- a/drivers/net/caif/caif_serial.c
+++ b/drivers/net/caif/caif_serial.c
@@ -270,7 +270,9 @@ static int caif_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct ser_device *ser;
 
-	BUG_ON(dev == NULL);
+	if (WARN_ON(!dev))
+		return -EINVAL;
+
 	ser = netdev_priv(dev);
 
 	/* Send flow off once, on high water mark */
-- 
cgit v1.2.3-59-g8ed1b


From 65d88d04114bca7d85faebd5fed61069cb2b632c Mon Sep 17 00:00:00 2001
From: "Jason A. Donenfeld" <Jason@zx2c4.com>
Date: Sun, 15 Dec 2019 22:08:00 +0100
Subject: wireguard: selftests: import harness makefile for test suite

WireGuard has been using this on build.wireguard.com for the last
several years with considerable success. It allows for very quick and
iterative development cycles, and supports several platforms.

To run the test suite on your current platform in QEMU:

  $ make -C tools/testing/selftests/wireguard/qemu -j$(nproc)

To run it with KASAN and such turned on:

  $ DEBUG_KERNEL=yes make -C tools/testing/selftests/wireguard/qemu -j$(nproc)

To run it emulated for another platform in QEMU:

  $ ARCH=arm make -C tools/testing/selftests/wireguard/qemu -j$(nproc)

At the moment, we support aarch64_be, aarch64, arm, armeb, i686, m68k,
mips64, mips64el, mips, mipsel, powerpc64le, powerpc, and x86_64.

The system supports incremental rebuilding, so it should be very fast to
change a single file and then test it out and have immediate feedback.

This requires for the right toolchain and qemu to be installed prior.
I've had success with those from musl.cc.

This is tailored for WireGuard at the moment, though later projects
might generalize it for other network testing.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/wireguard/qemu/.gitignore  |   2 +
 tools/testing/selftests/wireguard/qemu/Makefile    | 385 +++++++++++++++++++++
 .../selftests/wireguard/qemu/arch/aarch64.config   |   5 +
 .../wireguard/qemu/arch/aarch64_be.config          |   6 +
 .../selftests/wireguard/qemu/arch/arm.config       |   9 +
 .../selftests/wireguard/qemu/arch/armeb.config     |  10 +
 .../selftests/wireguard/qemu/arch/i686.config      |   5 +
 .../selftests/wireguard/qemu/arch/m68k.config      |   9 +
 .../selftests/wireguard/qemu/arch/mips.config      |  11 +
 .../selftests/wireguard/qemu/arch/mips64.config    |  14 +
 .../selftests/wireguard/qemu/arch/mips64el.config  |  15 +
 .../selftests/wireguard/qemu/arch/mipsel.config    |  12 +
 .../selftests/wireguard/qemu/arch/powerpc.config   |  10 +
 .../wireguard/qemu/arch/powerpc64le.config         |  12 +
 .../selftests/wireguard/qemu/arch/x86_64.config    |   5 +
 .../testing/selftests/wireguard/qemu/debug.config  |  67 ++++
 tools/testing/selftests/wireguard/qemu/init.c      | 284 +++++++++++++++
 .../testing/selftests/wireguard/qemu/kernel.config |  86 +++++
 18 files changed, 947 insertions(+)
 create mode 100644 tools/testing/selftests/wireguard/qemu/.gitignore
 create mode 100644 tools/testing/selftests/wireguard/qemu/Makefile
 create mode 100644 tools/testing/selftests/wireguard/qemu/arch/aarch64.config
 create mode 100644 tools/testing/selftests/wireguard/qemu/arch/aarch64_be.config
 create mode 100644 tools/testing/selftests/wireguard/qemu/arch/arm.config
 create mode 100644 tools/testing/selftests/wireguard/qemu/arch/armeb.config
 create mode 100644 tools/testing/selftests/wireguard/qemu/arch/i686.config
 create mode 100644 tools/testing/selftests/wireguard/qemu/arch/m68k.config
 create mode 100644 tools/testing/selftests/wireguard/qemu/arch/mips.config
 create mode 100644 tools/testing/selftests/wireguard/qemu/arch/mips64.config
 create mode 100644 tools/testing/selftests/wireguard/qemu/arch/mips64el.config
 create mode 100644 tools/testing/selftests/wireguard/qemu/arch/mipsel.config
 create mode 100644 tools/testing/selftests/wireguard/qemu/arch/powerpc.config
 create mode 100644 tools/testing/selftests/wireguard/qemu/arch/powerpc64le.config
 create mode 100644 tools/testing/selftests/wireguard/qemu/arch/x86_64.config
 create mode 100644 tools/testing/selftests/wireguard/qemu/debug.config
 create mode 100644 tools/testing/selftests/wireguard/qemu/init.c
 create mode 100644 tools/testing/selftests/wireguard/qemu/kernel.config

diff --git a/tools/testing/selftests/wireguard/qemu/.gitignore b/tools/testing/selftests/wireguard/qemu/.gitignore
new file mode 100644
index 000000000000..415b542a9d59
--- /dev/null
+++ b/tools/testing/selftests/wireguard/qemu/.gitignore
@@ -0,0 +1,2 @@
+build/
+distfiles/
diff --git a/tools/testing/selftests/wireguard/qemu/Makefile b/tools/testing/selftests/wireguard/qemu/Makefile
new file mode 100644
index 000000000000..6d51bf78eeff
--- /dev/null
+++ b/tools/testing/selftests/wireguard/qemu/Makefile
@@ -0,0 +1,385 @@
+# SPDX-License-Identifier: GPL-2.0
+#
+# Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+
+PWD := $(shell pwd)
+
+CHOST := $(shell gcc -dumpmachine)
+ifneq (,$(ARCH))
+CBUILD := $(subst -gcc,,$(lastword $(subst /, ,$(firstword $(wildcard $(foreach bindir,$(subst :, ,$(PATH)),$(bindir)/$(ARCH)-*-gcc))))))
+ifeq (,$(CBUILD))
+$(error The toolchain for $(ARCH) is not installed)
+endif
+else
+CBUILD := $(CHOST)
+ARCH := $(firstword $(subst -, ,$(CBUILD)))
+endif
+
+# Set these from the environment to override
+KERNEL_PATH ?= $(PWD)/../../../../..
+BUILD_PATH ?= $(PWD)/build/$(ARCH)
+DISTFILES_PATH ?= $(PWD)/distfiles
+NR_CPUS ?= 4
+
+MIRROR := https://download.wireguard.com/qemu-test/distfiles/
+
+default: qemu
+
+# variable name, tarball project name, version, tarball extension, default URI base
+define tar_download =
+$(1)_VERSION := $(3)
+$(1)_NAME := $(2)-$$($(1)_VERSION)
+$(1)_TAR := $(DISTFILES_PATH)/$$($(1)_NAME)$(4)
+$(1)_PATH := $(BUILD_PATH)/$$($(1)_NAME)
+$(call file_download,$$($(1)_NAME)$(4),$(5),$(6))
+endef
+
+define file_download =
+$(DISTFILES_PATH)/$(1):
+	mkdir -p $(DISTFILES_PATH)
+	flock -x $$@.lock -c '[ -f $$@ ] && exit 0; wget -O $$@.tmp $(MIRROR)$(1) || wget -t inf --retry-on-http-error=404 -O $$@.tmp $(2)$(1) || rm -f $$@.tmp'
+	if echo "$(3)  $$@.tmp" | sha256sum -c -; then mv $$@.tmp $$@; else rm -f $$@.tmp; exit 71; fi
+endef
+
+$(eval $(call tar_download,MUSL,musl,1.1.20,.tar.gz,https://www.musl-libc.org/releases/,44be8771d0e6c6b5f82dd15662eb2957c9a3173a19a8b49966ac0542bbd40d61))
+$(eval $(call tar_download,LIBMNL,libmnl,1.0.4,.tar.bz2,https://www.netfilter.org/projects/libmnl/files/,171f89699f286a5854b72b91d06e8f8e3683064c5901fb09d954a9ab6f551f81))
+$(eval $(call tar_download,IPERF,iperf,3.1.7,.tar.gz,http://downloads.es.net/pub/iperf/,a4ef73406fe92250602b8da2ae89ec53211f805df97a1d1d629db5a14043734f))
+$(eval $(call tar_download,BASH,bash,5.0,.tar.gz,https://ftp.gnu.org/gnu/bash/,b4a80f2ac66170b2913efbfb9f2594f1f76c7b1afd11f799e22035d63077fb4d))
+$(eval $(call tar_download,IPROUTE2,iproute2,5.1.0,.tar.gz,https://www.kernel.org/pub/linux/utils/net/iproute2/,9b43707d6075ecdca14803ca8ce0c8553848c49fa1586d12fd508d66577243f2))
+$(eval $(call tar_download,IPTABLES,iptables,1.6.1,.tar.bz2,https://www.netfilter.org/projects/iptables/files/,0fc2d7bd5d7be11311726466789d4c65fb4c8e096c9182b56ce97440864f0cf5))
+$(eval $(call tar_download,NMAP,nmap,7.60,.tar.bz2,https://nmap.org/dist/,a8796ecc4fa6c38aad6139d9515dc8113023a82e9d787e5a5fb5fa1b05516f21))
+$(eval $(call tar_download,IPUTILS,iputils,s20161105,.tar.gz,https://github.com/iputils/iputils/archive/s20161105.tar.gz/#,f813092f03d17294fd23544b129b95cdb87fe19f7970a51908a6b88509acad8a))
+$(eval $(call tar_download,WIREGUARD_TOOLS,WireGuard,0.0.20191212,.tar.xz,https://git.zx2c4.com/WireGuard/snapshot/,b0d718380f7a8822b2f12d75e462fa4eafa3a77871002981f367cd4fe2a1b071))
+
+KERNEL_BUILD_PATH := $(BUILD_PATH)/kernel$(if $(findstring yes,$(DEBUG_KERNEL)),-debug)
+rwildcard=$(foreach d,$(wildcard $1*),$(call rwildcard,$d/,$2) $(filter $(subst *,%,$2),$d))
+WIREGUARD_SOURCES := $(call rwildcard,$(KERNEL_PATH)/drivers/net/wireguard/,*)
+
+export CFLAGS ?= -O3 -pipe
+export LDFLAGS ?=
+export CPPFLAGS := -I$(BUILD_PATH)/include
+
+ifeq ($(CHOST),$(CBUILD))
+CROSS_COMPILE_FLAG := --host=$(CHOST)
+NOPIE_GCC := gcc -fno-PIE
+CFLAGS += -march=native
+STRIP := strip
+else
+$(info Cross compilation: building for $(CBUILD) using $(CHOST))
+CROSS_COMPILE_FLAG := --build=$(CBUILD) --host=$(CHOST)
+export CROSS_COMPILE=$(CBUILD)-
+NOPIE_GCC := $(CBUILD)-gcc -fno-PIE
+STRIP := $(CBUILD)-strip
+endif
+ifeq ($(ARCH),aarch64)
+QEMU_ARCH := aarch64
+KERNEL_ARCH := arm64
+KERNEL_BZIMAGE := $(KERNEL_BUILD_PATH)/arch/arm64/boot/Image
+ifeq ($(CHOST),$(CBUILD))
+QEMU_MACHINE := -cpu host -machine virt,gic_version=host,accel=kvm
+else
+QEMU_MACHINE := -cpu cortex-a53 -machine virt
+CFLAGS += -march=armv8-a -mtune=cortex-a53
+endif
+else ifeq ($(ARCH),aarch64_be)
+QEMU_ARCH := aarch64
+KERNEL_ARCH := arm64
+KERNEL_BZIMAGE := $(KERNEL_BUILD_PATH)/arch/arm64/boot/Image
+ifeq ($(CHOST),$(CBUILD))
+QEMU_MACHINE := -cpu host -machine virt,gic_version=host,accel=kvm
+else
+QEMU_MACHINE := -cpu cortex-a53 -machine virt
+CFLAGS += -march=armv8-a -mtune=cortex-a53
+endif
+else ifeq ($(ARCH),arm)
+QEMU_ARCH := arm
+KERNEL_ARCH := arm
+KERNEL_BZIMAGE := $(KERNEL_BUILD_PATH)/arch/arm/boot/zImage
+ifeq ($(CHOST),$(CBUILD))
+QEMU_MACHINE := -cpu host -machine virt,gic_version=host,accel=kvm
+else
+QEMU_MACHINE := -cpu cortex-a15 -machine virt
+CFLAGS += -march=armv7-a -mtune=cortex-a15 -mabi=aapcs-linux
+endif
+else ifeq ($(ARCH),armeb)
+QEMU_ARCH := arm
+KERNEL_ARCH := arm
+KERNEL_BZIMAGE := $(KERNEL_BUILD_PATH)/arch/arm/boot/zImage
+ifeq ($(CHOST),$(CBUILD))
+QEMU_MACHINE := -cpu host -machine virt,gic_version=host,accel=kvm
+else
+QEMU_MACHINE := -cpu cortex-a15 -machine virt
+CFLAGS += -march=armv7-a -mabi=aapcs-linux # We don't pass -mtune=cortex-a15 due to a compiler bug on big endian.
+LDFLAGS += -Wl,--be8
+endif
+else ifeq ($(ARCH),x86_64)
+QEMU_ARCH := x86_64
+KERNEL_ARCH := x86_64
+KERNEL_BZIMAGE := $(KERNEL_BUILD_PATH)/arch/x86/boot/bzImage
+ifeq ($(CHOST),$(CBUILD))
+QEMU_MACHINE := -cpu host -machine q35,accel=kvm
+else
+QEMU_MACHINE := -cpu Skylake-Server -machine q35
+CFLAGS += -march=skylake-avx512
+endif
+else ifeq ($(ARCH),i686)
+QEMU_ARCH := i386
+KERNEL_ARCH := x86
+KERNEL_BZIMAGE := $(KERNEL_BUILD_PATH)/arch/x86/boot/bzImage
+ifeq ($(subst i686,x86_64,$(CBUILD)),$(CHOST))
+QEMU_MACHINE := -cpu host -machine q35,accel=kvm
+else
+QEMU_MACHINE := -cpu coreduo -machine q35
+CFLAGS += -march=prescott
+endif
+else ifeq ($(ARCH),mips64)
+QEMU_ARCH := mips64
+KERNEL_ARCH := mips
+KERNEL_BZIMAGE := $(KERNEL_BUILD_PATH)/vmlinux
+ifeq ($(CHOST),$(CBUILD))
+QEMU_MACHINE := -cpu host -machine malta,accel=kvm
+CFLAGS += -EB
+else
+QEMU_MACHINE := -cpu MIPS64R2-generic -machine malta -smp 1
+CFLAGS += -march=mips64r2 -EB
+endif
+else ifeq ($(ARCH),mips64el)
+QEMU_ARCH := mips64el
+KERNEL_ARCH := mips
+KERNEL_BZIMAGE := $(KERNEL_BUILD_PATH)/vmlinux
+ifeq ($(CHOST),$(CBUILD))
+QEMU_MACHINE := -cpu host -machine malta,accel=kvm
+CFLAGS += -EL
+else
+QEMU_MACHINE := -cpu MIPS64R2-generic -machine malta -smp 1
+CFLAGS += -march=mips64r2 -EL
+endif
+else ifeq ($(ARCH),mips)
+QEMU_ARCH := mips
+KERNEL_ARCH := mips
+KERNEL_BZIMAGE := $(KERNEL_BUILD_PATH)/vmlinux
+ifeq ($(CHOST),$(CBUILD))
+QEMU_MACHINE := -cpu host -machine malta,accel=kvm
+CFLAGS += -EB
+else
+QEMU_MACHINE := -cpu 24Kf -machine malta -smp 1
+CFLAGS += -march=mips32r2 -EB
+endif
+else ifeq ($(ARCH),mipsel)
+QEMU_ARCH := mipsel
+KERNEL_ARCH := mips
+KERNEL_BZIMAGE := $(KERNEL_BUILD_PATH)/vmlinux
+ifeq ($(CHOST),$(CBUILD))
+QEMU_MACHINE := -cpu host -machine malta,accel=kvm
+CFLAGS += -EL
+else
+QEMU_MACHINE := -cpu 24Kf -machine malta -smp 1
+CFLAGS += -march=mips32r2 -EL
+endif
+else ifeq ($(ARCH),powerpc64le)
+QEMU_ARCH := ppc64
+KERNEL_ARCH := powerpc
+KERNEL_BZIMAGE := $(KERNEL_BUILD_PATH)/vmlinux
+ifeq ($(CHOST),$(CBUILD))
+QEMU_MACHINE := -cpu host,accel=kvm -machine pseries
+else
+QEMU_MACHINE := -machine pseries
+endif
+CFLAGS += -mcpu=powerpc64le -mlong-double-64
+else ifeq ($(ARCH),powerpc)
+QEMU_ARCH := ppc
+KERNEL_ARCH := powerpc
+KERNEL_BZIMAGE := $(KERNEL_BUILD_PATH)/arch/powerpc/boot/uImage
+ifeq ($(CHOST),$(CBUILD))
+QEMU_MACHINE := -cpu host,accel=kvm -machine ppce500
+else
+QEMU_MACHINE := -machine ppce500
+endif
+CFLAGS += -mcpu=powerpc -mlong-double-64 -msecure-plt
+else ifeq ($(ARCH),m68k)
+QEMU_ARCH := m68k
+KERNEL_ARCH := m68k
+KERNEL_BZIMAGE := $(KERNEL_BUILD_PATH)/vmlinux
+ifeq ($(CHOST),$(CBUILD))
+QEMU_MACHINE := -cpu host,accel=kvm -machine q800
+else
+QEMU_MACHINE := -machine q800
+endif
+else
+$(error I only build: x86_64, i686, arm, armeb, aarch64, aarch64_be, mips, mipsel, mips64, mips64el, powerpc64le, powerpc, m68k)
+endif
+
+REAL_CC := $(CBUILD)-gcc
+MUSL_CC := $(BUILD_PATH)/musl-gcc
+export CC := $(MUSL_CC)
+USERSPACE_DEPS := $(MUSL_CC) $(BUILD_PATH)/include/.installed $(BUILD_PATH)/include/linux/.installed
+
+build: $(KERNEL_BZIMAGE)
+qemu: $(KERNEL_BZIMAGE)
+	rm -f $(BUILD_PATH)/result
+	timeout --foreground 20m qemu-system-$(QEMU_ARCH) \
+		-nodefaults \
+		-nographic \
+		-smp $(NR_CPUS) \
+		$(QEMU_MACHINE) \
+		-m $$(grep -q CONFIG_DEBUG_KMEMLEAK=y $(KERNEL_BUILD_PATH)/.config && echo 1G || echo 256M) \
+		-serial stdio \
+		-serial file:$(BUILD_PATH)/result \
+		-no-reboot \
+		-monitor none \
+		-kernel $<
+	grep -Fq success $(BUILD_PATH)/result
+
+$(BUILD_PATH)/init-cpio-spec.txt:
+	mkdir -p $(BUILD_PATH)
+	echo "file /init $(BUILD_PATH)/init 755 0 0" > $@
+	echo "file /init.sh $(PWD)/../netns.sh 755 0 0" >> $@
+	echo "dir /dev 755 0 0" >> $@
+	echo "nod /dev/console 644 0 0 c 5 1" >> $@
+	echo "dir /bin 755 0 0" >> $@
+	echo "file /bin/iperf3 $(IPERF_PATH)/src/iperf3 755 0 0" >> $@
+	echo "file /bin/wg $(WIREGUARD_TOOLS_PATH)/src/tools/wg 755 0 0" >> $@
+	echo "file /bin/bash $(BASH_PATH)/bash 755 0 0" >> $@
+	echo "file /bin/ip $(IPROUTE2_PATH)/ip/ip 755 0 0" >> $@
+	echo "file /bin/ss $(IPROUTE2_PATH)/misc/ss 755 0 0" >> $@
+	echo "file /bin/ping $(IPUTILS_PATH)/ping 755 0 0" >> $@
+	echo "file /bin/ncat $(NMAP_PATH)/ncat/ncat 755 0 0" >> $@
+	echo "file /bin/xtables-multi $(IPTABLES_PATH)/iptables/xtables-multi 755 0 0" >> $@
+	echo "slink /bin/iptables xtables-multi 777 0 0" >> $@
+	echo "slink /bin/ping6 ping 777 0 0" >> $@
+	echo "dir /lib 755 0 0" >> $@
+	echo "file /lib/libc.so $(MUSL_PATH)/lib/libc.so 755 0 0" >> $@
+	echo "slink /lib/ld-linux.so.1 libc.so 777 0 0" >> $@
+
+$(KERNEL_BUILD_PATH)/.config: kernel.config arch/$(ARCH).config
+	mkdir -p $(KERNEL_BUILD_PATH)
+	cp kernel.config $(KERNEL_BUILD_PATH)/minimal.config
+	printf 'CONFIG_NR_CPUS=$(NR_CPUS)\nCONFIG_INITRAMFS_SOURCE="$(BUILD_PATH)/init-cpio-spec.txt"\n' >> $(KERNEL_BUILD_PATH)/minimal.config
+	cat arch/$(ARCH).config >> $(KERNEL_BUILD_PATH)/minimal.config
+	$(MAKE) -C $(KERNEL_PATH) O=$(KERNEL_BUILD_PATH) ARCH=$(KERNEL_ARCH) allnoconfig
+	cd $(KERNEL_BUILD_PATH) && ARCH=$(KERNEL_ARCH) $(KERNEL_PATH)/scripts/kconfig/merge_config.sh -n $(KERNEL_BUILD_PATH)/.config $(KERNEL_BUILD_PATH)/minimal.config
+	$(if $(findstring yes,$(DEBUG_KERNEL)),cp debug.config $(KERNEL_BUILD_PATH) && cd $(KERNEL_BUILD_PATH) && ARCH=$(KERNEL_ARCH) $(KERNEL_PATH)/scripts/kconfig/merge_config.sh -n $(KERNEL_BUILD_PATH)/.config debug.config,)
+
+$(KERNEL_BZIMAGE): $(KERNEL_BUILD_PATH)/.config $(BUILD_PATH)/init-cpio-spec.txt $(MUSL_PATH)/lib/libc.so $(IPERF_PATH)/src/iperf3 $(IPUTILS_PATH)/ping $(BASH_PATH)/bash $(IPROUTE2_PATH)/misc/ss $(IPROUTE2_PATH)/ip/ip $(IPTABLES_PATH)/iptables/xtables-multi $(NMAP_PATH)/ncat/ncat $(WIREGUARD_TOOLS_PATH)/src/tools/wg $(BUILD_PATH)/init ../netns.sh $(WIREGUARD_SOURCES)
+	$(MAKE) -C $(KERNEL_PATH) O=$(KERNEL_BUILD_PATH) ARCH=$(KERNEL_ARCH) CROSS_COMPILE=$(CROSS_COMPILE) CC="$(NOPIE_GCC)"
+
+$(BUILD_PATH)/include/linux/.installed: | $(KERNEL_BUILD_PATH)/.config
+	$(MAKE) -C $(KERNEL_PATH) O=$(KERNEL_BUILD_PATH) INSTALL_HDR_PATH=$(BUILD_PATH) ARCH=$(KERNEL_ARCH) CROSS_COMPILE=$(CROSS_COMPILE) headers_install
+	touch $@
+
+$(MUSL_PATH)/lib/libc.so: $(MUSL_TAR)
+	mkdir -p $(BUILD_PATH)
+	flock -s $<.lock tar -C $(BUILD_PATH) -xf $<
+	cd $(MUSL_PATH) && CC=$(REAL_CC) ./configure --prefix=/ --disable-static --build=$(CBUILD)
+	$(MAKE) -C $(MUSL_PATH)
+	$(STRIP) -s $@
+
+$(BUILD_PATH)/include/.installed: $(MUSL_PATH)/lib/libc.so
+	$(MAKE) -C $(MUSL_PATH) DESTDIR=$(BUILD_PATH) install-headers
+	touch $@
+
+$(MUSL_CC): $(MUSL_PATH)/lib/libc.so
+	sh $(MUSL_PATH)/tools/musl-gcc.specs.sh $(BUILD_PATH)/include $(MUSL_PATH)/lib /lib/ld-linux.so.1 > $(BUILD_PATH)/musl-gcc.specs
+	printf '#!/bin/sh\nexec "$(REAL_CC)" --specs="$(BUILD_PATH)/musl-gcc.specs" -fno-stack-protector -no-pie "$$@"\n' > $(BUILD_PATH)/musl-gcc
+	chmod +x $(BUILD_PATH)/musl-gcc
+
+$(IPERF_PATH)/.installed: $(IPERF_TAR)
+	mkdir -p $(BUILD_PATH)
+	flock -s $<.lock tar -C $(BUILD_PATH) -xf $<
+	sed -i '1s/^/#include <stdint.h>/' $(IPERF_PATH)/src/cjson.h $(IPERF_PATH)/src/timer.h
+	sed -i -r 's/-p?g//g' $(IPERF_PATH)/src/Makefile*
+	touch $@
+
+$(IPERF_PATH)/src/iperf3: | $(IPERF_PATH)/.installed $(USERSPACE_DEPS)
+	cd $(IPERF_PATH) && CFLAGS="$(CFLAGS) -D_GNU_SOURCE" ./configure --prefix=/ $(CROSS_COMPILE_FLAG) --enable-static --disable-shared
+	$(MAKE) -C $(IPERF_PATH)
+	$(STRIP) -s $@
+
+$(LIBMNL_PATH)/.installed: $(LIBMNL_TAR)
+	flock -s $<.lock tar -C $(BUILD_PATH) -xf $<
+	touch $@
+
+$(LIBMNL_PATH)/src/.libs/libmnl.a: | $(LIBMNL_PATH)/.installed $(USERSPACE_DEPS)
+	cd $(LIBMNL_PATH) && ./configure --prefix=/ $(CROSS_COMPILE_FLAG) --enable-static --disable-shared
+	$(MAKE) -C $(LIBMNL_PATH)
+	sed -i 's:prefix=.*:prefix=$(LIBMNL_PATH):' $(LIBMNL_PATH)/libmnl.pc
+
+$(WIREGUARD_TOOLS_PATH)/.installed: $(WIREGUARD_TOOLS_TAR)
+	flock -s $<.lock tar -C $(BUILD_PATH) -xf $<
+	touch $@
+
+$(WIREGUARD_TOOLS_PATH)/src/tools/wg: | $(WIREGUARD_TOOLS_PATH)/.installed $(LIBMNL_PATH)/src/.libs/libmnl.a $(USERSPACE_DEPS)
+	LDFLAGS="$(LDFLAGS) -L$(LIBMNL_PATH)/src/.libs" $(MAKE) -C $(WIREGUARD_TOOLS_PATH)/src/tools LIBMNL_CFLAGS="-I$(LIBMNL_PATH)/include" LIBMNL_LDLIBS="-lmnl" wg
+	$(STRIP) -s $@
+
+$(BUILD_PATH)/init: init.c | $(USERSPACE_DEPS)
+	mkdir -p $(BUILD_PATH)
+	$(MUSL_CC) -o $@ $(CFLAGS) $(LDFLAGS) -std=gnu11 $<
+	$(STRIP) -s $@
+
+$(IPUTILS_PATH)/.installed: $(IPUTILS_TAR)
+	mkdir -p $(BUILD_PATH)
+	flock -s $<.lock tar -C $(BUILD_PATH) -xf $<
+	touch $@
+
+$(IPUTILS_PATH)/ping: | $(IPUTILS_PATH)/.installed $(USERSPACE_DEPS)
+	$(MAKE) -C $(IPUTILS_PATH) USE_CAP=no USE_IDN=no USE_NETTLE=no USE_CRYPTO=no ping
+	$(STRIP) -s $@
+
+$(BASH_PATH)/.installed: $(BASH_TAR)
+	mkdir -p $(BUILD_PATH)
+	flock -s $<.lock tar -C $(BUILD_PATH) -xf $<
+	touch $@
+
+$(BASH_PATH)/bash: | $(BASH_PATH)/.installed $(USERSPACE_DEPS)
+	cd $(BASH_PATH) && ./configure --prefix=/ $(CROSS_COMPILE_FLAG) --without-bash-malloc --disable-debugger --disable-help-builtin --disable-history --disable-multibyte --disable-progcomp --disable-readline --disable-mem-scramble
+	$(MAKE) -C $(BASH_PATH)
+	$(STRIP) -s $@
+
+$(IPROUTE2_PATH)/.installed: $(IPROUTE2_TAR)
+	mkdir -p $(BUILD_PATH)
+	flock -s $<.lock tar -C $(BUILD_PATH) -xf $<
+	printf 'CC:=$(CC)\nPKG_CONFIG:=pkg-config\nTC_CONFIG_XT:=n\nTC_CONFIG_ATM:=n\nTC_CONFIG_IPSET:=n\nIP_CONFIG_SETNS:=y\nHAVE_ELF:=n\nHAVE_MNL:=y\nHAVE_BERKELEY_DB:=n\nHAVE_LATEX:=n\nHAVE_PDFLATEX:=n\nCFLAGS+=-DHAVE_SETNS -DHAVE_LIBMNL -I$(LIBMNL_PATH)/include\nLDLIBS+=-lmnl' > $(IPROUTE2_PATH)/config.mk
+	printf 'lib: snapshot\n\t$$(MAKE) -C lib\nip/ip: lib\n\t$$(MAKE) -C ip ip\nmisc/ss: lib\n\t$$(MAKE) -C misc ss\n' >> $(IPROUTE2_PATH)/Makefile
+	touch $@
+
+$(IPROUTE2_PATH)/ip/ip: | $(IPROUTE2_PATH)/.installed $(LIBMNL_PATH)/src/.libs/libmnl.a $(USERSPACE_DEPS)
+	LDFLAGS="$(LDFLAGS) -L$(LIBMNL_PATH)/src/.libs" PKG_CONFIG_LIBDIR="$(LIBMNL_PATH)" $(MAKE) -C $(IPROUTE2_PATH) PREFIX=/ ip/ip
+	$(STRIP) -s $(IPROUTE2_PATH)/ip/ip
+
+$(IPROUTE2_PATH)/misc/ss: | $(IPROUTE2_PATH)/.installed $(LIBMNL_PATH)/src/.libs/libmnl.a $(USERSPACE_DEPS)
+	LDFLAGS="$(LDFLAGS) -L$(LIBMNL_PATH)/src/.libs" PKG_CONFIG_LIBDIR="$(LIBMNL_PATH)" $(MAKE) -C $(IPROUTE2_PATH) PREFIX=/ misc/ss
+	$(STRIP) -s $(IPROUTE2_PATH)/misc/ss
+
+$(IPTABLES_PATH)/.installed: $(IPTABLES_TAR)
+	mkdir -p $(BUILD_PATH)
+	flock -s $<.lock tar -C $(BUILD_PATH) -xf $<
+	sed -i -e "/nfnetlink=[01]/s:=[01]:=0:" -e "/nfconntrack=[01]/s:=[01]:=0:" $(IPTABLES_PATH)/configure
+	touch $@
+
+$(IPTABLES_PATH)/iptables/xtables-multi: | $(IPTABLES_PATH)/.installed $(LIBMNL_PATH)/src/.libs/libmnl.a $(USERSPACE_DEPS)
+	cd $(IPTABLES_PATH) && PKG_CONFIG_LIBDIR="$(LIBMNL_PATH)" ./configure --prefix=/ $(CROSS_COMPILE_FLAG) --enable-static --disable-shared --disable-nftables --disable-bpf-compiler --disable-nfsynproxy --disable-libipq --with-kernel=$(BUILD_PATH)/include
+	$(MAKE) -C $(IPTABLES_PATH)
+	$(STRIP) -s $@
+
+$(NMAP_PATH)/.installed: $(NMAP_TAR)
+	mkdir -p $(BUILD_PATH)
+	flock -s $<.lock tar -C $(BUILD_PATH) -xf $<
+	touch $@
+
+$(NMAP_PATH)/ncat/ncat: | $(NMAP_PATH)/.installed $(USERSPACE_DEPS)
+	cd $(NMAP_PATH) && ./configure --prefix=/ $(CROSS_COMPILE_FLAG) --enable-static --disable-shared --without-ndiff --without-zenmap --without-nping --with-libpcap=included --with-libpcre=included --with-libdnet=included --without-liblua --with-liblinear=included --without-nmap-update --without-openssl --with-pcap=linux
+	$(MAKE) -C $(NMAP_PATH) build-ncat
+	$(STRIP) -s $@
+
+clean:
+	rm -rf $(BUILD_PATH)
+
+distclean: clean
+	rm -rf $(DISTFILES_PATH)
+
+menuconfig: $(KERNEL_BUILD_PATH)/.config
+	$(MAKE) -C $(KERNEL_PATH) O=$(KERNEL_BUILD_PATH) ARCH=$(KERNEL_ARCH) CROSS_COMPILE=$(CROSS_COMPILE) CC="$(NOPIE_GCC)" menuconfig
+
+.PHONY: qemu build clean distclean menuconfig
+.DELETE_ON_ERROR:
diff --git a/tools/testing/selftests/wireguard/qemu/arch/aarch64.config b/tools/testing/selftests/wireguard/qemu/arch/aarch64.config
new file mode 100644
index 000000000000..3d063bb247bb
--- /dev/null
+++ b/tools/testing/selftests/wireguard/qemu/arch/aarch64.config
@@ -0,0 +1,5 @@
+CONFIG_SERIAL_AMBA_PL011=y
+CONFIG_SERIAL_AMBA_PL011_CONSOLE=y
+CONFIG_CMDLINE_BOOL=y
+CONFIG_CMDLINE="console=ttyAMA0 wg.success=ttyAMA1"
+CONFIG_FRAME_WARN=1280
diff --git a/tools/testing/selftests/wireguard/qemu/arch/aarch64_be.config b/tools/testing/selftests/wireguard/qemu/arch/aarch64_be.config
new file mode 100644
index 000000000000..dbdc7e406a7b
--- /dev/null
+++ b/tools/testing/selftests/wireguard/qemu/arch/aarch64_be.config
@@ -0,0 +1,6 @@
+CONFIG_CPU_BIG_ENDIAN=y
+CONFIG_SERIAL_AMBA_PL011=y
+CONFIG_SERIAL_AMBA_PL011_CONSOLE=y
+CONFIG_CMDLINE_BOOL=y
+CONFIG_CMDLINE="console=ttyAMA0 wg.success=ttyAMA1"
+CONFIG_FRAME_WARN=1280
diff --git a/tools/testing/selftests/wireguard/qemu/arch/arm.config b/tools/testing/selftests/wireguard/qemu/arch/arm.config
new file mode 100644
index 000000000000..148f49905418
--- /dev/null
+++ b/tools/testing/selftests/wireguard/qemu/arch/arm.config
@@ -0,0 +1,9 @@
+CONFIG_MMU=y
+CONFIG_ARCH_MULTI_V7=y
+CONFIG_ARCH_VIRT=y
+CONFIG_THUMB2_KERNEL=n
+CONFIG_SERIAL_AMBA_PL011=y
+CONFIG_SERIAL_AMBA_PL011_CONSOLE=y
+CONFIG_CMDLINE_BOOL=y
+CONFIG_CMDLINE="console=ttyAMA0 wg.success=ttyAMA1"
+CONFIG_FRAME_WARN=1024
diff --git a/tools/testing/selftests/wireguard/qemu/arch/armeb.config b/tools/testing/selftests/wireguard/qemu/arch/armeb.config
new file mode 100644
index 000000000000..bd76b07d00a2
--- /dev/null
+++ b/tools/testing/selftests/wireguard/qemu/arch/armeb.config
@@ -0,0 +1,10 @@
+CONFIG_MMU=y
+CONFIG_ARCH_MULTI_V7=y
+CONFIG_ARCH_VIRT=y
+CONFIG_THUMB2_KERNEL=n
+CONFIG_SERIAL_AMBA_PL011=y
+CONFIG_SERIAL_AMBA_PL011_CONSOLE=y
+CONFIG_CMDLINE_BOOL=y
+CONFIG_CMDLINE="console=ttyAMA0 wg.success=ttyAMA1"
+CONFIG_CPU_BIG_ENDIAN=y
+CONFIG_FRAME_WARN=1024
diff --git a/tools/testing/selftests/wireguard/qemu/arch/i686.config b/tools/testing/selftests/wireguard/qemu/arch/i686.config
new file mode 100644
index 000000000000..a85025d7206e
--- /dev/null
+++ b/tools/testing/selftests/wireguard/qemu/arch/i686.config
@@ -0,0 +1,5 @@
+CONFIG_SERIAL_8250=y
+CONFIG_SERIAL_8250_CONSOLE=y
+CONFIG_CMDLINE_BOOL=y
+CONFIG_CMDLINE="console=ttyS0 wg.success=ttyS1"
+CONFIG_FRAME_WARN=1024
diff --git a/tools/testing/selftests/wireguard/qemu/arch/m68k.config b/tools/testing/selftests/wireguard/qemu/arch/m68k.config
new file mode 100644
index 000000000000..5381ea10896c
--- /dev/null
+++ b/tools/testing/selftests/wireguard/qemu/arch/m68k.config
@@ -0,0 +1,9 @@
+CONFIG_MMU=y
+CONFIG_M68040=y
+CONFIG_MAC=y
+CONFIG_SERIAL_PMACZILOG=y
+CONFIG_SERIAL_PMACZILOG_TTYS=y
+CONFIG_SERIAL_PMACZILOG_CONSOLE=y
+CONFIG_CMDLINE_BOOL=y
+CONFIG_CMDLINE="console=ttyS0 wg.success=ttyS1"
+CONFIG_FRAME_WARN=1024
diff --git a/tools/testing/selftests/wireguard/qemu/arch/mips.config b/tools/testing/selftests/wireguard/qemu/arch/mips.config
new file mode 100644
index 000000000000..df71d6b95546
--- /dev/null
+++ b/tools/testing/selftests/wireguard/qemu/arch/mips.config
@@ -0,0 +1,11 @@
+CONFIG_CPU_MIPS32_R2=y
+CONFIG_MIPS_MALTA=y
+CONFIG_MIPS_CPS=y
+CONFIG_MIPS_FP_SUPPORT=y
+CONFIG_POWER_RESET=y
+CONFIG_POWER_RESET_SYSCON=y
+CONFIG_SERIAL_8250=y
+CONFIG_SERIAL_8250_CONSOLE=y
+CONFIG_CMDLINE_BOOL=y
+CONFIG_CMDLINE="console=ttyS0 wg.success=ttyS1"
+CONFIG_FRAME_WARN=1024
diff --git a/tools/testing/selftests/wireguard/qemu/arch/mips64.config b/tools/testing/selftests/wireguard/qemu/arch/mips64.config
new file mode 100644
index 000000000000..90c783f725c4
--- /dev/null
+++ b/tools/testing/selftests/wireguard/qemu/arch/mips64.config
@@ -0,0 +1,14 @@
+CONFIG_64BIT=y
+CONFIG_CPU_MIPS64_R2=y
+CONFIG_MIPS32_N32=y
+CONFIG_CPU_HAS_MSA=y
+CONFIG_MIPS_MALTA=y
+CONFIG_MIPS_CPS=y
+CONFIG_MIPS_FP_SUPPORT=y
+CONFIG_POWER_RESET=y
+CONFIG_POWER_RESET_SYSCON=y
+CONFIG_SERIAL_8250=y
+CONFIG_SERIAL_8250_CONSOLE=y
+CONFIG_CMDLINE_BOOL=y
+CONFIG_CMDLINE="console=ttyS0 wg.success=ttyS1"
+CONFIG_FRAME_WARN=1280
diff --git a/tools/testing/selftests/wireguard/qemu/arch/mips64el.config b/tools/testing/selftests/wireguard/qemu/arch/mips64el.config
new file mode 100644
index 000000000000..435b0b43e00c
--- /dev/null
+++ b/tools/testing/selftests/wireguard/qemu/arch/mips64el.config
@@ -0,0 +1,15 @@
+CONFIG_64BIT=y
+CONFIG_CPU_MIPS64_R2=y
+CONFIG_MIPS32_N32=y
+CONFIG_CPU_HAS_MSA=y
+CONFIG_MIPS_MALTA=y
+CONFIG_CPU_LITTLE_ENDIAN=y
+CONFIG_MIPS_CPS=y
+CONFIG_MIPS_FP_SUPPORT=y
+CONFIG_POWER_RESET=y
+CONFIG_POWER_RESET_SYSCON=y
+CONFIG_SERIAL_8250=y
+CONFIG_SERIAL_8250_CONSOLE=y
+CONFIG_CMDLINE_BOOL=y
+CONFIG_CMDLINE="console=ttyS0 wg.success=ttyS1"
+CONFIG_FRAME_WARN=1280
diff --git a/tools/testing/selftests/wireguard/qemu/arch/mipsel.config b/tools/testing/selftests/wireguard/qemu/arch/mipsel.config
new file mode 100644
index 000000000000..62bb50c4a85f
--- /dev/null
+++ b/tools/testing/selftests/wireguard/qemu/arch/mipsel.config
@@ -0,0 +1,12 @@
+CONFIG_CPU_MIPS32_R2=y
+CONFIG_MIPS_MALTA=y
+CONFIG_CPU_LITTLE_ENDIAN=y
+CONFIG_MIPS_CPS=y
+CONFIG_MIPS_FP_SUPPORT=y
+CONFIG_POWER_RESET=y
+CONFIG_POWER_RESET_SYSCON=y
+CONFIG_SERIAL_8250=y
+CONFIG_SERIAL_8250_CONSOLE=y
+CONFIG_CMDLINE_BOOL=y
+CONFIG_CMDLINE="console=ttyS0 wg.success=ttyS1"
+CONFIG_FRAME_WARN=1024
diff --git a/tools/testing/selftests/wireguard/qemu/arch/powerpc.config b/tools/testing/selftests/wireguard/qemu/arch/powerpc.config
new file mode 100644
index 000000000000..57957093b71b
--- /dev/null
+++ b/tools/testing/selftests/wireguard/qemu/arch/powerpc.config
@@ -0,0 +1,10 @@
+CONFIG_PPC_QEMU_E500=y
+CONFIG_FSL_SOC_BOOKE=y
+CONFIG_PPC_85xx=y
+CONFIG_PHYS_64BIT=y
+CONFIG_SERIAL_8250=y
+CONFIG_SERIAL_8250_CONSOLE=y
+CONFIG_MATH_EMULATION=y
+CONFIG_CMDLINE_BOOL=y
+CONFIG_CMDLINE="console=ttyS0 wg.success=ttyS1"
+CONFIG_FRAME_WARN=1024
diff --git a/tools/testing/selftests/wireguard/qemu/arch/powerpc64le.config b/tools/testing/selftests/wireguard/qemu/arch/powerpc64le.config
new file mode 100644
index 000000000000..990c510a9cfa
--- /dev/null
+++ b/tools/testing/selftests/wireguard/qemu/arch/powerpc64le.config
@@ -0,0 +1,12 @@
+CONFIG_PPC64=y
+CONFIG_PPC_PSERIES=y
+CONFIG_ALTIVEC=y
+CONFIG_VSX=y
+CONFIG_PPC_OF_BOOT_TRAMPOLINE=y
+CONFIG_PPC_RADIX_MMU=y
+CONFIG_HVC_CONSOLE=y
+CONFIG_CPU_LITTLE_ENDIAN=y
+CONFIG_CMDLINE_BOOL=y
+CONFIG_CMDLINE="console=hvc0 wg.success=hvc1"
+CONFIG_SECTION_MISMATCH_WARN_ONLY=y
+CONFIG_FRAME_WARN=1280
diff --git a/tools/testing/selftests/wireguard/qemu/arch/x86_64.config b/tools/testing/selftests/wireguard/qemu/arch/x86_64.config
new file mode 100644
index 000000000000..00a1ef4869d5
--- /dev/null
+++ b/tools/testing/selftests/wireguard/qemu/arch/x86_64.config
@@ -0,0 +1,5 @@
+CONFIG_SERIAL_8250=y
+CONFIG_SERIAL_8250_CONSOLE=y
+CONFIG_CMDLINE_BOOL=y
+CONFIG_CMDLINE="console=ttyS0 wg.success=ttyS1"
+CONFIG_FRAME_WARN=1280
diff --git a/tools/testing/selftests/wireguard/qemu/debug.config b/tools/testing/selftests/wireguard/qemu/debug.config
new file mode 100644
index 000000000000..b9c72706fe4d
--- /dev/null
+++ b/tools/testing/selftests/wireguard/qemu/debug.config
@@ -0,0 +1,67 @@
+CONFIG_LOCALVERSION="-debug"
+CONFIG_ENABLE_WARN_DEPRECATED=y
+CONFIG_ENABLE_MUST_CHECK=y
+CONFIG_FRAME_POINTER=y
+CONFIG_STACK_VALIDATION=y
+CONFIG_DEBUG_KERNEL=y
+CONFIG_DEBUG_INFO=y
+CONFIG_DEBUG_INFO_DWARF4=y
+CONFIG_PAGE_EXTENSION=y
+CONFIG_PAGE_POISONING=y
+CONFIG_DEBUG_OBJECTS=y
+CONFIG_DEBUG_OBJECTS_FREE=y
+CONFIG_DEBUG_OBJECTS_TIMERS=y
+CONFIG_DEBUG_OBJECTS_WORK=y
+CONFIG_DEBUG_OBJECTS_RCU_HEAD=y
+CONFIG_DEBUG_OBJECTS_PERCPU_COUNTER=y
+CONFIG_DEBUG_OBJECTS_ENABLE_DEFAULT=1
+CONFIG_SLUB_DEBUG_ON=y
+CONFIG_DEBUG_VM=y
+CONFIG_DEBUG_MEMORY_INIT=y
+CONFIG_HAVE_DEBUG_STACKOVERFLOW=y
+CONFIG_DEBUG_STACKOVERFLOW=y
+CONFIG_HAVE_ARCH_KMEMCHECK=y
+CONFIG_HAVE_ARCH_KASAN=y
+CONFIG_KASAN=y
+CONFIG_KASAN_INLINE=y
+CONFIG_UBSAN=y
+CONFIG_UBSAN_SANITIZE_ALL=y
+CONFIG_UBSAN_NO_ALIGNMENT=y
+CONFIG_UBSAN_NULL=y
+CONFIG_DEBUG_KMEMLEAK=y
+CONFIG_DEBUG_KMEMLEAK_EARLY_LOG_SIZE=8192
+CONFIG_DEBUG_STACK_USAGE=y
+CONFIG_DEBUG_SHIRQ=y
+CONFIG_WQ_WATCHDOG=y
+CONFIG_SCHED_DEBUG=y
+CONFIG_SCHED_INFO=y
+CONFIG_SCHEDSTATS=y
+CONFIG_SCHED_STACK_END_CHECK=y
+CONFIG_DEBUG_TIMEKEEPING=y
+CONFIG_TIMER_STATS=y
+CONFIG_DEBUG_PREEMPT=y
+CONFIG_DEBUG_RT_MUTEXES=y
+CONFIG_DEBUG_SPINLOCK=y
+CONFIG_DEBUG_MUTEXES=y
+CONFIG_DEBUG_LOCK_ALLOC=y
+CONFIG_PROVE_LOCKING=y
+CONFIG_LOCKDEP=y
+CONFIG_DEBUG_ATOMIC_SLEEP=y
+CONFIG_TRACE_IRQFLAGS=y
+CONFIG_DEBUG_BUGVERBOSE=y
+CONFIG_DEBUG_LIST=y
+CONFIG_DEBUG_PI_LIST=y
+CONFIG_PROVE_RCU=y
+CONFIG_SPARSE_RCU_POINTER=y
+CONFIG_RCU_CPU_STALL_TIMEOUT=21
+CONFIG_RCU_TRACE=y
+CONFIG_RCU_EQS_DEBUG=y
+CONFIG_USER_STACKTRACE_SUPPORT=y
+CONFIG_DEBUG_SG=y
+CONFIG_DEBUG_NOTIFIERS=y
+CONFIG_DOUBLEFAULT=y
+CONFIG_X86_DEBUG_FPU=y
+CONFIG_DEBUG_SECTION_MISMATCH=y
+CONFIG_DEBUG_PAGEALLOC=y
+CONFIG_DEBUG_PAGEALLOC_ENABLE_DEFAULT=y
+CONFIG_DEBUG_WW_MUTEX_SLOWPATH=y
diff --git a/tools/testing/selftests/wireguard/qemu/init.c b/tools/testing/selftests/wireguard/qemu/init.c
new file mode 100644
index 000000000000..51e5ddedee88
--- /dev/null
+++ b/tools/testing/selftests/wireguard/qemu/init.c
@@ -0,0 +1,284 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#define _GNU_SOURCE
+#include <unistd.h>
+#include <errno.h>
+#include <string.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <stdbool.h>
+#include <fcntl.h>
+#include <sys/wait.h>
+#include <sys/mount.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <sys/types.h>
+#include <sys/io.h>
+#include <sys/ioctl.h>
+#include <sys/reboot.h>
+#include <sys/utsname.h>
+#include <sys/sendfile.h>
+#include <linux/random.h>
+#include <linux/version.h>
+
+__attribute__((noreturn)) static void poweroff(void)
+{
+	fflush(stdout);
+	fflush(stderr);
+	reboot(RB_AUTOBOOT);
+	sleep(30);
+	fprintf(stderr, "\x1b[37m\x1b[41m\x1b[1mFailed to power off!!!\x1b[0m\n");
+	exit(1);
+}
+
+static void panic(const char *what)
+{
+	fprintf(stderr, "\n\n\x1b[37m\x1b[41m\x1b[1mSOMETHING WENT HORRIBLY WRONG\x1b[0m\n\n    \x1b[31m\x1b[1m%s: %s\x1b[0m\n\n\x1b[37m\x1b[44m\x1b[1mPower off...\x1b[0m\n\n", what, strerror(errno));
+	poweroff();
+}
+
+#define pretty_message(msg) puts("\x1b[32m\x1b[1m" msg "\x1b[0m")
+
+static void print_banner(void)
+{
+	struct utsname utsname;
+	int len;
+
+	if (uname(&utsname) < 0)
+		panic("uname");
+
+	len = strlen("    WireGuard Test Suite on       ") + strlen(utsname.sysname) + strlen(utsname.release) + strlen(utsname.machine);
+	printf("\x1b[45m\x1b[33m\x1b[1m%*.s\x1b[0m\n\x1b[45m\x1b[33m\x1b[1m    WireGuard Test Suite on %s %s %s    \x1b[0m\n\x1b[45m\x1b[33m\x1b[1m%*.s\x1b[0m\n\n", len, "", utsname.sysname, utsname.release, utsname.machine, len, "");
+}
+
+static void seed_rng(void)
+{
+	int fd;
+	struct {
+		int entropy_count;
+		int buffer_size;
+		unsigned char buffer[256];
+	} entropy = {
+		.entropy_count = sizeof(entropy.buffer) * 8,
+		.buffer_size = sizeof(entropy.buffer),
+		.buffer = "Adding real entropy is not actually important for these tests. Don't try this at home, kids!"
+	};
+
+	if (mknod("/dev/urandom", S_IFCHR | 0644, makedev(1, 9)))
+		panic("mknod(/dev/urandom)");
+	fd = open("/dev/urandom", O_WRONLY);
+	if (fd < 0)
+		panic("open(urandom)");
+	for (int i = 0; i < 256; ++i) {
+		if (ioctl(fd, RNDADDENTROPY, &entropy) < 0)
+			panic("ioctl(urandom)");
+	}
+	close(fd);
+}
+
+static void mount_filesystems(void)
+{
+	pretty_message("[+] Mounting filesystems...");
+	mkdir("/dev", 0755);
+	mkdir("/proc", 0755);
+	mkdir("/sys", 0755);
+	mkdir("/tmp", 0755);
+	mkdir("/run", 0755);
+	mkdir("/var", 0755);
+	if (mount("none", "/dev", "devtmpfs", 0, NULL))
+		panic("devtmpfs mount");
+	if (mount("none", "/proc", "proc", 0, NULL))
+		panic("procfs mount");
+	if (mount("none", "/sys", "sysfs", 0, NULL))
+		panic("sysfs mount");
+	if (mount("none", "/tmp", "tmpfs", 0, NULL))
+		panic("tmpfs mount");
+	if (mount("none", "/run", "tmpfs", 0, NULL))
+		panic("tmpfs mount");
+	if (mount("none", "/sys/kernel/debug", "debugfs", 0, NULL))
+		; /* Not a problem if it fails.*/
+	if (symlink("/run", "/var/run"))
+		panic("run symlink");
+	if (symlink("/proc/self/fd", "/dev/fd"))
+		panic("fd symlink");
+}
+
+static void enable_logging(void)
+{
+	int fd;
+	pretty_message("[+] Enabling logging...");
+	fd = open("/proc/sys/kernel/printk", O_WRONLY);
+	if (fd >= 0) {
+		if (write(fd, "9\n", 2) != 2)
+			panic("write(printk)");
+		close(fd);
+	}
+	fd = open("/proc/sys/debug/exception-trace", O_WRONLY);
+	if (fd >= 0) {
+		if (write(fd, "1\n", 2) != 2)
+			panic("write(exception-trace)");
+		close(fd);
+	}
+	fd = open("/proc/sys/kernel/panic_on_warn", O_WRONLY);
+	if (fd >= 0) {
+		if (write(fd, "1\n", 2) != 2)
+			panic("write(panic_on_warn)");
+		close(fd);
+	}
+}
+
+static void kmod_selftests(void)
+{
+	FILE *file;
+	char line[2048], *start, *pass;
+	bool success = true;
+	pretty_message("[+] Module self-tests:");
+	file = fopen("/proc/kmsg", "r");
+	if (!file)
+		panic("fopen(kmsg)");
+	if (fcntl(fileno(file), F_SETFL, O_NONBLOCK) < 0)
+		panic("fcntl(kmsg, nonblock)");
+	while (fgets(line, sizeof(line), file)) {
+		start = strstr(line, "wireguard: ");
+		if (!start)
+			continue;
+		start += 11;
+		*strchrnul(start, '\n') = '\0';
+		if (strstr(start, "www.wireguard.com"))
+			break;
+		pass = strstr(start, ": pass");
+		if (!pass || pass[6] != '\0') {
+			success = false;
+			printf(" \x1b[31m*  %s\x1b[0m\n", start);
+		} else
+			printf(" \x1b[32m*  %s\x1b[0m\n", start);
+	}
+	fclose(file);
+	if (!success) {
+		puts("\x1b[31m\x1b[1m[-] Tests failed! \u2639\x1b[0m");
+		poweroff();
+	}
+}
+
+static void launch_tests(void)
+{
+	char cmdline[4096], *success_dev;
+	int status, fd;
+	pid_t pid;
+
+	pretty_message("[+] Launching tests...");
+	pid = fork();
+	if (pid == -1)
+		panic("fork");
+	else if (pid == 0) {
+		execl("/init.sh", "init", NULL);
+		panic("exec");
+	}
+	if (waitpid(pid, &status, 0) < 0)
+		panic("waitpid");
+	if (WIFEXITED(status) && WEXITSTATUS(status) == 0) {
+		pretty_message("[+] Tests successful! :-)");
+		fd = open("/proc/cmdline", O_RDONLY);
+		if (fd < 0)
+			panic("open(/proc/cmdline)");
+		if (read(fd, cmdline, sizeof(cmdline) - 1) <= 0)
+			panic("read(/proc/cmdline)");
+		cmdline[sizeof(cmdline) - 1] = '\0';
+		for (success_dev = strtok(cmdline, " \n"); success_dev; success_dev = strtok(NULL, " \n")) {
+			if (strncmp(success_dev, "wg.success=", 11))
+				continue;
+			memcpy(success_dev + 11 - 5, "/dev/", 5);
+			success_dev += 11 - 5;
+			break;
+		}
+		if (!success_dev || !strlen(success_dev))
+			panic("Unable to find success device");
+
+		fd = open(success_dev, O_WRONLY);
+		if (fd < 0)
+			panic("open(success_dev)");
+		if (write(fd, "success\n", 8) != 8)
+			panic("write(success_dev)");
+		close(fd);
+	} else {
+		const char *why = "unknown cause";
+		int what = -1;
+
+		if (WIFEXITED(status)) {
+			why = "exit code";
+			what = WEXITSTATUS(status);
+		} else if (WIFSIGNALED(status)) {
+			why = "signal";
+			what = WTERMSIG(status);
+		}
+		printf("\x1b[31m\x1b[1m[-] Tests failed with %s %d! \u2639\x1b[0m\n", why, what);
+	}
+}
+
+static void ensure_console(void)
+{
+	for (unsigned int i = 0; i < 1000; ++i) {
+		int fd = open("/dev/console", O_RDWR);
+		if (fd < 0) {
+			usleep(50000);
+			continue;
+		}
+		dup2(fd, 0);
+		dup2(fd, 1);
+		dup2(fd, 2);
+		close(fd);
+		if (write(1, "\0\0\0\0\n", 5) == 5)
+			return;
+	}
+	panic("Unable to open console device");
+}
+
+static void clear_leaks(void)
+{
+	int fd;
+
+	fd = open("/sys/kernel/debug/kmemleak", O_WRONLY);
+	if (fd < 0)
+		return;
+	pretty_message("[+] Starting memory leak detection...");
+	write(fd, "clear\n", 5);
+	close(fd);
+}
+
+static void check_leaks(void)
+{
+	int fd;
+
+	fd = open("/sys/kernel/debug/kmemleak", O_WRONLY);
+	if (fd < 0)
+		return;
+	pretty_message("[+] Scanning for memory leaks...");
+	sleep(2); /* Wait for any grace periods. */
+	write(fd, "scan\n", 5);
+	close(fd);
+
+	fd = open("/sys/kernel/debug/kmemleak", O_RDONLY);
+	if (fd < 0)
+		return;
+	if (sendfile(1, fd, NULL, 0x7ffff000) > 0)
+		panic("Memory leaks encountered");
+	close(fd);
+}
+
+int main(int argc, char *argv[])
+{
+	seed_rng();
+	ensure_console();
+	print_banner();
+	mount_filesystems();
+	kmod_selftests();
+	enable_logging();
+	clear_leaks();
+	launch_tests();
+	check_leaks();
+	poweroff();
+	return 1;
+}
diff --git a/tools/testing/selftests/wireguard/qemu/kernel.config b/tools/testing/selftests/wireguard/qemu/kernel.config
new file mode 100644
index 000000000000..9cca30206014
--- /dev/null
+++ b/tools/testing/selftests/wireguard/qemu/kernel.config
@@ -0,0 +1,86 @@
+CONFIG_LOCALVERSION=""
+CONFIG_NET=y
+CONFIG_NETDEVICES=y
+CONFIG_NET_CORE=y
+CONFIG_NET_IPIP=y
+CONFIG_DUMMY=y
+CONFIG_VETH=y
+CONFIG_MULTIUSER=y
+CONFIG_NAMESPACES=y
+CONFIG_NET_NS=y
+CONFIG_UNIX=y
+CONFIG_INET=y
+CONFIG_IPV6=y
+CONFIG_NETFILTER=y
+CONFIG_NETFILTER_ADVANCED=y
+CONFIG_NF_CONNTRACK=y
+CONFIG_NF_NAT=y
+CONFIG_NETFILTER_XTABLES=y
+CONFIG_NETFILTER_XT_NAT=y
+CONFIG_NETFILTER_XT_MATCH_LENGTH=y
+CONFIG_NF_CONNTRACK_IPV4=y
+CONFIG_NF_NAT_IPV4=y
+CONFIG_IP_NF_IPTABLES=y
+CONFIG_IP_NF_FILTER=y
+CONFIG_IP_NF_NAT=y
+CONFIG_IP_ADVANCED_ROUTER=y
+CONFIG_IP_MULTIPLE_TABLES=y
+CONFIG_IPV6_MULTIPLE_TABLES=y
+CONFIG_TTY=y
+CONFIG_BINFMT_ELF=y
+CONFIG_BINFMT_SCRIPT=y
+CONFIG_VDSO=y
+CONFIG_VIRTUALIZATION=y
+CONFIG_HYPERVISOR_GUEST=y
+CONFIG_PARAVIRT=y
+CONFIG_KVM_GUEST=y
+CONFIG_PARAVIRT_SPINLOCKS=y
+CONFIG_PRINTK=y
+CONFIG_KALLSYMS=y
+CONFIG_BUG=y
+CONFIG_CC_OPTIMIZE_FOR_PERFORMANCE=y
+CONFIG_EMBEDDED=n
+CONFIG_BASE_FULL=y
+CONFIG_FUTEX=y
+CONFIG_SHMEM=y
+CONFIG_SLUB=y
+CONFIG_SPARSEMEM_VMEMMAP=y
+CONFIG_SMP=y
+CONFIG_SCHED_SMT=y
+CONFIG_SCHED_MC=y
+CONFIG_NUMA=y
+CONFIG_PREEMPT=y
+CONFIG_NO_HZ=y
+CONFIG_NO_HZ_IDLE=y
+CONFIG_NO_HZ_FULL=n
+CONFIG_HZ_PERIODIC=n
+CONFIG_HIGH_RES_TIMERS=y
+CONFIG_ARCH_RANDOM=y
+CONFIG_FILE_LOCKING=y
+CONFIG_POSIX_TIMERS=y
+CONFIG_DEVTMPFS=y
+CONFIG_PROC_FS=y
+CONFIG_PROC_SYSCTL=y
+CONFIG_SYSFS=y
+CONFIG_TMPFS=y
+CONFIG_CONSOLE_LOGLEVEL_DEFAULT=15
+CONFIG_PRINTK_TIME=y
+CONFIG_BLK_DEV_INITRD=y
+CONFIG_LEGACY_VSYSCALL_NONE=y
+CONFIG_KERNEL_GZIP=y
+CONFIG_PANIC_ON_OOPS=y
+CONFIG_BUG_ON_DATA_CORRUPTION=y
+CONFIG_LOCKUP_DETECTOR=y
+CONFIG_SOFTLOCKUP_DETECTOR=y
+CONFIG_HARDLOCKUP_DETECTOR=y
+CONFIG_WQ_WATCHDOG=y
+CONFIG_DETECT_HUNG_TASK=y
+CONFIG_BOOTPARAM_HARDLOCKUP_PANIC=y
+CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC=y
+CONFIG_BOOTPARAM_HUNG_TASK_PANIC=y
+CONFIG_PANIC_TIMEOUT=-1
+CONFIG_STACKTRACE=y
+CONFIG_EARLY_PRINTK=y
+CONFIG_GDB_SCRIPTS=y
+CONFIG_WIREGUARD=y
+CONFIG_WIREGUARD_DEBUG=y
-- 
cgit v1.2.3-59-g8ed1b


From d7c68a38bb4f9b7c1a2e4a772872c752ee5c44a6 Mon Sep 17 00:00:00 2001
From: "Jason A. Donenfeld" <Jason@zx2c4.com>
Date: Sun, 15 Dec 2019 22:08:01 +0100
Subject: wireguard: Kconfig: select parent dependency for crypto

This fixes the crypto selection submenu depenencies. Otherwise, we'd
wind up issuing warnings in which certain dependencies we also select
couldn't be satisfied. This condition was triggered by the addition of
the test suite autobuilder in the previous commit.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/Kconfig | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index ffe8d4e2b206..01e2657e4c26 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -85,6 +85,8 @@ config WIREGUARD
 	select CRYPTO_POLY1305_X86_64 if X86 && 64BIT
 	select CRYPTO_BLAKE2S_X86 if X86 && 64BIT
 	select CRYPTO_CURVE25519_X86 if X86 && 64BIT
+	select ARM_CRYPTO if ARM
+	select ARM64_CRYPTO if ARM64
 	select CRYPTO_CHACHA20_NEON if (ARM || ARM64) && KERNEL_MODE_NEON
 	select CRYPTO_POLY1305_NEON if ARM64 && KERNEL_MODE_NEON
 	select CRYPTO_POLY1305_ARM if ARM
-- 
cgit v1.2.3-59-g8ed1b


From a2ec8b5706944d228181c8b91d815f41d6dd8e7b Mon Sep 17 00:00:00 2001
From: Josh Soref <jsoref@gmail.com>
Date: Sun, 15 Dec 2019 22:08:02 +0100
Subject: wireguard: global: fix spelling mistakes in comments

This fixes two spelling errors in source code comments.

Signed-off-by: Josh Soref <jsoref@gmail.com>
[Jason: rewrote commit message]
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/wireguard/receive.c | 2 +-
 include/uapi/linux/wireguard.h  | 8 ++++----
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireguard/receive.c b/drivers/net/wireguard/receive.c
index 7e675f541491..9c6bab9c981f 100644
--- a/drivers/net/wireguard/receive.c
+++ b/drivers/net/wireguard/receive.c
@@ -380,7 +380,7 @@ static void wg_packet_consume_data_done(struct wg_peer *peer,
 	/* We've already verified the Poly1305 auth tag, which means this packet
 	 * was not modified in transit. We can therefore tell the networking
 	 * stack that all checksums of every layer of encapsulation have already
-	 * been checked "by the hardware" and therefore is unneccessary to check
+	 * been checked "by the hardware" and therefore is unnecessary to check
 	 * again in software.
 	 */
 	skb->ip_summed = CHECKSUM_UNNECESSARY;
diff --git a/include/uapi/linux/wireguard.h b/include/uapi/linux/wireguard.h
index dd8a47c4ad11..ae88be14c947 100644
--- a/include/uapi/linux/wireguard.h
+++ b/include/uapi/linux/wireguard.h
@@ -18,13 +18,13 @@
  * one but not both of:
  *
  *    WGDEVICE_A_IFINDEX: NLA_U32
- *    WGDEVICE_A_IFNAME: NLA_NUL_STRING, maxlen IFNAMESIZ - 1
+ *    WGDEVICE_A_IFNAME: NLA_NUL_STRING, maxlen IFNAMSIZ - 1
  *
  * The kernel will then return several messages (NLM_F_MULTI) containing the
  * following tree of nested items:
  *
  *    WGDEVICE_A_IFINDEX: NLA_U32
- *    WGDEVICE_A_IFNAME: NLA_NUL_STRING, maxlen IFNAMESIZ - 1
+ *    WGDEVICE_A_IFNAME: NLA_NUL_STRING, maxlen IFNAMSIZ - 1
  *    WGDEVICE_A_PRIVATE_KEY: NLA_EXACT_LEN, len WG_KEY_LEN
  *    WGDEVICE_A_PUBLIC_KEY: NLA_EXACT_LEN, len WG_KEY_LEN
  *    WGDEVICE_A_LISTEN_PORT: NLA_U16
@@ -77,7 +77,7 @@
  * WGDEVICE_A_IFINDEX and WGDEVICE_A_IFNAME:
  *
  *    WGDEVICE_A_IFINDEX: NLA_U32
- *    WGDEVICE_A_IFNAME: NLA_NUL_STRING, maxlen IFNAMESIZ - 1
+ *    WGDEVICE_A_IFNAME: NLA_NUL_STRING, maxlen IFNAMSIZ - 1
  *    WGDEVICE_A_FLAGS: NLA_U32, 0 or WGDEVICE_F_REPLACE_PEERS if all current
  *                      peers should be removed prior to adding the list below.
  *    WGDEVICE_A_PRIVATE_KEY: len WG_KEY_LEN, all zeros to remove
@@ -121,7 +121,7 @@
  * filling in information not contained in the prior. Note that if
  * WGDEVICE_F_REPLACE_PEERS is specified in the first message, it probably
  * should not be specified in fragments that come after, so that the list
- * of peers is only cleared the first time but appened after. Likewise for
+ * of peers is only cleared the first time but appended after. Likewise for
  * peers, if WGPEER_F_REPLACE_ALLOWEDIPS is specified in the first message
  * of a peer, it likely should not be specified in subsequent fragments.
  *
-- 
cgit v1.2.3-59-g8ed1b


From 43967b6ff91e53bcce5ae08c16a0588a475b53a1 Mon Sep 17 00:00:00 2001
From: YueHaibing <yuehaibing@huawei.com>
Date: Sun, 15 Dec 2019 22:08:03 +0100
Subject: wireguard: main: remove unused include <linux/version.h>

Remove <linux/version.h> from the includes for main.c, which is unused.

Signed-off-by: YueHaibing <yuehaibing@huawei.com>
[Jason: reworded commit message]
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/wireguard/main.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/wireguard/main.c b/drivers/net/wireguard/main.c
index 10c0a40f6a9e..7a7d5f1a80fc 100644
--- a/drivers/net/wireguard/main.c
+++ b/drivers/net/wireguard/main.c
@@ -12,7 +12,6 @@
 
 #include <uapi/linux/wireguard.h>
 
-#include <linux/version.h>
 #include <linux/init.h>
 #include <linux/module.h>
 #include <linux/genetlink.h>
-- 
cgit v1.2.3-59-g8ed1b


From d89ee7d5c73af15c1c6f12b016cdf469742b5726 Mon Sep 17 00:00:00 2001
From: Wei Yongjun <weiyongjun1@huawei.com>
Date: Sun, 15 Dec 2019 22:08:04 +0100
Subject: wireguard: allowedips: use kfree_rcu() instead of call_rcu()

The callback function of call_rcu() just calls a kfree(), so we
can use kfree_rcu() instead of call_rcu() + callback function.

Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/wireguard/allowedips.c | 7 +------
 1 file changed, 1 insertion(+), 6 deletions(-)

diff --git a/drivers/net/wireguard/allowedips.c b/drivers/net/wireguard/allowedips.c
index 72667d5399c3..121d9ea0f135 100644
--- a/drivers/net/wireguard/allowedips.c
+++ b/drivers/net/wireguard/allowedips.c
@@ -31,11 +31,6 @@ static void copy_and_assign_cidr(struct allowedips_node *node, const u8 *src,
 #define CHOOSE_NODE(parent, key) \
 	parent->bit[(key[parent->bit_at_a] >> parent->bit_at_b) & 1]
 
-static void node_free_rcu(struct rcu_head *rcu)
-{
-	kfree(container_of(rcu, struct allowedips_node, rcu));
-}
-
 static void push_rcu(struct allowedips_node **stack,
 		     struct allowedips_node __rcu *p, unsigned int *len)
 {
@@ -112,7 +107,7 @@ static void walk_remove_by_peer(struct allowedips_node __rcu **top,
 				if (!node->bit[0] || !node->bit[1]) {
 					rcu_assign_pointer(*nptr, DEREF(
 					       &node->bit[!REF(node->bit[0])]));
-					call_rcu(&node->rcu, node_free_rcu);
+					kfree_rcu(node, rcu);
 					node = DEREF(nptr);
 				}
 			}
-- 
cgit v1.2.3-59-g8ed1b


From 13591a1c3899cfe6891ae0df46637097fdde36a4 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <seckelmann@datto.com>
Date: Tue, 10 Dec 2019 12:07:32 +0200
Subject: ath11k: register HE mesh capabilities

The capabilities for the HE mesh are generated from the capabilities
reported by the fw. But the firmware only reports the overall capabilities
and not the one which are specific for mesh. Some of them (TWT, MU UL/DL,
TB PPDU, ...) require an infrastructure setup with a main STA (AP)
controlling the operations. This is not the case for mesh and thus these
capabilities are removed from the list of capabilities.

Signed-off-by: Sven Eckelmann <seckelmann@datto.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/mac.c | 75 +++++++++++++++++++++++++++++++++++
 1 file changed, 75 insertions(+)

diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index 978d8768d68a..65f6a2585577 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -3346,6 +3346,77 @@ static void ath11k_gen_ppe_thresh(struct ath11k_ppe_threshold *fw_ppet,
 	}
 }
 
+static void
+ath11k_mac_filter_he_cap_mesh(struct ieee80211_he_cap_elem *he_cap_elem)
+{
+	u8 m;
+
+	m = IEEE80211_HE_MAC_CAP0_TWT_RES |
+	    IEEE80211_HE_MAC_CAP0_TWT_REQ;
+	he_cap_elem->mac_cap_info[0] &= ~m;
+
+	m = IEEE80211_HE_MAC_CAP2_TRS |
+	    IEEE80211_HE_MAC_CAP2_BCAST_TWT |
+	    IEEE80211_HE_MAC_CAP2_MU_CASCADING;
+	he_cap_elem->mac_cap_info[2] &= ~m;
+
+	m = IEEE80211_HE_MAC_CAP3_FLEX_TWT_SCHED |
+	    IEEE80211_HE_MAC_CAP2_BCAST_TWT |
+	    IEEE80211_HE_MAC_CAP2_MU_CASCADING;
+	he_cap_elem->mac_cap_info[3] &= ~m;
+
+	m = IEEE80211_HE_MAC_CAP4_BSRP_BQRP_A_MPDU_AGG |
+	    IEEE80211_HE_MAC_CAP4_BQR;
+	he_cap_elem->mac_cap_info[4] &= ~m;
+
+	m = IEEE80211_HE_MAC_CAP5_SUBCHAN_SELECVITE_TRANSMISSION |
+	    IEEE80211_HE_MAC_CAP5_UL_2x996_TONE_RU |
+	    IEEE80211_HE_MAC_CAP5_PUNCTURED_SOUNDING |
+	    IEEE80211_HE_MAC_CAP5_HT_VHT_TRIG_FRAME_RX;
+	he_cap_elem->mac_cap_info[5] &= ~m;
+
+	m = IEEE80211_HE_PHY_CAP2_UL_MU_FULL_MU_MIMO |
+	    IEEE80211_HE_PHY_CAP2_UL_MU_PARTIAL_MU_MIMO;
+	he_cap_elem->phy_cap_info[2] &= ~m;
+
+	m = IEEE80211_HE_PHY_CAP3_RX_HE_MU_PPDU_FROM_NON_AP_STA |
+	    IEEE80211_HE_PHY_CAP3_DCM_MAX_CONST_TX_MASK |
+	    IEEE80211_HE_PHY_CAP3_DCM_MAX_CONST_RX_MASK;
+	he_cap_elem->phy_cap_info[3] &= ~m;
+
+	m = IEEE80211_HE_PHY_CAP4_MU_BEAMFORMER;
+	he_cap_elem->phy_cap_info[4] &= ~m;
+
+	m = IEEE80211_HE_PHY_CAP5_NG16_MU_FEEDBACK;
+	he_cap_elem->phy_cap_info[5] &= ~m;
+
+	m = IEEE80211_HE_PHY_CAP6_CODEBOOK_SIZE_75_MU |
+	    IEEE80211_HE_PHY_CAP6_TRIG_MU_BEAMFORMER_FB |
+	    IEEE80211_HE_PHY_CAP6_TRIG_CQI_FB |
+	    IEEE80211_HE_PHY_CAP6_PARTIAL_BANDWIDTH_DL_MUMIMO;
+	he_cap_elem->phy_cap_info[6] &= ~m;
+
+	m = IEEE80211_HE_PHY_CAP7_SRP_BASED_SR |
+	    IEEE80211_HE_PHY_CAP7_POWER_BOOST_FACTOR_AR |
+	    IEEE80211_HE_PHY_CAP7_STBC_TX_ABOVE_80MHZ |
+	    IEEE80211_HE_PHY_CAP7_STBC_RX_ABOVE_80MHZ;
+	he_cap_elem->phy_cap_info[7] &= ~m;
+
+	m = IEEE80211_HE_PHY_CAP8_HE_ER_SU_PPDU_4XLTF_AND_08_US_GI |
+	    IEEE80211_HE_PHY_CAP8_20MHZ_IN_40MHZ_HE_PPDU_IN_2G |
+	    IEEE80211_HE_PHY_CAP8_20MHZ_IN_160MHZ_HE_PPDU |
+	    IEEE80211_HE_PHY_CAP8_80MHZ_IN_160MHZ_HE_PPDU;
+	he_cap_elem->phy_cap_info[8] &= ~m;
+
+	m = IEEE80211_HE_PHY_CAP9_LONGER_THAN_16_SIGB_OFDM_SYM |
+	    IEEE80211_HE_PHY_CAP9_NON_TRIGGERED_CQI_FEEDBACK |
+	    IEEE80211_HE_PHY_CAP9_RX_1024_QAM_LESS_THAN_242_TONE_RU |
+	    IEEE80211_HE_PHY_CAP9_TX_1024_QAM_LESS_THAN_242_TONE_RU |
+	    IEEE80211_HE_PHY_CAP9_RX_FULL_BW_SU_USING_MU_WITH_COMP_SIGB |
+	    IEEE80211_HE_PHY_CAP9_RX_FULL_BW_SU_USING_MU_WITH_NON_COMP_SIGB;
+	he_cap_elem->phy_cap_info[9] &= ~m;
+}
+
 static int ath11k_mac_copy_he_cap(struct ath11k *ar,
 				  struct ath11k_pdev_cap *cap,
 				  struct ieee80211_sband_iftype_data *data,
@@ -3362,6 +3433,7 @@ static int ath11k_mac_copy_he_cap(struct ath11k *ar,
 		switch (i) {
 		case NL80211_IFTYPE_STATION:
 		case NL80211_IFTYPE_AP:
+		case NL80211_IFTYPE_MESH_POINT:
 			break;
 
 		default:
@@ -3402,6 +3474,9 @@ static int ath11k_mac_copy_he_cap(struct ath11k *ar,
 			he_cap_elem->phy_cap_info[9] |=
 				IEEE80211_HE_PHY_CAP9_TX_1024_QAM_LESS_THAN_242_TONE_RU;
 			break;
+		case NL80211_IFTYPE_MESH_POINT:
+			ath11k_mac_filter_he_cap_mesh(he_cap_elem);
+			break;
 		}
 
 		he_cap->he_mcs_nss_supp.rx_mcs_80 =
-- 
cgit v1.2.3-59-g8ed1b


From a67bcec3569f12da7c99891fa68f227791630fa1 Mon Sep 17 00:00:00 2001
From: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Date: Tue, 10 Dec 2019 12:08:00 +0200
Subject: ath10k: Fix some typo in some warning messages

Fix some typo:
  s/to to/to/
  s/even/event/

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/mac.c      | 2 +-
 drivers/net/wireless/ath/ath10k/testmode.c | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index 767c7bf16975..0969e9bf5d5e 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -1098,7 +1098,7 @@ static int ath10k_monitor_vdev_stop(struct ath10k *ar)
 
 	ret = ath10k_wmi_vdev_stop(ar, ar->monitor_vdev_id);
 	if (ret)
-		ath10k_warn(ar, "failed to to request monitor vdev %i stop: %d\n",
+		ath10k_warn(ar, "failed to request monitor vdev %i stop: %d\n",
 			    ar->monitor_vdev_id, ret);
 
 	ret = ath10k_vdev_setup_sync(ar);
diff --git a/drivers/net/wireless/ath/ath10k/testmode.c b/drivers/net/wireless/ath/ath10k/testmode.c
index 1bffe3fbea3f..7a9b9bbcdbfc 100644
--- a/drivers/net/wireless/ath/ath10k/testmode.c
+++ b/drivers/net/wireless/ath/ath10k/testmode.c
@@ -65,7 +65,7 @@ bool ath10k_tm_event_wmi(struct ath10k *ar, u32 cmd_id, struct sk_buff *skb)
 	ret = nla_put_u32(nl_skb, ATH10K_TM_ATTR_CMD, ATH10K_TM_CMD_WMI);
 	if (ret) {
 		ath10k_warn(ar,
-			    "failed to to put testmode wmi event cmd attribute: %d\n",
+			    "failed to put testmode wmi event cmd attribute: %d\n",
 			    ret);
 		kfree_skb(nl_skb);
 		goto out;
@@ -74,7 +74,7 @@ bool ath10k_tm_event_wmi(struct ath10k *ar, u32 cmd_id, struct sk_buff *skb)
 	ret = nla_put_u32(nl_skb, ATH10K_TM_ATTR_WMI_CMDID, cmd_id);
 	if (ret) {
 		ath10k_warn(ar,
-			    "failed to to put testmode wmi even cmd_id: %d\n",
+			    "failed to put testmode wmi event cmd_id: %d\n",
 			    ret);
 		kfree_skb(nl_skb);
 		goto out;
-- 
cgit v1.2.3-59-g8ed1b


From 5d57c327584ae0f88eba61ec8faa9c621f40cffd Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Fri, 13 Dec 2019 18:22:02 +0000
Subject: net: phylink: improve clause 45 PHY ksettings_set implementation

While testing ethtool with the Methode DM7052 module, it was noticed
that attempting to set the advertising mask results in the mask being
truncated to the support offered by the currently chosen PHY interface
mode.

When a PHY dynamically changes the PHY interface mode, limiting the
advertising mask in this way is not correct - if the PHY happened to
negotiate 10GBASE-T, and selected 10GBASE-R as the host interface, we
don't want to restrict the advertisement to just 10GBASE-* modes.

Rework setting the advertisement to take account of this; do not pass
the requested advertisement through phylink_validate(), but rely on
the advertisement restriction (supported mask) set when the PHY was
initially setup.

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/phylink.c | 84 ++++++++++++++++++++++++++++++-----------------
 1 file changed, 53 insertions(+), 31 deletions(-)

diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index 1e0e32c466ee..8f3dc27db95d 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -1230,44 +1230,66 @@ int phylink_ethtool_ksettings_set(struct phylink *pl,
 		__set_bit(ETHTOOL_LINK_MODE_Autoneg_BIT, config.advertising);
 	}
 
-	if (phylink_validate(pl, support, &config))
-		return -EINVAL;
-
-	/* If autonegotiation is enabled, we must have an advertisement */
-	if (config.an_enabled && phylink_is_empty_linkmode(config.advertising))
-		return -EINVAL;
-
-	our_kset = *kset;
-	linkmode_copy(our_kset.link_modes.advertising, config.advertising);
-	our_kset.base.speed = config.speed;
-	our_kset.base.duplex = config.duplex;
-
-	/* If we have a PHY, configure the phy */
 	if (pl->phydev) {
+		/* If we have a PHY, we process the kset change via phylib.
+		 * phylib will call our link state function if the PHY
+		 * parameters have changed, which will trigger a resolve
+		 * and update the MAC configuration.
+		 */
+		our_kset = *kset;
+		linkmode_copy(our_kset.link_modes.advertising,
+			      config.advertising);
+		our_kset.base.speed = config.speed;
+		our_kset.base.duplex = config.duplex;
+
 		ret = phy_ethtool_ksettings_set(pl->phydev, &our_kset);
 		if (ret)
 			return ret;
-	}
 
-	mutex_lock(&pl->state_mutex);
-	/* Configure the MAC to match the new settings */
-	linkmode_copy(pl->link_config.advertising, our_kset.link_modes.advertising);
-	pl->link_config.interface = config.interface;
-	pl->link_config.speed = our_kset.base.speed;
-	pl->link_config.duplex = our_kset.base.duplex;
-	pl->link_config.an_enabled = our_kset.base.autoneg != AUTONEG_DISABLE;
+		mutex_lock(&pl->state_mutex);
+		/* Save the new configuration */
+		linkmode_copy(pl->link_config.advertising,
+			      our_kset.link_modes.advertising);
+		pl->link_config.interface = config.interface;
+		pl->link_config.speed = our_kset.base.speed;
+		pl->link_config.duplex = our_kset.base.duplex;
+		pl->link_config.an_enabled = our_kset.base.autoneg !=
+					     AUTONEG_DISABLE;
+		mutex_unlock(&pl->state_mutex);
+	} else {
+		/* For a fixed link, this isn't able to change any parameters,
+		 * which just leaves inband mode.
+		 */
+		if (phylink_validate(pl, support, &config))
+			return -EINVAL;
 
-	/* If we have a PHY, phylib will call our link state function if the
-	 * mode has changed, which will trigger a resolve and update the MAC
-	 * configuration. For a fixed link, this isn't able to change any
-	 * parameters, which just leaves inband mode.
-	 */
-	if (pl->cur_link_an_mode == MLO_AN_INBAND &&
-	    !test_bit(PHYLINK_DISABLE_STOPPED, &pl->phylink_disable_state)) {
-		phylink_mac_config(pl, &pl->link_config);
-		phylink_mac_an_restart(pl);
+		/* If autonegotiation is enabled, we must have an advertisement */
+		if (config.an_enabled &&
+		    phylink_is_empty_linkmode(config.advertising))
+			return -EINVAL;
+
+		mutex_lock(&pl->state_mutex);
+		linkmode_copy(pl->link_config.advertising, config.advertising);
+		pl->link_config.interface = config.interface;
+		pl->link_config.speed = config.speed;
+		pl->link_config.duplex = config.duplex;
+		pl->link_config.an_enabled = kset->base.autoneg !=
+					     AUTONEG_DISABLE;
+
+		if (pl->cur_link_an_mode == MLO_AN_INBAND &&
+		    !test_bit(PHYLINK_DISABLE_STOPPED,
+			      &pl->phylink_disable_state)) {
+			/* If in 802.3z mode, this updates the advertisement.
+			 *
+			 * If we are in SGMII mode without a PHY, there is no
+			 * advertisement; the only thing we have is the pause
+			 * modes which can only come from a PHY.
+			 */
+			phylink_mac_config(pl, &pl->link_config);
+			phylink_mac_an_restart(pl);
+		}
+		mutex_unlock(&pl->state_mutex);
 	}
-	mutex_unlock(&pl->state_mutex);
 
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From df3f57ac960590f4e87ec71549efd49bbfdb25d0 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Fri, 13 Dec 2019 18:22:07 +0000
Subject: net: phylink: extend clause 45 PHY validation workaround

Commit e45d1f5288b8 ("net: phylink: support Clause 45 PHYs on SFP+
modules") added a workaround to support clause 45 PHYs which
dynamically switch their interface mode on SFP+ modules.  This was
implemented by validating the PHYs supported/advertising using
PHY_INTERFACE_MODE_NA, rather than the specific interface mode that
we attached the PHY with.

However, we already have a situation where phylink is used to connect
a Marvell 88X3310 PHY which also behaves in exactly the same way, but
which seemingly doesn't need this.  The reason seems to be that the
mvpp2 driver sets a whole bunch of link modes for
PHY_INTERFACE_MODE_10GKR down to 10Mb/s, despite 10GBASE-R not actually
supporting anything but 10Gb/s speeds.

When testing with drivers that (correctly) take the mvneta approach,
where the validate() method only returns what can be supported /
advertised for the specified link mode, we find that Clause 45 PHYs do
not behave as we expect: their advertisement is restricted to what
the current link will support, rather than what the PHY supports
through its dynamic switching.

Extend this workaround to all such cases; if we have a Clause 45 PHY
attaching via any means, except in USXGMII, XAUI and RXAUI which are
all unable to support this dynamic switching or have other solutions
to it, then we need to validate using PHY_INTERFACE_MODE_NA.

This should allow mvpp2 to switch to a more conformant validate()
implementation.

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/phylink.c | 22 +++++++++++++---------
 1 file changed, 13 insertions(+), 9 deletions(-)

diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index 8f3dc27db95d..a917f95372cd 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -736,7 +736,19 @@ static int phylink_bringup_phy(struct phylink *pl, struct phy_device *phy,
 	memset(&config, 0, sizeof(config));
 	linkmode_copy(supported, phy->supported);
 	linkmode_copy(config.advertising, phy->advertising);
-	config.interface = interface;
+
+	/* Clause 45 PHYs switch their Serdes lane between several different
+	 * modes, normally 10GBASE-R, SGMII. Some use 2500BASE-X for 2.5G
+	 * speeds. We really need to know which interface modes the PHY and
+	 * MAC supports to properly work out which linkmodes can be supported.
+	 */
+	if (phy->is_c45 &&
+	    interface != PHY_INTERFACE_MODE_RXAUI &&
+	    interface != PHY_INTERFACE_MODE_XAUI &&
+	    interface != PHY_INTERFACE_MODE_USXGMII)
+		config.interface = PHY_INTERFACE_MODE_NA;
+	else
+		config.interface = interface;
 
 	ret = phylink_validate(pl, supported, &config);
 	if (ret)
@@ -1902,14 +1914,6 @@ static int phylink_sfp_connect_phy(void *upstream, struct phy_device *phy)
 	if (ret < 0)
 		return ret;
 
-	/* Clause 45 PHYs switch their Serdes lane between several different
-	 * modes, normally 10GBASE-R, SGMII. Some use 2500BASE-X for 2.5G
-	 * speeds.  We really need to know which interface modes the PHY and
-	 * MAC supports to properly work out which linkmodes can be supported.
-	 */
-	if (phy->is_c45)
-		interface = PHY_INTERFACE_MODE_NA;
-
 	ret = phylink_bringup_phy(pl, phy, interface);
 	if (ret)
 		phy_detach(phy);
-- 
cgit v1.2.3-59-g8ed1b


From ef8e0b808b908c49670ef00b95c7848e233ca3bd Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Fri, 13 Dec 2019 18:22:12 +0000
Subject: net: mvpp2: update mvpp2 validate() implementation

Fix up the mvpp2 validate implementation to adopt the same behaviour as
mvneta:
- only allow the link modes that the specified PHY interface mode
  supports with the exception of 1000base-X and 2500base-X.
- use the basex helper to deal with SFP modules that can be switched
  between 1000base-X vs 2500base-X.

This gives consistent behaviour between mvneta and mvpp2.

This commit depends on "net: phylink: extend clause 45 PHY validation
workaround" so is not marked for backporting to stable kernels.

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Acked-by: Antoine Tenart <antoine.tenart@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c | 22 ++++++++++++++++++----
 1 file changed, 18 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
index 62dc2f362a16..3d9eb5747d9e 100644
--- a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
+++ b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
@@ -4792,6 +4792,8 @@ static void mvpp2_phylink_validate(struct phylink_config *config,
 			phylink_set(mask, 10000baseER_Full);
 			phylink_set(mask, 10000baseKR_Full);
 		}
+		if (state->interface != PHY_INTERFACE_MODE_NA)
+			break;
 		/* Fall-through */
 	case PHY_INTERFACE_MODE_RGMII:
 	case PHY_INTERFACE_MODE_RGMII_ID:
@@ -4802,13 +4804,23 @@ static void mvpp2_phylink_validate(struct phylink_config *config,
 		phylink_set(mask, 10baseT_Full);
 		phylink_set(mask, 100baseT_Half);
 		phylink_set(mask, 100baseT_Full);
+		phylink_set(mask, 1000baseT_Full);
+		phylink_set(mask, 1000baseX_Full);
+		if (state->interface != PHY_INTERFACE_MODE_NA)
+			break;
 		/* Fall-through */
 	case PHY_INTERFACE_MODE_1000BASEX:
 	case PHY_INTERFACE_MODE_2500BASEX:
-		phylink_set(mask, 1000baseT_Full);
-		phylink_set(mask, 1000baseX_Full);
-		phylink_set(mask, 2500baseT_Full);
-		phylink_set(mask, 2500baseX_Full);
+		if (port->comphy ||
+		    state->interface != PHY_INTERFACE_MODE_2500BASEX) {
+			phylink_set(mask, 1000baseT_Full);
+			phylink_set(mask, 1000baseX_Full);
+		}
+		if (port->comphy ||
+		    state->interface == PHY_INTERFACE_MODE_2500BASEX) {
+			phylink_set(mask, 2500baseT_Full);
+			phylink_set(mask, 2500baseX_Full);
+		}
 		break;
 	default:
 		goto empty_set;
@@ -4817,6 +4829,8 @@ static void mvpp2_phylink_validate(struct phylink_config *config,
 	bitmap_and(supported, supported, mask, __ETHTOOL_LINK_MODE_MASK_NBITS);
 	bitmap_and(state->advertising, state->advertising, mask,
 		   __ETHTOOL_LINK_MODE_MASK_NBITS);
+
+	phylink_helper_basex_speed(state);
 	return;
 
 empty_set:
-- 
cgit v1.2.3-59-g8ed1b


From 71a55a2315b047352b3d65e2d24724207be85ae2 Mon Sep 17 00:00:00 2001
From: Biao Huang <biao.huang@mediatek.com>
Date: Mon, 16 Dec 2019 13:39:57 +0800
Subject: net-next: stmmac: mediatek: add more support for RMII

MT2712 SoC can provide the rmii reference clock, and the clock
will output from TXC pin only, which means ref_clk pin of external
PHY should connect to TXC pin in this case.
Add corresponding clock and timing settings.

Signed-off-by: Biao Huang <biao.huang@mediatek.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../net/ethernet/stmicro/stmmac/dwmac-mediatek.c   | 89 +++++++++++++++-------
 1 file changed, 60 insertions(+), 29 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-mediatek.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-mediatek.c
index bdb80421acac..9e4b83832938 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-mediatek.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-mediatek.c
@@ -55,6 +55,8 @@ struct mediatek_dwmac_plat_data {
 	struct regmap *peri_regmap;
 	struct device *dev;
 	phy_interface_t phy_mode;
+	int num_clks_to_config;
+	bool rmii_clk_from_mac;
 	bool rmii_rxc;
 };
 
@@ -73,21 +75,33 @@ struct mediatek_dwmac_variant {
 
 /* list of clocks required for mac */
 static const char * const mt2712_dwmac_clk_l[] = {
-	"axi", "apb", "mac_main", "ptp_ref"
+	"axi", "apb", "mac_main", "ptp_ref", "rmii_internal"
 };
 
 static int mt2712_set_interface(struct mediatek_dwmac_plat_data *plat)
 {
+	int rmii_clk_from_mac = plat->rmii_clk_from_mac ? RMII_CLK_SRC_INTERNAL : 0;
 	int rmii_rxc = plat->rmii_rxc ? RMII_CLK_SRC_RXC : 0;
 	u32 intf_val = 0;
 
+	/* The clock labeled as "rmii_internal" in mt2712_dwmac_clk_l is needed
+	 * only in RMII(when MAC provides the reference clock), and useless for
+	 * RGMII/MII/RMII(when PHY provides the reference clock).
+	 * num_clks_to_config indicates the real number of clocks should be
+	 * configured, equals to (plat->variant->num_clks - 1) in default for all the case,
+	 * then +1 for rmii_clk_from_mac case.
+	 */
+	plat->num_clks_to_config = plat->variant->num_clks - 1;
+
 	/* select phy interface in top control domain */
 	switch (plat->phy_mode) {
 	case PHY_INTERFACE_MODE_MII:
 		intf_val |= PHY_INTF_MII;
 		break;
 	case PHY_INTERFACE_MODE_RMII:
-		intf_val |= (PHY_INTF_RMII | rmii_rxc);
+		if (plat->rmii_clk_from_mac)
+			plat->num_clks_to_config++;
+		intf_val |= (PHY_INTF_RMII | rmii_rxc | rmii_clk_from_mac);
 		break;
 	case PHY_INTERFACE_MODE_RGMII:
 	case PHY_INTERFACE_MODE_RGMII_TXID:
@@ -173,35 +187,50 @@ static int mt2712_set_delay(struct mediatek_dwmac_plat_data *plat)
 		delay_val |= FIELD_PREP(ETH_DLY_RXC_INV, mac_delay->rx_inv);
 		break;
 	case PHY_INTERFACE_MODE_RMII:
-		/* the rmii reference clock is from external phy,
-		 * and the property "rmii_rxc" indicates which pin(TXC/RXC)
-		 * the reference clk is connected to. The reference clock is a
-		 * received signal, so rx_delay/rx_inv are used to indicate
-		 * the reference clock timing adjustment
-		 */
-		if (plat->rmii_rxc) {
-			/* the rmii reference clock from outside is connected
-			 * to RXC pin, the reference clock will be adjusted
-			 * by RXC delay macro circuit.
-			 */
-			delay_val |= FIELD_PREP(ETH_DLY_RXC_ENABLE, !!mac_delay->rx_delay);
-			delay_val |= FIELD_PREP(ETH_DLY_RXC_STAGES, mac_delay->rx_delay);
-			delay_val |= FIELD_PREP(ETH_DLY_RXC_INV, mac_delay->rx_inv);
-		} else {
-			/* the rmii reference clock from outside is connected
-			 * to TXC pin, the reference clock will be adjusted
-			 * by TXC delay macro circuit.
+		if (plat->rmii_clk_from_mac) {
+			/* case 1: mac provides the rmii reference clock,
+			 * and the clock output to TXC pin.
+			 * The egress timing can be adjusted by GTXC delay macro circuit.
+			 * The ingress timing can be adjusted by TXC delay macro circuit.
 			 */
 			delay_val |= FIELD_PREP(ETH_DLY_TXC_ENABLE, !!mac_delay->rx_delay);
 			delay_val |= FIELD_PREP(ETH_DLY_TXC_STAGES, mac_delay->rx_delay);
 			delay_val |= FIELD_PREP(ETH_DLY_TXC_INV, mac_delay->rx_inv);
+
+			delay_val |= FIELD_PREP(ETH_DLY_GTXC_ENABLE, !!mac_delay->tx_delay);
+			delay_val |= FIELD_PREP(ETH_DLY_GTXC_STAGES, mac_delay->tx_delay);
+			delay_val |= FIELD_PREP(ETH_DLY_GTXC_INV, mac_delay->tx_inv);
+		} else {
+			/* case 2: the rmii reference clock is from external phy,
+			 * and the property "rmii_rxc" indicates which pin(TXC/RXC)
+			 * the reference clk is connected to. The reference clock is a
+			 * received signal, so rx_delay/rx_inv are used to indicate
+			 * the reference clock timing adjustment
+			 */
+			if (plat->rmii_rxc) {
+				/* the rmii reference clock from outside is connected
+				 * to RXC pin, the reference clock will be adjusted
+				 * by RXC delay macro circuit.
+				 */
+				delay_val |= FIELD_PREP(ETH_DLY_RXC_ENABLE, !!mac_delay->rx_delay);
+				delay_val |= FIELD_PREP(ETH_DLY_RXC_STAGES, mac_delay->rx_delay);
+				delay_val |= FIELD_PREP(ETH_DLY_RXC_INV, mac_delay->rx_inv);
+			} else {
+				/* the rmii reference clock from outside is connected
+				 * to TXC pin, the reference clock will be adjusted
+				 * by TXC delay macro circuit.
+				 */
+				delay_val |= FIELD_PREP(ETH_DLY_TXC_ENABLE, !!mac_delay->rx_delay);
+				delay_val |= FIELD_PREP(ETH_DLY_TXC_STAGES, mac_delay->rx_delay);
+				delay_val |= FIELD_PREP(ETH_DLY_TXC_INV, mac_delay->rx_inv);
+			}
+			/* tx_inv will inverse the tx clock inside mac relateive to
+			 * reference clock from external phy,
+			 * and this bit is located in the same register with fine-tune
+			 */
+			if (mac_delay->tx_inv)
+				fine_val = ETH_RMII_DLY_TX_INV;
 		}
-		/* tx_inv will inverse the tx clock inside mac relateive to
-		 * reference clock from external phy,
-		 * and this bit is located in the same register with fine-tune
-		 */
-		if (mac_delay->tx_inv)
-			fine_val = ETH_RMII_DLY_TX_INV;
 		break;
 	case PHY_INTERFACE_MODE_RGMII:
 	case PHY_INTERFACE_MODE_RGMII_TXID:
@@ -278,6 +307,7 @@ static int mediatek_dwmac_config_dt(struct mediatek_dwmac_plat_data *plat)
 	mac_delay->tx_inv = of_property_read_bool(plat->np, "mediatek,txc-inverse");
 	mac_delay->rx_inv = of_property_read_bool(plat->np, "mediatek,rxc-inverse");
 	plat->rmii_rxc = of_property_read_bool(plat->np, "mediatek,rmii-rxc");
+	plat->rmii_clk_from_mac = of_property_read_bool(plat->np, "mediatek,rmii-clk-from-mac");
 
 	return 0;
 }
@@ -294,6 +324,8 @@ static int mediatek_dwmac_clk_init(struct mediatek_dwmac_plat_data *plat)
 	for (i = 0; i < num; i++)
 		plat->clks[i].id = variant->clk_list[i];
 
+	plat->num_clks_to_config = variant->num_clks;
+
 	return devm_clk_bulk_get(plat->dev, num, plat->clks);
 }
 
@@ -321,7 +353,7 @@ static int mediatek_dwmac_init(struct platform_device *pdev, void *priv)
 		return ret;
 	}
 
-	ret = clk_bulk_prepare_enable(variant->num_clks, plat->clks);
+	ret = clk_bulk_prepare_enable(plat->num_clks_to_config, plat->clks);
 	if (ret) {
 		dev_err(plat->dev, "failed to enable clks, err = %d\n", ret);
 		return ret;
@@ -336,9 +368,8 @@ static int mediatek_dwmac_init(struct platform_device *pdev, void *priv)
 static void mediatek_dwmac_exit(struct platform_device *pdev, void *priv)
 {
 	struct mediatek_dwmac_plat_data *plat = priv;
-	const struct mediatek_dwmac_variant *variant = plat->variant;
 
-	clk_bulk_disable_unprepare(variant->num_clks, plat->clks);
+	clk_bulk_disable_unprepare(plat->num_clks_to_config, plat->clks);
 
 	pm_runtime_put_sync(&pdev->dev);
 	pm_runtime_disable(&pdev->dev);
-- 
cgit v1.2.3-59-g8ed1b


From 882007ed783211c27e859234e78328d35ded03c7 Mon Sep 17 00:00:00 2001
From: Biao Huang <biao.huang@mediatek.com>
Date: Mon, 16 Dec 2019 13:39:58 +0800
Subject: net-next: dt-binding: dwmac-mediatek: add more description for RMII

MT2712 SoC can provide RMII reference clock,
so add corresponding description in dt-binding.

Signed-off-by: Biao Huang <biao.huang@mediatek.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../devicetree/bindings/net/mediatek-dwmac.txt     | 33 +++++++++++++++-------
 1 file changed, 23 insertions(+), 10 deletions(-)

diff --git a/Documentation/devicetree/bindings/net/mediatek-dwmac.txt b/Documentation/devicetree/bindings/net/mediatek-dwmac.txt
index 8a08621a5b54..afbcaebf062e 100644
--- a/Documentation/devicetree/bindings/net/mediatek-dwmac.txt
+++ b/Documentation/devicetree/bindings/net/mediatek-dwmac.txt
@@ -14,7 +14,7 @@ Required properties:
 	Should be "macirq" for the main MAC IRQ
 - clocks: Must contain a phandle for each entry in clock-names.
 - clock-names: The name of the clock listed in the clocks property. These are
-	"axi", "apb", "mac_main", "ptp_ref" for MT2712 SoC
+	"axi", "apb", "mac_main", "ptp_ref", "rmii_internal" for MT2712 SoC.
 - mac-address: See ethernet.txt in the same directory
 - phy-mode: See ethernet.txt in the same directory
 - mediatek,pericfg: A phandle to the syscon node that control ethernet
@@ -23,8 +23,10 @@ Required properties:
 Optional properties:
 - mediatek,tx-delay-ps: TX clock delay macro value. Default is 0.
 	It should be defined for RGMII/MII interface.
+	It should be defined for RMII interface when the reference clock is from MT2712 SoC.
 - mediatek,rx-delay-ps: RX clock delay macro value. Default is 0.
-	It should be defined for RGMII/MII/RMII interface.
+	It should be defined for RGMII/MII interface.
+	It should be defined for RMII interface.
 Both delay properties need to be a multiple of 170 for RGMII interface,
 or will round down. Range 0~31*170.
 Both delay properties need to be a multiple of 550 for MII/RMII interface,
@@ -34,13 +36,20 @@ or will round down. Range 0~31*550.
 	reference clock, which is from external PHYs, is connected to RXC pin
 	on MT2712 SoC.
 	Otherwise, is connected to TXC pin.
+- mediatek,rmii-clk-from-mac: boolean property, if present indicates that
+	MT2712 SoC provides the RMII reference clock, which outputs to TXC pin only.
 - mediatek,txc-inverse: boolean property, if present indicates that
 	1. tx clock will be inversed in MII/RGMII case,
 	2. tx clock inside MAC will be inversed relative to reference clock
 	   which is from external PHYs in RMII case, and it rarely happen.
+	3. the reference clock, which outputs to TXC pin will be inversed in RMII case
+	   when the reference clock is from MT2712 SoC.
 - mediatek,rxc-inverse: boolean property, if present indicates that
 	1. rx clock will be inversed in MII/RGMII case.
-	2. reference clock will be inversed when arrived at MAC in RMII case.
+	2. reference clock will be inversed when arrived at MAC in RMII case, when
+	   the reference clock is from external PHYs.
+	3. the inside clock, which be sent to MAC, will be inversed in RMII case when
+	   the reference clock is from MT2712 SoC.
 - assigned-clocks: mac_main and ptp_ref clocks
 - assigned-clock-parents: parent clocks of the assigned clocks
 
@@ -50,29 +59,33 @@ Example:
 		reg = <0 0x1101c000 0 0x1300>;
 		interrupts = <GIC_SPI 237 IRQ_TYPE_LEVEL_LOW>;
 		interrupt-names = "macirq";
-		phy-mode ="rgmii";
+		phy-mode ="rgmii-rxid";
 		mac-address = [00 55 7b b5 7d f7];
 		clock-names = "axi",
 			      "apb",
 			      "mac_main",
 			      "ptp_ref",
-			      "ptp_top";
+			      "rmii_internal";
 		clocks = <&pericfg CLK_PERI_GMAC>,
 			 <&pericfg CLK_PERI_GMAC_PCLK>,
 			 <&topckgen CLK_TOP_ETHER_125M_SEL>,
-			 <&topckgen CLK_TOP_ETHER_50M_SEL>;
+			 <&topckgen CLK_TOP_ETHER_50M_SEL>,
+			 <&topckgen CLK_TOP_ETHER_50M_RMII_SEL>;
 		assigned-clocks = <&topckgen CLK_TOP_ETHER_125M_SEL>,
-				  <&topckgen CLK_TOP_ETHER_50M_SEL>;
+				  <&topckgen CLK_TOP_ETHER_50M_SEL>,
+				  <&topckgen CLK_TOP_ETHER_50M_RMII_SEL>;
 		assigned-clock-parents = <&topckgen CLK_TOP_ETHERPLL_125M>,
-					 <&topckgen CLK_TOP_APLL1_D3>;
+					 <&topckgen CLK_TOP_APLL1_D3>,
+					 <&topckgen CLK_TOP_ETHERPLL_50M>;
+		power-domains = <&scpsys MT2712_POWER_DOMAIN_AUDIO>;
 		mediatek,pericfg = <&pericfg>;
 		mediatek,tx-delay-ps = <1530>;
 		mediatek,rx-delay-ps = <1530>;
 		mediatek,rmii-rxc;
 		mediatek,txc-inverse;
 		mediatek,rxc-inverse;
-		snps,txpbl = <32>;
-		snps,rxpbl = <32>;
+		snps,txpbl = <1>;
+		snps,rxpbl = <1>;
 		snps,reset-gpio = <&pio 87 GPIO_ACTIVE_LOW>;
 		snps,reset-active-low;
 	};
-- 
cgit v1.2.3-59-g8ed1b


From b59c43e03cee8537e0403be6cd4dda2cf5149455 Mon Sep 17 00:00:00 2001
From: Ajay Gupta <ajayg@nvidia.com>
Date: Sun, 15 Dec 2019 22:14:51 -0800
Subject: net: stmmac: dwc-qos: use generic device api

Use generic device api so that driver can work both with DT
or ACPI based devices.

Signed-off-by: Ajay Gupta <ajayg@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c   | 19 ++++++++++---------
 1 file changed, 10 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c
index dd9967aeda22..f87306b3cdae 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c
@@ -40,7 +40,7 @@ struct tegra_eqos {
 static int dwc_eth_dwmac_config_dt(struct platform_device *pdev,
 				   struct plat_stmmacenet_data *plat_dat)
 {
-	struct device_node *np = pdev->dev.of_node;
+	struct device *dev = &pdev->dev;
 	u32 burst_map = 0;
 	u32 bit_index = 0;
 	u32 a_index = 0;
@@ -52,9 +52,10 @@ static int dwc_eth_dwmac_config_dt(struct platform_device *pdev,
 			return -ENOMEM;
 	}
 
-	plat_dat->axi->axi_lpi_en = of_property_read_bool(np, "snps,en-lpi");
-	if (of_property_read_u32(np, "snps,write-requests",
-				 &plat_dat->axi->axi_wr_osr_lmt)) {
+	plat_dat->axi->axi_lpi_en = device_property_read_bool(dev,
+							      "snps,en-lpi");
+	if (device_property_read_u32(dev, "snps,write-requests",
+				     &plat_dat->axi->axi_wr_osr_lmt)) {
 		/**
 		 * Since the register has a reset value of 1, if property
 		 * is missing, default to 1.
@@ -68,8 +69,8 @@ static int dwc_eth_dwmac_config_dt(struct platform_device *pdev,
 		plat_dat->axi->axi_wr_osr_lmt--;
 	}
 
-	if (of_property_read_u32(np, "snps,read-requests",
-				 &plat_dat->axi->axi_rd_osr_lmt)) {
+	if (device_property_read_u32(dev, "snps,read-requests",
+				     &plat_dat->axi->axi_rd_osr_lmt)) {
 		/**
 		 * Since the register has a reset value of 1, if property
 		 * is missing, default to 1.
@@ -82,7 +83,7 @@ static int dwc_eth_dwmac_config_dt(struct platform_device *pdev,
 		 */
 		plat_dat->axi->axi_rd_osr_lmt--;
 	}
-	of_property_read_u32(np, "snps,burst-map", &burst_map);
+	device_property_read_u32(dev, "snps,burst-map", &burst_map);
 
 	/* converts burst-map bitmask to burst array */
 	for (bit_index = 0; bit_index < 7; bit_index++) {
@@ -421,7 +422,7 @@ static int dwc_eth_dwmac_probe(struct platform_device *pdev)
 	void *priv;
 	int ret;
 
-	data = of_device_get_match_data(&pdev->dev);
+	data = device_get_match_data(&pdev->dev);
 
 	memset(&stmmac_res, 0, sizeof(struct stmmac_resources));
 
@@ -478,7 +479,7 @@ static int dwc_eth_dwmac_remove(struct platform_device *pdev)
 	const struct dwc_eth_dwmac_data *data;
 	int err;
 
-	data = of_device_get_match_data(&pdev->dev);
+	data = device_get_match_data(&pdev->dev);
 
 	err = stmmac_dvr_remove(&pdev->dev);
 	if (err < 0)
-- 
cgit v1.2.3-59-g8ed1b


From 1d4605e0aff9ffe5163b84ccefaa8c6c2387504d Mon Sep 17 00:00:00 2001
From: Ajay Gupta <ajayg@nvidia.com>
Date: Sun, 15 Dec 2019 22:14:52 -0800
Subject: net: stmmac: dwc-qos: avoid clk and reset for acpi device

There are no clocks, resets or gpios referenced by Tegra ACPI
device so don't access clocks, resets or gpios interface with
ACPI device.

Clocks, resets and GPIOs for ACPI devices will be handled via
ACPI interface.

Signed-off-by: Ajay Gupta <ajayg@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c
index f87306b3cdae..2342d497348e 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c
@@ -271,6 +271,7 @@ static void *tegra_eqos_probe(struct platform_device *pdev,
 			      struct plat_stmmacenet_data *data,
 			      struct stmmac_resources *res)
 {
+	struct device *dev = &pdev->dev;
 	struct tegra_eqos *eqos;
 	int err;
 
@@ -283,6 +284,9 @@ static void *tegra_eqos_probe(struct platform_device *pdev,
 	eqos->dev = &pdev->dev;
 	eqos->regs = res->addr;
 
+	if (!is_of_node(dev->fwnode))
+		goto bypass_clk_reset_gpio;
+
 	eqos->clk_master = devm_clk_get(&pdev->dev, "master_bus");
 	if (IS_ERR(eqos->clk_master)) {
 		err = PTR_ERR(eqos->clk_master);
@@ -355,6 +359,7 @@ static void *tegra_eqos_probe(struct platform_device *pdev,
 
 	usleep_range(2000, 4000);
 
+bypass_clk_reset_gpio:
 	data->fix_mac_speed = tegra_eqos_fix_speed;
 	data->init = tegra_eqos_init;
 	data->bsp_priv = eqos;
-- 
cgit v1.2.3-59-g8ed1b


From 871185ace40df871a93866b2a7ce441276fc4ee8 Mon Sep 17 00:00:00 2001
From: Lukas Wunner <lukas@wunner.de>
Date: Wed, 20 Nov 2019 12:33:59 +0100
Subject: netfilter: Clean up unnecessary #ifdef

If CONFIG_NETFILTER_INGRESS is not enabled, nf_ingress() becomes a no-op
because it solely contains an if-clause calling nf_hook_ingress_active(),
for which an empty inline stub exists in <linux/netfilter_ingress.h>.

All the symbols used in the if-clause's body are still available even if
CONFIG_NETFILTER_INGRESS is not enabled.

The additional "#ifdef CONFIG_NETFILTER_INGRESS" in nf_ingress() is thus
unnecessary, so drop it.

Signed-off-by: Lukas Wunner <lukas@wunner.de>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/core/dev.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/net/core/dev.c b/net/core/dev.c
index 2c277b8aba38..1ccead4b19bf 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -4932,7 +4932,6 @@ static bool skb_pfmemalloc_protocol(struct sk_buff *skb)
 static inline int nf_ingress(struct sk_buff *skb, struct packet_type **pt_prev,
 			     int *ret, struct net_device *orig_dev)
 {
-#ifdef CONFIG_NETFILTER_INGRESS
 	if (nf_hook_ingress_active(skb)) {
 		int ingress_retval;
 
@@ -4946,7 +4945,6 @@ static inline int nf_ingress(struct sk_buff *skb, struct packet_type **pt_prev,
 		rcu_read_unlock();
 		return ingress_retval;
 	}
-#endif /* CONFIG_NETFILTER_INGRESS */
 	return 0;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 2f5e70c8ce47396bfa8f5c437574b569c02597bb Mon Sep 17 00:00:00 2001
From: Lukas Wunner <lukas@wunner.de>
Date: Wed, 20 Nov 2019 12:33:59 +0100
Subject: netfilter: Document ingress hook

Amend kerneldoc of struct net_device to fix a "make htmldocs" warning:

include/linux/netdevice.h:2045: warning: Function parameter or member 'nf_hooks_ingress' not described in 'net_device'

Reported-by: kbuild test robot <lkp@intel.com>
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/linux/netdevice.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 30745068fb39..0b097bbd3663 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1708,6 +1708,7 @@ enum netdev_priv_flags {
  *	@miniq_ingress:		ingress/clsact qdisc specific data for
  *				ingress processing
  *	@ingress_queue:		XXX: need comments on this one
+ *	@nf_hooks_ingress:	netfilter hooks executed for ingress packets
  *	@broadcast:		hw bcast address
  *
  *	@rx_cpu_rmap:	CPU reverse-mapping for RX completion interrupts,
-- 
cgit v1.2.3-59-g8ed1b


From cf3e204a1ca5442190018a317d9ec181b4639bd6 Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Fri, 13 Dec 2019 16:53:05 +0800
Subject: netfilter: nft_tunnel: no need to call htons() when dumping ports

info->key.tp_src and tp_dst are __be16, when using nla_put_be16()
to dump them, htons() is not needed, so remove it in this patch.

Fixes: af308b94a2a4 ("netfilter: nf_tables: add tunnel support")
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nft_tunnel.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/netfilter/nft_tunnel.c b/net/netfilter/nft_tunnel.c
index 3d4c2ae605a8..ef2065dd4f8a 100644
--- a/net/netfilter/nft_tunnel.c
+++ b/net/netfilter/nft_tunnel.c
@@ -501,8 +501,8 @@ static int nft_tunnel_opts_dump(struct sk_buff *skb,
 static int nft_tunnel_ports_dump(struct sk_buff *skb,
 				 struct ip_tunnel_info *info)
 {
-	if (nla_put_be16(skb, NFTA_TUNNEL_KEY_SPORT, htons(info->key.tp_src)) < 0 ||
-	    nla_put_be16(skb, NFTA_TUNNEL_KEY_DPORT, htons(info->key.tp_dst)) < 0)
+	if (nla_put_be16(skb, NFTA_TUNNEL_KEY_SPORT, info->key.tp_src) < 0 ||
+	    nla_put_be16(skb, NFTA_TUNNEL_KEY_DPORT, info->key.tp_dst) < 0)
 		return -1;
 
 	return 0;
-- 
cgit v1.2.3-59-g8ed1b


From 0705f95c332081036d85f26691e9d3cd7d901c31 Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Fri, 13 Dec 2019 16:53:06 +0800
Subject: netfilter: nft_tunnel: add the missing ERSPAN_VERSION nla_policy

ERSPAN_VERSION is an attribute parsed in kernel side, nla_policy
type should be added for it, like other attributes.

Fixes: af308b94a2a4 ("netfilter: nf_tables: add tunnel support")
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nft_tunnel.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/net/netfilter/nft_tunnel.c b/net/netfilter/nft_tunnel.c
index ef2065dd4f8a..6538895466e0 100644
--- a/net/netfilter/nft_tunnel.c
+++ b/net/netfilter/nft_tunnel.c
@@ -248,8 +248,9 @@ static int nft_tunnel_obj_vxlan_init(const struct nlattr *attr,
 }
 
 static const struct nla_policy nft_tunnel_opts_erspan_policy[NFTA_TUNNEL_KEY_ERSPAN_MAX + 1] = {
+	[NFTA_TUNNEL_KEY_ERSPAN_VERSION]	= { .type = NLA_U32 },
 	[NFTA_TUNNEL_KEY_ERSPAN_V1_INDEX]	= { .type = NLA_U32 },
-	[NFTA_TUNNEL_KEY_ERSPAN_V2_DIR]	= { .type = NLA_U8 },
+	[NFTA_TUNNEL_KEY_ERSPAN_V2_DIR]		= { .type = NLA_U8 },
 	[NFTA_TUNNEL_KEY_ERSPAN_V2_HWID]	= { .type = NLA_U8 },
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From 2149f36dbd44f2d6bf5357e5d096205b92cd854a Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Fri, 13 Dec 2019 16:53:07 +0800
Subject: netfilter: nft_tunnel: also dump ERSPAN_VERSION

This is not necessary, but it'll be easier to parse in userspace,
also given that other places like act_tunnel_key, cls_flower and
ip_tunnel_core are also doing so.

Signed-off-by: Xin Long <lucien.xin@gmail.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nft_tunnel.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/net/netfilter/nft_tunnel.c b/net/netfilter/nft_tunnel.c
index 6538895466e0..b3a9b10ff43d 100644
--- a/net/netfilter/nft_tunnel.c
+++ b/net/netfilter/nft_tunnel.c
@@ -479,6 +479,9 @@ static int nft_tunnel_opts_dump(struct sk_buff *skb,
 				 htonl(opts->u.vxlan.gbp)))
 			return -1;
 	} else if (opts->flags & TUNNEL_ERSPAN_OPT) {
+		if (nla_put_be32(skb, NFTA_TUNNEL_KEY_ERSPAN_VERSION,
+				 htonl(opts->u.erspan.version)))
+			return -1;
 		switch (opts->u.erspan.version) {
 		case ERSPAN_VERSION:
 			if (nla_put_be32(skb, NFTA_TUNNEL_KEY_ERSPAN_V1_INDEX,
-- 
cgit v1.2.3-59-g8ed1b


From 73239bd9707ab2f3b7621e50468c14410cf4e2c2 Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Fri, 13 Dec 2019 16:53:08 +0800
Subject: netfilter: nft_tunnel: also dump OPTS_ERSPAN/VXLAN

This patch is to add the nest attr OPTS_ERSPAN/VXLAN when dumping
KEY_OPTS, and it would be helpful when parsing in userpace. Also,
this is needed for supporting multiple geneve opts in the future
patches.

Signed-off-by: Xin Long <lucien.xin@gmail.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nft_tunnel.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/net/netfilter/nft_tunnel.c b/net/netfilter/nft_tunnel.c
index b3a9b10ff43d..eb1740236526 100644
--- a/net/netfilter/nft_tunnel.c
+++ b/net/netfilter/nft_tunnel.c
@@ -468,17 +468,24 @@ static int nft_tunnel_opts_dump(struct sk_buff *skb,
 				struct nft_tunnel_obj *priv)
 {
 	struct nft_tunnel_opts *opts = &priv->opts;
-	struct nlattr *nest;
+	struct nlattr *nest, *inner;
 
 	nest = nla_nest_start_noflag(skb, NFTA_TUNNEL_KEY_OPTS);
 	if (!nest)
 		return -1;
 
 	if (opts->flags & TUNNEL_VXLAN_OPT) {
+		inner = nla_nest_start_noflag(skb, NFTA_TUNNEL_KEY_OPTS_VXLAN);
+		if (!inner)
+			return -1;
 		if (nla_put_be32(skb, NFTA_TUNNEL_KEY_VXLAN_GBP,
 				 htonl(opts->u.vxlan.gbp)))
 			return -1;
+		nla_nest_end(skb, inner);
 	} else if (opts->flags & TUNNEL_ERSPAN_OPT) {
+		inner = nla_nest_start_noflag(skb, NFTA_TUNNEL_KEY_OPTS_ERSPAN);
+		if (!inner)
+			return -1;
 		if (nla_put_be32(skb, NFTA_TUNNEL_KEY_ERSPAN_VERSION,
 				 htonl(opts->u.erspan.version)))
 			return -1;
@@ -496,6 +503,7 @@ static int nft_tunnel_opts_dump(struct sk_buff *skb,
 				return -1;
 			break;
 		}
+		nla_nest_end(skb, inner);
 	}
 	nla_nest_end(skb, nest);
 
-- 
cgit v1.2.3-59-g8ed1b


From 7e03998429ee0a27c13b8690c11f9cf40f67f6e9 Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Fri, 13 Dec 2019 16:53:09 +0800
Subject: netfilter: nft_tunnel: add the missing nla_nest_cancel()

When nla_put_xxx() fails under nla_nest_start_noflag(),
nla_nest_cancel() should be called, so that the skb can
be trimmed properly.

Signed-off-by: Xin Long <lucien.xin@gmail.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nft_tunnel.c | 38 ++++++++++++++++++++++++++------------
 1 file changed, 26 insertions(+), 12 deletions(-)

diff --git a/net/netfilter/nft_tunnel.c b/net/netfilter/nft_tunnel.c
index eb1740236526..23cd163689d5 100644
--- a/net/netfilter/nft_tunnel.c
+++ b/net/netfilter/nft_tunnel.c
@@ -443,10 +443,15 @@ static int nft_tunnel_ip_dump(struct sk_buff *skb, struct ip_tunnel_info *info)
 		if (!nest)
 			return -1;
 
-		if (nla_put_in6_addr(skb, NFTA_TUNNEL_KEY_IP6_SRC, &info->key.u.ipv6.src) < 0 ||
-		    nla_put_in6_addr(skb, NFTA_TUNNEL_KEY_IP6_DST, &info->key.u.ipv6.dst) < 0 ||
-		    nla_put_be32(skb, NFTA_TUNNEL_KEY_IP6_FLOWLABEL, info->key.label))
+		if (nla_put_in6_addr(skb, NFTA_TUNNEL_KEY_IP6_SRC,
+				     &info->key.u.ipv6.src) < 0 ||
+		    nla_put_in6_addr(skb, NFTA_TUNNEL_KEY_IP6_DST,
+				     &info->key.u.ipv6.dst) < 0 ||
+		    nla_put_be32(skb, NFTA_TUNNEL_KEY_IP6_FLOWLABEL,
+				 info->key.label)) {
+			nla_nest_cancel(skb, nest);
 			return -1;
+		}
 
 		nla_nest_end(skb, nest);
 	} else {
@@ -454,9 +459,13 @@ static int nft_tunnel_ip_dump(struct sk_buff *skb, struct ip_tunnel_info *info)
 		if (!nest)
 			return -1;
 
-		if (nla_put_in_addr(skb, NFTA_TUNNEL_KEY_IP_SRC, info->key.u.ipv4.src) < 0 ||
-		    nla_put_in_addr(skb, NFTA_TUNNEL_KEY_IP_DST, info->key.u.ipv4.dst) < 0)
+		if (nla_put_in_addr(skb, NFTA_TUNNEL_KEY_IP_SRC,
+				    info->key.u.ipv4.src) < 0 ||
+		    nla_put_in_addr(skb, NFTA_TUNNEL_KEY_IP_DST,
+				    info->key.u.ipv4.dst) < 0) {
+			nla_nest_cancel(skb, nest);
 			return -1;
+		}
 
 		nla_nest_end(skb, nest);
 	}
@@ -477,37 +486,42 @@ static int nft_tunnel_opts_dump(struct sk_buff *skb,
 	if (opts->flags & TUNNEL_VXLAN_OPT) {
 		inner = nla_nest_start_noflag(skb, NFTA_TUNNEL_KEY_OPTS_VXLAN);
 		if (!inner)
-			return -1;
+			goto failure;
 		if (nla_put_be32(skb, NFTA_TUNNEL_KEY_VXLAN_GBP,
 				 htonl(opts->u.vxlan.gbp)))
-			return -1;
+			goto inner_failure;
 		nla_nest_end(skb, inner);
 	} else if (opts->flags & TUNNEL_ERSPAN_OPT) {
 		inner = nla_nest_start_noflag(skb, NFTA_TUNNEL_KEY_OPTS_ERSPAN);
 		if (!inner)
-			return -1;
+			goto failure;
 		if (nla_put_be32(skb, NFTA_TUNNEL_KEY_ERSPAN_VERSION,
 				 htonl(opts->u.erspan.version)))
-			return -1;
+			goto inner_failure;
 		switch (opts->u.erspan.version) {
 		case ERSPAN_VERSION:
 			if (nla_put_be32(skb, NFTA_TUNNEL_KEY_ERSPAN_V1_INDEX,
 					 opts->u.erspan.u.index))
-				return -1;
+				goto inner_failure;
 			break;
 		case ERSPAN_VERSION2:
 			if (nla_put_u8(skb, NFTA_TUNNEL_KEY_ERSPAN_V2_HWID,
 				       get_hwid(&opts->u.erspan.u.md2)) ||
 			    nla_put_u8(skb, NFTA_TUNNEL_KEY_ERSPAN_V2_DIR,
 				       opts->u.erspan.u.md2.dir))
-				return -1;
+				goto inner_failure;
 			break;
 		}
 		nla_nest_end(skb, inner);
 	}
 	nla_nest_end(skb, nest);
-
 	return 0;
+
+inner_failure:
+	nla_nest_cancel(skb, inner);
+failure:
+	nla_nest_cancel(skb, nest);
+	return -1;
 }
 
 static int nft_tunnel_ports_dump(struct sk_buff *skb,
-- 
cgit v1.2.3-59-g8ed1b


From 13d74c0a9708a4f1ab0164a800ce9ea3de32f47b Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Fri, 13 Dec 2019 01:58:15 +0100
Subject: netfilter: conntrack: remove two export symbols

Not used anywhere, remove them.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nf_conntrack_core.c   | 1 -
 net/netfilter/nf_conntrack_extend.c | 1 -
 2 files changed, 2 deletions(-)

diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index 0af1898af2b8..983a9481e8f8 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -2333,7 +2333,6 @@ int nf_conntrack_set_hashsize(const char *val, const struct kernel_param *kp)
 
 	return nf_conntrack_hash_resize(hashsize);
 }
-EXPORT_SYMBOL_GPL(nf_conntrack_set_hashsize);
 
 static __always_inline unsigned int total_extension_size(void)
 {
diff --git a/net/netfilter/nf_conntrack_extend.c b/net/netfilter/nf_conntrack_extend.c
index c24e5b64b00c..3dbe2329c3f1 100644
--- a/net/netfilter/nf_conntrack_extend.c
+++ b/net/netfilter/nf_conntrack_extend.c
@@ -37,7 +37,6 @@ void nf_ct_ext_destroy(struct nf_conn *ct)
 
 	kfree(ct->ext);
 }
-EXPORT_SYMBOL(nf_ct_ext_destroy);
 
 void *nf_ct_ext_add(struct nf_conn *ct, enum nf_ct_ext_id id, gfp_t gfp)
 {
-- 
cgit v1.2.3-59-g8ed1b


From b7ffa045e7000d5410bf206454e0cb8de0428ed5 Mon Sep 17 00:00:00 2001
From: Jon Maloy <jon.maloy@ericsson.com>
Date: Mon, 16 Dec 2019 19:21:02 +0100
Subject: tipc: don't send gap blocks in ACK messages

In the commit referred to below we eliminated sending of the 'gap'
indicator in regular ACK messages, reserving this to explicit NACK
ditto.

Unfortunately we missed to also eliminate building of the 'gap block'
area in ACK messages. This area is meant to report gaps in the
received packet sequence following the initial gap, so that lost
packets can be retransmitted earlier and received out-of-sequence
packets can be released earlier. However, the interpretation of those
blocks is dependent on a complete and correct sequence of gaps and
acks. Hence, when the initial gap indicator is missing a single gap
block will be interpreted as an acknowledgment of all preceding
packets. This may lead to packets being released prematurely from the
sender's transmit queue, with easily predicatble consequences.

We now fix this by not building any gap block area if there is no
initial gap to report.

Fixes: commit 02288248b051 ("tipc: eliminate gap indicator from ACK messages")
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/tipc/link.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/net/tipc/link.c b/net/tipc/link.c
index 94dd48cd70a3..467c53a1fb5c 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -250,7 +250,7 @@ static int tipc_link_build_nack_msg(struct tipc_link *l,
 static void tipc_link_build_bc_init_msg(struct tipc_link *l,
 					struct sk_buff_head *xmitq);
 static int tipc_link_release_pkts(struct tipc_link *l, u16 to);
-static u16 tipc_build_gap_ack_blks(struct tipc_link *l, void *data);
+static u16 tipc_build_gap_ack_blks(struct tipc_link *l, void *data, u16 gap);
 static int tipc_link_advance_transmq(struct tipc_link *l, u16 acked, u16 gap,
 				     struct tipc_gap_ack_blks *ga,
 				     struct sk_buff_head *xmitq);
@@ -1423,14 +1423,14 @@ static int tipc_link_release_pkts(struct tipc_link *l, u16 acked)
  *
  * returns the actual allocated memory size
  */
-static u16 tipc_build_gap_ack_blks(struct tipc_link *l, void *data)
+static u16 tipc_build_gap_ack_blks(struct tipc_link *l, void *data, u16 gap)
 {
 	struct sk_buff *skb = skb_peek(&l->deferdq);
 	struct tipc_gap_ack_blks *ga = data;
 	u16 len, expect, seqno = 0;
 	u8 n = 0;
 
-	if (!skb)
+	if (!skb || !gap)
 		goto exit;
 
 	expect = buf_seqno(skb);
@@ -1739,7 +1739,7 @@ static void tipc_link_build_proto_msg(struct tipc_link *l, int mtyp, bool probe,
 		msg_set_probe(hdr, probe);
 		msg_set_is_keepalive(hdr, probe || probe_reply);
 		if (l->peer_caps & TIPC_GAP_ACK_BLOCK)
-			glen = tipc_build_gap_ack_blks(l, data);
+			glen = tipc_build_gap_ack_blks(l, data, rcvgap);
 		tipc_mon_prep(l->net, data + glen, &dlen, mstate, l->bearer_id);
 		msg_set_size(hdr, INT_H_SIZE + glen + dlen);
 		skb_trim(skb, INT_H_SIZE + glen + dlen);
-- 
cgit v1.2.3-59-g8ed1b


From 8ae674964e67eb7deb3f0e489bfe9c102a7990b0 Mon Sep 17 00:00:00 2001
From: Florian Fainelli <f.fainelli@gmail.com>
Date: Mon, 16 Dec 2019 10:32:47 -0800
Subject: net: dsa: Make PHYLINK related function static again

Commit 77373d49de22 ("net: dsa: Move the phylink driver calls into
port.c") moved and exported a bunch of symbols, but they are not used
outside of net/dsa/port.c at the moment, so no reason to export them.

Reported-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Vivien Didelot <vivien.didelot@gmail.com>
Acked-by: Russell King <rmk+kernel@armlinux.org.uk>
Acked-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/dsa/dsa_priv.h | 16 ----------------
 net/dsa/port.c     | 38 ++++++++++++++++----------------------
 2 files changed, 16 insertions(+), 38 deletions(-)

diff --git a/net/dsa/dsa_priv.h b/net/dsa/dsa_priv.h
index 2dd86d9bcda9..09ea2fd78c74 100644
--- a/net/dsa/dsa_priv.h
+++ b/net/dsa/dsa_priv.h
@@ -150,22 +150,6 @@ int dsa_port_vid_add(struct dsa_port *dp, u16 vid, u16 flags);
 int dsa_port_vid_del(struct dsa_port *dp, u16 vid);
 int dsa_port_link_register_of(struct dsa_port *dp);
 void dsa_port_link_unregister_of(struct dsa_port *dp);
-void dsa_port_phylink_validate(struct phylink_config *config,
-			       unsigned long *supported,
-			       struct phylink_link_state *state);
-void dsa_port_phylink_mac_pcs_get_state(struct phylink_config *config,
-					struct phylink_link_state *state);
-void dsa_port_phylink_mac_config(struct phylink_config *config,
-				 unsigned int mode,
-				 const struct phylink_link_state *state);
-void dsa_port_phylink_mac_an_restart(struct phylink_config *config);
-void dsa_port_phylink_mac_link_down(struct phylink_config *config,
-				    unsigned int mode,
-				    phy_interface_t interface);
-void dsa_port_phylink_mac_link_up(struct phylink_config *config,
-				  unsigned int mode,
-				  phy_interface_t interface,
-				  struct phy_device *phydev);
 extern const struct phylink_mac_ops dsa_port_phylink_mac_ops;
 
 /* slave.c */
diff --git a/net/dsa/port.c b/net/dsa/port.c
index 46ac9ba21987..ffb5601f7ed6 100644
--- a/net/dsa/port.c
+++ b/net/dsa/port.c
@@ -415,9 +415,9 @@ static struct phy_device *dsa_port_get_phy_device(struct dsa_port *dp)
 	return phydev;
 }
 
-void dsa_port_phylink_validate(struct phylink_config *config,
-			       unsigned long *supported,
-			       struct phylink_link_state *state)
+static void dsa_port_phylink_validate(struct phylink_config *config,
+				      unsigned long *supported,
+				      struct phylink_link_state *state)
 {
 	struct dsa_port *dp = container_of(config, struct dsa_port, pl_config);
 	struct dsa_switch *ds = dp->ds;
@@ -427,10 +427,9 @@ void dsa_port_phylink_validate(struct phylink_config *config,
 
 	ds->ops->phylink_validate(ds, dp->index, supported, state);
 }
-EXPORT_SYMBOL_GPL(dsa_port_phylink_validate);
 
-void dsa_port_phylink_mac_pcs_get_state(struct phylink_config *config,
-					struct phylink_link_state *state)
+static void dsa_port_phylink_mac_pcs_get_state(struct phylink_config *config,
+					       struct phylink_link_state *state)
 {
 	struct dsa_port *dp = container_of(config, struct dsa_port, pl_config);
 	struct dsa_switch *ds = dp->ds;
@@ -444,11 +443,10 @@ void dsa_port_phylink_mac_pcs_get_state(struct phylink_config *config,
 	if (ds->ops->phylink_mac_link_state(ds, dp->index, state) < 0)
 		state->link = 0;
 }
-EXPORT_SYMBOL_GPL(dsa_port_phylink_mac_pcs_get_state);
 
-void dsa_port_phylink_mac_config(struct phylink_config *config,
-				 unsigned int mode,
-				 const struct phylink_link_state *state)
+static void dsa_port_phylink_mac_config(struct phylink_config *config,
+					unsigned int mode,
+					const struct phylink_link_state *state)
 {
 	struct dsa_port *dp = container_of(config, struct dsa_port, pl_config);
 	struct dsa_switch *ds = dp->ds;
@@ -458,9 +456,8 @@ void dsa_port_phylink_mac_config(struct phylink_config *config,
 
 	ds->ops->phylink_mac_config(ds, dp->index, mode, state);
 }
-EXPORT_SYMBOL_GPL(dsa_port_phylink_mac_config);
 
-void dsa_port_phylink_mac_an_restart(struct phylink_config *config)
+static void dsa_port_phylink_mac_an_restart(struct phylink_config *config)
 {
 	struct dsa_port *dp = container_of(config, struct dsa_port, pl_config);
 	struct dsa_switch *ds = dp->ds;
@@ -470,11 +467,10 @@ void dsa_port_phylink_mac_an_restart(struct phylink_config *config)
 
 	ds->ops->phylink_mac_an_restart(ds, dp->index);
 }
-EXPORT_SYMBOL_GPL(dsa_port_phylink_mac_an_restart);
 
-void dsa_port_phylink_mac_link_down(struct phylink_config *config,
-				    unsigned int mode,
-				    phy_interface_t interface)
+static void dsa_port_phylink_mac_link_down(struct phylink_config *config,
+					   unsigned int mode,
+					   phy_interface_t interface)
 {
 	struct dsa_port *dp = container_of(config, struct dsa_port, pl_config);
 	struct phy_device *phydev = NULL;
@@ -491,12 +487,11 @@ void dsa_port_phylink_mac_link_down(struct phylink_config *config,
 
 	ds->ops->phylink_mac_link_down(ds, dp->index, mode, interface);
 }
-EXPORT_SYMBOL_GPL(dsa_port_phylink_mac_link_down);
 
-void dsa_port_phylink_mac_link_up(struct phylink_config *config,
-				  unsigned int mode,
-				  phy_interface_t interface,
-				  struct phy_device *phydev)
+static void dsa_port_phylink_mac_link_up(struct phylink_config *config,
+					 unsigned int mode,
+					 phy_interface_t interface,
+					 struct phy_device *phydev)
 {
 	struct dsa_port *dp = container_of(config, struct dsa_port, pl_config);
 	struct dsa_switch *ds = dp->ds;
@@ -509,7 +504,6 @@ void dsa_port_phylink_mac_link_up(struct phylink_config *config,
 
 	ds->ops->phylink_mac_link_up(ds, dp->index, mode, interface, phydev);
 }
-EXPORT_SYMBOL_GPL(dsa_port_phylink_mac_link_up);
 
 const struct phylink_mac_ops dsa_port_phylink_mac_ops = {
 	.validate = dsa_port_phylink_validate,
-- 
cgit v1.2.3-59-g8ed1b


From 92f7440ecc9331aa1e274273d4a3cb5c58b69706 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Mon, 16 Dec 2019 22:14:25 -0800
Subject: selftests/bpf: More succinct Makefile output

Similarly to bpftool/libbpf output, make selftests/bpf output succinct
per-item output line. Output is roughly as follows:

$ make
...
  CLANG-LLC [test_maps] pyperf600.o
  CLANG-LLC [test_maps] strobemeta.o
  CLANG-LLC [test_maps] pyperf100.o
  EXTRA-OBJ [test_progs] cgroup_helpers.o
  EXTRA-OBJ [test_progs] trace_helpers.o
     BINARY test_align
     BINARY test_verifier_log
   GEN-SKEL [test_progs] fexit_bpf2bpf.skel.h
   GEN-SKEL [test_progs] test_global_data.skel.h
   GEN-SKEL [test_progs] sendmsg6_prog.skel.h
...

To see the actual command invocation, verbose mode can be turned on with V=1
argument:

$ make V=1

... very verbose output ...

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20191217061425.2346359-1-andriin@fb.com
---
 tools/testing/selftests/bpf/Makefile | 36 ++++++++++++++++++++++++++++++++++++
 1 file changed, 36 insertions(+)

diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index ba90621617a8..c652bd84ef0e 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -77,6 +77,24 @@ TEST_GEN_PROGS_EXTENDED = test_sock_addr test_skb_cgroup_id_user \
 
 TEST_CUSTOM_PROGS = urandom_read
 
+# Emit succinct information message describing current building step
+# $1 - generic step name (e.g., CC, LINK, etc);
+# $2 - optional "flavor" specifier; if provided, will be emitted as [flavor];
+# $3 - target (assumed to be file); only file name will be emitted;
+# $4 - optional extra arg, emitted as-is, if provided.
+ifeq ($(V),1)
+msg =
+else
+msg = @$(info $(1)$(if $(2), [$(2)]) $(notdir $(3)))$(if $(4), $(4))
+endif
+
+# override lib.mk's default rules
+OVERRIDE_TARGETS := 1
+override define CLEAN
+	$(call msg,    CLEAN)
+	$(RM) -r $(TEST_GEN_PROGS) $(TEST_GEN_PROGS_EXTENDED) $(TEST_GEN_FILES) $(EXTRA_CLEAN)
+endef
+
 include ../lib.mk
 
 # Define simple and short `make test_progs`, `make test_sysctl`, etc targets
@@ -89,10 +107,16 @@ $(notdir $(TEST_GEN_PROGS)						\
 	 $(TEST_GEN_PROGS_EXTENDED)					\
 	 $(TEST_CUSTOM_PROGS)): %: $(OUTPUT)/% ;
 
+$(OUTPUT)/%:%.c
+	$(call msg,     BINARY,,$@)
+	$(LINK.c) $^ $(LDLIBS) -o $@
+
 $(OUTPUT)/urandom_read: urandom_read.c
+	$(call msg,     BINARY,,$@)
 	$(CC) -o $@ $< -Wl,--build-id
 
 $(OUTPUT)/test_stub.o: test_stub.c
+	$(call msg,         CC,,$@)
 	$(CC) -c $(CFLAGS) -o $@ $<
 
 BPFOBJ := $(OUTPUT)/libbpf.a
@@ -167,24 +191,28 @@ $(OUTPUT)/flow_dissector_load.o: flow_dissector_load.h
 # $3 - CFLAGS
 # $4 - LDFLAGS
 define CLANG_BPF_BUILD_RULE
+	$(call msg,  CLANG-LLC,$(TRUNNER_BINARY),$2)
 	($(CLANG) $3 -O2 -target bpf -emit-llvm				\
 		-c $1 -o - || echo "BPF obj compilation failed") | 	\
 	$(LLC) -mattr=dwarfris -march=bpf -mcpu=probe $4 -filetype=obj -o $2
 endef
 # Similar to CLANG_BPF_BUILD_RULE, but with disabled alu32
 define CLANG_NOALU32_BPF_BUILD_RULE
+	$(call msg,  CLANG-LLC,$(TRUNNER_BINARY),$2)
 	($(CLANG) $3 -O2 -target bpf -emit-llvm				\
 		-c $1 -o - || echo "BPF obj compilation failed") | 	\
 	$(LLC) -march=bpf -mcpu=v2 $4 -filetype=obj -o $2
 endef
 # Similar to CLANG_BPF_BUILD_RULE, but using native Clang and bpf LLC
 define CLANG_NATIVE_BPF_BUILD_RULE
+	$(call msg,  CLANG-BPF,$(TRUNNER_BINARY),$2)
 	($(CLANG) $3 -O2 -emit-llvm					\
 		-c $1 -o - || echo "BPF obj compilation failed") | 	\
 	$(LLC) -march=bpf -mcpu=probe $4 -filetype=obj -o $2
 endef
 # Build BPF object using GCC
 define GCC_BPF_BUILD_RULE
+	$(call msg,    GCC-BPF,$(TRUNNER_BINARY),$2)
 	$(BPF_GCC) $3 $4 -O2 -c $1 -o $2
 endef
 
@@ -243,6 +271,7 @@ $(TRUNNER_BPF_OBJS): $(TRUNNER_OUTPUT)/%.o:				\
 $(TRUNNER_BPF_SKELS): $(TRUNNER_OUTPUT)/%.skel.h:			\
 		      $(TRUNNER_OUTPUT)/%.o				\
 		      | $(BPFTOOL) $(TRUNNER_OUTPUT)
+	$$(call msg,   GEN-SKEL,$(TRUNNER_BINARY),$$@)
 	$$(BPFTOOL) gen skeleton $$< > $$@
 endif
 
@@ -250,6 +279,7 @@ endif
 ifeq ($($(TRUNNER_TESTS_DIR)-tests-hdr),)
 $(TRUNNER_TESTS_DIR)-tests-hdr := y
 $(TRUNNER_TESTS_HDR): $(TRUNNER_TESTS_DIR)/*.c
+	$$(call msg,   TEST-HDR,$(TRUNNER_BINARY),$$@)
 	$$(shell ( cd $(TRUNNER_TESTS_DIR);				\
 		  echo '/* Generated header, do not edit */';		\
 		  ls *.c 2> /dev/null |					\
@@ -265,6 +295,7 @@ $(TRUNNER_TEST_OBJS): $(TRUNNER_OUTPUT)/%.test.o:			\
 		      $(TRUNNER_BPF_OBJS)				\
 		      $(TRUNNER_BPF_SKELS)				\
 		      $$(BPFOBJ) | $(TRUNNER_OUTPUT)
+	$$(call msg,   TEST-OBJ,$(TRUNNER_BINARY),$$@)
 	cd $$(@D) && $$(CC) $$(CFLAGS) -c $(CURDIR)/$$< $$(LDLIBS) -o $$(@F)
 
 $(TRUNNER_EXTRA_OBJS): $(TRUNNER_OUTPUT)/%.o:				\
@@ -272,17 +303,20 @@ $(TRUNNER_EXTRA_OBJS): $(TRUNNER_OUTPUT)/%.o:				\
 		       $(TRUNNER_EXTRA_HDRS)				\
 		       $(TRUNNER_TESTS_HDR)				\
 		       $$(BPFOBJ) | $(TRUNNER_OUTPUT)
+	$$(call msg,  EXTRA-OBJ,$(TRUNNER_BINARY),$$@)
 	$$(CC) $$(CFLAGS) -c $$< $$(LDLIBS) -o $$@
 
 # only copy extra resources if in flavored build
 $(TRUNNER_BINARY)-extras: $(TRUNNER_EXTRA_FILES) | $(TRUNNER_OUTPUT)
 ifneq ($2,)
+	$$(call msg,  EXTRAS-CP,$(TRUNNER_BINARY),$(TRUNNER_EXTRA_FILES))
 	cp -a $$^ $(TRUNNER_OUTPUT)/
 endif
 
 $(OUTPUT)/$(TRUNNER_BINARY): $(TRUNNER_TEST_OBJS)			\
 			     $(TRUNNER_EXTRA_OBJS) $$(BPFOBJ)		\
 			     | $(TRUNNER_BINARY)-extras
+	$$(call msg,     BINARY,,$$@)
 	$$(CC) $$(CFLAGS) $$(filter %.a %.o,$$^) $$(LDLIBS) -o $$@
 
 endef
@@ -334,10 +368,12 @@ verifier/tests.h: verifier/*.c
 		  echo '#endif' \
 		) > verifier/tests.h)
 $(OUTPUT)/test_verifier: test_verifier.c verifier/tests.h $(BPFOBJ) | $(OUTPUT)
+	$(call msg,     BINARY,,$@)
 	$(CC) $(CFLAGS) $(filter %.a %.o %.c,$^) $(LDLIBS) -o $@
 
 # Make sure we are able to include and link libbpf against c++.
 $(OUTPUT)/test_cpp: test_cpp.cpp $(BPFOBJ)
+	$(call msg,        CXX,,$@)
 	$(CXX) $(CFLAGS) $^ $(LDLIBS) -o $@
 
 EXTRA_CLEAN := $(TEST_CUSTOM_PROGS)					\
-- 
cgit v1.2.3-59-g8ed1b


From fb9a98e160f10555936e1c7041d4efda4954d1df Mon Sep 17 00:00:00 2001
From: Toke Høiland-Jørgensen <toke@redhat.com>
Date: Tue, 17 Dec 2019 12:28:10 +0100
Subject: libbpf: Fix libbpf_common.h when installing libbpf through 'make
 install'
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This fixes two issues with the newly introduced libbpf_common.h file:

- The header failed to include <string.h> for the definition of memset()
- The new file was not included in the install_headers rule in the Makefile

Both of these issues cause breakage when installing libbpf with 'make
install' and trying to use it in applications.

Fixes: 544402d4b493 ("libbpf: Extract common user-facing helpers")
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20191217112810.768078-1-toke@redhat.com
---
 tools/lib/bpf/Makefile        | 1 +
 tools/lib/bpf/libbpf_common.h | 2 ++
 2 files changed, 3 insertions(+)

diff --git a/tools/lib/bpf/Makefile b/tools/lib/bpf/Makefile
index a3718cb275f2..d4790121adf4 100644
--- a/tools/lib/bpf/Makefile
+++ b/tools/lib/bpf/Makefile
@@ -251,6 +251,7 @@ install_headers: bpf_helper_defs.h
 		$(call do_install,libbpf.h,$(prefix)/include/bpf,644); \
 		$(call do_install,btf.h,$(prefix)/include/bpf,644); \
 		$(call do_install,libbpf_util.h,$(prefix)/include/bpf,644); \
+		$(call do_install,libbpf_common.h,$(prefix)/include/bpf,644); \
 		$(call do_install,xsk.h,$(prefix)/include/bpf,644); \
 		$(call do_install,bpf_helpers.h,$(prefix)/include/bpf,644); \
 		$(call do_install,bpf_helper_defs.h,$(prefix)/include/bpf,644); \
diff --git a/tools/lib/bpf/libbpf_common.h b/tools/lib/bpf/libbpf_common.h
index 4fb833840961..a23ae1ac27eb 100644
--- a/tools/lib/bpf/libbpf_common.h
+++ b/tools/lib/bpf/libbpf_common.h
@@ -9,6 +9,8 @@
 #ifndef __LIBBPF_LIBBPF_COMMON_H
 #define __LIBBPF_LIBBPF_COMMON_H
 
+#include <string.h>
+
 #ifndef LIBBPF_API
 #define LIBBPF_API __attribute__((visibility("default")))
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From 4a3d6c6a6e4d652808729e7a2a8c3774a5a5c814 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Tue, 17 Dec 2019 15:42:28 -0800
Subject: libbpf: Reduce log level for custom section names

Libbpf is trying to recognize BPF program type based on its section name
during bpf_object__open() phase. This is not strictly enforced and user code
has ability to specify/override correct BPF program type after open.  But if
BPF program is using custom section name, libbpf will still emit warnings,
which can be quite annoying to users. This patch reduces log level of
information messages emitted by libbpf if section name is not canonical. User
can still get a list of all supported section names as debug-level message.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191217234228.1739308-1-andriin@fb.com
---
 tools/lib/bpf/libbpf.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 3fe42d6b0c2f..906bbbf7b2e4 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -5883,7 +5883,7 @@ int libbpf_prog_type_by_name(const char *name, enum bpf_prog_type *prog_type,
 		return 0;
 	}
 
-	pr_warn("failed to guess program type from ELF section '%s'\n", name);
+	pr_debug("failed to guess program type from ELF section '%s'\n", name);
 	type_names = libbpf_get_type_names(false);
 	if (type_names != NULL) {
 		pr_debug("supported section(type) names are:%s\n", type_names);
@@ -6001,10 +6001,10 @@ int libbpf_attach_type_by_name(const char *name,
 		*attach_type = section_defs[i].attach_type;
 		return 0;
 	}
-	pr_warn("failed to guess attach type based on ELF section name '%s'\n", name);
+	pr_debug("failed to guess attach type based on ELF section name '%s'\n", name);
 	type_names = libbpf_get_type_names(true);
 	if (type_names != NULL) {
-		pr_info("attachable section(type) names are:%s\n", type_names);
+		pr_debug("attachable section(type) names are:%s\n", type_names);
 		free(type_names);
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From 5dc7a8b2114479fd945956ece9875f747a996a8e Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Tue, 17 Dec 2019 21:25:50 -0800
Subject: bpftool, selftests/bpf: Embed object file inside skeleton

Embed contents of BPF object file used for BPF skeleton generation inside
skeleton itself. This allows to keep BPF object file and its skeleton in sync
at all times, and simpifies skeleton instantiation.

Also switch existing selftests to not require BPF_EMBED_OBJ anymore.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20191218052552.2915188-2-andriin@fb.com
---
 tools/bpf/bpftool/gen.c                            | 232 +++++++++++++--------
 .../selftests/bpf/prog_tests/attach_probe.c        |   4 +-
 .../testing/selftests/bpf/prog_tests/core_extern.c |   4 +-
 .../selftests/bpf/prog_tests/fentry_fexit.c        |  10 +-
 .../testing/selftests/bpf/prog_tests/fentry_test.c |   7 +-
 tools/testing/selftests/bpf/prog_tests/mmap.c      |   4 +-
 tools/testing/selftests/bpf/prog_tests/skeleton.c  |   4 +-
 .../selftests/bpf/prog_tests/stacktrace_build_id.c |   4 +-
 .../bpf/prog_tests/stacktrace_build_id_nmi.c       |   4 +-
 9 files changed, 154 insertions(+), 119 deletions(-)

diff --git a/tools/bpf/bpftool/gen.c b/tools/bpf/bpftool/gen.c
index f70088b4c19b..8d93c8f90f82 100644
--- a/tools/bpf/bpftool/gen.c
+++ b/tools/bpf/bpftool/gen.c
@@ -16,6 +16,7 @@
 #include <libbpf.h>
 #include <sys/types.h>
 #include <sys/stat.h>
+#include <sys/mman.h>
 #include <unistd.h>
 
 #include "btf.h"
@@ -261,14 +262,16 @@ static int codegen(const char *template, ...)
 static int do_skeleton(int argc, char **argv)
 {
 	char header_guard[MAX_OBJ_NAME_LEN + sizeof("__SKEL_H__")];
-	size_t i, map_cnt = 0, prog_cnt = 0;
-	char obj_name[MAX_OBJ_NAME_LEN];
+	size_t i, map_cnt = 0, prog_cnt = 0, file_sz, mmap_sz;
+	DECLARE_LIBBPF_OPTS(bpf_object_open_opts, opts);
+	char obj_name[MAX_OBJ_NAME_LEN], *obj_data;
+	struct bpf_object *obj = NULL;
 	const char *file, *ident;
 	struct bpf_program *prog;
-	struct bpf_object *obj;
+	int fd, len, err = -1;
 	struct bpf_map *map;
 	struct btf *btf;
-	int err = -1;
+	struct stat st;
 
 	if (!REQ_ARGS(1)) {
 		usage();
@@ -281,14 +284,31 @@ static int do_skeleton(int argc, char **argv)
 		return -1;
 	}
 
-	obj = bpf_object__open_file(file, NULL);
-	if (IS_ERR(obj)) {
-		p_err("failed to open BPF object file: %ld", PTR_ERR(obj));
+	if (stat(file, &st)) {
+		p_err("failed to stat() %s: %s", file, strerror(errno));
 		return -1;
 	}
-
+	file_sz = st.st_size;
+	mmap_sz = roundup(file_sz, sysconf(_SC_PAGE_SIZE));
+	fd = open(file, O_RDONLY);
+	if (fd < 0) {
+		p_err("failed to open() %s: %s", file, strerror(errno));
+		return -1;
+	}
+	obj_data = mmap(NULL, mmap_sz, PROT_READ, MAP_PRIVATE, fd, 0);
+	if (obj_data == MAP_FAILED) {
+		obj_data = NULL;
+		p_err("failed to mmap() %s: %s", file, strerror(errno));
+		goto out;
+	}
 	get_obj_name(obj_name, file);
-	get_header_guard(header_guard, obj_name);
+	opts.object_name = obj_name;
+	obj = bpf_object__open_mem(obj_data, file_sz, &opts);
+	if (IS_ERR(obj)) {
+		obj = NULL;
+		p_err("failed to open BPF object file: %ld", PTR_ERR(obj));
+		goto out;
+	}
 
 	bpf_object__for_each_map(map, obj) {
 		ident = get_map_ident(map);
@@ -303,8 +323,11 @@ static int do_skeleton(int argc, char **argv)
 		prog_cnt++;
 	}
 
+	get_header_guard(header_guard, obj_name);
 	codegen("\
 		\n\
+		/* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */   \n\
+									    \n\
 		/* THIS FILE IS AUTOGENERATED! */			    \n\
 		#ifndef %2$s						    \n\
 		#define %2$s						    \n\
@@ -356,19 +379,95 @@ static int do_skeleton(int argc, char **argv)
 		\n\
 		};							    \n\
 									    \n\
-		static inline struct bpf_object_skeleton *		    \n\
-		%1$s__create_skeleton(struct %1$s *obj, struct bpf_embed_data *embed)\n\
+		static void						    \n\
+		%1$s__destroy(struct %1$s *obj)				    \n\
+		{							    \n\
+			if (!obj)					    \n\
+				return;					    \n\
+			if (obj->skeleton)				    \n\
+				bpf_object__destroy_skeleton(obj->skeleton);\n\
+			free(obj);					    \n\
+		}							    \n\
+									    \n\
+		static inline int					    \n\
+		%1$s__create_skeleton(struct %1$s *obj);		    \n\
+									    \n\
+		static inline struct %1$s *				    \n\
+		%1$s__open_opts(const struct bpf_object_open_opts *opts)    \n\
+		{							    \n\
+			struct %1$s *obj;				    \n\
+									    \n\
+			obj = calloc(1, sizeof(*obj));			    \n\
+			if (!obj)					    \n\
+				return NULL;				    \n\
+			if (%1$s__create_skeleton(obj))			    \n\
+				goto err;				    \n\
+			if (bpf_object__open_skeleton(obj->skeleton, opts)) \n\
+				goto err;				    \n\
+									    \n\
+			return obj;					    \n\
+		err:							    \n\
+			%1$s__destroy(obj);				    \n\
+			return NULL;					    \n\
+		}							    \n\
+									    \n\
+		static inline struct %1$s *				    \n\
+		%1$s__open(void)					    \n\
+		{							    \n\
+			return %1$s__open_opts(NULL);			    \n\
+		}							    \n\
+									    \n\
+		static inline int					    \n\
+		%1$s__load(struct %1$s *obj)				    \n\
+		{							    \n\
+			return bpf_object__load_skeleton(obj->skeleton);    \n\
+		}							    \n\
+									    \n\
+		static inline struct %1$s *				    \n\
+		%1$s__open_and_load(void)				    \n\
+		{							    \n\
+			struct %1$s *obj;				    \n\
+									    \n\
+			obj = %1$s__open();				    \n\
+			if (!obj)					    \n\
+				return NULL;				    \n\
+			if (%1$s__load(obj)) {				    \n\
+				%1$s__destroy(obj);			    \n\
+				return NULL;				    \n\
+			}						    \n\
+			return obj;					    \n\
+		}							    \n\
+									    \n\
+		static inline int					    \n\
+		%1$s__attach(struct %1$s *obj)				    \n\
+		{							    \n\
+			return bpf_object__attach_skeleton(obj->skeleton);  \n\
+		}							    \n\
+									    \n\
+		static inline void					    \n\
+		%1$s__detach(struct %1$s *obj)				    \n\
+		{							    \n\
+			return bpf_object__detach_skeleton(obj->skeleton);  \n\
+		}							    \n\
+		",
+		obj_name
+	);
+
+	codegen("\
+		\n\
+									    \n\
+		static inline int					    \n\
+		%1$s__create_skeleton(struct %1$s *obj)			    \n\
 		{							    \n\
 			struct bpf_object_skeleton *s;			    \n\
 									    \n\
 			s = calloc(1, sizeof(*s));			    \n\
 			if (!s)						    \n\
-				return NULL;				    \n\
+				return -1;				    \n\
+			obj->skeleton = s;				    \n\
 									    \n\
 			s->sz = sizeof(*s);				    \n\
 			s->name = \"%1$s\";				    \n\
-			s->data = embed->data;				    \n\
-			s->data_sz = embed->size;			    \n\
 			s->obj = &obj->obj;				    \n\
 		",
 		obj_name
@@ -438,90 +537,45 @@ static int do_skeleton(int argc, char **argv)
 	codegen("\
 		\n\
 									    \n\
-			return s;					    \n\
-		err:							    \n\
-			bpf_object__destroy_skeleton(s);		    \n\
-			return NULL;					    \n\
-		}							    \n\
-									    \n\
-		static void						    \n\
-		%1$s__destroy(struct %1$s *obj)				    \n\
-		{							    \n\
-			if (!obj)					    \n\
-				return;					    \n\
-			if (obj->skeleton)				    \n\
-				bpf_object__destroy_skeleton(obj->skeleton);\n\
-			free(obj);					    \n\
-		}							    \n\
-									    \n\
-		static inline struct %1$s *				    \n\
-		%1$s__open_opts(struct bpf_embed_data *embed, const struct bpf_object_open_opts *opts)\n\
-		{							    \n\
-			struct %1$s *obj;				    \n\
-									    \n\
-			obj = calloc(1, sizeof(*obj));			    \n\
-			if (!obj)					    \n\
-				return NULL;				    \n\
-									    \n\
-			obj->skeleton = %1$s__create_skeleton(obj, embed);  \n\
-			if (!obj->skeleton)				    \n\
-				goto err;				    \n\
-									    \n\
-			if (bpf_object__open_skeleton(obj->skeleton, opts)) \n\
-				goto err;				    \n\
+			s->data_sz = %d;				    \n\
+			s->data = \"\\					    \n\
+		",
+		file_sz);
+
+	/* embed contents of BPF object file */
+	for (i = 0, len = 0; i < file_sz; i++) {
+		int w = obj_data[i] ? 4 : 2;
+
+		len += w;
+		if (len > 78) {
+			printf("\\\n");
+			len = w;
+		}
+		if (!obj_data[i])
+			printf("\\0");
+		else
+			printf("\\x%02x", (unsigned char)obj_data[i]);
+	}
+
+	codegen("\
+		\n\
+		\";							    \n\
 									    \n\
-			return obj;					    \n\
+			return 0;					    \n\
 		err:							    \n\
-			%1$s__destroy(obj);				    \n\
-			return NULL;					    \n\
-		}							    \n\
-									    \n\
-		static inline struct %1$s *				    \n\
-		%1$s__open(struct bpf_embed_data *embed)		    \n\
-		{							    \n\
-			return %1$s__open_opts(embed, NULL);		    \n\
-		}							    \n\
-									    \n\
-		static inline int					    \n\
-		%1$s__load(struct %1$s *obj)				    \n\
-		{							    \n\
-			return bpf_object__load_skeleton(obj->skeleton);    \n\
-		}							    \n\
-									    \n\
-		static inline struct %1$s *				    \n\
-		%1$s__open_and_load(struct bpf_embed_data *embed)	    \n\
-		{							    \n\
-			struct %1$s *obj;				    \n\
-									    \n\
-			obj = %1$s__open(embed);			    \n\
-			if (!obj)					    \n\
-				return NULL;				    \n\
-			if (%1$s__load(obj)) {				    \n\
-				%1$s__destroy(obj);			    \n\
-				return NULL;				    \n\
-			}						    \n\
-			return obj;					    \n\
-		}							    \n\
-									    \n\
-		static inline int					    \n\
-		%1$s__attach(struct %1$s *obj)				    \n\
-		{							    \n\
-			return bpf_object__attach_skeleton(obj->skeleton);  \n\
-		}							    \n\
-									    \n\
-		static inline void					    \n\
-		%1$s__detach(struct %1$s *obj)				    \n\
-		{							    \n\
-			return bpf_object__detach_skeleton(obj->skeleton);  \n\
+			bpf_object__destroy_skeleton(s);		    \n\
+			return -1;					    \n\
 		}							    \n\
 									    \n\
 		#endif /* %2$s */					    \n\
 		",
-		obj_name, header_guard
-	);
+		obj_name, header_guard);
 	err = 0;
 out:
 	bpf_object__close(obj);
+	if (obj_data)
+		munmap(obj_data, mmap_sz);
+	close(fd);
 	return err;
 }
 
diff --git a/tools/testing/selftests/bpf/prog_tests/attach_probe.c b/tools/testing/selftests/bpf/prog_tests/attach_probe.c
index 60da1d08daa0..5ed90ede2f1d 100644
--- a/tools/testing/selftests/bpf/prog_tests/attach_probe.c
+++ b/tools/testing/selftests/bpf/prog_tests/attach_probe.c
@@ -22,8 +22,6 @@ ssize_t get_base_addr() {
 	return -EINVAL;
 }
 
-BPF_EMBED_OBJ(probe, "test_attach_probe.o");
-
 void test_attach_probe(void)
 {
 	int duration = 0;
@@ -39,7 +37,7 @@ void test_attach_probe(void)
 		return;
 	uprobe_offset = (size_t)&get_base_addr - base_addr;
 
-	skel = test_attach_probe__open_and_load(&probe_embed);
+	skel = test_attach_probe__open_and_load();
 	if (CHECK(!skel, "skel_open", "failed to open skeleton\n"))
 		return;
 	if (CHECK(!skel->bss, "check_bss", ".bss wasn't mmap()-ed\n"))
diff --git a/tools/testing/selftests/bpf/prog_tests/core_extern.c b/tools/testing/selftests/bpf/prog_tests/core_extern.c
index 30a7972e9012..5f03dc1de29e 100644
--- a/tools/testing/selftests/bpf/prog_tests/core_extern.c
+++ b/tools/testing/selftests/bpf/prog_tests/core_extern.c
@@ -124,8 +124,6 @@ static struct test_case {
 	{ .name = "u64 (max+1)", .fails = 1, .cfg = CFG"CONFIG_ULONG=0x10000000000000000" },
 };
 
-BPF_EMBED_OBJ(core_extern, "test_core_extern.o");
-
 void test_core_extern(void)
 {
 	const uint32_t kern_ver = get_kernel_version();
@@ -159,7 +157,7 @@ void test_core_extern(void)
 			opts.kconfig_path = tmp_cfg_path;
 		}
 
-		skel = test_core_extern__open_opts(&core_extern_embed, &opts);
+		skel = test_core_extern__open_opts(&opts);
 		if (CHECK(!skel, "skel_open", "skeleton open failed\n"))
 			goto cleanup;
 		err = test_core_extern__load(skel);
diff --git a/tools/testing/selftests/bpf/prog_tests/fentry_fexit.c b/tools/testing/selftests/bpf/prog_tests/fentry_fexit.c
index 110fcf053fd0..235ac4f67f5b 100644
--- a/tools/testing/selftests/bpf/prog_tests/fentry_fexit.c
+++ b/tools/testing/selftests/bpf/prog_tests/fentry_fexit.c
@@ -5,10 +5,6 @@
 #include "fentry_test.skel.h"
 #include "fexit_test.skel.h"
 
-BPF_EMBED_OBJ(pkt_access, "test_pkt_access.o");
-BPF_EMBED_OBJ(fentry, "fentry_test.o");
-BPF_EMBED_OBJ(fexit, "fexit_test.o");
-
 void test_fentry_fexit(void)
 {
 	struct test_pkt_access *pkt_skel = NULL;
@@ -18,13 +14,13 @@ void test_fentry_fexit(void)
 	__u32 duration = 0, retval;
 	int err, pkt_fd, i;
 
-	pkt_skel = test_pkt_access__open_and_load(&pkt_access_embed);
+	pkt_skel = test_pkt_access__open_and_load();
 	if (CHECK(!pkt_skel, "pkt_skel_load", "pkt_access skeleton failed\n"))
 		return;
-	fentry_skel = fentry_test__open_and_load(&fentry_embed);
+	fentry_skel = fentry_test__open_and_load();
 	if (CHECK(!fentry_skel, "fentry_skel_load", "fentry skeleton failed\n"))
 		goto close_prog;
-	fexit_skel = fexit_test__open_and_load(&fexit_embed);
+	fexit_skel = fexit_test__open_and_load();
 	if (CHECK(!fexit_skel, "fexit_skel_load", "fexit skeleton failed\n"))
 		goto close_prog;
 
diff --git a/tools/testing/selftests/bpf/prog_tests/fentry_test.c b/tools/testing/selftests/bpf/prog_tests/fentry_test.c
index 46a4afdf507a..e1a379f5f7d2 100644
--- a/tools/testing/selftests/bpf/prog_tests/fentry_test.c
+++ b/tools/testing/selftests/bpf/prog_tests/fentry_test.c
@@ -4,9 +4,6 @@
 #include "test_pkt_access.skel.h"
 #include "fentry_test.skel.h"
 
-BPF_EMBED_OBJ_DECLARE(pkt_access);
-BPF_EMBED_OBJ_DECLARE(fentry);
-
 void test_fentry_test(void)
 {
 	struct test_pkt_access *pkt_skel = NULL;
@@ -15,10 +12,10 @@ void test_fentry_test(void)
 	__u32 duration, retval;
 	__u64 *result;
 
-	pkt_skel = test_pkt_access__open_and_load(&pkt_access_embed);
+	pkt_skel = test_pkt_access__open_and_load();
 	if (CHECK(!pkt_skel, "pkt_skel_load", "pkt_access skeleton failed\n"))
 		return;
-	fentry_skel = fentry_test__open_and_load(&fentry_embed);
+	fentry_skel = fentry_test__open_and_load();
 	if (CHECK(!fentry_skel, "fentry_skel_load", "fentry skeleton failed\n"))
 		goto cleanup;
 
diff --git a/tools/testing/selftests/bpf/prog_tests/mmap.c b/tools/testing/selftests/bpf/prog_tests/mmap.c
index 95a44d37ccea..16a814eb4d64 100644
--- a/tools/testing/selftests/bpf/prog_tests/mmap.c
+++ b/tools/testing/selftests/bpf/prog_tests/mmap.c
@@ -13,8 +13,6 @@ static size_t roundup_page(size_t sz)
 	return (sz + page_size - 1) / page_size * page_size;
 }
 
-BPF_EMBED_OBJ(test_mmap, "test_mmap.o");
-
 void test_mmap(void)
 {
 	const size_t bss_sz = roundup_page(sizeof(struct test_mmap__bss));
@@ -30,7 +28,7 @@ void test_mmap(void)
 	__u64 val = 0;
 
 
-	skel = test_mmap__open_and_load(&test_mmap_embed);
+	skel = test_mmap__open_and_load();
 	if (CHECK(!skel, "skel_open_and_load", "skeleton open/load failed\n"))
 		return;
 
diff --git a/tools/testing/selftests/bpf/prog_tests/skeleton.c b/tools/testing/selftests/bpf/prog_tests/skeleton.c
index 151cdad3ad0d..ec6f2aec3853 100644
--- a/tools/testing/selftests/bpf/prog_tests/skeleton.c
+++ b/tools/testing/selftests/bpf/prog_tests/skeleton.c
@@ -10,8 +10,6 @@ struct s {
 
 #include "test_skeleton.skel.h"
 
-BPF_EMBED_OBJ(skeleton, "test_skeleton.o");
-
 void test_skeleton(void)
 {
 	int duration = 0, err;
@@ -19,7 +17,7 @@ void test_skeleton(void)
 	struct test_skeleton__bss *bss;
 	struct test_skeleton__externs *exts;
 
-	skel = test_skeleton__open(&skeleton_embed);
+	skel = test_skeleton__open();
 	if (CHECK(!skel, "skel_open", "failed to open skeleton\n"))
 		return;
 
diff --git a/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id.c b/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id.c
index 4af8b8253f25..e8399ae50e77 100644
--- a/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id.c
+++ b/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id.c
@@ -2,8 +2,6 @@
 #include <test_progs.h>
 #include "test_stacktrace_build_id.skel.h"
 
-BPF_EMBED_OBJ(stacktrace_build_id, "test_stacktrace_build_id.o");
-
 void test_stacktrace_build_id(void)
 {
 
@@ -18,7 +16,7 @@ void test_stacktrace_build_id(void)
 	int retry = 1;
 
 retry:
-	skel = test_stacktrace_build_id__open_and_load(&stacktrace_build_id_embed);
+	skel = test_stacktrace_build_id__open_and_load();
 	if (CHECK(!skel, "skel_open_and_load", "skeleton open/load failed\n"))
 		return;
 
diff --git a/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id_nmi.c b/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id_nmi.c
index 32fb03881a7b..8974450a4bdb 100644
--- a/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id_nmi.c
+++ b/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id_nmi.c
@@ -15,8 +15,6 @@ static __u64 read_perf_max_sample_freq(void)
 	return sample_freq;
 }
 
-BPF_EMBED_OBJ_DECLARE(stacktrace_build_id);
-
 void test_stacktrace_build_id_nmi(void)
 {
 	int control_map_fd, stackid_hmap_fd, stackmap_fd;
@@ -37,7 +35,7 @@ void test_stacktrace_build_id_nmi(void)
 	attr.sample_freq = read_perf_max_sample_freq();
 
 retry:
-	skel = test_stacktrace_build_id__open(&stacktrace_build_id_embed);
+	skel = test_stacktrace_build_id__open();
 	if (CHECK(!skel, "skel_open", "skeleton open failed\n"))
 		return;
 
-- 
cgit v1.2.3-59-g8ed1b


From 3bf3c79b772f4155312c000310abab700aba2200 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Tue, 17 Dec 2019 21:25:51 -0800
Subject: libbpf: Remove BPF_EMBED_OBJ macro from libbpf.h

Drop BPF_EMBED_OBJ and struct bpf_embed_data now that skeleton automatically
embeds contents of its source object file. While BPF_EMBED_OBJ is useful
independently of skeleton, we are currently don't have any use cases utilizing
it, so let's remove them until/if we need it.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20191218052552.2915188-3-andriin@fb.com
---
 tools/lib/bpf/libbpf.h | 35 -----------------------------------
 1 file changed, 35 deletions(-)

diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index 6340823871e2..f7084235bae9 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -600,41 +600,6 @@ bpf_program__bpil_offs_to_addr(struct bpf_prog_info_linear *info_linear);
  */
 LIBBPF_API int libbpf_num_possible_cpus(void);
 
-struct bpf_embed_data {
-	void *data;
-	size_t size;
-};
-
-#define BPF_EMBED_OBJ_DECLARE(NAME)					\
-extern struct bpf_embed_data NAME##_embed;				\
-extern char NAME##_data[];						\
-extern char NAME##_data_end[];
-
-#define __BPF_EMBED_OBJ(NAME, PATH, SZ, ASM_TYPE)			\
-asm (									\
-"	.pushsection \".rodata\", \"a\", @progbits		\n"	\
-"	.global "#NAME"_data					\n"	\
-#NAME"_data:							\n"	\
-"	.incbin \"" PATH "\"					\n"	\
-"	.global "#NAME"_data_end				\n"	\
-#NAME"_data_end:						\n"	\
-"	.global "#NAME"_embed					\n"	\
-"	.type "#NAME"_embed, @object				\n"	\
-"	.size "#NAME"_size, "#SZ"				\n"	\
-"	.align 8,						\n"	\
-#NAME"_embed:							\n"	\
-"	"ASM_TYPE" "#NAME"_data					\n"	\
-"	"ASM_TYPE" "#NAME"_data_end - "#NAME"_data 		\n"	\
-"	.popsection						\n"	\
-);									\
-BPF_EMBED_OBJ_DECLARE(NAME)
-
-#if __SIZEOF_POINTER__ == 4
-#define BPF_EMBED_OBJ(NAME, PATH) __BPF_EMBED_OBJ(NAME, PATH, 8, ".long")
-#else
-#define BPF_EMBED_OBJ(NAME, PATH) __BPF_EMBED_OBJ(NAME, PATH, 16, ".quad")
-#endif
-
 struct bpf_map_skeleton {
 	const char *name;
 	struct bpf_map **map;
-- 
cgit v1.2.3-59-g8ed1b


From cb21ac58854605e6c15fd7da2c619823967b3140 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Tue, 17 Dec 2019 21:25:52 -0800
Subject: bpftool: Add gen subcommand manpage

Add bpftool-gen.rst describing skeleton on the high level. Also include
a small, but complete, example BPF app (BPF side, userspace side, generated
skeleton) in example section to demonstrate skeleton API and its usage.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20191218052552.2915188-4-andriin@fb.com
---
 tools/bpf/bpftool/Documentation/bpftool-gen.rst | 304 ++++++++++++++++++++++++
 tools/bpf/bpftool/Documentation/bpftool.rst     |   3 +-
 2 files changed, 306 insertions(+), 1 deletion(-)
 create mode 100644 tools/bpf/bpftool/Documentation/bpftool-gen.rst

diff --git a/tools/bpf/bpftool/Documentation/bpftool-gen.rst b/tools/bpf/bpftool/Documentation/bpftool-gen.rst
new file mode 100644
index 000000000000..b6a114bf908d
--- /dev/null
+++ b/tools/bpf/bpftool/Documentation/bpftool-gen.rst
@@ -0,0 +1,304 @@
+================
+bpftool-gen
+================
+-------------------------------------------------------------------------------
+tool for BPF code-generation
+-------------------------------------------------------------------------------
+
+:Manual section: 8
+
+SYNOPSIS
+========
+
+	**bpftool** [*OPTIONS*] **gen** *COMMAND*
+
+	*OPTIONS* := { { **-j** | **--json** } [{ **-p** | **--pretty** }] }
+
+	*COMMAND* := { **skeleton | **help** }
+
+GEN COMMANDS
+=============
+
+|	**bpftool** **gen skeleton** *FILE*
+|	**bpftool** **gen help**
+
+DESCRIPTION
+===========
+	**bpftool gen skeleton** *FILE*
+		  Generate BPF skeleton C header file for a given *FILE*.
+
+		  BPF skeleton is an alternative interface to existing libbpf
+		  APIs for working with BPF objects. Skeleton code is intended
+		  to significantly shorten and simplify code to load and work
+		  with BPF programs from userspace side. Generated code is
+		  tailored to specific input BPF object *FILE*, reflecting its
+		  structure by listing out available maps, program, variables,
+		  etc. Skeleton eliminates the need to lookup mentioned
+		  components by name. Instead, if skeleton instantiation
+		  succeeds, they are populated in skeleton structure as valid
+		  libbpf types (e.g., struct bpf_map pointer) and can be
+		  passed to existing generic libbpf APIs.
+
+		  In addition to simple and reliable access to maps and
+		  programs, skeleton provides a storage for BPF links (struct
+		  bpf_link) for each BPF program within BPF object. When
+		  requested, supported BPF programs will be automatically
+		  attached and resulting BPF links stored for further use by
+		  user in pre-allocated fields in skeleton struct. For BPF
+		  programs that can't be automatically attached by libbpf,
+		  user can attach them manually, but store resulting BPF link
+		  in per-program link field. All such set up links will be
+		  automatically destroyed on BPF skeleton destruction. This
+		  eliminates the need for users to manage links manually and
+		  rely on libbpf support to detach programs and free up
+		  resources.
+
+		  Another facility provided by BPF skeleton is an interface to
+		  global variables of all supported kinds: mutable, read-only,
+		  as well as extern ones. This interface allows to pre-setup
+		  initial values of variables before BPF object is loaded and
+		  verified by kernel. For non-read-only variables, the same
+		  interface can be used to fetch values of global variables on
+		  userspace side, even if they are modified by BPF code.
+
+		  During skeleton generation, contents of source BPF object
+		  *FILE* is embedded within generated code and is thus not
+		  necessary to keep around. This ensures skeleton and BPF
+		  object file are matching 1-to-1 and always stay in sync.
+		  Generated code is dual-licensed under LGPL-2.1 and
+		  BSD-2-Clause licenses.
+
+		  It is a design goal and guarantee that skeleton interfaces
+		  are interoperable with generic libbpf APIs. User should
+		  always be able to use skeleton API to create and load BPF
+		  object, and later use libbpf APIs to keep working with
+		  specific maps, programs, etc.
+
+		  As part of skeleton, few custom functions are generated.
+		  Each of them is prefixed with object name, derived from
+		  object file name. I.e., if BPF object file name is
+		  **example.o**, BPF object name will be **example**. The
+		  following custom functions are provided in such case:
+
+		  - **example__open** and **example__open_opts**.
+		    These functions are used to instantiate skeleton. It
+		    corresponds to libbpf's **bpf_object__open()** API.
+		    **_opts** variants accepts extra **bpf_object_open_opts**
+		    options.
+
+		  - **example__load**.
+		    This function creates maps, loads and verifies BPF
+		    programs, initializes global data maps. It corresponds to
+		    libppf's **bpf_object__load** API.
+
+		  - **example__open_and_load** combines **example__open** and
+		    **example__load** invocations in one commonly used
+		    operation.
+
+		  - **example__attach** and **example__detach**
+		    This pair of functions allow to attach and detach,
+		    correspondingly, already loaded BPF object. Only BPF
+		    programs of types supported by libbpf for auto-attachment
+		    will be auto-attached and their corresponding BPF links
+		    instantiated. For other BPF programs, user can manually
+		    create a BPF link and assign it to corresponding fields in
+		    skeleton struct. **example__detach** will detach both
+		    links created automatically, as well as those populated by
+		    user manually.
+
+		  - **example__destroy**
+		    Detach and unload BPF programs, free up all the resources
+		    used by skeleton and BPF object.
+
+		  If BPF object has global variables, corresponding structs
+		  with memory layout corresponding to global data data section
+		  layout will be created. Currently supported ones are: .data,
+		  .bss, .rodata, and .extern structs/data sections. These
+		  data sections/structs can be used to set up initial values of
+		  variables, if set before **example__load**. Afterwards, if
+		  target kernel supports memory-mapped BPF arrays, same
+		  structs can be used to fetch and update (non-read-only)
+		  data from userspace, with same simplicity as for BPF side.
+
+	**bpftool gen help**
+		  Print short help message.
+
+OPTIONS
+=======
+	-h, --help
+		  Print short generic help message (similar to **bpftool help**).
+
+	-V, --version
+		  Print version number (similar to **bpftool version**).
+
+	-j, --json
+		  Generate JSON output. For commands that cannot produce JSON,
+		  this option has no effect.
+
+	-p, --pretty
+		  Generate human-readable JSON output. Implies **-j**.
+
+	-d, --debug
+		  Print all logs available from libbpf, including debug-level
+		  information.
+
+EXAMPLES
+========
+**$ cat example.c**
+::
+
+  #include <stdbool.h>
+  #include <linux/ptrace.h>
+  #include <linux/bpf.h>
+  #include "bpf_helpers.h"
+
+  const volatile int param1 = 42;
+  bool global_flag = true;
+  struct { int x; } data = {};
+
+  struct {
+  	__uint(type, BPF_MAP_TYPE_HASH);
+  	__uint(max_entries, 128);
+  	__type(key, int);
+  	__type(value, long);
+  } my_map SEC(".maps");
+
+  SEC("raw_tp/sys_enter")
+  int handle_sys_enter(struct pt_regs *ctx)
+  {
+  	static long my_static_var;
+  	if (global_flag)
+  		my_static_var++;
+  	else
+  		data.x += param1;
+  	return 0;
+  }
+
+  SEC("raw_tp/sys_exit")
+  int handle_sys_exit(struct pt_regs *ctx)
+  {
+  	int zero = 0;
+  	bpf_map_lookup_elem(&my_map, &zero);
+  	return 0;
+  }
+
+This is example BPF application with two BPF programs and a mix of BPF maps
+and global variables.
+
+**$ bpftool gen skeleton example.o**
+::
+
+  /* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */
+
+  /* THIS FILE IS AUTOGENERATED! */
+  #ifndef __EXAMPLE_SKEL_H__
+  #define __EXAMPLE_SKEL_H__
+
+  #include <stdlib.h>
+  #include <libbpf.h>
+
+  struct example {
+  	struct bpf_object_skeleton *skeleton;
+  	struct bpf_object *obj;
+  	struct {
+  		struct bpf_map *rodata;
+  		struct bpf_map *data;
+  		struct bpf_map *bss;
+  		struct bpf_map *my_map;
+  	} maps;
+  	struct {
+  		struct bpf_program *handle_sys_enter;
+  		struct bpf_program *handle_sys_exit;
+  	} progs;
+  	struct {
+  		struct bpf_link *handle_sys_enter;
+  		struct bpf_link *handle_sys_exit;
+  	} links;
+  	struct example__bss {
+  		struct {
+  			int x;
+  		} data;
+  	} *bss;
+  	struct example__data {
+  		_Bool global_flag;
+  		long int handle_sys_enter_my_static_var;
+  	} *data;
+  	struct example__rodata {
+  		int param1;
+  	} *rodata;
+  };
+
+  static void example__destroy(struct example *obj);
+  static inline struct example *example__open_opts(
+                const struct bpf_object_open_opts *opts);
+  static inline struct example *example__open();
+  static inline int example__load(struct example *obj);
+  static inline struct example *example__open_and_load();
+  static inline int example__attach(struct example *obj);
+  static inline void example__detach(struct example *obj);
+
+  #endif /* __EXAMPLE_SKEL_H__ */
+
+**$ cat example_user.c**
+::
+
+  #include "example.skel.h"
+
+  int main()
+  {
+  	struct example *skel;
+  	int err = 0;
+
+  	skel = example__open();
+  	if (!skel)
+  		goto cleanup;
+
+  	skel->rodata->param1 = 128;
+
+  	err = example__load(skel);
+  	if (err)
+  		goto cleanup;
+
+  	err = example__attach(skel);
+  	if (err)
+  		goto cleanup;
+
+  	/* all libbpf APIs are usable */
+  	printf("my_map name: %s\n", bpf_map__name(skel->maps.my_map));
+  	printf("sys_enter prog FD: %d\n",
+  	       bpf_program__fd(skel->progs.handle_sys_enter));
+
+  	/* detach and re-attach sys_exit program */
+  	bpf_link__destroy(skel->links.handle_sys_exit);
+  	skel->links.handle_sys_exit =
+  		bpf_program__attach(skel->progs.handle_sys_exit);
+
+  	printf("my_static_var: %ld\n",
+  	       skel->bss->handle_sys_enter_my_static_var);
+
+  cleanup:
+  	example__destroy(skel);
+  	return err;
+  }
+
+**# ./example_user**
+::
+
+  my_map name: my_map
+  sys_enter prog FD: 8
+  my_static_var: 7
+
+This is a stripped-out version of skeleton generated for above example code.
+
+SEE ALSO
+========
+	**bpf**\ (2),
+	**bpf-helpers**\ (7),
+	**bpftool**\ (8),
+	**bpftool-map**\ (8),
+	**bpftool-prog**\ (8),
+	**bpftool-cgroup**\ (8),
+	**bpftool-feature**\ (8),
+	**bpftool-net**\ (8),
+	**bpftool-perf**\ (8),
+	**bpftool-btf**\ (8)
diff --git a/tools/bpf/bpftool/Documentation/bpftool.rst b/tools/bpf/bpftool/Documentation/bpftool.rst
index 6a9c52ef84a9..34239fda69ed 100644
--- a/tools/bpf/bpftool/Documentation/bpftool.rst
+++ b/tools/bpf/bpftool/Documentation/bpftool.rst
@@ -81,4 +81,5 @@ SEE ALSO
 	**bpftool-feature**\ (8),
 	**bpftool-net**\ (8),
 	**bpftool-perf**\ (8),
-	**bpftool-btf**\ (8)
+	**bpftool-btf**\ (8),
+	**bpftool-gen**\ (8),
-- 
cgit v1.2.3-59-g8ed1b


From 3646ae0de90eb271ef48ec1fa75e63873ab20fc6 Mon Sep 17 00:00:00 2001
From: Shahjada Abul Husain <shahjada@chelsio.com>
Date: Tue, 17 Dec 2019 12:12:08 +0530
Subject: cxgb4: fix missed high priority region calculation

commit c21939998802 ("cxgb4: add support for high priority filters")
has missed considering high priority region calculation in some code
paths. This patch fixes them.

Fixes: c21939998802 ("cxgb4: add support for high priority filters")
Signed-off-by: Shahjada Abul Husain <shahjada@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c   | 21 +++++++++++++++------
 .../net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c    |  3 ++-
 .../net/ethernet/chelsio/cxgb4/cxgb4_tc_matchall.c  |  2 +-
 3 files changed, 18 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c
index 6a5b0346dad9..44deabfa32b5 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c
@@ -369,12 +369,16 @@ static int get_filter_count(struct adapter *adapter, unsigned int fidx,
 			return -E2BIG;
 		}
 	} else {
-		if ((fidx != (adapter->tids.nftids +
-			      adapter->tids.nsftids - 1)) &&
-		    fidx >= adapter->tids.nftids)
+		if ((fidx != (adapter->tids.nftids + adapter->tids.nsftids +
+			      adapter->tids.nhpftids - 1)) &&
+		    fidx >= (adapter->tids.nftids + adapter->tids.nhpftids))
 			return -E2BIG;
 
-		f = &adapter->tids.ftid_tab[fidx];
+		if (fidx < adapter->tids.nhpftids)
+			f = &adapter->tids.hpftid_tab[fidx];
+		else
+			f = &adapter->tids.ftid_tab[fidx -
+						    adapter->tids.nhpftids];
 		if (!f->valid)
 			return -EINVAL;
 	}
@@ -480,6 +484,7 @@ int cxgb4_get_free_ftid(struct net_device *dev, int family)
 		ftid -= n;
 	}
 	spin_unlock_bh(&t->ftid_lock);
+	ftid += t->nhpftids;
 
 	return found ? ftid : -ENOMEM;
 }
@@ -815,10 +820,14 @@ int delete_filter(struct adapter *adapter, unsigned int fidx)
 	struct filter_entry *f;
 	int ret;
 
-	if (fidx >= adapter->tids.nftids + adapter->tids.nsftids)
+	if (fidx >= adapter->tids.nftids + adapter->tids.nsftids +
+		    adapter->tids.nhpftids)
 		return -EINVAL;
 
-	f = &adapter->tids.ftid_tab[fidx];
+	if (fidx < adapter->tids.nhpftids)
+		f = &adapter->tids.hpftid_tab[fidx];
+	else
+		f = &adapter->tids.ftid_tab[fidx - adapter->tids.nhpftids];
 	ret = writable_filter(f);
 	if (ret)
 		return ret;
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c
index 706b71e1b6c1..bb5513bdd293 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c
@@ -672,7 +672,8 @@ int cxgb4_tc_flower_replace(struct net_device *dev,
 		 * 0 to driver. However, the hardware TCAM index
 		 * starts from 0. Hence, the -1 here.
 		 */
-		if (cls->common.prio <= adap->tids.nftids) {
+		if (cls->common.prio <= (adap->tids.nftids +
+					 adap->tids.nhpftids)) {
 			fidx = cls->common.prio - 1;
 			if (fidx < adap->tids.nhpftids)
 				fs->prio = 1;
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_matchall.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_matchall.c
index 53e8db8b2546..24c3c2dc7171 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_matchall.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_matchall.c
@@ -137,7 +137,7 @@ static int cxgb4_matchall_alloc_filter(struct net_device *dev,
 	 * -1 here. 1 slot is enough to create a wildcard matchall
 	 * VIID rule.
 	 */
-	if (cls->common.prio <= adap->tids.nftids)
+	if (cls->common.prio <= (adap->tids.nftids + adap->tids.nhpftids))
 		fidx = cls->common.prio - 1;
 	else
 		fidx = cxgb4_get_free_ftid(dev, PF_INET);
-- 
cgit v1.2.3-59-g8ed1b


From 59437d78f088e506cabdd5e0f0601c531a9830b4 Mon Sep 17 00:00:00 2001
From: Shahjada Abul Husain <shahjada@chelsio.com>
Date: Tue, 17 Dec 2019 12:12:09 +0530
Subject: cxgb4/chtls: fix ULD connection failures due to wrong TID base

Currently, the hardware TID index is assumed to start from index 0.
However, with the following changeset,

commit c21939998802 ("cxgb4: add support for high priority filters")

hardware TID index can start after the high priority region, which
has introduced a regression resulting in connection failures for
ULDs.

So, fix all related code to properly recalculate the TID start index
based on whether high priority filters are enabled or not.

Fixes: c21939998802 ("cxgb4: add support for high priority filters")
Signed-off-by: Shahjada Abul Husain <shahjada@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/crypto/chelsio/chtls/chtls_cm.c            |  2 +-
 drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c | 22 ++++++++++------------
 drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c  | 12 +++++-------
 drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c    | 13 ++++++++-----
 drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h     |  9 ++++++++-
 5 files changed, 32 insertions(+), 26 deletions(-)

diff --git a/drivers/crypto/chelsio/chtls/chtls_cm.c b/drivers/crypto/chelsio/chtls/chtls_cm.c
index aca75237bbcf..72e5b0f65a91 100644
--- a/drivers/crypto/chelsio/chtls/chtls_cm.c
+++ b/drivers/crypto/chelsio/chtls/chtls_cm.c
@@ -1273,7 +1273,7 @@ static int chtls_pass_accept_req(struct chtls_dev *cdev, struct sk_buff *skb)
 	ctx = (struct listen_ctx *)data;
 	lsk = ctx->lsk;
 
-	if (unlikely(tid >= cdev->tids->ntids)) {
+	if (unlikely(tid_out_of_range(cdev->tids, tid))) {
 		pr_info("passive open TID %u too large\n", tid);
 		return 1;
 	}
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c
index b6a79536e143..9a96b01dad56 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c
@@ -3171,14 +3171,12 @@ static const struct file_operations mem_debugfs_fops = {
 
 static int tid_info_show(struct seq_file *seq, void *v)
 {
-	unsigned int tid_start = 0;
 	struct adapter *adap = seq->private;
-	const struct tid_info *t = &adap->tids;
-	enum chip_type chip = CHELSIO_CHIP_VERSION(adap->params.chip);
-
-	if (chip > CHELSIO_T5)
-		tid_start = t4_read_reg(adap, LE_DB_ACTIVE_TABLE_START_INDEX_A);
+	const struct tid_info *t;
+	enum chip_type chip;
 
+	t = &adap->tids;
+	chip = CHELSIO_CHIP_VERSION(adap->params.chip);
 	if (t4_read_reg(adap, LE_DB_CONFIG_A) & HASHEN_F) {
 		unsigned int sb;
 		seq_printf(seq, "Connections in use: %u\n",
@@ -3190,9 +3188,9 @@ static int tid_info_show(struct seq_file *seq, void *v)
 			sb = t4_read_reg(adap, LE_DB_SRVR_START_INDEX_A);
 
 		if (sb) {
-			seq_printf(seq, "TID range: %u..%u/%u..%u", tid_start,
+			seq_printf(seq, "TID range: %u..%u/%u..%u", t->tid_base,
 				   sb - 1, adap->tids.hash_base,
-				   t->ntids - 1);
+				   t->tid_base + t->ntids - 1);
 			seq_printf(seq, ", in use: %u/%u\n",
 				   atomic_read(&t->tids_in_use),
 				   atomic_read(&t->hash_tids_in_use));
@@ -3201,14 +3199,14 @@ static int tid_info_show(struct seq_file *seq, void *v)
 				   t->aftid_base,
 				   t->aftid_end,
 				   adap->tids.hash_base,
-				   t->ntids - 1);
+				   t->tid_base + t->ntids - 1);
 			seq_printf(seq, ", in use: %u/%u\n",
 				   atomic_read(&t->tids_in_use),
 				   atomic_read(&t->hash_tids_in_use));
 		} else {
 			seq_printf(seq, "TID range: %u..%u",
 				   adap->tids.hash_base,
-				   t->ntids - 1);
+				   t->tid_base + t->ntids - 1);
 			seq_printf(seq, ", in use: %u\n",
 				   atomic_read(&t->hash_tids_in_use));
 		}
@@ -3216,8 +3214,8 @@ static int tid_info_show(struct seq_file *seq, void *v)
 		seq_printf(seq, "Connections in use: %u\n",
 			   atomic_read(&t->conns_in_use));
 
-		seq_printf(seq, "TID range: %u..%u", tid_start,
-			   tid_start + t->ntids - 1);
+		seq_printf(seq, "TID range: %u..%u", t->tid_base,
+			   t->tid_base + t->ntids - 1);
 		seq_printf(seq, ", in use: %u\n",
 			   atomic_read(&t->tids_in_use));
 	}
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c
index 44deabfa32b5..2a2938bbb93a 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c
@@ -361,13 +361,11 @@ static int get_filter_count(struct adapter *adapter, unsigned int fidx,
 
 	tcb_base = t4_read_reg(adapter, TP_CMM_TCB_BASE_A);
 	if (is_hashfilter(adapter) && hash) {
-		if (fidx < adapter->tids.ntids) {
-			f = adapter->tids.tid_tab[fidx];
-			if (!f)
-				return -EINVAL;
-		} else {
+		if (tid_out_of_range(&adapter->tids, fidx))
 			return -E2BIG;
-		}
+		f = adapter->tids.tid_tab[fidx - adapter->tids.tid_base];
+		if (!f)
+			return -EINVAL;
 	} else {
 		if ((fidx != (adapter->tids.nftids + adapter->tids.nsftids +
 			      adapter->tids.nhpftids - 1)) &&
@@ -1615,7 +1613,7 @@ static int cxgb4_del_hash_filter(struct net_device *dev, int filter_id,
 	netdev_dbg(dev, "%s: filter_id = %d ; nftids = %d\n",
 		   __func__, filter_id, adapter->tids.nftids);
 
-	if (filter_id > adapter->tids.ntids)
+	if (tid_out_of_range(t, filter_id))
 		return -E2BIG;
 
 	f = lookup_tid(t, filter_id);
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
index be750f2de23c..1930e39f195e 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
@@ -1447,8 +1447,8 @@ static void mk_tid_release(struct sk_buff *skb, unsigned int chan,
 static void cxgb4_queue_tid_release(struct tid_info *t, unsigned int chan,
 				    unsigned int tid)
 {
-	void **p = &t->tid_tab[tid];
 	struct adapter *adap = container_of(t, struct adapter, tids);
+	void **p = &t->tid_tab[tid - t->tid_base];
 
 	spin_lock_bh(&adap->tid_release_lock);
 	*p = adap->tid_release_head;
@@ -1500,13 +1500,13 @@ static void process_tid_release_list(struct work_struct *work)
 void cxgb4_remove_tid(struct tid_info *t, unsigned int chan, unsigned int tid,
 		      unsigned short family)
 {
-	struct sk_buff *skb;
 	struct adapter *adap = container_of(t, struct adapter, tids);
+	struct sk_buff *skb;
 
-	WARN_ON(tid >= t->ntids);
+	WARN_ON(tid_out_of_range(&adap->tids, tid));
 
-	if (t->tid_tab[tid]) {
-		t->tid_tab[tid] = NULL;
+	if (t->tid_tab[tid - adap->tids.tid_base]) {
+		t->tid_tab[tid - adap->tids.tid_base] = NULL;
 		atomic_dec(&t->conns_in_use);
 		if (t->hash_base && (tid >= t->hash_base)) {
 			if (family == AF_INET6)
@@ -4727,6 +4727,9 @@ static int adap_init0(struct adapter *adap, int vpd_skip)
 			adap->rawf_start = val[0];
 			adap->rawf_cnt = val[1] - val[0] + 1;
 		}
+
+		adap->tids.tid_base =
+			t4_read_reg(adap, LE_DB_ACTIVE_TABLE_START_INDEX_A);
 	}
 
 	/* qids (ingress/egress) returned from firmware can be anywhere
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h
index e69de9a296ae..d9d27bc1ae67 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h
@@ -99,6 +99,7 @@ struct eotid_entry {
  */
 struct tid_info {
 	void **tid_tab;
+	unsigned int tid_base;
 	unsigned int ntids;
 
 	struct serv_entry *stid_tab;
@@ -152,9 +153,15 @@ struct tid_info {
 
 static inline void *lookup_tid(const struct tid_info *t, unsigned int tid)
 {
+	tid -= t->tid_base;
 	return tid < t->ntids ? t->tid_tab[tid] : NULL;
 }
 
+static inline bool tid_out_of_range(const struct tid_info *t, unsigned int tid)
+{
+	return ((tid - t->tid_base) >= t->ntids);
+}
+
 static inline void *lookup_atid(const struct tid_info *t, unsigned int atid)
 {
 	return atid < t->natids ? t->atid_tab[atid].data : NULL;
@@ -176,7 +183,7 @@ static inline void *lookup_stid(const struct tid_info *t, unsigned int stid)
 static inline void cxgb4_insert_tid(struct tid_info *t, void *data,
 				    unsigned int tid, unsigned short family)
 {
-	t->tid_tab[tid] = data;
+	t->tid_tab[tid - t->tid_base] = data;
 	if (t->hash_base && (tid >= t->hash_base)) {
 		if (family == AF_INET6)
 			atomic_add(2, &t->hash_tids_in_use);
-- 
cgit v1.2.3-59-g8ed1b


From 92fbeb43b6c44fb03e024c4c5b9f3d6b8ebd0cf0 Mon Sep 17 00:00:00 2001
From: Paul Durrant <pdurrant@amazon.com>
Date: Tue, 17 Dec 2019 13:32:16 +0000
Subject: xen-netback: move netback_probe() and netback_remove() to the end...

...of xenbus.c

This is a cosmetic function re-ordering to reduce churn in a subsequent
patch. Some style fix-up was done to make checkpatch.pl happier.

No functional change.

Signed-off-by: Paul Durrant <pdurrant@amazon.com>
Acked-by: Wei Liu <wei.liu@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/xen-netback/xenbus.c | 353 +++++++++++++++++++--------------------
 1 file changed, 174 insertions(+), 179 deletions(-)

diff --git a/drivers/net/xen-netback/xenbus.c b/drivers/net/xen-netback/xenbus.c
index f533b7372d59..bb61316d79de 100644
--- a/drivers/net/xen-netback/xenbus.c
+++ b/drivers/net/xen-netback/xenbus.c
@@ -195,185 +195,6 @@ static void xenvif_debugfs_delif(struct xenvif *vif)
 }
 #endif /* CONFIG_DEBUG_FS */
 
-static int netback_remove(struct xenbus_device *dev)
-{
-	struct backend_info *be = dev_get_drvdata(&dev->dev);
-
-	set_backend_state(be, XenbusStateClosed);
-
-	unregister_hotplug_status_watch(be);
-	if (be->vif) {
-		kobject_uevent(&dev->dev.kobj, KOBJ_OFFLINE);
-		xen_unregister_watchers(be->vif);
-		xenbus_rm(XBT_NIL, dev->nodename, "hotplug-status");
-		xenvif_free(be->vif);
-		be->vif = NULL;
-	}
-	kfree(be->hotplug_script);
-	kfree(be);
-	dev_set_drvdata(&dev->dev, NULL);
-	return 0;
-}
-
-
-/**
- * Entry point to this code when a new device is created.  Allocate the basic
- * structures and switch to InitWait.
- */
-static int netback_probe(struct xenbus_device *dev,
-			 const struct xenbus_device_id *id)
-{
-	const char *message;
-	struct xenbus_transaction xbt;
-	int err;
-	int sg;
-	const char *script;
-	struct backend_info *be = kzalloc(sizeof(struct backend_info),
-					  GFP_KERNEL);
-	if (!be) {
-		xenbus_dev_fatal(dev, -ENOMEM,
-				 "allocating backend structure");
-		return -ENOMEM;
-	}
-
-	be->dev = dev;
-	dev_set_drvdata(&dev->dev, be);
-
-	be->state = XenbusStateInitialising;
-	err = xenbus_switch_state(dev, XenbusStateInitialising);
-	if (err)
-		goto fail;
-
-	sg = 1;
-
-	do {
-		err = xenbus_transaction_start(&xbt);
-		if (err) {
-			xenbus_dev_fatal(dev, err, "starting transaction");
-			goto fail;
-		}
-
-		err = xenbus_printf(xbt, dev->nodename, "feature-sg", "%d", sg);
-		if (err) {
-			message = "writing feature-sg";
-			goto abort_transaction;
-		}
-
-		err = xenbus_printf(xbt, dev->nodename, "feature-gso-tcpv4",
-				    "%d", sg);
-		if (err) {
-			message = "writing feature-gso-tcpv4";
-			goto abort_transaction;
-		}
-
-		err = xenbus_printf(xbt, dev->nodename, "feature-gso-tcpv6",
-				    "%d", sg);
-		if (err) {
-			message = "writing feature-gso-tcpv6";
-			goto abort_transaction;
-		}
-
-		/* We support partial checksum setup for IPv6 packets */
-		err = xenbus_printf(xbt, dev->nodename,
-				    "feature-ipv6-csum-offload",
-				    "%d", 1);
-		if (err) {
-			message = "writing feature-ipv6-csum-offload";
-			goto abort_transaction;
-		}
-
-		/* We support rx-copy path. */
-		err = xenbus_printf(xbt, dev->nodename,
-				    "feature-rx-copy", "%d", 1);
-		if (err) {
-			message = "writing feature-rx-copy";
-			goto abort_transaction;
-		}
-
-		/*
-		 * We don't support rx-flip path (except old guests who don't
-		 * grok this feature flag).
-		 */
-		err = xenbus_printf(xbt, dev->nodename,
-				    "feature-rx-flip", "%d", 0);
-		if (err) {
-			message = "writing feature-rx-flip";
-			goto abort_transaction;
-		}
-
-		/* We support dynamic multicast-control. */
-		err = xenbus_printf(xbt, dev->nodename,
-				    "feature-multicast-control", "%d", 1);
-		if (err) {
-			message = "writing feature-multicast-control";
-			goto abort_transaction;
-		}
-
-		err = xenbus_printf(xbt, dev->nodename,
-				    "feature-dynamic-multicast-control",
-				    "%d", 1);
-		if (err) {
-			message = "writing feature-dynamic-multicast-control";
-			goto abort_transaction;
-		}
-
-		err = xenbus_transaction_end(xbt, 0);
-	} while (err == -EAGAIN);
-
-	if (err) {
-		xenbus_dev_fatal(dev, err, "completing transaction");
-		goto fail;
-	}
-
-	/*
-	 * Split event channels support, this is optional so it is not
-	 * put inside the above loop.
-	 */
-	err = xenbus_printf(XBT_NIL, dev->nodename,
-			    "feature-split-event-channels",
-			    "%u", separate_tx_rx_irq);
-	if (err)
-		pr_debug("Error writing feature-split-event-channels\n");
-
-	/* Multi-queue support: This is an optional feature. */
-	err = xenbus_printf(XBT_NIL, dev->nodename,
-			    "multi-queue-max-queues", "%u", xenvif_max_queues);
-	if (err)
-		pr_debug("Error writing multi-queue-max-queues\n");
-
-	err = xenbus_printf(XBT_NIL, dev->nodename,
-			    "feature-ctrl-ring",
-			    "%u", true);
-	if (err)
-		pr_debug("Error writing feature-ctrl-ring\n");
-
-	script = xenbus_read(XBT_NIL, dev->nodename, "script", NULL);
-	if (IS_ERR(script)) {
-		err = PTR_ERR(script);
-		xenbus_dev_fatal(dev, err, "reading script");
-		goto fail;
-	}
-
-	be->hotplug_script = script;
-
-
-	/* This kicks hotplug scripts, so do it immediately. */
-	err = backend_create_xenvif(be);
-	if (err)
-		goto fail;
-
-	return 0;
-
-abort_transaction:
-	xenbus_transaction_end(xbt, 1);
-	xenbus_dev_fatal(dev, err, "%s", message);
-fail:
-	pr_debug("failed\n");
-	netback_remove(dev);
-	return err;
-}
-
-
 /*
  * Handle the creation of the hotplug script environment.  We add the script
  * and vif variables to the environment, for the benefit of the vif-* hotplug
@@ -1128,6 +949,180 @@ static int read_xenbus_vif_flags(struct backend_info *be)
 	return 0;
 }
 
+static int netback_remove(struct xenbus_device *dev)
+{
+	struct backend_info *be = dev_get_drvdata(&dev->dev);
+
+	set_backend_state(be, XenbusStateClosed);
+
+	unregister_hotplug_status_watch(be);
+	if (be->vif) {
+		kobject_uevent(&dev->dev.kobj, KOBJ_OFFLINE);
+		xen_unregister_watchers(be->vif);
+		xenbus_rm(XBT_NIL, dev->nodename, "hotplug-status");
+		xenvif_free(be->vif);
+		be->vif = NULL;
+	}
+	kfree(be->hotplug_script);
+	kfree(be);
+	dev_set_drvdata(&dev->dev, NULL);
+	return 0;
+}
+
+/**
+ * Entry point to this code when a new device is created.  Allocate the basic
+ * structures and switch to InitWait.
+ */
+static int netback_probe(struct xenbus_device *dev,
+			 const struct xenbus_device_id *id)
+{
+	const char *message;
+	struct xenbus_transaction xbt;
+	int err;
+	int sg;
+	const char *script;
+	struct backend_info *be = kzalloc(sizeof(*be), GFP_KERNEL);
+
+	if (!be) {
+		xenbus_dev_fatal(dev, -ENOMEM,
+				 "allocating backend structure");
+		return -ENOMEM;
+	}
+
+	be->dev = dev;
+	dev_set_drvdata(&dev->dev, be);
+
+	be->state = XenbusStateInitialising;
+	err = xenbus_switch_state(dev, XenbusStateInitialising);
+	if (err)
+		goto fail;
+
+	sg = 1;
+
+	do {
+		err = xenbus_transaction_start(&xbt);
+		if (err) {
+			xenbus_dev_fatal(dev, err, "starting transaction");
+			goto fail;
+		}
+
+		err = xenbus_printf(xbt, dev->nodename, "feature-sg", "%d", sg);
+		if (err) {
+			message = "writing feature-sg";
+			goto abort_transaction;
+		}
+
+		err = xenbus_printf(xbt, dev->nodename, "feature-gso-tcpv4",
+				    "%d", sg);
+		if (err) {
+			message = "writing feature-gso-tcpv4";
+			goto abort_transaction;
+		}
+
+		err = xenbus_printf(xbt, dev->nodename, "feature-gso-tcpv6",
+				    "%d", sg);
+		if (err) {
+			message = "writing feature-gso-tcpv6";
+			goto abort_transaction;
+		}
+
+		/* We support partial checksum setup for IPv6 packets */
+		err = xenbus_printf(xbt, dev->nodename,
+				    "feature-ipv6-csum-offload",
+				    "%d", 1);
+		if (err) {
+			message = "writing feature-ipv6-csum-offload";
+			goto abort_transaction;
+		}
+
+		/* We support rx-copy path. */
+		err = xenbus_printf(xbt, dev->nodename,
+				    "feature-rx-copy", "%d", 1);
+		if (err) {
+			message = "writing feature-rx-copy";
+			goto abort_transaction;
+		}
+
+		/* We don't support rx-flip path (except old guests who
+		 * don't grok this feature flag).
+		 */
+		err = xenbus_printf(xbt, dev->nodename,
+				    "feature-rx-flip", "%d", 0);
+		if (err) {
+			message = "writing feature-rx-flip";
+			goto abort_transaction;
+		}
+
+		/* We support dynamic multicast-control. */
+		err = xenbus_printf(xbt, dev->nodename,
+				    "feature-multicast-control", "%d", 1);
+		if (err) {
+			message = "writing feature-multicast-control";
+			goto abort_transaction;
+		}
+
+		err = xenbus_printf(xbt, dev->nodename,
+				    "feature-dynamic-multicast-control",
+				    "%d", 1);
+		if (err) {
+			message = "writing feature-dynamic-multicast-control";
+			goto abort_transaction;
+		}
+
+		err = xenbus_transaction_end(xbt, 0);
+	} while (err == -EAGAIN);
+
+	if (err) {
+		xenbus_dev_fatal(dev, err, "completing transaction");
+		goto fail;
+	}
+
+	/* Split event channels support, this is optional so it is not
+	 * put inside the above loop.
+	 */
+	err = xenbus_printf(XBT_NIL, dev->nodename,
+			    "feature-split-event-channels",
+			    "%u", separate_tx_rx_irq);
+	if (err)
+		pr_debug("Error writing feature-split-event-channels\n");
+
+	/* Multi-queue support: This is an optional feature. */
+	err = xenbus_printf(XBT_NIL, dev->nodename,
+			    "multi-queue-max-queues", "%u", xenvif_max_queues);
+	if (err)
+		pr_debug("Error writing multi-queue-max-queues\n");
+
+	err = xenbus_printf(XBT_NIL, dev->nodename,
+			    "feature-ctrl-ring",
+			    "%u", true);
+	if (err)
+		pr_debug("Error writing feature-ctrl-ring\n");
+
+	script = xenbus_read(XBT_NIL, dev->nodename, "script", NULL);
+	if (IS_ERR(script)) {
+		err = PTR_ERR(script);
+		xenbus_dev_fatal(dev, err, "reading script");
+		goto fail;
+	}
+
+	be->hotplug_script = script;
+
+	/* This kicks hotplug scripts, so do it immediately. */
+	err = backend_create_xenvif(be);
+	if (err)
+		goto fail;
+
+	return 0;
+
+abort_transaction:
+	xenbus_transaction_end(xbt, 1);
+	xenbus_dev_fatal(dev, err, "%s", message);
+fail:
+	pr_debug("failed\n");
+	netback_remove(dev);
+	return err;
+}
+
 static const struct xenbus_device_id netback_ids[] = {
 	{ "vif" },
 	{ "" }
-- 
cgit v1.2.3-59-g8ed1b


From f55c3188df84248558f33b13bba4c2817dda78a7 Mon Sep 17 00:00:00 2001
From: Paul Durrant <pdurrant@amazon.com>
Date: Tue, 17 Dec 2019 13:32:17 +0000
Subject: xen-netback: switch state to InitWait at the end of
 netback_probe()...

...as the comment above the function states.

The switch to Initialising at the start of the function is somewhat bogus
as the toolstack will have set that initial state anyway. To behave
correctly, a backend should switch to InitWait once it has set up all
xenstore values that may be required by a initialising frontend. This
patch calls backend_switch_state() to make the transition at the
appropriate point.

NOTE: backend_switch_state() ignores errors from xenbus_switch_state()
      and so this patch removes an error path from netback_probe(). This
      means a failure to change state at this stage (in the absence of
      other failures) will leave the device instantiated. This is highly
      unlikley to happen as a failure to change state would indicate a
      failure to write to xenstore, and that will trigger other error
      paths. Also, a 'stuck' device can still be cleaned up using 'unbind'
      in any case.

Signed-off-by: Paul Durrant <pdurrant@amazon.com>
Acked-by: Wei Liu <wei.liu@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/xen-netback/xenbus.c | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

diff --git a/drivers/net/xen-netback/xenbus.c b/drivers/net/xen-netback/xenbus.c
index bb61316d79de..682e5e20971b 100644
--- a/drivers/net/xen-netback/xenbus.c
+++ b/drivers/net/xen-netback/xenbus.c
@@ -992,11 +992,6 @@ static int netback_probe(struct xenbus_device *dev,
 	be->dev = dev;
 	dev_set_drvdata(&dev->dev, be);
 
-	be->state = XenbusStateInitialising;
-	err = xenbus_switch_state(dev, XenbusStateInitialising);
-	if (err)
-		goto fail;
-
 	sg = 1;
 
 	do {
@@ -1098,6 +1093,8 @@ static int netback_probe(struct xenbus_device *dev,
 	if (err)
 		pr_debug("Error writing feature-ctrl-ring\n");
 
+	backend_switch_state(be, XenbusStateInitWait);
+
 	script = xenbus_read(XBT_NIL, dev->nodename, "script", NULL);
 	if (IS_ERR(script)) {
 		err = PTR_ERR(script);
-- 
cgit v1.2.3-59-g8ed1b


From 1f2565780e9b7218cf92c7630130e82dcc0fe9c2 Mon Sep 17 00:00:00 2001
From: Paul Durrant <pdurrant@amazon.com>
Date: Tue, 17 Dec 2019 13:32:18 +0000
Subject: xen-netback: remove 'hotplug-status' once it has served its purpose

Removing the 'hotplug-status' node in netback_remove() is wrong; the script
may not have completed. Only remove the node once the watch has fired and
has been unregistered.

Signed-off-by: Paul Durrant <pdurrant@amazon.com>
Acked-by: Wei Liu <wei.liu@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/xen-netback/xenbus.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/xen-netback/xenbus.c b/drivers/net/xen-netback/xenbus.c
index 682e5e20971b..17b4950ec051 100644
--- a/drivers/net/xen-netback/xenbus.c
+++ b/drivers/net/xen-netback/xenbus.c
@@ -648,6 +648,7 @@ static void hotplug_status_changed(struct xenbus_watch *watch,
 
 		/* Not interested in this watch anymore. */
 		unregister_hotplug_status_watch(be);
+		xenbus_rm(XBT_NIL, be->dev->nodename, "hotplug-status");
 	}
 	kfree(str);
 }
@@ -959,7 +960,6 @@ static int netback_remove(struct xenbus_device *dev)
 	if (be->vif) {
 		kobject_uevent(&dev->dev.kobj, KOBJ_OFFLINE);
 		xen_unregister_watchers(be->vif);
-		xenbus_rm(XBT_NIL, dev->nodename, "hotplug-status");
 		xenvif_free(be->vif);
 		be->vif = NULL;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 97c63746ebb9eebe34551fdb4370d6e3fff52059 Mon Sep 17 00:00:00 2001
From: John Crispin <john@phrozen.org>
Date: Fri, 13 Dec 2019 16:38:31 +0100
Subject: ath11k: add wmi helper for turning STA PS on/off

Add a WMI call helper to set the powersave mode of a STA interface.

Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/wmi.c | 30 ++++++++++++++++++++++++++++++
 drivers/net/wireless/ath/ath11k/wmi.h |  7 +++++++
 2 files changed, 37 insertions(+)

diff --git a/drivers/net/wireless/ath/ath11k/wmi.c b/drivers/net/wireless/ath/ath11k/wmi.c
index b05642617b78..bf3c246ce883 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.c
+++ b/drivers/net/wireless/ath/ath11k/wmi.c
@@ -1178,6 +1178,36 @@ int ath11k_wmi_pdev_set_param(struct ath11k *ar, u32 param_id,
 	return ret;
 }
 
+int ath11k_wmi_pdev_set_ps_mode(struct ath11k *ar, int vdev_id, u32 enable)
+{
+	struct ath11k_pdev_wmi *wmi = ar->wmi;
+	struct wmi_pdev_set_ps_mode_cmd *cmd;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_pdev_set_ps_mode_cmd *)skb->data;
+	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_STA_POWERSAVE_MODE_CMD) |
+			  FIELD_PREP(WMI_TLV_LEN, sizeof(*cmd) - TLV_HDR_SIZE);
+	cmd->vdev_id = vdev_id;
+	cmd->sta_ps_mode = enable;
+
+	ret = ath11k_wmi_cmd_send(wmi, skb, WMI_STA_POWERSAVE_MODE_CMDID);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to send WMI_PDEV_SET_PARAM cmd\n");
+		dev_kfree_skb(skb);
+	}
+
+	ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
+		   "WMI vdev set psmode %d vdev id %d\n",
+		   enable, vdev_id);
+
+	return ret;
+}
+
 int ath11k_wmi_pdev_suspend(struct ath11k *ar, u32 suspend_opt,
 			    u32 pdev_id)
 {
diff --git a/drivers/net/wireless/ath/ath11k/wmi.h b/drivers/net/wireless/ath/ath11k/wmi.h
index 4a518d406bc5..9919d8bf297b 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.h
+++ b/drivers/net/wireless/ath/ath11k/wmi.h
@@ -2827,6 +2827,12 @@ struct wmi_pdev_set_param_cmd {
 	u32 param_value;
 } __packed;
 
+struct wmi_pdev_set_ps_mode_cmd {
+	u32 tlv_header;
+	u32 vdev_id;
+	u32 sta_ps_mode;
+} __packed;
+
 struct wmi_pdev_suspend_cmd {
 	u32 tlv_header;
 	u32 pdev_id;
@@ -4682,6 +4688,7 @@ int ath11k_wmi_set_peer_param(struct ath11k *ar, const u8 *peer_addr,
 			      u32 vdev_id, u32 param_id, u32 param_val);
 int ath11k_wmi_pdev_set_param(struct ath11k *ar, u32 param_id,
 			      u32 param_value, u8 pdev_id);
+int ath11k_wmi_pdev_set_ps_mode(struct ath11k *ar, int vdev_id, u32 enable);
 int ath11k_wmi_wait_for_unified_ready(struct ath11k_base *ab);
 int ath11k_wmi_cmd_init(struct ath11k_base *ab);
 int ath11k_wmi_wait_for_service_ready(struct ath11k_base *ab);
-- 
cgit v1.2.3-59-g8ed1b


From 6bfebd4bf93928958cd57df9a8ccd5862f88468c Mon Sep 17 00:00:00 2001
From: John Crispin <john@phrozen.org>
Date: Fri, 13 Dec 2019 16:38:32 +0100
Subject: ath11k: disable PS for STA interfaces by default upon bringup

After applying this setting the TX performance issue of STA interfaces is
gone and we can see TX performance go up to ~900mbit on HE80.

Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/mac.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index 65f6a2585577..f5f7d8188e6e 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -4172,6 +4172,13 @@ static int ath11k_mac_op_add_interface(struct ieee80211_hw *hw,
 				    arvif->vdev_id, ret);
 			goto err_peer_del;
 		}
+
+		ret = ath11k_wmi_pdev_set_ps_mode(ar, arvif->vdev_id, false);
+		if (ret) {
+			ath11k_warn(ar->ab, "failed to disable vdev %d ps mode: %d\n",
+				    arvif->vdev_id, ret);
+			goto err_peer_del;
+		}
 		break;
 	default:
 		break;
-- 
cgit v1.2.3-59-g8ed1b


From bde4d95e92d56111c0548cb81dfe1ccb8b2bbfbd Mon Sep 17 00:00:00 2001
From: John Crispin <john@phrozen.org>
Date: Fri, 13 Dec 2019 16:38:33 +0100
Subject: ath11k: drop memset when setting up a tx cmd desc

There is no point in zero'ing out the structure if we set all values in the
following line.

Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/dp_tx.c  | 2 --
 drivers/net/wireless/ath/ath11k/hal_tx.c | 1 +
 2 files changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp_tx.c b/drivers/net/wireless/ath/ath11k/dp_tx.c
index a8b9557c2346..28ebc414533e 100644
--- a/drivers/net/wireless/ath/ath11k/dp_tx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_tx.c
@@ -167,8 +167,6 @@ int ath11k_dp_tx(struct ath11k *ar, struct ath11k_vif *arvif,
 	skb_cb->vif = arvif->vif;
 	skb_cb->ar = ar;
 
-	memset(cached_desc, 0, HAL_TCL_DESC_LEN);
-
 	ath11k_hal_tx_cmd_desc_setup(ab, cached_desc, &ti);
 
 	hal_ring_id = tx_ring->tcl_data_ring.ring_id;
diff --git a/drivers/net/wireless/ath/ath11k/hal_tx.c b/drivers/net/wireless/ath/ath11k/hal_tx.c
index cbe549798762..72a51ed65a51 100644
--- a/drivers/net/wireless/ath/ath11k/hal_tx.c
+++ b/drivers/net/wireless/ath/ath11k/hal_tx.c
@@ -71,6 +71,7 @@ void ath11k_hal_tx_cmd_desc_setup(struct ath11k_base *ab, void *cmd,
 				    ti->dscp_tid_tbl_idx) |
 			 FIELD_PREP(HAL_TCL_DATA_CMD_INFO3_SEARCH_INDEX,
 				    ti->bss_ast_hash);
+	tcl_cmd->info4 = 0;
 }
 
 /* Commit the descriptor to hardware */
-- 
cgit v1.2.3-59-g8ed1b


From 6bc9d6f786a56c4b8d267d7e9e52bbcde6a0b03a Mon Sep 17 00:00:00 2001
From: John Crispin <john@phrozen.org>
Date: Fri, 13 Dec 2019 16:38:34 +0100
Subject: ath11k: rename ath11k_wmi_base instances from wmi_sc to wmi_ab

This makes the code consistent with the recent sc to ab rename.

Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/core.c     |   4 +-
 drivers/net/wireless/ath/ath11k/core.h     |   2 +-
 drivers/net/wireless/ath/ath11k/htc.c      |   2 +-
 drivers/net/wireless/ath/ath11k/mac.c      |   4 +-
 drivers/net/wireless/ath/ath11k/testmode.c |   2 +-
 drivers/net/wireless/ath/ath11k/wmi.c      | 146 ++++++++++++++---------------
 drivers/net/wireless/ath/ath11k/wmi.h      |   2 +-
 7 files changed, 81 insertions(+), 81 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/core.c b/drivers/net/wireless/ath/ath11k/core.c
index 5cdc9b2aee51..9e823056e673 100644
--- a/drivers/net/wireless/ath/ath11k/core.c
+++ b/drivers/net/wireless/ath/ath11k/core.c
@@ -651,7 +651,7 @@ static void ath11k_core_restart(struct work_struct *work)
 		idr_destroy(&ar->txmgmt_idr);
 	}
 
-	wake_up(&ab->wmi_sc.tx_credits_wq);
+	wake_up(&ab->wmi_ab.tx_credits_wq);
 	wake_up(&ab->peer_mapping_wq);
 
 	ret = ath11k_core_reconfigure_on_crash(ab);
@@ -761,7 +761,7 @@ struct ath11k_base *ath11k_core_alloc(struct device *dev)
 
 	INIT_LIST_HEAD(&ab->peers);
 	init_waitqueue_head(&ab->peer_mapping_wq);
-	init_waitqueue_head(&ab->wmi_sc.tx_credits_wq);
+	init_waitqueue_head(&ab->wmi_ab.tx_credits_wq);
 	INIT_WORK(&ab->restart_work, ath11k_core_restart);
 	timer_setup(&ab->rx_replenish_retry, ath11k_ce_rx_replenish_retry, 0);
 	ab->dev = dev;
diff --git a/drivers/net/wireless/ath/ath11k/core.h b/drivers/net/wireless/ath/ath11k/core.h
index 06482a5c102d..25cdcf71d0c4 100644
--- a/drivers/net/wireless/ath/ath11k/core.h
+++ b/drivers/net/wireless/ath/ath11k/core.h
@@ -578,7 +578,7 @@ struct ath11k_base {
 	struct platform_device *pdev;
 	struct device *dev;
 	struct ath11k_qmi qmi;
-	struct ath11k_wmi_base wmi_sc;
+	struct ath11k_wmi_base wmi_ab;
 	struct completion fw_ready;
 	struct rproc *tgt_rproc;
 	int num_radios;
diff --git a/drivers/net/wireless/ath/ath11k/htc.c b/drivers/net/wireless/ath/ath11k/htc.c
index 9d4115f25ceb..8f54f58b83e6 100644
--- a/drivers/net/wireless/ath/ath11k/htc.c
+++ b/drivers/net/wireless/ath/ath11k/htc.c
@@ -736,7 +736,7 @@ int ath11k_htc_init(struct ath11k_base *ab)
 
 	htc->ab = ab;
 
-	switch (ab->wmi_sc.preferred_hw_mode) {
+	switch (ab->wmi_ab.preferred_hw_mode) {
 	case WMI_HOST_HW_MODE_SINGLE:
 		htc->wmi_ep_count = 1;
 		break;
diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index f5f7d8188e6e..3450c22cab64 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -5320,7 +5320,7 @@ ath11k_mac_update_bss_chan_survey(struct ath11k *ar,
 
 	lockdep_assert_held(&ar->conf_mutex);
 
-	if (!test_bit(WMI_TLV_SERVICE_BSS_CHANNEL_INFO_64, ar->ab->wmi_sc.svc_map) ||
+	if (!test_bit(WMI_TLV_SERVICE_BSS_CHANNEL_INFO_64, ar->ab->wmi_ab.svc_map) ||
 	    ar->rx_channel != channel)
 		return;
 
@@ -5838,7 +5838,7 @@ int ath11k_mac_allocate(struct ath11k_base *ab)
 		ar->pdev_idx = i;
 		ar->lmac_id = ath11k_core_get_hw_mac_id(ab, i);
 
-		ar->wmi = &ab->wmi_sc.wmi[i];
+		ar->wmi = &ab->wmi_ab.wmi[i];
 		/* FIXME wmi[0] is already initialized during attach,
 		 * Should we do this again?
 		 */
diff --git a/drivers/net/wireless/ath/ath11k/testmode.c b/drivers/net/wireless/ath/ath11k/testmode.c
index 932bee6d1740..d2dc9db01491 100644
--- a/drivers/net/wireless/ath/ath11k/testmode.c
+++ b/drivers/net/wireless/ath/ath11k/testmode.c
@@ -150,7 +150,7 @@ static int ath11k_tm_cmd_wmi(struct ath11k *ar, struct nlattr *tb[])
 
 	ath11k_dbg_dump(ar->ab, ATH11K_DBG_TESTMODE, NULL, "", buf, buf_len);
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, buf_len);
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, buf_len);
 	if (!skb) {
 		ret = -ENOMEM;
 		goto out;
diff --git a/drivers/net/wireless/ath/ath11k/wmi.c b/drivers/net/wireless/ath/ath11k/wmi.c
index bf3c246ce883..b16bfb50d73e 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.c
+++ b/drivers/net/wireless/ath/ath11k/wmi.c
@@ -205,7 +205,7 @@ static int ath11k_wmi_cmd_send_nowait(struct ath11k_pdev_wmi *wmi, struct sk_buf
 				      u32 cmd_id)
 {
 	struct ath11k_skb_cb *skb_cb = ATH11K_SKB_CB(skb);
-	struct ath11k_base *ab = wmi->wmi_sc->ab;
+	struct ath11k_base *ab = wmi->wmi_ab->ab;
 	struct wmi_cmd_hdr *cmd_hdr;
 	int ret;
 	u32 cmd = 0;
@@ -234,7 +234,7 @@ err_pull:
 int ath11k_wmi_cmd_send(struct ath11k_pdev_wmi *wmi, struct sk_buff *skb,
 			u32 cmd_id)
 {
-	struct ath11k_wmi_base *wmi_sc = wmi->wmi_sc;
+	struct ath11k_wmi_base *wmi_sc = wmi->wmi_ab;
 	int ret = -EOPNOTSUPP;
 
 	might_sleep();
@@ -448,7 +448,7 @@ static void ath11k_wmi_service_bitmap_copy(struct ath11k_pdev_wmi *wmi,
 	for (i = 0, j = 0; i < WMI_SERVICE_BM_SIZE && j < WMI_MAX_SERVICE; i++) {
 		do {
 			if (wmi_svc_bm[i] & BIT(j % WMI_SERVICE_BITS_IN_SIZE32))
-				set_bit(j, wmi->wmi_sc->svc_map);
+				set_bit(j, wmi->wmi_ab->svc_map);
 		} while (++j % WMI_SERVICE_BITS_IN_SIZE32);
 	}
 }
@@ -457,7 +457,7 @@ static int ath11k_wmi_tlv_svc_rdy_parse(struct ath11k_base *ab, u16 tag, u16 len
 					const void *ptr, void *data)
 {
 	struct wmi_tlv_svc_ready_parse *svc_ready = data;
-	struct ath11k_pdev_wmi *wmi_handle = &ab->wmi_sc.wmi[0];
+	struct ath11k_pdev_wmi *wmi_handle = &ab->wmi_ab.wmi[0];
 	u16 expect_len;
 
 	switch (tag) {
@@ -538,7 +538,7 @@ int ath11k_wmi_mgmt_send(struct ath11k *ar, u32 vdev_id, u32 buf_id,
 
 	len = sizeof(*cmd) + sizeof(*frame_tlv) + roundup(buf_len, 4);
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, len);
 	if (!skb)
 		return -ENOMEM;
 
@@ -590,7 +590,7 @@ int ath11k_wmi_vdev_create(struct ath11k *ar, u8 *macaddr,
 	len = sizeof(*cmd) + TLV_HDR_SIZE +
 		(WMI_NUM_SUPPORTED_BAND_MAX * sizeof(*txrx_streams));
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, len);
 	if (!skb)
 		return -ENOMEM;
 
@@ -656,7 +656,7 @@ int ath11k_wmi_vdev_delete(struct ath11k *ar, u8 vdev_id)
 	struct sk_buff *skb;
 	int ret;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, sizeof(*cmd));
 	if (!skb)
 		return -ENOMEM;
 
@@ -683,7 +683,7 @@ int ath11k_wmi_vdev_stop(struct ath11k *ar, u8 vdev_id)
 	struct sk_buff *skb;
 	int ret;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, sizeof(*cmd));
 	if (!skb)
 		return -ENOMEM;
 
@@ -711,7 +711,7 @@ int ath11k_wmi_vdev_down(struct ath11k *ar, u8 vdev_id)
 	struct sk_buff *skb;
 	int ret;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, sizeof(*cmd));
 	if (!skb)
 		return -ENOMEM;
 
@@ -789,7 +789,7 @@ int ath11k_wmi_vdev_start(struct ath11k *ar, struct wmi_vdev_start_req_arg *arg,
 
 	len = sizeof(*cmd) + sizeof(*chan) + TLV_HDR_SIZE;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, len);
 	if (!skb)
 		return -ENOMEM;
 
@@ -867,7 +867,7 @@ int ath11k_wmi_vdev_up(struct ath11k *ar, u32 vdev_id, u32 aid, const u8 *bssid)
 	struct sk_buff *skb;
 	int ret;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, sizeof(*cmd));
 	if (!skb)
 		return -ENOMEM;
 
@@ -901,7 +901,7 @@ int ath11k_wmi_send_peer_create_cmd(struct ath11k *ar,
 	struct sk_buff *skb;
 	int ret;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, sizeof(*cmd));
 	if (!skb)
 		return -ENOMEM;
 
@@ -934,7 +934,7 @@ int ath11k_wmi_send_peer_delete_cmd(struct ath11k *ar,
 	struct sk_buff *skb;
 	int ret;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, sizeof(*cmd));
 	if (!skb)
 		return -ENOMEM;
 
@@ -966,7 +966,7 @@ int ath11k_wmi_send_pdev_set_regdomain(struct ath11k *ar,
 	struct sk_buff *skb;
 	int ret;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, sizeof(*cmd));
 	if (!skb)
 		return -ENOMEM;
 
@@ -1006,7 +1006,7 @@ int ath11k_wmi_set_peer_param(struct ath11k *ar, const u8 *peer_addr,
 	struct sk_buff *skb;
 	int ret;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, sizeof(*cmd));
 	if (!skb)
 		return -ENOMEM;
 
@@ -1040,7 +1040,7 @@ int ath11k_wmi_send_peer_flush_tids_cmd(struct ath11k *ar,
 	struct sk_buff *skb;
 	int ret;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, sizeof(*cmd));
 	if (!skb)
 		return -ENOMEM;
 
@@ -1076,7 +1076,7 @@ int ath11k_wmi_peer_rx_reorder_queue_setup(struct ath11k *ar,
 	struct sk_buff *skb;
 	int ret;
 
-	skb = ath11k_wmi_alloc_skb(ar->wmi->wmi_sc, sizeof(*cmd));
+	skb = ath11k_wmi_alloc_skb(ar->wmi->wmi_ab, sizeof(*cmd));
 	if (!skb)
 		return -ENOMEM;
 
@@ -1118,7 +1118,7 @@ ath11k_wmi_rx_reord_queue_remove(struct ath11k *ar,
 	struct sk_buff *skb;
 	int ret;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, sizeof(*cmd));
 	if (!skb)
 		return -ENOMEM;
 
@@ -1154,7 +1154,7 @@ int ath11k_wmi_pdev_set_param(struct ath11k *ar, u32 param_id,
 	struct sk_buff *skb;
 	int ret;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, sizeof(*cmd));
 	if (!skb)
 		return -ENOMEM;
 
@@ -1185,7 +1185,7 @@ int ath11k_wmi_pdev_set_ps_mode(struct ath11k *ar, int vdev_id, u32 enable)
 	struct sk_buff *skb;
 	int ret;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, sizeof(*cmd));
 	if (!skb)
 		return -ENOMEM;
 
@@ -1216,7 +1216,7 @@ int ath11k_wmi_pdev_suspend(struct ath11k *ar, u32 suspend_opt,
 	struct sk_buff *skb;
 	int ret;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, sizeof(*cmd));
 	if (!skb)
 		return -ENOMEM;
 
@@ -1247,7 +1247,7 @@ int ath11k_wmi_pdev_resume(struct ath11k *ar, u32 pdev_id)
 	struct sk_buff *skb;
 	int ret;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, sizeof(*cmd));
 	if (!skb)
 		return -ENOMEM;
 
@@ -1281,7 +1281,7 @@ int ath11k_wmi_pdev_bss_chan_info_request(struct ath11k *ar,
 	struct sk_buff *skb;
 	int ret;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, sizeof(*cmd));
 	if (!skb)
 		return -ENOMEM;
 
@@ -1314,7 +1314,7 @@ int ath11k_wmi_send_set_ap_ps_param_cmd(struct ath11k *ar, u8 *peer_addr,
 	struct sk_buff *skb;
 	int ret;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, sizeof(*cmd));
 	if (!skb)
 		return -ENOMEM;
 
@@ -1349,7 +1349,7 @@ int ath11k_wmi_set_sta_ps_param(struct ath11k *ar, u32 vdev_id,
 	struct sk_buff *skb;
 	int ret;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, sizeof(*cmd));
 	if (!skb)
 		return -ENOMEM;
 
@@ -1384,7 +1384,7 @@ int ath11k_wmi_force_fw_hang_cmd(struct ath11k *ar, u32 type, u32 delay_time_ms)
 
 	len = sizeof(*cmd);
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, len);
 	if (!skb)
 		return -ENOMEM;
 
@@ -1412,7 +1412,7 @@ int ath11k_wmi_vdev_set_param_cmd(struct ath11k *ar, u32 vdev_id,
 	struct sk_buff *skb;
 	int ret;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, sizeof(*cmd));
 	if (!skb)
 		return -ENOMEM;
 
@@ -1446,7 +1446,7 @@ int ath11k_wmi_send_stats_request_cmd(struct ath11k *ar,
 	struct sk_buff *skb;
 	int ret;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, sizeof(*cmd));
 	if (!skb)
 		return -ENOMEM;
 
@@ -1479,7 +1479,7 @@ int ath11k_wmi_send_bcn_offload_control_cmd(struct ath11k *ar,
 	struct sk_buff *skb;
 	int ret;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, sizeof(*cmd));
 	if (!skb)
 		return -ENOMEM;
 
@@ -1520,7 +1520,7 @@ int ath11k_wmi_bcn_tmpl(struct ath11k *ar, u32 vdev_id,
 
 	len = sizeof(*cmd) + sizeof(*bcn_prb_info) + TLV_HDR_SIZE + aligned_len;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, len);
 	if (!skb)
 		return -ENOMEM;
 
@@ -1571,7 +1571,7 @@ int ath11k_wmi_vdev_install_key(struct ath11k *ar,
 
 	len = sizeof(*cmd) + TLV_HDR_SIZE + key_len_aligned;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, len);
 	if (!skb)
 		return -ENOMEM;
 
@@ -1717,7 +1717,7 @@ int ath11k_wmi_send_peer_assoc_cmd(struct ath11k *ar,
 	      sizeof(*mcs) + TLV_HDR_SIZE +
 	      (sizeof(*he_mcs) * param->peer_he_mcs_count);
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, len);
 	if (!skb)
 		return -ENOMEM;
 
@@ -1983,7 +1983,7 @@ int ath11k_wmi_send_scan_start_cmd(struct ath11k *ar,
 			roundup(params->extraie.len, sizeof(u32));
 	len += extraie_len_with_pad;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, len);
 	if (!skb)
 		return -ENOMEM;
 
@@ -2099,7 +2099,7 @@ int ath11k_wmi_send_scan_stop_cmd(struct ath11k *ar,
 	struct sk_buff *skb;
 	int ret;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, sizeof(*cmd));
 	if (!skb)
 		return -ENOMEM;
 
@@ -2155,7 +2155,7 @@ int ath11k_wmi_send_scan_chan_list_cmd(struct ath11k *ar,
 	len = sizeof(*cmd) + TLV_HDR_SIZE +
 		 sizeof(*chan_info) * chan_list->nallchans;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, len);
 	if (!skb)
 		return -ENOMEM;
 
@@ -2247,7 +2247,7 @@ int ath11k_wmi_send_wmm_update_cmd_tlv(struct ath11k *ar, u32 vdev_id,
 	struct sk_buff *skb;
 	int ret, ac;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, sizeof(*cmd));
 	if (!skb)
 		return -ENOMEM;
 
@@ -2314,7 +2314,7 @@ int ath11k_wmi_send_dfs_phyerr_offload_enable_cmd(struct ath11k *ar,
 	struct sk_buff *skb;
 	int ret;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, sizeof(*cmd));
 	if (!skb)
 		return -ENOMEM;
 
@@ -2351,7 +2351,7 @@ int ath11k_wmi_pdev_peer_pktlog_filter(struct ath11k *ar, u8 *addr, u8 enable)
 	int ret, len;
 
 	len = sizeof(*cmd) + sizeof(*info) + TLV_HDR_SIZE;
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, len);
 	if (!skb)
 		return -ENOMEM;
 
@@ -2397,7 +2397,7 @@ ath11k_wmi_send_init_country_cmd(struct ath11k *ar,
 	struct sk_buff *skb;
 	int ret;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, sizeof(*cmd));
 	if (!skb)
 		return -ENOMEM;
 
@@ -2449,7 +2449,7 @@ int ath11k_wmi_pdev_pktlog_enable(struct ath11k *ar, u32 pktlog_filter)
 	struct sk_buff *skb;
 	int ret;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, sizeof(*cmd));
 	if (!skb)
 		return -ENOMEM;
 
@@ -2479,7 +2479,7 @@ int ath11k_wmi_pdev_pktlog_disable(struct ath11k *ar)
 	struct sk_buff *skb;
 	int ret;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, sizeof(*cmd));
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, sizeof(*cmd));
 	if (!skb)
 		return -ENOMEM;
 
@@ -2504,14 +2504,14 @@ int
 ath11k_wmi_send_twt_enable_cmd(struct ath11k *ar, u32 pdev_id)
 {
 	struct ath11k_pdev_wmi *wmi = ar->wmi;
-	struct ath11k_base *ab = wmi->wmi_sc->ab;
+	struct ath11k_base *ab = wmi->wmi_ab->ab;
 	struct wmi_twt_enable_params_cmd *cmd;
 	struct sk_buff *skb;
 	int ret, len;
 
 	len = sizeof(*cmd);
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, len);
 	if (!skb)
 		return -ENOMEM;
 
@@ -2555,14 +2555,14 @@ int
 ath11k_wmi_send_twt_disable_cmd(struct ath11k *ar, u32 pdev_id)
 {
 	struct ath11k_pdev_wmi *wmi = ar->wmi;
-	struct ath11k_base *ab = wmi->wmi_sc->ab;
+	struct ath11k_base *ab = wmi->wmi_ab->ab;
 	struct wmi_twt_disable_params_cmd *cmd;
 	struct sk_buff *skb;
 	int ret, len;
 
 	len = sizeof(*cmd);
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, len);
 	if (!skb)
 		return -ENOMEM;
 
@@ -2585,14 +2585,14 @@ ath11k_wmi_send_obss_spr_cmd(struct ath11k *ar, u32 vdev_id,
 			     struct ieee80211_he_obss_pd *he_obss_pd)
 {
 	struct ath11k_pdev_wmi *wmi = ar->wmi;
-	struct ath11k_base *ab = wmi->wmi_sc->ab;
+	struct ath11k_base *ab = wmi->wmi_ab->ab;
 	struct wmi_obss_spatial_reuse_params_cmd *cmd;
 	struct sk_buff *skb;
 	int ret, len;
 
 	len = sizeof(*cmd);
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, len);
 	if (!skb)
 		return -ENOMEM;
 
@@ -2714,7 +2714,7 @@ ath11k_wmi_copy_resource_config(struct wmi_resource_config *wmi_cfg,
 static int ath11k_init_cmd_send(struct ath11k_pdev_wmi *wmi,
 				struct wmi_init_cmd_param *param)
 {
-	struct ath11k_base *ab = wmi->wmi_sc->ab;
+	struct ath11k_base *ab = wmi->wmi_ab->ab;
 	struct sk_buff *skb;
 	struct wmi_init_cmd *cmd;
 	struct wmi_resource_config *cfg;
@@ -2734,7 +2734,7 @@ static int ath11k_init_cmd_send(struct ath11k_pdev_wmi *wmi,
 	len = sizeof(*cmd) + TLV_HDR_SIZE + sizeof(*cfg) + hw_mode_len +
 	      (sizeof(*host_mem_chunks) * WMI_MAX_MEM_REQS);
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, len);
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, len);
 	if (!skb)
 		return -ENOMEM;
 
@@ -2829,7 +2829,7 @@ int ath11k_wmi_wait_for_service_ready(struct ath11k_base *ab)
 {
 	unsigned long time_left;
 
-	time_left = wait_for_completion_timeout(&ab->wmi_sc.service_ready,
+	time_left = wait_for_completion_timeout(&ab->wmi_ab.service_ready,
 						WMI_SERVICE_READY_TIMEOUT_HZ);
 	if (!time_left)
 		return -ETIMEDOUT;
@@ -2841,7 +2841,7 @@ int ath11k_wmi_wait_for_unified_ready(struct ath11k_base *ab)
 {
 	unsigned long time_left;
 
-	time_left = wait_for_completion_timeout(&ab->wmi_sc.unified_ready,
+	time_left = wait_for_completion_timeout(&ab->wmi_ab.unified_ready,
 						WMI_SERVICE_READY_TIMEOUT_HZ);
 	if (!time_left)
 		return -ETIMEDOUT;
@@ -2851,7 +2851,7 @@ int ath11k_wmi_wait_for_unified_ready(struct ath11k_base *ab)
 
 int ath11k_wmi_cmd_init(struct ath11k_base *ab)
 {
-	struct ath11k_wmi_base *wmi_sc = &ab->wmi_sc;
+	struct ath11k_wmi_base *wmi_sc = &ab->wmi_ab;
 	struct wmi_init_cmd_param init_param;
 	struct target_resource_config  config;
 
@@ -2971,16 +2971,16 @@ static int ath11k_wmi_tlv_hw_mode_caps(struct ath11k_base *soc,
 	while (i < svc_rdy_ext->n_hw_mode_caps) {
 		hw_mode_caps = &svc_rdy_ext->hw_mode_caps[i];
 		mode = hw_mode_caps->hw_mode_id;
-		pref = soc->wmi_sc.preferred_hw_mode;
+		pref = soc->wmi_ab.preferred_hw_mode;
 
 		if (ath11k_hw_mode_pri_map[mode] < ath11k_hw_mode_pri_map[pref]) {
 			svc_rdy_ext->pref_hw_mode_caps = *hw_mode_caps;
-			soc->wmi_sc.preferred_hw_mode = mode;
+			soc->wmi_ab.preferred_hw_mode = mode;
 		}
 		i++;
 	}
 
-	if (soc->wmi_sc.preferred_hw_mode == WMI_HOST_HW_MODE_MAX)
+	if (soc->wmi_ab.preferred_hw_mode == WMI_HOST_HW_MODE_MAX)
 		return -EINVAL;
 
 	return 0;
@@ -3030,7 +3030,7 @@ static int ath11k_wmi_tlv_ext_hal_reg_caps_parse(struct ath11k_base *soc,
 static int ath11k_wmi_tlv_ext_hal_reg_caps(struct ath11k_base *soc,
 					   u16 len, const void *ptr, void *data)
 {
-	struct ath11k_pdev_wmi *wmi_handle = &soc->wmi_sc.wmi[0];
+	struct ath11k_pdev_wmi *wmi_handle = &soc->wmi_ab.wmi[0];
 	struct wmi_tlv_svc_rdy_ext_parse *svc_rdy_ext = data;
 	struct ath11k_hal_reg_capabilities_ext reg_cap;
 	int ret;
@@ -3066,7 +3066,7 @@ static int ath11k_wmi_tlv_ext_soc_hal_reg_caps_parse(struct ath11k_base *soc,
 						     u16 len, const void *ptr,
 						     void *data)
 {
-	struct ath11k_pdev_wmi *wmi_handle = &soc->wmi_sc.wmi[0];
+	struct ath11k_pdev_wmi *wmi_handle = &soc->wmi_ab.wmi[0];
 	struct wmi_tlv_svc_rdy_ext_parse *svc_rdy_ext = data;
 	u8 hw_mode_id = svc_rdy_ext->pref_hw_mode_caps.hw_mode_id;
 	u32 phy_id_map;
@@ -3104,7 +3104,7 @@ static int ath11k_wmi_tlv_svc_rdy_ext_parse(struct ath11k_base *ab,
 					    u16 tag, u16 len,
 					    const void *ptr, void *data)
 {
-	struct ath11k_pdev_wmi *wmi_handle = &ab->wmi_sc.wmi[0];
+	struct ath11k_pdev_wmi *wmi_handle = &ab->wmi_ab.wmi[0];
 	struct wmi_tlv_svc_rdy_ext_parse *svc_rdy_ext = data;
 	int ret;
 
@@ -3156,7 +3156,7 @@ static int ath11k_wmi_tlv_svc_rdy_ext_parse(struct ath11k_base *ab,
 				return ret;
 
 			svc_rdy_ext->ext_hal_reg_done = true;
-			complete(&ab->wmi_sc.service_ready);
+			complete(&ab->wmi_ab.service_ready);
 		}
 		break;
 
@@ -4498,7 +4498,7 @@ unlock:
 static void ath11k_wmi_op_ep_tx_credits(struct ath11k_base *ab)
 {
 	/* try to send pending beacons first. they take priority */
-	wake_up(&ab->wmi_sc.tx_credits_wq);
+	wake_up(&ab->wmi_ab.tx_credits_wq);
 }
 
 static void ath11k_wmi_htc_tx_complete(struct ath11k_base *ab,
@@ -4668,7 +4668,7 @@ static int ath11k_ready_event(struct ath11k_base *ab, struct sk_buff *skb)
 		return ret;
 	}
 
-	complete(&ab->wmi_sc.unified_ready);
+	complete(&ab->wmi_ab.unified_ready);
 	return 0;
 }
 
@@ -5319,7 +5319,7 @@ static void ath11k_service_available_event(struct ath11k_base *ab, struct sk_buf
 		do {
 			if (ev->wmi_service_segment_bitmap[i] &
 			    BIT(j % WMI_AVAIL_SERVICE_BITS_IN_SIZE32))
-				set_bit(j, ab->wmi_sc.svc_map);
+				set_bit(j, ab->wmi_ab.svc_map);
 		} while (++j % WMI_AVAIL_SERVICE_BITS_IN_SIZE32);
 	}
 
@@ -5639,9 +5639,9 @@ static int ath11k_connect_pdev_htc_service(struct ath11k_base *ab,
 		return status;
 	}
 
-	ab->wmi_sc.wmi_endpoint_id[pdev_idx] = conn_resp.eid;
-	ab->wmi_sc.wmi[pdev_idx].eid = conn_resp.eid;
-	ab->wmi_sc.max_msg_len[pdev_idx] = conn_resp.max_msg_len;
+	ab->wmi_ab.wmi_endpoint_id[pdev_idx] = conn_resp.eid;
+	ab->wmi_ab.wmi[pdev_idx].eid = conn_resp.eid;
+	ab->wmi_ab.max_msg_len[pdev_idx] = conn_resp.max_msg_len;
 
 	return 0;
 }
@@ -5664,7 +5664,7 @@ ath11k_wmi_send_unit_test_cmd(struct ath11k *ar,
 	arg_len = sizeof(u32) * ut_cmd.num_args;
 	buf_len = sizeof(ut_cmd) + arg_len + TLV_HDR_SIZE;
 
-	skb = ath11k_wmi_alloc_skb(wmi->wmi_sc, buf_len);
+	skb = ath11k_wmi_alloc_skb(wmi->wmi_ab, buf_len);
 	if (!skb)
 		return -ENOMEM;
 
@@ -5771,11 +5771,11 @@ int ath11k_wmi_pdev_attach(struct ath11k_base *ab,
 	if (pdev_id >= MAX_RADIOS)
 		return -EINVAL;
 
-	wmi_handle = &ab->wmi_sc.wmi[pdev_id];
+	wmi_handle = &ab->wmi_ab.wmi[pdev_id];
 
-	wmi_handle->wmi_sc = &ab->wmi_sc;
+	wmi_handle->wmi_ab = &ab->wmi_ab;
 
-	ab->wmi_sc.ab = ab;
+	ab->wmi_ab.ab = ab;
 	/* TODO: Init remaining resource specific to pdev */
 
 	return 0;
@@ -5789,12 +5789,12 @@ int ath11k_wmi_attach(struct ath11k_base *ab)
 	if (ret)
 		return ret;
 
-	ab->wmi_sc.ab = ab;
-	ab->wmi_sc.preferred_hw_mode = WMI_HOST_HW_MODE_MAX;
+	ab->wmi_ab.ab = ab;
+	ab->wmi_ab.preferred_hw_mode = WMI_HOST_HW_MODE_MAX;
 
 	/* TODO: Init remaining wmi soc resources required */
-	init_completion(&ab->wmi_sc.service_ready);
-	init_completion(&ab->wmi_sc.unified_ready);
+	init_completion(&ab->wmi_ab.service_ready);
+	init_completion(&ab->wmi_ab.unified_ready);
 
 	return 0;
 }
diff --git a/drivers/net/wireless/ath/ath11k/wmi.h b/drivers/net/wireless/ath/ath11k/wmi.h
index 9919d8bf297b..ab983aac604b 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.h
+++ b/drivers/net/wireless/ath/ath11k/wmi.h
@@ -2354,7 +2354,7 @@ struct wmi_service_available_event {
 } __packed;
 
 struct ath11k_pdev_wmi {
-	struct ath11k_wmi_base *wmi_sc;
+	struct ath11k_wmi_base *wmi_ab;
 	enum ath11k_htc_ep_id eid;
 	const struct wmi_peer_flags_map *peer_flags;
 	u32 rx_decap_mode;
-- 
cgit v1.2.3-59-g8ed1b


From 8cfa7ef8127b00de1ce0275a85220c331fe9a4d0 Mon Sep 17 00:00:00 2001
From: John Crispin <john@phrozen.org>
Date: Fri, 13 Dec 2019 16:38:37 +0100
Subject: ath11k: move some tx_status parsing to debugfs code

Some of the fields are only used by debugfs. Move the parsing of these
from the data hot path to the debugfs code.

Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/debugfs_sta.c | 63 +++++++++++++++------------
 drivers/net/wireless/ath/ath11k/hal_tx.c      | 19 +-------
 drivers/net/wireless/ath/ath11k/hal_tx.h      |  7 +--
 3 files changed, 36 insertions(+), 53 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/debugfs_sta.c b/drivers/net/wireless/ath/ath11k/debugfs_sta.c
index 3c5f931e22a9..743760c9bcae 100644
--- a/drivers/net/wireless/ath/ath11k/debugfs_sta.c
+++ b/drivers/net/wireless/ath/ath11k/debugfs_sta.c
@@ -129,12 +129,16 @@ void ath11k_update_per_peer_stats_from_txcompl(struct ath11k *ar,
 {
 	struct ath11k_base *ab = ar->ab;
 	struct ath11k_per_peer_tx_stats *peer_stats = &ar->cached_stats;
+	enum hal_tx_rate_stats_pkt_type pkt_type;
+	enum hal_tx_rate_stats_sgi sgi;
+	enum hal_tx_rate_stats_bw bw;
 	struct ath11k_peer *peer;
 	struct ath11k_sta *arsta;
 	struct ieee80211_sta *sta;
 	u16 rate;
 	u8 rate_idx;
 	int ret;
+	u8 mcs;
 
 	rcu_read_lock();
 	spin_lock_bh(&ab->base_lock);
@@ -150,51 +154,52 @@ void ath11k_update_per_peer_stats_from_txcompl(struct ath11k *ar,
 	arsta = (struct ath11k_sta *)sta->drv_priv;
 
 	memset(&arsta->txrate, 0, sizeof(arsta->txrate));
-
-	if (ts->pkt_type == HAL_TX_RATE_STATS_PKT_TYPE_11A ||
-	    ts->pkt_type == HAL_TX_RATE_STATS_PKT_TYPE_11B) {
-		ret = ath11k_mac_hw_ratecode_to_legacy_rate(ts->mcs,
-							    ts->pkt_type,
+	pkt_type = FIELD_GET(HAL_TX_RATE_STATS_INFO0_PKT_TYPE,
+			     ts->rate_stats);
+	mcs = FIELD_GET(HAL_TX_RATE_STATS_INFO0_MCS,
+			ts->rate_stats);
+	sgi = FIELD_GET(HAL_TX_RATE_STATS_INFO0_SGI,
+			ts->rate_stats);
+	bw = FIELD_GET(HAL_TX_RATE_STATS_INFO0_BW, ts->rate_stats);
+
+	if (pkt_type == HAL_TX_RATE_STATS_PKT_TYPE_11A ||
+	    pkt_type == HAL_TX_RATE_STATS_PKT_TYPE_11B) {
+		ret = ath11k_mac_hw_ratecode_to_legacy_rate(mcs,
+							    pkt_type,
 							    &rate_idx,
 							    &rate);
-		if (ret < 0) {
-			spin_unlock_bh(&ab->base_lock);
-			rcu_read_unlock();
-			return;
-		}
+		if (ret < 0)
+			goto err_out;
 		arsta->txrate.legacy = rate;
-	} else if (ts->pkt_type == HAL_TX_RATE_STATS_PKT_TYPE_11N) {
-		if (ts->mcs > 7) {
-			ath11k_warn(ab, "Invalid HT mcs index %d\n", ts->mcs);
-			spin_unlock_bh(&ab->base_lock);
-			rcu_read_unlock();
-			return;
+	} else if (pkt_type == HAL_TX_RATE_STATS_PKT_TYPE_11N) {
+		if (mcs > 7) {
+			ath11k_warn(ab, "Invalid HT mcs index %d\n", mcs);
+			goto err_out;
 		}
 
-		arsta->txrate.mcs = ts->mcs + 8 * (arsta->last_txrate.nss - 1);
+		arsta->txrate.mcs = mcs + 8 * (arsta->last_txrate.nss - 1);
 		arsta->txrate.flags = RATE_INFO_FLAGS_MCS;
-		if (ts->sgi)
+		if (sgi)
 			arsta->txrate.flags |= RATE_INFO_FLAGS_SHORT_GI;
-	} else if (ts->pkt_type == HAL_TX_RATE_STATS_PKT_TYPE_11AC) {
-		if (ts->mcs > 9) {
-			ath11k_warn(ab, "Invalid VHT mcs index %d\n", ts->mcs);
-			spin_unlock_bh(&ab->base_lock);
-			rcu_read_unlock();
-			return;
+	} else if (pkt_type == HAL_TX_RATE_STATS_PKT_TYPE_11AC) {
+		if (mcs > 9) {
+			ath11k_warn(ab, "Invalid VHT mcs index %d\n", mcs);
+			goto err_out;
 		}
 
-		arsta->txrate.mcs = ts->mcs;
+		arsta->txrate.mcs = mcs;
 		arsta->txrate.flags = RATE_INFO_FLAGS_VHT_MCS;
-		if (ts->sgi)
+		if (sgi)
 			arsta->txrate.flags |= RATE_INFO_FLAGS_SHORT_GI;
-	} else {
-		/*TODO: update HE rates */
+	} else if (pkt_type == HAL_TX_RATE_STATS_PKT_TYPE_11AX) {
+		/* TODO */
 	}
 
 	arsta->txrate.nss = arsta->last_txrate.nss;
-	arsta->txrate.bw = ath11k_mac_bw_to_mac80211_bw(ts->bw);
+	arsta->txrate.bw = ath11k_mac_bw_to_mac80211_bw(bw);
 
 	ath11k_accumulate_per_peer_tx_stats(arsta, peer_stats, rate_idx);
+err_out:
 	spin_unlock_bh(&ab->base_lock);
 	rcu_read_unlock();
 }
diff --git a/drivers/net/wireless/ath/ath11k/hal_tx.c b/drivers/net/wireless/ath/ath11k/hal_tx.c
index 72a51ed65a51..c2a3a9d2e534 100644
--- a/drivers/net/wireless/ath/ath11k/hal_tx.c
+++ b/drivers/net/wireless/ath/ath11k/hal_tx.c
@@ -127,24 +127,7 @@ void ath11k_hal_tx_status_parse(struct ath11k_base *ab,
 	if (!(desc->rate_stats.info0 & HAL_TX_RATE_STATS_INFO0_VALID))
 		return;
 
-	ts->flags |= HAL_TX_STATUS_FLAGS_RATE_STATS_VALID;
-	ts->tsf = desc->rate_stats.tsf;
-	ts->bw = FIELD_GET(HAL_TX_RATE_STATS_INFO0_BW, desc->rate_stats.info0);
-	ts->pkt_type = FIELD_GET(HAL_TX_RATE_STATS_INFO0_PKT_TYPE,
-				 desc->rate_stats.info0);
-	if (desc->rate_stats.info0 & HAL_TX_RATE_STATS_INFO0_STBC)
-		ts->flags |= HAL_TX_STATUS_FLAGS_RATE_STBC;
-	if (desc->rate_stats.info0 & HAL_TX_RATE_STATS_INFO0_LDPC)
-		ts->flags |= HAL_TX_STATUS_FLAGS_RATE_LDPC;
-	if (desc->rate_stats.info0 & HAL_TX_RATE_STATS_INFO0_OFDMA_TX)
-		ts->flags |= HAL_TX_STATUS_FLAGS_OFDMA;
-
-	ts->sgi = FIELD_GET(HAL_TX_RATE_STATS_INFO0_SGI,
-			    desc->rate_stats.info0);
-	ts->mcs = FIELD_GET(HAL_TX_RATE_STATS_INFO0_MCS,
-			    desc->rate_stats.info0);
-	ts->num_tones_in_ru = FIELD_GET(HAL_TX_RATE_STATS_INFO0_TONES_IN_RU,
-					desc->rate_stats.info0);
+	ts->rate_stats = desc->rate_stats.info0;
 }
 
 void ath11k_hal_tx_set_dscp_tid_map(struct ath11k_base *ab, int id)
diff --git a/drivers/net/wireless/ath/ath11k/hal_tx.h b/drivers/net/wireless/ath/ath11k/hal_tx.h
index 5217eaf9da50..cf99e2b31f9f 100644
--- a/drivers/net/wireless/ath/ath11k/hal_tx.h
+++ b/drivers/net/wireless/ath/ath11k/hal_tx.h
@@ -51,17 +51,12 @@ struct hal_tx_status {
 	u32 desc_id;
 	enum hal_wbm_tqm_rel_reason status;
 	u8 ack_rssi;
-	enum hal_tx_rate_stats_bw bw;
-	enum hal_tx_rate_stats_pkt_type pkt_type;
-	enum hal_tx_rate_stats_sgi sgi;
-	u8 mcs;
-	u16 num_tones_in_ru;
 	u32 flags; /* %HAL_TX_STATUS_FLAGS_ */
-	u32 tsf;
 	u32 ppdu_id;
 	u8 try_cnt;
 	u8 tid;
 	u16 peer_id;
+	u32 rate_stats;
 };
 
 void ath11k_hal_tx_cmd_desc_setup(struct ath11k_base *ab, void *cmd,
-- 
cgit v1.2.3-59-g8ed1b


From d0998eb84ed3cc02c15f92b07223735ddbbe9b06 Mon Sep 17 00:00:00 2001
From: John Crispin <john@phrozen.org>
Date: Fri, 13 Dec 2019 16:38:38 +0100
Subject: ath11k: optimise ath11k_dp_tx_completion_handler

the current code does 4 memcpys for each completion frame.
1) duplicate the desc
2 + 3) inside kfifo insertion
4) kfifo remove

The code simply drops the kfifo and uses a trivial ring buffer. This
requires a single memcpy for insertion. There is no removal needed as
we can simply use the inserted data for processing. As the code runs
inside the NAPI context it is atomic and there is no need for most of
the locking.

Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/dp.c     | 16 +++++-------
 drivers/net/wireless/ath/ath11k/dp.h     | 11 ++++-----
 drivers/net/wireless/ath/ath11k/dp_tx.c  | 42 +++++++++++++-------------------
 drivers/net/wireless/ath/ath11k/hal_tx.c | 13 ----------
 drivers/net/wireless/ath/ath11k/hal_tx.h |  2 --
 5 files changed, 28 insertions(+), 56 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp.c b/drivers/net/wireless/ath/ath11k/dp.c
index b966a16a930f..b112825a52ed 100644
--- a/drivers/net/wireless/ath/ath11k/dp.c
+++ b/drivers/net/wireless/ath/ath11k/dp.c
@@ -3,7 +3,6 @@
  * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
  */
 
-#include <linux/kfifo.h>
 #include "core.h"
 #include "dp_tx.h"
 #include "hal_tx.h"
@@ -828,10 +827,7 @@ void ath11k_dp_free(struct ath11k_base *ab)
 			     ath11k_dp_tx_pending_cleanup, ab);
 		idr_destroy(&dp->tx_ring[i].txbuf_idr);
 		spin_unlock_bh(&dp->tx_ring[i].tx_idr_lock);
-
-		spin_lock_bh(&dp->tx_ring[i].tx_status_lock);
-		kfifo_free(&dp->tx_ring[i].tx_status_fifo);
-		spin_unlock_bh(&dp->tx_ring[i].tx_status_lock);
+		kfree(dp->tx_ring[i].tx_status);
 	}
 
 	/* Deinit any SOC level resource */
@@ -871,17 +867,17 @@ int ath11k_dp_alloc(struct ath11k_base *ab)
 	if (ret)
 		goto fail_link_desc_cleanup;
 
-	size = roundup_pow_of_two(DP_TX_COMP_RING_SIZE);
+	size = sizeof(struct hal_wbm_release_ring) * DP_TX_COMP_RING_SIZE;
 
 	for (i = 0; i < DP_TCL_NUM_RING_MAX; i++) {
 		idr_init(&dp->tx_ring[i].txbuf_idr);
 		spin_lock_init(&dp->tx_ring[i].tx_idr_lock);
 		dp->tx_ring[i].tcl_data_ring_id = i;
 
-		spin_lock_init(&dp->tx_ring[i].tx_status_lock);
-		ret = kfifo_alloc(&dp->tx_ring[i].tx_status_fifo, size,
-				  GFP_KERNEL);
-		if (ret)
+		dp->tx_ring[i].tx_status_head = 0;
+		dp->tx_ring[i].tx_status_tail = DP_TX_COMP_RING_SIZE - 1;
+		dp->tx_ring[i].tx_status = kmalloc(size, GFP_KERNEL);
+		if (!dp->tx_ring[i].tx_status)
 			goto fail_cmn_srng_cleanup;
 	}
 
diff --git a/drivers/net/wireless/ath/ath11k/dp.h b/drivers/net/wireless/ath/ath11k/dp.h
index f7e53509ae07..2f0980f2c762 100644
--- a/drivers/net/wireless/ath/ath11k/dp.h
+++ b/drivers/net/wireless/ath/ath11k/dp.h
@@ -6,7 +6,6 @@
 #ifndef ATH11K_DP_H
 #define ATH11K_DP_H
 
-#include <linux/kfifo.h>
 #include "hal_rx.h"
 
 struct ath11k_base;
@@ -58,6 +57,8 @@ struct dp_rxdma_ring {
 	int bufs_max;
 };
 
+#define ATH11K_TX_COMPL_NEXT(x)	(((x) + 1) % DP_TX_COMP_RING_SIZE)
+
 struct dp_tx_ring {
 	u8 tcl_data_ring_id;
 	struct dp_srng tcl_data_ring;
@@ -65,11 +66,9 @@ struct dp_tx_ring {
 	struct idr txbuf_idr;
 	/* Protects txbuf_idr and num_pending */
 	spinlock_t tx_idr_lock;
-	DECLARE_KFIFO_PTR(tx_status_fifo, struct hal_wbm_release_ring);
-	/* lock to protect tx_status_fifo because tx_status_fifo can be
-	 * accessed concurrently.
-	 */
-	spinlock_t tx_status_lock;
+	struct hal_wbm_release_ring *tx_status;
+	int tx_status_head;
+	int tx_status_tail;
 };
 
 struct ath11k_pdev_mon_stats {
diff --git a/drivers/net/wireless/ath/ath11k/dp_tx.c b/drivers/net/wireless/ath/ath11k/dp_tx.c
index 28ebc414533e..3eff9e0468c5 100644
--- a/drivers/net/wireless/ath/ath11k/dp_tx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_tx.c
@@ -79,7 +79,6 @@ int ath11k_dp_tx(struct ath11k *ar, struct ath11k_vif *arvif,
 	struct hal_srng *tcl_ring;
 	struct ieee80211_hdr *hdr = (void *)skb->data;
 	struct dp_tx_ring *tx_ring;
-	u8 cached_desc[HAL_TCL_DESC_LEN];
 	void *hal_tcl_desc;
 	u8 pool_id;
 	u8 hal_ring_id;
@@ -167,8 +166,6 @@ int ath11k_dp_tx(struct ath11k *ar, struct ath11k_vif *arvif,
 	skb_cb->vif = arvif->vif;
 	skb_cb->ar = ar;
 
-	ath11k_hal_tx_cmd_desc_setup(ab, cached_desc, &ti);
-
 	hal_ring_id = tx_ring->tcl_data_ring.ring_id;
 	tcl_ring = &ab->hal.srng_list[hal_ring_id];
 
@@ -188,7 +185,8 @@ int ath11k_dp_tx(struct ath11k *ar, struct ath11k_vif *arvif,
 		goto fail_unmap_dma;
 	}
 
-	ath11k_hal_tx_desc_sync(cached_desc, hal_tcl_desc);
+	ath11k_hal_tx_cmd_desc_setup(ab, hal_tcl_desc +
+					 sizeof(struct hal_tlv_hdr), &ti);
 
 	ath11k_hal_srng_access_end(ab, tcl_ring);
 
@@ -432,47 +430,45 @@ void ath11k_dp_tx_completion_handler(struct ath11k_base *ab, int ring_id)
 	int hal_ring_id = dp->tx_ring[ring_id].tcl_comp_ring.ring_id;
 	struct hal_srng *status_ring = &ab->hal.srng_list[hal_ring_id];
 	struct sk_buff *msdu;
-	struct hal_wbm_release_ring tx_status;
 	struct hal_tx_status ts;
 	struct dp_tx_ring *tx_ring = &dp->tx_ring[ring_id];
 	u32 *desc;
 	u32 msdu_id;
 	u8 mac_id;
 
-	spin_lock_bh(&status_ring->lock);
-
 	ath11k_hal_srng_access_begin(ab, status_ring);
 
-	spin_lock_bh(&tx_ring->tx_status_lock);
-	while (!kfifo_is_full(&tx_ring->tx_status_fifo) &&
+	while ((ATH11K_TX_COMPL_NEXT(tx_ring->tx_status_head) !=
+		tx_ring->tx_status_tail) &&
 	       (desc = ath11k_hal_srng_dst_get_next_entry(ab, status_ring))) {
-		ath11k_hal_tx_status_desc_sync((void *)desc,
-					       (void *)&tx_status);
-		kfifo_put(&tx_ring->tx_status_fifo, tx_status);
+		memcpy(&tx_ring->tx_status[tx_ring->tx_status_head],
+		       desc, sizeof(struct hal_wbm_release_ring));
+		tx_ring->tx_status_head =
+			ATH11K_TX_COMPL_NEXT(tx_ring->tx_status_head);
 	}
 
 	if ((ath11k_hal_srng_dst_peek(ab, status_ring) != NULL) &&
-	    kfifo_is_full(&tx_ring->tx_status_fifo)) {
+	    (ATH11K_TX_COMPL_NEXT(tx_ring->tx_status_head) == tx_ring->tx_status_tail)) {
 		/* TODO: Process pending tx_status messages when kfifo_is_full() */
 		ath11k_warn(ab, "Unable to process some of the tx_status ring desc because status_fifo is full\n");
 	}
 
-	spin_unlock_bh(&tx_ring->tx_status_lock);
-
 	ath11k_hal_srng_access_end(ab, status_ring);
-	spin_unlock_bh(&status_ring->lock);
 
-	spin_lock_bh(&tx_ring->tx_status_lock);
-	while (kfifo_get(&tx_ring->tx_status_fifo, &tx_status)) {
-		memset(&ts, 0, sizeof(ts));
-		ath11k_hal_tx_status_parse(ab, &tx_status, &ts);
+	while (ATH11K_TX_COMPL_NEXT(tx_ring->tx_status_tail) != tx_ring->tx_status_head) {
+		struct hal_wbm_release_ring *tx_status;
+
+		tx_ring->tx_status_tail =
+			ATH11K_TX_COMPL_NEXT(tx_ring->tx_status_tail);
+		tx_status = &tx_ring->tx_status[tx_ring->tx_status_tail];
+		ath11k_hal_tx_status_parse(ab, tx_status, &ts);
 
 		mac_id = FIELD_GET(DP_TX_DESC_ID_MAC_ID, ts.desc_id);
 		msdu_id = FIELD_GET(DP_TX_DESC_ID_MSDU_ID, ts.desc_id);
 
 		if (ts.buf_rel_source == HAL_WBM_REL_SRC_MODULE_FW) {
 			ath11k_dp_tx_process_htt_tx_complete(ab,
-							     (void *)&tx_status,
+							     (void *)tx_status,
 							     mac_id, msdu_id,
 							     tx_ring);
 			continue;
@@ -494,12 +490,8 @@ void ath11k_dp_tx_completion_handler(struct ath11k_base *ab, int ring_id)
 		if (atomic_dec_and_test(&ar->dp.num_tx_pending))
 			wake_up(&ar->dp.tx_empty_waitq);
 
-		/* TODO: Locking optimization so that tx_completion for an msdu
-		 * is not called with tx_status_lock acquired
-		 */
 		ath11k_dp_tx_complete_msdu(ar, msdu, &ts);
 	}
-	spin_unlock_bh(&tx_ring->tx_status_lock);
 }
 
 int ath11k_dp_tx_send_reo_cmd(struct ath11k_base *ab, struct dp_rx_tid *rx_tid,
diff --git a/drivers/net/wireless/ath/ath11k/hal_tx.c b/drivers/net/wireless/ath/ath11k/hal_tx.c
index c2a3a9d2e534..e8710bbbbc3a 100644
--- a/drivers/net/wireless/ath/ath11k/hal_tx.c
+++ b/drivers/net/wireless/ath/ath11k/hal_tx.c
@@ -74,19 +74,6 @@ void ath11k_hal_tx_cmd_desc_setup(struct ath11k_base *ab, void *cmd,
 	tcl_cmd->info4 = 0;
 }
 
-/* Commit the descriptor to hardware */
-void ath11k_hal_tx_desc_sync(void *tx_desc_cached, void *hw_desc)
-{
-	memcpy(hw_desc + sizeof(struct hal_tlv_hdr), tx_desc_cached,
-	       sizeof(struct hal_tcl_data_cmd));
-}
-
-/* Get the descriptor status from hardware */
-void ath11k_hal_tx_status_desc_sync(void *hw_desc, void *local_desc)
-{
-	memcpy(local_desc, hw_desc, HAL_TX_STATUS_DESC_LEN);
-}
-
 void ath11k_hal_tx_status_parse(struct ath11k_base *ab,
 				struct hal_wbm_release_ring *desc,
 				struct hal_tx_status *ts)
diff --git a/drivers/net/wireless/ath/ath11k/hal_tx.h b/drivers/net/wireless/ath/ath11k/hal_tx.h
index cf99e2b31f9f..5fe89b729a6e 100644
--- a/drivers/net/wireless/ath/ath11k/hal_tx.h
+++ b/drivers/net/wireless/ath/ath11k/hal_tx.h
@@ -61,11 +61,9 @@ struct hal_tx_status {
 
 void ath11k_hal_tx_cmd_desc_setup(struct ath11k_base *ab, void *cmd,
 				  struct hal_tx_info *ti);
-void ath11k_hal_tx_desc_sync(void *tx_desc_cached, void *hw_desc);
 void ath11k_hal_tx_status_parse(struct ath11k_base *ab,
 				struct hal_wbm_release_ring *desc,
 				struct hal_tx_status *ts);
-void ath11k_hal_tx_status_desc_sync(void *hw_desc, void *local_desc);
 void ath11k_hal_tx_set_dscp_tid_map(struct ath11k_base *ab, int id);
 int ath11k_hal_reo_cmd_send(struct ath11k_base *ab, struct hal_srng *srng,
 			    enum hal_reo_cmd_type type,
-- 
cgit v1.2.3-59-g8ed1b


From 2ad578fdb7b544e2000cbec7bc1c2796208f3fcf Mon Sep 17 00:00:00 2001
From: John Crispin <john@phrozen.org>
Date: Fri, 13 Dec 2019 16:38:39 +0100
Subject: ath11k: optimize ath11k_hal_tx_status_parse

Moving the function into dp_tx.c allows gcc to optimize the code better
and also avoid chace invalidates and context switches.

Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/dp_tx.c  | 40 ++++++++++++++++++++++++++---
 drivers/net/wireless/ath/ath11k/hal_tx.c | 43 --------------------------------
 drivers/net/wireless/ath/ath11k/hal_tx.h |  4 ---
 3 files changed, 37 insertions(+), 50 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp_tx.c b/drivers/net/wireless/ath/ath11k/dp_tx.c
index 3eff9e0468c5..e994021e9355 100644
--- a/drivers/net/wireless/ath/ath11k/dp_tx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_tx.c
@@ -423,6 +423,37 @@ exit:
 	rcu_read_unlock();
 }
 
+static inline void ath11k_dp_tx_status_parse(struct ath11k_base *ab,
+					     struct hal_wbm_release_ring *desc,
+					     struct hal_tx_status *ts)
+{
+	ts->buf_rel_source =
+		FIELD_GET(HAL_WBM_RELEASE_INFO0_REL_SRC_MODULE, desc->info0);
+	if (ts->buf_rel_source != HAL_WBM_REL_SRC_MODULE_FW &&
+	    ts->buf_rel_source != HAL_WBM_REL_SRC_MODULE_TQM)
+		return;
+
+	if (ts->buf_rel_source == HAL_WBM_REL_SRC_MODULE_FW)
+		return;
+
+	ts->status = FIELD_GET(HAL_WBM_RELEASE_INFO0_TQM_RELEASE_REASON,
+			       desc->info0);
+	ts->ppdu_id = FIELD_GET(HAL_WBM_RELEASE_INFO1_TQM_STATUS_NUMBER,
+				desc->info1);
+	ts->try_cnt = FIELD_GET(HAL_WBM_RELEASE_INFO1_TRANSMIT_COUNT,
+				desc->info1);
+	ts->ack_rssi = FIELD_GET(HAL_WBM_RELEASE_INFO2_ACK_FRAME_RSSI,
+				 desc->info2);
+	if (desc->info2 & HAL_WBM_RELEASE_INFO2_FIRST_MSDU)
+		ts->flags |= HAL_TX_STATUS_FLAGS_FIRST_MSDU;
+	ts->peer_id = FIELD_GET(HAL_WBM_RELEASE_INFO3_PEER_ID, desc->info3);
+	ts->tid = FIELD_GET(HAL_WBM_RELEASE_INFO3_TID, desc->info3);
+	if (desc->rate_stats.info0 & HAL_TX_RATE_STATS_INFO0_VALID)
+		ts->rate_stats = desc->rate_stats.info0;
+	else
+		ts->rate_stats = 0;
+}
+
 void ath11k_dp_tx_completion_handler(struct ath11k_base *ab, int ring_id)
 {
 	struct ath11k *ar;
@@ -457,14 +488,17 @@ void ath11k_dp_tx_completion_handler(struct ath11k_base *ab, int ring_id)
 
 	while (ATH11K_TX_COMPL_NEXT(tx_ring->tx_status_tail) != tx_ring->tx_status_head) {
 		struct hal_wbm_release_ring *tx_status;
+		u32 desc_id;
 
 		tx_ring->tx_status_tail =
 			ATH11K_TX_COMPL_NEXT(tx_ring->tx_status_tail);
 		tx_status = &tx_ring->tx_status[tx_ring->tx_status_tail];
-		ath11k_hal_tx_status_parse(ab, tx_status, &ts);
+		ath11k_dp_tx_status_parse(ab, tx_status, &ts);
 
-		mac_id = FIELD_GET(DP_TX_DESC_ID_MAC_ID, ts.desc_id);
-		msdu_id = FIELD_GET(DP_TX_DESC_ID_MSDU_ID, ts.desc_id);
+		desc_id = FIELD_GET(BUFFER_ADDR_INFO1_SW_COOKIE,
+				    tx_status->buf_addr_info.info1);
+		mac_id = FIELD_GET(DP_TX_DESC_ID_MAC_ID, desc_id);
+		msdu_id = FIELD_GET(DP_TX_DESC_ID_MSDU_ID, desc_id);
 
 		if (ts.buf_rel_source == HAL_WBM_REL_SRC_MODULE_FW) {
 			ath11k_dp_tx_process_htt_tx_complete(ab,
diff --git a/drivers/net/wireless/ath/ath11k/hal_tx.c b/drivers/net/wireless/ath/ath11k/hal_tx.c
index e8710bbbbc3a..e4aa7e8a1284 100644
--- a/drivers/net/wireless/ath/ath11k/hal_tx.c
+++ b/drivers/net/wireless/ath/ath11k/hal_tx.c
@@ -74,49 +74,6 @@ void ath11k_hal_tx_cmd_desc_setup(struct ath11k_base *ab, void *cmd,
 	tcl_cmd->info4 = 0;
 }
 
-void ath11k_hal_tx_status_parse(struct ath11k_base *ab,
-				struct hal_wbm_release_ring *desc,
-				struct hal_tx_status *ts)
-{
-	ts->buf_rel_source =
-		FIELD_GET(HAL_WBM_RELEASE_INFO0_REL_SRC_MODULE, desc->info0);
-	if (ts->buf_rel_source != HAL_WBM_REL_SRC_MODULE_FW &&
-	    ts->buf_rel_source != HAL_WBM_REL_SRC_MODULE_TQM)
-		return;
-
-	ts->desc_id = FIELD_GET(BUFFER_ADDR_INFO1_SW_COOKIE,
-				desc->buf_addr_info.info1);
-
-	if (ts->buf_rel_source == HAL_WBM_REL_SRC_MODULE_FW)
-		return;
-
-	ts->status = FIELD_GET(HAL_WBM_RELEASE_INFO0_TQM_RELEASE_REASON,
-			       desc->info0);
-	ts->ppdu_id = FIELD_GET(HAL_WBM_RELEASE_INFO1_TQM_STATUS_NUMBER,
-				desc->info1);
-	ts->try_cnt = FIELD_GET(HAL_WBM_RELEASE_INFO1_TRANSMIT_COUNT,
-				desc->info1);
-
-	ts->ack_rssi = FIELD_GET(HAL_WBM_RELEASE_INFO2_ACK_FRAME_RSSI,
-				 desc->info2);
-	if (desc->info2 & HAL_WBM_RELEASE_INFO2_FIRST_MSDU)
-		ts->flags |= HAL_TX_STATUS_FLAGS_FIRST_MSDU;
-
-	if (desc->info2 & HAL_WBM_RELEASE_INFO2_LAST_MSDU)
-		ts->flags |= HAL_TX_STATUS_FLAGS_LAST_MSDU;
-
-	if (desc->info2 & HAL_WBM_RELEASE_INFO2_MSDU_IN_AMSDU)
-		ts->flags |= HAL_TX_STATUS_FLAGS_MSDU_IN_AMSDU;
-
-	ts->peer_id = FIELD_GET(HAL_WBM_RELEASE_INFO3_PEER_ID, desc->info3);
-	ts->tid = FIELD_GET(HAL_WBM_RELEASE_INFO3_TID, desc->info3);
-
-	if (!(desc->rate_stats.info0 & HAL_TX_RATE_STATS_INFO0_VALID))
-		return;
-
-	ts->rate_stats = desc->rate_stats.info0;
-}
-
 void ath11k_hal_tx_set_dscp_tid_map(struct ath11k_base *ab, int id)
 {
 	u32 ctrl_reg_val;
diff --git a/drivers/net/wireless/ath/ath11k/hal_tx.h b/drivers/net/wireless/ath/ath11k/hal_tx.h
index 5fe89b729a6e..ce48a61bfb66 100644
--- a/drivers/net/wireless/ath/ath11k/hal_tx.h
+++ b/drivers/net/wireless/ath/ath11k/hal_tx.h
@@ -48,7 +48,6 @@ struct hal_tx_info {
 /* Tx status parsed from srng desc */
 struct hal_tx_status {
 	enum hal_wbm_rel_src_module buf_rel_source;
-	u32 desc_id;
 	enum hal_wbm_tqm_rel_reason status;
 	u8 ack_rssi;
 	u32 flags; /* %HAL_TX_STATUS_FLAGS_ */
@@ -61,9 +60,6 @@ struct hal_tx_status {
 
 void ath11k_hal_tx_cmd_desc_setup(struct ath11k_base *ab, void *cmd,
 				  struct hal_tx_info *ti);
-void ath11k_hal_tx_status_parse(struct ath11k_base *ab,
-				struct hal_wbm_release_ring *desc,
-				struct hal_tx_status *ts);
 void ath11k_hal_tx_set_dscp_tid_map(struct ath11k_base *ab, int id);
 int ath11k_hal_reo_cmd_send(struct ath11k_base *ab, struct hal_srng *srng,
 			    enum hal_reo_cmd_type type,
-- 
cgit v1.2.3-59-g8ed1b


From 485add35771b07e56833cff77e07936ba78e731c Mon Sep 17 00:00:00 2001
From: Pradeep Kumar Chitrapu <pradeepc@codeaurora.org>
Date: Fri, 13 Dec 2019 16:34:56 +0100
Subject: ath11k: fix pdev when invoking ath11k_wmi_send_twt_enable_cmd()

The code currently uses the wrong pdev id when enabling TWT. Fix this by
using the correct ones.

Fixes: e65a616f4e74 ("ath11k: add TWT support")
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Pradeep Kumar Chitrapu <pradeepc@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/mac.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index 3450c22cab64..b8cae42d7f7c 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -1917,9 +1917,9 @@ static void ath11k_mac_op_bss_info_changed(struct ieee80211_hw *hw,
 
 	if (changed & BSS_CHANGED_TWT) {
 		if (info->twt_requester || info->twt_responder)
-			ath11k_wmi_send_twt_enable_cmd(ar, ar->pdev_idx);
+			ath11k_wmi_send_twt_enable_cmd(ar, ar->pdev->pdev_id);
 		else
-			ath11k_wmi_send_twt_disable_cmd(ar, ar->pdev_idx);
+			ath11k_wmi_send_twt_disable_cmd(ar, ar->pdev->pdev_id);
 	}
 
 	if (changed & BSS_CHANGED_HE_OBSS_PD)
-- 
cgit v1.2.3-59-g8ed1b


From aacb46223e084d4e6c14b48962c4816b0baa1f0f Mon Sep 17 00:00:00 2001
From: Pradeep Kumar Chitrapu <pradeepc@codeaurora.org>
Date: Fri, 13 Dec 2019 16:34:57 +0100
Subject: ath11k: set the BA buffer size to 256 when HE is enabled

This patch sets the correct BA buffer size when we are in HE mode.
Without this change we are not able to receive 256 bitmapped BA frames.

Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Pradeep Kumar Chitrapu <pradeepc@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/mac.c | 14 ++++++++++++++
 drivers/net/wireless/ath/ath11k/wmi.h |  2 ++
 2 files changed, 16 insertions(+)

diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index b8cae42d7f7c..556eef9881a7 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -1758,6 +1758,20 @@ static void ath11k_mac_op_bss_info_changed(struct ieee80211_hw *hw,
 		if (ret)
 			ath11k_warn(ar->ab, "failed to update bcn template: %d\n",
 				    ret);
+
+		if (vif->bss_conf.he_support) {
+			ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id,
+							    WMI_VDEV_PARAM_BA_MODE,
+							    WMI_BA_MODE_BUFFER_SIZE_256);
+			if (ret)
+				ath11k_warn(ar->ab,
+					    "failed to set BA BUFFER SIZE 256 for vdev: %d\n",
+					    arvif->vdev_id);
+			else
+				ath11k_dbg(ar->ab, ATH11K_DBG_MAC,
+					   "Set BA BUFFER SIZE 256 for VDEV: %d\n",
+					   arvif->vdev_id);
+		}
 	}
 
 	if (changed & (BSS_CHANGED_BEACON_INFO | BSS_CHANGED_BEACON)) {
diff --git a/drivers/net/wireless/ath/ath11k/wmi.h b/drivers/net/wireless/ath/ath11k/wmi.h
index ab983aac604b..58a19857185e 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.h
+++ b/drivers/net/wireless/ath/ath11k/wmi.h
@@ -54,6 +54,7 @@ struct wmi_tlv {
 #define WLAN_SCAN_PARAMS_MAX_BSSID   4
 #define WLAN_SCAN_PARAMS_MAX_IE_LEN  256
 
+#define WMI_BA_MODE_BUFFER_SIZE_256  3
 /*
  * HW mode config type replicated from FW header
  * @WMI_HOST_HW_MODE_SINGLE: Only one PHY is active.
@@ -1003,6 +1004,7 @@ enum wmi_tlv_vdev_param {
 	WMI_VDEV_PARAM_HE_RANGE_EXT,
 	WMI_VDEV_PARAM_ENABLE_BCAST_PROBE_RESPONSE,
 	WMI_VDEV_PARAM_FILS_MAX_CHANNEL_GUARD_TIME,
+	WMI_VDEV_PARAM_BA_MODE = 0x7e,
 	WMI_VDEV_PARAM_SET_HE_SOUNDING_MODE = 0x87,
 	WMI_VDEV_PARAM_PROTOTYPE = 0x8000,
 	WMI_VDEV_PARAM_BSS_COLOR,
-- 
cgit v1.2.3-59-g8ed1b


From 20c3c4fd39677fd5d909b6953a21411c7d200d2f Mon Sep 17 00:00:00 2001
From: John Crispin <john@phrozen.org>
Date: Thu, 5 Dec 2019 07:26:44 +0100
Subject: ath11k: add some missing __packed qualifiers

A few of the WMI parameter structs were missing this.

Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/wmi.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/wmi.h b/drivers/net/wireless/ath/ath11k/wmi.h
index 58a19857185e..eed6a1af8a2a 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.h
+++ b/drivers/net/wireless/ath/ath11k/wmi.h
@@ -4564,12 +4564,12 @@ struct wmi_twt_enable_params_cmd {
 	u32 mode_check_interval;
 	u32 add_sta_slot_interval;
 	u32 remove_sta_slot_interval;
-};
+} __packed;
 
 struct wmi_twt_disable_params_cmd {
 	u32 tlv_header;
 	u32 pdev_id;
-};
+} __packed;
 
 struct wmi_obss_spatial_reuse_params_cmd {
 	u32 tlv_header;
@@ -4578,7 +4578,7 @@ struct wmi_obss_spatial_reuse_params_cmd {
 	s32 obss_min;
 	s32 obss_max;
 	u32 vdev_id;
-};
+} __packed;
 
 struct target_resource_config {
 	u32 num_vdevs;
-- 
cgit v1.2.3-59-g8ed1b


From 345a4f223a7c9c6f10fb5ed7ec468ef858e9ec8a Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.king@canonical.com>
Date: Wed, 11 Dec 2019 08:34:43 +0000
Subject: ath11k: fix several spelling mistakes

There are several spelling mistakes in warning and debug messages,
fix them.

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/debug.c           | 4 ++--
 drivers/net/wireless/ath/ath11k/debug_htt_stats.c | 2 +-
 drivers/net/wireless/ath/ath11k/wmi.c             | 2 +-
 3 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/debug.c b/drivers/net/wireless/ath/ath11k/debug.c
index c27fffd13a5d..e00b5739fb00 100644
--- a/drivers/net/wireless/ath/ath11k/debug.c
+++ b/drivers/net/wireless/ath/ath11k/debug.c
@@ -704,7 +704,7 @@ static ssize_t ath11k_write_extd_rx_stats(struct file *file,
 					       DP_RX_BUFFER_SIZE, &tlv_filter);
 
 	if (ret) {
-		ath11k_warn(ar->ab, "failed to set rx filter for moniter status ring\n");
+		ath11k_warn(ar->ab, "failed to set rx filter for monitor status ring\n");
 		goto exit;
 	}
 
@@ -948,7 +948,7 @@ static ssize_t ath11k_write_pktlog_filter(struct file *file,
 					       HAL_RXDMA_MONITOR_STATUS,
 					       DP_RX_BUFFER_SIZE, &tlv_filter);
 	if (ret) {
-		ath11k_warn(ar->ab, "failed to set rx filter for moniter status ring\n");
+		ath11k_warn(ar->ab, "failed to set rx filter for monitor status ring\n");
 		goto out;
 	}
 
diff --git a/drivers/net/wireless/ath/ath11k/debug_htt_stats.c b/drivers/net/wireless/ath/ath11k/debug_htt_stats.c
index 27b301bc1a1b..a824826f562c 100644
--- a/drivers/net/wireless/ath/ath11k/debug_htt_stats.c
+++ b/drivers/net/wireless/ath/ath11k/debug_htt_stats.c
@@ -3512,7 +3512,7 @@ htt_print_rx_pdev_fw_stats_phy_err_tlv(const void *tag_buf,
 	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_RX_PDEV_FW_STATS_PHY_ERR_TLV:");
 	len += HTT_DBG_OUT(buf + len, buf_len - len, "mac_id__word = %u",
 			   htt_stats_buf->mac_id__word);
-	len += HTT_DBG_OUT(buf + len, buf_len - len, "tota_phy_err_nct = %u",
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "total_phy_err_nct = %u",
 			   htt_stats_buf->total_phy_err_cnt);
 
 	ARRAY_TO_STRING(phy_errs,
diff --git a/drivers/net/wireless/ath/ath11k/wmi.c b/drivers/net/wireless/ath/ath11k/wmi.c
index b16bfb50d73e..473f67620b9b 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.c
+++ b/drivers/net/wireless/ath/ath11k/wmi.c
@@ -2574,7 +2574,7 @@ ath11k_wmi_send_twt_disable_cmd(struct ath11k *ar, u32 pdev_id)
 	ret = ath11k_wmi_cmd_send(wmi, skb,
 				  WMI_TWT_DISABLE_CMDID);
 	if (ret) {
-		ath11k_warn(ab, "Failed to send WMI_TWT_DIeABLE_CMDID");
+		ath11k_warn(ab, "Failed to send WMI_TWT_DISABLE_CMDID");
 		dev_kfree_skb(skb);
 	}
 	return ret;
-- 
cgit v1.2.3-59-g8ed1b


From 509421acab6926a0a3edc3120d8b2a62ef2ee7a9 Mon Sep 17 00:00:00 2001
From: Nathan Chancellor <natechancellor@gmail.com>
Date: Wed, 11 Dec 2019 12:22:52 -0700
Subject: ath11k: Remove unnecessary enum scan_priority

Clang warns:

drivers/net/wireless/ath/ath11k/wmi.c:1827:23: warning: implicit
conversion from enumeration type 'enum wmi_scan_priority' to different
enumeration type 'enum scan_priority' [-Wenum-conversion]
        arg->scan_priority = WMI_SCAN_PRIORITY_LOW;
                           ~ ^~~~~~~~~~~~~~~~~~~~~
1 warning generated.

wmi_scan_priority and scan_priority have the same values but the wmi one
has WMI prefixed to the names. Since that enum is already being used,
get rid of scan_priority and switch its one use to wmi_scan_priority to
fix this warning.

Fixes: d5c65159f289 ("ath11k: driver for Qualcomm IEEE 802.11ax devices")
Link: https://github.com/ClangBuiltLinux/linux/issues/808
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/wmi.h | 11 +----------
 1 file changed, 1 insertion(+), 10 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/wmi.h b/drivers/net/wireless/ath/ath11k/wmi.h
index eed6a1af8a2a..5b64eb2d56ef 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.h
+++ b/drivers/net/wireless/ath/ath11k/wmi.h
@@ -2904,15 +2904,6 @@ struct wmi_bcn_offload_ctrl_cmd {
 	u32 bcn_ctrl_op;
 } __packed;
 
-enum scan_priority {
-	SCAN_PRIORITY_VERY_LOW,
-	SCAN_PRIORITY_LOW,
-	SCAN_PRIORITY_MEDIUM,
-	SCAN_PRIORITY_HIGH,
-	SCAN_PRIORITY_VERY_HIGH,
-	SCAN_PRIORITY_COUNT,
-};
-
 enum scan_dwelltime_adaptive_mode {
 	SCAN_DWELL_MODE_DEFAULT = 0,
 	SCAN_DWELL_MODE_CONSERVATIVE = 1,
@@ -3064,7 +3055,7 @@ struct scan_req_params {
 	u32 scan_req_id;
 	u32 vdev_id;
 	u32 pdev_id;
-	enum scan_priority scan_priority;
+	enum wmi_scan_priority scan_priority;
 	union {
 		struct {
 			u32 scan_ev_started:1,
-- 
cgit v1.2.3-59-g8ed1b


From b5316db229d5afd7ba4cdba05b4c4b4c6307b5e7 Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.king@canonical.com>
Date: Thu, 12 Dec 2019 18:32:28 +0000
Subject: ath11k: fix memory leak on reg_info

Currently a return path is leaking the previously allocate reg_info. Fix
this by exiting via the return path mem_free.  Since the return value ret
is defaulted to 0, there is no need to re-assign ret to the 0 before
the goto.

Addresses-Coverity: ("Resource leak")
Fixes: d5c65159f289 ("ath11k: driver for Qualcomm IEEE 802.11ax devices")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/wmi.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath11k/wmi.c b/drivers/net/wireless/ath/ath11k/wmi.c
index 473f67620b9b..0cc58d55b308 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.c
+++ b/drivers/net/wireless/ath/ath11k/wmi.c
@@ -4552,7 +4552,7 @@ static int ath11k_reg_chan_list_event(struct ath11k_base *ab, struct sk_buff *sk
 	if (ab->default_regd[pdev_idx] && !ab->new_regd[pdev_idx] &&
 	    !memcmp((char *)ab->default_regd[pdev_idx]->alpha2,
 		    (char *)reg_info->alpha2, 2))
-		return 0;
+		goto mem_free;
 
 	/* Intersect new rules with default regd if a new country setting was
 	 * requested, i.e a default regd was already set during initialization
-- 
cgit v1.2.3-59-g8ed1b


From 08cc0f44f5ed5b5a91370f99929730dc3fe94b68 Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.king@canonical.com>
Date: Thu, 12 Dec 2019 19:10:44 +0000
Subject: ath11k: fix uninitialized variable radioup

The variable radioup is not uninitalized so it may contain a garbage
value and hence the detection of a radio that is not up is buggy.
Fix this by initializing it to zero.

Addresses-Coverity: ("Uninitalized scalar variable")
Fixes: d5c65159f289 ("ath11k: driver for Qualcomm IEEE 802.11ax devices")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/debug.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath11k/debug.c b/drivers/net/wireless/ath/ath11k/debug.c
index e00b5739fb00..f48daf17f2d2 100644
--- a/drivers/net/wireless/ath/ath11k/debug.c
+++ b/drivers/net/wireless/ath/ath11k/debug.c
@@ -541,7 +541,7 @@ static ssize_t ath11k_write_simulate_fw_crash(struct file *file,
 	struct ath11k *ar = ab->pdevs[0].ar;
 	char buf[32] = {0};
 	ssize_t rc;
-	int i, ret, radioup;
+	int i, ret, radioup = 0;
 
 	for (i = 0; i < ab->num_radios; i++) {
 		pdev = &ab->pdevs[i];
-- 
cgit v1.2.3-59-g8ed1b


From 4358bcb54bb9ede559e60685b9e544e723197ca1 Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.king@canonical.com>
Date: Thu, 12 Dec 2019 19:42:51 +0000
Subject: ath11k: fix missing free of skb on error return path

The error handling when the call to ath11k_hal_srng_get_entrysize fails
leaks skb, fix this by returning via the err_free return path that will
ensure the skb is free'd.

Addresses-Coverity: ("Resource leak")
Fixes: d5c65159f289 ("ath11k: driver for Qualcomm IEEE 802.11ax devices")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/dp_tx.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp_tx.c b/drivers/net/wireless/ath/ath11k/dp_tx.c
index e994021e9355..918305dda106 100644
--- a/drivers/net/wireless/ath/ath11k/dp_tx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_tx.c
@@ -669,8 +669,10 @@ int ath11k_dp_tx_htt_srng_setup(struct ath11k_base *ab, u32 ring_id,
 				 HAL_ADDR_MSB_REG_SHIFT;
 
 	ret = ath11k_hal_srng_get_entrysize(ring_type);
-	if (ret < 0)
-		return -EINVAL;
+	if (ret < 0) {
+		ret = -EINVAL;
+		goto err_free;
+	}
 
 	ring_entry_sz = ret;
 
-- 
cgit v1.2.3-59-g8ed1b


From cd21c827a1555276325880ea431a6a27a9186e30 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@oracle.com>
Date: Fri, 13 Dec 2019 13:47:54 +0300
Subject: ath11k: delete a stray unlock in ath11k_dbg_htt_stats_req()

The callers unlock this lock so this error path has a double unlock.

Fixes: d5c65159f289 ("ath11k: driver for Qualcomm IEEE 802.11ax devices")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/debug_htt_stats.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath11k/debug_htt_stats.c b/drivers/net/wireless/ath/ath11k/debug_htt_stats.c
index a824826f562c..090fffa5e53c 100644
--- a/drivers/net/wireless/ath/ath11k/debug_htt_stats.c
+++ b/drivers/net/wireless/ath/ath11k/debug_htt_stats.c
@@ -4287,7 +4287,6 @@ int ath11k_dbg_htt_stats_req(struct ath11k *ar)
 	ret = ath11k_dp_tx_htt_h2t_ext_stats_req(ar, type, &cfg_params, cookie);
 	if (ret) {
 		ath11k_warn(ar->ab, "failed to send htt stats request: %d\n", ret);
-		mutex_unlock(&ar->conf_mutex);
 		return ret;
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From c76fa846b0e1a1cdf6fa1e04c0c3985411cc973d Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@oracle.com>
Date: Fri, 13 Dec 2019 13:51:27 +0300
Subject: ath11k: checking for NULL vs IS_ERR()

The ath11k_ce_alloc_ring() function returns error pointers on error, not
NULL.  The rest of the driver assumes that "pipe->src_ring" is either
valid or NULL so this patch introduces a temporary varaible to avoid
leaving it as an error pointer.

Fixes: d5c65159f289 ("ath11k: driver for Qualcomm IEEE 802.11ax devices")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/ce.c | 23 +++++++++++++----------
 1 file changed, 13 insertions(+), 10 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/ce.c b/drivers/net/wireless/ath/ath11k/ce.c
index a6572b414303..cdd40c8fc867 100644
--- a/drivers/net/wireless/ath/ath11k/ce.c
+++ b/drivers/net/wireless/ath/ath11k/ce.c
@@ -441,6 +441,7 @@ static int ath11k_ce_alloc_pipe(struct ath11k_base *ab, int ce_id)
 {
 	struct ath11k_ce_pipe *pipe = &ab->ce.ce_pipe[ce_id];
 	const struct ce_attr *attr = &host_ce_config_wlan[ce_id];
+	struct ath11k_ce_ring *ring;
 	int nentries;
 	int desc_sz;
 
@@ -450,24 +451,26 @@ static int ath11k_ce_alloc_pipe(struct ath11k_base *ab, int ce_id)
 		pipe->send_cb = ath11k_ce_send_done_cb;
 		nentries = roundup_pow_of_two(attr->src_nentries);
 		desc_sz = ath11k_hal_ce_get_desc_size(HAL_CE_DESC_SRC);
-		pipe->src_ring = ath11k_ce_alloc_ring(ab, nentries, desc_sz);
-		if (!pipe->src_ring)
-			return -ENOMEM;
+		ring = ath11k_ce_alloc_ring(ab, nentries, desc_sz);
+		if (IS_ERR(ring))
+			return PTR_ERR(ring);
+		pipe->src_ring = ring;
 	}
 
 	if (attr->dest_nentries) {
 		pipe->recv_cb = attr->recv_cb;
 		nentries = roundup_pow_of_two(attr->dest_nentries);
 		desc_sz = ath11k_hal_ce_get_desc_size(HAL_CE_DESC_DST);
-		pipe->dest_ring = ath11k_ce_alloc_ring(ab, nentries, desc_sz);
-
-		if (!pipe->dest_ring)
-			return -ENOMEM;
+		ring = ath11k_ce_alloc_ring(ab, nentries, desc_sz);
+		if (IS_ERR(ring))
+			return PTR_ERR(ring);
+		pipe->dest_ring = ring;
 
 		desc_sz = ath11k_hal_ce_get_desc_size(HAL_CE_DESC_DST_STATUS);
-		pipe->status_ring = ath11k_ce_alloc_ring(ab, nentries, desc_sz);
-		if (!pipe->status_ring)
-			return -ENOMEM;
+		ring = ath11k_ce_alloc_ring(ab, nentries, desc_sz);
+		if (IS_ERR(ring))
+			return PTR_ERR(ring);
+		pipe->status_ring = ring;
 	}
 
 	return 0;
-- 
cgit v1.2.3-59-g8ed1b


From 269663f18474b739fcf45389c8fea82de3332e2f Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@oracle.com>
Date: Fri, 13 Dec 2019 13:52:19 +0300
Subject: ath11k: remove an unneeded NULL check

The list iterator is always non-NULL so it doesn't need to be checked.
I also removed the unnecessary initializer because the list iterator is
always initialized.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/dp_rx.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c b/drivers/net/wireless/ath/ath11k/dp_rx.c
index 1b3b65c0038c..3a3dc7680622 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
@@ -1225,12 +1225,12 @@ static
 struct htt_ppdu_stats_info *ath11k_dp_htt_get_ppdu_desc(struct ath11k *ar,
 							u32 ppdu_id)
 {
-	struct htt_ppdu_stats_info *ppdu_info = NULL;
+	struct htt_ppdu_stats_info *ppdu_info;
 
 	spin_lock_bh(&ar->data_lock);
 	if (!list_empty(&ar->ppdu_stats_info)) {
 		list_for_each_entry(ppdu_info, &ar->ppdu_stats_info, list) {
-			if (ppdu_info && ppdu_info->ppdu_id == ppdu_id) {
+			if (ppdu_info->ppdu_id == ppdu_id) {
 				spin_unlock_bh(&ar->data_lock);
 				return ppdu_info;
 			}
-- 
cgit v1.2.3-59-g8ed1b


From d1389e19e68270601c75acc14a890264f66b80a1 Mon Sep 17 00:00:00 2001
From: zhengbin <zhengbin13@huawei.com>
Date: Sat, 14 Dec 2019 18:22:34 +0800
Subject: ath11k: Remove unneeded semicolon

Fixes coccicheck warning:

drivers/net/wireless/ath/ath11k/wmi.h:2570:2-3: Unneeded semicolon

Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: zhengbin <zhengbin13@huawei.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/wmi.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath11k/wmi.h b/drivers/net/wireless/ath/ath11k/wmi.h
index 5b64eb2d56ef..780e72c2844a 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.h
+++ b/drivers/net/wireless/ath/ath11k/wmi.h
@@ -2569,7 +2569,7 @@ static inline const char *ath11k_wmi_phymode_str(enum wmi_phy_mode mode)
 		/* no default handler to allow compiler to check that the
 		 * enum is fully handled
 		 */
-	};
+	}
 
 	return "<unknown>";
 }
-- 
cgit v1.2.3-59-g8ed1b


From 0ef332951e856efa89507cdd13ba8f4fb8d4db12 Mon Sep 17 00:00:00 2001
From: Johan Hovold <johan@kernel.org>
Date: Tue, 10 Dec 2019 12:44:20 +0100
Subject: ath9k: fix storage endpoint lookup

Make sure to use the current alternate setting when verifying the
storage interface descriptors to avoid submitting an URB to an invalid
endpoint.

Failing to do so could cause the driver to misbehave or trigger a WARN()
in usb_submit_urb() that kernels with panic_on_warn set would choke on.

Fixes: 36bcce430657 ("ath9k_htc: Handle storage devices")
Cc: stable <stable@vger.kernel.org>     # 2.6.39
Signed-off-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath9k/hif_usb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath9k/hif_usb.c b/drivers/net/wireless/ath/ath9k/hif_usb.c
index fb649d85b8fc..dd0c32379375 100644
--- a/drivers/net/wireless/ath/ath9k/hif_usb.c
+++ b/drivers/net/wireless/ath/ath9k/hif_usb.c
@@ -1216,7 +1216,7 @@ err_fw:
 static int send_eject_command(struct usb_interface *interface)
 {
 	struct usb_device *udev = interface_to_usbdev(interface);
-	struct usb_host_interface *iface_desc = &interface->altsetting[0];
+	struct usb_host_interface *iface_desc = interface->cur_altsetting;
 	struct usb_endpoint_descriptor *endpoint;
 	unsigned char *cmd;
 	u8 bulk_out_ep;
-- 
cgit v1.2.3-59-g8ed1b


From f998f9fcf9033b1e7f76bf4c17969b74be664dce Mon Sep 17 00:00:00 2001
From: Eduardo Abinader <eduardoabinader@gmail.com>
Date: Tue, 17 Dec 2019 15:12:47 +0100
Subject: wcn36xx: disable HW_CONNECTION_MONITOR

Whenever the signal stregth decays smoothly and physical connnection
is already gone and no deauth has arrived, the qcom soc is not
able to indicate neither WCN36XX_HAL_MISSED_BEACON_IND nor
WCN36XX_HAL_MISSED_BEACON_IND. It was noticed that such situation gets
even more reproducible, when the driver fails to enter bmps mode - which is
highly likely to occur. Thus, in order to provide proper disconnection
of the connected STA, let mac80211 handle it, instead of wcn3xx driver.

Signed-off-by: Eduardo Abinader <eduardoabinader@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/wcn36xx/main.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/wireless/ath/wcn36xx/main.c b/drivers/net/wireless/ath/wcn36xx/main.c
index c30fdd0cbf1e..e49c306e0eef 100644
--- a/drivers/net/wireless/ath/wcn36xx/main.c
+++ b/drivers/net/wireless/ath/wcn36xx/main.c
@@ -1169,7 +1169,6 @@ static int wcn36xx_init_ieee80211(struct wcn36xx *wcn)
 
 	ieee80211_hw_set(wcn->hw, TIMING_BEACON_ONLY);
 	ieee80211_hw_set(wcn->hw, AMPDU_AGGREGATION);
-	ieee80211_hw_set(wcn->hw, CONNECTION_MONITOR);
 	ieee80211_hw_set(wcn->hw, SUPPORTS_PS);
 	ieee80211_hw_set(wcn->hw, SIGNAL_DBM);
 	ieee80211_hw_set(wcn->hw, HAS_RATE_CONTROL);
-- 
cgit v1.2.3-59-g8ed1b


From 5cc509aa83c6acd2c5cd94f99065c39d2bd0a490 Mon Sep 17 00:00:00 2001
From: Navid Emamdoost <navid.emamdoost@gmail.com>
Date: Fri, 22 Nov 2019 13:19:48 -0600
Subject: brcmfmac: Fix memory leak in brcmf_p2p_create_p2pdev()

In the implementation of brcmf_p2p_create_p2pdev() the allocated memory
for p2p_vif is leaked when the mac address is the same as primary
interface. To fix this, go to error path to release p2p_vif via
brcmf_free_vif().

Fixes: cb746e47837a ("brcmfmac: check p2pdev mac address uniqueness")
Signed-off-by: Navid Emamdoost <navid.emamdoost@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c
index 7ba9f6a68645..1f5deea5a288 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c
@@ -2092,7 +2092,8 @@ static struct wireless_dev *brcmf_p2p_create_p2pdev(struct brcmf_p2p_info *p2p,
 	/* firmware requires unique mac address for p2pdev interface */
 	if (addr && ether_addr_equal(addr, pri_ifp->mac_addr)) {
 		bphy_err(drvr, "discovery vif must be different from primary interface\n");
-		return ERR_PTR(-EINVAL);
+		err = -EINVAL;
+		goto fail;
 	}
 
 	brcmf_p2p_generate_bss_mac(p2p, addr);
-- 
cgit v1.2.3-59-g8ed1b


From b8e1ada81fb71f54c5577b46e562016623f2567c Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Mon, 25 Nov 2019 12:32:26 -0600
Subject: rtlwifi: rtl8192de: Remove unused GET_XXX and SET_XXX macros

As the first step in converting from macros that get/set information
in the RX and TX descriptors, unused macros are being removed.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 .../net/wireless/realtek/rtlwifi/rtl8192de/trx.h   | 329 +--------------------
 1 file changed, 3 insertions(+), 326 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.h
index 635989e15282..40746125f1ae 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.h
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.h
@@ -42,8 +42,6 @@
 	SET_BITS_OFFSET_LE(__pdesc, 0, 16, __val)
 #define SET_TX_DESC_OFFSET(__pdesc, __val)		\
 	SET_BITS_OFFSET_LE(__pdesc, 16, 8, __val)
-#define SET_TX_DESC_BMC(__pdesc, __val)			\
-	SET_BITS_OFFSET_LE(__pdesc, 24, 1, __val)
 #define SET_TX_DESC_HTC(__pdesc, __val)			\
 	SET_BITS_OFFSET_LE(__pdesc, 25, 1, __val)
 #define SET_TX_DESC_LAST_SEG(__pdesc, __val)		\
@@ -52,31 +50,9 @@
 	SET_BITS_OFFSET_LE(__pdesc, 27, 1, __val)
 #define SET_TX_DESC_LINIP(__pdesc, __val)		\
 	SET_BITS_OFFSET_LE(__pdesc, 28, 1, __val)
-#define SET_TX_DESC_NO_ACM(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc, 29, 1, __val)
-#define SET_TX_DESC_GF(__pdesc, __val)			\
-	SET_BITS_OFFSET_LE(__pdesc, 30, 1, __val)
 #define SET_TX_DESC_OWN(__pdesc, __val)			\
 	SET_BITS_OFFSET_LE(__pdesc, 31, 1, __val)
 
-#define GET_TX_DESC_PKT_SIZE(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc, 0, 16)
-#define GET_TX_DESC_OFFSET(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc, 16, 8)
-#define GET_TX_DESC_BMC(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc, 24, 1)
-#define GET_TX_DESC_HTC(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc, 25, 1)
-#define GET_TX_DESC_LAST_SEG(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc, 26, 1)
-#define GET_TX_DESC_FIRST_SEG(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc, 27, 1)
-#define GET_TX_DESC_LINIP(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc, 28, 1)
-#define GET_TX_DESC_NO_ACM(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc, 29, 1)
-#define GET_TX_DESC_GF(__pdesc)				\
-	SHIFT_AND_MASK_LE(__pdesc, 30, 1)
 #define GET_TX_DESC_OWN(__pdesc)			\
 	SHIFT_AND_MASK_LE(__pdesc, 31, 1)
 
@@ -84,136 +60,35 @@
 	SET_BITS_OFFSET_LE(__pdesc+4, 0, 5, __val)
 #define SET_TX_DESC_AGG_ENABLE(__pdesc, __val)		\
 	SET_BITS_OFFSET_LE(__pdesc+4, 5, 1, __val)
-#define SET_TX_DESC_BK(__pdesc, __val)			\
-	SET_BITS_OFFSET_LE(__pdesc+4, 6, 1, __val)
 #define SET_TX_DESC_RDG_ENABLE(__pdesc, __val)		\
 	SET_BITS_OFFSET_LE(__pdesc+4, 7, 1, __val)
 #define SET_TX_DESC_QUEUE_SEL(__pdesc, __val)		\
 	SET_BITS_OFFSET_LE(__pdesc+4, 8, 5, __val)
-#define SET_TX_DESC_RDG_NAV_EXT(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+4, 13, 1, __val)
-#define SET_TX_DESC_LSIG_TXOP_EN(__pdesc, __val)	\
-	SET_BITS_OFFSET_LE(__pdesc+4, 14, 1, __val)
-#define SET_TX_DESC_PIFS(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+4, 15, 1, __val)
 #define SET_TX_DESC_RATE_ID(__pdesc, __val)		\
 	SET_BITS_OFFSET_LE(__pdesc+4, 16, 4, __val)
-#define SET_TX_DESC_NAV_USE_HDR(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+4, 20, 1, __val)
-#define SET_TX_DESC_EN_DESC_ID(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+4, 21, 1, __val)
 #define SET_TX_DESC_SEC_TYPE(__pdesc, __val)		\
 	SET_BITS_OFFSET_LE(__pdesc+4, 22, 2, __val)
 #define SET_TX_DESC_PKT_OFFSET(__pdesc, __val)		\
 	SET_BITS_OFFSET_LE(__pdesc+4, 26, 8, __val)
 
-#define GET_TX_DESC_MACID(__pdesc)					\
-	SHIFT_AND_MASK_LE(__pdesc+4, 0, 5)
-#define GET_TX_DESC_AGG_ENABLE(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+4, 5, 1)
-#define GET_TX_DESC_AGG_BREAK(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+4, 6, 1)
-#define GET_TX_DESC_RDG_ENABLE(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+4, 7, 1)
-#define GET_TX_DESC_QUEUE_SEL(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+4, 8, 5)
-#define GET_TX_DESC_RDG_NAV_EXT(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+4, 13, 1)
-#define GET_TX_DESC_LSIG_TXOP_EN(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+4, 14, 1)
-#define GET_TX_DESC_PIFS(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+4, 15, 1)
-#define GET_TX_DESC_RATE_ID(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+4, 16, 4)
-#define GET_TX_DESC_NAV_USE_HDR(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+4, 20, 1)
-#define GET_TX_DESC_EN_DESC_ID(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+4, 21, 1)
-#define GET_TX_DESC_SEC_TYPE(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+4, 22, 2)
-#define GET_TX_DESC_PKT_OFFSET(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+4, 24, 8)
-
-#define SET_TX_DESC_RTS_RC(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+8, 0, 6, __val)
-#define SET_TX_DESC_DATA_RC(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+8, 6, 6, __val)
-#define SET_TX_DESC_BAR_RTY_TH(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+8, 14, 2, __val)
 #define SET_TX_DESC_MORE_FRAG(__pdesc, __val)		\
 	SET_BITS_OFFSET_LE(__pdesc+8, 17, 1, __val)
-#define SET_TX_DESC_RAW(__pdesc, __val)			\
-	SET_BITS_OFFSET_LE(__pdesc+8, 18, 1, __val)
-#define SET_TX_DESC_CCX(__pdesc, __val)			\
-	SET_BITS_OFFSET_LE(__pdesc+8, 19, 1, __val)
 #define SET_TX_DESC_AMPDU_DENSITY(__pdesc, __val)	\
 	SET_BITS_OFFSET_LE(__pdesc+8, 20, 3, __val)
-#define SET_TX_DESC_ANTSEL_A(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+8, 24, 1, __val)
-#define SET_TX_DESC_ANTSEL_B(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+8, 25, 1, __val)
-#define SET_TX_DESC_TX_ANT_CCK(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+8, 26, 2, __val)
-#define SET_TX_DESC_TX_ANTL(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+8, 28, 2, __val)
-#define SET_TX_DESC_TX_ANT_HT(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+8, 30, 2, __val)
-
-#define GET_TX_DESC_RTS_RC(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+8, 0, 6)
-#define GET_TX_DESC_DATA_RC(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+8, 6, 6)
-#define GET_TX_DESC_BAR_RTY_TH(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+8, 14, 2)
-#define GET_TX_DESC_MORE_FRAG(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+8, 17, 1)
-#define GET_TX_DESC_RAW(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+8, 18, 1)
-#define GET_TX_DESC_CCX(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+8, 19, 1)
-#define GET_TX_DESC_AMPDU_DENSITY(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+8, 20, 3)
-#define GET_TX_DESC_ANTSEL_A(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+8, 24, 1)
-#define GET_TX_DESC_ANTSEL_B(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+8, 25, 1)
-#define GET_TX_DESC_TX_ANT_CCK(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+8, 26, 2)
-#define GET_TX_DESC_TX_ANTL(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+8, 28, 2)
-#define GET_TX_DESC_TX_ANT_HT(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+8, 30, 2)
-
-#define SET_TX_DESC_NEXT_HEAP_PAGE(__pdesc, __val)	\
-	SET_BITS_OFFSET_LE(__pdesc+12, 0, 8, __val)
-#define SET_TX_DESC_TAIL_PAGE(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+12, 8, 8, __val)
+
 #define SET_TX_DESC_SEQ(__pdesc, __val)			\
 	SET_BITS_OFFSET_LE(__pdesc+12, 16, 12, __val)
 #define SET_TX_DESC_PKT_ID(__pdesc, __val)		\
 	SET_BITS_OFFSET_LE(__pdesc+12, 28, 4, __val)
 
-#define GET_TX_DESC_NEXT_HEAP_PAGE(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+12, 0, 8)
-#define GET_TX_DESC_TAIL_PAGE(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+12, 8, 8)
-#define GET_TX_DESC_SEQ(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+12, 16, 12)
-#define GET_TX_DESC_PKT_ID(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+12, 28, 4)
-
 #define SET_TX_DESC_RTS_RATE(__pdesc, __val)		\
 	SET_BITS_OFFSET_LE(__pdesc+16, 0, 5, __val)
-#define SET_TX_DESC_AP_DCFE(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+16, 5, 1, __val)
 #define SET_TX_DESC_QOS(__pdesc, __val)			\
 	SET_BITS_OFFSET_LE(__pdesc+16, 6, 1, __val)
 #define SET_TX_DESC_HWSEQ_EN(__pdesc, __val)		\
 	SET_BITS_OFFSET_LE(__pdesc+16, 7, 1, __val)
 #define SET_TX_DESC_USE_RATE(__pdesc, __val)		\
 	SET_BITS_OFFSET_LE(__pdesc+16, 8, 1, __val)
-#define SET_TX_DESC_DISABLE_RTS_FB(__pdesc, __val)	\
-	SET_BITS_OFFSET_LE(__pdesc+16, 9, 1, __val)
 #define SET_TX_DESC_DISABLE_FB(__pdesc, __val)		\
 	SET_BITS_OFFSET_LE(__pdesc+16, 10, 1, __val)
 #define SET_TX_DESC_CTS2SELF(__pdesc, __val)		\
@@ -222,18 +97,8 @@
 	SET_BITS_OFFSET_LE(__pdesc+16, 12, 1, __val)
 #define SET_TX_DESC_HW_RTS_ENABLE(__pdesc, __val)	\
 	SET_BITS_OFFSET_LE(__pdesc+16, 13, 1, __val)
-#define SET_TX_DESC_PORT_ID(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+16, 14, 1, __val)
-#define SET_TX_DESC_WAIT_DCTS(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+16, 18, 1, __val)
-#define SET_TX_DESC_CTS2AP_EN(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+16, 19, 1, __val)
 #define SET_TX_DESC_TX_SUB_CARRIER(__pdesc, __val)	\
 	SET_BITS_OFFSET_LE(__pdesc+16, 20, 2, __val)
-#define SET_TX_DESC_TX_STBC(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+16, 22, 2, __val)
-#define SET_TX_DESC_DATA_SHORT(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+16, 24, 1, __val)
 #define SET_TX_DESC_DATA_BW(__pdesc, __val)		\
 	SET_BITS_OFFSET_LE(__pdesc+16, 25, 1, __val)
 #define SET_TX_DESC_RTS_SHORT(__pdesc, __val)		\
@@ -245,158 +110,29 @@
 #define SET_TX_DESC_RTS_STBC(__pdesc, __val)		\
 	SET_BITS_OFFSET_LE(__pdesc+16, 30, 2, __val)
 
-#define GET_TX_DESC_RTS_RATE(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+16, 0, 5)
-#define GET_TX_DESC_AP_DCFE(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+16, 5, 1)
-#define GET_TX_DESC_QOS(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+16, 6, 1)
-#define GET_TX_DESC_HWSEQ_EN(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+16, 7, 1)
-#define GET_TX_DESC_USE_RATE(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+16, 8, 1)
-#define GET_TX_DESC_DISABLE_RTS_FB(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+16, 9, 1)
-#define GET_TX_DESC_DISABLE_FB(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+16, 10, 1)
-#define GET_TX_DESC_CTS2SELF(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+16, 11, 1)
-#define GET_TX_DESC_RTS_ENABLE(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+16, 12, 1)
-#define GET_TX_DESC_HW_RTS_ENABLE(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+16, 13, 1)
-#define GET_TX_DESC_PORT_ID(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+16, 14, 1)
-#define GET_TX_DESC_WAIT_DCTS(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+16, 18, 1)
-#define GET_TX_DESC_CTS2AP_EN(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+16, 19, 1)
-#define GET_TX_DESC_TX_SUB_CARRIER(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+16, 20, 2)
-#define GET_TX_DESC_TX_STBC(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+16, 22, 2)
-#define GET_TX_DESC_DATA_SHORT(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+16, 24, 1)
-#define GET_TX_DESC_DATA_BW(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+16, 25, 1)
-#define GET_TX_DESC_RTS_SHORT(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+16, 26, 1)
-#define GET_TX_DESC_RTS_BW(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+16, 27, 1)
-#define GET_TX_DESC_RTS_SC(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+16, 28, 2)
-#define GET_TX_DESC_RTS_STBC(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+16, 30, 2)
-
 #define SET_TX_DESC_TX_RATE(__pdesc, __val)		\
 	SET_BITS_OFFSET_LE(__pdesc+20, 0, 6, __val)
 #define SET_TX_DESC_DATA_SHORTGI(__pdesc, __val)	\
 	SET_BITS_OFFSET_LE(__pdesc+20, 6, 1, __val)
-#define SET_TX_DESC_CCX_TAG(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+20, 7, 1, __val)
 #define SET_TX_DESC_DATA_RATE_FB_LIMIT(__pdesc, __val)	\
 	SET_BITS_OFFSET_LE(__pdesc+20, 8, 5, __val)
 #define SET_TX_DESC_RTS_RATE_FB_LIMIT(__pdesc, __val)	\
 	SET_BITS_OFFSET_LE(__pdesc+20, 13, 4, __val)
-#define SET_TX_DESC_RETRY_LIMIT_ENABLE(__pdesc, __val)	\
-	SET_BITS_OFFSET_LE(__pdesc+20, 17, 1, __val)
-#define SET_TX_DESC_DATA_RETRY_LIMIT(__pdesc, __val)	\
-	SET_BITS_OFFSET_LE(__pdesc+20, 18, 6, __val)
-#define SET_TX_DESC_USB_TXAGG_NUM(__pdesc, __val)	\
-	SET_BITS_OFFSET_LE(__pdesc+20, 24, 8, __val)
-
-#define GET_TX_DESC_TX_RATE(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+20, 0, 6)
-#define GET_TX_DESC_DATA_SHORTGI(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+20, 6, 1)
-#define GET_TX_DESC_CCX_TAG(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+20, 7, 1)
-#define GET_TX_DESC_DATA_RATE_FB_LIMIT(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+20, 8, 5)
-#define GET_TX_DESC_RTS_RATE_FB_LIMIT(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+20, 13, 4)
-#define GET_TX_DESC_RETRY_LIMIT_ENABLE(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+20, 17, 1)
-#define GET_TX_DESC_DATA_RETRY_LIMIT(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+20, 18, 6)
-#define GET_TX_DESC_USB_TXAGG_NUM(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+20, 24, 8)
-
-#define SET_TX_DESC_TXAGC_A(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+24, 0, 5, __val)
-#define SET_TX_DESC_TXAGC_B(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+24, 5, 5, __val)
-#define SET_TX_DESC_USE_MAX_LEN(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+24, 10, 1, __val)
+
 #define SET_TX_DESC_MAX_AGG_NUM(__pdesc, __val)		\
 	SET_BITS_OFFSET_LE(__pdesc+24, 11, 5, __val)
-#define SET_TX_DESC_MCSG1_MAX_LEN(__pdesc, __val)	\
-	SET_BITS_OFFSET_LE(__pdesc+24, 16, 4, __val)
-#define SET_TX_DESC_MCSG2_MAX_LEN(__pdesc, __val)	\
-	SET_BITS_OFFSET_LE(__pdesc+24, 20, 4, __val)
-#define SET_TX_DESC_MCSG3_MAX_LEN(__pdesc, __val)	\
-	SET_BITS_OFFSET_LE(__pdesc+24, 24, 4, __val)
-#define SET_TX_DESC_MCS7_SGI_MAX_LEN(__pdesc, __val)	\
-	SET_BITS_OFFSET_LE(__pdesc+24, 28, 4, __val)
-
-#define GET_TX_DESC_TXAGC_A(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+24, 0, 5)
-#define GET_TX_DESC_TXAGC_B(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+24, 5, 5)
-#define GET_TX_DESC_USE_MAX_LEN(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+24, 10, 1)
-#define GET_TX_DESC_MAX_AGG_NUM(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+24, 11, 5)
-#define GET_TX_DESC_MCSG1_MAX_LEN(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+24, 16, 4)
-#define GET_TX_DESC_MCSG2_MAX_LEN(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+24, 20, 4)
-#define GET_TX_DESC_MCSG3_MAX_LEN(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+24, 24, 4)
-#define GET_TX_DESC_MCS7_SGI_MAX_LEN(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+24, 28, 4)
 
 #define SET_TX_DESC_TX_BUFFER_SIZE(__pdesc, __val)	\
 	SET_BITS_OFFSET_LE(__pdesc+28, 0, 16, __val)
-#define SET_TX_DESC_MCSG4_MAX_LEN(__pdesc, __val)	\
-	SET_BITS_OFFSET_LE(__pdesc+28, 16, 4, __val)
-#define SET_TX_DESC_MCSG5_MAX_LEN(__pdesc, __val)	\
-	SET_BITS_OFFSET_LE(__pdesc+28, 20, 4, __val)
-#define SET_TX_DESC_MCSG6_MAX_LEN(__pdesc, __val)	\
-	SET_BITS_OFFSET_LE(__pdesc+28, 24, 4, __val)
-#define SET_TX_DESC_MCS15_SGI_MAX_LEN(__pdesc, __val)	\
-	SET_BITS_OFFSET_LE(__pdesc+28, 28, 4, __val)
-
-#define GET_TX_DESC_TX_BUFFER_SIZE(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+28, 0, 16)
-#define GET_TX_DESC_MCSG4_MAX_LEN(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+28, 16, 4)
-#define GET_TX_DESC_MCSG5_MAX_LEN(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+28, 20, 4)
-#define GET_TX_DESC_MCSG6_MAX_LEN(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+28, 24, 4)
-#define GET_TX_DESC_MCS15_SGI_MAX_LEN(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+28, 28, 4)
 
 #define SET_TX_DESC_TX_BUFFER_ADDRESS(__pdesc, __val)	\
 	SET_BITS_OFFSET_LE(__pdesc+32, 0, 32, __val)
-#define SET_TX_DESC_TX_BUFFER_ADDRESS64(__pdesc, __val) \
-	SET_BITS_OFFSET_LE(__pdesc+36, 0, 32, __val)
 
 #define GET_TX_DESC_TX_BUFFER_ADDRESS(__pdesc)		\
 	SHIFT_AND_MASK_LE(__pdesc+32, 0, 32)
-#define GET_TX_DESC_TX_BUFFER_ADDRESS64(__pdesc)	\
-	SHIFT_AND_MASK_LE(__pdesc+36, 0, 32)
 
 #define SET_TX_DESC_NEXT_DESC_ADDRESS(__pdesc, __val)	\
 	SET_BITS_OFFSET_LE(__pdesc+40, 0, 32, __val)
-#define SET_TX_DESC_NEXT_DESC_ADDRESS64(__pdesc, __val) \
-	SET_BITS_OFFSET_LE(__pdesc+44, 0, 32, __val)
-
-#define GET_TX_DESC_NEXT_DESC_ADDRESS(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+40, 0, 32)
-#define GET_TX_DESC_NEXT_DESC_ADDRESS64(__pdesc)	\
-	SHIFT_AND_MASK_LE(__pdesc+44, 0, 32)
 
 #define GET_RX_DESC_PKT_LEN(__pdesc)			\
 	SHIFT_AND_MASK_LE(__pdesc, 0, 14)
@@ -406,22 +142,12 @@
 	SHIFT_AND_MASK_LE(__pdesc, 15, 1)
 #define GET_RX_DESC_DRV_INFO_SIZE(__pdesc)		\
 	SHIFT_AND_MASK_LE(__pdesc, 16, 4)
-#define GET_RX_DESC_SECURITY(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc, 20, 3)
-#define GET_RX_DESC_QOS(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc, 23, 1)
 #define GET_RX_DESC_SHIFT(__pdesc)			\
 	SHIFT_AND_MASK_LE(__pdesc, 24, 2)
 #define GET_RX_DESC_PHYST(__pdesc)			\
 	SHIFT_AND_MASK_LE(__pdesc, 26, 1)
 #define GET_RX_DESC_SWDEC(__pdesc)			\
 	SHIFT_AND_MASK_LE(__pdesc, 27, 1)
-#define GET_RX_DESC_LS(__pdesc)				\
-	SHIFT_AND_MASK_LE(__pdesc, 28, 1)
-#define GET_RX_DESC_FS(__pdesc)				\
-	SHIFT_AND_MASK_LE(__pdesc, 29, 1)
-#define GET_RX_DESC_EOR(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc, 30, 1)
 #define GET_RX_DESC_OWN(__pdesc)			\
 	SHIFT_AND_MASK_LE(__pdesc, 31, 1)
 
@@ -432,44 +158,10 @@
 #define SET_RX_DESC_OWN(__pdesc, __val)			\
 	SET_BITS_OFFSET_LE(__pdesc, 31, 1, __val)
 
-#define GET_RX_DESC_MACID(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+4, 0, 5)
-#define GET_RX_DESC_TID(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+4, 5, 4)
-#define GET_RX_DESC_HWRSVD(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+4, 9, 5)
 #define GET_RX_DESC_PAGGR(__pdesc)			\
 	SHIFT_AND_MASK_LE(__pdesc+4, 14, 1)
 #define GET_RX_DESC_FAGGR(__pdesc)			\
 	SHIFT_AND_MASK_LE(__pdesc+4, 15, 1)
-#define GET_RX_DESC_A1_FIT(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+4, 16, 4)
-#define GET_RX_DESC_A2_FIT(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+4, 20, 4)
-#define GET_RX_DESC_PAM(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+4, 24, 1)
-#define GET_RX_DESC_PWR(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+4, 25, 1)
-#define GET_RX_DESC_MD(__pdesc)				\
-	SHIFT_AND_MASK_LE(__pdesc+4, 26, 1)
-#define GET_RX_DESC_MF(__pdesc)				\
-	SHIFT_AND_MASK_LE(__pdesc+4, 27, 1)
-#define GET_RX_DESC_TYPE(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+4, 28, 2)
-#define GET_RX_DESC_MC(__pdesc)				\
-	SHIFT_AND_MASK_LE(__pdesc+4, 30, 1)
-#define GET_RX_DESC_BC(__pdesc)				\
-	SHIFT_AND_MASK_LE(__pdesc+4, 31, 1)
-#define GET_RX_DESC_SEQ(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+8, 0, 12)
-#define GET_RX_DESC_FRAG(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+8, 12, 4)
-#define GET_RX_DESC_NEXT_PKT_LEN(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+8, 16, 14)
-#define GET_RX_DESC_NEXT_IND(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+8, 30, 1)
-#define GET_RX_DESC_RSVD(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+8, 31, 1)
 
 #define GET_RX_DESC_RXMCS(__pdesc)			\
 	SHIFT_AND_MASK_LE(__pdesc+12, 0, 6)
@@ -479,29 +171,14 @@
 	SHIFT_AND_MASK_LE(__pdesc+12, 8, 1)
 #define GET_RX_DESC_BW(__pdesc)				\
 	SHIFT_AND_MASK_LE(__pdesc+12, 9, 1)
-#define GET_RX_DESC_HTC(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+12, 10, 1)
-#define GET_RX_DESC_HWPC_ERR(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+12, 14, 1)
-#define GET_RX_DESC_HWPC_IND(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+12, 15, 1)
-#define GET_RX_DESC_IV0(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+12, 16, 16)
-
-#define GET_RX_DESC_IV1(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+16, 0, 32)
+
 #define GET_RX_DESC_TSFL(__pdesc)			\
 	SHIFT_AND_MASK_LE(__pdesc+20, 0, 32)
 
 #define GET_RX_DESC_BUFF_ADDR(__pdesc)			\
 	SHIFT_AND_MASK_LE(__pdesc+24, 0, 32)
-#define GET_RX_DESC_BUFF_ADDR64(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+28, 0, 32)
-
 #define SET_RX_DESC_BUFF_ADDR(__pdesc, __val)		\
 	SET_BITS_OFFSET_LE(__pdesc+24, 0, 32, __val)
-#define SET_RX_DESC_BUFF_ADDR64(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+28, 0, 32, __val)
 
 #define CLEAR_PCI_TX_DESC_CONTENT(__pdesc, _size)	\
 	memset((void *)__pdesc, 0,			\
-- 
cgit v1.2.3-59-g8ed1b


From 08f6ef39bc3d40cc946dd3c3bbe50c088c498d9d Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Mon, 25 Nov 2019 12:32:27 -0600
Subject: rtlwifi: rtl8192de: Replace local bit manipulation macros

This driver uses a set of local macros to manipulate the RX and TX
descriptors, which are all little-endian quantities. These macros
are replaced by the bitfield macros le32p_replace_bits() and
le32_get_bits(). In several places, the macros operated on an entire
32-bit word. In these cases, a direct read or replacement is used.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 .../net/wireless/realtek/rtlwifi/rtl8192de/trx.h   | 160 +++++++++------------
 1 file changed, 69 insertions(+), 91 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.h
index 40746125f1ae..5ff1000b2400 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.h
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.h
@@ -14,171 +14,149 @@
 #define USB_HWDESC_HEADER_LEN			32
 #define CRCLENGTH				4
 
-/* Define a macro that takes a le32 word, converts it to host ordering,
- * right shifts by a specified count, creates a mask of the specified
- * bit count, and extracts that number of bits.
- */
-
-#define SHIFT_AND_MASK_LE(__pdesc, __shift, __mask)		\
-	((le32_to_cpu(*(((__le32 *)(__pdesc)))) >> (__shift)) &	\
-	BIT_LEN_MASK_32(__mask))
-
-/* Define a macro that clears a bit field in an le32 word and
- * sets the specified value into that bit field. The resulting
- * value remains in le32 ordering; however, it is properly converted
- * to host ordering for the clear and set operations before conversion
- * back to le32.
- */
-
-#define SET_BITS_OFFSET_LE(__pdesc, __shift, __len, __val)	\
-	(*(__le32 *)(__pdesc) =					\
-	(cpu_to_le32((le32_to_cpu(*((__le32 *)(__pdesc))) &	\
-	(~(BIT_OFFSET_LEN_MASK_32((__shift), __len)))) |		\
-	(((u32)(__val) & BIT_LEN_MASK_32(__len)) << (__shift)))));
-
 /* macros to read/write various fields in RX or TX descriptors */
 
 #define SET_TX_DESC_PKT_SIZE(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc, 0, 16, __val)
+	le32p_replace_bits((__le32 *)__pdesc, __val, GENMASK(15, 0))
 #define SET_TX_DESC_OFFSET(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc, 16, 8, __val)
+	le32p_replace_bits((__le32 *)__pdesc, __val, GENMASK(23, 16))
 #define SET_TX_DESC_HTC(__pdesc, __val)			\
-	SET_BITS_OFFSET_LE(__pdesc, 25, 1, __val)
+	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(25))
 #define SET_TX_DESC_LAST_SEG(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc, 26, 1, __val)
+	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(26))
 #define SET_TX_DESC_FIRST_SEG(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc, 27, 1, __val)
+	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(27))
 #define SET_TX_DESC_LINIP(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc, 28, 1, __val)
+	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(28))
 #define SET_TX_DESC_OWN(__pdesc, __val)			\
-	SET_BITS_OFFSET_LE(__pdesc, 31, 1, __val)
+	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(31))
 
 #define GET_TX_DESC_OWN(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc, 31, 1)
+	le32_get_bits(*(__le32 *)__pdesc, BIT(31))
 
 #define SET_TX_DESC_MACID(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+4, 0, 5, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, GENMASK(4, 0))
 #define SET_TX_DESC_AGG_ENABLE(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+4, 5, 1, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, BIT(5))
 #define SET_TX_DESC_RDG_ENABLE(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+4, 7, 1, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, BIT(7))
 #define SET_TX_DESC_QUEUE_SEL(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+4, 8, 5, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, GENMASK(12, 8))
 #define SET_TX_DESC_RATE_ID(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+4, 16, 4, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, GENMASK(19, 16))
 #define SET_TX_DESC_SEC_TYPE(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+4, 22, 2, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, GENMASK(23, 22))
 #define SET_TX_DESC_PKT_OFFSET(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+4, 26, 8, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, GENMASK(30, 26))
 
 #define SET_TX_DESC_MORE_FRAG(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+8, 17, 1, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 8), __val, BIT(17))
 #define SET_TX_DESC_AMPDU_DENSITY(__pdesc, __val)	\
-	SET_BITS_OFFSET_LE(__pdesc+8, 20, 3, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 8), __val, GENMASK(22, 20))
 
 #define SET_TX_DESC_SEQ(__pdesc, __val)			\
-	SET_BITS_OFFSET_LE(__pdesc+12, 16, 12, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 12), __val, GENMASK(27, 16))
 #define SET_TX_DESC_PKT_ID(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+12, 28, 4, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 12), __val, GENMASK(31, 28))
 
 #define SET_TX_DESC_RTS_RATE(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+16, 0, 5, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, GENMASK(4, 0))
 #define SET_TX_DESC_QOS(__pdesc, __val)			\
-	SET_BITS_OFFSET_LE(__pdesc+16, 6, 1, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(6))
 #define SET_TX_DESC_HWSEQ_EN(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+16, 7, 1, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(7))
 #define SET_TX_DESC_USE_RATE(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+16, 8, 1, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(8))
 #define SET_TX_DESC_DISABLE_FB(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+16, 10, 1, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(10))
 #define SET_TX_DESC_CTS2SELF(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+16, 11, 1, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(11))
 #define SET_TX_DESC_RTS_ENABLE(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+16, 12, 1, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(12))
 #define SET_TX_DESC_HW_RTS_ENABLE(__pdesc, __val)	\
-	SET_BITS_OFFSET_LE(__pdesc+16, 13, 1, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(13))
 #define SET_TX_DESC_TX_SUB_CARRIER(__pdesc, __val)	\
-	SET_BITS_OFFSET_LE(__pdesc+16, 20, 2, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, GENMASK(21, 20))
 #define SET_TX_DESC_DATA_BW(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+16, 25, 1, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(25))
 #define SET_TX_DESC_RTS_SHORT(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+16, 26, 1, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(26))
 #define SET_TX_DESC_RTS_BW(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+16, 27, 1, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(27))
 #define SET_TX_DESC_RTS_SC(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+16, 28, 2, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, GENMASK(29, 28))
 #define SET_TX_DESC_RTS_STBC(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+16, 30, 2, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, GENMASK(31, 30))
 
 #define SET_TX_DESC_TX_RATE(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+20, 0, 6, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 20), __val, GENMASK(5, 0))
 #define SET_TX_DESC_DATA_SHORTGI(__pdesc, __val)	\
-	SET_BITS_OFFSET_LE(__pdesc+20, 6, 1, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 20), __val, BIT(6))
 #define SET_TX_DESC_DATA_RATE_FB_LIMIT(__pdesc, __val)	\
-	SET_BITS_OFFSET_LE(__pdesc+20, 8, 5, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 20), __val, GENMASK(12, 8))
 #define SET_TX_DESC_RTS_RATE_FB_LIMIT(__pdesc, __val)	\
-	SET_BITS_OFFSET_LE(__pdesc+20, 13, 4, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 20), __val, GENMASK(16, 13))
 
 #define SET_TX_DESC_MAX_AGG_NUM(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+24, 11, 5, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 24), __val, GENMASK(15, 11))
 
 #define SET_TX_DESC_TX_BUFFER_SIZE(__pdesc, __val)	\
-	SET_BITS_OFFSET_LE(__pdesc+28, 0, 16, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 28), __val, GENMASK(15, 0))
 
 #define SET_TX_DESC_TX_BUFFER_ADDRESS(__pdesc, __val)	\
-	SET_BITS_OFFSET_LE(__pdesc+32, 0, 32, __val)
+	*(__le32 *)(__pdesc + 32) = cpu_to_le32(__val)
 
 #define GET_TX_DESC_TX_BUFFER_ADDRESS(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc+32, 0, 32)
+	le32_to_cpu(*(__le32 *)(__pdesc + 32))
 
 #define SET_TX_DESC_NEXT_DESC_ADDRESS(__pdesc, __val)	\
-	SET_BITS_OFFSET_LE(__pdesc+40, 0, 32, __val)
+	*(__le32 *)(__pdesc + 40) = cpu_to_le32(__val)
 
 #define GET_RX_DESC_PKT_LEN(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc, 0, 14)
+	le32_get_bits(*(__le32 *)__pdesc, GENMASK(13, 0))
 #define GET_RX_DESC_CRC32(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc, 14, 1)
+	le32_get_bits(*(__le32 *)__pdesc, BIT(14))
 #define GET_RX_DESC_ICV(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc, 15, 1)
+	le32_get_bits(*(__le32 *)__pdesc, BIT(15))
 #define GET_RX_DESC_DRV_INFO_SIZE(__pdesc)		\
-	SHIFT_AND_MASK_LE(__pdesc, 16, 4)
+	le32_get_bits(*(__le32 *)__pdesc, GENMASK(19, 16))
 #define GET_RX_DESC_SHIFT(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc, 24, 2)
+	le32_get_bits(*(__le32 *)__pdesc, GENMASK(25, 24))
 #define GET_RX_DESC_PHYST(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc, 26, 1)
+	le32_get_bits(*(__le32 *)__pdesc, BIT(26))
 #define GET_RX_DESC_SWDEC(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc, 27, 1)
+	le32_get_bits(*(__le32 *)__pdesc, BIT(27))
 #define GET_RX_DESC_OWN(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc, 31, 1)
+	le32_get_bits(*(__le32 *)__pdesc, BIT(31))
 
 #define SET_RX_DESC_PKT_LEN(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc, 0, 14, __val)
+	le32p_replace_bits((__le32 *)__pdesc, __val, GENMASK(13, 0))
 #define SET_RX_DESC_EOR(__pdesc, __val)			\
-	SET_BITS_OFFSET_LE(__pdesc, 30, 1, __val)
+	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(30))
 #define SET_RX_DESC_OWN(__pdesc, __val)			\
-	SET_BITS_OFFSET_LE(__pdesc, 31, 1, __val)
+	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(31))
 
 #define GET_RX_DESC_PAGGR(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+4, 14, 1)
+	le32_get_bits(*(__le32 *)(__pdesc + 4), BIT(14))
 #define GET_RX_DESC_FAGGR(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+4, 15, 1)
+	le32_get_bits(*(__le32 *)(__pdesc + 4), BIT(15))
 
 #define GET_RX_DESC_RXMCS(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+12, 0, 6)
+	le32_get_bits(*(__le32 *)(__pdesc + 12), GENMASK(5, 0))
 #define GET_RX_DESC_RXHT(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+12, 6, 1)
+	le32_get_bits(*(__le32 *)(__pdesc + 12), BIT(6))
 #define GET_RX_DESC_SPLCP(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+12, 8, 1)
+	le32_get_bits(*(__le32 *)(__pdesc + 12), BIT(8))
 #define GET_RX_DESC_BW(__pdesc)				\
-	SHIFT_AND_MASK_LE(__pdesc+12, 9, 1)
+	le32_get_bits(*(__le32 *)(__pdesc + 12), BIT(9))
 
 #define GET_RX_DESC_TSFL(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+20, 0, 32)
+	le32_to_cpu(*(__le32 *)(__pdesc + 20))
 
 #define GET_RX_DESC_BUFF_ADDR(__pdesc)			\
-	SHIFT_AND_MASK_LE(__pdesc+24, 0, 32)
+	le32_to_cpu(*(__le32 *)(__pdesc + 24))
 #define SET_RX_DESC_BUFF_ADDR(__pdesc, __val)		\
-	SET_BITS_OFFSET_LE(__pdesc+24, 0, 32, __val)
+	*(__le32 *)(__pdesc + 24) = cpu_to_le32(__val)
 
 #define CLEAR_PCI_TX_DESC_CONTENT(__pdesc, _size)	\
 	memset((void *)__pdesc, 0,			\
@@ -186,19 +164,19 @@
 
 /* For 92D early mode */
 #define SET_EARLYMODE_PKTNUM(__paddr, __value)		\
-	SET_BITS_OFFSET_LE(__paddr, 0, 3, __value)
+	le32p_replace_bits((__le32 *)__paddr, __value, GENMASK(2, 0))
 #define SET_EARLYMODE_LEN0(__paddr, __value)		\
-	SET_BITS_OFFSET_LE(__paddr, 4, 12, __value)
+	le32p_replace_bits((__le32 *)__paddr, __value, GENMASK(15, 4))
 #define SET_EARLYMODE_LEN1(__paddr, __value)		\
-	SET_BITS_OFFSET_LE(__paddr, 16, 12, __value)
+	le32p_replace_bits((__le32 *)__paddr, __value, GENMASK(27, 16))
 #define SET_EARLYMODE_LEN2_1(__paddr, __value)		\
-	SET_BITS_OFFSET_LE(__paddr, 28, 4, __value)
+	le32p_replace_bits((__le32 *)__paddr, __value, GENMASK(31, 28))
 #define SET_EARLYMODE_LEN2_2(__paddr, __value)		\
-	SET_BITS_OFFSET_LE(__paddr+4, 0, 8, __value)
+	le32p_replace_bits((__le32 *)(__paddr + 4), __value, GENMASK(7, 0))
 #define SET_EARLYMODE_LEN3(__paddr, __value)		\
-	SET_BITS_OFFSET_LE(__paddr+4, 8, 12, __value)
+	le32p_replace_bits((__le32 *)(__paddr + 4), __value, GENMASK(19, 8))
 #define SET_EARLYMODE_LEN4(__paddr, __value)		\
-	SET_BITS_OFFSET_LE(__paddr+4, 20, 12, __value)
+	le32p_replace_bits((__le32 *)(__paddr + 4), __value, GENMASK(31, 20))
 
 struct rx_fwinfo_92d {
 	u8 gain_trsw[4];
-- 
cgit v1.2.3-59-g8ed1b


From fa2b36a344f50ef6792315173567d9ce16832295 Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Mon, 25 Nov 2019 12:32:28 -0600
Subject: rtlwifi: rtl8192de: Convert macros that set descriptor

As a first step in the conversion, the macros that set the RX and TX
descriptors are converted to static inline routines, and the names are
changed from upper to lower case. To minimize the changes in a given
step, the input descriptor information is left as as a byte array
(u8 *), even though it should be a little-endian word array (__le32 *).
That will be changed in the next patch.

Several places where checkpatch.pl complains about a space after a cast
and other warnings are fixed.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 .../net/wireless/realtek/rtlwifi/rtl8192de/trx.c   | 229 +++++-----
 .../net/wireless/realtek/rtlwifi/rtl8192de/trx.h   | 498 ++++++++++++++-------
 2 files changed, 458 insertions(+), 269 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c
index 92c9fb45f800..18361f22337d 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c
@@ -438,49 +438,49 @@ static void _rtl92de_translate_rx_signal_stuff(struct ieee80211_hw *hw,
 
 bool rtl92de_rx_query_desc(struct ieee80211_hw *hw,	struct rtl_stats *stats,
 		struct ieee80211_rx_status *rx_status,
-		u8 *p_desc, struct sk_buff *skb)
+		u8 *pdesc, struct sk_buff *skb)
 {
 	struct rx_fwinfo_92d *p_drvinfo;
-	struct rx_desc_92d *pdesc = (struct rx_desc_92d *)p_desc;
-	u32 phystatus = GET_RX_DESC_PHYST(pdesc);
+	u32 phystatus = get_rx_desc_physt(pdesc);
 
-	stats->length = (u16) GET_RX_DESC_PKT_LEN(pdesc);
-	stats->rx_drvinfo_size = (u8) GET_RX_DESC_DRV_INFO_SIZE(pdesc) *
+	stats->length = (u16)get_rx_desc_pkt_len(pdesc);
+	stats->rx_drvinfo_size = (u8)get_rx_desc_drv_info_size(pdesc) *
 				 RX_DRV_INFO_SIZE_UNIT;
-	stats->rx_bufshift = (u8) (GET_RX_DESC_SHIFT(pdesc) & 0x03);
-	stats->icv = (u16) GET_RX_DESC_ICV(pdesc);
-	stats->crc = (u16) GET_RX_DESC_CRC32(pdesc);
+	stats->rx_bufshift = (u8)(get_rx_desc_shift(pdesc) & 0x03);
+	stats->icv = (u16)get_rx_desc_icv(pdesc);
+	stats->crc = (u16)get_rx_desc_crc32(pdesc);
 	stats->hwerror = (stats->crc | stats->icv);
-	stats->decrypted = !GET_RX_DESC_SWDEC(pdesc);
-	stats->rate = (u8) GET_RX_DESC_RXMCS(pdesc);
-	stats->shortpreamble = (u16) GET_RX_DESC_SPLCP(pdesc);
-	stats->isampdu = (bool) (GET_RX_DESC_PAGGR(pdesc) == 1);
-	stats->isfirst_ampdu = (bool) ((GET_RX_DESC_PAGGR(pdesc) == 1)
-					 && (GET_RX_DESC_FAGGR(pdesc) == 1));
-	stats->timestamp_low = GET_RX_DESC_TSFL(pdesc);
-	stats->rx_is40mhzpacket = (bool)GET_RX_DESC_BW(pdesc);
-	stats->is_ht = (bool)GET_RX_DESC_RXHT(pdesc);
+	stats->decrypted = !get_rx_desc_swdec(pdesc);
+	stats->rate = (u8)get_rx_desc_rxmcs(pdesc);
+	stats->shortpreamble = (u16)get_rx_desc_splcp(pdesc);
+	stats->isampdu = (bool)(get_rx_desc_paggr(pdesc) == 1);
+	stats->isfirst_ampdu = (bool)((get_rx_desc_paggr(pdesc) == 1) &&
+				      (get_rx_desc_faggr(pdesc) == 1));
+	stats->timestamp_low = get_rx_desc_tsfl(pdesc);
+	stats->rx_is40mhzpacket = (bool)get_rx_desc_bw(pdesc);
+	stats->is_ht = (bool)get_rx_desc_rxht(pdesc);
 	rx_status->freq = hw->conf.chandef.chan->center_freq;
 	rx_status->band = hw->conf.chandef.chan->band;
-	if (GET_RX_DESC_CRC32(pdesc))
+	if (get_rx_desc_crc32(pdesc))
 		rx_status->flag |= RX_FLAG_FAILED_FCS_CRC;
-	if (!GET_RX_DESC_SWDEC(pdesc))
+	if (!get_rx_desc_swdec(pdesc))
 		rx_status->flag |= RX_FLAG_DECRYPTED;
-	if (GET_RX_DESC_BW(pdesc))
+	if (get_rx_desc_bw(pdesc))
 		rx_status->bw = RATE_INFO_BW_40;
-	if (GET_RX_DESC_RXHT(pdesc))
+	if (get_rx_desc_rxht(pdesc))
 		rx_status->encoding = RX_ENC_HT;
 	rx_status->flag |= RX_FLAG_MACTIME_START;
 	if (stats->decrypted)
 		rx_status->flag |= RX_FLAG_DECRYPTED;
 	rx_status->rate_idx = rtlwifi_rate_mapping(hw, stats->is_ht,
 						   false, stats->rate);
-	rx_status->mactime = GET_RX_DESC_TSFL(pdesc);
+	rx_status->mactime = get_rx_desc_tsfl(pdesc);
 	if (phystatus) {
 		p_drvinfo = (struct rx_fwinfo_92d *)(skb->data +
 						     stats->rx_bufshift);
 		_rtl92de_translate_rx_signal_stuff(hw,
-						   skb, stats, pdesc,
+						   skb, stats,
+						   (struct rx_desc_92d *)pdesc,
 						   p_drvinfo);
 	}
 	/*rx_status->qual = stats->signal; */
@@ -493,13 +493,13 @@ static void _rtl92de_insert_emcontent(struct rtl_tcb_desc *ptcb_desc,
 {
 	memset(virtualaddress, 0, 8);
 
-	SET_EARLYMODE_PKTNUM(virtualaddress, ptcb_desc->empkt_num);
-	SET_EARLYMODE_LEN0(virtualaddress, ptcb_desc->empkt_len[0]);
-	SET_EARLYMODE_LEN1(virtualaddress, ptcb_desc->empkt_len[1]);
-	SET_EARLYMODE_LEN2_1(virtualaddress, ptcb_desc->empkt_len[2] & 0xF);
-	SET_EARLYMODE_LEN2_2(virtualaddress, ptcb_desc->empkt_len[2] >> 4);
-	SET_EARLYMODE_LEN3(virtualaddress, ptcb_desc->empkt_len[3]);
-	SET_EARLYMODE_LEN4(virtualaddress, ptcb_desc->empkt_len[4]);
+	set_earlymode_pktnum(virtualaddress, ptcb_desc->empkt_num);
+	set_earlymode_len0(virtualaddress, ptcb_desc->empkt_len[0]);
+	set_earlymode_len1(virtualaddress, ptcb_desc->empkt_len[1]);
+	set_earlymode_len2_1(virtualaddress, ptcb_desc->empkt_len[2] & 0xF);
+	set_earlymode_len2_2(virtualaddress, ptcb_desc->empkt_len[2] >> 4);
+	set_earlymode_len3(virtualaddress, ptcb_desc->empkt_len[3]);
+	set_earlymode_len4(virtualaddress, ptcb_desc->empkt_len[4]);
 }
 
 void rtl92de_tx_fill_desc(struct ieee80211_hw *hw,
@@ -556,8 +556,8 @@ void rtl92de_tx_fill_desc(struct ieee80211_hw *hw,
 	}
 	if (firstseg) {
 		if (rtlhal->earlymode_enable) {
-			SET_TX_DESC_PKT_OFFSET(pdesc, 1);
-			SET_TX_DESC_OFFSET(pdesc, USB_HWDESC_HEADER_LEN +
+			set_tx_desc_pkt_offset(pdesc, 1);
+			set_tx_desc_offset(pdesc, USB_HWDESC_HEADER_LEN +
 					   EM_HDR_LEN);
 			if (ptcb_desc->empkt_num) {
 				RT_TRACE(rtlpriv, COMP_SEND, DBG_LOUD,
@@ -567,60 +567,61 @@ void rtl92de_tx_fill_desc(struct ieee80211_hw *hw,
 							  (u8 *)(skb->data));
 			}
 		} else {
-			SET_TX_DESC_OFFSET(pdesc, USB_HWDESC_HEADER_LEN);
+			set_tx_desc_offset(pdesc, USB_HWDESC_HEADER_LEN);
 		}
 		/* 5G have no CCK rate */
 		if (rtlhal->current_bandtype == BAND_ON_5G)
 			if (ptcb_desc->hw_rate < DESC_RATE6M)
 				ptcb_desc->hw_rate = DESC_RATE6M;
-		SET_TX_DESC_TX_RATE(pdesc, ptcb_desc->hw_rate);
+		set_tx_desc_tx_rate(pdesc, ptcb_desc->hw_rate);
 		if (ptcb_desc->use_shortgi || ptcb_desc->use_shortpreamble)
-			SET_TX_DESC_DATA_SHORTGI(pdesc, 1);
+			set_tx_desc_data_shortgi(pdesc, 1);
 
 		if (rtlhal->macphymode == DUALMAC_DUALPHY &&
 			ptcb_desc->hw_rate == DESC_RATEMCS7)
-			SET_TX_DESC_DATA_SHORTGI(pdesc, 1);
+			set_tx_desc_data_shortgi(pdesc, 1);
 
 		if (info->flags & IEEE80211_TX_CTL_AMPDU) {
-			SET_TX_DESC_AGG_ENABLE(pdesc, 1);
-			SET_TX_DESC_MAX_AGG_NUM(pdesc, 0x14);
+			set_tx_desc_agg_enable(pdesc, 1);
+			set_tx_desc_max_agg_num(pdesc, 0x14);
 		}
-		SET_TX_DESC_SEQ(pdesc, seq_number);
-		SET_TX_DESC_RTS_ENABLE(pdesc, ((ptcb_desc->rts_enable &&
-				       !ptcb_desc->cts_enable) ? 1 : 0));
-		SET_TX_DESC_HW_RTS_ENABLE(pdesc, ((ptcb_desc->rts_enable
+		set_tx_desc_seq(pdesc, seq_number);
+		set_tx_desc_rts_enable(pdesc,
+				       ((ptcb_desc->rts_enable &&
+					!ptcb_desc->cts_enable) ? 1 : 0));
+		set_tx_desc_hw_rts_enable(pdesc, ((ptcb_desc->rts_enable
 					  || ptcb_desc->cts_enable) ? 1 : 0));
-		SET_TX_DESC_CTS2SELF(pdesc, ((ptcb_desc->cts_enable) ? 1 : 0));
-		SET_TX_DESC_RTS_STBC(pdesc, ((ptcb_desc->rts_stbc) ? 1 : 0));
+		set_tx_desc_cts2self(pdesc, ((ptcb_desc->cts_enable) ? 1 : 0));
+		set_tx_desc_rts_stbc(pdesc, ((ptcb_desc->rts_stbc) ? 1 : 0));
 		/* 5G have no CCK rate */
 		if (rtlhal->current_bandtype == BAND_ON_5G)
 			if (ptcb_desc->rts_rate < DESC_RATE6M)
 				ptcb_desc->rts_rate = DESC_RATE6M;
-		SET_TX_DESC_RTS_RATE(pdesc, ptcb_desc->rts_rate);
-		SET_TX_DESC_RTS_BW(pdesc, 0);
-		SET_TX_DESC_RTS_SC(pdesc, ptcb_desc->rts_sc);
-		SET_TX_DESC_RTS_SHORT(pdesc, ((ptcb_desc->rts_rate <=
+		set_tx_desc_rts_rate(pdesc, ptcb_desc->rts_rate);
+		set_tx_desc_rts_bw(pdesc, 0);
+		set_tx_desc_rts_sc(pdesc, ptcb_desc->rts_sc);
+		set_tx_desc_rts_short(pdesc, ((ptcb_desc->rts_rate <=
 			DESC_RATE54M) ?
 			(ptcb_desc->rts_use_shortpreamble ? 1 : 0) :
 			(ptcb_desc->rts_use_shortgi ? 1 : 0)));
 		if (bw_40) {
 			if (ptcb_desc->packet_bw) {
-				SET_TX_DESC_DATA_BW(pdesc, 1);
-				SET_TX_DESC_TX_SUB_CARRIER(pdesc, 3);
+				set_tx_desc_data_bw(pdesc, 1);
+				set_tx_desc_tx_sub_carrier(pdesc, 3);
 			} else {
-				SET_TX_DESC_DATA_BW(pdesc, 0);
-				SET_TX_DESC_TX_SUB_CARRIER(pdesc,
+				set_tx_desc_data_bw(pdesc, 0);
+				set_tx_desc_tx_sub_carrier(pdesc,
 							mac->cur_40_prime_sc);
 			}
 		} else {
-			SET_TX_DESC_DATA_BW(pdesc, 0);
-			SET_TX_DESC_TX_SUB_CARRIER(pdesc, 0);
+			set_tx_desc_data_bw(pdesc, 0);
+			set_tx_desc_tx_sub_carrier(pdesc, 0);
 		}
-		SET_TX_DESC_LINIP(pdesc, 0);
-		SET_TX_DESC_PKT_SIZE(pdesc, (u16) skb_len);
+		set_tx_desc_linip(pdesc, 0);
+		set_tx_desc_pkt_size(pdesc, (u16)skb_len);
 		if (sta) {
 			u8 ampdu_density = sta->ht_cap.ampdu_density;
-			SET_TX_DESC_AMPDU_DENSITY(pdesc, ampdu_density);
+			set_tx_desc_ampdu_density(pdesc, ampdu_density);
 		}
 		if (info->control.hw_key) {
 			struct ieee80211_key_conf *keyconf;
@@ -630,61 +631,61 @@ void rtl92de_tx_fill_desc(struct ieee80211_hw *hw,
 			case WLAN_CIPHER_SUITE_WEP40:
 			case WLAN_CIPHER_SUITE_WEP104:
 			case WLAN_CIPHER_SUITE_TKIP:
-				SET_TX_DESC_SEC_TYPE(pdesc, 0x1);
+				set_tx_desc_sec_type(pdesc, 0x1);
 				break;
 			case WLAN_CIPHER_SUITE_CCMP:
-				SET_TX_DESC_SEC_TYPE(pdesc, 0x3);
+				set_tx_desc_sec_type(pdesc, 0x3);
 				break;
 			default:
-				SET_TX_DESC_SEC_TYPE(pdesc, 0x0);
+				set_tx_desc_sec_type(pdesc, 0x0);
 				break;
 
 			}
 		}
-		SET_TX_DESC_PKT_ID(pdesc, 0);
-		SET_TX_DESC_QUEUE_SEL(pdesc, fw_qsel);
-		SET_TX_DESC_DATA_RATE_FB_LIMIT(pdesc, 0x1F);
-		SET_TX_DESC_RTS_RATE_FB_LIMIT(pdesc, 0xF);
-		SET_TX_DESC_DISABLE_FB(pdesc, ptcb_desc->disable_ratefallback ?
+		set_tx_desc_pkt_id(pdesc, 0);
+		set_tx_desc_queue_sel(pdesc, fw_qsel);
+		set_tx_desc_data_rate_fb_limit(pdesc, 0x1F);
+		set_tx_desc_rts_rate_fb_limit(pdesc, 0xF);
+		set_tx_desc_disable_fb(pdesc, ptcb_desc->disable_ratefallback ?
 				       1 : 0);
-		SET_TX_DESC_USE_RATE(pdesc, ptcb_desc->use_driver_rate ? 1 : 0);
+		set_tx_desc_use_rate(pdesc, ptcb_desc->use_driver_rate ? 1 : 0);
 
 		/* Set TxRate and RTSRate in TxDesc  */
 		/* This prevent Tx initial rate of new-coming packets */
 		/* from being overwritten by retried  packet rate.*/
 		if (!ptcb_desc->use_driver_rate) {
-			SET_TX_DESC_RTS_RATE(pdesc, 0x08);
-			/* SET_TX_DESC_TX_RATE(pdesc, 0x0b); */
+			set_tx_desc_rts_rate(pdesc, 0x08);
+			/* set_tx_desc_tx_rate(pdesc, 0x0b); */
 		}
 		if (ieee80211_is_data_qos(fc)) {
 			if (mac->rdg_en) {
 				RT_TRACE(rtlpriv, COMP_SEND, DBG_TRACE,
 					 "Enable RDG function\n");
-				SET_TX_DESC_RDG_ENABLE(pdesc, 1);
-				SET_TX_DESC_HTC(pdesc, 1);
+				set_tx_desc_rdg_enable(pdesc, 1);
+				set_tx_desc_htc(pdesc, 1);
 			}
 		}
 	}
 
-	SET_TX_DESC_FIRST_SEG(pdesc, (firstseg ? 1 : 0));
-	SET_TX_DESC_LAST_SEG(pdesc, (lastseg ? 1 : 0));
-	SET_TX_DESC_TX_BUFFER_SIZE(pdesc, (u16) buf_len);
-	SET_TX_DESC_TX_BUFFER_ADDRESS(pdesc, mapping);
+	set_tx_desc_first_seg(pdesc, (firstseg ? 1 : 0));
+	set_tx_desc_last_seg(pdesc, (lastseg ? 1 : 0));
+	set_tx_desc_tx_buffer_size(pdesc, (u16)buf_len);
+	set_tx_desc_tx_buffer_address(pdesc, mapping);
 	if (rtlpriv->dm.useramask) {
-		SET_TX_DESC_RATE_ID(pdesc, ptcb_desc->ratr_index);
-		SET_TX_DESC_MACID(pdesc, ptcb_desc->mac_id);
+		set_tx_desc_rate_id(pdesc, ptcb_desc->ratr_index);
+		set_tx_desc_macid(pdesc, ptcb_desc->mac_id);
 	} else {
-		SET_TX_DESC_RATE_ID(pdesc, 0xC + ptcb_desc->ratr_index);
-		SET_TX_DESC_MACID(pdesc, ptcb_desc->ratr_index);
+		set_tx_desc_rate_id(pdesc, 0xC + ptcb_desc->ratr_index);
+		set_tx_desc_macid(pdesc, ptcb_desc->ratr_index);
 	}
 	if (ieee80211_is_data_qos(fc))
-		SET_TX_DESC_QOS(pdesc, 1);
+		set_tx_desc_qos(pdesc, 1);
 
 	if ((!ieee80211_is_data_qos(fc)) && ppsc->fwctrl_lps) {
-		SET_TX_DESC_HWSEQ_EN(pdesc, 1);
-		SET_TX_DESC_PKT_ID(pdesc, 8);
+		set_tx_desc_hwseq_en(pdesc, 1);
+		set_tx_desc_pkt_id(pdesc, 8);
 	}
-	SET_TX_DESC_MORE_FRAG(pdesc, (lastseg ? 0 : 1));
+	set_tx_desc_more_frag(pdesc, (lastseg ? 0 : 1));
 	RT_TRACE(rtlpriv, COMP_SEND, DBG_TRACE, "\n");
 }
 
@@ -709,40 +710,40 @@ void rtl92de_tx_fill_cmddesc(struct ieee80211_hw *hw,
 	}
 	CLEAR_PCI_TX_DESC_CONTENT(pdesc, TX_DESC_SIZE);
 	if (firstseg)
-		SET_TX_DESC_OFFSET(pdesc, USB_HWDESC_HEADER_LEN);
+		set_tx_desc_offset(pdesc, USB_HWDESC_HEADER_LEN);
 	/* 5G have no CCK rate
 	 * Caution: The macros below are multi-line expansions.
 	 * The braces are needed no matter what checkpatch says
 	 */
 	if (rtlhal->current_bandtype == BAND_ON_5G) {
-		SET_TX_DESC_TX_RATE(pdesc, DESC_RATE6M);
+		set_tx_desc_tx_rate(pdesc, DESC_RATE6M);
 	} else {
-		SET_TX_DESC_TX_RATE(pdesc, DESC_RATE1M);
+		set_tx_desc_tx_rate(pdesc, DESC_RATE1M);
 	}
-	SET_TX_DESC_SEQ(pdesc, 0);
-	SET_TX_DESC_LINIP(pdesc, 0);
-	SET_TX_DESC_QUEUE_SEL(pdesc, fw_queue);
-	SET_TX_DESC_FIRST_SEG(pdesc, 1);
-	SET_TX_DESC_LAST_SEG(pdesc, 1);
-	SET_TX_DESC_TX_BUFFER_SIZE(pdesc, (u16)skb->len);
-	SET_TX_DESC_TX_BUFFER_ADDRESS(pdesc, mapping);
-	SET_TX_DESC_RATE_ID(pdesc, 7);
-	SET_TX_DESC_MACID(pdesc, 0);
-	SET_TX_DESC_PKT_SIZE(pdesc, (u16) (skb->len));
-	SET_TX_DESC_FIRST_SEG(pdesc, 1);
-	SET_TX_DESC_LAST_SEG(pdesc, 1);
-	SET_TX_DESC_OFFSET(pdesc, 0x20);
-	SET_TX_DESC_USE_RATE(pdesc, 1);
+	set_tx_desc_seq(pdesc, 0);
+	set_tx_desc_linip(pdesc, 0);
+	set_tx_desc_queue_sel(pdesc, fw_queue);
+	set_tx_desc_first_seg(pdesc, 1);
+	set_tx_desc_last_seg(pdesc, 1);
+	set_tx_desc_tx_buffer_size(pdesc, (u16)skb->len);
+	set_tx_desc_tx_buffer_address(pdesc, mapping);
+	set_tx_desc_rate_id(pdesc, 7);
+	set_tx_desc_macid(pdesc, 0);
+	set_tx_desc_pkt_size(pdesc, (u16)(skb->len));
+	set_tx_desc_first_seg(pdesc, 1);
+	set_tx_desc_last_seg(pdesc, 1);
+	set_tx_desc_offset(pdesc, 0x20);
+	set_tx_desc_use_rate(pdesc, 1);
 
 	if (!ieee80211_is_data_qos(fc) && ppsc->fwctrl_lps) {
-		SET_TX_DESC_HWSEQ_EN(pdesc, 1);
-		SET_TX_DESC_PKT_ID(pdesc, 8);
+		set_tx_desc_hwseq_en(pdesc, 1);
+		set_tx_desc_pkt_id(pdesc, 8);
 	}
 
 	RT_PRINT_DATA(rtlpriv, COMP_CMD, DBG_LOUD,
 		      "H2C Tx Cmd Content", pdesc, TX_DESC_SIZE);
 	wmb();
-	SET_TX_DESC_OWN(pdesc, 1);
+	set_tx_desc_own(pdesc, 1);
 }
 
 void rtl92de_set_desc(struct ieee80211_hw *hw, u8 *pdesc, bool istx,
@@ -752,10 +753,10 @@ void rtl92de_set_desc(struct ieee80211_hw *hw, u8 *pdesc, bool istx,
 		switch (desc_name) {
 		case HW_DESC_OWN:
 			wmb();
-			SET_TX_DESC_OWN(pdesc, 1);
+			set_tx_desc_own(pdesc, 1);
 			break;
 		case HW_DESC_TX_NEXTDESC_ADDR:
-			SET_TX_DESC_NEXT_DESC_ADDRESS(pdesc, *(u32 *) val);
+			set_tx_desc_next_desc_address(pdesc, *(u32 *)val);
 			break;
 		default:
 			WARN_ONCE(true, "rtl8192de: ERR txdesc :%d not processed\n",
@@ -766,16 +767,16 @@ void rtl92de_set_desc(struct ieee80211_hw *hw, u8 *pdesc, bool istx,
 		switch (desc_name) {
 		case HW_DESC_RXOWN:
 			wmb();
-			SET_RX_DESC_OWN(pdesc, 1);
+			set_rx_desc_own(pdesc, 1);
 			break;
 		case HW_DESC_RXBUFF_ADDR:
-			SET_RX_DESC_BUFF_ADDR(pdesc, *(u32 *) val);
+			set_rx_desc_buff_addr(pdesc, *(u32 *)val);
 			break;
 		case HW_DESC_RXPKT_LEN:
-			SET_RX_DESC_PKT_LEN(pdesc, *(u32 *) val);
+			set_rx_desc_pkt_len(pdesc, *(u32 *)val);
 			break;
 		case HW_DESC_RXERO:
-			SET_RX_DESC_EOR(pdesc, 1);
+			set_rx_desc_eor(pdesc, 1);
 			break;
 		default:
 			WARN_ONCE(true, "rtl8192de: ERR rxdesc :%d not processed\n",
@@ -793,10 +794,10 @@ u64 rtl92de_get_desc(struct ieee80211_hw *hw,
 	if (istx) {
 		switch (desc_name) {
 		case HW_DESC_OWN:
-			ret = GET_TX_DESC_OWN(p_desc);
+			ret = get_tx_desc_own(p_desc);
 			break;
 		case HW_DESC_TXBUFF_ADDR:
-			ret = GET_TX_DESC_TX_BUFFER_ADDRESS(p_desc);
+			ret = get_tx_desc_tx_buffer_address(p_desc);
 			break;
 		default:
 			WARN_ONCE(true, "rtl8192de: ERR txdesc :%d not processed\n",
@@ -806,13 +807,13 @@ u64 rtl92de_get_desc(struct ieee80211_hw *hw,
 	} else {
 		switch (desc_name) {
 		case HW_DESC_OWN:
-			ret = GET_RX_DESC_OWN(p_desc);
+			ret = get_rx_desc_own(p_desc);
 			break;
 		case HW_DESC_RXPKT_LEN:
-			ret = GET_RX_DESC_PKT_LEN(p_desc);
-			break;
+			ret = get_rx_desc_pkt_len(p_desc);
+		break;
 		case HW_DESC_RXBUFF_ADDR:
-			ret = GET_RX_DESC_BUFF_ADDR(p_desc);
+			ret = get_rx_desc_buff_addr(p_desc);
 			break;
 		default:
 			WARN_ONCE(true, "rtl8192de: ERR rxdesc :%d not processed\n",
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.h
index 5ff1000b2400..fe81196e46b7 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.h
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.h
@@ -16,167 +16,355 @@
 
 /* macros to read/write various fields in RX or TX descriptors */
 
-#define SET_TX_DESC_PKT_SIZE(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)__pdesc, __val, GENMASK(15, 0))
-#define SET_TX_DESC_OFFSET(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)__pdesc, __val, GENMASK(23, 16))
-#define SET_TX_DESC_HTC(__pdesc, __val)			\
-	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(25))
-#define SET_TX_DESC_LAST_SEG(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(26))
-#define SET_TX_DESC_FIRST_SEG(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(27))
-#define SET_TX_DESC_LINIP(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(28))
-#define SET_TX_DESC_OWN(__pdesc, __val)			\
-	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(31))
-
-#define GET_TX_DESC_OWN(__pdesc)			\
-	le32_get_bits(*(__le32 *)__pdesc, BIT(31))
-
-#define SET_TX_DESC_MACID(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, GENMASK(4, 0))
-#define SET_TX_DESC_AGG_ENABLE(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, BIT(5))
-#define SET_TX_DESC_RDG_ENABLE(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, BIT(7))
-#define SET_TX_DESC_QUEUE_SEL(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, GENMASK(12, 8))
-#define SET_TX_DESC_RATE_ID(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, GENMASK(19, 16))
-#define SET_TX_DESC_SEC_TYPE(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, GENMASK(23, 22))
-#define SET_TX_DESC_PKT_OFFSET(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, GENMASK(30, 26))
-
-#define SET_TX_DESC_MORE_FRAG(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)(__pdesc + 8), __val, BIT(17))
-#define SET_TX_DESC_AMPDU_DENSITY(__pdesc, __val)	\
-	le32p_replace_bits((__le32 *)(__pdesc + 8), __val, GENMASK(22, 20))
-
-#define SET_TX_DESC_SEQ(__pdesc, __val)			\
-	le32p_replace_bits((__le32 *)(__pdesc + 12), __val, GENMASK(27, 16))
-#define SET_TX_DESC_PKT_ID(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)(__pdesc + 12), __val, GENMASK(31, 28))
-
-#define SET_TX_DESC_RTS_RATE(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, GENMASK(4, 0))
-#define SET_TX_DESC_QOS(__pdesc, __val)			\
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(6))
-#define SET_TX_DESC_HWSEQ_EN(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(7))
-#define SET_TX_DESC_USE_RATE(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(8))
-#define SET_TX_DESC_DISABLE_FB(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(10))
-#define SET_TX_DESC_CTS2SELF(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(11))
-#define SET_TX_DESC_RTS_ENABLE(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(12))
-#define SET_TX_DESC_HW_RTS_ENABLE(__pdesc, __val)	\
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(13))
-#define SET_TX_DESC_TX_SUB_CARRIER(__pdesc, __val)	\
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, GENMASK(21, 20))
-#define SET_TX_DESC_DATA_BW(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(25))
-#define SET_TX_DESC_RTS_SHORT(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(26))
-#define SET_TX_DESC_RTS_BW(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(27))
-#define SET_TX_DESC_RTS_SC(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, GENMASK(29, 28))
-#define SET_TX_DESC_RTS_STBC(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, GENMASK(31, 30))
-
-#define SET_TX_DESC_TX_RATE(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)(__pdesc + 20), __val, GENMASK(5, 0))
-#define SET_TX_DESC_DATA_SHORTGI(__pdesc, __val)	\
-	le32p_replace_bits((__le32 *)(__pdesc + 20), __val, BIT(6))
-#define SET_TX_DESC_DATA_RATE_FB_LIMIT(__pdesc, __val)	\
-	le32p_replace_bits((__le32 *)(__pdesc + 20), __val, GENMASK(12, 8))
-#define SET_TX_DESC_RTS_RATE_FB_LIMIT(__pdesc, __val)	\
-	le32p_replace_bits((__le32 *)(__pdesc + 20), __val, GENMASK(16, 13))
-
-#define SET_TX_DESC_MAX_AGG_NUM(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)(__pdesc + 24), __val, GENMASK(15, 11))
-
-#define SET_TX_DESC_TX_BUFFER_SIZE(__pdesc, __val)	\
-	le32p_replace_bits((__le32 *)(__pdesc + 28), __val, GENMASK(15, 0))
-
-#define SET_TX_DESC_TX_BUFFER_ADDRESS(__pdesc, __val)	\
-	*(__le32 *)(__pdesc + 32) = cpu_to_le32(__val)
-
-#define GET_TX_DESC_TX_BUFFER_ADDRESS(__pdesc)		\
-	le32_to_cpu(*(__le32 *)(__pdesc + 32))
-
-#define SET_TX_DESC_NEXT_DESC_ADDRESS(__pdesc, __val)	\
-	*(__le32 *)(__pdesc + 40) = cpu_to_le32(__val)
-
-#define GET_RX_DESC_PKT_LEN(__pdesc)			\
-	le32_get_bits(*(__le32 *)__pdesc, GENMASK(13, 0))
-#define GET_RX_DESC_CRC32(__pdesc)			\
-	le32_get_bits(*(__le32 *)__pdesc, BIT(14))
-#define GET_RX_DESC_ICV(__pdesc)			\
-	le32_get_bits(*(__le32 *)__pdesc, BIT(15))
-#define GET_RX_DESC_DRV_INFO_SIZE(__pdesc)		\
-	le32_get_bits(*(__le32 *)__pdesc, GENMASK(19, 16))
-#define GET_RX_DESC_SHIFT(__pdesc)			\
-	le32_get_bits(*(__le32 *)__pdesc, GENMASK(25, 24))
-#define GET_RX_DESC_PHYST(__pdesc)			\
-	le32_get_bits(*(__le32 *)__pdesc, BIT(26))
-#define GET_RX_DESC_SWDEC(__pdesc)			\
-	le32_get_bits(*(__le32 *)__pdesc, BIT(27))
-#define GET_RX_DESC_OWN(__pdesc)			\
-	le32_get_bits(*(__le32 *)__pdesc, BIT(31))
-
-#define SET_RX_DESC_PKT_LEN(__pdesc, __val)		\
-	le32p_replace_bits((__le32 *)__pdesc, __val, GENMASK(13, 0))
-#define SET_RX_DESC_EOR(__pdesc, __val)			\
-	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(30))
-#define SET_RX_DESC_OWN(__pdesc, __val)			\
-	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(31))
-
-#define GET_RX_DESC_PAGGR(__pdesc)			\
-	le32_get_bits(*(__le32 *)(__pdesc + 4), BIT(14))
-#define GET_RX_DESC_FAGGR(__pdesc)			\
-	le32_get_bits(*(__le32 *)(__pdesc + 4), BIT(15))
-
-#define GET_RX_DESC_RXMCS(__pdesc)			\
-	le32_get_bits(*(__le32 *)(__pdesc + 12), GENMASK(5, 0))
-#define GET_RX_DESC_RXHT(__pdesc)			\
-	le32_get_bits(*(__le32 *)(__pdesc + 12), BIT(6))
-#define GET_RX_DESC_SPLCP(__pdesc)			\
-	le32_get_bits(*(__le32 *)(__pdesc + 12), BIT(8))
-#define GET_RX_DESC_BW(__pdesc)				\
-	le32_get_bits(*(__le32 *)(__pdesc + 12), BIT(9))
-
-#define GET_RX_DESC_TSFL(__pdesc)			\
-	le32_to_cpu(*(__le32 *)(__pdesc + 20))
-
-#define GET_RX_DESC_BUFF_ADDR(__pdesc)			\
-	le32_to_cpu(*(__le32 *)(__pdesc + 24))
-#define SET_RX_DESC_BUFF_ADDR(__pdesc, __val)		\
-	*(__le32 *)(__pdesc + 24) = cpu_to_le32(__val)
+static inline void set_tx_desc_pkt_size(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)__pdesc, __val, GENMASK(15, 0));
+}
+
+static inline void set_tx_desc_offset(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)__pdesc, __val, GENMASK(23, 16));
+}
+
+static inline void set_tx_desc_htc(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(25));
+}
+
+static inline void set_tx_desc_last_seg(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(26));
+}
+
+static inline void set_tx_desc_first_seg(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(27));
+}
+
+static inline void set_tx_desc_linip(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(28));
+}
+
+static inline void set_tx_desc_own(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(31));
+}
+
+static inline u32 get_tx_desc_own(u8 *__pdesc)
+{
+	return le32_get_bits(*(__le32 *)__pdesc, BIT(31));
+}
+
+static inline void set_tx_desc_macid(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, GENMASK(4, 0));
+}
+
+static inline void set_tx_desc_agg_enable(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, BIT(5));
+}
+
+static inline void set_tx_desc_rdg_enable(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, BIT(7));
+}
+
+static inline void set_tx_desc_queue_sel(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, GENMASK(12, 8));
+}
+
+static inline void set_tx_desc_rate_id(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, GENMASK(19, 16));
+}
+
+static inline void set_tx_desc_sec_type(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, GENMASK(23, 22));
+}
+
+static inline void set_tx_desc_pkt_offset(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, GENMASK(30, 26));
+}
+
+static inline void set_tx_desc_more_frag(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 8), __val, BIT(17));
+}
+
+static inline void set_tx_desc_ampdu_density(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 8), __val, GENMASK(22, 20));
+}
+
+static inline void set_tx_desc_seq(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 12), __val, GENMASK(27, 16));
+}
+
+static inline void set_tx_desc_pkt_id(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 12), __val, GENMASK(31, 28));
+}
+
+static inline void set_tx_desc_rts_rate(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, GENMASK(4, 0));
+}
+
+static inline void set_tx_desc_qos(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(6));
+}
+
+static inline void set_tx_desc_hwseq_en(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(7));
+}
+
+static inline void set_tx_desc_use_rate(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(8));
+}
+
+static inline void set_tx_desc_disable_fb(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(10));
+}
+
+static inline void set_tx_desc_cts2self(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(11));
+}
+
+static inline void set_tx_desc_rts_enable(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(12));
+}
+
+static inline void set_tx_desc_hw_rts_enable(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(13));
+}
+
+static inline void set_tx_desc_tx_sub_carrier(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, GENMASK(21, 20));
+}
+
+static inline void set_tx_desc_data_bw(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(25));
+}
+
+static inline void set_tx_desc_rts_short(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(26));
+}
+
+static inline void set_tx_desc_rts_bw(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(27));
+}
+
+static inline void set_tx_desc_rts_sc(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, GENMASK(29, 28));
+}
+
+static inline void set_tx_desc_rts_stbc(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, GENMASK(31, 30));
+}
+
+static inline void set_tx_desc_tx_rate(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 20), __val, GENMASK(5, 0));
+}
+
+static inline void set_tx_desc_data_shortgi(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 20), __val, BIT(6));
+}
+
+static inline void set_tx_desc_data_rate_fb_limit(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 20), __val, GENMASK(12, 8));
+}
+
+static inline void set_tx_desc_rts_rate_fb_limit(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 20), __val, GENMASK(16, 13));
+}
+
+static inline void set_tx_desc_max_agg_num(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 24), __val, GENMASK(15, 11));
+}
+
+static inline void set_tx_desc_tx_buffer_size(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)(__pdesc + 28), __val, GENMASK(15, 0));
+}
+
+static inline void set_tx_desc_tx_buffer_address(u8 *__pdesc, u32 __val)
+{
+	*(__le32 *)(__pdesc + 32) = cpu_to_le32(__val);
+}
+
+static inline u32 get_tx_desc_tx_buffer_address(u8 *__pdesc)
+{
+	return le32_to_cpu(*(__le32 *)(__pdesc + 32));
+}
+
+static inline void set_tx_desc_next_desc_address(u8 *__pdesc, u32 __val)
+{
+	*(__le32 *)(__pdesc + 40) = cpu_to_le32(__val);
+}
+
+static inline u32 get_rx_desc_pkt_len(u8 *__pdesc)
+{
+	return le32_get_bits(*(__le32 *)__pdesc, GENMASK(13, 0));
+}
+
+static inline u32 get_rx_desc_crc32(u8 *__pdesc)
+{
+	return le32_get_bits(*(__le32 *)__pdesc, BIT(14));
+}
+
+static inline u32 get_rx_desc_icv(u8 *__pdesc)
+{
+	return le32_get_bits(*(__le32 *)__pdesc, BIT(15));
+}
+
+static inline u32 get_rx_desc_drv_info_size(u8 *__pdesc)
+{
+	return le32_get_bits(*(__le32 *)__pdesc, GENMASK(19, 16));
+}
+
+static inline u32 get_rx_desc_shift(u8 *__pdesc)
+{
+	return le32_get_bits(*(__le32 *)__pdesc, GENMASK(25, 24));
+}
+
+static inline u32 get_rx_desc_physt(u8 *__pdesc)
+{
+	return le32_get_bits(*(__le32 *)__pdesc, BIT(26));
+}
+
+static inline u32 get_rx_desc_swdec(u8 *__pdesc)
+{
+	return le32_get_bits(*(__le32 *)__pdesc, BIT(27));
+}
+
+static inline u32 get_rx_desc_own(u8 *__pdesc)
+{
+	return le32_get_bits(*(__le32 *)__pdesc, BIT(31));
+}
+
+static inline void set_rx_desc_pkt_len(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)__pdesc, __val, GENMASK(13, 0));
+}
+
+static inline void set_rx_desc_eor(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(30));
+}
+
+static inline void set_rx_desc_own(u8 *__pdesc, u32 __val)
+{
+	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(31));
+}
+
+static inline u32 get_rx_desc_paggr(u8 *__pdesc)
+{
+	return le32_get_bits(*(__le32 *)(__pdesc + 4), BIT(14));
+}
+
+static inline u32 get_rx_desc_faggr(u8 *__pdesc)
+{
+	return le32_get_bits(*(__le32 *)(__pdesc + 4), BIT(15));
+}
+
+static inline u32 get_rx_desc_rxmcs(u8 *__pdesc)
+{
+	return le32_get_bits(*(__le32 *)(__pdesc + 12), GENMASK(5, 0));
+}
+
+static inline u32 get_rx_desc_rxht(u8 *__pdesc)
+{
+	return le32_get_bits(*(__le32 *)(__pdesc + 12), BIT(6));
+}
+
+static inline u32 get_rx_desc_splcp(u8 *__pdesc)
+{
+	return le32_get_bits(*(__le32 *)(__pdesc + 12), BIT(8));
+}
+
+static inline u32 get_rx_desc_bw(u8 *__pdesc)
+{
+	return le32_get_bits(*(__le32 *)(__pdesc + 12), BIT(9));
+}
+
+static inline u32 get_rx_desc_tsfl(u8 *__pdesc)
+{
+	return le32_to_cpu(*(__le32 *)(__pdesc + 20));
+}
+
+static inline u32 get_rx_desc_buff_addr(u8 *__pdesc)
+{
+	return le32_to_cpu(*(__le32 *)(__pdesc + 24));
+}
+
+static inline void set_rx_desc_buff_addr(u8 *__pdesc, u32 __val)
+{
+	*(__le32 *)(__pdesc + 24) = cpu_to_le32(__val);
+}
 
 #define CLEAR_PCI_TX_DESC_CONTENT(__pdesc, _size)	\
 	memset((void *)__pdesc, 0,			\
 	       min_t(size_t, _size, TX_DESC_NEXT_DESC_OFFSET))
 
 /* For 92D early mode */
-#define SET_EARLYMODE_PKTNUM(__paddr, __value)		\
-	le32p_replace_bits((__le32 *)__paddr, __value, GENMASK(2, 0))
-#define SET_EARLYMODE_LEN0(__paddr, __value)		\
-	le32p_replace_bits((__le32 *)__paddr, __value, GENMASK(15, 4))
-#define SET_EARLYMODE_LEN1(__paddr, __value)		\
-	le32p_replace_bits((__le32 *)__paddr, __value, GENMASK(27, 16))
-#define SET_EARLYMODE_LEN2_1(__paddr, __value)		\
-	le32p_replace_bits((__le32 *)__paddr, __value, GENMASK(31, 28))
-#define SET_EARLYMODE_LEN2_2(__paddr, __value)		\
-	le32p_replace_bits((__le32 *)(__paddr + 4), __value, GENMASK(7, 0))
-#define SET_EARLYMODE_LEN3(__paddr, __value)		\
-	le32p_replace_bits((__le32 *)(__paddr + 4), __value, GENMASK(19, 8))
-#define SET_EARLYMODE_LEN4(__paddr, __value)		\
-	le32p_replace_bits((__le32 *)(__paddr + 4), __value, GENMASK(31, 20))
+static inline void set_earlymode_pktnum(u8 *__paddr, u32 __value)
+{
+	le32p_replace_bits((__le32 *)__paddr, __value, GENMASK(2, 0));
+}
+
+static inline void set_earlymode_len0(u8 *__paddr, u32 __value)
+{
+	le32p_replace_bits((__le32 *)__paddr, __value, GENMASK(15, 4));
+}
+
+static inline void set_earlymode_len1(u8 *__paddr, u32 __value)
+{
+	le32p_replace_bits((__le32 *)__paddr, __value, GENMASK(27, 16));
+}
+
+static inline void set_earlymode_len2_1(u8 *__paddr, u32 __value)
+{
+	le32p_replace_bits((__le32 *)__paddr, __value, GENMASK(31, 28));
+}
+
+static inline void set_earlymode_len2_2(u8 *__paddr, u32 __value)
+{
+	le32p_replace_bits((__le32 *)(__paddr + 4), __value, GENMASK(7, 0));
+}
+
+static inline void set_earlymode_len3(u8 *__paddr, u32 __value)
+{
+	le32p_replace_bits((__le32 *)(__paddr + 4), __value, GENMASK(19, 8));
+}
+
+static inline void set_earlymode_len4(u8 *__paddr, u32 __value)
+{
+	le32p_replace_bits((__le32 *)(__paddr + 4), __value, GENMASK(31, 20));
+}
 
 struct rx_fwinfo_92d {
 	u8 gain_trsw[4];
-- 
cgit v1.2.3-59-g8ed1b


From 96c21a254053539bca12c5a750d9762ea419363d Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Mon, 25 Nov 2019 12:32:29 -0600
Subject: rtlwifi: rtl8192de: Convert inline routines to little-endian words

In this step, the read/write routines for the descriptors are converted
to use __le32 quantities, thus a lot of casts can be removed. Callback
routines still use the 8-bit arrays, but these are changed within the
specified routine.

Reported-by: kbuild test robot <lkp@intel.com>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 .../net/wireless/realtek/rtlwifi/rtl8192de/trx.c   |  25 +-
 .../net/wireless/realtek/rtlwifi/rtl8192de/trx.h   | 284 +++++++++++----------
 2 files changed, 159 insertions(+), 150 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c
index 18361f22337d..bb64612654b9 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c
@@ -438,8 +438,9 @@ static void _rtl92de_translate_rx_signal_stuff(struct ieee80211_hw *hw,
 
 bool rtl92de_rx_query_desc(struct ieee80211_hw *hw,	struct rtl_stats *stats,
 		struct ieee80211_rx_status *rx_status,
-		u8 *pdesc, struct sk_buff *skb)
+		u8 *pdesc8, struct sk_buff *skb)
 {
+	__le32 *pdesc = (__le32 *)pdesc8;
 	struct rx_fwinfo_92d *p_drvinfo;
 	u32 phystatus = get_rx_desc_physt(pdesc);
 
@@ -489,8 +490,10 @@ bool rtl92de_rx_query_desc(struct ieee80211_hw *hw,	struct rtl_stats *stats,
 }
 
 static void _rtl92de_insert_emcontent(struct rtl_tcb_desc *ptcb_desc,
-				      u8 *virtualaddress)
+				      u8 *virtualaddress8)
 {
+	__le32 *virtualaddress = (__le32 *)virtualaddress8;
+
 	memset(virtualaddress, 0, 8);
 
 	set_earlymode_pktnum(virtualaddress, ptcb_desc->empkt_num);
@@ -503,7 +506,7 @@ static void _rtl92de_insert_emcontent(struct rtl_tcb_desc *ptcb_desc,
 }
 
 void rtl92de_tx_fill_desc(struct ieee80211_hw *hw,
-			  struct ieee80211_hdr *hdr, u8 *pdesc_tx,
+			  struct ieee80211_hdr *hdr, u8 *pdesc8,
 			  u8 *pbd_desc_tx, struct ieee80211_tx_info *info,
 			  struct ieee80211_sta *sta,
 			  struct sk_buff *skb,
@@ -514,7 +517,7 @@ void rtl92de_tx_fill_desc(struct ieee80211_hw *hw,
 	struct rtl_pci *rtlpci = rtl_pcidev(rtl_pcipriv(hw));
 	struct rtl_hal *rtlhal = rtl_hal(rtlpriv);
 	struct rtl_ps_ctl *ppsc = rtl_psc(rtl_priv(hw));
-	u8 *pdesc = pdesc_tx;
+	__le32 *pdesc = (__le32 *)pdesc8;
 	u16 seq_number;
 	__le16 fc = hdr->frame_control;
 	unsigned int buf_len = 0;
@@ -549,7 +552,7 @@ void rtl92de_tx_fill_desc(struct ieee80211_hw *hw,
 			 "DMA mapping error\n");
 		return;
 	}
-	CLEAR_PCI_TX_DESC_CONTENT(pdesc, sizeof(struct tx_desc_92d));
+	clear_pci_tx_desc_content(pdesc, sizeof(struct tx_desc_92d));
 	if (ieee80211_is_nullfunc(fc) || ieee80211_is_ctl(fc)) {
 		firstseg = true;
 		lastseg = true;
@@ -690,7 +693,7 @@ void rtl92de_tx_fill_desc(struct ieee80211_hw *hw,
 }
 
 void rtl92de_tx_fill_cmddesc(struct ieee80211_hw *hw,
-			     u8 *pdesc, bool firstseg,
+			     u8 *pdesc8, bool firstseg,
 			     bool lastseg, struct sk_buff *skb)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
@@ -702,13 +705,14 @@ void rtl92de_tx_fill_cmddesc(struct ieee80211_hw *hw,
 		    skb->data, skb->len, PCI_DMA_TODEVICE);
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)(skb->data);
 	__le16 fc = hdr->frame_control;
+	__le32 *pdesc = (__le32 *)pdesc8;
 
 	if (pci_dma_mapping_error(rtlpci->pdev, mapping)) {
 		RT_TRACE(rtlpriv, COMP_SEND, DBG_TRACE,
 			 "DMA mapping error\n");
 		return;
 	}
-	CLEAR_PCI_TX_DESC_CONTENT(pdesc, TX_DESC_SIZE);
+	clear_pci_tx_desc_content(pdesc, TX_DESC_SIZE);
 	if (firstseg)
 		set_tx_desc_offset(pdesc, USB_HWDESC_HEADER_LEN);
 	/* 5G have no CCK rate
@@ -746,9 +750,11 @@ void rtl92de_tx_fill_cmddesc(struct ieee80211_hw *hw,
 	set_tx_desc_own(pdesc, 1);
 }
 
-void rtl92de_set_desc(struct ieee80211_hw *hw, u8 *pdesc, bool istx,
+void rtl92de_set_desc(struct ieee80211_hw *hw, u8 *pdesc8, bool istx,
 		      u8 desc_name, u8 *val)
 {
+	__le32  *pdesc = (__le32 *)pdesc8;
+
 	if (istx) {
 		switch (desc_name) {
 		case HW_DESC_OWN:
@@ -787,8 +793,9 @@ void rtl92de_set_desc(struct ieee80211_hw *hw, u8 *pdesc, bool istx,
 }
 
 u64 rtl92de_get_desc(struct ieee80211_hw *hw,
-		     u8 *p_desc, bool istx, u8 desc_name)
+		     u8 *p_desc8, bool istx, u8 desc_name)
 {
+	__le32 *p_desc = (__le32 *)p_desc8;
 	u32 ret = 0;
 
 	if (istx) {
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.h
index fe81196e46b7..d01578875cd5 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.h
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.h
@@ -16,354 +16,356 @@
 
 /* macros to read/write various fields in RX or TX descriptors */
 
-static inline void set_tx_desc_pkt_size(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_pkt_size(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)__pdesc, __val, GENMASK(15, 0));
+	le32p_replace_bits(__pdesc, __val, GENMASK(15, 0));
 }
 
-static inline void set_tx_desc_offset(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_offset(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)__pdesc, __val, GENMASK(23, 16));
+	le32p_replace_bits(__pdesc, __val, GENMASK(23, 16));
 }
 
-static inline void set_tx_desc_htc(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_htc(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(25));
+	le32p_replace_bits(__pdesc, __val, BIT(25));
 }
 
-static inline void set_tx_desc_last_seg(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_last_seg(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(26));
+	le32p_replace_bits(__pdesc, __val, BIT(26));
 }
 
-static inline void set_tx_desc_first_seg(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_first_seg(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(27));
+	le32p_replace_bits(__pdesc, __val, BIT(27));
 }
 
-static inline void set_tx_desc_linip(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_linip(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(28));
+	le32p_replace_bits(__pdesc, __val, BIT(28));
 }
 
-static inline void set_tx_desc_own(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_own(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(31));
+	le32p_replace_bits(__pdesc, __val, BIT(31));
 }
 
-static inline u32 get_tx_desc_own(u8 *__pdesc)
+static inline u32 get_tx_desc_own(__le32 *__pdesc)
 {
-	return le32_get_bits(*(__le32 *)__pdesc, BIT(31));
+	return le32_get_bits(*__pdesc, BIT(31));
 }
 
-static inline void set_tx_desc_macid(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_macid(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, GENMASK(4, 0));
+	le32p_replace_bits((__pdesc + 1), __val, GENMASK(4, 0));
 }
 
-static inline void set_tx_desc_agg_enable(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_agg_enable(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, BIT(5));
+	le32p_replace_bits((__pdesc + 1), __val, BIT(5));
 }
 
-static inline void set_tx_desc_rdg_enable(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_rdg_enable(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, BIT(7));
+	le32p_replace_bits((__pdesc + 1), __val, BIT(7));
 }
 
-static inline void set_tx_desc_queue_sel(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_queue_sel(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, GENMASK(12, 8));
+	le32p_replace_bits((__pdesc + 1), __val, GENMASK(12, 8));
 }
 
-static inline void set_tx_desc_rate_id(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_rate_id(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, GENMASK(19, 16));
+	le32p_replace_bits((__pdesc + 1), __val, GENMASK(19, 16));
 }
 
-static inline void set_tx_desc_sec_type(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_sec_type(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, GENMASK(23, 22));
+	le32p_replace_bits((__pdesc + 1), __val, GENMASK(23, 22));
 }
 
-static inline void set_tx_desc_pkt_offset(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_pkt_offset(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 4), __val, GENMASK(30, 26));
+	le32p_replace_bits((__pdesc + 1), __val, GENMASK(30, 26));
 }
 
-static inline void set_tx_desc_more_frag(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_more_frag(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 8), __val, BIT(17));
+	le32p_replace_bits((__pdesc + 2), __val, BIT(17));
 }
 
-static inline void set_tx_desc_ampdu_density(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_ampdu_density(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 8), __val, GENMASK(22, 20));
+	le32p_replace_bits((__pdesc + 2), __val, GENMASK(22, 20));
 }
 
-static inline void set_tx_desc_seq(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_seq(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 12), __val, GENMASK(27, 16));
+	le32p_replace_bits((__pdesc + 3), __val, GENMASK(27, 16));
 }
 
-static inline void set_tx_desc_pkt_id(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_pkt_id(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 12), __val, GENMASK(31, 28));
+	le32p_replace_bits((__pdesc + 3), __val, GENMASK(31, 28));
 }
 
-static inline void set_tx_desc_rts_rate(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_rts_rate(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, GENMASK(4, 0));
+	le32p_replace_bits((__pdesc + 4), __val, GENMASK(4, 0));
 }
 
-static inline void set_tx_desc_qos(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_qos(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(6));
+	le32p_replace_bits((__pdesc + 4), __val, BIT(6));
 }
 
-static inline void set_tx_desc_hwseq_en(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_hwseq_en(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(7));
+	le32p_replace_bits((__pdesc + 4), __val, BIT(7));
 }
 
-static inline void set_tx_desc_use_rate(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_use_rate(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(8));
+	le32p_replace_bits((__pdesc + 4), __val, BIT(8));
 }
 
-static inline void set_tx_desc_disable_fb(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_disable_fb(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(10));
+	le32p_replace_bits((__pdesc + 4), __val, BIT(10));
 }
 
-static inline void set_tx_desc_cts2self(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_cts2self(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(11));
+	le32p_replace_bits((__pdesc + 4), __val, BIT(11));
 }
 
-static inline void set_tx_desc_rts_enable(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_rts_enable(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(12));
+	le32p_replace_bits((__pdesc + 4), __val, BIT(12));
 }
 
-static inline void set_tx_desc_hw_rts_enable(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_hw_rts_enable(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(13));
+	le32p_replace_bits((__pdesc + 4), __val, BIT(13));
 }
 
-static inline void set_tx_desc_tx_sub_carrier(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_tx_sub_carrier(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, GENMASK(21, 20));
+	le32p_replace_bits((__pdesc + 4), __val, GENMASK(21, 20));
 }
 
-static inline void set_tx_desc_data_bw(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_data_bw(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(25));
+	le32p_replace_bits((__pdesc + 4), __val, BIT(25));
 }
 
-static inline void set_tx_desc_rts_short(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_rts_short(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(26));
+	le32p_replace_bits((__pdesc + 4), __val, BIT(26));
 }
 
-static inline void set_tx_desc_rts_bw(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_rts_bw(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, BIT(27));
+	le32p_replace_bits((__pdesc + 4), __val, BIT(27));
 }
 
-static inline void set_tx_desc_rts_sc(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_rts_sc(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, GENMASK(29, 28));
+	le32p_replace_bits((__pdesc + 4), __val, GENMASK(29, 28));
 }
 
-static inline void set_tx_desc_rts_stbc(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_rts_stbc(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 16), __val, GENMASK(31, 30));
+	le32p_replace_bits((__pdesc + 4), __val, GENMASK(31, 30));
 }
 
-static inline void set_tx_desc_tx_rate(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_tx_rate(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 20), __val, GENMASK(5, 0));
+	le32p_replace_bits((__pdesc + 5), __val, GENMASK(5, 0));
 }
 
-static inline void set_tx_desc_data_shortgi(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_data_shortgi(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 20), __val, BIT(6));
+	le32p_replace_bits((__pdesc + 5), __val, BIT(6));
 }
 
-static inline void set_tx_desc_data_rate_fb_limit(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_data_rate_fb_limit(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 20), __val, GENMASK(12, 8));
+	le32p_replace_bits((__pdesc + 5), __val, GENMASK(12, 8));
 }
 
-static inline void set_tx_desc_rts_rate_fb_limit(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_rts_rate_fb_limit(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 20), __val, GENMASK(16, 13));
+	le32p_replace_bits((__pdesc + 5), __val, GENMASK(16, 13));
 }
 
-static inline void set_tx_desc_max_agg_num(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_max_agg_num(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 24), __val, GENMASK(15, 11));
+	le32p_replace_bits((__pdesc + 6), __val, GENMASK(15, 11));
 }
 
-static inline void set_tx_desc_tx_buffer_size(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_tx_buffer_size(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)(__pdesc + 28), __val, GENMASK(15, 0));
+	le32p_replace_bits((__pdesc + 7), __val, GENMASK(15, 0));
 }
 
-static inline void set_tx_desc_tx_buffer_address(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_tx_buffer_address(__le32 *__pdesc, u32 __val)
 {
-	*(__le32 *)(__pdesc + 32) = cpu_to_le32(__val);
+	*(__pdesc + 8) = cpu_to_le32(__val);
 }
 
-static inline u32 get_tx_desc_tx_buffer_address(u8 *__pdesc)
+static inline u32 get_tx_desc_tx_buffer_address(__le32 *__pdesc)
 {
-	return le32_to_cpu(*(__le32 *)(__pdesc + 32));
+	return le32_to_cpu(*(__pdesc + 8));
 }
 
-static inline void set_tx_desc_next_desc_address(u8 *__pdesc, u32 __val)
+static inline void set_tx_desc_next_desc_address(__le32 *__pdesc, u32 __val)
 {
-	*(__le32 *)(__pdesc + 40) = cpu_to_le32(__val);
+	*(__pdesc + 10) = cpu_to_le32(__val);
 }
 
-static inline u32 get_rx_desc_pkt_len(u8 *__pdesc)
+static inline u32 get_rx_desc_pkt_len(__le32 *__pdesc)
 {
-	return le32_get_bits(*(__le32 *)__pdesc, GENMASK(13, 0));
+	return le32_get_bits(*__pdesc, GENMASK(13, 0));
 }
 
-static inline u32 get_rx_desc_crc32(u8 *__pdesc)
+static inline u32 get_rx_desc_crc32(__le32 *__pdesc)
 {
-	return le32_get_bits(*(__le32 *)__pdesc, BIT(14));
+	return le32_get_bits(*__pdesc, BIT(14));
 }
 
-static inline u32 get_rx_desc_icv(u8 *__pdesc)
+static inline u32 get_rx_desc_icv(__le32 *__pdesc)
 {
-	return le32_get_bits(*(__le32 *)__pdesc, BIT(15));
+	return le32_get_bits(*__pdesc, BIT(15));
 }
 
-static inline u32 get_rx_desc_drv_info_size(u8 *__pdesc)
+static inline u32 get_rx_desc_drv_info_size(__le32 *__pdesc)
 {
-	return le32_get_bits(*(__le32 *)__pdesc, GENMASK(19, 16));
+	return le32_get_bits(*__pdesc, GENMASK(19, 16));
 }
 
-static inline u32 get_rx_desc_shift(u8 *__pdesc)
+static inline u32 get_rx_desc_shift(__le32 *__pdesc)
 {
-	return le32_get_bits(*(__le32 *)__pdesc, GENMASK(25, 24));
+	return le32_get_bits(*__pdesc, GENMASK(25, 24));
 }
 
-static inline u32 get_rx_desc_physt(u8 *__pdesc)
+static inline u32 get_rx_desc_physt(__le32 *__pdesc)
 {
-	return le32_get_bits(*(__le32 *)__pdesc, BIT(26));
+	return le32_get_bits(*__pdesc, BIT(26));
 }
 
-static inline u32 get_rx_desc_swdec(u8 *__pdesc)
+static inline u32 get_rx_desc_swdec(__le32 *__pdesc)
 {
-	return le32_get_bits(*(__le32 *)__pdesc, BIT(27));
+	return le32_get_bits(*__pdesc, BIT(27));
 }
 
-static inline u32 get_rx_desc_own(u8 *__pdesc)
+static inline u32 get_rx_desc_own(__le32 *__pdesc)
 {
-	return le32_get_bits(*(__le32 *)__pdesc, BIT(31));
+	return le32_get_bits(*__pdesc, BIT(31));
 }
 
-static inline void set_rx_desc_pkt_len(u8 *__pdesc, u32 __val)
+static inline void set_rx_desc_pkt_len(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)__pdesc, __val, GENMASK(13, 0));
+	le32p_replace_bits(__pdesc, __val, GENMASK(13, 0));
 }
 
-static inline void set_rx_desc_eor(u8 *__pdesc, u32 __val)
+static inline void set_rx_desc_eor(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(30));
+	le32p_replace_bits(__pdesc, __val, BIT(30));
 }
 
-static inline void set_rx_desc_own(u8 *__pdesc, u32 __val)
+static inline void set_rx_desc_own(__le32 *__pdesc, u32 __val)
 {
-	le32p_replace_bits((__le32 *)__pdesc, __val, BIT(31));
+	le32p_replace_bits(__pdesc, __val, BIT(31));
 }
 
-static inline u32 get_rx_desc_paggr(u8 *__pdesc)
+static inline u32 get_rx_desc_paggr(__le32 *__pdesc)
 {
-	return le32_get_bits(*(__le32 *)(__pdesc + 4), BIT(14));
+	return le32_get_bits(*(__pdesc + 1), BIT(14));
 }
 
-static inline u32 get_rx_desc_faggr(u8 *__pdesc)
+static inline u32 get_rx_desc_faggr(__le32 *__pdesc)
 {
-	return le32_get_bits(*(__le32 *)(__pdesc + 4), BIT(15));
+	return le32_get_bits(*(__pdesc + 1), BIT(15));
 }
 
-static inline u32 get_rx_desc_rxmcs(u8 *__pdesc)
+static inline u32 get_rx_desc_rxmcs(__le32 *__pdesc)
 {
-	return le32_get_bits(*(__le32 *)(__pdesc + 12), GENMASK(5, 0));
+	return le32_get_bits(*(__pdesc + 3), GENMASK(5, 0));
 }
 
-static inline u32 get_rx_desc_rxht(u8 *__pdesc)
+static inline u32 get_rx_desc_rxht(__le32 *__pdesc)
 {
-	return le32_get_bits(*(__le32 *)(__pdesc + 12), BIT(6));
+	return le32_get_bits(*(__pdesc + 3), BIT(6));
 }
 
-static inline u32 get_rx_desc_splcp(u8 *__pdesc)
+static inline u32 get_rx_desc_splcp(__le32 *__pdesc)
 {
-	return le32_get_bits(*(__le32 *)(__pdesc + 12), BIT(8));
+	return le32_get_bits(*(__pdesc + 3), BIT(8));
 }
 
-static inline u32 get_rx_desc_bw(u8 *__pdesc)
+static inline u32 get_rx_desc_bw(__le32 *__pdesc)
 {
-	return le32_get_bits(*(__le32 *)(__pdesc + 12), BIT(9));
+	return le32_get_bits(*(__pdesc + 3), BIT(9));
 }
 
-static inline u32 get_rx_desc_tsfl(u8 *__pdesc)
+static inline u32 get_rx_desc_tsfl(__le32 *__pdesc)
 {
-	return le32_to_cpu(*(__le32 *)(__pdesc + 20));
+	return le32_to_cpu(*(__pdesc + 5));
 }
 
-static inline u32 get_rx_desc_buff_addr(u8 *__pdesc)
+static inline u32 get_rx_desc_buff_addr(__le32 *__pdesc)
 {
-	return le32_to_cpu(*(__le32 *)(__pdesc + 24));
+	return le32_to_cpu(*(__pdesc + 6));
 }
 
-static inline void set_rx_desc_buff_addr(u8 *__pdesc, u32 __val)
+static inline void set_rx_desc_buff_addr(__le32 *__pdesc, u32 __val)
 {
-	*(__le32 *)(__pdesc + 24) = cpu_to_le32(__val);
+	*(__pdesc + 6) = cpu_to_le32(__val);
 }
 
-#define CLEAR_PCI_TX_DESC_CONTENT(__pdesc, _size)	\
-	memset((void *)__pdesc, 0,			\
-	       min_t(size_t, _size, TX_DESC_NEXT_DESC_OFFSET))
+static inline void clear_pci_tx_desc_content(__le32 *__pdesc, u32 _size)
+{
+	memset((void *)__pdesc, 0,
+	       min_t(size_t, _size, TX_DESC_NEXT_DESC_OFFSET));
+}
 
 /* For 92D early mode */
-static inline void set_earlymode_pktnum(u8 *__paddr, u32 __value)
+static inline void set_earlymode_pktnum(__le32 *__paddr, u32 __value)
 {
-	le32p_replace_bits((__le32 *)__paddr, __value, GENMASK(2, 0));
+	le32p_replace_bits(__paddr, __value, GENMASK(2, 0));
 }
 
-static inline void set_earlymode_len0(u8 *__paddr, u32 __value)
+static inline void set_earlymode_len0(__le32 *__paddr, u32 __value)
 {
-	le32p_replace_bits((__le32 *)__paddr, __value, GENMASK(15, 4));
+	le32p_replace_bits(__paddr, __value, GENMASK(15, 4));
 }
 
-static inline void set_earlymode_len1(u8 *__paddr, u32 __value)
+static inline void set_earlymode_len1(__le32 *__paddr, u32 __value)
 {
-	le32p_replace_bits((__le32 *)__paddr, __value, GENMASK(27, 16));
+	le32p_replace_bits(__paddr, __value, GENMASK(27, 16));
 }
 
-static inline void set_earlymode_len2_1(u8 *__paddr, u32 __value)
+static inline void set_earlymode_len2_1(__le32 *__paddr, u32 __value)
 {
-	le32p_replace_bits((__le32 *)__paddr, __value, GENMASK(31, 28));
+	le32p_replace_bits(__paddr, __value, GENMASK(31, 28));
 }
 
-static inline void set_earlymode_len2_2(u8 *__paddr, u32 __value)
+static inline void set_earlymode_len2_2(__le32 *__paddr, u32 __value)
 {
-	le32p_replace_bits((__le32 *)(__paddr + 4), __value, GENMASK(7, 0));
+	le32p_replace_bits((__paddr + 1), __value, GENMASK(7, 0));
 }
 
-static inline void set_earlymode_len3(u8 *__paddr, u32 __value)
+static inline void set_earlymode_len3(__le32 *__paddr, u32 __value)
 {
-	le32p_replace_bits((__le32 *)(__paddr + 4), __value, GENMASK(19, 8));
+	le32p_replace_bits((__paddr + 1), __value, GENMASK(19, 8));
 }
 
-static inline void set_earlymode_len4(u8 *__paddr, u32 __value)
+static inline void set_earlymode_len4(__le32 *__paddr, u32 __value)
 {
-	le32p_replace_bits((__le32 *)(__paddr + 4), __value, GENMASK(31, 20));
+	le32p_replace_bits((__paddr + 1), __value, GENMASK(31, 20));
 }
 
 struct rx_fwinfo_92d {
-- 
cgit v1.2.3-59-g8ed1b


From 7031fb7f1cd12720921bf2c62e1c09aea7bcb890 Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Mon, 25 Nov 2019 16:09:26 -0600
Subject: rtlwifi: rtl8192de: Remove usage of private bit manipulation macros

Besides the previously changes macros in trx.h, there are some of these
local macros in the driver. These are replaced with standard bit
manipulations. Unused macros are deleted.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 .../net/wireless/realtek/rtlwifi/rtl8192de/fw.h    | 61 +++-------------------
 1 file changed, 7 insertions(+), 54 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/fw.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/fw.h
index bd8ea6d66dff..7f0a17c1a9ea 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/fw.h
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/fw.h
@@ -16,73 +16,26 @@
 		(GET_FIRMWARE_HDR_SIGNATURE(_pfwhdr) & 0xFFFF) == 0x92D2 ||  \
 		(GET_FIRMWARE_HDR_SIGNATURE(_pfwhdr) & 0xFFFF) == 0x92D3)
 
-/* Define a macro that takes an le32 word, converts it to host ordering,
- * right shifts by a specified count, creates a mask of the specified
- * bit count, and extracts that number of bits.
- */
-
-#define SHIFT_AND_MASK_LE(__pdesc, __shift, __mask)		\
-	((le32_to_cpu(*(((__le32 *)(__pdesc)))) >> (__shift)) &	\
-	BIT_LEN_MASK_32(__mask))
-
 /* Firmware Header(8-byte alinment required) */
 /* --- LONG WORD 0 ---- */
 #define GET_FIRMWARE_HDR_SIGNATURE(__fwhdr)		\
-	SHIFT_AND_MASK_LE(__fwhdr, 0, 16)
-#define GET_FIRMWARE_HDR_CATEGORY(__fwhdr)		\
-	SHIFT_AND_MASK_LE(__fwhdr, 16, 8)
-#define GET_FIRMWARE_HDR_FUNCTION(__fwhdr)		\
-	SHIFT_AND_MASK_LE(__fwhdr, 24, 8)
+	le32_get_bits(*(__le32 *)__fwhdr, GENMASK(15, 0))
 #define GET_FIRMWARE_HDR_VERSION(__fwhdr)		\
-	SHIFT_AND_MASK_LE(__fwhdr + 4, 0, 16)
+	le32_get_bits(*(__le32 *)(__fwhdr + 4), GENMASK(15, 0))
 #define GET_FIRMWARE_HDR_SUB_VER(__fwhdr)		\
-	SHIFT_AND_MASK_LE(__fwhdr + 4, 16, 8)
-#define GET_FIRMWARE_HDR_RSVD1(__fwhdr)			\
-	SHIFT_AND_MASK_LE(__fwhdr + 4, 24, 8)
-
-/* --- LONG WORD 1 ---- */
-#define GET_FIRMWARE_HDR_MONTH(__fwhdr)			\
-	SHIFT_AND_MASK_LE(__fwhdr + 8, 0, 8)
-#define GET_FIRMWARE_HDR_DATE(__fwhdr)			\
-	SHIFT_AND_MASK_LE(__fwhdr + 8, 8, 8)
-#define GET_FIRMWARE_HDR_HOUR(__fwhdr)			\
-	SHIFT_AND_MASK_LE(__fwhdr + 8, 16, 8)
-#define GET_FIRMWARE_HDR_MINUTE(__fwhdr)		\
-	SHIFT_AND_MASK_LE(__fwhdr + 8, 24, 8)
-#define GET_FIRMWARE_HDR_ROMCODE_SIZE(__fwhdr)		\
-	SHIFT_AND_MASK_LE(__fwhdr + 12, 0, 16)
-#define GET_FIRMWARE_HDR_RSVD2(__fwhdr)			\
-	SHIFT_AND_MASK_LE(__fwhdr + 12, 16, 16)
-
-/* --- LONG WORD 2 ---- */
-#define GET_FIRMWARE_HDR_SVN_IDX(__fwhdr)		\
-	SHIFT_AND_MASK_LE(__fwhdr + 16, 0, 32)
-#define GET_FIRMWARE_HDR_RSVD3(__fwhdr)			\
-	SHIFT_AND_MASK_LE(__fwhdr + 20, 0, 32)
-
-/* --- LONG WORD 3 ---- */
-#define GET_FIRMWARE_HDR_RSVD4(__fwhdr)			\
-	SHIFT_AND_MASK_LE(__fwhdr + 24, 0, 32)
-#define GET_FIRMWARE_HDR_RSVD5(__fwhdr)			\
-	SHIFT_AND_MASK_LE(__fwhdr + 28, 0, 32)
+	le32_get_bits(*(__le32 *)(__fwhdr + 4), GENMASK(23, 16))
 
 #define pagenum_128(_len) \
 	(u32)(((_len) >> 7) + ((_len) & 0x7F ? 1 : 0))
 
-#define SET_H2CCMD_PWRMODE_PARM_MODE(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE(__ph2ccmd, 0, 8, __val)
-#define SET_H2CCMD_PWRMODE_PARM_SMART_PS(__ph2ccmd, __val)	\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd) + 1, 0, 8, __val)
-#define SET_H2CCMD_PWRMODE_PARM_BCN_PASS_TIME(__ph2ccmd, __val)	\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd) + 2, 0, 8, __val)
 #define SET_H2CCMD_JOINBSSRPT_PARM_OPMODE(__ph2ccmd, __val)	\
-	SET_BITS_TO_LE_1BYTE(__ph2ccmd, 0, 8, __val)
+	*(u8 *)__ph2ccmd = __val;
 #define SET_H2CCMD_RSVDPAGE_LOC_PROBE_RSP(__ph2ccmd, __val)	\
-	SET_BITS_TO_LE_1BYTE(__ph2ccmd, 0, 8, __val)
+	*(u8 *)__ph2ccmd = __val;
 #define SET_H2CCMD_RSVDPAGE_LOC_PSPOLL(__ph2ccmd, __val)	\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd) + 1, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 1) = __val;
 #define SET_H2CCMD_RSVDPAGE_LOC_NULL_DATA(__ph2ccmd, __val)	\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd) + 2, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 2) = __val;
 
 int rtl92d_download_fw(struct ieee80211_hw *hw);
 void rtl92d_fill_h2c_cmd(struct ieee80211_hw *hw, u8 element_id,
-- 
cgit v1.2.3-59-g8ed1b


From 5ea5c93d937cb5fb5891835ab134221a2881e32a Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Mon, 25 Nov 2019 16:09:27 -0600
Subject: rtlwifi: rtl8188ee: Remove usage of private bit manipulation

Besides the previously changes macros for the RX and TX descriptors.
the local macros are used in other places. These are replaced with
standard bit manipulation macros.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 .../net/wireless/realtek/rtlwifi/rtl8188ee/fw.c    |  12 +--
 .../net/wireless/realtek/rtlwifi/rtl8188ee/fw.h    | 103 ++++++++-------------
 .../net/wireless/realtek/rtlwifi/rtl8188ee/hw.c    |  11 +--
 3 files changed, 49 insertions(+), 77 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/fw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/fw.c
index e2e0bfbc24fe..fc7b9ad7e5d0 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/fw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/fw.c
@@ -372,20 +372,20 @@ void rtl88e_set_fw_pwrmode_cmd(struct ieee80211_hw *hw, u8 mode)
 	u8 rlbm, power_state = 0;
 	RT_TRACE(rtlpriv, COMP_POWER, DBG_LOUD, "FW LPS mode = %d\n", mode);
 
-	SET_H2CCMD_PWRMODE_PARM_MODE(u1_h2c_set_pwrmode, ((mode) ? 1 : 0));
+	set_h2ccmd_pwrmode_parm_mode(u1_h2c_set_pwrmode, ((mode) ? 1 : 0));
 	rlbm = 0;/*YJ, temp, 120316. FW now not support RLBM=2.*/
-	SET_H2CCMD_PWRMODE_PARM_RLBM(u1_h2c_set_pwrmode, rlbm);
-	SET_H2CCMD_PWRMODE_PARM_SMART_PS(u1_h2c_set_pwrmode,
+	set_h2ccmd_pwrmode_parm_rlbm(u1_h2c_set_pwrmode, rlbm);
+	set_h2ccmd_pwrmode_parm_smart_ps(u1_h2c_set_pwrmode,
 		(rtlpriv->mac80211.p2p) ? ppsc->smart_ps : 1);
-	SET_H2CCMD_PWRMODE_PARM_AWAKE_INTERVAL(u1_h2c_set_pwrmode,
+	set_h2ccmd_pwrmode_parm_awake_interval(u1_h2c_set_pwrmode,
 		ppsc->reg_max_lps_awakeintvl);
-	SET_H2CCMD_PWRMODE_PARM_ALL_QUEUE_UAPSD(u1_h2c_set_pwrmode, 0);
+	set_h2ccmd_pwrmode_parm_all_queue_uapsd(u1_h2c_set_pwrmode, 0);
 	if (mode == FW_PS_ACTIVE_MODE)
 		power_state |= FW_PWR_STATE_ACTIVE;
 	else
 		power_state |= FW_PWR_STATE_RF_OFF;
 
-	SET_H2CCMD_PWRMODE_PARM_PWR_STATE(u1_h2c_set_pwrmode, power_state);
+	set_h2ccmd_pwrmode_parm_pwr_state(u1_h2c_set_pwrmode, power_state);
 
 	RT_PRINT_DATA(rtlpriv, COMP_CMD, DBG_DMESG,
 		      "rtl92c_set_fw_pwrmode(): u1_h2c_set_pwrmode\n",
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/fw.h b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/fw.h
index 39ddb7afea9d..79f095e47d71 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/fw.h
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/fw.h
@@ -169,82 +169,55 @@ enum rtl8188e_h2c_cmd {
 	SET_BITS_TO_LE_1BYTE((__cmd)+2, 0, 8, __value)
 
 
-#define SET_H2CCMD_PWRMODE_PARM_MODE(__ph2ccmd, __val)			\
-	SET_BITS_TO_LE_1BYTE(__ph2ccmd, 0, 8, __val)
-#define SET_H2CCMD_PWRMODE_PARM_RLBM(__cmd, __value)		\
-	SET_BITS_TO_LE_1BYTE((__cmd)+1, 0, 4, __value)
-#define SET_H2CCMD_PWRMODE_PARM_SMART_PS(__cmd, __value)		\
-	SET_BITS_TO_LE_1BYTE((__cmd)+1, 4, 4, __value)
-#define SET_H2CCMD_PWRMODE_PARM_AWAKE_INTERVAL(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE((__cmd)+2, 0, 8, __value)
-#define SET_H2CCMD_PWRMODE_PARM_ALL_QUEUE_UAPSD(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE((__cmd)+3, 0, 8, __value)
-#define SET_H2CCMD_PWRMODE_PARM_PWR_STATE(__cmd, __value)		\
-	SET_BITS_TO_LE_1BYTE((__cmd)+4, 0, 8, __value)
-#define GET_88E_H2CCMD_PWRMODE_PARM_MODE(__cmd)			\
-	LE_BITS_TO_1BYTE(__cmd, 0, 8)
+static inline void set_h2ccmd_pwrmode_parm_mode(u8 *__ph2ccmd, u8 __val)
+{
+	*(u8 *)(__ph2ccmd) = __val;
+}
+
+static inline void set_h2ccmd_pwrmode_parm_rlbm(u8 *__cmd, u8 __value)
+{
+	u8p_replace_bits(__cmd + 1, __value, GENMASK(3, 0));
+}
+
+static inline void set_h2ccmd_pwrmode_parm_smart_ps(u8 *__cmd, u8 __value)
+{
+	u8p_replace_bits(__cmd + 1, __value, GENMASK(7, 4));
+}
+
+static inline void set_h2ccmd_pwrmode_parm_awake_interval(u8 *__cmd, u8 __value)
+{
+	*(u8 *)(__cmd + 2) = __value;
+}
+
+static inline void set_h2ccmd_pwrmode_parm_all_queue_uapsd(u8 *__cmd,
+							   u8 __value)
+{
+	*(u8 *)(__cmd + 3) = __value;
+}
+
+static inline void set_h2ccmd_pwrmode_parm_pwr_state(u8 *__cmd, u8 __value)
+{
+	*(u8 *)(__cmd + 4) = __value;
+}
 
 #define SET_H2CCMD_JOINBSSRPT_PARM_OPMODE(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE(__ph2ccmd, 0, 8, __val)
+	*(u8 *)(__ph2ccmd) = __val;
 #define SET_H2CCMD_RSVDPAGE_LOC_PROBE_RSP(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE(__ph2ccmd, 0, 8, __val)
+	*(u8 *)(__ph2ccmd) = __val;
 #define SET_H2CCMD_RSVDPAGE_LOC_PSPOLL(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd)+1, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 1) = __val;
 #define SET_H2CCMD_RSVDPAGE_LOC_NULL_DATA(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd)+2, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 2) = __val;
 
 /* AP_OFFLOAD */
 #define SET_H2CCMD_AP_OFFLOAD_ON(__cmd, __value)			\
-	SET_BITS_TO_LE_1BYTE(__cmd, 0, 8, __value)
+	*(u8 *)__cmd = __value;
 #define SET_H2CCMD_AP_OFFLOAD_HIDDEN(__cmd, __value)		\
-	SET_BITS_TO_LE_1BYTE((__cmd)+1, 0, 8, __value)
+	*(u8 *)(__cmd + 1) = __value;
 #define SET_H2CCMD_AP_OFFLOAD_DENYANY(__cmd, __value)		\
-	SET_BITS_TO_LE_1BYTE((__cmd)+2, 0, 8, __value)
+	*(u8 *)(__cmd + 2) = __value;
 #define SET_H2CCMD_AP_OFFLOAD_WAKEUP_EVT_RPT(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE((__cmd)+3, 0, 8, __value)
-
-/* Keep Alive Control*/
-#define SET_88E_H2CCMD_KEEP_ALIVE_ENABLE(__cmd, __value)		\
-	SET_BITS_TO_LE_1BYTE(__cmd, 0, 1, __value)
-#define SET_88E_H2CCMD_KEEP_ALIVE_ACCPEPT_USER_DEFINED(__cmd, __value) \
-	SET_BITS_TO_LE_1BYTE(__cmd, 1, 1, __value)
-#define SET_88E_H2CCMD_KEEP_ALIVE_PERIOD(__cmd, __value)		\
-	SET_BITS_TO_LE_1BYTE((__cmd)+1, 0, 8, __value)
-
-/*REMOTE_WAKE_CTRL */
-#define SET_88E_H2CCMD_REMOTE_WAKE_CTRL_EN(__cmd, __value)		\
-	SET_BITS_TO_LE_1BYTE(__cmd, 0, 1, __value)
-#if (USE_OLD_WOWLAN_DEBUG_FW == 0)
-#define SET_88E_H2CCMD_REMOTE_WAKE_CTRL_ARP_OFFLOAD_EN(__cmd, __value) \
-	SET_BITS_TO_LE_1BYTE(__cmd, 1, 1, __value)
-#define SET_88E_H2CCMD_REMOTE_WAKE_CTRL_NDP_OFFLOAD_EN(__cmd, __value) \
-	SET_BITS_TO_LE_1BYTE(__cmd, 2, 1, __value)
-#define SET_88E_H2CCMD_REMOTE_WAKE_CTRL_GTK_OFFLOAD_EN(__cmd, __value) \
-	SET_BITS_TO_LE_1BYTE(__cmd, 3, 1, __value)
-#else
-#define SET_88E_H2_REM_WAKE_ENC_ALG(__cmd, __value)		\
-	SET_BITS_TO_LE_1BYTE((__cmd)+1, 0, 8, __value)
-#define SET_88E_H2CCMD_REMOTE_WAKE_CTRL_GROUP_ENC_ALG(__cmd, __value) \
-	SET_BITS_TO_LE_1BYTE((__cmd)+2, 0, 8, __value)
-#endif
-
-/* GTK_OFFLOAD */
-#define SET_88E_H2CCMD_AOAC_GLOBAL_INFO_PAIRWISE_ENC_ALG(__cmd, __value) \
-	SET_BITS_TO_LE_1BYTE(__cmd, 0, 8, __value)
-#define SET_88E_H2CCMD_AOAC_GLOBAL_INFO_GROUP_ENC_ALG(__cmd, __value) \
-	SET_BITS_TO_LE_1BYTE((__cmd)+1, 0, 8, __value)
-
-/* AOAC_RSVDPAGE_LOC */
-#define SET_88E_H2CCMD_AOAC_RSVD_LOC_REM_WAKE_CTRL_INFO(__cmd, __value) \
-	SET_BITS_TO_LE_1BYTE((__cmd), 0, 8, __value)
-#define SET_88E_H2CCMD_AOAC_RSVDPAGE_LOC_ARP_RSP(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE((__cmd)+1, 0, 8, __value)
-#define SET_88E_H2CCMD_AOAC_RSVDPAGE_LOC_NEIGHBOR_ADV(__cmd, __value) \
-	SET_BITS_TO_LE_1BYTE((__cmd)+2, 0, 8, __value)
-#define SET_88E_H2CCMD_AOAC_RSVDPAGE_LOC_GTK_RSP(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE((__cmd)+3, 0, 8, __value)
-#define SET_88E_H2CCMD_AOAC_RSVDPAGE_LOC_GTK_INFO(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE((__cmd)+4, 0, 8, __value)
+	*(u8 *)(__cmd + 3) = __value;
 
 int rtl88e_download_fw(struct ieee80211_hw *hw,
 		       bool buse_wake_on_wlan_fw);
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/hw.c
index f92e95f5494f..70716631de85 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/hw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/hw.c
@@ -2486,13 +2486,12 @@ void rtl8188ee_bt_hw_init(struct ieee80211_hw *hw)
 		if (rtlpriv->btcoexist.bt_ant_isolation)
 			rtl_write_byte(rtlpriv, REG_GPIO_MUXCFG, 0xa0);
 
-		u1_tmp = rtl_read_byte(rtlpriv, 0x4fd) &
-			 BIT_OFFSET_LEN_MASK_32(0, 1);
+		u1_tmp = rtl_read_byte(rtlpriv, 0x4fd) & BIT(0);
 		u1_tmp = u1_tmp |
 			 ((rtlpriv->btcoexist.bt_ant_isolation == 1) ?
-			 0 : BIT_OFFSET_LEN_MASK_32(1, 1)) |
+			 0 : BIT((1)) |
 			 ((rtlpriv->btcoexist.bt_service == BT_SCO) ?
-			 0 : BIT_OFFSET_LEN_MASK_32(2, 1));
+			 0 : BIT(2)));
 		rtl_write_byte(rtlpriv, 0x4fd, u1_tmp);
 
 		rtl_write_dword(rtlpriv, REG_BT_COEX_TABLE+4, 0xaaaa9aaa);
@@ -2502,11 +2501,11 @@ void rtl8188ee_bt_hw_init(struct ieee80211_hw *hw)
 		/* Config to 1T1R. */
 		if (rtlphy->rf_type == RF_1T1R) {
 			u1_tmp = rtl_read_byte(rtlpriv, ROFDM0_TRXPATHENABLE);
-			u1_tmp &= ~(BIT_OFFSET_LEN_MASK_32(1, 1));
+			u1_tmp &= ~(BIT(1));
 			rtl_write_byte(rtlpriv, ROFDM0_TRXPATHENABLE, u1_tmp);
 
 			u1_tmp = rtl_read_byte(rtlpriv, ROFDM1_TRXPATHENABLE);
-			u1_tmp &= ~(BIT_OFFSET_LEN_MASK_32(1, 1));
+			u1_tmp &= ~(BIT(1));
 			rtl_write_byte(rtlpriv, ROFDM1_TRXPATHENABLE, u1_tmp);
 		}
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 4c8692d4939d2704e9ba3894a5510d0e323b6fbd Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Mon, 25 Nov 2019 16:09:28 -0600
Subject: rtlwifi: rtl8192ce: rtl8192c_com: Remove usage of private bit
 manipulation macros

Besides the previously changes macros for the RX and TX descriptors.
the local macros are used in other places. These are replaced with
standard bit manipulation macros.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 .../net/wireless/realtek/rtlwifi/rtl8192c/dm_common.c | 19 +++++++++----------
 .../net/wireless/realtek/rtlwifi/rtl8192c/fw_common.h | 14 +++++++-------
 drivers/net/wireless/realtek/rtlwifi/rtl8192ce/hw.c   | 11 +++++------
 3 files changed, 21 insertions(+), 23 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192c/dm_common.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192c/dm_common.c
index 4bef237f488d..06fc9b5cdd8f 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192c/dm_common.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192c/dm_common.c
@@ -8,11 +8,11 @@
 #include "../base.h"
 #include "../core.h"
 
-#define BT_RSSI_STATE_NORMAL_POWER	BIT_OFFSET_LEN_MASK_32(0, 1)
-#define BT_RSSI_STATE_AMDPU_OFF		BIT_OFFSET_LEN_MASK_32(1, 1)
-#define BT_RSSI_STATE_SPECIAL_LOW	BIT_OFFSET_LEN_MASK_32(2, 1)
-#define BT_RSSI_STATE_BG_EDCA_LOW	BIT_OFFSET_LEN_MASK_32(3, 1)
-#define BT_RSSI_STATE_TXPOWER_LOW	BIT_OFFSET_LEN_MASK_32(4, 1)
+#define BT_RSSI_STATE_NORMAL_POWER	BIT(0)
+#define BT_RSSI_STATE_AMDPU_OFF		BIT(1)
+#define BT_RSSI_STATE_SPECIAL_LOW	BIT(2)
+#define BT_RSSI_STATE_BG_EDCA_LOW	BIT(3)
+#define BT_RSSI_STATE_TXPOWER_LOW	BIT(4)
 #define BT_MASK				0x00ffffff
 
 #define RTLPRIV			(struct rtl_priv *)
@@ -1515,7 +1515,7 @@ static bool rtl92c_bt_state_change(struct ieee80211_hw *hw)
 	    polling == 0xffffffff && bt_state == 0xff)
 		return false;
 
-	bt_state &= BIT_OFFSET_LEN_MASK_32(0, 1);
+	bt_state &= BIT(0);
 	if (bt_state != rtlpriv->btcoexist.bt_cur_state) {
 		rtlpriv->btcoexist.bt_cur_state = bt_state;
 
@@ -1524,8 +1524,7 @@ static bool rtl92c_bt_state_change(struct ieee80211_hw *hw)
 
 			bt_state = bt_state |
 			  ((rtlpriv->btcoexist.bt_ant_isolation == 1) ?
-			  0 : BIT_OFFSET_LEN_MASK_32(1, 1)) |
-			  BIT_OFFSET_LEN_MASK_32(2, 1);
+			  0 : BIT(1)) | BIT(2);
 			rtl_write_byte(rtlpriv, 0x4fd, bt_state);
 		}
 		return true;
@@ -1555,9 +1554,9 @@ static bool rtl92c_bt_state_change(struct ieee80211_hw *hw)
 			rtlpriv->btcoexist.bt_service = cur_service_type;
 			bt_state = bt_state |
 			   ((rtlpriv->btcoexist.bt_ant_isolation == 1) ?
-			   0 : BIT_OFFSET_LEN_MASK_32(1, 1)) |
+			   0 : BIT(1)) |
 			   ((rtlpriv->btcoexist.bt_service != BT_IDLE) ?
-			   0 : BIT_OFFSET_LEN_MASK_32(2, 1));
+			   0 : BIT(2));
 
 			/* Add interrupt migration when bt is not ini
 			 * idle state (no traffic). */
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192c/fw_common.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192c/fw_common.h
index 888d9fc94bc2..706fc753dfe6 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192c/fw_common.h
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192c/fw_common.h
@@ -46,19 +46,19 @@
 #define pagenum_128(_len)	(u32)(((_len)>>7) + ((_len)&0x7F ? 1 : 0))
 
 #define SET_H2CCMD_PWRMODE_PARM_MODE(__ph2ccmd, __val)			\
-	SET_BITS_TO_LE_1BYTE(__ph2ccmd, 0, 8, __val)
+	*(u8 *)(__ph2ccmd) = __val
 #define SET_H2CCMD_PWRMODE_PARM_SMART_PS(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd)+1, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 1) = __val
 #define SET_H2CCMD_PWRMODE_PARM_BCN_PASS_TIME(__ph2ccmd, __val)	\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd)+2, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 2) = __val
 #define SET_H2CCMD_JOINBSSRPT_PARM_OPMODE(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE(__ph2ccmd, 0, 8, __val)
+	*(u8 *)(__ph2ccmd) = __val
 #define SET_H2CCMD_RSVDPAGE_LOC_PROBE_RSP(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE(__ph2ccmd, 0, 8, __val)
+	*(u8 *)(__ph2ccmd) = __val
 #define SET_H2CCMD_RSVDPAGE_LOC_PSPOLL(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd)+1, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 1) = __val
 #define SET_H2CCMD_RSVDPAGE_LOC_NULL_DATA(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd)+2, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 2) = __val
 
 int rtl92c_download_fw(struct ieee80211_hw *hw);
 void rtl92c_fill_h2c_cmd(struct ieee80211_hw *hw, u8 element_id,
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/hw.c
index a52dd64d528d..6402a9e09be7 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/hw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/hw.c
@@ -2299,13 +2299,12 @@ void rtl8192ce_bt_hw_init(struct ieee80211_hw *hw)
 		if (rtlpriv->btcoexist.bt_ant_isolation)
 			rtl_write_byte(rtlpriv, REG_GPIO_MUXCFG, 0xa0);
 
-		u1_tmp = rtl_read_byte(rtlpriv, 0x4fd) &
-			 BIT_OFFSET_LEN_MASK_32(0, 1);
+		u1_tmp = rtl_read_byte(rtlpriv, 0x4fd) & BIT(0);
 		u1_tmp = u1_tmp |
 			 ((rtlpriv->btcoexist.bt_ant_isolation == 1) ?
-			 0 : BIT_OFFSET_LEN_MASK_32(1, 1)) |
+			 0 : BIT(1)) |
 			 ((rtlpriv->btcoexist.bt_service == BT_SCO) ?
-			 0 : BIT_OFFSET_LEN_MASK_32(2, 1));
+			 0 : BIT(2));
 		rtl_write_byte(rtlpriv, 0x4fd, u1_tmp);
 
 		rtl_write_dword(rtlpriv, REG_BT_COEX_TABLE+4, 0xaaaa9aaa);
@@ -2315,11 +2314,11 @@ void rtl8192ce_bt_hw_init(struct ieee80211_hw *hw)
 		/* Config to 1T1R. */
 		if (rtlphy->rf_type == RF_1T1R) {
 			u1_tmp = rtl_read_byte(rtlpriv, ROFDM0_TRXPATHENABLE);
-			u1_tmp &= ~(BIT_OFFSET_LEN_MASK_32(1, 1));
+			u1_tmp &= ~(BIT(1));
 			rtl_write_byte(rtlpriv, ROFDM0_TRXPATHENABLE, u1_tmp);
 
 			u1_tmp = rtl_read_byte(rtlpriv, ROFDM1_TRXPATHENABLE);
-			u1_tmp &= ~(BIT_OFFSET_LEN_MASK_32(1, 1));
+			u1_tmp &= ~(BIT(1));
 			rtl_write_byte(rtlpriv, ROFDM1_TRXPATHENABLE, u1_tmp);
 		}
 	}
-- 
cgit v1.2.3-59-g8ed1b


From f0dcd57e0b417152190e9f68e5ef981e1c316162 Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Mon, 25 Nov 2019 16:09:29 -0600
Subject: rtlwifi: Remove dependence on special bit manipulation macros for
 common driver

The special macros are used in several places in the base driver.
These are replaced by standard bit manipulation macros or code.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtlwifi/base.h |  4 ++--
 drivers/net/wireless/realtek/rtlwifi/ps.c   | 20 ++++++++++----------
 2 files changed, 12 insertions(+), 12 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/base.h b/drivers/net/wireless/realtek/rtlwifi/base.h
index e4a7e074ae3f..fa92e29fffda 100644
--- a/drivers/net/wireless/realtek/rtlwifi/base.h
+++ b/drivers/net/wireless/realtek/rtlwifi/base.h
@@ -61,9 +61,9 @@ enum ap_peer {
 	CP_MACADDR((u8 *)(_hdr)+FRAME_OFFSET_ADDRESS3, (u8 *)(_val))
 
 #define SET_TX_DESC_SPE_RPT(__pdesc, __val)			\
-	SET_BITS_TO_LE_4BYTE((__pdesc) + 8, 19, 1, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 8), __val, BIT(19))
 #define SET_TX_DESC_SW_DEFINE(__pdesc, __val)	\
-	SET_BITS_TO_LE_4BYTE((__pdesc) + 24, 0, 12, __val)
+	le32p_replace_bits((__le32 *)(__pdesc + 24), __val, GENMASK(11, 0))
 
 int rtl_init_core(struct ieee80211_hw *hw);
 void rtl_deinit_core(struct ieee80211_hw *hw);
diff --git a/drivers/net/wireless/realtek/rtlwifi/ps.c b/drivers/net/wireless/realtek/rtlwifi/ps.c
index e5e1ec5a41dc..bc0ac96ee615 100644
--- a/drivers/net/wireless/realtek/rtlwifi/ps.c
+++ b/drivers/net/wireless/realtek/rtlwifi/ps.c
@@ -737,7 +737,7 @@ static void rtl_p2p_noa_ie(struct ieee80211_hw *hw, void *data,
 	find_p2p_ie = true;
 	/*to find noa ie*/
 	while (ie + 1 < end) {
-		noa_len = READEF2BYTE((__le16 *)&ie[1]);
+		noa_len = le16_to_cpu(*((__le16 *)&ie[1]));
 		if (ie + 3 + ie[1] > end)
 			return;
 
@@ -766,16 +766,16 @@ static void rtl_p2p_noa_ie(struct ieee80211_hw *hw, void *data,
 				index = 5;
 				for (i = 0; i < noa_num; i++) {
 					p2pinfo->noa_count_type[i] =
-							READEF1BYTE(ie+index);
+					 *(u8 *)(ie + index);
 					index += 1;
 					p2pinfo->noa_duration[i] =
-						 READEF4BYTE((__le32 *)ie+index);
+					 le32_to_cpu(*(__le32 *)ie + index);
 					index += 4;
 					p2pinfo->noa_interval[i] =
-						 READEF4BYTE((__le32 *)ie+index);
+					 le32_to_cpu(*(__le32 *)ie + index);
 					index += 4;
 					p2pinfo->noa_start_time[i] =
-						 READEF4BYTE((__le32 *)ie+index);
+					 le32_to_cpu(*(__le32 *)ie + index);
 					index += 4;
 				}
 
@@ -832,7 +832,7 @@ static void rtl_p2p_action_ie(struct ieee80211_hw *hw, void *data,
 	RT_TRACE(rtlpriv, COMP_FW, DBG_LOUD, "action frame find P2P IE.\n");
 	/*to find noa ie*/
 	while (ie + 1 < end) {
-		noa_len = READEF2BYTE((__le16 *)&ie[1]);
+		noa_len = le16_to_cpu(*(__le16 *)&ie[1]);
 		if (ie + 3 + ie[1] > end)
 			return;
 
@@ -861,16 +861,16 @@ static void rtl_p2p_action_ie(struct ieee80211_hw *hw, void *data,
 				index = 5;
 				for (i = 0; i < noa_num; i++) {
 					p2pinfo->noa_count_type[i] =
-							READEF1BYTE(ie+index);
+					 *(u8 *)(ie + index);
 					index += 1;
 					p2pinfo->noa_duration[i] =
-							 READEF4BYTE((__le32 *)ie+index);
+					 le32_to_cpu(*(__le32 *)ie + index);
 					index += 4;
 					p2pinfo->noa_interval[i] =
-							 READEF4BYTE((__le32 *)ie+index);
+					 le32_to_cpu(*(__le32 *)ie + index);
 					index += 4;
 					p2pinfo->noa_start_time[i] =
-							 READEF4BYTE((__le32 *)ie+index);
+					 le32_to_cpu(*(__le32 *)ie + index);
 					index += 4;
 				}
 
-- 
cgit v1.2.3-59-g8ed1b


From 39f40710d0b5231e1f4adca16be4da0dcf6434cb Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Mon, 25 Nov 2019 16:09:30 -0600
Subject: rtlwifi: rtl88821ae: Remove usage of private bit manipulation macros

Besides the previously changed macros for the RX and TX descriptors.
the local macros are used in other places. These are replaced with
standard bit manipulation macros.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 .../net/wireless/realtek/rtlwifi/rtl8821ae/fw.h    | 102 ++++++++++-----------
 1 file changed, 47 insertions(+), 55 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/fw.h b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/fw.h
index e11e496b7277..c269942b3f4a 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/fw.h
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/fw.h
@@ -151,123 +151,115 @@ enum rtl8821a_h2c_cmd {
 #define pagenum_128(_len)	(u32)(((_len)>>7) + ((_len)&0x7F ? 1 : 0))
 
 #define SET_8812_H2CCMD_WOWLAN_FUNC_ENABLE(__cmd, __value)		\
-	SET_BITS_TO_LE_1BYTE(__cmd, 0, 1, __value)
+	u8p_replace_bits(__cmd, __value, BIT(0))
 #define SET_8812_H2CCMD_WOWLAN_PATTERN_MATCH_ENABLE(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE(__cmd, 1, 1, __value)
+	u8p_replace_bits(__cmd, __value, BIT(1))
 #define SET_8812_H2CCMD_WOWLAN_MAGIC_PKT_ENABLE(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE(__cmd, 2, 1, __value)
+	u8p_replace_bits(__cmd, __value, BIT(2))
 #define SET_8812_H2CCMD_WOWLAN_UNICAST_PKT_ENABLE(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE(__cmd, 3, 1, __value)
+	u8p_replace_bits(__cmd, __value, BIT(3))
 #define SET_8812_H2CCMD_WOWLAN_ALL_PKT_DROP(__cmd, __value)		\
-	SET_BITS_TO_LE_1BYTE(__cmd, 4, 1, __value)
+	u8p_replace_bits(__cmd, __value, BIT(4))
 #define SET_8812_H2CCMD_WOWLAN_GPIO_ACTIVE(__cmd, __value)		\
-	SET_BITS_TO_LE_1BYTE(__cmd, 5, 1, __value)
+	u8p_replace_bits(__cmd, __value, BIT(5))
 #define SET_8812_H2CCMD_WOWLAN_REKEY_WAKE_UP(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE(__cmd, 6, 1, __value)
+	u8p_replace_bits(__cmd, __value, BIT(6))
 #define SET_8812_H2CCMD_WOWLAN_DISCONNECT_WAKE_UP(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE(__cmd, 7, 1, __value)
+	u8p_replace_bits(__cmd, __value, BIT(7))
 #define SET_8812_H2CCMD_WOWLAN_GPIONUM(__cmd, __value)		\
-	SET_BITS_TO_LE_1BYTE((__cmd) + 1, 0, 8, __value)
+	*(u8 *)(__cmd + 1) = __value
 #define SET_8812_H2CCMD_WOWLAN_GPIO_DURATION(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE((__cmd) + 2, 0, 8, __value)
+	*(u8 *)(__cmd + 2) = __value
 
 #define SET_H2CCMD_PWRMODE_PARM_MODE(__ph2ccmd, __val)			\
-	SET_BITS_TO_LE_1BYTE(__ph2ccmd, 0, 8, __val)
+	*(u8 *)__ph2ccmd = __val
 #define SET_H2CCMD_PWRMODE_PARM_RLBM(__cmd, __value)		\
-	SET_BITS_TO_LE_1BYTE((__cmd)+1, 0, 4, __value)
+	u8p_replace_bits(__cmd + 1, __value, GENMASK(3, 0))
 #define SET_H2CCMD_PWRMODE_PARM_SMART_PS(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE((__cmd)+1, 4, 4, __value)
+	u8p_replace_bits(__cmd + 1, __value, GENMASK(7, 4))
 #define SET_H2CCMD_PWRMODE_PARM_AWAKE_INTERVAL(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE((__cmd)+2, 0, 8, __value)
+	*(u8 *)(__cmd + 2) = __value
 #define SET_H2CCMD_PWRMODE_PARM_ALL_QUEUE_UAPSD(__cmd, __value)		\
-	SET_BITS_TO_LE_1BYTE((__cmd)+3, 0, 8, __value)
+	*(u8 *)(__cmd + 3) = __value
 #define SET_H2CCMD_PWRMODE_PARM_PWR_STATE(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE((__cmd)+4, 0, 8, __value)
+	*(u8 *)(__cmd + 4) = __value
 #define SET_H2CCMD_PWRMODE_PARM_BYTE5(__cmd, __value)		\
-	SET_BITS_TO_LE_1BYTE((__cmd) + 5, 0, 8, __value)
-#define GET_8821AE_H2CCMD_PWRMODE_PARM_MODE(__cmd)		\
-	LE_BITS_TO_1BYTE(__cmd, 0, 8)
+	*(u8 *)(__cmd + 5) = __value
 
-#define SET_H2CCMD_JOINBSSRPT_PARM_OPMODE(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE(__ph2ccmd, 0, 8, __val)
 #define SET_H2CCMD_RSVDPAGE_LOC_PSPOLL(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd)+1, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 1) = __val
 #define SET_H2CCMD_RSVDPAGE_LOC_NULL_DATA(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd)+2, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 2) = __val
 #define SET_H2CCMD_RSVDPAGE_LOC_QOS_NULL_DATA(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd)+3, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 3) = __val
 #define SET_H2CCMD_RSVDPAGE_LOC_BT_QOS_NULL_DATA(__ph2ccmd, __val)	\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd) + 4, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 4) = __val
 
 /* _MEDIA_STATUS_RPT_PARM_CMD1 */
 #define SET_H2CCMD_MSRRPT_PARM_OPMODE(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE(__cmd, 0, 1, __value)
+	u8p_replace_bits(__cmd + 1, __value, BIT(0))
 #define SET_H2CCMD_MSRRPT_PARM_MACID_IND(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE(__cmd, 1, 1, __value)
-#define SET_H2CCMD_MSRRPT_PARM_MACID(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE(__cmd+1, 0, 8, __value)
-#define SET_H2CCMD_MSRRPT_PARM_MACID_END(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE(__cmd+2, 0, 8, __value)
+	u8p_replace_bits(__cmd + 1, __value, BIT(1))
 
 /* AP_OFFLOAD */
 #define SET_H2CCMD_AP_OFFLOAD_ON(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE(__cmd, 0, 8, __value)
+	*(u8 *)__cmd = __value
 #define SET_H2CCMD_AP_OFFLOAD_HIDDEN(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE((__cmd)+1, 0, 8, __value)
+	*(u8 *)(__cmd + 1) = __value
 #define SET_H2CCMD_AP_OFFLOAD_DENYANY(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE((__cmd)+2, 0, 8, __value)
+	*(u8 *)(__cmd + 2) = __value
 #define SET_H2CCMD_AP_OFFLOAD_WAKEUP_EVT_RPT(__cmd, __value) \
-	SET_BITS_TO_LE_1BYTE((__cmd)+3, 0, 8, __value)
+	*(u8 *)(__cmd + 3) = __value
 
 /* Keep Alive Control*/
 #define SET_8812_H2CCMD_KEEP_ALIVE_ENABLE(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE(__cmd, 0, 1, __value)
+	u8p_replace_bits(__cmd, __value, BIT(0))
 #define SET_8812_H2CCMD_KEEP_ALIVE_ACCPEPT_USER_DEFINED(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE(__cmd, 1, 1, __value)
+	u8p_replace_bits(__cmd, __value, BIT(1))
 #define SET_8812_H2CCMD_KEEP_ALIVE_PERIOD(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE((__cmd)+1, 0, 8, __value)
+	*(u8 *)(__cmd + 1) = __value
 
 /*REMOTE_WAKE_CTRL */
 #define SET_8812_H2CCMD_REMOTE_WAKECTRL_ENABLE(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE(__cmd, 0, 1, __value)
+	u8p_replace_bits(__cmd, __value, BIT(0))
 #define SET_8812_H2CCMD_REMOTE_WAKE_CTRL_ARP_OFFLOAD_EN(__cmd, __value)\
-	SET_BITS_TO_LE_1BYTE(__cmd, 1, 1, __value)
+	u8p_replace_bits(__cmd, __value, BIT(1))
 #define SET_8812_H2CCMD_REMOTE_WAKE_CTRL_NDP_OFFLOAD_EN(__cmd, __value)\
-	SET_BITS_TO_LE_1BYTE(__cmd, 2, 1, __value)
+	u8p_replace_bits(__cmd, __value, BIT(2))
 #define SET_8812_H2CCMD_REMOTE_WAKE_CTRL_GTK_OFFLOAD_EN(__cmd, __value)\
-	SET_BITS_TO_LE_1BYTE(__cmd, 3, 1, __value)
+	u8p_replace_bits(__cmd, __value, BIT(3))
 #define SET_8812_H2CCMD_REMOTE_WAKE_CTRL_REALWOWV2_EN(__cmd, __value)\
-	SET_BITS_TO_LE_1BYTE(__cmd, 6, 1, __value)
+	u8p_replace_bits(__cmd, __value, BIT(6))
 
 /* GTK_OFFLOAD */
 #define SET_8812_H2CCMD_AOAC_GLOBAL_INFO_PAIRWISE_ENC_ALG(__cmd, __value)\
-	SET_BITS_TO_LE_1BYTE(__cmd, 0, 8, __value)
+	*(u8 *)__cmd = __value
 #define SET_8812_H2CCMD_AOAC_GLOBAL_INFO_GROUP_ENC_ALG(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE((__cmd)+1, 0, 8, __value)
+	*(u8 *)(__cmd + 1) = __value
 
 /* AOAC_RSVDPAGE_LOC */
 #define SET_8821AE_H2CCMD_AOAC_RSVDPAGE_LOC_REMOTE_WAKE_CTRL_INFO(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE((__cmd), 0, 8, __value)
+	*(u8 *)__cmd = __value
 #define SET_8821AE_H2CCMD_AOAC_RSVDPAGE_LOC_ARP_RSP(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE((__cmd)+1, 0, 8, __value)
+	*(u8 *)(__cmd + 1) = __value
 #define SET_8821AE_H2CCMD_AOAC_RSVDPAGE_LOC_NEIGHBOR_ADV(__cmd, __value)\
-	SET_BITS_TO_LE_1BYTE((__cmd)+2, 0, 8, __value)
+	*(u8 *)(__cmd + 2) = __value
 #define SET_8821AE_H2CCMD_AOAC_RSVDPAGE_LOC_GTK_RSP(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE((__cmd)+3, 0, 8, __value)
+	*(u8 *)(__cmd + 3) = __value
 #define SET_8821AE_H2CCMD_AOAC_RSVDPAGE_LOC_GTK_INFO(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE((__cmd)+4, 0, 8, __value)
+	*(u8 *)(__cmd + 4) = __value
 #define SET_8821AE_H2CCMD_AOAC_RSVDPAGE_LOC_GTK_EXT_MEM(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE((__cmd)+5, 0, 8, __value)
+	*(u8 *)(__cmd + 5) = __value
 
 /* Disconnect_Decision_Control */
 #define SET_8812_H2CCMD_DISCONNECT_DECISION_CTRL_ENABLE(__cmd, __value)	\
-	SET_BITS_TO_LE_1BYTE(__cmd, 0, 1, __value)
+	u8p_replace_bits(__cmd, __value, BIT(0))
 #define SET_8812_H2CCMD_DISCONNECT_DECISION_CTRL_USER_SETTING(__cmd, __value)\
-	SET_BITS_TO_LE_1BYTE(__cmd, 1, 1, __value)
+	u8p_replace_bits(__cmd, __value, BIT(1))
 #define SET_8812_H2CCMD_DISCONNECT_DECISION_CTRL_CHECK_PERIOD(__cmd, __value)\
-	SET_BITS_TO_LE_1BYTE((__cmd)+1, 0, 8, __value) /* unit: beacon period */
+	*(u8 *)(__cmd + 1) = __value
 #define SET_8812_H2CCMD_DISCONNECT_DECISION_CTRL_TRYPKT_NUM(__cmd, __value)\
-	SET_BITS_TO_LE_1BYTE((__cmd)+2, 0, 8, __value)
+	*(u8 *)(__cmd + 2) = __value
 
 int rtl8821ae_download_fw(struct ieee80211_hw *hw, bool buse_wake_on_wlan_fw);
 #if (USE_SPECIFIC_FW_TO_SUPPORT_WOWLAN == 1)
-- 
cgit v1.2.3-59-g8ed1b


From 8e345762126374412e4ce783bf979c4fad0ea725 Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Mon, 25 Nov 2019 16:09:31 -0600
Subject: rtlwifi: rtl8192ee: Remove usage of private bit manipulation macros

Besides the previously changed macros for the RX and TX descriptors.
the local macros are used in other places. These are replaced with
standard bit manipulation macros.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 .../net/wireless/realtek/rtlwifi/rtl8192ee/fw.h    | 36 ++++++++++------------
 1 file changed, 16 insertions(+), 20 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/fw.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/fw.h
index cbfecea232a3..27ac4f8b9a9b 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/fw.h
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/fw.h
@@ -111,44 +111,40 @@ enum rtl8192e_h2c_cmd {
 	(u32)(((_len) >> 7) + ((_len) & 0x7F ? 1 : 0))
 
 #define SET_H2CCMD_PWRMODE_PARM_MODE(__ph2ccmd, __val)			\
-	SET_BITS_TO_LE_1BYTE(__ph2ccmd, 0, 8, __val)
+	*(u8 *)__ph2ccmd = __val;
 #define SET_H2CCMD_PWRMODE_PARM_RLBM(__cmd, __val)			\
-	SET_BITS_TO_LE_1BYTE((__cmd)+1, 0, 4, __val)
+	u8p_replace_bits(__cmd + 1, __val, GENMASK(3, 0))
 #define SET_H2CCMD_PWRMODE_PARM_SMART_PS(__cmd, __val)		\
-	SET_BITS_TO_LE_1BYTE((__cmd)+1, 4, 4, __val)
+	u8p_replace_bits(__cmd + 1, __val, GENMASK(7, 4))
 #define SET_H2CCMD_PWRMODE_PARM_AWAKE_INTERVAL(__cmd, __val)	\
-	SET_BITS_TO_LE_1BYTE((__cmd)+2, 0, 8, __val)
+	*(u8 *)(__cmd + 2) = __val;
 #define SET_H2CCMD_PWRMODE_PARM_ALL_QUEUE_UAPSD(__cmd, __val)	\
-	SET_BITS_TO_LE_1BYTE((__cmd)+3, 0, 8, __val)
+	*(u8 *)(__cmd + 3) = __val;
 #define SET_H2CCMD_PWRMODE_PARM_PWR_STATE(__cmd, __val)		\
-	SET_BITS_TO_LE_1BYTE((__cmd)+4, 0, 8, __val)
+	*(u8 *)(__cmd + 4) = __val;
 #define SET_H2CCMD_PWRMODE_PARM_BYTE5(__cmd, __val)		\
-	SET_BITS_TO_LE_1BYTE((__cmd) + 5, 0, 8, __val)
-#define GET_92E_H2CCMD_PWRMODE_PARM_MODE(__cmd)			\
-	LE_BITS_TO_1BYTE(__cmd, 0, 8)
+	*(u8 *)(__cmd + 5) = __val;
 
-#define SET_H2CCMD_JOINBSSRPT_PARM_OPMODE(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE(__ph2ccmd, 0, 8, __val)
 #define SET_H2CCMD_RSVDPAGE_LOC_PROBE_RSP(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE(__ph2ccmd, 0, 8, __val)
+	*(u8 *)__ph2ccmd = __val;
 #define SET_H2CCMD_RSVDPAGE_LOC_PSPOLL(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd)+1, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 1) = __val;
 #define SET_H2CCMD_RSVDPAGE_LOC_NULL_DATA(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd)+2, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 2) = __val;
 #define SET_H2CCMD_RSVDPAGE_LOC_QOS_NULL_DATA(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd) + 3, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 3) = __val;
 #define SET_H2CCMD_RSVDPAGE_LOC_BT_QOS_NULL_DATA(__ph2ccmd, __val)	\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd) + 4, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 4) = __val;
 
 /* _MEDIA_STATUS_RPT_PARM_CMD1 */
 #define SET_H2CCMD_MSRRPT_PARM_OPMODE(__cmd, __val)		\
-	SET_BITS_TO_LE_1BYTE(__cmd, 0, 1, __val)
+	u8p_replace_bits(__cmd, __val, BIT(0))
 #define SET_H2CCMD_MSRRPT_PARM_MACID_IND(__cmd, __val)		\
-	SET_BITS_TO_LE_1BYTE(__cmd, 1, 1, __val)
+	u8p_replace_bits(__cmd, __val, BIT(1))
 #define SET_H2CCMD_MSRRPT_PARM_MACID(__cmd, __val)		\
-	SET_BITS_TO_LE_1BYTE(__cmd+1, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 1) = __val;
 #define SET_H2CCMD_MSRRPT_PARM_MACID_END(__cmd, __val)		\
-	SET_BITS_TO_LE_1BYTE(__cmd+2, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 2) = __val;
 
 int rtl92ee_download_fw(struct ieee80211_hw *hw, bool buse_wake_on_wlan_fw);
 void rtl92ee_fill_h2c_cmd(struct ieee80211_hw *hw, u8 element_id,
-- 
cgit v1.2.3-59-g8ed1b


From 1626f192957552bd567261ad5103e7bd033e9ef0 Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Mon, 25 Nov 2019 16:09:32 -0600
Subject: rtlwifi: rtl8723ae: Remove usage of private bit manipulation macros

Besides the previously changed macros for the RX and TX descriptors.
the local macros are used in other places. These are replaced with
standard bit manipulation macros.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtlwifi/rtl8723ae/fw.h | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/fw.h b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/fw.h
index 5c843736de8d..3f9ed9b4428e 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/fw.h
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/fw.h
@@ -18,19 +18,19 @@
 #define pagenum_128(_len)	(u32)(((_len)>>7) + ((_len)&0x7F ? 1 : 0))
 
 #define SET_H2CCMD_PWRMODE_PARM_MODE(__ph2ccmd, __val)			\
-	SET_BITS_TO_LE_1BYTE(__ph2ccmd, 0, 8, __val)
+	*(u8 *)__ph2ccmd = __val
 #define SET_H2CCMD_PWRMODE_PARM_SMART_PS(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd)+1, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 1) = __val
 #define SET_H2CCMD_PWRMODE_PARM_BCN_PASS_TIME(__ph2ccmd, __val)	\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd)+2, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 2) = __val
 #define SET_H2CCMD_JOINBSSRPT_PARM_OPMODE(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE(__ph2ccmd, 0, 8, __val)
+	*(u8 *)__ph2ccmd = __val
 #define SET_H2CCMD_RSVDPAGE_LOC_PROBE_RSP(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE(__ph2ccmd, 0, 8, __val)
+	*(u8 *)__ph2ccmd = __val
 #define SET_H2CCMD_RSVDPAGE_LOC_PSPOLL(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd)+1, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 1) = __val
 #define SET_H2CCMD_RSVDPAGE_LOC_NULL_DATA(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd)+2, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 2) = __val
 
 void rtl8723e_fill_h2c_cmd(struct ieee80211_hw *hw, u8 element_id,
 			   u32 cmd_len, u8 *p_cmdbuffer);
-- 
cgit v1.2.3-59-g8ed1b


From 84bc9a58326a037a9472d8d899a22159542d706e Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Mon, 25 Nov 2019 16:09:33 -0600
Subject: rtlwifi: rtl8723be: Remove usage of private bit manipulation macros

Besides the previously changed macros for the RX and TX descriptors.
the local macros are used in other places. These are replaced with
standard bit manipulation macros.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 .../net/wireless/realtek/rtlwifi/rtl8723be/fw.h    | 30 ++++++++++------------
 1 file changed, 14 insertions(+), 16 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/fw.h b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/fw.h
index 97ea77464139..7c5e5e916274 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/fw.h
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/fw.h
@@ -83,37 +83,35 @@ enum rtl8723b_h2c_cmd {
 
 
 #define SET_H2CCMD_PWRMODE_PARM_MODE(__ph2ccmd, __val)			\
-	SET_BITS_TO_LE_1BYTE(__ph2ccmd, 0, 8, __val)
+	*(u8 *)__ph2ccmd = __val
 #define SET_H2CCMD_PWRMODE_PARM_RLBM(__ph2ccmd, __val)			\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd)+1, 0, 4, __val)
+	u8p_replace_bits(__ph2ccmd + 1, __val, GENMASK(3, 0))
 #define SET_H2CCMD_PWRMODE_PARM_SMART_PS(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd)+1, 4, 4, __val)
+	u8p_replace_bits(__ph2ccmd + 1, __val, GENMASK(7, 4))
 #define SET_H2CCMD_PWRMODE_PARM_AWAKE_INTERVAL(__ph2ccmd, __val)	\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd)+2, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 2) = __val
 #define SET_H2CCMD_PWRMODE_PARM_ALL_QUEUE_UAPSD(__ph2ccmd, __val)	\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd)+3, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 3) = __val
 #define SET_H2CCMD_PWRMODE_PARM_PWR_STATE(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd)+4, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 4) = __val
 #define SET_H2CCMD_PWRMODE_PARM_BYTE5(__ph2ccmd, __val)			\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd) + 5, 0, 8, __val)
-#define GET_88E_H2CCMD_PWRMODE_PARM_MODE(__ph2ccmd)			\
-	LE_BITS_TO_1BYTE(__ph2ccmd, 0, 8)
+	*(u8 *)(__ph2ccmd + 5) = __val
 
 #define SET_H2CCMD_MSRRPT_PARM_OPMODE(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE(__ph2ccmd, 0, 1, __val)
+	u8p_replace_bits(__ph2ccmd, __val, BIT(0))
 #define SET_H2CCMD_MSRRPT_PARM_MACID_IND(__ph2ccmd, __val)	\
-	SET_BITS_TO_LE_1BYTE(__ph2ccmd, 1, 1, __val)
+	u8p_replace_bits(__ph2ccmd, __val, BIT(1))
 
 #define SET_H2CCMD_RSVDPAGE_LOC_PROBE_RSP(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE(__ph2ccmd, 0, 8, __val)
+	*(u8 *)(__ph2ccmd) = __val
 #define SET_H2CCMD_RSVDPAGE_LOC_PSPOLL(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd)+1, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 1) = __val
 #define SET_H2CCMD_RSVDPAGE_LOC_NULL_DATA(__ph2ccmd, __val)		\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd)+2, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 2) = __val
 #define SET_H2CCMD_RSVDPAGE_LOC_QOS_NULL_DATA(__ph2ccmd, __val)	\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd) + 3, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 3) = __val
 #define SET_H2CCMD_RSVDPAGE_LOC_BT_QOS_NULL_DATA(__ph2ccmd, __val)	\
-	SET_BITS_TO_LE_1BYTE((__ph2ccmd) + 4, 0, 8, __val)
+	*(u8 *)(__ph2ccmd + 4) = __val
 
 
 void rtl8723be_fill_h2c_cmd(struct ieee80211_hw *hw, u8 element_id,
-- 
cgit v1.2.3-59-g8ed1b


From bb7e6d85536b4f1bfbc15e21dab829bf1377c8f8 Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Mon, 25 Nov 2019 16:09:34 -0600
Subject: rtlwifi: Remove last definitions of local bit manipulation macros

All usage of the local macros have been removed, thus the remaining
definitions are deleted.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtlwifi/wifi.h | 115 ----------------------------
 1 file changed, 115 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/wifi.h b/drivers/net/wireless/realtek/rtlwifi/wifi.h
index 3bdda1c98339..1cff9f07c9e9 100644
--- a/drivers/net/wireless/realtek/rtlwifi/wifi.h
+++ b/drivers/net/wireless/realtek/rtlwifi/wifi.h
@@ -2898,121 +2898,6 @@ enum bt_radio_shared {
  *	2. Before write integer to IO.
  *	3. After read integer from IO.
  ****************************************/
-/* Convert little data endian to host ordering */
-#define EF1BYTE(_val)		\
-	((u8)(_val))
-#define EF2BYTE(_val)		\
-	(le16_to_cpu(_val))
-#define EF4BYTE(_val)		\
-	(le32_to_cpu(_val))
-
-/* Read data from memory */
-#define READEF1BYTE(_ptr)      \
-	EF1BYTE(*((u8 *)(_ptr)))
-/* Read le16 data from memory and convert to host ordering */
-#define READEF2BYTE(_ptr)      \
-	EF2BYTE(*(_ptr))
-#define READEF4BYTE(_ptr)      \
-	EF4BYTE(*(_ptr))
-
-/* Create a bit mask
- * Examples:
- * BIT_LEN_MASK_32(0) => 0x00000000
- * BIT_LEN_MASK_32(1) => 0x00000001
- * BIT_LEN_MASK_32(2) => 0x00000003
- * BIT_LEN_MASK_32(32) => 0xFFFFFFFF
- */
-#define BIT_LEN_MASK_32(__bitlen)	 \
-	(0xFFFFFFFF >> (32 - (__bitlen)))
-#define BIT_LEN_MASK_16(__bitlen)	 \
-	(0xFFFF >> (16 - (__bitlen)))
-#define BIT_LEN_MASK_8(__bitlen) \
-	(0xFF >> (8 - (__bitlen)))
-
-/* Create an offset bit mask
- * Examples:
- * BIT_OFFSET_LEN_MASK_32(0, 2) => 0x00000003
- * BIT_OFFSET_LEN_MASK_32(16, 2) => 0x00030000
- */
-#define BIT_OFFSET_LEN_MASK_32(__bitoffset, __bitlen) \
-	(BIT_LEN_MASK_32(__bitlen) << (__bitoffset))
-#define BIT_OFFSET_LEN_MASK_16(__bitoffset, __bitlen) \
-	(BIT_LEN_MASK_16(__bitlen) << (__bitoffset))
-#define BIT_OFFSET_LEN_MASK_8(__bitoffset, __bitlen) \
-	(BIT_LEN_MASK_8(__bitlen) << (__bitoffset))
-
-/*Description:
- * Return 4-byte value in host byte ordering from
- * 4-byte pointer in little-endian system.
- */
-#define LE_P4BYTE_TO_HOST_4BYTE(__pstart) \
-	(EF4BYTE(*((__le32 *)(__pstart))))
-#define LE_P2BYTE_TO_HOST_2BYTE(__pstart) \
-	(EF2BYTE(*((__le16 *)(__pstart))))
-#define LE_P1BYTE_TO_HOST_1BYTE(__pstart) \
-	(EF1BYTE(*((u8 *)(__pstart))))
-
-/*Description:
- * Translate subfield (continuous bits in little-endian) of 4-byte
- * value to host byte ordering.
- */
-#define LE_BITS_TO_4BYTE(__pstart, __bitoffset, __bitlen) \
-	( \
-		(LE_P4BYTE_TO_HOST_4BYTE(__pstart) >> (__bitoffset))  & \
-		BIT_LEN_MASK_32(__bitlen) \
-	)
-#define LE_BITS_TO_2BYTE(__pstart, __bitoffset, __bitlen) \
-	( \
-		(LE_P2BYTE_TO_HOST_2BYTE(__pstart) >> (__bitoffset)) & \
-		BIT_LEN_MASK_16(__bitlen) \
-	)
-#define LE_BITS_TO_1BYTE(__pstart, __bitoffset, __bitlen) \
-	( \
-		(LE_P1BYTE_TO_HOST_1BYTE(__pstart) >> (__bitoffset)) & \
-		BIT_LEN_MASK_8(__bitlen) \
-	)
-
-/* Description:
- * Mask subfield (continuous bits in little-endian) of 4-byte value
- * and return the result in 4-byte value in host byte ordering.
- */
-#define LE_BITS_CLEARED_TO_4BYTE(__pstart, __bitoffset, __bitlen) \
-	( \
-		LE_P4BYTE_TO_HOST_4BYTE(__pstart)  & \
-		(~BIT_OFFSET_LEN_MASK_32(__bitoffset, __bitlen)) \
-	)
-#define LE_BITS_CLEARED_TO_2BYTE(__pstart, __bitoffset, __bitlen) \
-	( \
-		LE_P2BYTE_TO_HOST_2BYTE(__pstart) & \
-		(~BIT_OFFSET_LEN_MASK_16(__bitoffset, __bitlen)) \
-	)
-#define LE_BITS_CLEARED_TO_1BYTE(__pstart, __bitoffset, __bitlen) \
-	( \
-		LE_P1BYTE_TO_HOST_1BYTE(__pstart) & \
-		(~BIT_OFFSET_LEN_MASK_8(__bitoffset, __bitlen)) \
-	)
-
-/* Description:
- * Set subfield of little-endian 4-byte value to specified value.
- */
-#define SET_BITS_TO_LE_4BYTE(__pstart, __bitoffset, __bitlen, __val) \
-	*((__le32 *)(__pstart)) = \
-	cpu_to_le32( \
-		LE_BITS_CLEARED_TO_4BYTE(__pstart, __bitoffset, __bitlen) | \
-		((((u32)__val) & BIT_LEN_MASK_32(__bitlen)) << (__bitoffset)) \
-	)
-#define SET_BITS_TO_LE_2BYTE(__pstart, __bitoffset, __bitlen, __val) \
-	*((__le16 *)(__pstart)) = \
-	cpu_to_le16( \
-		LE_BITS_CLEARED_TO_2BYTE(__pstart, __bitoffset, __bitlen) | \
-		((((u16)__val) & BIT_LEN_MASK_16(__bitlen)) << (__bitoffset)) \
-	)
-#define SET_BITS_TO_LE_1BYTE(__pstart, __bitoffset, __bitlen, __val) \
-	*((u8 *)(__pstart)) = EF1BYTE \
-	( \
-		LE_BITS_CLEARED_TO_1BYTE(__pstart, __bitoffset, __bitlen) | \
-		((((u8)__val) & BIT_LEN_MASK_8(__bitlen)) << (__bitoffset)) \
-	)
 
 #define	N_BYTE_ALIGMENT(__value, __aligment) ((__aligment == 1) ? \
 	(__value) : (((__value + __aligment - 1) / __aligment) * __aligment))
-- 
cgit v1.2.3-59-g8ed1b


From 37bc6c72f5b76f3c6ae00e3001d3187100fbeb1b Mon Sep 17 00:00:00 2001
From: Austin Kim <austindh.kim@gmail.com>
Date: Wed, 27 Nov 2019 14:43:58 +0900
Subject: brcmsmac: Remove always false 'channel < 0' statement

As 'channel' is declared as u16, the following expression is always false.
   channel < 0

So we can remove unnecessary 'always false' statement.

Signed-off-by: Austin Kim <austindh.kim@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/broadcom/brcm80211/brcmsmac/main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/main.c b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/main.c
index 3f09d89ba922..7f2c15c799d2 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/main.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/main.c
@@ -5408,7 +5408,7 @@ int brcms_c_set_channel(struct brcms_c_info *wlc, u16 channel)
 {
 	u16 chspec = ch20mhz_chspec(channel);
 
-	if (channel < 0 || channel > MAXCHANNEL)
+	if (channel > MAXCHANNEL)
 		return -EINVAL;
 
 	if (!brcms_c_valid_chanspec_db(wlc->cmi, chspec))
-- 
cgit v1.2.3-59-g8ed1b


From 475eec112e4267232d10f4afe2f939a241692b6c Mon Sep 17 00:00:00 2001
From: Phong Tran <tranmanphong@gmail.com>
Date: Wed, 27 Nov 2019 00:55:26 +0700
Subject: b43legacy: Fix -Wcast-function-type

correct usage prototype of callback in tasklet_init().
Report by https://github.com/KSPP/linux/issues/20

Tested-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Phong Tran <tranmanphong@gmail.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/broadcom/b43legacy/main.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/broadcom/b43legacy/main.c b/drivers/net/wireless/broadcom/b43legacy/main.c
index 4325e91736eb..8b6b657c4b85 100644
--- a/drivers/net/wireless/broadcom/b43legacy/main.c
+++ b/drivers/net/wireless/broadcom/b43legacy/main.c
@@ -1275,8 +1275,9 @@ static void handle_irq_ucode_debug(struct b43legacy_wldev *dev)
 }
 
 /* Interrupt handler bottom-half */
-static void b43legacy_interrupt_tasklet(struct b43legacy_wldev *dev)
+static void b43legacy_interrupt_tasklet(unsigned long data)
 {
+	struct b43legacy_wldev *dev = (struct b43legacy_wldev *)data;
 	u32 reason;
 	u32 dma_reason[ARRAY_SIZE(dev->dma_reason)];
 	u32 merged_dma_reason = 0;
@@ -3741,7 +3742,7 @@ static int b43legacy_one_core_attach(struct ssb_device *dev,
 	b43legacy_set_status(wldev, B43legacy_STAT_UNINIT);
 	wldev->bad_frames_preempt = modparam_bad_frames_preempt;
 	tasklet_init(&wldev->isr_tasklet,
-		     (void (*)(unsigned long))b43legacy_interrupt_tasklet,
+		     b43legacy_interrupt_tasklet,
 		     (unsigned long)wldev);
 	if (modparam_pio)
 		wldev->__using_pio = true;
-- 
cgit v1.2.3-59-g8ed1b


From ebd77feb27e91bb5fe35a7818b7c13ea7435fb98 Mon Sep 17 00:00:00 2001
From: Phong Tran <tranmanphong@gmail.com>
Date: Wed, 27 Nov 2019 00:55:27 +0700
Subject: ipw2x00: Fix -Wcast-function-type

correct usage prototype of callback in tasklet_init().
Report by https://github.com/KSPP/linux/issues/20

Signed-off-by: Phong Tran <tranmanphong@gmail.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/intel/ipw2x00/ipw2100.c | 7 ++++---
 drivers/net/wireless/intel/ipw2x00/ipw2200.c | 5 +++--
 2 files changed, 7 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/intel/ipw2x00/ipw2100.c b/drivers/net/wireless/intel/ipw2x00/ipw2100.c
index c4c83ab60cbc..0579554ed4b3 100644
--- a/drivers/net/wireless/intel/ipw2x00/ipw2100.c
+++ b/drivers/net/wireless/intel/ipw2x00/ipw2100.c
@@ -3206,8 +3206,9 @@ static void ipw2100_tx_send_data(struct ipw2100_priv *priv)
 	}
 }
 
-static void ipw2100_irq_tasklet(struct ipw2100_priv *priv)
+static void ipw2100_irq_tasklet(unsigned long data)
 {
+	struct ipw2100_priv *priv = (struct ipw2100_priv *)data;
 	struct net_device *dev = priv->net_dev;
 	unsigned long flags;
 	u32 inta, tmp;
@@ -6006,7 +6007,7 @@ static void ipw2100_rf_kill(struct work_struct *work)
 	spin_unlock_irqrestore(&priv->low_lock, flags);
 }
 
-static void ipw2100_irq_tasklet(struct ipw2100_priv *priv);
+static void ipw2100_irq_tasklet(unsigned long data);
 
 static const struct net_device_ops ipw2100_netdev_ops = {
 	.ndo_open		= ipw2100_open,
@@ -6136,7 +6137,7 @@ static struct net_device *ipw2100_alloc_device(struct pci_dev *pci_dev,
 	INIT_DELAYED_WORK(&priv->rf_kill, ipw2100_rf_kill);
 	INIT_DELAYED_WORK(&priv->scan_event, ipw2100_scan_event);
 
-	tasklet_init(&priv->irq_tasklet, (void (*)(unsigned long))
+	tasklet_init(&priv->irq_tasklet,
 		     ipw2100_irq_tasklet, (unsigned long)priv);
 
 	/* NOTE:  We do not start the deferred work for status checks yet */
diff --git a/drivers/net/wireless/intel/ipw2x00/ipw2200.c b/drivers/net/wireless/intel/ipw2x00/ipw2200.c
index 31e43fc1d12b..5ef6f87a48ac 100644
--- a/drivers/net/wireless/intel/ipw2x00/ipw2200.c
+++ b/drivers/net/wireless/intel/ipw2x00/ipw2200.c
@@ -1945,8 +1945,9 @@ static void notify_wx_assoc_event(struct ipw_priv *priv)
 	wireless_send_event(priv->net_dev, SIOCGIWAP, &wrqu, NULL);
 }
 
-static void ipw_irq_tasklet(struct ipw_priv *priv)
+static void ipw_irq_tasklet(unsigned long data)
 {
+	struct ipw_priv *priv = (struct ipw_priv *)data;
 	u32 inta, inta_mask, handled = 0;
 	unsigned long flags;
 	int rc = 0;
@@ -10677,7 +10678,7 @@ static int ipw_setup_deferred_work(struct ipw_priv *priv)
 	INIT_WORK(&priv->qos_activate, ipw_bg_qos_activate);
 #endif				/* CONFIG_IPW2200_QOS */
 
-	tasklet_init(&priv->irq_tasklet, (void (*)(unsigned long))
+	tasklet_init(&priv->irq_tasklet,
 		     ipw_irq_tasklet, (unsigned long)priv);
 
 	return ret;
-- 
cgit v1.2.3-59-g8ed1b


From da5e57e8a6a3e69dac2937ba63fa86355628fbb2 Mon Sep 17 00:00:00 2001
From: Phong Tran <tranmanphong@gmail.com>
Date: Wed, 27 Nov 2019 00:55:28 +0700
Subject: iwlegacy: Fix -Wcast-function-type

correct usage prototype of callback in tasklet_init().
Report by https://github.com/KSPP/linux/issues/20

Signed-off-by: Phong Tran <tranmanphong@gmail.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/intel/iwlegacy/3945-mac.c | 5 +++--
 drivers/net/wireless/intel/iwlegacy/4965-mac.c | 5 +++--
 2 files changed, 6 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlegacy/3945-mac.c b/drivers/net/wireless/intel/iwlegacy/3945-mac.c
index 1168055da182..206b43b9dff8 100644
--- a/drivers/net/wireless/intel/iwlegacy/3945-mac.c
+++ b/drivers/net/wireless/intel/iwlegacy/3945-mac.c
@@ -1376,8 +1376,9 @@ il3945_dump_nic_error_log(struct il_priv *il)
 }
 
 static void
-il3945_irq_tasklet(struct il_priv *il)
+il3945_irq_tasklet(unsigned long data)
 {
+	struct il_priv *il = (struct il_priv *)data;
 	u32 inta, handled = 0;
 	u32 inta_fh;
 	unsigned long flags;
@@ -3401,7 +3402,7 @@ il3945_setup_deferred_work(struct il_priv *il)
 	timer_setup(&il->watchdog, il_bg_watchdog, 0);
 
 	tasklet_init(&il->irq_tasklet,
-		     (void (*)(unsigned long))il3945_irq_tasklet,
+		     il3945_irq_tasklet,
 		     (unsigned long)il);
 }
 
diff --git a/drivers/net/wireless/intel/iwlegacy/4965-mac.c b/drivers/net/wireless/intel/iwlegacy/4965-mac.c
index 3664f56f8cbd..d1e17589dbeb 100644
--- a/drivers/net/wireless/intel/iwlegacy/4965-mac.c
+++ b/drivers/net/wireless/intel/iwlegacy/4965-mac.c
@@ -4343,8 +4343,9 @@ il4965_synchronize_irq(struct il_priv *il)
 }
 
 static void
-il4965_irq_tasklet(struct il_priv *il)
+il4965_irq_tasklet(unsigned long data)
 {
+	struct il_priv *il = (struct il_priv *)data;
 	u32 inta, handled = 0;
 	u32 inta_fh;
 	unsigned long flags;
@@ -6237,7 +6238,7 @@ il4965_setup_deferred_work(struct il_priv *il)
 	timer_setup(&il->watchdog, il_bg_watchdog, 0);
 
 	tasklet_init(&il->irq_tasklet,
-		     (void (*)(unsigned long))il4965_irq_tasklet,
+		     il4965_irq_tasklet,
 		     (unsigned long)il);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From cb775c88da5d48a85d99d95219f637b6fad2e0e9 Mon Sep 17 00:00:00 2001
From: Phong Tran <tranmanphong@gmail.com>
Date: Wed, 27 Nov 2019 00:55:29 +0700
Subject: rtlwifi: rtl_pci: Fix -Wcast-function-type

correct usage prototype of callback in tasklet_init().
Report by https://github.com/KSPP/linux/issues/20

Signed-off-by: Phong Tran <tranmanphong@gmail.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtlwifi/pci.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/pci.c b/drivers/net/wireless/realtek/rtlwifi/pci.c
index f88d26535978..25335bd2873b 100644
--- a/drivers/net/wireless/realtek/rtlwifi/pci.c
+++ b/drivers/net/wireless/realtek/rtlwifi/pci.c
@@ -1061,13 +1061,15 @@ done:
 	return ret;
 }
 
-static void _rtl_pci_irq_tasklet(struct ieee80211_hw *hw)
+static void _rtl_pci_irq_tasklet(unsigned long data)
 {
+	struct ieee80211_hw *hw = (struct ieee80211_hw *)data;
 	_rtl_pci_tx_chk_waitq(hw);
 }
 
-static void _rtl_pci_prepare_bcn_tasklet(struct ieee80211_hw *hw)
+static void _rtl_pci_prepare_bcn_tasklet(unsigned long data)
 {
+	struct ieee80211_hw *hw = (struct ieee80211_hw *)data;
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 	struct rtl_pci *rtlpci = rtl_pcidev(rtl_pcipriv(hw));
 	struct rtl_mac *mac = rtl_mac(rtl_priv(hw));
@@ -1193,10 +1195,10 @@ static void _rtl_pci_init_struct(struct ieee80211_hw *hw,
 
 	/*task */
 	tasklet_init(&rtlpriv->works.irq_tasklet,
-		     (void (*)(unsigned long))_rtl_pci_irq_tasklet,
+		     _rtl_pci_irq_tasklet,
 		     (unsigned long)hw);
 	tasklet_init(&rtlpriv->works.irq_prepare_bcn_tasklet,
-		     (void (*)(unsigned long))_rtl_pci_prepare_bcn_tasklet,
+		     _rtl_pci_prepare_bcn_tasklet,
 		     (unsigned long)hw);
 	INIT_WORK(&rtlpriv->works.lps_change_work,
 		  rtl_lps_change_work_callback);
-- 
cgit v1.2.3-59-g8ed1b


From f427939391f290cbeabe0231eb8a116429d823f0 Mon Sep 17 00:00:00 2001
From: yu kuai <yukuai3@huawei.com>
Date: Wed, 27 Nov 2019 20:44:33 +0800
Subject: bcma: remove set but not used variable 'sizel'
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fixes gcc '-Wunused-but-set-variable' warning:

drivers/bcma/scan.c: In function ‘bcma_erom_get_addr_desc’:

drivers/bcma/scan.c:222:20: warning: variable ‘sizel’ set but
not used [-Wunused-but-set-variable]

It is never used, and so can be removed.

Fixes: 8369ae33b705 ("bcma: add Broadcom specific AMBA bus driver")
Signed-off-by: yu kuai <yukuai3@huawei.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/bcma/scan.c | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

diff --git a/drivers/bcma/scan.c b/drivers/bcma/scan.c
index 4a2d1b235fb5..1f2de714b401 100644
--- a/drivers/bcma/scan.c
+++ b/drivers/bcma/scan.c
@@ -219,7 +219,7 @@ static s32 bcma_erom_get_mst_port(struct bcma_bus *bus, u32 __iomem **eromptr)
 static u32 bcma_erom_get_addr_desc(struct bcma_bus *bus, u32 __iomem **eromptr,
 				  u32 type, u8 port)
 {
-	u32 addrl, addrh, sizel, sizeh = 0;
+	u32 addrl, addrh, sizeh = 0;
 	u32 size;
 
 	u32 ent = bcma_erom_get_ent(bus, eromptr);
@@ -239,12 +239,9 @@ static u32 bcma_erom_get_addr_desc(struct bcma_bus *bus, u32 __iomem **eromptr,
 
 	if ((ent & SCAN_ADDR_SZ) == SCAN_ADDR_SZ_SZD) {
 		size = bcma_erom_get_ent(bus, eromptr);
-		sizel = size & SCAN_SIZE_SZ;
 		if (size & SCAN_SIZE_SG32)
 			sizeh = bcma_erom_get_ent(bus, eromptr);
-	} else
-		sizel = SCAN_ADDR_SZ_BASE <<
-				((ent & SCAN_ADDR_SZ) >> SCAN_ADDR_SZ_SHIFT);
+	}
 
 	return addrl;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 57f9807d1ea2bd1154d9f58c39670fbc49b1975e Mon Sep 17 00:00:00 2001
From: Stanislaw Gruszka <sgruszka@redhat.com>
Date: Thu, 28 Nov 2019 10:11:23 +0100
Subject: rt2x00: implement reconfig_complete

mac80211 has .reconfig_complete that can be used as indicator of
HW restart end. This will allow to configure keys/IV differently
for HW restart and normal configuration.

Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ralink/rt2x00/rt2800pci.c |  1 +
 drivers/net/wireless/ralink/rt2x00/rt2800soc.c |  1 +
 drivers/net/wireless/ralink/rt2x00/rt2800usb.c |  1 +
 drivers/net/wireless/ralink/rt2x00/rt2x00.h    |  2 ++
 drivers/net/wireless/ralink/rt2x00/rt2x00dev.c | 11 -----------
 drivers/net/wireless/ralink/rt2x00/rt2x00mac.c | 20 ++++++++++++++++++++
 6 files changed, 25 insertions(+), 11 deletions(-)

diff --git a/drivers/net/wireless/ralink/rt2x00/rt2800pci.c b/drivers/net/wireless/ralink/rt2x00/rt2800pci.c
index a23c26574002..3868c07672bd 100644
--- a/drivers/net/wireless/ralink/rt2x00/rt2800pci.c
+++ b/drivers/net/wireless/ralink/rt2x00/rt2800pci.c
@@ -311,6 +311,7 @@ static const struct ieee80211_ops rt2800pci_mac80211_ops = {
 	.get_survey		= rt2800_get_survey,
 	.get_ringparam		= rt2x00mac_get_ringparam,
 	.tx_frames_pending	= rt2x00mac_tx_frames_pending,
+	.reconfig_complete	= rt2x00mac_reconfig_complete,
 };
 
 static const struct rt2800_ops rt2800pci_rt2800_ops = {
diff --git a/drivers/net/wireless/ralink/rt2x00/rt2800soc.c b/drivers/net/wireless/ralink/rt2x00/rt2800soc.c
index 7b931bb96a9e..bbfe1425c0ee 100644
--- a/drivers/net/wireless/ralink/rt2x00/rt2800soc.c
+++ b/drivers/net/wireless/ralink/rt2x00/rt2800soc.c
@@ -156,6 +156,7 @@ static const struct ieee80211_ops rt2800soc_mac80211_ops = {
 	.get_survey		= rt2800_get_survey,
 	.get_ringparam		= rt2x00mac_get_ringparam,
 	.tx_frames_pending	= rt2x00mac_tx_frames_pending,
+	.reconfig_complete	= rt2x00mac_reconfig_complete,
 };
 
 static const struct rt2800_ops rt2800soc_rt2800_ops = {
diff --git a/drivers/net/wireless/ralink/rt2x00/rt2800usb.c b/drivers/net/wireless/ralink/rt2x00/rt2800usb.c
index 0dfb55c69b73..4cc64fe481a7 100644
--- a/drivers/net/wireless/ralink/rt2x00/rt2800usb.c
+++ b/drivers/net/wireless/ralink/rt2x00/rt2800usb.c
@@ -654,6 +654,7 @@ static const struct ieee80211_ops rt2800usb_mac80211_ops = {
 	.get_survey		= rt2800_get_survey,
 	.get_ringparam		= rt2x00mac_get_ringparam,
 	.tx_frames_pending	= rt2x00mac_tx_frames_pending,
+	.reconfig_complete	= rt2x00mac_reconfig_complete,
 };
 
 static const struct rt2800_ops rt2800usb_rt2800_ops = {
diff --git a/drivers/net/wireless/ralink/rt2x00/rt2x00.h b/drivers/net/wireless/ralink/rt2x00/rt2x00.h
index a90a518b40d3..ea8a34ecae14 100644
--- a/drivers/net/wireless/ralink/rt2x00/rt2x00.h
+++ b/drivers/net/wireless/ralink/rt2x00/rt2x00.h
@@ -1439,6 +1439,8 @@ void rt2x00mac_tx(struct ieee80211_hw *hw,
 		  struct sk_buff *skb);
 int rt2x00mac_start(struct ieee80211_hw *hw);
 void rt2x00mac_stop(struct ieee80211_hw *hw);
+void rt2x00mac_reconfig_complete(struct ieee80211_hw *hw,
+				 enum ieee80211_reconfig_type reconfig_type);
 int rt2x00mac_add_interface(struct ieee80211_hw *hw,
 			    struct ieee80211_vif *vif);
 void rt2x00mac_remove_interface(struct ieee80211_hw *hw,
diff --git a/drivers/net/wireless/ralink/rt2x00/rt2x00dev.c b/drivers/net/wireless/ralink/rt2x00/rt2x00dev.c
index c3eab767bc21..7f9e43a4f805 100644
--- a/drivers/net/wireless/ralink/rt2x00/rt2x00dev.c
+++ b/drivers/net/wireless/ralink/rt2x00/rt2x00dev.c
@@ -1255,16 +1255,6 @@ int rt2x00lib_start(struct rt2x00_dev *rt2x00dev)
 {
 	int retval = 0;
 
-	if (test_bit(DEVICE_STATE_STARTED, &rt2x00dev->flags)) {
-		/*
-		 * This is special case for ieee80211_restart_hw(), otherwise
-		 * mac80211 never call start() two times in row without stop();
-		 */
-		set_bit(DEVICE_STATE_RESET, &rt2x00dev->flags);
-		rt2x00dev->ops->lib->pre_reset_hw(rt2x00dev);
-		rt2x00lib_stop(rt2x00dev);
-	}
-
 	/*
 	 * If this is the first interface which is added,
 	 * we should load the firmware now.
@@ -1292,7 +1282,6 @@ int rt2x00lib_start(struct rt2x00_dev *rt2x00dev)
 	set_bit(DEVICE_STATE_STARTED, &rt2x00dev->flags);
 
 out:
-	clear_bit(DEVICE_STATE_RESET, &rt2x00dev->flags);
 	return retval;
 }
 
diff --git a/drivers/net/wireless/ralink/rt2x00/rt2x00mac.c b/drivers/net/wireless/ralink/rt2x00/rt2x00mac.c
index beb20c5faf5f..32efbc8e9f92 100644
--- a/drivers/net/wireless/ralink/rt2x00/rt2x00mac.c
+++ b/drivers/net/wireless/ralink/rt2x00/rt2x00mac.c
@@ -165,6 +165,15 @@ int rt2x00mac_start(struct ieee80211_hw *hw)
 	if (!test_bit(DEVICE_STATE_PRESENT, &rt2x00dev->flags))
 		return 0;
 
+	if (test_bit(DEVICE_STATE_STARTED, &rt2x00dev->flags)) {
+		/*
+		 * This is special case for ieee80211_restart_hw(), otherwise
+		 * mac80211 never call start() two times in row without stop();
+		 */
+		set_bit(DEVICE_STATE_RESET, &rt2x00dev->flags);
+		rt2x00dev->ops->lib->pre_reset_hw(rt2x00dev);
+		rt2x00lib_stop(rt2x00dev);
+	}
 	return rt2x00lib_start(rt2x00dev);
 }
 EXPORT_SYMBOL_GPL(rt2x00mac_start);
@@ -180,6 +189,17 @@ void rt2x00mac_stop(struct ieee80211_hw *hw)
 }
 EXPORT_SYMBOL_GPL(rt2x00mac_stop);
 
+void
+rt2x00mac_reconfig_complete(struct ieee80211_hw *hw,
+			    enum ieee80211_reconfig_type reconfig_type)
+{
+	struct rt2x00_dev *rt2x00dev = hw->priv;
+
+	if (reconfig_type == IEEE80211_RECONFIG_TYPE_RESTART)
+		clear_bit(DEVICE_STATE_RESET, &rt2x00dev->flags);
+}
+EXPORT_SYMBOL_GPL(rt2x00mac_reconfig_complete);
+
 int rt2x00mac_add_interface(struct ieee80211_hw *hw,
 			    struct ieee80211_vif *vif)
 {
-- 
cgit v1.2.3-59-g8ed1b


From 2d52378a0ad0c2c3e8a2572aa883da46ddfaac4e Mon Sep 17 00:00:00 2001
From: Stanislaw Gruszka <sgruszka@redhat.com>
Date: Thu, 28 Nov 2019 10:11:24 +0100
Subject: rt2x00: use RESET state bit to prevent IV changes on restart

We can not set IV when we are under reset. Previously we read previous
IV value from registers and zeroed IV registers when key was deleted,
but this could possibly not work if DEL_KEY command was not performed.

Now when we have ->reconfig_complete ops and we fully mark mac80211
reset state via DEVICE_STATE_RESET flag, we can just skip setting IV
when keys are set during reset.

Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ralink/rt2x00/rt2800lib.c | 21 +++++++++------------
 1 file changed, 9 insertions(+), 12 deletions(-)

diff --git a/drivers/net/wireless/ralink/rt2x00/rt2800lib.c b/drivers/net/wireless/ralink/rt2x00/rt2800lib.c
index a36c3fea7495..6beac1f74e7c 100644
--- a/drivers/net/wireless/ralink/rt2x00/rt2800lib.c
+++ b/drivers/net/wireless/ralink/rt2x00/rt2800lib.c
@@ -1652,20 +1652,17 @@ static void rt2800_config_wcid_attr_cipher(struct rt2x00_dev *rt2x00dev,
 		rt2800_register_write(rt2x00dev, offset, reg);
 	}
 
-	offset = MAC_IVEIV_ENTRY(key->hw_key_idx);
+	if (test_bit(DEVICE_STATE_RESET, &rt2x00dev->flags))
+		return;
 
-	if (crypto->cmd == SET_KEY) {
-		rt2800_register_multiread(rt2x00dev, offset,
-					  &iveiv_entry, sizeof(iveiv_entry));
-		if ((crypto->cipher == CIPHER_TKIP) ||
-		    (crypto->cipher == CIPHER_TKIP_NO_MIC) ||
-		    (crypto->cipher == CIPHER_AES))
-			iveiv_entry.iv[3] |= 0x20;
-		iveiv_entry.iv[3] |= key->keyidx << 6;
-	} else {
-		memset(&iveiv_entry, 0, sizeof(iveiv_entry));
-	}
+	offset = MAC_IVEIV_ENTRY(key->hw_key_idx);
 
+	memset(&iveiv_entry, 0, sizeof(iveiv_entry));
+	if ((crypto->cipher == CIPHER_TKIP) ||
+	    (crypto->cipher == CIPHER_TKIP_NO_MIC) ||
+	    (crypto->cipher == CIPHER_AES))
+		iveiv_entry.iv[3] |= 0x20;
+	iveiv_entry.iv[3] |= key->keyidx << 6;
 	rt2800_register_multiwrite(rt2x00dev, offset,
 				   &iveiv_entry, sizeof(iveiv_entry));
 }
-- 
cgit v1.2.3-59-g8ed1b


From e93cd35101b61e4c79149be2cfc927c4b28dc60c Mon Sep 17 00:00:00 2001
From: Johan Hovold <johan@kernel.org>
Date: Thu, 28 Nov 2019 18:22:00 +0100
Subject: rsi: fix use-after-free on failed probe and unbind

Make sure to stop both URBs before returning after failed probe as well
as on disconnect to avoid use-after-free in the completion handler.

Reported-by: syzbot+b563b7f8dbe8223a51e8@syzkaller.appspotmail.com
Fixes: a4302bff28e2 ("rsi: add bluetooth rx endpoint")
Fixes: dad0d04fa7ba ("rsi: Add RS9113 wireless driver")
Cc: stable <stable@vger.kernel.org>     # 3.15
Cc: Siva Rebbagondla <siva.rebbagondla@redpinesignals.com>
Cc: Prameela Rani Garnepudi <prameela.j04cs@gmail.com>
Cc: Amitkumar Karwar <amit.karwar@redpinesignals.com>
Cc: Fariya Fatima <fariyaf@gmail.com>
Signed-off-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/rsi/rsi_91x_usb.c | 18 +++++++++++++++++-
 1 file changed, 17 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/rsi/rsi_91x_usb.c b/drivers/net/wireless/rsi/rsi_91x_usb.c
index 53f41fc2cadf..30bed719486e 100644
--- a/drivers/net/wireless/rsi/rsi_91x_usb.c
+++ b/drivers/net/wireless/rsi/rsi_91x_usb.c
@@ -292,6 +292,15 @@ out:
 		dev_kfree_skb(rx_cb->rx_skb);
 }
 
+static void rsi_rx_urb_kill(struct rsi_hw *adapter, u8 ep_num)
+{
+	struct rsi_91x_usbdev *dev = (struct rsi_91x_usbdev *)adapter->rsi_dev;
+	struct rx_usb_ctrl_block *rx_cb = &dev->rx_cb[ep_num - 1];
+	struct urb *urb = rx_cb->rx_urb;
+
+	usb_kill_urb(urb);
+}
+
 /**
  * rsi_rx_urb_submit() - This function submits the given URB to the USB stack.
  * @adapter: Pointer to the adapter structure.
@@ -823,10 +832,13 @@ static int rsi_probe(struct usb_interface *pfunction,
 	if (adapter->priv->coex_mode > 1) {
 		status = rsi_rx_urb_submit(adapter, BT_EP);
 		if (status)
-			goto err1;
+			goto err_kill_wlan_urb;
 	}
 
 	return 0;
+
+err_kill_wlan_urb:
+	rsi_rx_urb_kill(adapter, WLAN_EP);
 err1:
 	rsi_deinit_usb_interface(adapter);
 err:
@@ -857,6 +869,10 @@ static void rsi_disconnect(struct usb_interface *pfunction)
 		adapter->priv->bt_adapter = NULL;
 	}
 
+	if (adapter->priv->coex_mode > 1)
+		rsi_rx_urb_kill(adapter, BT_EP);
+	rsi_rx_urb_kill(adapter, WLAN_EP);
+
 	rsi_reset_card(adapter);
 	rsi_deinit_usb_interface(adapter);
 	rsi_91x_deinit(adapter);
-- 
cgit v1.2.3-59-g8ed1b


From 92aafe77123ab478e5f5095878856ab0424910da Mon Sep 17 00:00:00 2001
From: Johan Hovold <johan@kernel.org>
Date: Thu, 28 Nov 2019 18:22:01 +0100
Subject: rsi: fix use-after-free on probe errors

The driver would fail to stop the command timer in most error paths,
something which specifically could lead to the timer being freed while
still active on I/O errors during probe.

Fix this by making sure that each function starting the timer also stops
it in all relevant error paths.

Reported-by: syzbot+1d1597a5aa3679c65b9f@syzkaller.appspotmail.com
Fixes: b78e91bcfb33 ("rsi: Add new firmware loading method")
Cc: stable <stable@vger.kernel.org>     # 4.12
Cc: Prameela Rani Garnepudi <prameela.j04cs@gmail.com>
Cc: Amitkumar Karwar <amit.karwar@redpinesignals.com>
Signed-off-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/rsi/rsi_91x_hal.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/rsi/rsi_91x_hal.c b/drivers/net/wireless/rsi/rsi_91x_hal.c
index f84250bdb8cf..6f8d5f9a9f7e 100644
--- a/drivers/net/wireless/rsi/rsi_91x_hal.c
+++ b/drivers/net/wireless/rsi/rsi_91x_hal.c
@@ -622,6 +622,7 @@ static int bl_cmd(struct rsi_hw *adapter, u8 cmd, u8 exp_resp, char *str)
 	bl_start_cmd_timer(adapter, timeout);
 	status = bl_write_cmd(adapter, cmd, exp_resp, &regout_val);
 	if (status < 0) {
+		bl_stop_cmd_timer(adapter);
 		rsi_dbg(ERR_ZONE,
 			"%s: Command %s (%0x) writing failed..\n",
 			__func__, str, cmd);
@@ -737,10 +738,9 @@ static int ping_pong_write(struct rsi_hw *adapter, u8 cmd, u8 *addr, u32 size)
 	}
 
 	status = bl_cmd(adapter, cmd_req, cmd_resp, str);
-	if (status) {
-		bl_stop_cmd_timer(adapter);
+	if (status)
 		return status;
-	}
+
 	return 0;
 }
 
@@ -828,10 +828,9 @@ static int auto_fw_upgrade(struct rsi_hw *adapter, u8 *flash_content,
 
 	status = bl_cmd(adapter, EOF_REACHED, FW_LOADING_SUCCESSFUL,
 			"EOF_REACHED");
-	if (status) {
-		bl_stop_cmd_timer(adapter);
+	if (status)
 		return status;
-	}
+
 	rsi_dbg(INFO_ZONE, "FW loading is done and FW is running..\n");
 	return 0;
 }
@@ -849,6 +848,7 @@ static int rsi_hal_prepare_fwload(struct rsi_hw *adapter)
 						  &regout_val,
 						  RSI_COMMON_REG_SIZE);
 		if (status < 0) {
+			bl_stop_cmd_timer(adapter);
 			rsi_dbg(ERR_ZONE,
 				"%s: REGOUT read failed\n", __func__);
 			return status;
-- 
cgit v1.2.3-59-g8ed1b


From 47768297481184932844ab01a86752ba31a38861 Mon Sep 17 00:00:00 2001
From: Johan Hovold <johan@kernel.org>
Date: Thu, 28 Nov 2019 18:22:02 +0100
Subject: rsi: fix memory leak on failed URB submission

Make sure to free the skb on failed receive-URB submission (e.g. on
disconnect or currently also due to a missing endpoint).

Fixes: a1854fae1414 ("rsi: improve RX packet handling in USB interface")
Cc: stable <stable@vger.kernel.org>     # 4.17
Cc: Prameela Rani Garnepudi <prameela.j04cs@gmail.com>
Signed-off-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/rsi/rsi_91x_usb.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/rsi/rsi_91x_usb.c b/drivers/net/wireless/rsi/rsi_91x_usb.c
index 30bed719486e..2c869df1c62e 100644
--- a/drivers/net/wireless/rsi/rsi_91x_usb.c
+++ b/drivers/net/wireless/rsi/rsi_91x_usb.c
@@ -338,8 +338,10 @@ static int rsi_rx_urb_submit(struct rsi_hw *adapter, u8 ep_num)
 			  rx_cb);
 
 	status = usb_submit_urb(urb, GFP_KERNEL);
-	if (status)
+	if (status) {
 		rsi_dbg(ERR_ZONE, "%s: Failed in urb submission\n", __func__);
+		dev_kfree_skb(skb);
+	}
 
 	return status;
 }
-- 
cgit v1.2.3-59-g8ed1b


From b9b9f9fea21830f85cf0148cd8dce001ae55ead1 Mon Sep 17 00:00:00 2001
From: Johan Hovold <johan@kernel.org>
Date: Thu, 28 Nov 2019 18:22:03 +0100
Subject: rsi: fix non-atomic allocation in completion handler

USB completion handlers are called in atomic context and must
specifically not allocate memory using GFP_KERNEL.

Fixes: a1854fae1414 ("rsi: improve RX packet handling in USB interface")
Cc: stable <stable@vger.kernel.org> # 4.17
Cc: Prameela Rani Garnepudi <prameela.j04cs@gmail.com>
Signed-off-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/rsi/rsi_91x_usb.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/rsi/rsi_91x_usb.c b/drivers/net/wireless/rsi/rsi_91x_usb.c
index 2c869df1c62e..ead75574e10a 100644
--- a/drivers/net/wireless/rsi/rsi_91x_usb.c
+++ b/drivers/net/wireless/rsi/rsi_91x_usb.c
@@ -16,6 +16,7 @@
  */
 
 #include <linux/module.h>
+#include <linux/types.h>
 #include <net/rsi_91x.h>
 #include "rsi_usb.h"
 #include "rsi_hal.h"
@@ -29,7 +30,7 @@ MODULE_PARM_DESC(dev_oper_mode,
 		 "9[Wi-Fi STA + BT LE], 13[Wi-Fi STA + BT classic + BT LE]\n"
 		 "6[AP + BT classic], 14[AP + BT classic + BT LE]");
 
-static int rsi_rx_urb_submit(struct rsi_hw *adapter, u8 ep_num);
+static int rsi_rx_urb_submit(struct rsi_hw *adapter, u8 ep_num, gfp_t flags);
 
 /**
  * rsi_usb_card_write() - This function writes to the USB Card.
@@ -285,7 +286,7 @@ static void rsi_rx_done_handler(struct urb *urb)
 	status = 0;
 
 out:
-	if (rsi_rx_urb_submit(dev->priv, rx_cb->ep_num))
+	if (rsi_rx_urb_submit(dev->priv, rx_cb->ep_num, GFP_ATOMIC))
 		rsi_dbg(ERR_ZONE, "%s: Failed in urb submission", __func__);
 
 	if (status)
@@ -307,7 +308,7 @@ static void rsi_rx_urb_kill(struct rsi_hw *adapter, u8 ep_num)
  *
  * Return: 0 on success, a negative error code on failure.
  */
-static int rsi_rx_urb_submit(struct rsi_hw *adapter, u8 ep_num)
+static int rsi_rx_urb_submit(struct rsi_hw *adapter, u8 ep_num, gfp_t mem_flags)
 {
 	struct rsi_91x_usbdev *dev = (struct rsi_91x_usbdev *)adapter->rsi_dev;
 	struct rx_usb_ctrl_block *rx_cb = &dev->rx_cb[ep_num - 1];
@@ -337,7 +338,7 @@ static int rsi_rx_urb_submit(struct rsi_hw *adapter, u8 ep_num)
 			  rsi_rx_done_handler,
 			  rx_cb);
 
-	status = usb_submit_urb(urb, GFP_KERNEL);
+	status = usb_submit_urb(urb, mem_flags);
 	if (status) {
 		rsi_dbg(ERR_ZONE, "%s: Failed in urb submission\n", __func__);
 		dev_kfree_skb(skb);
@@ -827,12 +828,12 @@ static int rsi_probe(struct usb_interface *pfunction,
 		rsi_dbg(INIT_ZONE, "%s: Device Init Done\n", __func__);
 	}
 
-	status = rsi_rx_urb_submit(adapter, WLAN_EP);
+	status = rsi_rx_urb_submit(adapter, WLAN_EP, GFP_KERNEL);
 	if (status)
 		goto err1;
 
 	if (adapter->priv->coex_mode > 1) {
-		status = rsi_rx_urb_submit(adapter, BT_EP);
+		status = rsi_rx_urb_submit(adapter, BT_EP, GFP_KERNEL);
 		if (status)
 			goto err_kill_wlan_urb;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 960da557f435bfe14edb160a42b8ed3a2e255e73 Mon Sep 17 00:00:00 2001
From: Johan Hovold <johan@kernel.org>
Date: Thu, 28 Nov 2019 18:22:04 +0100
Subject: rsi: add missing endpoint sanity checks

The driver expects at least one bulk-in endpoint when in "wifi-alone"
operating mode and two bulk-in endpoints otherwise, and would otherwise
fail to to submit the corresponding bulk URB to the default pipe during
probe with a somewhat cryptic message:

	rsi_91x: rsi_rx_urb_submit: Failed in urb submission
	rsi_91x: rsi_probe: Failed in probe...Exiting
	RSI-USB WLAN: probe of 2-2.4:1.0 failed with error -8

The current endpoint sanity check looks broken and would only bail out
early if there was no bulk-in endpoint but at least one bulk-out
endpoint.

Tighten this check to always require at least one bulk-in and one
bulk-out endpoint, and add the missing sanity check for a Bluetooth
bulk-in endpoint when in a BT operating mode. Also make sure to log an
informative error message when the expected endpoints are missing.

Signed-off-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/rsi/rsi_91x_usb.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/rsi/rsi_91x_usb.c b/drivers/net/wireless/rsi/rsi_91x_usb.c
index ead75574e10a..396b9b81c1cd 100644
--- a/drivers/net/wireless/rsi/rsi_91x_usb.c
+++ b/drivers/net/wireless/rsi/rsi_91x_usb.c
@@ -149,9 +149,17 @@ static int rsi_find_bulk_in_and_out_endpoints(struct usb_interface *interface,
 			break;
 	}
 
-	if (!(dev->bulkin_endpoint_addr[0]) &&
-	    dev->bulkout_endpoint_addr[0])
+	if (!(dev->bulkin_endpoint_addr[0] && dev->bulkout_endpoint_addr[0])) {
+		dev_err(&interface->dev, "missing wlan bulk endpoints\n");
 		return -EINVAL;
+	}
+
+	if (adapter->priv->coex_mode > 1) {
+		if (!dev->bulkin_endpoint_addr[1]) {
+			dev_err(&interface->dev, "missing bt bulk-in endpoint\n");
+			return -EINVAL;
+		}
+	}
 
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 216b44000ada87a63891a8214c347e05a4aea8fe Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@oracle.com>
Date: Tue, 3 Dec 2019 12:58:55 +0300
Subject: brcmfmac: Fix use after free in brcmf_sdio_readframes()

The brcmu_pkt_buf_free_skb() function frees "pkt" so it leads to a
static checker warning:

    drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c:1974 brcmf_sdio_readframes()
    error: dereferencing freed memory 'pkt'

It looks like there was supposed to be a continue after we free "pkt".

Fixes: 4754fceeb9a6 ("brcmfmac: streamline SDIO read frame routine")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: Franky Lin <franky.lin@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
index 264ad63232f8..1dea0178832e 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
@@ -1935,6 +1935,7 @@ static uint brcmf_sdio_readframes(struct brcmf_sdio *bus, uint maxframes)
 					       BRCMF_SDIO_FT_NORMAL)) {
 				rd->len = 0;
 				brcmu_pkt_buf_free_skb(pkt);
+				continue;
 			}
 			bus->sdcnt.rx_readahead_cnt++;
 			if (rd->len != roundup(rd_new.len, 16)) {
-- 
cgit v1.2.3-59-g8ed1b


From ffc7b2826a3cd48369e0676b4dc083080ef96d7c Mon Sep 17 00:00:00 2001
From: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Date: Wed, 4 Dec 2019 06:43:55 +0100
Subject: rt2x00usb: Fix a warning message in 'rt2x00usb_watchdog_tx_dma()'

'forced' is duplicated in the message, axe one of them.

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ralink/rt2x00/rt2x00usb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ralink/rt2x00/rt2x00usb.c b/drivers/net/wireless/ralink/rt2x00/rt2x00usb.c
index bc2dfef0de22..92e9e023c349 100644
--- a/drivers/net/wireless/ralink/rt2x00/rt2x00usb.c
+++ b/drivers/net/wireless/ralink/rt2x00/rt2x00usb.c
@@ -522,7 +522,7 @@ EXPORT_SYMBOL_GPL(rt2x00usb_flush_queue);
 
 static void rt2x00usb_watchdog_tx_dma(struct data_queue *queue)
 {
-	rt2x00_warn(queue->rt2x00dev, "TX queue %d DMA timed out, invoke forced forced reset\n",
+	rt2x00_warn(queue->rt2x00dev, "TX queue %d DMA timed out, invoke forced reset\n",
 		    queue->qid);
 
 	rt2x00queue_stop_queue(queue);
-- 
cgit v1.2.3-59-g8ed1b


From d0b103a52b72a45b381332d4421ac3e2eaccecb7 Mon Sep 17 00:00:00 2001
From: Ganapathi Bhat <gbhat@marvell.com>
Date: Fri, 6 Dec 2019 13:47:48 +0530
Subject: MAINTAINERS: update Ganapathi Bhat's email address

I'd like to use this email-id from now on.

Signed-off-by: Ganapathi Bhat <gbhat@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 MAINTAINERS | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index 09a599fa3e1a..d09519805cc0 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -9956,7 +9956,7 @@ F:	drivers/net/ethernet/marvell/mvneta.*
 MARVELL MWIFIEX WIRELESS DRIVER
 M:	Amitkumar Karwar <amitkarwar@gmail.com>
 M:	Nishant Sarmukadam <nishants@marvell.com>
-M:	Ganapathi Bhat <gbhat@marvell.com>
+M:	Ganapathi Bhat <ganapathi.bhat@nxp.com>
 M:	Xinming Hu <huxinming820@gmail.com>
 L:	linux-wireless@vger.kernel.org
 S:	Maintained
-- 
cgit v1.2.3-59-g8ed1b


From 1c9f329b084b7b8ea6d60d91a202e884cdcf6aae Mon Sep 17 00:00:00 2001
From: Brian Norris <briannorris@chromium.org>
Date: Mon, 9 Dec 2019 16:39:11 -0800
Subject: mwifiex: delete unused mwifiex_get_intf_num()

Commit 7afb94da3cd8 ("mwifiex: update set_mac_address logic") fixed the
only user of this function, partly because the author seems to have
noticed that, as written, it's on the borderline between highly
misleading and buggy.

Anyway, no sense in keeping dead code around: let's drop it.

Fixes: 7afb94da3cd8 ("mwifiex: update set_mac_address logic")
Signed-off-by: Brian Norris <briannorris@chromium.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/marvell/mwifiex/main.h | 13 -------------
 1 file changed, 13 deletions(-)

diff --git a/drivers/net/wireless/marvell/mwifiex/main.h b/drivers/net/wireless/marvell/mwifiex/main.h
index 547ff3c578ee..fa5634af40f7 100644
--- a/drivers/net/wireless/marvell/mwifiex/main.h
+++ b/drivers/net/wireless/marvell/mwifiex/main.h
@@ -1295,19 +1295,6 @@ mwifiex_copy_rates(u8 *dest, u32 pos, u8 *src, int len)
 	return pos;
 }
 
-/* This function return interface number with the same bss_type.
- */
-static inline u8
-mwifiex_get_intf_num(struct mwifiex_adapter *adapter, u8 bss_type)
-{
-	u8 i, num = 0;
-
-	for (i = 0; i < adapter->priv_num; i++)
-		if (adapter->priv[i] && adapter->priv[i]->bss_type == bss_type)
-			num++;
-	return num;
-}
-
 /*
  * This function returns the correct private structure pointer based
  * upon the BSS type and BSS number.
-- 
cgit v1.2.3-59-g8ed1b


From 8d9627b05b2c33e4468e65739eb7caf9c3f274d8 Mon Sep 17 00:00:00 2001
From: Rafał Miłecki <rafal@milecki.pl>
Date: Tue, 10 Dec 2019 12:35:55 +0100
Subject: brcmfmac: set interface carrier to off by default
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

It's important as brcmfmac creates one main interface for each PHY and
doesn't allow deleting it. Not setting carrier could result in other
subsystems misbehaving (e.g. LEDs "netdev" trigger turning LED on).

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c
index 85cf96461dde..d3ddd97fe768 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c
@@ -661,6 +661,8 @@ int brcmf_net_attach(struct brcmf_if *ifp, bool rtnl_locked)
 		goto fail;
 	}
 
+	netif_carrier_off(ndev);
+
 	ndev->priv_destructor = brcmf_cfg80211_free_netdev;
 	brcmf_dbg(INFO, "%s: Broadcom Dongle Host Driver\n", ndev->name);
 	return 0;
-- 
cgit v1.2.3-59-g8ed1b


From af615abd15966387393471649185d15954e9cecd Mon Sep 17 00:00:00 2001
From: Johan Hovold <johan@kernel.org>
Date: Tue, 10 Dec 2019 12:44:21 +0100
Subject: at76c50x-usb: fix endpoint debug message

Make sure to use the current alternate setting, which may not be the
same as the first alternate setting, also when printing the number of
endpoints at probe.

Signed-off-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/atmel/at76c50x-usb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/atmel/at76c50x-usb.c b/drivers/net/wireless/atmel/at76c50x-usb.c
index db2c3b8d491e..3b2680772f03 100644
--- a/drivers/net/wireless/atmel/at76c50x-usb.c
+++ b/drivers/net/wireless/atmel/at76c50x-usb.c
@@ -2236,7 +2236,7 @@ static int at76_alloc_urbs(struct at76_priv *priv,
 	at76_dbg(DBG_PROC_ENTRY, "%s: ENTER", __func__);
 
 	at76_dbg(DBG_URB, "%s: NumEndpoints %d ", __func__,
-		 interface->altsetting[0].desc.bNumEndpoints);
+		 interface->cur_altsetting->desc.bNumEndpoints);
 
 	ep_in = NULL;
 	ep_out = NULL;
-- 
cgit v1.2.3-59-g8ed1b


From 3428fbcd6e6c0850b1a8b2a12082b7b2aabb3da3 Mon Sep 17 00:00:00 2001
From: Johan Hovold <johan@kernel.org>
Date: Tue, 10 Dec 2019 12:44:22 +0100
Subject: brcmfmac: fix interface sanity check

Make sure to use the current alternate setting when verifying the
interface descriptors to avoid binding to an invalid interface.

Failing to do so could cause the driver to misbehave or trigger a WARN()
in usb_submit_urb() that kernels with panic_on_warn set would choke on.

Fixes: 71bb244ba2fd ("brcm80211: fmac: add USB support for bcm43235/6/8 chipsets")
Cc: stable <stable@vger.kernel.org>     # 3.4
Cc: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c
index 06f3c01f10b3..7cdfde9b3dea 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c
@@ -1348,7 +1348,7 @@ brcmf_usb_probe(struct usb_interface *intf, const struct usb_device_id *id)
 		goto fail;
 	}
 
-	desc = &intf->altsetting[0].desc;
+	desc = &intf->cur_altsetting->desc;
 	if ((desc->bInterfaceClass != USB_CLASS_VENDOR_SPEC) ||
 	    (desc->bInterfaceSubClass != 2) ||
 	    (desc->bInterfaceProtocol != 0xff)) {
@@ -1361,7 +1361,7 @@ brcmf_usb_probe(struct usb_interface *intf, const struct usb_device_id *id)
 
 	num_of_eps = desc->bNumEndpoints;
 	for (ep = 0; ep < num_of_eps; ep++) {
-		endpoint = &intf->altsetting[0].endpoint[ep].desc;
+		endpoint = &intf->cur_altsetting->endpoint[ep].desc;
 		endpoint_num = usb_endpoint_num(endpoint);
 		if (!usb_endpoint_xfer_bulk(endpoint))
 			continue;
-- 
cgit v1.2.3-59-g8ed1b


From b73e05aa543cf8db4f4927e36952360d71291d41 Mon Sep 17 00:00:00 2001
From: Johan Hovold <johan@kernel.org>
Date: Tue, 10 Dec 2019 12:44:23 +0100
Subject: orinoco_usb: fix interface sanity check

Make sure to use the current alternate setting when verifying the
interface descriptors to avoid binding to an invalid interface.

Failing to do so could cause the driver to misbehave or trigger a WARN()
in usb_submit_urb() that kernels with panic_on_warn set would choke on.

Fixes: 9afac70a7305 ("orinoco: add orinoco_usb driver")
Cc: stable <stable@vger.kernel.org>     # 2.6.35
Signed-off-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/intersil/orinoco/orinoco_usb.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intersil/orinoco/orinoco_usb.c b/drivers/net/wireless/intersil/orinoco/orinoco_usb.c
index 40a8b941ad5c..8c79b963bcff 100644
--- a/drivers/net/wireless/intersil/orinoco/orinoco_usb.c
+++ b/drivers/net/wireless/intersil/orinoco/orinoco_usb.c
@@ -1608,9 +1608,9 @@ static int ezusb_probe(struct usb_interface *interface,
 	/* set up the endpoint information */
 	/* check out the endpoints */
 
-	iface_desc = &interface->altsetting[0].desc;
+	iface_desc = &interface->cur_altsetting->desc;
 	for (i = 0; i < iface_desc->bNumEndpoints; ++i) {
-		ep = &interface->altsetting[0].endpoint[i].desc;
+		ep = &interface->cur_altsetting->endpoint[i].desc;
 
 		if (usb_endpoint_is_bulk_in(ep)) {
 			/* we found a bulk in endpoint */
-- 
cgit v1.2.3-59-g8ed1b


From 39a4281c312f2d226c710bc656ce380c621a2b16 Mon Sep 17 00:00:00 2001
From: Johan Hovold <johan@kernel.org>
Date: Tue, 10 Dec 2019 12:44:24 +0100
Subject: rtl8xxxu: fix interface sanity check

Make sure to use the current alternate setting when verifying the
interface descriptors to avoid binding to an invalid interface.

Failing to do so could cause the driver to misbehave or trigger a WARN()
in usb_submit_urb() that kernels with panic_on_warn set would choke on.

Fixes: 26f1fad29ad9 ("New driver: rtl8xxxu (mac80211)")
Cc: stable <stable@vger.kernel.org>     # 4.4
Cc: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
index aa2bb2ae9809..54a1a4ea107b 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
@@ -6384,7 +6384,7 @@ static int rtl8xxxu_parse_usb(struct rtl8xxxu_priv *priv,
 	u8 dir, xtype, num;
 	int ret = 0;
 
-	host_interface = &interface->altsetting[0];
+	host_interface = interface->cur_altsetting;
 	interface_desc = &host_interface->desc;
 	endpoints = interface_desc->bNumEndpoints;
 
-- 
cgit v1.2.3-59-g8ed1b


From 3139b180906af43bc09bd3373fc2338a8271d9d9 Mon Sep 17 00:00:00 2001
From: Johan Hovold <johan@kernel.org>
Date: Tue, 10 Dec 2019 12:44:25 +0100
Subject: rsi_91x_usb: fix interface sanity check

Make sure to use the current alternate setting when verifying the
interface descriptors to avoid binding to an invalid interface.

Failing to do so could cause the driver to misbehave or trigger a WARN()
in usb_submit_urb() that kernels with panic_on_warn set would choke on.

Fixes: dad0d04fa7ba ("rsi: Add RS9113 wireless driver")
Cc: stable <stable@vger.kernel.org>     # 3.15
Cc: Fariya Fatima <fariyaf@gmail.com>
Signed-off-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/rsi/rsi_91x_usb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/rsi/rsi_91x_usb.c b/drivers/net/wireless/rsi/rsi_91x_usb.c
index 396b9b81c1cd..a62d41c0ccbc 100644
--- a/drivers/net/wireless/rsi/rsi_91x_usb.c
+++ b/drivers/net/wireless/rsi/rsi_91x_usb.c
@@ -118,7 +118,7 @@ static int rsi_find_bulk_in_and_out_endpoints(struct usb_interface *interface,
 	__le16 buffer_size;
 	int ii, bin_found = 0, bout_found = 0;
 
-	iface_desc = &(interface->altsetting[0]);
+	iface_desc = interface->cur_altsetting;
 
 	for (ii = 0; ii < iface_desc->desc.bNumEndpoints; ++ii) {
 		endpoint = &(iface_desc->endpoint[ii].desc);
-- 
cgit v1.2.3-59-g8ed1b


From 2d68bb2687abb747558b933e80845ff31570a49c Mon Sep 17 00:00:00 2001
From: Johan Hovold <johan@kernel.org>
Date: Tue, 10 Dec 2019 12:44:26 +0100
Subject: zd1211rw: fix storage endpoint lookup

Make sure to use the current alternate setting when verifying the
storage interface descriptors to avoid submitting an URB to an invalid
endpoint.

Failing to do so could cause the driver to misbehave or trigger a WARN()
in usb_submit_urb() that kernels with panic_on_warn set would choke on.

Fixes: a1030e92c150 ("[PATCH] zd1211rw: Convert installer CDROM device into WLAN device")
Cc: stable <stable@vger.kernel.org>     # 2.6.19
Signed-off-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/zydas/zd1211rw/zd_usb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/zydas/zd1211rw/zd_usb.c b/drivers/net/wireless/zydas/zd1211rw/zd_usb.c
index 7b5c2fe5bd4d..8ff0374126e4 100644
--- a/drivers/net/wireless/zydas/zd1211rw/zd_usb.c
+++ b/drivers/net/wireless/zydas/zd1211rw/zd_usb.c
@@ -1263,7 +1263,7 @@ static void print_id(struct usb_device *udev)
 static int eject_installer(struct usb_interface *intf)
 {
 	struct usb_device *udev = interface_to_usbdev(intf);
-	struct usb_host_interface *iface_desc = &intf->altsetting[0];
+	struct usb_host_interface *iface_desc = intf->cur_altsetting;
 	struct usb_endpoint_descriptor *endpoint;
 	unsigned char *cmd;
 	u8 bulk_out_ep;
-- 
cgit v1.2.3-59-g8ed1b


From f365f2f67b6a6e896864dbe92b0e5bbb34147eea Mon Sep 17 00:00:00 2001
From: Michael Straube <straube.linux@gmail.com>
Date: Wed, 11 Dec 2019 16:47:50 +0100
Subject: rtlwifi: rtl8192ce: use generic rtl_query_rxpwrpercentage

Function _rtl92c_query_rxpwrpercentage is identical to the generic
version rtl_query_rxpwrpercentage. Remove _rtl92c_query_rxpwrpercentage
and use the generic function.

Signed-off-by: Michael Straube <straube.linux@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtlwifi/rtl8192ce/trx.c | 14 ++------------
 1 file changed, 2 insertions(+), 12 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/trx.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/trx.c
index fc9a3aae047f..dec66b3ac365 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/trx.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/trx.c
@@ -23,16 +23,6 @@ static u8 _rtl92ce_map_hwqueue_to_fwqueue(struct sk_buff *skb, u8 hw_queue)
 	return skb->priority;
 }
 
-static u8 _rtl92c_query_rxpwrpercentage(s8 antpower)
-{
-	if ((antpower <= -100) || (antpower >= 20))
-		return 0;
-	else if (antpower >= 0)
-		return 100;
-	else
-		return 100 + antpower;
-}
-
 static long _rtl92ce_signal_scale_mapping(struct ieee80211_hw *hw,
 		long currsig)
 {
@@ -194,7 +184,7 @@ static void _rtl92ce_query_rxphystatus(struct ieee80211_hw *hw,
 			rx_pwr[i] =
 			    ((p_drvinfo->gain_trsw[i] & 0x3f) * 2) - 110;
 			/* Translate DBM to percentage. */
-			rssi = _rtl92c_query_rxpwrpercentage(rx_pwr[i]);
+			rssi = rtl_query_rxpwrpercentage(rx_pwr[i]);
 			total_rssi += rssi;
 			/* Get Rx snr value in DB */
 			rtlpriv->stats.rx_snr_db[i] =
@@ -209,7 +199,7 @@ static void _rtl92ce_query_rxphystatus(struct ieee80211_hw *hw,
 		 * hardware (for rate adaptive)
 		 */
 		rx_pwr_all = ((p_drvinfo->pwdb_all >> 1) & 0x7f) - 110;
-		pwdb_all = _rtl92c_query_rxpwrpercentage(rx_pwr_all);
+		pwdb_all = rtl_query_rxpwrpercentage(rx_pwr_all);
 		pstats->rx_pwdb_all = pwdb_all;
 		pstats->rxpower = rx_pwr_all;
 		pstats->recvsignalpower = rx_pwr_all;
-- 
cgit v1.2.3-59-g8ed1b


From b8a19dd64501e4b5019eed27f98b2f8675513f6d Mon Sep 17 00:00:00 2001
From: Michael Straube <straube.linux@gmail.com>
Date: Wed, 11 Dec 2019 16:47:51 +0100
Subject: rtlwifi: rtl8192cu: use generic rtl_query_rxpwrpercentage

Function _rtl92c_query_rxpwrpercentage is identical to the generic
version rtl_query_rxpwrpercentage. Remove _rtl92c_query_rxpwrpercentage
and use the generic function.

Signed-off-by: Michael Straube <straube.linux@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtlwifi/rtl8192cu/mac.c | 16 +++-------------
 1 file changed, 3 insertions(+), 13 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/mac.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/mac.c
index cec19b32c7e2..c8652283a516 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/mac.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/mac.c
@@ -567,16 +567,6 @@ void rtl92c_set_min_space(struct ieee80211_hw *hw, bool is2T)
 
 /*==============================================================*/
 
-static u8 _rtl92c_query_rxpwrpercentage(s8 antpower)
-{
-	if ((antpower <= -100) || (antpower >= 20))
-		return 0;
-	else if (antpower >= 0)
-		return 100;
-	else
-		return 100 + antpower;
-}
-
 static long _rtl92c_signal_scale_mapping(struct ieee80211_hw *hw,
 		long currsig)
 {
@@ -678,7 +668,7 @@ static void _rtl92c_query_rxphystatus(struct ieee80211_hw *hw,
 				break;
 			}
 		}
-		pwdb_all = _rtl92c_query_rxpwrpercentage(rx_pwr_all);
+		pwdb_all = rtl_query_rxpwrpercentage(rx_pwr_all);
 		pstats->rx_pwdb_all = pwdb_all;
 		pstats->recvsignalpower = rx_pwr_all;
 		if (packet_match_bssid) {
@@ -707,7 +697,7 @@ static void _rtl92c_query_rxphystatus(struct ieee80211_hw *hw,
 				rf_rx_num++;
 			rx_pwr[i] =
 			    ((p_drvinfo->gain_trsw[i] & 0x3f) * 2) - 110;
-			rssi = _rtl92c_query_rxpwrpercentage(rx_pwr[i]);
+			rssi = rtl_query_rxpwrpercentage(rx_pwr[i]);
 			total_rssi += rssi;
 			rtlpriv->stats.rx_snr_db[i] =
 			    (long)(p_drvinfo->rxsnr[i] / 2);
@@ -716,7 +706,7 @@ static void _rtl92c_query_rxphystatus(struct ieee80211_hw *hw,
 				pstats->rx_mimo_signalstrength[i] = (u8) rssi;
 		}
 		rx_pwr_all = ((p_drvinfo->pwdb_all >> 1) & 0x7f) - 110;
-		pwdb_all = _rtl92c_query_rxpwrpercentage(rx_pwr_all);
+		pwdb_all = rtl_query_rxpwrpercentage(rx_pwr_all);
 		pstats->rx_pwdb_all = pwdb_all;
 		pstats->rxpower = rx_pwr_all;
 		pstats->recvsignalpower = rx_pwr_all;
-- 
cgit v1.2.3-59-g8ed1b


From 5a87ae1d676ed31ec30779f6d98ea68afc5cbde7 Mon Sep 17 00:00:00 2001
From: Michael Straube <straube.linux@gmail.com>
Date: Wed, 11 Dec 2019 16:47:52 +0100
Subject: rtlwifi: rtl8192de: use generic rtl_query_rxpwrpercentage

Function _rtl92d_query_rxpwrpercentage is identical to the generic
version rtl_query_rxpwrpercentage. Remove _rtl92d_query_rxpwrpercentage
and use the generic function.

Signed-off-by: Michael Straube <straube.linux@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c | 16 +++-------------
 1 file changed, 3 insertions(+), 13 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c
index bb64612654b9..0fb6ef7e2a0e 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c
@@ -23,16 +23,6 @@ static u8 _rtl92de_map_hwqueue_to_fwqueue(struct sk_buff *skb, u8 hw_queue)
 	return skb->priority;
 }
 
-static u8 _rtl92d_query_rxpwrpercentage(s8 antpower)
-{
-	if ((antpower <= -100) || (antpower >= 20))
-		return 0;
-	else if (antpower >= 0)
-		return 100;
-	else
-		return 100 + antpower;
-}
-
 static long _rtl92de_translate_todbm(struct ieee80211_hw *hw,
 				     u8 signal_strength_index)
 {
@@ -141,7 +131,7 @@ static void _rtl92de_query_rxphystatus(struct ieee80211_hw *hw,
 				break;
 			}
 		}
-		pwdb_all = _rtl92d_query_rxpwrpercentage(rx_pwr_all);
+		pwdb_all = rtl_query_rxpwrpercentage(rx_pwr_all);
 		/* CCK gain is smaller than OFDM/MCS gain,  */
 		/* so we add gain diff by experiences, the val is 6 */
 		pwdb_all += 6;
@@ -183,7 +173,7 @@ static void _rtl92de_query_rxphystatus(struct ieee80211_hw *hw,
 				rf_rx_num++;
 			rx_pwr[i] = ((p_drvinfo->gain_trsw[i] & 0x3f) * 2)
 				    - 110;
-			rssi = _rtl92d_query_rxpwrpercentage(rx_pwr[i]);
+			rssi = rtl_query_rxpwrpercentage(rx_pwr[i]);
 			total_rssi += rssi;
 			rtlpriv->stats.rx_snr_db[i] =
 					 (long)(p_drvinfo->rxsnr[i] / 2);
@@ -191,7 +181,7 @@ static void _rtl92de_query_rxphystatus(struct ieee80211_hw *hw,
 				pstats->rx_mimo_signalstrength[i] = (u8) rssi;
 		}
 		rx_pwr_all = ((p_drvinfo->pwdb_all >> 1) & 0x7f) - 106;
-		pwdb_all = _rtl92d_query_rxpwrpercentage(rx_pwr_all);
+		pwdb_all = rtl_query_rxpwrpercentage(rx_pwr_all);
 		pstats->rx_pwdb_all = pwdb_all;
 		pstats->rxpower = rx_pwr_all;
 		pstats->recvsignalpower = rx_pwr_all;
-- 
cgit v1.2.3-59-g8ed1b


From d01b26734297b9fc5aaf110569d86254ede1db39 Mon Sep 17 00:00:00 2001
From: Michael Straube <straube.linux@gmail.com>
Date: Wed, 11 Dec 2019 16:47:53 +0100
Subject: rtlwifi: rtl8192ce: use generic rtl_signal_scale_mapping

Function _rtl92ce_signal_scale_mapping is identical to the generic
version rtl_signal_scale_mapping. Remove _rtl92ce_signal_scale_mapping
and use the generic function.

Signed-off-by: Michael Straube <straube.linux@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 .../net/wireless/realtek/rtlwifi/rtl8192ce/trx.c   | 34 ++--------------------
 1 file changed, 2 insertions(+), 32 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/trx.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/trx.c
index dec66b3ac365..8fc3cb824066 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/trx.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/trx.c
@@ -23,35 +23,6 @@ static u8 _rtl92ce_map_hwqueue_to_fwqueue(struct sk_buff *skb, u8 hw_queue)
 	return skb->priority;
 }
 
-static long _rtl92ce_signal_scale_mapping(struct ieee80211_hw *hw,
-		long currsig)
-{
-	long retsig;
-
-	if (currsig >= 61 && currsig <= 100)
-		retsig = 90 + ((currsig - 60) / 4);
-	else if (currsig >= 41 && currsig <= 60)
-		retsig = 78 + ((currsig - 40) / 2);
-	else if (currsig >= 31 && currsig <= 40)
-		retsig = 66 + (currsig - 30);
-	else if (currsig >= 21 && currsig <= 30)
-		retsig = 54 + (currsig - 20);
-	else if (currsig >= 5 && currsig <= 20)
-		retsig = 42 + (((currsig - 5) * 2) / 3);
-	else if (currsig == 4)
-		retsig = 36;
-	else if (currsig == 3)
-		retsig = 27;
-	else if (currsig == 2)
-		retsig = 18;
-	else if (currsig == 1)
-		retsig = 9;
-	else
-		retsig = currsig;
-
-	return retsig;
-}
-
 static void _rtl92ce_query_rxphystatus(struct ieee80211_hw *hw,
 				       struct rtl_stats *pstats,
 				       struct rx_desc_92c *pdesc,
@@ -231,11 +202,10 @@ static void _rtl92ce_query_rxphystatus(struct ieee80211_hw *hw,
 	 */
 	if (is_cck_rate)
 		pstats->signalstrength =
-		    (u8)(_rtl92ce_signal_scale_mapping(hw, pwdb_all));
+		    (u8)(rtl_signal_scale_mapping(hw, pwdb_all));
 	else if (rf_rx_num != 0)
 		pstats->signalstrength =
-		    (u8)(_rtl92ce_signal_scale_mapping
-			  (hw, total_rssi /= rf_rx_num));
+		    (u8)(rtl_signal_scale_mapping(hw, total_rssi /= rf_rx_num));
 }
 
 static void _rtl92ce_translate_rx_signal_stuff(struct ieee80211_hw *hw,
-- 
cgit v1.2.3-59-g8ed1b


From fef91a2b5f5f6631dd672063b8879ed5f2501c86 Mon Sep 17 00:00:00 2001
From: Michael Straube <straube.linux@gmail.com>
Date: Wed, 11 Dec 2019 16:47:54 +0100
Subject: rtlwifi: rtl8192cu: use generic rtl_signal_scale_mapping

Function _rtl92c_signal_scale_mapping is identical to the generic
version rtl_signal_scale_mapping. Remove _rtl92c_signal_scale_mapping
and use the generic function.

Signed-off-by: Michael Straube <straube.linux@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 .../net/wireless/realtek/rtlwifi/rtl8192cu/mac.c   | 33 ++--------------------
 1 file changed, 2 insertions(+), 31 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/mac.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/mac.c
index c8652283a516..b4b67341dc83 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/mac.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/mac.c
@@ -567,34 +567,6 @@ void rtl92c_set_min_space(struct ieee80211_hw *hw, bool is2T)
 
 /*==============================================================*/
 
-static long _rtl92c_signal_scale_mapping(struct ieee80211_hw *hw,
-		long currsig)
-{
-	long retsig;
-
-	if (currsig >= 61 && currsig <= 100)
-		retsig = 90 + ((currsig - 60) / 4);
-	else if (currsig >= 41 && currsig <= 60)
-		retsig = 78 + ((currsig - 40) / 2);
-	else if (currsig >= 31 && currsig <= 40)
-		retsig = 66 + (currsig - 30);
-	else if (currsig >= 21 && currsig <= 30)
-		retsig = 54 + (currsig - 20);
-	else if (currsig >= 5 && currsig <= 20)
-		retsig = 42 + (((currsig - 5) * 2) / 3);
-	else if (currsig == 4)
-		retsig = 36;
-	else if (currsig == 3)
-		retsig = 27;
-	else if (currsig == 2)
-		retsig = 18;
-	else if (currsig == 1)
-		retsig = 9;
-	else
-		retsig = currsig;
-	return retsig;
-}
-
 static void _rtl92c_query_rxphystatus(struct ieee80211_hw *hw,
 				      struct rtl_stats *pstats,
 				      struct rx_desc_92c *p_desc,
@@ -729,11 +701,10 @@ static void _rtl92c_query_rxphystatus(struct ieee80211_hw *hw,
 	}
 	if (is_cck_rate)
 		pstats->signalstrength =
-		    (u8) (_rtl92c_signal_scale_mapping(hw, pwdb_all));
+		    (u8)(rtl_signal_scale_mapping(hw, pwdb_all));
 	else if (rf_rx_num != 0)
 		pstats->signalstrength =
-		    (u8) (_rtl92c_signal_scale_mapping
-			  (hw, total_rssi /= rf_rx_num));
+		    (u8)(rtl_signal_scale_mapping(hw, total_rssi /= rf_rx_num));
 }
 
 void rtl92c_translate_rx_signal_stuff(struct ieee80211_hw *hw,
-- 
cgit v1.2.3-59-g8ed1b


From 716c733f52ecf7f587e617c2739e7a7410301968 Mon Sep 17 00:00:00 2001
From: Michael Straube <straube.linux@gmail.com>
Date: Wed, 11 Dec 2019 16:47:55 +0100
Subject: rtlwifi: rtl8192de: use generic rtl_signal_scale_mapping

Function _rtl92de_signal_scale_mapping is identical to the generic
version rtl_signal_scale_mapping. Remove _rtl92de_signal_scale_mapping
and use the generic function.

Signed-off-by: Michael Straube <straube.linux@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 .../net/wireless/realtek/rtlwifi/rtl8192de/trx.c   | 31 ++--------------------
 1 file changed, 2 insertions(+), 29 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c
index 0fb6ef7e2a0e..ab5b05ef168e 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c
@@ -33,33 +33,6 @@ static long _rtl92de_translate_todbm(struct ieee80211_hw *hw,
 	return signal_power;
 }
 
-static long _rtl92de_signal_scale_mapping(struct ieee80211_hw *hw, long currsig)
-{
-	long retsig;
-
-	if (currsig >= 61 && currsig <= 100)
-		retsig = 90 + ((currsig - 60) / 4);
-	else if (currsig >= 41 && currsig <= 60)
-		retsig = 78 + ((currsig - 40) / 2);
-	else if (currsig >= 31 && currsig <= 40)
-		retsig = 66 + (currsig - 30);
-	else if (currsig >= 21 && currsig <= 30)
-		retsig = 54 + (currsig - 20);
-	else if (currsig >= 5 && currsig <= 20)
-		retsig = 42 + (((currsig - 5) * 2) / 3);
-	else if (currsig == 4)
-		retsig = 36;
-	else if (currsig == 3)
-		retsig = 27;
-	else if (currsig == 2)
-		retsig = 18;
-	else if (currsig == 1)
-		retsig = 9;
-	else
-		retsig = currsig;
-	return retsig;
-}
-
 static void _rtl92de_query_rxphystatus(struct ieee80211_hw *hw,
 				       struct rtl_stats *pstats,
 				       struct rx_desc_92d *pdesc,
@@ -202,10 +175,10 @@ static void _rtl92de_query_rxphystatus(struct ieee80211_hw *hw,
 		}
 	}
 	if (is_cck_rate)
-		pstats->signalstrength = (u8)(_rtl92de_signal_scale_mapping(hw,
+		pstats->signalstrength = (u8)(rtl_signal_scale_mapping(hw,
 				pwdb_all));
 	else if (rf_rx_num != 0)
-		pstats->signalstrength = (u8)(_rtl92de_signal_scale_mapping(hw,
+		pstats->signalstrength = (u8)(rtl_signal_scale_mapping(hw,
 				total_rssi /= rf_rx_num));
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 1b8d2e0a9e4221b99eea375c079507ce8ef655f5 Mon Sep 17 00:00:00 2001
From: Wright Feng <wright.feng@cypress.com>
Date: Thu, 12 Dec 2019 00:52:45 +0100
Subject: brcmfmac: reset two D11 cores if chip has two D11 cores

There are two D11 cores in RSDB chips like 4359. We have to reset two
D11 cores simutaneously before firmware download, or the firmware may
not be initialized correctly and cause "fw initialized failed" error.

Signed-off-by: Wright Feng <wright.feng@cypress.com>
Signed-off-by: Soeren Moch <smoch@web.de>
Reviewed-by: Chi-Hsien Lin <chi-hsien.lin@cypress.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 .../wireless/broadcom/brcm80211/brcmfmac/chip.c    | 50 ++++++++++++++++++++++
 .../wireless/broadcom/brcm80211/brcmfmac/chip.h    |  1 +
 .../wireless/broadcom/brcm80211/brcmfmac/pcie.c    |  2 +-
 3 files changed, 52 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
index a795d781b4c5..0b5fbe5d8270 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
@@ -433,11 +433,25 @@ static void brcmf_chip_ai_resetcore(struct brcmf_core_priv *core, u32 prereset,
 {
 	struct brcmf_chip_priv *ci;
 	int count;
+	struct brcmf_core *d11core2 = NULL;
+	struct brcmf_core_priv *d11priv2 = NULL;
 
 	ci = core->chip;
 
+	/* special handle two D11 cores reset */
+	if (core->pub.id == BCMA_CORE_80211) {
+		d11core2 = brcmf_chip_get_d11core(&ci->pub, 1);
+		if (d11core2) {
+			brcmf_dbg(INFO, "found two d11 cores, reset both\n");
+			d11priv2 = container_of(d11core2,
+						struct brcmf_core_priv, pub);
+		}
+	}
+
 	/* must disable first to work for arbitrary current core state */
 	brcmf_chip_ai_coredisable(core, prereset, reset);
+	if (d11priv2)
+		brcmf_chip_ai_coredisable(d11priv2, prereset, reset);
 
 	count = 0;
 	while (ci->ops->read32(ci->ctx, core->wrapbase + BCMA_RESET_CTL) &
@@ -449,9 +463,30 @@ static void brcmf_chip_ai_resetcore(struct brcmf_core_priv *core, u32 prereset,
 		usleep_range(40, 60);
 	}
 
+	if (d11priv2) {
+		count = 0;
+		while (ci->ops->read32(ci->ctx,
+				       d11priv2->wrapbase + BCMA_RESET_CTL) &
+				       BCMA_RESET_CTL_RESET) {
+			ci->ops->write32(ci->ctx,
+					 d11priv2->wrapbase + BCMA_RESET_CTL,
+					 0);
+			count++;
+			if (count > 50)
+				break;
+			usleep_range(40, 60);
+		}
+	}
+
 	ci->ops->write32(ci->ctx, core->wrapbase + BCMA_IOCTL,
 			 postreset | BCMA_IOCTL_CLK);
 	ci->ops->read32(ci->ctx, core->wrapbase + BCMA_IOCTL);
+
+	if (d11priv2) {
+		ci->ops->write32(ci->ctx, d11priv2->wrapbase + BCMA_IOCTL,
+				 postreset | BCMA_IOCTL_CLK);
+		ci->ops->read32(ci->ctx, d11priv2->wrapbase + BCMA_IOCTL);
+	}
 }
 
 char *brcmf_chip_name(u32 id, u32 rev, char *buf, uint len)
@@ -1109,6 +1144,21 @@ void brcmf_chip_detach(struct brcmf_chip *pub)
 	kfree(chip);
 }
 
+struct brcmf_core *brcmf_chip_get_d11core(struct brcmf_chip *pub, u8 unit)
+{
+	struct brcmf_chip_priv *chip;
+	struct brcmf_core_priv *core;
+
+	chip = container_of(pub, struct brcmf_chip_priv, pub);
+	list_for_each_entry(core, &chip->cores, list) {
+		if (core->pub.id == BCMA_CORE_80211) {
+			if (unit-- == 0)
+				return &core->pub;
+		}
+	}
+	return NULL;
+}
+
 struct brcmf_core *brcmf_chip_get_core(struct brcmf_chip *pub, u16 coreid)
 {
 	struct brcmf_chip_priv *chip;
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.h
index 7b00f6a59e89..8fa38658e727 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.h
@@ -74,6 +74,7 @@ struct brcmf_chip *brcmf_chip_attach(void *ctx,
 				     const struct brcmf_buscore_ops *ops);
 void brcmf_chip_detach(struct brcmf_chip *chip);
 struct brcmf_core *brcmf_chip_get_core(struct brcmf_chip *chip, u16 coreid);
+struct brcmf_core *brcmf_chip_get_d11core(struct brcmf_chip *pub, u8 unit);
 struct brcmf_core *brcmf_chip_get_chipcommon(struct brcmf_chip *chip);
 struct brcmf_core *brcmf_chip_get_pmu(struct brcmf_chip *pub);
 bool brcmf_chip_iscoreup(struct brcmf_core *core);
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
index f64ce5074a55..7ac72804e285 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
@@ -78,7 +78,7 @@ static const struct brcmf_firmware_mapping brcmf_pcie_fwnames[] = {
 	BRCMF_FW_ENTRY(BRCM_CC_4371_CHIP_ID, 0xFFFFFFFF, 4371),
 };
 
-#define BRCMF_PCIE_FW_UP_TIMEOUT		2000 /* msec */
+#define BRCMF_PCIE_FW_UP_TIMEOUT		5000 /* msec */
 
 #define BRCMF_PCIE_REG_MAP_SIZE			(32 * 1024)
 
-- 
cgit v1.2.3-59-g8ed1b


From 172f6854551d48d1c9530f84513b421db944e714 Mon Sep 17 00:00:00 2001
From: Chung-Hsien Hsu <stanley.hsu@cypress.com>
Date: Thu, 12 Dec 2019 00:52:46 +0100
Subject: brcmfmac: set F2 blocksize and watermark for 4359

Set F2 blocksize to 256 bytes and watermark to 0x40 for 4359. Also
enable and configure F1 MesBusyCtrl. It fixes DMA error while having
UDP bi-directional traffic.

Signed-off-by: Chung-Hsien Hsu <stanley.hsu@cypress.com>
[slightly adapted for rebase on mainline linux]
Signed-off-by: Soeren Moch <smoch@web.de>
Reviewed-by: Chi-Hsien Lin <chi-hsien.lin@cypress.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c |  6 +++++-
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c   | 15 +++++++++++++++
 2 files changed, 20 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
index 96fd8e2bf773..68baf0189305 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
@@ -43,6 +43,7 @@
 
 #define SDIO_FUNC1_BLOCKSIZE		64
 #define SDIO_FUNC2_BLOCKSIZE		512
+#define SDIO_4359_FUNC2_BLOCKSIZE	256
 /* Maximum milliseconds to wait for F2 to come up */
 #define SDIO_WAIT_F2RDY	3000
 
@@ -903,6 +904,7 @@ static void brcmf_sdiod_host_fixup(struct mmc_host *host)
 static int brcmf_sdiod_probe(struct brcmf_sdio_dev *sdiodev)
 {
 	int ret = 0;
+	unsigned int f2_blksz = SDIO_FUNC2_BLOCKSIZE;
 
 	sdio_claim_host(sdiodev->func1);
 
@@ -912,7 +914,9 @@ static int brcmf_sdiod_probe(struct brcmf_sdio_dev *sdiodev)
 		sdio_release_host(sdiodev->func1);
 		goto out;
 	}
-	ret = sdio_set_block_size(sdiodev->func2, SDIO_FUNC2_BLOCKSIZE);
+	if (sdiodev->func2->device == SDIO_DEVICE_ID_BROADCOM_4359)
+		f2_blksz = SDIO_4359_FUNC2_BLOCKSIZE;
+	ret = sdio_set_block_size(sdiodev->func2, f2_blksz);
 	if (ret) {
 		brcmf_err("Failed to set F2 blocksize\n");
 		sdio_release_host(sdiodev->func1);
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
index 1dea0178832e..4d5b7770d4ea 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
@@ -42,6 +42,8 @@
 #define DEFAULT_F2_WATERMARK    0x8
 #define CY_4373_F2_WATERMARK    0x40
 #define CY_43012_F2_WATERMARK    0x60
+#define CY_4359_F2_WATERMARK	0x40
+#define CY_4359_F1_MESBUSYCTRL	(CY_4359_F2_WATERMARK | SBSDIO_MESBUSYCTRL_ENAB)
 
 #ifdef DEBUG
 
@@ -4206,6 +4208,19 @@ static void brcmf_sdio_firmware_callback(struct device *dev, int err,
 			brcmf_sdiod_writeb(sdiod, SBSDIO_DEVICE_CTL, devctl,
 					   &err);
 			break;
+		case SDIO_DEVICE_ID_BROADCOM_4359:
+			brcmf_dbg(INFO, "set F2 watermark to 0x%x*4 bytes\n",
+				  CY_4359_F2_WATERMARK);
+			brcmf_sdiod_writeb(sdiod, SBSDIO_WATERMARK,
+					   CY_4359_F2_WATERMARK, &err);
+			devctl = brcmf_sdiod_readb(sdiod, SBSDIO_DEVICE_CTL,
+						   &err);
+			devctl |= SBSDIO_DEVCTL_F2WM_ENAB;
+			brcmf_sdiod_writeb(sdiod, SBSDIO_DEVICE_CTL, devctl,
+					   &err);
+			brcmf_sdiod_writeb(sdiod, SBSDIO_FUNC1_MESBUSYCTRL,
+					   CY_4359_F1_MESBUSYCTRL, &err);
+			break;
 		default:
 			brcmf_sdiod_writeb(sdiod, SBSDIO_WATERMARK,
 					   DEFAULT_F2_WATERMARK, &err);
-- 
cgit v1.2.3-59-g8ed1b


From 6647274ed995a172369cb04754eb5f8b85f68f6d Mon Sep 17 00:00:00 2001
From: Soeren Moch <smoch@web.de>
Date: Thu, 12 Dec 2019 00:52:47 +0100
Subject: brcmfmac: fix rambase for 4359/9

Newer 4359 chip revisions need a different rambase address.
This fixes firmware download on such devices which fails otherwise.

Signed-off-by: Soeren Moch <smoch@web.de>
Acked-by: Chi-Hsien Lin <chi-hsien.lin@cypress.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
index 0b5fbe5d8270..baf72e3984fc 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
@@ -712,7 +712,6 @@ static u32 brcmf_chip_tcm_rambase(struct brcmf_chip_priv *ci)
 	case BRCM_CC_43569_CHIP_ID:
 	case BRCM_CC_43570_CHIP_ID:
 	case BRCM_CC_4358_CHIP_ID:
-	case BRCM_CC_4359_CHIP_ID:
 	case BRCM_CC_43602_CHIP_ID:
 	case BRCM_CC_4371_CHIP_ID:
 		return 0x180000;
@@ -722,6 +721,8 @@ static u32 brcmf_chip_tcm_rambase(struct brcmf_chip_priv *ci)
 	case BRCM_CC_4366_CHIP_ID:
 	case BRCM_CC_43664_CHIP_ID:
 		return 0x200000;
+	case BRCM_CC_4359_CHIP_ID:
+		return (ci->pub.chiprev < 9) ? 0x180000 : 0x160000;
 	case CY_CC_4373_CHIP_ID:
 		return 0x160000;
 	default:
-- 
cgit v1.2.3-59-g8ed1b


From c12c8913d79c49ceccb38f42714d25b783833758 Mon Sep 17 00:00:00 2001
From: Soeren Moch <smoch@web.de>
Date: Thu, 12 Dec 2019 00:52:48 +0100
Subject: brcmfmac: make errors when setting roaming parameters non-fatal

4359 dongles do not support setting roaming parameters (error -52).
Do not fail the 80211 configuration in this case.

Signed-off-by: Soeren Moch <smoch@web.de>
Acked-by: Chi-Hsien Lin <chi-hsien.lin@cypress.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
index 5598bbd09b62..0cf13cea1dbe 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
@@ -6012,19 +6012,17 @@ static s32 brcmf_dongle_roam(struct brcmf_if *ifp)
 	roamtrigger[1] = cpu_to_le32(BRCM_BAND_ALL);
 	err = brcmf_fil_cmd_data_set(ifp, BRCMF_C_SET_ROAM_TRIGGER,
 				     (void *)roamtrigger, sizeof(roamtrigger));
-	if (err) {
+	if (err)
 		bphy_err(drvr, "WLC_SET_ROAM_TRIGGER error (%d)\n", err);
-		goto roam_setup_done;
-	}
 
 	roam_delta[0] = cpu_to_le32(WL_ROAM_DELTA);
 	roam_delta[1] = cpu_to_le32(BRCM_BAND_ALL);
 	err = brcmf_fil_cmd_data_set(ifp, BRCMF_C_SET_ROAM_DELTA,
 				     (void *)roam_delta, sizeof(roam_delta));
-	if (err) {
+	if (err)
 		bphy_err(drvr, "WLC_SET_ROAM_DELTA error (%d)\n", err);
-		goto roam_setup_done;
-	}
+
+	return 0;
 
 roam_setup_done:
 	return err;
-- 
cgit v1.2.3-59-g8ed1b


From d4aef159394d5940bd7158ab789969dab82f7c76 Mon Sep 17 00:00:00 2001
From: Soeren Moch <smoch@web.de>
Date: Thu, 12 Dec 2019 00:52:49 +0100
Subject: brcmfmac: add support for BCM4359 SDIO chipset

BCM4359 is a 2x2 802.11 abgn+ac Dual-Band HT80 combo chip and it
supports Real Simultaneous Dual Band feature.

Based on a similar patch by: Wright Feng <wright.feng@cypress.com>

Signed-off-by: Soeren Moch <smoch@web.de>
Acked-by: Chi-Hsien Lin <chi-hsien.lin@cypress.com>
Acked-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c | 2 ++
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c   | 1 +
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c   | 2 ++
 include/linux/mmc/sdio_ids.h                              | 2 ++
 4 files changed, 7 insertions(+)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
index 68baf0189305..f4c53ab46058 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
@@ -973,8 +973,10 @@ static const struct sdio_device_id brcmf_sdmmc_ids[] = {
 	BRCMF_SDIO_DEVICE(SDIO_DEVICE_ID_BROADCOM_43455),
 	BRCMF_SDIO_DEVICE(SDIO_DEVICE_ID_BROADCOM_4354),
 	BRCMF_SDIO_DEVICE(SDIO_DEVICE_ID_BROADCOM_4356),
+	BRCMF_SDIO_DEVICE(SDIO_DEVICE_ID_BROADCOM_4359),
 	BRCMF_SDIO_DEVICE(SDIO_DEVICE_ID_CYPRESS_4373),
 	BRCMF_SDIO_DEVICE(SDIO_DEVICE_ID_CYPRESS_43012),
+	BRCMF_SDIO_DEVICE(SDIO_DEVICE_ID_CYPRESS_89359),
 	{ /* end: all zeroes */ }
 };
 MODULE_DEVICE_TABLE(sdio, brcmf_sdmmc_ids);
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
index baf72e3984fc..282d0bc14e8e 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
@@ -1408,6 +1408,7 @@ bool brcmf_chip_sr_capable(struct brcmf_chip *pub)
 		addr = CORE_CC_REG(base, sr_control0);
 		reg = chip->ops->read32(chip->ctx, addr);
 		return (reg & CC_SR_CTL0_ENABLE_MASK) != 0;
+	case BRCM_CC_4359_CHIP_ID:
 	case CY_CC_43012_CHIP_ID:
 		addr = CORE_CC_REG(pmu->base, retention_ctl);
 		reg = chip->ops->read32(chip->ctx, addr);
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
index 4d5b7770d4ea..f9df95bc7fa1 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
@@ -616,6 +616,7 @@ BRCMF_FW_DEF(43455, "brcmfmac43455-sdio");
 BRCMF_FW_DEF(43456, "brcmfmac43456-sdio");
 BRCMF_FW_DEF(4354, "brcmfmac4354-sdio");
 BRCMF_FW_DEF(4356, "brcmfmac4356-sdio");
+BRCMF_FW_DEF(4359, "brcmfmac4359-sdio");
 BRCMF_FW_DEF(4373, "brcmfmac4373-sdio");
 BRCMF_FW_DEF(43012, "brcmfmac43012-sdio");
 
@@ -638,6 +639,7 @@ static const struct brcmf_firmware_mapping brcmf_sdio_fwnames[] = {
 	BRCMF_FW_ENTRY(BRCM_CC_4345_CHIP_ID, 0xFFFFFDC0, 43455),
 	BRCMF_FW_ENTRY(BRCM_CC_4354_CHIP_ID, 0xFFFFFFFF, 4354),
 	BRCMF_FW_ENTRY(BRCM_CC_4356_CHIP_ID, 0xFFFFFFFF, 4356),
+	BRCMF_FW_ENTRY(BRCM_CC_4359_CHIP_ID, 0xFFFFFFFF, 4359),
 	BRCMF_FW_ENTRY(CY_CC_4373_CHIP_ID, 0xFFFFFFFF, 4373),
 	BRCMF_FW_ENTRY(CY_CC_43012_CHIP_ID, 0xFFFFFFFF, 43012)
 };
diff --git a/include/linux/mmc/sdio_ids.h b/include/linux/mmc/sdio_ids.h
index 08b25c02b5a1..2e9a6e4634eb 100644
--- a/include/linux/mmc/sdio_ids.h
+++ b/include/linux/mmc/sdio_ids.h
@@ -41,8 +41,10 @@
 #define SDIO_DEVICE_ID_BROADCOM_43455		0xa9bf
 #define SDIO_DEVICE_ID_BROADCOM_4354		0x4354
 #define SDIO_DEVICE_ID_BROADCOM_4356		0x4356
+#define SDIO_DEVICE_ID_BROADCOM_4359		0x4359
 #define SDIO_DEVICE_ID_CYPRESS_4373		0x4373
 #define SDIO_DEVICE_ID_CYPRESS_43012		43012
+#define SDIO_DEVICE_ID_CYPRESS_89359		0x4355
 
 #define SDIO_VENDOR_ID_INTEL			0x0089
 #define SDIO_DEVICE_ID_INTEL_IWMC3200WIMAX	0x1402
-- 
cgit v1.2.3-59-g8ed1b


From 837482e69a3f0d7cbc73922020012f83635f5ddb Mon Sep 17 00:00:00 2001
From: Wright Feng <wright.feng@cypress.com>
Date: Thu, 12 Dec 2019 00:52:50 +0100
Subject: brcmfmac: add RSDB condition when setting interface combinations

With firmware RSDB feature
1. The maximum support interface is four.
2. The maximum difference channel is two.
3. The maximum interfaces of {station/p2p client/AP} are two.
4. The maximum interface of p2p device is one.

Signed-off-by: Wright Feng <wright.feng@cypress.com>
Signed-off-by: Soeren Moch <smoch@web.de>
Reviewed-by: Chi-Hsien Lin <chi-hsien.lin@cypress.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 .../broadcom/brcm80211/brcmfmac/cfg80211.c         | 54 ++++++++++++++++++----
 1 file changed, 46 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
index 0cf13cea1dbe..9d9dc9195e9e 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
@@ -6520,6 +6520,9 @@ brcmf_txrx_stypes[NUM_NL80211_IFTYPES] = {
  *	#STA <= 1, #AP <= 1, channels = 1, 2 total
  *	#AP <= 4, matching BI, channels = 1, 4 total
  *
+ * no p2p and rsdb:
+ *	#STA <= 2, #AP <= 2, channels = 2, 4 total
+ *
  * p2p, no mchan, and mbss:
  *
  *	#STA <= 1, #P2P-DEV <= 1, #{P2P-CL, P2P-GO} <= 1, channels = 1, 3 total
@@ -6531,6 +6534,10 @@ brcmf_txrx_stypes[NUM_NL80211_IFTYPES] = {
  *	#STA <= 1, #P2P-DEV <= 1, #{P2P-CL, P2P-GO} <= 1, channels = 2, 3 total
  *	#STA <= 1, #P2P-DEV <= 1, #AP <= 1, #P2P-CL <= 1, channels = 1, 4 total
  *	#AP <= 4, matching BI, channels = 1, 4 total
+ *
+ * p2p, rsdb, and no mbss:
+ *	#STA <= 2, #P2P-DEV <= 1, #{P2P-CL, P2P-GO} <= 2, AP <= 2,
+ *	 channels = 2, 4 total
  */
 static int brcmf_setup_ifmodes(struct wiphy *wiphy, struct brcmf_if *ifp)
 {
@@ -6538,13 +6545,14 @@ static int brcmf_setup_ifmodes(struct wiphy *wiphy, struct brcmf_if *ifp)
 	struct ieee80211_iface_limit *c0_limits = NULL;
 	struct ieee80211_iface_limit *p2p_limits = NULL;
 	struct ieee80211_iface_limit *mbss_limits = NULL;
-	bool mbss, p2p;
+	bool mbss, p2p, rsdb;
 	int i, c, n_combos;
 
 	mbss = brcmf_feat_is_enabled(ifp, BRCMF_FEAT_MBSS);
 	p2p = brcmf_feat_is_enabled(ifp, BRCMF_FEAT_P2P);
+	rsdb = brcmf_feat_is_enabled(ifp, BRCMF_FEAT_RSDB);
 
-	n_combos = 1 + !!p2p + !!mbss;
+	n_combos = 1 + !!(p2p && !rsdb) + !!mbss;
 	combo = kcalloc(n_combos, sizeof(*combo), GFP_KERNEL);
 	if (!combo)
 		goto err;
@@ -6555,16 +6563,36 @@ static int brcmf_setup_ifmodes(struct wiphy *wiphy, struct brcmf_if *ifp)
 
 	c = 0;
 	i = 0;
-	c0_limits = kcalloc(p2p ? 3 : 2, sizeof(*c0_limits), GFP_KERNEL);
+	if (p2p && rsdb)
+		c0_limits = kcalloc(4, sizeof(*c0_limits), GFP_KERNEL);
+	else if (p2p)
+		c0_limits = kcalloc(3, sizeof(*c0_limits), GFP_KERNEL);
+	else
+		c0_limits = kcalloc(2, sizeof(*c0_limits), GFP_KERNEL);
 	if (!c0_limits)
 		goto err;
-	c0_limits[i].max = 1;
-	c0_limits[i++].types = BIT(NL80211_IFTYPE_STATION);
-	if (p2p) {
+	if (p2p && rsdb) {
+		combo[c].num_different_channels = 2;
+		wiphy->interface_modes |= BIT(NL80211_IFTYPE_P2P_CLIENT) |
+					  BIT(NL80211_IFTYPE_P2P_GO) |
+					  BIT(NL80211_IFTYPE_P2P_DEVICE);
+		c0_limits[i].max = 2;
+		c0_limits[i++].types = BIT(NL80211_IFTYPE_STATION);
+		c0_limits[i].max = 1;
+		c0_limits[i++].types = BIT(NL80211_IFTYPE_P2P_DEVICE);
+		c0_limits[i].max = 2;
+		c0_limits[i++].types = BIT(NL80211_IFTYPE_P2P_CLIENT) |
+				       BIT(NL80211_IFTYPE_P2P_GO);
+		c0_limits[i].max = 2;
+		c0_limits[i++].types = BIT(NL80211_IFTYPE_AP);
+		combo[c].max_interfaces = 5;
+	} else if (p2p) {
 		if (brcmf_feat_is_enabled(ifp, BRCMF_FEAT_MCHAN))
 			combo[c].num_different_channels = 2;
 		else
 			combo[c].num_different_channels = 1;
+		c0_limits[i].max = 1;
+		c0_limits[i++].types = BIT(NL80211_IFTYPE_STATION);
 		wiphy->interface_modes |= BIT(NL80211_IFTYPE_P2P_CLIENT) |
 					  BIT(NL80211_IFTYPE_P2P_GO) |
 					  BIT(NL80211_IFTYPE_P2P_DEVICE);
@@ -6573,16 +6601,26 @@ static int brcmf_setup_ifmodes(struct wiphy *wiphy, struct brcmf_if *ifp)
 		c0_limits[i].max = 1;
 		c0_limits[i++].types = BIT(NL80211_IFTYPE_P2P_CLIENT) |
 				       BIT(NL80211_IFTYPE_P2P_GO);
+		combo[c].max_interfaces = i;
+	} else if (rsdb) {
+		combo[c].num_different_channels = 2;
+		c0_limits[i].max = 2;
+		c0_limits[i++].types = BIT(NL80211_IFTYPE_STATION);
+		c0_limits[i].max = 2;
+		c0_limits[i++].types = BIT(NL80211_IFTYPE_AP);
+		combo[c].max_interfaces = 3;
 	} else {
 		combo[c].num_different_channels = 1;
 		c0_limits[i].max = 1;
+		c0_limits[i++].types = BIT(NL80211_IFTYPE_STATION);
+		c0_limits[i].max = 1;
 		c0_limits[i++].types = BIT(NL80211_IFTYPE_AP);
+		combo[c].max_interfaces = i;
 	}
-	combo[c].max_interfaces = i;
 	combo[c].n_limits = i;
 	combo[c].limits = c0_limits;
 
-	if (p2p) {
+	if (p2p && !rsdb) {
 		c++;
 		i = 0;
 		p2p_limits = kcalloc(4, sizeof(*p2p_limits), GFP_KERNEL);
-- 
cgit v1.2.3-59-g8ed1b


From 2635853ce4ab7654a77ab7080fb56de83408606b Mon Sep 17 00:00:00 2001
From: Wright Feng <wright.feng@cypress.com>
Date: Thu, 12 Dec 2019 00:52:51 +0100
Subject: brcmfmac: not set mbss in vif if firmware does not support MBSS

With RSDB mode, FMAC and firmware are able to create 2 or more AP,
so we should not set mbss in vif structure if firmware does not
support MBSS feature.

Signed-off-by: Wright Feng <wright.feng@cypress.com>
Signed-off-by: Soeren Moch <smoch@web.de>
Reviewed-by: Chi-Hsien Lin <chi-hsien.lin@cypress.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
index 9d9dc9195e9e..6eb3064c3721 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
@@ -5363,6 +5363,7 @@ struct brcmf_cfg80211_vif *brcmf_alloc_vif(struct brcmf_cfg80211_info *cfg,
 	struct brcmf_cfg80211_vif *vif_walk;
 	struct brcmf_cfg80211_vif *vif;
 	bool mbss;
+	struct brcmf_if *ifp = brcmf_get_ifp(cfg->pub, 0);
 
 	brcmf_dbg(TRACE, "allocating virtual interface (size=%zu)\n",
 		  sizeof(*vif));
@@ -5375,7 +5376,8 @@ struct brcmf_cfg80211_vif *brcmf_alloc_vif(struct brcmf_cfg80211_info *cfg,
 
 	brcmf_init_prof(&vif->profile);
 
-	if (type == NL80211_IFTYPE_AP) {
+	if (type == NL80211_IFTYPE_AP &&
+	    brcmf_feat_is_enabled(ifp, BRCMF_FEAT_MBSS)) {
 		mbss = false;
 		list_for_each_entry(vif_walk, &cfg->vif_list, list) {
 			if (vif_walk->wdev.iftype == NL80211_IFTYPE_AP) {
-- 
cgit v1.2.3-59-g8ed1b


From 4282dc057d750c6a7dd92953564b15c26b54c22c Mon Sep 17 00:00:00 2001
From: Navid Emamdoost <navid.emamdoost@gmail.com>
Date: Sat, 14 Dec 2019 19:51:14 -0600
Subject: brcmfmac: Fix memory leak in brcmf_usbdev_qinit

In the implementation of brcmf_usbdev_qinit() the allocated memory for
reqs is leaking if usb_alloc_urb() fails. Release reqs in the error
handling path.

Fixes: 71bb244ba2fd ("brcm80211: fmac: add USB support for bcm43235/6/8 chipsets")
Signed-off-by: Navid Emamdoost <navid.emamdoost@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c
index 7cdfde9b3dea..575ed19e9195 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c
@@ -430,6 +430,7 @@ fail:
 			usb_free_urb(req->urb);
 		list_del(q->next);
 	}
+	kfree(reqs);
 	return NULL;
 
 }
-- 
cgit v1.2.3-59-g8ed1b


From a32de68edab7b73ded850bcf76cdf6858e92a7e5 Mon Sep 17 00:00:00 2001
From: Dmitry Osipenko <digetx@gmail.com>
Date: Sun, 15 Dec 2019 21:42:24 +0300
Subject: brcmfmac: Keep OOB wake-interrupt disabled when it shouldn't be
 enabled

NVIDIA Tegra SoCs do not like when OOB wake is enabled and WiFi interface
is in DOWN state during suspend. This results in a CPU hang on programming
OOB wake-up state of the GPIO controller during of system's suspend.

The solution is trivial: don't enable wake for the OOB interrupt when it
should be disabled.

This fixes hang on Tegra20 (Acer A500) and Tegra30 (Nexus 7) devices which
are using BCM4329 and BCM4330 WiFi chips respectively.

Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c | 10 +++++-----
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.h   |  1 -
 2 files changed, 5 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
index f4c53ab46058..b684a5b6d904 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
@@ -120,7 +120,7 @@ int brcmf_sdiod_intr_register(struct brcmf_sdio_dev *sdiodev)
 			brcmf_err("enable_irq_wake failed %d\n", ret);
 			return ret;
 		}
-		sdiodev->irq_wake = true;
+		disable_irq_wake(pdata->oob_irq_nr);
 
 		sdio_claim_host(sdiodev->func1);
 
@@ -179,10 +179,6 @@ void brcmf_sdiod_intr_unregister(struct brcmf_sdio_dev *sdiodev)
 		sdio_release_host(sdiodev->func1);
 
 		sdiodev->oob_irq_requested = false;
-		if (sdiodev->irq_wake) {
-			disable_irq_wake(pdata->oob_irq_nr);
-			sdiodev->irq_wake = false;
-		}
 		free_irq(pdata->oob_irq_nr, &sdiodev->func1->dev);
 		sdiodev->irq_en = false;
 		sdiodev->oob_irq_requested = false;
@@ -1173,6 +1169,10 @@ static int brcmf_ops_sdio_resume(struct device *dev)
 		if (ret)
 			brcmf_err("Failed to probe device on resume\n");
 	} else {
+		if (sdiodev->wowl_enabled &&
+		    sdiodev->settings->bus.sdio.oob_irq_supported)
+			disable_irq_wake(sdiodev->settings->bus.sdio.oob_irq_nr);
+
 		brcmf_sdiod_freezer_off(sdiodev);
 	}
 
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.h
index 0bd47c119dae..163fd664780a 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.h
@@ -178,7 +178,6 @@ struct brcmf_sdio_dev {
 	bool sd_irq_requested;
 	bool irq_en;			/* irq enable flags */
 	spinlock_t irq_en_lock;
-	bool irq_wake;			/* irq wake enable flags */
 	bool sg_support;
 	uint max_request_size;
 	ushort max_segment_count;
-- 
cgit v1.2.3-59-g8ed1b


From c705f9fc6a1736dcf6ec01f8206707c108dca824 Mon Sep 17 00:00:00 2001
From: Aditya Pakki <pakki001@umn.edu>
Date: Sun, 15 Dec 2019 13:58:58 -0600
Subject: orinoco: avoid assertion in case of NULL pointer

In ezusb_init, if upriv is NULL, the code crashes. However, the caller
in ezusb_probe can handle the error and print the failure message.
The patch replaces the BUG_ON call to error return.

Signed-off-by: Aditya Pakki <pakki001@umn.edu>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/intersil/orinoco/orinoco_usb.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intersil/orinoco/orinoco_usb.c b/drivers/net/wireless/intersil/orinoco/orinoco_usb.c
index 8c79b963bcff..e753f43e0162 100644
--- a/drivers/net/wireless/intersil/orinoco/orinoco_usb.c
+++ b/drivers/net/wireless/intersil/orinoco/orinoco_usb.c
@@ -1361,7 +1361,8 @@ static int ezusb_init(struct hermes *hw)
 	int retval;
 
 	BUG_ON(in_interrupt());
-	BUG_ON(!upriv);
+	if (!upriv)
+		return -EINVAL;
 
 	upriv->reply_count = 0;
 	/* Write the MAGIC number on the simulated registers to keep
-- 
cgit v1.2.3-59-g8ed1b


From 7d0b447a3f82f5a1a70bae20c362f332e9fa91ea Mon Sep 17 00:00:00 2001
From: Jose Abreu <Jose.Abreu@synopsys.com>
Date: Wed, 18 Dec 2019 11:24:43 +0100
Subject: net: stmmac: Print more information in DebugFS DMA Capabilities file

DMA Capabilites have grown but the DebugFS that shows this info has not
been updated. Lets add the missing information.

Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 31 ++++++++++++++++++++++-
 1 file changed, 30 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index da80866d0371..3299f5bb02e2 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -4238,9 +4238,38 @@ static int stmmac_dma_cap_show(struct seq_file *seq, void *v)
 		   priv->dma_cap.number_rx_channel);
 	seq_printf(seq, "\tNumber of Additional TX channel: %d\n",
 		   priv->dma_cap.number_tx_channel);
+	seq_printf(seq, "\tNumber of Additional RX queues: %d\n",
+		   priv->dma_cap.number_rx_queues);
+	seq_printf(seq, "\tNumber of Additional TX queues: %d\n",
+		   priv->dma_cap.number_tx_queues);
 	seq_printf(seq, "\tEnhanced descriptors: %s\n",
 		   (priv->dma_cap.enh_desc) ? "Y" : "N");
-
+	seq_printf(seq, "\tTX Fifo Size: %d\n", priv->dma_cap.tx_fifo_size);
+	seq_printf(seq, "\tRX Fifo Size: %d\n", priv->dma_cap.rx_fifo_size);
+	seq_printf(seq, "\tHash Table Size: %d\n", priv->dma_cap.hash_tb_sz);
+	seq_printf(seq, "\tTSO: %s\n", priv->dma_cap.tsoen ? "Y" : "N");
+	seq_printf(seq, "\tNumber of PPS Outputs: %d\n",
+		   priv->dma_cap.pps_out_num);
+	seq_printf(seq, "\tSafety Features: %s\n",
+		   priv->dma_cap.asp ? "Y" : "N");
+	seq_printf(seq, "\tFlexible RX Parser: %s\n",
+		   priv->dma_cap.frpsel ? "Y" : "N");
+	seq_printf(seq, "\tEnhanced Addressing: %d\n",
+		   priv->dma_cap.addr64);
+	seq_printf(seq, "\tReceive Side Scaling: %s\n",
+		   priv->dma_cap.rssen ? "Y" : "N");
+	seq_printf(seq, "\tVLAN Hash Filtering: %s\n",
+		   priv->dma_cap.vlhash ? "Y" : "N");
+	seq_printf(seq, "\tSplit Header: %s\n",
+		   priv->dma_cap.sphen ? "Y" : "N");
+	seq_printf(seq, "\tVLAN TX Insertion: %s\n",
+		   priv->dma_cap.vlins ? "Y" : "N");
+	seq_printf(seq, "\tDouble VLAN: %s\n",
+		   priv->dma_cap.dvlan ? "Y" : "N");
+	seq_printf(seq, "\tNumber of L3/L4 Filters: %d\n",
+		   priv->dma_cap.l3l4fnum);
+	seq_printf(seq, "\tARP Offloading: %s\n",
+		   priv->dma_cap.arpoffsel ? "Y" : "N");
 	return 0;
 }
 DEFINE_SHOW_ATTRIBUTE(stmmac_dma_cap);
-- 
cgit v1.2.3-59-g8ed1b


From 021bd5e369709717231338a182f00611794fb588 Mon Sep 17 00:00:00 2001
From: Jose Abreu <Jose.Abreu@synopsys.com>
Date: Wed, 18 Dec 2019 11:24:44 +0100
Subject: net: stmmac: Let TX and RX interrupts be independently
 enabled/disabled

By using this mechanism we can get rid of the not so nice method of
scheduling TX NAPI when the RX was scheduled. No bandwidth reduction was
seen with this change.

Changes from v1:
- Remove useless comment (Jakub)
- Do not bind the TX clean to NAPI budget (Jakub)

Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c  | 24 ++++++++--
 drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h   | 11 +++--
 drivers/net/ethernet/stmicro/stmmac/dwmac4_lib.c   | 47 ++++++++++++++----
 drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h    |  6 ++-
 drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c    | 22 +++++++--
 drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h     |  2 +
 drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c | 24 ++++++++--
 drivers/net/ethernet/stmicro/stmmac/hwif.h         |  6 ++-
 drivers/net/ethernet/stmicro/stmmac/stmmac.h       |  1 +
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c  | 55 +++++++++++++---------
 10 files changed, 149 insertions(+), 49 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c
index 1c8d84ed8410..6f834302fda3 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c
@@ -335,14 +335,30 @@ static void sun8i_dwmac_dump_mac_regs(struct mac_device_info *hw,
 	}
 }
 
-static void sun8i_dwmac_enable_dma_irq(void __iomem *ioaddr, u32 chan)
+static void sun8i_dwmac_enable_dma_irq(void __iomem *ioaddr, u32 chan,
+				       bool rx, bool tx)
 {
-	writel(EMAC_RX_INT | EMAC_TX_INT, ioaddr + EMAC_INT_EN);
+	u32 value = readl(ioaddr + EMAC_INT_EN);
+
+	if (rx)
+		value |= EMAC_RX_INT;
+	if (tx)
+		value |= EMAC_TX_INT;
+
+	writel(value, ioaddr + EMAC_INT_EN);
 }
 
-static void sun8i_dwmac_disable_dma_irq(void __iomem *ioaddr, u32 chan)
+static void sun8i_dwmac_disable_dma_irq(void __iomem *ioaddr, u32 chan,
+					bool rx, bool tx)
 {
-	writel(0, ioaddr + EMAC_INT_EN);
+	u32 value = readl(ioaddr + EMAC_INT_EN);
+
+	if (rx)
+		value &= ~EMAC_RX_INT;
+	if (tx)
+		value &= ~EMAC_TX_INT;
+
+	writel(value, ioaddr + EMAC_INT_EN);
 }
 
 static void sun8i_dwmac_dma_start_tx(void __iomem *ioaddr, u32 chan)
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h b/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h
index 589931795847..bcb6d5190f3d 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h
@@ -168,6 +168,8 @@
 /* DMA default interrupt mask for 4.00 */
 #define DMA_CHAN_INTR_DEFAULT_MASK	(DMA_CHAN_INTR_NORMAL | \
 					 DMA_CHAN_INTR_ABNORMAL)
+#define DMA_CHAN_INTR_DEFAULT_RX	(DMA_CHAN_INTR_ENA_RIE)
+#define DMA_CHAN_INTR_DEFAULT_TX	(DMA_CHAN_INTR_ENA_TIE)
 
 #define DMA_CHAN_INTR_NORMAL_4_10	(DMA_CHAN_INTR_ENA_NIE_4_10 | \
 					 DMA_CHAN_INTR_ENA_RIE | \
@@ -178,6 +180,8 @@
 /* DMA default interrupt mask for 4.10a */
 #define DMA_CHAN_INTR_DEFAULT_MASK_4_10	(DMA_CHAN_INTR_NORMAL_4_10 | \
 					 DMA_CHAN_INTR_ABNORMAL_4_10)
+#define DMA_CHAN_INTR_DEFAULT_RX_4_10	(DMA_CHAN_INTR_ENA_RIE)
+#define DMA_CHAN_INTR_DEFAULT_TX_4_10	(DMA_CHAN_INTR_ENA_TIE)
 
 /* channel 0 specific fields */
 #define DMA_CHAN0_DBG_STAT_TPS		GENMASK(15, 12)
@@ -186,9 +190,10 @@
 #define DMA_CHAN0_DBG_STAT_RPS_SHIFT	8
 
 int dwmac4_dma_reset(void __iomem *ioaddr);
-void dwmac4_enable_dma_irq(void __iomem *ioaddr, u32 chan);
-void dwmac410_enable_dma_irq(void __iomem *ioaddr, u32 chan);
-void dwmac4_disable_dma_irq(void __iomem *ioaddr, u32 chan);
+void dwmac4_enable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx);
+void dwmac410_enable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx);
+void dwmac4_disable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx);
+void dwmac410_disable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx);
 void dwmac4_dma_start_tx(void __iomem *ioaddr, u32 chan);
 void dwmac4_dma_stop_tx(void __iomem *ioaddr, u32 chan);
 void dwmac4_dma_start_rx(void __iomem *ioaddr, u32 chan);
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_lib.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_lib.c
index f2a29a90e085..9becca280074 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_lib.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_lib.c
@@ -97,21 +97,52 @@ void dwmac4_set_rx_ring_len(void __iomem *ioaddr, u32 len, u32 chan)
 	writel(len, ioaddr + DMA_CHAN_RX_RING_LEN(chan));
 }
 
-void dwmac4_enable_dma_irq(void __iomem *ioaddr, u32 chan)
+void dwmac4_enable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx)
 {
-	writel(DMA_CHAN_INTR_DEFAULT_MASK, ioaddr +
-	       DMA_CHAN_INTR_ENA(chan));
+	u32 value = readl(ioaddr + DMA_CHAN_INTR_ENA(chan));
+
+	if (rx)
+		value |= DMA_CHAN_INTR_DEFAULT_RX;
+	if (tx)
+		value |= DMA_CHAN_INTR_DEFAULT_TX;
+
+	writel(value, ioaddr + DMA_CHAN_INTR_ENA(chan));
 }
 
-void dwmac410_enable_dma_irq(void __iomem *ioaddr, u32 chan)
+void dwmac410_enable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx)
 {
-	writel(DMA_CHAN_INTR_DEFAULT_MASK_4_10,
-	       ioaddr + DMA_CHAN_INTR_ENA(chan));
+	u32 value = readl(ioaddr + DMA_CHAN_INTR_ENA(chan));
+
+	if (rx)
+		value |= DMA_CHAN_INTR_DEFAULT_RX_4_10;
+	if (tx)
+		value |= DMA_CHAN_INTR_DEFAULT_TX_4_10;
+
+	writel(value, ioaddr + DMA_CHAN_INTR_ENA(chan));
 }
 
-void dwmac4_disable_dma_irq(void __iomem *ioaddr, u32 chan)
+void dwmac4_disable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx)
 {
-	writel(0, ioaddr + DMA_CHAN_INTR_ENA(chan));
+	u32 value = readl(ioaddr + DMA_CHAN_INTR_ENA(chan));
+
+	if (rx)
+		value &= ~DMA_CHAN_INTR_DEFAULT_RX;
+	if (tx)
+		value &= ~DMA_CHAN_INTR_DEFAULT_TX;
+
+	writel(value, ioaddr + DMA_CHAN_INTR_ENA(chan));
+}
+
+void dwmac410_disable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx)
+{
+	u32 value = readl(ioaddr + DMA_CHAN_INTR_ENA(chan));
+
+	if (rx)
+		value &= ~DMA_CHAN_INTR_DEFAULT_RX_4_10;
+	if (tx)
+		value &= ~DMA_CHAN_INTR_DEFAULT_TX_4_10;
+
+	writel(value, ioaddr + DMA_CHAN_INTR_ENA(chan));
 }
 
 int dwmac4_dma_interrupt(void __iomem *ioaddr,
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h b/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h
index 292b880f3f9f..e5dbd0bc257e 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h
@@ -96,6 +96,8 @@
 
 /* DMA default interrupt mask */
 #define DMA_INTR_DEFAULT_MASK	(DMA_INTR_NORMAL | DMA_INTR_ABNORMAL)
+#define DMA_INTR_DEFAULT_RX	(DMA_INTR_ENA_RIE)
+#define DMA_INTR_DEFAULT_TX	(DMA_INTR_ENA_TIE)
 
 /* DMA Status register defines */
 #define DMA_STATUS_GLPII	0x40000000	/* GMAC LPI interrupt */
@@ -130,8 +132,8 @@
 #define NUM_DWMAC1000_DMA_REGS	23
 
 void dwmac_enable_dma_transmission(void __iomem *ioaddr);
-void dwmac_enable_dma_irq(void __iomem *ioaddr, u32 chan);
-void dwmac_disable_dma_irq(void __iomem *ioaddr, u32 chan);
+void dwmac_enable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx);
+void dwmac_disable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx);
 void dwmac_dma_start_tx(void __iomem *ioaddr, u32 chan);
 void dwmac_dma_stop_tx(void __iomem *ioaddr, u32 chan);
 void dwmac_dma_start_rx(void __iomem *ioaddr, u32 chan);
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c b/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c
index 1bc25aa86dbd..688d36095333 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c
@@ -37,14 +37,28 @@ void dwmac_enable_dma_transmission(void __iomem *ioaddr)
 	writel(1, ioaddr + DMA_XMT_POLL_DEMAND);
 }
 
-void dwmac_enable_dma_irq(void __iomem *ioaddr, u32 chan)
+void dwmac_enable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx)
 {
-	writel(DMA_INTR_DEFAULT_MASK, ioaddr + DMA_INTR_ENA);
+	u32 value = readl(ioaddr + DMA_INTR_ENA);
+
+	if (rx)
+		value |= DMA_INTR_DEFAULT_RX;
+	if (tx)
+		value |= DMA_INTR_DEFAULT_TX;
+
+	writel(value, ioaddr + DMA_INTR_ENA);
 }
 
-void dwmac_disable_dma_irq(void __iomem *ioaddr, u32 chan)
+void dwmac_disable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx)
 {
-	writel(0, ioaddr + DMA_INTR_ENA);
+	u32 value = readl(ioaddr + DMA_INTR_ENA);
+
+	if (rx)
+		value &= ~DMA_INTR_DEFAULT_RX;
+	if (tx)
+		value &= ~DMA_INTR_DEFAULT_TX;
+
+	writel(value, ioaddr + DMA_INTR_ENA);
 }
 
 void dwmac_dma_start_tx(void __iomem *ioaddr, u32 chan)
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h
index 3b6e559aa0b9..158cf4ad1596 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h
+++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h
@@ -361,6 +361,8 @@
 #define XGMAC_TIE			BIT(0)
 #define XGMAC_DMA_INT_DEFAULT_EN	(XGMAC_NIE | XGMAC_AIE | XGMAC_RBUE | \
 					XGMAC_RIE | XGMAC_TIE)
+#define XGMAC_DMA_INT_DEFAULT_RX	(XGMAC_RBUE | XGMAC_RIE)
+#define XGMAC_DMA_INT_DEFAULT_TX	(XGMAC_TIE)
 #define XGMAC_DMA_CH_Rx_WATCHDOG(x)	(0x0000313c + (0x80 * (x)))
 #define XGMAC_RWT			GENMASK(7, 0)
 #define XGMAC_DMA_CH_STATUS(x)		(0x00003160 + (0x80 * (x)))
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c
index 22a7f0cc1b90..ae066f4e99a8 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c
@@ -248,14 +248,30 @@ static void dwxgmac2_dma_tx_mode(void __iomem *ioaddr, int mode,
 	writel(value, ioaddr +  XGMAC_MTL_TXQ_OPMODE(channel));
 }
 
-static void dwxgmac2_enable_dma_irq(void __iomem *ioaddr, u32 chan)
+static void dwxgmac2_enable_dma_irq(void __iomem *ioaddr, u32 chan,
+				    bool rx, bool tx)
 {
-	writel(XGMAC_DMA_INT_DEFAULT_EN, ioaddr + XGMAC_DMA_CH_INT_EN(chan));
+	u32 value = readl(ioaddr + XGMAC_DMA_CH_INT_EN(chan));
+
+	if (rx)
+		value |= XGMAC_DMA_INT_DEFAULT_RX;
+	if (tx)
+		value |= XGMAC_DMA_INT_DEFAULT_TX;
+
+	writel(value, ioaddr + XGMAC_DMA_CH_INT_EN(chan));
 }
 
-static void dwxgmac2_disable_dma_irq(void __iomem *ioaddr, u32 chan)
+static void dwxgmac2_disable_dma_irq(void __iomem *ioaddr, u32 chan,
+				     bool rx, bool tx)
 {
-	writel(0, ioaddr + XGMAC_DMA_CH_INT_EN(chan));
+	u32 value = readl(ioaddr + XGMAC_DMA_CH_INT_EN(chan));
+
+	if (rx)
+		value &= ~XGMAC_DMA_INT_DEFAULT_RX;
+	if (tx)
+		value &= ~XGMAC_DMA_INT_DEFAULT_TX;
+
+	writel(value, ioaddr + XGMAC_DMA_CH_INT_EN(chan));
 }
 
 static void dwxgmac2_dma_start_tx(void __iomem *ioaddr, u32 chan)
diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.h b/drivers/net/ethernet/stmicro/stmmac/hwif.h
index aa5b917398fe..098fbe7a5862 100644
--- a/drivers/net/ethernet/stmicro/stmmac/hwif.h
+++ b/drivers/net/ethernet/stmicro/stmmac/hwif.h
@@ -187,8 +187,10 @@ struct stmmac_dma_ops {
 	void (*dma_diagnostic_fr) (void *data, struct stmmac_extra_stats *x,
 				   void __iomem *ioaddr);
 	void (*enable_dma_transmission) (void __iomem *ioaddr);
-	void (*enable_dma_irq)(void __iomem *ioaddr, u32 chan);
-	void (*disable_dma_irq)(void __iomem *ioaddr, u32 chan);
+	void (*enable_dma_irq)(void __iomem *ioaddr, u32 chan,
+			       bool rx, bool tx);
+	void (*disable_dma_irq)(void __iomem *ioaddr, u32 chan,
+				bool rx, bool tx);
 	void (*start_tx)(void __iomem *ioaddr, u32 chan);
 	void (*stop_tx)(void __iomem *ioaddr, u32 chan);
 	void (*start_rx)(void __iomem *ioaddr, u32 chan);
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
index d993fc7e82c3..f98c5eefb382 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
@@ -88,6 +88,7 @@ struct stmmac_channel {
 	struct napi_struct rx_napi ____cacheline_aligned_in_smp;
 	struct napi_struct tx_napi ____cacheline_aligned_in_smp;
 	struct stmmac_priv *priv_data;
+	spinlock_t lock;
 	u32 index;
 };
 
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 3299f5bb02e2..d1a22c4fe67b 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -2069,17 +2069,25 @@ static int stmmac_napi_check(struct stmmac_priv *priv, u32 chan)
 	int status = stmmac_dma_interrupt_status(priv, priv->ioaddr,
 						 &priv->xstats, chan);
 	struct stmmac_channel *ch = &priv->channel[chan];
+	unsigned long flags;
 
 	if ((status & handle_rx) && (chan < priv->plat->rx_queues_to_use)) {
 		if (napi_schedule_prep(&ch->rx_napi)) {
-			stmmac_disable_dma_irq(priv, priv->ioaddr, chan);
+			spin_lock_irqsave(&ch->lock, flags);
+			stmmac_disable_dma_irq(priv, priv->ioaddr, chan, 1, 0);
+			spin_unlock_irqrestore(&ch->lock, flags);
 			__napi_schedule_irqoff(&ch->rx_napi);
-			status |= handle_tx;
 		}
 	}
 
-	if ((status & handle_tx) && (chan < priv->plat->tx_queues_to_use))
-		napi_schedule_irqoff(&ch->tx_napi);
+	if ((status & handle_tx) && (chan < priv->plat->tx_queues_to_use)) {
+		if (napi_schedule_prep(&ch->tx_napi)) {
+			spin_lock_irqsave(&ch->lock, flags);
+			stmmac_disable_dma_irq(priv, priv->ioaddr, chan, 0, 1);
+			spin_unlock_irqrestore(&ch->lock, flags);
+			__napi_schedule_irqoff(&ch->tx_napi);
+		}
+	}
 
 	return status;
 }
@@ -2274,14 +2282,14 @@ static void stmmac_tx_timer(struct timer_list *t)
 
 	ch = &priv->channel[tx_q->queue_index];
 
-	/*
-	 * If NAPI is already running we can miss some events. Let's rearm
-	 * the timer and try again.
-	 */
-	if (likely(napi_schedule_prep(&ch->tx_napi)))
+	if (likely(napi_schedule_prep(&ch->tx_napi))) {
+		unsigned long flags;
+
+		spin_lock_irqsave(&ch->lock, flags);
+		stmmac_disable_dma_irq(priv, priv->ioaddr, ch->index, 0, 1);
+		spin_unlock_irqrestore(&ch->lock, flags);
 		__napi_schedule(&ch->tx_napi);
-	else
-		mod_timer(&tx_q->txtimer, STMMAC_COAL_TIMER(10));
+	}
 }
 
 /**
@@ -3751,8 +3759,14 @@ static int stmmac_napi_poll_rx(struct napi_struct *napi, int budget)
 	priv->xstats.napi_poll++;
 
 	work_done = stmmac_rx(priv, budget, chan);
-	if (work_done < budget && napi_complete_done(napi, work_done))
-		stmmac_enable_dma_irq(priv, priv->ioaddr, chan);
+	if (work_done < budget && napi_complete_done(napi, work_done)) {
+		unsigned long flags;
+
+		spin_lock_irqsave(&ch->lock, flags);
+		stmmac_enable_dma_irq(priv, priv->ioaddr, chan, 1, 0);
+		spin_unlock_irqrestore(&ch->lock, flags);
+	}
+
 	return work_done;
 }
 
@@ -3761,7 +3775,6 @@ static int stmmac_napi_poll_tx(struct napi_struct *napi, int budget)
 	struct stmmac_channel *ch =
 		container_of(napi, struct stmmac_channel, tx_napi);
 	struct stmmac_priv *priv = ch->priv_data;
-	struct stmmac_tx_queue *tx_q;
 	u32 chan = ch->index;
 	int work_done;
 
@@ -3770,15 +3783,12 @@ static int stmmac_napi_poll_tx(struct napi_struct *napi, int budget)
 	work_done = stmmac_tx_clean(priv, DMA_TX_SIZE, chan);
 	work_done = min(work_done, budget);
 
-	if (work_done < budget)
-		napi_complete_done(napi, work_done);
+	if (work_done < budget && napi_complete_done(napi, work_done)) {
+		unsigned long flags;
 
-	/* Force transmission restart */
-	tx_q = &priv->tx_queue[chan];
-	if (tx_q->cur_tx != tx_q->dirty_tx) {
-		stmmac_enable_dma_transmission(priv, priv->ioaddr);
-		stmmac_set_tx_tail_ptr(priv, priv->ioaddr, tx_q->tx_tail_addr,
-				       chan);
+		spin_lock_irqsave(&ch->lock, flags);
+		stmmac_enable_dma_irq(priv, priv->ioaddr, chan, 0, 1);
+		spin_unlock_irqrestore(&ch->lock, flags);
 	}
 
 	return work_done;
@@ -4714,6 +4724,7 @@ int stmmac_dvr_probe(struct device *device,
 	for (queue = 0; queue < maxq; queue++) {
 		struct stmmac_channel *ch = &priv->channel[queue];
 
+		spin_lock_init(&ch->lock);
 		ch->priv_data = priv;
 		ch->index = queue;
 
-- 
cgit v1.2.3-59-g8ed1b


From 3755b21b04b096347605d17008c6b6a8cf3f59e4 Mon Sep 17 00:00:00 2001
From: Jose Abreu <Jose.Abreu@synopsys.com>
Date: Wed, 18 Dec 2019 11:24:45 +0100
Subject: net: stmmac: Always use TX coalesce timer value when rescheduling

When we have pending packets we re-arm the TX timer with a magic value.

This changes the re-arm of the timer from 10us to the user-defined
coalesce value. As we support different speeds, having a magic value of
10us can be either too short or to large depending on the speed so we
let user configure it. The default value of the timer is 1ms but it can
be reconfigured by ethtool.

Changes from v1:
- Reword commit message (Jakub)

Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index d1a22c4fe67b..ecb89c609fb2 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -1975,7 +1975,7 @@ static int stmmac_tx_clean(struct stmmac_priv *priv, int budget, u32 queue)
 
 	/* We still have pending packets, let's call for a new scheduling */
 	if (tx_q->dirty_tx != tx_q->cur_tx)
-		mod_timer(&tx_q->txtimer, STMMAC_COAL_TIMER(10));
+		mod_timer(&tx_q->txtimer, STMMAC_COAL_TIMER(priv->tx_coal_timer));
 
 	__netif_tx_unlock_bh(netdev_get_tx_queue(priv->dev, queue));
 
-- 
cgit v1.2.3-59-g8ed1b


From 504723af0d85434be5fb6f2dde0b62644a7f1ead Mon Sep 17 00:00:00 2001
From: Jose Abreu <joabreu@synopsys.com>
Date: Wed, 18 Dec 2019 11:33:05 +0100
Subject: net: stmmac: Add basic EST support for GMAC5+

Adds the support for EST in GMAC5+ cores. This feature allows to offload
scheduling of queues opening time to the IP.

Signed-off-by: Jose Abreu <joabreu@synopsys.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/stmicro/stmmac/common.h      |  4 +
 drivers/net/ethernet/stmicro/stmmac/dwmac4.h      |  9 +++
 drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c |  2 +
 drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c  |  3 +
 drivers/net/ethernet/stmicro/stmmac/dwmac5.c      | 95 +++++++++++++++++++++++
 drivers/net/ethernet/stmicro/stmmac/dwmac5.h      | 19 +++++
 drivers/net/ethernet/stmicro/stmmac/hwif.h        |  5 ++
 include/linux/stmmac.h                            | 13 ++++
 8 files changed, 150 insertions(+)

diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h
index b210e987a1db..2f1fc93a9b1e 100644
--- a/drivers/net/ethernet/stmicro/stmmac/common.h
+++ b/drivers/net/ethernet/stmicro/stmmac/common.h
@@ -363,6 +363,10 @@ struct dma_features {
 	unsigned int dvlan;
 	unsigned int l3l4fnum;
 	unsigned int arpoffsel;
+	/* TSN Features */
+	unsigned int estwid;
+	unsigned int estdep;
+	unsigned int estsel;
 };
 
 /* GMAC TX FIFO is 8K, Rx FIFO is 16K */
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4.h b/drivers/net/ethernet/stmicro/stmmac/dwmac4.h
index 2dc70d104161..f7f5a2751147 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac4.h
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4.h
@@ -176,6 +176,8 @@ enum power_event {
 #define GMAC_CONFIG_SARC		GENMASK(30, 28)
 #define GMAC_CONFIG_SARC_SHIFT		28
 #define GMAC_CONFIG_IPC			BIT(27)
+#define GMAC_CONFIG_IPG			GENMASK(26, 24)
+#define GMAC_CONFIG_IPG_SHIFT		24
 #define GMAC_CONFIG_2K			BIT(22)
 #define GMAC_CONFIG_ACS			BIT(20)
 #define GMAC_CONFIG_BE			BIT(18)
@@ -183,6 +185,7 @@ enum power_event {
 #define GMAC_CONFIG_JE			BIT(16)
 #define GMAC_CONFIG_PS			BIT(15)
 #define GMAC_CONFIG_FES			BIT(14)
+#define GMAC_CONFIG_FES_SHIFT		14
 #define GMAC_CONFIG_DM			BIT(13)
 #define GMAC_CONFIG_LM			BIT(12)
 #define GMAC_CONFIG_DCRS		BIT(9)
@@ -190,6 +193,9 @@ enum power_event {
 #define GMAC_CONFIG_RE			BIT(0)
 
 /* MAC extended config */
+#define GMAC_CONFIG_EIPG		GENMASK(29, 25)
+#define GMAC_CONFIG_EIPG_SHIFT		25
+#define GMAC_CONFIG_EIPG_EN		BIT(24)
 #define GMAC_CONFIG_HDSMS		GENMASK(22, 20)
 #define GMAC_CONFIG_HDSMS_SHIFT		20
 #define GMAC_CONFIG_HDSMS_256		(0x2 << GMAC_CONFIG_HDSMS_SHIFT)
@@ -231,6 +237,9 @@ enum power_event {
 
 /* MAC HW features3 bitmap */
 #define GMAC_HW_FEAT_ASP		GENMASK(29, 28)
+#define GMAC_HW_FEAT_ESTWID		GENMASK(21, 20)
+#define GMAC_HW_FEAT_ESTDEP		GENMASK(19, 17)
+#define GMAC_HW_FEAT_ESTSEL		BIT(16)
 #define GMAC_HW_FEAT_FRPES		GENMASK(14, 13)
 #define GMAC_HW_FEAT_FRPBS		GENMASK(12, 11)
 #define GMAC_HW_FEAT_FRPSEL		BIT(10)
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c
index 40ca00e596dd..8df7496411a0 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c
@@ -984,6 +984,7 @@ const struct stmmac_ops dwmac410_ops = {
 	.set_arp_offload = dwmac4_set_arp_offload,
 	.config_l3_filter = dwmac4_config_l3_filter,
 	.config_l4_filter = dwmac4_config_l4_filter,
+	.est_configure = dwmac5_est_configure,
 };
 
 const struct stmmac_ops dwmac510_ops = {
@@ -1027,6 +1028,7 @@ const struct stmmac_ops dwmac510_ops = {
 	.set_arp_offload = dwmac4_set_arp_offload,
 	.config_l3_filter = dwmac4_config_l3_filter,
 	.config_l4_filter = dwmac4_config_l4_filter,
+	.est_configure = dwmac5_est_configure,
 };
 
 int dwmac4_setup(struct stmmac_priv *priv)
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c
index c15409030710..3552ce1201ff 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c
@@ -404,6 +404,9 @@ static void dwmac4_get_hw_feature(void __iomem *ioaddr,
 
 	/* 5.10 Features */
 	dma_cap->asp = (hw_cap & GMAC_HW_FEAT_ASP) >> 28;
+	dma_cap->estwid = (hw_cap & GMAC_HW_FEAT_ESTWID) >> 20;
+	dma_cap->estdep = (hw_cap & GMAC_HW_FEAT_ESTDEP) >> 17;
+	dma_cap->estsel = (hw_cap & GMAC_HW_FEAT_ESTSEL) >> 16;
 	dma_cap->frpes = (hw_cap & GMAC_HW_FEAT_FRPES) >> 13;
 	dma_cap->frpbs = (hw_cap & GMAC_HW_FEAT_FRPBS) >> 11;
 	dma_cap->frpsel = (hw_cap & GMAC_HW_FEAT_FRPSEL) >> 10;
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac5.c b/drivers/net/ethernet/stmicro/stmmac/dwmac5.c
index e436fa160c7d..8047f402fb3f 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac5.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac5.c
@@ -550,3 +550,98 @@ int dwmac5_flex_pps_config(void __iomem *ioaddr, int index,
 	writel(val, ioaddr + MAC_PPS_CONTROL);
 	return 0;
 }
+
+static int dwmac5_est_write(void __iomem *ioaddr, u32 reg, u32 val, bool gcl)
+{
+	u32 ctrl;
+
+	writel(val, ioaddr + MTL_EST_GCL_DATA);
+
+	ctrl = (reg << ADDR_SHIFT);
+	ctrl |= gcl ? 0 : GCRR;
+
+	writel(ctrl, ioaddr + MTL_EST_GCL_CONTROL);
+
+	ctrl |= SRWO;
+	writel(ctrl, ioaddr + MTL_EST_GCL_CONTROL);
+
+	return readl_poll_timeout(ioaddr + MTL_EST_GCL_CONTROL,
+				  ctrl, !(ctrl & SRWO), 100, 5000);
+}
+
+int dwmac5_est_configure(void __iomem *ioaddr, struct stmmac_est *cfg,
+			 unsigned int ptp_rate)
+{
+	u32 speed, total_offset, offset, ctrl, ctr_low;
+	u32 extcfg = readl(ioaddr + GMAC_EXT_CONFIG);
+	u32 mac_cfg = readl(ioaddr + GMAC_CONFIG);
+	int i, ret = 0x0;
+	u64 total_ctr;
+
+	if (extcfg & GMAC_CONFIG_EIPG_EN) {
+		offset = (extcfg & GMAC_CONFIG_EIPG) >> GMAC_CONFIG_EIPG_SHIFT;
+		offset = 104 + (offset * 8);
+	} else {
+		offset = (mac_cfg & GMAC_CONFIG_IPG) >> GMAC_CONFIG_IPG_SHIFT;
+		offset = 96 - (offset * 8);
+	}
+
+	speed = mac_cfg & (GMAC_CONFIG_PS | GMAC_CONFIG_FES);
+	speed = speed >> GMAC_CONFIG_FES_SHIFT;
+
+	switch (speed) {
+	case 0x0:
+		offset = offset * 1000; /* 1G */
+		break;
+	case 0x1:
+		offset = offset * 400; /* 2.5G */
+		break;
+	case 0x2:
+		offset = offset * 100000; /* 10M */
+		break;
+	case 0x3:
+		offset = offset * 10000; /* 100M */
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	offset = offset / 1000;
+
+	ret |= dwmac5_est_write(ioaddr, BTR_LOW, cfg->btr[0], false);
+	ret |= dwmac5_est_write(ioaddr, BTR_HIGH, cfg->btr[1], false);
+	ret |= dwmac5_est_write(ioaddr, TER, cfg->ter, false);
+	ret |= dwmac5_est_write(ioaddr, LLR, cfg->gcl_size, false);
+	if (ret)
+		return ret;
+
+	total_offset = 0;
+	for (i = 0; i < cfg->gcl_size; i++) {
+		ret = dwmac5_est_write(ioaddr, i, cfg->gcl[i] + offset, true);
+		if (ret)
+			return ret;
+
+		total_offset += offset;
+	}
+
+	total_ctr = cfg->ctr[0] + cfg->ctr[1] * 1000000000;
+	total_ctr += total_offset;
+
+	ctr_low = do_div(total_ctr, 1000000000);
+
+	ret |= dwmac5_est_write(ioaddr, CTR_LOW, ctr_low, false);
+	ret |= dwmac5_est_write(ioaddr, CTR_HIGH, total_ctr, false);
+	if (ret)
+		return ret;
+
+	ctrl = readl(ioaddr + MTL_EST_CONTROL);
+	ctrl &= ~PTOV;
+	ctrl |= ((1000000000 / ptp_rate) * 6) << PTOV_SHIFT;
+	if (cfg->enable)
+		ctrl |= EEST | SSWL;
+	else
+		ctrl &= ~EEST;
+
+	writel(ctrl, ioaddr + MTL_EST_CONTROL);
+	return 0;
+}
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac5.h b/drivers/net/ethernet/stmicro/stmmac/dwmac5.h
index 23fecf68f781..70e6d8837dd9 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac5.h
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac5.h
@@ -30,6 +30,23 @@
 #define MAC_PPSx_INTERVAL(x)		(0x00000b88 + ((x) * 0x10))
 #define MAC_PPSx_WIDTH(x)		(0x00000b8c + ((x) * 0x10))
 
+#define MTL_EST_CONTROL			0x00000c50
+#define PTOV				GENMASK(31, 24)
+#define PTOV_SHIFT			24
+#define SSWL				BIT(1)
+#define EEST				BIT(0)
+#define MTL_EST_GCL_CONTROL		0x00000c80
+#define BTR_LOW				0x0
+#define BTR_HIGH			0x1
+#define CTR_LOW				0x2
+#define CTR_HIGH			0x3
+#define TER				0x4
+#define LLR				0x5
+#define ADDR_SHIFT			8
+#define GCRR				BIT(2)
+#define SRWO				BIT(0)
+#define MTL_EST_GCL_DATA		0x00000c84
+
 #define MTL_RXP_CONTROL_STATUS		0x00000ca0
 #define RXPI				BIT(31)
 #define NPE				GENMASK(23, 16)
@@ -83,5 +100,7 @@ int dwmac5_rxp_config(void __iomem *ioaddr, struct stmmac_tc_entry *entries,
 int dwmac5_flex_pps_config(void __iomem *ioaddr, int index,
 			   struct stmmac_pps_cfg *cfg, bool enable,
 			   u32 sub_second_inc, u32 systime_flags);
+int dwmac5_est_configure(void __iomem *ioaddr, struct stmmac_est *cfg,
+			 unsigned int ptp_rate);
 
 #endif /* __DWMAC5_H__ */
diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.h b/drivers/net/ethernet/stmicro/stmmac/hwif.h
index 098fbe7a5862..9fd34df6eb16 100644
--- a/drivers/net/ethernet/stmicro/stmmac/hwif.h
+++ b/drivers/net/ethernet/stmicro/stmmac/hwif.h
@@ -276,6 +276,7 @@ struct stmmac_safety_stats;
 struct stmmac_tc_entry;
 struct stmmac_pps_cfg;
 struct stmmac_rss;
+struct stmmac_est;
 
 /* Helpers to program the MAC core */
 struct stmmac_ops {
@@ -373,6 +374,8 @@ struct stmmac_ops {
 				bool en, bool udp, bool sa, bool inv,
 				u32 match);
 	void (*set_arp_offload)(struct mac_device_info *hw, bool en, u32 addr);
+	int (*est_configure)(void __iomem *ioaddr, struct stmmac_est *cfg,
+			     unsigned int ptp_rate);
 };
 
 #define stmmac_core_init(__priv, __args...) \
@@ -459,6 +462,8 @@ struct stmmac_ops {
 	stmmac_do_callback(__priv, mac, config_l4_filter, __args)
 #define stmmac_set_arp_offload(__priv, __args...) \
 	stmmac_do_void_callback(__priv, mac, set_arp_offload, __args)
+#define stmmac_est_configure(__priv, __args...) \
+	stmmac_do_callback(__priv, mac, est_configure, __args)
 
 /* PTP and HW Timer helpers */
 struct stmmac_hwtimestamp {
diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h
index d4bcd9387136..0531afa9b21e 100644
--- a/include/linux/stmmac.h
+++ b/include/linux/stmmac.h
@@ -109,6 +109,18 @@ struct stmmac_axi {
 	bool axi_rb;
 };
 
+#define EST_GCL		1024
+struct stmmac_est {
+	int enable;
+	u32 btr_offset[2];
+	u32 btr[2];
+	u32 ctr[2];
+	u32 ter;
+	u32 gcl_unaligned[EST_GCL];
+	u32 gcl[EST_GCL];
+	u32 gcl_size;
+};
+
 struct stmmac_rxq_cfg {
 	u8 mode_to_use;
 	u32 chan;
@@ -139,6 +151,7 @@ struct plat_stmmacenet_data {
 	struct device_node *phylink_node;
 	struct device_node *mdio_node;
 	struct stmmac_dma_cfg *dma_cfg;
+	struct stmmac_est *est;
 	int clk_csr;
 	int has_gmac;
 	int enh_desc;
-- 
cgit v1.2.3-59-g8ed1b


From 8572aec3d0dc43045254fd1bf581fb980bfdbc4b Mon Sep 17 00:00:00 2001
From: Jose Abreu <Jose.Abreu@synopsys.com>
Date: Wed, 18 Dec 2019 11:33:06 +0100
Subject: net: stmmac: Add basic EST support for XGMAC

Adds the support for EST in XGMAC cores. This feature allows to offload
scheduling of queues opening time to the IP.

Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h     | 19 ++++++++
 .../net/ethernet/stmicro/stmmac/dwxgmac2_core.c    | 52 ++++++++++++++++++++++
 drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c |  3 ++
 3 files changed, 74 insertions(+)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h
index 158cf4ad1596..e1e79c90c865 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h
+++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h
@@ -136,6 +136,9 @@
 #define XGMAC_HWFEAT_TXQCNT		GENMASK(9, 6)
 #define XGMAC_HWFEAT_RXQCNT		GENMASK(3, 0)
 #define XGMAC_HW_FEATURE3		0x00000128
+#define XGMAC_HWFEAT_ESTWID		GENMASK(24, 23)
+#define XGMAC_HWFEAT_ESTDEP		GENMASK(22, 20)
+#define XGMAC_HWFEAT_ESTSEL		BIT(19)
 #define XGMAC_HWFEAT_ASP		GENMASK(15, 14)
 #define XGMAC_HWFEAT_DVLAN		BIT(13)
 #define XGMAC_HWFEAT_FRPES		GENMASK(12, 11)
@@ -237,6 +240,22 @@
 #define XGMAC_TC_PRTY_MAP1		0x00001044
 #define XGMAC_PSTC(x)			GENMASK((x) * 8 + 7, (x) * 8)
 #define XGMAC_PSTC_SHIFT(x)		((x) * 8)
+#define XGMAC_MTL_EST_CONTROL		0x00001050
+#define XGMAC_PTOV			GENMASK(31, 23)
+#define XGMAC_PTOV_SHIFT		23
+#define XGMAC_SSWL			BIT(1)
+#define XGMAC_EEST			BIT(0)
+#define XGMAC_MTL_EST_GCL_CONTROL	0x00001080
+#define XGMAC_BTR_LOW			0x0
+#define XGMAC_BTR_HIGH			0x1
+#define XGMAC_CTR_LOW			0x2
+#define XGMAC_CTR_HIGH			0x3
+#define XGMAC_TER			0x4
+#define XGMAC_LLR			0x5
+#define XGMAC_ADDR_SHIFT		8
+#define XGMAC_GCRR			BIT(2)
+#define XGMAC_SRWO			BIT(0)
+#define XGMAC_MTL_EST_GCL_DATA		0x00001084
 #define XGMAC_MTL_RXP_CONTROL_STATUS	0x000010a0
 #define XGMAC_RXPI			BIT(31)
 #define XGMAC_NPE			GENMASK(23, 16)
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c
index 082f5ee9e525..2f6e960947d9 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c
@@ -1359,6 +1359,57 @@ static void dwxgmac2_set_arp_offload(struct mac_device_info *hw, bool en,
 	writel(value, ioaddr + XGMAC_RX_CONFIG);
 }
 
+static int dwxgmac3_est_write(void __iomem *ioaddr, u32 reg, u32 val, bool gcl)
+{
+	u32 ctrl;
+
+	writel(val, ioaddr + XGMAC_MTL_EST_GCL_DATA);
+
+	ctrl = (reg << XGMAC_ADDR_SHIFT);
+	ctrl |= gcl ? 0 : XGMAC_GCRR;
+
+	writel(ctrl, ioaddr + XGMAC_MTL_EST_GCL_CONTROL);
+
+	ctrl |= XGMAC_SRWO;
+	writel(ctrl, ioaddr + XGMAC_MTL_EST_GCL_CONTROL);
+
+	return readl_poll_timeout_atomic(ioaddr + XGMAC_MTL_EST_GCL_CONTROL,
+					 ctrl, !(ctrl & XGMAC_SRWO), 100, 5000);
+}
+
+static int dwxgmac3_est_configure(void __iomem *ioaddr, struct stmmac_est *cfg,
+				  unsigned int ptp_rate)
+{
+	int i, ret = 0x0;
+	u32 ctrl;
+
+	ret |= dwxgmac3_est_write(ioaddr, XGMAC_BTR_LOW, cfg->btr[0], false);
+	ret |= dwxgmac3_est_write(ioaddr, XGMAC_BTR_HIGH, cfg->btr[1], false);
+	ret |= dwxgmac3_est_write(ioaddr, XGMAC_TER, cfg->ter, false);
+	ret |= dwxgmac3_est_write(ioaddr, XGMAC_LLR, cfg->gcl_size, false);
+	ret |= dwxgmac3_est_write(ioaddr, XGMAC_CTR_LOW, cfg->ctr[0], false);
+	ret |= dwxgmac3_est_write(ioaddr, XGMAC_CTR_HIGH, cfg->ctr[1], false);
+	if (ret)
+		return ret;
+
+	for (i = 0; i < cfg->gcl_size; i++) {
+		ret = dwxgmac3_est_write(ioaddr, i, cfg->gcl[i], true);
+		if (ret)
+			return ret;
+	}
+
+	ctrl = readl(ioaddr + XGMAC_MTL_EST_CONTROL);
+	ctrl &= ~XGMAC_PTOV;
+	ctrl |= ((1000000000 / ptp_rate) * 9) << XGMAC_PTOV_SHIFT;
+	if (cfg->enable)
+		ctrl |= XGMAC_EEST | XGMAC_SSWL;
+	else
+		ctrl &= ~XGMAC_EEST;
+
+	writel(ctrl, ioaddr + XGMAC_MTL_EST_CONTROL);
+	return 0;
+}
+
 const struct stmmac_ops dwxgmac210_ops = {
 	.core_init = dwxgmac2_core_init,
 	.set_mac = dwxgmac2_set_mac,
@@ -1402,6 +1453,7 @@ const struct stmmac_ops dwxgmac210_ops = {
 	.config_l3_filter = dwxgmac2_config_l3_filter,
 	.config_l4_filter = dwxgmac2_config_l4_filter,
 	.set_arp_offload = dwxgmac2_set_arp_offload,
+	.est_configure = dwxgmac3_est_configure,
 };
 
 int dwxgmac2_setup(struct stmmac_priv *priv)
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c
index ae066f4e99a8..6ff2795a5117 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c
@@ -429,6 +429,9 @@ static void dwxgmac2_get_hw_feature(void __iomem *ioaddr,
 
 	/* MAC HW feature 3 */
 	hw_cap = readl(ioaddr + XGMAC_HW_FEATURE3);
+	dma_cap->estwid = (hw_cap & XGMAC_HWFEAT_ESTWID) >> 23;
+	dma_cap->estdep = (hw_cap & XGMAC_HWFEAT_ESTDEP) >> 20;
+	dma_cap->estsel = (hw_cap & XGMAC_HWFEAT_ESTSEL) >> 19;
 	dma_cap->asp = (hw_cap & XGMAC_HWFEAT_ASP) >> 14;
 	dma_cap->dvlan = (hw_cap & XGMAC_HWFEAT_DVLAN) >> 13;
 	dma_cap->frpes = (hw_cap & XGMAC_HWFEAT_FRPES) >> 11;
-- 
cgit v1.2.3-59-g8ed1b


From b60189e0392fa06348911077ef281eb2b1047b6a Mon Sep 17 00:00:00 2001
From: Jose Abreu <Jose.Abreu@synopsys.com>
Date: Wed, 18 Dec 2019 11:33:07 +0100
Subject: net: stmmac: Integrate EST with TAPRIO scheduler API

Now that we have the EST code for XGMAC and QoS we can use it with the
TAPRIO scheduler. Integrate it into the main driver and use the API to
configure the EST feature.

Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/stmicro/stmmac/hwif.h        |   5 +
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c |   2 +
 drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c   | 108 ++++++++++++++++++++++
 3 files changed, 115 insertions(+)

diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.h b/drivers/net/ethernet/stmicro/stmmac/hwif.h
index 9fd34df6eb16..22f79c0a9343 100644
--- a/drivers/net/ethernet/stmicro/stmmac/hwif.h
+++ b/drivers/net/ethernet/stmicro/stmmac/hwif.h
@@ -521,6 +521,7 @@ struct stmmac_priv;
 struct tc_cls_u32_offload;
 struct tc_cbs_qopt_offload;
 struct flow_cls_offload;
+struct tc_taprio_qopt_offload;
 
 struct stmmac_tc_ops {
 	int (*init)(struct stmmac_priv *priv);
@@ -530,6 +531,8 @@ struct stmmac_tc_ops {
 			 struct tc_cbs_qopt_offload *qopt);
 	int (*setup_cls)(struct stmmac_priv *priv,
 			 struct flow_cls_offload *cls);
+	int (*setup_taprio)(struct stmmac_priv *priv,
+			    struct tc_taprio_qopt_offload *qopt);
 };
 
 #define stmmac_tc_init(__priv, __args...) \
@@ -540,6 +543,8 @@ struct stmmac_tc_ops {
 	stmmac_do_callback(__priv, tc, setup_cbs, __args)
 #define stmmac_tc_setup_cls(__priv, __args...) \
 	stmmac_do_callback(__priv, tc, setup_cls, __args)
+#define stmmac_tc_setup_taprio(__priv, __args...) \
+	stmmac_do_callback(__priv, tc, setup_taprio, __args)
 
 struct stmmac_counters;
 
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index ecb89c609fb2..18a959589cbc 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -4076,6 +4076,8 @@ static int stmmac_setup_tc(struct net_device *ndev, enum tc_setup_type type,
 						  priv, priv, true);
 	case TC_SETUP_QDISC_CBS:
 		return stmmac_tc_setup_cbs(priv, priv, type_data);
+	case TC_SETUP_QDISC_TAPRIO:
+		return stmmac_tc_setup_taprio(priv, priv, type_data);
 	default:
 		return -EOPNOTSUPP;
 	}
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c
index 7d972e0fd2b0..58d4ce094381 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c
@@ -591,9 +591,117 @@ static int tc_setup_cls(struct stmmac_priv *priv,
 	return ret;
 }
 
+static int tc_setup_taprio(struct stmmac_priv *priv,
+			   struct tc_taprio_qopt_offload *qopt)
+{
+	u32 size, wid = priv->dma_cap.estwid, dep = priv->dma_cap.estdep;
+	struct plat_stmmacenet_data *plat = priv->plat;
+	struct timespec64 time;
+	int i, ret = 0;
+
+	if (!priv->dma_cap.estsel)
+		return -EOPNOTSUPP;
+
+	switch (wid) {
+	case 0x1:
+		wid = 16;
+		break;
+	case 0x2:
+		wid = 20;
+		break;
+	case 0x3:
+		wid = 24;
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	switch (dep) {
+	case 0x1:
+		dep = 64;
+		break;
+	case 0x2:
+		dep = 128;
+		break;
+	case 0x3:
+		dep = 256;
+		break;
+	case 0x4:
+		dep = 512;
+		break;
+	case 0x5:
+		dep = 1024;
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	if (!qopt->enable)
+		goto disable;
+	if (qopt->num_entries >= dep)
+		return -EINVAL;
+	if (!qopt->base_time)
+		return -ERANGE;
+	if (!qopt->cycle_time)
+		return -ERANGE;
+
+	if (!plat->est) {
+		plat->est = devm_kzalloc(priv->device, sizeof(*plat->est),
+					 GFP_KERNEL);
+		if (!plat->est)
+			return -ENOMEM;
+	} else {
+		memset(plat->est, 0, sizeof(*plat->est));
+	}
+
+	size = qopt->num_entries;
+
+	priv->plat->est->gcl_size = size;
+	priv->plat->est->enable = qopt->enable;
+
+	for (i = 0; i < size; i++) {
+		s64 delta_ns = qopt->entries[i].interval;
+		u32 gates = qopt->entries[i].gate_mask;
+
+		if (delta_ns > GENMASK(wid, 0))
+			return -ERANGE;
+		if (gates > GENMASK(31 - wid, 0))
+			return -ERANGE;
+		if (qopt->entries[i].command != TC_TAPRIO_CMD_SET_GATES)
+			return -EOPNOTSUPP;
+
+		priv->plat->est->gcl[i] = delta_ns | (gates << wid);
+	}
+
+	/* Adjust for real system time */
+	time = ktime_to_timespec64(qopt->base_time);
+	priv->plat->est->btr[0] = (u32)time.tv_nsec;
+	priv->plat->est->btr[1] = (u32)time.tv_sec;
+
+	priv->plat->est->ctr[0] = (u32)(qopt->cycle_time % NSEC_PER_SEC);
+	priv->plat->est->ctr[1] = (u32)(qopt->cycle_time / NSEC_PER_SEC);
+
+	ret = stmmac_est_configure(priv, priv->ioaddr, priv->plat->est,
+				   priv->plat->clk_ptp_rate);
+	if (ret) {
+		netdev_err(priv->dev, "failed to configure EST\n");
+		goto disable;
+	}
+
+	netdev_info(priv->dev, "configured EST\n");
+	return 0;
+
+disable:
+	priv->plat->est->enable = false;
+	stmmac_est_configure(priv, priv->ioaddr, priv->plat->est,
+			     priv->plat->clk_ptp_rate);
+	return ret;
+}
+
 const struct stmmac_tc_ops dwmac510_tc_ops = {
 	.init = tc_init,
 	.setup_cls_u32 = tc_setup_cls_u32,
 	.setup_cbs = tc_setup_cbs,
 	.setup_cls = tc_setup_cls,
+	.setup_taprio = tc_setup_taprio,
 };
-- 
cgit v1.2.3-59-g8ed1b


From 1ac14241543c0108e9fd23aca4927a148a44c815 Mon Sep 17 00:00:00 2001
From: Jose Abreu <Jose.Abreu@synopsys.com>
Date: Wed, 18 Dec 2019 11:33:08 +0100
Subject: net: stmmac: Add Frame Preemption support using TAPRIO API

Adds the support for Frame Preemption using TAPRIO API. This works along
with EST feature and allows to select if preemptable traffic shall be
sent during specific queues opening time.

Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/stmicro/stmmac/common.h    |  1 +
 drivers/net/ethernet/stmicro/stmmac/hwif.h      |  4 ++++
 drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c | 29 ++++++++++++++++++++++++-
 3 files changed, 33 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h
index 2f1fc93a9b1e..09c72025ed3e 100644
--- a/drivers/net/ethernet/stmicro/stmmac/common.h
+++ b/drivers/net/ethernet/stmicro/stmmac/common.h
@@ -367,6 +367,7 @@ struct dma_features {
 	unsigned int estwid;
 	unsigned int estdep;
 	unsigned int estsel;
+	unsigned int fpesel;
 };
 
 /* GMAC TX FIFO is 8K, Rx FIFO is 16K */
diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.h b/drivers/net/ethernet/stmicro/stmmac/hwif.h
index 22f79c0a9343..905a6f0edaca 100644
--- a/drivers/net/ethernet/stmicro/stmmac/hwif.h
+++ b/drivers/net/ethernet/stmicro/stmmac/hwif.h
@@ -376,6 +376,8 @@ struct stmmac_ops {
 	void (*set_arp_offload)(struct mac_device_info *hw, bool en, u32 addr);
 	int (*est_configure)(void __iomem *ioaddr, struct stmmac_est *cfg,
 			     unsigned int ptp_rate);
+	void (*fpe_configure)(void __iomem *ioaddr, u32 num_txq, u32 num_rxq,
+			      bool enable);
 };
 
 #define stmmac_core_init(__priv, __args...) \
@@ -464,6 +466,8 @@ struct stmmac_ops {
 	stmmac_do_void_callback(__priv, mac, set_arp_offload, __args)
 #define stmmac_est_configure(__priv, __args...) \
 	stmmac_do_callback(__priv, mac, est_configure, __args)
+#define stmmac_fpe_configure(__priv, __args...) \
+	stmmac_do_void_callback(__priv, mac, fpe_configure, __args)
 
 /* PTP and HW Timer helpers */
 struct stmmac_hwtimestamp {
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c
index 58d4ce094381..8ff8f9b9bb22 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c
@@ -597,6 +597,7 @@ static int tc_setup_taprio(struct stmmac_priv *priv,
 	u32 size, wid = priv->dma_cap.estwid, dep = priv->dma_cap.estdep;
 	struct plat_stmmacenet_data *plat = priv->plat;
 	struct timespec64 time;
+	bool fpe = false;
 	int i, ret = 0;
 
 	if (!priv->dma_cap.estsel)
@@ -667,8 +668,23 @@ static int tc_setup_taprio(struct stmmac_priv *priv,
 			return -ERANGE;
 		if (gates > GENMASK(31 - wid, 0))
 			return -ERANGE;
-		if (qopt->entries[i].command != TC_TAPRIO_CMD_SET_GATES)
+
+		switch (qopt->entries[i].command) {
+		case TC_TAPRIO_CMD_SET_GATES:
+			if (fpe)
+				return -EINVAL;
+			break;
+		case TC_TAPRIO_CMD_SET_AND_HOLD:
+			gates |= BIT(0);
+			fpe = true;
+			break;
+		case TC_TAPRIO_CMD_SET_AND_RELEASE:
+			gates &= ~BIT(0);
+			fpe = true;
+			break;
+		default:
 			return -EOPNOTSUPP;
+		}
 
 		priv->plat->est->gcl[i] = delta_ns | (gates << wid);
 	}
@@ -681,6 +697,17 @@ static int tc_setup_taprio(struct stmmac_priv *priv,
 	priv->plat->est->ctr[0] = (u32)(qopt->cycle_time % NSEC_PER_SEC);
 	priv->plat->est->ctr[1] = (u32)(qopt->cycle_time / NSEC_PER_SEC);
 
+	if (fpe && !priv->dma_cap.fpesel)
+		return -EOPNOTSUPP;
+
+	ret = stmmac_fpe_configure(priv, priv->ioaddr,
+				   priv->plat->tx_queues_to_use,
+				   priv->plat->rx_queues_to_use, fpe);
+	if (ret && fpe) {
+		netdev_err(priv->dev, "failed to enable Frame Preemption\n");
+		return ret;
+	}
+
 	ret = stmmac_est_configure(priv, priv->ioaddr, priv->plat->est,
 				   priv->plat->clk_ptp_rate);
 	if (ret) {
-- 
cgit v1.2.3-59-g8ed1b


From 7c7282746883e48aff54ff6d55e6d8674a0c414a Mon Sep 17 00:00:00 2001
From: Jose Abreu <Jose.Abreu@synopsys.com>
Date: Wed, 18 Dec 2019 11:33:09 +0100
Subject: net: stmmac: gmac5+: Add support for Frame Preemption

Adds the HW specific support for Frame Preemption on GMAC5+ cores.

Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac4.h      |  3 +++
 drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c |  2 ++
 drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c  |  1 +
 drivers/net/ethernet/stmicro/stmmac/dwmac5.c      | 23 +++++++++++++++++++++++
 drivers/net/ethernet/stmicro/stmmac/dwmac5.h      |  5 +++++
 5 files changed, 34 insertions(+)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4.h b/drivers/net/ethernet/stmicro/stmmac/dwmac4.h
index f7f5a2751147..2e6b60a476c6 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac4.h
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4.h
@@ -64,6 +64,8 @@
 #define GMAC_RXQCTRL_MCBCQEN_SHIFT	20
 #define GMAC_RXQCTRL_TACPQE		BIT(21)
 #define GMAC_RXQCTRL_TACPQE_SHIFT	21
+#define GMAC_RXQCTRL_FPRQ		GENMASK(26, 24)
+#define GMAC_RXQCTRL_FPRQ_SHIFT		24
 
 /* MAC Packet Filtering */
 #define GMAC_PACKET_FILTER_PR		BIT(0)
@@ -237,6 +239,7 @@ enum power_event {
 
 /* MAC HW features3 bitmap */
 #define GMAC_HW_FEAT_ASP		GENMASK(29, 28)
+#define GMAC_HW_FEAT_FPESEL		BIT(26)
 #define GMAC_HW_FEAT_ESTWID		GENMASK(21, 20)
 #define GMAC_HW_FEAT_ESTDEP		GENMASK(19, 17)
 #define GMAC_HW_FEAT_ESTSEL		BIT(16)
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c
index 8df7496411a0..f0c0ea616032 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c
@@ -985,6 +985,7 @@ const struct stmmac_ops dwmac410_ops = {
 	.config_l3_filter = dwmac4_config_l3_filter,
 	.config_l4_filter = dwmac4_config_l4_filter,
 	.est_configure = dwmac5_est_configure,
+	.fpe_configure = dwmac5_fpe_configure,
 };
 
 const struct stmmac_ops dwmac510_ops = {
@@ -1029,6 +1030,7 @@ const struct stmmac_ops dwmac510_ops = {
 	.config_l3_filter = dwmac4_config_l3_filter,
 	.config_l4_filter = dwmac4_config_l4_filter,
 	.est_configure = dwmac5_est_configure,
+	.fpe_configure = dwmac5_fpe_configure,
 };
 
 int dwmac4_setup(struct stmmac_priv *priv)
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c
index 3552ce1201ff..213d44482ffa 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c
@@ -404,6 +404,7 @@ static void dwmac4_get_hw_feature(void __iomem *ioaddr,
 
 	/* 5.10 Features */
 	dma_cap->asp = (hw_cap & GMAC_HW_FEAT_ASP) >> 28;
+	dma_cap->fpesel = (hw_cap & GMAC_HW_FEAT_FPESEL) >> 26;
 	dma_cap->estwid = (hw_cap & GMAC_HW_FEAT_ESTWID) >> 20;
 	dma_cap->estdep = (hw_cap & GMAC_HW_FEAT_ESTDEP) >> 17;
 	dma_cap->estsel = (hw_cap & GMAC_HW_FEAT_ESTSEL) >> 16;
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac5.c b/drivers/net/ethernet/stmicro/stmmac/dwmac5.c
index 8047f402fb3f..5d4a3c2458ea 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac5.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac5.c
@@ -645,3 +645,26 @@ int dwmac5_est_configure(void __iomem *ioaddr, struct stmmac_est *cfg,
 	writel(ctrl, ioaddr + MTL_EST_CONTROL);
 	return 0;
 }
+
+void dwmac5_fpe_configure(void __iomem *ioaddr, u32 num_txq, u32 num_rxq,
+			  bool enable)
+{
+	u32 value;
+
+	if (!enable) {
+		value = readl(ioaddr + MAC_FPE_CTRL_STS);
+
+		value &= ~EFPE;
+
+		writel(value, ioaddr + MAC_FPE_CTRL_STS);
+	}
+
+	value = readl(ioaddr + GMAC_RXQ_CTRL1);
+	value &= ~GMAC_RXQCTRL_FPRQ;
+	value |= (num_rxq - 1) << GMAC_RXQCTRL_FPRQ_SHIFT;
+	writel(value, ioaddr + GMAC_RXQ_CTRL1);
+
+	value = readl(ioaddr + MAC_FPE_CTRL_STS);
+	value |= EFPE;
+	writel(value, ioaddr + MAC_FPE_CTRL_STS);
+}
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac5.h b/drivers/net/ethernet/stmicro/stmmac/dwmac5.h
index 70e6d8837dd9..3e8faa96b4d4 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac5.h
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac5.h
@@ -11,6 +11,9 @@
 #define PRTYEN				BIT(1)
 #define TMOUTEN				BIT(0)
 
+#define MAC_FPE_CTRL_STS		0x00000234
+#define EFPE				BIT(0)
+
 #define MAC_PPS_CONTROL			0x00000b70
 #define PPS_MAXIDX(x)			((((x) + 1) * 8) - 1)
 #define PPS_MINIDX(x)			((x) * 8)
@@ -102,5 +105,7 @@ int dwmac5_flex_pps_config(void __iomem *ioaddr, int index,
 			   u32 sub_second_inc, u32 systime_flags);
 int dwmac5_est_configure(void __iomem *ioaddr, struct stmmac_est *cfg,
 			 unsigned int ptp_rate);
+void dwmac5_fpe_configure(void __iomem *ioaddr, u32 num_txq, u32 num_rxq,
+			  bool enable);
 
 #endif /* __DWMAC5_H__ */
-- 
cgit v1.2.3-59-g8ed1b


From f0e56c8d8f7d0e03e571e21c778f02d226aaeb97 Mon Sep 17 00:00:00 2001
From: Jose Abreu <Jose.Abreu@synopsys.com>
Date: Wed, 18 Dec 2019 11:33:10 +0100
Subject: net: stmmac: xgmac3+: Add support for Frame Preemption

Adds the HW specific support for Frame Preemption on XGMAC3+ cores.

Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h     |  6 ++++++
 .../net/ethernet/stmicro/stmmac/dwxgmac2_core.c    | 24 ++++++++++++++++++++++
 drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c |  1 +
 3 files changed, 31 insertions(+)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h
index e1e79c90c865..174b903a8211 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h
+++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h
@@ -73,6 +73,9 @@
 #define XGMAC_RXQ_CTRL0			0x000000a0
 #define XGMAC_RXQEN(x)			GENMASK((x) * 2 + 1, (x) * 2)
 #define XGMAC_RXQEN_SHIFT(x)		((x) * 2)
+#define XGMAC_RXQ_CTRL1			0x000000a4
+#define XGMAC_RQ			GENMASK(7, 4)
+#define XGMAC_RQ_SHIFT			4
 #define XGMAC_RXQ_CTRL2			0x000000a8
 #define XGMAC_RXQ_CTRL3			0x000000ac
 #define XGMAC_PSRQ(x)			GENMASK((x) * 8 + 7, (x) * 8)
@@ -136,6 +139,7 @@
 #define XGMAC_HWFEAT_TXQCNT		GENMASK(9, 6)
 #define XGMAC_HWFEAT_RXQCNT		GENMASK(3, 0)
 #define XGMAC_HW_FEATURE3		0x00000128
+#define XGMAC_HWFEAT_FPESEL		BIT(26)
 #define XGMAC_HWFEAT_ESTWID		GENMASK(24, 23)
 #define XGMAC_HWFEAT_ESTDEP		GENMASK(22, 20)
 #define XGMAC_HWFEAT_ESTSEL		BIT(19)
@@ -151,6 +155,8 @@
 #define XGMAC_MDIO_ADDR			0x00000200
 #define XGMAC_MDIO_DATA			0x00000204
 #define XGMAC_MDIO_C22P			0x00000220
+#define XGMAC_FPE_CTRL_STS		0x00000280
+#define XGMAC_EFPE			BIT(0)
 #define XGMAC_ADDRx_HIGH(x)		(0x00000300 + (x) * 0x8)
 #define XGMAC_ADDR_MAX			32
 #define XGMAC_AE			BIT(31)
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c
index 2f6e960947d9..307105e8dea0 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c
@@ -1410,6 +1410,29 @@ static int dwxgmac3_est_configure(void __iomem *ioaddr, struct stmmac_est *cfg,
 	return 0;
 }
 
+static void dwxgmac3_fpe_configure(void __iomem *ioaddr, u32 num_txq,
+				   u32 num_rxq, bool enable)
+{
+	u32 value;
+
+	if (!enable) {
+		value = readl(ioaddr + XGMAC_FPE_CTRL_STS);
+
+		value &= ~XGMAC_EFPE;
+
+		writel(value, ioaddr + XGMAC_FPE_CTRL_STS);
+	}
+
+	value = readl(ioaddr + XGMAC_RXQ_CTRL1);
+	value &= ~XGMAC_RQ;
+	value |= (num_rxq - 1) << XGMAC_RQ_SHIFT;
+	writel(value, ioaddr + XGMAC_RXQ_CTRL1);
+
+	value = readl(ioaddr + XGMAC_FPE_CTRL_STS);
+	value |= XGMAC_EFPE;
+	writel(value, ioaddr + XGMAC_FPE_CTRL_STS);
+}
+
 const struct stmmac_ops dwxgmac210_ops = {
 	.core_init = dwxgmac2_core_init,
 	.set_mac = dwxgmac2_set_mac,
@@ -1454,6 +1477,7 @@ const struct stmmac_ops dwxgmac210_ops = {
 	.config_l4_filter = dwxgmac2_config_l4_filter,
 	.set_arp_offload = dwxgmac2_set_arp_offload,
 	.est_configure = dwxgmac3_est_configure,
+	.fpe_configure = dwxgmac3_fpe_configure,
 };
 
 int dwxgmac2_setup(struct stmmac_priv *priv)
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c
index 6ff2795a5117..c1ca73ebb0e7 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c
@@ -429,6 +429,7 @@ static void dwxgmac2_get_hw_feature(void __iomem *ioaddr,
 
 	/* MAC HW feature 3 */
 	hw_cap = readl(ioaddr + XGMAC_HW_FEATURE3);
+	dma_cap->fpesel = (hw_cap & XGMAC_HWFEAT_FPESEL) >> 26;
 	dma_cap->estwid = (hw_cap & XGMAC_HWFEAT_ESTWID) >> 23;
 	dma_cap->estdep = (hw_cap & XGMAC_HWFEAT_ESTDEP) >> 20;
 	dma_cap->estsel = (hw_cap & XGMAC_HWFEAT_ESTSEL) >> 19;
-- 
cgit v1.2.3-59-g8ed1b


From ea77b8c813c6e56b20bb88567b63acc147cefb57 Mon Sep 17 00:00:00 2001
From: Jose Abreu <Jose.Abreu@synopsys.com>
Date: Wed, 18 Dec 2019 11:33:11 +0100
Subject: net: stmmac: mmc: Add Frame Preemption counters on GMAC5+ cores

This can be useful for debug. Add these counters on GMAC5+ cores just
like we did for XGMAC.

Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/stmicro/stmmac/mmc_core.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/drivers/net/ethernet/stmicro/stmmac/mmc_core.c b/drivers/net/ethernet/stmicro/stmmac/mmc_core.c
index 252cf48c5816..a57b0fa815ab 100644
--- a/drivers/net/ethernet/stmicro/stmmac/mmc_core.c
+++ b/drivers/net/ethernet/stmicro/stmmac/mmc_core.c
@@ -119,6 +119,13 @@
 #define MMC_RX_ICMP_GD_OCTETS		0x180
 #define MMC_RX_ICMP_ERR_OCTETS		0x184
 
+#define MMC_TX_FPE_FRAG			0x1a8
+#define MMC_TX_HOLD_REQ			0x1ac
+#define MMC_RX_PKT_ASSEMBLY_ERR		0x1c8
+#define MMC_RX_PKT_SMD_ERR		0x1cc
+#define MMC_RX_PKT_ASSEMBLY_OK		0x1d0
+#define MMC_RX_FPE_FRAG			0x1d4
+
 /* XGMAC MMC Registers */
 #define MMC_XGMAC_TX_OCTET_GB		0x14
 #define MMC_XGMAC_TX_PKT_GB		0x1c
@@ -315,6 +322,15 @@ static void dwmac_mmc_read(void __iomem *mmcaddr, struct stmmac_counters *mmc)
 	mmc->mmc_rx_tcp_err_octets += readl(mmcaddr + MMC_RX_TCP_ERR_OCTETS);
 	mmc->mmc_rx_icmp_gd_octets += readl(mmcaddr + MMC_RX_ICMP_GD_OCTETS);
 	mmc->mmc_rx_icmp_err_octets += readl(mmcaddr + MMC_RX_ICMP_ERR_OCTETS);
+
+	mmc->mmc_tx_fpe_fragment_cntr += readl(mmcaddr + MMC_TX_FPE_FRAG);
+	mmc->mmc_tx_hold_req_cntr += readl(mmcaddr + MMC_TX_HOLD_REQ);
+	mmc->mmc_rx_packet_assembly_err_cntr +=
+		readl(mmcaddr + MMC_RX_PKT_ASSEMBLY_ERR);
+	mmc->mmc_rx_packet_smd_err_cntr += readl(mmcaddr + MMC_RX_PKT_SMD_ERR);
+	mmc->mmc_rx_packet_assembly_ok_cntr +=
+		readl(mmcaddr + MMC_RX_PKT_ASSEMBLY_OK);
+	mmc->mmc_rx_fpe_fragment_cntr += readl(mmcaddr + MMC_RX_FPE_FRAG);
 }
 
 const struct stmmac_mmc_ops dwmac_mmc_ops = {
-- 
cgit v1.2.3-59-g8ed1b


From 871602b1079234b2a9279062eaeadb3aad7bc863 Mon Sep 17 00:00:00 2001
From: Julian Wiedmann <jwi@linux.ibm.com>
Date: Wed, 18 Dec 2019 17:34:42 +0100
Subject: s390/qeth: only handle IRQs while device is online

A qeth device that's offline should not be receiving any IRQs - all
pending IOs have been terminated, and we avoid starting any new ones.

So rather than immediately registering the IRQ handler when the device
is probed, only register it while the device is online.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/s390/net/qeth_core_main.c | 68 ++++++++++++++++++---------------------
 1 file changed, 31 insertions(+), 37 deletions(-)

diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index ce7ff1abbef3..3d2374801308 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -972,8 +972,6 @@ static void qeth_irq(struct ccw_device *cdev, unsigned long intparm,
 	/* while we hold the ccwdev lock, this stays valid: */
 	gdev = dev_get_drvdata(&cdev->dev);
 	card = dev_get_drvdata(&gdev->dev);
-	if (!card)
-		return;
 
 	QETH_CARD_TEXT(card, 5, "irq");
 
@@ -1198,31 +1196,6 @@ static void qeth_free_buffer_pool(struct qeth_card *card)
 	}
 }
 
-static void qeth_clean_channel(struct qeth_channel *channel)
-{
-	struct ccw_device *cdev = channel->ccwdev;
-
-	QETH_DBF_TEXT(SETUP, 2, "freech");
-
-	spin_lock_irq(get_ccwdev_lock(cdev));
-	cdev->handler = NULL;
-	spin_unlock_irq(get_ccwdev_lock(cdev));
-}
-
-static void qeth_setup_channel(struct qeth_channel *channel)
-{
-	struct ccw_device *cdev = channel->ccwdev;
-
-	QETH_DBF_TEXT(SETUP, 2, "setupch");
-
-	channel->state = CH_STATE_DOWN;
-	atomic_set(&channel->irq_pending, 0);
-
-	spin_lock_irq(get_ccwdev_lock(cdev));
-	cdev->handler = qeth_irq;
-	spin_unlock_irq(get_ccwdev_lock(cdev));
-}
-
 static int qeth_osa_set_output_queues(struct qeth_card *card, bool single)
 {
 	unsigned int count = single ? 1 : card->dev->num_tx_queues;
@@ -1395,9 +1368,6 @@ static struct qeth_card *qeth_alloc_card(struct ccwgroup_device *gdev)
 	if (!card->read_cmd)
 		goto out_read_cmd;
 
-	qeth_setup_channel(&card->read);
-	qeth_setup_channel(&card->write);
-	qeth_setup_channel(&card->data);
 	card->qeth_service_level.seq_print = qeth_core_sl_print;
 	register_service_level(&card->qeth_service_level);
 	return card;
@@ -1467,12 +1437,38 @@ int qeth_stop_channel(struct qeth_channel *channel)
 			channel->active_cmd);
 		channel->active_cmd = NULL;
 	}
+	cdev->handler = NULL;
 	spin_unlock_irq(get_ccwdev_lock(cdev));
 
 	return rc;
 }
 EXPORT_SYMBOL_GPL(qeth_stop_channel);
 
+static int qeth_start_channel(struct qeth_channel *channel)
+{
+	struct ccw_device *cdev = channel->ccwdev;
+	int rc;
+
+	channel->state = CH_STATE_DOWN;
+	atomic_set(&channel->irq_pending, 0);
+
+	spin_lock_irq(get_ccwdev_lock(cdev));
+	cdev->handler = qeth_irq;
+	spin_unlock_irq(get_ccwdev_lock(cdev));
+
+	rc = ccw_device_set_online(cdev);
+	if (rc)
+		goto err;
+
+	return 0;
+
+err:
+	spin_lock_irq(get_ccwdev_lock(cdev));
+	cdev->handler = NULL;
+	spin_unlock_irq(get_ccwdev_lock(cdev));
+	return rc;
+}
+
 static int qeth_halt_channels(struct qeth_card *card)
 {
 	int rc1 = 0, rc2 = 0, rc3 = 0;
@@ -4706,7 +4702,7 @@ static void qeth_determine_capabilities(struct qeth_card *card)
 	QETH_CARD_TEXT(card, 2, "detcapab");
 	if (!ddev->online) {
 		ddev_offline = 1;
-		rc = ccw_device_set_online(ddev);
+		rc = qeth_start_channel(channel);
 		if (rc) {
 			QETH_CARD_TEXT_(card, 2, "3err%d", rc);
 			goto out;
@@ -4881,9 +4877,6 @@ out_free_nothing:
 static void qeth_core_free_card(struct qeth_card *card)
 {
 	QETH_CARD_TEXT(card, 2, "freecrd");
-	qeth_clean_channel(&card->read);
-	qeth_clean_channel(&card->write);
-	qeth_clean_channel(&card->data);
 	qeth_put_cmd(card->read_cmd);
 	destroy_workqueue(card->event_wq);
 	unregister_service_level(&card->qeth_service_level);
@@ -4946,13 +4939,14 @@ retry:
 	qeth_stop_channel(&card->write);
 	qeth_stop_channel(&card->read);
 	qdio_free(CARD_DDEV(card));
-	rc = ccw_device_set_online(CARD_RDEV(card));
+
+	rc = qeth_start_channel(&card->read);
 	if (rc)
 		goto retriable;
-	rc = ccw_device_set_online(CARD_WDEV(card));
+	rc = qeth_start_channel(&card->write);
 	if (rc)
 		goto retriable;
-	rc = ccw_device_set_online(CARD_DDEV(card));
+	rc = qeth_start_channel(&card->data);
 	if (rc)
 		goto retriable;
 retriable:
-- 
cgit v1.2.3-59-g8ed1b


From 7359393f3c19c9351e81345ef29f4f47229a6416 Mon Sep 17 00:00:00 2001
From: Julian Wiedmann <jwi@linux.ibm.com>
Date: Wed, 18 Dec 2019 17:34:43 +0100
Subject: s390/qeth: wake up all waiters from qeth_irq()

card->wait_q is shared by different users, for different wake-up
conditions. qeth_irq() can potentially trigger multiple of these
conditions:
1) A change to channel->irq_pending, which qeth_send_control_data() is
   waiting for.
2) A change to card->state, which qeth_clear_channel() and
   qeth_halt_channel() are waiting for.

As qeth_irq() does only a single wake_up(), we might miss to wake up
a second eligible waiter. Luckily all waiters are guarded with a
timeout, so this situation should recover on its own eventually.

To make things work robustly, add an additional wake_up() for changes
to channel->state. And extract a helper that updates
channel->irq_pending along with the needed wake_up().

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/s390/net/qeth_core.h      |  7 +++++++
 drivers/s390/net/qeth_core_main.c | 27 +++++++++++----------------
 2 files changed, 18 insertions(+), 16 deletions(-)

diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h
index a23875a4fa2b..b54ef12840a2 100644
--- a/drivers/s390/net/qeth_core.h
+++ b/drivers/s390/net/qeth_core.h
@@ -862,6 +862,13 @@ static inline bool qeth_card_hw_is_reachable(struct qeth_card *card)
 	return card->state == CARD_STATE_SOFTSETUP;
 }
 
+static inline void qeth_unlock_channel(struct qeth_card *card,
+				       struct qeth_channel *channel)
+{
+	atomic_set(&channel->irq_pending, 0);
+	wake_up(&card->wait_q);
+}
+
 struct qeth_trap_id {
 	__u16 lparnr;
 	char vmname[8];
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index 3d2374801308..38f3ed7567bc 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -520,11 +520,10 @@ static int __qeth_issue_next_read(struct qeth_card *card)
 	} else {
 		QETH_DBF_MESSAGE(2, "error %i on device %x when starting next read ccw!\n",
 				 rc, CARD_DEVID(card));
-		atomic_set(&channel->irq_pending, 0);
+		qeth_unlock_channel(card, channel);
 		qeth_put_cmd(iob);
 		card->read_or_write_problem = 1;
 		qeth_schedule_recovery(card);
-		wake_up(&card->wait_q);
 	}
 	return rc;
 }
@@ -1001,24 +1000,25 @@ static void qeth_irq(struct ccw_device *cdev, unsigned long intparm,
 	}
 
 	channel->active_cmd = NULL;
+	qeth_unlock_channel(card, channel);
 
 	rc = qeth_check_irb_error(card, cdev, irb);
 	if (rc) {
 		/* IO was terminated, free its resources. */
 		if (iob)
 			qeth_cancel_cmd(iob, rc);
-		atomic_set(&channel->irq_pending, 0);
-		wake_up(&card->wait_q);
 		return;
 	}
 
-	atomic_set(&channel->irq_pending, 0);
-
-	if (irb->scsw.cmd.fctl & (SCSW_FCTL_CLEAR_FUNC))
+	if (irb->scsw.cmd.fctl & SCSW_FCTL_CLEAR_FUNC) {
 		channel->state = CH_STATE_STOPPED;
+		wake_up(&card->wait_q);
+	}
 
-	if (irb->scsw.cmd.fctl & (SCSW_FCTL_HALT_FUNC))
+	if (irb->scsw.cmd.fctl & SCSW_FCTL_HALT_FUNC) {
 		channel->state = CH_STATE_HALTED;
+		wake_up(&card->wait_q);
+	}
 
 	if (iob && (irb->scsw.cmd.fctl & (SCSW_FCTL_CLEAR_FUNC |
 					  SCSW_FCTL_HALT_FUNC))) {
@@ -1052,7 +1052,7 @@ static void qeth_irq(struct ccw_device *cdev, unsigned long intparm,
 				qeth_cancel_cmd(iob, rc);
 			qeth_clear_ipacmd_list(card);
 			qeth_schedule_recovery(card);
-			goto out;
+			return;
 		}
 	}
 
@@ -1060,16 +1060,12 @@ static void qeth_irq(struct ccw_device *cdev, unsigned long intparm,
 		/* sanity check: */
 		if (irb->scsw.cmd.count > iob->length) {
 			qeth_cancel_cmd(iob, -EIO);
-			goto out;
+			return;
 		}
 		if (iob->callback)
 			iob->callback(card, iob,
 				      iob->length - irb->scsw.cmd.count);
 	}
-
-out:
-	wake_up(&card->wait_q);
-	return;
 }
 
 static void qeth_notify_skbs(struct qeth_qdio_out_q *q,
@@ -1780,8 +1776,7 @@ static int qeth_send_control_data(struct qeth_card *card,
 		QETH_CARD_TEXT_(card, 2, " err%d", rc);
 		qeth_dequeue_cmd(card, iob);
 		qeth_put_cmd(iob);
-		atomic_set(&channel->irq_pending, 0);
-		wake_up(&card->wait_q);
+		qeth_unlock_channel(card, channel);
 		goto out;
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From e6b1b7da247b329bb9e62adc97ef28678199cf35 Mon Sep 17 00:00:00 2001
From: Julian Wiedmann <jwi@linux.ibm.com>
Date: Wed, 18 Dec 2019 17:34:44 +0100
Subject: s390/qeth: overhaul L3 IP address dump code

The current code that dumps the RXIP/VIPA/IPATO addresses via sysfs
first checks whether the buffer still provides sufficient space to hold
another formatted address.
But the maximum length of an formatted IPv4 address is 15 characters,
not 12. So we underestimate the max required length and if the buffer
was previously filled to _just_ the right level, a formatted address can
end up being truncated.

Revamp these code paths to use the _actually_ required length of the
formatted IP address, and while at it suppress a gratuitous newline.

Also use scnprintf() to format the output. In case of a truncation, this
would allow us to return the number of characters that were actually
written.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/s390/net/qeth_l3.h      |  3 +-
 drivers/s390/net/qeth_l3_main.c | 20 ++++---------
 drivers/s390/net/qeth_l3_sys.c  | 66 +++++++++++++++++++++++------------------
 3 files changed, 44 insertions(+), 45 deletions(-)

diff --git a/drivers/s390/net/qeth_l3.h b/drivers/s390/net/qeth_l3.h
index 5db04fe472c0..2383ffad0a4a 100644
--- a/drivers/s390/net/qeth_l3.h
+++ b/drivers/s390/net/qeth_l3.h
@@ -102,7 +102,8 @@ struct qeth_ipato_entry {
 
 extern const struct attribute_group *qeth_l3_attr_groups[];
 
-void qeth_l3_ipaddr_to_string(enum qeth_prot_versions, const __u8 *, char *);
+int qeth_l3_ipaddr_to_string(enum qeth_prot_versions proto, const u8 *addr,
+			     char *buf);
 int qeth_l3_create_device_attributes(struct device *);
 void qeth_l3_remove_device_attributes(struct device *);
 int qeth_l3_setrouting_v4(struct qeth_card *);
diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index 27126330a4b0..3710761831cc 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -44,23 +44,13 @@ static int qeth_l3_register_addr_entry(struct qeth_card *,
 static int qeth_l3_deregister_addr_entry(struct qeth_card *,
 		struct qeth_ipaddr *);
 
-static void qeth_l3_ipaddr4_to_string(const __u8 *addr, char *buf)
-{
-	sprintf(buf, "%pI4", addr);
-}
-
-static void qeth_l3_ipaddr6_to_string(const __u8 *addr, char *buf)
-{
-	sprintf(buf, "%pI6", addr);
-}
-
-void qeth_l3_ipaddr_to_string(enum qeth_prot_versions proto, const __u8 *addr,
-				char *buf)
+int qeth_l3_ipaddr_to_string(enum qeth_prot_versions proto, const u8 *addr,
+			     char *buf)
 {
 	if (proto == QETH_PROT_IPV4)
-		qeth_l3_ipaddr4_to_string(addr, buf);
-	else if (proto == QETH_PROT_IPV6)
-		qeth_l3_ipaddr6_to_string(addr, buf);
+		return sprintf(buf, "%pI4", addr);
+	else
+		return sprintf(buf, "%pI6", addr);
 }
 
 static struct qeth_ipaddr *qeth_l3_find_addr_by_ip(struct qeth_card *card,
diff --git a/drivers/s390/net/qeth_l3_sys.c b/drivers/s390/net/qeth_l3_sys.c
index f9067ed6c7d3..aa9f3fc45447 100644
--- a/drivers/s390/net/qeth_l3_sys.c
+++ b/drivers/s390/net/qeth_l3_sys.c
@@ -386,30 +386,35 @@ static ssize_t qeth_l3_dev_ipato_add_show(char *buf, struct qeth_card *card,
 			enum qeth_prot_versions proto)
 {
 	struct qeth_ipato_entry *ipatoe;
-	char addr_str[40];
-	int entry_len; /* length of 1 entry string, differs between v4 and v6 */
-	int i = 0;
+	int str_len = 0;
 
-	entry_len = (proto == QETH_PROT_IPV4)? 12 : 40;
-	/* add strlen for "/<mask>\n" */
-	entry_len += (proto == QETH_PROT_IPV4)? 5 : 6;
 	mutex_lock(&card->ip_lock);
 	list_for_each_entry(ipatoe, &card->ipato.entries, entry) {
+		char addr_str[40];
+		int entry_len;
+
 		if (ipatoe->proto != proto)
 			continue;
-		/* String must not be longer than PAGE_SIZE. So we check if
-		 * string length gets near PAGE_SIZE. Then we can savely display
-		 * the next IPv6 address (worst case, compared to IPv4) */
-		if ((PAGE_SIZE - i) <= entry_len)
+
+		entry_len = qeth_l3_ipaddr_to_string(proto, ipatoe->addr,
+						     addr_str);
+		if (entry_len < 0)
+			continue;
+
+		/* Append /%mask to the entry: */
+		entry_len += 1 + ((proto == QETH_PROT_IPV4) ? 2 : 3);
+		/* Enough room to format %entry\n into null terminated page? */
+		if (entry_len + 1 > PAGE_SIZE - str_len - 1)
 			break;
-		qeth_l3_ipaddr_to_string(proto, ipatoe->addr, addr_str);
-		i += snprintf(buf + i, PAGE_SIZE - i,
-			      "%s/%i\n", addr_str, ipatoe->mask_bits);
+
+		entry_len = scnprintf(buf, PAGE_SIZE - str_len,
+				      "%s/%i\n", addr_str, ipatoe->mask_bits);
+		str_len += entry_len;
+		buf += entry_len;
 	}
 	mutex_unlock(&card->ip_lock);
-	i += snprintf(buf + i, PAGE_SIZE - i, "\n");
 
-	return i;
+	return str_len ? str_len : scnprintf(buf, PAGE_SIZE, "\n");
 }
 
 static ssize_t qeth_l3_dev_ipato_add4_show(struct device *dev,
@@ -607,31 +612,34 @@ static ssize_t qeth_l3_dev_ip_add_show(struct device *dev, char *buf,
 {
 	struct qeth_card *card = dev_get_drvdata(dev);
 	struct qeth_ipaddr *ipaddr;
-	char addr_str[40];
 	int str_len = 0;
-	int entry_len; /* length of 1 entry string, differs between v4 and v6 */
 	int i;
 
-	entry_len = (proto == QETH_PROT_IPV4)? 12 : 40;
-	entry_len += 2; /* \n + terminator */
 	mutex_lock(&card->ip_lock);
 	hash_for_each(card->ip_htable, i, ipaddr, hnode) {
+		char addr_str[40];
+		int entry_len;
+
 		if (ipaddr->proto != proto || ipaddr->type != type)
 			continue;
-		/* String must not be longer than PAGE_SIZE. So we check if
-		 * string length gets near PAGE_SIZE. Then we can savely display
-		 * the next IPv6 address (worst case, compared to IPv4) */
-		if ((PAGE_SIZE - str_len) <= entry_len)
+
+		entry_len = qeth_l3_ipaddr_to_string(proto, (u8 *)&ipaddr->u,
+						     addr_str);
+		if (entry_len < 0)
+			continue;
+
+		/* Enough room to format %addr\n into null terminated page? */
+		if (entry_len + 1 > PAGE_SIZE - str_len - 1)
 			break;
-		qeth_l3_ipaddr_to_string(proto, (const u8 *)&ipaddr->u,
-			addr_str);
-		str_len += snprintf(buf + str_len, PAGE_SIZE - str_len, "%s\n",
-				    addr_str);
+
+		entry_len = scnprintf(buf, PAGE_SIZE - str_len, "%s\n",
+				      addr_str);
+		str_len += entry_len;
+		buf += entry_len;
 	}
 	mutex_unlock(&card->ip_lock);
-	str_len += snprintf(buf + str_len, PAGE_SIZE - str_len, "\n");
 
-	return str_len;
+	return str_len ? str_len : scnprintf(buf, PAGE_SIZE, "\n");
 }
 
 static ssize_t qeth_l3_dev_vipa_add4_show(struct device *dev,
-- 
cgit v1.2.3-59-g8ed1b


From 2390166a6b457d3849dd4067c970f714b65bd9c3 Mon Sep 17 00:00:00 2001
From: Julian Wiedmann <jwi@linux.ibm.com>
Date: Wed, 18 Dec 2019 17:34:45 +0100
Subject: s390/qeth: clean up L3 sysfs code

Consolidate some duplicated code for adding RXIP/VIPA addresses, and
move the locking to where it's actually needed.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/s390/net/qeth_l3_main.c |  12 ++++-
 drivers/s390/net/qeth_l3_sys.c  | 106 +++++++++-------------------------------
 2 files changed, 33 insertions(+), 85 deletions(-)

diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index 3710761831cc..8db548340632 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -571,6 +571,7 @@ int qeth_l3_add_ipato_entry(struct qeth_card *card,
 
 	QETH_CARD_TEXT(card, 2, "addipato");
 
+	mutex_lock(&card->conf_mutex);
 	mutex_lock(&card->ip_lock);
 
 	list_for_each_entry(ipatoe, &card->ipato.entries, entry) {
@@ -590,6 +591,7 @@ int qeth_l3_add_ipato_entry(struct qeth_card *card,
 	}
 
 	mutex_unlock(&card->ip_lock);
+	mutex_unlock(&card->conf_mutex);
 
 	return rc;
 }
@@ -603,6 +605,7 @@ int qeth_l3_del_ipato_entry(struct qeth_card *card,
 
 	QETH_CARD_TEXT(card, 2, "delipato");
 
+	mutex_lock(&card->conf_mutex);
 	mutex_lock(&card->ip_lock);
 
 	list_for_each_entry_safe(ipatoe, tmp, &card->ipato.entries, entry) {
@@ -619,6 +622,8 @@ int qeth_l3_del_ipato_entry(struct qeth_card *card,
 	}
 
 	mutex_unlock(&card->ip_lock);
+	mutex_unlock(&card->conf_mutex);
+
 	return rc;
 }
 
@@ -627,6 +632,7 @@ int qeth_l3_modify_rxip_vipa(struct qeth_card *card, bool add, const u8 *ip,
 			     enum qeth_prot_versions proto)
 {
 	struct qeth_ipaddr addr;
+	int rc;
 
 	qeth_l3_init_ipaddr(&addr, type, proto);
 	if (proto == QETH_PROT_IPV4)
@@ -634,7 +640,11 @@ int qeth_l3_modify_rxip_vipa(struct qeth_card *card, bool add, const u8 *ip,
 	else
 		memcpy(&addr.u.a6.addr, ip, 16);
 
-	return qeth_l3_modify_ip(card, &addr, add);
+	mutex_lock(&card->conf_mutex);
+	rc = qeth_l3_modify_ip(card, &addr, add);
+	mutex_unlock(&card->conf_mutex);
+
+	return rc;
 }
 
 int qeth_l3_modify_hsuid(struct qeth_card *card, bool add)
diff --git a/drivers/s390/net/qeth_l3_sys.c b/drivers/s390/net/qeth_l3_sys.c
index aa9f3fc45447..96c73965eb68 100644
--- a/drivers/s390/net/qeth_l3_sys.c
+++ b/drivers/s390/net/qeth_l3_sys.c
@@ -460,16 +460,14 @@ static ssize_t qeth_l3_dev_ipato_add_store(const char *buf, size_t count,
 	int mask_bits;
 	int rc = 0;
 
-	mutex_lock(&card->conf_mutex);
 	rc = qeth_l3_parse_ipatoe(buf, proto, addr, &mask_bits);
 	if (rc)
-		goto out;
+		return rc;
 
 	ipatoe = kzalloc(sizeof(struct qeth_ipato_entry), GFP_KERNEL);
-	if (!ipatoe) {
-		rc = -ENOMEM;
-		goto out;
-	}
+	if (!ipatoe)
+		return -ENOMEM;
+
 	ipatoe->proto = proto;
 	memcpy(ipatoe->addr, addr, (proto == QETH_PROT_IPV4)? 4:16);
 	ipatoe->mask_bits = mask_bits;
@@ -477,8 +475,7 @@ static ssize_t qeth_l3_dev_ipato_add_store(const char *buf, size_t count,
 	rc = qeth_l3_add_ipato_entry(card, ipatoe);
 	if (rc)
 		kfree(ipatoe);
-out:
-	mutex_unlock(&card->conf_mutex);
+
 	return rc ? rc : count;
 }
 
@@ -501,11 +498,9 @@ static ssize_t qeth_l3_dev_ipato_del_store(const char *buf, size_t count,
 	int mask_bits;
 	int rc = 0;
 
-	mutex_lock(&card->conf_mutex);
 	rc = qeth_l3_parse_ipatoe(buf, proto, addr, &mask_bits);
 	if (!rc)
 		rc = qeth_l3_del_ipato_entry(card, proto, addr, mask_bits);
-	mutex_unlock(&card->conf_mutex);
 	return rc ? rc : count;
 }
 
@@ -650,63 +645,34 @@ static ssize_t qeth_l3_dev_vipa_add4_show(struct device *dev,
 				       QETH_IP_TYPE_VIPA);
 }
 
-static int qeth_l3_parse_vipae(const char *buf, enum qeth_prot_versions proto,
-		 u8 *addr)
-{
-	if (qeth_l3_string_to_ipaddr(buf, proto, addr)) {
-		return -EINVAL;
-	}
-	return 0;
-}
-
-static ssize_t qeth_l3_dev_vipa_add_store(const char *buf, size_t count,
-			struct qeth_card *card, enum qeth_prot_versions proto)
+static ssize_t qeth_l3_vipa_store(struct device *dev, const char *buf, bool add,
+				  size_t count, enum qeth_prot_versions proto)
 {
+	struct qeth_card *card = dev_get_drvdata(dev);
 	u8 addr[16] = {0, };
 	int rc;
 
-	mutex_lock(&card->conf_mutex);
-	rc = qeth_l3_parse_vipae(buf, proto, addr);
+	rc = qeth_l3_string_to_ipaddr(buf, proto, addr);
 	if (!rc)
-		rc = qeth_l3_modify_rxip_vipa(card, true, addr,
+		rc = qeth_l3_modify_rxip_vipa(card, add, addr,
 					      QETH_IP_TYPE_VIPA, proto);
-	mutex_unlock(&card->conf_mutex);
 	return rc ? rc : count;
 }
 
 static ssize_t qeth_l3_dev_vipa_add4_store(struct device *dev,
 		struct device_attribute *attr, const char *buf, size_t count)
 {
-	struct qeth_card *card = dev_get_drvdata(dev);
-
-	return qeth_l3_dev_vipa_add_store(buf, count, card, QETH_PROT_IPV4);
+	return qeth_l3_vipa_store(dev, buf, true, count, QETH_PROT_IPV4);
 }
 
 static QETH_DEVICE_ATTR(vipa_add4, add4, 0644,
 			qeth_l3_dev_vipa_add4_show,
 			qeth_l3_dev_vipa_add4_store);
 
-static ssize_t qeth_l3_dev_vipa_del_store(const char *buf, size_t count,
-			 struct qeth_card *card, enum qeth_prot_versions proto)
-{
-	u8 addr[16];
-	int rc;
-
-	mutex_lock(&card->conf_mutex);
-	rc = qeth_l3_parse_vipae(buf, proto, addr);
-	if (!rc)
-		rc = qeth_l3_modify_rxip_vipa(card, false, addr,
-					      QETH_IP_TYPE_VIPA, proto);
-	mutex_unlock(&card->conf_mutex);
-	return rc ? rc : count;
-}
-
 static ssize_t qeth_l3_dev_vipa_del4_store(struct device *dev,
 		struct device_attribute *attr, const char *buf, size_t count)
 {
-	struct qeth_card *card = dev_get_drvdata(dev);
-
-	return qeth_l3_dev_vipa_del_store(buf, count, card, QETH_PROT_IPV4);
+	return qeth_l3_vipa_store(dev, buf, true, count, QETH_PROT_IPV4);
 }
 
 static QETH_DEVICE_ATTR(vipa_del4, del4, 0200, NULL,
@@ -723,9 +689,7 @@ static ssize_t qeth_l3_dev_vipa_add6_show(struct device *dev,
 static ssize_t qeth_l3_dev_vipa_add6_store(struct device *dev,
 		struct device_attribute *attr, const char *buf, size_t count)
 {
-	struct qeth_card *card = dev_get_drvdata(dev);
-
-	return qeth_l3_dev_vipa_add_store(buf, count, card, QETH_PROT_IPV6);
+	return qeth_l3_vipa_store(dev, buf, true, count, QETH_PROT_IPV6);
 }
 
 static QETH_DEVICE_ATTR(vipa_add6, add6, 0644,
@@ -735,9 +699,7 @@ static QETH_DEVICE_ATTR(vipa_add6, add6, 0644,
 static ssize_t qeth_l3_dev_vipa_del6_store(struct device *dev,
 		struct device_attribute *attr, const char *buf, size_t count)
 {
-	struct qeth_card *card = dev_get_drvdata(dev);
-
-	return qeth_l3_dev_vipa_del_store(buf, count, card, QETH_PROT_IPV6);
+	return qeth_l3_vipa_store(dev, buf, false, count, QETH_PROT_IPV6);
 }
 
 static QETH_DEVICE_ATTR(vipa_del6, del6, 0200, NULL,
@@ -790,54 +752,34 @@ static int qeth_l3_parse_rxipe(const char *buf, enum qeth_prot_versions proto,
 	return 0;
 }
 
-static ssize_t qeth_l3_dev_rxip_add_store(const char *buf, size_t count,
-			struct qeth_card *card, enum qeth_prot_versions proto)
+static ssize_t qeth_l3_rxip_store(struct device *dev, const char *buf, bool add,
+				  size_t count, enum qeth_prot_versions proto)
 {
+	struct qeth_card *card = dev_get_drvdata(dev);
 	u8 addr[16] = {0, };
 	int rc;
 
-	mutex_lock(&card->conf_mutex);
 	rc = qeth_l3_parse_rxipe(buf, proto, addr);
 	if (!rc)
-		rc = qeth_l3_modify_rxip_vipa(card, true, addr,
+		rc = qeth_l3_modify_rxip_vipa(card, add, addr,
 					      QETH_IP_TYPE_RXIP, proto);
-	mutex_unlock(&card->conf_mutex);
 	return rc ? rc : count;
 }
 
 static ssize_t qeth_l3_dev_rxip_add4_store(struct device *dev,
 		struct device_attribute *attr, const char *buf, size_t count)
 {
-	struct qeth_card *card = dev_get_drvdata(dev);
-
-	return qeth_l3_dev_rxip_add_store(buf, count, card, QETH_PROT_IPV4);
+	return qeth_l3_rxip_store(dev, buf, true, count, QETH_PROT_IPV4);
 }
 
 static QETH_DEVICE_ATTR(rxip_add4, add4, 0644,
 			qeth_l3_dev_rxip_add4_show,
 			qeth_l3_dev_rxip_add4_store);
 
-static ssize_t qeth_l3_dev_rxip_del_store(const char *buf, size_t count,
-			struct qeth_card *card, enum qeth_prot_versions proto)
-{
-	u8 addr[16];
-	int rc;
-
-	mutex_lock(&card->conf_mutex);
-	rc = qeth_l3_parse_rxipe(buf, proto, addr);
-	if (!rc)
-		rc = qeth_l3_modify_rxip_vipa(card, false, addr,
-					      QETH_IP_TYPE_RXIP, proto);
-	mutex_unlock(&card->conf_mutex);
-	return rc ? rc : count;
-}
-
 static ssize_t qeth_l3_dev_rxip_del4_store(struct device *dev,
 		struct device_attribute *attr, const char *buf, size_t count)
 {
-	struct qeth_card *card = dev_get_drvdata(dev);
-
-	return qeth_l3_dev_rxip_del_store(buf, count, card, QETH_PROT_IPV4);
+	return qeth_l3_rxip_store(dev, buf, false, count, QETH_PROT_IPV4);
 }
 
 static QETH_DEVICE_ATTR(rxip_del4, del4, 0200, NULL,
@@ -854,9 +796,7 @@ static ssize_t qeth_l3_dev_rxip_add6_show(struct device *dev,
 static ssize_t qeth_l3_dev_rxip_add6_store(struct device *dev,
 		struct device_attribute *attr, const char *buf, size_t count)
 {
-	struct qeth_card *card = dev_get_drvdata(dev);
-
-	return qeth_l3_dev_rxip_add_store(buf, count, card, QETH_PROT_IPV6);
+	return qeth_l3_rxip_store(dev, buf, true, count, QETH_PROT_IPV6);
 }
 
 static QETH_DEVICE_ATTR(rxip_add6, add6, 0644,
@@ -866,9 +806,7 @@ static QETH_DEVICE_ATTR(rxip_add6, add6, 0644,
 static ssize_t qeth_l3_dev_rxip_del6_store(struct device *dev,
 		struct device_attribute *attr, const char *buf, size_t count)
 {
-	struct qeth_card *card = dev_get_drvdata(dev);
-
-	return qeth_l3_dev_rxip_del_store(buf, count, card, QETH_PROT_IPV6);
+	return qeth_l3_rxip_store(dev, buf, false, count, QETH_PROT_IPV6);
 }
 
 static QETH_DEVICE_ATTR(rxip_del6, del6, 0200, NULL,
-- 
cgit v1.2.3-59-g8ed1b


From 490df97142fe38f7af9ff8d4df8e9ece41df38ec Mon Sep 17 00:00:00 2001
From: Julian Wiedmann <jwi@linux.ibm.com>
Date: Wed, 18 Dec 2019 17:34:46 +0100
Subject: s390/qeth: remove open-coded inet_make_mask()

Use inet_make_mask() to replace some complicated bit-fiddling.

Also use the right data types to replace some raw memcpy calls with
proper assignments.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/s390/net/qeth_core_mpc.h |  8 ++++----
 drivers/s390/net/qeth_l3.h       |  2 +-
 drivers/s390/net/qeth_l3_main.c  | 34 +++++++++++++++-------------------
 3 files changed, 20 insertions(+), 24 deletions(-)

diff --git a/drivers/s390/net/qeth_core_mpc.h b/drivers/s390/net/qeth_core_mpc.h
index 88f4dc140751..f13225b43781 100644
--- a/drivers/s390/net/qeth_core_mpc.h
+++ b/drivers/s390/net/qeth_core_mpc.h
@@ -338,14 +338,14 @@ enum qeth_card_info_port_speed {
 
 /* (SET)DELIP(M) IPA stuff ***************************************************/
 struct qeth_ipacmd_setdelip4 {
-	__u8   ip_addr[4];
-	__u8   mask[4];
+	__be32 addr;
+	__be32 mask;
 	__u32  flags;
 } __attribute__ ((packed));
 
 struct qeth_ipacmd_setdelip6 {
-	__u8   ip_addr[16];
-	__u8   mask[16];
+	struct in6_addr addr;
+	struct in6_addr prefix;
 	__u32  flags;
 } __attribute__ ((packed));
 
diff --git a/drivers/s390/net/qeth_l3.h b/drivers/s390/net/qeth_l3.h
index 2383ffad0a4a..89fb91dad12e 100644
--- a/drivers/s390/net/qeth_l3.h
+++ b/drivers/s390/net/qeth_l3.h
@@ -35,7 +35,7 @@ struct qeth_ipaddr {
 	union {
 		struct {
 			__be32 addr;
-			unsigned int mask;
+			__be32 mask;
 		} a4;
 		struct {
 			struct in6_addr addr;
diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index 8db548340632..bd2273f9ca35 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -369,17 +369,16 @@ static int qeth_l3_send_setdelmc(struct qeth_card *card,
 	return qeth_send_ipa_cmd(card, iob, qeth_l3_setdelip_cb, NULL);
 }
 
-static void qeth_l3_fill_netmask(u8 *netmask, unsigned int len)
+static void qeth_l3_set_ipv6_prefix(struct in6_addr *prefix, unsigned int len)
 {
-	int i, j;
-	for (i = 0; i < 16; i++) {
-		j = (len) - (i * 8);
-		if (j >= 8)
-			netmask[i] = 0xff;
-		else if (j > 0)
-			netmask[i] = (u8)(0xFF00 >> j);
-		else
-			netmask[i] = 0;
+	unsigned int i = 0;
+
+	while (len && i < 4) {
+		int mask_len = min_t(int, len, 32);
+
+		prefix->s6_addr32[i] = inet_make_mask(mask_len);
+		len -= mask_len;
+		i++;
 	}
 }
 
@@ -402,7 +401,6 @@ static int qeth_l3_send_setdelip(struct qeth_card *card,
 {
 	struct qeth_cmd_buffer *iob;
 	struct qeth_ipa_cmd *cmd;
-	__u8 netmask[16];
 	u32 flags;
 
 	QETH_CARD_TEXT(card, 4, "setdelip");
@@ -417,15 +415,13 @@ static int qeth_l3_send_setdelip(struct qeth_card *card,
 	QETH_CARD_TEXT_(card, 4, "flags%02X", flags);
 
 	if (addr->proto == QETH_PROT_IPV6) {
-		memcpy(cmd->data.setdelip6.ip_addr, &addr->u.a6.addr,
-		       sizeof(struct in6_addr));
-		qeth_l3_fill_netmask(netmask, addr->u.a6.pfxlen);
-		memcpy(cmd->data.setdelip6.mask, netmask,
-		       sizeof(struct in6_addr));
+		cmd->data.setdelip6.addr = addr->u.a6.addr;
+		qeth_l3_set_ipv6_prefix(&cmd->data.setdelip6.prefix,
+					addr->u.a6.pfxlen);
 		cmd->data.setdelip6.flags = flags;
 	} else {
-		memcpy(cmd->data.setdelip4.ip_addr, &addr->u.a4.addr, 4);
-		memcpy(cmd->data.setdelip4.mask, &addr->u.a4.mask, 4);
+		cmd->data.setdelip4.addr = addr->u.a4.addr;
+		cmd->data.setdelip4.mask = addr->u.a4.mask;
 		cmd->data.setdelip4.flags = flags;
 	}
 
@@ -2436,7 +2432,7 @@ static int qeth_l3_ip_event(struct notifier_block *this,
 
 	qeth_l3_init_ipaddr(&addr, QETH_IP_TYPE_NORMAL, QETH_PROT_IPV4);
 	addr.u.a4.addr = ifa->ifa_address;
-	addr.u.a4.mask = be32_to_cpu(ifa->ifa_mask);
+	addr.u.a4.mask = ifa->ifa_mask;
 
 	return qeth_l3_handle_ip_event(card, &addr, event);
 }
-- 
cgit v1.2.3-59-g8ed1b


From b6beb62a5261f4ba2e73a39889f19e5af3ce545a Mon Sep 17 00:00:00 2001
From: Julian Wiedmann <jwi@linux.ibm.com>
Date: Wed, 18 Dec 2019 17:34:47 +0100
Subject: s390/qeth: don't raise NETDEV_REBOOT event from L3 offline path

This is a leftover from back when a recovery action didn't go through
dev_close(), and was meant to shoot down all remaining af_iucv sockets
on the interface.

Now that the offline path always calls dev_close(), the
NETDEV_GOING_DOWN event from __dev_close_many() is sufficient and this
hack can be removed.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/s390/net/qeth_l3_main.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index bd2273f9ca35..8bd3fc73a4fb 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -2288,12 +2288,6 @@ static int __qeth_l3_set_offline(struct ccwgroup_device *cgdev,
 	rtnl_unlock();
 
 	qeth_l3_stop_card(card);
-	if (card->options.cq == QETH_CQ_ENABLED) {
-		rtnl_lock();
-		call_netdevice_notifiers(NETDEV_REBOOT, card->dev);
-		rtnl_unlock();
-	}
-
 	rc  = qeth_stop_channel(&card->data);
 	rc2 = qeth_stop_channel(&card->write);
 	rc3 = qeth_stop_channel(&card->read);
-- 
cgit v1.2.3-59-g8ed1b


From adee2592b6c0e96d8b079c7c9116264293e923e6 Mon Sep 17 00:00:00 2001
From: Julian Wiedmann <jwi@linux.ibm.com>
Date: Wed, 18 Dec 2019 17:34:48 +0100
Subject: s390/qeth: stop yielding the ip_lock during IPv4 registration

As commit df2a2a5225cc ("s390/qeth: convert IP table spinlock to mutex")
converted the ip_lock to a mutex, we no longer have to yield it while
the subsequent IO sleep-waits for completion.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/s390/net/qeth_l3.h      |  1 -
 drivers/s390/net/qeth_l3_main.c | 34 ++--------------------------------
 2 files changed, 2 insertions(+), 33 deletions(-)

diff --git a/drivers/s390/net/qeth_l3.h b/drivers/s390/net/qeth_l3.h
index 89fb91dad12e..6ccfe2121095 100644
--- a/drivers/s390/net/qeth_l3.h
+++ b/drivers/s390/net/qeth_l3.h
@@ -23,7 +23,6 @@ struct qeth_ipaddr {
 	struct hlist_node hnode;
 	enum qeth_ip_types type;
 	u8 is_multicast:1;
-	u8 in_progress:1;
 	u8 disp_flag:2;
 	u8 ipato:1;			/* ucast only */
 
diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index 8bd3fc73a4fb..789d3b2ba0de 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -151,8 +151,6 @@ static int qeth_l3_delete_ip(struct qeth_card *card,
 	addr->ref_counter--;
 	if (addr->type == QETH_IP_TYPE_NORMAL && addr->ref_counter > 0)
 		return rc;
-	if (addr->in_progress)
-		return -EINPROGRESS;
 
 	if (qeth_card_hw_is_reachable(card))
 		rc = qeth_l3_deregister_addr_entry(card, addr);
@@ -213,29 +211,10 @@ static int qeth_l3_add_ip(struct qeth_card *card, struct qeth_ipaddr *tmp_addr)
 			return 0;
 		}
 
-		/* qeth_l3_register_addr_entry can go to sleep
-		 * if we add a IPV4 addr. It is caused by the reason
-		 * that SETIP ipa cmd starts ARP staff for IPV4 addr.
-		 * Thus we should unlock spinlock, and make a protection
-		 * using in_progress variable to indicate that there is
-		 * an hardware operation with this IPV4 address
-		 */
-		if (addr->proto == QETH_PROT_IPV4) {
-			addr->in_progress = 1;
-			mutex_unlock(&card->ip_lock);
-			rc = qeth_l3_register_addr_entry(card, addr);
-			mutex_lock(&card->ip_lock);
-			addr->in_progress = 0;
-		} else
-			rc = qeth_l3_register_addr_entry(card, addr);
+		rc = qeth_l3_register_addr_entry(card, addr);
 
 		if (!rc || rc == -EADDRINUSE || rc == -ENETDOWN) {
 			addr->disp_flag = QETH_DISP_ADDR_DO_NOTHING;
-			if (addr->ref_counter < 1) {
-				qeth_l3_deregister_addr_entry(card, addr);
-				hash_del(&addr->hnode);
-				kfree(addr);
-			}
 		} else {
 			hash_del(&addr->hnode);
 			kfree(addr);
@@ -303,19 +282,10 @@ static void qeth_l3_recover_ip(struct qeth_card *card)
 
 	hash_for_each_safe(card->ip_htable, i, tmp, addr, hnode) {
 		if (addr->disp_flag == QETH_DISP_ADDR_ADD) {
-			if (addr->proto == QETH_PROT_IPV4) {
-				addr->in_progress = 1;
-				mutex_unlock(&card->ip_lock);
-				rc = qeth_l3_register_addr_entry(card, addr);
-				mutex_lock(&card->ip_lock);
-				addr->in_progress = 0;
-			} else
-				rc = qeth_l3_register_addr_entry(card, addr);
+			rc = qeth_l3_register_addr_entry(card, addr);
 
 			if (!rc) {
 				addr->disp_flag = QETH_DISP_ADDR_DO_NOTHING;
-				if (addr->ref_counter < 1)
-					qeth_l3_delete_ip(card, addr);
 			} else {
 				hash_del(&addr->hnode);
 				kfree(addr);
-- 
cgit v1.2.3-59-g8ed1b


From 52f82bf16b6e754c2a9971420ccc8209ff52081c Mon Sep 17 00:00:00 2001
From: Julian Wiedmann <jwi@linux.ibm.com>
Date: Wed, 18 Dec 2019 17:34:49 +0100
Subject: s390/qeth: consolidate helpers for capability checking

Convert the old code to use struct qeth_ipa_caps, and while at it remove
all unused helper macros.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/s390/net/qeth_core.h      | 47 +++------------------------------------
 drivers/s390/net/qeth_core_main.c | 28 +++++++++++------------
 drivers/s390/net/qeth_core_mpc.h  | 13 +++++++++--
 3 files changed, 27 insertions(+), 61 deletions(-)

diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h
index b54ef12840a2..6e16b19732f6 100644
--- a/drivers/s390/net/qeth_core.h
+++ b/drivers/s390/net/qeth_core.h
@@ -125,12 +125,6 @@ struct qeth_routing_info {
 	enum qeth_routing_types type;
 };
 
-/* IPA stuff */
-struct qeth_ipa_info {
-	__u32 supported_funcs;
-	__u32 enabled_funcs;
-};
-
 /* SETBRIDGEPORT stuff */
 enum qeth_sbp_roles {
 	QETH_SBP_ROLE_NONE	= 0,
@@ -169,41 +163,6 @@ struct qeth_vnicc_info {
 	bool rx_bcast_enabled;
 };
 
-static inline int qeth_is_adp_supported(struct qeth_ipa_info *ipa,
-		enum qeth_ipa_setadp_cmd func)
-{
-	return (ipa->supported_funcs & func);
-}
-
-static inline int qeth_is_ipa_supported(struct qeth_ipa_info *ipa,
-		enum qeth_ipa_funcs func)
-{
-	return (ipa->supported_funcs & func);
-}
-
-static inline int qeth_is_ipa_enabled(struct qeth_ipa_info *ipa,
-		enum qeth_ipa_funcs func)
-{
-	return (ipa->supported_funcs & ipa->enabled_funcs & func);
-}
-
-#define qeth_adp_supported(c, f) \
-	qeth_is_adp_supported(&c->options.adp, f)
-#define qeth_is_supported(c, f) \
-	qeth_is_ipa_supported(&c->options.ipa4, f)
-#define qeth_is_enabled(c, f) \
-	qeth_is_ipa_enabled(&c->options.ipa4, f)
-#define qeth_is_supported6(c, f) \
-	qeth_is_ipa_supported(&c->options.ipa6, f)
-#define qeth_is_enabled6(c, f) \
-	qeth_is_ipa_enabled(&c->options.ipa6, f)
-#define qeth_is_ipafunc_supported(c, prot, f) \
-	 ((prot == QETH_PROT_IPV6) ? \
-		qeth_is_supported6(c, f) : qeth_is_supported(c, f))
-#define qeth_is_ipafunc_enabled(c, prot, f) \
-	 ((prot == QETH_PROT_IPV6) ? \
-		qeth_is_enabled6(c, f) : qeth_is_enabled(c, f))
-
 #define QETH_IDX_FUNC_LEVEL_OSD		 0x0101
 #define QETH_IDX_FUNC_LEVEL_IQD		 0x4108
 
@@ -735,11 +694,11 @@ enum qeth_discipline_id {
 };
 
 struct qeth_card_options {
+	struct qeth_ipa_caps ipa4;
+	struct qeth_ipa_caps ipa6;
 	struct qeth_routing_info route4;
-	struct qeth_ipa_info ipa4;
-	struct qeth_ipa_info adp; /*Adapter parameters*/
 	struct qeth_routing_info route6;
-	struct qeth_ipa_info ipa6;
+	struct qeth_ipa_caps adp; /* Adapter parameters */
 	struct qeth_sbp_info sbp; /* SETBRIDGEPORT options */
 	struct qeth_vnicc_info vnicc; /* VNICC options */
 	int fake_broadcast;
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index 38f3ed7567bc..6b55271091cb 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -2862,7 +2862,7 @@ static int qeth_query_setadapterparms_cb(struct qeth_card *card,
 		      cmd->data.setadapterparms.data.query_cmds_supp.lan_type;
 		QETH_CARD_TEXT_(card, 2, "lnk %d", card->info.link_type);
 	}
-	card->options.adp.supported_funcs =
+	card->options.adp.supported =
 		cmd->data.setadapterparms.data.query_cmds_supp.supported_cmds;
 	return 0;
 }
@@ -2918,8 +2918,8 @@ static int qeth_query_ipassists_cb(struct qeth_card *card,
 	case IPA_RC_NOTSUPP:
 	case IPA_RC_L2_UNSUPPORTED_CMD:
 		QETH_CARD_TEXT(card, 2, "ipaunsup");
-		card->options.ipa4.supported_funcs |= IPA_SETADAPTERPARMS;
-		card->options.ipa6.supported_funcs |= IPA_SETADAPTERPARMS;
+		card->options.ipa4.supported |= IPA_SETADAPTERPARMS;
+		card->options.ipa6.supported |= IPA_SETADAPTERPARMS;
 		return -EOPNOTSUPP;
 	default:
 		QETH_DBF_MESSAGE(1, "IPA_CMD_QIPASSIST on device %x: Unhandled rc=%#x\n",
@@ -2927,13 +2927,11 @@ static int qeth_query_ipassists_cb(struct qeth_card *card,
 		return -EIO;
 	}
 
-	if (cmd->hdr.prot_version == QETH_PROT_IPV4) {
-		card->options.ipa4.supported_funcs = cmd->hdr.ipa_supported;
-		card->options.ipa4.enabled_funcs = cmd->hdr.ipa_enabled;
-	} else if (cmd->hdr.prot_version == QETH_PROT_IPV6) {
-		card->options.ipa6.supported_funcs = cmd->hdr.ipa_supported;
-		card->options.ipa6.enabled_funcs = cmd->hdr.ipa_enabled;
-	} else
+	if (cmd->hdr.prot_version == QETH_PROT_IPV4)
+		card->options.ipa4 = cmd->hdr.assists;
+	else if (cmd->hdr.prot_version == QETH_PROT_IPV6)
+		card->options.ipa6 = cmd->hdr.assists;
+	else
 		QETH_DBF_MESSAGE(1, "IPA_CMD_QIPASSIST on device %x: Flawed LIC detected\n",
 				 CARD_DEVID(card));
 	return 0;
@@ -5002,9 +5000,9 @@ retriable:
 		*carrier_ok = true;
 	}
 
-	card->options.ipa4.supported_funcs = 0;
-	card->options.ipa6.supported_funcs = 0;
-	card->options.adp.supported_funcs = 0;
+	card->options.ipa4.supported = 0;
+	card->options.ipa6.supported = 0;
+	card->options.adp.supported = 0;
 	card->options.sbp.supported_funcs = 0;
 	card->info.diagass_support = 0;
 	rc = qeth_query_ipassists(card, QETH_PROT_IPV4);
@@ -5421,9 +5419,9 @@ int qeth_setassparms_cb(struct qeth_card *card,
 
 	cmd->hdr.return_code = cmd->data.setassparms.hdr.return_code;
 	if (cmd->hdr.prot_version == QETH_PROT_IPV4)
-		card->options.ipa4.enabled_funcs = cmd->hdr.ipa_enabled;
+		card->options.ipa4.enabled = cmd->hdr.assists.enabled;
 	if (cmd->hdr.prot_version == QETH_PROT_IPV6)
-		card->options.ipa6.enabled_funcs = cmd->hdr.ipa_enabled;
+		card->options.ipa6.enabled = cmd->hdr.assists.enabled;
 	return 0;
 }
 EXPORT_SYMBOL_GPL(qeth_setassparms_cb);
diff --git a/drivers/s390/net/qeth_core_mpc.h b/drivers/s390/net/qeth_core_mpc.h
index f13225b43781..f4dc37e28ac7 100644
--- a/drivers/s390/net/qeth_core_mpc.h
+++ b/drivers/s390/net/qeth_core_mpc.h
@@ -53,6 +53,16 @@ static inline bool qeth_ipa_caps_enabled(struct qeth_ipa_caps *caps, u32 mask)
 	return (caps->enabled & mask) == mask;
 }
 
+#define qeth_adp_supported(c, f) \
+	qeth_ipa_caps_supported(&c->options.adp, f)
+#define qeth_is_supported(c, f) \
+	qeth_ipa_caps_supported(&c->options.ipa4, f)
+#define qeth_is_supported6(c, f) \
+	qeth_ipa_caps_supported(&c->options.ipa6, f)
+#define qeth_is_ipafunc_supported(c, prot, f) \
+	 ((prot == QETH_PROT_IPV6) ? qeth_is_supported6(c, f) : \
+				     qeth_is_supported(c, f))
+
 enum qeth_card_types {
 	QETH_CARD_TYPE_OSD     = 1,
 	QETH_CARD_TYPE_IQD     = 5,
@@ -766,8 +776,7 @@ struct qeth_ipacmd_hdr {
 	__u8   prim_version_no;
 	__u8   param_count;
 	__u16  prot_version;
-	__u32  ipa_supported;
-	__u32  ipa_enabled;
+	struct qeth_ipa_caps assists;
 } __attribute__ ((packed));
 
 /* The IPA command itself */
-- 
cgit v1.2.3-59-g8ed1b


From 334b49de120dd6deb28c31eedb7ca81351f7464e Mon Sep 17 00:00:00 2001
From: Julian Wiedmann <jwi@linux.ibm.com>
Date: Wed, 18 Dec 2019 17:34:50 +0100
Subject: s390/qeth: make use of napi_schedule_irqoff()

qeth_qdio_start_poll() is called from the qdio layer's IRQ handler,
while IRQs are masked.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/s390/net/qeth_core_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index 6b55271091cb..78349355c582 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -3402,7 +3402,7 @@ static void qeth_qdio_start_poll(struct ccw_device *ccwdev, int queue,
 	struct qeth_card *card = (struct qeth_card *)card_ptr;
 
 	if (card->dev->flags & IFF_UP)
-		napi_schedule(&card->napi);
+		napi_schedule_irqoff(&card->napi);
 }
 
 int qeth_configure_cq(struct qeth_card *card, enum qeth_cq cq)
-- 
cgit v1.2.3-59-g8ed1b


From cbd22f172df782d6b2150044d0ff2051bad2a00f Mon Sep 17 00:00:00 2001
From: Kevin 'ldir' Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
Date: Wed, 18 Dec 2019 14:05:13 +0000
Subject: sch_cake: drop unused variable tin_quantum_prio
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Turns out tin_quantum_prio isn't used anymore and is a leftover from a
previous implementation of diffserv tins.  Since the variable isn't used
in any calculations it can be eliminated.

Drop variable and places where it was set.  Rename remaining variable
and consolidate naming of intermediate variables that set it.

Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sched/sch_cake.c | 59 ++++++++++++++++------------------------------------
 1 file changed, 18 insertions(+), 41 deletions(-)

diff --git a/net/sched/sch_cake.c b/net/sched/sch_cake.c
index e0f40400f679..6cc3ab145513 100644
--- a/net/sched/sch_cake.c
+++ b/net/sched/sch_cake.c
@@ -173,8 +173,7 @@ struct cake_tin_data {
 	u64	tin_rate_bps;
 	u16	tin_rate_shft;
 
-	u16	tin_quantum_prio;
-	u16	tin_quantum_band;
+	u16	tin_quantum;
 	s32	tin_deficit;
 	u32	tin_backlog;
 	u32	tin_dropped;
@@ -1919,7 +1918,7 @@ begin:
 		while (b->tin_deficit < 0 ||
 		       !(b->sparse_flow_count + b->bulk_flow_count)) {
 			if (b->tin_deficit <= 0)
-				b->tin_deficit += b->tin_quantum_band;
+				b->tin_deficit += b->tin_quantum;
 			if (b->sparse_flow_count + b->bulk_flow_count)
 				empty = false;
 
@@ -2241,8 +2240,7 @@ static int cake_config_besteffort(struct Qdisc *sch)
 
 	cake_set_rate(b, rate, mtu,
 		      us_to_ns(q->target), us_to_ns(q->interval));
-	b->tin_quantum_band = 65535;
-	b->tin_quantum_prio = 65535;
+	b->tin_quantum = 65535;
 
 	return 0;
 }
@@ -2253,8 +2251,7 @@ static int cake_config_precedence(struct Qdisc *sch)
 	struct cake_sched_data *q = qdisc_priv(sch);
 	u32 mtu = psched_mtu(qdisc_dev(sch));
 	u64 rate = q->rate_bps;
-	u32 quantum1 = 256;
-	u32 quantum2 = 256;
+	u32 quantum = 256;
 	u32 i;
 
 	q->tin_cnt = 8;
@@ -2267,18 +2264,14 @@ static int cake_config_precedence(struct Qdisc *sch)
 		cake_set_rate(b, rate, mtu, us_to_ns(q->target),
 			      us_to_ns(q->interval));
 
-		b->tin_quantum_prio = max_t(u16, 1U, quantum1);
-		b->tin_quantum_band = max_t(u16, 1U, quantum2);
+		b->tin_quantum = max_t(u16, 1U, quantum);
 
 		/* calculate next class's parameters */
 		rate  *= 7;
 		rate >>= 3;
 
-		quantum1  *= 3;
-		quantum1 >>= 1;
-
-		quantum2  *= 7;
-		quantum2 >>= 3;
+		quantum  *= 7;
+		quantum >>= 3;
 	}
 
 	return 0;
@@ -2347,8 +2340,7 @@ static int cake_config_diffserv8(struct Qdisc *sch)
 	struct cake_sched_data *q = qdisc_priv(sch);
 	u32 mtu = psched_mtu(qdisc_dev(sch));
 	u64 rate = q->rate_bps;
-	u32 quantum1 = 256;
-	u32 quantum2 = 256;
+	u32 quantum = 256;
 	u32 i;
 
 	q->tin_cnt = 8;
@@ -2364,18 +2356,14 @@ static int cake_config_diffserv8(struct Qdisc *sch)
 		cake_set_rate(b, rate, mtu, us_to_ns(q->target),
 			      us_to_ns(q->interval));
 
-		b->tin_quantum_prio = max_t(u16, 1U, quantum1);
-		b->tin_quantum_band = max_t(u16, 1U, quantum2);
+		b->tin_quantum = max_t(u16, 1U, quantum);
 
 		/* calculate next class's parameters */
 		rate  *= 7;
 		rate >>= 3;
 
-		quantum1  *= 3;
-		quantum1 >>= 1;
-
-		quantum2  *= 7;
-		quantum2 >>= 3;
+		quantum  *= 7;
+		quantum >>= 3;
 	}
 
 	return 0;
@@ -2414,17 +2402,11 @@ static int cake_config_diffserv4(struct Qdisc *sch)
 	cake_set_rate(&q->tins[3], rate >> 2, mtu,
 		      us_to_ns(q->target), us_to_ns(q->interval));
 
-	/* priority weights */
-	q->tins[0].tin_quantum_prio = quantum;
-	q->tins[1].tin_quantum_prio = quantum >> 4;
-	q->tins[2].tin_quantum_prio = quantum << 2;
-	q->tins[3].tin_quantum_prio = quantum << 4;
-
 	/* bandwidth-sharing weights */
-	q->tins[0].tin_quantum_band = quantum;
-	q->tins[1].tin_quantum_band = quantum >> 4;
-	q->tins[2].tin_quantum_band = quantum >> 1;
-	q->tins[3].tin_quantum_band = quantum >> 2;
+	q->tins[0].tin_quantum = quantum;
+	q->tins[1].tin_quantum = quantum >> 4;
+	q->tins[2].tin_quantum = quantum >> 1;
+	q->tins[3].tin_quantum = quantum >> 2;
 
 	return 0;
 }
@@ -2455,15 +2437,10 @@ static int cake_config_diffserv3(struct Qdisc *sch)
 	cake_set_rate(&q->tins[2], rate >> 2, mtu,
 		      us_to_ns(q->target), us_to_ns(q->interval));
 
-	/* priority weights */
-	q->tins[0].tin_quantum_prio = quantum;
-	q->tins[1].tin_quantum_prio = quantum >> 4;
-	q->tins[2].tin_quantum_prio = quantum << 4;
-
 	/* bandwidth-sharing weights */
-	q->tins[0].tin_quantum_band = quantum;
-	q->tins[1].tin_quantum_band = quantum >> 4;
-	q->tins[2].tin_quantum_band = quantum >> 2;
+	q->tins[0].tin_quantum = quantum;
+	q->tins[1].tin_quantum = quantum >> 4;
+	q->tins[2].tin_quantum = quantum >> 2;
 
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 9586a992fb752b14ac18fc86fe086b0e3372fff4 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@mellanox.com>
Date: Wed, 18 Dec 2019 14:55:08 +0000
Subject: net: pkt_cls: Clarify a comment

The bit about negating HW backlog left me scratching my head. Clarify the
comment.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/pkt_cls.h | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h
index e553fc80eb23..a7c5d492bc04 100644
--- a/include/net/pkt_cls.h
+++ b/include/net/pkt_cls.h
@@ -791,9 +791,8 @@ enum tc_prio_command {
 struct tc_prio_qopt_offload_params {
 	int bands;
 	u8 priomap[TC_PRIO_MAX + 1];
-	/* In case that a prio qdisc is offloaded and now is changed to a
-	 * non-offloadedable config, it needs to update the backlog & qlen
-	 * values to negate the HW backlog & qlen values (and only them).
+	/* At the point of un-offloading the Qdisc, the reported backlog and
+	 * qlen need to be reduced by the portion that is in HW.
 	 */
 	struct gnet_stats_queue *qstats;
 };
-- 
cgit v1.2.3-59-g8ed1b


From 5bc146c90e9eed409210809fae3c0efebef68ae1 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@mellanox.com>
Date: Wed, 18 Dec 2019 14:55:10 +0000
Subject: mlxsw: spectrum_qdisc: Clarify a comment

Expand the comment at mlxsw_sp_qdisc_prio_graft() to make the problem that
this function is trying to handle clearer.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../net/ethernet/mellanox/mlxsw/spectrum_qdisc.c   | 31 +++++++++++++++++-----
 1 file changed, 24 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c
index 68cc6737d45c..135fef6c54b1 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c
@@ -631,10 +631,30 @@ static struct mlxsw_sp_qdisc_ops mlxsw_sp_qdisc_ops_prio = {
 	.clean_stats = mlxsw_sp_setup_tc_qdisc_prio_clean_stats,
 };
 
-/* Grafting is not supported in mlxsw. It will result in un-offloading of the
- * grafted qdisc as well as the qdisc in the qdisc new location.
- * (However, if the graft is to the location where the qdisc is already at, it
- * will be ignored completely and won't cause un-offloading).
+/* Linux allows linking of Qdiscs to arbitrary classes (so long as the resulting
+ * graph is free of cycles). These operations do not change the parent handle
+ * though, which means it can be incomplete (if there is more than one class
+ * where the Qdisc in question is grafted) or outright wrong (if the Qdisc was
+ * linked to a different class and then removed from the original class).
+ *
+ * E.g. consider this sequence of operations:
+ *
+ *  # tc qdisc add dev swp1 root handle 1: prio
+ *  # tc qdisc add dev swp1 parent 1:3 handle 13: red limit 1000000 avpkt 10000
+ *  RED: set bandwidth to 10Mbit
+ *  # tc qdisc link dev swp1 handle 13: parent 1:2
+ *
+ * At this point, both 1:2 and 1:3 have the same RED Qdisc instance as their
+ * child. But RED will still only claim that 1:3 is its parent. If it's removed
+ * from that band, its only parent will be 1:2, but it will continue to claim
+ * that it is in fact 1:3.
+ *
+ * The notification for child Qdisc replace (e.g. TC_RED_REPLACE) comes before
+ * the notification for parent graft (e.g. TC_PRIO_GRAFT). We take the replace
+ * notification to offload the child Qdisc, based on its parent handle, and use
+ * the graft operation to validate that the class where the child is actually
+ * grafted corresponds to the parent handle. If the two don't match, we
+ * unoffload the child.
  */
 static int
 mlxsw_sp_qdisc_prio_graft(struct mlxsw_sp_port *mlxsw_sp_port,
@@ -644,9 +664,6 @@ mlxsw_sp_qdisc_prio_graft(struct mlxsw_sp_port *mlxsw_sp_port,
 	int tclass_num = MLXSW_SP_PRIO_BAND_TO_TCLASS(p->band);
 	struct mlxsw_sp_qdisc *old_qdisc;
 
-	/* Check if the grafted qdisc is already in its "new" location. If so -
-	 * nothing needs to be done.
-	 */
 	if (p->band < IEEE_8021QAZ_MAX_TCS &&
 	    mlxsw_sp_port->tclass_qdiscs[tclass_num].handle == p->child_handle)
 		return 0;
-- 
cgit v1.2.3-59-g8ed1b


From 9cf9b925d5387b501732dcbc8fe3c77104d423f8 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@mellanox.com>
Date: Wed, 18 Dec 2019 14:55:11 +0000
Subject: mlxsw: spectrum: Rename MLXSW_REG_QEEC_HIERARCY_* enumerators

These enums want to be named MLXSW_REG_QEEC_HIERARCHY_, but due to a typo
lack the second H. That is confusing and complicates searching.

But actually the enumerators should be named _HR_, because that is how
their enum type is called. So rename them as appropriate.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/reg.h          | 11 +++++------
 drivers/net/ethernet/mellanox/mlxsw/spectrum.c     | 19 +++++++++----------
 drivers/net/ethernet/mellanox/mlxsw/spectrum_dcb.c |  8 ++++----
 3 files changed, 18 insertions(+), 20 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h
index 5294a1622643..86a2d575ae73 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/reg.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h
@@ -3477,10 +3477,10 @@ MLXSW_REG_DEFINE(qeec, MLXSW_REG_QEEC_ID, MLXSW_REG_QEEC_LEN);
 MLXSW_ITEM32(reg, qeec, local_port, 0x00, 16, 8);
 
 enum mlxsw_reg_qeec_hr {
-	MLXSW_REG_QEEC_HIERARCY_PORT,
-	MLXSW_REG_QEEC_HIERARCY_GROUP,
-	MLXSW_REG_QEEC_HIERARCY_SUBGROUP,
-	MLXSW_REG_QEEC_HIERARCY_TC,
+	MLXSW_REG_QEEC_HR_PORT,
+	MLXSW_REG_QEEC_HR_GROUP,
+	MLXSW_REG_QEEC_HR_SUBGROUP,
+	MLXSW_REG_QEEC_HR_TC,
 };
 
 /* reg_qeec_element_hierarchy
@@ -3618,8 +3618,7 @@ static inline void mlxsw_reg_qeec_ptps_pack(char *payload, u8 local_port,
 {
 	MLXSW_REG_ZERO(qeec, payload);
 	mlxsw_reg_qeec_local_port_set(payload, local_port);
-	mlxsw_reg_qeec_element_hierarchy_set(payload,
-					     MLXSW_REG_QEEC_HIERARCY_PORT);
+	mlxsw_reg_qeec_element_hierarchy_set(payload, MLXSW_REG_QEEC_HR_PORT);
 	mlxsw_reg_qeec_ptps_set(payload, ptps);
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index 556dca328bb5..0d8fce749248 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -3602,26 +3602,25 @@ static int mlxsw_sp_port_ets_init(struct mlxsw_sp_port *mlxsw_sp_port)
 	 * one subgroup, which are all member in the same group.
 	 */
 	err = mlxsw_sp_port_ets_set(mlxsw_sp_port,
-				    MLXSW_REG_QEEC_HIERARCY_GROUP, 0, 0, false,
-				    0);
+				    MLXSW_REG_QEEC_HR_GROUP, 0, 0, false, 0);
 	if (err)
 		return err;
 	for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) {
 		err = mlxsw_sp_port_ets_set(mlxsw_sp_port,
-					    MLXSW_REG_QEEC_HIERARCY_SUBGROUP, i,
+					    MLXSW_REG_QEEC_HR_SUBGROUP, i,
 					    0, false, 0);
 		if (err)
 			return err;
 	}
 	for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) {
 		err = mlxsw_sp_port_ets_set(mlxsw_sp_port,
-					    MLXSW_REG_QEEC_HIERARCY_TC, i, i,
+					    MLXSW_REG_QEEC_HR_TC, i, i,
 					    false, 0);
 		if (err)
 			return err;
 
 		err = mlxsw_sp_port_ets_set(mlxsw_sp_port,
-					    MLXSW_REG_QEEC_HIERARCY_TC,
+					    MLXSW_REG_QEEC_HR_TC,
 					    i + 8, i,
 					    true, 100);
 		if (err)
@@ -3633,13 +3632,13 @@ static int mlxsw_sp_port_ets_init(struct mlxsw_sp_port *mlxsw_sp_port)
 	 * for the initial configuration.
 	 */
 	err = mlxsw_sp_port_ets_maxrate_set(mlxsw_sp_port,
-					    MLXSW_REG_QEEC_HIERARCY_PORT, 0, 0,
+					    MLXSW_REG_QEEC_HR_PORT, 0, 0,
 					    MLXSW_REG_QEEC_MAS_DIS);
 	if (err)
 		return err;
 	for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) {
 		err = mlxsw_sp_port_ets_maxrate_set(mlxsw_sp_port,
-						    MLXSW_REG_QEEC_HIERARCY_SUBGROUP,
+						    MLXSW_REG_QEEC_HR_SUBGROUP,
 						    i, 0,
 						    MLXSW_REG_QEEC_MAS_DIS);
 		if (err)
@@ -3647,14 +3646,14 @@ static int mlxsw_sp_port_ets_init(struct mlxsw_sp_port *mlxsw_sp_port)
 	}
 	for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) {
 		err = mlxsw_sp_port_ets_maxrate_set(mlxsw_sp_port,
-						    MLXSW_REG_QEEC_HIERARCY_TC,
+						    MLXSW_REG_QEEC_HR_TC,
 						    i, i,
 						    MLXSW_REG_QEEC_MAS_DIS);
 		if (err)
 			return err;
 
 		err = mlxsw_sp_port_ets_maxrate_set(mlxsw_sp_port,
-						    MLXSW_REG_QEEC_HIERARCY_TC,
+						    MLXSW_REG_QEEC_HR_TC,
 						    i + 8, i,
 						    MLXSW_REG_QEEC_MAS_DIS);
 		if (err)
@@ -3664,7 +3663,7 @@ static int mlxsw_sp_port_ets_init(struct mlxsw_sp_port *mlxsw_sp_port)
 	/* Configure the min shaper for multicast TCs. */
 	for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) {
 		err = mlxsw_sp_port_min_bw_set(mlxsw_sp_port,
-					       MLXSW_REG_QEEC_HIERARCY_TC,
+					       MLXSW_REG_QEEC_HR_TC,
 					       i + 8, i,
 					       MLXSW_REG_QEEC_MIS_MIN);
 		if (err)
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_dcb.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_dcb.c
index 21296fa7f7fb..fe3bbba90659 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_dcb.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_dcb.c
@@ -160,7 +160,7 @@ static int __mlxsw_sp_dcbnl_ieee_setets(struct mlxsw_sp_port *mlxsw_sp_port,
 		u8 weight = ets->tc_tx_bw[i];
 
 		err = mlxsw_sp_port_ets_set(mlxsw_sp_port,
-					    MLXSW_REG_QEEC_HIERARCY_SUBGROUP, i,
+					    MLXSW_REG_QEEC_HR_SUBGROUP, i,
 					    0, dwrr, weight);
 		if (err) {
 			netdev_err(dev, "Failed to link subgroup ETS element %d to group\n",
@@ -198,7 +198,7 @@ err_port_ets_set:
 		u8 weight = my_ets->tc_tx_bw[i];
 
 		err = mlxsw_sp_port_ets_set(mlxsw_sp_port,
-					    MLXSW_REG_QEEC_HIERARCY_SUBGROUP, i,
+					    MLXSW_REG_QEEC_HR_SUBGROUP, i,
 					    0, dwrr, weight);
 	}
 	return err;
@@ -507,7 +507,7 @@ static int mlxsw_sp_dcbnl_ieee_setmaxrate(struct net_device *dev,
 
 	for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) {
 		err = mlxsw_sp_port_ets_maxrate_set(mlxsw_sp_port,
-						    MLXSW_REG_QEEC_HIERARCY_SUBGROUP,
+						    MLXSW_REG_QEEC_HR_SUBGROUP,
 						    i, 0,
 						    maxrate->tc_maxrate[i]);
 		if (err) {
@@ -523,7 +523,7 @@ static int mlxsw_sp_dcbnl_ieee_setmaxrate(struct net_device *dev,
 err_port_ets_maxrate_set:
 	for (i--; i >= 0; i--)
 		mlxsw_sp_port_ets_maxrate_set(mlxsw_sp_port,
-					      MLXSW_REG_QEEC_HIERARCY_SUBGROUP,
+					      MLXSW_REG_QEEC_HR_SUBGROUP,
 					      i, 0, my_maxrate->tc_maxrate[i]);
 	return err;
 }
-- 
cgit v1.2.3-59-g8ed1b


From dcc68b4d8084e1ac9af0d4022d6b1aff6a139a33 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@mellanox.com>
Date: Wed, 18 Dec 2019 14:55:13 +0000
Subject: net: sch_ets: Add a new Qdisc

Introduces a new Qdisc, which is based on 802.1Q-2014 wording. It is
PRIO-like in how it is configured, meaning one needs to specify how many
bands there are, how many are strict and how many are dwrr, quanta for the
latter, and priomap.

The new Qdisc operates like the PRIO / DRR combo would when configured as
per the standard. The strict classes, if any, are tried for traffic first.
When there's no traffic in any of the strict queues, the ETS ones (if any)
are treated in the same way as in DRR.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/uapi/linux/pkt_sched.h |  17 +
 net/sched/Kconfig              |  17 +
 net/sched/Makefile             |   1 +
 net/sched/sch_ets.c            | 733 +++++++++++++++++++++++++++++++++++++++++
 4 files changed, 768 insertions(+)
 create mode 100644 net/sched/sch_ets.c

diff --git a/include/uapi/linux/pkt_sched.h b/include/uapi/linux/pkt_sched.h
index 9f1a72876212..bf5a5b1dfb0b 100644
--- a/include/uapi/linux/pkt_sched.h
+++ b/include/uapi/linux/pkt_sched.h
@@ -1187,4 +1187,21 @@ enum {
 
 #define TCA_TAPRIO_ATTR_MAX (__TCA_TAPRIO_ATTR_MAX - 1)
 
+/* ETS */
+
+#define TCQ_ETS_MAX_BANDS 16
+
+enum {
+	TCA_ETS_UNSPEC,
+	TCA_ETS_NBANDS,		/* u8 */
+	TCA_ETS_NSTRICT,	/* u8 */
+	TCA_ETS_QUANTA,		/* nested TCA_ETS_QUANTA_BAND */
+	TCA_ETS_QUANTA_BAND,	/* u32 */
+	TCA_ETS_PRIOMAP,	/* nested TCA_ETS_PRIOMAP_BAND */
+	TCA_ETS_PRIOMAP_BAND,	/* u8 */
+	__TCA_ETS_MAX,
+};
+
+#define TCA_ETS_MAX (__TCA_ETS_MAX - 1)
+
 #endif
diff --git a/net/sched/Kconfig b/net/sched/Kconfig
index 2985509147a2..b1e7ec726958 100644
--- a/net/sched/Kconfig
+++ b/net/sched/Kconfig
@@ -409,6 +409,23 @@ config NET_SCH_PLUG
 	  To compile this code as a module, choose M here: the
 	  module will be called sch_plug.
 
+config NET_SCH_ETS
+	tristate "Enhanced transmission selection scheduler (ETS)"
+	help
+          The Enhanced Transmission Selection scheduler is a classful
+          queuing discipline that merges functionality of PRIO and DRR
+          qdiscs in one scheduler. ETS makes it easy to configure a set of
+          strict and bandwidth-sharing bands to implement the transmission
+          selection described in 802.1Qaz.
+
+	  Say Y here if you want to use the ETS packet scheduling
+	  algorithm.
+
+	  To compile this driver as a module, choose M here: the module
+	  will be called sch_ets.
+
+	  If unsure, say N.
+
 menuconfig NET_SCH_DEFAULT
 	bool "Allow override default queue discipline"
 	---help---
diff --git a/net/sched/Makefile b/net/sched/Makefile
index 415d1e1f237e..bc8856b865ff 100644
--- a/net/sched/Makefile
+++ b/net/sched/Makefile
@@ -48,6 +48,7 @@ obj-$(CONFIG_NET_SCH_ATM)	+= sch_atm.o
 obj-$(CONFIG_NET_SCH_NETEM)	+= sch_netem.o
 obj-$(CONFIG_NET_SCH_DRR)	+= sch_drr.o
 obj-$(CONFIG_NET_SCH_PLUG)	+= sch_plug.o
+obj-$(CONFIG_NET_SCH_ETS)	+= sch_ets.o
 obj-$(CONFIG_NET_SCH_MQPRIO)	+= sch_mqprio.o
 obj-$(CONFIG_NET_SCH_SKBPRIO)	+= sch_skbprio.o
 obj-$(CONFIG_NET_SCH_CHOKE)	+= sch_choke.o
diff --git a/net/sched/sch_ets.c b/net/sched/sch_ets.c
new file mode 100644
index 000000000000..e6194b23e9b0
--- /dev/null
+++ b/net/sched/sch_ets.c
@@ -0,0 +1,733 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * net/sched/sch_ets.c         Enhanced Transmission Selection scheduler
+ *
+ * Description
+ * -----------
+ *
+ * The Enhanced Transmission Selection scheduler is a classful queuing
+ * discipline that merges functionality of PRIO and DRR qdiscs in one scheduler.
+ * ETS makes it easy to configure a set of strict and bandwidth-sharing bands to
+ * implement the transmission selection described in 802.1Qaz.
+ *
+ * Although ETS is technically classful, it's not possible to add and remove
+ * classes at will. Instead one specifies number of classes, how many are
+ * PRIO-like and how many DRR-like, and quanta for the latter.
+ *
+ * Algorithm
+ * ---------
+ *
+ * The strict classes, if any, are tried for traffic first: first band 0, if it
+ * has no traffic then band 1, etc.
+ *
+ * When there is no traffic in any of the strict queues, the bandwidth-sharing
+ * ones are tried next. Each band is assigned a deficit counter, initialized to
+ * "quantum" of that band. ETS maintains a list of active bandwidth-sharing
+ * bands whose qdiscs are non-empty. A packet is dequeued from the band at the
+ * head of the list if the packet size is smaller or equal to the deficit
+ * counter. If the counter is too small, it is increased by "quantum" and the
+ * scheduler moves on to the next band in the active list.
+ */
+
+#include <linux/module.h>
+#include <net/gen_stats.h>
+#include <net/netlink.h>
+#include <net/pkt_cls.h>
+#include <net/pkt_sched.h>
+#include <net/sch_generic.h>
+
+struct ets_class {
+	struct list_head alist; /* In struct ets_sched.active. */
+	struct Qdisc *qdisc;
+	u32 quantum;
+	u32 deficit;
+	struct gnet_stats_basic_packed bstats;
+	struct gnet_stats_queue qstats;
+};
+
+struct ets_sched {
+	struct list_head active;
+	struct tcf_proto __rcu *filter_list;
+	struct tcf_block *block;
+	unsigned int nbands;
+	unsigned int nstrict;
+	u8 prio2band[TC_PRIO_MAX + 1];
+	struct ets_class classes[TCQ_ETS_MAX_BANDS];
+};
+
+static const struct nla_policy ets_policy[TCA_ETS_MAX + 1] = {
+	[TCA_ETS_NBANDS] = { .type = NLA_U8 },
+	[TCA_ETS_NSTRICT] = { .type = NLA_U8 },
+	[TCA_ETS_QUANTA] = { .type = NLA_NESTED },
+	[TCA_ETS_PRIOMAP] = { .type = NLA_NESTED },
+};
+
+static const struct nla_policy ets_priomap_policy[TCA_ETS_MAX + 1] = {
+	[TCA_ETS_PRIOMAP_BAND] = { .type = NLA_U8 },
+};
+
+static const struct nla_policy ets_quanta_policy[TCA_ETS_MAX + 1] = {
+	[TCA_ETS_QUANTA_BAND] = { .type = NLA_U32 },
+};
+
+static const struct nla_policy ets_class_policy[TCA_ETS_MAX + 1] = {
+	[TCA_ETS_QUANTA_BAND] = { .type = NLA_U32 },
+};
+
+static int ets_quantum_parse(struct Qdisc *sch, const struct nlattr *attr,
+			     unsigned int *quantum,
+			     struct netlink_ext_ack *extack)
+{
+	*quantum = nla_get_u32(attr);
+	if (!*quantum) {
+		NL_SET_ERR_MSG(extack, "ETS quantum cannot be zero");
+		return -EINVAL;
+	}
+	return 0;
+}
+
+static struct ets_class *
+ets_class_from_arg(struct Qdisc *sch, unsigned long arg)
+{
+	struct ets_sched *q = qdisc_priv(sch);
+
+	return &q->classes[arg - 1];
+}
+
+static u32 ets_class_id(struct Qdisc *sch, const struct ets_class *cl)
+{
+	struct ets_sched *q = qdisc_priv(sch);
+	int band = cl - q->classes;
+
+	return TC_H_MAKE(sch->handle, band + 1);
+}
+
+static bool ets_class_is_strict(struct ets_sched *q, const struct ets_class *cl)
+{
+	unsigned int band = cl - q->classes;
+
+	return band < q->nstrict;
+}
+
+static int ets_class_change(struct Qdisc *sch, u32 classid, u32 parentid,
+			    struct nlattr **tca, unsigned long *arg,
+			    struct netlink_ext_ack *extack)
+{
+	struct ets_class *cl = ets_class_from_arg(sch, *arg);
+	struct ets_sched *q = qdisc_priv(sch);
+	struct nlattr *opt = tca[TCA_OPTIONS];
+	struct nlattr *tb[TCA_ETS_MAX + 1];
+	unsigned int quantum;
+	int err;
+
+	/* Classes can be added and removed only through Qdisc_ops.change
+	 * interface.
+	 */
+	if (!cl) {
+		NL_SET_ERR_MSG(extack, "Fine-grained class addition and removal is not supported");
+		return -EOPNOTSUPP;
+	}
+
+	if (!opt) {
+		NL_SET_ERR_MSG(extack, "ETS options are required for this operation");
+		return -EINVAL;
+	}
+
+	err = nla_parse_nested(tb, TCA_ETS_MAX, opt, ets_class_policy, extack);
+	if (err < 0)
+		return err;
+
+	if (!tb[TCA_ETS_QUANTA_BAND])
+		/* Nothing to configure. */
+		return 0;
+
+	if (ets_class_is_strict(q, cl)) {
+		NL_SET_ERR_MSG(extack, "Strict bands do not have a configurable quantum");
+		return -EINVAL;
+	}
+
+	err = ets_quantum_parse(sch, tb[TCA_ETS_QUANTA_BAND], &quantum,
+				extack);
+	if (err)
+		return err;
+
+	sch_tree_lock(sch);
+	cl->quantum = quantum;
+	sch_tree_unlock(sch);
+	return 0;
+}
+
+static int ets_class_graft(struct Qdisc *sch, unsigned long arg,
+			   struct Qdisc *new, struct Qdisc **old,
+			   struct netlink_ext_ack *extack)
+{
+	struct ets_class *cl = ets_class_from_arg(sch, arg);
+
+	if (!new) {
+		new = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops,
+					ets_class_id(sch, cl), NULL);
+		if (!new)
+			new = &noop_qdisc;
+		else
+			qdisc_hash_add(new, true);
+	}
+
+	*old = qdisc_replace(sch, new, &cl->qdisc);
+	return 0;
+}
+
+static struct Qdisc *ets_class_leaf(struct Qdisc *sch, unsigned long arg)
+{
+	struct ets_class *cl = ets_class_from_arg(sch, arg);
+
+	return cl->qdisc;
+}
+
+static unsigned long ets_class_find(struct Qdisc *sch, u32 classid)
+{
+	unsigned long band = TC_H_MIN(classid);
+	struct ets_sched *q = qdisc_priv(sch);
+
+	if (band - 1 >= q->nbands)
+		return 0;
+	return band;
+}
+
+static void ets_class_qlen_notify(struct Qdisc *sch, unsigned long arg)
+{
+	struct ets_class *cl = ets_class_from_arg(sch, arg);
+	struct ets_sched *q = qdisc_priv(sch);
+
+	/* We get notified about zero-length child Qdiscs as well if they are
+	 * offloaded. Those aren't on the active list though, so don't attempt
+	 * to remove them.
+	 */
+	if (!ets_class_is_strict(q, cl) && sch->q.qlen)
+		list_del(&cl->alist);
+}
+
+static int ets_class_dump(struct Qdisc *sch, unsigned long arg,
+			  struct sk_buff *skb, struct tcmsg *tcm)
+{
+	struct ets_class *cl = ets_class_from_arg(sch, arg);
+	struct ets_sched *q = qdisc_priv(sch);
+	struct nlattr *nest;
+
+	tcm->tcm_parent = TC_H_ROOT;
+	tcm->tcm_handle = ets_class_id(sch, cl);
+	tcm->tcm_info = cl->qdisc->handle;
+
+	nest = nla_nest_start_noflag(skb, TCA_OPTIONS);
+	if (!nest)
+		goto nla_put_failure;
+	if (!ets_class_is_strict(q, cl)) {
+		if (nla_put_u32(skb, TCA_ETS_QUANTA_BAND, cl->quantum))
+			goto nla_put_failure;
+	}
+	return nla_nest_end(skb, nest);
+
+nla_put_failure:
+	nla_nest_cancel(skb, nest);
+	return -EMSGSIZE;
+}
+
+static int ets_class_dump_stats(struct Qdisc *sch, unsigned long arg,
+				struct gnet_dump *d)
+{
+	struct ets_class *cl = ets_class_from_arg(sch, arg);
+	struct Qdisc *cl_q = cl->qdisc;
+
+	if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch),
+				  d, NULL, &cl_q->bstats) < 0 ||
+	    qdisc_qstats_copy(d, cl_q) < 0)
+		return -1;
+
+	return 0;
+}
+
+static void ets_qdisc_walk(struct Qdisc *sch, struct qdisc_walker *arg)
+{
+	struct ets_sched *q = qdisc_priv(sch);
+	int i;
+
+	if (arg->stop)
+		return;
+
+	for (i = 0; i < q->nbands; i++) {
+		if (arg->count < arg->skip) {
+			arg->count++;
+			continue;
+		}
+		if (arg->fn(sch, i + 1, arg) < 0) {
+			arg->stop = 1;
+			break;
+		}
+		arg->count++;
+	}
+}
+
+static struct tcf_block *
+ets_qdisc_tcf_block(struct Qdisc *sch, unsigned long cl,
+		    struct netlink_ext_ack *extack)
+{
+	struct ets_sched *q = qdisc_priv(sch);
+
+	if (cl) {
+		NL_SET_ERR_MSG(extack, "ETS classid must be zero");
+		return NULL;
+	}
+
+	return q->block;
+}
+
+static unsigned long ets_qdisc_bind_tcf(struct Qdisc *sch, unsigned long parent,
+					u32 classid)
+{
+	return ets_class_find(sch, classid);
+}
+
+static void ets_qdisc_unbind_tcf(struct Qdisc *sch, unsigned long arg)
+{
+}
+
+static struct ets_class *ets_classify(struct sk_buff *skb, struct Qdisc *sch,
+				      int *qerr)
+{
+	struct ets_sched *q = qdisc_priv(sch);
+	u32 band = skb->priority;
+	struct tcf_result res;
+	struct tcf_proto *fl;
+	int err;
+
+	*qerr = NET_XMIT_SUCCESS | __NET_XMIT_BYPASS;
+	if (TC_H_MAJ(skb->priority) != sch->handle) {
+		fl = rcu_dereference_bh(q->filter_list);
+		err = tcf_classify(skb, fl, &res, false);
+#ifdef CONFIG_NET_CLS_ACT
+		switch (err) {
+		case TC_ACT_STOLEN:
+		case TC_ACT_QUEUED:
+		case TC_ACT_TRAP:
+			*qerr = NET_XMIT_SUCCESS | __NET_XMIT_STOLEN;
+			/* fall through */
+		case TC_ACT_SHOT:
+			return NULL;
+		}
+#endif
+		if (!fl || err < 0) {
+			if (TC_H_MAJ(band))
+				band = 0;
+			return &q->classes[q->prio2band[band & TC_PRIO_MAX]];
+		}
+		band = res.classid;
+	}
+	band = TC_H_MIN(band) - 1;
+	if (band >= q->nbands)
+		return &q->classes[q->prio2band[0]];
+	return &q->classes[band];
+}
+
+static int ets_qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch,
+			     struct sk_buff **to_free)
+{
+	unsigned int len = qdisc_pkt_len(skb);
+	struct ets_sched *q = qdisc_priv(sch);
+	struct ets_class *cl;
+	int err = 0;
+	bool first;
+
+	cl = ets_classify(skb, sch, &err);
+	if (!cl) {
+		if (err & __NET_XMIT_BYPASS)
+			qdisc_qstats_drop(sch);
+		__qdisc_drop(skb, to_free);
+		return err;
+	}
+
+	first = !cl->qdisc->q.qlen;
+	err = qdisc_enqueue(skb, cl->qdisc, to_free);
+	if (unlikely(err != NET_XMIT_SUCCESS)) {
+		if (net_xmit_drop_count(err)) {
+			cl->qstats.drops++;
+			qdisc_qstats_drop(sch);
+		}
+		return err;
+	}
+
+	if (first && !ets_class_is_strict(q, cl)) {
+		list_add_tail(&cl->alist, &q->active);
+		cl->deficit = cl->quantum;
+	}
+
+	sch->qstats.backlog += len;
+	sch->q.qlen++;
+	return err;
+}
+
+static struct sk_buff *
+ets_qdisc_dequeue_skb(struct Qdisc *sch, struct sk_buff *skb)
+{
+	qdisc_bstats_update(sch, skb);
+	qdisc_qstats_backlog_dec(sch, skb);
+	sch->q.qlen--;
+	return skb;
+}
+
+static struct sk_buff *ets_qdisc_dequeue(struct Qdisc *sch)
+{
+	struct ets_sched *q = qdisc_priv(sch);
+	struct ets_class *cl;
+	struct sk_buff *skb;
+	unsigned int band;
+	unsigned int len;
+
+	while (1) {
+		for (band = 0; band < q->nstrict; band++) {
+			cl = &q->classes[band];
+			skb = qdisc_dequeue_peeked(cl->qdisc);
+			if (skb)
+				return ets_qdisc_dequeue_skb(sch, skb);
+		}
+
+		if (list_empty(&q->active))
+			goto out;
+
+		cl = list_first_entry(&q->active, struct ets_class, alist);
+		skb = cl->qdisc->ops->peek(cl->qdisc);
+		if (!skb) {
+			qdisc_warn_nonwc(__func__, cl->qdisc);
+			goto out;
+		}
+
+		len = qdisc_pkt_len(skb);
+		if (len <= cl->deficit) {
+			cl->deficit -= len;
+			skb = qdisc_dequeue_peeked(cl->qdisc);
+			if (unlikely(!skb))
+				goto out;
+			if (cl->qdisc->q.qlen == 0)
+				list_del(&cl->alist);
+			return ets_qdisc_dequeue_skb(sch, skb);
+		}
+
+		cl->deficit += cl->quantum;
+		list_move_tail(&cl->alist, &q->active);
+	}
+out:
+	return NULL;
+}
+
+static int ets_qdisc_priomap_parse(struct nlattr *priomap_attr,
+				   unsigned int nbands, u8 *priomap,
+				   struct netlink_ext_ack *extack)
+{
+	const struct nlattr *attr;
+	int prio = 0;
+	u8 band;
+	int rem;
+	int err;
+
+	err = __nla_validate_nested(priomap_attr, TCA_ETS_MAX,
+				    ets_priomap_policy, NL_VALIDATE_STRICT,
+				    extack);
+	if (err)
+		return err;
+
+	nla_for_each_nested(attr, priomap_attr, rem) {
+		switch (nla_type(attr)) {
+		case TCA_ETS_PRIOMAP_BAND:
+			if (prio > TC_PRIO_MAX) {
+				NL_SET_ERR_MSG_MOD(extack, "Too many priorities in ETS priomap");
+				return -EINVAL;
+			}
+			band = nla_get_u8(attr);
+			if (band >= nbands) {
+				NL_SET_ERR_MSG_MOD(extack, "Invalid band number in ETS priomap");
+				return -EINVAL;
+			}
+			priomap[prio++] = band;
+			break;
+		default:
+			WARN_ON_ONCE(1); /* Validate should have caught this. */
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
+static int ets_qdisc_quanta_parse(struct Qdisc *sch, struct nlattr *quanta_attr,
+				  unsigned int nbands, unsigned int nstrict,
+				  unsigned int *quanta,
+				  struct netlink_ext_ack *extack)
+{
+	const struct nlattr *attr;
+	int band = nstrict;
+	int rem;
+	int err;
+
+	err = __nla_validate_nested(quanta_attr, TCA_ETS_MAX,
+				    ets_quanta_policy, NL_VALIDATE_STRICT,
+				    extack);
+	if (err < 0)
+		return err;
+
+	nla_for_each_nested(attr, quanta_attr, rem) {
+		switch (nla_type(attr)) {
+		case TCA_ETS_QUANTA_BAND:
+			if (band >= nbands) {
+				NL_SET_ERR_MSG_MOD(extack, "ETS quanta has more values than bands");
+				return -EINVAL;
+			}
+			err = ets_quantum_parse(sch, attr, &quanta[band++],
+						extack);
+			if (err)
+				return err;
+			break;
+		default:
+			WARN_ON_ONCE(1); /* Validate should have caught this. */
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
+static int ets_qdisc_change(struct Qdisc *sch, struct nlattr *opt,
+			    struct netlink_ext_ack *extack)
+{
+	unsigned int quanta[TCQ_ETS_MAX_BANDS] = {0};
+	struct Qdisc *queues[TCQ_ETS_MAX_BANDS];
+	struct ets_sched *q = qdisc_priv(sch);
+	struct nlattr *tb[TCA_ETS_MAX + 1];
+	unsigned int oldbands = q->nbands;
+	u8 priomap[TC_PRIO_MAX + 1];
+	unsigned int nstrict = 0;
+	unsigned int nbands;
+	unsigned int i;
+	int err;
+
+	if (!opt) {
+		NL_SET_ERR_MSG(extack, "ETS options are required for this operation");
+		return -EINVAL;
+	}
+
+	err = nla_parse_nested(tb, TCA_ETS_MAX, opt, ets_policy, extack);
+	if (err < 0)
+		return err;
+
+	if (!tb[TCA_ETS_NBANDS]) {
+		NL_SET_ERR_MSG_MOD(extack, "Number of bands is a required argument");
+		return -EINVAL;
+	}
+	nbands = nla_get_u8(tb[TCA_ETS_NBANDS]);
+	if (nbands < 1 || nbands > TCQ_ETS_MAX_BANDS) {
+		NL_SET_ERR_MSG_MOD(extack, "Invalid number of bands");
+		return -EINVAL;
+	}
+	/* Unless overridden, traffic goes to the last band. */
+	memset(priomap, nbands - 1, sizeof(priomap));
+
+	if (tb[TCA_ETS_NSTRICT]) {
+		nstrict = nla_get_u8(tb[TCA_ETS_NSTRICT]);
+		if (nstrict > nbands) {
+			NL_SET_ERR_MSG_MOD(extack, "Invalid number of strict bands");
+			return -EINVAL;
+		}
+	}
+
+	if (tb[TCA_ETS_PRIOMAP]) {
+		err = ets_qdisc_priomap_parse(tb[TCA_ETS_PRIOMAP],
+					      nbands, priomap, extack);
+		if (err)
+			return err;
+	}
+
+	if (tb[TCA_ETS_QUANTA]) {
+		err = ets_qdisc_quanta_parse(sch, tb[TCA_ETS_QUANTA],
+					     nbands, nstrict, quanta, extack);
+		if (err)
+			return err;
+	}
+	/* If there are more bands than strict + quanta provided, the remaining
+	 * ones are ETS with quantum of MTU. Initialize the missing values here.
+	 */
+	for (i = nstrict; i < nbands; i++) {
+		if (!quanta[i])
+			quanta[i] = psched_mtu(qdisc_dev(sch));
+	}
+
+	/* Before commit, make sure we can allocate all new qdiscs */
+	for (i = oldbands; i < nbands; i++) {
+		queues[i] = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops,
+					      ets_class_id(sch, &q->classes[i]),
+					      extack);
+		if (!queues[i]) {
+			while (i > oldbands)
+				qdisc_put(queues[--i]);
+			return -ENOMEM;
+		}
+	}
+
+	sch_tree_lock(sch);
+
+	q->nbands = nbands;
+	q->nstrict = nstrict;
+	memcpy(q->prio2band, priomap, sizeof(priomap));
+
+	for (i = q->nbands; i < oldbands; i++)
+		qdisc_tree_flush_backlog(q->classes[i].qdisc);
+
+	for (i = 0; i < q->nbands; i++)
+		q->classes[i].quantum = quanta[i];
+
+	for (i = oldbands; i < q->nbands; i++) {
+		q->classes[i].qdisc = queues[i];
+		if (q->classes[i].qdisc != &noop_qdisc)
+			qdisc_hash_add(q->classes[i].qdisc, true);
+	}
+
+	sch_tree_unlock(sch);
+
+	for (i = q->nbands; i < oldbands; i++) {
+		qdisc_put(q->classes[i].qdisc);
+		memset(&q->classes[i], 0, sizeof(q->classes[i]));
+	}
+	return 0;
+}
+
+static int ets_qdisc_init(struct Qdisc *sch, struct nlattr *opt,
+			  struct netlink_ext_ack *extack)
+{
+	struct ets_sched *q = qdisc_priv(sch);
+	int err;
+
+	if (!opt)
+		return -EINVAL;
+
+	err = tcf_block_get(&q->block, &q->filter_list, sch, extack);
+	if (err)
+		return err;
+
+	INIT_LIST_HEAD(&q->active);
+	return ets_qdisc_change(sch, opt, extack);
+}
+
+static void ets_qdisc_reset(struct Qdisc *sch)
+{
+	struct ets_sched *q = qdisc_priv(sch);
+	int band;
+
+	for (band = q->nstrict; band < q->nbands; band++) {
+		if (q->classes[band].qdisc->q.qlen)
+			list_del(&q->classes[band].alist);
+	}
+	for (band = 0; band < q->nbands; band++)
+		qdisc_reset(q->classes[band].qdisc);
+	sch->qstats.backlog = 0;
+	sch->q.qlen = 0;
+}
+
+static void ets_qdisc_destroy(struct Qdisc *sch)
+{
+	struct ets_sched *q = qdisc_priv(sch);
+	int band;
+
+	tcf_block_put(q->block);
+	for (band = 0; band < q->nbands; band++)
+		qdisc_put(q->classes[band].qdisc);
+}
+
+static int ets_qdisc_dump(struct Qdisc *sch, struct sk_buff *skb)
+{
+	struct ets_sched *q = qdisc_priv(sch);
+	struct nlattr *opts;
+	struct nlattr *nest;
+	int band;
+	int prio;
+
+	opts = nla_nest_start_noflag(skb, TCA_OPTIONS);
+	if (!opts)
+		goto nla_err;
+
+	if (nla_put_u8(skb, TCA_ETS_NBANDS, q->nbands))
+		goto nla_err;
+
+	if (q->nstrict &&
+	    nla_put_u8(skb, TCA_ETS_NSTRICT, q->nstrict))
+		goto nla_err;
+
+	if (q->nbands > q->nstrict) {
+		nest = nla_nest_start(skb, TCA_ETS_QUANTA);
+		if (!nest)
+			goto nla_err;
+
+		for (band = q->nstrict; band < q->nbands; band++) {
+			if (nla_put_u32(skb, TCA_ETS_QUANTA_BAND,
+					q->classes[band].quantum))
+				goto nla_err;
+		}
+
+		nla_nest_end(skb, nest);
+	}
+
+	nest = nla_nest_start(skb, TCA_ETS_PRIOMAP);
+	if (!nest)
+		goto nla_err;
+
+	for (prio = 0; prio <= TC_PRIO_MAX; prio++) {
+		if (nla_put_u8(skb, TCA_ETS_PRIOMAP_BAND, q->prio2band[prio]))
+			goto nla_err;
+	}
+
+	nla_nest_end(skb, nest);
+
+	return nla_nest_end(skb, opts);
+
+nla_err:
+	nla_nest_cancel(skb, opts);
+	return -EMSGSIZE;
+}
+
+static const struct Qdisc_class_ops ets_class_ops = {
+	.change		= ets_class_change,
+	.graft		= ets_class_graft,
+	.leaf		= ets_class_leaf,
+	.find		= ets_class_find,
+	.qlen_notify	= ets_class_qlen_notify,
+	.dump		= ets_class_dump,
+	.dump_stats	= ets_class_dump_stats,
+	.walk		= ets_qdisc_walk,
+	.tcf_block	= ets_qdisc_tcf_block,
+	.bind_tcf	= ets_qdisc_bind_tcf,
+	.unbind_tcf	= ets_qdisc_unbind_tcf,
+};
+
+static struct Qdisc_ops ets_qdisc_ops __read_mostly = {
+	.cl_ops		= &ets_class_ops,
+	.id		= "ets",
+	.priv_size	= sizeof(struct ets_sched),
+	.enqueue	= ets_qdisc_enqueue,
+	.dequeue	= ets_qdisc_dequeue,
+	.peek		= qdisc_peek_dequeued,
+	.change		= ets_qdisc_change,
+	.init		= ets_qdisc_init,
+	.reset		= ets_qdisc_reset,
+	.destroy	= ets_qdisc_destroy,
+	.dump		= ets_qdisc_dump,
+	.owner		= THIS_MODULE,
+};
+
+static int __init ets_init(void)
+{
+	return register_qdisc(&ets_qdisc_ops);
+}
+
+static void __exit ets_exit(void)
+{
+	unregister_qdisc(&ets_qdisc_ops);
+}
+
+module_init(ets_init);
+module_exit(ets_exit);
+MODULE_LICENSE("GPL");
-- 
cgit v1.2.3-59-g8ed1b


From d35eb52bd2ac7557b62bda52668f2e64dde2cf90 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@mellanox.com>
Date: Wed, 18 Dec 2019 14:55:15 +0000
Subject: net: sch_ets: Make the ETS qdisc offloadable

Add hooks at appropriate points to make it possible to offload the ETS
Qdisc.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  1 +
 include/net/pkt_cls.h     | 31 ++++++++++++++++
 net/sched/sch_ets.c       | 95 +++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 127 insertions(+)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 30745068fb39..7a8ed11f5d45 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -849,6 +849,7 @@ enum tc_setup_type {
 	TC_SETUP_QDISC_GRED,
 	TC_SETUP_QDISC_TAPRIO,
 	TC_SETUP_FT,
+	TC_SETUP_QDISC_ETS,
 };
 
 /* These structures hold the attributes of bpf state that are being passed
diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h
index a7c5d492bc04..47b115e2012a 100644
--- a/include/net/pkt_cls.h
+++ b/include/net/pkt_cls.h
@@ -823,4 +823,35 @@ struct tc_root_qopt_offload {
 	bool ingress;
 };
 
+enum tc_ets_command {
+	TC_ETS_REPLACE,
+	TC_ETS_DESTROY,
+	TC_ETS_STATS,
+	TC_ETS_GRAFT,
+};
+
+struct tc_ets_qopt_offload_replace_params {
+	unsigned int bands;
+	u8 priomap[TC_PRIO_MAX + 1];
+	unsigned int quanta[TCQ_ETS_MAX_BANDS];	/* 0 for strict bands. */
+	unsigned int weights[TCQ_ETS_MAX_BANDS];
+	struct gnet_stats_queue *qstats;
+};
+
+struct tc_ets_qopt_offload_graft_params {
+	u8 band;
+	u32 child_handle;
+};
+
+struct tc_ets_qopt_offload {
+	enum tc_ets_command command;
+	u32 handle;
+	u32 parent;
+	union {
+		struct tc_ets_qopt_offload_replace_params replace_params;
+		struct tc_qopt_offload_stats stats;
+		struct tc_ets_qopt_offload_graft_params graft_params;
+	};
+};
+
 #endif
diff --git a/net/sched/sch_ets.c b/net/sched/sch_ets.c
index e6194b23e9b0..a87e9159338c 100644
--- a/net/sched/sch_ets.c
+++ b/net/sched/sch_ets.c
@@ -102,6 +102,91 @@ static u32 ets_class_id(struct Qdisc *sch, const struct ets_class *cl)
 	return TC_H_MAKE(sch->handle, band + 1);
 }
 
+static void ets_offload_change(struct Qdisc *sch)
+{
+	struct net_device *dev = qdisc_dev(sch);
+	struct ets_sched *q = qdisc_priv(sch);
+	struct tc_ets_qopt_offload qopt;
+	unsigned int w_psum_prev = 0;
+	unsigned int q_psum = 0;
+	unsigned int q_sum = 0;
+	unsigned int quantum;
+	unsigned int w_psum;
+	unsigned int weight;
+	unsigned int i;
+
+	if (!tc_can_offload(dev) || !dev->netdev_ops->ndo_setup_tc)
+		return;
+
+	qopt.command = TC_ETS_REPLACE;
+	qopt.handle = sch->handle;
+	qopt.parent = sch->parent;
+	qopt.replace_params.bands = q->nbands;
+	qopt.replace_params.qstats = &sch->qstats;
+	memcpy(&qopt.replace_params.priomap,
+	       q->prio2band, sizeof(q->prio2band));
+
+	for (i = 0; i < q->nbands; i++)
+		q_sum += q->classes[i].quantum;
+
+	for (i = 0; i < q->nbands; i++) {
+		quantum = q->classes[i].quantum;
+		q_psum += quantum;
+		w_psum = quantum ? q_psum * 100 / q_sum : 0;
+		weight = w_psum - w_psum_prev;
+		w_psum_prev = w_psum;
+
+		qopt.replace_params.quanta[i] = quantum;
+		qopt.replace_params.weights[i] = weight;
+	}
+
+	dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_QDISC_ETS, &qopt);
+}
+
+static void ets_offload_destroy(struct Qdisc *sch)
+{
+	struct net_device *dev = qdisc_dev(sch);
+	struct tc_ets_qopt_offload qopt;
+
+	if (!tc_can_offload(dev) || !dev->netdev_ops->ndo_setup_tc)
+		return;
+
+	qopt.command = TC_ETS_DESTROY;
+	qopt.handle = sch->handle;
+	qopt.parent = sch->parent;
+	dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_QDISC_ETS, &qopt);
+}
+
+static void ets_offload_graft(struct Qdisc *sch, struct Qdisc *new,
+			      struct Qdisc *old, unsigned long arg,
+			      struct netlink_ext_ack *extack)
+{
+	struct net_device *dev = qdisc_dev(sch);
+	struct tc_ets_qopt_offload qopt;
+
+	qopt.command = TC_ETS_GRAFT;
+	qopt.handle = sch->handle;
+	qopt.parent = sch->parent;
+	qopt.graft_params.band = arg - 1;
+	qopt.graft_params.child_handle = new->handle;
+
+	qdisc_offload_graft_helper(dev, sch, new, old, TC_SETUP_QDISC_ETS,
+				   &qopt, extack);
+}
+
+static int ets_offload_dump(struct Qdisc *sch)
+{
+	struct tc_ets_qopt_offload qopt;
+
+	qopt.command = TC_ETS_STATS;
+	qopt.handle = sch->handle;
+	qopt.parent = sch->parent;
+	qopt.stats.bstats = &sch->bstats;
+	qopt.stats.qstats = &sch->qstats;
+
+	return qdisc_offload_dump_helper(sch, TC_SETUP_QDISC_ETS, &qopt);
+}
+
 static bool ets_class_is_strict(struct ets_sched *q, const struct ets_class *cl)
 {
 	unsigned int band = cl - q->classes;
@@ -154,6 +239,8 @@ static int ets_class_change(struct Qdisc *sch, u32 classid, u32 parentid,
 	sch_tree_lock(sch);
 	cl->quantum = quantum;
 	sch_tree_unlock(sch);
+
+	ets_offload_change(sch);
 	return 0;
 }
 
@@ -173,6 +260,7 @@ static int ets_class_graft(struct Qdisc *sch, unsigned long arg,
 	}
 
 	*old = qdisc_replace(sch, new, &cl->qdisc);
+	ets_offload_graft(sch, new, *old, arg, extack);
 	return 0;
 }
 
@@ -589,6 +677,7 @@ static int ets_qdisc_change(struct Qdisc *sch, struct nlattr *opt,
 
 	sch_tree_unlock(sch);
 
+	ets_offload_change(sch);
 	for (i = q->nbands; i < oldbands; i++) {
 		qdisc_put(q->classes[i].qdisc);
 		memset(&q->classes[i], 0, sizeof(q->classes[i]));
@@ -633,6 +722,7 @@ static void ets_qdisc_destroy(struct Qdisc *sch)
 	struct ets_sched *q = qdisc_priv(sch);
 	int band;
 
+	ets_offload_destroy(sch);
 	tcf_block_put(q->block);
 	for (band = 0; band < q->nbands; band++)
 		qdisc_put(q->classes[band].qdisc);
@@ -645,6 +735,11 @@ static int ets_qdisc_dump(struct Qdisc *sch, struct sk_buff *skb)
 	struct nlattr *nest;
 	int band;
 	int prio;
+	int err;
+
+	err = ets_offload_dump(sch);
+	if (err)
+		return err;
 
 	opts = nla_nest_start_noflag(skb, TCA_OPTIONS);
 	if (!opts)
-- 
cgit v1.2.3-59-g8ed1b


From 7917f52ae188b87de627a892b57646abfe5a82bc Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@mellanox.com>
Date: Wed, 18 Dec 2019 14:55:17 +0000
Subject: mlxsw: spectrum_qdisc: Generalize PRIO offload to support ETS

Thanks to the similarity between PRIO and ETS it is possible to simply
reuse most of the code for offloading PRIO Qdisc. Extract the common
functionality into separate functions, making the current PRIO handlers
thin API adapters.

Extend the new functions to pass quanta for individual bands, which allows
configuring a subset of bands as WRR. Invoke mlxsw_sp_port_ets_set() as
appropriate to de/configure WRR-ness and weight of individual bands.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../net/ethernet/mellanox/mlxsw/spectrum_qdisc.c   | 104 ++++++++++++++++-----
 1 file changed, 81 insertions(+), 23 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c
index 135fef6c54b1..d513af49c0a8 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c
@@ -471,14 +471,16 @@ int mlxsw_sp_setup_tc_red(struct mlxsw_sp_port *mlxsw_sp_port,
 }
 
 static int
-mlxsw_sp_qdisc_prio_destroy(struct mlxsw_sp_port *mlxsw_sp_port,
-			    struct mlxsw_sp_qdisc *mlxsw_sp_qdisc)
+__mlxsw_sp_qdisc_ets_destroy(struct mlxsw_sp_port *mlxsw_sp_port)
 {
 	int i;
 
 	for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) {
 		mlxsw_sp_port_prio_tc_set(mlxsw_sp_port, i,
 					  MLXSW_SP_PORT_DEFAULT_TCLASS);
+		mlxsw_sp_port_ets_set(mlxsw_sp_port,
+				      MLXSW_REG_QEEC_HR_SUBGROUP,
+				      i, 0, false, 0);
 		mlxsw_sp_qdisc_destroy(mlxsw_sp_port,
 				       &mlxsw_sp_port->tclass_qdiscs[i]);
 		mlxsw_sp_port->tclass_qdiscs[i].prio_bitmap = 0;
@@ -488,36 +490,58 @@ mlxsw_sp_qdisc_prio_destroy(struct mlxsw_sp_port *mlxsw_sp_port,
 }
 
 static int
-mlxsw_sp_qdisc_prio_check_params(struct mlxsw_sp_port *mlxsw_sp_port,
-				 struct mlxsw_sp_qdisc *mlxsw_sp_qdisc,
-				 void *params)
+mlxsw_sp_qdisc_prio_destroy(struct mlxsw_sp_port *mlxsw_sp_port,
+			    struct mlxsw_sp_qdisc *mlxsw_sp_qdisc)
 {
-	struct tc_prio_qopt_offload_params *p = params;
+	return __mlxsw_sp_qdisc_ets_destroy(mlxsw_sp_port);
+}
 
-	if (p->bands > IEEE_8021QAZ_MAX_TCS)
+static int
+__mlxsw_sp_qdisc_ets_check_params(unsigned int nbands)
+{
+	if (nbands > IEEE_8021QAZ_MAX_TCS)
 		return -EOPNOTSUPP;
 
 	return 0;
 }
 
 static int
-mlxsw_sp_qdisc_prio_replace(struct mlxsw_sp_port *mlxsw_sp_port,
-			    struct mlxsw_sp_qdisc *mlxsw_sp_qdisc,
-			    void *params)
+mlxsw_sp_qdisc_prio_check_params(struct mlxsw_sp_port *mlxsw_sp_port,
+				 struct mlxsw_sp_qdisc *mlxsw_sp_qdisc,
+				 void *params)
 {
 	struct tc_prio_qopt_offload_params *p = params;
+
+	return __mlxsw_sp_qdisc_ets_check_params(p->bands);
+}
+
+static int
+__mlxsw_sp_qdisc_ets_replace(struct mlxsw_sp_port *mlxsw_sp_port,
+			     unsigned int nbands,
+			     const unsigned int *quanta,
+			     const unsigned int *weights,
+			     const u8 *priomap)
+{
 	struct mlxsw_sp_qdisc *child_qdisc;
 	int tclass, i, band, backlog;
 	u8 old_priomap;
 	int err;
 
-	for (band = 0; band < p->bands; band++) {
+	for (band = 0; band < nbands; band++) {
 		tclass = MLXSW_SP_PRIO_BAND_TO_TCLASS(band);
 		child_qdisc = &mlxsw_sp_port->tclass_qdiscs[tclass];
 		old_priomap = child_qdisc->prio_bitmap;
 		child_qdisc->prio_bitmap = 0;
+
+		err = mlxsw_sp_port_ets_set(mlxsw_sp_port,
+					    MLXSW_REG_QEEC_HR_SUBGROUP,
+					    tclass, 0, !!quanta[band],
+					    weights[band]);
+		if (err)
+			return err;
+
 		for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) {
-			if (p->priomap[i] == band) {
+			if (priomap[i] == band) {
 				child_qdisc->prio_bitmap |= BIT(i);
 				if (BIT(i) & old_priomap)
 					continue;
@@ -540,21 +564,46 @@ mlxsw_sp_qdisc_prio_replace(struct mlxsw_sp_port *mlxsw_sp_port,
 		child_qdisc = &mlxsw_sp_port->tclass_qdiscs[tclass];
 		child_qdisc->prio_bitmap = 0;
 		mlxsw_sp_qdisc_destroy(mlxsw_sp_port, child_qdisc);
+		mlxsw_sp_port_ets_set(mlxsw_sp_port,
+				      MLXSW_REG_QEEC_HR_SUBGROUP,
+				      tclass, 0, false, 0);
 	}
 	return 0;
 }
 
+static int
+mlxsw_sp_qdisc_prio_replace(struct mlxsw_sp_port *mlxsw_sp_port,
+			    struct mlxsw_sp_qdisc *mlxsw_sp_qdisc,
+			    void *params)
+{
+	struct tc_prio_qopt_offload_params *p = params;
+	unsigned int zeroes[TCQ_ETS_MAX_BANDS] = {0};
+
+	return __mlxsw_sp_qdisc_ets_replace(mlxsw_sp_port, p->bands,
+					    zeroes, zeroes, p->priomap);
+}
+
+static void
+__mlxsw_sp_qdisc_ets_unoffload(struct mlxsw_sp_port *mlxsw_sp_port,
+			       struct mlxsw_sp_qdisc *mlxsw_sp_qdisc,
+			       struct gnet_stats_queue *qstats)
+{
+	u64 backlog;
+
+	backlog = mlxsw_sp_cells_bytes(mlxsw_sp_port->mlxsw_sp,
+				       mlxsw_sp_qdisc->stats_base.backlog);
+	qstats->backlog -= backlog;
+}
+
 static void
 mlxsw_sp_qdisc_prio_unoffload(struct mlxsw_sp_port *mlxsw_sp_port,
 			      struct mlxsw_sp_qdisc *mlxsw_sp_qdisc,
 			      void *params)
 {
 	struct tc_prio_qopt_offload_params *p = params;
-	u64 backlog;
 
-	backlog = mlxsw_sp_cells_bytes(mlxsw_sp_port->mlxsw_sp,
-				       mlxsw_sp_qdisc->stats_base.backlog);
-	p->qstats->backlog -= backlog;
+	__mlxsw_sp_qdisc_ets_unoffload(mlxsw_sp_port, mlxsw_sp_qdisc,
+				       p->qstats);
 }
 
 static int
@@ -657,22 +706,22 @@ static struct mlxsw_sp_qdisc_ops mlxsw_sp_qdisc_ops_prio = {
  * unoffload the child.
  */
 static int
-mlxsw_sp_qdisc_prio_graft(struct mlxsw_sp_port *mlxsw_sp_port,
-			  struct mlxsw_sp_qdisc *mlxsw_sp_qdisc,
-			  struct tc_prio_qopt_offload_graft_params *p)
+__mlxsw_sp_qdisc_ets_graft(struct mlxsw_sp_port *mlxsw_sp_port,
+			   struct mlxsw_sp_qdisc *mlxsw_sp_qdisc,
+			   u8 band, u32 child_handle)
 {
-	int tclass_num = MLXSW_SP_PRIO_BAND_TO_TCLASS(p->band);
+	int tclass_num = MLXSW_SP_PRIO_BAND_TO_TCLASS(band);
 	struct mlxsw_sp_qdisc *old_qdisc;
 
-	if (p->band < IEEE_8021QAZ_MAX_TCS &&
-	    mlxsw_sp_port->tclass_qdiscs[tclass_num].handle == p->child_handle)
+	if (band < IEEE_8021QAZ_MAX_TCS &&
+	    mlxsw_sp_port->tclass_qdiscs[tclass_num].handle == child_handle)
 		return 0;
 
 	/* See if the grafted qdisc is already offloaded on any tclass. If so,
 	 * unoffload it.
 	 */
 	old_qdisc = mlxsw_sp_qdisc_find_by_handle(mlxsw_sp_port,
-						  p->child_handle);
+						  child_handle);
 	if (old_qdisc)
 		mlxsw_sp_qdisc_destroy(mlxsw_sp_port, old_qdisc);
 
@@ -681,6 +730,15 @@ mlxsw_sp_qdisc_prio_graft(struct mlxsw_sp_port *mlxsw_sp_port,
 	return -EOPNOTSUPP;
 }
 
+static int
+mlxsw_sp_qdisc_prio_graft(struct mlxsw_sp_port *mlxsw_sp_port,
+			  struct mlxsw_sp_qdisc *mlxsw_sp_qdisc,
+			  struct tc_prio_qopt_offload_graft_params *p)
+{
+	return __mlxsw_sp_qdisc_ets_graft(mlxsw_sp_port, mlxsw_sp_qdisc,
+					  p->band, p->child_handle);
+}
+
 int mlxsw_sp_setup_tc_prio(struct mlxsw_sp_port *mlxsw_sp_port,
 			   struct tc_prio_qopt_offload *p)
 {
-- 
cgit v1.2.3-59-g8ed1b


From 19f405b988e7bdee75e4936d27d2909c11574a7c Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@mellanox.com>
Date: Wed, 18 Dec 2019 14:55:19 +0000
Subject: mlxsw: spectrum_qdisc: Support offloading of ETS Qdisc

Handle TC_SETUP_QDISC_ETS, add a new ops structure for the ETS Qdisc.
Invoke the extended prio handlers implemented in the previous patch. For
stats ops, invoke directly the prio callbacks, which are not sensitive to
differences between PRIO and ETS.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum.c     |  2 +
 drivers/net/ethernet/mellanox/mlxsw/spectrum.h     |  2 +
 .../net/ethernet/mellanox/mlxsw/spectrum_qdisc.c   | 84 ++++++++++++++++++++++
 3 files changed, 88 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index 0d8fce749248..ea632042e609 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -1796,6 +1796,8 @@ static int mlxsw_sp_setup_tc(struct net_device *dev, enum tc_setup_type type,
 		return mlxsw_sp_setup_tc_red(mlxsw_sp_port, type_data);
 	case TC_SETUP_QDISC_PRIO:
 		return mlxsw_sp_setup_tc_prio(mlxsw_sp_port, type_data);
+	case TC_SETUP_QDISC_ETS:
+		return mlxsw_sp_setup_tc_ets(mlxsw_sp_port, type_data);
 	default:
 		return -EOPNOTSUPP;
 	}
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
index 347bec9d1ecf..948ef4720d40 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
@@ -852,6 +852,8 @@ int mlxsw_sp_setup_tc_red(struct mlxsw_sp_port *mlxsw_sp_port,
 			  struct tc_red_qopt_offload *p);
 int mlxsw_sp_setup_tc_prio(struct mlxsw_sp_port *mlxsw_sp_port,
 			   struct tc_prio_qopt_offload *p);
+int mlxsw_sp_setup_tc_ets(struct mlxsw_sp_port *mlxsw_sp_port,
+			  struct tc_ets_qopt_offload *p);
 
 /* spectrum_fid.c */
 bool mlxsw_sp_fid_is_dummy(struct mlxsw_sp *mlxsw_sp, u16 fid_index);
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c
index d513af49c0a8..81a2c087f534 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c
@@ -18,6 +18,7 @@ enum mlxsw_sp_qdisc_type {
 	MLXSW_SP_QDISC_NO_QDISC,
 	MLXSW_SP_QDISC_RED,
 	MLXSW_SP_QDISC_PRIO,
+	MLXSW_SP_QDISC_ETS,
 };
 
 struct mlxsw_sp_qdisc_ops {
@@ -680,6 +681,55 @@ static struct mlxsw_sp_qdisc_ops mlxsw_sp_qdisc_ops_prio = {
 	.clean_stats = mlxsw_sp_setup_tc_qdisc_prio_clean_stats,
 };
 
+static int
+mlxsw_sp_qdisc_ets_check_params(struct mlxsw_sp_port *mlxsw_sp_port,
+				struct mlxsw_sp_qdisc *mlxsw_sp_qdisc,
+				void *params)
+{
+	struct tc_ets_qopt_offload_replace_params *p = params;
+
+	return __mlxsw_sp_qdisc_ets_check_params(p->bands);
+}
+
+static int
+mlxsw_sp_qdisc_ets_replace(struct mlxsw_sp_port *mlxsw_sp_port,
+			   struct mlxsw_sp_qdisc *mlxsw_sp_qdisc,
+			   void *params)
+{
+	struct tc_ets_qopt_offload_replace_params *p = params;
+
+	return __mlxsw_sp_qdisc_ets_replace(mlxsw_sp_port, p->bands,
+					    p->quanta, p->weights, p->priomap);
+}
+
+static void
+mlxsw_sp_qdisc_ets_unoffload(struct mlxsw_sp_port *mlxsw_sp_port,
+			     struct mlxsw_sp_qdisc *mlxsw_sp_qdisc,
+			     void *params)
+{
+	struct tc_ets_qopt_offload_replace_params *p = params;
+
+	__mlxsw_sp_qdisc_ets_unoffload(mlxsw_sp_port, mlxsw_sp_qdisc,
+				       p->qstats);
+}
+
+static int
+mlxsw_sp_qdisc_ets_destroy(struct mlxsw_sp_port *mlxsw_sp_port,
+			   struct mlxsw_sp_qdisc *mlxsw_sp_qdisc)
+{
+	return __mlxsw_sp_qdisc_ets_destroy(mlxsw_sp_port);
+}
+
+static struct mlxsw_sp_qdisc_ops mlxsw_sp_qdisc_ops_ets = {
+	.type = MLXSW_SP_QDISC_ETS,
+	.check_params = mlxsw_sp_qdisc_ets_check_params,
+	.replace = mlxsw_sp_qdisc_ets_replace,
+	.unoffload = mlxsw_sp_qdisc_ets_unoffload,
+	.destroy = mlxsw_sp_qdisc_ets_destroy,
+	.get_stats = mlxsw_sp_qdisc_get_prio_stats,
+	.clean_stats = mlxsw_sp_setup_tc_qdisc_prio_clean_stats,
+};
+
 /* Linux allows linking of Qdiscs to arbitrary classes (so long as the resulting
  * graph is free of cycles). These operations do not change the parent handle
  * though, which means it can be incomplete (if there is more than one class
@@ -772,6 +822,40 @@ int mlxsw_sp_setup_tc_prio(struct mlxsw_sp_port *mlxsw_sp_port,
 	}
 }
 
+int mlxsw_sp_setup_tc_ets(struct mlxsw_sp_port *mlxsw_sp_port,
+			  struct tc_ets_qopt_offload *p)
+{
+	struct mlxsw_sp_qdisc *mlxsw_sp_qdisc;
+
+	mlxsw_sp_qdisc = mlxsw_sp_qdisc_find(mlxsw_sp_port, p->parent, true);
+	if (!mlxsw_sp_qdisc)
+		return -EOPNOTSUPP;
+
+	if (p->command == TC_ETS_REPLACE)
+		return mlxsw_sp_qdisc_replace(mlxsw_sp_port, p->handle,
+					      mlxsw_sp_qdisc,
+					      &mlxsw_sp_qdisc_ops_ets,
+					      &p->replace_params);
+
+	if (!mlxsw_sp_qdisc_compare(mlxsw_sp_qdisc, p->handle,
+				    MLXSW_SP_QDISC_ETS))
+		return -EOPNOTSUPP;
+
+	switch (p->command) {
+	case TC_ETS_DESTROY:
+		return mlxsw_sp_qdisc_destroy(mlxsw_sp_port, mlxsw_sp_qdisc);
+	case TC_ETS_STATS:
+		return mlxsw_sp_qdisc_get_stats(mlxsw_sp_port, mlxsw_sp_qdisc,
+						&p->stats);
+	case TC_ETS_GRAFT:
+		return __mlxsw_sp_qdisc_ets_graft(mlxsw_sp_port, mlxsw_sp_qdisc,
+						  p->graft_params.band,
+						  p->graft_params.child_handle);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
 int mlxsw_sp_tc_qdisc_init(struct mlxsw_sp_port *mlxsw_sp_port)
 {
 	struct mlxsw_sp_qdisc *mlxsw_sp_qdisc;
-- 
cgit v1.2.3-59-g8ed1b


From 4cf9b8f9929285b91f5f22652ad5175a2206303b Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@mellanox.com>
Date: Wed, 18 Dec 2019 14:55:21 +0000
Subject: selftests: forwarding: Move start_/stop_traffic from mlxsw to lib.sh

These two functions are used for starting several streams of traffic, and
then stopping them later. They will be handy for the test coverage of ETS
Qdisc. Move them from mlxsw-specific qos_lib.sh to the generic lib.sh.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/drivers/net/mlxsw/qos_lib.sh | 18 ------------------
 tools/testing/selftests/net/forwarding/lib.sh        | 18 ++++++++++++++++++
 2 files changed, 18 insertions(+), 18 deletions(-)

diff --git a/tools/testing/selftests/drivers/net/mlxsw/qos_lib.sh b/tools/testing/selftests/drivers/net/mlxsw/qos_lib.sh
index e80be65799ad..75a3fb3b5663 100644
--- a/tools/testing/selftests/drivers/net/mlxsw/qos_lib.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/qos_lib.sh
@@ -24,24 +24,6 @@ rate()
 	echo $((8 * (t1 - t0) / interval))
 }
 
-start_traffic()
-{
-	local h_in=$1; shift    # Where the traffic egresses the host
-	local sip=$1; shift
-	local dip=$1; shift
-	local dmac=$1; shift
-
-	$MZ $h_in -p 8000 -A $sip -B $dip -c 0 \
-		-a own -b $dmac -t udp -q &
-	sleep 1
-}
-
-stop_traffic()
-{
-	# Suppress noise from killing mausezahn.
-	{ kill %% && wait %%; } 2>/dev/null
-}
-
 check_rate()
 {
 	local rate=$1; shift
diff --git a/tools/testing/selftests/net/forwarding/lib.sh b/tools/testing/selftests/net/forwarding/lib.sh
index 1f64e7348f69..a0b09bb6995e 100644
--- a/tools/testing/selftests/net/forwarding/lib.sh
+++ b/tools/testing/selftests/net/forwarding/lib.sh
@@ -1065,3 +1065,21 @@ flood_test()
 	flood_unicast_test $br_port $host1_if $host2_if
 	flood_multicast_test $br_port $host1_if $host2_if
 }
+
+start_traffic()
+{
+	local h_in=$1; shift    # Where the traffic egresses the host
+	local sip=$1; shift
+	local dip=$1; shift
+	local dmac=$1; shift
+
+	$MZ $h_in -p 8000 -A $sip -B $dip -c 0 \
+		-a own -b $dmac -t udp -q &
+	sleep 1
+}
+
+stop_traffic()
+{
+	# Suppress noise from killing mausezahn.
+	{ kill %% && wait %%; } 2>/dev/null
+}
-- 
cgit v1.2.3-59-g8ed1b


From ddd3fd750ffee2c562fa88ea6ed1f90a02818303 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@mellanox.com>
Date: Wed, 18 Dec 2019 14:55:22 +0000
Subject: selftests: forwarding: sch_ets: Add test coverage for ETS Qdisc

This tests the newly-added ETS Qdisc. It runs two to three streams of
traffic, each with a different priority. ETS Qdisc is supposed to allocate
bandwidth according to the DRR algorithm and given weights. After running
the traffic for a while, counters are compared for each stream to check
that the expected ratio is in fact observed.

In order for the DRR process to kick in, a traffic bottleneck must exist in
the first place. In slow path, such bottleneck can be implemented by
wrapping the ETS Qdisc inside a TBF or other shaper. This might however
make the configuration unoffloadable. Instead, on HW datapath, the
bottleneck would be set up by lowering port speed and configuring shared
buffer suitably.

Therefore the test is structured as a core component that implements the
testing, with two wrapper scripts that implement the details of slow path
resp. fast path configuration.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../testing/selftests/drivers/net/mlxsw/qos_lib.sh |  28 ++
 .../testing/selftests/drivers/net/mlxsw/sch_ets.sh |  67 +++++
 tools/testing/selftests/net/forwarding/sch_ets.sh  |  44 +++
 .../selftests/net/forwarding/sch_ets_core.sh       | 300 +++++++++++++++++++++
 .../selftests/net/forwarding/sch_ets_tests.sh      | 227 ++++++++++++++++
 5 files changed, 666 insertions(+)
 create mode 100755 tools/testing/selftests/drivers/net/mlxsw/sch_ets.sh
 create mode 100755 tools/testing/selftests/net/forwarding/sch_ets.sh
 create mode 100644 tools/testing/selftests/net/forwarding/sch_ets_core.sh
 create mode 100644 tools/testing/selftests/net/forwarding/sch_ets_tests.sh

diff --git a/tools/testing/selftests/drivers/net/mlxsw/qos_lib.sh b/tools/testing/selftests/drivers/net/mlxsw/qos_lib.sh
index 75a3fb3b5663..a5937069ac16 100644
--- a/tools/testing/selftests/drivers/net/mlxsw/qos_lib.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/qos_lib.sh
@@ -78,3 +78,31 @@ measure_rate()
 	echo $ir $er
 	return $ret
 }
+
+bail_on_lldpad()
+{
+	if systemctl is-active --quiet lldpad; then
+
+		cat >/dev/stderr <<-EOF
+		WARNING: lldpad is running
+
+			lldpad will likely configure DCB, and this test will
+			configure Qdiscs. mlxsw does not support both at the
+			same time, one of them is arbitrarily going to overwrite
+			the other. That will cause spurious failures (or,
+			unlikely, passes) of this test.
+		EOF
+
+		if [[ -z $ALLOW_LLDPAD ]]; then
+			cat >/dev/stderr <<-EOF
+
+				If you want to run the test anyway, please set
+				an environment variable ALLOW_LLDPAD to a
+				non-empty string.
+			EOF
+			exit 1
+		else
+			return
+		fi
+	fi
+}
diff --git a/tools/testing/selftests/drivers/net/mlxsw/sch_ets.sh b/tools/testing/selftests/drivers/net/mlxsw/sch_ets.sh
new file mode 100755
index 000000000000..c9fc4d4885c1
--- /dev/null
+++ b/tools/testing/selftests/drivers/net/mlxsw/sch_ets.sh
@@ -0,0 +1,67 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+# A driver for the ETS selftest that implements testing in offloaded datapath.
+lib_dir=$(dirname $0)/../../../net/forwarding
+source $lib_dir/sch_ets_core.sh
+source $lib_dir/devlink_lib.sh
+source qos_lib.sh
+
+ALL_TESTS="
+	ping_ipv4
+	priomap_mode
+	ets_test_strict
+	ets_test_mixed
+	ets_test_dwrr
+"
+
+switch_create()
+{
+	ets_switch_create
+
+	# Create a bottleneck so that the DWRR process can kick in.
+	ethtool -s $h2 speed 1000 autoneg off
+	ethtool -s $swp2 speed 1000 autoneg off
+
+	# Set the ingress quota high and use the three egress TCs to limit the
+	# amount of traffic that is admitted to the shared buffers. This makes
+	# sure that there is always enough traffic of all types to select from
+	# for the DWRR process.
+	devlink_port_pool_th_set $swp1 0 12
+	devlink_tc_bind_pool_th_set $swp1 0 ingress 0 12
+	devlink_port_pool_th_set $swp2 4 12
+	devlink_tc_bind_pool_th_set $swp2 7 egress 4 5
+	devlink_tc_bind_pool_th_set $swp2 6 egress 4 5
+	devlink_tc_bind_pool_th_set $swp2 5 egress 4 5
+
+	# Note: sch_ets_core.sh uses VLAN ingress-qos-map to assign packet
+	# priorities at $swp1 based on their 802.1p headers. ingress-qos-map is
+	# not offloaded by mlxsw as of this writing, but the mapping used is
+	# 1:1, which is the mapping currently hard-coded by the driver.
+}
+
+switch_destroy()
+{
+	devlink_tc_bind_pool_th_restore $swp2 5 egress
+	devlink_tc_bind_pool_th_restore $swp2 6 egress
+	devlink_tc_bind_pool_th_restore $swp2 7 egress
+	devlink_port_pool_th_restore $swp2 4
+	devlink_tc_bind_pool_th_restore $swp1 0 ingress
+	devlink_port_pool_th_restore $swp1 0
+
+	ethtool -s $swp2 autoneg on
+	ethtool -s $h2 autoneg on
+
+	ets_switch_destroy
+}
+
+# Callback from sch_ets_tests.sh
+get_stats()
+{
+	local band=$1; shift
+
+	ethtool_stats_get "$h2" rx_octets_prio_$band
+}
+
+bail_on_lldpad
+ets_run
diff --git a/tools/testing/selftests/net/forwarding/sch_ets.sh b/tools/testing/selftests/net/forwarding/sch_ets.sh
new file mode 100755
index 000000000000..40e0ad1bc4f2
--- /dev/null
+++ b/tools/testing/selftests/net/forwarding/sch_ets.sh
@@ -0,0 +1,44 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+# A driver for the ETS selftest that implements testing in slowpath.
+lib_dir=.
+source sch_ets_core.sh
+
+ALL_TESTS="
+	ping_ipv4
+	priomap_mode
+	ets_test_strict
+	ets_test_mixed
+	ets_test_dwrr
+	classifier_mode
+	ets_test_strict
+	ets_test_mixed
+	ets_test_dwrr
+"
+
+switch_create()
+{
+	ets_switch_create
+
+	# Create a bottleneck so that the DWRR process can kick in.
+	tc qdisc add dev $swp2 root handle 1: tbf \
+	   rate 1Gbit burst 1Mbit latency 100ms
+	PARENT="parent 1:"
+}
+
+switch_destroy()
+{
+	ets_switch_destroy
+	tc qdisc del dev $swp2 root
+}
+
+# Callback from sch_ets_tests.sh
+get_stats()
+{
+	local stream=$1; shift
+
+	link_stats_get $h2.1$stream rx bytes
+}
+
+ets_run
diff --git a/tools/testing/selftests/net/forwarding/sch_ets_core.sh b/tools/testing/selftests/net/forwarding/sch_ets_core.sh
new file mode 100644
index 000000000000..f906fcc66572
--- /dev/null
+++ b/tools/testing/selftests/net/forwarding/sch_ets_core.sh
@@ -0,0 +1,300 @@
+# SPDX-License-Identifier: GPL-2.0
+
+# This is a template for ETS Qdisc test.
+#
+# This test sends from H1 several traffic streams with 802.1p-tagged packets.
+# The tags are used at $swp1 to prioritize the traffic. Each stream is then
+# queued at a different ETS band according to the assigned priority. After
+# runnig for a while, counters at H2 are consulted to determine whether the
+# traffic scheduling was according to the ETS configuration.
+#
+# This template is supposed to be embedded by a test driver, which implements
+# statistics collection, any HW-specific stuff, and prominently configures the
+# system to assure that there is overcommitment at $swp2. That is necessary so
+# that the ETS traffic selection algorithm kicks in and has to schedule some
+# traffic at the expense of other.
+#
+# A driver for veth-based testing is in sch_ets.sh, an example of a driver for
+# an offloaded data path is in selftests/drivers/net/mlxsw/sch_ets.sh.
+#
+# +---------------------------------------------------------------------+
+# | H1                                                                  |
+# |     + $h1.10              + $h1.11              + $h1.12            |
+# |     | 192.0.2.1/28        | 192.0.2.17/28       | 192.0.2.33/28     |
+# |     | egress-qos-map      | egress-qos-map      | egress-qos-map    |
+# |     |  0:0                |  0:1                |  0:2              |
+# |     \____________________ | ____________________/                   |
+# |                          \|/                                        |
+# |                           + $h1                                     |
+# +---------------------------|-----------------------------------------+
+#                             |
+# +---------------------------|-----------------------------------------+
+# | SW                        + $swp1                                   |
+# |                           | >1Gbps                                  |
+# |      ____________________/|\____________________                    |
+# |     /                     |                     \                   |
+# |  +--|----------------+ +--|----------------+ +--|----------------+  |
+# |  |  + $swp1.10       | |  + $swp1.11       | |  + $swp1.12       |  |
+# |  |    ingress-qos-map| |    ingress-qos-map| |    ingress-qos-map|  |
+# |  |     0:0 1:1 2:2   | |     0:0 1:1 2:2   | |     0:0 1:1 2:2   |  |
+# |  |                   | |                   | |                   |  |
+# |  |    BR10           | |    BR11           | |    BR12           |  |
+# |  |                   | |                   | |                   |  |
+# |  |  + $swp2.10       | |  + $swp2.11       | |  + $swp2.12       |  |
+# |  +--|----------------+ +--|----------------+ +--|----------------+  |
+# |     \____________________ | ____________________/                   |
+# |                          \|/                                        |
+# |                           + $swp2                                   |
+# |                           | 1Gbps (ethtool or HTB qdisc)            |
+# |                           | qdisc ets quanta $W0 $W1 $W2            |
+# |                           |           priomap 0 1 2                 |
+# +---------------------------|-----------------------------------------+
+#                             |
+# +---------------------------|-----------------------------------------+
+# | H2                        + $h2                                     |
+# |      ____________________/|\____________________                    |
+# |     /                     |                     \                   |
+# |     + $h2.10              + $h2.11              + $h2.12            |
+# |       192.0.2.2/28          192.0.2.18/28         192.0.2.34/28     |
+# +---------------------------------------------------------------------+
+
+NUM_NETIFS=4
+CHECK_TC=yes
+source $lib_dir/lib.sh
+source $lib_dir/sch_ets_tests.sh
+
+PARENT=root
+QDISC_DEV=
+
+sip()
+{
+	echo 192.0.2.$((16 * $1 + 1))
+}
+
+dip()
+{
+	echo 192.0.2.$((16 * $1 + 2))
+}
+
+# Callback from sch_ets_tests.sh
+ets_start_traffic()
+{
+	local dst_mac=$(mac_get $h2)
+	local i=$1; shift
+
+	start_traffic $h1.1$i $(sip $i) $(dip $i) $dst_mac
+}
+
+ETS_CHANGE_QDISC=
+
+priomap_mode()
+{
+	echo "Running in priomap mode"
+	ets_delete_qdisc
+	ETS_CHANGE_QDISC=ets_change_qdisc_priomap
+}
+
+classifier_mode()
+{
+	echo "Running in classifier mode"
+	ets_delete_qdisc
+	ETS_CHANGE_QDISC=ets_change_qdisc_classifier
+}
+
+ets_change_qdisc_priomap()
+{
+	local dev=$1; shift
+	local nstrict=$1; shift
+	local priomap=$1; shift
+	local quanta=("${@}")
+
+	local op=$(if [[ -n $QDISC_DEV ]]; then echo change; else echo add; fi)
+
+	tc qdisc $op dev $dev $PARENT handle 10: ets			       \
+		$(if ((nstrict)); then echo strict $nstrict; fi)	       \
+		$(if ((${#quanta[@]})); then echo quanta ${quanta[@]}; fi)     \
+		priomap $priomap
+	QDISC_DEV=$dev
+}
+
+ets_change_qdisc_classifier()
+{
+	local dev=$1; shift
+	local nstrict=$1; shift
+	local priomap=$1; shift
+	local quanta=("${@}")
+
+	local op=$(if [[ -n $QDISC_DEV ]]; then echo change; else echo add; fi)
+
+	tc qdisc $op dev $dev $PARENT handle 10: ets			       \
+		$(if ((nstrict)); then echo strict $nstrict; fi)	       \
+		$(if ((${#quanta[@]})); then echo quanta ${quanta[@]}; fi)
+
+	if [[ $op == add ]]; then
+		local prio=0
+		local band
+
+		for band in $priomap; do
+			tc filter add dev $dev parent 10: basic \
+				match "meta(priority eq $prio)" \
+				flowid 10:$((band + 1))
+			((prio++))
+		done
+	fi
+	QDISC_DEV=$dev
+}
+
+# Callback from sch_ets_tests.sh
+ets_change_qdisc()
+{
+	if [[ -z "$ETS_CHANGE_QDISC" ]]; then
+		exit 1
+	fi
+	$ETS_CHANGE_QDISC "$@"
+}
+
+ets_delete_qdisc()
+{
+	if [[ -n $QDISC_DEV ]]; then
+		tc qdisc del dev $QDISC_DEV $PARENT
+		QDISC_DEV=
+	fi
+}
+
+h1_create()
+{
+	local i;
+
+	simple_if_init $h1
+	mtu_set $h1 9900
+	for i in {0..2}; do
+		vlan_create $h1 1$i v$h1 $(sip $i)/28
+		ip link set dev $h1.1$i type vlan egress 0:$i
+	done
+}
+
+h1_destroy()
+{
+	local i
+
+	for i in {0..2}; do
+		vlan_destroy $h1 1$i
+	done
+	mtu_restore $h1
+	simple_if_fini $h1
+}
+
+h2_create()
+{
+	local i
+
+	simple_if_init $h2
+	mtu_set $h2 9900
+	for i in {0..2}; do
+		vlan_create $h2 1$i v$h2 $(dip $i)/28
+	done
+}
+
+h2_destroy()
+{
+	local i
+
+	for i in {0..2}; do
+		vlan_destroy $h2 1$i
+	done
+	mtu_restore $h2
+	simple_if_fini $h2
+}
+
+ets_switch_create()
+{
+	local i
+
+	ip link set dev $swp1 up
+	mtu_set $swp1 9900
+
+	ip link set dev $swp2 up
+	mtu_set $swp2 9900
+
+	for i in {0..2}; do
+		vlan_create $swp1 1$i
+		ip link set dev $swp1.1$i type vlan ingress 0:0 1:1 2:2
+
+		vlan_create $swp2 1$i
+
+		ip link add dev br1$i type bridge
+		ip link set dev $swp1.1$i master br1$i
+		ip link set dev $swp2.1$i master br1$i
+
+		ip link set dev br1$i up
+		ip link set dev $swp1.1$i up
+		ip link set dev $swp2.1$i up
+	done
+}
+
+ets_switch_destroy()
+{
+	local i
+
+	ets_delete_qdisc
+
+	for i in {0..2}; do
+		ip link del dev br1$i
+		vlan_destroy $swp2 1$i
+		vlan_destroy $swp1 1$i
+	done
+
+	mtu_restore $swp2
+	ip link set dev $swp2 down
+
+	mtu_restore $swp1
+	ip link set dev $swp1 down
+}
+
+setup_prepare()
+{
+	h1=${NETIFS[p1]}
+	swp1=${NETIFS[p2]}
+
+	swp2=${NETIFS[p3]}
+	h2=${NETIFS[p4]}
+
+	put=$swp2
+	hut=$h2
+
+	vrf_prepare
+
+	h1_create
+	h2_create
+	switch_create
+}
+
+cleanup()
+{
+	pre_cleanup
+
+	switch_destroy
+	h2_destroy
+	h1_destroy
+
+	vrf_cleanup
+}
+
+ping_ipv4()
+{
+	ping_test $h1.10 $(dip 0) " vlan 10"
+	ping_test $h1.11 $(dip 1) " vlan 11"
+	ping_test $h1.12 $(dip 2) " vlan 12"
+}
+
+ets_run()
+{
+	trap cleanup EXIT
+
+	setup_prepare
+	setup_wait
+
+	tests_run
+
+	exit $EXIT_STATUS
+}
diff --git a/tools/testing/selftests/net/forwarding/sch_ets_tests.sh b/tools/testing/selftests/net/forwarding/sch_ets_tests.sh
new file mode 100644
index 000000000000..3c3b204d47e8
--- /dev/null
+++ b/tools/testing/selftests/net/forwarding/sch_ets_tests.sh
@@ -0,0 +1,227 @@
+# SPDX-License-Identifier: GPL-2.0
+
+# Global interface:
+#  $put -- port under test (e.g. $swp2)
+#  get_stats($band) -- A function to collect stats for band
+#  ets_start_traffic($band) -- Start traffic for this band
+#  ets_change_qdisc($op, $dev, $nstrict, $quanta...) -- Add or change qdisc
+
+# WS describes the Qdisc configuration. It has one value per band (so the
+# number of array elements indicates the number of bands). If the value is
+# 0, it is a strict band, otherwise the it's a DRR band and the value is
+# that band's quantum.
+declare -a WS
+
+qdisc_describe()
+{
+	local nbands=${#WS[@]}
+	local nstrict=0
+	local i
+
+	for ((i = 0; i < nbands; i++)); do
+		if ((!${WS[$i]})); then
+			: $((nstrict++))
+		fi
+	done
+
+	echo -n "ets bands $nbands"
+	if ((nstrict)); then
+		echo -n " strict $nstrict"
+	fi
+	if ((nstrict < nbands)); then
+		echo -n " quanta"
+		for ((i = nstrict; i < nbands; i++)); do
+			echo -n " ${WS[$i]}"
+		done
+	fi
+}
+
+__strict_eval()
+{
+	local desc=$1; shift
+	local d=$1; shift
+	local total=$1; shift
+	local above=$1; shift
+
+	RET=0
+
+	if ((! total)); then
+		check_err 1 "No traffic observed"
+		log_test "$desc"
+		return
+	fi
+
+	local ratio=$(echo "scale=2; 100 * $d / $total" | bc -l)
+	if ((above)); then
+		test $(echo "$ratio > 95.0" | bc -l) -eq 1
+		check_err $? "Not enough traffic"
+		log_test "$desc"
+		log_info "Expected ratio >95% Measured ratio $ratio"
+	else
+		test $(echo "$ratio < 5" | bc -l) -eq 1
+		check_err $? "Too much traffic"
+		log_test "$desc"
+		log_info "Expected ratio <5% Measured ratio $ratio"
+	fi
+}
+
+strict_eval()
+{
+	__strict_eval "$@" 1
+}
+
+notraf_eval()
+{
+	__strict_eval "$@" 0
+}
+
+__ets_dwrr_test()
+{
+	local -a streams=("$@")
+
+	local low_stream=${streams[0]}
+	local seen_strict=0
+	local -a t0 t1 d
+	local stream
+	local total
+	local i
+
+	echo "Testing $(qdisc_describe), streams ${streams[@]}"
+
+	for stream in ${streams[@]}; do
+		ets_start_traffic $stream
+	done
+
+	sleep 10
+
+	t0=($(for stream in ${streams[@]}; do
+		  get_stats $stream
+	      done))
+
+	sleep 10
+
+	t1=($(for stream in ${streams[@]}; do
+		  get_stats $stream
+	      done))
+	d=($(for ((i = 0; i < ${#streams[@]}; i++)); do
+		 echo $((${t1[$i]} - ${t0[$i]}))
+	     done))
+	total=$(echo ${d[@]} | sed 's/ /+/g' | bc)
+
+	for ((i = 0; i < ${#streams[@]}; i++)); do
+		local stream=${streams[$i]}
+		if ((seen_strict)); then
+			notraf_eval "band $stream" ${d[$i]} $total
+		elif ((${WS[$stream]} == 0)); then
+			strict_eval "band $stream" ${d[$i]} $total
+			seen_strict=1
+		elif ((stream == low_stream)); then
+			# Low stream is used as DWRR evaluation reference.
+			continue
+		else
+			multipath_eval "bands $low_stream:$stream" \
+				       ${WS[$low_stream]} ${WS[$stream]} \
+				       ${d[0]} ${d[$i]}
+		fi
+	done
+
+	for stream in ${streams[@]}; do
+		stop_traffic
+	done
+}
+
+ets_dwrr_test_012()
+{
+	__ets_dwrr_test 0 1 2
+}
+
+ets_dwrr_test_01()
+{
+	__ets_dwrr_test 0 1
+}
+
+ets_dwrr_test_12()
+{
+	__ets_dwrr_test 1 2
+}
+
+ets_qdisc_setup()
+{
+	local dev=$1; shift
+	local nstrict=$1; shift
+	local -a quanta=("$@")
+
+	local ndwrr=${#quanta[@]}
+	local nbands=$((nstrict + ndwrr))
+	local nstreams=$(if ((nbands > 3)); then echo 3; else echo $nbands; fi)
+	local priomap=$(seq 0 $((nstreams - 1)))
+	local i
+
+	WS=($(
+		for ((i = 0; i < nstrict; i++)); do
+			echo 0
+		done
+		for ((i = 0; i < ndwrr; i++)); do
+			echo ${quanta[$i]}
+		done
+	))
+
+	ets_change_qdisc $dev $nstrict "$priomap" ${quanta[@]}
+}
+
+ets_set_dwrr_uniform()
+{
+	ets_qdisc_setup $put 0 3300 3300 3300
+}
+
+ets_set_dwrr_varying()
+{
+	ets_qdisc_setup $put 0 5000 3500 1500
+}
+
+ets_set_strict()
+{
+	ets_qdisc_setup $put 3
+}
+
+ets_set_mixed()
+{
+	ets_qdisc_setup $put 1 5000 2500 1500
+}
+
+ets_change_quantum()
+{
+	tc class change dev $put classid 10:2 ets quantum 8000
+	WS[1]=8000
+}
+
+ets_set_dwrr_two_bands()
+{
+	ets_qdisc_setup $put 0 5000 2500
+}
+
+ets_test_strict()
+{
+	ets_set_strict
+	ets_dwrr_test_01
+	ets_dwrr_test_12
+}
+
+ets_test_mixed()
+{
+	ets_set_mixed
+	ets_dwrr_test_01
+	ets_dwrr_test_12
+}
+
+ets_test_dwrr()
+{
+	ets_set_dwrr_uniform
+	ets_dwrr_test_012
+	ets_set_dwrr_varying
+	ets_dwrr_test_012
+	ets_change_quantum
+	ets_dwrr_test_012
+	ets_set_dwrr_two_bands
+	ets_dwrr_test_01
+}
-- 
cgit v1.2.3-59-g8ed1b


From 82c664b69c8b3729dfcf33e9abfd952dad35fcca Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@mellanox.com>
Date: Wed, 18 Dec 2019 14:55:24 +0000
Subject: selftests: qdiscs: Add test coverage for ETS Qdisc

Add TDC coverage for the new ETS Qdisc.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../selftests/tc-testing/tc-tests/qdiscs/ets.json  | 940 +++++++++++++++++++++
 1 file changed, 940 insertions(+)
 create mode 100644 tools/testing/selftests/tc-testing/tc-tests/qdiscs/ets.json

diff --git a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/ets.json b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/ets.json
new file mode 100644
index 000000000000..180593010675
--- /dev/null
+++ b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/ets.json
@@ -0,0 +1,940 @@
+[
+    {
+        "id": "e90e",
+        "name": "Add ETS qdisc using bands",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets bands 2",
+        "expExitCode": "0",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets 1: root .* bands 2",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "b059",
+        "name": "Add ETS qdisc using quanta",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets quanta 1000 900 800 700",
+        "expExitCode": "0",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets 1: root .*bands 4 quanta 1000 900 800 700",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "e8e7",
+        "name": "Add ETS qdisc using strict",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets strict 3",
+        "expExitCode": "0",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets 1: root .*bands 3 strict 3",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "233c",
+        "name": "Add ETS qdisc using bands + quanta",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets bands 4 quanta 1000 900 800 700",
+        "expExitCode": "0",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets 1: root .*bands 4 quanta 1000 900 800 700 priomap",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "3d35",
+        "name": "Add ETS qdisc using bands + strict",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets bands 3 strict 3",
+        "expExitCode": "0",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets 1: root .*bands 3 strict 3 priomap",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "7f3b",
+        "name": "Add ETS qdisc using strict + quanta",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets strict 3 quanta 1500 750",
+        "expExitCode": "0",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets 1: root .*bands 5 strict 3 quanta 1500 750 priomap",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "4593",
+        "name": "Add ETS qdisc using strict 0 + quanta",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets strict 0 quanta 1500 750",
+        "expExitCode": "0",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets 1: root .*bands 2 quanta 1500 750 priomap",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "8938",
+        "name": "Add ETS qdisc using bands + strict + quanta",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets bands 5 strict 3 quanta 1500 750",
+        "expExitCode": "0",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets 1: root .*bands 5 .*strict 3 quanta 1500 750 priomap",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "0782",
+        "name": "Add ETS qdisc with more bands than quanta",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets bands 2 quanta 1000",
+        "expExitCode": "0",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets 1: root .*bands 2 .*quanta 1000 [1-9][0-9]* priomap",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "501b",
+        "name": "Add ETS qdisc with more bands than strict",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets bands 3 strict 1",
+        "expExitCode": "0",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets 1: root .*bands 3 strict 1 quanta ([1-9][0-9]* ){2}priomap",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "671a",
+        "name": "Add ETS qdisc with more bands than strict + quanta",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets bands 3 strict 1 quanta 1000",
+        "expExitCode": "0",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets 1: root .*bands 3 strict 1 quanta 1000 [1-9][0-9]* priomap",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "2a23",
+        "name": "Add ETS qdisc with 16 bands",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets bands 16",
+        "expExitCode": "0",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets 1: root .* bands 16",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "8daf",
+        "name": "Add ETS qdisc with 17 bands",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets bands 17",
+        "expExitCode": "1",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets",
+        "matchCount": "0",
+        "teardown": [
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "7f95",
+        "name": "Add ETS qdisc with 17 strict",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets strict 17",
+        "expExitCode": "1",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets",
+        "matchCount": "0",
+        "teardown": [
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "837a",
+        "name": "Add ETS qdisc with 16 quanta",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets quanta 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16",
+        "expExitCode": "0",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets 1: root .* bands 16",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "65b6",
+        "name": "Add ETS qdisc with 17 quanta",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets quanta 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17",
+        "expExitCode": "2",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets",
+        "matchCount": "0",
+        "teardown": [
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "b9e9",
+        "name": "Add ETS qdisc with 16 strict + quanta",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets strict 8 quanta 1 2 3 4 5 6 7 8",
+        "expExitCode": "0",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets 1: root .* bands 16",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "9877",
+        "name": "Add ETS qdisc with 17 strict + quanta",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets strict 9 quanta 1 2 3 4 5 6 7 8",
+        "expExitCode": "2",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets",
+        "matchCount": "0",
+        "teardown": [
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "c696",
+        "name": "Add ETS qdisc with priomap",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets bands 5 priomap 0 0 1 0 1 2 0 1 2 3 0 1 2 3 4 0",
+        "expExitCode": "0",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets 1: root .*priomap 0 0 1 0 1 2 0 1 2 3 0 1 2 3 4 0",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "30c4",
+        "name": "Add ETS qdisc with quanta + priomap",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets quanta 1000 2000 3000 4000 5000 priomap 0 0 1 0 1 2 0 1 2 3 0 1 2 3 4 0",
+        "expExitCode": "0",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets 1: root .*quanta 1000 2000 3000 4000 5000 priomap 0 0 1 0 1 2 0 1 2 3 0 1 2 3 4 0",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "e8ac",
+        "name": "Add ETS qdisc with strict + priomap",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets strict 5 priomap 0 0 1 0 1 2 0 1 2 3 0 1 2 3 4 0",
+        "expExitCode": "0",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets 1: root .*bands 5 strict 5 priomap 0 0 1 0 1 2 0 1 2 3 0 1 2 3 4 0",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "5a7e",
+        "name": "Add ETS qdisc with quanta + strict + priomap",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets strict 2 quanta 1000 2000 3000 priomap 0 0 1 0 1 2 0 1 2 3 0 1 2 3 4 0",
+        "expExitCode": "0",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets 1: root .*strict 2 quanta 1000 2000 3000 priomap 0 0 1 0 1 2 0 1 2 3 0 1 2 3 4 0",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "cb8b",
+        "name": "Show ETS class :1",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets quanta 4000 3000 2000",
+        "expExitCode": "0",
+        "verifyCmd": "$TC class show dev $DUMMY classid 1:1",
+        "matchPattern": "class ets 1:1 root quantum 4000",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "1b4e",
+        "name": "Show ETS class :2",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets quanta 4000 3000 2000",
+        "expExitCode": "0",
+        "verifyCmd": "$TC class show dev $DUMMY classid 1:2",
+        "matchPattern": "class ets 1:2 root quantum 3000",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "f642",
+        "name": "Show ETS class :3",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets quanta 4000 3000 2000",
+        "expExitCode": "0",
+        "verifyCmd": "$TC class show dev $DUMMY classid 1:3",
+        "matchPattern": "class ets 1:3 root quantum 2000",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "0a5f",
+        "name": "Show ETS strict class",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets strict 3",
+        "expExitCode": "0",
+        "verifyCmd": "$TC class show dev $DUMMY classid 1:1",
+        "matchPattern": "class ets 1:1 root $",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "f7c8",
+        "name": "Add ETS qdisc with too many quanta",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets bands 2 quanta 1000 2000 3000",
+        "expExitCode": "1",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets",
+        "matchCount": "0",
+        "teardown": [
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "2389",
+        "name": "Add ETS qdisc with too many strict",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets bands 2 strict 3",
+        "expExitCode": "1",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets",
+        "matchCount": "0",
+        "teardown": [
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "fe3c",
+        "name": "Add ETS qdisc with too many strict + quanta",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets bands 4 strict 2 quanta 1000 2000 3000",
+        "expExitCode": "1",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets",
+        "matchCount": "0",
+        "teardown": [
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "cb04",
+        "name": "Add ETS qdisc with excess priomap elements",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets bands 5 priomap 0 0 1 0 1 2 0 1 2 3 0 1 2 3 4 0 1 2",
+        "expExitCode": "1",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets",
+        "matchCount": "0",
+        "teardown": [
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "c32e",
+        "name": "Add ETS qdisc with priomap above bands",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets bands 2 priomap 0 1 2",
+        "expExitCode": "1",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets",
+        "matchCount": "0",
+        "teardown": [
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "744c",
+        "name": "Add ETS qdisc with priomap above quanta",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets quanta 1000 500 priomap 0 1 2",
+        "expExitCode": "1",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets",
+        "matchCount": "0",
+        "teardown": [
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "7b33",
+        "name": "Add ETS qdisc with priomap above strict",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets strict 2 priomap 0 1 2",
+        "expExitCode": "1",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets",
+        "matchCount": "0",
+        "teardown": [
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "dbe6",
+        "name": "Add ETS qdisc with priomap above strict + quanta",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets strict 1 quanta 1000 500 priomap 0 1 2 3",
+        "expExitCode": "1",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets",
+        "matchCount": "0",
+        "teardown": [
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "bdb2",
+        "name": "Add ETS qdisc with priomap within bands with strict + quanta",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets bands 4 strict 1 quanta 1000 500 priomap 0 1 2 3",
+        "expExitCode": "0",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets",
+        "matchCount": "1",
+        "teardown": [
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "39a3",
+        "name": "Add ETS qdisc with priomap above bands with strict + quanta",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets bands 4 strict 1 quanta 1000 500 priomap 0 1 2 3 4",
+        "expExitCode": "1",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets",
+        "matchCount": "0",
+        "teardown": [
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "557c",
+        "name": "Unset priorities default to the last band",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets bands 4 priomap 0 0 0 0",
+        "expExitCode": "0",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets .*priomap 0 0 0 0 3 3 3 3 3 3 3 3 3 3 3 3",
+        "matchCount": "1",
+        "teardown": [
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "a347",
+        "name": "Unset priorities default to the last band -- no priomap",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets bands 4",
+        "expExitCode": "0",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets .*priomap 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3",
+        "matchCount": "1",
+        "teardown": [
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "39c4",
+        "name": "Add ETS qdisc with too few bands",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets bands 0",
+        "expExitCode": "1",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets",
+        "matchCount": "0",
+        "teardown": [
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "930b",
+        "name": "Add ETS qdisc with too many bands",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets bands 17",
+        "expExitCode": "1",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets",
+        "matchCount": "0",
+        "teardown": [
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "406a",
+        "name": "Add ETS qdisc without parameters",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets",
+        "expExitCode": "1",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets",
+        "matchCount": "0",
+        "teardown": [
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "e51a",
+        "name": "Zero element in quanta",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets quanta 1000 0 800 700",
+        "expExitCode": "1",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets",
+        "matchCount": "0",
+        "teardown": [
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "e7f2",
+        "name": "Sole zero element in quanta",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets quanta 0",
+        "expExitCode": "1",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets",
+        "matchCount": "0",
+        "teardown": [
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "d6e6",
+        "name": "No values after the quanta keyword",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root ets quanta",
+        "expExitCode": "255",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets",
+        "matchCount": "0",
+        "teardown": [
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "28c6",
+        "name": "Change ETS band quantum",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true",
+            "$TC qdisc add dev $DUMMY handle 1: root ets quanta 1000 2000 3000"
+        ],
+        "cmdUnderTest": "$TC class change dev $DUMMY classid 1:1 ets quantum 1500",
+        "expExitCode": "0",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets 1: root .*quanta 1500 2000 3000 priomap ",
+        "matchCount": "1",
+        "teardown": [
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "4714",
+        "name": "Change ETS band without quantum",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true",
+            "$TC qdisc add dev $DUMMY handle 1: root ets quanta 1000 2000 3000"
+        ],
+        "cmdUnderTest": "$TC class change dev $DUMMY classid 1:1 ets",
+        "expExitCode": "0",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets 1: root .*quanta 1000 2000 3000 priomap ",
+        "matchCount": "1",
+        "teardown": [
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "6979",
+        "name": "Change quantum of a strict ETS band",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true",
+            "$TC qdisc add dev $DUMMY handle 1: root ets strict 5"
+        ],
+        "cmdUnderTest": "$TC class change dev $DUMMY classid 1:2 ets quantum 1500",
+        "expExitCode": "2",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets .*bands 5 .*strict 5",
+        "matchCount": "1",
+        "teardown": [
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "9a7d",
+        "name": "Change ETS strict band without quantum",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true",
+            "$TC qdisc add dev $DUMMY handle 1: root ets strict 5"
+        ],
+        "cmdUnderTest": "$TC class change dev $DUMMY classid 1:2 ets",
+        "expExitCode": "0",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc ets .*bands 5 .*strict 5",
+        "matchCount": "1",
+        "teardown": [
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    }
+]
-- 
cgit v1.2.3-59-g8ed1b


From a1ec57c020201ba29608a5a3588832d07de1a518 Mon Sep 17 00:00:00 2001
From: Jose Abreu <Jose.Abreu@synopsys.com>
Date: Wed, 18 Dec 2019 23:55:01 +0100
Subject: net: stmmac: tc: Fix TAPRIO division operation

For ARCHs that don't support 64 bits division we need to use the
helpers.

Fixes: b60189e0392f ("net: stmmac: Integrate EST with TAPRIO scheduler API")
Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c
index 8ff8f9b9bb22..6c4686b77516 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c
@@ -599,6 +599,7 @@ static int tc_setup_taprio(struct stmmac_priv *priv,
 	struct timespec64 time;
 	bool fpe = false;
 	int i, ret = 0;
+	u64 ctr;
 
 	if (!priv->dma_cap.estsel)
 		return -EOPNOTSUPP;
@@ -694,8 +695,9 @@ static int tc_setup_taprio(struct stmmac_priv *priv,
 	priv->plat->est->btr[0] = (u32)time.tv_nsec;
 	priv->plat->est->btr[1] = (u32)time.tv_sec;
 
-	priv->plat->est->ctr[0] = (u32)(qopt->cycle_time % NSEC_PER_SEC);
-	priv->plat->est->ctr[1] = (u32)(qopt->cycle_time / NSEC_PER_SEC);
+	ctr = qopt->cycle_time;
+	priv->plat->est->ctr[0] = do_div(ctr, NSEC_PER_SEC);
+	priv->plat->est->ctr[1] = (u32)ctr;
 
 	if (fpe && !priv->dma_cap.fpesel)
 		return -EOPNOTSUPP;
-- 
cgit v1.2.3-59-g8ed1b


From 7c43e0d6a526e7734eb854fe242886f52ccd06ac Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Wed, 18 Dec 2019 13:43:14 -0800
Subject: bpftool: Simplify format string to not use positional args

Change format string referring to just single argument out of two available.
Some versions of libc can reject such format string.

Reported-by: Nikita Shirokov <tehnerd@tehnerd.com>
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20191218214314.2403729-1-andriin@fb.com
---
 tools/bpf/bpftool/gen.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tools/bpf/bpftool/gen.c b/tools/bpf/bpftool/gen.c
index 8d93c8f90f82..851c465f99dc 100644
--- a/tools/bpf/bpftool/gen.c
+++ b/tools/bpf/bpftool/gen.c
@@ -567,9 +567,9 @@ static int do_skeleton(int argc, char **argv)
 			return -1;					    \n\
 		}							    \n\
 									    \n\
-		#endif /* %2$s */					    \n\
+		#endif /* %s */						    \n\
 		",
-		obj_name, header_guard);
+		header_guard);
 	err = 0;
 out:
 	bpf_object__close(obj);
-- 
cgit v1.2.3-59-g8ed1b


From dacce6412e09b5dce19514e2c4e2a8aab0eb217f Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Wed, 18 Dec 2019 14:17:07 -0800
Subject: bpftool: Work-around rst2man conversion bug

Work-around what appears to be a bug in rst2man convertion tool, used to
create man pages out of reStructureText-formatted documents. If text line
starts with dot, rst2man will put it in resulting man file verbatim. This
seems to cause man tool to interpret it as a directive/command (e.g., `.bs`), and
subsequently not render entire line because it's unrecognized one.

Enclose '.xxx' words in extra formatting to work around.

Fixes: cb21ac588546 ("bpftool: Add gen subcommand manpage")
Reported-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com
Link: https://lore.kernel.org/bpf/20191218221707.2552199-1-andriin@fb.com
---
 tools/bpf/bpftool/Documentation/bpftool-gen.rst | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

diff --git a/tools/bpf/bpftool/Documentation/bpftool-gen.rst b/tools/bpf/bpftool/Documentation/bpftool-gen.rst
index b6a114bf908d..86a87da97d0b 100644
--- a/tools/bpf/bpftool/Documentation/bpftool-gen.rst
+++ b/tools/bpf/bpftool/Documentation/bpftool-gen.rst
@@ -112,13 +112,14 @@ DESCRIPTION
 
 		  If BPF object has global variables, corresponding structs
 		  with memory layout corresponding to global data data section
-		  layout will be created. Currently supported ones are: .data,
-		  .bss, .rodata, and .extern structs/data sections. These
-		  data sections/structs can be used to set up initial values of
-		  variables, if set before **example__load**. Afterwards, if
-		  target kernel supports memory-mapped BPF arrays, same
-		  structs can be used to fetch and update (non-read-only)
-		  data from userspace, with same simplicity as for BPF side.
+		  layout will be created. Currently supported ones are: *.data*,
+		  *.bss*, *.rodata*, and *.kconfig* structs/data sections.
+		  These data sections/structs can be used to set up initial
+		  values of variables, if set before **example__load**.
+		  Afterwards, if target kernel supports memory-mapped BPF
+		  arrays, same structs can be used to fetch and update
+		  (non-read-only) data from userspace, with same simplicity
+		  as for BPF side.
 
 	**bpftool gen help**
 		  Print short help message.
-- 
cgit v1.2.3-59-g8ed1b


From 6de6c1f840c051017f2308503858ff19344c56b3 Mon Sep 17 00:00:00 2001
From: "Nikita V. Shirokov" <tehnerd@tehnerd.com>
Date: Wed, 18 Dec 2019 12:57:47 -0800
Subject: bpf: Allow to change skb mark in test_run

allow to pass skb's mark field into bpf_prog_test_run ctx
for BPF_PROG_TYPE_SCHED_CLS prog type. that would allow
to test bpf programs which are doing decision based on this
field

Signed-off-by: Nikita V. Shirokov <tehnerd@tehnerd.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 net/bpf/test_run.c                               | 10 +++++++++-
 tools/testing/selftests/bpf/prog_tests/skb_ctx.c |  5 +++++
 tools/testing/selftests/bpf/progs/test_skb_ctx.c |  1 +
 3 files changed, 15 insertions(+), 1 deletion(-)

diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c
index 93a9c87787e0..d555c0d8657d 100644
--- a/net/bpf/test_run.c
+++ b/net/bpf/test_run.c
@@ -251,7 +251,13 @@ static int convert___skb_to_skb(struct sk_buff *skb, struct __sk_buff *__skb)
 		return 0;
 
 	/* make sure the fields we don't use are zeroed */
-	if (!range_is_zero(__skb, 0, offsetof(struct __sk_buff, priority)))
+	if (!range_is_zero(__skb, 0, offsetof(struct __sk_buff, mark)))
+		return -EINVAL;
+
+	/* mark is allowed */
+
+	if (!range_is_zero(__skb, offsetofend(struct __sk_buff, mark),
+			   offsetof(struct __sk_buff, priority)))
 		return -EINVAL;
 
 	/* priority is allowed */
@@ -274,6 +280,7 @@ static int convert___skb_to_skb(struct sk_buff *skb, struct __sk_buff *__skb)
 			   sizeof(struct __sk_buff)))
 		return -EINVAL;
 
+	skb->mark = __skb->mark;
 	skb->priority = __skb->priority;
 	skb->tstamp = __skb->tstamp;
 	memcpy(&cb->data, __skb->cb, QDISC_CB_PRIV_LEN);
@@ -301,6 +308,7 @@ static void convert_skb_to___skb(struct sk_buff *skb, struct __sk_buff *__skb)
 	if (!__skb)
 		return;
 
+	__skb->mark = skb->mark;
 	__skb->priority = skb->priority;
 	__skb->tstamp = skb->tstamp;
 	memcpy(__skb->cb, &cb->data, QDISC_CB_PRIV_LEN);
diff --git a/tools/testing/selftests/bpf/prog_tests/skb_ctx.c b/tools/testing/selftests/bpf/prog_tests/skb_ctx.c
index edf5e8c7d400..c6d6b685a946 100644
--- a/tools/testing/selftests/bpf/prog_tests/skb_ctx.c
+++ b/tools/testing/selftests/bpf/prog_tests/skb_ctx.c
@@ -13,6 +13,7 @@ void test_skb_ctx(void)
 		.tstamp = 7,
 		.wire_len = 100,
 		.gso_segs = 8,
+		.mark = 9,
 	};
 	struct bpf_prog_test_run_attr tattr = {
 		.data_in = &pkt_v4,
@@ -93,4 +94,8 @@ void test_skb_ctx(void)
 		   "ctx_out_tstamp",
 		   "skb->tstamp == %lld, expected %d\n",
 		   skb.tstamp, 8);
+	CHECK_ATTR(skb.mark != 10,
+		   "ctx_out_mark",
+		   "skb->mark == %u, expected %d\n",
+		   skb.mark, 10);
 }
diff --git a/tools/testing/selftests/bpf/progs/test_skb_ctx.c b/tools/testing/selftests/bpf/progs/test_skb_ctx.c
index 534fbf9a7344..e18da87fe84f 100644
--- a/tools/testing/selftests/bpf/progs/test_skb_ctx.c
+++ b/tools/testing/selftests/bpf/progs/test_skb_ctx.c
@@ -17,6 +17,7 @@ int process(struct __sk_buff *skb)
 	}
 	skb->priority++;
 	skb->tstamp++;
+	skb->mark++;
 
 	if (skb->wire_len != 100)
 		return 1;
-- 
cgit v1.2.3-59-g8ed1b


From d69587062c347314a019cf6ee27f2e4b494868e1 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Wed, 18 Dec 2019 14:50:39 -0800
Subject: libbpf: Add bpf_link__disconnect() API to preserve underlying BPF
 resource

There are cases in which BPF resource (program, map, etc) has to outlive
userspace program that "installed" it in the system in the first place.
When BPF program is attached, libbpf returns bpf_link object, which
is supposed to be destroyed after no longer necessary through
bpf_link__destroy() API. Currently, bpf_link destruction causes both automatic
detachment and frees up any resources allocated to for bpf_link in-memory
representation. This is inconvenient for the case described above because of
coupling of detachment and resource freeing.

This patch introduces bpf_link__disconnect() API call, which marks bpf_link as
disconnected from its underlying BPF resouces. This means that when bpf_link
is destroyed later, all its memory resources will be freed, but BPF resource
itself won't be detached.

This design allows to follow strict and resource-leak-free design by default,
while giving easy and straightforward way for user code to opt for keeping BPF
resource attached beyond lifetime of a bpf_link. For some BPF programs (i.e.,
FS-based tracepoints, kprobes, raw tracepoint, etc), user has to make sure to
pin BPF program to prevent kernel to automatically detach it on process exit.
This should typically be achived by pinning BPF program (or map in some cases)
in BPF FS.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20191218225039.2668205-1-andriin@fb.com
---
 tools/lib/bpf/libbpf.c   | 40 ++++++++++++++++++++++++++++++----------
 tools/lib/bpf/libbpf.h   |  1 +
 tools/lib/bpf/libbpf.map |  1 +
 3 files changed, 32 insertions(+), 10 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 906bbbf7b2e4..2a341a5b9f63 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -6245,17 +6245,37 @@ int bpf_prog_load_xattr(const struct bpf_prog_load_attr *attr,
 }
 
 struct bpf_link {
+	int (*detach)(struct bpf_link *link);
 	int (*destroy)(struct bpf_link *link);
+	bool disconnected;
 };
 
+/* Release "ownership" of underlying BPF resource (typically, BPF program
+ * attached to some BPF hook, e.g., tracepoint, kprobe, etc). Disconnected
+ * link, when destructed through bpf_link__destroy() call won't attempt to
+ * detach/unregisted that BPF resource. This is useful in situations where,
+ * say, attached BPF program has to outlive userspace program that attached it
+ * in the system. Depending on type of BPF program, though, there might be
+ * additional steps (like pinning BPF program in BPF FS) necessary to ensure
+ * exit of userspace program doesn't trigger automatic detachment and clean up
+ * inside the kernel.
+ */
+void bpf_link__disconnect(struct bpf_link *link)
+{
+	link->disconnected = true;
+}
+
 int bpf_link__destroy(struct bpf_link *link)
 {
-	int err;
+	int err = 0;
 
 	if (!link)
 		return 0;
 
-	err = link->destroy(link);
+	if (!link->disconnected && link->detach)
+		err = link->detach(link);
+	if (link->destroy)
+		link->destroy(link);
 	free(link);
 
 	return err;
@@ -6266,7 +6286,7 @@ struct bpf_link_fd {
 	int fd; /* hook FD */
 };
 
-static int bpf_link__destroy_perf_event(struct bpf_link *link)
+static int bpf_link__detach_perf_event(struct bpf_link *link)
 {
 	struct bpf_link_fd *l = (void *)link;
 	int err;
@@ -6298,10 +6318,10 @@ struct bpf_link *bpf_program__attach_perf_event(struct bpf_program *prog,
 		return ERR_PTR(-EINVAL);
 	}
 
-	link = malloc(sizeof(*link));
+	link = calloc(1, sizeof(*link));
 	if (!link)
 		return ERR_PTR(-ENOMEM);
-	link->link.destroy = &bpf_link__destroy_perf_event;
+	link->link.detach = &bpf_link__detach_perf_event;
 	link->fd = pfd;
 
 	if (ioctl(pfd, PERF_EVENT_IOC_SET_BPF, prog_fd) < 0) {
@@ -6608,7 +6628,7 @@ out:
 	return link;
 }
 
-static int bpf_link__destroy_fd(struct bpf_link *link)
+static int bpf_link__detach_fd(struct bpf_link *link)
 {
 	struct bpf_link_fd *l = (void *)link;
 
@@ -6629,10 +6649,10 @@ struct bpf_link *bpf_program__attach_raw_tracepoint(struct bpf_program *prog,
 		return ERR_PTR(-EINVAL);
 	}
 
-	link = malloc(sizeof(*link));
+	link = calloc(1, sizeof(*link));
 	if (!link)
 		return ERR_PTR(-ENOMEM);
-	link->link.destroy = &bpf_link__destroy_fd;
+	link->link.detach = &bpf_link__detach_fd;
 
 	pfd = bpf_raw_tracepoint_open(tp_name, prog_fd);
 	if (pfd < 0) {
@@ -6668,10 +6688,10 @@ struct bpf_link *bpf_program__attach_trace(struct bpf_program *prog)
 		return ERR_PTR(-EINVAL);
 	}
 
-	link = malloc(sizeof(*link));
+	link = calloc(1, sizeof(*link));
 	if (!link)
 		return ERR_PTR(-ENOMEM);
-	link->link.destroy = &bpf_link__destroy_fd;
+	link->link.detach = &bpf_link__detach_fd;
 
 	pfd = bpf_raw_tracepoint_open(NULL, prog_fd);
 	if (pfd < 0) {
diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index f7084235bae9..ad8c1c127933 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -215,6 +215,7 @@ LIBBPF_API void bpf_program__unload(struct bpf_program *prog);
 
 struct bpf_link;
 
+LIBBPF_API void bpf_link__disconnect(struct bpf_link *link);
 LIBBPF_API int bpf_link__destroy(struct bpf_link *link);
 
 LIBBPF_API struct bpf_link *
diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
index 1376d992a703..e3a471f38a71 100644
--- a/tools/lib/bpf/libbpf.map
+++ b/tools/lib/bpf/libbpf.map
@@ -212,6 +212,7 @@ LIBBPF_0.0.6 {
 LIBBPF_0.0.7 {
 	global:
 		btf_dump__emit_type_decl;
+		bpf_link__disconnect;
 		bpf_object__find_program_by_name;
 		bpf_object__attach_skeleton;
 		bpf_object__destroy_skeleton;
-- 
cgit v1.2.3-59-g8ed1b


From 81bfdd087bf31a87c5ff25cc7004d5308954a35c Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Wed, 18 Dec 2019 16:28:34 -0800
Subject: libbpf: Put Kconfig externs into .kconfig section

Move Kconfig-provided externs into custom .kconfig section. Add __kconfig into
bpf_helpers.h for user convenience. Update selftests accordingly.

Suggested-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191219002837.3074619-2-andriin@fb.com
---
 tools/bpf/bpftool/gen.c                            |  8 +--
 tools/lib/bpf/bpf_helpers.h                        |  2 +
 tools/lib/bpf/libbpf.c                             | 58 +++++++++++++---------
 tools/testing/selftests/bpf/prog_tests/skeleton.c  | 16 +++---
 .../testing/selftests/bpf/progs/test_core_extern.c | 20 ++++----
 tools/testing/selftests/bpf/progs/test_skeleton.c  |  4 +-
 6 files changed, 60 insertions(+), 48 deletions(-)

diff --git a/tools/bpf/bpftool/gen.c b/tools/bpf/bpftool/gen.c
index 851c465f99dc..a14d8bc5d31d 100644
--- a/tools/bpf/bpftool/gen.c
+++ b/tools/bpf/bpftool/gen.c
@@ -83,8 +83,8 @@ static const char *get_map_ident(const struct bpf_map *map)
 		return "rodata";
 	else if (str_has_suffix(name, ".bss"))
 		return "bss";
-	else if (str_has_suffix(name, ".extern"))
-		return "externs"; /* extern is a C keyword */
+	else if (str_has_suffix(name, ".kconfig"))
+		return "kconfig";
 	else
 		return NULL;
 }
@@ -112,8 +112,8 @@ static int codegen_datasec_def(struct bpf_object *obj,
 		sec_ident = "bss";
 	else if (strcmp(sec_name, ".rodata") == 0)
 		sec_ident = "rodata";
-	else if (strcmp(sec_name, ".extern") == 0)
-		sec_ident = "externs"; /* extern is a C keyword */
+	else if (strcmp(sec_name, ".kconfig") == 0)
+		sec_ident = "kconfig";
 	else
 		return 0;
 
diff --git a/tools/lib/bpf/bpf_helpers.h b/tools/lib/bpf/bpf_helpers.h
index aa46700075e1..f69cc208778a 100644
--- a/tools/lib/bpf/bpf_helpers.h
+++ b/tools/lib/bpf/bpf_helpers.h
@@ -53,4 +53,6 @@ enum libbpf_tristate {
 	TRI_MODULE = 2,
 };
 
+#define __kconfig __attribute__((section(".kconfig")))
+
 #endif
diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 2a341a5b9f63..0772eb3254e2 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -231,21 +231,21 @@ struct bpf_program {
 #define DATA_SEC ".data"
 #define BSS_SEC ".bss"
 #define RODATA_SEC ".rodata"
-#define EXTERN_SEC ".extern"
+#define KCONFIG_SEC ".kconfig"
 
 enum libbpf_map_type {
 	LIBBPF_MAP_UNSPEC,
 	LIBBPF_MAP_DATA,
 	LIBBPF_MAP_BSS,
 	LIBBPF_MAP_RODATA,
-	LIBBPF_MAP_EXTERN,
+	LIBBPF_MAP_KCONFIG,
 };
 
 static const char * const libbpf_type_to_btf_name[] = {
 	[LIBBPF_MAP_DATA]	= DATA_SEC,
 	[LIBBPF_MAP_BSS]	= BSS_SEC,
 	[LIBBPF_MAP_RODATA]	= RODATA_SEC,
-	[LIBBPF_MAP_EXTERN]	= EXTERN_SEC,
+	[LIBBPF_MAP_KCONFIG]	= KCONFIG_SEC,
 };
 
 struct bpf_map {
@@ -305,7 +305,7 @@ struct bpf_object {
 	char *kconfig_path;
 	struct extern_desc *externs;
 	int nr_extern;
-	int extern_map_idx;
+	int kconfig_map_idx;
 
 	bool loaded;
 	bool has_pseudo_calls;
@@ -606,7 +606,7 @@ static struct bpf_object *bpf_object__new(const char *path,
 	obj->efile.data_shndx = -1;
 	obj->efile.rodata_shndx = -1;
 	obj->efile.bss_shndx = -1;
-	obj->extern_map_idx = -1;
+	obj->kconfig_map_idx = -1;
 
 	obj->kern_version = get_kernel_version();
 	obj->loaded = false;
@@ -902,11 +902,25 @@ static size_t bpf_map_mmap_sz(const struct bpf_map *map)
 	return map_sz;
 }
 
+static char *internal_map_name(struct bpf_object *obj,
+			       enum libbpf_map_type type)
+{
+	char map_name[BPF_OBJ_NAME_LEN];
+	const char *sfx = libbpf_type_to_btf_name[type];
+	int sfx_len = max((size_t)7, strlen(sfx));
+	int pfx_len = min((size_t)BPF_OBJ_NAME_LEN - sfx_len - 1,
+			  strlen(obj->name));
+
+	snprintf(map_name, sizeof(map_name), "%.*s%.*s", pfx_len, obj->name,
+		 sfx_len, libbpf_type_to_btf_name[type]);
+
+	return strdup(map_name);
+}
+
 static int
 bpf_object__init_internal_map(struct bpf_object *obj, enum libbpf_map_type type,
 			      int sec_idx, void *data, size_t data_sz)
 {
-	char map_name[BPF_OBJ_NAME_LEN];
 	struct bpf_map_def *def;
 	struct bpf_map *map;
 	int err;
@@ -918,9 +932,7 @@ bpf_object__init_internal_map(struct bpf_object *obj, enum libbpf_map_type type,
 	map->libbpf_type = type;
 	map->sec_idx = sec_idx;
 	map->sec_offset = 0;
-	snprintf(map_name, sizeof(map_name), "%.8s%.7s", obj->name,
-		 libbpf_type_to_btf_name[type]);
-	map->name = strdup(map_name);
+	map->name = internal_map_name(obj, type);
 	if (!map->name) {
 		pr_warn("failed to alloc map name\n");
 		return -ENOMEM;
@@ -931,12 +943,12 @@ bpf_object__init_internal_map(struct bpf_object *obj, enum libbpf_map_type type,
 	def->key_size = sizeof(int);
 	def->value_size = data_sz;
 	def->max_entries = 1;
-	def->map_flags = type == LIBBPF_MAP_RODATA || type == LIBBPF_MAP_EXTERN
+	def->map_flags = type == LIBBPF_MAP_RODATA || type == LIBBPF_MAP_KCONFIG
 			 ? BPF_F_RDONLY_PROG : 0;
 	def->map_flags |= BPF_F_MMAPABLE;
 
 	pr_debug("map '%s' (global data): at sec_idx %d, offset %zu, flags %x.\n",
-		 map_name, map->sec_idx, map->sec_offset, def->map_flags);
+		 map->name, map->sec_idx, map->sec_offset, def->map_flags);
 
 	map->mmaped = mmap(NULL, bpf_map_mmap_sz(map), PROT_READ | PROT_WRITE,
 			   MAP_SHARED | MAP_ANONYMOUS, -1, 0);
@@ -1232,7 +1244,7 @@ out:
 	return err;
 }
 
-static int bpf_object__init_extern_map(struct bpf_object *obj)
+static int bpf_object__init_kconfig_map(struct bpf_object *obj)
 {
 	struct extern_desc *last_ext;
 	size_t map_sz;
@@ -1244,13 +1256,13 @@ static int bpf_object__init_extern_map(struct bpf_object *obj)
 	last_ext = &obj->externs[obj->nr_extern - 1];
 	map_sz = last_ext->data_off + last_ext->sz;
 
-	err = bpf_object__init_internal_map(obj, LIBBPF_MAP_EXTERN,
+	err = bpf_object__init_internal_map(obj, LIBBPF_MAP_KCONFIG,
 					    obj->efile.symbols_shndx,
 					    NULL, map_sz);
 	if (err)
 		return err;
 
-	obj->extern_map_idx = obj->nr_maps - 1;
+	obj->kconfig_map_idx = obj->nr_maps - 1;
 
 	return 0;
 }
@@ -1742,7 +1754,7 @@ static int bpf_object__init_maps(struct bpf_object *obj,
 	err = bpf_object__init_user_maps(obj, strict);
 	err = err ?: bpf_object__init_user_btf_maps(obj, strict, pin_root_path);
 	err = err ?: bpf_object__init_global_data_maps(obj);
-	err = err ?: bpf_object__init_extern_map(obj);
+	err = err ?: bpf_object__init_kconfig_map(obj);
 	if (err)
 		return err;
 
@@ -2269,9 +2281,9 @@ static int bpf_object__collect_externs(struct bpf_object *obj)
 			 i, ext->sym_idx, ext->data_off, ext->name);
 	}
 
-	btf_id = btf__find_by_name(obj->btf, EXTERN_SEC);
+	btf_id = btf__find_by_name(obj->btf, KCONFIG_SEC);
 	if (btf_id <= 0) {
-		pr_warn("no BTF info found for '%s' datasec\n", EXTERN_SEC);
+		pr_warn("no BTF info found for '%s' datasec\n", KCONFIG_SEC);
 		return -ESRCH;
 	}
 
@@ -2361,7 +2373,7 @@ bpf_object__section_to_libbpf_map_type(const struct bpf_object *obj, int shndx)
 	else if (shndx == obj->efile.rodata_shndx)
 		return LIBBPF_MAP_RODATA;
 	else if (shndx == obj->efile.symbols_shndx)
-		return LIBBPF_MAP_EXTERN;
+		return LIBBPF_MAP_KCONFIG;
 	else
 		return LIBBPF_MAP_UNSPEC;
 }
@@ -2908,8 +2920,8 @@ bpf_object__populate_internal_map(struct bpf_object *obj, struct bpf_map *map)
 		return err;
 	}
 
-	/* Freeze .rodata and .extern map as read-only from syscall side. */
-	if (map_type == LIBBPF_MAP_RODATA || map_type == LIBBPF_MAP_EXTERN) {
+	/* Freeze .rodata and .kconfig map as read-only from syscall side. */
+	if (map_type == LIBBPF_MAP_RODATA || map_type == LIBBPF_MAP_KCONFIG) {
 		err = bpf_map_freeze(map->fd);
 		if (err) {
 			err = -errno;
@@ -4264,7 +4276,7 @@ bpf_program__relocate(struct bpf_program *prog, struct bpf_object *obj)
 			break;
 		case RELO_EXTERN:
 			insn[0].src_reg = BPF_PSEUDO_MAP_VALUE;
-			insn[0].imm = obj->maps[obj->extern_map_idx].fd;
+			insn[0].imm = obj->maps[obj->kconfig_map_idx].fd;
 			insn[1].imm = relo->sym_off;
 			break;
 		case RELO_CALL:
@@ -4743,7 +4755,7 @@ static int bpf_object__resolve_externs(struct bpf_object *obj,
 	if (obj->nr_extern == 0)
 		return 0;
 
-	data = obj->maps[obj->extern_map_idx].mmaped;
+	data = obj->maps[obj->kconfig_map_idx].mmaped;
 
 	for (i = 0; i < obj->nr_extern; i++) {
 		ext = &obj->externs[i];
@@ -7546,7 +7558,7 @@ int bpf_object__open_skeleton(struct bpf_object_skeleton *s,
 		}
 
 		/* externs shouldn't be pre-setup from user code */
-		if (mmaped && (*map)->libbpf_type != LIBBPF_MAP_EXTERN)
+		if (mmaped && (*map)->libbpf_type != LIBBPF_MAP_KCONFIG)
 			*mmaped = (*map)->mmaped;
 	}
 
diff --git a/tools/testing/selftests/bpf/prog_tests/skeleton.c b/tools/testing/selftests/bpf/prog_tests/skeleton.c
index ec6f2aec3853..9264a2736018 100644
--- a/tools/testing/selftests/bpf/prog_tests/skeleton.c
+++ b/tools/testing/selftests/bpf/prog_tests/skeleton.c
@@ -15,20 +15,18 @@ void test_skeleton(void)
 	int duration = 0, err;
 	struct test_skeleton* skel;
 	struct test_skeleton__bss *bss;
-	struct test_skeleton__externs *exts;
+	struct test_skeleton__kconfig *kcfg;
 
 	skel = test_skeleton__open();
 	if (CHECK(!skel, "skel_open", "failed to open skeleton\n"))
 		return;
 
-	printf("EXTERNS BEFORE: %p\n", skel->externs);
-	if (CHECK(skel->externs, "skel_externs", "externs are mmaped()!\n"))
+	if (CHECK(skel->kconfig, "skel_kconfig", "kconfig is mmaped()!\n"))
 		goto cleanup;
 
 	err = test_skeleton__load(skel);
 	if (CHECK(err, "skel_load", "failed to load skeleton: %d\n", err))
 		goto cleanup;
-	printf("EXTERNS AFTER: %p\n", skel->externs);
 
 	bss = skel->bss;
 	bss->in1 = 1;
@@ -37,7 +35,7 @@ void test_skeleton(void)
 	bss->in4 = 4;
 	bss->in5.a = 5;
 	bss->in5.b = 6;
-	exts = skel->externs;
+	kcfg = skel->kconfig;
 
 	err = test_skeleton__attach(skel);
 	if (CHECK(err, "skel_attach", "skeleton attach failed: %d\n", err))
@@ -55,10 +53,10 @@ void test_skeleton(void)
 	CHECK(bss->handler_out5.b != 6, "res6", "got %lld != exp %d\n",
 	      bss->handler_out5.b, 6);
 
-	CHECK(bss->bpf_syscall != exts->CONFIG_BPF_SYSCALL, "ext1",
-	      "got %d != exp %d\n", bss->bpf_syscall, exts->CONFIG_BPF_SYSCALL);
-	CHECK(bss->kern_ver != exts->LINUX_KERNEL_VERSION, "ext2",
-	      "got %d != exp %d\n", bss->kern_ver, exts->LINUX_KERNEL_VERSION);
+	CHECK(bss->bpf_syscall != kcfg->CONFIG_BPF_SYSCALL, "ext1",
+	      "got %d != exp %d\n", bss->bpf_syscall, kcfg->CONFIG_BPF_SYSCALL);
+	CHECK(bss->kern_ver != kcfg->LINUX_KERNEL_VERSION, "ext2",
+	      "got %d != exp %d\n", bss->kern_ver, kcfg->LINUX_KERNEL_VERSION);
 
 cleanup:
 	test_skeleton__destroy(skel);
diff --git a/tools/testing/selftests/bpf/progs/test_core_extern.c b/tools/testing/selftests/bpf/progs/test_core_extern.c
index e12f09f9e881..9bfc91d9d004 100644
--- a/tools/testing/selftests/bpf/progs/test_core_extern.c
+++ b/tools/testing/selftests/bpf/progs/test_core_extern.c
@@ -10,16 +10,16 @@
 /* non-existing BPF helper, to test dead code elimination */
 static int (*bpf_missing_helper)(const void *arg1, int arg2) = (void *) 999;
 
-extern int LINUX_KERNEL_VERSION;
-extern bool CONFIG_BPF_SYSCALL; /* strong */
-extern enum libbpf_tristate CONFIG_TRISTATE __weak;
-extern bool CONFIG_BOOL __weak;
-extern char CONFIG_CHAR __weak;
-extern uint16_t CONFIG_USHORT __weak;
-extern int CONFIG_INT __weak;
-extern uint64_t CONFIG_ULONG __weak;
-extern const char CONFIG_STR[8] __weak;
-extern uint64_t CONFIG_MISSING __weak;
+extern int LINUX_KERNEL_VERSION __kconfig;
+extern bool CONFIG_BPF_SYSCALL __kconfig; /* strong */
+extern enum libbpf_tristate CONFIG_TRISTATE __kconfig __weak;
+extern bool CONFIG_BOOL __kconfig __weak;
+extern char CONFIG_CHAR __kconfig __weak;
+extern uint16_t CONFIG_USHORT __kconfig __weak;
+extern int CONFIG_INT __kconfig __weak;
+extern uint64_t CONFIG_ULONG __kconfig __weak;
+extern const char CONFIG_STR[8] __kconfig __weak;
+extern uint64_t CONFIG_MISSING __kconfig __weak;
 
 uint64_t kern_ver = -1;
 uint64_t bpf_syscall = -1;
diff --git a/tools/testing/selftests/bpf/progs/test_skeleton.c b/tools/testing/selftests/bpf/progs/test_skeleton.c
index 9caa44758ea2..4f69aac5635f 100644
--- a/tools/testing/selftests/bpf/progs/test_skeleton.c
+++ b/tools/testing/selftests/bpf/progs/test_skeleton.c
@@ -21,8 +21,8 @@ char out3 = 0;
 long long out4 = 0;
 int out1 = 0;
 
-extern bool CONFIG_BPF_SYSCALL;
-extern int LINUX_KERNEL_VERSION;
+extern bool CONFIG_BPF_SYSCALL __kconfig;
+extern int LINUX_KERNEL_VERSION __kconfig;
 bool bpf_syscall = 0;
 int kern_ver = 0;
 
-- 
cgit v1.2.3-59-g8ed1b


From 8601fd422148a8f7ff5f7eaf75b6703d5166332c Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Wed, 18 Dec 2019 16:28:35 -0800
Subject: libbpf: Allow to augment system Kconfig through extra optional config

Instead of all or nothing approach of overriding Kconfig file location, allow
to extend it with extra values and override chosen subset of values though
optional user-provided extra config, passed as a string through open options'
.kconfig option. If same config key is present in both user-supplied config
and Kconfig, user-supplied one wins. This allows applications to more easily
test various conditions despite host kernel's real configuration. If all of
BPF object's __kconfig externs are satisfied from user-supplied config, system
Kconfig won't be read at all.

Simplify selftests by not needing to create temporary Kconfig files.

Suggested-by: Alexei Starovoitov <ast@fb.com>
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191219002837.3074619-3-andriin@fb.com
---
 tools/lib/bpf/libbpf.c                             | 204 +++++++++++++--------
 tools/lib/bpf/libbpf.h                             |   8 +-
 .../testing/selftests/bpf/prog_tests/core_extern.c |  32 +---
 3 files changed, 132 insertions(+), 112 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 0772eb3254e2..7b74382c7977 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -302,7 +302,7 @@ struct bpf_object {
 	size_t nr_maps;
 	size_t maps_cap;
 
-	char *kconfig_path;
+	char *kconfig;
 	struct extern_desc *externs;
 	int nr_extern;
 	int kconfig_map_idx;
@@ -1149,98 +1149,129 @@ static int set_ext_value_num(struct extern_desc *ext, void *ext_val,
 	return 0;
 }
 
-static int bpf_object__read_kernel_config(struct bpf_object *obj,
-					  const char *config_path,
-					  void *data)
+static int bpf_object__process_kconfig_line(struct bpf_object *obj,
+					    char *buf, void *data)
 {
-	char buf[PATH_MAX], *sep, *value;
 	struct extern_desc *ext;
+	char *sep, *value;
 	int len, err = 0;
 	void *ext_val;
 	__u64 num;
-	gzFile file;
 
-	if (config_path) {
-		file = gzopen(config_path, "r");
-	} else {
-		struct utsname uts;
+	if (strncmp(buf, "CONFIG_", 7))
+		return 0;
 
-		uname(&uts);
-		len = snprintf(buf, PATH_MAX, "/boot/config-%s", uts.release);
-		if (len < 0)
-			return -EINVAL;
-		else if (len >= PATH_MAX)
-			return -ENAMETOOLONG;
-		/* gzopen also accepts uncompressed files. */
-		file = gzopen(buf, "r");
-		if (!file)
-			file = gzopen("/proc/config.gz", "r");
+	sep = strchr(buf, '=');
+	if (!sep) {
+		pr_warn("failed to parse '%s': no separator\n", buf);
+		return -EINVAL;
+	}
+
+	/* Trim ending '\n' */
+	len = strlen(buf);
+	if (buf[len - 1] == '\n')
+		buf[len - 1] = '\0';
+	/* Split on '=' and ensure that a value is present. */
+	*sep = '\0';
+	if (!sep[1]) {
+		*sep = '=';
+		pr_warn("failed to parse '%s': no value\n", buf);
+		return -EINVAL;
+	}
+
+	ext = find_extern_by_name(obj, buf);
+	if (!ext || ext->is_set)
+		return 0;
+
+	ext_val = data + ext->data_off;
+	value = sep + 1;
+
+	switch (*value) {
+	case 'y': case 'n': case 'm':
+		err = set_ext_value_tri(ext, ext_val, *value);
+		break;
+	case '"':
+		err = set_ext_value_str(ext, ext_val, value);
+		break;
+	default:
+		/* assume integer */
+		err = parse_u64(value, &num);
+		if (err) {
+			pr_warn("extern %s=%s should be integer\n",
+				ext->name, value);
+			return err;
+		}
+		err = set_ext_value_num(ext, ext_val, num);
+		break;
 	}
+	if (err)
+		return err;
+	pr_debug("extern %s=%s\n", ext->name, value);
+	return 0;
+}
+
+static int bpf_object__read_kconfig_file(struct bpf_object *obj, void *data)
+{
+	char buf[PATH_MAX];
+	struct utsname uts;
+	int len, err = 0;
+	gzFile file;
+
+	uname(&uts);
+	len = snprintf(buf, PATH_MAX, "/boot/config-%s", uts.release);
+	if (len < 0)
+		return -EINVAL;
+	else if (len >= PATH_MAX)
+		return -ENAMETOOLONG;
+
+	/* gzopen also accepts uncompressed files. */
+	file = gzopen(buf, "r");
+	if (!file)
+		file = gzopen("/proc/config.gz", "r");
+
 	if (!file) {
-		pr_warn("failed to read kernel config at '%s'\n", config_path);
+		pr_warn("failed to open system Kconfig\n");
 		return -ENOENT;
 	}
 
 	while (gzgets(file, buf, sizeof(buf))) {
-		if (strncmp(buf, "CONFIG_", 7))
-			continue;
-
-		sep = strchr(buf, '=');
-		if (!sep) {
-			err = -EINVAL;
-			pr_warn("failed to parse '%s': no separator\n", buf);
-			goto out;
-		}
-		/* Trim ending '\n' */
-		len = strlen(buf);
-		if (buf[len - 1] == '\n')
-			buf[len - 1] = '\0';
-		/* Split on '=' and ensure that a value is present. */
-		*sep = '\0';
-		if (!sep[1]) {
-			err = -EINVAL;
-			*sep = '=';
-			pr_warn("failed to parse '%s': no value\n", buf);
+		err = bpf_object__process_kconfig_line(obj, buf, data);
+		if (err) {
+			pr_warn("error parsing system Kconfig line '%s': %d\n",
+				buf, err);
 			goto out;
 		}
+	}
 
-		ext = find_extern_by_name(obj, buf);
-		if (!ext)
-			continue;
-		if (ext->is_set) {
-			err = -EINVAL;
-			pr_warn("re-defining extern '%s' not allowed\n", buf);
-			goto out;
-		}
+out:
+	gzclose(file);
+	return err;
+}
 
-		ext_val = data + ext->data_off;
-		value = sep + 1;
+static int bpf_object__read_kconfig_mem(struct bpf_object *obj,
+					const char *config, void *data)
+{
+	char buf[PATH_MAX];
+	int err = 0;
+	FILE *file;
 
-		switch (*value) {
-		case 'y': case 'n': case 'm':
-			err = set_ext_value_tri(ext, ext_val, *value);
-			break;
-		case '"':
-			err = set_ext_value_str(ext, ext_val, value);
-			break;
-		default:
-			/* assume integer */
-			err = parse_u64(value, &num);
-			if (err) {
-				pr_warn("extern %s=%s should be integer\n",
-					ext->name, value);
-				goto out;
-			}
-			err = set_ext_value_num(ext, ext_val, num);
+	file = fmemopen((void *)config, strlen(config), "r");
+	if (!file) {
+		err = -errno;
+		pr_warn("failed to open in-memory Kconfig: %d\n", err);
+		return err;
+	}
+
+	while (fgets(buf, sizeof(buf), file)) {
+		err = bpf_object__process_kconfig_line(obj, buf, data);
+		if (err) {
+			pr_warn("error parsing in-memory Kconfig line '%s': %d\n",
+				buf, err);
 			break;
 		}
-		if (err)
-			goto out;
-		pr_debug("extern %s=%s\n", ext->name, value);
 	}
 
-out:
-	gzclose(file);
+	fclose(file);
 	return err;
 }
 
@@ -4567,7 +4598,7 @@ static struct bpf_object *
 __bpf_object__open(const char *path, const void *obj_buf, size_t obj_buf_sz,
 		   const struct bpf_object_open_opts *opts)
 {
-	const char *obj_name, *kconfig_path;
+	const char *obj_name, *kconfig;
 	struct bpf_program *prog;
 	struct bpf_object *obj;
 	char tmp_name[64];
@@ -4599,10 +4630,10 @@ __bpf_object__open(const char *path, const void *obj_buf, size_t obj_buf_sz,
 		return obj;
 
 	obj->relaxed_core_relocs = OPTS_GET(opts, relaxed_core_relocs, false);
-	kconfig_path = OPTS_GET(opts, kconfig_path, NULL);
-	if (kconfig_path) {
-		obj->kconfig_path = strdup(kconfig_path);
-		if (!obj->kconfig_path)
+	kconfig = OPTS_GET(opts, kconfig, NULL);
+	if (kconfig) {
+		obj->kconfig = strdup(kconfig);
+		if (!obj->kconfig)
 			return ERR_PTR(-ENOMEM);
 	}
 
@@ -4745,7 +4776,7 @@ static int bpf_object__sanitize_maps(struct bpf_object *obj)
 }
 
 static int bpf_object__resolve_externs(struct bpf_object *obj,
-				       const char *config_path)
+				       const char *extra_kconfig)
 {
 	bool need_config = false;
 	struct extern_desc *ext;
@@ -4779,8 +4810,21 @@ static int bpf_object__resolve_externs(struct bpf_object *obj,
 			return -EINVAL;
 		}
 	}
+	if (need_config && extra_kconfig) {
+		err = bpf_object__read_kconfig_mem(obj, extra_kconfig, data);
+		if (err)
+			return -EINVAL;
+		need_config = false;
+		for (i = 0; i < obj->nr_extern; i++) {
+			ext = &obj->externs[i];
+			if (!ext->is_set) {
+				need_config = true;
+				break;
+			}
+		}
+	}
 	if (need_config) {
-		err = bpf_object__read_kernel_config(obj, config_path, data);
+		err = bpf_object__read_kconfig_file(obj, data);
 		if (err)
 			return -EINVAL;
 	}
@@ -4818,7 +4862,7 @@ int bpf_object__load_xattr(struct bpf_object_load_attr *attr)
 	obj->loaded = true;
 
 	err = bpf_object__probe_caps(obj);
-	err = err ? : bpf_object__resolve_externs(obj, obj->kconfig_path);
+	err = err ? : bpf_object__resolve_externs(obj, obj->kconfig);
 	err = err ? : bpf_object__sanitize_and_load_btf(obj);
 	err = err ? : bpf_object__sanitize_maps(obj);
 	err = err ? : bpf_object__create_maps(obj);
@@ -5412,7 +5456,7 @@ void bpf_object__close(struct bpf_object *obj)
 		zfree(&map->pin_path);
 	}
 
-	zfree(&obj->kconfig_path);
+	zfree(&obj->kconfig);
 	zfree(&obj->externs);
 	obj->nr_extern = 0;
 
diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index ad8c1c127933..fe592ef48f1b 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -85,12 +85,12 @@ struct bpf_object_open_opts {
 	 */
 	const char *pin_root_path;
 	__u32 attach_prog_fd;
-	/* kernel config file path override (for CONFIG_ externs); can point
-	 * to either uncompressed text file or .gz file
+	/* Additional kernel config content that augments and overrides
+	 * system Kconfig for CONFIG_xxx externs.
 	 */
-	const char *kconfig_path;
+	const char *kconfig;
 };
-#define bpf_object_open_opts__last_field kconfig_path
+#define bpf_object_open_opts__last_field kconfig
 
 LIBBPF_API struct bpf_object *bpf_object__open(const char *path);
 LIBBPF_API struct bpf_object *
diff --git a/tools/testing/selftests/bpf/prog_tests/core_extern.c b/tools/testing/selftests/bpf/prog_tests/core_extern.c
index 5f03dc1de29e..b093787e9448 100644
--- a/tools/testing/selftests/bpf/prog_tests/core_extern.c
+++ b/tools/testing/selftests/bpf/prog_tests/core_extern.c
@@ -23,19 +23,13 @@ static uint32_t get_kernel_version(void)
 static struct test_case {
 	const char *name;
 	const char *cfg;
-	const char *cfg_path;
 	bool fails;
 	struct test_core_extern__data data;
 } test_cases[] = {
-	{ .name = "default search path", .cfg_path = NULL,
-	  .data = { .bpf_syscall = true } },
-	{ .name = "/proc/config.gz", .cfg_path = "/proc/config.gz",
-	  .data = { .bpf_syscall = true } },
-	{ .name = "missing config", .fails = true,
-	  .cfg_path = "/proc/invalid-config.gz" },
+	{ .name = "default search path", .data = { .bpf_syscall = true } },
 	{
 		.name = "custom values",
-		.cfg = "CONFIG_BPF_SYSCALL=y\n"
+		.cfg = "CONFIG_BPF_SYSCALL=n\n"
 		       "CONFIG_TRISTATE=m\n"
 		       "CONFIG_BOOL=y\n"
 		       "CONFIG_CHAR=100\n"
@@ -45,7 +39,7 @@ static struct test_case {
 		       "CONFIG_STR=\"abracad\"\n"
 		       "CONFIG_MISSING=0",
 		.data = {
-			.bpf_syscall = true,
+			.bpf_syscall = false,
 			.tristate_val = TRI_MODULE,
 			.bool_val = true,
 			.char_val = 100,
@@ -133,30 +127,14 @@ void test_core_extern(void)
 	int n = sizeof(*skel->data) / sizeof(uint64_t);
 
 	for (i = 0; i < ARRAY_SIZE(test_cases); i++) {
-		char tmp_cfg_path[] = "/tmp/test_core_extern_cfg.XXXXXX";
 		struct test_case *t = &test_cases[i];
 		DECLARE_LIBBPF_OPTS(bpf_object_open_opts, opts,
-			.kconfig_path = t->cfg_path,
+			.kconfig = t->cfg,
 		);
 
 		if (!test__start_subtest(t->name))
 			continue;
 
-		if (t->cfg) {
-			size_t n = strlen(t->cfg) + 1;
-			int fd = mkstemp(tmp_cfg_path);
-			int written;
-
-			if (CHECK(fd < 0, "mkstemp", "errno: %d\n", errno))
-				continue;
-			printf("using '%s' as config file\n", tmp_cfg_path);
-			written = write(fd, t->cfg, n);
-			close(fd);
-			if (CHECK_FAIL(written != n))
-				goto cleanup;
-			opts.kconfig_path = tmp_cfg_path;
-		}
-
 		skel = test_core_extern__open_opts(&opts);
 		if (CHECK(!skel, "skel_open", "skeleton open failed\n"))
 			goto cleanup;
@@ -185,8 +163,6 @@ void test_core_extern(void)
 			       j, exp[j], got[j]);
 		}
 cleanup:
-		if (t->cfg)
-			unlink(tmp_cfg_path);
 		test_core_extern__destroy(skel);
 		skel = NULL;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 630628cb7dc39780660d8fcedc66e0298a82f9da Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Wed, 18 Dec 2019 16:28:36 -0800
Subject: libbpf: BTF is required when externs are present

BTF is required to get type information about extern variables.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191219002837.3074619-4-andriin@fb.com
---
 tools/lib/bpf/libbpf.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 7b74382c7977..6340b81b555b 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -1887,7 +1887,8 @@ static void bpf_object__sanitize_btf_ext(struct bpf_object *obj)
 
 static bool bpf_object__is_btf_mandatory(const struct bpf_object *obj)
 {
-	return obj->efile.btf_maps_shndx >= 0;
+	return obj->efile.btf_maps_shndx >= 0 ||
+	       obj->nr_extern > 0;
 }
 
 static int bpf_object__init_btf(struct bpf_object *obj,
-- 
cgit v1.2.3-59-g8ed1b


From f1003b787c00fbaa4b11619c6b23a885bfce8f07 Mon Sep 17 00:00:00 2001
From: Björn Töpel <bjorn.topel@gmail.com>
Date: Mon, 16 Dec 2019 10:13:35 +0100
Subject: riscv, bpf: Fix broken BPF tail calls
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The BPF JIT incorrectly clobbered the a0 register, and did not flag
usage of s5 register when BPF stack was being used.

Fixes: 2353ecc6f91f ("bpf, riscv: add BPF JIT for RV64G")
Signed-off-by: Björn Töpel <bjorn.topel@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191216091343.23260-2-bjorn.topel@gmail.com
---
 arch/riscv/net/bpf_jit_comp.c | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

diff --git a/arch/riscv/net/bpf_jit_comp.c b/arch/riscv/net/bpf_jit_comp.c
index 5451ef3845f2..1606ebd49666 100644
--- a/arch/riscv/net/bpf_jit_comp.c
+++ b/arch/riscv/net/bpf_jit_comp.c
@@ -120,6 +120,11 @@ static bool seen_reg(int reg, struct rv_jit_context *ctx)
 	return false;
 }
 
+static void mark_fp(struct rv_jit_context *ctx)
+{
+	__set_bit(RV_CTX_F_SEEN_S5, &ctx->flags);
+}
+
 static void mark_call(struct rv_jit_context *ctx)
 {
 	__set_bit(RV_CTX_F_SEEN_CALL, &ctx->flags);
@@ -596,7 +601,8 @@ static void __build_epilogue(u8 reg, struct rv_jit_context *ctx)
 
 	emit(rv_addi(RV_REG_SP, RV_REG_SP, stack_adjust), ctx);
 	/* Set return value. */
-	emit(rv_addi(RV_REG_A0, RV_REG_A5, 0), ctx);
+	if (reg == RV_REG_RA)
+		emit(rv_addi(RV_REG_A0, RV_REG_A5, 0), ctx);
 	emit(rv_jalr(RV_REG_ZERO, reg, 0), ctx);
 }
 
@@ -1426,6 +1432,10 @@ static void build_prologue(struct rv_jit_context *ctx)
 {
 	int stack_adjust = 0, store_offset, bpf_stack_adjust;
 
+	bpf_stack_adjust = round_up(ctx->prog->aux->stack_depth, 16);
+	if (bpf_stack_adjust)
+		mark_fp(ctx);
+
 	if (seen_reg(RV_REG_RA, ctx))
 		stack_adjust += 8;
 	stack_adjust += 8; /* RV_REG_FP */
@@ -1443,7 +1453,6 @@ static void build_prologue(struct rv_jit_context *ctx)
 		stack_adjust += 8;
 
 	stack_adjust = round_up(stack_adjust, 16);
-	bpf_stack_adjust = round_up(ctx->prog->aux->stack_depth, 16);
 	stack_adjust += bpf_stack_adjust;
 
 	store_offset = stack_adjust - 8;
-- 
cgit v1.2.3-59-g8ed1b


From 7d1ef13fea2b66bb466592a7855b33181c85ec5f Mon Sep 17 00:00:00 2001
From: Björn Töpel <bjorn.topel@gmail.com>
Date: Mon, 16 Dec 2019 10:13:36 +0100
Subject: riscv, bpf: Add support for far branching
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This commit adds branch relaxation to the BPF JIT, and with that
support for far (offset greater than 12b) branching.

The branch relaxation requires more than two passes to converge. For
most programs it is three passes, but for larger programs it can be
more.

Signed-off-by: Björn Töpel <bjorn.topel@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Luke Nelson <lukenels@cs.washington.edu>
Link: https://lore.kernel.org/bpf/20191216091343.23260-3-bjorn.topel@gmail.com
---
 arch/riscv/net/bpf_jit_comp.c | 352 ++++++++++++++++++++++--------------------
 1 file changed, 188 insertions(+), 164 deletions(-)

diff --git a/arch/riscv/net/bpf_jit_comp.c b/arch/riscv/net/bpf_jit_comp.c
index 1606ebd49666..e599458a9bcd 100644
--- a/arch/riscv/net/bpf_jit_comp.c
+++ b/arch/riscv/net/bpf_jit_comp.c
@@ -461,6 +461,11 @@ static u32 rv_amoadd_d(u8 rd, u8 rs2, u8 rs1, u8 aq, u8 rl)
 	return rv_amo_insn(0, aq, rl, rs2, rs1, 3, rd, 0x2f);
 }
 
+static u32 rv_auipc(u8 rd, u32 imm31_12)
+{
+	return rv_u_insn(imm31_12, rd, 0x17);
+}
+
 static bool is_12b_int(s64 val)
 {
 	return -(1 << 11) <= val && val < (1 << 11);
@@ -484,7 +489,7 @@ static bool is_32b_int(s64 val)
 static int is_12b_check(int off, int insn)
 {
 	if (!is_12b_int(off)) {
-		pr_err("bpf-jit: insn=%d offset=%d not supported yet!\n",
+		pr_err("bpf-jit: insn=%d 12b < offset=%d not supported yet!\n",
 		       insn, (int)off);
 		return -1;
 	}
@@ -494,7 +499,7 @@ static int is_12b_check(int off, int insn)
 static int is_13b_check(int off, int insn)
 {
 	if (!is_13b_int(off)) {
-		pr_err("bpf-jit: insn=%d offset=%d not supported yet!\n",
+		pr_err("bpf-jit: insn=%d 13b < offset=%d not supported yet!\n",
 		       insn, (int)off);
 		return -1;
 	}
@@ -504,7 +509,7 @@ static int is_13b_check(int off, int insn)
 static int is_21b_check(int off, int insn)
 {
 	if (!is_21b_int(off)) {
-		pr_err("bpf-jit: insn=%d offset=%d not supported yet!\n",
+		pr_err("bpf-jit: insn=%d 21b < offset=%d not supported yet!\n",
 		       insn, (int)off);
 		return -1;
 	}
@@ -550,10 +555,13 @@ static void emit_imm(u8 rd, s64 val, struct rv_jit_context *ctx)
 		emit(rv_addi(rd, rd, lower), ctx);
 }
 
-static int rv_offset(int bpf_to, int bpf_from, struct rv_jit_context *ctx)
+static int rv_offset(int insn, int off, struct rv_jit_context *ctx)
 {
-	int from = ctx->offset[bpf_from] - 1, to = ctx->offset[bpf_to];
+	int from, to;
 
+	off++; /* BPF branch is from PC+1, RV is from PC */
+	from = (insn > 0) ? ctx->offset[insn - 1] : 0;
+	to = (insn + off > 0) ? ctx->offset[insn + off - 1] : 0;
 	return (to - from) << 2;
 }
 
@@ -606,6 +614,109 @@ static void __build_epilogue(u8 reg, struct rv_jit_context *ctx)
 	emit(rv_jalr(RV_REG_ZERO, reg, 0), ctx);
 }
 
+/* return -1 or inverted cond */
+static int invert_bpf_cond(u8 cond)
+{
+	switch (cond) {
+	case BPF_JEQ:
+		return BPF_JNE;
+	case BPF_JGT:
+		return BPF_JLE;
+	case BPF_JLT:
+		return BPF_JGE;
+	case BPF_JGE:
+		return BPF_JLT;
+	case BPF_JLE:
+		return BPF_JGT;
+	case BPF_JNE:
+		return BPF_JEQ;
+	case BPF_JSGT:
+		return BPF_JSLE;
+	case BPF_JSLT:
+		return BPF_JSGE;
+	case BPF_JSGE:
+		return BPF_JSLT;
+	case BPF_JSLE:
+		return BPF_JSGT;
+	}
+	return -1;
+}
+
+static void emit_bcc(u8 cond, u8 rd, u8 rs, int rvoff,
+		     struct rv_jit_context *ctx)
+{
+	switch (cond) {
+	case BPF_JEQ:
+		emit(rv_beq(rd, rs, rvoff >> 1), ctx);
+		return;
+	case BPF_JGT:
+		emit(rv_bltu(rs, rd, rvoff >> 1), ctx);
+		return;
+	case BPF_JLT:
+		emit(rv_bltu(rd, rs, rvoff >> 1), ctx);
+		return;
+	case BPF_JGE:
+		emit(rv_bgeu(rd, rs, rvoff >> 1), ctx);
+		return;
+	case BPF_JLE:
+		emit(rv_bgeu(rs, rd, rvoff >> 1), ctx);
+		return;
+	case BPF_JNE:
+		emit(rv_bne(rd, rs, rvoff >> 1), ctx);
+		return;
+	case BPF_JSGT:
+		emit(rv_blt(rs, rd, rvoff >> 1), ctx);
+		return;
+	case BPF_JSLT:
+		emit(rv_blt(rd, rs, rvoff >> 1), ctx);
+		return;
+	case BPF_JSGE:
+		emit(rv_bge(rd, rs, rvoff >> 1), ctx);
+		return;
+	case BPF_JSLE:
+		emit(rv_bge(rs, rd, rvoff >> 1), ctx);
+	}
+}
+
+static void emit_branch(u8 cond, u8 rd, u8 rs, int rvoff,
+			struct rv_jit_context *ctx)
+{
+	s64 upper, lower;
+
+	if (is_13b_int(rvoff)) {
+		emit_bcc(cond, rd, rs, rvoff, ctx);
+		return;
+	}
+
+	/* Adjust for jal */
+	rvoff -= 4;
+
+	/* Transform, e.g.:
+	 *   bne rd,rs,foo
+	 * to
+	 *   beq rd,rs,<.L1>
+	 *   (auipc foo)
+	 *   jal(r) foo
+	 * .L1
+	 */
+	cond = invert_bpf_cond(cond);
+	if (is_21b_int(rvoff)) {
+		emit_bcc(cond, rd, rs, 8, ctx);
+		emit(rv_jal(RV_REG_ZERO, rvoff >> 1), ctx);
+		return;
+	}
+
+	/* 32b No need for an additional rvoff adjustment, since we
+	 * get that from the auipc at PC', where PC = PC' + 4.
+	 */
+	upper = (rvoff + (1 << 11)) >> 12;
+	lower = rvoff & 0xfff;
+
+	emit_bcc(cond, rd, rs, 12, ctx);
+	emit(rv_auipc(RV_REG_T1, upper), ctx);
+	emit(rv_jalr(RV_REG_ZERO, RV_REG_T1, lower), ctx);
+}
+
 static void emit_zext_32(u8 reg, struct rv_jit_context *ctx)
 {
 	emit(rv_slli(reg, reg, 32), ctx);
@@ -693,13 +804,6 @@ static void init_regs(u8 *rd, u8 *rs, const struct bpf_insn *insn,
 		*rs = bpf_to_rv_reg(insn->src_reg, ctx);
 }
 
-static int rv_offset_check(int *rvoff, s16 off, int insn,
-			   struct rv_jit_context *ctx)
-{
-	*rvoff = rv_offset(insn + off, insn, ctx);
-	return is_13b_check(*rvoff, insn);
-}
-
 static void emit_zext_32_rd_rs(u8 *rd, u8 *rs, struct rv_jit_context *ctx)
 {
 	emit(rv_addi(RV_REG_T2, *rd, 0), ctx);
@@ -732,13 +836,19 @@ static void emit_sext_32_rd(u8 *rd, struct rv_jit_context *ctx)
 	*rd = RV_REG_T2;
 }
 
+static bool is_signed_bpf_cond(u8 cond)
+{
+	return cond == BPF_JSGT || cond == BPF_JSLT ||
+		cond == BPF_JSGE || cond == BPF_JSLE;
+}
+
 static int emit_insn(const struct bpf_insn *insn, struct rv_jit_context *ctx,
 		     bool extra_pass)
 {
 	bool is64 = BPF_CLASS(insn->code) == BPF_ALU64 ||
 		    BPF_CLASS(insn->code) == BPF_JMP;
+	int s, e, rvoff, i = insn - ctx->prog->insnsi;
 	struct bpf_prog_aux *aux = ctx->prog->aux;
-	int rvoff, i = insn - ctx->prog->insnsi;
 	u8 rd = -1, rs = -1, code = insn->code;
 	s16 off = insn->off;
 	s32 imm = insn->imm;
@@ -1006,7 +1116,7 @@ out_be:
 
 	/* JUMP off */
 	case BPF_JMP | BPF_JA:
-		rvoff = rv_offset(i + off, i, ctx);
+		rvoff = rv_offset(i, off, ctx);
 		if (!is_21b_int(rvoff)) {
 			pr_err("bpf-jit: insn=%d offset=%d not supported yet!\n",
 			       i, rvoff);
@@ -1019,194 +1129,96 @@ out_be:
 	/* IF (dst COND src) JUMP off */
 	case BPF_JMP | BPF_JEQ | BPF_X:
 	case BPF_JMP32 | BPF_JEQ | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_zext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_beq(rd, rs, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JGT | BPF_X:
 	case BPF_JMP32 | BPF_JGT | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_zext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_bltu(rs, rd, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JLT | BPF_X:
 	case BPF_JMP32 | BPF_JLT | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_zext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_bltu(rd, rs, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JGE | BPF_X:
 	case BPF_JMP32 | BPF_JGE | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_zext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_bgeu(rd, rs, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JLE | BPF_X:
 	case BPF_JMP32 | BPF_JLE | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_zext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_bgeu(rs, rd, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JNE | BPF_X:
 	case BPF_JMP32 | BPF_JNE | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_zext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_bne(rd, rs, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSGT | BPF_X:
 	case BPF_JMP32 | BPF_JSGT | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_sext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_blt(rs, rd, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSLT | BPF_X:
 	case BPF_JMP32 | BPF_JSLT | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_sext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_blt(rd, rs, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSGE | BPF_X:
 	case BPF_JMP32 | BPF_JSGE | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_sext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_bge(rd, rs, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSLE | BPF_X:
 	case BPF_JMP32 | BPF_JSLE | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_sext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_bge(rs, rd, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSET | BPF_X:
 	case BPF_JMP32 | BPF_JSET | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_zext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_and(RV_REG_T1, rd, rs), ctx);
-		emit(rv_bne(RV_REG_T1, RV_REG_ZERO, rvoff >> 1), ctx);
+		rvoff = rv_offset(i, off, ctx);
+		if (!is64) {
+			s = ctx->ninsns;
+			if (is_signed_bpf_cond(BPF_OP(code)))
+				emit_sext_32_rd_rs(&rd, &rs, ctx);
+			else
+				emit_zext_32_rd_rs(&rd, &rs, ctx);
+			e = ctx->ninsns;
+
+			/* Adjust for extra insns */
+			rvoff -= (e - s) << 2;
+		}
+
+		if (BPF_OP(code) == BPF_JSET) {
+			/* Adjust for and */
+			rvoff -= 4;
+			emit(rv_and(RV_REG_T1, rd, rs), ctx);
+			emit_branch(BPF_JNE, RV_REG_T1, RV_REG_ZERO, rvoff,
+				    ctx);
+		} else {
+			emit_branch(BPF_OP(code), rd, rs, rvoff, ctx);
+		}
 		break;
 
 	/* IF (dst COND imm) JUMP off */
 	case BPF_JMP | BPF_JEQ | BPF_K:
 	case BPF_JMP32 | BPF_JEQ | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_zext_32_rd_t1(&rd, ctx);
-		emit(rv_beq(rd, RV_REG_T1, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JGT | BPF_K:
 	case BPF_JMP32 | BPF_JGT | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_zext_32_rd_t1(&rd, ctx);
-		emit(rv_bltu(RV_REG_T1, rd, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JLT | BPF_K:
 	case BPF_JMP32 | BPF_JLT | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_zext_32_rd_t1(&rd, ctx);
-		emit(rv_bltu(rd, RV_REG_T1, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JGE | BPF_K:
 	case BPF_JMP32 | BPF_JGE | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_zext_32_rd_t1(&rd, ctx);
-		emit(rv_bgeu(rd, RV_REG_T1, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JLE | BPF_K:
 	case BPF_JMP32 | BPF_JLE | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_zext_32_rd_t1(&rd, ctx);
-		emit(rv_bgeu(RV_REG_T1, rd, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JNE | BPF_K:
 	case BPF_JMP32 | BPF_JNE | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_zext_32_rd_t1(&rd, ctx);
-		emit(rv_bne(rd, RV_REG_T1, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSGT | BPF_K:
 	case BPF_JMP32 | BPF_JSGT | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_sext_32_rd(&rd, ctx);
-		emit(rv_blt(RV_REG_T1, rd, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSLT | BPF_K:
 	case BPF_JMP32 | BPF_JSLT | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_sext_32_rd(&rd, ctx);
-		emit(rv_blt(rd, RV_REG_T1, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSGE | BPF_K:
 	case BPF_JMP32 | BPF_JSGE | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_sext_32_rd(&rd, ctx);
-		emit(rv_bge(rd, RV_REG_T1, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSLE | BPF_K:
 	case BPF_JMP32 | BPF_JSLE | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_sext_32_rd(&rd, ctx);
-		emit(rv_bge(RV_REG_T1, rd, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSET | BPF_K:
 	case BPF_JMP32 | BPF_JSET | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
+		rvoff = rv_offset(i, off, ctx);
+		s = ctx->ninsns;
 		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_zext_32_rd_t1(&rd, ctx);
-		emit(rv_and(RV_REG_T1, rd, RV_REG_T1), ctx);
-		emit(rv_bne(RV_REG_T1, RV_REG_ZERO, rvoff >> 1), ctx);
+		if (!is64) {
+			if (is_signed_bpf_cond(BPF_OP(code)))
+				emit_sext_32_rd(&rd, ctx);
+			else
+				emit_zext_32_rd_t1(&rd, ctx);
+		}
+		e = ctx->ninsns;
+
+		/* Adjust for extra insns */
+		rvoff -= (e - s) << 2;
+
+		if (BPF_OP(code) == BPF_JSET) {
+			/* Adjust for and */
+			rvoff -= 4;
+			emit(rv_and(RV_REG_T1, rd, RV_REG_T1), ctx);
+			emit_branch(BPF_JNE, RV_REG_T1, RV_REG_ZERO, rvoff,
+				    ctx);
+		} else {
+			emit_branch(BPF_OP(code), rd, RV_REG_T1, rvoff, ctx);
+		}
 		break;
 
 	/* function call */
@@ -1557,6 +1569,7 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *prog)
 {
 	bool tmp_blinded = false, extra_pass = false;
 	struct bpf_prog *tmp, *orig_prog = prog;
+	int pass = 0, prev_ninsns = 0, i;
 	struct rv_jit_data *jit_data;
 	struct rv_jit_context *ctx;
 	unsigned int image_size;
@@ -1596,15 +1609,25 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *prog)
 		prog = orig_prog;
 		goto out_offset;
 	}
+	for (i = 0; i < prog->len; i++) {
+		prev_ninsns += 32;
+		ctx->offset[i] = prev_ninsns;
+	}
 
-	/* First pass generates the ctx->offset, but does not emit an image. */
-	if (build_body(ctx, extra_pass)) {
-		prog = orig_prog;
-		goto out_offset;
+	for (i = 0; i < 16; i++) {
+		pass++;
+		ctx->ninsns = 0;
+		if (build_body(ctx, extra_pass)) {
+			prog = orig_prog;
+			goto out_offset;
+		}
+		build_prologue(ctx);
+		ctx->epilogue_offset = ctx->ninsns;
+		build_epilogue(ctx);
+		if (ctx->ninsns == prev_ninsns)
+			break;
+		prev_ninsns = ctx->ninsns;
 	}
-	build_prologue(ctx);
-	ctx->epilogue_offset = ctx->ninsns;
-	build_epilogue(ctx);
 
 	/* Allocate image, now that we know the size. */
 	image_size = sizeof(u32) * ctx->ninsns;
@@ -1619,6 +1642,7 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *prog)
 	/* Second, real pass, that acutally emits the image. */
 	ctx->insns = (u32 *)jit_data->image;
 skip_init_ctx:
+	pass++;
 	ctx->ninsns = 0;
 
 	build_prologue(ctx);
@@ -1630,7 +1654,7 @@ skip_init_ctx:
 	build_epilogue(ctx);
 
 	if (bpf_jit_enable > 1)
-		bpf_jit_dump(prog->len, image_size, 2, ctx->insns);
+		bpf_jit_dump(prog->len, image_size, pass, ctx->insns);
 
 	prog->bpf_func = (void *)ctx->insns;
 	prog->jited = 1;
-- 
cgit v1.2.3-59-g8ed1b


From 29d92edd9ee8b37bd8e9a0ba7fd549f874e0d069 Mon Sep 17 00:00:00 2001
From: Björn Töpel <bjorn.topel@gmail.com>
Date: Mon, 16 Dec 2019 10:13:37 +0100
Subject: riscv, bpf: Add support for far branching when emitting tail call
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Start use the emit_branch() function in the tail call emitter in order
to support far branching.

Signed-off-by: Björn Töpel <bjorn.topel@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191216091343.23260-4-bjorn.topel@gmail.com
---
 arch/riscv/net/bpf_jit_comp.c | 22 +++-------------------
 1 file changed, 3 insertions(+), 19 deletions(-)

diff --git a/arch/riscv/net/bpf_jit_comp.c b/arch/riscv/net/bpf_jit_comp.c
index e599458a9bcd..c38c95df3440 100644
--- a/arch/riscv/net/bpf_jit_comp.c
+++ b/arch/riscv/net/bpf_jit_comp.c
@@ -496,16 +496,6 @@ static int is_12b_check(int off, int insn)
 	return 0;
 }
 
-static int is_13b_check(int off, int insn)
-{
-	if (!is_13b_int(off)) {
-		pr_err("bpf-jit: insn=%d 13b < offset=%d not supported yet!\n",
-		       insn, (int)off);
-		return -1;
-	}
-	return 0;
-}
-
 static int is_21b_check(int off, int insn)
 {
 	if (!is_21b_int(off)) {
@@ -744,18 +734,14 @@ static int emit_bpf_tail_call(int insn, struct rv_jit_context *ctx)
 		return -1;
 	emit(rv_lwu(RV_REG_T1, off, RV_REG_A1), ctx);
 	off = (tc_ninsn - (ctx->ninsns - start_insn)) << 2;
-	if (is_13b_check(off, insn))
-		return -1;
-	emit(rv_bgeu(RV_REG_A2, RV_REG_T1, off >> 1), ctx);
+	emit_branch(BPF_JGE, RV_REG_A2, RV_REG_T1, off, ctx);
 
 	/* if (--TCC < 0)
 	 *     goto out;
 	 */
 	emit(rv_addi(RV_REG_T1, tcc, -1), ctx);
 	off = (tc_ninsn - (ctx->ninsns - start_insn)) << 2;
-	if (is_13b_check(off, insn))
-		return -1;
-	emit(rv_blt(RV_REG_T1, RV_REG_ZERO, off >> 1), ctx);
+	emit_branch(BPF_JSLT, RV_REG_T1, RV_REG_ZERO, off, ctx);
 
 	/* prog = array->ptrs[index];
 	 * if (!prog)
@@ -768,9 +754,7 @@ static int emit_bpf_tail_call(int insn, struct rv_jit_context *ctx)
 		return -1;
 	emit(rv_ld(RV_REG_T2, off, RV_REG_T2), ctx);
 	off = (tc_ninsn - (ctx->ninsns - start_insn)) << 2;
-	if (is_13b_check(off, insn))
-		return -1;
-	emit(rv_beq(RV_REG_T2, RV_REG_ZERO, off >> 1), ctx);
+	emit_branch(BPF_JEQ, RV_REG_T2, RV_REG_ZERO, off, ctx);
 
 	/* goto *(prog->bpf_func + 4); */
 	off = offsetof(struct bpf_prog, bpf_func);
-- 
cgit v1.2.3-59-g8ed1b


From 33203c02f2f8ae8e19d7dbbabf55cfd0025efede Mon Sep 17 00:00:00 2001
From: Björn Töpel <bjorn.topel@gmail.com>
Date: Mon, 16 Dec 2019 10:13:38 +0100
Subject: riscv, bpf: Add support for far jumps and exits
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This commit add support for far (offset > 21b) jumps and exits.

Signed-off-by: Björn Töpel <bjorn.topel@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Luke Nelson <lukenels@cs.washington.edu>
Link: https://lore.kernel.org/bpf/20191216091343.23260-5-bjorn.topel@gmail.com
---
 arch/riscv/net/bpf_jit_comp.c | 37 +++++++++++++++++--------------------
 1 file changed, 17 insertions(+), 20 deletions(-)

diff --git a/arch/riscv/net/bpf_jit_comp.c b/arch/riscv/net/bpf_jit_comp.c
index c38c95df3440..2fc0f24ad30f 100644
--- a/arch/riscv/net/bpf_jit_comp.c
+++ b/arch/riscv/net/bpf_jit_comp.c
@@ -496,16 +496,6 @@ static int is_12b_check(int off, int insn)
 	return 0;
 }
 
-static int is_21b_check(int off, int insn)
-{
-	if (!is_21b_int(off)) {
-		pr_err("bpf-jit: insn=%d 21b < offset=%d not supported yet!\n",
-		       insn, (int)off);
-		return -1;
-	}
-	return 0;
-}
-
 static void emit_imm(u8 rd, s64 val, struct rv_jit_context *ctx)
 {
 	/* Note that the immediate from the add is sign-extended,
@@ -820,6 +810,21 @@ static void emit_sext_32_rd(u8 *rd, struct rv_jit_context *ctx)
 	*rd = RV_REG_T2;
 }
 
+static void emit_jump_and_link(u8 rd, int rvoff, struct rv_jit_context *ctx)
+{
+	s64 upper, lower;
+
+	if (is_21b_int(rvoff)) {
+		emit(rv_jal(rd, rvoff >> 1), ctx);
+		return;
+	}
+
+	upper = (rvoff + (1 << 11)) >> 12;
+	lower = rvoff & 0xfff;
+	emit(rv_auipc(RV_REG_T1, upper), ctx);
+	emit(rv_jalr(rd, RV_REG_T1, lower), ctx);
+}
+
 static bool is_signed_bpf_cond(u8 cond)
 {
 	return cond == BPF_JSGT || cond == BPF_JSLT ||
@@ -1101,13 +1106,7 @@ out_be:
 	/* JUMP off */
 	case BPF_JMP | BPF_JA:
 		rvoff = rv_offset(i, off, ctx);
-		if (!is_21b_int(rvoff)) {
-			pr_err("bpf-jit: insn=%d offset=%d not supported yet!\n",
-			       i, rvoff);
-			return -1;
-		}
-
-		emit(rv_jal(RV_REG_ZERO, rvoff >> 1), ctx);
+		emit_jump_and_link(RV_REG_ZERO, rvoff, ctx);
 		break;
 
 	/* IF (dst COND src) JUMP off */
@@ -1245,9 +1244,7 @@ out_be:
 			break;
 
 		rvoff = epilogue_offset(ctx);
-		if (is_21b_check(rvoff, i))
-			return -1;
-		emit(rv_jal(RV_REG_ZERO, rvoff >> 1), ctx);
+		emit_jump_and_link(RV_REG_ZERO, rvoff, ctx);
 		break;
 
 	/* dst = imm64 */
-- 
cgit v1.2.3-59-g8ed1b


From fe8322b866d560a3b3535ed9aa49df59809decbe Mon Sep 17 00:00:00 2001
From: Björn Töpel <bjorn.topel@gmail.com>
Date: Mon, 16 Dec 2019 10:13:39 +0100
Subject: riscv, bpf: Optimize BPF tail calls
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Remove one addi, and instead use the offset part of jalr.

Signed-off-by: Björn Töpel <bjorn.topel@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191216091343.23260-6-bjorn.topel@gmail.com
---
 arch/riscv/net/bpf_jit_comp.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/arch/riscv/net/bpf_jit_comp.c b/arch/riscv/net/bpf_jit_comp.c
index 2fc0f24ad30f..8aa19c846881 100644
--- a/arch/riscv/net/bpf_jit_comp.c
+++ b/arch/riscv/net/bpf_jit_comp.c
@@ -552,7 +552,7 @@ static int epilogue_offset(struct rv_jit_context *ctx)
 	return (to - from) << 2;
 }
 
-static void __build_epilogue(u8 reg, struct rv_jit_context *ctx)
+static void __build_epilogue(bool is_tail_call, struct rv_jit_context *ctx)
 {
 	int stack_adjust = ctx->stack_size, store_offset = stack_adjust - 8;
 
@@ -589,9 +589,11 @@ static void __build_epilogue(u8 reg, struct rv_jit_context *ctx)
 
 	emit(rv_addi(RV_REG_SP, RV_REG_SP, stack_adjust), ctx);
 	/* Set return value. */
-	if (reg == RV_REG_RA)
+	if (!is_tail_call)
 		emit(rv_addi(RV_REG_A0, RV_REG_A5, 0), ctx);
-	emit(rv_jalr(RV_REG_ZERO, reg, 0), ctx);
+	emit(rv_jalr(RV_REG_ZERO, is_tail_call ? RV_REG_T3 : RV_REG_RA,
+		     is_tail_call ? 4 : 0), /* skip TCC init */
+	     ctx);
 }
 
 /* return -1 or inverted cond */
@@ -751,9 +753,8 @@ static int emit_bpf_tail_call(int insn, struct rv_jit_context *ctx)
 	if (is_12b_check(off, insn))
 		return -1;
 	emit(rv_ld(RV_REG_T3, off, RV_REG_T2), ctx);
-	emit(rv_addi(RV_REG_T3, RV_REG_T3, 4), ctx);
 	emit(rv_addi(RV_REG_TCC, RV_REG_T1, 0), ctx);
-	__build_epilogue(RV_REG_T3, ctx);
+	__build_epilogue(true, ctx);
 	return 0;
 }
 
@@ -1504,7 +1505,7 @@ static void build_prologue(struct rv_jit_context *ctx)
 
 static void build_epilogue(struct rv_jit_context *ctx)
 {
-	__build_epilogue(RV_REG_RA, ctx);
+	__build_epilogue(false, ctx);
 }
 
 static int build_body(struct rv_jit_context *ctx, bool extra_pass)
-- 
cgit v1.2.3-59-g8ed1b


From 7f3631e88ee6fb13beac333a0e9e605d2414d17f Mon Sep 17 00:00:00 2001
From: Björn Töpel <bjorn.topel@gmail.com>
Date: Mon, 16 Dec 2019 10:13:40 +0100
Subject: riscv, bpf: Provide RISC-V specific JIT image alloc/free
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This commit makes sure that the JIT images is kept close to the kernel
text, so BPF calls can use relative calling with auipc/jalr or jal
instead of loading the full 64-bit address and jalr.

The BPF JIT image region is 128 MB before the kernel text.

Signed-off-by: Björn Töpel <bjorn.topel@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191216091343.23260-7-bjorn.topel@gmail.com
---
 arch/riscv/include/asm/pgtable.h |  4 ++++
 arch/riscv/net/bpf_jit_comp.c    | 13 +++++++++++++
 2 files changed, 17 insertions(+)

diff --git a/arch/riscv/include/asm/pgtable.h b/arch/riscv/include/asm/pgtable.h
index 7ff0ed4f292e..cc3f49415620 100644
--- a/arch/riscv/include/asm/pgtable.h
+++ b/arch/riscv/include/asm/pgtable.h
@@ -404,6 +404,10 @@ static inline int ptep_clear_flush_young(struct vm_area_struct *vma,
 #define VMALLOC_END      (PAGE_OFFSET - 1)
 #define VMALLOC_START    (PAGE_OFFSET - VMALLOC_SIZE)
 
+#define BPF_JIT_REGION_SIZE	(SZ_128M)
+#define BPF_JIT_REGION_START	(PAGE_OFFSET - BPF_JIT_REGION_SIZE)
+#define BPF_JIT_REGION_END	(VMALLOC_END)
+
 /*
  * Roughly size the vmemmap space to be large enough to fit enough
  * struct pages to map half the virtual address space. Then
diff --git a/arch/riscv/net/bpf_jit_comp.c b/arch/riscv/net/bpf_jit_comp.c
index 8aa19c846881..46cff093f526 100644
--- a/arch/riscv/net/bpf_jit_comp.c
+++ b/arch/riscv/net/bpf_jit_comp.c
@@ -1656,3 +1656,16 @@ out:
 					   tmp : orig_prog);
 	return prog;
 }
+
+void *bpf_jit_alloc_exec(unsigned long size)
+{
+	return __vmalloc_node_range(size, PAGE_SIZE, BPF_JIT_REGION_START,
+				    BPF_JIT_REGION_END, GFP_KERNEL,
+				    PAGE_KERNEL_EXEC, 0, NUMA_NO_NODE,
+				    __builtin_return_address(0));
+}
+
+void bpf_jit_free_exec(void *addr)
+{
+	return vfree(addr);
+}
-- 
cgit v1.2.3-59-g8ed1b


From e368b64f8b0c8066115ad1da71f83f20f7bb43af Mon Sep 17 00:00:00 2001
From: Björn Töpel <bjorn.topel@gmail.com>
Date: Mon, 16 Dec 2019 10:13:41 +0100
Subject: riscv, bpf: Optimize calls
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Instead of using emit_imm() and emit_jalr() which can expand to six
instructions, start using jal or auipc+jalr.

Signed-off-by: Björn Töpel <bjorn.topel@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191216091343.23260-8-bjorn.topel@gmail.com
---
 arch/riscv/net/bpf_jit_comp.c | 101 ++++++++++++++++++++++++++----------------
 1 file changed, 64 insertions(+), 37 deletions(-)

diff --git a/arch/riscv/net/bpf_jit_comp.c b/arch/riscv/net/bpf_jit_comp.c
index 46cff093f526..8d7e3343a08c 100644
--- a/arch/riscv/net/bpf_jit_comp.c
+++ b/arch/riscv/net/bpf_jit_comp.c
@@ -811,11 +811,12 @@ static void emit_sext_32_rd(u8 *rd, struct rv_jit_context *ctx)
 	*rd = RV_REG_T2;
 }
 
-static void emit_jump_and_link(u8 rd, int rvoff, struct rv_jit_context *ctx)
+static void emit_jump_and_link(u8 rd, s64 rvoff, bool force_jalr,
+			       struct rv_jit_context *ctx)
 {
 	s64 upper, lower;
 
-	if (is_21b_int(rvoff)) {
+	if (rvoff && is_21b_int(rvoff) && !force_jalr) {
 		emit(rv_jal(rd, rvoff >> 1), ctx);
 		return;
 	}
@@ -832,6 +833,28 @@ static bool is_signed_bpf_cond(u8 cond)
 		cond == BPF_JSGE || cond == BPF_JSLE;
 }
 
+static int emit_call(bool fixed, u64 addr, struct rv_jit_context *ctx)
+{
+	s64 off = 0;
+	u64 ip;
+	u8 rd;
+
+	if (addr && ctx->insns) {
+		ip = (u64)(long)(ctx->insns + ctx->ninsns);
+		off = addr - ip;
+		if (!is_32b_int(off)) {
+			pr_err("bpf-jit: target call addr %pK is out of range\n",
+			       (void *)addr);
+			return -ERANGE;
+		}
+	}
+
+	emit_jump_and_link(RV_REG_RA, off, !fixed, ctx);
+	rd = bpf_to_rv_reg(BPF_REG_0, ctx);
+	emit(rv_addi(rd, RV_REG_A0, 0), ctx);
+	return 0;
+}
+
 static int emit_insn(const struct bpf_insn *insn, struct rv_jit_context *ctx,
 		     bool extra_pass)
 {
@@ -1107,7 +1130,7 @@ out_be:
 	/* JUMP off */
 	case BPF_JMP | BPF_JA:
 		rvoff = rv_offset(i, off, ctx);
-		emit_jump_and_link(RV_REG_ZERO, rvoff, ctx);
+		emit_jump_and_link(RV_REG_ZERO, rvoff, false, ctx);
 		break;
 
 	/* IF (dst COND src) JUMP off */
@@ -1209,7 +1232,7 @@ out_be:
 	case BPF_JMP | BPF_CALL:
 	{
 		bool fixed;
-		int i, ret;
+		int ret;
 		u64 addr;
 
 		mark_call(ctx);
@@ -1217,20 +1240,9 @@ out_be:
 					    &fixed);
 		if (ret < 0)
 			return ret;
-		if (fixed) {
-			emit_imm(RV_REG_T1, addr, ctx);
-		} else {
-			i = ctx->ninsns;
-			emit_imm(RV_REG_T1, addr, ctx);
-			for (i = ctx->ninsns - i; i < 8; i++) {
-				/* nop */
-				emit(rv_addi(RV_REG_ZERO, RV_REG_ZERO, 0),
-				     ctx);
-			}
-		}
-		emit(rv_jalr(RV_REG_RA, RV_REG_T1, 0), ctx);
-		rd = bpf_to_rv_reg(BPF_REG_0, ctx);
-		emit(rv_addi(rd, RV_REG_A0, 0), ctx);
+		ret = emit_call(fixed, addr, ctx);
+		if (ret)
+			return ret;
 		break;
 	}
 	/* tail call */
@@ -1245,7 +1257,7 @@ out_be:
 			break;
 
 		rvoff = epilogue_offset(ctx);
-		emit_jump_and_link(RV_REG_ZERO, rvoff, ctx);
+		emit_jump_and_link(RV_REG_ZERO, rvoff, false, ctx);
 		break;
 
 	/* dst = imm64 */
@@ -1508,7 +1520,7 @@ static void build_epilogue(struct rv_jit_context *ctx)
 	__build_epilogue(false, ctx);
 }
 
-static int build_body(struct rv_jit_context *ctx, bool extra_pass)
+static int build_body(struct rv_jit_context *ctx, bool extra_pass, int *offset)
 {
 	const struct bpf_prog *prog = ctx->prog;
 	int i;
@@ -1520,12 +1532,12 @@ static int build_body(struct rv_jit_context *ctx, bool extra_pass)
 		ret = emit_insn(insn, ctx, extra_pass);
 		if (ret > 0) {
 			i++;
-			if (ctx->insns == NULL)
-				ctx->offset[i] = ctx->ninsns;
+			if (offset)
+				offset[i] = ctx->ninsns;
 			continue;
 		}
-		if (ctx->insns == NULL)
-			ctx->offset[i] = ctx->ninsns;
+		if (offset)
+			offset[i] = ctx->ninsns;
 		if (ret)
 			return ret;
 	}
@@ -1553,8 +1565,8 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *prog)
 	struct bpf_prog *tmp, *orig_prog = prog;
 	int pass = 0, prev_ninsns = 0, i;
 	struct rv_jit_data *jit_data;
+	unsigned int image_size = 0;
 	struct rv_jit_context *ctx;
-	unsigned int image_size;
 
 	if (!prog->jit_requested)
 		return orig_prog;
@@ -1599,36 +1611,51 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *prog)
 	for (i = 0; i < 16; i++) {
 		pass++;
 		ctx->ninsns = 0;
-		if (build_body(ctx, extra_pass)) {
+		if (build_body(ctx, extra_pass, ctx->offset)) {
 			prog = orig_prog;
 			goto out_offset;
 		}
 		build_prologue(ctx);
 		ctx->epilogue_offset = ctx->ninsns;
 		build_epilogue(ctx);
-		if (ctx->ninsns == prev_ninsns)
-			break;
+
+		if (ctx->ninsns == prev_ninsns) {
+			if (jit_data->header)
+				break;
+
+			image_size = sizeof(u32) * ctx->ninsns;
+			jit_data->header =
+				bpf_jit_binary_alloc(image_size,
+						     &jit_data->image,
+						     sizeof(u32),
+						     bpf_fill_ill_insns);
+			if (!jit_data->header) {
+				prog = orig_prog;
+				goto out_offset;
+			}
+
+			ctx->insns = (u32 *)jit_data->image;
+			/* Now, when the image is allocated, the image
+			 * can potentially shrink more (auipc/jalr ->
+			 * jal).
+			 */
+		}
 		prev_ninsns = ctx->ninsns;
 	}
 
-	/* Allocate image, now that we know the size. */
-	image_size = sizeof(u32) * ctx->ninsns;
-	jit_data->header = bpf_jit_binary_alloc(image_size, &jit_data->image,
-						sizeof(u32),
-						bpf_fill_ill_insns);
-	if (!jit_data->header) {
+	if (i == 16) {
+		pr_err("bpf-jit: image did not converge in <%d passes!\n", i);
+		bpf_jit_binary_free(jit_data->header);
 		prog = orig_prog;
 		goto out_offset;
 	}
 
-	/* Second, real pass, that acutally emits the image. */
-	ctx->insns = (u32 *)jit_data->image;
 skip_init_ctx:
 	pass++;
 	ctx->ninsns = 0;
 
 	build_prologue(ctx);
-	if (build_body(ctx, extra_pass)) {
+	if (build_body(ctx, extra_pass, NULL)) {
 		bpf_jit_binary_free(jit_data->header);
 		prog = orig_prog;
 		goto out_offset;
-- 
cgit v1.2.3-59-g8ed1b


From eb9928bed003dd61a443d0ba51ae066429fbe735 Mon Sep 17 00:00:00 2001
From: Björn Töpel <bjorn.topel@gmail.com>
Date: Mon, 16 Dec 2019 10:13:42 +0100
Subject: riscv, bpf: Add missing uapi header for BPF_PROG_TYPE_PERF_EVENT
 programs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add missing uapi header the BPF_PROG_TYPE_PERF_EVENT programs by
exporting struct user_regs_struct instead of struct pt_regs which is
in-kernel only.

Signed-off-by: Björn Töpel <bjorn.topel@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191216091343.23260-9-bjorn.topel@gmail.com
---
 arch/riscv/include/uapi/asm/bpf_perf_event.h | 9 +++++++++
 tools/include/uapi/asm/bpf_perf_event.h      | 2 ++
 2 files changed, 11 insertions(+)
 create mode 100644 arch/riscv/include/uapi/asm/bpf_perf_event.h

diff --git a/arch/riscv/include/uapi/asm/bpf_perf_event.h b/arch/riscv/include/uapi/asm/bpf_perf_event.h
new file mode 100644
index 000000000000..6cb1c2823288
--- /dev/null
+++ b/arch/riscv/include/uapi/asm/bpf_perf_event.h
@@ -0,0 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
+#ifndef _UAPI__ASM_BPF_PERF_EVENT_H__
+#define _UAPI__ASM_BPF_PERF_EVENT_H__
+
+#include <asm/ptrace.h>
+
+typedef struct user_regs_struct bpf_user_pt_regs_t;
+
+#endif /* _UAPI__ASM_BPF_PERF_EVENT_H__ */
diff --git a/tools/include/uapi/asm/bpf_perf_event.h b/tools/include/uapi/asm/bpf_perf_event.h
index 13a58531e6fa..39acc149d843 100644
--- a/tools/include/uapi/asm/bpf_perf_event.h
+++ b/tools/include/uapi/asm/bpf_perf_event.h
@@ -2,6 +2,8 @@
 #include "../../arch/arm64/include/uapi/asm/bpf_perf_event.h"
 #elif defined(__s390__)
 #include "../../arch/s390/include/uapi/asm/bpf_perf_event.h"
+#elif defined(__riscv)
+#include "../../arch/riscv/include/uapi/asm/bpf_perf_event.h"
 #else
 #include <uapi/asm-generic/bpf_perf_event.h>
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From 34bfc10a6e7e573563a700239d302ab5944b3397 Mon Sep 17 00:00:00 2001
From: Björn Töpel <bjorn.topel@gmail.com>
Date: Mon, 16 Dec 2019 10:13:43 +0100
Subject: riscv, perf: Add arch specific perf_arch_bpf_user_pt_regs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

RISC-V was missing a proper perf_arch_bpf_user_pt_regs macro for
CONFIG_PERF_EVENT builds.

Signed-off-by: Björn Töpel <bjorn.topel@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191216091343.23260-10-bjorn.topel@gmail.com
---
 arch/riscv/include/asm/perf_event.h | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/arch/riscv/include/asm/perf_event.h b/arch/riscv/include/asm/perf_event.h
index aefbfaa6a781..0234048b12bc 100644
--- a/arch/riscv/include/asm/perf_event.h
+++ b/arch/riscv/include/asm/perf_event.h
@@ -82,4 +82,8 @@ struct riscv_pmu {
 	int		irq;
 };
 
+#ifdef CONFIG_PERF_EVENTS
+#define perf_arch_bpf_user_pt_regs(regs) (struct user_regs_struct *)regs
+#endif
+
 #endif /* _ASM_RISCV_PERF_EVENT_H */
-- 
cgit v1.2.3-59-g8ed1b


From 12dd14b230b3c742b80272ecb8a83cdf824625ca Mon Sep 17 00:00:00 2001
From: Toke Høiland-Jørgensen <toke@redhat.com>
Date: Thu, 19 Dec 2019 13:07:14 +0100
Subject: libbpf: Add missing newline in opts validation macro
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The error log output in the opts validation macro was missing a newline.

Fixes: 2ce8450ef5a3 ("libbpf: add bpf_object__open_{file, mem} w/ extensible opts")
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191219120714.928380-1-toke@redhat.com
---
 tools/lib/bpf/libbpf_internal.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/lib/bpf/libbpf_internal.h b/tools/lib/bpf/libbpf_internal.h
index 7d71621bb7e8..8c3afbd97747 100644
--- a/tools/lib/bpf/libbpf_internal.h
+++ b/tools/lib/bpf/libbpf_internal.h
@@ -76,7 +76,7 @@ static inline bool libbpf_validate_opts(const char *opts,
 
 		for (i = opts_sz; i < user_sz; i++) {
 			if (opts[i]) {
-				pr_warn("%s has non-zero extra bytes",
+				pr_warn("%s has non-zero extra bytes\n",
 					type_name);
 				return false;
 			}
-- 
cgit v1.2.3-59-g8ed1b


From 580205dd4fe800b1e95be8b6df9e2991f975a8ad Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Wed, 18 Dec 2019 18:04:42 -0800
Subject: selftests/bpf: Fix test_attach_probe

Fix two issues in test_attach_probe:

1. it was not able to parse /proc/self/maps beyond the first line,
   since %s means parse string until white space.
2. offset has to be accounted for otherwise uprobed address is incorrect.

Fixes: 1e8611bbdfc9 ("selftests/bpf: add kprobe/uprobe selftests")
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Yonghong Song <yhs@fb.com>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/20191219020442.1922617-1-ast@kernel.org
---
 tools/testing/selftests/bpf/prog_tests/attach_probe.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/attach_probe.c b/tools/testing/selftests/bpf/prog_tests/attach_probe.c
index 5ed90ede2f1d..a0ee87c8e1ea 100644
--- a/tools/testing/selftests/bpf/prog_tests/attach_probe.c
+++ b/tools/testing/selftests/bpf/prog_tests/attach_probe.c
@@ -3,7 +3,7 @@
 #include "test_attach_probe.skel.h"
 
 ssize_t get_base_addr() {
-	size_t start;
+	size_t start, offset;
 	char buf[256];
 	FILE *f;
 
@@ -11,10 +11,11 @@ ssize_t get_base_addr() {
 	if (!f)
 		return -errno;
 
-	while (fscanf(f, "%zx-%*x %s %*s\n", &start, buf) == 2) {
+	while (fscanf(f, "%zx-%*x %s %zx %*[^\n]\n",
+		      &start, buf, &offset) == 3) {
 		if (strcmp(buf, "r-xp") == 0) {
 			fclose(f);
-			return start;
+			return start - offset;
 		}
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From b5c7d0d0f7d4a30ae96c1350df677cb849060b7c Mon Sep 17 00:00:00 2001
From: Toke Høiland-Jørgensen <toke@redhat.com>
Date: Thu, 19 Dec 2019 10:02:36 +0100
Subject: libbpf: Fix printing of ulimit value
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Naresh pointed out that libbpf builds fail on 32-bit architectures because
rlimit.rlim_cur is defined as 'unsigned long long' on those architectures.
Fix this by using %zu in printf and casting to size_t.

Fixes: dc3a2d254782 ("libbpf: Print hint about ulimit when getting permission denied error")
Reported-by: Naresh Kamboju <naresh.kamboju@linaro.org>
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191219090236.905059-1-toke@redhat.com
---
 tools/lib/bpf/libbpf.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 6340b81b555b..febbaac3daf4 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -117,7 +117,7 @@ static void pr_perm_msg(int err)
 		return;
 
 	if (limit.rlim_cur < 1024)
-		snprintf(buf, sizeof(buf), "%lu bytes", limit.rlim_cur);
+		snprintf(buf, sizeof(buf), "%zu bytes", (size_t)limit.rlim_cur);
 	else if (limit.rlim_cur < 1024*1024)
 		snprintf(buf, sizeof(buf), "%.1f KiB", (double)limit.rlim_cur / 1024);
 	else
-- 
cgit v1.2.3-59-g8ed1b


From 7745ff9842617323adbe24e71c495d5ebd9aa352 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Wed, 18 Dec 2019 21:21:03 -0800
Subject: libbpf: Fix another __u64 printf warning

Fix yet another printf warning for %llu specifier on ppc64le. This time size_t
casting won't work, so cast to verbose `unsigned long long`.

Fixes: 166750bc1dd2 ("libbpf: Support libbpf-provided extern variables")
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191219052103.3515-1-andriin@fb.com
---
 tools/lib/bpf/libbpf.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index febbaac3daf4..9576a90c5a1c 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -1129,12 +1129,12 @@ static int set_ext_value_num(struct extern_desc *ext, void *ext_val,
 {
 	if (ext->type != EXT_INT && ext->type != EXT_CHAR) {
 		pr_warn("extern %s=%llu should be integer\n",
-			ext->name, value);
+			ext->name, (unsigned long long)value);
 		return -EINVAL;
 	}
 	if (!is_ext_value_in_range(ext, value)) {
 		pr_warn("extern %s=%llu value doesn't fit in %d bytes\n",
-			ext->name, value, ext->sz);
+			ext->name, (unsigned long long)value, ext->sz);
 		return -ERANGE;
 	}
 	switch (ext->sz) {
-- 
cgit v1.2.3-59-g8ed1b


From 7be3c2331d506a61da78cfe426a037a3c7de7c71 Mon Sep 17 00:00:00 2001
From: Ahmad Masri <amasri@codeaurora.org>
Date: Wed, 18 Dec 2019 20:10:04 +0200
Subject: wil6210: dump Rx status message on errors

Dump all the Rx status message on different errors to allow more
visibility of the case.

Signed-off-by: Ahmad Masri <amasri@codeaurora.org>
Signed-off-by: Maya Erez <merez@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/wil6210/txrx_edma.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/net/wireless/ath/wil6210/txrx_edma.c b/drivers/net/wireless/ath/wil6210/txrx_edma.c
index 02548d40253c..7bfe867c7509 100644
--- a/drivers/net/wireless/ath/wil6210/txrx_edma.c
+++ b/drivers/net/wireless/ath/wil6210/txrx_edma.c
@@ -903,6 +903,11 @@ again:
 	if (unlikely(!wil_val_in_range(buff_id, 1, wil->rx_buff_mgmt.size))) {
 		wil_err(wil, "Corrupt buff_id=%d, sring->swhead=%d\n",
 			buff_id, sring->swhead);
+		print_hex_dump(KERN_ERR, "RxS ", DUMP_PREFIX_OFFSET, 16, 1,
+			       msg, wil->use_compressed_rx_status ?
+			       sizeof(struct wil_rx_status_compressed) :
+			       sizeof(struct wil_rx_status_extended), false);
+
 		wil_rx_status_reset_buff_id(sring);
 		wil_sring_advance_swhead(sring);
 		sring->invalid_buff_id_cnt++;
@@ -963,6 +968,11 @@ again:
 
 	if (unlikely(dmalen > sz)) {
 		wil_err(wil, "Rx size too large: %d bytes!\n", dmalen);
+		print_hex_dump(KERN_ERR, "RxS ", DUMP_PREFIX_OFFSET, 16, 1,
+			       msg, wil->use_compressed_rx_status ?
+			       sizeof(struct wil_rx_status_compressed) :
+			       sizeof(struct wil_rx_status_extended), false);
+
 		stats->rx_large_frame++;
 		rxdata->skipping = true;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From d3214d42806676e47eec2e5a5ed041ec40b9dd35 Mon Sep 17 00:00:00 2001
From: Alexei Avshalom Lazar <ailizaro@codeaurora.org>
Date: Wed, 18 Dec 2019 20:10:07 +0200
Subject: wil6210: minimize the time that mem_lock is held

mem_lock is taken for the entire wil_reset().
Optimize this by taking mem_lock just before device is
being reset and release the lock after FW download.

Signed-off-by: Alexei Avshalom Lazar <ailizaro@codeaurora.org>
Signed-off-by: Maya Erez <merez@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/wil6210/main.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/ath/wil6210/main.c b/drivers/net/wireless/ath/wil6210/main.c
index 06091d8a9e23..6d39547e20fd 100644
--- a/drivers/net/wireless/ath/wil6210/main.c
+++ b/drivers/net/wireless/ath/wil6210/main.c
@@ -1654,6 +1654,8 @@ int wil_reset(struct wil6210_priv *wil, bool load_fw)
 	/* Disable device led before reset*/
 	wmi_led_cfg(wil, false);
 
+	down_write(&wil->mem_lock);
+
 	/* prevent NAPI from being scheduled and prevent wmi commands */
 	mutex_lock(&wil->wmi_mutex);
 	if (test_bit(wil_status_suspending, wil->status))
@@ -1702,6 +1704,7 @@ int wil_reset(struct wil6210_priv *wil, bool load_fw)
 
 		if  (wil->secured_boot) {
 			wil_err(wil, "secured boot is not supported\n");
+			up_write(&wil->mem_lock);
 			return -ENOTSUPP;
 		}
 
@@ -1737,6 +1740,8 @@ int wil_reset(struct wil6210_priv *wil, bool load_fw)
 
 	clear_bit(wil_status_resetting, wil->status);
 
+	up_write(&wil->mem_lock);
+
 	if (load_fw) {
 		wil_unmask_irq(wil);
 
@@ -1786,6 +1791,7 @@ int wil_reset(struct wil6210_priv *wil, bool load_fw)
 	return rc;
 
 out:
+	up_write(&wil->mem_lock);
 	clear_bit(wil_status_resetting, wil->status);
 	return rc;
 }
@@ -1811,9 +1817,7 @@ int __wil_up(struct wil6210_priv *wil)
 
 	WARN_ON(!mutex_is_locked(&wil->mutex));
 
-	down_write(&wil->mem_lock);
 	rc = wil_reset(wil, true);
-	up_write(&wil->mem_lock);
 	if (rc)
 		return rc;
 
@@ -1905,9 +1909,7 @@ int __wil_down(struct wil6210_priv *wil)
 	wil_abort_scan_all_vifs(wil, false);
 	mutex_unlock(&wil->vif_mutex);
 
-	down_write(&wil->mem_lock);
 	rc = wil_reset(wil, false);
-	up_write(&wil->mem_lock);
 
 	return rc;
 }
-- 
cgit v1.2.3-59-g8ed1b


From c715b38473c842db34c408ede2d5727d2b50cf1c Mon Sep 17 00:00:00 2001
From: Alexei Avshalom Lazar <ailizaro@codeaurora.org>
Date: Wed, 18 Dec 2019 20:10:10 +0200
Subject: wil6210: take mem_lock for writing in crash dump collection

On some crash dump cases mem_lock is already taken, error
returns and crash dump copy fails.
In this case wait until mem_lock available instead of failing
the operation.
Also take the mem_lock for writing to prevent other threads from
altering the state of the device while collecting crash dump.

Signed-off-by: Alexei Avshalom Lazar <ailizaro@codeaurora.org>
Signed-off-by: Maya Erez <merez@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/wil6210/wil_crash_dump.c | 17 ++++++++++++-----
 1 file changed, 12 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/ath/wil6210/wil_crash_dump.c b/drivers/net/wireless/ath/wil6210/wil_crash_dump.c
index 1332eb8c831f..89c12cb2aaab 100644
--- a/drivers/net/wireless/ath/wil6210/wil_crash_dump.c
+++ b/drivers/net/wireless/ath/wil6210/wil_crash_dump.c
@@ -46,7 +46,7 @@ static int wil_fw_get_crash_dump_bounds(struct wil6210_priv *wil,
 
 int wil_fw_copy_crash_dump(struct wil6210_priv *wil, void *dest, u32 size)
 {
-	int i, rc;
+	int i;
 	const struct fw_map *map;
 	void *data;
 	u32 host_min, dump_size, offset, len;
@@ -62,9 +62,15 @@ int wil_fw_copy_crash_dump(struct wil6210_priv *wil, void *dest, u32 size)
 		return -EINVAL;
 	}
 
-	rc = wil_mem_access_lock(wil);
-	if (rc)
-		return rc;
+	down_write(&wil->mem_lock);
+
+	if (test_bit(wil_status_suspending, wil->status) ||
+	    test_bit(wil_status_suspended, wil->status)) {
+		wil_err(wil,
+			"suspend/resume in progress. cannot copy crash dump\n");
+		up_write(&wil->mem_lock);
+		return -EBUSY;
+	}
 
 	/* copy to crash dump area */
 	for (i = 0; i < ARRAY_SIZE(fw_mapping); i++) {
@@ -84,7 +90,8 @@ int wil_fw_copy_crash_dump(struct wil6210_priv *wil, void *dest, u32 size)
 		wil_memcpy_fromio_32((void * __force)(dest + offset),
 				     (const void __iomem * __force)data, len);
 	}
-	wil_mem_access_unlock(wil);
+
+	up_write(&wil->mem_lock);
 
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 5c5d88a29dd4fe8154c8d262163a96d0add70917 Mon Sep 17 00:00:00 2001
From: Alexei Avshalom Lazar <ailizaro@codeaurora.org>
Date: Wed, 18 Dec 2019 20:10:13 +0200
Subject: wil6210: add verification for cid upper bound

max_assoc_sta can receive values (from the user or from the FW)
that are higher than WIL6210_MAX_CID.
Verify that cid doesn't exceed the upper bound of WIL6210_MAX_CID.

Signed-off-by: Alexei Avshalom Lazar <ailizaro@codeaurora.org>
Signed-off-by: Maya Erez <merez@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/wil6210/wil6210.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/wil6210/wil6210.h b/drivers/net/wireless/ath/wil6210/wil6210.h
index 97626bfd4dac..bf00f0693a3d 100644
--- a/drivers/net/wireless/ath/wil6210/wil6210.h
+++ b/drivers/net/wireless/ath/wil6210/wil6210.h
@@ -1148,7 +1148,7 @@ static inline void wil_c(struct wil6210_priv *wil, u32 reg, u32 val)
  */
 static inline bool wil_cid_valid(struct wil6210_priv *wil, int cid)
 {
-	return (cid >= 0 && cid < wil->max_assoc_sta);
+	return (cid >= 0 && cid < wil->max_assoc_sta && cid < WIL6210_MAX_CID);
 }
 
 void wil_get_board_file(struct wil6210_priv *wil, char *buf, size_t len);
-- 
cgit v1.2.3-59-g8ed1b


From c99d5f8d61784f152fb3c2e217fa17b05bbd640a Mon Sep 17 00:00:00 2001
From: Dedy Lansky <dlansky@codeaurora.org>
Date: Wed, 18 Dec 2019 20:10:16 +0200
Subject: wil6210: reduce ucode_debug memory region

ucode_debug memory region defined as 4K bytes. Fix this according to
Talyn device memory map.

Signed-off-by: Dedy Lansky <dlansky@codeaurora.org>
Signed-off-by: Maya Erez <merez@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/wil6210/wmi.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/wil6210/wmi.c b/drivers/net/wireless/ath/wil6210/wmi.c
index 7a0d934eb271..3aca86b3071d 100644
--- a/drivers/net/wireless/ath/wil6210/wmi.c
+++ b/drivers/net/wireless/ath/wil6210/wmi.c
@@ -196,8 +196,8 @@ const struct fw_map talyn_mb_fw_mapping[] = {
 	{0x8c0000, 0x8c0210, 0x8c0000, "dum_user_rgf", true, true},
 	/* DMA OFU 296b */
 	{0x8c2000, 0x8c2128, 0x8c2000, "dma_ofu", true, true},
-	/* ucode debug 4k */
-	{0x8c3000, 0x8c4000, 0x8c3000, "ucode_debug", true, true},
+	/* ucode debug 256b */
+	{0x8c3000, 0x8c3100, 0x8c3000, "ucode_debug", true, true},
 	/* upper area 1536k */
 	{0x900000, 0xa80000, 0x900000, "upper", true, true},
 	/* UCODE areas - accessible by debugfs blobs but not by
-- 
cgit v1.2.3-59-g8ed1b


From 18beb61d848d267371a769114d90bd1be6239eda Mon Sep 17 00:00:00 2001
From: Dedy Lansky <dlansky@codeaurora.org>
Date: Wed, 18 Dec 2019 20:10:18 +0200
Subject: wil6210: fix MID valid bits in Rx status message

Fix incorrect definitions of MAC ID bits inside Rx status message.

Signed-off-by: Dedy Lansky <dlansky@codeaurora.org>
Signed-off-by: Maya Erez <merez@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/wil6210/txrx_edma.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/ath/wil6210/txrx_edma.h b/drivers/net/wireless/ath/wil6210/txrx_edma.h
index c744c65225da..c736f7413a35 100644
--- a/drivers/net/wireless/ath/wil6210/txrx_edma.h
+++ b/drivers/net/wireless/ath/wil6210/txrx_edma.h
@@ -46,7 +46,7 @@
 
 #define WIL_RX_EDMA_DLPF_LU_MISS_TID_POS	5
 
-#define WIL_RX_EDMA_MID_VALID_BIT		BIT(22)
+#define WIL_RX_EDMA_MID_VALID_BIT		BIT(20)
 
 #define WIL_EDMA_DESC_TX_MAC_CFG_0_QID_POS 16
 #define WIL_EDMA_DESC_TX_MAC_CFG_0_QID_LEN 6
@@ -244,8 +244,8 @@ struct wil_ring_tx_status {
  *		     calculated, Bit1- L4Err - TCP/UDP Checksum Error
  *	bit      7 : Reserved:1
  *	bit  8..19 : Flow ID:12 - MSDU flow ID
- *	bit 20..21 : MID:2 - The MAC ID
- *	bit     22 : MID_V:1 - The MAC ID field is valid
+ *	bit     20 : MID_V:1 - The MAC ID field is valid
+ *	bit 21..22 : MID:2 - The MAC ID
  *	bit     23 : L3T:1 - IP types: 0-IPv6, 1-IPv4
  *	bit     24 : L4T:1 - Layer 4 Type: 0-UDP, 1-TCP
  *	bit     25 : BC:1 - The received MPDU is broadcast
@@ -479,7 +479,7 @@ static inline int wil_rx_status_get_mid(void *msg)
 		return 0; /* use the default MID */
 
 	return WIL_GET_BITS(((struct wil_rx_status_compressed *)msg)->d0,
-			    20, 21);
+			    21, 22);
 }
 
 static inline int wil_rx_status_get_error(void *msg)
-- 
cgit v1.2.3-59-g8ed1b


From 5e5f069c30ccdbc09bd548858ee75d0487e74490 Mon Sep 17 00:00:00 2001
From: Ahmad Masri <amasri@codeaurora.org>
Date: Wed, 18 Dec 2019 20:10:21 +0200
Subject: wil6210: support set_multicast_to_unicast cfg80211 operation

Wil6210 AP has a separate ring for transmitting multicast packets,
multicast packets are transmitted without an ack from the receiver side.
Therefore, 802.11 spec defines some low MCS rates for multicat packets.
However, there is no guarantee that these packets were really received
and handled on the client side.

Some applications that rely on multicast packets, may prefer to
transmit these packets as a unicast to ensure reliability, and also
to ensure better performance with high MCS rates.
multicast to unicast is done by duplicating multicast packets to all
clients and changing the DA (multicast) to the MAC address of the
client.
see NL80211_CMD_SET_MULTICAST_TO_UNICAST for more info.

Signed-off-by: Ahmad Masri <amasri@codeaurora.org>
Signed-off-by: Maya Erez <merez@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/wil6210/cfg80211.c | 16 +++++++++++++++
 drivers/net/wireless/ath/wil6210/txrx.c     | 32 ++++++++++++++++++++++++++++-
 drivers/net/wireless/ath/wil6210/wil6210.h  |  1 +
 3 files changed, 48 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/wil6210/cfg80211.c b/drivers/net/wireless/ath/wil6210/cfg80211.c
index 7d6f14420855..5a3aff131472 100644
--- a/drivers/net/wireless/ath/wil6210/cfg80211.c
+++ b/drivers/net/wireless/ath/wil6210/cfg80211.c
@@ -2579,6 +2579,21 @@ wil_cfg80211_update_ft_ies(struct wiphy *wiphy, struct net_device *dev,
 	return rc;
 }
 
+static int wil_cfg80211_set_multicast_to_unicast(struct wiphy *wiphy,
+						 struct net_device *dev,
+						 const bool enabled)
+{
+	struct wil6210_priv *wil = wiphy_to_wil(wiphy);
+
+	if (wil->multicast_to_unicast == enabled)
+		return 0;
+
+	wil_info(wil, "set multicast to unicast, enabled=%d\n", enabled);
+	wil->multicast_to_unicast = enabled;
+
+	return 0;
+}
+
 static const struct cfg80211_ops wil_cfg80211_ops = {
 	.add_virtual_intf = wil_cfg80211_add_iface,
 	.del_virtual_intf = wil_cfg80211_del_iface,
@@ -2615,6 +2630,7 @@ static const struct cfg80211_ops wil_cfg80211_ops = {
 	.sched_scan_start = wil_cfg80211_sched_scan_start,
 	.sched_scan_stop = wil_cfg80211_sched_scan_stop,
 	.update_ft_ies = wil_cfg80211_update_ft_ies,
+	.set_multicast_to_unicast = wil_cfg80211_set_multicast_to_unicast,
 };
 
 static void wil_wiphy_init(struct wiphy *wiphy)
diff --git a/drivers/net/wireless/ath/wil6210/txrx.c b/drivers/net/wireless/ath/wil6210/txrx.c
index 8ebc6d59aa74..17118d643d7e 100644
--- a/drivers/net/wireless/ath/wil6210/txrx.c
+++ b/drivers/net/wireless/ath/wil6210/txrx.c
@@ -10,6 +10,7 @@
 #include <linux/moduleparam.h>
 #include <linux/ip.h>
 #include <linux/ipv6.h>
+#include <linux/if_vlan.h>
 #include <net/ipv6.h>
 #include <linux/prefetch.h>
 
@@ -1529,6 +1530,35 @@ static struct wil_ring *wil_find_tx_bcast_1(struct wil6210_priv *wil,
 	return v;
 }
 
+/* apply multicast to unicast only for ARP and IP packets
+ * (see NL80211_CMD_SET_MULTICAST_TO_UNICAST for more info)
+ */
+static bool wil_check_multicast_to_unicast(struct wil6210_priv *wil,
+					   struct sk_buff *skb)
+{
+	const struct ethhdr *eth = (void *)skb->data;
+	const struct vlan_ethhdr *ethvlan = (void *)skb->data;
+	__be16 ethertype;
+
+	if (!wil->multicast_to_unicast)
+		return false;
+
+	/* multicast to unicast conversion only for some payload */
+	ethertype = eth->h_proto;
+	if (ethertype == htons(ETH_P_8021Q) && skb->len >= VLAN_ETH_HLEN)
+		ethertype = ethvlan->h_vlan_encapsulated_proto;
+	switch (ethertype) {
+	case htons(ETH_P_ARP):
+	case htons(ETH_P_IP):
+	case htons(ETH_P_IPV6):
+		break;
+	default:
+		return false;
+	}
+
+	return true;
+}
+
 static void wil_set_da_for_vring(struct wil6210_priv *wil,
 				 struct sk_buff *skb, int vring_index)
 {
@@ -2336,7 +2366,7 @@ netdev_tx_t wil_start_xmit(struct sk_buff *skb, struct net_device *ndev)
 		/* in STA mode (ESS), all to same VRING (to AP) */
 		ring = wil_find_tx_ring_sta(wil, vif, skb);
 	} else if (bcast) {
-		if (vif->pbss)
+		if (vif->pbss || wil_check_multicast_to_unicast(wil, skb))
 			/* in pbss, no bcast VRING - duplicate skb in
 			 * all stations VRINGs
 			 */
diff --git a/drivers/net/wireless/ath/wil6210/wil6210.h b/drivers/net/wireless/ath/wil6210/wil6210.h
index bf00f0693a3d..21399ced2a5d 100644
--- a/drivers/net/wireless/ath/wil6210/wil6210.h
+++ b/drivers/net/wireless/ath/wil6210/wil6210.h
@@ -1059,6 +1059,7 @@ struct wil6210_priv {
 
 	u32 max_agg_wsize;
 	u32 max_ampdu_size;
+	u8 multicast_to_unicast;
 };
 
 #define wil_to_wiphy(i) (i->wiphy)
-- 
cgit v1.2.3-59-g8ed1b


From 4315a74a482f86750a1e78ac536c5baeb05ff6b1 Mon Sep 17 00:00:00 2001
From: Dedy Lansky <dlansky@codeaurora.org>
Date: Wed, 18 Dec 2019 20:10:24 +0200
Subject: wil6210: add support for set_cqm_rssi_config

set_cqm_rssi_config() is used by the kernel to configure connection
quality monitor RSSI threshold.
wil6210 uses WMI_SET_LINK_MONITOR_CMDID to set the RSSI threshold to
FW which in turn reports RSSI threshold changes with
WMI_LINK_MONITOR_EVENTID.

Signed-off-by: Dedy Lansky <dlansky@codeaurora.org>
Signed-off-by: Maya Erez <merez@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/wil6210/cfg80211.c | 18 +++++++
 drivers/net/wireless/ath/wil6210/wil6210.h  |  3 ++
 drivers/net/wireless/ath/wil6210/wmi.c      | 76 +++++++++++++++++++++++++++++
 drivers/net/wireless/ath/wil6210/wmi.h      | 33 +++++++++++++
 4 files changed, 130 insertions(+)

diff --git a/drivers/net/wireless/ath/wil6210/cfg80211.c b/drivers/net/wireless/ath/wil6210/cfg80211.c
index 5a3aff131472..0851d2bede89 100644
--- a/drivers/net/wireless/ath/wil6210/cfg80211.c
+++ b/drivers/net/wireless/ath/wil6210/cfg80211.c
@@ -2594,6 +2594,23 @@ static int wil_cfg80211_set_multicast_to_unicast(struct wiphy *wiphy,
 	return 0;
 }
 
+static int wil_cfg80211_set_cqm_rssi_config(struct wiphy *wiphy,
+					    struct net_device *dev,
+					    s32 rssi_thold, u32 rssi_hyst)
+{
+	struct wil6210_priv *wil = wiphy_to_wil(wiphy);
+	int rc;
+
+	wil->cqm_rssi_thold = rssi_thold;
+
+	rc = wmi_set_cqm_rssi_config(wil, rssi_thold, rssi_hyst);
+	if (rc)
+		/* reset stored value upon failure */
+		wil->cqm_rssi_thold = 0;
+
+	return rc;
+}
+
 static const struct cfg80211_ops wil_cfg80211_ops = {
 	.add_virtual_intf = wil_cfg80211_add_iface,
 	.del_virtual_intf = wil_cfg80211_del_iface,
@@ -2625,6 +2642,7 @@ static const struct cfg80211_ops wil_cfg80211_ops = {
 	.start_p2p_device = wil_cfg80211_start_p2p_device,
 	.stop_p2p_device = wil_cfg80211_stop_p2p_device,
 	.set_power_mgmt = wil_cfg80211_set_power_mgmt,
+	.set_cqm_rssi_config = wil_cfg80211_set_cqm_rssi_config,
 	.suspend = wil_cfg80211_suspend,
 	.resume = wil_cfg80211_resume,
 	.sched_scan_start = wil_cfg80211_sched_scan_start,
diff --git a/drivers/net/wireless/ath/wil6210/wil6210.h b/drivers/net/wireless/ath/wil6210/wil6210.h
index 21399ced2a5d..5dc881d3c057 100644
--- a/drivers/net/wireless/ath/wil6210/wil6210.h
+++ b/drivers/net/wireless/ath/wil6210/wil6210.h
@@ -1060,6 +1060,7 @@ struct wil6210_priv {
 	u32 max_agg_wsize;
 	u32 max_ampdu_size;
 	u8 multicast_to_unicast;
+	s32 cqm_rssi_thold;
 };
 
 #define wil_to_wiphy(i) (i->wiphy)
@@ -1441,4 +1442,6 @@ int wmi_addba_rx_resp_edma(struct wil6210_priv *wil, u8 mid, u8 cid,
 void update_supported_bands(struct wil6210_priv *wil);
 
 void wil_clear_fw_log_addr(struct wil6210_priv *wil);
+int wmi_set_cqm_rssi_config(struct wil6210_priv *wil,
+			    s32 rssi_thold, u32 rssi_hyst);
 #endif /* __WIL6210_H__ */
diff --git a/drivers/net/wireless/ath/wil6210/wmi.c b/drivers/net/wireless/ath/wil6210/wmi.c
index 3aca86b3071d..dcba0a4c47b4 100644
--- a/drivers/net/wireless/ath/wil6210/wmi.c
+++ b/drivers/net/wireless/ath/wil6210/wmi.c
@@ -476,6 +476,8 @@ static const char *cmdid2name(u16 cmdid)
 		return "WMI_RBUFCAP_CFG_CMD";
 	case WMI_TEMP_SENSE_ALL_CMDID:
 		return "WMI_TEMP_SENSE_ALL_CMDID";
+	case WMI_SET_LINK_MONITOR_CMDID:
+		return "WMI_SET_LINK_MONITOR_CMD";
 	default:
 		return "Untracked CMD";
 	}
@@ -624,6 +626,10 @@ static const char *eventid2name(u16 eventid)
 		return "WMI_RBUFCAP_CFG_EVENT";
 	case WMI_TEMP_SENSE_ALL_DONE_EVENTID:
 		return "WMI_TEMP_SENSE_ALL_DONE_EVENTID";
+	case WMI_SET_LINK_MONITOR_EVENTID:
+		return "WMI_SET_LINK_MONITOR_EVENT";
+	case WMI_LINK_MONITOR_EVENTID:
+		return "WMI_LINK_MONITOR_EVENT";
 	default:
 		return "Untracked EVENT";
 	}
@@ -1836,6 +1842,32 @@ fail:
 	wil6210_disconnect(vif, NULL, WLAN_REASON_PREV_AUTH_NOT_VALID);
 }
 
+static void
+wmi_evt_link_monitor(struct wil6210_vif *vif, int id, void *d, int len)
+{
+	struct wil6210_priv *wil = vif_to_wil(vif);
+	struct net_device *ndev = vif_to_ndev(vif);
+	struct wmi_link_monitor_event *evt = d;
+	enum nl80211_cqm_rssi_threshold_event event_type;
+
+	if (len < sizeof(*evt)) {
+		wil_err(wil, "link monitor event too short %d\n", len);
+		return;
+	}
+
+	wil_dbg_wmi(wil, "link monitor event, type %d rssi %d (stored %d)\n",
+		    evt->type, evt->rssi_level, wil->cqm_rssi_thold);
+
+	if (evt->type != WMI_LINK_MONITOR_NOTIF_RSSI_THRESHOLD_EVT)
+		/* ignore */
+		return;
+
+	event_type = (evt->rssi_level > wil->cqm_rssi_thold ?
+		      NL80211_CQM_RSSI_THRESHOLD_EVENT_HIGH :
+		      NL80211_CQM_RSSI_THRESHOLD_EVENT_LOW);
+	cfg80211_cqm_rssi_notify(ndev, event_type, evt->rssi_level, GFP_KERNEL);
+}
+
 /**
  * Some events are ignored for purpose; and need not be interpreted as
  * "unhandled events"
@@ -1869,6 +1901,7 @@ static const struct {
 	{WMI_LINK_STATS_EVENTID,		wmi_evt_link_stats},
 	{WMI_FT_AUTH_STATUS_EVENTID,		wmi_evt_auth_status},
 	{WMI_FT_REASSOC_STATUS_EVENTID,		wmi_evt_reassoc_status},
+	{WMI_LINK_MONITOR_EVENTID,		wmi_evt_link_monitor},
 };
 
 /*
@@ -3981,3 +4014,46 @@ int wmi_link_stats_cfg(struct wil6210_vif *vif, u32 type, u8 cid, u32 interval)
 
 	return 0;
 }
+
+int wmi_set_cqm_rssi_config(struct wil6210_priv *wil,
+			    s32 rssi_thold, u32 rssi_hyst)
+{
+	struct net_device *ndev = wil->main_ndev;
+	struct wil6210_vif *vif = ndev_to_vif(ndev);
+	int rc;
+	struct {
+		struct wmi_set_link_monitor_cmd cmd;
+		s8 rssi_thold;
+	} __packed cmd = {
+		.cmd = {
+			.rssi_hyst = rssi_hyst,
+			.rssi_thresholds_list_size = 1,
+		},
+		.rssi_thold = rssi_thold,
+	};
+	struct {
+		struct wmi_cmd_hdr hdr;
+		struct wmi_set_link_monitor_event evt;
+	} __packed reply = {
+		.evt = {.status = WMI_FW_STATUS_FAILURE},
+	};
+
+	if (rssi_thold > S8_MAX || rssi_thold < S8_MIN || rssi_hyst > U8_MAX)
+		return -EINVAL;
+
+	rc = wmi_call(wil, WMI_SET_LINK_MONITOR_CMDID, vif->mid, &cmd,
+		      sizeof(cmd), WMI_SET_LINK_MONITOR_EVENTID,
+		      &reply, sizeof(reply), WIL_WMI_CALL_GENERAL_TO_MS);
+	if (rc) {
+		wil_err(wil, "WMI_SET_LINK_MONITOR_CMDID failed, rc %d\n", rc);
+		return rc;
+	}
+
+	if (reply.evt.status != WMI_FW_STATUS_SUCCESS) {
+		wil_err(wil, "WMI_SET_LINK_MONITOR_CMDID failed, status %d\n",
+			reply.evt.status);
+		return -EINVAL;
+	}
+
+	return 0;
+}
diff --git a/drivers/net/wireless/ath/wil6210/wmi.h b/drivers/net/wireless/ath/wil6210/wmi.h
index 6bd4ccee28ab..e3558136e0c4 100644
--- a/drivers/net/wireless/ath/wil6210/wmi.h
+++ b/drivers/net/wireless/ath/wil6210/wmi.h
@@ -192,6 +192,7 @@ enum wmi_command_id {
 	WMI_RCP_ADDBA_RESP_EDMA_CMDID			= 0x83B,
 	WMI_LINK_MAINTAIN_CFG_WRITE_CMDID		= 0x842,
 	WMI_LINK_MAINTAIN_CFG_READ_CMDID		= 0x843,
+	WMI_SET_LINK_MONITOR_CMDID			= 0x845,
 	WMI_SET_SECTORS_CMDID				= 0x849,
 	WMI_MAINTAIN_PAUSE_CMDID			= 0x850,
 	WMI_MAINTAIN_RESUME_CMDID			= 0x851,
@@ -1973,6 +1974,7 @@ enum wmi_event_id {
 	WMI_REPORT_STATISTICS_EVENTID			= 0x100B,
 	WMI_FT_AUTH_STATUS_EVENTID			= 0x100C,
 	WMI_FT_REASSOC_STATUS_EVENTID			= 0x100D,
+	WMI_LINK_MONITOR_EVENTID			= 0x100E,
 	WMI_RADAR_GENERAL_CONFIG_EVENTID		= 0x1100,
 	WMI_RADAR_CONFIG_SELECT_EVENTID			= 0x1101,
 	WMI_RADAR_PARAMS_CONFIG_EVENTID			= 0x1102,
@@ -2024,6 +2026,7 @@ enum wmi_event_id {
 	WMI_TX_MGMT_PACKET_EVENTID			= 0x1841,
 	WMI_LINK_MAINTAIN_CFG_WRITE_DONE_EVENTID	= 0x1842,
 	WMI_LINK_MAINTAIN_CFG_READ_DONE_EVENTID		= 0x1843,
+	WMI_SET_LINK_MONITOR_EVENTID			= 0x1845,
 	WMI_RF_XPM_READ_RESULT_EVENTID			= 0x1856,
 	WMI_RF_XPM_WRITE_RESULT_EVENTID			= 0x1857,
 	WMI_LED_CFG_DONE_EVENTID			= 0x1858,
@@ -3312,6 +3315,36 @@ struct wmi_link_maintain_cfg_read_cmd {
 	__le32 cid;
 } __packed;
 
+/* WMI_SET_LINK_MONITOR_CMDID */
+struct wmi_set_link_monitor_cmd {
+	u8 rssi_hyst;
+	u8 reserved[12];
+	u8 rssi_thresholds_list_size;
+	s8 rssi_thresholds_list[0];
+} __packed;
+
+/* wmi_link_monitor_event_type */
+enum wmi_link_monitor_event_type {
+	WMI_LINK_MONITOR_NOTIF_RSSI_THRESHOLD_EVT	= 0x00,
+	WMI_LINK_MONITOR_NOTIF_TX_ERR_EVT		= 0x01,
+	WMI_LINK_MONITOR_NOTIF_THERMAL_EVT		= 0x02,
+};
+
+/* WMI_SET_LINK_MONITOR_EVENTID */
+struct wmi_set_link_monitor_event {
+	/* wmi_fw_status */
+	u8 status;
+	u8 reserved[3];
+} __packed;
+
+/* WMI_LINK_MONITOR_EVENTID */
+struct wmi_link_monitor_event {
+	/* link_monitor_event_type */
+	u8 type;
+	s8 rssi_level;
+	u8 reserved[2];
+} __packed;
+
 /* WMI_LINK_MAINTAIN_CFG_WRITE_DONE_EVENTID */
 struct wmi_link_maintain_cfg_write_done_event {
 	/* requested connection ID */
-- 
cgit v1.2.3-59-g8ed1b


From 77ea8b455ca1d93221bad979a59271233ecb757e Mon Sep 17 00:00:00 2001
From: John Crispin <john@phrozen.org>
Date: Thu, 5 Dec 2019 07:26:45 +0100
Subject: ath11k: explicitly cast wmi commands to their correct struct type

Three of the WMI command handlers were not casting to the right data type.
Lets make the code consistent with the other handlers.

Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/wmi.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/wmi.c b/drivers/net/wireless/ath/ath11k/wmi.c
index 0cc58d55b308..a9b301ceb24b 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.c
+++ b/drivers/net/wireless/ath/ath11k/wmi.c
@@ -2515,7 +2515,7 @@ ath11k_wmi_send_twt_enable_cmd(struct ath11k *ar, u32 pdev_id)
 	if (!skb)
 		return -ENOMEM;
 
-	cmd = (void *)skb->data;
+	cmd = (struct wmi_twt_enable_params_cmd *)skb->data;
 	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_TWT_ENABLE_CMD) |
 			  FIELD_PREP(WMI_TLV_LEN, len - TLV_HDR_SIZE);
 	cmd->pdev_id = pdev_id;
@@ -2566,7 +2566,7 @@ ath11k_wmi_send_twt_disable_cmd(struct ath11k *ar, u32 pdev_id)
 	if (!skb)
 		return -ENOMEM;
 
-	cmd = (void *)skb->data;
+	cmd = (struct wmi_twt_disable_params_cmd *)skb->data;
 	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG, WMI_TAG_TWT_DISABLE_CMD) |
 			  FIELD_PREP(WMI_TLV_LEN, len - TLV_HDR_SIZE);
 	cmd->pdev_id = pdev_id;
@@ -2596,7 +2596,7 @@ ath11k_wmi_send_obss_spr_cmd(struct ath11k *ar, u32 vdev_id,
 	if (!skb)
 		return -ENOMEM;
 
-	cmd = (void *)skb->data;
+	cmd = (struct wmi_obss_spatial_reuse_params_cmd *)skb->data;
 	cmd->tlv_header = FIELD_PREP(WMI_TLV_TAG,
 				     WMI_TAG_OBSS_SPATIAL_REUSE_SET_CMD) |
 			  FIELD_PREP(WMI_TLV_LEN, len - TLV_HDR_SIZE);
-- 
cgit v1.2.3-59-g8ed1b


From ca0e477931c5a725bb887c764a49eb5d69ab7a67 Mon Sep 17 00:00:00 2001
From: Kees Cook <keescook@chromium.org>
Date: Wed, 18 Dec 2019 16:58:02 -0800
Subject: ath11k: Use sizeof_field() instead of FIELD_SIZEOF()

The FIELD_SIZEOF() macro was redundant, and is being removed from the
kernel. Since commit c593642c8be0 ("treewide: Use sizeof_field() macro")
this is one of the last users of the old macro, so replace it.

Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/wmi.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath11k/wmi.h b/drivers/net/wireless/ath/ath11k/wmi.h
index 780e72c2844a..1fde15c762ad 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.h
+++ b/drivers/net/wireless/ath/ath11k/wmi.h
@@ -44,7 +44,7 @@ struct wmi_tlv {
 
 #define WMI_TLV_LEN	GENMASK(15, 0)
 #define WMI_TLV_TAG	GENMASK(31, 16)
-#define TLV_HDR_SIZE	FIELD_SIZEOF(struct wmi_tlv, header)
+#define TLV_HDR_SIZE	sizeof_field(struct wmi_tlv, header)
 
 #define WMI_CMD_HDR_CMD_ID      GENMASK(23, 0)
 #define WMI_MAX_MEM_REQS        32
-- 
cgit v1.2.3-59-g8ed1b


From c48f16b42a3bf71eeb9c6141369b001097f961e6 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Tue, 17 Dec 2019 13:39:00 +0000
Subject: net: phy: remove redundant .aneg_done initialisers

Remove initialisers that set .aneg_done to genphy_aneg_done - this is
the default for clause 22 PHYs, so the initialiser is redundant.

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/mscc.c       | 6 ------
 drivers/net/phy/phy_device.c | 1 -
 2 files changed, 7 deletions(-)

diff --git a/drivers/net/phy/mscc.c b/drivers/net/phy/mscc.c
index d5f8f351d9ef..50214c081164 100644
--- a/drivers/net/phy/mscc.c
+++ b/drivers/net/phy/mscc.c
@@ -2404,7 +2404,6 @@ static struct phy_driver vsc85xx_driver[] = {
 	.soft_reset	= &genphy_soft_reset,
 	.config_init	= &vsc85xx_config_init,
 	.config_aneg    = &vsc85xx_config_aneg,
-	.aneg_done	= &genphy_aneg_done,
 	.read_status	= &vsc85xx_read_status,
 	.ack_interrupt	= &vsc85xx_ack_interrupt,
 	.config_intr	= &vsc85xx_config_intr,
@@ -2429,7 +2428,6 @@ static struct phy_driver vsc85xx_driver[] = {
 	.soft_reset	= &genphy_soft_reset,
 	.config_init    = &vsc85xx_config_init,
 	.config_aneg    = &vsc85xx_config_aneg,
-	.aneg_done	= &genphy_aneg_done,
 	.read_status	= &vsc85xx_read_status,
 	.ack_interrupt  = &vsc85xx_ack_interrupt,
 	.config_intr    = &vsc85xx_config_intr,
@@ -2454,7 +2452,6 @@ static struct phy_driver vsc85xx_driver[] = {
 	.soft_reset	= &genphy_soft_reset,
 	.config_init	= &vsc85xx_config_init,
 	.config_aneg	= &vsc85xx_config_aneg,
-	.aneg_done	= &genphy_aneg_done,
 	.read_status	= &vsc85xx_read_status,
 	.ack_interrupt	= &vsc85xx_ack_interrupt,
 	.config_intr	= &vsc85xx_config_intr,
@@ -2479,7 +2476,6 @@ static struct phy_driver vsc85xx_driver[] = {
 	.soft_reset	= &genphy_soft_reset,
 	.config_init    = &vsc85xx_config_init,
 	.config_aneg    = &vsc85xx_config_aneg,
-	.aneg_done	= &genphy_aneg_done,
 	.read_status	= &vsc85xx_read_status,
 	.ack_interrupt  = &vsc85xx_ack_interrupt,
 	.config_intr    = &vsc85xx_config_intr,
@@ -2504,7 +2500,6 @@ static struct phy_driver vsc85xx_driver[] = {
 	.soft_reset	= &genphy_soft_reset,
 	.config_init    = &vsc8584_config_init,
 	.config_aneg    = &vsc85xx_config_aneg,
-	.aneg_done	= &genphy_aneg_done,
 	.read_status	= &vsc85xx_read_status,
 	.ack_interrupt  = &vsc85xx_ack_interrupt,
 	.config_intr    = &vsc85xx_config_intr,
@@ -2530,7 +2525,6 @@ static struct phy_driver vsc85xx_driver[] = {
 	.soft_reset	= &genphy_soft_reset,
 	.config_init    = &vsc8584_config_init,
 	.config_aneg    = &vsc85xx_config_aneg,
-	.aneg_done	= &genphy_aneg_done,
 	.read_status	= &vsc85xx_read_status,
 	.ack_interrupt  = &vsc85xx_ack_interrupt,
 	.config_intr    = &vsc85xx_config_intr,
diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c
index 0887ed2bb050..acbdced20c11 100644
--- a/drivers/net/phy/phy_device.c
+++ b/drivers/net/phy/phy_device.c
@@ -2575,7 +2575,6 @@ static struct phy_driver genphy_driver = {
 	.name		= "Generic PHY",
 	.soft_reset	= genphy_no_soft_reset,
 	.get_features	= genphy_read_abilities,
-	.aneg_done	= genphy_aneg_done,
 	.suspend	= genphy_suspend,
 	.resume		= genphy_resume,
 	.set_loopback   = genphy_loopback,
-- 
cgit v1.2.3-59-g8ed1b


From af006240c6cfea9f52fc197ac26e6cade4a8623d Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Tue, 17 Dec 2019 13:39:06 +0000
Subject: net: phy: use phy_resolve_aneg_pause()

Several drivers code their own version of this, working from the LPA
register, after setting the ethtool link partner advertisement bitmask.
Use the generic function instead.

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/lxt.c      | 5 +----
 drivers/net/phy/marvell.c  | 5 +----
 drivers/net/phy/uPD60620.c | 7 +------
 3 files changed, 3 insertions(+), 14 deletions(-)

diff --git a/drivers/net/phy/lxt.c b/drivers/net/phy/lxt.c
index 356bd6472f49..30172852e36e 100644
--- a/drivers/net/phy/lxt.c
+++ b/drivers/net/phy/lxt.c
@@ -190,10 +190,7 @@ static int lxt973a2_read_status(struct phy_device *phydev)
 				phydev->duplex = DUPLEX_FULL;
 		}
 
-		if (phydev->duplex == DUPLEX_FULL) {
-			phydev->pause = lpa & LPA_PAUSE_CAP ? 1 : 0;
-			phydev->asym_pause = lpa & LPA_PAUSE_ASYM ? 1 : 0;
-		}
+		phy_resolve_aneg_pause(phydev);
 	} else {
 		int bmcr = phy_read(phydev, MII_BMCR);
 
diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c
index b1fbd1937328..d807c3e5cc56 100644
--- a/drivers/net/phy/marvell.c
+++ b/drivers/net/phy/marvell.c
@@ -1384,10 +1384,7 @@ static int marvell_read_status_page_an(struct phy_device *phydev,
 		mii_lpa_to_linkmode_lpa_t(phydev->lp_advertising, lpa);
 		mii_stat1000_mod_linkmode_lpa_t(phydev->lp_advertising, lpagb);
 
-		if (phydev->duplex == DUPLEX_FULL) {
-			phydev->pause = lpa & LPA_PAUSE_CAP ? 1 : 0;
-			phydev->asym_pause = lpa & LPA_PAUSE_ASYM ? 1 : 0;
-		}
+		phy_resolve_aneg_pause(phydev);
 	} else {
 		/* The fiber link is only 1000M capable */
 		fiber_lpa_mod_linkmode_lpa_t(phydev->lp_advertising, lpa);
diff --git a/drivers/net/phy/uPD60620.c b/drivers/net/phy/uPD60620.c
index a32b3fd8a370..38834347a427 100644
--- a/drivers/net/phy/uPD60620.c
+++ b/drivers/net/phy/uPD60620.c
@@ -68,12 +68,7 @@ static int upd60620_read_status(struct phy_device *phydev)
 			mii_lpa_to_linkmode_lpa_t(phydev->lp_advertising,
 						  phy_state);
 
-			if (phydev->duplex == DUPLEX_FULL) {
-				if (phy_state & LPA_PAUSE_CAP)
-					phydev->pause = 1;
-				if (phy_state & LPA_PAUSE_ASYM)
-					phydev->asym_pause = 1;
-			}
+			phy_resolve_aneg_pause(phydev);
 		}
 	}
 	return 0;
-- 
cgit v1.2.3-59-g8ed1b


From 2a10ab043ac5a658225ee77852db7942de9ac4c5 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Tue, 17 Dec 2019 13:39:11 +0000
Subject: net: phy: add genphy_check_and_restart_aneg()

Add a helper for restarting autonegotiation(), similar to the clause 45
variant.  Use it in __genphy_config_aneg()

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/phy_device.c | 48 ++++++++++++++++++++++++++++----------------
 include/linux/phy.h          |  1 +
 2 files changed, 32 insertions(+), 17 deletions(-)

diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c
index acbdced20c11..c61f2295e21e 100644
--- a/drivers/net/phy/phy_device.c
+++ b/drivers/net/phy/phy_device.c
@@ -1770,6 +1770,36 @@ int genphy_restart_aneg(struct phy_device *phydev)
 }
 EXPORT_SYMBOL(genphy_restart_aneg);
 
+/**
+ * genphy_check_and_restart_aneg - Enable and restart auto-negotiation
+ * @phydev: target phy_device struct
+ * @restart: whether aneg restart is requested
+ *
+ * Check, and restart auto-negotiation if needed.
+ */
+int genphy_check_and_restart_aneg(struct phy_device *phydev, bool restart)
+{
+	int ret = 0;
+
+	if (!restart) {
+		/* Advertisement hasn't changed, but maybe aneg was never on to
+		 * begin with?  Or maybe phy was isolated?
+		 */
+		ret = phy_read(phydev, MII_BMCR);
+		if (ret < 0)
+			return ret;
+
+		if (!(ret & BMCR_ANENABLE) || (ret & BMCR_ISOLATE))
+			restart = true;
+	}
+
+	if (restart)
+		ret = genphy_restart_aneg(phydev);
+
+	return ret;
+}
+EXPORT_SYMBOL(genphy_check_and_restart_aneg);
+
 /**
  * __genphy_config_aneg - restart auto-negotiation or write BMCR
  * @phydev: target phy_device struct
@@ -1795,23 +1825,7 @@ int __genphy_config_aneg(struct phy_device *phydev, bool changed)
 	else if (err)
 		changed = true;
 
-	if (!changed) {
-		/* Advertisement hasn't changed, but maybe aneg was never on to
-		 * begin with?  Or maybe phy was isolated?
-		 */
-		int ctl = phy_read(phydev, MII_BMCR);
-
-		if (ctl < 0)
-			return ctl;
-
-		if (!(ctl & BMCR_ANENABLE) || (ctl & BMCR_ISOLATE))
-			changed = true; /* do restart aneg */
-	}
-
-	/* Only restart aneg if we are advertising something different
-	 * than we were before.
-	 */
-	return changed ? genphy_restart_aneg(phydev) : 0;
+	return genphy_check_and_restart_aneg(phydev, changed);
 }
 EXPORT_SYMBOL(__genphy_config_aneg);
 
diff --git a/include/linux/phy.h b/include/linux/phy.h
index 5032d453ac66..1c4f97d2631d 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -1094,6 +1094,7 @@ void phy_attached_info(struct phy_device *phydev);
 int genphy_read_abilities(struct phy_device *phydev);
 int genphy_setup_forced(struct phy_device *phydev);
 int genphy_restart_aneg(struct phy_device *phydev);
+int genphy_check_and_restart_aneg(struct phy_device *phydev, bool restart);
 int genphy_config_eee_advert(struct phy_device *phydev);
 int __genphy_config_aneg(struct phy_device *phydev, bool changed);
 int genphy_aneg_done(struct phy_device *phydev);
-- 
cgit v1.2.3-59-g8ed1b


From 0efc286a923874f0c243e5766cce54e9429ed949 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Tue, 17 Dec 2019 13:39:16 +0000
Subject: net: phy: provide and use genphy_read_status_fixed()

There are two drivers and generic code which contain exactly the same
code to read the status of a PHY operating without autonegotiation
enabled. Rather than duplicate this code, provide a helper to read
this information.

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/lxt.c        | 19 +++--------------
 drivers/net/phy/marvell.c    | 19 ++++-------------
 drivers/net/phy/phy_device.c | 49 +++++++++++++++++++++++++++++---------------
 include/linux/phy.h          |  1 +
 4 files changed, 41 insertions(+), 47 deletions(-)

diff --git a/drivers/net/phy/lxt.c b/drivers/net/phy/lxt.c
index 30172852e36e..fec58ad69e02 100644
--- a/drivers/net/phy/lxt.c
+++ b/drivers/net/phy/lxt.c
@@ -192,22 +192,9 @@ static int lxt973a2_read_status(struct phy_device *phydev)
 
 		phy_resolve_aneg_pause(phydev);
 	} else {
-		int bmcr = phy_read(phydev, MII_BMCR);
-
-		if (bmcr < 0)
-			return bmcr;
-
-		if (bmcr & BMCR_FULLDPLX)
-			phydev->duplex = DUPLEX_FULL;
-		else
-			phydev->duplex = DUPLEX_HALF;
-
-		if (bmcr & BMCR_SPEED1000)
-			phydev->speed = SPEED_1000;
-		else if (bmcr & BMCR_SPEED100)
-			phydev->speed = SPEED_100;
-		else
-			phydev->speed = SPEED_10;
+		err = genphy_read_status_fixed(phydev);
+		if (err < 0)
+			return err;
 
 		phydev->pause = phydev->asym_pause = 0;
 		linkmode_zero(phydev->lp_advertising);
diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c
index d807c3e5cc56..5e49639dc8dd 100644
--- a/drivers/net/phy/marvell.c
+++ b/drivers/net/phy/marvell.c
@@ -1407,22 +1407,11 @@ static int marvell_read_status_page_an(struct phy_device *phydev,
 
 static int marvell_read_status_page_fixed(struct phy_device *phydev)
 {
-	int bmcr = phy_read(phydev, MII_BMCR);
-
-	if (bmcr < 0)
-		return bmcr;
-
-	if (bmcr & BMCR_FULLDPLX)
-		phydev->duplex = DUPLEX_FULL;
-	else
-		phydev->duplex = DUPLEX_HALF;
+	int err;
 
-	if (bmcr & BMCR_SPEED1000)
-		phydev->speed = SPEED_1000;
-	else if (bmcr & BMCR_SPEED100)
-		phydev->speed = SPEED_100;
-	else
-		phydev->speed = SPEED_10;
+	err = genphy_read_status_fixed(phydev);
+	if (err < 0)
+		return err;
 
 	phydev->pause = 0;
 	phydev->asym_pause = 0;
diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c
index c61f2295e21e..ec19efe7ea99 100644
--- a/drivers/net/phy/phy_device.c
+++ b/drivers/net/phy/phy_device.c
@@ -1992,6 +1992,36 @@ int genphy_read_lpa(struct phy_device *phydev)
 }
 EXPORT_SYMBOL(genphy_read_lpa);
 
+/**
+ * genphy_read_status_fixed - read the link parameters for !aneg mode
+ * @phydev: target phy_device struct
+ *
+ * Read the current duplex and speed state for a PHY operating with
+ * autonegotiation disabled.
+ */
+int genphy_read_status_fixed(struct phy_device *phydev)
+{
+	int bmcr = phy_read(phydev, MII_BMCR);
+
+	if (bmcr < 0)
+		return bmcr;
+
+	if (bmcr & BMCR_FULLDPLX)
+		phydev->duplex = DUPLEX_FULL;
+	else
+		phydev->duplex = DUPLEX_HALF;
+
+	if (bmcr & BMCR_SPEED1000)
+		phydev->speed = SPEED_1000;
+	else if (bmcr & BMCR_SPEED100)
+		phydev->speed = SPEED_100;
+	else
+		phydev->speed = SPEED_10;
+
+	return 0;
+}
+EXPORT_SYMBOL(genphy_read_status_fixed);
+
 /**
  * genphy_read_status - check the link status and update current link state
  * @phydev: target phy_device struct
@@ -2026,22 +2056,9 @@ int genphy_read_status(struct phy_device *phydev)
 	if (phydev->autoneg == AUTONEG_ENABLE && phydev->autoneg_complete) {
 		phy_resolve_aneg_linkmode(phydev);
 	} else if (phydev->autoneg == AUTONEG_DISABLE) {
-		int bmcr = phy_read(phydev, MII_BMCR);
-
-		if (bmcr < 0)
-			return bmcr;
-
-		if (bmcr & BMCR_FULLDPLX)
-			phydev->duplex = DUPLEX_FULL;
-		else
-			phydev->duplex = DUPLEX_HALF;
-
-		if (bmcr & BMCR_SPEED1000)
-			phydev->speed = SPEED_1000;
-		else if (bmcr & BMCR_SPEED100)
-			phydev->speed = SPEED_100;
-		else
-			phydev->speed = SPEED_10;
+		err = genphy_read_status_fixed(phydev);
+		if (err < 0)
+			return err;
 	}
 
 	return 0;
diff --git a/include/linux/phy.h b/include/linux/phy.h
index 1c4f97d2631d..b2105e0d72d3 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -1100,6 +1100,7 @@ int __genphy_config_aneg(struct phy_device *phydev, bool changed);
 int genphy_aneg_done(struct phy_device *phydev);
 int genphy_update_link(struct phy_device *phydev);
 int genphy_read_lpa(struct phy_device *phydev);
+int genphy_read_status_fixed(struct phy_device *phydev);
 int genphy_read_status(struct phy_device *phydev);
 int genphy_suspend(struct phy_device *phydev);
 int genphy_resume(struct phy_device *phydev);
-- 
cgit v1.2.3-59-g8ed1b


From fcf1f59afc67dc8c214a1b085527ed1fc513c3a2 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Tue, 17 Dec 2019 13:39:21 +0000
Subject: net: phy: marvell: rearrange to use genphy_read_lpa()

Rearrange the Marvell PHY driver to use genphy_read_lpa() rather than
open-coding this functionality.

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/marvell.c | 66 +++++++++++++++++++++++------------------------
 1 file changed, 32 insertions(+), 34 deletions(-)

diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c
index 5e49639dc8dd..ac9c21b4e6c9 100644
--- a/drivers/net/phy/marvell.c
+++ b/drivers/net/phy/marvell.c
@@ -1343,52 +1343,30 @@ static int marvell_read_status_page_an(struct phy_device *phydev,
 {
 	int status;
 	int lpa;
-	int lpagb;
+	int err;
 
 	status = phy_read(phydev, MII_M1011_PHY_STATUS);
 	if (status < 0)
 		return status;
 
-	lpa = phy_read(phydev, MII_LPA);
-	if (lpa < 0)
-		return lpa;
-
-	lpagb = phy_read(phydev, MII_STAT1000);
-	if (lpagb < 0)
-		return lpagb;
-
-	if (status & MII_M1011_PHY_STATUS_FULLDUPLEX)
-		phydev->duplex = DUPLEX_FULL;
-	else
-		phydev->duplex = DUPLEX_HALF;
-
-	status = status & MII_M1011_PHY_STATUS_SPD_MASK;
-	phydev->pause = 0;
-	phydev->asym_pause = 0;
-
-	switch (status) {
-	case MII_M1011_PHY_STATUS_1000:
-		phydev->speed = SPEED_1000;
-		break;
-
-	case MII_M1011_PHY_STATUS_100:
-		phydev->speed = SPEED_100;
-		break;
-
-	default:
-		phydev->speed = SPEED_10;
-		break;
-	}
-
 	if (!fiber) {
-		mii_lpa_to_linkmode_lpa_t(phydev->lp_advertising, lpa);
-		mii_stat1000_mod_linkmode_lpa_t(phydev->lp_advertising, lpagb);
+		err = genphy_read_lpa(phydev);
+		if (err < 0)
+			return err;
 
+		phydev->pause = 0;
+		phydev->asym_pause = 0;
 		phy_resolve_aneg_pause(phydev);
 	} else {
+		lpa = phy_read(phydev, MII_LPA);
+		if (lpa < 0)
+			return lpa;
+
 		/* The fiber link is only 1000M capable */
 		fiber_lpa_mod_linkmode_lpa_t(phydev->lp_advertising, lpa);
 
+		phydev->pause = 0;
+		phydev->asym_pause = 0;
 		if (phydev->duplex == DUPLEX_FULL) {
 			if (!(lpa & LPA_PAUSE_FIBER)) {
 				phydev->pause = 0;
@@ -1402,6 +1380,26 @@ static int marvell_read_status_page_an(struct phy_device *phydev,
 			}
 		}
 	}
+
+	if (status & MII_M1011_PHY_STATUS_FULLDUPLEX)
+		phydev->duplex = DUPLEX_FULL;
+	else
+		phydev->duplex = DUPLEX_HALF;
+
+	switch (status & MII_M1011_PHY_STATUS_SPD_MASK) {
+	case MII_M1011_PHY_STATUS_1000:
+		phydev->speed = SPEED_1000;
+		break;
+
+	case MII_M1011_PHY_STATUS_100:
+		phydev->speed = SPEED_100;
+		break;
+
+	default:
+		phydev->speed = SPEED_10;
+		break;
+	}
+
 	return 0;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 98f92831c561e43e8d4dafa7165e996af4dc8ac3 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Tue, 17 Dec 2019 13:39:26 +0000
Subject: net: phy: marvell: initialise link partner state earlier

Move the initialisation of the link partner state earlier, inside
marvell_read_status_page(), so we don't have the same initialisation
scattered amongst the other files.  This is in a similar place to
the genphy implementation, so would result in the same behaviour if
a PHY read error occurs.

This allows us to get rid of marvell_read_status_page_fixed(), which
became a pointless wrapper around genphy_read_status_fixed().

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/marvell.c | 25 +++++--------------------
 1 file changed, 5 insertions(+), 20 deletions(-)

diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c
index ac9c21b4e6c9..b50c8a4710b4 100644
--- a/drivers/net/phy/marvell.c
+++ b/drivers/net/phy/marvell.c
@@ -1354,8 +1354,6 @@ static int marvell_read_status_page_an(struct phy_device *phydev,
 		if (err < 0)
 			return err;
 
-		phydev->pause = 0;
-		phydev->asym_pause = 0;
 		phy_resolve_aneg_pause(phydev);
 	} else {
 		lpa = phy_read(phydev, MII_LPA);
@@ -1365,8 +1363,6 @@ static int marvell_read_status_page_an(struct phy_device *phydev,
 		/* The fiber link is only 1000M capable */
 		fiber_lpa_mod_linkmode_lpa_t(phydev->lp_advertising, lpa);
 
-		phydev->pause = 0;
-		phydev->asym_pause = 0;
 		if (phydev->duplex == DUPLEX_FULL) {
 			if (!(lpa & LPA_PAUSE_FIBER)) {
 				phydev->pause = 0;
@@ -1403,21 +1399,6 @@ static int marvell_read_status_page_an(struct phy_device *phydev,
 	return 0;
 }
 
-static int marvell_read_status_page_fixed(struct phy_device *phydev)
-{
-	int err;
-
-	err = genphy_read_status_fixed(phydev);
-	if (err < 0)
-		return err;
-
-	phydev->pause = 0;
-	phydev->asym_pause = 0;
-	linkmode_zero(phydev->lp_advertising);
-
-	return 0;
-}
-
 /* marvell_read_status_page
  *
  * Description:
@@ -1443,10 +1424,14 @@ static int marvell_read_status_page(struct phy_device *phydev, int page)
 	if (err)
 		return err;
 
+	linkmode_zero(phydev->lp_advertising);
+	phydev->pause = 0;
+	phydev->asym_pause = 0;
+
 	if (phydev->autoneg == AUTONEG_ENABLE)
 		err = marvell_read_status_page_an(phydev, fiber);
 	else
-		err = marvell_read_status_page_fixed(phydev);
+		err = genphy_read_status_fixed(phydev);
 
 	return err;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 760fa78f3509d3d414804c4017de3d0dba995081 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Tue, 17 Dec 2019 13:39:31 +0000
Subject: net: phy: marvell: use positive logic for link state

Rather than using negative logic:

	if (there is no link)
		set link = 0
	else
		set link = 1

use the more natural positive logic:

	if (there is link)
		set link = 1
	else
		set link = 0

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/marvell.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c
index b50c8a4710b4..5900281f4c2c 100644
--- a/drivers/net/phy/marvell.c
+++ b/drivers/net/phy/marvell.c
@@ -174,7 +174,6 @@
 #define ADVERTISE_PAUSE_FIBER		0x180
 #define ADVERTISE_PAUSE_ASYM_FIBER	0x100
 
-#define REGISTER_LINK_STATUS	0x400
 #define NB_FIBER_STATS	1
 
 MODULE_DESCRIPTION("Marvell PHY driver");
@@ -1327,10 +1326,10 @@ static int marvell_update_link(struct phy_device *phydev, int fiber)
 		if (status < 0)
 			return status;
 
-		if ((status & REGISTER_LINK_STATUS) == 0)
-			phydev->link = 0;
-		else
+		if (status & MII_M1011_PHY_STATUS_LINK)
 			phydev->link = 1;
+		else
+			phydev->link = 0;
 	} else {
 		return genphy_update_link(phydev);
 	}
-- 
cgit v1.2.3-59-g8ed1b


From d2004e27eb352a5961cae7cc50563dbdc5386ba7 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Tue, 17 Dec 2019 13:39:36 +0000
Subject: net: phy: marvell: consolidate phy status reading

marvell_read_status_page_an() always reads the PHY status register, but
marvell_update_link() has already done this.  Rather than wastefully
reading the register twice in quick succession, read it once in
marvell_read_status_page() and use the result for both.

This makes marvell_update_link() rather pointless, so move it into
marvell_read_status_page().

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/marvell.c | 60 ++++++++++++++---------------------------------
 1 file changed, 17 insertions(+), 43 deletions(-)

diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c
index 5900281f4c2c..9c26e8ebd143 100644
--- a/drivers/net/phy/marvell.c
+++ b/drivers/net/phy/marvell.c
@@ -1307,47 +1307,12 @@ static void fiber_lpa_mod_linkmode_lpa_t(unsigned long *advertising, u32 lpa)
 			 advertising, lpa & LPA_FIBER_1000FULL);
 }
 
-/**
- * marvell_update_link - update link status in real time in @phydev
- * @phydev: target phy_device struct
- *
- * Description: Update the value in phydev->link to reflect the
- *   current link value.
- */
-static int marvell_update_link(struct phy_device *phydev, int fiber)
-{
-	int status;
-
-	/* Use the generic register for copper link, or specific
-	 * register for fiber case
-	 */
-	if (fiber) {
-		status = phy_read(phydev, MII_M1011_PHY_STATUS);
-		if (status < 0)
-			return status;
-
-		if (status & MII_M1011_PHY_STATUS_LINK)
-			phydev->link = 1;
-		else
-			phydev->link = 0;
-	} else {
-		return genphy_update_link(phydev);
-	}
-
-	return 0;
-}
-
 static int marvell_read_status_page_an(struct phy_device *phydev,
-				       int fiber)
+				       int fiber, int status)
 {
-	int status;
 	int lpa;
 	int err;
 
-	status = phy_read(phydev, MII_M1011_PHY_STATUS);
-	if (status < 0)
-		return status;
-
 	if (!fiber) {
 		err = genphy_read_lpa(phydev);
 		if (err < 0)
@@ -1408,27 +1373,36 @@ static int marvell_read_status_page_an(struct phy_device *phydev,
  */
 static int marvell_read_status_page(struct phy_device *phydev, int page)
 {
+	int status;
 	int fiber;
 	int err;
 
-	/* Detect and update the link, but return if there
-	 * was an error
+	status = phy_read(phydev, MII_M1011_PHY_STATUS);
+	if (status < 0)
+		return status;
+
+	/* Use the generic register for copper link status,
+	 * and the PHY status register for fiber link status.
 	 */
+	if (page == MII_MARVELL_FIBER_PAGE) {
+		phydev->link = !!(status & MII_M1011_PHY_STATUS_LINK);
+	} else {
+		err = genphy_update_link(phydev);
+		if (err)
+			return err;
+	}
+
 	if (page == MII_MARVELL_FIBER_PAGE)
 		fiber = 1;
 	else
 		fiber = 0;
 
-	err = marvell_update_link(phydev, fiber);
-	if (err)
-		return err;
-
 	linkmode_zero(phydev->lp_advertising);
 	phydev->pause = 0;
 	phydev->asym_pause = 0;
 
 	if (phydev->autoneg == AUTONEG_ENABLE)
-		err = marvell_read_status_page_an(phydev, fiber);
+		err = marvell_read_status_page_an(phydev, fiber, status);
 	else
 		err = genphy_read_status_fixed(phydev);
 
-- 
cgit v1.2.3-59-g8ed1b


From 20ecf424d0dbb820a337e994c8feec18c1fa77c9 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Tue, 17 Dec 2019 13:39:42 +0000
Subject: net: phy: marvell: use existing clause 37 definitions

Use existing clause 37 advertising/link partner definitions rather than
private ones for the advertisement registers.

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/marvell.c | 26 ++++++++------------------
 1 file changed, 8 insertions(+), 18 deletions(-)

diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c
index 9c26e8ebd143..c6c7cde84c1c 100644
--- a/drivers/net/phy/marvell.c
+++ b/drivers/net/phy/marvell.c
@@ -162,18 +162,9 @@
 #define MII_88E1510_GEN_CTRL_REG_1_MODE_SGMII	0x1	/* SGMII to copper */
 #define MII_88E1510_GEN_CTRL_REG_1_RESET	0x8000	/* Soft reset */
 
-#define LPA_FIBER_1000HALF	0x40
-#define LPA_FIBER_1000FULL	0x20
-
 #define LPA_PAUSE_FIBER		0x180
 #define LPA_PAUSE_ASYM_FIBER	0x100
 
-#define ADVERTISE_FIBER_1000HALF	0x40
-#define ADVERTISE_FIBER_1000FULL	0x20
-
-#define ADVERTISE_PAUSE_FIBER		0x180
-#define ADVERTISE_PAUSE_ASYM_FIBER	0x100
-
 #define NB_FIBER_STATS	1
 
 MODULE_DESCRIPTION("Marvell PHY driver");
@@ -496,16 +487,15 @@ static inline u32 linkmode_adv_to_fiber_adv_t(unsigned long *advertise)
 	u32 result = 0;
 
 	if (linkmode_test_bit(ETHTOOL_LINK_MODE_1000baseT_Half_BIT, advertise))
-		result |= ADVERTISE_FIBER_1000HALF;
+		result |= ADVERTISE_1000XHALF;
 	if (linkmode_test_bit(ETHTOOL_LINK_MODE_1000baseT_Full_BIT, advertise))
-		result |= ADVERTISE_FIBER_1000FULL;
+		result |= ADVERTISE_1000XFULL;
 
 	if (linkmode_test_bit(ETHTOOL_LINK_MODE_Asym_Pause_BIT, advertise) &&
 	    linkmode_test_bit(ETHTOOL_LINK_MODE_Pause_BIT, advertise))
-		result |= LPA_PAUSE_ASYM_FIBER;
+		result |= ADVERTISE_1000XPSE_ASYM;
 	else if (linkmode_test_bit(ETHTOOL_LINK_MODE_Pause_BIT, advertise))
-		result |= (ADVERTISE_PAUSE_FIBER
-			   & (~ADVERTISE_PAUSE_ASYM_FIBER));
+		result |= ADVERTISE_1000XPAUSE;
 
 	return result;
 }
@@ -538,8 +528,8 @@ static int marvell_config_aneg_fiber(struct phy_device *phydev)
 		return adv;
 
 	oldadv = adv;
-	adv &= ~(ADVERTISE_FIBER_1000HALF | ADVERTISE_FIBER_1000FULL
-		| LPA_PAUSE_FIBER);
+	adv &= ~(ADVERTISE_1000XHALF | ADVERTISE_1000XFULL |
+		 ADVERTISE_1000XPAUSE | ADVERTISE_1000XPSE_ASYM);
 	adv |= linkmode_adv_to_fiber_adv_t(phydev->advertising);
 
 	if (adv != oldadv) {
@@ -1301,10 +1291,10 @@ static int m88e6390_config_aneg(struct phy_device *phydev)
 static void fiber_lpa_mod_linkmode_lpa_t(unsigned long *advertising, u32 lpa)
 {
 	linkmode_mod_bit(ETHTOOL_LINK_MODE_1000baseT_Half_BIT,
-			 advertising, lpa & LPA_FIBER_1000HALF);
+			 advertising, lpa & LPA_1000XHALF);
 
 	linkmode_mod_bit(ETHTOOL_LINK_MODE_1000baseT_Full_BIT,
-			 advertising, lpa & LPA_FIBER_1000FULL);
+			 advertising, lpa & LPA_1000XFULL);
 }
 
 static int marvell_read_status_page_an(struct phy_device *phydev,
-- 
cgit v1.2.3-59-g8ed1b


From 9f4bae704f93c9b83196781de3df8a4bf1855a4b Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Tue, 17 Dec 2019 13:39:47 +0000
Subject: net: phy: marvell: use phy_modify_changed()

Use phy_modify_changed() to change the fiber advertisement register
rather than open coding this functionality.

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/marvell.c | 26 ++++++++++----------------
 1 file changed, 10 insertions(+), 16 deletions(-)

diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c
index c6c7cde84c1c..7da64208365b 100644
--- a/drivers/net/phy/marvell.c
+++ b/drivers/net/phy/marvell.c
@@ -513,7 +513,7 @@ static int marvell_config_aneg_fiber(struct phy_device *phydev)
 {
 	int changed = 0;
 	int err;
-	int adv, oldadv;
+	u16 adv;
 
 	if (phydev->autoneg != AUTONEG_ENABLE)
 		return genphy_setup_forced(phydev);
@@ -522,23 +522,17 @@ static int marvell_config_aneg_fiber(struct phy_device *phydev)
 	linkmode_and(phydev->advertising, phydev->advertising,
 		     phydev->supported);
 
-	/* Setup fiber advertisement */
-	adv = phy_read(phydev, MII_ADVERTISE);
-	if (adv < 0)
-		return adv;
-
-	oldadv = adv;
-	adv &= ~(ADVERTISE_1000XHALF | ADVERTISE_1000XFULL |
-		 ADVERTISE_1000XPAUSE | ADVERTISE_1000XPSE_ASYM);
-	adv |= linkmode_adv_to_fiber_adv_t(phydev->advertising);
-
-	if (adv != oldadv) {
-		err = phy_write(phydev, MII_ADVERTISE, adv);
-		if (err < 0)
-			return err;
+	adv = linkmode_adv_to_fiber_adv_t(phydev->advertising);
 
+	/* Setup fiber advertisement */
+	err = phy_modify_changed(phydev, MII_ADVERTISE,
+				 ADVERTISE_1000XHALF | ADVERTISE_1000XFULL |
+				 ADVERTISE_1000XPAUSE | ADVERTISE_1000XPSE_ASYM,
+				 adv);
+	if (err < 0)
+		return err;
+	if (err > 0)
 		changed = 1;
-	}
 
 	if (changed == 0) {
 		/* Advertisement hasn't changed, but maybe aneg was never on to
-- 
cgit v1.2.3-59-g8ed1b


From b5abac2d2d17ce1b30c267dd72fb91c8bc21bee1 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Tue, 17 Dec 2019 13:39:52 +0000
Subject: net: phy: marvell: use genphy_check_and_restart_aneg()

Use the helper to check and restart autonegotiation for the marvell
fiber page negotiation setting.

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/marvell.c | 21 +--------------------
 1 file changed, 1 insertion(+), 20 deletions(-)

diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c
index 7da64208365b..28e33ece4ce1 100644
--- a/drivers/net/phy/marvell.c
+++ b/drivers/net/phy/marvell.c
@@ -534,26 +534,7 @@ static int marvell_config_aneg_fiber(struct phy_device *phydev)
 	if (err > 0)
 		changed = 1;
 
-	if (changed == 0) {
-		/* Advertisement hasn't changed, but maybe aneg was never on to
-		 * begin with?	Or maybe phy was isolated?
-		 */
-		int ctl = phy_read(phydev, MII_BMCR);
-
-		if (ctl < 0)
-			return ctl;
-
-		if (!(ctl & BMCR_ANENABLE) || (ctl & BMCR_ISOLATE))
-			changed = 1; /* do restart aneg */
-	}
-
-	/* Only restart aneg if we are advertising something different
-	 * than we were before.
-	 */
-	if (changed > 0)
-		changed = genphy_restart_aneg(phydev);
-
-	return changed;
+	return genphy_check_and_restart_aneg(phydev, changed);
 }
 
 static int m88e1510_config_aneg(struct phy_device *phydev)
-- 
cgit v1.2.3-59-g8ed1b


From 5bf2fc1f9c88397b125d5ec5f65b1ed9300ba59d Mon Sep 17 00:00:00 2001
From: Aditya Pakki <pakki001@umn.edu>
Date: Thu, 19 Dec 2019 11:57:35 -0600
Subject: bpf: Remove unnecessary assertion on fp_old

The two callers of bpf_prog_realloc - bpf_patch_insn_single and
bpf_migrate_filter dereference the struct fp_old, before passing
it to the function. Thus assertion to check fp_old is unnecessary
and can be removed.

Signed-off-by: Aditya Pakki <pakki001@umn.edu>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191219175735.19231-1-pakki001@umn.edu
---
 kernel/bpf/core.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c
index 2ff01a716128..7622dfc36705 100644
--- a/kernel/bpf/core.c
+++ b/kernel/bpf/core.c
@@ -222,8 +222,6 @@ struct bpf_prog *bpf_prog_realloc(struct bpf_prog *fp_old, unsigned int size,
 	u32 pages, delta;
 	int ret;
 
-	BUG_ON(fp_old == NULL);
-
 	size = round_up(size, PAGE_SIZE);
 	pages = size / PAGE_SIZE;
 	if (pages <= fp_old->pages)
-- 
cgit v1.2.3-59-g8ed1b


From 0dea4d039a7fdd97790a71da37c1ce67a5b52769 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Tue, 17 Dec 2019 13:50:29 +0000
Subject: net: sfp: report error on failure to read sfp soft status

Report a rate-limited error if we fail to read the SFP soft status,
and preserve the current status in that case. This avoids I2C bus
errors from triggering a link flap.

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/sfp.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/drivers/net/phy/sfp.c b/drivers/net/phy/sfp.c
index e54aef921038..73c2969f11a4 100644
--- a/drivers/net/phy/sfp.c
+++ b/drivers/net/phy/sfp.c
@@ -442,13 +442,20 @@ static unsigned int sfp_soft_get_state(struct sfp *sfp)
 {
 	unsigned int state = 0;
 	u8 status;
+	int ret;
 
-	if (sfp_read(sfp, true, SFP_STATUS, &status, sizeof(status)) ==
-		     sizeof(status)) {
+	ret = sfp_read(sfp, true, SFP_STATUS, &status, sizeof(status));
+	if (ret == sizeof(status)) {
 		if (status & SFP_STATUS_RX_LOS)
 			state |= SFP_F_LOS;
 		if (status & SFP_STATUS_TX_FAULT)
 			state |= SFP_F_TX_FAULT;
+	} else {
+		dev_err_ratelimited(sfp->dev,
+				    "failed to read SFP soft status: %d\n",
+				    ret);
+		/* Preserve the current state */
+		state = sfp->state;
 	}
 
 	return state & sfp->state_soft_mask;
-- 
cgit v1.2.3-59-g8ed1b


From 6791c102602b428b729890f6a38a319dfe225aba Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Tue, 17 Dec 2019 15:47:36 +0000
Subject: net: mvpp2: cycle comphy to power it down

Presently, at boot time, the comphys are enabled. For firmware
compatibility reasons, the comphy driver does not power down the
comphys at boot. Consequently, the ethernet comphys are left active
until the network interfaces are brought through an up/down cycle.

If the port is never used, the port wastes power needlessly. Arrange
for the ethernet comphys to be cycled by the mvpp2 driver as if the
interface went through an up/down cycle during driver probe, thereby
powering them down.

This saves:
  270mW per 10G SFP+ port on the Macchiatobin Single Shot (eth0/eth1)
  370mW per 10G PHY port on the Macchiatobin Double Shot (eth0/eth1)
  160mW on the SFP port on either Macchiatobin flavour (eth3)

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Acked-by: Antoine Tenart <antoine.tenart@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
index 3d9eb5747d9e..28eba99888c3 100644
--- a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
+++ b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
@@ -5425,6 +5425,16 @@ static int mvpp2_port_probe(struct platform_device *pdev,
 		port->phylink = NULL;
 	}
 
+	/* Cycle the comphy to power it down, saving 270mW per port -
+	 * don't worry about an error powering it up. When the comphy
+	 * driver does this, we can remove this code.
+	 */
+	if (port->comphy) {
+		err = mvpp22_comphy_init(port);
+		if (err == 0)
+			phy_power_off(port->comphy);
+	}
+
 	err = register_netdev(dev);
 	if (err < 0) {
 		dev_err(&pdev->dev, "failed to register netdev\n");
-- 
cgit v1.2.3-59-g8ed1b


From a886ca6fcfffd337482352f383c1002c72786b17 Mon Sep 17 00:00:00 2001
From: Aditya Pakki <pakki001@umn.edu>
Date: Tue, 17 Dec 2019 15:06:19 -0600
Subject: hdlcdrv: replace unnecessary assertion in hdlcdrv_register

In hdlcdrv_register, failure to register the driver causes a crash.
The three callers of hdlcdrv_register all pass valid pointers and
do not fail. The patch eliminates the unnecessary BUG_ON assertion.

Signed-off-by: Aditya Pakki <pakki001@umn.edu>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/hamradio/hdlcdrv.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/hamradio/hdlcdrv.c b/drivers/net/hamradio/hdlcdrv.c
index df495b5595f5..e7413a643929 100644
--- a/drivers/net/hamradio/hdlcdrv.c
+++ b/drivers/net/hamradio/hdlcdrv.c
@@ -687,8 +687,6 @@ struct net_device *hdlcdrv_register(const struct hdlcdrv_ops *ops,
 	struct hdlcdrv_state *s;
 	int err;
 
-	BUG_ON(ops == NULL);
-
 	if (privsize < sizeof(struct hdlcdrv_state))
 		privsize = sizeof(struct hdlcdrv_state);
 
-- 
cgit v1.2.3-59-g8ed1b


From 31c491e56ad1ad769e09f6971d267834bfc9bc39 Mon Sep 17 00:00:00 2001
From: John Hurley <john.hurley@netronome.com>
Date: Tue, 17 Dec 2019 21:57:16 +0000
Subject: nfp: flower: pass flow rule pointer directly to match functions

In kernel 5.1, the flow offload API was introduced along with a helper
function to extract the flow_rule from the TC offload struct. Each of the
match helper functions are passed the offload struct and extract the flow
rule to a local variable.

Simplify the code while also removing the extra compat and local variable
calls by extracting the rule once in the main match handler, and passing
a reference to the rule direct to each helper.

This patch does not change driver functionality.

Signed-off-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/netronome/nfp/flower/match.c | 76 ++++++++---------------
 1 file changed, 27 insertions(+), 49 deletions(-)

diff --git a/drivers/net/ethernet/netronome/nfp/flower/match.c b/drivers/net/ethernet/netronome/nfp/flower/match.c
index 9cc3ba17ff69..1079f37dbc78 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/match.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/match.c
@@ -10,9 +10,8 @@
 static void
 nfp_flower_compile_meta_tci(struct nfp_flower_meta_tci *ext,
 			    struct nfp_flower_meta_tci *msk,
-			    struct flow_cls_offload *flow, u8 key_type)
+			    struct flow_rule *rule, u8 key_type)
 {
-	struct flow_rule *rule = flow_cls_offload_flow_rule(flow);
 	u16 tmp_tci;
 
 	memset(ext, 0, sizeof(struct nfp_flower_meta_tci));
@@ -77,11 +76,8 @@ nfp_flower_compile_port(struct nfp_flower_in_port *frame, u32 cmsg_port,
 
 static void
 nfp_flower_compile_mac(struct nfp_flower_mac_mpls *ext,
-		       struct nfp_flower_mac_mpls *msk,
-		       struct flow_cls_offload *flow)
+		       struct nfp_flower_mac_mpls *msk, struct flow_rule *rule)
 {
-	struct flow_rule *rule = flow_cls_offload_flow_rule(flow);
-
 	memset(ext, 0, sizeof(struct nfp_flower_mac_mpls));
 	memset(msk, 0, sizeof(struct nfp_flower_mac_mpls));
 
@@ -130,10 +126,8 @@ nfp_flower_compile_mac(struct nfp_flower_mac_mpls *ext,
 static void
 nfp_flower_compile_tport(struct nfp_flower_tp_ports *ext,
 			 struct nfp_flower_tp_ports *msk,
-			 struct flow_cls_offload *flow)
+			 struct flow_rule *rule)
 {
-	struct flow_rule *rule = flow_cls_offload_flow_rule(flow);
-
 	memset(ext, 0, sizeof(struct nfp_flower_tp_ports));
 	memset(msk, 0, sizeof(struct nfp_flower_tp_ports));
 
@@ -150,11 +144,8 @@ nfp_flower_compile_tport(struct nfp_flower_tp_ports *ext,
 
 static void
 nfp_flower_compile_ip_ext(struct nfp_flower_ip_ext *ext,
-			  struct nfp_flower_ip_ext *msk,
-			  struct flow_cls_offload *flow)
+			  struct nfp_flower_ip_ext *msk, struct flow_rule *rule)
 {
-	struct flow_rule *rule = flow_cls_offload_flow_rule(flow);
-
 	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_BASIC)) {
 		struct flow_match_basic match;
 
@@ -224,10 +215,8 @@ nfp_flower_compile_ip_ext(struct nfp_flower_ip_ext *ext,
 
 static void
 nfp_flower_compile_ipv4(struct nfp_flower_ipv4 *ext,
-			struct nfp_flower_ipv4 *msk,
-			struct flow_cls_offload *flow)
+			struct nfp_flower_ipv4 *msk, struct flow_rule *rule)
 {
-	struct flow_rule *rule = flow_cls_offload_flow_rule(flow);
 	struct flow_match_ipv4_addrs match;
 
 	memset(ext, 0, sizeof(struct nfp_flower_ipv4));
@@ -241,16 +230,13 @@ nfp_flower_compile_ipv4(struct nfp_flower_ipv4 *ext,
 		msk->ipv4_dst = match.mask->dst;
 	}
 
-	nfp_flower_compile_ip_ext(&ext->ip_ext, &msk->ip_ext, flow);
+	nfp_flower_compile_ip_ext(&ext->ip_ext, &msk->ip_ext, rule);
 }
 
 static void
 nfp_flower_compile_ipv6(struct nfp_flower_ipv6 *ext,
-			struct nfp_flower_ipv6 *msk,
-			struct flow_cls_offload *flow)
+			struct nfp_flower_ipv6 *msk, struct flow_rule *rule)
 {
-	struct flow_rule *rule = flow_cls_offload_flow_rule(flow);
-
 	memset(ext, 0, sizeof(struct nfp_flower_ipv6));
 	memset(msk, 0, sizeof(struct nfp_flower_ipv6));
 
@@ -264,16 +250,15 @@ nfp_flower_compile_ipv6(struct nfp_flower_ipv6 *ext,
 		msk->ipv6_dst = match.mask->dst;
 	}
 
-	nfp_flower_compile_ip_ext(&ext->ip_ext, &msk->ip_ext, flow);
+	nfp_flower_compile_ip_ext(&ext->ip_ext, &msk->ip_ext, rule);
 }
 
 static int
-nfp_flower_compile_geneve_opt(void *ext, void *msk,
-			      struct flow_cls_offload *flow)
+nfp_flower_compile_geneve_opt(void *ext, void *msk, struct flow_rule *rule)
 {
 	struct flow_match_enc_opts match;
 
-	flow_rule_match_enc_opts(flow->rule, &match);
+	flow_rule_match_enc_opts(rule, &match);
 	memcpy(ext, match.key->data, match.key->len);
 	memcpy(msk, match.mask->data, match.mask->len);
 
@@ -283,10 +268,8 @@ nfp_flower_compile_geneve_opt(void *ext, void *msk,
 static void
 nfp_flower_compile_tun_ipv4_addrs(struct nfp_flower_tun_ipv4 *ext,
 				  struct nfp_flower_tun_ipv4 *msk,
-				  struct flow_cls_offload *flow)
+				  struct flow_rule *rule)
 {
-	struct flow_rule *rule = flow_cls_offload_flow_rule(flow);
-
 	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_IPV4_ADDRS)) {
 		struct flow_match_ipv4_addrs match;
 
@@ -301,10 +284,8 @@ nfp_flower_compile_tun_ipv4_addrs(struct nfp_flower_tun_ipv4 *ext,
 static void
 nfp_flower_compile_tun_ip_ext(struct nfp_flower_tun_ip_ext *ext,
 			      struct nfp_flower_tun_ip_ext *msk,
-			      struct flow_cls_offload *flow)
+			      struct flow_rule *rule)
 {
-	struct flow_rule *rule = flow_cls_offload_flow_rule(flow);
-
 	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_IP)) {
 		struct flow_match_ip match;
 
@@ -319,10 +300,8 @@ nfp_flower_compile_tun_ip_ext(struct nfp_flower_tun_ip_ext *ext,
 static void
 nfp_flower_compile_ipv4_gre_tun(struct nfp_flower_ipv4_gre_tun *ext,
 				struct nfp_flower_ipv4_gre_tun *msk,
-				struct flow_cls_offload *flow)
+				struct flow_rule *rule)
 {
-	struct flow_rule *rule = flow_cls_offload_flow_rule(flow);
-
 	memset(ext, 0, sizeof(struct nfp_flower_ipv4_gre_tun));
 	memset(msk, 0, sizeof(struct nfp_flower_ipv4_gre_tun));
 
@@ -341,17 +320,15 @@ nfp_flower_compile_ipv4_gre_tun(struct nfp_flower_ipv4_gre_tun *ext,
 		msk->tun_flags = cpu_to_be16(NFP_FL_GRE_FLAG_KEY);
 	}
 
-	nfp_flower_compile_tun_ipv4_addrs(&ext->ipv4, &msk->ipv4, flow);
-	nfp_flower_compile_tun_ip_ext(&ext->ip_ext, &msk->ip_ext, flow);
+	nfp_flower_compile_tun_ipv4_addrs(&ext->ipv4, &msk->ipv4, rule);
+	nfp_flower_compile_tun_ip_ext(&ext->ip_ext, &msk->ip_ext, rule);
 }
 
 static void
 nfp_flower_compile_ipv4_udp_tun(struct nfp_flower_ipv4_udp_tun *ext,
 				struct nfp_flower_ipv4_udp_tun *msk,
-				struct flow_cls_offload *flow)
+				struct flow_rule *rule)
 {
-	struct flow_rule *rule = flow_cls_offload_flow_rule(flow);
-
 	memset(ext, 0, sizeof(struct nfp_flower_ipv4_udp_tun));
 	memset(msk, 0, sizeof(struct nfp_flower_ipv4_udp_tun));
 
@@ -366,8 +343,8 @@ nfp_flower_compile_ipv4_udp_tun(struct nfp_flower_ipv4_udp_tun *ext,
 		msk->tun_id = cpu_to_be32(temp_vni);
 	}
 
-	nfp_flower_compile_tun_ipv4_addrs(&ext->ipv4, &msk->ipv4, flow);
-	nfp_flower_compile_tun_ip_ext(&ext->ip_ext, &msk->ip_ext, flow);
+	nfp_flower_compile_tun_ipv4_addrs(&ext->ipv4, &msk->ipv4, rule);
+	nfp_flower_compile_tun_ip_ext(&ext->ip_ext, &msk->ip_ext, rule);
 }
 
 int nfp_flower_compile_flow_match(struct nfp_app *app,
@@ -378,6 +355,7 @@ int nfp_flower_compile_flow_match(struct nfp_app *app,
 				  enum nfp_flower_tun_type tun_type,
 				  struct netlink_ext_ack *extack)
 {
+	struct flow_rule *rule = flow_cls_offload_flow_rule(flow);
 	u32 port_id;
 	int err;
 	u8 *ext;
@@ -393,7 +371,7 @@ int nfp_flower_compile_flow_match(struct nfp_app *app,
 
 	nfp_flower_compile_meta_tci((struct nfp_flower_meta_tci *)ext,
 				    (struct nfp_flower_meta_tci *)msk,
-				    flow, key_ls->key_layer);
+				    rule, key_ls->key_layer);
 	ext += sizeof(struct nfp_flower_meta_tci);
 	msk += sizeof(struct nfp_flower_meta_tci);
 
@@ -425,7 +403,7 @@ int nfp_flower_compile_flow_match(struct nfp_app *app,
 	if (NFP_FLOWER_LAYER_MAC & key_ls->key_layer) {
 		nfp_flower_compile_mac((struct nfp_flower_mac_mpls *)ext,
 				       (struct nfp_flower_mac_mpls *)msk,
-				       flow);
+				       rule);
 		ext += sizeof(struct nfp_flower_mac_mpls);
 		msk += sizeof(struct nfp_flower_mac_mpls);
 	}
@@ -433,7 +411,7 @@ int nfp_flower_compile_flow_match(struct nfp_app *app,
 	if (NFP_FLOWER_LAYER_TP & key_ls->key_layer) {
 		nfp_flower_compile_tport((struct nfp_flower_tp_ports *)ext,
 					 (struct nfp_flower_tp_ports *)msk,
-					 flow);
+					 rule);
 		ext += sizeof(struct nfp_flower_tp_ports);
 		msk += sizeof(struct nfp_flower_tp_ports);
 	}
@@ -441,7 +419,7 @@ int nfp_flower_compile_flow_match(struct nfp_app *app,
 	if (NFP_FLOWER_LAYER_IPV4 & key_ls->key_layer) {
 		nfp_flower_compile_ipv4((struct nfp_flower_ipv4 *)ext,
 					(struct nfp_flower_ipv4 *)msk,
-					flow);
+					rule);
 		ext += sizeof(struct nfp_flower_ipv4);
 		msk += sizeof(struct nfp_flower_ipv4);
 	}
@@ -449,7 +427,7 @@ int nfp_flower_compile_flow_match(struct nfp_app *app,
 	if (NFP_FLOWER_LAYER_IPV6 & key_ls->key_layer) {
 		nfp_flower_compile_ipv6((struct nfp_flower_ipv6 *)ext,
 					(struct nfp_flower_ipv6 *)msk,
-					flow);
+					rule);
 		ext += sizeof(struct nfp_flower_ipv6);
 		msk += sizeof(struct nfp_flower_ipv6);
 	}
@@ -457,7 +435,7 @@ int nfp_flower_compile_flow_match(struct nfp_app *app,
 	if (key_ls->key_layer_two & NFP_FLOWER_LAYER2_GRE) {
 		__be32 tun_dst;
 
-		nfp_flower_compile_ipv4_gre_tun((void *)ext, (void *)msk, flow);
+		nfp_flower_compile_ipv4_gre_tun((void *)ext, (void *)msk, rule);
 		tun_dst = ((struct nfp_flower_ipv4_gre_tun *)ext)->ipv4.dst;
 		ext += sizeof(struct nfp_flower_ipv4_gre_tun);
 		msk += sizeof(struct nfp_flower_ipv4_gre_tun);
@@ -473,7 +451,7 @@ int nfp_flower_compile_flow_match(struct nfp_app *app,
 	    key_ls->key_layer_two & NFP_FLOWER_LAYER2_GENEVE) {
 		__be32 tun_dst;
 
-		nfp_flower_compile_ipv4_udp_tun((void *)ext, (void *)msk, flow);
+		nfp_flower_compile_ipv4_udp_tun((void *)ext, (void *)msk, rule);
 		tun_dst = ((struct nfp_flower_ipv4_udp_tun *)ext)->ipv4.dst;
 		ext += sizeof(struct nfp_flower_ipv4_udp_tun);
 		msk += sizeof(struct nfp_flower_ipv4_udp_tun);
@@ -485,7 +463,7 @@ int nfp_flower_compile_flow_match(struct nfp_app *app,
 		nfp_tunnel_add_ipv4_off(app, tun_dst);
 
 		if (key_ls->key_layer_two & NFP_FLOWER_LAYER2_GENEVE_OP) {
-			err = nfp_flower_compile_geneve_opt(ext, msk, flow);
+			err = nfp_flower_compile_geneve_opt(ext, msk, rule);
 			if (err)
 				return err;
 		}
-- 
cgit v1.2.3-59-g8ed1b


From 77a5f73a8b337c22ccadefbed30d2a36425388d1 Mon Sep 17 00:00:00 2001
From: John Hurley <john.hurley@netronome.com>
Date: Tue, 17 Dec 2019 21:57:17 +0000
Subject: nfp: flower: move udp tunnel key match compilation to helper function

IPv4 UDP and GRE tunnel match rule compile helpers share functions for
compiling fields such as IP addresses. However, they handle fields such
tunnel IDs differently.

Create new helper functions for compiling GRE and UDP tunnel key data.
This is in preparation for supporting IPv6 tunnels where these new
functions can be reused.

This patch does not change functionality.

Signed-off-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/netronome/nfp/flower/match.c | 57 ++++++++++++++---------
 1 file changed, 35 insertions(+), 22 deletions(-)

diff --git a/drivers/net/ethernet/netronome/nfp/flower/match.c b/drivers/net/ethernet/netronome/nfp/flower/match.c
index 1079f37dbc78..1cf8eaeb7ec4 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/match.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/match.c
@@ -297,6 +297,38 @@ nfp_flower_compile_tun_ip_ext(struct nfp_flower_tun_ip_ext *ext,
 	}
 }
 
+static void
+nfp_flower_compile_tun_udp_key(__be32 *key, __be32 *key_msk,
+			       struct flow_rule *rule)
+{
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_KEYID)) {
+		struct flow_match_enc_keyid match;
+		u32 vni;
+
+		flow_rule_match_enc_keyid(rule, &match);
+		vni = be32_to_cpu(match.key->keyid) << NFP_FL_TUN_VNI_OFFSET;
+		*key = cpu_to_be32(vni);
+		vni = be32_to_cpu(match.mask->keyid) << NFP_FL_TUN_VNI_OFFSET;
+		*key_msk = cpu_to_be32(vni);
+	}
+}
+
+static void
+nfp_flower_compile_tun_gre_key(__be32 *key, __be32 *key_msk, __be16 *flags,
+			       __be16 *flags_msk, struct flow_rule *rule)
+{
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_KEYID)) {
+		struct flow_match_enc_keyid match;
+
+		flow_rule_match_enc_keyid(rule, &match);
+		*key = match.key->keyid;
+		*key_msk = match.mask->keyid;
+
+		*flags = cpu_to_be16(NFP_FL_GRE_FLAG_KEY);
+		*flags_msk = cpu_to_be16(NFP_FL_GRE_FLAG_KEY);
+	}
+}
+
 static void
 nfp_flower_compile_ipv4_gre_tun(struct nfp_flower_ipv4_gre_tun *ext,
 				struct nfp_flower_ipv4_gre_tun *msk,
@@ -309,19 +341,10 @@ nfp_flower_compile_ipv4_gre_tun(struct nfp_flower_ipv4_gre_tun *ext,
 	ext->ethertype = cpu_to_be16(ETH_P_TEB);
 	msk->ethertype = cpu_to_be16(~0);
 
-	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_KEYID)) {
-		struct flow_match_enc_keyid match;
-
-		flow_rule_match_enc_keyid(rule, &match);
-		ext->tun_key = match.key->keyid;
-		msk->tun_key = match.mask->keyid;
-
-		ext->tun_flags = cpu_to_be16(NFP_FL_GRE_FLAG_KEY);
-		msk->tun_flags = cpu_to_be16(NFP_FL_GRE_FLAG_KEY);
-	}
-
 	nfp_flower_compile_tun_ipv4_addrs(&ext->ipv4, &msk->ipv4, rule);
 	nfp_flower_compile_tun_ip_ext(&ext->ip_ext, &msk->ip_ext, rule);
+	nfp_flower_compile_tun_gre_key(&ext->tun_key, &msk->tun_key,
+				       &ext->tun_flags, &msk->tun_flags, rule);
 }
 
 static void
@@ -332,19 +355,9 @@ nfp_flower_compile_ipv4_udp_tun(struct nfp_flower_ipv4_udp_tun *ext,
 	memset(ext, 0, sizeof(struct nfp_flower_ipv4_udp_tun));
 	memset(msk, 0, sizeof(struct nfp_flower_ipv4_udp_tun));
 
-	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_KEYID)) {
-		struct flow_match_enc_keyid match;
-		u32 temp_vni;
-
-		flow_rule_match_enc_keyid(rule, &match);
-		temp_vni = be32_to_cpu(match.key->keyid) << NFP_FL_TUN_VNI_OFFSET;
-		ext->tun_id = cpu_to_be32(temp_vni);
-		temp_vni = be32_to_cpu(match.mask->keyid) << NFP_FL_TUN_VNI_OFFSET;
-		msk->tun_id = cpu_to_be32(temp_vni);
-	}
-
 	nfp_flower_compile_tun_ipv4_addrs(&ext->ipv4, &msk->ipv4, rule);
 	nfp_flower_compile_tun_ip_ext(&ext->ip_ext, &msk->ip_ext, rule);
+	nfp_flower_compile_tun_udp_key(&ext->tun_id, &msk->tun_id, rule);
 }
 
 int nfp_flower_compile_flow_match(struct nfp_app *app,
-- 
cgit v1.2.3-59-g8ed1b


From 78a722af4ad9cb2d82f230f697220af5aa6723c1 Mon Sep 17 00:00:00 2001
From: John Hurley <john.hurley@netronome.com>
Date: Tue, 17 Dec 2019 21:57:18 +0000
Subject: nfp: flower: compile match for IPv6 tunnels

IPv6 tunnel matches are now supported by firmware. Modify the NFP driver
to compile these match rules. IPv6 matches are handled similar to IPv4
tunnels with the difference the address length. The type of tunnel is
indicated by the same bitmap that is used in IPv4 with an extra bit
signifying that the IPv6 variation should be used.

Only compile IPv6 tunnel matches when the fw features symbol indicated
that they are compatible with the currently loaded fw.

Signed-off-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/netronome/nfp/flower/cmsg.h   |  83 +++++++++++++++
 drivers/net/ethernet/netronome/nfp/flower/main.h   |   1 +
 drivers/net/ethernet/netronome/nfp/flower/match.c  | 111 ++++++++++++++++-----
 .../net/ethernet/netronome/nfp/flower/offload.c    | 102 ++++++++++++++-----
 4 files changed, 246 insertions(+), 51 deletions(-)

diff --git a/drivers/net/ethernet/netronome/nfp/flower/cmsg.h b/drivers/net/ethernet/netronome/nfp/flower/cmsg.h
index 7eb2ec8969c3..8d12805baa0d 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/cmsg.h
+++ b/drivers/net/ethernet/netronome/nfp/flower/cmsg.h
@@ -26,6 +26,7 @@
 #define NFP_FLOWER_LAYER2_GRE		BIT(0)
 #define NFP_FLOWER_LAYER2_GENEVE	BIT(5)
 #define NFP_FLOWER_LAYER2_GENEVE_OP	BIT(6)
+#define NFP_FLOWER_LAYER2_TUN_IPV6	BIT(7)
 
 #define NFP_FLOWER_MASK_VLAN_PRIO	GENMASK(15, 13)
 #define NFP_FLOWER_MASK_VLAN_PRESENT	BIT(12)
@@ -63,6 +64,7 @@
 #define NFP_FL_MAX_GENEVE_OPT_ACT	32
 #define NFP_FL_MAX_GENEVE_OPT_CNT	64
 #define NFP_FL_MAX_GENEVE_OPT_KEY	32
+#define NFP_FL_MAX_GENEVE_OPT_KEY_V6	8
 
 /* Action opcodes */
 #define NFP_FL_ACTION_OPCODE_OUTPUT		0
@@ -387,6 +389,11 @@ struct nfp_flower_tun_ipv4 {
 	__be32 dst;
 };
 
+struct nfp_flower_tun_ipv6 {
+	struct in6_addr src;
+	struct in6_addr dst;
+};
+
 struct nfp_flower_tun_ip_ext {
 	u8 tos;
 	u8 ttl;
@@ -416,6 +423,42 @@ struct nfp_flower_ipv4_udp_tun {
 	__be32 tun_id;
 };
 
+/* Flow Frame IPv6 UDP TUNNEL --> Tunnel details (11W/44B)
+ * -----------------------------------------------------------------
+ *    3                   2                   1
+ *  1 0 9 8 7 6 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |                  ipv6_addr_src,   31 - 0                      |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |                  ipv6_addr_src,  63 - 32                      |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |                  ipv6_addr_src,  95 - 64                      |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |                  ipv6_addr_src, 127 - 96                      |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |                  ipv6_addr_dst,   31 - 0                      |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |                  ipv6_addr_dst,  63 - 32                      |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |                  ipv6_addr_dst,  95 - 64                      |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |                  ipv6_addr_dst, 127 - 96                      |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |           Reserved            |      tos      |      ttl      |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |                            Reserved                           |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |                     VNI                       |   Reserved    |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ */
+struct nfp_flower_ipv6_udp_tun {
+	struct nfp_flower_tun_ipv6 ipv6;
+	__be16 reserved1;
+	struct nfp_flower_tun_ip_ext ip_ext;
+	__be32 reserved2;
+	__be32 tun_id;
+};
+
 /* Flow Frame GRE TUNNEL --> Tunnel details (6W/24B)
  * -----------------------------------------------------------------
  *    3                   2                   1
@@ -445,6 +488,46 @@ struct nfp_flower_ipv4_gre_tun {
 	__be32 reserved2;
 };
 
+/* Flow Frame GRE TUNNEL V6 --> Tunnel details (12W/48B)
+ * -----------------------------------------------------------------
+ *    3                   2                   1
+ *  1 0 9 8 7 6 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |                  ipv6_addr_src,   31 - 0                      |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |                  ipv6_addr_src,  63 - 32                      |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |                  ipv6_addr_src,  95 - 64                      |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |                  ipv6_addr_src, 127 - 96                      |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |                  ipv6_addr_dst,   31 - 0                      |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |                  ipv6_addr_dst,  63 - 32                      |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |                  ipv6_addr_dst,  95 - 64                      |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |                  ipv6_addr_dst, 127 - 96                      |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |           tun_flags           |       tos     |       ttl     |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |            Reserved           |           Ethertype           |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |                              Key                              |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |                           Reserved                            |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ */
+struct nfp_flower_ipv6_gre_tun {
+	struct nfp_flower_tun_ipv6 ipv6;
+	__be16 tun_flags;
+	struct nfp_flower_tun_ip_ext ip_ext;
+	__be16 reserved1;
+	__be16 ethertype;
+	__be32 tun_key;
+	__be32 reserved2;
+};
+
 struct nfp_flower_geneve_options {
 	u8 data[NFP_FL_MAX_GENEVE_OPT_KEY];
 };
diff --git a/drivers/net/ethernet/netronome/nfp/flower/main.h b/drivers/net/ethernet/netronome/nfp/flower/main.h
index 31d94592a7c0..386d6297e0f1 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/main.h
+++ b/drivers/net/ethernet/netronome/nfp/flower/main.h
@@ -43,6 +43,7 @@ struct nfp_app;
 #define NFP_FL_FEATS_VF_RLIM		BIT(4)
 #define NFP_FL_FEATS_FLOW_MOD		BIT(5)
 #define NFP_FL_FEATS_PRE_TUN_RULES	BIT(6)
+#define NFP_FL_FEATS_IPV6_TUN		BIT(7)
 #define NFP_FL_FEATS_FLOW_MERGE		BIT(30)
 #define NFP_FL_FEATS_LAG		BIT(31)
 
diff --git a/drivers/net/ethernet/netronome/nfp/flower/match.c b/drivers/net/ethernet/netronome/nfp/flower/match.c
index 1cf8eaeb7ec4..2410ead05a06 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/match.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/match.c
@@ -281,6 +281,22 @@ nfp_flower_compile_tun_ipv4_addrs(struct nfp_flower_tun_ipv4 *ext,
 	}
 }
 
+static void
+nfp_flower_compile_tun_ipv6_addrs(struct nfp_flower_tun_ipv6 *ext,
+				  struct nfp_flower_tun_ipv6 *msk,
+				  struct flow_rule *rule)
+{
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_IPV6_ADDRS)) {
+		struct flow_match_ipv6_addrs match;
+
+		flow_rule_match_enc_ipv6_addrs(rule, &match);
+		ext->src = match.key->src;
+		ext->dst = match.key->dst;
+		msk->src = match.mask->src;
+		msk->dst = match.mask->dst;
+	}
+}
+
 static void
 nfp_flower_compile_tun_ip_ext(struct nfp_flower_tun_ip_ext *ext,
 			      struct nfp_flower_tun_ip_ext *msk,
@@ -360,6 +376,37 @@ nfp_flower_compile_ipv4_udp_tun(struct nfp_flower_ipv4_udp_tun *ext,
 	nfp_flower_compile_tun_udp_key(&ext->tun_id, &msk->tun_id, rule);
 }
 
+static void
+nfp_flower_compile_ipv6_udp_tun(struct nfp_flower_ipv6_udp_tun *ext,
+				struct nfp_flower_ipv6_udp_tun *msk,
+				struct flow_rule *rule)
+{
+	memset(ext, 0, sizeof(struct nfp_flower_ipv6_udp_tun));
+	memset(msk, 0, sizeof(struct nfp_flower_ipv6_udp_tun));
+
+	nfp_flower_compile_tun_ipv6_addrs(&ext->ipv6, &msk->ipv6, rule);
+	nfp_flower_compile_tun_ip_ext(&ext->ip_ext, &msk->ip_ext, rule);
+	nfp_flower_compile_tun_udp_key(&ext->tun_id, &msk->tun_id, rule);
+}
+
+static void
+nfp_flower_compile_ipv6_gre_tun(struct nfp_flower_ipv6_gre_tun *ext,
+				struct nfp_flower_ipv6_gre_tun *msk,
+				struct flow_rule *rule)
+{
+	memset(ext, 0, sizeof(struct nfp_flower_ipv6_gre_tun));
+	memset(msk, 0, sizeof(struct nfp_flower_ipv6_gre_tun));
+
+	/* NVGRE is the only supported GRE tunnel type */
+	ext->ethertype = cpu_to_be16(ETH_P_TEB);
+	msk->ethertype = cpu_to_be16(~0);
+
+	nfp_flower_compile_tun_ipv6_addrs(&ext->ipv6, &msk->ipv6, rule);
+	nfp_flower_compile_tun_ip_ext(&ext->ip_ext, &msk->ip_ext, rule);
+	nfp_flower_compile_tun_gre_key(&ext->tun_key, &msk->tun_key,
+				       &ext->tun_flags, &msk->tun_flags, rule);
+}
+
 int nfp_flower_compile_flow_match(struct nfp_app *app,
 				  struct flow_cls_offload *flow,
 				  struct nfp_fl_key_ls *key_ls,
@@ -446,34 +493,50 @@ int nfp_flower_compile_flow_match(struct nfp_app *app,
 	}
 
 	if (key_ls->key_layer_two & NFP_FLOWER_LAYER2_GRE) {
-		__be32 tun_dst;
-
-		nfp_flower_compile_ipv4_gre_tun((void *)ext, (void *)msk, rule);
-		tun_dst = ((struct nfp_flower_ipv4_gre_tun *)ext)->ipv4.dst;
-		ext += sizeof(struct nfp_flower_ipv4_gre_tun);
-		msk += sizeof(struct nfp_flower_ipv4_gre_tun);
-
-		/* Store the tunnel destination in the rule data.
-		 * This must be present and be an exact match.
-		 */
-		nfp_flow->nfp_tun_ipv4_addr = tun_dst;
-		nfp_tunnel_add_ipv4_off(app, tun_dst);
+		if (key_ls->key_layer_two & NFP_FLOWER_LAYER2_TUN_IPV6) {
+			nfp_flower_compile_ipv6_gre_tun((void *)ext,
+							(void *)msk, rule);
+			ext += sizeof(struct nfp_flower_ipv6_gre_tun);
+			msk += sizeof(struct nfp_flower_ipv6_gre_tun);
+		} else {
+			__be32 dst;
+
+			nfp_flower_compile_ipv4_gre_tun((void *)ext,
+							(void *)msk, rule);
+			dst = ((struct nfp_flower_ipv4_gre_tun *)ext)->ipv4.dst;
+			ext += sizeof(struct nfp_flower_ipv4_gre_tun);
+			msk += sizeof(struct nfp_flower_ipv4_gre_tun);
+
+			/* Store the tunnel destination in the rule data.
+			 * This must be present and be an exact match.
+			 */
+			nfp_flow->nfp_tun_ipv4_addr = dst;
+			nfp_tunnel_add_ipv4_off(app, dst);
+		}
 	}
 
 	if (key_ls->key_layer & NFP_FLOWER_LAYER_VXLAN ||
 	    key_ls->key_layer_two & NFP_FLOWER_LAYER2_GENEVE) {
-		__be32 tun_dst;
-
-		nfp_flower_compile_ipv4_udp_tun((void *)ext, (void *)msk, rule);
-		tun_dst = ((struct nfp_flower_ipv4_udp_tun *)ext)->ipv4.dst;
-		ext += sizeof(struct nfp_flower_ipv4_udp_tun);
-		msk += sizeof(struct nfp_flower_ipv4_udp_tun);
-
-		/* Store the tunnel destination in the rule data.
-		 * This must be present and be an exact match.
-		 */
-		nfp_flow->nfp_tun_ipv4_addr = tun_dst;
-		nfp_tunnel_add_ipv4_off(app, tun_dst);
+		if (key_ls->key_layer_two & NFP_FLOWER_LAYER2_TUN_IPV6) {
+			nfp_flower_compile_ipv6_udp_tun((void *)ext,
+							(void *)msk, rule);
+			ext += sizeof(struct nfp_flower_ipv6_udp_tun);
+			msk += sizeof(struct nfp_flower_ipv6_udp_tun);
+		} else {
+			__be32 dst;
+
+			nfp_flower_compile_ipv4_udp_tun((void *)ext,
+							(void *)msk, rule);
+			dst = ((struct nfp_flower_ipv4_udp_tun *)ext)->ipv4.dst;
+			ext += sizeof(struct nfp_flower_ipv4_udp_tun);
+			msk += sizeof(struct nfp_flower_ipv4_udp_tun);
+
+			/* Store the tunnel destination in the rule data.
+			 * This must be present and be an exact match.
+			 */
+			nfp_flow->nfp_tun_ipv4_addr = dst;
+			nfp_tunnel_add_ipv4_off(app, dst);
+		}
 
 		if (key_ls->key_layer_two & NFP_FLOWER_LAYER2_GENEVE_OP) {
 			err = nfp_flower_compile_geneve_opt(ext, msk, rule);
diff --git a/drivers/net/ethernet/netronome/nfp/flower/offload.c b/drivers/net/ethernet/netronome/nfp/flower/offload.c
index 987ae221f6be..099735086fb2 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/offload.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/offload.c
@@ -54,6 +54,10 @@
 	(BIT(FLOW_DISSECTOR_KEY_ENC_CONTROL) | \
 	 BIT(FLOW_DISSECTOR_KEY_ENC_IPV4_ADDRS))
 
+#define NFP_FLOWER_WHITELIST_TUN_DISSECTOR_V6_R \
+	(BIT(FLOW_DISSECTOR_KEY_ENC_CONTROL) | \
+	 BIT(FLOW_DISSECTOR_KEY_ENC_IPV6_ADDRS))
+
 #define NFP_FLOWER_MERGE_FIELDS \
 	(NFP_FLOWER_LAYER_PORT | \
 	 NFP_FLOWER_LAYER_MAC | \
@@ -146,10 +150,11 @@ static bool nfp_flower_check_higher_than_l3(struct flow_cls_offload *f)
 
 static int
 nfp_flower_calc_opt_layer(struct flow_dissector_key_enc_opts *enc_opts,
-			  u32 *key_layer_two, int *key_size,
+			  u32 *key_layer_two, int *key_size, bool ipv6,
 			  struct netlink_ext_ack *extack)
 {
-	if (enc_opts->len > NFP_FL_MAX_GENEVE_OPT_KEY) {
+	if (enc_opts->len > NFP_FL_MAX_GENEVE_OPT_KEY ||
+	    (ipv6 && enc_opts->len > NFP_FL_MAX_GENEVE_OPT_KEY_V6)) {
 		NL_SET_ERR_MSG_MOD(extack, "unsupported offload: geneve options exceed maximum length");
 		return -EOPNOTSUPP;
 	}
@@ -167,7 +172,7 @@ nfp_flower_calc_udp_tun_layer(struct flow_dissector_key_ports *enc_ports,
 			      struct flow_dissector_key_enc_opts *enc_op,
 			      u32 *key_layer_two, u8 *key_layer, int *key_size,
 			      struct nfp_flower_priv *priv,
-			      enum nfp_flower_tun_type *tun_type,
+			      enum nfp_flower_tun_type *tun_type, bool ipv6,
 			      struct netlink_ext_ack *extack)
 {
 	int err;
@@ -176,7 +181,15 @@ nfp_flower_calc_udp_tun_layer(struct flow_dissector_key_ports *enc_ports,
 	case htons(IANA_VXLAN_UDP_PORT):
 		*tun_type = NFP_FL_TUNNEL_VXLAN;
 		*key_layer |= NFP_FLOWER_LAYER_VXLAN;
-		*key_size += sizeof(struct nfp_flower_ipv4_udp_tun);
+
+		if (ipv6) {
+			*key_layer |= NFP_FLOWER_LAYER_EXT_META;
+			*key_size += sizeof(struct nfp_flower_ext_meta);
+			*key_layer_two |= NFP_FLOWER_LAYER2_TUN_IPV6;
+			*key_size += sizeof(struct nfp_flower_ipv6_udp_tun);
+		} else {
+			*key_size += sizeof(struct nfp_flower_ipv4_udp_tun);
+		}
 
 		if (enc_op) {
 			NL_SET_ERR_MSG_MOD(extack, "unsupported offload: encap options not supported on vxlan tunnels");
@@ -192,7 +205,13 @@ nfp_flower_calc_udp_tun_layer(struct flow_dissector_key_ports *enc_ports,
 		*key_layer |= NFP_FLOWER_LAYER_EXT_META;
 		*key_size += sizeof(struct nfp_flower_ext_meta);
 		*key_layer_two |= NFP_FLOWER_LAYER2_GENEVE;
-		*key_size += sizeof(struct nfp_flower_ipv4_udp_tun);
+
+		if (ipv6) {
+			*key_layer_two |= NFP_FLOWER_LAYER2_TUN_IPV6;
+			*key_size += sizeof(struct nfp_flower_ipv6_udp_tun);
+		} else {
+			*key_size += sizeof(struct nfp_flower_ipv4_udp_tun);
+		}
 
 		if (!enc_op)
 			break;
@@ -200,8 +219,8 @@ nfp_flower_calc_udp_tun_layer(struct flow_dissector_key_ports *enc_ports,
 			NL_SET_ERR_MSG_MOD(extack, "unsupported offload: loaded firmware does not support geneve option offload");
 			return -EOPNOTSUPP;
 		}
-		err = nfp_flower_calc_opt_layer(enc_op, key_layer_two,
-						key_size, extack);
+		err = nfp_flower_calc_opt_layer(enc_op, key_layer_two, key_size,
+						ipv6, extack);
 		if (err)
 			return err;
 		break;
@@ -237,6 +256,8 @@ nfp_flower_calculate_key_layers(struct nfp_app *app,
 
 	/* If any tun dissector is used then the required set must be used. */
 	if (dissector->used_keys & NFP_FLOWER_WHITELIST_TUN_DISSECTOR &&
+	    (dissector->used_keys & NFP_FLOWER_WHITELIST_TUN_DISSECTOR_V6_R)
+	    != NFP_FLOWER_WHITELIST_TUN_DISSECTOR_V6_R &&
 	    (dissector->used_keys & NFP_FLOWER_WHITELIST_TUN_DISSECTOR_R)
 	    != NFP_FLOWER_WHITELIST_TUN_DISSECTOR_R) {
 		NL_SET_ERR_MSG_MOD(extack, "unsupported offload: tunnel match not supported");
@@ -268,8 +289,10 @@ nfp_flower_calculate_key_layers(struct nfp_app *app,
 	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_CONTROL)) {
 		struct flow_match_enc_opts enc_op = { NULL, NULL };
 		struct flow_match_ipv4_addrs ipv4_addrs;
+		struct flow_match_ipv6_addrs ipv6_addrs;
 		struct flow_match_control enc_ctl;
 		struct flow_match_ports enc_ports;
+		bool ipv6_tun = false;
 
 		flow_rule_match_enc_control(rule, &enc_ctl);
 
@@ -277,38 +300,62 @@ nfp_flower_calculate_key_layers(struct nfp_app *app,
 			NL_SET_ERR_MSG_MOD(extack, "unsupported offload: wildcarded protocols on tunnels are not supported");
 			return -EOPNOTSUPP;
 		}
-		if (enc_ctl.key->addr_type != FLOW_DISSECTOR_KEY_IPV4_ADDRS) {
-			NL_SET_ERR_MSG_MOD(extack, "unsupported offload: only IPv4 tunnels are supported");
+
+		ipv6_tun = enc_ctl.key->addr_type ==
+				FLOW_DISSECTOR_KEY_IPV6_ADDRS;
+		if (ipv6_tun &&
+		    !(priv->flower_ext_feats & NFP_FL_FEATS_IPV6_TUN)) {
+			NL_SET_ERR_MSG_MOD(extack, "unsupported offload: firmware does not support IPv6 tunnels");
 			return -EOPNOTSUPP;
 		}
 
-		/* These fields are already verified as used. */
-		flow_rule_match_enc_ipv4_addrs(rule, &ipv4_addrs);
-		if (ipv4_addrs.mask->dst != cpu_to_be32(~0)) {
-			NL_SET_ERR_MSG_MOD(extack, "unsupported offload: only an exact match IPv4 destination address is supported");
+		if (!ipv6_tun &&
+		    enc_ctl.key->addr_type != FLOW_DISSECTOR_KEY_IPV4_ADDRS) {
+			NL_SET_ERR_MSG_MOD(extack, "unsupported offload: tunnel address type not IPv4 or IPv6");
 			return -EOPNOTSUPP;
 		}
 
+		if (ipv6_tun) {
+			flow_rule_match_enc_ipv6_addrs(rule, &ipv6_addrs);
+			if (memchr_inv(&ipv6_addrs.mask->dst, 0xff,
+				       sizeof(ipv6_addrs.mask->dst))) {
+				NL_SET_ERR_MSG_MOD(extack, "unsupported offload: only an exact match IPv6 destination address is supported");
+				return -EOPNOTSUPP;
+			}
+		} else {
+			flow_rule_match_enc_ipv4_addrs(rule, &ipv4_addrs);
+			if (ipv4_addrs.mask->dst != cpu_to_be32(~0)) {
+				NL_SET_ERR_MSG_MOD(extack, "unsupported offload: only an exact match IPv4 destination address is supported");
+				return -EOPNOTSUPP;
+			}
+		}
+
 		if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_OPTS))
 			flow_rule_match_enc_opts(rule, &enc_op);
 
-
 		if (!flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_PORTS)) {
 			/* check if GRE, which has no enc_ports */
-			if (netif_is_gretap(netdev)) {
-				*tun_type = NFP_FL_TUNNEL_GRE;
-				key_layer |= NFP_FLOWER_LAYER_EXT_META;
-				key_size += sizeof(struct nfp_flower_ext_meta);
-				key_layer_two |= NFP_FLOWER_LAYER2_GRE;
-				key_size +=
-					sizeof(struct nfp_flower_ipv4_gre_tun);
+			if (!netif_is_gretap(netdev)) {
+				NL_SET_ERR_MSG_MOD(extack, "unsupported offload: an exact match on L4 destination port is required for non-GRE tunnels");
+				return -EOPNOTSUPP;
+			}
 
-				if (enc_op.key) {
-					NL_SET_ERR_MSG_MOD(extack, "unsupported offload: encap options not supported on GRE tunnels");
-					return -EOPNOTSUPP;
-				}
+			*tun_type = NFP_FL_TUNNEL_GRE;
+			key_layer |= NFP_FLOWER_LAYER_EXT_META;
+			key_size += sizeof(struct nfp_flower_ext_meta);
+			key_layer_two |= NFP_FLOWER_LAYER2_GRE;
+
+			if (ipv6_tun) {
+				key_layer_two |= NFP_FLOWER_LAYER2_TUN_IPV6;
+				key_size +=
+					sizeof(struct nfp_flower_ipv6_udp_tun);
 			} else {
-				NL_SET_ERR_MSG_MOD(extack, "unsupported offload: an exact match on L4 destination port is required for non-GRE tunnels");
+				key_size +=
+					sizeof(struct nfp_flower_ipv4_udp_tun);
+			}
+
+			if (enc_op.key) {
+				NL_SET_ERR_MSG_MOD(extack, "unsupported offload: encap options not supported on GRE tunnels");
 				return -EOPNOTSUPP;
 			}
 		} else {
@@ -323,7 +370,8 @@ nfp_flower_calculate_key_layers(struct nfp_app *app,
 							    &key_layer_two,
 							    &key_layer,
 							    &key_size, priv,
-							    tun_type, extack);
+							    tun_type, ipv6_tun,
+							    extack);
 			if (err)
 				return err;
 
-- 
cgit v1.2.3-59-g8ed1b


From cfa189934231a2c7922ad9e15561b5b01ea5e82e Mon Sep 17 00:00:00 2001
From: John Hurley <john.hurley@netronome.com>
Date: Tue, 17 Dec 2019 21:57:19 +0000
Subject: nfp: flower: offload list of IPv6 tunnel endpoint addresses

Fw requires a list of IPv6 addresses that are used as tunnel endpoints to
enable correct decap of tunneled packets.

Store a list of IPv6 endpoints used in rules with a ref counter to track
how many times it is in use. Offload the entire list any time a new IPv6
address is added or when an address is removed (ref count is 0).

Signed-off-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/netronome/nfp/flower/cmsg.h   |  1 +
 drivers/net/ethernet/netronome/nfp/flower/main.h   | 21 +++++
 drivers/net/ethernet/netronome/nfp/flower/match.c  | 24 ++++++
 .../net/ethernet/netronome/nfp/flower/offload.c    |  6 ++
 .../ethernet/netronome/nfp/flower/tunnel_conf.c    | 90 +++++++++++++++++++++-
 5 files changed, 141 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/netronome/nfp/flower/cmsg.h b/drivers/net/ethernet/netronome/nfp/flower/cmsg.h
index 8d12805baa0d..cbb94cf1d282 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/cmsg.h
+++ b/drivers/net/ethernet/netronome/nfp/flower/cmsg.h
@@ -568,6 +568,7 @@ enum nfp_flower_cmsg_type_port {
 	NFP_FLOWER_CMSG_TYPE_QOS_DEL =		19,
 	NFP_FLOWER_CMSG_TYPE_QOS_STATS =	20,
 	NFP_FLOWER_CMSG_TYPE_PRE_TUN_RULE =	21,
+	NFP_FLOWER_CMSG_TYPE_TUN_IPS_V6 =	22,
 	NFP_FLOWER_CMSG_TYPE_MAX =		32,
 };
 
diff --git a/drivers/net/ethernet/netronome/nfp/flower/main.h b/drivers/net/ethernet/netronome/nfp/flower/main.h
index 386d6297e0f1..bd288e236db7 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/main.h
+++ b/drivers/net/ethernet/netronome/nfp/flower/main.h
@@ -63,8 +63,10 @@ struct nfp_fl_stats_id {
  * struct nfp_fl_tunnel_offloads - priv data for tunnel offloads
  * @offloaded_macs:	Hashtable of the offloaded MAC addresses
  * @ipv4_off_list:	List of IPv4 addresses to offload
+ * @ipv6_off_list:	List of IPv6 addresses to offload
  * @neigh_off_list:	List of neighbour offloads
  * @ipv4_off_lock:	Lock for the IPv4 address list
+ * @ipv6_off_lock:	Lock for the IPv6 address list
  * @neigh_off_lock:	Lock for the neighbour address list
  * @mac_off_ids:	IDA to manage id assignment for offloaded MACs
  * @neigh_nb:		Notifier to monitor neighbour state
@@ -72,8 +74,10 @@ struct nfp_fl_stats_id {
 struct nfp_fl_tunnel_offloads {
 	struct rhashtable offloaded_macs;
 	struct list_head ipv4_off_list;
+	struct list_head ipv6_off_list;
 	struct list_head neigh_off_list;
 	struct mutex ipv4_off_lock;
+	struct mutex ipv6_off_lock;
 	spinlock_t neigh_off_lock;
 	struct ida mac_off_ids;
 	struct notifier_block neigh_nb;
@@ -274,12 +278,25 @@ struct nfp_fl_stats {
 	u64 used;
 };
 
+/**
+ * struct nfp_ipv6_addr_entry - cached IPv6 addresses
+ * @ipv6_addr:	IP address
+ * @ref_count:	number of rules currently using this IP
+ * @list:	list pointer
+ */
+struct nfp_ipv6_addr_entry {
+	struct in6_addr ipv6_addr;
+	int ref_count;
+	struct list_head list;
+};
+
 struct nfp_fl_payload {
 	struct nfp_fl_rule_metadata meta;
 	unsigned long tc_flower_cookie;
 	struct rhash_head fl_node;
 	struct rcu_head rcu;
 	__be32 nfp_tun_ipv4_addr;
+	struct nfp_ipv6_addr_entry *nfp_tun_ipv6;
 	struct net_device *ingress_dev;
 	char *unmasked_data;
 	char *mask_data;
@@ -397,6 +414,10 @@ int nfp_tunnel_mac_event_handler(struct nfp_app *app,
 				 unsigned long event, void *ptr);
 void nfp_tunnel_del_ipv4_off(struct nfp_app *app, __be32 ipv4);
 void nfp_tunnel_add_ipv4_off(struct nfp_app *app, __be32 ipv4);
+void
+nfp_tunnel_put_ipv6_off(struct nfp_app *app, struct nfp_ipv6_addr_entry *entry);
+struct nfp_ipv6_addr_entry *
+nfp_tunnel_add_ipv6_off(struct nfp_app *app, struct in6_addr *ipv6);
 void nfp_tunnel_request_route(struct nfp_app *app, struct sk_buff *skb);
 void nfp_tunnel_keep_alive(struct nfp_app *app, struct sk_buff *skb);
 void nfp_flower_lag_init(struct nfp_fl_lag *lag);
diff --git a/drivers/net/ethernet/netronome/nfp/flower/match.c b/drivers/net/ethernet/netronome/nfp/flower/match.c
index 2410ead05a06..546bc01d507d 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/match.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/match.c
@@ -494,10 +494,22 @@ int nfp_flower_compile_flow_match(struct nfp_app *app,
 
 	if (key_ls->key_layer_two & NFP_FLOWER_LAYER2_GRE) {
 		if (key_ls->key_layer_two & NFP_FLOWER_LAYER2_TUN_IPV6) {
+			struct nfp_flower_ipv6_gre_tun *gre_match;
+			struct nfp_ipv6_addr_entry *entry;
+			struct in6_addr *dst;
+
 			nfp_flower_compile_ipv6_gre_tun((void *)ext,
 							(void *)msk, rule);
+			gre_match = (struct nfp_flower_ipv6_gre_tun *)ext;
+			dst = &gre_match->ipv6.dst;
 			ext += sizeof(struct nfp_flower_ipv6_gre_tun);
 			msk += sizeof(struct nfp_flower_ipv6_gre_tun);
+
+			entry = nfp_tunnel_add_ipv6_off(app, dst);
+			if (!entry)
+				return -EOPNOTSUPP;
+
+			nfp_flow->nfp_tun_ipv6 = entry;
 		} else {
 			__be32 dst;
 
@@ -518,10 +530,22 @@ int nfp_flower_compile_flow_match(struct nfp_app *app,
 	if (key_ls->key_layer & NFP_FLOWER_LAYER_VXLAN ||
 	    key_ls->key_layer_two & NFP_FLOWER_LAYER2_GENEVE) {
 		if (key_ls->key_layer_two & NFP_FLOWER_LAYER2_TUN_IPV6) {
+			struct nfp_flower_ipv6_udp_tun *udp_match;
+			struct nfp_ipv6_addr_entry *entry;
+			struct in6_addr *dst;
+
 			nfp_flower_compile_ipv6_udp_tun((void *)ext,
 							(void *)msk, rule);
+			udp_match = (struct nfp_flower_ipv6_udp_tun *)ext;
+			dst = &udp_match->ipv6.dst;
 			ext += sizeof(struct nfp_flower_ipv6_udp_tun);
 			msk += sizeof(struct nfp_flower_ipv6_udp_tun);
+
+			entry = nfp_tunnel_add_ipv6_off(app, dst);
+			if (!entry)
+				return -EOPNOTSUPP;
+
+			nfp_flow->nfp_tun_ipv6 = entry;
 		} else {
 			__be32 dst;
 
diff --git a/drivers/net/ethernet/netronome/nfp/flower/offload.c b/drivers/net/ethernet/netronome/nfp/flower/offload.c
index 099735086fb2..83ada1b6ad88 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/offload.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/offload.c
@@ -539,6 +539,7 @@ nfp_flower_allocate_new(struct nfp_fl_key_ls *key_layer)
 		goto err_free_mask;
 
 	flow_pay->nfp_tun_ipv4_addr = 0;
+	flow_pay->nfp_tun_ipv6 = NULL;
 	flow_pay->meta.flags = 0;
 	INIT_LIST_HEAD(&flow_pay->linked_flows);
 	flow_pay->in_hw = false;
@@ -1243,6 +1244,8 @@ err_remove_rhash:
 err_release_metadata:
 	nfp_modify_flow_metadata(app, flow_pay);
 err_destroy_flow:
+	if (flow_pay->nfp_tun_ipv6)
+		nfp_tunnel_put_ipv6_off(app, flow_pay->nfp_tun_ipv6);
 	kfree(flow_pay->action_data);
 	kfree(flow_pay->mask_data);
 	kfree(flow_pay->unmasked_data);
@@ -1359,6 +1362,9 @@ nfp_flower_del_offload(struct nfp_app *app, struct net_device *netdev,
 	if (nfp_flow->nfp_tun_ipv4_addr)
 		nfp_tunnel_del_ipv4_off(app, nfp_flow->nfp_tun_ipv4_addr);
 
+	if (nfp_flow->nfp_tun_ipv6)
+		nfp_tunnel_put_ipv6_off(app, nfp_flow->nfp_tun_ipv6);
+
 	if (!nfp_flow->in_hw) {
 		err = 0;
 		goto err_free_merge_flow;
diff --git a/drivers/net/ethernet/netronome/nfp/flower/tunnel_conf.c b/drivers/net/ethernet/netronome/nfp/flower/tunnel_conf.c
index 2600ce476d6b..eddb52db6dce 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/tunnel_conf.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/tunnel_conf.c
@@ -116,6 +116,18 @@ struct nfp_ipv4_addr_entry {
 	struct list_head list;
 };
 
+#define NFP_FL_IPV6_ADDRS_MAX        4
+
+/**
+ * struct nfp_tun_ipv6_addr - set the IP address list on the NFP
+ * @count:	number of IPs populated in the array
+ * @ipv6_addr:	array of IPV6_ADDRS_MAX 128 bit IPv6 addresses
+ */
+struct nfp_tun_ipv6_addr {
+	__be32 count;
+	struct in6_addr ipv6_addr[NFP_FL_IPV6_ADDRS_MAX];
+};
+
 #define NFP_TUN_MAC_OFFLOAD_DEL_FLAG	0x2
 
 /**
@@ -502,6 +514,78 @@ void nfp_tunnel_del_ipv4_off(struct nfp_app *app, __be32 ipv4)
 	nfp_tun_write_ipv4_list(app);
 }
 
+static void nfp_tun_write_ipv6_list(struct nfp_app *app)
+{
+	struct nfp_flower_priv *priv = app->priv;
+	struct nfp_ipv6_addr_entry *entry;
+	struct nfp_tun_ipv6_addr payload;
+	int count = 0;
+
+	memset(&payload, 0, sizeof(struct nfp_tun_ipv6_addr));
+	mutex_lock(&priv->tun.ipv6_off_lock);
+	list_for_each_entry(entry, &priv->tun.ipv6_off_list, list) {
+		if (count >= NFP_FL_IPV6_ADDRS_MAX) {
+			nfp_flower_cmsg_warn(app, "Too many IPv6 tunnel endpoint addresses, some cannot be offloaded.\n");
+			break;
+		}
+		payload.ipv6_addr[count++] = entry->ipv6_addr;
+	}
+	mutex_unlock(&priv->tun.ipv6_off_lock);
+	payload.count = cpu_to_be32(count);
+
+	nfp_flower_xmit_tun_conf(app, NFP_FLOWER_CMSG_TYPE_TUN_IPS_V6,
+				 sizeof(struct nfp_tun_ipv6_addr),
+				 &payload, GFP_KERNEL);
+}
+
+struct nfp_ipv6_addr_entry *
+nfp_tunnel_add_ipv6_off(struct nfp_app *app, struct in6_addr *ipv6)
+{
+	struct nfp_flower_priv *priv = app->priv;
+	struct nfp_ipv6_addr_entry *entry;
+
+	mutex_lock(&priv->tun.ipv6_off_lock);
+	list_for_each_entry(entry, &priv->tun.ipv6_off_list, list)
+		if (!memcmp(&entry->ipv6_addr, ipv6, sizeof(*ipv6))) {
+			entry->ref_count++;
+			mutex_unlock(&priv->tun.ipv6_off_lock);
+			return entry;
+		}
+
+	entry = kmalloc(sizeof(*entry), GFP_KERNEL);
+	if (!entry) {
+		mutex_unlock(&priv->tun.ipv6_off_lock);
+		nfp_flower_cmsg_warn(app, "Mem error when offloading IP address.\n");
+		return NULL;
+	}
+	entry->ipv6_addr = *ipv6;
+	entry->ref_count = 1;
+	list_add_tail(&entry->list, &priv->tun.ipv6_off_list);
+	mutex_unlock(&priv->tun.ipv6_off_lock);
+
+	nfp_tun_write_ipv6_list(app);
+
+	return entry;
+}
+
+void
+nfp_tunnel_put_ipv6_off(struct nfp_app *app, struct nfp_ipv6_addr_entry *entry)
+{
+	struct nfp_flower_priv *priv = app->priv;
+	bool freed = false;
+
+	mutex_lock(&priv->tun.ipv6_off_lock);
+	if (!--entry->ref_count) {
+		list_del(&entry->list);
+		kfree(entry);
+		freed = true;
+	}
+	mutex_unlock(&priv->tun.ipv6_off_lock);
+
+	if (freed)
+		nfp_tun_write_ipv6_list(app);
+}
+
 static int
 __nfp_tunnel_offload_mac(struct nfp_app *app, u8 *mac, u16 idx, bool del)
 {
@@ -1013,9 +1097,11 @@ int nfp_tunnel_config_start(struct nfp_app *app)
 
 	ida_init(&priv->tun.mac_off_ids);
 
-	/* Initialise priv data for IPv4 offloading. */
+	/* Initialise priv data for IPv4/v6 offloading. */
 	mutex_init(&priv->tun.ipv4_off_lock);
 	INIT_LIST_HEAD(&priv->tun.ipv4_off_list);
+	mutex_init(&priv->tun.ipv6_off_lock);
+	INIT_LIST_HEAD(&priv->tun.ipv6_off_list);
 
 	/* Initialise priv data for neighbour offloading. */
 	spin_lock_init(&priv->tun.neigh_off_lock);
@@ -1050,6 +1136,8 @@ void nfp_tunnel_config_stop(struct nfp_app *app)
 		kfree(ip_entry);
 	}
 
+	mutex_destroy(&priv->tun.ipv6_off_lock);
+
 	/* Free any memory that may be occupied by the route list. */
 	list_for_each_safe(ptr, storage, &priv->tun.neigh_off_list) {
 		route_entry = list_entry(ptr, struct nfp_ipv4_route_entry,
-- 
cgit v1.2.3-59-g8ed1b


From 1922c9a4e273fd0934f82d9d80c69d53658a1c9d Mon Sep 17 00:00:00 2001
From: John Hurley <john.hurley@netronome.com>
Date: Tue, 17 Dec 2019 21:57:20 +0000
Subject: nfp: flower: modify pre-tunnel and set tunnel action for ipv6

The IPv4 set tunnel action allows the setting of tunnel metadata such as
the TTL and ToS values. The pre-tunnel action includes the destination IP
address and is used to calculate the next hop from from the neighbour
table.

Much of the IPv4 tunnel actions can be reused for IPv6 tunnels. Change the
names of associated functions and structs to remove the IPv4 identifier
and make minor modifcations to support IPv6 tunnel actions.

Ensure the pre-tunnel action contains the IPv6 address along with an
identifying flag when an IPv6 tunnel action is required.

Signed-off-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/netronome/nfp/flower/action.c | 65 ++++++++++++++++------
 drivers/net/ethernet/netronome/nfp/flower/cmsg.h   | 19 ++++---
 .../net/ethernet/netronome/nfp/flower/offload.c    |  8 +--
 3 files changed, 62 insertions(+), 30 deletions(-)

diff --git a/drivers/net/ethernet/netronome/nfp/flower/action.c b/drivers/net/ethernet/netronome/nfp/flower/action.c
index 1b019fdfcd97..c06600fb47ff 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/action.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/action.c
@@ -22,8 +22,9 @@
 #define NFP_FL_TUNNEL_CSUM			cpu_to_be16(0x01)
 #define NFP_FL_TUNNEL_KEY			cpu_to_be16(0x04)
 #define NFP_FL_TUNNEL_GENEVE_OPT		cpu_to_be16(0x0800)
-#define NFP_FL_SUPPORTED_TUNNEL_INFO_FLAGS	IP_TUNNEL_INFO_TX
-#define NFP_FL_SUPPORTED_IPV4_UDP_TUN_FLAGS	(NFP_FL_TUNNEL_CSUM | \
+#define NFP_FL_SUPPORTED_TUNNEL_INFO_FLAGS	(IP_TUNNEL_INFO_TX | \
+						 IP_TUNNEL_INFO_IPV6)
+#define NFP_FL_SUPPORTED_UDP_TUN_FLAGS		(NFP_FL_TUNNEL_CSUM | \
 						 NFP_FL_TUNNEL_KEY | \
 						 NFP_FL_TUNNEL_GENEVE_OPT)
 
@@ -394,19 +395,26 @@ nfp_fl_push_geneve_options(struct nfp_fl_payload *nfp_fl, int *list_len,
 }
 
 static int
-nfp_fl_set_ipv4_tun(struct nfp_app *app, struct nfp_fl_set_ipv4_tun *set_tun,
-		    const struct flow_action_entry *act,
-		    struct nfp_fl_pre_tunnel *pre_tun,
-		    enum nfp_flower_tun_type tun_type,
-		    struct net_device *netdev, struct netlink_ext_ack *extack)
+nfp_fl_set_tun(struct nfp_app *app, struct nfp_fl_set_tun *set_tun,
+	       const struct flow_action_entry *act,
+	       struct nfp_fl_pre_tunnel *pre_tun,
+	       enum nfp_flower_tun_type tun_type,
+	       struct net_device *netdev, struct netlink_ext_ack *extack)
 {
-	size_t act_size = sizeof(struct nfp_fl_set_ipv4_tun);
 	const struct ip_tunnel_info *ip_tun = act->tunnel;
+	bool ipv6 = ip_tunnel_info_af(ip_tun) == AF_INET6;
+	size_t act_size = sizeof(struct nfp_fl_set_tun);
 	struct nfp_flower_priv *priv = app->priv;
 	u32 tmp_set_ip_tun_type_index = 0;
 	/* Currently support one pre-tunnel so index is always 0. */
 	int pretun_idx = 0;
 
+	if (!IS_ENABLED(CONFIG_IPV6) && ipv6)
+		return -EOPNOTSUPP;
+
+	if (ipv6 && !(priv->flower_ext_feats & NFP_FL_FEATS_IPV6_TUN))
+		return -EOPNOTSUPP;
+
 	BUILD_BUG_ON(NFP_FL_TUNNEL_CSUM != TUNNEL_CSUM ||
 		     NFP_FL_TUNNEL_KEY	!= TUNNEL_KEY ||
 		     NFP_FL_TUNNEL_GENEVE_OPT != TUNNEL_GENEVE_OPT);
@@ -417,19 +425,35 @@ nfp_fl_set_ipv4_tun(struct nfp_app *app, struct nfp_fl_set_ipv4_tun *set_tun,
 		return -EOPNOTSUPP;
 	}
 
-	set_tun->head.jump_id = NFP_FL_ACTION_OPCODE_SET_IPV4_TUNNEL;
+	set_tun->head.jump_id = NFP_FL_ACTION_OPCODE_SET_TUNNEL;
 	set_tun->head.len_lw = act_size >> NFP_FL_LW_SIZ;
 
 	/* Set tunnel type and pre-tunnel index. */
 	tmp_set_ip_tun_type_index |=
-		FIELD_PREP(NFP_FL_IPV4_TUNNEL_TYPE, tun_type) |
-		FIELD_PREP(NFP_FL_IPV4_PRE_TUN_INDEX, pretun_idx);
+		FIELD_PREP(NFP_FL_TUNNEL_TYPE, tun_type) |
+		FIELD_PREP(NFP_FL_PRE_TUN_INDEX, pretun_idx);
 
 	set_tun->tun_type_index = cpu_to_be32(tmp_set_ip_tun_type_index);
 	set_tun->tun_id = ip_tun->key.tun_id;
 
 	if (ip_tun->key.ttl) {
 		set_tun->ttl = ip_tun->key.ttl;
+#ifdef CONFIG_IPV6
+	} else if (ipv6) {
+		struct net *net = dev_net(netdev);
+		struct flowi6 flow = {};
+		struct dst_entry *dst;
+
+		flow.daddr = ip_tun->key.u.ipv6.dst;
+		flow.flowi4_proto = IPPROTO_UDP;
+		dst = ipv6_stub->ipv6_dst_lookup_flow(net, NULL, &flow, NULL);
+		if (!IS_ERR(dst)) {
+			set_tun->ttl = ip6_dst_hoplimit(dst);
+			dst_release(dst);
+		} else {
+			set_tun->ttl = net->ipv6.devconf_all->hop_limit;
+		}
+#endif
 	} else {
 		struct net *net = dev_net(netdev);
 		struct flowi4 flow = {};
@@ -455,7 +479,7 @@ nfp_fl_set_ipv4_tun(struct nfp_app *app, struct nfp_fl_set_ipv4_tun *set_tun,
 	set_tun->tos = ip_tun->key.tos;
 
 	if (!(ip_tun->key.tun_flags & NFP_FL_TUNNEL_KEY) ||
-	    ip_tun->key.tun_flags & ~NFP_FL_SUPPORTED_IPV4_UDP_TUN_FLAGS) {
+	    ip_tun->key.tun_flags & ~NFP_FL_SUPPORTED_UDP_TUN_FLAGS) {
 		NL_SET_ERR_MSG_MOD(extack, "unsupported offload: loaded firmware does not support tunnel flag offload");
 		return -EOPNOTSUPP;
 	}
@@ -467,7 +491,12 @@ nfp_fl_set_ipv4_tun(struct nfp_app *app, struct nfp_fl_set_ipv4_tun *set_tun,
 	}
 
 	/* Complete pre_tunnel action. */
-	pre_tun->ipv4_dst = ip_tun->key.u.ipv4.dst;
+	if (ipv6) {
+		pre_tun->flags |= cpu_to_be16(NFP_FL_PRE_TUN_IPV6);
+		pre_tun->ipv6_dst = ip_tun->key.u.ipv6.dst;
+	} else {
+		pre_tun->ipv4_dst = ip_tun->key.u.ipv4.dst;
+	}
 
 	return 0;
 }
@@ -956,8 +985,8 @@ nfp_flower_loop_action(struct nfp_app *app, const struct flow_action_entry *act,
 		       struct nfp_flower_pedit_acts *set_act, bool *pkt_host,
 		       struct netlink_ext_ack *extack, int act_idx)
 {
-	struct nfp_fl_set_ipv4_tun *set_tun;
 	struct nfp_fl_pre_tunnel *pre_tun;
+	struct nfp_fl_set_tun *set_tun;
 	struct nfp_fl_push_vlan *psh_v;
 	struct nfp_fl_push_mpls *psh_m;
 	struct nfp_fl_pop_vlan *pop_v;
@@ -1032,7 +1061,7 @@ nfp_flower_loop_action(struct nfp_app *app, const struct flow_action_entry *act,
 		 * If none, the packet falls back before applying other actions.
 		 */
 		if (*a_len + sizeof(struct nfp_fl_pre_tunnel) +
-		    sizeof(struct nfp_fl_set_ipv4_tun) > NFP_FL_MAX_A_SIZ) {
+		    sizeof(struct nfp_fl_set_tun) > NFP_FL_MAX_A_SIZ) {
 			NL_SET_ERR_MSG_MOD(extack, "unsupported offload: maximum allowed action list size exceeded at tunnel encap");
 			return -EOPNOTSUPP;
 		}
@@ -1046,11 +1075,11 @@ nfp_flower_loop_action(struct nfp_app *app, const struct flow_action_entry *act,
 			return err;
 
 		set_tun = (void *)&nfp_fl->action_data[*a_len];
-		err = nfp_fl_set_ipv4_tun(app, set_tun, act, pre_tun,
-					  *tun_type, netdev, extack);
+		err = nfp_fl_set_tun(app, set_tun, act, pre_tun, *tun_type,
+				     netdev, extack);
 		if (err)
 			return err;
-		*a_len += sizeof(struct nfp_fl_set_ipv4_tun);
+		*a_len += sizeof(struct nfp_fl_set_tun);
 		}
 		break;
 	case FLOW_ACTION_TUNNEL_DECAP:
diff --git a/drivers/net/ethernet/netronome/nfp/flower/cmsg.h b/drivers/net/ethernet/netronome/nfp/flower/cmsg.h
index cbb94cf1d282..7537695c2334 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/cmsg.h
+++ b/drivers/net/ethernet/netronome/nfp/flower/cmsg.h
@@ -72,7 +72,7 @@
 #define NFP_FL_ACTION_OPCODE_POP_VLAN		2
 #define NFP_FL_ACTION_OPCODE_PUSH_MPLS		3
 #define NFP_FL_ACTION_OPCODE_POP_MPLS		4
-#define NFP_FL_ACTION_OPCODE_SET_IPV4_TUNNEL	6
+#define NFP_FL_ACTION_OPCODE_SET_TUNNEL		6
 #define NFP_FL_ACTION_OPCODE_SET_ETHERNET	7
 #define NFP_FL_ACTION_OPCODE_SET_MPLS		8
 #define NFP_FL_ACTION_OPCODE_SET_IPV4_ADDRS	9
@@ -101,8 +101,8 @@
 
 /* Tunnel ports */
 #define NFP_FL_PORT_TYPE_TUN		0x50000000
-#define NFP_FL_IPV4_TUNNEL_TYPE		GENMASK(7, 4)
-#define NFP_FL_IPV4_PRE_TUN_INDEX	GENMASK(2, 0)
+#define NFP_FL_TUNNEL_TYPE		GENMASK(7, 4)
+#define NFP_FL_PRE_TUN_INDEX		GENMASK(2, 0)
 
 #define NFP_FLOWER_WORKQ_MAX_SKBS	30000
 
@@ -208,13 +208,16 @@ struct nfp_fl_pre_lag {
 
 struct nfp_fl_pre_tunnel {
 	struct nfp_fl_act_head head;
-	__be16 reserved;
-	__be32 ipv4_dst;
-	/* reserved for use with IPv6 addresses */
-	__be32 extra[3];
+	__be16 flags;
+	union {
+		__be32 ipv4_dst;
+		struct in6_addr ipv6_dst;
+	};
 };
 
-struct nfp_fl_set_ipv4_tun {
+#define NFP_FL_PRE_TUN_IPV6	BIT(0)
+
+struct nfp_fl_set_tun {
 	struct nfp_fl_act_head head;
 	__be16 reserved;
 	__be64 tun_id __packed;
diff --git a/drivers/net/ethernet/netronome/nfp/flower/offload.c b/drivers/net/ethernet/netronome/nfp/flower/offload.c
index 83ada1b6ad88..d51210549299 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/offload.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/offload.c
@@ -591,7 +591,7 @@ nfp_flower_update_merge_with_actions(struct nfp_fl_payload *flow,
 		case NFP_FL_ACTION_OPCODE_POP_VLAN:
 			merge->tci = cpu_to_be16(0);
 			break;
-		case NFP_FL_ACTION_OPCODE_SET_IPV4_TUNNEL:
+		case NFP_FL_ACTION_OPCODE_SET_TUNNEL:
 			/* New tunnel header means l2 to l4 can be matched. */
 			eth_broadcast_addr(&merge->l2.mac_dst[0]);
 			eth_broadcast_addr(&merge->l2.mac_src[0]);
@@ -814,15 +814,15 @@ nfp_fl_verify_post_tun_acts(char *acts, int len, struct nfp_fl_push_vlan **vlan)
 static int
 nfp_fl_push_vlan_after_tun(char *acts, int len, struct nfp_fl_push_vlan *vlan)
 {
-	struct nfp_fl_set_ipv4_tun *tun;
+	struct nfp_fl_set_tun *tun;
 	struct nfp_fl_act_head *a;
 	unsigned int act_off = 0;
 
 	while (act_off < len) {
 		a = (struct nfp_fl_act_head *)&acts[act_off];
 
-		if (a->jump_id == NFP_FL_ACTION_OPCODE_SET_IPV4_TUNNEL) {
-			tun = (struct nfp_fl_set_ipv4_tun *)a;
+		if (a->jump_id == NFP_FL_ACTION_OPCODE_SET_TUNNEL) {
+			tun = (struct nfp_fl_set_tun *)a;
 			tun->outer_vlan_tpid = vlan->vlan_tpid;
 			tun->outer_vlan_tci = vlan->vlan_tci;
 
-- 
cgit v1.2.3-59-g8ed1b


From 50b1c86ab0a07509fd65c92acacf281fb72a285b Mon Sep 17 00:00:00 2001
From: John Hurley <john.hurley@netronome.com>
Date: Tue, 17 Dec 2019 21:57:21 +0000
Subject: nfp: flower: handle ipv6 tunnel no neigh request

When fw does not know the next hop for an IPv6 tunnel, it sends a request
to the driver.

Handle this request by doing a route lookup on the IPv6 address and
offloading the next hop to the fw neighbour table.

Similar functions already exist to handle IPv4 no neighbour requests. To
avoid confusion, append these functions with the _ipv4 tag. There is no
change in functionality with this.

Signed-off-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/netronome/nfp/flower/cmsg.c   |   8 +-
 drivers/net/ethernet/netronome/nfp/flower/cmsg.h   |   2 +
 drivers/net/ethernet/netronome/nfp/flower/main.h   |   3 +-
 .../ethernet/netronome/nfp/flower/tunnel_conf.c    | 111 ++++++++++++++++++++-
 4 files changed, 116 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ethernet/netronome/nfp/flower/cmsg.c b/drivers/net/ethernet/netronome/nfp/flower/cmsg.c
index 05981b54eaab..00b904ea4ebc 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/cmsg.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/cmsg.c
@@ -270,7 +270,10 @@ nfp_flower_cmsg_process_one_rx(struct nfp_app *app, struct sk_buff *skb)
 		}
 		goto err_default;
 	case NFP_FLOWER_CMSG_TYPE_NO_NEIGH:
-		nfp_tunnel_request_route(app, skb);
+		nfp_tunnel_request_route_v4(app, skb);
+		break;
+	case NFP_FLOWER_CMSG_TYPE_NO_NEIGH_V6:
+		nfp_tunnel_request_route_v6(app, skb);
 		break;
 	case NFP_FLOWER_CMSG_TYPE_ACTIVE_TUNS:
 		nfp_tunnel_keep_alive(app, skb);
@@ -361,7 +364,8 @@ void nfp_flower_cmsg_rx(struct nfp_app *app, struct sk_buff *skb)
 		   nfp_flower_process_mtu_ack(app, skb)) {
 		/* Handle MTU acks outside wq to prevent RTNL conflict. */
 		dev_consume_skb_any(skb);
-	} else if (cmsg_hdr->type == NFP_FLOWER_CMSG_TYPE_TUN_NEIGH) {
+	} else if (cmsg_hdr->type == NFP_FLOWER_CMSG_TYPE_TUN_NEIGH ||
+		   cmsg_hdr->type == NFP_FLOWER_CMSG_TYPE_TUN_NEIGH_V6) {
 		/* Acks from the NFP that the route is added - ignore. */
 		dev_consume_skb_any(skb);
 	} else if (cmsg_hdr->type == NFP_FLOWER_CMSG_TYPE_PORT_REIFY) {
diff --git a/drivers/net/ethernet/netronome/nfp/flower/cmsg.h b/drivers/net/ethernet/netronome/nfp/flower/cmsg.h
index 7537695c2334..5a8ad5814060 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/cmsg.h
+++ b/drivers/net/ethernet/netronome/nfp/flower/cmsg.h
@@ -572,6 +572,8 @@ enum nfp_flower_cmsg_type_port {
 	NFP_FLOWER_CMSG_TYPE_QOS_STATS =	20,
 	NFP_FLOWER_CMSG_TYPE_PRE_TUN_RULE =	21,
 	NFP_FLOWER_CMSG_TYPE_TUN_IPS_V6 =	22,
+	NFP_FLOWER_CMSG_TYPE_NO_NEIGH_V6 =	23,
+	NFP_FLOWER_CMSG_TYPE_TUN_NEIGH_V6 =	24,
 	NFP_FLOWER_CMSG_TYPE_MAX =		32,
 };
 
diff --git a/drivers/net/ethernet/netronome/nfp/flower/main.h b/drivers/net/ethernet/netronome/nfp/flower/main.h
index bd288e236db7..01e1886e9d85 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/main.h
+++ b/drivers/net/ethernet/netronome/nfp/flower/main.h
@@ -418,7 +418,8 @@ void
 nfp_tunnel_put_ipv6_off(struct nfp_app *app, struct nfp_ipv6_addr_entry *entry);
 struct nfp_ipv6_addr_entry *
 nfp_tunnel_add_ipv6_off(struct nfp_app *app, struct in6_addr *ipv6);
-void nfp_tunnel_request_route(struct nfp_app *app, struct sk_buff *skb);
+void nfp_tunnel_request_route_v4(struct nfp_app *app, struct sk_buff *skb);
+void nfp_tunnel_request_route_v6(struct nfp_app *app, struct sk_buff *skb);
 void nfp_tunnel_keep_alive(struct nfp_app *app, struct sk_buff *skb);
 void nfp_flower_lag_init(struct nfp_fl_lag *lag);
 void nfp_flower_lag_cleanup(struct nfp_fl_lag *lag);
diff --git a/drivers/net/ethernet/netronome/nfp/flower/tunnel_conf.c b/drivers/net/ethernet/netronome/nfp/flower/tunnel_conf.c
index eddb52db6dce..09807b31b830 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/tunnel_conf.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/tunnel_conf.c
@@ -70,6 +70,22 @@ struct nfp_tun_neigh {
 	__be32 port_id;
 };
 
+/**
+ * struct nfp_tun_neigh_v6 - neighbour/route entry on the NFP
+ * @dst_ipv6:	destination IPv6 address
+ * @src_ipv6:	source IPv6 address
+ * @dst_addr:	destination MAC address
+ * @src_addr:	source MAC address
+ * @port_id:	NFP port to output packet on - associated with source IPv6
+ */
+struct nfp_tun_neigh_v6 {
+	struct in6_addr dst_ipv6;
+	struct in6_addr src_ipv6;
+	u8 dst_addr[ETH_ALEN];
+	u8 src_addr[ETH_ALEN];
+	__be32 port_id;
+};
+
 /**
  * struct nfp_tun_req_route_ipv4 - NFP requests a route/neighbour lookup
  * @ingress_port:	ingress port of packet that signalled request
@@ -82,6 +98,16 @@ struct nfp_tun_req_route_ipv4 {
 	__be32 reserved[2];
 };
 
+/**
+ * struct nfp_tun_req_route_ipv6 - NFP requests an IPv6 route/neighbour lookup
+ * @ingress_port:	ingress port of packet that signalled request
+ * @ipv6_addr:		destination ipv6 address for route
+ */
+struct nfp_tun_req_route_ipv6 {
+	__be32 ingress_port;
+	struct in6_addr ipv6_addr;
+};
+
 /**
  * struct nfp_ipv4_route_entry - routes that are offloaded to the NFP
  * @ipv4_addr:	destination of route
@@ -299,8 +325,8 @@ static void nfp_tun_del_route_from_cache(struct nfp_app *app, __be32 ipv4_addr)
 }
 
 static void
-nfp_tun_write_neigh(struct net_device *netdev, struct nfp_app *app,
-		    struct flowi4 *flow, struct neighbour *neigh, gfp_t flag)
+nfp_tun_write_neigh_v4(struct net_device *netdev, struct nfp_app *app,
+		       struct flowi4 *flow, struct neighbour *neigh, gfp_t flag)
 {
 	struct nfp_tun_neigh payload;
 	u32 port_id;
@@ -334,6 +360,39 @@ send_msg:
 				 (unsigned char *)&payload, flag);
 }
 
+static void
+nfp_tun_write_neigh_v6(struct net_device *netdev, struct nfp_app *app,
+		       struct flowi6 *flow, struct neighbour *neigh, gfp_t flag)
+{
+	struct nfp_tun_neigh_v6 payload;
+	u32 port_id;
+
+	port_id = nfp_flower_get_port_id_from_netdev(app, netdev);
+	if (!port_id)
+		return;
+
+	memset(&payload, 0, sizeof(struct nfp_tun_neigh_v6));
+	payload.dst_ipv6 = flow->daddr;
+
+	/* If entry has expired send dst IP with all other fields 0. */
+	if (!(neigh->nud_state & NUD_VALID) || neigh->dead) {
+		/* Trigger probe to verify invalid neighbour state. */
+		neigh_event_send(neigh, NULL);
+		goto send_msg;
+	}
+
+	/* Have a valid neighbour so populate rest of entry. */
+	payload.src_ipv6 = flow->saddr;
+	ether_addr_copy(payload.src_addr, netdev->dev_addr);
+	neigh_ha_snapshot(payload.dst_addr, neigh, netdev);
+	payload.port_id = cpu_to_be32(port_id);
+
+send_msg:
+	nfp_flower_xmit_tun_conf(app, NFP_FLOWER_CMSG_TYPE_TUN_NEIGH_V6,
+				 sizeof(struct nfp_tun_neigh_v6),
+				 (unsigned char *)&payload, flag);
+}
+
 static int
 nfp_tun_neigh_event_handler(struct notifier_block *nb, unsigned long event,
 			    void *ptr)
@@ -384,12 +443,12 @@ nfp_tun_neigh_event_handler(struct notifier_block *nb, unsigned long event,
 #endif
 
 	flow.flowi4_proto = IPPROTO_UDP;
-	nfp_tun_write_neigh(n->dev, app, &flow, n, GFP_ATOMIC);
+	nfp_tun_write_neigh_v4(n->dev, app, &flow, n, GFP_ATOMIC);
 
 	return NOTIFY_OK;
 }
 
-void nfp_tunnel_request_route(struct nfp_app *app, struct sk_buff *skb)
+void nfp_tunnel_request_route_v4(struct nfp_app *app, struct sk_buff *skb)
 {
 	struct nfp_tun_req_route_ipv4 *payload;
 	struct net_device *netdev;
@@ -423,7 +482,7 @@ void nfp_tunnel_request_route(struct nfp_app *app, struct sk_buff *skb)
 	ip_rt_put(rt);
 	if (!n)
 		goto fail_rcu_unlock;
-	nfp_tun_write_neigh(n->dev, app, &flow, n, GFP_ATOMIC);
+	nfp_tun_write_neigh_v4(n->dev, app, &flow, n, GFP_ATOMIC);
 	neigh_release(n);
 	rcu_read_unlock();
 	return;
@@ -433,6 +492,48 @@ fail_rcu_unlock:
 	nfp_flower_cmsg_warn(app, "Requested route not found.\n");
 }
 
+void nfp_tunnel_request_route_v6(struct nfp_app *app, struct sk_buff *skb)
+{
+	struct nfp_tun_req_route_ipv6 *payload;
+	struct net_device *netdev;
+	struct flowi6 flow = {};
+	struct dst_entry *dst;
+	struct neighbour *n;
+
+	payload = nfp_flower_cmsg_get_data(skb);
+
+	rcu_read_lock();
+	netdev = nfp_app_dev_get(app, be32_to_cpu(payload->ingress_port), NULL);
+	if (!netdev)
+		goto fail_rcu_unlock;
+
+	flow.daddr = payload->ipv6_addr;
+	flow.flowi6_proto = IPPROTO_UDP;
+
+#if IS_ENABLED(CONFIG_INET) && IS_ENABLED(CONFIG_IPV6)
+	dst = ipv6_stub->ipv6_dst_lookup_flow(dev_net(netdev), NULL, &flow,
+					      NULL);
+	if (IS_ERR(dst))
+		goto fail_rcu_unlock;
+#else
+	goto fail_rcu_unlock;
+#endif
+
+	n = dst_neigh_lookup(dst, &flow.daddr);
+	dst_release(dst);
+	if (!n)
+		goto fail_rcu_unlock;
+
+	nfp_tun_write_neigh_v6(n->dev, app, &flow, n, GFP_ATOMIC);
+	neigh_release(n);
+	rcu_read_unlock();
+	return;
+
+fail_rcu_unlock:
+	rcu_read_unlock();
+	nfp_flower_cmsg_warn(app, "Requested IPv6 route not found.\n");
+}
+
 static void nfp_tun_write_ipv4_list(struct nfp_app *app)
 {
 	struct nfp_flower_priv *priv = app->priv;
-- 
cgit v1.2.3-59-g8ed1b


From 6c463a059f626bc2115b17bee97f014f59dd844a Mon Sep 17 00:00:00 2001
From: John Hurley <john.hurley@netronome.com>
Date: Tue, 17 Dec 2019 21:57:22 +0000
Subject: nfp: flower: handle notifiers for ipv6 route changes

A notifier is used to track route changes in the kernel. If a change is
made to a route that is offloaded to fw then an update is sent to the NIC.
The driver tracks all routes that are offloaded to determine if a kernel
change is of interest.

Extend the notifier to track IPv6 route changes and create a new list that
stores offloaded IPv6 routes. Modify the IPv4 route helper functions to
accept varying address lengths. This way, the same core functions can be
used to handle IPv4 and IPv6.

Signed-off-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/netronome/nfp/flower/main.h   |  12 +-
 .../ethernet/netronome/nfp/flower/tunnel_conf.c    | 237 +++++++++++++++------
 2 files changed, 181 insertions(+), 68 deletions(-)

diff --git a/drivers/net/ethernet/netronome/nfp/flower/main.h b/drivers/net/ethernet/netronome/nfp/flower/main.h
index 01e1886e9d85..7a4603261026 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/main.h
+++ b/drivers/net/ethernet/netronome/nfp/flower/main.h
@@ -64,10 +64,12 @@ struct nfp_fl_stats_id {
  * @offloaded_macs:	Hashtable of the offloaded MAC addresses
  * @ipv4_off_list:	List of IPv4 addresses to offload
  * @ipv6_off_list:	List of IPv6 addresses to offload
- * @neigh_off_list:	List of neighbour offloads
+ * @neigh_off_list_v4:	List of IPv4 neighbour offloads
+ * @neigh_off_list_v6:	List of IPv6 neighbour offloads
  * @ipv4_off_lock:	Lock for the IPv4 address list
  * @ipv6_off_lock:	Lock for the IPv6 address list
- * @neigh_off_lock:	Lock for the neighbour address list
+ * @neigh_off_lock_v4:	Lock for the IPv4 neighbour address list
+ * @neigh_off_lock_v6:	Lock for the IPv6 neighbour address list
  * @mac_off_ids:	IDA to manage id assignment for offloaded MACs
  * @neigh_nb:		Notifier to monitor neighbour state
  */
@@ -75,10 +77,12 @@ struct nfp_fl_tunnel_offloads {
 	struct rhashtable offloaded_macs;
 	struct list_head ipv4_off_list;
 	struct list_head ipv6_off_list;
-	struct list_head neigh_off_list;
+	struct list_head neigh_off_list_v4;
+	struct list_head neigh_off_list_v6;
 	struct mutex ipv4_off_lock;
 	struct mutex ipv6_off_lock;
-	spinlock_t neigh_off_lock;
+	spinlock_t neigh_off_lock_v4;
+	spinlock_t neigh_off_lock_v6;
 	struct ida mac_off_ids;
 	struct notifier_block neigh_nb;
 };
diff --git a/drivers/net/ethernet/netronome/nfp/flower/tunnel_conf.c b/drivers/net/ethernet/netronome/nfp/flower/tunnel_conf.c
index 09807b31b830..b179b63c8456 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/tunnel_conf.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/tunnel_conf.c
@@ -109,13 +109,13 @@ struct nfp_tun_req_route_ipv6 {
 };
 
 /**
- * struct nfp_ipv4_route_entry - routes that are offloaded to the NFP
- * @ipv4_addr:	destination of route
+ * struct nfp_offloaded_route - routes that are offloaded to the NFP
  * @list:	list pointer
+ * @ip_add:	destination of route - can be IPv4 or IPv6
  */
-struct nfp_ipv4_route_entry {
-	__be32 ipv4_addr;
+struct nfp_offloaded_route {
 	struct list_head list;
+	u8 ip_add[];
 };
 
 #define NFP_FL_IPV4_ADDRS_MAX        32
@@ -262,66 +262,121 @@ nfp_flower_xmit_tun_conf(struct nfp_app *app, u8 mtype, u16 plen, void *pdata,
 	return 0;
 }
 
-static bool nfp_tun_has_route(struct nfp_app *app, __be32 ipv4_addr)
+static bool
+__nfp_tun_has_route(struct list_head *route_list, spinlock_t *list_lock,
+		    void *add, int add_len)
 {
-	struct nfp_flower_priv *priv = app->priv;
-	struct nfp_ipv4_route_entry *entry;
-	struct list_head *ptr, *storage;
+	struct nfp_offloaded_route *entry;
 
-	spin_lock_bh(&priv->tun.neigh_off_lock);
-	list_for_each_safe(ptr, storage, &priv->tun.neigh_off_list) {
-		entry = list_entry(ptr, struct nfp_ipv4_route_entry, list);
-		if (entry->ipv4_addr == ipv4_addr) {
-			spin_unlock_bh(&priv->tun.neigh_off_lock);
+	spin_lock_bh(list_lock);
+	list_for_each_entry(entry, route_list, list)
+		if (!memcmp(entry->ip_add, add, add_len)) {
+			spin_unlock_bh(list_lock);
 			return true;
 		}
-	}
-	spin_unlock_bh(&priv->tun.neigh_off_lock);
+	spin_unlock_bh(list_lock);
 	return false;
 }
 
-static void nfp_tun_add_route_to_cache(struct nfp_app *app, __be32 ipv4_addr)
+static int
+__nfp_tun_add_route_to_cache(struct list_head *route_list,
+			     spinlock_t *list_lock, void *add, int add_len)
 {
-	struct nfp_flower_priv *priv = app->priv;
-	struct nfp_ipv4_route_entry *entry;
-	struct list_head *ptr, *storage;
+	struct nfp_offloaded_route *entry;
 
-	spin_lock_bh(&priv->tun.neigh_off_lock);
-	list_for_each_safe(ptr, storage, &priv->tun.neigh_off_list) {
-		entry = list_entry(ptr, struct nfp_ipv4_route_entry, list);
-		if (entry->ipv4_addr == ipv4_addr) {
-			spin_unlock_bh(&priv->tun.neigh_off_lock);
-			return;
+	spin_lock_bh(list_lock);
+	list_for_each_entry(entry, route_list, list)
+		if (!memcmp(entry->ip_add, add, add_len)) {
+			spin_unlock_bh(list_lock);
+			return 0;
 		}
-	}
-	entry = kmalloc(sizeof(*entry), GFP_ATOMIC);
+
+	entry = kmalloc(sizeof(*entry) + add_len, GFP_ATOMIC);
 	if (!entry) {
-		spin_unlock_bh(&priv->tun.neigh_off_lock);
-		nfp_flower_cmsg_warn(app, "Mem error when storing new route.\n");
-		return;
+		spin_unlock_bh(list_lock);
+		return -ENOMEM;
 	}
 
-	entry->ipv4_addr = ipv4_addr;
-	list_add_tail(&entry->list, &priv->tun.neigh_off_list);
-	spin_unlock_bh(&priv->tun.neigh_off_lock);
+	memcpy(entry->ip_add, add, add_len);
+	list_add_tail(&entry->list, route_list);
+	spin_unlock_bh(list_lock);
+
+	return 0;
 }
 
-static void nfp_tun_del_route_from_cache(struct nfp_app *app, __be32 ipv4_addr)
+static void
+__nfp_tun_del_route_from_cache(struct list_head *route_list,
+			       spinlock_t *list_lock, void *add, int add_len)
 {
-	struct nfp_flower_priv *priv = app->priv;
-	struct nfp_ipv4_route_entry *entry;
-	struct list_head *ptr, *storage;
+	struct nfp_offloaded_route *entry;
 
-	spin_lock_bh(&priv->tun.neigh_off_lock);
-	list_for_each_safe(ptr, storage, &priv->tun.neigh_off_list) {
-		entry = list_entry(ptr, struct nfp_ipv4_route_entry, list);
-		if (entry->ipv4_addr == ipv4_addr) {
+	spin_lock_bh(list_lock);
+	list_for_each_entry(entry, route_list, list)
+		if (!memcmp(entry->ip_add, add, add_len)) {
 			list_del(&entry->list);
 			kfree(entry);
 			break;
 		}
-	}
-	spin_unlock_bh(&priv->tun.neigh_off_lock);
+	spin_unlock_bh(list_lock);
+}
+
+static bool nfp_tun_has_route_v4(struct nfp_app *app, __be32 *ipv4_addr)
+{
+	struct nfp_flower_priv *priv = app->priv;
+
+	return __nfp_tun_has_route(&priv->tun.neigh_off_list_v4,
+				   &priv->tun.neigh_off_lock_v4, ipv4_addr,
+				   sizeof(*ipv4_addr));
+}
+
+static bool
+nfp_tun_has_route_v6(struct nfp_app *app, struct in6_addr *ipv6_addr)
+{
+	struct nfp_flower_priv *priv = app->priv;
+
+	return __nfp_tun_has_route(&priv->tun.neigh_off_list_v6,
+				   &priv->tun.neigh_off_lock_v6, ipv6_addr,
+				   sizeof(*ipv6_addr));
+}
+
+static void
+nfp_tun_add_route_to_cache_v4(struct nfp_app *app, __be32 *ipv4_addr)
+{
+	struct nfp_flower_priv *priv = app->priv;
+
+	__nfp_tun_add_route_to_cache(&priv->tun.neigh_off_list_v4,
+				     &priv->tun.neigh_off_lock_v4, ipv4_addr,
+				     sizeof(*ipv4_addr));
+}
+
+static void
+nfp_tun_add_route_to_cache_v6(struct nfp_app *app, struct in6_addr *ipv6_addr)
+{
+	struct nfp_flower_priv *priv = app->priv;
+
+	__nfp_tun_add_route_to_cache(&priv->tun.neigh_off_list_v6,
+				     &priv->tun.neigh_off_lock_v6, ipv6_addr,
+				     sizeof(*ipv6_addr));
+}
+
+static void
+nfp_tun_del_route_from_cache_v4(struct nfp_app *app, __be32 *ipv4_addr)
+{
+	struct nfp_flower_priv *priv = app->priv;
+
+	__nfp_tun_del_route_from_cache(&priv->tun.neigh_off_list_v4,
+				       &priv->tun.neigh_off_lock_v4, ipv4_addr,
+				       sizeof(*ipv4_addr));
+}
+
+static void
+nfp_tun_del_route_from_cache_v6(struct nfp_app *app, struct in6_addr *ipv6_addr)
+{
+	struct nfp_flower_priv *priv = app->priv;
+
+	__nfp_tun_del_route_from_cache(&priv->tun.neigh_off_list_v6,
+				       &priv->tun.neigh_off_lock_v6, ipv6_addr,
+				       sizeof(*ipv6_addr));
 }
 
 static void
@@ -340,7 +395,7 @@ nfp_tun_write_neigh_v4(struct net_device *netdev, struct nfp_app *app,
 
 	/* If entry has expired send dst IP with all other fields 0. */
 	if (!(neigh->nud_state & NUD_VALID) || neigh->dead) {
-		nfp_tun_del_route_from_cache(app, payload.dst_ipv4);
+		nfp_tun_del_route_from_cache_v4(app, &payload.dst_ipv4);
 		/* Trigger ARP to verify invalid neighbour state. */
 		neigh_event_send(neigh, NULL);
 		goto send_msg;
@@ -352,7 +407,7 @@ nfp_tun_write_neigh_v4(struct net_device *netdev, struct nfp_app *app,
 	neigh_ha_snapshot(payload.dst_addr, neigh, netdev);
 	payload.port_id = cpu_to_be32(port_id);
 	/* Add destination of new route to NFP cache. */
-	nfp_tun_add_route_to_cache(app, payload.dst_ipv4);
+	nfp_tun_add_route_to_cache_v4(app, &payload.dst_ipv4);
 
 send_msg:
 	nfp_flower_xmit_tun_conf(app, NFP_FLOWER_CMSG_TYPE_TUN_NEIGH,
@@ -376,6 +431,7 @@ nfp_tun_write_neigh_v6(struct net_device *netdev, struct nfp_app *app,
 
 	/* If entry has expired send dst IP with all other fields 0. */
 	if (!(neigh->nud_state & NUD_VALID) || neigh->dead) {
+		nfp_tun_del_route_from_cache_v6(app, &payload.dst_ipv6);
 		/* Trigger probe to verify invalid neighbour state. */
 		neigh_event_send(neigh, NULL);
 		goto send_msg;
@@ -386,6 +442,8 @@ nfp_tun_write_neigh_v6(struct net_device *netdev, struct nfp_app *app,
 	ether_addr_copy(payload.src_addr, netdev->dev_addr);
 	neigh_ha_snapshot(payload.dst_addr, neigh, netdev);
 	payload.port_id = cpu_to_be32(port_id);
+	/* Add destination of new route to NFP cache. */
+	nfp_tun_add_route_to_cache_v6(app, &payload.dst_ipv6);
 
 send_msg:
 	nfp_flower_xmit_tun_conf(app, NFP_FLOWER_CMSG_TYPE_TUN_NEIGH_V6,
@@ -399,10 +457,12 @@ nfp_tun_neigh_event_handler(struct notifier_block *nb, unsigned long event,
 {
 	struct nfp_flower_priv *app_priv;
 	struct netevent_redirect *redir;
-	struct flowi4 flow = {};
+	struct flowi4 flow4 = {};
+	struct flowi6 flow6 = {};
 	struct neighbour *n;
 	struct nfp_app *app;
 	struct rtable *rt;
+	bool ipv6 = false;
 	int err;
 
 	switch (event) {
@@ -417,7 +477,13 @@ nfp_tun_neigh_event_handler(struct notifier_block *nb, unsigned long event,
 		return NOTIFY_DONE;
 	}
 
-	flow.daddr = *(__be32 *)n->primary_key;
+	if (n->tbl->family == AF_INET6)
+		ipv6 = true;
+
+	if (ipv6)
+		flow6.daddr = *(struct in6_addr *)n->primary_key;
+	else
+		flow4.daddr = *(__be32 *)n->primary_key;
 
 	app_priv = container_of(nb, struct nfp_flower_priv, tun.neigh_nb);
 	app = app_priv->app;
@@ -427,23 +493,41 @@ nfp_tun_neigh_event_handler(struct notifier_block *nb, unsigned long event,
 		return NOTIFY_DONE;
 
 	/* Only concerned with changes to routes already added to NFP. */
-	if (!nfp_tun_has_route(app, flow.daddr))
+	if ((ipv6 && !nfp_tun_has_route_v6(app, &flow6.daddr)) ||
+	    (!ipv6 && !nfp_tun_has_route_v4(app, &flow4.daddr)))
 		return NOTIFY_DONE;
 
 #if IS_ENABLED(CONFIG_INET)
-	/* Do a route lookup to populate flow data. */
-	rt = ip_route_output_key(dev_net(n->dev), &flow);
-	err = PTR_ERR_OR_ZERO(rt);
-	if (err)
+	if (ipv6) {
+#if IS_ENABLED(CONFIG_IPV6)
+		struct dst_entry *dst;
+
+		dst = ipv6_stub->ipv6_dst_lookup_flow(dev_net(n->dev), NULL,
+						      &flow6, NULL);
+		if (IS_ERR(dst))
+			return NOTIFY_DONE;
+
+		dst_release(dst);
+		flow6.flowi6_proto = IPPROTO_UDP;
+		nfp_tun_write_neigh_v6(n->dev, app, &flow6, n, GFP_ATOMIC);
+#else
 		return NOTIFY_DONE;
+#endif /* CONFIG_IPV6 */
+	} else {
+		/* Do a route lookup to populate flow data. */
+		rt = ip_route_output_key(dev_net(n->dev), &flow4);
+		err = PTR_ERR_OR_ZERO(rt);
+		if (err)
+			return NOTIFY_DONE;
 
-	ip_rt_put(rt);
+		ip_rt_put(rt);
+
+		flow4.flowi4_proto = IPPROTO_UDP;
+		nfp_tun_write_neigh_v4(n->dev, app, &flow4, n, GFP_ATOMIC);
+	}
 #else
 	return NOTIFY_DONE;
-#endif
-
-	flow.flowi4_proto = IPPROTO_UDP;
-	nfp_tun_write_neigh_v4(n->dev, app, &flow, n, GFP_ATOMIC);
+#endif /* CONFIG_INET */
 
 	return NOTIFY_OK;
 }
@@ -1205,8 +1289,10 @@ int nfp_tunnel_config_start(struct nfp_app *app)
 	INIT_LIST_HEAD(&priv->tun.ipv6_off_list);
 
 	/* Initialise priv data for neighbour offloading. */
-	spin_lock_init(&priv->tun.neigh_off_lock);
-	INIT_LIST_HEAD(&priv->tun.neigh_off_list);
+	spin_lock_init(&priv->tun.neigh_off_lock_v4);
+	INIT_LIST_HEAD(&priv->tun.neigh_off_list_v4);
+	spin_lock_init(&priv->tun.neigh_off_lock_v6);
+	INIT_LIST_HEAD(&priv->tun.neigh_off_list_v6);
 	priv->tun.neigh_nb.notifier_call = nfp_tun_neigh_event_handler;
 
 	err = register_netevent_notifier(&priv->tun.neigh_nb);
@@ -1221,9 +1307,11 @@ int nfp_tunnel_config_start(struct nfp_app *app)
 
 void nfp_tunnel_config_stop(struct nfp_app *app)
 {
+	struct nfp_offloaded_route *route_entry, *temp;
 	struct nfp_flower_priv *priv = app->priv;
-	struct nfp_ipv4_route_entry *route_entry;
 	struct nfp_ipv4_addr_entry *ip_entry;
+	struct nfp_tun_neigh_v6 ipv6_route;
+	struct nfp_tun_neigh ipv4_route;
 	struct list_head *ptr, *storage;
 
 	unregister_netevent_notifier(&priv->tun.neigh_nb);
@@ -1239,12 +1327,33 @@ void nfp_tunnel_config_stop(struct nfp_app *app)
 
 	mutex_destroy(&priv->tun.ipv6_off_lock);
 
-	/* Free any memory that may be occupied by the route list. */
-	list_for_each_safe(ptr, storage, &priv->tun.neigh_off_list) {
-		route_entry = list_entry(ptr, struct nfp_ipv4_route_entry,
-					 list);
+	/* Free memory in the route list and remove entries from fw cache. */
+	list_for_each_entry_safe(route_entry, temp,
+				 &priv->tun.neigh_off_list_v4, list) {
+		memset(&ipv4_route, 0, sizeof(ipv4_route));
+		memcpy(&ipv4_route.dst_ipv4, &route_entry->ip_add,
+		       sizeof(ipv4_route.dst_ipv4));
 		list_del(&route_entry->list);
 		kfree(route_entry);
+
+		nfp_flower_xmit_tun_conf(app, NFP_FLOWER_CMSG_TYPE_TUN_NEIGH,
+					 sizeof(struct nfp_tun_neigh),
+					 (unsigned char *)&ipv4_route,
+					 GFP_KERNEL);
+	}
+
+	list_for_each_entry_safe(route_entry, temp,
+				 &priv->tun.neigh_off_list_v6, list) {
+		memset(&ipv6_route, 0, sizeof(ipv6_route));
+		memcpy(&ipv6_route.dst_ipv6, &route_entry->ip_add,
+		       sizeof(ipv6_route.dst_ipv6));
+		list_del(&route_entry->list);
+		kfree(route_entry);
+
+		nfp_flower_xmit_tun_conf(app, NFP_FLOWER_CMSG_TYPE_TUN_NEIGH_V6,
+					 sizeof(struct nfp_tun_neigh),
+					 (unsigned char *)&ipv6_route,
+					 GFP_KERNEL);
 	}
 
 	/* Destroy rhash. Entries should be cleaned on netdev notifier unreg. */
-- 
cgit v1.2.3-59-g8ed1b


From 9ea9bfa12240650feba7648dbb7f334dac8ba757 Mon Sep 17 00:00:00 2001
From: John Hurley <john.hurley@netronome.com>
Date: Tue, 17 Dec 2019 21:57:23 +0000
Subject: nfp: flower: support ipv6 tunnel keep-alive messages from fw

FW sends an update of IPv6 tunnels that are active in a given period. Use
this information to update the kernel table so that neighbour entries do
not time out when active on the NIC.

Signed-off-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/netronome/nfp/flower/cmsg.c   |  3 ++
 drivers/net/ethernet/netronome/nfp/flower/cmsg.h   |  1 +
 drivers/net/ethernet/netronome/nfp/flower/main.h   |  1 +
 .../ethernet/netronome/nfp/flower/tunnel_conf.c    | 62 ++++++++++++++++++++++
 4 files changed, 67 insertions(+)

diff --git a/drivers/net/ethernet/netronome/nfp/flower/cmsg.c b/drivers/net/ethernet/netronome/nfp/flower/cmsg.c
index 00b904ea4ebc..a595ddb92bff 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/cmsg.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/cmsg.c
@@ -278,6 +278,9 @@ nfp_flower_cmsg_process_one_rx(struct nfp_app *app, struct sk_buff *skb)
 	case NFP_FLOWER_CMSG_TYPE_ACTIVE_TUNS:
 		nfp_tunnel_keep_alive(app, skb);
 		break;
+	case NFP_FLOWER_CMSG_TYPE_ACTIVE_TUNS_V6:
+		nfp_tunnel_keep_alive_v6(app, skb);
+		break;
 	case NFP_FLOWER_CMSG_TYPE_QOS_STATS:
 		nfp_flower_stats_rlim_reply(app, skb);
 		break;
diff --git a/drivers/net/ethernet/netronome/nfp/flower/cmsg.h b/drivers/net/ethernet/netronome/nfp/flower/cmsg.h
index 5a8ad5814060..9b50d76bbc09 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/cmsg.h
+++ b/drivers/net/ethernet/netronome/nfp/flower/cmsg.h
@@ -574,6 +574,7 @@ enum nfp_flower_cmsg_type_port {
 	NFP_FLOWER_CMSG_TYPE_TUN_IPS_V6 =	22,
 	NFP_FLOWER_CMSG_TYPE_NO_NEIGH_V6 =	23,
 	NFP_FLOWER_CMSG_TYPE_TUN_NEIGH_V6 =	24,
+	NFP_FLOWER_CMSG_TYPE_ACTIVE_TUNS_V6 =	25,
 	NFP_FLOWER_CMSG_TYPE_MAX =		32,
 };
 
diff --git a/drivers/net/ethernet/netronome/nfp/flower/main.h b/drivers/net/ethernet/netronome/nfp/flower/main.h
index 7a4603261026..ddd7b7f5f5ab 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/main.h
+++ b/drivers/net/ethernet/netronome/nfp/flower/main.h
@@ -425,6 +425,7 @@ nfp_tunnel_add_ipv6_off(struct nfp_app *app, struct in6_addr *ipv6);
 void nfp_tunnel_request_route_v4(struct nfp_app *app, struct sk_buff *skb);
 void nfp_tunnel_request_route_v6(struct nfp_app *app, struct sk_buff *skb);
 void nfp_tunnel_keep_alive(struct nfp_app *app, struct sk_buff *skb);
+void nfp_tunnel_keep_alive_v6(struct nfp_app *app, struct sk_buff *skb);
 void nfp_flower_lag_init(struct nfp_fl_lag *lag);
 void nfp_flower_lag_cleanup(struct nfp_fl_lag *lag);
 int nfp_flower_lag_reset(struct nfp_fl_lag *lag);
diff --git a/drivers/net/ethernet/netronome/nfp/flower/tunnel_conf.c b/drivers/net/ethernet/netronome/nfp/flower/tunnel_conf.c
index b179b63c8456..2df3deedf9fd 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/tunnel_conf.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/tunnel_conf.c
@@ -54,6 +54,25 @@ struct nfp_tun_active_tuns {
 	} tun_info[];
 };
 
+/**
+ * struct nfp_tun_active_tuns_v6 - periodic message of active IPv6 tunnels
+ * @seq:		sequence number of the message
+ * @count:		number of tunnels report in message
+ * @flags:		options part of the request
+ * @tun_info.ipv6:		dest IPv6 address of active route
+ * @tun_info.egress_port:	port the encapsulated packet egressed
+ * @tun_info:		tunnels that have sent traffic in reported period
+ */
+struct nfp_tun_active_tuns_v6 {
+	__be32 seq;
+	__be32 count;
+	__be32 flags;
+	struct route_ip_info_v6 {
+		struct in6_addr ipv6;
+		__be32 egress_port;
+	} tun_info[];
+};
+
 /**
  * struct nfp_tun_neigh - neighbour/route entry on the NFP
  * @dst_ipv4:	destination IPv4 address
@@ -244,6 +263,49 @@ void nfp_tunnel_keep_alive(struct nfp_app *app, struct sk_buff *skb)
 	rcu_read_unlock();
 }
 
+void nfp_tunnel_keep_alive_v6(struct nfp_app *app, struct sk_buff *skb)
+{
+#if IS_ENABLED(CONFIG_IPV6)
+	struct nfp_tun_active_tuns_v6 *payload;
+	struct net_device *netdev;
+	int count, i, pay_len;
+	struct neighbour *n;
+	void *ipv6_add;
+	u32 port;
+
+	payload = nfp_flower_cmsg_get_data(skb);
+	count = be32_to_cpu(payload->count);
+	if (count > NFP_FL_IPV6_ADDRS_MAX) {
+		nfp_flower_cmsg_warn(app, "IPv6 tunnel keep-alive request exceeds max routes.\n");
+		return;
+	}
+
+	pay_len = nfp_flower_cmsg_get_data_len(skb);
+	if (pay_len != struct_size(payload, tun_info, count)) {
+		nfp_flower_cmsg_warn(app, "Corruption in tunnel keep-alive message.\n");
+		return;
+	}
+
+	rcu_read_lock();
+	for (i = 0; i < count; i++) {
+		ipv6_add = &payload->tun_info[i].ipv6;
+		port = be32_to_cpu(payload->tun_info[i].egress_port);
+		netdev = nfp_app_dev_get(app, port, NULL);
+		if (!netdev)
+			continue;
+
+		n = neigh_lookup(&nd_tbl, ipv6_add, netdev);
+		if (!n)
+			continue;
+
+		/* Update the used timestamp of neighbour */
+		neigh_event_send(n, NULL);
+		neigh_release(n);
+	}
+	rcu_read_unlock();
+#endif
+}
+
 static int
 nfp_flower_xmit_tun_conf(struct nfp_app *app, u8 mtype, u16 plen, void *pdata,
 			 gfp_t flag)
-- 
cgit v1.2.3-59-g8ed1b


From 783461604f7e3c7b4df9c653bf3db165cb9e3326 Mon Sep 17 00:00:00 2001
From: John Hurley <john.hurley@netronome.com>
Date: Tue, 17 Dec 2019 21:57:24 +0000
Subject: nfp: flower: update flow merge code to support IPv6 tunnels

Both pre-tunnel match rules and flow merge functions parse compiled
match/action fields for validation.

Update these validation functions to include IPv6 match and action fields.

Signed-off-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../net/ethernet/netronome/nfp/flower/offload.c    | 28 ++++++++++++++++++----
 1 file changed, 23 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/netronome/nfp/flower/offload.c b/drivers/net/ethernet/netronome/nfp/flower/offload.c
index d51210549299..7ca5c1becfcf 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/offload.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/offload.c
@@ -68,7 +68,8 @@
 #define NFP_FLOWER_PRE_TUN_RULE_FIELDS \
 	(NFP_FLOWER_LAYER_PORT | \
 	 NFP_FLOWER_LAYER_MAC | \
-	 NFP_FLOWER_LAYER_IPV4)
+	 NFP_FLOWER_LAYER_IPV4 | \
+	 NFP_FLOWER_LAYER_IPV6)
 
 struct nfp_flower_merge_check {
 	union {
@@ -566,10 +567,12 @@ nfp_flower_update_merge_with_actions(struct nfp_fl_payload *flow,
 	struct nfp_fl_set_ip4_addrs *ipv4_add;
 	struct nfp_fl_set_ipv6_addr *ipv6_add;
 	struct nfp_fl_push_vlan *push_vlan;
+	struct nfp_fl_pre_tunnel *pre_tun;
 	struct nfp_fl_set_tport *tport;
 	struct nfp_fl_set_eth *eth;
 	struct nfp_fl_act_head *a;
 	unsigned int act_off = 0;
+	bool ipv6_tun = false;
 	u8 act_id = 0;
 	u8 *ports;
 	int i;
@@ -597,8 +600,12 @@ nfp_flower_update_merge_with_actions(struct nfp_fl_payload *flow,
 			eth_broadcast_addr(&merge->l2.mac_src[0]);
 			memset(&merge->l4, 0xff,
 			       sizeof(struct nfp_flower_tp_ports));
-			memset(&merge->ipv4, 0xff,
-			       sizeof(struct nfp_flower_ipv4));
+			if (ipv6_tun)
+				memset(&merge->ipv6, 0xff,
+				       sizeof(struct nfp_flower_ipv6));
+			else
+				memset(&merge->ipv4, 0xff,
+				       sizeof(struct nfp_flower_ipv4));
 			break;
 		case NFP_FL_ACTION_OPCODE_SET_ETHERNET:
 			eth = (struct nfp_fl_set_eth *)a;
@@ -646,6 +653,10 @@ nfp_flower_update_merge_with_actions(struct nfp_fl_payload *flow,
 				ports[i] |= tport->tp_port_mask[i];
 			break;
 		case NFP_FL_ACTION_OPCODE_PRE_TUNNEL:
+			pre_tun = (struct nfp_fl_pre_tunnel *)a;
+			ipv6_tun = be16_to_cpu(pre_tun->flags) &
+					NFP_FL_PRE_TUN_IPV6;
+			break;
 		case NFP_FL_ACTION_OPCODE_PRE_LAG:
 		case NFP_FL_ACTION_OPCODE_PUSH_GENEVE:
 			break;
@@ -1107,15 +1118,22 @@ nfp_flower_validate_pre_tun_rule(struct nfp_app *app,
 		return -EOPNOTSUPP;
 	}
 
-	if (key_layer & NFP_FLOWER_LAYER_IPV4) {
+	if (key_layer & NFP_FLOWER_LAYER_IPV4 ||
+	    key_layer & NFP_FLOWER_LAYER_IPV6) {
+		/* Flags and proto fields have same offset in IPv4 and IPv6. */
 		int ip_flags = offsetof(struct nfp_flower_ipv4, ip_ext.flags);
 		int ip_proto = offsetof(struct nfp_flower_ipv4, ip_ext.proto);
+		int size;
 		int i;
 
+		size = key_layer & NFP_FLOWER_LAYER_IPV4 ?
+			sizeof(struct nfp_flower_ipv4) :
+			sizeof(struct nfp_flower_ipv6);
+
 		mask += sizeof(struct nfp_flower_mac_mpls);
 
 		/* Ensure proto and flags are the only IP layer fields. */
-		for (i = 0; i < sizeof(struct nfp_flower_ipv4); i++)
+		for (i = 0; i < size; i++)
 			if (mask[i] && i != ip_flags && i != ip_proto) {
 				NL_SET_ERR_MSG_MOD(extack, "unsupported pre-tunnel rule: only flags and proto can be matched in ip header");
 				return -EOPNOTSUPP;
-- 
cgit v1.2.3-59-g8ed1b


From 572c7245182a200e3ac69bb86a0e60cc34c34dc9 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <jakub.kicinski@netronome.com>
Date: Tue, 17 Dec 2019 14:12:00 -0800
Subject: nfp: pass packet pointer to nfp_net_parse_meta()

Make nfp_net_parse_meta() take a packet pointer and return
a drop/no drop decision. Right now it returns the end of
metadata and caller compares it to the packet pointer.

Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/netronome/nfp/nfp_net_common.c | 18 ++++++++----------
 1 file changed, 8 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
index d4eeb3b3cf35..780bd1daa601 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
@@ -1661,9 +1661,9 @@ nfp_net_set_hash_desc(struct net_device *netdev, struct nfp_meta_parsed *meta,
 			 &rx_hash->hash);
 }
 
-static void *
+static bool
 nfp_net_parse_meta(struct net_device *netdev, struct nfp_meta_parsed *meta,
-		   void *data, int meta_len)
+		   void *data, void *pkt, int meta_len)
 {
 	u32 meta_info;
 
@@ -1694,13 +1694,13 @@ nfp_net_parse_meta(struct net_device *netdev, struct nfp_meta_parsed *meta,
 			data += 4;
 			break;
 		default:
-			return NULL;
+			return true;
 		}
 
 		meta_info >>= NFP_NET_META_FIELD_SIZE;
 	}
 
-	return data;
+	return data != pkt;
 }
 
 static void
@@ -1885,12 +1885,10 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
 			nfp_net_set_hash_desc(dp->netdev, &meta,
 					      rxbuf->frag + meta_off, rxd);
 		} else if (meta_len) {
-			void *end;
-
-			end = nfp_net_parse_meta(dp->netdev, &meta,
-						 rxbuf->frag + meta_off,
-						 meta_len);
-			if (unlikely(end != rxbuf->frag + pkt_off)) {
+			if (unlikely(nfp_net_parse_meta(dp->netdev, &meta,
+							rxbuf->frag + meta_off,
+							rxbuf->frag + pkt_off,
+							meta_len))) {
 				nn_dp_warn(dp, "invalid RX packet metadata\n");
 				nfp_net_rx_drop(dp, r_vec, rx_ring, rxbuf,
 						NULL);
-- 
cgit v1.2.3-59-g8ed1b


From 8d5a49e9e31ba1ddd34a54b2351d068a90c78707 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <jakub.kicinski@netronome.com>
Date: Tue, 17 Dec 2019 14:12:01 -0800
Subject: net/tls: add helper for testing if socket is RX offloaded

There is currently no way for driver to reliably check that
the socket it has looked up is in fact RX offloaded. Add
a helper. This allows drivers to catch misbehaving firmware.

Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tls.h    | 9 +++++++++
 net/tls/tls_device.c | 5 +++--
 2 files changed, 12 insertions(+), 2 deletions(-)

diff --git a/include/net/tls.h b/include/net/tls.h
index df630f5fc723..bf9eb4823933 100644
--- a/include/net/tls.h
+++ b/include/net/tls.h
@@ -641,6 +641,7 @@ int tls_sw_fallback_init(struct sock *sk,
 #ifdef CONFIG_TLS_DEVICE
 void tls_device_init(void);
 void tls_device_cleanup(void);
+void tls_device_sk_destruct(struct sock *sk);
 int tls_set_device_offload(struct sock *sk, struct tls_context *ctx);
 void tls_device_free_resources_tx(struct sock *sk);
 int tls_set_device_offload_rx(struct sock *sk, struct tls_context *ctx);
@@ -649,6 +650,14 @@ void tls_device_rx_resync_new_rec(struct sock *sk, u32 rcd_len, u32 seq);
 void tls_offload_tx_resync_request(struct sock *sk, u32 got_seq, u32 exp_seq);
 int tls_device_decrypted(struct sock *sk, struct tls_context *tls_ctx,
 			 struct sk_buff *skb, struct strp_msg *rxm);
+
+static inline bool tls_is_sk_rx_device_offloaded(struct sock *sk)
+{
+	if (!sk_fullsock(sk) ||
+	    smp_load_acquire(&sk->sk_destruct) != tls_device_sk_destruct)
+		return false;
+	return tls_get_ctx(sk)->rx_conf == TLS_HW;
+}
 #else
 static inline void tls_device_init(void) {}
 static inline void tls_device_cleanup(void) {}
diff --git a/net/tls/tls_device.c b/net/tls/tls_device.c
index cd91ad812291..1ba5a92832bb 100644
--- a/net/tls/tls_device.c
+++ b/net/tls/tls_device.c
@@ -178,7 +178,7 @@ static void tls_icsk_clean_acked(struct sock *sk, u32 acked_seq)
  * socket and no in-flight SKBs associated with this
  * socket, so it is safe to free all the resources.
  */
-static void tls_device_sk_destruct(struct sock *sk)
+void tls_device_sk_destruct(struct sock *sk)
 {
 	struct tls_context *tls_ctx = tls_get_ctx(sk);
 	struct tls_offload_context_tx *ctx = tls_offload_ctx_tx(tls_ctx);
@@ -196,6 +196,7 @@ static void tls_device_sk_destruct(struct sock *sk)
 	if (refcount_dec_and_test(&tls_ctx->refcount))
 		tls_device_queue_ctx_destruction(tls_ctx);
 }
+EXPORT_SYMBOL_GPL(tls_device_sk_destruct);
 
 void tls_device_free_resources_tx(struct sock *sk)
 {
@@ -903,7 +904,7 @@ static void tls_device_attach(struct tls_context *ctx, struct sock *sk,
 		spin_unlock_irq(&tls_device_lock);
 
 		ctx->sk_destruct = sk->sk_destruct;
-		sk->sk_destruct = tls_device_sk_destruct;
+		smp_store_release(&sk->sk_destruct, tls_device_sk_destruct);
 	}
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 6a35ddc5445a8291ced6247a67977e110275acde Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <jakub.kicinski@netronome.com>
Date: Tue, 17 Dec 2019 14:12:02 -0800
Subject: nfp: tls: implement the stream sync RX resync

The simple RX resync strategy controlled by the kernel does not
guarantee as good results as if the device helps by detecting
the potential record boundaries and keeping track of them.

We've called this strategy stream scan in the tls-offload doc.

Implement this strategy for the NFP. The device sends a request
for record boundary confirmation, which is then recorded in
per-TLS socket state and responded to once record is reached.
Because the device keeps track of records passing after the
request was sent the response is not as latency sensitive as
when kernel just tries to tell the device the information
about the next record.

Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/netronome/nfp/ccm.h           |  1 +
 drivers/net/ethernet/netronome/nfp/crypto/crypto.h | 15 ++++
 drivers/net/ethernet/netronome/nfp/crypto/fw.h     |  8 ++
 drivers/net/ethernet/netronome/nfp/crypto/tls.c    | 89 ++++++++++++++++++++--
 drivers/net/ethernet/netronome/nfp/nfp_net.h       |  6 ++
 .../net/ethernet/netronome/nfp/nfp_net_common.c    | 11 ++-
 drivers/net/ethernet/netronome/nfp/nfp_net_ctrl.c  | 41 ++++++++--
 drivers/net/ethernet/netronome/nfp/nfp_net_ctrl.h  |  9 +++
 .../net/ethernet/netronome/nfp/nfp_net_ethtool.c   |  8 +-
 9 files changed, 172 insertions(+), 16 deletions(-)

diff --git a/drivers/net/ethernet/netronome/nfp/ccm.h b/drivers/net/ethernet/netronome/nfp/ccm.h
index a460c75522be..d81d450be50e 100644
--- a/drivers/net/ethernet/netronome/nfp/ccm.h
+++ b/drivers/net/ethernet/netronome/nfp/ccm.h
@@ -26,6 +26,7 @@ enum nfp_ccm_type {
 	NFP_CCM_TYPE_CRYPTO_ADD		= 10,
 	NFP_CCM_TYPE_CRYPTO_DEL		= 11,
 	NFP_CCM_TYPE_CRYPTO_UPDATE	= 12,
+	NFP_CCM_TYPE_CRYPTO_RESYNC	= 13,
 	__NFP_CCM_TYPE_MAX,
 };
 
diff --git a/drivers/net/ethernet/netronome/nfp/crypto/crypto.h b/drivers/net/ethernet/netronome/nfp/crypto/crypto.h
index 60372ddf69f0..bffe58bb2f27 100644
--- a/drivers/net/ethernet/netronome/nfp/crypto/crypto.h
+++ b/drivers/net/ethernet/netronome/nfp/crypto/crypto.h
@@ -4,6 +4,10 @@
 #ifndef NFP_CRYPTO_H
 #define NFP_CRYPTO_H 1
 
+struct net_device;
+struct nfp_net;
+struct nfp_net_tls_resync_req;
+
 struct nfp_net_tls_offload_ctx {
 	__be32 fw_handle[2];
 
@@ -17,11 +21,22 @@ struct nfp_net_tls_offload_ctx {
 
 #ifdef CONFIG_TLS_DEVICE
 int nfp_net_tls_init(struct nfp_net *nn);
+int nfp_net_tls_rx_resync_req(struct net_device *netdev,
+			      struct nfp_net_tls_resync_req *req,
+			      void *pkt, unsigned int pkt_len);
 #else
 static inline int nfp_net_tls_init(struct nfp_net *nn)
 {
 	return 0;
 }
+
+static inline int
+nfp_net_tls_rx_resync_req(struct net_device *netdev,
+			  struct nfp_net_tls_resync_req *req,
+			  void *pkt, unsigned int pkt_len)
+{
+	return -EOPNOTSUPP;
+}
 #endif
 
 #endif
diff --git a/drivers/net/ethernet/netronome/nfp/crypto/fw.h b/drivers/net/ethernet/netronome/nfp/crypto/fw.h
index 67413d946c4a..8d1458896bcb 100644
--- a/drivers/net/ethernet/netronome/nfp/crypto/fw.h
+++ b/drivers/net/ethernet/netronome/nfp/crypto/fw.h
@@ -9,6 +9,14 @@
 #define NFP_NET_CRYPTO_OP_TLS_1_2_AES_GCM_128_ENC	0
 #define NFP_NET_CRYPTO_OP_TLS_1_2_AES_GCM_128_DEC	1
 
+struct nfp_net_tls_resync_req {
+	__be32 fw_handle[2];
+	__be32 tcp_seq;
+	u8 l3_offset;
+	u8 l4_offset;
+	u8 resv[2];
+};
+
 struct nfp_crypto_reply_simple {
 	struct nfp_ccm_hdr hdr;
 	__be32 error;
diff --git a/drivers/net/ethernet/netronome/nfp/crypto/tls.c b/drivers/net/ethernet/netronome/nfp/crypto/tls.c
index 96a96b35c0ca..7c50e3dfb9d5 100644
--- a/drivers/net/ethernet/netronome/nfp/crypto/tls.c
+++ b/drivers/net/ethernet/netronome/nfp/crypto/tls.c
@@ -5,6 +5,7 @@
 #include <linux/ipv6.h>
 #include <linux/skbuff.h>
 #include <linux/string.h>
+#include <net/inet6_hashtables.h>
 #include <net/tls.h>
 
 #include "../ccm.h"
@@ -391,8 +392,9 @@ nfp_net_tls_add(struct net_device *netdev, struct sock *sk,
 	if (direction == TLS_OFFLOAD_CTX_DIR_TX)
 		return 0;
 
-	tls_offload_rx_resync_set_type(sk,
-				       TLS_OFFLOAD_SYNC_TYPE_CORE_NEXT_HINT);
+	if (!nn->tlv_caps.tls_resync_ss)
+		tls_offload_rx_resync_set_type(sk, TLS_OFFLOAD_SYNC_TYPE_CORE_NEXT_HINT);
+
 	return 0;
 
 err_fw_remove:
@@ -424,6 +426,7 @@ nfp_net_tls_resync(struct net_device *netdev, struct sock *sk, u32 seq,
 	struct nfp_net *nn = netdev_priv(netdev);
 	struct nfp_net_tls_offload_ctx *ntls;
 	struct nfp_crypto_req_update *req;
+	enum nfp_ccm_type type;
 	struct sk_buff *skb;
 	gfp_t flags;
 	int err;
@@ -442,15 +445,18 @@ nfp_net_tls_resync(struct net_device *netdev, struct sock *sk, u32 seq,
 	req->tcp_seq = cpu_to_be32(seq);
 	memcpy(req->rec_no, rcd_sn, sizeof(req->rec_no));
 
+	type = NFP_CCM_TYPE_CRYPTO_UPDATE;
 	if (direction == TLS_OFFLOAD_CTX_DIR_TX) {
-		err = nfp_net_tls_communicate_simple(nn, skb, "sync",
-						     NFP_CCM_TYPE_CRYPTO_UPDATE);
+		err = nfp_net_tls_communicate_simple(nn, skb, "sync", type);
 		if (err)
 			return err;
 		ntls->next_seq = seq;
 	} else {
-		nfp_ccm_mbox_post(nn, skb, NFP_CCM_TYPE_CRYPTO_UPDATE,
+		if (nn->tlv_caps.tls_resync_ss)
+			type = NFP_CCM_TYPE_CRYPTO_RESYNC;
+		nfp_ccm_mbox_post(nn, skb, type,
 				  sizeof(struct nfp_crypto_reply_simple));
+		atomic_inc(&nn->ktls_rx_resync_sent);
 	}
 
 	return 0;
@@ -462,6 +468,79 @@ static const struct tlsdev_ops nfp_net_tls_ops = {
 	.tls_dev_resync = nfp_net_tls_resync,
 };
 
+int nfp_net_tls_rx_resync_req(struct net_device *netdev,
+			      struct nfp_net_tls_resync_req *req,
+			      void *pkt, unsigned int pkt_len)
+{
+	struct nfp_net *nn = netdev_priv(netdev);
+	struct nfp_net_tls_offload_ctx *ntls;
+	struct ipv6hdr *ipv6h;
+	struct tcphdr *th;
+	struct iphdr *iph;
+	struct sock *sk;
+	__be32 tcp_seq;
+	int err;
+
+	iph = pkt + req->l3_offset;
+	ipv6h = pkt + req->l3_offset;
+	th = pkt + req->l4_offset;
+
+	if ((u8 *)&th[1] > (u8 *)pkt + pkt_len) {
+		netdev_warn_once(netdev, "invalid TLS RX resync request (l3_off: %hhu l4_off: %hhu pkt_len: %u)\n",
+				 req->l3_offset, req->l4_offset, pkt_len);
+		err = -EINVAL;
+		goto err_cnt_ign;
+	}
+
+	switch (iph->version) {
+	case 4:
+		sk = inet_lookup_established(dev_net(netdev), &tcp_hashinfo,
+					     iph->saddr, th->source, iph->daddr,
+					     th->dest, netdev->ifindex);
+		break;
+#if IS_ENABLED(CONFIG_IPV6)
+	case 6:
+		sk = __inet6_lookup_established(dev_net(netdev), &tcp_hashinfo,
+						&ipv6h->saddr, th->source,
+						&ipv6h->daddr, ntohs(th->dest),
+						netdev->ifindex, 0);
+		break;
+#endif
+	default:
+		netdev_warn_once(netdev, "invalid TLS RX resync request (l3_off: %hhu l4_off: %hhu ipver: %u)\n",
+				 req->l3_offset, req->l4_offset, iph->version);
+		err = -EINVAL;
+		goto err_cnt_ign;
+	}
+
+	err = 0;
+	if (!sk)
+		goto err_cnt_ign;
+	if (!tls_is_sk_rx_device_offloaded(sk) ||
+	    sk->sk_shutdown & RCV_SHUTDOWN)
+		goto err_put_sock;
+
+	ntls = tls_driver_ctx(sk, TLS_OFFLOAD_CTX_DIR_RX);
+	/* some FW versions can't report the handle and report 0s */
+	if (memchr_inv(&req->fw_handle, 0, sizeof(req->fw_handle)) &&
+	    memcmp(&req->fw_handle, &ntls->fw_handle, sizeof(ntls->fw_handle)))
+		goto err_put_sock;
+
+	/* copy to ensure alignment */
+	memcpy(&tcp_seq, &req->tcp_seq, sizeof(tcp_seq));
+	tls_offload_rx_resync_request(sk, tcp_seq);
+	atomic_inc(&nn->ktls_rx_resync_req);
+
+	sock_gen_put(sk);
+	return 0;
+
+err_put_sock:
+	sock_gen_put(sk);
+err_cnt_ign:
+	atomic_inc(&nn->ktls_rx_resync_ign);
+	return err;
+}
+
 static int nfp_net_tls_reset(struct nfp_net *nn)
 {
 	struct nfp_crypto_req_reset *req;
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net.h b/drivers/net/ethernet/netronome/nfp/nfp_net.h
index 250f510b1d21..ff4438478ea9 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net.h
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net.h
@@ -586,6 +586,9 @@ struct nfp_net_dp {
  * @ktls_conn_id_gen:	Trivial generator for kTLS connection ids (for TX)
  * @ktls_no_space:	Counter of firmware rejecting kTLS connection due to
  *			lack of space
+ * @ktls_rx_resync_req:	Counter of TLS RX resync requested
+ * @ktls_rx_resync_ign:	Counter of TLS RX resync requests ignored
+ * @ktls_rx_resync_sent:    Counter of TLS RX resync completed
  * @mbox_cmsg:		Common Control Message via vNIC mailbox state
  * @mbox_cmsg.queue:	CCM mbox queue of pending messages
  * @mbox_cmsg.wq:	CCM mbox wait queue of waiting processes
@@ -674,6 +677,9 @@ struct nfp_net {
 	atomic64_t ktls_conn_id_gen;
 
 	atomic_t ktls_no_space;
+	atomic_t ktls_rx_resync_req;
+	atomic_t ktls_rx_resync_ign;
+	atomic_t ktls_rx_resync_sent;
 
 	struct {
 		struct sk_buff_head queue;
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
index 780bd1daa601..9bfb3b077bc1 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
@@ -47,6 +47,7 @@
 #include "nfp_net_sriov.h"
 #include "nfp_port.h"
 #include "crypto/crypto.h"
+#include "crypto/fw.h"
 
 /**
  * nfp_net_get_fw_version() - Read and parse the FW version
@@ -1663,7 +1664,7 @@ nfp_net_set_hash_desc(struct net_device *netdev, struct nfp_meta_parsed *meta,
 
 static bool
 nfp_net_parse_meta(struct net_device *netdev, struct nfp_meta_parsed *meta,
-		   void *data, void *pkt, int meta_len)
+		   void *data, void *pkt, unsigned int pkt_len, int meta_len)
 {
 	u32 meta_info;
 
@@ -1693,6 +1694,12 @@ nfp_net_parse_meta(struct net_device *netdev, struct nfp_meta_parsed *meta,
 				(__force __wsum)__get_unaligned_cpu32(data);
 			data += 4;
 			break;
+		case NFP_NET_META_RESYNC_INFO:
+			if (nfp_net_tls_rx_resync_req(netdev, data, pkt,
+						      pkt_len))
+				return NULL;
+			data += sizeof(struct nfp_net_tls_resync_req);
+			break;
 		default:
 			return true;
 		}
@@ -1888,7 +1895,7 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
 			if (unlikely(nfp_net_parse_meta(dp->netdev, &meta,
 							rxbuf->frag + meta_off,
 							rxbuf->frag + pkt_off,
-							meta_len))) {
+							pkt_len, meta_len))) {
 				nn_dp_warn(dp, "invalid RX packet metadata\n");
 				nfp_net_rx_drop(dp, r_vec, rx_ring, rxbuf,
 						NULL);
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_ctrl.c b/drivers/net/ethernet/netronome/nfp/nfp_net_ctrl.c
index 45756648a85c..c3a763134e79 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_ctrl.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_ctrl.c
@@ -17,6 +17,30 @@ static void nfp_net_tlv_caps_reset(struct nfp_net_tlv_caps *caps)
 	caps->mbox_len = NFP_NET_CFG_MBOX_VAL_MAX_SZ;
 }
 
+static bool
+nfp_net_tls_parse_crypto_ops(struct device *dev, struct nfp_net_tlv_caps *caps,
+			     u8 __iomem *ctrl_mem, u8 __iomem *data,
+			     unsigned int length, unsigned int offset,
+			     bool rx_stream_scan)
+{
+	/* Ignore the legacy TLV if new one was already parsed */
+	if (caps->tls_resync_ss && !rx_stream_scan)
+		return true;
+
+	if (length < 32) {
+		dev_err(dev,
+			"CRYPTO OPS TLV should be at least 32B, is %dB offset:%u\n",
+			length, offset);
+		return false;
+	}
+
+	caps->crypto_ops = readl(data);
+	caps->crypto_enable_off = data - ctrl_mem + 16;
+	caps->tls_resync_ss = rx_stream_scan;
+
+	return true;
+}
+
 int nfp_net_tlv_caps_parse(struct device *dev, u8 __iomem *ctrl_mem,
 			   struct nfp_net_tlv_caps *caps)
 {
@@ -104,15 +128,10 @@ int nfp_net_tlv_caps_parse(struct device *dev, u8 __iomem *ctrl_mem,
 				caps->mbox_cmsg_types = readl(data);
 			break;
 		case NFP_NET_CFG_TLV_TYPE_CRYPTO_OPS:
-			if (length < 32) {
-				dev_err(dev,
-					"CRYPTO OPS TLV should be at least 32B, is %dB offset:%u\n",
-					length, offset);
+			if (!nfp_net_tls_parse_crypto_ops(dev, caps, ctrl_mem,
+							  data, length, offset,
+							  false))
 				return -EINVAL;
-			}
-
-			caps->crypto_ops = readl(data);
-			caps->crypto_enable_off = data - ctrl_mem + 16;
 			break;
 		case NFP_NET_CFG_TLV_TYPE_VNIC_STATS:
 			if ((data - ctrl_mem) % 8) {
@@ -123,6 +142,12 @@ int nfp_net_tlv_caps_parse(struct device *dev, u8 __iomem *ctrl_mem,
 			caps->vnic_stats_off = data - ctrl_mem;
 			caps->vnic_stats_cnt = length / 10;
 			break;
+		case NFP_NET_CFG_TLV_TYPE_CRYPTO_OPS_RX_SCAN:
+			if (!nfp_net_tls_parse_crypto_ops(dev, caps, ctrl_mem,
+							  data, length, offset,
+							  true))
+				return -EINVAL;
+			break;
 		default:
 			if (!FIELD_GET(NFP_NET_CFG_TLV_HEADER_REQUIRED, hdr))
 				break;
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_ctrl.h b/drivers/net/ethernet/netronome/nfp/nfp_net_ctrl.h
index c38cc36a2a70..3d61a8cb60b0 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_ctrl.h
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_ctrl.h
@@ -45,6 +45,7 @@
 #define NFP_NET_META_PORTID		5
 #define NFP_NET_META_CSUM		6 /* checksum complete type */
 #define NFP_NET_META_CONN_HANDLE	7
+#define NFP_NET_META_RESYNC_INFO	8 /* RX resync info request */
 
 #define NFP_META_PORT_ID_CTRL		~0U
 
@@ -479,6 +480,7 @@
  * 8 words, bitmaps of supported and enabled crypto operations.
  * First 16B (4 words) contains a bitmap of supported crypto operations,
  * and next 16B contain the enabled operations.
+ * This capability is made obsolete by ones with better sync methods.
  *
  * %NFP_NET_CFG_TLV_TYPE_VNIC_STATS:
  * Variable, per-vNIC statistics, data should be 8B aligned (FW should insert
@@ -490,6 +492,10 @@
  * This TLV overwrites %NFP_NET_CFG_STATS_* values (statistics in this TLV
  * duplicate the old ones, so driver should be careful not to unnecessarily
  * render both).
+ *
+ * %NFP_NET_CFG_TLV_TYPE_CRYPTO_OPS_RX_SCAN:
+ * Same as %NFP_NET_CFG_TLV_TYPE_CRYPTO_OPS, but crypto TLS does stream scan
+ * RX sync, rather than kernel-assisted sync.
  */
 #define NFP_NET_CFG_TLV_TYPE_UNKNOWN		0
 #define NFP_NET_CFG_TLV_TYPE_RESERVED		1
@@ -502,6 +508,7 @@
 #define NFP_NET_CFG_TLV_TYPE_MBOX_CMSG_TYPES	10
 #define NFP_NET_CFG_TLV_TYPE_CRYPTO_OPS		11 /* see crypto/fw.h */
 #define NFP_NET_CFG_TLV_TYPE_VNIC_STATS		12
+#define NFP_NET_CFG_TLV_TYPE_CRYPTO_OPS_RX_SCAN	13
 
 struct device;
 
@@ -516,6 +523,7 @@ struct device;
  * @crypto_enable_off:	offset of crypto ops enable region
  * @vnic_stats_off:	offset of vNIC stats area
  * @vnic_stats_cnt:	number of vNIC stats
+ * @tls_resync_ss:	TLS resync will be performed via stream scan
  */
 struct nfp_net_tlv_caps {
 	u32 me_freq_mhz;
@@ -527,6 +535,7 @@ struct nfp_net_tlv_caps {
 	unsigned int crypto_enable_off;
 	unsigned int vnic_stats_off;
 	unsigned int vnic_stats_cnt;
+	unsigned int tls_resync_ss:1;
 };
 
 int nfp_net_tlv_caps_parse(struct device *dev, u8 __iomem *ctrl_mem,
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
index b386a221c599..d648e32c0520 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
@@ -174,7 +174,7 @@ static const char nfp_tlv_stat_names[][ETH_GSTRING_LEN] = {
 #define NN_ET_SWITCH_STATS_LEN 9
 #define NN_RVEC_GATHER_STATS	13
 #define NN_RVEC_PER_Q_STATS	3
-#define NN_CTRL_PATH_STATS	1
+#define NN_CTRL_PATH_STATS	4
 
 #define SFP_SFF_REV_COMPLIANCE	1
 
@@ -476,6 +476,9 @@ static u8 *nfp_vnic_get_sw_stats_strings(struct net_device *netdev, u8 *data)
 	data = nfp_pr_et(data, "tx_tls_drop_no_sync_data");
 
 	data = nfp_pr_et(data, "hw_tls_no_space");
+	data = nfp_pr_et(data, "rx_tls_resync_req_ok");
+	data = nfp_pr_et(data, "rx_tls_resync_req_ign");
+	data = nfp_pr_et(data, "rx_tls_resync_sent");
 
 	return data;
 }
@@ -524,6 +527,9 @@ static u64 *nfp_vnic_get_sw_stats(struct net_device *netdev, u64 *data)
 		*data++ = gathered_stats[j];
 
 	*data++ = atomic_read(&nn->ktls_no_space);
+	*data++ = atomic_read(&nn->ktls_rx_resync_req);
+	*data++ = atomic_read(&nn->ktls_rx_resync_ign);
+	*data++ = atomic_read(&nn->ktls_rx_resync_sent);
 
 	return data;
 }
-- 
cgit v1.2.3-59-g8ed1b


From d63b542c625c4af8f78f227b2749b87ca6acdb37 Mon Sep 17 00:00:00 2001
From: Florian Fainelli <f.fainelli@gmail.com>
Date: Tue, 17 Dec 2019 16:29:50 -0800
Subject: net: systemport: Set correct DMA mask

SYSTEMPORT is capabable of doing up to 40-bit of physical addresses, set
an appropriate DMA mask to permit that.

Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bcmsysport.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/net/ethernet/broadcom/bcmsysport.c b/drivers/net/ethernet/broadcom/bcmsysport.c
index 8e3152779a61..1907e47fd0af 100644
--- a/drivers/net/ethernet/broadcom/bcmsysport.c
+++ b/drivers/net/ethernet/broadcom/bcmsysport.c
@@ -2427,6 +2427,14 @@ static int bcm_sysport_probe(struct platform_device *pdev)
 	if (!of_id || !of_id->data)
 		return -EINVAL;
 
+	ret = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(40));
+	if (ret)
+		ret = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
+	if (ret) {
+		dev_err(&pdev->dev, "unable to set DMA mask: %d\n", ret);
+		return ret;
+	}
+
 	/* Fairly quickly we need to know the type of adapter we have */
 	params = of_id->data;
 
-- 
cgit v1.2.3-59-g8ed1b


From 99d55638d4b053c549c924959747c58392b2dbea Mon Sep 17 00:00:00 2001
From: Doug Berger <opendmb@gmail.com>
Date: Tue, 17 Dec 2019 16:51:08 -0800
Subject: net: bcmgenet: enable NETIF_F_HIGHDMA flag

This commit configures the DMA masks for the GENET driver and
sets the NETIF_F_HIGHDMA flag to report support of the feature.

Signed-off-by: Doug Berger <opendmb@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/genet/bcmgenet.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
index 32f1245a69e2..d9defb8b1e5f 100644
--- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c
+++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
@@ -2,7 +2,7 @@
 /*
  * Broadcom GENET (Gigabit Ethernet) controller driver
  *
- * Copyright (c) 2014-2017 Broadcom
+ * Copyright (c) 2014-2019 Broadcom
  */
 
 #define pr_fmt(fmt)				"bcmgenet: " fmt
@@ -3537,7 +3537,7 @@ static int bcmgenet_probe(struct platform_device *pdev)
 
 	/* Set hardware features */
 	dev->hw_features |= NETIF_F_SG | NETIF_F_IP_CSUM |
-		NETIF_F_IPV6_CSUM | NETIF_F_RXCSUM;
+		NETIF_F_IPV6_CSUM | NETIF_F_HIGHDMA | NETIF_F_RXCSUM;
 
 	/* Request the WOL interrupt and advertise suspend if available */
 	priv->wol_irq_disabled = true;
@@ -3574,6 +3574,14 @@ static int bcmgenet_probe(struct platform_device *pdev)
 
 	bcmgenet_set_hw_params(priv);
 
+	err = -EIO;
+	if (priv->hw_params->flags & GENET_HAS_40BITS)
+		err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(40));
+	if (err)
+		err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
+	if (err)
+		goto err;
+
 	/* Mii wait queue */
 	init_waitqueue_head(&priv->wq);
 	/* Always use RX_BUF_LENGTH (2KB) buffer for all chips */
-- 
cgit v1.2.3-59-g8ed1b


From dd8e911b7f2af2315184a8e551ca77ff226a4ec5 Mon Sep 17 00:00:00 2001
From: Doug Berger <opendmb@gmail.com>
Date: Tue, 17 Dec 2019 16:51:09 -0800
Subject: net: bcmgenet: enable NETIF_F_HW_CSUM feature

The GENET hardware should be capable of generating IP checksums
using the NETIF_F_HW_CSUM feature, so switch to using that feature
instead of the depricated NETIF_F_IP_CSUM and NETIF_F_IPV6_CSUM.

Signed-off-by: Doug Berger <opendmb@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/genet/bcmgenet.c | 29 +++++++++++---------------
 1 file changed, 12 insertions(+), 17 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
index d9defb8b1e5f..cd07b3ad1d53 100644
--- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c
+++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
@@ -549,7 +549,7 @@ static int bcmgenet_set_tx_csum(struct net_device *dev,
 	tbuf_ctrl = bcmgenet_tbuf_ctrl_get(priv);
 	rbuf_ctrl = bcmgenet_rbuf_readl(priv, RBUF_CTRL);
 
-	desc_64b_en = !!(wanted & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM));
+	desc_64b_en = !!(wanted & NETIF_F_HW_CSUM);
 
 	/* enable 64 bytes descriptor in both directions (RBUF and TBUF) */
 	if (desc_64b_en) {
@@ -574,7 +574,7 @@ static int bcmgenet_set_features(struct net_device *dev,
 	netdev_features_t wanted = dev->wanted_features;
 	int ret = 0;
 
-	if (changed & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM))
+	if (changed & NETIF_F_HW_CSUM)
 		ret = bcmgenet_set_tx_csum(dev, wanted);
 	if (changed & (NETIF_F_RXCSUM))
 		ret = bcmgenet_set_rx_csum(dev, wanted);
@@ -1516,24 +1516,19 @@ static struct sk_buff *bcmgenet_put_tx_csum(struct net_device *dev,
 			ip_proto = ipv6_hdr(skb)->nexthdr;
 			break;
 		default:
-			return skb;
+			/* don't use UDP flag */
+			ip_proto = 0;
+			break;
 		}
 
 		offset = skb_checksum_start_offset(skb) - sizeof(*status);
 		tx_csum_info = (offset << STATUS_TX_CSUM_START_SHIFT) |
-				(offset + skb->csum_offset);
+				(offset + skb->csum_offset) |
+				STATUS_TX_CSUM_LV;
 
-		/* Set the length valid bit for TCP and UDP and just set
-		 * the special UDP flag for IPv4, else just set to 0.
-		 */
-		if (ip_proto == IPPROTO_TCP || ip_proto == IPPROTO_UDP) {
-			tx_csum_info |= STATUS_TX_CSUM_LV;
-			if (ip_proto == IPPROTO_UDP &&
-			    ip_ver == htons(ETH_P_IP))
-				tx_csum_info |= STATUS_TX_CSUM_PROTO_UDP;
-		} else {
-			tx_csum_info = 0;
-		}
+		/* Set the special UDP flag for UDP */
+		if (ip_proto == IPPROTO_UDP)
+			tx_csum_info |= STATUS_TX_CSUM_PROTO_UDP;
 
 		status->tx_csum_info = tx_csum_info;
 	}
@@ -3536,8 +3531,8 @@ static int bcmgenet_probe(struct platform_device *pdev)
 	priv->msg_enable = netif_msg_init(-1, GENET_MSG_DEFAULT);
 
 	/* Set hardware features */
-	dev->hw_features |= NETIF_F_SG | NETIF_F_IP_CSUM |
-		NETIF_F_IPV6_CSUM | NETIF_F_HIGHDMA | NETIF_F_RXCSUM;
+	dev->hw_features |= NETIF_F_SG | NETIF_F_HIGHDMA | NETIF_F_HW_CSUM |
+		NETIF_F_RXCSUM;
 
 	/* Request the WOL interrupt and advertise suspend if available */
 	priv->wol_irq_disabled = true;
-- 
cgit v1.2.3-59-g8ed1b


From 810155397890ecdea23868ef0ed10cb80134997f Mon Sep 17 00:00:00 2001
From: Doug Berger <opendmb@gmail.com>
Date: Tue, 17 Dec 2019 16:51:10 -0800
Subject: net: bcmgenet: use CHECKSUM_COMPLETE for NETIF_F_RXCSUM

This commit updates the Rx checksum offload behavior of the driver
to use the more generic CHECKSUM_COMPLETE method that supports all
protocols over the CHECKSUM_UNNECESSARY method that only applies
to some protocols known by the hardware.

This behavior is perceived to be superior.

Signed-off-by: Doug Berger <opendmb@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/genet/bcmgenet.c | 19 +++++++------------
 drivers/net/ethernet/broadcom/genet/bcmgenet.h |  2 +-
 2 files changed, 8 insertions(+), 13 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
index cd07b3ad1d53..13cbe5828adb 100644
--- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c
+++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
@@ -521,7 +521,7 @@ static int bcmgenet_set_rx_csum(struct net_device *dev,
 
 	/* enable rx checksumming */
 	if (rx_csum_en)
-		rbuf_chk_ctrl |= RBUF_RXCHK_EN;
+		rbuf_chk_ctrl |= RBUF_RXCHK_EN | RBUF_L3_PARSE_DIS;
 	else
 		rbuf_chk_ctrl &= ~RBUF_RXCHK_EN;
 	priv->desc_rxchk_en = rx_csum_en;
@@ -1739,7 +1739,6 @@ static unsigned int bcmgenet_desc_rx(struct bcmgenet_rx_ring *ring,
 	unsigned int bytes_processed = 0;
 	unsigned int p_index, mask;
 	unsigned int discards;
-	unsigned int chksum_ok = 0;
 
 	/* Clear status before servicing to reduce spurious interrupts */
 	if (ring->index == DESC_INDEX) {
@@ -1790,9 +1789,15 @@ static unsigned int bcmgenet_desc_rx(struct bcmgenet_rx_ring *ring,
 				dmadesc_get_length_status(priv, cb->bd_addr);
 		} else {
 			struct status_64 *status;
+			__be16 rx_csum;
 
 			status = (struct status_64 *)skb->data;
 			dma_length_status = status->length_status;
+			rx_csum = (__force __be16)(status->rx_csum & 0xffff);
+			if (priv->desc_rxchk_en) {
+				skb->csum = (__force __wsum)ntohs(rx_csum);
+				skb->ip_summed = CHECKSUM_COMPLETE;
+			}
 		}
 
 		/* DMA flags and length are still valid no matter how
@@ -1835,18 +1840,12 @@ static unsigned int bcmgenet_desc_rx(struct bcmgenet_rx_ring *ring,
 			goto next;
 		} /* error packet */
 
-		chksum_ok = (dma_flag & priv->dma_rx_chk_bit) &&
-			     priv->desc_rxchk_en;
-
 		skb_put(skb, len);
 		if (priv->desc_64b_en) {
 			skb_pull(skb, 64);
 			len -= 64;
 		}
 
-		if (likely(chksum_ok))
-			skb->ip_summed = CHECKSUM_UNNECESSARY;
-
 		/* remove hardware 2bytes added for IP alignment */
 		skb_pull(skb, 2);
 		len -= 2;
@@ -3322,19 +3321,15 @@ static void bcmgenet_set_hw_params(struct bcmgenet_priv *priv)
 	if (GENET_IS_V5(priv) || GENET_IS_V4(priv)) {
 		bcmgenet_dma_regs = bcmgenet_dma_regs_v3plus;
 		genet_dma_ring_regs = genet_dma_ring_regs_v4;
-		priv->dma_rx_chk_bit = DMA_RX_CHK_V3PLUS;
 	} else if (GENET_IS_V3(priv)) {
 		bcmgenet_dma_regs = bcmgenet_dma_regs_v3plus;
 		genet_dma_ring_regs = genet_dma_ring_regs_v123;
-		priv->dma_rx_chk_bit = DMA_RX_CHK_V3PLUS;
 	} else if (GENET_IS_V2(priv)) {
 		bcmgenet_dma_regs = bcmgenet_dma_regs_v2;
 		genet_dma_ring_regs = genet_dma_ring_regs_v123;
-		priv->dma_rx_chk_bit = DMA_RX_CHK_V12;
 	} else if (GENET_IS_V1(priv)) {
 		bcmgenet_dma_regs = bcmgenet_dma_regs_v1;
 		genet_dma_ring_regs = genet_dma_ring_regs_v123;
-		priv->dma_rx_chk_bit = DMA_RX_CHK_V12;
 	}
 
 	/* enum genet_version starts at 1 */
diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.h b/drivers/net/ethernet/broadcom/genet/bcmgenet.h
index a5659197598f..d33c0d093f82 100644
--- a/drivers/net/ethernet/broadcom/genet/bcmgenet.h
+++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.h
@@ -251,6 +251,7 @@ struct bcmgenet_mib_counters {
 #define RBUF_CHK_CTRL			0x14
 #define  RBUF_RXCHK_EN			(1 << 0)
 #define  RBUF_SKIP_FCS			(1 << 4)
+#define  RBUF_L3_PARSE_DIS		(1 << 5)
 
 #define RBUF_ENERGY_CTRL		0x9c
 #define  RBUF_EEE_EN			(1 << 0)
@@ -663,7 +664,6 @@ struct bcmgenet_priv {
 	bool desc_rxchk_en;
 	bool crc_fwd_en;
 
-	unsigned int dma_rx_chk_bit;
 	u32 dma_max_burst_length;
 
 	u32 msg_enable;
-- 
cgit v1.2.3-59-g8ed1b


From f63db4efdd50a2b67b97336fde67661af8af112b Mon Sep 17 00:00:00 2001
From: Doug Berger <opendmb@gmail.com>
Date: Tue, 17 Dec 2019 16:51:11 -0800
Subject: net: bcmgenet: Refactor bcmgenet_set_features()

In preparation for unconditionally enabling TX and RX checksum
offloads, refactor bcmgenet_set_features() a bit such that
__netdev_update_features() during register_netdev() can make sure
that features are correctly programmed during network device
registration.

Since we can now be called during register_netdev() with clocks
gated, we need to temporarily turn them on/off in order to have a
successful register programming.

We also move the CRC forward setting read into
bcmgenet_set_features() since priv->crc_fwd_en matters while
turning on RX checksum offload, that way we are guaranteed they
are in sync in case we ever add support for NETIF_F_RXFCS at some
point in the future.

Signed-off-by: Doug Berger <opendmb@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/genet/bcmgenet.c | 38 +++++++++++++-------------
 1 file changed, 19 insertions(+), 19 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
index 13cbe5828adb..811713e3d230 100644
--- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c
+++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
@@ -508,8 +508,8 @@ static int bcmgenet_set_link_ksettings(struct net_device *dev,
 	return phy_ethtool_ksettings_set(dev->phydev, cmd);
 }
 
-static int bcmgenet_set_rx_csum(struct net_device *dev,
-				netdev_features_t wanted)
+static void bcmgenet_set_rx_csum(struct net_device *dev,
+				 netdev_features_t wanted)
 {
 	struct bcmgenet_priv *priv = netdev_priv(dev);
 	u32 rbuf_chk_ctrl;
@@ -535,12 +535,10 @@ static int bcmgenet_set_rx_csum(struct net_device *dev,
 		rbuf_chk_ctrl &= ~RBUF_SKIP_FCS;
 
 	bcmgenet_rbuf_writel(priv, rbuf_chk_ctrl, RBUF_CHK_CTRL);
-
-	return 0;
 }
 
-static int bcmgenet_set_tx_csum(struct net_device *dev,
-				netdev_features_t wanted)
+static void bcmgenet_set_tx_csum(struct net_device *dev,
+				 netdev_features_t wanted)
 {
 	struct bcmgenet_priv *priv = netdev_priv(dev);
 	bool desc_64b_en;
@@ -563,21 +561,27 @@ static int bcmgenet_set_tx_csum(struct net_device *dev,
 
 	bcmgenet_tbuf_ctrl_set(priv, tbuf_ctrl);
 	bcmgenet_rbuf_writel(priv, rbuf_ctrl, RBUF_CTRL);
-
-	return 0;
 }
 
 static int bcmgenet_set_features(struct net_device *dev,
 				 netdev_features_t features)
 {
-	netdev_features_t changed = features ^ dev->features;
-	netdev_features_t wanted = dev->wanted_features;
-	int ret = 0;
+	struct bcmgenet_priv *priv = netdev_priv(dev);
+	u32 reg;
+	int ret;
+
+	ret = clk_prepare_enable(priv->clk);
+	if (ret)
+		return ret;
+
+	/* Make sure we reflect the value of CRC_CMD_FWD */
+	reg = bcmgenet_umac_readl(priv, UMAC_CMD);
+	priv->crc_fwd_en = !!(reg & CMD_CRC_FWD);
 
-	if (changed & NETIF_F_HW_CSUM)
-		ret = bcmgenet_set_tx_csum(dev, wanted);
-	if (changed & (NETIF_F_RXCSUM))
-		ret = bcmgenet_set_rx_csum(dev, wanted);
+	bcmgenet_set_tx_csum(dev, features);
+	bcmgenet_set_rx_csum(dev, features);
+
+	clk_disable_unprepare(priv->clk);
 
 	return ret;
 }
@@ -2880,10 +2884,6 @@ static int bcmgenet_open(struct net_device *dev)
 
 	init_umac(priv);
 
-	/* Make sure we reflect the value of CRC_CMD_FWD */
-	reg = bcmgenet_umac_readl(priv, UMAC_CMD);
-	priv->crc_fwd_en = !!(reg & CMD_CRC_FWD);
-
 	bcmgenet_set_hw_addr(priv, dev->dev_addr);
 
 	if (priv->internal_phy) {
-- 
cgit v1.2.3-59-g8ed1b


From 206f54b66cbf6f71e9e86f50f60ffdf7f565c3b7 Mon Sep 17 00:00:00 2001
From: Doug Berger <opendmb@gmail.com>
Date: Tue, 17 Dec 2019 16:51:12 -0800
Subject: net: bcmgenet: Utilize bcmgenet_set_features() during resume/open

During driver resume and open, the HW may have lost its context/state,
utilize bcmgenet_set_features() to make sure we do restore the correct
set of features that were previously configured.

Signed-off-by: Doug Berger <opendmb@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/genet/bcmgenet.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
index 811713e3d230..0e57effd5b19 100644
--- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c
+++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
@@ -2884,6 +2884,11 @@ static int bcmgenet_open(struct net_device *dev)
 
 	init_umac(priv);
 
+	/* Apply features again in case we changed them while interface was
+	 * down
+	 */
+	bcmgenet_set_features(dev, dev->features);
+
 	bcmgenet_set_hw_addr(priv, dev->dev_addr);
 
 	if (priv->internal_phy) {
@@ -3687,6 +3692,9 @@ static int bcmgenet_resume(struct device *d)
 	genphy_config_aneg(dev->phydev);
 	bcmgenet_mii_config(priv->dev, false);
 
+	/* Restore enabled features */
+	bcmgenet_set_features(dev, dev->features);
+
 	bcmgenet_set_hw_addr(priv, dev->dev_addr);
 
 	if (priv->internal_phy) {
-- 
cgit v1.2.3-59-g8ed1b


From ae895c49905cd99aca24d23361b72ed559b30f4f Mon Sep 17 00:00:00 2001
From: Doug Berger <opendmb@gmail.com>
Date: Tue, 17 Dec 2019 16:51:13 -0800
Subject: net: bcmgenet: Turn on offloads by default

We can turn on the RX/TX checksum offloads and the scatter/gather
features by default and make sure that those are properly reflected
back to e.g: stacked devices such as VLAN.

Signed-off-by: Doug Berger <opendmb@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/genet/bcmgenet.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
index 0e57effd5b19..13e9154db253 100644
--- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c
+++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
@@ -3530,9 +3530,11 @@ static int bcmgenet_probe(struct platform_device *pdev)
 
 	priv->msg_enable = netif_msg_init(-1, GENET_MSG_DEFAULT);
 
-	/* Set hardware features */
-	dev->hw_features |= NETIF_F_SG | NETIF_F_HIGHDMA | NETIF_F_HW_CSUM |
-		NETIF_F_RXCSUM;
+	/* Set default features */
+	dev->features |= NETIF_F_SG | NETIF_F_HIGHDMA | NETIF_F_HW_CSUM |
+			 NETIF_F_RXCSUM;
+	dev->hw_features |= dev->features;
+	dev->vlan_features |= dev->features;
 
 	/* Request the WOL interrupt and advertise suspend if available */
 	priv->wol_irq_disabled = true;
-- 
cgit v1.2.3-59-g8ed1b


From e3fa85883d1566d788960c0f31866a26c898c070 Mon Sep 17 00:00:00 2001
From: Doug Berger <opendmb@gmail.com>
Date: Tue, 17 Dec 2019 16:51:14 -0800
Subject: net: bcmgenet: Be drop monitor friendly while re-allocating headroom

During bcmgenet_put_tx_csum() make sure we differentiate a SKB
headroom re-allocation failure from the normal swap and replace
path.

Signed-off-by: Doug Berger <opendmb@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/genet/bcmgenet.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
index 13e9154db253..e2bca19bf10b 100644
--- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c
+++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
@@ -1499,11 +1499,12 @@ static struct sk_buff *bcmgenet_put_tx_csum(struct net_device *dev,
 		 * enough headroom for us to insert 64B status block.
 		 */
 		new_skb = skb_realloc_headroom(skb, sizeof(*status));
-		dev_kfree_skb(skb);
 		if (!new_skb) {
+			dev_kfree_skb_any(skb);
 			dev->stats.tx_dropped++;
 			return NULL;
 		}
+		dev_consume_skb_any(skb);
 		skb = new_skb;
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From f1af17c0b7a51357699bba7c8cba1276c9486907 Mon Sep 17 00:00:00 2001
From: Doug Berger <opendmb@gmail.com>
Date: Tue, 17 Dec 2019 16:51:15 -0800
Subject: net: bcmgenet: Add software counters to track reallocations

When inserting the TSB, keep track of how many times we had to do
it and if there was a failure in doing so, this helps profile the
driver for possibly incorrect headroom settings.

Signed-off-by: Doug Berger <opendmb@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/genet/bcmgenet.c | 6 ++++++
 drivers/net/ethernet/broadcom/genet/bcmgenet.h | 2 ++
 2 files changed, 8 insertions(+)

diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
index e2bca19bf10b..3ee7917e3fc0 100644
--- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c
+++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
@@ -861,6 +861,9 @@ static const struct bcmgenet_stats bcmgenet_gstrings_stats[] = {
 	STAT_GENET_SOFT_MIB("alloc_rx_buff_failed", mib.alloc_rx_buff_failed),
 	STAT_GENET_SOFT_MIB("rx_dma_failed", mib.rx_dma_failed),
 	STAT_GENET_SOFT_MIB("tx_dma_failed", mib.tx_dma_failed),
+	STAT_GENET_SOFT_MIB("tx_realloc_tsb", mib.tx_realloc_tsb),
+	STAT_GENET_SOFT_MIB("tx_realloc_tsb_failed",
+			    mib.tx_realloc_tsb_failed),
 	/* Per TX queues */
 	STAT_GENET_Q(0),
 	STAT_GENET_Q(1),
@@ -1487,6 +1490,7 @@ static void bcmgenet_tx_reclaim_all(struct net_device *dev)
 static struct sk_buff *bcmgenet_put_tx_csum(struct net_device *dev,
 					    struct sk_buff *skb)
 {
+	struct bcmgenet_priv *priv = netdev_priv(dev);
 	struct status_64 *status = NULL;
 	struct sk_buff *new_skb;
 	u16 offset;
@@ -1501,11 +1505,13 @@ static struct sk_buff *bcmgenet_put_tx_csum(struct net_device *dev,
 		new_skb = skb_realloc_headroom(skb, sizeof(*status));
 		if (!new_skb) {
 			dev_kfree_skb_any(skb);
+			priv->mib.tx_realloc_tsb_failed++;
 			dev->stats.tx_dropped++;
 			return NULL;
 		}
 		dev_consume_skb_any(skb);
 		skb = new_skb;
+		priv->mib.tx_realloc_tsb++;
 	}
 
 	skb_push(skb, sizeof(*status));
diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.h b/drivers/net/ethernet/broadcom/genet/bcmgenet.h
index d33c0d093f82..61a6fe9f4cec 100644
--- a/drivers/net/ethernet/broadcom/genet/bcmgenet.h
+++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.h
@@ -144,6 +144,8 @@ struct bcmgenet_mib_counters {
 	u32	alloc_rx_buff_failed;
 	u32	rx_dma_failed;
 	u32	tx_dma_failed;
+	u32	tx_realloc_tsb;
+	u32	tx_realloc_tsb_failed;
 };
 
 #define UMAC_HD_BKP_CTRL		0x004
-- 
cgit v1.2.3-59-g8ed1b


From 5080832627b65e3772a35d1dced68c64e2b24442 Mon Sep 17 00:00:00 2001
From: Nathan Chancellor <natechancellor@gmail.com>
Date: Tue, 17 Dec 2019 18:21:52 -0700
Subject: NFC: pn544: Adjust indentation in pn544_hci_check_presence

Clang warns

../drivers/nfc/pn544/pn544.c:696:4: warning: misleading indentation;
statement is not part of the previous 'if' [-Wmisleading-indentation]
                 return nfc_hci_send_cmd(hdev, NFC_HCI_RF_READER_A_GATE,
                 ^
../drivers/nfc/pn544/pn544.c:692:3: note: previous statement is here
                if (target->nfcid1_len != 4 && target->nfcid1_len != 7 &&
                ^
1 warning generated.

This warning occurs because there is a space after the tab on this line.
Remove it so that the indentation is consistent with the Linux kernel
coding style and clang no longer warns.

Fixes: da052850b911 ("NFC: Add pn544 presence check for different targets")
Link: https://github.com/ClangBuiltLinux/linux/issues/814
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/nfc/pn544/pn544.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/nfc/pn544/pn544.c b/drivers/nfc/pn544/pn544.c
index cda996f6954e..2b83156efe3f 100644
--- a/drivers/nfc/pn544/pn544.c
+++ b/drivers/nfc/pn544/pn544.c
@@ -693,7 +693,7 @@ static int pn544_hci_check_presence(struct nfc_hci_dev *hdev,
 		    target->nfcid1_len != 10)
 			return -EOPNOTSUPP;
 
-		 return nfc_hci_send_cmd(hdev, NFC_HCI_RF_READER_A_GATE,
+		return nfc_hci_send_cmd(hdev, NFC_HCI_RF_READER_A_GATE,
 				     PN544_RF_READER_CMD_ACTIVATE_NEXT,
 				     target->nfcid1, target->nfcid1_len, NULL);
 	} else if (target->supported_protocols & (NFC_PROTO_JEWEL_MASK |
-- 
cgit v1.2.3-59-g8ed1b


From 0536b85239b8440735cdd910aae0eb076ebbb439 Mon Sep 17 00:00:00 2001
From: Björn Töpel <bjorn.topel@intel.com>
Date: Thu, 19 Dec 2019 07:09:59 +0100
Subject: xdp: Simplify devmap cleanup
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

After the RCU flavor consolidation [1], call_rcu() and
synchronize_rcu() waits for preempt-disable regions (NAPI) in addition
to the read-side critical sections. As a result of this, the cleanup
code in devmap can be simplified

* There is no longer a need to flush in __dev_map_entry_free, since we
  know that this has been done when the call_rcu() callback is
  triggered.

* When freeing the map, there is no need to explicitly wait for a
  flush. It's guaranteed to be done after the synchronize_rcu() call
  in dev_map_free(). The rcu_barrier() is still needed, so that the
  map is not freed prior the elements.

[1] https://lwn.net/Articles/777036/

Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/bpf/20191219061006.21980-2-bjorn.topel@gmail.com
---
 kernel/bpf/devmap.c | 43 +++++--------------------------------------
 1 file changed, 5 insertions(+), 38 deletions(-)

diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index 3d3d61b5985b..b7595de6a91a 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -201,7 +201,7 @@ static struct bpf_map *dev_map_alloc(union bpf_attr *attr)
 static void dev_map_free(struct bpf_map *map)
 {
 	struct bpf_dtab *dtab = container_of(map, struct bpf_dtab, map);
-	int i, cpu;
+	int i;
 
 	/* At this point bpf_prog->aux->refcnt == 0 and this map->refcnt == 0,
 	 * so the programs (can be more than one that used this map) were
@@ -221,18 +221,6 @@ static void dev_map_free(struct bpf_map *map)
 	/* Make sure prior __dev_map_entry_free() have completed. */
 	rcu_barrier();
 
-	/* To ensure all pending flush operations have completed wait for flush
-	 * list to empty on _all_ cpus.
-	 * Because the above synchronize_rcu() ensures the map is disconnected
-	 * from the program we can assume no new items will be added.
-	 */
-	for_each_online_cpu(cpu) {
-		struct list_head *flush_list = per_cpu_ptr(dtab->flush_list, cpu);
-
-		while (!list_empty(flush_list))
-			cond_resched();
-	}
-
 	if (dtab->map.map_type == BPF_MAP_TYPE_DEVMAP_HASH) {
 		for (i = 0; i < dtab->n_buckets; i++) {
 			struct bpf_dtab_netdev *dev;
@@ -345,8 +333,7 @@ static int dev_map_hash_get_next_key(struct bpf_map *map, void *key,
 	return -ENOENT;
 }
 
-static int bq_xmit_all(struct xdp_bulk_queue *bq, u32 flags,
-		       bool in_napi_ctx)
+static int bq_xmit_all(struct xdp_bulk_queue *bq, u32 flags)
 {
 	struct bpf_dtab_netdev *obj = bq->obj;
 	struct net_device *dev = obj->dev;
@@ -384,11 +371,7 @@ error:
 	for (i = 0; i < bq->count; i++) {
 		struct xdp_frame *xdpf = bq->q[i];
 
-		/* RX path under NAPI protection, can return frames faster */
-		if (likely(in_napi_ctx))
-			xdp_return_frame_rx_napi(xdpf);
-		else
-			xdp_return_frame(xdpf);
+		xdp_return_frame_rx_napi(xdpf);
 		drops++;
 	}
 	goto out;
@@ -409,7 +392,7 @@ void __dev_map_flush(struct bpf_map *map)
 
 	rcu_read_lock();
 	list_for_each_entry_safe(bq, tmp, flush_list, flush_node)
-		bq_xmit_all(bq, XDP_XMIT_FLUSH, true);
+		bq_xmit_all(bq, XDP_XMIT_FLUSH);
 	rcu_read_unlock();
 }
 
@@ -440,7 +423,7 @@ static int bq_enqueue(struct bpf_dtab_netdev *obj, struct xdp_frame *xdpf,
 	struct xdp_bulk_queue *bq = this_cpu_ptr(obj->bulkq);
 
 	if (unlikely(bq->count == DEV_MAP_BULK_SIZE))
-		bq_xmit_all(bq, 0, true);
+		bq_xmit_all(bq, 0);
 
 	/* Ingress dev_rx will be the same for all xdp_frame's in
 	 * bulk_queue, because bq stored per-CPU and must be flushed
@@ -509,27 +492,11 @@ static void *dev_map_hash_lookup_elem(struct bpf_map *map, void *key)
 	return dev ? &dev->ifindex : NULL;
 }
 
-static void dev_map_flush_old(struct bpf_dtab_netdev *dev)
-{
-	if (dev->dev->netdev_ops->ndo_xdp_xmit) {
-		struct xdp_bulk_queue *bq;
-		int cpu;
-
-		rcu_read_lock();
-		for_each_online_cpu(cpu) {
-			bq = per_cpu_ptr(dev->bulkq, cpu);
-			bq_xmit_all(bq, XDP_XMIT_FLUSH, false);
-		}
-		rcu_read_unlock();
-	}
-}
-
 static void __dev_map_entry_free(struct rcu_head *rcu)
 {
 	struct bpf_dtab_netdev *dev;
 
 	dev = container_of(rcu, struct bpf_dtab_netdev, rcu);
-	dev_map_flush_old(dev);
 	free_percpu(dev->bulkq);
 	dev_put(dev->dev);
 	kfree(dev);
-- 
cgit v1.2.3-59-g8ed1b


From 4bc188c7f23a5a308d7f15dda1b6a286d74e8954 Mon Sep 17 00:00:00 2001
From: Björn Töpel <bjorn.topel@intel.com>
Date: Thu, 19 Dec 2019 07:10:00 +0100
Subject: xdp: Simplify cpumap cleanup
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

After the RCU flavor consolidation [1], call_rcu() and
synchronize_rcu() waits for preempt-disable regions (NAPI) in addition
to the read-side critical sections. As a result of this, the cleanup
code in cpumap can be simplified

* There is no longer a need to flush in __cpu_map_entry_free, since we
  know that this has been done when the call_rcu() callback is
  triggered.

* When freeing the map, there is no need to explicitly wait for a
  flush. It's guaranteed to be done after the synchronize_rcu() call
  in cpu_map_free().

[1] https://lwn.net/Articles/777036/

Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/bpf/20191219061006.21980-3-bjorn.topel@gmail.com
---
 kernel/bpf/cpumap.c | 34 +++++-----------------------------
 1 file changed, 5 insertions(+), 29 deletions(-)

diff --git a/kernel/bpf/cpumap.c b/kernel/bpf/cpumap.c
index ef49e17ae47c..04c8eb11cd90 100644
--- a/kernel/bpf/cpumap.c
+++ b/kernel/bpf/cpumap.c
@@ -75,7 +75,7 @@ struct bpf_cpu_map {
 	struct list_head __percpu *flush_list;
 };
 
-static int bq_flush_to_queue(struct xdp_bulk_queue *bq, bool in_napi_ctx);
+static int bq_flush_to_queue(struct xdp_bulk_queue *bq);
 
 static struct bpf_map *cpu_map_alloc(union bpf_attr *attr)
 {
@@ -399,7 +399,6 @@ free_rcu:
 static void __cpu_map_entry_free(struct rcu_head *rcu)
 {
 	struct bpf_cpu_map_entry *rcpu;
-	int cpu;
 
 	/* This cpu_map_entry have been disconnected from map and one
 	 * RCU graze-period have elapsed.  Thus, XDP cannot queue any
@@ -408,13 +407,6 @@ static void __cpu_map_entry_free(struct rcu_head *rcu)
 	 */
 	rcpu = container_of(rcu, struct bpf_cpu_map_entry, rcu);
 
-	/* Flush remaining packets in percpu bulkq */
-	for_each_online_cpu(cpu) {
-		struct xdp_bulk_queue *bq = per_cpu_ptr(rcpu->bulkq, cpu);
-
-		/* No concurrent bq_enqueue can run at this point */
-		bq_flush_to_queue(bq, false);
-	}
 	free_percpu(rcpu->bulkq);
 	/* Cannot kthread_stop() here, last put free rcpu resources */
 	put_cpu_map_entry(rcpu);
@@ -507,7 +499,6 @@ static int cpu_map_update_elem(struct bpf_map *map, void *key, void *value,
 static void cpu_map_free(struct bpf_map *map)
 {
 	struct bpf_cpu_map *cmap = container_of(map, struct bpf_cpu_map, map);
-	int cpu;
 	u32 i;
 
 	/* At this point bpf_prog->aux->refcnt == 0 and this map->refcnt == 0,
@@ -522,18 +513,6 @@ static void cpu_map_free(struct bpf_map *map)
 	bpf_clear_redirect_map(map);
 	synchronize_rcu();
 
-	/* To ensure all pending flush operations have completed wait for flush
-	 * list be empty on _all_ cpus. Because the above synchronize_rcu()
-	 * ensures the map is disconnected from the program we can assume no new
-	 * items will be added to the list.
-	 */
-	for_each_online_cpu(cpu) {
-		struct list_head *flush_list = per_cpu_ptr(cmap->flush_list, cpu);
-
-		while (!list_empty(flush_list))
-			cond_resched();
-	}
-
 	/* For cpu_map the remote CPUs can still be using the entries
 	 * (struct bpf_cpu_map_entry).
 	 */
@@ -599,7 +578,7 @@ const struct bpf_map_ops cpu_map_ops = {
 	.map_check_btf		= map_check_no_btf,
 };
 
-static int bq_flush_to_queue(struct xdp_bulk_queue *bq, bool in_napi_ctx)
+static int bq_flush_to_queue(struct xdp_bulk_queue *bq)
 {
 	struct bpf_cpu_map_entry *rcpu = bq->obj;
 	unsigned int processed = 0, drops = 0;
@@ -620,10 +599,7 @@ static int bq_flush_to_queue(struct xdp_bulk_queue *bq, bool in_napi_ctx)
 		err = __ptr_ring_produce(q, xdpf);
 		if (err) {
 			drops++;
-			if (likely(in_napi_ctx))
-				xdp_return_frame_rx_napi(xdpf);
-			else
-				xdp_return_frame(xdpf);
+			xdp_return_frame_rx_napi(xdpf);
 		}
 		processed++;
 	}
@@ -646,7 +622,7 @@ static int bq_enqueue(struct bpf_cpu_map_entry *rcpu, struct xdp_frame *xdpf)
 	struct xdp_bulk_queue *bq = this_cpu_ptr(rcpu->bulkq);
 
 	if (unlikely(bq->count == CPU_MAP_BULK_SIZE))
-		bq_flush_to_queue(bq, true);
+		bq_flush_to_queue(bq);
 
 	/* Notice, xdp_buff/page MUST be queued here, long enough for
 	 * driver to code invoking us to finished, due to driver
@@ -688,7 +664,7 @@ void __cpu_map_flush(struct bpf_map *map)
 	struct xdp_bulk_queue *bq, *tmp;
 
 	list_for_each_entry_safe(bq, tmp, flush_list, flush_node) {
-		bq_flush_to_queue(bq, true);
+		bq_flush_to_queue(bq);
 
 		/* If already running, costs spin_lock_irqsave + smb_mb */
 		wake_up_process(bq->obj->kthread);
-- 
cgit v1.2.3-59-g8ed1b


From fb5aacdf3603ccbafe1da74eecd132eb4a31e53f Mon Sep 17 00:00:00 2001
From: Björn Töpel <bjorn.topel@intel.com>
Date: Thu, 19 Dec 2019 07:10:01 +0100
Subject: xdp: Fix graze->grace type-o in cpumap comments
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Simple spelling fix.

Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/bpf/20191219061006.21980-4-bjorn.topel@gmail.com
---
 kernel/bpf/cpumap.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/kernel/bpf/cpumap.c b/kernel/bpf/cpumap.c
index 04c8eb11cd90..f9deed659798 100644
--- a/kernel/bpf/cpumap.c
+++ b/kernel/bpf/cpumap.c
@@ -401,7 +401,7 @@ static void __cpu_map_entry_free(struct rcu_head *rcu)
 	struct bpf_cpu_map_entry *rcpu;
 
 	/* This cpu_map_entry have been disconnected from map and one
-	 * RCU graze-period have elapsed.  Thus, XDP cannot queue any
+	 * RCU grace-period have elapsed.  Thus, XDP cannot queue any
 	 * new packets and cannot change/set flush_needed that can
 	 * find this entry.
 	 */
@@ -428,7 +428,7 @@ static void __cpu_map_entry_free(struct rcu_head *rcu)
  * percpu bulkq to queue.  Due to caller map_delete_elem() disable
  * preemption, cannot call kthread_stop() to make sure queue is empty.
  * Instead a work_queue is started for stopping kthread,
- * cpu_map_kthread_stop, which waits for an RCU graze period before
+ * cpu_map_kthread_stop, which waits for an RCU grace period before
  * stopping kthread, emptying the queue.
  */
 static void __cpu_map_entry_replace(struct bpf_cpu_map *cmap,
@@ -523,7 +523,7 @@ static void cpu_map_free(struct bpf_map *map)
 		if (!rcpu)
 			continue;
 
-		/* bq flush and cleanup happens after RCU graze-period */
+		/* bq flush and cleanup happens after RCU grace-period */
 		__cpu_map_entry_replace(cmap, i, NULL); /* call_rcu */
 	}
 	free_percpu(cmap->flush_list);
-- 
cgit v1.2.3-59-g8ed1b


From e312b9e706ed6d94f6cc9088fcd9fbd81de4525c Mon Sep 17 00:00:00 2001
From: Björn Töpel <bjorn.topel@intel.com>
Date: Thu, 19 Dec 2019 07:10:02 +0100
Subject: xsk: Make xskmap flush_list common for all map instances
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The xskmap flush list is used to track entries that need to flushed
from via the xdp_do_flush_map() function. This list used to be
per-map, but there is really no reason for that. Instead make the
flush list global for all xskmaps, which simplifies __xsk_map_flush()
and xsk_map_alloc().

Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/bpf/20191219061006.21980-5-bjorn.topel@gmail.com
---
 include/net/xdp_sock.h | 11 ++++-------
 kernel/bpf/xskmap.c    | 18 +++---------------
 net/core/filter.c      |  9 ++++-----
 net/xdp/xsk.c          | 17 +++++++++--------
 4 files changed, 20 insertions(+), 35 deletions(-)

diff --git a/include/net/xdp_sock.h b/include/net/xdp_sock.h
index e3780e4b74e1..48594740d67c 100644
--- a/include/net/xdp_sock.h
+++ b/include/net/xdp_sock.h
@@ -72,7 +72,6 @@ struct xdp_umem {
 
 struct xsk_map {
 	struct bpf_map map;
-	struct list_head __percpu *flush_list;
 	spinlock_t lock; /* Synchronize map updates */
 	struct xdp_sock *xsk_map[];
 };
@@ -139,9 +138,8 @@ void xsk_map_try_sock_delete(struct xsk_map *map, struct xdp_sock *xs,
 			     struct xdp_sock **map_entry);
 int xsk_map_inc(struct xsk_map *map);
 void xsk_map_put(struct xsk_map *map);
-int __xsk_map_redirect(struct bpf_map *map, struct xdp_buff *xdp,
-		       struct xdp_sock *xs);
-void __xsk_map_flush(struct bpf_map *map);
+int __xsk_map_redirect(struct xdp_sock *xs, struct xdp_buff *xdp);
+void __xsk_map_flush(void);
 
 static inline struct xdp_sock *__xsk_map_lookup_elem(struct bpf_map *map,
 						     u32 key)
@@ -369,13 +367,12 @@ static inline u64 xsk_umem_adjust_offset(struct xdp_umem *umem, u64 handle,
 	return 0;
 }
 
-static inline int __xsk_map_redirect(struct bpf_map *map, struct xdp_buff *xdp,
-				     struct xdp_sock *xs)
+static inline int __xsk_map_redirect(struct xdp_sock *xs, struct xdp_buff *xdp)
 {
 	return -EOPNOTSUPP;
 }
 
-static inline void __xsk_map_flush(struct bpf_map *map)
+static inline void __xsk_map_flush(void)
 {
 }
 
diff --git a/kernel/bpf/xskmap.c b/kernel/bpf/xskmap.c
index 90c4fce1c981..2cc5c8f4c800 100644
--- a/kernel/bpf/xskmap.c
+++ b/kernel/bpf/xskmap.c
@@ -72,9 +72,9 @@ static void xsk_map_sock_delete(struct xdp_sock *xs,
 static struct bpf_map *xsk_map_alloc(union bpf_attr *attr)
 {
 	struct bpf_map_memory mem;
-	int cpu, err, numa_node;
+	int err, numa_node;
 	struct xsk_map *m;
-	u64 cost, size;
+	u64 size;
 
 	if (!capable(CAP_NET_ADMIN))
 		return ERR_PTR(-EPERM);
@@ -86,9 +86,8 @@ static struct bpf_map *xsk_map_alloc(union bpf_attr *attr)
 
 	numa_node = bpf_map_attr_numa_node(attr);
 	size = struct_size(m, xsk_map, attr->max_entries);
-	cost = size + array_size(sizeof(*m->flush_list), num_possible_cpus());
 
-	err = bpf_map_charge_init(&mem, cost);
+	err = bpf_map_charge_init(&mem, size);
 	if (err < 0)
 		return ERR_PTR(err);
 
@@ -102,16 +101,6 @@ static struct bpf_map *xsk_map_alloc(union bpf_attr *attr)
 	bpf_map_charge_move(&m->map.memory, &mem);
 	spin_lock_init(&m->lock);
 
-	m->flush_list = alloc_percpu(struct list_head);
-	if (!m->flush_list) {
-		bpf_map_charge_finish(&m->map.memory);
-		bpf_map_area_free(m);
-		return ERR_PTR(-ENOMEM);
-	}
-
-	for_each_possible_cpu(cpu)
-		INIT_LIST_HEAD(per_cpu_ptr(m->flush_list, cpu));
-
 	return &m->map;
 }
 
@@ -121,7 +110,6 @@ static void xsk_map_free(struct bpf_map *map)
 
 	bpf_clear_redirect_map(map);
 	synchronize_net();
-	free_percpu(m->flush_list);
 	bpf_map_area_free(m);
 }
 
diff --git a/net/core/filter.c b/net/core/filter.c
index a411f7835dee..c51678c473c5 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -3511,8 +3511,7 @@ err:
 
 static int __bpf_tx_xdp_map(struct net_device *dev_rx, void *fwd,
 			    struct bpf_map *map,
-			    struct xdp_buff *xdp,
-			    u32 index)
+			    struct xdp_buff *xdp)
 {
 	int err;
 
@@ -3537,7 +3536,7 @@ static int __bpf_tx_xdp_map(struct net_device *dev_rx, void *fwd,
 	case BPF_MAP_TYPE_XSKMAP: {
 		struct xdp_sock *xs = fwd;
 
-		err = __xsk_map_redirect(map, xdp, xs);
+		err = __xsk_map_redirect(xs, xdp);
 		return err;
 	}
 	default:
@@ -3562,7 +3561,7 @@ void xdp_do_flush_map(void)
 			__cpu_map_flush(map);
 			break;
 		case BPF_MAP_TYPE_XSKMAP:
-			__xsk_map_flush(map);
+			__xsk_map_flush();
 			break;
 		default:
 			break;
@@ -3619,7 +3618,7 @@ static int xdp_do_redirect_map(struct net_device *dev, struct xdp_buff *xdp,
 	if (ri->map_to_flush && unlikely(ri->map_to_flush != map))
 		xdp_do_flush_map();
 
-	err = __bpf_tx_xdp_map(dev, fwd, map, xdp, index);
+	err = __bpf_tx_xdp_map(dev, fwd, map, xdp);
 	if (unlikely(err))
 		goto err;
 
diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c
index 956793893c9d..e45c27f5cfca 100644
--- a/net/xdp/xsk.c
+++ b/net/xdp/xsk.c
@@ -31,6 +31,8 @@
 
 #define TX_BATCH_SIZE 16
 
+static DEFINE_PER_CPU(struct list_head, xskmap_flush_list);
+
 bool xsk_is_setup_for_bpf_map(struct xdp_sock *xs)
 {
 	return READ_ONCE(xs->rx) &&  READ_ONCE(xs->umem) &&
@@ -264,11 +266,9 @@ out_unlock:
 	return err;
 }
 
-int __xsk_map_redirect(struct bpf_map *map, struct xdp_buff *xdp,
-		       struct xdp_sock *xs)
+int __xsk_map_redirect(struct xdp_sock *xs, struct xdp_buff *xdp)
 {
-	struct xsk_map *m = container_of(map, struct xsk_map, map);
-	struct list_head *flush_list = this_cpu_ptr(m->flush_list);
+	struct list_head *flush_list = this_cpu_ptr(&xskmap_flush_list);
 	int err;
 
 	err = xsk_rcv(xs, xdp);
@@ -281,10 +281,9 @@ int __xsk_map_redirect(struct bpf_map *map, struct xdp_buff *xdp,
 	return 0;
 }
 
-void __xsk_map_flush(struct bpf_map *map)
+void __xsk_map_flush(void)
 {
-	struct xsk_map *m = container_of(map, struct xsk_map, map);
-	struct list_head *flush_list = this_cpu_ptr(m->flush_list);
+	struct list_head *flush_list = this_cpu_ptr(&xskmap_flush_list);
 	struct xdp_sock *xs, *tmp;
 
 	list_for_each_entry_safe(xs, tmp, flush_list, flush_node) {
@@ -1177,7 +1176,7 @@ static struct pernet_operations xsk_net_ops = {
 
 static int __init xsk_init(void)
 {
-	int err;
+	int err, cpu;
 
 	err = proto_register(&xsk_proto, 0 /* no slab */);
 	if (err)
@@ -1195,6 +1194,8 @@ static int __init xsk_init(void)
 	if (err)
 		goto out_pernet;
 
+	for_each_possible_cpu(cpu)
+		INIT_LIST_HEAD(&per_cpu(xskmap_flush_list, cpu));
 	return 0;
 
 out_pernet:
-- 
cgit v1.2.3-59-g8ed1b


From 96360004b8628541f5d05a845ea213267db0b1a2 Mon Sep 17 00:00:00 2001
From: Björn Töpel <bjorn.topel@intel.com>
Date: Thu, 19 Dec 2019 07:10:03 +0100
Subject: xdp: Make devmap flush_list common for all map instances
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The devmap flush list is used to track entries that need to flushed
from via the xdp_do_flush_map() function. This list used to be
per-map, but there is really no reason for that. Instead make the
flush list global for all devmaps, which simplifies __dev_map_flush()
and dev_map_init_map().

Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/bpf/20191219061006.21980-6-bjorn.topel@gmail.com
---
 include/linux/bpf.h |  4 ++--
 kernel/bpf/devmap.c | 35 +++++++++++++----------------------
 net/core/filter.c   |  2 +-
 3 files changed, 16 insertions(+), 25 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index d467983e61bb..31191804ca09 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -959,7 +959,7 @@ struct sk_buff;
 
 struct bpf_dtab_netdev *__dev_map_lookup_elem(struct bpf_map *map, u32 key);
 struct bpf_dtab_netdev *__dev_map_hash_lookup_elem(struct bpf_map *map, u32 key);
-void __dev_map_flush(struct bpf_map *map);
+void __dev_map_flush(void);
 int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
 		    struct net_device *dev_rx);
 int dev_map_generic_redirect(struct bpf_dtab_netdev *dst, struct sk_buff *skb,
@@ -1068,7 +1068,7 @@ static inline struct net_device  *__dev_map_hash_lookup_elem(struct bpf_map *map
 	return NULL;
 }
 
-static inline void __dev_map_flush(struct bpf_map *map)
+static inline void __dev_map_flush(void)
 {
 }
 
diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index b7595de6a91a..da9c832fc5c8 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -75,7 +75,6 @@ struct bpf_dtab_netdev {
 struct bpf_dtab {
 	struct bpf_map map;
 	struct bpf_dtab_netdev **netdev_map; /* DEVMAP type only */
-	struct list_head __percpu *flush_list;
 	struct list_head list;
 
 	/* these are only used for DEVMAP_HASH type maps */
@@ -85,6 +84,7 @@ struct bpf_dtab {
 	u32 n_buckets;
 };
 
+static DEFINE_PER_CPU(struct list_head, dev_map_flush_list);
 static DEFINE_SPINLOCK(dev_map_lock);
 static LIST_HEAD(dev_map_list);
 
@@ -109,8 +109,8 @@ static inline struct hlist_head *dev_map_index_hash(struct bpf_dtab *dtab,
 
 static int dev_map_init_map(struct bpf_dtab *dtab, union bpf_attr *attr)
 {
-	int err, cpu;
-	u64 cost;
+	u64 cost = 0;
+	int err;
 
 	/* check sanity of attributes */
 	if (attr->max_entries == 0 || attr->key_size != 4 ||
@@ -125,9 +125,6 @@ static int dev_map_init_map(struct bpf_dtab *dtab, union bpf_attr *attr)
 
 	bpf_map_init_from_attr(&dtab->map, attr);
 
-	/* make sure page count doesn't overflow */
-	cost = (u64) sizeof(struct list_head) * num_possible_cpus();
-
 	if (attr->map_type == BPF_MAP_TYPE_DEVMAP_HASH) {
 		dtab->n_buckets = roundup_pow_of_two(dtab->map.max_entries);
 
@@ -143,17 +140,10 @@ static int dev_map_init_map(struct bpf_dtab *dtab, union bpf_attr *attr)
 	if (err)
 		return -EINVAL;
 
-	dtab->flush_list = alloc_percpu(struct list_head);
-	if (!dtab->flush_list)
-		goto free_charge;
-
-	for_each_possible_cpu(cpu)
-		INIT_LIST_HEAD(per_cpu_ptr(dtab->flush_list, cpu));
-
 	if (attr->map_type == BPF_MAP_TYPE_DEVMAP_HASH) {
 		dtab->dev_index_head = dev_map_create_hash(dtab->n_buckets);
 		if (!dtab->dev_index_head)
-			goto free_percpu;
+			goto free_charge;
 
 		spin_lock_init(&dtab->index_lock);
 	} else {
@@ -161,13 +151,11 @@ static int dev_map_init_map(struct bpf_dtab *dtab, union bpf_attr *attr)
 						      sizeof(struct bpf_dtab_netdev *),
 						      dtab->map.numa_node);
 		if (!dtab->netdev_map)
-			goto free_percpu;
+			goto free_charge;
 	}
 
 	return 0;
 
-free_percpu:
-	free_percpu(dtab->flush_list);
 free_charge:
 	bpf_map_charge_finish(&dtab->map.memory);
 	return -ENOMEM;
@@ -254,7 +242,6 @@ static void dev_map_free(struct bpf_map *map)
 		bpf_map_area_free(dtab->netdev_map);
 	}
 
-	free_percpu(dtab->flush_list);
 	kfree(dtab);
 }
 
@@ -384,10 +371,9 @@ error:
  * net device can be torn down. On devmap tear down we ensure the flush list
  * is empty before completing to ensure all flush operations have completed.
  */
-void __dev_map_flush(struct bpf_map *map)
+void __dev_map_flush(void)
 {
-	struct bpf_dtab *dtab = container_of(map, struct bpf_dtab, map);
-	struct list_head *flush_list = this_cpu_ptr(dtab->flush_list);
+	struct list_head *flush_list = this_cpu_ptr(&dev_map_flush_list);
 	struct xdp_bulk_queue *bq, *tmp;
 
 	rcu_read_lock();
@@ -419,7 +405,7 @@ static int bq_enqueue(struct bpf_dtab_netdev *obj, struct xdp_frame *xdpf,
 		      struct net_device *dev_rx)
 
 {
-	struct list_head *flush_list = this_cpu_ptr(obj->dtab->flush_list);
+	struct list_head *flush_list = this_cpu_ptr(&dev_map_flush_list);
 	struct xdp_bulk_queue *bq = this_cpu_ptr(obj->bulkq);
 
 	if (unlikely(bq->count == DEV_MAP_BULK_SIZE))
@@ -777,10 +763,15 @@ static struct notifier_block dev_map_notifier = {
 
 static int __init dev_map_init(void)
 {
+	int cpu;
+
 	/* Assure tracepoint shadow struct _bpf_dtab_netdev is in sync */
 	BUILD_BUG_ON(offsetof(struct bpf_dtab_netdev, dev) !=
 		     offsetof(struct _bpf_dtab_netdev, dev));
 	register_netdevice_notifier(&dev_map_notifier);
+
+	for_each_possible_cpu(cpu)
+		INIT_LIST_HEAD(&per_cpu(dev_map_flush_list, cpu));
 	return 0;
 }
 
diff --git a/net/core/filter.c b/net/core/filter.c
index c51678c473c5..b7570cb84902 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -3555,7 +3555,7 @@ void xdp_do_flush_map(void)
 		switch (map->map_type) {
 		case BPF_MAP_TYPE_DEVMAP:
 		case BPF_MAP_TYPE_DEVMAP_HASH:
-			__dev_map_flush(map);
+			__dev_map_flush();
 			break;
 		case BPF_MAP_TYPE_CPUMAP:
 			__cpu_map_flush(map);
-- 
cgit v1.2.3-59-g8ed1b


From cdfafe98cabefeedbbc65af5c191c59745c03298 Mon Sep 17 00:00:00 2001
From: Björn Töpel <bjorn.topel@intel.com>
Date: Thu, 19 Dec 2019 07:10:04 +0100
Subject: xdp: Make cpumap flush_list common for all map instances
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The cpumap flush list is used to track entries that need to flushed
from via the xdp_do_flush_map() function. This list used to be
per-map, but there is really no reason for that. Instead make the
flush list global for all devmaps, which simplifies __cpu_map_flush()
and cpu_map_alloc().

Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/bpf/20191219061006.21980-7-bjorn.topel@gmail.com
---
 include/linux/bpf.h |  4 ++--
 kernel/bpf/cpumap.c | 36 ++++++++++++++++++------------------
 net/core/filter.c   |  2 +-
 3 files changed, 21 insertions(+), 21 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 31191804ca09..8f3e00c84f39 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -966,7 +966,7 @@ int dev_map_generic_redirect(struct bpf_dtab_netdev *dst, struct sk_buff *skb,
 			     struct bpf_prog *xdp_prog);
 
 struct bpf_cpu_map_entry *__cpu_map_lookup_elem(struct bpf_map *map, u32 key);
-void __cpu_map_flush(struct bpf_map *map);
+void __cpu_map_flush(void);
 int cpu_map_enqueue(struct bpf_cpu_map_entry *rcpu, struct xdp_buff *xdp,
 		    struct net_device *dev_rx);
 
@@ -1097,7 +1097,7 @@ struct bpf_cpu_map_entry *__cpu_map_lookup_elem(struct bpf_map *map, u32 key)
 	return NULL;
 }
 
-static inline void __cpu_map_flush(struct bpf_map *map)
+static inline void __cpu_map_flush(void)
 {
 }
 
diff --git a/kernel/bpf/cpumap.c b/kernel/bpf/cpumap.c
index f9deed659798..70f71b154fa5 100644
--- a/kernel/bpf/cpumap.c
+++ b/kernel/bpf/cpumap.c
@@ -72,17 +72,18 @@ struct bpf_cpu_map {
 	struct bpf_map map;
 	/* Below members specific for map type */
 	struct bpf_cpu_map_entry **cpu_map;
-	struct list_head __percpu *flush_list;
 };
 
+static DEFINE_PER_CPU(struct list_head, cpu_map_flush_list);
+
 static int bq_flush_to_queue(struct xdp_bulk_queue *bq);
 
 static struct bpf_map *cpu_map_alloc(union bpf_attr *attr)
 {
 	struct bpf_cpu_map *cmap;
 	int err = -ENOMEM;
-	int ret, cpu;
 	u64 cost;
+	int ret;
 
 	if (!capable(CAP_SYS_ADMIN))
 		return ERR_PTR(-EPERM);
@@ -106,7 +107,6 @@ static struct bpf_map *cpu_map_alloc(union bpf_attr *attr)
 
 	/* make sure page count doesn't overflow */
 	cost = (u64) cmap->map.max_entries * sizeof(struct bpf_cpu_map_entry *);
-	cost += sizeof(struct list_head) * num_possible_cpus();
 
 	/* Notice returns -EPERM on if map size is larger than memlock limit */
 	ret = bpf_map_charge_init(&cmap->map.memory, cost);
@@ -115,23 +115,14 @@ static struct bpf_map *cpu_map_alloc(union bpf_attr *attr)
 		goto free_cmap;
 	}
 
-	cmap->flush_list = alloc_percpu(struct list_head);
-	if (!cmap->flush_list)
-		goto free_charge;
-
-	for_each_possible_cpu(cpu)
-		INIT_LIST_HEAD(per_cpu_ptr(cmap->flush_list, cpu));
-
 	/* Alloc array for possible remote "destination" CPUs */
 	cmap->cpu_map = bpf_map_area_alloc(cmap->map.max_entries *
 					   sizeof(struct bpf_cpu_map_entry *),
 					   cmap->map.numa_node);
 	if (!cmap->cpu_map)
-		goto free_percpu;
+		goto free_charge;
 
 	return &cmap->map;
-free_percpu:
-	free_percpu(cmap->flush_list);
 free_charge:
 	bpf_map_charge_finish(&cmap->map.memory);
 free_cmap:
@@ -526,7 +517,6 @@ static void cpu_map_free(struct bpf_map *map)
 		/* bq flush and cleanup happens after RCU grace-period */
 		__cpu_map_entry_replace(cmap, i, NULL); /* call_rcu */
 	}
-	free_percpu(cmap->flush_list);
 	bpf_map_area_free(cmap->cpu_map);
 	kfree(cmap);
 }
@@ -618,7 +608,7 @@ static int bq_flush_to_queue(struct xdp_bulk_queue *bq)
  */
 static int bq_enqueue(struct bpf_cpu_map_entry *rcpu, struct xdp_frame *xdpf)
 {
-	struct list_head *flush_list = this_cpu_ptr(rcpu->cmap->flush_list);
+	struct list_head *flush_list = this_cpu_ptr(&cpu_map_flush_list);
 	struct xdp_bulk_queue *bq = this_cpu_ptr(rcpu->bulkq);
 
 	if (unlikely(bq->count == CPU_MAP_BULK_SIZE))
@@ -657,10 +647,9 @@ int cpu_map_enqueue(struct bpf_cpu_map_entry *rcpu, struct xdp_buff *xdp,
 	return 0;
 }
 
-void __cpu_map_flush(struct bpf_map *map)
+void __cpu_map_flush(void)
 {
-	struct bpf_cpu_map *cmap = container_of(map, struct bpf_cpu_map, map);
-	struct list_head *flush_list = this_cpu_ptr(cmap->flush_list);
+	struct list_head *flush_list = this_cpu_ptr(&cpu_map_flush_list);
 	struct xdp_bulk_queue *bq, *tmp;
 
 	list_for_each_entry_safe(bq, tmp, flush_list, flush_node) {
@@ -670,3 +659,14 @@ void __cpu_map_flush(struct bpf_map *map)
 		wake_up_process(bq->obj->kthread);
 	}
 }
+
+static int __init cpu_map_init(void)
+{
+	int cpu;
+
+	for_each_possible_cpu(cpu)
+		INIT_LIST_HEAD(&per_cpu(cpu_map_flush_list, cpu));
+	return 0;
+}
+
+subsys_initcall(cpu_map_init);
diff --git a/net/core/filter.c b/net/core/filter.c
index b7570cb84902..c706325b3e66 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -3558,7 +3558,7 @@ void xdp_do_flush_map(void)
 			__dev_map_flush();
 			break;
 		case BPF_MAP_TYPE_CPUMAP:
-			__cpu_map_flush(map);
+			__cpu_map_flush();
 			break;
 		case BPF_MAP_TYPE_XSKMAP:
 			__xsk_map_flush();
-- 
cgit v1.2.3-59-g8ed1b


From 332f22a60e4c3492d4953cd6f7aaa4e8bd0bba97 Mon Sep 17 00:00:00 2001
From: Björn Töpel <bjorn.topel@intel.com>
Date: Thu, 19 Dec 2019 07:10:05 +0100
Subject: xdp: Remove map_to_flush and map swap detection
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Now that all XDP maps that can be used with bpf_redirect_map() tracks
entries to be flushed in a global fashion, there is not need to track
that the map has changed and flush from xdp_do_generic_map()
anymore. All entries will be flushed in xdp_do_flush_map().

This means that the map_to_flush can be removed, and the corresponding
checks. Moving the flush logic to one place, xdp_do_flush_map(), give
a bulking behavior and performance boost.

Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/bpf/20191219061006.21980-8-bjorn.topel@gmail.com
---
 include/linux/filter.h |  1 -
 net/core/filter.c      | 27 +++------------------------
 2 files changed, 3 insertions(+), 25 deletions(-)

diff --git a/include/linux/filter.h b/include/linux/filter.h
index 37ac7025031d..69d6706fc889 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h
@@ -592,7 +592,6 @@ struct bpf_redirect_info {
 	u32 tgt_index;
 	void *tgt_value;
 	struct bpf_map *map;
-	struct bpf_map *map_to_flush;
 	u32 kern_flags;
 };
 
diff --git a/net/core/filter.c b/net/core/filter.c
index c706325b3e66..d9caa3e57ea1 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -3547,26 +3547,9 @@ static int __bpf_tx_xdp_map(struct net_device *dev_rx, void *fwd,
 
 void xdp_do_flush_map(void)
 {
-	struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info);
-	struct bpf_map *map = ri->map_to_flush;
-
-	ri->map_to_flush = NULL;
-	if (map) {
-		switch (map->map_type) {
-		case BPF_MAP_TYPE_DEVMAP:
-		case BPF_MAP_TYPE_DEVMAP_HASH:
-			__dev_map_flush();
-			break;
-		case BPF_MAP_TYPE_CPUMAP:
-			__cpu_map_flush();
-			break;
-		case BPF_MAP_TYPE_XSKMAP:
-			__xsk_map_flush();
-			break;
-		default:
-			break;
-		}
-	}
+	__dev_map_flush();
+	__cpu_map_flush();
+	__xsk_map_flush();
 }
 EXPORT_SYMBOL_GPL(xdp_do_flush_map);
 
@@ -3615,14 +3598,10 @@ static int xdp_do_redirect_map(struct net_device *dev, struct xdp_buff *xdp,
 	ri->tgt_value = NULL;
 	WRITE_ONCE(ri->map, NULL);
 
-	if (ri->map_to_flush && unlikely(ri->map_to_flush != map))
-		xdp_do_flush_map();
-
 	err = __bpf_tx_xdp_map(dev, fwd, map, xdp);
 	if (unlikely(err))
 		goto err;
 
-	ri->map_to_flush = map;
 	_trace_xdp_redirect_map(dev, xdp_prog, fwd, map, index);
 	return 0;
 err:
-- 
cgit v1.2.3-59-g8ed1b


From 1170beaa3fa3c3381fd820e9d05b84d168fe1dab Mon Sep 17 00:00:00 2001
From: Björn Töpel <bjorn.topel@intel.com>
Date: Thu, 19 Dec 2019 07:10:06 +0100
Subject: xdp: Simplify __bpf_tx_xdp_map()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The explicit error checking is not needed. Simply return the error
instead.

Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/bpf/20191219061006.21980-9-bjorn.topel@gmail.com
---
 net/core/filter.c | 33 +++++++--------------------------
 1 file changed, 7 insertions(+), 26 deletions(-)

diff --git a/net/core/filter.c b/net/core/filter.c
index d9caa3e57ea1..217af9974c86 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -3510,35 +3510,16 @@ err:
 }
 
 static int __bpf_tx_xdp_map(struct net_device *dev_rx, void *fwd,
-			    struct bpf_map *map,
-			    struct xdp_buff *xdp)
+			    struct bpf_map *map, struct xdp_buff *xdp)
 {
-	int err;
-
 	switch (map->map_type) {
 	case BPF_MAP_TYPE_DEVMAP:
-	case BPF_MAP_TYPE_DEVMAP_HASH: {
-		struct bpf_dtab_netdev *dst = fwd;
-
-		err = dev_map_enqueue(dst, xdp, dev_rx);
-		if (unlikely(err))
-			return err;
-		break;
-	}
-	case BPF_MAP_TYPE_CPUMAP: {
-		struct bpf_cpu_map_entry *rcpu = fwd;
-
-		err = cpu_map_enqueue(rcpu, xdp, dev_rx);
-		if (unlikely(err))
-			return err;
-		break;
-	}
-	case BPF_MAP_TYPE_XSKMAP: {
-		struct xdp_sock *xs = fwd;
-
-		err = __xsk_map_redirect(xs, xdp);
-		return err;
-	}
+	case BPF_MAP_TYPE_DEVMAP_HASH:
+		return dev_map_enqueue(fwd, xdp, dev_rx);
+	case BPF_MAP_TYPE_CPUMAP:
+		return cpu_map_enqueue(fwd, xdp, dev_rx);
+	case BPF_MAP_TYPE_XSKMAP:
+		return __xsk_map_redirect(fwd, xdp);
 	default:
 		break;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 1020c1f24a946e7d5d8a67db741b20efcd2cefc5 Mon Sep 17 00:00:00 2001
From: Andrey Ignatov <rdna@fb.com>
Date: Wed, 18 Dec 2019 23:44:33 -0800
Subject: bpf: Simplify __cgroup_bpf_attach

__cgroup_bpf_attach has a lot of identical code to handle two scenarios:
BPF_F_ALLOW_MULTI is set and unset.

Simplify it by splitting the two main steps:

* First, the decision is made whether a new bpf_prog_list entry should
  be allocated or existing entry should be reused for the new program.
  This decision is saved in replace_pl pointer;

* Next, replace_pl pointer is used to handle both possible states of
  BPF_F_ALLOW_MULTI flag (set / unset) instead of doing similar work for
  them separately.

This splitting, in turn, allows to make further simplifications:

* The check for attaching same program twice in BPF_F_ALLOW_MULTI mode
  can be done before allocating cgroup storage, so that if user tries to
  attach same program twice no alloc/free happens as it was before;

* pl_was_allocated becomes redundant so it's removed.

Signed-off-by: Andrey Ignatov <rdna@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/c6193db6fe630797110b0d3ff06c125d093b834c.1576741281.git.rdna@fb.com
---
 kernel/bpf/cgroup.c | 62 ++++++++++++++++++++---------------------------------
 1 file changed, 23 insertions(+), 39 deletions(-)

diff --git a/kernel/bpf/cgroup.c b/kernel/bpf/cgroup.c
index 9f90d3c92bda..e8cbdd1be687 100644
--- a/kernel/bpf/cgroup.c
+++ b/kernel/bpf/cgroup.c
@@ -295,9 +295,8 @@ int __cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog,
 	struct bpf_prog *old_prog = NULL;
 	struct bpf_cgroup_storage *storage[MAX_BPF_CGROUP_STORAGE_TYPE],
 		*old_storage[MAX_BPF_CGROUP_STORAGE_TYPE] = {NULL};
+	struct bpf_prog_list *pl, *replace_pl = NULL;
 	enum bpf_cgroup_storage_type stype;
-	struct bpf_prog_list *pl;
-	bool pl_was_allocated;
 	int err;
 
 	if ((flags & BPF_F_ALLOW_OVERRIDE) && (flags & BPF_F_ALLOW_MULTI))
@@ -317,6 +316,16 @@ int __cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog,
 	if (prog_list_length(progs) >= BPF_CGROUP_MAX_PROGS)
 		return -E2BIG;
 
+	if (flags & BPF_F_ALLOW_MULTI) {
+		list_for_each_entry(pl, progs, node) {
+			if (pl->prog == prog)
+				/* disallow attaching the same prog twice */
+				return -EINVAL;
+		}
+	} else if (!list_empty(progs)) {
+		replace_pl = list_first_entry(progs, typeof(*pl), node);
+	}
+
 	for_each_cgroup_storage_type(stype) {
 		storage[stype] = bpf_cgroup_storage_alloc(prog, stype);
 		if (IS_ERR(storage[stype])) {
@@ -327,52 +336,27 @@ int __cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog,
 		}
 	}
 
-	if (flags & BPF_F_ALLOW_MULTI) {
-		list_for_each_entry(pl, progs, node) {
-			if (pl->prog == prog) {
-				/* disallow attaching the same prog twice */
-				for_each_cgroup_storage_type(stype)
-					bpf_cgroup_storage_free(storage[stype]);
-				return -EINVAL;
-			}
+	if (replace_pl) {
+		pl = replace_pl;
+		old_prog = pl->prog;
+		for_each_cgroup_storage_type(stype) {
+			old_storage[stype] = pl->storage[stype];
+			bpf_cgroup_storage_unlink(old_storage[stype]);
 		}
-
+	} else {
 		pl = kmalloc(sizeof(*pl), GFP_KERNEL);
 		if (!pl) {
 			for_each_cgroup_storage_type(stype)
 				bpf_cgroup_storage_free(storage[stype]);
 			return -ENOMEM;
 		}
-
-		pl_was_allocated = true;
-		pl->prog = prog;
-		for_each_cgroup_storage_type(stype)
-			pl->storage[stype] = storage[stype];
 		list_add_tail(&pl->node, progs);
-	} else {
-		if (list_empty(progs)) {
-			pl = kmalloc(sizeof(*pl), GFP_KERNEL);
-			if (!pl) {
-				for_each_cgroup_storage_type(stype)
-					bpf_cgroup_storage_free(storage[stype]);
-				return -ENOMEM;
-			}
-			pl_was_allocated = true;
-			list_add_tail(&pl->node, progs);
-		} else {
-			pl = list_first_entry(progs, typeof(*pl), node);
-			old_prog = pl->prog;
-			for_each_cgroup_storage_type(stype) {
-				old_storage[stype] = pl->storage[stype];
-				bpf_cgroup_storage_unlink(old_storage[stype]);
-			}
-			pl_was_allocated = false;
-		}
-		pl->prog = prog;
-		for_each_cgroup_storage_type(stype)
-			pl->storage[stype] = storage[stype];
 	}
 
+	pl->prog = prog;
+	for_each_cgroup_storage_type(stype)
+		pl->storage[stype] = storage[stype];
+
 	cgrp->bpf.flags[type] = flags;
 
 	err = update_effective_progs(cgrp, type);
@@ -401,7 +385,7 @@ cleanup:
 		pl->storage[stype] = old_storage[stype];
 		bpf_cgroup_storage_link(old_storage[stype], cgrp, type);
 	}
-	if (pl_was_allocated) {
+	if (!replace_pl) {
 		list_del(&pl->node);
 		kfree(pl);
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 9fab329d6a04c0a52a84d207b5e0d83aeb660aa0 Mon Sep 17 00:00:00 2001
From: Andrey Ignatov <rdna@fb.com>
Date: Wed, 18 Dec 2019 23:44:34 -0800
Subject: bpf: Remove unused new_flags in hierarchy_allows_attach()

new_flags is unused, remove it.

Signed-off-by: Andrey Ignatov <rdna@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/2c49b30ab750f93cfef04a1e40b097d70c3a39a1.1576741281.git.rdna@fb.com
---
 kernel/bpf/cgroup.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/kernel/bpf/cgroup.c b/kernel/bpf/cgroup.c
index e8cbdd1be687..283efe3ce052 100644
--- a/kernel/bpf/cgroup.c
+++ b/kernel/bpf/cgroup.c
@@ -103,8 +103,7 @@ static u32 prog_list_length(struct list_head *head)
  * if parent has overridable or multi-prog, allow attaching
  */
 static bool hierarchy_allows_attach(struct cgroup *cgrp,
-				    enum bpf_attach_type type,
-				    u32 new_flags)
+				    enum bpf_attach_type type)
 {
 	struct cgroup *p;
 
@@ -303,7 +302,7 @@ int __cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog,
 		/* invalid combination */
 		return -EINVAL;
 
-	if (!hierarchy_allows_attach(cgrp, type, flags))
+	if (!hierarchy_allows_attach(cgrp, type))
 		return -EPERM;
 
 	if (!list_empty(progs) && cgrp->bpf.flags[type] != flags)
-- 
cgit v1.2.3-59-g8ed1b


From 7dd68b3279f1792103d12e69933db3128c6d416e Mon Sep 17 00:00:00 2001
From: Andrey Ignatov <rdna@fb.com>
Date: Wed, 18 Dec 2019 23:44:35 -0800
Subject: bpf: Support replacing cgroup-bpf program in MULTI mode

The common use-case in production is to have multiple cgroup-bpf
programs per attach type that cover multiple use-cases. Such programs
are attached with BPF_F_ALLOW_MULTI and can be maintained by different
people.

Order of programs usually matters, for example imagine two egress
programs: the first one drops packets and the second one counts packets.
If they're swapped the result of counting program will be different.

It brings operational challenges with updating cgroup-bpf program(s)
attached with BPF_F_ALLOW_MULTI since there is no way to replace a
program:

* One way to update is to detach all programs first and then attach the
  new version(s) again in the right order. This introduces an
  interruption in the work a program is doing and may not be acceptable
  (e.g. if it's egress firewall);

* Another way is attach the new version of a program first and only then
  detach the old version. This introduces the time interval when two
  versions of same program are working, what may not be acceptable if a
  program is not idempotent. It also imposes additional burden on
  program developers to make sure that two versions of their program can
  co-exist.

Solve the problem by introducing a "replace" mode in BPF_PROG_ATTACH
command for cgroup-bpf programs being attached with BPF_F_ALLOW_MULTI
flag. This mode is enabled by newly introduced BPF_F_REPLACE attach flag
and bpf_attr.replace_bpf_fd attribute to pass fd of the old program to
replace

That way user can replace any program among those attached with
BPF_F_ALLOW_MULTI flag without the problems described above.

Details of the new API:

* If BPF_F_REPLACE is set but replace_bpf_fd doesn't have valid
  descriptor of BPF program, BPF_PROG_ATTACH will return corresponding
  error (EINVAL or EBADF).

* If replace_bpf_fd has valid descriptor of BPF program but such a
  program is not attached to specified cgroup, BPF_PROG_ATTACH will
  return ENOENT.

BPF_F_REPLACE is introduced to make the user intent clear, since
replace_bpf_fd alone can't be used for this (its default value, 0, is a
valid fd). BPF_F_REPLACE also makes it possible to extend the API in the
future (e.g. add BPF_F_BEFORE and BPF_F_AFTER if needed).

Signed-off-by: Andrey Ignatov <rdna@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Acked-by: Andrii Narkyiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/30cd850044a0057bdfcaaf154b7d2f39850ba813.1576741281.git.rdna@fb.com
---
 include/linux/bpf-cgroup.h     |  4 +++-
 include/uapi/linux/bpf.h       | 10 ++++++++++
 kernel/bpf/cgroup.c            | 30 ++++++++++++++++++++++++++----
 kernel/bpf/syscall.c           |  4 ++--
 kernel/cgroup/cgroup.c         |  5 +++--
 tools/include/uapi/linux/bpf.h | 10 ++++++++++
 6 files changed, 54 insertions(+), 9 deletions(-)

diff --git a/include/linux/bpf-cgroup.h b/include/linux/bpf-cgroup.h
index 169fd25f6bc2..18f6a6da7c3c 100644
--- a/include/linux/bpf-cgroup.h
+++ b/include/linux/bpf-cgroup.h
@@ -85,6 +85,7 @@ int cgroup_bpf_inherit(struct cgroup *cgrp);
 void cgroup_bpf_offline(struct cgroup *cgrp);
 
 int __cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog,
+			struct bpf_prog *replace_prog,
 			enum bpf_attach_type type, u32 flags);
 int __cgroup_bpf_detach(struct cgroup *cgrp, struct bpf_prog *prog,
 			enum bpf_attach_type type);
@@ -93,7 +94,8 @@ int __cgroup_bpf_query(struct cgroup *cgrp, const union bpf_attr *attr,
 
 /* Wrapper for __cgroup_bpf_*() protected by cgroup_mutex */
 int cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog,
-		      enum bpf_attach_type type, u32 flags);
+		      struct bpf_prog *replace_prog, enum bpf_attach_type type,
+		      u32 flags);
 int cgroup_bpf_detach(struct cgroup *cgrp, struct bpf_prog *prog,
 		      enum bpf_attach_type type, u32 flags);
 int cgroup_bpf_query(struct cgroup *cgrp, const union bpf_attr *attr,
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index dbbcf0b02970..7df436da542d 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -231,6 +231,11 @@ enum bpf_attach_type {
  * When children program makes decision (like picking TCP CA or sock bind)
  * parent program has a chance to override it.
  *
+ * With BPF_F_ALLOW_MULTI a new program is added to the end of the list of
+ * programs for a cgroup. Though it's possible to replace an old program at
+ * any position by also specifying BPF_F_REPLACE flag and position itself in
+ * replace_bpf_fd attribute. Old program at this position will be released.
+ *
  * A cgroup with MULTI or OVERRIDE flag allows any attach flags in sub-cgroups.
  * A cgroup with NONE doesn't allow any programs in sub-cgroups.
  * Ex1:
@@ -249,6 +254,7 @@ enum bpf_attach_type {
  */
 #define BPF_F_ALLOW_OVERRIDE	(1U << 0)
 #define BPF_F_ALLOW_MULTI	(1U << 1)
+#define BPF_F_REPLACE		(1U << 2)
 
 /* If BPF_F_STRICT_ALIGNMENT is used in BPF_PROG_LOAD command, the
  * verifier will perform strict alignment checking as if the kernel
@@ -442,6 +448,10 @@ union bpf_attr {
 		__u32		attach_bpf_fd;	/* eBPF program to attach */
 		__u32		attach_type;
 		__u32		attach_flags;
+		__u32		replace_bpf_fd;	/* previously attached eBPF
+						 * program to replace if
+						 * BPF_F_REPLACE is used
+						 */
 	};
 
 	struct { /* anonymous struct used by BPF_PROG_TEST_RUN command */
diff --git a/kernel/bpf/cgroup.c b/kernel/bpf/cgroup.c
index 283efe3ce052..45346c79613a 100644
--- a/kernel/bpf/cgroup.c
+++ b/kernel/bpf/cgroup.c
@@ -282,14 +282,17 @@ cleanup:
  *                         propagate the change to descendants
  * @cgrp: The cgroup which descendants to traverse
  * @prog: A program to attach
+ * @replace_prog: Previously attached program to replace if BPF_F_REPLACE is set
  * @type: Type of attach operation
  * @flags: Option flags
  *
  * Must be called with cgroup_mutex held.
  */
 int __cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog,
+			struct bpf_prog *replace_prog,
 			enum bpf_attach_type type, u32 flags)
 {
+	u32 saved_flags = (flags & (BPF_F_ALLOW_OVERRIDE | BPF_F_ALLOW_MULTI));
 	struct list_head *progs = &cgrp->bpf.progs[type];
 	struct bpf_prog *old_prog = NULL;
 	struct bpf_cgroup_storage *storage[MAX_BPF_CGROUP_STORAGE_TYPE],
@@ -298,14 +301,15 @@ int __cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog,
 	enum bpf_cgroup_storage_type stype;
 	int err;
 
-	if ((flags & BPF_F_ALLOW_OVERRIDE) && (flags & BPF_F_ALLOW_MULTI))
+	if (((flags & BPF_F_ALLOW_OVERRIDE) && (flags & BPF_F_ALLOW_MULTI)) ||
+	    ((flags & BPF_F_REPLACE) && !(flags & BPF_F_ALLOW_MULTI)))
 		/* invalid combination */
 		return -EINVAL;
 
 	if (!hierarchy_allows_attach(cgrp, type))
 		return -EPERM;
 
-	if (!list_empty(progs) && cgrp->bpf.flags[type] != flags)
+	if (!list_empty(progs) && cgrp->bpf.flags[type] != saved_flags)
 		/* Disallow attaching non-overridable on top
 		 * of existing overridable in this cgroup.
 		 * Disallow attaching multi-prog if overridable or none
@@ -320,7 +324,12 @@ int __cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog,
 			if (pl->prog == prog)
 				/* disallow attaching the same prog twice */
 				return -EINVAL;
+			if (pl->prog == replace_prog)
+				replace_pl = pl;
 		}
+		if ((flags & BPF_F_REPLACE) && !replace_pl)
+			/* prog to replace not found for cgroup */
+			return -ENOENT;
 	} else if (!list_empty(progs)) {
 		replace_pl = list_first_entry(progs, typeof(*pl), node);
 	}
@@ -356,7 +365,7 @@ int __cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog,
 	for_each_cgroup_storage_type(stype)
 		pl->storage[stype] = storage[stype];
 
-	cgrp->bpf.flags[type] = flags;
+	cgrp->bpf.flags[type] = saved_flags;
 
 	err = update_effective_progs(cgrp, type);
 	if (err)
@@ -522,6 +531,7 @@ int __cgroup_bpf_query(struct cgroup *cgrp, const union bpf_attr *attr,
 int cgroup_bpf_prog_attach(const union bpf_attr *attr,
 			   enum bpf_prog_type ptype, struct bpf_prog *prog)
 {
+	struct bpf_prog *replace_prog = NULL;
 	struct cgroup *cgrp;
 	int ret;
 
@@ -529,8 +539,20 @@ int cgroup_bpf_prog_attach(const union bpf_attr *attr,
 	if (IS_ERR(cgrp))
 		return PTR_ERR(cgrp);
 
-	ret = cgroup_bpf_attach(cgrp, prog, attr->attach_type,
+	if ((attr->attach_flags & BPF_F_ALLOW_MULTI) &&
+	    (attr->attach_flags & BPF_F_REPLACE)) {
+		replace_prog = bpf_prog_get_type(attr->replace_bpf_fd, ptype);
+		if (IS_ERR(replace_prog)) {
+			cgroup_put(cgrp);
+			return PTR_ERR(replace_prog);
+		}
+	}
+
+	ret = cgroup_bpf_attach(cgrp, prog, replace_prog, attr->attach_type,
 				attr->attach_flags);
+
+	if (replace_prog)
+		bpf_prog_put(replace_prog);
 	cgroup_put(cgrp);
 	return ret;
 }
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index b08c362f4e02..81ee8595dfee 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -2073,10 +2073,10 @@ static int bpf_prog_attach_check_attach_type(const struct bpf_prog *prog,
 	}
 }
 
-#define BPF_PROG_ATTACH_LAST_FIELD attach_flags
+#define BPF_PROG_ATTACH_LAST_FIELD replace_bpf_fd
 
 #define BPF_F_ATTACH_MASK \
-	(BPF_F_ALLOW_OVERRIDE | BPF_F_ALLOW_MULTI)
+	(BPF_F_ALLOW_OVERRIDE | BPF_F_ALLOW_MULTI | BPF_F_REPLACE)
 
 static int bpf_prog_attach(const union bpf_attr *attr)
 {
diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
index 735af8f15f95..725365df066d 100644
--- a/kernel/cgroup/cgroup.c
+++ b/kernel/cgroup/cgroup.c
@@ -6288,12 +6288,13 @@ void cgroup_sk_free(struct sock_cgroup_data *skcd)
 
 #ifdef CONFIG_CGROUP_BPF
 int cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog,
-		      enum bpf_attach_type type, u32 flags)
+		      struct bpf_prog *replace_prog, enum bpf_attach_type type,
+		      u32 flags)
 {
 	int ret;
 
 	mutex_lock(&cgroup_mutex);
-	ret = __cgroup_bpf_attach(cgrp, prog, type, flags);
+	ret = __cgroup_bpf_attach(cgrp, prog, replace_prog, type, flags);
 	mutex_unlock(&cgroup_mutex);
 	return ret;
 }
diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index dbbcf0b02970..7df436da542d 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -231,6 +231,11 @@ enum bpf_attach_type {
  * When children program makes decision (like picking TCP CA or sock bind)
  * parent program has a chance to override it.
  *
+ * With BPF_F_ALLOW_MULTI a new program is added to the end of the list of
+ * programs for a cgroup. Though it's possible to replace an old program at
+ * any position by also specifying BPF_F_REPLACE flag and position itself in
+ * replace_bpf_fd attribute. Old program at this position will be released.
+ *
  * A cgroup with MULTI or OVERRIDE flag allows any attach flags in sub-cgroups.
  * A cgroup with NONE doesn't allow any programs in sub-cgroups.
  * Ex1:
@@ -249,6 +254,7 @@ enum bpf_attach_type {
  */
 #define BPF_F_ALLOW_OVERRIDE	(1U << 0)
 #define BPF_F_ALLOW_MULTI	(1U << 1)
+#define BPF_F_REPLACE		(1U << 2)
 
 /* If BPF_F_STRICT_ALIGNMENT is used in BPF_PROG_LOAD command, the
  * verifier will perform strict alignment checking as if the kernel
@@ -442,6 +448,10 @@ union bpf_attr {
 		__u32		attach_bpf_fd;	/* eBPF program to attach */
 		__u32		attach_type;
 		__u32		attach_flags;
+		__u32		replace_bpf_fd;	/* previously attached eBPF
+						 * program to replace if
+						 * BPF_F_REPLACE is used
+						 */
 	};
 
 	struct { /* anonymous struct used by BPF_PROG_TEST_RUN command */
-- 
cgit v1.2.3-59-g8ed1b


From cdbee3839cd91a4577425c43cf064b9523926159 Mon Sep 17 00:00:00 2001
From: Andrey Ignatov <rdna@fb.com>
Date: Wed, 18 Dec 2019 23:44:36 -0800
Subject: libbpf: Introduce bpf_prog_attach_xattr

Introduce a new bpf_prog_attach_xattr function that, in addition to
program fd, target fd and attach type, accepts an extendable struct
bpf_prog_attach_opts.

bpf_prog_attach_opts relies on DECLARE_LIBBPF_OPTS macro to maintain
backward and forward compatibility and has the following "optional"
attach attributes:

* existing attach_flags, since it's not required when attaching in NONE
  mode. Even though it's quite often used in MULTI and OVERRIDE mode it
  seems to be a good idea to reduce number of arguments to
  bpf_prog_attach_xattr;

* newly introduced attribute of BPF_PROG_ATTACH command: replace_prog_fd
  that is fd of previously attached cgroup-bpf program to replace if
  BPF_F_REPLACE flag is used.

The new function is named to be consistent with other xattr-functions
(bpf_prog_test_run_xattr, bpf_create_map_xattr, bpf_load_program_xattr).

The struct bpf_prog_attach_opts is supposed to be used with
DECLARE_LIBBPF_OPTS macro.

Signed-off-by: Andrey Ignatov <rdna@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/bd6e0732303eb14e4b79cb128268d9e9ad6db208.1576741281.git.rdna@fb.com
---
 tools/lib/bpf/bpf.c      | 17 ++++++++++++++++-
 tools/lib/bpf/bpf.h      | 11 +++++++++++
 tools/lib/bpf/libbpf.map |  1 +
 3 files changed, 28 insertions(+), 1 deletion(-)

diff --git a/tools/lib/bpf/bpf.c b/tools/lib/bpf/bpf.c
index 98596e15390f..a787d53699c8 100644
--- a/tools/lib/bpf/bpf.c
+++ b/tools/lib/bpf/bpf.c
@@ -466,14 +466,29 @@ int bpf_obj_get(const char *pathname)
 
 int bpf_prog_attach(int prog_fd, int target_fd, enum bpf_attach_type type,
 		    unsigned int flags)
+{
+	DECLARE_LIBBPF_OPTS(bpf_prog_attach_opts, opts,
+		.flags = flags,
+	);
+
+	return bpf_prog_attach_xattr(prog_fd, target_fd, type, &opts);
+}
+
+int bpf_prog_attach_xattr(int prog_fd, int target_fd,
+			  enum bpf_attach_type type,
+			  const struct bpf_prog_attach_opts *opts)
 {
 	union bpf_attr attr;
 
+	if (!OPTS_VALID(opts, bpf_prog_attach_opts))
+		return -EINVAL;
+
 	memset(&attr, 0, sizeof(attr));
 	attr.target_fd	   = target_fd;
 	attr.attach_bpf_fd = prog_fd;
 	attr.attach_type   = type;
-	attr.attach_flags  = flags;
+	attr.attach_flags  = OPTS_GET(opts, flags, 0);
+	attr.replace_bpf_fd = OPTS_GET(opts, replace_prog_fd, 0);
 
 	return sys_bpf(BPF_PROG_ATTACH, &attr, sizeof(attr));
 }
diff --git a/tools/lib/bpf/bpf.h b/tools/lib/bpf/bpf.h
index 269807ce9ef5..f0ab8519986e 100644
--- a/tools/lib/bpf/bpf.h
+++ b/tools/lib/bpf/bpf.h
@@ -126,8 +126,19 @@ LIBBPF_API int bpf_map_get_next_key(int fd, const void *key, void *next_key);
 LIBBPF_API int bpf_map_freeze(int fd);
 LIBBPF_API int bpf_obj_pin(int fd, const char *pathname);
 LIBBPF_API int bpf_obj_get(const char *pathname);
+
+struct bpf_prog_attach_opts {
+	size_t sz; /* size of this struct for forward/backward compatibility */
+	unsigned int flags;
+	int replace_prog_fd;
+};
+#define bpf_prog_attach_opts__last_field replace_prog_fd
+
 LIBBPF_API int bpf_prog_attach(int prog_fd, int attachable_fd,
 			       enum bpf_attach_type type, unsigned int flags);
+LIBBPF_API int bpf_prog_attach_xattr(int prog_fd, int attachable_fd,
+				     enum bpf_attach_type type,
+				     const struct bpf_prog_attach_opts *opts);
 LIBBPF_API int bpf_prog_detach(int attachable_fd, enum bpf_attach_type type);
 LIBBPF_API int bpf_prog_detach2(int prog_fd, int attachable_fd,
 				enum bpf_attach_type type);
diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
index e3a471f38a71..e9713a574243 100644
--- a/tools/lib/bpf/libbpf.map
+++ b/tools/lib/bpf/libbpf.map
@@ -219,6 +219,7 @@ LIBBPF_0.0.7 {
 		bpf_object__detach_skeleton;
 		bpf_object__load_skeleton;
 		bpf_object__open_skeleton;
+		bpf_prog_attach_xattr;
 		bpf_program__attach;
 		bpf_program__name;
 		btf__align_of;
-- 
cgit v1.2.3-59-g8ed1b


From 257c88559f360ef40d251942f1f9f0c55f5f91ca Mon Sep 17 00:00:00 2001
From: Andrey Ignatov <rdna@fb.com>
Date: Wed, 18 Dec 2019 23:44:37 -0800
Subject: selftests/bpf: Convert test_cgroup_attach to prog_tests

Convert test_cgroup_attach to prog_tests.

This change does a lot of things but in many cases it's pretty expensive
to separate them, so they go in one commit. Nevertheless the logic is
ketp as is and changes made are just moving things around, simplifying
them (w/o changing the meaning of the tests) and making prog_tests
compatible:

* split the 3 tests in the file into 3 separate files in prog_tests/;

* rename the test functions to test_<file_base_name>;

* remove unused includes, constants, variables and functions from every
  test;

* replace `if`-s with or `if (CHECK())` where additional context should
  be logged and with `if (CHECK_FAIL())` where line number is enough;

* switch from `log_err()` to logging via `CHECK()`;

* replace `assert`-s with `CHECK_FAIL()` to avoid crashing the whole
  test_progs if one assertion fails;

* replace cgroup_helpers with test__join_cgroup() in
  cgroup_attach_override only, other tests need more fine-grained
  control for cgroup creation/deletion so cgroup_helpers are still used
  there;

* simplify cgroup_attach_autodetach by switching to easiest possible
  program since this test doesn't really need such a complicated program
  as cgroup_attach_multi does;

* remove test_cgroup_attach.c itself.

Signed-off-by: Andrey Ignatov <rdna@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/0ff19cc64d2dc5cf404349f07131119480e10e32.1576741281.git.rdna@fb.com
---
 tools/testing/selftests/bpf/.gitignore             |   1 -
 tools/testing/selftests/bpf/Makefile               |   3 +-
 .../bpf/prog_tests/cgroup_attach_autodetach.c      | 111 ++++
 .../selftests/bpf/prog_tests/cgroup_attach_multi.c | 238 +++++++++
 .../bpf/prog_tests/cgroup_attach_override.c        | 148 ++++++
 tools/testing/selftests/bpf/test_cgroup_attach.c   | 571 ---------------------
 6 files changed, 498 insertions(+), 574 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/cgroup_attach_autodetach.c
 create mode 100644 tools/testing/selftests/bpf/prog_tests/cgroup_attach_multi.c
 create mode 100644 tools/testing/selftests/bpf/prog_tests/cgroup_attach_override.c
 delete mode 100644 tools/testing/selftests/bpf/test_cgroup_attach.c

diff --git a/tools/testing/selftests/bpf/.gitignore b/tools/testing/selftests/bpf/.gitignore
index ce5af95ede42..b139b3d75ebb 100644
--- a/tools/testing/selftests/bpf/.gitignore
+++ b/tools/testing/selftests/bpf/.gitignore
@@ -21,7 +21,6 @@ test_lirc_mode2_user
 get_cgroup_id_user
 test_skb_cgroup_id_user
 test_socket_cookie
-test_cgroup_attach
 test_cgroup_storage
 test_select_reuseport
 test_flow_dissector
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index c652bd84ef0e..866fc1cadd7c 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -32,7 +32,7 @@ TEST_GEN_PROGS = test_verifier test_tag test_maps test_lru_map test_lpm_map test
 	test_sock test_btf test_sockmap get_cgroup_id_user test_socket_cookie \
 	test_cgroup_storage \
 	test_netcnt test_tcpnotify_user test_sock_fields test_sysctl test_hashmap \
-	test_cgroup_attach test_progs-no_alu32
+	test_progs-no_alu32
 
 # Also test bpf-gcc, if present
 ifneq ($(BPF_GCC),)
@@ -136,7 +136,6 @@ $(OUTPUT)/test_cgroup_storage: cgroup_helpers.c
 $(OUTPUT)/test_netcnt: cgroup_helpers.c
 $(OUTPUT)/test_sock_fields: cgroup_helpers.c
 $(OUTPUT)/test_sysctl: cgroup_helpers.c
-$(OUTPUT)/test_cgroup_attach: cgroup_helpers.c
 
 .PHONY: force
 
diff --git a/tools/testing/selftests/bpf/prog_tests/cgroup_attach_autodetach.c b/tools/testing/selftests/bpf/prog_tests/cgroup_attach_autodetach.c
new file mode 100644
index 000000000000..5b13f2c6c402
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/cgroup_attach_autodetach.c
@@ -0,0 +1,111 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <test_progs.h>
+
+#include "cgroup_helpers.h"
+
+#define PING_CMD	"ping -q -c1 -w1 127.0.0.1 > /dev/null"
+
+char bpf_log_buf[BPF_LOG_BUF_SIZE];
+
+static int prog_load(void)
+{
+	struct bpf_insn prog[] = {
+		BPF_MOV64_IMM(BPF_REG_0, 1), /* r0 = 1 */
+		BPF_EXIT_INSN(),
+	};
+	size_t insns_cnt = sizeof(prog) / sizeof(struct bpf_insn);
+
+	return bpf_load_program(BPF_PROG_TYPE_CGROUP_SKB,
+			       prog, insns_cnt, "GPL", 0,
+			       bpf_log_buf, BPF_LOG_BUF_SIZE);
+}
+
+void test_cgroup_attach_autodetach(void)
+{
+	__u32 duration = 0, prog_cnt = 4, attach_flags;
+	int allow_prog[2] = {-1};
+	__u32 prog_ids[2] = {0};
+	void *ptr = NULL;
+	int cg = 0, i;
+	int attempts;
+
+	for (i = 0; i < ARRAY_SIZE(allow_prog); i++) {
+		allow_prog[i] = prog_load();
+		if (CHECK(allow_prog[i] < 0, "prog_load",
+			  "verifier output:\n%s\n-------\n", bpf_log_buf))
+			goto err;
+	}
+
+	if (CHECK_FAIL(setup_cgroup_environment()))
+		goto err;
+
+	/* create a cgroup, attach two programs and remember their ids */
+	cg = create_and_get_cgroup("/cg_autodetach");
+	if (CHECK_FAIL(cg < 0))
+		goto err;
+
+	if (CHECK_FAIL(join_cgroup("/cg_autodetach")))
+		goto err;
+
+	for (i = 0; i < ARRAY_SIZE(allow_prog); i++)
+		if (CHECK(bpf_prog_attach(allow_prog[i], cg,
+					  BPF_CGROUP_INET_EGRESS,
+					  BPF_F_ALLOW_MULTI),
+			  "prog_attach", "prog[%d], errno=%d\n", i, errno))
+			goto err;
+
+	/* make sure that programs are attached and run some traffic */
+	if (CHECK(bpf_prog_query(cg, BPF_CGROUP_INET_EGRESS, 0, &attach_flags,
+				 prog_ids, &prog_cnt),
+		  "prog_query", "errno=%d\n", errno))
+		goto err;
+	if (CHECK_FAIL(system(PING_CMD)))
+		goto err;
+
+	/* allocate some memory (4Mb) to pin the original cgroup */
+	ptr = malloc(4 * (1 << 20));
+	if (CHECK_FAIL(!ptr))
+		goto err;
+
+	/* close programs and cgroup fd */
+	for (i = 0; i < ARRAY_SIZE(allow_prog); i++) {
+		close(allow_prog[i]);
+		allow_prog[i] = -1;
+	}
+
+	close(cg);
+	cg = 0;
+
+	/* leave the cgroup and remove it. don't detach programs */
+	cleanup_cgroup_environment();
+
+	/* wait for the asynchronous auto-detachment.
+	 * wait for no more than 5 sec and give up.
+	 */
+	for (i = 0; i < ARRAY_SIZE(prog_ids); i++) {
+		for (attempts = 5; attempts >= 0; attempts--) {
+			int fd = bpf_prog_get_fd_by_id(prog_ids[i]);
+
+			if (fd < 0)
+				break;
+
+			/* don't leave the fd open */
+			close(fd);
+
+			if (CHECK_FAIL(!attempts))
+				goto err;
+
+			sleep(1);
+		}
+	}
+
+err:
+	for (i = 0; i < ARRAY_SIZE(allow_prog); i++)
+		if (allow_prog[i] >= 0)
+			close(allow_prog[i]);
+	if (cg)
+		close(cg);
+	free(ptr);
+	cleanup_cgroup_environment();
+}
diff --git a/tools/testing/selftests/bpf/prog_tests/cgroup_attach_multi.c b/tools/testing/selftests/bpf/prog_tests/cgroup_attach_multi.c
new file mode 100644
index 000000000000..4eaab7435044
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/cgroup_attach_multi.c
@@ -0,0 +1,238 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <test_progs.h>
+
+#include "cgroup_helpers.h"
+
+#define PING_CMD	"ping -q -c1 -w1 127.0.0.1 > /dev/null"
+
+char bpf_log_buf[BPF_LOG_BUF_SIZE];
+
+static int map_fd = -1;
+
+static int prog_load_cnt(int verdict, int val)
+{
+	int cgroup_storage_fd, percpu_cgroup_storage_fd;
+
+	if (map_fd < 0)
+		map_fd = bpf_create_map(BPF_MAP_TYPE_ARRAY, 4, 8, 1, 0);
+	if (map_fd < 0) {
+		printf("failed to create map '%s'\n", strerror(errno));
+		return -1;
+	}
+
+	cgroup_storage_fd = bpf_create_map(BPF_MAP_TYPE_CGROUP_STORAGE,
+				sizeof(struct bpf_cgroup_storage_key), 8, 0, 0);
+	if (cgroup_storage_fd < 0) {
+		printf("failed to create map '%s'\n", strerror(errno));
+		return -1;
+	}
+
+	percpu_cgroup_storage_fd = bpf_create_map(
+		BPF_MAP_TYPE_PERCPU_CGROUP_STORAGE,
+		sizeof(struct bpf_cgroup_storage_key), 8, 0, 0);
+	if (percpu_cgroup_storage_fd < 0) {
+		printf("failed to create map '%s'\n", strerror(errno));
+		return -1;
+	}
+
+	struct bpf_insn prog[] = {
+		BPF_MOV32_IMM(BPF_REG_0, 0),
+		BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_0, -4), /* *(u32 *)(fp - 4) = r0 */
+		BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
+		BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4), /* r2 = fp - 4 */
+		BPF_LD_MAP_FD(BPF_REG_1, map_fd),
+		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
+		BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
+		BPF_MOV64_IMM(BPF_REG_1, val), /* r1 = 1 */
+		BPF_RAW_INSN(BPF_STX | BPF_XADD | BPF_DW, BPF_REG_0, BPF_REG_1, 0, 0), /* xadd r0 += r1 */
+
+		BPF_LD_MAP_FD(BPF_REG_1, cgroup_storage_fd),
+		BPF_MOV64_IMM(BPF_REG_2, 0),
+		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_local_storage),
+		BPF_MOV64_IMM(BPF_REG_1, val),
+		BPF_RAW_INSN(BPF_STX | BPF_XADD | BPF_W, BPF_REG_0, BPF_REG_1, 0, 0),
+
+		BPF_LD_MAP_FD(BPF_REG_1, percpu_cgroup_storage_fd),
+		BPF_MOV64_IMM(BPF_REG_2, 0),
+		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_local_storage),
+		BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_0, 0),
+		BPF_ALU64_IMM(BPF_ADD, BPF_REG_3, 0x1),
+		BPF_STX_MEM(BPF_W, BPF_REG_0, BPF_REG_3, 0),
+
+		BPF_MOV64_IMM(BPF_REG_0, verdict), /* r0 = verdict */
+		BPF_EXIT_INSN(),
+	};
+	size_t insns_cnt = sizeof(prog) / sizeof(struct bpf_insn);
+	int ret;
+
+	ret = bpf_load_program(BPF_PROG_TYPE_CGROUP_SKB,
+			       prog, insns_cnt, "GPL", 0,
+			       bpf_log_buf, BPF_LOG_BUF_SIZE);
+
+	close(cgroup_storage_fd);
+	return ret;
+}
+
+void test_cgroup_attach_multi(void)
+{
+	__u32 prog_ids[4], prog_cnt = 0, attach_flags, saved_prog_id;
+	int cg1 = 0, cg2 = 0, cg3 = 0, cg4 = 0, cg5 = 0, key = 0;
+	int allow_prog[6] = {-1};
+	unsigned long long value;
+	__u32 duration = 0;
+	int i = 0;
+
+	for (i = 0; i < ARRAY_SIZE(allow_prog); i++) {
+		allow_prog[i] = prog_load_cnt(1, 1 << i);
+		if (CHECK(allow_prog[i] < 0, "prog_load",
+			  "verifier output:\n%s\n-------\n", bpf_log_buf))
+			goto err;
+	}
+
+	if (CHECK_FAIL(setup_cgroup_environment()))
+		goto err;
+
+	cg1 = create_and_get_cgroup("/cg1");
+	if (CHECK_FAIL(cg1 < 0))
+		goto err;
+	cg2 = create_and_get_cgroup("/cg1/cg2");
+	if (CHECK_FAIL(cg2 < 0))
+		goto err;
+	cg3 = create_and_get_cgroup("/cg1/cg2/cg3");
+	if (CHECK_FAIL(cg3 < 0))
+		goto err;
+	cg4 = create_and_get_cgroup("/cg1/cg2/cg3/cg4");
+	if (CHECK_FAIL(cg4 < 0))
+		goto err;
+	cg5 = create_and_get_cgroup("/cg1/cg2/cg3/cg4/cg5");
+	if (CHECK_FAIL(cg5 < 0))
+		goto err;
+
+	if (CHECK_FAIL(join_cgroup("/cg1/cg2/cg3/cg4/cg5")))
+		goto err;
+
+	if (CHECK(bpf_prog_attach(allow_prog[0], cg1, BPF_CGROUP_INET_EGRESS,
+				  BPF_F_ALLOW_MULTI),
+		  "prog0_attach_to_cg1_multi", "errno=%d\n", errno))
+		goto err;
+
+	if (CHECK(!bpf_prog_attach(allow_prog[0], cg1, BPF_CGROUP_INET_EGRESS,
+				   BPF_F_ALLOW_MULTI),
+		  "fail_same_prog_attach_to_cg1", "unexpected success\n"))
+		goto err;
+
+	if (CHECK(bpf_prog_attach(allow_prog[1], cg1, BPF_CGROUP_INET_EGRESS,
+				  BPF_F_ALLOW_MULTI),
+		  "prog1_attach_to_cg1_multi", "errno=%d\n", errno))
+		goto err;
+
+	if (CHECK(bpf_prog_attach(allow_prog[2], cg2, BPF_CGROUP_INET_EGRESS,
+				  BPF_F_ALLOW_OVERRIDE),
+		  "prog2_attach_to_cg2_override", "errno=%d\n", errno))
+		goto err;
+
+	if (CHECK(bpf_prog_attach(allow_prog[3], cg3, BPF_CGROUP_INET_EGRESS,
+				  BPF_F_ALLOW_MULTI),
+		  "prog3_attach_to_cg3_multi", "errno=%d\n", errno))
+		goto err;
+
+	if (CHECK(bpf_prog_attach(allow_prog[4], cg4, BPF_CGROUP_INET_EGRESS,
+			    BPF_F_ALLOW_OVERRIDE),
+		  "prog4_attach_to_cg4_override", "errno=%d\n", errno))
+		goto err;
+
+	if (CHECK(bpf_prog_attach(allow_prog[5], cg5, BPF_CGROUP_INET_EGRESS, 0),
+		  "prog5_attach_to_cg5_none", "errno=%d\n", errno))
+		goto err;
+
+	CHECK_FAIL(system(PING_CMD));
+	CHECK_FAIL(bpf_map_lookup_elem(map_fd, &key, &value));
+	CHECK_FAIL(value != 1 + 2 + 8 + 32);
+
+	/* query the number of effective progs in cg5 */
+	CHECK_FAIL(bpf_prog_query(cg5, BPF_CGROUP_INET_EGRESS,
+				  BPF_F_QUERY_EFFECTIVE, NULL, NULL, &prog_cnt));
+	CHECK_FAIL(prog_cnt != 4);
+	/* retrieve prog_ids of effective progs in cg5 */
+	CHECK_FAIL(bpf_prog_query(cg5, BPF_CGROUP_INET_EGRESS,
+				  BPF_F_QUERY_EFFECTIVE, &attach_flags,
+				  prog_ids, &prog_cnt));
+	CHECK_FAIL(prog_cnt != 4);
+	CHECK_FAIL(attach_flags != 0);
+	saved_prog_id = prog_ids[0];
+	/* check enospc handling */
+	prog_ids[0] = 0;
+	prog_cnt = 2;
+	CHECK_FAIL(bpf_prog_query(cg5, BPF_CGROUP_INET_EGRESS,
+				  BPF_F_QUERY_EFFECTIVE, &attach_flags,
+				  prog_ids, &prog_cnt) != -1);
+	CHECK_FAIL(errno != ENOSPC);
+	CHECK_FAIL(prog_cnt != 4);
+	/* check that prog_ids are returned even when buffer is too small */
+	CHECK_FAIL(prog_ids[0] != saved_prog_id);
+	/* retrieve prog_id of single attached prog in cg5 */
+	prog_ids[0] = 0;
+	CHECK_FAIL(bpf_prog_query(cg5, BPF_CGROUP_INET_EGRESS, 0, NULL,
+				  prog_ids, &prog_cnt));
+	CHECK_FAIL(prog_cnt != 1);
+	CHECK_FAIL(prog_ids[0] != saved_prog_id);
+
+	/* detach bottom program and ping again */
+	if (CHECK(bpf_prog_detach2(-1, cg5, BPF_CGROUP_INET_EGRESS),
+		  "prog_detach_from_cg5", "errno=%d\n", errno))
+		goto err;
+
+	value = 0;
+	CHECK_FAIL(bpf_map_update_elem(map_fd, &key, &value, 0));
+	CHECK_FAIL(system(PING_CMD));
+	CHECK_FAIL(bpf_map_lookup_elem(map_fd, &key, &value));
+	CHECK_FAIL(value != 1 + 2 + 8 + 16);
+
+	/* detach 3rd from bottom program and ping again */
+	if (CHECK(!bpf_prog_detach2(0, cg3, BPF_CGROUP_INET_EGRESS),
+		  "fail_prog_detach_from_cg3", "unexpected success\n"))
+		goto err;
+
+	if (CHECK(bpf_prog_detach2(allow_prog[3], cg3, BPF_CGROUP_INET_EGRESS),
+		  "prog3_detach_from_cg3", "errno=%d\n", errno))
+		goto err;
+
+	value = 0;
+	CHECK_FAIL(bpf_map_update_elem(map_fd, &key, &value, 0));
+	CHECK_FAIL(system(PING_CMD));
+	CHECK_FAIL(bpf_map_lookup_elem(map_fd, &key, &value));
+	CHECK_FAIL(value != 1 + 2 + 16);
+
+	/* detach 2nd from bottom program and ping again */
+	if (CHECK(bpf_prog_detach2(-1, cg4, BPF_CGROUP_INET_EGRESS),
+		  "prog_detach_from_cg4", "errno=%d\n", errno))
+		goto err;
+
+	value = 0;
+	CHECK_FAIL(bpf_map_update_elem(map_fd, &key, &value, 0));
+	CHECK_FAIL(system(PING_CMD));
+	CHECK_FAIL(bpf_map_lookup_elem(map_fd, &key, &value));
+	CHECK_FAIL(value != 1 + 2 + 4);
+
+	prog_cnt = 4;
+	CHECK_FAIL(bpf_prog_query(cg5, BPF_CGROUP_INET_EGRESS,
+				  BPF_F_QUERY_EFFECTIVE, &attach_flags,
+				  prog_ids, &prog_cnt));
+	CHECK_FAIL(prog_cnt != 3);
+	CHECK_FAIL(attach_flags != 0);
+	CHECK_FAIL(bpf_prog_query(cg5, BPF_CGROUP_INET_EGRESS, 0, NULL,
+				  prog_ids, &prog_cnt));
+	CHECK_FAIL(prog_cnt != 0);
+
+err:
+	for (i = 0; i < ARRAY_SIZE(allow_prog); i++)
+		if (allow_prog[i] >= 0)
+			close(allow_prog[i]);
+	close(cg1);
+	close(cg2);
+	close(cg3);
+	close(cg4);
+	close(cg5);
+	cleanup_cgroup_environment();
+}
diff --git a/tools/testing/selftests/bpf/prog_tests/cgroup_attach_override.c b/tools/testing/selftests/bpf/prog_tests/cgroup_attach_override.c
new file mode 100644
index 000000000000..9d8cb48b99de
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/cgroup_attach_override.c
@@ -0,0 +1,148 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <test_progs.h>
+
+#include "cgroup_helpers.h"
+
+#define FOO		"/foo"
+#define BAR		"/foo/bar/"
+#define PING_CMD	"ping -q -c1 -w1 127.0.0.1 > /dev/null"
+
+char bpf_log_buf[BPF_LOG_BUF_SIZE];
+
+static int prog_load(int verdict)
+{
+	struct bpf_insn prog[] = {
+		BPF_MOV64_IMM(BPF_REG_0, verdict), /* r0 = verdict */
+		BPF_EXIT_INSN(),
+	};
+	size_t insns_cnt = sizeof(prog) / sizeof(struct bpf_insn);
+
+	return bpf_load_program(BPF_PROG_TYPE_CGROUP_SKB,
+			       prog, insns_cnt, "GPL", 0,
+			       bpf_log_buf, BPF_LOG_BUF_SIZE);
+}
+
+void test_cgroup_attach_override(void)
+{
+	int drop_prog = -1, allow_prog = -1, foo = -1, bar = -1;
+	__u32 duration = 0;
+
+	allow_prog = prog_load(1);
+	if (CHECK(allow_prog < 0, "prog_load_allow",
+		  "verifier output:\n%s\n-------\n", bpf_log_buf))
+		goto err;
+
+	drop_prog = prog_load(0);
+	if (CHECK(drop_prog < 0, "prog_load_drop",
+		  "verifier output:\n%s\n-------\n", bpf_log_buf))
+		goto err;
+
+	foo = test__join_cgroup(FOO);
+	if (CHECK(foo < 0, "cgroup_join_foo", "cgroup setup failed\n"))
+		goto err;
+
+	if (CHECK(bpf_prog_attach(drop_prog, foo, BPF_CGROUP_INET_EGRESS,
+				  BPF_F_ALLOW_OVERRIDE),
+		  "prog_attach_drop_foo_override",
+		  "attach prog to %s failed, errno=%d\n", FOO, errno))
+		goto err;
+
+	if (CHECK(!system(PING_CMD), "ping_fail",
+		  "ping unexpectedly succeeded\n"))
+		goto err;
+
+	bar = test__join_cgroup(BAR);
+	if (CHECK(bar < 0, "cgroup_join_bar", "cgroup setup failed\n"))
+		goto err;
+
+	if (CHECK(!system(PING_CMD), "ping_fail",
+		  "ping unexpectedly succeeded\n"))
+		goto err;
+
+	if (CHECK(bpf_prog_attach(allow_prog, bar, BPF_CGROUP_INET_EGRESS,
+				  BPF_F_ALLOW_OVERRIDE),
+		  "prog_attach_allow_bar_override",
+		  "attach prog to %s failed, errno=%d\n", BAR, errno))
+		goto err;
+
+	if (CHECK(system(PING_CMD), "ping_ok", "ping failed\n"))
+		goto err;
+
+	if (CHECK(bpf_prog_detach(bar, BPF_CGROUP_INET_EGRESS),
+		  "prog_detach_bar",
+		  "detach prog from %s failed, errno=%d\n", BAR, errno))
+		goto err;
+
+	if (CHECK(!system(PING_CMD), "ping_fail",
+		  "ping unexpectedly succeeded\n"))
+		goto err;
+
+	if (CHECK(bpf_prog_attach(allow_prog, bar, BPF_CGROUP_INET_EGRESS,
+				  BPF_F_ALLOW_OVERRIDE),
+		  "prog_attach_allow_bar_override",
+		  "attach prog to %s failed, errno=%d\n", BAR, errno))
+		goto err;
+
+	if (CHECK(bpf_prog_detach(foo, BPF_CGROUP_INET_EGRESS),
+		  "prog_detach_foo",
+		  "detach prog from %s failed, errno=%d\n", FOO, errno))
+		goto err;
+
+	if (CHECK(system(PING_CMD), "ping_ok", "ping failed\n"))
+		goto err;
+
+	if (CHECK(bpf_prog_attach(allow_prog, bar, BPF_CGROUP_INET_EGRESS,
+				  BPF_F_ALLOW_OVERRIDE),
+		  "prog_attach_allow_bar_override",
+		  "attach prog to %s failed, errno=%d\n", BAR, errno))
+		goto err;
+
+	if (CHECK(!bpf_prog_attach(allow_prog, bar, BPF_CGROUP_INET_EGRESS, 0),
+		  "fail_prog_attach_allow_bar_none",
+		  "attach prog to %s unexpectedly succeeded\n", BAR))
+		goto err;
+
+	if (CHECK(bpf_prog_detach(bar, BPF_CGROUP_INET_EGRESS),
+		  "prog_detach_bar",
+		  "detach prog from %s failed, errno=%d\n", BAR, errno))
+		goto err;
+
+	if (CHECK(!bpf_prog_detach(foo, BPF_CGROUP_INET_EGRESS),
+		  "fail_prog_detach_foo",
+		  "double detach from %s unexpectedly succeeded\n", FOO))
+		goto err;
+
+	if (CHECK(bpf_prog_attach(allow_prog, foo, BPF_CGROUP_INET_EGRESS, 0),
+		  "prog_attach_allow_foo_none",
+		  "attach prog to %s failed, errno=%d\n", FOO, errno))
+		goto err;
+
+	if (CHECK(!bpf_prog_attach(allow_prog, bar, BPF_CGROUP_INET_EGRESS, 0),
+		  "fail_prog_attach_allow_bar_none",
+		  "attach prog to %s unexpectedly succeeded\n", BAR))
+		goto err;
+
+	if (CHECK(!bpf_prog_attach(allow_prog, bar, BPF_CGROUP_INET_EGRESS,
+				   BPF_F_ALLOW_OVERRIDE),
+		  "fail_prog_attach_allow_bar_override",
+		  "attach prog to %s unexpectedly succeeded\n", BAR))
+		goto err;
+
+	if (CHECK(!bpf_prog_attach(allow_prog, foo, BPF_CGROUP_INET_EGRESS,
+				   BPF_F_ALLOW_OVERRIDE),
+		  "fail_prog_attach_allow_foo_override",
+		  "attach prog to %s unexpectedly succeeded\n", FOO))
+		goto err;
+
+	if (CHECK(bpf_prog_attach(drop_prog, foo, BPF_CGROUP_INET_EGRESS, 0),
+		  "prog_attach_drop_foo_none",
+		  "attach prog to %s failed, errno=%d\n", FOO, errno))
+		goto err;
+
+err:
+	close(foo);
+	close(bar);
+	close(allow_prog);
+	close(drop_prog);
+}
diff --git a/tools/testing/selftests/bpf/test_cgroup_attach.c b/tools/testing/selftests/bpf/test_cgroup_attach.c
deleted file mode 100644
index 7671909ee1cb..000000000000
--- a/tools/testing/selftests/bpf/test_cgroup_attach.c
+++ /dev/null
@@ -1,571 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-
-/* eBPF example program:
- *
- * - Creates arraymap in kernel with 4 bytes keys and 8 byte values
- *
- * - Loads eBPF program
- *
- *   The eBPF program accesses the map passed in to store two pieces of
- *   information. The number of invocations of the program, which maps
- *   to the number of packets received, is stored to key 0. Key 1 is
- *   incremented on each iteration by the number of bytes stored in
- *   the skb. The program also stores the number of received bytes
- *   in the cgroup storage.
- *
- * - Attaches the new program to a cgroup using BPF_PROG_ATTACH
- *
- * - Every second, reads map[0] and map[1] to see how many bytes and
- *   packets were seen on any socket of tasks in the given cgroup.
- */
-
-#define _GNU_SOURCE
-
-#include <stdio.h>
-#include <stdlib.h>
-#include <assert.h>
-#include <sys/resource.h>
-#include <sys/time.h>
-#include <unistd.h>
-#include <linux/filter.h>
-
-#include <linux/bpf.h>
-#include <bpf/bpf.h>
-
-#include "bpf_util.h"
-#include "bpf_rlimit.h"
-#include "cgroup_helpers.h"
-
-#define FOO		"/foo"
-#define BAR		"/foo/bar/"
-#define PING_CMD	"ping -q -c1 -w1 127.0.0.1 > /dev/null"
-
-char bpf_log_buf[BPF_LOG_BUF_SIZE];
-
-#ifdef DEBUG
-#define debug(args...) printf(args)
-#else
-#define debug(args...)
-#endif
-
-static int prog_load(int verdict)
-{
-	int ret;
-	struct bpf_insn prog[] = {
-		BPF_MOV64_IMM(BPF_REG_0, verdict), /* r0 = verdict */
-		BPF_EXIT_INSN(),
-	};
-	size_t insns_cnt = sizeof(prog) / sizeof(struct bpf_insn);
-
-	ret = bpf_load_program(BPF_PROG_TYPE_CGROUP_SKB,
-			       prog, insns_cnt, "GPL", 0,
-			       bpf_log_buf, BPF_LOG_BUF_SIZE);
-
-	if (ret < 0) {
-		log_err("Loading program");
-		printf("Output from verifier:\n%s\n-------\n", bpf_log_buf);
-		return 0;
-	}
-	return ret;
-}
-
-static int test_foo_bar(void)
-{
-	int drop_prog, allow_prog, foo = 0, bar = 0, rc = 0;
-
-	allow_prog = prog_load(1);
-	if (!allow_prog)
-		goto err;
-
-	drop_prog = prog_load(0);
-	if (!drop_prog)
-		goto err;
-
-	if (setup_cgroup_environment())
-		goto err;
-
-	/* Create cgroup /foo, get fd, and join it */
-	foo = create_and_get_cgroup(FOO);
-	if (foo < 0)
-		goto err;
-
-	if (join_cgroup(FOO))
-		goto err;
-
-	if (bpf_prog_attach(drop_prog, foo, BPF_CGROUP_INET_EGRESS,
-			    BPF_F_ALLOW_OVERRIDE)) {
-		log_err("Attaching prog to /foo");
-		goto err;
-	}
-
-	debug("Attached DROP prog. This ping in cgroup /foo should fail...\n");
-	assert(system(PING_CMD) != 0);
-
-	/* Create cgroup /foo/bar, get fd, and join it */
-	bar = create_and_get_cgroup(BAR);
-	if (bar < 0)
-		goto err;
-
-	if (join_cgroup(BAR))
-		goto err;
-
-	debug("Attached DROP prog. This ping in cgroup /foo/bar should fail...\n");
-	assert(system(PING_CMD) != 0);
-
-	if (bpf_prog_attach(allow_prog, bar, BPF_CGROUP_INET_EGRESS,
-			    BPF_F_ALLOW_OVERRIDE)) {
-		log_err("Attaching prog to /foo/bar");
-		goto err;
-	}
-
-	debug("Attached PASS prog. This ping in cgroup /foo/bar should pass...\n");
-	assert(system(PING_CMD) == 0);
-
-	if (bpf_prog_detach(bar, BPF_CGROUP_INET_EGRESS)) {
-		log_err("Detaching program from /foo/bar");
-		goto err;
-	}
-
-	debug("Detached PASS from /foo/bar while DROP is attached to /foo.\n"
-	       "This ping in cgroup /foo/bar should fail...\n");
-	assert(system(PING_CMD) != 0);
-
-	if (bpf_prog_attach(allow_prog, bar, BPF_CGROUP_INET_EGRESS,
-			    BPF_F_ALLOW_OVERRIDE)) {
-		log_err("Attaching prog to /foo/bar");
-		goto err;
-	}
-
-	if (bpf_prog_detach(foo, BPF_CGROUP_INET_EGRESS)) {
-		log_err("Detaching program from /foo");
-		goto err;
-	}
-
-	debug("Attached PASS from /foo/bar and detached DROP from /foo.\n"
-	       "This ping in cgroup /foo/bar should pass...\n");
-	assert(system(PING_CMD) == 0);
-
-	if (bpf_prog_attach(allow_prog, bar, BPF_CGROUP_INET_EGRESS,
-			    BPF_F_ALLOW_OVERRIDE)) {
-		log_err("Attaching prog to /foo/bar");
-		goto err;
-	}
-
-	if (!bpf_prog_attach(allow_prog, bar, BPF_CGROUP_INET_EGRESS, 0)) {
-		errno = 0;
-		log_err("Unexpected success attaching prog to /foo/bar");
-		goto err;
-	}
-
-	if (bpf_prog_detach(bar, BPF_CGROUP_INET_EGRESS)) {
-		log_err("Detaching program from /foo/bar");
-		goto err;
-	}
-
-	if (!bpf_prog_detach(foo, BPF_CGROUP_INET_EGRESS)) {
-		errno = 0;
-		log_err("Unexpected success in double detach from /foo");
-		goto err;
-	}
-
-	if (bpf_prog_attach(allow_prog, foo, BPF_CGROUP_INET_EGRESS, 0)) {
-		log_err("Attaching non-overridable prog to /foo");
-		goto err;
-	}
-
-	if (!bpf_prog_attach(allow_prog, bar, BPF_CGROUP_INET_EGRESS, 0)) {
-		errno = 0;
-		log_err("Unexpected success attaching non-overridable prog to /foo/bar");
-		goto err;
-	}
-
-	if (!bpf_prog_attach(allow_prog, bar, BPF_CGROUP_INET_EGRESS,
-			     BPF_F_ALLOW_OVERRIDE)) {
-		errno = 0;
-		log_err("Unexpected success attaching overridable prog to /foo/bar");
-		goto err;
-	}
-
-	if (!bpf_prog_attach(allow_prog, foo, BPF_CGROUP_INET_EGRESS,
-			     BPF_F_ALLOW_OVERRIDE)) {
-		errno = 0;
-		log_err("Unexpected success attaching overridable prog to /foo");
-		goto err;
-	}
-
-	if (bpf_prog_attach(drop_prog, foo, BPF_CGROUP_INET_EGRESS, 0)) {
-		log_err("Attaching different non-overridable prog to /foo");
-		goto err;
-	}
-
-	goto out;
-
-err:
-	rc = 1;
-
-out:
-	close(foo);
-	close(bar);
-	cleanup_cgroup_environment();
-	if (!rc)
-		printf("#override:PASS\n");
-	else
-		printf("#override:FAIL\n");
-	return rc;
-}
-
-static int map_fd = -1;
-
-static int prog_load_cnt(int verdict, int val)
-{
-	int cgroup_storage_fd, percpu_cgroup_storage_fd;
-
-	if (map_fd < 0)
-		map_fd = bpf_create_map(BPF_MAP_TYPE_ARRAY, 4, 8, 1, 0);
-	if (map_fd < 0) {
-		printf("failed to create map '%s'\n", strerror(errno));
-		return -1;
-	}
-
-	cgroup_storage_fd = bpf_create_map(BPF_MAP_TYPE_CGROUP_STORAGE,
-				sizeof(struct bpf_cgroup_storage_key), 8, 0, 0);
-	if (cgroup_storage_fd < 0) {
-		printf("failed to create map '%s'\n", strerror(errno));
-		return -1;
-	}
-
-	percpu_cgroup_storage_fd = bpf_create_map(
-		BPF_MAP_TYPE_PERCPU_CGROUP_STORAGE,
-		sizeof(struct bpf_cgroup_storage_key), 8, 0, 0);
-	if (percpu_cgroup_storage_fd < 0) {
-		printf("failed to create map '%s'\n", strerror(errno));
-		return -1;
-	}
-
-	struct bpf_insn prog[] = {
-		BPF_MOV32_IMM(BPF_REG_0, 0),
-		BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_0, -4), /* *(u32 *)(fp - 4) = r0 */
-		BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-		BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4), /* r2 = fp - 4 */
-		BPF_LD_MAP_FD(BPF_REG_1, map_fd),
-		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-		BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
-		BPF_MOV64_IMM(BPF_REG_1, val), /* r1 = 1 */
-		BPF_RAW_INSN(BPF_STX | BPF_XADD | BPF_DW, BPF_REG_0, BPF_REG_1, 0, 0), /* xadd r0 += r1 */
-
-		BPF_LD_MAP_FD(BPF_REG_1, cgroup_storage_fd),
-		BPF_MOV64_IMM(BPF_REG_2, 0),
-		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_local_storage),
-		BPF_MOV64_IMM(BPF_REG_1, val),
-		BPF_RAW_INSN(BPF_STX | BPF_XADD | BPF_W, BPF_REG_0, BPF_REG_1, 0, 0),
-
-		BPF_LD_MAP_FD(BPF_REG_1, percpu_cgroup_storage_fd),
-		BPF_MOV64_IMM(BPF_REG_2, 0),
-		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_local_storage),
-		BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_0, 0),
-		BPF_ALU64_IMM(BPF_ADD, BPF_REG_3, 0x1),
-		BPF_STX_MEM(BPF_W, BPF_REG_0, BPF_REG_3, 0),
-
-		BPF_MOV64_IMM(BPF_REG_0, verdict), /* r0 = verdict */
-		BPF_EXIT_INSN(),
-	};
-	size_t insns_cnt = sizeof(prog) / sizeof(struct bpf_insn);
-	int ret;
-
-	ret = bpf_load_program(BPF_PROG_TYPE_CGROUP_SKB,
-			       prog, insns_cnt, "GPL", 0,
-			       bpf_log_buf, BPF_LOG_BUF_SIZE);
-
-	if (ret < 0) {
-		log_err("Loading program");
-		printf("Output from verifier:\n%s\n-------\n", bpf_log_buf);
-		return 0;
-	}
-	close(cgroup_storage_fd);
-	return ret;
-}
-
-
-static int test_multiprog(void)
-{
-	__u32 prog_ids[4], prog_cnt = 0, attach_flags, saved_prog_id;
-	int cg1 = 0, cg2 = 0, cg3 = 0, cg4 = 0, cg5 = 0, key = 0;
-	int drop_prog, allow_prog[6] = {}, rc = 0;
-	unsigned long long value;
-	int i = 0;
-
-	for (i = 0; i < 6; i++) {
-		allow_prog[i] = prog_load_cnt(1, 1 << i);
-		if (!allow_prog[i])
-			goto err;
-	}
-	drop_prog = prog_load_cnt(0, 1);
-	if (!drop_prog)
-		goto err;
-
-	if (setup_cgroup_environment())
-		goto err;
-
-	cg1 = create_and_get_cgroup("/cg1");
-	if (cg1 < 0)
-		goto err;
-	cg2 = create_and_get_cgroup("/cg1/cg2");
-	if (cg2 < 0)
-		goto err;
-	cg3 = create_and_get_cgroup("/cg1/cg2/cg3");
-	if (cg3 < 0)
-		goto err;
-	cg4 = create_and_get_cgroup("/cg1/cg2/cg3/cg4");
-	if (cg4 < 0)
-		goto err;
-	cg5 = create_and_get_cgroup("/cg1/cg2/cg3/cg4/cg5");
-	if (cg5 < 0)
-		goto err;
-
-	if (join_cgroup("/cg1/cg2/cg3/cg4/cg5"))
-		goto err;
-
-	if (bpf_prog_attach(allow_prog[0], cg1, BPF_CGROUP_INET_EGRESS,
-			    BPF_F_ALLOW_MULTI)) {
-		log_err("Attaching prog to cg1");
-		goto err;
-	}
-	if (!bpf_prog_attach(allow_prog[0], cg1, BPF_CGROUP_INET_EGRESS,
-			     BPF_F_ALLOW_MULTI)) {
-		log_err("Unexpected success attaching the same prog to cg1");
-		goto err;
-	}
-	if (bpf_prog_attach(allow_prog[1], cg1, BPF_CGROUP_INET_EGRESS,
-			    BPF_F_ALLOW_MULTI)) {
-		log_err("Attaching prog2 to cg1");
-		goto err;
-	}
-	if (bpf_prog_attach(allow_prog[2], cg2, BPF_CGROUP_INET_EGRESS,
-			    BPF_F_ALLOW_OVERRIDE)) {
-		log_err("Attaching prog to cg2");
-		goto err;
-	}
-	if (bpf_prog_attach(allow_prog[3], cg3, BPF_CGROUP_INET_EGRESS,
-			    BPF_F_ALLOW_MULTI)) {
-		log_err("Attaching prog to cg3");
-		goto err;
-	}
-	if (bpf_prog_attach(allow_prog[4], cg4, BPF_CGROUP_INET_EGRESS,
-			    BPF_F_ALLOW_OVERRIDE)) {
-		log_err("Attaching prog to cg4");
-		goto err;
-	}
-	if (bpf_prog_attach(allow_prog[5], cg5, BPF_CGROUP_INET_EGRESS, 0)) {
-		log_err("Attaching prog to cg5");
-		goto err;
-	}
-	assert(system(PING_CMD) == 0);
-	assert(bpf_map_lookup_elem(map_fd, &key, &value) == 0);
-	assert(value == 1 + 2 + 8 + 32);
-
-	/* query the number of effective progs in cg5 */
-	assert(bpf_prog_query(cg5, BPF_CGROUP_INET_EGRESS, BPF_F_QUERY_EFFECTIVE,
-			      NULL, NULL, &prog_cnt) == 0);
-	assert(prog_cnt == 4);
-	/* retrieve prog_ids of effective progs in cg5 */
-	assert(bpf_prog_query(cg5, BPF_CGROUP_INET_EGRESS, BPF_F_QUERY_EFFECTIVE,
-			      &attach_flags, prog_ids, &prog_cnt) == 0);
-	assert(prog_cnt == 4);
-	assert(attach_flags == 0);
-	saved_prog_id = prog_ids[0];
-	/* check enospc handling */
-	prog_ids[0] = 0;
-	prog_cnt = 2;
-	assert(bpf_prog_query(cg5, BPF_CGROUP_INET_EGRESS, BPF_F_QUERY_EFFECTIVE,
-			      &attach_flags, prog_ids, &prog_cnt) == -1 &&
-	       errno == ENOSPC);
-	assert(prog_cnt == 4);
-	/* check that prog_ids are returned even when buffer is too small */
-	assert(prog_ids[0] == saved_prog_id);
-	/* retrieve prog_id of single attached prog in cg5 */
-	prog_ids[0] = 0;
-	assert(bpf_prog_query(cg5, BPF_CGROUP_INET_EGRESS, 0,
-			      NULL, prog_ids, &prog_cnt) == 0);
-	assert(prog_cnt == 1);
-	assert(prog_ids[0] == saved_prog_id);
-
-	/* detach bottom program and ping again */
-	if (bpf_prog_detach2(-1, cg5, BPF_CGROUP_INET_EGRESS)) {
-		log_err("Detaching prog from cg5");
-		goto err;
-	}
-	value = 0;
-	assert(bpf_map_update_elem(map_fd, &key, &value, 0) == 0);
-	assert(system(PING_CMD) == 0);
-	assert(bpf_map_lookup_elem(map_fd, &key, &value) == 0);
-	assert(value == 1 + 2 + 8 + 16);
-
-	/* detach 3rd from bottom program and ping again */
-	errno = 0;
-	if (!bpf_prog_detach2(0, cg3, BPF_CGROUP_INET_EGRESS)) {
-		log_err("Unexpected success on detach from cg3");
-		goto err;
-	}
-	if (bpf_prog_detach2(allow_prog[3], cg3, BPF_CGROUP_INET_EGRESS)) {
-		log_err("Detaching from cg3");
-		goto err;
-	}
-	value = 0;
-	assert(bpf_map_update_elem(map_fd, &key, &value, 0) == 0);
-	assert(system(PING_CMD) == 0);
-	assert(bpf_map_lookup_elem(map_fd, &key, &value) == 0);
-	assert(value == 1 + 2 + 16);
-
-	/* detach 2nd from bottom program and ping again */
-	if (bpf_prog_detach2(-1, cg4, BPF_CGROUP_INET_EGRESS)) {
-		log_err("Detaching prog from cg4");
-		goto err;
-	}
-	value = 0;
-	assert(bpf_map_update_elem(map_fd, &key, &value, 0) == 0);
-	assert(system(PING_CMD) == 0);
-	assert(bpf_map_lookup_elem(map_fd, &key, &value) == 0);
-	assert(value == 1 + 2 + 4);
-
-	prog_cnt = 4;
-	assert(bpf_prog_query(cg5, BPF_CGROUP_INET_EGRESS, BPF_F_QUERY_EFFECTIVE,
-			      &attach_flags, prog_ids, &prog_cnt) == 0);
-	assert(prog_cnt == 3);
-	assert(attach_flags == 0);
-	assert(bpf_prog_query(cg5, BPF_CGROUP_INET_EGRESS, 0,
-			      NULL, prog_ids, &prog_cnt) == 0);
-	assert(prog_cnt == 0);
-	goto out;
-err:
-	rc = 1;
-
-out:
-	for (i = 0; i < 6; i++)
-		if (allow_prog[i] > 0)
-			close(allow_prog[i]);
-	close(cg1);
-	close(cg2);
-	close(cg3);
-	close(cg4);
-	close(cg5);
-	cleanup_cgroup_environment();
-	if (!rc)
-		printf("#multi:PASS\n");
-	else
-		printf("#multi:FAIL\n");
-	return rc;
-}
-
-static int test_autodetach(void)
-{
-	__u32 prog_cnt = 4, attach_flags;
-	int allow_prog[2] = {0};
-	__u32 prog_ids[2] = {0};
-	int cg = 0, i, rc = -1;
-	void *ptr = NULL;
-	int attempts;
-
-	for (i = 0; i < ARRAY_SIZE(allow_prog); i++) {
-		allow_prog[i] = prog_load_cnt(1, 1 << i);
-		if (!allow_prog[i])
-			goto err;
-	}
-
-	if (setup_cgroup_environment())
-		goto err;
-
-	/* create a cgroup, attach two programs and remember their ids */
-	cg = create_and_get_cgroup("/cg_autodetach");
-	if (cg < 0)
-		goto err;
-
-	if (join_cgroup("/cg_autodetach"))
-		goto err;
-
-	for (i = 0; i < ARRAY_SIZE(allow_prog); i++) {
-		if (bpf_prog_attach(allow_prog[i], cg, BPF_CGROUP_INET_EGRESS,
-				    BPF_F_ALLOW_MULTI)) {
-			log_err("Attaching prog[%d] to cg:egress", i);
-			goto err;
-		}
-	}
-
-	/* make sure that programs are attached and run some traffic */
-	assert(bpf_prog_query(cg, BPF_CGROUP_INET_EGRESS, 0, &attach_flags,
-			      prog_ids, &prog_cnt) == 0);
-	assert(system(PING_CMD) == 0);
-
-	/* allocate some memory (4Mb) to pin the original cgroup */
-	ptr = malloc(4 * (1 << 20));
-	if (!ptr)
-		goto err;
-
-	/* close programs and cgroup fd */
-	for (i = 0; i < ARRAY_SIZE(allow_prog); i++) {
-		close(allow_prog[i]);
-		allow_prog[i] = 0;
-	}
-
-	close(cg);
-	cg = 0;
-
-	/* leave the cgroup and remove it. don't detach programs */
-	cleanup_cgroup_environment();
-
-	/* wait for the asynchronous auto-detachment.
-	 * wait for no more than 5 sec and give up.
-	 */
-	for (i = 0; i < ARRAY_SIZE(prog_ids); i++) {
-		for (attempts = 5; attempts >= 0; attempts--) {
-			int fd = bpf_prog_get_fd_by_id(prog_ids[i]);
-
-			if (fd < 0)
-				break;
-
-			/* don't leave the fd open */
-			close(fd);
-
-			if (!attempts)
-				goto err;
-
-			sleep(1);
-		}
-	}
-
-	rc = 0;
-err:
-	for (i = 0; i < ARRAY_SIZE(allow_prog); i++)
-		if (allow_prog[i] > 0)
-			close(allow_prog[i]);
-	if (cg)
-		close(cg);
-	free(ptr);
-	cleanup_cgroup_environment();
-	if (!rc)
-		printf("#autodetach:PASS\n");
-	else
-		printf("#autodetach:FAIL\n");
-	return rc;
-}
-
-int main(void)
-{
-	int (*tests[])(void) = {
-		test_foo_bar,
-		test_multiprog,
-		test_autodetach,
-	};
-	int errors = 0;
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(tests); i++)
-		if (tests[i]())
-			errors++;
-
-	if (errors)
-		printf("test_cgroup_attach:FAIL\n");
-	else
-		printf("test_cgroup_attach:PASS\n");
-
-	return errors ? EXIT_FAILURE : EXIT_SUCCESS;
-}
-- 
cgit v1.2.3-59-g8ed1b


From 06ac0186bd242d44f5703ce64ee0077b317c86bb Mon Sep 17 00:00:00 2001
From: Andrey Ignatov <rdna@fb.com>
Date: Wed, 18 Dec 2019 23:44:38 -0800
Subject: selftests/bpf: Test BPF_F_REPLACE in cgroup_attach_multi

Test replacing a cgroup-bpf program attached with BPF_F_ALLOW_MULTI and
possible failure modes: invalid combination of flags, invalid
replace_bpf_fd, replacing a non-attachd to specified cgroup program.

Example of program replacing:

  # gdb -q --args ./test_progs --name=cgroup_attach_multi
  ...
  Breakpoint 1, test_cgroup_attach_multi () at cgroup_attach_multi.c:227
  (gdb)
  [1]+  Stopped                 gdb -q --args ./test_progs --name=cgroup_attach_multi
  # bpftool c s /mnt/cgroup2/cgroup-test-work-dir/cg1
  ID       AttachType      AttachFlags     Name
  2133     egress          multi
  2134     egress          multi
  # fg
  gdb -q --args ./test_progs --name=cgroup_attach_multi
  (gdb) c
  Continuing.

  Breakpoint 2, test_cgroup_attach_multi () at cgroup_attach_multi.c:233
  (gdb)
  [1]+  Stopped                 gdb -q --args ./test_progs --name=cgroup_attach_multi
  # bpftool c s /mnt/cgroup2/cgroup-test-work-dir/cg1
  ID       AttachType      AttachFlags     Name
  2139     egress          multi
  2134     egress          multi

Signed-off-by: Andrey Ignatov <rdna@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/7b9b83e8d5fb82e15b034341bd40b6fb2431eeba.1576741281.git.rdna@fb.com
---
 .../selftests/bpf/prog_tests/cgroup_attach_multi.c | 53 ++++++++++++++++++++--
 1 file changed, 50 insertions(+), 3 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/cgroup_attach_multi.c b/tools/testing/selftests/bpf/prog_tests/cgroup_attach_multi.c
index 4eaab7435044..2ff21dbce179 100644
--- a/tools/testing/selftests/bpf/prog_tests/cgroup_attach_multi.c
+++ b/tools/testing/selftests/bpf/prog_tests/cgroup_attach_multi.c
@@ -78,7 +78,8 @@ void test_cgroup_attach_multi(void)
 {
 	__u32 prog_ids[4], prog_cnt = 0, attach_flags, saved_prog_id;
 	int cg1 = 0, cg2 = 0, cg3 = 0, cg4 = 0, cg5 = 0, key = 0;
-	int allow_prog[6] = {-1};
+	DECLARE_LIBBPF_OPTS(bpf_prog_attach_opts, attach_opts);
+	int allow_prog[7] = {-1};
 	unsigned long long value;
 	__u32 duration = 0;
 	int i = 0;
@@ -189,6 +190,52 @@ void test_cgroup_attach_multi(void)
 	CHECK_FAIL(bpf_map_lookup_elem(map_fd, &key, &value));
 	CHECK_FAIL(value != 1 + 2 + 8 + 16);
 
+	/* test replace */
+
+	attach_opts.flags = BPF_F_ALLOW_OVERRIDE | BPF_F_REPLACE;
+	attach_opts.replace_prog_fd = allow_prog[0];
+	if (CHECK(!bpf_prog_attach_xattr(allow_prog[6], cg1,
+					 BPF_CGROUP_INET_EGRESS, &attach_opts),
+		  "fail_prog_replace_override", "unexpected success\n"))
+		goto err;
+	CHECK_FAIL(errno != EINVAL);
+
+	attach_opts.flags = BPF_F_REPLACE;
+	if (CHECK(!bpf_prog_attach_xattr(allow_prog[6], cg1,
+					 BPF_CGROUP_INET_EGRESS, &attach_opts),
+		  "fail_prog_replace_no_multi", "unexpected success\n"))
+		goto err;
+	CHECK_FAIL(errno != EINVAL);
+
+	attach_opts.flags = BPF_F_ALLOW_MULTI | BPF_F_REPLACE;
+	attach_opts.replace_prog_fd = -1;
+	if (CHECK(!bpf_prog_attach_xattr(allow_prog[6], cg1,
+					 BPF_CGROUP_INET_EGRESS, &attach_opts),
+		  "fail_prog_replace_bad_fd", "unexpected success\n"))
+		goto err;
+	CHECK_FAIL(errno != EBADF);
+
+	/* replacing a program that is not attached to cgroup should fail  */
+	attach_opts.replace_prog_fd = allow_prog[3];
+	if (CHECK(!bpf_prog_attach_xattr(allow_prog[6], cg1,
+					 BPF_CGROUP_INET_EGRESS, &attach_opts),
+		  "fail_prog_replace_no_ent", "unexpected success\n"))
+		goto err;
+	CHECK_FAIL(errno != ENOENT);
+
+	/* replace 1st from the top program */
+	attach_opts.replace_prog_fd = allow_prog[0];
+	if (CHECK(bpf_prog_attach_xattr(allow_prog[6], cg1,
+					BPF_CGROUP_INET_EGRESS, &attach_opts),
+		  "prog_replace", "errno=%d\n", errno))
+		goto err;
+
+	value = 0;
+	CHECK_FAIL(bpf_map_update_elem(map_fd, &key, &value, 0));
+	CHECK_FAIL(system(PING_CMD));
+	CHECK_FAIL(bpf_map_lookup_elem(map_fd, &key, &value));
+	CHECK_FAIL(value != 64 + 2 + 8 + 16);
+
 	/* detach 3rd from bottom program and ping again */
 	if (CHECK(!bpf_prog_detach2(0, cg3, BPF_CGROUP_INET_EGRESS),
 		  "fail_prog_detach_from_cg3", "unexpected success\n"))
@@ -202,7 +249,7 @@ void test_cgroup_attach_multi(void)
 	CHECK_FAIL(bpf_map_update_elem(map_fd, &key, &value, 0));
 	CHECK_FAIL(system(PING_CMD));
 	CHECK_FAIL(bpf_map_lookup_elem(map_fd, &key, &value));
-	CHECK_FAIL(value != 1 + 2 + 16);
+	CHECK_FAIL(value != 64 + 2 + 16);
 
 	/* detach 2nd from bottom program and ping again */
 	if (CHECK(bpf_prog_detach2(-1, cg4, BPF_CGROUP_INET_EGRESS),
@@ -213,7 +260,7 @@ void test_cgroup_attach_multi(void)
 	CHECK_FAIL(bpf_map_update_elem(map_fd, &key, &value, 0));
 	CHECK_FAIL(system(PING_CMD));
 	CHECK_FAIL(bpf_map_lookup_elem(map_fd, &key, &value));
-	CHECK_FAIL(value != 1 + 2 + 4);
+	CHECK_FAIL(value != 64 + 2 + 4);
 
 	prog_cnt = 4;
 	CHECK_FAIL(bpf_prog_query(cg5, BPF_CGROUP_INET_EGRESS,
-- 
cgit v1.2.3-59-g8ed1b


From 1019f9efcb9fa0cea4b361604b438fef3ea3e394 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 30 Sep 2019 14:45:54 +0200
Subject: iwlwifi: mvm: update powersave correctly for D3

This fixes a long-standing bug - we haven't been able to check the
firmware image that was loaded for D3/not-D3 since the introduction
of the unified image...

Fix this by keeping a status flag for D3 instead of checking for
the firmware image that's loaded.

This reduces occurrences of checks for IWL_UCODE_WOWLAN to just the
code that actually loads the image or deals with it in other ways.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/d3.c    |  6 ++++++
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h   |  2 ++
 drivers/net/wireless/intel/iwlwifi/mvm/power.c | 27 ++++++++++++--------------
 3 files changed, 20 insertions(+), 15 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
index 43ebb2149b63..8878409d2f07 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
@@ -989,6 +989,8 @@ static int __iwl_mvm_suspend(struct ieee80211_hw *hw,
 
 	mutex_lock(&mvm->mutex);
 
+	set_bit(IWL_MVM_STATUS_IN_D3, &mvm->status);
+
 	vif = iwl_mvm_get_bss_vif(mvm);
 	if (IS_ERR_OR_NULL(vif)) {
 		ret = 1;
@@ -1083,6 +1085,8 @@ static int __iwl_mvm_suspend(struct ieee80211_hw *hw,
 				ieee80211_restart_hw(mvm->hw);
 			}
 		}
+
+		clear_bit(IWL_MVM_STATUS_IN_D3, &mvm->status);
 	}
  out_noreset:
 	mutex_unlock(&mvm->mutex);
@@ -1929,6 +1933,8 @@ static int __iwl_mvm_resume(struct iwl_mvm *mvm, bool test)
 
 	mutex_lock(&mvm->mutex);
 
+	clear_bit(IWL_MVM_STATUS_IN_D3, &mvm->status);
+
 	/* get the BSS vif pointer again */
 	vif = iwl_mvm_get_bss_vif(mvm);
 	if (IS_ERR_OR_NULL(vif))
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 3ec8de00f3aa..3571b71fc51e 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1160,6 +1160,7 @@ struct iwl_mvm {
  * @IWL_MVM_STATUS_ROC_AUX_RUNNING: AUX remain-on-channel is running
  * @IWL_MVM_STATUS_FIRMWARE_RUNNING: firmware is running
  * @IWL_MVM_STATUS_NEED_FLUSH_P2P: need to flush P2P bcast STA
+ * @IWL_MVM_STATUS_IN_D3: in D3 (or at least about to go into it)
  */
 enum iwl_mvm_status {
 	IWL_MVM_STATUS_HW_RFKILL,
@@ -1170,6 +1171,7 @@ enum iwl_mvm_status {
 	IWL_MVM_STATUS_ROC_AUX_RUNNING,
 	IWL_MVM_STATUS_FIRMWARE_RUNNING,
 	IWL_MVM_STATUS_NEED_FLUSH_P2P,
+	IWL_MVM_STATUS_IN_D3,
 };
 
 /* Keep track of completed init configuration */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/power.c b/drivers/net/wireless/intel/iwlwifi/mvm/power.c
index 25d7faea1c62..c146303ec73b 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/power.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/power.c
@@ -198,7 +198,7 @@ static void iwl_mvm_power_configure_uapsd(struct iwl_mvm *mvm,
 		if (!mvmvif->queue_params[ac].uapsd)
 			continue;
 
-		if (mvm->fwrt.cur_fw_img != IWL_UCODE_WOWLAN)
+		if (!test_bit(IWL_MVM_STATUS_IN_D3, &mvm->status))
 			cmd->flags |=
 				cpu_to_le16(POWER_FLAGS_ADVANCE_PM_ENA_MSK);
 
@@ -233,15 +233,15 @@ static void iwl_mvm_power_configure_uapsd(struct iwl_mvm *mvm,
 		cmd->flags |= cpu_to_le16(POWER_FLAGS_SNOOZE_ENA_MSK);
 		cmd->snooze_interval = cpu_to_le16(IWL_MVM_PS_SNOOZE_INTERVAL);
 		cmd->snooze_window =
-			(mvm->fwrt.cur_fw_img == IWL_UCODE_WOWLAN) ?
+			test_bit(IWL_MVM_STATUS_IN_D3, &mvm->status) ?
 				cpu_to_le16(IWL_MVM_WOWLAN_PS_SNOOZE_WINDOW) :
 				cpu_to_le16(IWL_MVM_PS_SNOOZE_WINDOW);
 	}
 
 	cmd->uapsd_max_sp = mvm->hw->uapsd_max_sp_len;
 
-	if (mvm->fwrt.cur_fw_img == IWL_UCODE_WOWLAN || cmd->flags &
-	    cpu_to_le16(POWER_FLAGS_SNOOZE_ENA_MSK)) {
+	if (test_bit(IWL_MVM_STATUS_IN_D3, &mvm->status) ||
+	    cmd->flags & cpu_to_le16(POWER_FLAGS_SNOOZE_ENA_MSK)) {
 		cmd->rx_data_timeout_uapsd =
 			cpu_to_le32(IWL_MVM_WOWLAN_PS_RX_DATA_TIMEOUT);
 		cmd->tx_data_timeout_uapsd =
@@ -354,8 +354,7 @@ static bool iwl_mvm_power_is_radar(struct ieee80211_vif *vif)
 
 static void iwl_mvm_power_config_skip_dtim(struct iwl_mvm *mvm,
 					   struct ieee80211_vif *vif,
-					   struct iwl_mac_power_cmd *cmd,
-					   bool host_awake)
+					   struct iwl_mac_power_cmd *cmd)
 {
 	int dtimper = vif->bss_conf.dtim_period ?: 1;
 	int skip;
@@ -370,7 +369,7 @@ static void iwl_mvm_power_config_skip_dtim(struct iwl_mvm *mvm,
 	if (dtimper >= 10)
 		return;
 
-	if (host_awake) {
+	if (!test_bit(IWL_MVM_STATUS_IN_D3, &mvm->status)) {
 		if (iwlmvm_mod_params.power_scheme != IWL_POWER_SCHEME_LP)
 			return;
 		skip = 2;
@@ -390,8 +389,7 @@ static void iwl_mvm_power_config_skip_dtim(struct iwl_mvm *mvm,
 
 static void iwl_mvm_power_build_cmd(struct iwl_mvm *mvm,
 				    struct ieee80211_vif *vif,
-				    struct iwl_mac_power_cmd *cmd,
-				    bool host_awake)
+				    struct iwl_mac_power_cmd *cmd)
 {
 	int dtimper, bi;
 	int keep_alive;
@@ -437,9 +435,9 @@ static void iwl_mvm_power_build_cmd(struct iwl_mvm *mvm,
 		cmd->lprx_rssi_threshold = POWER_LPRX_RSSI_THRESHOLD;
 	}
 
-	iwl_mvm_power_config_skip_dtim(mvm, vif, cmd, host_awake);
+	iwl_mvm_power_config_skip_dtim(mvm, vif, cmd);
 
-	if (!host_awake) {
+	if (test_bit(IWL_MVM_STATUS_IN_D3, &mvm->status)) {
 		cmd->rx_data_timeout =
 			cpu_to_le32(IWL_MVM_WOWLAN_PS_RX_DATA_TIMEOUT);
 		cmd->tx_data_timeout =
@@ -512,8 +510,7 @@ static int iwl_mvm_power_send_cmd(struct iwl_mvm *mvm,
 {
 	struct iwl_mac_power_cmd cmd = {};
 
-	iwl_mvm_power_build_cmd(mvm, vif, &cmd,
-				mvm->fwrt.cur_fw_img != IWL_UCODE_WOWLAN);
+	iwl_mvm_power_build_cmd(mvm, vif, &cmd);
 	iwl_mvm_power_log(mvm, &cmd);
 #ifdef CONFIG_IWLWIFI_DEBUGFS
 	memcpy(&iwl_mvm_vif_from_mac80211(vif)->mac_pwr_cmd, &cmd, sizeof(cmd));
@@ -536,7 +533,7 @@ int iwl_mvm_power_update_device(struct iwl_mvm *mvm)
 		cmd.flags |= cpu_to_le16(DEVICE_POWER_FLAGS_POWER_SAVE_ENA_MSK);
 
 #ifdef CONFIG_IWLWIFI_DEBUGFS
-	if ((mvm->fwrt.cur_fw_img == IWL_UCODE_WOWLAN) ?
+	if (test_bit(IWL_MVM_STATUS_IN_D3, &mvm->status) ?
 			mvm->disable_power_off_d3 : mvm->disable_power_off)
 		cmd.flags &=
 			cpu_to_le16(~DEVICE_POWER_FLAGS_POWER_SAVE_ENA_MSK);
@@ -943,7 +940,7 @@ static int iwl_mvm_power_set_ba(struct iwl_mvm *mvm,
 	if (!mvmvif->bf_data.bf_enabled)
 		return 0;
 
-	if (mvm->fwrt.cur_fw_img == IWL_UCODE_WOWLAN)
+	if (test_bit(IWL_MVM_STATUS_IN_D3, &mvm->status))
 		cmd.ba_escape_timer = cpu_to_le32(IWL_BA_ESCAPE_TIMER_D3);
 
 	mvmvif->bf_data.ba_enabled = !(!mvmvif->pm_enabled ||
-- 
cgit v1.2.3-59-g8ed1b


From efaaa97c0e7f14c48ee406a37ba25ff3915c6e50 Mon Sep 17 00:00:00 2001
From: Luca Coelho <luciano.coelho@intel.com>
Date: Tue, 1 Oct 2019 14:42:28 +0300
Subject: iwlwifi: dbg_ini: don't skip a TX FIFO when dumping

Before we start looping over the internal TX FIFOs increase the fifo
number, but that's incorrect and causes a FIFO to be skipped.  This is
probably due to a copy and paste from the previous loop.

Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/dbg.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/dbg.c b/drivers/net/wireless/intel/iwlwifi/fw/dbg.c
index ed90dd104366..c8bc6b9b53a4 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/dbg.c
+++ b/drivers/net/wireless/intel/iwlwifi/fw/dbg.c
@@ -1230,13 +1230,15 @@ static bool iwl_ini_txf_iter(struct iwl_fw_runtime *fwrt,
 			iter->lmac = 0;
 	}
 
-	if (!iter->internal_txf)
+	if (!iter->internal_txf) {
 		for (iter->fifo++; iter->fifo < txf_num; iter->fifo++) {
 			iter->fifo_size =
 				cfg->lmac[iter->lmac].txfifo_size[iter->fifo];
 			if (iter->fifo_size && (lmac_bitmap & BIT(iter->fifo)))
 				return true;
 		}
+		iter->fifo--;
+	}
 
 	iter->internal_txf = 1;
 
-- 
cgit v1.2.3-59-g8ed1b


From 5b5b9d35e05a01afacda745a0877105d33f7a768 Mon Sep 17 00:00:00 2001
From: Tova Mussai <tova.mussai@intel.com>
Date: Thu, 3 Oct 2019 20:17:07 +0300
Subject: iwlwifi: scan: remove support for fw scan api v11

The fw already support scan api v12,
v11 is not needed anymore.

Signed-off-by: Tova Mussai <tova.mussai@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/api/scan.h | 41 ---------------------
 drivers/net/wireless/intel/iwlwifi/mvm/scan.c    | 47 ------------------------
 2 files changed, 88 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h b/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h
index 408798f351c6..1b2b5fa56e19 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h
@@ -921,21 +921,6 @@ struct iwl_scan_probe_params_v4 {
 
 #define SCAN_MAX_NUM_CHANS_V3 67
 
-/**
- * struct iwl_scan_channel_params_v3
- * @flags: channel flags &enum iwl_scan_channel_flags
- * @count: num of channels in scan request
- * @reserved: for future use and alignment
- * @channel_config: array of explicit channel configurations
- *                  for 2.4Ghz and 5.2Ghz bands
- */
-struct iwl_scan_channel_params_v3 {
-	u8 flags;
-	u8 count;
-	__le16 reserved;
-	struct iwl_scan_channel_cfg_umac channel_config[SCAN_MAX_NUM_CHANS_V3];
-} __packed; /* SCAN_CHANNEL_PARAMS_API_S_VER_3 */
-
 /**
  * struct iwl_scan_channel_params_v4
  * @flags: channel flags &enum iwl_scan_channel_flags
@@ -1010,20 +995,6 @@ struct iwl_scan_periodic_parms_v1 {
 	__le16 reserved;
 } __packed; /* SCAN_PERIODIC_PARAMS_API_S_VER_1 */
 
-/**
- * struct iwl_scan_req_params_v11
- * @general_params: &struct iwl_scan_general_params_v10
- * @channel_params: &struct iwl_scan_channel_params_v3
- * @periodic_params: &struct iwl_scan_periodic_parms_v1
- * @probe_params: &struct iwl_scan_probe_params_v3
- */
-struct iwl_scan_req_params_v11 {
-	struct iwl_scan_general_params_v10 general_params;
-	struct iwl_scan_channel_params_v3 channel_params;
-	struct iwl_scan_periodic_parms_v1 periodic_params;
-	struct iwl_scan_probe_params_v3 probe_params;
-} __packed; /* SCAN_REQUEST_PARAMS_API_S_VER_11 */
-
 /**
  * struct iwl_scan_req_params_v12
  * @general_params: &struct iwl_scan_general_params_v10
@@ -1052,18 +1023,6 @@ struct iwl_scan_req_params_v13 {
 	struct iwl_scan_probe_params_v4 probe_params;
 } __packed; /* SCAN_REQUEST_PARAMS_API_S_VER_13 */
 
-/**
- * struct iwl_scan_req_umac_v11
- * @uid: scan id, &enum iwl_umac_scan_uid_offsets
- * @ooc_priority: out of channel priority - &enum iwl_scan_priority
- * @scan_params: scan parameters
- */
-struct iwl_scan_req_umac_v11 {
-	__le32 uid;
-	__le32 ooc_priority;
-	struct iwl_scan_req_params_v11 scan_params;
-} __packed; /* SCAN_REQUEST_CMD_UMAC_API_S_VER_11 */
-
 /**
  * struct iwl_scan_req_umac_v12
  * @uid: scan id, &enum iwl_umac_scan_uid_offsets
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
index a046ac9fa852..c0c89d460700 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
@@ -1906,20 +1906,6 @@ iwl_mvm_scan_umac_fill_probe_p_v4(struct iwl_mvm_scan_params *params,
 	iwl_scan_build_ssids(params, pp->direct_scan, bitmap_ssid);
 }
 
-static void
-iwl_mvm_scan_umac_fill_ch_p_v3(struct iwl_mvm *mvm,
-			       struct iwl_mvm_scan_params *params,
-			       struct ieee80211_vif *vif,
-			       struct iwl_scan_channel_params_v3 *cp)
-{
-	cp->flags = iwl_mvm_scan_umac_chan_flags_v2(mvm, params, vif);
-	cp->count = params->n_channels;
-
-	iwl_mvm_umac_scan_cfg_channels(mvm, params->channels,
-				       params->n_channels, 0,
-				       cp->channel_config);
-}
-
 static void
 iwl_mvm_scan_umac_fill_ch_p_v4(struct iwl_mvm *mvm,
 			       struct iwl_mvm_scan_params *params,
@@ -1937,37 +1923,6 @@ iwl_mvm_scan_umac_fill_ch_p_v4(struct iwl_mvm *mvm,
 					  vif->type);
 }
 
-static int iwl_mvm_scan_umac_v11(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
-				 struct iwl_mvm_scan_params *params, int type,
-				 int uid)
-{
-	struct iwl_scan_req_umac_v11 *cmd = mvm->scan_cmd;
-	struct iwl_scan_req_params_v11 *scan_p = &cmd->scan_params;
-	int ret;
-	u16 gen_flags;
-
-	mvm->scan_uid_status[uid] = type;
-
-	cmd->ooc_priority = cpu_to_le32(iwl_mvm_scan_umac_ooc_priority(params));
-	cmd->uid = cpu_to_le32(uid);
-
-	gen_flags = iwl_mvm_scan_umac_flags_v2(mvm, params, vif, type);
-	iwl_mvm_scan_umac_fill_general_p_v10(mvm, params, vif,
-					     &scan_p->general_params,
-					     gen_flags);
-
-	 ret = iwl_mvm_fill_scan_sched_params(params,
-					      scan_p->periodic_params.schedule,
-					      &scan_p->periodic_params.delay);
-	if (ret)
-		return ret;
-
-	iwl_mvm_scan_umac_fill_probe_p_v3(params, &scan_p->probe_params);
-	iwl_mvm_scan_umac_fill_ch_p_v3(mvm, params, vif,
-				       &scan_p->channel_params);
-
-	return 0;
-}
 
 static int iwl_mvm_scan_umac_v12(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 				 struct iwl_mvm_scan_params *params, int type,
@@ -2152,7 +2107,6 @@ static const struct iwl_scan_umac_handler iwl_scan_umac_handlers[] = {
 	/* set the newest version first to shorten the list traverse time */
 	IWL_SCAN_UMAC_HANDLER(13),
 	IWL_SCAN_UMAC_HANDLER(12),
-	IWL_SCAN_UMAC_HANDLER(11),
 };
 
 static int iwl_mvm_build_scan_cmd(struct iwl_mvm *mvm,
@@ -2511,7 +2465,6 @@ static int iwl_scan_req_umac_get_size(u8 scan_ver)
 	switch (scan_ver) {
 		IWL_SCAN_REQ_UMAC_HANDLE_SIZE(13);
 		IWL_SCAN_REQ_UMAC_HANDLE_SIZE(12);
-		IWL_SCAN_REQ_UMAC_HANDLE_SIZE(11);
 	}
 
 	return 0;
-- 
cgit v1.2.3-59-g8ed1b


From 484b165306e18fd9a0d960e6ed2d6eff31665522 Mon Sep 17 00:00:00 2001
From: Magnus Karlsson <magnus.karlsson@intel.com>
Date: Thu, 19 Dec 2019 13:39:20 +0100
Subject: xsk: Eliminate the lazy update threshold

The lazy update threshold was introduced to keep the producer and
consumer some distance apart in the completion ring. This was
important in the beginning of the development of AF_XDP as the ring
format as that point in time was very sensitive to the producer and
consumer being on the same cache line. This is not the case
anymore as the current ring format does not degrade in any noticeable
way when this happens. Moreover, this threshold makes it impossible
to run the system with rings that have less than 128 entries.

So let us remove this threshold and just get one entry from the ring
as in all other functions. This will enable us to remove this function
in a later commit. Note that xskq_produce_addr_lazy followed by
xskq_produce_flush_addr_n are still not the same function as
xskq_produce_addr() as it operates on another cached pointer.

Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-2-git-send-email-magnus.karlsson@intel.com
---
 net/xdp/xsk_queue.h | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/net/xdp/xsk_queue.h b/net/xdp/xsk_queue.h
index eddae4688862..a2f0ba65559f 100644
--- a/net/xdp/xsk_queue.h
+++ b/net/xdp/xsk_queue.h
@@ -11,7 +11,6 @@
 #include <net/xdp_sock.h>
 
 #define RX_BATCH_SIZE 16
-#define LAZY_UPDATE_THRESHOLD 128
 
 struct xdp_ring {
 	u32 producer ____cacheline_aligned_in_smp;
@@ -239,7 +238,7 @@ static inline int xskq_produce_addr_lazy(struct xsk_queue *q, u64 addr)
 {
 	struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring;
 
-	if (xskq_nb_free(q, q->prod_head, LAZY_UPDATE_THRESHOLD) == 0)
+	if (xskq_nb_free(q, q->prod_head, 1) == 0)
 		return -ENOSPC;
 
 	/* A, matches D */
-- 
cgit v1.2.3-59-g8ed1b


From 11cc2d21499cabe7e7964389634ed1de3ee91d33 Mon Sep 17 00:00:00 2001
From: Magnus Karlsson <magnus.karlsson@intel.com>
Date: Thu, 19 Dec 2019 13:39:21 +0100
Subject: xsk: Simplify detection of empty and full rings

In order to set the correct return flags for poll, the xsk code has to
check if the Rx queue is empty and if the Tx queue is full. This code
was unnecessarily large and complex as it used the functions that are
used to update the local state from the global state (xskq_nb_free and
xskq_nb_avail). Since we are not doing this nor updating any data
dependent on this state, we can simplify the functions. Another
benefit from this is that we can also simplify the xskq_nb_free and
xskq_nb_avail functions in a later commit.

Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-3-git-send-email-magnus.karlsson@intel.com
---
 net/xdp/xsk_queue.h | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/net/xdp/xsk_queue.h b/net/xdp/xsk_queue.h
index a2f0ba65559f..b28f9da7b317 100644
--- a/net/xdp/xsk_queue.h
+++ b/net/xdp/xsk_queue.h
@@ -362,12 +362,15 @@ static inline void xskq_produce_flush_desc(struct xsk_queue *q)
 
 static inline bool xskq_full_desc(struct xsk_queue *q)
 {
-	return xskq_nb_avail(q, q->nentries) == q->nentries;
+	/* No barriers needed since data is not accessed */
+	return READ_ONCE(q->ring->producer) - READ_ONCE(q->ring->consumer) ==
+		q->nentries;
 }
 
 static inline bool xskq_empty_desc(struct xsk_queue *q)
 {
-	return xskq_nb_free(q, q->prod_tail, q->nentries) == q->nentries;
+	/* No barriers needed since data is not accessed */
+	return READ_ONCE(q->ring->consumer) == READ_ONCE(q->ring->producer);
 }
 
 void xskq_set_umem(struct xsk_queue *q, u64 size, u64 chunk_mask);
-- 
cgit v1.2.3-59-g8ed1b


From d7012f05e3ca5aba92770c370895092d4882e8c2 Mon Sep 17 00:00:00 2001
From: Magnus Karlsson <magnus.karlsson@intel.com>
Date: Thu, 19 Dec 2019 13:39:22 +0100
Subject: xsk: Consolidate to one single cached producer pointer

Currently, the xsk ring code has two cached producer pointers:
prod_head and prod_tail. This patch consolidates these two into a
single one called cached_prod to make the code simpler and easier to
maintain. This will be in line with the user space part of the the
code found in libbpf, that only uses a single cached pointer.

The Rx path only uses the two top level functions
xskq_produce_batch_desc and xskq_produce_flush_desc and they both use
prod_head and never prod_tail. So just move them over to
cached_prod.

The Tx XDP_DRV path uses xskq_produce_addr_lazy and
xskq_produce_flush_addr_n and unnecessarily operates on both prod_tail
and prod_head, so move them over to just use cached_prod by skipping
the intermediate step of updating prod_tail.

The Tx path in XDP_SKB mode uses xskq_reserve_addr and
xskq_produce_addr. They currently use both cached pointers, but we can
operate on the global producer pointer in xskq_produce_addr since it
has to be updated anyway, thus eliminating the use of both cached
pointers. We can also remove the xskq_nb_free in xskq_produce_addr
since it is already called in xskq_reserve_addr. No need to do it
twice.

When there is only one cached producer pointer, we can also simplify
xskq_nb_free by removing one argument.

Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-4-git-send-email-magnus.karlsson@intel.com
---
 net/xdp/xsk_queue.h | 47 +++++++++++++++++++++--------------------------
 1 file changed, 21 insertions(+), 26 deletions(-)

diff --git a/net/xdp/xsk_queue.h b/net/xdp/xsk_queue.h
index b28f9da7b317..7ad80748f209 100644
--- a/net/xdp/xsk_queue.h
+++ b/net/xdp/xsk_queue.h
@@ -35,8 +35,7 @@ struct xsk_queue {
 	u64 size;
 	u32 ring_mask;
 	u32 nentries;
-	u32 prod_head;
-	u32 prod_tail;
+	u32 cached_prod;
 	u32 cons_head;
 	u32 cons_tail;
 	struct xdp_ring *ring;
@@ -94,39 +93,39 @@ static inline u64 xskq_nb_invalid_descs(struct xsk_queue *q)
 
 static inline u32 xskq_nb_avail(struct xsk_queue *q, u32 dcnt)
 {
-	u32 entries = q->prod_tail - q->cons_tail;
+	u32 entries = q->cached_prod - q->cons_tail;
 
 	if (entries == 0) {
 		/* Refresh the local pointer */
-		q->prod_tail = READ_ONCE(q->ring->producer);
-		entries = q->prod_tail - q->cons_tail;
+		q->cached_prod = READ_ONCE(q->ring->producer);
+		entries = q->cached_prod - q->cons_tail;
 	}
 
 	return (entries > dcnt) ? dcnt : entries;
 }
 
-static inline u32 xskq_nb_free(struct xsk_queue *q, u32 producer, u32 dcnt)
+static inline u32 xskq_nb_free(struct xsk_queue *q, u32 dcnt)
 {
-	u32 free_entries = q->nentries - (producer - q->cons_tail);
+	u32 free_entries = q->nentries - (q->cached_prod - q->cons_tail);
 
 	if (free_entries >= dcnt)
 		return free_entries;
 
 	/* Refresh the local tail pointer */
 	q->cons_tail = READ_ONCE(q->ring->consumer);
-	return q->nentries - (producer - q->cons_tail);
+	return q->nentries - (q->cached_prod - q->cons_tail);
 }
 
 static inline bool xskq_has_addrs(struct xsk_queue *q, u32 cnt)
 {
-	u32 entries = q->prod_tail - q->cons_tail;
+	u32 entries = q->cached_prod - q->cons_tail;
 
 	if (entries >= cnt)
 		return true;
 
 	/* Refresh the local pointer. */
-	q->prod_tail = READ_ONCE(q->ring->producer);
-	entries = q->prod_tail - q->cons_tail;
+	q->cached_prod = READ_ONCE(q->ring->producer);
+	entries = q->cached_prod - q->cons_tail;
 
 	return entries >= cnt;
 }
@@ -220,17 +219,15 @@ static inline void xskq_discard_addr(struct xsk_queue *q)
 static inline int xskq_produce_addr(struct xsk_queue *q, u64 addr)
 {
 	struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring;
-
-	if (xskq_nb_free(q, q->prod_tail, 1) == 0)
-		return -ENOSPC;
+	unsigned int idx = q->ring->producer;
 
 	/* A, matches D */
-	ring->desc[q->prod_tail++ & q->ring_mask] = addr;
+	ring->desc[idx++ & q->ring_mask] = addr;
 
 	/* Order producer and data */
 	smp_wmb(); /* B, matches C */
 
-	WRITE_ONCE(q->ring->producer, q->prod_tail);
+	WRITE_ONCE(q->ring->producer, idx);
 	return 0;
 }
 
@@ -238,11 +235,11 @@ static inline int xskq_produce_addr_lazy(struct xsk_queue *q, u64 addr)
 {
 	struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring;
 
-	if (xskq_nb_free(q, q->prod_head, 1) == 0)
+	if (xskq_nb_free(q, 1) == 0)
 		return -ENOSPC;
 
 	/* A, matches D */
-	ring->desc[q->prod_head++ & q->ring_mask] = addr;
+	ring->desc[q->cached_prod++ & q->ring_mask] = addr;
 	return 0;
 }
 
@@ -252,17 +249,16 @@ static inline void xskq_produce_flush_addr_n(struct xsk_queue *q,
 	/* Order producer and data */
 	smp_wmb(); /* B, matches C */
 
-	q->prod_tail += nb_entries;
-	WRITE_ONCE(q->ring->producer, q->prod_tail);
+	WRITE_ONCE(q->ring->producer, q->ring->producer + nb_entries);
 }
 
 static inline int xskq_reserve_addr(struct xsk_queue *q)
 {
-	if (xskq_nb_free(q, q->prod_head, 1) == 0)
+	if (xskq_nb_free(q, 1) == 0)
 		return -ENOSPC;
 
 	/* A, matches D */
-	q->prod_head++;
+	q->cached_prod++;
 	return 0;
 }
 
@@ -340,11 +336,11 @@ static inline int xskq_produce_batch_desc(struct xsk_queue *q,
 	struct xdp_rxtx_ring *ring = (struct xdp_rxtx_ring *)q->ring;
 	unsigned int idx;
 
-	if (xskq_nb_free(q, q->prod_head, 1) == 0)
+	if (xskq_nb_free(q, 1) == 0)
 		return -ENOSPC;
 
 	/* A, matches D */
-	idx = (q->prod_head++) & q->ring_mask;
+	idx = q->cached_prod++ & q->ring_mask;
 	ring->desc[idx].addr = addr;
 	ring->desc[idx].len = len;
 
@@ -356,8 +352,7 @@ static inline void xskq_produce_flush_desc(struct xsk_queue *q)
 	/* Order producer and data */
 	smp_wmb(); /* B, matches C */
 
-	q->prod_tail = q->prod_head;
-	WRITE_ONCE(q->ring->producer, q->prod_tail);
+	WRITE_ONCE(q->ring->producer, q->cached_prod);
 }
 
 static inline bool xskq_full_desc(struct xsk_queue *q)
-- 
cgit v1.2.3-59-g8ed1b


From 59e35e552529b858f35b30bc5a803ea532ca17f1 Mon Sep 17 00:00:00 2001
From: Magnus Karlsson <magnus.karlsson@intel.com>
Date: Thu, 19 Dec 2019 13:39:23 +0100
Subject: xsk: Standardize naming of producer ring access functions

Adopt the naming of the producer ring access functions to have a
similar naming convention as the functions in libbpf, but adapted to
the kernel. You first reserve a number of entries that you later
submit to the global state of the ring. This is much clearer, IMO,
than the one that was in the kernel part. Once renamed, we also
discover that two functions are actually the same, so remove one of
them. Some of the primitive ring submission operations are also the
same so break these out into __xskq_prod_submit that the upper level
ring access functions can use.

Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-5-git-send-email-magnus.karlsson@intel.com
---
 net/xdp/xsk.c       | 20 +++++++++---------
 net/xdp/xsk_queue.h | 58 +++++++++++++++++++++++++----------------------------
 2 files changed, 37 insertions(+), 41 deletions(-)

diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c
index e45c27f5cfca..5f2123fe630c 100644
--- a/net/xdp/xsk.c
+++ b/net/xdp/xsk.c
@@ -167,7 +167,7 @@ static int __xsk_rcv(struct xdp_sock *xs, struct xdp_buff *xdp, u32 len)
 
 	offset += metalen;
 	addr = xsk_umem_adjust_offset(xs->umem, addr, offset);
-	err = xskq_produce_batch_desc(xs->rx, addr, len);
+	err = xskq_prod_reserve_desc(xs->rx, addr, len);
 	if (!err) {
 		xskq_discard_addr(xs->umem->fq);
 		xdp_return_buff(xdp);
@@ -180,7 +180,7 @@ static int __xsk_rcv(struct xdp_sock *xs, struct xdp_buff *xdp, u32 len)
 
 static int __xsk_rcv_zc(struct xdp_sock *xs, struct xdp_buff *xdp, u32 len)
 {
-	int err = xskq_produce_batch_desc(xs->rx, (u64)xdp->handle, len);
+	int err = xskq_prod_reserve_desc(xs->rx, xdp->handle, len);
 
 	if (err)
 		xs->rx_dropped++;
@@ -216,7 +216,7 @@ static int xsk_rcv(struct xdp_sock *xs, struct xdp_buff *xdp)
 
 static void xsk_flush(struct xdp_sock *xs)
 {
-	xskq_produce_flush_desc(xs->rx);
+	xskq_prod_submit(xs->rx);
 	xs->sk.sk_data_ready(&xs->sk);
 }
 
@@ -247,12 +247,12 @@ int xsk_generic_rcv(struct xdp_sock *xs, struct xdp_buff *xdp)
 	memcpy(buffer, xdp->data_meta, len + metalen);
 
 	addr = xsk_umem_adjust_offset(xs->umem, addr, metalen);
-	err = xskq_produce_batch_desc(xs->rx, addr, len);
+	err = xskq_prod_reserve_desc(xs->rx, addr, len);
 	if (err)
 		goto out_drop;
 
 	xskq_discard_addr(xs->umem->fq);
-	xskq_produce_flush_desc(xs->rx);
+	xskq_prod_submit(xs->rx);
 
 	spin_unlock_bh(&xs->rx_lock);
 
@@ -294,7 +294,7 @@ void __xsk_map_flush(void)
 
 void xsk_umem_complete_tx(struct xdp_umem *umem, u32 nb_entries)
 {
-	xskq_produce_flush_addr_n(umem->cq, nb_entries);
+	xskq_prod_submit_n(umem->cq, nb_entries);
 }
 EXPORT_SYMBOL(xsk_umem_complete_tx);
 
@@ -319,7 +319,7 @@ bool xsk_umem_consume_tx(struct xdp_umem *umem, struct xdp_desc *desc)
 		if (!xskq_peek_desc(xs->tx, desc, umem))
 			continue;
 
-		if (xskq_produce_addr_lazy(umem->cq, desc->addr))
+		if (xskq_prod_reserve_addr(umem->cq, desc->addr))
 			goto out;
 
 		xskq_discard_desc(xs->tx);
@@ -348,7 +348,7 @@ static void xsk_destruct_skb(struct sk_buff *skb)
 	unsigned long flags;
 
 	spin_lock_irqsave(&xs->tx_completion_lock, flags);
-	WARN_ON_ONCE(xskq_produce_addr(xs->umem->cq, addr));
+	xskq_prod_submit_addr(xs->umem->cq, addr);
 	spin_unlock_irqrestore(&xs->tx_completion_lock, flags);
 
 	sock_wfree(skb);
@@ -389,7 +389,7 @@ static int xsk_generic_xmit(struct sock *sk)
 		addr = desc.addr;
 		buffer = xdp_umem_get_data(xs->umem, addr);
 		err = skb_store_bits(skb, 0, buffer, len);
-		if (unlikely(err) || xskq_reserve_addr(xs->umem->cq)) {
+		if (unlikely(err) || xskq_prod_reserve(xs->umem->cq)) {
 			kfree_skb(skb);
 			goto out;
 		}
@@ -470,7 +470,7 @@ static __poll_t xsk_poll(struct file *file, struct socket *sock,
 			__xsk_sendmsg(sk);
 	}
 
-	if (xs->rx && !xskq_empty_desc(xs->rx))
+	if (xs->rx && !xskq_prod_is_empty(xs->rx))
 		mask |= EPOLLIN | EPOLLRDNORM;
 	if (xs->tx && !xskq_full_desc(xs->tx))
 		mask |= EPOLLOUT | EPOLLWRNORM;
diff --git a/net/xdp/xsk_queue.h b/net/xdp/xsk_queue.h
index 7ad80748f209..1b9a350f2e66 100644
--- a/net/xdp/xsk_queue.h
+++ b/net/xdp/xsk_queue.h
@@ -216,22 +216,17 @@ static inline void xskq_discard_addr(struct xsk_queue *q)
 	q->cons_tail++;
 }
 
-static inline int xskq_produce_addr(struct xsk_queue *q, u64 addr)
+static inline int xskq_prod_reserve(struct xsk_queue *q)
 {
-	struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring;
-	unsigned int idx = q->ring->producer;
+	if (xskq_nb_free(q, 1) == 0)
+		return -ENOSPC;
 
 	/* A, matches D */
-	ring->desc[idx++ & q->ring_mask] = addr;
-
-	/* Order producer and data */
-	smp_wmb(); /* B, matches C */
-
-	WRITE_ONCE(q->ring->producer, idx);
+	q->cached_prod++;
 	return 0;
 }
 
-static inline int xskq_produce_addr_lazy(struct xsk_queue *q, u64 addr)
+static inline int xskq_prod_reserve_addr(struct xsk_queue *q, u64 addr)
 {
 	struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring;
 
@@ -243,23 +238,32 @@ static inline int xskq_produce_addr_lazy(struct xsk_queue *q, u64 addr)
 	return 0;
 }
 
-static inline void xskq_produce_flush_addr_n(struct xsk_queue *q,
-					     u32 nb_entries)
+static inline void __xskq_prod_submit(struct xsk_queue *q, u32 idx)
 {
 	/* Order producer and data */
 	smp_wmb(); /* B, matches C */
 
-	WRITE_ONCE(q->ring->producer, q->ring->producer + nb_entries);
+	WRITE_ONCE(q->ring->producer, idx);
 }
 
-static inline int xskq_reserve_addr(struct xsk_queue *q)
+static inline void xskq_prod_submit(struct xsk_queue *q)
 {
-	if (xskq_nb_free(q, 1) == 0)
-		return -ENOSPC;
+	__xskq_prod_submit(q, q->cached_prod);
+}
 
-	/* A, matches D */
-	q->cached_prod++;
-	return 0;
+static inline void xskq_prod_submit_addr(struct xsk_queue *q, u64 addr)
+{
+	struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring;
+	u32 idx = q->ring->producer;
+
+	ring->desc[idx++ & q->ring_mask] = addr;
+
+	__xskq_prod_submit(q, idx);
+}
+
+static inline void xskq_prod_submit_n(struct xsk_queue *q, u32 nb_entries)
+{
+	__xskq_prod_submit(q, q->ring->producer + nb_entries);
 }
 
 /* Rx/Tx queue */
@@ -330,11 +334,11 @@ static inline void xskq_discard_desc(struct xsk_queue *q)
 	q->cons_tail++;
 }
 
-static inline int xskq_produce_batch_desc(struct xsk_queue *q,
-					  u64 addr, u32 len)
+static inline int xskq_prod_reserve_desc(struct xsk_queue *q,
+					 u64 addr, u32 len)
 {
 	struct xdp_rxtx_ring *ring = (struct xdp_rxtx_ring *)q->ring;
-	unsigned int idx;
+	u32 idx;
 
 	if (xskq_nb_free(q, 1) == 0)
 		return -ENOSPC;
@@ -347,14 +351,6 @@ static inline int xskq_produce_batch_desc(struct xsk_queue *q,
 	return 0;
 }
 
-static inline void xskq_produce_flush_desc(struct xsk_queue *q)
-{
-	/* Order producer and data */
-	smp_wmb(); /* B, matches C */
-
-	WRITE_ONCE(q->ring->producer, q->cached_prod);
-}
-
 static inline bool xskq_full_desc(struct xsk_queue *q)
 {
 	/* No barriers needed since data is not accessed */
@@ -362,7 +358,7 @@ static inline bool xskq_full_desc(struct xsk_queue *q)
 		q->nentries;
 }
 
-static inline bool xskq_empty_desc(struct xsk_queue *q)
+static inline bool xskq_prod_is_empty(struct xsk_queue *q)
 {
 	/* No barriers needed since data is not accessed */
 	return READ_ONCE(q->ring->consumer) == READ_ONCE(q->ring->producer);
-- 
cgit v1.2.3-59-g8ed1b


From 4b638f13bab4dbfe8569c84e374e8201a427115c Mon Sep 17 00:00:00 2001
From: Magnus Karlsson <magnus.karlsson@intel.com>
Date: Thu, 19 Dec 2019 13:39:24 +0100
Subject: xsk: Eliminate the RX batch size

In the xsk consumer ring code there is a variable called RX_BATCH_SIZE
that dictates the minimum number of entries that we try to grab from
the fill and Tx rings. In fact, the code always try to grab the
maximum amount of entries from these rings. The only thing this
variable does is to throw an error if there is less than 16 (as it is
defined) entries on the ring. There is no reason to do this and it
will just lead to weird behavior from user space's point of view. So
eliminate this variable.

With this change, we will be able to simplify the xskq_nb_free and
xskq_nb_avail code in the next commit.

Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-6-git-send-email-magnus.karlsson@intel.com
---
 net/xdp/xsk_queue.h | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/net/xdp/xsk_queue.h b/net/xdp/xsk_queue.h
index 1b9a350f2e66..6ca5fed87852 100644
--- a/net/xdp/xsk_queue.h
+++ b/net/xdp/xsk_queue.h
@@ -10,8 +10,6 @@
 #include <linux/if_xdp.h>
 #include <net/xdp_sock.h>
 
-#define RX_BATCH_SIZE 16
-
 struct xdp_ring {
 	u32 producer ____cacheline_aligned_in_smp;
 	u32 consumer ____cacheline_aligned_in_smp;
@@ -202,7 +200,7 @@ static inline u64 *xskq_peek_addr(struct xsk_queue *q, u64 *addr,
 	if (q->cons_tail == q->cons_head) {
 		smp_mb(); /* D, matches A */
 		WRITE_ONCE(q->ring->consumer, q->cons_tail);
-		q->cons_head = q->cons_tail + xskq_nb_avail(q, RX_BATCH_SIZE);
+		q->cons_head = q->cons_tail + xskq_nb_avail(q, 1);
 
 		/* Order consumer and data */
 		smp_rmb();
@@ -320,7 +318,7 @@ static inline struct xdp_desc *xskq_peek_desc(struct xsk_queue *q,
 	if (q->cons_tail == q->cons_head) {
 		smp_mb(); /* D, matches A */
 		WRITE_ONCE(q->ring->consumer, q->cons_tail);
-		q->cons_head = q->cons_tail + xskq_nb_avail(q, RX_BATCH_SIZE);
+		q->cons_head = q->cons_tail + xskq_nb_avail(q, 1);
 
 		/* Order consumer and data */
 		smp_rmb(); /* C, matches B */
-- 
cgit v1.2.3-59-g8ed1b


From df0ae6f78a45e5696427779fc3379c5d75f5d4a5 Mon Sep 17 00:00:00 2001
From: Magnus Karlsson <magnus.karlsson@intel.com>
Date: Thu, 19 Dec 2019 13:39:25 +0100
Subject: xsk: Simplify xskq_nb_avail and xskq_nb_free

At this point, there are no users of the functions xskq_nb_avail and
xskq_nb_free that take any other number of entries argument than 1, so
let us get rid of the second argument that takes the number of
entries.

Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-7-git-send-email-magnus.karlsson@intel.com
---
 net/xdp/xsk_queue.h | 24 +++++++++++++-----------
 1 file changed, 13 insertions(+), 11 deletions(-)

diff --git a/net/xdp/xsk_queue.h b/net/xdp/xsk_queue.h
index 6ca5fed87852..8bfa2ee6864c 100644
--- a/net/xdp/xsk_queue.h
+++ b/net/xdp/xsk_queue.h
@@ -89,7 +89,7 @@ static inline u64 xskq_nb_invalid_descs(struct xsk_queue *q)
 	return q ? q->invalid_descs : 0;
 }
 
-static inline u32 xskq_nb_avail(struct xsk_queue *q, u32 dcnt)
+static inline u32 xskq_nb_avail(struct xsk_queue *q)
 {
 	u32 entries = q->cached_prod - q->cons_tail;
 
@@ -99,19 +99,21 @@ static inline u32 xskq_nb_avail(struct xsk_queue *q, u32 dcnt)
 		entries = q->cached_prod - q->cons_tail;
 	}
 
-	return (entries > dcnt) ? dcnt : entries;
+	return entries;
 }
 
-static inline u32 xskq_nb_free(struct xsk_queue *q, u32 dcnt)
+static inline bool xskq_prod_is_full(struct xsk_queue *q)
 {
 	u32 free_entries = q->nentries - (q->cached_prod - q->cons_tail);
 
-	if (free_entries >= dcnt)
-		return free_entries;
+	if (free_entries)
+		return false;
 
 	/* Refresh the local tail pointer */
 	q->cons_tail = READ_ONCE(q->ring->consumer);
-	return q->nentries - (q->cached_prod - q->cons_tail);
+	free_entries = q->nentries - (q->cached_prod - q->cons_tail);
+
+	return !free_entries;
 }
 
 static inline bool xskq_has_addrs(struct xsk_queue *q, u32 cnt)
@@ -200,7 +202,7 @@ static inline u64 *xskq_peek_addr(struct xsk_queue *q, u64 *addr,
 	if (q->cons_tail == q->cons_head) {
 		smp_mb(); /* D, matches A */
 		WRITE_ONCE(q->ring->consumer, q->cons_tail);
-		q->cons_head = q->cons_tail + xskq_nb_avail(q, 1);
+		q->cons_head = q->cons_tail + xskq_nb_avail(q);
 
 		/* Order consumer and data */
 		smp_rmb();
@@ -216,7 +218,7 @@ static inline void xskq_discard_addr(struct xsk_queue *q)
 
 static inline int xskq_prod_reserve(struct xsk_queue *q)
 {
-	if (xskq_nb_free(q, 1) == 0)
+	if (xskq_prod_is_full(q))
 		return -ENOSPC;
 
 	/* A, matches D */
@@ -228,7 +230,7 @@ static inline int xskq_prod_reserve_addr(struct xsk_queue *q, u64 addr)
 {
 	struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring;
 
-	if (xskq_nb_free(q, 1) == 0)
+	if (xskq_prod_is_full(q))
 		return -ENOSPC;
 
 	/* A, matches D */
@@ -318,7 +320,7 @@ static inline struct xdp_desc *xskq_peek_desc(struct xsk_queue *q,
 	if (q->cons_tail == q->cons_head) {
 		smp_mb(); /* D, matches A */
 		WRITE_ONCE(q->ring->consumer, q->cons_tail);
-		q->cons_head = q->cons_tail + xskq_nb_avail(q, 1);
+		q->cons_head = q->cons_tail + xskq_nb_avail(q);
 
 		/* Order consumer and data */
 		smp_rmb(); /* C, matches B */
@@ -338,7 +340,7 @@ static inline int xskq_prod_reserve_desc(struct xsk_queue *q,
 	struct xdp_rxtx_ring *ring = (struct xdp_rxtx_ring *)q->ring;
 	u32 idx;
 
-	if (xskq_nb_free(q, 1) == 0)
+	if (xskq_prod_is_full(q))
 		return -ENOSPC;
 
 	/* A, matches D */
-- 
cgit v1.2.3-59-g8ed1b


From c5ed924b54c892ee637d2e6889ef83341835a560 Mon Sep 17 00:00:00 2001
From: Magnus Karlsson <magnus.karlsson@intel.com>
Date: Thu, 19 Dec 2019 13:39:26 +0100
Subject: xsk: Simplify the consumer ring access functions

Simplify and refactor consumer ring functions. The consumer first
"peeks" to find descriptors or addresses that are available to
read from the ring, then reads them and finally "releases" these
descriptors once it is done. The two local variables cons_tail
and cons_head are turned into one single variable called
cached_cons. cached_tail referred to the cached value of the
global consumer pointer and will be stored in cached_cons. For
cached_head, we just use cached_prod instead as it was not used
for a consumer queue before. It also better reflects what it
really is now: a cached copy of the producer pointer.

The names of the functions are also renamed in the same manner as
the producer functions. The new functions are called xskq_cons_
followed by what it does.

Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-8-git-send-email-magnus.karlsson@intel.com
---
 net/xdp/xsk.c       |  24 ++++++-------
 net/xdp/xsk_queue.h | 102 ++++++++++++++++++++++++----------------------------
 2 files changed, 58 insertions(+), 68 deletions(-)

diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c
index 5f2123fe630c..cd84b9d0f1e1 100644
--- a/net/xdp/xsk.c
+++ b/net/xdp/xsk.c
@@ -41,19 +41,19 @@ bool xsk_is_setup_for_bpf_map(struct xdp_sock *xs)
 
 bool xsk_umem_has_addrs(struct xdp_umem *umem, u32 cnt)
 {
-	return xskq_has_addrs(umem->fq, cnt);
+	return xskq_cons_has_entries(umem->fq, cnt);
 }
 EXPORT_SYMBOL(xsk_umem_has_addrs);
 
 u64 *xsk_umem_peek_addr(struct xdp_umem *umem, u64 *addr)
 {
-	return xskq_peek_addr(umem->fq, addr, umem);
+	return xskq_cons_peek_addr(umem->fq, addr, umem);
 }
 EXPORT_SYMBOL(xsk_umem_peek_addr);
 
 void xsk_umem_discard_addr(struct xdp_umem *umem)
 {
-	xskq_discard_addr(umem->fq);
+	xskq_cons_release(umem->fq);
 }
 EXPORT_SYMBOL(xsk_umem_discard_addr);
 
@@ -148,7 +148,7 @@ static int __xsk_rcv(struct xdp_sock *xs, struct xdp_buff *xdp, u32 len)
 	u32 metalen;
 	int err;
 
-	if (!xskq_peek_addr(xs->umem->fq, &addr, xs->umem) ||
+	if (!xskq_cons_peek_addr(xs->umem->fq, &addr, xs->umem) ||
 	    len > xs->umem->chunk_size_nohr - XDP_PACKET_HEADROOM) {
 		xs->rx_dropped++;
 		return -ENOSPC;
@@ -169,7 +169,7 @@ static int __xsk_rcv(struct xdp_sock *xs, struct xdp_buff *xdp, u32 len)
 	addr = xsk_umem_adjust_offset(xs->umem, addr, offset);
 	err = xskq_prod_reserve_desc(xs->rx, addr, len);
 	if (!err) {
-		xskq_discard_addr(xs->umem->fq);
+		xskq_cons_release(xs->umem->fq);
 		xdp_return_buff(xdp);
 		return 0;
 	}
@@ -236,7 +236,7 @@ int xsk_generic_rcv(struct xdp_sock *xs, struct xdp_buff *xdp)
 		goto out_unlock;
 	}
 
-	if (!xskq_peek_addr(xs->umem->fq, &addr, xs->umem) ||
+	if (!xskq_cons_peek_addr(xs->umem->fq, &addr, xs->umem) ||
 	    len > xs->umem->chunk_size_nohr - XDP_PACKET_HEADROOM) {
 		err = -ENOSPC;
 		goto out_drop;
@@ -251,7 +251,7 @@ int xsk_generic_rcv(struct xdp_sock *xs, struct xdp_buff *xdp)
 	if (err)
 		goto out_drop;
 
-	xskq_discard_addr(xs->umem->fq);
+	xskq_cons_release(xs->umem->fq);
 	xskq_prod_submit(xs->rx);
 
 	spin_unlock_bh(&xs->rx_lock);
@@ -316,13 +316,13 @@ bool xsk_umem_consume_tx(struct xdp_umem *umem, struct xdp_desc *desc)
 
 	rcu_read_lock();
 	list_for_each_entry_rcu(xs, &umem->xsk_list, list) {
-		if (!xskq_peek_desc(xs->tx, desc, umem))
+		if (!xskq_cons_peek_desc(xs->tx, desc, umem))
 			continue;
 
 		if (xskq_prod_reserve_addr(umem->cq, desc->addr))
 			goto out;
 
-		xskq_discard_desc(xs->tx);
+		xskq_cons_release(xs->tx);
 		rcu_read_unlock();
 		return true;
 	}
@@ -368,7 +368,7 @@ static int xsk_generic_xmit(struct sock *sk)
 	if (xs->queue_id >= xs->dev->real_num_tx_queues)
 		goto out;
 
-	while (xskq_peek_desc(xs->tx, &desc, xs->umem)) {
+	while (xskq_cons_peek_desc(xs->tx, &desc, xs->umem)) {
 		char *buffer;
 		u64 addr;
 		u32 len;
@@ -401,7 +401,7 @@ static int xsk_generic_xmit(struct sock *sk)
 		skb->destructor = xsk_destruct_skb;
 
 		err = dev_direct_xmit(skb, xs->queue_id);
-		xskq_discard_desc(xs->tx);
+		xskq_cons_release(xs->tx);
 		/* Ignore NET_XMIT_CN as packet might have been sent */
 		if (err == NET_XMIT_DROP || err == NETDEV_TX_BUSY) {
 			/* SKB completed but not sent */
@@ -472,7 +472,7 @@ static __poll_t xsk_poll(struct file *file, struct socket *sock,
 
 	if (xs->rx && !xskq_prod_is_empty(xs->rx))
 		mask |= EPOLLIN | EPOLLRDNORM;
-	if (xs->tx && !xskq_full_desc(xs->tx))
+	if (xs->tx && !xskq_cons_is_full(xs->tx))
 		mask |= EPOLLOUT | EPOLLWRNORM;
 
 	return mask;
diff --git a/net/xdp/xsk_queue.h b/net/xdp/xsk_queue.h
index 8bfa2ee6864c..1436116767ea 100644
--- a/net/xdp/xsk_queue.h
+++ b/net/xdp/xsk_queue.h
@@ -34,8 +34,7 @@ struct xsk_queue {
 	u32 ring_mask;
 	u32 nentries;
 	u32 cached_prod;
-	u32 cons_head;
-	u32 cons_tail;
+	u32 cached_cons;
 	struct xdp_ring *ring;
 	u64 invalid_descs;
 };
@@ -89,43 +88,48 @@ static inline u64 xskq_nb_invalid_descs(struct xsk_queue *q)
 	return q ? q->invalid_descs : 0;
 }
 
-static inline u32 xskq_nb_avail(struct xsk_queue *q)
+static inline void __xskq_cons_release(struct xsk_queue *q)
 {
-	u32 entries = q->cached_prod - q->cons_tail;
+	smp_mb(); /* D, matches A */
+	WRITE_ONCE(q->ring->consumer, q->cached_cons);
+}
 
-	if (entries == 0) {
-		/* Refresh the local pointer */
-		q->cached_prod = READ_ONCE(q->ring->producer);
-		entries = q->cached_prod - q->cons_tail;
-	}
+static inline void __xskq_cons_peek(struct xsk_queue *q)
+{
+	/* Refresh the local pointer */
+	q->cached_prod = READ_ONCE(q->ring->producer);
+	smp_rmb(); /* C, matches B */
+}
 
-	return entries;
+static inline void xskq_cons_get_entries(struct xsk_queue *q)
+{
+	__xskq_cons_release(q);
+	__xskq_cons_peek(q);
 }
 
 static inline bool xskq_prod_is_full(struct xsk_queue *q)
 {
-	u32 free_entries = q->nentries - (q->cached_prod - q->cons_tail);
+	u32 free_entries = q->nentries - (q->cached_prod - q->cached_cons);
 
 	if (free_entries)
 		return false;
 
 	/* Refresh the local tail pointer */
-	q->cons_tail = READ_ONCE(q->ring->consumer);
-	free_entries = q->nentries - (q->cached_prod - q->cons_tail);
+	q->cached_cons = READ_ONCE(q->ring->consumer);
+	free_entries = q->nentries - (q->cached_prod - q->cached_cons);
 
 	return !free_entries;
 }
 
-static inline bool xskq_has_addrs(struct xsk_queue *q, u32 cnt)
+static inline bool xskq_cons_has_entries(struct xsk_queue *q, u32 cnt)
 {
-	u32 entries = q->cached_prod - q->cons_tail;
+	u32 entries = q->cached_prod - q->cached_cons;
 
 	if (entries >= cnt)
 		return true;
 
-	/* Refresh the local pointer. */
-	q->cached_prod = READ_ONCE(q->ring->producer);
-	entries = q->cached_prod - q->cons_tail;
+	__xskq_cons_peek(q);
+	entries = q->cached_prod - q->cached_cons;
 
 	return entries >= cnt;
 }
@@ -172,9 +176,10 @@ static inline bool xskq_is_valid_addr_unaligned(struct xsk_queue *q, u64 addr,
 static inline u64 *xskq_validate_addr(struct xsk_queue *q, u64 *addr,
 				      struct xdp_umem *umem)
 {
-	while (q->cons_tail != q->cons_head) {
-		struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring;
-		unsigned int idx = q->cons_tail & q->ring_mask;
+	struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring;
+
+	while (q->cached_cons != q->cached_prod) {
+		u32 idx = q->cached_cons & q->ring_mask;
 
 		*addr = READ_ONCE(ring->desc[idx]) & q->chunk_mask;
 
@@ -190,30 +195,27 @@ static inline u64 *xskq_validate_addr(struct xsk_queue *q, u64 *addr,
 			return addr;
 
 out:
-		q->cons_tail++;
+		q->cached_cons++;
 	}
 
 	return NULL;
 }
 
-static inline u64 *xskq_peek_addr(struct xsk_queue *q, u64 *addr,
-				  struct xdp_umem *umem)
+static inline u64 *xskq_cons_peek_addr(struct xsk_queue *q, u64 *addr,
+				       struct xdp_umem *umem)
 {
-	if (q->cons_tail == q->cons_head) {
-		smp_mb(); /* D, matches A */
-		WRITE_ONCE(q->ring->consumer, q->cons_tail);
-		q->cons_head = q->cons_tail + xskq_nb_avail(q);
-
-		/* Order consumer and data */
-		smp_rmb();
-	}
-
+	if (q->cached_prod == q->cached_cons)
+		xskq_cons_get_entries(q);
 	return xskq_validate_addr(q, addr, umem);
 }
 
-static inline void xskq_discard_addr(struct xsk_queue *q)
+static inline void xskq_cons_release(struct xsk_queue *q)
 {
-	q->cons_tail++;
+	/* To improve performance, only update local state here.
+	 * Do the actual release operation when we get new entries
+	 * from the ring in xskq_cons_get_entries() instead.
+	 */
+	q->cached_cons++;
 }
 
 static inline int xskq_prod_reserve(struct xsk_queue *q)
@@ -299,41 +301,29 @@ static inline struct xdp_desc *xskq_validate_desc(struct xsk_queue *q,
 						  struct xdp_desc *desc,
 						  struct xdp_umem *umem)
 {
-	while (q->cons_tail != q->cons_head) {
+	while (q->cached_cons != q->cached_prod) {
 		struct xdp_rxtx_ring *ring = (struct xdp_rxtx_ring *)q->ring;
-		unsigned int idx = q->cons_tail & q->ring_mask;
+		u32 idx = q->cached_cons & q->ring_mask;
 
 		*desc = READ_ONCE(ring->desc[idx]);
 		if (xskq_is_valid_desc(q, desc, umem))
 			return desc;
 
-		q->cons_tail++;
+		q->cached_cons++;
 	}
 
 	return NULL;
 }
 
-static inline struct xdp_desc *xskq_peek_desc(struct xsk_queue *q,
-					      struct xdp_desc *desc,
-					      struct xdp_umem *umem)
+static inline struct xdp_desc *xskq_cons_peek_desc(struct xsk_queue *q,
+						   struct xdp_desc *desc,
+						   struct xdp_umem *umem)
 {
-	if (q->cons_tail == q->cons_head) {
-		smp_mb(); /* D, matches A */
-		WRITE_ONCE(q->ring->consumer, q->cons_tail);
-		q->cons_head = q->cons_tail + xskq_nb_avail(q);
-
-		/* Order consumer and data */
-		smp_rmb(); /* C, matches B */
-	}
-
+	if (q->cached_prod == q->cached_cons)
+		xskq_cons_get_entries(q);
 	return xskq_validate_desc(q, desc, umem);
 }
 
-static inline void xskq_discard_desc(struct xsk_queue *q)
-{
-	q->cons_tail++;
-}
-
 static inline int xskq_prod_reserve_desc(struct xsk_queue *q,
 					 u64 addr, u32 len)
 {
@@ -351,7 +341,7 @@ static inline int xskq_prod_reserve_desc(struct xsk_queue *q,
 	return 0;
 }
 
-static inline bool xskq_full_desc(struct xsk_queue *q)
+static inline bool xskq_cons_is_full(struct xsk_queue *q)
 {
 	/* No barriers needed since data is not accessed */
 	return READ_ONCE(q->ring->producer) - READ_ONCE(q->ring->consumer) ==
-- 
cgit v1.2.3-59-g8ed1b


From 03896ef1f0cb23d2742ddf486c531c700a2da7d6 Mon Sep 17 00:00:00 2001
From: Magnus Karlsson <magnus.karlsson@intel.com>
Date: Thu, 19 Dec 2019 13:39:27 +0100
Subject: xsk: Change names of validation functions

Change the names of the validation functions to better reflect what
they are doing. The uppermost ones are reading entries from the rings
and only the bottom ones validate entries. So xskq_cons_read_ is a
better prefix name.

Also change the xskq_cons_read_ functions to return a bool
as the the descriptor or address is already returned by reference
in the parameters. Everyone is using the return value as a bool
anyway.

Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-9-git-send-email-magnus.karlsson@intel.com
---
 include/net/xdp_sock.h |  4 ++--
 net/xdp/xsk.c          |  4 ++--
 net/xdp/xsk_queue.h    | 59 ++++++++++++++++++++++++++------------------------
 3 files changed, 35 insertions(+), 32 deletions(-)

diff --git a/include/net/xdp_sock.h b/include/net/xdp_sock.h
index 48594740d67c..63f005830866 100644
--- a/include/net/xdp_sock.h
+++ b/include/net/xdp_sock.h
@@ -118,7 +118,7 @@ int xsk_generic_rcv(struct xdp_sock *xs, struct xdp_buff *xdp);
 bool xsk_is_setup_for_bpf_map(struct xdp_sock *xs);
 /* Used from netdev driver */
 bool xsk_umem_has_addrs(struct xdp_umem *umem, u32 cnt);
-u64 *xsk_umem_peek_addr(struct xdp_umem *umem, u64 *addr);
+bool xsk_umem_peek_addr(struct xdp_umem *umem, u64 *addr);
 void xsk_umem_discard_addr(struct xdp_umem *umem);
 void xsk_umem_complete_tx(struct xdp_umem *umem, u32 nb_entries);
 bool xsk_umem_consume_tx(struct xdp_umem *umem, struct xdp_desc *desc);
@@ -197,7 +197,7 @@ static inline bool xsk_umem_has_addrs_rq(struct xdp_umem *umem, u32 cnt)
 	return xsk_umem_has_addrs(umem, cnt - rq->length);
 }
 
-static inline u64 *xsk_umem_peek_addr_rq(struct xdp_umem *umem, u64 *addr)
+static inline bool xsk_umem_peek_addr_rq(struct xdp_umem *umem, u64 *addr)
 {
 	struct xdp_umem_fq_reuse *rq = umem->fq_reuse;
 
diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c
index cd84b9d0f1e1..4e932a930b3a 100644
--- a/net/xdp/xsk.c
+++ b/net/xdp/xsk.c
@@ -45,7 +45,7 @@ bool xsk_umem_has_addrs(struct xdp_umem *umem, u32 cnt)
 }
 EXPORT_SYMBOL(xsk_umem_has_addrs);
 
-u64 *xsk_umem_peek_addr(struct xdp_umem *umem, u64 *addr)
+bool xsk_umem_peek_addr(struct xdp_umem *umem, u64 *addr)
 {
 	return xskq_cons_peek_addr(umem->fq, addr, umem);
 }
@@ -126,7 +126,7 @@ static void __xsk_rcv_memcpy(struct xdp_umem *umem, u64 addr, void *from_buf,
 	void *to_buf = xdp_umem_get_data(umem, addr);
 
 	addr = xsk_umem_add_offset_to_addr(addr);
-	if (xskq_crosses_non_contig_pg(umem, addr, len + metalen)) {
+	if (xskq_cons_crosses_non_contig_pg(umem, addr, len + metalen)) {
 		void *next_pg_addr = umem->pages[(addr >> PAGE_SHIFT) + 1].addr;
 		u64 page_start = addr & ~(PAGE_SIZE - 1);
 		u64 first_len = PAGE_SIZE - (addr - page_start);
diff --git a/net/xdp/xsk_queue.h b/net/xdp/xsk_queue.h
index 1436116767ea..6d04a96dd092 100644
--- a/net/xdp/xsk_queue.h
+++ b/net/xdp/xsk_queue.h
@@ -136,8 +136,9 @@ static inline bool xskq_cons_has_entries(struct xsk_queue *q, u32 cnt)
 
 /* UMEM queue */
 
-static inline bool xskq_crosses_non_contig_pg(struct xdp_umem *umem, u64 addr,
-					      u64 length)
+static inline bool xskq_cons_crosses_non_contig_pg(struct xdp_umem *umem,
+						   u64 addr,
+						   u64 length)
 {
 	bool cross_pg = (addr & (PAGE_SIZE - 1)) + length > PAGE_SIZE;
 	bool next_pg_contig =
@@ -147,7 +148,7 @@ static inline bool xskq_crosses_non_contig_pg(struct xdp_umem *umem, u64 addr,
 	return cross_pg && !next_pg_contig;
 }
 
-static inline bool xskq_is_valid_addr(struct xsk_queue *q, u64 addr)
+static inline bool xskq_cons_is_valid_addr(struct xsk_queue *q, u64 addr)
 {
 	if (addr >= q->size) {
 		q->invalid_descs++;
@@ -157,7 +158,8 @@ static inline bool xskq_is_valid_addr(struct xsk_queue *q, u64 addr)
 	return true;
 }
 
-static inline bool xskq_is_valid_addr_unaligned(struct xsk_queue *q, u64 addr,
+static inline bool xskq_cons_is_valid_unaligned(struct xsk_queue *q,
+						u64 addr,
 						u64 length,
 						struct xdp_umem *umem)
 {
@@ -165,7 +167,7 @@ static inline bool xskq_is_valid_addr_unaligned(struct xsk_queue *q, u64 addr,
 
 	addr = xsk_umem_add_offset_to_addr(addr);
 	if (base_addr >= q->size || addr >= q->size ||
-	    xskq_crosses_non_contig_pg(umem, addr, length)) {
+	    xskq_cons_crosses_non_contig_pg(umem, addr, length)) {
 		q->invalid_descs++;
 		return false;
 	}
@@ -173,8 +175,8 @@ static inline bool xskq_is_valid_addr_unaligned(struct xsk_queue *q, u64 addr,
 	return true;
 }
 
-static inline u64 *xskq_validate_addr(struct xsk_queue *q, u64 *addr,
-				      struct xdp_umem *umem)
+static inline bool xskq_cons_read_addr(struct xsk_queue *q, u64 *addr,
+				       struct xdp_umem *umem)
 {
 	struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring;
 
@@ -184,29 +186,29 @@ static inline u64 *xskq_validate_addr(struct xsk_queue *q, u64 *addr,
 		*addr = READ_ONCE(ring->desc[idx]) & q->chunk_mask;
 
 		if (umem->flags & XDP_UMEM_UNALIGNED_CHUNK_FLAG) {
-			if (xskq_is_valid_addr_unaligned(q, *addr,
+			if (xskq_cons_is_valid_unaligned(q, *addr,
 							 umem->chunk_size_nohr,
 							 umem))
-				return addr;
+				return true;
 			goto out;
 		}
 
-		if (xskq_is_valid_addr(q, *addr))
-			return addr;
+		if (xskq_cons_is_valid_addr(q, *addr))
+			return true;
 
 out:
 		q->cached_cons++;
 	}
 
-	return NULL;
+	return false;
 }
 
-static inline u64 *xskq_cons_peek_addr(struct xsk_queue *q, u64 *addr,
+static inline bool xskq_cons_peek_addr(struct xsk_queue *q, u64 *addr,
 				       struct xdp_umem *umem)
 {
 	if (q->cached_prod == q->cached_cons)
 		xskq_cons_get_entries(q);
-	return xskq_validate_addr(q, addr, umem);
+	return xskq_cons_read_addr(q, addr, umem);
 }
 
 static inline void xskq_cons_release(struct xsk_queue *q)
@@ -270,11 +272,12 @@ static inline void xskq_prod_submit_n(struct xsk_queue *q, u32 nb_entries)
 
 /* Rx/Tx queue */
 
-static inline bool xskq_is_valid_desc(struct xsk_queue *q, struct xdp_desc *d,
-				      struct xdp_umem *umem)
+static inline bool xskq_cons_is_valid_desc(struct xsk_queue *q,
+					   struct xdp_desc *d,
+					   struct xdp_umem *umem)
 {
 	if (umem->flags & XDP_UMEM_UNALIGNED_CHUNK_FLAG) {
-		if (!xskq_is_valid_addr_unaligned(q, d->addr, d->len, umem))
+		if (!xskq_cons_is_valid_unaligned(q, d->addr, d->len, umem))
 			return false;
 
 		if (d->len > umem->chunk_size_nohr || d->options) {
@@ -285,7 +288,7 @@ static inline bool xskq_is_valid_desc(struct xsk_queue *q, struct xdp_desc *d,
 		return true;
 	}
 
-	if (!xskq_is_valid_addr(q, d->addr))
+	if (!xskq_cons_is_valid_addr(q, d->addr))
 		return false;
 
 	if (((d->addr + d->len) & q->chunk_mask) != (d->addr & q->chunk_mask) ||
@@ -297,31 +300,31 @@ static inline bool xskq_is_valid_desc(struct xsk_queue *q, struct xdp_desc *d,
 	return true;
 }
 
-static inline struct xdp_desc *xskq_validate_desc(struct xsk_queue *q,
-						  struct xdp_desc *desc,
-						  struct xdp_umem *umem)
+static inline bool xskq_cons_read_desc(struct xsk_queue *q,
+				       struct xdp_desc *desc,
+				       struct xdp_umem *umem)
 {
 	while (q->cached_cons != q->cached_prod) {
 		struct xdp_rxtx_ring *ring = (struct xdp_rxtx_ring *)q->ring;
 		u32 idx = q->cached_cons & q->ring_mask;
 
 		*desc = READ_ONCE(ring->desc[idx]);
-		if (xskq_is_valid_desc(q, desc, umem))
-			return desc;
+		if (xskq_cons_is_valid_desc(q, desc, umem))
+			return true;
 
 		q->cached_cons++;
 	}
 
-	return NULL;
+	return false;
 }
 
-static inline struct xdp_desc *xskq_cons_peek_desc(struct xsk_queue *q,
-						   struct xdp_desc *desc,
-						   struct xdp_umem *umem)
+static inline bool xskq_cons_peek_desc(struct xsk_queue *q,
+				       struct xdp_desc *desc,
+				       struct xdp_umem *umem)
 {
 	if (q->cached_prod == q->cached_cons)
 		xskq_cons_get_entries(q);
-	return xskq_validate_desc(q, desc, umem);
+	return xskq_cons_read_desc(q, desc, umem);
 }
 
 static inline int xskq_prod_reserve_desc(struct xsk_queue *q,
-- 
cgit v1.2.3-59-g8ed1b


From f8509aa078de0842ec1817e8026e58620cd05d3b Mon Sep 17 00:00:00 2001
From: Magnus Karlsson <magnus.karlsson@intel.com>
Date: Thu, 19 Dec 2019 13:39:28 +0100
Subject: xsk: ixgbe: i40e: ice: mlx5: Xsk_umem_discard_addr to
 xsk_umem_release_addr

Change the name of xsk_umem_discard_addr to xsk_umem_release_addr to
better reflect the new naming of the AF_XDP queue manipulation
functions. As this functions is used by drivers implementing support
for AF_XDP zero-copy, it requires a name change to these drivers. The
function xsk_umem_release_addr_rq has also changed name in the same
fashion.

Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-10-git-send-email-magnus.karlsson@intel.com
---
 drivers/net/ethernet/intel/i40e/i40e_xsk.c          |  4 ++--
 drivers/net/ethernet/intel/ice/ice_xsk.c            |  4 ++--
 drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c        |  4 ++--
 drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c |  2 +-
 include/net/xdp_sock.h                              | 10 +++++-----
 net/xdp/xsk.c                                       |  4 ++--
 6 files changed, 14 insertions(+), 14 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_xsk.c b/drivers/net/ethernet/intel/i40e/i40e_xsk.c
index d07e1a890428..20d6f11a3468 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_xsk.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_xsk.c
@@ -269,7 +269,7 @@ static bool i40e_alloc_buffer_zc(struct i40e_ring *rx_ring,
 
 	bi->handle = xsk_umem_adjust_offset(umem, handle, umem->headroom);
 
-	xsk_umem_discard_addr(umem);
+	xsk_umem_release_addr(umem);
 	return true;
 }
 
@@ -306,7 +306,7 @@ static bool i40e_alloc_buffer_slow_zc(struct i40e_ring *rx_ring,
 
 	bi->handle = xsk_umem_adjust_offset(umem, handle, umem->headroom);
 
-	xsk_umem_discard_addr_rq(umem);
+	xsk_umem_release_addr_rq(umem);
 	return true;
 }
 
diff --git a/drivers/net/ethernet/intel/ice/ice_xsk.c b/drivers/net/ethernet/intel/ice/ice_xsk.c
index cf9b8b22d24f..0af1f0b951c0 100644
--- a/drivers/net/ethernet/intel/ice/ice_xsk.c
+++ b/drivers/net/ethernet/intel/ice/ice_xsk.c
@@ -555,7 +555,7 @@ ice_alloc_buf_fast_zc(struct ice_ring *rx_ring, struct ice_rx_buf *rx_buf)
 
 	rx_buf->handle = handle + umem->headroom;
 
-	xsk_umem_discard_addr(umem);
+	xsk_umem_release_addr(umem);
 	return true;
 }
 
@@ -591,7 +591,7 @@ ice_alloc_buf_slow_zc(struct ice_ring *rx_ring, struct ice_rx_buf *rx_buf)
 
 	rx_buf->handle = handle + umem->headroom;
 
-	xsk_umem_discard_addr_rq(umem);
+	xsk_umem_release_addr_rq(umem);
 	return true;
 }
 
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c
index d6feaacfbf89..1501d0a22078 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c
@@ -277,7 +277,7 @@ static bool ixgbe_alloc_buffer_zc(struct ixgbe_ring *rx_ring,
 
 	bi->handle = xsk_umem_adjust_offset(umem, handle, umem->headroom);
 
-	xsk_umem_discard_addr(umem);
+	xsk_umem_release_addr(umem);
 	return true;
 }
 
@@ -304,7 +304,7 @@ static bool ixgbe_alloc_buffer_slow_zc(struct ixgbe_ring *rx_ring,
 
 	bi->handle = xsk_umem_adjust_offset(umem, handle, umem->headroom);
 
-	xsk_umem_discard_addr_rq(umem);
+	xsk_umem_release_addr_rq(umem);
 	return true;
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c
index 475b6bd5d29b..62fc8a128a8d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c
@@ -35,7 +35,7 @@ int mlx5e_xsk_page_alloc_umem(struct mlx5e_rq *rq,
 	 */
 	dma_info->addr = xdp_umem_get_dma(umem, handle);
 
-	xsk_umem_discard_addr_rq(umem);
+	xsk_umem_release_addr_rq(umem);
 
 	dma_sync_single_for_device(rq->pdev, dma_info->addr, PAGE_SIZE,
 				   DMA_BIDIRECTIONAL);
diff --git a/include/net/xdp_sock.h b/include/net/xdp_sock.h
index 63f005830866..e86ec48ef627 100644
--- a/include/net/xdp_sock.h
+++ b/include/net/xdp_sock.h
@@ -119,7 +119,7 @@ bool xsk_is_setup_for_bpf_map(struct xdp_sock *xs);
 /* Used from netdev driver */
 bool xsk_umem_has_addrs(struct xdp_umem *umem, u32 cnt);
 bool xsk_umem_peek_addr(struct xdp_umem *umem, u64 *addr);
-void xsk_umem_discard_addr(struct xdp_umem *umem);
+void xsk_umem_release_addr(struct xdp_umem *umem);
 void xsk_umem_complete_tx(struct xdp_umem *umem, u32 nb_entries);
 bool xsk_umem_consume_tx(struct xdp_umem *umem, struct xdp_desc *desc);
 void xsk_umem_consume_tx_done(struct xdp_umem *umem);
@@ -208,12 +208,12 @@ static inline bool xsk_umem_peek_addr_rq(struct xdp_umem *umem, u64 *addr)
 	return addr;
 }
 
-static inline void xsk_umem_discard_addr_rq(struct xdp_umem *umem)
+static inline void xsk_umem_release_addr_rq(struct xdp_umem *umem)
 {
 	struct xdp_umem_fq_reuse *rq = umem->fq_reuse;
 
 	if (!rq->length)
-		xsk_umem_discard_addr(umem);
+		xsk_umem_release_addr(umem);
 	else
 		rq->length--;
 }
@@ -258,7 +258,7 @@ static inline u64 *xsk_umem_peek_addr(struct xdp_umem *umem, u64 *addr)
 	return NULL;
 }
 
-static inline void xsk_umem_discard_addr(struct xdp_umem *umem)
+static inline void xsk_umem_release_addr(struct xdp_umem *umem)
 {
 }
 
@@ -332,7 +332,7 @@ static inline u64 *xsk_umem_peek_addr_rq(struct xdp_umem *umem, u64 *addr)
 	return NULL;
 }
 
-static inline void xsk_umem_discard_addr_rq(struct xdp_umem *umem)
+static inline void xsk_umem_release_addr_rq(struct xdp_umem *umem)
 {
 }
 
diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c
index 4e932a930b3a..55092feeaf36 100644
--- a/net/xdp/xsk.c
+++ b/net/xdp/xsk.c
@@ -51,11 +51,11 @@ bool xsk_umem_peek_addr(struct xdp_umem *umem, u64 *addr)
 }
 EXPORT_SYMBOL(xsk_umem_peek_addr);
 
-void xsk_umem_discard_addr(struct xdp_umem *umem)
+void xsk_umem_release_addr(struct xdp_umem *umem)
 {
 	xskq_cons_release(umem->fq);
 }
-EXPORT_SYMBOL(xsk_umem_discard_addr);
+EXPORT_SYMBOL(xsk_umem_release_addr);
 
 void xsk_set_rx_need_wakeup(struct xdp_umem *umem)
 {
-- 
cgit v1.2.3-59-g8ed1b


From c34787fcc90f0732ff00754057f11780007002e4 Mon Sep 17 00:00:00 2001
From: Magnus Karlsson <magnus.karlsson@intel.com>
Date: Thu, 19 Dec 2019 13:39:29 +0100
Subject: xsk: Remove unnecessary READ_ONCE of data

There are two unnecessary READ_ONCE of descriptor data. These are not
needed since the data is written by the producer before it signals
that the data is available by incrementing the producer pointer. As the
access to this producer pointer is serialized and the consumer always
reads the descriptor after it has read and synchronized with the
producer counter, the write of the descriptor will have fully
completed and it does not matter if the consumer has any read tearing.

Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-11-git-send-email-magnus.karlsson@intel.com
---
 net/xdp/xsk_queue.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/xdp/xsk_queue.h b/net/xdp/xsk_queue.h
index 6d04a96dd092..4c049cacc693 100644
--- a/net/xdp/xsk_queue.h
+++ b/net/xdp/xsk_queue.h
@@ -183,7 +183,7 @@ static inline bool xskq_cons_read_addr(struct xsk_queue *q, u64 *addr,
 	while (q->cached_cons != q->cached_prod) {
 		u32 idx = q->cached_cons & q->ring_mask;
 
-		*addr = READ_ONCE(ring->desc[idx]) & q->chunk_mask;
+		*addr = ring->desc[idx] & q->chunk_mask;
 
 		if (umem->flags & XDP_UMEM_UNALIGNED_CHUNK_FLAG) {
 			if (xskq_cons_is_valid_unaligned(q, *addr,
@@ -308,7 +308,7 @@ static inline bool xskq_cons_read_desc(struct xsk_queue *q,
 		struct xdp_rxtx_ring *ring = (struct xdp_rxtx_ring *)q->ring;
 		u32 idx = q->cached_cons & q->ring_mask;
 
-		*desc = READ_ONCE(ring->desc[idx]);
+		*desc = ring->desc[idx];
 		if (xskq_cons_is_valid_desc(q, desc, umem))
 			return true;
 
-- 
cgit v1.2.3-59-g8ed1b


From 15d8c9162ced1789d25261859a7b37db8426e409 Mon Sep 17 00:00:00 2001
From: Magnus Karlsson <magnus.karlsson@intel.com>
Date: Thu, 19 Dec 2019 13:39:30 +0100
Subject: xsk: Add function naming comments and reorder functions

Add comments on how the ring access functions are named and how they
are supposed to be used for producers and consumers. The functions are
also reordered so that the consumer functions are in the beginning and
the producer functions in the end, for easier reference. Put this in a
separate patch as the diff might look a little odd, but no
functionality has changed in this patch.

Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-12-git-send-email-magnus.karlsson@intel.com
---
 net/xdp/xsk.c       |  10 ++
 net/xdp/xsk_queue.h | 293 ++++++++++++++++++++++++++++------------------------
 2 files changed, 167 insertions(+), 136 deletions(-)

diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c
index 55092feeaf36..5560d60c1ff5 100644
--- a/net/xdp/xsk.c
+++ b/net/xdp/xsk.c
@@ -319,6 +319,11 @@ bool xsk_umem_consume_tx(struct xdp_umem *umem, struct xdp_desc *desc)
 		if (!xskq_cons_peek_desc(xs->tx, desc, umem))
 			continue;
 
+		/* This is the backpreassure mechanism for the Tx path.
+		 * Reserve space in the completion queue and only proceed
+		 * if there is space in it. This avoids having to implement
+		 * any buffering in the Tx path.
+		 */
 		if (xskq_prod_reserve_addr(umem->cq, desc->addr))
 			goto out;
 
@@ -389,6 +394,11 @@ static int xsk_generic_xmit(struct sock *sk)
 		addr = desc.addr;
 		buffer = xdp_umem_get_data(xs->umem, addr);
 		err = skb_store_bits(skb, 0, buffer, len);
+		/* This is the backpreassure mechanism for the Tx path.
+		 * Reserve space in the completion queue and only proceed
+		 * if there is space in it. This avoids having to implement
+		 * any buffering in the Tx path.
+		 */
 		if (unlikely(err) || xskq_prod_reserve(xs->umem->cq)) {
 			kfree_skb(skb);
 			goto out;
diff --git a/net/xdp/xsk_queue.h b/net/xdp/xsk_queue.h
index 4c049cacc693..bec2af11853a 100644
--- a/net/xdp/xsk_queue.h
+++ b/net/xdp/xsk_queue.h
@@ -81,60 +81,27 @@ struct xsk_queue {
  * now and again after circling through the ring.
  */
 
-/* Common functions operating for both RXTX and umem queues */
-
-static inline u64 xskq_nb_invalid_descs(struct xsk_queue *q)
-{
-	return q ? q->invalid_descs : 0;
-}
-
-static inline void __xskq_cons_release(struct xsk_queue *q)
-{
-	smp_mb(); /* D, matches A */
-	WRITE_ONCE(q->ring->consumer, q->cached_cons);
-}
-
-static inline void __xskq_cons_peek(struct xsk_queue *q)
-{
-	/* Refresh the local pointer */
-	q->cached_prod = READ_ONCE(q->ring->producer);
-	smp_rmb(); /* C, matches B */
-}
-
-static inline void xskq_cons_get_entries(struct xsk_queue *q)
-{
-	__xskq_cons_release(q);
-	__xskq_cons_peek(q);
-}
-
-static inline bool xskq_prod_is_full(struct xsk_queue *q)
-{
-	u32 free_entries = q->nentries - (q->cached_prod - q->cached_cons);
-
-	if (free_entries)
-		return false;
-
-	/* Refresh the local tail pointer */
-	q->cached_cons = READ_ONCE(q->ring->consumer);
-	free_entries = q->nentries - (q->cached_prod - q->cached_cons);
-
-	return !free_entries;
-}
-
-static inline bool xskq_cons_has_entries(struct xsk_queue *q, u32 cnt)
-{
-	u32 entries = q->cached_prod - q->cached_cons;
-
-	if (entries >= cnt)
-		return true;
-
-	__xskq_cons_peek(q);
-	entries = q->cached_prod - q->cached_cons;
-
-	return entries >= cnt;
-}
+/* The operations on the rings are the following:
+ *
+ * producer                           consumer
+ *
+ * RESERVE entries                    PEEK in the ring for entries
+ * WRITE data into the ring           READ data from the ring
+ * SUBMIT entries                     RELEASE entries
+ *
+ * The producer reserves one or more entries in the ring. It can then
+ * fill in these entries and finally submit them so that they can be
+ * seen and read by the consumer.
+ *
+ * The consumer peeks into the ring to see if the producer has written
+ * any new entries. If so, the producer can then read these entries
+ * and when it is done reading them release them back to the producer
+ * so that the producer can use these slots to fill in new entries.
+ *
+ * The function names below reflect these operations.
+ */
 
-/* UMEM queue */
+/* Functions that read and validate content from consumer rings. */
 
 static inline bool xskq_cons_crosses_non_contig_pg(struct xdp_umem *umem,
 						   u64 addr,
@@ -148,16 +115,6 @@ static inline bool xskq_cons_crosses_non_contig_pg(struct xdp_umem *umem,
 	return cross_pg && !next_pg_contig;
 }
 
-static inline bool xskq_cons_is_valid_addr(struct xsk_queue *q, u64 addr)
-{
-	if (addr >= q->size) {
-		q->invalid_descs++;
-		return false;
-	}
-
-	return true;
-}
-
 static inline bool xskq_cons_is_valid_unaligned(struct xsk_queue *q,
 						u64 addr,
 						u64 length,
@@ -175,6 +132,16 @@ static inline bool xskq_cons_is_valid_unaligned(struct xsk_queue *q,
 	return true;
 }
 
+static inline bool xskq_cons_is_valid_addr(struct xsk_queue *q, u64 addr)
+{
+	if (addr >= q->size) {
+		q->invalid_descs++;
+		return false;
+	}
+
+	return true;
+}
+
 static inline bool xskq_cons_read_addr(struct xsk_queue *q, u64 *addr,
 				       struct xdp_umem *umem)
 {
@@ -203,75 +170,6 @@ out:
 	return false;
 }
 
-static inline bool xskq_cons_peek_addr(struct xsk_queue *q, u64 *addr,
-				       struct xdp_umem *umem)
-{
-	if (q->cached_prod == q->cached_cons)
-		xskq_cons_get_entries(q);
-	return xskq_cons_read_addr(q, addr, umem);
-}
-
-static inline void xskq_cons_release(struct xsk_queue *q)
-{
-	/* To improve performance, only update local state here.
-	 * Do the actual release operation when we get new entries
-	 * from the ring in xskq_cons_get_entries() instead.
-	 */
-	q->cached_cons++;
-}
-
-static inline int xskq_prod_reserve(struct xsk_queue *q)
-{
-	if (xskq_prod_is_full(q))
-		return -ENOSPC;
-
-	/* A, matches D */
-	q->cached_prod++;
-	return 0;
-}
-
-static inline int xskq_prod_reserve_addr(struct xsk_queue *q, u64 addr)
-{
-	struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring;
-
-	if (xskq_prod_is_full(q))
-		return -ENOSPC;
-
-	/* A, matches D */
-	ring->desc[q->cached_prod++ & q->ring_mask] = addr;
-	return 0;
-}
-
-static inline void __xskq_prod_submit(struct xsk_queue *q, u32 idx)
-{
-	/* Order producer and data */
-	smp_wmb(); /* B, matches C */
-
-	WRITE_ONCE(q->ring->producer, idx);
-}
-
-static inline void xskq_prod_submit(struct xsk_queue *q)
-{
-	__xskq_prod_submit(q, q->cached_prod);
-}
-
-static inline void xskq_prod_submit_addr(struct xsk_queue *q, u64 addr)
-{
-	struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring;
-	u32 idx = q->ring->producer;
-
-	ring->desc[idx++ & q->ring_mask] = addr;
-
-	__xskq_prod_submit(q, idx);
-}
-
-static inline void xskq_prod_submit_n(struct xsk_queue *q, u32 nb_entries)
-{
-	__xskq_prod_submit(q, q->ring->producer + nb_entries);
-}
-
-/* Rx/Tx queue */
-
 static inline bool xskq_cons_is_valid_desc(struct xsk_queue *q,
 					   struct xdp_desc *d,
 					   struct xdp_umem *umem)
@@ -318,6 +216,48 @@ static inline bool xskq_cons_read_desc(struct xsk_queue *q,
 	return false;
 }
 
+/* Functions for consumers */
+
+static inline void __xskq_cons_release(struct xsk_queue *q)
+{
+	smp_mb(); /* D, matches A */
+	WRITE_ONCE(q->ring->consumer, q->cached_cons);
+}
+
+static inline void __xskq_cons_peek(struct xsk_queue *q)
+{
+	/* Refresh the local pointer */
+	q->cached_prod = READ_ONCE(q->ring->producer);
+	smp_rmb(); /* C, matches B */
+}
+
+static inline void xskq_cons_get_entries(struct xsk_queue *q)
+{
+	__xskq_cons_release(q);
+	__xskq_cons_peek(q);
+}
+
+static inline bool xskq_cons_has_entries(struct xsk_queue *q, u32 cnt)
+{
+	u32 entries = q->cached_prod - q->cached_cons;
+
+	if (entries >= cnt)
+		return true;
+
+	__xskq_cons_peek(q);
+	entries = q->cached_prod - q->cached_cons;
+
+	return entries >= cnt;
+}
+
+static inline bool xskq_cons_peek_addr(struct xsk_queue *q, u64 *addr,
+				       struct xdp_umem *umem)
+{
+	if (q->cached_prod == q->cached_cons)
+		xskq_cons_get_entries(q);
+	return xskq_cons_read_addr(q, addr, umem);
+}
+
 static inline bool xskq_cons_peek_desc(struct xsk_queue *q,
 				       struct xdp_desc *desc,
 				       struct xdp_umem *umem)
@@ -327,6 +267,60 @@ static inline bool xskq_cons_peek_desc(struct xsk_queue *q,
 	return xskq_cons_read_desc(q, desc, umem);
 }
 
+static inline void xskq_cons_release(struct xsk_queue *q)
+{
+	/* To improve performance, only update local state here.
+	 * Reflect this to global state when we get new entries
+	 * from the ring in xskq_cons_get_entries().
+	 */
+	q->cached_cons++;
+}
+
+static inline bool xskq_cons_is_full(struct xsk_queue *q)
+{
+	/* No barriers needed since data is not accessed */
+	return READ_ONCE(q->ring->producer) - READ_ONCE(q->ring->consumer) ==
+		q->nentries;
+}
+
+/* Functions for producers */
+
+static inline bool xskq_prod_is_full(struct xsk_queue *q)
+{
+	u32 free_entries = q->nentries - (q->cached_prod - q->cached_cons);
+
+	if (free_entries)
+		return false;
+
+	/* Refresh the local tail pointer */
+	q->cached_cons = READ_ONCE(q->ring->consumer);
+	free_entries = q->nentries - (q->cached_prod - q->cached_cons);
+
+	return !free_entries;
+}
+
+static inline int xskq_prod_reserve(struct xsk_queue *q)
+{
+	if (xskq_prod_is_full(q))
+		return -ENOSPC;
+
+	/* A, matches D */
+	q->cached_prod++;
+	return 0;
+}
+
+static inline int xskq_prod_reserve_addr(struct xsk_queue *q, u64 addr)
+{
+	struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring;
+
+	if (xskq_prod_is_full(q))
+		return -ENOSPC;
+
+	/* A, matches D */
+	ring->desc[q->cached_prod++ & q->ring_mask] = addr;
+	return 0;
+}
+
 static inline int xskq_prod_reserve_desc(struct xsk_queue *q,
 					 u64 addr, u32 len)
 {
@@ -344,11 +338,31 @@ static inline int xskq_prod_reserve_desc(struct xsk_queue *q,
 	return 0;
 }
 
-static inline bool xskq_cons_is_full(struct xsk_queue *q)
+static inline void __xskq_prod_submit(struct xsk_queue *q, u32 idx)
 {
-	/* No barriers needed since data is not accessed */
-	return READ_ONCE(q->ring->producer) - READ_ONCE(q->ring->consumer) ==
-		q->nentries;
+	smp_wmb(); /* B, matches C */
+
+	WRITE_ONCE(q->ring->producer, idx);
+}
+
+static inline void xskq_prod_submit(struct xsk_queue *q)
+{
+	__xskq_prod_submit(q, q->cached_prod);
+}
+
+static inline void xskq_prod_submit_addr(struct xsk_queue *q, u64 addr)
+{
+	struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring;
+	u32 idx = q->ring->producer;
+
+	ring->desc[idx++ & q->ring_mask] = addr;
+
+	__xskq_prod_submit(q, idx);
+}
+
+static inline void xskq_prod_submit_n(struct xsk_queue *q, u32 nb_entries)
+{
+	__xskq_prod_submit(q, q->ring->producer + nb_entries);
 }
 
 static inline bool xskq_prod_is_empty(struct xsk_queue *q)
@@ -357,6 +371,13 @@ static inline bool xskq_prod_is_empty(struct xsk_queue *q)
 	return READ_ONCE(q->ring->consumer) == READ_ONCE(q->ring->producer);
 }
 
+/* For both producers and consumers */
+
+static inline u64 xskq_nb_invalid_descs(struct xsk_queue *q)
+{
+	return q ? q->invalid_descs : 0;
+}
+
 void xskq_set_umem(struct xsk_queue *q, u64 size, u64 chunk_mask);
 struct xsk_queue *xskq_create(u32 nentries, bool umem_queue);
 void xskq_destroy(struct xsk_queue *q_ops);
-- 
cgit v1.2.3-59-g8ed1b


From 1d9cb1f381860b529edec57cf7a08133f40366eb Mon Sep 17 00:00:00 2001
From: Magnus Karlsson <magnus.karlsson@intel.com>
Date: Thu, 19 Dec 2019 13:39:31 +0100
Subject: xsk: Use struct_size() helper

Improve readability and maintainability by using the struct_size()
helper when allocating the AF_XDP rings.

Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-13-git-send-email-magnus.karlsson@intel.com
---
 net/xdp/xsk_queue.c | 15 +++++++--------
 1 file changed, 7 insertions(+), 8 deletions(-)

diff --git a/net/xdp/xsk_queue.c b/net/xdp/xsk_queue.c
index b66504592d9b..c90e9c1e3c63 100644
--- a/net/xdp/xsk_queue.c
+++ b/net/xdp/xsk_queue.c
@@ -18,14 +18,14 @@ void xskq_set_umem(struct xsk_queue *q, u64 size, u64 chunk_mask)
 	q->chunk_mask = chunk_mask;
 }
 
-static u32 xskq_umem_get_ring_size(struct xsk_queue *q)
+static size_t xskq_get_ring_size(struct xsk_queue *q, bool umem_queue)
 {
-	return sizeof(struct xdp_umem_ring) + q->nentries * sizeof(u64);
-}
+	struct xdp_umem_ring *umem_ring;
+	struct xdp_rxtx_ring *rxtx_ring;
 
-static u32 xskq_rxtx_get_ring_size(struct xsk_queue *q)
-{
-	return sizeof(struct xdp_ring) + q->nentries * sizeof(struct xdp_desc);
+	if (umem_queue)
+		return struct_size(umem_ring, desc, q->nentries);
+	return struct_size(rxtx_ring, desc, q->nentries);
 }
 
 struct xsk_queue *xskq_create(u32 nentries, bool umem_queue)
@@ -43,8 +43,7 @@ struct xsk_queue *xskq_create(u32 nentries, bool umem_queue)
 
 	gfp_flags = GFP_KERNEL | __GFP_ZERO | __GFP_NOWARN |
 		    __GFP_COMP  | __GFP_NORETRY;
-	size = umem_queue ? xskq_umem_get_ring_size(q) :
-	       xskq_rxtx_get_ring_size(q);
+	size = xskq_get_ring_size(q, umem_queue);
 
 	q->ring = (struct xdp_ring *)__get_free_pages(gfp_flags,
 						      get_order(size));
-- 
cgit v1.2.3-59-g8ed1b


From 478bee0df0ec9067c12e7d058d78721a7e7a1b29 Mon Sep 17 00:00:00 2001
From: Andrey Ignatov <rdna@fb.com>
Date: Thu, 19 Dec 2019 16:05:11 -0800
Subject: selftests/bpf: Preserve errno in test_progs CHECK macros

It's follow-up for discussion [1]

CHECK and CHECK_FAIL macros in test_progs.h can affect errno in some
circumstances, e.g. if some code accidentally closes stdout. It makes
checking errno in patterns like this unreliable:

	if (CHECK(!bpf_prog_attach_xattr(...), "tag", "msg"))
		goto err;
	CHECK_FAIL(errno != ENOENT);

, since by CHECK_FAIL time errno could be affected not only by
bpf_prog_attach_xattr but by CHECK as well.

Fix it by saving and restoring errno in the macros. There is no "Fixes"
tag since no problems were discovered yet and it's rather precaution.

test_progs was run with this change and no difference was identified.

[1] https://lore.kernel.org/bpf/20191219210907.GD16266@rdna-mbp.dhcp.thefacebook.com/

Signed-off-by: Andrey Ignatov <rdna@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/20191220000511.1684853-1-rdna@fb.com
---
 tools/testing/selftests/bpf/test_progs.h | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/tools/testing/selftests/bpf/test_progs.h b/tools/testing/selftests/bpf/test_progs.h
index 8477df835979..de1fdaa4e7b4 100644
--- a/tools/testing/selftests/bpf/test_progs.h
+++ b/tools/testing/selftests/bpf/test_progs.h
@@ -100,6 +100,7 @@ extern struct ipv6_packet pkt_v6;
 
 #define _CHECK(condition, tag, duration, format...) ({			\
 	int __ret = !!(condition);					\
+	int __save_errno = errno;					\
 	if (__ret) {							\
 		test__fail();						\
 		printf("%s:FAIL:%s ", __func__, tag);			\
@@ -108,15 +109,18 @@ extern struct ipv6_packet pkt_v6;
 		printf("%s:PASS:%s %d nsec\n",				\
 		       __func__, tag, duration);			\
 	}								\
+	errno = __save_errno;						\
 	__ret;								\
 })
 
 #define CHECK_FAIL(condition) ({					\
 	int __ret = !!(condition);					\
+	int __save_errno = errno;					\
 	if (__ret) {							\
 		test__fail();						\
 		printf("%s:FAIL:%d\n", __func__, __LINE__);		\
 	}								\
+	errno = __save_errno;						\
 	__ret;								\
 })
 
-- 
cgit v1.2.3-59-g8ed1b


From d3f11b018f6ce278e683008e9c225fe87afc532d Mon Sep 17 00:00:00 2001
From: Jay Jayatheerthan <jay.jayatheerthan@intel.com>
Date: Fri, 20 Dec 2019 14:25:25 +0530
Subject: samples/bpf: xdpsock: Add duration option to specify how long to run

The application now supports '-d' or '--duration' option to specify number of
seconds to run. This is used in tx, rx and l2fwd features. If this option is
not provided, the application runs forever.

Signed-off-by: Jay Jayatheerthan <jay.jayatheerthan@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191220085530.4980-2-jay.jayatheerthan@intel.com
---
 samples/bpf/xdpsock_user.c | 37 +++++++++++++++++++++++++++++++++++--
 1 file changed, 35 insertions(+), 2 deletions(-)

diff --git a/samples/bpf/xdpsock_user.c b/samples/bpf/xdpsock_user.c
index e7829e5baaff..e188a79a9c31 100644
--- a/samples/bpf/xdpsock_user.c
+++ b/samples/bpf/xdpsock_user.c
@@ -65,6 +65,9 @@ static u32 opt_xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
 static const char *opt_if = "";
 static int opt_ifindex;
 static int opt_queue;
+static unsigned long opt_duration;
+static unsigned long start_time;
+static bool benchmark_done;
 static int opt_poll;
 static int opt_interval = 1;
 static u32 opt_xdp_bind_flags = XDP_USE_NEED_WAKEUP;
@@ -167,10 +170,21 @@ static void dump_stats(void)
 	}
 }
 
+static bool is_benchmark_done(void)
+{
+	if (opt_duration > 0) {
+		unsigned long dt = (get_nsecs() - start_time);
+
+		if (dt >= opt_duration)
+			benchmark_done = true;
+	}
+	return benchmark_done;
+}
+
 static void *poller(void *arg)
 {
 	(void)arg;
-	for (;;) {
+	while (!is_benchmark_done()) {
 		sleep(opt_interval);
 		dump_stats();
 	}
@@ -375,6 +389,7 @@ static struct option long_options[] = {
 	{"unaligned", no_argument, 0, 'u'},
 	{"shared-umem", no_argument, 0, 'M'},
 	{"force", no_argument, 0, 'F'},
+	{"duration", required_argument, 0, 'd'},
 	{0, 0, 0, 0}
 };
 
@@ -399,6 +414,8 @@ static void usage(const char *prog)
 		"  -u, --unaligned	Enable unaligned chunk placement\n"
 		"  -M, --shared-umem	Enable XDP_SHARED_UMEM\n"
 		"  -F, --force		Force loading the XDP prog\n"
+		"  -d, --duration=n	Duration in secs to run command.\n"
+		"			Default: forever.\n"
 		"\n";
 	fprintf(stderr, str, prog, XSK_UMEM__DEFAULT_FRAME_SIZE);
 	exit(EXIT_FAILURE);
@@ -411,7 +428,7 @@ static void parse_command_line(int argc, char **argv)
 	opterr = 0;
 
 	for (;;) {
-		c = getopt_long(argc, argv, "Frtli:q:psSNn:czf:muM",
+		c = getopt_long(argc, argv, "Frtli:q:psSNn:czf:muMd:",
 				long_options, &option_index);
 		if (c == -1)
 			break;
@@ -469,6 +486,10 @@ static void parse_command_line(int argc, char **argv)
 		case 'M':
 			opt_num_xsks = MAX_SOCKS;
 			break;
+		case 'd':
+			opt_duration = atoi(optarg);
+			opt_duration *= 1000000000;
+			break;
 		default:
 			usage(basename(argv[0]));
 		}
@@ -622,6 +643,9 @@ static void rx_drop_all(void)
 
 		for (i = 0; i < num_socks; i++)
 			rx_drop(xsks[i], fds);
+
+		if (benchmark_done)
+			break;
 	}
 }
 
@@ -671,6 +695,9 @@ static void tx_only_all(void)
 
 		for (i = 0; i < num_socks; i++)
 			tx_only(xsks[i], frame_nb[i]);
+
+		if (benchmark_done)
+			break;
 	}
 }
 
@@ -739,6 +766,9 @@ static void l2fwd_all(void)
 
 		for (i = 0; i < num_socks; i++)
 			l2fwd(xsks[i], fds);
+
+		if (benchmark_done)
+			break;
 	}
 }
 
@@ -852,6 +882,7 @@ int main(int argc, char **argv)
 		exit_with_error(ret);
 
 	prev_time = get_nsecs();
+	start_time = prev_time;
 
 	if (opt_bench == BENCH_RXDROP)
 		rx_drop_all();
@@ -860,5 +891,7 @@ int main(int argc, char **argv)
 	else
 		l2fwd_all();
 
+	pthread_join(pt, NULL);
+
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 695255882bdf63da240db33d0f2aa9ccca1cbe67 Mon Sep 17 00:00:00 2001
From: Jay Jayatheerthan <jay.jayatheerthan@intel.com>
Date: Fri, 20 Dec 2019 14:25:26 +0530
Subject: samples/bpf: xdpsock: Use common code to handle signal and main exit

Add code to do cleanup for signals and application completion in a unified
fashion. The signal handler sets benckmark_done flag terminating the
threads. The cleanup is called before returning from main() function.

Signed-off-by: Jay Jayatheerthan <jay.jayatheerthan@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191220085530.4980-3-jay.jayatheerthan@intel.com
---
 samples/bpf/xdpsock_user.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/samples/bpf/xdpsock_user.c b/samples/bpf/xdpsock_user.c
index e188a79a9c31..7febc3d519a1 100644
--- a/samples/bpf/xdpsock_user.c
+++ b/samples/bpf/xdpsock_user.c
@@ -209,6 +209,11 @@ static void remove_xdp_program(void)
 }
 
 static void int_exit(int sig)
+{
+	benchmark_done = true;
+}
+
+static void xdpsock_cleanup(void)
 {
 	struct xsk_umem *umem = xsks[0]->umem->umem;
 	int i;
@@ -218,8 +223,6 @@ static void int_exit(int sig)
 		xsk_socket__delete(xsks[i]->xsk);
 	(void)xsk_umem__delete(umem);
 	remove_xdp_program();
-
-	exit(EXIT_SUCCESS);
 }
 
 static void __exit_with_error(int error, const char *file, const char *func,
@@ -893,5 +896,7 @@ int main(int argc, char **argv)
 
 	pthread_join(pt, NULL);
 
+	xdpsock_cleanup();
+
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From cd9e72b6f21044b36a096833003811c2b2038455 Mon Sep 17 00:00:00 2001
From: Jay Jayatheerthan <jay.jayatheerthan@intel.com>
Date: Fri, 20 Dec 2019 14:25:27 +0530
Subject: samples/bpf: xdpsock: Add option to specify batch size

New option to specify batch size for tx, rx and l2fwd has been added. This
allows fine tuning to maximize performance. It is specified using '-b' or
'--batch_size' options. When not specified default is 64.

Signed-off-by: Jay Jayatheerthan <jay.jayatheerthan@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191220085530.4980-4-jay.jayatheerthan@intel.com
---
 samples/bpf/xdpsock_user.c | 52 ++++++++++++++++++++++++++--------------------
 1 file changed, 29 insertions(+), 23 deletions(-)

diff --git a/samples/bpf/xdpsock_user.c b/samples/bpf/xdpsock_user.c
index 7febc3d519a1..1ba3e7142f39 100644
--- a/samples/bpf/xdpsock_user.c
+++ b/samples/bpf/xdpsock_user.c
@@ -45,7 +45,6 @@
 #endif
 
 #define NUM_FRAMES (4 * 1024)
-#define BATCH_SIZE 64
 
 #define DEBUG_HEXDUMP 0
 
@@ -68,6 +67,7 @@ static int opt_queue;
 static unsigned long opt_duration;
 static unsigned long start_time;
 static bool benchmark_done;
+static u32 opt_batch_size = 64;
 static int opt_poll;
 static int opt_interval = 1;
 static u32 opt_xdp_bind_flags = XDP_USE_NEED_WAKEUP;
@@ -237,7 +237,6 @@ static void __exit_with_error(int error, const char *file, const char *func,
 
 #define exit_with_error(error) __exit_with_error(error, __FILE__, __func__, \
 						 __LINE__)
-
 static const char pkt_data[] =
 	"\x3c\xfd\xfe\x9e\x7f\x71\xec\xb1\xd7\x98\x3a\xc0\x08\x00\x45\x00"
 	"\x00\x2e\x00\x00\x00\x00\x40\x11\x88\x97\x05\x08\x07\x08\xc8\x14"
@@ -291,11 +290,10 @@ static void hex_dump(void *pkt, size_t length, u64 addr)
 	printf("\n");
 }
 
-static size_t gen_eth_frame(struct xsk_umem_info *umem, u64 addr)
+static void gen_eth_frame(struct xsk_umem_info *umem, u64 addr)
 {
 	memcpy(xsk_umem__get_data(umem->buffer, addr), pkt_data,
 	       sizeof(pkt_data) - 1);
-	return sizeof(pkt_data) - 1;
 }
 
 static struct xsk_umem_info *xsk_configure_umem(void *buffer, u64 size)
@@ -393,6 +391,7 @@ static struct option long_options[] = {
 	{"shared-umem", no_argument, 0, 'M'},
 	{"force", no_argument, 0, 'F'},
 	{"duration", required_argument, 0, 'd'},
+	{"batch-size", required_argument, 0, 'b'},
 	{0, 0, 0, 0}
 };
 
@@ -419,8 +418,11 @@ static void usage(const char *prog)
 		"  -F, --force		Force loading the XDP prog\n"
 		"  -d, --duration=n	Duration in secs to run command.\n"
 		"			Default: forever.\n"
+		"  -b, --batch-size=n	Batch size for sending or receiving\n"
+		"			packets. Default: %d\n"
 		"\n";
-	fprintf(stderr, str, prog, XSK_UMEM__DEFAULT_FRAME_SIZE);
+	fprintf(stderr, str, prog, XSK_UMEM__DEFAULT_FRAME_SIZE,
+		opt_batch_size);
 	exit(EXIT_FAILURE);
 }
 
@@ -431,7 +433,7 @@ static void parse_command_line(int argc, char **argv)
 	opterr = 0;
 
 	for (;;) {
-		c = getopt_long(argc, argv, "Frtli:q:psSNn:czf:muMd:",
+		c = getopt_long(argc, argv, "Frtli:q:pSNn:czf:muMd:b:",
 				long_options, &option_index);
 		if (c == -1)
 			break;
@@ -493,6 +495,9 @@ static void parse_command_line(int argc, char **argv)
 			opt_duration = atoi(optarg);
 			opt_duration *= 1000000000;
 			break;
+		case 'b':
+			opt_batch_size = atoi(optarg);
+			break;
 		default:
 			usage(basename(argv[0]));
 		}
@@ -540,7 +545,7 @@ static inline void complete_tx_l2fwd(struct xsk_socket_info *xsk,
 	if (!opt_need_wakeup || xsk_ring_prod__needs_wakeup(&xsk->tx))
 		kick_tx(xsk);
 
-	ndescs = (xsk->outstanding_tx > BATCH_SIZE) ? BATCH_SIZE :
+	ndescs = (xsk->outstanding_tx > opt_batch_size) ? opt_batch_size :
 		xsk->outstanding_tx;
 
 	/* re-add completed Tx buffers */
@@ -580,7 +585,7 @@ static inline void complete_tx_only(struct xsk_socket_info *xsk)
 	if (!opt_need_wakeup || xsk_ring_prod__needs_wakeup(&xsk->tx))
 		kick_tx(xsk);
 
-	rcvd = xsk_ring_cons__peek(&xsk->umem->cq, BATCH_SIZE, &idx);
+	rcvd = xsk_ring_cons__peek(&xsk->umem->cq, opt_batch_size, &idx);
 	if (rcvd > 0) {
 		xsk_ring_cons__release(&xsk->umem->cq, rcvd);
 		xsk->outstanding_tx -= rcvd;
@@ -594,7 +599,7 @@ static void rx_drop(struct xsk_socket_info *xsk, struct pollfd *fds)
 	u32 idx_rx = 0, idx_fq = 0;
 	int ret;
 
-	rcvd = xsk_ring_cons__peek(&xsk->rx, BATCH_SIZE, &idx_rx);
+	rcvd = xsk_ring_cons__peek(&xsk->rx, opt_batch_size, &idx_rx);
 	if (!rcvd) {
 		if (xsk_ring_prod__needs_wakeup(&xsk->umem->fq))
 			ret = poll(fds, num_socks, opt_timeout);
@@ -655,23 +660,24 @@ static void rx_drop_all(void)
 static void tx_only(struct xsk_socket_info *xsk, u32 frame_nb)
 {
 	u32 idx;
+	unsigned int i;
 
-	if (xsk_ring_prod__reserve(&xsk->tx, BATCH_SIZE, &idx) == BATCH_SIZE) {
-		unsigned int i;
-
-		for (i = 0; i < BATCH_SIZE; i++) {
-			xsk_ring_prod__tx_desc(&xsk->tx, idx + i)->addr	=
-				(frame_nb + i) << XSK_UMEM__DEFAULT_FRAME_SHIFT;
-			xsk_ring_prod__tx_desc(&xsk->tx, idx + i)->len =
-				sizeof(pkt_data) - 1;
-		}
+	while (xsk_ring_prod__reserve(&xsk->tx, opt_batch_size, &idx) <
+				      opt_batch_size) {
+		complete_tx_only(xsk);
+	}
 
-		xsk_ring_prod__submit(&xsk->tx, BATCH_SIZE);
-		xsk->outstanding_tx += BATCH_SIZE;
-		frame_nb += BATCH_SIZE;
-		frame_nb %= NUM_FRAMES;
+	for (i = 0; i < opt_batch_size; i++) {
+		struct xdp_desc *tx_desc = xsk_ring_prod__tx_desc(&xsk->tx,
+								  idx + i);
+		tx_desc->addr = (frame_nb + i) << XSK_UMEM__DEFAULT_FRAME_SHIFT;
+		tx_desc->len = sizeof(pkt_data) - 1;
 	}
 
+	xsk_ring_prod__submit(&xsk->tx, opt_batch_size);
+	xsk->outstanding_tx += opt_batch_size;
+	frame_nb += opt_batch_size;
+	frame_nb %= NUM_FRAMES;
 	complete_tx_only(xsk);
 }
 
@@ -712,7 +718,7 @@ static void l2fwd(struct xsk_socket_info *xsk, struct pollfd *fds)
 
 	complete_tx_l2fwd(xsk, fds);
 
-	rcvd = xsk_ring_cons__peek(&xsk->rx, BATCH_SIZE, &idx_rx);
+	rcvd = xsk_ring_cons__peek(&xsk->rx, opt_batch_size, &idx_rx);
 	if (!rcvd) {
 		if (xsk_ring_prod__needs_wakeup(&xsk->umem->fq))
 			ret = poll(fds, num_socks, opt_timeout);
-- 
cgit v1.2.3-59-g8ed1b


From ece6e9694751a4f0b99372724a0705a0217132b3 Mon Sep 17 00:00:00 2001
From: Jay Jayatheerthan <jay.jayatheerthan@intel.com>
Date: Fri, 20 Dec 2019 14:25:28 +0530
Subject: samples/bpf: xdpsock: Add option to specify number of packets to send

Use '-C' or '--tx-pkt-count' to specify number of packets to send.
If it is not specified, the application sends packets forever. If packet
count is not a multiple of batch size, last batch sent is less than the
batch size.

Signed-off-by: Jay Jayatheerthan <jay.jayatheerthan@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191220085530.4980-5-jay.jayatheerthan@intel.com
---
 samples/bpf/xdpsock_user.c | 73 +++++++++++++++++++++++++++++++++++++---------
 1 file changed, 59 insertions(+), 14 deletions(-)

diff --git a/samples/bpf/xdpsock_user.c b/samples/bpf/xdpsock_user.c
index 1ba3e7142f39..f96ce3055d46 100644
--- a/samples/bpf/xdpsock_user.c
+++ b/samples/bpf/xdpsock_user.c
@@ -68,6 +68,7 @@ static unsigned long opt_duration;
 static unsigned long start_time;
 static bool benchmark_done;
 static u32 opt_batch_size = 64;
+static int opt_pkt_count;
 static int opt_poll;
 static int opt_interval = 1;
 static u32 opt_xdp_bind_flags = XDP_USE_NEED_WAKEUP;
@@ -392,6 +393,7 @@ static struct option long_options[] = {
 	{"force", no_argument, 0, 'F'},
 	{"duration", required_argument, 0, 'd'},
 	{"batch-size", required_argument, 0, 'b'},
+	{"tx-pkt-count", required_argument, 0, 'C'},
 	{0, 0, 0, 0}
 };
 
@@ -420,6 +422,8 @@ static void usage(const char *prog)
 		"			Default: forever.\n"
 		"  -b, --batch-size=n	Batch size for sending or receiving\n"
 		"			packets. Default: %d\n"
+		"  -C, --tx-pkt-count=n	Number of packets to send.\n"
+		"			Default: Continuous packets.\n"
 		"\n";
 	fprintf(stderr, str, prog, XSK_UMEM__DEFAULT_FRAME_SIZE,
 		opt_batch_size);
@@ -433,7 +437,7 @@ static void parse_command_line(int argc, char **argv)
 	opterr = 0;
 
 	for (;;) {
-		c = getopt_long(argc, argv, "Frtli:q:pSNn:czf:muMd:b:",
+		c = getopt_long(argc, argv, "Frtli:q:pSNn:czf:muMd:b:C:",
 				long_options, &option_index);
 		if (c == -1)
 			break;
@@ -498,6 +502,9 @@ static void parse_command_line(int argc, char **argv)
 		case 'b':
 			opt_batch_size = atoi(optarg);
 			break;
+		case 'C':
+			opt_pkt_count = atoi(optarg);
+			break;
 		default:
 			usage(basename(argv[0]));
 		}
@@ -574,7 +581,8 @@ static inline void complete_tx_l2fwd(struct xsk_socket_info *xsk,
 	}
 }
 
-static inline void complete_tx_only(struct xsk_socket_info *xsk)
+static inline void complete_tx_only(struct xsk_socket_info *xsk,
+				    int batch_size)
 {
 	unsigned int rcvd;
 	u32 idx;
@@ -585,7 +593,7 @@ static inline void complete_tx_only(struct xsk_socket_info *xsk)
 	if (!opt_need_wakeup || xsk_ring_prod__needs_wakeup(&xsk->tx))
 		kick_tx(xsk);
 
-	rcvd = xsk_ring_cons__peek(&xsk->umem->cq, opt_batch_size, &idx);
+	rcvd = xsk_ring_cons__peek(&xsk->umem->cq, batch_size, &idx);
 	if (rcvd > 0) {
 		xsk_ring_cons__release(&xsk->umem->cq, rcvd);
 		xsk->outstanding_tx -= rcvd;
@@ -657,34 +665,62 @@ static void rx_drop_all(void)
 	}
 }
 
-static void tx_only(struct xsk_socket_info *xsk, u32 frame_nb)
+static void tx_only(struct xsk_socket_info *xsk, u32 frame_nb, int batch_size)
 {
 	u32 idx;
 	unsigned int i;
 
-	while (xsk_ring_prod__reserve(&xsk->tx, opt_batch_size, &idx) <
-				      opt_batch_size) {
-		complete_tx_only(xsk);
+	while (xsk_ring_prod__reserve(&xsk->tx, batch_size, &idx) <
+				      batch_size) {
+		complete_tx_only(xsk, batch_size);
 	}
 
-	for (i = 0; i < opt_batch_size; i++) {
+	for (i = 0; i < batch_size; i++) {
 		struct xdp_desc *tx_desc = xsk_ring_prod__tx_desc(&xsk->tx,
 								  idx + i);
 		tx_desc->addr = (frame_nb + i) << XSK_UMEM__DEFAULT_FRAME_SHIFT;
 		tx_desc->len = sizeof(pkt_data) - 1;
 	}
 
-	xsk_ring_prod__submit(&xsk->tx, opt_batch_size);
-	xsk->outstanding_tx += opt_batch_size;
-	frame_nb += opt_batch_size;
+	xsk_ring_prod__submit(&xsk->tx, batch_size);
+	xsk->outstanding_tx += batch_size;
+	frame_nb += batch_size;
 	frame_nb %= NUM_FRAMES;
-	complete_tx_only(xsk);
+	complete_tx_only(xsk, batch_size);
+}
+
+static inline int get_batch_size(int pkt_cnt)
+{
+	if (!opt_pkt_count)
+		return opt_batch_size;
+
+	if (pkt_cnt + opt_batch_size <= opt_pkt_count)
+		return opt_batch_size;
+
+	return opt_pkt_count - pkt_cnt;
+}
+
+static void complete_tx_only_all(void)
+{
+	bool pending;
+	int i;
+
+	do {
+		pending = false;
+		for (i = 0; i < num_socks; i++) {
+			if (xsks[i]->outstanding_tx) {
+				complete_tx_only(xsks[i], opt_batch_size);
+				pending = !!xsks[i]->outstanding_tx;
+			}
+		}
+	} while (pending);
 }
 
 static void tx_only_all(void)
 {
 	struct pollfd fds[MAX_SOCKS] = {};
 	u32 frame_nb[MAX_SOCKS] = {};
+	int pkt_cnt = 0;
 	int i, ret;
 
 	for (i = 0; i < num_socks; i++) {
@@ -692,7 +728,9 @@ static void tx_only_all(void)
 		fds[0].events = POLLOUT;
 	}
 
-	for (;;) {
+	while ((opt_pkt_count && pkt_cnt < opt_pkt_count) || !opt_pkt_count) {
+		int batch_size = get_batch_size(pkt_cnt);
+
 		if (opt_poll) {
 			ret = poll(fds, num_socks, opt_timeout);
 			if (ret <= 0)
@@ -703,11 +741,16 @@ static void tx_only_all(void)
 		}
 
 		for (i = 0; i < num_socks; i++)
-			tx_only(xsks[i], frame_nb[i]);
+			tx_only(xsks[i], frame_nb[i], batch_size);
+
+		pkt_cnt += batch_size;
 
 		if (benchmark_done)
 			break;
 	}
+
+	if (opt_pkt_count)
+		complete_tx_only_all();
 }
 
 static void l2fwd(struct xsk_socket_info *xsk, struct pollfd *fds)
@@ -900,6 +943,8 @@ int main(int argc, char **argv)
 	else
 		l2fwd_all();
 
+	benchmark_done = true;
+
 	pthread_join(pt, NULL);
 
 	xdpsock_cleanup();
-- 
cgit v1.2.3-59-g8ed1b


From 4a3c23ae3acc6185a9d418830f22672a52ff986a Mon Sep 17 00:00:00 2001
From: Jay Jayatheerthan <jay.jayatheerthan@intel.com>
Date: Fri, 20 Dec 2019 14:25:29 +0530
Subject: samples/bpf: xdpsock: Add option to specify tx packet size

New option '-s' or '--tx-pkt-size' has been added to specify the transmit
packet size. The packet size ranges from 47 to 4096 bytes. When this
option is not provided, it defaults to 64 byte packet.

The code uses struct ethhdr, struct iphdr and struct udphdr to form the
transmit packet. The MAC address, IP address and UDP ports are set to default
values.

The code calculates IP and UDP checksums before sending the packet.
Checksum calculation code in Linux kernel is used for this purpose.
The Ethernet FCS is not filled by the code.

Signed-off-by: Jay Jayatheerthan <jay.jayatheerthan@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191220085530.4980-6-jay.jayatheerthan@intel.com
---
 samples/bpf/xdpsock_user.c | 276 +++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 265 insertions(+), 11 deletions(-)

diff --git a/samples/bpf/xdpsock_user.c b/samples/bpf/xdpsock_user.c
index f96ce3055d46..2297158a32bd 100644
--- a/samples/bpf/xdpsock_user.c
+++ b/samples/bpf/xdpsock_user.c
@@ -10,6 +10,9 @@
 #include <linux/if_link.h>
 #include <linux/if_xdp.h>
 #include <linux/if_ether.h>
+#include <linux/ip.h>
+#include <linux/udp.h>
+#include <arpa/inet.h>
 #include <locale.h>
 #include <net/ethernet.h>
 #include <net/if.h>
@@ -45,11 +48,14 @@
 #endif
 
 #define NUM_FRAMES (4 * 1024)
+#define MIN_PKT_SIZE 64
 
 #define DEBUG_HEXDUMP 0
 
 typedef __u64 u64;
 typedef __u32 u32;
+typedef __u16 u16;
+typedef __u8  u8;
 
 static unsigned long prev_time;
 
@@ -69,6 +75,8 @@ static unsigned long start_time;
 static bool benchmark_done;
 static u32 opt_batch_size = 64;
 static int opt_pkt_count;
+static u16 opt_pkt_size = MIN_PKT_SIZE;
+static u32 pkt_fill_pattern = 0x12345678;
 static int opt_poll;
 static int opt_interval = 1;
 static u32 opt_xdp_bind_flags = XDP_USE_NEED_WAKEUP;
@@ -238,12 +246,6 @@ static void __exit_with_error(int error, const char *file, const char *func,
 
 #define exit_with_error(error) __exit_with_error(error, __FILE__, __func__, \
 						 __LINE__)
-static const char pkt_data[] =
-	"\x3c\xfd\xfe\x9e\x7f\x71\xec\xb1\xd7\x98\x3a\xc0\x08\x00\x45\x00"
-	"\x00\x2e\x00\x00\x00\x00\x40\x11\x88\x97\x05\x08\x07\x08\xc8\x14"
-	"\x1e\x04\x10\x92\x10\x92\x00\x1a\x6d\xa3\x34\x33\x1f\x69\x40\x6b"
-	"\x54\x59\xb6\x14\x2d\x11\x44\xbf\xaf\xd9\xbe\xaa";
-
 static void swap_mac_addresses(void *data)
 {
 	struct ether_header *eth = (struct ether_header *)data;
@@ -291,10 +293,243 @@ static void hex_dump(void *pkt, size_t length, u64 addr)
 	printf("\n");
 }
 
+static void *memset32_htonl(void *dest, u32 val, u32 size)
+{
+	u32 *ptr = (u32 *)dest;
+	int i;
+
+	val = htonl(val);
+
+	for (i = 0; i < (size & (~0x3)); i += 4)
+		ptr[i >> 2] = val;
+
+	for (; i < size; i++)
+		((char *)dest)[i] = ((char *)&val)[i & 3];
+
+	return dest;
+}
+
+/*
+ * This function code has been taken from
+ * Linux kernel lib/checksum.c
+ */
+static inline unsigned short from32to16(unsigned int x)
+{
+	/* add up 16-bit and 16-bit for 16+c bit */
+	x = (x & 0xffff) + (x >> 16);
+	/* add up carry.. */
+	x = (x & 0xffff) + (x >> 16);
+	return x;
+}
+
+/*
+ * This function code has been taken from
+ * Linux kernel lib/checksum.c
+ */
+static unsigned int do_csum(const unsigned char *buff, int len)
+{
+	unsigned int result = 0;
+	int odd;
+
+	if (len <= 0)
+		goto out;
+	odd = 1 & (unsigned long)buff;
+	if (odd) {
+#ifdef __LITTLE_ENDIAN
+		result += (*buff << 8);
+#else
+		result = *buff;
+#endif
+		len--;
+		buff++;
+	}
+	if (len >= 2) {
+		if (2 & (unsigned long)buff) {
+			result += *(unsigned short *)buff;
+			len -= 2;
+			buff += 2;
+		}
+		if (len >= 4) {
+			const unsigned char *end = buff +
+						   ((unsigned int)len & ~3);
+			unsigned int carry = 0;
+
+			do {
+				unsigned int w = *(unsigned int *)buff;
+
+				buff += 4;
+				result += carry;
+				result += w;
+				carry = (w > result);
+			} while (buff < end);
+			result += carry;
+			result = (result & 0xffff) + (result >> 16);
+		}
+		if (len & 2) {
+			result += *(unsigned short *)buff;
+			buff += 2;
+		}
+	}
+	if (len & 1)
+#ifdef __LITTLE_ENDIAN
+		result += *buff;
+#else
+		result += (*buff << 8);
+#endif
+	result = from32to16(result);
+	if (odd)
+		result = ((result >> 8) & 0xff) | ((result & 0xff) << 8);
+out:
+	return result;
+}
+
+__sum16 ip_fast_csum(const void *iph, unsigned int ihl);
+
+/*
+ *	This is a version of ip_compute_csum() optimized for IP headers,
+ *	which always checksum on 4 octet boundaries.
+ *	This function code has been taken from
+ *	Linux kernel lib/checksum.c
+ */
+__sum16 ip_fast_csum(const void *iph, unsigned int ihl)
+{
+	return (__force __sum16)~do_csum(iph, ihl * 4);
+}
+
+/*
+ * Fold a partial checksum
+ * This function code has been taken from
+ * Linux kernel include/asm-generic/checksum.h
+ */
+static inline __sum16 csum_fold(__wsum csum)
+{
+	u32 sum = (__force u32)csum;
+
+	sum = (sum & 0xffff) + (sum >> 16);
+	sum = (sum & 0xffff) + (sum >> 16);
+	return (__force __sum16)~sum;
+}
+
+/*
+ * This function code has been taken from
+ * Linux kernel lib/checksum.c
+ */
+static inline u32 from64to32(u64 x)
+{
+	/* add up 32-bit and 32-bit for 32+c bit */
+	x = (x & 0xffffffff) + (x >> 32);
+	/* add up carry.. */
+	x = (x & 0xffffffff) + (x >> 32);
+	return (u32)x;
+}
+
+__wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
+			  __u32 len, __u8 proto, __wsum sum);
+
+/*
+ * This function code has been taken from
+ * Linux kernel lib/checksum.c
+ */
+__wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
+			  __u32 len, __u8 proto, __wsum sum)
+{
+	unsigned long long s = (__force u32)sum;
+
+	s += (__force u32)saddr;
+	s += (__force u32)daddr;
+#ifdef __BIG_ENDIAN__
+	s += proto + len;
+#else
+	s += (proto + len) << 8;
+#endif
+	return (__force __wsum)from64to32(s);
+}
+
+/*
+ * This function has been taken from
+ * Linux kernel include/asm-generic/checksum.h
+ */
+static inline __sum16
+csum_tcpudp_magic(__be32 saddr, __be32 daddr, __u32 len,
+		  __u8 proto, __wsum sum)
+{
+	return csum_fold(csum_tcpudp_nofold(saddr, daddr, len, proto, sum));
+}
+
+static inline u16 udp_csum(u32 saddr, u32 daddr, u32 len,
+			   u8 proto, u16 *udp_pkt)
+{
+	u32 csum = 0;
+	u32 cnt = 0;
+
+	/* udp hdr and data */
+	for (; cnt < len; cnt += 2)
+		csum += udp_pkt[cnt >> 1];
+
+	return csum_tcpudp_magic(saddr, daddr, len, proto, csum);
+}
+
+#define ETH_FCS_SIZE 4
+
+#define PKT_HDR_SIZE (sizeof(struct ethhdr) + sizeof(struct iphdr) + \
+		      sizeof(struct udphdr))
+
+#define PKT_SIZE		(opt_pkt_size - ETH_FCS_SIZE)
+#define IP_PKT_SIZE		(PKT_SIZE - sizeof(struct ethhdr))
+#define UDP_PKT_SIZE		(IP_PKT_SIZE - sizeof(struct iphdr))
+#define UDP_PKT_DATA_SIZE	(UDP_PKT_SIZE - sizeof(struct udphdr))
+
+static u8 pkt_data[XSK_UMEM__DEFAULT_FRAME_SIZE];
+
+static void gen_eth_hdr_data(void)
+{
+	struct udphdr *udp_hdr = (struct udphdr *)(pkt_data +
+						   sizeof(struct ethhdr) +
+						   sizeof(struct iphdr));
+	struct iphdr *ip_hdr = (struct iphdr *)(pkt_data +
+						sizeof(struct ethhdr));
+	struct ethhdr *eth_hdr = (struct ethhdr *)pkt_data;
+
+	/* ethernet header */
+	memcpy(eth_hdr->h_dest, "\x3c\xfd\xfe\x9e\x7f\x71", ETH_ALEN);
+	memcpy(eth_hdr->h_source, "\xec\xb1\xd7\x98\x3a\xc0", ETH_ALEN);
+	eth_hdr->h_proto = htons(ETH_P_IP);
+
+	/* IP header */
+	ip_hdr->version = IPVERSION;
+	ip_hdr->ihl = 0x5; /* 20 byte header */
+	ip_hdr->tos = 0x0;
+	ip_hdr->tot_len = htons(IP_PKT_SIZE);
+	ip_hdr->id = 0;
+	ip_hdr->frag_off = 0;
+	ip_hdr->ttl = IPDEFTTL;
+	ip_hdr->protocol = IPPROTO_UDP;
+	ip_hdr->saddr = htonl(0x0a0a0a10);
+	ip_hdr->daddr = htonl(0x0a0a0a20);
+
+	/* IP header checksum */
+	ip_hdr->check = 0;
+	ip_hdr->check = ip_fast_csum((const void *)ip_hdr, ip_hdr->ihl);
+
+	/* UDP header */
+	udp_hdr->source = htons(0x1000);
+	udp_hdr->dest = htons(0x1000);
+	udp_hdr->len = htons(UDP_PKT_SIZE);
+
+	/* UDP data */
+	memset32_htonl(pkt_data + PKT_HDR_SIZE, pkt_fill_pattern,
+		       UDP_PKT_DATA_SIZE);
+
+	/* UDP header checksum */
+	udp_hdr->check = 0;
+	udp_hdr->check = udp_csum(ip_hdr->saddr, ip_hdr->daddr, UDP_PKT_SIZE,
+				  IPPROTO_UDP, (u16 *)udp_hdr);
+}
+
 static void gen_eth_frame(struct xsk_umem_info *umem, u64 addr)
 {
 	memcpy(xsk_umem__get_data(umem->buffer, addr), pkt_data,
-	       sizeof(pkt_data) - 1);
+	       PKT_SIZE);
 }
 
 static struct xsk_umem_info *xsk_configure_umem(void *buffer, u64 size)
@@ -394,6 +629,7 @@ static struct option long_options[] = {
 	{"duration", required_argument, 0, 'd'},
 	{"batch-size", required_argument, 0, 'b'},
 	{"tx-pkt-count", required_argument, 0, 'C'},
+	{"tx-pkt-size", required_argument, 0, 's'},
 	{0, 0, 0, 0}
 };
 
@@ -424,9 +660,14 @@ static void usage(const char *prog)
 		"			packets. Default: %d\n"
 		"  -C, --tx-pkt-count=n	Number of packets to send.\n"
 		"			Default: Continuous packets.\n"
+		"  -s, --tx-pkt-size=n	Transmit packet size.\n"
+		"			(Default: %d bytes)\n"
+		"			Min size: %d, Max size %d.\n"
 		"\n";
 	fprintf(stderr, str, prog, XSK_UMEM__DEFAULT_FRAME_SIZE,
-		opt_batch_size);
+		opt_batch_size, MIN_PKT_SIZE, MIN_PKT_SIZE,
+		XSK_UMEM__DEFAULT_FRAME_SIZE);
+
 	exit(EXIT_FAILURE);
 }
 
@@ -437,7 +678,7 @@ static void parse_command_line(int argc, char **argv)
 	opterr = 0;
 
 	for (;;) {
-		c = getopt_long(argc, argv, "Frtli:q:pSNn:czf:muMd:b:C:",
+		c = getopt_long(argc, argv, "Frtli:q:pSNn:czf:muMd:b:C:s:",
 				long_options, &option_index);
 		if (c == -1)
 			break;
@@ -505,6 +746,16 @@ static void parse_command_line(int argc, char **argv)
 		case 'C':
 			opt_pkt_count = atoi(optarg);
 			break;
+		case 's':
+			opt_pkt_size = atoi(optarg);
+			if (opt_pkt_size > (XSK_UMEM__DEFAULT_FRAME_SIZE) ||
+			    opt_pkt_size < MIN_PKT_SIZE) {
+				fprintf(stderr,
+					"ERROR: Invalid frame size %d\n",
+					opt_pkt_size);
+				usage(basename(argv[0]));
+			}
+			break;
 		default:
 			usage(basename(argv[0]));
 		}
@@ -679,7 +930,7 @@ static void tx_only(struct xsk_socket_info *xsk, u32 frame_nb, int batch_size)
 		struct xdp_desc *tx_desc = xsk_ring_prod__tx_desc(&xsk->tx,
 								  idx + i);
 		tx_desc->addr = (frame_nb + i) << XSK_UMEM__DEFAULT_FRAME_SHIFT;
-		tx_desc->len = sizeof(pkt_data) - 1;
+		tx_desc->len = PKT_SIZE;
 	}
 
 	xsk_ring_prod__submit(&xsk->tx, batch_size);
@@ -916,9 +1167,12 @@ int main(int argc, char **argv)
 	for (i = 0; i < opt_num_xsks; i++)
 		xsks[num_socks++] = xsk_configure_socket(umem, rx, tx);
 
-	if (opt_bench == BENCH_TXONLY)
+	if (opt_bench == BENCH_TXONLY) {
+		gen_eth_hdr_data();
+
 		for (i = 0; i < NUM_FRAMES; i++)
 			gen_eth_frame(umem, i * opt_xsk_frame_size);
+	}
 
 	if (opt_num_xsks > 1 && opt_bench != BENCH_TXONLY)
 		enter_xsks_into_map(obj);
-- 
cgit v1.2.3-59-g8ed1b


From 46e3268eaaca9f8a0f145872b96fe6d54a232890 Mon Sep 17 00:00:00 2001
From: Jay Jayatheerthan <jay.jayatheerthan@intel.com>
Date: Fri, 20 Dec 2019 14:25:30 +0530
Subject: samples/bpf: xdpsock: Add option to specify transmit fill pattern

The UDP payload fill pattern can be specified using '-P' or '--tx-pkt-pattern'
option. It is an unsigned 32 bit field and defaulted to 0x12345678.

The IP and UDP checksum is calculated by the code as per the content of
the packet before transmission.

Signed-off-by: Jay Jayatheerthan <jay.jayatheerthan@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191220085530.4980-7-jay.jayatheerthan@intel.com
---
 samples/bpf/xdpsock_user.c | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

diff --git a/samples/bpf/xdpsock_user.c b/samples/bpf/xdpsock_user.c
index 2297158a32bd..d74c4c83fc93 100644
--- a/samples/bpf/xdpsock_user.c
+++ b/samples/bpf/xdpsock_user.c
@@ -76,7 +76,7 @@ static bool benchmark_done;
 static u32 opt_batch_size = 64;
 static int opt_pkt_count;
 static u16 opt_pkt_size = MIN_PKT_SIZE;
-static u32 pkt_fill_pattern = 0x12345678;
+static u32 opt_pkt_fill_pattern = 0x12345678;
 static int opt_poll;
 static int opt_interval = 1;
 static u32 opt_xdp_bind_flags = XDP_USE_NEED_WAKEUP;
@@ -517,7 +517,7 @@ static void gen_eth_hdr_data(void)
 	udp_hdr->len = htons(UDP_PKT_SIZE);
 
 	/* UDP data */
-	memset32_htonl(pkt_data + PKT_HDR_SIZE, pkt_fill_pattern,
+	memset32_htonl(pkt_data + PKT_HDR_SIZE, opt_pkt_fill_pattern,
 		       UDP_PKT_DATA_SIZE);
 
 	/* UDP header checksum */
@@ -630,6 +630,7 @@ static struct option long_options[] = {
 	{"batch-size", required_argument, 0, 'b'},
 	{"tx-pkt-count", required_argument, 0, 'C'},
 	{"tx-pkt-size", required_argument, 0, 's'},
+	{"tx-pkt-pattern", required_argument, 0, 'P'},
 	{0, 0, 0, 0}
 };
 
@@ -663,10 +664,11 @@ static void usage(const char *prog)
 		"  -s, --tx-pkt-size=n	Transmit packet size.\n"
 		"			(Default: %d bytes)\n"
 		"			Min size: %d, Max size %d.\n"
+		"  -P, --tx-pkt-pattern=nPacket fill pattern. Default: 0x%x\n"
 		"\n";
 	fprintf(stderr, str, prog, XSK_UMEM__DEFAULT_FRAME_SIZE,
 		opt_batch_size, MIN_PKT_SIZE, MIN_PKT_SIZE,
-		XSK_UMEM__DEFAULT_FRAME_SIZE);
+		XSK_UMEM__DEFAULT_FRAME_SIZE, opt_pkt_fill_pattern);
 
 	exit(EXIT_FAILURE);
 }
@@ -678,7 +680,7 @@ static void parse_command_line(int argc, char **argv)
 	opterr = 0;
 
 	for (;;) {
-		c = getopt_long(argc, argv, "Frtli:q:pSNn:czf:muMd:b:C:s:",
+		c = getopt_long(argc, argv, "Frtli:q:pSNn:czf:muMd:b:C:s:P:",
 				long_options, &option_index);
 		if (c == -1)
 			break;
@@ -756,6 +758,9 @@ static void parse_command_line(int argc, char **argv)
 				usage(basename(argv[0]));
 			}
 			break;
+		case 'P':
+			opt_pkt_fill_pattern = strtol(optarg, NULL, 16);
+			break;
 		default:
 			usage(basename(argv[0]));
 		}
-- 
cgit v1.2.3-59-g8ed1b


From f9e6bfdbaf0cf304d72c70a05d81acac01a04f48 Mon Sep 17 00:00:00 2001
From: Jesper Dangaard Brouer <brouer@redhat.com>
Date: Fri, 20 Dec 2019 17:19:36 +0100
Subject: samples/bpf: Xdp_redirect_cpu fix missing tracepoint attach

When sample xdp_redirect_cpu was converted to use libbpf, the
tracepoints used by this sample were not getting attached automatically
like with bpf_load.c. The BPF-maps was still getting loaded, thus
nobody notice that the tracepoints were not updating these maps.

This fix doesn't use the new skeleton code, as this bug was introduced
in v5.1 and stable might want to backport this. E.g. Red Hat QA uses
this sample as part of their testing.

Fixes: bbaf6029c49c ("samples/bpf: Convert XDP samples to libbpf usage")
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/157685877642.26195.2798780195186786841.stgit@firesoul
---
 samples/bpf/xdp_redirect_cpu_user.c | 59 ++++++++++++++++++++++++++++++++++---
 1 file changed, 55 insertions(+), 4 deletions(-)

diff --git a/samples/bpf/xdp_redirect_cpu_user.c b/samples/bpf/xdp_redirect_cpu_user.c
index 72af628529b5..79a2fb7d16cb 100644
--- a/samples/bpf/xdp_redirect_cpu_user.c
+++ b/samples/bpf/xdp_redirect_cpu_user.c
@@ -16,6 +16,10 @@ static const char *__doc__ =
 #include <getopt.h>
 #include <net/if.h>
 #include <time.h>
+#include <linux/limits.h>
+
+#define __must_check
+#include <linux/err.h>
 
 #include <arpa/inet.h>
 #include <linux/if_link.h>
@@ -46,6 +50,10 @@ static int cpus_count_map_fd;
 static int cpus_iterator_map_fd;
 static int exception_cnt_map_fd;
 
+#define NUM_TP 5
+struct bpf_link *tp_links[NUM_TP] = { 0 };
+static int tp_cnt = 0;
+
 /* Exit return codes */
 #define EXIT_OK		0
 #define EXIT_FAIL		1
@@ -88,6 +96,10 @@ static void int_exit(int sig)
 			printf("program on interface changed, not removing\n");
 		}
 	}
+	/* Detach tracepoints */
+	while (tp_cnt)
+		bpf_link__destroy(tp_links[--tp_cnt]);
+
 	exit(EXIT_OK);
 }
 
@@ -588,23 +600,61 @@ static void stats_poll(int interval, bool use_separators, char *prog_name,
 	free_stats_record(prev);
 }
 
+static struct bpf_link * attach_tp(struct bpf_object *obj,
+				   const char *tp_category,
+				   const char* tp_name)
+{
+	struct bpf_program *prog;
+	struct bpf_link *link;
+	char sec_name[PATH_MAX];
+	int len;
+
+	len = snprintf(sec_name, PATH_MAX, "tracepoint/%s/%s",
+		       tp_category, tp_name);
+	if (len < 0)
+		exit(EXIT_FAIL);
+
+	prog = bpf_object__find_program_by_title(obj, sec_name);
+	if (!prog) {
+		fprintf(stderr, "ERR: finding progsec: %s\n", sec_name);
+		exit(EXIT_FAIL_BPF);
+	}
+
+	link = bpf_program__attach_tracepoint(prog, tp_category, tp_name);
+	if (IS_ERR(link))
+		exit(EXIT_FAIL_BPF);
+
+	return link;
+}
+
+static void init_tracepoints(struct bpf_object *obj) {
+	tp_links[tp_cnt++] = attach_tp(obj, "xdp", "xdp_redirect_err");
+	tp_links[tp_cnt++] = attach_tp(obj, "xdp", "xdp_redirect_map_err");
+	tp_links[tp_cnt++] = attach_tp(obj, "xdp", "xdp_exception");
+	tp_links[tp_cnt++] = attach_tp(obj, "xdp", "xdp_cpumap_enqueue");
+	tp_links[tp_cnt++] = attach_tp(obj, "xdp", "xdp_cpumap_kthread");
+}
+
 static int init_map_fds(struct bpf_object *obj)
 {
-	cpu_map_fd = bpf_object__find_map_fd_by_name(obj, "cpu_map");
-	rx_cnt_map_fd = bpf_object__find_map_fd_by_name(obj, "rx_cnt");
+	/* Maps updated by tracepoints */
 	redirect_err_cnt_map_fd =
 		bpf_object__find_map_fd_by_name(obj, "redirect_err_cnt");
+	exception_cnt_map_fd =
+		bpf_object__find_map_fd_by_name(obj, "exception_cnt");
 	cpumap_enqueue_cnt_map_fd =
 		bpf_object__find_map_fd_by_name(obj, "cpumap_enqueue_cnt");
 	cpumap_kthread_cnt_map_fd =
 		bpf_object__find_map_fd_by_name(obj, "cpumap_kthread_cnt");
+
+	/* Maps used by XDP */
+	rx_cnt_map_fd = bpf_object__find_map_fd_by_name(obj, "rx_cnt");
+	cpu_map_fd = bpf_object__find_map_fd_by_name(obj, "cpu_map");
 	cpus_available_map_fd =
 		bpf_object__find_map_fd_by_name(obj, "cpus_available");
 	cpus_count_map_fd = bpf_object__find_map_fd_by_name(obj, "cpus_count");
 	cpus_iterator_map_fd =
 		bpf_object__find_map_fd_by_name(obj, "cpus_iterator");
-	exception_cnt_map_fd =
-		bpf_object__find_map_fd_by_name(obj, "exception_cnt");
 
 	if (cpu_map_fd < 0 || rx_cnt_map_fd < 0 ||
 	    redirect_err_cnt_map_fd < 0 || cpumap_enqueue_cnt_map_fd < 0 ||
@@ -662,6 +712,7 @@ int main(int argc, char **argv)
 			strerror(errno));
 		return EXIT_FAIL;
 	}
+	init_tracepoints(obj);
 	if (init_map_fds(obj) < 0) {
 		fprintf(stderr, "bpf_object__find_map_fd_by_name failed\n");
 		return EXIT_FAIL;
-- 
cgit v1.2.3-59-g8ed1b


From dba261f35d16a8e4b70962d0f8391290510a8201 Mon Sep 17 00:00:00 2001
From: Oleksij Rempel <o.rempel@pengutronix.de>
Date: Wed, 18 Dec 2019 09:02:12 +0100
Subject: dt-bindings: net: dsa: qca, ar9331 switch documentation

Atheros AR9331 has built-in 5 port switch. The switch can be configured
to use all 5 or 4 ports. One of built-in PHYs can be used by first built-in
ethernet controller or to be used directly by the switch over second ethernet
controller.

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../devicetree/bindings/net/dsa/ar9331.txt         | 148 +++++++++++++++++++++
 1 file changed, 148 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/net/dsa/ar9331.txt

diff --git a/Documentation/devicetree/bindings/net/dsa/ar9331.txt b/Documentation/devicetree/bindings/net/dsa/ar9331.txt
new file mode 100644
index 000000000000..320607cbbb17
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/dsa/ar9331.txt
@@ -0,0 +1,148 @@
+Atheros AR9331 built-in switch
+=============================
+
+It is a switch built-in to Atheros AR9331 WiSoC and addressable over internal
+MDIO bus. All PHYs are built-in as well.
+
+Required properties:
+
+ - compatible: should be: "qca,ar9331-switch"
+ - reg: Address on the MII bus for the switch.
+ - resets : Must contain an entry for each entry in reset-names.
+ - reset-names : Must include the following entries: "switch"
+ - interrupt-parent: Phandle to the parent interrupt controller
+ - interrupts: IRQ line for the switch
+ - interrupt-controller: Indicates the switch is itself an interrupt
+   controller. This is used for the PHY interrupts.
+ - #interrupt-cells: must be 1
+ - mdio: Container of PHY and devices on the switches MDIO bus.
+
+See Documentation/devicetree/bindings/net/dsa/dsa.txt for a list of additional
+required and optional properties.
+Examples:
+
+eth0: ethernet@19000000 {
+	compatible = "qca,ar9330-eth";
+	reg = <0x19000000 0x200>;
+	interrupts = <4>;
+
+	resets = <&rst 9>, <&rst 22>;
+	reset-names = "mac", "mdio";
+	clocks = <&pll ATH79_CLK_AHB>, <&pll ATH79_CLK_AHB>;
+	clock-names = "eth", "mdio";
+
+	phy-mode = "mii";
+	phy-handle = <&phy_port4>;
+};
+
+eth1: ethernet@1a000000 {
+	compatible = "qca,ar9330-eth";
+	reg = <0x1a000000 0x200>;
+	interrupts = <5>;
+	resets = <&rst 13>, <&rst 23>;
+	reset-names = "mac", "mdio";
+	clocks = <&pll ATH79_CLK_AHB>, <&pll ATH79_CLK_AHB>;
+	clock-names = "eth", "mdio";
+
+	phy-mode = "gmii";
+
+	fixed-link {
+		speed = <1000>;
+		full-duplex;
+	};
+
+	mdio {
+		#address-cells = <1>;
+		#size-cells = <0>;
+
+		switch10: switch@10 {
+			#address-cells = <1>;
+			#size-cells = <0>;
+
+			compatible = "qca,ar9331-switch";
+			reg = <0x10>;
+			resets = <&rst 8>;
+			reset-names = "switch";
+
+			interrupt-parent = <&miscintc>;
+			interrupts = <12>;
+
+			interrupt-controller;
+			#interrupt-cells = <1>;
+
+			ports {
+				#address-cells = <1>;
+				#size-cells = <0>;
+
+				switch_port0: port@0 {
+					reg = <0x0>;
+					label = "cpu";
+					ethernet = <&eth1>;
+
+					phy-mode = "gmii";
+
+					fixed-link {
+						speed = <1000>;
+						full-duplex;
+					};
+				};
+
+				switch_port1: port@1 {
+					reg = <0x1>;
+					phy-handle = <&phy_port0>;
+					phy-mode = "internal";
+				};
+
+				switch_port2: port@2 {
+					reg = <0x2>;
+					phy-handle = <&phy_port1>;
+					phy-mode = "internal";
+				};
+
+				switch_port3: port@3 {
+					reg = <0x3>;
+					phy-handle = <&phy_port2>;
+					phy-mode = "internal";
+				};
+
+				switch_port4: port@4 {
+					reg = <0x4>;
+					phy-handle = <&phy_port3>;
+					phy-mode = "internal";
+				};
+			};
+
+			mdio {
+				#address-cells = <1>;
+				#size-cells = <0>;
+
+				interrupt-parent = <&switch10>;
+
+				phy_port0: phy@0 {
+					reg = <0x0>;
+					interrupts = <0>;
+				};
+
+				phy_port1: phy@1 {
+					reg = <0x1>;
+					interrupts = <0>;
+				};
+
+				phy_port2: phy@2 {
+					reg = <0x2>;
+					interrupts = <0>;
+				};
+
+				phy_port3: phy@3 {
+					reg = <0x3>;
+					interrupts = <0>;
+				};
+
+				phy_port4: phy@4 {
+					reg = <0x4>;
+					interrupts = <0>;
+				};
+			};
+		};
+	};
+};
-- 
cgit v1.2.3-59-g8ed1b


From 4eb7ae7a301d3586c3351e81d5c3cfe2304a1a6a Mon Sep 17 00:00:00 2001
From: Oleksij Rempel <o.rempel@pengutronix.de>
Date: Wed, 18 Dec 2019 09:02:13 +0100
Subject: MIPS: ath79: ar9331: add ar9331-switch node

Add switch node supported by dsa ar9331 driver.

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 arch/mips/boot/dts/qca/ar9331.dtsi           | 119 ++++++++++++++++++++++++++-
 arch/mips/boot/dts/qca/ar9331_dpt_module.dts |  13 +++
 2 files changed, 131 insertions(+), 1 deletion(-)

diff --git a/arch/mips/boot/dts/qca/ar9331.dtsi b/arch/mips/boot/dts/qca/ar9331.dtsi
index 5cfc9d347826..8f5aed760abb 100644
--- a/arch/mips/boot/dts/qca/ar9331.dtsi
+++ b/arch/mips/boot/dts/qca/ar9331.dtsi
@@ -126,6 +126,9 @@
 			clocks = <&pll ATH79_CLK_AHB>, <&pll ATH79_CLK_AHB>;
 			clock-names = "eth", "mdio";
 
+			phy-mode = "mii";
+			phy-handle = <&phy_port4>;
+
 			status = "disabled";
 		};
 
@@ -133,13 +136,127 @@
 			compatible = "qca,ar9330-eth";
 			reg = <0x1a000000 0x200>;
 			interrupts = <5>;
-
 			resets = <&rst 13>, <&rst 23>;
 			reset-names = "mac", "mdio";
 			clocks = <&pll ATH79_CLK_AHB>, <&pll ATH79_CLK_AHB>;
 			clock-names = "eth", "mdio";
 
+			phy-mode = "gmii";
+
 			status = "disabled";
+
+			fixed-link {
+				speed = <1000>;
+				full-duplex;
+			};
+
+			mdio {
+				#address-cells = <1>;
+				#size-cells = <0>;
+
+				switch10: switch@10 {
+					#address-cells = <1>;
+					#size-cells = <0>;
+
+					compatible = "qca,ar9331-switch";
+					reg = <0x10>;
+					resets = <&rst 8>;
+					reset-names = "switch";
+
+					interrupt-parent = <&miscintc>;
+					interrupts = <12>;
+
+					interrupt-controller;
+					#interrupt-cells = <1>;
+
+					ports {
+						#address-cells = <1>;
+						#size-cells = <0>;
+
+						switch_port0: port@0 {
+							reg = <0x0>;
+							label = "cpu";
+							ethernet = <&eth1>;
+
+							phy-mode = "gmii";
+
+							fixed-link {
+								speed = <1000>;
+								full-duplex;
+							};
+						};
+
+						switch_port1: port@1 {
+							reg = <0x1>;
+							phy-handle = <&phy_port0>;
+							phy-mode = "internal";
+
+							status = "disabled";
+						};
+
+						switch_port2: port@2 {
+							reg = <0x2>;
+							phy-handle = <&phy_port1>;
+							phy-mode = "internal";
+
+							status = "disabled";
+						};
+
+						switch_port3: port@3 {
+							reg = <0x3>;
+							phy-handle = <&phy_port2>;
+							phy-mode = "internal";
+
+							status = "disabled";
+						};
+
+						switch_port4: port@4 {
+							reg = <0x4>;
+							phy-handle = <&phy_port3>;
+							phy-mode = "internal";
+
+							status = "disabled";
+						};
+					};
+
+					mdio {
+						#address-cells = <1>;
+						#size-cells = <0>;
+
+						interrupt-parent = <&switch10>;
+
+						phy_port0: phy@0 {
+							reg = <0x0>;
+							interrupts = <0>;
+							status = "disabled";
+						};
+
+						phy_port1: phy@1 {
+							reg = <0x1>;
+							interrupts = <0>;
+							status = "disabled";
+						};
+
+						phy_port2: phy@2 {
+							reg = <0x2>;
+							interrupts = <0>;
+							status = "disabled";
+						};
+
+						phy_port3: phy@3 {
+							reg = <0x3>;
+							interrupts = <0>;
+							status = "disabled";
+						};
+
+						phy_port4: phy@4 {
+							reg = <0x4>;
+							interrupts = <0>;
+							status = "disabled";
+						};
+					};
+				};
+			};
 		};
 
 		usb: usb@1b000100 {
diff --git a/arch/mips/boot/dts/qca/ar9331_dpt_module.dts b/arch/mips/boot/dts/qca/ar9331_dpt_module.dts
index 77bab823eb3b..0f2b20044834 100644
--- a/arch/mips/boot/dts/qca/ar9331_dpt_module.dts
+++ b/arch/mips/boot/dts/qca/ar9331_dpt_module.dts
@@ -84,3 +84,16 @@
 &eth1 {
 	status = "okay";
 };
+
+&switch_port1 {
+	label = "lan0";
+	status = "okay";
+};
+
+&phy_port0 {
+	status = "okay";
+};
+
+&phy_port4 {
+	status = "okay";
+};
-- 
cgit v1.2.3-59-g8ed1b


From 48fda74f0a9377ce2145ac5f06b6db0274880826 Mon Sep 17 00:00:00 2001
From: Oleksij Rempel <o.rempel@pengutronix.de>
Date: Wed, 18 Dec 2019 09:02:14 +0100
Subject: net: dsa: add support for Atheros AR9331 TAG format

Add support for tag format used in Atheros AR9331 built-in switch.

Reviewed-by: Vivien Didelot <vivien.didelot@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/dsa.h    |  2 ++
 net/dsa/Kconfig      |  6 ++++
 net/dsa/Makefile     |  1 +
 net/dsa/tag_ar9331.c | 96 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 105 insertions(+)
 create mode 100644 net/dsa/tag_ar9331.c

diff --git a/include/net/dsa.h b/include/net/dsa.h
index 6767dc3f66c0..da5578db228e 100644
--- a/include/net/dsa.h
+++ b/include/net/dsa.h
@@ -43,6 +43,7 @@ struct phylink_link_state;
 #define DSA_TAG_PROTO_SJA1105_VALUE		13
 #define DSA_TAG_PROTO_KSZ8795_VALUE		14
 #define DSA_TAG_PROTO_OCELOT_VALUE		15
+#define DSA_TAG_PROTO_AR9331_VALUE		16
 
 enum dsa_tag_protocol {
 	DSA_TAG_PROTO_NONE		= DSA_TAG_PROTO_NONE_VALUE,
@@ -61,6 +62,7 @@ enum dsa_tag_protocol {
 	DSA_TAG_PROTO_SJA1105		= DSA_TAG_PROTO_SJA1105_VALUE,
 	DSA_TAG_PROTO_KSZ8795		= DSA_TAG_PROTO_KSZ8795_VALUE,
 	DSA_TAG_PROTO_OCELOT		= DSA_TAG_PROTO_OCELOT_VALUE,
+	DSA_TAG_PROTO_AR9331		= DSA_TAG_PROTO_AR9331_VALUE,
 };
 
 struct packet_type;
diff --git a/net/dsa/Kconfig b/net/dsa/Kconfig
index 1e6c3cac11e6..92663dcb3aa2 100644
--- a/net/dsa/Kconfig
+++ b/net/dsa/Kconfig
@@ -29,6 +29,12 @@ config NET_DSA_TAG_8021Q
 
 	  Drivers which use these helpers should select this as dependency.
 
+config NET_DSA_TAG_AR9331
+	tristate "Tag driver for Atheros AR9331 SoC with built-in switch"
+	help
+	  Say Y or M if you want to enable support for tagging frames for
+	  the Atheros AR9331 SoC with built-in switch.
+
 config NET_DSA_TAG_BRCM_COMMON
 	tristate
 	default n
diff --git a/net/dsa/Makefile b/net/dsa/Makefile
index 9a482c38bdb1..108486cfdeef 100644
--- a/net/dsa/Makefile
+++ b/net/dsa/Makefile
@@ -5,6 +5,7 @@ dsa_core-y += dsa.o dsa2.o master.o port.o slave.o switch.o
 
 # tagging formats
 obj-$(CONFIG_NET_DSA_TAG_8021Q) += tag_8021q.o
+obj-$(CONFIG_NET_DSA_TAG_AR9331) += tag_ar9331.o
 obj-$(CONFIG_NET_DSA_TAG_BRCM_COMMON) += tag_brcm.o
 obj-$(CONFIG_NET_DSA_TAG_DSA) += tag_dsa.o
 obj-$(CONFIG_NET_DSA_TAG_EDSA) += tag_edsa.o
diff --git a/net/dsa/tag_ar9331.c b/net/dsa/tag_ar9331.c
new file mode 100644
index 000000000000..466ffa92a474
--- /dev/null
+++ b/net/dsa/tag_ar9331.c
@@ -0,0 +1,96 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (c) 2019 Pengutronix, Oleksij Rempel <kernel@pengutronix.de>
+ */
+
+
+#include <linux/bitfield.h>
+#include <linux/etherdevice.h>
+
+#include "dsa_priv.h"
+
+#define AR9331_HDR_LEN			2
+#define AR9331_HDR_VERSION		1
+
+#define AR9331_HDR_VERSION_MASK		GENMASK(15, 14)
+#define AR9331_HDR_PRIORITY_MASK	GENMASK(13, 12)
+#define AR9331_HDR_TYPE_MASK		GENMASK(10, 8)
+#define AR9331_HDR_BROADCAST		BIT(7)
+#define AR9331_HDR_FROM_CPU		BIT(6)
+/* AR9331_HDR_RESERVED - not used or may be version field.
+ * According to the AR8216 doc it should 0b10. On AR9331 it is 0b11 on RX path
+ * and should be set to 0b11 to make it work.
+ */
+#define AR9331_HDR_RESERVED_MASK	GENMASK(5, 4)
+#define AR9331_HDR_PORT_NUM_MASK	GENMASK(3, 0)
+
+static struct sk_buff *ar9331_tag_xmit(struct sk_buff *skb,
+				       struct net_device *dev)
+{
+	struct dsa_port *dp = dsa_slave_to_port(dev);
+	__le16 *phdr;
+	u16 hdr;
+
+	if (skb_cow_head(skb, 0) < 0)
+		return NULL;
+
+	phdr = skb_push(skb, AR9331_HDR_LEN);
+
+	hdr = FIELD_PREP(AR9331_HDR_VERSION_MASK, AR9331_HDR_VERSION);
+	hdr |= AR9331_HDR_FROM_CPU | dp->index;
+	/* 0b10 for AR8216 and 0b11 for AR9331 */
+	hdr |= AR9331_HDR_RESERVED_MASK;
+
+	phdr[0] = cpu_to_le16(hdr);
+
+	return skb;
+}
+
+static struct sk_buff *ar9331_tag_rcv(struct sk_buff *skb,
+				      struct net_device *ndev,
+				      struct packet_type *pt)
+{
+	u8 ver, port;
+	u16 hdr;
+
+	if (unlikely(!pskb_may_pull(skb, AR9331_HDR_LEN)))
+		return NULL;
+
+	hdr = le16_to_cpu(*(__le16 *)skb_mac_header(skb));
+
+	ver = FIELD_GET(AR9331_HDR_VERSION_MASK, hdr);
+	if (unlikely(ver != AR9331_HDR_VERSION)) {
+		netdev_warn_once(ndev, "%s:%i wrong header version 0x%2x\n",
+				 __func__, __LINE__, hdr);
+		return NULL;
+	}
+
+	if (unlikely(hdr & AR9331_HDR_FROM_CPU)) {
+		netdev_warn_once(ndev, "%s:%i packet should not be from cpu 0x%2x\n",
+				 __func__, __LINE__, hdr);
+		return NULL;
+	}
+
+	skb_pull_rcsum(skb, AR9331_HDR_LEN);
+
+	/* Get source port information */
+	port = FIELD_GET(AR9331_HDR_PORT_NUM_MASK, hdr);
+
+	skb->dev = dsa_master_find_slave(ndev, 0, port);
+	if (!skb->dev)
+		return NULL;
+
+	return skb;
+}
+
+static const struct dsa_device_ops ar9331_netdev_ops = {
+	.name	= "ar9331",
+	.proto	= DSA_TAG_PROTO_AR9331,
+	.xmit	= ar9331_tag_xmit,
+	.rcv	= ar9331_tag_rcv,
+	.overhead = AR9331_HDR_LEN,
+};
+
+MODULE_LICENSE("GPL v2");
+MODULE_ALIAS_DSA_TAG_DRIVER(DSA_TAG_PROTO_AR9331);
+module_dsa_tag_driver(ar9331_netdev_ops);
-- 
cgit v1.2.3-59-g8ed1b


From ec6698c272debe16c8825c88a5b3c3f68a12ef4a Mon Sep 17 00:00:00 2001
From: Oleksij Rempel <o.rempel@pengutronix.de>
Date: Wed, 18 Dec 2019 09:02:15 +0100
Subject: net: dsa: add support for Atheros AR9331 built-in switch

Provide basic support for Atheros AR9331 built-in switch. So far it
works as port multiplexer without any hardware offloading support.

Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Vivien Didelot <vivien.didelot@gmail.com>
Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/Kconfig      |   2 +
 drivers/net/dsa/Makefile     |   1 +
 drivers/net/dsa/qca/Kconfig  |   9 +
 drivers/net/dsa/qca/Makefile |   2 +
 drivers/net/dsa/qca/ar9331.c | 855 +++++++++++++++++++++++++++++++++++++++++++
 5 files changed, 869 insertions(+)
 create mode 100644 drivers/net/dsa/qca/Kconfig
 create mode 100644 drivers/net/dsa/qca/Makefile
 create mode 100644 drivers/net/dsa/qca/ar9331.c

diff --git a/drivers/net/dsa/Kconfig b/drivers/net/dsa/Kconfig
index c7667645f04a..cbd74a72d0a1 100644
--- a/drivers/net/dsa/Kconfig
+++ b/drivers/net/dsa/Kconfig
@@ -54,6 +54,8 @@ source "drivers/net/dsa/mv88e6xxx/Kconfig"
 
 source "drivers/net/dsa/ocelot/Kconfig"
 
+source "drivers/net/dsa/qca/Kconfig"
+
 source "drivers/net/dsa/sja1105/Kconfig"
 
 config NET_DSA_QCA8K
diff --git a/drivers/net/dsa/Makefile b/drivers/net/dsa/Makefile
index 9d384a32b3a2..4a943ccc2ca4 100644
--- a/drivers/net/dsa/Makefile
+++ b/drivers/net/dsa/Makefile
@@ -21,4 +21,5 @@ obj-y				+= b53/
 obj-y				+= microchip/
 obj-y				+= mv88e6xxx/
 obj-y				+= ocelot/
+obj-y				+= qca/
 obj-y				+= sja1105/
diff --git a/drivers/net/dsa/qca/Kconfig b/drivers/net/dsa/qca/Kconfig
new file mode 100644
index 000000000000..e3c8d715a18f
--- /dev/null
+++ b/drivers/net/dsa/qca/Kconfig
@@ -0,0 +1,9 @@
+# SPDX-License-Identifier: GPL-2.0-only
+config NET_DSA_AR9331
+	tristate "Qualcomm Atheros AR9331 Ethernet switch support"
+	depends on NET_DSA
+	select NET_DSA_TAG_AR9331
+	select REGMAP
+	---help---
+	  This enables support for the Qualcomm Atheros AR9331 built-in Ethernet
+	  switch.
diff --git a/drivers/net/dsa/qca/Makefile b/drivers/net/dsa/qca/Makefile
new file mode 100644
index 000000000000..274022319066
--- /dev/null
+++ b/drivers/net/dsa/qca/Makefile
@@ -0,0 +1,2 @@
+# SPDX-License-Identifier: GPL-2.0-only
+obj-$(CONFIG_NET_DSA_AR9331)	+= ar9331.o
diff --git a/drivers/net/dsa/qca/ar9331.c b/drivers/net/dsa/qca/ar9331.c
new file mode 100644
index 000000000000..0d1a7cd85fe8
--- /dev/null
+++ b/drivers/net/dsa/qca/ar9331.c
@@ -0,0 +1,855 @@
+// SPDX-License-Identifier: GPL-2.0-only
+// Copyright (c) 2019 Pengutronix, Oleksij Rempel <kernel@pengutronix.de>
+/*
+ *                   +----------------------+
+ * GMAC1----RGMII----|--MAC0                |
+ *      \---MDIO1----|--REGs                |----MDIO3----\
+ *                   |                      |             |  +------+
+ *                   |                      |             +--|      |
+ *                   |                 MAC1-|----RMII--M-----| PHY0 |-o P0
+ *                   |                      |          |  |  +------+
+ *                   |                      |          |  +--|      |
+ *                   |                 MAC2-|----RMII--------| PHY1 |-o P1
+ *                   |                      |          |  |  +------+
+ *                   |                      |          |  +--|      |
+ *                   |                 MAC3-|----RMII--------| PHY2 |-o P2
+ *                   |                      |          |  |  +------+
+ *                   |                      |          |  +--|      |
+ *                   |                 MAC4-|----RMII--------| PHY3 |-o P3
+ *                   |                      |          |  |  +------+
+ *                   |                      |          |  +--|      |
+ *                   |                 MAC5-|--+-RMII--M-----|-PHY4-|-o P4
+ *                   |                      |  |       |     +------+
+ *                   +----------------------+  |       \--CFG_SW_PHY_SWAP
+ * GMAC0---------------RMII--------------------/        \-CFG_SW_PHY_ADDR_SWAP
+ *      \---MDIO0--NC
+ *
+ * GMAC0 and MAC5 are connected together and use same PHY. Depending on
+ * configuration it can be PHY4 (default) or PHY0. Only GMAC0 or MAC5 can be
+ * used at same time. If GMAC0 is used (default) then MAC5 should be disabled.
+ *
+ * CFG_SW_PHY_SWAP - swap connections of PHY0 and PHY4. If this bit is not set
+ * PHY4 is connected to GMAC0/MAC5 bundle and PHY0 is connected to MAC1. If this
+ * bit is set, PHY4 is connected to MAC1 and PHY0 is connected to GMAC0/MAC5
+ * bundle.
+ *
+ * CFG_SW_PHY_ADDR_SWAP - swap addresses of PHY0 and PHY4
+ *
+ * CFG_SW_PHY_SWAP and CFG_SW_PHY_ADDR_SWAP are part of SoC specific register
+ * set and not related to switch internal registers.
+ */
+
+#include <linux/bitfield.h>
+#include <linux/module.h>
+#include <linux/of_irq.h>
+#include <linux/of_mdio.h>
+#include <linux/regmap.h>
+#include <linux/reset.h>
+#include <net/dsa.h>
+
+#define AR9331_SW_NAME				"ar9331_switch"
+#define AR9331_SW_PORTS				6
+
+/* dummy reg to change page */
+#define AR9331_SW_REG_PAGE			0x40000
+
+/* Global Interrupt */
+#define AR9331_SW_REG_GINT			0x10
+#define AR9331_SW_REG_GINT_MASK			0x14
+#define AR9331_SW_GINT_PHY_INT			BIT(2)
+
+#define AR9331_SW_REG_FLOOD_MASK		0x2c
+#define AR9331_SW_FLOOD_MASK_BROAD_TO_CPU	BIT(26)
+
+#define AR9331_SW_REG_GLOBAL_CTRL		0x30
+#define AR9331_SW_GLOBAL_CTRL_MFS_M		GENMASK(13, 0)
+
+#define AR9331_SW_REG_MDIO_CTRL			0x98
+#define AR9331_SW_MDIO_CTRL_BUSY		BIT(31)
+#define AR9331_SW_MDIO_CTRL_MASTER_EN		BIT(30)
+#define AR9331_SW_MDIO_CTRL_CMD_READ		BIT(27)
+#define AR9331_SW_MDIO_CTRL_PHY_ADDR_M		GENMASK(25, 21)
+#define AR9331_SW_MDIO_CTRL_REG_ADDR_M		GENMASK(20, 16)
+#define AR9331_SW_MDIO_CTRL_DATA_M		GENMASK(16, 0)
+
+#define AR9331_SW_REG_PORT_STATUS(_port)	(0x100 + (_port) * 0x100)
+
+/* FLOW_LINK_EN - enable mac flow control config auto-neg with phy.
+ * If not set, mac can be config by software.
+ */
+#define AR9331_SW_PORT_STATUS_FLOW_LINK_EN	BIT(12)
+
+/* LINK_EN - If set, MAC is configured from PHY link status.
+ * If not set, MAC should be configured by software.
+ */
+#define AR9331_SW_PORT_STATUS_LINK_EN		BIT(9)
+#define AR9331_SW_PORT_STATUS_DUPLEX_MODE	BIT(6)
+#define AR9331_SW_PORT_STATUS_RX_FLOW_EN	BIT(5)
+#define AR9331_SW_PORT_STATUS_TX_FLOW_EN	BIT(4)
+#define AR9331_SW_PORT_STATUS_RXMAC		BIT(3)
+#define AR9331_SW_PORT_STATUS_TXMAC		BIT(2)
+#define AR9331_SW_PORT_STATUS_SPEED_M		GENMASK(1, 0)
+#define AR9331_SW_PORT_STATUS_SPEED_1000	2
+#define AR9331_SW_PORT_STATUS_SPEED_100		1
+#define AR9331_SW_PORT_STATUS_SPEED_10		0
+
+#define AR9331_SW_PORT_STATUS_MAC_MASK \
+	(AR9331_SW_PORT_STATUS_TXMAC | AR9331_SW_PORT_STATUS_RXMAC)
+
+#define AR9331_SW_PORT_STATUS_LINK_MASK \
+	(AR9331_SW_PORT_STATUS_LINK_EN | AR9331_SW_PORT_STATUS_FLOW_LINK_EN | \
+	 AR9331_SW_PORT_STATUS_DUPLEX_MODE | \
+	 AR9331_SW_PORT_STATUS_RX_FLOW_EN | AR9331_SW_PORT_STATUS_TX_FLOW_EN | \
+	 AR9331_SW_PORT_STATUS_SPEED_M)
+
+/* Phy bypass mode
+ * ------------------------------------------------------------------------
+ * Bit:   | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 |10 |11 |12 |13 |14 |15 |
+ *
+ * real   | start |   OP  | PhyAddr           |  Reg Addr         |  TA   |
+ * atheros| start |   OP  | 2'b00 |PhyAdd[2:0]|  Reg Addr[4:0]    |  TA   |
+ *
+ *
+ * Bit:   |16 |17 |18 |19 |20 |21 |22 |23 |24 |25 |26 |27 |28 |29 |30 |31 |
+ * real   |  Data                                                         |
+ * atheros|  Data                                                         |
+ *
+ * ------------------------------------------------------------------------
+ * Page address mode
+ * ------------------------------------------------------------------------
+ * Bit:   | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 |10 |11 |12 |13 |14 |15 |
+ * real   | start |   OP  | PhyAddr           |  Reg Addr         |  TA   |
+ * atheros| start |   OP  | 2'b11 |                          8'b0 |  TA   |
+ *
+ * Bit:   |16 |17 |18 |19 |20 |21 |22 |23 |24 |25 |26 |27 |28 |29 |30 |31 |
+ * real   |  Data                                                         |
+ * atheros|                       | Page [9:0]                            |
+ */
+/* In case of Page Address mode, Bit[18:9] of 32 bit register address should be
+ * written to bits[9:0] of mdio data register.
+ */
+#define AR9331_SW_ADDR_PAGE			GENMASK(18, 9)
+
+/* ------------------------------------------------------------------------
+ * Normal register access mode
+ * ------------------------------------------------------------------------
+ * Bit:   | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 |10 |11 |12 |13 |14 |15 |
+ * real   | start |   OP  | PhyAddr           |  Reg Addr         |  TA   |
+ * atheros| start |   OP  | 2'b10 |  low_addr[7:0]                |  TA   |
+ *
+ * Bit:   |16 |17 |18 |19 |20 |21 |22 |23 |24 |25 |26 |27 |28 |29 |30 |31 |
+ * real   |  Data                                                         |
+ * atheros|  Data                                                         |
+ * ------------------------------------------------------------------------
+ */
+#define AR9331_SW_LOW_ADDR_PHY			GENMASK(8, 6)
+#define AR9331_SW_LOW_ADDR_REG			GENMASK(5, 1)
+
+#define AR9331_SW_MDIO_PHY_MODE_M		GENMASK(4, 3)
+#define AR9331_SW_MDIO_PHY_MODE_PAGE		3
+#define AR9331_SW_MDIO_PHY_MODE_REG		2
+#define AR9331_SW_MDIO_PHY_MODE_BYPASS		0
+#define AR9331_SW_MDIO_PHY_ADDR_M		GENMASK(2, 0)
+
+/* Empirical determined values */
+#define AR9331_SW_MDIO_POLL_SLEEP_US		1
+#define AR9331_SW_MDIO_POLL_TIMEOUT_US		20
+
+struct ar9331_sw_priv {
+	struct device *dev;
+	struct dsa_switch ds;
+	struct dsa_switch_ops ops;
+	struct irq_domain *irqdomain;
+	struct mii_bus *mbus; /* mdio master */
+	struct mii_bus *sbus; /* mdio slave */
+	struct regmap *regmap;
+	struct reset_control *sw_reset;
+};
+
+/* Warning: switch reset will reset last AR9331_SW_MDIO_PHY_MODE_PAGE request
+ * If some kind of optimization is used, the request should be repeated.
+ */
+static int ar9331_sw_reset(struct ar9331_sw_priv *priv)
+{
+	int ret;
+
+	ret = reset_control_assert(priv->sw_reset);
+	if (ret)
+		goto error;
+
+	/* AR9331 doc do not provide any information about proper reset
+	 * sequence. The AR8136 (the closes switch to the AR9331) doc says:
+	 * reset duration should be greater than 10ms. So, let's use this value
+	 * for now.
+	 */
+	usleep_range(10000, 15000);
+	ret = reset_control_deassert(priv->sw_reset);
+	if (ret)
+		goto error;
+	/* There is no information on how long should we wait after reset.
+	 * AR8136 has an EEPROM and there is an Interrupt for EEPROM load
+	 * status. AR9331 has no EEPROM support.
+	 * For now, do not wait. In case AR8136 will be needed, the after
+	 * reset delay can be added as well.
+	 */
+
+	return 0;
+error:
+	dev_err_ratelimited(priv->dev, "%s: %i\n", __func__, ret);
+	return ret;
+}
+
+static int ar9331_sw_mbus_write(struct mii_bus *mbus, int port, int regnum,
+				u16 data)
+{
+	struct ar9331_sw_priv *priv = mbus->priv;
+	struct regmap *regmap = priv->regmap;
+	u32 val;
+	int ret;
+
+	ret = regmap_write(regmap, AR9331_SW_REG_MDIO_CTRL,
+			   AR9331_SW_MDIO_CTRL_BUSY |
+			   AR9331_SW_MDIO_CTRL_MASTER_EN |
+			   FIELD_PREP(AR9331_SW_MDIO_CTRL_PHY_ADDR_M, port) |
+			   FIELD_PREP(AR9331_SW_MDIO_CTRL_REG_ADDR_M, regnum) |
+			   FIELD_PREP(AR9331_SW_MDIO_CTRL_DATA_M, data));
+	if (ret)
+		goto error;
+
+	ret = regmap_read_poll_timeout(regmap, AR9331_SW_REG_MDIO_CTRL, val,
+				       !(val & AR9331_SW_MDIO_CTRL_BUSY),
+				       AR9331_SW_MDIO_POLL_SLEEP_US,
+				       AR9331_SW_MDIO_POLL_TIMEOUT_US);
+	if (ret)
+		goto error;
+
+	return 0;
+error:
+	dev_err_ratelimited(priv->dev, "PHY write error: %i\n", ret);
+	return ret;
+}
+
+static int ar9331_sw_mbus_read(struct mii_bus *mbus, int port, int regnum)
+{
+	struct ar9331_sw_priv *priv = mbus->priv;
+	struct regmap *regmap = priv->regmap;
+	u32 val;
+	int ret;
+
+	ret = regmap_write(regmap, AR9331_SW_REG_MDIO_CTRL,
+			   AR9331_SW_MDIO_CTRL_BUSY |
+			   AR9331_SW_MDIO_CTRL_MASTER_EN |
+			   AR9331_SW_MDIO_CTRL_CMD_READ |
+			   FIELD_PREP(AR9331_SW_MDIO_CTRL_PHY_ADDR_M, port) |
+			   FIELD_PREP(AR9331_SW_MDIO_CTRL_REG_ADDR_M, regnum));
+	if (ret)
+		goto error;
+
+	ret = regmap_read_poll_timeout(regmap, AR9331_SW_REG_MDIO_CTRL, val,
+				       !(val & AR9331_SW_MDIO_CTRL_BUSY),
+				       AR9331_SW_MDIO_POLL_SLEEP_US,
+				       AR9331_SW_MDIO_POLL_TIMEOUT_US);
+	if (ret)
+		goto error;
+
+	ret = regmap_read(regmap, AR9331_SW_REG_MDIO_CTRL, &val);
+	if (ret)
+		goto error;
+
+	return FIELD_GET(AR9331_SW_MDIO_CTRL_DATA_M, val);
+
+error:
+	dev_err_ratelimited(priv->dev, "PHY read error: %i\n", ret);
+	return ret;
+}
+
+static int ar9331_sw_mbus_init(struct ar9331_sw_priv *priv)
+{
+	struct device *dev = priv->dev;
+	static struct mii_bus *mbus;
+	struct device_node *np, *mnp;
+	int ret;
+
+	np = dev->of_node;
+
+	mbus = devm_mdiobus_alloc(dev);
+	if (!mbus)
+		return -ENOMEM;
+
+	mbus->name = np->full_name;
+	snprintf(mbus->id, MII_BUS_ID_SIZE, "%pOF", np);
+
+	mbus->read = ar9331_sw_mbus_read;
+	mbus->write = ar9331_sw_mbus_write;
+	mbus->priv = priv;
+	mbus->parent = dev;
+
+	mnp = of_get_child_by_name(np, "mdio");
+	if (!mnp)
+		return -ENODEV;
+
+	ret = of_mdiobus_register(mbus, mnp);
+	of_node_put(mnp);
+	if (ret)
+		return ret;
+
+	priv->mbus = mbus;
+
+	return 0;
+}
+
+static int ar9331_sw_setup(struct dsa_switch *ds)
+{
+	struct ar9331_sw_priv *priv = (struct ar9331_sw_priv *)ds->priv;
+	struct regmap *regmap = priv->regmap;
+	int ret;
+
+	ret = ar9331_sw_reset(priv);
+	if (ret)
+		return ret;
+
+	/* Reset will set proper defaults. CPU - Port0 will be enabled and
+	 * configured. All other ports (ports 1 - 5) are disabled
+	 */
+	ret = ar9331_sw_mbus_init(priv);
+	if (ret)
+		return ret;
+
+	/* Do not drop broadcast frames */
+	ret = regmap_write_bits(regmap, AR9331_SW_REG_FLOOD_MASK,
+				AR9331_SW_FLOOD_MASK_BROAD_TO_CPU,
+				AR9331_SW_FLOOD_MASK_BROAD_TO_CPU);
+	if (ret)
+		goto error;
+
+	/* Set max frame size to the maximum supported value */
+	ret = regmap_write_bits(regmap, AR9331_SW_REG_GLOBAL_CTRL,
+				AR9331_SW_GLOBAL_CTRL_MFS_M,
+				AR9331_SW_GLOBAL_CTRL_MFS_M);
+	if (ret)
+		goto error;
+
+	return 0;
+error:
+	dev_err_ratelimited(priv->dev, "%s: %i\n", __func__, ret);
+	return ret;
+}
+
+static void ar9331_sw_port_disable(struct dsa_switch *ds, int port)
+{
+	struct ar9331_sw_priv *priv = (struct ar9331_sw_priv *)ds->priv;
+	struct regmap *regmap = priv->regmap;
+	int ret;
+
+	ret = regmap_write(regmap, AR9331_SW_REG_PORT_STATUS(port), 0);
+	if (ret)
+		dev_err_ratelimited(priv->dev, "%s: %i\n", __func__, ret);
+}
+
+static enum dsa_tag_protocol ar9331_sw_get_tag_protocol(struct dsa_switch *ds,
+							int port)
+{
+	return DSA_TAG_PROTO_AR9331;
+}
+
+static void ar9331_sw_phylink_validate(struct dsa_switch *ds, int port,
+				       unsigned long *supported,
+				       struct phylink_link_state *state)
+{
+	__ETHTOOL_DECLARE_LINK_MODE_MASK(mask) = { 0, };
+
+	switch (port) {
+	case 0:
+		if (state->interface != PHY_INTERFACE_MODE_GMII)
+			goto unsupported;
+
+		phylink_set(mask, 1000baseT_Full);
+		phylink_set(mask, 1000baseT_Half);
+		break;
+	case 1:
+	case 2:
+	case 3:
+	case 4:
+	case 5:
+		if (state->interface != PHY_INTERFACE_MODE_INTERNAL)
+			goto unsupported;
+		break;
+	default:
+		bitmap_zero(supported, __ETHTOOL_LINK_MODE_MASK_NBITS);
+		dev_err(ds->dev, "Unsupported port: %i\n", port);
+		return;
+	}
+
+	phylink_set_port_modes(mask);
+	phylink_set(mask, Pause);
+	phylink_set(mask, Asym_Pause);
+
+	phylink_set(mask, 10baseT_Half);
+	phylink_set(mask, 10baseT_Full);
+	phylink_set(mask, 100baseT_Half);
+	phylink_set(mask, 100baseT_Full);
+
+	bitmap_and(supported, supported, mask,
+		   __ETHTOOL_LINK_MODE_MASK_NBITS);
+	bitmap_and(state->advertising, state->advertising, mask,
+		   __ETHTOOL_LINK_MODE_MASK_NBITS);
+
+	return;
+
+unsupported:
+	bitmap_zero(supported, __ETHTOOL_LINK_MODE_MASK_NBITS);
+	dev_err(ds->dev, "Unsupported interface: %d, port: %d\n",
+		state->interface, port);
+}
+
+static void ar9331_sw_phylink_mac_config(struct dsa_switch *ds, int port,
+					 unsigned int mode,
+					 const struct phylink_link_state *state)
+{
+	struct ar9331_sw_priv *priv = (struct ar9331_sw_priv *)ds->priv;
+	struct regmap *regmap = priv->regmap;
+	int ret;
+	u32 val;
+
+	switch (state->speed) {
+	case SPEED_1000:
+		val = AR9331_SW_PORT_STATUS_SPEED_1000;
+		break;
+	case SPEED_100:
+		val = AR9331_SW_PORT_STATUS_SPEED_100;
+		break;
+	case SPEED_10:
+		val = AR9331_SW_PORT_STATUS_SPEED_10;
+		break;
+	default:
+		return;
+	}
+
+	if (state->duplex)
+		val |= AR9331_SW_PORT_STATUS_DUPLEX_MODE;
+
+	if (state->pause & MLO_PAUSE_TX)
+		val |= AR9331_SW_PORT_STATUS_TX_FLOW_EN;
+
+	if (state->pause & MLO_PAUSE_RX)
+		val |= AR9331_SW_PORT_STATUS_RX_FLOW_EN;
+
+	ret = regmap_update_bits(regmap, AR9331_SW_REG_PORT_STATUS(port),
+				 AR9331_SW_PORT_STATUS_LINK_MASK, val);
+	if (ret)
+		dev_err_ratelimited(priv->dev, "%s: %i\n", __func__, ret);
+}
+
+static void ar9331_sw_phylink_mac_link_down(struct dsa_switch *ds, int port,
+					    unsigned int mode,
+					    phy_interface_t interface)
+{
+	struct ar9331_sw_priv *priv = (struct ar9331_sw_priv *)ds->priv;
+	struct regmap *regmap = priv->regmap;
+	int ret;
+
+	ret = regmap_update_bits(regmap, AR9331_SW_REG_PORT_STATUS(port),
+				 AR9331_SW_PORT_STATUS_MAC_MASK, 0);
+	if (ret)
+		dev_err_ratelimited(priv->dev, "%s: %i\n", __func__, ret);
+}
+
+static void ar9331_sw_phylink_mac_link_up(struct dsa_switch *ds, int port,
+					  unsigned int mode,
+					  phy_interface_t interface,
+					  struct phy_device *phydev)
+{
+	struct ar9331_sw_priv *priv = (struct ar9331_sw_priv *)ds->priv;
+	struct regmap *regmap = priv->regmap;
+	int ret;
+
+	ret = regmap_update_bits(regmap, AR9331_SW_REG_PORT_STATUS(port),
+				 AR9331_SW_PORT_STATUS_MAC_MASK,
+				 AR9331_SW_PORT_STATUS_MAC_MASK);
+	if (ret)
+		dev_err_ratelimited(priv->dev, "%s: %i\n", __func__, ret);
+}
+
+static const struct dsa_switch_ops ar9331_sw_ops = {
+	.get_tag_protocol	= ar9331_sw_get_tag_protocol,
+	.setup			= ar9331_sw_setup,
+	.port_disable		= ar9331_sw_port_disable,
+	.phylink_validate	= ar9331_sw_phylink_validate,
+	.phylink_mac_config	= ar9331_sw_phylink_mac_config,
+	.phylink_mac_link_down	= ar9331_sw_phylink_mac_link_down,
+	.phylink_mac_link_up	= ar9331_sw_phylink_mac_link_up,
+};
+
+static irqreturn_t ar9331_sw_irq(int irq, void *data)
+{
+	struct ar9331_sw_priv *priv = data;
+	struct regmap *regmap = priv->regmap;
+	u32 stat;
+	int ret;
+
+	ret = regmap_read(regmap, AR9331_SW_REG_GINT, &stat);
+	if (ret) {
+		dev_err(priv->dev, "can't read interrupt status\n");
+		return IRQ_NONE;
+	}
+
+	if (!stat)
+		return IRQ_NONE;
+
+	if (stat & AR9331_SW_GINT_PHY_INT) {
+		int child_irq;
+
+		child_irq = irq_find_mapping(priv->irqdomain, 0);
+		handle_nested_irq(child_irq);
+	}
+
+	ret = regmap_write(regmap, AR9331_SW_REG_GINT, stat);
+	if (ret) {
+		dev_err(priv->dev, "can't write interrupt status\n");
+		return IRQ_NONE;
+	}
+
+	return IRQ_HANDLED;
+}
+
+static void ar9331_sw_mask_irq(struct irq_data *d)
+{
+	struct ar9331_sw_priv *priv = irq_data_get_irq_chip_data(d);
+	struct regmap *regmap = priv->regmap;
+	int ret;
+
+	ret = regmap_update_bits(regmap, AR9331_SW_REG_GINT_MASK,
+				 AR9331_SW_GINT_PHY_INT, 0);
+	if (ret)
+		dev_err(priv->dev, "could not mask IRQ\n");
+}
+
+static void ar9331_sw_unmask_irq(struct irq_data *d)
+{
+	struct ar9331_sw_priv *priv = irq_data_get_irq_chip_data(d);
+	struct regmap *regmap = priv->regmap;
+	int ret;
+
+	ret = regmap_update_bits(regmap, AR9331_SW_REG_GINT_MASK,
+				 AR9331_SW_GINT_PHY_INT,
+				 AR9331_SW_GINT_PHY_INT);
+	if (ret)
+		dev_err(priv->dev, "could not unmask IRQ\n");
+}
+
+static struct irq_chip ar9331_sw_irq_chip = {
+	.name = AR9331_SW_NAME,
+	.irq_mask = ar9331_sw_mask_irq,
+	.irq_unmask = ar9331_sw_unmask_irq,
+};
+
+static int ar9331_sw_irq_map(struct irq_domain *domain, unsigned int irq,
+			     irq_hw_number_t hwirq)
+{
+	irq_set_chip_data(irq, domain->host_data);
+	irq_set_chip_and_handler(irq, &ar9331_sw_irq_chip, handle_simple_irq);
+	irq_set_nested_thread(irq, 1);
+	irq_set_noprobe(irq);
+
+	return 0;
+}
+
+static void ar9331_sw_irq_unmap(struct irq_domain *d, unsigned int irq)
+{
+	irq_set_nested_thread(irq, 0);
+	irq_set_chip_and_handler(irq, NULL, NULL);
+	irq_set_chip_data(irq, NULL);
+}
+
+static const struct irq_domain_ops ar9331_sw_irqdomain_ops = {
+	.map = ar9331_sw_irq_map,
+	.unmap = ar9331_sw_irq_unmap,
+	.xlate = irq_domain_xlate_onecell,
+};
+
+static int ar9331_sw_irq_init(struct ar9331_sw_priv *priv)
+{
+	struct device_node *np = priv->dev->of_node;
+	struct device *dev = priv->dev;
+	int ret, irq;
+
+	irq = of_irq_get(np, 0);
+	if (irq <= 0) {
+		dev_err(dev, "failed to get parent IRQ\n");
+		return irq ? irq : -EINVAL;
+	}
+
+	ret = devm_request_threaded_irq(dev, irq, NULL, ar9331_sw_irq,
+					IRQF_ONESHOT, AR9331_SW_NAME, priv);
+	if (ret) {
+		dev_err(dev, "unable to request irq: %d\n", ret);
+		return ret;
+	}
+
+	priv->irqdomain = irq_domain_add_linear(np, 1, &ar9331_sw_irqdomain_ops,
+						priv);
+	if (!priv->irqdomain) {
+		dev_err(dev, "failed to create IRQ domain\n");
+		return -EINVAL;
+	}
+
+	irq_set_parent(irq_create_mapping(priv->irqdomain, 0), irq);
+
+	return 0;
+}
+
+static int __ar9331_mdio_write(struct mii_bus *sbus, u8 mode, u16 reg, u16 val)
+{
+	u8 r, p;
+
+	p = FIELD_PREP(AR9331_SW_MDIO_PHY_MODE_M, mode) |
+		FIELD_GET(AR9331_SW_LOW_ADDR_PHY, reg);
+	r = FIELD_GET(AR9331_SW_LOW_ADDR_REG, reg);
+
+	return mdiobus_write(sbus, p, r, val);
+}
+
+static int __ar9331_mdio_read(struct mii_bus *sbus, u16 reg)
+{
+	u8 r, p;
+
+	p = FIELD_PREP(AR9331_SW_MDIO_PHY_MODE_M, AR9331_SW_MDIO_PHY_MODE_REG) |
+		FIELD_GET(AR9331_SW_LOW_ADDR_PHY, reg);
+	r = FIELD_GET(AR9331_SW_LOW_ADDR_REG, reg);
+
+	return mdiobus_read(sbus, p, r);
+}
+
+static int ar9331_mdio_read(void *ctx, const void *reg_buf, size_t reg_len,
+			    void *val_buf, size_t val_len)
+{
+	struct ar9331_sw_priv *priv = ctx;
+	struct mii_bus *sbus = priv->sbus;
+	u32 reg = *(u32 *)reg_buf;
+	int ret;
+
+	if (reg == AR9331_SW_REG_PAGE) {
+		/* We cannot read the page selector register from hardware and
+		 * we cache its value in regmap. Return all bits set here,
+		 * that regmap will always write the page on first use.
+		 */
+		*(u32 *)val_buf = GENMASK(9, 0);
+		return 0;
+	}
+
+	ret = __ar9331_mdio_read(sbus, reg);
+	if (ret < 0)
+		goto error;
+
+	*(u32 *)val_buf = ret;
+	ret = __ar9331_mdio_read(sbus, reg + 2);
+	if (ret < 0)
+		goto error;
+
+	*(u32 *)val_buf |= ret << 16;
+
+	return 0;
+error:
+	dev_err_ratelimited(&sbus->dev, "Bus error. Failed to read register.\n");
+	return ret;
+}
+
+static int ar9331_mdio_write(void *ctx, u32 reg, u32 val)
+{
+	struct ar9331_sw_priv *priv = (struct ar9331_sw_priv *)ctx;
+	struct mii_bus *sbus = priv->sbus;
+	int ret;
+
+	if (reg == AR9331_SW_REG_PAGE) {
+		ret = __ar9331_mdio_write(sbus, AR9331_SW_MDIO_PHY_MODE_PAGE,
+					  0, val);
+		if (ret < 0)
+			goto error;
+
+		return 0;
+	}
+
+	ret = __ar9331_mdio_write(sbus, AR9331_SW_MDIO_PHY_MODE_REG, reg, val);
+	if (ret < 0)
+		goto error;
+
+	ret = __ar9331_mdio_write(sbus, AR9331_SW_MDIO_PHY_MODE_REG, reg + 2,
+				  val >> 16);
+	if (ret < 0)
+		goto error;
+
+	return 0;
+error:
+	dev_err_ratelimited(&sbus->dev, "Bus error. Failed to write register.\n");
+	return ret;
+}
+
+static int ar9331_sw_bus_write(void *context, const void *data, size_t count)
+{
+	u32 reg = *(u32 *)data;
+	u32 val = *((u32 *)data + 1);
+
+	return ar9331_mdio_write(context, reg, val);
+}
+
+static const struct regmap_range ar9331_valid_regs[] = {
+	regmap_reg_range(0x0, 0x0),
+	regmap_reg_range(0x10, 0x14),
+	regmap_reg_range(0x20, 0x24),
+	regmap_reg_range(0x2c, 0x30),
+	regmap_reg_range(0x40, 0x44),
+	regmap_reg_range(0x50, 0x78),
+	regmap_reg_range(0x80, 0x98),
+
+	regmap_reg_range(0x100, 0x120),
+	regmap_reg_range(0x200, 0x220),
+	regmap_reg_range(0x300, 0x320),
+	regmap_reg_range(0x400, 0x420),
+	regmap_reg_range(0x500, 0x520),
+	regmap_reg_range(0x600, 0x620),
+
+	regmap_reg_range(0x20000, 0x200a4),
+	regmap_reg_range(0x20100, 0x201a4),
+	regmap_reg_range(0x20200, 0x202a4),
+	regmap_reg_range(0x20300, 0x203a4),
+	regmap_reg_range(0x20400, 0x204a4),
+	regmap_reg_range(0x20500, 0x205a4),
+
+	/* dummy page selector reg */
+	regmap_reg_range(AR9331_SW_REG_PAGE, AR9331_SW_REG_PAGE),
+};
+
+static const struct regmap_range ar9331_nonvolatile_regs[] = {
+	regmap_reg_range(AR9331_SW_REG_PAGE, AR9331_SW_REG_PAGE),
+};
+
+static const struct regmap_range_cfg ar9331_regmap_range[] = {
+	{
+		.selector_reg = AR9331_SW_REG_PAGE,
+		.selector_mask = GENMASK(9, 0),
+		.selector_shift = 0,
+
+		.window_start = 0,
+		.window_len = 512,
+
+		.range_min = 0,
+		.range_max = AR9331_SW_REG_PAGE - 4,
+	},
+};
+
+static const struct regmap_access_table ar9331_register_set = {
+	.yes_ranges = ar9331_valid_regs,
+	.n_yes_ranges = ARRAY_SIZE(ar9331_valid_regs),
+};
+
+static const struct regmap_access_table ar9331_volatile_set = {
+	.no_ranges = ar9331_nonvolatile_regs,
+	.n_no_ranges = ARRAY_SIZE(ar9331_nonvolatile_regs),
+};
+
+static const struct regmap_config ar9331_mdio_regmap_config = {
+	.reg_bits = 32,
+	.val_bits = 32,
+	.reg_stride = 4,
+	.max_register = AR9331_SW_REG_PAGE,
+
+	.ranges = ar9331_regmap_range,
+	.num_ranges = ARRAY_SIZE(ar9331_regmap_range),
+
+	.volatile_table = &ar9331_volatile_set,
+	.wr_table = &ar9331_register_set,
+	.rd_table = &ar9331_register_set,
+
+	.cache_type = REGCACHE_RBTREE,
+};
+
+static struct regmap_bus ar9331_sw_bus = {
+	.reg_format_endian_default = REGMAP_ENDIAN_NATIVE,
+	.val_format_endian_default = REGMAP_ENDIAN_NATIVE,
+	.read = ar9331_mdio_read,
+	.write = ar9331_sw_bus_write,
+	.max_raw_read = 4,
+	.max_raw_write = 4,
+};
+
+static int ar9331_sw_probe(struct mdio_device *mdiodev)
+{
+	struct ar9331_sw_priv *priv;
+	struct dsa_switch *ds;
+	int ret;
+
+	priv = devm_kzalloc(&mdiodev->dev, sizeof(*priv), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+
+	priv->regmap = devm_regmap_init(&mdiodev->dev, &ar9331_sw_bus, priv,
+					&ar9331_mdio_regmap_config);
+	if (IS_ERR(priv->regmap)) {
+		ret = PTR_ERR(priv->regmap);
+		dev_err(&mdiodev->dev, "regmap init failed: %d\n", ret);
+		return ret;
+	}
+
+	priv->sw_reset = devm_reset_control_get(&mdiodev->dev, "switch");
+	if (IS_ERR(priv->sw_reset)) {
+		dev_err(&mdiodev->dev, "missing switch reset\n");
+		return PTR_ERR(priv->sw_reset);
+	}
+
+	priv->sbus = mdiodev->bus;
+	priv->dev = &mdiodev->dev;
+
+	ret = ar9331_sw_irq_init(priv);
+	if (ret)
+		return ret;
+
+	ds = &priv->ds;
+	ds->dev = &mdiodev->dev;
+	ds->num_ports = AR9331_SW_PORTS;
+	ds->priv = priv;
+	priv->ops = ar9331_sw_ops;
+	ds->ops = &priv->ops;
+	dev_set_drvdata(&mdiodev->dev, priv);
+
+	ret = dsa_register_switch(ds);
+	if (ret)
+		goto err_remove_irq;
+
+	return 0;
+
+err_remove_irq:
+	irq_domain_remove(priv->irqdomain);
+
+	return ret;
+}
+
+static void ar9331_sw_remove(struct mdio_device *mdiodev)
+{
+	struct ar9331_sw_priv *priv = dev_get_drvdata(&mdiodev->dev);
+
+	irq_domain_remove(priv->irqdomain);
+	mdiobus_unregister(priv->mbus);
+	dsa_unregister_switch(&priv->ds);
+
+	reset_control_assert(priv->sw_reset);
+}
+
+static const struct of_device_id ar9331_sw_of_match[] = {
+	{ .compatible = "qca,ar9331-switch" },
+	{ },
+};
+
+static struct mdio_driver ar9331_sw_mdio_driver = {
+	.probe = ar9331_sw_probe,
+	.remove = ar9331_sw_remove,
+	.mdiodrv.driver = {
+		.name = AR9331_SW_NAME,
+		.of_match_table = ar9331_sw_of_match,
+	},
+};
+
+mdio_module_driver(ar9331_sw_mdio_driver);
+
+MODULE_AUTHOR("Oleksij Rempel <kernel@pengutronix.de>");
+MODULE_DESCRIPTION("Driver for Atheros AR9331 switch");
+MODULE_LICENSE("GPL v2");
-- 
cgit v1.2.3-59-g8ed1b


From 43985468df6757bc079ca0be44edf4da997891e0 Mon Sep 17 00:00:00 2001
From: Stefan Hajnoczi <stefanha@redhat.com>
Date: Wed, 18 Dec 2019 19:06:58 +0100
Subject: VSOCK: fix header include in vsock_diag_test

The vsock_diag_test program directly included ../../../include/uapi/
headers from the source tree.  Tests are supposed to use the
usr/include/linux/ headers that have been prepared with make
headers_install instead.

Suggested-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/vsock/Makefile          | 2 +-
 tools/testing/vsock/README            | 2 +-
 tools/testing/vsock/vsock_diag_test.c | 5 ++---
 3 files changed, 4 insertions(+), 5 deletions(-)

diff --git a/tools/testing/vsock/Makefile b/tools/testing/vsock/Makefile
index 5be687b1e16c..d41a4e13960a 100644
--- a/tools/testing/vsock/Makefile
+++ b/tools/testing/vsock/Makefile
@@ -3,7 +3,7 @@ all: test
 test: vsock_diag_test
 vsock_diag_test: vsock_diag_test.o timeout.o control.o
 
-CFLAGS += -g -O2 -Werror -Wall -I. -I../../include/uapi -I../../include -Wno-pointer-sign -fno-strict-overflow -fno-strict-aliasing -fno-common -MMD -U_FORTIFY_SOURCE -D_GNU_SOURCE
+CFLAGS += -g -O2 -Werror -Wall -I. -I../../include -I../../../usr/include -Wno-pointer-sign -fno-strict-overflow -fno-strict-aliasing -fno-common -MMD -U_FORTIFY_SOURCE -D_GNU_SOURCE
 .PHONY: all test clean
 clean:
 	${RM} *.o *.d vsock_diag_test
diff --git a/tools/testing/vsock/README b/tools/testing/vsock/README
index 2cc6d7302db6..cf7dc64273bf 100644
--- a/tools/testing/vsock/README
+++ b/tools/testing/vsock/README
@@ -10,7 +10,7 @@ The following tests are available:
 The following prerequisite steps are not automated and must be performed prior
 to running tests:
 
-1. Build the kernel and these tests.
+1. Build the kernel, make headers_install, and build these tests.
 2. Install the kernel and tests on the host.
 3. Install the kernel and tests inside the guest.
 4. Boot the guest and ensure that the AF_VSOCK transport is enabled.
diff --git a/tools/testing/vsock/vsock_diag_test.c b/tools/testing/vsock/vsock_diag_test.c
index c481101364a4..fc391e041954 100644
--- a/tools/testing/vsock/vsock_diag_test.c
+++ b/tools/testing/vsock/vsock_diag_test.c
@@ -21,12 +21,11 @@
 #include <linux/list.h>
 #include <linux/net.h>
 #include <linux/netlink.h>
+#include <linux/vm_sockets.h>
 #include <linux/sock_diag.h>
+#include <linux/vm_sockets_diag.h>
 #include <netinet/tcp.h>
 
-#include "../../../include/uapi/linux/vm_sockets.h"
-#include "../../../include/uapi/linux/vm_sockets_diag.h"
-
 #include "timeout.h"
 #include "control.h"
 
-- 
cgit v1.2.3-59-g8ed1b


From ce1c7f84e276e8610ffa582517202ff76374cbc7 Mon Sep 17 00:00:00 2001
From: Stefan Hajnoczi <stefanha@redhat.com>
Date: Wed, 18 Dec 2019 19:06:59 +0100
Subject: VSOCK: add SPDX identifiers to vsock tests

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/vsock/control.h | 1 +
 tools/testing/vsock/timeout.h | 1 +
 2 files changed, 2 insertions(+)

diff --git a/tools/testing/vsock/control.h b/tools/testing/vsock/control.h
index 54a07efd267c..dac3964a891d 100644
--- a/tools/testing/vsock/control.h
+++ b/tools/testing/vsock/control.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 #ifndef CONTROL_H
 #define CONTROL_H
 
diff --git a/tools/testing/vsock/timeout.h b/tools/testing/vsock/timeout.h
index 77db9ce9860a..ecb7c840e65a 100644
--- a/tools/testing/vsock/timeout.h
+++ b/tools/testing/vsock/timeout.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
 #ifndef TIMEOUT_H
 #define TIMEOUT_H
 
-- 
cgit v1.2.3-59-g8ed1b


From df7e0e0d237e5a4ee262befb11b599058a178966 Mon Sep 17 00:00:00 2001
From: Stefan Hajnoczi <stefanha@redhat.com>
Date: Wed, 18 Dec 2019 19:07:00 +0100
Subject: VSOCK: extract utility functions from vsock_diag_test.c

Move useful functions into a separate file in preparation for more
vsock test programs.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/vsock/Makefile          |  2 +-
 tools/testing/vsock/util.c            | 66 +++++++++++++++++++++++++
 tools/testing/vsock/util.h            | 36 ++++++++++++++
 tools/testing/vsock/vsock_diag_test.c | 92 +++++++++--------------------------
 4 files changed, 125 insertions(+), 71 deletions(-)
 create mode 100644 tools/testing/vsock/util.c
 create mode 100644 tools/testing/vsock/util.h

diff --git a/tools/testing/vsock/Makefile b/tools/testing/vsock/Makefile
index d41a4e13960a..a916878a2d8c 100644
--- a/tools/testing/vsock/Makefile
+++ b/tools/testing/vsock/Makefile
@@ -1,7 +1,7 @@
 # SPDX-License-Identifier: GPL-2.0-only
 all: test
 test: vsock_diag_test
-vsock_diag_test: vsock_diag_test.o timeout.o control.o
+vsock_diag_test: vsock_diag_test.o timeout.o control.o util.o
 
 CFLAGS += -g -O2 -Werror -Wall -I. -I../../include -I../../../usr/include -Wno-pointer-sign -fno-strict-overflow -fno-strict-aliasing -fno-common -MMD -U_FORTIFY_SOURCE -D_GNU_SOURCE
 .PHONY: all test clean
diff --git a/tools/testing/vsock/util.c b/tools/testing/vsock/util.c
new file mode 100644
index 000000000000..f40f45b36d2f
--- /dev/null
+++ b/tools/testing/vsock/util.c
@@ -0,0 +1,66 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * vsock test utilities
+ *
+ * Copyright (C) 2017 Red Hat, Inc.
+ *
+ * Author: Stefan Hajnoczi <stefanha@redhat.com>
+ */
+
+#include <errno.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <signal.h>
+
+#include "timeout.h"
+#include "util.h"
+
+/* Install signal handlers */
+void init_signals(void)
+{
+	struct sigaction act = {
+		.sa_handler = sigalrm,
+	};
+
+	sigaction(SIGALRM, &act, NULL);
+	signal(SIGPIPE, SIG_IGN);
+}
+
+/* Parse a CID in string representation */
+unsigned int parse_cid(const char *str)
+{
+	char *endptr = NULL;
+	unsigned long n;
+
+	errno = 0;
+	n = strtoul(str, &endptr, 10);
+	if (errno || *endptr != '\0') {
+		fprintf(stderr, "malformed CID \"%s\"\n", str);
+		exit(EXIT_FAILURE);
+	}
+	return n;
+}
+
+/* Run test cases.  The program terminates if a failure occurs. */
+void run_tests(const struct test_case *test_cases,
+	       const struct test_opts *opts)
+{
+	int i;
+
+	for (i = 0; test_cases[i].name; i++) {
+		void (*run)(const struct test_opts *opts);
+
+		printf("%s...", test_cases[i].name);
+		fflush(stdout);
+
+		if (opts->mode == TEST_MODE_CLIENT)
+			run = test_cases[i].run_client;
+		else
+			run = test_cases[i].run_server;
+
+		if (run)
+			run(opts);
+
+		printf("ok\n");
+	}
+}
diff --git a/tools/testing/vsock/util.h b/tools/testing/vsock/util.h
new file mode 100644
index 000000000000..033e7d59a42a
--- /dev/null
+++ b/tools/testing/vsock/util.h
@@ -0,0 +1,36 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+#ifndef UTIL_H
+#define UTIL_H
+
+/* Tests can either run as the client or the server */
+enum test_mode {
+	TEST_MODE_UNSET,
+	TEST_MODE_CLIENT,
+	TEST_MODE_SERVER
+};
+
+/* Test runner options */
+struct test_opts {
+	enum test_mode mode;
+	unsigned int peer_cid;
+};
+
+/* A test case definition.  Test functions must print failures to stderr and
+ * terminate with exit(EXIT_FAILURE).
+ */
+struct test_case {
+	const char *name; /* human-readable name */
+
+	/* Called when test mode is TEST_MODE_CLIENT */
+	void (*run_client)(const struct test_opts *opts);
+
+	/* Called when test mode is TEST_MODE_SERVER */
+	void (*run_server)(const struct test_opts *opts);
+};
+
+void init_signals(void);
+unsigned int parse_cid(const char *str);
+void run_tests(const struct test_case *test_cases,
+	       const struct test_opts *opts);
+
+#endif /* UTIL_H */
diff --git a/tools/testing/vsock/vsock_diag_test.c b/tools/testing/vsock/vsock_diag_test.c
index fc391e041954..944c8a72eed7 100644
--- a/tools/testing/vsock/vsock_diag_test.c
+++ b/tools/testing/vsock/vsock_diag_test.c
@@ -9,12 +9,10 @@
 
 #include <getopt.h>
 #include <stdio.h>
-#include <stdbool.h>
 #include <stdlib.h>
 #include <string.h>
 #include <errno.h>
 #include <unistd.h>
-#include <signal.h>
 #include <sys/socket.h>
 #include <sys/stat.h>
 #include <sys/types.h>
@@ -28,12 +26,7 @@
 
 #include "timeout.h"
 #include "control.h"
-
-enum test_mode {
-	TEST_MODE_UNSET,
-	TEST_MODE_CLIENT,
-	TEST_MODE_SERVER
-};
+#include "util.h"
 
 /* Per-socket status */
 struct vsock_stat {
@@ -334,7 +327,7 @@ static void free_sock_stat(struct list_head *sockets)
 		free(st);
 }
 
-static void test_no_sockets(unsigned int peer_cid)
+static void test_no_sockets(const struct test_opts *opts)
 {
 	LIST_HEAD(sockets);
 
@@ -345,7 +338,7 @@ static void test_no_sockets(unsigned int peer_cid)
 	free_sock_stat(&sockets);
 }
 
-static void test_listen_socket_server(unsigned int peer_cid)
+static void test_listen_socket_server(const struct test_opts *opts)
 {
 	union {
 		struct sockaddr sa;
@@ -383,7 +376,7 @@ static void test_listen_socket_server(unsigned int peer_cid)
 	free_sock_stat(&sockets);
 }
 
-static void test_connect_client(unsigned int peer_cid)
+static void test_connect_client(const struct test_opts *opts)
 {
 	union {
 		struct sockaddr sa;
@@ -392,7 +385,7 @@ static void test_connect_client(unsigned int peer_cid)
 		.svm = {
 			.svm_family = AF_VSOCK,
 			.svm_port = 1234,
-			.svm_cid = peer_cid,
+			.svm_cid = opts->peer_cid,
 		},
 	};
 	int fd;
@@ -429,7 +422,7 @@ static void test_connect_client(unsigned int peer_cid)
 	free_sock_stat(&sockets);
 }
 
-static void test_connect_server(unsigned int peer_cid)
+static void test_connect_server(const struct test_opts *opts)
 {
 	union {
 		struct sockaddr sa;
@@ -481,9 +474,9 @@ static void test_connect_server(unsigned int peer_cid)
 			clientaddr.sa.sa_family);
 		exit(EXIT_FAILURE);
 	}
-	if (clientaddr.svm.svm_cid != peer_cid) {
+	if (clientaddr.svm.svm_cid != opts->peer_cid) {
 		fprintf(stderr, "expected peer CID %u from accept(2), got %u\n",
-			peer_cid, clientaddr.svm.svm_cid);
+			opts->peer_cid, clientaddr.svm.svm_cid);
 		exit(EXIT_FAILURE);
 	}
 
@@ -502,11 +495,7 @@ static void test_connect_server(unsigned int peer_cid)
 	free_sock_stat(&sockets);
 }
 
-static struct {
-	const char *name;
-	void (*run_client)(unsigned int peer_cid);
-	void (*run_server)(unsigned int peer_cid);
-} test_cases[] = {
+static struct test_case test_cases[] = {
 	{
 		.name = "No sockets",
 		.run_server = test_no_sockets,
@@ -523,30 +512,6 @@ static struct {
 	{},
 };
 
-static void init_signals(void)
-{
-	struct sigaction act = {
-		.sa_handler = sigalrm,
-	};
-
-	sigaction(SIGALRM, &act, NULL);
-	signal(SIGPIPE, SIG_IGN);
-}
-
-static unsigned int parse_cid(const char *str)
-{
-	char *endptr = NULL;
-	unsigned long int n;
-
-	errno = 0;
-	n = strtoul(str, &endptr, 10);
-	if (errno || *endptr != '\0') {
-		fprintf(stderr, "malformed CID \"%s\"\n", str);
-		exit(EXIT_FAILURE);
-	}
-	return n;
-}
-
 static const char optstring[] = "";
 static const struct option longopts[] = {
 	{
@@ -601,9 +566,10 @@ int main(int argc, char **argv)
 {
 	const char *control_host = NULL;
 	const char *control_port = NULL;
-	int mode = TEST_MODE_UNSET;
-	unsigned int peer_cid = VMADDR_CID_ANY;
-	int i;
+	struct test_opts opts = {
+		.mode = TEST_MODE_UNSET,
+		.peer_cid = VMADDR_CID_ANY,
+	};
 
 	init_signals();
 
@@ -619,16 +585,16 @@ int main(int argc, char **argv)
 			break;
 		case 'm':
 			if (strcmp(optarg, "client") == 0)
-				mode = TEST_MODE_CLIENT;
+				opts.mode = TEST_MODE_CLIENT;
 			else if (strcmp(optarg, "server") == 0)
-				mode = TEST_MODE_SERVER;
+				opts.mode = TEST_MODE_SERVER;
 			else {
 				fprintf(stderr, "--mode must be \"client\" or \"server\"\n");
 				return EXIT_FAILURE;
 			}
 			break;
 		case 'p':
-			peer_cid = parse_cid(optarg);
+			opts.peer_cid = parse_cid(optarg);
 			break;
 		case 'P':
 			control_port = optarg;
@@ -641,35 +607,21 @@ int main(int argc, char **argv)
 
 	if (!control_port)
 		usage();
-	if (mode == TEST_MODE_UNSET)
+	if (opts.mode == TEST_MODE_UNSET)
 		usage();
-	if (peer_cid == VMADDR_CID_ANY)
+	if (opts.peer_cid == VMADDR_CID_ANY)
 		usage();
 
 	if (!control_host) {
-		if (mode != TEST_MODE_SERVER)
+		if (opts.mode != TEST_MODE_SERVER)
 			usage();
 		control_host = "0.0.0.0";
 	}
 
-	control_init(control_host, control_port, mode == TEST_MODE_SERVER);
-
-	for (i = 0; test_cases[i].name; i++) {
-		void (*run)(unsigned int peer_cid);
+	control_init(control_host, control_port,
+		     opts.mode == TEST_MODE_SERVER);
 
-		printf("%s...", test_cases[i].name);
-		fflush(stdout);
-
-		if (mode == TEST_MODE_CLIENT)
-			run = test_cases[i].run_client;
-		else
-			run = test_cases[i].run_server;
-
-		if (run)
-			run(peer_cid);
-
-		printf("ok\n");
-	}
+	run_tests(test_cases, &opts);
 
 	control_cleanup();
 	return EXIT_SUCCESS;
-- 
cgit v1.2.3-59-g8ed1b


From 9bb8a29d28e01bf7dcdb67f71edbca9d3826d004 Mon Sep 17 00:00:00 2001
From: Stefan Hajnoczi <stefanha@redhat.com>
Date: Wed, 18 Dec 2019 19:07:01 +0100
Subject: VSOCK: extract connect/accept functions from vsock_diag_test.c

Many test cases will need to connect to the server or accept incoming
connections.  This patch extracts these operations into utility
functions that can be reused.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/vsock/util.c            | 108 ++++++++++++++++++++++++++++++++++
 tools/testing/vsock/util.h            |   6 ++
 tools/testing/vsock/vsock_diag_test.c |  81 ++-----------------------
 3 files changed, 119 insertions(+), 76 deletions(-)

diff --git a/tools/testing/vsock/util.c b/tools/testing/vsock/util.c
index f40f45b36d2f..f838bcee3589 100644
--- a/tools/testing/vsock/util.c
+++ b/tools/testing/vsock/util.c
@@ -11,8 +11,10 @@
 #include <stdio.h>
 #include <stdlib.h>
 #include <signal.h>
+#include <unistd.h>
 
 #include "timeout.h"
+#include "control.h"
 #include "util.h"
 
 /* Install signal handlers */
@@ -41,6 +43,112 @@ unsigned int parse_cid(const char *str)
 	return n;
 }
 
+/* Connect to <cid, port> and return the file descriptor. */
+int vsock_stream_connect(unsigned int cid, unsigned int port)
+{
+	union {
+		struct sockaddr sa;
+		struct sockaddr_vm svm;
+	} addr = {
+		.svm = {
+			.svm_family = AF_VSOCK,
+			.svm_port = port,
+			.svm_cid = cid,
+		},
+	};
+	int ret;
+	int fd;
+
+	control_expectln("LISTENING");
+
+	fd = socket(AF_VSOCK, SOCK_STREAM, 0);
+
+	timeout_begin(TIMEOUT);
+	do {
+		ret = connect(fd, &addr.sa, sizeof(addr.svm));
+		timeout_check("connect");
+	} while (ret < 0 && errno == EINTR);
+	timeout_end();
+
+	if (ret < 0) {
+		int old_errno = errno;
+
+		close(fd);
+		fd = -1;
+		errno = old_errno;
+	}
+	return fd;
+}
+
+/* Listen on <cid, port> and return the first incoming connection.  The remote
+ * address is stored to clientaddrp.  clientaddrp may be NULL.
+ */
+int vsock_stream_accept(unsigned int cid, unsigned int port,
+			struct sockaddr_vm *clientaddrp)
+{
+	union {
+		struct sockaddr sa;
+		struct sockaddr_vm svm;
+	} addr = {
+		.svm = {
+			.svm_family = AF_VSOCK,
+			.svm_port = port,
+			.svm_cid = cid,
+		},
+	};
+	union {
+		struct sockaddr sa;
+		struct sockaddr_vm svm;
+	} clientaddr;
+	socklen_t clientaddr_len = sizeof(clientaddr.svm);
+	int fd;
+	int client_fd;
+	int old_errno;
+
+	fd = socket(AF_VSOCK, SOCK_STREAM, 0);
+
+	if (bind(fd, &addr.sa, sizeof(addr.svm)) < 0) {
+		perror("bind");
+		exit(EXIT_FAILURE);
+	}
+
+	if (listen(fd, 1) < 0) {
+		perror("listen");
+		exit(EXIT_FAILURE);
+	}
+
+	control_writeln("LISTENING");
+
+	timeout_begin(TIMEOUT);
+	do {
+		client_fd = accept(fd, &clientaddr.sa, &clientaddr_len);
+		timeout_check("accept");
+	} while (client_fd < 0 && errno == EINTR);
+	timeout_end();
+
+	old_errno = errno;
+	close(fd);
+	errno = old_errno;
+
+	if (client_fd < 0)
+		return client_fd;
+
+	if (clientaddr_len != sizeof(clientaddr.svm)) {
+		fprintf(stderr, "unexpected addrlen from accept(2), %zu\n",
+			(size_t)clientaddr_len);
+		exit(EXIT_FAILURE);
+	}
+	if (clientaddr.sa.sa_family != AF_VSOCK) {
+		fprintf(stderr, "expected AF_VSOCK from accept(2), got %d\n",
+			clientaddr.sa.sa_family);
+		exit(EXIT_FAILURE);
+	}
+
+	if (clientaddrp)
+		*clientaddrp = clientaddr.svm;
+	return client_fd;
+}
+
 /* Run test cases.  The program terminates if a failure occurs. */
 void run_tests(const struct test_case *test_cases,
 	       const struct test_opts *opts)
diff --git a/tools/testing/vsock/util.h b/tools/testing/vsock/util.h
index 033e7d59a42a..1786305cfddd 100644
--- a/tools/testing/vsock/util.h
+++ b/tools/testing/vsock/util.h
@@ -2,6 +2,9 @@
 #ifndef UTIL_H
 #define UTIL_H
 
+#include <sys/socket.h>
+#include <linux/vm_sockets.h>
+
 /* Tests can either run as the client or the server */
 enum test_mode {
 	TEST_MODE_UNSET,
@@ -30,6 +33,9 @@ struct test_case {
 
 void init_signals(void);
 unsigned int parse_cid(const char *str);
+int vsock_stream_connect(unsigned int cid, unsigned int port);
+int vsock_stream_accept(unsigned int cid, unsigned int port,
+			struct sockaddr_vm *clientaddrp);
 void run_tests(const struct test_case *test_cases,
 	       const struct test_opts *opts);
 
diff --git a/tools/testing/vsock/vsock_diag_test.c b/tools/testing/vsock/vsock_diag_test.c
index 944c8a72eed7..abd7dc2a9631 100644
--- a/tools/testing/vsock/vsock_diag_test.c
+++ b/tools/testing/vsock/vsock_diag_test.c
@@ -13,13 +13,11 @@
 #include <string.h>
 #include <errno.h>
 #include <unistd.h>
-#include <sys/socket.h>
 #include <sys/stat.h>
 #include <sys/types.h>
 #include <linux/list.h>
 #include <linux/net.h>
 #include <linux/netlink.h>
-#include <linux/vm_sockets.h>
 #include <linux/sock_diag.h>
 #include <linux/vm_sockets_diag.h>
 #include <netinet/tcp.h>
@@ -378,33 +376,12 @@ static void test_listen_socket_server(const struct test_opts *opts)
 
 static void test_connect_client(const struct test_opts *opts)
 {
-	union {
-		struct sockaddr sa;
-		struct sockaddr_vm svm;
-	} addr = {
-		.svm = {
-			.svm_family = AF_VSOCK,
-			.svm_port = 1234,
-			.svm_cid = opts->peer_cid,
-		},
-	};
 	int fd;
-	int ret;
 	LIST_HEAD(sockets);
 	struct vsock_stat *st;
 
-	control_expectln("LISTENING");
-
-	fd = socket(AF_VSOCK, SOCK_STREAM, 0);
-
-	timeout_begin(TIMEOUT);
-	do {
-		ret = connect(fd, &addr.sa, sizeof(addr.svm));
-		timeout_check("connect");
-	} while (ret < 0 && errno == EINTR);
-	timeout_end();
-
-	if (ret < 0) {
+	fd = vsock_stream_connect(opts->peer_cid, 1234);
+	if (fd < 0) {
 		perror("connect");
 		exit(EXIT_FAILURE);
 	}
@@ -424,66 +401,19 @@ static void test_connect_client(const struct test_opts *opts)
 
 static void test_connect_server(const struct test_opts *opts)
 {
-	union {
-		struct sockaddr sa;
-		struct sockaddr_vm svm;
-	} addr = {
-		.svm = {
-			.svm_family = AF_VSOCK,
-			.svm_port = 1234,
-			.svm_cid = VMADDR_CID_ANY,
-		},
-	};
-	union {
-		struct sockaddr sa;
-		struct sockaddr_vm svm;
-	} clientaddr;
-	socklen_t clientaddr_len = sizeof(clientaddr.svm);
-	LIST_HEAD(sockets);
 	struct vsock_stat *st;
-	int fd;
+	LIST_HEAD(sockets);
 	int client_fd;
 
-	fd = socket(AF_VSOCK, SOCK_STREAM, 0);
-
-	if (bind(fd, &addr.sa, sizeof(addr.svm)) < 0) {
-		perror("bind");
-		exit(EXIT_FAILURE);
-	}
-
-	if (listen(fd, 1) < 0) {
-		perror("listen");
-		exit(EXIT_FAILURE);
-	}
-
-	control_writeln("LISTENING");
-
-	timeout_begin(TIMEOUT);
-	do {
-		client_fd = accept(fd, &clientaddr.sa, &clientaddr_len);
-		timeout_check("accept");
-	} while (client_fd < 0 && errno == EINTR);
-	timeout_end();
-
+	client_fd = vsock_stream_accept(VMADDR_CID_ANY, 1234, NULL);
 	if (client_fd < 0) {
 		perror("accept");
 		exit(EXIT_FAILURE);
 	}
-	if (clientaddr.sa.sa_family != AF_VSOCK) {
-		fprintf(stderr, "expected AF_VSOCK from accept(2), got %d\n",
-			clientaddr.sa.sa_family);
-		exit(EXIT_FAILURE);
-	}
-	if (clientaddr.svm.svm_cid != opts->peer_cid) {
-		fprintf(stderr, "expected peer CID %u from accept(2), got %u\n",
-			opts->peer_cid, clientaddr.svm.svm_cid);
-		exit(EXIT_FAILURE);
-	}
 
 	read_vsock_stat(&sockets);
 
-	check_num_sockets(&sockets, 2);
-	find_vsock_stat(&sockets, fd);
+	check_num_sockets(&sockets, 1);
 	st = find_vsock_stat(&sockets, client_fd);
 	check_socket_state(st, TCP_ESTABLISHED);
 
@@ -491,7 +421,6 @@ static void test_connect_server(const struct test_opts *opts)
 	control_expectln("DONE");
 
 	close(client_fd);
-	close(fd);
 	free_sock_stat(&sockets);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 2f65b44e199c6521be8dbc7458250c74a86ce2aa Mon Sep 17 00:00:00 2001
From: Stefan Hajnoczi <stefanha@redhat.com>
Date: Wed, 18 Dec 2019 19:07:02 +0100
Subject: VSOCK: add full barrier between test cases

See code comment for details.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/vsock/util.c | 18 ++++++++++++++++--
 1 file changed, 16 insertions(+), 2 deletions(-)

diff --git a/tools/testing/vsock/util.c b/tools/testing/vsock/util.c
index f838bcee3589..4280a56ba677 100644
--- a/tools/testing/vsock/util.c
+++ b/tools/testing/vsock/util.c
@@ -161,10 +161,24 @@ void run_tests(const struct test_case *test_cases,
 		printf("%s...", test_cases[i].name);
 		fflush(stdout);
 
-		if (opts->mode == TEST_MODE_CLIENT)
+		if (opts->mode == TEST_MODE_CLIENT) {
+			/* Full barrier before executing the next test.  This
+			 * ensures that client and server are executing the
+			 * same test case.  In particular, it means whoever is
+			 * faster will not see the peer still executing the
+			 * last test.  This is important because port numbers
+			 * can be used by multiple test cases.
+			 */
+			control_expectln("NEXT");
+			control_writeln("NEXT");
+
 			run = test_cases[i].run_client;
-		else
+		} else {
+			control_writeln("NEXT");
+			control_expectln("NEXT");
+
 			run = test_cases[i].run_server;
+		}
 
 		if (run)
 			run(opts);
-- 
cgit v1.2.3-59-g8ed1b


From 092f32ae628ed4874caf349cc0cfa0626cc3c5b0 Mon Sep 17 00:00:00 2001
From: Stefan Hajnoczi <stefanha@redhat.com>
Date: Wed, 18 Dec 2019 19:07:03 +0100
Subject: VSOCK: add send_byte()/recv_byte() test utilities

Test cases will want to transfer data.  This patch adds utility
functions to do this.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/vsock/util.c | 103 +++++++++++++++++++++++++++++++++++++++++++++
 tools/testing/vsock/util.h |   2 +
 2 files changed, 105 insertions(+)

diff --git a/tools/testing/vsock/util.c b/tools/testing/vsock/util.c
index 4280a56ba677..6026ef3ce512 100644
--- a/tools/testing/vsock/util.c
+++ b/tools/testing/vsock/util.c
@@ -9,6 +9,7 @@
 
 #include <errno.h>
 #include <stdio.h>
+#include <stdint.h>
 #include <stdlib.h>
 #include <signal.h>
 #include <unistd.h>
@@ -149,6 +150,108 @@ int vsock_stream_accept(unsigned int cid, unsigned int port,
 	return client_fd;
 }
 
+/* Transmit one byte and check the return value.
+ *
+ * expected_ret:
+ *  <0 Negative errno (for testing errors)
+ *   0 End-of-file
+ *   1 Success
+ */
+void send_byte(int fd, int expected_ret, int flags)
+{
+	const uint8_t byte = 'A';
+	ssize_t nwritten;
+
+	timeout_begin(TIMEOUT);
+	do {
+		nwritten = send(fd, &byte, sizeof(byte), flags);
+		timeout_check("write");
+	} while (nwritten < 0 && errno == EINTR);
+	timeout_end();
+
+	if (expected_ret < 0) {
+		if (nwritten != -1) {
+			fprintf(stderr, "bogus send(2) return value %zd\n",
+				nwritten);
+			exit(EXIT_FAILURE);
+		}
+		if (errno != -expected_ret) {
+			perror("write");
+			exit(EXIT_FAILURE);
+		}
+		return;
+	}
+
+	if (nwritten < 0) {
+		perror("write");
+		exit(EXIT_FAILURE);
+	}
+	if (nwritten == 0) {
+		if (expected_ret == 0)
+			return;
+
+		fprintf(stderr, "unexpected EOF while sending byte\n");
+		exit(EXIT_FAILURE);
+	}
+	if (nwritten != sizeof(byte)) {
+		fprintf(stderr, "bogus send(2) return value %zd\n", nwritten);
+		exit(EXIT_FAILURE);
+	}
+}
+
+/* Receive one byte and check the return value.
+ *
+ * expected_ret:
+ *  <0 Negative errno (for testing errors)
+ *   0 End-of-file
+ *   1 Success
+ */
+void recv_byte(int fd, int expected_ret, int flags)
+{
+	uint8_t byte;
+	ssize_t nread;
+
+	timeout_begin(TIMEOUT);
+	do {
+		nread = recv(fd, &byte, sizeof(byte), flags);
+		timeout_check("read");
+	} while (nread < 0 && errno == EINTR);
+	timeout_end();
+
+	if (expected_ret < 0) {
+		if (nread != -1) {
+			fprintf(stderr, "bogus recv(2) return value %zd\n",
+				nread);
+			exit(EXIT_FAILURE);
+		}
+		if (errno != -expected_ret) {
+			perror("read");
+			exit(EXIT_FAILURE);
+		}
+		return;
+	}
+
+	if (nread < 0) {
+		perror("read");
+		exit(EXIT_FAILURE);
+	}
+	if (nread == 0) {
+		if (expected_ret == 0)
+			return;
+
+		fprintf(stderr, "unexpected EOF while receiving byte\n");
+		exit(EXIT_FAILURE);
+	}
+	if (nread != sizeof(byte)) {
+		fprintf(stderr, "bogus recv(2) return value %zd\n", nread);
+		exit(EXIT_FAILURE);
+	}
+	if (byte != 'A') {
+		fprintf(stderr, "unexpected byte read %c\n", byte);
+		exit(EXIT_FAILURE);
+	}
+}
+
 /* Run test cases.  The program terminates if a failure occurs. */
 void run_tests(const struct test_case *test_cases,
 	       const struct test_opts *opts)
diff --git a/tools/testing/vsock/util.h b/tools/testing/vsock/util.h
index 1786305cfddd..4df12e4b5ebe 100644
--- a/tools/testing/vsock/util.h
+++ b/tools/testing/vsock/util.h
@@ -36,6 +36,8 @@ unsigned int parse_cid(const char *str);
 int vsock_stream_connect(unsigned int cid, unsigned int port);
 int vsock_stream_accept(unsigned int cid, unsigned int port,
 			struct sockaddr_vm *clientaddrp);
+void send_byte(int fd, int expected_ret, int flags);
+void recv_byte(int fd, int expected_ret, int flags);
 void run_tests(const struct test_case *test_cases,
 	       const struct test_opts *opts);
 
-- 
cgit v1.2.3-59-g8ed1b


From cdbcc18de10a7d72b821fe40fbc2d0ec61e39448 Mon Sep 17 00:00:00 2001
From: Stefan Hajnoczi <stefanha@redhat.com>
Date: Wed, 18 Dec 2019 19:07:04 +0100
Subject: VSOCK: add AF_VSOCK test cases

The vsock_test.c program runs a test suite of AF_VSOCK test cases.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/vsock/.gitignore   |   1 +
 tools/testing/vsock/Makefile     |   5 +-
 tools/testing/vsock/README       |   1 +
 tools/testing/vsock/vsock_test.c | 312 +++++++++++++++++++++++++++++++++++++++
 4 files changed, 317 insertions(+), 2 deletions(-)
 create mode 100644 tools/testing/vsock/vsock_test.c

diff --git a/tools/testing/vsock/.gitignore b/tools/testing/vsock/.gitignore
index dc5f11faf530..7f7a2ccc30c4 100644
--- a/tools/testing/vsock/.gitignore
+++ b/tools/testing/vsock/.gitignore
@@ -1,2 +1,3 @@
 *.d
+vsock_test
 vsock_diag_test
diff --git a/tools/testing/vsock/Makefile b/tools/testing/vsock/Makefile
index a916878a2d8c..f8293c6910c9 100644
--- a/tools/testing/vsock/Makefile
+++ b/tools/testing/vsock/Makefile
@@ -1,10 +1,11 @@
 # SPDX-License-Identifier: GPL-2.0-only
 all: test
-test: vsock_diag_test
+test: vsock_test vsock_diag_test
+vsock_test: vsock_test.o timeout.o control.o util.o
 vsock_diag_test: vsock_diag_test.o timeout.o control.o util.o
 
 CFLAGS += -g -O2 -Werror -Wall -I. -I../../include -I../../../usr/include -Wno-pointer-sign -fno-strict-overflow -fno-strict-aliasing -fno-common -MMD -U_FORTIFY_SOURCE -D_GNU_SOURCE
 .PHONY: all test clean
 clean:
-	${RM} *.o *.d vsock_diag_test
+	${RM} *.o *.d vsock_test vsock_diag_test
 -include *.d
diff --git a/tools/testing/vsock/README b/tools/testing/vsock/README
index cf7dc64273bf..4d5045e7d2c3 100644
--- a/tools/testing/vsock/README
+++ b/tools/testing/vsock/README
@@ -5,6 +5,7 @@ Hyper-V.
 
 The following tests are available:
 
+  * vsock_test - core AF_VSOCK socket functionality
   * vsock_diag_test - vsock_diag.ko module for listing open sockets
 
 The following prerequisite steps are not automated and must be performed prior
diff --git a/tools/testing/vsock/vsock_test.c b/tools/testing/vsock/vsock_test.c
new file mode 100644
index 000000000000..fae8ddc3ef72
--- /dev/null
+++ b/tools/testing/vsock/vsock_test.c
@@ -0,0 +1,312 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * vsock_test - vsock.ko test suite
+ *
+ * Copyright (C) 2017 Red Hat, Inc.
+ *
+ * Author: Stefan Hajnoczi <stefanha@redhat.com>
+ */
+
+#include <getopt.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <errno.h>
+#include <unistd.h>
+
+#include "timeout.h"
+#include "control.h"
+#include "util.h"
+
+static void test_stream_connection_reset(const struct test_opts *opts)
+{
+	union {
+		struct sockaddr sa;
+		struct sockaddr_vm svm;
+	} addr = {
+		.svm = {
+			.svm_family = AF_VSOCK,
+			.svm_port = 1234,
+			.svm_cid = opts->peer_cid,
+		},
+	};
+	int ret;
+	int fd;
+
+	fd = socket(AF_VSOCK, SOCK_STREAM, 0);
+
+	timeout_begin(TIMEOUT);
+	do {
+		ret = connect(fd, &addr.sa, sizeof(addr.svm));
+		timeout_check("connect");
+	} while (ret < 0 && errno == EINTR);
+	timeout_end();
+
+	if (ret != -1) {
+		fprintf(stderr, "expected connect(2) failure, got %d\n", ret);
+		exit(EXIT_FAILURE);
+	}
+	if (errno != ECONNRESET) {
+		fprintf(stderr, "unexpected connect(2) errno %d\n", errno);
+		exit(EXIT_FAILURE);
+	}
+
+	close(fd);
+}
+
+static void test_stream_client_close_client(const struct test_opts *opts)
+{
+	int fd;
+
+	fd = vsock_stream_connect(opts->peer_cid, 1234);
+	if (fd < 0) {
+		perror("connect");
+		exit(EXIT_FAILURE);
+	}
+
+	send_byte(fd, 1, 0);
+	close(fd);
+	control_writeln("CLOSED");
+}
+
+static void test_stream_client_close_server(const struct test_opts *opts)
+{
+	int fd;
+
+	fd = vsock_stream_accept(VMADDR_CID_ANY, 1234, NULL);
+	if (fd < 0) {
+		perror("accept");
+		exit(EXIT_FAILURE);
+	}
+
+	control_expectln("CLOSED");
+
+	send_byte(fd, -EPIPE, 0);
+	recv_byte(fd, 1, 0);
+	recv_byte(fd, 0, 0);
+	close(fd);
+}
+
+static void test_stream_server_close_client(const struct test_opts *opts)
+{
+	int fd;
+
+	fd = vsock_stream_connect(opts->peer_cid, 1234);
+	if (fd < 0) {
+		perror("connect");
+		exit(EXIT_FAILURE);
+	}
+
+	control_expectln("CLOSED");
+
+	send_byte(fd, -EPIPE, 0);
+	recv_byte(fd, 1, 0);
+	recv_byte(fd, 0, 0);
+	close(fd);
+}
+
+static void test_stream_server_close_server(const struct test_opts *opts)
+{
+	int fd;
+
+	fd = vsock_stream_accept(VMADDR_CID_ANY, 1234, NULL);
+	if (fd < 0) {
+		perror("accept");
+		exit(EXIT_FAILURE);
+	}
+
+	send_byte(fd, 1, 0);
+	close(fd);
+	control_writeln("CLOSED");
+}
+
+/* With the standard socket sizes, VMCI is able to support about 100
+ * concurrent stream connections.
+ */
+#define MULTICONN_NFDS 100
+
+static void test_stream_multiconn_client(const struct test_opts *opts)
+{
+	int fds[MULTICONN_NFDS];
+	int i;
+
+	for (i = 0; i < MULTICONN_NFDS; i++) {
+		fds[i] = vsock_stream_connect(opts->peer_cid, 1234);
+		if (fds[i] < 0) {
+			perror("connect");
+			exit(EXIT_FAILURE);
+		}
+	}
+
+	for (i = 0; i < MULTICONN_NFDS; i++) {
+		if (i % 2)
+			recv_byte(fds[i], 1, 0);
+		else
+			send_byte(fds[i], 1, 0);
+	}
+
+	for (i = 0; i < MULTICONN_NFDS; i++)
+		close(fds[i]);
+}
+
+static void test_stream_multiconn_server(const struct test_opts *opts)
+{
+	int fds[MULTICONN_NFDS];
+	int i;
+
+	for (i = 0; i < MULTICONN_NFDS; i++) {
+		fds[i] = vsock_stream_accept(VMADDR_CID_ANY, 1234, NULL);
+		if (fds[i] < 0) {
+			perror("accept");
+			exit(EXIT_FAILURE);
+		}
+	}
+
+	for (i = 0; i < MULTICONN_NFDS; i++) {
+		if (i % 2)
+			send_byte(fds[i], 1, 0);
+		else
+			recv_byte(fds[i], 1, 0);
+	}
+
+	for (i = 0; i < MULTICONN_NFDS; i++)
+		close(fds[i]);
+}
+
+static struct test_case test_cases[] = {
+	{
+		.name = "SOCK_STREAM connection reset",
+		.run_client = test_stream_connection_reset,
+	},
+	{
+		.name = "SOCK_STREAM client close",
+		.run_client = test_stream_client_close_client,
+		.run_server = test_stream_client_close_server,
+	},
+	{
+		.name = "SOCK_STREAM server close",
+		.run_client = test_stream_server_close_client,
+		.run_server = test_stream_server_close_server,
+	},
+	{
+		.name = "SOCK_STREAM multiple connections",
+		.run_client = test_stream_multiconn_client,
+		.run_server = test_stream_multiconn_server,
+	},
+	{},
+};
+
+static const char optstring[] = "";
+static const struct option longopts[] = {
+	{
+		.name = "control-host",
+		.has_arg = required_argument,
+		.val = 'H',
+	},
+	{
+		.name = "control-port",
+		.has_arg = required_argument,
+		.val = 'P',
+	},
+	{
+		.name = "mode",
+		.has_arg = required_argument,
+		.val = 'm',
+	},
+	{
+		.name = "peer-cid",
+		.has_arg = required_argument,
+		.val = 'p',
+	},
+	{
+		.name = "help",
+		.has_arg = no_argument,
+		.val = '?',
+	},
+	{},
+};
+
+static void usage(void)
+{
+	fprintf(stderr, "Usage: vsock_test [--help] [--control-host=<host>] --control-port=<port> --mode=client|server --peer-cid=<cid>\n"
+		"\n"
+		"  Server: vsock_test --control-port=1234 --mode=server --peer-cid=3\n"
+		"  Client: vsock_test --control-host=192.168.0.1 --control-port=1234 --mode=client --peer-cid=2\n"
+		"\n"
+		"Run vsock.ko tests.  Must be launched in both guest\n"
+		"and host.  One side must use --mode=client and\n"
+		"the other side must use --mode=server.\n"
+		"\n"
+		"A TCP control socket connection is used to coordinate tests\n"
+		"between the client and the server.  The server requires a\n"
+		"listen address and the client requires an address to\n"
+		"connect to.\n"
+		"\n"
+		"The CID of the other side must be given with --peer-cid=<cid>.\n");
+	exit(EXIT_FAILURE);
+}
+
+int main(int argc, char **argv)
+{
+	const char *control_host = NULL;
+	const char *control_port = NULL;
+	struct test_opts opts = {
+		.mode = TEST_MODE_UNSET,
+		.peer_cid = VMADDR_CID_ANY,
+	};
+
+	init_signals();
+
+	for (;;) {
+		int opt = getopt_long(argc, argv, optstring, longopts, NULL);
+
+		if (opt == -1)
+			break;
+
+		switch (opt) {
+		case 'H':
+			control_host = optarg;
+			break;
+		case 'm':
+			if (strcmp(optarg, "client") == 0)
+				opts.mode = TEST_MODE_CLIENT;
+			else if (strcmp(optarg, "server") == 0)
+				opts.mode = TEST_MODE_SERVER;
+			else {
+				fprintf(stderr, "--mode must be \"client\" or \"server\"\n");
+				return EXIT_FAILURE;
+			}
+			break;
+		case 'p':
+			opts.peer_cid = parse_cid(optarg);
+			break;
+		case 'P':
+			control_port = optarg;
+			break;
+		case '?':
+		default:
+			usage();
+		}
+	}
+
+	if (!control_port)
+		usage();
+	if (opts.mode == TEST_MODE_UNSET)
+		usage();
+	if (opts.peer_cid == VMADDR_CID_ANY)
+		usage();
+
+	if (!control_host) {
+		if (opts.mode != TEST_MODE_SERVER)
+			usage();
+		control_host = "0.0.0.0";
+	}
+
+	control_init(control_host, control_port,
+		     opts.mode == TEST_MODE_SERVER);
+
+	run_tests(test_cases, &opts);
+
+	control_cleanup();
+	return EXIT_SUCCESS;
+}
-- 
cgit v1.2.3-59-g8ed1b


From 770ce0078cbf97262e86c9cc210684ce3b4266f5 Mon Sep 17 00:00:00 2001
From: Stefano Garzarella <sgarzare@redhat.com>
Date: Wed, 18 Dec 2019 19:07:05 +0100
Subject: vsock_test: wait for the remote to close the connection

Before check if a send returns -EPIPE, we need to make sure the
connection is closed.
To do that, we use epoll API to wait EPOLLRDHUP or EPOLLHUP events
on the socket.

Reported-by: Jorgen Hansen <jhansen@vmware.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/vsock/util.c       | 39 +++++++++++++++++++++++++++++++++++++++
 tools/testing/vsock/util.h       |  1 +
 tools/testing/vsock/vsock_test.c | 12 ++++++++----
 3 files changed, 48 insertions(+), 4 deletions(-)

diff --git a/tools/testing/vsock/util.c b/tools/testing/vsock/util.c
index 6026ef3ce512..b132c96c87fc 100644
--- a/tools/testing/vsock/util.c
+++ b/tools/testing/vsock/util.c
@@ -13,6 +13,8 @@
 #include <stdlib.h>
 #include <signal.h>
 #include <unistd.h>
+#include <assert.h>
+#include <sys/epoll.h>
 
 #include "timeout.h"
 #include "control.h"
@@ -44,6 +46,43 @@ unsigned int parse_cid(const char *str)
 	return n;
 }
 
+/* Wait for the remote to close the connection */
+void vsock_wait_remote_close(int fd)
+{
+	struct epoll_event ev;
+	int epollfd, nfds;
+
+	epollfd = epoll_create1(0);
+	if (epollfd == -1) {
+		perror("epoll_create1");
+		exit(EXIT_FAILURE);
+	}
+
+	ev.events = EPOLLRDHUP | EPOLLHUP;
+	ev.data.fd = fd;
+	if (epoll_ctl(epollfd, EPOLL_CTL_ADD, fd, &ev) == -1) {
+		perror("epoll_ctl");
+		exit(EXIT_FAILURE);
+	}
+
+	nfds = epoll_wait(epollfd, &ev, 1, TIMEOUT * 1000);
+	if (nfds == -1) {
+		perror("epoll_wait");
+		exit(EXIT_FAILURE);
+	}
+
+	if (nfds == 0) {
+		fprintf(stderr, "epoll_wait timed out\n");
+		exit(EXIT_FAILURE);
+	}
+
+	assert(nfds == 1);
+	assert(ev.events & (EPOLLRDHUP | EPOLLHUP));
+	assert(ev.data.fd == fd);
+
+	close(epollfd);
+}
+
 /* Connect to <cid, port> and return the file descriptor. */
 int vsock_stream_connect(unsigned int cid, unsigned int port)
 {
diff --git a/tools/testing/vsock/util.h b/tools/testing/vsock/util.h
index 4df12e4b5ebe..331e945f3ae6 100644
--- a/tools/testing/vsock/util.h
+++ b/tools/testing/vsock/util.h
@@ -36,6 +36,7 @@ unsigned int parse_cid(const char *str);
 int vsock_stream_connect(unsigned int cid, unsigned int port);
 int vsock_stream_accept(unsigned int cid, unsigned int port,
 			struct sockaddr_vm *clientaddrp);
+void vsock_wait_remote_close(int fd);
 void send_byte(int fd, int expected_ret, int flags);
 void recv_byte(int fd, int expected_ret, int flags);
 void run_tests(const struct test_case *test_cases,
diff --git a/tools/testing/vsock/vsock_test.c b/tools/testing/vsock/vsock_test.c
index fae8ddc3ef72..629d7ce58202 100644
--- a/tools/testing/vsock/vsock_test.c
+++ b/tools/testing/vsock/vsock_test.c
@@ -66,7 +66,6 @@ static void test_stream_client_close_client(const struct test_opts *opts)
 
 	send_byte(fd, 1, 0);
 	close(fd);
-	control_writeln("CLOSED");
 }
 
 static void test_stream_client_close_server(const struct test_opts *opts)
@@ -79,7 +78,10 @@ static void test_stream_client_close_server(const struct test_opts *opts)
 		exit(EXIT_FAILURE);
 	}
 
-	control_expectln("CLOSED");
+	/* Wait for the remote to close the connection, before check
+	 * -EPIPE error on send.
+	 */
+	vsock_wait_remote_close(fd);
 
 	send_byte(fd, -EPIPE, 0);
 	recv_byte(fd, 1, 0);
@@ -97,7 +99,10 @@ static void test_stream_server_close_client(const struct test_opts *opts)
 		exit(EXIT_FAILURE);
 	}
 
-	control_expectln("CLOSED");
+	/* Wait for the remote to close the connection, before check
+	 * -EPIPE error on send.
+	 */
+	vsock_wait_remote_close(fd);
 
 	send_byte(fd, -EPIPE, 0);
 	recv_byte(fd, 1, 0);
@@ -117,7 +122,6 @@ static void test_stream_server_close_server(const struct test_opts *opts)
 
 	send_byte(fd, 1, 0);
 	close(fd);
-	control_writeln("CLOSED");
 }
 
 /* With the standard socket sizes, VMCI is able to support about 100
-- 
cgit v1.2.3-59-g8ed1b


From 5a2b2425634dd8d340d3889f2317a752a1a87618 Mon Sep 17 00:00:00 2001
From: Stefano Garzarella <sgarzare@redhat.com>
Date: Wed, 18 Dec 2019 19:07:06 +0100
Subject: testing/vsock: add parameters to list and skip tests

Some tests can fail with transports that have a slightly
different behavior, so let's add the possibility to specify
which tests to skip.

Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/vsock/control.c         | 15 ++++++-
 tools/testing/vsock/control.h         |  1 +
 tools/testing/vsock/util.c            | 75 ++++++++++++++++++++++++++++-------
 tools/testing/vsock/util.h            |  6 ++-
 tools/testing/vsock/vsock_diag_test.c | 19 ++++++++-
 tools/testing/vsock/vsock_test.c      | 20 +++++++++-
 6 files changed, 116 insertions(+), 20 deletions(-)

diff --git a/tools/testing/vsock/control.c b/tools/testing/vsock/control.c
index 45f328c6ff23..4874872fc5a3 100644
--- a/tools/testing/vsock/control.c
+++ b/tools/testing/vsock/control.c
@@ -205,11 +205,22 @@ void control_expectln(const char *str)
 	char *line;
 
 	line = control_readln();
-	if (strcmp(str, line) != 0) {
+
+	control_cmpln(line, str, true);
+
+	free(line);
+}
+
+bool control_cmpln(char *line, const char *str, bool fail)
+{
+	if (strcmp(str, line) == 0)
+		return true;
+
+	if (fail) {
 		fprintf(stderr, "expected \"%s\" on control socket, got \"%s\"\n",
 			str, line);
 		exit(EXIT_FAILURE);
 	}
 
-	free(line);
+	return false;
 }
diff --git a/tools/testing/vsock/control.h b/tools/testing/vsock/control.h
index dac3964a891d..51814b4f9ac1 100644
--- a/tools/testing/vsock/control.h
+++ b/tools/testing/vsock/control.h
@@ -10,5 +10,6 @@ void control_cleanup(void);
 void control_writeln(const char *str);
 char *control_readln(void);
 void control_expectln(const char *str);
+bool control_cmpln(char *line, const char *str, bool fail);
 
 #endif /* CONTROL_H */
diff --git a/tools/testing/vsock/util.c b/tools/testing/vsock/util.c
index b132c96c87fc..93cbd6f603f9 100644
--- a/tools/testing/vsock/util.c
+++ b/tools/testing/vsock/util.c
@@ -299,32 +299,77 @@ void run_tests(const struct test_case *test_cases,
 
 	for (i = 0; test_cases[i].name; i++) {
 		void (*run)(const struct test_opts *opts);
+		char *line;
 
-		printf("%s...", test_cases[i].name);
+		printf("%d - %s...", i, test_cases[i].name);
 		fflush(stdout);
 
-		if (opts->mode == TEST_MODE_CLIENT) {
-			/* Full barrier before executing the next test.  This
-			 * ensures that client and server are executing the
-			 * same test case.  In particular, it means whoever is
-			 * faster will not see the peer still executing the
-			 * last test.  This is important because port numbers
-			 * can be used by multiple test cases.
-			 */
-			control_expectln("NEXT");
+		/* Full barrier before executing the next test.  This
+		 * ensures that client and server are executing the
+		 * same test case.  In particular, it means whoever is
+		 * faster will not see the peer still executing the
+		 * last test.  This is important because port numbers
+		 * can be used by multiple test cases.
+		 */
+		if (test_cases[i].skip)
+			control_writeln("SKIP");
+		else
 			control_writeln("NEXT");
 
-			run = test_cases[i].run_client;
-		} else {
-			control_writeln("NEXT");
-			control_expectln("NEXT");
+		line = control_readln();
+		if (control_cmpln(line, "SKIP", false) || test_cases[i].skip) {
 
-			run = test_cases[i].run_server;
+			printf("skipped\n");
+
+			free(line);
+			continue;
 		}
 
+		control_cmpln(line, "NEXT", true);
+		free(line);
+
+		if (opts->mode == TEST_MODE_CLIENT)
+			run = test_cases[i].run_client;
+		else
+			run = test_cases[i].run_server;
+
 		if (run)
 			run(opts);
 
 		printf("ok\n");
 	}
 }
+
+void list_tests(const struct test_case *test_cases)
+{
+	int i;
+
+	printf("ID\tTest name\n");
+
+	for (i = 0; test_cases[i].name; i++)
+		printf("%d\t%s\n", i, test_cases[i].name);
+
+	exit(EXIT_FAILURE);
+}
+
+void skip_test(struct test_case *test_cases, size_t test_cases_len,
+	       const char *test_id_str)
+{
+	unsigned long test_id;
+	char *endptr = NULL;
+
+	errno = 0;
+	test_id = strtoul(test_id_str, &endptr, 10);
+	if (errno || *endptr != '\0') {
+		fprintf(stderr, "malformed test ID \"%s\"\n", test_id_str);
+		exit(EXIT_FAILURE);
+	}
+
+	if (test_id >= test_cases_len) {
+		fprintf(stderr, "test ID (%lu) larger than the max allowed (%lu)\n",
+			test_id, test_cases_len - 1);
+		exit(EXIT_FAILURE);
+	}
+
+	test_cases[test_id].skip = true;
+}
diff --git a/tools/testing/vsock/util.h b/tools/testing/vsock/util.h
index 331e945f3ae6..e53dd09d26d9 100644
--- a/tools/testing/vsock/util.h
+++ b/tools/testing/vsock/util.h
@@ -29,6 +29,8 @@ struct test_case {
 
 	/* Called when test mode is TEST_MODE_SERVER */
 	void (*run_server)(const struct test_opts *opts);
+
+	bool skip;
 };
 
 void init_signals(void);
@@ -41,5 +43,7 @@ void send_byte(int fd, int expected_ret, int flags);
 void recv_byte(int fd, int expected_ret, int flags);
 void run_tests(const struct test_case *test_cases,
 	       const struct test_opts *opts);
-
+void list_tests(const struct test_case *test_cases);
+void skip_test(struct test_case *test_cases, size_t test_cases_len,
+	       const char *test_id_str);
 #endif /* UTIL_H */
diff --git a/tools/testing/vsock/vsock_diag_test.c b/tools/testing/vsock/vsock_diag_test.c
index abd7dc2a9631..b82483627259 100644
--- a/tools/testing/vsock/vsock_diag_test.c
+++ b/tools/testing/vsock/vsock_diag_test.c
@@ -463,6 +463,16 @@ static const struct option longopts[] = {
 		.has_arg = required_argument,
 		.val = 'p',
 	},
+	{
+		.name = "list",
+		.has_arg = no_argument,
+		.val = 'l',
+	},
+	{
+		.name = "skip",
+		.has_arg = required_argument,
+		.val = 's',
+	},
 	{
 		.name = "help",
 		.has_arg = no_argument,
@@ -473,7 +483,7 @@ static const struct option longopts[] = {
 
 static void usage(void)
 {
-	fprintf(stderr, "Usage: vsock_diag_test [--help] [--control-host=<host>] --control-port=<port> --mode=client|server --peer-cid=<cid>\n"
+	fprintf(stderr, "Usage: vsock_diag_test [--help] [--control-host=<host>] --control-port=<port> --mode=client|server --peer-cid=<cid> [--list] [--skip=<test_id>]\n"
 		"\n"
 		"  Server: vsock_diag_test --control-port=1234 --mode=server --peer-cid=3\n"
 		"  Client: vsock_diag_test --control-host=192.168.0.1 --control-port=1234 --mode=client --peer-cid=2\n"
@@ -528,6 +538,13 @@ int main(int argc, char **argv)
 		case 'P':
 			control_port = optarg;
 			break;
+		case 'l':
+			list_tests(test_cases);
+			break;
+		case 's':
+			skip_test(test_cases, ARRAY_SIZE(test_cases) - 1,
+				  optarg);
+			break;
 		case '?':
 		default:
 			usage();
diff --git a/tools/testing/vsock/vsock_test.c b/tools/testing/vsock/vsock_test.c
index 629d7ce58202..3ac56651f3f9 100644
--- a/tools/testing/vsock/vsock_test.c
+++ b/tools/testing/vsock/vsock_test.c
@@ -13,6 +13,7 @@
 #include <string.h>
 #include <errno.h>
 #include <unistd.h>
+#include <linux/kernel.h>
 
 #include "timeout.h"
 #include "control.h"
@@ -222,6 +223,16 @@ static const struct option longopts[] = {
 		.has_arg = required_argument,
 		.val = 'p',
 	},
+	{
+		.name = "list",
+		.has_arg = no_argument,
+		.val = 'l',
+	},
+	{
+		.name = "skip",
+		.has_arg = required_argument,
+		.val = 's',
+	},
 	{
 		.name = "help",
 		.has_arg = no_argument,
@@ -232,7 +243,7 @@ static const struct option longopts[] = {
 
 static void usage(void)
 {
-	fprintf(stderr, "Usage: vsock_test [--help] [--control-host=<host>] --control-port=<port> --mode=client|server --peer-cid=<cid>\n"
+	fprintf(stderr, "Usage: vsock_test [--help] [--control-host=<host>] --control-port=<port> --mode=client|server --peer-cid=<cid> [--list] [--skip=<test_id>]\n"
 		"\n"
 		"  Server: vsock_test --control-port=1234 --mode=server --peer-cid=3\n"
 		"  Client: vsock_test --control-host=192.168.0.1 --control-port=1234 --mode=client --peer-cid=2\n"
@@ -287,6 +298,13 @@ int main(int argc, char **argv)
 		case 'P':
 			control_port = optarg;
 			break;
+		case 'l':
+			list_tests(test_cases);
+			break;
+		case 's':
+			skip_test(test_cases, ARRAY_SIZE(test_cases) - 1,
+				  optarg);
+			break;
 		case '?':
 		default:
 			usage();
-- 
cgit v1.2.3-59-g8ed1b


From 8d00b93fece1cc3547a53abd6ef3358d23ff4d25 Mon Sep 17 00:00:00 2001
From: Stefano Garzarella <sgarzare@redhat.com>
Date: Wed, 18 Dec 2019 19:07:07 +0100
Subject: testing/vsock: print list of options and description

Since we now have several options, in the help we print the list
of all supported options and a brief description of them.

Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/vsock/vsock_diag_test.c | 13 ++++++++++++-
 tools/testing/vsock/vsock_test.c      | 13 ++++++++++++-
 2 files changed, 24 insertions(+), 2 deletions(-)

diff --git a/tools/testing/vsock/vsock_diag_test.c b/tools/testing/vsock/vsock_diag_test.c
index b82483627259..cec6f5a738e1 100644
--- a/tools/testing/vsock/vsock_diag_test.c
+++ b/tools/testing/vsock/vsock_diag_test.c
@@ -497,7 +497,18 @@ static void usage(void)
 		"listen address and the client requires an address to\n"
 		"connect to.\n"
 		"\n"
-		"The CID of the other side must be given with --peer-cid=<cid>.\n");
+		"The CID of the other side must be given with --peer-cid=<cid>.\n"
+		"\n"
+		"Options:\n"
+		"  --help                 This help message\n"
+		"  --control-host <host>  Server IP address to connect to\n"
+		"  --control-port <port>  Server port to listen on/connect to\n"
+		"  --mode client|server   Server or client mode\n"
+		"  --peer-cid <cid>       CID of the other side\n"
+		"  --list                 List of tests that will be executed\n"
+		"  --skip <test_id>       Test ID to skip;\n"
+		"                         use multiple --skip options to skip more tests\n"
+		);
 	exit(EXIT_FAILURE);
 }
 
diff --git a/tools/testing/vsock/vsock_test.c b/tools/testing/vsock/vsock_test.c
index 3ac56651f3f9..a63e05d6a0f9 100644
--- a/tools/testing/vsock/vsock_test.c
+++ b/tools/testing/vsock/vsock_test.c
@@ -257,7 +257,18 @@ static void usage(void)
 		"listen address and the client requires an address to\n"
 		"connect to.\n"
 		"\n"
-		"The CID of the other side must be given with --peer-cid=<cid>.\n");
+		"The CID of the other side must be given with --peer-cid=<cid>.\n"
+		"\n"
+		"Options:\n"
+		"  --help                 This help message\n"
+		"  --control-host <host>  Server IP address to connect to\n"
+		"  --control-port <port>  Server port to listen on/connect to\n"
+		"  --mode client|server   Server or client mode\n"
+		"  --peer-cid <cid>       CID of the other side\n"
+		"  --list                 List of tests that will be executed\n"
+		"  --skip <test_id>       Test ID to skip;\n"
+		"                         use multiple --skip options to skip more tests\n"
+		);
 	exit(EXIT_FAILURE);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From d6269a93ffc37effac85903122023a251dfa4aca Mon Sep 17 00:00:00 2001
From: Stefano Garzarella <sgarzare@redhat.com>
Date: Wed, 18 Dec 2019 19:07:08 +0100
Subject: vsock_test: add SOCK_STREAM MSG_PEEK test

Test if the MSG_PEEK flags of recv(2) works as expected.

Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/vsock/vsock_test.c | 34 ++++++++++++++++++++++++++++++++++
 1 file changed, 34 insertions(+)

diff --git a/tools/testing/vsock/vsock_test.c b/tools/testing/vsock/vsock_test.c
index a63e05d6a0f9..1d8b93f1af31 100644
--- a/tools/testing/vsock/vsock_test.c
+++ b/tools/testing/vsock/vsock_test.c
@@ -178,6 +178,35 @@ static void test_stream_multiconn_server(const struct test_opts *opts)
 		close(fds[i]);
 }
 
+static void test_stream_msg_peek_client(const struct test_opts *opts)
+{
+	int fd;
+
+	fd = vsock_stream_connect(opts->peer_cid, 1234);
+	if (fd < 0) {
+		perror("connect");
+		exit(EXIT_FAILURE);
+	}
+
+	send_byte(fd, 1, 0);
+	close(fd);
+}
+
+static void test_stream_msg_peek_server(const struct test_opts *opts)
+{
+	int fd;
+
+	fd = vsock_stream_accept(VMADDR_CID_ANY, 1234, NULL);
+	if (fd < 0) {
+		perror("accept");
+		exit(EXIT_FAILURE);
+	}
+
+	recv_byte(fd, 1, MSG_PEEK);
+	recv_byte(fd, 1, 0);
+	close(fd);
+}
+
 static struct test_case test_cases[] = {
 	{
 		.name = "SOCK_STREAM connection reset",
@@ -198,6 +227,11 @@ static struct test_case test_cases[] = {
 		.run_client = test_stream_multiconn_client,
 		.run_server = test_stream_multiconn_server,
 	},
+	{
+		.name = "SOCK_STREAM MSG_PEEK",
+		.run_client = test_stream_msg_peek_client,
+		.run_server = test_stream_msg_peek_server,
+	},
 	{},
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From e1b5e598e5a51b453328879682b178b4acc15105 Mon Sep 17 00:00:00 2001
From: John Rutherford <john.rutherford@dektech.com.au>
Date: Thu, 19 Dec 2019 16:03:57 +1100
Subject: tipc: make legacy address flag readable over netlink

To enable iproute2/tipc to generate backwards compatible
printouts and validate command parameters for nodes using a
<z.c.n> node address, it needs to be able to read the legacy
address flag from the kernel.  The legacy address flag records
the way in which the node identity was originally specified.

The legacy address flag is requested by the netlink message
TIPC_NL_ADDR_LEGACY_GET.  If the flag is set the attribute
TIPC_NLA_NET_ADDR_LEGACY is set in the return message.

Signed-off-by: John Rutherford <john.rutherford@dektech.com.au>
Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/uapi/linux/tipc_netlink.h |  2 ++
 net/tipc/net.c                    | 56 +++++++++++++++++++++++++++++++++++++++
 net/tipc/net.h                    |  1 +
 net/tipc/netlink.c                |  6 +++++
 4 files changed, 65 insertions(+)

diff --git a/include/uapi/linux/tipc_netlink.h b/include/uapi/linux/tipc_netlink.h
index 6c2194ab745b..dc0d23a50e69 100644
--- a/include/uapi/linux/tipc_netlink.h
+++ b/include/uapi/linux/tipc_netlink.h
@@ -65,6 +65,7 @@ enum {
 	TIPC_NL_UDP_GET_REMOTEIP,
 	TIPC_NL_KEY_SET,
 	TIPC_NL_KEY_FLUSH,
+	TIPC_NL_ADDR_LEGACY_GET,
 
 	__TIPC_NL_CMD_MAX,
 	TIPC_NL_CMD_MAX = __TIPC_NL_CMD_MAX - 1
@@ -176,6 +177,7 @@ enum {
 	TIPC_NLA_NET_ADDR,		/* u32 */
 	TIPC_NLA_NET_NODEID,		/* u64 */
 	TIPC_NLA_NET_NODEID_W1,		/* u64 */
+	TIPC_NLA_NET_ADDR_LEGACY,	/* flag */
 
 	__TIPC_NLA_NET_MAX,
 	TIPC_NLA_NET_MAX = __TIPC_NLA_NET_MAX - 1
diff --git a/net/tipc/net.c b/net/tipc/net.c
index 2de3cec9929d..85400e4242de 100644
--- a/net/tipc/net.c
+++ b/net/tipc/net.c
@@ -302,3 +302,59 @@ int tipc_nl_net_set(struct sk_buff *skb, struct genl_info *info)
 
 	return err;
 }
+
+static int __tipc_nl_addr_legacy_get(struct net *net, struct tipc_nl_msg *msg)
+{
+	struct tipc_net *tn = tipc_net(net);
+	struct nlattr *attrs;
+	void *hdr;
+
+	hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_family,
+			  0, TIPC_NL_ADDR_LEGACY_GET);
+	if (!hdr)
+		return -EMSGSIZE;
+
+	attrs = nla_nest_start(msg->skb, TIPC_NLA_NET);
+	if (!attrs)
+		goto msg_full;
+
+	if (tn->legacy_addr_format)
+		if (nla_put_flag(msg->skb, TIPC_NLA_NET_ADDR_LEGACY))
+			goto attr_msg_full;
+
+	nla_nest_end(msg->skb, attrs);
+	genlmsg_end(msg->skb, hdr);
+
+	return 0;
+
+attr_msg_full:
+	nla_nest_cancel(msg->skb, attrs);
+msg_full:
+	genlmsg_cancel(msg->skb, hdr);
+
+	return -EMSGSIZE;
+}
+
+int tipc_nl_net_addr_legacy_get(struct sk_buff *skb, struct genl_info *info)
+{
+	struct net *net = sock_net(skb->sk);
+	struct tipc_nl_msg msg;
+	struct sk_buff *rep;
+	int err;
+
+	rep = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL);
+	if (!rep)
+		return -ENOMEM;
+
+	msg.skb = rep;
+	msg.portid = info->snd_portid;
+	msg.seq = info->snd_seq;
+
+	err = __tipc_nl_addr_legacy_get(net, &msg);
+	if (err) {
+		nlmsg_free(msg.skb);
+		return err;
+	}
+
+	return genlmsg_reply(msg.skb, info);
+}
diff --git a/net/tipc/net.h b/net/tipc/net.h
index b7f2e364eb99..6740d97c706e 100644
--- a/net/tipc/net.h
+++ b/net/tipc/net.h
@@ -47,5 +47,6 @@ void tipc_net_stop(struct net *net);
 int tipc_nl_net_dump(struct sk_buff *skb, struct netlink_callback *cb);
 int tipc_nl_net_set(struct sk_buff *skb, struct genl_info *info);
 int __tipc_nl_net_set(struct sk_buff *skb, struct genl_info *info);
+int tipc_nl_net_addr_legacy_get(struct sk_buff *skb, struct genl_info *info);
 
 #endif
diff --git a/net/tipc/netlink.c b/net/tipc/netlink.c
index e53231bd23b4..7c35094c20b8 100644
--- a/net/tipc/netlink.c
+++ b/net/tipc/netlink.c
@@ -83,6 +83,7 @@ const struct nla_policy tipc_nl_net_policy[TIPC_NLA_NET_MAX + 1] = {
 	[TIPC_NLA_NET_ADDR]		= { .type = NLA_U32 },
 	[TIPC_NLA_NET_NODEID]		= { .type = NLA_U64 },
 	[TIPC_NLA_NET_NODEID_W1]	= { .type = NLA_U64 },
+	[TIPC_NLA_NET_ADDR_LEGACY]	= { .type = NLA_FLAG }
 };
 
 const struct nla_policy tipc_nl_link_policy[TIPC_NLA_LINK_MAX + 1] = {
@@ -273,6 +274,11 @@ static const struct genl_ops tipc_genl_v2_ops[] = {
 		.doit	= tipc_nl_node_flush_key,
 	},
 #endif
+	{
+		.cmd	= TIPC_NL_ADDR_LEGACY_GET,
+		.validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP,
+		.doit	= tipc_nl_net_addr_legacy_get,
+	},
 };
 
 struct genl_family tipc_genl_family __ro_after_init = {
-- 
cgit v1.2.3-59-g8ed1b


From b2598318eb455c306e703ab30e18b88a1dc7df97 Mon Sep 17 00:00:00 2001
From: Yunsheng Lin <linyunsheng@huawei.com>
Date: Thu, 19 Dec 2019 14:57:40 +0800
Subject: net: hns3: check FE bit before calling hns3_add_frag()

A BD with FE bit means that it is the last BD of a packet,
currently the FE bit is checked before calling hns3_add_frag(),
which is unnecessary because the FE bit may have been checked
in some case.

This patch checks the FE bit before calling hns3_add_frag()
after processing the first BD of a SKB and adjust the location
of memcpy() to reduce duplication.

Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hns3_enet.c | 46 +++++++------------------
 1 file changed, 13 insertions(+), 33 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
index b113b895dab7..5679a0cef975 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
@@ -2788,7 +2788,6 @@ static bool hns3_parse_vlan_tag(struct hns3_enet_ring *ring,
 static int hns3_alloc_skb(struct hns3_enet_ring *ring, unsigned int length,
 			  unsigned char *va)
 {
-#define HNS3_NEED_ADD_FRAG	1
 	struct hns3_desc_cb *desc_cb = &ring->desc_cb[ring->next_to_clean];
 	struct net_device *netdev = ring_to_netdev(ring);
 	struct sk_buff *skb;
@@ -2832,33 +2831,19 @@ static int hns3_alloc_skb(struct hns3_enet_ring *ring, unsigned int length,
 			    desc_cb);
 	ring_ptr_move_fw(ring, next_to_clean);
 
-	return HNS3_NEED_ADD_FRAG;
+	return 0;
 }
 
-static int hns3_add_frag(struct hns3_enet_ring *ring, struct hns3_desc *desc,
-			 bool pending)
+static int hns3_add_frag(struct hns3_enet_ring *ring)
 {
 	struct sk_buff *skb = ring->skb;
 	struct sk_buff *head_skb = skb;
 	struct sk_buff *new_skb;
 	struct hns3_desc_cb *desc_cb;
-	struct hns3_desc *pre_desc;
+	struct hns3_desc *desc;
 	u32 bd_base_info;
-	int pre_bd;
-
-	/* if there is pending bd, the SW param next_to_clean has moved
-	 * to next and the next is NULL
-	 */
-	if (pending) {
-		pre_bd = (ring->next_to_clean - 1 + ring->desc_num) %
-			 ring->desc_num;
-		pre_desc = &ring->desc[pre_bd];
-		bd_base_info = le32_to_cpu(pre_desc->rx.bd_base_info);
-	} else {
-		bd_base_info = le32_to_cpu(desc->rx.bd_base_info);
-	}
 
-	while (!(bd_base_info & BIT(HNS3_RXD_FE_B))) {
+	do {
 		desc = &ring->desc[ring->next_to_clean];
 		desc_cb = &ring->desc_cb[ring->next_to_clean];
 		bd_base_info = le32_to_cpu(desc->rx.bd_base_info);
@@ -2895,7 +2880,7 @@ static int hns3_add_frag(struct hns3_enet_ring *ring, struct hns3_desc *desc,
 		hns3_nic_reuse_page(skb, ring->frag_num++, ring, 0, desc_cb);
 		ring_ptr_move_fw(ring, next_to_clean);
 		ring->pending_buf++;
-	}
+	} while (!(bd_base_info & BIT(HNS3_RXD_FE_B)));
 
 	return 0;
 }
@@ -3063,28 +3048,23 @@ static int hns3_handle_rx_bd(struct hns3_enet_ring *ring)
 
 		if (ret < 0) /* alloc buffer fail */
 			return ret;
-		if (ret > 0) { /* need add frag */
-			ret = hns3_add_frag(ring, desc, false);
+		if (!(bd_base_info & BIT(HNS3_RXD_FE_B))) { /* need add frag */
+			ret = hns3_add_frag(ring);
 			if (ret)
 				return ret;
-
-			/* As the head data may be changed when GRO enable, copy
-			 * the head data in after other data rx completed
-			 */
-			memcpy(skb->data, ring->va,
-			       ALIGN(ring->pull_len, sizeof(long)));
 		}
 	} else {
-		ret = hns3_add_frag(ring, desc, true);
+		ret = hns3_add_frag(ring);
 		if (ret)
 			return ret;
+	}
 
-		/* As the head data may be changed when GRO enable, copy
-		 * the head data in after other data rx completed
-		 */
+	/* As the head data may be changed when GRO enable, copy
+	 * the head data in after other data rx completed
+	 */
+	if (skb->len > HNS3_RX_HEAD_SIZE)
 		memcpy(skb->data, ring->va,
 		       ALIGN(ring->pull_len, sizeof(long)));
-	}
 
 	ret = hns3_handle_bdinfo(ring, skb);
 	if (unlikely(ret)) {
-- 
cgit v1.2.3-59-g8ed1b


From 31c92ccecaab75af4a746af6e071314bd3d8ded3 Mon Sep 17 00:00:00 2001
From: Huazhong Tan <tanhuazhong@huawei.com>
Date: Thu, 19 Dec 2019 14:57:41 +0800
Subject: net: hns3: remove useless mutex vport_cfg_mutex in the struct
 hclge_dev

Mutex vport_cfg_mutex has been used to protect uc_mac_list,
mc_mac_list and vlan_list from being modified by unloading
or reset task at the same time. But now unloading will
set up HCLGE_STATE_REMOVING flag and call cancel_work_sync to
break down this race condition, so this mutex is unnecessary.

Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 9 ---------
 drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h | 2 --
 drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c  | 2 --
 3 files changed, 13 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
index 4e7a07877e5d..76b4418d937c 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
@@ -7534,7 +7534,6 @@ void hclge_uninit_vport_mac_table(struct hclge_dev *hdev)
 	struct hclge_vport *vport;
 	int i;
 
-	mutex_lock(&hdev->vport_cfg_mutex);
 	for (i = 0; i < hdev->num_alloc_vport; i++) {
 		vport = &hdev->vport[i];
 		list_for_each_entry_safe(mac, tmp, &vport->uc_mac_list, node) {
@@ -7547,7 +7546,6 @@ void hclge_uninit_vport_mac_table(struct hclge_dev *hdev)
 			kfree(mac);
 		}
 	}
-	mutex_unlock(&hdev->vport_cfg_mutex);
 }
 
 static int hclge_get_mac_ethertype_cmd_status(struct hclge_dev *hdev,
@@ -8308,7 +8306,6 @@ void hclge_uninit_vport_vlan_table(struct hclge_dev *hdev)
 	struct hclge_vport *vport;
 	int i;
 
-	mutex_lock(&hdev->vport_cfg_mutex);
 	for (i = 0; i < hdev->num_alloc_vport; i++) {
 		vport = &hdev->vport[i];
 		list_for_each_entry_safe(vlan, tmp, &vport->vlan_list, node) {
@@ -8316,7 +8313,6 @@ void hclge_uninit_vport_vlan_table(struct hclge_dev *hdev)
 			kfree(vlan);
 		}
 	}
-	mutex_unlock(&hdev->vport_cfg_mutex);
 }
 
 static void hclge_restore_vlan_table(struct hnae3_handle *handle)
@@ -8328,7 +8324,6 @@ static void hclge_restore_vlan_table(struct hnae3_handle *handle)
 	u16 state, vlan_id;
 	int i;
 
-	mutex_lock(&hdev->vport_cfg_mutex);
 	for (i = 0; i < hdev->num_alloc_vport; i++) {
 		vport = &hdev->vport[i];
 		vlan_proto = vport->port_base_vlan_cfg.vlan_info.vlan_proto;
@@ -8354,8 +8349,6 @@ static void hclge_restore_vlan_table(struct hnae3_handle *handle)
 				break;
 		}
 	}
-
-	mutex_unlock(&hdev->vport_cfg_mutex);
 }
 
 int hclge_en_hw_strip_rxvtag(struct hnae3_handle *handle, bool enable)
@@ -9390,7 +9383,6 @@ static int hclge_init_ae_dev(struct hnae3_ae_dev *ae_dev)
 	hdev->mps = ETH_FRAME_LEN + ETH_FCS_LEN + 2 * VLAN_HLEN;
 
 	mutex_init(&hdev->vport_lock);
-	mutex_init(&hdev->vport_cfg_mutex);
 	spin_lock_init(&hdev->fd_rule_lock);
 
 	ret = hclge_pci_init(hdev);
@@ -9943,7 +9935,6 @@ static void hclge_uninit_ae_dev(struct hnae3_ae_dev *ae_dev)
 	mutex_destroy(&hdev->vport_lock);
 	hclge_uninit_vport_mac_table(hdev);
 	hclge_uninit_vport_vlan_table(hdev);
-	mutex_destroy(&hdev->vport_cfg_mutex);
 	ae_dev->priv = NULL;
 }
 
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h
index 3a913976b820..4e5cfda16eb7 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h
@@ -821,8 +821,6 @@ struct hclge_dev {
 	u16 share_umv_size;
 	struct mutex umv_mutex; /* protect share_umv_size */
 
-	struct mutex vport_cfg_mutex;   /* Protect stored vf table */
-
 	DECLARE_KFIFO(mac_tnl_log, struct hclge_mac_tnl_stats,
 		      HCLGE_MAC_TNL_LOG_SIZE);
 
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c
index a79d1512119d..f905dd3386b3 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c
@@ -797,13 +797,11 @@ void hclge_mbx_handler(struct hclge_dev *hdev)
 			hclge_get_link_mode(vport, req);
 			break;
 		case HCLGE_MBX_GET_VF_FLR_STATUS:
-			mutex_lock(&hdev->vport_cfg_mutex);
 			hclge_rm_vport_all_mac_table(vport, true,
 						     HCLGE_MAC_ADDR_UC);
 			hclge_rm_vport_all_mac_table(vport, true,
 						     HCLGE_MAC_ADDR_MC);
 			hclge_rm_vport_all_vlan_table(vport, true);
-			mutex_unlock(&hdev->vport_cfg_mutex);
 			break;
 		case HCLGE_MBX_GET_MEDIA_TYPE:
 			ret = hclge_get_vf_media_type(vport, req);
-- 
cgit v1.2.3-59-g8ed1b


From a3374d7d335c24225f7d65bc0abb632d81dacdf6 Mon Sep 17 00:00:00 2001
From: Huazhong Tan <tanhuazhong@huawei.com>
Date: Thu, 19 Dec 2019 14:57:42 +0800
Subject: net: hns3: optimization for CMDQ uninitialization

When uninitializing CMDQ, HCLGE_STATE_CMD_DISABLE will
be set up firstly, then the driver does not send command
anymore. So, hclge_free_cmd_desc can be called without
holding ring->lock. hclge_destroy_cmd_queue() and
hclge_destroy_queue() are unnecessary now, so removes them,
the VF driver has implemented currently.

BTW, the VF driver should set up HCLGEVF_STATE_CMD_DISABLE
as well in the hclgevf_cmd_uninit(), just likes what the PF
driver does.

Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.c   | 16 ++--------------
 drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.c |  2 +-
 2 files changed, 3 insertions(+), 15 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.c
index 940ead3970d1..7f509eff562e 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.c
@@ -479,19 +479,6 @@ static void hclge_cmd_uninit_regs(struct hclge_hw *hw)
 	hclge_write_dev(hw, HCLGE_NIC_CRQ_TAIL_REG, 0);
 }
 
-static void hclge_destroy_queue(struct hclge_cmq_ring *ring)
-{
-	spin_lock(&ring->lock);
-	hclge_free_cmd_desc(ring);
-	spin_unlock(&ring->lock);
-}
-
-static void hclge_destroy_cmd_queue(struct hclge_hw *hw)
-{
-	hclge_destroy_queue(&hw->cmq.csq);
-	hclge_destroy_queue(&hw->cmq.crq);
-}
-
 void hclge_cmd_uninit(struct hclge_dev *hdev)
 {
 	spin_lock_bh(&hdev->hw.cmq.csq.lock);
@@ -501,5 +488,6 @@ void hclge_cmd_uninit(struct hclge_dev *hdev)
 	spin_unlock(&hdev->hw.cmq.crq.lock);
 	spin_unlock_bh(&hdev->hw.cmq.csq.lock);
 
-	hclge_destroy_cmd_queue(&hdev->hw);
+	hclge_free_cmd_desc(&hdev->hw.cmq.csq);
+	hclge_free_cmd_desc(&hdev->hw.cmq.crq);
 }
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.c
index af2245e3bb95..f38d236ebf4f 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.c
@@ -443,7 +443,7 @@ void hclgevf_cmd_uninit(struct hclgevf_dev *hdev)
 {
 	spin_lock_bh(&hdev->hw.cmq.csq.lock);
 	spin_lock(&hdev->hw.cmq.crq.lock);
-	clear_bit(HCLGEVF_STATE_CMD_DISABLE, &hdev->state);
+	set_bit(HCLGEVF_STATE_CMD_DISABLE, &hdev->state);
 	hclgevf_cmd_uninit_regs(&hdev->hw);
 	spin_unlock(&hdev->hw.cmq.crq.lock);
 	spin_unlock_bh(&hdev->hw.cmq.csq.lock);
-- 
cgit v1.2.3-59-g8ed1b


From 44b6b88336a08be87a02d69f6a4754c86cda97f1 Mon Sep 17 00:00:00 2001
From: Yufeng Mo <moyufeng@huawei.com>
Date: Thu, 19 Dec 2019 14:57:43 +0800
Subject: net: hns3: get FD rules location before dump in debugfs

Currently, the dump FD tcam mode in debugfs is to query all FD tcams,
including empty rules, which is unnecessary. This patch modify to
find the position of useful rules before dump FD tcam, so that it does
not need to query empty rules.

This patch also modifies some help information in debugfs.

Signed-off-by: Yufeng Mo <moyufeng@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c | 74 ++++++++++++++++++++--
 1 file changed, 67 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c
index 92a99e4fdda5..f3d4cbd28913 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c
@@ -886,8 +886,8 @@ static void hclge_dbg_dump_mng_table(struct hclge_dev *hdev)
 	}
 }
 
-static void hclge_dbg_fd_tcam_read(struct hclge_dev *hdev, u8 stage,
-				   bool sel_x, u32 loc)
+static int hclge_dbg_fd_tcam_read(struct hclge_dev *hdev, u8 stage,
+				  bool sel_x, u32 loc)
 {
 	struct hclge_fd_tcam_config_1_cmd *req1;
 	struct hclge_fd_tcam_config_2_cmd *req2;
@@ -912,7 +912,7 @@ static void hclge_dbg_fd_tcam_read(struct hclge_dev *hdev, u8 stage,
 
 	ret = hclge_cmd_send(&hdev->hw, desc, 3);
 	if (ret)
-		return;
+		return ret;
 
 	dev_info(&hdev->pdev->dev, " read result tcam key %s(%u):\n",
 		 sel_x ? "x" : "y", loc);
@@ -931,16 +931,76 @@ static void hclge_dbg_fd_tcam_read(struct hclge_dev *hdev, u8 stage,
 	req = (u32 *)req3->tcam_data;
 	for (i = 0; i < 5; i++)
 		dev_info(&hdev->pdev->dev, "%08x\n", *req++);
+
+	return ret;
+}
+
+static int hclge_dbg_get_rules_location(struct hclge_dev *hdev, u16 *rule_locs)
+{
+	struct hclge_fd_rule *rule;
+	struct hlist_node *node;
+	int cnt = 0;
+
+	spin_lock_bh(&hdev->fd_rule_lock);
+	hlist_for_each_entry_safe(rule, node, &hdev->fd_rule_list, rule_node) {
+		rule_locs[cnt] = rule->location;
+		cnt++;
+	}
+	spin_unlock_bh(&hdev->fd_rule_lock);
+
+	if (cnt != hdev->hclge_fd_rule_num)
+		return -EINVAL;
+
+	return cnt;
 }
 
 static void hclge_dbg_fd_tcam(struct hclge_dev *hdev)
 {
-	u32 i;
+	int i, ret, rule_cnt;
+	u16 *rule_locs;
+
+	if (!hnae3_dev_fd_supported(hdev)) {
+		dev_err(&hdev->pdev->dev,
+			"Only FD-supported dev supports dump fd tcam\n");
+		return;
+	}
+
+	if (!hdev->hclge_fd_rule_num ||
+	    !hdev->fd_cfg.rule_num[HCLGE_FD_STAGE_1])
+		return;
+
+	rule_locs = kcalloc(hdev->fd_cfg.rule_num[HCLGE_FD_STAGE_1],
+			    sizeof(u16), GFP_KERNEL);
+	if (!rule_locs)
+		return;
 
-	for (i = 0; i < hdev->fd_cfg.rule_num[0]; i++) {
-		hclge_dbg_fd_tcam_read(hdev, 0, true, i);
-		hclge_dbg_fd_tcam_read(hdev, 0, false, i);
+	rule_cnt = hclge_dbg_get_rules_location(hdev, rule_locs);
+	if (rule_cnt <= 0) {
+		dev_err(&hdev->pdev->dev,
+			"failed to get rule number, ret = %d\n", rule_cnt);
+		kfree(rule_locs);
+		return;
 	}
+
+	for (i = 0; i < rule_cnt; i++) {
+		ret = hclge_dbg_fd_tcam_read(hdev, 0, true, rule_locs[i]);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"failed to get fd tcam key x, ret = %d\n", ret);
+			kfree(rule_locs);
+			return;
+		}
+
+		ret = hclge_dbg_fd_tcam_read(hdev, 0, false, rule_locs[i]);
+		if (ret) {
+			dev_err(&hdev->pdev->dev,
+				"failed to get fd tcam key y, ret = %d\n", ret);
+			kfree(rule_locs);
+			return;
+		}
+	}
+
+	kfree(rule_locs);
 }
 
 void hclge_dbg_dump_rst_info(struct hclge_dev *hdev)
-- 
cgit v1.2.3-59-g8ed1b


From 2a7556bb2b7308c6e0fe646c1ddbbf9522bf1120 Mon Sep 17 00:00:00 2001
From: Yunsheng Lin <linyunsheng@huawei.com>
Date: Thu, 19 Dec 2019 14:57:44 +0800
Subject: net: hns3: implement ndo_features_check ops for hns3 driver

The function netif_skb_features() will disable the TSO feature
by using dflt_features_check() if the driver does not implement
ndo_features_check ops, which may cause performance degradation
problem when hns3 hardware can do multiple tagged TSO.

Also, the HNS3 hardware only supports checksum on the SKB with
a max header len of 480 bytes, so remove the checksum and TSO
related features when the header len is over 480 bytes.

Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hns3_enet.c | 32 +++++++++++++++++++++++++
 1 file changed, 32 insertions(+)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
index 5679a0cef975..840a1fbc4d2e 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
@@ -1556,6 +1556,37 @@ static int hns3_nic_set_features(struct net_device *netdev,
 	return 0;
 }
 
+static netdev_features_t hns3_features_check(struct sk_buff *skb,
+					     struct net_device *dev,
+					     netdev_features_t features)
+{
+#define HNS3_MAX_HDR_LEN	480U
+#define HNS3_MAX_L4_HDR_LEN	60U
+
+	size_t len;
+
+	if (skb->ip_summed != CHECKSUM_PARTIAL)
+		return features;
+
+	if (skb->encapsulation)
+		len = skb_inner_transport_header(skb) - skb->data;
+	else
+		len = skb_transport_header(skb) - skb->data;
+
+	/* Assume L4 is 60 byte as TCP is the only protocol with a
+	 * a flexible value, and it's max len is 60 bytes.
+	 */
+	len += HNS3_MAX_L4_HDR_LEN;
+
+	/* Hardware only supports checksum on the skb with a max header
+	 * len of 480 bytes.
+	 */
+	if (len > HNS3_MAX_HDR_LEN)
+		features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+
+	return features;
+}
+
 static void hns3_nic_get_stats64(struct net_device *netdev,
 				 struct rtnl_link_stats64 *stats)
 {
@@ -1970,6 +2001,7 @@ static const struct net_device_ops hns3_nic_netdev_ops = {
 	.ndo_do_ioctl		= hns3_nic_do_ioctl,
 	.ndo_change_mtu		= hns3_nic_change_mtu,
 	.ndo_set_features	= hns3_nic_set_features,
+	.ndo_features_check	= hns3_features_check,
 	.ndo_get_stats64	= hns3_nic_get_stats64,
 	.ndo_setup_tc		= hns3_nic_setup_tc,
 	.ndo_set_rx_mode	= hns3_nic_set_rx_mode,
-- 
cgit v1.2.3-59-g8ed1b


From 89b40c7fe2a8b30ef6d46d45ae8455adcd9c1b47 Mon Sep 17 00:00:00 2001
From: Huazhong Tan <tanhuazhong@huawei.com>
Date: Thu, 19 Dec 2019 14:57:45 +0800
Subject: net: hns3: add some VF VLAN information for command "ip link show"

This patch adds some VF VLAN information for command "ip link show".

Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
index 76b4418d937c..d1aafea918b0 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
@@ -2944,6 +2944,9 @@ static int hclge_get_vf_config(struct hnae3_handle *handle, int vf,
 	ivf->trusted = vport->vf_info.trusted;
 	ivf->min_tx_rate = 0;
 	ivf->max_tx_rate = vport->vf_info.max_tx_rate;
+	ivf->vlan = vport->port_base_vlan_cfg.vlan_info.vlan_tag;
+	ivf->vlan_proto = htons(vport->port_base_vlan_cfg.vlan_info.vlan_proto);
+	ivf->qos = vport->port_base_vlan_cfg.vlan_info.qos;
 	ether_addr_copy(ivf->mac, vport->vf_info.mac);
 
 	return 0;
-- 
cgit v1.2.3-59-g8ed1b


From ff7dfcdd684b6277ed02c1025746a94cb1bfc397 Mon Sep 17 00:00:00 2001
From: Huazhong Tan <tanhuazhong@huawei.com>
Date: Thu, 19 Dec 2019 14:57:46 +0800
Subject: net: hns3: add a log for getting chain failure in
 hns3_nic_uninit_vector_data()

Since the mapping can be overwritten, when fail to get
the chain between vector and ring, we should go on to
deal with the remaining options. For debugging, this
patch adds log info for this failure.

Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hns3_enet.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
index 840a1fbc4d2e..aee5facc89b5 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
@@ -3602,7 +3602,12 @@ static void hns3_nic_uninit_vector_data(struct hns3_nic_priv *priv)
 		if (!tqp_vector->rx_group.ring && !tqp_vector->tx_group.ring)
 			continue;
 
-		hns3_get_vector_ring_chain(tqp_vector, &vector_ring_chain);
+		/* Since the mapping can be overwritten, when fail to get the
+		 * chain between vector and ring, we should go on to deal with
+		 * the remaining options.
+		 */
+		if (hns3_get_vector_ring_chain(tqp_vector, &vector_ring_chain))
+			dev_warn(priv->dev, "failed to get ring chain\n");
 
 		h->ae_algo->ops->unmap_ring_from_vector(h,
 			tqp_vector->vector_irq, &vector_ring_chain);
-- 
cgit v1.2.3-59-g8ed1b


From d9b81c963a4c4a0a6dd5d57c9d74d9dced77020d Mon Sep 17 00:00:00 2001
From: Huazhong Tan <tanhuazhong@huawei.com>
Date: Thu, 19 Dec 2019 14:57:47 +0800
Subject: net: hns3: only print misc interrupt status when handling fails

Printing misc interrupt status of hardware error event in the
IRQ handler is unnecessary, since hclge_handle_hw_msix_error()
will print out the detail information for this hardware error
when handling success. So, this patch removes the print in
IRQ handler, and prints it when hclge_handle_hw_msix_error()
fails.

Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
index d1aafea918b0..a510f005209c 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
@@ -3005,8 +3005,6 @@ static u32 hclge_check_event_cause(struct hclge_dev *hdev, u32 *clearval)
 
 	/* check for vector0 msix event source */
 	if (msix_src_reg & HCLGE_VECTOR0_REG_MSIX_MASK) {
-		dev_info(&hdev->pdev->dev, "received event 0x%x\n",
-			 msix_src_reg);
 		*clearval = msix_src_reg;
 		return HCLGE_VECTOR0_EVENT_ERR;
 	}
@@ -3505,10 +3503,15 @@ static enum hnae3_reset_type hclge_get_reset_level(struct hnae3_ae_dev *ae_dev,
 
 	/* first, resolve any unknown reset type to the known type(s) */
 	if (test_bit(HNAE3_UNKNOWN_RESET, addr)) {
+		u32 msix_sts_reg = hclge_read_dev(&hdev->hw,
+					HCLGE_VECTOR0_PF_OTHER_INT_STS_REG);
 		/* we will intentionally ignore any errors from this function
 		 *  as we will end up in *some* reset request in any case
 		 */
-		hclge_handle_hw_msix_error(hdev, addr);
+		if (hclge_handle_hw_msix_error(hdev, addr))
+			dev_info(&hdev->pdev->dev, "received msix interrupt 0x%x\n",
+				 msix_sts_reg);
+
 		clear_bit(HNAE3_UNKNOWN_RESET, addr);
 		/* We defered the clearing of the error event which caused
 		 * interrupt since it was not posssible to do that in
-- 
cgit v1.2.3-59-g8ed1b


From b59ec4cac5160370d908c60934e799bc6da7d784 Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Wed, 2 Oct 2019 12:20:32 +0300
Subject: iwlwifi: mvm: add support for location range request version 8

The new API requires the driver to set the frame format
(legacy, HT, VHT etc.) to be used for the measurement.

The new API also supports 11az and secured measurement, but
these are not supported by the driver for now.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
 .../net/wireless/intel/iwlwifi/fw/api/location.h   | 104 ++++++++-
 .../net/wireless/intel/iwlwifi/mvm/ftm-initiator.c | 239 ++++++++++++++++-----
 2 files changed, 290 insertions(+), 53 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/location.h b/drivers/net/wireless/intel/iwlwifi/fw/api/location.h
index 7a0fe5adefa5..a51fda4c9588 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/location.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/location.h
@@ -403,7 +403,7 @@ enum iwl_initiator_ap_flags {
 };
 
 /**
- * struct iwl_tof_range_req_ap_entry - AP configuration parameters
+ * struct iwl_tof_range_req_ap_entry_v3 - AP configuration parameters
  * @initiator_ap_flags: see &enum iwl_initiator_ap_flags.
  * @channel_num: AP Channel number
  * @bandwidth: AP bandwidth. One of iwl_tof_bandwidth.
@@ -420,7 +420,7 @@ enum iwl_initiator_ap_flags {
  * @reserved: For alignment and future use
  * @tsf_delta: not in use
  */
-struct iwl_tof_range_req_ap_entry {
+struct iwl_tof_range_req_ap_entry_v3 {
 	__le32 initiator_ap_flags;
 	u8 channel_num;
 	u8 bandwidth;
@@ -434,6 +434,72 @@ struct iwl_tof_range_req_ap_entry {
 	__le32 tsf_delta;
 } __packed; /* LOCATION_RANGE_REQ_AP_ENTRY_CMD_API_S_VER_3 */
 
+/**
+ * enum iwl_location_frame_format - location frame formats
+ * @IWL_LOCATION_FRAME_FORMAT_LEGACY: legacy
+ * @IWL_LOCATION_FRAME_FORMAT_HT: HT
+ * @IWL_LOCATION_FRAME_FORMAT_VHT: VHT
+ * @IWL_LOCATION_FRAME_FORMAT_HE: HE
+ */
+enum iwl_location_frame_format {
+	IWL_LOCATION_FRAME_FORMAT_LEGACY,
+	IWL_LOCATION_FRAME_FORMAT_HT,
+	IWL_LOCATION_FRAME_FORMAT_VHT,
+	IWL_LOCATION_FRAME_FORMAT_HE,
+};
+
+/**
+ * enum iwl_location_bw - location bandwidth selection
+ * @IWL_LOCATION_BW_20MHZ: 20MHz
+ * @IWL_LOCATION_BW_40MHZ: 40MHz
+ * @IWL_LOCATION_BW_80MHZ: 80MHz
+ */
+enum iwl_location_bw {
+	IWL_LOCATION_BW_20MHZ,
+	IWL_LOCATION_BW_40MHZ,
+	IWL_LOCATION_BW_80MHZ,
+};
+
+#define HLTK_11AZ_LEN	32
+#define TK_11AZ_LEN	32
+
+#define LOCATION_BW_POS	4
+
+/**
+ * struct iwl_tof_range_req_ap_entry - AP configuration parameters
+ * @initiator_ap_flags: see &enum iwl_initiator_ap_flags.
+ * @channel_num: AP Channel number
+ * @format_bw: bits 0 - 3: &enum iwl_location_frame_format.
+ *             bits 4 - 7: &enum iwl_location_bw.
+ * @ctrl_ch_position: Coding of the control channel position relative to the
+ *	center frequency, see iwl_mvm_get_ctrl_pos().
+ * @ftmr_max_retries: Max number of retries to send the FTMR in case of no
+ *	reply from the AP.
+ * @bssid: AP's BSSID
+ * @burst_period: Recommended value to be sent to the AP. Measurement
+ *	periodicity In units of 100ms. ignored if num_of_bursts_exp = 0
+ * @samples_per_burst: the number of FTMs pairs in single Burst (1-31);
+ * @num_of_bursts: Recommended value to be sent to the AP. 2s Exponent of
+ *	the number of measurement iterations (min 2^0 = 1, max 2^14)
+ * @reserved: For alignment and future use
+ * @hltk: HLTK to be used for secured 11az measurement
+ * @tk: TK to be used for secured 11az measurement
+ */
+struct iwl_tof_range_req_ap_entry {
+	__le32 initiator_ap_flags;
+	u8 channel_num;
+	u8 format_bw;
+	u8 ctrl_ch_position;
+	u8 ftmr_max_retries;
+	u8 bssid[ETH_ALEN];
+	__le16 burst_period;
+	u8 samples_per_burst;
+	u8 num_of_bursts;
+	__le16 reserved;
+	u8 hltk[HLTK_11AZ_LEN];
+	u8 tk[TK_11AZ_LEN];
+} __packed; /* LOCATION_RANGE_REQ_AP_ENTRY_CMD_API_S_VER_4 */
+
 /**
  * enum iwl_tof_response_mode
  * @IWL_MVM_TOF_RESPONSE_ASAP: report each AP measurement separately as soon as
@@ -535,6 +601,38 @@ struct iwl_tof_range_req_cmd_v5 {
 } __packed;
 /* LOCATION_RANGE_REQ_CMD_API_S_VER_5 */
 
+/**
+ * struct iwl_tof_range_req_cmd_v7 - start measurement cmd
+ * @initiator_flags: see flags @ iwl_tof_initiator_flags
+ * @request_id: A Token incremented per request. The same Token will be
+ *		sent back in the range response
+ * @num_of_ap: Number of APs to measure (error if > IWL_MVM_TOF_MAX_APS)
+ * @range_req_bssid: ranging request BSSID
+ * @macaddr_mask: Bits set to 0 shall be copied from the MAC address template.
+ *		  Bits set to 1 shall be randomized by the UMAC
+ * @macaddr_template: MAC address template to use for non-randomized bits
+ * @req_timeout_ms: Requested timeout of the response in units of milliseconds.
+ *	This is the session time for completing the measurement.
+ * @tsf_mac_id: report the measurement start time for each ap in terms of the
+ *	TSF of this mac id. 0xff to disable TSF reporting.
+ * @common_calib: The common calib value to inject to this measurement calc
+ * @specific_calib: The specific calib value to inject to this measurement calc
+ * @ap: per-AP request data, see &struct iwl_tof_range_req_ap_entry_v2.
+ */
+struct iwl_tof_range_req_cmd_v7 {
+	__le32 initiator_flags;
+	u8 request_id;
+	u8 num_of_ap;
+	u8 range_req_bssid[ETH_ALEN];
+	u8 macaddr_mask[ETH_ALEN];
+	u8 macaddr_template[ETH_ALEN];
+	__le32 req_timeout_ms;
+	__le32 tsf_mac_id;
+	__le16 common_calib;
+	__le16 specific_calib;
+	struct iwl_tof_range_req_ap_entry_v3 ap[IWL_MVM_TOF_MAX_APS];
+} __packed; /* LOCATION_RANGE_REQ_CMD_API_S_VER_7 */
+
 /**
  * struct iwl_tof_range_req_cmd - start measurement cmd
  * @initiator_flags: see flags @ iwl_tof_initiator_flags
@@ -565,7 +663,7 @@ struct iwl_tof_range_req_cmd {
 	__le16 common_calib;
 	__le16 specific_calib;
 	struct iwl_tof_range_req_ap_entry ap[IWL_MVM_TOF_MAX_APS];
-} __packed; /* LOCATION_RANGE_REQ_CMD_API_S_VER_7 */
+} __packed; /* LOCATION_RANGE_REQ_CMD_API_S_VER_8 */
 
 /*
  * enum iwl_tof_range_request_status - status of the sent request
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c b/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c
index 9f4b117db9d7..f783d6d53b6f 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c
@@ -208,10 +208,11 @@ static void iwl_mvm_ftm_cmd(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	cmd->tsf_mac_id = cpu_to_le32(0xff);
 }
 
-static int iwl_mvm_ftm_target_chandef(struct iwl_mvm *mvm,
-				      struct cfg80211_pmsr_request_peer *peer,
-				      u8 *channel, u8 *bandwidth,
-				      u8 *ctrl_ch_position)
+static int
+iwl_mvm_ftm_target_chandef_v1(struct iwl_mvm *mvm,
+			      struct cfg80211_pmsr_request_peer *peer,
+			      u8 *channel, u8 *bandwidth,
+			      u8 *ctrl_ch_position)
 {
 	u32 freq = peer->chandef.chan->center_freq;
 
@@ -242,6 +243,45 @@ static int iwl_mvm_ftm_target_chandef(struct iwl_mvm *mvm,
 	return 0;
 }
 
+static int
+iwl_mvm_ftm_target_chandef_v2(struct iwl_mvm *mvm,
+			      struct cfg80211_pmsr_request_peer *peer,
+			      u8 *channel, u8 *format_bw,
+			      u8 *ctrl_ch_position)
+{
+	u32 freq = peer->chandef.chan->center_freq;
+
+	*channel = ieee80211_frequency_to_channel(freq);
+
+	switch (peer->chandef.width) {
+	case NL80211_CHAN_WIDTH_20_NOHT:
+		*format_bw = IWL_LOCATION_FRAME_FORMAT_LEGACY;
+		*format_bw |= IWL_LOCATION_BW_20MHZ << LOCATION_BW_POS;
+		break;
+	case NL80211_CHAN_WIDTH_20:
+		*format_bw = IWL_LOCATION_FRAME_FORMAT_HT;
+		*format_bw |= IWL_LOCATION_BW_20MHZ << LOCATION_BW_POS;
+		break;
+	case NL80211_CHAN_WIDTH_40:
+		*format_bw = IWL_LOCATION_FRAME_FORMAT_HT;
+		*format_bw |= IWL_LOCATION_BW_40MHZ << LOCATION_BW_POS;
+		break;
+	case NL80211_CHAN_WIDTH_80:
+		*format_bw = IWL_LOCATION_FRAME_FORMAT_VHT;
+		*format_bw |= IWL_LOCATION_BW_80MHZ << LOCATION_BW_POS;
+		break;
+	default:
+		IWL_ERR(mvm, "Unsupported BW in FTM request (%d)\n",
+			peer->chandef.width);
+		return -EINVAL;
+	}
+
+	*ctrl_ch_position = (peer->chandef.width > NL80211_CHAN_WIDTH_20) ?
+		iwl_mvm_get_ctrl_pos(&peer->chandef) : 0;
+
+	return 0;
+}
+
 static int
 iwl_mvm_ftm_put_target_v2(struct iwl_mvm *mvm,
 			  struct cfg80211_pmsr_request_peer *peer,
@@ -249,9 +289,9 @@ iwl_mvm_ftm_put_target_v2(struct iwl_mvm *mvm,
 {
 	int ret;
 
-	ret = iwl_mvm_ftm_target_chandef(mvm, peer, &target->channel_num,
-					 &target->bandwidth,
-					 &target->ctrl_ch_position);
+	ret = iwl_mvm_ftm_target_chandef_v1(mvm, peer, &target->channel_num,
+					    &target->bandwidth,
+					    &target->ctrl_ch_position);
 	if (ret)
 		return ret;
 
@@ -278,18 +318,11 @@ iwl_mvm_ftm_put_target_v2(struct iwl_mvm *mvm,
 #define FTM_PUT_FLAG(flag)	(target->initiator_ap_flags |= \
 				 cpu_to_le32(IWL_INITIATOR_AP_FLAGS_##flag))
 
-static int iwl_mvm_ftm_put_target(struct iwl_mvm *mvm,
-				  struct cfg80211_pmsr_request_peer *peer,
-				  struct iwl_tof_range_req_ap_entry *target)
+static void
+iwl_mvm_ftm_put_target_common(struct iwl_mvm *mvm,
+			      struct cfg80211_pmsr_request_peer *peer,
+			      struct iwl_tof_range_req_ap_entry *target)
 {
-	int ret;
-
-	ret = iwl_mvm_ftm_target_chandef(mvm, peer, &target->channel_num,
-					 &target->bandwidth,
-					 &target->ctrl_ch_position);
-	if (ret)
-		return ret;
-
 	memcpy(target->bssid, peer->addr, ETH_ALEN);
 	target->burst_period =
 		cpu_to_le16(peer->ftm.burst_period);
@@ -314,60 +347,166 @@ static int iwl_mvm_ftm_put_target(struct iwl_mvm *mvm,
 		FTM_PUT_FLAG(ALGO_LR);
 	else if (IWL_MVM_FTM_INITIATOR_ALGO == IWL_TOF_ALGO_TYPE_FFT)
 		FTM_PUT_FLAG(ALGO_FFT);
+}
+
+static int
+iwl_mvm_ftm_put_target_v3(struct iwl_mvm *mvm,
+			  struct cfg80211_pmsr_request_peer *peer,
+			  struct iwl_tof_range_req_ap_entry_v3 *target)
+{
+	int ret;
+
+	ret = iwl_mvm_ftm_target_chandef_v1(mvm, peer, &target->channel_num,
+					    &target->bandwidth,
+					    &target->ctrl_ch_position);
+	if (ret)
+		return ret;
+
+	/*
+	 * Versions 3 and 4 has some common fields, so
+	 * iwl_mvm_ftm_put_target_common() can be used for version 7 too.
+	 */
+	iwl_mvm_ftm_put_target_common(mvm, peer, (void *)target);
 
 	return 0;
 }
 
-int iwl_mvm_ftm_start(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
-		      struct cfg80211_pmsr_request *req)
+static int iwl_mvm_ftm_put_target_v4(struct iwl_mvm *mvm,
+				     struct cfg80211_pmsr_request_peer *peer,
+				     struct iwl_tof_range_req_ap_entry *target)
+{
+	int ret;
+
+	ret = iwl_mvm_ftm_target_chandef_v2(mvm, peer, &target->channel_num,
+					    &target->format_bw,
+					    &target->ctrl_ch_position);
+	if (ret)
+		return ret;
+
+	iwl_mvm_ftm_put_target_common(mvm, peer, target);
+
+	return 0;
+}
+
+static int iwl_mvm_ftm_send_cmd(struct iwl_mvm *mvm, struct iwl_host_cmd *hcmd)
+{
+	u32 status;
+	int err = iwl_mvm_send_cmd_status(mvm, hcmd, &status);
+
+	if (!err && status) {
+		IWL_ERR(mvm, "FTM range request command failure, status: %u\n",
+			status);
+		err = iwl_ftm_range_request_status_to_err(status);
+	}
+
+	return err;
+}
+
+static int iwl_mvm_ftm_start_v5(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+				struct cfg80211_pmsr_request *req)
 {
 	struct iwl_tof_range_req_cmd_v5 cmd_v5;
-	struct iwl_tof_range_req_cmd cmd;
-	bool new_api = fw_has_api(&mvm->fw->ucode_capa,
-				  IWL_UCODE_TLV_API_FTM_NEW_RANGE_REQ);
-	u8 num_of_ap;
 	struct iwl_host_cmd hcmd = {
 		.id = iwl_cmd_id(TOF_RANGE_REQ_CMD, LOCATION_GROUP, 0),
 		.dataflags[0] = IWL_HCMD_DFL_DUP,
+		.data[0] = &cmd_v5,
+		.len[0] = sizeof(cmd_v5),
 	};
-	u32 status = 0;
-	int err, i;
+	u8 i;
+	int err;
 
-	lockdep_assert_held(&mvm->mutex);
+	iwl_mvm_ftm_cmd_v5(mvm, vif, &cmd_v5, req);
 
-	if (mvm->ftm_initiator.req)
-		return -EBUSY;
+	for (i = 0; i < cmd_v5.num_of_ap; i++) {
+		struct cfg80211_pmsr_request_peer *peer = &req->peers[i];
 
-	if (new_api) {
-		iwl_mvm_ftm_cmd(mvm, vif, &cmd, req);
-		hcmd.data[0] = &cmd;
-		hcmd.len[0] = sizeof(cmd);
-		num_of_ap = cmd.num_of_ap;
-	} else {
-		iwl_mvm_ftm_cmd_v5(mvm, vif, &cmd_v5, req);
-		hcmd.data[0] = &cmd_v5;
-		hcmd.len[0] = sizeof(cmd_v5);
-		num_of_ap = cmd_v5.num_of_ap;
+		err = iwl_mvm_ftm_put_target_v2(mvm, peer, &cmd_v5.ap[i]);
+		if (err)
+			return err;
 	}
 
-	for (i = 0; i < num_of_ap; i++) {
+	return iwl_mvm_ftm_send_cmd(mvm, &hcmd);
+}
+
+static int iwl_mvm_ftm_start_v7(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+				struct cfg80211_pmsr_request *req)
+{
+	struct iwl_tof_range_req_cmd_v7 cmd_v7;
+	struct iwl_host_cmd hcmd = {
+		.id = iwl_cmd_id(TOF_RANGE_REQ_CMD, LOCATION_GROUP, 0),
+		.dataflags[0] = IWL_HCMD_DFL_DUP,
+		.data[0] = &cmd_v7,
+		.len[0] = sizeof(cmd_v7),
+	};
+	u8 i;
+	int err;
+
+	/*
+	 * Versions 7 and 8 has the same structure except from the responders
+	 * list, so iwl_mvm_ftm_cmd() can be used for version 7 too.
+	 */
+	iwl_mvm_ftm_cmd(mvm, vif, (void *)&cmd_v7, req);
+
+	for (i = 0; i < cmd_v7.num_of_ap; i++) {
 		struct cfg80211_pmsr_request_peer *peer = &req->peers[i];
 
-		if (new_api)
-			err = iwl_mvm_ftm_put_target(mvm, peer, &cmd.ap[i]);
-		else
-			err = iwl_mvm_ftm_put_target_v2(mvm, peer,
-							&cmd_v5.ap[i]);
+		err = iwl_mvm_ftm_put_target_v3(mvm, peer, &cmd_v7.ap[i]);
+		if (err)
+			return err;
+	}
+
+	return iwl_mvm_ftm_send_cmd(mvm, &hcmd);
+}
 
+static int iwl_mvm_ftm_start_v8(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+				struct cfg80211_pmsr_request *req)
+{
+	struct iwl_tof_range_req_cmd cmd;
+	struct iwl_host_cmd hcmd = {
+		.id = iwl_cmd_id(TOF_RANGE_REQ_CMD, LOCATION_GROUP, 0),
+		.dataflags[0] = IWL_HCMD_DFL_DUP,
+		.data[0] = &cmd,
+		.len[0] = sizeof(cmd),
+	};
+	u8 i;
+	int err;
+
+	iwl_mvm_ftm_cmd(mvm, vif, &cmd, req);
+
+	for (i = 0; i < cmd.num_of_ap; i++) {
+		struct cfg80211_pmsr_request_peer *peer = &req->peers[i];
+
+		err = iwl_mvm_ftm_put_target_v4(mvm, peer, &cmd.ap[i]);
 		if (err)
 			return err;
 	}
 
-	err = iwl_mvm_send_cmd_status(mvm, &hcmd, &status);
-	if (!err && status) {
-		IWL_ERR(mvm, "FTM range request command failure, status: %u\n",
-			status);
-		err = iwl_ftm_range_request_status_to_err(status);
+	return iwl_mvm_ftm_send_cmd(mvm, &hcmd);
+}
+
+int iwl_mvm_ftm_start(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+		      struct cfg80211_pmsr_request *req)
+{
+	bool new_api = fw_has_api(&mvm->fw->ucode_capa,
+				  IWL_UCODE_TLV_API_FTM_NEW_RANGE_REQ);
+	int err;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	if (mvm->ftm_initiator.req)
+		return -EBUSY;
+
+	if (new_api) {
+		u8 cmd_ver = iwl_mvm_lookup_cmd_ver(mvm->fw, LOCATION_GROUP,
+						    TOF_RANGE_REQ_CMD);
+
+		if (cmd_ver == 8)
+			err = iwl_mvm_ftm_start_v8(mvm, vif, req);
+		else
+			err = iwl_mvm_ftm_start_v7(mvm, vif, req);
+
+	} else {
+		err = iwl_mvm_ftm_start_v5(mvm, vif, req);
 	}
 
 	if (!err) {
-- 
cgit v1.2.3-59-g8ed1b


From b1a6db1320e94bc8e924918fc236b5df97c1cca7 Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Wed, 2 Oct 2019 13:44:27 +0300
Subject: iwlwifi: mvm: add support for responder config command version 7

The new API requires the driver to config the supported frame format
(legacy, HT, VHT etc.).

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
 .../net/wireless/intel/iwlwifi/fw/api/location.h   | 40 ++++++++-
 .../net/wireless/intel/iwlwifi/mvm/ftm-responder.c | 95 +++++++++++++++++-----
 2 files changed, 113 insertions(+), 22 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/location.h b/drivers/net/wireless/intel/iwlwifi/fw/api/location.h
index a51fda4c9588..a0d6802c2715 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/location.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/location.h
@@ -240,7 +240,7 @@ enum iwl_tof_responder_cfg_flags {
 };
 
 /**
- * struct iwl_tof_responder_config_cmd - ToF AP mode (for debug)
+ * struct iwl_tof_responder_config_cmd_v6 - ToF AP mode (for debug)
  * @cmd_valid_fields: &iwl_tof_responder_cmd_valid_field
  * @responder_cfg_flags: &iwl_tof_responder_cfg_flags
  * @bandwidth: current AP Bandwidth: &enum iwl_tof_bandwidth
@@ -258,7 +258,7 @@ enum iwl_tof_responder_cfg_flags {
  * @bssid: Current AP BSSID
  * @reserved2: reserved
  */
-struct iwl_tof_responder_config_cmd {
+struct iwl_tof_responder_config_cmd_v6 {
 	__le32 cmd_valid_fields;
 	__le32 responder_cfg_flags;
 	u8 bandwidth;
@@ -274,6 +274,42 @@ struct iwl_tof_responder_config_cmd {
 	__le16 reserved2;
 } __packed; /* TOF_RESPONDER_CONFIG_CMD_API_S_VER_6 */
 
+/**
+ * struct iwl_tof_responder_config_cmd - ToF AP mode (for debug)
+ * @cmd_valid_fields: &iwl_tof_responder_cmd_valid_field
+ * @responder_cfg_flags: &iwl_tof_responder_cfg_flags
+ * @format_bw: bits 0 - 3: &enum iwl_location_frame_format.
+ *             bits 4 - 7: &enum iwl_location_bw.
+ * @rate: current AP rate
+ * @channel_num: current AP Channel
+ * @ctrl_ch_position: coding of the control channel position relative to
+ *	the center frequency, see iwl_mvm_get_ctrl_pos()
+ * @sta_id: index of the AP STA when in AP mode
+ * @reserved1: reserved
+ * @toa_offset: Artificial addition [pSec] for the ToA - to be used for debug
+ *	purposes, simulating station movement by adding various values
+ *	to this field
+ * @common_calib: XVT: common calibration value
+ * @specific_calib: XVT: specific calibration value
+ * @bssid: Current AP BSSID
+ * @reserved2: reserved
+ */
+struct iwl_tof_responder_config_cmd {
+	__le32 cmd_valid_fields;
+	__le32 responder_cfg_flags;
+	u8 format_bw;
+	u8 rate;
+	u8 channel_num;
+	u8 ctrl_ch_position;
+	u8 sta_id;
+	u8 reserved1;
+	__le16 toa_offset;
+	__le16 common_calib;
+	__le16 specific_calib;
+	u8 bssid[ETH_ALEN];
+	__le16 reserved2;
+} __packed; /* TOF_RESPONDER_CONFIG_CMD_API_S_VER_6 */
+
 #define IWL_LCI_CIVIC_IE_MAX_SIZE	400
 
 /**
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ftm-responder.c b/drivers/net/wireless/intel/iwlwifi/mvm/ftm-responder.c
index 1513b8b4062f..834564198409 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ftm-responder.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ftm-responder.c
@@ -6,7 +6,7 @@
  * GPL LICENSE SUMMARY
  *
  * Copyright(c) 2015 - 2017 Intel Deutschland GmbH
- * Copyright (C) 2018 Intel Corporation
+ * Copyright (C) 2018 - 2019 Intel Corporation
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -27,7 +27,7 @@
  * BSD LICENSE
  *
  * Copyright(c) 2015 - 2017 Intel Deutschland GmbH
- * Copyright (C) 2018 Intel Corporation
+ * Copyright (C) 2018 - 2019 Intel Corporation
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -62,12 +62,72 @@
 #include "mvm.h"
 #include "constants.h"
 
+static int iwl_mvm_ftm_responder_set_bw_v1(struct cfg80211_chan_def *chandef,
+					   u8 *bw, u8 *ctrl_ch_position)
+{
+	switch (chandef->width) {
+	case NL80211_CHAN_WIDTH_20_NOHT:
+		*bw = IWL_TOF_BW_20_LEGACY;
+		break;
+	case NL80211_CHAN_WIDTH_20:
+		*bw = IWL_TOF_BW_20_HT;
+		break;
+	case NL80211_CHAN_WIDTH_40:
+		*bw = IWL_TOF_BW_40;
+		*ctrl_ch_position = iwl_mvm_get_ctrl_pos(chandef);
+		break;
+	case NL80211_CHAN_WIDTH_80:
+		*bw = IWL_TOF_BW_80;
+		*ctrl_ch_position = iwl_mvm_get_ctrl_pos(chandef);
+		break;
+	default:
+		return -ENOTSUPP;
+	}
+
+	return 0;
+}
+
+static int iwl_mvm_ftm_responder_set_bw_v2(struct cfg80211_chan_def *chandef,
+					   u8 *format_bw,
+					   u8 *ctrl_ch_position)
+{
+	switch (chandef->width) {
+	case NL80211_CHAN_WIDTH_20_NOHT:
+		*format_bw = IWL_LOCATION_FRAME_FORMAT_LEGACY;
+		*format_bw |= IWL_LOCATION_BW_20MHZ << LOCATION_BW_POS;
+		break;
+	case NL80211_CHAN_WIDTH_20:
+		*format_bw = IWL_LOCATION_FRAME_FORMAT_HT;
+		*format_bw |= IWL_LOCATION_BW_20MHZ << LOCATION_BW_POS;
+		break;
+	case NL80211_CHAN_WIDTH_40:
+		*format_bw = IWL_LOCATION_FRAME_FORMAT_HT;
+		*format_bw |= IWL_LOCATION_BW_40MHZ << LOCATION_BW_POS;
+		*ctrl_ch_position = iwl_mvm_get_ctrl_pos(chandef);
+		break;
+	case NL80211_CHAN_WIDTH_80:
+		*format_bw = IWL_LOCATION_FRAME_FORMAT_VHT;
+		*format_bw |= IWL_LOCATION_BW_80MHZ << LOCATION_BW_POS;
+		*ctrl_ch_position = iwl_mvm_get_ctrl_pos(chandef);
+		break;
+	default:
+		return -ENOTSUPP;
+	}
+
+	return 0;
+}
+
 static int
 iwl_mvm_ftm_responder_cmd(struct iwl_mvm *mvm,
 			  struct ieee80211_vif *vif,
 			  struct cfg80211_chan_def *chandef)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	/*
+	 * The command structure is the same for versions 6 and 7, (only the
+	 * field interpretation is different), so the same struct can be use
+	 * for all cases.
+	 */
 	struct iwl_tof_responder_config_cmd cmd = {
 		.channel_num = chandef->chan->hw_value,
 		.cmd_valid_fields =
@@ -76,27 +136,22 @@ iwl_mvm_ftm_responder_cmd(struct iwl_mvm *mvm,
 				    IWL_TOF_RESPONDER_CMD_VALID_STA_ID),
 		.sta_id = mvmvif->bcast_sta.sta_id,
 	};
+	u8 cmd_ver = iwl_mvm_lookup_cmd_ver(mvm->fw, LOCATION_GROUP,
+					    TOF_RESPONDER_CONFIG_CMD);
+	int err;
 
 	lockdep_assert_held(&mvm->mutex);
 
-	switch (chandef->width) {
-	case NL80211_CHAN_WIDTH_20_NOHT:
-		cmd.bandwidth = IWL_TOF_BW_20_LEGACY;
-		break;
-	case NL80211_CHAN_WIDTH_20:
-		cmd.bandwidth = IWL_TOF_BW_20_HT;
-		break;
-	case NL80211_CHAN_WIDTH_40:
-		cmd.bandwidth = IWL_TOF_BW_40;
-		cmd.ctrl_ch_position = iwl_mvm_get_ctrl_pos(chandef);
-		break;
-	case NL80211_CHAN_WIDTH_80:
-		cmd.bandwidth = IWL_TOF_BW_80;
-		cmd.ctrl_ch_position = iwl_mvm_get_ctrl_pos(chandef);
-		break;
-	default:
-		WARN_ON(1);
-		return -EINVAL;
+	if (cmd_ver == 7)
+		err = iwl_mvm_ftm_responder_set_bw_v2(chandef, &cmd.format_bw,
+						      &cmd.ctrl_ch_position);
+	else
+		err = iwl_mvm_ftm_responder_set_bw_v1(chandef, &cmd.format_bw,
+						      &cmd.ctrl_ch_position);
+
+	if (err) {
+		IWL_ERR(mvm, "Failed to set responder bandwidth\n");
+		return err;
 	}
 
 	memcpy(cmd.bssid, vif->addr, ETH_ALEN);
-- 
cgit v1.2.3-59-g8ed1b


From c042f0c77f3df39ac4d530f5720ed64ff2f38375 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Fri, 27 Sep 2019 10:36:02 +0200
Subject: iwlwifi: allocate more receive buffers for HE devices

For HE-capable devices, we need to allocate more receive buffers as
there could be 256 frames aggregated into a single A-MPDU, and then
they might contain A-MSDUs as well. Until 22000 family, the devices
are able to put multiple frames into a single RB and the default RB
size is 4k, but starting from AX210 family this is no longer true.
On the other hand, those newer devices only use 2k receive buffers
(by default).

Modify the code and configuration to allocate an appropriate number
of RBs depending on the device capabilities:

 * 4096 for AX210 HE devices, which use 2k buffers by default,
 * 2048 for 22000 family devices which use 4k buffers by default,
 * 512 for existing 9000 family devices, which doesn't really
   change anything since that's the default before this patch,
 * 512 also for AX210/22000 family devices that don't do HE.

Theoretically, for devices lower than AX210, we wouldn't have to
allocate that many RBs if the RB size was manually increased, but
to support that the code got more complex, and it didn't really
seem necessary as that's a use case for monitor mode only, where
hopefully the wasted memory isn't really much of a concern.

Note that AX210 devices actually support bigger than 12-bit VID,
which is required here as we want to allocate 4096 buffers plus
some for quick recycling, so adjust the code for that as well.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/cfg/22000.c     | 55 ++++++++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/cfg/9000.c      |  1 +
 drivers/net/wireless/intel/iwlwifi/iwl-config.h    |  3 ++
 .../net/wireless/intel/iwlwifi/iwl-context-info.h  | 20 ++++----
 drivers/net/wireless/intel/iwlwifi/iwl-fh.h        |  5 +-
 .../wireless/intel/iwlwifi/pcie/ctxt-info-gen3.c   |  2 +-
 .../net/wireless/intel/iwlwifi/pcie/ctxt-info.c    | 11 +++--
 drivers/net/wireless/intel/iwlwifi/pcie/drv.c      | 13 +++++
 drivers/net/wireless/intel/iwlwifi/pcie/internal.h |  5 +-
 drivers/net/wireless/intel/iwlwifi/pcie/rx.c       | 34 ++++++++-----
 10 files changed, 116 insertions(+), 33 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
index 56dc335a788c..bd687e745dc9 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
@@ -236,10 +236,22 @@ static const struct iwl_ht_params iwl_22000_ht_params = {
 		},							\
 	}
 
+/*
+ * If the device doesn't support HE, no need to have that many buffers.
+ * 22000 devices can split multiple frames into a single RB, so fewer are
+ * needed; AX210 cannot (but use smaller RBs by default) - these sizes
+ * were picked according to 8 MSDUs inside 256 A-MSDUs in an A-MPDU, with
+ * additional overhead to account for processing time.
+ */
+#define IWL_NUM_RBDS_NON_HE		512
+#define IWL_NUM_RBDS_22000_HE		2048
+#define IWL_NUM_RBDS_AX210_HE		4096
+
 const struct iwl_cfg iwl22000_2ac_cfg_hr = {
 	.name = "Intel(R) Dual Band Wireless AC 22000",
 	.fw_name_pre = IWL_22000_HR_FW_PRE,
 	IWL_DEVICE_22500,
+	.num_rbds = IWL_NUM_RBDS_NON_HE,
 };
 
 const struct iwl_cfg iwl22000_2ac_cfg_hr_cdb = {
@@ -247,12 +259,14 @@ const struct iwl_cfg iwl22000_2ac_cfg_hr_cdb = {
 	.fw_name_pre = IWL_22000_HR_CDB_FW_PRE,
 	IWL_DEVICE_22500,
 	.cdb = true,
+	.num_rbds = IWL_NUM_RBDS_NON_HE,
 };
 
 const struct iwl_cfg iwl22000_2ac_cfg_jf = {
 	.name = "Intel(R) Dual Band Wireless AC 22000",
 	.fw_name_pre = IWL_22000_JF_FW_PRE,
 	IWL_DEVICE_22500,
+	.num_rbds = IWL_NUM_RBDS_NON_HE,
 };
 
 const struct iwl_cfg iwl_ax101_cfg_qu_hr = {
@@ -266,6 +280,7 @@ const struct iwl_cfg iwl_ax101_cfg_qu_hr = {
 	 */
 	.max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT,
 	.tx_with_siso_diversity = true,
+	.num_rbds = IWL_NUM_RBDS_22000_HE,
 };
 
 const struct iwl_cfg iwl_ax201_cfg_qu_hr = {
@@ -278,6 +293,7 @@ const struct iwl_cfg iwl_ax201_cfg_qu_hr = {
 	 * HT size; mac80211 would otherwise pick the HE max (256) by default.
 	 */
 	.max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT,
+	.num_rbds = IWL_NUM_RBDS_22000_HE,
 };
 
 const struct iwl_cfg iwl_ax101_cfg_qu_c0_hr_b0 = {
@@ -290,6 +306,7 @@ const struct iwl_cfg iwl_ax101_cfg_qu_c0_hr_b0 = {
 	 * HT size; mac80211 would otherwise pick the HE max (256) by default.
 	 */
 	.max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT,
+	.num_rbds = IWL_NUM_RBDS_22000_HE,
 };
 
 const struct iwl_cfg iwl_ax201_cfg_qu_c0_hr_b0 = {
@@ -302,6 +319,7 @@ const struct iwl_cfg iwl_ax201_cfg_qu_c0_hr_b0 = {
 	 * HT size; mac80211 would otherwise pick the HE max (256) by default.
 	 */
 	.max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT,
+	.num_rbds = IWL_NUM_RBDS_22000_HE,
 };
 
 const struct iwl_cfg iwl_ax101_cfg_quz_hr = {
@@ -314,6 +332,7 @@ const struct iwl_cfg iwl_ax101_cfg_quz_hr = {
 	 * HT size; mac80211 would otherwise pick the HE max (256) by default.
 	 */
 	.max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT,
+	.num_rbds = IWL_NUM_RBDS_22000_HE,
 };
 
 const struct iwl_cfg iwl_ax201_cfg_quz_hr = {
@@ -326,6 +345,7 @@ const struct iwl_cfg iwl_ax201_cfg_quz_hr = {
          * HT size; mac80211 would otherwise pick the HE max (256) by default.
          */
 	.max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT,
+	.num_rbds = IWL_NUM_RBDS_22000_HE,
 };
 
 const struct iwl_cfg iwl_ax1650s_cfg_quz_hr = {
@@ -338,6 +358,7 @@ const struct iwl_cfg iwl_ax1650s_cfg_quz_hr = {
          * HT size; mac80211 would otherwise pick the HE max (256) by default.
          */
 	.max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT,
+	.num_rbds = IWL_NUM_RBDS_22000_HE,
 };
 
 const struct iwl_cfg iwl_ax1650i_cfg_quz_hr = {
@@ -350,6 +371,7 @@ const struct iwl_cfg iwl_ax1650i_cfg_quz_hr = {
          * HT size; mac80211 would otherwise pick the HE max (256) by default.
          */
 	.max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT,
+	.num_rbds = IWL_NUM_RBDS_22000_HE,
 };
 
 const struct iwl_cfg iwl_ax200_cfg_cc = {
@@ -363,6 +385,7 @@ const struct iwl_cfg iwl_ax200_cfg_cc = {
 	 */
 	.max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT,
 	.trans.bisr_workaround = 1,
+	.num_rbds = IWL_NUM_RBDS_22000_HE,
 };
 
 const struct iwl_cfg killer1650x_2ax_cfg = {
@@ -376,6 +399,7 @@ const struct iwl_cfg killer1650x_2ax_cfg = {
 	 */
 	.max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT,
 	.trans.bisr_workaround = 1,
+	.num_rbds = IWL_NUM_RBDS_22000_HE,
 };
 
 const struct iwl_cfg killer1650w_2ax_cfg = {
@@ -389,6 +413,7 @@ const struct iwl_cfg killer1650w_2ax_cfg = {
 	 */
 	.max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT,
 	.trans.bisr_workaround = 1,
+	.num_rbds = IWL_NUM_RBDS_22000_HE,
 };
 
 /*
@@ -400,48 +425,56 @@ const struct iwl_cfg iwl9461_2ac_cfg_qu_b0_jf_b0 = {
 	.name = "Intel(R) Wireless-AC 9461",
 	.fw_name_pre = IWL_QU_B_JF_B_FW_PRE,
 	IWL_DEVICE_22500,
+	.num_rbds = IWL_NUM_RBDS_NON_HE,
 };
 
 const struct iwl_cfg iwl9462_2ac_cfg_qu_b0_jf_b0 = {
 	.name = "Intel(R) Wireless-AC 9462",
 	.fw_name_pre = IWL_QU_B_JF_B_FW_PRE,
 	IWL_DEVICE_22500,
+	.num_rbds = IWL_NUM_RBDS_NON_HE,
 };
 
 const struct iwl_cfg iwl9560_2ac_cfg_qu_b0_jf_b0 = {
 	.name = "Intel(R) Wireless-AC 9560",
 	.fw_name_pre = IWL_QU_B_JF_B_FW_PRE,
 	IWL_DEVICE_22500,
+	.num_rbds = IWL_NUM_RBDS_NON_HE,
 };
 
 const struct iwl_cfg iwl9560_2ac_160_cfg_qu_b0_jf_b0 = {
 	.name = "Intel(R) Wireless-AC 9560 160MHz",
 	.fw_name_pre = IWL_QU_B_JF_B_FW_PRE,
 	IWL_DEVICE_22500,
+	.num_rbds = IWL_NUM_RBDS_NON_HE,
 };
 
 const struct iwl_cfg iwl9461_2ac_cfg_qu_c0_jf_b0 = {
 	.name = "Intel(R) Wireless-AC 9461",
 	.fw_name_pre = IWL_QU_C_JF_B_FW_PRE,
 	IWL_DEVICE_22500,
+	.num_rbds = IWL_NUM_RBDS_NON_HE,
 };
 
 const struct iwl_cfg iwl9462_2ac_cfg_qu_c0_jf_b0 = {
 	.name = "Intel(R) Wireless-AC 9462",
 	.fw_name_pre = IWL_QU_C_JF_B_FW_PRE,
 	IWL_DEVICE_22500,
+	.num_rbds = IWL_NUM_RBDS_NON_HE,
 };
 
 const struct iwl_cfg iwl9560_2ac_cfg_qu_c0_jf_b0 = {
 	.name = "Intel(R) Wireless-AC 9560",
 	.fw_name_pre = IWL_QU_C_JF_B_FW_PRE,
 	IWL_DEVICE_22500,
+	.num_rbds = IWL_NUM_RBDS_NON_HE,
 };
 
 const struct iwl_cfg iwl9560_2ac_160_cfg_qu_c0_jf_b0 = {
 	.name = "Intel(R) Wireless-AC 9560 160MHz",
 	.fw_name_pre = IWL_QU_C_JF_B_FW_PRE,
 	IWL_DEVICE_22500,
+	.num_rbds = IWL_NUM_RBDS_NON_HE,
 };
 
 const struct iwl_cfg iwl9560_2ac_cfg_qnj_jf_b0 = {
@@ -454,6 +487,7 @@ const struct iwl_cfg iwl9560_2ac_cfg_qnj_jf_b0 = {
 	 * HT size; mac80211 would otherwise pick the HE max (256) by default.
 	 */
 	.max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT,
+	.num_rbds = IWL_NUM_RBDS_NON_HE,
 };
 
 const struct iwl_cfg iwl9560_2ac_cfg_quz_a0_jf_b0_soc = {
@@ -468,6 +502,7 @@ const struct iwl_cfg iwl9560_2ac_cfg_quz_a0_jf_b0_soc = {
 	.max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT,
 	.integrated = true,
 	.soc_latency = 5000,
+	.num_rbds = IWL_NUM_RBDS_NON_HE,
 };
 
 const struct iwl_cfg iwl9560_2ac_160_cfg_quz_a0_jf_b0_soc = {
@@ -482,6 +517,7 @@ const struct iwl_cfg iwl9560_2ac_160_cfg_quz_a0_jf_b0_soc = {
 	.max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT,
 	.integrated = true,
 	.soc_latency = 5000,
+	.num_rbds = IWL_NUM_RBDS_NON_HE,
 };
 
 const struct iwl_cfg iwl9461_2ac_cfg_quz_a0_jf_b0_soc = {
@@ -496,6 +532,7 @@ const struct iwl_cfg iwl9461_2ac_cfg_quz_a0_jf_b0_soc = {
 	.max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT,
 	.integrated = true,
 	.soc_latency = 5000,
+	.num_rbds = IWL_NUM_RBDS_NON_HE,
 };
 
 const struct iwl_cfg iwl9462_2ac_cfg_quz_a0_jf_b0_soc = {
@@ -510,6 +547,7 @@ const struct iwl_cfg iwl9462_2ac_cfg_quz_a0_jf_b0_soc = {
 	.max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT,
 	.integrated = true,
 	.soc_latency = 5000,
+	.num_rbds = IWL_NUM_RBDS_NON_HE,
 };
 
 const struct iwl_cfg iwl9560_killer_s_2ac_cfg_quz_a0_jf_b0_soc = {
@@ -524,6 +562,7 @@ const struct iwl_cfg iwl9560_killer_s_2ac_cfg_quz_a0_jf_b0_soc = {
 	.max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT,
 	.integrated = true,
 	.soc_latency = 5000,
+	.num_rbds = IWL_NUM_RBDS_NON_HE,
 };
 
 const struct iwl_cfg iwl9560_killer_i_2ac_cfg_quz_a0_jf_b0_soc = {
@@ -538,18 +577,21 @@ const struct iwl_cfg iwl9560_killer_i_2ac_cfg_quz_a0_jf_b0_soc = {
 	.max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT,
 	.integrated = true,
 	.soc_latency = 5000,
+	.num_rbds = IWL_NUM_RBDS_NON_HE,
 };
 
 const struct iwl_cfg killer1550i_2ac_cfg_qu_b0_jf_b0 = {
 	.name = "Killer (R) Wireless-AC 1550i Wireless Network Adapter (9560NGW)",
 	.fw_name_pre = IWL_QU_B_JF_B_FW_PRE,
 	IWL_DEVICE_22500,
+	.num_rbds = IWL_NUM_RBDS_NON_HE,
 };
 
 const struct iwl_cfg killer1550s_2ac_cfg_qu_b0_jf_b0 = {
 	.name = "Killer (R) Wireless-AC 1550s Wireless Network Adapter (9560NGW)",
 	.fw_name_pre = IWL_QU_B_JF_B_FW_PRE,
 	IWL_DEVICE_22500,
+	.num_rbds = IWL_NUM_RBDS_NON_HE,
 };
 
 const struct iwl_cfg killer1650s_2ax_cfg_qu_b0_hr_b0 = {
@@ -562,6 +604,7 @@ const struct iwl_cfg killer1650s_2ax_cfg_qu_b0_hr_b0 = {
 	 * HT size; mac80211 would otherwise pick the HE max (256) by default.
 	 */
 	.max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT,
+	.num_rbds = IWL_NUM_RBDS_22000_HE,
 };
 
 const struct iwl_cfg killer1650i_2ax_cfg_qu_b0_hr_b0 = {
@@ -574,6 +617,7 @@ const struct iwl_cfg killer1650i_2ax_cfg_qu_b0_hr_b0 = {
 	 * HT size; mac80211 would otherwise pick the HE max (256) by default.
 	 */
 	.max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT,
+	.num_rbds = IWL_NUM_RBDS_22000_HE,
 };
 
 const struct iwl_cfg killer1650s_2ax_cfg_qu_c0_hr_b0 = {
@@ -586,6 +630,7 @@ const struct iwl_cfg killer1650s_2ax_cfg_qu_c0_hr_b0 = {
 	 * HT size; mac80211 would otherwise pick the HE max (256) by default.
 	 */
 	.max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT,
+	.num_rbds = IWL_NUM_RBDS_22000_HE,
 };
 
 const struct iwl_cfg killer1650i_2ax_cfg_qu_c0_hr_b0 = {
@@ -598,6 +643,7 @@ const struct iwl_cfg killer1650i_2ax_cfg_qu_c0_hr_b0 = {
 	 * HT size; mac80211 would otherwise pick the HE max (256) by default.
 	 */
 	.max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT,
+	.num_rbds = IWL_NUM_RBDS_22000_HE,
 };
 
 const struct iwl_cfg iwl22000_2ax_cfg_jf = {
@@ -610,6 +656,7 @@ const struct iwl_cfg iwl22000_2ax_cfg_jf = {
 	 * HT size; mac80211 would otherwise pick the HE max (256) by default.
 	 */
 	.max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT,
+	.num_rbds = IWL_NUM_RBDS_22000_HE,
 };
 
 const struct iwl_cfg iwl22000_2ax_cfg_qnj_hr_a0_f0 = {
@@ -622,6 +669,7 @@ const struct iwl_cfg iwl22000_2ax_cfg_qnj_hr_a0_f0 = {
 	 * HT size; mac80211 would otherwise pick the HE max (256) by default.
 	 */
 	.max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT,
+	.num_rbds = IWL_NUM_RBDS_22000_HE,
 };
 
 const struct iwl_cfg iwl22000_2ax_cfg_qnj_hr_b0 = {
@@ -634,6 +682,7 @@ const struct iwl_cfg iwl22000_2ax_cfg_qnj_hr_b0 = {
 	 * HT size; mac80211 would otherwise pick the HE max (256) by default.
 	 */
 	.max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT,
+	.num_rbds = IWL_NUM_RBDS_22000_HE,
 };
 
 const struct iwl_cfg iwl22000_2ax_cfg_qnj_hr_a0 = {
@@ -646,18 +695,21 @@ const struct iwl_cfg iwl22000_2ax_cfg_qnj_hr_a0 = {
 	 * HT size; mac80211 would otherwise pick the HE max (256) by default.
 	 */
 	.max_tx_agg_size = IEEE80211_MAX_AMPDU_BUF_HT,
+	.num_rbds = IWL_NUM_RBDS_22000_HE,
 };
 
 const struct iwl_cfg iwlax210_2ax_cfg_so_jf_a0 = {
 	.name = "Intel(R) Wireless-AC 9560 160MHz",
 	.fw_name_pre = IWL_22000_SO_A_JF_B_FW_PRE,
 	IWL_DEVICE_AX210,
+	.num_rbds = IWL_NUM_RBDS_NON_HE,
 };
 
 const struct iwl_cfg iwlax210_2ax_cfg_so_hr_a0 = {
 	.name = "Intel(R) Wi-Fi 7 AX210 160MHz",
 	.fw_name_pre = IWL_22000_SO_A_HR_B_FW_PRE,
 	IWL_DEVICE_AX210,
+	.num_rbds = IWL_NUM_RBDS_AX210_HE,
 };
 
 const struct iwl_cfg iwlax211_2ax_cfg_so_gf_a0 = {
@@ -665,6 +717,7 @@ const struct iwl_cfg iwlax211_2ax_cfg_so_gf_a0 = {
 	.fw_name_pre = IWL_22000_SO_A_GF_A_FW_PRE,
 	.uhb_supported = true,
 	IWL_DEVICE_AX210,
+	.num_rbds = IWL_NUM_RBDS_AX210_HE,
 };
 
 const struct iwl_cfg iwlax210_2ax_cfg_ty_gf_a0 = {
@@ -672,12 +725,14 @@ const struct iwl_cfg iwlax210_2ax_cfg_ty_gf_a0 = {
 	.fw_name_pre = IWL_22000_TY_A_GF_A_FW_PRE,
 	.uhb_supported = true,
 	IWL_DEVICE_AX210,
+	.num_rbds = IWL_NUM_RBDS_AX210_HE,
 };
 
 const struct iwl_cfg iwlax411_2ax_cfg_so_gf4_a0 = {
 	.name = "Intel(R) Wi-Fi 7 AX411 160MHz",
 	.fw_name_pre = IWL_22000_SO_A_GF4_A_FW_PRE,
 	IWL_DEVICE_AX210,
+	.num_rbds = IWL_NUM_RBDS_AX210_HE,
 };
 
 MODULE_FIRMWARE(IWL_22000_HR_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/9000.c b/drivers/net/wireless/intel/iwlwifi/cfg/9000.c
index e9155b9b5ee4..ac05c1a6ec6b 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/9000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/9000.c
@@ -138,6 +138,7 @@ static const struct iwl_tt_params iwl9000_tt_params = {
 	.thermal_params = &iwl9000_tt_params,				\
 	.apmg_not_supported = true,					\
 	.trans.mq_rx_supported = true,					\
+	.num_rbds = 512,						\
 	.vht_mu_mimo_supported = true,					\
 	.mac_addr_from_csr = true,					\
 	.trans.rf_id = true,						\
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
index 317eac066082..f8a93324d7c0 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
@@ -431,6 +431,8 @@ struct iwl_fw_mon_regs {
  * @uhb_supported: ultra high band channels supported
  * @min_256_ba_txq_size: minimum number of slots required in a TX queue which
  *	supports 256 BA aggregation
+ * @num_rbds: number of receive buffer descriptors to use
+ *	(only used for multi-queue capable devices)
  *
  * We enable the driver to be backward compatible wrt. hardware features.
  * API differences in uCode shouldn't be handled here but through TLVs
@@ -485,6 +487,7 @@ struct iwl_cfg {
 	u8 max_vht_ampdu_exponent;
 	u8 ucode_api_max;
 	u8 ucode_api_min;
+	u16 num_rbds;
 	u32 min_umac_error_event_table;
 	u32 extra_phy_cfg_flags;
 	u32 d3_debug_data_base_addr;
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-context-info.h b/drivers/net/wireless/intel/iwlwifi/iwl-context-info.h
index 5ed07e37e3ee..eeaa8cbdddce 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-context-info.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-context-info.h
@@ -6,7 +6,7 @@
  * GPL LICENSE SUMMARY
  *
  * Copyright(c) 2017 Intel Deutschland GmbH
- * Copyright(c) 2018 Intel Corporation
+ * Copyright(c) 2018 - 2019 Intel Corporation
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -20,7 +20,7 @@
  * BSD LICENSE
  *
  * Copyright(c) 2017 Intel Deutschland GmbH
- * Copyright(c) 2018 Intel Corporation
+ * Copyright(c) 2018 - 2019 Intel Corporation
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -64,12 +64,12 @@
  *	the init done for driver command that configures several system modes
  * @IWL_CTXT_INFO_EARLY_DEBUG: enable early debug
  * @IWL_CTXT_INFO_ENABLE_CDMP: enable core dump
- * @IWL_CTXT_INFO_RB_CB_SIZE_POS: position of the RBD Cyclic Buffer Size
+ * @IWL_CTXT_INFO_RB_CB_SIZE: mask of the RBD Cyclic Buffer Size
  *	exponent, the actual size is 2**value, valid sizes are 8-2048.
  *	The value is four bits long. Maximum valid exponent is 12
  * @IWL_CTXT_INFO_TFD_FORMAT_LONG: use long TFD Format (the
  *	default is short format - not supported by the driver)
- * @IWL_CTXT_INFO_RB_SIZE_POS: RB size position
+ * @IWL_CTXT_INFO_RB_SIZE: RB size mask
  *	(values are IWL_CTXT_INFO_RB_SIZE_*K)
  * @IWL_CTXT_INFO_RB_SIZE_1K: Value for 1K RB size
  * @IWL_CTXT_INFO_RB_SIZE_2K: Value for 2K RB size
@@ -83,12 +83,12 @@
  * @IWL_CTXT_INFO_RB_SIZE_32K: Value for 32K RB size
  */
 enum iwl_context_info_flags {
-	IWL_CTXT_INFO_AUTO_FUNC_INIT	= BIT(0),
-	IWL_CTXT_INFO_EARLY_DEBUG	= BIT(1),
-	IWL_CTXT_INFO_ENABLE_CDMP	= BIT(2),
-	IWL_CTXT_INFO_RB_CB_SIZE_POS	= 4,
-	IWL_CTXT_INFO_TFD_FORMAT_LONG	= BIT(8),
-	IWL_CTXT_INFO_RB_SIZE_POS	= 9,
+	IWL_CTXT_INFO_AUTO_FUNC_INIT	= 0x0001,
+	IWL_CTXT_INFO_EARLY_DEBUG	= 0x0002,
+	IWL_CTXT_INFO_ENABLE_CDMP	= 0x0004,
+	IWL_CTXT_INFO_RB_CB_SIZE	= 0x00f0,
+	IWL_CTXT_INFO_TFD_FORMAT_LONG	= 0x0100,
+	IWL_CTXT_INFO_RB_SIZE		= 0x1e00,
 	IWL_CTXT_INFO_RB_SIZE_1K	= 0x1,
 	IWL_CTXT_INFO_RB_SIZE_2K	= 0x2,
 	IWL_CTXT_INFO_RB_SIZE_4K	= 0x4,
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-fh.h b/drivers/net/wireless/intel/iwlwifi/iwl-fh.h
index 8836f85afe85..bf673ce5f183 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-fh.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-fh.h
@@ -611,10 +611,7 @@ static inline unsigned int FH_MEM_CBBC_QUEUE(struct iwl_trans *trans,
  */
 #define FH_TX_CHICKEN_BITS_SCD_AUTO_RETRY_EN	(0x00000002)
 
-#define MQ_RX_TABLE_SIZE	512
-#define MQ_RX_TABLE_MASK	(MQ_RX_TABLE_SIZE - 1)
-#define MQ_RX_NUM_RBDS		(MQ_RX_TABLE_SIZE - 1)
-#define RX_POOL_SIZE		(MQ_RX_NUM_RBDS +	\
+#define RX_POOL_SIZE(rbds)	((rbds) - 1 +	\
 				 IWL_MAX_RX_HW_QUEUES *	\
 				 (RX_CLAIM_REQ_ALLOC - RX_POST_REQ_ALLOC))
 /* cb size is the exponent */
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/ctxt-info-gen3.c b/drivers/net/wireless/intel/iwlwifi/pcie/ctxt-info-gen3.c
index a4e09a5b1816..01f248ba8fec 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/ctxt-info-gen3.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/ctxt-info-gen3.c
@@ -206,7 +206,7 @@ int iwl_pcie_ctxt_info_gen3_init(struct iwl_trans *trans,
 	ctxt_info_gen3->mtr_size =
 		cpu_to_le16(TFD_QUEUE_CB_SIZE(cmdq_size));
 	ctxt_info_gen3->mcr_size =
-		cpu_to_le16(RX_QUEUE_CB_SIZE(MQ_RX_TABLE_SIZE));
+		cpu_to_le16(RX_QUEUE_CB_SIZE(trans->cfg->num_rbds));
 
 	trans_pcie->ctxt_info_gen3 = ctxt_info_gen3;
 	trans_pcie->prph_info = prph_info;
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/ctxt-info.c b/drivers/net/wireless/intel/iwlwifi/pcie/ctxt-info.c
index d38cefbb779e..c34451e7e7ef 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/ctxt-info.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/ctxt-info.c
@@ -193,11 +193,12 @@ int iwl_pcie_ctxt_info_init(struct iwl_trans *trans,
 		rb_size = IWL_CTXT_INFO_RB_SIZE_4K;
 	}
 
-	BUILD_BUG_ON(RX_QUEUE_CB_SIZE(MQ_RX_TABLE_SIZE) > 0xF);
-	control_flags = IWL_CTXT_INFO_TFD_FORMAT_LONG |
-			(RX_QUEUE_CB_SIZE(MQ_RX_TABLE_SIZE) <<
-			 IWL_CTXT_INFO_RB_CB_SIZE_POS) |
-			(rb_size << IWL_CTXT_INFO_RB_SIZE_POS);
+	WARN_ON(RX_QUEUE_CB_SIZE(trans->cfg->num_rbds) > 12);
+	control_flags = IWL_CTXT_INFO_TFD_FORMAT_LONG;
+	control_flags |=
+		u32_encode_bits(RX_QUEUE_CB_SIZE(trans->cfg->num_rbds),
+				IWL_CTXT_INFO_RB_CB_SIZE);
+	control_flags |= u32_encode_bits(rb_size, IWL_CTXT_INFO_RB_SIZE);
 	ctxt_info->control.control_flags = cpu_to_le32(control_flags);
 
 	/* initialize RX default queue */
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
index 040cec17d3ad..cdf112d6c0f9 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
@@ -995,6 +995,7 @@ static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	const struct iwl_cfg *cfg = (struct iwl_cfg *)(ent->driver_data);
 	const struct iwl_cfg *cfg_7265d __maybe_unused = NULL;
 	struct iwl_trans *iwl_trans;
+	struct iwl_trans_pcie *trans_pcie;
 	unsigned long flags;
 	int ret;
 
@@ -1002,6 +1003,8 @@ static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (IS_ERR(iwl_trans))
 		return PTR_ERR(iwl_trans);
 
+	trans_pcie = IWL_TRANS_GET_PCIE_TRANS(iwl_trans);
+
 	/* the trans_cfg should never change, so set it now */
 	iwl_trans->trans_cfg = &cfg->trans;
 
@@ -1129,6 +1132,16 @@ static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	/* now set the real cfg we decided to use */
 	iwl_trans->cfg = cfg;
 
+	if (cfg->trans.mq_rx_supported) {
+		if (WARN_ON(!cfg->num_rbds)) {
+			ret = -EINVAL;
+			goto out_free_trans;
+		}
+		trans_pcie->num_rx_bufs = cfg->num_rbds;
+	} else {
+		trans_pcie->num_rx_bufs = RX_QUEUE_SIZE;
+	}
+
 	if (iwl_trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_8000 &&
 	    iwl_trans_grab_nic_access(iwl_trans, &flags)) {
 		u32 hw_step;
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
index a091690f6c79..c45df9a48559 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
@@ -513,8 +513,8 @@ struct cont_rec {
  */
 struct iwl_trans_pcie {
 	struct iwl_rxq *rxq;
-	struct iwl_rx_mem_buffer rx_pool[RX_POOL_SIZE];
-	struct iwl_rx_mem_buffer *global_table[RX_POOL_SIZE];
+	struct iwl_rx_mem_buffer *rx_pool;
+	struct iwl_rx_mem_buffer **global_table;
 	struct iwl_rb_allocator rba;
 	union {
 		struct iwl_context_info *ctxt_info;
@@ -573,6 +573,7 @@ struct iwl_trans_pcie {
 	u8 no_reclaim_cmds[MAX_NO_RECLAIM_CMDS];
 	u8 max_tbs;
 	u16 tfd_size;
+	u16 num_rx_bufs;
 
 	enum iwl_amsdu_size rx_buf_size;
 	bool bc_table_dword;
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
index 452da44a21e0..f03d3cfb50da 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
@@ -510,7 +510,7 @@ void iwl_pcie_free_rbs_pool(struct iwl_trans *trans)
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	int i;
 
-	for (i = 0; i < RX_POOL_SIZE; i++) {
+	for (i = 0; i < RX_POOL_SIZE(trans_pcie->num_rx_bufs); i++) {
 		if (!trans_pcie->rx_pool[i].page)
 			continue;
 		dma_unmap_page(trans->dev, trans_pcie->rx_pool[i].page_dma,
@@ -738,7 +738,7 @@ static int iwl_pcie_alloc_rxq_dma(struct iwl_trans *trans,
 
 	spin_lock_init(&rxq->lock);
 	if (trans->trans_cfg->mq_rx_supported)
-		rxq->queue_size = MQ_RX_TABLE_SIZE;
+		rxq->queue_size = trans->cfg->num_rbds;
 	else
 		rxq->queue_size = RX_QUEUE_SIZE;
 
@@ -807,8 +807,18 @@ static int iwl_pcie_rx_alloc(struct iwl_trans *trans)
 
 	trans_pcie->rxq = kcalloc(trans->num_rx_queues, sizeof(struct iwl_rxq),
 				  GFP_KERNEL);
-	if (!trans_pcie->rxq)
-		return -ENOMEM;
+	trans_pcie->rx_pool = kcalloc(RX_POOL_SIZE(trans_pcie->num_rx_bufs),
+				      sizeof(trans_pcie->rx_pool[0]),
+				      GFP_KERNEL);
+	trans_pcie->global_table =
+		kcalloc(RX_POOL_SIZE(trans_pcie->num_rx_bufs),
+			sizeof(trans_pcie->global_table[0]),
+			GFP_KERNEL);
+	if (!trans_pcie->rxq || !trans_pcie->rx_pool ||
+	    !trans_pcie->global_table) {
+		ret = -ENOMEM;
+		goto err;
+	}
 
 	spin_lock_init(&rba->lock);
 
@@ -845,6 +855,8 @@ err:
 		trans_pcie->base_rb_stts = NULL;
 		trans_pcie->base_rb_stts_dma = 0;
 	}
+	kfree(trans_pcie->rx_pool);
+	kfree(trans_pcie->global_table);
 	kfree(trans_pcie->rxq);
 
 	return ret;
@@ -1081,12 +1093,11 @@ static int _iwl_pcie_rx_init(struct iwl_trans *trans)
 
 	/* move the pool to the default queue and allocator ownerships */
 	queue_size = trans->trans_cfg->mq_rx_supported ?
-		     MQ_RX_NUM_RBDS : RX_QUEUE_SIZE;
+			trans_pcie->num_rx_bufs - 1 : RX_QUEUE_SIZE;
 	allocator_pool_size = trans->num_rx_queues *
 		(RX_CLAIM_REQ_ALLOC - RX_POST_REQ_ALLOC);
 	num_alloc = queue_size + allocator_pool_size;
-	BUILD_BUG_ON(ARRAY_SIZE(trans_pcie->global_table) !=
-		     ARRAY_SIZE(trans_pcie->rx_pool));
+
 	for (i = 0; i < num_alloc; i++) {
 		struct iwl_rx_mem_buffer *rxb = &trans_pcie->rx_pool[i];
 
@@ -1177,6 +1188,8 @@ void iwl_pcie_rx_free(struct iwl_trans *trans)
 		if (rxq->napi.poll)
 			netif_napi_del(&rxq->napi);
 	}
+	kfree(trans_pcie->rx_pool);
+	kfree(trans_pcie->global_table);
 	kfree(trans_pcie->rxq);
 }
 
@@ -1390,13 +1403,12 @@ static struct iwl_rx_mem_buffer *iwl_pcie_get_rxb(struct iwl_trans *trans,
 		return rxb;
 	}
 
-	/* used_bd is a 32/16 bit but only 12 are used to retrieve the vid */
 	if (trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_AX210)
-		vid = le16_to_cpu(rxq->cd[i].rbid) & 0x0FFF;
+		vid = le16_to_cpu(rxq->cd[i].rbid);
 	else
-		vid = le32_to_cpu(rxq->bd_32[i]) & 0x0FFF;
+		vid = le32_to_cpu(rxq->bd_32[i]) & 0x0FFF; /* 12-bit VID */
 
-	if (!vid || vid > ARRAY_SIZE(trans_pcie->global_table))
+	if (!vid || vid > RX_POOL_SIZE(trans_pcie->num_rx_bufs))
 		goto out_err;
 
 	rxb = trans_pcie->global_table[vid - 1];
-- 
cgit v1.2.3-59-g8ed1b


From 80084e3591390367f56161a48b6818b3a3fd64df Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 2 Oct 2019 11:33:46 +0200
Subject: iwlwifi: pcie: map only used part of RX buffers

We don't need to map *everything* of the RX buffers, we won't use
that much, map only the part we're going to use. This save some
IOMMU space (if applicable and it can deal with that) and also
prepares a bit for mapping partial pages for 2K buffers later.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-trans.h     | 18 ++++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/pcie/internal.h |  2 ++
 drivers/net/wireless/intel/iwlwifi/pcie/rx.c       | 13 ++++++-------
 drivers/net/wireless/intel/iwlwifi/pcie/trans.c    |  4 +++-
 4 files changed, 29 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
index 8cadad7364ac..872cda2be0f6 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
@@ -358,6 +358,24 @@ iwl_trans_get_rb_size_order(enum iwl_amsdu_size rb_size)
 	}
 }
 
+static inline int
+iwl_trans_get_rb_size(enum iwl_amsdu_size rb_size)
+{
+	switch (rb_size) {
+	case IWL_AMSDU_2K:
+		return 2 * 1024;
+	case IWL_AMSDU_4K:
+		return 4 * 1024;
+	case IWL_AMSDU_8K:
+		return 8 * 1024;
+	case IWL_AMSDU_12K:
+		return 12 * 1024;
+	default:
+		WARN_ON(1);
+		return 0;
+	}
+}
+
 struct iwl_hcmd_names {
 	u8 cmd_id;
 	const char *const cmd_name;
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
index c45df9a48559..c7d094cf1e4e 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
@@ -491,6 +491,7 @@ struct cont_rec {
  * @sw_csum_tx: if true, then the transport will compute the csum of the TXed
  *	frame.
  * @rx_page_order: page order for receive buffer size
+ * @rx_buf_bytes: RX buffer (RB) size in bytes
  * @reg_lock: protect hw register access
  * @mutex: to protect stop_device / start_fw / start_hw
  * @cmd_in_flight: true when we have a host command in flight
@@ -581,6 +582,7 @@ struct iwl_trans_pcie {
 	bool sw_csum_tx;
 	bool pcie_dbg_dumped_once;
 	u32 rx_page_order;
+	u32 rx_buf_bytes;
 
 	/*protect hw register */
 	spinlock_t reg_lock;
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
index f03d3cfb50da..20f93386fc4d 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
@@ -485,7 +485,7 @@ void iwl_pcie_rxq_alloc_rbs(struct iwl_trans *trans, gfp_t priority,
 		/* Get physical address of the RB */
 		rxb->page_dma =
 			dma_map_page(trans->dev, page, 0,
-				     PAGE_SIZE << trans_pcie->rx_page_order,
+				     trans_pcie->rx_buf_bytes,
 				     DMA_FROM_DEVICE);
 		if (dma_mapping_error(trans->dev, rxb->page_dma)) {
 			rxb->page = NULL;
@@ -514,8 +514,7 @@ void iwl_pcie_free_rbs_pool(struct iwl_trans *trans)
 		if (!trans_pcie->rx_pool[i].page)
 			continue;
 		dma_unmap_page(trans->dev, trans_pcie->rx_pool[i].page_dma,
-			       PAGE_SIZE << trans_pcie->rx_page_order,
-			       DMA_FROM_DEVICE);
+			       trans_pcie->rx_buf_bytes, DMA_FROM_DEVICE);
 		__free_pages(trans_pcie->rx_pool[i].page,
 			     trans_pcie->rx_page_order);
 		trans_pcie->rx_pool[i].page = NULL;
@@ -575,8 +574,8 @@ static void iwl_pcie_rx_allocator(struct iwl_trans *trans)
 
 			/* Get physical address of the RB */
 			rxb->page_dma = dma_map_page(trans->dev, page, 0,
-					PAGE_SIZE << trans_pcie->rx_page_order,
-					DMA_FROM_DEVICE);
+						     trans_pcie->rx_buf_bytes,
+						     DMA_FROM_DEVICE);
 			if (dma_mapping_error(trans->dev, rxb->page_dma)) {
 				rxb->page = NULL;
 				__free_pages(page, trans_pcie->rx_page_order);
@@ -1248,7 +1247,7 @@ static void iwl_pcie_rx_handle_rb(struct iwl_trans *trans,
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	struct iwl_txq *txq = trans_pcie->txq[trans_pcie->cmd_queue];
 	bool page_stolen = false;
-	int max_len = PAGE_SIZE << trans_pcie->rx_page_order;
+	int max_len = trans_pcie->rx_buf_bytes;
 	u32 offset = 0;
 
 	if (WARN_ON(!rxb))
@@ -1369,7 +1368,7 @@ static void iwl_pcie_rx_handle_rb(struct iwl_trans *trans,
 	if (rxb->page != NULL) {
 		rxb->page_dma =
 			dma_map_page(trans->dev, rxb->page, 0,
-				     PAGE_SIZE << trans_pcie->rx_page_order,
+				     trans_pcie->rx_buf_bytes,
 				     DMA_FROM_DEVICE);
 		if (dma_mapping_error(trans->dev, rxb->page_dma)) {
 			/*
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
index af9bc6b64542..d7617eb52c02 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
@@ -1915,6 +1915,8 @@ static void iwl_trans_pcie_configure(struct iwl_trans *trans,
 	trans_pcie->rx_buf_size = trans_cfg->rx_buf_size;
 	trans_pcie->rx_page_order =
 		iwl_trans_get_rb_size_order(trans_pcie->rx_buf_size);
+	trans_pcie->rx_buf_bytes =
+		iwl_trans_get_rb_size(trans_pcie->rx_buf_size);
 
 	trans_pcie->bc_table_dword = trans_cfg->bc_table_dword;
 	trans_pcie->scd_set_active = trans_cfg->scd_set_active;
@@ -2933,7 +2935,7 @@ static u32 iwl_trans_pcie_dump_rbs(struct iwl_trans *trans,
 				   int allocated_rb_nums)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
-	int max_len = PAGE_SIZE << trans_pcie->rx_page_order;
+	int max_len = trans_pcie->rx_buf_bytes;
 	/* Dump RBs is supported only for pre-9000 devices (1 queue) */
 	struct iwl_rxq *rxq = &trans_pcie->rxq[0];
 	u32 i, r, j, rb_len = 0;
-- 
cgit v1.2.3-59-g8ed1b


From cfdc20efebdc1dc918338818b10f1d95b58b2ca5 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 2 Oct 2019 12:11:54 +0200
Subject: iwlwifi: pcie: use partial pages if applicable

If we have only 2k RBs like on the latest (AX210) hardware, then
even on x86 where PAGE_SIZE is 4k we currently waste half of the
memory.

If this is the case, return partial pages from the allocator and
track the offset in each RBD (to be able to find the data in them
and remap them later.)

This might also address other platforms with larger PAGE_SIZE by
putting more RBs into a single large page.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/pcie/internal.h | 14 +++++
 drivers/net/wireless/intel/iwlwifi/pcie/rx.c       | 59 ++++++++++++++++++----
 drivers/net/wireless/intel/iwlwifi/pcie/trans.c    | 10 ++--
 3 files changed, 70 insertions(+), 13 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
index c7d094cf1e4e..2ac730859eac 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
@@ -106,6 +106,8 @@ struct iwl_host_cmd;
  * @page: driver's pointer to the rxb page
  * @invalid: rxb is in driver ownership - not owned by HW
  * @vid: index of this rxb in the global table
+ * @offset: indicates which offset of the page (in bytes)
+ *	this buffer uses (if multiple RBs fit into one page)
  */
 struct iwl_rx_mem_buffer {
 	dma_addr_t page_dma;
@@ -113,6 +115,7 @@ struct iwl_rx_mem_buffer {
 	u16 vid;
 	bool invalid;
 	struct list_head list;
+	u32 offset;
 };
 
 /**
@@ -511,6 +514,11 @@ struct cont_rec {
  * @in_rescan: true if we have triggered a device rescan
  * @base_rb_stts: base virtual address of receive buffer status for all queues
  * @base_rb_stts_dma: base physical address of receive buffer status
+ * @supported_dma_mask: DMA mask to validate the actual address against,
+ *	will be DMA_BIT_MASK(11) or DMA_BIT_MASK(12) depending on the device
+ * @alloc_page_lock: spinlock for the page allocator
+ * @alloc_page: allocated page to still use parts of
+ * @alloc_page_used: how much of the allocated page was already used (bytes)
  */
 struct iwl_trans_pcie {
 	struct iwl_rxq *rxq;
@@ -583,6 +591,12 @@ struct iwl_trans_pcie {
 	bool pcie_dbg_dumped_once;
 	u32 rx_page_order;
 	u32 rx_buf_bytes;
+	u32 supported_dma_mask;
+
+	/* allocator lock for the two values below */
+	spinlock_t alloc_page_lock;
+	struct page *alloc_page;
+	u32 alloc_page_used;
 
 	/*protect hw register */
 	spinlock_t reg_lock;
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
index 20f93386fc4d..498a14ebac47 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
@@ -298,6 +298,7 @@ static void iwl_pcie_restock_bd(struct iwl_trans *trans,
 static void iwl_pcie_rxmq_restock(struct iwl_trans *trans,
 				  struct iwl_rxq *rxq)
 {
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	struct iwl_rx_mem_buffer *rxb;
 
 	/*
@@ -318,8 +319,8 @@ static void iwl_pcie_rxmq_restock(struct iwl_trans *trans,
 				       list);
 		list_del(&rxb->list);
 		rxb->invalid = false;
-		/* 12 first bits are expected to be empty */
-		WARN_ON(rxb->page_dma & DMA_BIT_MASK(12));
+		/* some low bits are expected to be unset (depending on hw) */
+		WARN_ON(rxb->page_dma & trans_pcie->supported_dma_mask);
 		/* Point to Rx buffer via next RBD in circular buffer */
 		iwl_pcie_restock_bd(trans, rxq, rxb);
 		rxq->write = (rxq->write + 1) & (rxq->queue_size - 1);
@@ -412,15 +413,34 @@ void iwl_pcie_rxq_restock(struct iwl_trans *trans, struct iwl_rxq *rxq)
  *
  */
 static struct page *iwl_pcie_rx_alloc_page(struct iwl_trans *trans,
-					   gfp_t priority)
+					   u32 *offset, gfp_t priority)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	unsigned int rbsize = iwl_trans_get_rb_size(trans_pcie->rx_buf_size);
+	unsigned int allocsize = PAGE_SIZE << trans_pcie->rx_page_order;
 	struct page *page;
 	gfp_t gfp_mask = priority;
 
 	if (trans_pcie->rx_page_order > 0)
 		gfp_mask |= __GFP_COMP;
 
+	if (trans_pcie->alloc_page) {
+		spin_lock_bh(&trans_pcie->alloc_page_lock);
+		/* recheck */
+		if (trans_pcie->alloc_page) {
+			*offset = trans_pcie->alloc_page_used;
+			page = trans_pcie->alloc_page;
+			trans_pcie->alloc_page_used += rbsize;
+			if (trans_pcie->alloc_page_used >= allocsize)
+				trans_pcie->alloc_page = NULL;
+			else
+				get_page(page);
+			spin_unlock_bh(&trans_pcie->alloc_page_lock);
+			return page;
+		}
+		spin_unlock_bh(&trans_pcie->alloc_page_lock);
+	}
+
 	/* Alloc a new receive buffer */
 	page = alloc_pages(gfp_mask, trans_pcie->rx_page_order);
 	if (!page) {
@@ -436,6 +456,18 @@ static struct page *iwl_pcie_rx_alloc_page(struct iwl_trans *trans,
 				 "Failed to alloc_pages\n");
 		return NULL;
 	}
+
+	if (2 * rbsize <= allocsize) {
+		spin_lock_bh(&trans_pcie->alloc_page_lock);
+		if (!trans_pcie->alloc_page) {
+			get_page(page);
+			trans_pcie->alloc_page = page;
+			trans_pcie->alloc_page_used = rbsize;
+		}
+		spin_unlock_bh(&trans_pcie->alloc_page_lock);
+	}
+
+	*offset = 0;
 	return page;
 }
 
@@ -456,6 +488,8 @@ void iwl_pcie_rxq_alloc_rbs(struct iwl_trans *trans, gfp_t priority,
 	struct page *page;
 
 	while (1) {
+		unsigned int offset;
+
 		spin_lock(&rxq->lock);
 		if (list_empty(&rxq->rx_used)) {
 			spin_unlock(&rxq->lock);
@@ -463,8 +497,7 @@ void iwl_pcie_rxq_alloc_rbs(struct iwl_trans *trans, gfp_t priority,
 		}
 		spin_unlock(&rxq->lock);
 
-		/* Alloc a new receive buffer */
-		page = iwl_pcie_rx_alloc_page(trans, priority);
+		page = iwl_pcie_rx_alloc_page(trans, &offset, priority);
 		if (!page)
 			return;
 
@@ -482,9 +515,10 @@ void iwl_pcie_rxq_alloc_rbs(struct iwl_trans *trans, gfp_t priority,
 
 		BUG_ON(rxb->page);
 		rxb->page = page;
+		rxb->offset = offset;
 		/* Get physical address of the RB */
 		rxb->page_dma =
-			dma_map_page(trans->dev, page, 0,
+			dma_map_page(trans->dev, page, rxb->offset,
 				     trans_pcie->rx_buf_bytes,
 				     DMA_FROM_DEVICE);
 		if (dma_mapping_error(trans->dev, rxb->page_dma)) {
@@ -567,13 +601,15 @@ static void iwl_pcie_rx_allocator(struct iwl_trans *trans)
 			BUG_ON(rxb->page);
 
 			/* Alloc a new receive buffer */
-			page = iwl_pcie_rx_alloc_page(trans, gfp_mask);
+			page = iwl_pcie_rx_alloc_page(trans, &rxb->offset,
+						      gfp_mask);
 			if (!page)
 				continue;
 			rxb->page = page;
 
 			/* Get physical address of the RB */
-			rxb->page_dma = dma_map_page(trans->dev, page, 0,
+			rxb->page_dma = dma_map_page(trans->dev, page,
+						     rxb->offset,
 						     trans_pcie->rx_buf_bytes,
 						     DMA_FROM_DEVICE);
 			if (dma_mapping_error(trans->dev, rxb->page_dma)) {
@@ -1190,6 +1226,9 @@ void iwl_pcie_rx_free(struct iwl_trans *trans)
 	kfree(trans_pcie->rx_pool);
 	kfree(trans_pcie->global_table);
 	kfree(trans_pcie->rxq);
+
+	if (trans_pcie->alloc_page)
+		__free_pages(trans_pcie->alloc_page, trans_pcie->rx_page_order);
 }
 
 static void iwl_pcie_rx_move_to_allocator(struct iwl_rxq *rxq,
@@ -1261,7 +1300,7 @@ static void iwl_pcie_rx_handle_rb(struct iwl_trans *trans,
 		bool reclaim;
 		int index, cmd_index, len;
 		struct iwl_rx_cmd_buffer rxcb = {
-			._offset = offset,
+			._offset = rxb->offset + offset,
 			._rx_page_order = trans_pcie->rx_page_order,
 			._page = rxb->page,
 			._page_stolen = false,
@@ -1367,7 +1406,7 @@ static void iwl_pcie_rx_handle_rb(struct iwl_trans *trans,
 	 * rx_free list for reuse later. */
 	if (rxb->page != NULL) {
 		rxb->page_dma =
-			dma_map_page(trans->dev, rxb->page, 0,
+			dma_map_page(trans->dev, rxb->page, rxb->offset,
 				     trans_pcie->rx_buf_bytes,
 				     DMA_FROM_DEVICE);
 		if (dma_mapping_error(trans->dev, rxb->page_dma)) {
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
index d7617eb52c02..cb4915451792 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
@@ -1917,6 +1917,9 @@ static void iwl_trans_pcie_configure(struct iwl_trans *trans,
 		iwl_trans_get_rb_size_order(trans_pcie->rx_buf_size);
 	trans_pcie->rx_buf_bytes =
 		iwl_trans_get_rb_size(trans_pcie->rx_buf_size);
+	trans_pcie->supported_dma_mask = DMA_BIT_MASK(12);
+	if (trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_AX210)
+		trans_pcie->supported_dma_mask = DMA_BIT_MASK(11);
 
 	trans_pcie->bc_table_dword = trans_cfg->bc_table_dword;
 	trans_pcie->scd_set_active = trans_cfg->scd_set_active;
@@ -2961,9 +2964,9 @@ static u32 iwl_trans_pcie_dump_rbs(struct iwl_trans *trans,
 		rb->index = cpu_to_le32(i);
 		memcpy(rb->data, page_address(rxb->page), max_len);
 		/* remap the page for the free benefit */
-		rxb->page_dma = dma_map_page(trans->dev, rxb->page, 0,
-						     max_len,
-						     DMA_FROM_DEVICE);
+		rxb->page_dma = dma_map_page(trans->dev, rxb->page,
+					     rxb->offset, max_len,
+					     DMA_FROM_DEVICE);
 
 		*data = iwl_fw_error_next_data(*data);
 	}
@@ -3454,6 +3457,7 @@ struct iwl_trans *iwl_trans_pcie_alloc(struct pci_dev *pdev,
 	trans_pcie->opmode_down = true;
 	spin_lock_init(&trans_pcie->irq_lock);
 	spin_lock_init(&trans_pcie->reg_lock);
+	spin_lock_init(&trans_pcie->alloc_page_lock);
 	mutex_init(&trans_pcie->mutex);
 	init_waitqueue_head(&trans_pcie->ucode_write_waitq);
 
-- 
cgit v1.2.3-59-g8ed1b


From 0e00270863daaf803b32c0ff64b73dabdefd612a Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 8 Oct 2019 16:03:15 +0200
Subject: iwlwifi: pcie: validate queue ID before array deref/bit ops

Validate that the queue ID is in range before trying to use it as
an index or for test_bit() - the previous bug showed that this has
in fact happened, and it was lucky that we caught it there, had the
bit been set then we'd have actually used the value despite being
far out of range.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c | 16 +++++++++++++++-
 1 file changed, 15 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c
index 8ca0250de99e..c3683cddc301 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c
@@ -587,6 +587,10 @@ int iwl_trans_pcie_gen2_tx(struct iwl_trans *trans, struct sk_buff *skb,
 	int idx;
 	void *tfd;
 
+	if (WARN_ONCE(txq_id >= IWL_MAX_TVQM_QUEUES,
+		      "queue %d out of range", txq_id))
+		return -EINVAL;
+
 	if (WARN_ONCE(!test_bit(txq_id, trans_pcie->queue_used),
 		      "TX on unused queue %d\n", txq_id))
 		return -EINVAL;
@@ -1101,9 +1105,15 @@ void iwl_pcie_gen2_txq_free_memory(struct iwl_trans *trans,
 static void iwl_pcie_gen2_txq_free(struct iwl_trans *trans, int txq_id)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
-	struct iwl_txq *txq = trans_pcie->txq[txq_id];
+	struct iwl_txq *txq;
 	int i;
 
+	if (WARN_ONCE(txq_id >= IWL_MAX_TVQM_QUEUES,
+		      "queue %d out of range", txq_id))
+		return;
+
+	txq = trans_pcie->txq[txq_id];
+
 	if (WARN_ON(!txq))
 		return;
 
@@ -1258,6 +1268,10 @@ void iwl_trans_pcie_dyn_txq_free(struct iwl_trans *trans, int queue)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 
+	if (WARN(queue >= IWL_MAX_TVQM_QUEUES,
+		 "queue %d out of range", queue))
+		return;
+
 	/*
 	 * Upon HW Rfkill - we stop the device, and then stop the queues
 	 * in the op_mode. Just for the sake of the simplicity of the op_mode,
-- 
cgit v1.2.3-59-g8ed1b


From b81b7bd02eda86ae3cb40eb594f5ccd6ca790d8b Mon Sep 17 00:00:00 2001
From: Luca Coelho <luciano.coelho@intel.com>
Date: Thu, 10 Oct 2019 10:13:49 +0300
Subject: iwlwifi: remove some outdated iwl22000 configurations

A few configuration structures were either not referenced anymore or
assigned to devices IDs that were not in use anymore.  Remove them.

Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/cfg/22000.c  | 22 ----------------------
 drivers/net/wireless/intel/iwlwifi/iwl-config.h |  3 ---
 drivers/net/wireless/intel/iwlwifi/pcie/drv.c   |  6 ------
 3 files changed, 31 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
index bd687e745dc9..1d3dc7cf4d63 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
@@ -247,28 +247,6 @@ static const struct iwl_ht_params iwl_22000_ht_params = {
 #define IWL_NUM_RBDS_22000_HE		2048
 #define IWL_NUM_RBDS_AX210_HE		4096
 
-const struct iwl_cfg iwl22000_2ac_cfg_hr = {
-	.name = "Intel(R) Dual Band Wireless AC 22000",
-	.fw_name_pre = IWL_22000_HR_FW_PRE,
-	IWL_DEVICE_22500,
-	.num_rbds = IWL_NUM_RBDS_NON_HE,
-};
-
-const struct iwl_cfg iwl22000_2ac_cfg_hr_cdb = {
-	.name = "Intel(R) Dual Band Wireless AC 22000",
-	.fw_name_pre = IWL_22000_HR_CDB_FW_PRE,
-	IWL_DEVICE_22500,
-	.cdb = true,
-	.num_rbds = IWL_NUM_RBDS_NON_HE,
-};
-
-const struct iwl_cfg iwl22000_2ac_cfg_jf = {
-	.name = "Intel(R) Dual Band Wireless AC 22000",
-	.fw_name_pre = IWL_22000_JF_FW_PRE,
-	IWL_DEVICE_22500,
-	.num_rbds = IWL_NUM_RBDS_NON_HE,
-};
-
 const struct iwl_cfg iwl_ax101_cfg_qu_hr = {
 	.name = "Intel(R) Wi-Fi 6 AX101",
 	.fw_name_pre = IWL_22000_QU_B_HR_B_FW_PRE,
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
index f8a93324d7c0..31a6f286f7f7 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
@@ -598,9 +598,6 @@ extern const struct iwl_cfg iwl9560_2ac_cfg_shared_clk;
 extern const struct iwl_cfg iwl9560_2ac_160_cfg_shared_clk;
 extern const struct iwl_cfg iwl9560_killer_2ac_cfg_shared_clk;
 extern const struct iwl_cfg iwl9560_killer_s_2ac_cfg_shared_clk;
-extern const struct iwl_cfg iwl22000_2ac_cfg_hr;
-extern const struct iwl_cfg iwl22000_2ac_cfg_hr_cdb;
-extern const struct iwl_cfg iwl22000_2ac_cfg_jf;
 extern const struct iwl_cfg iwl_ax101_cfg_qu_hr;
 extern const struct iwl_cfg iwl_ax101_cfg_qu_c0_hr_b0;
 extern const struct iwl_cfg iwl_ax101_cfg_quz_hr;
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
index cdf112d6c0f9..bac3d3fa0dec 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
@@ -910,7 +910,6 @@ static const struct pci_device_id iwl_hw_card_ids[] = {
 	{IWL_PCI_DEVICE(0x2720, 0x0074, iwl_ax201_cfg_qu_hr)},
 	{IWL_PCI_DEVICE(0x2720, 0x0078, iwl_ax201_cfg_qu_hr)},
 	{IWL_PCI_DEVICE(0x2720, 0x007C, iwl_ax201_cfg_qu_hr)},
-	{IWL_PCI_DEVICE(0x2720, 0x0090, iwl22000_2ac_cfg_hr_cdb)},
 	{IWL_PCI_DEVICE(0x2720, 0x0244, iwl_ax101_cfg_qu_hr)},
 	{IWL_PCI_DEVICE(0x2720, 0x0310, iwl_ax201_cfg_qu_hr)},
 	{IWL_PCI_DEVICE(0x2720, 0x0A10, iwl_ax201_cfg_qu_hr)},
@@ -1083,11 +1082,6 @@ static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		else if ((hw_status & UMAG_GEN_HW_IS_FPGA) &&
 			 CSR_HW_RF_STEP(iwl_trans->hw_rf_id) == SILICON_A_STEP) {
 			cfg = &iwl22000_2ax_cfg_qnj_hr_a0_f0;
-		} else {
-			/*
-			 * a step no FPGA
-			 */
-			cfg = &iwl22000_2ac_cfg_hr;
 		}
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From 6dece0e99faa2aef6c21406dcd95e53d1a6f99d5 Mon Sep 17 00:00:00 2001
From: Luca Coelho <luciano.coelho@intel.com>
Date: Thu, 10 Oct 2019 11:59:30 +0300
Subject: iwlwifi: remove CSR registers abstraction

We needed this abstraction for some CSR registers for
IWL_DEVICE_22560, but that has been removed, so we don't need the
abstraction anymore.  Remove it.

Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/cfg/1000.c      |  6 +--
 drivers/net/wireless/intel/iwlwifi/cfg/2000.c      | 12 ++----
 drivers/net/wireless/intel/iwlwifi/cfg/22000.c     |  2 -
 drivers/net/wireless/intel/iwlwifi/cfg/5000.c      |  7 +--
 drivers/net/wireless/intel/iwlwifi/cfg/6000.c      | 19 +++-----
 drivers/net/wireless/intel/iwlwifi/cfg/7000.c      |  3 +-
 drivers/net/wireless/intel/iwlwifi/cfg/8000.c      |  3 +-
 drivers/net/wireless/intel/iwlwifi/cfg/9000.c      |  1 -
 drivers/net/wireless/intel/iwlwifi/iwl-config.h    | 50 ----------------------
 drivers/net/wireless/intel/iwlwifi/iwl-csr.h       | 27 +++++++++++-
 drivers/net/wireless/intel/iwlwifi/iwl-io.c        | 37 ++--------------
 drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c | 14 ++----
 drivers/net/wireless/intel/iwlwifi/pcie/drv.c      |  6 ---
 drivers/net/wireless/intel/iwlwifi/pcie/rx.c       |  2 +-
 .../net/wireless/intel/iwlwifi/pcie/trans-gen2.c   |  5 +--
 drivers/net/wireless/intel/iwlwifi/pcie/trans.c    | 37 +++++++---------
 drivers/net/wireless/intel/iwlwifi/pcie/tx.c       | 12 +++---
 17 files changed, 74 insertions(+), 169 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/1000.c b/drivers/net/wireless/intel/iwlwifi/cfg/1000.c
index b92255b91b72..8a4579bb10d3 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/1000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/1000.c
@@ -77,8 +77,7 @@ static const struct iwl_eeprom_params iwl1000_eeprom_params = {
 	.trans.base_params = &iwl1000_base_params,		\
 	.eeprom_params = &iwl1000_eeprom_params,		\
 	.led_mode = IWL_LED_BLINK,				\
-	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K,	\
-	.trans.csr = &iwl_csr_v1
+	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K
 
 const struct iwl_cfg iwl1000_bgn_cfg = {
 	.name = "Intel(R) Centrino(R) Wireless-N 1000 BGN",
@@ -104,8 +103,7 @@ const struct iwl_cfg iwl1000_bg_cfg = {
 	.eeprom_params = &iwl1000_eeprom_params,		\
 	.led_mode = IWL_LED_RF_STATE,				\
 	.rx_with_siso_diversity = true,				\
-	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K,	\
-	.trans.csr = &iwl_csr_v1
+	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K
 
 const struct iwl_cfg iwl100_bgn_cfg = {
 	.name = "Intel(R) Centrino(R) Wireless-N 100 BGN",
diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/2000.c b/drivers/net/wireless/intel/iwlwifi/cfg/2000.c
index 2b1ae0cecc83..7140a5f3ea8b 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/2000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/2000.c
@@ -103,8 +103,7 @@ static const struct iwl_eeprom_params iwl20x0_eeprom_params = {
 	.trans.base_params = &iwl2000_base_params,		\
 	.eeprom_params = &iwl20x0_eeprom_params,		\
 	.led_mode = IWL_LED_RF_STATE,				\
-	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K,	\
-	.trans.csr = &iwl_csr_v1
+	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K
 
 
 const struct iwl_cfg iwl2000_2bgn_cfg = {
@@ -131,8 +130,7 @@ const struct iwl_cfg iwl2000_2bgn_d_cfg = {
 	.trans.base_params = &iwl2030_base_params,		\
 	.eeprom_params = &iwl20x0_eeprom_params,		\
 	.led_mode = IWL_LED_RF_STATE,				\
-	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K,	\
-	.trans.csr = &iwl_csr_v1
+	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K
 
 const struct iwl_cfg iwl2030_2bgn_cfg = {
 	.name = "Intel(R) Centrino(R) Wireless-N 2230 BGN",
@@ -153,8 +151,7 @@ const struct iwl_cfg iwl2030_2bgn_cfg = {
 	.eeprom_params = &iwl20x0_eeprom_params,		\
 	.led_mode = IWL_LED_RF_STATE,				\
 	.rx_with_siso_diversity = true,				\
-	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K,	\
-	.trans.csr = &iwl_csr_v1
+	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K
 
 const struct iwl_cfg iwl105_bgn_cfg = {
 	.name = "Intel(R) Centrino(R) Wireless-N 105 BGN",
@@ -181,8 +178,7 @@ const struct iwl_cfg iwl105_bgn_d_cfg = {
 	.eeprom_params = &iwl20x0_eeprom_params,		\
 	.led_mode = IWL_LED_RF_STATE,				\
 	.rx_with_siso_diversity = true,				\
-	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K,	\
-	.trans.csr = &iwl_csr_v1
+	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K
 
 const struct iwl_cfg iwl135_bgn_cfg = {
 	.name = "Intel(R) Centrino(R) Wireless-N 135 BGN",
diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
index 1d3dc7cf4d63..837c06891f5e 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
@@ -199,7 +199,6 @@ static const struct iwl_ht_params iwl_22000_ht_params = {
 	IWL_DEVICE_22000_COMMON,					\
 	.trans.device_family = IWL_DEVICE_FAMILY_22000,			\
 	.trans.base_params = &iwl_22000_base_params,			\
-	.trans.csr = &iwl_csr_v1,					\
 	.gp2_reg_addr = 0xa02c68,					\
 	.mon_dram_regs = {						\
 		.write_ptr = {						\
@@ -217,7 +216,6 @@ static const struct iwl_ht_params iwl_22000_ht_params = {
 	.trans.umac_prph_offset = 0x300000,				\
 	.trans.device_family = IWL_DEVICE_FAMILY_AX210,			\
 	.trans.base_params = &iwl_ax210_base_params,			\
-	.trans.csr = &iwl_csr_v1,					\
 	.min_txq_size = 128,						\
 	.gp2_reg_addr = 0xd02c68,					\
 	.min_256_ba_txq_size = 512,					\
diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/5000.c b/drivers/net/wireless/intel/iwlwifi/cfg/5000.c
index aab4495c6085..3591336dc644 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/5000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/5000.c
@@ -75,8 +75,7 @@ static const struct iwl_eeprom_params iwl5000_eeprom_params = {
 	.trans.base_params = &iwl5000_base_params,		\
 	.eeprom_params = &iwl5000_eeprom_params,		\
 	.led_mode = IWL_LED_BLINK,				\
-	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K,	\
-	.trans.csr = &iwl_csr_v1
+	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K
 
 const struct iwl_cfg iwl5300_agn_cfg = {
 	.name = "Intel(R) Ultimate N WiFi Link 5300 AGN",
@@ -125,7 +124,6 @@ const struct iwl_cfg iwl5350_agn_cfg = {
 	.ht_params = &iwl5000_ht_params,
 	.led_mode = IWL_LED_BLINK,
 	.internal_wimax_coex = true,
-	.trans.csr = &iwl_csr_v1,
 };
 
 #define IWL_DEVICE_5150						\
@@ -141,8 +139,7 @@ const struct iwl_cfg iwl5350_agn_cfg = {
 	.eeprom_params = &iwl5000_eeprom_params,		\
 	.led_mode = IWL_LED_BLINK,				\
 	.internal_wimax_coex = true,				\
-	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K,	\
-	.trans.csr = &iwl_csr_v1
+	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K
 
 const struct iwl_cfg iwl5150_agn_cfg = {
 	.name = "Intel(R) WiMAX/WiFi Link 5150 AGN",
diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/6000.c b/drivers/net/wireless/intel/iwlwifi/cfg/6000.c
index 39ea81903dbe..b4a8a6804c39 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/6000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/6000.c
@@ -124,8 +124,7 @@ static const struct iwl_eeprom_params iwl6000_eeprom_params = {
 	.trans.base_params = &iwl6000_g2_base_params,		\
 	.eeprom_params = &iwl6000_eeprom_params,		\
 	.led_mode = IWL_LED_RF_STATE,				\
-	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K,	\
-	.trans.csr = &iwl_csr_v1
+	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K
 
 const struct iwl_cfg iwl6005_2agn_cfg = {
 	.name = "Intel(R) Centrino(R) Advanced-N 6205 AGN",
@@ -179,8 +178,7 @@ const struct iwl_cfg iwl6005_2agn_mow2_cfg = {
 	.trans.base_params = &iwl6000_g2_base_params,		\
 	.eeprom_params = &iwl6000_eeprom_params,		\
 	.led_mode = IWL_LED_RF_STATE,				\
-	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K,	\
-	.trans.csr = &iwl_csr_v1
+	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K
 
 const struct iwl_cfg iwl6030_2agn_cfg = {
 	.name = "Intel(R) Centrino(R) Advanced-N 6230 AGN",
@@ -216,8 +214,7 @@ const struct iwl_cfg iwl6030_2bg_cfg = {
 	.trans.base_params = &iwl6000_g2_base_params,		\
 	.eeprom_params = &iwl6000_eeprom_params,		\
 	.led_mode = IWL_LED_RF_STATE,				\
-	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K,	\
-	.trans.csr = &iwl_csr_v1
+	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K
 
 const struct iwl_cfg iwl6035_2agn_cfg = {
 	.name = "Intel(R) Centrino(R) Advanced-N 6235 AGN",
@@ -272,8 +269,7 @@ const struct iwl_cfg iwl130_bg_cfg = {
 	.trans.base_params = &iwl6000_base_params,		\
 	.eeprom_params = &iwl6000_eeprom_params,		\
 	.led_mode = IWL_LED_BLINK,				\
-	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K,	\
-	.trans.csr = &iwl_csr_v1
+	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K
 
 const struct iwl_cfg iwl6000i_2agn_cfg = {
 	.name = "Intel(R) Centrino(R) Advanced-N 6200 AGN",
@@ -306,8 +302,7 @@ const struct iwl_cfg iwl6000i_2bg_cfg = {
 	.eeprom_params = &iwl6000_eeprom_params,		\
 	.led_mode = IWL_LED_BLINK,				\
 	.internal_wimax_coex = true,				\
-	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K,	\
-	.trans.csr = &iwl_csr_v1
+	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K
 
 const struct iwl_cfg iwl6050_2agn_cfg = {
 	.name = "Intel(R) Centrino(R) Advanced-N + WiMAX 6250 AGN",
@@ -333,8 +328,7 @@ const struct iwl_cfg iwl6050_2abg_cfg = {
 	.eeprom_params = &iwl6000_eeprom_params,		\
 	.led_mode = IWL_LED_BLINK,				\
 	.internal_wimax_coex = true,				\
-	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K,	\
-	.trans.csr = &iwl_csr_v1
+	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K
 
 const struct iwl_cfg iwl6150_bgn_cfg = {
 	.name = "Intel(R) Centrino(R) Wireless-N + WiMAX 6150 BGN",
@@ -361,7 +355,6 @@ const struct iwl_cfg iwl6000_3agn_cfg = {
 	.eeprom_params = &iwl6000_eeprom_params,
 	.ht_params = &iwl6000_ht_params,
 	.led_mode = IWL_LED_BLINK,
-	.trans.csr = &iwl_csr_v1,
 };
 
 MODULE_FIRMWARE(IWL6000_MODULE_FIRMWARE(IWL6000_UCODE_API_MAX));
diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/7000.c b/drivers/net/wireless/intel/iwlwifi/cfg/7000.c
index deb520aeb3f8..b72993e07fc2 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/7000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/7000.c
@@ -154,8 +154,7 @@ static const struct iwl_ht_params iwl7000_ht_params = {
 	.nvm_hw_section_num = 0,				\
 	.non_shared_ant = ANT_A,				\
 	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K,	\
-	.dccm_offset = IWL7000_DCCM_OFFSET,			\
-	.trans.csr = &iwl_csr_v1
+	.dccm_offset = IWL7000_DCCM_OFFSET
 
 #define IWL_DEVICE_7000						\
 	IWL_DEVICE_7000_COMMON,					\
diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/8000.c b/drivers/net/wireless/intel/iwlwifi/cfg/8000.c
index b3cc477140c0..280d84fa5cb1 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/8000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/8000.c
@@ -151,8 +151,7 @@ static const struct iwl_tt_params iwl8000_tt_params = {
 	.apmg_not_supported = true,					\
 	.nvm_type = IWL_NVM_EXT,					\
 	.dbgc_supported = true,						\
-	.min_umac_error_event_table = 0x800000,				\
-	.trans.csr = &iwl_csr_v1
+	.min_umac_error_event_table = 0x800000
 
 #define IWL_DEVICE_8000							\
 	IWL_DEVICE_8000_COMMON,						\
diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/9000.c b/drivers/net/wireless/intel/iwlwifi/cfg/9000.c
index ac05c1a6ec6b..79390f880efb 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/9000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/9000.c
@@ -145,7 +145,6 @@ static const struct iwl_tt_params iwl9000_tt_params = {
 	.nvm_type = IWL_NVM_EXT,					\
 	.dbgc_supported = true,						\
 	.min_umac_error_event_table = 0x800000,				\
-	.trans.csr = &iwl_csr_v1,					\
 	.d3_debug_data_base_addr = 0x401000,				\
 	.d3_debug_data_length = 92 * 1024,				\
 	.ht_params = &iwl9000_ht_params,				\
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
index 31a6f286f7f7..bb91c36c9d8b 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
@@ -284,52 +284,6 @@ struct iwl_pwr_tx_backoff {
 	u32 backoff;
 };
 
-/**
- * struct iwl_csr_params
- *
- * @flag_sw_reset: reset the device
- * @flag_mac_clock_ready:
- *	Indicates MAC (ucode processor, etc.) is powered up and can run.
- *	Internal resources are accessible.
- *	NOTE:  This does not indicate that the processor is actually running.
- *	NOTE:  This does not indicate that device has completed
- *	       init or post-power-down restore of internal SRAM memory.
- *	       Use CSR_UCODE_DRV_GP1_BIT_MAC_SLEEP as indication that
- *	       SRAM is restored and uCode is in normal operation mode.
- *	       This note is relevant only for pre 5xxx devices.
- *	NOTE:  After device reset, this bit remains "0" until host sets
- *	       INIT_DONE
- * @flag_init_done: Host sets this to put device into fully operational
- *	D0 power mode. Host resets this after SW_RESET to put device into
- *	low power mode.
- * @flag_mac_access_req: Host sets this to request and maintain MAC wakeup,
- *	to allow host access to device-internal resources. Host must wait for
- *	mac_clock_ready (and !GOING_TO_SLEEP) before accessing non-CSR device
- *	registers.
- * @flag_val_mac_access_en: mac access is enabled
- * @flag_master_dis: disable master
- * @flag_stop_master: stop master
- * @addr_sw_reset: address for resetting the device
- * @mac_addr0_otp: first part of MAC address from OTP
- * @mac_addr1_otp: second part of MAC address from OTP
- * @mac_addr0_strap: first part of MAC address from strap
- * @mac_addr1_strap: second part of MAC address from strap
- */
-struct iwl_csr_params {
-	u8 flag_sw_reset;
-	u8 flag_mac_clock_ready;
-	u8 flag_init_done;
-	u8 flag_mac_access_req;
-	u8 flag_val_mac_access_en;
-	u8 flag_master_dis;
-	u8 flag_stop_master;
-	u8 addr_sw_reset;
-	u32 mac_addr0_otp;
-	u32 mac_addr1_otp;
-	u32 mac_addr0_strap;
-	u32 mac_addr1_strap;
-};
-
 /**
  * struct iwl_cfg_trans - information needed to start the trans
  *
@@ -348,7 +302,6 @@ struct iwl_csr_params {
  */
 struct iwl_cfg_trans_params {
 	const struct iwl_base_params *base_params;
-	const struct iwl_csr_params *csr;
 	enum iwl_device_family device_family;
 	u32 umac_prph_offset;
 	u32 rf_id:1,
@@ -499,9 +452,6 @@ struct iwl_cfg {
 	const struct iwl_fw_mon_regs mon_smem_regs;
 };
 
-extern const struct iwl_csr_params iwl_csr_v1;
-extern const struct iwl_csr_params iwl_csr_v2;
-
 /*
  * This list declares the config structures for all devices.
  */
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-csr.h b/drivers/net/wireless/intel/iwlwifi/iwl-csr.h
index 92d9898ab7c2..2c150778d144 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-csr.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-csr.h
@@ -256,6 +256,7 @@
 /* RESET */
 #define CSR_RESET_REG_FLAG_NEVO_RESET                (0x00000001)
 #define CSR_RESET_REG_FLAG_FORCE_NMI                 (0x00000002)
+#define CSR_RESET_REG_FLAG_SW_RESET		     (0x00000080)
 #define CSR_RESET_REG_FLAG_MASTER_DISABLED           (0x00000100)
 #define CSR_RESET_REG_FLAG_STOP_MASTER               (0x00000200)
 #define CSR_RESET_LINK_PWR_MGMT_DISABLED             (0x80000000)
@@ -278,11 +279,35 @@
  *     4:  GOING_TO_SLEEP
  *         Indicates MAC is entering a power-saving sleep power-down.
  *         Not a good time to access device-internal resources.
+ *     3:  MAC_ACCESS_REQ
+ *         Host sets this to request and maintain MAC wakeup, to allow host
+ *         access to device-internal resources.  Host must wait for
+ *         MAC_CLOCK_READY (and !GOING_TO_SLEEP) before accessing non-CSR
+ *         device registers.
+ *     2:  INIT_DONE
+ *         Host sets this to put device into fully operational D0 power mode.
+ *         Host resets this after SW_RESET to put device into low power mode.
+ *     0:  MAC_CLOCK_READY
+ *         Indicates MAC (ucode processor, etc.) is powered up and can run.
+ *         Internal resources are accessible.
+ *         NOTE:  This does not indicate that the processor is actually running.
+ *         NOTE:  This does not indicate that device has completed
+ *                init or post-power-down restore of internal SRAM memory.
+ *                Use CSR_UCODE_DRV_GP1_BIT_MAC_SLEEP as indication that
+ *                SRAM is restored and uCode is in normal operation mode.
+ *                Later devices (5xxx/6xxx/1xxx) use non-volatile SRAM, and
+ *                do not need to save/restore it.
+ *         NOTE:  After device reset, this bit remains "0" until host sets
+ *                INIT_DONE
  */
+#define CSR_GP_CNTRL_REG_FLAG_MAC_CLOCK_READY	     (0x00000001)
 #define CSR_GP_CNTRL_REG_FLAG_INIT_DONE		     (0x00000004)
-#define CSR_GP_CNTRL_REG_FLAG_GOING_TO_SLEEP         (0x00000010)
+#define CSR_GP_CNTRL_REG_FLAG_MAC_ACCESS_REQ	     (0x00000008)
+#define CSR_GP_CNTRL_REG_FLAG_GOING_TO_SLEEP	     (0x00000010)
 #define CSR_GP_CNTRL_REG_FLAG_XTAL_ON		     (0x00000400)
 
+#define CSR_GP_CNTRL_REG_VAL_MAC_ACCESS_EN	     (0x00000001)
+
 #define CSR_GP_CNTRL_REG_MSK_POWER_SAVE_TYPE         (0x07000000)
 #define CSR_GP_CNTRL_REG_FLAG_RFKILL_WAKE_L1A_EN     (0x04000000)
 #define CSR_GP_CNTRL_REG_FLAG_HW_RF_KILL_SW          (0x08000000)
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-io.c b/drivers/net/wireless/intel/iwlwifi/iwl-io.c
index 1b7414bf7bef..2139f0b8f2bb 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-io.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-io.c
@@ -70,36 +70,6 @@
 #include "iwl-prph.h"
 #include "iwl-fh.h"
 
-const struct iwl_csr_params iwl_csr_v1 = {
-	.flag_mac_clock_ready = 0,
-	.flag_val_mac_access_en = 0,
-	.flag_init_done = 2,
-	.flag_mac_access_req = 3,
-	.flag_sw_reset = 7,
-	.flag_master_dis = 8,
-	.flag_stop_master = 9,
-	.addr_sw_reset = CSR_BASE + 0x020,
-	.mac_addr0_otp = 0x380,
-	.mac_addr1_otp = 0x384,
-	.mac_addr0_strap = 0x388,
-	.mac_addr1_strap = 0x38C
-};
-
-const struct iwl_csr_params iwl_csr_v2 = {
-	.flag_init_done = 6,
-	.flag_mac_clock_ready = 20,
-	.flag_val_mac_access_en = 20,
-	.flag_mac_access_req = 21,
-	.flag_master_dis = 28,
-	.flag_stop_master = 29,
-	.flag_sw_reset = 31,
-	.addr_sw_reset = CSR_BASE + 0x024,
-	.mac_addr0_otp = 0x30,
-	.mac_addr1_otp = 0x34,
-	.mac_addr0_strap = 0x38,
-	.mac_addr1_strap = 0x3C
-};
-
 void iwl_write8(struct iwl_trans *trans, u32 ofs, u8 val)
 {
 	trace_iwlwifi_dev_iowrite8(trans->dev, ofs, val);
@@ -506,8 +476,7 @@ int iwl_finish_nic_init(struct iwl_trans *trans,
 	 * Set "initialization complete" bit to move adapter from
 	 * D0U* --> D0A* (powered-up active) state.
 	 */
-	iwl_set_bit(trans, CSR_GP_CNTRL,
-		    BIT(cfg_trans->csr->flag_init_done));
+	iwl_set_bit(trans, CSR_GP_CNTRL, CSR_GP_CNTRL_REG_FLAG_INIT_DONE);
 
 	if (cfg_trans->device_family == IWL_DEVICE_FAMILY_8000)
 		udelay(2);
@@ -518,8 +487,8 @@ int iwl_finish_nic_init(struct iwl_trans *trans,
 	 * and accesses to uCode SRAM.
 	 */
 	err = iwl_poll_bit(trans, CSR_GP_CNTRL,
-			   BIT(cfg_trans->csr->flag_mac_clock_ready),
-			   BIT(cfg_trans->csr->flag_mac_clock_ready),
+			   CSR_GP_CNTRL_REG_FLAG_MAC_CLOCK_READY,
+			   CSR_GP_CNTRL_REG_FLAG_MAC_CLOCK_READY,
 			   25000);
 	if (err < 0)
 		IWL_DEBUG_INFO(trans, "Failed to wake NIC\n");
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
index 1e240a2a8329..4e706c684b1d 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
@@ -801,12 +801,8 @@ static void iwl_flip_hw_address(__le32 mac_addr0, __le32 mac_addr1, u8 *dest)
 static void iwl_set_hw_address_from_csr(struct iwl_trans *trans,
 					struct iwl_nvm_data *data)
 {
-	__le32 mac_addr0 =
-		cpu_to_le32(iwl_read32(trans,
-				       trans->trans_cfg->csr->mac_addr0_strap));
-	__le32 mac_addr1 =
-		cpu_to_le32(iwl_read32(trans,
-				       trans->trans_cfg->csr->mac_addr1_strap));
+	__le32 mac_addr0 = cpu_to_le32(iwl_read32(trans, CSR_MAC_ADDR0_STRAP));
+	__le32 mac_addr1 = cpu_to_le32(iwl_read32(trans, CSR_MAC_ADDR1_STRAP));
 
 	iwl_flip_hw_address(mac_addr0, mac_addr1, data->hw_addr);
 	/*
@@ -816,10 +812,8 @@ static void iwl_set_hw_address_from_csr(struct iwl_trans *trans,
 	if (is_valid_ether_addr(data->hw_addr))
 		return;
 
-	mac_addr0 = cpu_to_le32(iwl_read32(trans,
-					trans->trans_cfg->csr->mac_addr0_otp));
-	mac_addr1 = cpu_to_le32(iwl_read32(trans,
-					trans->trans_cfg->csr->mac_addr1_otp));
+	mac_addr0 = cpu_to_le32(iwl_read32(trans, CSR_MAC_ADDR0_OTP));
+	mac_addr1 = cpu_to_le32(iwl_read32(trans, CSR_MAC_ADDR1_OTP));
 
 	iwl_flip_hw_address(mac_addr0, mac_addr1, data->hw_addr);
 }
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
index bac3d3fa0dec..79c24c696c8f 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
@@ -1007,12 +1007,6 @@ static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	/* the trans_cfg should never change, so set it now */
 	iwl_trans->trans_cfg = &cfg->trans;
 
-	if (WARN_ONCE(!iwl_trans->trans_cfg->csr,
-		      "CSR addresses aren't configured\n")) {
-		ret = -EINVAL;
-		goto out_free_trans;
-	}
-
 #if IS_ENABLED(CONFIG_IWLMVM)
 	/*
 	 * special-case 7265D, it has the same PCI IDs.
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
index 498a14ebac47..12fec6d59eb9 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
@@ -240,7 +240,7 @@ static void iwl_pcie_rxq_inc_wr_ptr(struct iwl_trans *trans,
 			IWL_DEBUG_INFO(trans, "Rx queue requesting wakeup, GP1 = 0x%x\n",
 				       reg);
 			iwl_set_bit(trans, CSR_GP_CNTRL,
-				    BIT(trans->trans_cfg->csr->flag_mac_access_req));
+				    CSR_GP_CNTRL_REG_FLAG_MAC_ACCESS_REQ);
 			rxq->need_update = true;
 			return;
 		}
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c
index 0252716c0b24..c1bb7150489a 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c
@@ -157,8 +157,7 @@ static void iwl_pcie_gen2_apm_stop(struct iwl_trans *trans, bool op_mode_leave)
 	 * Clear "initialization complete" bit to move adapter from
 	 * D0A* (powered-up Active) --> D0U* (Uninitialized) state.
 	 */
-	iwl_clear_bit(trans, CSR_GP_CNTRL,
-		      BIT(trans->trans_cfg->csr->flag_init_done));
+	iwl_clear_bit(trans, CSR_GP_CNTRL, CSR_GP_CNTRL_REG_FLAG_INIT_DONE);
 }
 
 void _iwl_trans_pcie_gen2_stop_device(struct iwl_trans *trans)
@@ -200,7 +199,7 @@ void _iwl_trans_pcie_gen2_stop_device(struct iwl_trans *trans)
 
 	/* Make sure (redundant) we've released our request to stay awake */
 	iwl_clear_bit(trans, CSR_GP_CNTRL,
-		      BIT(trans->trans_cfg->csr->flag_mac_access_req));
+		      CSR_GP_CNTRL_REG_FLAG_MAC_ACCESS_REQ);
 
 	/* Stop the device, and put it in low power state */
 	iwl_pcie_gen2_apm_stop(trans, false);
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
index cb4915451792..3db1313f9270 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
@@ -183,8 +183,7 @@ out:
 static void iwl_trans_pcie_sw_reset(struct iwl_trans *trans)
 {
 	/* Reset entire device - do controller reset (results in SHRD_HW_RST) */
-	iwl_set_bit(trans, trans->trans_cfg->csr->addr_sw_reset,
-		    BIT(trans->trans_cfg->csr->flag_sw_reset));
+	iwl_set_bit(trans, CSR_RESET, CSR_RESET_REG_FLAG_SW_RESET);
 	usleep_range(5000, 6000);
 }
 
@@ -487,8 +486,7 @@ static void iwl_pcie_apm_lp_xtal_enable(struct iwl_trans *trans)
 	 * Clear "initialization complete" bit to move adapter from
 	 * D0A* (powered-up Active) --> D0U* (Uninitialized) state.
 	 */
-	iwl_clear_bit(trans, CSR_GP_CNTRL,
-		      BIT(trans->trans_cfg->csr->flag_init_done));
+	iwl_clear_bit(trans, CSR_GP_CNTRL, CSR_GP_CNTRL_REG_FLAG_INIT_DONE);
 
 	/* Activates XTAL resources monitor */
 	__iwl_trans_pcie_set_bit(trans, CSR_MONITOR_CFG_REG,
@@ -510,12 +508,11 @@ void iwl_pcie_apm_stop_master(struct iwl_trans *trans)
 	int ret;
 
 	/* stop device's busmaster DMA activity */
-	iwl_set_bit(trans, trans->trans_cfg->csr->addr_sw_reset,
-		    BIT(trans->trans_cfg->csr->flag_stop_master));
+	iwl_set_bit(trans, CSR_RESET, CSR_RESET_REG_FLAG_STOP_MASTER);
 
-	ret = iwl_poll_bit(trans, trans->trans_cfg->csr->addr_sw_reset,
-			   BIT(trans->trans_cfg->csr->flag_master_dis),
-			   BIT(trans->trans_cfg->csr->flag_master_dis), 100);
+	ret = iwl_poll_bit(trans, CSR_RESET,
+			   CSR_RESET_REG_FLAG_MASTER_DISABLED,
+			   CSR_RESET_REG_FLAG_MASTER_DISABLED, 100);
 	if (ret < 0)
 		IWL_WARN(trans, "Master Disable Timed Out, 100 usec\n");
 
@@ -564,8 +561,7 @@ static void iwl_pcie_apm_stop(struct iwl_trans *trans, bool op_mode_leave)
 	 * Clear "initialization complete" bit to move adapter from
 	 * D0A* (powered-up Active) --> D0U* (Uninitialized) state.
 	 */
-	iwl_clear_bit(trans, CSR_GP_CNTRL,
-		      BIT(trans->trans_cfg->csr->flag_init_done));
+	iwl_clear_bit(trans, CSR_GP_CNTRL, CSR_GP_CNTRL_REG_FLAG_INIT_DONE);
 }
 
 static int iwl_pcie_nic_init(struct iwl_trans *trans)
@@ -1270,7 +1266,7 @@ static void _iwl_trans_pcie_stop_device(struct iwl_trans *trans)
 
 	/* Make sure (redundant) we've released our request to stay awake */
 	iwl_clear_bit(trans, CSR_GP_CNTRL,
-		      BIT(trans->trans_cfg->csr->flag_mac_access_req));
+		      CSR_GP_CNTRL_REG_FLAG_MAC_ACCESS_REQ);
 
 	/* Stop the device, and put it in low power state */
 	iwl_pcie_apm_stop(trans, false);
@@ -1494,9 +1490,8 @@ void iwl_pcie_d3_complete_suspend(struct iwl_trans *trans,
 	iwl_pcie_synchronize_irqs(trans);
 
 	iwl_clear_bit(trans, CSR_GP_CNTRL,
-		      BIT(trans->trans_cfg->csr->flag_mac_access_req));
-	iwl_clear_bit(trans, CSR_GP_CNTRL,
-		      BIT(trans->trans_cfg->csr->flag_init_done));
+		      CSR_GP_CNTRL_REG_FLAG_MAC_ACCESS_REQ);
+	iwl_clear_bit(trans, CSR_GP_CNTRL, CSR_GP_CNTRL_REG_FLAG_INIT_DONE);
 
 	if (reset) {
 		/*
@@ -1561,7 +1556,7 @@ static int iwl_trans_pcie_d3_resume(struct iwl_trans *trans,
 	}
 
 	iwl_set_bit(trans, CSR_GP_CNTRL,
-		    BIT(trans->trans_cfg->csr->flag_mac_access_req));
+		    CSR_GP_CNTRL_REG_FLAG_MAC_ACCESS_REQ);
 
 	ret = iwl_finish_nic_init(trans, trans->trans_cfg);
 	if (ret)
@@ -1583,7 +1578,7 @@ static int iwl_trans_pcie_d3_resume(struct iwl_trans *trans,
 
 	if (!reset) {
 		iwl_clear_bit(trans, CSR_GP_CNTRL,
-			      BIT(trans->trans_cfg->csr->flag_mac_access_req));
+			      CSR_GP_CNTRL_REG_FLAG_MAC_ACCESS_REQ);
 	} else {
 		iwl_trans_pcie_tx_reset(trans);
 
@@ -2029,7 +2024,7 @@ static bool iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans,
 
 	/* this bit wakes up the NIC */
 	__iwl_trans_pcie_set_bit(trans, CSR_GP_CNTRL,
-				 BIT(trans->trans_cfg->csr->flag_mac_access_req));
+				 CSR_GP_CNTRL_REG_FLAG_MAC_ACCESS_REQ);
 	if (trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_8000)
 		udelay(2);
 
@@ -2054,8 +2049,8 @@ static bool iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans,
 	 * and do not save/restore SRAM when power cycling.
 	 */
 	ret = iwl_poll_bit(trans, CSR_GP_CNTRL,
-			   BIT(trans->trans_cfg->csr->flag_val_mac_access_en),
-			   (BIT(trans->trans_cfg->csr->flag_mac_clock_ready) |
+			   CSR_GP_CNTRL_REG_VAL_MAC_ACCESS_EN,
+			   (CSR_GP_CNTRL_REG_FLAG_MAC_CLOCK_READY |
 			    CSR_GP_CNTRL_REG_FLAG_GOING_TO_SLEEP), 15000);
 	if (unlikely(ret < 0)) {
 		u32 cntrl = iwl_read32(trans, CSR_GP_CNTRL);
@@ -2137,7 +2132,7 @@ static void iwl_trans_pcie_release_nic_access(struct iwl_trans *trans,
 		goto out;
 
 	__iwl_trans_pcie_clear_bit(trans, CSR_GP_CNTRL,
-				   BIT(trans->trans_cfg->csr->flag_mac_access_req));
+				   CSR_GP_CNTRL_REG_FLAG_MAC_ACCESS_REQ);
 	/*
 	 * Above we read the CSR_GP_CNTRL register, which will flush
 	 * any previous writes, but we need the write that clears the
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
index f21f16ab2a97..9ac03122653c 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
@@ -306,7 +306,7 @@ static void iwl_pcie_txq_inc_wr_ptr(struct iwl_trans *trans,
 			IWL_DEBUG_INFO(trans, "Tx queue %d requesting wakeup, GP1 = 0x%x\n",
 				       txq_id, reg);
 			iwl_set_bit(trans, CSR_GP_CNTRL,
-				    BIT(trans->trans_cfg->csr->flag_mac_access_req));
+				    CSR_GP_CNTRL_REG_FLAG_MAC_ACCESS_REQ);
 			txq->need_update = true;
 			return;
 		}
@@ -646,7 +646,7 @@ static void iwl_pcie_clear_cmd_in_flight(struct iwl_trans *trans)
 
 	trans_pcie->cmd_hold_nic_awake = false;
 	__iwl_trans_pcie_clear_bit(trans, CSR_GP_CNTRL,
-				   BIT(trans->trans_cfg->csr->flag_mac_access_req));
+				   CSR_GP_CNTRL_REG_FLAG_MAC_ACCESS_REQ);
 }
 
 /*
@@ -1255,16 +1255,16 @@ static int iwl_pcie_set_cmd_in_flight(struct iwl_trans *trans,
 	if (trans->trans_cfg->base_params->apmg_wake_up_wa &&
 	    !trans_pcie->cmd_hold_nic_awake) {
 		__iwl_trans_pcie_set_bit(trans, CSR_GP_CNTRL,
-					 BIT(trans->trans_cfg->csr->flag_mac_access_req));
+					 CSR_GP_CNTRL_REG_FLAG_MAC_ACCESS_REQ);
 
 		ret = iwl_poll_bit(trans, CSR_GP_CNTRL,
-				   BIT(trans->trans_cfg->csr->flag_val_mac_access_en),
-				   (BIT(trans->trans_cfg->csr->flag_mac_clock_ready) |
+				   CSR_GP_CNTRL_REG_VAL_MAC_ACCESS_EN,
+				   (CSR_GP_CNTRL_REG_FLAG_MAC_CLOCK_READY |
 				    CSR_GP_CNTRL_REG_FLAG_GOING_TO_SLEEP),
 				   15000);
 		if (ret < 0) {
 			__iwl_trans_pcie_clear_bit(trans, CSR_GP_CNTRL,
-					BIT(trans->trans_cfg->csr->flag_mac_access_req));
+					CSR_GP_CNTRL_REG_FLAG_MAC_ACCESS_REQ);
 			IWL_ERR(trans, "Failed to wake NIC for hcmd\n");
 			return -EIO;
 		}
-- 
cgit v1.2.3-59-g8ed1b


From e7dbfed1adb0e9123f582293e79760dc52b00d74 Mon Sep 17 00:00:00 2001
From: Martin Varghese <martin.varghese@nokia.com>
Date: Sat, 21 Dec 2019 08:50:01 +0530
Subject: net: skb_mpls_push() modified to allow MPLS header push at start of
 packet.

The existing skb_mpls_push() implementation always inserts mpls header
after the mac header. L2 VPN use cases requires MPLS header to be
inserted before the ethernet header as the ethernet packet gets tunnelled
inside MPLS header in those cases.

Signed-off-by: Martin Varghese <martin.varghese@nokia.com>
Acked-by: Pravin B Shelar <pshelar@ovn.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/skbuff.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 973a71f4bc89..d90c8276a9ee 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -5472,12 +5472,15 @@ static void skb_mod_eth_type(struct sk_buff *skb, struct ethhdr *hdr,
 }
 
 /**
- * skb_mpls_push() - push a new MPLS header after the mac header
+ * skb_mpls_push() - push a new MPLS header after mac_len bytes from start of
+ *                   the packet
  *
  * @skb: buffer
  * @mpls_lse: MPLS label stack entry to push
  * @mpls_proto: ethertype of the new MPLS header (expects 0x8847 or 0x8848)
  * @mac_len: length of the MAC header
+ * @ethernet: flag to indicate if the resulting packet after skb_mpls_push is
+ *            ethernet
  *
  * Expects skb->data at mac header.
  *
@@ -5501,7 +5504,7 @@ int skb_mpls_push(struct sk_buff *skb, __be32 mpls_lse, __be16 mpls_proto,
 		return err;
 
 	if (!skb->inner_protocol) {
-		skb_set_inner_network_header(skb, mac_len);
+		skb_set_inner_network_header(skb, skb_network_offset(skb));
 		skb_set_inner_protocol(skb, skb->protocol);
 	}
 
@@ -5510,6 +5513,7 @@ int skb_mpls_push(struct sk_buff *skb, __be32 mpls_lse, __be16 mpls_proto,
 		mac_len);
 	skb_reset_mac_header(skb);
 	skb_set_network_header(skb, mac_len);
+	skb_reset_mac_len(skb);
 
 	lse = mpls_hdr(skb);
 	lse->label_stack_entry = mpls_lse;
-- 
cgit v1.2.3-59-g8ed1b


From 76f99f987f2b7f95d43857b6e0362bd8dec9890c Mon Sep 17 00:00:00 2001
From: Martin Varghese <martin.varghese@nokia.com>
Date: Sat, 21 Dec 2019 08:50:23 +0530
Subject: net: Rephrased comments section of skb_mpls_pop()

Rephrased comments section of skb_mpls_pop() to align it with
comments section of skb_mpls_push().

Signed-off-by: Martin Varghese <martin.varghese@nokia.com>
Acked-by: Pravin B Shelar <pshelar@ovn.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/skbuff.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index d90c8276a9ee..44b0894d8ae1 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -5533,7 +5533,7 @@ EXPORT_SYMBOL_GPL(skb_mpls_push);
  * @skb: buffer
  * @next_proto: ethertype of header after popped MPLS header
  * @mac_len: length of the MAC header
- * @ethernet: flag to indicate if ethernet header is present in packet
+ * @ethernet: flag to indicate if the packet is ethernet
  *
  * Expects skb->data at mac header.
  *
-- 
cgit v1.2.3-59-g8ed1b


From f66b53fdbb22ced1a323b22b9de84a61aacd8d18 Mon Sep 17 00:00:00 2001
From: Martin Varghese <martin.varghese@nokia.com>
Date: Sat, 21 Dec 2019 08:50:46 +0530
Subject: openvswitch: New MPLS actions for layer 2 tunnelling

The existing PUSH MPLS action inserts MPLS header between ethernet header
and the IP header. Though this behaviour is fine for L3 VPN where an IP
packet is encapsulated inside a MPLS tunnel, it does not suffice the L2
VPN (l2 tunnelling) requirements. In L2 VPN the MPLS header should
encapsulate the ethernet packet.

The new mpls action ADD_MPLS inserts MPLS header at the start of the
packet or at the start of the l3 header depending on the value of l3 tunnel
flag in the ADD_MPLS arguments.

POP_MPLS action is extended to support ethertype 0x6558.

Signed-off-by: Martin Varghese <martin.varghese@nokia.com>
Acked-by: Pravin B Shelar <pshelar@ovn.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/uapi/linux/openvswitch.h | 31 +++++++++++++++++++++++++++++++
 net/openvswitch/actions.c        | 30 ++++++++++++++++++++++++------
 net/openvswitch/flow_netlink.c   | 34 ++++++++++++++++++++++++++++++++++
 3 files changed, 89 insertions(+), 6 deletions(-)

diff --git a/include/uapi/linux/openvswitch.h b/include/uapi/linux/openvswitch.h
index a87b44cd5590..ae2bff14e7e1 100644
--- a/include/uapi/linux/openvswitch.h
+++ b/include/uapi/linux/openvswitch.h
@@ -672,6 +672,32 @@ struct ovs_action_push_mpls {
 	__be16 mpls_ethertype; /* Either %ETH_P_MPLS_UC or %ETH_P_MPLS_MC */
 };
 
+/**
+ * struct ovs_action_add_mpls - %OVS_ACTION_ATTR_ADD_MPLS action
+ * argument.
+ * @mpls_lse: MPLS label stack entry to push.
+ * @mpls_ethertype: Ethertype to set in the encapsulating ethernet frame.
+ * @tun_flags: MPLS tunnel attributes.
+ *
+ * The only values @mpls_ethertype should ever be given are %ETH_P_MPLS_UC and
+ * %ETH_P_MPLS_MC, indicating MPLS unicast or multicast. Other are rejected.
+ */
+struct ovs_action_add_mpls {
+	__be32 mpls_lse;
+	__be16 mpls_ethertype; /* Either %ETH_P_MPLS_UC or %ETH_P_MPLS_MC */
+	__u16 tun_flags;
+};
+
+#define OVS_MPLS_L3_TUNNEL_FLAG_MASK  (1 << 0) /* Flag to specify the place of
+						* insertion of MPLS header.
+						* When false, the MPLS header
+						* will be inserted at the start
+						* of the packet.
+						* When true, the MPLS header
+						* will be inserted at the start
+						* of the l3 header.
+						*/
+
 /**
  * struct ovs_action_push_vlan - %OVS_ACTION_ATTR_PUSH_VLAN action argument.
  * @vlan_tpid: Tag protocol identifier (TPID) to push.
@@ -892,6 +918,10 @@ struct check_pkt_len_arg {
  * @OVS_ACTION_ATTR_CHECK_PKT_LEN: Check the packet length and execute a set
  * of actions if greater than the specified packet length, else execute
  * another set of actions.
+ * @OVS_ACTION_ATTR_ADD_MPLS: Push a new MPLS label stack entry at the
+ * start of the packet or at the start of the l3 header depending on the value
+ * of l3 tunnel flag in the tun_flags field of OVS_ACTION_ATTR_ADD_MPLS
+ * argument.
  *
  * Only a single header can be set with a single %OVS_ACTION_ATTR_SET.  Not all
  * fields within a header are modifiable, e.g. the IPv4 protocol and fragment
@@ -927,6 +957,7 @@ enum ovs_action_attr {
 	OVS_ACTION_ATTR_METER,        /* u32 meter ID. */
 	OVS_ACTION_ATTR_CLONE,        /* Nested OVS_CLONE_ATTR_*.  */
 	OVS_ACTION_ATTR_CHECK_PKT_LEN, /* Nested OVS_CHECK_PKT_LEN_ATTR_*. */
+	OVS_ACTION_ATTR_ADD_MPLS,     /* struct ovs_action_add_mpls. */
 
 	__OVS_ACTION_ATTR_MAX,	      /* Nothing past this will be accepted
 				       * from userspace. */
diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c
index 4c8395462303..7fbfe2adfffa 100644
--- a/net/openvswitch/actions.c
+++ b/net/openvswitch/actions.c
@@ -161,16 +161,17 @@ static int do_execute_actions(struct datapath *dp, struct sk_buff *skb,
 			      const struct nlattr *attr, int len);
 
 static int push_mpls(struct sk_buff *skb, struct sw_flow_key *key,
-		     const struct ovs_action_push_mpls *mpls)
+		     __be32 mpls_lse, __be16 mpls_ethertype, __u16 mac_len)
 {
 	int err;
 
-	err = skb_mpls_push(skb, mpls->mpls_lse, mpls->mpls_ethertype,
-			    skb->mac_len,
-			    ovs_key_mac_proto(key) == MAC_PROTO_ETHERNET);
+	err = skb_mpls_push(skb, mpls_lse, mpls_ethertype, mac_len, !!mac_len);
 	if (err)
 		return err;
 
+	if (!mac_len)
+		key->mac_proto = MAC_PROTO_NONE;
+
 	invalidate_flow_key(key);
 	return 0;
 }
@@ -185,6 +186,9 @@ static int pop_mpls(struct sk_buff *skb, struct sw_flow_key *key,
 	if (err)
 		return err;
 
+	if (ethertype == htons(ETH_P_TEB))
+		key->mac_proto = MAC_PROTO_ETHERNET;
+
 	invalidate_flow_key(key);
 	return 0;
 }
@@ -1229,10 +1233,24 @@ static int do_execute_actions(struct datapath *dp, struct sk_buff *skb,
 			execute_hash(skb, key, a);
 			break;
 
-		case OVS_ACTION_ATTR_PUSH_MPLS:
-			err = push_mpls(skb, key, nla_data(a));
+		case OVS_ACTION_ATTR_PUSH_MPLS: {
+			struct ovs_action_push_mpls *mpls = nla_data(a);
+
+			err = push_mpls(skb, key, mpls->mpls_lse,
+					mpls->mpls_ethertype, skb->mac_len);
 			break;
+		}
+		case OVS_ACTION_ATTR_ADD_MPLS: {
+			struct ovs_action_add_mpls *mpls = nla_data(a);
+			__u16 mac_len = 0;
+
+			if (mpls->tun_flags & OVS_MPLS_L3_TUNNEL_FLAG_MASK)
+				mac_len = skb->mac_len;
 
+			err = push_mpls(skb, key, mpls->mpls_lse,
+					mpls->mpls_ethertype, mac_len);
+			break;
+		}
 		case OVS_ACTION_ATTR_POP_MPLS:
 			err = pop_mpls(skb, key, nla_get_be16(a));
 			break;
diff --git a/net/openvswitch/flow_netlink.c b/net/openvswitch/flow_netlink.c
index 65c2e3458ff5..7da4230627f5 100644
--- a/net/openvswitch/flow_netlink.c
+++ b/net/openvswitch/flow_netlink.c
@@ -79,6 +79,7 @@ static bool actions_may_change_flow(const struct nlattr *actions)
 		case OVS_ACTION_ATTR_SET_MASKED:
 		case OVS_ACTION_ATTR_METER:
 		case OVS_ACTION_ATTR_CHECK_PKT_LEN:
+		case OVS_ACTION_ATTR_ADD_MPLS:
 		default:
 			return true;
 		}
@@ -3005,6 +3006,7 @@ static int __ovs_nla_copy_actions(struct net *net, const struct nlattr *attr,
 			[OVS_ACTION_ATTR_METER] = sizeof(u32),
 			[OVS_ACTION_ATTR_CLONE] = (u32)-1,
 			[OVS_ACTION_ATTR_CHECK_PKT_LEN] = (u32)-1,
+			[OVS_ACTION_ATTR_ADD_MPLS] = sizeof(struct ovs_action_add_mpls),
 		};
 		const struct ovs_action_push_vlan *vlan;
 		int type = nla_type(a);
@@ -3072,6 +3074,33 @@ static int __ovs_nla_copy_actions(struct net *net, const struct nlattr *attr,
 		case OVS_ACTION_ATTR_RECIRC:
 			break;
 
+		case OVS_ACTION_ATTR_ADD_MPLS: {
+			const struct ovs_action_add_mpls *mpls = nla_data(a);
+
+			if (!eth_p_mpls(mpls->mpls_ethertype))
+				return -EINVAL;
+
+			if (mpls->tun_flags & OVS_MPLS_L3_TUNNEL_FLAG_MASK) {
+				if (vlan_tci & htons(VLAN_CFI_MASK) ||
+				    (eth_type != htons(ETH_P_IP) &&
+				     eth_type != htons(ETH_P_IPV6) &&
+				     eth_type != htons(ETH_P_ARP) &&
+				     eth_type != htons(ETH_P_RARP) &&
+				     !eth_p_mpls(eth_type)))
+					return -EINVAL;
+				mpls_label_count++;
+			} else {
+				if (mac_proto == MAC_PROTO_ETHERNET) {
+					mpls_label_count = 1;
+					mac_proto = MAC_PROTO_NONE;
+				} else {
+					mpls_label_count++;
+				}
+			}
+			eth_type = mpls->mpls_ethertype;
+			break;
+		}
+
 		case OVS_ACTION_ATTR_PUSH_MPLS: {
 			const struct ovs_action_push_mpls *mpls = nla_data(a);
 
@@ -3109,6 +3138,11 @@ static int __ovs_nla_copy_actions(struct net *net, const struct nlattr *attr,
 			 * recirculation.
 			 */
 			proto = nla_get_be16(a);
+
+			if (proto == htons(ETH_P_TEB) &&
+			    mac_proto != MAC_PROTO_NONE)
+				return -EINVAL;
+
 			mpls_label_count--;
 
 			if (!eth_p_mpls(proto) || !mpls_label_count)
-- 
cgit v1.2.3-59-g8ed1b


From 3127f7c9b7da3f64a7b970bfbfa307f48adbb72e Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Sat, 21 Dec 2019 14:03:38 +0100
Subject: r8169: factor out rtl8168h_2_get_adc_bias_ioffset

Simplify and factor out this magic from rtl8168h_2_hw_phy_config()
and name it based on the vendor driver.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 39 ++++++++++++++++---------------
 1 file changed, 20 insertions(+), 19 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index a887b685d070..c845a5850ce9 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -3204,11 +3204,26 @@ static void rtl8168h_1_hw_phy_config(struct rtl8169_private *tp)
 	rtl_enable_eee(tp);
 }
 
+static u16 rtl8168h_2_get_adc_bias_ioffset(struct rtl8169_private *tp)
+{
+	u16 data1, data2, ioffset;
+
+	r8168_mac_ocp_write(tp, 0xdd02, 0x807d);
+	data1 = r8168_mac_ocp_read(tp, 0xdd02);
+	data2 = r8168_mac_ocp_read(tp, 0xdd00);
+
+	ioffset = (data2 >> 1) & 0x7ff8;
+	ioffset |= data2 & 0x0007;
+	if (data1 & BIT(7))
+		ioffset |= BIT(15);
+
+	return ioffset;
+}
+
 static void rtl8168h_2_hw_phy_config(struct rtl8169_private *tp)
 {
-	u16 ioffset_p3, ioffset_p2, ioffset_p1, ioffset_p0;
 	struct phy_device *phydev = tp->phydev;
-	u16 rlen;
+	u16 ioffset, rlen;
 	u32 data;
 
 	rtl_apply_firmware(tp);
@@ -3223,23 +3238,9 @@ static void rtl8168h_2_hw_phy_config(struct rtl8169_private *tp)
 	/* enable GPHY 10M */
 	phy_modify_paged(tp->phydev, 0x0a44, 0x11, 0, BIT(11));
 
-	r8168_mac_ocp_write(tp, 0xdd02, 0x807d);
-	data = r8168_mac_ocp_read(tp, 0xdd02);
-	ioffset_p3 = ((data & 0x80)>>7);
-	ioffset_p3 <<= 3;
-
-	data = r8168_mac_ocp_read(tp, 0xdd00);
-	ioffset_p3 |= ((data & (0xe000))>>13);
-	ioffset_p2 = ((data & (0x1e00))>>9);
-	ioffset_p1 = ((data & (0x01e0))>>5);
-	ioffset_p0 = ((data & 0x0010)>>4);
-	ioffset_p0 <<= 3;
-	ioffset_p0 |= (data & (0x07));
-	data = (ioffset_p3<<12)|(ioffset_p2<<8)|(ioffset_p1<<4)|(ioffset_p0);
-
-	if ((ioffset_p3 != 0x0f) || (ioffset_p2 != 0x0f) ||
-	    (ioffset_p1 != 0x0f) || (ioffset_p0 != 0x0f))
-		phy_write_paged(phydev, 0x0bcf, 0x16, data);
+	ioffset = rtl8168h_2_get_adc_bias_ioffset(tp);
+	if (ioffset != 0xffff)
+		phy_write_paged(phydev, 0x0bcf, 0x16, ioffset);
 
 	/* Modify rlen (TX LPF corner frequency) level */
 	data = phy_read_paged(phydev, 0x0bcd, 0x16);
-- 
cgit v1.2.3-59-g8ed1b


From 8722709b049473f9bd3ba9b817d52d9ab0a8e20b Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Sat, 21 Dec 2019 14:11:08 +0100
Subject: r8169: remove MAC workaround in rtl8168e_2_hw_phy_config

Due to recent changes we don't need the call to rtl_rar_exgmac_set()
and longer at this place. It's called from rtl_rar_set() which is
called in rtl_init_mac_address() and rtl8169_resume().

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index c845a5850ce9..38a09b5ee0b4 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -2961,9 +2961,6 @@ static void rtl8168e_2_hw_phy_config(struct rtl8169_private *tp)
 	rtl_writephy(tp, 0x1f, 0x0005);
 	rtl_w0w1_phy(tp, 0x01, 0x0100, 0x0000);
 	rtl_writephy(tp, 0x1f, 0x0000);
-
-	/* Broken BIOS workaround: feed GigaMAC registers with MAC address. */
-	rtl_rar_exgmac_set(tp, tp->dev->dev_addr);
 }
 
 static void rtl8168f_hw_phy_config(struct rtl8169_private *tp)
-- 
cgit v1.2.3-59-g8ed1b


From af7797785d61e1053d23764b5d3fbdf2ab926f31 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Sat, 21 Dec 2019 14:15:21 +0100
Subject: r8169: move enabling EEE to rtl8169_init_phy

Simplify the code by moving the call to rtl_enable_eee() from the
individual PHY configs to rtl8169_init_phy().

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 14 +++-----------
 1 file changed, 3 insertions(+), 11 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 38a09b5ee0b4..0161d839fa87 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -2951,7 +2951,6 @@ static void rtl8168e_2_hw_phy_config(struct rtl8169_private *tp)
 	r8168d_phy_param(phydev, 0x8b85, 0x0000, 0x4000);
 
 	rtl8168f_config_eee_phy(tp);
-	rtl_enable_eee(tp);
 
 	/* Green feature */
 	rtl_writephy(tp, 0x1f, 0x0003);
@@ -2978,7 +2977,6 @@ static void rtl8168f_hw_phy_config(struct rtl8169_private *tp)
 	r8168d_phy_param(phydev, 0x8b86, 0x0000, 0x0001);
 
 	rtl8168f_config_eee_phy(tp);
-	rtl_enable_eee(tp);
 }
 
 static void rtl8168f_1_hw_phy_config(struct rtl8169_private *tp)
@@ -3124,14 +3122,12 @@ static void rtl8168g_1_hw_phy_config(struct rtl8169_private *tp)
 
 	rtl8168g_disable_aldps(tp);
 	rtl8168g_config_eee_phy(tp);
-	rtl_enable_eee(tp);
 }
 
 static void rtl8168g_2_hw_phy_config(struct rtl8169_private *tp)
 {
 	rtl_apply_firmware(tp);
 	rtl8168g_config_eee_phy(tp);
-	rtl_enable_eee(tp);
 }
 
 static void rtl8168h_1_hw_phy_config(struct rtl8169_private *tp)
@@ -3198,7 +3194,6 @@ static void rtl8168h_1_hw_phy_config(struct rtl8169_private *tp)
 
 	rtl8168g_disable_aldps(tp);
 	rtl8168h_config_eee_phy(tp);
-	rtl_enable_eee(tp);
 }
 
 static u16 rtl8168h_2_get_adc_bias_ioffset(struct rtl8169_private *tp)
@@ -3253,7 +3248,6 @@ static void rtl8168h_2_hw_phy_config(struct rtl8169_private *tp)
 
 	rtl8168g_disable_aldps(tp);
 	rtl8168g_config_eee_phy(tp);
-	rtl_enable_eee(tp);
 }
 
 static void rtl8168ep_1_hw_phy_config(struct rtl8169_private *tp)
@@ -3276,7 +3270,6 @@ static void rtl8168ep_1_hw_phy_config(struct rtl8169_private *tp)
 
 	rtl8168g_disable_aldps(tp);
 	rtl8168g_config_eee_phy(tp);
-	rtl_enable_eee(tp);
 }
 
 static void rtl8168ep_2_hw_phy_config(struct rtl8169_private *tp)
@@ -3328,7 +3321,6 @@ static void rtl8168ep_2_hw_phy_config(struct rtl8169_private *tp)
 
 	rtl8168g_disable_aldps(tp);
 	rtl8168g_config_eee_phy(tp);
-	rtl_enable_eee(tp);
 }
 
 static void rtl8117_hw_phy_config(struct rtl8169_private *tp)
@@ -3368,7 +3360,6 @@ static void rtl8117_hw_phy_config(struct rtl8169_private *tp)
 
 	rtl8168g_disable_aldps(tp);
 	rtl8168h_config_eee_phy(tp);
-	rtl_enable_eee(tp);
 }
 
 static void rtl8102e_hw_phy_config(struct rtl8169_private *tp)
@@ -3475,7 +3466,6 @@ static void rtl8125_1_hw_phy_config(struct rtl8169_private *tp)
 	phy_modify_paged(phydev, 0xa44, 0x11, 0x0000, 0x0800);
 
 	rtl8125_config_eee_phy(tp);
-	rtl_enable_eee(tp);
 }
 
 static void rtl8125_2_hw_phy_config(struct rtl8169_private *tp)
@@ -3541,7 +3531,6 @@ static void rtl8125_2_hw_phy_config(struct rtl8169_private *tp)
 	phy_modify_paged(phydev, 0xa44, 0x11, 0x0000, 0x0800);
 
 	rtl8125_config_eee_phy(tp);
-	rtl_enable_eee(tp);
 }
 
 static void rtl_hw_phy_config(struct net_device *dev)
@@ -3630,6 +3619,9 @@ static void rtl8169_init_phy(struct net_device *dev, struct rtl8169_private *tp)
 	/* We may have called phy_speed_down before */
 	phy_speed_up(tp->phydev);
 
+	if (rtl_supports_eee(tp))
+		rtl_enable_eee(tp);
+
 	genphy_soft_reset(tp->phydev);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 6b722237b656d045c0b9bab9966a5e46604258ba Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Mon, 23 Dec 2019 15:28:12 +0200
Subject: net: fib_notifier: Add temporary events to the FIB notification chain

Subsequent patches are going to simplify the IPv6 route offload API,
which will only use three events - replace, delete and append.

Introduce a temporary version of replace and delete in order to make the
conversion easier to review. Note that append does not need a temporary
version, as it is currently not used.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/fib_notifier.h | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/include/net/fib_notifier.h b/include/net/fib_notifier.h
index 6d59221ff05a..b3c54325caec 100644
--- a/include/net/fib_notifier.h
+++ b/include/net/fib_notifier.h
@@ -23,6 +23,8 @@ enum fib_event_type {
 	FIB_EVENT_NH_DEL,
 	FIB_EVENT_VIF_ADD,
 	FIB_EVENT_VIF_DEL,
+	FIB_EVENT_ENTRY_REPLACE_TMP,
+	FIB_EVENT_ENTRY_DEL_TMP,
 };
 
 struct fib_notifier_ops {
-- 
cgit v1.2.3-59-g8ed1b


From c10c4279c778df03f404a4d6906d7d4b840eb95f Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Mon, 23 Dec 2019 15:28:13 +0200
Subject: ipv6: Notify newly added route if should be offloaded

fib6_add_rt2node() takes care of adding a single route ('struct
fib6_info') to a FIB node. The route in question should only be notified
in case it is added as the first route in the node (lowest metric) or if
it is added as a sibling route to the first route in the node.

The first criterion can be tested by checking if the route is pointed to
by 'fn->leaf'. The second criterion can be tested by checking the new
'notify_sibling_rt' variable that is set when the route is added as a
sibling to the first route in the node.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/ip6_fib.c | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 7bae6a91b487..045bcaf5e770 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -1039,6 +1039,7 @@ static int fib6_add_rt2node(struct fib6_node *fn, struct fib6_info *rt,
 		   (info->nlh->nlmsg_flags & NLM_F_CREATE));
 	int found = 0;
 	bool rt_can_ecmp = rt6_qualify_for_ecmp(rt);
+	bool notify_sibling_rt = false;
 	u16 nlflags = NLM_F_EXCL;
 	int err;
 
@@ -1130,6 +1131,7 @@ next_iter:
 
 		/* Find the first route that have the same metric */
 		sibling = leaf;
+		notify_sibling_rt = true;
 		while (sibling) {
 			if (sibling->fib6_metric == rt->fib6_metric &&
 			    rt6_qualify_for_ecmp(sibling)) {
@@ -1139,6 +1141,7 @@ next_iter:
 			}
 			sibling = rcu_dereference_protected(sibling->fib6_next,
 				    lockdep_is_held(&rt->fib6_table->tb6_lock));
+			notify_sibling_rt = false;
 		}
 		/* For each sibling in the list, increment the counter of
 		 * siblings. BUG() if counters does not match, list of siblings
@@ -1166,6 +1169,21 @@ add:
 		nlflags |= NLM_F_CREATE;
 
 		if (!info->skip_notify_kernel) {
+			enum fib_event_type fib_event;
+
+			if (notify_sibling_rt)
+				fib_event = FIB_EVENT_ENTRY_APPEND;
+			else
+				fib_event = FIB_EVENT_ENTRY_REPLACE_TMP;
+			/* The route should only be notified if it is the first
+			 * route in the node or if it is added as a sibling
+			 * route to the first route in the node.
+			 */
+			if (notify_sibling_rt || ins == &fn->leaf)
+				err = call_fib6_entry_notifiers(info->nl_net,
+								fib_event, rt,
+								extack);
+
 			err = call_fib6_entry_notifiers(info->nl_net,
 							FIB_EVENT_ENTRY_ADD,
 							rt, extack);
-- 
cgit v1.2.3-59-g8ed1b


From 51bf7f387fdfe5ec8c33734b3124ccec83c8d0c3 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Mon, 23 Dec 2019 15:28:14 +0200
Subject: ipv6: Notify route if replacing currently offloaded one

Similar to the corresponding IPv4 patch, only notify the new route if it
is replacing the currently offloaded one. Meaning, the one pointed to by
'fn->leaf'.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/ip6_fib.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 045bcaf5e770..7cf9554888b0 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -1231,6 +1231,13 @@ add:
 		}
 
 		if (!info->skip_notify_kernel) {
+			enum fib_event_type fib_event;
+
+			fib_event = FIB_EVENT_ENTRY_REPLACE_TMP;
+			if (ins == &fn->leaf)
+				err = call_fib6_entry_notifiers(info->nl_net,
+								fib_event, rt,
+								extack);
 			err = call_fib6_entry_notifiers(info->nl_net,
 							FIB_EVENT_ENTRY_REPLACE,
 							rt, extack);
-- 
cgit v1.2.3-59-g8ed1b


From 0ee0f47c26b2f909ec95d7d8fed8119e288c4dd3 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Mon, 23 Dec 2019 15:28:15 +0200
Subject: ipv6: Notify multipath route if should be offloaded

In a similar fashion to previous patches, only notify the new multipath
route if it is the first route in the node or if it was appended to such
route.

The type of the notification (replace vs. append) is determined based on
the number of routes added ('nhn') and the number of sibling routes. If
the two do not match, then an append notification should be sent.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/route.c | 48 ++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 48 insertions(+)

diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index b59940416cb5..c0809f52f9ef 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -5017,6 +5017,32 @@ static void ip6_route_mpath_notify(struct fib6_info *rt,
 		inet6_rt_notify(RTM_NEWROUTE, rt, info, nlflags);
 }
 
+static bool ip6_route_mpath_should_notify(const struct fib6_info *rt)
+{
+	bool rt_can_ecmp = rt6_qualify_for_ecmp(rt);
+	bool should_notify = false;
+	struct fib6_info *leaf;
+	struct fib6_node *fn;
+
+	rcu_read_lock();
+	fn = rcu_dereference(rt->fib6_node);
+	if (!fn)
+		goto out;
+
+	leaf = rcu_dereference(fn->leaf);
+	if (!leaf)
+		goto out;
+
+	if (rt == leaf ||
+	    (rt_can_ecmp && rt->fib6_metric == leaf->fib6_metric &&
+	     rt6_qualify_for_ecmp(leaf)))
+		should_notify = true;
+out:
+	rcu_read_unlock();
+
+	return should_notify;
+}
+
 static int ip6_route_multipath_add(struct fib6_config *cfg,
 				   struct netlink_ext_ack *extack)
 {
@@ -5147,6 +5173,28 @@ static int ip6_route_multipath_add(struct fib6_config *cfg,
 		nhn++;
 	}
 
+	/* An in-kernel notification should only be sent in case the new
+	 * multipath route is added as the first route in the node, or if
+	 * it was appended to it. We pass 'rt_notif' since it is the first
+	 * sibling and might allow us to skip some checks in the replace case.
+	 */
+	if (ip6_route_mpath_should_notify(rt_notif)) {
+		enum fib_event_type fib_event;
+
+		if (rt_notif->fib6_nsiblings != nhn - 1)
+			fib_event = FIB_EVENT_ENTRY_APPEND;
+		else
+			fib_event = FIB_EVENT_ENTRY_REPLACE_TMP;
+
+		err = call_fib6_multipath_entry_notifiers(info->nl_net,
+							  fib_event, rt_notif,
+							  nhn - 1, extack);
+		if (err) {
+			/* Delete all the siblings that were just added */
+			err_nh = NULL;
+			goto add_errout;
+		}
+	}
 	event_type = replace ? FIB_EVENT_ENTRY_REPLACE : FIB_EVENT_ENTRY_ADD;
 	err = call_fib6_multipath_entry_notifiers(info->nl_net, event_type,
 						  rt_notif, nhn - 1, extack);
-- 
cgit v1.2.3-59-g8ed1b


From 9c6ecd3cf62d0eb57539f966d7ad617ad3a59f4b Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Mon, 23 Dec 2019 15:28:16 +0200
Subject: ipv6: Only Replay routes of interest to new listeners

When a new listener is registered to the FIB notification chain it
receives a dump of all the available routes in the system. Instead, make
sure to only replay the IPv6 routes that are actually used in the data
path and are of any interest to the new listener.

This is done by iterating over all the routing tables in the given
namespace, but from each traversed node only the first route ('leaf') is
notified. Multipath routes are notified in a single notification instead
of one for each nexthop.

Add fib6_rt_dump_tmp() to do that. Later on in the patch set it will be
renamed to fib6_rt_dump() instead of the existing one.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/ip6_fib.c | 40 ++++++++++++++++++++++++++++++++++++++++
 1 file changed, 40 insertions(+)

diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 7cf9554888b0..51cf848e38f0 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -370,6 +370,21 @@ static int call_fib6_entry_notifier(struct notifier_block *nb,
 	return call_fib6_notifier(nb, event_type, &info.info);
 }
 
+static int call_fib6_multipath_entry_notifier(struct notifier_block *nb,
+					      enum fib_event_type event_type,
+					      struct fib6_info *rt,
+					      unsigned int nsiblings,
+					      struct netlink_ext_ack *extack)
+{
+	struct fib6_entry_notifier_info info = {
+		.info.extack = extack,
+		.rt = rt,
+		.nsiblings = nsiblings,
+	};
+
+	return call_fib6_notifier(nb, event_type, &info.info);
+}
+
 int call_fib6_entry_notifiers(struct net *net,
 			      enum fib_event_type event_type,
 			      struct fib6_info *rt,
@@ -414,16 +429,41 @@ static int fib6_rt_dump(struct fib6_info *rt, struct fib6_dump_arg *arg)
 					rt, arg->extack);
 }
 
+static int fib6_rt_dump_tmp(struct fib6_info *rt, struct fib6_dump_arg *arg)
+{
+	enum fib_event_type fib_event = FIB_EVENT_ENTRY_REPLACE_TMP;
+	int err;
+
+	if (!rt || rt == arg->net->ipv6.fib6_null_entry)
+		return 0;
+
+	if (rt->fib6_nsiblings)
+		err = call_fib6_multipath_entry_notifier(arg->nb, fib_event,
+							 rt,
+							 rt->fib6_nsiblings,
+							 arg->extack);
+	else
+		err = call_fib6_entry_notifier(arg->nb, fib_event, rt,
+					       arg->extack);
+
+	return err;
+}
+
 static int fib6_node_dump(struct fib6_walker *w)
 {
 	struct fib6_info *rt;
 	int err = 0;
 
+	err = fib6_rt_dump_tmp(w->leaf, w->args);
+	if (err)
+		goto out;
+
 	for_each_fib6_walker_rt(w) {
 		err = fib6_rt_dump(rt, w->args);
 		if (err)
 			break;
 	}
+out:
 	w->leaf = NULL;
 	return err;
 }
-- 
cgit v1.2.3-59-g8ed1b


From d2f0c9b11410f9c6a07c126f8a215b0b81cdcf6c Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Mon, 23 Dec 2019 15:28:17 +0200
Subject: ipv6: Handle route deletion notification

For the purpose of route offload, when a single route is deleted, it is
only of interest if it is the first route in the node or if it is
sibling to such a route.

In the first case, distinguish between several possibilities:

1. Route is the last route in the node. Emit a delete notification

2. Route is followed by a non-multipath route. Emit a replace
notification for the non-multipath route.

3. Route is followed by a multipath route. Emit a replace notification
for the multipath route.

In the second case, only emit a delete notification to ensure the route
is no longer used as a valid nexthop.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_fib.h |  1 +
 net/ipv6/ip6_fib.c    | 44 +++++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 44 insertions(+), 1 deletion(-)

diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index f1535f172935..b579faea41e9 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -487,6 +487,7 @@ int call_fib6_multipath_entry_notifiers(struct net *net,
 					struct fib6_info *rt,
 					unsigned int nsiblings,
 					struct netlink_ext_ack *extack);
+int call_fib6_entry_notifiers_replace(struct net *net, struct fib6_info *rt);
 void fib6_rt_update(struct net *net, struct fib6_info *rt,
 		    struct nl_info *info);
 void inet6_rt_notify(int event, struct fib6_info *rt, struct nl_info *info,
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 51cf848e38f0..67ddee539f77 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -415,6 +415,18 @@ int call_fib6_multipath_entry_notifiers(struct net *net,
 	return call_fib6_notifiers(net, event_type, &info.info);
 }
 
+int call_fib6_entry_notifiers_replace(struct net *net, struct fib6_info *rt)
+{
+	struct fib6_entry_notifier_info info = {
+		.rt = rt,
+		.nsiblings = rt->fib6_nsiblings,
+	};
+
+	rt->fib6_table->fib_seq++;
+	return call_fib6_notifiers(net, FIB_EVENT_ENTRY_REPLACE_TMP,
+				   &info.info);
+}
+
 struct fib6_dump_arg {
 	struct net *net;
 	struct notifier_block *nb;
@@ -1910,13 +1922,29 @@ static struct fib6_node *fib6_repair_tree(struct net *net,
 static void fib6_del_route(struct fib6_table *table, struct fib6_node *fn,
 			   struct fib6_info __rcu **rtp, struct nl_info *info)
 {
+	struct fib6_info *leaf, *replace_rt = NULL;
 	struct fib6_walker *w;
 	struct fib6_info *rt = rcu_dereference_protected(*rtp,
 				    lockdep_is_held(&table->tb6_lock));
 	struct net *net = info->nl_net;
+	bool notify_del = false;
 
 	RT6_TRACE("fib6_del_route\n");
 
+	/* If the deleted route is the first in the node and it is not part of
+	 * a multipath route, then we need to replace it with the next route
+	 * in the node, if exists.
+	 */
+	leaf = rcu_dereference_protected(fn->leaf,
+					 lockdep_is_held(&table->tb6_lock));
+	if (leaf == rt && !rt->fib6_nsiblings) {
+		if (rcu_access_pointer(rt->fib6_next))
+			replace_rt = rcu_dereference_protected(rt->fib6_next,
+					    lockdep_is_held(&table->tb6_lock));
+		else
+			notify_del = true;
+	}
+
 	/* Unlink it */
 	*rtp = rt->fib6_next;
 	rt->fib6_node = NULL;
@@ -1934,6 +1962,14 @@ static void fib6_del_route(struct fib6_table *table, struct fib6_node *fn,
 	if (rt->fib6_nsiblings) {
 		struct fib6_info *sibling, *next_sibling;
 
+		/* The route is deleted from a multipath route. If this
+		 * multipath route is the first route in the node, then we need
+		 * to emit a delete notification. Otherwise, we need to skip
+		 * the notification.
+		 */
+		if (rt->fib6_metric == leaf->fib6_metric &&
+		    rt6_qualify_for_ecmp(leaf))
+			notify_del = true;
 		list_for_each_entry_safe(sibling, next_sibling,
 					 &rt->fib6_siblings, fib6_siblings)
 			sibling->fib6_nsiblings--;
@@ -1969,8 +2005,14 @@ static void fib6_del_route(struct fib6_table *table, struct fib6_node *fn,
 
 	fib6_purge_rt(rt, fn, net);
 
-	if (!info->skip_notify_kernel)
+	if (!info->skip_notify_kernel) {
+		if (notify_del)
+			call_fib6_entry_notifiers(net, FIB_EVENT_ENTRY_DEL_TMP,
+						  rt, NULL);
+		else if (replace_rt)
+			call_fib6_entry_notifiers_replace(net, replace_rt);
 		call_fib6_entry_notifiers(net, FIB_EVENT_ENTRY_DEL, rt, NULL);
+	}
 	if (!info->skip_notify)
 		inet6_rt_notify(RTM_DELROUTE, rt, info, 0);
 
-- 
cgit v1.2.3-59-g8ed1b


From 0284696b97b2fad1b220871559dff410cc3187e0 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Mon, 23 Dec 2019 15:28:18 +0200
Subject: ipv6: Handle multipath route deletion notification

When an entire multipath route is deleted, only emit a notification if
it is the first route in the node. Emit a replace notification in case
the last sibling is followed by another route. Otherwise, emit a delete
notification.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/route.c | 26 ++++++++++++++++++++++++++
 1 file changed, 26 insertions(+)

diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index c0809f52f9ef..646716a47cc9 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -3749,6 +3749,7 @@ static int __ip6_del_rt_siblings(struct fib6_info *rt, struct fib6_config *cfg)
 
 	if (rt->fib6_nsiblings && cfg->fc_delete_all_nh) {
 		struct fib6_info *sibling, *next_sibling;
+		struct fib6_node *fn;
 
 		/* prefer to send a single notification with all hops */
 		skb = nlmsg_new(rt6_nlmsg_size(rt), gfp_any());
@@ -3764,7 +3765,32 @@ static int __ip6_del_rt_siblings(struct fib6_info *rt, struct fib6_config *cfg)
 				info->skip_notify = 1;
 		}
 
+		/* 'rt' points to the first sibling route. If it is not the
+		 * leaf, then we do not need to send a notification. Otherwise,
+		 * we need to check if the last sibling has a next route or not
+		 * and emit a replace or delete notification, respectively.
+		 */
 		info->skip_notify_kernel = 1;
+		fn = rcu_dereference_protected(rt->fib6_node,
+					    lockdep_is_held(&table->tb6_lock));
+		if (rcu_access_pointer(fn->leaf) == rt) {
+			struct fib6_info *last_sibling, *replace_rt;
+
+			last_sibling = list_last_entry(&rt->fib6_siblings,
+						       struct fib6_info,
+						       fib6_siblings);
+			replace_rt = rcu_dereference_protected(
+					    last_sibling->fib6_next,
+					    lockdep_is_held(&table->tb6_lock));
+			if (replace_rt)
+				call_fib6_entry_notifiers_replace(net,
+								  replace_rt);
+			else
+				call_fib6_multipath_entry_notifiers(net,
+						       FIB_EVENT_ENTRY_DEL_TMP,
+						       rt, rt->fib6_nsiblings,
+						       NULL);
+		}
 		call_fib6_multipath_entry_notifiers(net,
 						    FIB_EVENT_ENTRY_DEL,
 						    rt,
-- 
cgit v1.2.3-59-g8ed1b


From dacad7b34b59fb06a242ef2cf64205f04f7c5339 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Mon, 23 Dec 2019 15:28:19 +0200
Subject: mlxsw: spectrum_router: Start using new IPv6 route notifications

With the new notifications mlxsw does not need to handle identical
routes itself, as this is taken care of by the core IPv6 code.

Instead, mlxsw only needs to take care of inserting and removing routes
from the device.

Convert mlxsw to use the new IPv6 route notifications and simplify the
code.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../net/ethernet/mellanox/mlxsw/spectrum_router.c  | 225 +++++++--------------
 1 file changed, 76 insertions(+), 149 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
index 0678815efbf3..295cdcb1c4c0 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
@@ -4989,13 +4989,6 @@ static void mlxsw_sp_rt6_destroy(struct mlxsw_sp_rt6 *mlxsw_sp_rt6)
 	kfree(mlxsw_sp_rt6);
 }
 
-static bool mlxsw_sp_fib6_rt_can_mp(const struct fib6_info *rt)
-{
-	/* RTF_CACHE routes are ignored */
-	return !(rt->fib6_flags & RTF_ADDRCONF) &&
-		rt->fib6_nh->fib_nh_gw_family;
-}
-
 static struct fib6_info *
 mlxsw_sp_fib6_entry_rt(const struct mlxsw_sp_fib6_entry *fib6_entry)
 {
@@ -5003,37 +4996,6 @@ mlxsw_sp_fib6_entry_rt(const struct mlxsw_sp_fib6_entry *fib6_entry)
 				list)->rt;
 }
 
-static struct mlxsw_sp_fib6_entry *
-mlxsw_sp_fib6_node_mp_entry_find(const struct mlxsw_sp_fib_node *fib_node,
-				 const struct fib6_info *nrt, bool replace)
-{
-	struct mlxsw_sp_fib6_entry *fib6_entry;
-
-	if (!mlxsw_sp_fib6_rt_can_mp(nrt) || replace)
-		return NULL;
-
-	list_for_each_entry(fib6_entry, &fib_node->entry_list, common.list) {
-		struct fib6_info *rt = mlxsw_sp_fib6_entry_rt(fib6_entry);
-
-		/* RT6_TABLE_LOCAL and RT6_TABLE_MAIN share the same
-		 * virtual router.
-		 */
-		if (rt->fib6_table->tb6_id > nrt->fib6_table->tb6_id)
-			continue;
-		if (rt->fib6_table->tb6_id != nrt->fib6_table->tb6_id)
-			break;
-		if (rt->fib6_metric < nrt->fib6_metric)
-			continue;
-		if (rt->fib6_metric == nrt->fib6_metric &&
-		    mlxsw_sp_fib6_rt_can_mp(rt))
-			return fib6_entry;
-		if (rt->fib6_metric > nrt->fib6_metric)
-			break;
-	}
-
-	return NULL;
-}
-
 static struct mlxsw_sp_rt6 *
 mlxsw_sp_fib6_entry_rt_find(const struct mlxsw_sp_fib6_entry *fib6_entry,
 			    const struct fib6_info *rt)
@@ -5424,86 +5386,13 @@ static void mlxsw_sp_fib6_entry_destroy(struct mlxsw_sp *mlxsw_sp,
 	kfree(fib6_entry);
 }
 
-static struct mlxsw_sp_fib6_entry *
-mlxsw_sp_fib6_node_entry_find(const struct mlxsw_sp_fib_node *fib_node,
-			      const struct fib6_info *nrt, bool replace)
-{
-	struct mlxsw_sp_fib6_entry *fib6_entry, *fallback = NULL;
-
-	list_for_each_entry(fib6_entry, &fib_node->entry_list, common.list) {
-		struct fib6_info *rt = mlxsw_sp_fib6_entry_rt(fib6_entry);
-
-		if (rt->fib6_table->tb6_id > nrt->fib6_table->tb6_id)
-			continue;
-		if (rt->fib6_table->tb6_id != nrt->fib6_table->tb6_id)
-			break;
-		if (replace && rt->fib6_metric == nrt->fib6_metric) {
-			if (mlxsw_sp_fib6_rt_can_mp(rt) ==
-			    mlxsw_sp_fib6_rt_can_mp(nrt))
-				return fib6_entry;
-			if (mlxsw_sp_fib6_rt_can_mp(nrt))
-				fallback = fallback ?: fib6_entry;
-		}
-		if (rt->fib6_metric > nrt->fib6_metric)
-			return fallback ?: fib6_entry;
-	}
-
-	return fallback;
-}
-
-static int
-mlxsw_sp_fib6_node_list_insert(struct mlxsw_sp_fib6_entry *new6_entry,
-			       bool *p_replace)
-{
-	struct mlxsw_sp_fib_node *fib_node = new6_entry->common.fib_node;
-	struct fib6_info *nrt = mlxsw_sp_fib6_entry_rt(new6_entry);
-	struct mlxsw_sp_fib6_entry *fib6_entry;
-
-	fib6_entry = mlxsw_sp_fib6_node_entry_find(fib_node, nrt, *p_replace);
-
-	if (*p_replace && !fib6_entry)
-		*p_replace = false;
-
-	if (fib6_entry) {
-		list_add_tail(&new6_entry->common.list,
-			      &fib6_entry->common.list);
-	} else {
-		struct mlxsw_sp_fib6_entry *last;
-
-		list_for_each_entry(last, &fib_node->entry_list, common.list) {
-			struct fib6_info *rt = mlxsw_sp_fib6_entry_rt(last);
-
-			if (nrt->fib6_table->tb6_id > rt->fib6_table->tb6_id)
-				break;
-			fib6_entry = last;
-		}
-
-		if (fib6_entry)
-			list_add(&new6_entry->common.list,
-				 &fib6_entry->common.list);
-		else
-			list_add(&new6_entry->common.list,
-				 &fib_node->entry_list);
-	}
-
-	return 0;
-}
-
-static void
-mlxsw_sp_fib6_node_list_remove(struct mlxsw_sp_fib6_entry *fib6_entry)
-{
-	list_del(&fib6_entry->common.list);
-}
-
 static int mlxsw_sp_fib6_node_entry_link(struct mlxsw_sp *mlxsw_sp,
-					 struct mlxsw_sp_fib6_entry *fib6_entry,
-					 bool *p_replace)
+					 struct mlxsw_sp_fib6_entry *fib6_entry)
 {
+	struct mlxsw_sp_fib_node *fib_node = fib6_entry->common.fib_node;
 	int err;
 
-	err = mlxsw_sp_fib6_node_list_insert(fib6_entry, p_replace);
-	if (err)
-		return err;
+	list_add(&fib6_entry->common.list, &fib_node->entry_list);
 
 	err = mlxsw_sp_fib_node_entry_add(mlxsw_sp, &fib6_entry->common);
 	if (err)
@@ -5512,7 +5401,7 @@ static int mlxsw_sp_fib6_node_entry_link(struct mlxsw_sp *mlxsw_sp,
 	return 0;
 
 err_fib_node_entry_add:
-	mlxsw_sp_fib6_node_list_remove(fib6_entry);
+	list_del(&fib6_entry->common.list);
 	return err;
 }
 
@@ -5521,7 +5410,7 @@ mlxsw_sp_fib6_node_entry_unlink(struct mlxsw_sp *mlxsw_sp,
 				struct mlxsw_sp_fib6_entry *fib6_entry)
 {
 	mlxsw_sp_fib_node_entry_del(mlxsw_sp, &fib6_entry->common);
-	mlxsw_sp_fib6_node_list_remove(fib6_entry);
+	list_del(&fib6_entry->common.list);
 }
 
 static struct mlxsw_sp_fib6_entry *
@@ -5557,15 +5446,15 @@ mlxsw_sp_fib6_entry_lookup(struct mlxsw_sp *mlxsw_sp,
 }
 
 static void mlxsw_sp_fib6_entry_replace(struct mlxsw_sp *mlxsw_sp,
-					struct mlxsw_sp_fib6_entry *fib6_entry,
-					bool replace)
+					struct mlxsw_sp_fib6_entry *fib6_entry)
 {
 	struct mlxsw_sp_fib_node *fib_node = fib6_entry->common.fib_node;
 	struct mlxsw_sp_fib6_entry *replaced;
 
-	if (!replace)
+	if (list_is_singular(&fib_node->entry_list))
 		return;
 
+	/* We inserted the new entry before replaced one */
 	replaced = list_next_entry(fib6_entry, common.list);
 
 	mlxsw_sp_fib6_node_entry_unlink(mlxsw_sp, replaced);
@@ -5573,9 +5462,9 @@ static void mlxsw_sp_fib6_entry_replace(struct mlxsw_sp *mlxsw_sp,
 	mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
 }
 
-static int mlxsw_sp_router_fib6_add(struct mlxsw_sp *mlxsw_sp,
-				    struct fib6_info **rt_arr,
-				    unsigned int nrt6, bool replace)
+static int mlxsw_sp_router_fib6_replace(struct mlxsw_sp *mlxsw_sp,
+					struct fib6_info **rt_arr,
+					unsigned int nrt6)
 {
 	struct mlxsw_sp_fib6_entry *fib6_entry;
 	struct mlxsw_sp_fib_node *fib_node;
@@ -5599,18 +5488,6 @@ static int mlxsw_sp_router_fib6_add(struct mlxsw_sp *mlxsw_sp,
 	if (IS_ERR(fib_node))
 		return PTR_ERR(fib_node);
 
-	/* Before creating a new entry, try to append route to an existing
-	 * multipath entry.
-	 */
-	fib6_entry = mlxsw_sp_fib6_node_mp_entry_find(fib_node, rt, replace);
-	if (fib6_entry) {
-		err = mlxsw_sp_fib6_entry_nexthop_add(mlxsw_sp, fib6_entry,
-						      rt_arr, nrt6);
-		if (err)
-			goto err_fib6_entry_nexthop_add;
-		return 0;
-	}
-
 	fib6_entry = mlxsw_sp_fib6_entry_create(mlxsw_sp, fib_node, rt_arr,
 						nrt6);
 	if (IS_ERR(fib6_entry)) {
@@ -5618,17 +5495,61 @@ static int mlxsw_sp_router_fib6_add(struct mlxsw_sp *mlxsw_sp,
 		goto err_fib6_entry_create;
 	}
 
-	err = mlxsw_sp_fib6_node_entry_link(mlxsw_sp, fib6_entry, &replace);
+	err = mlxsw_sp_fib6_node_entry_link(mlxsw_sp, fib6_entry);
 	if (err)
 		goto err_fib6_node_entry_link;
 
-	mlxsw_sp_fib6_entry_replace(mlxsw_sp, fib6_entry, replace);
+	mlxsw_sp_fib6_entry_replace(mlxsw_sp, fib6_entry);
 
 	return 0;
 
 err_fib6_node_entry_link:
 	mlxsw_sp_fib6_entry_destroy(mlxsw_sp, fib6_entry);
 err_fib6_entry_create:
+	mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
+	return err;
+}
+
+static int mlxsw_sp_router_fib6_append(struct mlxsw_sp *mlxsw_sp,
+				       struct fib6_info **rt_arr,
+				       unsigned int nrt6)
+{
+	struct mlxsw_sp_fib6_entry *fib6_entry;
+	struct mlxsw_sp_fib_node *fib_node;
+	struct fib6_info *rt = rt_arr[0];
+	int err;
+
+	if (mlxsw_sp->router->aborted)
+		return 0;
+
+	if (rt->fib6_src.plen)
+		return -EINVAL;
+
+	if (mlxsw_sp_fib6_rt_should_ignore(rt))
+		return 0;
+
+	fib_node = mlxsw_sp_fib_node_get(mlxsw_sp, rt->fib6_table->tb6_id,
+					 &rt->fib6_dst.addr,
+					 sizeof(rt->fib6_dst.addr),
+					 rt->fib6_dst.plen,
+					 MLXSW_SP_L3_PROTO_IPV6);
+	if (IS_ERR(fib_node))
+		return PTR_ERR(fib_node);
+
+	if (WARN_ON_ONCE(list_empty(&fib_node->entry_list))) {
+		mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
+		return -EINVAL;
+	}
+
+	fib6_entry = list_first_entry(&fib_node->entry_list,
+				      struct mlxsw_sp_fib6_entry, common.list);
+	err = mlxsw_sp_fib6_entry_nexthop_add(mlxsw_sp, fib6_entry, rt_arr,
+					      nrt6);
+	if (err)
+		goto err_fib6_entry_nexthop_add;
+
+	return 0;
+
 err_fib6_entry_nexthop_add:
 	mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
 	return err;
@@ -6039,25 +5960,29 @@ static void mlxsw_sp_router_fib6_event_work(struct work_struct *work)
 	struct mlxsw_sp_fib_event_work *fib_work =
 		container_of(work, struct mlxsw_sp_fib_event_work, work);
 	struct mlxsw_sp *mlxsw_sp = fib_work->mlxsw_sp;
-	bool replace;
 	int err;
 
 	rtnl_lock();
 	mlxsw_sp_span_respin(mlxsw_sp);
 
 	switch (fib_work->event) {
-	case FIB_EVENT_ENTRY_REPLACE: /* fall through */
-	case FIB_EVENT_ENTRY_ADD:
-		replace = fib_work->event == FIB_EVENT_ENTRY_REPLACE;
-		err = mlxsw_sp_router_fib6_add(mlxsw_sp,
-					       fib_work->fib6_work.rt_arr,
-					       fib_work->fib6_work.nrt6,
-					       replace);
+	case FIB_EVENT_ENTRY_REPLACE_TMP:
+		err = mlxsw_sp_router_fib6_replace(mlxsw_sp,
+						   fib_work->fib6_work.rt_arr,
+						   fib_work->fib6_work.nrt6);
 		if (err)
 			mlxsw_sp_router_fib_abort(mlxsw_sp);
 		mlxsw_sp_router_fib6_work_fini(&fib_work->fib6_work);
 		break;
-	case FIB_EVENT_ENTRY_DEL:
+	case FIB_EVENT_ENTRY_APPEND:
+		err = mlxsw_sp_router_fib6_append(mlxsw_sp,
+						  fib_work->fib6_work.rt_arr,
+						  fib_work->fib6_work.nrt6);
+		if (err)
+			mlxsw_sp_router_fib_abort(mlxsw_sp);
+		mlxsw_sp_router_fib6_work_fini(&fib_work->fib6_work);
+		break;
+	case FIB_EVENT_ENTRY_DEL_TMP:
 		mlxsw_sp_router_fib6_del(mlxsw_sp,
 					 fib_work->fib6_work.rt_arr,
 					 fib_work->fib6_work.nrt6);
@@ -6143,9 +6068,9 @@ static int mlxsw_sp_router_fib6_event(struct mlxsw_sp_fib_event_work *fib_work,
 	int err;
 
 	switch (fib_work->event) {
-	case FIB_EVENT_ENTRY_REPLACE: /* fall through */
-	case FIB_EVENT_ENTRY_ADD: /* fall through */
-	case FIB_EVENT_ENTRY_DEL:
+	case FIB_EVENT_ENTRY_REPLACE_TMP: /* fall through */
+	case FIB_EVENT_ENTRY_APPEND: /* fall through */
+	case FIB_EVENT_ENTRY_DEL_TMP:
 		fen6_info = container_of(info, struct fib6_entry_notifier_info,
 					 info);
 		err = mlxsw_sp_router_fib6_work_init(&fib_work->fib6_work,
@@ -6248,7 +6173,9 @@ static int mlxsw_sp_router_fib_event(struct notifier_block *nb,
 						     router->mlxsw_sp);
 		return notifier_from_errno(err);
 	case FIB_EVENT_ENTRY_ADD: /* fall through */
-	case FIB_EVENT_ENTRY_REPLACE:
+	case FIB_EVENT_ENTRY_REPLACE: /* fall through */
+	case FIB_EVENT_ENTRY_REPLACE_TMP: /* fall through */
+	case FIB_EVENT_ENTRY_APPEND:
 		if (router->aborted) {
 			NL_SET_ERR_MSG_MOD(info->extack, "FIB offload was aborted. Not configuring route");
 			return notifier_from_errno(-EINVAL);
-- 
cgit v1.2.3-59-g8ed1b


From caafb2509fac1432849650826953dd88b7cbe374 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Mon, 23 Dec 2019 15:28:20 +0200
Subject: ipv6: Remove old route notifications and convert listeners

Now that mlxsw is converted to use the new FIB notifications it is
possible to delete the old ones and use the new replace / append /
delete notifications.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../net/ethernet/mellanox/mlxsw/spectrum_router.c  |  9 ++--
 drivers/net/netdevsim/fib.c                        |  1 -
 include/net/fib_notifier.h                         |  2 -
 net/ipv6/ip6_fib.c                                 | 61 ++++++----------------
 net/ipv6/route.c                                   | 18 +------
 5 files changed, 21 insertions(+), 70 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
index 295cdcb1c4c0..f62e8d67348c 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
@@ -5966,7 +5966,7 @@ static void mlxsw_sp_router_fib6_event_work(struct work_struct *work)
 	mlxsw_sp_span_respin(mlxsw_sp);
 
 	switch (fib_work->event) {
-	case FIB_EVENT_ENTRY_REPLACE_TMP:
+	case FIB_EVENT_ENTRY_REPLACE:
 		err = mlxsw_sp_router_fib6_replace(mlxsw_sp,
 						   fib_work->fib6_work.rt_arr,
 						   fib_work->fib6_work.nrt6);
@@ -5982,7 +5982,7 @@ static void mlxsw_sp_router_fib6_event_work(struct work_struct *work)
 			mlxsw_sp_router_fib_abort(mlxsw_sp);
 		mlxsw_sp_router_fib6_work_fini(&fib_work->fib6_work);
 		break;
-	case FIB_EVENT_ENTRY_DEL_TMP:
+	case FIB_EVENT_ENTRY_DEL:
 		mlxsw_sp_router_fib6_del(mlxsw_sp,
 					 fib_work->fib6_work.rt_arr,
 					 fib_work->fib6_work.nrt6);
@@ -6068,9 +6068,9 @@ static int mlxsw_sp_router_fib6_event(struct mlxsw_sp_fib_event_work *fib_work,
 	int err;
 
 	switch (fib_work->event) {
-	case FIB_EVENT_ENTRY_REPLACE_TMP: /* fall through */
+	case FIB_EVENT_ENTRY_REPLACE: /* fall through */
 	case FIB_EVENT_ENTRY_APPEND: /* fall through */
-	case FIB_EVENT_ENTRY_DEL_TMP:
+	case FIB_EVENT_ENTRY_DEL:
 		fen6_info = container_of(info, struct fib6_entry_notifier_info,
 					 info);
 		err = mlxsw_sp_router_fib6_work_init(&fib_work->fib6_work,
@@ -6174,7 +6174,6 @@ static int mlxsw_sp_router_fib_event(struct notifier_block *nb,
 		return notifier_from_errno(err);
 	case FIB_EVENT_ENTRY_ADD: /* fall through */
 	case FIB_EVENT_ENTRY_REPLACE: /* fall through */
-	case FIB_EVENT_ENTRY_REPLACE_TMP: /* fall through */
 	case FIB_EVENT_ENTRY_APPEND:
 		if (router->aborted) {
 			NL_SET_ERR_MSG_MOD(info->extack, "FIB offload was aborted. Not configuring route");
diff --git a/drivers/net/netdevsim/fib.c b/drivers/net/netdevsim/fib.c
index 4e02a4231fcb..b5df308b4e33 100644
--- a/drivers/net/netdevsim/fib.c
+++ b/drivers/net/netdevsim/fib.c
@@ -178,7 +178,6 @@ static int nsim_fib_event_nb(struct notifier_block *nb, unsigned long event,
 		break;
 
 	case FIB_EVENT_ENTRY_REPLACE:  /* fall through */
-	case FIB_EVENT_ENTRY_ADD:  /* fall through */
 	case FIB_EVENT_ENTRY_DEL:
 		err = nsim_fib_event(data, info, event != FIB_EVENT_ENTRY_DEL);
 		break;
diff --git a/include/net/fib_notifier.h b/include/net/fib_notifier.h
index b3c54325caec..6d59221ff05a 100644
--- a/include/net/fib_notifier.h
+++ b/include/net/fib_notifier.h
@@ -23,8 +23,6 @@ enum fib_event_type {
 	FIB_EVENT_NH_DEL,
 	FIB_EVENT_VIF_ADD,
 	FIB_EVENT_VIF_DEL,
-	FIB_EVENT_ENTRY_REPLACE_TMP,
-	FIB_EVENT_ENTRY_DEL_TMP,
 };
 
 struct fib_notifier_ops {
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 67ddee539f77..b1e9a10e1133 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -423,8 +423,7 @@ int call_fib6_entry_notifiers_replace(struct net *net, struct fib6_info *rt)
 	};
 
 	rt->fib6_table->fib_seq++;
-	return call_fib6_notifiers(net, FIB_EVENT_ENTRY_REPLACE_TMP,
-				   &info.info);
+	return call_fib6_notifiers(net, FIB_EVENT_ENTRY_REPLACE, &info.info);
 }
 
 struct fib6_dump_arg {
@@ -435,15 +434,7 @@ struct fib6_dump_arg {
 
 static int fib6_rt_dump(struct fib6_info *rt, struct fib6_dump_arg *arg)
 {
-	if (rt == arg->net->ipv6.fib6_null_entry)
-		return 0;
-	return call_fib6_entry_notifier(arg->nb, FIB_EVENT_ENTRY_ADD,
-					rt, arg->extack);
-}
-
-static int fib6_rt_dump_tmp(struct fib6_info *rt, struct fib6_dump_arg *arg)
-{
-	enum fib_event_type fib_event = FIB_EVENT_ENTRY_REPLACE_TMP;
+	enum fib_event_type fib_event = FIB_EVENT_ENTRY_REPLACE;
 	int err;
 
 	if (!rt || rt == arg->net->ipv6.fib6_null_entry)
@@ -463,19 +454,9 @@ static int fib6_rt_dump_tmp(struct fib6_info *rt, struct fib6_dump_arg *arg)
 
 static int fib6_node_dump(struct fib6_walker *w)
 {
-	struct fib6_info *rt;
-	int err = 0;
-
-	err = fib6_rt_dump_tmp(w->leaf, w->args);
-	if (err)
-		goto out;
+	int err;
 
-	for_each_fib6_walker_rt(w) {
-		err = fib6_rt_dump(rt, w->args);
-		if (err)
-			break;
-	}
-out:
+	err = fib6_rt_dump(w->leaf, w->args);
 	w->leaf = NULL;
 	return err;
 }
@@ -1220,25 +1201,21 @@ next_iter:
 add:
 		nlflags |= NLM_F_CREATE;
 
-		if (!info->skip_notify_kernel) {
+		/* The route should only be notified if it is the first
+		 * route in the node or if it is added as a sibling
+		 * route to the first route in the node.
+		 */
+		if (!info->skip_notify_kernel &&
+		    (notify_sibling_rt || ins == &fn->leaf)) {
 			enum fib_event_type fib_event;
 
 			if (notify_sibling_rt)
 				fib_event = FIB_EVENT_ENTRY_APPEND;
 			else
-				fib_event = FIB_EVENT_ENTRY_REPLACE_TMP;
-			/* The route should only be notified if it is the first
-			 * route in the node or if it is added as a sibling
-			 * route to the first route in the node.
-			 */
-			if (notify_sibling_rt || ins == &fn->leaf)
-				err = call_fib6_entry_notifiers(info->nl_net,
-								fib_event, rt,
-								extack);
-
+				fib_event = FIB_EVENT_ENTRY_REPLACE;
 			err = call_fib6_entry_notifiers(info->nl_net,
-							FIB_EVENT_ENTRY_ADD,
-							rt, extack);
+							fib_event, rt,
+							extack);
 			if (err) {
 				struct fib6_info *sibling, *next_sibling;
 
@@ -1282,14 +1259,7 @@ add:
 			return -ENOENT;
 		}
 
-		if (!info->skip_notify_kernel) {
-			enum fib_event_type fib_event;
-
-			fib_event = FIB_EVENT_ENTRY_REPLACE_TMP;
-			if (ins == &fn->leaf)
-				err = call_fib6_entry_notifiers(info->nl_net,
-								fib_event, rt,
-								extack);
+		if (!info->skip_notify_kernel && ins == &fn->leaf) {
 			err = call_fib6_entry_notifiers(info->nl_net,
 							FIB_EVENT_ENTRY_REPLACE,
 							rt, extack);
@@ -2007,11 +1977,10 @@ static void fib6_del_route(struct fib6_table *table, struct fib6_node *fn,
 
 	if (!info->skip_notify_kernel) {
 		if (notify_del)
-			call_fib6_entry_notifiers(net, FIB_EVENT_ENTRY_DEL_TMP,
+			call_fib6_entry_notifiers(net, FIB_EVENT_ENTRY_DEL,
 						  rt, NULL);
 		else if (replace_rt)
 			call_fib6_entry_notifiers_replace(net, replace_rt);
-		call_fib6_entry_notifiers(net, FIB_EVENT_ENTRY_DEL, rt, NULL);
 	}
 	if (!info->skip_notify)
 		inet6_rt_notify(RTM_DELROUTE, rt, info, 0);
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 646716a47cc9..4b8659e077d3 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -3787,15 +3787,10 @@ static int __ip6_del_rt_siblings(struct fib6_info *rt, struct fib6_config *cfg)
 								  replace_rt);
 			else
 				call_fib6_multipath_entry_notifiers(net,
-						       FIB_EVENT_ENTRY_DEL_TMP,
+						       FIB_EVENT_ENTRY_DEL,
 						       rt, rt->fib6_nsiblings,
 						       NULL);
 		}
-		call_fib6_multipath_entry_notifiers(net,
-						    FIB_EVENT_ENTRY_DEL,
-						    rt,
-						    rt->fib6_nsiblings,
-						    NULL);
 		list_for_each_entry_safe(sibling, next_sibling,
 					 &rt->fib6_siblings,
 					 fib6_siblings) {
@@ -5074,7 +5069,6 @@ static int ip6_route_multipath_add(struct fib6_config *cfg,
 {
 	struct fib6_info *rt_notif = NULL, *rt_last = NULL;
 	struct nl_info *info = &cfg->fc_nlinfo;
-	enum fib_event_type event_type;
 	struct fib6_config r_cfg;
 	struct rtnexthop *rtnh;
 	struct fib6_info *rt;
@@ -5210,7 +5204,7 @@ static int ip6_route_multipath_add(struct fib6_config *cfg,
 		if (rt_notif->fib6_nsiblings != nhn - 1)
 			fib_event = FIB_EVENT_ENTRY_APPEND;
 		else
-			fib_event = FIB_EVENT_ENTRY_REPLACE_TMP;
+			fib_event = FIB_EVENT_ENTRY_REPLACE;
 
 		err = call_fib6_multipath_entry_notifiers(info->nl_net,
 							  fib_event, rt_notif,
@@ -5221,14 +5215,6 @@ static int ip6_route_multipath_add(struct fib6_config *cfg,
 			goto add_errout;
 		}
 	}
-	event_type = replace ? FIB_EVENT_ENTRY_REPLACE : FIB_EVENT_ENTRY_ADD;
-	err = call_fib6_multipath_entry_notifiers(info->nl_net, event_type,
-						  rt_notif, nhn - 1, extack);
-	if (err) {
-		/* Delete all the siblings that were just added */
-		err_nh = NULL;
-		goto add_errout;
-	}
 
 	/* success ... tell user about new route */
 	ip6_route_mpath_notify(rt_notif, rt_last, info, nlflags);
-- 
cgit v1.2.3-59-g8ed1b


From 0e5dafc8a6e540c0145b61545c557c43be70af10 Mon Sep 17 00:00:00 2001
From: Richard Cochran <richardcochran@gmail.com>
Date: Wed, 25 Dec 2019 18:16:09 -0800
Subject: net: phy: Introduce helper functions for time stamping support.

Some parts of the networking stack and at least one driver test fields
within the 'struct phy_device' in order to query time stamping
capabilities and to invoke time stamping methods.  This patch adds a
functional interface around the time stamping fields.  This will allow
insulating the callers from future changes to the details of the time
stamping implemenation.

Signed-off-by: Richard Cochran <richardcochran@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/phy.h | 60 +++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 60 insertions(+)

diff --git a/include/linux/phy.h b/include/linux/phy.h
index 7d530b3f8855..0248f5e9939d 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -936,6 +936,66 @@ static inline bool phy_polling_mode(struct phy_device *phydev)
 	return phydev->irq == PHY_POLL;
 }
 
+/**
+ * phy_has_hwtstamp - Tests whether a PHY time stamp configuration.
+ * @phydev: the phy_device struct
+ */
+static inline bool phy_has_hwtstamp(struct phy_device *phydev)
+{
+	return phydev && phydev->drv && phydev->drv->hwtstamp;
+}
+
+/**
+ * phy_has_rxtstamp - Tests whether a PHY supports receive time stamping.
+ * @phydev: the phy_device struct
+ */
+static inline bool phy_has_rxtstamp(struct phy_device *phydev)
+{
+	return phydev && phydev->drv && phydev->drv->rxtstamp;
+}
+
+/**
+ * phy_has_tsinfo - Tests whether a PHY reports time stamping and/or
+ * PTP hardware clock capabilities.
+ * @phydev: the phy_device struct
+ */
+static inline bool phy_has_tsinfo(struct phy_device *phydev)
+{
+	return phydev && phydev->drv && phydev->drv->ts_info;
+}
+
+/**
+ * phy_has_txtstamp - Tests whether a PHY supports transmit time stamping.
+ * @phydev: the phy_device struct
+ */
+static inline bool phy_has_txtstamp(struct phy_device *phydev)
+{
+	return phydev && phydev->drv && phydev->drv->txtstamp;
+}
+
+static inline int phy_hwtstamp(struct phy_device *phydev, struct ifreq *ifr)
+{
+	return phydev->drv->hwtstamp(phydev, ifr);
+}
+
+static inline bool phy_rxtstamp(struct phy_device *phydev, struct sk_buff *skb,
+				int type)
+{
+	return phydev->drv->rxtstamp(phydev, skb, type);
+}
+
+static inline int phy_ts_info(struct phy_device *phydev,
+			      struct ethtool_ts_info *tsinfo)
+{
+	return phydev->drv->ts_info(phydev, tsinfo);
+}
+
+static inline void phy_txtstamp(struct phy_device *phydev, struct sk_buff *skb,
+				int type)
+{
+	phydev->drv->txtstamp(phydev, skb, type);
+}
+
 /**
  * phy_is_internal - Convenience function for testing if a PHY is internal
  * @phydev: the phy_device struct
-- 
cgit v1.2.3-59-g8ed1b


From d25de984aaee5265ff0c2c4eb49e474096d29827 Mon Sep 17 00:00:00 2001
From: Richard Cochran <richardcochran@gmail.com>
Date: Wed, 25 Dec 2019 18:16:10 -0800
Subject: net: macvlan: Use the PHY time stamping interface.

The macvlan layer tests fields of the phy_device in order to determine
whether to invoke the PHY's tsinfo ethtool callback.  This patch
replaces the open coded logic with an invocation of the proper
methods.

Signed-off-by: Richard Cochran <richardcochran@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/macvlan.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c
index 05631d97eeb4..d066cf58c926 100644
--- a/drivers/net/macvlan.c
+++ b/drivers/net/macvlan.c
@@ -1036,8 +1036,8 @@ static int macvlan_ethtool_get_ts_info(struct net_device *dev,
 	const struct ethtool_ops *ops = real_dev->ethtool_ops;
 	struct phy_device *phydev = real_dev->phydev;
 
-	if (phydev && phydev->drv && phydev->drv->ts_info) {
-		 return phydev->drv->ts_info(phydev, info);
+	if (phy_has_tsinfo(phydev)) {
+		return phy_ts_info(phydev, info);
 	} else if (ops->get_ts_info) {
 		return ops->get_ts_info(real_dev, info);
 	} else {
-- 
cgit v1.2.3-59-g8ed1b


From dfe6d68fc4049c799b2194d88523679a694ead75 Mon Sep 17 00:00:00 2001
From: Richard Cochran <richardcochran@gmail.com>
Date: Wed, 25 Dec 2019 18:16:11 -0800
Subject: net: vlan: Use the PHY time stamping interface.

The vlan layer tests fields of the phy_device in order to determine
whether to invoke the PHY's tsinfo ethtool callback.  This patch
replaces the open coded logic with an invocation of the proper
methods.

Signed-off-by: Richard Cochran <richardcochran@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/8021q/vlan_dev.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index e5bff5cc6f97..5ff8059837b4 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -646,8 +646,8 @@ static int vlan_ethtool_get_ts_info(struct net_device *dev,
 	const struct ethtool_ops *ops = vlan->real_dev->ethtool_ops;
 	struct phy_device *phydev = vlan->real_dev->phydev;
 
-	if (phydev && phydev->drv && phydev->drv->ts_info) {
-		 return phydev->drv->ts_info(phydev, info);
+	if (phy_has_tsinfo(phydev)) {
+		return phy_ts_info(phydev, info);
 	} else if (ops->get_ts_info) {
 		return ops->get_ts_info(vlan->real_dev, info);
 	} else {
-- 
cgit v1.2.3-59-g8ed1b


From 7774ee23689d3c56ecb5eff6cd1bf6060bdd3d8e Mon Sep 17 00:00:00 2001
From: Richard Cochran <richardcochran@gmail.com>
Date: Wed, 25 Dec 2019 18:16:12 -0800
Subject: net: ethtool: Use the PHY time stamping interface.

The ethtool layer tests fields of the phy_device in order to determine
whether to invoke the PHY's tsinfo ethtool callback.  This patch
replaces the open coded logic with an invocation of the proper
methods.

Signed-off-by: Richard Cochran <richardcochran@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ethtool/ioctl.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
index aed2c2cf1623..88f7cddf5a6f 100644
--- a/net/ethtool/ioctl.c
+++ b/net/ethtool/ioctl.c
@@ -2096,8 +2096,8 @@ static int ethtool_get_ts_info(struct net_device *dev, void __user *useraddr)
 	memset(&info, 0, sizeof(info));
 	info.cmd = ETHTOOL_GET_TS_INFO;
 
-	if (phydev && phydev->drv && phydev->drv->ts_info) {
-		err = phydev->drv->ts_info(phydev, &info);
+	if (phy_has_tsinfo(phydev)) {
+		err = phy_ts_info(phydev, &info);
 	} else if (ops->get_ts_info) {
 		err = ops->get_ts_info(dev, &info);
 	} else {
-- 
cgit v1.2.3-59-g8ed1b


From bfd57b5900212acd5a14b9bea9ef49a5606d31a6 Mon Sep 17 00:00:00 2001
From: Richard Cochran <richardcochran@gmail.com>
Date: Wed, 25 Dec 2019 18:16:13 -0800
Subject: net: netcp_ethss: Use the PHY time stamping interface.

The netcp_ethss driver tests fields of the phy_device in order to
determine whether to defer to the PHY's time stamping functionality.
This patch replaces the open coded logic with an invocation of the
proper methods.

Signed-off-by: Richard Cochran <richardcochran@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/ti/netcp_ethss.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/ti/netcp_ethss.c b/drivers/net/ethernet/ti/netcp_ethss.c
index d6a192c1f337..fb36115e9c51 100644
--- a/drivers/net/ethernet/ti/netcp_ethss.c
+++ b/drivers/net/ethernet/ti/netcp_ethss.c
@@ -2533,8 +2533,6 @@ static int gbe_del_vid(void *intf_priv, int vid)
 }
 
 #if IS_ENABLED(CONFIG_TI_CPTS)
-#define HAS_PHY_TXTSTAMP(p) ((p)->drv && (p)->drv->txtstamp)
-#define HAS_PHY_RXTSTAMP(p) ((p)->drv && (p)->drv->rxtstamp)
 
 static void gbe_txtstamp(void *context, struct sk_buff *skb)
 {
@@ -2566,7 +2564,7 @@ static int gbe_txtstamp_mark_pkt(struct gbe_intf *gbe_intf,
 	 * We mark it here because skb_tx_timestamp() is called
 	 * after all the txhooks are called.
 	 */
-	if (phydev && HAS_PHY_TXTSTAMP(phydev)) {
+	if (phy_has_txtstamp(phydev)) {
 		skb_shinfo(p_info->skb)->tx_flags |= SKBTX_IN_PROGRESS;
 		return 0;
 	}
@@ -2588,7 +2586,7 @@ static int gbe_rxtstamp(struct gbe_intf *gbe_intf, struct netcp_packet *p_info)
 	if (p_info->rxtstamp_complete)
 		return 0;
 
-	if (phydev && HAS_PHY_RXTSTAMP(phydev)) {
+	if (phy_has_rxtstamp(phydev)) {
 		p_info->rxtstamp_complete = true;
 		return 0;
 	}
@@ -2830,7 +2828,7 @@ static int gbe_ioctl(void *intf_priv, struct ifreq *req, int cmd)
 	struct gbe_intf *gbe_intf = intf_priv;
 	struct phy_device *phy = gbe_intf->slave->phy;
 
-	if (!phy || !phy->drv->hwtstamp) {
+	if (!phy_has_hwtstamp(phy)) {
 		switch (cmd) {
 		case SIOCGHWTSTAMP:
 			return gbe_hwtstamp_get(gbe_intf, req);
-- 
cgit v1.2.3-59-g8ed1b


From 12d0efb9e6d90e6ff5c94e46d1ea273a41e36454 Mon Sep 17 00:00:00 2001
From: Richard Cochran <richardcochran@gmail.com>
Date: Wed, 25 Dec 2019 18:16:14 -0800
Subject: net: phy: dp83640: Move the probe and remove methods around.

An upcoming patch will change how the PHY time stamping functions are
registered with the networking stack, and adapting this driver would
entail adding forward declarations for four time stamping methods.
However, forward declarations are considered to be stylistic defects.
This patch avoids the issue by moving the probe and remove methods
immediately above the phy_driver interface structure.

Signed-off-by: Richard Cochran <richardcochran@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/dp83640.c | 180 +++++++++++++++++++++++-----------------------
 1 file changed, 90 insertions(+), 90 deletions(-)

diff --git a/drivers/net/phy/dp83640.c b/drivers/net/phy/dp83640.c
index 8f241b57fcf6..b58abdb5491e 100644
--- a/drivers/net/phy/dp83640.c
+++ b/drivers/net/phy/dp83640.c
@@ -1131,96 +1131,6 @@ static void dp83640_clock_put(struct dp83640_clock *clock)
 	mutex_unlock(&clock->clock_lock);
 }
 
-static int dp83640_probe(struct phy_device *phydev)
-{
-	struct dp83640_clock *clock;
-	struct dp83640_private *dp83640;
-	int err = -ENOMEM, i;
-
-	if (phydev->mdio.addr == BROADCAST_ADDR)
-		return 0;
-
-	clock = dp83640_clock_get_bus(phydev->mdio.bus);
-	if (!clock)
-		goto no_clock;
-
-	dp83640 = kzalloc(sizeof(struct dp83640_private), GFP_KERNEL);
-	if (!dp83640)
-		goto no_memory;
-
-	dp83640->phydev = phydev;
-	INIT_DELAYED_WORK(&dp83640->ts_work, rx_timestamp_work);
-
-	INIT_LIST_HEAD(&dp83640->rxts);
-	INIT_LIST_HEAD(&dp83640->rxpool);
-	for (i = 0; i < MAX_RXTS; i++)
-		list_add(&dp83640->rx_pool_data[i].list, &dp83640->rxpool);
-
-	phydev->priv = dp83640;
-
-	spin_lock_init(&dp83640->rx_lock);
-	skb_queue_head_init(&dp83640->rx_queue);
-	skb_queue_head_init(&dp83640->tx_queue);
-
-	dp83640->clock = clock;
-
-	if (choose_this_phy(clock, phydev)) {
-		clock->chosen = dp83640;
-		clock->ptp_clock = ptp_clock_register(&clock->caps,
-						      &phydev->mdio.dev);
-		if (IS_ERR(clock->ptp_clock)) {
-			err = PTR_ERR(clock->ptp_clock);
-			goto no_register;
-		}
-	} else
-		list_add_tail(&dp83640->list, &clock->phylist);
-
-	dp83640_clock_put(clock);
-	return 0;
-
-no_register:
-	clock->chosen = NULL;
-	kfree(dp83640);
-no_memory:
-	dp83640_clock_put(clock);
-no_clock:
-	return err;
-}
-
-static void dp83640_remove(struct phy_device *phydev)
-{
-	struct dp83640_clock *clock;
-	struct list_head *this, *next;
-	struct dp83640_private *tmp, *dp83640 = phydev->priv;
-
-	if (phydev->mdio.addr == BROADCAST_ADDR)
-		return;
-
-	enable_status_frames(phydev, false);
-	cancel_delayed_work_sync(&dp83640->ts_work);
-
-	skb_queue_purge(&dp83640->rx_queue);
-	skb_queue_purge(&dp83640->tx_queue);
-
-	clock = dp83640_clock_get(dp83640->clock);
-
-	if (dp83640 == clock->chosen) {
-		ptp_clock_unregister(clock->ptp_clock);
-		clock->chosen = NULL;
-	} else {
-		list_for_each_safe(this, next, &clock->phylist) {
-			tmp = list_entry(this, struct dp83640_private, list);
-			if (tmp == dp83640) {
-				list_del_init(&tmp->list);
-				break;
-			}
-		}
-	}
-
-	dp83640_clock_put(clock);
-	kfree(dp83640);
-}
-
 static int dp83640_soft_reset(struct phy_device *phydev)
 {
 	int ret;
@@ -1526,6 +1436,96 @@ static int dp83640_ts_info(struct phy_device *dev, struct ethtool_ts_info *info)
 	return 0;
 }
 
+static int dp83640_probe(struct phy_device *phydev)
+{
+	struct dp83640_clock *clock;
+	struct dp83640_private *dp83640;
+	int err = -ENOMEM, i;
+
+	if (phydev->mdio.addr == BROADCAST_ADDR)
+		return 0;
+
+	clock = dp83640_clock_get_bus(phydev->mdio.bus);
+	if (!clock)
+		goto no_clock;
+
+	dp83640 = kzalloc(sizeof(struct dp83640_private), GFP_KERNEL);
+	if (!dp83640)
+		goto no_memory;
+
+	dp83640->phydev = phydev;
+	INIT_DELAYED_WORK(&dp83640->ts_work, rx_timestamp_work);
+
+	INIT_LIST_HEAD(&dp83640->rxts);
+	INIT_LIST_HEAD(&dp83640->rxpool);
+	for (i = 0; i < MAX_RXTS; i++)
+		list_add(&dp83640->rx_pool_data[i].list, &dp83640->rxpool);
+
+	phydev->priv = dp83640;
+
+	spin_lock_init(&dp83640->rx_lock);
+	skb_queue_head_init(&dp83640->rx_queue);
+	skb_queue_head_init(&dp83640->tx_queue);
+
+	dp83640->clock = clock;
+
+	if (choose_this_phy(clock, phydev)) {
+		clock->chosen = dp83640;
+		clock->ptp_clock = ptp_clock_register(&clock->caps,
+						      &phydev->mdio.dev);
+		if (IS_ERR(clock->ptp_clock)) {
+			err = PTR_ERR(clock->ptp_clock);
+			goto no_register;
+		}
+	} else
+		list_add_tail(&dp83640->list, &clock->phylist);
+
+	dp83640_clock_put(clock);
+	return 0;
+
+no_register:
+	clock->chosen = NULL;
+	kfree(dp83640);
+no_memory:
+	dp83640_clock_put(clock);
+no_clock:
+	return err;
+}
+
+static void dp83640_remove(struct phy_device *phydev)
+{
+	struct dp83640_clock *clock;
+	struct list_head *this, *next;
+	struct dp83640_private *tmp, *dp83640 = phydev->priv;
+
+	if (phydev->mdio.addr == BROADCAST_ADDR)
+		return;
+
+	enable_status_frames(phydev, false);
+	cancel_delayed_work_sync(&dp83640->ts_work);
+
+	skb_queue_purge(&dp83640->rx_queue);
+	skb_queue_purge(&dp83640->tx_queue);
+
+	clock = dp83640_clock_get(dp83640->clock);
+
+	if (dp83640 == clock->chosen) {
+		ptp_clock_unregister(clock->ptp_clock);
+		clock->chosen = NULL;
+	} else {
+		list_for_each_safe(this, next, &clock->phylist) {
+			tmp = list_entry(this, struct dp83640_private, list);
+			if (tmp == dp83640) {
+				list_del_init(&tmp->list);
+				break;
+			}
+		}
+	}
+
+	dp83640_clock_put(clock);
+	kfree(dp83640);
+}
+
 static struct phy_driver dp83640_driver = {
 	.phy_id		= DP83640_PHY_ID,
 	.phy_id_mask	= 0xfffffff0,
-- 
cgit v1.2.3-59-g8ed1b


From 4715f65ffa0520af0680dbfbedbe349f175adaf4 Mon Sep 17 00:00:00 2001
From: Richard Cochran <richardcochran@gmail.com>
Date: Wed, 25 Dec 2019 18:16:15 -0800
Subject: net: Introduce a new MII time stamping interface.

Currently the stack supports time stamping in PHY devices.  However,
there are newer, non-PHY devices that can snoop an MII bus and provide
time stamps.  In order to support such devices, this patch introduces
a new interface to be used by both PHY and non-PHY devices.

In addition, the one and only user of the old PHY time stamping API is
converted to the new interface.

Signed-off-by: Richard Cochran <richardcochran@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/dp83640.c       | 39 ++++++++++++++++-----------
 drivers/net/phy/phy.c           |  4 +--
 drivers/net/phy/phy_device.c    |  2 ++
 include/linux/mii_timestamper.h | 58 +++++++++++++++++++++++++++++++++++++++++
 include/linux/phy.h             | 41 +++++++----------------------
 net/core/timestamping.c         | 20 +++++++-------
 6 files changed, 106 insertions(+), 58 deletions(-)
 create mode 100644 include/linux/mii_timestamper.h

diff --git a/drivers/net/phy/dp83640.c b/drivers/net/phy/dp83640.c
index b58abdb5491e..ac72a324fcd1 100644
--- a/drivers/net/phy/dp83640.c
+++ b/drivers/net/phy/dp83640.c
@@ -98,6 +98,7 @@ struct dp83640_private {
 	struct list_head list;
 	struct dp83640_clock *clock;
 	struct phy_device *phydev;
+	struct mii_timestamper mii_ts;
 	struct delayed_work ts_work;
 	int hwts_tx_en;
 	int hwts_rx_en;
@@ -1229,9 +1230,10 @@ static int dp83640_config_intr(struct phy_device *phydev)
 	}
 }
 
-static int dp83640_hwtstamp(struct phy_device *phydev, struct ifreq *ifr)
+static int dp83640_hwtstamp(struct mii_timestamper *mii_ts, struct ifreq *ifr)
 {
-	struct dp83640_private *dp83640 = phydev->priv;
+	struct dp83640_private *dp83640 =
+		container_of(mii_ts, struct dp83640_private, mii_ts);
 	struct hwtstamp_config cfg;
 	u16 txcfg0, rxcfg0;
 
@@ -1307,8 +1309,8 @@ static int dp83640_hwtstamp(struct phy_device *phydev, struct ifreq *ifr)
 
 	mutex_lock(&dp83640->clock->extreg_lock);
 
-	ext_write(0, phydev, PAGE5, PTP_TXCFG0, txcfg0);
-	ext_write(0, phydev, PAGE5, PTP_RXCFG0, rxcfg0);
+	ext_write(0, dp83640->phydev, PAGE5, PTP_TXCFG0, txcfg0);
+	ext_write(0, dp83640->phydev, PAGE5, PTP_RXCFG0, rxcfg0);
 
 	mutex_unlock(&dp83640->clock->extreg_lock);
 
@@ -1338,10 +1340,11 @@ static void rx_timestamp_work(struct work_struct *work)
 		schedule_delayed_work(&dp83640->ts_work, SKB_TIMESTAMP_TIMEOUT);
 }
 
-static bool dp83640_rxtstamp(struct phy_device *phydev,
+static bool dp83640_rxtstamp(struct mii_timestamper *mii_ts,
 			     struct sk_buff *skb, int type)
 {
-	struct dp83640_private *dp83640 = phydev->priv;
+	struct dp83640_private *dp83640 =
+		container_of(mii_ts, struct dp83640_private, mii_ts);
 	struct dp83640_skb_info *skb_info = (struct dp83640_skb_info *)skb->cb;
 	struct list_head *this, *next;
 	struct rxts *rxts;
@@ -1387,11 +1390,12 @@ static bool dp83640_rxtstamp(struct phy_device *phydev,
 	return true;
 }
 
-static void dp83640_txtstamp(struct phy_device *phydev,
+static void dp83640_txtstamp(struct mii_timestamper *mii_ts,
 			     struct sk_buff *skb, int type)
 {
 	struct dp83640_skb_info *skb_info = (struct dp83640_skb_info *)skb->cb;
-	struct dp83640_private *dp83640 = phydev->priv;
+	struct dp83640_private *dp83640 =
+		container_of(mii_ts, struct dp83640_private, mii_ts);
 
 	switch (dp83640->hwts_tx_en) {
 
@@ -1414,9 +1418,11 @@ static void dp83640_txtstamp(struct phy_device *phydev,
 	}
 }
 
-static int dp83640_ts_info(struct phy_device *dev, struct ethtool_ts_info *info)
+static int dp83640_ts_info(struct mii_timestamper *mii_ts,
+			   struct ethtool_ts_info *info)
 {
-	struct dp83640_private *dp83640 = dev->priv;
+	struct dp83640_private *dp83640 =
+		container_of(mii_ts, struct dp83640_private, mii_ts);
 
 	info->so_timestamping =
 		SOF_TIMESTAMPING_TX_HARDWARE |
@@ -1454,13 +1460,18 @@ static int dp83640_probe(struct phy_device *phydev)
 		goto no_memory;
 
 	dp83640->phydev = phydev;
-	INIT_DELAYED_WORK(&dp83640->ts_work, rx_timestamp_work);
+	dp83640->mii_ts.rxtstamp = dp83640_rxtstamp;
+	dp83640->mii_ts.txtstamp = dp83640_txtstamp;
+	dp83640->mii_ts.hwtstamp = dp83640_hwtstamp;
+	dp83640->mii_ts.ts_info  = dp83640_ts_info;
 
+	INIT_DELAYED_WORK(&dp83640->ts_work, rx_timestamp_work);
 	INIT_LIST_HEAD(&dp83640->rxts);
 	INIT_LIST_HEAD(&dp83640->rxpool);
 	for (i = 0; i < MAX_RXTS; i++)
 		list_add(&dp83640->rx_pool_data[i].list, &dp83640->rxpool);
 
+	phydev->mii_ts = &dp83640->mii_ts;
 	phydev->priv = dp83640;
 
 	spin_lock_init(&dp83640->rx_lock);
@@ -1501,6 +1512,8 @@ static void dp83640_remove(struct phy_device *phydev)
 	if (phydev->mdio.addr == BROADCAST_ADDR)
 		return;
 
+	phydev->mii_ts = NULL;
+
 	enable_status_frames(phydev, false);
 	cancel_delayed_work_sync(&dp83640->ts_work);
 
@@ -1537,10 +1550,6 @@ static struct phy_driver dp83640_driver = {
 	.config_init	= dp83640_config_init,
 	.ack_interrupt  = dp83640_ack_interrupt,
 	.config_intr    = dp83640_config_intr,
-	.ts_info	= dp83640_ts_info,
-	.hwtstamp	= dp83640_hwtstamp,
-	.rxtstamp	= dp83640_rxtstamp,
-	.txtstamp	= dp83640_txtstamp,
 };
 
 static int __init dp83640_init(void)
diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c
index 80be4d691e5b..541ed01496bf 100644
--- a/drivers/net/phy/phy.c
+++ b/drivers/net/phy/phy.c
@@ -422,8 +422,8 @@ int phy_mii_ioctl(struct phy_device *phydev, struct ifreq *ifr, int cmd)
 		return 0;
 
 	case SIOCSHWTSTAMP:
-		if (phydev->drv && phydev->drv->hwtstamp)
-			return phydev->drv->hwtstamp(phydev, ifr);
+		if (phydev->mii_ts && phydev->mii_ts->hwtstamp)
+			return phydev->mii_ts->hwtstamp(phydev->mii_ts, ifr);
 		/* fall through */
 
 	default:
diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c
index 94961e7592e4..e55119cd6c86 100644
--- a/drivers/net/phy/phy_device.c
+++ b/drivers/net/phy/phy_device.c
@@ -919,6 +919,8 @@ static void phy_link_change(struct phy_device *phydev, bool up, bool do_carrier)
 			netif_carrier_off(netdev);
 	}
 	phydev->adjust_link(netdev);
+	if (phydev->mii_ts && phydev->mii_ts->link_state)
+		phydev->mii_ts->link_state(phydev->mii_ts, phydev);
 }
 
 /**
diff --git a/include/linux/mii_timestamper.h b/include/linux/mii_timestamper.h
new file mode 100644
index 000000000000..36002386029c
--- /dev/null
+++ b/include/linux/mii_timestamper.h
@@ -0,0 +1,58 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Support for generic time stamping devices on MII buses.
+ * Copyright (C) 2018 Richard Cochran <richardcochran@gmail.com>
+ */
+#ifndef _LINUX_MII_TIMESTAMPER_H
+#define _LINUX_MII_TIMESTAMPER_H
+
+#include <linux/device.h>
+#include <linux/ethtool.h>
+#include <linux/skbuff.h>
+
+struct phy_device;
+
+/**
+ * struct mii_timestamper - Callback interface to MII time stamping devices.
+ *
+ * @rxtstamp:	Requests a Rx timestamp for 'skb'.  If the skb is accepted,
+ *		the MII time stamping device promises to deliver it using
+ *		netif_rx() as soon as a timestamp becomes available. One of
+ *		the PTP_CLASS_ values is passed in 'type'.  The function
+ *		must return true if the skb is accepted for delivery.
+ *
+ * @txtstamp:	Requests a Tx timestamp for 'skb'.  The MII time stamping
+ *		device promises to deliver it using skb_complete_tx_timestamp()
+ *		as soon as a timestamp becomes available. One of the PTP_CLASS_
+ *		values is passed in 'type'.
+ *
+ * @hwtstamp:	Handles SIOCSHWTSTAMP ioctl for hardware time stamping.
+ *
+ * @link_state: Allows the device to respond to changes in the link
+ *		state.  The caller invokes this function while holding
+ *		the phy_device mutex.
+ *
+ * @ts_info:	Handles ethtool queries for hardware time stamping.
+ *
+ * Drivers for PHY time stamping devices should embed their
+ * mii_timestamper within a private structure, obtaining a reference
+ * to it using container_of().
+ */
+struct mii_timestamper {
+	bool (*rxtstamp)(struct mii_timestamper *mii_ts,
+			 struct sk_buff *skb, int type);
+
+	void (*txtstamp)(struct mii_timestamper *mii_ts,
+			 struct sk_buff *skb, int type);
+
+	int  (*hwtstamp)(struct mii_timestamper *mii_ts,
+			 struct ifreq *ifreq);
+
+	void (*link_state)(struct mii_timestamper *mii_ts,
+			   struct phy_device *phydev);
+
+	int  (*ts_info)(struct mii_timestamper *mii_ts,
+			struct ethtool_ts_info *ts_info);
+};
+
+#endif
diff --git a/include/linux/phy.h b/include/linux/phy.h
index 0248f5e9939d..30e599c454db 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -17,6 +17,7 @@
 #include <linux/linkmode.h>
 #include <linux/mdio.h>
 #include <linux/mii.h>
+#include <linux/mii_timestamper.h>
 #include <linux/module.h>
 #include <linux/timer.h>
 #include <linux/workqueue.h>
@@ -441,6 +442,7 @@ struct phy_device {
 	struct sfp_bus *sfp_bus;
 	struct phylink *phylink;
 	struct net_device *attached_dev;
+	struct mii_timestamper *mii_ts;
 
 	u8 mdix;
 	u8 mdix_ctrl;
@@ -546,29 +548,6 @@ struct phy_driver {
 	 */
 	int (*match_phy_device)(struct phy_device *phydev);
 
-	/* Handles ethtool queries for hardware time stamping. */
-	int (*ts_info)(struct phy_device *phydev, struct ethtool_ts_info *ti);
-
-	/* Handles SIOCSHWTSTAMP ioctl for hardware time stamping. */
-	int  (*hwtstamp)(struct phy_device *phydev, struct ifreq *ifr);
-
-	/*
-	 * Requests a Rx timestamp for 'skb'. If the skb is accepted,
-	 * the phy driver promises to deliver it using netif_rx() as
-	 * soon as a timestamp becomes available. One of the
-	 * PTP_CLASS_ values is passed in 'type'. The function must
-	 * return true if the skb is accepted for delivery.
-	 */
-	bool (*rxtstamp)(struct phy_device *dev, struct sk_buff *skb, int type);
-
-	/*
-	 * Requests a Tx timestamp for 'skb'. The phy driver promises
-	 * to deliver it using skb_complete_tx_timestamp() as soon as a
-	 * timestamp becomes available. One of the PTP_CLASS_ values
-	 * is passed in 'type'.
-	 */
-	void (*txtstamp)(struct phy_device *dev, struct sk_buff *skb, int type);
-
 	/* Some devices (e.g. qnap TS-119P II) require PHY register changes to
 	 * enable Wake on LAN, so set_wol is provided to be called in the
 	 * ethernet driver's set_wol function. */
@@ -942,7 +921,7 @@ static inline bool phy_polling_mode(struct phy_device *phydev)
  */
 static inline bool phy_has_hwtstamp(struct phy_device *phydev)
 {
-	return phydev && phydev->drv && phydev->drv->hwtstamp;
+	return phydev && phydev->mii_ts && phydev->mii_ts->hwtstamp;
 }
 
 /**
@@ -951,7 +930,7 @@ static inline bool phy_has_hwtstamp(struct phy_device *phydev)
  */
 static inline bool phy_has_rxtstamp(struct phy_device *phydev)
 {
-	return phydev && phydev->drv && phydev->drv->rxtstamp;
+	return phydev && phydev->mii_ts && phydev->mii_ts->rxtstamp;
 }
 
 /**
@@ -961,7 +940,7 @@ static inline bool phy_has_rxtstamp(struct phy_device *phydev)
  */
 static inline bool phy_has_tsinfo(struct phy_device *phydev)
 {
-	return phydev && phydev->drv && phydev->drv->ts_info;
+	return phydev && phydev->mii_ts && phydev->mii_ts->ts_info;
 }
 
 /**
@@ -970,30 +949,30 @@ static inline bool phy_has_tsinfo(struct phy_device *phydev)
  */
 static inline bool phy_has_txtstamp(struct phy_device *phydev)
 {
-	return phydev && phydev->drv && phydev->drv->txtstamp;
+	return phydev && phydev->mii_ts && phydev->mii_ts->txtstamp;
 }
 
 static inline int phy_hwtstamp(struct phy_device *phydev, struct ifreq *ifr)
 {
-	return phydev->drv->hwtstamp(phydev, ifr);
+	return phydev->mii_ts->hwtstamp(phydev->mii_ts, ifr);
 }
 
 static inline bool phy_rxtstamp(struct phy_device *phydev, struct sk_buff *skb,
 				int type)
 {
-	return phydev->drv->rxtstamp(phydev, skb, type);
+	return phydev->mii_ts->rxtstamp(phydev->mii_ts, skb, type);
 }
 
 static inline int phy_ts_info(struct phy_device *phydev,
 			      struct ethtool_ts_info *tsinfo)
 {
-	return phydev->drv->ts_info(phydev, tsinfo);
+	return phydev->mii_ts->ts_info(phydev->mii_ts, tsinfo);
 }
 
 static inline void phy_txtstamp(struct phy_device *phydev, struct sk_buff *skb,
 				int type)
 {
-	phydev->drv->txtstamp(phydev, skb, type);
+	phydev->mii_ts->txtstamp(phydev->mii_ts, skb, type);
 }
 
 /**
diff --git a/net/core/timestamping.c b/net/core/timestamping.c
index 7911235706a9..04840697fe79 100644
--- a/net/core/timestamping.c
+++ b/net/core/timestamping.c
@@ -13,7 +13,7 @@
 static unsigned int classify(const struct sk_buff *skb)
 {
 	if (likely(skb->dev && skb->dev->phydev &&
-		   skb->dev->phydev->drv))
+		   skb->dev->phydev->mii_ts))
 		return ptp_classify_raw(skb);
 	else
 		return PTP_CLASS_NONE;
@@ -21,7 +21,7 @@ static unsigned int classify(const struct sk_buff *skb)
 
 void skb_clone_tx_timestamp(struct sk_buff *skb)
 {
-	struct phy_device *phydev;
+	struct mii_timestamper *mii_ts;
 	struct sk_buff *clone;
 	unsigned int type;
 
@@ -32,22 +32,22 @@ void skb_clone_tx_timestamp(struct sk_buff *skb)
 	if (type == PTP_CLASS_NONE)
 		return;
 
-	phydev = skb->dev->phydev;
-	if (likely(phydev->drv->txtstamp)) {
+	mii_ts = skb->dev->phydev->mii_ts;
+	if (likely(mii_ts->txtstamp)) {
 		clone = skb_clone_sk(skb);
 		if (!clone)
 			return;
-		phydev->drv->txtstamp(phydev, clone, type);
+		mii_ts->txtstamp(mii_ts, clone, type);
 	}
 }
 EXPORT_SYMBOL_GPL(skb_clone_tx_timestamp);
 
 bool skb_defer_rx_timestamp(struct sk_buff *skb)
 {
-	struct phy_device *phydev;
+	struct mii_timestamper *mii_ts;
 	unsigned int type;
 
-	if (!skb->dev || !skb->dev->phydev || !skb->dev->phydev->drv)
+	if (!skb->dev || !skb->dev->phydev || !skb->dev->phydev->mii_ts)
 		return false;
 
 	if (skb_headroom(skb) < ETH_HLEN)
@@ -62,9 +62,9 @@ bool skb_defer_rx_timestamp(struct sk_buff *skb)
 	if (type == PTP_CLASS_NONE)
 		return false;
 
-	phydev = skb->dev->phydev;
-	if (likely(phydev->drv->rxtstamp))
-		return phydev->drv->rxtstamp(phydev, skb, type);
+	mii_ts = skb->dev->phydev->mii_ts;
+	if (likely(mii_ts->rxtstamp))
+		return mii_ts->rxtstamp(mii_ts, skb, type);
 
 	return false;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 767ff483731502a0fc34f34a3a0851aca175eb71 Mon Sep 17 00:00:00 2001
From: Richard Cochran <richardcochran@gmail.com>
Date: Wed, 25 Dec 2019 18:16:16 -0800
Subject: net: Add a layer for non-PHY MII time stamping drivers.

While PHY time stamping drivers can simply attach their interface
directly to the PHY instance, stand alone drivers require support in
order to manage their services.  Non-PHY MII time stamping drivers
have a control interface over another bus like I2C, SPI, UART, or via
a memory mapped peripheral.  The controller device will be associated
with one or more time stamping channels, each of which sits snoops in
on a MII bus.

This patch provides a glue layer that will enable time stamping
channels to find their controlling device.

Signed-off-by: Richard Cochran <richardcochran@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/Makefile          |   2 +
 drivers/net/phy/mii_timestamper.c | 125 ++++++++++++++++++++++++++++++++++++++
 include/linux/mii_timestamper.h   |  63 +++++++++++++++++++
 net/Kconfig                       |   7 ++-
 4 files changed, 194 insertions(+), 3 deletions(-)
 create mode 100644 drivers/net/phy/mii_timestamper.c

diff --git a/drivers/net/phy/Makefile b/drivers/net/phy/Makefile
index d846b4dc1c68..fe5badf13b65 100644
--- a/drivers/net/phy/Makefile
+++ b/drivers/net/phy/Makefile
@@ -43,6 +43,8 @@ obj-$(CONFIG_MDIO_SUN4I)	+= mdio-sun4i.o
 obj-$(CONFIG_MDIO_THUNDER)	+= mdio-thunder.o
 obj-$(CONFIG_MDIO_XGENE)	+= mdio-xgene.o
 
+obj-$(CONFIG_NETWORK_PHY_TIMESTAMPING) += mii_timestamper.o
+
 obj-$(CONFIG_SFP)		+= sfp.o
 sfp-obj-$(CONFIG_SFP)		+= sfp-bus.o
 obj-y				+= $(sfp-obj-y) $(sfp-obj-m)
diff --git a/drivers/net/phy/mii_timestamper.c b/drivers/net/phy/mii_timestamper.c
new file mode 100644
index 000000000000..2f12c5d901df
--- /dev/null
+++ b/drivers/net/phy/mii_timestamper.c
@@ -0,0 +1,125 @@
+// SPDX-License-Identifier: GPL-2.0
+//
+// Support for generic time stamping devices on MII buses.
+// Copyright (C) 2018 Richard Cochran <richardcochran@gmail.com>
+//
+
+#include <linux/mii_timestamper.h>
+
+static LIST_HEAD(mii_timestamping_devices);
+static DEFINE_MUTEX(tstamping_devices_lock);
+
+struct mii_timestamping_desc {
+	struct list_head list;
+	struct mii_timestamping_ctrl *ctrl;
+	struct device *device;
+};
+
+/**
+ * register_mii_tstamp_controller() - registers an MII time stamping device.
+ *
+ * @device:	The device to be registered.
+ * @ctrl:	Pointer to device's control interface.
+ *
+ * Returns zero on success or non-zero on failure.
+ */
+int register_mii_tstamp_controller(struct device *device,
+				   struct mii_timestamping_ctrl *ctrl)
+{
+	struct mii_timestamping_desc *desc;
+
+	desc = kzalloc(sizeof(*desc), GFP_KERNEL);
+	if (!desc)
+		return -ENOMEM;
+
+	INIT_LIST_HEAD(&desc->list);
+	desc->ctrl = ctrl;
+	desc->device = device;
+
+	mutex_lock(&tstamping_devices_lock);
+	list_add_tail(&mii_timestamping_devices, &desc->list);
+	mutex_unlock(&tstamping_devices_lock);
+
+	return 0;
+}
+EXPORT_SYMBOL(register_mii_tstamp_controller);
+
+/**
+ * unregister_mii_tstamp_controller() - unregisters an MII time stamping device.
+ *
+ * @device:	A device previously passed to register_mii_tstamp_controller().
+ */
+void unregister_mii_tstamp_controller(struct device *device)
+{
+	struct mii_timestamping_desc *desc;
+	struct list_head *this, *next;
+
+	mutex_lock(&tstamping_devices_lock);
+	list_for_each_safe(this, next, &mii_timestamping_devices) {
+		desc = list_entry(this, struct mii_timestamping_desc, list);
+		if (desc->device == device) {
+			list_del_init(&desc->list);
+			kfree(desc);
+			break;
+		}
+	}
+	mutex_unlock(&tstamping_devices_lock);
+}
+EXPORT_SYMBOL(unregister_mii_tstamp_controller);
+
+/**
+ * register_mii_timestamper - Enables a given port of an MII time stamper.
+ *
+ * @node:	The device tree node of the MII time stamp controller.
+ * @port:	The index of the port to be enabled.
+ *
+ * Returns a valid interface on success or ERR_PTR otherwise.
+ */
+struct mii_timestamper *register_mii_timestamper(struct device_node *node,
+						 unsigned int port)
+{
+	struct mii_timestamper *mii_ts = NULL;
+	struct mii_timestamping_desc *desc;
+	struct list_head *this;
+
+	mutex_lock(&tstamping_devices_lock);
+	list_for_each(this, &mii_timestamping_devices) {
+		desc = list_entry(this, struct mii_timestamping_desc, list);
+		if (desc->device->of_node == node) {
+			mii_ts = desc->ctrl->probe_channel(desc->device, port);
+			if (!IS_ERR(mii_ts)) {
+				mii_ts->device = desc->device;
+				get_device(desc->device);
+			}
+			break;
+		}
+	}
+	mutex_unlock(&tstamping_devices_lock);
+
+	return mii_ts ? mii_ts : ERR_PTR(-EPROBE_DEFER);
+}
+EXPORT_SYMBOL(register_mii_timestamper);
+
+/**
+ * unregister_mii_timestamper - Disables a given MII time stamper.
+ *
+ * @mii_ts:	An interface obtained via register_mii_timestamper().
+ *
+ */
+void unregister_mii_timestamper(struct mii_timestamper *mii_ts)
+{
+	struct mii_timestamping_desc *desc;
+	struct list_head *this;
+
+	mutex_lock(&tstamping_devices_lock);
+	list_for_each(this, &mii_timestamping_devices) {
+		desc = list_entry(this, struct mii_timestamping_desc, list);
+		if (desc->device == mii_ts->device) {
+			desc->ctrl->release_channel(desc->device, mii_ts);
+			put_device(desc->device);
+			break;
+		}
+	}
+	mutex_unlock(&tstamping_devices_lock);
+}
+EXPORT_SYMBOL(unregister_mii_timestamper);
diff --git a/include/linux/mii_timestamper.h b/include/linux/mii_timestamper.h
index 36002386029c..fa940bbaf8ae 100644
--- a/include/linux/mii_timestamper.h
+++ b/include/linux/mii_timestamper.h
@@ -33,10 +33,15 @@ struct phy_device;
  *		the phy_device mutex.
  *
  * @ts_info:	Handles ethtool queries for hardware time stamping.
+ * @device:	Remembers the device to which the instance belongs.
  *
  * Drivers for PHY time stamping devices should embed their
  * mii_timestamper within a private structure, obtaining a reference
  * to it using container_of().
+ *
+ * Drivers for non-PHY time stamping devices should return a pointer
+ * to a mii_timestamper from the probe_channel() callback of their
+ * mii_timestamping_ctrl interface.
  */
 struct mii_timestamper {
 	bool (*rxtstamp)(struct mii_timestamper *mii_ts,
@@ -53,6 +58,64 @@ struct mii_timestamper {
 
 	int  (*ts_info)(struct mii_timestamper *mii_ts,
 			struct ethtool_ts_info *ts_info);
+
+	struct device *device;
+};
+
+/**
+ * struct mii_timestamping_ctrl - MII time stamping controller interface.
+ *
+ * @probe_channel:	Callback into the controller driver announcing the
+ *			presence of the 'port' channel.  The 'device' field
+ *			had been passed to register_mii_tstamp_controller().
+ *			The driver must return either a pointer to a valid
+ *			MII timestamper instance or PTR_ERR.
+ *
+ * @release_channel:	Releases an instance obtained via .probe_channel.
+ */
+struct mii_timestamping_ctrl {
+	struct mii_timestamper *(*probe_channel)(struct device *device,
+						 unsigned int port);
+	void (*release_channel)(struct device *device,
+				struct mii_timestamper *mii_ts);
 };
 
+#ifdef CONFIG_NETWORK_PHY_TIMESTAMPING
+
+int register_mii_tstamp_controller(struct device *device,
+				   struct mii_timestamping_ctrl *ctrl);
+
+void unregister_mii_tstamp_controller(struct device *device);
+
+struct mii_timestamper *register_mii_timestamper(struct device_node *node,
+						 unsigned int port);
+
+void unregister_mii_timestamper(struct mii_timestamper *mii_ts);
+
+#else
+
+static inline
+int register_mii_tstamp_controller(struct device *device,
+				   struct mii_timestamping_ctrl *ctrl)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline void unregister_mii_tstamp_controller(struct device *device)
+{
+}
+
+static inline
+struct mii_timestamper *register_mii_timestamper(struct device_node *node,
+						 unsigned int port)
+{
+	return NULL;
+}
+
+static inline void unregister_mii_timestamper(struct mii_timestamper *mii_ts)
+{
+}
+
+#endif
+
 #endif
diff --git a/net/Kconfig b/net/Kconfig
index bd191f978a23..52af65e5d28c 100644
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -108,9 +108,10 @@ config NETWORK_PHY_TIMESTAMPING
 	bool "Timestamping in PHY devices"
 	select NET_PTP_CLASSIFY
 	help
-	  This allows timestamping of network packets by PHYs with
-	  hardware timestamping capabilities. This option adds some
-	  overhead in the transmit and receive paths.
+	  This allows timestamping of network packets by PHYs (or
+	  other MII bus snooping devices) with hardware timestamping
+	  capabilities. This option adds some overhead in the transmit
+	  and receive paths.
 
 	  If you are unsure how to answer this question, answer N.
 
-- 
cgit v1.2.3-59-g8ed1b


From 25d12e1dde28fafd2ac37afadd24252fa19b80cd Mon Sep 17 00:00:00 2001
From: Richard Cochran <richardcochran@gmail.com>
Date: Wed, 25 Dec 2019 18:16:17 -0800
Subject: dt-bindings: ptp: Introduce MII time stamping devices.

This patch add a new binding that allows non-PHY MII time stamping
devices to find their buses.  The new documentation covers both the
generic binding and one upcoming user.

Signed-off-by: Richard Cochran <richardcochran@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/devicetree/bindings/ptp/ptp-ines.txt | 35 ++++++++++++++++++
 .../devicetree/bindings/ptp/timestamper.txt        | 42 ++++++++++++++++++++++
 2 files changed, 77 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/ptp/ptp-ines.txt
 create mode 100644 Documentation/devicetree/bindings/ptp/timestamper.txt

diff --git a/Documentation/devicetree/bindings/ptp/ptp-ines.txt b/Documentation/devicetree/bindings/ptp/ptp-ines.txt
new file mode 100644
index 000000000000..4c242bd1ce9c
--- /dev/null
+++ b/Documentation/devicetree/bindings/ptp/ptp-ines.txt
@@ -0,0 +1,35 @@
+ZHAW InES PTP time stamping IP core
+
+The IP core needs two different kinds of nodes.  The control node
+lives somewhere in the memory map and specifies the address of the
+control registers.  There can be up to three port handles placed as
+attributes of PHY nodes.  These associate a particular MII bus with a
+port index within the IP core.
+
+Required properties of the control node:
+
+- compatible:		"ines,ptp-ctrl"
+- reg:			physical address and size of the register bank
+
+Required format of the port handle within the PHY node:
+
+- timestamper:		provides control node reference and
+			the port channel within the IP core
+
+Example:
+
+	tstamper: timestamper@60000000 {
+		compatible = "ines,ptp-ctrl";
+		reg = <0x60000000 0x80>;
+	};
+
+	ethernet@80000000 {
+		...
+		mdio {
+			...
+			ethernet-phy@3 {
+				...
+				timestamper = <&tstamper 0>;
+			};
+		};
+	};
diff --git a/Documentation/devicetree/bindings/ptp/timestamper.txt b/Documentation/devicetree/bindings/ptp/timestamper.txt
new file mode 100644
index 000000000000..fc550ce4d4ea
--- /dev/null
+++ b/Documentation/devicetree/bindings/ptp/timestamper.txt
@@ -0,0 +1,42 @@
+Time stamps from MII bus snooping devices
+
+This binding supports non-PHY devices that snoop the MII bus and
+provide time stamps.  In contrast to PHY time stamping drivers (which
+can simply attach their interface directly to the PHY instance), stand
+alone MII time stamping drivers use this binding to specify the
+connection between the snooping device and a given network interface.
+
+Non-PHY MII time stamping drivers typically talk to the control
+interface over another bus like I2C, SPI, UART, or via a memory mapped
+peripheral.  This controller device is associated with one or more
+time stamping channels, each of which snoops on a MII bus.
+
+The "timestamper" property lives in a phy node and links a time
+stamping channel from the controller device to that phy's MII bus.
+
+Example:
+
+	tstamper: timestamper@10000000 {
+		compatible = "ines,ptp-ctrl";
+		reg = <0x10000000 0x80>;
+	};
+
+	ethernet@20000000 {
+		mdio {
+			ethernet-phy@1 {
+				timestamper = <&tstamper 0>;
+			};
+		};
+	};
+
+	ethernet@30000000 {
+		mdio {
+			ethernet-phy@2 {
+				timestamper = <&tstamper 1>;
+			};
+		};
+	};
+
+In this example, time stamps from the MII bus attached to phy@1 will
+appear on time stamp channel 0 (zero), and those from phy@2 appear on
+channel 1.
-- 
cgit v1.2.3-59-g8ed1b


From 1dca22b1842136f57e9bd60fefd1715c22ff1874 Mon Sep 17 00:00:00 2001
From: Richard Cochran <richardcochran@gmail.com>
Date: Wed, 25 Dec 2019 18:16:18 -0800
Subject: net: mdio: of: Register discovered MII time stampers.

When parsing a PHY node, register its time stamper, if any, and attach
the instance to the PHY device.

Signed-off-by: Richard Cochran <richardcochran@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/phy_device.c |  3 +++
 drivers/of/of_mdio.c         | 30 +++++++++++++++++++++++++++++-
 2 files changed, 32 insertions(+), 1 deletion(-)

diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c
index e55119cd6c86..e5dc9f87f495 100644
--- a/drivers/net/phy/phy_device.c
+++ b/drivers/net/phy/phy_device.c
@@ -881,6 +881,9 @@ EXPORT_SYMBOL(phy_device_register);
  */
 void phy_device_remove(struct phy_device *phydev)
 {
+	if (phydev->mii_ts)
+		unregister_mii_timestamper(phydev->mii_ts);
+
 	device_del(&phydev->mdio.dev);
 
 	/* Assert the reset signal */
diff --git a/drivers/of/of_mdio.c b/drivers/of/of_mdio.c
index fc757ef6eadc..f5c2a5487761 100644
--- a/drivers/of/of_mdio.c
+++ b/drivers/of/of_mdio.c
@@ -42,14 +42,37 @@ static int of_get_phy_id(struct device_node *device, u32 *phy_id)
 	return -EINVAL;
 }
 
+static struct mii_timestamper *of_find_mii_timestamper(struct device_node *node)
+{
+	struct of_phandle_args arg;
+	int err;
+
+	err = of_parse_phandle_with_fixed_args(node, "timestamper", 1, 0, &arg);
+
+	if (err == -ENOENT)
+		return NULL;
+	else if (err)
+		return ERR_PTR(err);
+
+	if (arg.args_count != 1)
+		return ERR_PTR(-EINVAL);
+
+	return register_mii_timestamper(arg.np, arg.args[0]);
+}
+
 static int of_mdiobus_register_phy(struct mii_bus *mdio,
 				    struct device_node *child, u32 addr)
 {
+	struct mii_timestamper *mii_ts;
 	struct phy_device *phy;
 	bool is_c45;
 	int rc;
 	u32 phy_id;
 
+	mii_ts = of_find_mii_timestamper(child);
+	if (IS_ERR(mii_ts))
+		return PTR_ERR(mii_ts);
+
 	is_c45 = of_device_is_compatible(child,
 					 "ethernet-phy-ieee802.3-c45");
 
@@ -57,11 +80,14 @@ static int of_mdiobus_register_phy(struct mii_bus *mdio,
 		phy = phy_device_create(mdio, addr, phy_id, 0, NULL);
 	else
 		phy = get_phy_device(mdio, addr, is_c45);
-	if (IS_ERR(phy))
+	if (IS_ERR(phy)) {
+		unregister_mii_timestamper(mii_ts);
 		return PTR_ERR(phy);
+	}
 
 	rc = of_irq_get(child, 0);
 	if (rc == -EPROBE_DEFER) {
+		unregister_mii_timestamper(mii_ts);
 		phy_device_free(phy);
 		return rc;
 	}
@@ -90,10 +116,12 @@ static int of_mdiobus_register_phy(struct mii_bus *mdio,
 	 * register it */
 	rc = phy_device_register(phy);
 	if (rc) {
+		unregister_mii_timestamper(mii_ts);
 		phy_device_free(phy);
 		of_node_put(child);
 		return rc;
 	}
+	phy->mii_ts = mii_ts;
 
 	dev_dbg(&mdio->dev, "registered phy %pOFn at address %i\n",
 		child, addr);
-- 
cgit v1.2.3-59-g8ed1b


From b6fd7b96366769651ab23988607ce9c5c9042cdb Mon Sep 17 00:00:00 2001
From: Richard Cochran <richardcochran@gmail.com>
Date: Wed, 25 Dec 2019 18:16:19 -0800
Subject: net: Introduce peer to peer one step PTP time stamping.

The 1588 standard defines one step operation for both Sync and
PDelay_Resp messages.  Up until now, hardware with P2P one step has
been rare, and kernel support was lacking.  This patch adds support of
the mode in anticipation of new hardware developments.

Signed-off-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c   | 1 +
 drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c | 1 +
 drivers/net/ethernet/microchip/lan743x_ptp.c       | 3 +++
 drivers/net/ethernet/qlogic/qede/qede_ptp.c        | 1 +
 include/uapi/linux/net_tstamp.h                    | 8 ++++++++
 net/core/dev_ioctl.c                               | 1 +
 6 files changed, 15 insertions(+)

diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index cff64e43bdd8..741d865e4afc 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -15410,6 +15410,7 @@ int bnx2x_configure_ptp_filters(struct bnx2x *bp)
 		REG_WR(bp, rule, BNX2X_PTP_TX_ON_RULE_MASK);
 		break;
 	case HWTSTAMP_TX_ONESTEP_SYNC:
+	case HWTSTAMP_TX_ONESTEP_P2P:
 		BNX2X_ERR("One-step timestamping is not supported\n");
 		return -ERANGE;
 	}
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c
index ec2ff3d7f41c..4aaaa4937b1a 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c
@@ -920,6 +920,7 @@ static int mlxsw_sp_ptp_get_message_types(const struct hwtstamp_config *config,
 		egr_types = 0xff;
 		break;
 	case HWTSTAMP_TX_ONESTEP_SYNC:
+	case HWTSTAMP_TX_ONESTEP_P2P:
 		return -ERANGE;
 	}
 
diff --git a/drivers/net/ethernet/microchip/lan743x_ptp.c b/drivers/net/ethernet/microchip/lan743x_ptp.c
index afe52463dc57..9399f6a98748 100644
--- a/drivers/net/ethernet/microchip/lan743x_ptp.c
+++ b/drivers/net/ethernet/microchip/lan743x_ptp.c
@@ -1265,6 +1265,9 @@ int lan743x_ptp_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
 
 		lan743x_ptp_set_sync_ts_insert(adapter, true);
 		break;
+	case HWTSTAMP_TX_ONESTEP_P2P:
+		ret = -ERANGE;
+		break;
 	default:
 		netif_warn(adapter, drv, adapter->netdev,
 			   "  tx_type = %d, UNKNOWN\n", config.tx_type);
diff --git a/drivers/net/ethernet/qlogic/qede/qede_ptp.c b/drivers/net/ethernet/qlogic/qede/qede_ptp.c
index f815435cf106..4c7f7a7fc151 100644
--- a/drivers/net/ethernet/qlogic/qede/qede_ptp.c
+++ b/drivers/net/ethernet/qlogic/qede/qede_ptp.c
@@ -247,6 +247,7 @@ static int qede_ptp_cfg_filters(struct qede_dev *edev)
 		break;
 
 	case HWTSTAMP_TX_ONESTEP_SYNC:
+	case HWTSTAMP_TX_ONESTEP_P2P:
 		DP_ERR(edev, "One-step timestamping is not supported\n");
 		return -ERANGE;
 	}
diff --git a/include/uapi/linux/net_tstamp.h b/include/uapi/linux/net_tstamp.h
index e5b39721c6e4..f96e650d0af9 100644
--- a/include/uapi/linux/net_tstamp.h
+++ b/include/uapi/linux/net_tstamp.h
@@ -90,6 +90,14 @@ enum hwtstamp_tx_types {
 	 * queue.
 	 */
 	HWTSTAMP_TX_ONESTEP_SYNC,
+
+	/*
+	 * Same as HWTSTAMP_TX_ONESTEP_SYNC, but also enables time
+	 * stamp insertion directly into PDelay_Resp packets. In this
+	 * case, neither transmitted Sync nor PDelay_Resp packets will
+	 * receive a time stamp via the socket error queue.
+	 */
+	HWTSTAMP_TX_ONESTEP_P2P,
 };
 
 /* possible values for hwtstamp_config->rx_filter */
diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c
index 5163d900bb4f..dbaebbe573f0 100644
--- a/net/core/dev_ioctl.c
+++ b/net/core/dev_ioctl.c
@@ -187,6 +187,7 @@ static int net_hwtstamp_validate(struct ifreq *ifr)
 	case HWTSTAMP_TX_OFF:
 	case HWTSTAMP_TX_ON:
 	case HWTSTAMP_TX_ONESTEP_SYNC:
+	case HWTSTAMP_TX_ONESTEP_P2P:
 		tx_type_valid = 1;
 		break;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From bad1eaa6ac312ffd7aa46dd5a4d9843b824aa023 Mon Sep 17 00:00:00 2001
From: Richard Cochran <richardcochran@gmail.com>
Date: Wed, 25 Dec 2019 18:16:20 -0800
Subject: ptp: Add a driver for InES time stamping IP core.

The InES at the ZHAW offers a PTP time stamping IP core.  The FPGA
logic recognizes and time stamps PTP frames on the MII bus.  This
patch adds a driver for the core along with a device tree binding to
allow hooking the driver to MII buses.

Signed-off-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/ptp/Kconfig    |  10 +
 drivers/ptp/Makefile   |   1 +
 drivers/ptp/ptp_ines.c | 852 +++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 863 insertions(+)
 create mode 100644 drivers/ptp/ptp_ines.c

diff --git a/drivers/ptp/Kconfig b/drivers/ptp/Kconfig
index b0d1b8d264fa..c382158f587d 100644
--- a/drivers/ptp/Kconfig
+++ b/drivers/ptp/Kconfig
@@ -89,6 +89,16 @@ config DP83640_PHY
 	  In order for this to work, your MAC driver must also
 	  implement the skb_tx_timestamp() function.
 
+config PTP_1588_CLOCK_INES
+	tristate "ZHAW InES PTP time stamping IP core"
+	depends on NETWORK_PHY_TIMESTAMPING
+	depends on PHYLIB
+	depends on PTP_1588_CLOCK
+	help
+	  This driver adds support for using the ZHAW InES 1588 IP
+	  core.  This clock is only useful if the MII bus of your MAC
+	  is wired up to the core.
+
 config PTP_1588_CLOCK_PCH
 	tristate "Intel PCH EG20T as PTP clock"
 	depends on X86_32 || COMPILE_TEST
diff --git a/drivers/ptp/Makefile b/drivers/ptp/Makefile
index 69a06f86a450..3fb91bebbaf7 100644
--- a/drivers/ptp/Makefile
+++ b/drivers/ptp/Makefile
@@ -6,6 +6,7 @@
 ptp-y					:= ptp_clock.o ptp_chardev.o ptp_sysfs.o
 obj-$(CONFIG_PTP_1588_CLOCK)		+= ptp.o
 obj-$(CONFIG_PTP_1588_CLOCK_DTE)	+= ptp_dte.o
+obj-$(CONFIG_PTP_1588_CLOCK_INES)	+= ptp_ines.o
 obj-$(CONFIG_PTP_1588_CLOCK_IXP46X)	+= ptp_ixp46x.o
 obj-$(CONFIG_PTP_1588_CLOCK_PCH)	+= ptp_pch.o
 obj-$(CONFIG_PTP_1588_CLOCK_KVM)	+= ptp_kvm.o
diff --git a/drivers/ptp/ptp_ines.c b/drivers/ptp/ptp_ines.c
new file mode 100644
index 000000000000..dfda54cbd866
--- /dev/null
+++ b/drivers/ptp/ptp_ines.c
@@ -0,0 +1,852 @@
+// SPDX-License-Identifier: GPL-2.0
+//
+// Copyright (C) 2018 MOSER-BAER AG
+//
+
+#define pr_fmt(fmt) "InES_PTP: " fmt
+
+#include <linux/ethtool.h>
+#include <linux/export.h>
+#include <linux/if_vlan.h>
+#include <linux/mii_timestamper.h>
+#include <linux/module.h>
+#include <linux/net_tstamp.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/of_irq.h>
+#include <linux/phy.h>
+#include <linux/platform_device.h>
+#include <linux/ptp_classify.h>
+#include <linux/ptp_clock_kernel.h>
+#include <linux/stddef.h>
+
+MODULE_DESCRIPTION("Driver for the ZHAW InES PTP time stamping IP core");
+MODULE_AUTHOR("Richard Cochran <richardcochran@gmail.com>");
+MODULE_VERSION("1.0");
+MODULE_LICENSE("GPL");
+
+/* GLOBAL register */
+#define MCAST_MAC_SELECT_SHIFT	2
+#define MCAST_MAC_SELECT_MASK	0x3
+#define IO_RESET		BIT(1)
+#define PTP_RESET		BIT(0)
+
+/* VERSION register */
+#define IF_MAJOR_VER_SHIFT	12
+#define IF_MAJOR_VER_MASK	0xf
+#define IF_MINOR_VER_SHIFT	8
+#define IF_MINOR_VER_MASK	0xf
+#define FPGA_MAJOR_VER_SHIFT	4
+#define FPGA_MAJOR_VER_MASK	0xf
+#define FPGA_MINOR_VER_SHIFT	0
+#define FPGA_MINOR_VER_MASK	0xf
+
+/* INT_STAT register */
+#define RX_INTR_STATUS_3	BIT(5)
+#define RX_INTR_STATUS_2	BIT(4)
+#define RX_INTR_STATUS_1	BIT(3)
+#define TX_INTR_STATUS_3	BIT(2)
+#define TX_INTR_STATUS_2	BIT(1)
+#define TX_INTR_STATUS_1	BIT(0)
+
+/* INT_MSK register */
+#define RX_INTR_MASK_3		BIT(5)
+#define RX_INTR_MASK_2		BIT(4)
+#define RX_INTR_MASK_1		BIT(3)
+#define TX_INTR_MASK_3		BIT(2)
+#define TX_INTR_MASK_2		BIT(1)
+#define TX_INTR_MASK_1		BIT(0)
+
+/* BUF_STAT register */
+#define RX_FIFO_NE_3		BIT(5)
+#define RX_FIFO_NE_2		BIT(4)
+#define RX_FIFO_NE_1		BIT(3)
+#define TX_FIFO_NE_3		BIT(2)
+#define TX_FIFO_NE_2		BIT(1)
+#define TX_FIFO_NE_1		BIT(0)
+
+/* PORT_CONF register */
+#define CM_ONE_STEP		BIT(6)
+#define PHY_SPEED_SHIFT		4
+#define PHY_SPEED_MASK		0x3
+#define P2P_DELAY_WR_POS_SHIFT	2
+#define P2P_DELAY_WR_POS_MASK	0x3
+#define PTP_MODE_SHIFT		0
+#define PTP_MODE_MASK		0x3
+
+/* TS_STAT_TX register */
+#define TS_ENABLE		BIT(15)
+#define DATA_READ_POS_SHIFT	8
+#define DATA_READ_POS_MASK	0x1f
+#define DISCARDED_EVENTS_SHIFT	4
+#define DISCARDED_EVENTS_MASK	0xf
+
+#define INES_N_PORTS		3
+#define INES_REGISTER_SIZE	0x80
+#define INES_PORT_OFFSET	0x20
+#define INES_PORT_SIZE		0x20
+#define INES_FIFO_DEPTH		90
+#define INES_MAX_EVENTS		100
+
+#define BC_PTP_V1		0
+#define BC_PTP_V2		1
+#define TC_E2E_PTP_V2		2
+#define TC_P2P_PTP_V2		3
+
+#define OFF_PTP_CLOCK_ID	20
+#define OFF_PTP_PORT_NUM	28
+
+#define PHY_SPEED_10		0
+#define PHY_SPEED_100		1
+#define PHY_SPEED_1000		2
+
+#define PORT_CONF \
+	((PHY_SPEED_1000 << PHY_SPEED_SHIFT) | (BC_PTP_V2 << PTP_MODE_SHIFT))
+
+#define ines_read32(s, r)	__raw_readl((void __iomem *)&s->regs->r)
+#define ines_write32(s, v, r)	__raw_writel(v, (void __iomem *)&s->regs->r)
+
+#define MESSAGE_TYPE_SYNC		1
+#define MESSAGE_TYPE_P_DELAY_REQ	2
+#define MESSAGE_TYPE_P_DELAY_RESP	3
+#define MESSAGE_TYPE_DELAY_REQ		4
+
+#define SYNC				0x0
+#define DELAY_REQ			0x1
+#define PDELAY_REQ			0x2
+#define PDELAY_RESP			0x3
+
+static LIST_HEAD(ines_clocks);
+static DEFINE_MUTEX(ines_clocks_lock);
+
+struct ines_global_regs {
+	u32 id;
+	u32 test;
+	u32 global;
+	u32 version;
+	u32 test2;
+	u32 int_stat;
+	u32 int_msk;
+	u32 buf_stat;
+};
+
+struct ines_port_registers {
+	u32 port_conf;
+	u32 p_delay;
+	u32 ts_stat_tx;
+	u32 ts_stat_rx;
+	u32 ts_tx;
+	u32 ts_rx;
+};
+
+struct ines_timestamp {
+	struct list_head list;
+	unsigned long	tmo;
+	u16		tag;
+	u64		sec;
+	u64		nsec;
+	u64		clkid;
+	u16		portnum;
+	u16		seqid;
+};
+
+struct ines_port {
+	struct ines_port_registers	*regs;
+	struct mii_timestamper		mii_ts;
+	struct ines_clock		*clock;
+	bool				rxts_enabled;
+	bool				txts_enabled;
+	unsigned int			index;
+	struct delayed_work		ts_work;
+	/* lock protects event list and tx_skb */
+	spinlock_t			lock;
+	struct sk_buff			*tx_skb;
+	struct list_head		events;
+	struct list_head		pool;
+	struct ines_timestamp		pool_data[INES_MAX_EVENTS];
+};
+
+struct ines_clock {
+	struct ines_port		port[INES_N_PORTS];
+	struct ines_global_regs __iomem	*regs;
+	void __iomem			*base;
+	struct device_node		*node;
+	struct device			*dev;
+	struct list_head		list;
+};
+
+static bool ines_match(struct sk_buff *skb, unsigned int ptp_class,
+		       struct ines_timestamp *ts, struct device *dev);
+static int ines_rxfifo_read(struct ines_port *port);
+static u64 ines_rxts64(struct ines_port *port, unsigned int words);
+static bool ines_timestamp_expired(struct ines_timestamp *ts);
+static u64 ines_txts64(struct ines_port *port, unsigned int words);
+static void ines_txtstamp_work(struct work_struct *work);
+static bool is_sync_pdelay_resp(struct sk_buff *skb, int type);
+static u8 tag_to_msgtype(u8 tag);
+
+static void ines_clock_cleanup(struct ines_clock *clock)
+{
+	struct ines_port *port;
+	int i;
+
+	for (i = 0; i < INES_N_PORTS; i++) {
+		port = &clock->port[i];
+		cancel_delayed_work_sync(&port->ts_work);
+	}
+}
+
+static int ines_clock_init(struct ines_clock *clock, struct device *device,
+			   void __iomem *addr)
+{
+	struct device_node *node = device->of_node;
+	unsigned long port_addr;
+	struct ines_port *port;
+	int i, j;
+
+	INIT_LIST_HEAD(&clock->list);
+	clock->node = node;
+	clock->dev  = device;
+	clock->base = addr;
+	clock->regs = clock->base;
+
+	for (i = 0; i < INES_N_PORTS; i++) {
+		port = &clock->port[i];
+		port_addr = (unsigned long) clock->base +
+			INES_PORT_OFFSET + i * INES_PORT_SIZE;
+		port->regs = (struct ines_port_registers *) port_addr;
+		port->clock = clock;
+		port->index = i;
+		INIT_DELAYED_WORK(&port->ts_work, ines_txtstamp_work);
+		spin_lock_init(&port->lock);
+		INIT_LIST_HEAD(&port->events);
+		INIT_LIST_HEAD(&port->pool);
+		for (j = 0; j < INES_MAX_EVENTS; j++)
+			list_add(&port->pool_data[j].list, &port->pool);
+	}
+
+	ines_write32(clock, 0xBEEF, test);
+	ines_write32(clock, 0xBEEF, test2);
+
+	dev_dbg(device, "ID      0x%x\n", ines_read32(clock, id));
+	dev_dbg(device, "TEST    0x%x\n", ines_read32(clock, test));
+	dev_dbg(device, "VERSION 0x%x\n", ines_read32(clock, version));
+	dev_dbg(device, "TEST2   0x%x\n", ines_read32(clock, test2));
+
+	for (i = 0; i < INES_N_PORTS; i++) {
+		port = &clock->port[i];
+		ines_write32(port, PORT_CONF, port_conf);
+	}
+
+	return 0;
+}
+
+static struct ines_port *ines_find_port(struct device_node *node, u32 index)
+{
+	struct ines_port *port = NULL;
+	struct ines_clock *clock;
+	struct list_head *this;
+
+	mutex_lock(&ines_clocks_lock);
+	list_for_each(this, &ines_clocks) {
+		clock = list_entry(this, struct ines_clock, list);
+		if (clock->node == node) {
+			port = &clock->port[index];
+			break;
+		}
+	}
+	mutex_unlock(&ines_clocks_lock);
+	return port;
+}
+
+static u64 ines_find_rxts(struct ines_port *port, struct sk_buff *skb, int type)
+{
+	struct list_head *this, *next;
+	struct ines_timestamp *ts;
+	unsigned long flags;
+	u64 ns = 0;
+
+	if (type == PTP_CLASS_NONE)
+		return 0;
+
+	spin_lock_irqsave(&port->lock, flags);
+	ines_rxfifo_read(port);
+	list_for_each_safe(this, next, &port->events) {
+		ts = list_entry(this, struct ines_timestamp, list);
+		if (ines_timestamp_expired(ts)) {
+			list_del_init(&ts->list);
+			list_add(&ts->list, &port->pool);
+			continue;
+		}
+		if (ines_match(skb, type, ts, port->clock->dev)) {
+			ns = ts->sec * 1000000000ULL + ts->nsec;
+			list_del_init(&ts->list);
+			list_add(&ts->list, &port->pool);
+			break;
+		}
+	}
+	spin_unlock_irqrestore(&port->lock, flags);
+
+	return ns;
+}
+
+static u64 ines_find_txts(struct ines_port *port, struct sk_buff *skb)
+{
+	unsigned int class = ptp_classify_raw(skb), i;
+	u32 data_rd_pos, buf_stat, mask, ts_stat_tx;
+	struct ines_timestamp ts;
+	unsigned long flags;
+	u64 ns = 0;
+
+	mask = TX_FIFO_NE_1 << port->index;
+
+	spin_lock_irqsave(&port->lock, flags);
+
+	for (i = 0; i < INES_FIFO_DEPTH; i++) {
+
+		buf_stat = ines_read32(port->clock, buf_stat);
+		if (!(buf_stat & mask)) {
+			dev_dbg(port->clock->dev,
+				  "Tx timestamp FIFO unexpectedly empty\n");
+			break;
+		}
+		ts_stat_tx = ines_read32(port, ts_stat_tx);
+		data_rd_pos = (ts_stat_tx >> DATA_READ_POS_SHIFT) &
+			DATA_READ_POS_MASK;
+		if (data_rd_pos) {
+			dev_err(port->clock->dev,
+				"unexpected Tx read pos %u\n", data_rd_pos);
+			break;
+		}
+
+		ts.tag     = ines_read32(port, ts_tx);
+		ts.sec     = ines_txts64(port, 3);
+		ts.nsec    = ines_txts64(port, 2);
+		ts.clkid   = ines_txts64(port, 4);
+		ts.portnum = ines_read32(port, ts_tx);
+		ts.seqid   = ines_read32(port, ts_tx);
+
+		if (ines_match(skb, class, &ts, port->clock->dev)) {
+			ns = ts.sec * 1000000000ULL + ts.nsec;
+			break;
+		}
+	}
+
+	spin_unlock_irqrestore(&port->lock, flags);
+	return ns;
+}
+
+static int ines_hwtstamp(struct mii_timestamper *mii_ts, struct ifreq *ifr)
+{
+	struct ines_port *port = container_of(mii_ts, struct ines_port, mii_ts);
+	u32 cm_one_step = 0, port_conf, ts_stat_rx, ts_stat_tx;
+	struct hwtstamp_config cfg;
+	unsigned long flags;
+
+	if (copy_from_user(&cfg, ifr->ifr_data, sizeof(cfg)))
+		return -EFAULT;
+
+	/* reserved for future extensions */
+	if (cfg.flags)
+		return -EINVAL;
+
+	switch (cfg.tx_type) {
+	case HWTSTAMP_TX_OFF:
+		ts_stat_tx = 0;
+		break;
+	case HWTSTAMP_TX_ON:
+		ts_stat_tx = TS_ENABLE;
+		break;
+	case HWTSTAMP_TX_ONESTEP_P2P:
+		ts_stat_tx = TS_ENABLE;
+		cm_one_step = CM_ONE_STEP;
+		break;
+	default:
+		return -ERANGE;
+	}
+
+	switch (cfg.rx_filter) {
+	case HWTSTAMP_FILTER_NONE:
+		ts_stat_rx = 0;
+		break;
+	case HWTSTAMP_FILTER_ALL:
+	case HWTSTAMP_FILTER_PTP_V1_L4_EVENT:
+	case HWTSTAMP_FILTER_PTP_V1_L4_SYNC:
+	case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ:
+		return -ERANGE;
+	case HWTSTAMP_FILTER_PTP_V2_L4_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_L4_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ:
+	case HWTSTAMP_FILTER_PTP_V2_L2_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_L2_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ:
+	case HWTSTAMP_FILTER_PTP_V2_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ:
+		ts_stat_rx = TS_ENABLE;
+		cfg.rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT;
+		break;
+	default:
+		return -ERANGE;
+	}
+
+	spin_lock_irqsave(&port->lock, flags);
+
+	port_conf = ines_read32(port, port_conf);
+	port_conf &= ~CM_ONE_STEP;
+	port_conf |= cm_one_step;
+
+	ines_write32(port, port_conf, port_conf);
+	ines_write32(port, ts_stat_rx, ts_stat_rx);
+	ines_write32(port, ts_stat_tx, ts_stat_tx);
+
+	port->rxts_enabled = ts_stat_rx == TS_ENABLE ? true : false;
+	port->txts_enabled = ts_stat_tx == TS_ENABLE ? true : false;
+
+	spin_unlock_irqrestore(&port->lock, flags);
+
+	return copy_to_user(ifr->ifr_data, &cfg, sizeof(cfg)) ? -EFAULT : 0;
+}
+
+static void ines_link_state(struct mii_timestamper *mii_ts,
+			    struct phy_device *phydev)
+{
+	struct ines_port *port = container_of(mii_ts, struct ines_port, mii_ts);
+	u32 port_conf, speed_conf;
+	unsigned long flags;
+
+	switch (phydev->speed) {
+	case SPEED_10:
+		speed_conf = PHY_SPEED_10 << PHY_SPEED_SHIFT;
+		break;
+	case SPEED_100:
+		speed_conf = PHY_SPEED_100 << PHY_SPEED_SHIFT;
+		break;
+	case SPEED_1000:
+		speed_conf = PHY_SPEED_1000 << PHY_SPEED_SHIFT;
+		break;
+	default:
+		dev_err(port->clock->dev, "bad speed: %d\n", phydev->speed);
+		return;
+	}
+	spin_lock_irqsave(&port->lock, flags);
+
+	port_conf = ines_read32(port, port_conf);
+	port_conf &= ~(0x3 << PHY_SPEED_SHIFT);
+	port_conf |= speed_conf;
+
+	ines_write32(port, port_conf, port_conf);
+
+	spin_unlock_irqrestore(&port->lock, flags);
+}
+
+static bool ines_match(struct sk_buff *skb, unsigned int ptp_class,
+		       struct ines_timestamp *ts, struct device *dev)
+{
+	u8 *msgtype, *data = skb_mac_header(skb);
+	unsigned int offset = 0;
+	__be16 *portn, *seqid;
+	__be64 *clkid;
+
+	if (unlikely(ptp_class & PTP_CLASS_V1))
+		return false;
+
+	if (ptp_class & PTP_CLASS_VLAN)
+		offset += VLAN_HLEN;
+
+	switch (ptp_class & PTP_CLASS_PMASK) {
+	case PTP_CLASS_IPV4:
+		offset += ETH_HLEN + IPV4_HLEN(data + offset) + UDP_HLEN;
+		break;
+	case PTP_CLASS_IPV6:
+		offset += ETH_HLEN + IP6_HLEN + UDP_HLEN;
+		break;
+	case PTP_CLASS_L2:
+		offset += ETH_HLEN;
+		break;
+	default:
+		return false;
+	}
+
+	if (skb->len + ETH_HLEN < offset + OFF_PTP_SEQUENCE_ID + sizeof(*seqid))
+		return false;
+
+	msgtype = data + offset;
+	clkid = (__be64 *)(data + offset + OFF_PTP_CLOCK_ID);
+	portn = (__be16 *)(data + offset + OFF_PTP_PORT_NUM);
+	seqid = (__be16 *)(data + offset + OFF_PTP_SEQUENCE_ID);
+
+	if (tag_to_msgtype(ts->tag & 0x7) != (*msgtype & 0xf)) {
+		dev_dbg(dev, "msgtype mismatch ts %hhu != skb %hhu\n",
+			  tag_to_msgtype(ts->tag & 0x7), *msgtype & 0xf);
+		return false;
+	}
+	if (cpu_to_be64(ts->clkid) != *clkid) {
+		dev_dbg(dev, "clkid mismatch ts %llx != skb %llx\n",
+			  cpu_to_be64(ts->clkid), *clkid);
+		return false;
+	}
+	if (ts->portnum != ntohs(*portn)) {
+		dev_dbg(dev, "portn mismatch ts %hu != skb %hu\n",
+			  ts->portnum, ntohs(*portn));
+		return false;
+	}
+	if (ts->seqid != ntohs(*seqid)) {
+		dev_dbg(dev, "seqid mismatch ts %hu != skb %hu\n",
+			  ts->seqid, ntohs(*seqid));
+		return false;
+	}
+
+	return true;
+}
+
+static bool ines_rxtstamp(struct mii_timestamper *mii_ts,
+			  struct sk_buff *skb, int type)
+{
+	struct ines_port *port = container_of(mii_ts, struct ines_port, mii_ts);
+	struct skb_shared_hwtstamps *ssh;
+	u64 ns;
+
+	if (!port->rxts_enabled)
+		return false;
+
+	ns = ines_find_rxts(port, skb, type);
+	if (!ns)
+		return false;
+
+	ssh = skb_hwtstamps(skb);
+	ssh->hwtstamp = ns_to_ktime(ns);
+	netif_rx(skb);
+
+	return true;
+}
+
+static int ines_rxfifo_read(struct ines_port *port)
+{
+	u32 data_rd_pos, buf_stat, mask, ts_stat_rx;
+	struct ines_timestamp *ts;
+	unsigned int i;
+
+	mask = RX_FIFO_NE_1 << port->index;
+
+	for (i = 0; i < INES_FIFO_DEPTH; i++) {
+		if (list_empty(&port->pool)) {
+			dev_err(port->clock->dev, "event pool is empty\n");
+			return -1;
+		}
+		buf_stat = ines_read32(port->clock, buf_stat);
+		if (!(buf_stat & mask))
+			break;
+
+		ts_stat_rx = ines_read32(port, ts_stat_rx);
+		data_rd_pos = (ts_stat_rx >> DATA_READ_POS_SHIFT) &
+			DATA_READ_POS_MASK;
+		if (data_rd_pos) {
+			dev_err(port->clock->dev, "unexpected Rx read pos %u\n",
+				data_rd_pos);
+			break;
+		}
+
+		ts = list_first_entry(&port->pool, struct ines_timestamp, list);
+		ts->tmo     = jiffies + HZ;
+		ts->tag     = ines_read32(port, ts_rx);
+		ts->sec     = ines_rxts64(port, 3);
+		ts->nsec    = ines_rxts64(port, 2);
+		ts->clkid   = ines_rxts64(port, 4);
+		ts->portnum = ines_read32(port, ts_rx);
+		ts->seqid   = ines_read32(port, ts_rx);
+
+		list_del_init(&ts->list);
+		list_add_tail(&ts->list, &port->events);
+	}
+
+	return 0;
+}
+
+static u64 ines_rxts64(struct ines_port *port, unsigned int words)
+{
+	unsigned int i;
+	u64 result;
+	u16 word;
+
+	word = ines_read32(port, ts_rx);
+	result = word;
+	words--;
+	for (i = 0; i < words; i++) {
+		word = ines_read32(port, ts_rx);
+		result <<= 16;
+		result |= word;
+	}
+	return result;
+}
+
+static bool ines_timestamp_expired(struct ines_timestamp *ts)
+{
+	return time_after(jiffies, ts->tmo);
+}
+
+static int ines_ts_info(struct mii_timestamper *mii_ts,
+			struct ethtool_ts_info *info)
+{
+	info->so_timestamping =
+		SOF_TIMESTAMPING_TX_HARDWARE |
+		SOF_TIMESTAMPING_TX_SOFTWARE |
+		SOF_TIMESTAMPING_RX_HARDWARE |
+		SOF_TIMESTAMPING_RX_SOFTWARE |
+		SOF_TIMESTAMPING_SOFTWARE |
+		SOF_TIMESTAMPING_RAW_HARDWARE;
+
+	info->phc_index = -1;
+
+	info->tx_types =
+		(1 << HWTSTAMP_TX_OFF) |
+		(1 << HWTSTAMP_TX_ON) |
+		(1 << HWTSTAMP_TX_ONESTEP_P2P);
+
+	info->rx_filters =
+		(1 << HWTSTAMP_FILTER_NONE) |
+		(1 << HWTSTAMP_FILTER_PTP_V2_EVENT);
+
+	return 0;
+}
+
+static u64 ines_txts64(struct ines_port *port, unsigned int words)
+{
+	unsigned int i;
+	u64 result;
+	u16 word;
+
+	word = ines_read32(port, ts_tx);
+	result = word;
+	words--;
+	for (i = 0; i < words; i++) {
+		word = ines_read32(port, ts_tx);
+		result <<= 16;
+		result |= word;
+	}
+	return result;
+}
+
+static bool ines_txts_onestep(struct ines_port *port, struct sk_buff *skb, int type)
+{
+	unsigned long flags;
+	u32 port_conf;
+
+	spin_lock_irqsave(&port->lock, flags);
+	port_conf = ines_read32(port, port_conf);
+	spin_unlock_irqrestore(&port->lock, flags);
+
+	if (port_conf & CM_ONE_STEP)
+		return is_sync_pdelay_resp(skb, type);
+
+	return false;
+}
+
+static void ines_txtstamp(struct mii_timestamper *mii_ts,
+			  struct sk_buff *skb, int type)
+{
+	struct ines_port *port = container_of(mii_ts, struct ines_port, mii_ts);
+	struct sk_buff *old_skb = NULL;
+	unsigned long flags;
+
+	if (!port->txts_enabled || ines_txts_onestep(port, skb, type)) {
+		kfree_skb(skb);
+		return;
+	}
+
+	spin_lock_irqsave(&port->lock, flags);
+
+	if (port->tx_skb)
+		old_skb = port->tx_skb;
+
+	port->tx_skb = skb;
+
+	spin_unlock_irqrestore(&port->lock, flags);
+
+	if (old_skb)
+		kfree_skb(old_skb);
+
+	schedule_delayed_work(&port->ts_work, 1);
+}
+
+static void ines_txtstamp_work(struct work_struct *work)
+{
+	struct ines_port *port =
+		container_of(work, struct ines_port, ts_work.work);
+	struct skb_shared_hwtstamps ssh;
+	struct sk_buff *skb;
+	unsigned long flags;
+	u64 ns;
+
+	spin_lock_irqsave(&port->lock, flags);
+	skb = port->tx_skb;
+	port->tx_skb = NULL;
+	spin_unlock_irqrestore(&port->lock, flags);
+
+	ns = ines_find_txts(port, skb);
+	if (!ns) {
+		kfree_skb(skb);
+		return;
+	}
+	ssh.hwtstamp = ns_to_ktime(ns);
+	skb_complete_tx_timestamp(skb, &ssh);
+}
+
+static bool is_sync_pdelay_resp(struct sk_buff *skb, int type)
+{
+	u8 *data = skb->data, *msgtype;
+	unsigned int offset = 0;
+
+	if (type & PTP_CLASS_VLAN)
+		offset += VLAN_HLEN;
+
+	switch (type & PTP_CLASS_PMASK) {
+	case PTP_CLASS_IPV4:
+		offset += ETH_HLEN + IPV4_HLEN(data + offset) + UDP_HLEN;
+		break;
+	case PTP_CLASS_IPV6:
+		offset += ETH_HLEN + IP6_HLEN + UDP_HLEN;
+		break;
+	case PTP_CLASS_L2:
+		offset += ETH_HLEN;
+		break;
+	default:
+		return 0;
+	}
+
+	if (type & PTP_CLASS_V1)
+		offset += OFF_PTP_CONTROL;
+
+	if (skb->len < offset + 1)
+		return 0;
+
+	msgtype = data + offset;
+
+	switch ((*msgtype & 0xf)) {
+	case SYNC:
+	case PDELAY_RESP:
+		return true;
+	default:
+		return false;
+	}
+}
+
+static u8 tag_to_msgtype(u8 tag)
+{
+	switch (tag) {
+	case MESSAGE_TYPE_SYNC:
+		return SYNC;
+	case MESSAGE_TYPE_P_DELAY_REQ:
+		return PDELAY_REQ;
+	case MESSAGE_TYPE_P_DELAY_RESP:
+		return PDELAY_RESP;
+	case MESSAGE_TYPE_DELAY_REQ:
+		return DELAY_REQ;
+	}
+	return 0xf;
+}
+
+static struct mii_timestamper *ines_ptp_probe_channel(struct device *device,
+						      unsigned int index)
+{
+	struct device_node *node = device->of_node;
+	struct ines_port *port;
+
+	if (index > INES_N_PORTS - 1) {
+		dev_err(device, "bad port index %u\n", index);
+		return ERR_PTR(-EINVAL);
+	}
+	port = ines_find_port(node, index);
+	if (!port) {
+		dev_err(device, "missing port index %u\n", index);
+		return ERR_PTR(-ENODEV);
+	}
+	port->mii_ts.rxtstamp = ines_rxtstamp;
+	port->mii_ts.txtstamp = ines_txtstamp;
+	port->mii_ts.hwtstamp = ines_hwtstamp;
+	port->mii_ts.link_state = ines_link_state;
+	port->mii_ts.ts_info = ines_ts_info;
+
+	return &port->mii_ts;
+}
+
+static void ines_ptp_release_channel(struct device *device,
+				     struct mii_timestamper *mii_ts)
+{
+}
+
+static struct mii_timestamping_ctrl ines_ctrl = {
+	.probe_channel = ines_ptp_probe_channel,
+	.release_channel = ines_ptp_release_channel,
+};
+
+static int ines_ptp_ctrl_probe(struct platform_device *pld)
+{
+	struct ines_clock *clock;
+	struct resource *res;
+	void __iomem *addr;
+	int err = 0;
+
+	res = platform_get_resource(pld, IORESOURCE_MEM, 0);
+	if (!res) {
+		dev_err(&pld->dev, "missing memory resource\n");
+		return -EINVAL;
+	}
+	addr = devm_ioremap_resource(&pld->dev, res);
+	if (IS_ERR(addr)) {
+		err = PTR_ERR(addr);
+		goto out;
+	}
+	clock = kzalloc(sizeof(*clock), GFP_KERNEL);
+	if (!clock) {
+		err = -ENOMEM;
+		goto out;
+	}
+	if (ines_clock_init(clock, &pld->dev, addr)) {
+		kfree(clock);
+		err = -ENOMEM;
+		goto out;
+	}
+	err = register_mii_tstamp_controller(&pld->dev, &ines_ctrl);
+	if (err) {
+		kfree(clock);
+		goto out;
+	}
+	mutex_lock(&ines_clocks_lock);
+	list_add_tail(&ines_clocks, &clock->list);
+	mutex_unlock(&ines_clocks_lock);
+
+	dev_set_drvdata(&pld->dev, clock);
+out:
+	return err;
+}
+
+static int ines_ptp_ctrl_remove(struct platform_device *pld)
+{
+	struct ines_clock *clock = dev_get_drvdata(&pld->dev);
+
+	unregister_mii_tstamp_controller(&pld->dev);
+	mutex_lock(&ines_clocks_lock);
+	list_del(&clock->list);
+	mutex_unlock(&ines_clocks_lock);
+	ines_clock_cleanup(clock);
+	kfree(clock);
+	return 0;
+}
+
+static const struct of_device_id ines_ptp_ctrl_of_match[] = {
+	{ .compatible = "ines,ptp-ctrl" },
+	{ }
+};
+
+MODULE_DEVICE_TABLE(of, ines_ptp_ctrl_of_match);
+
+static struct platform_driver ines_ptp_ctrl_driver = {
+	.probe  = ines_ptp_ctrl_probe,
+	.remove = ines_ptp_ctrl_remove,
+	.driver = {
+		.name = "ines_ptp_ctrl",
+		.of_match_table = of_match_ptr(ines_ptp_ctrl_of_match),
+	},
+};
+module_platform_driver(ines_ptp_ctrl_driver);
-- 
cgit v1.2.3-59-g8ed1b


From 8ab9da573dad95536d8a92f19d7967e8142cd827 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Mon, 23 Dec 2019 10:03:05 -0800
Subject: libbpf: Support CO-RE relocations for LDX/ST/STX instructions

Clang patch [0] enables emitting relocatable generic ALU/ALU64 instructions
(i.e, shifts and arithmetic operations), as well as generic load/store
instructions. The former ones are already supported by libbpf as is. This
patch adds further support for load/store instructions. Relocatable field
offset is encoded in BPF instruction's 16-bit offset section and are adjusted
by libbpf based on target kernel BTF.

These Clang changes and corresponding libbpf changes allow for more succinct
generated BPF code by encoding relocatable field reads as a single
ST/LDX/STX instruction. It also enables relocatable access to BPF context.
Previously, if context struct (e.g., __sk_buff) was accessed with CO-RE
relocations (e.g., due to preserve_access_index attribute), it would be
rejected by BPF verifier due to modified context pointer dereference. With
Clang patch, such context accesses are both relocatable and have a fixed
offset from the point of view of BPF verifier.

  [0] https://reviews.llvm.org/D71790

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20191223180305.86417-1-andriin@fb.com
---
 tools/lib/bpf/libbpf.c | 31 ++++++++++++++++++++++++++++---
 1 file changed, 28 insertions(+), 3 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 9576a90c5a1c..7513165b104f 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -18,6 +18,7 @@
 #include <stdarg.h>
 #include <libgen.h>
 #include <inttypes.h>
+#include <limits.h>
 #include <string.h>
 #include <unistd.h>
 #include <endian.h>
@@ -3810,11 +3811,13 @@ static int bpf_core_reloc_insn(struct bpf_program *prog,
 	insn = &prog->insns[insn_idx];
 	class = BPF_CLASS(insn->code);
 
-	if (class == BPF_ALU || class == BPF_ALU64) {
+	switch (class) {
+	case BPF_ALU:
+	case BPF_ALU64:
 		if (BPF_SRC(insn->code) != BPF_K)
 			return -EINVAL;
 		if (!failed && validate && insn->imm != orig_val) {
-			pr_warn("prog '%s': unexpected insn #%d value: got %u, exp %u -> %u\n",
+			pr_warn("prog '%s': unexpected insn #%d (ALU/ALU64) value: got %u, exp %u -> %u\n",
 				bpf_program__title(prog, false), insn_idx,
 				insn->imm, orig_val, new_val);
 			return -EINVAL;
@@ -3824,7 +3827,29 @@ static int bpf_core_reloc_insn(struct bpf_program *prog,
 		pr_debug("prog '%s': patched insn #%d (ALU/ALU64)%s imm %u -> %u\n",
 			 bpf_program__title(prog, false), insn_idx,
 			 failed ? " w/ failed reloc" : "", orig_val, new_val);
-	} else {
+		break;
+	case BPF_LDX:
+	case BPF_ST:
+	case BPF_STX:
+		if (!failed && validate && insn->off != orig_val) {
+			pr_warn("prog '%s': unexpected insn #%d (LD/LDX/ST/STX) value: got %u, exp %u -> %u\n",
+				bpf_program__title(prog, false), insn_idx,
+				insn->off, orig_val, new_val);
+			return -EINVAL;
+		}
+		if (new_val > SHRT_MAX) {
+			pr_warn("prog '%s': insn #%d (LD/LDX/ST/STX) value too big: %u\n",
+				bpf_program__title(prog, false), insn_idx,
+				new_val);
+			return -ERANGE;
+		}
+		orig_val = insn->off;
+		insn->off = new_val;
+		pr_debug("prog '%s': patched insn #%d (LD/LDX/ST/STX)%s off %u -> %u\n",
+			 bpf_program__title(prog, false), insn_idx,
+			 failed ? " w/ failed reloc" : "", orig_val, new_val);
+		break;
+	default:
 		pr_warn("prog '%s': trying to relocate unrecognized insn #%d, code:%x, src:%x, dst:%x, off:%x, imm:%x\n",
 			bpf_program__title(prog, false),
 			insn_idx, insn->code, insn->src_reg, insn->dst_reg,
-- 
cgit v1.2.3-59-g8ed1b


From 1162f844030ac1ac7321b5e8f6c9badc7a11428f Mon Sep 17 00:00:00 2001
From: Hechao Li <hechaol@fb.com>
Date: Mon, 23 Dec 2019 17:17:42 -0800
Subject: bpf: Print error message for bpftool cgroup show

Currently, when bpftool cgroup show <path> has an error, no error
message is printed. This is confusing because the user may think the
result is empty.

Before the change:

$ bpftool cgroup show /sys/fs/cgroup
ID       AttachType      AttachFlags     Name
$ echo $?
255

After the change:
$ ./bpftool cgroup show /sys/fs/cgroup
Error: can't query bpf programs attached to /sys/fs/cgroup: Operation
not permitted

v2: Rename check_query_cgroup_progs to cgroup_has_attached_progs

Signed-off-by: Hechao Li <hechaol@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191224011742.3714301-1-hechaol@fb.com
---
 tools/bpf/bpftool/cgroup.c | 56 ++++++++++++++++++++++++++++++++--------------
 1 file changed, 39 insertions(+), 17 deletions(-)

diff --git a/tools/bpf/bpftool/cgroup.c b/tools/bpf/bpftool/cgroup.c
index 1ef45e55039e..2f017caa678d 100644
--- a/tools/bpf/bpftool/cgroup.c
+++ b/tools/bpf/bpftool/cgroup.c
@@ -117,6 +117,25 @@ static int count_attached_bpf_progs(int cgroup_fd, enum bpf_attach_type type)
 	return prog_cnt;
 }
 
+static int cgroup_has_attached_progs(int cgroup_fd)
+{
+	enum bpf_attach_type type;
+	bool no_prog = true;
+
+	for (type = 0; type < __MAX_BPF_ATTACH_TYPE; type++) {
+		int count = count_attached_bpf_progs(cgroup_fd, type);
+
+		if (count < 0 && errno != EINVAL)
+			return -1;
+
+		if (count > 0) {
+			no_prog = false;
+			break;
+		}
+	}
+
+	return no_prog ? 0 : 1;
+}
 static int show_attached_bpf_progs(int cgroup_fd, enum bpf_attach_type type,
 				   int level)
 {
@@ -161,6 +180,7 @@ static int show_attached_bpf_progs(int cgroup_fd, enum bpf_attach_type type,
 static int do_show(int argc, char **argv)
 {
 	enum bpf_attach_type type;
+	int has_attached_progs;
 	const char *path;
 	int cgroup_fd;
 	int ret = -1;
@@ -192,6 +212,16 @@ static int do_show(int argc, char **argv)
 		goto exit;
 	}
 
+	has_attached_progs = cgroup_has_attached_progs(cgroup_fd);
+	if (has_attached_progs < 0) {
+		p_err("can't query bpf programs attached to %s: %s",
+		      path, strerror(errno));
+		goto exit_cgroup;
+	} else if (!has_attached_progs) {
+		ret = 0;
+		goto exit_cgroup;
+	}
+
 	if (json_output)
 		jsonw_start_array(json_wtr);
 	else
@@ -212,6 +242,7 @@ static int do_show(int argc, char **argv)
 	if (json_output)
 		jsonw_end_array(json_wtr);
 
+exit_cgroup:
 	close(cgroup_fd);
 exit:
 	return ret;
@@ -228,7 +259,7 @@ static int do_show_tree_fn(const char *fpath, const struct stat *sb,
 			   int typeflag, struct FTW *ftw)
 {
 	enum bpf_attach_type type;
-	bool skip = true;
+	int has_attached_progs;
 	int cgroup_fd;
 
 	if (typeflag != FTW_D)
@@ -240,22 +271,13 @@ static int do_show_tree_fn(const char *fpath, const struct stat *sb,
 		return SHOW_TREE_FN_ERR;
 	}
 
-	for (type = 0; type < __MAX_BPF_ATTACH_TYPE; type++) {
-		int count = count_attached_bpf_progs(cgroup_fd, type);
-
-		if (count < 0 && errno != EINVAL) {
-			p_err("can't query bpf programs attached to %s: %s",
-			      fpath, strerror(errno));
-			close(cgroup_fd);
-			return SHOW_TREE_FN_ERR;
-		}
-		if (count > 0) {
-			skip = false;
-			break;
-		}
-	}
-
-	if (skip) {
+	has_attached_progs = cgroup_has_attached_progs(cgroup_fd);
+	if (has_attached_progs < 0) {
+		p_err("can't query bpf programs attached to %s: %s",
+		      fpath, strerror(errno));
+		close(cgroup_fd);
+		return SHOW_TREE_FN_ERR;
+	} else if (!has_attached_progs) {
 		close(cgroup_fd);
 		return 0;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From db8f6f5c8de6dae924a68858ad6a4217f735be13 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Wed, 18 Dec 2019 12:05:13 +0100
Subject: netfilter: nft_meta: move time handling to helper

reduce size of the (large) meta evaluation function.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nft_meta.c | 28 ++++++++++++++++++++++------
 1 file changed, 22 insertions(+), 6 deletions(-)

diff --git a/net/netfilter/nft_meta.c b/net/netfilter/nft_meta.c
index 9740b554fdb3..ba74f3ee7264 100644
--- a/net/netfilter/nft_meta.c
+++ b/net/netfilter/nft_meta.c
@@ -33,8 +33,9 @@
 
 static DEFINE_PER_CPU(struct rnd_state, nft_prandom_state);
 
-static u8 nft_meta_weekday(time64_t secs)
+static u8 nft_meta_weekday(void)
 {
+	time64_t secs = ktime_get_real_seconds();
 	unsigned int dse;
 	u8 wday;
 
@@ -56,6 +57,25 @@ static u32 nft_meta_hour(time64_t secs)
 		+ tm.tm_sec;
 }
 
+static noinline_for_stack void
+nft_meta_get_eval_time(enum nft_meta_keys key,
+		       u32 *dest)
+{
+	switch (key) {
+	case NFT_META_TIME_NS:
+		nft_reg_store64(dest, ktime_get_real_ns());
+		break;
+	case NFT_META_TIME_DAY:
+		nft_reg_store8(dest, nft_meta_weekday());
+		break;
+	case NFT_META_TIME_HOUR:
+		*dest = nft_meta_hour(ktime_get_real_seconds());
+		break;
+	default:
+		break;
+	}
+}
+
 void nft_meta_get_eval(const struct nft_expr *expr,
 		       struct nft_regs *regs,
 		       const struct nft_pktinfo *pkt)
@@ -247,13 +267,9 @@ void nft_meta_get_eval(const struct nft_expr *expr,
 		strncpy((char *)dest, out->rtnl_link_ops->kind, IFNAMSIZ);
 		break;
 	case NFT_META_TIME_NS:
-		nft_reg_store64(dest, ktime_get_real_ns());
-		break;
 	case NFT_META_TIME_DAY:
-		nft_reg_store8(dest, nft_meta_weekday(ktime_get_real_seconds()));
-		break;
 	case NFT_META_TIME_HOUR:
-		*dest = nft_meta_hour(ktime_get_real_seconds());
+		nft_meta_get_eval_time(priv->key, dest);
 		break;
 	default:
 		WARN_ON(1);
-- 
cgit v1.2.3-59-g8ed1b


From 4a54594abdbee230a0471aa137b3d5405c897661 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Wed, 18 Dec 2019 12:05:14 +0100
Subject: netfilter: nft_meta: move pkttype handling to helper

When pkttype is loopback, nft_meta performs guesswork to detect
broad/multicast packets. Place this in a helper, this is hardly a hot path.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nft_meta.c | 90 +++++++++++++++++++++++++++---------------------
 1 file changed, 51 insertions(+), 39 deletions(-)

diff --git a/net/netfilter/nft_meta.c b/net/netfilter/nft_meta.c
index ba74f3ee7264..fe49b27dfa87 100644
--- a/net/netfilter/nft_meta.c
+++ b/net/netfilter/nft_meta.c
@@ -76,6 +76,56 @@ nft_meta_get_eval_time(enum nft_meta_keys key,
 	}
 }
 
+static noinline bool
+nft_meta_get_eval_pkttype_lo(const struct nft_pktinfo *pkt,
+			     u32 *dest)
+{
+	const struct sk_buff *skb = pkt->skb;
+
+	switch (nft_pf(pkt)) {
+	case NFPROTO_IPV4:
+		if (ipv4_is_multicast(ip_hdr(skb)->daddr))
+			nft_reg_store8(dest, PACKET_MULTICAST);
+		else
+			nft_reg_store8(dest, PACKET_BROADCAST);
+		break;
+	case NFPROTO_IPV6:
+		nft_reg_store8(dest, PACKET_MULTICAST);
+		break;
+	case NFPROTO_NETDEV:
+		switch (skb->protocol) {
+		case htons(ETH_P_IP): {
+			int noff = skb_network_offset(skb);
+			struct iphdr *iph, _iph;
+
+			iph = skb_header_pointer(skb, noff,
+						 sizeof(_iph), &_iph);
+			if (!iph)
+				return false;
+
+			if (ipv4_is_multicast(iph->daddr))
+				nft_reg_store8(dest, PACKET_MULTICAST);
+			else
+				nft_reg_store8(dest, PACKET_BROADCAST);
+
+			break;
+		}
+		case htons(ETH_P_IPV6):
+			nft_reg_store8(dest, PACKET_MULTICAST);
+			break;
+		default:
+			WARN_ON_ONCE(1);
+			return false;
+		}
+		break;
+	default:
+		WARN_ON_ONCE(1);
+		return false;
+	}
+
+	return true;
+}
+
 void nft_meta_get_eval(const struct nft_expr *expr,
 		       struct nft_regs *regs,
 		       const struct nft_pktinfo *pkt)
@@ -183,46 +233,8 @@ void nft_meta_get_eval(const struct nft_expr *expr,
 			break;
 		}
 
-		switch (nft_pf(pkt)) {
-		case NFPROTO_IPV4:
-			if (ipv4_is_multicast(ip_hdr(skb)->daddr))
-				nft_reg_store8(dest, PACKET_MULTICAST);
-			else
-				nft_reg_store8(dest, PACKET_BROADCAST);
-			break;
-		case NFPROTO_IPV6:
-			nft_reg_store8(dest, PACKET_MULTICAST);
-			break;
-		case NFPROTO_NETDEV:
-			switch (skb->protocol) {
-			case htons(ETH_P_IP): {
-				int noff = skb_network_offset(skb);
-				struct iphdr *iph, _iph;
-
-				iph = skb_header_pointer(skb, noff,
-							 sizeof(_iph), &_iph);
-				if (!iph)
-					goto err;
-
-				if (ipv4_is_multicast(iph->daddr))
-					nft_reg_store8(dest, PACKET_MULTICAST);
-				else
-					nft_reg_store8(dest, PACKET_BROADCAST);
-
-				break;
-			}
-			case htons(ETH_P_IPV6):
-				nft_reg_store8(dest, PACKET_MULTICAST);
-				break;
-			default:
-				WARN_ON_ONCE(1);
-				goto err;
-			}
-			break;
-		default:
-			WARN_ON_ONCE(1);
+		if (!nft_meta_get_eval_pkttype_lo(pkt, dest))
 			goto err;
-		}
 		break;
 	case NFT_META_CPU:
 		*dest = raw_smp_processor_id();
-- 
cgit v1.2.3-59-g8ed1b


From 726b44f044e8e67cbe2209c1a5704aca981be3b2 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Wed, 18 Dec 2019 12:05:15 +0100
Subject: netfilter: nft_meta: move sk uid/git handling to helper

Not a hot path.  Also, both have copy&paste case statements,
so use a common helper for both.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nft_meta.c | 65 +++++++++++++++++++++++++++---------------------
 1 file changed, 36 insertions(+), 29 deletions(-)

diff --git a/net/netfilter/nft_meta.c b/net/netfilter/nft_meta.c
index fe49b27dfa87..1b32440ec2e6 100644
--- a/net/netfilter/nft_meta.c
+++ b/net/netfilter/nft_meta.c
@@ -126,6 +126,41 @@ nft_meta_get_eval_pkttype_lo(const struct nft_pktinfo *pkt,
 	return true;
 }
 
+static noinline bool
+nft_meta_get_eval_skugid(enum nft_meta_keys key,
+			 u32 *dest,
+			 const struct nft_pktinfo *pkt)
+{
+	struct sock *sk = skb_to_full_sk(pkt->skb);
+	struct socket *sock;
+
+	if (!sk || !sk_fullsock(sk) || !net_eq(nft_net(pkt), sock_net(sk)))
+		return false;
+
+	read_lock_bh(&sk->sk_callback_lock);
+	sock = sk->sk_socket;
+	if (!sock || !sock->file) {
+		read_unlock_bh(&sk->sk_callback_lock);
+		return false;
+	}
+
+	switch (key) {
+	case NFT_META_SKUID:
+		*dest = from_kuid_munged(&init_user_ns,
+					 sock->file->f_cred->fsuid);
+		break;
+	case NFT_META_SKGID:
+		*dest =	from_kgid_munged(&init_user_ns,
+					 sock->file->f_cred->fsgid);
+		break;
+	default:
+		break;
+	}
+
+	read_unlock_bh(&sk->sk_callback_lock);
+	return true;
+}
+
 void nft_meta_get_eval(const struct nft_expr *expr,
 		       struct nft_regs *regs,
 		       const struct nft_pktinfo *pkt)
@@ -180,37 +215,9 @@ void nft_meta_get_eval(const struct nft_expr *expr,
 		nft_reg_store16(dest, out->type);
 		break;
 	case NFT_META_SKUID:
-		sk = skb_to_full_sk(skb);
-		if (!sk || !sk_fullsock(sk) ||
-		    !net_eq(nft_net(pkt), sock_net(sk)))
-			goto err;
-
-		read_lock_bh(&sk->sk_callback_lock);
-		if (sk->sk_socket == NULL ||
-		    sk->sk_socket->file == NULL) {
-			read_unlock_bh(&sk->sk_callback_lock);
-			goto err;
-		}
-
-		*dest =	from_kuid_munged(&init_user_ns,
-				sk->sk_socket->file->f_cred->fsuid);
-		read_unlock_bh(&sk->sk_callback_lock);
-		break;
 	case NFT_META_SKGID:
-		sk = skb_to_full_sk(skb);
-		if (!sk || !sk_fullsock(sk) ||
-		    !net_eq(nft_net(pkt), sock_net(sk)))
+		if (!nft_meta_get_eval_skugid(priv->key, dest, pkt))
 			goto err;
-
-		read_lock_bh(&sk->sk_callback_lock);
-		if (sk->sk_socket == NULL ||
-		    sk->sk_socket->file == NULL) {
-			read_unlock_bh(&sk->sk_callback_lock);
-			goto err;
-		}
-		*dest =	from_kgid_munged(&init_user_ns,
-				 sk->sk_socket->file->f_cred->fsgid);
-		read_unlock_bh(&sk->sk_callback_lock);
 		break;
 #ifdef CONFIG_IP_ROUTE_CLASSID
 	case NFT_META_RTCLASSID: {
-- 
cgit v1.2.3-59-g8ed1b


From b1327fbc29915ef5bf0eec5abc2e2e67983cb037 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Wed, 18 Dec 2019 12:05:16 +0100
Subject: netfilter: nft_meta: move cgroup handling to helper

Reduce size of main eval function.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nft_meta.c | 20 +++++++++++++++-----
 1 file changed, 15 insertions(+), 5 deletions(-)

diff --git a/net/netfilter/nft_meta.c b/net/netfilter/nft_meta.c
index 1b32440ec2e6..3fca1c3ec361 100644
--- a/net/netfilter/nft_meta.c
+++ b/net/netfilter/nft_meta.c
@@ -161,6 +161,20 @@ nft_meta_get_eval_skugid(enum nft_meta_keys key,
 	return true;
 }
 
+#ifdef CONFIG_CGROUP_NET_CLASSID
+static noinline bool
+nft_meta_get_eval_cgroup(u32 *dest, const struct nft_pktinfo *pkt)
+{
+	struct sock *sk = skb_to_full_sk(pkt->skb);
+
+	if (!sk || !sk_fullsock(sk) || !net_eq(nft_net(pkt), sock_net(sk)))
+		return false;
+
+	*dest = sock_cgroup_classid(&sk->sk_cgrp_data);
+	return true;
+}
+#endif
+
 void nft_meta_get_eval(const struct nft_expr *expr,
 		       struct nft_regs *regs,
 		       const struct nft_pktinfo *pkt)
@@ -168,7 +182,6 @@ void nft_meta_get_eval(const struct nft_expr *expr,
 	const struct nft_meta *priv = nft_expr_priv(expr);
 	const struct sk_buff *skb = pkt->skb;
 	const struct net_device *in = nft_in(pkt), *out = nft_out(pkt);
-	struct sock *sk;
 	u32 *dest = &regs->data[priv->dreg];
 
 	switch (priv->key) {
@@ -258,11 +271,8 @@ void nft_meta_get_eval(const struct nft_expr *expr,
 		break;
 #ifdef CONFIG_CGROUP_NET_CLASSID
 	case NFT_META_CGROUP:
-		sk = skb_to_full_sk(skb);
-		if (!sk || !sk_fullsock(sk) ||
-		    !net_eq(nft_net(pkt), sock_net(sk)))
+		if (!nft_meta_get_eval_cgroup(dest, pkt))
 			goto err;
-		*dest = sock_cgroup_classid(&sk->sk_cgrp_data);
 		break;
 #endif
 	case NFT_META_PRANDOM: {
-- 
cgit v1.2.3-59-g8ed1b


From a4150a1faa3604beef25ed2374d537f86059ad52 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Wed, 18 Dec 2019 12:05:17 +0100
Subject: netfilter: nft_meta: move interface kind handling to helper

checkpatch complains about == NULL checks in original code,
so use !in instead.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nft_meta.c | 31 +++++++++++++++++++++++++------
 1 file changed, 25 insertions(+), 6 deletions(-)

diff --git a/net/netfilter/nft_meta.c b/net/netfilter/nft_meta.c
index 3fca1c3ec361..2f7cc64b0c15 100644
--- a/net/netfilter/nft_meta.c
+++ b/net/netfilter/nft_meta.c
@@ -175,6 +175,30 @@ nft_meta_get_eval_cgroup(u32 *dest, const struct nft_pktinfo *pkt)
 }
 #endif
 
+static noinline bool nft_meta_get_eval_kind(enum nft_meta_keys key,
+					    u32 *dest,
+					    const struct nft_pktinfo *pkt)
+{
+	const struct net_device *in = nft_in(pkt), *out = nft_out(pkt);
+
+	switch (key) {
+	case NFT_META_IIFKIND:
+		if (!in || !in->rtnl_link_ops)
+			return false;
+		strncpy((char *)dest, in->rtnl_link_ops->kind, IFNAMSIZ);
+		break;
+	case NFT_META_OIFKIND:
+		if (!out || !out->rtnl_link_ops)
+			return false;
+		strncpy((char *)dest, out->rtnl_link_ops->kind, IFNAMSIZ);
+		break;
+	default:
+		return false;
+	}
+
+	return true;
+}
+
 void nft_meta_get_eval(const struct nft_expr *expr,
 		       struct nft_regs *regs,
 		       const struct nft_pktinfo *pkt)
@@ -286,14 +310,9 @@ void nft_meta_get_eval(const struct nft_expr *expr,
 		break;
 #endif
 	case NFT_META_IIFKIND:
-		if (in == NULL || in->rtnl_link_ops == NULL)
-			goto err;
-		strncpy((char *)dest, in->rtnl_link_ops->kind, IFNAMSIZ);
-		break;
 	case NFT_META_OIFKIND:
-		if (out == NULL || out->rtnl_link_ops == NULL)
+		if (!nft_meta_get_eval_kind(priv->key, dest, pkt))
 			goto err;
-		strncpy((char *)dest, out->rtnl_link_ops->kind, IFNAMSIZ);
 		break;
 	case NFT_META_TIME_NS:
 	case NFT_META_TIME_DAY:
-- 
cgit v1.2.3-59-g8ed1b


From 8724e819cc9a8fab4c18e791cb0bd602fb294971 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Wed, 18 Dec 2019 12:05:18 +0100
Subject: netfilter: nft_meta: move all interface related keys to helper

Reduces repetiveness and reduces size of meta eval function.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nft_meta.c | 95 +++++++++++++++++++++++++++++++++++-------------
 1 file changed, 70 insertions(+), 25 deletions(-)

diff --git a/net/netfilter/nft_meta.c b/net/netfilter/nft_meta.c
index 2f7cc64b0c15..022f1473ddd1 100644
--- a/net/netfilter/nft_meta.c
+++ b/net/netfilter/nft_meta.c
@@ -199,13 +199,79 @@ static noinline bool nft_meta_get_eval_kind(enum nft_meta_keys key,
 	return true;
 }
 
+static void nft_meta_store_ifindex(u32 *dest, const struct net_device *dev)
+{
+	*dest = dev ? dev->ifindex : 0;
+}
+
+static void nft_meta_store_ifname(u32 *dest, const struct net_device *dev)
+{
+	strncpy((char *)dest, dev ? dev->name : "", IFNAMSIZ);
+}
+
+static bool nft_meta_store_iftype(u32 *dest, const struct net_device *dev)
+{
+	if (!dev)
+		return false;
+
+	nft_reg_store16(dest, dev->type);
+	return true;
+}
+
+static bool nft_meta_store_ifgroup(u32 *dest, const struct net_device *dev)
+{
+	if (!dev)
+		return false;
+
+	*dest = dev->group;
+	return true;
+}
+
+static bool nft_meta_get_eval_ifname(enum nft_meta_keys key, u32 *dest,
+				     const struct nft_pktinfo *pkt)
+{
+	switch (key) {
+	case NFT_META_IIFNAME:
+		nft_meta_store_ifname(dest, nft_in(pkt));
+		break;
+	case NFT_META_OIFNAME:
+		nft_meta_store_ifname(dest, nft_out(pkt));
+		break;
+	case NFT_META_IIF:
+		nft_meta_store_ifindex(dest, nft_in(pkt));
+		break;
+	case NFT_META_OIF:
+		nft_meta_store_ifindex(dest, nft_out(pkt));
+		break;
+	case NFT_META_IIFTYPE:
+		if (!nft_meta_store_iftype(dest, nft_in(pkt)))
+			return false;
+		break;
+	case NFT_META_OIFTYPE:
+		if (!nft_meta_store_iftype(dest, nft_out(pkt)))
+			return false;
+		break;
+	case NFT_META_IIFGROUP:
+		if (!nft_meta_store_ifgroup(dest, nft_out(pkt)))
+			return false;
+		break;
+	case NFT_META_OIFGROUP:
+		if (!nft_meta_store_ifgroup(dest, nft_out(pkt)))
+			return false;
+		break;
+	default:
+		return false;
+	}
+
+	return true;
+}
+
 void nft_meta_get_eval(const struct nft_expr *expr,
 		       struct nft_regs *regs,
 		       const struct nft_pktinfo *pkt)
 {
 	const struct nft_meta *priv = nft_expr_priv(expr);
 	const struct sk_buff *skb = pkt->skb;
-	const struct net_device *in = nft_in(pkt), *out = nft_out(pkt);
 	u32 *dest = &regs->data[priv->dreg];
 
 	switch (priv->key) {
@@ -230,26 +296,15 @@ void nft_meta_get_eval(const struct nft_expr *expr,
 		*dest = skb->mark;
 		break;
 	case NFT_META_IIF:
-		*dest = in ? in->ifindex : 0;
-		break;
 	case NFT_META_OIF:
-		*dest = out ? out->ifindex : 0;
-		break;
 	case NFT_META_IIFNAME:
-		strncpy((char *)dest, in ? in->name : "", IFNAMSIZ);
-		break;
 	case NFT_META_OIFNAME:
-		strncpy((char *)dest, out ? out->name : "", IFNAMSIZ);
-		break;
 	case NFT_META_IIFTYPE:
-		if (in == NULL)
-			goto err;
-		nft_reg_store16(dest, in->type);
-		break;
 	case NFT_META_OIFTYPE:
-		if (out == NULL)
+	case NFT_META_IIFGROUP:
+	case NFT_META_OIFGROUP:
+		if (!nft_meta_get_eval_ifname(priv->key, dest, pkt))
 			goto err;
-		nft_reg_store16(dest, out->type);
 		break;
 	case NFT_META_SKUID:
 	case NFT_META_SKGID:
@@ -283,16 +338,6 @@ void nft_meta_get_eval(const struct nft_expr *expr,
 	case NFT_META_CPU:
 		*dest = raw_smp_processor_id();
 		break;
-	case NFT_META_IIFGROUP:
-		if (in == NULL)
-			goto err;
-		*dest = in->group;
-		break;
-	case NFT_META_OIFGROUP:
-		if (out == NULL)
-			goto err;
-		*dest = out->group;
-		break;
 #ifdef CONFIG_CGROUP_NET_CLASSID
 	case NFT_META_CGROUP:
 		if (!nft_meta_get_eval_cgroup(dest, pkt))
-- 
cgit v1.2.3-59-g8ed1b


From 6b2faee0ca91b63bd5a3b2087d4f25c76b983961 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Wed, 18 Dec 2019 12:05:19 +0100
Subject: netfilter: nft_meta: place prandom handling in a helper

Move this out of the main eval loop, the numgen expression
provides a better alternative to meta random.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nft_meta.c | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

diff --git a/net/netfilter/nft_meta.c b/net/netfilter/nft_meta.c
index 022f1473ddd1..ac6fc95387dc 100644
--- a/net/netfilter/nft_meta.c
+++ b/net/netfilter/nft_meta.c
@@ -266,6 +266,13 @@ static bool nft_meta_get_eval_ifname(enum nft_meta_keys key, u32 *dest,
 	return true;
 }
 
+static noinline u32 nft_prandom_u32(void)
+{
+	struct rnd_state *state = this_cpu_ptr(&nft_prandom_state);
+
+	return prandom_u32_state(state);
+}
+
 void nft_meta_get_eval(const struct nft_expr *expr,
 		       struct nft_regs *regs,
 		       const struct nft_pktinfo *pkt)
@@ -344,11 +351,9 @@ void nft_meta_get_eval(const struct nft_expr *expr,
 			goto err;
 		break;
 #endif
-	case NFT_META_PRANDOM: {
-		struct rnd_state *state = this_cpu_ptr(&nft_prandom_state);
-		*dest = prandom_u32_state(state);
+	case NFT_META_PRANDOM:
+		*dest = nft_prandom_u32();
 		break;
-	}
 #ifdef CONFIG_XFRM
 	case NFT_META_SECPATH:
 		nft_reg_store8(dest, secpath_exists(skb));
-- 
cgit v1.2.3-59-g8ed1b


From 01a0fc82252d82eda50d4e1252b826a3ef7afb3d Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Wed, 18 Dec 2019 12:05:20 +0100
Subject: netfilter: nft_meta: place rtclassid handling in a helper

skb_dst is an inline helper with a WARN_ON(), so this is a bit more code
than it looks like.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nft_meta.c | 22 ++++++++++++++++------
 1 file changed, 16 insertions(+), 6 deletions(-)

diff --git a/net/netfilter/nft_meta.c b/net/netfilter/nft_meta.c
index ac6fc95387dc..fb1a571db924 100644
--- a/net/netfilter/nft_meta.c
+++ b/net/netfilter/nft_meta.c
@@ -273,6 +273,20 @@ static noinline u32 nft_prandom_u32(void)
 	return prandom_u32_state(state);
 }
 
+#ifdef CONFIG_IP_ROUTE_CLASSID
+static noinline bool
+nft_meta_get_eval_rtclassid(const struct sk_buff *skb, u32 *dest)
+{
+	const struct dst_entry *dst = skb_dst(skb);
+
+	if (!dst)
+		return false;
+
+	*dest = dst->tclassid;
+	return true;
+}
+#endif
+
 void nft_meta_get_eval(const struct nft_expr *expr,
 		       struct nft_regs *regs,
 		       const struct nft_pktinfo *pkt)
@@ -319,14 +333,10 @@ void nft_meta_get_eval(const struct nft_expr *expr,
 			goto err;
 		break;
 #ifdef CONFIG_IP_ROUTE_CLASSID
-	case NFT_META_RTCLASSID: {
-		const struct dst_entry *dst = skb_dst(skb);
-
-		if (dst == NULL)
+	case NFT_META_RTCLASSID:
+		if (!nft_meta_get_eval_rtclassid(skb, dest))
 			goto err;
-		*dest = dst->tclassid;
 		break;
-	}
 #endif
 #ifdef CONFIG_NETWORK_SECMARK
 	case NFT_META_SECMARK:
-- 
cgit v1.2.3-59-g8ed1b


From c14ceb0ec727187f71a487a592ffa91767fed66e Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Wed, 18 Dec 2019 12:05:21 +0100
Subject: netfilter: nft_meta: add support for slave device ifindex matching

Allow to match on vrf slave ifindex or name.

In case there was no slave interface involved, store 0 in the
destination register just like existing iif/oif matching.

sdif(name) is restricted to the ipv4/ipv6 input and forward hooks,
as it depends on ip(6) stack parsing/storing info in skb->cb[].

Cc: Martin Willi <martin@strongswan.org>
Cc: David Ahern <dsahern@kernel.org>
Cc: Shrijeet Mukherjee <shrijeet@gmail.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/uapi/linux/netfilter/nf_tables.h |  4 ++
 net/netfilter/nft_meta.c                 | 76 +++++++++++++++++++++++++++++---
 2 files changed, 73 insertions(+), 7 deletions(-)

diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h
index bb9b049310df..e237ecbdcd8a 100644
--- a/include/uapi/linux/netfilter/nf_tables.h
+++ b/include/uapi/linux/netfilter/nf_tables.h
@@ -805,6 +805,8 @@ enum nft_exthdr_attributes {
  * @NFT_META_TIME_NS: time since epoch (in nanoseconds)
  * @NFT_META_TIME_DAY: day of week (from 0 = Sunday to 6 = Saturday)
  * @NFT_META_TIME_HOUR: hour of day (in seconds)
+ * @NFT_META_SDIF: slave device interface index
+ * @NFT_META_SDIFNAME: slave device interface name
  */
 enum nft_meta_keys {
 	NFT_META_LEN,
@@ -840,6 +842,8 @@ enum nft_meta_keys {
 	NFT_META_TIME_NS,
 	NFT_META_TIME_DAY,
 	NFT_META_TIME_HOUR,
+	NFT_META_SDIF,
+	NFT_META_SDIFNAME,
 };
 
 /**
diff --git a/net/netfilter/nft_meta.c b/net/netfilter/nft_meta.c
index fb1a571db924..951b6e87ed5d 100644
--- a/net/netfilter/nft_meta.c
+++ b/net/netfilter/nft_meta.c
@@ -17,6 +17,7 @@
 #include <linux/smp.h>
 #include <linux/static_key.h>
 #include <net/dst.h>
+#include <net/ip.h>
 #include <net/sock.h>
 #include <net/tcp_states.h> /* for TCP_TIME_WAIT */
 #include <net/netfilter/nf_tables.h>
@@ -287,6 +288,28 @@ nft_meta_get_eval_rtclassid(const struct sk_buff *skb, u32 *dest)
 }
 #endif
 
+static noinline u32 nft_meta_get_eval_sdif(const struct nft_pktinfo *pkt)
+{
+	switch (nft_pf(pkt)) {
+	case NFPROTO_IPV4:
+		return inet_sdif(pkt->skb);
+	case NFPROTO_IPV6:
+		return inet6_sdif(pkt->skb);
+	}
+
+	return 0;
+}
+
+static noinline void
+nft_meta_get_eval_sdifname(u32 *dest, const struct nft_pktinfo *pkt)
+{
+	u32 sdif = nft_meta_get_eval_sdif(pkt);
+	const struct net_device *dev;
+
+	dev = sdif ? dev_get_by_index_rcu(nft_net(pkt), sdif) : NULL;
+	nft_meta_store_ifname(dest, dev);
+}
+
 void nft_meta_get_eval(const struct nft_expr *expr,
 		       struct nft_regs *regs,
 		       const struct nft_pktinfo *pkt)
@@ -379,6 +402,12 @@ void nft_meta_get_eval(const struct nft_expr *expr,
 	case NFT_META_TIME_HOUR:
 		nft_meta_get_eval_time(priv->key, dest);
 		break;
+	case NFT_META_SDIF:
+		*dest = nft_meta_get_eval_sdif(pkt);
+		break;
+	case NFT_META_SDIFNAME:
+		nft_meta_get_eval_sdifname(dest, pkt);
+		break;
 	default:
 		WARN_ON(1);
 		goto err;
@@ -459,6 +488,7 @@ int nft_meta_get_init(const struct nft_ctx *ctx,
 	case NFT_META_MARK:
 	case NFT_META_IIF:
 	case NFT_META_OIF:
+	case NFT_META_SDIF:
 	case NFT_META_SKUID:
 	case NFT_META_SKGID:
 #ifdef CONFIG_IP_ROUTE_CLASSID
@@ -480,6 +510,7 @@ int nft_meta_get_init(const struct nft_ctx *ctx,
 	case NFT_META_OIFNAME:
 	case NFT_META_IIFKIND:
 	case NFT_META_OIFKIND:
+	case NFT_META_SDIFNAME:
 		len = IFNAMSIZ;
 		break;
 	case NFT_META_PRANDOM:
@@ -510,16 +541,28 @@ int nft_meta_get_init(const struct nft_ctx *ctx,
 }
 EXPORT_SYMBOL_GPL(nft_meta_get_init);
 
-static int nft_meta_get_validate(const struct nft_ctx *ctx,
-				 const struct nft_expr *expr,
-				 const struct nft_data **data)
+static int nft_meta_get_validate_sdif(const struct nft_ctx *ctx)
 {
-#ifdef CONFIG_XFRM
-	const struct nft_meta *priv = nft_expr_priv(expr);
 	unsigned int hooks;
 
-	if (priv->key != NFT_META_SECPATH)
-		return 0;
+	switch (ctx->family) {
+	case NFPROTO_IPV4:
+	case NFPROTO_IPV6:
+	case NFPROTO_INET:
+		hooks = (1 << NF_INET_LOCAL_IN) |
+			(1 << NF_INET_FORWARD);
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	return nft_chain_validate_hooks(ctx->chain, hooks);
+}
+
+static int nft_meta_get_validate_xfrm(const struct nft_ctx *ctx)
+{
+#ifdef CONFIG_XFRM
+	unsigned int hooks;
 
 	switch (ctx->family) {
 	case NFPROTO_NETDEV:
@@ -542,6 +585,25 @@ static int nft_meta_get_validate(const struct nft_ctx *ctx,
 #endif
 }
 
+static int nft_meta_get_validate(const struct nft_ctx *ctx,
+				 const struct nft_expr *expr,
+				 const struct nft_data **data)
+{
+	const struct nft_meta *priv = nft_expr_priv(expr);
+
+	switch (priv->key) {
+	case NFT_META_SECPATH:
+		return nft_meta_get_validate_xfrm(ctx);
+	case NFT_META_SDIF:
+	case NFT_META_SDIFNAME:
+		return nft_meta_get_validate_sdif(ctx);
+	default:
+		break;
+	}
+
+	return 0;
+}
+
 int nft_meta_set_validate(const struct nft_ctx *ctx,
 			  const struct nft_expr *expr,
 			  const struct nft_data **data)
-- 
cgit v1.2.3-59-g8ed1b


From f643ee295c1c63bc117fb052d4da681354d6f732 Mon Sep 17 00:00:00 2001
From: Kevin Kou <qdkevin.kou@gmail.com>
Date: Thu, 26 Dec 2019 12:29:17 +0000
Subject: sctp: move trace_sctp_probe_path into sctp_outq_sack

The original patch bringed in the "SCTP ACK tracking trace event"
feature was committed at Dec.20, 2017, it replaced jprobe usage
with trace events, and bringed in two trace events, one is
TRACE_EVENT(sctp_probe), another one is TRACE_EVENT(sctp_probe_path).
The original patch intended to trigger the trace_sctp_probe_path in
TRACE_EVENT(sctp_probe) as below code,

+TRACE_EVENT(sctp_probe,
+
+	TP_PROTO(const struct sctp_endpoint *ep,
+		 const struct sctp_association *asoc,
+		 struct sctp_chunk *chunk),
+
+	TP_ARGS(ep, asoc, chunk),
+
+	TP_STRUCT__entry(
+		__field(__u64, asoc)
+		__field(__u32, mark)
+		__field(__u16, bind_port)
+		__field(__u16, peer_port)
+		__field(__u32, pathmtu)
+		__field(__u32, rwnd)
+		__field(__u16, unack_data)
+	),
+
+	TP_fast_assign(
+		struct sk_buff *skb = chunk->skb;
+
+		__entry->asoc = (unsigned long)asoc;
+		__entry->mark = skb->mark;
+		__entry->bind_port = ep->base.bind_addr.port;
+		__entry->peer_port = asoc->peer.port;
+		__entry->pathmtu = asoc->pathmtu;
+		__entry->rwnd = asoc->peer.rwnd;
+		__entry->unack_data = asoc->unack_data;
+
+		if (trace_sctp_probe_path_enabled()) {
+			struct sctp_transport *sp;
+
+			list_for_each_entry(sp, &asoc->peer.transport_addr_list,
+					    transports) {
+				trace_sctp_probe_path(sp, asoc);
+			}
+		}
+	),

But I found it did not work when I did testing, and trace_sctp_probe_path
had no output, I finally found that there is trace buffer lock
operation(trace_event_buffer_reserve) in include/trace/trace_events.h:

static notrace void							\
trace_event_raw_event_##call(void *__data, proto)			\
{									\
	struct trace_event_file *trace_file = __data;			\
	struct trace_event_data_offsets_##call __maybe_unused __data_offsets;\
	struct trace_event_buffer fbuffer;				\
	struct trace_event_raw_##call *entry;				\
	int __data_size;						\
									\
	if (trace_trigger_soft_disabled(trace_file))			\
		return;							\
									\
	__data_size = trace_event_get_offsets_##call(&__data_offsets, args); \
									\
	entry = trace_event_buffer_reserve(&fbuffer, trace_file,	\
				 sizeof(*entry) + __data_size);		\
									\
	if (!entry)							\
		return;							\
									\
	tstruct								\
									\
	{ assign; }							\
									\
	trace_event_buffer_commit(&fbuffer);				\
}

The reason caused no output of trace_sctp_probe_path is that
trace_sctp_probe_path written in TP_fast_assign part of
TRACE_EVENT(sctp_probe), and it will be placed( { assign; } ) after the
trace_event_buffer_reserve() when compiler expands Macro,

        entry = trace_event_buffer_reserve(&fbuffer, trace_file,        \
                                 sizeof(*entry) + __data_size);         \
                                                                        \
        if (!entry)                                                     \
                return;                                                 \
                                                                        \
        tstruct                                                         \
                                                                        \
        { assign; }                                                     \

so trace_sctp_probe_path finally can not acquire trace_event_buffer
and return no output, that is to say the nest of tracepoint entry function
is not allowed. The function call flow is:

trace_sctp_probe()
-> trace_event_raw_event_sctp_probe()
 -> lock buffer
 -> trace_sctp_probe_path()
   -> trace_event_raw_event_sctp_probe_path()  --nested
   -> buffer has been locked and return no output.

This patch is to remove trace_sctp_probe_path from the TP_fast_assign
part of TRACE_EVENT(sctp_probe) to avoid the nest of entry function,
and trigger sctp_probe_path_trace in sctp_outq_sack.

After this patch, you can enable both events individually,
  # cd /sys/kernel/debug/tracing
  # echo 1 > events/sctp/sctp_probe/enable
  # echo 1 > events/sctp/sctp_probe_path/enable

Or, you can enable all the events under sctp.

  # echo 1 > events/sctp/enable

Signed-off-by: Kevin Kou <qdkevin.kou@gmail.com>
Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/trace/events/sctp.h | 9 ---------
 net/sctp/outqueue.c         | 6 ++++++
 2 files changed, 6 insertions(+), 9 deletions(-)

diff --git a/include/trace/events/sctp.h b/include/trace/events/sctp.h
index 7475c7be165a..d4aac3436595 100644
--- a/include/trace/events/sctp.h
+++ b/include/trace/events/sctp.h
@@ -75,15 +75,6 @@ TRACE_EVENT(sctp_probe,
 		__entry->pathmtu = asoc->pathmtu;
 		__entry->rwnd = asoc->peer.rwnd;
 		__entry->unack_data = asoc->unack_data;
-
-		if (trace_sctp_probe_path_enabled()) {
-			struct sctp_transport *sp;
-
-			list_for_each_entry(sp, &asoc->peer.transport_addr_list,
-					    transports) {
-				trace_sctp_probe_path(sp, asoc);
-			}
-		}
 	),
 
 	TP_printk("asoc=%#llx mark=%#x bind_port=%d peer_port=%d pathmtu=%d "
diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c
index a031d1134c60..6b0b3bad4daa 100644
--- a/net/sctp/outqueue.c
+++ b/net/sctp/outqueue.c
@@ -36,6 +36,7 @@
 #include <net/sctp/sctp.h>
 #include <net/sctp/sm.h>
 #include <net/sctp/stream_sched.h>
+#include <trace/events/sctp.h>
 
 /* Declare internal functions here.  */
 static int sctp_acked(struct sctp_sackhdr *sack, __u32 tsn);
@@ -1238,6 +1239,11 @@ int sctp_outq_sack(struct sctp_outq *q, struct sctp_chunk *chunk)
 	/* Grab the association's destination address list. */
 	transport_list = &asoc->peer.transport_addr_list;
 
+	/* SCTP path tracepoint for congestion control debugging. */
+	list_for_each_entry(transport, transport_list, transports) {
+		trace_sctp_probe_path(transport, asoc);
+	}
+
 	sack_ctsn = ntohl(sack->cum_tsn_ack);
 	gap_ack_blocks = ntohs(sack->num_gap_ack_blocks);
 	asoc->stats.gapcnt += gap_ack_blocks;
-- 
cgit v1.2.3-59-g8ed1b


From c1e469902640ed0a82d0b8df5951199c17a49e3c Mon Sep 17 00:00:00 2001
From: Andy Roulin <aroulin@cumulusnetworks.com>
Date: Thu, 26 Dec 2019 05:41:57 -0800
Subject: bonding: rename AD_STATE_* to LACP_STATE_*

As the LACP actor/partner state is now part of the uapi, rename the
3ad state defines with LACP prefix. The LACP prefix is preferred over
BOND_3AD as the LACP standard moved to 802.1AX.

Fixes: 826f66b30c2e3 ("bonding: move 802.3ad port state flags to uapi")
Signed-off-by: Andy Roulin <aroulin@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/bonding/bond_3ad.c  | 112 ++++++++++++++++++++--------------------
 include/uapi/linux/if_bonding.h |  16 +++---
 2 files changed, 64 insertions(+), 64 deletions(-)

diff --git a/drivers/net/bonding/bond_3ad.c b/drivers/net/bonding/bond_3ad.c
index 34bfe99641a3..31e43a2197a3 100644
--- a/drivers/net/bonding/bond_3ad.c
+++ b/drivers/net/bonding/bond_3ad.c
@@ -447,8 +447,8 @@ static void __choose_matched(struct lacpdu *lacpdu, struct port *port)
 	     MAC_ADDRESS_EQUAL(&(lacpdu->partner_system), &(port->actor_system)) &&
 	     (ntohs(lacpdu->partner_system_priority) == port->actor_system_priority) &&
 	     (ntohs(lacpdu->partner_key) == port->actor_oper_port_key) &&
-	     ((lacpdu->partner_state & AD_STATE_AGGREGATION) == (port->actor_oper_port_state & AD_STATE_AGGREGATION))) ||
-	    ((lacpdu->actor_state & AD_STATE_AGGREGATION) == 0)
+	     ((lacpdu->partner_state & LACP_STATE_AGGREGATION) == (port->actor_oper_port_state & LACP_STATE_AGGREGATION))) ||
+	    ((lacpdu->actor_state & LACP_STATE_AGGREGATION) == 0)
 		) {
 		port->sm_vars |= AD_PORT_MATCHED;
 	} else {
@@ -482,18 +482,18 @@ static void __record_pdu(struct lacpdu *lacpdu, struct port *port)
 		partner->port_state = lacpdu->actor_state;
 
 		/* set actor_oper_port_state.defaulted to FALSE */
-		port->actor_oper_port_state &= ~AD_STATE_DEFAULTED;
+		port->actor_oper_port_state &= ~LACP_STATE_DEFAULTED;
 
 		/* set the partner sync. to on if the partner is sync,
 		 * and the port is matched
 		 */
 		if ((port->sm_vars & AD_PORT_MATCHED) &&
-		    (lacpdu->actor_state & AD_STATE_SYNCHRONIZATION)) {
-			partner->port_state |= AD_STATE_SYNCHRONIZATION;
+		    (lacpdu->actor_state & LACP_STATE_SYNCHRONIZATION)) {
+			partner->port_state |= LACP_STATE_SYNCHRONIZATION;
 			slave_dbg(port->slave->bond->dev, port->slave->dev,
 				  "partner sync=1\n");
 		} else {
-			partner->port_state &= ~AD_STATE_SYNCHRONIZATION;
+			partner->port_state &= ~LACP_STATE_SYNCHRONIZATION;
 			slave_dbg(port->slave->bond->dev, port->slave->dev,
 				  "partner sync=0\n");
 		}
@@ -516,7 +516,7 @@ static void __record_default(struct port *port)
 		       sizeof(struct port_params));
 
 		/* set actor_oper_port_state.defaulted to true */
-		port->actor_oper_port_state |= AD_STATE_DEFAULTED;
+		port->actor_oper_port_state |= LACP_STATE_DEFAULTED;
 	}
 }
 
@@ -546,7 +546,7 @@ static void __update_selected(struct lacpdu *lacpdu, struct port *port)
 		    !MAC_ADDRESS_EQUAL(&lacpdu->actor_system, &partner->system) ||
 		    ntohs(lacpdu->actor_system_priority) != partner->system_priority ||
 		    ntohs(lacpdu->actor_key) != partner->key ||
-		    (lacpdu->actor_state & AD_STATE_AGGREGATION) != (partner->port_state & AD_STATE_AGGREGATION)) {
+		    (lacpdu->actor_state & LACP_STATE_AGGREGATION) != (partner->port_state & LACP_STATE_AGGREGATION)) {
 			port->sm_vars &= ~AD_PORT_SELECTED;
 		}
 	}
@@ -578,8 +578,8 @@ static void __update_default_selected(struct port *port)
 		    !MAC_ADDRESS_EQUAL(&admin->system, &oper->system) ||
 		    admin->system_priority != oper->system_priority ||
 		    admin->key != oper->key ||
-		    (admin->port_state & AD_STATE_AGGREGATION)
-			!= (oper->port_state & AD_STATE_AGGREGATION)) {
+		    (admin->port_state & LACP_STATE_AGGREGATION)
+			!= (oper->port_state & LACP_STATE_AGGREGATION)) {
 			port->sm_vars &= ~AD_PORT_SELECTED;
 		}
 	}
@@ -609,10 +609,10 @@ static void __update_ntt(struct lacpdu *lacpdu, struct port *port)
 		    !MAC_ADDRESS_EQUAL(&(lacpdu->partner_system), &(port->actor_system)) ||
 		    (ntohs(lacpdu->partner_system_priority) != port->actor_system_priority) ||
 		    (ntohs(lacpdu->partner_key) != port->actor_oper_port_key) ||
-		    ((lacpdu->partner_state & AD_STATE_LACP_ACTIVITY) != (port->actor_oper_port_state & AD_STATE_LACP_ACTIVITY)) ||
-		    ((lacpdu->partner_state & AD_STATE_LACP_TIMEOUT) != (port->actor_oper_port_state & AD_STATE_LACP_TIMEOUT)) ||
-		    ((lacpdu->partner_state & AD_STATE_SYNCHRONIZATION) != (port->actor_oper_port_state & AD_STATE_SYNCHRONIZATION)) ||
-		    ((lacpdu->partner_state & AD_STATE_AGGREGATION) != (port->actor_oper_port_state & AD_STATE_AGGREGATION))
+		    ((lacpdu->partner_state & LACP_STATE_LACP_ACTIVITY) != (port->actor_oper_port_state & LACP_STATE_LACP_ACTIVITY)) ||
+		    ((lacpdu->partner_state & LACP_STATE_LACP_TIMEOUT) != (port->actor_oper_port_state & LACP_STATE_LACP_TIMEOUT)) ||
+		    ((lacpdu->partner_state & LACP_STATE_SYNCHRONIZATION) != (port->actor_oper_port_state & LACP_STATE_SYNCHRONIZATION)) ||
+		    ((lacpdu->partner_state & LACP_STATE_AGGREGATION) != (port->actor_oper_port_state & LACP_STATE_AGGREGATION))
 		   ) {
 			port->ntt = true;
 		}
@@ -968,7 +968,7 @@ static void ad_mux_machine(struct port *port, bool *update_slave_arr)
 			 * edable port will take place only after this timer)
 			 */
 			if ((port->sm_vars & AD_PORT_SELECTED) &&
-			    (port->partner_oper.port_state & AD_STATE_SYNCHRONIZATION) &&
+			    (port->partner_oper.port_state & LACP_STATE_SYNCHRONIZATION) &&
 			    !__check_agg_selection_timer(port)) {
 				if (port->aggregator->is_active)
 					port->sm_mux_state =
@@ -986,14 +986,14 @@ static void ad_mux_machine(struct port *port, bool *update_slave_arr)
 				port->sm_mux_state = AD_MUX_DETACHED;
 			} else if (port->aggregator->is_active) {
 				port->actor_oper_port_state |=
-				    AD_STATE_SYNCHRONIZATION;
+				    LACP_STATE_SYNCHRONIZATION;
 			}
 			break;
 		case AD_MUX_COLLECTING_DISTRIBUTING:
 			if (!(port->sm_vars & AD_PORT_SELECTED) ||
 			    (port->sm_vars & AD_PORT_STANDBY) ||
-			    !(port->partner_oper.port_state & AD_STATE_SYNCHRONIZATION) ||
-			    !(port->actor_oper_port_state & AD_STATE_SYNCHRONIZATION)) {
+			    !(port->partner_oper.port_state & LACP_STATE_SYNCHRONIZATION) ||
+			    !(port->actor_oper_port_state & LACP_STATE_SYNCHRONIZATION)) {
 				port->sm_mux_state = AD_MUX_ATTACHED;
 			} else {
 				/* if port state hasn't changed make
@@ -1022,11 +1022,11 @@ static void ad_mux_machine(struct port *port, bool *update_slave_arr)
 			  port->sm_mux_state);
 		switch (port->sm_mux_state) {
 		case AD_MUX_DETACHED:
-			port->actor_oper_port_state &= ~AD_STATE_SYNCHRONIZATION;
+			port->actor_oper_port_state &= ~LACP_STATE_SYNCHRONIZATION;
 			ad_disable_collecting_distributing(port,
 							   update_slave_arr);
-			port->actor_oper_port_state &= ~AD_STATE_COLLECTING;
-			port->actor_oper_port_state &= ~AD_STATE_DISTRIBUTING;
+			port->actor_oper_port_state &= ~LACP_STATE_COLLECTING;
+			port->actor_oper_port_state &= ~LACP_STATE_DISTRIBUTING;
 			port->ntt = true;
 			break;
 		case AD_MUX_WAITING:
@@ -1035,20 +1035,20 @@ static void ad_mux_machine(struct port *port, bool *update_slave_arr)
 		case AD_MUX_ATTACHED:
 			if (port->aggregator->is_active)
 				port->actor_oper_port_state |=
-				    AD_STATE_SYNCHRONIZATION;
+				    LACP_STATE_SYNCHRONIZATION;
 			else
 				port->actor_oper_port_state &=
-				    ~AD_STATE_SYNCHRONIZATION;
-			port->actor_oper_port_state &= ~AD_STATE_COLLECTING;
-			port->actor_oper_port_state &= ~AD_STATE_DISTRIBUTING;
+				    ~LACP_STATE_SYNCHRONIZATION;
+			port->actor_oper_port_state &= ~LACP_STATE_COLLECTING;
+			port->actor_oper_port_state &= ~LACP_STATE_DISTRIBUTING;
 			ad_disable_collecting_distributing(port,
 							   update_slave_arr);
 			port->ntt = true;
 			break;
 		case AD_MUX_COLLECTING_DISTRIBUTING:
-			port->actor_oper_port_state |= AD_STATE_COLLECTING;
-			port->actor_oper_port_state |= AD_STATE_DISTRIBUTING;
-			port->actor_oper_port_state |= AD_STATE_SYNCHRONIZATION;
+			port->actor_oper_port_state |= LACP_STATE_COLLECTING;
+			port->actor_oper_port_state |= LACP_STATE_DISTRIBUTING;
+			port->actor_oper_port_state |= LACP_STATE_SYNCHRONIZATION;
 			ad_enable_collecting_distributing(port,
 							  update_slave_arr);
 			port->ntt = true;
@@ -1146,7 +1146,7 @@ static void ad_rx_machine(struct lacpdu *lacpdu, struct port *port)
 				port->sm_vars |= AD_PORT_LACP_ENABLED;
 			port->sm_vars &= ~AD_PORT_SELECTED;
 			__record_default(port);
-			port->actor_oper_port_state &= ~AD_STATE_EXPIRED;
+			port->actor_oper_port_state &= ~LACP_STATE_EXPIRED;
 			port->sm_rx_state = AD_RX_PORT_DISABLED;
 
 			/* Fall Through */
@@ -1156,9 +1156,9 @@ static void ad_rx_machine(struct lacpdu *lacpdu, struct port *port)
 		case AD_RX_LACP_DISABLED:
 			port->sm_vars &= ~AD_PORT_SELECTED;
 			__record_default(port);
-			port->partner_oper.port_state &= ~AD_STATE_AGGREGATION;
+			port->partner_oper.port_state &= ~LACP_STATE_AGGREGATION;
 			port->sm_vars |= AD_PORT_MATCHED;
-			port->actor_oper_port_state &= ~AD_STATE_EXPIRED;
+			port->actor_oper_port_state &= ~LACP_STATE_EXPIRED;
 			break;
 		case AD_RX_EXPIRED:
 			/* Reset of the Synchronization flag (Standard 43.4.12)
@@ -1167,19 +1167,19 @@ static void ad_rx_machine(struct lacpdu *lacpdu, struct port *port)
 			 * case of EXPIRED even if LINK_DOWN didn't arrive for
 			 * the port.
 			 */
-			port->partner_oper.port_state &= ~AD_STATE_SYNCHRONIZATION;
+			port->partner_oper.port_state &= ~LACP_STATE_SYNCHRONIZATION;
 			port->sm_vars &= ~AD_PORT_MATCHED;
-			port->partner_oper.port_state |= AD_STATE_LACP_TIMEOUT;
-			port->partner_oper.port_state |= AD_STATE_LACP_ACTIVITY;
+			port->partner_oper.port_state |= LACP_STATE_LACP_TIMEOUT;
+			port->partner_oper.port_state |= LACP_STATE_LACP_ACTIVITY;
 			port->sm_rx_timer_counter = __ad_timer_to_ticks(AD_CURRENT_WHILE_TIMER, (u16)(AD_SHORT_TIMEOUT));
-			port->actor_oper_port_state |= AD_STATE_EXPIRED;
+			port->actor_oper_port_state |= LACP_STATE_EXPIRED;
 			port->sm_vars |= AD_PORT_CHURNED;
 			break;
 		case AD_RX_DEFAULTED:
 			__update_default_selected(port);
 			__record_default(port);
 			port->sm_vars |= AD_PORT_MATCHED;
-			port->actor_oper_port_state &= ~AD_STATE_EXPIRED;
+			port->actor_oper_port_state &= ~LACP_STATE_EXPIRED;
 			break;
 		case AD_RX_CURRENT:
 			/* detect loopback situation */
@@ -1192,8 +1192,8 @@ static void ad_rx_machine(struct lacpdu *lacpdu, struct port *port)
 			__update_selected(lacpdu, port);
 			__update_ntt(lacpdu, port);
 			__record_pdu(lacpdu, port);
-			port->sm_rx_timer_counter = __ad_timer_to_ticks(AD_CURRENT_WHILE_TIMER, (u16)(port->actor_oper_port_state & AD_STATE_LACP_TIMEOUT));
-			port->actor_oper_port_state &= ~AD_STATE_EXPIRED;
+			port->sm_rx_timer_counter = __ad_timer_to_ticks(AD_CURRENT_WHILE_TIMER, (u16)(port->actor_oper_port_state & LACP_STATE_LACP_TIMEOUT));
+			port->actor_oper_port_state &= ~LACP_STATE_EXPIRED;
 			break;
 		default:
 			break;
@@ -1221,7 +1221,7 @@ static void ad_churn_machine(struct port *port)
 	if (port->sm_churn_actor_timer_counter &&
 	    !(--port->sm_churn_actor_timer_counter) &&
 	    port->sm_churn_actor_state == AD_CHURN_MONITOR) {
-		if (port->actor_oper_port_state & AD_STATE_SYNCHRONIZATION) {
+		if (port->actor_oper_port_state & LACP_STATE_SYNCHRONIZATION) {
 			port->sm_churn_actor_state = AD_NO_CHURN;
 		} else {
 			port->churn_actor_count++;
@@ -1231,7 +1231,7 @@ static void ad_churn_machine(struct port *port)
 	if (port->sm_churn_partner_timer_counter &&
 	    !(--port->sm_churn_partner_timer_counter) &&
 	    port->sm_churn_partner_state == AD_CHURN_MONITOR) {
-		if (port->partner_oper.port_state & AD_STATE_SYNCHRONIZATION) {
+		if (port->partner_oper.port_state & LACP_STATE_SYNCHRONIZATION) {
 			port->sm_churn_partner_state = AD_NO_CHURN;
 		} else {
 			port->churn_partner_count++;
@@ -1288,7 +1288,7 @@ static void ad_periodic_machine(struct port *port)
 
 	/* check if port was reinitialized */
 	if (((port->sm_vars & AD_PORT_BEGIN) || !(port->sm_vars & AD_PORT_LACP_ENABLED) || !port->is_enabled) ||
-	    (!(port->actor_oper_port_state & AD_STATE_LACP_ACTIVITY) && !(port->partner_oper.port_state & AD_STATE_LACP_ACTIVITY))
+	    (!(port->actor_oper_port_state & LACP_STATE_LACP_ACTIVITY) && !(port->partner_oper.port_state & LACP_STATE_LACP_ACTIVITY))
 	   ) {
 		port->sm_periodic_state = AD_NO_PERIODIC;
 	}
@@ -1305,11 +1305,11 @@ static void ad_periodic_machine(struct port *port)
 			switch (port->sm_periodic_state) {
 			case AD_FAST_PERIODIC:
 				if (!(port->partner_oper.port_state
-				      & AD_STATE_LACP_TIMEOUT))
+				      & LACP_STATE_LACP_TIMEOUT))
 					port->sm_periodic_state = AD_SLOW_PERIODIC;
 				break;
 			case AD_SLOW_PERIODIC:
-				if ((port->partner_oper.port_state & AD_STATE_LACP_TIMEOUT)) {
+				if ((port->partner_oper.port_state & LACP_STATE_LACP_TIMEOUT)) {
 					port->sm_periodic_timer_counter = 0;
 					port->sm_periodic_state = AD_PERIODIC_TX;
 				}
@@ -1325,7 +1325,7 @@ static void ad_periodic_machine(struct port *port)
 			break;
 		case AD_PERIODIC_TX:
 			if (!(port->partner_oper.port_state &
-			    AD_STATE_LACP_TIMEOUT))
+			    LACP_STATE_LACP_TIMEOUT))
 				port->sm_periodic_state = AD_SLOW_PERIODIC;
 			else
 				port->sm_periodic_state = AD_FAST_PERIODIC;
@@ -1532,7 +1532,7 @@ static void ad_port_selection_logic(struct port *port, bool *update_slave_arr)
 	ad_agg_selection_logic(aggregator, update_slave_arr);
 
 	if (!port->aggregator->is_active)
-		port->actor_oper_port_state &= ~AD_STATE_SYNCHRONIZATION;
+		port->actor_oper_port_state &= ~LACP_STATE_SYNCHRONIZATION;
 }
 
 /* Decide if "agg" is a better choice for the new active aggregator that
@@ -1838,13 +1838,13 @@ static void ad_initialize_port(struct port *port, int lacp_fast)
 		port->actor_port_priority = 0xff;
 		port->actor_port_aggregator_identifier = 0;
 		port->ntt = false;
-		port->actor_admin_port_state = AD_STATE_AGGREGATION |
-					       AD_STATE_LACP_ACTIVITY;
-		port->actor_oper_port_state  = AD_STATE_AGGREGATION |
-					       AD_STATE_LACP_ACTIVITY;
+		port->actor_admin_port_state = LACP_STATE_AGGREGATION |
+					       LACP_STATE_LACP_ACTIVITY;
+		port->actor_oper_port_state  = LACP_STATE_AGGREGATION |
+					       LACP_STATE_LACP_ACTIVITY;
 
 		if (lacp_fast)
-			port->actor_oper_port_state |= AD_STATE_LACP_TIMEOUT;
+			port->actor_oper_port_state |= LACP_STATE_LACP_TIMEOUT;
 
 		memcpy(&port->partner_admin, &tmpl, sizeof(tmpl));
 		memcpy(&port->partner_oper, &tmpl, sizeof(tmpl));
@@ -2095,10 +2095,10 @@ void bond_3ad_unbind_slave(struct slave *slave)
 		  aggregator->aggregator_identifier);
 
 	/* Tell the partner that this port is not suitable for aggregation */
-	port->actor_oper_port_state &= ~AD_STATE_SYNCHRONIZATION;
-	port->actor_oper_port_state &= ~AD_STATE_COLLECTING;
-	port->actor_oper_port_state &= ~AD_STATE_DISTRIBUTING;
-	port->actor_oper_port_state &= ~AD_STATE_AGGREGATION;
+	port->actor_oper_port_state &= ~LACP_STATE_SYNCHRONIZATION;
+	port->actor_oper_port_state &= ~LACP_STATE_COLLECTING;
+	port->actor_oper_port_state &= ~LACP_STATE_DISTRIBUTING;
+	port->actor_oper_port_state &= ~LACP_STATE_AGGREGATION;
 	__update_lacpdu_from_port(port);
 	ad_lacpdu_send(port);
 
@@ -2685,9 +2685,9 @@ void bond_3ad_update_lacp_rate(struct bonding *bond)
 	bond_for_each_slave(bond, slave, iter) {
 		port = &(SLAVE_AD_INFO(slave)->port);
 		if (lacp_fast)
-			port->actor_oper_port_state |= AD_STATE_LACP_TIMEOUT;
+			port->actor_oper_port_state |= LACP_STATE_LACP_TIMEOUT;
 		else
-			port->actor_oper_port_state &= ~AD_STATE_LACP_TIMEOUT;
+			port->actor_oper_port_state &= ~LACP_STATE_LACP_TIMEOUT;
 	}
 	spin_unlock_bh(&bond->mode_lock);
 }
diff --git a/include/uapi/linux/if_bonding.h b/include/uapi/linux/if_bonding.h
index 6829213a54c5..45f3750aa861 100644
--- a/include/uapi/linux/if_bonding.h
+++ b/include/uapi/linux/if_bonding.h
@@ -96,14 +96,14 @@
 #define BOND_XMIT_POLICY_ENCAP34	4 /* encapsulated layer 3+4 */
 
 /* 802.3ad port state definitions (43.4.2.2 in the 802.3ad standard) */
-#define AD_STATE_LACP_ACTIVITY   0x1
-#define AD_STATE_LACP_TIMEOUT    0x2
-#define AD_STATE_AGGREGATION     0x4
-#define AD_STATE_SYNCHRONIZATION 0x8
-#define AD_STATE_COLLECTING      0x10
-#define AD_STATE_DISTRIBUTING    0x20
-#define AD_STATE_DEFAULTED       0x40
-#define AD_STATE_EXPIRED         0x80
+#define LACP_STATE_LACP_ACTIVITY   0x1
+#define LACP_STATE_LACP_TIMEOUT    0x2
+#define LACP_STATE_AGGREGATION     0x4
+#define LACP_STATE_SYNCHRONIZATION 0x8
+#define LACP_STATE_COLLECTING      0x10
+#define LACP_STATE_DISTRIBUTING    0x20
+#define LACP_STATE_DEFAULTED       0x40
+#define LACP_STATE_EXPIRED         0x80
 
 typedef struct ifbond {
 	__s32 bond_mode;
-- 
cgit v1.2.3-59-g8ed1b


From 231c8d2bbc4feac4c5db6afbda2484f24661736b Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Thu, 26 Dec 2019 18:41:13 +0200
Subject: mlxsw: spectrum_router: Remove unnecessary checks

Now that the networking stack takes care of only notifying the routes of
interest, we do not need to maintain a list of identical routes.

Remove the check that tests if the route is the first route in the FIB
node.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c | 16 +---------------
 1 file changed, 1 insertion(+), 15 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
index f62e8d67348c..87a010cb43b1 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
@@ -3231,10 +3231,6 @@ mlxsw_sp_nexthop_group_update(struct mlxsw_sp *mlxsw_sp,
 	return 0;
 }
 
-static bool
-mlxsw_sp_fib_node_entry_is_first(const struct mlxsw_sp_fib_node *fib_node,
-				 const struct mlxsw_sp_fib_entry *fib_entry);
-
 static int
 mlxsw_sp_nexthop_fib_entries_update(struct mlxsw_sp *mlxsw_sp,
 				    struct mlxsw_sp_nexthop_group *nh_grp)
@@ -3243,9 +3239,6 @@ mlxsw_sp_nexthop_fib_entries_update(struct mlxsw_sp *mlxsw_sp,
 	int err;
 
 	list_for_each_entry(fib_entry, &nh_grp->fib_list, nexthop_group_node) {
-		if (!mlxsw_sp_fib_node_entry_is_first(fib_entry->fib_node,
-						      fib_entry))
-			continue;
 		err = mlxsw_sp_fib_entry_update(mlxsw_sp, fib_entry);
 		if (err)
 			return err;
@@ -3263,12 +3256,8 @@ mlxsw_sp_nexthop_fib_entries_refresh(struct mlxsw_sp_nexthop_group *nh_grp)
 	enum mlxsw_reg_ralue_op op = MLXSW_REG_RALUE_OP_WRITE_WRITE;
 	struct mlxsw_sp_fib_entry *fib_entry;
 
-	list_for_each_entry(fib_entry, &nh_grp->fib_list, nexthop_group_node) {
-		if (!mlxsw_sp_fib_node_entry_is_first(fib_entry->fib_node,
-						      fib_entry))
-			continue;
+	list_for_each_entry(fib_entry, &nh_grp->fib_list, nexthop_group_node)
 		mlxsw_sp_fib_entry_offload_refresh(fib_entry, op, 0);
-	}
 }
 
 static void mlxsw_sp_adj_grp_size_round_up(u16 *p_adj_grp_size)
@@ -4785,9 +4774,6 @@ static int mlxsw_sp_fib_node_entry_add(struct mlxsw_sp *mlxsw_sp,
 {
 	struct mlxsw_sp_fib_node *fib_node = fib_entry->fib_node;
 
-	if (!mlxsw_sp_fib_node_entry_is_first(fib_node, fib_entry))
-		return 0;
-
 	/* To prevent packet loss, overwrite the previously offloaded
 	 * entry.
 	 */
-- 
cgit v1.2.3-59-g8ed1b


From 0d2fb5aa93a634316aa5b061170e9732a84d1217 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Thu, 26 Dec 2019 18:41:14 +0200
Subject: mlxsw: spectrum_router: Eliminate dead code

Since the driver no longer maintains a list of identical routes there is
no route to promote when a route is deleted.

Remove that code that took care of it.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c | 10 ----------
 1 file changed, 10 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
index 87a010cb43b1..0439b2399a53 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
@@ -4795,16 +4795,6 @@ static void mlxsw_sp_fib_node_entry_del(struct mlxsw_sp *mlxsw_sp,
 	if (!mlxsw_sp_fib_node_entry_is_first(fib_node, fib_entry))
 		return;
 
-	/* Promote the next entry by overwriting the deleted entry */
-	if (!list_is_singular(&fib_node->entry_list)) {
-		struct mlxsw_sp_fib_entry *n = list_next_entry(fib_entry, list);
-		enum mlxsw_reg_ralue_op op = MLXSW_REG_RALUE_OP_WRITE_DELETE;
-
-		mlxsw_sp_fib_entry_update(mlxsw_sp, n);
-		mlxsw_sp_fib_entry_offload_refresh(fib_entry, op, 0);
-		return;
-	}
-
 	mlxsw_sp_fib_entry_del(mlxsw_sp, fib_entry);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 0705297e511a334e99edf7da0d9a79f4d6b1c028 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Thu, 26 Dec 2019 18:41:15 +0200
Subject: mlxsw: spectrum_router: Make route creation and destruction symmetric

Host routes that perform decapsulation of IP in IP tunnels have a
special adjacency entry linked to them. This entry stores information
such as the expected underlay source IP. When the route is deleted this
entry needs to be freed.

The allocation of the adjacency entry happens in
mlxsw_sp_fib4_entry_type_set(), but it is freed in
mlxsw_sp_fib4_node_entry_unlink().

Create a new function - mlxsw_sp_fib4_entry_type_unset() - and free the
adjacency entry there.

This will allow us to consolidate mlxsw_sp_fib{4,6}_node_entry_unlink()
in the next patch.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c | 18 +++++++++++++++---
 1 file changed, 15 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
index 0439b2399a53..940832761d1b 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
@@ -4480,6 +4480,19 @@ mlxsw_sp_fib4_entry_type_set(struct mlxsw_sp *mlxsw_sp,
 	}
 }
 
+static void
+mlxsw_sp_fib4_entry_type_unset(struct mlxsw_sp *mlxsw_sp,
+			       struct mlxsw_sp_fib_entry *fib_entry)
+{
+	switch (fib_entry->type) {
+	case MLXSW_SP_FIB_ENTRY_TYPE_IPIP_DECAP:
+		mlxsw_sp_fib_entry_decap_fini(mlxsw_sp, fib_entry);
+		break;
+	default:
+		break;
+	}
+}
+
 static struct mlxsw_sp_fib4_entry *
 mlxsw_sp_fib4_entry_create(struct mlxsw_sp *mlxsw_sp,
 			   struct mlxsw_sp_fib_node *fib_node,
@@ -4512,6 +4525,7 @@ mlxsw_sp_fib4_entry_create(struct mlxsw_sp *mlxsw_sp,
 	return fib4_entry;
 
 err_nexthop4_group_get:
+	mlxsw_sp_fib4_entry_type_unset(mlxsw_sp, fib_entry);
 err_fib4_entry_type_set:
 	kfree(fib4_entry);
 	return ERR_PTR(err);
@@ -4521,6 +4535,7 @@ static void mlxsw_sp_fib4_entry_destroy(struct mlxsw_sp *mlxsw_sp,
 					struct mlxsw_sp_fib4_entry *fib4_entry)
 {
 	mlxsw_sp_nexthop4_group_put(mlxsw_sp, &fib4_entry->common);
+	mlxsw_sp_fib4_entry_type_unset(mlxsw_sp, &fib4_entry->common);
 	kfree(fib4_entry);
 }
 
@@ -4823,9 +4838,6 @@ mlxsw_sp_fib4_node_entry_unlink(struct mlxsw_sp *mlxsw_sp,
 {
 	mlxsw_sp_fib_node_entry_del(mlxsw_sp, &fib4_entry->common);
 	list_del(&fib4_entry->common.list);
-
-	if (fib4_entry->common.type == MLXSW_SP_FIB_ENTRY_TYPE_IPIP_DECAP)
-		mlxsw_sp_fib_entry_decap_fini(mlxsw_sp, &fib4_entry->common);
 }
 
 static void mlxsw_sp_fib4_entry_replace(struct mlxsw_sp *mlxsw_sp,
-- 
cgit v1.2.3-59-g8ed1b


From b04720aee9a61b0b2f03a81d575fdcfb84c43c34 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Thu, 26 Dec 2019 18:41:16 +0200
Subject: mlxsw: spectrum_router: Consolidate identical functions

After the last patch mlxsw_sp_fib{4,6}_node_entry_link() and
mlxsw_sp_fib{4,6}_node_entry_unlink() are identical and can therefore be
consolidated into the same common function.

Perform the consolidation.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../net/ethernet/mellanox/mlxsw/spectrum_router.c  | 71 +++++++---------------
 1 file changed, 22 insertions(+), 49 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
index 940832761d1b..f332c55fc83e 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
@@ -4813,31 +4813,31 @@ static void mlxsw_sp_fib_node_entry_del(struct mlxsw_sp *mlxsw_sp,
 	mlxsw_sp_fib_entry_del(mlxsw_sp, fib_entry);
 }
 
-static int mlxsw_sp_fib4_node_entry_link(struct mlxsw_sp *mlxsw_sp,
-					 struct mlxsw_sp_fib4_entry *fib4_entry)
+static int mlxsw_sp_fib_node_entry_link(struct mlxsw_sp *mlxsw_sp,
+					struct mlxsw_sp_fib_entry *fib_entry)
 {
-	struct mlxsw_sp_fib_node *fib_node = fib4_entry->common.fib_node;
+	struct mlxsw_sp_fib_node *fib_node = fib_entry->fib_node;
 	int err;
 
-	list_add(&fib4_entry->common.list, &fib_node->entry_list);
+	list_add(&fib_entry->list, &fib_node->entry_list);
 
-	err = mlxsw_sp_fib_node_entry_add(mlxsw_sp, &fib4_entry->common);
+	err = mlxsw_sp_fib_node_entry_add(mlxsw_sp, fib_entry);
 	if (err)
 		goto err_fib_node_entry_add;
 
 	return 0;
 
 err_fib_node_entry_add:
-	list_del(&fib4_entry->common.list);
+	list_del(&fib_entry->list);
 	return err;
 }
 
 static void
-mlxsw_sp_fib4_node_entry_unlink(struct mlxsw_sp *mlxsw_sp,
-				struct mlxsw_sp_fib4_entry *fib4_entry)
+mlxsw_sp_fib_node_entry_unlink(struct mlxsw_sp *mlxsw_sp,
+			       struct mlxsw_sp_fib_entry *fib_entry)
 {
-	mlxsw_sp_fib_node_entry_del(mlxsw_sp, &fib4_entry->common);
-	list_del(&fib4_entry->common.list);
+	mlxsw_sp_fib_node_entry_del(mlxsw_sp, fib_entry);
+	list_del(&fib_entry->list);
 }
 
 static void mlxsw_sp_fib4_entry_replace(struct mlxsw_sp *mlxsw_sp,
@@ -4852,7 +4852,7 @@ static void mlxsw_sp_fib4_entry_replace(struct mlxsw_sp *mlxsw_sp,
 	/* We inserted the new entry before replaced one */
 	replaced = list_next_entry(fib4_entry, common.list);
 
-	mlxsw_sp_fib4_node_entry_unlink(mlxsw_sp, replaced);
+	mlxsw_sp_fib_node_entry_unlink(mlxsw_sp, &replaced->common);
 	mlxsw_sp_fib4_entry_destroy(mlxsw_sp, replaced);
 	mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
 }
@@ -4884,17 +4884,17 @@ mlxsw_sp_router_fib4_replace(struct mlxsw_sp *mlxsw_sp,
 		goto err_fib4_entry_create;
 	}
 
-	err = mlxsw_sp_fib4_node_entry_link(mlxsw_sp, fib4_entry);
+	err = mlxsw_sp_fib_node_entry_link(mlxsw_sp, &fib4_entry->common);
 	if (err) {
 		dev_warn(mlxsw_sp->bus_info->dev, "Failed to link FIB entry to node\n");
-		goto err_fib4_node_entry_link;
+		goto err_fib_node_entry_link;
 	}
 
 	mlxsw_sp_fib4_entry_replace(mlxsw_sp, fib4_entry);
 
 	return 0;
 
-err_fib4_node_entry_link:
+err_fib_node_entry_link:
 	mlxsw_sp_fib4_entry_destroy(mlxsw_sp, fib4_entry);
 err_fib4_entry_create:
 	mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
@@ -4915,7 +4915,7 @@ static void mlxsw_sp_router_fib4_del(struct mlxsw_sp *mlxsw_sp,
 		return;
 	fib_node = fib4_entry->common.fib_node;
 
-	mlxsw_sp_fib4_node_entry_unlink(mlxsw_sp, fib4_entry);
+	mlxsw_sp_fib_node_entry_unlink(mlxsw_sp, &fib4_entry->common);
 	mlxsw_sp_fib4_entry_destroy(mlxsw_sp, fib4_entry);
 	mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
 }
@@ -5374,33 +5374,6 @@ static void mlxsw_sp_fib6_entry_destroy(struct mlxsw_sp *mlxsw_sp,
 	kfree(fib6_entry);
 }
 
-static int mlxsw_sp_fib6_node_entry_link(struct mlxsw_sp *mlxsw_sp,
-					 struct mlxsw_sp_fib6_entry *fib6_entry)
-{
-	struct mlxsw_sp_fib_node *fib_node = fib6_entry->common.fib_node;
-	int err;
-
-	list_add(&fib6_entry->common.list, &fib_node->entry_list);
-
-	err = mlxsw_sp_fib_node_entry_add(mlxsw_sp, &fib6_entry->common);
-	if (err)
-		goto err_fib_node_entry_add;
-
-	return 0;
-
-err_fib_node_entry_add:
-	list_del(&fib6_entry->common.list);
-	return err;
-}
-
-static void
-mlxsw_sp_fib6_node_entry_unlink(struct mlxsw_sp *mlxsw_sp,
-				struct mlxsw_sp_fib6_entry *fib6_entry)
-{
-	mlxsw_sp_fib_node_entry_del(mlxsw_sp, &fib6_entry->common);
-	list_del(&fib6_entry->common.list);
-}
-
 static struct mlxsw_sp_fib6_entry *
 mlxsw_sp_fib6_entry_lookup(struct mlxsw_sp *mlxsw_sp,
 			   const struct fib6_info *rt)
@@ -5445,7 +5418,7 @@ static void mlxsw_sp_fib6_entry_replace(struct mlxsw_sp *mlxsw_sp,
 	/* We inserted the new entry before replaced one */
 	replaced = list_next_entry(fib6_entry, common.list);
 
-	mlxsw_sp_fib6_node_entry_unlink(mlxsw_sp, replaced);
+	mlxsw_sp_fib_node_entry_unlink(mlxsw_sp, &replaced->common);
 	mlxsw_sp_fib6_entry_destroy(mlxsw_sp, replaced);
 	mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
 }
@@ -5483,15 +5456,15 @@ static int mlxsw_sp_router_fib6_replace(struct mlxsw_sp *mlxsw_sp,
 		goto err_fib6_entry_create;
 	}
 
-	err = mlxsw_sp_fib6_node_entry_link(mlxsw_sp, fib6_entry);
+	err = mlxsw_sp_fib_node_entry_link(mlxsw_sp, &fib6_entry->common);
 	if (err)
-		goto err_fib6_node_entry_link;
+		goto err_fib_node_entry_link;
 
 	mlxsw_sp_fib6_entry_replace(mlxsw_sp, fib6_entry);
 
 	return 0;
 
-err_fib6_node_entry_link:
+err_fib_node_entry_link:
 	mlxsw_sp_fib6_entry_destroy(mlxsw_sp, fib6_entry);
 err_fib6_entry_create:
 	mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
@@ -5577,7 +5550,7 @@ static void mlxsw_sp_router_fib6_del(struct mlxsw_sp *mlxsw_sp,
 
 	fib_node = fib6_entry->common.fib_node;
 
-	mlxsw_sp_fib6_node_entry_unlink(mlxsw_sp, fib6_entry);
+	mlxsw_sp_fib_node_entry_unlink(mlxsw_sp, &fib6_entry->common);
 	mlxsw_sp_fib6_entry_destroy(mlxsw_sp, fib6_entry);
 	mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
 }
@@ -5737,7 +5710,7 @@ static void mlxsw_sp_fib4_node_flush(struct mlxsw_sp *mlxsw_sp,
 				 common.list) {
 		bool do_break = &tmp->common.list == &fib_node->entry_list;
 
-		mlxsw_sp_fib4_node_entry_unlink(mlxsw_sp, fib4_entry);
+		mlxsw_sp_fib_node_entry_unlink(mlxsw_sp, &fib4_entry->common);
 		mlxsw_sp_fib4_entry_destroy(mlxsw_sp, fib4_entry);
 		mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
 		/* Break when entry list is empty and node was freed.
@@ -5758,7 +5731,7 @@ static void mlxsw_sp_fib6_node_flush(struct mlxsw_sp *mlxsw_sp,
 				 common.list) {
 		bool do_break = &tmp->common.list == &fib_node->entry_list;
 
-		mlxsw_sp_fib6_node_entry_unlink(mlxsw_sp, fib6_entry);
+		mlxsw_sp_fib_node_entry_unlink(mlxsw_sp, &fib6_entry->common);
 		mlxsw_sp_fib6_entry_destroy(mlxsw_sp, fib6_entry);
 		mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
 		if (do_break)
-- 
cgit v1.2.3-59-g8ed1b


From 7c4a7ec8558dfc770c27931b43f01aede613faac Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Thu, 26 Dec 2019 18:41:17 +0200
Subject: mlxsw: spectrum_router: Remove FIB entry list from FIB node

As explained in previous patches, the driver no longer needs to maintain
a list of identical FIB entries (i.e, same {tb_id, prefix, prefix
length}) and therefore each FIB node can only store one FIB entry.

Remove the FIB entry list and simplify the code.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../net/ethernet/mellanox/mlxsw/spectrum_router.c  | 225 +++++++--------------
 1 file changed, 74 insertions(+), 151 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
index f332c55fc83e..da1c8342c8f5 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
@@ -382,9 +382,10 @@ enum mlxsw_sp_fib_entry_type {
 };
 
 struct mlxsw_sp_nexthop_group;
+struct mlxsw_sp_fib_entry;
 
 struct mlxsw_sp_fib_node {
-	struct list_head entry_list;
+	struct mlxsw_sp_fib_entry *fib_entry;
 	struct list_head list;
 	struct rhash_head ht_node;
 	struct mlxsw_sp_fib *fib;
@@ -397,7 +398,6 @@ struct mlxsw_sp_fib_entry_decap {
 };
 
 struct mlxsw_sp_fib_entry {
-	struct list_head list;
 	struct mlxsw_sp_fib_node *fib_node;
 	enum mlxsw_sp_fib_entry_type type;
 	struct list_head nexthop_group_node;
@@ -1162,7 +1162,6 @@ mlxsw_sp_router_ip2me_fib_entry_find(struct mlxsw_sp *mlxsw_sp, u32 tb_id,
 				     const union mlxsw_sp_l3addr *addr,
 				     enum mlxsw_sp_fib_entry_type type)
 {
-	struct mlxsw_sp_fib_entry *fib_entry;
 	struct mlxsw_sp_fib_node *fib_node;
 	unsigned char addr_prefix_len;
 	struct mlxsw_sp_fib *fib;
@@ -1191,15 +1190,10 @@ mlxsw_sp_router_ip2me_fib_entry_find(struct mlxsw_sp *mlxsw_sp, u32 tb_id,
 
 	fib_node = mlxsw_sp_fib_node_lookup(fib, addrp, addr_len,
 					    addr_prefix_len);
-	if (!fib_node || list_empty(&fib_node->entry_list))
+	if (!fib_node || fib_node->fib_entry->type != type)
 		return NULL;
 
-	fib_entry = list_first_entry(&fib_node->entry_list,
-				     struct mlxsw_sp_fib_entry, list);
-	if (fib_entry->type != type)
-		return NULL;
-
-	return fib_entry;
+	return fib_node->fib_entry;
 }
 
 /* Given an IPIP entry, find the corresponding decap route. */
@@ -1209,7 +1203,6 @@ mlxsw_sp_ipip_entry_find_decap(struct mlxsw_sp *mlxsw_sp,
 {
 	static struct mlxsw_sp_fib_node *fib_node;
 	const struct mlxsw_sp_ipip_ops *ipip_ops;
-	struct mlxsw_sp_fib_entry *fib_entry;
 	unsigned char saddr_prefix_len;
 	union mlxsw_sp_l3addr saddr;
 	struct mlxsw_sp_fib *ul_fib;
@@ -1244,15 +1237,11 @@ mlxsw_sp_ipip_entry_find_decap(struct mlxsw_sp *mlxsw_sp,
 
 	fib_node = mlxsw_sp_fib_node_lookup(ul_fib, saddrp, saddr_len,
 					    saddr_prefix_len);
-	if (!fib_node || list_empty(&fib_node->entry_list))
+	if (!fib_node ||
+	    fib_node->fib_entry->type != MLXSW_SP_FIB_ENTRY_TYPE_TRAP)
 		return NULL;
 
-	fib_entry = list_first_entry(&fib_node->entry_list,
-				     struct mlxsw_sp_fib_entry, list);
-	if (fib_entry->type != MLXSW_SP_FIB_ENTRY_TYPE_TRAP)
-		return NULL;
-
-	return fib_entry;
+	return fib_node->fib_entry;
 }
 
 static struct mlxsw_sp_ipip_entry *
@@ -4559,15 +4548,14 @@ mlxsw_sp_fib4_entry_lookup(struct mlxsw_sp *mlxsw_sp,
 	if (!fib_node)
 		return NULL;
 
-	list_for_each_entry(fib4_entry, &fib_node->entry_list, common.list) {
-		if (fib4_entry->tb_id == fen_info->tb_id &&
-		    fib4_entry->tos == fen_info->tos &&
-		    fib4_entry->type == fen_info->type &&
-		    mlxsw_sp_nexthop4_group_fi(fib4_entry->common.nh_group) ==
-		    fen_info->fi) {
-			return fib4_entry;
-		}
-	}
+	fib4_entry = container_of(fib_node->fib_entry,
+				  struct mlxsw_sp_fib4_entry, common);
+	if (fib4_entry->tb_id == fen_info->tb_id &&
+	    fib4_entry->tos == fen_info->tos &&
+	    fib4_entry->type == fen_info->type &&
+	    mlxsw_sp_nexthop4_group_fi(fib4_entry->common.nh_group) ==
+	    fen_info->fi)
+		return fib4_entry;
 
 	return NULL;
 }
@@ -4615,7 +4603,6 @@ mlxsw_sp_fib_node_create(struct mlxsw_sp_fib *fib, const void *addr,
 	if (!fib_node)
 		return NULL;
 
-	INIT_LIST_HEAD(&fib_node->entry_list);
 	list_add(&fib_node->list, &fib->node_list);
 	memcpy(fib_node->key.addr, addr, addr_len);
 	fib_node->key.prefix_len = prefix_len;
@@ -4626,18 +4613,9 @@ mlxsw_sp_fib_node_create(struct mlxsw_sp_fib *fib, const void *addr,
 static void mlxsw_sp_fib_node_destroy(struct mlxsw_sp_fib_node *fib_node)
 {
 	list_del(&fib_node->list);
-	WARN_ON(!list_empty(&fib_node->entry_list));
 	kfree(fib_node);
 }
 
-static bool
-mlxsw_sp_fib_node_entry_is_first(const struct mlxsw_sp_fib_node *fib_node,
-				 const struct mlxsw_sp_fib_entry *fib_entry)
-{
-	return list_first_entry(&fib_node->entry_list,
-				struct mlxsw_sp_fib_entry, list) == fib_entry;
-}
-
 static int mlxsw_sp_fib_lpm_tree_link(struct mlxsw_sp *mlxsw_sp,
 				      struct mlxsw_sp_fib_node *fib_node)
 {
@@ -4777,58 +4755,29 @@ static void mlxsw_sp_fib_node_put(struct mlxsw_sp *mlxsw_sp,
 {
 	struct mlxsw_sp_vr *vr = fib_node->fib->vr;
 
-	if (!list_empty(&fib_node->entry_list))
+	if (fib_node->fib_entry)
 		return;
 	mlxsw_sp_fib_node_fini(mlxsw_sp, fib_node);
 	mlxsw_sp_fib_node_destroy(fib_node);
 	mlxsw_sp_vr_put(mlxsw_sp, vr);
 }
 
-static int mlxsw_sp_fib_node_entry_add(struct mlxsw_sp *mlxsw_sp,
-				       struct mlxsw_sp_fib_entry *fib_entry)
-{
-	struct mlxsw_sp_fib_node *fib_node = fib_entry->fib_node;
-
-	/* To prevent packet loss, overwrite the previously offloaded
-	 * entry.
-	 */
-	if (!list_is_singular(&fib_node->entry_list)) {
-		enum mlxsw_reg_ralue_op op = MLXSW_REG_RALUE_OP_WRITE_DELETE;
-		struct mlxsw_sp_fib_entry *n = list_next_entry(fib_entry, list);
-
-		mlxsw_sp_fib_entry_offload_refresh(n, op, 0);
-	}
-
-	return mlxsw_sp_fib_entry_update(mlxsw_sp, fib_entry);
-}
-
-static void mlxsw_sp_fib_node_entry_del(struct mlxsw_sp *mlxsw_sp,
-					struct mlxsw_sp_fib_entry *fib_entry)
-{
-	struct mlxsw_sp_fib_node *fib_node = fib_entry->fib_node;
-
-	if (!mlxsw_sp_fib_node_entry_is_first(fib_node, fib_entry))
-		return;
-
-	mlxsw_sp_fib_entry_del(mlxsw_sp, fib_entry);
-}
-
 static int mlxsw_sp_fib_node_entry_link(struct mlxsw_sp *mlxsw_sp,
 					struct mlxsw_sp_fib_entry *fib_entry)
 {
 	struct mlxsw_sp_fib_node *fib_node = fib_entry->fib_node;
 	int err;
 
-	list_add(&fib_entry->list, &fib_node->entry_list);
+	fib_node->fib_entry = fib_entry;
 
-	err = mlxsw_sp_fib_node_entry_add(mlxsw_sp, fib_entry);
+	err = mlxsw_sp_fib_entry_update(mlxsw_sp, fib_entry);
 	if (err)
-		goto err_fib_node_entry_add;
+		goto err_fib_entry_update;
 
 	return 0;
 
-err_fib_node_entry_add:
-	list_del(&fib_entry->list);
+err_fib_entry_update:
+	fib_node->fib_entry = NULL;
 	return err;
 }
 
@@ -4836,32 +4785,18 @@ static void
 mlxsw_sp_fib_node_entry_unlink(struct mlxsw_sp *mlxsw_sp,
 			       struct mlxsw_sp_fib_entry *fib_entry)
 {
-	mlxsw_sp_fib_node_entry_del(mlxsw_sp, fib_entry);
-	list_del(&fib_entry->list);
-}
-
-static void mlxsw_sp_fib4_entry_replace(struct mlxsw_sp *mlxsw_sp,
-					struct mlxsw_sp_fib4_entry *fib4_entry)
-{
-	struct mlxsw_sp_fib_node *fib_node = fib4_entry->common.fib_node;
-	struct mlxsw_sp_fib4_entry *replaced;
-
-	if (list_is_singular(&fib_node->entry_list))
-		return;
-
-	/* We inserted the new entry before replaced one */
-	replaced = list_next_entry(fib4_entry, common.list);
+	struct mlxsw_sp_fib_node *fib_node = fib_entry->fib_node;
 
-	mlxsw_sp_fib_node_entry_unlink(mlxsw_sp, &replaced->common);
-	mlxsw_sp_fib4_entry_destroy(mlxsw_sp, replaced);
-	mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
+	mlxsw_sp_fib_entry_del(mlxsw_sp, fib_entry);
+	fib_node->fib_entry = NULL;
 }
 
 static int
 mlxsw_sp_router_fib4_replace(struct mlxsw_sp *mlxsw_sp,
 			     const struct fib_entry_notifier_info *fen_info)
 {
-	struct mlxsw_sp_fib4_entry *fib4_entry;
+	struct mlxsw_sp_fib4_entry *fib4_entry, *fib4_replaced;
+	struct mlxsw_sp_fib_entry *replaced;
 	struct mlxsw_sp_fib_node *fib_node;
 	int err;
 
@@ -4884,17 +4819,26 @@ mlxsw_sp_router_fib4_replace(struct mlxsw_sp *mlxsw_sp,
 		goto err_fib4_entry_create;
 	}
 
+	replaced = fib_node->fib_entry;
 	err = mlxsw_sp_fib_node_entry_link(mlxsw_sp, &fib4_entry->common);
 	if (err) {
 		dev_warn(mlxsw_sp->bus_info->dev, "Failed to link FIB entry to node\n");
 		goto err_fib_node_entry_link;
 	}
 
-	mlxsw_sp_fib4_entry_replace(mlxsw_sp, fib4_entry);
+	/* Nothing to replace */
+	if (!replaced)
+		return 0;
+
+	mlxsw_sp_fib_entry_offload_unset(replaced);
+	fib4_replaced = container_of(replaced, struct mlxsw_sp_fib4_entry,
+				     common);
+	mlxsw_sp_fib4_entry_destroy(mlxsw_sp, fib4_replaced);
 
 	return 0;
 
 err_fib_node_entry_link:
+	fib_node->fib_entry = replaced;
 	mlxsw_sp_fib4_entry_destroy(mlxsw_sp, fib4_entry);
 err_fib4_entry_create:
 	mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
@@ -5201,16 +5145,16 @@ mlxsw_sp_nexthop6_group_update(struct mlxsw_sp *mlxsw_sp,
 	 * currently associated with it in the device's table is that
 	 * of the old group. Start using the new one instead.
 	 */
-	err = mlxsw_sp_fib_node_entry_add(mlxsw_sp, &fib6_entry->common);
+	err = mlxsw_sp_fib_entry_update(mlxsw_sp, &fib6_entry->common);
 	if (err)
-		goto err_fib_node_entry_add;
+		goto err_fib_entry_update;
 
 	if (list_empty(&old_nh_grp->fib_list))
 		mlxsw_sp_nexthop6_group_destroy(mlxsw_sp, old_nh_grp);
 
 	return 0;
 
-err_fib_node_entry_add:
+err_fib_entry_update:
 	mlxsw_sp_nexthop6_group_put(mlxsw_sp, &fib6_entry->common);
 err_nexthop6_group_get:
 	list_add_tail(&fib6_entry->common.nexthop_group_node,
@@ -5381,6 +5325,7 @@ mlxsw_sp_fib6_entry_lookup(struct mlxsw_sp *mlxsw_sp,
 	struct mlxsw_sp_fib6_entry *fib6_entry;
 	struct mlxsw_sp_fib_node *fib_node;
 	struct mlxsw_sp_fib *fib;
+	struct fib6_info *cmp_rt;
 	struct mlxsw_sp_vr *vr;
 
 	vr = mlxsw_sp_vr_find(mlxsw_sp, rt->fib6_table->tb6_id);
@@ -5394,40 +5339,23 @@ mlxsw_sp_fib6_entry_lookup(struct mlxsw_sp *mlxsw_sp,
 	if (!fib_node)
 		return NULL;
 
-	list_for_each_entry(fib6_entry, &fib_node->entry_list, common.list) {
-		struct fib6_info *iter_rt = mlxsw_sp_fib6_entry_rt(fib6_entry);
-
-		if (rt->fib6_table->tb6_id == iter_rt->fib6_table->tb6_id &&
-		    rt->fib6_metric == iter_rt->fib6_metric &&
-		    mlxsw_sp_fib6_entry_rt_find(fib6_entry, rt))
-			return fib6_entry;
-	}
+	fib6_entry = container_of(fib_node->fib_entry,
+				  struct mlxsw_sp_fib6_entry, common);
+	cmp_rt = mlxsw_sp_fib6_entry_rt(fib6_entry);
+	if (rt->fib6_table->tb6_id == cmp_rt->fib6_table->tb6_id &&
+	    rt->fib6_metric == cmp_rt->fib6_metric &&
+	    mlxsw_sp_fib6_entry_rt_find(fib6_entry, rt))
+		return fib6_entry;
 
 	return NULL;
 }
 
-static void mlxsw_sp_fib6_entry_replace(struct mlxsw_sp *mlxsw_sp,
-					struct mlxsw_sp_fib6_entry *fib6_entry)
-{
-	struct mlxsw_sp_fib_node *fib_node = fib6_entry->common.fib_node;
-	struct mlxsw_sp_fib6_entry *replaced;
-
-	if (list_is_singular(&fib_node->entry_list))
-		return;
-
-	/* We inserted the new entry before replaced one */
-	replaced = list_next_entry(fib6_entry, common.list);
-
-	mlxsw_sp_fib_node_entry_unlink(mlxsw_sp, &replaced->common);
-	mlxsw_sp_fib6_entry_destroy(mlxsw_sp, replaced);
-	mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
-}
-
 static int mlxsw_sp_router_fib6_replace(struct mlxsw_sp *mlxsw_sp,
 					struct fib6_info **rt_arr,
 					unsigned int nrt6)
 {
-	struct mlxsw_sp_fib6_entry *fib6_entry;
+	struct mlxsw_sp_fib6_entry *fib6_entry, *fib6_replaced;
+	struct mlxsw_sp_fib_entry *replaced;
 	struct mlxsw_sp_fib_node *fib_node;
 	struct fib6_info *rt = rt_arr[0];
 	int err;
@@ -5456,15 +5384,24 @@ static int mlxsw_sp_router_fib6_replace(struct mlxsw_sp *mlxsw_sp,
 		goto err_fib6_entry_create;
 	}
 
+	replaced = fib_node->fib_entry;
 	err = mlxsw_sp_fib_node_entry_link(mlxsw_sp, &fib6_entry->common);
 	if (err)
 		goto err_fib_node_entry_link;
 
-	mlxsw_sp_fib6_entry_replace(mlxsw_sp, fib6_entry);
+	/* Nothing to replace */
+	if (!replaced)
+		return 0;
+
+	mlxsw_sp_fib_entry_offload_unset(replaced);
+	fib6_replaced = container_of(replaced, struct mlxsw_sp_fib6_entry,
+				     common);
+	mlxsw_sp_fib6_entry_destroy(mlxsw_sp, fib6_replaced);
 
 	return 0;
 
 err_fib_node_entry_link:
+	fib_node->fib_entry = replaced;
 	mlxsw_sp_fib6_entry_destroy(mlxsw_sp, fib6_entry);
 err_fib6_entry_create:
 	mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
@@ -5497,13 +5434,13 @@ static int mlxsw_sp_router_fib6_append(struct mlxsw_sp *mlxsw_sp,
 	if (IS_ERR(fib_node))
 		return PTR_ERR(fib_node);
 
-	if (WARN_ON_ONCE(list_empty(&fib_node->entry_list))) {
+	if (WARN_ON_ONCE(!fib_node->fib_entry)) {
 		mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
 		return -EINVAL;
 	}
 
-	fib6_entry = list_first_entry(&fib_node->entry_list,
-				      struct mlxsw_sp_fib6_entry, common.list);
+	fib6_entry = container_of(fib_node->fib_entry,
+				  struct mlxsw_sp_fib6_entry, common);
 	err = mlxsw_sp_fib6_entry_nexthop_add(mlxsw_sp, fib6_entry, rt_arr,
 					      nrt6);
 	if (err)
@@ -5704,39 +5641,25 @@ static int mlxsw_sp_router_set_abort_trap(struct mlxsw_sp *mlxsw_sp)
 static void mlxsw_sp_fib4_node_flush(struct mlxsw_sp *mlxsw_sp,
 				     struct mlxsw_sp_fib_node *fib_node)
 {
-	struct mlxsw_sp_fib4_entry *fib4_entry, *tmp;
-
-	list_for_each_entry_safe(fib4_entry, tmp, &fib_node->entry_list,
-				 common.list) {
-		bool do_break = &tmp->common.list == &fib_node->entry_list;
+	struct mlxsw_sp_fib4_entry *fib4_entry;
 
-		mlxsw_sp_fib_node_entry_unlink(mlxsw_sp, &fib4_entry->common);
-		mlxsw_sp_fib4_entry_destroy(mlxsw_sp, fib4_entry);
-		mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
-		/* Break when entry list is empty and node was freed.
-		 * Otherwise, we'll access freed memory in the next
-		 * iteration.
-		 */
-		if (do_break)
-			break;
-	}
+	fib4_entry = container_of(fib_node->fib_entry,
+				  struct mlxsw_sp_fib4_entry, common);
+	mlxsw_sp_fib_node_entry_unlink(mlxsw_sp, fib_node->fib_entry);
+	mlxsw_sp_fib4_entry_destroy(mlxsw_sp, fib4_entry);
+	mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
 }
 
 static void mlxsw_sp_fib6_node_flush(struct mlxsw_sp *mlxsw_sp,
 				     struct mlxsw_sp_fib_node *fib_node)
 {
-	struct mlxsw_sp_fib6_entry *fib6_entry, *tmp;
-
-	list_for_each_entry_safe(fib6_entry, tmp, &fib_node->entry_list,
-				 common.list) {
-		bool do_break = &tmp->common.list == &fib_node->entry_list;
+	struct mlxsw_sp_fib6_entry *fib6_entry;
 
-		mlxsw_sp_fib_node_entry_unlink(mlxsw_sp, &fib6_entry->common);
-		mlxsw_sp_fib6_entry_destroy(mlxsw_sp, fib6_entry);
-		mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
-		if (do_break)
-			break;
-	}
+	fib6_entry = container_of(fib_node->fib_entry,
+				  struct mlxsw_sp_fib6_entry, common);
+	mlxsw_sp_fib_node_entry_unlink(mlxsw_sp, fib_node->fib_entry);
+	mlxsw_sp_fib6_entry_destroy(mlxsw_sp, fib6_entry);
+	mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
 }
 
 static void mlxsw_sp_fib_node_flush(struct mlxsw_sp *mlxsw_sp,
-- 
cgit v1.2.3-59-g8ed1b


From 3aec743d69822d22d4a5b60deb9518ed8be6fa67 Mon Sep 17 00:00:00 2001
From: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date: Thu, 26 Dec 2019 19:51:47 +0100
Subject: net: phy: realtek: add logging for the RGMII TX delay configuration

RGMII requires a delay of 2ns between the data and the clock signal.
There are at least three ways this can happen. One possibility is by
having the PHY generate this delay.
This is a common source for problems (for example with slow TX speeds or
packet loss when sending data). The TX delay configuration of the
RTL8211F PHY can be set either by pin-strappping the RXD1 pin (HIGH
means enabled, LOW means disabled) or through configuring a paged
register. The setting from the RXD1 pin is also reflected in the
register.

Add debug logging to the TX delay configuration on RTL8211F so it's
easier to spot these issues (for example if the TX delay is enabled for
both, the RTL8211F PHY and the MAC).
This is especially helpful because there is no public datasheet for the
RTL8211F PHY available with all the RX/TX delay specifics.

Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/realtek.c | 19 ++++++++++++++++++-
 1 file changed, 18 insertions(+), 1 deletion(-)

diff --git a/drivers/net/phy/realtek.c b/drivers/net/phy/realtek.c
index 476db5345e1a..879ca37c8508 100644
--- a/drivers/net/phy/realtek.c
+++ b/drivers/net/phy/realtek.c
@@ -171,7 +171,9 @@ static int rtl8211c_config_init(struct phy_device *phydev)
 
 static int rtl8211f_config_init(struct phy_device *phydev)
 {
+	struct device *dev = &phydev->mdio.dev;
 	u16 val;
+	int ret;
 
 	/* enable TX-delay for rgmii-{id,txid}, and disable it for rgmii and
 	 * rgmii-rxid. The RX-delay can be enabled by the external RXDLY pin.
@@ -189,7 +191,22 @@ static int rtl8211f_config_init(struct phy_device *phydev)
 		return 0;
 	}
 
-	return phy_modify_paged(phydev, 0xd08, 0x11, RTL8211F_TX_DELAY, val);
+	ret = phy_modify_paged_changed(phydev, 0xd08, 0x11, RTL8211F_TX_DELAY,
+				       val);
+	if (ret < 0) {
+		dev_err(dev, "Failed to update the TX delay register\n");
+		return ret;
+	} else if (ret) {
+		dev_dbg(dev,
+			"%s 2ns TX delay (and changing the value from pin-strapping RXD1 or the bootloader)\n",
+			val ? "Enabling" : "Disabling");
+	} else {
+		dev_dbg(dev,
+			"2ns TX delay was already %s (by pin-strapping RXD1 or bootloader configuration)\n",
+			val ? "enabled" : "disabled");
+	}
+
+	return 0;
 }
 
 static int rtl8211e_config_init(struct phy_device *phydev)
-- 
cgit v1.2.3-59-g8ed1b


From 1b3047b5208a80e7213bc4621f6fa4e5feb8dee4 Mon Sep 17 00:00:00 2001
From: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date: Thu, 26 Dec 2019 19:51:48 +0100
Subject: net: phy: realtek: add support for configuring the RX delay on
 RTL8211F

On RTL8211F the RX and TX delays (2ns) can be configured in two ways:
- pin strapping (RXD1 for the TX delay and RXD0 for the RX delay, LOW
  means "off" and HIGH means "on") which is read during PHY reset
- using software to configure the TX and RX delay registers

So far only the configuration using pin strapping has been supported.
Add support for enabling or disabling the RGMII RX delay based on the
phy-mode to be able to get the RX delay into a known state. This is
important because the RX delay has to be coordinated between the PHY,
MAC and the PCB design (trace length). With an invalid RX delay applied
(for example if both PHY and MAC add a 2ns RX delay) Ethernet may not
work at all.

Also add debug logging when configuring the RX delay (just like the TX
delay) because this is a common source of problems.

Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/realtek.c | 46 ++++++++++++++++++++++++++++++++++++----------
 1 file changed, 36 insertions(+), 10 deletions(-)

diff --git a/drivers/net/phy/realtek.c b/drivers/net/phy/realtek.c
index 879ca37c8508..f5fa2fff3ddc 100644
--- a/drivers/net/phy/realtek.c
+++ b/drivers/net/phy/realtek.c
@@ -29,6 +29,8 @@
 #define RTL8211F_INSR				0x1d
 
 #define RTL8211F_TX_DELAY			BIT(8)
+#define RTL8211F_RX_DELAY			BIT(3)
+
 #define RTL8211E_TX_DELAY			BIT(1)
 #define RTL8211E_RX_DELAY			BIT(2)
 #define RTL8211E_MODE_MII_GMII			BIT(3)
@@ -172,38 +174,62 @@ static int rtl8211c_config_init(struct phy_device *phydev)
 static int rtl8211f_config_init(struct phy_device *phydev)
 {
 	struct device *dev = &phydev->mdio.dev;
-	u16 val;
+	u16 val_txdly, val_rxdly;
 	int ret;
 
-	/* enable TX-delay for rgmii-{id,txid}, and disable it for rgmii and
-	 * rgmii-rxid. The RX-delay can be enabled by the external RXDLY pin.
-	 */
 	switch (phydev->interface) {
 	case PHY_INTERFACE_MODE_RGMII:
+		val_txdly = 0;
+		val_rxdly = 0;
+		break;
+
 	case PHY_INTERFACE_MODE_RGMII_RXID:
-		val = 0;
+		val_txdly = 0;
+		val_rxdly = RTL8211F_RX_DELAY;
 		break;
-	case PHY_INTERFACE_MODE_RGMII_ID:
+
 	case PHY_INTERFACE_MODE_RGMII_TXID:
-		val = RTL8211F_TX_DELAY;
+		val_txdly = RTL8211F_TX_DELAY;
+		val_rxdly = 0;
+		break;
+
+	case PHY_INTERFACE_MODE_RGMII_ID:
+		val_txdly = RTL8211F_TX_DELAY;
+		val_rxdly = RTL8211F_RX_DELAY;
 		break;
+
 	default: /* the rest of the modes imply leaving delay as is. */
 		return 0;
 	}
 
 	ret = phy_modify_paged_changed(phydev, 0xd08, 0x11, RTL8211F_TX_DELAY,
-				       val);
+				       val_txdly);
 	if (ret < 0) {
 		dev_err(dev, "Failed to update the TX delay register\n");
 		return ret;
 	} else if (ret) {
 		dev_dbg(dev,
 			"%s 2ns TX delay (and changing the value from pin-strapping RXD1 or the bootloader)\n",
-			val ? "Enabling" : "Disabling");
+			val_txdly ? "Enabling" : "Disabling");
 	} else {
 		dev_dbg(dev,
 			"2ns TX delay was already %s (by pin-strapping RXD1 or bootloader configuration)\n",
-			val ? "enabled" : "disabled");
+			val_txdly ? "enabled" : "disabled");
+	}
+
+	ret = phy_modify_paged_changed(phydev, 0xd08, 0x15, RTL8211F_RX_DELAY,
+				       val_rxdly);
+	if (ret < 0) {
+		dev_err(dev, "Failed to update the RX delay register\n");
+		return ret;
+	} else if (ret) {
+		dev_dbg(dev,
+			"%s 2ns RX delay (and changing the value from pin-strapping RXD0 or the bootloader)\n",
+			val_rxdly ? "Enabling" : "Disabling");
+	} else {
+		dev_dbg(dev,
+			"2ns RX delay was already %s (by pin-strapping RXD0 or bootloader configuration)\n",
+			val_rxdly ? "enabled" : "disabled");
 	}
 
 	return 0;
-- 
cgit v1.2.3-59-g8ed1b


From 9476654bd5e8ad42abe8ee9f9e90069ff8e60c17 Mon Sep 17 00:00:00 2001
From: Paul Durrant <pdurrant@amazon.com>
Date: Mon, 23 Dec 2019 09:59:23 +0000
Subject: xen-netback: support dynamic unbind/bind

By re-attaching RX, TX, and CTL rings during connect() rather than
assuming they are freshly allocated (i.e. assuming the counters are zero),
and avoiding forcing state to Closed in netback_remove() it is possible
for vif instances to be unbound and re-bound from and to (respectively) a
running guest.

Dynamic unbind/bind is a highly useful feature for a backend module as it
allows it to be unloaded and re-loaded (i.e. updated) without requiring
domUs to be halted.

This has been tested by running iperf as a server in the test VM and
then running a client against it in a continuous loop, whilst also
running:

while true;
  do echo vif-$DOMID-$VIF >unbind;
  echo down;
  rmmod xen-netback;
  echo unloaded;
  modprobe xen-netback;
  cd $(pwd);
  brctl addif xenbr0 vif$DOMID.$VIF;
  ip link set vif$DOMID.$VIF up;
  echo up;
  sleep 5;
  done

in dom0 from /sys/bus/xen-backend/drivers/vif to continuously unbind,
unload, re-load, re-bind and re-plumb the backend.

Clearly a performance drop was seen but no TCP connection resets were
observed during this test and moreover a parallel SSH connection into the
guest remained perfectly usable throughout.

Signed-off-by: Paul Durrant <pdurrant@amazon.com>
Reviewed-by: Wei Liu <wei.liu@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/xen-netback/interface.c | 10 +++++++++-
 drivers/net/xen-netback/netback.c   | 20 +++++++++++++++++---
 drivers/net/xen-netback/xenbus.c    |  5 ++---
 3 files changed, 28 insertions(+), 7 deletions(-)

diff --git a/drivers/net/xen-netback/interface.c b/drivers/net/xen-netback/interface.c
index f15ba3de6195..0c8a02a1ead7 100644
--- a/drivers/net/xen-netback/interface.c
+++ b/drivers/net/xen-netback/interface.c
@@ -585,6 +585,7 @@ int xenvif_connect_ctrl(struct xenvif *vif, grant_ref_t ring_ref,
 	struct net_device *dev = vif->dev;
 	void *addr;
 	struct xen_netif_ctrl_sring *shared;
+	RING_IDX rsp_prod, req_prod;
 	int err;
 
 	err = xenbus_map_ring_valloc(xenvif_to_xenbus_device(vif),
@@ -593,7 +594,14 @@ int xenvif_connect_ctrl(struct xenvif *vif, grant_ref_t ring_ref,
 		goto err;
 
 	shared = (struct xen_netif_ctrl_sring *)addr;
-	BACK_RING_INIT(&vif->ctrl, shared, XEN_PAGE_SIZE);
+	rsp_prod = READ_ONCE(shared->rsp_prod);
+	req_prod = READ_ONCE(shared->req_prod);
+
+	BACK_RING_ATTACH(&vif->ctrl, shared, rsp_prod, XEN_PAGE_SIZE);
+
+	err = -EIO;
+	if (req_prod - rsp_prod > RING_SIZE(&vif->ctrl))
+		goto err_unmap;
 
 	err = bind_interdomain_evtchn_to_irq(vif->domid, evtchn);
 	if (err < 0)
diff --git a/drivers/net/xen-netback/netback.c b/drivers/net/xen-netback/netback.c
index 0020b2e8c279..315dfc6ea297 100644
--- a/drivers/net/xen-netback/netback.c
+++ b/drivers/net/xen-netback/netback.c
@@ -1453,7 +1453,7 @@ int xenvif_map_frontend_data_rings(struct xenvif_queue *queue,
 	void *addr;
 	struct xen_netif_tx_sring *txs;
 	struct xen_netif_rx_sring *rxs;
-
+	RING_IDX rsp_prod, req_prod;
 	int err = -ENOMEM;
 
 	err = xenbus_map_ring_valloc(xenvif_to_xenbus_device(queue->vif),
@@ -1462,7 +1462,14 @@ int xenvif_map_frontend_data_rings(struct xenvif_queue *queue,
 		goto err;
 
 	txs = (struct xen_netif_tx_sring *)addr;
-	BACK_RING_INIT(&queue->tx, txs, XEN_PAGE_SIZE);
+	rsp_prod = READ_ONCE(txs->rsp_prod);
+	req_prod = READ_ONCE(txs->req_prod);
+
+	BACK_RING_ATTACH(&queue->tx, txs, rsp_prod, XEN_PAGE_SIZE);
+
+	err = -EIO;
+	if (req_prod - rsp_prod > RING_SIZE(&queue->tx))
+		goto err;
 
 	err = xenbus_map_ring_valloc(xenvif_to_xenbus_device(queue->vif),
 				     &rx_ring_ref, 1, &addr);
@@ -1470,7 +1477,14 @@ int xenvif_map_frontend_data_rings(struct xenvif_queue *queue,
 		goto err;
 
 	rxs = (struct xen_netif_rx_sring *)addr;
-	BACK_RING_INIT(&queue->rx, rxs, XEN_PAGE_SIZE);
+	rsp_prod = READ_ONCE(rxs->rsp_prod);
+	req_prod = READ_ONCE(rxs->req_prod);
+
+	BACK_RING_ATTACH(&queue->rx, rxs, rsp_prod, XEN_PAGE_SIZE);
+
+	err = -EIO;
+	if (req_prod - rsp_prod > RING_SIZE(&queue->rx))
+		goto err;
 
 	return 0;
 
diff --git a/drivers/net/xen-netback/xenbus.c b/drivers/net/xen-netback/xenbus.c
index 17b4950ec051..286054b60d47 100644
--- a/drivers/net/xen-netback/xenbus.c
+++ b/drivers/net/xen-netback/xenbus.c
@@ -954,12 +954,10 @@ static int netback_remove(struct xenbus_device *dev)
 {
 	struct backend_info *be = dev_get_drvdata(&dev->dev);
 
-	set_backend_state(be, XenbusStateClosed);
-
 	unregister_hotplug_status_watch(be);
 	if (be->vif) {
 		kobject_uevent(&dev->dev.kobj, KOBJ_OFFLINE);
-		xen_unregister_watchers(be->vif);
+		backend_disconnect(be);
 		xenvif_free(be->vif);
 		be->vif = NULL;
 	}
@@ -1131,6 +1129,7 @@ static struct xenbus_driver netback_driver = {
 	.remove = netback_remove,
 	.uevent = netback_uevent,
 	.otherend_changed = frontend_changed,
+	.allow_rebind = true,
 };
 
 int xenvif_xenbus_init(void)
-- 
cgit v1.2.3-59-g8ed1b


From 0914d2bb11cc182039084ac3b961dc359b647468 Mon Sep 17 00:00:00 2001
From: Mao Wenan <maowenan@huawei.com>
Date: Mon, 23 Dec 2019 18:42:57 +0800
Subject: af_packet: refactoring code for prb_calc_retire_blk_tmo

If __ethtool_get_link_ksettings() is failed and with
non-zero value, prb_calc_retire_blk_tmo() should return
DEFAULT_PRB_RETIRE_TOV firstly.

This patch is to refactory code and make it more readable.

Signed-off-by: Mao Wenan <maowenan@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/packet/af_packet.c | 30 ++++++++++++------------------
 1 file changed, 12 insertions(+), 18 deletions(-)

diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 118cd66b7516..3bec515ccde3 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -520,7 +520,7 @@ static int prb_calc_retire_blk_tmo(struct packet_sock *po,
 				int blk_size_in_bytes)
 {
 	struct net_device *dev;
-	unsigned int mbits = 0, msec = 0, div = 0, tmo = 0;
+	unsigned int mbits, div;
 	struct ethtool_link_ksettings ecmd;
 	int err;
 
@@ -532,31 +532,25 @@ static int prb_calc_retire_blk_tmo(struct packet_sock *po,
 	}
 	err = __ethtool_get_link_ksettings(dev, &ecmd);
 	rtnl_unlock();
-	if (!err) {
-		/*
-		 * If the link speed is so slow you don't really
-		 * need to worry about perf anyways
-		 */
-		if (ecmd.base.speed < SPEED_1000 ||
-		    ecmd.base.speed == SPEED_UNKNOWN) {
-			return DEFAULT_PRB_RETIRE_TOV;
-		} else {
-			msec = 1;
-			div = ecmd.base.speed / 1000;
-		}
-	} else
+	if (err)
 		return DEFAULT_PRB_RETIRE_TOV;
 
+	/* If the link speed is so slow you don't really
+	 * need to worry about perf anyways
+	 */
+	if (ecmd.base.speed < SPEED_1000 ||
+	    ecmd.base.speed == SPEED_UNKNOWN)
+		return DEFAULT_PRB_RETIRE_TOV;
+
+	div = ecmd.base.speed / 1000;
 	mbits = (blk_size_in_bytes * 8) / (1024 * 1024);
 
 	if (div)
 		mbits /= div;
 
-	tmo = mbits * msec;
-
 	if (div)
-		return tmo+1;
-	return tmo;
+		return mbits + 1;
+	return mbits;
 }
 
 static void prb_init_ft_ops(struct tpacket_kbdq_core *p1,
-- 
cgit v1.2.3-59-g8ed1b


From c04b116a2866b3e3a6b909a576769ccde3d622ac Mon Sep 17 00:00:00 2001
From: Julian Wiedmann <jwi@linux.ibm.com>
Date: Mon, 23 Dec 2019 15:22:25 +0100
Subject: s390/qeth: consolidate RX code

To reduce the path length and levels of indirection, move the RX
processing from the sub-drivers into the core.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/s390/net/qeth_core.h      |  16 ----
 drivers/s390/net/qeth_core_main.c | 180 ++++++++++++++++++++++++++++++++------
 drivers/s390/net/qeth_l2_main.c   |  40 ---------
 drivers/s390/net/qeth_l3_main.c   |  91 -------------------
 4 files changed, 153 insertions(+), 174 deletions(-)

diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h
index 6e16b19732f6..4ab3be814ea7 100644
--- a/drivers/s390/net/qeth_core.h
+++ b/drivers/s390/net/qeth_core.h
@@ -728,7 +728,6 @@ struct qeth_osn_info {
 
 struct qeth_discipline {
 	const struct device_type *devtype;
-	int (*process_rx_buffer)(struct qeth_card *card, int budget, int *done);
 	int (*recover)(void *ptr);
 	int (*setup) (struct ccwgroup_device *);
 	void (*remove) (struct ccwgroup_device *);
@@ -923,18 +922,6 @@ static inline struct dst_entry *qeth_dst_check_rcu(struct sk_buff *skb, int ipv)
 	return dst;
 }
 
-static inline void qeth_rx_csum(struct qeth_card *card, struct sk_buff *skb,
-				u8 flags)
-{
-	if ((card->dev->features & NETIF_F_RXCSUM) &&
-	    (flags & QETH_HDR_EXT_CSUM_TRANSP_REQ)) {
-		skb->ip_summed = CHECKSUM_UNNECESSARY;
-		QETH_CARD_STAT_INC(card, rx_skb_csum);
-	} else {
-		skb->ip_summed = CHECKSUM_NONE;
-	}
-}
-
 static inline void qeth_tx_csum(struct sk_buff *skb, u8 *flags, int ipv)
 {
 	*flags |= QETH_HDR_EXT_CSUM_TRANSP_REQ;
@@ -1031,9 +1018,6 @@ struct qeth_cmd_buffer *qeth_get_diag_cmd(struct qeth_card *card,
 void qeth_notify_cmd(struct qeth_cmd_buffer *iob, int reason);
 void qeth_put_cmd(struct qeth_cmd_buffer *iob);
 
-struct sk_buff *qeth_core_get_next_skb(struct qeth_card *,
-		struct qeth_qdio_buffer *, struct qdio_buffer_element **, int *,
-		struct qeth_hdr **);
 void qeth_schedule_recovery(struct qeth_card *);
 int qeth_poll(struct napi_struct *napi, int budget);
 void qeth_clear_ipacmd_list(struct qeth_card *);
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index 41f37ce89a4f..8e2c0588525f 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -5046,6 +5046,114 @@ out:
 }
 EXPORT_SYMBOL_GPL(qeth_core_hardsetup_card);
 
+#if IS_ENABLED(CONFIG_QETH_L3)
+static void qeth_l3_rebuild_skb(struct qeth_card *card, struct sk_buff *skb,
+				struct qeth_hdr *hdr)
+{
+	struct af_iucv_trans_hdr *iucv = (struct af_iucv_trans_hdr *) skb->data;
+	struct qeth_hdr_layer3 *l3_hdr = &hdr->hdr.l3;
+	struct net_device *dev = skb->dev;
+
+	if (IS_IQD(card) && iucv->magic == ETH_P_AF_IUCV) {
+		dev_hard_header(skb, dev, ETH_P_AF_IUCV, dev->dev_addr,
+				"FAKELL", skb->len);
+		return;
+	}
+
+	if (!(l3_hdr->flags & QETH_HDR_PASSTHRU)) {
+		u16 prot = (l3_hdr->flags & QETH_HDR_IPV6) ? ETH_P_IPV6 :
+							     ETH_P_IP;
+		unsigned char tg_addr[ETH_ALEN];
+
+		skb_reset_network_header(skb);
+		switch (l3_hdr->flags & QETH_HDR_CAST_MASK) {
+		case QETH_CAST_MULTICAST:
+			if (prot == ETH_P_IP)
+				ip_eth_mc_map(ip_hdr(skb)->daddr, tg_addr);
+			else
+				ipv6_eth_mc_map(&ipv6_hdr(skb)->daddr, tg_addr);
+			QETH_CARD_STAT_INC(card, rx_multicast);
+			break;
+		case QETH_CAST_BROADCAST:
+			ether_addr_copy(tg_addr, dev->broadcast);
+			QETH_CARD_STAT_INC(card, rx_multicast);
+			break;
+		default:
+			if (card->options.sniffer)
+				skb->pkt_type = PACKET_OTHERHOST;
+			ether_addr_copy(tg_addr, dev->dev_addr);
+		}
+
+		if (l3_hdr->ext_flags & QETH_HDR_EXT_SRC_MAC_ADDR)
+			dev_hard_header(skb, dev, prot, tg_addr,
+					&l3_hdr->next_hop.rx.src_mac, skb->len);
+		else
+			dev_hard_header(skb, dev, prot, tg_addr, "FAKELL",
+					skb->len);
+	}
+
+	/* copy VLAN tag from hdr into skb */
+	if (!card->options.sniffer &&
+	    (l3_hdr->ext_flags & (QETH_HDR_EXT_VLAN_FRAME |
+				  QETH_HDR_EXT_INCLUDE_VLAN_TAG))) {
+		u16 tag = (l3_hdr->ext_flags & QETH_HDR_EXT_VLAN_FRAME) ?
+				l3_hdr->vlan_id :
+				l3_hdr->next_hop.rx.vlan_id;
+
+		__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), tag);
+	}
+}
+#endif
+
+static void qeth_receive_skb(struct qeth_card *card, struct sk_buff *skb,
+			     struct qeth_hdr *hdr)
+{
+	bool is_cso;
+
+	switch (hdr->hdr.l2.id) {
+	case QETH_HEADER_TYPE_OSN:
+		skb_push(skb, sizeof(*hdr));
+		skb_copy_to_linear_data(skb, hdr, sizeof(*hdr));
+		QETH_CARD_STAT_ADD(card, rx_bytes, skb->len);
+		QETH_CARD_STAT_INC(card, rx_packets);
+
+		card->osn_info.data_cb(skb);
+		return;
+#if IS_ENABLED(CONFIG_QETH_L3)
+	case QETH_HEADER_TYPE_LAYER3:
+		qeth_l3_rebuild_skb(card, skb, hdr);
+		is_cso = hdr->hdr.l3.ext_flags & QETH_HDR_EXT_CSUM_TRANSP_REQ;
+		break;
+#endif
+	case QETH_HEADER_TYPE_LAYER2:
+		is_cso = hdr->hdr.l2.flags[1] & QETH_HDR_EXT_CSUM_TRANSP_REQ;
+		break;
+	default:
+		/* never happens */
+		dev_kfree_skb_any(skb);
+		return;
+	}
+
+	if (is_cso && (card->dev->features & NETIF_F_RXCSUM)) {
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+		QETH_CARD_STAT_INC(card, rx_skb_csum);
+	} else {
+		skb->ip_summed = CHECKSUM_NONE;
+	}
+
+	skb->protocol = eth_type_trans(skb, skb->dev);
+
+	QETH_CARD_STAT_ADD(card, rx_bytes, skb->len);
+	QETH_CARD_STAT_INC(card, rx_packets);
+	if (skb_is_nonlinear(skb)) {
+		QETH_CARD_STAT_INC(card, rx_sg_skbs);
+		QETH_CARD_STAT_ADD(card, rx_sg_frags,
+				   skb_shinfo(skb)->nr_frags);
+	}
+
+	napi_gro_receive(&card->napi, skb);
+}
+
 static void qeth_create_skb_frag(struct sk_buff *skb, char *data, int data_len)
 {
 	struct page *page = virt_to_page(data);
@@ -5062,10 +5170,10 @@ static inline int qeth_is_last_sbale(struct qdio_buffer_element *sbale)
 	return (sbale->eflags & SBAL_EFLAGS_LAST_ENTRY);
 }
 
-struct sk_buff *qeth_core_get_next_skb(struct qeth_card *card,
-		struct qeth_qdio_buffer *qethbuffer,
-		struct qdio_buffer_element **__element, int *__offset,
-		struct qeth_hdr **hdr)
+static int qeth_extract_skb(struct qeth_card *card,
+			    struct qeth_qdio_buffer *qethbuffer,
+			    struct qdio_buffer_element **__element,
+			    int *__offset)
 {
 	struct qdio_buffer_element *element = *__element;
 	struct qdio_buffer *buffer = qethbuffer->buffer;
@@ -5073,6 +5181,7 @@ struct sk_buff *qeth_core_get_next_skb(struct qeth_card *card,
 	int offset = *__offset;
 	bool use_rx_sg = false;
 	unsigned int headroom;
+	struct qeth_hdr *hdr;
 	struct sk_buff *skb;
 	int skb_len = 0;
 
@@ -5080,42 +5189,42 @@ next_packet:
 	/* qeth_hdr must not cross element boundaries */
 	while (element->length < offset + sizeof(struct qeth_hdr)) {
 		if (qeth_is_last_sbale(element))
-			return NULL;
+			return -ENODATA;
 		element++;
 		offset = 0;
 	}
-	*hdr = element->addr + offset;
 
-	offset += sizeof(struct qeth_hdr);
+	hdr = element->addr + offset;
+	offset += sizeof(*hdr);
 	skb = NULL;
 
-	switch ((*hdr)->hdr.l2.id) {
+	switch (hdr->hdr.l2.id) {
 	case QETH_HEADER_TYPE_LAYER2:
-		skb_len = (*hdr)->hdr.l2.pkt_length;
+		skb_len = hdr->hdr.l2.pkt_length;
 		linear_len = ETH_HLEN;
 		headroom = 0;
 		break;
 	case QETH_HEADER_TYPE_LAYER3:
-		skb_len = (*hdr)->hdr.l3.length;
+		skb_len = hdr->hdr.l3.length;
 		if (!IS_LAYER3(card)) {
 			QETH_CARD_STAT_INC(card, rx_dropped_notsupp);
 			goto walk_packet;
 		}
 
-		if ((*hdr)->hdr.l3.flags & QETH_HDR_PASSTHRU) {
+		if (hdr->hdr.l3.flags & QETH_HDR_PASSTHRU) {
 			linear_len = ETH_HLEN;
 			headroom = 0;
 			break;
 		}
 
-		if ((*hdr)->hdr.l3.flags & QETH_HDR_IPV6)
+		if (hdr->hdr.l3.flags & QETH_HDR_IPV6)
 			linear_len = sizeof(struct ipv6hdr);
 		else
 			linear_len = sizeof(struct iphdr);
 		headroom = ETH_HLEN;
 		break;
 	case QETH_HEADER_TYPE_OSN:
-		skb_len = (*hdr)->hdr.osn.pdu_length;
+		skb_len = hdr->hdr.osn.pdu_length;
 		if (!IS_OSN(card)) {
 			QETH_CARD_STAT_INC(card, rx_dropped_notsupp);
 			goto walk_packet;
@@ -5125,13 +5234,13 @@ next_packet:
 		headroom = sizeof(struct qeth_hdr);
 		break;
 	default:
-		if ((*hdr)->hdr.l2.id & QETH_HEADER_MASK_INVAL)
+		if (hdr->hdr.l2.id & QETH_HEADER_MASK_INVAL)
 			QETH_CARD_STAT_INC(card, rx_frame_errors);
 		else
 			QETH_CARD_STAT_INC(card, rx_dropped_notsupp);
 
 		/* Can't determine packet length, drop the whole buffer. */
-		return NULL;
+		return -EPROTONOSUPPORT;
 	}
 
 	if (skb_len < linear_len) {
@@ -5195,7 +5304,7 @@ walk_packet:
 					QETH_CARD_STAT_INC(card,
 							   rx_length_errors);
 				}
-				return NULL;
+				return -EMSGSIZE;
 			}
 			element++;
 			offset = 0;
@@ -5208,22 +5317,40 @@ walk_packet:
 
 	*__element = element;
 	*__offset = offset;
-	if (use_rx_sg) {
-		QETH_CARD_STAT_INC(card, rx_sg_skbs);
-		QETH_CARD_STAT_ADD(card, rx_sg_frags,
-				   skb_shinfo(skb)->nr_frags);
+
+	qeth_receive_skb(card, skb, hdr);
+	return 0;
+}
+
+static int qeth_extract_skbs(struct qeth_card *card, int budget,
+			     struct qeth_qdio_buffer *buf, bool *done)
+{
+	int work_done = 0;
+
+	WARN_ON_ONCE(!budget);
+	*done = false;
+
+	while (budget) {
+		if (qeth_extract_skb(card, buf, &card->rx.b_element,
+				     &card->rx.e_offset)) {
+			*done = true;
+			break;
+		}
+
+		work_done++;
+		budget--;
 	}
-	return skb;
+
+	return work_done;
 }
-EXPORT_SYMBOL_GPL(qeth_core_get_next_skb);
 
 int qeth_poll(struct napi_struct *napi, int budget)
 {
 	struct qeth_card *card = container_of(napi, struct qeth_card, napi);
 	int work_done = 0;
 	struct qeth_qdio_buffer *buffer;
-	int done;
 	int new_budget = budget;
+	bool done;
 
 	while (1) {
 		if (!card->rx.b_count) {
@@ -5246,11 +5373,10 @@ int qeth_poll(struct napi_struct *napi, int budget)
 			if (!(card->rx.qdio_err &&
 			    qeth_check_qdio_errors(card, buffer->buffer,
 			    card->rx.qdio_err, "qinerr")))
-				work_done +=
-					card->discipline->process_rx_buffer(
-						card, new_budget, &done);
+				work_done += qeth_extract_skbs(card, new_budget,
+							       buffer, &done);
 			else
-				done = 1;
+				done = true;
 
 			if (done) {
 				QETH_CARD_STAT_INC(card, rx_bufs);
diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c
index 8c95e6019bac..69d147c4d2ca 100644
--- a/drivers/s390/net/qeth_l2_main.c
+++ b/drivers/s390/net/qeth_l2_main.c
@@ -298,45 +298,6 @@ static void qeth_l2_stop_card(struct qeth_card *card)
 	card->info.promisc_mode = 0;
 }
 
-static int qeth_l2_process_inbound_buffer(struct qeth_card *card,
-				int budget, int *done)
-{
-	int work_done = 0;
-	struct sk_buff *skb;
-	struct qeth_hdr *hdr;
-	unsigned int len;
-
-	*done = 0;
-	WARN_ON_ONCE(!budget);
-	while (budget) {
-		skb = qeth_core_get_next_skb(card,
-			&card->qdio.in_q->bufs[card->rx.b_index],
-			&card->rx.b_element, &card->rx.e_offset, &hdr);
-		if (!skb) {
-			*done = 1;
-			break;
-		}
-
-		if (hdr->hdr.l2.id == QETH_HEADER_TYPE_LAYER2) {
-			skb->protocol = eth_type_trans(skb, skb->dev);
-			qeth_rx_csum(card, skb, hdr->hdr.l2.flags[1]);
-			len = skb->len;
-			napi_gro_receive(&card->napi, skb);
-		} else {
-			skb_push(skb, sizeof(*hdr));
-			skb_copy_to_linear_data(skb, hdr, sizeof(*hdr));
-			len = skb->len;
-			card->osn_info.data_cb(skb);
-		}
-
-		work_done++;
-		budget--;
-		QETH_CARD_STAT_INC(card, rx_packets);
-		QETH_CARD_STAT_ADD(card, rx_bytes, len);
-	}
-	return work_done;
-}
-
 static int qeth_l2_request_initial_mac(struct qeth_card *card)
 {
 	int rc = 0;
@@ -961,7 +922,6 @@ static int qeth_l2_control_event(struct qeth_card *card,
 
 struct qeth_discipline qeth_l2_discipline = {
 	.devtype = &qeth_l2_devtype,
-	.process_rx_buffer = qeth_l2_process_inbound_buffer,
 	.recover = qeth_l2_recover,
 	.setup = qeth_l2_probe_device,
 	.remove = qeth_l2_remove_device,
diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index 04d7777b7d16..8a1535ee1467 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -1164,96 +1164,6 @@ static int qeth_l3_vlan_rx_kill_vid(struct net_device *dev,
 	return 0;
 }
 
-static void qeth_l3_rebuild_skb(struct qeth_card *card, struct sk_buff *skb,
-				struct qeth_hdr *hdr)
-{
-	struct af_iucv_trans_hdr *iucv = (struct af_iucv_trans_hdr *) skb->data;
-	struct net_device *dev = skb->dev;
-
-	if (IS_IQD(card) && iucv->magic == ETH_P_AF_IUCV) {
-		dev_hard_header(skb, dev, ETH_P_AF_IUCV, dev->dev_addr,
-				"FAKELL", skb->len);
-		return;
-	}
-
-	if (!(hdr->hdr.l3.flags & QETH_HDR_PASSTHRU)) {
-		u16 prot = (hdr->hdr.l3.flags & QETH_HDR_IPV6) ? ETH_P_IPV6 :
-								 ETH_P_IP;
-		unsigned char tg_addr[ETH_ALEN];
-
-		skb_reset_network_header(skb);
-		switch (hdr->hdr.l3.flags & QETH_HDR_CAST_MASK) {
-		case QETH_CAST_MULTICAST:
-			if (prot == ETH_P_IP)
-				ip_eth_mc_map(ip_hdr(skb)->daddr, tg_addr);
-			else
-				ipv6_eth_mc_map(&ipv6_hdr(skb)->daddr, tg_addr);
-			QETH_CARD_STAT_INC(card, rx_multicast);
-			break;
-		case QETH_CAST_BROADCAST:
-			ether_addr_copy(tg_addr, card->dev->broadcast);
-			QETH_CARD_STAT_INC(card, rx_multicast);
-			break;
-		default:
-			if (card->options.sniffer)
-				skb->pkt_type = PACKET_OTHERHOST;
-			ether_addr_copy(tg_addr, card->dev->dev_addr);
-		}
-
-		if (hdr->hdr.l3.ext_flags & QETH_HDR_EXT_SRC_MAC_ADDR)
-			card->dev->header_ops->create(skb, card->dev, prot,
-				tg_addr, &hdr->hdr.l3.next_hop.rx.src_mac,
-				skb->len);
-		else
-			card->dev->header_ops->create(skb, card->dev, prot,
-				tg_addr, "FAKELL", skb->len);
-	}
-
-	/* copy VLAN tag from hdr into skb */
-	if (!card->options.sniffer &&
-	    (hdr->hdr.l3.ext_flags & (QETH_HDR_EXT_VLAN_FRAME |
-				      QETH_HDR_EXT_INCLUDE_VLAN_TAG))) {
-		u16 tag = (hdr->hdr.l3.ext_flags & QETH_HDR_EXT_VLAN_FRAME) ?
-				hdr->hdr.l3.vlan_id :
-				hdr->hdr.l3.next_hop.rx.vlan_id;
-		__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), tag);
-	}
-
-	qeth_rx_csum(card, skb, hdr->hdr.l3.ext_flags);
-}
-
-static int qeth_l3_process_inbound_buffer(struct qeth_card *card,
-				int budget, int *done)
-{
-	int work_done = 0;
-	struct sk_buff *skb;
-	struct qeth_hdr *hdr;
-
-	*done = 0;
-	WARN_ON_ONCE(!budget);
-	while (budget) {
-		skb = qeth_core_get_next_skb(card,
-			&card->qdio.in_q->bufs[card->rx.b_index],
-			&card->rx.b_element, &card->rx.e_offset, &hdr);
-		if (!skb) {
-			*done = 1;
-			break;
-		}
-
-		if (hdr->hdr.l3.id == QETH_HEADER_TYPE_LAYER3)
-			qeth_l3_rebuild_skb(card, skb, hdr);
-
-		skb->protocol = eth_type_trans(skb, skb->dev);
-		QETH_CARD_STAT_INC(card, rx_packets);
-		QETH_CARD_STAT_ADD(card, rx_bytes, skb->len);
-
-		napi_gro_receive(&card->napi, skb);
-		work_done++;
-		budget--;
-	}
-	return work_done;
-}
-
 static void qeth_l3_stop_card(struct qeth_card *card)
 {
 	QETH_CARD_TEXT(card, 2, "stopcard");
@@ -2317,7 +2227,6 @@ static int qeth_l3_control_event(struct qeth_card *card,
 
 struct qeth_discipline qeth_l3_discipline = {
 	.devtype = &qeth_l3_devtype,
-	.process_rx_buffer = qeth_l3_process_inbound_buffer,
 	.recover = qeth_l3_recover,
 	.setup = qeth_l3_probe_device,
 	.remove = qeth_l3_remove_device,
-- 
cgit v1.2.3-59-g8ed1b


From dcdcf8670dad7547cff59ef9ddd7312fc706464a Mon Sep 17 00:00:00 2001
From: Julian Wiedmann <jwi@linux.ibm.com>
Date: Mon, 23 Dec 2019 15:22:26 +0100
Subject: s390/qeth: use napi_gro_frags() for SG skbs

For non-linear packets, get the skb for attaching the page fragments
from napi_get_frags() so that it can be recycled during GRO.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/s390/net/qeth_core_main.c | 67 +++++++++++++++++++++++++++++----------
 1 file changed, 50 insertions(+), 17 deletions(-)

diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index 8e2c0588525f..b32b50384c5c 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -5106,8 +5106,9 @@ static void qeth_l3_rebuild_skb(struct qeth_card *card, struct sk_buff *skb,
 #endif
 
 static void qeth_receive_skb(struct qeth_card *card, struct sk_buff *skb,
-			     struct qeth_hdr *hdr)
+			     struct qeth_hdr *hdr, bool uses_frags)
 {
+	struct napi_struct *napi = &card->napi;
 	bool is_cso;
 
 	switch (hdr->hdr.l2.id) {
@@ -5130,7 +5131,10 @@ static void qeth_receive_skb(struct qeth_card *card, struct sk_buff *skb,
 		break;
 	default:
 		/* never happens */
-		dev_kfree_skb_any(skb);
+		if (uses_frags)
+			napi_free_frags(napi);
+		else
+			dev_kfree_skb_any(skb);
 		return;
 	}
 
@@ -5141,8 +5145,6 @@ static void qeth_receive_skb(struct qeth_card *card, struct sk_buff *skb,
 		skb->ip_summed = CHECKSUM_NONE;
 	}
 
-	skb->protocol = eth_type_trans(skb, skb->dev);
-
 	QETH_CARD_STAT_ADD(card, rx_bytes, skb->len);
 	QETH_CARD_STAT_INC(card, rx_packets);
 	if (skb_is_nonlinear(skb)) {
@@ -5151,7 +5153,12 @@ static void qeth_receive_skb(struct qeth_card *card, struct sk_buff *skb,
 				   skb_shinfo(skb)->nr_frags);
 	}
 
-	napi_gro_receive(&card->napi, skb);
+	if (uses_frags) {
+		napi_gro_frags(napi);
+	} else {
+		skb->protocol = eth_type_trans(skb, skb->dev);
+		napi_gro_receive(napi, skb);
+	}
 }
 
 static void qeth_create_skb_frag(struct sk_buff *skb, char *data, int data_len)
@@ -5177,7 +5184,9 @@ static int qeth_extract_skb(struct qeth_card *card,
 {
 	struct qdio_buffer_element *element = *__element;
 	struct qdio_buffer *buffer = qethbuffer->buffer;
+	struct napi_struct *napi = &card->napi;
 	unsigned int linear_len = 0;
+	bool uses_frags = false;
 	int offset = *__offset;
 	bool use_rx_sg = false;
 	unsigned int headroom;
@@ -5253,21 +5262,42 @@ next_packet:
 		     !atomic_read(&card->force_alloc_skb) &&
 		     !IS_OSN(card));
 
-	if (use_rx_sg && qethbuffer->rx_skb) {
+	if (use_rx_sg) {
 		/* QETH_CQ_ENABLED only: */
-		skb = qethbuffer->rx_skb;
-		qethbuffer->rx_skb = NULL;
-	} else {
-		if (!use_rx_sg)
-			linear_len = skb_len;
-		skb = napi_alloc_skb(&card->napi, linear_len + headroom);
+		if (qethbuffer->rx_skb) {
+			skb = qethbuffer->rx_skb;
+			qethbuffer->rx_skb = NULL;
+			goto use_skb;
+		}
+
+		skb = napi_get_frags(napi);
+		if (!skb) {
+			/* -ENOMEM, no point in falling back further. */
+			QETH_CARD_STAT_INC(card, rx_dropped_nomem);
+			goto walk_packet;
+		}
+
+		if (skb_tailroom(skb) >= linear_len + headroom) {
+			uses_frags = true;
+			goto use_skb;
+		}
+
+		netdev_info_once(card->dev,
+				 "Insufficient linear space in NAPI frags skb, need %u but have %u\n",
+				 linear_len + headroom, skb_tailroom(skb));
+		/* Shouldn't happen. Don't optimize, fall back to linear skb. */
 	}
 
-	if (!skb)
+	linear_len = skb_len;
+	skb = napi_alloc_skb(napi, linear_len + headroom);
+	if (!skb) {
 		QETH_CARD_STAT_INC(card, rx_dropped_nomem);
-	else if (headroom)
-		skb_reserve(skb, headroom);
+		goto walk_packet;
+	}
 
+use_skb:
+	if (headroom)
+		skb_reserve(skb, headroom);
 walk_packet:
 	while (skb_len) {
 		int data_len = min(skb_len, (int)(element->length - offset));
@@ -5300,7 +5330,10 @@ walk_packet:
 				QETH_CARD_TEXT(card, 4, "unexeob");
 				QETH_CARD_HEX(card, 2, buffer, sizeof(void *));
 				if (skb) {
-					dev_kfree_skb_any(skb);
+					if (uses_frags)
+						napi_free_frags(napi);
+					else
+						dev_kfree_skb_any(skb);
 					QETH_CARD_STAT_INC(card,
 							   rx_length_errors);
 				}
@@ -5318,7 +5351,7 @@ walk_packet:
 	*__element = element;
 	*__offset = offset;
 
-	qeth_receive_skb(card, skb, hdr);
+	qeth_receive_skb(card, skb, hdr, uses_frags);
 	return 0;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 8ca8559ff3cefd78e12b9658f21cc8bcc02407ec Mon Sep 17 00:00:00 2001
From: Julian Wiedmann <jwi@linux.ibm.com>
Date: Mon, 23 Dec 2019 15:22:27 +0100
Subject: s390/qeth: remove QETH_RX_PULL_LEN

Since commit f677fcb9aeb6 ("s390/qeth: ensure linear access to packet headers"),
the CQ-specific skbs are allocated with a slightly bigger linear part
than necessary. Shrink it down to the maximum that's needed by
qeth_extract_skb().

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/s390/net/qeth_core.h      | 1 -
 drivers/s390/net/qeth_core_main.c | 6 ++++--
 2 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h
index 4ab3be814ea7..7c37e94fb28b 100644
--- a/drivers/s390/net/qeth_core.h
+++ b/drivers/s390/net/qeth_core.h
@@ -221,7 +221,6 @@ struct qeth_vnicc_info {
 
 /* large receive scatter gather copy break */
 #define QETH_RX_SG_CB (PAGE_SIZE >> 1)
-#define QETH_RX_PULL_LEN 256
 
 struct qeth_hdr_layer3 {
 	__u8  id;
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index b32b50384c5c..3be3d13f8d65 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -2627,7 +2627,8 @@ static int qeth_init_input_buffer(struct qeth_card *card,
 
 	if ((card->options.cq == QETH_CQ_ENABLED) && (!buf->rx_skb)) {
 		buf->rx_skb = netdev_alloc_skb(card->dev,
-					       QETH_RX_PULL_LEN + ETH_HLEN);
+					       ETH_HLEN +
+					       sizeof(struct ipv6hdr));
 		if (!buf->rx_skb)
 			return 1;
 	}
@@ -5264,7 +5265,8 @@ next_packet:
 
 	if (use_rx_sg) {
 		/* QETH_CQ_ENABLED only: */
-		if (qethbuffer->rx_skb) {
+		if (qethbuffer->rx_skb &&
+		    skb_tailroom(qethbuffer->rx_skb) >= linear_len + headroom) {
 			skb = qethbuffer->rx_skb;
 			qethbuffer->rx_skb = NULL;
 			goto use_skb;
-- 
cgit v1.2.3-59-g8ed1b


From 7c8dce4b166113743adad131b5a24c4acc12f92c Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Thu, 26 Dec 2019 13:02:53 -0800
Subject: bpftool: Make skeleton C code compilable with C++ compiler

When auto-generated BPF skeleton C code is included from C++ application, it
triggers compilation error due to void * being implicitly casted to whatever
target pointer type. This is supported by C, but not C++. To solve this
problem, add explicit casts, where necessary.

To ensure issues like this are captured going forward, add skeleton usage in
test_cpp test.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191226210253.3132060-1-andriin@fb.com
---
 tools/bpf/bpftool/gen.c                  | 10 +++++-----
 tools/testing/selftests/bpf/Makefile     |  2 +-
 tools/testing/selftests/bpf/test_cpp.cpp | 10 ++++++++++
 3 files changed, 16 insertions(+), 6 deletions(-)

diff --git a/tools/bpf/bpftool/gen.c b/tools/bpf/bpftool/gen.c
index a14d8bc5d31d..7ce09a9a6999 100644
--- a/tools/bpf/bpftool/gen.c
+++ b/tools/bpf/bpftool/gen.c
@@ -397,7 +397,7 @@ static int do_skeleton(int argc, char **argv)
 		{							    \n\
 			struct %1$s *obj;				    \n\
 									    \n\
-			obj = calloc(1, sizeof(*obj));			    \n\
+			obj = (typeof(obj))calloc(1, sizeof(*obj));	    \n\
 			if (!obj)					    \n\
 				return NULL;				    \n\
 			if (%1$s__create_skeleton(obj))			    \n\
@@ -461,7 +461,7 @@ static int do_skeleton(int argc, char **argv)
 		{							    \n\
 			struct bpf_object_skeleton *s;			    \n\
 									    \n\
-			s = calloc(1, sizeof(*s));			    \n\
+			s = (typeof(s))calloc(1, sizeof(*s));		    \n\
 			if (!s)						    \n\
 				return -1;				    \n\
 			obj->skeleton = s;				    \n\
@@ -479,7 +479,7 @@ static int do_skeleton(int argc, char **argv)
 				/* maps */				    \n\
 				s->map_cnt = %zu;			    \n\
 				s->map_skel_sz = sizeof(*s->maps);	    \n\
-				s->maps = calloc(s->map_cnt, s->map_skel_sz);\n\
+				s->maps = (typeof(s->maps))calloc(s->map_cnt, s->map_skel_sz);\n\
 				if (!s->maps)				    \n\
 					goto err;			    \n\
 			",
@@ -515,7 +515,7 @@ static int do_skeleton(int argc, char **argv)
 				/* programs */				    \n\
 				s->prog_cnt = %zu;			    \n\
 				s->prog_skel_sz = sizeof(*s->progs);	    \n\
-				s->progs = calloc(s->prog_cnt, s->prog_skel_sz);\n\
+				s->progs = (typeof(s->progs))calloc(s->prog_cnt, s->prog_skel_sz);\n\
 				if (!s->progs)				    \n\
 					goto err;			    \n\
 			",
@@ -538,7 +538,7 @@ static int do_skeleton(int argc, char **argv)
 		\n\
 									    \n\
 			s->data_sz = %d;				    \n\
-			s->data = \"\\					    \n\
+			s->data = (void *)\"\\				    \n\
 		",
 		file_sz);
 
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index 866fc1cadd7c..41691fb067da 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -371,7 +371,7 @@ $(OUTPUT)/test_verifier: test_verifier.c verifier/tests.h $(BPFOBJ) | $(OUTPUT)
 	$(CC) $(CFLAGS) $(filter %.a %.o %.c,$^) $(LDLIBS) -o $@
 
 # Make sure we are able to include and link libbpf against c++.
-$(OUTPUT)/test_cpp: test_cpp.cpp $(BPFOBJ)
+$(OUTPUT)/test_cpp: test_cpp.cpp $(OUTPUT)/test_core_extern.skel.h $(BPFOBJ)
 	$(call msg,        CXX,,$@)
 	$(CXX) $(CFLAGS) $^ $(LDLIBS) -o $@
 
diff --git a/tools/testing/selftests/bpf/test_cpp.cpp b/tools/testing/selftests/bpf/test_cpp.cpp
index f0eb2727b766..6fe23a10d48a 100644
--- a/tools/testing/selftests/bpf/test_cpp.cpp
+++ b/tools/testing/selftests/bpf/test_cpp.cpp
@@ -1,12 +1,16 @@
 /* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */
+#include <iostream>
 #include "libbpf.h"
 #include "bpf.h"
 #include "btf.h"
+#include "test_core_extern.skel.h"
 
 /* do nothing, just make sure we can link successfully */
 
 int main(int argc, char *argv[])
 {
+	struct test_core_extern *skel;
+
 	/* libbpf.h */
 	libbpf_set_print(NULL);
 
@@ -16,5 +20,11 @@ int main(int argc, char *argv[])
 	/* btf.h */
 	btf__new(NULL, 0);
 
+	/* BPF skeleton */
+	skel = test_core_extern__open_and_load();
+	test_core_extern__destroy(skel);
+
+	std::cout << "DONE!" << std::endl;
+
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 473900a504e510cf9175876de8892ad1e3e7efab Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Mon, 23 Dec 2019 12:27:50 -0800
Subject: tcp_cubic: optimize hystart_update()

We do not care which bit in ca->found is set.

We avoid accessing hystart and hystart_detect unless really needed,
possibly avoiding one cache line miss.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/tcp_cubic.c | 9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)

diff --git a/net/ipv4/tcp_cubic.c b/net/ipv4/tcp_cubic.c
index 1b3d032a4df2..297936033c33 100644
--- a/net/ipv4/tcp_cubic.c
+++ b/net/ipv4/tcp_cubic.c
@@ -381,9 +381,6 @@ static void hystart_update(struct sock *sk, u32 delay)
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct bictcp *ca = inet_csk_ca(sk);
 
-	if (ca->found & hystart_detect)
-		return;
-
 	if (hystart_detect & HYSTART_ACK_TRAIN) {
 		u32 now = bictcp_clock();
 
@@ -391,7 +388,7 @@ static void hystart_update(struct sock *sk, u32 delay)
 		if ((s32)(now - ca->last_ack) <= hystart_ack_delta) {
 			ca->last_ack = now;
 			if ((s32)(now - ca->round_start) > ca->delay_min >> 4) {
-				ca->found |= HYSTART_ACK_TRAIN;
+				ca->found = 1;
 				NET_INC_STATS(sock_net(sk),
 					      LINUX_MIB_TCPHYSTARTTRAINDETECT);
 				NET_ADD_STATS(sock_net(sk),
@@ -412,7 +409,7 @@ static void hystart_update(struct sock *sk, u32 delay)
 		} else {
 			if (ca->curr_rtt > ca->delay_min +
 			    HYSTART_DELAY_THRESH(ca->delay_min >> 3)) {
-				ca->found |= HYSTART_DELAY;
+				ca->found = 1;
 				NET_INC_STATS(sock_net(sk),
 					      LINUX_MIB_TCPHYSTARTDELAYDETECT);
 				NET_ADD_STATS(sock_net(sk),
@@ -450,7 +447,7 @@ static void bictcp_acked(struct sock *sk, const struct ack_sample *sample)
 		ca->delay_min = delay;
 
 	/* hystart triggers when cwnd is larger than some threshold */
-	if (hystart && tcp_in_slow_start(tp) &&
+	if (!ca->found && hystart && tcp_in_slow_start(tp) &&
 	    tp->snd_cwnd >= hystart_low_window)
 		hystart_update(sk, delay);
 }
-- 
cgit v1.2.3-59-g8ed1b


From 35821fc2b41c51161e503bade3630603668dd3af Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Mon, 23 Dec 2019 12:27:51 -0800
Subject: tcp_cubic: remove one conditional from hystart_update()

If we initialize ca->curr_rtt to ~0U, we do not need to test
for zero value in hystart_update()

We only read ca->curr_rtt if at least HYSTART_MIN_SAMPLES have
been processed, and thus ca->curr_rtt will have a sane value.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/tcp_cubic.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/ipv4/tcp_cubic.c b/net/ipv4/tcp_cubic.c
index 297936033c33..5eff762acd7f 100644
--- a/net/ipv4/tcp_cubic.c
+++ b/net/ipv4/tcp_cubic.c
@@ -133,7 +133,7 @@ static inline void bictcp_hystart_reset(struct sock *sk)
 
 	ca->round_start = ca->last_ack = bictcp_clock();
 	ca->end_seq = tp->snd_nxt;
-	ca->curr_rtt = 0;
+	ca->curr_rtt = ~0U;
 	ca->sample_cnt = 0;
 }
 
@@ -402,7 +402,7 @@ static void hystart_update(struct sock *sk, u32 delay)
 	if (hystart_detect & HYSTART_DELAY) {
 		/* obtain the minimum delay of more than sampling packets */
 		if (ca->sample_cnt < HYSTART_MIN_SAMPLES) {
-			if (ca->curr_rtt == 0 || ca->curr_rtt > delay)
+			if (ca->curr_rtt > delay)
 				ca->curr_rtt = delay;
 
 			ca->sample_cnt++;
-- 
cgit v1.2.3-59-g8ed1b


From cff04e2da308c522f654237b45dd64248fe8d1fa Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Mon, 23 Dec 2019 12:27:52 -0800
Subject: tcp_cubic: switch bictcp_clock() to usec resolution

Current 1ms clock feeds ca->round_start, ca->delay_min,
ca->last_ack.

This is quite problematic for data-center flows, where delay_min
is way below 1 ms.

This means Hystart Train detection triggers every time jiffies value
is updated, since "((s32)(now - ca->round_start) > ca->delay_min >> 4)"
expression becomes true.

This kind of random behavior can be solved by reusing the existing
usec timestamp that TCP keeps in tp->tcp_mstamp

Note that a followup patch will tweak things a bit, because
during slow start, GRO aggregation on receivers naturally
increases the RTT as TSO packets gradually come to ~64KB size.

To recap, right after this patch CUBIC Hystart train detection
is more aggressive, since short RTT flows might exit slow start at
cwnd = 20, instead of being possibly unbounded.

Following patch will address this problem.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/tcp_cubic.c | 35 ++++++++++++++---------------------
 1 file changed, 14 insertions(+), 21 deletions(-)

diff --git a/net/ipv4/tcp_cubic.c b/net/ipv4/tcp_cubic.c
index 5eff762acd7f..068775b91fb5 100644
--- a/net/ipv4/tcp_cubic.c
+++ b/net/ipv4/tcp_cubic.c
@@ -40,8 +40,8 @@
 
 /* Number of delay samples for detecting the increase of delay */
 #define HYSTART_MIN_SAMPLES	8
-#define HYSTART_DELAY_MIN	(4U<<3)
-#define HYSTART_DELAY_MAX	(16U<<3)
+#define HYSTART_DELAY_MIN	(4000U)	/* 4 ms */
+#define HYSTART_DELAY_MAX	(16000U)	/* 16 ms */
 #define HYSTART_DELAY_THRESH(x)	clamp(x, HYSTART_DELAY_MIN, HYSTART_DELAY_MAX)
 
 static int fast_convergence __read_mostly = 1;
@@ -53,7 +53,7 @@ static int tcp_friendliness __read_mostly = 1;
 static int hystart __read_mostly = 1;
 static int hystart_detect __read_mostly = HYSTART_ACK_TRAIN | HYSTART_DELAY;
 static int hystart_low_window __read_mostly = 16;
-static int hystart_ack_delta __read_mostly = 2;
+static int hystart_ack_delta_us __read_mostly = 2000;
 
 static u32 cube_rtt_scale __read_mostly;
 static u32 beta_scale __read_mostly;
@@ -77,8 +77,8 @@ MODULE_PARM_DESC(hystart_detect, "hybrid slow start detection mechanisms"
 		 " 1: packet-train 2: delay 3: both packet-train and delay");
 module_param(hystart_low_window, int, 0644);
 MODULE_PARM_DESC(hystart_low_window, "lower bound cwnd for hybrid slow start");
-module_param(hystart_ack_delta, int, 0644);
-MODULE_PARM_DESC(hystart_ack_delta, "spacing between ack's indicating train (msecs)");
+module_param(hystart_ack_delta_us, int, 0644);
+MODULE_PARM_DESC(hystart_ack_delta_us, "spacing between ack's indicating train (usecs)");
 
 /* BIC TCP Parameters */
 struct bictcp {
@@ -89,7 +89,7 @@ struct bictcp {
 	u32	bic_origin_point;/* origin point of bic function */
 	u32	bic_K;		/* time to origin point
 				   from the beginning of the current epoch */
-	u32	delay_min;	/* min delay (msec << 3) */
+	u32	delay_min;	/* min delay (usec) */
 	u32	epoch_start;	/* beginning of an epoch */
 	u32	ack_cnt;	/* number of acks */
 	u32	tcp_cwnd;	/* estimated tcp cwnd */
@@ -117,13 +117,9 @@ static inline void bictcp_reset(struct bictcp *ca)
 	ca->found = 0;
 }
 
-static inline u32 bictcp_clock(void)
+static inline u32 bictcp_clock_us(const struct sock *sk)
 {
-#if HZ < 1000
-	return ktime_to_ms(ktime_get_real());
-#else
-	return jiffies_to_msecs(jiffies);
-#endif
+	return tcp_sk(sk)->tcp_mstamp;
 }
 
 static inline void bictcp_hystart_reset(struct sock *sk)
@@ -131,7 +127,7 @@ static inline void bictcp_hystart_reset(struct sock *sk)
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct bictcp *ca = inet_csk_ca(sk);
 
-	ca->round_start = ca->last_ack = bictcp_clock();
+	ca->round_start = ca->last_ack = bictcp_clock_us(sk);
 	ca->end_seq = tp->snd_nxt;
 	ca->curr_rtt = ~0U;
 	ca->sample_cnt = 0;
@@ -276,7 +272,7 @@ static inline void bictcp_update(struct bictcp *ca, u32 cwnd, u32 acked)
 	 */
 
 	t = (s32)(tcp_jiffies32 - ca->epoch_start);
-	t += msecs_to_jiffies(ca->delay_min >> 3);
+	t += usecs_to_jiffies(ca->delay_min);
 	/* change the unit from HZ to bictcp_HZ */
 	t <<= BICTCP_HZ;
 	do_div(t, HZ);
@@ -382,12 +378,12 @@ static void hystart_update(struct sock *sk, u32 delay)
 	struct bictcp *ca = inet_csk_ca(sk);
 
 	if (hystart_detect & HYSTART_ACK_TRAIN) {
-		u32 now = bictcp_clock();
+		u32 now = bictcp_clock_us(sk);
 
 		/* first detection parameter - ack-train detection */
-		if ((s32)(now - ca->last_ack) <= hystart_ack_delta) {
+		if ((s32)(now - ca->last_ack) <= hystart_ack_delta_us) {
 			ca->last_ack = now;
-			if ((s32)(now - ca->round_start) > ca->delay_min >> 4) {
+			if ((s32)(now - ca->round_start) > ca->delay_min >> 1) {
 				ca->found = 1;
 				NET_INC_STATS(sock_net(sk),
 					      LINUX_MIB_TCPHYSTARTTRAINDETECT);
@@ -421,9 +417,6 @@ static void hystart_update(struct sock *sk, u32 delay)
 	}
 }
 
-/* Track delayed acknowledgment ratio using sliding window
- * ratio = (15*ratio + sample) / 16
- */
 static void bictcp_acked(struct sock *sk, const struct ack_sample *sample)
 {
 	const struct tcp_sock *tp = tcp_sk(sk);
@@ -438,7 +431,7 @@ static void bictcp_acked(struct sock *sk, const struct ack_sample *sample)
 	if (ca->epoch_start && (s32)(tcp_jiffies32 - ca->epoch_start) < HZ)
 		return;
 
-	delay = (sample->rtt_us << 3) / USEC_PER_MSEC;
+	delay = sample->rtt_us;
 	if (delay == 0)
 		delay = 1;
 
-- 
cgit v1.2.3-59-g8ed1b


From 42f3a8aaae66d31d87850fb4b02979a0fc5dc541 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Mon, 23 Dec 2019 12:27:53 -0800
Subject: tcp_cubic: tweak Hystart detection for short RTT flows

After switching ca->delay_min to usec resolution, we exit
slow start prematurely for very low RTT flows, setting
snd_ssthresh to 20.

The reason is that delay_min is fed with RTT of small packet
trains. Then as cwnd is increased, TCP sends bigger TSO packets.

LRO/GRO aggregation and/or interrupt mitigation strategies
on receiver tend to inflate RTT samples.

Fix this by adding to delay_min the expected delay of
two TSO packets, given current pacing rate.

Tested:

Sender uses pfifo_fast qdisc

Before :
$ nstat -n;for f in {1..10}; do ./super_netperf 1 -H lpaa24 -l -4000000; done;nstat|egrep "Hystart"
  11348
  11707
  11562
  11428
  11773
  11534
   9878
  11693
  10597
  10968
TcpExtTCPHystartTrainDetect     10                 0.0
TcpExtTCPHystartTrainCwnd       200                0.0

After :
$ nstat -n;for f in {1..10}; do ./super_netperf 1 -H lpaa24 -l -4000000; done;nstat|egrep "Hystart"
  14877
  14517
  15797
  18466
  17376
  14833
  17558
  17933
  16039
  18059
TcpExtTCPHystartTrainDetect     10                 0.0
TcpExtTCPHystartTrainCwnd       1670               0.0

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/tcp_cubic.c | 23 +++++++++++++++++++++--
 1 file changed, 21 insertions(+), 2 deletions(-)

diff --git a/net/ipv4/tcp_cubic.c b/net/ipv4/tcp_cubic.c
index 068775b91fb5..0e5428ed04fe 100644
--- a/net/ipv4/tcp_cubic.c
+++ b/net/ipv4/tcp_cubic.c
@@ -436,8 +436,27 @@ static void bictcp_acked(struct sock *sk, const struct ack_sample *sample)
 		delay = 1;
 
 	/* first time call or link delay decreases */
-	if (ca->delay_min == 0 || ca->delay_min > delay)
-		ca->delay_min = delay;
+	if (ca->delay_min == 0 || ca->delay_min > delay) {
+		unsigned long rate = READ_ONCE(sk->sk_pacing_rate);
+
+		/* Account for TSO/GRO delays.
+		 * Otherwise short RTT flows could get too small ssthresh,
+		 * since during slow start we begin with small TSO packets
+		 * and could lower ca->delay_min too much.
+		 * Ideally even with a very small RTT we would like to have
+		 * at least one TSO packet being sent and received by GRO,
+		 * and another one in qdisc layer.
+		 * We apply another 100% factor because @rate is doubled at
+		 * this point.
+		 * We cap the cushion to 1ms.
+		 */
+		if (rate)
+			delay += min_t(u64, USEC_PER_MSEC,
+				       div64_ul((u64)GSO_MAX_SIZE *
+						4 * USEC_PER_SEC, rate));
+		if (ca->delay_min == 0 || ca->delay_min > delay)
+			ca->delay_min = delay;
+	}
 
 	/* hystart triggers when cwnd is larger than some threshold */
 	if (!ca->found && hystart && tcp_in_slow_start(tp) &&
-- 
cgit v1.2.3-59-g8ed1b


From ede656e8465839530c3287c7f54adf75dc2b9563 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Mon, 23 Dec 2019 12:27:54 -0800
Subject: tcp_cubic: make Hystart aware of pacing

For years we disabled Hystart ACK train detection at Google
because it was fooled by TCP pacing.

ACK train detection uses a simple heuristic, detecting if
we receive ACK past half the RTT, to exit slow start before
hitting the bottleneck and experience massive drops.

But pacing by design might delay packets up to RTT/2,
so we need to tweak the Hystart logic to be aware of this
extra delay.

Tested:
 Added a 100 usec delay at receiver.

Before:
nstat -n;for f in {1..10}; do ./super_netperf 1 -H lpaa24 -l -4000000; done;nstat|egrep "Hystart"
   9117
   7057
   9553
   8300
   7030
   6849
   9533
  10126
   6876
   8473
TcpExtTCPHystartTrainDetect     10                 0.0
TcpExtTCPHystartTrainCwnd       1230               0.0

After :
nstat -n;for f in {1..10}; do ./super_netperf 1 -H lpaa24 -l -4000000; done;nstat|egrep "Hystart"
   9845
  10103
  10866
  11096
  11936
  11487
  11773
  12188
  11066
  11894
TcpExtTCPHystartTrainDetect     10                 0.0
TcpExtTCPHystartTrainCwnd       6462               0.0

Disabling Hystart ACK Train detection gives similar numbers

echo 2 >/sys/module/tcp_cubic/parameters/hystart_detect
nstat -n;for f in {1..10}; do ./super_netperf 1 -H lpaa24 -l -4000000; done;nstat|egrep "Hystart"
  11173
  10954
  12455
  10627
  11578
  11583
  11222
  10880
  10665
  11366

Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/tcp_cubic.c | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/net/ipv4/tcp_cubic.c b/net/ipv4/tcp_cubic.c
index 0e5428ed04fe..d02bb283c689 100644
--- a/net/ipv4/tcp_cubic.c
+++ b/net/ipv4/tcp_cubic.c
@@ -376,6 +376,7 @@ static void hystart_update(struct sock *sk, u32 delay)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct bictcp *ca = inet_csk_ca(sk);
+	u32 threshold;
 
 	if (hystart_detect & HYSTART_ACK_TRAIN) {
 		u32 now = bictcp_clock_us(sk);
@@ -383,7 +384,17 @@ static void hystart_update(struct sock *sk, u32 delay)
 		/* first detection parameter - ack-train detection */
 		if ((s32)(now - ca->last_ack) <= hystart_ack_delta_us) {
 			ca->last_ack = now;
-			if ((s32)(now - ca->round_start) > ca->delay_min >> 1) {
+
+			threshold = ca->delay_min;
+			/* Hystart ack train triggers if we get ack past
+			 * ca->delay_min/2.
+			 * Pacing might have delayed packets up to RTT/2
+			 * during slow start.
+			 */
+			if (sk->sk_pacing_status == SK_PACING_NONE)
+				threshold >>= 1;
+
+			if ((s32)(now - ca->round_start) > threshold) {
 				ca->found = 1;
 				NET_INC_STATS(sock_net(sk),
 					      LINUX_MIB_TCPHYSTARTTRAINDETECT);
-- 
cgit v1.2.3-59-g8ed1b


From 8a3f44a0bb76c4e69dea3ad398ada41ae4ffc3ea Mon Sep 17 00:00:00 2001
From: Xu Wang <vulab@iscas.ac.cn>
Date: Tue, 24 Dec 2019 09:37:04 +0000
Subject: ppp: Remove redundant BUG_ON() check in ppp_pernet

Passing NULL to ppp_pernet causes a crash via BUG_ON.
Dereferencing net in net_generic() also has the same effect.
This patch removes the redundant BUG_ON check on the same parameter.

Signed-off-by: Xu Wang <vulab@iscas.ac.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ppp/ppp_generic.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/ppp/ppp_generic.c b/drivers/net/ppp/ppp_generic.c
index 3bf8a8b42983..22cc2cb9d878 100644
--- a/drivers/net/ppp/ppp_generic.c
+++ b/drivers/net/ppp/ppp_generic.c
@@ -296,8 +296,6 @@ static struct class *ppp_class;
 /* per net-namespace data */
 static inline struct ppp_net *ppp_pernet(struct net *net)
 {
-	BUG_ON(!net);
-
 	return net_generic(net, ppp_net_id);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From c8f957df6e338406128cadb2ee30e38f9a816ed6 Mon Sep 17 00:00:00 2001
From: Mao Wenan <maowenan@huawei.com>
Date: Tue, 24 Dec 2019 19:58:12 +0800
Subject: net: dsa: qca: ar9331: drop pointless static qualifier in
 ar9331_sw_mbus_init

There is no need to set variable 'mbus' static
since new value always be assigned before use it.

Signed-off-by: Mao Wenan <maowenan@huawei.com>
Reviewed-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/qca/ar9331.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/dsa/qca/ar9331.c b/drivers/net/dsa/qca/ar9331.c
index 0d1a7cd85fe8..da3bece75e21 100644
--- a/drivers/net/dsa/qca/ar9331.c
+++ b/drivers/net/dsa/qca/ar9331.c
@@ -266,7 +266,7 @@ error:
 static int ar9331_sw_mbus_init(struct ar9331_sw_priv *priv)
 {
 	struct device *dev = priv->dev;
-	static struct mii_bus *mbus;
+	struct mii_bus *mbus;
 	struct device_node *np, *mnp;
 	int ret;
 
-- 
cgit v1.2.3-59-g8ed1b


From cad451dd2427d6ddd38a411e688def499daa798d Mon Sep 17 00:00:00 2001
From: YueHaibing <yuehaibing@huawei.com>
Date: Tue, 24 Dec 2019 20:51:28 +0800
Subject: net: ena: remove set but not used variable 'rx_ring'

drivers/net/ethernet/amazon/ena/ena_netdev.c: In function ena_xdp_xmit_buff:
drivers/net/ethernet/amazon/ena/ena_netdev.c:316:19: warning:
 variable rx_ring set but not used [-Wunused-but-set-variable]

commit 548c4940b9f1 ("net: ena: Implement XDP_TX action")
left behind this unused variable.

Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/amazon/ena/ena_netdev.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c
index 081acf077bb5..894e8c1a8cf1 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.c
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c
@@ -313,7 +313,6 @@ static int ena_xdp_xmit_buff(struct net_device *dev,
 	struct ena_com_tx_ctx ena_tx_ctx = {0};
 	struct ena_tx_buffer *tx_info;
 	struct ena_ring *xdp_ring;
-	struct ena_ring *rx_ring;
 	u16 next_to_use, req_id;
 	int rc;
 	void *push_hdr;
@@ -324,8 +323,6 @@ static int ena_xdp_xmit_buff(struct net_device *dev,
 	req_id = xdp_ring->free_ids[next_to_use];
 	tx_info = &xdp_ring->tx_buffer_info[req_id];
 	tx_info->num_of_bufs = 0;
-	rx_ring = &xdp_ring->adapter->rx_ring[qid -
-		  xdp_ring->adapter->xdp_first_ring];
 	page_ref_inc(rx_info->page);
 	tx_info->xdp_rx_page = rx_info->page;
 
-- 
cgit v1.2.3-59-g8ed1b


From 0df952873636a6df4b7469a4eb4da1e477031e3a Mon Sep 17 00:00:00 2001
From: Nikita Yushchenko <nikita.yoush@cogentembedded.com>
Date: Wed, 25 Dec 2019 08:22:38 +0300
Subject: mv88e6xxx: Add serdes Rx statistics

If packet checker is enabled in the serdes, then Rx counter registers
start working, and no side effects have been detected.

This patch enables packet checker automatically when powering serdes on,
and exposes Rx counter registers via ethtool statistics interface.

Code partially basded by older attempt by Andrew Lunn.

Signed-off-by: Nikita Yushchenko <nikita.yoush@cogentembedded.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/mv88e6xxx/chip.c   |   3 ++
 drivers/net/dsa/mv88e6xxx/serdes.c | 100 +++++++++++++++++++++++++++++++++++--
 drivers/net/dsa/mv88e6xxx/serdes.h |   9 ++++
 3 files changed, 109 insertions(+), 3 deletions(-)

diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c
index 3bd988529178..5eeeb6566196 100644
--- a/drivers/net/dsa/mv88e6xxx/chip.c
+++ b/drivers/net/dsa/mv88e6xxx/chip.c
@@ -4424,6 +4424,9 @@ static const struct mv88e6xxx_ops mv88e6390_ops = {
 	.gpio_ops = &mv88e6352_gpio_ops,
 	.avb_ops = &mv88e6390_avb_ops,
 	.ptp_ops = &mv88e6352_ptp_ops,
+	.serdes_get_sset_count = mv88e6390_serdes_get_sset_count,
+	.serdes_get_strings = mv88e6390_serdes_get_strings,
+	.serdes_get_stats = mv88e6390_serdes_get_stats,
 	.phylink_validate = mv88e6390_phylink_validate,
 };
 
diff --git a/drivers/net/dsa/mv88e6xxx/serdes.c b/drivers/net/dsa/mv88e6xxx/serdes.c
index 902feb398746..8d8b3b74aee1 100644
--- a/drivers/net/dsa/mv88e6xxx/serdes.c
+++ b/drivers/net/dsa/mv88e6xxx/serdes.c
@@ -405,22 +405,116 @@ static int mv88e6390_serdes_power_sgmii(struct mv88e6xxx_chip *chip, u8 lane,
 	return err;
 }
 
+struct mv88e6390_serdes_hw_stat {
+	char string[ETH_GSTRING_LEN];
+	int reg;
+};
+
+static struct mv88e6390_serdes_hw_stat mv88e6390_serdes_hw_stats[] = {
+	{ "serdes_rx_pkts", 0xf021 },
+	{ "serdes_rx_bytes", 0xf024 },
+	{ "serdes_rx_pkts_error", 0xf027 },
+};
+
+int mv88e6390_serdes_get_sset_count(struct mv88e6xxx_chip *chip, int port)
+{
+	if (mv88e6390_serdes_get_lane(chip, port) == 0)
+		return 0;
+
+	return ARRAY_SIZE(mv88e6390_serdes_hw_stats);
+}
+
+int mv88e6390_serdes_get_strings(struct mv88e6xxx_chip *chip,
+				 int port, uint8_t *data)
+{
+	struct mv88e6390_serdes_hw_stat *stat;
+	int i;
+
+	if (mv88e6390_serdes_get_lane(chip, port) == 0)
+		return 0;
+
+	for (i = 0; i < ARRAY_SIZE(mv88e6390_serdes_hw_stats); i++) {
+		stat = &mv88e6390_serdes_hw_stats[i];
+		memcpy(data + i * ETH_GSTRING_LEN, stat->string,
+		       ETH_GSTRING_LEN);
+	}
+	return ARRAY_SIZE(mv88e6390_serdes_hw_stats);
+}
+
+static uint64_t mv88e6390_serdes_get_stat(struct mv88e6xxx_chip *chip, int lane,
+					  struct mv88e6390_serdes_hw_stat *stat)
+{
+	u16 reg[3];
+	int err, i;
+
+	for (i = 0; i < 3; i++) {
+		err = mv88e6390_serdes_read(chip, lane, MDIO_MMD_PHYXS,
+					    stat->reg + i, &reg[i]);
+		if (err) {
+			dev_err(chip->dev, "failed to read statistic\n");
+			return 0;
+		}
+	}
+
+	return reg[0] | ((u64)reg[1] << 16) | ((u64)reg[2] << 32);
+}
+
+int mv88e6390_serdes_get_stats(struct mv88e6xxx_chip *chip, int port,
+			       uint64_t *data)
+{
+	struct mv88e6390_serdes_hw_stat *stat;
+	int lane;
+	int i;
+
+	lane = mv88e6390_serdes_get_lane(chip, port);
+	if (lane == 0)
+		return 0;
+
+	for (i = 0; i < ARRAY_SIZE(mv88e6390_serdes_hw_stats); i++) {
+		stat = &mv88e6390_serdes_hw_stats[i];
+		data[i] = mv88e6390_serdes_get_stat(chip, lane, stat);
+	}
+
+	return ARRAY_SIZE(mv88e6390_serdes_hw_stats);
+}
+
+static int mv88e6390_serdes_enable_checker(struct mv88e6xxx_chip *chip, u8 lane)
+{
+	u16 reg;
+	int err;
+
+	err = mv88e6390_serdes_read(chip, lane, MDIO_MMD_PHYXS,
+				    MV88E6390_PG_CONTROL, &reg);
+	if (err)
+		return err;
+
+	reg |= MV88E6390_PG_CONTROL_ENABLE_PC;
+	return mv88e6390_serdes_write(chip, lane, MDIO_MMD_PHYXS,
+				      MV88E6390_PG_CONTROL, reg);
+}
+
 int mv88e6390_serdes_power(struct mv88e6xxx_chip *chip, int port, u8 lane,
 			   bool up)
 {
 	u8 cmode = chip->ports[port].cmode;
+	int err = 0;
 
 	switch (cmode) {
 	case MV88E6XXX_PORT_STS_CMODE_SGMII:
 	case MV88E6XXX_PORT_STS_CMODE_1000BASEX:
 	case MV88E6XXX_PORT_STS_CMODE_2500BASEX:
-		return mv88e6390_serdes_power_sgmii(chip, lane, up);
+		err = mv88e6390_serdes_power_sgmii(chip, lane, up);
+		break;
 	case MV88E6XXX_PORT_STS_CMODE_XAUI:
 	case MV88E6XXX_PORT_STS_CMODE_RXAUI:
-		return mv88e6390_serdes_power_10g(chip, lane, up);
+		err = mv88e6390_serdes_power_10g(chip, lane, up);
+		break;
 	}
 
-	return 0;
+	if (!err && up)
+		err = mv88e6390_serdes_enable_checker(chip, lane);
+
+	return err;
 }
 
 static void mv88e6390_serdes_irq_link_sgmii(struct mv88e6xxx_chip *chip,
diff --git a/drivers/net/dsa/mv88e6xxx/serdes.h b/drivers/net/dsa/mv88e6xxx/serdes.h
index bd8df36ab537..d16ef4da20b0 100644
--- a/drivers/net/dsa/mv88e6xxx/serdes.h
+++ b/drivers/net/dsa/mv88e6xxx/serdes.h
@@ -74,6 +74,10 @@
 #define MV88E6390_SGMII_PHY_STATUS_SPD_DPL_VALID BIT(11)
 #define MV88E6390_SGMII_PHY_STATUS_LINK		BIT(10)
 
+/* Packet generator pad packet checker */
+#define MV88E6390_PG_CONTROL		0xf010
+#define MV88E6390_PG_CONTROL_ENABLE_PC		BIT(0)
+
 u8 mv88e6341_serdes_get_lane(struct mv88e6xxx_chip *chip, int port);
 u8 mv88e6352_serdes_get_lane(struct mv88e6xxx_chip *chip, int port);
 u8 mv88e6390_serdes_get_lane(struct mv88e6xxx_chip *chip, int port);
@@ -99,6 +103,11 @@ int mv88e6352_serdes_get_strings(struct mv88e6xxx_chip *chip,
 				 int port, uint8_t *data);
 int mv88e6352_serdes_get_stats(struct mv88e6xxx_chip *chip, int port,
 			       uint64_t *data);
+int mv88e6390_serdes_get_sset_count(struct mv88e6xxx_chip *chip, int port);
+int mv88e6390_serdes_get_strings(struct mv88e6xxx_chip *chip,
+				 int port, uint8_t *data);
+int mv88e6390_serdes_get_stats(struct mv88e6xxx_chip *chip, int port,
+			       uint64_t *data);
 
 /* Return the (first) SERDES lane address a port is using, 0 otherwise. */
 static inline u8 mv88e6xxx_serdes_get_lane(struct mv88e6xxx_chip *chip,
-- 
cgit v1.2.3-59-g8ed1b


From 356b23c073dd063427102329b296061855b912d9 Mon Sep 17 00:00:00 2001
From: Kevin Kou <qdkevin.kou@gmail.com>
Date: Wed, 25 Dec 2019 08:27:25 +0000
Subject: sctp: do trace_sctp_probe after SACK validation and check

The function sctp_sf_eat_sack_6_2 now performs the Verification
Tag validation, Chunk length validation, Bogu check, and also
the detection of out-of-order SACK based on the RFC2960
Section 6.2 at the beginning, and finally performs the further
processing of SACK. The trace_sctp_probe now triggered before
the above necessary validation and check.

this patch is to do the trace_sctp_probe after the chunk sanity
tests, but keep doing trace if the SACK received is out of order,
for the out-of-order SACK is valuable to congestion control
debugging.

v1->v2:
 - keep doing SCTP trace if the SACK is out of order as Marcelo's
   suggestion.
v2->v3:
 - regenerate the patch as v2 generated on top of v1, and add
   'net-next' tag to the new one as Marcelo's comments.

Signed-off-by: Kevin Kou <qdkevin.kou@gmail.com>
Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sctp/sm_statefuns.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c
index 42558fa3f3e4..748e3b19ec1d 100644
--- a/net/sctp/sm_statefuns.c
+++ b/net/sctp/sm_statefuns.c
@@ -3281,8 +3281,6 @@ enum sctp_disposition sctp_sf_eat_sack_6_2(struct net *net,
 	struct sctp_sackhdr *sackh;
 	__u32 ctsn;
 
-	trace_sctp_probe(ep, asoc, chunk);
-
 	if (!sctp_vtag_verify(chunk, asoc))
 		return sctp_sf_pdiscard(net, ep, asoc, type, arg, commands);
 
@@ -3299,6 +3297,15 @@ enum sctp_disposition sctp_sf_eat_sack_6_2(struct net *net,
 	chunk->subh.sack_hdr = sackh;
 	ctsn = ntohl(sackh->cum_tsn_ack);
 
+	/* If Cumulative TSN Ack beyond the max tsn currently
+	 * send, terminating the association and respond to the
+	 * sender with an ABORT.
+	 */
+	if (TSN_lte(asoc->next_tsn, ctsn))
+		return sctp_sf_violation_ctsn(net, ep, asoc, type, arg, commands);
+
+	trace_sctp_probe(ep, asoc, chunk);
+
 	/* i) If Cumulative TSN Ack is less than the Cumulative TSN
 	 *     Ack Point, then drop the SACK.  Since Cumulative TSN
 	 *     Ack is monotonically increasing, a SACK whose
@@ -3312,13 +3319,6 @@ enum sctp_disposition sctp_sf_eat_sack_6_2(struct net *net,
 		return SCTP_DISPOSITION_DISCARD;
 	}
 
-	/* If Cumulative TSN Ack beyond the max tsn currently
-	 * send, terminating the association and respond to the
-	 * sender with an ABORT.
-	 */
-	if (!TSN_lt(ctsn, asoc->next_tsn))
-		return sctp_sf_violation_ctsn(net, ep, asoc, type, arg, commands);
-
 	/* Return this SACK for further processing.  */
 	sctp_add_cmd_sf(commands, SCTP_CMD_PROCESS_SACK, SCTP_CHUNK(chunk));
 
-- 
cgit v1.2.3-59-g8ed1b


From 2b4a8990b7df55875745a80a609a1ceaaf51f322 Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Fri, 27 Dec 2019 15:55:18 +0100
Subject: ethtool: introduce ethtool netlink interface

Basic genetlink and init infrastructure for the netlink interface, register
genetlink family "ethtool". Add CONFIG_ETHTOOL_NETLINK Kconfig option to
make the build optional. Add initial overall interface description into
Documentation/networking/ethtool-netlink.rst, further patches will add more
detailed information.

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/ethtool-netlink.rst | 216 +++++++++++++++++++++++++++
 Documentation/networking/index.rst           |   1 +
 include/linux/ethtool_netlink.h              |   9 ++
 include/uapi/linux/ethtool_netlink.h         |  36 +++++
 net/Kconfig                                  |   8 +
 net/ethtool/Makefile                         |   6 +-
 net/ethtool/netlink.c                        |  33 ++++
 net/ethtool/netlink.h                        |  10 ++
 8 files changed, 318 insertions(+), 1 deletion(-)
 create mode 100644 Documentation/networking/ethtool-netlink.rst
 create mode 100644 include/linux/ethtool_netlink.h
 create mode 100644 include/uapi/linux/ethtool_netlink.h
 create mode 100644 net/ethtool/netlink.c
 create mode 100644 net/ethtool/netlink.h

diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
new file mode 100644
index 000000000000..9448442ad293
--- /dev/null
+++ b/Documentation/networking/ethtool-netlink.rst
@@ -0,0 +1,216 @@
+=============================
+Netlink interface for ethtool
+=============================
+
+
+Basic information
+=================
+
+Netlink interface for ethtool uses generic netlink family ``ethtool``
+(userspace application should use macros ``ETHTOOL_GENL_NAME`` and
+``ETHTOOL_GENL_VERSION`` defined in ``<linux/ethtool_netlink.h>`` uapi
+header). This family does not use a specific header, all information in
+requests and replies is passed using netlink attributes.
+
+The ethtool netlink interface uses extended ACK for error and warning
+reporting, userspace application developers are encouraged to make these
+messages available to user in a suitable way.
+
+Requests can be divided into three categories: "get" (retrieving information),
+"set" (setting parameters) and "action" (invoking an action).
+
+All "set" and "action" type requests require admin privileges
+(``CAP_NET_ADMIN`` in the namespace). Most "get" type requests are allowed for
+anyone but there are exceptions (where the response contains sensitive
+information). In some cases, the request as such is allowed for anyone but
+unprivileged users have attributes with sensitive information (e.g.
+wake-on-lan password) omitted.
+
+
+Conventions
+===========
+
+Attributes which represent a boolean value usually use NLA_U8 type so that we
+can distinguish three states: "on", "off" and "not present" (meaning the
+information is not available in "get" requests or value is not to be changed
+in "set" requests). For these attributes, the "true" value should be passed as
+number 1 but any non-zero value should be understood as "true" by recipient.
+In the tables below, "bool" denotes NLA_U8 attributes interpreted in this way.
+
+In the message structure descriptions below, if an attribute name is suffixed
+with "+", parent nest can contain multiple attributes of the same type. This
+implements an array of entries.
+
+
+Request header
+==============
+
+Each request or reply message contains a nested attribute with common header.
+Structure of this header is
+
+  ==============================  ======  =============================
+  ``ETHTOOL_A_HEADER_DEV_INDEX``  u32     device ifindex
+  ``ETHTOOL_A_HEADER_DEV_NAME``   string  device name
+  ``ETHTOOL_A_HEADER_FLAGS``      u32     flags common for all requests
+  ==============================  ======  =============================
+
+``ETHTOOL_A_HEADER_DEV_INDEX`` and ``ETHTOOL_A_HEADER_DEV_NAME`` identify the
+device message relates to. One of them is sufficient in requests, if both are
+used, they must identify the same device. Some requests, e.g. global string
+sets, do not require device identification. Most ``GET`` requests also allow
+dump requests without device identification to query the same information for
+all devices providing it (each device in a separate message).
+
+``ETHTOOL_A_HEADER_FLAGS`` is a bitmap of request flags common for all request
+types. The interpretation of these flags is the same for all request types but
+the flags may not apply to requests. Recognized flags are:
+
+  =================================  ===================================
+  ``ETHTOOL_FLAG_COMPACT_BITSETS``   use compact format bitsets in reply
+  ``ETHTOOL_FLAG_OMIT_REPLY``        omit optional reply (_SET and _ACT)
+  =================================  ===================================
+
+New request flags should follow the general idea that if the flag is not set,
+the behaviour is backward compatible, i.e. requests from old clients not aware
+of the flag should be interpreted the way the client expects. A client must
+not set flags it does not understand.
+
+
+List of message types
+=====================
+
+All constants identifying message types use ``ETHTOOL_CMD_`` prefix and suffix
+according to message purpose:
+
+  ==============    ======================================
+  ``_GET``          userspace request to retrieve data
+  ``_SET``          userspace request to set data
+  ``_ACT``          userspace request to perform an action
+  ``_GET_REPLY``    kernel reply to a ``GET`` request
+  ``_SET_REPLY``    kernel reply to a ``SET`` request
+  ``_ACT_REPLY``    kernel reply to an ``ACT`` request
+  ``_NTF``          kernel notification
+  ==============    ======================================
+
+``GET`` requests are sent by userspace applications to retrieve device
+information. They usually do not contain any message specific attributes.
+Kernel replies with corresponding "GET_REPLY" message. For most types, ``GET``
+request with ``NLM_F_DUMP`` and no device identification can be used to query
+the information for all devices supporting the request.
+
+If the data can be also modified, corresponding ``SET`` message with the same
+layout as corresponding ``GET_REPLY`` is used to request changes. Only
+attributes where a change is requested are included in such request (also, not
+all attributes may be changed). Replies to most ``SET`` request consist only
+of error code and extack; if kernel provides additional data, it is sent in
+the form of corresponding ``SET_REPLY`` message which can be suppressed by
+setting ``ETHTOOL_FLAG_OMIT_REPLY`` flag in request header.
+
+Data modification also triggers sending a ``NTF`` message with a notification.
+These usually bear only a subset of attributes which was affected by the
+change. The same notification is issued if the data is modified using other
+means (mostly ioctl ethtool interface). Unlike notifications from ethtool
+netlink code which are only sent if something actually changed, notifications
+triggered by ioctl interface may be sent even if the request did not actually
+change any data.
+
+``ACT`` messages request kernel (driver) to perform a specific action. If some
+information is reported by kernel (which can be suppressed by setting
+``ETHTOOL_FLAG_OMIT_REPLY`` flag in request header), the reply takes form of
+an ``ACT_REPLY`` message. Performing an action also triggers a notification
+(``NTF`` message).
+
+Later sections describe the format and semantics of these messages.
+
+
+Request translation
+===================
+
+The following table maps ioctl commands to netlink commands providing their
+functionality. Entries with "n/a" in right column are commands which do not
+have their netlink replacement yet.
+
+  =================================== =====================================
+  ioctl command                       netlink command
+  =================================== =====================================
+  ``ETHTOOL_GSET``                    n/a
+  ``ETHTOOL_SSET``                    n/a
+  ``ETHTOOL_GDRVINFO``                n/a
+  ``ETHTOOL_GREGS``                   n/a
+  ``ETHTOOL_GWOL``                    n/a
+  ``ETHTOOL_SWOL``                    n/a
+  ``ETHTOOL_GMSGLVL``                 n/a
+  ``ETHTOOL_SMSGLVL``                 n/a
+  ``ETHTOOL_NWAY_RST``                n/a
+  ``ETHTOOL_GLINK``                   n/a
+  ``ETHTOOL_GEEPROM``                 n/a
+  ``ETHTOOL_SEEPROM``                 n/a
+  ``ETHTOOL_GCOALESCE``               n/a
+  ``ETHTOOL_SCOALESCE``               n/a
+  ``ETHTOOL_GRINGPARAM``              n/a
+  ``ETHTOOL_SRINGPARAM``              n/a
+  ``ETHTOOL_GPAUSEPARAM``             n/a
+  ``ETHTOOL_SPAUSEPARAM``             n/a
+  ``ETHTOOL_GRXCSUM``                 n/a
+  ``ETHTOOL_SRXCSUM``                 n/a
+  ``ETHTOOL_GTXCSUM``                 n/a
+  ``ETHTOOL_STXCSUM``                 n/a
+  ``ETHTOOL_GSG``                     n/a
+  ``ETHTOOL_SSG``                     n/a
+  ``ETHTOOL_TEST``                    n/a
+  ``ETHTOOL_GSTRINGS``                n/a
+  ``ETHTOOL_PHYS_ID``                 n/a
+  ``ETHTOOL_GSTATS``                  n/a
+  ``ETHTOOL_GTSO``                    n/a
+  ``ETHTOOL_STSO``                    n/a
+  ``ETHTOOL_GPERMADDR``               rtnetlink ``RTM_GETLINK``
+  ``ETHTOOL_GUFO``                    n/a
+  ``ETHTOOL_SUFO``                    n/a
+  ``ETHTOOL_GGSO``                    n/a
+  ``ETHTOOL_SGSO``                    n/a
+  ``ETHTOOL_GFLAGS``                  n/a
+  ``ETHTOOL_SFLAGS``                  n/a
+  ``ETHTOOL_GPFLAGS``                 n/a
+  ``ETHTOOL_SPFLAGS``                 n/a
+  ``ETHTOOL_GRXFH``                   n/a
+  ``ETHTOOL_SRXFH``                   n/a
+  ``ETHTOOL_GGRO``                    n/a
+  ``ETHTOOL_SGRO``                    n/a
+  ``ETHTOOL_GRXRINGS``                n/a
+  ``ETHTOOL_GRXCLSRLCNT``             n/a
+  ``ETHTOOL_GRXCLSRULE``              n/a
+  ``ETHTOOL_GRXCLSRLALL``             n/a
+  ``ETHTOOL_SRXCLSRLDEL``             n/a
+  ``ETHTOOL_SRXCLSRLINS``             n/a
+  ``ETHTOOL_FLASHDEV``                n/a
+  ``ETHTOOL_RESET``                   n/a
+  ``ETHTOOL_SRXNTUPLE``               n/a
+  ``ETHTOOL_GRXNTUPLE``               n/a
+  ``ETHTOOL_GSSET_INFO``              n/a
+  ``ETHTOOL_GRXFHINDIR``              n/a
+  ``ETHTOOL_SRXFHINDIR``              n/a
+  ``ETHTOOL_GFEATURES``               n/a
+  ``ETHTOOL_SFEATURES``               n/a
+  ``ETHTOOL_GCHANNELS``               n/a
+  ``ETHTOOL_SCHANNELS``               n/a
+  ``ETHTOOL_SET_DUMP``                n/a
+  ``ETHTOOL_GET_DUMP_FLAG``           n/a
+  ``ETHTOOL_GET_DUMP_DATA``           n/a
+  ``ETHTOOL_GET_TS_INFO``             n/a
+  ``ETHTOOL_GMODULEINFO``             n/a
+  ``ETHTOOL_GMODULEEEPROM``           n/a
+  ``ETHTOOL_GEEE``                    n/a
+  ``ETHTOOL_SEEE``                    n/a
+  ``ETHTOOL_GRSSH``                   n/a
+  ``ETHTOOL_SRSSH``                   n/a
+  ``ETHTOOL_GTUNABLE``                n/a
+  ``ETHTOOL_STUNABLE``                n/a
+  ``ETHTOOL_GPHYSTATS``               n/a
+  ``ETHTOOL_PERQUEUE``                n/a
+  ``ETHTOOL_GLINKSETTINGS``           n/a
+  ``ETHTOOL_SLINKSETTINGS``           n/a
+  ``ETHTOOL_PHY_GTUNABLE``            n/a
+  ``ETHTOOL_PHY_STUNABLE``            n/a
+  ``ETHTOOL_GFECPARAM``               n/a
+  ``ETHTOOL_SFECPARAM``               n/a
+  =================================== =====================================
diff --git a/Documentation/networking/index.rst b/Documentation/networking/index.rst
index 5acab1290e03..bee73be7af93 100644
--- a/Documentation/networking/index.rst
+++ b/Documentation/networking/index.rst
@@ -16,6 +16,7 @@ Contents:
    devlink-info-versions
    devlink-trap
    devlink-trap-netdevsim
+   ethtool-netlink
    ieee802154
    j1939
    kapi
diff --git a/include/linux/ethtool_netlink.h b/include/linux/ethtool_netlink.h
new file mode 100644
index 000000000000..f27e92b5f344
--- /dev/null
+++ b/include/linux/ethtool_netlink.h
@@ -0,0 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+
+#ifndef _LINUX_ETHTOOL_NETLINK_H_
+#define _LINUX_ETHTOOL_NETLINK_H_
+
+#include <uapi/linux/ethtool_netlink.h>
+#include <linux/ethtool.h>
+
+#endif /* _LINUX_ETHTOOL_NETLINK_H_ */
diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
new file mode 100644
index 000000000000..3c93276ba066
--- /dev/null
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -0,0 +1,36 @@
+/* SPDX-License-Identifier: GPL-2.0-only WITH Linux-syscall-note */
+/*
+ * include/uapi/linux/ethtool_netlink.h - netlink interface for ethtool
+ *
+ * See Documentation/networking/ethtool-netlink.txt in kernel source tree for
+ * doucumentation of the interface.
+ */
+
+#ifndef _UAPI_LINUX_ETHTOOL_NETLINK_H_
+#define _UAPI_LINUX_ETHTOOL_NETLINK_H_
+
+#include <linux/ethtool.h>
+
+/* message types - userspace to kernel */
+enum {
+	ETHTOOL_MSG_USER_NONE,
+
+	/* add new constants above here */
+	__ETHTOOL_MSG_USER_CNT,
+	ETHTOOL_MSG_USER_MAX = __ETHTOOL_MSG_USER_CNT - 1
+};
+
+/* message types - kernel to userspace */
+enum {
+	ETHTOOL_MSG_KERNEL_NONE,
+
+	/* add new constants above here */
+	__ETHTOOL_MSG_KERNEL_CNT,
+	ETHTOOL_MSG_KERNEL_MAX = __ETHTOOL_MSG_KERNEL_CNT - 1
+};
+
+/* generic netlink info */
+#define ETHTOOL_GENL_NAME "ethtool"
+#define ETHTOOL_GENL_VERSION 1
+
+#endif /* _UAPI_LINUX_ETHTOOL_NETLINK_H_ */
diff --git a/net/Kconfig b/net/Kconfig
index 52af65e5d28c..54916b7adb9b 100644
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -449,6 +449,14 @@ config FAILOVER
 	  migration of VMs with direct attached VFs by failing over to the
 	  paravirtual datapath when the VF is unplugged.
 
+config ETHTOOL_NETLINK
+	bool "Netlink interface for ethtool"
+	default y
+	help
+	  An alternative userspace interface for ethtool based on generic
+	  netlink. It provides better extensibility and some new features,
+	  e.g. notification messages.
+
 endif   # if NET
 
 # Used by archs to tell that they support BPF JIT compiler plus which flavour.
diff --git a/net/ethtool/Makefile b/net/ethtool/Makefile
index f68387618973..59d5ee230c29 100644
--- a/net/ethtool/Makefile
+++ b/net/ethtool/Makefile
@@ -1,3 +1,7 @@
 # SPDX-License-Identifier: GPL-2.0-only
 
-obj-y		+= ioctl.o common.o
+obj-y				+= ioctl.o common.o
+
+obj-$(CONFIG_ETHTOOL_NETLINK)	+= ethtool_nl.o
+
+ethtool_nl-y	:= netlink.o
diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
new file mode 100644
index 000000000000..59e1ebde2f15
--- /dev/null
+++ b/net/ethtool/netlink.c
@@ -0,0 +1,33 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include <linux/ethtool_netlink.h>
+#include "netlink.h"
+
+/* genetlink setup */
+
+static const struct genl_ops ethtool_genl_ops[] = {
+};
+
+static struct genl_family ethtool_genl_family = {
+	.name		= ETHTOOL_GENL_NAME,
+	.version	= ETHTOOL_GENL_VERSION,
+	.netnsok	= true,
+	.parallel_ops	= true,
+	.ops		= ethtool_genl_ops,
+	.n_ops		= ARRAY_SIZE(ethtool_genl_ops),
+};
+
+/* module setup */
+
+static int __init ethnl_init(void)
+{
+	int ret;
+
+	ret = genl_register_family(&ethtool_genl_family);
+	if (WARN(ret < 0, "ethtool: genetlink family registration failed"))
+		return ret;
+
+	return 0;
+}
+
+subsys_initcall(ethnl_init);
diff --git a/net/ethtool/netlink.h b/net/ethtool/netlink.h
new file mode 100644
index 000000000000..e4220780d368
--- /dev/null
+++ b/net/ethtool/netlink.h
@@ -0,0 +1,10 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+
+#ifndef _NET_ETHTOOL_NETLINK_H
+#define _NET_ETHTOOL_NETLINK_H
+
+#include <linux/ethtool_netlink.h>
+#include <linux/netdevice.h>
+#include <net/genetlink.h>
+
+#endif /* _NET_ETHTOOL_NETLINK_H */
-- 
cgit v1.2.3-59-g8ed1b


From 041b1c5d4a53e97fc9e029ae32469552ca12cb9b Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Fri, 27 Dec 2019 15:55:23 +0100
Subject: ethtool: helper functions for netlink interface

Add common request/reply header definition and helpers to parse request
header and fill reply header. Provide ethnl_update_* helpers to update
structure members from request attributes (to be used for *_SET requests).

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/uapi/linux/ethtool_netlink.h |  21 ++++
 net/ethtool/netlink.c                | 166 ++++++++++++++++++++++++++++
 net/ethtool/netlink.h                | 205 +++++++++++++++++++++++++++++++++++
 3 files changed, 392 insertions(+)

diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
index 3c93276ba066..82fc3b5f41c9 100644
--- a/include/uapi/linux/ethtool_netlink.h
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -29,6 +29,27 @@ enum {
 	ETHTOOL_MSG_KERNEL_MAX = __ETHTOOL_MSG_KERNEL_CNT - 1
 };
 
+/* request header */
+
+/* use compact bitsets in reply */
+#define ETHTOOL_FLAG_COMPACT_BITSETS	(1 << 0)
+/* provide optional reply for SET or ACT requests */
+#define ETHTOOL_FLAG_OMIT_REPLY	(1 << 1)
+
+#define ETHTOOL_FLAG_ALL (ETHTOOL_FLAG_COMPACT_BITSETS | \
+			  ETHTOOL_FLAG_OMIT_REPLY)
+
+enum {
+	ETHTOOL_A_HEADER_UNSPEC,
+	ETHTOOL_A_HEADER_DEV_INDEX,		/* u32 */
+	ETHTOOL_A_HEADER_DEV_NAME,		/* string */
+	ETHTOOL_A_HEADER_FLAGS,			/* u32 - ETHTOOL_FLAG_* */
+
+	/* add new constants above here */
+	__ETHTOOL_A_HEADER_CNT,
+	ETHTOOL_A_HEADER_MAX = __ETHTOOL_A_HEADER_CNT - 1
+};
+
 /* generic netlink info */
 #define ETHTOOL_GENL_NAME "ethtool"
 #define ETHTOOL_GENL_VERSION 1
diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index 59e1ebde2f15..aef882e0c3f5 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -1,8 +1,174 @@
 // SPDX-License-Identifier: GPL-2.0-only
 
+#include <net/sock.h>
 #include <linux/ethtool_netlink.h>
 #include "netlink.h"
 
+static struct genl_family ethtool_genl_family;
+
+static const struct nla_policy ethnl_header_policy[ETHTOOL_A_HEADER_MAX + 1] = {
+	[ETHTOOL_A_HEADER_UNSPEC]	= { .type = NLA_REJECT },
+	[ETHTOOL_A_HEADER_DEV_INDEX]	= { .type = NLA_U32 },
+	[ETHTOOL_A_HEADER_DEV_NAME]	= { .type = NLA_NUL_STRING,
+					    .len = ALTIFNAMSIZ - 1 },
+	[ETHTOOL_A_HEADER_FLAGS]	= { .type = NLA_U32 },
+};
+
+/**
+ * ethnl_parse_header() - parse request header
+ * @req_info:    structure to put results into
+ * @header:      nest attribute with request header
+ * @net:         request netns
+ * @extack:      netlink extack for error reporting
+ * @require_dev: fail if no device identified in header
+ *
+ * Parse request header in nested attribute @nest and puts results into
+ * the structure pointed to by @req_info. Extack from @info is used for error
+ * reporting. If req_info->dev is not null on return, reference to it has
+ * been taken. If error is returned, *req_info is null initialized and no
+ * reference is held.
+ *
+ * Return: 0 on success or negative error code
+ */
+int ethnl_parse_header(struct ethnl_req_info *req_info,
+		       const struct nlattr *header, struct net *net,
+		       struct netlink_ext_ack *extack, bool require_dev)
+{
+	struct nlattr *tb[ETHTOOL_A_HEADER_MAX + 1];
+	const struct nlattr *devname_attr;
+	struct net_device *dev = NULL;
+	int ret;
+
+	if (!header) {
+		NL_SET_ERR_MSG(extack, "request header missing");
+		return -EINVAL;
+	}
+	ret = nla_parse_nested(tb, ETHTOOL_A_HEADER_MAX, header,
+			       ethnl_header_policy, extack);
+	if (ret < 0)
+		return ret;
+	devname_attr = tb[ETHTOOL_A_HEADER_DEV_NAME];
+
+	if (tb[ETHTOOL_A_HEADER_DEV_INDEX]) {
+		u32 ifindex = nla_get_u32(tb[ETHTOOL_A_HEADER_DEV_INDEX]);
+
+		dev = dev_get_by_index(net, ifindex);
+		if (!dev) {
+			NL_SET_ERR_MSG_ATTR(extack,
+					    tb[ETHTOOL_A_HEADER_DEV_INDEX],
+					    "no device matches ifindex");
+			return -ENODEV;
+		}
+		/* if both ifindex and ifname are passed, they must match */
+		if (devname_attr &&
+		    strncmp(dev->name, nla_data(devname_attr), IFNAMSIZ)) {
+			dev_put(dev);
+			NL_SET_ERR_MSG_ATTR(extack, header,
+					    "ifindex and name do not match");
+			return -ENODEV;
+		}
+	} else if (devname_attr) {
+		dev = dev_get_by_name(net, nla_data(devname_attr));
+		if (!dev) {
+			NL_SET_ERR_MSG_ATTR(extack, devname_attr,
+					    "no device matches name");
+			return -ENODEV;
+		}
+	} else if (require_dev) {
+		NL_SET_ERR_MSG_ATTR(extack, header,
+				    "neither ifindex nor name specified");
+		return -EINVAL;
+	}
+
+	if (dev && !netif_device_present(dev)) {
+		dev_put(dev);
+		NL_SET_ERR_MSG(extack, "device not present");
+		return -ENODEV;
+	}
+
+	req_info->dev = dev;
+	if (tb[ETHTOOL_A_HEADER_FLAGS])
+		req_info->flags = nla_get_u32(tb[ETHTOOL_A_HEADER_FLAGS]);
+
+	return 0;
+}
+
+/**
+ * ethnl_fill_reply_header() - Put common header into a reply message
+ * @skb:      skb with the message
+ * @dev:      network device to describe in header
+ * @attrtype: attribute type to use for the nest
+ *
+ * Create a nested attribute with attributes describing given network device.
+ *
+ * Return: 0 on success, error value (-EMSGSIZE only) on error
+ */
+int ethnl_fill_reply_header(struct sk_buff *skb, struct net_device *dev,
+			    u16 attrtype)
+{
+	struct nlattr *nest;
+
+	if (!dev)
+		return 0;
+	nest = nla_nest_start(skb, attrtype);
+	if (!nest)
+		return -EMSGSIZE;
+
+	if (nla_put_u32(skb, ETHTOOL_A_HEADER_DEV_INDEX, (u32)dev->ifindex) ||
+	    nla_put_string(skb, ETHTOOL_A_HEADER_DEV_NAME, dev->name))
+		goto nla_put_failure;
+	/* If more attributes are put into reply header, ethnl_header_size()
+	 * must be updated to account for them.
+	 */
+
+	nla_nest_end(skb, nest);
+	return 0;
+
+nla_put_failure:
+	nla_nest_cancel(skb, nest);
+	return -EMSGSIZE;
+}
+
+/**
+ * ethnl_reply_init() - Create skb for a reply and fill device identification
+ * @payload: payload length (without netlink and genetlink header)
+ * @dev:     device the reply is about (may be null)
+ * @cmd:     ETHTOOL_MSG_* message type for reply
+ * @info:    genetlink info of the received packet we respond to
+ * @ehdrp:   place to store payload pointer returned by genlmsg_new()
+ *
+ * Return: pointer to allocated skb on success, NULL on error
+ */
+struct sk_buff *ethnl_reply_init(size_t payload, struct net_device *dev, u8 cmd,
+				 u16 hdr_attrtype, struct genl_info *info,
+				 void **ehdrp)
+{
+	struct sk_buff *skb;
+
+	skb = genlmsg_new(payload, GFP_KERNEL);
+	if (!skb)
+		goto err;
+	*ehdrp = genlmsg_put_reply(skb, info, &ethtool_genl_family, 0, cmd);
+	if (!*ehdrp)
+		goto err_free;
+
+	if (dev) {
+		int ret;
+
+		ret = ethnl_fill_reply_header(skb, dev, hdr_attrtype);
+		if (ret < 0)
+			goto err_free;
+	}
+	return skb;
+
+err_free:
+	nlmsg_free(skb);
+err:
+	if (info)
+		GENL_SET_ERR_MSG(info, "failed to setup reply message");
+	return NULL;
+}
+
 /* genetlink setup */
 
 static const struct genl_ops ethtool_genl_ops[] = {
diff --git a/net/ethtool/netlink.h b/net/ethtool/netlink.h
index e4220780d368..05d7183da894 100644
--- a/net/ethtool/netlink.h
+++ b/net/ethtool/netlink.h
@@ -6,5 +6,210 @@
 #include <linux/ethtool_netlink.h>
 #include <linux/netdevice.h>
 #include <net/genetlink.h>
+#include <net/sock.h>
+
+struct ethnl_req_info;
+
+int ethnl_parse_header(struct ethnl_req_info *req_info,
+		       const struct nlattr *nest, struct net *net,
+		       struct netlink_ext_ack *extack, bool require_dev);
+int ethnl_fill_reply_header(struct sk_buff *skb, struct net_device *dev,
+			    u16 attrtype);
+struct sk_buff *ethnl_reply_init(size_t payload, struct net_device *dev, u8 cmd,
+				 u16 hdr_attrtype, struct genl_info *info,
+				 void **ehdrp);
+
+/**
+ * ethnl_strz_size() - calculate attribute length for fixed size string
+ * @s: ETH_GSTRING_LEN sized string (may not be null terminated)
+ *
+ * Return: total length of an attribute with null terminated string from @s
+ */
+static inline int ethnl_strz_size(const char *s)
+{
+	return nla_total_size(strnlen(s, ETH_GSTRING_LEN) + 1);
+}
+
+/**
+ * ethnl_put_strz() - put string attribute with fixed size string
+ * @skb:     skb with the message
+ * @attrype: attribute type
+ * @s:       ETH_GSTRING_LEN sized string (may not be null terminated)
+ *
+ * Puts an attribute with null terminated string from @s into the message.
+ *
+ * Return: 0 on success, negative error code on failure
+ */
+static inline int ethnl_put_strz(struct sk_buff *skb, u16 attrtype,
+				 const char *s)
+{
+	unsigned int len = strnlen(s, ETH_GSTRING_LEN);
+	struct nlattr *attr;
+
+	attr = nla_reserve(skb, attrtype, len + 1);
+	if (!attr)
+		return -EMSGSIZE;
+
+	memcpy(nla_data(attr), s, len);
+	((char *)nla_data(attr))[len] = '\0';
+	return 0;
+}
+
+/**
+ * ethnl_update_u32() - update u32 value from NLA_U32 attribute
+ * @dst:  value to update
+ * @attr: netlink attribute with new value or null
+ * @mod:  pointer to bool for modification tracking
+ *
+ * Copy the u32 value from NLA_U32 netlink attribute @attr into variable
+ * pointed to by @dst; do nothing if @attr is null. Bool pointed to by @mod
+ * is set to true if this function changed the value of *dst, otherwise it
+ * is left as is.
+ */
+static inline void ethnl_update_u32(u32 *dst, const struct nlattr *attr,
+				    bool *mod)
+{
+	u32 val;
+
+	if (!attr)
+		return;
+	val = nla_get_u32(attr);
+	if (*dst == val)
+		return;
+
+	*dst = val;
+	*mod = true;
+}
+
+/**
+ * ethnl_update_u8() - update u8 value from NLA_U8 attribute
+ * @dst:  value to update
+ * @attr: netlink attribute with new value or null
+ * @mod:  pointer to bool for modification tracking
+ *
+ * Copy the u8 value from NLA_U8 netlink attribute @attr into variable
+ * pointed to by @dst; do nothing if @attr is null. Bool pointed to by @mod
+ * is set to true if this function changed the value of *dst, otherwise it
+ * is left as is.
+ */
+static inline void ethnl_update_u8(u8 *dst, const struct nlattr *attr,
+				   bool *mod)
+{
+	u8 val;
+
+	if (!attr)
+		return;
+	val = nla_get_u8(attr);
+	if (*dst == val)
+		return;
+
+	*dst = val;
+	*mod = true;
+}
+
+/**
+ * ethnl_update_bool32() - update u32 used as bool from NLA_U8 attribute
+ * @dst:  value to update
+ * @attr: netlink attribute with new value or null
+ * @mod:  pointer to bool for modification tracking
+ *
+ * Use the u8 value from NLA_U8 netlink attribute @attr to set u32 variable
+ * pointed to by @dst to 0 (if zero) or 1 (if not); do nothing if @attr is
+ * null. Bool pointed to by @mod is set to true if this function changed the
+ * logical value of *dst, otherwise it is left as is.
+ */
+static inline void ethnl_update_bool32(u32 *dst, const struct nlattr *attr,
+				       bool *mod)
+{
+	u8 val;
+
+	if (!attr)
+		return;
+	val = !!nla_get_u8(attr);
+	if (!!*dst == val)
+		return;
+
+	*dst = val;
+	*mod = true;
+}
+
+/**
+ * ethnl_update_binary() - update binary data from NLA_BINARY atribute
+ * @dst:  value to update
+ * @len:  destination buffer length
+ * @attr: netlink attribute with new value or null
+ * @mod:  pointer to bool for modification tracking
+ *
+ * Use the u8 value from NLA_U8 netlink attribute @attr to rewrite data block
+ * of length @len at @dst by attribute payload; do nothing if @attr is null.
+ * Bool pointed to by @mod is set to true if this function changed the logical
+ * value of *dst, otherwise it is left as is.
+ */
+static inline void ethnl_update_binary(void *dst, unsigned int len,
+				       const struct nlattr *attr, bool *mod)
+{
+	if (!attr)
+		return;
+	if (nla_len(attr) < len)
+		len = nla_len(attr);
+	if (!memcmp(dst, nla_data(attr), len))
+		return;
+
+	memcpy(dst, nla_data(attr), len);
+	*mod = true;
+}
+
+/**
+ * ethnl_update_bitfield32() - update u32 value from NLA_BITFIELD32 attribute
+ * @dst:  value to update
+ * @attr: netlink attribute with new value or null
+ * @mod:  pointer to bool for modification tracking
+ *
+ * Update bits in u32 value which are set in attribute's mask to values from
+ * attribute's value. Do nothing if @attr is null or the value wouldn't change;
+ * otherwise, set bool pointed to by @mod to true.
+ */
+static inline void ethnl_update_bitfield32(u32 *dst, const struct nlattr *attr,
+					   bool *mod)
+{
+	struct nla_bitfield32 change;
+	u32 newval;
+
+	if (!attr)
+		return;
+	change = nla_get_bitfield32(attr);
+	newval = (*dst & ~change.selector) | (change.value & change.selector);
+	if (*dst == newval)
+		return;
+
+	*dst = newval;
+	*mod = true;
+}
+
+/**
+ * ethnl_reply_header_size() - total size of reply header
+ *
+ * This is an upper estimate so that we do not need to hold RTNL lock longer
+ * than necessary (to prevent rename between size estimate and composing the
+ * message). Accounts only for device ifindex and name as those are the only
+ * attributes ethnl_fill_reply_header() puts into the reply header.
+ */
+static inline unsigned int ethnl_reply_header_size(void)
+{
+	return nla_total_size(nla_total_size(sizeof(u32)) +
+			      nla_total_size(IFNAMSIZ));
+}
+
+/**
+ * struct ethnl_req_info - base type of request information for GET requests
+ * @dev:   network device the request is for (may be null)
+ * @flags: request flags common for all request types
+ *
+ * This is a common base, additional members may follow after this structure.
+ */
+struct ethnl_req_info {
+	struct net_device	*dev;
+	u32			flags;
+};
 
 #endif /* _NET_ETHTOOL_NETLINK_H */
-- 
cgit v1.2.3-59-g8ed1b


From 10b518d4e6dd5390e40f7d8de0f08753c1195a7e Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Fri, 27 Dec 2019 15:55:28 +0100
Subject: ethtool: netlink bitset handling

The ethtool netlink code uses common framework for passing arbitrary
length bit sets to allow future extensions. A bitset can be a list (only
one bitmap) or can consist of value and mask pair (used e.g. when client
want to modify only some bits). A bitset can use one of two formats:
verbose (bit by bit) or compact.

Verbose format consists of bitset size (number of bits), list flag and
an array of bit nests, telling which bits are part of the list or which
bits are in the mask and which of them are to be set. In requests, bits
can be identified by index (position) or by name. In replies, kernel
provides both index and name. Verbose format is suitable for "one shot"
applications like standard ethtool command as it avoids the need to
either keep bit names (e.g. link modes) in sync with kernel or having to
add an extra roundtrip for string set request (e.g. for private flags).

Compact format uses one (list) or two (value/mask) arrays of 32-bit
words to store the bitmap(s). It is more suitable for long running
applications (ethtool in monitor mode or network management daemons)
which can retrieve the names once and then pass only compact bitmaps to
save space.

Userspace requests can use either format; ETHTOOL_FLAG_COMPACT_BITSETS
flag in request header tells kernel which format to use in reply.
Notifications always use compact format.

As some code uses arrays of unsigned long for internal representation and
some arrays of u32 (or even a single u32), two sets of parse/compose
helpers are introduced. To avoid code duplication, helpers for unsigned
long arrays are implemented as wrappers around helpers for u32 arrays.
There are two reasons for this choice: (1) u32 arrays are more frequent in
ethtool code and (2) unsigned long array can be always interpreted as an
u32 array on little endian 64-bit and all 32-bit architectures while we
would need special handling for odd number of u32 words in the opposite
direction.

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/ethtool-netlink.rst |  84 +++
 include/uapi/linux/ethtool_netlink.h         |  35 ++
 net/ethtool/Makefile                         |   2 +-
 net/ethtool/bitset.c                         | 735 +++++++++++++++++++++++++++
 net/ethtool/bitset.h                         |  28 +
 5 files changed, 883 insertions(+), 1 deletion(-)
 create mode 100644 net/ethtool/bitset.c
 create mode 100644 net/ethtool/bitset.h

diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
index 9448442ad293..7797f1237472 100644
--- a/Documentation/networking/ethtool-netlink.rst
+++ b/Documentation/networking/ethtool-netlink.rst
@@ -76,6 +76,90 @@ of the flag should be interpreted the way the client expects. A client must
 not set flags it does not understand.
 
 
+Bit sets
+========
+
+For short bitmaps of (reasonably) fixed length, standard ``NLA_BITFIELD32``
+type is used. For arbitrary length bitmaps, ethtool netlink uses a nested
+attribute with contents of one of two forms: compact (two binary bitmaps
+representing bit values and mask of affected bits) and bit-by-bit (list of
+bits identified by either index or name).
+
+Verbose (bit-by-bit) bitsets allow sending symbolic names for bits together
+with their values which saves a round trip (when the bitset is passed in a
+request) or at least a second request (when the bitset is in a reply). This is
+useful for one shot applications like traditional ethtool command. On the
+other hand, long running applications like ethtool monitor (displaying
+notifications) or network management daemons may prefer fetching the names
+only once and using compact form to save message size. Notifications from
+ethtool netlink interface always use compact form for bitsets.
+
+A bitset can represent either a value/mask pair (``ETHTOOL_A_BITSET_NOMASK``
+not set) or a single bitmap (``ETHTOOL_A_BITSET_NOMASK`` set). In requests
+modifying a bitmap, the former changes the bit set in mask to values set in
+value and preserves the rest; the latter sets the bits set in the bitmap and
+clears the rest.
+
+Compact form: nested (bitset) atrribute contents:
+
+  ============================  ======  ============================
+  ``ETHTOOL_A_BITSET_NOMASK``   flag    no mask, only a list
+  ``ETHTOOL_A_BITSET_SIZE``     u32     number of significant bits
+  ``ETHTOOL_A_BITSET_VALUE``    binary  bitmap of bit values
+  ``ETHTOOL_A_BITSET_MASK``     binary  bitmap of valid bits
+  ============================  ======  ============================
+
+Value and mask must have length at least ``ETHTOOL_A_BITSET_SIZE`` bits
+rounded up to a multiple of 32 bits. They consist of 32-bit words in host byte
+order, words ordered from least significant to most significant (i.e. the same
+way as bitmaps are passed with ioctl interface).
+
+For compact form, ``ETHTOOL_A_BITSET_SIZE`` and ``ETHTOOL_A_BITSET_VALUE`` are
+mandatory. ``ETHTOOL_A_BITSET_MASK`` attribute is mandatory if
+``ETHTOOL_A_BITSET_NOMASK`` is not set (bitset represents a value/mask pair);
+if ``ETHTOOL_A_BITSET_NOMASK`` is not set, ``ETHTOOL_A_BITSET_MASK`` is not
+allowed (bitset represents a single bitmap.
+
+Kernel bit set length may differ from userspace length if older application is
+used on newer kernel or vice versa. If userspace bitmap is longer, an error is
+issued only if the request actually tries to set values of some bits not
+recognized by kernel.
+
+Bit-by-bit form: nested (bitset) attribute contents:
+
+ +------------------------------------+--------+-----------------------------+
+ | ``ETHTOOL_A_BITSET_NOMASK``        | flag   | no mask, only a list        |
+ +------------------------------------+--------+-----------------------------+
+ | ``ETHTOOL_A_BITSET_SIZE``          | u32    | number of significant bits  |
+ +------------------------------------+--------+-----------------------------+
+ | ``ETHTOOL_A_BITSET_BITS``          | nested | array of bits               |
+ +-+----------------------------------+--------+-----------------------------+
+ | | ``ETHTOOL_A_BITSET_BITS_BIT+``   | nested | one bit                     |
+ +-+-+--------------------------------+--------+-----------------------------+
+ | | | ``ETHTOOL_A_BITSET_BIT_INDEX`` | u32    | bit index (0 for LSB)       |
+ +-+-+--------------------------------+--------+-----------------------------+
+ | | | ``ETHTOOL_A_BITSET_BIT_NAME``  | string | bit name                    |
+ +-+-+--------------------------------+--------+-----------------------------+
+ | | | ``ETHTOOL_A_BITSET_BIT_VALUE`` | flag   | present if bit is set       |
+ +-+-+--------------------------------+--------+-----------------------------+
+
+Bit size is optional for bit-by-bit form. ``ETHTOOL_A_BITSET_BITS`` nest can
+only contain ``ETHTOOL_A_BITSET_BITS_BIT`` attributes but there can be an
+arbitrary number of them.  A bit may be identified by its index or by its
+name. When used in requests, listed bits are set to 0 or 1 according to
+``ETHTOOL_A_BITSET_BIT_VALUE``, the rest is preserved. A request fails if
+index exceeds kernel bit length or if name is not recognized.
+
+When ``ETHTOOL_A_BITSET_NOMASK`` flag is present, bitset is interpreted as
+a simple bitmap. ``ETHTOOL_A_BITSET_BIT_VALUE`` attributes are not used in
+such case. Such bitset represents a bitmap with listed bits set and the rest
+zero.
+
+In requests, application can use either form. Form used by kernel in reply is
+determined by ``ETHTOOL_FLAG_COMPACT_BITSETS`` flag in flags field of request
+header. Semantics of value and mask depends on the attribute.
+
+
 List of message types
 =====================
 
diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
index 82fc3b5f41c9..951203049615 100644
--- a/include/uapi/linux/ethtool_netlink.h
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -50,6 +50,41 @@ enum {
 	ETHTOOL_A_HEADER_MAX = __ETHTOOL_A_HEADER_CNT - 1
 };
 
+/* bit sets */
+
+enum {
+	ETHTOOL_A_BITSET_BIT_UNSPEC,
+	ETHTOOL_A_BITSET_BIT_INDEX,		/* u32 */
+	ETHTOOL_A_BITSET_BIT_NAME,		/* string */
+	ETHTOOL_A_BITSET_BIT_VALUE,		/* flag */
+
+	/* add new constants above here */
+	__ETHTOOL_A_BITSET_BIT_CNT,
+	ETHTOOL_A_BITSET_BIT_MAX = __ETHTOOL_A_BITSET_BIT_CNT - 1
+};
+
+enum {
+	ETHTOOL_A_BITSET_BITS_UNSPEC,
+	ETHTOOL_A_BITSET_BITS_BIT,		/* nest - _A_BITSET_BIT_* */
+
+	/* add new constants above here */
+	__ETHTOOL_A_BITSET_BITS_CNT,
+	ETHTOOL_A_BITSET_BITS_MAX = __ETHTOOL_A_BITSET_BITS_CNT - 1
+};
+
+enum {
+	ETHTOOL_A_BITSET_UNSPEC,
+	ETHTOOL_A_BITSET_NOMASK,		/* flag */
+	ETHTOOL_A_BITSET_SIZE,			/* u32 */
+	ETHTOOL_A_BITSET_BITS,			/* nest - _A_BITSET_BITS_* */
+	ETHTOOL_A_BITSET_VALUE,			/* binary */
+	ETHTOOL_A_BITSET_MASK,			/* binary */
+
+	/* add new constants above here */
+	__ETHTOOL_A_BITSET_CNT,
+	ETHTOOL_A_BITSET_MAX = __ETHTOOL_A_BITSET_CNT - 1
+};
+
 /* generic netlink info */
 #define ETHTOOL_GENL_NAME "ethtool"
 #define ETHTOOL_GENL_VERSION 1
diff --git a/net/ethtool/Makefile b/net/ethtool/Makefile
index 59d5ee230c29..a7e6c2c85db9 100644
--- a/net/ethtool/Makefile
+++ b/net/ethtool/Makefile
@@ -4,4 +4,4 @@ obj-y				+= ioctl.o common.o
 
 obj-$(CONFIG_ETHTOOL_NETLINK)	+= ethtool_nl.o
 
-ethtool_nl-y	:= netlink.o
+ethtool_nl-y	:= netlink.o bitset.o
diff --git a/net/ethtool/bitset.c b/net/ethtool/bitset.c
new file mode 100644
index 000000000000..fce45dac4205
--- /dev/null
+++ b/net/ethtool/bitset.c
@@ -0,0 +1,735 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include <linux/ethtool_netlink.h>
+#include <linux/bitmap.h>
+#include "netlink.h"
+#include "bitset.h"
+
+/* Some bitmaps are internally represented as an array of unsigned long, some
+ * as an array of u32 (some even as single u32 for now). To avoid the need of
+ * wrappers on caller side, we provide two set of functions: those with "32"
+ * suffix in their names expect u32 based bitmaps, those without it expect
+ * unsigned long bitmaps.
+ */
+
+static u32 ethnl_lower_bits(unsigned int n)
+{
+	return ~(u32)0 >> (32 - n % 32);
+}
+
+static u32 ethnl_upper_bits(unsigned int n)
+{
+	return ~(u32)0 << (n % 32);
+}
+
+/**
+ * ethnl_bitmap32_clear() - Clear u32 based bitmap
+ * @dst:   bitmap to clear
+ * @start: beginning of the interval
+ * @end:   end of the interval
+ * @mod:   set if bitmap was modified
+ *
+ * Clear @nbits bits of a bitmap with indices @start <= i < @end
+ */
+static void ethnl_bitmap32_clear(u32 *dst, unsigned int start, unsigned int end,
+				 bool *mod)
+{
+	unsigned int start_word = start / 32;
+	unsigned int end_word = end / 32;
+	unsigned int i;
+	u32 mask;
+
+	if (end <= start)
+		return;
+
+	if (start % 32) {
+		mask = ethnl_upper_bits(start);
+		if (end_word == start_word) {
+			mask &= ethnl_lower_bits(end);
+			if (dst[start_word] & mask) {
+				dst[start_word] &= ~mask;
+				*mod = true;
+			}
+			return;
+		}
+		if (dst[start_word] & mask) {
+			dst[start_word] &= ~mask;
+			*mod = true;
+		}
+		start_word++;
+	}
+
+	for (i = start_word; i < end_word; i++) {
+		if (dst[i]) {
+			dst[i] = 0;
+			*mod = true;
+		}
+	}
+	if (end % 32) {
+		mask = ethnl_lower_bits(end);
+		if (dst[end_word] & mask) {
+			dst[end_word] &= ~mask;
+			*mod = true;
+		}
+	}
+}
+
+/**
+ * ethnl_bitmap32_not_zero() - Check if any bit is set in an interval
+ * @map:   bitmap to test
+ * @start: beginning of the interval
+ * @end:   end of the interval
+ *
+ * Return: true if there is non-zero bit with  index @start <= i < @end,
+ *         false if the whole interval is zero
+ */
+static bool ethnl_bitmap32_not_zero(const u32 *map, unsigned int start,
+				    unsigned int end)
+{
+	unsigned int start_word = start / 32;
+	unsigned int end_word = end / 32;
+	u32 mask;
+
+	if (end <= start)
+		return true;
+
+	if (start % 32) {
+		mask = ethnl_upper_bits(start);
+		if (end_word == start_word) {
+			mask &= ethnl_lower_bits(end);
+			return map[start_word] & mask;
+		}
+		if (map[start_word] & mask)
+			return true;
+		start_word++;
+	}
+
+	if (!memchr_inv(map + start_word, '\0',
+			(end_word - start_word) * sizeof(u32)))
+		return true;
+	if (end % 32 == 0)
+		return true;
+	return map[end_word] & ethnl_lower_bits(end);
+}
+
+/**
+ * ethnl_bitmap32_update() - Modify u32 based bitmap according to value/mask
+ *			     pair
+ * @dst:   bitmap to update
+ * @nbits: bit size of the bitmap
+ * @value: values to set
+ * @mask:  mask of bits to set
+ * @mod:   set to true if bitmap is modified, preserve if not
+ *
+ * Set bits in @dst bitmap which are set in @mask to values from @value, leave
+ * the rest untouched. If destination bitmap was modified, set @mod to true,
+ * leave as it is if not.
+ */
+static void ethnl_bitmap32_update(u32 *dst, unsigned int nbits,
+				  const u32 *value, const u32 *mask, bool *mod)
+{
+	while (nbits > 0) {
+		u32 real_mask = mask ? *mask : ~(u32)0;
+		u32 new_value;
+
+		if (nbits < 32)
+			real_mask &= ethnl_lower_bits(nbits);
+		new_value = (*dst & ~real_mask) | (*value & real_mask);
+		if (new_value != *dst) {
+			*dst = new_value;
+			*mod = true;
+		}
+
+		if (nbits <= 32)
+			break;
+		dst++;
+		nbits -= 32;
+		value++;
+		if (mask)
+			mask++;
+	}
+}
+
+static bool ethnl_bitmap32_test_bit(const u32 *map, unsigned int index)
+{
+	return map[index / 32] & (1U << (index % 32));
+}
+
+/**
+ * ethnl_bitset32_size() - Calculate size of bitset nested attribute
+ * @val:     value bitmap (u32 based)
+ * @mask:    mask bitmap (u32 based, optional)
+ * @nbits:   bit length of the bitset
+ * @names:   array of bit names (optional)
+ * @compact: assume compact format for output
+ *
+ * Estimate length of netlink attribute composed by a later call to
+ * ethnl_put_bitset32() call with the same arguments.
+ *
+ * Return: negative error code or attribute length estimate
+ */
+int ethnl_bitset32_size(const u32 *val, const u32 *mask, unsigned int nbits,
+			ethnl_string_array_t names, bool compact)
+{
+	unsigned int len = 0;
+
+	/* list flag */
+	if (!mask)
+		len += nla_total_size(sizeof(u32));
+	/* size */
+	len += nla_total_size(sizeof(u32));
+
+	if (compact) {
+		unsigned int nwords = DIV_ROUND_UP(nbits, 32);
+
+		/* value, mask */
+		len += (mask ? 2 : 1) * nla_total_size(nwords * sizeof(u32));
+	} else {
+		unsigned int bits_len = 0;
+		unsigned int bit_len, i;
+
+		for (i = 0; i < nbits; i++) {
+			const char *name = names ? names[i] : NULL;
+
+			if (!ethnl_bitmap32_test_bit(mask ?: val, i))
+				continue;
+			/* index */
+			bit_len = nla_total_size(sizeof(u32));
+			/* name */
+			if (name)
+				bit_len += ethnl_strz_size(name);
+			/* value */
+			if (mask && ethnl_bitmap32_test_bit(val, i))
+				bit_len += nla_total_size(0);
+
+			/* bit nest */
+			bits_len += nla_total_size(bit_len);
+		}
+		/* bits nest */
+		len += nla_total_size(bits_len);
+	}
+
+	/* outermost nest */
+	return nla_total_size(len);
+}
+
+/**
+ * ethnl_put_bitset32() - Put a bitset nest into a message
+ * @skb:      skb with the message
+ * @attrtype: attribute type for the bitset nest
+ * @val:      value bitmap (u32 based)
+ * @mask:     mask bitmap (u32 based, optional)
+ * @nbits:    bit length of the bitset
+ * @names:    array of bit names (optional)
+ * @compact:  use compact format for the output
+ *
+ * Compose a nested attribute representing a bitset. If @mask is null, simple
+ * bitmap (bit list) is created, if @mask is provided, represent a value/mask
+ * pair. Bit names are only used in verbose mode and when provided by calller.
+ *
+ * Return: 0 on success, negative error value on error
+ */
+int ethnl_put_bitset32(struct sk_buff *skb, int attrtype, const u32 *val,
+		       const u32 *mask, unsigned int nbits,
+		       ethnl_string_array_t names, bool compact)
+{
+	struct nlattr *nest;
+	struct nlattr *attr;
+
+	nest = nla_nest_start(skb, attrtype);
+	if (!nest)
+		return -EMSGSIZE;
+
+	if (!mask && nla_put_flag(skb, ETHTOOL_A_BITSET_NOMASK))
+		goto nla_put_failure;
+	if (nla_put_u32(skb, ETHTOOL_A_BITSET_SIZE, nbits))
+		goto nla_put_failure;
+	if (compact) {
+		unsigned int nwords = DIV_ROUND_UP(nbits, 32);
+		unsigned int nbytes = nwords * sizeof(u32);
+		u32 *dst;
+
+		attr = nla_reserve(skb, ETHTOOL_A_BITSET_VALUE, nbytes);
+		if (!attr)
+			goto nla_put_failure;
+		dst = nla_data(attr);
+		memcpy(dst, val, nbytes);
+		if (nbits % 32)
+			dst[nwords - 1] &= ethnl_lower_bits(nbits);
+
+		if (mask) {
+			attr = nla_reserve(skb, ETHTOOL_A_BITSET_MASK, nbytes);
+			if (!attr)
+				goto nla_put_failure;
+			dst = nla_data(attr);
+			memcpy(dst, mask, nbytes);
+			if (nbits % 32)
+				dst[nwords - 1] &= ethnl_lower_bits(nbits);
+		}
+	} else {
+		struct nlattr *bits;
+		unsigned int i;
+
+		bits = nla_nest_start(skb, ETHTOOL_A_BITSET_BITS);
+		if (!bits)
+			goto nla_put_failure;
+		for (i = 0; i < nbits; i++) {
+			const char *name = names ? names[i] : NULL;
+
+			if (!ethnl_bitmap32_test_bit(mask ?: val, i))
+				continue;
+			attr = nla_nest_start(skb, ETHTOOL_A_BITSET_BITS_BIT);
+			if (!attr)
+				goto nla_put_failure;
+			if (nla_put_u32(skb, ETHTOOL_A_BITSET_BIT_INDEX, i))
+				goto nla_put_failure;
+			if (name &&
+			    ethnl_put_strz(skb, ETHTOOL_A_BITSET_BIT_NAME, name))
+				goto nla_put_failure;
+			if (mask && ethnl_bitmap32_test_bit(val, i) &&
+			    nla_put_flag(skb, ETHTOOL_A_BITSET_BIT_VALUE))
+				goto nla_put_failure;
+			nla_nest_end(skb, attr);
+		}
+		nla_nest_end(skb, bits);
+	}
+
+	nla_nest_end(skb, nest);
+	return 0;
+
+nla_put_failure:
+	nla_nest_cancel(skb, nest);
+	return -EMSGSIZE;
+}
+
+static const struct nla_policy bitset_policy[ETHTOOL_A_BITSET_MAX + 1] = {
+	[ETHTOOL_A_BITSET_UNSPEC]	= { .type = NLA_REJECT },
+	[ETHTOOL_A_BITSET_NOMASK]	= { .type = NLA_FLAG },
+	[ETHTOOL_A_BITSET_SIZE]		= { .type = NLA_U32 },
+	[ETHTOOL_A_BITSET_BITS]		= { .type = NLA_NESTED },
+	[ETHTOOL_A_BITSET_VALUE]	= { .type = NLA_BINARY },
+	[ETHTOOL_A_BITSET_MASK]		= { .type = NLA_BINARY },
+};
+
+static const struct nla_policy bit_policy[ETHTOOL_A_BITSET_BIT_MAX + 1] = {
+	[ETHTOOL_A_BITSET_BIT_UNSPEC]	= { .type = NLA_REJECT },
+	[ETHTOOL_A_BITSET_BIT_INDEX]	= { .type = NLA_U32 },
+	[ETHTOOL_A_BITSET_BIT_NAME]	= { .type = NLA_NUL_STRING },
+	[ETHTOOL_A_BITSET_BIT_VALUE]	= { .type = NLA_FLAG },
+};
+
+/**
+ * ethnl_bitset_is_compact() - check if bitset attribute represents a compact
+ *			       bitset
+ * @bitset:  nested attribute representing a bitset
+ * @compact: pointer for return value
+ *
+ * Return: 0 on success, negative error code on failure
+ */
+int ethnl_bitset_is_compact(const struct nlattr *bitset, bool *compact)
+{
+	struct nlattr *tb[ETHTOOL_A_BITSET_MAX + 1];
+	int ret;
+
+	ret = nla_parse_nested(tb, ETHTOOL_A_BITSET_MAX, bitset,
+			       bitset_policy, NULL);
+	if (ret < 0)
+		return ret;
+
+	if (tb[ETHTOOL_A_BITSET_BITS]) {
+		if (tb[ETHTOOL_A_BITSET_VALUE] || tb[ETHTOOL_A_BITSET_MASK])
+			return -EINVAL;
+		*compact = false;
+		return 0;
+	}
+	if (!tb[ETHTOOL_A_BITSET_SIZE] || !tb[ETHTOOL_A_BITSET_VALUE])
+		return -EINVAL;
+
+	*compact = true;
+	return 0;
+}
+
+/**
+ * ethnl_name_to_idx() - look up string index for a name
+ * @names:   array of ETH_GSTRING_LEN sized strings
+ * @n_names: number of strings in the array
+ * @name:    name to look up
+ *
+ * Return: index of the string if found, -ENOENT if not found
+ */
+static int ethnl_name_to_idx(ethnl_string_array_t names, unsigned int n_names,
+			     const char *name)
+{
+	unsigned int i;
+
+	if (!names)
+		return -ENOENT;
+
+	for (i = 0; i < n_names; i++) {
+		/* names[i] may not be null terminated */
+		if (!strncmp(names[i], name, ETH_GSTRING_LEN) &&
+		    strlen(name) <= ETH_GSTRING_LEN)
+			return i;
+	}
+
+	return -ENOENT;
+}
+
+static int ethnl_parse_bit(unsigned int *index, bool *val, unsigned int nbits,
+			   const struct nlattr *bit_attr, bool no_mask,
+			   ethnl_string_array_t names,
+			   struct netlink_ext_ack *extack)
+{
+	struct nlattr *tb[ETHTOOL_A_BITSET_BIT_MAX + 1];
+	int ret, idx;
+
+	ret = nla_parse_nested(tb, ETHTOOL_A_BITSET_BIT_MAX, bit_attr,
+			       bit_policy, extack);
+	if (ret < 0)
+		return ret;
+
+	if (tb[ETHTOOL_A_BITSET_BIT_INDEX]) {
+		const char *name;
+
+		idx = nla_get_u32(tb[ETHTOOL_A_BITSET_BIT_INDEX]);
+		if (idx >= nbits) {
+			NL_SET_ERR_MSG_ATTR(extack,
+					    tb[ETHTOOL_A_BITSET_BIT_INDEX],
+					    "bit index too high");
+			return -EOPNOTSUPP;
+		}
+		name = names ? names[idx] : NULL;
+		if (tb[ETHTOOL_A_BITSET_BIT_NAME] && name &&
+		    strncmp(nla_data(tb[ETHTOOL_A_BITSET_BIT_NAME]), name,
+			    nla_len(tb[ETHTOOL_A_BITSET_BIT_NAME]))) {
+			NL_SET_ERR_MSG_ATTR(extack, bit_attr,
+					    "bit index and name mismatch");
+			return -EINVAL;
+		}
+	} else if (tb[ETHTOOL_A_BITSET_BIT_NAME]) {
+		idx = ethnl_name_to_idx(names, nbits,
+					nla_data(tb[ETHTOOL_A_BITSET_BIT_NAME]));
+		if (idx < 0) {
+			NL_SET_ERR_MSG_ATTR(extack,
+					    tb[ETHTOOL_A_BITSET_BIT_NAME],
+					    "bit name not found");
+			return -EOPNOTSUPP;
+		}
+	} else {
+		NL_SET_ERR_MSG_ATTR(extack, bit_attr,
+				    "neither bit index nor name specified");
+		return -EINVAL;
+	}
+
+	*index = idx;
+	*val = no_mask || tb[ETHTOOL_A_BITSET_BIT_VALUE];
+	return 0;
+}
+
+static int
+ethnl_update_bitset32_verbose(u32 *bitmap, unsigned int nbits,
+			      const struct nlattr *attr, struct nlattr **tb,
+			      ethnl_string_array_t names,
+			      struct netlink_ext_ack *extack, bool *mod)
+{
+	struct nlattr *bit_attr;
+	bool no_mask;
+	int rem;
+	int ret;
+
+	if (tb[ETHTOOL_A_BITSET_VALUE]) {
+		NL_SET_ERR_MSG_ATTR(extack, tb[ETHTOOL_A_BITSET_VALUE],
+				    "value only allowed in compact bitset");
+		return -EINVAL;
+	}
+	if (tb[ETHTOOL_A_BITSET_MASK]) {
+		NL_SET_ERR_MSG_ATTR(extack, tb[ETHTOOL_A_BITSET_MASK],
+				    "mask only allowed in compact bitset");
+		return -EINVAL;
+	}
+	no_mask = tb[ETHTOOL_A_BITSET_NOMASK];
+
+	nla_for_each_nested(bit_attr, tb[ETHTOOL_A_BITSET_BITS], rem) {
+		bool old_val, new_val;
+		unsigned int idx;
+
+		if (nla_type(bit_attr) != ETHTOOL_A_BITSET_BITS_BIT) {
+			NL_SET_ERR_MSG_ATTR(extack, bit_attr,
+					    "only ETHTOOL_A_BITSET_BITS_BIT allowed in ETHTOOL_A_BITSET_BITS");
+			return -EINVAL;
+		}
+		ret = ethnl_parse_bit(&idx, &new_val, nbits, bit_attr, no_mask,
+				      names, extack);
+		if (ret < 0)
+			return ret;
+		old_val = bitmap[idx / 32] & ((u32)1 << (idx % 32));
+		if (new_val != old_val) {
+			if (new_val)
+				bitmap[idx / 32] |= ((u32)1 << (idx % 32));
+			else
+				bitmap[idx / 32] &= ~((u32)1 << (idx % 32));
+			*mod = true;
+		}
+	}
+
+	return 0;
+}
+
+static int ethnl_compact_sanity_checks(unsigned int nbits,
+				       const struct nlattr *nest,
+				       struct nlattr **tb,
+				       struct netlink_ext_ack *extack)
+{
+	bool no_mask = tb[ETHTOOL_A_BITSET_NOMASK];
+	unsigned int attr_nbits, attr_nwords;
+	const struct nlattr *test_attr;
+
+	if (no_mask && tb[ETHTOOL_A_BITSET_MASK]) {
+		NL_SET_ERR_MSG_ATTR(extack, tb[ETHTOOL_A_BITSET_MASK],
+				    "mask not allowed in list bitset");
+		return -EINVAL;
+	}
+	if (!tb[ETHTOOL_A_BITSET_SIZE]) {
+		NL_SET_ERR_MSG_ATTR(extack, nest,
+				    "missing size in compact bitset");
+		return -EINVAL;
+	}
+	if (!tb[ETHTOOL_A_BITSET_VALUE]) {
+		NL_SET_ERR_MSG_ATTR(extack, nest,
+				    "missing value in compact bitset");
+		return -EINVAL;
+	}
+	if (!no_mask && !tb[ETHTOOL_A_BITSET_MASK]) {
+		NL_SET_ERR_MSG_ATTR(extack, nest,
+				    "missing mask in compact nonlist bitset");
+		return -EINVAL;
+	}
+
+	attr_nbits = nla_get_u32(tb[ETHTOOL_A_BITSET_SIZE]);
+	attr_nwords = DIV_ROUND_UP(attr_nbits, 32);
+	if (nla_len(tb[ETHTOOL_A_BITSET_VALUE]) != attr_nwords * sizeof(u32)) {
+		NL_SET_ERR_MSG_ATTR(extack, tb[ETHTOOL_A_BITSET_VALUE],
+				    "bitset value length does not match size");
+		return -EINVAL;
+	}
+	if (tb[ETHTOOL_A_BITSET_MASK] &&
+	    nla_len(tb[ETHTOOL_A_BITSET_MASK]) != attr_nwords * sizeof(u32)) {
+		NL_SET_ERR_MSG_ATTR(extack, tb[ETHTOOL_A_BITSET_MASK],
+				    "bitset mask length does not match size");
+		return -EINVAL;
+	}
+	if (attr_nbits <= nbits)
+		return 0;
+
+	test_attr = no_mask ? tb[ETHTOOL_A_BITSET_VALUE] :
+			      tb[ETHTOOL_A_BITSET_MASK];
+	if (ethnl_bitmap32_not_zero(nla_data(test_attr), nbits, attr_nbits)) {
+		NL_SET_ERR_MSG_ATTR(extack, test_attr,
+				    "cannot modify bits past kernel bitset size");
+		return -EINVAL;
+	}
+	return 0;
+}
+
+/**
+ * ethnl_update_bitset32() - Apply a bitset nest to a u32 based bitmap
+ * @bitmap:  bitmap to update
+ * @nbits:   size of the updated bitmap in bits
+ * @attr:    nest attribute to parse and apply
+ * @names:   array of bit names; may be null for compact format
+ * @extack:  extack for error reporting
+ * @mod:     set this to true if bitmap is modified, leave as it is if not
+ *
+ * Apply bitset netsted attribute to a bitmap. If the attribute represents
+ * a bit list, @bitmap is set to its contents; otherwise, bits in mask are
+ * set to values from value. Bitmaps in the attribute may be longer than
+ * @nbits but the message must not request modifying any bits past @nbits.
+ *
+ * Return: negative error code on failure, 0 on success
+ */
+int ethnl_update_bitset32(u32 *bitmap, unsigned int nbits,
+			  const struct nlattr *attr, ethnl_string_array_t names,
+			  struct netlink_ext_ack *extack, bool *mod)
+{
+	struct nlattr *tb[ETHTOOL_A_BITSET_MAX + 1];
+	unsigned int change_bits;
+	bool no_mask;
+	int ret;
+
+	if (!attr)
+		return 0;
+	ret = nla_parse_nested(tb, ETHTOOL_A_BITSET_MAX, attr, bitset_policy,
+			       extack);
+	if (ret < 0)
+		return ret;
+
+	if (tb[ETHTOOL_A_BITSET_BITS])
+		return ethnl_update_bitset32_verbose(bitmap, nbits, attr, tb,
+						     names, extack, mod);
+	ret = ethnl_compact_sanity_checks(nbits, attr, tb, extack);
+	if (ret < 0)
+		return ret;
+
+	no_mask = tb[ETHTOOL_A_BITSET_NOMASK];
+	change_bits = min_t(unsigned int,
+			    nla_get_u32(tb[ETHTOOL_A_BITSET_SIZE]), nbits);
+	ethnl_bitmap32_update(bitmap, change_bits,
+			      nla_data(tb[ETHTOOL_A_BITSET_VALUE]),
+			      no_mask ? NULL :
+					nla_data(tb[ETHTOOL_A_BITSET_MASK]),
+			      mod);
+	if (no_mask && change_bits < nbits)
+		ethnl_bitmap32_clear(bitmap, change_bits, nbits, mod);
+
+	return 0;
+}
+
+#if BITS_PER_LONG == 64 && defined(__BIG_ENDIAN)
+
+/* 64-bit big endian architectures are the only case when u32 based bitmaps
+ * and unsigned long based bitmaps have different memory layout so that we
+ * cannot simply cast the latter to the former and need actual wrappers
+ * converting the latter to the former.
+ *
+ * To reduce the number of slab allocations, the wrappers use fixed size local
+ * variables for bitmaps up to ETHNL_SMALL_BITMAP_BITS bits which is the
+ * majority of bitmaps used by ethtool.
+ */
+#define ETHNL_SMALL_BITMAP_BITS 128
+#define ETHNL_SMALL_BITMAP_WORDS DIV_ROUND_UP(ETHNL_SMALL_BITMAP_BITS, 32)
+
+int ethnl_bitset_size(const unsigned long *val, const unsigned long *mask,
+		      unsigned int nbits, ethnl_string_array_t names,
+		      bool compact)
+{
+	u32 small_mask32[ETHNL_SMALL_BITMAP_WORDS];
+	u32 small_val32[ETHNL_SMALL_BITMAP_WORDS];
+	u32 *mask32;
+	u32 *val32;
+	int ret;
+
+	if (nbits > ETHNL_SMALL_BITMAP_BITS) {
+		unsigned int nwords = DIV_ROUND_UP(nbits, 32);
+
+		val32 = kmalloc_array(2 * nwords, sizeof(u32), GFP_KERNEL);
+		if (!val32)
+			return -ENOMEM;
+		mask32 = val32 + nwords;
+	} else {
+		val32 = small_val32;
+		mask32 = small_mask32;
+	}
+
+	bitmap_to_arr32(val32, val, nbits);
+	if (mask)
+		bitmap_to_arr32(mask32, mask, nbits);
+	else
+		mask32 = NULL;
+	ret = ethnl_bitset32_size(val32, mask32, nbits, names, compact);
+
+	if (nbits > ETHNL_SMALL_BITMAP_BITS)
+		kfree(val32);
+
+	return ret;
+}
+
+int ethnl_put_bitset(struct sk_buff *skb, int attrtype,
+		     const unsigned long *val, const unsigned long *mask,
+		     unsigned int nbits, ethnl_string_array_t names,
+		     bool compact)
+{
+	u32 small_mask32[ETHNL_SMALL_BITMAP_WORDS];
+	u32 small_val32[ETHNL_SMALL_BITMAP_WORDS];
+	u32 *mask32;
+	u32 *val32;
+	int ret;
+
+	if (nbits > ETHNL_SMALL_BITMAP_BITS) {
+		unsigned int nwords = DIV_ROUND_UP(nbits, 32);
+
+		val32 = kmalloc_array(2 * nwords, sizeof(u32), GFP_KERNEL);
+		if (!val32)
+			return -ENOMEM;
+		mask32 = val32 + nwords;
+	} else {
+		val32 = small_val32;
+		mask32 = small_mask32;
+	}
+
+	bitmap_to_arr32(val32, val, nbits);
+	if (mask)
+		bitmap_to_arr32(mask32, mask, nbits);
+	else
+		mask32 = NULL;
+	ret = ethnl_put_bitset32(skb, attrtype, val32, mask32, nbits, names,
+				 compact);
+
+	if (nbits > ETHNL_SMALL_BITMAP_BITS)
+		kfree(val32);
+
+	return ret;
+}
+
+int ethnl_update_bitset(unsigned long *bitmap, unsigned int nbits,
+			const struct nlattr *attr, ethnl_string_array_t names,
+			struct netlink_ext_ack *extack, bool *mod)
+{
+	u32 small_bitmap32[ETHNL_SMALL_BITMAP_WORDS];
+	u32 *bitmap32 = small_bitmap32;
+	bool u32_mod = false;
+	int ret;
+
+	if (nbits > ETHNL_SMALL_BITMAP_BITS) {
+		unsigned int dst_words = DIV_ROUND_UP(nbits, 32);
+
+		bitmap32 = kmalloc_array(dst_words, sizeof(u32), GFP_KERNEL);
+		if (!bitmap32)
+			return -ENOMEM;
+	}
+
+	bitmap_to_arr32(bitmap32, bitmap, nbits);
+	ret = ethnl_update_bitset32(bitmap32, nbits, attr, names, extack,
+				    &u32_mod);
+	if (u32_mod) {
+		bitmap_from_arr32(bitmap, bitmap32, nbits);
+		*mod = true;
+	}
+
+	if (nbits > ETHNL_SMALL_BITMAP_BITS)
+		kfree(bitmap32);
+
+	return ret;
+}
+
+#else
+
+/* On little endian 64-bit and all 32-bit architectures, an unsigned long
+ * based bitmap can be interpreted as u32 based one using a simple cast.
+ */
+
+int ethnl_bitset_size(const unsigned long *val, const unsigned long *mask,
+		      unsigned int nbits, ethnl_string_array_t names,
+		      bool compact)
+{
+	return ethnl_bitset32_size((const u32 *)val, (const u32 *)mask, nbits,
+				   names, compact);
+}
+
+int ethnl_put_bitset(struct sk_buff *skb, int attrtype,
+		     const unsigned long *val, const unsigned long *mask,
+		     unsigned int nbits, ethnl_string_array_t names,
+		     bool compact)
+{
+	return ethnl_put_bitset32(skb, attrtype, (const u32 *)val,
+				  (const u32 *)mask, nbits, names, compact);
+}
+
+int ethnl_update_bitset(unsigned long *bitmap, unsigned int nbits,
+			const struct nlattr *attr, ethnl_string_array_t names,
+			struct netlink_ext_ack *extack, bool *mod)
+{
+	return ethnl_update_bitset32((u32 *)bitmap, nbits, attr, names, extack,
+				     mod);
+}
+
+#endif /* BITS_PER_LONG == 64 && defined(__BIG_ENDIAN) */
diff --git a/net/ethtool/bitset.h b/net/ethtool/bitset.h
new file mode 100644
index 000000000000..b8247e34109d
--- /dev/null
+++ b/net/ethtool/bitset.h
@@ -0,0 +1,28 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+
+#ifndef _NET_ETHTOOL_BITSET_H
+#define _NET_ETHTOOL_BITSET_H
+
+typedef const char (*const ethnl_string_array_t)[ETH_GSTRING_LEN];
+
+int ethnl_bitset_is_compact(const struct nlattr *bitset, bool *compact);
+int ethnl_bitset_size(const unsigned long *val, const unsigned long *mask,
+		      unsigned int nbits, ethnl_string_array_t names,
+		      bool compact);
+int ethnl_bitset32_size(const u32 *val, const u32 *mask, unsigned int nbits,
+			ethnl_string_array_t names, bool compact);
+int ethnl_put_bitset(struct sk_buff *skb, int attrtype,
+		     const unsigned long *val, const unsigned long *mask,
+		     unsigned int nbits, ethnl_string_array_t names,
+		     bool compact);
+int ethnl_put_bitset32(struct sk_buff *skb, int attrtype, const u32 *val,
+		       const u32 *mask, unsigned int nbits,
+		       ethnl_string_array_t names, bool compact);
+int ethnl_update_bitset(unsigned long *bitmap, unsigned int nbits,
+			const struct nlattr *attr, ethnl_string_array_t names,
+			struct netlink_ext_ack *extack, bool *mod);
+int ethnl_update_bitset32(u32 *bitmap, unsigned int nbits,
+			  const struct nlattr *attr, ethnl_string_array_t names,
+			  struct netlink_ext_ack *extack, bool *mod);
+
+#endif /* _NET_ETHTOOL_BITSET_H */
-- 
cgit v1.2.3-59-g8ed1b


From 6b08d6c146f4c5ed451c45339c10feb06d619db2 Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Fri, 27 Dec 2019 15:55:33 +0100
Subject: ethtool: support for netlink notifications

Add infrastructure for ethtool netlink notifications. There is only one
multicast group "monitor" which is used to notify userspace about changes
and actions performed. Notification messages (types using suffix _NTF)
share the format with replies to GET requests.

Notifications are supposed to be broadcasted on every configuration change,
whether it is done using the netlink interface or ioctl one. Netlink SET
requests only trigger a notification if some data is actually changed.

To trigger an ethtool notification, both ethtool netlink and external code
use ethtool_notify() helper. This helper requires RTNL to be held and may
sleep. Handlers sending messages for specific notification message types
are registered in ethnl_notify_handlers array. As notifications can be
triggered from other code, ethnl_ok flag is used to prevent an attempt to
send notification before genetlink family is registered.

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/ethtool_netlink.h      |  5 +++++
 include/linux/netdevice.h            |  9 +++++++++
 include/uapi/linux/ethtool_netlink.h |  2 ++
 net/ethtool/netlink.c                | 32 ++++++++++++++++++++++++++++++++
 4 files changed, 48 insertions(+)

diff --git a/include/linux/ethtool_netlink.h b/include/linux/ethtool_netlink.h
index f27e92b5f344..c98f6852c8eb 100644
--- a/include/linux/ethtool_netlink.h
+++ b/include/linux/ethtool_netlink.h
@@ -5,5 +5,10 @@
 
 #include <uapi/linux/ethtool_netlink.h>
 #include <linux/ethtool.h>
+#include <linux/netdevice.h>
+
+enum ethtool_multicast_groups {
+	ETHNL_MCGRP_MONITOR,
+};
 
 #endif /* _LINUX_ETHTOOL_NETLINK_H_ */
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 469a297b58c0..f007155ae8f4 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -4393,6 +4393,15 @@ struct netdev_notifier_bonding_info {
 void netdev_bonding_info_change(struct net_device *dev,
 				struct netdev_bonding_info *bonding_info);
 
+#if IS_ENABLED(CONFIG_ETHTOOL_NETLINK)
+void ethtool_notify(struct net_device *dev, unsigned int cmd, const void *data);
+#else
+static inline void ethtool_notify(struct net_device *dev, unsigned int cmd,
+				  const void *data)
+{
+}
+#endif
+
 static inline
 struct sk_buff *skb_gso_segment(struct sk_buff *skb, netdev_features_t features)
 {
diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
index 951203049615..d530ccb6f7e6 100644
--- a/include/uapi/linux/ethtool_netlink.h
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -89,4 +89,6 @@ enum {
 #define ETHTOOL_GENL_NAME "ethtool"
 #define ETHTOOL_GENL_VERSION 1
 
+#define ETHTOOL_MCGRP_MONITOR_NAME "monitor"
+
 #endif /* _UAPI_LINUX_ETHTOOL_NETLINK_H_ */
diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index aef882e0c3f5..c0f25c8f3565 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -6,6 +6,8 @@
 
 static struct genl_family ethtool_genl_family;
 
+static bool ethnl_ok __read_mostly;
+
 static const struct nla_policy ethnl_header_policy[ETHTOOL_A_HEADER_MAX + 1] = {
 	[ETHTOOL_A_HEADER_UNSPEC]	= { .type = NLA_REJECT },
 	[ETHTOOL_A_HEADER_DEV_INDEX]	= { .type = NLA_U32 },
@@ -169,11 +171,38 @@ err:
 	return NULL;
 }
 
+/* notifications */
+
+typedef void (*ethnl_notify_handler_t)(struct net_device *dev, unsigned int cmd,
+				       const void *data);
+
+static const ethnl_notify_handler_t ethnl_notify_handlers[] = {
+};
+
+void ethtool_notify(struct net_device *dev, unsigned int cmd, const void *data)
+{
+	if (unlikely(!ethnl_ok))
+		return;
+	ASSERT_RTNL();
+
+	if (likely(cmd < ARRAY_SIZE(ethnl_notify_handlers) &&
+		   ethnl_notify_handlers[cmd]))
+		ethnl_notify_handlers[cmd](dev, cmd, data);
+	else
+		WARN_ONCE(1, "notification %u not implemented (dev=%s)\n",
+			  cmd, netdev_name(dev));
+}
+EXPORT_SYMBOL(ethtool_notify);
+
 /* genetlink setup */
 
 static const struct genl_ops ethtool_genl_ops[] = {
 };
 
+static const struct genl_multicast_group ethtool_nl_mcgrps[] = {
+	[ETHNL_MCGRP_MONITOR] = { .name = ETHTOOL_MCGRP_MONITOR_NAME },
+};
+
 static struct genl_family ethtool_genl_family = {
 	.name		= ETHTOOL_GENL_NAME,
 	.version	= ETHTOOL_GENL_VERSION,
@@ -181,6 +210,8 @@ static struct genl_family ethtool_genl_family = {
 	.parallel_ops	= true,
 	.ops		= ethtool_genl_ops,
 	.n_ops		= ARRAY_SIZE(ethtool_genl_ops),
+	.mcgrps		= ethtool_nl_mcgrps,
+	.n_mcgrps	= ARRAY_SIZE(ethtool_nl_mcgrps),
 };
 
 /* module setup */
@@ -192,6 +223,7 @@ static int __init ethnl_init(void)
 	ret = genl_register_family(&ethtool_genl_family);
 	if (WARN(ret < 0, "ethtool: genetlink family registration failed"))
 		return ret;
+	ethnl_ok = true;
 
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 728480f1244200fe294073afeb612c583a2080d2 Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Fri, 27 Dec 2019 15:55:38 +0100
Subject: ethtool: default handlers for GET requests

Significant part of GET request processing is common for most request
types but unfortunately it cannot be easily separated from type specific
code as we need to alternate between common actions (parsing common request
header, allocating message and filling netlink/genetlink headers etc.) and
specific actions (querying the device, composing the reply). The processing
also happens in three different situations: "do" request, "dump" request
and notification, each doing things in slightly different way.

The request specific code is implemented in four or five callbacks defined
in an instance of struct get_request_ops:

  parse_request() - parse incoming message
  prepare_data()  - retrieve data from driver or NIC
  reply_size()    - estimate reply message size
  fill_reply()    - compose reply message
  cleanup_data()  - (optional) clean up additional data

Other members of struct get_request_ops describe the data structure holding
information from client request and data used to compose the message. The
default handlers ethnl_default_doit(), ethnl_default_dumpit(),
ethnl_default_start() and ethnl_default_done() can be then used in genl_ops
handler. Notification handler will be introduced in a later patch.

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ethtool/netlink.c | 321 ++++++++++++++++++++++++++++++++++++++++++++++++++
 net/ethtool/netlink.h | 116 +++++++++++++++++-
 2 files changed, 436 insertions(+), 1 deletion(-)

diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index c0f25c8f3565..ae63e8423072 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -171,6 +171,327 @@ err:
 	return NULL;
 }
 
+/* GET request helpers */
+
+/**
+ * struct ethnl_dump_ctx - context structure for generic dumpit() callback
+ * @ops:      request ops of currently processed message type
+ * @req_info: parsed request header of processed request
+ * @pos_hash: saved iteration position - hashbucket
+ * @pos_idx:  saved iteration position - index
+ *
+ * These parameters are kept in struct netlink_callback as context preserved
+ * between iterations. They are initialized by ethnl_default_start() and used
+ * in ethnl_default_dumpit() and ethnl_default_done().
+ */
+struct ethnl_dump_ctx {
+	const struct ethnl_request_ops	*ops;
+	struct ethnl_req_info		*req_info;
+	struct ethnl_reply_data		*reply_data;
+	int				pos_hash;
+	int				pos_idx;
+};
+
+static const struct ethnl_request_ops *
+ethnl_default_requests[__ETHTOOL_MSG_USER_CNT] = {
+};
+
+static struct ethnl_dump_ctx *ethnl_dump_context(struct netlink_callback *cb)
+{
+	return (struct ethnl_dump_ctx *)cb->ctx;
+}
+
+/**
+ * ethnl_default_parse() - Parse request message
+ * @req_info:    pointer to structure to put data into
+ * @nlhdr:       pointer to request message header
+ * @net:         request netns
+ * @request_ops: struct request_ops for request type
+ * @extack:      netlink extack for error reporting
+ * @require_dev: fail if no device identified in header
+ *
+ * Parse universal request header and call request specific ->parse_request()
+ * callback (if defined) to parse the rest of the message.
+ *
+ * Return: 0 on success or negative error code
+ */
+static int ethnl_default_parse(struct ethnl_req_info *req_info,
+			       const struct nlmsghdr *nlhdr, struct net *net,
+			       const struct ethnl_request_ops *request_ops,
+			       struct netlink_ext_ack *extack, bool require_dev)
+{
+	struct nlattr **tb;
+	int ret;
+
+	tb = kmalloc_array(request_ops->max_attr + 1, sizeof(tb[0]),
+			   GFP_KERNEL);
+	if (!tb)
+		return -ENOMEM;
+
+	ret = nlmsg_parse(nlhdr, GENL_HDRLEN, tb, request_ops->max_attr,
+			  request_ops->request_policy, extack);
+	if (ret < 0)
+		goto out;
+	ret = ethnl_parse_header(req_info, tb[request_ops->hdr_attr], net,
+				 extack, require_dev);
+	if (ret < 0)
+		goto out;
+
+	if (request_ops->parse_request) {
+		ret = request_ops->parse_request(req_info, tb, extack);
+		if (ret < 0)
+			goto out;
+	}
+
+	ret = 0;
+out:
+	kfree(tb);
+	return ret;
+}
+
+/**
+ * ethnl_init_reply_data() - Initialize reply data for GET request
+ * @req_info: pointer to embedded struct ethnl_req_info
+ * @ops:      instance of struct ethnl_request_ops describing the layout
+ * @dev:      network device to initialize the reply for
+ *
+ * Fills the reply data part with zeros and sets the dev member. Must be called
+ * before calling the ->fill_reply() callback (for each iteration when handling
+ * dump requests).
+ */
+static void ethnl_init_reply_data(struct ethnl_reply_data *reply_data,
+				  const struct ethnl_request_ops *ops,
+				  struct net_device *dev)
+{
+	memset(reply_data, 0, ops->reply_data_size);
+	reply_data->dev = dev;
+}
+
+/* default ->doit() handler for GET type requests */
+static int ethnl_default_doit(struct sk_buff *skb, struct genl_info *info)
+{
+	struct ethnl_reply_data *reply_data = NULL;
+	struct ethnl_req_info *req_info = NULL;
+	const u8 cmd = info->genlhdr->cmd;
+	const struct ethnl_request_ops *ops;
+	struct sk_buff *rskb;
+	void *reply_payload;
+	int reply_len;
+	int ret;
+
+	ops = ethnl_default_requests[cmd];
+	if (WARN_ONCE(!ops, "cmd %u has no ethnl_request_ops\n", cmd))
+		return -EOPNOTSUPP;
+	req_info = kzalloc(ops->req_info_size, GFP_KERNEL);
+	if (!req_info)
+		return -ENOMEM;
+	reply_data = kmalloc(ops->reply_data_size, GFP_KERNEL);
+	if (!reply_data) {
+		kfree(req_info);
+		return -ENOMEM;
+	}
+
+	ret = ethnl_default_parse(req_info, info->nlhdr, genl_info_net(info), ops,
+				  info->extack, !ops->allow_nodev_do);
+	if (ret < 0)
+		goto err_dev;
+	ethnl_init_reply_data(reply_data, ops, req_info->dev);
+
+	rtnl_lock();
+	ret = ops->prepare_data(req_info, reply_data, info);
+	rtnl_unlock();
+	if (ret < 0)
+		goto err_cleanup;
+	reply_len = ops->reply_size(req_info, reply_data);
+	if (ret < 0)
+		goto err_cleanup;
+	ret = -ENOMEM;
+	rskb = ethnl_reply_init(reply_len, req_info->dev, ops->reply_cmd,
+				ops->hdr_attr, info, &reply_payload);
+	if (!rskb)
+		goto err_cleanup;
+	ret = ops->fill_reply(rskb, req_info, reply_data);
+	if (ret < 0)
+		goto err_msg;
+	if (ops->cleanup_data)
+		ops->cleanup_data(reply_data);
+
+	genlmsg_end(rskb, reply_payload);
+	if (req_info->dev)
+		dev_put(req_info->dev);
+	kfree(reply_data);
+	kfree(req_info);
+	return genlmsg_reply(rskb, info);
+
+err_msg:
+	WARN_ONCE(ret == -EMSGSIZE, "calculated message payload length (%d) not sufficient\n", reply_len);
+	nlmsg_free(rskb);
+err_cleanup:
+	if (ops->cleanup_data)
+		ops->cleanup_data(reply_data);
+err_dev:
+	if (req_info->dev)
+		dev_put(req_info->dev);
+	kfree(reply_data);
+	kfree(req_info);
+	return ret;
+}
+
+static int ethnl_default_dump_one(struct sk_buff *skb, struct net_device *dev,
+				  const struct ethnl_dump_ctx *ctx)
+{
+	int ret;
+
+	ethnl_init_reply_data(ctx->reply_data, ctx->ops, dev);
+	rtnl_lock();
+	ret = ctx->ops->prepare_data(ctx->req_info, ctx->reply_data, NULL);
+	rtnl_unlock();
+	if (ret < 0)
+		goto out;
+	ret = ethnl_fill_reply_header(skb, dev, ctx->ops->hdr_attr);
+	if (ret < 0)
+		goto out;
+	ret = ctx->ops->fill_reply(skb, ctx->req_info, ctx->reply_data);
+
+out:
+	if (ctx->ops->cleanup_data)
+		ctx->ops->cleanup_data(ctx->reply_data);
+	ctx->reply_data->dev = NULL;
+	return ret;
+}
+
+/* Default ->dumpit() handler for GET requests. Device iteration copied from
+ * rtnl_dump_ifinfo(); we have to be more careful about device hashtable
+ * persistence as we cannot guarantee to hold RTNL lock through the whole
+ * function as rtnetnlink does.
+ */
+static int ethnl_default_dumpit(struct sk_buff *skb,
+				struct netlink_callback *cb)
+{
+	struct ethnl_dump_ctx *ctx = ethnl_dump_context(cb);
+	struct net *net = sock_net(skb->sk);
+	int s_idx = ctx->pos_idx;
+	int h, idx = 0;
+	int ret = 0;
+	void *ehdr;
+
+	rtnl_lock();
+	for (h = ctx->pos_hash; h < NETDEV_HASHENTRIES; h++, s_idx = 0) {
+		struct hlist_head *head;
+		struct net_device *dev;
+		unsigned int seq;
+
+		head = &net->dev_index_head[h];
+
+restart_chain:
+		seq = net->dev_base_seq;
+		cb->seq = seq;
+		idx = 0;
+		hlist_for_each_entry(dev, head, index_hlist) {
+			if (idx < s_idx)
+				goto cont;
+			dev_hold(dev);
+			rtnl_unlock();
+
+			ehdr = genlmsg_put(skb, NETLINK_CB(cb->skb).portid,
+					   cb->nlh->nlmsg_seq,
+					   &ethtool_genl_family, 0,
+					   ctx->ops->reply_cmd);
+			if (!ehdr) {
+				dev_put(dev);
+				ret = -EMSGSIZE;
+				goto out;
+			}
+			ret = ethnl_default_dump_one(skb, dev, ctx);
+			dev_put(dev);
+			if (ret < 0) {
+				genlmsg_cancel(skb, ehdr);
+				if (ret == -EOPNOTSUPP)
+					goto lock_and_cont;
+				if (likely(skb->len))
+					ret = skb->len;
+				goto out;
+			}
+			genlmsg_end(skb, ehdr);
+lock_and_cont:
+			rtnl_lock();
+			if (net->dev_base_seq != seq) {
+				s_idx = idx + 1;
+				goto restart_chain;
+			}
+cont:
+			idx++;
+		}
+
+	}
+	rtnl_unlock();
+
+out:
+	ctx->pos_hash = h;
+	ctx->pos_idx = idx;
+	nl_dump_check_consistent(cb, nlmsg_hdr(skb));
+
+	return ret;
+}
+
+/* generic ->start() handler for GET requests */
+static int ethnl_default_start(struct netlink_callback *cb)
+{
+	struct ethnl_dump_ctx *ctx = ethnl_dump_context(cb);
+	struct ethnl_reply_data *reply_data;
+	const struct ethnl_request_ops *ops;
+	struct ethnl_req_info *req_info;
+	struct genlmsghdr *ghdr;
+	int ret;
+
+	BUILD_BUG_ON(sizeof(*ctx) > sizeof(cb->ctx));
+
+	ghdr = nlmsg_data(cb->nlh);
+	ops = ethnl_default_requests[ghdr->cmd];
+	if (WARN_ONCE(!ops, "cmd %u has no ethnl_request_ops\n", ghdr->cmd))
+		return -EOPNOTSUPP;
+	req_info = kzalloc(ops->req_info_size, GFP_KERNEL);
+	if (!req_info)
+		return -ENOMEM;
+	reply_data = kmalloc(ops->reply_data_size, GFP_KERNEL);
+	if (!reply_data) {
+		kfree(req_info);
+		return -ENOMEM;
+	}
+
+	ret = ethnl_default_parse(req_info, cb->nlh, sock_net(cb->skb->sk), ops,
+				  cb->extack, false);
+	if (req_info->dev) {
+		/* We ignore device specification in dump requests but as the
+		 * same parser as for non-dump (doit) requests is used, it
+		 * would take reference to the device if it finds one
+		 */
+		dev_put(req_info->dev);
+		req_info->dev = NULL;
+	}
+	if (ret < 0)
+		return ret;
+
+	ctx->ops = ops;
+	ctx->req_info = req_info;
+	ctx->reply_data = reply_data;
+	ctx->pos_hash = 0;
+	ctx->pos_idx = 0;
+
+	return 0;
+}
+
+/* default ->done() handler for GET requests */
+static int ethnl_default_done(struct netlink_callback *cb)
+{
+	struct ethnl_dump_ctx *ctx = ethnl_dump_context(cb);
+
+	kfree(ctx->reply_data);
+	kfree(ctx->req_info);
+
+	return 0;
+}
+
 /* notifications */
 
 typedef void (*ethnl_notify_handler_t)(struct net_device *dev, unsigned int cmd,
diff --git a/net/ethtool/netlink.h b/net/ethtool/netlink.h
index 05d7183da894..6ec0dd06277f 100644
--- a/net/ethtool/netlink.h
+++ b/net/ethtool/netlink.h
@@ -200,16 +200,130 @@ static inline unsigned int ethnl_reply_header_size(void)
 			      nla_total_size(IFNAMSIZ));
 }
 
+/* GET request handling */
+
+/* Unified processing of GET requests uses two data structures: request info
+ * and reply data. Request info holds information parsed from client request
+ * and its stays constant through all request processing. Reply data holds data
+ * retrieved from ethtool_ops callbacks or other internal sources which is used
+ * to compose the reply. When processing a dump request, request info is filled
+ * only once (when the request message is parsed) but reply data is filled for
+ * each reply message.
+ *
+ * Both structures consist of part common for all request types (struct
+ * ethnl_req_info and struct ethnl_reply_data defined below) and optional
+ * parts specific for each request type. Common part always starts at offset 0.
+ */
+
 /**
  * struct ethnl_req_info - base type of request information for GET requests
  * @dev:   network device the request is for (may be null)
  * @flags: request flags common for all request types
  *
- * This is a common base, additional members may follow after this structure.
+ * This is a common base for request specific structures holding data from
+ * parsed userspace request. These always embed struct ethnl_req_info at
+ * zero offset.
  */
 struct ethnl_req_info {
 	struct net_device	*dev;
 	u32			flags;
 };
 
+/**
+ * struct ethnl_reply_data - base type of reply data for GET requests
+ * @dev:       device for current reply message; in single shot requests it is
+ *             equal to &ethnl_req_info.dev; in dumps it's different for each
+ *             reply message
+ *
+ * This is a common base for request specific structures holding data for
+ * kernel reply message. These always embed struct ethnl_reply_data at zero
+ * offset.
+ */
+struct ethnl_reply_data {
+	struct net_device		*dev;
+};
+
+static inline int ethnl_ops_begin(struct net_device *dev)
+{
+	if (dev && dev->ethtool_ops->begin)
+		return dev->ethtool_ops->begin(dev);
+	else
+		return 0;
+}
+
+static inline void ethnl_ops_complete(struct net_device *dev)
+{
+	if (dev && dev->ethtool_ops->complete)
+		dev->ethtool_ops->complete(dev);
+}
+
+/**
+ * struct ethnl_request_ops - unified handling of GET requests
+ * @request_cmd:      command id for request (GET)
+ * @reply_cmd:        command id for reply (GET_REPLY)
+ * @hdr_attr:         attribute type for request header
+ * @max_attr:         maximum (top level) attribute type
+ * @req_info_size:    size of request info
+ * @reply_data_size:  size of reply data
+ * @request_policy:   netlink policy for message contents
+ * @allow_nodev_do:   allow non-dump request with no device identification
+ * @parse_request:
+ *	Parse request except common header (struct ethnl_req_info). Common
+ *	header is already filled on entry, the rest up to @repdata_offset
+ *	is zero initialized. This callback should only modify type specific
+ *	request info by parsed attributes from request message.
+ * @prepare_data:
+ *	Retrieve and prepare data needed to compose a reply message. Calls to
+ *	ethtool_ops handlers are limited to this callback. Common reply data
+ *	(struct ethnl_reply_data) is filled on entry, type specific part after
+ *	it is zero initialized. This callback should only modify the type
+ *	specific part of reply data. Device identification from struct
+ *	ethnl_reply_data is to be used as for dump requests, it iterates
+ *	through network devices while dev member of struct ethnl_req_info
+ *	points to the device from client request.
+ * @reply_size:
+ *	Estimate reply message size. Returned value must be sufficient for
+ *	message payload without common reply header. The callback may returned
+ *	estimate higher than actual message size if exact calculation would
+ *	not be worth the saved memory space.
+ * @fill_reply:
+ *	Fill reply message payload (except for common header) from reply data.
+ *	The callback must not generate more payload than previously called
+ *	->reply_size() estimated.
+ * @cleanup_data:
+ *	Optional cleanup called when reply data is no longer needed. Can be
+ *	used e.g. to free any additional data structures outside the main
+ *	structure which were allocated by ->prepare_data(). When processing
+ *	dump requests, ->cleanup() is called for each message.
+ *
+ * Description of variable parts of GET request handling when using the
+ * unified infrastructure. When used, a pointer to an instance of this
+ * structure is to be added to &ethnl_default_requests array and generic
+ * handlers ethnl_default_doit(), ethnl_default_dumpit(),
+ * ethnl_default_start() and ethnl_default_done() used in @ethtool_genl_ops.
+ */
+struct ethnl_request_ops {
+	u8			request_cmd;
+	u8			reply_cmd;
+	u16			hdr_attr;
+	unsigned int		max_attr;
+	unsigned int		req_info_size;
+	unsigned int		reply_data_size;
+	const struct nla_policy *request_policy;
+	bool			allow_nodev_do;
+
+	int (*parse_request)(struct ethnl_req_info *req_info,
+			     struct nlattr **tb,
+			     struct netlink_ext_ack *extack);
+	int (*prepare_data)(const struct ethnl_req_info *req_info,
+			    struct ethnl_reply_data *reply_data,
+			    struct genl_info *info);
+	int (*reply_size)(const struct ethnl_req_info *req_info,
+			  const struct ethnl_reply_data *reply_data);
+	int (*fill_reply)(struct sk_buff *skb,
+			  const struct ethnl_req_info *req_info,
+			  const struct ethnl_reply_data *reply_data);
+	void (*cleanup_data)(struct ethnl_reply_data *reply_data);
+};
+
 #endif /* _NET_ETHTOOL_NETLINK_H */
-- 
cgit v1.2.3-59-g8ed1b


From 71921690f9745fef60a2bad425f30adf8cdc9da0 Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Fri, 27 Dec 2019 15:55:43 +0100
Subject: ethtool: provide string sets with STRSET_GET request

Requests a contents of one or more string sets, i.e. indexed arrays of
strings; this information is provided by ETHTOOL_GSSET_INFO and
ETHTOOL_GSTRINGS commands of ioctl interface. Unlike ioctl interface, all
information can be retrieved with one request and mulitple string sets can
be requested at once.

There are three types of requests:

  - no NLM_F_DUMP, no device: get "global" stringsets
  - no NLM_F_DUMP, with device: get string sets related to the device
  - NLM_F_DUMP, no device: get device related string sets for all devices

Client can request either all string sets of given type (global or device
related) or only specific sets. With ETHTOOL_A_STRSET_COUNTS flag set, only
set sizes (numbers of strings) are returned.

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/ethtool-netlink.rst |  75 ++++-
 include/uapi/linux/ethtool.h                 |   3 +
 include/uapi/linux/ethtool_netlink.h         |  56 ++++
 net/ethtool/Makefile                         |   2 +-
 net/ethtool/netlink.c                        |   8 +
 net/ethtool/netlink.h                        |   4 +
 net/ethtool/strset.c                         | 425 +++++++++++++++++++++++++++
 7 files changed, 570 insertions(+), 3 deletions(-)
 create mode 100644 net/ethtool/strset.c

diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
index 7797f1237472..3912cb0eb9c6 100644
--- a/Documentation/networking/ethtool-netlink.rst
+++ b/Documentation/networking/ethtool-netlink.rst
@@ -176,6 +176,18 @@ according to message purpose:
   ``_NTF``          kernel notification
   ==============    ======================================
 
+Userspace to kernel:
+
+  ===================================== ================================
+  ``ETHTOOL_MSG_STRSET_GET``            get string set
+  ===================================== ================================
+
+Kernel to userspace:
+
+  ===================================== ================================
+  ``ETHTOOL_MSG_STRSET_GET_REPLY``      string set contents
+  ===================================== ================================
+
 ``GET`` requests are sent by userspace applications to retrieve device
 information. They usually do not contain any message specific attributes.
 Kernel replies with corresponding "GET_REPLY" message. For most types, ``GET``
@@ -207,6 +219,65 @@ an ``ACT_REPLY`` message. Performing an action also triggers a notification
 Later sections describe the format and semantics of these messages.
 
 
+STRSET_GET
+==========
+
+Requests contents of a string set as provided by ioctl commands
+``ETHTOOL_GSSET_INFO`` and ``ETHTOOL_GSTRINGS.`` String sets are not user
+writeable so that the corresponding ``STRSET_SET`` message is only used in
+kernel replies. There are two types of string sets: global (independent of
+a device, e.g. device feature names) and device specific (e.g. device private
+flags).
+
+Request contents:
+
+ +---------------------------------------+--------+------------------------+
+ | ``ETHTOOL_A_STRSET_HEADER``           | nested | request header         |
+ +---------------------------------------+--------+------------------------+
+ | ``ETHTOOL_A_STRSET_STRINGSETS``       | nested | string set to request  |
+ +-+-------------------------------------+--------+------------------------+
+ | | ``ETHTOOL_A_STRINGSETS_STRINGSET+`` | nested | one string set         |
+ +-+-+-----------------------------------+--------+------------------------+
+ | | | ``ETHTOOL_A_STRINGSET_ID``        | u32    | set id                 |
+ +-+-+-----------------------------------+--------+------------------------+
+
+Kernel response contents:
+
+ +---------------------------------------+--------+-----------------------+
+ | ``ETHTOOL_A_STRSET_HEADER``           | nested | reply header          |
+ +---------------------------------------+--------+-----------------------+
+ | ``ETHTOOL_A_STRSET_STRINGSETS``       | nested | array of string sets  |
+ +-+-------------------------------------+--------+-----------------------+
+ | | ``ETHTOOL_A_STRINGSETS_STRINGSET+`` | nested | one string set        |
+ +-+-+-----------------------------------+--------+-----------------------+
+ | | | ``ETHTOOL_A_STRINGSET_ID``        | u32    | set id                |
+ +-+-+-----------------------------------+--------+-----------------------+
+ | | | ``ETHTOOL_A_STRINGSET_COUNT``     | u32    | number of strings     |
+ +-+-+-----------------------------------+--------+-----------------------+
+ | | | ``ETHTOOL_A_STRINGSET_STRINGS``   | nested | array of strings      |
+ +-+-+-+---------------------------------+--------+-----------------------+
+ | | | | ``ETHTOOL_A_STRINGS_STRING+``   | nested | one string            |
+ +-+-+-+-+-------------------------------+--------+-----------------------+
+ | | | | | ``ETHTOOL_A_STRING_INDEX``    | u32    | string index          |
+ +-+-+-+-+-------------------------------+--------+-----------------------+
+ | | | | | ``ETHTOOL_A_STRING_VALUE``    | string | string value          |
+ +-+-+-+-+-------------------------------+--------+-----------------------+
+ | ``ETHTOOL_A_STRSET_COUNTS_ONLY``      | flag   | return only counts    |
+ +---------------------------------------+--------+-----------------------+
+
+Device identification in request header is optional. Depending on its presence
+a and ``NLM_F_DUMP`` flag, there are three type of ``STRSET_GET`` requests:
+
+ - no ``NLM_F_DUMP,`` no device: get "global" stringsets
+ - no ``NLM_F_DUMP``, with device: get string sets related to the device
+ - ``NLM_F_DUMP``, no device: get device related string sets for all devices
+
+If there is no ``ETHTOOL_A_STRSET_STRINGSETS`` array, all string sets of
+requested type are returned, otherwise only those specified in the request.
+Flag ``ETHTOOL_A_STRSET_COUNTS_ONLY`` tells kernel to only return string
+counts of the sets, not the actual strings.
+
+
 Request translation
 ===================
 
@@ -242,7 +313,7 @@ have their netlink replacement yet.
   ``ETHTOOL_GSG``                     n/a
   ``ETHTOOL_SSG``                     n/a
   ``ETHTOOL_TEST``                    n/a
-  ``ETHTOOL_GSTRINGS``                n/a
+  ``ETHTOOL_GSTRINGS``                ``ETHTOOL_MSG_STRSET_GET``
   ``ETHTOOL_PHYS_ID``                 n/a
   ``ETHTOOL_GSTATS``                  n/a
   ``ETHTOOL_GTSO``                    n/a
@@ -270,7 +341,7 @@ have their netlink replacement yet.
   ``ETHTOOL_RESET``                   n/a
   ``ETHTOOL_SRXNTUPLE``               n/a
   ``ETHTOOL_GRXNTUPLE``               n/a
-  ``ETHTOOL_GSSET_INFO``              n/a
+  ``ETHTOOL_GSSET_INFO``              ``ETHTOOL_MSG_STRSET_GET``
   ``ETHTOOL_GRXFHINDIR``              n/a
   ``ETHTOOL_SRXFHINDIR``              n/a
   ``ETHTOOL_GFEATURES``               n/a
diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h
index f44155840b07..116bcbf09c74 100644
--- a/include/uapi/linux/ethtool.h
+++ b/include/uapi/linux/ethtool.h
@@ -606,6 +606,9 @@ enum ethtool_stringset {
 	ETH_SS_PHY_STATS,
 	ETH_SS_PHY_TUNABLES,
 	ETH_SS_LINK_MODES,
+
+	/* add new constants above here */
+	ETH_SS_COUNT
 };
 
 /**
diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
index d530ccb6f7e6..cabef1fec42a 100644
--- a/include/uapi/linux/ethtool_netlink.h
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -14,6 +14,7 @@
 /* message types - userspace to kernel */
 enum {
 	ETHTOOL_MSG_USER_NONE,
+	ETHTOOL_MSG_STRSET_GET,
 
 	/* add new constants above here */
 	__ETHTOOL_MSG_USER_CNT,
@@ -23,6 +24,7 @@ enum {
 /* message types - kernel to userspace */
 enum {
 	ETHTOOL_MSG_KERNEL_NONE,
+	ETHTOOL_MSG_STRSET_GET_REPLY,
 
 	/* add new constants above here */
 	__ETHTOOL_MSG_KERNEL_CNT,
@@ -85,6 +87,60 @@ enum {
 	ETHTOOL_A_BITSET_MAX = __ETHTOOL_A_BITSET_CNT - 1
 };
 
+/* string sets */
+
+enum {
+	ETHTOOL_A_STRING_UNSPEC,
+	ETHTOOL_A_STRING_INDEX,			/* u32 */
+	ETHTOOL_A_STRING_VALUE,			/* string */
+
+	/* add new constants above here */
+	__ETHTOOL_A_STRING_CNT,
+	ETHTOOL_A_STRING_MAX = __ETHTOOL_A_STRING_CNT - 1
+};
+
+enum {
+	ETHTOOL_A_STRINGS_UNSPEC,
+	ETHTOOL_A_STRINGS_STRING,		/* nest - _A_STRINGS_* */
+
+	/* add new constants above here */
+	__ETHTOOL_A_STRINGS_CNT,
+	ETHTOOL_A_STRINGS_MAX = __ETHTOOL_A_STRINGS_CNT - 1
+};
+
+enum {
+	ETHTOOL_A_STRINGSET_UNSPEC,
+	ETHTOOL_A_STRINGSET_ID,			/* u32 */
+	ETHTOOL_A_STRINGSET_COUNT,		/* u32 */
+	ETHTOOL_A_STRINGSET_STRINGS,		/* nest - _A_STRINGS_* */
+
+	/* add new constants above here */
+	__ETHTOOL_A_STRINGSET_CNT,
+	ETHTOOL_A_STRINGSET_MAX = __ETHTOOL_A_STRINGSET_CNT - 1
+};
+
+enum {
+	ETHTOOL_A_STRINGSETS_UNSPEC,
+	ETHTOOL_A_STRINGSETS_STRINGSET,		/* nest - _A_STRINGSET_* */
+
+	/* add new constants above here */
+	__ETHTOOL_A_STRINGSETS_CNT,
+	ETHTOOL_A_STRINGSETS_MAX = __ETHTOOL_A_STRINGSETS_CNT - 1
+};
+
+/* STRSET */
+
+enum {
+	ETHTOOL_A_STRSET_UNSPEC,
+	ETHTOOL_A_STRSET_HEADER,		/* nest - _A_HEADER_* */
+	ETHTOOL_A_STRSET_STRINGSETS,		/* nest - _A_STRINGSETS_* */
+	ETHTOOL_A_STRSET_COUNTS_ONLY,		/* flag */
+
+	/* add new constants above here */
+	__ETHTOOL_A_STRSET_CNT,
+	ETHTOOL_A_STRSET_MAX = __ETHTOOL_A_STRSET_CNT - 1
+};
+
 /* generic netlink info */
 #define ETHTOOL_GENL_NAME "ethtool"
 #define ETHTOOL_GENL_VERSION 1
diff --git a/net/ethtool/Makefile b/net/ethtool/Makefile
index a7e6c2c85db9..efcc42c34d62 100644
--- a/net/ethtool/Makefile
+++ b/net/ethtool/Makefile
@@ -4,4 +4,4 @@ obj-y				+= ioctl.o common.o
 
 obj-$(CONFIG_ETHTOOL_NETLINK)	+= ethtool_nl.o
 
-ethtool_nl-y	:= netlink.o bitset.o
+ethtool_nl-y	:= netlink.o bitset.o strset.o
diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index ae63e8423072..7dc082bde670 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -194,6 +194,7 @@ struct ethnl_dump_ctx {
 
 static const struct ethnl_request_ops *
 ethnl_default_requests[__ETHTOOL_MSG_USER_CNT] = {
+	[ETHTOOL_MSG_STRSET_GET]	= &ethnl_strset_request_ops,
 };
 
 static struct ethnl_dump_ctx *ethnl_dump_context(struct netlink_callback *cb)
@@ -518,6 +519,13 @@ EXPORT_SYMBOL(ethtool_notify);
 /* genetlink setup */
 
 static const struct genl_ops ethtool_genl_ops[] = {
+	{
+		.cmd	= ETHTOOL_MSG_STRSET_GET,
+		.doit	= ethnl_default_doit,
+		.start	= ethnl_default_start,
+		.dumpit	= ethnl_default_dumpit,
+		.done	= ethnl_default_done,
+	},
 };
 
 static const struct genl_multicast_group ethtool_nl_mcgrps[] = {
diff --git a/net/ethtool/netlink.h b/net/ethtool/netlink.h
index 6ec0dd06277f..44e9f63aefb7 100644
--- a/net/ethtool/netlink.h
+++ b/net/ethtool/netlink.h
@@ -326,4 +326,8 @@ struct ethnl_request_ops {
 	void (*cleanup_data)(struct ethnl_reply_data *reply_data);
 };
 
+/* request handlers */
+
+extern const struct ethnl_request_ops ethnl_strset_request_ops;
+
 #endif /* _NET_ETHTOOL_NETLINK_H */
diff --git a/net/ethtool/strset.c b/net/ethtool/strset.c
new file mode 100644
index 000000000000..9f2243329015
--- /dev/null
+++ b/net/ethtool/strset.c
@@ -0,0 +1,425 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include <linux/ethtool.h>
+#include <linux/phy.h>
+#include "netlink.h"
+#include "common.h"
+
+struct strset_info {
+	bool per_dev;
+	bool free_strings;
+	unsigned int count;
+	const char (*strings)[ETH_GSTRING_LEN];
+};
+
+static const struct strset_info info_template[] = {
+	[ETH_SS_TEST] = {
+		.per_dev	= true,
+	},
+	[ETH_SS_STATS] = {
+		.per_dev	= true,
+	},
+	[ETH_SS_PRIV_FLAGS] = {
+		.per_dev	= true,
+	},
+	[ETH_SS_FEATURES] = {
+		.per_dev	= false,
+		.count		= ARRAY_SIZE(netdev_features_strings),
+		.strings	= netdev_features_strings,
+	},
+	[ETH_SS_RSS_HASH_FUNCS] = {
+		.per_dev	= false,
+		.count		= ARRAY_SIZE(rss_hash_func_strings),
+		.strings	= rss_hash_func_strings,
+	},
+	[ETH_SS_TUNABLES] = {
+		.per_dev	= false,
+		.count		= ARRAY_SIZE(tunable_strings),
+		.strings	= tunable_strings,
+	},
+	[ETH_SS_PHY_STATS] = {
+		.per_dev	= true,
+	},
+	[ETH_SS_PHY_TUNABLES] = {
+		.per_dev	= false,
+		.count		= ARRAY_SIZE(phy_tunable_strings),
+		.strings	= phy_tunable_strings,
+	},
+	[ETH_SS_LINK_MODES] = {
+		.per_dev	= false,
+		.count		= __ETHTOOL_LINK_MODE_MASK_NBITS,
+		.strings	= link_mode_names,
+	},
+};
+
+struct strset_req_info {
+	struct ethnl_req_info		base;
+	u32				req_ids;
+	bool				counts_only;
+};
+
+#define STRSET_REQINFO(__req_base) \
+	container_of(__req_base, struct strset_req_info, base)
+
+struct strset_reply_data {
+	struct ethnl_reply_data		base;
+	struct strset_info		sets[ETH_SS_COUNT];
+};
+
+#define STRSET_REPDATA(__reply_base) \
+	container_of(__reply_base, struct strset_reply_data, base)
+
+static const struct nla_policy strset_get_policy[ETHTOOL_A_STRSET_MAX + 1] = {
+	[ETHTOOL_A_STRSET_UNSPEC]	= { .type = NLA_REJECT },
+	[ETHTOOL_A_STRSET_HEADER]	= { .type = NLA_NESTED },
+	[ETHTOOL_A_STRSET_STRINGSETS]	= { .type = NLA_NESTED },
+};
+
+static const struct nla_policy
+get_stringset_policy[ETHTOOL_A_STRINGSET_MAX + 1] = {
+	[ETHTOOL_A_STRINGSET_UNSPEC]	= { .type = NLA_REJECT },
+	[ETHTOOL_A_STRINGSET_ID]	= { .type = NLA_U32 },
+	[ETHTOOL_A_STRINGSET_COUNT]	= { .type = NLA_REJECT },
+	[ETHTOOL_A_STRINGSET_STRINGS]	= { .type = NLA_REJECT },
+};
+
+/**
+ * strset_include() - test if a string set should be included in reply
+ * @data: pointer to request data structure
+ * @id:   id of string set to check (ETH_SS_* constants)
+ */
+static bool strset_include(const struct strset_req_info *info,
+			   const struct strset_reply_data *data, u32 id)
+{
+	bool per_dev;
+
+	BUILD_BUG_ON(ETH_SS_COUNT >= BITS_PER_BYTE * sizeof(info->req_ids));
+
+	if (info->req_ids)
+		return info->req_ids & (1U << id);
+	per_dev = data->sets[id].per_dev;
+	if (!per_dev && !data->sets[id].strings)
+		return false;
+
+	return data->base.dev ? per_dev : !per_dev;
+}
+
+static int strset_get_id(const struct nlattr *nest, u32 *val,
+			 struct netlink_ext_ack *extack)
+{
+	struct nlattr *tb[ETHTOOL_A_STRINGSET_MAX + 1];
+	int ret;
+
+	ret = nla_parse_nested(tb, ETHTOOL_A_STRINGSET_MAX, nest,
+			       get_stringset_policy, extack);
+	if (ret < 0)
+		return ret;
+	if (!tb[ETHTOOL_A_STRINGSET_ID])
+		return -EINVAL;
+
+	*val = nla_get_u32(tb[ETHTOOL_A_STRINGSET_ID]);
+	return 0;
+}
+
+static const struct nla_policy
+strset_stringsets_policy[ETHTOOL_A_STRINGSETS_MAX + 1] = {
+	[ETHTOOL_A_STRINGSETS_UNSPEC]		= { .type = NLA_REJECT },
+	[ETHTOOL_A_STRINGSETS_STRINGSET]	= { .type = NLA_NESTED },
+};
+
+static int strset_parse_request(struct ethnl_req_info *req_base,
+				struct nlattr **tb,
+				struct netlink_ext_ack *extack)
+{
+	struct strset_req_info *req_info = STRSET_REQINFO(req_base);
+	struct nlattr *nest = tb[ETHTOOL_A_STRSET_STRINGSETS];
+	struct nlattr *attr;
+	int rem, ret;
+
+	if (!nest)
+		return 0;
+	ret = nla_validate_nested(nest, ETHTOOL_A_STRINGSETS_MAX,
+				  strset_stringsets_policy, extack);
+	if (ret < 0)
+		return ret;
+
+	req_info->counts_only = tb[ETHTOOL_A_STRSET_COUNTS_ONLY];
+	nla_for_each_nested(attr, nest, rem) {
+		u32 id;
+
+		if (WARN_ONCE(nla_type(attr) != ETHTOOL_A_STRINGSETS_STRINGSET,
+			      "unexpected attrtype %u in ETHTOOL_A_STRSET_STRINGSETS\n",
+			      nla_type(attr)))
+			return -EINVAL;
+
+		ret = strset_get_id(attr, &id, extack);
+		if (ret < 0)
+			return ret;
+		if (ret >= ETH_SS_COUNT) {
+			NL_SET_ERR_MSG_ATTR(extack, attr,
+					    "unknown string set id");
+			return -EOPNOTSUPP;
+		}
+
+		req_info->req_ids |= (1U << id);
+	}
+
+	return 0;
+}
+
+static void strset_cleanup_data(struct ethnl_reply_data *reply_base)
+{
+	struct strset_reply_data *data = STRSET_REPDATA(reply_base);
+	unsigned int i;
+
+	for (i = 0; i < ETH_SS_COUNT; i++)
+		if (data->sets[i].free_strings) {
+			kfree(data->sets[i].strings);
+			data->sets[i].strings = NULL;
+			data->sets[i].free_strings = false;
+		}
+}
+
+static int strset_prepare_set(struct strset_info *info, struct net_device *dev,
+			      unsigned int id, bool counts_only)
+{
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+	void *strings;
+	int count, ret;
+
+	if (id == ETH_SS_PHY_STATS && dev->phydev &&
+	    !ops->get_ethtool_phy_stats)
+		ret = phy_ethtool_get_sset_count(dev->phydev);
+	else if (ops->get_sset_count && ops->get_strings)
+		ret = ops->get_sset_count(dev, id);
+	else
+		ret = -EOPNOTSUPP;
+	if (ret <= 0) {
+		info->count = 0;
+		return 0;
+	}
+
+	count = ret;
+	if (!counts_only) {
+		strings = kcalloc(count, ETH_GSTRING_LEN, GFP_KERNEL);
+		if (!strings)
+			return -ENOMEM;
+		if (id == ETH_SS_PHY_STATS && dev->phydev &&
+		    !ops->get_ethtool_phy_stats)
+			phy_ethtool_get_strings(dev->phydev, strings);
+		else
+			ops->get_strings(dev, id, strings);
+		info->strings = strings;
+		info->free_strings = true;
+	}
+	info->count = count;
+
+	return 0;
+}
+
+static int strset_prepare_data(const struct ethnl_req_info *req_base,
+			       struct ethnl_reply_data *reply_base,
+			       struct genl_info *info)
+{
+	const struct strset_req_info *req_info = STRSET_REQINFO(req_base);
+	struct strset_reply_data *data = STRSET_REPDATA(reply_base);
+	struct net_device *dev = reply_base->dev;
+	unsigned int i;
+	int ret;
+
+	BUILD_BUG_ON(ARRAY_SIZE(info_template) != ETH_SS_COUNT);
+	memcpy(&data->sets, &info_template, sizeof(data->sets));
+
+	if (!dev) {
+		for (i = 0; i < ETH_SS_COUNT; i++) {
+			if ((req_info->req_ids & (1U << i)) &&
+			    data->sets[i].per_dev) {
+				if (info)
+					GENL_SET_ERR_MSG(info, "requested per device strings without dev");
+				return -EINVAL;
+			}
+		}
+	}
+
+	ret = ethnl_ops_begin(dev);
+	if (ret < 0)
+		goto err_strset;
+	for (i = 0; i < ETH_SS_COUNT; i++) {
+		if (!strset_include(req_info, data, i) ||
+		    !data->sets[i].per_dev)
+			continue;
+
+		ret = strset_prepare_set(&data->sets[i], dev, i,
+					 req_info->counts_only);
+		if (ret < 0)
+			goto err_ops;
+	}
+	ethnl_ops_complete(dev);
+
+	return 0;
+err_ops:
+	ethnl_ops_complete(dev);
+err_strset:
+	strset_cleanup_data(reply_base);
+	return ret;
+}
+
+/* calculate size of ETHTOOL_A_STRSET_STRINGSET nest for one string set */
+static int strset_set_size(const struct strset_info *info, bool counts_only)
+{
+	unsigned int len = 0;
+	unsigned int i;
+
+	if (info->count == 0)
+		return 0;
+	if (counts_only)
+		return nla_total_size(2 * nla_total_size(sizeof(u32)));
+
+	for (i = 0; i < info->count; i++) {
+		const char *str = info->strings[i];
+
+		/* ETHTOOL_A_STRING_INDEX, ETHTOOL_A_STRING_VALUE, nest */
+		len += nla_total_size(nla_total_size(sizeof(u32)) +
+				      ethnl_strz_size(str));
+	}
+	/* ETHTOOL_A_STRINGSET_ID, ETHTOOL_A_STRINGSET_COUNT */
+	len = 2 * nla_total_size(sizeof(u32)) + nla_total_size(len);
+
+	return nla_total_size(len);
+}
+
+static int strset_reply_size(const struct ethnl_req_info *req_base,
+			     const struct ethnl_reply_data *reply_base)
+{
+	const struct strset_req_info *req_info = STRSET_REQINFO(req_base);
+	const struct strset_reply_data *data = STRSET_REPDATA(reply_base);
+	unsigned int i;
+	int len = 0;
+	int ret;
+
+	len += ethnl_reply_header_size();
+	for (i = 0; i < ETH_SS_COUNT; i++) {
+		const struct strset_info *set_info = &data->sets[i];
+
+		if (!strset_include(req_info, data, i))
+			continue;
+
+		ret = strset_set_size(set_info, req_info->counts_only);
+		if (ret < 0)
+			return ret;
+		len += ret;
+	}
+
+	return len;
+}
+
+/* fill one string into reply */
+static int strset_fill_string(struct sk_buff *skb,
+			      const struct strset_info *set_info, u32 idx)
+{
+	struct nlattr *string_attr;
+	const char *value;
+
+	value = set_info->strings[idx];
+
+	string_attr = nla_nest_start(skb, ETHTOOL_A_STRINGS_STRING);
+	if (!string_attr)
+		return -EMSGSIZE;
+	if (nla_put_u32(skb, ETHTOOL_A_STRING_INDEX, idx) ||
+	    ethnl_put_strz(skb, ETHTOOL_A_STRING_VALUE, value))
+		goto nla_put_failure;
+	nla_nest_end(skb, string_attr);
+
+	return 0;
+nla_put_failure:
+	nla_nest_cancel(skb, string_attr);
+	return -EMSGSIZE;
+}
+
+/* fill one string set into reply */
+static int strset_fill_set(struct sk_buff *skb,
+			   const struct strset_info *set_info, u32 id,
+			   bool counts_only)
+{
+	struct nlattr *stringset_attr;
+	struct nlattr *strings_attr;
+	unsigned int i;
+
+	if (!set_info->per_dev && !set_info->strings)
+		return -EOPNOTSUPP;
+	if (set_info->count == 0)
+		return 0;
+	stringset_attr = nla_nest_start(skb, ETHTOOL_A_STRINGSETS_STRINGSET);
+	if (!stringset_attr)
+		return -EMSGSIZE;
+
+	if (nla_put_u32(skb, ETHTOOL_A_STRINGSET_ID, id) ||
+	    nla_put_u32(skb, ETHTOOL_A_STRINGSET_COUNT, set_info->count))
+		goto nla_put_failure;
+
+	if (!counts_only) {
+		strings_attr = nla_nest_start(skb, ETHTOOL_A_STRINGSET_STRINGS);
+		if (!strings_attr)
+			goto nla_put_failure;
+		for (i = 0; i < set_info->count; i++) {
+			if (strset_fill_string(skb, set_info, i) < 0)
+				goto nla_put_failure;
+		}
+		nla_nest_end(skb, strings_attr);
+	}
+
+	nla_nest_end(skb, stringset_attr);
+	return 0;
+
+nla_put_failure:
+	nla_nest_cancel(skb, stringset_attr);
+	return -EMSGSIZE;
+}
+
+static int strset_fill_reply(struct sk_buff *skb,
+			     const struct ethnl_req_info *req_base,
+			     const struct ethnl_reply_data *reply_base)
+{
+	const struct strset_req_info *req_info = STRSET_REQINFO(req_base);
+	const struct strset_reply_data *data = STRSET_REPDATA(reply_base);
+	struct nlattr *nest;
+	unsigned int i;
+	int ret;
+
+	nest = nla_nest_start(skb, ETHTOOL_A_STRSET_STRINGSETS);
+	if (!nest)
+		return -EMSGSIZE;
+
+	for (i = 0; i < ETH_SS_COUNT; i++) {
+		if (strset_include(req_info, data, i)) {
+			ret = strset_fill_set(skb, &data->sets[i], i,
+					      req_info->counts_only);
+			if (ret < 0)
+				goto nla_put_failure;
+		}
+	}
+
+	nla_nest_end(skb, nest);
+	return 0;
+
+nla_put_failure:
+	nla_nest_cancel(skb, nest);
+	return ret;
+}
+
+const struct ethnl_request_ops ethnl_strset_request_ops = {
+	.request_cmd		= ETHTOOL_MSG_STRSET_GET,
+	.reply_cmd		= ETHTOOL_MSG_STRSET_GET_REPLY,
+	.hdr_attr		= ETHTOOL_A_STRSET_HEADER,
+	.max_attr		= ETHTOOL_A_STRSET_MAX,
+	.req_info_size		= sizeof(struct strset_req_info),
+	.reply_data_size	= sizeof(struct strset_reply_data),
+	.request_policy		= strset_get_policy,
+	.allow_nodev_do		= true,
+
+	.parse_request		= strset_parse_request,
+	.prepare_data		= strset_prepare_data,
+	.reply_size		= strset_reply_size,
+	.fill_reply		= strset_fill_reply,
+	.cleanup_data		= strset_cleanup_data,
+};
-- 
cgit v1.2.3-59-g8ed1b


From 459e0b81b37043545d90629fdfb243444151e77d Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Fri, 27 Dec 2019 15:55:48 +0100
Subject: ethtool: provide link settings with LINKINFO_GET request

Implement LINKINFO_GET netlink request to get basic link settings provided
by ETHTOOL_GLINKSETTINGS and ETHTOOL_GSET ioctl commands.

This request provides settings not directly related to autonegotiation and
link mode selection: physical port, phy MDIO address, MDI(-X) status,
MDI(-X) control and transceiver.

LINKINFO_GET request can be used with NLM_F_DUMP (without device
identification) to request the information for all devices in current
network namespace providing the data.

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/ethtool-netlink.rst | 37 ++++++++++-
 include/uapi/linux/ethtool_netlink.h         | 18 ++++++
 net/ethtool/Makefile                         |  2 +-
 net/ethtool/common.c                         | 48 ++++++++++++++
 net/ethtool/common.h                         |  4 ++
 net/ethtool/ioctl.c                          | 48 --------------
 net/ethtool/linkinfo.c                       | 94 ++++++++++++++++++++++++++++
 net/ethtool/netlink.c                        |  8 +++
 net/ethtool/netlink.h                        |  1 +
 9 files changed, 209 insertions(+), 51 deletions(-)
 create mode 100644 net/ethtool/linkinfo.c

diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
index 3912cb0eb9c6..60684786a92c 100644
--- a/Documentation/networking/ethtool-netlink.rst
+++ b/Documentation/networking/ethtool-netlink.rst
@@ -180,12 +180,14 @@ Userspace to kernel:
 
   ===================================== ================================
   ``ETHTOOL_MSG_STRSET_GET``            get string set
+  ``ETHTOOL_MSG_LINKINFO_GET``          get link settings
   ===================================== ================================
 
 Kernel to userspace:
 
   ===================================== ================================
   ``ETHTOOL_MSG_STRSET_GET_REPLY``      string set contents
+  ``ETHTOOL_MSG_LINKINFO_GET_REPLY``    link settings
   ===================================== ================================
 
 ``GET`` requests are sent by userspace applications to retrieve device
@@ -278,6 +280,37 @@ Flag ``ETHTOOL_A_STRSET_COUNTS_ONLY`` tells kernel to only return string
 counts of the sets, not the actual strings.
 
 
+LINKINFO_GET
+============
+
+Requests link settings as provided by ``ETHTOOL_GLINKSETTINGS`` except for
+link modes and autonegotiation related information. The request does not use
+any attributes.
+
+Request contents:
+
+  ====================================  ======  ==========================
+  ``ETHTOOL_A_LINKINFO_HEADER``         nested  request header
+  ====================================  ======  ==========================
+
+Kernel response contents:
+
+  ====================================  ======  ==========================
+  ``ETHTOOL_A_LINKINFO_HEADER``         nested  reply header
+  ``ETHTOOL_A_LINKINFO_PORT``           u8      physical port
+  ``ETHTOOL_A_LINKINFO_PHYADDR``        u8      phy MDIO address
+  ``ETHTOOL_A_LINKINFO_TP_MDIX``        u8      MDI(-X) status
+  ``ETHTOOL_A_LINKINFO_TP_MDIX_CTRL``   u8      MDI(-X) control
+  ``ETHTOOL_A_LINKINFO_TRANSCEIVER``    u8      transceiver
+  ====================================  ======  ==========================
+
+Attributes and their values have the same meaning as matching members of the
+corresponding ioctl structures.
+
+``LINKINFO_GET`` allows dump requests (kernel returns reply message for all
+devices supporting the request).
+
+
 Request translation
 ===================
 
@@ -288,7 +321,7 @@ have their netlink replacement yet.
   =================================== =====================================
   ioctl command                       netlink command
   =================================== =====================================
-  ``ETHTOOL_GSET``                    n/a
+  ``ETHTOOL_GSET``                    ``ETHTOOL_MSG_LINKINFO_GET``
   ``ETHTOOL_SSET``                    n/a
   ``ETHTOOL_GDRVINFO``                n/a
   ``ETHTOOL_GREGS``                   n/a
@@ -362,7 +395,7 @@ have their netlink replacement yet.
   ``ETHTOOL_STUNABLE``                n/a
   ``ETHTOOL_GPHYSTATS``               n/a
   ``ETHTOOL_PERQUEUE``                n/a
-  ``ETHTOOL_GLINKSETTINGS``           n/a
+  ``ETHTOOL_GLINKSETTINGS``           ``ETHTOOL_MSG_LINKINFO_GET``
   ``ETHTOOL_SLINKSETTINGS``           n/a
   ``ETHTOOL_PHY_GTUNABLE``            n/a
   ``ETHTOOL_PHY_STUNABLE``            n/a
diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
index cabef1fec42a..1966532993e5 100644
--- a/include/uapi/linux/ethtool_netlink.h
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -15,6 +15,7 @@
 enum {
 	ETHTOOL_MSG_USER_NONE,
 	ETHTOOL_MSG_STRSET_GET,
+	ETHTOOL_MSG_LINKINFO_GET,
 
 	/* add new constants above here */
 	__ETHTOOL_MSG_USER_CNT,
@@ -25,6 +26,7 @@ enum {
 enum {
 	ETHTOOL_MSG_KERNEL_NONE,
 	ETHTOOL_MSG_STRSET_GET_REPLY,
+	ETHTOOL_MSG_LINKINFO_GET_REPLY,
 
 	/* add new constants above here */
 	__ETHTOOL_MSG_KERNEL_CNT,
@@ -141,6 +143,22 @@ enum {
 	ETHTOOL_A_STRSET_MAX = __ETHTOOL_A_STRSET_CNT - 1
 };
 
+/* LINKINFO */
+
+enum {
+	ETHTOOL_A_LINKINFO_UNSPEC,
+	ETHTOOL_A_LINKINFO_HEADER,		/* nest - _A_HEADER_* */
+	ETHTOOL_A_LINKINFO_PORT,		/* u8 */
+	ETHTOOL_A_LINKINFO_PHYADDR,		/* u8 */
+	ETHTOOL_A_LINKINFO_TP_MDIX,		/* u8 */
+	ETHTOOL_A_LINKINFO_TP_MDIX_CTRL,	/* u8 */
+	ETHTOOL_A_LINKINFO_TRANSCEIVER,		/* u8 */
+
+	/* add new constants above here */
+	__ETHTOOL_A_LINKINFO_CNT,
+	ETHTOOL_A_LINKINFO_MAX = __ETHTOOL_A_LINKINFO_CNT - 1
+};
+
 /* generic netlink info */
 #define ETHTOOL_GENL_NAME "ethtool"
 #define ETHTOOL_GENL_VERSION 1
diff --git a/net/ethtool/Makefile b/net/ethtool/Makefile
index efcc42c34d62..765736ec52c0 100644
--- a/net/ethtool/Makefile
+++ b/net/ethtool/Makefile
@@ -4,4 +4,4 @@ obj-y				+= ioctl.o common.o
 
 obj-$(CONFIG_ETHTOOL_NETLINK)	+= ethtool_nl.o
 
-ethtool_nl-y	:= netlink.o bitset.o strset.o
+ethtool_nl-y	:= netlink.o bitset.o strset.o linkinfo.o
diff --git a/net/ethtool/common.c b/net/ethtool/common.c
index 0a8728565356..1d4a0aeff2cb 100644
--- a/net/ethtool/common.c
+++ b/net/ethtool/common.c
@@ -169,3 +169,51 @@ const char link_mode_names[][ETH_GSTRING_LEN] = {
 	__DEFINE_LINK_MODE_NAME(400000, CR8, Full),
 };
 static_assert(ARRAY_SIZE(link_mode_names) == __ETHTOOL_LINK_MODE_MASK_NBITS);
+
+/* return false if legacy contained non-0 deprecated fields
+ * maxtxpkt/maxrxpkt. rest of ksettings always updated
+ */
+bool
+convert_legacy_settings_to_link_ksettings(
+	struct ethtool_link_ksettings *link_ksettings,
+	const struct ethtool_cmd *legacy_settings)
+{
+	bool retval = true;
+
+	memset(link_ksettings, 0, sizeof(*link_ksettings));
+
+	/* This is used to tell users that driver is still using these
+	 * deprecated legacy fields, and they should not use
+	 * %ETHTOOL_GLINKSETTINGS/%ETHTOOL_SLINKSETTINGS
+	 */
+	if (legacy_settings->maxtxpkt ||
+	    legacy_settings->maxrxpkt)
+		retval = false;
+
+	ethtool_convert_legacy_u32_to_link_mode(
+		link_ksettings->link_modes.supported,
+		legacy_settings->supported);
+	ethtool_convert_legacy_u32_to_link_mode(
+		link_ksettings->link_modes.advertising,
+		legacy_settings->advertising);
+	ethtool_convert_legacy_u32_to_link_mode(
+		link_ksettings->link_modes.lp_advertising,
+		legacy_settings->lp_advertising);
+	link_ksettings->base.speed
+		= ethtool_cmd_speed(legacy_settings);
+	link_ksettings->base.duplex
+		= legacy_settings->duplex;
+	link_ksettings->base.port
+		= legacy_settings->port;
+	link_ksettings->base.phy_address
+		= legacy_settings->phy_address;
+	link_ksettings->base.autoneg
+		= legacy_settings->autoneg;
+	link_ksettings->base.mdio_support
+		= legacy_settings->mdio_support;
+	link_ksettings->base.eth_tp_mdix
+		= legacy_settings->eth_tp_mdix;
+	link_ksettings->base.eth_tp_mdix_ctrl
+		= legacy_settings->eth_tp_mdix_ctrl;
+	return retval;
+}
diff --git a/net/ethtool/common.h b/net/ethtool/common.h
index bbb788908cb1..c8a237402729 100644
--- a/net/ethtool/common.h
+++ b/net/ethtool/common.h
@@ -19,4 +19,8 @@ extern const char
 phy_tunable_strings[__ETHTOOL_PHY_TUNABLE_COUNT][ETH_GSTRING_LEN];
 extern const char link_mode_names[][ETH_GSTRING_LEN];
 
+bool convert_legacy_settings_to_link_ksettings(
+	struct ethtool_link_ksettings *link_ksettings,
+	const struct ethtool_cmd *legacy_settings);
+
 #endif /* _ETHTOOL_COMMON_H */
diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
index 88f7cddf5a6f..8a0a13b478e0 100644
--- a/net/ethtool/ioctl.c
+++ b/net/ethtool/ioctl.c
@@ -358,54 +358,6 @@ bool ethtool_convert_link_mode_to_legacy_u32(u32 *legacy_u32,
 }
 EXPORT_SYMBOL(ethtool_convert_link_mode_to_legacy_u32);
 
-/* return false if legacy contained non-0 deprecated fields
- * maxtxpkt/maxrxpkt. rest of ksettings always updated
- */
-static bool
-convert_legacy_settings_to_link_ksettings(
-	struct ethtool_link_ksettings *link_ksettings,
-	const struct ethtool_cmd *legacy_settings)
-{
-	bool retval = true;
-
-	memset(link_ksettings, 0, sizeof(*link_ksettings));
-
-	/* This is used to tell users that driver is still using these
-	 * deprecated legacy fields, and they should not use
-	 * %ETHTOOL_GLINKSETTINGS/%ETHTOOL_SLINKSETTINGS
-	 */
-	if (legacy_settings->maxtxpkt ||
-	    legacy_settings->maxrxpkt)
-		retval = false;
-
-	ethtool_convert_legacy_u32_to_link_mode(
-		link_ksettings->link_modes.supported,
-		legacy_settings->supported);
-	ethtool_convert_legacy_u32_to_link_mode(
-		link_ksettings->link_modes.advertising,
-		legacy_settings->advertising);
-	ethtool_convert_legacy_u32_to_link_mode(
-		link_ksettings->link_modes.lp_advertising,
-		legacy_settings->lp_advertising);
-	link_ksettings->base.speed
-		= ethtool_cmd_speed(legacy_settings);
-	link_ksettings->base.duplex
-		= legacy_settings->duplex;
-	link_ksettings->base.port
-		= legacy_settings->port;
-	link_ksettings->base.phy_address
-		= legacy_settings->phy_address;
-	link_ksettings->base.autoneg
-		= legacy_settings->autoneg;
-	link_ksettings->base.mdio_support
-		= legacy_settings->mdio_support;
-	link_ksettings->base.eth_tp_mdix
-		= legacy_settings->eth_tp_mdix;
-	link_ksettings->base.eth_tp_mdix_ctrl
-		= legacy_settings->eth_tp_mdix_ctrl;
-	return retval;
-}
-
 /* return false if ksettings link modes had higher bits
  * set. legacy_settings always updated (best effort)
  */
diff --git a/net/ethtool/linkinfo.c b/net/ethtool/linkinfo.c
new file mode 100644
index 000000000000..f52a31af6fff
--- /dev/null
+++ b/net/ethtool/linkinfo.c
@@ -0,0 +1,94 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include "netlink.h"
+#include "common.h"
+
+struct linkinfo_req_info {
+	struct ethnl_req_info		base;
+};
+
+struct linkinfo_reply_data {
+	struct ethnl_reply_data		base;
+	struct ethtool_link_ksettings	ksettings;
+	struct ethtool_link_settings	*lsettings;
+};
+
+#define LINKINFO_REPDATA(__reply_base) \
+	container_of(__reply_base, struct linkinfo_reply_data, base)
+
+static const struct nla_policy
+linkinfo_get_policy[ETHTOOL_A_LINKINFO_MAX + 1] = {
+	[ETHTOOL_A_LINKINFO_UNSPEC]		= { .type = NLA_REJECT },
+	[ETHTOOL_A_LINKINFO_HEADER]		= { .type = NLA_NESTED },
+	[ETHTOOL_A_LINKINFO_PORT]		= { .type = NLA_REJECT },
+	[ETHTOOL_A_LINKINFO_PHYADDR]		= { .type = NLA_REJECT },
+	[ETHTOOL_A_LINKINFO_TP_MDIX]		= { .type = NLA_REJECT },
+	[ETHTOOL_A_LINKINFO_TP_MDIX_CTRL]	= { .type = NLA_REJECT },
+	[ETHTOOL_A_LINKINFO_TRANSCEIVER]	= { .type = NLA_REJECT },
+};
+
+static int linkinfo_prepare_data(const struct ethnl_req_info *req_base,
+				 struct ethnl_reply_data *reply_base,
+				 struct genl_info *info)
+{
+	struct linkinfo_reply_data *data = LINKINFO_REPDATA(reply_base);
+	struct net_device *dev = reply_base->dev;
+	int ret;
+
+	data->lsettings = &data->ksettings.base;
+
+	ret = ethnl_ops_begin(dev);
+	if (ret < 0)
+		return ret;
+	ret = __ethtool_get_link_ksettings(dev, &data->ksettings);
+	if (ret < 0 && info)
+		GENL_SET_ERR_MSG(info, "failed to retrieve link settings");
+	ethnl_ops_complete(dev);
+
+	return ret;
+}
+
+static int linkinfo_reply_size(const struct ethnl_req_info *req_base,
+			       const struct ethnl_reply_data *reply_base)
+{
+	return nla_total_size(sizeof(u8)) /* LINKINFO_PORT */
+		+ nla_total_size(sizeof(u8)) /* LINKINFO_PHYADDR */
+		+ nla_total_size(sizeof(u8)) /* LINKINFO_TP_MDIX */
+		+ nla_total_size(sizeof(u8)) /* LINKINFO_TP_MDIX_CTRL */
+		+ nla_total_size(sizeof(u8)) /* LINKINFO_TRANSCEIVER */
+		+ 0;
+}
+
+static int linkinfo_fill_reply(struct sk_buff *skb,
+			       const struct ethnl_req_info *req_base,
+			       const struct ethnl_reply_data *reply_base)
+{
+	const struct linkinfo_reply_data *data = LINKINFO_REPDATA(reply_base);
+
+	if (nla_put_u8(skb, ETHTOOL_A_LINKINFO_PORT, data->lsettings->port) ||
+	    nla_put_u8(skb, ETHTOOL_A_LINKINFO_PHYADDR,
+		       data->lsettings->phy_address) ||
+	    nla_put_u8(skb, ETHTOOL_A_LINKINFO_TP_MDIX,
+		       data->lsettings->eth_tp_mdix) ||
+	    nla_put_u8(skb, ETHTOOL_A_LINKINFO_TP_MDIX_CTRL,
+		       data->lsettings->eth_tp_mdix_ctrl) ||
+	    nla_put_u8(skb, ETHTOOL_A_LINKINFO_TRANSCEIVER,
+		       data->lsettings->transceiver))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+const struct ethnl_request_ops ethnl_linkinfo_request_ops = {
+	.request_cmd		= ETHTOOL_MSG_LINKINFO_GET,
+	.reply_cmd		= ETHTOOL_MSG_LINKINFO_GET_REPLY,
+	.hdr_attr		= ETHTOOL_A_LINKINFO_HEADER,
+	.max_attr		= ETHTOOL_A_LINKINFO_MAX,
+	.req_info_size		= sizeof(struct linkinfo_req_info),
+	.reply_data_size	= sizeof(struct linkinfo_reply_data),
+	.request_policy		= linkinfo_get_policy,
+
+	.prepare_data		= linkinfo_prepare_data,
+	.reply_size		= linkinfo_reply_size,
+	.fill_reply		= linkinfo_fill_reply,
+};
diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index 7dc082bde670..ce86d97c922e 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -195,6 +195,7 @@ struct ethnl_dump_ctx {
 static const struct ethnl_request_ops *
 ethnl_default_requests[__ETHTOOL_MSG_USER_CNT] = {
 	[ETHTOOL_MSG_STRSET_GET]	= &ethnl_strset_request_ops,
+	[ETHTOOL_MSG_LINKINFO_GET]	= &ethnl_linkinfo_request_ops,
 };
 
 static struct ethnl_dump_ctx *ethnl_dump_context(struct netlink_callback *cb)
@@ -526,6 +527,13 @@ static const struct genl_ops ethtool_genl_ops[] = {
 		.dumpit	= ethnl_default_dumpit,
 		.done	= ethnl_default_done,
 	},
+	{
+		.cmd	= ETHTOOL_MSG_LINKINFO_GET,
+		.doit	= ethnl_default_doit,
+		.start	= ethnl_default_start,
+		.dumpit	= ethnl_default_dumpit,
+		.done	= ethnl_default_done,
+	},
 };
 
 static const struct genl_multicast_group ethtool_nl_mcgrps[] = {
diff --git a/net/ethtool/netlink.h b/net/ethtool/netlink.h
index 44e9f63aefb7..9fc8f94d8dce 100644
--- a/net/ethtool/netlink.h
+++ b/net/ethtool/netlink.h
@@ -329,5 +329,6 @@ struct ethnl_request_ops {
 /* request handlers */
 
 extern const struct ethnl_request_ops ethnl_strset_request_ops;
+extern const struct ethnl_request_ops ethnl_linkinfo_request_ops;
 
 #endif /* _NET_ETHTOOL_NETLINK_H */
-- 
cgit v1.2.3-59-g8ed1b


From a53f3d41e4d3df46aba254ba51e7fbe45470fece Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Fri, 27 Dec 2019 15:55:53 +0100
Subject: ethtool: set link settings with LINKINFO_SET request

Implement LINKINFO_SET netlink request to set link settings queried by
LINKINFO_GET message.

Only physical port, phy MDIO address and MDI(-X) control can be set,
attempt to modify MDI(-X) status and transceiver is rejected.

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/ethtool-netlink.rst | 24 +++++++++-
 include/uapi/linux/ethtool_netlink.h         |  1 +
 net/ethtool/linkinfo.c                       | 71 ++++++++++++++++++++++++++++
 net/ethtool/netlink.c                        |  5 ++
 net/ethtool/netlink.h                        |  2 +
 5 files changed, 101 insertions(+), 2 deletions(-)

diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
index 60684786a92c..39c4aba324c3 100644
--- a/Documentation/networking/ethtool-netlink.rst
+++ b/Documentation/networking/ethtool-netlink.rst
@@ -181,6 +181,7 @@ Userspace to kernel:
   ===================================== ================================
   ``ETHTOOL_MSG_STRSET_GET``            get string set
   ``ETHTOOL_MSG_LINKINFO_GET``          get link settings
+  ``ETHTOOL_MSG_LINKINFO_SET``          set link settings
   ===================================== ================================
 
 Kernel to userspace:
@@ -311,6 +312,25 @@ corresponding ioctl structures.
 devices supporting the request).
 
 
+LINKINFO_SET
+============
+
+``LINKINFO_SET`` request allows setting some of the attributes reported by
+``LINKINFO_GET``.
+
+Request contents:
+
+  ====================================  ======  ==========================
+  ``ETHTOOL_A_LINKINFO_HEADER``         nested  request header
+  ``ETHTOOL_A_LINKINFO_PORT``           u8      physical port
+  ``ETHTOOL_A_LINKINFO_PHYADDR``        u8      phy MDIO address
+  ``ETHTOOL_A_LINKINFO_TP_MDIX_CTRL``   u8      MDI(-X) control
+  ====================================  ======  ==========================
+
+MDI(-X) status and transceiver cannot be set, request with the corresponding
+attributes is rejected.
+
+
 Request translation
 ===================
 
@@ -322,7 +342,7 @@ have their netlink replacement yet.
   ioctl command                       netlink command
   =================================== =====================================
   ``ETHTOOL_GSET``                    ``ETHTOOL_MSG_LINKINFO_GET``
-  ``ETHTOOL_SSET``                    n/a
+  ``ETHTOOL_SSET``                    ``ETHTOOL_MSG_LINKINFO_SET``
   ``ETHTOOL_GDRVINFO``                n/a
   ``ETHTOOL_GREGS``                   n/a
   ``ETHTOOL_GWOL``                    n/a
@@ -396,7 +416,7 @@ have their netlink replacement yet.
   ``ETHTOOL_GPHYSTATS``               n/a
   ``ETHTOOL_PERQUEUE``                n/a
   ``ETHTOOL_GLINKSETTINGS``           ``ETHTOOL_MSG_LINKINFO_GET``
-  ``ETHTOOL_SLINKSETTINGS``           n/a
+  ``ETHTOOL_SLINKSETTINGS``           ``ETHTOOL_MSG_LINKINFO_SET``
   ``ETHTOOL_PHY_GTUNABLE``            n/a
   ``ETHTOOL_PHY_STUNABLE``            n/a
   ``ETHTOOL_GFECPARAM``               n/a
diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
index 1966532993e5..5b7806a5bef8 100644
--- a/include/uapi/linux/ethtool_netlink.h
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -16,6 +16,7 @@ enum {
 	ETHTOOL_MSG_USER_NONE,
 	ETHTOOL_MSG_STRSET_GET,
 	ETHTOOL_MSG_LINKINFO_GET,
+	ETHTOOL_MSG_LINKINFO_SET,
 
 	/* add new constants above here */
 	__ETHTOOL_MSG_USER_CNT,
diff --git a/net/ethtool/linkinfo.c b/net/ethtool/linkinfo.c
index f52a31af6fff..8a5f68f92425 100644
--- a/net/ethtool/linkinfo.c
+++ b/net/ethtool/linkinfo.c
@@ -92,3 +92,74 @@ const struct ethnl_request_ops ethnl_linkinfo_request_ops = {
 	.reply_size		= linkinfo_reply_size,
 	.fill_reply		= linkinfo_fill_reply,
 };
+
+/* LINKINFO_SET */
+
+static const struct nla_policy
+linkinfo_set_policy[ETHTOOL_A_LINKINFO_MAX + 1] = {
+	[ETHTOOL_A_LINKINFO_UNSPEC]		= { .type = NLA_REJECT },
+	[ETHTOOL_A_LINKINFO_HEADER]		= { .type = NLA_NESTED },
+	[ETHTOOL_A_LINKINFO_PORT]		= { .type = NLA_U8 },
+	[ETHTOOL_A_LINKINFO_PHYADDR]		= { .type = NLA_U8 },
+	[ETHTOOL_A_LINKINFO_TP_MDIX]		= { .type = NLA_REJECT },
+	[ETHTOOL_A_LINKINFO_TP_MDIX_CTRL]	= { .type = NLA_U8 },
+	[ETHTOOL_A_LINKINFO_TRANSCEIVER]	= { .type = NLA_REJECT },
+};
+
+int ethnl_set_linkinfo(struct sk_buff *skb, struct genl_info *info)
+{
+	struct nlattr *tb[ETHTOOL_A_LINKINFO_MAX + 1];
+	struct ethtool_link_ksettings ksettings = {};
+	struct ethtool_link_settings *lsettings;
+	struct ethnl_req_info req_info = {};
+	struct net_device *dev;
+	bool mod = false;
+	int ret;
+
+	ret = nlmsg_parse(info->nlhdr, GENL_HDRLEN, tb,
+			  ETHTOOL_A_LINKINFO_MAX, linkinfo_set_policy,
+			  info->extack);
+	if (ret < 0)
+		return ret;
+	ret = ethnl_parse_header(&req_info, tb[ETHTOOL_A_LINKINFO_HEADER],
+				 genl_info_net(info), info->extack, true);
+	if (ret < 0)
+		return ret;
+	dev = req_info.dev;
+	if (!dev->ethtool_ops->get_link_ksettings ||
+	    !dev->ethtool_ops->set_link_ksettings)
+		return -EOPNOTSUPP;
+
+	rtnl_lock();
+	ret = ethnl_ops_begin(dev);
+	if (ret < 0)
+		goto out_rtnl;
+
+	ret = __ethtool_get_link_ksettings(dev, &ksettings);
+	if (ret < 0) {
+		if (info)
+			GENL_SET_ERR_MSG(info, "failed to retrieve link settings");
+		goto out_ops;
+	}
+	lsettings = &ksettings.base;
+
+	ethnl_update_u8(&lsettings->port, tb[ETHTOOL_A_LINKINFO_PORT], &mod);
+	ethnl_update_u8(&lsettings->phy_address, tb[ETHTOOL_A_LINKINFO_PHYADDR],
+			&mod);
+	ethnl_update_u8(&lsettings->eth_tp_mdix_ctrl,
+			tb[ETHTOOL_A_LINKINFO_TP_MDIX_CTRL], &mod);
+	ret = 0;
+	if (!mod)
+		goto out_ops;
+
+	ret = dev->ethtool_ops->set_link_ksettings(dev, &ksettings);
+	if (ret < 0)
+		GENL_SET_ERR_MSG(info, "link settings update failed");
+
+out_ops:
+	ethnl_ops_complete(dev);
+out_rtnl:
+	rtnl_unlock();
+	dev_put(dev);
+	return ret;
+}
diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index ce86d97c922e..7867425956f6 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -534,6 +534,11 @@ static const struct genl_ops ethtool_genl_ops[] = {
 		.dumpit	= ethnl_default_dumpit,
 		.done	= ethnl_default_done,
 	},
+	{
+		.cmd	= ETHTOOL_MSG_LINKINFO_SET,
+		.flags	= GENL_UNS_ADMIN_PERM,
+		.doit	= ethnl_set_linkinfo,
+	},
 };
 
 static const struct genl_multicast_group ethtool_nl_mcgrps[] = {
diff --git a/net/ethtool/netlink.h b/net/ethtool/netlink.h
index 9fc8f94d8dce..bbe5fe60a023 100644
--- a/net/ethtool/netlink.h
+++ b/net/ethtool/netlink.h
@@ -331,4 +331,6 @@ struct ethnl_request_ops {
 extern const struct ethnl_request_ops ethnl_strset_request_ops;
 extern const struct ethnl_request_ops ethnl_linkinfo_request_ops;
 
+int ethnl_set_linkinfo(struct sk_buff *skb, struct genl_info *info);
+
 #endif /* _NET_ETHTOOL_NETLINK_H */
-- 
cgit v1.2.3-59-g8ed1b


From 5cf2a548bcbd3da3e79adb0b64ef96dbb9988d78 Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Fri, 27 Dec 2019 15:55:58 +0100
Subject: ethtool: add default notification handler

The ethtool netlink notifications have the same format as related GET
replies so that if generic GET handling framework is used to process GET
requests, its callbacks and instance of struct get_request_ops can be
also used to compose corresponding notification message.

Provide function ethnl_std_notify() to be used as notification handler in
ethnl_notify_handlers table.

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ethtool/netlink.c | 89 +++++++++++++++++++++++++++++++++++++++++++++++++++
 net/ethtool/netlink.h |  4 ++-
 2 files changed, 92 insertions(+), 1 deletion(-)

diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index 7867425956f6..057b67f8ba8c 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -7,6 +7,7 @@
 static struct genl_family ethtool_genl_family;
 
 static bool ethnl_ok __read_mostly;
+static u32 ethnl_bcast_seq;
 
 static const struct nla_policy ethnl_header_policy[ETHTOOL_A_HEADER_MAX + 1] = {
 	[ETHTOOL_A_HEADER_UNSPEC]	= { .type = NLA_REJECT },
@@ -171,6 +172,18 @@ err:
 	return NULL;
 }
 
+static void *ethnl_bcastmsg_put(struct sk_buff *skb, u8 cmd)
+{
+	return genlmsg_put(skb, 0, ++ethnl_bcast_seq, &ethtool_genl_family, 0,
+			   cmd);
+}
+
+static int ethnl_multicast(struct sk_buff *skb, struct net_device *dev)
+{
+	return genlmsg_multicast_netns(&ethtool_genl_family, dev_net(dev), skb,
+				       0, ETHNL_MCGRP_MONITOR, GFP_KERNEL);
+}
+
 /* GET request helpers */
 
 /**
@@ -494,6 +507,82 @@ static int ethnl_default_done(struct netlink_callback *cb)
 	return 0;
 }
 
+static const struct ethnl_request_ops *
+ethnl_default_notify_ops[ETHTOOL_MSG_KERNEL_MAX + 1] = {
+};
+
+/* default notification handler */
+static void ethnl_default_notify(struct net_device *dev, unsigned int cmd,
+				 const void *data)
+{
+	struct ethnl_reply_data *reply_data;
+	const struct ethnl_request_ops *ops;
+	struct ethnl_req_info *req_info;
+	struct sk_buff *skb;
+	void *reply_payload;
+	int reply_len;
+	int ret;
+
+	if (WARN_ONCE(cmd > ETHTOOL_MSG_KERNEL_MAX ||
+		      !ethnl_default_notify_ops[cmd],
+		      "unexpected notification type %u\n", cmd))
+		return;
+	ops = ethnl_default_notify_ops[cmd];
+	req_info = kzalloc(ops->req_info_size, GFP_KERNEL);
+	if (!req_info)
+		return;
+	reply_data = kmalloc(ops->reply_data_size, GFP_KERNEL);
+	if (!reply_data) {
+		kfree(req_info);
+		return;
+	}
+
+	req_info->dev = dev;
+	req_info->flags |= ETHTOOL_FLAG_COMPACT_BITSETS;
+
+	ethnl_init_reply_data(reply_data, ops, dev);
+	ret = ops->prepare_data(req_info, reply_data, NULL);
+	if (ret < 0)
+		goto err_cleanup;
+	reply_len = ops->reply_size(req_info, reply_data);
+	if (ret < 0)
+		goto err_cleanup;
+	ret = -ENOMEM;
+	skb = genlmsg_new(reply_len, GFP_KERNEL);
+	if (!skb)
+		goto err_cleanup;
+	reply_payload = ethnl_bcastmsg_put(skb, cmd);
+	if (!reply_payload)
+		goto err_skb;
+	ret = ethnl_fill_reply_header(skb, dev, ops->hdr_attr);
+	if (ret < 0)
+		goto err_msg;
+	ret = ops->fill_reply(skb, req_info, reply_data);
+	if (ret < 0)
+		goto err_msg;
+	if (ops->cleanup_data)
+		ops->cleanup_data(reply_data);
+
+	genlmsg_end(skb, reply_payload);
+	kfree(reply_data);
+	kfree(req_info);
+	ethnl_multicast(skb, dev);
+	return;
+
+err_msg:
+	WARN_ONCE(ret == -EMSGSIZE,
+		  "calculated message payload length (%d) not sufficient\n",
+		  reply_len);
+err_skb:
+	nlmsg_free(skb);
+err_cleanup:
+	if (ops->cleanup_data)
+		ops->cleanup_data(reply_data);
+	kfree(reply_data);
+	kfree(req_info);
+	return;
+}
+
 /* notifications */
 
 typedef void (*ethnl_notify_handler_t)(struct net_device *dev, unsigned int cmd,
diff --git a/net/ethtool/netlink.h b/net/ethtool/netlink.h
index bbe5fe60a023..5d56d7779a06 100644
--- a/net/ethtool/netlink.h
+++ b/net/ethtool/netlink.h
@@ -300,7 +300,9 @@ static inline void ethnl_ops_complete(struct net_device *dev)
  * unified infrastructure. When used, a pointer to an instance of this
  * structure is to be added to &ethnl_default_requests array and generic
  * handlers ethnl_default_doit(), ethnl_default_dumpit(),
- * ethnl_default_start() and ethnl_default_done() used in @ethtool_genl_ops.
+ * ethnl_default_start() and ethnl_default_done() used in @ethtool_genl_ops;
+ * ethnl_default_notify() can be used in @ethnl_notify_handlers to send
+ * notifications of the corresponding type.
  */
 struct ethnl_request_ops {
 	u8			request_cmd;
-- 
cgit v1.2.3-59-g8ed1b


From 73286734c1b0d009fd317c2a74e173cb22233dad Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Fri, 27 Dec 2019 15:56:03 +0100
Subject: ethtool: add LINKINFO_NTF notification

Send ETHTOOL_MSG_LINKINFO_NTF notification message whenever device link
settings are modified using ETHTOOL_MSG_LINKINFO_SET netlink message or
ETHTOOL_SLINKSETTINGS or ETHTOOL_SSET ioctl commands.

The notification message has the same format as reply to LINKINFO_GET
request. ETHTOOL_MSG_LINKINFO_SET netlink request only triggers the
notification if there is a change but the ioctl command handlers do not
check if there is an actual change and trigger the notification whenever
the commands are executed.

As all work is done by ethnl_default_notify() handler and callback
functions introduced to handle LINKINFO_GET requests, all that remains is
adding entries for ETHTOOL_MSG_LINKINFO_NTF into ethnl_notify_handlers and
ethnl_default_notify_ops lookup tables and calls to ethtool_notify() where
needed.

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/ethtool-netlink.rst |  1 +
 include/uapi/linux/ethtool_netlink.h         |  1 +
 net/ethtool/ioctl.c                          | 12 ++++++++++--
 net/ethtool/linkinfo.c                       |  2 ++
 net/ethtool/netlink.c                        |  2 ++
 5 files changed, 16 insertions(+), 2 deletions(-)

diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
index 39c4aba324c3..34254482d295 100644
--- a/Documentation/networking/ethtool-netlink.rst
+++ b/Documentation/networking/ethtool-netlink.rst
@@ -189,6 +189,7 @@ Kernel to userspace:
   ===================================== ================================
   ``ETHTOOL_MSG_STRSET_GET_REPLY``      string set contents
   ``ETHTOOL_MSG_LINKINFO_GET_REPLY``    link settings
+  ``ETHTOOL_MSG_LINKINFO_NTF``          link settings notification
   ===================================== ================================
 
 ``GET`` requests are sent by userspace applications to retrieve device
diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
index 5b7806a5bef8..d530fa30de36 100644
--- a/include/uapi/linux/ethtool_netlink.h
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -28,6 +28,7 @@ enum {
 	ETHTOOL_MSG_KERNEL_NONE,
 	ETHTOOL_MSG_STRSET_GET_REPLY,
 	ETHTOOL_MSG_LINKINFO_GET_REPLY,
+	ETHTOOL_MSG_LINKINFO_NTF,
 
 	/* add new constants above here */
 	__ETHTOOL_MSG_KERNEL_CNT,
diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
index 8a0a13b478e0..11a467294a33 100644
--- a/net/ethtool/ioctl.c
+++ b/net/ethtool/ioctl.c
@@ -26,6 +26,7 @@
 #include <net/devlink.h>
 #include <net/xdp_sock.h>
 #include <net/flow_offload.h>
+#include <linux/ethtool_netlink.h>
 
 #include "common.h"
 
@@ -571,7 +572,10 @@ static int ethtool_set_link_ksettings(struct net_device *dev,
 	    != link_ksettings.base.link_mode_masks_nwords)
 		return -EINVAL;
 
-	return dev->ethtool_ops->set_link_ksettings(dev, &link_ksettings);
+	err = dev->ethtool_ops->set_link_ksettings(dev, &link_ksettings);
+	if (err >= 0)
+		ethtool_notify(dev, ETHTOOL_MSG_LINKINFO_NTF, NULL);
+	return err;
 }
 
 /* Query device for its ethtool_cmd settings.
@@ -620,6 +624,7 @@ static int ethtool_set_settings(struct net_device *dev, void __user *useraddr)
 {
 	struct ethtool_link_ksettings link_ksettings;
 	struct ethtool_cmd cmd;
+	int ret;
 
 	ASSERT_RTNL();
 
@@ -632,7 +637,10 @@ static int ethtool_set_settings(struct net_device *dev, void __user *useraddr)
 		return -EINVAL;
 	link_ksettings.base.link_mode_masks_nwords =
 		__ETHTOOL_LINK_MODE_MASK_NU32;
-	return dev->ethtool_ops->set_link_ksettings(dev, &link_ksettings);
+	ret = dev->ethtool_ops->set_link_ksettings(dev, &link_ksettings);
+	if (ret >= 0)
+		ethtool_notify(dev, ETHTOOL_MSG_LINKINFO_NTF, NULL);
+	return ret;
 }
 
 static noinline_for_stack int ethtool_get_drvinfo(struct net_device *dev,
diff --git a/net/ethtool/linkinfo.c b/net/ethtool/linkinfo.c
index 8a5f68f92425..5d16cb4e8693 100644
--- a/net/ethtool/linkinfo.c
+++ b/net/ethtool/linkinfo.c
@@ -155,6 +155,8 @@ int ethnl_set_linkinfo(struct sk_buff *skb, struct genl_info *info)
 	ret = dev->ethtool_ops->set_link_ksettings(dev, &ksettings);
 	if (ret < 0)
 		GENL_SET_ERR_MSG(info, "link settings update failed");
+	else
+		ethtool_notify(dev, ETHTOOL_MSG_LINKINFO_NTF, NULL);
 
 out_ops:
 	ethnl_ops_complete(dev);
diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index 057b67f8ba8c..942da4ebdfe9 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -509,6 +509,7 @@ static int ethnl_default_done(struct netlink_callback *cb)
 
 static const struct ethnl_request_ops *
 ethnl_default_notify_ops[ETHTOOL_MSG_KERNEL_MAX + 1] = {
+	[ETHTOOL_MSG_LINKINFO_NTF]	= &ethnl_linkinfo_request_ops,
 };
 
 /* default notification handler */
@@ -589,6 +590,7 @@ typedef void (*ethnl_notify_handler_t)(struct net_device *dev, unsigned int cmd,
 				       const void *data);
 
 static const ethnl_notify_handler_t ethnl_notify_handlers[] = {
+	[ETHTOOL_MSG_LINKINFO_NTF]	= ethnl_default_notify,
 };
 
 void ethtool_notify(struct net_device *dev, unsigned int cmd, const void *data)
-- 
cgit v1.2.3-59-g8ed1b


From f625aa9be8c10f2e4dc677837e240730a25feda7 Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Fri, 27 Dec 2019 15:56:08 +0100
Subject: ethtool: provide link mode information with LINKMODES_GET request

Implement LINKMODES_GET netlink request to get link modes related
information provided by ETHTOOL_GLINKSETTINGS and ETHTOOL_GSET ioctl
commands.

This request provides supported, advertised and peer advertised link modes,
autonegotiation flag, speed and duplex.

LINKMODES_GET request can be used with NLM_F_DUMP (without device
identification) to request the information for all devices in current
network namespace providing the data.

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/ethtool-netlink.rst |  36 +++++++
 include/linux/ethtool_netlink.h              |   3 +
 include/uapi/linux/ethtool_netlink.h         |  18 ++++
 net/ethtool/Makefile                         |   2 +-
 net/ethtool/linkmodes.c                      | 140 +++++++++++++++++++++++++++
 net/ethtool/netlink.c                        |   8 ++
 net/ethtool/netlink.h                        |   1 +
 7 files changed, 207 insertions(+), 1 deletion(-)
 create mode 100644 net/ethtool/linkmodes.c

diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
index 34254482d295..04c55be0264c 100644
--- a/Documentation/networking/ethtool-netlink.rst
+++ b/Documentation/networking/ethtool-netlink.rst
@@ -182,6 +182,7 @@ Userspace to kernel:
   ``ETHTOOL_MSG_STRSET_GET``            get string set
   ``ETHTOOL_MSG_LINKINFO_GET``          get link settings
   ``ETHTOOL_MSG_LINKINFO_SET``          set link settings
+  ``ETHTOOL_MSG_LINKMODES_GET``         get link modes info
   ===================================== ================================
 
 Kernel to userspace:
@@ -190,6 +191,7 @@ Kernel to userspace:
   ``ETHTOOL_MSG_STRSET_GET_REPLY``      string set contents
   ``ETHTOOL_MSG_LINKINFO_GET_REPLY``    link settings
   ``ETHTOOL_MSG_LINKINFO_NTF``          link settings notification
+  ``ETHTOOL_MSG_LINKMODES_GET_REPLY``   link modes info
   ===================================== ================================
 
 ``GET`` requests are sent by userspace applications to retrieve device
@@ -332,6 +334,38 @@ MDI(-X) status and transceiver cannot be set, request with the corresponding
 attributes is rejected.
 
 
+LINKMODES_GET
+=============
+
+Requests link modes (supported, advertised and peer advertised) and related
+information (autonegotiation status, link speed and duplex) as provided by
+``ETHTOOL_GLINKSETTINGS``. The request does not use any attributes.
+
+Request contents:
+
+  ====================================  ======  ==========================
+  ``ETHTOOL_A_LINKMODES_HEADER``        nested  request header
+  ====================================  ======  ==========================
+
+Kernel response contents:
+
+  ====================================  ======  ==========================
+  ``ETHTOOL_A_LINKMODES_HEADER``        nested  reply header
+  ``ETHTOOL_A_LINKMODES_AUTONEG``       u8      autonegotiation status
+  ``ETHTOOL_A_LINKMODES_OURS``          bitset  advertised link modes
+  ``ETHTOOL_A_LINKMODES_PEER``          bitset  partner link modes
+  ``ETHTOOL_A_LINKMODES_SPEED``         u32     link speed (Mb/s)
+  ``ETHTOOL_A_LINKMODES_DUPLEX``        u8      duplex mode
+  ====================================  ======  ==========================
+
+For ``ETHTOOL_A_LINKMODES_OURS``, value represents advertised modes and mask
+represents supported modes. ``ETHTOOL_A_LINKMODES_PEER`` in the reply is a bit
+list.
+
+``LINKMODES_GET`` allows dump requests (kernel returns reply messages for all
+devices supporting the request).
+
+
 Request translation
 ===================
 
@@ -343,6 +377,7 @@ have their netlink replacement yet.
   ioctl command                       netlink command
   =================================== =====================================
   ``ETHTOOL_GSET``                    ``ETHTOOL_MSG_LINKINFO_GET``
+                                      ``ETHTOOL_MSG_LINKMODES_GET``
   ``ETHTOOL_SSET``                    ``ETHTOOL_MSG_LINKINFO_SET``
   ``ETHTOOL_GDRVINFO``                n/a
   ``ETHTOOL_GREGS``                   n/a
@@ -417,6 +452,7 @@ have their netlink replacement yet.
   ``ETHTOOL_GPHYSTATS``               n/a
   ``ETHTOOL_PERQUEUE``                n/a
   ``ETHTOOL_GLINKSETTINGS``           ``ETHTOOL_MSG_LINKINFO_GET``
+                                      ``ETHTOOL_MSG_LINKMODES_GET``
   ``ETHTOOL_SLINKSETTINGS``           ``ETHTOOL_MSG_LINKINFO_SET``
   ``ETHTOOL_PHY_GTUNABLE``            n/a
   ``ETHTOOL_PHY_STUNABLE``            n/a
diff --git a/include/linux/ethtool_netlink.h b/include/linux/ethtool_netlink.h
index c98f6852c8eb..d01b77887f82 100644
--- a/include/linux/ethtool_netlink.h
+++ b/include/linux/ethtool_netlink.h
@@ -7,6 +7,9 @@
 #include <linux/ethtool.h>
 #include <linux/netdevice.h>
 
+#define __ETHTOOL_LINK_MODE_MASK_NWORDS \
+	DIV_ROUND_UP(__ETHTOOL_LINK_MODE_MASK_NBITS, 32)
+
 enum ethtool_multicast_groups {
 	ETHNL_MCGRP_MONITOR,
 };
diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
index d530fa30de36..dc1cae052eee 100644
--- a/include/uapi/linux/ethtool_netlink.h
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -17,6 +17,7 @@ enum {
 	ETHTOOL_MSG_STRSET_GET,
 	ETHTOOL_MSG_LINKINFO_GET,
 	ETHTOOL_MSG_LINKINFO_SET,
+	ETHTOOL_MSG_LINKMODES_GET,
 
 	/* add new constants above here */
 	__ETHTOOL_MSG_USER_CNT,
@@ -29,6 +30,7 @@ enum {
 	ETHTOOL_MSG_STRSET_GET_REPLY,
 	ETHTOOL_MSG_LINKINFO_GET_REPLY,
 	ETHTOOL_MSG_LINKINFO_NTF,
+	ETHTOOL_MSG_LINKMODES_GET_REPLY,
 
 	/* add new constants above here */
 	__ETHTOOL_MSG_KERNEL_CNT,
@@ -161,6 +163,22 @@ enum {
 	ETHTOOL_A_LINKINFO_MAX = __ETHTOOL_A_LINKINFO_CNT - 1
 };
 
+/* LINKMODES */
+
+enum {
+	ETHTOOL_A_LINKMODES_UNSPEC,
+	ETHTOOL_A_LINKMODES_HEADER,		/* nest - _A_HEADER_* */
+	ETHTOOL_A_LINKMODES_AUTONEG,		/* u8 */
+	ETHTOOL_A_LINKMODES_OURS,		/* bitset */
+	ETHTOOL_A_LINKMODES_PEER,		/* bitset */
+	ETHTOOL_A_LINKMODES_SPEED,		/* u32 */
+	ETHTOOL_A_LINKMODES_DUPLEX,		/* u8 */
+
+	/* add new constants above here */
+	__ETHTOOL_A_LINKMODES_CNT,
+	ETHTOOL_A_LINKMODES_MAX = __ETHTOOL_A_LINKMODES_CNT - 1
+};
+
 /* generic netlink info */
 #define ETHTOOL_GENL_NAME "ethtool"
 #define ETHTOOL_GENL_VERSION 1
diff --git a/net/ethtool/Makefile b/net/ethtool/Makefile
index 765736ec52c0..8023da6672ce 100644
--- a/net/ethtool/Makefile
+++ b/net/ethtool/Makefile
@@ -4,4 +4,4 @@ obj-y				+= ioctl.o common.o
 
 obj-$(CONFIG_ETHTOOL_NETLINK)	+= ethtool_nl.o
 
-ethtool_nl-y	:= netlink.o bitset.o strset.o linkinfo.o
+ethtool_nl-y	:= netlink.o bitset.o strset.o linkinfo.o linkmodes.o
diff --git a/net/ethtool/linkmodes.c b/net/ethtool/linkmodes.c
new file mode 100644
index 000000000000..81856fa1e632
--- /dev/null
+++ b/net/ethtool/linkmodes.c
@@ -0,0 +1,140 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include "netlink.h"
+#include "common.h"
+#include "bitset.h"
+
+struct linkmodes_req_info {
+	struct ethnl_req_info		base;
+};
+
+struct linkmodes_reply_data {
+	struct ethnl_reply_data		base;
+	struct ethtool_link_ksettings	ksettings;
+	struct ethtool_link_settings	*lsettings;
+	bool				peer_empty;
+};
+
+#define LINKMODES_REPDATA(__reply_base) \
+	container_of(__reply_base, struct linkmodes_reply_data, base)
+
+static const struct nla_policy
+linkmodes_get_policy[ETHTOOL_A_LINKMODES_MAX + 1] = {
+	[ETHTOOL_A_LINKMODES_UNSPEC]		= { .type = NLA_REJECT },
+	[ETHTOOL_A_LINKMODES_HEADER]		= { .type = NLA_NESTED },
+	[ETHTOOL_A_LINKMODES_AUTONEG]		= { .type = NLA_REJECT },
+	[ETHTOOL_A_LINKMODES_OURS]		= { .type = NLA_REJECT },
+	[ETHTOOL_A_LINKMODES_PEER]		= { .type = NLA_REJECT },
+	[ETHTOOL_A_LINKMODES_SPEED]		= { .type = NLA_REJECT },
+	[ETHTOOL_A_LINKMODES_DUPLEX]		= { .type = NLA_REJECT },
+};
+
+static int linkmodes_prepare_data(const struct ethnl_req_info *req_base,
+				  struct ethnl_reply_data *reply_base,
+				  struct genl_info *info)
+{
+	struct linkmodes_reply_data *data = LINKMODES_REPDATA(reply_base);
+	struct net_device *dev = reply_base->dev;
+	int ret;
+
+	data->lsettings = &data->ksettings.base;
+
+	ret = ethnl_ops_begin(dev);
+	if (ret < 0)
+		return ret;
+
+	ret = __ethtool_get_link_ksettings(dev, &data->ksettings);
+	if (ret < 0 && info) {
+		GENL_SET_ERR_MSG(info, "failed to retrieve link settings");
+		goto out;
+	}
+
+	data->peer_empty =
+		bitmap_empty(data->ksettings.link_modes.lp_advertising,
+			     __ETHTOOL_LINK_MODE_MASK_NBITS);
+
+out:
+	ethnl_ops_complete(dev);
+	return ret;
+}
+
+static int linkmodes_reply_size(const struct ethnl_req_info *req_base,
+				const struct ethnl_reply_data *reply_base)
+{
+	const struct linkmodes_reply_data *data = LINKMODES_REPDATA(reply_base);
+	const struct ethtool_link_ksettings *ksettings = &data->ksettings;
+	bool compact = req_base->flags & ETHTOOL_FLAG_COMPACT_BITSETS;
+	int len, ret;
+
+	len = nla_total_size(sizeof(u8)) /* LINKMODES_AUTONEG */
+		+ nla_total_size(sizeof(u32)) /* LINKMODES_SPEED */
+		+ nla_total_size(sizeof(u8)) /* LINKMODES_DUPLEX */
+		+ 0;
+	ret = ethnl_bitset_size(ksettings->link_modes.advertising,
+				ksettings->link_modes.supported,
+				__ETHTOOL_LINK_MODE_MASK_NBITS,
+				link_mode_names, compact);
+	if (ret < 0)
+		return ret;
+	len += ret;
+	if (!data->peer_empty) {
+		ret = ethnl_bitset_size(ksettings->link_modes.lp_advertising,
+					NULL, __ETHTOOL_LINK_MODE_MASK_NBITS,
+					link_mode_names, compact);
+		if (ret < 0)
+			return ret;
+		len += ret;
+	}
+
+	return len;
+}
+
+static int linkmodes_fill_reply(struct sk_buff *skb,
+				const struct ethnl_req_info *req_base,
+				const struct ethnl_reply_data *reply_base)
+{
+	const struct linkmodes_reply_data *data = LINKMODES_REPDATA(reply_base);
+	const struct ethtool_link_ksettings *ksettings = &data->ksettings;
+	const struct ethtool_link_settings *lsettings = &ksettings->base;
+	bool compact = req_base->flags & ETHTOOL_FLAG_COMPACT_BITSETS;
+	int ret;
+
+	if (nla_put_u8(skb, ETHTOOL_A_LINKMODES_AUTONEG, lsettings->autoneg))
+		return -EMSGSIZE;
+
+	ret = ethnl_put_bitset(skb, ETHTOOL_A_LINKMODES_OURS,
+			       ksettings->link_modes.advertising,
+			       ksettings->link_modes.supported,
+			       __ETHTOOL_LINK_MODE_MASK_NBITS, link_mode_names,
+			       compact);
+	if (ret < 0)
+		return -EMSGSIZE;
+	if (!data->peer_empty) {
+		ret = ethnl_put_bitset(skb, ETHTOOL_A_LINKMODES_PEER,
+				       ksettings->link_modes.lp_advertising,
+				       NULL, __ETHTOOL_LINK_MODE_MASK_NBITS,
+				       link_mode_names, compact);
+		if (ret < 0)
+			return -EMSGSIZE;
+	}
+
+	if (nla_put_u32(skb, ETHTOOL_A_LINKMODES_SPEED, lsettings->speed) ||
+	    nla_put_u8(skb, ETHTOOL_A_LINKMODES_DUPLEX, lsettings->duplex))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+const struct ethnl_request_ops ethnl_linkmodes_request_ops = {
+	.request_cmd		= ETHTOOL_MSG_LINKMODES_GET,
+	.reply_cmd		= ETHTOOL_MSG_LINKMODES_GET_REPLY,
+	.hdr_attr		= ETHTOOL_A_LINKMODES_HEADER,
+	.max_attr		= ETHTOOL_A_LINKMODES_MAX,
+	.req_info_size		= sizeof(struct linkmodes_req_info),
+	.reply_data_size	= sizeof(struct linkmodes_reply_data),
+	.request_policy		= linkmodes_get_policy,
+
+	.prepare_data		= linkmodes_prepare_data,
+	.reply_size		= linkmodes_reply_size,
+	.fill_reply		= linkmodes_fill_reply,
+};
diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index 942da4ebdfe9..703ff3a227a4 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -209,6 +209,7 @@ static const struct ethnl_request_ops *
 ethnl_default_requests[__ETHTOOL_MSG_USER_CNT] = {
 	[ETHTOOL_MSG_STRSET_GET]	= &ethnl_strset_request_ops,
 	[ETHTOOL_MSG_LINKINFO_GET]	= &ethnl_linkinfo_request_ops,
+	[ETHTOOL_MSG_LINKMODES_GET]	= &ethnl_linkmodes_request_ops,
 };
 
 static struct ethnl_dump_ctx *ethnl_dump_context(struct netlink_callback *cb)
@@ -630,6 +631,13 @@ static const struct genl_ops ethtool_genl_ops[] = {
 		.flags	= GENL_UNS_ADMIN_PERM,
 		.doit	= ethnl_set_linkinfo,
 	},
+	{
+		.cmd	= ETHTOOL_MSG_LINKMODES_GET,
+		.doit	= ethnl_default_doit,
+		.start	= ethnl_default_start,
+		.dumpit	= ethnl_default_dumpit,
+		.done	= ethnl_default_done,
+	},
 };
 
 static const struct genl_multicast_group ethtool_nl_mcgrps[] = {
diff --git a/net/ethtool/netlink.h b/net/ethtool/netlink.h
index 5d56d7779a06..256d38972d1e 100644
--- a/net/ethtool/netlink.h
+++ b/net/ethtool/netlink.h
@@ -332,6 +332,7 @@ struct ethnl_request_ops {
 
 extern const struct ethnl_request_ops ethnl_strset_request_ops;
 extern const struct ethnl_request_ops ethnl_linkinfo_request_ops;
+extern const struct ethnl_request_ops ethnl_linkmodes_request_ops;
 
 int ethnl_set_linkinfo(struct sk_buff *skb, struct genl_info *info);
 
-- 
cgit v1.2.3-59-g8ed1b


From bfbcfe2032e70bd8598d680d39ac177d507e39ac Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Fri, 27 Dec 2019 15:56:13 +0100
Subject: ethtool: set link modes related data with LINKMODES_SET request

Implement LINKMODES_SET netlink request to set advertised linkmodes and
related attributes as ETHTOOL_SLINKSETTINGS and ETHTOOL_SSET commands do.

The request allows setting autonegotiation flag, speed, duplex and
advertised link modes.

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/ethtool-netlink.rst |  27 +++
 include/uapi/linux/ethtool_netlink.h         |   1 +
 net/ethtool/linkmodes.c                      | 235 +++++++++++++++++++++++++++
 net/ethtool/netlink.c                        |   5 +
 net/ethtool/netlink.h                        |   1 +
 5 files changed, 269 insertions(+)

diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
index 04c55be0264c..625c80183563 100644
--- a/Documentation/networking/ethtool-netlink.rst
+++ b/Documentation/networking/ethtool-netlink.rst
@@ -183,6 +183,7 @@ Userspace to kernel:
   ``ETHTOOL_MSG_LINKINFO_GET``          get link settings
   ``ETHTOOL_MSG_LINKINFO_SET``          set link settings
   ``ETHTOOL_MSG_LINKMODES_GET``         get link modes info
+  ``ETHTOOL_MSG_LINKMODES_SET``         set link modes info
   ===================================== ================================
 
 Kernel to userspace:
@@ -366,6 +367,30 @@ list.
 devices supporting the request).
 
 
+LINKMODES_SET
+=============
+
+Request contents:
+
+  ====================================  ======  ==========================
+  ``ETHTOOL_A_LINKMODES_HEADER``        nested  request header
+  ``ETHTOOL_A_LINKMODES_AUTONEG``       u8      autonegotiation status
+  ``ETHTOOL_A_LINKMODES_OURS``          bitset  advertised link modes
+  ``ETHTOOL_A_LINKMODES_PEER``          bitset  partner link modes
+  ``ETHTOOL_A_LINKMODES_SPEED``         u32     link speed (Mb/s)
+  ``ETHTOOL_A_LINKMODES_DUPLEX``        u8      duplex mode
+  ====================================  ======  ==========================
+
+``ETHTOOL_A_LINKMODES_OURS`` bit set allows setting advertised link modes. If
+autonegotiation is on (either set now or kept from before), advertised modes
+are not changed (no ``ETHTOOL_A_LINKMODES_OURS`` attribute) and at least one
+of speed and duplex is specified, kernel adjusts advertised modes to all
+supported modes matching speed, duplex or both (whatever is specified). This
+autoselection is done on ethtool side with ioctl interface, netlink interface
+is supposed to allow requesting changes without knowing what exactly kernel
+supports.
+
+
 Request translation
 ===================
 
@@ -379,6 +404,7 @@ have their netlink replacement yet.
   ``ETHTOOL_GSET``                    ``ETHTOOL_MSG_LINKINFO_GET``
                                       ``ETHTOOL_MSG_LINKMODES_GET``
   ``ETHTOOL_SSET``                    ``ETHTOOL_MSG_LINKINFO_SET``
+                                      ``ETHTOOL_MSG_LINKMODES_SET``
   ``ETHTOOL_GDRVINFO``                n/a
   ``ETHTOOL_GREGS``                   n/a
   ``ETHTOOL_GWOL``                    n/a
@@ -454,6 +480,7 @@ have their netlink replacement yet.
   ``ETHTOOL_GLINKSETTINGS``           ``ETHTOOL_MSG_LINKINFO_GET``
                                       ``ETHTOOL_MSG_LINKMODES_GET``
   ``ETHTOOL_SLINKSETTINGS``           ``ETHTOOL_MSG_LINKINFO_SET``
+                                      ``ETHTOOL_MSG_LINKMODES_SET``
   ``ETHTOOL_PHY_GTUNABLE``            n/a
   ``ETHTOOL_PHY_STUNABLE``            n/a
   ``ETHTOOL_GFECPARAM``               n/a
diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
index dc1cae052eee..cddf978b98df 100644
--- a/include/uapi/linux/ethtool_netlink.h
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -18,6 +18,7 @@ enum {
 	ETHTOOL_MSG_LINKINFO_GET,
 	ETHTOOL_MSG_LINKINFO_SET,
 	ETHTOOL_MSG_LINKMODES_GET,
+	ETHTOOL_MSG_LINKMODES_SET,
 
 	/* add new constants above here */
 	__ETHTOOL_MSG_USER_CNT,
diff --git a/net/ethtool/linkmodes.c b/net/ethtool/linkmodes.c
index 81856fa1e632..790b60771d0e 100644
--- a/net/ethtool/linkmodes.c
+++ b/net/ethtool/linkmodes.c
@@ -138,3 +138,238 @@ const struct ethnl_request_ops ethnl_linkmodes_request_ops = {
 	.reply_size		= linkmodes_reply_size,
 	.fill_reply		= linkmodes_fill_reply,
 };
+
+/* LINKMODES_SET */
+
+struct link_mode_info {
+	int				speed;
+	u8				duplex;
+};
+
+#define __DEFINE_LINK_MODE_PARAMS(_speed, _type, _duplex) \
+	[ETHTOOL_LINK_MODE(_speed, _type, _duplex)] = { \
+		.speed	= SPEED_ ## _speed, \
+		.duplex	= __DUPLEX_ ## _duplex \
+	}
+#define __DUPLEX_Half DUPLEX_HALF
+#define __DUPLEX_Full DUPLEX_FULL
+#define __DEFINE_SPECIAL_MODE_PARAMS(_mode) \
+	[ETHTOOL_LINK_MODE_ ## _mode ## _BIT] = { \
+		.speed	= SPEED_UNKNOWN, \
+		.duplex	= DUPLEX_UNKNOWN, \
+	}
+
+static const struct link_mode_info link_mode_params[] = {
+	__DEFINE_LINK_MODE_PARAMS(10, T, Half),
+	__DEFINE_LINK_MODE_PARAMS(10, T, Full),
+	__DEFINE_LINK_MODE_PARAMS(100, T, Half),
+	__DEFINE_LINK_MODE_PARAMS(100, T, Full),
+	__DEFINE_LINK_MODE_PARAMS(1000, T, Half),
+	__DEFINE_LINK_MODE_PARAMS(1000, T, Full),
+	__DEFINE_SPECIAL_MODE_PARAMS(Autoneg),
+	__DEFINE_SPECIAL_MODE_PARAMS(TP),
+	__DEFINE_SPECIAL_MODE_PARAMS(AUI),
+	__DEFINE_SPECIAL_MODE_PARAMS(MII),
+	__DEFINE_SPECIAL_MODE_PARAMS(FIBRE),
+	__DEFINE_SPECIAL_MODE_PARAMS(BNC),
+	__DEFINE_LINK_MODE_PARAMS(10000, T, Full),
+	__DEFINE_SPECIAL_MODE_PARAMS(Pause),
+	__DEFINE_SPECIAL_MODE_PARAMS(Asym_Pause),
+	__DEFINE_LINK_MODE_PARAMS(2500, X, Full),
+	__DEFINE_SPECIAL_MODE_PARAMS(Backplane),
+	__DEFINE_LINK_MODE_PARAMS(1000, KX, Full),
+	__DEFINE_LINK_MODE_PARAMS(10000, KX4, Full),
+	__DEFINE_LINK_MODE_PARAMS(10000, KR, Full),
+	[ETHTOOL_LINK_MODE_10000baseR_FEC_BIT] = {
+		.speed	= SPEED_10000,
+		.duplex = DUPLEX_FULL,
+	},
+	__DEFINE_LINK_MODE_PARAMS(20000, MLD2, Full),
+	__DEFINE_LINK_MODE_PARAMS(20000, KR2, Full),
+	__DEFINE_LINK_MODE_PARAMS(40000, KR4, Full),
+	__DEFINE_LINK_MODE_PARAMS(40000, CR4, Full),
+	__DEFINE_LINK_MODE_PARAMS(40000, SR4, Full),
+	__DEFINE_LINK_MODE_PARAMS(40000, LR4, Full),
+	__DEFINE_LINK_MODE_PARAMS(56000, KR4, Full),
+	__DEFINE_LINK_MODE_PARAMS(56000, CR4, Full),
+	__DEFINE_LINK_MODE_PARAMS(56000, SR4, Full),
+	__DEFINE_LINK_MODE_PARAMS(56000, LR4, Full),
+	__DEFINE_LINK_MODE_PARAMS(25000, CR, Full),
+	__DEFINE_LINK_MODE_PARAMS(25000, KR, Full),
+	__DEFINE_LINK_MODE_PARAMS(25000, SR, Full),
+	__DEFINE_LINK_MODE_PARAMS(50000, CR2, Full),
+	__DEFINE_LINK_MODE_PARAMS(50000, KR2, Full),
+	__DEFINE_LINK_MODE_PARAMS(100000, KR4, Full),
+	__DEFINE_LINK_MODE_PARAMS(100000, SR4, Full),
+	__DEFINE_LINK_MODE_PARAMS(100000, CR4, Full),
+	__DEFINE_LINK_MODE_PARAMS(100000, LR4_ER4, Full),
+	__DEFINE_LINK_MODE_PARAMS(50000, SR2, Full),
+	__DEFINE_LINK_MODE_PARAMS(1000, X, Full),
+	__DEFINE_LINK_MODE_PARAMS(10000, CR, Full),
+	__DEFINE_LINK_MODE_PARAMS(10000, SR, Full),
+	__DEFINE_LINK_MODE_PARAMS(10000, LR, Full),
+	__DEFINE_LINK_MODE_PARAMS(10000, LRM, Full),
+	__DEFINE_LINK_MODE_PARAMS(10000, ER, Full),
+	__DEFINE_LINK_MODE_PARAMS(2500, T, Full),
+	__DEFINE_LINK_MODE_PARAMS(5000, T, Full),
+	__DEFINE_SPECIAL_MODE_PARAMS(FEC_NONE),
+	__DEFINE_SPECIAL_MODE_PARAMS(FEC_RS),
+	__DEFINE_SPECIAL_MODE_PARAMS(FEC_BASER),
+	__DEFINE_LINK_MODE_PARAMS(50000, KR, Full),
+	__DEFINE_LINK_MODE_PARAMS(50000, SR, Full),
+	__DEFINE_LINK_MODE_PARAMS(50000, CR, Full),
+	__DEFINE_LINK_MODE_PARAMS(50000, LR_ER_FR, Full),
+	__DEFINE_LINK_MODE_PARAMS(50000, DR, Full),
+	__DEFINE_LINK_MODE_PARAMS(100000, KR2, Full),
+	__DEFINE_LINK_MODE_PARAMS(100000, SR2, Full),
+	__DEFINE_LINK_MODE_PARAMS(100000, CR2, Full),
+	__DEFINE_LINK_MODE_PARAMS(100000, LR2_ER2_FR2, Full),
+	__DEFINE_LINK_MODE_PARAMS(100000, DR2, Full),
+	__DEFINE_LINK_MODE_PARAMS(200000, KR4, Full),
+	__DEFINE_LINK_MODE_PARAMS(200000, SR4, Full),
+	__DEFINE_LINK_MODE_PARAMS(200000, LR4_ER4_FR4, Full),
+	__DEFINE_LINK_MODE_PARAMS(200000, DR4, Full),
+	__DEFINE_LINK_MODE_PARAMS(200000, CR4, Full),
+	__DEFINE_LINK_MODE_PARAMS(100, T1, Full),
+	__DEFINE_LINK_MODE_PARAMS(1000, T1, Full),
+	__DEFINE_LINK_MODE_PARAMS(400000, KR8, Full),
+	__DEFINE_LINK_MODE_PARAMS(400000, SR8, Full),
+	__DEFINE_LINK_MODE_PARAMS(400000, LR8_ER8_FR8, Full),
+	__DEFINE_LINK_MODE_PARAMS(400000, DR8, Full),
+	__DEFINE_LINK_MODE_PARAMS(400000, CR8, Full),
+};
+
+static const struct nla_policy
+linkmodes_set_policy[ETHTOOL_A_LINKMODES_MAX + 1] = {
+	[ETHTOOL_A_LINKMODES_UNSPEC]		= { .type = NLA_REJECT },
+	[ETHTOOL_A_LINKMODES_HEADER]		= { .type = NLA_NESTED },
+	[ETHTOOL_A_LINKMODES_AUTONEG]		= { .type = NLA_U8 },
+	[ETHTOOL_A_LINKMODES_OURS]		= { .type = NLA_NESTED },
+	[ETHTOOL_A_LINKMODES_PEER]		= { .type = NLA_REJECT },
+	[ETHTOOL_A_LINKMODES_SPEED]		= { .type = NLA_U32 },
+	[ETHTOOL_A_LINKMODES_DUPLEX]		= { .type = NLA_U8 },
+};
+
+/* Set advertised link modes to all supported modes matching requested speed
+ * and duplex values. Called when autonegotiation is on, speed or duplex is
+ * requested but no link mode change. This is done in userspace with ioctl()
+ * interface, move it into kernel for netlink.
+ * Returns true if advertised modes bitmap was modified.
+ */
+static bool ethnl_auto_linkmodes(struct ethtool_link_ksettings *ksettings,
+				 bool req_speed, bool req_duplex)
+{
+	unsigned long *advertising = ksettings->link_modes.advertising;
+	unsigned long *supported = ksettings->link_modes.supported;
+	DECLARE_BITMAP(old_adv, __ETHTOOL_LINK_MODE_MASK_NBITS);
+	unsigned int i;
+
+	BUILD_BUG_ON(ARRAY_SIZE(link_mode_params) !=
+		     __ETHTOOL_LINK_MODE_MASK_NBITS);
+
+	bitmap_copy(old_adv, advertising, __ETHTOOL_LINK_MODE_MASK_NBITS);
+
+	for (i = 0; i < __ETHTOOL_LINK_MODE_MASK_NBITS; i++) {
+		const struct link_mode_info *info = &link_mode_params[i];
+
+		if (info->speed == SPEED_UNKNOWN)
+			continue;
+		if (test_bit(i, supported) &&
+		    (!req_speed || info->speed == ksettings->base.speed) &&
+		    (!req_duplex || info->duplex == ksettings->base.duplex))
+			set_bit(i, advertising);
+		else
+			clear_bit(i, advertising);
+	}
+
+	return !bitmap_equal(old_adv, advertising,
+			     __ETHTOOL_LINK_MODE_MASK_NBITS);
+}
+
+static int ethnl_update_linkmodes(struct genl_info *info, struct nlattr **tb,
+				  struct ethtool_link_ksettings *ksettings,
+				  bool *mod)
+{
+	struct ethtool_link_settings *lsettings = &ksettings->base;
+	bool req_speed, req_duplex;
+	int ret;
+
+	*mod = false;
+	req_speed = tb[ETHTOOL_A_LINKMODES_SPEED];
+	req_duplex = tb[ETHTOOL_A_LINKMODES_DUPLEX];
+
+	ethnl_update_u8(&lsettings->autoneg, tb[ETHTOOL_A_LINKMODES_AUTONEG],
+			mod);
+	ret = ethnl_update_bitset(ksettings->link_modes.advertising,
+				  __ETHTOOL_LINK_MODE_MASK_NBITS,
+				  tb[ETHTOOL_A_LINKMODES_OURS], link_mode_names,
+				  info->extack, mod);
+	if (ret < 0)
+		return ret;
+	ethnl_update_u32(&lsettings->speed, tb[ETHTOOL_A_LINKMODES_SPEED],
+			 mod);
+	ethnl_update_u8(&lsettings->duplex, tb[ETHTOOL_A_LINKMODES_DUPLEX],
+			mod);
+
+	if (!tb[ETHTOOL_A_LINKMODES_OURS] && lsettings->autoneg &&
+	    (req_speed || req_duplex) &&
+	    ethnl_auto_linkmodes(ksettings, req_speed, req_duplex))
+		*mod = true;
+
+	return 0;
+}
+
+int ethnl_set_linkmodes(struct sk_buff *skb, struct genl_info *info)
+{
+	struct nlattr *tb[ETHTOOL_A_LINKMODES_MAX + 1];
+	struct ethtool_link_ksettings ksettings = {};
+	struct ethtool_link_settings *lsettings;
+	struct ethnl_req_info req_info = {};
+	struct net_device *dev;
+	bool mod = false;
+	int ret;
+
+	ret = nlmsg_parse(info->nlhdr, GENL_HDRLEN, tb,
+			  ETHTOOL_A_LINKMODES_MAX, linkmodes_set_policy,
+			  info->extack);
+	if (ret < 0)
+		return ret;
+	ret = ethnl_parse_header(&req_info, tb[ETHTOOL_A_LINKMODES_HEADER],
+				 genl_info_net(info), info->extack, true);
+	if (ret < 0)
+		return ret;
+	dev = req_info.dev;
+	if (!dev->ethtool_ops->get_link_ksettings ||
+	    !dev->ethtool_ops->set_link_ksettings)
+		return -EOPNOTSUPP;
+
+	rtnl_lock();
+	ret = ethnl_ops_begin(dev);
+	if (ret < 0)
+		goto out_rtnl;
+
+	ret = __ethtool_get_link_ksettings(dev, &ksettings);
+	if (ret < 0) {
+		if (info)
+			GENL_SET_ERR_MSG(info, "failed to retrieve link settings");
+		goto out_ops;
+	}
+	lsettings = &ksettings.base;
+
+	ret = ethnl_update_linkmodes(info, tb, &ksettings, &mod);
+	if (ret < 0)
+		goto out_ops;
+
+	if (mod) {
+		ret = dev->ethtool_ops->set_link_ksettings(dev, &ksettings);
+		if (ret < 0)
+			GENL_SET_ERR_MSG(info, "link settings update failed");
+	}
+
+out_ops:
+	ethnl_ops_complete(dev);
+out_rtnl:
+	rtnl_unlock();
+	dev_put(dev);
+	return ret;
+}
diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index 703ff3a227a4..5f28f3cb022d 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -638,6 +638,11 @@ static const struct genl_ops ethtool_genl_ops[] = {
 		.dumpit	= ethnl_default_dumpit,
 		.done	= ethnl_default_done,
 	},
+	{
+		.cmd	= ETHTOOL_MSG_LINKMODES_SET,
+		.flags	= GENL_UNS_ADMIN_PERM,
+		.doit	= ethnl_set_linkmodes,
+	},
 };
 
 static const struct genl_multicast_group ethtool_nl_mcgrps[] = {
diff --git a/net/ethtool/netlink.h b/net/ethtool/netlink.h
index 256d38972d1e..1269cca8a002 100644
--- a/net/ethtool/netlink.h
+++ b/net/ethtool/netlink.h
@@ -335,5 +335,6 @@ extern const struct ethnl_request_ops ethnl_linkinfo_request_ops;
 extern const struct ethnl_request_ops ethnl_linkmodes_request_ops;
 
 int ethnl_set_linkinfo(struct sk_buff *skb, struct genl_info *info);
+int ethnl_set_linkmodes(struct sk_buff *skb, struct genl_info *info);
 
 #endif /* _NET_ETHTOOL_NETLINK_H */
-- 
cgit v1.2.3-59-g8ed1b


From 1b1b1847c8505df2e1dd8804838526bed22a8bd4 Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Fri, 27 Dec 2019 15:56:18 +0100
Subject: ethtool: add LINKMODES_NTF notification

Send ETHTOOL_MSG_LINKMODES_NTF notification message whenever device link
settings or advertised modes are modified using ETHTOOL_MSG_LINKMODES_SET
netlink message or ETHTOOL_SLINKSETTINGS or ETHTOOL_SSET ioctl commands.

The notification message has the same format as reply to LINKMODES_GET
request. ETHTOOL_MSG_LINKMODES_SET netlink request only triggers the
notification if there is a change but the ioctl command handlers do not
check if there is an actual change and trigger the notification whenever
the commands are executed.

As all work is done by ethnl_default_notify() handler and callback
functions introduced to handle LINKMODES_GET requests, all that remains is
adding entries for ETHTOOL_MSG_LINKMODES_NTF into ethnl_notify_handlers and
ethnl_default_notify_ops lookup tables and calls to ethtool_notify() where
needed.

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/ethtool-netlink.rst | 1 +
 include/uapi/linux/ethtool_netlink.h         | 1 +
 net/ethtool/ioctl.c                          | 8 ++++++--
 net/ethtool/linkmodes.c                      | 2 ++
 net/ethtool/netlink.c                        | 2 ++
 5 files changed, 12 insertions(+), 2 deletions(-)

diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
index 625c80183563..9d96d51e9360 100644
--- a/Documentation/networking/ethtool-netlink.rst
+++ b/Documentation/networking/ethtool-netlink.rst
@@ -193,6 +193,7 @@ Kernel to userspace:
   ``ETHTOOL_MSG_LINKINFO_GET_REPLY``    link settings
   ``ETHTOOL_MSG_LINKINFO_NTF``          link settings notification
   ``ETHTOOL_MSG_LINKMODES_GET_REPLY``   link modes info
+  ``ETHTOOL_MSG_LINKMODES_NTF``         link modes notification
   ===================================== ================================
 
 ``GET`` requests are sent by userspace applications to retrieve device
diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
index cddf978b98df..35948df6d6e3 100644
--- a/include/uapi/linux/ethtool_netlink.h
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -32,6 +32,7 @@ enum {
 	ETHTOOL_MSG_LINKINFO_GET_REPLY,
 	ETHTOOL_MSG_LINKINFO_NTF,
 	ETHTOOL_MSG_LINKMODES_GET_REPLY,
+	ETHTOOL_MSG_LINKMODES_NTF,
 
 	/* add new constants above here */
 	__ETHTOOL_MSG_KERNEL_CNT,
diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
index 11a467294a33..36e2ef2d900d 100644
--- a/net/ethtool/ioctl.c
+++ b/net/ethtool/ioctl.c
@@ -573,8 +573,10 @@ static int ethtool_set_link_ksettings(struct net_device *dev,
 		return -EINVAL;
 
 	err = dev->ethtool_ops->set_link_ksettings(dev, &link_ksettings);
-	if (err >= 0)
+	if (err >= 0) {
 		ethtool_notify(dev, ETHTOOL_MSG_LINKINFO_NTF, NULL);
+		ethtool_notify(dev, ETHTOOL_MSG_LINKMODES_NTF, NULL);
+	}
 	return err;
 }
 
@@ -638,8 +640,10 @@ static int ethtool_set_settings(struct net_device *dev, void __user *useraddr)
 	link_ksettings.base.link_mode_masks_nwords =
 		__ETHTOOL_LINK_MODE_MASK_NU32;
 	ret = dev->ethtool_ops->set_link_ksettings(dev, &link_ksettings);
-	if (ret >= 0)
+	if (ret >= 0) {
 		ethtool_notify(dev, ETHTOOL_MSG_LINKINFO_NTF, NULL);
+		ethtool_notify(dev, ETHTOOL_MSG_LINKMODES_NTF, NULL);
+	}
 	return ret;
 }
 
diff --git a/net/ethtool/linkmodes.c b/net/ethtool/linkmodes.c
index 790b60771d0e..0b99f494ad3b 100644
--- a/net/ethtool/linkmodes.c
+++ b/net/ethtool/linkmodes.c
@@ -364,6 +364,8 @@ int ethnl_set_linkmodes(struct sk_buff *skb, struct genl_info *info)
 		ret = dev->ethtool_ops->set_link_ksettings(dev, &ksettings);
 		if (ret < 0)
 			GENL_SET_ERR_MSG(info, "link settings update failed");
+		else
+			ethtool_notify(dev, ETHTOOL_MSG_LINKMODES_NTF, NULL);
 	}
 
 out_ops:
diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index 5f28f3cb022d..1b5e1bd26504 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -511,6 +511,7 @@ static int ethnl_default_done(struct netlink_callback *cb)
 static const struct ethnl_request_ops *
 ethnl_default_notify_ops[ETHTOOL_MSG_KERNEL_MAX + 1] = {
 	[ETHTOOL_MSG_LINKINFO_NTF]	= &ethnl_linkinfo_request_ops,
+	[ETHTOOL_MSG_LINKMODES_NTF]	= &ethnl_linkmodes_request_ops,
 };
 
 /* default notification handler */
@@ -592,6 +593,7 @@ typedef void (*ethnl_notify_handler_t)(struct net_device *dev, unsigned int cmd,
 
 static const ethnl_notify_handler_t ethnl_notify_handlers[] = {
 	[ETHTOOL_MSG_LINKINFO_NTF]	= ethnl_default_notify,
+	[ETHTOOL_MSG_LINKMODES_NTF]	= ethnl_default_notify,
 };
 
 void ethtool_notify(struct net_device *dev, unsigned int cmd, const void *data)
-- 
cgit v1.2.3-59-g8ed1b


From 3d2b847fb99cf2b28aa046e486636e555bc6ed1c Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Fri, 27 Dec 2019 15:56:23 +0100
Subject: ethtool: provide link state with LINKSTATE_GET request

Implement LINKSTATE_GET netlink request to get link state information.

At the moment, only link up flag as provided by ETHTOOL_GLINK ioctl command
is returned.

LINKSTATE_GET request can be used with NLM_F_DUMP (without device
identification) to request the information for all devices in current
network namespace providing the data.

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/ethtool-netlink.rst | 33 ++++++++++++-
 include/uapi/linux/ethtool_netlink.h         | 14 ++++++
 net/ethtool/Makefile                         |  3 +-
 net/ethtool/common.c                         |  8 +++
 net/ethtool/common.h                         |  3 ++
 net/ethtool/ioctl.c                          |  8 +--
 net/ethtool/linkstate.c                      | 74 ++++++++++++++++++++++++++++
 net/ethtool/netlink.c                        |  8 +++
 net/ethtool/netlink.h                        |  1 +
 9 files changed, 146 insertions(+), 6 deletions(-)
 create mode 100644 net/ethtool/linkstate.c

diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
index 9d96d51e9360..c60afba69e3c 100644
--- a/Documentation/networking/ethtool-netlink.rst
+++ b/Documentation/networking/ethtool-netlink.rst
@@ -184,6 +184,7 @@ Userspace to kernel:
   ``ETHTOOL_MSG_LINKINFO_SET``          set link settings
   ``ETHTOOL_MSG_LINKMODES_GET``         get link modes info
   ``ETHTOOL_MSG_LINKMODES_SET``         set link modes info
+  ``ETHTOOL_MSG_LINKSTATE_GET``         get link state
   ===================================== ================================
 
 Kernel to userspace:
@@ -194,6 +195,7 @@ Kernel to userspace:
   ``ETHTOOL_MSG_LINKINFO_NTF``          link settings notification
   ``ETHTOOL_MSG_LINKMODES_GET_REPLY``   link modes info
   ``ETHTOOL_MSG_LINKMODES_NTF``         link modes notification
+  ``ETHTOOL_MSG_LINKSTATE_GET_REPLY``   link state info
   ===================================== ================================
 
 ``GET`` requests are sent by userspace applications to retrieve device
@@ -392,6 +394,35 @@ is supposed to allow requesting changes without knowing what exactly kernel
 supports.
 
 
+LINKSTATE_GET
+=============
+
+Requests link state information. At the moment, only link up/down flag (as
+provided by ``ETHTOOL_GLINK`` ioctl command) is provided but some future
+extensions are planned (e.g. link down reason). This request does not have any
+attributes.
+
+Request contents:
+
+  ====================================  ======  ==========================
+  ``ETHTOOL_A_LINKSTATE_HEADER``        nested  request header
+  ====================================  ======  ==========================
+
+Kernel response contents:
+
+  ====================================  ======  ==========================
+  ``ETHTOOL_A_LINKSTATE_HEADER``        nested  reply header
+  ``ETHTOOL_A_LINKSTATE_LINK``          bool    link state (up/down)
+  ====================================  ======  ==========================
+
+For most NIC drivers, the value of ``ETHTOOL_A_LINKSTATE_LINK`` returns
+carrier flag provided by ``netif_carrier_ok()`` but there are drivers which
+define their own handler.
+
+``LINKSTATE_GET`` allows dump requests (kernel returns reply messages for all
+devices supporting the request).
+
+
 Request translation
 ===================
 
@@ -413,7 +444,7 @@ have their netlink replacement yet.
   ``ETHTOOL_GMSGLVL``                 n/a
   ``ETHTOOL_SMSGLVL``                 n/a
   ``ETHTOOL_NWAY_RST``                n/a
-  ``ETHTOOL_GLINK``                   n/a
+  ``ETHTOOL_GLINK``                   ``ETHTOOL_MSG_LINKSTATE_GET``
   ``ETHTOOL_GEEPROM``                 n/a
   ``ETHTOOL_SEEPROM``                 n/a
   ``ETHTOOL_GCOALESCE``               n/a
diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
index 35948df6d6e3..02f82f42a889 100644
--- a/include/uapi/linux/ethtool_netlink.h
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -19,6 +19,7 @@ enum {
 	ETHTOOL_MSG_LINKINFO_SET,
 	ETHTOOL_MSG_LINKMODES_GET,
 	ETHTOOL_MSG_LINKMODES_SET,
+	ETHTOOL_MSG_LINKSTATE_GET,
 
 	/* add new constants above here */
 	__ETHTOOL_MSG_USER_CNT,
@@ -33,6 +34,7 @@ enum {
 	ETHTOOL_MSG_LINKINFO_NTF,
 	ETHTOOL_MSG_LINKMODES_GET_REPLY,
 	ETHTOOL_MSG_LINKMODES_NTF,
+	ETHTOOL_MSG_LINKSTATE_GET_REPLY,
 
 	/* add new constants above here */
 	__ETHTOOL_MSG_KERNEL_CNT,
@@ -181,6 +183,18 @@ enum {
 	ETHTOOL_A_LINKMODES_MAX = __ETHTOOL_A_LINKMODES_CNT - 1
 };
 
+/* LINKSTATE */
+
+enum {
+	ETHTOOL_A_LINKSTATE_UNSPEC,
+	ETHTOOL_A_LINKSTATE_HEADER,		/* nest - _A_HEADER_* */
+	ETHTOOL_A_LINKSTATE_LINK,		/* u8 */
+
+	/* add new constants above here */
+	__ETHTOOL_A_LINKSTATE_CNT,
+	ETHTOOL_A_LINKSTATE_MAX = __ETHTOOL_A_LINKSTATE_CNT - 1
+};
+
 /* generic netlink info */
 #define ETHTOOL_GENL_NAME "ethtool"
 #define ETHTOOL_GENL_VERSION 1
diff --git a/net/ethtool/Makefile b/net/ethtool/Makefile
index 8023da6672ce..9a1332fb0cc6 100644
--- a/net/ethtool/Makefile
+++ b/net/ethtool/Makefile
@@ -4,4 +4,5 @@ obj-y				+= ioctl.o common.o
 
 obj-$(CONFIG_ETHTOOL_NETLINK)	+= ethtool_nl.o
 
-ethtool_nl-y	:= netlink.o bitset.o strset.o linkinfo.o linkmodes.o
+ethtool_nl-y	:= netlink.o bitset.o strset.o linkinfo.o linkmodes.o \
+		   linkstate.o
diff --git a/net/ethtool/common.c b/net/ethtool/common.c
index 1d4a0aeff2cb..e621b1694d2f 100644
--- a/net/ethtool/common.c
+++ b/net/ethtool/common.c
@@ -217,3 +217,11 @@ convert_legacy_settings_to_link_ksettings(
 		= legacy_settings->eth_tp_mdix_ctrl;
 	return retval;
 }
+
+int __ethtool_get_link(struct net_device *dev)
+{
+	if (!dev->ethtool_ops->get_link)
+		return -EOPNOTSUPP;
+
+	return netif_running(dev) && dev->ethtool_ops->get_link(dev);
+}
diff --git a/net/ethtool/common.h b/net/ethtool/common.h
index c8a237402729..5c5f7dc90cd4 100644
--- a/net/ethtool/common.h
+++ b/net/ethtool/common.h
@@ -3,6 +3,7 @@
 #ifndef _ETHTOOL_COMMON_H
 #define _ETHTOOL_COMMON_H
 
+#include <linux/netdevice.h>
 #include <linux/ethtool.h>
 
 /* compose link mode index from speed, type and duplex */
@@ -19,6 +20,8 @@ extern const char
 phy_tunable_strings[__ETHTOOL_PHY_TUNABLE_COUNT][ETH_GSTRING_LEN];
 extern const char link_mode_names[][ETH_GSTRING_LEN];
 
+int __ethtool_get_link(struct net_device *dev);
+
 bool convert_legacy_settings_to_link_ksettings(
 	struct ethtool_link_ksettings *link_ksettings,
 	const struct ethtool_cmd *legacy_settings);
diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
index 36e2ef2d900d..182bffbffa78 100644
--- a/net/ethtool/ioctl.c
+++ b/net/ethtool/ioctl.c
@@ -1365,12 +1365,12 @@ static int ethtool_nway_reset(struct net_device *dev)
 static int ethtool_get_link(struct net_device *dev, char __user *useraddr)
 {
 	struct ethtool_value edata = { .cmd = ETHTOOL_GLINK };
+	int link = __ethtool_get_link(dev);
 
-	if (!dev->ethtool_ops->get_link)
-		return -EOPNOTSUPP;
-
-	edata.data = netif_running(dev) && dev->ethtool_ops->get_link(dev);
+	if (link < 0)
+		return link;
 
+	edata.data = link;
 	if (copy_to_user(useraddr, &edata, sizeof(edata)))
 		return -EFAULT;
 	return 0;
diff --git a/net/ethtool/linkstate.c b/net/ethtool/linkstate.c
new file mode 100644
index 000000000000..2740cde0a182
--- /dev/null
+++ b/net/ethtool/linkstate.c
@@ -0,0 +1,74 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include "netlink.h"
+#include "common.h"
+
+struct linkstate_req_info {
+	struct ethnl_req_info		base;
+};
+
+struct linkstate_reply_data {
+	struct ethnl_reply_data		base;
+	int				link;
+};
+
+#define LINKSTATE_REPDATA(__reply_base) \
+	container_of(__reply_base, struct linkstate_reply_data, base)
+
+static const struct nla_policy
+linkstate_get_policy[ETHTOOL_A_LINKSTATE_MAX + 1] = {
+	[ETHTOOL_A_LINKSTATE_UNSPEC]		= { .type = NLA_REJECT },
+	[ETHTOOL_A_LINKSTATE_HEADER]		= { .type = NLA_NESTED },
+	[ETHTOOL_A_LINKSTATE_LINK]		= { .type = NLA_REJECT },
+};
+
+static int linkstate_prepare_data(const struct ethnl_req_info *req_base,
+				  struct ethnl_reply_data *reply_base,
+				  struct genl_info *info)
+{
+	struct linkstate_reply_data *data = LINKSTATE_REPDATA(reply_base);
+	struct net_device *dev = reply_base->dev;
+	int ret;
+
+	ret = ethnl_ops_begin(dev);
+	if (ret < 0)
+		return ret;
+	data->link = __ethtool_get_link(dev);
+	ethnl_ops_complete(dev);
+
+	return 0;
+}
+
+static int linkstate_reply_size(const struct ethnl_req_info *req_base,
+				const struct ethnl_reply_data *reply_base)
+{
+	return nla_total_size(sizeof(u8)) /* LINKSTATE_LINK */
+		+ 0;
+}
+
+static int linkstate_fill_reply(struct sk_buff *skb,
+				const struct ethnl_req_info *req_base,
+				const struct ethnl_reply_data *reply_base)
+{
+	struct linkstate_reply_data *data = LINKSTATE_REPDATA(reply_base);
+
+	if (data->link >= 0 &&
+	    nla_put_u8(skb, ETHTOOL_A_LINKSTATE_LINK, !!data->link))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+const struct ethnl_request_ops ethnl_linkstate_request_ops = {
+	.request_cmd		= ETHTOOL_MSG_LINKSTATE_GET,
+	.reply_cmd		= ETHTOOL_MSG_LINKSTATE_GET_REPLY,
+	.hdr_attr		= ETHTOOL_A_LINKSTATE_HEADER,
+	.max_attr		= ETHTOOL_A_LINKSTATE_MAX,
+	.req_info_size		= sizeof(struct linkstate_req_info),
+	.reply_data_size	= sizeof(struct linkstate_reply_data),
+	.request_policy		= linkstate_get_policy,
+
+	.prepare_data		= linkstate_prepare_data,
+	.reply_size		= linkstate_reply_size,
+	.fill_reply		= linkstate_fill_reply,
+};
diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index 1b5e1bd26504..4ca96c7b86b3 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -210,6 +210,7 @@ ethnl_default_requests[__ETHTOOL_MSG_USER_CNT] = {
 	[ETHTOOL_MSG_STRSET_GET]	= &ethnl_strset_request_ops,
 	[ETHTOOL_MSG_LINKINFO_GET]	= &ethnl_linkinfo_request_ops,
 	[ETHTOOL_MSG_LINKMODES_GET]	= &ethnl_linkmodes_request_ops,
+	[ETHTOOL_MSG_LINKSTATE_GET]	= &ethnl_linkstate_request_ops,
 };
 
 static struct ethnl_dump_ctx *ethnl_dump_context(struct netlink_callback *cb)
@@ -645,6 +646,13 @@ static const struct genl_ops ethtool_genl_ops[] = {
 		.flags	= GENL_UNS_ADMIN_PERM,
 		.doit	= ethnl_set_linkmodes,
 	},
+	{
+		.cmd	= ETHTOOL_MSG_LINKSTATE_GET,
+		.doit	= ethnl_default_doit,
+		.start	= ethnl_default_start,
+		.dumpit	= ethnl_default_dumpit,
+		.done	= ethnl_default_done,
+	},
 };
 
 static const struct genl_multicast_group ethtool_nl_mcgrps[] = {
diff --git a/net/ethtool/netlink.h b/net/ethtool/netlink.h
index 1269cca8a002..da9d6521a4eb 100644
--- a/net/ethtool/netlink.h
+++ b/net/ethtool/netlink.h
@@ -333,6 +333,7 @@ struct ethnl_request_ops {
 extern const struct ethnl_request_ops ethnl_strset_request_ops;
 extern const struct ethnl_request_ops ethnl_linkinfo_request_ops;
 extern const struct ethnl_request_ops ethnl_linkmodes_request_ops;
+extern const struct ethnl_request_ops ethnl_linkstate_request_ops;
 
 int ethnl_set_linkinfo(struct sk_buff *skb, struct genl_info *info);
 int ethnl_set_linkmodes(struct sk_buff *skb, struct genl_info *info);
-- 
cgit v1.2.3-59-g8ed1b


From 270fe2ceda66b6964d4c6f261d7f562a02c1c786 Mon Sep 17 00:00:00 2001
From: Chen Zhou <chenzhou10@huawei.com>
Date: Sat, 28 Dec 2019 11:09:47 +0800
Subject: net/wan/fsl_ucc_hdlc: remove set but not used variables 'ut_info' and
 'ret'

Fixes gcc '-Wunused-but-set-variable' warning:

drivers/net/wan/fsl_ucc_hdlc.c: In function ucc_hdlc_irq_handler:
drivers/net/wan/fsl_ucc_hdlc.c:643:23:
	warning: variable ut_info set but not used [-Wunused-but-set-variable]
drivers/net/wan/fsl_ucc_hdlc.c: In function uhdlc_suspend:
drivers/net/wan/fsl_ucc_hdlc.c:880:23:
	warning: variable ut_info set but not used [-Wunused-but-set-variable]
drivers/net/wan/fsl_ucc_hdlc.c: In function uhdlc_resume:
drivers/net/wan/fsl_ucc_hdlc.c:925:6:
	warning: variable ret set but not used [-Wunused-but-set-variable]

Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Chen Zhou <chenzhou10@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/wan/fsl_ucc_hdlc.c | 14 +++++---------
 1 file changed, 5 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wan/fsl_ucc_hdlc.c b/drivers/net/wan/fsl_ucc_hdlc.c
index 308384756e6f..c28f8409067e 100644
--- a/drivers/net/wan/fsl_ucc_hdlc.c
+++ b/drivers/net/wan/fsl_ucc_hdlc.c
@@ -635,11 +635,9 @@ static irqreturn_t ucc_hdlc_irq_handler(int irq, void *dev_id)
 	struct ucc_hdlc_private *priv = (struct ucc_hdlc_private *)dev_id;
 	struct net_device *dev = priv->ndev;
 	struct ucc_fast_private *uccf;
-	struct ucc_tdm_info *ut_info;
 	u32 ucce;
 	u32 uccm;
 
-	ut_info = priv->ut_info;
 	uccf = priv->uccf;
 
 	ucce = ioread32be(uccf->p_ucce);
@@ -872,7 +870,6 @@ static void resume_clk_config(struct ucc_hdlc_private *priv)
 static int uhdlc_suspend(struct device *dev)
 {
 	struct ucc_hdlc_private *priv = dev_get_drvdata(dev);
-	struct ucc_tdm_info *ut_info;
 	struct ucc_fast __iomem *uf_regs;
 
 	if (!priv)
@@ -884,7 +881,6 @@ static int uhdlc_suspend(struct device *dev)
 	netif_device_detach(priv->ndev);
 	napi_disable(&priv->napi);
 
-	ut_info = priv->ut_info;
 	uf_regs = priv->uf_regs;
 
 	/* backup gumr guemr*/
@@ -917,7 +913,7 @@ static int uhdlc_resume(struct device *dev)
 	struct ucc_fast __iomem *uf_regs;
 	struct ucc_fast_private *uccf;
 	struct ucc_fast_info *uf_info;
-	int ret, i;
+	int i;
 	u32 cecr_subblock;
 	u16 bd_status;
 
@@ -962,16 +958,16 @@ static int uhdlc_resume(struct device *dev)
 
 	/* Write to QE CECR, UCCx channel to Stop Transmission */
 	cecr_subblock = ucc_fast_get_qe_cr_subblock(uf_info->ucc_num);
-	ret = qe_issue_cmd(QE_STOP_TX, cecr_subblock,
-			   (u8)QE_CR_PROTOCOL_UNSPECIFIED, 0);
+	qe_issue_cmd(QE_STOP_TX, cecr_subblock,
+		     (u8)QE_CR_PROTOCOL_UNSPECIFIED, 0);
 
 	/* Set UPSMR normal mode */
 	iowrite32be(0, &uf_regs->upsmr);
 
 	/* init parameter base */
 	cecr_subblock = ucc_fast_get_qe_cr_subblock(uf_info->ucc_num);
-	ret = qe_issue_cmd(QE_ASSIGN_PAGE_TO_DEVICE, cecr_subblock,
-			   QE_CR_PROTOCOL_UNSPECIFIED, priv->ucc_pram_offset);
+	qe_issue_cmd(QE_ASSIGN_PAGE_TO_DEVICE, cecr_subblock,
+		     QE_CR_PROTOCOL_UNSPECIFIED, priv->ucc_pram_offset);
 
 	priv->ucc_pram = (struct ucc_hdlc_param __iomem *)
 				qe_muram_addr(priv->ucc_pram_offset);
-- 
cgit v1.2.3-59-g8ed1b


From c26a2c2ddc0115eb088873f5c309cf46b982f522 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <olteanv@gmail.com>
Date: Sat, 28 Dec 2019 15:30:45 +0200
Subject: gianfar: Fix TX timestamping with a stacked DSA driver

The driver wrongly assumes that it is the only entity that can set the
SKBTX_IN_PROGRESS bit of the current skb. Therefore, in the
gfar_clean_tx_ring function, where the TX timestamp is collected if
necessary, the aforementioned bit is used to discriminate whether or not
the TX timestamp should be delivered to the socket's error queue.

But a stacked driver such as a DSA switch can also set the
SKBTX_IN_PROGRESS bit, which is actually exactly what it should do in
order to denote that the hardware timestamping process is undergoing.

Therefore, gianfar would misinterpret the "in progress" bit as being its
own, and deliver a second skb clone in the socket's error queue,
completely throwing off a PTP process which is not expecting to receive
it, _even though_ TX timestamping is not enabled for gianfar.

There have been discussions [0] as to whether non-MAC drivers need or
not to set SKBTX_IN_PROGRESS at all (whose purpose is to avoid sending 2
timestamps, a sw and a hw one, to applications which only expect one).
But as of this patch, there are at least 2 PTP drivers that would break
in conjunction with gianfar: the sja1105 DSA switch and the felix
switch, by way of its ocelot core driver.

So regardless of that conclusion, fix the gianfar driver to not do stuff
based on flags set by others and not intended for it.

[0]: https://www.spinics.net/lists/netdev/msg619699.html

Fixes: f0ee7acfcdd4 ("gianfar: Add hardware TX timestamping support")
Signed-off-by: Vladimir Oltean <olteanv@gmail.com>
Acked-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/freescale/gianfar.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net/ethernet/freescale/gianfar.c
index b636d83a7ee9..f7e5cafe89a9 100644
--- a/drivers/net/ethernet/freescale/gianfar.c
+++ b/drivers/net/ethernet/freescale/gianfar.c
@@ -2205,13 +2205,17 @@ static void gfar_clean_tx_ring(struct gfar_priv_tx_q *tx_queue)
 	skb_dirtytx = tx_queue->skb_dirtytx;
 
 	while ((skb = tx_queue->tx_skbuff[skb_dirtytx])) {
+		bool do_tstamp;
+
+		do_tstamp = (skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) &&
+			    priv->hwts_tx_en;
 
 		frags = skb_shinfo(skb)->nr_frags;
 
 		/* When time stamping, one additional TxBD must be freed.
 		 * Also, we need to dma_unmap_single() the TxPAL.
 		 */
-		if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS))
+		if (unlikely(do_tstamp))
 			nr_txbds = frags + 2;
 		else
 			nr_txbds = frags + 1;
@@ -2225,7 +2229,7 @@ static void gfar_clean_tx_ring(struct gfar_priv_tx_q *tx_queue)
 		    (lstatus & BD_LENGTH_MASK))
 			break;
 
-		if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS)) {
+		if (unlikely(do_tstamp)) {
 			next = next_txbd(bdp, base, tx_ring_size);
 			buflen = be16_to_cpu(next->length) +
 				 GMAC_FCB_LEN + GMAC_TXPAL_LEN;
@@ -2235,7 +2239,7 @@ static void gfar_clean_tx_ring(struct gfar_priv_tx_q *tx_queue)
 		dma_unmap_single(priv->dev, be32_to_cpu(bdp->bufPtr),
 				 buflen, DMA_TO_DEVICE);
 
-		if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS)) {
+		if (unlikely(do_tstamp)) {
 			struct skb_shared_hwtstamps shhwtstamps;
 			u64 *ns = (u64 *)(((uintptr_t)skb->data + 0x10) &
 					  ~0x7UL);
-- 
cgit v1.2.3-59-g8ed1b


From f685e609a301f171b62ee52a69acc3a74c2c04aa Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <olteanv@gmail.com>
Date: Sat, 28 Dec 2019 15:30:46 +0200
Subject: net: dsa: Deny PTP on master if switch supports it

It is possible to kill PTP on a DSA switch completely and absolutely,
until a reboot, with a simple command:

tcpdump -i eth2 -j adapter_unsynced

where eth2 is the switch's DSA master.

Why? Well, in short, the PTP API in place today is a bit rudimentary and
relies on applications to retrieve the TX timestamps by polling the
error queue and looking at the cmsg structure. But there is no timestamp
identification of any sorts (except whether it's HW or SW), you don't
know how many more timestamps are there to come, which one is this one,
from whom it is, etc. In other words, the SO_TIMESTAMPING API is
fundamentally limited in that you can get a single HW timestamp from the
stack.

And the "-j adapter_unsynced" flag of tcpdump enables hardware
timestamping.

So let's imagine what happens when the DSA master decides it wants to
deliver TX timestamps to the skb's socket too:
- The timestamp that the user space sees is taken by the DSA master.
  Whereas the RX timestamp will eventually be overwritten by the DSA
  switch. So the RX and TX timestamps will be in different time bases
  (aka garbage).
- The user space applications have no way to deal with the second (real)
  TX timestamp finally delivered by the DSA switch, or even to know to
  wait for it.

Take ptp4l from the linuxptp project, for example. This is its behavior
after running tcpdump, before the patch:

ptp4l[172]: [6469.594] Unexpected data on socket err queue:
ptp4l[172]: [6469.693] rms    8 max   16 freq -21257 +/-  11 delay   748 +/-   0
ptp4l[172]: [6469.711] Unexpected data on socket err queue:
ptp4l[172]: 0020 00 00 00 1f 7b ff fe 63 02 48 00 03 aa 05 00 fd
ptp4l[172]: 0030 00 00 00 00 00 00 00 00 00 00
ptp4l[172]: [6469.721] Unexpected data on socket err queue:
ptp4l[172]: 0000 01 80 c2 00 00 0e 00 1f 7b 63 02 48 88 f7 10 02
ptp4l[172]: 0010 00 2c 00 00 02 00 00 00 00 00 00 00 00 00 00 00
ptp4l[172]: 0020 00 00 00 1f 7b ff fe 63 02 48 00 01 c6 b1 00 fd
ptp4l[172]: 0030 00 00 00 00 00 00 00 00 00 00
ptp4l[172]: [6469.838] Unexpected data on socket err queue:
ptp4l[172]: 0000 01 80 c2 00 00 0e 00 1f 7b 63 02 48 88 f7 10 02
ptp4l[172]: 0010 00 2c 00 00 02 00 00 00 00 00 00 00 00 00 00 00
ptp4l[172]: 0020 00 00 00 1f 7b ff fe 63 02 48 00 03 aa 06 00 fd
ptp4l[172]: 0030 00 00 00 00 00 00 00 00 00 00
ptp4l[172]: [6469.848] Unexpected data on socket err queue:
ptp4l[172]: 0000 01 80 c2 00 00 0e 00 1f 7b 63 02 48 88 f7 13 02
ptp4l[172]: 0010 00 36 00 00 02 00 00 00 00 00 00 00 00 00 00 00
ptp4l[172]: 0020 00 00 00 1f 7b ff fe 63 02 48 00 04 1a 45 05 7f
ptp4l[172]: 0030 00 00 5e 05 41 32 27 c2 1a 68 00 04 9f ff fe 05
ptp4l[172]: 0040 de 06 00 01
ptp4l[172]: [6469.855] Unexpected data on socket err queue:
ptp4l[172]: 0000 01 80 c2 00 00 0e 00 1f 7b 63 02 48 88 f7 10 02
ptp4l[172]: 0010 00 2c 00 00 02 00 00 00 00 00 00 00 00 00 00 00
ptp4l[172]: 0020 00 00 00 1f 7b ff fe 63 02 48 00 01 c6 b2 00 fd
ptp4l[172]: 0030 00 00 00 00 00 00 00 00 00 00
ptp4l[172]: [6469.974] Unexpected data on socket err queue:
ptp4l[172]: 0000 01 80 c2 00 00 0e 00 1f 7b 63 02 48 88 f7 10 02
ptp4l[172]: 0010 00 2c 00 00 02 00 00 00 00 00 00 00 00 00 00 00
ptp4l[172]: 0020 00 00 00 1f 7b ff fe 63 02 48 00 03 aa 07 00 fd
ptp4l[172]: 0030 00 00 00 00 00 00 00 00 00 00

The ptp4l program itself is heavily patched to show this (more details
here [0]). Otherwise, by default it just hangs.

On the other hand, with the DSA patch to disallow HW timestamping
applied:

tcpdump -i eth2 -j adapter_unsynced
tcpdump: SIOCSHWTSTAMP failed: Device or resource busy

So it is a fact of life that PTP timestamping on the DSA master is
incompatible with timestamping on the switch MAC, at least with the
current API. And if the switch supports PTP, taking the timestamps from
the switch MAC is highly preferable anyway, due to the fact that those
don't contain the queuing latencies of the switch. So just disallow PTP
on the DSA master if there is any PTP-capable switch attached.

[0]: https://sourceforge.net/p/linuxptp/mailman/message/36880648/

Fixes: 0336369d3a4d ("net: dsa: forward hardware timestamping ioctls to switch driver")
Signed-off-by: Vladimir Oltean <olteanv@gmail.com>
Acked-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/dsa/master.c | 30 ++++++++++++++++++++++++++++++
 1 file changed, 30 insertions(+)

diff --git a/net/dsa/master.c b/net/dsa/master.c
index 3255dfc97f86..bd44bde272f4 100644
--- a/net/dsa/master.c
+++ b/net/dsa/master.c
@@ -197,6 +197,35 @@ static int dsa_master_get_phys_port_name(struct net_device *dev,
 	return 0;
 }
 
+static int dsa_master_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
+{
+	struct dsa_port *cpu_dp = dev->dsa_ptr;
+	struct dsa_switch *ds = cpu_dp->ds;
+	struct dsa_switch_tree *dst;
+	int err = -EOPNOTSUPP;
+	struct dsa_port *dp;
+
+	dst = ds->dst;
+
+	switch (cmd) {
+	case SIOCGHWTSTAMP:
+	case SIOCSHWTSTAMP:
+		/* Deny PTP operations on master if there is at least one
+		 * switch in the tree that is PTP capable.
+		 */
+		list_for_each_entry(dp, &dst->ports, list)
+			if (dp->ds->ops->port_hwtstamp_get ||
+			    dp->ds->ops->port_hwtstamp_set)
+				return -EBUSY;
+		break;
+	}
+
+	if (cpu_dp->orig_ndo_ops && cpu_dp->orig_ndo_ops->ndo_do_ioctl)
+		err = cpu_dp->orig_ndo_ops->ndo_do_ioctl(dev, ifr, cmd);
+
+	return err;
+}
+
 static int dsa_master_ethtool_setup(struct net_device *dev)
 {
 	struct dsa_port *cpu_dp = dev->dsa_ptr;
@@ -249,6 +278,7 @@ static int dsa_master_ndo_setup(struct net_device *dev)
 		memcpy(ops, cpu_dp->orig_ndo_ops, sizeof(*ops));
 
 	ops->ndo_get_phys_port_name = dsa_master_get_phys_port_name;
+	ops->ndo_do_ioctl = dsa_master_ioctl;
 
 	dev->netdev_ops  = ops;
 
-- 
cgit v1.2.3-59-g8ed1b


From f278b99ca6b2d91a5744588d81bae297179b0d1f Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Mon, 30 Dec 2019 06:06:19 -0800
Subject: tcp_cubic: refactor code to perform a divide only when needed

Neal Cardwell suggested to not change ca->delay_min
and apply the ack delay cushion only when Hystart ACK train
is still under consideration. This should avoid a 64bit
divide unless needed.

Tested:

40Gbit(mlx4) testbed (with sch_fq as packet scheduler)

$ echo -n 'file tcp_cubic.c +p'  >/sys/kernel/debug/dynamic_debug/control
$ nstat -n;for f in {1..10}; do ./super_netperf 1 -H lpaa24 -l -4000000; done;nstat|egrep "Hystart"
  14815
  16280
  15293
  15563
  11574
  15145
  14789
  18548
  16972
  12520
TcpExtTCPHystartTrainDetect     10                 0.0
TcpExtTCPHystartTrainCwnd       1396               0.0
$ dmesg | tail -10
[ 4873.951350] hystart_ack_train (116 > 93) delay_min 24 (+ ack_delay 69) cwnd 80
[ 4875.155379] hystart_ack_train (55 > 50) delay_min 21 (+ ack_delay 29) cwnd 160
[ 4876.333921] hystart_ack_train (69 > 62) delay_min 23 (+ ack_delay 39) cwnd 130
[ 4877.519037] hystart_ack_train (69 > 60) delay_min 22 (+ ack_delay 38) cwnd 130
[ 4878.701559] hystart_ack_train (87 > 63) delay_min 24 (+ ack_delay 39) cwnd 160
[ 4879.844597] hystart_ack_train (93 > 50) delay_min 21 (+ ack_delay 29) cwnd 216
[ 4880.956650] hystart_ack_train (74 > 67) delay_min 20 (+ ack_delay 47) cwnd 108
[ 4882.098500] hystart_ack_train (61 > 57) delay_min 23 (+ ack_delay 34) cwnd 130
[ 4883.262056] hystart_ack_train (72 > 67) delay_min 21 (+ ack_delay 46) cwnd 130
[ 4884.418760] hystart_ack_train (74 > 67) delay_min 29 (+ ack_delay 38) cwnd 152

10Gbit(bnx2x) testbed (with sch_fq as packet scheduler)

$ echo -n 'file tcp_cubic.c +p'  >/sys/kernel/debug/dynamic_debug/control
$ nstat -n;for f in {1..10}; do ./super_netperf 1 -H lpk52 -l -4000000; done;nstat|egrep "Hystart"
   7050
   7065
   7100
   6900
   7202
   7263
   7189
   6869
   7463
   7034
TcpExtTCPHystartTrainDetect     10                 0.0
TcpExtTCPHystartTrainCwnd       3199               0.0
$ dmesg | tail -10
[  176.920012] hystart_ack_train (161 > 141) delay_min 83 (+ ack_delay 58) cwnd 264
[  179.144645] hystart_ack_train (164 > 159) delay_min 120 (+ ack_delay 39) cwnd 444
[  181.354527] hystart_ack_train (214 > 168) delay_min 125 (+ ack_delay 43) cwnd 436
[  183.539565] hystart_ack_train (170 > 147) delay_min 96 (+ ack_delay 51) cwnd 326
[  185.727309] hystart_ack_train (177 > 160) delay_min 61 (+ ack_delay 99) cwnd 128
[  187.947142] hystart_ack_train (184 > 167) delay_min 123 (+ ack_delay 44) cwnd 367
[  190.166680] hystart_ack_train (230 > 153) delay_min 116 (+ ack_delay 37) cwnd 444
[  192.327285] hystart_ack_train (210 > 206) delay_min 86 (+ ack_delay 120) cwnd 152
[  194.511392] hystart_ack_train (173 > 151) delay_min 94 (+ ack_delay 57) cwnd 239
[  196.736023] hystart_ack_train (149 > 146) delay_min 105 (+ ack_delay 41) cwnd 399

Fixes: 42f3a8aaae66 ("tcp_cubic: tweak Hystart detection for short RTT flows")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Neal Cardwell <ncardwell@google.com>
Link: https://www.spinics.net/lists/netdev/msg621886.html
Link: https://www.spinics.net/lists/netdev/msg621797.html
Acked-by: Neal Cardwell <ncardwell@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/tcp_cubic.c | 51 ++++++++++++++++++++++++++++-----------------------
 1 file changed, 28 insertions(+), 23 deletions(-)

diff --git a/net/ipv4/tcp_cubic.c b/net/ipv4/tcp_cubic.c
index d02bb283c689..8f8eefd3a3ce 100644
--- a/net/ipv4/tcp_cubic.c
+++ b/net/ipv4/tcp_cubic.c
@@ -372,6 +372,26 @@ static void bictcp_state(struct sock *sk, u8 new_state)
 	}
 }
 
+/* Account for TSO/GRO delays.
+ * Otherwise short RTT flows could get too small ssthresh, since during
+ * slow start we begin with small TSO packets and ca->delay_min would
+ * not account for long aggregation delay when TSO packets get bigger.
+ * Ideally even with a very small RTT we would like to have at least one
+ * TSO packet being sent and received by GRO, and another one in qdisc layer.
+ * We apply another 100% factor because @rate is doubled at this point.
+ * We cap the cushion to 1ms.
+ */
+static u32 hystart_ack_delay(struct sock *sk)
+{
+	unsigned long rate;
+
+	rate = READ_ONCE(sk->sk_pacing_rate);
+	if (!rate)
+		return 0;
+	return min_t(u64, USEC_PER_MSEC,
+		     div64_ul((u64)GSO_MAX_SIZE * 4 * USEC_PER_SEC, rate));
+}
+
 static void hystart_update(struct sock *sk, u32 delay)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
@@ -385,7 +405,8 @@ static void hystart_update(struct sock *sk, u32 delay)
 		if ((s32)(now - ca->last_ack) <= hystart_ack_delta_us) {
 			ca->last_ack = now;
 
-			threshold = ca->delay_min;
+			threshold = ca->delay_min + hystart_ack_delay(sk);
+
 			/* Hystart ack train triggers if we get ack past
 			 * ca->delay_min/2.
 			 * Pacing might have delayed packets up to RTT/2
@@ -396,6 +417,9 @@ static void hystart_update(struct sock *sk, u32 delay)
 
 			if ((s32)(now - ca->round_start) > threshold) {
 				ca->found = 1;
+				pr_debug("hystart_ack_train (%u > %u) delay_min %u (+ ack_delay %u) cwnd %u\n",
+					 now - ca->round_start, threshold,
+					 ca->delay_min, hystart_ack_delay(sk), tp->snd_cwnd);
 				NET_INC_STATS(sock_net(sk),
 					      LINUX_MIB_TCPHYSTARTTRAINDETECT);
 				NET_ADD_STATS(sock_net(sk),
@@ -447,30 +471,11 @@ static void bictcp_acked(struct sock *sk, const struct ack_sample *sample)
 		delay = 1;
 
 	/* first time call or link delay decreases */
-	if (ca->delay_min == 0 || ca->delay_min > delay) {
-		unsigned long rate = READ_ONCE(sk->sk_pacing_rate);
-
-		/* Account for TSO/GRO delays.
-		 * Otherwise short RTT flows could get too small ssthresh,
-		 * since during slow start we begin with small TSO packets
-		 * and could lower ca->delay_min too much.
-		 * Ideally even with a very small RTT we would like to have
-		 * at least one TSO packet being sent and received by GRO,
-		 * and another one in qdisc layer.
-		 * We apply another 100% factor because @rate is doubled at
-		 * this point.
-		 * We cap the cushion to 1ms.
-		 */
-		if (rate)
-			delay += min_t(u64, USEC_PER_MSEC,
-				       div64_ul((u64)GSO_MAX_SIZE *
-						4 * USEC_PER_SEC, rate));
-		if (ca->delay_min == 0 || ca->delay_min > delay)
-			ca->delay_min = delay;
-	}
+	if (ca->delay_min == 0 || ca->delay_min > delay)
+		ca->delay_min = delay;
 
 	/* hystart triggers when cwnd is larger than some threshold */
-	if (!ca->found && hystart && tcp_in_slow_start(tp) &&
+	if (!ca->found && tcp_in_slow_start(tp) && hystart &&
 	    tp->snd_cwnd >= hystart_low_window)
 		hystart_update(sk, delay);
 }
-- 
cgit v1.2.3-59-g8ed1b


From db99d5126fb9067c3123b40a33c9936235f059fa Mon Sep 17 00:00:00 2001
From: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Date: Thu, 26 Dec 2019 16:02:24 +0100
Subject: sfc: avoid duplicate error handling code in
 'efx_ef10_sriov_set_vf_mac()'

'eth_zero_addr()' is already called in the error handling path. This is
harmless, but there is no point in calling it twice, so remove one.

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/ef10_sriov.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/sfc/ef10_sriov.c b/drivers/net/ethernet/sfc/ef10_sriov.c
index 52bd43f45761..14393767ef9f 100644
--- a/drivers/net/ethernet/sfc/ef10_sriov.c
+++ b/drivers/net/ethernet/sfc/ef10_sriov.c
@@ -522,10 +522,9 @@ int efx_ef10_sriov_set_vf_mac(struct efx_nic *efx, int vf_i, u8 *mac)
 
 	if (!is_zero_ether_addr(mac)) {
 		rc = efx_ef10_vport_add_mac(efx, vf->vport_id, mac);
-		if (rc) {
-			eth_zero_addr(vf->mac);
+		if (rc)
 			goto fail;
-		}
+
 		if (vf->efx)
 			ether_addr_copy(vf->efx->net_dev->dev_addr, mac);
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 544fed47af4d2174ac0b550e9c8da15c2dfdb117 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <olteanv@gmail.com>
Date: Fri, 27 Dec 2019 15:02:28 +0200
Subject: ptp: introduce ptp_cancel_worker_sync

In order to effectively use the PTP kernel thread for tasks such as
timestamping packets, allow the user control over stopping it, which is
needed e.g. when the timestamping queues must be drained.

Signed-off-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/ptp/ptp_clock.c          | 6 ++++++
 include/linux/ptp_clock_kernel.h | 9 +++++++++
 2 files changed, 15 insertions(+)

diff --git a/drivers/ptp/ptp_clock.c b/drivers/ptp/ptp_clock.c
index e60eab7f8a61..4f0d91a76dcb 100644
--- a/drivers/ptp/ptp_clock.c
+++ b/drivers/ptp/ptp_clock.c
@@ -371,6 +371,12 @@ int ptp_schedule_worker(struct ptp_clock *ptp, unsigned long delay)
 }
 EXPORT_SYMBOL(ptp_schedule_worker);
 
+void ptp_cancel_worker_sync(struct ptp_clock *ptp)
+{
+	kthread_cancel_delayed_work_sync(&ptp->aux_work);
+}
+EXPORT_SYMBOL(ptp_cancel_worker_sync);
+
 /* module operations */
 
 static void __exit ptp_exit(void)
diff --git a/include/linux/ptp_clock_kernel.h b/include/linux/ptp_clock_kernel.h
index 93cc4f1d444a..c64a1ef87240 100644
--- a/include/linux/ptp_clock_kernel.h
+++ b/include/linux/ptp_clock_kernel.h
@@ -243,6 +243,13 @@ int ptp_find_pin(struct ptp_clock *ptp,
 
 int ptp_schedule_worker(struct ptp_clock *ptp, unsigned long delay);
 
+/**
+ * ptp_cancel_worker_sync() - cancel ptp auxiliary clock
+ *
+ * @ptp:     The clock obtained from ptp_clock_register().
+ */
+void ptp_cancel_worker_sync(struct ptp_clock *ptp);
+
 #else
 static inline struct ptp_clock *ptp_clock_register(struct ptp_clock_info *info,
 						   struct device *parent)
@@ -260,6 +267,8 @@ static inline int ptp_find_pin(struct ptp_clock *ptp,
 static inline int ptp_schedule_worker(struct ptp_clock *ptp,
 				      unsigned long delay)
 { return -EOPNOTSUPP; }
+static inline void ptp_cancel_worker_sync(struct ptp_clock *ptp)
+{ }
 
 #endif
 
-- 
cgit v1.2.3-59-g8ed1b


From 1e762bd278d2a70bc74b9cbee7f1e93bd4704fe2 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <olteanv@gmail.com>
Date: Fri, 27 Dec 2019 15:02:29 +0200
Subject: net: dsa: sja1105: Use PTP core's dedicated kernel thread for RX
 timestamping

And move the queue of skb's waiting for RX timestamps into the ptp_data
structure, since it isn't needed if PTP is not compiled.

Signed-off-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/sja1105/sja1105_ptp.c | 33 +++++++++++++++------------------
 drivers/net/dsa/sja1105/sja1105_ptp.h |  1 +
 include/linux/dsa/sja1105.h           |  2 --
 3 files changed, 16 insertions(+), 20 deletions(-)

diff --git a/drivers/net/dsa/sja1105/sja1105_ptp.c b/drivers/net/dsa/sja1105/sja1105_ptp.c
index 54258a25031d..387b22a86a53 100644
--- a/drivers/net/dsa/sja1105/sja1105_ptp.c
+++ b/drivers/net/dsa/sja1105/sja1105_ptp.c
@@ -367,22 +367,16 @@ static int sja1105_ptpclkval_write(struct sja1105_private *priv, u64 ticks,
 				ptp_sts);
 }
 
-#define rxtstamp_to_tagger(d) \
-	container_of((d), struct sja1105_tagger_data, rxtstamp_work)
-#define tagger_to_sja1105(d) \
-	container_of((d), struct sja1105_private, tagger_data)
-
-static void sja1105_rxtstamp_work(struct work_struct *work)
+static long sja1105_rxtstamp_work(struct ptp_clock_info *ptp)
 {
-	struct sja1105_tagger_data *tagger_data = rxtstamp_to_tagger(work);
-	struct sja1105_private *priv = tagger_to_sja1105(tagger_data);
-	struct sja1105_ptp_data *ptp_data = &priv->ptp_data;
+	struct sja1105_ptp_data *ptp_data = ptp_caps_to_data(ptp);
+	struct sja1105_private *priv = ptp_data_to_sja1105(ptp_data);
 	struct dsa_switch *ds = priv->ds;
 	struct sk_buff *skb;
 
 	mutex_lock(&ptp_data->lock);
 
-	while ((skb = skb_dequeue(&tagger_data->skb_rxtstamp_queue)) != NULL) {
+	while ((skb = skb_dequeue(&ptp_data->skb_rxtstamp_queue)) != NULL) {
 		struct skb_shared_hwtstamps *shwt = skb_hwtstamps(skb);
 		u64 ticks, ts;
 		int rc;
@@ -404,6 +398,9 @@ static void sja1105_rxtstamp_work(struct work_struct *work)
 	}
 
 	mutex_unlock(&ptp_data->lock);
+
+	/* Don't restart */
+	return -1;
 }
 
 /* Called from dsa_skb_defer_rx_timestamp */
@@ -411,16 +408,16 @@ bool sja1105_port_rxtstamp(struct dsa_switch *ds, int port,
 			   struct sk_buff *skb, unsigned int type)
 {
 	struct sja1105_private *priv = ds->priv;
-	struct sja1105_tagger_data *tagger_data = &priv->tagger_data;
+	struct sja1105_ptp_data *ptp_data = &priv->ptp_data;
 
-	if (!test_bit(SJA1105_HWTS_RX_EN, &tagger_data->state))
+	if (!test_bit(SJA1105_HWTS_RX_EN, &priv->tagger_data.state))
 		return false;
 
 	/* We need to read the full PTP clock to reconstruct the Rx
 	 * timestamp. For that we need a sleepable context.
 	 */
-	skb_queue_tail(&tagger_data->skb_rxtstamp_queue, skb);
-	schedule_work(&tagger_data->rxtstamp_work);
+	skb_queue_tail(&ptp_data->skb_rxtstamp_queue, skb);
+	ptp_schedule_worker(ptp_data->clock, 0);
 	return true;
 }
 
@@ -628,11 +625,11 @@ int sja1105_ptp_clock_register(struct dsa_switch *ds)
 		.adjtime	= sja1105_ptp_adjtime,
 		.gettimex64	= sja1105_ptp_gettimex,
 		.settime64	= sja1105_ptp_settime,
+		.do_aux_work	= sja1105_rxtstamp_work,
 		.max_adj	= SJA1105_MAX_ADJ_PPB,
 	};
 
-	skb_queue_head_init(&tagger_data->skb_rxtstamp_queue);
-	INIT_WORK(&tagger_data->rxtstamp_work, sja1105_rxtstamp_work);
+	skb_queue_head_init(&ptp_data->skb_rxtstamp_queue);
 	spin_lock_init(&tagger_data->meta_lock);
 
 	ptp_data->clock = ptp_clock_register(&ptp_data->caps, ds->dev);
@@ -653,8 +650,8 @@ void sja1105_ptp_clock_unregister(struct dsa_switch *ds)
 	if (IS_ERR_OR_NULL(ptp_data->clock))
 		return;
 
-	cancel_work_sync(&priv->tagger_data.rxtstamp_work);
-	skb_queue_purge(&priv->tagger_data.skb_rxtstamp_queue);
+	ptp_cancel_worker_sync(ptp_data->clock);
+	skb_queue_purge(&ptp_data->skb_rxtstamp_queue);
 	ptp_clock_unregister(ptp_data->clock);
 	ptp_data->clock = NULL;
 }
diff --git a/drivers/net/dsa/sja1105/sja1105_ptp.h b/drivers/net/dsa/sja1105/sja1105_ptp.h
index 470f44b76318..6f4a19eec709 100644
--- a/drivers/net/dsa/sja1105/sja1105_ptp.h
+++ b/drivers/net/dsa/sja1105/sja1105_ptp.h
@@ -30,6 +30,7 @@ struct sja1105_ptp_cmd {
 };
 
 struct sja1105_ptp_data {
+	struct sk_buff_head skb_rxtstamp_queue;
 	struct ptp_clock_info caps;
 	struct ptp_clock *clock;
 	struct sja1105_ptp_cmd cmd;
diff --git a/include/linux/dsa/sja1105.h b/include/linux/dsa/sja1105.h
index 897e799dbcb9..c0b6a603ea8c 100644
--- a/include/linux/dsa/sja1105.h
+++ b/include/linux/dsa/sja1105.h
@@ -37,8 +37,6 @@
  * the structure defined in struct sja1105_private.
  */
 struct sja1105_tagger_data {
-	struct sk_buff_head skb_rxtstamp_queue;
-	struct work_struct rxtstamp_work;
 	struct sk_buff *stampable_skb;
 	/* Protects concurrent access to the meta state machine
 	 * from taggers running on multiple ports on SMP systems
-- 
cgit v1.2.3-59-g8ed1b


From 19d1f0ed74b75279f46549de80fa2adf80ef9995 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <olteanv@gmail.com>
Date: Fri, 27 Dec 2019 15:02:30 +0200
Subject: net: dsa: sja1105: Empty the RX timestamping queue on PTP settings
 change

When disabling PTP timestamping, don't reset the switch with the new
static config until all existing PTP frames have been timestamped on the
RX path or dropped. There's nothing we can do with these afterwards.

Signed-off-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/sja1105/sja1105_ptp.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/dsa/sja1105/sja1105_ptp.c b/drivers/net/dsa/sja1105/sja1105_ptp.c
index 387b22a86a53..e6b611470043 100644
--- a/drivers/net/dsa/sja1105/sja1105_ptp.c
+++ b/drivers/net/dsa/sja1105/sja1105_ptp.c
@@ -83,6 +83,7 @@ static int sja1105_init_avb_params(struct sja1105_private *priv,
 static int sja1105_change_rxtstamping(struct sja1105_private *priv,
 				      bool on)
 {
+	struct sja1105_ptp_data *ptp_data = &priv->ptp_data;
 	struct sja1105_general_params_entry *general_params;
 	struct sja1105_table *table;
 	int rc;
@@ -101,6 +102,8 @@ static int sja1105_change_rxtstamping(struct sja1105_private *priv,
 		kfree_skb(priv->tagger_data.stampable_skb);
 		priv->tagger_data.stampable_skb = NULL;
 	}
+	ptp_cancel_worker_sync(ptp_data->clock);
+	skb_queue_purge(&ptp_data->skb_rxtstamp_queue);
 
 	return sja1105_static_config_reload(priv, SJA1105_RX_HWTSTAMPING);
 }
-- 
cgit v1.2.3-59-g8ed1b


From f398efc14a9277b55defff71f59a46fdf13c713f Mon Sep 17 00:00:00 2001
From: Kevin Kou <qdkevin.kou@gmail.com>
Date: Fri, 27 Dec 2019 13:11:16 +0000
Subject: sctp: add enabled check for path tracepoint loop.

sctp_outq_sack is the main function handles SACK, it is called very
frequently. As the commit "move trace_sctp_probe_path into sctp_outq_sack"
added below code to this function, sctp tracepoint is disabled most of time,
but the loop of transport list will be always called even though the
tracepoint is disabled, this is unnecessary.

+	/* SCTP path tracepoint for congestion control debugging. */
+	list_for_each_entry(transport, transport_list, transports) {
+		trace_sctp_probe_path(transport, asoc);
+	}

This patch is to add tracepoint enabled check at outside of the loop of
transport list, and avoid traversing the loop when trace is disabled,
it is a small optimization.

Signed-off-by: Kevin Kou <qdkevin.kou@gmail.com>
Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sctp/outqueue.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c
index 6b0b3bad4daa..577e3bc4ee6f 100644
--- a/net/sctp/outqueue.c
+++ b/net/sctp/outqueue.c
@@ -1240,8 +1240,9 @@ int sctp_outq_sack(struct sctp_outq *q, struct sctp_chunk *chunk)
 	transport_list = &asoc->peer.transport_addr_list;
 
 	/* SCTP path tracepoint for congestion control debugging. */
-	list_for_each_entry(transport, transport_list, transports) {
-		trace_sctp_probe_path(transport, asoc);
+	if (trace_sctp_probe_path_enabled()) {
+		list_for_each_entry(transport, transport_list, transports)
+			trace_sctp_probe_path(transport, asoc);
 	}
 
 	sack_ctsn = ntohl(sack->cum_tsn_ack);
-- 
cgit v1.2.3-59-g8ed1b


From 9e860947d8d7a1504476ac49abfce90a4ce600f3 Mon Sep 17 00:00:00 2001
From: Vijay Khemka <vijaykhemka@fb.com>
Date: Fri, 27 Dec 2019 14:43:49 -0800
Subject: net/ncsi: Fix gma flag setting after response

gma_flag was set at the time of GMA command request but it should
only be set after getting successful response. Movinng this flag
setting in GMA response handler.

This flag is used mainly for not repeating GMA command once
received MAC address.

Signed-off-by: Vijay Khemka <vijaykhemka@fb.com>
Reviewed-by: Samuel Mendoza-Jonas <sam@mendozajonas.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ncsi/ncsi-manage.c | 3 ---
 net/ncsi/ncsi-rsp.c    | 6 ++++++
 2 files changed, 6 insertions(+), 3 deletions(-)

diff --git a/net/ncsi/ncsi-manage.c b/net/ncsi/ncsi-manage.c
index 70fe02697544..e20b81514029 100644
--- a/net/ncsi/ncsi-manage.c
+++ b/net/ncsi/ncsi-manage.c
@@ -764,9 +764,6 @@ static int ncsi_gma_handler(struct ncsi_cmd_arg *nca, unsigned int mf_id)
 		return -1;
 	}
 
-	/* Set the flag for GMA command which should only be called once */
-	nca->ndp->gma_flag = 1;
-
 	/* Get Mac address from NCSI device */
 	return nch->handler(nca);
 }
diff --git a/net/ncsi/ncsi-rsp.c b/net/ncsi/ncsi-rsp.c
index d5611f04926d..a94bb59793f0 100644
--- a/net/ncsi/ncsi-rsp.c
+++ b/net/ncsi/ncsi-rsp.c
@@ -627,6 +627,9 @@ static int ncsi_rsp_handler_oem_mlx_gma(struct ncsi_request *nr)
 	saddr.sa_family = ndev->type;
 	ndev->priv_flags |= IFF_LIVE_ADDR_CHANGE;
 	memcpy(saddr.sa_data, &rsp->data[MLX_MAC_ADDR_OFFSET], ETH_ALEN);
+	/* Set the flag for GMA command which should only be called once */
+	ndp->gma_flag = 1;
+
 	ret = ops->ndo_set_mac_address(ndev, &saddr);
 	if (ret < 0)
 		netdev_warn(ndev, "NCSI: 'Writing mac address to device failed\n");
@@ -671,6 +674,9 @@ static int ncsi_rsp_handler_oem_bcm_gma(struct ncsi_request *nr)
 	if (!is_valid_ether_addr((const u8 *)saddr.sa_data))
 		return -ENXIO;
 
+	/* Set the flag for GMA command which should only be called once */
+	ndp->gma_flag = 1;
+
 	ret = ops->ndo_set_mac_address(ndev, &saddr);
 	if (ret < 0)
 		netdev_warn(ndev, "NCSI: 'Writing mac address to device failed\n");
-- 
cgit v1.2.3-59-g8ed1b


From 6506f52dcbf8c2afca47bd7a48c18459e1f04f86 Mon Sep 17 00:00:00 2001
From: Robert Beckett <bob.beckett@collabora.com>
Date: Tue, 22 Oct 2019 16:31:41 +0100
Subject: igb: dont drop packets if rx flow control is enabled

If Rx flow control has been enabled (via autoneg or forced), packets
should not be dropped due to Rx descriptor ring exhaustion. Instead
pause frames should be used to apply back pressure. This only applies
if VFs are not in use.

Move SRRCTL setup to its own function for easy reuse and only set drop
enable bit if Rx flow control is not enabled.

Since v1: always enable dropping of packets if VFs in use.

Signed-off-by: Robert Beckett <bob.beckett@collabora.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igb/igb.h         |  1 +
 drivers/net/ethernet/intel/igb/igb_ethtool.c |  8 +++++
 drivers/net/ethernet/intel/igb/igb_main.c    | 47 +++++++++++++++++++---------
 3 files changed, 42 insertions(+), 14 deletions(-)

diff --git a/drivers/net/ethernet/intel/igb/igb.h b/drivers/net/ethernet/intel/igb/igb.h
index ca54e268d157..49b5fa9d4783 100644
--- a/drivers/net/ethernet/intel/igb/igb.h
+++ b/drivers/net/ethernet/intel/igb/igb.h
@@ -661,6 +661,7 @@ void igb_configure_tx_ring(struct igb_adapter *, struct igb_ring *);
 void igb_configure_rx_ring(struct igb_adapter *, struct igb_ring *);
 void igb_setup_tctl(struct igb_adapter *);
 void igb_setup_rctl(struct igb_adapter *);
+void igb_setup_srrctl(struct igb_adapter *, struct igb_ring *);
 netdev_tx_t igb_xmit_frame_ring(struct sk_buff *, struct igb_ring *);
 void igb_alloc_rx_buffers(struct igb_ring *, u16);
 void igb_update_stats(struct igb_adapter *);
diff --git a/drivers/net/ethernet/intel/igb/igb_ethtool.c b/drivers/net/ethernet/intel/igb/igb_ethtool.c
index 4690d6c87f39..43c438365389 100644
--- a/drivers/net/ethernet/intel/igb/igb_ethtool.c
+++ b/drivers/net/ethernet/intel/igb/igb_ethtool.c
@@ -396,6 +396,7 @@ static int igb_set_pauseparam(struct net_device *netdev,
 	struct igb_adapter *adapter = netdev_priv(netdev);
 	struct e1000_hw *hw = &adapter->hw;
 	int retval = 0;
+	int i;
 
 	/* 100basefx does not support setting link flow control */
 	if (hw->dev_spec._82575.eth_flags.e100_base_fx)
@@ -428,6 +429,13 @@ static int igb_set_pauseparam(struct net_device *netdev,
 
 		retval = ((hw->phy.media_type == e1000_media_type_copper) ?
 			  igb_force_mac_fc(hw) : igb_setup_link(hw));
+
+		/* Make sure SRRCTL considers new fc settings for each ring */
+		for (i = 0; i < adapter->num_rx_queues; i++) {
+			struct igb_ring *ring = adapter->rx_ring[i];
+
+			igb_setup_srrctl(adapter, ring);
+		}
 	}
 
 	clear_bit(__IGB_RESETTING, &adapter->state);
diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index d11e64a58ed1..b46bff8fe056 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -4467,6 +4467,37 @@ static inline void igb_set_vmolr(struct igb_adapter *adapter,
 	wr32(E1000_VMOLR(vfn), vmolr);
 }
 
+/**
+ *  igb_setup_srrctl - configure the split and replication receive control
+ *                     registers
+ *  @adapter: Board private structure
+ *  @ring: receive ring to be configured
+ **/
+void igb_setup_srrctl(struct igb_adapter *adapter, struct igb_ring *ring)
+{
+	struct e1000_hw *hw = &adapter->hw;
+	int reg_idx = ring->reg_idx;
+	u32 srrctl = 0;
+
+	srrctl = IGB_RX_HDR_LEN << E1000_SRRCTL_BSIZEHDRSIZE_SHIFT;
+	if (ring_uses_large_buffer(ring))
+		srrctl |= IGB_RXBUFFER_3072 >> E1000_SRRCTL_BSIZEPKT_SHIFT;
+	else
+		srrctl |= IGB_RXBUFFER_2048 >> E1000_SRRCTL_BSIZEPKT_SHIFT;
+	srrctl |= E1000_SRRCTL_DESCTYPE_ADV_ONEBUF;
+	if (hw->mac.type >= e1000_82580)
+		srrctl |= E1000_SRRCTL_TIMESTAMP;
+	/* Only set Drop Enable if VFs allocated, or we are supporting multiple
+	 * queues and rx flow control is disabled
+	 */
+	if (adapter->vfs_allocated_count ||
+	    (!(hw->fc.current_mode & e1000_fc_rx_pause) &&
+	     adapter->num_rx_queues > 1))
+		srrctl |= E1000_SRRCTL_DROP_EN;
+
+	wr32(E1000_SRRCTL(reg_idx), srrctl);
+}
+
 /**
  *  igb_configure_rx_ring - Configure a receive ring after Reset
  *  @adapter: board private structure
@@ -4481,7 +4512,7 @@ void igb_configure_rx_ring(struct igb_adapter *adapter,
 	union e1000_adv_rx_desc *rx_desc;
 	u64 rdba = ring->dma;
 	int reg_idx = ring->reg_idx;
-	u32 srrctl = 0, rxdctl = 0;
+	u32 rxdctl = 0;
 
 	/* disable the queue */
 	wr32(E1000_RXDCTL(reg_idx), 0);
@@ -4499,19 +4530,7 @@ void igb_configure_rx_ring(struct igb_adapter *adapter,
 	writel(0, ring->tail);
 
 	/* set descriptor configuration */
-	srrctl = IGB_RX_HDR_LEN << E1000_SRRCTL_BSIZEHDRSIZE_SHIFT;
-	if (ring_uses_large_buffer(ring))
-		srrctl |= IGB_RXBUFFER_3072 >> E1000_SRRCTL_BSIZEPKT_SHIFT;
-	else
-		srrctl |= IGB_RXBUFFER_2048 >> E1000_SRRCTL_BSIZEPKT_SHIFT;
-	srrctl |= E1000_SRRCTL_DESCTYPE_ADV_ONEBUF;
-	if (hw->mac.type >= e1000_82580)
-		srrctl |= E1000_SRRCTL_TIMESTAMP;
-	/* Only set Drop Enable if we are supporting multiple queues */
-	if (adapter->vfs_allocated_count || adapter->num_rx_queues > 1)
-		srrctl |= E1000_SRRCTL_DROP_EN;
-
-	wr32(E1000_SRRCTL(reg_idx), srrctl);
+	igb_setup_srrctl(adapter, ring);
 
 	/* set filtering for VMDQ pools */
 	igb_set_vmolr(adapter, reg_idx & 0x7, true);
-- 
cgit v1.2.3-59-g8ed1b


From b7b462454a5c2aed1d8aecd165d1dae9017bd5fd Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Thu, 31 Oct 2019 08:06:07 +0200
Subject: igc: Add scatter gather support

Scatter gather is used to do DMA data transfers of data that is written to
noncontiguous areas of memory.
This patch enables scatter gather support.

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_main.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 9700527dd797..75dcaa87d06b 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -4433,6 +4433,7 @@ static int igc_probe(struct pci_dev *pdev,
 		goto err_sw_init;
 
 	/* Add supported features to the features list*/
+	netdev->features |= NETIF_F_SG;
 	netdev->features |= NETIF_F_RXCSUM;
 	netdev->features |= NETIF_F_HW_CSUM;
 	netdev->features |= NETIF_F_SCTP_CRC;
-- 
cgit v1.2.3-59-g8ed1b


From 21da01fd3bb1124b46e6b15657de36dffcc58af9 Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Tue, 5 Nov 2019 11:44:13 +0200
Subject: igc: Improve the DMA mapping flow

Improve the probe flow and set both the DMA mask and the coherent
to the same thing. Make the flow optimized and cleared.

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_main.c | 24 +++++++++---------------
 1 file changed, 9 insertions(+), 15 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 75dcaa87d06b..91acce2e1d1c 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -4345,32 +4345,26 @@ static int igc_probe(struct pci_dev *pdev,
 	struct net_device *netdev;
 	struct igc_hw *hw;
 	const struct igc_info *ei = igc_info_tbl[ent->driver_data];
-	int err;
+	int err, pci_using_dac;
 
 	err = pci_enable_device_mem(pdev);
 	if (err)
 		return err;
 
-	err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(64));
+	pci_using_dac = 0;
+	err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
 	if (!err) {
-		err = dma_set_coherent_mask(&pdev->dev,
-					    DMA_BIT_MASK(64));
+		pci_using_dac = 1;
 	} else {
-		err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(32));
+		err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
 		if (err) {
-			err = dma_set_coherent_mask(&pdev->dev,
-						    DMA_BIT_MASK(32));
-			if (err) {
-				dev_err(&pdev->dev, "igc: Wrong DMA config\n");
-				goto err_dma;
-			}
+			dev_err(&pdev->dev,
+				"No usable DMA configuration, aborting\n");
+			goto err_dma;
 		}
 	}
 
-	err = pci_request_selected_regions(pdev,
-					   pci_select_bars(pdev,
-							   IORESOURCE_MEM),
-					   igc_driver_name);
+	err = pci_request_mem_regions(pdev, igc_driver_name);
 	if (err)
 		goto err_pci_reg;
 
-- 
cgit v1.2.3-59-g8ed1b


From faf4dd52e9e34a36254a6ad43369064b6928d504 Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Sun, 10 Nov 2019 16:05:20 +0200
Subject: igc: Prefer to use the pci_release_mem_regions method

Use the pci_release_mem_regions method instead of the
pci_release_selected_regions method

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_main.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 91acce2e1d1c..de8e4b570fa6 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -4527,8 +4527,7 @@ err_sw_init:
 err_ioremap:
 	free_netdev(netdev);
 err_alloc_etherdev:
-	pci_release_selected_regions(pdev,
-				     pci_select_bars(pdev, IORESOURCE_MEM));
+	pci_release_mem_regions(pdev);
 err_pci_reg:
 err_dma:
 	pci_disable_device(pdev);
-- 
cgit v1.2.3-59-g8ed1b


From 57cd472c2baa1d1dc7f037d2c6d6d7d29be29d5d Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Sun, 10 Nov 2019 17:57:44 +0200
Subject: igc: Remove excess parameter description from igc_is_non_eop

The function description for igc_is_non_eop includes an extra @skb
parameter description. This parameter doesn't exist on the function, so
remove it.

Suggested-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_main.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index de8e4b570fa6..ac420a9b4d2b 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -1485,7 +1485,6 @@ static bool igc_can_reuse_rx_page(struct igc_rx_buffer *rx_buffer)
  * igc_is_non_eop - process handling of non-EOP buffers
  * @rx_ring: Rx ring being processed
  * @rx_desc: Rx descriptor for current buffer
- * @skb: current socket buffer containing buffer in progress
  *
  * This function updates next to clean.  If the buffer is an EOP buffer
  * this function exits returning false, otherwise it will place the
-- 
cgit v1.2.3-59-g8ed1b


From 085c858950419f5a7eea351c2e5cb999b4caad2d Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Mon, 11 Nov 2019 09:04:12 +0200
Subject: igc: Fix the parameter description for igc_alloc_rx_buffers

The function description for igc_alloc_rx_buffers has not reflected
the function meaning. Add meaningful description.

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_main.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index ac420a9b4d2b..833770fd16d2 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -1566,7 +1566,8 @@ static void igc_put_rx_buffer(struct igc_ring *rx_ring,
 
 /**
  * igc_alloc_rx_buffers - Replace used receive buffers; packet split
- * @adapter: address of board private structure
+ * @rx_ring: rx descriptor ring
+ * @cleaned_count: number of buffers to clean
  */
 static void igc_alloc_rx_buffers(struct igc_ring *rx_ring, u16 cleaned_count)
 {
-- 
cgit v1.2.3-59-g8ed1b


From 86efeccd5a7adcfbe12c0ba558615e53a33b67fc Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Tue, 12 Nov 2019 17:13:32 +0200
Subject: igc: Fix parameter descriptions for a several functions

igc_watchdog, igc_set_interrupt_capability, igc_init_interrupt_scheme,
__igc_open and __igc_close parameter descriptions has not reflected
functions meaning. Add meaningful description.

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_main.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 833770fd16d2..d3a45d3cbcd9 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -3109,7 +3109,7 @@ bool igc_has_link(struct igc_adapter *adapter)
 
 /**
  * igc_watchdog - Timer Call-back
- * @data: pointer to adapter cast into an unsigned long
+ * @t: timer for the watchdog
  */
 static void igc_watchdog(struct timer_list *t)
 {
@@ -3638,6 +3638,7 @@ static int igc_poll(struct napi_struct *napi, int budget)
 /**
  * igc_set_interrupt_capability - set MSI or MSI-X if supported
  * @adapter: Pointer to adapter structure
+ * @msix: boolean value for MSI-X capability
  *
  * Attempt to configure interrupts using the best available
  * capabilities of the hardware and kernel.
@@ -3906,6 +3907,7 @@ static void igc_cache_ring_register(struct igc_adapter *adapter)
 /**
  * igc_init_interrupt_scheme - initialize interrupts, allocate queues/vectors
  * @adapter: Pointer to adapter structure
+ * @msix: boolean for MSI-X capability
  *
  * This function initializes the interrupts and allocates all of the queues.
  */
@@ -4073,8 +4075,9 @@ static void igc_write_itr(struct igc_q_vector *q_vector)
 }
 
 /**
- * igc_open - Called when a network interface is made active
+ * __igc_open - Called when a network interface is made active
  * @netdev: network interface device structure
+ * @resuming: boolean indicating if the device is resuming
  *
  * Returns 0 on success, negative value on failure
  *
@@ -4164,8 +4167,9 @@ static int igc_open(struct net_device *netdev)
 }
 
 /**
- * igc_close - Disables a network interface
+ * __igc_close - Disables a network interface
  * @netdev: network interface device structure
+ * @suspending: boolean indicating the device is suspending
  *
  * Returns 0, this is not allowed to fail
  *
-- 
cgit v1.2.3-59-g8ed1b


From 4439dc427d83c67aaa8447091c479c482ff85af9 Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Wed, 13 Nov 2019 11:27:29 +0200
Subject: igc: Add 64 bit DMA access support

On relevant platforms ndo_start_xmit can handle socket buffer
fragments in high memory

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_main.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index d3a45d3cbcd9..b85009837cdf 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -4445,6 +4445,9 @@ static int igc_probe(struct pci_dev *pdev,
 	netdev->hw_features |= NETIF_F_NTUPLE;
 	netdev->hw_features |= netdev->features;
 
+	if (pci_using_dac)
+		netdev->features |= NETIF_F_HIGHDMA;
+
 	/* MTU range: 68 - 9216 */
 	netdev->min_mtu = ETH_MIN_MTU;
 	netdev->max_mtu = MAX_STD_JUMBO_FRAME_SIZE;
-- 
cgit v1.2.3-59-g8ed1b


From 9513d2a5dc7f3f2c037bfd2ea35264cc783b954e Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Thu, 14 Nov 2019 09:54:46 +0200
Subject: igc: Add legacy power management support

Add suspend, resume, runtime_suspend, runtime_resume and
runtime_idle callbacks implementation.

Reported-by: kbuild test robot <lpk@intel.com>
Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc.h         |   2 +
 drivers/net/ethernet/intel/igc/igc_defines.h |  31 ++++
 drivers/net/ethernet/intel/igc/igc_main.c    | 204 +++++++++++++++++++++++++++
 drivers/net/ethernet/intel/igc/igc_regs.h    |   9 ++
 4 files changed, 246 insertions(+)

diff --git a/drivers/net/ethernet/intel/igc/igc.h b/drivers/net/ethernet/intel/igc/igc.h
index 0868677d43ed..612fe9ec81a4 100644
--- a/drivers/net/ethernet/intel/igc/igc.h
+++ b/drivers/net/ethernet/intel/igc/igc.h
@@ -370,6 +370,8 @@ struct igc_adapter {
 	struct timer_list dma_err_timer;
 	struct timer_list phy_info_timer;
 
+	u32 wol;
+	u32 en_mng_pt;
 	u16 link_speed;
 	u16 link_duplex;
 
diff --git a/drivers/net/ethernet/intel/igc/igc_defines.h b/drivers/net/ethernet/intel/igc/igc_defines.h
index f3788f0b95b4..50dffd5db606 100644
--- a/drivers/net/ethernet/intel/igc/igc_defines.h
+++ b/drivers/net/ethernet/intel/igc/igc_defines.h
@@ -10,6 +10,37 @@
 
 #define IGC_CTRL_EXT_DRV_LOAD	0x10000000 /* Drv loaded bit for FW */
 
+/* Definitions for power management and wakeup registers */
+/* Wake Up Control */
+#define IGC_WUC_PME_EN	0x00000002 /* PME Enable */
+
+/* Wake Up Filter Control */
+#define IGC_WUFC_LNKC	0x00000001 /* Link Status Change Wakeup Enable */
+#define IGC_WUFC_MC	0x00000008 /* Directed Multicast Wakeup Enable */
+
+#define IGC_CTRL_ADVD3WUC	0x00100000  /* D3 WUC */
+
+/* Wake Up Status */
+#define IGC_WUS_EX	0x00000004 /* Directed Exact */
+#define IGC_WUS_ARPD	0x00000020 /* Directed ARP Request */
+#define IGC_WUS_IPV4	0x00000040 /* Directed IPv4 */
+#define IGC_WUS_IPV6	0x00000080 /* Directed IPv6 */
+#define IGC_WUS_NSD	0x00000400 /* Directed IPv6 Neighbor Solicitation */
+
+/* Packet types that are enabled for wake packet delivery */
+#define WAKE_PKT_WUS ( \
+	IGC_WUS_EX   | \
+	IGC_WUS_ARPD | \
+	IGC_WUS_IPV4 | \
+	IGC_WUS_IPV6 | \
+	IGC_WUS_NSD)
+
+/* Wake Up Packet Length */
+#define IGC_WUPL_MASK	0x00000FFF
+
+/* Wake Up Packet Memory stores the first 128 bytes of the wake up packet */
+#define IGC_WUPM_BYTES	128
+
 /* Physical Func Reset Done Indication */
 #define IGC_CTRL_EXT_LINK_MODE_MASK	0x00C00000
 
diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index b85009837cdf..ca7b8d6791f1 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -8,6 +8,7 @@
 #include <linux/tcp.h>
 #include <linux/udp.h>
 #include <linux/ip.h>
+#include <linux/pm_runtime.h>
 
 #include <net/ipv6.h>
 
@@ -4581,11 +4582,214 @@ static void igc_remove(struct pci_dev *pdev)
 	pci_disable_device(pdev);
 }
 
+static int __igc_shutdown(struct pci_dev *pdev, bool *enable_wake,
+			  bool runtime)
+{
+	struct net_device *netdev = pci_get_drvdata(pdev);
+	struct igc_adapter *adapter = netdev_priv(netdev);
+	u32 wufc = runtime ? IGC_WUFC_LNKC : adapter->wol;
+	struct igc_hw *hw = &adapter->hw;
+	u32 ctrl, rctl, status;
+	bool wake;
+
+	rtnl_lock();
+	netif_device_detach(netdev);
+
+	if (netif_running(netdev))
+		__igc_close(netdev, true);
+
+	igc_clear_interrupt_scheme(adapter);
+	rtnl_unlock();
+
+	status = rd32(IGC_STATUS);
+	if (status & IGC_STATUS_LU)
+		wufc &= ~IGC_WUFC_LNKC;
+
+	if (wufc) {
+		igc_setup_rctl(adapter);
+		igc_set_rx_mode(netdev);
+
+		/* turn on all-multi mode if wake on multicast is enabled */
+		if (wufc & IGC_WUFC_MC) {
+			rctl = rd32(IGC_RCTL);
+			rctl |= IGC_RCTL_MPE;
+			wr32(IGC_RCTL, rctl);
+		}
+
+		ctrl = rd32(IGC_CTRL);
+		ctrl |= IGC_CTRL_ADVD3WUC;
+		wr32(IGC_CTRL, ctrl);
+
+		/* Allow time for pending master requests to run */
+		igc_disable_pcie_master(hw);
+
+		wr32(IGC_WUC, IGC_WUC_PME_EN);
+		wr32(IGC_WUFC, wufc);
+	} else {
+		wr32(IGC_WUC, 0);
+		wr32(IGC_WUFC, 0);
+	}
+
+	wake = wufc || adapter->en_mng_pt;
+	if (!wake)
+		igc_power_down_link(adapter);
+	else
+		igc_power_up_link(adapter);
+
+	if (enable_wake)
+		*enable_wake = wake;
+
+	/* Release control of h/w to f/w.  If f/w is AMT enabled, this
+	 * would have already happened in close and is redundant.
+	 */
+	igc_release_hw_control(adapter);
+
+	pci_disable_device(pdev);
+
+	return 0;
+}
+
+#ifdef CONFIG_PM
+static int __maybe_unused igc_runtime_suspend(struct device *dev)
+{
+	return __igc_shutdown(to_pci_dev(dev), NULL, 1);
+}
+
+static void igc_deliver_wake_packet(struct net_device *netdev)
+{
+	struct igc_adapter *adapter = netdev_priv(netdev);
+	struct igc_hw *hw = &adapter->hw;
+	struct sk_buff *skb;
+	u32 wupl;
+
+	wupl = rd32(IGC_WUPL) & IGC_WUPL_MASK;
+
+	/* WUPM stores only the first 128 bytes of the wake packet.
+	 * Read the packet only if we have the whole thing.
+	 */
+	if (wupl == 0 || wupl > IGC_WUPM_BYTES)
+		return;
+
+	skb = netdev_alloc_skb_ip_align(netdev, IGC_WUPM_BYTES);
+	if (!skb)
+		return;
+
+	skb_put(skb, wupl);
+
+	/* Ensure reads are 32-bit aligned */
+	wupl = roundup(wupl, 4);
+
+	memcpy_fromio(skb->data, hw->hw_addr + IGC_WUPM_REG(0), wupl);
+
+	skb->protocol = eth_type_trans(skb, netdev);
+	netif_rx(skb);
+}
+
+static int __maybe_unused igc_resume(struct device *dev)
+{
+	struct pci_dev *pdev = to_pci_dev(dev);
+	struct net_device *netdev = pci_get_drvdata(pdev);
+	struct igc_adapter *adapter = netdev_priv(netdev);
+	struct igc_hw *hw = &adapter->hw;
+	u32 err, val;
+
+	pci_set_power_state(pdev, PCI_D0);
+	pci_restore_state(pdev);
+	pci_save_state(pdev);
+
+	if (!pci_device_is_present(pdev))
+		return -ENODEV;
+	err = pci_enable_device_mem(pdev);
+	if (err) {
+		dev_err(&pdev->dev,
+			"igc: Cannot enable PCI device from suspend\n");
+		return err;
+	}
+	pci_set_master(pdev);
+
+	pci_enable_wake(pdev, PCI_D3hot, 0);
+	pci_enable_wake(pdev, PCI_D3cold, 0);
+
+	if (igc_init_interrupt_scheme(adapter, true)) {
+		dev_err(&pdev->dev, "Unable to allocate memory for queues\n");
+		return -ENOMEM;
+	}
+
+	igc_reset(adapter);
+
+	/* let the f/w know that the h/w is now under the control of the
+	 * driver.
+	 */
+	igc_get_hw_control(adapter);
+
+	val = rd32(IGC_WUS);
+	if (val & WAKE_PKT_WUS)
+		igc_deliver_wake_packet(netdev);
+
+	wr32(IGC_WUS, ~0);
+
+	rtnl_lock();
+	if (!err && netif_running(netdev))
+		err = __igc_open(netdev, true);
+
+	if (!err)
+		netif_device_attach(netdev);
+	rtnl_unlock();
+
+	return err;
+}
+
+static int __maybe_unused igc_runtime_resume(struct device *dev)
+{
+	return igc_resume(dev);
+}
+
+static int __maybe_unused igc_suspend(struct device *dev)
+{
+	return __igc_shutdown(to_pci_dev(dev), NULL, 0);
+}
+
+static int __maybe_unused igc_runtime_idle(struct device *dev)
+{
+	struct net_device *netdev = dev_get_drvdata(dev);
+	struct igc_adapter *adapter = netdev_priv(netdev);
+
+	if (!igc_has_link(adapter))
+		pm_schedule_suspend(dev, MSEC_PER_SEC * 5);
+
+	return -EBUSY;
+}
+#endif /* CONFIG_PM */
+
+static void igc_shutdown(struct pci_dev *pdev)
+{
+	bool wake;
+
+	__igc_shutdown(pdev, &wake, 0);
+
+	if (system_state == SYSTEM_POWER_OFF) {
+		pci_wake_from_d3(pdev, wake);
+		pci_set_power_state(pdev, PCI_D3hot);
+	}
+}
+
+#ifdef CONFIG_PM
+static const struct dev_pm_ops igc_pm_ops = {
+	SET_SYSTEM_SLEEP_PM_OPS(igc_suspend, igc_resume)
+	SET_RUNTIME_PM_OPS(igc_runtime_suspend, igc_runtime_resume,
+			   igc_runtime_idle)
+};
+#endif
+
 static struct pci_driver igc_driver = {
 	.name     = igc_driver_name,
 	.id_table = igc_pci_tbl,
 	.probe    = igc_probe,
 	.remove   = igc_remove,
+#ifdef CONFIG_PM
+	.driver.pm = &igc_pm_ops,
+#endif
+	.shutdown = igc_shutdown,
 };
 
 void igc_set_flag_queue_pairs(struct igc_adapter *adapter,
diff --git a/drivers/net/ethernet/intel/igc/igc_regs.h b/drivers/net/ethernet/intel/igc/igc_regs.h
index 50d7c04dccf5..93a9139f08c5 100644
--- a/drivers/net/ethernet/intel/igc/igc_regs.h
+++ b/drivers/net/ethernet/intel/igc/igc_regs.h
@@ -215,6 +215,15 @@
 /* Shadow Ram Write Register - RW */
 #define IGC_SRWR	0x12018
 
+/* Wake Up registers */
+#define IGC_WUC		0x05800  /* Wakeup Control - RW */
+#define IGC_WUFC	0x05808  /* Wakeup Filter Control - RW */
+#define IGC_WUS		0x05810  /* Wakeup Status - R/W1C */
+#define IGC_WUPL	0x05900  /* Wakeup Packet Length - RW */
+
+/* Wake Up packet memory */
+#define IGC_WUPM_REG(_i)	(0x05A00 + ((_i) * 4))
+
 /* forward declaration */
 struct igc_hw;
 u32 igc_rd32(struct igc_hw *hw, u32 reg);
-- 
cgit v1.2.3-59-g8ed1b


From c557a4b3f73837bf074e643e163a37f848c9781c Mon Sep 17 00:00:00 2001
From: Alexander Duyck <alexander.h.duyck@linux.intel.com>
Date: Thu, 31 Oct 2019 09:58:51 -0700
Subject: e1000e: Use netdev_info instead of pr_info for link messages

Replace the pr_info calls with netdev_info in all cases related to the
netdevice link state.

As a result of this patch the link messages will change as shown below.
Before:
e1000e: ens3 NIC Link is Down
e1000e: ens3 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: Rx/Tx

After:
e1000e 0000:00:03.0 ens3: NIC Link is Down
e1000e 0000:00:03.0 ens3: NIC Link is Up 1000 Mbps Full Duplex, Flow Control: Rx/Tx

Suggested-by: Joe Perches <joe@perches.com>
Signed-off-by: Alexander Duyck <alexander.h.duyck@linux.intel.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/e1000e/netdev.c | 17 +++++++++--------
 1 file changed, 9 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c
index 4c220600ea9a..8797913b2702 100644
--- a/drivers/net/ethernet/intel/e1000e/netdev.c
+++ b/drivers/net/ethernet/intel/e1000e/netdev.c
@@ -4723,7 +4723,7 @@ int e1000e_close(struct net_device *netdev)
 		e1000_free_irq(adapter);
 
 		/* Link status message must follow this format */
-		pr_info("%s NIC Link is Down\n", netdev->name);
+		netdev_info(netdev, "NIC Link is Down\n");
 	}
 
 	napi_disable(&adapter->napi);
@@ -5073,12 +5073,13 @@ static void e1000_print_link_info(struct e1000_adapter *adapter)
 	u32 ctrl = er32(CTRL);
 
 	/* Link status message must follow this format for user tools */
-	pr_info("%s NIC Link is Up %d Mbps %s Duplex, Flow Control: %s\n",
-		adapter->netdev->name, adapter->link_speed,
-		adapter->link_duplex == FULL_DUPLEX ? "Full" : "Half",
-		(ctrl & E1000_CTRL_TFCE) && (ctrl & E1000_CTRL_RFCE) ? "Rx/Tx" :
-		(ctrl & E1000_CTRL_RFCE) ? "Rx" :
-		(ctrl & E1000_CTRL_TFCE) ? "Tx" : "None");
+	netdev_info(adapter->netdev,
+		    "NIC Link is Up %d Mbps %s Duplex, Flow Control: %s\n",
+		    adapter->link_speed,
+		    adapter->link_duplex == FULL_DUPLEX ? "Full" : "Half",
+		    (ctrl & E1000_CTRL_TFCE) && (ctrl & E1000_CTRL_RFCE) ? "Rx/Tx" :
+		    (ctrl & E1000_CTRL_RFCE) ? "Rx" :
+		    (ctrl & E1000_CTRL_TFCE) ? "Tx" : "None");
 }
 
 static bool e1000e_has_link(struct e1000_adapter *adapter)
@@ -5307,7 +5308,7 @@ static void e1000_watchdog_task(struct work_struct *work)
 			adapter->link_speed = 0;
 			adapter->link_duplex = 0;
 			/* Link status message must follow this format */
-			pr_info("%s NIC Link is Down\n", adapter->netdev->name);
+			netdev_info(netdev, "NIC Link is Down\n");
 			netif_carrier_off(netdev);
 			netif_stop_queue(netdev);
 			if (!test_bit(__E1000_DOWN, &adapter->state))
-- 
cgit v1.2.3-59-g8ed1b


From 684ea87cc312c98386c667d1046c61eb92ea8379 Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Mon, 2 Dec 2019 09:56:26 +0200
Subject: igc: Remove serdes comments from a description of methods

Serdes interface is not applicable for i225 devices.
Remove this from comments and make comments more clearly.

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_main.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index ca7b8d6791f1..25989c087aca 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -115,7 +115,7 @@ void igc_reset(struct igc_adapter *adapter)
 }
 
 /**
- * igc_power_up_link - Power up the phy/serdes link
+ * igc_power_up_link - Power up the phy link
  * @adapter: address of board private structure
  */
 static void igc_power_up_link(struct igc_adapter *adapter)
@@ -129,7 +129,7 @@ static void igc_power_up_link(struct igc_adapter *adapter)
 }
 
 /**
- * igc_power_down_link - Power down the phy/serdes link
+ * igc_power_down_link - Power down the phy link
  * @adapter: address of board private structure
  */
 static void igc_power_down_link(struct igc_adapter *adapter)
-- 
cgit v1.2.3-59-g8ed1b


From 68e039f966cb577c91649a02591646ac3919f8c9 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Wed, 1 Jan 2020 00:00:01 +0100
Subject: batman-adv: Update copyright years for 2020

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Simon Wunderlich <sw@simonwunderlich.de>
---
 include/uapi/linux/batadv_packet.h     | 2 +-
 include/uapi/linux/batman_adv.h        | 2 +-
 net/batman-adv/Kconfig                 | 2 +-
 net/batman-adv/Makefile                | 2 +-
 net/batman-adv/bat_algo.c              | 2 +-
 net/batman-adv/bat_algo.h              | 2 +-
 net/batman-adv/bat_iv_ogm.c            | 2 +-
 net/batman-adv/bat_iv_ogm.h            | 2 +-
 net/batman-adv/bat_v.c                 | 2 +-
 net/batman-adv/bat_v.h                 | 2 +-
 net/batman-adv/bat_v_elp.c             | 2 +-
 net/batman-adv/bat_v_elp.h             | 2 +-
 net/batman-adv/bat_v_ogm.c             | 2 +-
 net/batman-adv/bat_v_ogm.h             | 2 +-
 net/batman-adv/bitarray.c              | 2 +-
 net/batman-adv/bitarray.h              | 2 +-
 net/batman-adv/bridge_loop_avoidance.c | 2 +-
 net/batman-adv/bridge_loop_avoidance.h | 2 +-
 net/batman-adv/debugfs.c               | 2 +-
 net/batman-adv/debugfs.h               | 2 +-
 net/batman-adv/distributed-arp-table.c | 2 +-
 net/batman-adv/distributed-arp-table.h | 2 +-
 net/batman-adv/fragmentation.c         | 2 +-
 net/batman-adv/fragmentation.h         | 2 +-
 net/batman-adv/gateway_client.c        | 2 +-
 net/batman-adv/gateway_client.h        | 2 +-
 net/batman-adv/gateway_common.c        | 2 +-
 net/batman-adv/gateway_common.h        | 2 +-
 net/batman-adv/hard-interface.c        | 2 +-
 net/batman-adv/hard-interface.h        | 2 +-
 net/batman-adv/hash.c                  | 2 +-
 net/batman-adv/hash.h                  | 2 +-
 net/batman-adv/icmp_socket.c           | 2 +-
 net/batman-adv/icmp_socket.h           | 2 +-
 net/batman-adv/log.c                   | 2 +-
 net/batman-adv/log.h                   | 2 +-
 net/batman-adv/main.c                  | 2 +-
 net/batman-adv/main.h                  | 2 +-
 net/batman-adv/multicast.c             | 2 +-
 net/batman-adv/multicast.h             | 2 +-
 net/batman-adv/netlink.c               | 2 +-
 net/batman-adv/netlink.h               | 2 +-
 net/batman-adv/network-coding.c        | 2 +-
 net/batman-adv/network-coding.h        | 2 +-
 net/batman-adv/originator.c            | 2 +-
 net/batman-adv/originator.h            | 2 +-
 net/batman-adv/routing.c               | 2 +-
 net/batman-adv/routing.h               | 2 +-
 net/batman-adv/send.c                  | 2 +-
 net/batman-adv/send.h                  | 2 +-
 net/batman-adv/soft-interface.c        | 2 +-
 net/batman-adv/soft-interface.h        | 2 +-
 net/batman-adv/sysfs.c                 | 2 +-
 net/batman-adv/sysfs.h                 | 2 +-
 net/batman-adv/tp_meter.c              | 2 +-
 net/batman-adv/tp_meter.h              | 2 +-
 net/batman-adv/trace.c                 | 2 +-
 net/batman-adv/trace.h                 | 2 +-
 net/batman-adv/translation-table.c     | 2 +-
 net/batman-adv/translation-table.h     | 2 +-
 net/batman-adv/tvlv.c                  | 2 +-
 net/batman-adv/tvlv.h                  | 2 +-
 net/batman-adv/types.h                 | 2 +-
 63 files changed, 63 insertions(+), 63 deletions(-)

diff --git a/include/uapi/linux/batadv_packet.h b/include/uapi/linux/batadv_packet.h
index 2a15f01c2243..0ae34c85ef9e 100644
--- a/include/uapi/linux/batadv_packet.h
+++ b/include/uapi/linux/batadv_packet.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: (GPL-2.0 WITH Linux-syscall-note) */
-/* Copyright (C) 2007-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  */
diff --git a/include/uapi/linux/batman_adv.h b/include/uapi/linux/batman_adv.h
index 67f4636758af..617c180ff0c8 100644
--- a/include/uapi/linux/batman_adv.h
+++ b/include/uapi/linux/batman_adv.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: MIT */
-/* Copyright (C) 2016-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2016-2020  B.A.T.M.A.N. contributors:
  *
  * Matthias Schiffer
  */
diff --git a/net/batman-adv/Kconfig b/net/batman-adv/Kconfig
index d5028af750d5..045b2b183213 100644
--- a/net/batman-adv/Kconfig
+++ b/net/batman-adv/Kconfig
@@ -1,5 +1,5 @@
 # SPDX-License-Identifier: GPL-2.0
-# Copyright (C) 2007-2019  B.A.T.M.A.N. contributors:
+# Copyright (C) 2007-2020  B.A.T.M.A.N. contributors:
 #
 # Marek Lindner, Simon Wunderlich
 
diff --git a/net/batman-adv/Makefile b/net/batman-adv/Makefile
index fd63e116d9ff..daa49af7ff40 100644
--- a/net/batman-adv/Makefile
+++ b/net/batman-adv/Makefile
@@ -1,5 +1,5 @@
 # SPDX-License-Identifier: GPL-2.0
-# Copyright (C) 2007-2019  B.A.T.M.A.N. contributors:
+# Copyright (C) 2007-2020  B.A.T.M.A.N. contributors:
 #
 # Marek Lindner, Simon Wunderlich
 
diff --git a/net/batman-adv/bat_algo.c b/net/batman-adv/bat_algo.c
index fa39eaaab9d7..382fbe51fd34 100644
--- a/net/batman-adv/bat_algo.c
+++ b/net/batman-adv/bat_algo.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2007-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  */
diff --git a/net/batman-adv/bat_algo.h b/net/batman-adv/bat_algo.h
index 37898da8ad48..686a60bc9492 100644
--- a/net/batman-adv/bat_algo.h
+++ b/net/batman-adv/bat_algo.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2011-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2011-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Linus Lüssing
  */
diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c
index 5b0b20e6da95..f0209505e41a 100644
--- a/net/batman-adv/bat_iv_ogm.c
+++ b/net/batman-adv/bat_iv_ogm.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2007-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  */
diff --git a/net/batman-adv/bat_iv_ogm.h b/net/batman-adv/bat_iv_ogm.h
index c7a9ba305bfc..0c57c1000c64 100644
--- a/net/batman-adv/bat_iv_ogm.h
+++ b/net/batman-adv/bat_iv_ogm.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2007-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  */
diff --git a/net/batman-adv/bat_v.c b/net/batman-adv/bat_v.c
index 4ff6cf1ecae7..0ecaf1bb0068 100644
--- a/net/batman-adv/bat_v.c
+++ b/net/batman-adv/bat_v.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2013-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2013-2020  B.A.T.M.A.N. contributors:
  *
  * Linus Lüssing, Marek Lindner
  */
diff --git a/net/batman-adv/bat_v.h b/net/batman-adv/bat_v.h
index 37833db098e6..5e0be10bc84e 100644
--- a/net/batman-adv/bat_v.h
+++ b/net/batman-adv/bat_v.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2011-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2011-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Linus Lüssing
  */
diff --git a/net/batman-adv/bat_v_elp.c b/net/batman-adv/bat_v_elp.c
index 6536e8cc53a0..1e3172db7492 100644
--- a/net/batman-adv/bat_v_elp.c
+++ b/net/batman-adv/bat_v_elp.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2011-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2011-2020  B.A.T.M.A.N. contributors:
  *
  * Linus Lüssing, Marek Lindner
  */
diff --git a/net/batman-adv/bat_v_elp.h b/net/batman-adv/bat_v_elp.h
index 1a29505f4f66..4358d436be2a 100644
--- a/net/batman-adv/bat_v_elp.h
+++ b/net/batman-adv/bat_v_elp.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2013-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2013-2020  B.A.T.M.A.N. contributors:
  *
  * Linus Lüssing, Marek Lindner
  */
diff --git a/net/batman-adv/bat_v_ogm.c b/net/batman-adv/bat_v_ogm.c
index 714ce56cfcc8..969466218999 100644
--- a/net/batman-adv/bat_v_ogm.c
+++ b/net/batman-adv/bat_v_ogm.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2013-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2013-2020  B.A.T.M.A.N. contributors:
  *
  * Antonio Quartulli
  */
diff --git a/net/batman-adv/bat_v_ogm.h b/net/batman-adv/bat_v_ogm.h
index bf16d040461d..0ae2575f70bb 100644
--- a/net/batman-adv/bat_v_ogm.h
+++ b/net/batman-adv/bat_v_ogm.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2013-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2013-2020  B.A.T.M.A.N. contributors:
  *
  * Antonio Quartulli
  */
diff --git a/net/batman-adv/bitarray.c b/net/batman-adv/bitarray.c
index 7f04a6acf14e..4bc695cda397 100644
--- a/net/batman-adv/bitarray.c
+++ b/net/batman-adv/bitarray.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2006-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2006-2020  B.A.T.M.A.N. contributors:
  *
  * Simon Wunderlich, Marek Lindner
  */
diff --git a/net/batman-adv/bitarray.h b/net/batman-adv/bitarray.h
index 84ad2d2b6ac9..533c6d44cb58 100644
--- a/net/batman-adv/bitarray.h
+++ b/net/batman-adv/bitarray.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2006-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2006-2020  B.A.T.M.A.N. contributors:
  *
  * Simon Wunderlich, Marek Lindner
  */
diff --git a/net/batman-adv/bridge_loop_avoidance.c b/net/batman-adv/bridge_loop_avoidance.c
index 0eff33580f95..41cc87f06b14 100644
--- a/net/batman-adv/bridge_loop_avoidance.c
+++ b/net/batman-adv/bridge_loop_avoidance.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2011-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2011-2020  B.A.T.M.A.N. contributors:
  *
  * Simon Wunderlich
  */
diff --git a/net/batman-adv/bridge_loop_avoidance.h b/net/batman-adv/bridge_loop_avoidance.h
index 02b24a861a85..41edb2c4a327 100644
--- a/net/batman-adv/bridge_loop_avoidance.h
+++ b/net/batman-adv/bridge_loop_avoidance.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2011-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2011-2020  B.A.T.M.A.N. contributors:
  *
  * Simon Wunderlich
  */
diff --git a/net/batman-adv/debugfs.c b/net/batman-adv/debugfs.c
index 38c4d8e51155..452856c27d20 100644
--- a/net/batman-adv/debugfs.c
+++ b/net/batman-adv/debugfs.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2010-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2010-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner
  */
diff --git a/net/batman-adv/debugfs.h b/net/batman-adv/debugfs.h
index 1c5afd301ce9..7e2e8f586f42 100644
--- a/net/batman-adv/debugfs.h
+++ b/net/batman-adv/debugfs.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2010-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2010-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner
  */
diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c
index 5004e38fe792..906011b60d66 100644
--- a/net/batman-adv/distributed-arp-table.c
+++ b/net/batman-adv/distributed-arp-table.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2011-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2011-2020  B.A.T.M.A.N. contributors:
  *
  * Antonio Quartulli
  */
diff --git a/net/batman-adv/distributed-arp-table.h b/net/batman-adv/distributed-arp-table.h
index 67c7729add55..2bff2f4a325c 100644
--- a/net/batman-adv/distributed-arp-table.h
+++ b/net/batman-adv/distributed-arp-table.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2011-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2011-2020  B.A.T.M.A.N. contributors:
  *
  * Antonio Quartulli
  */
diff --git a/net/batman-adv/fragmentation.c b/net/batman-adv/fragmentation.c
index 385fccdcf69d..7cad97644d05 100644
--- a/net/batman-adv/fragmentation.c
+++ b/net/batman-adv/fragmentation.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2013-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2013-2020  B.A.T.M.A.N. contributors:
  *
  * Martin Hundebøll <martin@hundeboll.net>
  */
diff --git a/net/batman-adv/fragmentation.h b/net/batman-adv/fragmentation.h
index abfe8c6556de..881ef328b6cd 100644
--- a/net/batman-adv/fragmentation.h
+++ b/net/batman-adv/fragmentation.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2013-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2013-2020  B.A.T.M.A.N. contributors:
  *
  * Martin Hundebøll <martin@hundeboll.net>
  */
diff --git a/net/batman-adv/gateway_client.c b/net/batman-adv/gateway_client.c
index 47df4c678988..e22e49289677 100644
--- a/net/batman-adv/gateway_client.c
+++ b/net/batman-adv/gateway_client.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2009-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2009-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner
  */
diff --git a/net/batman-adv/gateway_client.h b/net/batman-adv/gateway_client.h
index 0be8e7178ec7..88b5dba84354 100644
--- a/net/batman-adv/gateway_client.h
+++ b/net/batman-adv/gateway_client.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2009-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2009-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner
  */
diff --git a/net/batman-adv/gateway_common.c b/net/batman-adv/gateway_common.c
index fc55750542e4..16cd9450ceb1 100644
--- a/net/batman-adv/gateway_common.c
+++ b/net/batman-adv/gateway_common.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2009-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2009-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner
  */
diff --git a/net/batman-adv/gateway_common.h b/net/batman-adv/gateway_common.h
index 211b14b37db8..c3a0c5a7f7e9 100644
--- a/net/batman-adv/gateway_common.h
+++ b/net/batman-adv/gateway_common.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2009-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2009-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner
  */
diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c
index afb52282d5bd..c7e98a40dd33 100644
--- a/net/batman-adv/hard-interface.c
+++ b/net/batman-adv/hard-interface.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2007-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  */
diff --git a/net/batman-adv/hard-interface.h b/net/batman-adv/hard-interface.h
index bbb8a6f18d6b..bad2e50135e8 100644
--- a/net/batman-adv/hard-interface.h
+++ b/net/batman-adv/hard-interface.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2007-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  */
diff --git a/net/batman-adv/hash.c b/net/batman-adv/hash.c
index a9d4e176f4de..68638e0450a6 100644
--- a/net/batman-adv/hash.c
+++ b/net/batman-adv/hash.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2006-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2006-2020  B.A.T.M.A.N. contributors:
  *
  * Simon Wunderlich, Marek Lindner
  */
diff --git a/net/batman-adv/hash.h b/net/batman-adv/hash.h
index 57877f0b78e0..91ae9f32b580 100644
--- a/net/batman-adv/hash.h
+++ b/net/batman-adv/hash.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2006-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2006-2020  B.A.T.M.A.N. contributors:
  *
  * Simon Wunderlich, Marek Lindner
  */
diff --git a/net/batman-adv/icmp_socket.c b/net/batman-adv/icmp_socket.c
index 0a70b66e8770..ccb535c77e5d 100644
--- a/net/batman-adv/icmp_socket.c
+++ b/net/batman-adv/icmp_socket.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2007-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner
  */
diff --git a/net/batman-adv/icmp_socket.h b/net/batman-adv/icmp_socket.h
index 27fafff586df..6abd0f4742ef 100644
--- a/net/batman-adv/icmp_socket.h
+++ b/net/batman-adv/icmp_socket.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2007-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner
  */
diff --git a/net/batman-adv/log.c b/net/batman-adv/log.c
index 11941cf1adcc..a67b2b091447 100644
--- a/net/batman-adv/log.c
+++ b/net/batman-adv/log.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2010-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2010-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner
  */
diff --git a/net/batman-adv/log.h b/net/batman-adv/log.h
index 41fd900121c5..f9884dc56cf3 100644
--- a/net/batman-adv/log.h
+++ b/net/batman-adv/log.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2007-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  */
diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c
index 4a89177def64..0b840e1e3dfa 100644
--- a/net/batman-adv/main.c
+++ b/net/batman-adv/main.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2007-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  */
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index fd8c0728ddc7..692306df7b6f 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2007-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  */
diff --git a/net/batman-adv/multicast.c b/net/batman-adv/multicast.c
index f9ec8e7507b6..9ebdc1e864b9 100644
--- a/net/batman-adv/multicast.c
+++ b/net/batman-adv/multicast.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2014-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2014-2020  B.A.T.M.A.N. contributors:
  *
  * Linus Lüssing
  */
diff --git a/net/batman-adv/multicast.h b/net/batman-adv/multicast.h
index 5d9e2bb29c97..ebf825991ecd 100644
--- a/net/batman-adv/multicast.h
+++ b/net/batman-adv/multicast.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2014-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2014-2020  B.A.T.M.A.N. contributors:
  *
  * Linus Lüssing
  */
diff --git a/net/batman-adv/netlink.c b/net/batman-adv/netlink.c
index 7e052d6f759b..02ed073f95a9 100644
--- a/net/batman-adv/netlink.c
+++ b/net/batman-adv/netlink.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2016-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2016-2020  B.A.T.M.A.N. contributors:
  *
  * Matthias Schiffer
  */
diff --git a/net/batman-adv/netlink.h b/net/batman-adv/netlink.h
index ddc674e47dbb..7ee48f916997 100644
--- a/net/batman-adv/netlink.h
+++ b/net/batman-adv/netlink.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2016-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2016-2020  B.A.T.M.A.N. contributors:
  *
  * Matthias Schiffer
  */
diff --git a/net/batman-adv/network-coding.c b/net/batman-adv/network-coding.c
index 580609389f0f..8f0717c3f7b5 100644
--- a/net/batman-adv/network-coding.c
+++ b/net/batman-adv/network-coding.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2012-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2012-2020  B.A.T.M.A.N. contributors:
  *
  * Martin Hundebøll, Jeppe Ledet-Pedersen
  */
diff --git a/net/batman-adv/network-coding.h b/net/batman-adv/network-coding.h
index 753fa49723cf..334289084127 100644
--- a/net/batman-adv/network-coding.h
+++ b/net/batman-adv/network-coding.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2012-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2012-2020  B.A.T.M.A.N. contributors:
  *
  * Martin Hundebøll, Jeppe Ledet-Pedersen
  */
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index 38613487fb1b..5b0c2fffc214 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2009-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2009-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  */
diff --git a/net/batman-adv/originator.h b/net/batman-adv/originator.h
index 512a1f99dd75..7bc01c138b3a 100644
--- a/net/batman-adv/originator.h
+++ b/net/batman-adv/originator.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2007-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  */
diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c
index f0f864820dea..3632bd976c56 100644
--- a/net/batman-adv/routing.c
+++ b/net/batman-adv/routing.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2007-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  */
diff --git a/net/batman-adv/routing.h b/net/batman-adv/routing.h
index c20feac95107..2ed49db6eff5 100644
--- a/net/batman-adv/routing.h
+++ b/net/batman-adv/routing.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2007-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  */
diff --git a/net/batman-adv/send.c b/net/batman-adv/send.c
index 3ce5f7bad369..7f8ade04e08e 100644
--- a/net/batman-adv/send.c
+++ b/net/batman-adv/send.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2007-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  */
diff --git a/net/batman-adv/send.h b/net/batman-adv/send.h
index 5fc0fd1e5d08..0d36e15589f6 100644
--- a/net/batman-adv/send.h
+++ b/net/batman-adv/send.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2007-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  */
diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c
index 832e156c519e..5f05a728f347 100644
--- a/net/batman-adv/soft-interface.c
+++ b/net/batman-adv/soft-interface.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2007-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  */
diff --git a/net/batman-adv/soft-interface.h b/net/batman-adv/soft-interface.h
index 29139ad769fe..534e08d6ad91 100644
--- a/net/batman-adv/soft-interface.h
+++ b/net/batman-adv/soft-interface.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2007-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner
  */
diff --git a/net/batman-adv/sysfs.c b/net/batman-adv/sysfs.c
index e5bbc28ed12c..c45962d8527b 100644
--- a/net/batman-adv/sysfs.c
+++ b/net/batman-adv/sysfs.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2010-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2010-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner
  */
diff --git a/net/batman-adv/sysfs.h b/net/batman-adv/sysfs.h
index 5e466093dfa5..d987f8b30a98 100644
--- a/net/batman-adv/sysfs.h
+++ b/net/batman-adv/sysfs.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2010-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2010-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner
  */
diff --git a/net/batman-adv/tp_meter.c b/net/batman-adv/tp_meter.c
index dd6a9a40dbb9..bd2ac570c42c 100644
--- a/net/batman-adv/tp_meter.c
+++ b/net/batman-adv/tp_meter.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2012-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2012-2020  B.A.T.M.A.N. contributors:
  *
  * Edo Monticelli, Antonio Quartulli
  */
diff --git a/net/batman-adv/tp_meter.h b/net/batman-adv/tp_meter.h
index 78d310da0ad3..140105215aa2 100644
--- a/net/batman-adv/tp_meter.h
+++ b/net/batman-adv/tp_meter.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2012-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2012-2020  B.A.T.M.A.N. contributors:
  *
  * Edo Monticelli, Antonio Quartulli
  */
diff --git a/net/batman-adv/trace.c b/net/batman-adv/trace.c
index 3cedd2c36528..3444d9e4e90d 100644
--- a/net/batman-adv/trace.c
+++ b/net/batman-adv/trace.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2010-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2010-2020  B.A.T.M.A.N. contributors:
  *
  * Sven Eckelmann
  */
diff --git a/net/batman-adv/trace.h b/net/batman-adv/trace.h
index d8f764521c0b..f631b1e01b89 100644
--- a/net/batman-adv/trace.h
+++ b/net/batman-adv/trace.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2010-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2010-2020  B.A.T.M.A.N. contributors:
  *
  * Sven Eckelmann
  */
diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c
index 8a482c5ec67b..852932838ddc 100644
--- a/net/batman-adv/translation-table.c
+++ b/net/batman-adv/translation-table.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2007-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich, Antonio Quartulli
  */
diff --git a/net/batman-adv/translation-table.h b/net/batman-adv/translation-table.h
index 4a98860d7f0e..b24d35b9226a 100644
--- a/net/batman-adv/translation-table.h
+++ b/net/batman-adv/translation-table.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2007-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich, Antonio Quartulli
  */
diff --git a/net/batman-adv/tvlv.c b/net/batman-adv/tvlv.c
index aae63f0d21eb..0963a43ad996 100644
--- a/net/batman-adv/tvlv.c
+++ b/net/batman-adv/tvlv.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/* Copyright (C) 2007-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  */
diff --git a/net/batman-adv/tvlv.h b/net/batman-adv/tvlv.h
index 36985000a0a8..d509d00c7a23 100644
--- a/net/batman-adv/tvlv.h
+++ b/net/batman-adv/tvlv.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2007-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  */
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index bdf9827b5f63..4a17a66cc572 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (C) 2007-2019  B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2020  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  */
-- 
cgit v1.2.3-59-g8ed1b


From b2e55ca89245fccd459a2d56850822a69a6f91da Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Wed, 1 Jan 2020 00:51:33 +0100
Subject: batman-adv: Disable CONFIG_BATMAN_ADV_SYSFS by default

The sysfs support in batman-adv is deprecated since a while and will be
removed completely next year.

All tools which were known to the batman-adv development team are
supporting the batman-adv netlink interface since a while. Thus
disabling CONFIG_BATMAN_ADV_SYSFS by default should not cause problems on
most systems. It is still possible to enable it in case it is still
required in a specific setup.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Simon Wunderlich <sw@simonwunderlich.de>
---
 net/batman-adv/Kconfig | 1 -
 1 file changed, 1 deletion(-)

diff --git a/net/batman-adv/Kconfig b/net/batman-adv/Kconfig
index 045b2b183213..c762758a4649 100644
--- a/net/batman-adv/Kconfig
+++ b/net/batman-adv/Kconfig
@@ -100,7 +100,6 @@ config BATMAN_ADV_DEBUG
 config BATMAN_ADV_SYSFS
 	bool "batman-adv sysfs entries"
 	depends on BATMAN_ADV
-	default y
 	help
 	  Say Y here if you want to enable batman-adv device configuration and
 	  status interface through sysfs attributes. It is replaced by the
-- 
cgit v1.2.3-59-g8ed1b


From 44768decb7c0a6c7c01995dbacc864f3921b6dac Mon Sep 17 00:00:00 2001
From: Jesper Dangaard Brouer <brouer@redhat.com>
Date: Fri, 27 Dec 2019 18:13:18 +0100
Subject: page_pool: handle page recycle for NUMA_NO_NODE condition

The check in pool_page_reusable (page_to_nid(page) == pool->p.nid) is
not valid if page_pool was configured with pool->p.nid = NUMA_NO_NODE.

The goal of the NUMA changes in commit d5394610b1ba ("page_pool: Don't
recycle non-reusable pages"), were to have RX-pages that belongs to the
same NUMA node as the CPU processing RX-packet during softirq/NAPI. As
illustrated by the performance measurements.

This patch moves the NAPI checks out of fast-path, and at the same time
solves the NUMA_NO_NODE issue.

First realize that alloc_pages_node() with pool->p.nid = NUMA_NO_NODE
will lookup current CPU nid (Numa ID) via numa_mem_id(), which is used
as the the preferred nid.  It is only in rare situations, where
e.g. NUMA zone runs dry, that page gets doesn't get allocated from
preferred nid.  The page_pool API allows drivers to control the nid
themselves via controlling pool->p.nid.

This patch moves the NAPI check to when alloc cache is refilled, via
dequeuing/consuming pages from the ptr_ring. Thus, we can allow placing
pages from remote NUMA into the ptr_ring, as the dequeue/consume step
will check the NUMA node. All current drivers using page_pool will
alloc/refill RX-ring from same CPU running softirq/NAPI process.

Drivers that control the nid explicitly, also use page_pool_update_nid
when changing nid runtime.  To speed up transision to new nid the alloc
cache is now flushed on nid changes.  This force pages to come from
ptr_ring, which does the appropate nid check.

For the NUMA_NO_NODE case, when a NIC IRQ is moved to another NUMA
node, we accept that transitioning the alloc cache doesn't happen
immediately. The preferred nid change runtime via consulting
numa_mem_id() based on the CPU processing RX-packets.

Notice, to avoid stressing the page buddy allocator and avoid doing too
much work under softirq with preempt disabled, the NUMA check at
ptr_ring dequeue will break the refill cycle, when detecting a NUMA
mismatch. This will cause a slower transition, but its done on purpose.

Fixes: d5394610b1ba ("page_pool: Don't recycle non-reusable pages")
Reported-by: Li RongQing <lirongqing@baidu.com>
Reported-by: Yunsheng Lin <linyunsheng@huawei.com>
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/page_pool.c | 80 +++++++++++++++++++++++++++++++++++++++-------------
 1 file changed, 61 insertions(+), 19 deletions(-)

diff --git a/net/core/page_pool.c b/net/core/page_pool.c
index a6aefe989043..515bfb97fda1 100644
--- a/net/core/page_pool.c
+++ b/net/core/page_pool.c
@@ -96,10 +96,60 @@ struct page_pool *page_pool_create(const struct page_pool_params *params)
 }
 EXPORT_SYMBOL(page_pool_create);
 
+static void __page_pool_return_page(struct page_pool *pool, struct page *page);
+
+noinline
+static struct page *page_pool_refill_alloc_cache(struct page_pool *pool,
+						 bool refill)
+{
+	struct ptr_ring *r = &pool->ring;
+	struct page *page;
+	int pref_nid; /* preferred NUMA node */
+
+	/* Quicker fallback, avoid locks when ring is empty */
+	if (__ptr_ring_empty(r))
+		return NULL;
+
+	/* Softirq guarantee CPU and thus NUMA node is stable. This,
+	 * assumes CPU refilling driver RX-ring will also run RX-NAPI.
+	 */
+	pref_nid = (pool->p.nid == NUMA_NO_NODE) ? numa_mem_id() : pool->p.nid;
+
+	/* Slower-path: Get pages from locked ring queue */
+	spin_lock(&r->consumer_lock);
+
+	/* Refill alloc array, but only if NUMA match */
+	do {
+		page = __ptr_ring_consume(r);
+		if (unlikely(!page))
+			break;
+
+		if (likely(page_to_nid(page) == pref_nid)) {
+			pool->alloc.cache[pool->alloc.count++] = page;
+		} else {
+			/* NUMA mismatch;
+			 * (1) release 1 page to page-allocator and
+			 * (2) break out to fallthrough to alloc_pages_node.
+			 * This limit stress on page buddy alloactor.
+			 */
+			__page_pool_return_page(pool, page);
+			page = NULL;
+			break;
+		}
+	} while (pool->alloc.count < PP_ALLOC_CACHE_REFILL &&
+		 refill);
+
+	/* Return last page */
+	if (likely(pool->alloc.count > 0))
+		page = pool->alloc.cache[--pool->alloc.count];
+
+	spin_unlock(&r->consumer_lock);
+	return page;
+}
+
 /* fast path */
 static struct page *__page_pool_get_cached(struct page_pool *pool)
 {
-	struct ptr_ring *r = &pool->ring;
 	bool refill = false;
 	struct page *page;
 
@@ -113,20 +163,7 @@ static struct page *__page_pool_get_cached(struct page_pool *pool)
 		refill = true;
 	}
 
-	/* Quicker fallback, avoid locks when ring is empty */
-	if (__ptr_ring_empty(r))
-		return NULL;
-
-	/* Slow-path: Get page from locked ring queue,
-	 * refill alloc array if requested.
-	 */
-	spin_lock(&r->consumer_lock);
-	page = __ptr_ring_consume(r);
-	if (refill)
-		pool->alloc.count = __ptr_ring_consume_batched(r,
-							pool->alloc.cache,
-							PP_ALLOC_CACHE_REFILL);
-	spin_unlock(&r->consumer_lock);
+	page = page_pool_refill_alloc_cache(pool, refill);
 	return page;
 }
 
@@ -311,13 +348,10 @@ static bool __page_pool_recycle_direct(struct page *page,
 
 /* page is NOT reusable when:
  * 1) allocated when system is under some pressure. (page_is_pfmemalloc)
- * 2) belongs to a different NUMA node than pool->p.nid.
- *
- * To update pool->p.nid users must call page_pool_update_nid.
  */
 static bool pool_page_reusable(struct page_pool *pool, struct page *page)
 {
-	return !page_is_pfmemalloc(page) && page_to_nid(page) == pool->p.nid;
+	return !page_is_pfmemalloc(page);
 }
 
 void __page_pool_put_page(struct page_pool *pool, struct page *page,
@@ -484,7 +518,15 @@ EXPORT_SYMBOL(page_pool_destroy);
 /* Caller must provide appropriate safe context, e.g. NAPI. */
 void page_pool_update_nid(struct page_pool *pool, int new_nid)
 {
+	struct page *page;
+
 	trace_page_pool_update_nid(pool, new_nid);
 	pool->p.nid = new_nid;
+
+	/* Flush pool alloc cache, as refill will check NUMA node */
+	while (pool->alloc.count) {
+		page = pool->alloc.cache[--pool->alloc.count];
+		__page_pool_return_page(pool, page);
+	}
 }
 EXPORT_SYMBOL(page_pool_update_nid);
-- 
cgit v1.2.3-59-g8ed1b


From f13fc10785bc04f83fe0d7730beb8ff4be563a20 Mon Sep 17 00:00:00 2001
From: Jesper Dangaard Brouer <brouer@redhat.com>
Date: Fri, 27 Dec 2019 18:13:24 +0100
Subject: page_pool: help compiler remove code in case CONFIG_NUMA=n

When kernel is compiled without NUMA support, then page_pool NUMA
config setting (pool->p.nid) doesn't make any practical sense. The
compiler cannot see that it can remove the code paths.

This patch avoids reading pool->p.nid setting in case of !CONFIG_NUMA,
in allocation and numa check code, which helps compiler to see the
optimisation potential. It leaves update code intact to keep API the
same.

 $ ./scripts/bloat-o-meter net/core/page_pool.o-numa-enabled \
                           net/core/page_pool.o-numa-disabled
 add/remove: 0/0 grow/shrink: 0/3 up/down: 0/-113 (-113)
 Function                                     old     new   delta
 page_pool_create                             401     398      -3
 __page_pool_alloc_pages_slow                 439     426     -13
 page_pool_refill_alloc_cache                 425     328     -97
 Total: Before=3611, After=3498, chg -3.13%

Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/page_pool.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/net/core/page_pool.c b/net/core/page_pool.c
index 515bfb97fda1..9b7cbe35df37 100644
--- a/net/core/page_pool.c
+++ b/net/core/page_pool.c
@@ -113,7 +113,12 @@ static struct page *page_pool_refill_alloc_cache(struct page_pool *pool,
 	/* Softirq guarantee CPU and thus NUMA node is stable. This,
 	 * assumes CPU refilling driver RX-ring will also run RX-NAPI.
 	 */
+#ifdef CONFIG_NUMA
 	pref_nid = (pool->p.nid == NUMA_NO_NODE) ? numa_mem_id() : pool->p.nid;
+#else
+	/* Ignore pool->p.nid setting if !CONFIG_NUMA, helps compiler */
+	pref_nid = numa_mem_id(); /* will be zero like page_to_nid() */
+#endif
 
 	/* Slower-path: Get pages from locked ring queue */
 	spin_lock(&r->consumer_lock);
@@ -200,7 +205,11 @@ static struct page *__page_pool_alloc_pages_slow(struct page_pool *pool,
 	 */
 
 	/* Cache was empty, do real allocation */
+#ifdef CONFIG_NUMA
 	page = alloc_pages_node(pool->p.nid, gfp, pool->p.order);
+#else
+	page = alloc_pages(gfp, pool->p.order);
+#endif
 	if (!page)
 		return NULL;
 
-- 
cgit v1.2.3-59-g8ed1b


From d8446884f8d0b2b9f124b98cdf2d9b8c5fe128ba Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@mellanox.com>
Date: Sun, 29 Dec 2019 13:48:27 +0200
Subject: mlxsw: reg: Add QoS Port DSCP to Priority Mapping Register

Add QPDP. This register controls the port default Switch Priority and
Color. The default Switch Priority and Color are used for frames where the
trust state uses default values. Currently there are two cases where this
applies: a port is in trust-PCP state, but a packet arrives untagged; and a
port is in trust-DSCP state, but a non-IP packet arrives.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/reg.h | 33 +++++++++++++++++++++++++++++++
 1 file changed, 33 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h
index 2b8b4ed6bbef..fd59280cf979 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/reg.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h
@@ -3748,6 +3748,38 @@ mlxsw_reg_qpdsm_prio_pack(char *payload, unsigned short prio, u8 dscp)
 	mlxsw_reg_qpdsm_prio_entry_color2_dscp_set(payload, prio, dscp);
 }
 
+/* QPDP - QoS Port DSCP to Priority Mapping Register
+ * -------------------------------------------------
+ * This register controls the port default Switch Priority and Color. The
+ * default Switch Priority and Color are used for frames where the trust state
+ * uses default values. All member ports of a LAG should be configured with the
+ * same default values.
+ */
+#define MLXSW_REG_QPDP_ID 0x4007
+#define MLXSW_REG_QPDP_LEN 0x8
+
+MLXSW_REG_DEFINE(qpdp, MLXSW_REG_QPDP_ID, MLXSW_REG_QPDP_LEN);
+
+/* reg_qpdp_local_port
+ * Local Port. Supported for data packets from CPU port.
+ * Access: Index
+ */
+MLXSW_ITEM32(reg, qpdp, local_port, 0x00, 16, 8);
+
+/* reg_qpdp_switch_prio
+ * Default port Switch Priority (default 0)
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, qpdp, switch_prio, 0x04, 0, 4);
+
+static inline void mlxsw_reg_qpdp_pack(char *payload, u8 local_port,
+				       u8 switch_prio)
+{
+	MLXSW_REG_ZERO(qpdp, payload);
+	mlxsw_reg_qpdp_local_port_set(payload, local_port);
+	mlxsw_reg_qpdp_switch_prio_set(payload, switch_prio);
+}
+
 /* QPDPM - QoS Port DSCP to Priority Mapping Register
  * --------------------------------------------------
  * This register controls the mapping from DSCP field to
@@ -10580,6 +10612,7 @@ static const struct mlxsw_reg_info *mlxsw_reg_infos[] = {
 	MLXSW_REG(qeec),
 	MLXSW_REG(qrwe),
 	MLXSW_REG(qpdsm),
+	MLXSW_REG(qpdp),
 	MLXSW_REG(qpdpm),
 	MLXSW_REG(qtctm),
 	MLXSW_REG(qpsc),
-- 
cgit v1.2.3-59-g8ed1b


From 379a00dd21618f4354d41b1ac148188e005dbc05 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@mellanox.com>
Date: Sun, 29 Dec 2019 13:48:28 +0200
Subject: mlxsw: spectrum_dcb: Allow setting default port priority

When APP TLV selector 1 (EtherType) is used with PID of 0, the
corresponding entry specifies "default application priority [...] when
application priority is not otherwise specified."

mlxsw currently supports this type of APP entry, but uses it only as a
fallback for unspecified DSCP rules. However non-IP traffic is prioritized
according to port-default priority, not according to the DSCP-to-prio
tables, and thus it's currently not possible to prioritize such traffic
correctly.

Extend the use of the abovementioned APP entry to also set default port
priority.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum_dcb.c | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_dcb.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_dcb.c
index fe3bbba90659..db66f2b56a6d 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_dcb.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_dcb.c
@@ -368,6 +368,17 @@ err_update_qrwe:
 	return err;
 }
 
+static int
+mlxsw_sp_port_dcb_app_update_qpdp(struct mlxsw_sp_port *mlxsw_sp_port,
+				  u8 default_prio)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char qpdp_pl[MLXSW_REG_QPDP_LEN];
+
+	mlxsw_reg_qpdp_pack(qpdp_pl, mlxsw_sp_port->local_port, default_prio);
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(qpdp), qpdp_pl);
+}
+
 static int
 mlxsw_sp_port_dcb_app_update_qpdpm(struct mlxsw_sp_port *mlxsw_sp_port,
 				   struct dcb_ieee_app_dscp_map *map)
@@ -405,6 +416,12 @@ static int mlxsw_sp_port_dcb_app_update(struct mlxsw_sp_port *mlxsw_sp_port)
 	int err;
 
 	default_prio = mlxsw_sp_port_dcb_app_default_prio(mlxsw_sp_port);
+	err = mlxsw_sp_port_dcb_app_update_qpdp(mlxsw_sp_port, default_prio);
+	if (err) {
+		netdev_err(mlxsw_sp_port->dev, "Couldn't configure port default priority\n");
+		return err;
+	}
+
 	have_dscp = mlxsw_sp_port_dcb_app_prio_dscp_map(mlxsw_sp_port,
 							&prio_map);
 
-- 
cgit v1.2.3-59-g8ed1b


From c5341bcc337ca762faa58790c2065c76cd05f2e6 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@mellanox.com>
Date: Sun, 29 Dec 2019 13:48:29 +0200
Subject: selftests: mlxsw: Add a self-test for port-default priority

Send non-IP traffic to a port and observe that it gets prioritized
according to the lldptool app=$prio,1,0 rules.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../selftests/drivers/net/mlxsw/qos_defprio.sh     | 176 +++++++++++++++++++++
 tools/testing/selftests/net/forwarding/lib.sh      |  23 +++
 2 files changed, 199 insertions(+)
 create mode 100755 tools/testing/selftests/drivers/net/mlxsw/qos_defprio.sh

diff --git a/tools/testing/selftests/drivers/net/mlxsw/qos_defprio.sh b/tools/testing/selftests/drivers/net/mlxsw/qos_defprio.sh
new file mode 100755
index 000000000000..eff6393ce974
--- /dev/null
+++ b/tools/testing/selftests/drivers/net/mlxsw/qos_defprio.sh
@@ -0,0 +1,176 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+# Test for port-default priority. Non-IP packets ingress $swp1 and are
+# prioritized according to the default priority specified at the port.
+# rx_octets_prio_* counters are used to verify the prioritization.
+#
+# +-----------------------+
+# | H1                    |
+# |    + $h1              |
+# |    | 192.0.2.1/28     |
+# +----|------------------+
+#      |
+# +----|------------------+
+# | SW |                  |
+# |    + $swp1            |
+# |      192.0.2.2/28     |
+# |      APP=<prio>,1,0   |
+# +-----------------------+
+
+ALL_TESTS="
+	ping_ipv4
+	test_defprio
+"
+
+lib_dir=$(dirname $0)/../../../net/forwarding
+
+NUM_NETIFS=2
+: ${HIT_TIMEOUT:=1000} # ms
+source $lib_dir/lib.sh
+
+declare -a APP
+
+defprio_install()
+{
+	local dev=$1; shift
+	local prio=$1; shift
+	local app="app=$prio,1,0"
+
+	lldptool -T -i $dev -V APP $app >/dev/null
+	lldpad_app_wait_set $dev
+	APP[$prio]=$app
+}
+
+defprio_uninstall()
+{
+	local dev=$1; shift
+	local prio=$1; shift
+	local app=${APP[$prio]}
+
+	lldptool -T -i $dev -V APP -d $app >/dev/null
+	lldpad_app_wait_del
+	unset APP[$prio]
+}
+
+defprio_flush()
+{
+	local dev=$1; shift
+	local prio
+
+	if ((${#APP[@]})); then
+		lldptool -T -i $dev -V APP -d ${APP[@]} >/dev/null
+	fi
+	lldpad_app_wait_del
+	APP=()
+}
+
+h1_create()
+{
+	simple_if_init $h1 192.0.2.1/28
+}
+
+h1_destroy()
+{
+	simple_if_fini $h1 192.0.2.1/28
+}
+
+switch_create()
+{
+	ip link set dev $swp1 up
+	ip addr add dev $swp1 192.0.2.2/28
+}
+
+switch_destroy()
+{
+	defprio_flush $swp1
+	ip addr del dev $swp1 192.0.2.2/28
+	ip link set dev $swp1 down
+}
+
+setup_prepare()
+{
+	h1=${NETIFS[p1]}
+	swp1=${NETIFS[p2]}
+
+	vrf_prepare
+
+	h1_create
+	switch_create
+}
+
+cleanup()
+{
+	pre_cleanup
+
+	switch_destroy
+	h1_destroy
+
+	vrf_cleanup
+}
+
+ping_ipv4()
+{
+	ping_test $h1 192.0.2.2
+}
+
+wait_for_packets()
+{
+	local t0=$1; shift
+	local prio_observe=$1; shift
+
+	local t1=$(ethtool_stats_get $swp1 rx_frames_prio_$prio_observe)
+	local delta=$((t1 - t0))
+	echo $delta
+	((delta >= 10))
+}
+
+__test_defprio()
+{
+	local prio_install=$1; shift
+	local prio_observe=$1; shift
+	local delta
+	local key
+	local i
+
+	RET=0
+
+	defprio_install $swp1 $prio_install
+
+	local t0=$(ethtool_stats_get $swp1 rx_frames_prio_$prio_observe)
+	mausezahn -q $h1 -d 100m -c 10 -t arp reply
+	delta=$(busywait "$HIT_TIMEOUT" wait_for_packets $t0 $prio_observe)
+
+	check_err $? "Default priority $prio_install/$prio_observe: Expected to capture 10 packets, got $delta."
+	log_test "Default priority $prio_install/$prio_observe"
+
+	defprio_uninstall $swp1 $prio_install
+}
+
+test_defprio()
+{
+	local prio
+
+	for prio in {0..7}; do
+		__test_defprio $prio $prio
+	done
+
+	defprio_install $swp1 3
+	__test_defprio 0 3
+	__test_defprio 1 3
+	__test_defprio 2 3
+	__test_defprio 4 4
+	__test_defprio 5 5
+	__test_defprio 6 6
+	__test_defprio 7 7
+	defprio_uninstall $swp1 3
+}
+
+trap cleanup EXIT
+
+setup_prepare
+setup_wait
+
+tests_run
+
+exit $EXIT_STATUS
diff --git a/tools/testing/selftests/net/forwarding/lib.sh b/tools/testing/selftests/net/forwarding/lib.sh
index a0b09bb6995e..8dc5fac98cbc 100644
--- a/tools/testing/selftests/net/forwarding/lib.sh
+++ b/tools/testing/selftests/net/forwarding/lib.sh
@@ -225,6 +225,29 @@ log_info()
 	echo "INFO: $msg"
 }
 
+busywait()
+{
+	local timeout=$1; shift
+
+	local start_time="$(date -u +%s%3N)"
+	while true
+	do
+		local out
+		out=$("$@")
+		local ret=$?
+		if ((!ret)); then
+			echo -n "$out"
+			return 0
+		fi
+
+		local current_time="$(date -u +%s%3N)"
+		if ((current_time - start_time > timeout)); then
+			echo -n "$out"
+			return 1
+		fi
+	done
+}
+
 setup_wait_dev()
 {
 	local dev=$1; shift
-- 
cgit v1.2.3-59-g8ed1b


From 98c8147648fa1cdb803fe51fa0a6ff439ed4769f Mon Sep 17 00:00:00 2001
From: Niu Xilei <niu_xilei@163.com>
Date: Mon, 30 Dec 2019 17:52:22 +0800
Subject: vxlan: Fix alignment and code style of vxlan.c

Fixed Coding function and style issues

Signed-off-by: Niu Xilei <niu_xilei@163.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/vxlan.c | 21 +++++++++++----------
 1 file changed, 11 insertions(+), 10 deletions(-)

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 3ec6b506033d..e95e6585ab82 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1060,6 +1060,7 @@ static int vxlan_fdb_parse(struct nlattr *tb[], struct vxlan_dev *vxlan,
 			return err;
 	} else {
 		union vxlan_addr *remote = &vxlan->default_dst.remote_ip;
+
 		if (remote->sa.sa_family == AF_INET) {
 			ip->sin.sin_addr.s_addr = htonl(INADDR_ANY);
 			ip->sa.sa_family = AF_INET;
@@ -1696,7 +1697,7 @@ static int vxlan_rcv(struct sock *sk, struct sk_buff *skb)
 
 	if (__iptunnel_pull_header(skb, VXLAN_HLEN, protocol, raw_proto,
 				   !net_eq(vxlan->net, dev_net(vxlan->dev))))
-			goto drop;
+		goto drop;
 
 	if (vxlan_collect_metadata(vs)) {
 		struct metadata_dst *tun_dst;
@@ -4128,30 +4129,30 @@ static int vxlan_fill_info(struct sk_buff *skb, const struct net_device *dev)
 	    nla_put_u8(skb, IFLA_VXLAN_DF, vxlan->cfg.df) ||
 	    nla_put_be32(skb, IFLA_VXLAN_LABEL, vxlan->cfg.label) ||
 	    nla_put_u8(skb, IFLA_VXLAN_LEARNING,
-			!!(vxlan->cfg.flags & VXLAN_F_LEARN)) ||
+		       !!(vxlan->cfg.flags & VXLAN_F_LEARN)) ||
 	    nla_put_u8(skb, IFLA_VXLAN_PROXY,
-			!!(vxlan->cfg.flags & VXLAN_F_PROXY)) ||
+		       !!(vxlan->cfg.flags & VXLAN_F_PROXY)) ||
 	    nla_put_u8(skb, IFLA_VXLAN_RSC,
 		       !!(vxlan->cfg.flags & VXLAN_F_RSC)) ||
 	    nla_put_u8(skb, IFLA_VXLAN_L2MISS,
-			!!(vxlan->cfg.flags & VXLAN_F_L2MISS)) ||
+		       !!(vxlan->cfg.flags & VXLAN_F_L2MISS)) ||
 	    nla_put_u8(skb, IFLA_VXLAN_L3MISS,
-			!!(vxlan->cfg.flags & VXLAN_F_L3MISS)) ||
+		       !!(vxlan->cfg.flags & VXLAN_F_L3MISS)) ||
 	    nla_put_u8(skb, IFLA_VXLAN_COLLECT_METADATA,
 		       !!(vxlan->cfg.flags & VXLAN_F_COLLECT_METADATA)) ||
 	    nla_put_u32(skb, IFLA_VXLAN_AGEING, vxlan->cfg.age_interval) ||
 	    nla_put_u32(skb, IFLA_VXLAN_LIMIT, vxlan->cfg.addrmax) ||
 	    nla_put_be16(skb, IFLA_VXLAN_PORT, vxlan->cfg.dst_port) ||
 	    nla_put_u8(skb, IFLA_VXLAN_UDP_CSUM,
-			!(vxlan->cfg.flags & VXLAN_F_UDP_ZERO_CSUM_TX)) ||
+		       !(vxlan->cfg.flags & VXLAN_F_UDP_ZERO_CSUM_TX)) ||
 	    nla_put_u8(skb, IFLA_VXLAN_UDP_ZERO_CSUM6_TX,
-			!!(vxlan->cfg.flags & VXLAN_F_UDP_ZERO_CSUM6_TX)) ||
+		       !!(vxlan->cfg.flags & VXLAN_F_UDP_ZERO_CSUM6_TX)) ||
 	    nla_put_u8(skb, IFLA_VXLAN_UDP_ZERO_CSUM6_RX,
-			!!(vxlan->cfg.flags & VXLAN_F_UDP_ZERO_CSUM6_RX)) ||
+		       !!(vxlan->cfg.flags & VXLAN_F_UDP_ZERO_CSUM6_RX)) ||
 	    nla_put_u8(skb, IFLA_VXLAN_REMCSUM_TX,
-			!!(vxlan->cfg.flags & VXLAN_F_REMCSUM_TX)) ||
+		       !!(vxlan->cfg.flags & VXLAN_F_REMCSUM_TX)) ||
 	    nla_put_u8(skb, IFLA_VXLAN_REMCSUM_RX,
-			!!(vxlan->cfg.flags & VXLAN_F_REMCSUM_RX)))
+		       !!(vxlan->cfg.flags & VXLAN_F_REMCSUM_RX)))
 		goto nla_put_failure;
 
 	if (nla_put(skb, IFLA_VXLAN_PORT_RANGE, sizeof(ports), &ports))
-- 
cgit v1.2.3-59-g8ed1b


From cea9760950a50fa13b6535fbaa6fa91223394431 Mon Sep 17 00:00:00 2001
From: David Ahern <dsahern@gmail.com>
Date: Mon, 30 Dec 2019 14:14:25 -0800
Subject: ipv4/tcp: Use local variable for tcp_md5_addr

Extract the typecast to (union tcp_md5_addr *) to a local variable
rather than the current long, inline declaration with function calls.

No functional change intended.

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/tcp_ipv4.c | 43 ++++++++++++++++++++++++++-----------------
 1 file changed, 26 insertions(+), 17 deletions(-)

diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 6f52e5288a6f..c2bfff528578 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -701,9 +701,13 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb)
 	rcu_read_lock();
 	hash_location = tcp_parse_md5sig_option(th);
 	if (sk && sk_fullsock(sk)) {
-		key = tcp_md5_do_lookup(sk, (union tcp_md5_addr *)
-					&ip_hdr(skb)->saddr, AF_INET);
+		const union tcp_md5_addr *addr;
+
+		addr = (union tcp_md5_addr *)&ip_hdr(skb)->saddr;
+		key = tcp_md5_do_lookup(sk, addr, AF_INET);
 	} else if (hash_location) {
+		const union tcp_md5_addr *addr;
+
 		/*
 		 * active side is lost. Try to find listening socket through
 		 * source port, and then find md5 key through listening socket.
@@ -720,8 +724,8 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb)
 		if (!sk1)
 			goto out;
 
-		key = tcp_md5_do_lookup(sk1, (union tcp_md5_addr *)
-					&ip_hdr(skb)->saddr, AF_INET);
+		addr = (union tcp_md5_addr *)&ip_hdr(skb)->saddr;
+		key = tcp_md5_do_lookup(sk1, addr, AF_INET);
 		if (!key)
 			goto out;
 
@@ -905,6 +909,8 @@ static void tcp_v4_timewait_ack(struct sock *sk, struct sk_buff *skb)
 static void tcp_v4_reqsk_send_ack(const struct sock *sk, struct sk_buff *skb,
 				  struct request_sock *req)
 {
+	const union tcp_md5_addr *addr;
+
 	/* sk->sk_state == TCP_LISTEN -> for regular TCP_SYN_RECV
 	 * sk->sk_state == TCP_SYN_RECV -> for Fast Open.
 	 */
@@ -916,14 +922,14 @@ static void tcp_v4_reqsk_send_ack(const struct sock *sk, struct sk_buff *skb,
 	 * exception of <SYN> segments, MUST be right-shifted by
 	 * Rcv.Wind.Shift bits:
 	 */
+	addr = (union tcp_md5_addr *)&ip_hdr(skb)->saddr;
 	tcp_v4_send_ack(sk, skb, seq,
 			tcp_rsk(req)->rcv_nxt,
 			req->rsk_rcv_wnd >> inet_rsk(req)->rcv_wscale,
 			tcp_time_stamp_raw() + tcp_rsk(req)->ts_off,
 			req->ts_recent,
 			0,
-			tcp_md5_do_lookup(sk, (union tcp_md5_addr *)&ip_hdr(skb)->saddr,
-					  AF_INET),
+			tcp_md5_do_lookup(sk, addr, AF_INET),
 			inet_rsk(req)->no_srccheck ? IP_REPLY_ARG_NOSRCCHECK : 0,
 			ip_hdr(skb)->tos);
 }
@@ -1149,6 +1155,7 @@ static int tcp_v4_parse_md5_keys(struct sock *sk, int optname,
 {
 	struct tcp_md5sig cmd;
 	struct sockaddr_in *sin = (struct sockaddr_in *)&cmd.tcpm_addr;
+	const union tcp_md5_addr *addr;
 	u8 prefixlen = 32;
 
 	if (optlen < sizeof(cmd))
@@ -1167,16 +1174,16 @@ static int tcp_v4_parse_md5_keys(struct sock *sk, int optname,
 			return -EINVAL;
 	}
 
+	addr = (union tcp_md5_addr *)&sin->sin_addr.s_addr;
+
 	if (!cmd.tcpm_keylen)
-		return tcp_md5_do_del(sk, (union tcp_md5_addr *)&sin->sin_addr.s_addr,
-				      AF_INET, prefixlen);
+		return tcp_md5_do_del(sk, addr, AF_INET, prefixlen);
 
 	if (cmd.tcpm_keylen > TCP_MD5SIG_MAXKEYLEN)
 		return -EINVAL;
 
-	return tcp_md5_do_add(sk, (union tcp_md5_addr *)&sin->sin_addr.s_addr,
-			      AF_INET, prefixlen, cmd.tcpm_key, cmd.tcpm_keylen,
-			      GFP_KERNEL);
+	return tcp_md5_do_add(sk, addr, AF_INET, prefixlen,
+			      cmd.tcpm_key, cmd.tcpm_keylen, GFP_KERNEL);
 }
 
 static int tcp_v4_md5_hash_headers(struct tcp_md5sig_pool *hp,
@@ -1301,11 +1308,12 @@ static bool tcp_v4_inbound_md5_hash(const struct sock *sk,
 	struct tcp_md5sig_key *hash_expected;
 	const struct iphdr *iph = ip_hdr(skb);
 	const struct tcphdr *th = tcp_hdr(skb);
+	const union tcp_md5_addr *addr;
 	int genhash;
 	unsigned char newhash[16];
 
-	hash_expected = tcp_md5_do_lookup(sk, (union tcp_md5_addr *)&iph->saddr,
-					  AF_INET);
+	addr = (union tcp_md5_addr *)&iph->saddr;
+	hash_expected = tcp_md5_do_lookup(sk, addr, AF_INET);
 	hash_location = tcp_parse_md5sig_option(th);
 
 	/* We've parsed the options - do we have a hash? */
@@ -1419,6 +1427,7 @@ struct sock *tcp_v4_syn_recv_sock(const struct sock *sk, struct sk_buff *skb,
 	struct tcp_sock *newtp;
 	struct sock *newsk;
 #ifdef CONFIG_TCP_MD5SIG
+	const union tcp_md5_addr *addr;
 	struct tcp_md5sig_key *key;
 #endif
 	struct ip_options_rcu *inet_opt;
@@ -1468,8 +1477,8 @@ struct sock *tcp_v4_syn_recv_sock(const struct sock *sk, struct sk_buff *skb,
 
 #ifdef CONFIG_TCP_MD5SIG
 	/* Copy over the MD5 key from the original socket */
-	key = tcp_md5_do_lookup(sk, (union tcp_md5_addr *)&newinet->inet_daddr,
-				AF_INET);
+	addr = (union tcp_md5_addr *)&newinet->inet_daddr;
+	key = tcp_md5_do_lookup(sk, addr, AF_INET);
 	if (key) {
 		/*
 		 * We're using one, so create a matching key
@@ -1477,8 +1486,8 @@ struct sock *tcp_v4_syn_recv_sock(const struct sock *sk, struct sk_buff *skb,
 		 * memory, then we end up not copying the key
 		 * across. Shucks.
 		 */
-		tcp_md5_do_add(newsk, (union tcp_md5_addr *)&newinet->inet_daddr,
-			       AF_INET, 32, key->key, key->keylen, GFP_ATOMIC);
+		tcp_md5_do_add(newsk, addr, AF_INET, 32,
+			       key->key, key->keylen, GFP_ATOMIC);
 		sk_nocaps_add(newsk, NETIF_F_GSO_MASK);
 	}
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From d14c77e0b24f6d1394d9bd4c4ba629808b656baf Mon Sep 17 00:00:00 2001
From: David Ahern <dsahern@gmail.com>
Date: Mon, 30 Dec 2019 14:14:26 -0800
Subject: ipv6/tcp: Pass dif and sdif to tcp_v6_inbound_md5_hash

The original ingress device index is saved to the cb space of the skb
and the cb is moved during tcp processing. Since tcp_v6_inbound_md5_hash
can be called before and after the cb move, pass dif and sdif to it so
the caller can save both prior to the cb move. Both are used by a later
patch.

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/tcp_ipv6.c | 15 +++++++++------
 1 file changed, 9 insertions(+), 6 deletions(-)

diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index df5fd9109696..e94f23b61b62 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -698,7 +698,8 @@ clear_hash_noput:
 #endif
 
 static bool tcp_v6_inbound_md5_hash(const struct sock *sk,
-				    const struct sk_buff *skb)
+				    const struct sk_buff *skb,
+				    int dif, int sdif)
 {
 #ifdef CONFIG_TCP_MD5SIG
 	const __u8 *hash_location = NULL;
@@ -953,6 +954,9 @@ static void tcp_v6_send_reset(const struct sock *sk, struct sk_buff *skb)
 	if (sk && sk_fullsock(sk)) {
 		key = tcp_v6_md5_do_lookup(sk, &ipv6h->saddr);
 	} else if (hash_location) {
+		int dif = tcp_v6_iif_l3_slave(skb);
+		int sdif = tcp_v6_sdif(skb);
+
 		/*
 		 * active side is lost. Try to find listening socket through
 		 * source port, and then find md5 key through listening socket.
@@ -964,9 +968,7 @@ static void tcp_v6_send_reset(const struct sock *sk, struct sk_buff *skb)
 					   &tcp_hashinfo, NULL, 0,
 					   &ipv6h->saddr,
 					   th->source, &ipv6h->daddr,
-					   ntohs(th->source),
-					   tcp_v6_iif_l3_slave(skb),
-					   tcp_v6_sdif(skb));
+					   ntohs(th->source), dif, sdif);
 		if (!sk1)
 			goto out;
 
@@ -1480,6 +1482,7 @@ INDIRECT_CALLABLE_SCOPE int tcp_v6_rcv(struct sk_buff *skb)
 {
 	struct sk_buff *skb_to_free;
 	int sdif = inet6_sdif(skb);
+	int dif = inet6_iif(skb);
 	const struct tcphdr *th;
 	const struct ipv6hdr *hdr;
 	bool refcounted;
@@ -1528,7 +1531,7 @@ process:
 		struct sock *nsk;
 
 		sk = req->rsk_listener;
-		if (tcp_v6_inbound_md5_hash(sk, skb)) {
+		if (tcp_v6_inbound_md5_hash(sk, skb, dif, sdif)) {
 			sk_drops_add(sk, skb);
 			reqsk_put(req);
 			goto discard_it;
@@ -1583,7 +1586,7 @@ process:
 	if (!xfrm6_policy_check(sk, XFRM_POLICY_IN, skb))
 		goto discard_and_relse;
 
-	if (tcp_v6_inbound_md5_hash(sk, skb))
+	if (tcp_v6_inbound_md5_hash(sk, skb, dif, sdif))
 		goto discard_and_relse;
 
 	if (tcp_filter(sk, skb))
-- 
cgit v1.2.3-59-g8ed1b


From 534322ca3daf56a27c9712ae9b1fef4daede5516 Mon Sep 17 00:00:00 2001
From: David Ahern <dsahern@gmail.com>
Date: Mon, 30 Dec 2019 14:14:27 -0800
Subject: ipv4/tcp: Pass dif and sdif to tcp_v4_inbound_md5_hash

The original ingress device index is saved to the cb space of the skb
and the cb is moved during tcp processing. Since tcp_v4_inbound_md5_hash
can be called before and after the cb move, pass dif and sdif to it so
the caller can save both prior to the cb move. Both are used by a later
patch.

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/tcp_ipv4.c | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index c2bfff528578..93f220793add 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -707,6 +707,8 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb)
 		key = tcp_md5_do_lookup(sk, addr, AF_INET);
 	} else if (hash_location) {
 		const union tcp_md5_addr *addr;
+		int sdif = tcp_v4_sdif(skb);
+		int dif = inet_iif(skb);
 
 		/*
 		 * active side is lost. Try to find listening socket through
@@ -718,8 +720,7 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb)
 		sk1 = __inet_lookup_listener(net, &tcp_hashinfo, NULL, 0,
 					     ip_hdr(skb)->saddr,
 					     th->source, ip_hdr(skb)->daddr,
-					     ntohs(th->source), inet_iif(skb),
-					     tcp_v4_sdif(skb));
+					     ntohs(th->source), dif, sdif);
 		/* don't send rst if it can't find key */
 		if (!sk1)
 			goto out;
@@ -1293,7 +1294,8 @@ EXPORT_SYMBOL(tcp_v4_md5_hash_skb);
 
 /* Called with rcu_read_lock() */
 static bool tcp_v4_inbound_md5_hash(const struct sock *sk,
-				    const struct sk_buff *skb)
+				    const struct sk_buff *skb,
+				    int dif, int sdif)
 {
 #ifdef CONFIG_TCP_MD5SIG
 	/*
@@ -1817,6 +1819,7 @@ int tcp_v4_rcv(struct sk_buff *skb)
 	struct net *net = dev_net(skb->dev);
 	struct sk_buff *skb_to_free;
 	int sdif = inet_sdif(skb);
+	int dif = inet_iif(skb);
 	const struct iphdr *iph;
 	const struct tcphdr *th;
 	bool refcounted;
@@ -1865,7 +1868,7 @@ process:
 		struct sock *nsk;
 
 		sk = req->rsk_listener;
-		if (unlikely(tcp_v4_inbound_md5_hash(sk, skb))) {
+		if (unlikely(tcp_v4_inbound_md5_hash(sk, skb, dif, sdif))) {
 			sk_drops_add(sk, skb);
 			reqsk_put(req);
 			goto discard_it;
@@ -1923,7 +1926,7 @@ process:
 	if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb))
 		goto discard_and_relse;
 
-	if (tcp_v4_inbound_md5_hash(sk, skb))
+	if (tcp_v4_inbound_md5_hash(sk, skb, dif, sdif))
 		goto discard_and_relse;
 
 	nf_reset_ct(skb);
-- 
cgit v1.2.3-59-g8ed1b


From dea53bb80e07b9e1641b865493908c20cb8df2ac Mon Sep 17 00:00:00 2001
From: David Ahern <dsahern@gmail.com>
Date: Mon, 30 Dec 2019 14:14:28 -0800
Subject: tcp: Add l3index to tcp_md5sig_key and md5 functions

Add l3index to tcp_md5sig_key to represent the L3 domain of a key, and
add l3index to tcp_md5_do_add and tcp_md5_do_del to fill in the key.

With the key now based on an l3index, add the new parameter to the
lookup functions and consider the l3index when looking for a match.

The l3index comes from the skb when processing ingress packets leveraging
the helpers created for socket lookups, tcp_v4_sdif and inet_iif (and the
v6 variants). When the sdif index is set it means the packet ingressed a
device that is part of an L3 domain and inet_iif points to the VRF device.
For egress, the L3 domain is determined from the socket binding and
sk_bound_dev_if.

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h   | 24 +++++++++---------
 net/ipv4/tcp_ipv4.c | 68 +++++++++++++++++++++++++++++++++++---------------
 net/ipv6/tcp_ipv6.c | 72 +++++++++++++++++++++++++++++++++++++++--------------
 3 files changed, 113 insertions(+), 51 deletions(-)

diff --git a/include/net/tcp.h b/include/net/tcp.h
index e460ea7f767b..7df37e2fddca 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1532,8 +1532,9 @@ struct tcp_md5sig_key {
 	struct hlist_node	node;
 	u8			keylen;
 	u8			family; /* AF_INET or AF_INET6 */
-	union tcp_md5_addr	addr;
 	u8			prefixlen;
+	union tcp_md5_addr	addr;
+	int			l3index; /* set if key added with L3 scope */
 	u8			key[TCP_MD5SIG_MAXKEYLEN];
 	struct rcu_head		rcu;
 };
@@ -1577,34 +1578,33 @@ struct tcp_md5sig_pool {
 int tcp_v4_md5_hash_skb(char *md5_hash, const struct tcp_md5sig_key *key,
 			const struct sock *sk, const struct sk_buff *skb);
 int tcp_md5_do_add(struct sock *sk, const union tcp_md5_addr *addr,
-		   int family, u8 prefixlen, const u8 *newkey, u8 newkeylen,
-		   gfp_t gfp);
+		   int family, u8 prefixlen, int l3index,
+		   const u8 *newkey, u8 newkeylen, gfp_t gfp);
 int tcp_md5_do_del(struct sock *sk, const union tcp_md5_addr *addr,
-		   int family, u8 prefixlen);
+		   int family, u8 prefixlen, int l3index);
 struct tcp_md5sig_key *tcp_v4_md5_lookup(const struct sock *sk,
 					 const struct sock *addr_sk);
 
 #ifdef CONFIG_TCP_MD5SIG
 #include <linux/jump_label.h>
 extern struct static_key_false tcp_md5_needed;
-struct tcp_md5sig_key *__tcp_md5_do_lookup(const struct sock *sk,
+struct tcp_md5sig_key *__tcp_md5_do_lookup(const struct sock *sk, int l3index,
 					   const union tcp_md5_addr *addr,
 					   int family);
 static inline struct tcp_md5sig_key *
-tcp_md5_do_lookup(const struct sock *sk,
-		  const union tcp_md5_addr *addr,
-		  int family)
+tcp_md5_do_lookup(const struct sock *sk, int l3index,
+		  const union tcp_md5_addr *addr, int family)
 {
 	if (!static_branch_unlikely(&tcp_md5_needed))
 		return NULL;
-	return __tcp_md5_do_lookup(sk, addr, family);
+	return __tcp_md5_do_lookup(sk, l3index, addr, family);
 }
 
 #define tcp_twsk_md5_key(twsk)	((twsk)->tw_md5_key)
 #else
-static inline struct tcp_md5sig_key *tcp_md5_do_lookup(const struct sock *sk,
-					 const union tcp_md5_addr *addr,
-					 int family)
+static inline struct tcp_md5sig_key *
+tcp_md5_do_lookup(const struct sock *sk, int l3index,
+		  const union tcp_md5_addr *addr, int family)
 {
 	return NULL;
 }
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 93f220793add..30b3f19d6301 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -702,13 +702,19 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb)
 	hash_location = tcp_parse_md5sig_option(th);
 	if (sk && sk_fullsock(sk)) {
 		const union tcp_md5_addr *addr;
+		int l3index;
 
+		/* sdif set, means packet ingressed via a device
+		 * in an L3 domain and inet_iif is set to it.
+		 */
+		l3index = tcp_v4_sdif(skb) ? inet_iif(skb) : 0;
 		addr = (union tcp_md5_addr *)&ip_hdr(skb)->saddr;
-		key = tcp_md5_do_lookup(sk, addr, AF_INET);
+		key = tcp_md5_do_lookup(sk, l3index, addr, AF_INET);
 	} else if (hash_location) {
 		const union tcp_md5_addr *addr;
 		int sdif = tcp_v4_sdif(skb);
 		int dif = inet_iif(skb);
+		int l3index;
 
 		/*
 		 * active side is lost. Try to find listening socket through
@@ -725,8 +731,12 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb)
 		if (!sk1)
 			goto out;
 
+		/* sdif set, means packet ingressed via a device
+		 * in an L3 domain and dif is set to it.
+		 */
+		l3index = sdif ? dif : 0;
 		addr = (union tcp_md5_addr *)&ip_hdr(skb)->saddr;
-		key = tcp_md5_do_lookup(sk1, addr, AF_INET);
+		key = tcp_md5_do_lookup(sk1, l3index, addr, AF_INET);
 		if (!key)
 			goto out;
 
@@ -911,6 +921,7 @@ static void tcp_v4_reqsk_send_ack(const struct sock *sk, struct sk_buff *skb,
 				  struct request_sock *req)
 {
 	const union tcp_md5_addr *addr;
+	int l3index;
 
 	/* sk->sk_state == TCP_LISTEN -> for regular TCP_SYN_RECV
 	 * sk->sk_state == TCP_SYN_RECV -> for Fast Open.
@@ -924,13 +935,14 @@ static void tcp_v4_reqsk_send_ack(const struct sock *sk, struct sk_buff *skb,
 	 * Rcv.Wind.Shift bits:
 	 */
 	addr = (union tcp_md5_addr *)&ip_hdr(skb)->saddr;
+	l3index = tcp_v4_sdif(skb) ? inet_iif(skb) : 0;
 	tcp_v4_send_ack(sk, skb, seq,
 			tcp_rsk(req)->rcv_nxt,
 			req->rsk_rcv_wnd >> inet_rsk(req)->rcv_wscale,
 			tcp_time_stamp_raw() + tcp_rsk(req)->ts_off,
 			req->ts_recent,
 			0,
-			tcp_md5_do_lookup(sk, addr, AF_INET),
+			tcp_md5_do_lookup(sk, l3index, addr, AF_INET),
 			inet_rsk(req)->no_srccheck ? IP_REPLY_ARG_NOSRCCHECK : 0,
 			ip_hdr(skb)->tos);
 }
@@ -990,7 +1002,7 @@ DEFINE_STATIC_KEY_FALSE(tcp_md5_needed);
 EXPORT_SYMBOL(tcp_md5_needed);
 
 /* Find the Key structure for an address.  */
-struct tcp_md5sig_key *__tcp_md5_do_lookup(const struct sock *sk,
+struct tcp_md5sig_key *__tcp_md5_do_lookup(const struct sock *sk, int l3index,
 					   const union tcp_md5_addr *addr,
 					   int family)
 {
@@ -1010,7 +1022,8 @@ struct tcp_md5sig_key *__tcp_md5_do_lookup(const struct sock *sk,
 	hlist_for_each_entry_rcu(key, &md5sig->head, node) {
 		if (key->family != family)
 			continue;
-
+		if (key->l3index && key->l3index != l3index)
+			continue;
 		if (family == AF_INET) {
 			mask = inet_make_mask(key->prefixlen);
 			match = (key->addr.a4.s_addr & mask) ==
@@ -1034,7 +1047,8 @@ EXPORT_SYMBOL(__tcp_md5_do_lookup);
 
 static struct tcp_md5sig_key *tcp_md5_do_lookup_exact(const struct sock *sk,
 						      const union tcp_md5_addr *addr,
-						      int family, u8 prefixlen)
+						      int family, u8 prefixlen,
+						      int l3index)
 {
 	const struct tcp_sock *tp = tcp_sk(sk);
 	struct tcp_md5sig_key *key;
@@ -1053,6 +1067,8 @@ static struct tcp_md5sig_key *tcp_md5_do_lookup_exact(const struct sock *sk,
 	hlist_for_each_entry_rcu(key, &md5sig->head, node) {
 		if (key->family != family)
 			continue;
+		if (key->l3index && key->l3index != l3index)
+			continue;
 		if (!memcmp(&key->addr, addr, size) &&
 		    key->prefixlen == prefixlen)
 			return key;
@@ -1064,23 +1080,26 @@ struct tcp_md5sig_key *tcp_v4_md5_lookup(const struct sock *sk,
 					 const struct sock *addr_sk)
 {
 	const union tcp_md5_addr *addr;
+	int l3index;
 
+	l3index = l3mdev_master_ifindex_by_index(sock_net(sk),
+						 addr_sk->sk_bound_dev_if);
 	addr = (const union tcp_md5_addr *)&addr_sk->sk_daddr;
-	return tcp_md5_do_lookup(sk, addr, AF_INET);
+	return tcp_md5_do_lookup(sk, l3index, addr, AF_INET);
 }
 EXPORT_SYMBOL(tcp_v4_md5_lookup);
 
 /* This can be called on a newly created socket, from other files */
 int tcp_md5_do_add(struct sock *sk, const union tcp_md5_addr *addr,
-		   int family, u8 prefixlen, const u8 *newkey, u8 newkeylen,
-		   gfp_t gfp)
+		   int family, u8 prefixlen, int l3index,
+		   const u8 *newkey, u8 newkeylen, gfp_t gfp)
 {
 	/* Add Key to the list */
 	struct tcp_md5sig_key *key;
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct tcp_md5sig_info *md5sig;
 
-	key = tcp_md5_do_lookup_exact(sk, addr, family, prefixlen);
+	key = tcp_md5_do_lookup_exact(sk, addr, family, prefixlen, l3index);
 	if (key) {
 		/* Pre-existing entry - just update that one. */
 		memcpy(key->key, newkey, newkeylen);
@@ -1112,6 +1131,7 @@ int tcp_md5_do_add(struct sock *sk, const union tcp_md5_addr *addr,
 	key->keylen = newkeylen;
 	key->family = family;
 	key->prefixlen = prefixlen;
+	key->l3index = l3index;
 	memcpy(&key->addr, addr,
 	       (family == AF_INET6) ? sizeof(struct in6_addr) :
 				      sizeof(struct in_addr));
@@ -1121,11 +1141,11 @@ int tcp_md5_do_add(struct sock *sk, const union tcp_md5_addr *addr,
 EXPORT_SYMBOL(tcp_md5_do_add);
 
 int tcp_md5_do_del(struct sock *sk, const union tcp_md5_addr *addr, int family,
-		   u8 prefixlen)
+		   u8 prefixlen, int l3index)
 {
 	struct tcp_md5sig_key *key;
 
-	key = tcp_md5_do_lookup_exact(sk, addr, family, prefixlen);
+	key = tcp_md5_do_lookup_exact(sk, addr, family, prefixlen, l3index);
 	if (!key)
 		return -ENOENT;
 	hlist_del_rcu(&key->node);
@@ -1158,6 +1178,7 @@ static int tcp_v4_parse_md5_keys(struct sock *sk, int optname,
 	struct sockaddr_in *sin = (struct sockaddr_in *)&cmd.tcpm_addr;
 	const union tcp_md5_addr *addr;
 	u8 prefixlen = 32;
+	int l3index = 0;
 
 	if (optlen < sizeof(cmd))
 		return -EINVAL;
@@ -1178,12 +1199,12 @@ static int tcp_v4_parse_md5_keys(struct sock *sk, int optname,
 	addr = (union tcp_md5_addr *)&sin->sin_addr.s_addr;
 
 	if (!cmd.tcpm_keylen)
-		return tcp_md5_do_del(sk, addr, AF_INET, prefixlen);
+		return tcp_md5_do_del(sk, addr, AF_INET, prefixlen, l3index);
 
 	if (cmd.tcpm_keylen > TCP_MD5SIG_MAXKEYLEN)
 		return -EINVAL;
 
-	return tcp_md5_do_add(sk, addr, AF_INET, prefixlen,
+	return tcp_md5_do_add(sk, addr, AF_INET, prefixlen, l3index,
 			      cmd.tcpm_key, cmd.tcpm_keylen, GFP_KERNEL);
 }
 
@@ -1311,11 +1332,16 @@ static bool tcp_v4_inbound_md5_hash(const struct sock *sk,
 	const struct iphdr *iph = ip_hdr(skb);
 	const struct tcphdr *th = tcp_hdr(skb);
 	const union tcp_md5_addr *addr;
-	int genhash;
 	unsigned char newhash[16];
+	int genhash, l3index;
+
+	/* sdif set, means packet ingressed via a device
+	 * in an L3 domain and dif is set to the l3mdev
+	 */
+	l3index = sdif ? dif : 0;
 
 	addr = (union tcp_md5_addr *)&iph->saddr;
-	hash_expected = tcp_md5_do_lookup(sk, addr, AF_INET);
+	hash_expected = tcp_md5_do_lookup(sk, l3index, addr, AF_INET);
 	hash_location = tcp_parse_md5sig_option(th);
 
 	/* We've parsed the options - do we have a hash? */
@@ -1341,11 +1367,11 @@ static bool tcp_v4_inbound_md5_hash(const struct sock *sk,
 
 	if (genhash || memcmp(hash_location, newhash, 16) != 0) {
 		NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMD5FAILURE);
-		net_info_ratelimited("MD5 Hash failed for (%pI4, %d)->(%pI4, %d)%s\n",
+		net_info_ratelimited("MD5 Hash failed for (%pI4, %d)->(%pI4, %d)%s L3 index %d\n",
 				     &iph->saddr, ntohs(th->source),
 				     &iph->daddr, ntohs(th->dest),
 				     genhash ? " tcp_v4_calc_md5_hash failed"
-				     : "");
+				     : "", l3index);
 		return true;
 	}
 	return false;
@@ -1431,6 +1457,7 @@ struct sock *tcp_v4_syn_recv_sock(const struct sock *sk, struct sk_buff *skb,
 #ifdef CONFIG_TCP_MD5SIG
 	const union tcp_md5_addr *addr;
 	struct tcp_md5sig_key *key;
+	int l3index;
 #endif
 	struct ip_options_rcu *inet_opt;
 
@@ -1478,9 +1505,10 @@ struct sock *tcp_v4_syn_recv_sock(const struct sock *sk, struct sk_buff *skb,
 	tcp_initialize_rcv_mss(newsk);
 
 #ifdef CONFIG_TCP_MD5SIG
+	l3index = l3mdev_master_ifindex_by_index(sock_net(sk), ireq->ir_iif);
 	/* Copy over the MD5 key from the original socket */
 	addr = (union tcp_md5_addr *)&newinet->inet_daddr;
-	key = tcp_md5_do_lookup(sk, addr, AF_INET);
+	key = tcp_md5_do_lookup(sk, l3index, addr, AF_INET);
 	if (key) {
 		/*
 		 * We're using one, so create a matching key
@@ -1488,7 +1516,7 @@ struct sock *tcp_v4_syn_recv_sock(const struct sock *sk, struct sk_buff *skb,
 		 * memory, then we end up not copying the key
 		 * across. Shucks.
 		 */
-		tcp_md5_do_add(newsk, addr, AF_INET, 32,
+		tcp_md5_do_add(newsk, addr, AF_INET, 32, l3index,
 			       key->key, key->keylen, GFP_ATOMIC);
 		sk_nocaps_add(newsk, NETIF_F_GSO_MASK);
 	}
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index e94f23b61b62..71ad7d89be0f 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -81,7 +81,8 @@ static const struct tcp_sock_af_ops tcp_sock_ipv6_specific;
 static const struct tcp_sock_af_ops tcp_sock_ipv6_mapped_specific;
 #else
 static struct tcp_md5sig_key *tcp_v6_md5_do_lookup(const struct sock *sk,
-						   const struct in6_addr *addr)
+						   const struct in6_addr *addr,
+						   int l3index)
 {
 	return NULL;
 }
@@ -532,15 +533,22 @@ static void tcp_v6_reqsk_destructor(struct request_sock *req)
 
 #ifdef CONFIG_TCP_MD5SIG
 static struct tcp_md5sig_key *tcp_v6_md5_do_lookup(const struct sock *sk,
-						   const struct in6_addr *addr)
+						   const struct in6_addr *addr,
+						   int l3index)
 {
-	return tcp_md5_do_lookup(sk, (union tcp_md5_addr *)addr, AF_INET6);
+	return tcp_md5_do_lookup(sk, l3index,
+				 (union tcp_md5_addr *)addr, AF_INET6);
 }
 
 static struct tcp_md5sig_key *tcp_v6_md5_lookup(const struct sock *sk,
 						const struct sock *addr_sk)
 {
-	return tcp_v6_md5_do_lookup(sk, &addr_sk->sk_v6_daddr);
+	int l3index;
+
+	l3index = l3mdev_master_ifindex_by_index(sock_net(sk),
+						 addr_sk->sk_bound_dev_if);
+	return tcp_v6_md5_do_lookup(sk, &addr_sk->sk_v6_daddr,
+				    l3index);
 }
 
 static int tcp_v6_parse_md5_keys(struct sock *sk, int optname,
@@ -548,6 +556,7 @@ static int tcp_v6_parse_md5_keys(struct sock *sk, int optname,
 {
 	struct tcp_md5sig cmd;
 	struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)&cmd.tcpm_addr;
+	int l3index = 0;
 	u8 prefixlen;
 
 	if (optlen < sizeof(cmd))
@@ -572,9 +581,9 @@ static int tcp_v6_parse_md5_keys(struct sock *sk, int optname,
 	if (!cmd.tcpm_keylen) {
 		if (ipv6_addr_v4mapped(&sin6->sin6_addr))
 			return tcp_md5_do_del(sk, (union tcp_md5_addr *)&sin6->sin6_addr.s6_addr32[3],
-					      AF_INET, prefixlen);
+					      AF_INET, prefixlen, l3index);
 		return tcp_md5_do_del(sk, (union tcp_md5_addr *)&sin6->sin6_addr,
-				      AF_INET6, prefixlen);
+				      AF_INET6, prefixlen, l3index);
 	}
 
 	if (cmd.tcpm_keylen > TCP_MD5SIG_MAXKEYLEN)
@@ -582,12 +591,13 @@ static int tcp_v6_parse_md5_keys(struct sock *sk, int optname,
 
 	if (ipv6_addr_v4mapped(&sin6->sin6_addr))
 		return tcp_md5_do_add(sk, (union tcp_md5_addr *)&sin6->sin6_addr.s6_addr32[3],
-				      AF_INET, prefixlen, cmd.tcpm_key,
-				      cmd.tcpm_keylen, GFP_KERNEL);
+				      AF_INET, prefixlen, l3index,
+				      cmd.tcpm_key, cmd.tcpm_keylen,
+				      GFP_KERNEL);
 
 	return tcp_md5_do_add(sk, (union tcp_md5_addr *)&sin6->sin6_addr,
-			      AF_INET6, prefixlen, cmd.tcpm_key,
-			      cmd.tcpm_keylen, GFP_KERNEL);
+			      AF_INET6, prefixlen, l3index,
+			      cmd.tcpm_key, cmd.tcpm_keylen, GFP_KERNEL);
 }
 
 static int tcp_v6_md5_hash_headers(struct tcp_md5sig_pool *hp,
@@ -706,10 +716,15 @@ static bool tcp_v6_inbound_md5_hash(const struct sock *sk,
 	struct tcp_md5sig_key *hash_expected;
 	const struct ipv6hdr *ip6h = ipv6_hdr(skb);
 	const struct tcphdr *th = tcp_hdr(skb);
-	int genhash;
+	int genhash, l3index;
 	u8 newhash[16];
 
-	hash_expected = tcp_v6_md5_do_lookup(sk, &ip6h->saddr);
+	/* sdif set, means packet ingressed via a device
+	 * in an L3 domain and dif is set to the l3mdev
+	 */
+	l3index = sdif ? dif : 0;
+
+	hash_expected = tcp_v6_md5_do_lookup(sk, &ip6h->saddr, l3index);
 	hash_location = tcp_parse_md5sig_option(th);
 
 	/* We've parsed the options - do we have a hash? */
@@ -733,10 +748,10 @@ static bool tcp_v6_inbound_md5_hash(const struct sock *sk,
 
 	if (genhash || memcmp(hash_location, newhash, 16) != 0) {
 		NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMD5FAILURE);
-		net_info_ratelimited("MD5 Hash %s for [%pI6c]:%u->[%pI6c]:%u\n",
+		net_info_ratelimited("MD5 Hash %s for [%pI6c]:%u->[%pI6c]:%u L3 index %d\n",
 				     genhash ? "failed" : "mismatch",
 				     &ip6h->saddr, ntohs(th->source),
-				     &ip6h->daddr, ntohs(th->dest));
+				     &ip6h->daddr, ntohs(th->dest), l3index);
 		return true;
 	}
 #endif
@@ -952,10 +967,17 @@ static void tcp_v6_send_reset(const struct sock *sk, struct sk_buff *skb)
 	rcu_read_lock();
 	hash_location = tcp_parse_md5sig_option(th);
 	if (sk && sk_fullsock(sk)) {
-		key = tcp_v6_md5_do_lookup(sk, &ipv6h->saddr);
+		int l3index;
+
+		/* sdif set, means packet ingressed via a device
+		 * in an L3 domain and inet_iif is set to it.
+		 */
+		l3index = tcp_v6_sdif(skb) ? tcp_v6_iif_l3_slave(skb) : 0;
+		key = tcp_v6_md5_do_lookup(sk, &ipv6h->saddr, l3index);
 	} else if (hash_location) {
 		int dif = tcp_v6_iif_l3_slave(skb);
 		int sdif = tcp_v6_sdif(skb);
+		int l3index;
 
 		/*
 		 * active side is lost. Try to find listening socket through
@@ -972,7 +994,12 @@ static void tcp_v6_send_reset(const struct sock *sk, struct sk_buff *skb)
 		if (!sk1)
 			goto out;
 
-		key = tcp_v6_md5_do_lookup(sk1, &ipv6h->saddr);
+		/* sdif set, means packet ingressed via a device
+		 * in an L3 domain and dif is set to it.
+		 */
+		l3index = tcp_v6_sdif(skb) ? dif : 0;
+
+		key = tcp_v6_md5_do_lookup(sk1, &ipv6h->saddr, l3index);
 		if (!key)
 			goto out;
 
@@ -1042,6 +1069,10 @@ static void tcp_v6_timewait_ack(struct sock *sk, struct sk_buff *skb)
 static void tcp_v6_reqsk_send_ack(const struct sock *sk, struct sk_buff *skb,
 				  struct request_sock *req)
 {
+	int l3index;
+
+	l3index = tcp_v6_sdif(skb) ? tcp_v6_iif_l3_slave(skb) : 0;
+
 	/* sk->sk_state == TCP_LISTEN -> for regular TCP_SYN_RECV
 	 * sk->sk_state == TCP_SYN_RECV -> for Fast Open.
 	 */
@@ -1056,7 +1087,7 @@ static void tcp_v6_reqsk_send_ack(const struct sock *sk, struct sk_buff *skb,
 			req->rsk_rcv_wnd >> inet_rsk(req)->rcv_wscale,
 			tcp_time_stamp_raw() + tcp_rsk(req)->ts_off,
 			req->ts_recent, sk->sk_bound_dev_if,
-			tcp_v6_md5_do_lookup(sk, &ipv6_hdr(skb)->saddr),
+			tcp_v6_md5_do_lookup(sk, &ipv6_hdr(skb)->saddr, l3index),
 			0, 0, sk->sk_priority);
 }
 
@@ -1128,6 +1159,7 @@ static struct sock *tcp_v6_syn_recv_sock(const struct sock *sk, struct sk_buff *
 	struct sock *newsk;
 #ifdef CONFIG_TCP_MD5SIG
 	struct tcp_md5sig_key *key;
+	int l3index;
 #endif
 	struct flowi6 fl6;
 
@@ -1271,8 +1303,10 @@ static struct sock *tcp_v6_syn_recv_sock(const struct sock *sk, struct sk_buff *
 	newinet->inet_rcv_saddr = LOOPBACK4_IPV6;
 
 #ifdef CONFIG_TCP_MD5SIG
+	l3index = l3mdev_master_ifindex_by_index(sock_net(sk), ireq->ir_iif);
+
 	/* Copy over the MD5 key from the original socket */
-	key = tcp_v6_md5_do_lookup(sk, &newsk->sk_v6_daddr);
+	key = tcp_v6_md5_do_lookup(sk, &newsk->sk_v6_daddr, l3index);
 	if (key) {
 		/* We're using one, so create a matching key
 		 * on the newsk structure. If we fail to get
@@ -1280,7 +1314,7 @@ static struct sock *tcp_v6_syn_recv_sock(const struct sock *sk, struct sk_buff *
 		 * across. Shucks.
 		 */
 		tcp_md5_do_add(newsk, (union tcp_md5_addr *)&newsk->sk_v6_daddr,
-			       AF_INET6, 128, key->key, key->keylen,
+			       AF_INET6, 128, l3index, key->key, key->keylen,
 			       sk_gfp_mask(sk, GFP_ATOMIC));
 	}
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From 6b102db50cdde3ba2f78631ed21222edf3a5fb51 Mon Sep 17 00:00:00 2001
From: David Ahern <dsahern@gmail.com>
Date: Mon, 30 Dec 2019 14:14:29 -0800
Subject: net: Add device index to tcp_md5sig

Add support for userspace to specify a device index to limit the scope
of an entry via the TCP_MD5SIG_EXT setsockopt. The existing __tcpm_pad
is renamed to tcpm_ifindex and the new field is only checked if the new
TCP_MD5SIG_FLAG_IFINDEX is set in tcpm_flags. For now, the device index
must point to an L3 master device (e.g., VRF). The API and error
handling are setup to allow the constraint to be relaxed in the future
to any device index.

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/uapi/linux/tcp.h |  5 +++--
 net/ipv4/tcp_ipv4.c      | 18 ++++++++++++++++++
 net/ipv6/tcp_ipv6.c      | 20 +++++++++++++++++++-
 3 files changed, 40 insertions(+), 3 deletions(-)

diff --git a/include/uapi/linux/tcp.h b/include/uapi/linux/tcp.h
index 74af1f759cee..d87184e673ca 100644
--- a/include/uapi/linux/tcp.h
+++ b/include/uapi/linux/tcp.h
@@ -317,14 +317,15 @@ enum {
 #define TCP_MD5SIG_MAXKEYLEN	80
 
 /* tcp_md5sig extension flags for TCP_MD5SIG_EXT */
-#define TCP_MD5SIG_FLAG_PREFIX		1	/* address prefix length */
+#define TCP_MD5SIG_FLAG_PREFIX		0x1	/* address prefix length */
+#define TCP_MD5SIG_FLAG_IFINDEX		0x2	/* ifindex set */
 
 struct tcp_md5sig {
 	struct __kernel_sockaddr_storage tcpm_addr;	/* address associated */
 	__u8	tcpm_flags;				/* extension flags */
 	__u8	tcpm_prefixlen;				/* address prefix */
 	__u16	tcpm_keylen;				/* key length */
-	__u32	__tcpm_pad;				/* zero */
+	int	tcpm_ifindex;				/* device index for scope */
 	__u8	tcpm_key[TCP_MD5SIG_MAXKEYLEN];		/* key (binary) */
 };
 
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 30b3f19d6301..4adac9c75343 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1196,6 +1196,24 @@ static int tcp_v4_parse_md5_keys(struct sock *sk, int optname,
 			return -EINVAL;
 	}
 
+	if (optname == TCP_MD5SIG_EXT &&
+	    cmd.tcpm_flags & TCP_MD5SIG_FLAG_IFINDEX) {
+		struct net_device *dev;
+
+		rcu_read_lock();
+		dev = dev_get_by_index_rcu(sock_net(sk), cmd.tcpm_ifindex);
+		if (dev && netif_is_l3_master(dev))
+			l3index = dev->ifindex;
+
+		rcu_read_unlock();
+
+		/* ok to reference set/not set outside of rcu;
+		 * right now device MUST be an L3 master
+		 */
+		if (!dev || !l3index)
+			return -EINVAL;
+	}
+
 	addr = (union tcp_md5_addr *)&sin->sin_addr.s_addr;
 
 	if (!cmd.tcpm_keylen)
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 71ad7d89be0f..95e4e1e95db2 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -578,10 +578,28 @@ static int tcp_v6_parse_md5_keys(struct sock *sk, int optname,
 		prefixlen = ipv6_addr_v4mapped(&sin6->sin6_addr) ? 32 : 128;
 	}
 
+	if (optname == TCP_MD5SIG_EXT &&
+	    cmd.tcpm_flags & TCP_MD5SIG_FLAG_IFINDEX) {
+		struct net_device *dev;
+
+		rcu_read_lock();
+		dev = dev_get_by_index_rcu(sock_net(sk), cmd.tcpm_ifindex);
+		if (dev && netif_is_l3_master(dev))
+			l3index = dev->ifindex;
+		rcu_read_unlock();
+
+		/* ok to reference set/not set outside of rcu;
+		 * right now device MUST be an L3 master
+		 */
+		if (!dev || !l3index)
+			return -EINVAL;
+	}
+
 	if (!cmd.tcpm_keylen) {
 		if (ipv6_addr_v4mapped(&sin6->sin6_addr))
 			return tcp_md5_do_del(sk, (union tcp_md5_addr *)&sin6->sin6_addr.s6_addr32[3],
-					      AF_INET, prefixlen, l3index);
+					      AF_INET, prefixlen,
+					      l3index);
 		return tcp_md5_do_del(sk, (union tcp_md5_addr *)&sin6->sin6_addr,
 				      AF_INET6, prefixlen, l3index);
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 1bfb45d860d8360322656131eac51772fec6fc8c Mon Sep 17 00:00:00 2001
From: David Ahern <dsahern@gmail.com>
Date: Mon, 30 Dec 2019 14:14:30 -0800
Subject: nettest: Return 1 on MD5 failure for server mode

On failure to set MD5 password, do_server should return 1 so that the
program exits with 1 rather than 255. This used for negative testing
when adding MD5 with device option.

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/net/nettest.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/net/nettest.c b/tools/testing/selftests/net/nettest.c
index c08f4db8330d..bb6bb1ad11e2 100644
--- a/tools/testing/selftests/net/nettest.c
+++ b/tools/testing/selftests/net/nettest.c
@@ -1194,7 +1194,7 @@ static int do_server(struct sock_args *args)
 
 	if (args->password && tcp_md5_remote(lsd, args)) {
 		close(lsd);
-		return -1;
+		return 1;
 	}
 
 	while (1) {
-- 
cgit v1.2.3-59-g8ed1b


From eb09cf03b9bde3a08015b46119d0f594b250001f Mon Sep 17 00:00:00 2001
From: David Ahern <dsahern@gmail.com>
Date: Mon, 30 Dec 2019 14:14:31 -0800
Subject: nettest: Add support for TCP_MD5 extensions

Update nettest to implement TCP_MD5SIG_EXT for a prefix and a device.

Add a new option, -m, to specify a prefix and length to use with MD5
auth. The device option comes from the existing -d option. If either
are set and MD5 auth is requested, TCP_MD5SIG_EXT is used instead of
TCP_MD5SIG.

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/net/nettest.c | 82 +++++++++++++++++++++++++++++------
 1 file changed, 69 insertions(+), 13 deletions(-)

diff --git a/tools/testing/selftests/net/nettest.c b/tools/testing/selftests/net/nettest.c
index bb6bb1ad11e2..93208caacbe6 100644
--- a/tools/testing/selftests/net/nettest.c
+++ b/tools/testing/selftests/net/nettest.c
@@ -74,7 +74,14 @@ struct sock_args {
 	int use_cmsg;
 	const char *dev;
 	int ifindex;
+
 	const char *password;
+	/* prefix for MD5 password */
+	union {
+		struct sockaddr_in v4;
+		struct sockaddr_in6 v6;
+	} md5_prefix;
+	unsigned int prefix_len;
 
 	/* expected addresses and device index for connection */
 	int expected_ifindex;
@@ -200,20 +207,33 @@ static void log_address(const char *desc, struct sockaddr *sa)
 	fflush(stdout);
 }
 
-static int tcp_md5sig(int sd, void *addr, socklen_t alen, const char *password)
+static int tcp_md5sig(int sd, void *addr, socklen_t alen, struct sock_args *args)
 {
-	struct tcp_md5sig md5sig;
-	int keylen = password ? strlen(password) : 0;
+	int keylen = strlen(args->password);
+	struct tcp_md5sig md5sig = {};
+	int opt = TCP_MD5SIG;
 	int rc;
 
-	memset(&md5sig, 0, sizeof(md5sig));
-	memcpy(&md5sig.tcpm_addr, addr, alen);
 	md5sig.tcpm_keylen = keylen;
+	memcpy(md5sig.tcpm_key, args->password, keylen);
 
-	if (keylen)
-		memcpy(md5sig.tcpm_key, password, keylen);
+	if (args->prefix_len) {
+		opt = TCP_MD5SIG_EXT;
+		md5sig.tcpm_flags |= TCP_MD5SIG_FLAG_PREFIX;
+
+		md5sig.tcpm_prefixlen = args->prefix_len;
+		addr = &args->md5_prefix;
+	}
+	memcpy(&md5sig.tcpm_addr, addr, alen);
 
-	rc = setsockopt(sd, IPPROTO_TCP, TCP_MD5SIG, &md5sig, sizeof(md5sig));
+	if (args->ifindex) {
+		opt = TCP_MD5SIG_EXT;
+		md5sig.tcpm_flags |= TCP_MD5SIG_FLAG_IFINDEX;
+
+		md5sig.tcpm_ifindex = args->ifindex;
+	}
+
+	rc = setsockopt(sd, IPPROTO_TCP, opt, &md5sig, sizeof(md5sig));
 	if (rc < 0) {
 		/* ENOENT is harmless. Returned when a password is cleared */
 		if (errno == ENOENT)
@@ -254,7 +274,7 @@ static int tcp_md5_remote(int sd, struct sock_args *args)
 		exit(1);
 	}
 
-	if (tcp_md5sig(sd, addr, alen, args->password))
+	if (tcp_md5sig(sd, addr, alen, args))
 		return -1;
 
 	return 0;
@@ -1313,7 +1333,7 @@ static int connectsock(void *addr, socklen_t alen, struct sock_args *args)
 	if (args->type != SOCK_STREAM)
 		goto out;
 
-	if (args->password && tcp_md5sig(sd, addr, alen, args->password))
+	if (args->password && tcp_md5sig(sd, addr, alen, args))
 		goto err;
 
 	if (args->bind_test_only)
@@ -1405,16 +1425,18 @@ enum addr_type {
 	ADDR_TYPE_MCAST,
 	ADDR_TYPE_EXPECTED_LOCAL,
 	ADDR_TYPE_EXPECTED_REMOTE,
+	ADDR_TYPE_MD5_PREFIX,
 };
 
 static int convert_addr(struct sock_args *args, const char *_str,
 			enum addr_type atype)
 {
+	int pfx_len_max = args->version == AF_INET6 ? 128 : 32;
 	int family = args->version;
+	char *str, *dev, *sep;
 	struct in6_addr *in6;
 	struct in_addr  *in;
 	const char *desc;
-	char *str, *dev;
 	void *addr;
 	int rc = 0;
 
@@ -1443,6 +1465,30 @@ static int convert_addr(struct sock_args *args, const char *_str,
 		desc = "expected remote";
 		addr = &args->expected_raddr;
 		break;
+	case ADDR_TYPE_MD5_PREFIX:
+		desc = "md5 prefix";
+		if (family == AF_INET) {
+			args->md5_prefix.v4.sin_family = AF_INET;
+			addr = &args->md5_prefix.v4.sin_addr;
+		} else if (family == AF_INET6) {
+			args->md5_prefix.v6.sin6_family = AF_INET6;
+			addr = &args->md5_prefix.v6.sin6_addr;
+		} else
+			return 1;
+
+		sep = strchr(str, '/');
+		if (sep) {
+			*sep = '\0';
+			sep++;
+			if (str_to_uint(sep, 1, pfx_len_max,
+					&args->prefix_len) != 0) {
+				fprintf(stderr, "Invalid port\n");
+				return 1;
+			}
+		} else {
+			args->prefix_len = pfx_len_max;
+		}
+		break;
 	default:
 		log_error("unknown address type");
 		exit(1);
@@ -1522,7 +1568,7 @@ static char *random_msg(int len)
 	return m;
 }
 
-#define GETOPT_STR  "sr:l:p:t:g:P:DRn:M:d:SCi6L:0:1:2:Fbq"
+#define GETOPT_STR  "sr:l:p:t:g:P:DRn:M:m:d:SCi6L:0:1:2:Fbq"
 
 static void print_usage(char *prog)
 {
@@ -1551,6 +1597,7 @@ static void print_usage(char *prog)
 	"    -n num        number of times to send message\n"
 	"\n"
 	"    -M password   use MD5 sum protection\n"
+	"    -m prefix/len prefix and length to use for MD5 key\n"
 	"    -g grp        multicast group (e.g., 239.1.1.1)\n"
 	"    -i            interactive mode (default is echo and terminate)\n"
 	"\n"
@@ -1642,6 +1689,10 @@ int main(int argc, char *argv[])
 		case 'M':
 			args.password = optarg;
 			break;
+		case 'm':
+			if (convert_addr(&args, optarg, ADDR_TYPE_MD5_PREFIX) < 0)
+				return 1;
+			break;
 		case 'S':
 			args.use_setsockopt = 1;
 			break;
@@ -1706,11 +1757,16 @@ int main(int argc, char *argv[])
 	}
 
 	if (args.password &&
-	    (!args.has_remote_ip || args.type != SOCK_STREAM)) {
+	    ((!args.has_remote_ip && !args.prefix_len) || args.type != SOCK_STREAM)) {
 		log_error("MD5 passwords apply to TCP only and require a remote ip for the password\n");
 		return 1;
 	}
 
+	if (args.prefix_len && !args.password) {
+		log_error("Prefix range for MD5 protection specified without a password\n");
+		return 1;
+	}
+
 	if ((args.use_setsockopt || args.use_cmsg) && !args.ifindex) {
 		fprintf(stderr, "Device binding not specified\n");
 		return 1;
-- 
cgit v1.2.3-59-g8ed1b


From f0bee1ebb5594e9a9d2b85e365d10038ef16dce9 Mon Sep 17 00:00:00 2001
From: David Ahern <dsahern@gmail.com>
Date: Mon, 30 Dec 2019 14:14:32 -0800
Subject: fcnal-test: Add TCP MD5 tests

Add tests for existing TCP MD5 APIs - both single address
config and the new extended API for prefixes.

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/net/fcnal-test.sh | 145 ++++++++++++++++++++++++++++++
 1 file changed, 145 insertions(+)

diff --git a/tools/testing/selftests/net/fcnal-test.sh b/tools/testing/selftests/net/fcnal-test.sh
index 9fd3a0b97f0d..e630c6a7ee72 100755
--- a/tools/testing/selftests/net/fcnal-test.sh
+++ b/tools/testing/selftests/net/fcnal-test.sh
@@ -45,17 +45,22 @@ VRF_TABLE=1101
 NSA_IP=172.16.1.1
 NSB_IP=172.16.1.2
 VRF_IP=172.16.3.1
+NS_NET=172.16.1.0/24
 
 # IPv6 config
 NSA_IP6=2001:db8:1::1
 NSB_IP6=2001:db8:1::2
 VRF_IP6=2001:db8:3::1
+NS_NET6=2001:db8:1::/120
 
 NSA_LO_IP=172.16.2.1
 NSB_LO_IP=172.16.2.2
 NSA_LO_IP6=2001:db8:2::1
 NSB_LO_IP6=2001:db8:2::2
 
+MD5_PW=abc123
+MD5_WRONG_PW=abc1234
+
 MCAST=ff02::1
 # set after namespace create
 NSA_LINKIP6=
@@ -714,6 +719,74 @@ ipv4_ping()
 ################################################################################
 # IPv4 TCP
 
+#
+# MD5 tests without VRF
+#
+ipv4_tcp_md5_novrf()
+{
+	#
+	# single address
+	#
+
+	# basic use case
+	log_start
+	run_cmd nettest -s -M ${MD5_PW} -r ${NSB_IP} &
+	sleep 1
+	run_cmd_nsb nettest -r ${NSA_IP} -M ${MD5_PW}
+	log_test $? 0 "MD5: Single address config"
+
+	# client sends MD5, server not configured
+	log_start
+	show_hint "Should timeout due to MD5 mismatch"
+	run_cmd nettest -s &
+	sleep 1
+	run_cmd_nsb nettest -r ${NSA_IP} -M ${MD5_PW}
+	log_test $? 2 "MD5: Server no config, client uses password"
+
+	# wrong password
+	log_start
+	show_hint "Should timeout since client uses wrong password"
+	run_cmd nettest -s -M ${MD5_PW} -r ${NSB_IP} &
+	sleep 1
+	run_cmd_nsb nettest -r ${NSA_IP} -M ${MD5_WRONG_PW}
+	log_test $? 2 "MD5: Client uses wrong password"
+
+	# client from different address
+	log_start
+	show_hint "Should timeout due to MD5 mismatch"
+	run_cmd nettest -s -M ${MD5_PW} -r ${NSB_LO_IP} &
+	sleep 1
+	run_cmd_nsb nettest -r ${NSA_IP} -M ${MD5_PW}
+	log_test $? 2 "MD5: Client address does not match address configured with password"
+
+	#
+	# MD5 extension - prefix length
+	#
+
+	# client in prefix
+	log_start
+	run_cmd nettest -s -M ${MD5_PW} -m ${NS_NET} &
+	sleep 1
+	run_cmd_nsb nettest  -r ${NSA_IP} -M ${MD5_PW}
+	log_test $? 0 "MD5: Prefix config"
+
+	# client in prefix, wrong password
+	log_start
+	show_hint "Should timeout since client uses wrong password"
+	run_cmd nettest -s -M ${MD5_PW} -m ${NS_NET} &
+	sleep 1
+	run_cmd_nsb nettest -r ${NSA_IP} -M ${MD5_WRONG_PW}
+	log_test $? 2 "MD5: Prefix config, client uses wrong password"
+
+	# client outside of prefix
+	log_start
+	show_hint "Should timeout due to MD5 mismatch"
+	run_cmd nettest -s -M ${MD5_PW} -m ${NS_NET} &
+	sleep 1
+	run_cmd_nsb nettest -l ${NSB_LO_IP} -r ${NSA_IP} -M ${MD5_PW}
+	log_test $? 2 "MD5: Prefix config, client address not in configured prefix"
+}
+
 ipv4_tcp_novrf()
 {
 	local a
@@ -831,6 +904,8 @@ ipv4_tcp_novrf()
 	show_hint "Should fail 'Connection refused'"
 	run_cmd nettest -d ${NSA_DEV} -r ${a}
 	log_test_addr ${a} $? 1 "No server, device client, local conn"
+
+	ipv4_tcp_md5_novrf
 }
 
 ipv4_tcp_vrf()
@@ -1961,6 +2036,74 @@ ipv6_ping()
 ################################################################################
 # IPv6 TCP
 
+#
+# MD5 tests without VRF
+#
+ipv6_tcp_md5_novrf()
+{
+	#
+	# single address
+	#
+
+	# basic use case
+	log_start
+	run_cmd nettest -6 -s -M ${MD5_PW} -r ${NSB_IP6} &
+	sleep 1
+	run_cmd_nsb nettest -6 -r ${NSA_IP6} -M ${MD5_PW}
+	log_test $? 0 "MD5: Single address config"
+
+	# client sends MD5, server not configured
+	log_start
+	show_hint "Should timeout due to MD5 mismatch"
+	run_cmd nettest -6 -s &
+	sleep 1
+	run_cmd_nsb nettest -6 -r ${NSA_IP6} -M ${MD5_PW}
+	log_test $? 2 "MD5: Server no config, client uses password"
+
+	# wrong password
+	log_start
+	show_hint "Should timeout since client uses wrong password"
+	run_cmd nettest -6 -s -M ${MD5_PW} -r ${NSB_IP6} &
+	sleep 1
+	run_cmd_nsb nettest -6 -r ${NSA_IP6} -M ${MD5_WRONG_PW}
+	log_test $? 2 "MD5: Client uses wrong password"
+
+	# client from different address
+	log_start
+	show_hint "Should timeout due to MD5 mismatch"
+	run_cmd nettest -6 -s -M ${MD5_PW} -r ${NSB_LO_IP6} &
+	sleep 1
+	run_cmd_nsb nettest -6 -r ${NSA_IP6} -M ${MD5_PW}
+	log_test $? 2 "MD5: Client address does not match address configured with password"
+
+	#
+	# MD5 extension - prefix length
+	#
+
+	# client in prefix
+	log_start
+	run_cmd nettest -6 -s -M ${MD5_PW} -m ${NS_NET6} &
+	sleep 1
+	run_cmd_nsb nettest -6  -r ${NSA_IP6} -M ${MD5_PW}
+	log_test $? 0 "MD5: Prefix config"
+
+	# client in prefix, wrong password
+	log_start
+	show_hint "Should timeout since client uses wrong password"
+	run_cmd nettest -6 -s -M ${MD5_PW} -m ${NS_NET6} &
+	sleep 1
+	run_cmd_nsb nettest -6 -r ${NSA_IP6} -M ${MD5_WRONG_PW}
+	log_test $? 2 "MD5: Prefix config, client uses wrong password"
+
+	# client outside of prefix
+	log_start
+	show_hint "Should timeout due to MD5 mismatch"
+	run_cmd nettest -6 -s -M ${MD5_PW} -m ${NS_NET6} &
+	sleep 1
+	run_cmd_nsb nettest -6 -l ${NSB_LO_IP6} -r ${NSA_IP6} -M ${MD5_PW}
+	log_test $? 2 "MD5: Prefix config, client address not in configured prefix"
+}
+
 ipv6_tcp_novrf()
 {
 	local a
@@ -2077,6 +2220,8 @@ ipv6_tcp_novrf()
 		run_cmd nettest -6 -d ${NSA_DEV} -r ${a}
 		log_test_addr ${a} $? 1 "No server, device client, local conn"
 	done
+
+	ipv6_tcp_md5_novrf
 }
 
 ipv6_tcp_vrf()
-- 
cgit v1.2.3-59-g8ed1b


From 5cad8bce26e01238f82d391acc7f70dd83f84a91 Mon Sep 17 00:00:00 2001
From: David Ahern <dsahern@gmail.com>
Date: Mon, 30 Dec 2019 14:14:33 -0800
Subject: fcnal-test: Add TCP MD5 tests for VRF

Add tests for new TCP MD5 API for L3 domains (VRF).

A new namespace is added to create a duplicate configuration between
the VRF and default VRF to verify overlapping config is handled properly.

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/net/fcnal-test.sh | 313 ++++++++++++++++++++++++++++++
 1 file changed, 313 insertions(+)

diff --git a/tools/testing/selftests/net/fcnal-test.sh b/tools/testing/selftests/net/fcnal-test.sh
index e630c6a7ee72..15cb8ab68c7c 100755
--- a/tools/testing/selftests/net/fcnal-test.sh
+++ b/tools/testing/selftests/net/fcnal-test.sh
@@ -32,12 +32,17 @@
 #      lo2: 127.0.0.1/8, ::1/128
 #           172.16.2.2/32, 2001:db8:2::2/128
 #
+# ns-A to ns-C connection - only for VRF and same config
+# as ns-A to ns-B
+#
 # server / client nomenclature relative to ns-A
 
 VERBOSE=0
 
 NSA_DEV=eth1
+NSA_DEV2=eth2
 NSB_DEV=eth1
+NSC_DEV=eth2
 VRF=red
 VRF_TABLE=1101
 
@@ -68,9 +73,11 @@ NSB_LINKIP6=
 
 NSA=ns-A
 NSB=ns-B
+NSC=ns-C
 
 NSA_CMD="ip netns exec ${NSA}"
 NSB_CMD="ip netns exec ${NSB}"
+NSC_CMD="ip netns exec ${NSC}"
 
 which ping6 > /dev/null 2>&1 && ping6=$(which ping6) || ping6=$(which ping)
 
@@ -200,6 +207,11 @@ run_cmd_nsb()
 	do_run_cmd ${NSB_CMD} $*
 }
 
+run_cmd_nsc()
+{
+	do_run_cmd ${NSC_CMD} $*
+}
+
 setup_cmd()
 {
 	local cmd="$*"
@@ -406,6 +418,7 @@ cleanup()
 	fi
 
 	ip netns del ${NSB}
+	ip netns del ${NSC} >/dev/null 2>&1
 }
 
 setup()
@@ -437,6 +450,12 @@ setup()
 
 		ip -netns ${NSB} ro add ${VRF_IP}/32 via ${NSA_IP} dev ${NSB_DEV}
 		ip -netns ${NSB} -6 ro add ${VRF_IP6}/128 via ${NSA_IP6} dev ${NSB_DEV}
+
+		# some VRF tests use ns-C which has the same config as
+		# ns-B but for a device NOT in the VRF
+		create_ns ${NSC} "-" "-"
+		connect_ns ${NSA} ${NSA_DEV2} ${NSA_IP}/24 ${NSA_IP6}/64 \
+			   ${NSC} ${NSC_DEV} ${NSB_IP}/24 ${NSB_IP6}/64
 	else
 		ip -netns ${NSA} ro add ${NSB_LO_IP}/32 via ${NSB_IP} dev ${NSA_DEV}
 		ip -netns ${NSA} ro add ${NSB_LO_IP6}/128 via ${NSB_IP6} dev ${NSA_DEV}
@@ -787,6 +806,150 @@ ipv4_tcp_md5_novrf()
 	log_test $? 2 "MD5: Prefix config, client address not in configured prefix"
 }
 
+#
+# MD5 tests with VRF
+#
+ipv4_tcp_md5()
+{
+	#
+	# single address
+	#
+
+	# basic use case
+	log_start
+	run_cmd nettest -s -d ${VRF} -M ${MD5_PW} -r ${NSB_IP} &
+	sleep 1
+	run_cmd_nsb nettest -r ${NSA_IP} -M ${MD5_PW}
+	log_test $? 0 "MD5: VRF: Single address config"
+
+	# client sends MD5, server not configured
+	log_start
+	show_hint "Should timeout since server does not have MD5 auth"
+	run_cmd nettest -s -d ${VRF} &
+	sleep 1
+	run_cmd_nsb nettest -r ${NSA_IP} -M ${MD5_PW}
+	log_test $? 2 "MD5: VRF: Server no config, client uses password"
+
+	# wrong password
+	log_start
+	show_hint "Should timeout since client uses wrong password"
+	run_cmd nettest -s -d ${VRF} -M ${MD5_PW} -r ${NSB_IP} &
+	sleep 1
+	run_cmd_nsb nettest -r ${NSA_IP} -M ${MD5_WRONG_PW}
+	log_test $? 2 "MD5: VRF: Client uses wrong password"
+
+	# client from different address
+	log_start
+	show_hint "Should timeout since server config differs from client"
+	run_cmd nettest -s -d ${VRF} -M ${MD5_PW} -r ${NSB_LO_IP} &
+	sleep 1
+	run_cmd_nsb nettest -r ${NSA_IP} -M ${MD5_PW}
+	log_test $? 2 "MD5: VRF: Client address does not match address configured with password"
+
+	#
+	# MD5 extension - prefix length
+	#
+
+	# client in prefix
+	log_start
+	run_cmd nettest -s -d ${VRF} -M ${MD5_PW} -m ${NS_NET} &
+	sleep 1
+	run_cmd_nsb nettest  -r ${NSA_IP} -M ${MD5_PW}
+	log_test $? 0 "MD5: VRF: Prefix config"
+
+	# client in prefix, wrong password
+	log_start
+	show_hint "Should timeout since client uses wrong password"
+	run_cmd nettest -s -d ${VRF} -M ${MD5_PW} -m ${NS_NET} &
+	sleep 1
+	run_cmd_nsb nettest -r ${NSA_IP} -M ${MD5_WRONG_PW}
+	log_test $? 2 "MD5: VRF: Prefix config, client uses wrong password"
+
+	# client outside of prefix
+	log_start
+	show_hint "Should timeout since client address is outside of prefix"
+	run_cmd nettest -s -d ${VRF} -M ${MD5_PW} -m ${NS_NET} &
+	sleep 1
+	run_cmd_nsb nettest -l ${NSB_LO_IP} -r ${NSA_IP} -M ${MD5_PW}
+	log_test $? 2 "MD5: VRF: Prefix config, client address not in configured prefix"
+
+	#
+	# duplicate config between default VRF and a VRF
+	#
+
+	log_start
+	run_cmd nettest -s -d ${VRF} -M ${MD5_PW} -r ${NSB_IP} &
+	run_cmd nettest -s -M ${MD5_WRONG_PW} -r ${NSB_IP} &
+	sleep 1
+	run_cmd_nsb nettest  -r ${NSA_IP} -M ${MD5_PW}
+	log_test $? 0 "MD5: VRF: Single address config in default VRF and VRF, conn in VRF"
+
+	log_start
+	run_cmd nettest -s -d ${VRF} -M ${MD5_PW} -r ${NSB_IP} &
+	run_cmd nettest -s -M ${MD5_WRONG_PW} -r ${NSB_IP} &
+	sleep 1
+	run_cmd_nsc nettest  -r ${NSA_IP} -M ${MD5_WRONG_PW}
+	log_test $? 0 "MD5: VRF: Single address config in default VRF and VRF, conn in default VRF"
+
+	log_start
+	show_hint "Should timeout since client in default VRF uses VRF password"
+	run_cmd nettest -s -d ${VRF} -M ${MD5_PW} -r ${NSB_IP} &
+	run_cmd nettest -s -M ${MD5_WRONG_PW} -r ${NSB_IP} &
+	sleep 1
+	run_cmd_nsc nettest -r ${NSA_IP} -M ${MD5_PW}
+	log_test $? 2 "MD5: VRF: Single address config in default VRF and VRF, conn in default VRF with VRF pw"
+
+	log_start
+	show_hint "Should timeout since client in VRF uses default VRF password"
+	run_cmd nettest -s -d ${VRF} -M ${MD5_PW} -r ${NSB_IP} &
+	run_cmd nettest -s -M ${MD5_WRONG_PW} -r ${NSB_IP} &
+	sleep 1
+	run_cmd_nsb nettest -r ${NSA_IP} -M ${MD5_WRONG_PW}
+	log_test $? 2 "MD5: VRF: Single address config in default VRF and VRF, conn in VRF with default VRF pw"
+
+	log_start
+	run_cmd nettest -s -d ${VRF} -M ${MD5_PW} -m ${NS_NET} &
+	run_cmd nettest -s -M ${MD5_WRONG_PW} -m ${NS_NET} &
+	sleep 1
+	run_cmd_nsb nettest  -r ${NSA_IP} -M ${MD5_PW}
+	log_test $? 0 "MD5: VRF: Prefix config in default VRF and VRF, conn in VRF"
+
+	log_start
+	run_cmd nettest -s -d ${VRF} -M ${MD5_PW} -m ${NS_NET} &
+	run_cmd nettest -s -M ${MD5_WRONG_PW} -m ${NS_NET} &
+	sleep 1
+	run_cmd_nsc nettest  -r ${NSA_IP} -M ${MD5_WRONG_PW}
+	log_test $? 0 "MD5: VRF: Prefix config in default VRF and VRF, conn in default VRF"
+
+	log_start
+	show_hint "Should timeout since client in default VRF uses VRF password"
+	run_cmd nettest -s -d ${VRF} -M ${MD5_PW} -m ${NS_NET} &
+	run_cmd nettest -s -M ${MD5_WRONG_PW} -m ${NS_NET} &
+	sleep 1
+	run_cmd_nsc nettest -r ${NSA_IP} -M ${MD5_PW}
+	log_test $? 2 "MD5: VRF: Prefix config in default VRF and VRF, conn in default VRF with VRF pw"
+
+	log_start
+	show_hint "Should timeout since client in VRF uses default VRF password"
+	run_cmd nettest -s -d ${VRF} -M ${MD5_PW} -m ${NS_NET} &
+	run_cmd nettest -s -M ${MD5_WRONG_PW} -m ${NS_NET} &
+	sleep 1
+	run_cmd_nsb nettest -r ${NSA_IP} -M ${MD5_WRONG_PW}
+	log_test $? 2 "MD5: VRF: Prefix config in default VRF and VRF, conn in VRF with default VRF pw"
+
+	#
+	# negative tests
+	#
+	log_start
+	run_cmd nettest -s -d ${NSA_DEV} -M ${MD5_PW} -r ${NSB_IP}
+	log_test $? 1 "MD5: VRF: Device must be a VRF - single address"
+
+	log_start
+	run_cmd nettest -s -d ${NSA_DEV} -M ${MD5_PW} -m ${NS_NET}
+	log_test $? 1 "MD5: VRF: Device must be a VRF - prefix"
+
+}
+
 ipv4_tcp_novrf()
 {
 	local a
@@ -958,6 +1121,9 @@ ipv4_tcp_vrf()
 	run_cmd nettest -r ${a} -d ${NSA_DEV}
 	log_test_addr ${a} $? 1 "Global server, local connection"
 
+	# run MD5 tests
+	ipv4_tcp_md5
+
 	#
 	# enable VRF global server
 	#
@@ -2104,6 +2270,150 @@ ipv6_tcp_md5_novrf()
 	log_test $? 2 "MD5: Prefix config, client address not in configured prefix"
 }
 
+#
+# MD5 tests with VRF
+#
+ipv6_tcp_md5()
+{
+	#
+	# single address
+	#
+
+	# basic use case
+	log_start
+	run_cmd nettest -6 -s -d ${VRF} -M ${MD5_PW} -r ${NSB_IP6} &
+	sleep 1
+	run_cmd_nsb nettest -6 -r ${NSA_IP6} -M ${MD5_PW}
+	log_test $? 0 "MD5: VRF: Single address config"
+
+	# client sends MD5, server not configured
+	log_start
+	show_hint "Should timeout since server does not have MD5 auth"
+	run_cmd nettest -6 -s -d ${VRF} &
+	sleep 1
+	run_cmd_nsb nettest -6 -r ${NSA_IP6} -M ${MD5_PW}
+	log_test $? 2 "MD5: VRF: Server no config, client uses password"
+
+	# wrong password
+	log_start
+	show_hint "Should timeout since client uses wrong password"
+	run_cmd nettest -6 -s -d ${VRF} -M ${MD5_PW} -r ${NSB_IP6} &
+	sleep 1
+	run_cmd_nsb nettest -6 -r ${NSA_IP6} -M ${MD5_WRONG_PW}
+	log_test $? 2 "MD5: VRF: Client uses wrong password"
+
+	# client from different address
+	log_start
+	show_hint "Should timeout since server config differs from client"
+	run_cmd nettest -6 -s -d ${VRF} -M ${MD5_PW} -r ${NSB_LO_IP6} &
+	sleep 1
+	run_cmd_nsb nettest -6 -r ${NSA_IP6} -M ${MD5_PW}
+	log_test $? 2 "MD5: VRF: Client address does not match address configured with password"
+
+	#
+	# MD5 extension - prefix length
+	#
+
+	# client in prefix
+	log_start
+	run_cmd nettest -6 -s -d ${VRF} -M ${MD5_PW} -m ${NS_NET6} &
+	sleep 1
+	run_cmd_nsb nettest -6  -r ${NSA_IP6} -M ${MD5_PW}
+	log_test $? 0 "MD5: VRF: Prefix config"
+
+	# client in prefix, wrong password
+	log_start
+	show_hint "Should timeout since client uses wrong password"
+	run_cmd nettest -6 -s -d ${VRF} -M ${MD5_PW} -m ${NS_NET6} &
+	sleep 1
+	run_cmd_nsb nettest -6 -r ${NSA_IP6} -M ${MD5_WRONG_PW}
+	log_test $? 2 "MD5: VRF: Prefix config, client uses wrong password"
+
+	# client outside of prefix
+	log_start
+	show_hint "Should timeout since client address is outside of prefix"
+	run_cmd nettest -6 -s -d ${VRF} -M ${MD5_PW} -m ${NS_NET6} &
+	sleep 1
+	run_cmd_nsb nettest -6 -l ${NSB_LO_IP6} -r ${NSA_IP6} -M ${MD5_PW}
+	log_test $? 2 "MD5: VRF: Prefix config, client address not in configured prefix"
+
+	#
+	# duplicate config between default VRF and a VRF
+	#
+
+	log_start
+	run_cmd nettest -6 -s -d ${VRF} -M ${MD5_PW} -r ${NSB_IP6} &
+	run_cmd nettest -6 -s -M ${MD5_WRONG_PW} -r ${NSB_IP6} &
+	sleep 1
+	run_cmd_nsb nettest -6  -r ${NSA_IP6} -M ${MD5_PW}
+	log_test $? 0 "MD5: VRF: Single address config in default VRF and VRF, conn in VRF"
+
+	log_start
+	run_cmd nettest -6 -s -d ${VRF} -M ${MD5_PW} -r ${NSB_IP6} &
+	run_cmd nettest -6 -s -M ${MD5_WRONG_PW} -r ${NSB_IP6} &
+	sleep 1
+	run_cmd_nsc nettest -6  -r ${NSA_IP6} -M ${MD5_WRONG_PW}
+	log_test $? 0 "MD5: VRF: Single address config in default VRF and VRF, conn in default VRF"
+
+	log_start
+	show_hint "Should timeout since client in default VRF uses VRF password"
+	run_cmd nettest -6 -s -d ${VRF} -M ${MD5_PW} -r ${NSB_IP6} &
+	run_cmd nettest -6 -s -M ${MD5_WRONG_PW} -r ${NSB_IP6} &
+	sleep 1
+	run_cmd_nsc nettest -6 -r ${NSA_IP6} -M ${MD5_PW}
+	log_test $? 2 "MD5: VRF: Single address config in default VRF and VRF, conn in default VRF with VRF pw"
+
+	log_start
+	show_hint "Should timeout since client in VRF uses default VRF password"
+	run_cmd nettest -6 -s -d ${VRF} -M ${MD5_PW} -r ${NSB_IP6} &
+	run_cmd nettest -6 -s -M ${MD5_WRONG_PW} -r ${NSB_IP6} &
+	sleep 1
+	run_cmd_nsb nettest -6 -r ${NSA_IP6} -M ${MD5_WRONG_PW}
+	log_test $? 2 "MD5: VRF: Single address config in default VRF and VRF, conn in VRF with default VRF pw"
+
+	log_start
+	run_cmd nettest -6 -s -d ${VRF} -M ${MD5_PW} -m ${NS_NET6} &
+	run_cmd nettest -6 -s -M ${MD5_WRONG_PW} -m ${NS_NET6} &
+	sleep 1
+	run_cmd_nsb nettest -6  -r ${NSA_IP6} -M ${MD5_PW}
+	log_test $? 0 "MD5: VRF: Prefix config in default VRF and VRF, conn in VRF"
+
+	log_start
+	run_cmd nettest -6 -s -d ${VRF} -M ${MD5_PW} -m ${NS_NET6} &
+	run_cmd nettest -6 -s -M ${MD5_WRONG_PW} -m ${NS_NET6} &
+	sleep 1
+	run_cmd_nsc nettest -6  -r ${NSA_IP6} -M ${MD5_WRONG_PW}
+	log_test $? 0 "MD5: VRF: Prefix config in default VRF and VRF, conn in default VRF"
+
+	log_start
+	show_hint "Should timeout since client in default VRF uses VRF password"
+	run_cmd nettest -6 -s -d ${VRF} -M ${MD5_PW} -m ${NS_NET6} &
+	run_cmd nettest -6 -s -M ${MD5_WRONG_PW} -m ${NS_NET6} &
+	sleep 1
+	run_cmd_nsc nettest -6 -r ${NSA_IP6} -M ${MD5_PW}
+	log_test $? 2 "MD5: VRF: Prefix config in default VRF and VRF, conn in default VRF with VRF pw"
+
+	log_start
+	show_hint "Should timeout since client in VRF uses default VRF password"
+	run_cmd nettest -6 -s -d ${VRF} -M ${MD5_PW} -m ${NS_NET6} &
+	run_cmd nettest -6 -s -M ${MD5_WRONG_PW} -m ${NS_NET6} &
+	sleep 1
+	run_cmd_nsb nettest -6 -r ${NSA_IP6} -M ${MD5_WRONG_PW}
+	log_test $? 2 "MD5: VRF: Prefix config in default VRF and VRF, conn in VRF with default VRF pw"
+
+	#
+	# negative tests
+	#
+	log_start
+	run_cmd nettest -6 -s -d ${NSA_DEV} -M ${MD5_PW} -r ${NSB_IP6}
+	log_test $? 1 "MD5: VRF: Device must be a VRF - single address"
+
+	log_start
+	run_cmd nettest -6 -s -d ${NSA_DEV} -M ${MD5_PW} -m ${NS_NET6}
+	log_test $? 1 "MD5: VRF: Device must be a VRF - prefix"
+
+}
+
 ipv6_tcp_novrf()
 {
 	local a
@@ -2290,6 +2600,9 @@ ipv6_tcp_vrf()
 	run_cmd nettest -6 -r ${a} -d ${NSA_DEV}
 	log_test_addr ${a} $? 1 "Global server, local connection"
 
+	# run MD5 tests
+	ipv6_tcp_md5
+
 	#
 	# enable VRF global server
 	#
-- 
cgit v1.2.3-59-g8ed1b


From edf4579123e234ada9fe12fbc8436923fcceb963 Mon Sep 17 00:00:00 2001
From: Ben Hutchings <ben@decadent.org.uk>
Date: Tue, 31 Dec 2019 16:59:08 +0000
Subject: sfc: Remove unnecessary dependencies on I2C

Only the SFC4000 code, now moved to sfc-falcon, needed I2C.

Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Acked-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/Kconfig      | 2 --
 drivers/net/ethernet/sfc/net_driver.h | 1 -
 drivers/net/ethernet/sfc/nic.h        | 1 -
 3 files changed, 4 deletions(-)

diff --git a/drivers/net/ethernet/sfc/Kconfig b/drivers/net/ethernet/sfc/Kconfig
index 5f36774bf4b8..ea5a9220196c 100644
--- a/drivers/net/ethernet/sfc/Kconfig
+++ b/drivers/net/ethernet/sfc/Kconfig
@@ -21,8 +21,6 @@ config SFC
 	depends on PCI
 	select MDIO
 	select CRC32
-	select I2C
-	select I2C_ALGOBIT
 	imply PTP_1588_CLOCK
 	---help---
 	  This driver supports 10/40-gigabit Ethernet cards based on
diff --git a/drivers/net/ethernet/sfc/net_driver.h b/drivers/net/ethernet/sfc/net_driver.h
index dfd5182d9e47..709172a6995e 100644
--- a/drivers/net/ethernet/sfc/net_driver.h
+++ b/drivers/net/ethernet/sfc/net_driver.h
@@ -24,7 +24,6 @@
 #include <linux/mutex.h>
 #include <linux/rwsem.h>
 #include <linux/vmalloc.h>
-#include <linux/i2c.h>
 #include <linux/mtd/mtd.h>
 #include <net/busy_poll.h>
 #include <net/xdp.h>
diff --git a/drivers/net/ethernet/sfc/nic.h b/drivers/net/ethernet/sfc/nic.h
index 1f7c5717de75..bf0bdb22cc64 100644
--- a/drivers/net/ethernet/sfc/nic.h
+++ b/drivers/net/ethernet/sfc/nic.h
@@ -9,7 +9,6 @@
 #define EFX_NIC_H
 
 #include <linux/net_tstamp.h>
-#include <linux/i2c-algo-bit.h>
 #include "net_driver.h"
 #include "efx.h"
 #include "mcdi.h"
-- 
cgit v1.2.3-59-g8ed1b


From 6485f9ae3b96edeb2513528a6ad4fd77ee7bafc2 Mon Sep 17 00:00:00 2001
From: Julia Lawall <Julia.Lawall@inria.fr>
Date: Wed, 1 Jan 2020 08:43:31 +0100
Subject: ptp: ptp_clockmatrix: constify copied structure

The idtcm_caps structure is only copied into another structure,
so make it const.

The opportunity for this change was found using Coccinelle.

Signed-off-by: Julia Lawall <Julia.Lawall@inria.fr>
Acked-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/ptp/ptp_clockmatrix.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/ptp/ptp_clockmatrix.c b/drivers/ptp/ptp_clockmatrix.c
index a5110b7b4ece..e85836715f0b 100644
--- a/drivers/ptp/ptp_clockmatrix.c
+++ b/drivers/ptp/ptp_clockmatrix.c
@@ -1102,7 +1102,7 @@ static void idtcm_display_version_info(struct idtcm *idtcm)
 		 product_id, hw_rev_id, bond_id, csr_id, irq_id);
 }
 
-static struct ptp_clock_info idtcm_caps = {
+static const struct ptp_clock_info idtcm_caps = {
 	.owner		= THIS_MODULE,
 	.max_adj	= 244000,
 	.n_per_out	= 1,
-- 
cgit v1.2.3-59-g8ed1b


From a02158d62faa19d3cad8caf481253998a6c47934 Mon Sep 17 00:00:00 2001
From: Julia Lawall <Julia.Lawall@inria.fr>
Date: Wed, 1 Jan 2020 18:49:44 +0100
Subject: fsl/fman: use resource_size

Use resource_size rather than a verbose computation on
the end and start fields.

The semantic patch that makes these changes is as follows:
(http://coccinelle.lip6.fr/)

<smpl>
@@ struct resource ptr; @@
- (ptr.end + 1 - ptr.start)
+ resource_size(&ptr)

@@ struct resource *ptr; @@
- (ptr->end + 1 - ptr->start)
+ resource_size(ptr)
</smpl>

Signed-off-by: Julia Lawall <Julia.Lawall@inria.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/freescale/fman/mac.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/freescale/fman/mac.c b/drivers/net/ethernet/freescale/fman/mac.c
index f0806ace1ae2..55f2122c3217 100644
--- a/drivers/net/ethernet/freescale/fman/mac.c
+++ b/drivers/net/ethernet/freescale/fman/mac.c
@@ -692,7 +692,7 @@ static int mac_probe(struct platform_device *_of_dev)
 
 	mac_dev->res = __devm_request_region(dev,
 					     fman_get_mem_region(priv->fman),
-					     res.start, res.end + 1 - res.start,
+					     res.start, resource_size(&res),
 					     "mac");
 	if (!mac_dev->res) {
 		dev_err(dev, "__devm_request_mem_region(mac) failed\n");
@@ -701,7 +701,7 @@ static int mac_probe(struct platform_device *_of_dev)
 	}
 
 	priv->vaddr = devm_ioremap(dev, mac_dev->res->start,
-				   mac_dev->res->end + 1 - mac_dev->res->start);
+				   resource_size(mac_dev->res));
 	if (!priv->vaddr) {
 		dev_err(dev, "devm_ioremap() failed\n");
 		err = -EIO;
-- 
cgit v1.2.3-59-g8ed1b


From 0d08c9ec7d6eb9b4c8639396b1cc6f692cbb7955 Mon Sep 17 00:00:00 2001
From: Po Liu <po.liu@nxp.com>
Date: Thu, 2 Jan 2020 04:59:24 +0000
Subject: enetc: add support time specific departure base on the qos etf

ENETC implement time specific departure capability, which enables
the user to specify when a frame can be transmitted. When this
capability is enabled, the device will delay the transmission of
the frame so that it can be transmitted at the precisely specified time.
The delay departure time up to 0.5 seconds in the future. If the
departure time in the transmit BD has not yet been reached, based
on the current time, the packet will not be transmitted.

This driver was loaded by Qos driver ETF. User could load it by tc
commands. Here are the example commands:

tc qdisc add dev eth0 root handle 1: mqprio \
	   num_tc 8 map 0 1 2 3 4 5 6 7 hw 1
tc qdisc replace dev eth0 parent 1:8 etf \
	   clockid CLOCK_TAI delta 30000  offload

These example try to set queue mapping first and then set queue 7
with 30us ahead dequeue time.

Then user send test frame should set SO_TXTIME feature for socket.

There are also some limitations for this feature in hardware:
- Transmit checksum offloads and time specific departure operation
are mutually exclusive.
- Time Aware Shaper feature (Qbv) offload and time specific departure
operation are mutually exclusive.

Signed-off-by: Po Liu <Po.Liu@nxp.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/freescale/enetc/enetc.c     | 12 ++++++++
 drivers/net/ethernet/freescale/enetc/enetc.h     |  3 ++
 drivers/net/ethernet/freescale/enetc/enetc_hw.h  | 10 ++++++-
 drivers/net/ethernet/freescale/enetc/enetc_qos.c | 35 ++++++++++++++++++++++++
 4 files changed, 59 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/freescale/enetc/enetc.c b/drivers/net/ethernet/freescale/enetc/enetc.c
index 2ee4a2cd4780..1f79e36116a3 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc.c
+++ b/drivers/net/ethernet/freescale/enetc/enetc.c
@@ -149,11 +149,21 @@ static int enetc_map_tx_buffs(struct enetc_bdr *tx_ring, struct sk_buff *skb,
 
 	if (enetc_tx_csum(skb, &temp_bd))
 		flags |= ENETC_TXBD_FLAGS_CSUM | ENETC_TXBD_FLAGS_L4CS;
+	else if (tx_ring->tsd_enable)
+		flags |= ENETC_TXBD_FLAGS_TSE | ENETC_TXBD_FLAGS_TXSTART;
 
 	/* first BD needs frm_len and offload flags set */
 	temp_bd.frm_len = cpu_to_le16(skb->len);
 	temp_bd.flags = flags;
 
+	if (flags & ENETC_TXBD_FLAGS_TSE) {
+		u32 temp;
+
+		temp = (skb->skb_mstamp_ns >> 5 & ENETC_TXBD_TXSTART_MASK)
+			| (flags << ENETC_TXBD_FLAGS_OFFSET);
+		temp_bd.txstart = cpu_to_le32(temp);
+	}
+
 	if (flags & ENETC_TXBD_FLAGS_EX) {
 		u8 e_flags = 0;
 		*txbd = temp_bd;
@@ -1505,6 +1515,8 @@ int enetc_setup_tc(struct net_device *ndev, enum tc_setup_type type,
 		return enetc_setup_tc_taprio(ndev, type_data);
 	case TC_SETUP_QDISC_CBS:
 		return enetc_setup_tc_cbs(ndev, type_data);
+	case TC_SETUP_QDISC_ETF:
+		return enetc_setup_tc_txtime(ndev, type_data);
 	default:
 		return -EOPNOTSUPP;
 	}
diff --git a/drivers/net/ethernet/freescale/enetc/enetc.h b/drivers/net/ethernet/freescale/enetc/enetc.h
index 7ee0da6d0015..dd4a227ffc7a 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc.h
+++ b/drivers/net/ethernet/freescale/enetc/enetc.h
@@ -72,6 +72,7 @@ struct enetc_bdr {
 	struct enetc_ring_stats stats;
 
 	dma_addr_t bd_dma_base;
+	u8 tsd_enable; /* Time specific departure */
 } ____cacheline_aligned_in_smp;
 
 static inline void enetc_bdr_idx_inc(struct enetc_bdr *bdr, int *i)
@@ -256,8 +257,10 @@ int enetc_send_cmd(struct enetc_si *si, struct enetc_cbd *cbd);
 int enetc_setup_tc_taprio(struct net_device *ndev, void *type_data);
 void enetc_sched_speed_set(struct net_device *ndev);
 int enetc_setup_tc_cbs(struct net_device *ndev, void *type_data);
+int enetc_setup_tc_txtime(struct net_device *ndev, void *type_data);
 #else
 #define enetc_setup_tc_taprio(ndev, type_data) -EOPNOTSUPP
 #define enetc_sched_speed_set(ndev) (void)0
 #define enetc_setup_tc_cbs(ndev, type_data) -EOPNOTSUPP
+#define enetc_setup_tc_txtime(ndev, type_data) -EOPNOTSUPP
 #endif
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_hw.h b/drivers/net/ethernet/freescale/enetc/enetc_hw.h
index 51f543ef37a8..8375cd886dba 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc_hw.h
+++ b/drivers/net/ethernet/freescale/enetc/enetc_hw.h
@@ -358,6 +358,7 @@ union enetc_tx_bd {
 				u8 l4_csoff;
 				u8 flags;
 			}; /* default layout */
+			__le32 txstart;
 			__le32 lstatus;
 		};
 	};
@@ -378,11 +379,14 @@ union enetc_tx_bd {
 };
 
 #define ENETC_TXBD_FLAGS_L4CS	BIT(0)
+#define ENETC_TXBD_FLAGS_TSE	BIT(1)
 #define ENETC_TXBD_FLAGS_W	BIT(2)
 #define ENETC_TXBD_FLAGS_CSUM	BIT(3)
+#define ENETC_TXBD_FLAGS_TXSTART BIT(4)
 #define ENETC_TXBD_FLAGS_EX	BIT(6)
 #define ENETC_TXBD_FLAGS_F	BIT(7)
-
+#define ENETC_TXBD_TXSTART_MASK GENMASK(24, 0)
+#define ENETC_TXBD_FLAGS_OFFSET 24
 static inline void enetc_clear_tx_bd(union enetc_tx_bd *txbd)
 {
 	memset(txbd, 0, sizeof(*txbd));
@@ -615,3 +619,7 @@ struct enetc_cbd {
 /* Port time gating capability register */
 #define ENETC_QBV_PTGCAPR_OFFSET	0x11a08
 #define ENETC_QBV_MAX_GCL_LEN_MASK	GENMASK(15, 0)
+
+/* Port time specific departure */
+#define ENETC_PTCTSDR(n)	(0x1210 + 4 * (n))
+#define ENETC_TSDE		BIT(31)
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_qos.c b/drivers/net/ethernet/freescale/enetc/enetc_qos.c
index 9190ffc9f6b2..e910aaf0f5ec 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc_qos.c
+++ b/drivers/net/ethernet/freescale/enetc/enetc_qos.c
@@ -156,6 +156,11 @@ int enetc_setup_tc_taprio(struct net_device *ndev, void *type_data)
 	int err;
 	int i;
 
+	/* TSD and Qbv are mutually exclusive in hardware */
+	for (i = 0; i < priv->num_tx_rings; i++)
+		if (priv->tx_ring[i]->tsd_enable)
+			return -EBUSY;
+
 	for (i = 0; i < priv->num_tx_rings; i++)
 		enetc_set_bdr_prio(&priv->si->hw,
 				   priv->tx_ring[i]->index,
@@ -297,3 +302,33 @@ int enetc_setup_tc_cbs(struct net_device *ndev, void *type_data)
 
 	return 0;
 }
+
+int enetc_setup_tc_txtime(struct net_device *ndev, void *type_data)
+{
+	struct enetc_ndev_priv *priv = netdev_priv(ndev);
+	struct tc_etf_qopt_offload *qopt = type_data;
+	u8 tc_nums = netdev_get_num_tc(ndev);
+	int tc;
+
+	if (!tc_nums)
+		return -EOPNOTSUPP;
+
+	tc = qopt->queue;
+
+	if (tc < 0 || tc > priv->num_tx_rings)
+		return -EINVAL;
+
+	/* Do not support TXSTART and TX CSUM offload simutaniously */
+	if (ndev->features & NETIF_F_CSUM_MASK)
+		return -EBUSY;
+
+	/* TSD and Qbv are mutually exclusive in hardware */
+	if (enetc_rd(&priv->si->hw, ENETC_QBV_PTGCR_OFFSET) & ENETC_QBV_TGE)
+		return -EBUSY;
+
+	priv->tx_ring[tc]->tsd_enable = qopt->enable;
+	enetc_port_wr(&priv->si->hw, ENETC_PTCTSDR(tc),
+		      qopt->enable ? ENETC_TSDE : 0);
+
+	return 0;
+}
-- 
cgit v1.2.3-59-g8ed1b


From d0e8bcafc8aff5553beffe55046795f9bab9fe7b Mon Sep 17 00:00:00 2001
From: Mao Wenan <maowenan@huawei.com>
Date: Thu, 2 Jan 2020 22:02:27 +0800
Subject: tcp: use REXMIT_NEW instead of magic number

REXMIT_NEW is a macro for "FRTO-style
transmit of unsent/new packets", this patch
makes it more readable.

Signed-off-by: Mao Wenan <maowenan@huawei.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/tcp_input.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 88b987ca9ebb..1d1e3493965f 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -3550,7 +3550,7 @@ static void tcp_xmit_recovery(struct sock *sk, int rexmit)
 	if (rexmit == REXMIT_NONE || sk->sk_state == TCP_SYN_SENT)
 		return;
 
-	if (unlikely(rexmit == 2)) {
+	if (unlikely(rexmit == REXMIT_NEW)) {
 		__tcp_push_pending_frames(sk, tcp_current_mss(sk),
 					  TCP_NAGLE_OFF);
 		if (after(tp->snd_nxt, tp->high_seq))
-- 
cgit v1.2.3-59-g8ed1b


From 4a883ccfba3adefc7123ec1a6da3c7301a56ce97 Mon Sep 17 00:00:00 2001
From: YueHaibing <yuehaibing@huawei.com>
Date: Fri, 3 Jan 2020 03:48:56 +0000
Subject: ethtool: remove set but not used variable 'lsettings'

Fixes gcc '-Wunused-but-set-variable' warning:

net/ethtool/linkmodes.c: In function 'ethnl_set_linkmodes':
net/ethtool/linkmodes.c:326:32: warning:
 variable 'lsettings' set but not used [-Wunused-but-set-variable]
  struct ethtool_link_settings *lsettings;
                                ^
It is never used, so remove it.

Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Reviewed-by: Michal Kubecek <mkubecek@suse.cz>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ethtool/linkmodes.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/net/ethtool/linkmodes.c b/net/ethtool/linkmodes.c
index 0b99f494ad3b..96f20be64553 100644
--- a/net/ethtool/linkmodes.c
+++ b/net/ethtool/linkmodes.c
@@ -323,7 +323,6 @@ int ethnl_set_linkmodes(struct sk_buff *skb, struct genl_info *info)
 {
 	struct nlattr *tb[ETHTOOL_A_LINKMODES_MAX + 1];
 	struct ethtool_link_ksettings ksettings = {};
-	struct ethtool_link_settings *lsettings;
 	struct ethnl_req_info req_info = {};
 	struct net_device *dev;
 	bool mod = false;
@@ -354,7 +353,6 @@ int ethnl_set_linkmodes(struct sk_buff *skb, struct genl_info *info)
 			GENL_SET_ERR_MSG(info, "failed to retrieve link settings");
 		goto out_ops;
 	}
-	lsettings = &ksettings.base;
 
 	ret = ethnl_update_linkmodes(info, tb, &ksettings, &mod);
 	if (ret < 0)
-- 
cgit v1.2.3-59-g8ed1b


From b39c78b2aa09cae05f3a48c11f67b3add0d604de Mon Sep 17 00:00:00 2001
From: Li RongQing <lirongqing@baidu.com>
Date: Fri, 3 Jan 2020 11:51:00 +0800
Subject: net: remove the check argument from __skb_gro_checksum_convert

The argument is always ignored, so remove it.

Signed-off-by: Li RongQing <lirongqing@baidu.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 6 +++---
 net/ipv4/gre_offload.c    | 2 +-
 net/ipv4/udp_offload.c    | 2 +-
 net/ipv6/udp_offload.c    | 2 +-
 4 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 2fd19fb8826d..2741aa35bec6 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -2826,16 +2826,16 @@ static inline bool __skb_gro_checksum_convert_check(struct sk_buff *skb)
 }
 
 static inline void __skb_gro_checksum_convert(struct sk_buff *skb,
-					      __sum16 check, __wsum pseudo)
+					      __wsum pseudo)
 {
 	NAPI_GRO_CB(skb)->csum = ~pseudo;
 	NAPI_GRO_CB(skb)->csum_valid = 1;
 }
 
-#define skb_gro_checksum_try_convert(skb, proto, check, compute_pseudo)	\
+#define skb_gro_checksum_try_convert(skb, proto, compute_pseudo)	\
 do {									\
 	if (__skb_gro_checksum_convert_check(skb))			\
-		__skb_gro_checksum_convert(skb, check,			\
+		__skb_gro_checksum_convert(skb, 			\
 					   compute_pseudo(skb, proto));	\
 } while (0)
 
diff --git a/net/ipv4/gre_offload.c b/net/ipv4/gre_offload.c
index 4de7e962d3da..2e6d1b7a7bc9 100644
--- a/net/ipv4/gre_offload.c
+++ b/net/ipv4/gre_offload.c
@@ -174,7 +174,7 @@ static struct sk_buff *gre_gro_receive(struct list_head *head,
 		if (skb_gro_checksum_simple_validate(skb))
 			goto out_unlock;
 
-		skb_gro_checksum_try_convert(skb, IPPROTO_GRE, 0,
+		skb_gro_checksum_try_convert(skb, IPPROTO_GRE,
 					     null_compute_pseudo);
 	}
 
diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c
index a3908e55ed89..b25e42100ceb 100644
--- a/net/ipv4/udp_offload.c
+++ b/net/ipv4/udp_offload.c
@@ -480,7 +480,7 @@ struct sk_buff *udp4_gro_receive(struct list_head *head, struct sk_buff *skb)
 						 inet_gro_compute_pseudo))
 		goto flush;
 	else if (uh->check)
-		skb_gro_checksum_try_convert(skb, IPPROTO_UDP, uh->check,
+		skb_gro_checksum_try_convert(skb, IPPROTO_UDP,
 					     inet_gro_compute_pseudo);
 skip:
 	NAPI_GRO_CB(skb)->is_ipv6 = 0;
diff --git a/net/ipv6/udp_offload.c b/net/ipv6/udp_offload.c
index 64b8f05d6735..f0d5fc27d0b5 100644
--- a/net/ipv6/udp_offload.c
+++ b/net/ipv6/udp_offload.c
@@ -127,7 +127,7 @@ struct sk_buff *udp6_gro_receive(struct list_head *head, struct sk_buff *skb)
 						 ip6_gro_compute_pseudo))
 		goto flush;
 	else if (uh->check)
-		skb_gro_checksum_try_convert(skb, IPPROTO_UDP, uh->check,
+		skb_gro_checksum_try_convert(skb, IPPROTO_UDP,
 					     ip6_gro_compute_pseudo);
 
 skip:
-- 
cgit v1.2.3-59-g8ed1b


From 1f623431100138ffaaae36e122f2727c13de640a Mon Sep 17 00:00:00 2001
From: Xu Wang <vulab@iscas.ac.cn>
Date: Fri, 3 Jan 2020 09:20:40 +0000
Subject: net: Remove redundant BUG_ON() check in phonet_pernet

Passing NULL to phonet_pernet causes a crash via BUG_ON.
Dereferencing net in net_generic() also has the same effect.
This patch removes the redundant BUG_ON check on the same parameter.

Signed-off-by: Xu Wang <vulab@iscas.ac.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/phonet/pn_dev.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/net/phonet/pn_dev.c b/net/phonet/pn_dev.c
index 49bd76a87461..ac0fae06cc15 100644
--- a/net/phonet/pn_dev.c
+++ b/net/phonet/pn_dev.c
@@ -35,8 +35,6 @@ static unsigned int phonet_net_id __read_mostly;
 
 static struct phonet_net *phonet_pernet(struct net *net)
 {
-	BUG_ON(!net);
-
 	return net_generic(net, phonet_net_id);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From d2e9d229cfbd73baf33a817a0212233cea7df59a Mon Sep 17 00:00:00 2001
From: Xu Wang <vulab@iscas.ac.cn>
Date: Fri, 3 Jan 2020 09:28:16 +0000
Subject: l2tp: Remove redundant BUG_ON() check in l2tp_pernet

Passing NULL to l2tp_pernet causes a crash via BUG_ON.
Dereferencing net in net_generic() also has the same effect.
This patch removes the redundant BUG_ON check on the same parameter.

Signed-off-by: Xu Wang <vulab@iscas.ac.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/l2tp/l2tp_core.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c
index f82ea12bac37..c99223cb3338 100644
--- a/net/l2tp/l2tp_core.c
+++ b/net/l2tp/l2tp_core.c
@@ -122,8 +122,6 @@ static inline struct l2tp_tunnel *l2tp_tunnel(struct sock *sk)
 
 static inline struct l2tp_net *l2tp_pernet(const struct net *net)
 {
-	BUG_ON(!net);
-
 	return net_generic(net, l2tp_net_id);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 06f6646191fd458c5a649e052a20a9aecb3ac54c Mon Sep 17 00:00:00 2001
From: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Date: Fri, 3 Jan 2020 13:40:32 +0200
Subject: net: netsec: Change page pool nid to NUMA_NO_NODE

The current driver only exists on a non NUMA aware machine.
With 44768decb7c0 ("page_pool: handle page recycle for NUMA_NO_NODE condition")
applied we can safely change that to NUMA_NO_NODE and accommodate future
NUMA aware hardware using netsec network interface

Signed-off-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/socionext/netsec.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/socionext/netsec.c b/drivers/net/ethernet/socionext/netsec.c
index 869a498e3b5e..b5a9e947a4a8 100644
--- a/drivers/net/ethernet/socionext/netsec.c
+++ b/drivers/net/ethernet/socionext/netsec.c
@@ -1279,7 +1279,7 @@ static int netsec_setup_rx_dring(struct netsec_priv *priv)
 	/* internal DMA mapping in page_pool */
 	pp_params.flags = PP_FLAG_DMA_MAP;
 	pp_params.pool_size = DESC_NUM;
-	pp_params.nid = cpu_to_node(0);
+	pp_params.nid = NUMA_NO_NODE;
 	pp_params.dev = priv->dev;
 	pp_params.dma_dir = xdp_prog ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE;
 
-- 
cgit v1.2.3-59-g8ed1b


From 3c85efb8f15ffa5bd165881b9fd1f9e5dd1d705f Mon Sep 17 00:00:00 2001
From: yu kuai <yukuai3@huawei.com>
Date: Fri, 3 Jan 2020 20:04:37 +0800
Subject: bna: remove set but not used variable 'pgoff'
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

drivers/net/ethernet/brocade/bna/bfa_ioc.c: In function
‘bfa_ioc_fwver_clear’:
drivers/net/ethernet/brocade/bna/bfa_ioc.c:1127:13: warning: variable
‘pgoff’ set but not used [-Wunused-but-set-variable]

It is never used, and so can be removed.

Signed-off-by: yu kuai <yukuai3@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/brocade/bna/bfa_ioc.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/brocade/bna/bfa_ioc.c b/drivers/net/ethernet/brocade/bna/bfa_ioc.c
index 4042c2185e98..e17bfc87da90 100644
--- a/drivers/net/ethernet/brocade/bna/bfa_ioc.c
+++ b/drivers/net/ethernet/brocade/bna/bfa_ioc.c
@@ -1124,11 +1124,10 @@ bfa_nw_ioc_sem_release(void __iomem *sem_reg)
 static void
 bfa_ioc_fwver_clear(struct bfa_ioc *ioc)
 {
-	u32 pgnum, pgoff, loff = 0;
+	u32 pgnum, loff = 0;
 	int i;
 
 	pgnum = PSS_SMEM_PGNUM(ioc->ioc_regs.smem_pg0, loff);
-	pgoff = PSS_SMEM_PGOFF(loff);
 	writel(pgnum, ioc->ioc_regs.host_page_num_fn);
 
 	for (i = 0; i < (sizeof(struct bfi_ioc_image_hdr) / sizeof(u32)); i++) {
-- 
cgit v1.2.3-59-g8ed1b


From a54e3b8cffd6d2f12c7fcf14f6d13ab997c0f160 Mon Sep 17 00:00:00 2001
From: Brett Creeley <brett.creeley@intel.com>
Date: Thu, 12 Dec 2019 03:12:53 -0800
Subject: ice: Support UDP segmentation offload

Based on the work done by Alex Duyck on other Intel drivers, add code to
support UDP segmentation offload (USO) for the ice driver.

Signed-off-by: Brett Creeley <brett.creeley@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_main.c |  3 ++-
 drivers/net/ethernet/intel/ice/ice_txrx.c | 15 ++++++++++++---
 2 files changed, 14 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 4d5220c9c721..1fc6c350487b 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -2330,7 +2330,8 @@ static void ice_set_netdev_features(struct net_device *netdev)
 			 NETIF_F_HW_VLAN_CTAG_TX     |
 			 NETIF_F_HW_VLAN_CTAG_RX;
 
-	tso_features = NETIF_F_TSO;
+	tso_features = NETIF_F_TSO		|
+		       NETIF_F_GSO_UDP_L4;
 
 	/* set features that user can change */
 	netdev->hw_features = dflt_features | csumo_features |
diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c
index 2c212f64d99f..b77514bbd7ba 100644
--- a/drivers/net/ethernet/intel/ice/ice_txrx.c
+++ b/drivers/net/ethernet/intel/ice/ice_txrx.c
@@ -1925,6 +1925,7 @@ int ice_tso(struct ice_tx_buf *first, struct ice_tx_offload_params *off)
 	} ip;
 	union {
 		struct tcphdr *tcp;
+		struct udphdr *udp;
 		unsigned char *hdr;
 	} l4;
 	u64 cd_mss, cd_tso_len;
@@ -1958,10 +1959,18 @@ int ice_tso(struct ice_tx_buf *first, struct ice_tx_offload_params *off)
 
 	/* remove payload length from checksum */
 	paylen = skb->len - l4_start;
-	csum_replace_by_diff(&l4.tcp->check, (__force __wsum)htonl(paylen));
 
-	/* compute length of segmentation header */
-	off->header_len = (l4.tcp->doff * 4) + l4_start;
+	if (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4) {
+		csum_replace_by_diff(&l4.udp->check,
+				     (__force __wsum)htonl(paylen));
+		/* compute length of UDP segmentation header */
+		off->header_len = sizeof(l4.udp) + l4_start;
+	} else {
+		csum_replace_by_diff(&l4.tcp->check,
+				     (__force __wsum)htonl(paylen));
+		/* compute length of TCP segmentation header */
+		off->header_len = (l4.tcp->doff * 4) + l4_start;
+	}
 
 	/* update gso_segs and bytecount */
 	first->gso_segs = skb_shinfo(skb)->gso_segs;
-- 
cgit v1.2.3-59-g8ed1b


From cd6d6b83316a9cd2cb97df2bb0bbe617e832d0df Mon Sep 17 00:00:00 2001
From: Brett Creeley <brett.creeley@intel.com>
Date: Thu, 12 Dec 2019 03:12:54 -0800
Subject: ice: Fix VF spoofchk

There are many things wrong with the function
ice_set_vf_spoofchk().

1. The VSI being modified is the PF VSI, not the VF VSI.
2. We are enabling Rx VLAN pruning instead of Tx VLAN anti-spoof.
3. The spoofchk setting for each VF is not initialized correctly
   or re-initialized correctly on reset.

To fix [1] we need to make sure we are modifying the VF VSI.
This is done by using the vf->lan_vsi_idx to index into the PF's
VSI array.

To fix [2] replace setting Rx VLAN pruning in ice_set_vf_spoofchk()
with setting Tx VLAN anti-spoof.

To Fix [3] we need to make sure the initial VSI settings match what
is done in ice_set_vf_spoofchk() for spoofchk=on. Also make sure
this also works for VF reset. This was done by modifying ice_vsi_init()
to account for the current spoofchk state of the VF VSI.

Because of these changes, Tx VLAN anti-spoof needs to be removed
from ice_cfg_vlan_pruning(). This is okay for the VF because this
is now controlled from the admin enabling/disabling spoofchk. For the
PF, Tx VLAN anti-spoof should not be set. This change requires us to
call ice_set_vf_spoofchk() when configuring promiscuous mode for
the VF which requires ice_set_vf_spoofchk() to move in order to prevent
a forward declaration prototype.

Also, add VLAN 0 by default when allocating a VF since the PF is unaware
if the guest OS is running the 8021q module. Without this, MDD events will
trigger on untagged traffic because spoofcheck is enabled by default. Due
to this change, ignore add/delete messages for VLAN 0 from VIRTCHNL since
this is added/deleted during VF initialization/teardown respectively and
should not be modified.

Signed-off-by: Brett Creeley <brett.creeley@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ice/ice.h             |   1 +
 drivers/net/ethernet/intel/ice/ice_lib.c         |  34 ++--
 drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c | 207 ++++++++++++++---------
 drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h |   1 -
 4 files changed, 146 insertions(+), 97 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h
index f972dce8aebb..54b2c349c195 100644
--- a/drivers/net/ethernet/intel/ice/ice.h
+++ b/drivers/net/ethernet/intel/ice/ice.h
@@ -275,6 +275,7 @@ struct ice_vsi {
 	u8 current_isup:1;		 /* Sync 'link up' logging */
 	u8 stat_offsets_loaded:1;
 	u8 vlan_ena:1;
+	u16 num_vlan;
 
 	/* queue information */
 	u8 tx_mapping_mode;		 /* ICE_MAP_MODE_[CONTIG|SCATTER] */
diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c
index e7449248fab4..a47241e92d0b 100644
--- a/drivers/net/ethernet/intel/ice/ice_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_lib.c
@@ -817,12 +817,23 @@ static int ice_vsi_init(struct ice_vsi *vsi, bool init_vsi)
 		ctxt->info.valid_sections |=
 			cpu_to_le16(ICE_AQ_VSI_PROP_RXQ_MAP_VALID);
 
-	/* Enable MAC Antispoof with new VSI being initialized or updated */
-	if (vsi->type == ICE_VSI_VF && pf->vf[vsi->vf_id].spoofchk) {
+	/* enable/disable MAC and VLAN anti-spoof when spoofchk is on/off
+	 * respectively
+	 */
+	if (vsi->type == ICE_VSI_VF) {
 		ctxt->info.valid_sections |=
 			cpu_to_le16(ICE_AQ_VSI_PROP_SECURITY_VALID);
-		ctxt->info.sec_flags |=
-			ICE_AQ_VSI_SEC_FLAG_ENA_MAC_ANTI_SPOOF;
+		if (pf->vf[vsi->vf_id].spoofchk) {
+			ctxt->info.sec_flags |=
+				ICE_AQ_VSI_SEC_FLAG_ENA_MAC_ANTI_SPOOF |
+				(ICE_AQ_VSI_SEC_TX_VLAN_PRUNE_ENA <<
+				 ICE_AQ_VSI_SEC_TX_PRUNE_ENA_S);
+		} else {
+			ctxt->info.sec_flags &=
+				~(ICE_AQ_VSI_SEC_FLAG_ENA_MAC_ANTI_SPOOF |
+				  (ICE_AQ_VSI_SEC_TX_VLAN_PRUNE_ENA <<
+				   ICE_AQ_VSI_SEC_TX_PRUNE_ENA_S));
+		}
 	}
 
 	/* Allow control frames out of main VSI */
@@ -1636,22 +1647,14 @@ int ice_cfg_vlan_pruning(struct ice_vsi *vsi, bool ena, bool vlan_promisc)
 
 	ctxt->info = vsi->info;
 
-	if (ena) {
-		ctxt->info.sec_flags |=
-			ICE_AQ_VSI_SEC_TX_VLAN_PRUNE_ENA <<
-			ICE_AQ_VSI_SEC_TX_PRUNE_ENA_S;
+	if (ena)
 		ctxt->info.sw_flags2 |= ICE_AQ_VSI_SW_FLAG_RX_VLAN_PRUNE_ENA;
-	} else {
-		ctxt->info.sec_flags &=
-			~(ICE_AQ_VSI_SEC_TX_VLAN_PRUNE_ENA <<
-			  ICE_AQ_VSI_SEC_TX_PRUNE_ENA_S);
+	else
 		ctxt->info.sw_flags2 &= ~ICE_AQ_VSI_SW_FLAG_RX_VLAN_PRUNE_ENA;
-	}
 
 	if (!vlan_promisc)
 		ctxt->info.valid_sections =
-			cpu_to_le16(ICE_AQ_VSI_PROP_SECURITY_VALID |
-				    ICE_AQ_VSI_PROP_SW_VALID);
+			cpu_to_le16(ICE_AQ_VSI_PROP_SW_VALID);
 
 	status = ice_update_vsi(&pf->hw, vsi->idx, ctxt, NULL);
 	if (status) {
@@ -1661,7 +1664,6 @@ int ice_cfg_vlan_pruning(struct ice_vsi *vsi, bool ena, bool vlan_promisc)
 		goto err_out;
 	}
 
-	vsi->info.sec_flags = ctxt->info.sec_flags;
 	vsi->info.sw_flags2 = ctxt->info.sw_flags2;
 
 	kfree(ctxt);
diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c
index edb374296d1f..3860e9c964b9 100644
--- a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c
@@ -991,10 +991,17 @@ static void ice_cleanup_and_realloc_vf(struct ice_vf *vf)
 
 	/* reallocate VF resources to finish resetting the VSI state */
 	if (!ice_alloc_vf_res(vf)) {
+		struct ice_vsi *vsi;
+
 		ice_ena_vf_mappings(vf);
 		set_bit(ICE_VF_STATE_ACTIVE, vf->vf_states);
 		clear_bit(ICE_VF_STATE_DIS, vf->vf_states);
-		vf->num_vlan = 0;
+
+		vsi = pf->vsi[vf->lan_vsi_idx];
+		if (ice_vsi_add_vlan(vsi, 0))
+			dev_warn(ice_pf_to_dev(pf),
+				 "Failed to add VLAN 0 filter for VF %d, MDD events will trigger. Reset the VF, disable spoofchk, or enable 8021q module on the guest",
+				 vf->vf_id);
 	}
 
 	/* Tell the VF driver the reset is done. This needs to be done only
@@ -1023,7 +1030,7 @@ ice_vf_set_vsi_promisc(struct ice_vf *vf, struct ice_vsi *vsi, u8 promisc_m,
 	struct ice_hw *hw;
 
 	hw = &pf->hw;
-	if (vf->num_vlan) {
+	if (vsi->num_vlan) {
 		status = ice_set_vlan_vsi_promisc(hw, vsi->idx, promisc_m,
 						  rm_promisc);
 	} else if (vf->port_vlan_id) {
@@ -1273,7 +1280,7 @@ static bool ice_reset_vf(struct ice_vf *vf, bool is_vflr)
 	 */
 	if (test_bit(ICE_VF_STATE_UC_PROMISC, vf->vf_states) ||
 	    test_bit(ICE_VF_STATE_MC_PROMISC, vf->vf_states)) {
-		if (vf->port_vlan_id ||  vf->num_vlan)
+		if (vf->port_vlan_id || vsi->num_vlan)
 			promisc_m = ICE_UCAST_VLAN_PROMISC_BITS;
 		else
 			promisc_m = ICE_UCAST_PROMISC_BITS;
@@ -1917,6 +1924,89 @@ error_param:
 				     NULL, 0);
 }
 
+/**
+ * ice_set_vf_spoofchk
+ * @netdev: network interface device structure
+ * @vf_id: VF identifier
+ * @ena: flag to enable or disable feature
+ *
+ * Enable or disable VF spoof checking
+ */
+int ice_set_vf_spoofchk(struct net_device *netdev, int vf_id, bool ena)
+{
+	struct ice_netdev_priv *np = netdev_priv(netdev);
+	struct ice_pf *pf = np->vsi->back;
+	struct ice_vsi_ctx *ctx;
+	struct ice_vsi *vf_vsi;
+	enum ice_status status;
+	struct device *dev;
+	struct ice_vf *vf;
+	int ret = 0;
+
+	dev = ice_pf_to_dev(pf);
+	if (ice_validate_vf_id(pf, vf_id))
+		return -EINVAL;
+
+	vf = &pf->vf[vf_id];
+
+	if (ice_check_vf_init(pf, vf))
+		return -EBUSY;
+
+	vf_vsi = pf->vsi[vf->lan_vsi_idx];
+	if (!vf_vsi) {
+		netdev_err(netdev, "VSI %d for VF %d is null\n",
+			   vf->lan_vsi_idx, vf->vf_id);
+		return -EINVAL;
+	}
+
+	if (vf_vsi->type != ICE_VSI_VF) {
+		netdev_err(netdev,
+			   "Type %d of VSI %d for VF %d is no ICE_VSI_VF\n",
+			   vf_vsi->type, vf_vsi->vsi_num, vf->vf_id);
+		return -ENODEV;
+	}
+
+	if (ena == vf->spoofchk) {
+		dev_dbg(dev, "VF spoofchk already %s\n", ena ? "ON" : "OFF");
+		return 0;
+	}
+
+	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
+	if (!ctx)
+		return -ENOMEM;
+
+	ctx->info.sec_flags = vf_vsi->info.sec_flags;
+	ctx->info.valid_sections = cpu_to_le16(ICE_AQ_VSI_PROP_SECURITY_VALID);
+	if (ena) {
+		ctx->info.sec_flags |=
+			ICE_AQ_VSI_SEC_FLAG_ENA_MAC_ANTI_SPOOF |
+			(ICE_AQ_VSI_SEC_TX_VLAN_PRUNE_ENA <<
+			 ICE_AQ_VSI_SEC_TX_PRUNE_ENA_S);
+	} else {
+		ctx->info.sec_flags &=
+			~(ICE_AQ_VSI_SEC_FLAG_ENA_MAC_ANTI_SPOOF |
+			  (ICE_AQ_VSI_SEC_TX_VLAN_PRUNE_ENA <<
+			   ICE_AQ_VSI_SEC_TX_PRUNE_ENA_S));
+	}
+
+	status = ice_update_vsi(&pf->hw, vf_vsi->idx, ctx, NULL);
+	if (status) {
+		dev_err(dev,
+			"Failed to %sable spoofchk on VF %d VSI %d\n error %d",
+			ena ? "en" : "dis", vf->vf_id, vf_vsi->vsi_num, status);
+		ret = -EIO;
+		goto out;
+	}
+
+	/* only update spoofchk state and VSI context on success */
+	vf_vsi->info.sec_flags = ctx->info.sec_flags;
+	vf->spoofchk = ena;
+
+out:
+	kfree(ctx);
+	return ret;
+}
+
 /**
  * ice_vc_get_stats_msg
  * @vf: pointer to the VF info
@@ -2744,17 +2834,6 @@ static int ice_vc_process_vlan_msg(struct ice_vf *vf, u8 *msg, bool add_v)
 		goto error_param;
 	}
 
-	if (add_v && !ice_is_vf_trusted(vf) &&
-	    vf->num_vlan >= ICE_MAX_VLAN_PER_VF) {
-		dev_info(dev,
-			 "VF-%d is not trusted, switch the VF to trusted mode, in order to add more VLAN addresses\n",
-			 vf->vf_id);
-		/* There is no need to let VF know about being not trusted,
-		 * so we can just return success message here
-		 */
-		goto error_param;
-	}
-
 	for (i = 0; i < vfl->num_elements; i++) {
 		if (vfl->vlan_id[i] > ICE_MAX_VLANID) {
 			v_ret = VIRTCHNL_STATUS_ERR_PARAM;
@@ -2771,6 +2850,17 @@ static int ice_vc_process_vlan_msg(struct ice_vf *vf, u8 *msg, bool add_v)
 		goto error_param;
 	}
 
+	if (add_v && !ice_is_vf_trusted(vf) &&
+	    vsi->num_vlan >= ICE_MAX_VLAN_PER_VF) {
+		dev_info(dev,
+			 "VF-%d is not trusted, switch the VF to trusted mode, in order to add more VLAN addresses\n",
+			 vf->vf_id);
+		/* There is no need to let VF know about being not trusted,
+		 * so we can just return success message here
+		 */
+		goto error_param;
+	}
+
 	if (vsi->info.pvid) {
 		v_ret = VIRTCHNL_STATUS_ERR_PARAM;
 		goto error_param;
@@ -2785,7 +2875,7 @@ static int ice_vc_process_vlan_msg(struct ice_vf *vf, u8 *msg, bool add_v)
 			u16 vid = vfl->vlan_id[i];
 
 			if (!ice_is_vf_trusted(vf) &&
-			    vf->num_vlan >= ICE_MAX_VLAN_PER_VF) {
+			    vsi->num_vlan >= ICE_MAX_VLAN_PER_VF) {
 				dev_info(dev,
 					 "VF-%d is not trusted, switch the VF to trusted mode, in order to add more VLAN addresses\n",
 					 vf->vf_id);
@@ -2796,12 +2886,20 @@ static int ice_vc_process_vlan_msg(struct ice_vf *vf, u8 *msg, bool add_v)
 				goto error_param;
 			}
 
-			if (ice_vsi_add_vlan(vsi, vid)) {
+			/* we add VLAN 0 by default for each VF so we can enable
+			 * Tx VLAN anti-spoof without triggering MDD events so
+			 * we don't need to add it again here
+			 */
+			if (!vid)
+				continue;
+
+			status = ice_vsi_add_vlan(vsi, vid);
+			if (status) {
 				v_ret = VIRTCHNL_STATUS_ERR_PARAM;
 				goto error_param;
 			}
 
-			vf->num_vlan++;
+			vsi->num_vlan++;
 			/* Enable VLAN pruning when VLAN is added */
 			if (!vlan_promisc) {
 				status = ice_cfg_vlan_pruning(vsi, true, false);
@@ -2837,21 +2935,29 @@ static int ice_vc_process_vlan_msg(struct ice_vf *vf, u8 *msg, bool add_v)
 		 */
 		int num_vf_vlan;
 
-		num_vf_vlan = vf->num_vlan;
+		num_vf_vlan = vsi->num_vlan;
 		for (i = 0; i < vfl->num_elements && i < num_vf_vlan; i++) {
 			u16 vid = vfl->vlan_id[i];
 
+			/* we add VLAN 0 by default for each VF so we can enable
+			 * Tx VLAN anti-spoof without triggering MDD events so
+			 * we don't want a VIRTCHNL request to remove it
+			 */
+			if (!vid)
+				continue;
+
 			/* Make sure ice_vsi_kill_vlan is successful before
 			 * updating VLAN information
 			 */
-			if (ice_vsi_kill_vlan(vsi, vid)) {
+			status = ice_vsi_kill_vlan(vsi, vid);
+			if (status) {
 				v_ret = VIRTCHNL_STATUS_ERR_PARAM;
 				goto error_param;
 			}
 
-			vf->num_vlan--;
+			vsi->num_vlan--;
 			/* Disable VLAN pruning when the last VLAN is removed */
-			if (!vf->num_vlan)
+			if (!vsi->num_vlan)
 				ice_cfg_vlan_pruning(vsi, false, false);
 
 			/* Disable Unicast/Multicast VLAN promiscuous mode */
@@ -3164,65 +3270,6 @@ ice_get_vf_cfg(struct net_device *netdev, int vf_id, struct ifla_vf_info *ivi)
 	return 0;
 }
 
-/**
- * ice_set_vf_spoofchk
- * @netdev: network interface device structure
- * @vf_id: VF identifier
- * @ena: flag to enable or disable feature
- *
- * Enable or disable VF spoof checking
- */
-int ice_set_vf_spoofchk(struct net_device *netdev, int vf_id, bool ena)
-{
-	struct ice_pf *pf = ice_netdev_to_pf(netdev);
-	struct ice_vsi *vsi = pf->vsi[0];
-	struct ice_vsi_ctx *ctx;
-	enum ice_status status;
-	struct device *dev;
-	struct ice_vf *vf;
-	int ret = 0;
-
-	dev = ice_pf_to_dev(pf);
-	if (ice_validate_vf_id(pf, vf_id))
-		return -EINVAL;
-
-	vf = &pf->vf[vf_id];
-	if (ice_check_vf_init(pf, vf))
-		return -EBUSY;
-
-	if (ena == vf->spoofchk) {
-		dev_dbg(dev, "VF spoofchk already %s\n",
-			ena ? "ON" : "OFF");
-		return 0;
-	}
-
-	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
-	if (!ctx)
-		return -ENOMEM;
-
-	ctx->info.valid_sections = cpu_to_le16(ICE_AQ_VSI_PROP_SECURITY_VALID);
-
-	if (ena) {
-		ctx->info.sec_flags |= ICE_AQ_VSI_SEC_FLAG_ENA_MAC_ANTI_SPOOF;
-		ctx->info.sw_flags2 |= ICE_AQ_VSI_SW_FLAG_RX_PRUNE_EN_M;
-	}
-
-	status = ice_update_vsi(&pf->hw, vsi->idx, ctx, NULL);
-	if (status) {
-		dev_dbg(dev,
-			"Error %d, failed to update VSI* parameters\n", status);
-		ret = -EIO;
-		goto out;
-	}
-
-	vf->spoofchk = ena;
-	vsi->info.sec_flags = ctx->info.sec_flags;
-	vsi->info.sw_flags2 = ctx->info.sw_flags2;
-out:
-	kfree(ctx);
-	return ret;
-}
-
 /**
  * ice_wait_on_vf_reset
  * @vf: The VF being resseting
diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h
index 88aa65d5cb31..611f45100438 100644
--- a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h
+++ b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h
@@ -91,7 +91,6 @@ struct ice_vf {
 	unsigned long vf_caps;		/* VF's adv. capabilities */
 	u8 num_req_qs;			/* num of queue pairs requested by VF */
 	u16 num_mac;
-	u16 num_vlan;
 	u16 num_vf_qs;			/* num of queue configured per VF */
 	u16 num_qs_ena;			/* total num of Tx/Rx queue enabled */
 };
-- 
cgit v1.2.3-59-g8ed1b


From fc0f39bcb50cd4a414ee5d5fef5208a7d04c4c32 Mon Sep 17 00:00:00 2001
From: Brett Creeley <brett.creeley@intel.com>
Date: Thu, 12 Dec 2019 03:12:55 -0800
Subject: ice: Add code to keep track of current dflt_vsi

We can't have more than one default VSI so prevent another VSI from
overwriting the current dflt_vsi. This was achieved by adding the
following functions:

ice_is_dflt_vsi_in_use()
- Used to check if the default VSI is already being used.

ice_is_vsi_dflt_vsi()
- Used to check if VSI passed in is in fact the default VSI.

ice_set_dflt_vsi()
- Used to set the default VSI via a switch rule

ice_clear_dflt_vsi()
- Used to clear the default VSI via a switch rule.

Also, there was no need to introduce any locking because all mailbox
events and synchronization of switch filters for the PF happen in the
service task.

Signed-off-by: Brett Creeley <brett.creeley@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ice/ice.h      |   2 +
 drivers/net/ethernet/intel/ice/ice_lib.c  | 118 ++++++++++++++++++++++++++++++
 drivers/net/ethernet/intel/ice/ice_lib.h  |   8 ++
 drivers/net/ethernet/intel/ice/ice_main.c |  43 +++++++----
 4 files changed, 155 insertions(+), 16 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h
index 54b2c349c195..1376219a6c7e 100644
--- a/drivers/net/ethernet/intel/ice/ice.h
+++ b/drivers/net/ethernet/intel/ice/ice.h
@@ -174,6 +174,8 @@ struct ice_sw {
 	struct ice_pf *pf;
 	u16 sw_id;		/* switch ID for this switch */
 	u16 bridge_mode;	/* VEB/VEPA/Port Virtualizer */
+	struct ice_vsi *dflt_vsi;	/* default VSI for this switch */
+	u8 dflt_vsi_ena:1;	/* true if above dflt_vsi is enabled */
 };
 
 enum ice_state {
diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c
index a47241e92d0b..b5cd275606bb 100644
--- a/drivers/net/ethernet/intel/ice/ice_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_lib.c
@@ -2742,3 +2742,121 @@ cfg_mac_fltr_exit:
 	ice_free_fltr_list(&vsi->back->pdev->dev, &tmp_add_list);
 	return status;
 }
+
+/**
+ * ice_is_dflt_vsi_in_use - check if the default forwarding VSI is being used
+ * @sw: switch to check if its default forwarding VSI is free
+ *
+ * Return true if the default forwarding VSI is already being used, else returns
+ * false signalling that it's available to use.
+ */
+bool ice_is_dflt_vsi_in_use(struct ice_sw *sw)
+{
+	return (sw->dflt_vsi && sw->dflt_vsi_ena);
+}
+
+/**
+ * ice_is_vsi_dflt_vsi - check if the VSI passed in is the default VSI
+ * @sw: switch for the default forwarding VSI to compare against
+ * @vsi: VSI to compare against default forwarding VSI
+ *
+ * If this VSI passed in is the default forwarding VSI then return true, else
+ * return false
+ */
+bool ice_is_vsi_dflt_vsi(struct ice_sw *sw, struct ice_vsi *vsi)
+{
+	return (sw->dflt_vsi == vsi && sw->dflt_vsi_ena);
+}
+
+/**
+ * ice_set_dflt_vsi - set the default forwarding VSI
+ * @sw: switch used to assign the default forwarding VSI
+ * @vsi: VSI getting set as the default forwarding VSI on the switch
+ *
+ * If the VSI passed in is already the default VSI and it's enabled just return
+ * success.
+ *
+ * If there is already a default VSI on the switch and it's enabled then return
+ * -EEXIST since there can only be one default VSI per switch.
+ *
+ *  Otherwise try to set the VSI passed in as the switch's default VSI and
+ *  return the result.
+ */
+int ice_set_dflt_vsi(struct ice_sw *sw, struct ice_vsi *vsi)
+{
+	enum ice_status status;
+	struct device *dev;
+
+	if (!sw || !vsi)
+		return -EINVAL;
+
+	dev = ice_pf_to_dev(vsi->back);
+
+	/* the VSI passed in is already the default VSI */
+	if (ice_is_vsi_dflt_vsi(sw, vsi)) {
+		dev_dbg(dev, "VSI %d passed in is already the default forwarding VSI, nothing to do\n",
+			vsi->vsi_num);
+		return 0;
+	}
+
+	/* another VSI is already the default VSI for this switch */
+	if (ice_is_dflt_vsi_in_use(sw)) {
+		dev_err(dev,
+			"Default forwarding VSI %d already in use, disable it and try again\n",
+			sw->dflt_vsi->vsi_num);
+		return -EEXIST;
+	}
+
+	status = ice_cfg_dflt_vsi(&vsi->back->hw, vsi->idx, true, ICE_FLTR_RX);
+	if (status) {
+		dev_err(dev,
+			"Failed to set VSI %d as the default forwarding VSI, error %d\n",
+			vsi->vsi_num, status);
+		return -EIO;
+	}
+
+	sw->dflt_vsi = vsi;
+	sw->dflt_vsi_ena = true;
+
+	return 0;
+}
+
+/**
+ * ice_clear_dflt_vsi - clear the default forwarding VSI
+ * @sw: switch used to clear the default VSI
+ *
+ * If the switch has no default VSI or it's not enabled then return error.
+ *
+ * Otherwise try to clear the default VSI and return the result.
+ */
+int ice_clear_dflt_vsi(struct ice_sw *sw)
+{
+	struct ice_vsi *dflt_vsi;
+	enum ice_status status;
+	struct device *dev;
+
+	if (!sw)
+		return -EINVAL;
+
+	dev = ice_pf_to_dev(sw->pf);
+
+	dflt_vsi = sw->dflt_vsi;
+
+	/* there is no default VSI configured */
+	if (!ice_is_dflt_vsi_in_use(sw))
+		return -ENODEV;
+
+	status = ice_cfg_dflt_vsi(&dflt_vsi->back->hw, dflt_vsi->idx, false,
+				  ICE_FLTR_RX);
+	if (status) {
+		dev_err(dev,
+			"Failed to clear the default forwarding VSI %d, error %d\n",
+			dflt_vsi->vsi_num, status);
+		return -EIO;
+	}
+
+	sw->dflt_vsi = NULL;
+	sw->dflt_vsi_ena = false;
+
+	return 0;
+}
diff --git a/drivers/net/ethernet/intel/ice/ice_lib.h b/drivers/net/ethernet/intel/ice/ice_lib.h
index 6e31e30aba39..68fd0d4505c2 100644
--- a/drivers/net/ethernet/intel/ice/ice_lib.h
+++ b/drivers/net/ethernet/intel/ice/ice_lib.h
@@ -103,4 +103,12 @@ enum ice_status
 ice_vsi_cfg_mac_fltr(struct ice_vsi *vsi, const u8 *macaddr, bool set);
 
 bool ice_is_safe_mode(struct ice_pf *pf);
+
+bool ice_is_dflt_vsi_in_use(struct ice_sw *sw);
+
+bool ice_is_vsi_dflt_vsi(struct ice_sw *sw, struct ice_vsi *vsi);
+
+int ice_set_dflt_vsi(struct ice_sw *sw, struct ice_vsi *vsi);
+
+int ice_clear_dflt_vsi(struct ice_sw *sw);
 #endif /* !_ICE_LIB_H_ */
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 1fc6c350487b..639de467f120 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -379,25 +379,29 @@ static int ice_vsi_sync_fltr(struct ice_vsi *vsi)
 		clear_bit(ICE_VSI_FLAG_PROMISC_CHANGED, vsi->flags);
 		if (vsi->current_netdev_flags & IFF_PROMISC) {
 			/* Apply Rx filter rule to get traffic from wire */
-			status = ice_cfg_dflt_vsi(hw, vsi->idx, true,
-						  ICE_FLTR_RX);
-			if (status) {
-				netdev_err(netdev, "Error setting default VSI %i Rx rule\n",
-					   vsi->vsi_num);
-				vsi->current_netdev_flags &= ~IFF_PROMISC;
-				err = -EIO;
-				goto out_promisc;
+			if (!ice_is_dflt_vsi_in_use(pf->first_sw)) {
+				err = ice_set_dflt_vsi(pf->first_sw, vsi);
+				if (err && err != -EEXIST) {
+					netdev_err(netdev,
+						   "Error %d setting default VSI %i Rx rule\n",
+						   err, vsi->vsi_num);
+					vsi->current_netdev_flags &=
+						~IFF_PROMISC;
+					goto out_promisc;
+				}
 			}
 		} else {
 			/* Clear Rx filter to remove traffic from wire */
-			status = ice_cfg_dflt_vsi(hw, vsi->idx, false,
-						  ICE_FLTR_RX);
-			if (status) {
-				netdev_err(netdev, "Error clearing default VSI %i Rx rule\n",
-					   vsi->vsi_num);
-				vsi->current_netdev_flags |= IFF_PROMISC;
-				err = -EIO;
-				goto out_promisc;
+			if (ice_is_vsi_dflt_vsi(pf->first_sw, vsi)) {
+				err = ice_clear_dflt_vsi(pf->first_sw);
+				if (err) {
+					netdev_err(netdev,
+						   "Error %d clearing default VSI %i Rx rule\n",
+						   err, vsi->vsi_num);
+					vsi->current_netdev_flags |=
+						IFF_PROMISC;
+					goto out_promisc;
+				}
 			}
 		}
 	}
@@ -4671,6 +4675,13 @@ static void ice_rebuild(struct ice_pf *pf, enum ice_reset_req reset_type)
 		goto err_init_ctrlq;
 	}
 
+	if (pf->first_sw->dflt_vsi_ena)
+		dev_info(dev,
+			 "Clearing default VSI, re-enable after reset completes\n");
+	/* clear the default VSI configuration if it exists */
+	pf->first_sw->dflt_vsi = NULL;
+	pf->first_sw->dflt_vsi_ena = false;
+
 	ice_clear_pxe_mode(hw);
 
 	ret = ice_get_caps(hw);
-- 
cgit v1.2.3-59-g8ed1b


From 005881bcf99d11f1e409e286f57fc10ee41af224 Mon Sep 17 00:00:00 2001
From: Brett Creeley <brett.creeley@intel.com>
Date: Thu, 12 Dec 2019 03:12:56 -0800
Subject: ice: Add ice_for_each_vf() macro

Currently we do "for (i = 0; i < pf->num_alloc_vfs; i++)" all over the
place. Many other places use macros to contain this repeated for loop,
So create the macro ice_for_each_vf(pf, i) that does the same thing.

There were a couple places we were using one loop variable and a VF
iterator, which were changed to using a local variable within the
ice_for_each_vf() macro.

Also in ice_alloc_vfs() we were setting pf->num_alloc_vfs after doing
"for (i = 0; i < num_alloc_vfs; i++)". Instead assign pf->num_alloc_vfs
right after allocating memory for the pf->vf array.

Signed-off-by: Brett Creeley <brett.creeley@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_ethtool.c     |  7 +++++--
 drivers/net/ethernet/intel/ice/ice_main.c        |  4 ++--
 drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c | 23 ++++++++++++-----------
 drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h |  3 +++
 4 files changed, 22 insertions(+), 15 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c
index 9ebd93e79aeb..b815f1df5698 100644
--- a/drivers/net/ethernet/intel/ice/ice_ethtool.c
+++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c
@@ -283,12 +283,15 @@ out:
  */
 static bool ice_active_vfs(struct ice_pf *pf)
 {
-	struct ice_vf *vf = pf->vf;
 	int i;
 
-	for (i = 0; i < pf->num_alloc_vfs; i++, vf++)
+	ice_for_each_vf(pf, i) {
+		struct ice_vf *vf = &pf->vf[i];
+
 		if (test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states))
 			return true;
+	}
+
 	return false;
 }
 
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 639de467f120..0c46e7fe7250 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -476,7 +476,7 @@ ice_prepare_for_reset(struct ice_pf *pf)
 		ice_vc_notify_reset(pf);
 
 	/* Disable VFs until reset is completed */
-	for (i = 0; i < pf->num_alloc_vfs; i++)
+	ice_for_each_vf(pf, i)
 		ice_set_vf_state_qs_dis(&pf->vf[i]);
 
 	/* clear SW filtering DB */
@@ -1295,7 +1295,7 @@ static void ice_handle_mdd_event(struct ice_pf *pf)
 	}
 
 	/* check to see if one of the VFs caused the MDD */
-	for (i = 0; i < pf->num_alloc_vfs; i++) {
+	ice_for_each_vf(pf, i) {
 		struct ice_vf *vf = &pf->vf[i];
 
 		bool vf_mdd_detected = false;
diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c
index 3860e9c964b9..88cb0f9e928e 100644
--- a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c
@@ -78,10 +78,11 @@ ice_vc_vf_broadcast(struct ice_pf *pf, enum virtchnl_ops v_opcode,
 		    enum virtchnl_status_code v_retval, u8 *msg, u16 msglen)
 {
 	struct ice_hw *hw = &pf->hw;
-	struct ice_vf *vf = pf->vf;
 	int i;
 
-	for (i = 0; i < pf->num_alloc_vfs; i++, vf++) {
+	ice_for_each_vf(pf, i) {
+		struct ice_vf *vf = &pf->vf[i];
+
 		/* Not all vfs are enabled so skip the ones that are not */
 		if (!test_bit(ICE_VF_STATE_INIT, vf->vf_states) &&
 		    !test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states))
@@ -331,7 +332,7 @@ void ice_free_vfs(struct ice_pf *pf)
 		usleep_range(1000, 2000);
 
 	/* Avoid wait time by stopping all VFs at the same time */
-	for (i = 0; i < pf->num_alloc_vfs; i++)
+	ice_for_each_vf(pf, i)
 		if (test_bit(ICE_VF_STATE_QS_ENA, pf->vf[i].vf_states))
 			ice_dis_vf_qs(&pf->vf[i]);
 
@@ -1077,7 +1078,7 @@ static bool ice_config_res_vfs(struct ice_pf *pf)
 		ice_irq_dynamic_ena(hw, NULL, NULL);
 
 	/* Finish resetting each VF and allocate resources */
-	for (v = 0; v < pf->num_alloc_vfs; v++) {
+	ice_for_each_vf(pf, v) {
 		struct ice_vf *vf = &pf->vf[v];
 
 		vf->num_vf_qs = pf->num_vf_qps;
@@ -1120,10 +1121,10 @@ bool ice_reset_all_vfs(struct ice_pf *pf, bool is_vflr)
 		return false;
 
 	/* Begin reset on all VFs at once */
-	for (v = 0; v < pf->num_alloc_vfs; v++)
+	ice_for_each_vf(pf, v)
 		ice_trigger_vf_reset(&pf->vf[v], is_vflr, true);
 
-	for (v = 0; v < pf->num_alloc_vfs; v++) {
+	ice_for_each_vf(pf, v) {
 		struct ice_vsi *vsi;
 
 		vf = &pf->vf[v];
@@ -1168,7 +1169,7 @@ bool ice_reset_all_vfs(struct ice_pf *pf, bool is_vflr)
 		dev_warn(dev, "VF reset check timeout\n");
 
 	/* free VF resources to begin resetting the VSI state */
-	for (v = 0; v < pf->num_alloc_vfs; v++) {
+	ice_for_each_vf(pf, v) {
 		vf = &pf->vf[v];
 
 		ice_free_vf_res(vf);
@@ -1308,7 +1309,7 @@ void ice_vc_notify_link_state(struct ice_pf *pf)
 {
 	int i;
 
-	for (i = 0; i < pf->num_alloc_vfs; i++)
+	ice_for_each_vf(pf, i)
 		ice_vc_notify_vf_link_state(&pf->vf[i]);
 }
 
@@ -1392,9 +1393,10 @@ static int ice_alloc_vfs(struct ice_pf *pf, u16 num_alloc_vfs)
 		goto err_pci_disable_sriov;
 	}
 	pf->vf = vfs;
+	pf->num_alloc_vfs = num_alloc_vfs;
 
 	/* apply default profile */
-	for (i = 0; i < num_alloc_vfs; i++) {
+	ice_for_each_vf(pf, i) {
 		vfs[i].pf = pf;
 		vfs[i].vf_sw_id = pf->first_sw;
 		vfs[i].vf_id = i;
@@ -1403,7 +1405,6 @@ static int ice_alloc_vfs(struct ice_pf *pf, u16 num_alloc_vfs)
 		set_bit(ICE_VIRTCHNL_VF_CAP_L2, &vfs[i].vf_caps);
 		vfs[i].spoofchk = true;
 	}
-	pf->num_alloc_vfs = num_alloc_vfs;
 
 	/* VF resources get allocated with initialization */
 	if (!ice_config_res_vfs(pf)) {
@@ -1542,7 +1543,7 @@ void ice_process_vflr_event(struct ice_pf *pf)
 	    !pf->num_alloc_vfs)
 		return;
 
-	for (vf_id = 0; vf_id < pf->num_alloc_vfs; vf_id++) {
+	ice_for_each_vf(pf, vf_id) {
 		struct ice_vf *vf = &pf->vf[vf_id];
 		u32 reg_idx, bit_idx;
 
diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h
index 611f45100438..4647d636ed36 100644
--- a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h
+++ b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h
@@ -40,6 +40,9 @@
 #define ICE_DFLT_INTR_PER_VF		(ICE_DFLT_QS_PER_VF + 1)
 #define ICE_MAX_VF_RESET_WAIT		15
 
+#define ice_for_each_vf(pf, i) \
+	for ((i) = 0; (i) < (pf)->num_alloc_vfs; (i)++)
+
 /* Specific VF states */
 enum ice_vf_states {
 	ICE_VF_STATE_INIT = 0,		/* PF is initializing VF */
-- 
cgit v1.2.3-59-g8ed1b


From 118e0e10025866f92fc9f6403926cfed033e4078 Mon Sep 17 00:00:00 2001
From: Michal Swiatkowski <michal.swiatkowski@intel.com>
Date: Thu, 12 Dec 2019 03:12:57 -0800
Subject: ice: Set default value for ITR in alloc function

When the user sets itr_setting to zero from ethtool -C, the driver changes
this value to default in ice_cfg_itr (for example after changing ring
param). Remove code that sets default value in ice_cfg_itr and move it to
place where the driver allocates q_vectors.

Signed-off-by: Michal Swiatkowski <michal.swiatkowski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_base.c | 13 ++++---------
 1 file changed, 4 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_base.c b/drivers/net/ethernet/intel/ice/ice_base.c
index 77d6a0291e97..d4559d45288f 100644
--- a/drivers/net/ethernet/intel/ice/ice_base.c
+++ b/drivers/net/ethernet/intel/ice/ice_base.c
@@ -93,7 +93,8 @@ static int ice_pf_rxq_wait(struct ice_pf *pf, int pf_q, bool ena)
  * @vsi: the VSI being configured
  * @v_idx: index of the vector in the VSI struct
  *
- * We allocate one q_vector. If allocation fails we return -ENOMEM.
+ * We allocate one q_vector and set default value for ITR setting associated
+ * with this q_vector. If allocation fails we return -ENOMEM.
  */
 static int ice_vsi_alloc_q_vector(struct ice_vsi *vsi, int v_idx)
 {
@@ -108,6 +109,8 @@ static int ice_vsi_alloc_q_vector(struct ice_vsi *vsi, int v_idx)
 
 	q_vector->vsi = vsi;
 	q_vector->v_idx = v_idx;
+	q_vector->tx.itr_setting = ICE_DFLT_TX_ITR;
+	q_vector->rx.itr_setting = ICE_DFLT_RX_ITR;
 	if (vsi->type == ICE_VSI_VF)
 		goto out;
 	/* only set affinity_mask if the CPU is online */
@@ -674,10 +677,6 @@ void ice_cfg_itr(struct ice_hw *hw, struct ice_q_vector *q_vector)
 	if (q_vector->num_ring_rx) {
 		struct ice_ring_container *rc = &q_vector->rx;
 
-		/* if this value is set then don't overwrite with default */
-		if (!rc->itr_setting)
-			rc->itr_setting = ICE_DFLT_RX_ITR;
-
 		rc->target_itr = ITR_TO_REG(rc->itr_setting);
 		rc->next_update = jiffies + 1;
 		rc->current_itr = rc->target_itr;
@@ -688,10 +687,6 @@ void ice_cfg_itr(struct ice_hw *hw, struct ice_q_vector *q_vector)
 	if (q_vector->num_ring_tx) {
 		struct ice_ring_container *rc = &q_vector->tx;
 
-		/* if this value is set then don't overwrite with default */
-		if (!rc->itr_setting)
-			rc->itr_setting = ICE_DFLT_TX_ITR;
-
 		rc->target_itr = ITR_TO_REG(rc->itr_setting);
 		rc->next_update = jiffies + 1;
 		rc->current_itr = rc->target_itr;
-- 
cgit v1.2.3-59-g8ed1b


From 61dc79ced7aaff5420a4d2f7f63925c607fa9395 Mon Sep 17 00:00:00 2001
From: Michal Swiatkowski <michal.swiatkowski@intel.com>
Date: Thu, 12 Dec 2019 03:12:58 -0800
Subject: ice: Restore interrupt throttle settings after VSI rebuild

After each rebuild driver deallocates q_vectors, so the interrupt
throttle rate (ITR) settings get lost.

Create a function to save and restore ITR for each queue. If a user
increases the number of queues, restore all the previous queue
settings for each existing queue, and the additional queues will
get the default setting.

Signed-off-by: Michal Swiatkowski <michal.swiatkowski@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_lib.c  | 102 ++++++++++++++++++++++++++++++
 drivers/net/ethernet/intel/ice/ice_txrx.h |   6 ++
 2 files changed, 108 insertions(+)

diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c
index b5cd275606bb..4cfad81ba496 100644
--- a/drivers/net/ethernet/intel/ice/ice_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_lib.c
@@ -2403,6 +2403,97 @@ int ice_vsi_release(struct ice_vsi *vsi)
 	return 0;
 }
 
+/**
+ * ice_vsi_rebuild_update_coalesce - set coalesce for a q_vector
+ * @q_vector: pointer to q_vector which is being updated
+ * @coalesce: pointer to array of struct with stored coalesce
+ *
+ * Set coalesce param in q_vector and update these parameters in HW.
+ */
+static void
+ice_vsi_rebuild_update_coalesce(struct ice_q_vector *q_vector,
+				struct ice_coalesce_stored *coalesce)
+{
+	struct ice_ring_container *rx_rc = &q_vector->rx;
+	struct ice_ring_container *tx_rc = &q_vector->tx;
+	struct ice_hw *hw = &q_vector->vsi->back->hw;
+
+	tx_rc->itr_setting = coalesce->itr_tx;
+	rx_rc->itr_setting = coalesce->itr_rx;
+
+	/* dynamic ITR values will be updated during Tx/Rx */
+	if (!ITR_IS_DYNAMIC(tx_rc->itr_setting))
+		wr32(hw, GLINT_ITR(tx_rc->itr_idx, q_vector->reg_idx),
+		     ITR_REG_ALIGN(tx_rc->itr_setting) >>
+		     ICE_ITR_GRAN_S);
+	if (!ITR_IS_DYNAMIC(rx_rc->itr_setting))
+		wr32(hw, GLINT_ITR(rx_rc->itr_idx, q_vector->reg_idx),
+		     ITR_REG_ALIGN(rx_rc->itr_setting) >>
+		     ICE_ITR_GRAN_S);
+
+	q_vector->intrl = coalesce->intrl;
+	wr32(hw, GLINT_RATE(q_vector->reg_idx),
+	     ice_intrl_usec_to_reg(q_vector->intrl, hw->intrl_gran));
+}
+
+/**
+ * ice_vsi_rebuild_get_coalesce - get coalesce from all q_vectors
+ * @vsi: VSI connected with q_vectors
+ * @coalesce: array of struct with stored coalesce
+ *
+ * Returns array size.
+ */
+static int
+ice_vsi_rebuild_get_coalesce(struct ice_vsi *vsi,
+			     struct ice_coalesce_stored *coalesce)
+{
+	int i;
+
+	ice_for_each_q_vector(vsi, i) {
+		struct ice_q_vector *q_vector = vsi->q_vectors[i];
+
+		coalesce[i].itr_tx = q_vector->tx.itr_setting;
+		coalesce[i].itr_rx = q_vector->rx.itr_setting;
+		coalesce[i].intrl = q_vector->intrl;
+	}
+
+	return vsi->num_q_vectors;
+}
+
+/**
+ * ice_vsi_rebuild_set_coalesce - set coalesce from earlier saved arrays
+ * @vsi: VSI connected with q_vectors
+ * @coalesce: pointer to array of struct with stored coalesce
+ * @size: size of coalesce array
+ *
+ * Before this function, ice_vsi_rebuild_get_coalesce should be called to save
+ * ITR params in arrays. If size is 0 or coalesce wasn't stored set coalesce
+ * to default value.
+ */
+static void
+ice_vsi_rebuild_set_coalesce(struct ice_vsi *vsi,
+			     struct ice_coalesce_stored *coalesce, int size)
+{
+	int i;
+
+	if ((size && !coalesce) || !vsi)
+		return;
+
+	for (i = 0; i < size && i < vsi->num_q_vectors; i++)
+		ice_vsi_rebuild_update_coalesce(vsi->q_vectors[i],
+						&coalesce[i]);
+
+	for (; i < vsi->num_q_vectors; i++) {
+		struct ice_coalesce_stored coalesce_dflt = {
+			.itr_tx = ICE_DFLT_TX_ITR,
+			.itr_rx = ICE_DFLT_RX_ITR,
+			.intrl = 0
+		};
+		ice_vsi_rebuild_update_coalesce(vsi->q_vectors[i],
+						&coalesce_dflt);
+	}
+}
+
 /**
  * ice_vsi_rebuild - Rebuild VSI after reset
  * @vsi: VSI to be rebuild
@@ -2413,6 +2504,8 @@ int ice_vsi_release(struct ice_vsi *vsi)
 int ice_vsi_rebuild(struct ice_vsi *vsi, bool init_vsi)
 {
 	u16 max_txqs[ICE_MAX_TRAFFIC_CLASS] = { 0 };
+	struct ice_coalesce_stored *coalesce;
+	int prev_num_q_vectors = 0;
 	struct ice_vf *vf = NULL;
 	enum ice_status status;
 	struct ice_pf *pf;
@@ -2425,6 +2518,11 @@ int ice_vsi_rebuild(struct ice_vsi *vsi, bool init_vsi)
 	if (vsi->type == ICE_VSI_VF)
 		vf = &pf->vf[vsi->vf_id];
 
+	coalesce = kcalloc(vsi->num_q_vectors,
+			   sizeof(struct ice_coalesce_stored), GFP_KERNEL);
+	if (coalesce)
+		prev_num_q_vectors = ice_vsi_rebuild_get_coalesce(vsi,
+								  coalesce);
 	ice_rm_vsi_lan_cfg(vsi->port_info, vsi->idx);
 	ice_vsi_free_q_vectors(vsi);
 
@@ -2537,6 +2635,9 @@ int ice_vsi_rebuild(struct ice_vsi *vsi, bool init_vsi)
 			return ice_schedule_reset(pf, ICE_RESET_PFR);
 		}
 	}
+	ice_vsi_rebuild_set_coalesce(vsi, coalesce, prev_num_q_vectors);
+	kfree(coalesce);
+
 	return 0;
 
 err_vectors:
@@ -2551,6 +2652,7 @@ err_rings:
 err_vsi:
 	ice_vsi_clear(vsi);
 	set_bit(__ICE_RESET_FAILED, pf->state);
+	kfree(coalesce);
 	return ret;
 }
 
diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.h b/drivers/net/ethernet/intel/ice/ice_txrx.h
index a84cc0e6dd27..a86270696df1 100644
--- a/drivers/net/ethernet/intel/ice/ice_txrx.h
+++ b/drivers/net/ethernet/intel/ice/ice_txrx.h
@@ -341,6 +341,12 @@ struct ice_ring_container {
 	u16 itr_setting;
 };
 
+struct ice_coalesce_stored {
+	u16 itr_tx;
+	u16 itr_rx;
+	u8 intrl;
+};
+
 /* iterator for handling rings in ring container */
 #define ice_for_each_ring(pos, head) \
 	for (pos = (head).ring; pos; pos = pos->next)
-- 
cgit v1.2.3-59-g8ed1b


From 11c25c2f2eafcb414a9cd782fb1ea207c03be84d Mon Sep 17 00:00:00 2001
From: Michal Swiatkowski <michal.swiatkowski@intel.com>
Date: Thu, 12 Dec 2019 03:12:59 -0800
Subject: ice: Return error on not supported ethtool -C parameters

Check for all unused parameters, if ethtool sent one of them,
print info about that and return error.

Signed-off-by: Michal Swiatkowski <michal.swiatkowski@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_ethtool.c | 50 ++++++++++++++++++++++++++++
 1 file changed, 50 insertions(+)

diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c
index b815f1df5698..f395457b728f 100644
--- a/drivers/net/ethernet/intel/ice/ice_ethtool.c
+++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c
@@ -3587,6 +3587,53 @@ ice_set_q_coalesce(struct ice_vsi *vsi, struct ethtool_coalesce *ec, int q_num)
 	return 0;
 }
 
+/**
+ * ice_is_coalesce_param_invalid - check for unsupported coalesce parameters
+ * @netdev: pointer to the netdev associated with this query
+ * @ec: ethtool structure to fill with driver's coalesce settings
+ *
+ * Print netdev info if driver doesn't support one of the parameters
+ * and return error. When any parameters will be implemented, remove only
+ * this parameter from param array.
+ */
+static int
+ice_is_coalesce_param_invalid(struct net_device *netdev,
+			      struct ethtool_coalesce *ec)
+{
+	struct ice_ethtool_not_used {
+		u32 value;
+		const char *name;
+	} param[] = {
+		{ec->stats_block_coalesce_usecs, "stats-block-usecs"},
+		{ec->rate_sample_interval, "sample-interval"},
+		{ec->pkt_rate_low, "pkt-rate-low"},
+		{ec->pkt_rate_high, "pkt-rate-high"},
+		{ec->rx_max_coalesced_frames, "rx-frames"},
+		{ec->rx_coalesce_usecs_irq, "rx-usecs-irq"},
+		{ec->rx_max_coalesced_frames_irq, "rx-frames-irq"},
+		{ec->tx_max_coalesced_frames, "tx-frames"},
+		{ec->tx_coalesce_usecs_irq, "tx-usecs-irq"},
+		{ec->tx_max_coalesced_frames_irq, "tx-frames-irq"},
+		{ec->rx_coalesce_usecs_low, "rx-usecs-low"},
+		{ec->rx_max_coalesced_frames_low, "rx-frames-low"},
+		{ec->tx_coalesce_usecs_low, "tx-usecs-low"},
+		{ec->tx_max_coalesced_frames_low, "tx-frames-low"},
+		{ec->rx_max_coalesced_frames_high, "rx-frames-high"},
+		{ec->tx_max_coalesced_frames_high, "tx-frames-high"}
+	};
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(param); i++) {
+		if (param[i].value) {
+			netdev_info(netdev, "Setting %s not supported\n",
+				    param[i].name);
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
 /**
  * __ice_set_coalesce - set ITR/INTRL values for the device
  * @netdev: pointer to the netdev associated with this query
@@ -3603,6 +3650,9 @@ __ice_set_coalesce(struct net_device *netdev, struct ethtool_coalesce *ec,
 	struct ice_netdev_priv *np = netdev_priv(netdev);
 	struct ice_vsi *vsi = np->vsi;
 
+	if (ice_is_coalesce_param_invalid(netdev, ec))
+		return -EINVAL;
+
 	if (q_num < 0) {
 		int v_idx;
 
-- 
cgit v1.2.3-59-g8ed1b


From f57a683dedbbe80af69708ae862f7ac88a441e5c Mon Sep 17 00:00:00 2001
From: Vignesh Sridhar <vignesh.sridhar@intel.com>
Date: Thu, 12 Dec 2019 03:13:00 -0800
Subject: ice: Remove Rx flex descriptor programming

Remove Rx flex descriptor metadata and flag programming; per specification
these registers cannot be written to as they are read only.

Signed-off-by: Vignesh Sridhar <vignesh.sridhar@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_common.c     | 104 ------------------------
 drivers/net/ethernet/intel/ice/ice_hw_autogen.h |   9 --
 2 files changed, 113 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index fb1d930470c7..a03b4fdc01e6 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -7,25 +7,6 @@
 
 #define ICE_PF_RESET_WAIT_COUNT	200
 
-#define ICE_PROG_FLEX_ENTRY(hw, rxdid, mdid, idx) \
-	wr32((hw), GLFLXP_RXDID_FLX_WRD_##idx(rxdid), \
-	     ((ICE_RX_OPC_MDID << \
-	       GLFLXP_RXDID_FLX_WRD_##idx##_RXDID_OPCODE_S) & \
-	      GLFLXP_RXDID_FLX_WRD_##idx##_RXDID_OPCODE_M) | \
-	     (((mdid) << GLFLXP_RXDID_FLX_WRD_##idx##_PROT_MDID_S) & \
-	      GLFLXP_RXDID_FLX_WRD_##idx##_PROT_MDID_M))
-
-#define ICE_PROG_FLG_ENTRY(hw, rxdid, flg_0, flg_1, flg_2, flg_3, idx) \
-	wr32((hw), GLFLXP_RXDID_FLAGS(rxdid, idx), \
-	     (((flg_0) << GLFLXP_RXDID_FLAGS_FLEXIFLAG_4N_S) & \
-	      GLFLXP_RXDID_FLAGS_FLEXIFLAG_4N_M) | \
-	     (((flg_1) << GLFLXP_RXDID_FLAGS_FLEXIFLAG_4N_1_S) & \
-	      GLFLXP_RXDID_FLAGS_FLEXIFLAG_4N_1_M) | \
-	     (((flg_2) << GLFLXP_RXDID_FLAGS_FLEXIFLAG_4N_2_S) & \
-	      GLFLXP_RXDID_FLAGS_FLEXIFLAG_4N_2_M) | \
-	     (((flg_3) << GLFLXP_RXDID_FLAGS_FLEXIFLAG_4N_3_S) & \
-	      GLFLXP_RXDID_FLAGS_FLEXIFLAG_4N_3_M))
-
 /**
  * ice_set_mac_type - Sets MAC type
  * @hw: pointer to the HW structure
@@ -347,88 +328,6 @@ ice_aq_get_link_info(struct ice_port_info *pi, bool ena_lse,
 	return 0;
 }
 
-/**
- * ice_init_flex_flags
- * @hw: pointer to the hardware structure
- * @prof_id: Rx Descriptor Builder profile ID
- *
- * Function to initialize Rx flex flags
- */
-static void ice_init_flex_flags(struct ice_hw *hw, enum ice_rxdid prof_id)
-{
-	u8 idx = 0;
-
-	/* Flex-flag fields (0-2) are programmed with FLG64 bits with layout:
-	 * flexiflags0[5:0] - TCP flags, is_packet_fragmented, is_packet_UDP_GRE
-	 * flexiflags1[3:0] - Not used for flag programming
-	 * flexiflags2[7:0] - Tunnel and VLAN types
-	 * 2 invalid fields in last index
-	 */
-	switch (prof_id) {
-	/* Rx flex flags are currently programmed for the NIC profiles only.
-	 * Different flag bit programming configurations can be added per
-	 * profile as needed.
-	 */
-	case ICE_RXDID_FLEX_NIC:
-	case ICE_RXDID_FLEX_NIC_2:
-		ICE_PROG_FLG_ENTRY(hw, prof_id, ICE_FLG_PKT_FRG,
-				   ICE_FLG_UDP_GRE, ICE_FLG_PKT_DSI,
-				   ICE_FLG_FIN, idx++);
-		/* flex flag 1 is not used for flexi-flag programming, skipping
-		 * these four FLG64 bits.
-		 */
-		ICE_PROG_FLG_ENTRY(hw, prof_id, ICE_FLG_SYN, ICE_FLG_RST,
-				   ICE_FLG_PKT_DSI, ICE_FLG_PKT_DSI, idx++);
-		ICE_PROG_FLG_ENTRY(hw, prof_id, ICE_FLG_PKT_DSI,
-				   ICE_FLG_PKT_DSI, ICE_FLG_EVLAN_x8100,
-				   ICE_FLG_EVLAN_x9100, idx++);
-		ICE_PROG_FLG_ENTRY(hw, prof_id, ICE_FLG_VLAN_x8100,
-				   ICE_FLG_TNL_VLAN, ICE_FLG_TNL_MAC,
-				   ICE_FLG_TNL0, idx++);
-		ICE_PROG_FLG_ENTRY(hw, prof_id, ICE_FLG_TNL1, ICE_FLG_TNL2,
-				   ICE_FLG_PKT_DSI, ICE_FLG_PKT_DSI, idx);
-		break;
-
-	default:
-		ice_debug(hw, ICE_DBG_INIT,
-			  "Flag programming for profile ID %d not supported\n",
-			  prof_id);
-	}
-}
-
-/**
- * ice_init_flex_flds
- * @hw: pointer to the hardware structure
- * @prof_id: Rx Descriptor Builder profile ID
- *
- * Function to initialize flex descriptors
- */
-static void ice_init_flex_flds(struct ice_hw *hw, enum ice_rxdid prof_id)
-{
-	enum ice_flex_rx_mdid mdid;
-
-	switch (prof_id) {
-	case ICE_RXDID_FLEX_NIC:
-	case ICE_RXDID_FLEX_NIC_2:
-		ICE_PROG_FLEX_ENTRY(hw, prof_id, ICE_RX_MDID_HASH_LOW, 0);
-		ICE_PROG_FLEX_ENTRY(hw, prof_id, ICE_RX_MDID_HASH_HIGH, 1);
-		ICE_PROG_FLEX_ENTRY(hw, prof_id, ICE_RX_MDID_FLOW_ID_LOWER, 2);
-
-		mdid = (prof_id == ICE_RXDID_FLEX_NIC_2) ?
-			ICE_RX_MDID_SRC_VSI : ICE_RX_MDID_FLOW_ID_HIGH;
-
-		ICE_PROG_FLEX_ENTRY(hw, prof_id, mdid, 3);
-
-		ice_init_flex_flags(hw, prof_id);
-		break;
-
-	default:
-		ice_debug(hw, ICE_DBG_INIT,
-			  "Field init for profile ID %d not supported\n",
-			  prof_id);
-	}
-}
-
 /**
  * ice_init_fltr_mgmt_struct - initializes filter management list and locks
  * @hw: pointer to the HW struct
@@ -882,9 +781,6 @@ enum ice_status ice_init_hw(struct ice_hw *hw)
 
 	if (status)
 		goto err_unroll_fltr_mgmt_struct;
-
-	ice_init_flex_flds(hw, ICE_RXDID_FLEX_NIC);
-	ice_init_flex_flds(hw, ICE_RXDID_FLEX_NIC_2);
 	status = ice_init_hw_tbls(hw);
 	if (status)
 		goto err_unroll_fltr_mgmt_struct;
diff --git a/drivers/net/ethernet/intel/ice/ice_hw_autogen.h b/drivers/net/ethernet/intel/ice/ice_hw_autogen.h
index e8f32350fed2..f2cababf2561 100644
--- a/drivers/net/ethernet/intel/ice/ice_hw_autogen.h
+++ b/drivers/net/ethernet/intel/ice/ice_hw_autogen.h
@@ -60,15 +60,6 @@
 #define PRTDCB_GENS_DCBX_STATUS_M		ICE_M(0x7, 0)
 #define GL_PREEXT_L2_PMASK0(_i)			(0x0020F0FC + ((_i) * 4))
 #define GL_PREEXT_L2_PMASK1(_i)			(0x0020F108 + ((_i) * 4))
-#define GLFLXP_RXDID_FLAGS(_i, _j)		(0x0045D000 + ((_i) * 4 + (_j) * 256))
-#define GLFLXP_RXDID_FLAGS_FLEXIFLAG_4N_S	0
-#define GLFLXP_RXDID_FLAGS_FLEXIFLAG_4N_M	ICE_M(0x3F, 0)
-#define GLFLXP_RXDID_FLAGS_FLEXIFLAG_4N_1_S	8
-#define GLFLXP_RXDID_FLAGS_FLEXIFLAG_4N_1_M	ICE_M(0x3F, 8)
-#define GLFLXP_RXDID_FLAGS_FLEXIFLAG_4N_2_S	16
-#define GLFLXP_RXDID_FLAGS_FLEXIFLAG_4N_2_M	ICE_M(0x3F, 16)
-#define GLFLXP_RXDID_FLAGS_FLEXIFLAG_4N_3_S	24
-#define GLFLXP_RXDID_FLAGS_FLEXIFLAG_4N_3_M	ICE_M(0x3F, 24)
 #define GLFLXP_RXDID_FLX_WRD_0(_i)		(0x0045c800 + ((_i) * 4))
 #define GLFLXP_RXDID_FLX_WRD_0_PROT_MDID_S	0
 #define GLFLXP_RXDID_FLX_WRD_0_PROT_MDID_M	ICE_M(0xFF, 0)
-- 
cgit v1.2.3-59-g8ed1b


From 26a91525cca8e50854706a9e71ce38795ba1fbcb Mon Sep 17 00:00:00 2001
From: Brett Creeley <brett.creeley@intel.com>
Date: Thu, 12 Dec 2019 03:13:01 -0800
Subject: ice: Fix VF link state when it's IFLA_VF_LINK_STATE_AUTO

Currently the flow for ice_set_vf_link_state() is not configuring link
the same as all other VF link configuration flows. Fix this by only
setting the necessary VF members in ice_set_vf_link_state() and then
call ice_vc_notify_link_state() to actually configure link for the
VF. This made ice_set_pfe_link_forced() unnecessary, so it was
deleted. Also, this commonizes the link flows for the VF to all call
ice_vc_notify_link_state().

Signed-off-by: Brett Creeley <brett.creeley@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_main.c        |  3 +-
 drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c | 56 +++++-------------------
 2 files changed, 12 insertions(+), 47 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 0c46e7fe7250..3adeb7059ac5 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -844,8 +844,7 @@ ice_link_event(struct ice_pf *pf, struct ice_port_info *pi, bool link_up,
 	ice_vsi_link_event(vsi, link_up);
 	ice_print_link_msg(vsi, link_up);
 
-	if (pf->num_alloc_vfs)
-		ice_vc_notify_link_state(pf);
+	ice_vc_notify_link_state(pf);
 
 	return result;
 }
diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c
index 88cb0f9e928e..f080f3af182a 100644
--- a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c
@@ -121,26 +121,6 @@ ice_set_pfe_link(struct ice_vf *vf, struct virtchnl_pf_event *pfe,
 	}
 }
 
-/**
- * ice_set_pfe_link_forced - Force the virtchnl_pf_event link speed/status
- * @vf: pointer to the VF structure
- * @pfe: pointer to the virtchnl_pf_event to set link speed/status for
- * @link_up: whether or not to set the link up/down
- */
-static void
-ice_set_pfe_link_forced(struct ice_vf *vf, struct virtchnl_pf_event *pfe,
-			bool link_up)
-{
-	u16 link_speed;
-
-	if (link_up)
-		link_speed = ICE_AQ_LINK_SPEED_100GB;
-	else
-		link_speed = ICE_AQ_LINK_SPEED_UNKNOWN;
-
-	ice_set_pfe_link(vf, pfe, link_speed, link_up);
-}
-
 /**
  * ice_vc_notify_vf_link_state - Inform a VF of link status
  * @vf: pointer to the VF structure
@@ -161,13 +141,17 @@ static void ice_vc_notify_vf_link_state(struct ice_vf *vf)
 	pfe.severity = PF_EVENT_SEVERITY_INFO;
 
 	/* Always report link is down if the VF queues aren't enabled */
-	if (!vf->num_qs_ena)
+	if (!vf->num_qs_ena) {
 		ice_set_pfe_link(vf, &pfe, ICE_AQ_LINK_SPEED_UNKNOWN, false);
-	else if (vf->link_forced)
-		ice_set_pfe_link_forced(vf, &pfe, vf->link_up);
-	else
-		ice_set_pfe_link(vf, &pfe, ls->link_speed, ls->link_info &
-				 ICE_AQ_LINK_UP);
+	} else if (vf->link_forced) {
+		u16 link_speed = vf->link_up ?
+			ls->link_speed : ICE_AQ_LINK_SPEED_UNKNOWN;
+
+		ice_set_pfe_link(vf, &pfe, link_speed, vf->link_up);
+	} else {
+		ice_set_pfe_link(vf, &pfe, ls->link_speed,
+				 ls->link_info & ICE_AQ_LINK_UP);
+	}
 
 	ice_aq_send_msg_to_vf(hw, vf->vf_id, VIRTCHNL_OP_EVENT,
 			      VIRTCHNL_STATUS_SUCCESS, (u8 *)&pfe,
@@ -3392,28 +3376,18 @@ int ice_set_vf_trust(struct net_device *netdev, int vf_id, bool trusted)
 int ice_set_vf_link_state(struct net_device *netdev, int vf_id, int link_state)
 {
 	struct ice_pf *pf = ice_netdev_to_pf(netdev);
-	struct virtchnl_pf_event pfe = { 0 };
-	struct ice_link_status *ls;
 	struct ice_vf *vf;
-	struct ice_hw *hw;
 
 	if (ice_validate_vf_id(pf, vf_id))
 		return -EINVAL;
 
 	vf = &pf->vf[vf_id];
-	hw = &pf->hw;
-	ls = &pf->hw.port_info->phy.link_info;
-
 	if (ice_check_vf_init(pf, vf))
 		return -EBUSY;
 
-	pfe.event = VIRTCHNL_EVENT_LINK_CHANGE;
-	pfe.severity = PF_EVENT_SEVERITY_INFO;
-
 	switch (link_state) {
 	case IFLA_VF_LINK_STATE_AUTO:
 		vf->link_forced = false;
-		vf->link_up = ls->link_info & ICE_AQ_LINK_UP;
 		break;
 	case IFLA_VF_LINK_STATE_ENABLE:
 		vf->link_forced = true;
@@ -3427,15 +3401,7 @@ int ice_set_vf_link_state(struct net_device *netdev, int vf_id, int link_state)
 		return -EINVAL;
 	}
 
-	if (vf->link_forced)
-		ice_set_pfe_link_forced(vf, &pfe, vf->link_up);
-	else
-		ice_set_pfe_link(vf, &pfe, ls->link_speed, vf->link_up);
-
-	/* Notify the VF of its new link state */
-	ice_aq_send_msg_to_vf(hw, vf->vf_id, VIRTCHNL_OP_EVENT,
-			      VIRTCHNL_STATUS_SUCCESS, (u8 *)&pfe,
-			      sizeof(pfe), NULL);
+	ice_vc_notify_vf_link_state(vf);
 
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From ed4c068d46f69f6f248c6e5a22f69cc84bdb5bb6 Mon Sep 17 00:00:00 2001
From: Brett Creeley <brett.creeley@intel.com>
Date: Thu, 12 Dec 2019 03:13:02 -0800
Subject: ice: Enable ip link show on the PF to display VF unicast MAC(s)

Currently when there are SR-IOV VF(s) and the user does "ip link show <pf
interface>" the VF unicast MAC addresses all show 00:00:00:00:00:00
if the unicast MAC was set via VIRTCHNL (i.e. not administratively set
by the host PF).

This is misleading to the host administrator. Fix this by setting the
VF's dflt_lan_addr.addr when the VF's unicast MAC address is
configured via VIRTCHNL. There are a couple cases where we don't allow
the dflt_lan_addr.addr field to be written. First, If the VF's
pf_set_mac field is true and the VF is not trusted, then we don't allow
the dflt_lan_addr.addr to be modified. Second, if the
dflt_lan_addr.addr has already been set (i.e. via VIRTCHNL).

Also a small refactor was done to separate the flow for add and delete
MAC addresses in order to simplify the logic for error conditions
and set/clear the VF's dflt_lan_addr.addr field.

Signed-off-by: Brett Creeley <brett.creeley@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c | 199 +++++++++++------------
 1 file changed, 99 insertions(+), 100 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c
index f080f3af182a..82b1e7a4cb92 100644
--- a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c
@@ -34,37 +34,6 @@ static int ice_check_vf_init(struct ice_pf *pf, struct ice_vf *vf)
 	return 0;
 }
 
-/**
- * ice_err_to_virt err - translate errors for VF return code
- * @ice_err: error return code
- */
-static enum virtchnl_status_code ice_err_to_virt_err(enum ice_status ice_err)
-{
-	switch (ice_err) {
-	case ICE_SUCCESS:
-		return VIRTCHNL_STATUS_SUCCESS;
-	case ICE_ERR_BAD_PTR:
-	case ICE_ERR_INVAL_SIZE:
-	case ICE_ERR_DEVICE_NOT_SUPPORTED:
-	case ICE_ERR_PARAM:
-	case ICE_ERR_CFG:
-		return VIRTCHNL_STATUS_ERR_PARAM;
-	case ICE_ERR_NO_MEMORY:
-		return VIRTCHNL_STATUS_ERR_NO_MEMORY;
-	case ICE_ERR_NOT_READY:
-	case ICE_ERR_RESET_FAILED:
-	case ICE_ERR_FW_API_VER:
-	case ICE_ERR_AQ_ERROR:
-	case ICE_ERR_AQ_TIMEOUT:
-	case ICE_ERR_AQ_FULL:
-	case ICE_ERR_AQ_NO_WORK:
-	case ICE_ERR_AQ_EMPTY:
-		return VIRTCHNL_STATUS_ERR_ADMIN_QUEUE_ERROR;
-	default:
-		return VIRTCHNL_STATUS_ERR_NOT_SUPPORTED;
-	}
-}
-
 /**
  * ice_vc_vf_broadcast - Broadcast a message to all VFs on PF
  * @pf: pointer to the PF structure
@@ -2483,6 +2452,83 @@ static bool ice_can_vf_change_mac(struct ice_vf *vf)
 	return true;
 }
 
+/**
+ * ice_vc_add_mac_addr - attempt to add the MAC address passed in
+ * @vf: pointer to the VF info
+ * @vsi: pointer to the VF's VSI
+ * @mac_addr: MAC address to add
+ */
+static int
+ice_vc_add_mac_addr(struct ice_vf *vf, struct ice_vsi *vsi, u8 *mac_addr)
+{
+	struct device *dev = ice_pf_to_dev(vf->pf);
+	enum ice_status status;
+
+	/* default unicast MAC already added */
+	if (ether_addr_equal(mac_addr, vf->dflt_lan_addr.addr))
+		return 0;
+
+	if (is_unicast_ether_addr(mac_addr) && !ice_can_vf_change_mac(vf)) {
+		dev_err(dev, "VF attempting to override administratively set MAC address, bring down and up the VF interface to resume normal operation\n");
+		return -EPERM;
+	}
+
+	status = ice_vsi_cfg_mac_fltr(vsi, mac_addr, true);
+	if (status == ICE_ERR_ALREADY_EXISTS) {
+		dev_err(dev, "MAC %pM already exists for VF %d\n", mac_addr,
+			vf->vf_id);
+		return -EEXIST;
+	} else if (status) {
+		dev_err(dev, "Failed to add MAC %pM for VF %d\n, error %d\n",
+			mac_addr, vf->vf_id, status);
+		return -EIO;
+	}
+
+	/* only set dflt_lan_addr once */
+	if (is_zero_ether_addr(vf->dflt_lan_addr.addr) &&
+	    is_unicast_ether_addr(mac_addr))
+		ether_addr_copy(vf->dflt_lan_addr.addr, mac_addr);
+
+	vf->num_mac++;
+
+	return 0;
+}
+
+/**
+ * ice_vc_del_mac_addr - attempt to delete the MAC address passed in
+ * @vf: pointer to the VF info
+ * @vsi: pointer to the VF's VSI
+ * @mac_addr: MAC address to delete
+ */
+static int
+ice_vc_del_mac_addr(struct ice_vf *vf, struct ice_vsi *vsi, u8 *mac_addr)
+{
+	struct device *dev = ice_pf_to_dev(vf->pf);
+	enum ice_status status;
+
+	if (!ice_can_vf_change_mac(vf) &&
+	    ether_addr_equal(mac_addr, vf->dflt_lan_addr.addr))
+		return 0;
+
+	status = ice_vsi_cfg_mac_fltr(vsi, mac_addr, false);
+	if (status == ICE_ERR_DOES_NOT_EXIST) {
+		dev_err(dev, "MAC %pM does not exist for VF %d\n", mac_addr,
+			vf->vf_id);
+		return -ENOENT;
+	} else if (status) {
+		dev_err(dev, "Failed to delete MAC %pM for VF %d, error %d\n",
+			mac_addr, vf->vf_id, status);
+		return -EIO;
+	}
+
+	if (ether_addr_equal(mac_addr, vf->dflt_lan_addr.addr))
+		eth_zero_addr(vf->dflt_lan_addr.addr);
+
+	vf->num_mac--;
+
+	return 0;
+}
+
 /**
  * ice_vc_handle_mac_addr_msg
  * @vf: pointer to the VF info
@@ -2494,23 +2540,23 @@ static bool ice_can_vf_change_mac(struct ice_vf *vf)
 static int
 ice_vc_handle_mac_addr_msg(struct ice_vf *vf, u8 *msg, bool set)
 {
+	int (*ice_vc_cfg_mac)
+		(struct ice_vf *vf, struct ice_vsi *vsi, u8 *mac_addr);
 	enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS;
 	struct virtchnl_ether_addr_list *al =
 	    (struct virtchnl_ether_addr_list *)msg;
 	struct ice_pf *pf = vf->pf;
 	enum virtchnl_ops vc_op;
-	enum ice_status status;
 	struct ice_vsi *vsi;
-	struct device *dev;
-	int mac_count = 0;
 	int i;
 
-	dev = ice_pf_to_dev(pf);
-
-	if (set)
+	if (set) {
 		vc_op = VIRTCHNL_OP_ADD_ETH_ADDR;
-	else
+		ice_vc_cfg_mac = ice_vc_add_mac_addr;
+	} else {
 		vc_op = VIRTCHNL_OP_DEL_ETH_ADDR;
+		ice_vc_cfg_mac = ice_vc_del_mac_addr;
+	}
 
 	if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states) ||
 	    !ice_vc_isvalid_vsi_id(vf, al->vsi_id)) {
@@ -2518,14 +2564,15 @@ ice_vc_handle_mac_addr_msg(struct ice_vf *vf, u8 *msg, bool set)
 		goto handle_mac_exit;
 	}
 
+	/* If this VF is not privileged, then we can't add more than a
+	 * limited number of addresses. Check to make sure that the
+	 * additions do not push us over the limit.
+	 */
 	if (set && !ice_is_vf_trusted(vf) &&
 	    (vf->num_mac + al->num_elements) > ICE_MAX_MACADDR_PER_VF) {
-		dev_err(dev,
+		dev_err(ice_pf_to_dev(pf),
 			"Can't add more MAC addresses, because VF-%d is not trusted, switch the VF to trusted mode in order to add more functionalities\n",
 			vf->vf_id);
-		/* There is no need to let VF know about not being trusted
-		 * to add more MAC addr, so we can just return success message.
-		 */
 		v_ret = VIRTCHNL_STATUS_ERR_PARAM;
 		goto handle_mac_exit;
 	}
@@ -2537,70 +2584,22 @@ ice_vc_handle_mac_addr_msg(struct ice_vf *vf, u8 *msg, bool set)
 	}
 
 	for (i = 0; i < al->num_elements; i++) {
-		u8 *maddr = al->list[i].addr;
+		u8 *mac_addr = al->list[i].addr;
+		int result;
 
-		if (ether_addr_equal(maddr, vf->dflt_lan_addr.addr) ||
-		    is_broadcast_ether_addr(maddr)) {
-			if (set) {
-				/* VF is trying to add filters that the PF
-				 * already added. Just continue.
-				 */
-				dev_info(dev,
-					 "MAC %pM already set for VF %d\n",
-					 maddr, vf->vf_id);
-				continue;
-			} else {
-				/* VF can't remove dflt_lan_addr/bcast MAC */
-				dev_err(dev,
-					"VF can't remove default MAC address or MAC %pM programmed by PF for VF %d\n",
-					maddr, vf->vf_id);
-				continue;
-			}
-		}
-
-		/* check for the invalid cases and bail if necessary */
-		if (is_zero_ether_addr(maddr)) {
-			dev_err(dev,
-				"invalid MAC %pM provided for VF %d\n",
-				maddr, vf->vf_id);
-			v_ret = VIRTCHNL_STATUS_ERR_PARAM;
-			goto handle_mac_exit;
-		}
-
-		if (is_unicast_ether_addr(maddr) &&
-		    !ice_can_vf_change_mac(vf)) {
-			dev_err(dev,
-				"can't change unicast MAC for untrusted VF %d\n",
-				vf->vf_id);
-			v_ret = VIRTCHNL_STATUS_ERR_PARAM;
-			goto handle_mac_exit;
-		}
+		if (is_broadcast_ether_addr(mac_addr) ||
+		    is_zero_ether_addr(mac_addr))
+			continue;
 
-		/* program the updated filter list */
-		status = ice_vsi_cfg_mac_fltr(vsi, maddr, set);
-		if (status == ICE_ERR_DOES_NOT_EXIST ||
-		    status == ICE_ERR_ALREADY_EXISTS) {
-			dev_info(dev,
-				 "can't %s MAC filters %pM for VF %d, error %d\n",
-				 set ? "add" : "remove", maddr, vf->vf_id,
-				 status);
-		} else if (status) {
-			dev_err(dev,
-				"can't %s MAC filters for VF %d, error %d\n",
-				set ? "add" : "remove", vf->vf_id, status);
-			v_ret = ice_err_to_virt_err(status);
+		result = ice_vc_cfg_mac(vf, vsi, mac_addr);
+		if (result == -EEXIST || result == -ENOENT) {
+			continue;
+		} else if (result) {
+			v_ret = VIRTCHNL_STATUS_ERR_ADMIN_QUEUE_ERROR;
 			goto handle_mac_exit;
 		}
-
-		mac_count++;
 	}
 
-	/* Track number of MAC filters programmed for the VF VSI */
-	if (set)
-		vf->num_mac += mac_count;
-	else
-		vf->num_mac -= mac_count;
-
 handle_mac_exit:
 	/* send the response to the VF */
 	return ice_vc_send_msg_to_vf(vf, vc_op, v_ret, NULL, 0);
-- 
cgit v1.2.3-59-g8ed1b


From bda5b7db823c0f49babc260f027d7fa169476423 Mon Sep 17 00:00:00 2001
From: Tony Nguyen <anthony.l.nguyen@intel.com>
Date: Thu, 12 Dec 2019 03:13:03 -0800
Subject: ice: Demote MTU change print to debug

Following the changes of commit 12299132b3d3 ("net: ethernet: intel: Demote
MTU change prints to debug"), change the MTU change message to netdev_dbg()

Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 3adeb7059ac5..2c9350fcbb59 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -4836,7 +4836,7 @@ static int ice_change_mtu(struct net_device *netdev, int new_mtu)
 		}
 	}
 
-	netdev_info(netdev, "changed MTU to %d\n", new_mtu);
+	netdev_dbg(netdev, "changed MTU to %d\n", new_mtu);
 	return 0;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From ac614b13fe71ec561b4a540c9f3ef095da5a7114 Mon Sep 17 00:00:00 2001
From: Bruce Allan <bruce.w.allan@intel.com>
Date: Thu, 12 Dec 2019 03:13:04 -0800
Subject: ice: suppress checked_return error

Coverity reports an error that is not really an error; suppress it.

Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_dcb_lib.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/net/ethernet/intel/ice/ice_dcb_lib.c b/drivers/net/ethernet/intel/ice/ice_dcb_lib.c
index d3d3ec29def9..0664e5b8d130 100644
--- a/drivers/net/ethernet/intel/ice/ice_dcb_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_dcb_lib.c
@@ -396,6 +396,12 @@ dcb_error:
 	prev_cfg->etscfg.tcbwtable[0] = ICE_TC_MAX_BW;
 	prev_cfg->etscfg.tsatable[0] = ICE_IEEE_TSA_ETS;
 	memcpy(&prev_cfg->etsrec, &prev_cfg->etscfg, sizeof(prev_cfg->etsrec));
+	/* Coverity warns the return code of ice_pf_dcb_cfg() is not checked
+	 * here as is done for other calls to that function. That check is
+	 * not necessary since this is in this function's error cleanup path.
+	 * Suppress the Coverity warning with the following comment...
+	 */
+	/* coverity[check_return] */
 	ice_pf_dcb_cfg(pf, prev_cfg, false);
 	kfree(prev_cfg);
 }
-- 
cgit v1.2.3-59-g8ed1b


From 1f45ebe0d8fbe6178670b663005f38ef8535db5d Mon Sep 17 00:00:00 2001
From: Mitch Williams <mitch.a.williams@intel.com>
Date: Thu, 12 Dec 2019 03:13:05 -0800
Subject: ice: add extra check for null Rx descriptor

In the case where the hardware gives us a null Rx descriptor, it is
theoretically possible that we could call one of our skb-construction
functions with no data pointer, which would cause a panic.

In real life, this will never happen - we only get null RX
descriptors as the final descriptor in a chain of otherwise-valid
descriptors. When this happens, the skb will be extant and we'll just
call ice_add_rx_frag(), which can deal with empty data buffers.

Unfortunately, Coverity does not have intimate knowledge of our
hardware, so we must add a check here.

Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_txrx.c | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c
index b77514bbd7ba..fd17ace6b226 100644
--- a/drivers/net/ethernet/intel/ice/ice_txrx.c
+++ b/drivers/net/ethernet/intel/ice/ice_txrx.c
@@ -1071,13 +1071,16 @@ static int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget)
 		ice_put_rx_buf(rx_ring, rx_buf);
 		continue;
 construct_skb:
-		if (skb)
+		if (skb) {
 			ice_add_rx_frag(rx_ring, rx_buf, skb, size);
-		else if (ice_ring_uses_build_skb(rx_ring))
-			skb = ice_build_skb(rx_ring, rx_buf, &xdp);
-		else
+		} else if (likely(xdp.data)) {
+			if (ice_ring_uses_build_skb(rx_ring))
+				skb = ice_build_skb(rx_ring, rx_buf, &xdp);
+			else
+				skb = ice_construct_skb(rx_ring, rx_buf, &xdp);
+		} else {
 			skb = ice_construct_skb(rx_ring, rx_buf, &xdp);
-
+		}
 		/* exit if we failed to retrieve a buffer */
 		if (!skb) {
 			rx_ring->rx_stats.alloc_buf_failed++;
-- 
cgit v1.2.3-59-g8ed1b


From 65bb559b6c11dfe3311d1b77ba5a4b8b34a3ec75 Mon Sep 17 00:00:00 2001
From: Krzysztof Kazimierczak <krzysztof.kazimierczak@intel.com>
Date: Thu, 12 Dec 2019 03:13:06 -0800
Subject: ice: Add a boundary check in ice_xsk_umem()

In ice_xsk_umem(), variable qid which is later used as an array index,
is not validated for a possible boundary exceedance. Because of that,
a calling function might receive an invalid address, which causes
general protection fault when dereferenced.

To address this, add a boundary check to see if qid is greater than the
size of a UMEM array. Also, don't let user change vsi->num_xsk_umems
just by trying to setup a second UMEM if its value is already set up
(i.e. UMEM region has already been allocated for this VSI).

While at it, make sure that ring->zca.free pointer is always zeroed out
if there is no UMEM on a specified ring.

Signed-off-by: Krzysztof Kazimierczak <krzysztof.kazimierczak@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ice/ice.h      | 5 +++--
 drivers/net/ethernet/intel/ice/ice_base.c | 1 +
 drivers/net/ethernet/intel/ice/ice_xsk.c  | 3 ++-
 3 files changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h
index 1376219a6c7e..cb10abb14e11 100644
--- a/drivers/net/ethernet/intel/ice/ice.h
+++ b/drivers/net/ethernet/intel/ice/ice.h
@@ -465,12 +465,13 @@ static inline void ice_set_ring_xdp(struct ice_ring *ring)
 static inline struct xdp_umem *ice_xsk_umem(struct ice_ring *ring)
 {
 	struct xdp_umem **umems = ring->vsi->xsk_umems;
-	int qid = ring->q_index;
+	u16 qid = ring->q_index;
 
 	if (ice_ring_is_xdp(ring))
 		qid -= ring->vsi->num_xdp_txq;
 
-	if (!umems || !umems[qid] || !ice_is_xdp_ena_vsi(ring->vsi))
+	if (qid >= ring->vsi->num_xsk_umems || !umems || !umems[qid] ||
+	    !ice_is_xdp_ena_vsi(ring->vsi))
 		return NULL;
 
 	return umems[qid];
diff --git a/drivers/net/ethernet/intel/ice/ice_base.c b/drivers/net/ethernet/intel/ice/ice_base.c
index d4559d45288f..25f2e0fb3833 100644
--- a/drivers/net/ethernet/intel/ice/ice_base.c
+++ b/drivers/net/ethernet/intel/ice/ice_base.c
@@ -326,6 +326,7 @@ int ice_setup_rx_ctx(struct ice_ring *ring)
 			dev_info(&vsi->back->pdev->dev, "Registered XDP mem model MEM_TYPE_ZERO_COPY on Rx ring %d\n",
 				 ring->q_index);
 		} else {
+			ring->zca.free = NULL;
 			if (!xdp_rxq_info_is_reg(&ring->xdp_rxq))
 				xdp_rxq_info_reg(&ring->xdp_rxq,
 						 ring->netdev,
diff --git a/drivers/net/ethernet/intel/ice/ice_xsk.c b/drivers/net/ethernet/intel/ice/ice_xsk.c
index 0af1f0b951c0..0e401f116d54 100644
--- a/drivers/net/ethernet/intel/ice/ice_xsk.c
+++ b/drivers/net/ethernet/intel/ice/ice_xsk.c
@@ -414,7 +414,8 @@ ice_xsk_umem_enable(struct ice_vsi *vsi, struct xdp_umem *umem, u16 qid)
 	if (vsi->type != ICE_VSI_PF)
 		return -EINVAL;
 
-	vsi->num_xsk_umems = min_t(u16, vsi->num_rxq, vsi->num_txq);
+	if (!vsi->num_xsk_umems)
+		vsi->num_xsk_umems = min_t(u16, vsi->num_rxq, vsi->num_txq);
 	if (qid >= vsi->num_xsk_umems)
 		return -EINVAL;
 
-- 
cgit v1.2.3-59-g8ed1b


From 91125399345a3e28876f707b32d068c308b1e590 Mon Sep 17 00:00:00 2001
From: Krzysztof Kazimierczak <krzysztof.kazimierczak@intel.com>
Date: Thu, 12 Dec 2019 03:13:07 -0800
Subject: ice: Suppress Coverity warnings for xdp_rxq_info_reg

Coverity reports some of the calls to xdp_rxq_info_reg() as potential
issues, because the driver does not check its return value. However,
those calls are wrapped with "if (!xdp_rxq_info_is_reg(&ring->xdp_rxq))"
and this check alone is enough to be sure that the function will never
fail.

All possible states of xdp_rxq_info are:
 - NEW,
 - REGISTERED,
 - UNREGISTERED,
 - UNUSED.

The driver won't mark a queue as UNUSED under no circumstance, so the
return value can be ignored safely.

Add comments for Coverity right above calls to xdp_rxq_info_reg() to
suppress the warnings.

Signed-off-by: Krzysztof Kazimierczak <krzysztof.kazimierczak@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_base.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/ethernet/intel/ice/ice_base.c b/drivers/net/ethernet/intel/ice/ice_base.c
index 25f2e0fb3833..d8e975cceb21 100644
--- a/drivers/net/ethernet/intel/ice/ice_base.c
+++ b/drivers/net/ethernet/intel/ice/ice_base.c
@@ -302,6 +302,7 @@ int ice_setup_rx_ctx(struct ice_ring *ring)
 
 	if (ring->vsi->type == ICE_VSI_PF) {
 		if (!xdp_rxq_info_is_reg(&ring->xdp_rxq))
+			/* coverity[check_return] */
 			xdp_rxq_info_reg(&ring->xdp_rxq, ring->netdev,
 					 ring->q_index);
 
@@ -328,6 +329,7 @@ int ice_setup_rx_ctx(struct ice_ring *ring)
 		} else {
 			ring->zca.free = NULL;
 			if (!xdp_rxq_info_is_reg(&ring->xdp_rxq))
+				/* coverity[check_return] */
 				xdp_rxq_info_reg(&ring->xdp_rxq,
 						 ring->netdev,
 						 ring->q_index);
-- 
cgit v1.2.3-59-g8ed1b


From 5d9e618cbb54f5cbaef7f6f44ceb919978f0e5bc Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Tue, 17 Dec 2019 02:55:43 -0800
Subject: ice: Add device ids for E822 devices

Add support for E822 devices

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_devids.h | 18 ++++++++++++++++++
 drivers/net/ethernet/intel/ice/ice_main.c   |  9 +++++++++
 drivers/net/ethernet/intel/ice/ice_nvm.c    | 12 ++++++++++++
 3 files changed, 39 insertions(+)

diff --git a/drivers/net/ethernet/intel/ice/ice_devids.h b/drivers/net/ethernet/intel/ice/ice_devids.h
index f8d5c661d0ba..ce63017c56c7 100644
--- a/drivers/net/ethernet/intel/ice/ice_devids.h
+++ b/drivers/net/ethernet/intel/ice/ice_devids.h
@@ -11,5 +11,23 @@
 #define ICE_DEV_ID_E810C_QSFP		0x1592
 /* Intel(R) Ethernet Controller E810-C for SFP */
 #define ICE_DEV_ID_E810C_SFP		0x1593
+/* Intel(R) Ethernet Connection E822-C for backplane */
+#define ICE_DEV_ID_E822C_BACKPLANE	0x1890
+/* Intel(R) Ethernet Connection E822-C for QSFP */
+#define ICE_DEV_ID_E822C_QSFP		0x1891
+/* Intel(R) Ethernet Connection E822-C for SFP */
+#define ICE_DEV_ID_E822C_SFP		0x1892
+/* Intel(R) Ethernet Connection E822-C/X557-AT 10GBASE-T */
+#define ICE_DEV_ID_E822C_10G_BASE_T	0x1893
+/* Intel(R) Ethernet Connection E822-C 1GbE */
+#define ICE_DEV_ID_E822C_SGMII		0x1894
+/* Intel(R) Ethernet Connection E822-X for backplane */
+#define ICE_DEV_ID_E822X_BACKPLANE	0x1897
+/* Intel(R) Ethernet Connection E822-L for SFP */
+#define ICE_DEV_ID_E822L_SFP		0x1898
+/* Intel(R) Ethernet Connection E822-L/X557-AT 10GBASE-T */
+#define ICE_DEV_ID_E822L_10G_BASE_T	0x1899
+/* Intel(R) Ethernet Connection E822-L 1GbE */
+#define ICE_DEV_ID_E822L_SGMII		0x189A
 
 #endif /* _ICE_DEVIDS_H_ */
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 2c9350fcbb59..bf539483e25e 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -3572,6 +3572,15 @@ static const struct pci_device_id ice_pci_tbl[] = {
 	{ PCI_VDEVICE(INTEL, ICE_DEV_ID_E810C_BACKPLANE), 0 },
 	{ PCI_VDEVICE(INTEL, ICE_DEV_ID_E810C_QSFP), 0 },
 	{ PCI_VDEVICE(INTEL, ICE_DEV_ID_E810C_SFP), 0 },
+	{ PCI_VDEVICE(INTEL, ICE_DEV_ID_E822C_BACKPLANE), 0 },
+	{ PCI_VDEVICE(INTEL, ICE_DEV_ID_E822C_QSFP), 0 },
+	{ PCI_VDEVICE(INTEL, ICE_DEV_ID_E822C_SFP), 0 },
+	{ PCI_VDEVICE(INTEL, ICE_DEV_ID_E822C_10G_BASE_T), 0 },
+	{ PCI_VDEVICE(INTEL, ICE_DEV_ID_E822C_SGMII), 0 },
+	{ PCI_VDEVICE(INTEL, ICE_DEV_ID_E822X_BACKPLANE), 0 },
+	{ PCI_VDEVICE(INTEL, ICE_DEV_ID_E822L_SFP), 0 },
+	{ PCI_VDEVICE(INTEL, ICE_DEV_ID_E822L_10G_BASE_T), 0 },
+	{ PCI_VDEVICE(INTEL, ICE_DEV_ID_E822L_SGMII), 0 },
 	/* required last entry */
 	{ 0, }
 };
diff --git a/drivers/net/ethernet/intel/ice/ice_nvm.c b/drivers/net/ethernet/intel/ice/ice_nvm.c
index 57c73f613f32..7525ac50742e 100644
--- a/drivers/net/ethernet/intel/ice/ice_nvm.c
+++ b/drivers/net/ethernet/intel/ice/ice_nvm.c
@@ -289,6 +289,18 @@ enum ice_status ice_init_nvm(struct ice_hw *hw)
 
 	nvm->eetrack = (eetrack_hi << 16) | eetrack_lo;
 
+	/* the following devices do not have boot_cfg_tlv yet */
+	if (hw->device_id == ICE_DEV_ID_E822C_BACKPLANE ||
+	    hw->device_id == ICE_DEV_ID_E822C_QSFP ||
+	    hw->device_id == ICE_DEV_ID_E822C_10G_BASE_T ||
+	    hw->device_id == ICE_DEV_ID_E822C_SGMII ||
+	    hw->device_id == ICE_DEV_ID_E822C_SFP ||
+	    hw->device_id == ICE_DEV_ID_E822X_BACKPLANE ||
+	    hw->device_id == ICE_DEV_ID_E822L_SFP ||
+	    hw->device_id == ICE_DEV_ID_E822L_10G_BASE_T ||
+	    hw->device_id == ICE_DEV_ID_E822L_SGMII)
+		return status;
+
 	status = ice_get_pfa_module_tlv(hw, &boot_cfg_tlv, &boot_cfg_tlv_len,
 					ICE_SR_BOOT_CFG_PTR);
 	if (status) {
-- 
cgit v1.2.3-59-g8ed1b


From 36278a5d4d354e5d5610aa728831db9e03cc3d8d Mon Sep 17 00:00:00 2001
From: Alain Michaud <alainm@chromium.org>
Date: Wed, 11 Dec 2019 01:54:43 +0000
Subject: Bluetooth: Adding a bt_dev_warn_ratelimited macro.

The macro will be used to display rate limited warning messages in the
log.

Signed-off-by: Alain Michaud <alainm@chromium.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 include/net/bluetooth/bluetooth.h |  4 ++++
 net/bluetooth/lib.c               | 16 ++++++++++++++++
 2 files changed, 20 insertions(+)

diff --git a/include/net/bluetooth/bluetooth.h b/include/net/bluetooth/bluetooth.h
index fabee6db0abb..bd2675266859 100644
--- a/include/net/bluetooth/bluetooth.h
+++ b/include/net/bluetooth/bluetooth.h
@@ -129,6 +129,8 @@ void bt_warn(const char *fmt, ...);
 __printf(1, 2)
 void bt_err(const char *fmt, ...);
 __printf(1, 2)
+void bt_warn_ratelimited(const char *fmt, ...);
+__printf(1, 2)
 void bt_err_ratelimited(const char *fmt, ...);
 
 #define BT_INFO(fmt, ...)	bt_info(fmt "\n", ##__VA_ARGS__)
@@ -147,6 +149,8 @@ void bt_err_ratelimited(const char *fmt, ...);
 #define bt_dev_dbg(hdev, fmt, ...)				\
 	BT_DBG("%s: " fmt, (hdev)->name, ##__VA_ARGS__)
 
+#define bt_dev_warn_ratelimited(hdev, fmt, ...)			\
+	bt_warn_ratelimited("%s: " fmt, (hdev)->name, ##__VA_ARGS__)
 #define bt_dev_err_ratelimited(hdev, fmt, ...)			\
 	BT_ERR_RATELIMITED("%s: " fmt, (hdev)->name, ##__VA_ARGS__)
 
diff --git a/net/bluetooth/lib.c b/net/bluetooth/lib.c
index 63e65d9b4b24..c09e0a3a0ed9 100644
--- a/net/bluetooth/lib.c
+++ b/net/bluetooth/lib.c
@@ -183,6 +183,22 @@ void bt_err(const char *format, ...)
 }
 EXPORT_SYMBOL(bt_err);
 
+void bt_warn_ratelimited(const char *format, ...)
+{
+	struct va_format vaf;
+	va_list args;
+
+	va_start(args, format);
+
+	vaf.fmt = format;
+	vaf.va = &args;
+
+	pr_warn_ratelimited("%pV", &vaf);
+
+	va_end(args);
+}
+EXPORT_SYMBOL(bt_warn_ratelimited);
+
 void bt_err_ratelimited(const char *format, ...)
 {
 	struct va_format vaf;
-- 
cgit v1.2.3-59-g8ed1b


From 657cc646475b721f5c5bab82e7fd43302c7c8358 Mon Sep 17 00:00:00 2001
From: Marcel Holtmann <marcel@holtmann.org>
Date: Wed, 11 Dec 2019 11:34:36 +0100
Subject: Bluetooth: Remove usage of BT_ERR_RATELIMITED macro

The macro is really not needed and can be replaced with either usage of
bt_err_ratelimited or bt_dev_err_ratelimited.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
---
 include/net/bluetooth/bluetooth.h |  4 +---
 net/bluetooth/hci_event.c         | 14 ++++++--------
 2 files changed, 7 insertions(+), 11 deletions(-)

diff --git a/include/net/bluetooth/bluetooth.h b/include/net/bluetooth/bluetooth.h
index bd2675266859..e42bb8e03c09 100644
--- a/include/net/bluetooth/bluetooth.h
+++ b/include/net/bluetooth/bluetooth.h
@@ -138,8 +138,6 @@ void bt_err_ratelimited(const char *fmt, ...);
 #define BT_ERR(fmt, ...)	bt_err(fmt "\n", ##__VA_ARGS__)
 #define BT_DBG(fmt, ...)	pr_debug(fmt "\n", ##__VA_ARGS__)
 
-#define BT_ERR_RATELIMITED(fmt, ...) bt_err_ratelimited(fmt "\n", ##__VA_ARGS__)
-
 #define bt_dev_info(hdev, fmt, ...)				\
 	BT_INFO("%s: " fmt, (hdev)->name, ##__VA_ARGS__)
 #define bt_dev_warn(hdev, fmt, ...)				\
@@ -152,7 +150,7 @@ void bt_err_ratelimited(const char *fmt, ...);
 #define bt_dev_warn_ratelimited(hdev, fmt, ...)			\
 	bt_warn_ratelimited("%s: " fmt, (hdev)->name, ##__VA_ARGS__)
 #define bt_dev_err_ratelimited(hdev, fmt, ...)			\
-	BT_ERR_RATELIMITED("%s: " fmt, (hdev)->name, ##__VA_ARGS__)
+	bt_err_ratelimited("%s: " fmt, (hdev)->name, ##__VA_ARGS__)
 
 /* Connection and socket states */
 enum {
diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c
index c1d3a303d97f..1941f120a376 100644
--- a/net/bluetooth/hci_event.c
+++ b/net/bluetooth/hci_event.c
@@ -5451,7 +5451,7 @@ static void hci_le_adv_report_evt(struct hci_dev *hdev, struct sk_buff *skb)
 	hci_dev_unlock(hdev);
 }
 
-static u8 ext_evt_type_to_legacy(u16 evt_type)
+static u8 ext_evt_type_to_legacy(struct hci_dev *hdev, u16 evt_type)
 {
 	if (evt_type & LE_EXT_ADV_LEGACY_PDU) {
 		switch (evt_type) {
@@ -5468,10 +5468,7 @@ static u8 ext_evt_type_to_legacy(u16 evt_type)
 			return LE_ADV_SCAN_RSP;
 		}
 
-		BT_ERR_RATELIMITED("Unknown advertising packet type: 0x%02x",
-				   evt_type);
-
-		return LE_ADV_INVALID;
+		goto invalid;
 	}
 
 	if (evt_type & LE_EXT_ADV_CONN_IND) {
@@ -5491,8 +5488,9 @@ static u8 ext_evt_type_to_legacy(u16 evt_type)
 	    evt_type & LE_EXT_ADV_DIRECT_IND)
 		return LE_ADV_NONCONN_IND;
 
-	BT_ERR_RATELIMITED("Unknown advertising packet type: 0x%02x",
-				   evt_type);
+invalid:
+	bt_dev_err_ratelimited(hdev, "Unknown advertising packet type: 0x%02x",
+			       evt_type);
 
 	return LE_ADV_INVALID;
 }
@@ -5510,7 +5508,7 @@ static void hci_le_ext_adv_report_evt(struct hci_dev *hdev, struct sk_buff *skb)
 		u16 evt_type;
 
 		evt_type = __le16_to_cpu(ev->evt_type);
-		legacy_evt_type = ext_evt_type_to_legacy(evt_type);
+		legacy_evt_type = ext_evt_type_to_legacy(hdev, evt_type);
 		if (legacy_evt_type != LE_ADV_INVALID) {
 			process_adv_report(hdev, legacy_evt_type, &ev->bdaddr,
 					   ev->bdaddr_type, NULL, 0, ev->rssi,
-- 
cgit v1.2.3-59-g8ed1b


From 88d1cc96c4660f59d32a45810f49943e84c4bd54 Mon Sep 17 00:00:00 2001
From: Dmitry Osipenko <digetx@gmail.com>
Date: Sun, 15 Dec 2019 21:52:52 +0300
Subject: Bluetooth: hci_bcm: Add device-tree compatible for BCM4329

Driver supports BCM4329, but there is no device-tree compatible for
that chip. Let's add it in order to allow boards to specify Bluetooth
in theirs device-trees, in particular this is useful for NVIDIA Tegra20
boards.

Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 drivers/bluetooth/hci_bcm.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/bluetooth/hci_bcm.c b/drivers/bluetooth/hci_bcm.c
index f8f5c593a05c..bbfaf0c1529d 100644
--- a/drivers/bluetooth/hci_bcm.c
+++ b/drivers/bluetooth/hci_bcm.c
@@ -1467,6 +1467,7 @@ static struct bcm_device_data bcm4354_device_data = {
 
 static const struct of_device_id bcm_bluetooth_of_match[] = {
 	{ .compatible = "brcm,bcm20702a1" },
+	{ .compatible = "brcm,bcm4329-bt" },
 	{ .compatible = "brcm,bcm4345c5" },
 	{ .compatible = "brcm,bcm4330-bt" },
 	{ .compatible = "brcm,bcm43438-bt" },
-- 
cgit v1.2.3-59-g8ed1b


From 8b17fb9628dd91cf48b8d40fba4b4f7499bf2ee7 Mon Sep 17 00:00:00 2001
From: Dmitry Osipenko <digetx@gmail.com>
Date: Sun, 15 Dec 2019 21:52:53 +0300
Subject: dt-bindings: net: broadcom-bluetooth: Document BCM4329 support

The BCM4329 is a 802.11 a/b/g/n WiFi + Bluetooth 2.1 chip which is found
in Azurewave AW-NH611 WiFi+BT module.

Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 Documentation/devicetree/bindings/net/broadcom-bluetooth.txt | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/devicetree/bindings/net/broadcom-bluetooth.txt b/Documentation/devicetree/bindings/net/broadcom-bluetooth.txt
index b5eadee4a9a7..c44a30dbe43d 100644
--- a/Documentation/devicetree/bindings/net/broadcom-bluetooth.txt
+++ b/Documentation/devicetree/bindings/net/broadcom-bluetooth.txt
@@ -11,6 +11,7 @@ Required properties:
 
  - compatible: should contain one of the following:
    * "brcm,bcm20702a1"
+   * "brcm,bcm4329-bt"
    * "brcm,bcm4330-bt"
    * "brcm,bcm43438-bt"
    * "brcm,bcm4345c5"
-- 
cgit v1.2.3-59-g8ed1b


From 9f3565b89c43d2a0082bccbc3918f2ebae739533 Mon Sep 17 00:00:00 2001
From: Rocky Liao <rjliao@codeaurora.org>
Date: Fri, 13 Dec 2019 16:50:45 +0800
Subject: Bluetooth: hci_qca: Replace of_device_get_match_data with
 device_get_match_data

Replace of_device_get_match_data with device_get_match_data to make driver
work across platforms.

Signed-off-by: Rocky Liao <rjliao@codeaurora.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 drivers/bluetooth/hci_qca.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c
index f10bdf8e1fc5..b602ed01505b 100644
--- a/drivers/bluetooth/hci_qca.c
+++ b/drivers/bluetooth/hci_qca.c
@@ -1501,7 +1501,7 @@ static int qca_serdev_probe(struct serdev_device *serdev)
 		return -ENOMEM;
 
 	qcadev->serdev_hu.serdev = serdev;
-	data = of_device_get_match_data(&serdev->dev);
+	data = device_get_match_data(&serdev->dev);
 	serdev_device_set_drvdata(serdev, qcadev);
 	device_property_read_string(&serdev->dev, "firmware-name",
 					 &qcadev->firmware_name);
-- 
cgit v1.2.3-59-g8ed1b


From 5bd64c645fc660734f172be4bba449952ed84ce2 Mon Sep 17 00:00:00 2001
From: Abhishek Pandit-Subedi <abhishekpandit@chromium.org>
Date: Thu, 26 Dec 2019 11:03:40 -0800
Subject: Bluetooth: btbcm: Add missing static inline in header

This fixes a double definition error when CONFIG_BT_BCM is not set.

Fixes: 528379902337 ("Bluetooth: btbcm: Support pcm configuration")
Signed-off-by: Abhishek Pandit-Subedi <abhishekpandit@chromium.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 drivers/bluetooth/btbcm.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/bluetooth/btbcm.h b/drivers/bluetooth/btbcm.h
index 3c7dd0765837..014ef847a486 100644
--- a/drivers/bluetooth/btbcm.h
+++ b/drivers/bluetooth/btbcm.h
@@ -78,13 +78,13 @@ static inline int btbcm_set_bdaddr(struct hci_dev *hdev, const bdaddr_t *bdaddr)
 	return -EOPNOTSUPP;
 }
 
-int btbcm_read_pcm_int_params(struct hci_dev *hdev,
+static inline int btbcm_read_pcm_int_params(struct hci_dev *hdev,
 			      struct bcm_set_pcm_int_params *params)
 {
 	return -EOPNOTSUPP;
 }
 
-int btbcm_write_pcm_int_params(struct hci_dev *hdev,
+static inline int btbcm_write_pcm_int_params(struct hci_dev *hdev,
 			       const struct bcm_set_pcm_int_params *params)
 {
 	return -EOPNOTSUPP;
-- 
cgit v1.2.3-59-g8ed1b


From e601daed271e9eb1b923972a0a1af65f8c7bb77b Mon Sep 17 00:00:00 2001
From: Stefan Wahren <wahrenst@gmx.net>
Date: Wed, 1 Jan 2020 15:01:34 +0100
Subject: Bluetooth: hci_bcm: Drive RTS only for BCM43438
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The commit 3347a80965b3 ("Bluetooth: hci_bcm: Fix RTS handling during
startup") is causing at least a regression for AP6256 on Orange Pi 3.
So do the RTS line handing during startup only on the necessary platform.

Fixes: 3347a80965b3 ("Bluetooth: hci_bcm: Fix RTS handling during startup")
Reported-by: Ondřej Jirman <megous@megous.com>
Signed-off-by: Stefan Wahren <wahrenst@gmx.net>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 drivers/bluetooth/hci_bcm.c | 21 +++++++++++++++++----
 1 file changed, 17 insertions(+), 4 deletions(-)

diff --git a/drivers/bluetooth/hci_bcm.c b/drivers/bluetooth/hci_bcm.c
index bbfaf0c1529d..769bb4404bd1 100644
--- a/drivers/bluetooth/hci_bcm.c
+++ b/drivers/bluetooth/hci_bcm.c
@@ -53,6 +53,7 @@
  */
 struct bcm_device_data {
 	bool	no_early_set_baudrate;
+	bool	drive_rts_on_open;
 };
 
 /**
@@ -122,6 +123,7 @@ struct bcm_device {
 	bool			is_suspended;
 #endif
 	bool			no_early_set_baudrate;
+	bool			drive_rts_on_open;
 	u8			pcm_int_params[5];
 };
 
@@ -456,7 +458,9 @@ static int bcm_open(struct hci_uart *hu)
 
 out:
 	if (bcm->dev) {
-		hci_uart_set_flow_control(hu, true);
+		if (bcm->dev->drive_rts_on_open)
+			hci_uart_set_flow_control(hu, true);
+
 		hu->init_speed = bcm->dev->init_speed;
 
 		/* If oper_speed is set, ldisc/serdev will set the baudrate
@@ -466,7 +470,10 @@ out:
 			hu->oper_speed = bcm->dev->oper_speed;
 
 		err = bcm_gpio_set_power(bcm->dev, true);
-		hci_uart_set_flow_control(hu, false);
+
+		if (bcm->dev->drive_rts_on_open)
+			hci_uart_set_flow_control(hu, false);
+
 		if (err)
 			goto err_unset_hu;
 	}
@@ -1447,8 +1454,10 @@ static int bcm_serdev_probe(struct serdev_device *serdev)
 		dev_err(&serdev->dev, "Failed to power down\n");
 
 	data = device_get_match_data(bcmdev->dev);
-	if (data)
+	if (data) {
 		bcmdev->no_early_set_baudrate = data->no_early_set_baudrate;
+		bcmdev->drive_rts_on_open = data->drive_rts_on_open;
+	}
 
 	return hci_uart_register_device(&bcmdev->serdev_hu, &bcm_proto);
 }
@@ -1465,12 +1474,16 @@ static struct bcm_device_data bcm4354_device_data = {
 	.no_early_set_baudrate = true,
 };
 
+static struct bcm_device_data bcm43438_device_data = {
+	.drive_rts_on_open = true,
+};
+
 static const struct of_device_id bcm_bluetooth_of_match[] = {
 	{ .compatible = "brcm,bcm20702a1" },
 	{ .compatible = "brcm,bcm4329-bt" },
 	{ .compatible = "brcm,bcm4345c5" },
 	{ .compatible = "brcm,bcm4330-bt" },
-	{ .compatible = "brcm,bcm43438-bt" },
+	{ .compatible = "brcm,bcm43438-bt", .data = &bcm43438_device_data },
 	{ .compatible = "brcm,bcm43540-bt", .data = &bcm4354_device_data },
 	{ .compatible = "brcm,bcm4335a0" },
 	{ },
-- 
cgit v1.2.3-59-g8ed1b


From d841502c79e3fda2ba0e8d64f9eb00e9dd884af0 Mon Sep 17 00:00:00 2001
From: Balakrishna Godavarthi <bgodavar@codeaurora.org>
Date: Thu, 2 Jan 2020 20:19:11 +0530
Subject: Bluetooth: hci_qca: Collect controller memory dump during SSR

We will collect the ramdump of BT controller when hardware error event
received before rebooting the HCI layer. Before restarting a subsystem
or a process running on a subsystem, it is often required to request
either a subsystem or a process to perform proper cache dump and
software failure reason into a memory buffer which application
processor can retrieve afterwards. SW developers can often provide
initial investigation by looking into that debugging information.

Signed-off-by: Balakrishna Godavarthi <bgodavar@codeaurora.org>
Signed-off-by: Venkata Lakshmi Narayana Gubba <gubbaven@codeaurora.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 drivers/bluetooth/hci_qca.c | 296 +++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 291 insertions(+), 5 deletions(-)

diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c
index b602ed01505b..9392cc7f9908 100644
--- a/drivers/bluetooth/hci_qca.c
+++ b/drivers/bluetooth/hci_qca.c
@@ -20,6 +20,7 @@
 #include <linux/completion.h>
 #include <linux/debugfs.h>
 #include <linux/delay.h>
+#include <linux/devcoredump.h>
 #include <linux/device.h>
 #include <linux/gpio/consumer.h>
 #include <linux/mod_devicetable.h>
@@ -46,6 +47,7 @@
 #define IBS_BTSOC_TX_IDLE_TIMEOUT_MS	40
 #define IBS_HOST_TX_IDLE_TIMEOUT_MS	2000
 #define CMD_TRANS_TIMEOUT_MS		100
+#define MEMDUMP_TIMEOUT_MS		8000
 
 /* susclk rate */
 #define SUSCLK_RATE_32KHZ	32768
@@ -53,12 +55,21 @@
 /* Controller debug log header */
 #define QCA_DEBUG_HANDLE	0x2EDC
 
+/* Controller dump header */
+#define QCA_SSR_DUMP_HANDLE		0x0108
+#define QCA_DUMP_PACKET_SIZE		255
+#define QCA_LAST_SEQUENCE_NUM		0xFFFF
+#define QCA_CRASHBYTE_PACKET_LEN	1096
+#define QCA_MEMDUMP_BYTE		0xFB
+
 enum qca_flags {
 	QCA_IBS_ENABLED,
 	QCA_DROP_VENDOR_EVENT,
 	QCA_SUSPENDING,
+	QCA_MEMDUMP_COLLECTION
 };
 
+
 /* HCI_IBS transmit side sleep protocol states */
 enum tx_ibs_states {
 	HCI_IBS_TX_ASLEEP,
@@ -81,11 +92,40 @@ enum hci_ibs_clock_state_vote {
 	HCI_IBS_RX_VOTE_CLOCK_OFF,
 };
 
+/* Controller memory dump states */
+enum qca_memdump_states {
+	QCA_MEMDUMP_IDLE,
+	QCA_MEMDUMP_COLLECTING,
+	QCA_MEMDUMP_COLLECTED,
+	QCA_MEMDUMP_TIMEOUT,
+};
+
+struct qca_memdump_data {
+	char *memdump_buf_head;
+	char *memdump_buf_tail;
+	u32 current_seq_no;
+	u32 received_dump;
+};
+
+struct qca_memdump_event_hdr {
+	__u8    evt;
+	__u8    plen;
+	__u16   opcode;
+	__u16   seq_no;
+	__u8    reserved;
+} __packed;
+
+
+struct qca_dump_size {
+	u32 dump_size;
+} __packed;
+
 struct qca_data {
 	struct hci_uart *hu;
 	struct sk_buff *rx_skb;
 	struct sk_buff_head txq;
 	struct sk_buff_head tx_wait_q;	/* HCI_IBS wait queue	*/
+	struct sk_buff_head rx_memdump_q;	/* Memdump wait queue	*/
 	spinlock_t hci_ibs_lock;	/* HCI_IBS state lock	*/
 	u8 tx_ibs_state;	/* HCI_IBS transmit side power state*/
 	u8 rx_ibs_state;	/* HCI_IBS receive side power state */
@@ -95,14 +135,18 @@ struct qca_data {
 	u32 tx_idle_delay;
 	struct timer_list wake_retrans_timer;
 	u32 wake_retrans;
+	struct timer_list memdump_timer;
 	struct workqueue_struct *workqueue;
 	struct work_struct ws_awake_rx;
 	struct work_struct ws_awake_device;
 	struct work_struct ws_rx_vote_off;
 	struct work_struct ws_tx_vote_off;
+	struct work_struct ctrl_memdump_evt;
+	struct qca_memdump_data *qca_memdump;
 	unsigned long flags;
 	struct completion drop_ev_comp;
 	wait_queue_head_t suspend_wait_q;
+	enum qca_memdump_states memdump_state;
 
 	/* For debugging purpose */
 	u64 ibs_sent_wacks;
@@ -167,6 +211,7 @@ static int qca_regulator_enable(struct qca_serdev *qcadev);
 static void qca_regulator_disable(struct qca_serdev *qcadev);
 static void qca_power_shutdown(struct hci_uart *hu);
 static int qca_power_off(struct hci_dev *hdev);
+static void qca_controller_memdump(struct work_struct *work);
 
 static enum qca_btsoc_type qca_soc_type(struct hci_uart *hu)
 {
@@ -474,6 +519,23 @@ static void hci_ibs_wake_retrans_timeout(struct timer_list *t)
 		hci_uart_tx_wakeup(hu);
 }
 
+static void hci_memdump_timeout(struct timer_list *t)
+{
+	struct qca_data *qca = from_timer(qca, t, tx_idle_timer);
+	struct hci_uart *hu = qca->hu;
+	struct qca_memdump_data *qca_memdump = qca->qca_memdump;
+	char *memdump_buf = qca_memdump->memdump_buf_tail;
+
+	bt_dev_err(hu->hdev, "clearing allocated memory due to memdump timeout");
+	/* Inject hw error event to reset the device and driver. */
+	hci_reset_dev(hu->hdev);
+	kfree(memdump_buf);
+	kfree(qca_memdump);
+	qca->memdump_state = QCA_MEMDUMP_TIMEOUT;
+	del_timer(&qca->memdump_timer);
+	cancel_work_sync(&qca->ctrl_memdump_evt);
+}
+
 /* Initialize protocol */
 static int qca_open(struct hci_uart *hu)
 {
@@ -492,6 +554,7 @@ static int qca_open(struct hci_uart *hu)
 
 	skb_queue_head_init(&qca->txq);
 	skb_queue_head_init(&qca->tx_wait_q);
+	skb_queue_head_init(&qca->rx_memdump_q);
 	spin_lock_init(&qca->hci_ibs_lock);
 	qca->workqueue = alloc_ordered_workqueue("qca_wq", 0);
 	if (!qca->workqueue) {
@@ -504,7 +567,7 @@ static int qca_open(struct hci_uart *hu)
 	INIT_WORK(&qca->ws_awake_device, qca_wq_awake_device);
 	INIT_WORK(&qca->ws_rx_vote_off, qca_wq_serial_rx_clock_vote_off);
 	INIT_WORK(&qca->ws_tx_vote_off, qca_wq_serial_tx_clock_vote_off);
-
+	INIT_WORK(&qca->ctrl_memdump_evt, qca_controller_memdump);
 	init_waitqueue_head(&qca->suspend_wait_q);
 
 	qca->hu = hu;
@@ -544,6 +607,7 @@ static int qca_open(struct hci_uart *hu)
 
 	timer_setup(&qca->tx_idle_timer, hci_ibs_tx_idle_timeout, 0);
 	qca->tx_idle_delay = IBS_HOST_TX_IDLE_TIMEOUT_MS;
+	timer_setup(&qca->memdump_timer, hci_memdump_timeout, 0);
 
 	BT_DBG("HCI_UART_QCA open, tx_idle_delay=%u, wake_retrans=%u",
 	       qca->tx_idle_delay, qca->wake_retrans);
@@ -622,8 +686,10 @@ static int qca_close(struct hci_uart *hu)
 
 	skb_queue_purge(&qca->tx_wait_q);
 	skb_queue_purge(&qca->txq);
+	skb_queue_purge(&qca->rx_memdump_q);
 	del_timer(&qca->tx_idle_timer);
 	del_timer(&qca->wake_retrans_timer);
+	del_timer(&qca->memdump_timer);
 	destroy_workqueue(qca->workqueue);
 	qca->hu = NULL;
 
@@ -900,6 +966,126 @@ static int qca_recv_acl_data(struct hci_dev *hdev, struct sk_buff *skb)
 	return hci_recv_frame(hdev, skb);
 }
 
+static void qca_controller_memdump(struct work_struct *work)
+{
+	struct qca_data *qca = container_of(work, struct qca_data,
+					    ctrl_memdump_evt);
+	struct hci_uart *hu = qca->hu;
+	struct sk_buff *skb;
+	struct qca_memdump_event_hdr *cmd_hdr;
+	struct qca_memdump_data *qca_memdump = qca->qca_memdump;
+	struct qca_dump_size *dump;
+	char *memdump_buf;
+	char nullBuff[QCA_DUMP_PACKET_SIZE] = { 0 };
+	u16 opcode, seq_no;
+	u32 dump_size;
+
+	while ((skb = skb_dequeue(&qca->rx_memdump_q))) {
+
+		if (!qca_memdump) {
+			qca_memdump = kzalloc(sizeof(struct qca_memdump_data),
+					      GFP_ATOMIC);
+			if (!qca_memdump)
+				return;
+
+			qca->qca_memdump = qca_memdump;
+		}
+
+		qca->memdump_state = QCA_MEMDUMP_COLLECTING;
+		cmd_hdr = (void *) skb->data;
+		opcode = __le16_to_cpu(cmd_hdr->opcode);
+		seq_no = __le16_to_cpu(cmd_hdr->seq_no);
+		skb_pull(skb, sizeof(struct qca_memdump_event_hdr));
+
+		if (!seq_no) {
+
+			/* This is the first frame of memdump packet from
+			 * the controller, Disable IBS to recevie dump
+			 * with out any interruption, ideally time required for
+			 * the controller to send the dump is 8 seconds. let us
+			 * start timer to handle this asynchronous activity.
+			 */
+			clear_bit(QCA_IBS_ENABLED, &qca->flags);
+			set_bit(QCA_MEMDUMP_COLLECTION, &qca->flags);
+			dump = (void *) skb->data;
+			dump_size = __le32_to_cpu(dump->dump_size);
+			if (!(dump_size)) {
+				bt_dev_err(hu->hdev, "Rx invalid memdump size");
+				kfree_skb(skb);
+				return;
+			}
+
+			bt_dev_info(hu->hdev, "QCA collecting dump of size:%u",
+				    dump_size);
+			mod_timer(&qca->memdump_timer, (jiffies +
+				  msecs_to_jiffies(MEMDUMP_TIMEOUT_MS)));
+
+			skb_pull(skb, sizeof(dump_size));
+			memdump_buf = vmalloc(dump_size);
+			qca_memdump->memdump_buf_head = memdump_buf;
+			qca_memdump->memdump_buf_tail = memdump_buf;
+		}
+
+		memdump_buf = qca_memdump->memdump_buf_tail;
+
+		/* If sequence no 0 is missed then there is no point in
+		 * accepting the other sequences.
+		 */
+		if (!memdump_buf) {
+			bt_dev_err(hu->hdev, "QCA: Discarding other packets");
+			kfree(qca_memdump);
+			kfree_skb(skb);
+			qca->qca_memdump = NULL;
+			return;
+		}
+
+		/* There could be chance of missing some packets from
+		 * the controller. In such cases let us store the dummy
+		 * packets in the buffer.
+		 */
+		while ((seq_no > qca_memdump->current_seq_no + 1) &&
+			seq_no != QCA_LAST_SEQUENCE_NUM) {
+			bt_dev_err(hu->hdev, "QCA controller missed packet:%d",
+				   qca_memdump->current_seq_no);
+			memcpy(memdump_buf, nullBuff, QCA_DUMP_PACKET_SIZE);
+			memdump_buf = memdump_buf + QCA_DUMP_PACKET_SIZE;
+			qca_memdump->received_dump += QCA_DUMP_PACKET_SIZE;
+			qca_memdump->current_seq_no++;
+		}
+
+		memcpy(memdump_buf, (unsigned char *) skb->data, skb->len);
+		memdump_buf = memdump_buf + skb->len;
+		qca_memdump->memdump_buf_tail = memdump_buf;
+		qca_memdump->current_seq_no = seq_no + 1;
+		qca_memdump->received_dump += skb->len;
+		qca->qca_memdump = qca_memdump;
+		kfree_skb(skb);
+		if (seq_no == QCA_LAST_SEQUENCE_NUM) {
+			bt_dev_info(hu->hdev, "QCA writing crash dump of size %d bytes",
+				   qca_memdump->received_dump);
+			memdump_buf = qca_memdump->memdump_buf_head;
+			dev_coredumpv(&hu->serdev->dev, memdump_buf,
+				      qca_memdump->received_dump, GFP_KERNEL);
+			del_timer(&qca->memdump_timer);
+			kfree(qca->qca_memdump);
+			qca->qca_memdump = NULL;
+			qca->memdump_state = QCA_MEMDUMP_COLLECTED;
+		}
+	}
+
+}
+
+int qca_controller_memdump_event(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	struct hci_uart *hu = hci_get_drvdata(hdev);
+	struct qca_data *qca = hu->priv;
+
+	skb_queue_tail(&qca->rx_memdump_q, skb);
+	queue_work(qca->workqueue, &qca->ctrl_memdump_evt);
+
+	return 0;
+}
+
 static int qca_recv_event(struct hci_dev *hdev, struct sk_buff *skb)
 {
 	struct hci_uart *hu = hci_get_drvdata(hdev);
@@ -925,6 +1111,14 @@ static int qca_recv_event(struct hci_dev *hdev, struct sk_buff *skb)
 
 		return 0;
 	}
+	/* We receive chip memory dump as an event packet, With a dedicated
+	 * handler followed by a hardware error event. When this event is
+	 * received we store dump into a file before closing hci. This
+	 * dump will help in triaging the issues.
+	 */
+	if ((skb->data[0] == HCI_VENDOR_PKT) &&
+	    (get_unaligned_be16(skb->data + 2) == QCA_SSR_DUMP_HANDLE))
+		return qca_controller_memdump_event(hdev, skb);
 
 	return hci_recv_frame(hdev, skb);
 }
@@ -1203,6 +1397,91 @@ error:
 	return ret;
 }
 
+static int qca_send_crashbuffer(struct hci_uart *hu)
+{
+	struct qca_data *qca = hu->priv;
+	struct sk_buff *skb;
+
+	skb = bt_skb_alloc(QCA_CRASHBYTE_PACKET_LEN, GFP_KERNEL);
+	if (!skb) {
+		bt_dev_err(hu->hdev, "Failed to allocate memory for skb packet");
+		return -ENOMEM;
+	}
+
+	/* We forcefully crash the controller, by sending 0xfb byte for
+	 * 1024 times. We also might have chance of losing data, To be
+	 * on safer side we send 1096 bytes to the SoC.
+	 */
+	memset(skb_put(skb, QCA_CRASHBYTE_PACKET_LEN), QCA_MEMDUMP_BYTE,
+	       QCA_CRASHBYTE_PACKET_LEN);
+	hci_skb_pkt_type(skb) = HCI_COMMAND_PKT;
+	bt_dev_info(hu->hdev, "crash the soc to collect controller dump");
+	skb_queue_tail(&qca->txq, skb);
+	hci_uart_tx_wakeup(hu);
+
+	return 0;
+}
+
+static void qca_wait_for_dump_collection(struct hci_dev *hdev)
+{
+	struct hci_uart *hu = hci_get_drvdata(hdev);
+	struct qca_data *qca = hu->priv;
+	struct qca_memdump_data *qca_memdump = qca->qca_memdump;
+	char *memdump_buf = NULL;
+
+	wait_on_bit_timeout(&qca->flags, QCA_MEMDUMP_COLLECTION,
+			    TASK_UNINTERRUPTIBLE, MEMDUMP_TIMEOUT_MS);
+
+	clear_bit(QCA_MEMDUMP_COLLECTION, &qca->flags);
+	if (qca->memdump_state == QCA_MEMDUMP_IDLE) {
+		bt_dev_err(hu->hdev, "Clearing the buffers due to timeout");
+		if (qca_memdump)
+			memdump_buf = qca_memdump->memdump_buf_tail;
+		kfree(memdump_buf);
+		kfree(qca_memdump);
+		qca->memdump_state = QCA_MEMDUMP_TIMEOUT;
+		del_timer(&qca->memdump_timer);
+		cancel_work_sync(&qca->ctrl_memdump_evt);
+	}
+}
+
+static void qca_hw_error(struct hci_dev *hdev, u8 code)
+{
+	struct hci_uart *hu = hci_get_drvdata(hdev);
+	struct qca_data *qca = hu->priv;
+
+	bt_dev_info(hdev, "mem_dump_status: %d", qca->memdump_state);
+
+	if (qca->memdump_state == QCA_MEMDUMP_IDLE) {
+		/* If hardware error event received for other than QCA
+		 * soc memory dump event, then we need to crash the SOC
+		 * and wait here for 8 seconds to get the dump packets.
+		 * This will block main thread to be on hold until we
+		 * collect dump.
+		 */
+		set_bit(QCA_MEMDUMP_COLLECTION, &qca->flags);
+		qca_send_crashbuffer(hu);
+		qca_wait_for_dump_collection(hdev);
+	} else if (qca->memdump_state == QCA_MEMDUMP_COLLECTING) {
+		/* Let us wait here until memory dump collected or
+		 * memory dump timer expired.
+		 */
+		bt_dev_info(hdev, "waiting for dump to complete");
+		qca_wait_for_dump_collection(hdev);
+	}
+}
+
+static void qca_cmd_timeout(struct hci_dev *hdev)
+{
+	struct hci_uart *hu = hci_get_drvdata(hdev);
+	struct qca_data *qca = hu->priv;
+
+	if (qca->memdump_state == QCA_MEMDUMP_IDLE)
+		qca_send_crashbuffer(hu);
+	else
+		bt_dev_info(hdev, "Dump collection is in process");
+}
+
 static int qca_wcn3990_init(struct hci_uart *hu)
 {
 	struct qca_serdev *qcadev;
@@ -1320,6 +1599,8 @@ static int qca_setup(struct hci_uart *hu)
 	if (!ret) {
 		set_bit(QCA_IBS_ENABLED, &qca->flags);
 		qca_debugfs_init(hdev);
+		hu->hdev->hw_error = qca_hw_error;
+		hu->hdev->cmd_timeout = qca_cmd_timeout;
 	} else if (ret == -ENOENT) {
 		/* No patch/nvm-config found, run with original fw/config */
 		ret = 0;
@@ -1408,17 +1689,22 @@ static void qca_power_shutdown(struct hci_uart *hu)
 	host_set_baudrate(hu, 2400);
 	qca_send_power_pulse(hu, false);
 	qca_regulator_disable(qcadev);
+	hu->hdev->hw_error = NULL;
+	hu->hdev->cmd_timeout = NULL;
 }
 
 static int qca_power_off(struct hci_dev *hdev)
 {
 	struct hci_uart *hu = hci_get_drvdata(hdev);
+	struct qca_data *qca = hu->priv;
 
-	/* Perform pre shutdown command */
-	qca_send_pre_shutdown_cmd(hdev);
-
-	usleep_range(8000, 10000);
+	/* Stop sending shutdown command if soc crashes. */
+	if (qca->memdump_state == QCA_MEMDUMP_IDLE) {
+		qca_send_pre_shutdown_cmd(hdev);
+		usleep_range(8000, 10000);
+	}
 
+	qca->memdump_state = QCA_MEMDUMP_IDLE;
 	qca_power_shutdown(hu);
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 1efd927d660e6ab02a9cd32fbbe3c7dc47980132 Mon Sep 17 00:00:00 2001
From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Date: Thu, 2 Jan 2020 15:00:55 -0800
Subject: Bluetooth: Add support for LE PHY Update Complete event

This handles LE PHY Update Complete event and store both tx_phy and
rx_phy into hci_conn.

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 include/net/bluetooth/hci.h      |  8 ++++++++
 include/net/bluetooth/hci_core.h |  2 ++
 net/bluetooth/hci_event.c        | 27 +++++++++++++++++++++++++++
 3 files changed, 37 insertions(+)

diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h
index 5bc1e30dedde..07b6ecedc6ce 100644
--- a/include/net/bluetooth/hci.h
+++ b/include/net/bluetooth/hci.h
@@ -2186,6 +2186,14 @@ struct hci_ev_le_direct_adv_info {
 	__s8	 rssi;
 } __packed;
 
+#define HCI_EV_LE_PHY_UPDATE_COMPLETE	0x0c
+struct hci_ev_le_phy_update_complete {
+	__u8  status;
+	__u16 handle;
+	__u8  tx_phy;
+	__u8  rx_phy;
+} __packed;
+
 #define HCI_EV_LE_EXT_ADV_REPORT    0x0d
 struct hci_ev_le_ext_adv_report {
 	__le16 	 evt_type;
diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
index b689aceb636b..faebe3859931 100644
--- a/include/net/bluetooth/hci_core.h
+++ b/include/net/bluetooth/hci_core.h
@@ -493,6 +493,8 @@ struct hci_conn {
 	__u16		le_supv_timeout;
 	__u8		le_adv_data[HCI_MAX_AD_LENGTH];
 	__u8		le_adv_data_len;
+	__u8		le_tx_phy;
+	__u8		le_rx_phy;
 	__s8		rssi;
 	__s8		tx_power;
 	__s8		max_tx_power;
diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c
index 1941f120a376..6ddc4a74a5e4 100644
--- a/net/bluetooth/hci_event.c
+++ b/net/bluetooth/hci_event.c
@@ -5718,6 +5718,29 @@ static void hci_le_direct_adv_report_evt(struct hci_dev *hdev,
 	hci_dev_unlock(hdev);
 }
 
+static void hci_le_phy_update_evt(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	struct hci_ev_le_phy_update_complete *ev = (void *) skb->data;
+	struct hci_conn *conn;
+
+	BT_DBG("%s status 0x%2.2x", hdev->name, ev->status);
+
+	if (!ev->status)
+		return;
+
+	hci_dev_lock(hdev);
+
+	conn = hci_conn_hash_lookup_handle(hdev, __le16_to_cpu(ev->handle));
+	if (!conn)
+		goto unlock;
+
+	conn->le_tx_phy = ev->tx_phy;
+	conn->le_rx_phy = ev->rx_phy;
+
+unlock:
+	hci_dev_unlock(hdev);
+}
+
 static void hci_le_meta_evt(struct hci_dev *hdev, struct sk_buff *skb)
 {
 	struct hci_ev_le_meta *le_ev = (void *) skb->data;
@@ -5753,6 +5776,10 @@ static void hci_le_meta_evt(struct hci_dev *hdev, struct sk_buff *skb)
 		hci_le_direct_adv_report_evt(hdev, skb);
 		break;
 
+	case HCI_EV_LE_PHY_UPDATE_COMPLETE:
+		hci_le_phy_update_evt(hdev, skb);
+		break;
+
 	case HCI_EV_LE_EXT_ADV_REPORT:
 		hci_le_ext_adv_report_evt(hdev, skb);
 		break;
-- 
cgit v1.2.3-59-g8ed1b


From 4b6e228e297b73451f3a4b12fb7d0b24d9d32e6f Mon Sep 17 00:00:00 2001
From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Date: Thu, 2 Jan 2020 15:00:57 -0800
Subject: Bluetooth: Auto tune if input MTU is set to 0

This enables the code to set the input MTU using the underline link
packet types when set to 0, previously this would likely be rejected by
the remote peer since it would be bellow the minimal of 48 for BR/EDR
or 23 for LE, that way it shall be safe to use 0 without causing any
side effects.

This is convenient for the likes of A2DP transport, see:

https://habr.com/en/post/456182/

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 net/bluetooth/l2cap_core.c | 54 ++++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 52 insertions(+), 2 deletions(-)

diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c
index a845786258a0..1bca608e0170 100644
--- a/net/bluetooth/l2cap_core.c
+++ b/net/bluetooth/l2cap_core.c
@@ -1289,6 +1289,9 @@ static void l2cap_le_connect(struct l2cap_chan *chan)
 	if (test_and_set_bit(FLAG_LE_CONN_REQ_SENT, &chan->flags))
 		return;
 
+	if (!chan->imtu)
+		chan->imtu = chan->conn->mtu;
+
 	l2cap_le_flowctl_init(chan, 0);
 
 	req.psm     = chan->psm;
@@ -3226,6 +3229,49 @@ static inline void l2cap_txwin_setup(struct l2cap_chan *chan)
 	chan->ack_win = chan->tx_win;
 }
 
+static void l2cap_mtu_auto(struct l2cap_chan *chan)
+{
+	struct hci_conn *conn = chan->conn->hcon;
+
+	chan->imtu = L2CAP_DEFAULT_MIN_MTU;
+
+	/* The 2-DH1 packet has between 2 and 56 information bytes
+	 * (including the 2-byte payload header)
+	 */
+	if (!(conn->pkt_type & HCI_2DH1))
+		chan->imtu = 54;
+
+	/* The 3-DH1 packet has between 2 and 85 information bytes
+	 * (including the 2-byte payload header)
+	 */
+	if (!(conn->pkt_type & HCI_3DH1))
+		chan->imtu = 83;
+
+	/* The 2-DH3 packet has between 2 and 369 information bytes
+	 * (including the 2-byte payload header)
+	 */
+	if (!(conn->pkt_type & HCI_2DH3))
+		chan->imtu = 367;
+
+	/* The 3-DH3 packet has between 2 and 554 information bytes
+	 * (including the 2-byte payload header)
+	 */
+	if (!(conn->pkt_type & HCI_3DH3))
+		chan->imtu = 552;
+
+	/* The 2-DH5 packet has between 2 and 681 information bytes
+	 * (including the 2-byte payload header)
+	 */
+	if (!(conn->pkt_type & HCI_2DH5))
+		chan->imtu = 679;
+
+	/* The 3-DH5 packet has between 2 and 1023 information bytes
+	 * (including the 2-byte payload header)
+	 */
+	if (!(conn->pkt_type & HCI_3DH5))
+		chan->imtu = 1021;
+}
+
 static int l2cap_build_conf_req(struct l2cap_chan *chan, void *data, size_t data_size)
 {
 	struct l2cap_conf_req *req = data;
@@ -3255,8 +3301,12 @@ static int l2cap_build_conf_req(struct l2cap_chan *chan, void *data, size_t data
 	}
 
 done:
-	if (chan->imtu != L2CAP_DEFAULT_MTU)
-		l2cap_add_conf_opt(&ptr, L2CAP_CONF_MTU, 2, chan->imtu, endptr - ptr);
+	if (chan->imtu != L2CAP_DEFAULT_MTU) {
+		if (!chan->imtu)
+			l2cap_mtu_auto(chan);
+		l2cap_add_conf_opt(&ptr, L2CAP_CONF_MTU, 2, chan->imtu,
+				   endptr - ptr);
+	}
 
 	switch (chan->mode) {
 	case L2CAP_MODE_BASIC:
-- 
cgit v1.2.3-59-g8ed1b


From 3e8bac0cf131353ddf69f35c89b650c4f5a34a2f Mon Sep 17 00:00:00 2001
From: Andrei Otcheretianski <andrei.otcheretianski@intel.com>
Date: Sun, 27 Oct 2019 14:01:43 +0200
Subject: iwlwifi: mvm: Update BEACON_TEMPLATE_CMD firmware API

The new API version adds support for FILS discovery frames.
It adds a new flag and a field for short SSID configuration.
The new API is backward compatible, so we can just switch to it.

Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/api/tx.h | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/tx.h b/drivers/net/wireless/intel/iwlwifi/fw/api/tx.h
index f89a9e16a8c0..f1d1fe96fecc 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/tx.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/tx.h
@@ -813,6 +813,7 @@ enum iwl_mac_beacon_flags {
 	IWL_MAC_BEACON_ANT_A	= BIT(9),
 	IWL_MAC_BEACON_ANT_B	= BIT(10),
 	IWL_MAC_BEACON_ANT_C	= BIT(11),
+	IWL_MAC_BEACON_FILS	= BIT(12),
 };
 
 /**
@@ -820,6 +821,7 @@ enum iwl_mac_beacon_flags {
  * @byte_cnt: byte count of the beacon frame.
  * @flags: least significant byte for rate code. The most significant byte
  *	is &enum iwl_mac_beacon_flags.
+ * @short_ssid: Short SSID
  * @reserved: reserved
  * @template_id: currently equal to the mac context id of the coresponding mac.
  * @tim_idx: the offset of the tim IE in the beacon
@@ -831,14 +833,15 @@ enum iwl_mac_beacon_flags {
 struct iwl_mac_beacon_cmd {
 	__le16 byte_cnt;
 	__le16 flags;
-	__le64 reserved;
+	__le32 short_ssid;
+	__le32 reserved;
 	__le32 template_id;
 	__le32 tim_idx;
 	__le32 tim_size;
 	__le32 ecsa_offset;
 	__le32 csa_offset;
 	struct ieee80211_hdr frame[0];
-} __packed; /* BEACON_TEMPLATE_CMD_API_S_VER_9 */
+} __packed; /* BEACON_TEMPLATE_CMD_API_S_VER_10 */
 
 struct iwl_beacon_notif {
 	struct iwl_mvm_tx_resp beacon_notify_hdr;
-- 
cgit v1.2.3-59-g8ed1b


From 10137f07571287aaa602f492d11db062f2dd3a29 Mon Sep 17 00:00:00 2001
From: Luca Coelho <luciano.coelho@intel.com>
Date: Mon, 28 Oct 2019 11:18:05 +0200
Subject: iwlwifi: yoyo: don't allow changing the domain via debugfs

We don't want to allow changing the domain via debugfs so that we can
apply the domain to all TLV types more easily (doing some at runtime
is difficult due to buffer allocations etc.).  Change the
fw_dbg_domain debugfs file to be read-only and remove the write
function.

Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/debugfs.c  | 29 ++----------------------
 drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c |  3 ++-
 drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.h |  1 -
 3 files changed, 4 insertions(+), 29 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/debugfs.c b/drivers/net/wireless/intel/iwlwifi/fw/debugfs.c
index ca3b1a461dea..89f74116569d 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/debugfs.c
+++ b/drivers/net/wireless/intel/iwlwifi/fw/debugfs.c
@@ -320,31 +320,6 @@ out:
 
 FWRT_DEBUGFS_WRITE_FILE_OPS(send_hcmd, 512);
 
-static ssize_t iwl_dbgfs_fw_dbg_domain_write(struct iwl_fw_runtime *fwrt,
-					     char *buf, size_t count)
-{
-	u32 new_domain;
-	int ret;
-
-	if (!iwl_trans_fw_running(fwrt->trans))
-		return -EIO;
-
-	ret = kstrtou32(buf, 0, &new_domain);
-	if (ret)
-		return ret;
-
-	if (new_domain != fwrt->trans->dbg.domains_bitmap) {
-		ret = iwl_dbg_tlv_gen_active_trigs(fwrt, new_domain);
-		if (ret)
-			return ret;
-
-		iwl_dbg_tlv_time_point(fwrt, IWL_FW_INI_TIME_POINT_PERIODIC,
-				       NULL);
-	}
-
-	return count;
-}
-
 static ssize_t iwl_dbgfs_fw_dbg_domain_read(struct iwl_fw_runtime *fwrt,
 					    size_t size, char *buf)
 {
@@ -352,7 +327,7 @@ static ssize_t iwl_dbgfs_fw_dbg_domain_read(struct iwl_fw_runtime *fwrt,
 			 fwrt->trans->dbg.domains_bitmap);
 }
 
-FWRT_DEBUGFS_READ_WRITE_FILE_OPS(fw_dbg_domain, 20);
+FWRT_DEBUGFS_READ_FILE_OPS(fw_dbg_domain, 20);
 
 void iwl_fwrt_dbgfs_register(struct iwl_fw_runtime *fwrt,
 			    struct dentry *dbgfs_dir)
@@ -360,5 +335,5 @@ void iwl_fwrt_dbgfs_register(struct iwl_fw_runtime *fwrt,
 	INIT_DELAYED_WORK(&fwrt->timestamp.wk, iwl_fw_timestamp_marker_wk);
 	FWRT_DEBUGFS_ADD_FILE(timestamp_marker, dbgfs_dir, 0200);
 	FWRT_DEBUGFS_ADD_FILE(send_hcmd, dbgfs_dir, 0200);
-	FWRT_DEBUGFS_ADD_FILE(fw_dbg_domain, dbgfs_dir, 0600);
+	FWRT_DEBUGFS_ADD_FILE(fw_dbg_domain, dbgfs_dir, 0400);
 }
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c b/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
index f266647dc08c..79c8c8057c2d 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
@@ -913,7 +913,8 @@ iwl_dbg_tlv_gen_active_trig_list(struct iwl_fw_runtime *fwrt,
 	}
 }
 
-int iwl_dbg_tlv_gen_active_trigs(struct iwl_fw_runtime *fwrt, u32 new_domain)
+static int iwl_dbg_tlv_gen_active_trigs(struct iwl_fw_runtime *fwrt,
+					u32 new_domain)
 {
 	int i;
 
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.h b/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.h
index f18946872569..1360676b3b21 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.h
@@ -105,7 +105,6 @@ void iwl_dbg_tlv_init(struct iwl_trans *trans);
 void iwl_dbg_tlv_time_point(struct iwl_fw_runtime *fwrt,
 			    enum iwl_fw_ini_time_point tp_id,
 			    union iwl_dbg_tlv_tp_data *tp_data);
-int iwl_dbg_tlv_gen_active_trigs(struct iwl_fw_runtime *fwrt, u32 new_domain);
 void iwl_dbg_tlv_del_timers(struct iwl_trans *trans);
 
 #endif /* __iwl_dbg_tlv_h__*/
-- 
cgit v1.2.3-59-g8ed1b


From 058c411d19b21b8e5581cdeddae2392d5f421daa Mon Sep 17 00:00:00 2001
From: Luca Coelho <luciano.coelho@intel.com>
Date: Mon, 28 Oct 2019 12:39:10 +0200
Subject: iwlwifi: yoyo: remove unnecessary active triggers status flag

Now that we can't change the domain at runtime anymore, we don't have
to protect the active trigger status.  Remove it.  Additionally, we
don't need to flush the dumps at this point anymore, since this only
runs during initialization and there shouldn't be any dumps running.

Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/dbg.c      |  3 ---
 drivers/net/wireless/intel/iwlwifi/fw/runtime.h  | 11 -----------
 drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c | 14 +-------------
 3 files changed, 1 insertion(+), 27 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/dbg.c b/drivers/net/wireless/intel/iwlwifi/fw/dbg.c
index c8bc6b9b53a4..09daf0f1f6f8 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/dbg.c
+++ b/drivers/net/wireless/intel/iwlwifi/fw/dbg.c
@@ -2353,9 +2353,6 @@ int iwl_fw_dbg_ini_collect(struct iwl_fw_runtime *fwrt,
 	u32 occur, delay;
 	unsigned long idx;
 
-	if (test_bit(STATUS_GEN_ACTIVE_TRIGS, &fwrt->status))
-		return -EBUSY;
-
 	if (!iwl_fw_ini_trigger_on(fwrt, trig)) {
 		IWL_WARN(fwrt, "WRT: Trigger %d is not active, aborting dump\n",
 			 tp_id);
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/runtime.h b/drivers/net/wireless/intel/iwlwifi/fw/runtime.h
index c24575ff0e54..1b4bca717e79 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/runtime.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/runtime.h
@@ -129,14 +129,6 @@ struct iwl_txf_iter_data {
 	u8 internal_txf;
 };
 
-/**
- * enum iwl_fw_runtime_status - fw runtime status flags
- * @STATUS_GEN_ACTIVE_TRIGS: generating active trigger list
- */
-enum iwl_fw_runtime_status {
-	STATUS_GEN_ACTIVE_TRIGS,
-};
-
 /**
  * struct iwl_fw_runtime - runtime data for firmware
  * @fw: firmware image
@@ -150,7 +142,6 @@ enum iwl_fw_runtime_status {
  * @smem_cfg: saved firmware SMEM configuration
  * @cur_fw_img: current firmware image, must be maintained by
  *	the driver by calling &iwl_fw_set_current_image()
- * @status: &enum iwl_fw_runtime_status
  * @dump: debug dump data
  */
 struct iwl_fw_runtime {
@@ -171,8 +162,6 @@ struct iwl_fw_runtime {
 	/* memory configuration */
 	struct iwl_fwrt_shared_mem_cfg smem_cfg;
 
-	unsigned long status;
-
 	/* debug */
 	struct {
 		const struct iwl_fw_dump_desc *desc;
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c b/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
index 79c8c8057c2d..4acb24f8fc11 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
@@ -891,15 +891,10 @@ static void
 iwl_dbg_tlv_gen_active_trig_list(struct iwl_fw_runtime *fwrt,
 				 struct iwl_dbg_tlv_time_point_data *tp)
 {
-	struct iwl_dbg_tlv_node *node, *tmp;
+	struct iwl_dbg_tlv_node *node;
 	struct list_head *trig_list = &tp->trig_list;
 	struct list_head *active_trig_list = &tp->active_trig_list;
 
-	list_for_each_entry_safe(node, tmp, active_trig_list, list) {
-		list_del(&node->list);
-		kfree(node);
-	}
-
 	list_for_each_entry(node, trig_list, list) {
 		struct iwl_ucode_tlv *tlv = &node->tlv;
 		struct iwl_fw_ini_trigger_tlv *trig = (void *)tlv->data;
@@ -918,11 +913,6 @@ static int iwl_dbg_tlv_gen_active_trigs(struct iwl_fw_runtime *fwrt,
 {
 	int i;
 
-	if (test_and_set_bit(STATUS_GEN_ACTIVE_TRIGS, &fwrt->status))
-		return -EBUSY;
-
-	iwl_fw_flush_dumps(fwrt);
-
 	fwrt->trans->dbg.domains_bitmap = new_domain;
 
 	IWL_DEBUG_FW(fwrt,
@@ -936,8 +926,6 @@ static int iwl_dbg_tlv_gen_active_trigs(struct iwl_fw_runtime *fwrt,
 		iwl_dbg_tlv_gen_active_trig_list(fwrt, tp);
 	}
 
-	clear_bit(STATUS_GEN_ACTIVE_TRIGS, &fwrt->status);
-
 	return 0;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 5cbc310810090cf5a295835113d3b9e5dbbe587f Mon Sep 17 00:00:00 2001
From: Luca Coelho <luciano.coelho@intel.com>
Date: Mon, 28 Oct 2019 12:57:00 +0200
Subject: iwlwifi: yoyo: remove the iwl_dbg_tlv_gen_active_trigs() function

We only call this function from a single place and it's very
very small and self-contained anyway, so remove the function and move
the code into the caller.

Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c | 34 +++++++++---------------
 1 file changed, 12 insertions(+), 22 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c b/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
index 4acb24f8fc11..31a0e885faf6 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
@@ -908,27 +908,6 @@ iwl_dbg_tlv_gen_active_trig_list(struct iwl_fw_runtime *fwrt,
 	}
 }
 
-static int iwl_dbg_tlv_gen_active_trigs(struct iwl_fw_runtime *fwrt,
-					u32 new_domain)
-{
-	int i;
-
-	fwrt->trans->dbg.domains_bitmap = new_domain;
-
-	IWL_DEBUG_FW(fwrt,
-		     "WRT: Generating active triggers list, domain 0x%x\n",
-		     fwrt->trans->dbg.domains_bitmap);
-
-	for (i = 0; i < ARRAY_SIZE(fwrt->trans->dbg.time_point); i++) {
-		struct iwl_dbg_tlv_time_point_data *tp =
-			&fwrt->trans->dbg.time_point[i];
-
-		iwl_dbg_tlv_gen_active_trig_list(fwrt, tp);
-	}
-
-	return 0;
-}
-
 static bool iwl_dbg_tlv_check_fw_pkt(struct iwl_fw_runtime *fwrt,
 				     struct iwl_fwrt_dump_data *dump_data,
 				     union iwl_dbg_tlv_tp_data *tp_data,
@@ -1002,7 +981,18 @@ static void iwl_dbg_tlv_init_cfg(struct iwl_fw_runtime *fwrt)
 	enum iwl_fw_ini_buffer_location *ini_dest = &fwrt->trans->dbg.ini_dest;
 	int ret, i;
 
-	iwl_dbg_tlv_gen_active_trigs(fwrt, IWL_FW_DBG_DOMAIN);
+	fwrt->trans->dbg.domains_bitmap = IWL_FW_DBG_DOMAIN;
+
+	IWL_DEBUG_FW(fwrt,
+		     "WRT: Generating active triggers list, domain 0x%x\n",
+		     fwrt->trans->dbg.domains_bitmap);
+
+	for (i = 0; i < ARRAY_SIZE(fwrt->trans->dbg.time_point); i++) {
+		struct iwl_dbg_tlv_time_point_data *tp =
+			&fwrt->trans->dbg.time_point[i];
+
+		iwl_dbg_tlv_gen_active_trig_list(fwrt, tp);
+	}
 
 	*ini_dest = IWL_FW_INI_LOCATION_INVALID;
 	for (i = 0; i < IWL_FW_INI_ALLOCATION_NUM; i++) {
-- 
cgit v1.2.3-59-g8ed1b


From e6ec237d3068dd023f5a20fd6c11527e57961019 Mon Sep 17 00:00:00 2001
From: Oren Givon <oren.givon@intel.com>
Date: Tue, 22 Oct 2019 14:14:24 +0300
Subject: iwlwifi: add new iwlax411 struct for type SoSnj

Add new struct for SoSnj and add uhb support for ax411 structs.

Signed-off-by: Oren Givon <oren.givon@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/cfg/22000.c  | 10 ++++++++++
 drivers/net/wireless/intel/iwlwifi/iwl-config.h |  1 +
 2 files changed, 11 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
index 837c06891f5e..a22a830019c0 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
@@ -92,6 +92,7 @@
 #define IWL_22000_SO_A_GF_A_FW_PRE      "iwlwifi-so-a0-gf-a0-"
 #define IWL_22000_TY_A_GF_A_FW_PRE      "iwlwifi-ty-a0-gf-a0-"
 #define IWL_22000_SO_A_GF4_A_FW_PRE     "iwlwifi-so-a0-gf4-a0-"
+#define IWL_22000_SOSNJ_A_GF4_A_FW_PRE  "iwlwifi-SoSnj-a0-gf4-a0-"
 
 #define IWL_22000_HR_MODULE_FIRMWARE(api) \
 	IWL_22000_HR_FW_PRE __stringify(api) ".ucode"
@@ -707,6 +708,15 @@ const struct iwl_cfg iwlax210_2ax_cfg_ty_gf_a0 = {
 const struct iwl_cfg iwlax411_2ax_cfg_so_gf4_a0 = {
 	.name = "Intel(R) Wi-Fi 7 AX411 160MHz",
 	.fw_name_pre = IWL_22000_SO_A_GF4_A_FW_PRE,
+	.uhb_supported = true,
+	IWL_DEVICE_AX210,
+	.num_rbds = IWL_NUM_RBDS_AX210_HE,
+};
+
+const struct iwl_cfg iwlax411_2ax_cfg_sosnj_gf4_a0 = {
+	.name = "Intel(R) Wi-Fi 7 AX411 160MHz",
+	.fw_name_pre = IWL_22000_SOSNJ_A_GF4_A_FW_PRE,
+	.uhb_supported = true,
 	IWL_DEVICE_AX210,
 	.num_rbds = IWL_NUM_RBDS_AX210_HE,
 };
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
index bb91c36c9d8b..bbeb5fc3caad 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
@@ -586,6 +586,7 @@ extern const struct iwl_cfg iwlax210_2ax_cfg_so_hr_a0;
 extern const struct iwl_cfg iwlax211_2ax_cfg_so_gf_a0;
 extern const struct iwl_cfg iwlax210_2ax_cfg_ty_gf_a0;
 extern const struct iwl_cfg iwlax411_2ax_cfg_so_gf4_a0;
+extern const struct iwl_cfg iwlax411_2ax_cfg_sosnj_gf4_a0;
 #endif /* CPTCFG_IWLMVM || CPTCFG_IWLFMAC */
 
 #endif /* __IWL_CONFIG_H__ */
-- 
cgit v1.2.3-59-g8ed1b


From e701da0c2320148a378ed1d191b6c4f7e8e82c4c Mon Sep 17 00:00:00 2001
From: Luca Coelho <luciano.coelho@intel.com>
Date: Wed, 30 Oct 2019 09:40:12 +0200
Subject: iwlwifi: yoyo: check for the domain on all TLV types during init

Now that we don't have dynamically changing domains anymore, we can
simply skip all the TLVs with domains that are not enabled.  To do so,
remove the checks from the functions that handle the TLVs when a
timepoint is reached to the top allocation function.

Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/runtime.h  |  2 +-
 drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c | 22 +++++++++-------------
 drivers/net/wireless/intel/iwlwifi/iwl-drv.c     |  2 ++
 drivers/net/wireless/intel/iwlwifi/iwl-trans.h   |  2 ++
 4 files changed, 14 insertions(+), 14 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/runtime.h b/drivers/net/wireless/intel/iwlwifi/fw/runtime.h
index 1b4bca717e79..f8c6ed823bc5 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/runtime.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/runtime.h
@@ -69,7 +69,7 @@
 #include "iwl-eeprom-parse.h"
 #include "fw/acpi.h"
 
-#define IWL_FW_DBG_DOMAIN		IWL_FW_INI_DOMAIN_ALWAYS_ON
+#define IWL_FW_DBG_DOMAIN		IWL_TRANS_FW_DBG_DOMAIN(fwrt->trans)
 
 struct iwl_fw_runtime_ops {
 	int (*dump_start)(void *ctx);
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c b/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
index 31a0e885faf6..eaf64ca3a134 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
@@ -290,10 +290,19 @@ void iwl_dbg_tlv_alloc(struct iwl_trans *trans, struct iwl_ucode_tlv *tlv,
 	struct iwl_fw_ini_header *hdr = (void *)&tlv->data[0];
 	u32 type = le32_to_cpu(tlv->type);
 	u32 tlv_idx = type - IWL_UCODE_TLV_DEBUG_BASE;
+	u32 domain = le32_to_cpu(hdr->domain);
 	enum iwl_ini_cfg_state *cfg_state = ext ?
 		&trans->dbg.external_ini_cfg : &trans->dbg.internal_ini_cfg;
 	int ret;
 
+	if (domain != IWL_FW_INI_DOMAIN_ALWAYS_ON &&
+	    !(domain & trans->dbg.domains_bitmap)) {
+		IWL_DEBUG_FW(trans,
+			     "WRT: Skipping TLV with disabled domain 0x%0x (0x%0x)\n",
+			     domain, trans->dbg.domains_bitmap);
+		return;
+	}
+
 	if (tlv_idx >= ARRAY_SIZE(dbg_tlv_alloc) || !dbg_tlv_alloc[tlv_idx]) {
 		IWL_ERR(trans, "WRT: Unsupported TLV type 0x%x\n", type);
 		goto out_err;
@@ -660,7 +669,6 @@ static void iwl_dbg_tlv_send_hcmds(struct iwl_fw_runtime *fwrt,
 	list_for_each_entry(node, hcmd_list, list) {
 		struct iwl_fw_ini_hcmd_tlv *hcmd = (void *)node->tlv.data;
 		struct iwl_fw_ini_hcmd *hcmd_data = &hcmd->hcmd;
-		u32 domain = le32_to_cpu(hcmd->hdr.domain);
 		u16 hcmd_len = le32_to_cpu(node->tlv.length) - sizeof(*hcmd);
 		struct iwl_host_cmd cmd = {
 			.id = WIDE_ID(hcmd_data->group, hcmd_data->id),
@@ -668,10 +676,6 @@ static void iwl_dbg_tlv_send_hcmds(struct iwl_fw_runtime *fwrt,
 			.data = { hcmd_data->data, },
 		};
 
-		if (domain != IWL_FW_INI_DOMAIN_ALWAYS_ON &&
-		    !(domain & fwrt->trans->dbg.domains_bitmap))
-			continue;
-
 		iwl_trans_send_cmd(fwrt->trans, &cmd);
 	}
 }
@@ -897,12 +901,6 @@ iwl_dbg_tlv_gen_active_trig_list(struct iwl_fw_runtime *fwrt,
 
 	list_for_each_entry(node, trig_list, list) {
 		struct iwl_ucode_tlv *tlv = &node->tlv;
-		struct iwl_fw_ini_trigger_tlv *trig = (void *)tlv->data;
-		u32 domain = le32_to_cpu(trig->hdr.domain);
-
-		if (domain != IWL_FW_INI_DOMAIN_ALWAYS_ON &&
-		    !(domain & fwrt->trans->dbg.domains_bitmap))
-			continue;
 
 		iwl_dbg_tlv_add_active_trigger(fwrt, active_trig_list, tlv);
 	}
@@ -981,8 +979,6 @@ static void iwl_dbg_tlv_init_cfg(struct iwl_fw_runtime *fwrt)
 	enum iwl_fw_ini_buffer_location *ini_dest = &fwrt->trans->dbg.ini_dest;
 	int ret, i;
 
-	fwrt->trans->dbg.domains_bitmap = IWL_FW_DBG_DOMAIN;
-
 	IWL_DEBUG_FW(fwrt,
 		     "WRT: Generating active triggers list, domain 0x%x\n",
 		     fwrt->trans->dbg.domains_bitmap);
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
index 4096ccf58b07..c6cb5e1a1bd9 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
@@ -1647,6 +1647,8 @@ struct iwl_drv *iwl_drv_start(struct iwl_trans *trans)
 	drv->trans->dbgfs_dir = debugfs_create_dir("trans", drv->dbgfs_drv);
 #endif
 
+	drv->trans->dbg.domains_bitmap = IWL_TRANS_FW_DBG_DOMAIN(drv->trans);
+
 	ret = iwl_request_firmware(drv, true);
 	if (ret) {
 		IWL_ERR(trans, "Couldn't request the fw\n");
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
index 872cda2be0f6..58bee4442e20 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
@@ -112,6 +112,8 @@
  *	6) Eventually, the free function will be called.
  */
 
+#define IWL_TRANS_FW_DBG_DOMAIN(trans)	IWL_FW_INI_DOMAIN_ALWAYS_ON
+
 #define FH_RSCSR_FRAME_SIZE_MSK		0x00003FFF	/* bits 0-13 */
 #define FH_RSCSR_FRAME_INVALID		0x55550000
 #define FH_RSCSR_FRAME_ALIGN		0x40
-- 
cgit v1.2.3-59-g8ed1b


From 55b514b479ff8a6e2473e7ef2349d49380c5fa61 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 31 Oct 2019 10:28:32 +0100
Subject: iwlwifi: incorporate firmware filename into version

We have many different firmware images with the same version,
and it's sometimes cumbersome to figure out which image was
really used, especially as the marketing strings that we do
print out can be the same for (slightly) different hardware
using different firmware images.

Incorporate the firmware filename into the fw_version so it's
printed out all the time. Unfortunately, this will make the
string be longer than the 32 characters for ethtool, but we
almost never really use ethtool, so strip the "iwlwifi-"
prefix (if not overridden), and the remaining data that may
then be stripped at the end is not usually useful anyway.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/img.h  |  2 +-
 drivers/net/wireless/intel/iwlwifi/iwl-drv.c | 24 ++++++++++++++++++------
 2 files changed, 19 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/img.h b/drivers/net/wireless/intel/iwlwifi/fw/img.h
index 994880a83652..90ca5f929cf9 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/img.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/img.h
@@ -251,7 +251,7 @@ struct iwl_fw_dbg {
 struct iwl_fw {
 	u32 ucode_ver;
 
-	char fw_version[ETHTOOL_FWVERS_LEN];
+	char fw_version[64];
 
 	/* ucode images */
 	struct fw_img img[IWL_UCODE_TYPE_MAX];
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
index c6cb5e1a1bd9..773de0b724be 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
@@ -493,6 +493,16 @@ static void iwl_set_ucode_capabilities(struct iwl_drv *drv, const u8 *data,
 	}
 }
 
+static const char *iwl_reduced_fw_name(struct iwl_drv *drv)
+{
+	const char *name = drv->firmware_name;
+
+	if (strncmp(name, "iwlwifi-", 8) == 0)
+		name += 8;
+
+	return name;
+}
+
 static int iwl_parse_v1_v2_firmware(struct iwl_drv *drv,
 				    const struct firmware *ucode_raw,
 				    struct iwl_firmware_pieces *pieces)
@@ -551,12 +561,12 @@ static int iwl_parse_v1_v2_firmware(struct iwl_drv *drv,
 
 	snprintf(drv->fw.fw_version,
 		 sizeof(drv->fw.fw_version),
-		 "%u.%u.%u.%u%s",
+		 "%u.%u.%u.%u%s %s",
 		 IWL_UCODE_MAJOR(drv->fw.ucode_ver),
 		 IWL_UCODE_MINOR(drv->fw.ucode_ver),
 		 IWL_UCODE_API(drv->fw.ucode_ver),
 		 IWL_UCODE_SERIAL(drv->fw.ucode_ver),
-		 buildstr);
+		 buildstr, iwl_reduced_fw_name(drv));
 
 	/* Verify size of file vs. image size info in file's header */
 
@@ -636,12 +646,12 @@ static int iwl_parse_tlv_firmware(struct iwl_drv *drv,
 
 	snprintf(drv->fw.fw_version,
 		 sizeof(drv->fw.fw_version),
-		 "%u.%u.%u.%u%s",
+		 "%u.%u.%u.%u%s %s",
 		 IWL_UCODE_MAJOR(drv->fw.ucode_ver),
 		 IWL_UCODE_MINOR(drv->fw.ucode_ver),
 		 IWL_UCODE_API(drv->fw.ucode_ver),
 		 IWL_UCODE_SERIAL(drv->fw.ucode_ver),
-		 buildstr);
+		 buildstr, iwl_reduced_fw_name(drv));
 
 	data = ucode->data;
 
@@ -895,11 +905,13 @@ static int iwl_parse_tlv_firmware(struct iwl_drv *drv,
 			if (major >= 35)
 				snprintf(drv->fw.fw_version,
 					 sizeof(drv->fw.fw_version),
-					"%u.%08x.%u", major, minor, local_comp);
+					"%u.%08x.%u %s", major, minor,
+					local_comp, iwl_reduced_fw_name(drv));
 			else
 				snprintf(drv->fw.fw_version,
 					 sizeof(drv->fw.fw_version),
-					"%u.%u.%u", major, minor, local_comp);
+					"%u.%u.%u %s", major, minor,
+					local_comp, iwl_reduced_fw_name(drv));
 			break;
 			}
 		case IWL_UCODE_TLV_FW_DBG_DEST: {
-- 
cgit v1.2.3-59-g8ed1b


From 20f5aef55838241e797ba685235d91a3449e9f91 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 29 Oct 2019 15:45:40 +0100
Subject: iwlwifi: mvm: print out extended secboot status before dump

Print out the secure boot status, extended by the PCs of LMACs
and the UMAC. This needs to be before dumping, as dumping will
corrupt the PC (if the NMI is handled), so move that down.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-prph.h |  3 +++
 drivers/net/wireless/intel/iwlwifi/mvm/fw.c   | 27 ++++++++++++++++++++-------
 2 files changed, 23 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-prph.h b/drivers/net/wireless/intel/iwlwifi/iwl-prph.h
index 14c8ba23f3b9..c21d99418af9 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-prph.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-prph.h
@@ -430,6 +430,9 @@ enum aux_misc_master1_en {
 #define UMAG_SB_CPU_1_STATUS		0xA038C0
 #define UMAG_SB_CPU_2_STATUS		0xA038C4
 #define UMAG_GEN_HW_STATUS		0xA038C8
+#define UREG_UMAC_CURRENT_PC		0xa05c18
+#define UREG_LMAC1_CURRENT_PC		0xa05c1c
+#define UREG_LMAC2_CURRENT_PC		0xa05c20
 
 /* For UMAG_GEN_HW_STATUS reg check */
 enum {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index dd685f7eb410..67d414e5db19 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -353,22 +353,35 @@ static int iwl_mvm_load_ucode_wait_alive(struct iwl_mvm *mvm,
 	if (ret) {
 		struct iwl_trans *trans = mvm->trans;
 
-		if (ret == -ETIMEDOUT)
-			iwl_fw_dbg_error_collect(&mvm->fwrt,
-						 FW_DBG_TRIGGER_ALIVE_TIMEOUT);
-
-		if (trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_22000)
+		if (trans->trans_cfg->device_family >=
+					IWL_DEVICE_FAMILY_22000) {
 			IWL_ERR(mvm,
 				"SecBoot CPU1 Status: 0x%x, CPU2 Status: 0x%x\n",
 				iwl_read_umac_prph(trans, UMAG_SB_CPU_1_STATUS),
 				iwl_read_umac_prph(trans,
 						   UMAG_SB_CPU_2_STATUS));
-		else if (trans->trans_cfg->device_family >=
-			 IWL_DEVICE_FAMILY_8000)
+			IWL_ERR(mvm, "UMAC PC: 0x%x\n",
+				iwl_read_umac_prph(trans,
+						   UREG_UMAC_CURRENT_PC));
+			IWL_ERR(mvm, "LMAC PC: 0x%x\n",
+				iwl_read_umac_prph(trans,
+						   UREG_LMAC1_CURRENT_PC));
+			if (iwl_mvm_is_cdb_supported(mvm))
+				IWL_ERR(mvm, "LMAC2 PC: 0x%x\n",
+					iwl_read_umac_prph(trans,
+						UREG_LMAC2_CURRENT_PC));
+		} else if (trans->trans_cfg->device_family >=
+			   IWL_DEVICE_FAMILY_8000) {
 			IWL_ERR(mvm,
 				"SecBoot CPU1 Status: 0x%x, CPU2 Status: 0x%x\n",
 				iwl_read_prph(trans, SB_CPU_1_STATUS),
 				iwl_read_prph(trans, SB_CPU_2_STATUS));
+		}
+
+		if (ret == -ETIMEDOUT)
+			iwl_fw_dbg_error_collect(&mvm->fwrt,
+						 FW_DBG_TRIGGER_ALIVE_TIMEOUT);
+
 		iwl_fw_set_current_image(&mvm->fwrt, old_type);
 		return ret;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 87f3073c16326dc6916e277df8961abb6ff17a5f Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 31 Oct 2019 11:04:43 +0100
Subject: iwlwifi: prph: remove some unused register definitions

There are some unused register definitions, remove them.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-prph.h | 7 -------
 1 file changed, 7 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-prph.h b/drivers/net/wireless/intel/iwlwifi/iwl-prph.h
index c21d99418af9..1136d9784f9d 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-prph.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-prph.h
@@ -411,13 +411,6 @@ enum {
 	HW_STEP_LOCATION_BITS = 24,
 };
 
-#define AUX_MISC_MASTER1_EN		0xA20818
-enum aux_misc_master1_en {
-	AUX_MISC_MASTER1_EN_SBE_MSK	= 0x1,
-};
-
-#define AUX_MISC_MASTER1_SMPHR_STATUS	0xA20800
-#define RSA_ENABLE			0xA24B08
 #define PREG_AUX_BUS_WPROT_0		0xA04CC0
 
 /* device family 9000 WPROT register */
-- 
cgit v1.2.3-59-g8ed1b


From b3bd6416fc77b5056d2dda7a4d5814ec81af7ecd Mon Sep 17 00:00:00 2001
From: Luca Coelho <luciano.coelho@intel.com>
Date: Thu, 10 Oct 2019 15:57:35 +0300
Subject: iwlwifi: assume the driver_data is a trans_cfg, but allow full cfg

With the new concept of separating the trans-specific (trans_cfg) data
from the rest of the cfg, we will start mapping only the trans_cfg
part to the PCI device ID/subsystem device ID.  So we can assume that
the data passed to the probe function contains the trans_cfg, but
since the full cfg still contains the trans_cfg at the beginning, we
can allow a full cfg to be passed as well.  This makes it easier to
convert the existing tables one by one.

Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/pcie/drv.c | 90 ++++++++++++++++-----------
 1 file changed, 55 insertions(+), 35 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
index 79c24c696c8f..5eb528d2111f 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
@@ -991,21 +991,32 @@ MODULE_DEVICE_TABLE(pci, iwl_hw_card_ids);
 
 static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
-	const struct iwl_cfg *cfg = (struct iwl_cfg *)(ent->driver_data);
+	const struct iwl_cfg_trans_params *trans =
+		(struct iwl_cfg_trans_params *)(ent->driver_data);
 	const struct iwl_cfg *cfg_7265d __maybe_unused = NULL;
 	struct iwl_trans *iwl_trans;
 	struct iwl_trans_pcie *trans_pcie;
 	unsigned long flags;
 	int ret;
+	/*
+	 * This is needed for backwards compatibility with the old
+	 * tables, so we don't need to change all the config structs
+	 * at the same time.  The cfg is used to compare with the old
+	 * full cfg structs.
+	 */
+	const struct iwl_cfg *cfg = (struct iwl_cfg *)(ent->driver_data);
 
-	iwl_trans = iwl_trans_pcie_alloc(pdev, ent, &cfg->trans);
+	/* make sure trans is the first element in iwl_cfg */
+	BUILD_BUG_ON(offsetof(struct iwl_cfg, trans));
+
+	iwl_trans = iwl_trans_pcie_alloc(pdev, ent, trans);
 	if (IS_ERR(iwl_trans))
 		return PTR_ERR(iwl_trans);
 
 	trans_pcie = IWL_TRANS_GET_PCIE_TRANS(iwl_trans);
 
 	/* the trans_cfg should never change, so set it now */
-	iwl_trans->trans_cfg = &cfg->trans;
+	iwl_trans->trans_cfg = trans;
 
 #if IS_ENABLED(CONFIG_IWLMVM)
 	/*
@@ -1023,22 +1034,22 @@ static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		cfg_7265d = &iwl7265d_n_cfg;
 	if (cfg_7265d &&
 	    (iwl_trans->hw_rev & CSR_HW_REV_TYPE_MSK) == CSR_HW_REV_TYPE_7265D)
-		cfg = cfg_7265d;
+		iwl_trans->cfg = cfg_7265d;
 
 	iwl_trans->hw_rf_id = iwl_read32(iwl_trans, CSR_HW_RF_ID);
 
 	if (cfg == &iwlax210_2ax_cfg_so_hr_a0) {
 		if (iwl_trans->hw_rev == CSR_HW_REV_TYPE_TY) {
-			cfg = &iwlax210_2ax_cfg_ty_gf_a0;
+			iwl_trans->cfg = &iwlax210_2ax_cfg_ty_gf_a0;
 		} else if (CSR_HW_RF_ID_TYPE_CHIP_ID(iwl_trans->hw_rf_id) ==
 			   CSR_HW_RF_ID_TYPE_CHIP_ID(CSR_HW_RF_ID_TYPE_JF)) {
-			cfg = &iwlax210_2ax_cfg_so_jf_a0;
+			iwl_trans->cfg = &iwlax210_2ax_cfg_so_jf_a0;
 		} else if (CSR_HW_RF_ID_TYPE_CHIP_ID(iwl_trans->hw_rf_id) ==
 			   CSR_HW_RF_ID_TYPE_CHIP_ID(CSR_HW_RF_ID_TYPE_GF)) {
-			cfg = &iwlax211_2ax_cfg_so_gf_a0;
+			iwl_trans->cfg = &iwlax211_2ax_cfg_so_gf_a0;
 		} else if (CSR_HW_RF_ID_TYPE_CHIP_ID(iwl_trans->hw_rf_id) ==
 			   CSR_HW_RF_ID_TYPE_CHIP_ID(CSR_HW_RF_ID_TYPE_GF4)) {
-			cfg = &iwlax411_2ax_cfg_so_gf4_a0;
+			iwl_trans->cfg = &iwlax411_2ax_cfg_so_gf4_a0;
 		}
 	} else if (cfg == &iwl_ax101_cfg_qu_hr) {
 		if ((CSR_HW_RF_ID_TYPE_CHIP_ID(iwl_trans->hw_rf_id) ==
@@ -1046,13 +1057,17 @@ static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		     iwl_trans->hw_rev == CSR_HW_REV_TYPE_QNJ_B0) ||
 		    (CSR_HW_RF_ID_TYPE_CHIP_ID(iwl_trans->hw_rf_id) ==
 		     CSR_HW_RF_ID_TYPE_CHIP_ID(CSR_HW_RF_ID_TYPE_HR1))) {
-			cfg = &iwl22000_2ax_cfg_qnj_hr_b0;
+			iwl_trans->cfg = &iwl22000_2ax_cfg_qnj_hr_b0;
+		} else if (CSR_HW_RF_ID_TYPE_CHIP_ID(iwl_trans->hw_rf_id) ==
+		    CSR_HW_RF_ID_TYPE_CHIP_ID(CSR_HW_RF_ID_TYPE_HR) &&
+		    iwl_trans->hw_rev == CSR_HW_REV_TYPE_QUZ) {
+			iwl_trans->cfg = &iwl_ax101_cfg_quz_hr;
 		} else if (CSR_HW_RF_ID_TYPE_CHIP_ID(iwl_trans->hw_rf_id) ==
 			   CSR_HW_RF_ID_TYPE_CHIP_ID(CSR_HW_RF_ID_TYPE_HR)) {
-			cfg = &iwl_ax101_cfg_qu_hr;
+			iwl_trans->cfg = &iwl_ax101_cfg_qu_hr;
 		} else if (CSR_HW_RF_ID_TYPE_CHIP_ID(iwl_trans->hw_rf_id) ==
 			   CSR_HW_RF_ID_TYPE_CHIP_ID(CSR_HW_RF_ID_TYPE_JF)) {
-			cfg = &iwl22000_2ax_cfg_jf;
+			iwl_trans->cfg = &iwl22000_2ax_cfg_jf;
 		} else if (CSR_HW_RF_ID_TYPE_CHIP_ID(iwl_trans->hw_rf_id) ==
 			   CSR_HW_RF_ID_TYPE_CHIP_ID(CSR_HW_RF_ID_TYPE_HRCDB)) {
 			IWL_ERR(iwl_trans, "RF ID HRCDB is not supported\n");
@@ -1069,14 +1084,11 @@ static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 		hw_status = iwl_read_prph(iwl_trans, UMAG_GEN_HW_STATUS);
 		if (CSR_HW_RF_STEP(iwl_trans->hw_rf_id) == SILICON_B_STEP)
-			/*
-			 * b step fw is the same for physical card and fpga
-			 */
-			cfg = &iwl22000_2ax_cfg_qnj_hr_b0;
+			iwl_trans->cfg = &iwl22000_2ax_cfg_qnj_hr_b0;
 		else if ((hw_status & UMAG_GEN_HW_IS_FPGA) &&
-			 CSR_HW_RF_STEP(iwl_trans->hw_rf_id) == SILICON_A_STEP) {
-			cfg = &iwl22000_2ax_cfg_qnj_hr_a0_f0;
-		}
+			 CSR_HW_RF_STEP(iwl_trans->hw_rf_id) ==
+			 SILICON_A_STEP)
+			iwl_trans->cfg = &iwl22000_2ax_cfg_qnj_hr_a0_f0;
 	}
 
 	/*
@@ -1087,45 +1099,53 @@ static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	 */
 	if (iwl_trans->hw_rev == CSR_HW_REV_TYPE_QU_C0) {
 		if (cfg == &iwl_ax101_cfg_qu_hr)
-			cfg = &iwl_ax101_cfg_qu_c0_hr_b0;
+			iwl_trans->cfg = &iwl_ax101_cfg_qu_c0_hr_b0;
 		else if (cfg == &iwl_ax201_cfg_qu_hr)
-			cfg = &iwl_ax201_cfg_qu_c0_hr_b0;
+			iwl_trans->cfg = &iwl_ax201_cfg_qu_c0_hr_b0;
 		else if (cfg == &iwl9461_2ac_cfg_qu_b0_jf_b0)
-			cfg = &iwl9461_2ac_cfg_qu_c0_jf_b0;
+			iwl_trans->cfg = &iwl9461_2ac_cfg_qu_c0_jf_b0;
 		else if (cfg == &iwl9462_2ac_cfg_qu_b0_jf_b0)
-			cfg = &iwl9462_2ac_cfg_qu_c0_jf_b0;
+			iwl_trans->cfg = &iwl9462_2ac_cfg_qu_c0_jf_b0;
 		else if (cfg == &iwl9560_2ac_cfg_qu_b0_jf_b0)
-			cfg = &iwl9560_2ac_cfg_qu_c0_jf_b0;
+			iwl_trans->cfg = &iwl9560_2ac_cfg_qu_c0_jf_b0;
 		else if (cfg == &iwl9560_2ac_160_cfg_qu_b0_jf_b0)
-			cfg = &iwl9560_2ac_160_cfg_qu_c0_jf_b0;
+			iwl_trans->cfg = &iwl9560_2ac_160_cfg_qu_c0_jf_b0;
+		else if (cfg == &killer1650s_2ax_cfg_qu_b0_hr_b0)
+			iwl_trans->cfg = &killer1650s_2ax_cfg_qu_c0_hr_b0;
+		else if (cfg == &killer1650i_2ax_cfg_qu_b0_hr_b0)
+			iwl_trans->cfg = &killer1650i_2ax_cfg_qu_c0_hr_b0;
 	}
 
 	/* same thing for QuZ... */
 	if (iwl_trans->hw_rev == CSR_HW_REV_TYPE_QUZ) {
-		if (iwl_trans->cfg == &iwl_ax101_cfg_qu_hr)
+		if (cfg == &iwl_ax101_cfg_qu_hr)
 			iwl_trans->cfg = &iwl_ax101_cfg_quz_hr;
-		else if (iwl_trans->cfg == &iwl_ax201_cfg_qu_hr)
+		else if (cfg == &iwl_ax201_cfg_qu_hr)
 			iwl_trans->cfg = &iwl_ax201_cfg_quz_hr;
-		else if (iwl_trans->cfg == &iwl9461_2ac_cfg_qu_b0_jf_b0)
+		else if (cfg == &iwl9461_2ac_cfg_qu_b0_jf_b0)
 			iwl_trans->cfg = &iwl9461_2ac_cfg_quz_a0_jf_b0_soc;
-		else if (iwl_trans->cfg == &iwl9462_2ac_cfg_qu_b0_jf_b0)
+		else if (cfg == &iwl9462_2ac_cfg_qu_b0_jf_b0)
 			iwl_trans->cfg = &iwl9462_2ac_cfg_quz_a0_jf_b0_soc;
-		else if (iwl_trans->cfg == &iwl9560_2ac_cfg_qu_b0_jf_b0)
+		else if (cfg == &iwl9560_2ac_cfg_qu_b0_jf_b0)
 			iwl_trans->cfg = &iwl9560_2ac_cfg_quz_a0_jf_b0_soc;
-		else if (iwl_trans->cfg == &iwl9560_2ac_160_cfg_qu_b0_jf_b0)
+		else if (cfg == &iwl9560_2ac_160_cfg_qu_b0_jf_b0)
 			iwl_trans->cfg = &iwl9560_2ac_160_cfg_quz_a0_jf_b0_soc;
 	}
 
 #endif
-	/* now set the real cfg we decided to use */
-	iwl_trans->cfg = cfg;
+	/*
+	 * If we didn't set the cfg yet, assume the trans is actually
+	 * a full cfg from the old tables.
+	 */
+	if (!iwl_trans->cfg)
+		iwl_trans->cfg = cfg;
 
-	if (cfg->trans.mq_rx_supported) {
-		if (WARN_ON(!cfg->num_rbds)) {
+	if (iwl_trans->trans_cfg->mq_rx_supported) {
+		if (WARN_ON(!iwl_trans->cfg->num_rbds)) {
 			ret = -EINVAL;
 			goto out_free_trans;
 		}
-		trans_pcie->num_rx_bufs = cfg->num_rbds;
+		trans_pcie->num_rx_bufs = iwl_trans->cfg->num_rbds;
 	} else {
 		trans_pcie->num_rx_bufs = RX_QUEUE_SIZE;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 2a612a60ab440e6480c77b73403dfee061f74e4b Mon Sep 17 00:00:00 2001
From: Luca Coelho <luciano.coelho@intel.com>
Date: Thu, 10 Oct 2019 16:30:11 +0300
Subject: iwlwifi: implement a new device configuration table

Add a new device table that contains information that can be checked
at runtime in order to decide which configuration to use.  This allows
us to map the full cfg independently from the tran-specific
configuration.

This is the first step in creating the new table.  Subsequent patches
will add the possibility of checking different values at runtime in
order to make the decision.

Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/cfg/9000.c   |  6 ++++
 drivers/net/wireless/intel/iwlwifi/iwl-config.h | 10 ++++++
 drivers/net/wireless/intel/iwlwifi/pcie/drv.c   | 42 +++++++++++++++++++------
 3 files changed, 49 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/9000.c b/drivers/net/wireless/intel/iwlwifi/cfg/9000.c
index 79390f880efb..e7b95842bb60 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/9000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/9000.c
@@ -171,6 +171,12 @@ static const struct iwl_tt_params iwl9000_tt_params = {
 		},							\
 	}
 
+const struct iwl_cfg_trans_params iwl9000_trans_cfg = {
+	.device_family = IWL_DEVICE_FAMILY_9000,
+	.base_params = &iwl9000_base_params,
+	.mq_rx_supported = true,
+	.rf_id = true,
+};
 
 const struct iwl_cfg iwl9160_2ac_cfg = {
 	.name = "Intel(R) Dual Band Wireless AC 9160",
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
index bbeb5fc3caad..0af1a5184d82 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
@@ -452,9 +452,19 @@ struct iwl_cfg {
 	const struct iwl_fw_mon_regs mon_smem_regs;
 };
 
+#define IWL_CFG_ANY (~0)
+
+struct iwl_dev_info {
+	u16 device;
+	u16 subdevice;
+	const struct iwl_cfg *cfg;
+};
+
 /*
  * This list declares the config structures for all devices.
  */
+extern const struct iwl_cfg_trans_params iwl9000_trans_cfg;
+
 #if IS_ENABLED(CONFIG_IWLDVM)
 extern const struct iwl_cfg iwl5300_agn_cfg;
 extern const struct iwl_cfg iwl5100_agn_cfg;
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
index 5eb528d2111f..c2cb6283bc5a 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
@@ -565,14 +565,8 @@ static const struct pci_device_id iwl_hw_card_ids[] = {
 	{IWL_PCI_DEVICE(0x06F0, 0x40A4, iwl9462_2ac_cfg_quz_a0_jf_b0_soc)},
 	{IWL_PCI_DEVICE(0x06F0, 0x4234, iwl9560_2ac_cfg_quz_a0_jf_b0_soc)},
 	{IWL_PCI_DEVICE(0x06F0, 0x42A4, iwl9462_2ac_cfg_quz_a0_jf_b0_soc)},
-	{IWL_PCI_DEVICE(0x2526, 0x0010, iwl9260_2ac_160_cfg)},
-	{IWL_PCI_DEVICE(0x2526, 0x0014, iwl9260_2ac_160_cfg)},
-	{IWL_PCI_DEVICE(0x2526, 0x0018, iwl9260_2ac_160_cfg)},
-	{IWL_PCI_DEVICE(0x2526, 0x001C, iwl9260_2ac_160_cfg)},
-	{IWL_PCI_DEVICE(0x2526, 0x0030, iwl9560_2ac_160_cfg)},
+
 	{IWL_PCI_DEVICE(0x2526, 0x0034, iwl9560_2ac_cfg)},
-	{IWL_PCI_DEVICE(0x2526, 0x0038, iwl9560_2ac_160_cfg)},
-	{IWL_PCI_DEVICE(0x2526, 0x003C, iwl9560_2ac_160_cfg)},
 	{IWL_PCI_DEVICE(0x2526, 0x0060, iwl9461_2ac_cfg_soc)},
 	{IWL_PCI_DEVICE(0x2526, 0x0064, iwl9461_2ac_cfg_soc)},
 	{IWL_PCI_DEVICE(0x2526, 0x00A0, iwl9462_2ac_cfg_soc)},
@@ -601,7 +595,6 @@ static const struct pci_device_id iwl_hw_card_ids[] = {
 	{IWL_PCI_DEVICE(0x2526, 0x4010, iwl9260_2ac_160_cfg)},
 	{IWL_PCI_DEVICE(0x2526, 0x4018, iwl9260_2ac_160_cfg)},
 	{IWL_PCI_DEVICE(0x2526, 0x401C, iwl9260_2ac_160_cfg)},
-	{IWL_PCI_DEVICE(0x2526, 0x4030, iwl9560_2ac_160_cfg)},
 	{IWL_PCI_DEVICE(0x2526, 0x4034, iwl9560_2ac_160_cfg_soc)},
 	{IWL_PCI_DEVICE(0x2526, 0x40A4, iwl9462_2ac_cfg_soc)},
 	{IWL_PCI_DEVICE(0x2526, 0x4234, iwl9560_2ac_cfg_soc)},
@@ -613,6 +606,8 @@ static const struct pci_device_id iwl_hw_card_ids[] = {
 	{IWL_PCI_DEVICE(0x2526, 0xA014, iwl9260_2ac_160_cfg)},
 	{IWL_PCI_DEVICE(0x2526, 0xE010, iwl9260_2ac_160_cfg)},
 	{IWL_PCI_DEVICE(0x2526, 0xE014, iwl9260_2ac_160_cfg)},
+	{IWL_PCI_DEVICE(0x2526, PCI_ANY_ID, iwl9000_trans_cfg)},
+
 	{IWL_PCI_DEVICE(0x271B, 0x0010, iwl9160_2ac_cfg)},
 	{IWL_PCI_DEVICE(0x271B, 0x0014, iwl9160_2ac_cfg)},
 	{IWL_PCI_DEVICE(0x271B, 0x0210, iwl9160_2ac_cfg)},
@@ -986,6 +981,22 @@ static const struct pci_device_id iwl_hw_card_ids[] = {
 };
 MODULE_DEVICE_TABLE(pci, iwl_hw_card_ids);
 
+#define IWL_DEV_INFO(_device, _subdevice, _cfg) \
+	{.device = (_device), .subdevice = (_subdevice), .cfg = &(_cfg)}
+
+static const struct iwl_dev_info iwl_dev_info_table[] = {
+#if IS_ENABLED(CONFIG_IWLMVM)
+	IWL_DEV_INFO(0x2526, 0x0010, iwl9260_2ac_160_cfg),
+	IWL_DEV_INFO(0x2526, 0x0014, iwl9260_2ac_160_cfg),
+	IWL_DEV_INFO(0x2526, 0x0018, iwl9260_2ac_160_cfg),
+	IWL_DEV_INFO(0x2526, 0x001C, iwl9260_2ac_160_cfg),
+	IWL_DEV_INFO(0x2526, 0x0030, iwl9560_2ac_160_cfg),
+	IWL_DEV_INFO(0x2526, 0x0038, iwl9560_2ac_160_cfg),
+	IWL_DEV_INFO(0x2526, 0x003C, iwl9560_2ac_160_cfg),
+	IWL_DEV_INFO(0x2526, 0x4030, iwl9560_2ac_160_cfg),
+#endif /* CONFIG_IWLMVM */
+};
+
 /* PCI registers */
 #define PCI_CFG_RETRY_TIMEOUT	0x041
 
@@ -997,7 +1008,7 @@ static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	struct iwl_trans *iwl_trans;
 	struct iwl_trans_pcie *trans_pcie;
 	unsigned long flags;
-	int ret;
+	int i, ret;
 	/*
 	 * This is needed for backwards compatibility with the old
 	 * tables, so we don't need to change all the config structs
@@ -1018,6 +1029,18 @@ static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	/* the trans_cfg should never change, so set it now */
 	iwl_trans->trans_cfg = trans;
 
+	for (i = 0; i < ARRAY_SIZE(iwl_dev_info_table); i++) {
+		const struct iwl_dev_info *dev_info = &iwl_dev_info_table[i];
+
+		if ((dev_info->device == IWL_CFG_ANY ||
+		     dev_info->device == pdev->device) &&
+		    (dev_info->subdevice == IWL_CFG_ANY ||
+		     dev_info->subdevice == pdev->subsystem_device)) {
+			iwl_trans->cfg = dev_info->cfg;
+			goto found;
+		}
+	}
+
 #if IS_ENABLED(CONFIG_IWLMVM)
 	/*
 	 * special-case 7265D, it has the same PCI IDs.
@@ -1140,6 +1163,7 @@ static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (!iwl_trans->cfg)
 		iwl_trans->cfg = cfg;
 
+found:
 	if (iwl_trans->trans_cfg->mq_rx_supported) {
 		if (WARN_ON(!iwl_trans->cfg->num_rbds)) {
 			ret = -EINVAL;
-- 
cgit v1.2.3-59-g8ed1b


From 0b295a1eb81f37dc7d4f4f2ee9ef375fb36ab5d8 Mon Sep 17 00:00:00 2001
From: Luca Coelho <luciano.coelho@intel.com>
Date: Thu, 10 Oct 2019 18:29:26 +0300
Subject: iwlwifi: add device name to device_info

We have a lot of mostly duplicated data structures that are repeated
only because the device name string is different.  To avoid this, move
the string from the cfg to the trans structure and add it
independently from the rest of the configuration to the PCI mapping
tables.

Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/cfg/9000.c    |  4 ++-
 drivers/net/wireless/intel/iwlwifi/dvm/main.c    |  4 +--
 drivers/net/wireless/intel/iwlwifi/fw/dbg.c      |  2 +-
 drivers/net/wireless/intel/iwlwifi/iwl-config.h  |  3 ++
 drivers/net/wireless/intel/iwlwifi/iwl-trans.h   |  2 ++
 drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c |  2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/nvm.c     |  2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/ops.c     |  2 +-
 drivers/net/wireless/intel/iwlwifi/pcie/drv.c    | 44 +++++++++++++-----------
 9 files changed, 38 insertions(+), 27 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/9000.c b/drivers/net/wireless/intel/iwlwifi/cfg/9000.c
index e7b95842bb60..379ea788e424 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/9000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/9000.c
@@ -190,8 +190,10 @@ const struct iwl_cfg iwl9260_2ac_cfg = {
 	IWL_DEVICE_9000,
 };
 
+const char iwl9260_160_name[] = "Intel(R) Wireless-AC 9260 160MHz";
+const char iwl9560_160_name[] = "Intel(R) Wireless-AC 9560 160MHz";
+
 const struct iwl_cfg iwl9260_2ac_160_cfg = {
-	.name = "Intel(R) Wireless-AC 9260 160MHz",
 	.fw_name_pre = IWL9260_FW_PRE,
 	IWL_DEVICE_9000,
 };
diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/main.c b/drivers/net/wireless/intel/iwlwifi/dvm/main.c
index 4f2789bb3b5b..598ee7315558 100644
--- a/drivers/net/wireless/intel/iwlwifi/dvm/main.c
+++ b/drivers/net/wireless/intel/iwlwifi/dvm/main.c
@@ -1255,7 +1255,7 @@ static struct iwl_op_mode *iwl_op_mode_dvm_start(struct iwl_trans *trans,
 	 ************************/
 	hw = iwl_alloc_all();
 	if (!hw) {
-		pr_err("%s: Cannot allocate network device\n", cfg->name);
+		pr_err("%s: Cannot allocate network device\n", trans->name);
 		goto out;
 	}
 
@@ -1390,7 +1390,7 @@ static struct iwl_op_mode *iwl_op_mode_dvm_start(struct iwl_trans *trans,
 	 * 2. Read REV register
 	 ***********************/
 	IWL_INFO(priv, "Detected %s, REV=0x%X\n",
-		priv->cfg->name, priv->trans->hw_rev);
+		priv->trans->name, priv->trans->hw_rev);
 
 	if (iwl_trans_start_hw(priv->trans))
 		goto out_free_hw;
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/dbg.c b/drivers/net/wireless/intel/iwlwifi/fw/dbg.c
index 09daf0f1f6f8..89d796821023 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/dbg.c
+++ b/drivers/net/wireless/intel/iwlwifi/fw/dbg.c
@@ -929,7 +929,7 @@ iwl_fw_error_dump_file(struct iwl_fw_runtime *fwrt,
 			cpu_to_le32(CSR_HW_REV_STEP(fwrt->trans->hw_rev));
 		memcpy(dump_info->fw_human_readable, fwrt->fw->human_readable,
 		       sizeof(dump_info->fw_human_readable));
-		strncpy(dump_info->dev_human_readable, fwrt->trans->cfg->name,
+		strncpy(dump_info->dev_human_readable, fwrt->trans->name,
 			sizeof(dump_info->dev_human_readable) - 1);
 		strncpy(dump_info->bus_human_readable, fwrt->dev->bus->name,
 			sizeof(dump_info->bus_human_readable) - 1);
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
index 0af1a5184d82..be6a2bf9ce74 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
@@ -458,12 +458,15 @@ struct iwl_dev_info {
 	u16 device;
 	u16 subdevice;
 	const struct iwl_cfg *cfg;
+	const char *name;
 };
 
 /*
  * This list declares the config structures for all devices.
  */
 extern const struct iwl_cfg_trans_params iwl9000_trans_cfg;
+extern const char iwl9260_160_name[];
+extern const char iwl9560_160_name[];
 
 #if IS_ENABLED(CONFIG_IWLDVM)
 extern const struct iwl_cfg iwl5300_agn_cfg;
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
index 58bee4442e20..ab547be4994b 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
@@ -859,6 +859,8 @@ struct iwl_trans {
 
 	enum iwl_plat_pm_mode system_pm_mode;
 
+	const char *name;
+
 	/* pointer to trans specific struct */
 	/*Ensure that this pointer will always be aligned to sizeof pointer */
 	char trans_specific[0] __aligned(sizeof(void *));
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
index aa659162a7c2..190cf15b825c 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
@@ -752,7 +752,7 @@ static ssize_t iwl_dbgfs_fw_ver_read(struct file *file, char __user *user_buf,
 	pos += scnprintf(pos, endpos - pos, "FW: %s\n",
 			 mvm->fwrt.fw->human_readable);
 	pos += scnprintf(pos, endpos - pos, "Device: %s\n",
-			 mvm->fwrt.trans->cfg->name);
+			 mvm->fwrt.trans->name);
 	pos += scnprintf(pos, endpos - pos, "Bus: %s\n",
 			 mvm->fwrt.dev->bus->name);
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c b/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c
index 945c1ea5cda8..c46c00706cd5 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c
@@ -178,7 +178,7 @@ static int iwl_nvm_read_chunk(struct iwl_mvm *mvm, u16 section,
 		} else {
 			IWL_DEBUG_EEPROM(mvm->trans->dev,
 					 "NVM access command failed with status %d (device: %s)\n",
-					 ret, mvm->cfg->name);
+					 ret, mvm->trans->name);
 			ret = -ENODATA;
 		}
 		goto exit;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index 1b07a8e8f069..dfe02440d474 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -830,7 +830,7 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 	}
 
 	IWL_INFO(mvm, "Detected %s, REV=0x%X\n",
-		 mvm->cfg->name, mvm->trans->hw_rev);
+		 mvm->trans->name, mvm->trans->hw_rev);
 
 	if (iwlwifi_mod_params.nvm_file)
 		mvm->nvm_file_name = iwlwifi_mod_params.nvm_file;
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
index c2cb6283bc5a..9d8ffbecf5be 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
@@ -592,20 +592,10 @@ static const struct pci_device_id iwl_hw_card_ids[] = {
 	{IWL_PCI_DEVICE(0x2526, 0x1610, iwl9270_2ac_cfg)},
 	{IWL_PCI_DEVICE(0x2526, 0x2030, iwl9560_2ac_160_cfg_soc)},
 	{IWL_PCI_DEVICE(0x2526, 0x2034, iwl9560_2ac_160_cfg_soc)},
-	{IWL_PCI_DEVICE(0x2526, 0x4010, iwl9260_2ac_160_cfg)},
-	{IWL_PCI_DEVICE(0x2526, 0x4018, iwl9260_2ac_160_cfg)},
-	{IWL_PCI_DEVICE(0x2526, 0x401C, iwl9260_2ac_160_cfg)},
 	{IWL_PCI_DEVICE(0x2526, 0x4034, iwl9560_2ac_160_cfg_soc)},
 	{IWL_PCI_DEVICE(0x2526, 0x40A4, iwl9462_2ac_cfg_soc)},
 	{IWL_PCI_DEVICE(0x2526, 0x4234, iwl9560_2ac_cfg_soc)},
 	{IWL_PCI_DEVICE(0x2526, 0x42A4, iwl9462_2ac_cfg_soc)},
-	{IWL_PCI_DEVICE(0x2526, 0x6010, iwl9260_2ac_160_cfg)},
-	{IWL_PCI_DEVICE(0x2526, 0x6014, iwl9260_2ac_160_cfg)},
-	{IWL_PCI_DEVICE(0x2526, 0x8014, iwl9260_2ac_160_cfg)},
-	{IWL_PCI_DEVICE(0x2526, 0x8010, iwl9260_2ac_160_cfg)},
-	{IWL_PCI_DEVICE(0x2526, 0xA014, iwl9260_2ac_160_cfg)},
-	{IWL_PCI_DEVICE(0x2526, 0xE010, iwl9260_2ac_160_cfg)},
-	{IWL_PCI_DEVICE(0x2526, 0xE014, iwl9260_2ac_160_cfg)},
 	{IWL_PCI_DEVICE(0x2526, PCI_ANY_ID, iwl9000_trans_cfg)},
 
 	{IWL_PCI_DEVICE(0x271B, 0x0010, iwl9160_2ac_cfg)},
@@ -981,19 +971,28 @@ static const struct pci_device_id iwl_hw_card_ids[] = {
 };
 MODULE_DEVICE_TABLE(pci, iwl_hw_card_ids);
 
-#define IWL_DEV_INFO(_device, _subdevice, _cfg) \
-	{.device = (_device), .subdevice = (_subdevice), .cfg = &(_cfg)}
+#define IWL_DEV_INFO(_device, _subdevice, _cfg, _name)			 \
+	{ .device = (_device), .subdevice = (_subdevice), .cfg = &(_cfg), \
+	  .name = _name }
 
 static const struct iwl_dev_info iwl_dev_info_table[] = {
 #if IS_ENABLED(CONFIG_IWLMVM)
-	IWL_DEV_INFO(0x2526, 0x0010, iwl9260_2ac_160_cfg),
-	IWL_DEV_INFO(0x2526, 0x0014, iwl9260_2ac_160_cfg),
-	IWL_DEV_INFO(0x2526, 0x0018, iwl9260_2ac_160_cfg),
-	IWL_DEV_INFO(0x2526, 0x001C, iwl9260_2ac_160_cfg),
-	IWL_DEV_INFO(0x2526, 0x0030, iwl9560_2ac_160_cfg),
-	IWL_DEV_INFO(0x2526, 0x0038, iwl9560_2ac_160_cfg),
-	IWL_DEV_INFO(0x2526, 0x003C, iwl9560_2ac_160_cfg),
-	IWL_DEV_INFO(0x2526, 0x4030, iwl9560_2ac_160_cfg),
+	IWL_DEV_INFO(0x2526, 0x0010, iwl9260_2ac_160_cfg, iwl9260_160_name),
+	IWL_DEV_INFO(0x2526, 0x0014, iwl9260_2ac_160_cfg, iwl9260_160_name),
+	IWL_DEV_INFO(0x2526, 0x0018, iwl9260_2ac_160_cfg, iwl9260_160_name),
+	IWL_DEV_INFO(0x2526, 0x001C, iwl9260_2ac_160_cfg, iwl9260_160_name),
+	IWL_DEV_INFO(0x2526, 0x6010, iwl9260_2ac_160_cfg, iwl9260_160_name),
+	IWL_DEV_INFO(0x2526, 0x6014, iwl9260_2ac_160_cfg, iwl9260_160_name),
+	IWL_DEV_INFO(0x2526, 0x8014, iwl9260_2ac_160_cfg, iwl9260_160_name),
+	IWL_DEV_INFO(0x2526, 0x8010, iwl9260_2ac_160_cfg, iwl9260_160_name),
+	IWL_DEV_INFO(0x2526, 0xA014, iwl9260_2ac_160_cfg, iwl9260_160_name),
+	IWL_DEV_INFO(0x2526, 0xE010, iwl9260_2ac_160_cfg, iwl9260_160_name),
+	IWL_DEV_INFO(0x2526, 0xE014, iwl9260_2ac_160_cfg, iwl9260_160_name),
+
+	IWL_DEV_INFO(0x2526, 0x0030, iwl9560_2ac_160_cfg, iwl9560_160_name),
+	IWL_DEV_INFO(0x2526, 0x0038, iwl9560_2ac_160_cfg, iwl9560_160_name),
+	IWL_DEV_INFO(0x2526, 0x003C, iwl9560_2ac_160_cfg, iwl9560_160_name),
+	IWL_DEV_INFO(0x2526, 0x4030, iwl9560_2ac_160_cfg, iwl9560_160_name),
 #endif /* CONFIG_IWLMVM */
 };
 
@@ -1037,6 +1036,7 @@ static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		    (dev_info->subdevice == IWL_CFG_ANY ||
 		     dev_info->subdevice == pdev->subsystem_device)) {
 			iwl_trans->cfg = dev_info->cfg;
+			iwl_trans->name = dev_info->name;
 			goto found;
 		}
 	}
@@ -1164,6 +1164,10 @@ static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		iwl_trans->cfg = cfg;
 
 found:
+	/* if we don't have a name yet, copy name from the old cfg */
+	if (!iwl_trans->name)
+		iwl_trans->name = iwl_trans->cfg->name;
+
 	if (iwl_trans->trans_cfg->mq_rx_supported) {
 		if (WARN_ON(!iwl_trans->cfg->num_rbds)) {
 			ret = -EINVAL;
-- 
cgit v1.2.3-59-g8ed1b


From 14504ac5048e1773da16e741b2e659c8b7257592 Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Tue, 19 Nov 2019 13:43:32 +0200
Subject: igc: Remove no need declaration of the igc_clean_tx_ring

We want to avoid forward-declarations of function if possible.
Rearrange the igc_clean_tx_ring function implementation.

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_main.c | 75 +++++++++++++++----------------
 1 file changed, 37 insertions(+), 38 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 25989c087aca..c156cd68532b 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -52,7 +52,6 @@ static const struct pci_device_id igc_pci_tbl[] = {
 MODULE_DEVICE_TABLE(pci, igc_pci_tbl);
 
 /* forward declaration */
-static void igc_clean_tx_ring(struct igc_ring *tx_ring);
 static int igc_sw_init(struct igc_adapter *);
 static void igc_configure(struct igc_adapter *adapter);
 static void igc_power_down_link(struct igc_adapter *adapter);
@@ -176,43 +175,6 @@ static void igc_get_hw_control(struct igc_adapter *adapter)
 	     ctrl_ext | IGC_CTRL_EXT_DRV_LOAD);
 }
 
-/**
- * igc_free_tx_resources - Free Tx Resources per Queue
- * @tx_ring: Tx descriptor ring for a specific queue
- *
- * Free all transmit software resources
- */
-void igc_free_tx_resources(struct igc_ring *tx_ring)
-{
-	igc_clean_tx_ring(tx_ring);
-
-	vfree(tx_ring->tx_buffer_info);
-	tx_ring->tx_buffer_info = NULL;
-
-	/* if not set, then don't free */
-	if (!tx_ring->desc)
-		return;
-
-	dma_free_coherent(tx_ring->dev, tx_ring->size,
-			  tx_ring->desc, tx_ring->dma);
-
-	tx_ring->desc = NULL;
-}
-
-/**
- * igc_free_all_tx_resources - Free Tx Resources for All Queues
- * @adapter: board private structure
- *
- * Free all transmit software resources
- */
-static void igc_free_all_tx_resources(struct igc_adapter *adapter)
-{
-	int i;
-
-	for (i = 0; i < adapter->num_tx_queues; i++)
-		igc_free_tx_resources(adapter->tx_ring[i]);
-}
-
 /**
  * igc_clean_tx_ring - Free Tx Buffers
  * @tx_ring: ring to be cleaned
@@ -274,6 +236,43 @@ static void igc_clean_tx_ring(struct igc_ring *tx_ring)
 	tx_ring->next_to_clean = 0;
 }
 
+/**
+ * igc_free_tx_resources - Free Tx Resources per Queue
+ * @tx_ring: Tx descriptor ring for a specific queue
+ *
+ * Free all transmit software resources
+ */
+void igc_free_tx_resources(struct igc_ring *tx_ring)
+{
+	igc_clean_tx_ring(tx_ring);
+
+	vfree(tx_ring->tx_buffer_info);
+	tx_ring->tx_buffer_info = NULL;
+
+	/* if not set, then don't free */
+	if (!tx_ring->desc)
+		return;
+
+	dma_free_coherent(tx_ring->dev, tx_ring->size,
+			  tx_ring->desc, tx_ring->dma);
+
+	tx_ring->desc = NULL;
+}
+
+/**
+ * igc_free_all_tx_resources - Free Tx Resources for All Queues
+ * @adapter: board private structure
+ *
+ * Free all transmit software resources
+ */
+static void igc_free_all_tx_resources(struct igc_adapter *adapter)
+{
+	int i;
+
+	for (i = 0; i < adapter->num_tx_queues; i++)
+		igc_free_tx_resources(adapter->tx_ring[i]);
+}
+
 /**
  * igc_clean_all_tx_rings - Free Tx Buffers for all queues
  * @adapter: board private structure
-- 
cgit v1.2.3-59-g8ed1b


From 1ab69fb357099d7bd869cf9332f0cb137640c6a4 Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Tue, 19 Nov 2019 13:43:41 +0200
Subject: igc: Remove no need declaration of the igc_power_down_link

We want to avoid forward-declarations of function if possible.
Rearrange the igc_power_down_link function implementation.

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_main.c | 21 ++++++++++-----------
 1 file changed, 10 insertions(+), 11 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index c156cd68532b..17b83638efc9 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -54,7 +54,6 @@ MODULE_DEVICE_TABLE(pci, igc_pci_tbl);
 /* forward declaration */
 static int igc_sw_init(struct igc_adapter *);
 static void igc_configure(struct igc_adapter *adapter);
-static void igc_power_down_link(struct igc_adapter *adapter);
 static void igc_set_default_mac_filter(struct igc_adapter *adapter);
 static void igc_set_rx_mode(struct net_device *netdev);
 static void igc_write_itr(struct igc_q_vector *q_vector);
@@ -76,6 +75,16 @@ enum latency_range {
 	latency_invalid = 255
 };
 
+/**
+ * igc_power_down_link - Power down the phy/serdes link
+ * @adapter: address of board private structure
+ */
+static void igc_power_down_link(struct igc_adapter *adapter)
+{
+	if (adapter->hw.phy.media_type == igc_media_type_copper)
+		igc_power_down_phy_copper_base(&adapter->hw);
+}
+
 void igc_reset(struct igc_adapter *adapter)
 {
 	struct pci_dev *pdev = adapter->pdev;
@@ -127,16 +136,6 @@ static void igc_power_up_link(struct igc_adapter *adapter)
 	igc_setup_link(&adapter->hw);
 }
 
-/**
- * igc_power_down_link - Power down the phy link
- * @adapter: address of board private structure
- */
-static void igc_power_down_link(struct igc_adapter *adapter)
-{
-	if (adapter->hw.phy.media_type == igc_media_type_copper)
-		igc_power_down_phy_copper_base(&adapter->hw);
-}
-
 /**
  * igc_release_hw_control - release control of the h/w to f/w
  * @adapter: address of board private structure
-- 
cgit v1.2.3-59-g8ed1b


From 3988d8bfd71395b838ed0a887f7b8c2392fe6082 Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Tue, 19 Nov 2019 13:43:51 +0200
Subject: igc: Remove no need declaration of the igc_set_default_mac_filter

We want to avoid forward-declarations of function if possible.
Rearrange the igc_set_default_mac_filter function implementation.

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_main.c | 91 +++++++++++++++----------------
 1 file changed, 45 insertions(+), 46 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 17b83638efc9..95a62f838668 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -54,7 +54,6 @@ MODULE_DEVICE_TABLE(pci, igc_pci_tbl);
 /* forward declaration */
 static int igc_sw_init(struct igc_adapter *);
 static void igc_configure(struct igc_adapter *adapter);
-static void igc_set_default_mac_filter(struct igc_adapter *adapter);
 static void igc_set_rx_mode(struct net_device *netdev);
 static void igc_write_itr(struct igc_q_vector *q_vector);
 static void igc_assign_vector(struct igc_q_vector *q_vector, int msix_vector);
@@ -769,6 +768,51 @@ static void igc_setup_tctl(struct igc_adapter *adapter)
 	wr32(IGC_TCTL, tctl);
 }
 
+/**
+ * igc_rar_set_index - Sync RAL[index] and RAH[index] registers with MAC table
+ * @adapter: address of board private structure
+ * @index: Index of the RAR entry which need to be synced with MAC table
+ */
+static void igc_rar_set_index(struct igc_adapter *adapter, u32 index)
+{
+	u8 *addr = adapter->mac_table[index].addr;
+	struct igc_hw *hw = &adapter->hw;
+	u32 rar_low, rar_high;
+
+	/* HW expects these to be in network order when they are plugged
+	 * into the registers which are little endian.  In order to guarantee
+	 * that ordering we need to do an leXX_to_cpup here in order to be
+	 * ready for the byteswap that occurs with writel
+	 */
+	rar_low = le32_to_cpup((__le32 *)(addr));
+	rar_high = le16_to_cpup((__le16 *)(addr + 4));
+
+	/* Indicate to hardware the Address is Valid. */
+	if (adapter->mac_table[index].state & IGC_MAC_STATE_IN_USE) {
+		if (is_valid_ether_addr(addr))
+			rar_high |= IGC_RAH_AV;
+
+		rar_high |= IGC_RAH_POOL_1 <<
+			adapter->mac_table[index].queue;
+	}
+
+	wr32(IGC_RAL(index), rar_low);
+	wrfl();
+	wr32(IGC_RAH(index), rar_high);
+	wrfl();
+}
+
+/* Set default MAC address for the PF in the first RAR entry */
+static void igc_set_default_mac_filter(struct igc_adapter *adapter)
+{
+	struct igc_mac_addr *mac_table = &adapter->mac_table[0];
+
+	ether_addr_copy(mac_table->addr, adapter->hw.mac.addr);
+	mac_table->state = IGC_MAC_STATE_DEFAULT | IGC_MAC_STATE_IN_USE;
+
+	igc_rar_set_index(adapter, 0);
+}
+
 /**
  * igc_set_mac - Change the Ethernet Address of the NIC
  * @netdev: network interface device structure
@@ -2430,51 +2474,6 @@ static void igc_configure(struct igc_adapter *adapter)
 	}
 }
 
-/**
- * igc_rar_set_index - Sync RAL[index] and RAH[index] registers with MAC table
- * @adapter: address of board private structure
- * @index: Index of the RAR entry which need to be synced with MAC table
- */
-static void igc_rar_set_index(struct igc_adapter *adapter, u32 index)
-{
-	u8 *addr = adapter->mac_table[index].addr;
-	struct igc_hw *hw = &adapter->hw;
-	u32 rar_low, rar_high;
-
-	/* HW expects these to be in network order when they are plugged
-	 * into the registers which are little endian.  In order to guarantee
-	 * that ordering we need to do an leXX_to_cpup here in order to be
-	 * ready for the byteswap that occurs with writel
-	 */
-	rar_low = le32_to_cpup((__le32 *)(addr));
-	rar_high = le16_to_cpup((__le16 *)(addr + 4));
-
-	/* Indicate to hardware the Address is Valid. */
-	if (adapter->mac_table[index].state & IGC_MAC_STATE_IN_USE) {
-		if (is_valid_ether_addr(addr))
-			rar_high |= IGC_RAH_AV;
-
-		rar_high |= IGC_RAH_POOL_1 <<
-			adapter->mac_table[index].queue;
-	}
-
-	wr32(IGC_RAL(index), rar_low);
-	wrfl();
-	wr32(IGC_RAH(index), rar_high);
-	wrfl();
-}
-
-/* Set default MAC address for the PF in the first RAR entry */
-static void igc_set_default_mac_filter(struct igc_adapter *adapter)
-{
-	struct igc_mac_addr *mac_table = &adapter->mac_table[0];
-
-	ether_addr_copy(mac_table->addr, adapter->hw.mac.addr);
-	mac_table->state = IGC_MAC_STATE_DEFAULT | IGC_MAC_STATE_IN_USE;
-
-	igc_rar_set_index(adapter, 0);
-}
-
 /* If the filter to be added and an already existing filter express
  * the same address and address type, it should be possible to only
  * override the other configurations, for example the queue to steer
-- 
cgit v1.2.3-59-g8ed1b


From 1a7c0f2efd5559408c01c22474127d95274637a8 Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Tue, 19 Nov 2019 13:44:01 +0200
Subject: igc: Remove no need declaration of the igc_configure

We want to avoid forward-declarations of function if possible.
Rearrange the igc_configure function implementation.

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_main.c | 91 +++++++++++++++----------------
 1 file changed, 45 insertions(+), 46 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 95a62f838668..1023c9226a8e 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -53,7 +53,6 @@ MODULE_DEVICE_TABLE(pci, igc_pci_tbl);
 
 /* forward declaration */
 static int igc_sw_init(struct igc_adapter *);
-static void igc_configure(struct igc_adapter *adapter);
 static void igc_set_rx_mode(struct net_device *netdev);
 static void igc_write_itr(struct igc_q_vector *q_vector);
 static void igc_assign_vector(struct igc_q_vector *q_vector, int msix_vector);
@@ -1985,6 +1984,51 @@ static bool igc_clean_tx_irq(struct igc_q_vector *q_vector, int napi_budget)
 	return !!budget;
 }
 
+static void igc_nfc_filter_restore(struct igc_adapter *adapter)
+{
+	struct igc_nfc_filter *rule;
+
+	spin_lock(&adapter->nfc_lock);
+
+	hlist_for_each_entry(rule, &adapter->nfc_filter_list, nfc_node)
+		igc_add_filter(adapter, rule);
+
+	spin_unlock(&adapter->nfc_lock);
+}
+
+/**
+ * igc_configure - configure the hardware for RX and TX
+ * @adapter: private board structure
+ */
+static void igc_configure(struct igc_adapter *adapter)
+{
+	struct net_device *netdev = adapter->netdev;
+	int i = 0;
+
+	igc_get_hw_control(adapter);
+	igc_set_rx_mode(netdev);
+
+	igc_setup_tctl(adapter);
+	igc_setup_mrqc(adapter);
+	igc_setup_rctl(adapter);
+
+	igc_nfc_filter_restore(adapter);
+	igc_configure_tx(adapter);
+	igc_configure_rx(adapter);
+
+	igc_rx_fifo_flush_base(&adapter->hw);
+
+	/* call igc_desc_unused which always leaves
+	 * at least 1 descriptor unused to make sure
+	 * next_to_use != next_to_clean
+	 */
+	for (i = 0; i < adapter->num_rx_queues; i++) {
+		struct igc_ring *ring = adapter->rx_ring[i];
+
+		igc_alloc_rx_buffers(ring, igc_desc_unused(ring));
+	}
+}
+
 /**
  * igc_up - Open the interface and prepare it to handle traffic
  * @adapter: board private structure
@@ -2206,18 +2250,6 @@ static void igc_nfc_filter_exit(struct igc_adapter *adapter)
 	spin_unlock(&adapter->nfc_lock);
 }
 
-static void igc_nfc_filter_restore(struct igc_adapter *adapter)
-{
-	struct igc_nfc_filter *rule;
-
-	spin_lock(&adapter->nfc_lock);
-
-	hlist_for_each_entry(rule, &adapter->nfc_filter_list, nfc_node)
-		igc_add_filter(adapter, rule);
-
-	spin_unlock(&adapter->nfc_lock);
-}
-
 /**
  * igc_down - Close the interface
  * @adapter: board private structure
@@ -2441,39 +2473,6 @@ igc_features_check(struct sk_buff *skb, struct net_device *dev,
 	return features;
 }
 
-/**
- * igc_configure - configure the hardware for RX and TX
- * @adapter: private board structure
- */
-static void igc_configure(struct igc_adapter *adapter)
-{
-	struct net_device *netdev = adapter->netdev;
-	int i = 0;
-
-	igc_get_hw_control(adapter);
-	igc_set_rx_mode(netdev);
-
-	igc_setup_tctl(adapter);
-	igc_setup_mrqc(adapter);
-	igc_setup_rctl(adapter);
-
-	igc_nfc_filter_restore(adapter);
-	igc_configure_tx(adapter);
-	igc_configure_rx(adapter);
-
-	igc_rx_fifo_flush_base(&adapter->hw);
-
-	/* call igc_desc_unused which always leaves
-	 * at least 1 descriptor unused to make sure
-	 * next_to_use != next_to_clean
-	 */
-	for (i = 0; i < adapter->num_rx_queues; i++) {
-		struct igc_ring *ring = adapter->rx_ring[i];
-
-		igc_alloc_rx_buffers(ring, igc_desc_unused(ring));
-	}
-}
-
 /* If the filter to be added and an already existing filter express
  * the same address and address type, it should be possible to only
  * override the other configurations, for example the queue to steer
-- 
cgit v1.2.3-59-g8ed1b


From aac8f68c5687c3b3a592ae05f66a78837430330e Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Tue, 19 Nov 2019 13:44:11 +0200
Subject: igc: Remove no need declaration of the igc_alloc_mapped_page

We want to avoid forward-declarations of function if possible.
Rearrange the igc_alloc_mapped_page function implementation.

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_main.c | 94 +++++++++++++++----------------
 1 file changed, 46 insertions(+), 48 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 1023c9226a8e..4ad06952056b 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -63,8 +63,6 @@ static void igc_free_q_vectors(struct igc_adapter *adapter);
 static void igc_irq_disable(struct igc_adapter *adapter);
 static void igc_irq_enable(struct igc_adapter *adapter);
 static void igc_configure_msix(struct igc_adapter *adapter);
-static bool igc_alloc_mapped_page(struct igc_ring *rx_ring,
-				  struct igc_rx_buffer *bi);
 
 enum latency_range {
 	lowest_latency = 0,
@@ -1606,6 +1604,52 @@ static void igc_put_rx_buffer(struct igc_ring *rx_ring,
 	rx_buffer->page = NULL;
 }
 
+static inline unsigned int igc_rx_offset(struct igc_ring *rx_ring)
+{
+	return ring_uses_build_skb(rx_ring) ? IGC_SKB_PAD : 0;
+}
+
+static bool igc_alloc_mapped_page(struct igc_ring *rx_ring,
+				  struct igc_rx_buffer *bi)
+{
+	struct page *page = bi->page;
+	dma_addr_t dma;
+
+	/* since we are recycling buffers we should seldom need to alloc */
+	if (likely(page))
+		return true;
+
+	/* alloc new page for storage */
+	page = dev_alloc_pages(igc_rx_pg_order(rx_ring));
+	if (unlikely(!page)) {
+		rx_ring->rx_stats.alloc_failed++;
+		return false;
+	}
+
+	/* map page for use */
+	dma = dma_map_page_attrs(rx_ring->dev, page, 0,
+				 igc_rx_pg_size(rx_ring),
+				 DMA_FROM_DEVICE,
+				 IGC_RX_DMA_ATTR);
+
+	/* if mapping failed free memory back to system since
+	 * there isn't much point in holding memory we can't use
+	 */
+	if (dma_mapping_error(rx_ring->dev, dma)) {
+		__free_page(page);
+
+		rx_ring->rx_stats.alloc_failed++;
+		return false;
+	}
+
+	bi->dma = dma;
+	bi->page = page;
+	bi->page_offset = igc_rx_offset(rx_ring);
+	bi->pagecnt_bias = 1;
+
+	return true;
+}
+
 /**
  * igc_alloc_rx_buffers - Replace used receive buffers; packet split
  * @rx_ring: rx descriptor ring
@@ -1767,52 +1811,6 @@ static int igc_clean_rx_irq(struct igc_q_vector *q_vector, const int budget)
 	return total_packets;
 }
 
-static inline unsigned int igc_rx_offset(struct igc_ring *rx_ring)
-{
-	return ring_uses_build_skb(rx_ring) ? IGC_SKB_PAD : 0;
-}
-
-static bool igc_alloc_mapped_page(struct igc_ring *rx_ring,
-				  struct igc_rx_buffer *bi)
-{
-	struct page *page = bi->page;
-	dma_addr_t dma;
-
-	/* since we are recycling buffers we should seldom need to alloc */
-	if (likely(page))
-		return true;
-
-	/* alloc new page for storage */
-	page = dev_alloc_pages(igc_rx_pg_order(rx_ring));
-	if (unlikely(!page)) {
-		rx_ring->rx_stats.alloc_failed++;
-		return false;
-	}
-
-	/* map page for use */
-	dma = dma_map_page_attrs(rx_ring->dev, page, 0,
-				 igc_rx_pg_size(rx_ring),
-				 DMA_FROM_DEVICE,
-				 IGC_RX_DMA_ATTR);
-
-	/* if mapping failed free memory back to system since
-	 * there isn't much point in holding memory we can't use
-	 */
-	if (dma_mapping_error(rx_ring->dev, dma)) {
-		__free_page(page);
-
-		rx_ring->rx_stats.alloc_failed++;
-		return false;
-	}
-
-	bi->dma = dma;
-	bi->page = page;
-	bi->page_offset = igc_rx_offset(rx_ring);
-	bi->pagecnt_bias = 1;
-
-	return true;
-}
-
 /**
  * igc_clean_tx_irq - Reclaim resources after transmit completes
  * @q_vector: pointer to q_vector containing needed info
-- 
cgit v1.2.3-59-g8ed1b


From 0411d368bc8a43f33f2e012b19304cc9df8f03ba Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Tue, 19 Nov 2019 13:44:19 +0200
Subject: igc: Remove no need declaration of the igc_set_interrupt_capability

We want to avoid forward-declarations of function if possible.
Rearrange the igc_set_interrupt_capability function implementation.

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_main.c | 142 +++++++++++++++---------------
 1 file changed, 70 insertions(+), 72 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 4ad06952056b..452e26202c9e 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -57,8 +57,6 @@ static void igc_set_rx_mode(struct net_device *netdev);
 static void igc_write_itr(struct igc_q_vector *q_vector);
 static void igc_assign_vector(struct igc_q_vector *q_vector, int msix_vector);
 static void igc_free_q_vector(struct igc_adapter *adapter, int v_idx);
-static void igc_set_interrupt_capability(struct igc_adapter *adapter,
-					 bool msix);
 static void igc_free_q_vectors(struct igc_adapter *adapter);
 static void igc_irq_disable(struct igc_adapter *adapter);
 static void igc_irq_enable(struct igc_adapter *adapter);
@@ -2998,6 +2996,76 @@ static void igc_reset_interrupt_capability(struct igc_adapter *adapter)
 		igc_reset_q_vector(adapter, v_idx);
 }
 
+/**
+ * igc_set_interrupt_capability - set MSI or MSI-X if supported
+ * @adapter: Pointer to adapter structure
+ * @msix: boolean value for MSI-X capability
+ *
+ * Attempt to configure interrupts using the best available
+ * capabilities of the hardware and kernel.
+ */
+static void igc_set_interrupt_capability(struct igc_adapter *adapter,
+					 bool msix)
+{
+	int numvecs, i;
+	int err;
+
+	if (!msix)
+		goto msi_only;
+	adapter->flags |= IGC_FLAG_HAS_MSIX;
+
+	/* Number of supported queues. */
+	adapter->num_rx_queues = adapter->rss_queues;
+
+	adapter->num_tx_queues = adapter->rss_queues;
+
+	/* start with one vector for every Rx queue */
+	numvecs = adapter->num_rx_queues;
+
+	/* if Tx handler is separate add 1 for every Tx queue */
+	if (!(adapter->flags & IGC_FLAG_QUEUE_PAIRS))
+		numvecs += adapter->num_tx_queues;
+
+	/* store the number of vectors reserved for queues */
+	adapter->num_q_vectors = numvecs;
+
+	/* add 1 vector for link status interrupts */
+	numvecs++;
+
+	adapter->msix_entries = kcalloc(numvecs, sizeof(struct msix_entry),
+					GFP_KERNEL);
+
+	if (!adapter->msix_entries)
+		return;
+
+	/* populate entry values */
+	for (i = 0; i < numvecs; i++)
+		adapter->msix_entries[i].entry = i;
+
+	err = pci_enable_msix_range(adapter->pdev,
+				    adapter->msix_entries,
+				    numvecs,
+				    numvecs);
+	if (err > 0)
+		return;
+
+	kfree(adapter->msix_entries);
+	adapter->msix_entries = NULL;
+
+	igc_reset_interrupt_capability(adapter);
+
+msi_only:
+	adapter->flags &= ~IGC_FLAG_HAS_MSIX;
+
+	adapter->rss_queues = 1;
+	adapter->flags |= IGC_FLAG_QUEUE_PAIRS;
+	adapter->num_rx_queues = 1;
+	adapter->num_tx_queues = 1;
+	adapter->num_q_vectors = 1;
+	if (!pci_enable_msi(adapter->pdev))
+		adapter->flags |= IGC_FLAG_HAS_MSI;
+}
+
 /**
  * igc_clear_interrupt_scheme - reset the device to a state of no interrupts
  * @adapter: Pointer to adapter structure
@@ -3630,76 +3698,6 @@ static int igc_poll(struct napi_struct *napi, int budget)
 	return min(work_done, budget - 1);
 }
 
-/**
- * igc_set_interrupt_capability - set MSI or MSI-X if supported
- * @adapter: Pointer to adapter structure
- * @msix: boolean value for MSI-X capability
- *
- * Attempt to configure interrupts using the best available
- * capabilities of the hardware and kernel.
- */
-static void igc_set_interrupt_capability(struct igc_adapter *adapter,
-					 bool msix)
-{
-	int numvecs, i;
-	int err;
-
-	if (!msix)
-		goto msi_only;
-	adapter->flags |= IGC_FLAG_HAS_MSIX;
-
-	/* Number of supported queues. */
-	adapter->num_rx_queues = adapter->rss_queues;
-
-	adapter->num_tx_queues = adapter->rss_queues;
-
-	/* start with one vector for every Rx queue */
-	numvecs = adapter->num_rx_queues;
-
-	/* if Tx handler is separate add 1 for every Tx queue */
-	if (!(adapter->flags & IGC_FLAG_QUEUE_PAIRS))
-		numvecs += adapter->num_tx_queues;
-
-	/* store the number of vectors reserved for queues */
-	adapter->num_q_vectors = numvecs;
-
-	/* add 1 vector for link status interrupts */
-	numvecs++;
-
-	adapter->msix_entries = kcalloc(numvecs, sizeof(struct msix_entry),
-					GFP_KERNEL);
-
-	if (!adapter->msix_entries)
-		return;
-
-	/* populate entry values */
-	for (i = 0; i < numvecs; i++)
-		adapter->msix_entries[i].entry = i;
-
-	err = pci_enable_msix_range(adapter->pdev,
-				    adapter->msix_entries,
-				    numvecs,
-				    numvecs);
-	if (err > 0)
-		return;
-
-	kfree(adapter->msix_entries);
-	adapter->msix_entries = NULL;
-
-	igc_reset_interrupt_capability(adapter);
-
-msi_only:
-	adapter->flags &= ~IGC_FLAG_HAS_MSIX;
-
-	adapter->rss_queues = 1;
-	adapter->flags |= IGC_FLAG_QUEUE_PAIRS;
-	adapter->num_rx_queues = 1;
-	adapter->num_tx_queues = 1;
-	adapter->num_q_vectors = 1;
-	if (!pci_enable_msi(adapter->pdev))
-		adapter->flags |= IGC_FLAG_HAS_MSI;
-}
-
 static void igc_add_ring(struct igc_ring *ring,
 			 struct igc_ring_container *head)
 {
-- 
cgit v1.2.3-59-g8ed1b


From 86a4de661b7bb27f208c6c327870bcdbbe9cb152 Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Tue, 19 Nov 2019 13:44:29 +0200
Subject: igc: Remove no need declaration of the igc_set_rx_mode

We want to avoid forward-declarations of function if possible.
Rearrange the igc_set_rx_mode function implementation.

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_main.c | 353 +++++++++++++++---------------
 1 file changed, 176 insertions(+), 177 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 452e26202c9e..13491ad7d251 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -53,7 +53,6 @@ MODULE_DEVICE_TABLE(pci, igc_pci_tbl);
 
 /* forward declaration */
 static int igc_sw_init(struct igc_adapter *);
-static void igc_set_rx_mode(struct net_device *netdev);
 static void igc_write_itr(struct igc_q_vector *q_vector);
 static void igc_assign_vector(struct igc_q_vector *q_vector, int msix_vector);
 static void igc_free_q_vector(struct igc_adapter *adapter, int v_idx);
@@ -1992,6 +1991,182 @@ static void igc_nfc_filter_restore(struct igc_adapter *adapter)
 	spin_unlock(&adapter->nfc_lock);
 }
 
+/* If the filter to be added and an already existing filter express
+ * the same address and address type, it should be possible to only
+ * override the other configurations, for example the queue to steer
+ * traffic.
+ */
+static bool igc_mac_entry_can_be_used(const struct igc_mac_addr *entry,
+				      const u8 *addr, const u8 flags)
+{
+	if (!(entry->state & IGC_MAC_STATE_IN_USE))
+		return true;
+
+	if ((entry->state & IGC_MAC_STATE_SRC_ADDR) !=
+	    (flags & IGC_MAC_STATE_SRC_ADDR))
+		return false;
+
+	if (!ether_addr_equal(addr, entry->addr))
+		return false;
+
+	return true;
+}
+
+/* Add a MAC filter for 'addr' directing matching traffic to 'queue',
+ * 'flags' is used to indicate what kind of match is made, match is by
+ * default for the destination address, if matching by source address
+ * is desired the flag IGC_MAC_STATE_SRC_ADDR can be used.
+ */
+static int igc_add_mac_filter(struct igc_adapter *adapter,
+			      const u8 *addr, const u8 queue)
+{
+	struct igc_hw *hw = &adapter->hw;
+	int rar_entries = hw->mac.rar_entry_count;
+	int i;
+
+	if (is_zero_ether_addr(addr))
+		return -EINVAL;
+
+	/* Search for the first empty entry in the MAC table.
+	 * Do not touch entries at the end of the table reserved for the VF MAC
+	 * addresses.
+	 */
+	for (i = 0; i < rar_entries; i++) {
+		if (!igc_mac_entry_can_be_used(&adapter->mac_table[i],
+					       addr, 0))
+			continue;
+
+		ether_addr_copy(adapter->mac_table[i].addr, addr);
+		adapter->mac_table[i].queue = queue;
+		adapter->mac_table[i].state |= IGC_MAC_STATE_IN_USE;
+
+		igc_rar_set_index(adapter, i);
+		return i;
+	}
+
+	return -ENOSPC;
+}
+
+/* Remove a MAC filter for 'addr' directing matching traffic to
+ * 'queue', 'flags' is used to indicate what kind of match need to be
+ * removed, match is by default for the destination address, if
+ * matching by source address is to be removed the flag
+ * IGC_MAC_STATE_SRC_ADDR can be used.
+ */
+static int igc_del_mac_filter(struct igc_adapter *adapter,
+			      const u8 *addr, const u8 queue)
+{
+	struct igc_hw *hw = &adapter->hw;
+	int rar_entries = hw->mac.rar_entry_count;
+	int i;
+
+	if (is_zero_ether_addr(addr))
+		return -EINVAL;
+
+	/* Search for matching entry in the MAC table based on given address
+	 * and queue. Do not touch entries at the end of the table reserved
+	 * for the VF MAC addresses.
+	 */
+	for (i = 0; i < rar_entries; i++) {
+		if (!(adapter->mac_table[i].state & IGC_MAC_STATE_IN_USE))
+			continue;
+		if (adapter->mac_table[i].state != 0)
+			continue;
+		if (adapter->mac_table[i].queue != queue)
+			continue;
+		if (!ether_addr_equal(adapter->mac_table[i].addr, addr))
+			continue;
+
+		/* When a filter for the default address is "deleted",
+		 * we return it to its initial configuration
+		 */
+		if (adapter->mac_table[i].state & IGC_MAC_STATE_DEFAULT) {
+			adapter->mac_table[i].state =
+				IGC_MAC_STATE_DEFAULT | IGC_MAC_STATE_IN_USE;
+			adapter->mac_table[i].queue = 0;
+		} else {
+			adapter->mac_table[i].state = 0;
+			adapter->mac_table[i].queue = 0;
+			memset(adapter->mac_table[i].addr, 0, ETH_ALEN);
+		}
+
+		igc_rar_set_index(adapter, i);
+		return 0;
+	}
+
+	return -ENOENT;
+}
+
+static int igc_uc_sync(struct net_device *netdev, const unsigned char *addr)
+{
+	struct igc_adapter *adapter = netdev_priv(netdev);
+	int ret;
+
+	ret = igc_add_mac_filter(adapter, addr, adapter->num_rx_queues);
+
+	return min_t(int, ret, 0);
+}
+
+static int igc_uc_unsync(struct net_device *netdev, const unsigned char *addr)
+{
+	struct igc_adapter *adapter = netdev_priv(netdev);
+
+	igc_del_mac_filter(adapter, addr, adapter->num_rx_queues);
+
+	return 0;
+}
+
+/**
+ * igc_set_rx_mode - Secondary Unicast, Multicast and Promiscuous mode set
+ * @netdev: network interface device structure
+ *
+ * The set_rx_mode entry point is called whenever the unicast or multicast
+ * address lists or the network interface flags are updated.  This routine is
+ * responsible for configuring the hardware for proper unicast, multicast,
+ * promiscuous mode, and all-multi behavior.
+ */
+static void igc_set_rx_mode(struct net_device *netdev)
+{
+	struct igc_adapter *adapter = netdev_priv(netdev);
+	struct igc_hw *hw = &adapter->hw;
+	u32 rctl = 0, rlpml = MAX_JUMBO_FRAME_SIZE;
+	int count;
+
+	/* Check for Promiscuous and All Multicast modes */
+	if (netdev->flags & IFF_PROMISC) {
+		rctl |= IGC_RCTL_UPE | IGC_RCTL_MPE;
+	} else {
+		if (netdev->flags & IFF_ALLMULTI) {
+			rctl |= IGC_RCTL_MPE;
+		} else {
+			/* Write addresses to the MTA, if the attempt fails
+			 * then we should just turn on promiscuous mode so
+			 * that we can at least receive multicast traffic
+			 */
+			count = igc_write_mc_addr_list(netdev);
+			if (count < 0)
+				rctl |= IGC_RCTL_MPE;
+		}
+	}
+
+	/* Write addresses to available RAR registers, if there is not
+	 * sufficient space to store all the addresses then enable
+	 * unicast promiscuous mode
+	 */
+	if (__dev_uc_sync(netdev, igc_uc_sync, igc_uc_unsync))
+		rctl |= IGC_RCTL_UPE;
+
+	/* update state of unicast and multicast */
+	rctl |= rd32(IGC_RCTL) & ~(IGC_RCTL_UPE | IGC_RCTL_MPE);
+	wr32(IGC_RCTL, rctl);
+
+#if (PAGE_SIZE < 8192)
+	if (adapter->max_frame_size <= IGC_MAX_FRAME_BUILD_SKB)
+		rlpml = IGC_MAX_FRAME_BUILD_SKB;
+#endif
+	wr32(IGC_RLPML, rlpml);
+}
+
 /**
  * igc_configure - configure the hardware for RX and TX
  * @adapter: private board structure
@@ -2469,27 +2644,6 @@ igc_features_check(struct sk_buff *skb, struct net_device *dev,
 	return features;
 }
 
-/* If the filter to be added and an already existing filter express
- * the same address and address type, it should be possible to only
- * override the other configurations, for example the queue to steer
- * traffic.
- */
-static bool igc_mac_entry_can_be_used(const struct igc_mac_addr *entry,
-				      const u8 *addr, const u8 flags)
-{
-	if (!(entry->state & IGC_MAC_STATE_IN_USE))
-		return true;
-
-	if ((entry->state & IGC_MAC_STATE_SRC_ADDR) !=
-	    (flags & IGC_MAC_STATE_SRC_ADDR))
-		return false;
-
-	if (!ether_addr_equal(addr, entry->addr))
-		return false;
-
-	return true;
-}
-
 /* Add a MAC filter for 'addr' directing matching traffic to 'queue',
  * 'flags' is used to indicate what kind of match is made, match is by
  * default for the destination address, if matching by source address
@@ -2590,161 +2744,6 @@ int igc_del_mac_steering_filter(struct igc_adapter *adapter,
 					IGC_MAC_STATE_QUEUE_STEERING | flags);
 }
 
-/* Add a MAC filter for 'addr' directing matching traffic to 'queue',
- * 'flags' is used to indicate what kind of match is made, match is by
- * default for the destination address, if matching by source address
- * is desired the flag IGC_MAC_STATE_SRC_ADDR can be used.
- */
-static int igc_add_mac_filter(struct igc_adapter *adapter,
-			      const u8 *addr, const u8 queue)
-{
-	struct igc_hw *hw = &adapter->hw;
-	int rar_entries = hw->mac.rar_entry_count;
-	int i;
-
-	if (is_zero_ether_addr(addr))
-		return -EINVAL;
-
-	/* Search for the first empty entry in the MAC table.
-	 * Do not touch entries at the end of the table reserved for the VF MAC
-	 * addresses.
-	 */
-	for (i = 0; i < rar_entries; i++) {
-		if (!igc_mac_entry_can_be_used(&adapter->mac_table[i],
-					       addr, 0))
-			continue;
-
-		ether_addr_copy(adapter->mac_table[i].addr, addr);
-		adapter->mac_table[i].queue = queue;
-		adapter->mac_table[i].state |= IGC_MAC_STATE_IN_USE;
-
-		igc_rar_set_index(adapter, i);
-		return i;
-	}
-
-	return -ENOSPC;
-}
-
-/* Remove a MAC filter for 'addr' directing matching traffic to
- * 'queue', 'flags' is used to indicate what kind of match need to be
- * removed, match is by default for the destination address, if
- * matching by source address is to be removed the flag
- * IGC_MAC_STATE_SRC_ADDR can be used.
- */
-static int igc_del_mac_filter(struct igc_adapter *adapter,
-			      const u8 *addr, const u8 queue)
-{
-	struct igc_hw *hw = &adapter->hw;
-	int rar_entries = hw->mac.rar_entry_count;
-	int i;
-
-	if (is_zero_ether_addr(addr))
-		return -EINVAL;
-
-	/* Search for matching entry in the MAC table based on given address
-	 * and queue. Do not touch entries at the end of the table reserved
-	 * for the VF MAC addresses.
-	 */
-	for (i = 0; i < rar_entries; i++) {
-		if (!(adapter->mac_table[i].state & IGC_MAC_STATE_IN_USE))
-			continue;
-		if (adapter->mac_table[i].state != 0)
-			continue;
-		if (adapter->mac_table[i].queue != queue)
-			continue;
-		if (!ether_addr_equal(adapter->mac_table[i].addr, addr))
-			continue;
-
-		/* When a filter for the default address is "deleted",
-		 * we return it to its initial configuration
-		 */
-		if (adapter->mac_table[i].state & IGC_MAC_STATE_DEFAULT) {
-			adapter->mac_table[i].state =
-				IGC_MAC_STATE_DEFAULT | IGC_MAC_STATE_IN_USE;
-			adapter->mac_table[i].queue = 0;
-		} else {
-			adapter->mac_table[i].state = 0;
-			adapter->mac_table[i].queue = 0;
-			memset(adapter->mac_table[i].addr, 0, ETH_ALEN);
-		}
-
-		igc_rar_set_index(adapter, i);
-		return 0;
-	}
-
-	return -ENOENT;
-}
-
-static int igc_uc_sync(struct net_device *netdev, const unsigned char *addr)
-{
-	struct igc_adapter *adapter = netdev_priv(netdev);
-	int ret;
-
-	ret = igc_add_mac_filter(adapter, addr, adapter->num_rx_queues);
-
-	return min_t(int, ret, 0);
-}
-
-static int igc_uc_unsync(struct net_device *netdev, const unsigned char *addr)
-{
-	struct igc_adapter *adapter = netdev_priv(netdev);
-
-	igc_del_mac_filter(adapter, addr, adapter->num_rx_queues);
-
-	return 0;
-}
-
-/**
- * igc_set_rx_mode - Secondary Unicast, Multicast and Promiscuous mode set
- * @netdev: network interface device structure
- *
- * The set_rx_mode entry point is called whenever the unicast or multicast
- * address lists or the network interface flags are updated.  This routine is
- * responsible for configuring the hardware for proper unicast, multicast,
- * promiscuous mode, and all-multi behavior.
- */
-static void igc_set_rx_mode(struct net_device *netdev)
-{
-	struct igc_adapter *adapter = netdev_priv(netdev);
-	struct igc_hw *hw = &adapter->hw;
-	u32 rctl = 0, rlpml = MAX_JUMBO_FRAME_SIZE;
-	int count;
-
-	/* Check for Promiscuous and All Multicast modes */
-	if (netdev->flags & IFF_PROMISC) {
-		rctl |= IGC_RCTL_UPE | IGC_RCTL_MPE;
-	} else {
-		if (netdev->flags & IFF_ALLMULTI) {
-			rctl |= IGC_RCTL_MPE;
-		} else {
-			/* Write addresses to the MTA, if the attempt fails
-			 * then we should just turn on promiscuous mode so
-			 * that we can at least receive multicast traffic
-			 */
-			count = igc_write_mc_addr_list(netdev);
-			if (count < 0)
-				rctl |= IGC_RCTL_MPE;
-		}
-	}
-
-	/* Write addresses to available RAR registers, if there is not
-	 * sufficient space to store all the addresses then enable
-	 * unicast promiscuous mode
-	 */
-	if (__dev_uc_sync(netdev, igc_uc_sync, igc_uc_unsync))
-		rctl |= IGC_RCTL_UPE;
-
-	/* update state of unicast and multicast */
-	rctl |= rd32(IGC_RCTL) & ~(IGC_RCTL_UPE | IGC_RCTL_MPE);
-	wr32(IGC_RCTL, rctl);
-
-#if (PAGE_SIZE < 8192)
-	if (adapter->max_frame_size <= IGC_MAX_FRAME_BUILD_SKB)
-		rlpml = IGC_MAX_FRAME_BUILD_SKB;
-#endif
-	wr32(IGC_RLPML, rlpml);
-}
-
 /**
  * igc_msix_other - msix other interrupt handler
  * @irq: interrupt number
-- 
cgit v1.2.3-59-g8ed1b


From a146ea022f1d02df0116fae9d5d334b75f022a24 Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Tue, 19 Nov 2019 13:44:38 +0200
Subject: igc: Remove no need declaration of the igc_configure_msix

We want to avoid forward-declarations of function if possible.
Rearrange the igc_configure_msix function implementation.

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_main.c | 89 +++++++++++++++----------------
 1 file changed, 44 insertions(+), 45 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 13491ad7d251..6e35ac5bd2fc 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -59,7 +59,6 @@ static void igc_free_q_vector(struct igc_adapter *adapter, int v_idx);
 static void igc_free_q_vectors(struct igc_adapter *adapter);
 static void igc_irq_disable(struct igc_adapter *adapter);
 static void igc_irq_enable(struct igc_adapter *adapter);
-static void igc_configure_msix(struct igc_adapter *adapter);
 
 enum latency_range {
 	lowest_latency = 0,
@@ -2200,6 +2199,50 @@ static void igc_configure(struct igc_adapter *adapter)
 	}
 }
 
+/**
+ * igc_configure_msix - Configure MSI-X hardware
+ * @adapter: Pointer to adapter structure
+ *
+ * igc_configure_msix sets up the hardware to properly
+ * generate MSI-X interrupts.
+ */
+static void igc_configure_msix(struct igc_adapter *adapter)
+{
+	struct igc_hw *hw = &adapter->hw;
+	int i, vector = 0;
+	u32 tmp;
+
+	adapter->eims_enable_mask = 0;
+
+	/* set vector for other causes, i.e. link changes */
+	switch (hw->mac.type) {
+	case igc_i225:
+		/* Turn on MSI-X capability first, or our settings
+		 * won't stick.  And it will take days to debug.
+		 */
+		wr32(IGC_GPIE, IGC_GPIE_MSIX_MODE |
+		     IGC_GPIE_PBA | IGC_GPIE_EIAME |
+		     IGC_GPIE_NSICR);
+
+		/* enable msix_other interrupt */
+		adapter->eims_other = BIT(vector);
+		tmp = (vector++ | IGC_IVAR_VALID) << 8;
+
+		wr32(IGC_IVAR_MISC, tmp);
+		break;
+	default:
+		/* do nothing, since nothing else supports MSI-X */
+		break;
+	} /* switch (hw->mac.type) */
+
+	adapter->eims_enable_mask |= adapter->eims_other;
+
+	for (i = 0; i < adapter->num_q_vectors; i++)
+		igc_assign_vector(adapter->q_vector[i], vector++);
+
+	wrfl();
+}
+
 /**
  * igc_up - Open the interface and prepare it to handle traffic
  * @adapter: board private structure
@@ -2837,50 +2880,6 @@ static void igc_assign_vector(struct igc_q_vector *q_vector, int msix_vector)
 	q_vector->set_itr = 1;
 }
 
-/**
- * igc_configure_msix - Configure MSI-X hardware
- * @adapter: Pointer to adapter structure
- *
- * igc_configure_msix sets up the hardware to properly
- * generate MSI-X interrupts.
- */
-static void igc_configure_msix(struct igc_adapter *adapter)
-{
-	struct igc_hw *hw = &adapter->hw;
-	int i, vector = 0;
-	u32 tmp;
-
-	adapter->eims_enable_mask = 0;
-
-	/* set vector for other causes, i.e. link changes */
-	switch (hw->mac.type) {
-	case igc_i225:
-		/* Turn on MSI-X capability first, or our settings
-		 * won't stick.  And it will take days to debug.
-		 */
-		wr32(IGC_GPIE, IGC_GPIE_MSIX_MODE |
-		     IGC_GPIE_PBA | IGC_GPIE_EIAME |
-		     IGC_GPIE_NSICR);
-
-		/* enable msix_other interrupt */
-		adapter->eims_other = BIT(vector);
-		tmp = (vector++ | IGC_IVAR_VALID) << 8;
-
-		wr32(IGC_IVAR_MISC, tmp);
-		break;
-	default:
-		/* do nothing, since nothing else supports MSI-X */
-		break;
-	} /* switch (hw->mac.type) */
-
-	adapter->eims_enable_mask |= adapter->eims_other;
-
-	for (i = 0; i < adapter->num_q_vectors; i++)
-		igc_assign_vector(adapter->q_vector[i], vector++);
-
-	wrfl();
-}
-
 static irqreturn_t igc_msix_ring(int irq, void *data)
 {
 	struct igc_q_vector *q_vector = data;
-- 
cgit v1.2.3-59-g8ed1b


From fccf939e2709bf82fd58df73061a63e5016eecbe Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Tue, 19 Nov 2019 13:44:47 +0200
Subject: igc: Remove no need declaration of the igc_irq_enable

We want to avoid forward-declarations of function if possible.
Rearrange the igc_irq_enable function implementation.

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_main.c | 47 +++++++++++++++----------------
 1 file changed, 23 insertions(+), 24 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 6e35ac5bd2fc..9597c5753d6b 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -58,7 +58,6 @@ static void igc_assign_vector(struct igc_q_vector *q_vector, int msix_vector);
 static void igc_free_q_vector(struct igc_adapter *adapter, int v_idx);
 static void igc_free_q_vectors(struct igc_adapter *adapter);
 static void igc_irq_disable(struct igc_adapter *adapter);
-static void igc_irq_enable(struct igc_adapter *adapter);
 
 enum latency_range {
 	lowest_latency = 0,
@@ -2243,6 +2242,29 @@ static void igc_configure_msix(struct igc_adapter *adapter)
 	wrfl();
 }
 
+/**
+ * igc_irq_enable - Enable default interrupt generation settings
+ * @adapter: board private structure
+ */
+static void igc_irq_enable(struct igc_adapter *adapter)
+{
+	struct igc_hw *hw = &adapter->hw;
+
+	if (adapter->msix_entries) {
+		u32 ims = IGC_IMS_LSC | IGC_IMS_DOUTSYNC | IGC_IMS_DRSTA;
+		u32 regval = rd32(IGC_EIAC);
+
+		wr32(IGC_EIAC, regval | adapter->eims_enable_mask);
+		regval = rd32(IGC_EIAM);
+		wr32(IGC_EIAM, regval | adapter->eims_enable_mask);
+		wr32(IGC_EIMS, adapter->eims_enable_mask);
+		wr32(IGC_IMS, ims);
+	} else {
+		wr32(IGC_IMS, IMS_ENABLE_MASK | IGC_IMS_DRSTA);
+		wr32(IGC_IAM, IMS_ENABLE_MASK | IGC_IMS_DRSTA);
+	}
+}
+
 /**
  * igc_up - Open the interface and prepare it to handle traffic
  * @adapter: board private structure
@@ -3972,29 +3994,6 @@ static void igc_irq_disable(struct igc_adapter *adapter)
 	}
 }
 
-/**
- * igc_irq_enable - Enable default interrupt generation settings
- * @adapter: board private structure
- */
-static void igc_irq_enable(struct igc_adapter *adapter)
-{
-	struct igc_hw *hw = &adapter->hw;
-
-	if (adapter->msix_entries) {
-		u32 ims = IGC_IMS_LSC | IGC_IMS_DOUTSYNC | IGC_IMS_DRSTA;
-		u32 regval = rd32(IGC_EIAC);
-
-		wr32(IGC_EIAC, regval | adapter->eims_enable_mask);
-		regval = rd32(IGC_EIAM);
-		wr32(IGC_EIAM, regval | adapter->eims_enable_mask);
-		wr32(IGC_EIMS, adapter->eims_enable_mask);
-		wr32(IGC_IMS, ims);
-	} else {
-		wr32(IGC_IMS, IMS_ENABLE_MASK | IGC_IMS_DRSTA);
-		wr32(IGC_IAM, IMS_ENABLE_MASK | IGC_IMS_DRSTA);
-	}
-}
-
 /**
  * igc_request_irq - initialize interrupts
  * @adapter: Pointer to adapter structure
-- 
cgit v1.2.3-59-g8ed1b


From 35f9a78ac93361c2992719f9e00d60c158a32874 Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Tue, 19 Nov 2019 13:44:58 +0200
Subject: igc: Remove no need declaration of the igc_irq_disable

We want to avoid forward-declarations of function if possible.
Rearrange the igc_irq_disable function implementation.

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_main.c | 67 +++++++++++++++----------------
 1 file changed, 33 insertions(+), 34 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 9597c5753d6b..4cfb9856ee13 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -57,7 +57,6 @@ static void igc_write_itr(struct igc_q_vector *q_vector);
 static void igc_assign_vector(struct igc_q_vector *q_vector, int msix_vector);
 static void igc_free_q_vector(struct igc_adapter *adapter, int v_idx);
 static void igc_free_q_vectors(struct igc_adapter *adapter);
-static void igc_irq_disable(struct igc_adapter *adapter);
 
 enum latency_range {
 	lowest_latency = 0,
@@ -2265,6 +2264,39 @@ static void igc_irq_enable(struct igc_adapter *adapter)
 	}
 }
 
+/**
+ * igc_irq_disable - Mask off interrupt generation on the NIC
+ * @adapter: board private structure
+ */
+static void igc_irq_disable(struct igc_adapter *adapter)
+{
+	struct igc_hw *hw = &adapter->hw;
+
+	if (adapter->msix_entries) {
+		u32 regval = rd32(IGC_EIAM);
+
+		wr32(IGC_EIAM, regval & ~adapter->eims_enable_mask);
+		wr32(IGC_EIMC, adapter->eims_enable_mask);
+		regval = rd32(IGC_EIAC);
+		wr32(IGC_EIAC, regval & ~adapter->eims_enable_mask);
+	}
+
+	wr32(IGC_IAM, 0);
+	wr32(IGC_IMC, ~0);
+	wrfl();
+
+	if (adapter->msix_entries) {
+		int vector = 0, i;
+
+		synchronize_irq(adapter->msix_entries[vector++].vector);
+
+		for (i = 0; i < adapter->num_q_vectors; i++)
+			synchronize_irq(adapter->msix_entries[vector++].vector);
+	} else {
+		synchronize_irq(adapter->pdev->irq);
+	}
+}
+
 /**
  * igc_up - Open the interface and prepare it to handle traffic
  * @adapter: board private structure
@@ -3961,39 +3993,6 @@ static void igc_free_irq(struct igc_adapter *adapter)
 	}
 }
 
-/**
- * igc_irq_disable - Mask off interrupt generation on the NIC
- * @adapter: board private structure
- */
-static void igc_irq_disable(struct igc_adapter *adapter)
-{
-	struct igc_hw *hw = &adapter->hw;
-
-	if (adapter->msix_entries) {
-		u32 regval = rd32(IGC_EIAM);
-
-		wr32(IGC_EIAM, regval & ~adapter->eims_enable_mask);
-		wr32(IGC_EIMC, adapter->eims_enable_mask);
-		regval = rd32(IGC_EIAC);
-		wr32(IGC_EIAC, regval & ~adapter->eims_enable_mask);
-	}
-
-	wr32(IGC_IAM, 0);
-	wr32(IGC_IMC, ~0);
-	wrfl();
-
-	if (adapter->msix_entries) {
-		int vector = 0, i;
-
-		synchronize_irq(adapter->msix_entries[vector++].vector);
-
-		for (i = 0; i < adapter->num_q_vectors; i++)
-			synchronize_irq(adapter->msix_entries[vector++].vector);
-	} else {
-		synchronize_irq(adapter->pdev->irq);
-	}
-}
-
 /**
  * igc_request_irq - initialize interrupts
  * @adapter: Pointer to adapter structure
-- 
cgit v1.2.3-59-g8ed1b


From a8c4873b566bcbf024988334bcbfd0a31667064f Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Tue, 19 Nov 2019 13:45:07 +0200
Subject: igc: Remove no need declaration of the igc_free_q_vectors

We want to avoid forward-declarations of function if possible.
Rearrange the igc_free_q_vectors function implementation.

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_main.c | 27 +++++++++++++--------------
 1 file changed, 13 insertions(+), 14 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 4cfb9856ee13..58347135b0fc 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -56,7 +56,6 @@ static int igc_sw_init(struct igc_adapter *);
 static void igc_write_itr(struct igc_q_vector *q_vector);
 static void igc_assign_vector(struct igc_q_vector *q_vector, int msix_vector);
 static void igc_free_q_vector(struct igc_adapter *adapter, int v_idx);
-static void igc_free_q_vectors(struct igc_adapter *adapter);
 
 enum latency_range {
 	lowest_latency = 0,
@@ -3118,19 +3117,6 @@ msi_only:
 		adapter->flags |= IGC_FLAG_HAS_MSI;
 }
 
-/**
- * igc_clear_interrupt_scheme - reset the device to a state of no interrupts
- * @adapter: Pointer to adapter structure
- *
- * This function resets the device so that it has 0 rx queues, tx queues, and
- * MSI-X interrupts allocated.
- */
-static void igc_clear_interrupt_scheme(struct igc_adapter *adapter)
-{
-	igc_free_q_vectors(adapter);
-	igc_reset_interrupt_capability(adapter);
-}
-
 /**
  * igc_free_q_vectors - Free memory allocated for interrupt vectors
  * @adapter: board private structure to initialize
@@ -3153,6 +3139,19 @@ static void igc_free_q_vectors(struct igc_adapter *adapter)
 	}
 }
 
+/**
+ * igc_clear_interrupt_scheme - reset the device to a state of no interrupts
+ * @adapter: Pointer to adapter structure
+ *
+ * This function resets the device so that it has 0 rx queues, tx queues, and
+ * MSI-X interrupts allocated.
+ */
+static void igc_clear_interrupt_scheme(struct igc_adapter *adapter)
+{
+	igc_free_q_vectors(adapter);
+	igc_reset_interrupt_capability(adapter);
+}
+
 /**
  * igc_free_q_vector - Free memory allocated for specific interrupt vector
  * @adapter: board private structure to initialize
-- 
cgit v1.2.3-59-g8ed1b


From f7bcca5d9a8cf087b3f202ad45907b02437faf77 Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Tue, 19 Nov 2019 13:45:16 +0200
Subject: igc: Remove no need declaration of the igc_free_q_vector

We want to avoid forward-declarations of function if possible.
Rearrange the igc_free_q_vector function implementation.

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_main.c | 41 +++++++++++++++----------------
 1 file changed, 20 insertions(+), 21 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 58347135b0fc..368e537d1be2 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -55,7 +55,6 @@ MODULE_DEVICE_TABLE(pci, igc_pci_tbl);
 static int igc_sw_init(struct igc_adapter *);
 static void igc_write_itr(struct igc_q_vector *q_vector);
 static void igc_assign_vector(struct igc_q_vector *q_vector, int msix_vector);
-static void igc_free_q_vector(struct igc_adapter *adapter, int v_idx);
 
 enum latency_range {
 	lowest_latency = 0,
@@ -3117,6 +3116,26 @@ msi_only:
 		adapter->flags |= IGC_FLAG_HAS_MSI;
 }
 
+/**
+ * igc_free_q_vector - Free memory allocated for specific interrupt vector
+ * @adapter: board private structure to initialize
+ * @v_idx: Index of vector to be freed
+ *
+ * This function frees the memory allocated to the q_vector.
+ */
+static void igc_free_q_vector(struct igc_adapter *adapter, int v_idx)
+{
+	struct igc_q_vector *q_vector = adapter->q_vector[v_idx];
+
+	adapter->q_vector[v_idx] = NULL;
+
+	/* igc_get_stats64() might access the rings on this vector,
+	 * we must wait a grace period before freeing it.
+	 */
+	if (q_vector)
+		kfree_rcu(q_vector, rcu);
+}
+
 /**
  * igc_free_q_vectors - Free memory allocated for interrupt vectors
  * @adapter: board private structure to initialize
@@ -3152,26 +3171,6 @@ static void igc_clear_interrupt_scheme(struct igc_adapter *adapter)
 	igc_reset_interrupt_capability(adapter);
 }
 
-/**
- * igc_free_q_vector - Free memory allocated for specific interrupt vector
- * @adapter: board private structure to initialize
- * @v_idx: Index of vector to be freed
- *
- * This function frees the memory allocated to the q_vector.
- */
-static void igc_free_q_vector(struct igc_adapter *adapter, int v_idx)
-{
-	struct igc_q_vector *q_vector = adapter->q_vector[v_idx];
-
-	adapter->q_vector[v_idx] = NULL;
-
-	/* igc_get_stats64() might access the rings on this vector,
-	 * we must wait a grace period before freeing it.
-	 */
-	if (q_vector)
-		kfree_rcu(q_vector, rcu);
-}
-
 /* Need to wait a few seconds after link up to get diagnostic information from
  * the phy
  */
-- 
cgit v1.2.3-59-g8ed1b


From f817fa0555c8885060801b645885bcaf44c76b35 Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Tue, 19 Nov 2019 13:45:23 +0200
Subject: igc: Remove no need declaration of the igc_assign_vector

We want to avoid forward-declarations of function if possible.
Rearrange the igc_assign_vector function implementation.

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_main.c | 123 +++++++++++++++---------------
 1 file changed, 61 insertions(+), 62 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 368e537d1be2..4fb40c6aa583 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -54,7 +54,6 @@ MODULE_DEVICE_TABLE(pci, igc_pci_tbl);
 /* forward declaration */
 static int igc_sw_init(struct igc_adapter *);
 static void igc_write_itr(struct igc_q_vector *q_vector);
-static void igc_assign_vector(struct igc_q_vector *q_vector, int msix_vector);
 
 enum latency_range {
 	lowest_latency = 0,
@@ -2195,6 +2194,67 @@ static void igc_configure(struct igc_adapter *adapter)
 	}
 }
 
+/**
+ * igc_write_ivar - configure ivar for given MSI-X vector
+ * @hw: pointer to the HW structure
+ * @msix_vector: vector number we are allocating to a given ring
+ * @index: row index of IVAR register to write within IVAR table
+ * @offset: column offset of in IVAR, should be multiple of 8
+ *
+ * The IVAR table consists of 2 columns,
+ * each containing an cause allocation for an Rx and Tx ring, and a
+ * variable number of rows depending on the number of queues supported.
+ */
+static void igc_write_ivar(struct igc_hw *hw, int msix_vector,
+			   int index, int offset)
+{
+	u32 ivar = array_rd32(IGC_IVAR0, index);
+
+	/* clear any bits that are currently set */
+	ivar &= ~((u32)0xFF << offset);
+
+	/* write vector and valid bit */
+	ivar |= (msix_vector | IGC_IVAR_VALID) << offset;
+
+	array_wr32(IGC_IVAR0, index, ivar);
+}
+
+static void igc_assign_vector(struct igc_q_vector *q_vector, int msix_vector)
+{
+	struct igc_adapter *adapter = q_vector->adapter;
+	struct igc_hw *hw = &adapter->hw;
+	int rx_queue = IGC_N0_QUEUE;
+	int tx_queue = IGC_N0_QUEUE;
+
+	if (q_vector->rx.ring)
+		rx_queue = q_vector->rx.ring->reg_idx;
+	if (q_vector->tx.ring)
+		tx_queue = q_vector->tx.ring->reg_idx;
+
+	switch (hw->mac.type) {
+	case igc_i225:
+		if (rx_queue > IGC_N0_QUEUE)
+			igc_write_ivar(hw, msix_vector,
+				       rx_queue >> 1,
+				       (rx_queue & 0x1) << 4);
+		if (tx_queue > IGC_N0_QUEUE)
+			igc_write_ivar(hw, msix_vector,
+				       tx_queue >> 1,
+				       ((tx_queue & 0x1) << 4) + 8);
+		q_vector->eims_value = BIT(msix_vector);
+		break;
+	default:
+		WARN_ONCE(hw->mac.type != igc_i225, "Wrong MAC type\n");
+		break;
+	}
+
+	/* add q_vector eims value to global eims_enable_mask */
+	adapter->eims_enable_mask |= q_vector->eims_value;
+
+	/* configure q_vector to set itr on first interrupt */
+	q_vector->set_itr = 1;
+}
+
 /**
  * igc_configure_msix - Configure MSI-X hardware
  * @adapter: Pointer to adapter structure
@@ -2871,67 +2931,6 @@ static irqreturn_t igc_msix_other(int irq, void *data)
 	return IRQ_HANDLED;
 }
 
-/**
- * igc_write_ivar - configure ivar for given MSI-X vector
- * @hw: pointer to the HW structure
- * @msix_vector: vector number we are allocating to a given ring
- * @index: row index of IVAR register to write within IVAR table
- * @offset: column offset of in IVAR, should be multiple of 8
- *
- * The IVAR table consists of 2 columns,
- * each containing an cause allocation for an Rx and Tx ring, and a
- * variable number of rows depending on the number of queues supported.
- */
-static void igc_write_ivar(struct igc_hw *hw, int msix_vector,
-			   int index, int offset)
-{
-	u32 ivar = array_rd32(IGC_IVAR0, index);
-
-	/* clear any bits that are currently set */
-	ivar &= ~((u32)0xFF << offset);
-
-	/* write vector and valid bit */
-	ivar |= (msix_vector | IGC_IVAR_VALID) << offset;
-
-	array_wr32(IGC_IVAR0, index, ivar);
-}
-
-static void igc_assign_vector(struct igc_q_vector *q_vector, int msix_vector)
-{
-	struct igc_adapter *adapter = q_vector->adapter;
-	struct igc_hw *hw = &adapter->hw;
-	int rx_queue = IGC_N0_QUEUE;
-	int tx_queue = IGC_N0_QUEUE;
-
-	if (q_vector->rx.ring)
-		rx_queue = q_vector->rx.ring->reg_idx;
-	if (q_vector->tx.ring)
-		tx_queue = q_vector->tx.ring->reg_idx;
-
-	switch (hw->mac.type) {
-	case igc_i225:
-		if (rx_queue > IGC_N0_QUEUE)
-			igc_write_ivar(hw, msix_vector,
-				       rx_queue >> 1,
-				       (rx_queue & 0x1) << 4);
-		if (tx_queue > IGC_N0_QUEUE)
-			igc_write_ivar(hw, msix_vector,
-				       tx_queue >> 1,
-				       ((tx_queue & 0x1) << 4) + 8);
-		q_vector->eims_value = BIT(msix_vector);
-		break;
-	default:
-		WARN_ONCE(hw->mac.type != igc_i225, "Wrong MAC type\n");
-		break;
-	}
-
-	/* add q_vector eims value to global eims_enable_mask */
-	adapter->eims_enable_mask |= q_vector->eims_value;
-
-	/* configure q_vector to set itr on first interrupt */
-	q_vector->set_itr = 1;
-}
-
 static irqreturn_t igc_msix_ring(int irq, void *data)
 {
 	struct igc_q_vector *q_vector = data;
-- 
cgit v1.2.3-59-g8ed1b


From 55cd7386c5b759ba4a736dee85cbad825013fd20 Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Tue, 19 Nov 2019 13:45:32 +0200
Subject: igc: Remove no need declaration of the igc_write_itr

We want to avoid forward-declarations of function if possible.
Rearrange the igc_write_itr function implementation.

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_main.c | 33 +++++++++++++++----------------
 1 file changed, 16 insertions(+), 17 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 4fb40c6aa583..c7b9bbbef3aa 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -53,7 +53,6 @@ MODULE_DEVICE_TABLE(pci, igc_pci_tbl);
 
 /* forward declaration */
 static int igc_sw_init(struct igc_adapter *);
-static void igc_write_itr(struct igc_q_vector *q_vector);
 
 enum latency_range {
 	lowest_latency = 0,
@@ -2931,6 +2930,22 @@ static irqreturn_t igc_msix_other(int irq, void *data)
 	return IRQ_HANDLED;
 }
 
+static void igc_write_itr(struct igc_q_vector *q_vector)
+{
+	u32 itr_val = q_vector->itr_val & IGC_QVECTOR_MASK;
+
+	if (!q_vector->set_itr)
+		return;
+
+	if (!itr_val)
+		itr_val = IGC_ITR_VAL_MASK;
+
+	itr_val |= IGC_EITR_CNT_IGNR;
+
+	writel(itr_val, q_vector->itr_register);
+	q_vector->set_itr = 0;
+}
+
 static irqreturn_t igc_msix_ring(int irq, void *data)
 {
 	struct igc_q_vector *q_vector = data;
@@ -4044,22 +4059,6 @@ request_done:
 	return err;
 }
 
-static void igc_write_itr(struct igc_q_vector *q_vector)
-{
-	u32 itr_val = q_vector->itr_val & IGC_QVECTOR_MASK;
-
-	if (!q_vector->set_itr)
-		return;
-
-	if (!itr_val)
-		itr_val = IGC_ITR_VAL_MASK;
-
-	itr_val |= IGC_EITR_CNT_IGNR;
-
-	writel(itr_val, q_vector->itr_register);
-	q_vector->set_itr = 0;
-}
-
 /**
  * __igc_open - Called when a network interface is made active
  * @netdev: network interface device structure
-- 
cgit v1.2.3-59-g8ed1b


From 63c92c9d2e536c91ace17dec7534883f40421753 Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Tue, 19 Nov 2019 13:45:40 +0200
Subject: igc: Remove no need declaration of the igc_sw_init

We want to avoid forward-declarations of function if possible.
Rearrange the igc_sw_init function implementation.

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_main.c | 2733 ++++++++++++++---------------
 1 file changed, 1365 insertions(+), 1368 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index c7b9bbbef3aa..24fbbc928d66 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -51,9 +51,6 @@ static const struct pci_device_id igc_pci_tbl[] = {
 
 MODULE_DEVICE_TABLE(pci, igc_pci_tbl);
 
-/* forward declaration */
-static int igc_sw_init(struct igc_adapter *);
-
 enum latency_range {
 	lowest_latency = 0,
 	low_latency = 1,
@@ -2354,1230 +2351,1302 @@ static void igc_irq_disable(struct igc_adapter *adapter)
 	}
 }
 
+void igc_set_flag_queue_pairs(struct igc_adapter *adapter,
+			      const u32 max_rss_queues)
+{
+	/* Determine if we need to pair queues. */
+	/* If rss_queues > half of max_rss_queues, pair the queues in
+	 * order to conserve interrupts due to limited supply.
+	 */
+	if (adapter->rss_queues > (max_rss_queues / 2))
+		adapter->flags |= IGC_FLAG_QUEUE_PAIRS;
+	else
+		adapter->flags &= ~IGC_FLAG_QUEUE_PAIRS;
+}
+
+unsigned int igc_get_max_rss_queues(struct igc_adapter *adapter)
+{
+	unsigned int max_rss_queues;
+
+	/* Determine the maximum number of RSS queues supported. */
+	max_rss_queues = IGC_MAX_RX_QUEUES;
+
+	return max_rss_queues;
+}
+
+static void igc_init_queue_configuration(struct igc_adapter *adapter)
+{
+	u32 max_rss_queues;
+
+	max_rss_queues = igc_get_max_rss_queues(adapter);
+	adapter->rss_queues = min_t(u32, max_rss_queues, num_online_cpus());
+
+	igc_set_flag_queue_pairs(adapter, max_rss_queues);
+}
+
 /**
- * igc_up - Open the interface and prepare it to handle traffic
- * @adapter: board private structure
+ * igc_reset_q_vector - Reset config for interrupt vector
+ * @adapter: board private structure to initialize
+ * @v_idx: Index of vector to be reset
+ *
+ * If NAPI is enabled it will delete any references to the
+ * NAPI struct. This is preparation for igc_free_q_vector.
  */
-void igc_up(struct igc_adapter *adapter)
+static void igc_reset_q_vector(struct igc_adapter *adapter, int v_idx)
 {
-	struct igc_hw *hw = &adapter->hw;
-	int i = 0;
+	struct igc_q_vector *q_vector = adapter->q_vector[v_idx];
 
-	/* hardware has been reset, we need to reload some things */
-	igc_configure(adapter);
+	/* if we're coming from igc_set_interrupt_capability, the vectors are
+	 * not yet allocated
+	 */
+	if (!q_vector)
+		return;
 
-	clear_bit(__IGC_DOWN, &adapter->state);
+	if (q_vector->tx.ring)
+		adapter->tx_ring[q_vector->tx.ring->queue_index] = NULL;
 
-	for (i = 0; i < adapter->num_q_vectors; i++)
-		napi_enable(&adapter->q_vector[i]->napi);
+	if (q_vector->rx.ring)
+		adapter->rx_ring[q_vector->rx.ring->queue_index] = NULL;
 
-	if (adapter->msix_entries)
-		igc_configure_msix(adapter);
-	else
-		igc_assign_vector(adapter->q_vector[0], 0);
+	netif_napi_del(&q_vector->napi);
+}
 
-	/* Clear any pending interrupts. */
-	rd32(IGC_ICR);
-	igc_irq_enable(adapter);
+/**
+ * igc_free_q_vector - Free memory allocated for specific interrupt vector
+ * @adapter: board private structure to initialize
+ * @v_idx: Index of vector to be freed
+ *
+ * This function frees the memory allocated to the q_vector.
+ */
+static void igc_free_q_vector(struct igc_adapter *adapter, int v_idx)
+{
+	struct igc_q_vector *q_vector = adapter->q_vector[v_idx];
 
-	netif_tx_start_all_queues(adapter->netdev);
+	adapter->q_vector[v_idx] = NULL;
 
-	/* start the watchdog. */
-	hw->mac.get_link_status = 1;
-	schedule_work(&adapter->watchdog_task);
+	/* igc_get_stats64() might access the rings on this vector,
+	 * we must wait a grace period before freeing it.
+	 */
+	if (q_vector)
+		kfree_rcu(q_vector, rcu);
 }
 
 /**
- * igc_update_stats - Update the board statistics counters
- * @adapter: board private structure
+ * igc_free_q_vectors - Free memory allocated for interrupt vectors
+ * @adapter: board private structure to initialize
+ *
+ * This function frees the memory allocated to the q_vectors.  In addition if
+ * NAPI is enabled it will delete any references to the NAPI struct prior
+ * to freeing the q_vector.
  */
-void igc_update_stats(struct igc_adapter *adapter)
+static void igc_free_q_vectors(struct igc_adapter *adapter)
 {
-	struct rtnl_link_stats64 *net_stats = &adapter->stats64;
-	struct pci_dev *pdev = adapter->pdev;
-	struct igc_hw *hw = &adapter->hw;
-	u64 _bytes, _packets;
-	u64 bytes, packets;
-	unsigned int start;
-	u32 mpc;
-	int i;
+	int v_idx = adapter->num_q_vectors;
 
-	/* Prevent stats update while adapter is being reset, or if the pci
-	 * connection is down.
-	 */
-	if (adapter->link_speed == 0)
-		return;
-	if (pci_channel_offline(pdev))
-		return;
+	adapter->num_tx_queues = 0;
+	adapter->num_rx_queues = 0;
+	adapter->num_q_vectors = 0;
 
-	packets = 0;
-	bytes = 0;
+	while (v_idx--) {
+		igc_reset_q_vector(adapter, v_idx);
+		igc_free_q_vector(adapter, v_idx);
+	}
+}
 
-	rcu_read_lock();
-	for (i = 0; i < adapter->num_rx_queues; i++) {
-		struct igc_ring *ring = adapter->rx_ring[i];
-		u32 rqdpc = rd32(IGC_RQDPC(i));
+/**
+ * igc_update_itr - update the dynamic ITR value based on statistics
+ * @q_vector: pointer to q_vector
+ * @ring_container: ring info to update the itr for
+ *
+ * Stores a new ITR value based on packets and byte
+ * counts during the last interrupt.  The advantage of per interrupt
+ * computation is faster updates and more accurate ITR for the current
+ * traffic pattern.  Constants in this function were computed
+ * based on theoretical maximum wire speed and thresholds were set based
+ * on testing data as well as attempting to minimize response time
+ * while increasing bulk throughput.
+ * NOTE: These calculations are only valid when operating in a single-
+ * queue environment.
+ */
+static void igc_update_itr(struct igc_q_vector *q_vector,
+			   struct igc_ring_container *ring_container)
+{
+	unsigned int packets = ring_container->total_packets;
+	unsigned int bytes = ring_container->total_bytes;
+	u8 itrval = ring_container->itr;
 
-		if (hw->mac.type >= igc_i225)
-			wr32(IGC_RQDPC(i), 0);
+	/* no packets, exit with status unchanged */
+	if (packets == 0)
+		return;
 
-		if (rqdpc) {
-			ring->rx_stats.drops += rqdpc;
-			net_stats->rx_fifo_errors += rqdpc;
+	switch (itrval) {
+	case lowest_latency:
+		/* handle TSO and jumbo frames */
+		if (bytes / packets > 8000)
+			itrval = bulk_latency;
+		else if ((packets < 5) && (bytes > 512))
+			itrval = low_latency;
+		break;
+	case low_latency:  /* 50 usec aka 20000 ints/s */
+		if (bytes > 10000) {
+			/* this if handles the TSO accounting */
+			if (bytes / packets > 8000)
+				itrval = bulk_latency;
+			else if ((packets < 10) || ((bytes / packets) > 1200))
+				itrval = bulk_latency;
+			else if ((packets > 35))
+				itrval = lowest_latency;
+		} else if (bytes / packets > 2000) {
+			itrval = bulk_latency;
+		} else if (packets <= 2 && bytes < 512) {
+			itrval = lowest_latency;
 		}
-
-		do {
-			start = u64_stats_fetch_begin_irq(&ring->rx_syncp);
-			_bytes = ring->rx_stats.bytes;
-			_packets = ring->rx_stats.packets;
-		} while (u64_stats_fetch_retry_irq(&ring->rx_syncp, start));
-		bytes += _bytes;
-		packets += _packets;
+		break;
+	case bulk_latency: /* 250 usec aka 4000 ints/s */
+		if (bytes > 25000) {
+			if (packets > 35)
+				itrval = low_latency;
+		} else if (bytes < 1500) {
+			itrval = low_latency;
+		}
+		break;
 	}
 
-	net_stats->rx_bytes = bytes;
-	net_stats->rx_packets = packets;
+	/* clear work counters since we have the values we need */
+	ring_container->total_bytes = 0;
+	ring_container->total_packets = 0;
 
-	packets = 0;
-	bytes = 0;
-	for (i = 0; i < adapter->num_tx_queues; i++) {
-		struct igc_ring *ring = adapter->tx_ring[i];
+	/* write updated itr to ring container */
+	ring_container->itr = itrval;
+}
 
-		do {
-			start = u64_stats_fetch_begin_irq(&ring->tx_syncp);
-			_bytes = ring->tx_stats.bytes;
-			_packets = ring->tx_stats.packets;
-		} while (u64_stats_fetch_retry_irq(&ring->tx_syncp, start));
-		bytes += _bytes;
-		packets += _packets;
+static void igc_set_itr(struct igc_q_vector *q_vector)
+{
+	struct igc_adapter *adapter = q_vector->adapter;
+	u32 new_itr = q_vector->itr_val;
+	u8 current_itr = 0;
+
+	/* for non-gigabit speeds, just fix the interrupt rate at 4000 */
+	switch (adapter->link_speed) {
+	case SPEED_10:
+	case SPEED_100:
+		current_itr = 0;
+		new_itr = IGC_4K_ITR;
+		goto set_itr_now;
+	default:
+		break;
 	}
-	net_stats->tx_bytes = bytes;
-	net_stats->tx_packets = packets;
-	rcu_read_unlock();
 
-	/* read stats registers */
-	adapter->stats.crcerrs += rd32(IGC_CRCERRS);
-	adapter->stats.gprc += rd32(IGC_GPRC);
-	adapter->stats.gorc += rd32(IGC_GORCL);
-	rd32(IGC_GORCH); /* clear GORCL */
-	adapter->stats.bprc += rd32(IGC_BPRC);
-	adapter->stats.mprc += rd32(IGC_MPRC);
-	adapter->stats.roc += rd32(IGC_ROC);
+	igc_update_itr(q_vector, &q_vector->tx);
+	igc_update_itr(q_vector, &q_vector->rx);
 
-	adapter->stats.prc64 += rd32(IGC_PRC64);
-	adapter->stats.prc127 += rd32(IGC_PRC127);
-	adapter->stats.prc255 += rd32(IGC_PRC255);
-	adapter->stats.prc511 += rd32(IGC_PRC511);
-	adapter->stats.prc1023 += rd32(IGC_PRC1023);
-	adapter->stats.prc1522 += rd32(IGC_PRC1522);
-	adapter->stats.symerrs += rd32(IGC_SYMERRS);
-	adapter->stats.sec += rd32(IGC_SEC);
+	current_itr = max(q_vector->rx.itr, q_vector->tx.itr);
 
-	mpc = rd32(IGC_MPC);
-	adapter->stats.mpc += mpc;
-	net_stats->rx_fifo_errors += mpc;
-	adapter->stats.scc += rd32(IGC_SCC);
-	adapter->stats.ecol += rd32(IGC_ECOL);
-	adapter->stats.mcc += rd32(IGC_MCC);
-	adapter->stats.latecol += rd32(IGC_LATECOL);
-	adapter->stats.dc += rd32(IGC_DC);
-	adapter->stats.rlec += rd32(IGC_RLEC);
-	adapter->stats.xonrxc += rd32(IGC_XONRXC);
-	adapter->stats.xontxc += rd32(IGC_XONTXC);
-	adapter->stats.xoffrxc += rd32(IGC_XOFFRXC);
-	adapter->stats.xofftxc += rd32(IGC_XOFFTXC);
-	adapter->stats.fcruc += rd32(IGC_FCRUC);
-	adapter->stats.gptc += rd32(IGC_GPTC);
-	adapter->stats.gotc += rd32(IGC_GOTCL);
-	rd32(IGC_GOTCH); /* clear GOTCL */
-	adapter->stats.rnbc += rd32(IGC_RNBC);
-	adapter->stats.ruc += rd32(IGC_RUC);
-	adapter->stats.rfc += rd32(IGC_RFC);
-	adapter->stats.rjc += rd32(IGC_RJC);
-	adapter->stats.tor += rd32(IGC_TORH);
-	adapter->stats.tot += rd32(IGC_TOTH);
-	adapter->stats.tpr += rd32(IGC_TPR);
-
-	adapter->stats.ptc64 += rd32(IGC_PTC64);
-	adapter->stats.ptc127 += rd32(IGC_PTC127);
-	adapter->stats.ptc255 += rd32(IGC_PTC255);
-	adapter->stats.ptc511 += rd32(IGC_PTC511);
-	adapter->stats.ptc1023 += rd32(IGC_PTC1023);
-	adapter->stats.ptc1522 += rd32(IGC_PTC1522);
-
-	adapter->stats.mptc += rd32(IGC_MPTC);
-	adapter->stats.bptc += rd32(IGC_BPTC);
-
-	adapter->stats.tpt += rd32(IGC_TPT);
-	adapter->stats.colc += rd32(IGC_COLC);
-
-	adapter->stats.algnerrc += rd32(IGC_ALGNERRC);
-
-	adapter->stats.tsctc += rd32(IGC_TSCTC);
-	adapter->stats.tsctfc += rd32(IGC_TSCTFC);
-
-	adapter->stats.iac += rd32(IGC_IAC);
-	adapter->stats.icrxoc += rd32(IGC_ICRXOC);
-	adapter->stats.icrxptc += rd32(IGC_ICRXPTC);
-	adapter->stats.icrxatc += rd32(IGC_ICRXATC);
-	adapter->stats.ictxptc += rd32(IGC_ICTXPTC);
-	adapter->stats.ictxatc += rd32(IGC_ICTXATC);
-	adapter->stats.ictxqec += rd32(IGC_ICTXQEC);
-	adapter->stats.ictxqmtc += rd32(IGC_ICTXQMTC);
-	adapter->stats.icrxdmtc += rd32(IGC_ICRXDMTC);
-
-	/* Fill out the OS statistics structure */
-	net_stats->multicast = adapter->stats.mprc;
-	net_stats->collisions = adapter->stats.colc;
-
-	/* Rx Errors */
-
-	/* RLEC on some newer hardware can be incorrect so build
-	 * our own version based on RUC and ROC
-	 */
-	net_stats->rx_errors = adapter->stats.rxerrc +
-		adapter->stats.crcerrs + adapter->stats.algnerrc +
-		adapter->stats.ruc + adapter->stats.roc +
-		adapter->stats.cexterr;
-	net_stats->rx_length_errors = adapter->stats.ruc +
-				      adapter->stats.roc;
-	net_stats->rx_crc_errors = adapter->stats.crcerrs;
-	net_stats->rx_frame_errors = adapter->stats.algnerrc;
-	net_stats->rx_missed_errors = adapter->stats.mpc;
-
-	/* Tx Errors */
-	net_stats->tx_errors = adapter->stats.ecol +
-			       adapter->stats.latecol;
-	net_stats->tx_aborted_errors = adapter->stats.ecol;
-	net_stats->tx_window_errors = adapter->stats.latecol;
-	net_stats->tx_carrier_errors = adapter->stats.tncrs;
+	/* conservative mode (itr 3) eliminates the lowest_latency setting */
+	if (current_itr == lowest_latency &&
+	    ((q_vector->rx.ring && adapter->rx_itr_setting == 3) ||
+	    (!q_vector->rx.ring && adapter->tx_itr_setting == 3)))
+		current_itr = low_latency;
 
-	/* Tx Dropped needs to be maintained elsewhere */
+	switch (current_itr) {
+	/* counts and packets in update_itr are dependent on these numbers */
+	case lowest_latency:
+		new_itr = IGC_70K_ITR; /* 70,000 ints/sec */
+		break;
+	case low_latency:
+		new_itr = IGC_20K_ITR; /* 20,000 ints/sec */
+		break;
+	case bulk_latency:
+		new_itr = IGC_4K_ITR;  /* 4,000 ints/sec */
+		break;
+	default:
+		break;
+	}
 
-	/* Management Stats */
-	adapter->stats.mgptc += rd32(IGC_MGTPTC);
-	adapter->stats.mgprc += rd32(IGC_MGTPRC);
-	adapter->stats.mgpdc += rd32(IGC_MGTPDC);
+set_itr_now:
+	if (new_itr != q_vector->itr_val) {
+		/* this attempts to bias the interrupt rate towards Bulk
+		 * by adding intermediate steps when interrupt rate is
+		 * increasing
+		 */
+		new_itr = new_itr > q_vector->itr_val ?
+			  max((new_itr * q_vector->itr_val) /
+			  (new_itr + (q_vector->itr_val >> 2)),
+			  new_itr) : new_itr;
+		/* Don't write the value here; it resets the adapter's
+		 * internal timer, and causes us to delay far longer than
+		 * we should between interrupts.  Instead, we write the ITR
+		 * value at the beginning of the next interrupt so the timing
+		 * ends up being correct.
+		 */
+		q_vector->itr_val = new_itr;
+		q_vector->set_itr = 1;
+	}
 }
 
-static void igc_nfc_filter_exit(struct igc_adapter *adapter)
+static void igc_reset_interrupt_capability(struct igc_adapter *adapter)
 {
-	struct igc_nfc_filter *rule;
-
-	spin_lock(&adapter->nfc_lock);
-
-	hlist_for_each_entry(rule, &adapter->nfc_filter_list, nfc_node)
-		igc_erase_filter(adapter, rule);
+	int v_idx = adapter->num_q_vectors;
 
-	hlist_for_each_entry(rule, &adapter->cls_flower_list, nfc_node)
-		igc_erase_filter(adapter, rule);
+	if (adapter->msix_entries) {
+		pci_disable_msix(adapter->pdev);
+		kfree(adapter->msix_entries);
+		adapter->msix_entries = NULL;
+	} else if (adapter->flags & IGC_FLAG_HAS_MSI) {
+		pci_disable_msi(adapter->pdev);
+	}
 
-	spin_unlock(&adapter->nfc_lock);
+	while (v_idx--)
+		igc_reset_q_vector(adapter, v_idx);
 }
 
 /**
- * igc_down - Close the interface
- * @adapter: board private structure
+ * igc_set_interrupt_capability - set MSI or MSI-X if supported
+ * @adapter: Pointer to adapter structure
+ * @msix: boolean value for MSI-X capability
+ *
+ * Attempt to configure interrupts using the best available
+ * capabilities of the hardware and kernel.
  */
-void igc_down(struct igc_adapter *adapter)
+static void igc_set_interrupt_capability(struct igc_adapter *adapter,
+					 bool msix)
 {
-	struct net_device *netdev = adapter->netdev;
-	struct igc_hw *hw = &adapter->hw;
-	u32 tctl, rctl;
-	int i = 0;
-
-	set_bit(__IGC_DOWN, &adapter->state);
-
-	/* disable receives in the hardware */
-	rctl = rd32(IGC_RCTL);
-	wr32(IGC_RCTL, rctl & ~IGC_RCTL_EN);
-	/* flush and sleep below */
-
-	igc_nfc_filter_exit(adapter);
-
-	/* set trans_start so we don't get spurious watchdogs during reset */
-	netif_trans_update(netdev);
+	int numvecs, i;
+	int err;
 
-	netif_carrier_off(netdev);
-	netif_tx_stop_all_queues(netdev);
+	if (!msix)
+		goto msi_only;
+	adapter->flags |= IGC_FLAG_HAS_MSIX;
 
-	/* disable transmits in the hardware */
-	tctl = rd32(IGC_TCTL);
-	tctl &= ~IGC_TCTL_EN;
-	wr32(IGC_TCTL, tctl);
-	/* flush both disables and wait for them to finish */
-	wrfl();
-	usleep_range(10000, 20000);
+	/* Number of supported queues. */
+	adapter->num_rx_queues = adapter->rss_queues;
 
-	igc_irq_disable(adapter);
+	adapter->num_tx_queues = adapter->rss_queues;
 
-	adapter->flags &= ~IGC_FLAG_NEED_LINK_UPDATE;
+	/* start with one vector for every Rx queue */
+	numvecs = adapter->num_rx_queues;
 
-	for (i = 0; i < adapter->num_q_vectors; i++) {
-		if (adapter->q_vector[i]) {
-			napi_synchronize(&adapter->q_vector[i]->napi);
-			napi_disable(&adapter->q_vector[i]->napi);
-		}
-	}
+	/* if Tx handler is separate add 1 for every Tx queue */
+	if (!(adapter->flags & IGC_FLAG_QUEUE_PAIRS))
+		numvecs += adapter->num_tx_queues;
 
-	del_timer_sync(&adapter->watchdog_timer);
-	del_timer_sync(&adapter->phy_info_timer);
+	/* store the number of vectors reserved for queues */
+	adapter->num_q_vectors = numvecs;
 
-	/* record the stats before reset*/
-	spin_lock(&adapter->stats64_lock);
-	igc_update_stats(adapter);
-	spin_unlock(&adapter->stats64_lock);
+	/* add 1 vector for link status interrupts */
+	numvecs++;
 
-	adapter->link_speed = 0;
-	adapter->link_duplex = 0;
+	adapter->msix_entries = kcalloc(numvecs, sizeof(struct msix_entry),
+					GFP_KERNEL);
 
-	if (!pci_channel_offline(adapter->pdev))
-		igc_reset(adapter);
+	if (!adapter->msix_entries)
+		return;
 
-	/* clear VLAN promisc flag so VFTA will be updated if necessary */
-	adapter->flags &= ~IGC_FLAG_VLAN_PROMISC;
+	/* populate entry values */
+	for (i = 0; i < numvecs; i++)
+		adapter->msix_entries[i].entry = i;
 
-	igc_clean_all_tx_rings(adapter);
-	igc_clean_all_rx_rings(adapter);
-}
+	err = pci_enable_msix_range(adapter->pdev,
+				    adapter->msix_entries,
+				    numvecs,
+				    numvecs);
+	if (err > 0)
+		return;
 
-void igc_reinit_locked(struct igc_adapter *adapter)
-{
-	WARN_ON(in_interrupt());
-	while (test_and_set_bit(__IGC_RESETTING, &adapter->state))
-		usleep_range(1000, 2000);
-	igc_down(adapter);
-	igc_up(adapter);
-	clear_bit(__IGC_RESETTING, &adapter->state);
-}
+	kfree(adapter->msix_entries);
+	adapter->msix_entries = NULL;
 
-static void igc_reset_task(struct work_struct *work)
-{
-	struct igc_adapter *adapter;
+	igc_reset_interrupt_capability(adapter);
 
-	adapter = container_of(work, struct igc_adapter, reset_task);
+msi_only:
+	adapter->flags &= ~IGC_FLAG_HAS_MSIX;
 
-	netdev_err(adapter->netdev, "Reset adapter\n");
-	igc_reinit_locked(adapter);
+	adapter->rss_queues = 1;
+	adapter->flags |= IGC_FLAG_QUEUE_PAIRS;
+	adapter->num_rx_queues = 1;
+	adapter->num_tx_queues = 1;
+	adapter->num_q_vectors = 1;
+	if (!pci_enable_msi(adapter->pdev))
+		adapter->flags |= IGC_FLAG_HAS_MSI;
 }
 
 /**
- * igc_change_mtu - Change the Maximum Transfer Unit
- * @netdev: network interface device structure
- * @new_mtu: new value for maximum frame size
+ * igc_update_ring_itr - update the dynamic ITR value based on packet size
+ * @q_vector: pointer to q_vector
  *
- * Returns 0 on success, negative on failure
+ * Stores a new ITR value based on strictly on packet size.  This
+ * algorithm is less sophisticated than that used in igc_update_itr,
+ * due to the difficulty of synchronizing statistics across multiple
+ * receive rings.  The divisors and thresholds used by this function
+ * were determined based on theoretical maximum wire speed and testing
+ * data, in order to minimize response time while increasing bulk
+ * throughput.
+ * NOTE: This function is called only when operating in a multiqueue
+ * receive environment.
  */
-static int igc_change_mtu(struct net_device *netdev, int new_mtu)
+static void igc_update_ring_itr(struct igc_q_vector *q_vector)
 {
-	int max_frame = new_mtu + ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN;
-	struct igc_adapter *adapter = netdev_priv(netdev);
+	struct igc_adapter *adapter = q_vector->adapter;
+	int new_val = q_vector->itr_val;
+	int avg_wire_size = 0;
+	unsigned int packets;
 
-	/* adjust max frame to be at least the size of a standard frame */
-	if (max_frame < (ETH_FRAME_LEN + ETH_FCS_LEN))
-		max_frame = ETH_FRAME_LEN + ETH_FCS_LEN;
+	/* For non-gigabit speeds, just fix the interrupt rate at 4000
+	 * ints/sec - ITR timer value of 120 ticks.
+	 */
+	switch (adapter->link_speed) {
+	case SPEED_10:
+	case SPEED_100:
+		new_val = IGC_4K_ITR;
+		goto set_itr_val;
+	default:
+		break;
+	}
 
-	while (test_and_set_bit(__IGC_RESETTING, &adapter->state))
-		usleep_range(1000, 2000);
+	packets = q_vector->rx.total_packets;
+	if (packets)
+		avg_wire_size = q_vector->rx.total_bytes / packets;
 
-	/* igc_down has a dependency on max_frame_size */
-	adapter->max_frame_size = max_frame;
+	packets = q_vector->tx.total_packets;
+	if (packets)
+		avg_wire_size = max_t(u32, avg_wire_size,
+				      q_vector->tx.total_bytes / packets);
 
-	if (netif_running(netdev))
-		igc_down(adapter);
+	/* if avg_wire_size isn't set no work was done */
+	if (!avg_wire_size)
+		goto clear_counts;
 
-	netdev_dbg(netdev, "changing MTU from %d to %d\n",
-		   netdev->mtu, new_mtu);
-	netdev->mtu = new_mtu;
+	/* Add 24 bytes to size to account for CRC, preamble, and gap */
+	avg_wire_size += 24;
 
-	if (netif_running(netdev))
-		igc_up(adapter);
+	/* Don't starve jumbo frames */
+	avg_wire_size = min(avg_wire_size, 3000);
+
+	/* Give a little boost to mid-size frames */
+	if (avg_wire_size > 300 && avg_wire_size < 1200)
+		new_val = avg_wire_size / 3;
 	else
-		igc_reset(adapter);
+		new_val = avg_wire_size / 2;
 
-	clear_bit(__IGC_RESETTING, &adapter->state);
+	/* conservative mode (itr 3) eliminates the lowest_latency setting */
+	if (new_val < IGC_20K_ITR &&
+	    ((q_vector->rx.ring && adapter->rx_itr_setting == 3) ||
+	    (!q_vector->rx.ring && adapter->tx_itr_setting == 3)))
+		new_val = IGC_20K_ITR;
 
-	return 0;
+set_itr_val:
+	if (new_val != q_vector->itr_val) {
+		q_vector->itr_val = new_val;
+		q_vector->set_itr = 1;
+	}
+clear_counts:
+	q_vector->rx.total_bytes = 0;
+	q_vector->rx.total_packets = 0;
+	q_vector->tx.total_bytes = 0;
+	q_vector->tx.total_packets = 0;
 }
 
-/**
- * igc_get_stats - Get System Network Statistics
- * @netdev: network interface device structure
- *
- * Returns the address of the device statistics structure.
- * The statistics are updated here and also from the timer callback.
- */
-static struct net_device_stats *igc_get_stats(struct net_device *netdev)
+static void igc_ring_irq_enable(struct igc_q_vector *q_vector)
 {
-	struct igc_adapter *adapter = netdev_priv(netdev);
+	struct igc_adapter *adapter = q_vector->adapter;
+	struct igc_hw *hw = &adapter->hw;
 
-	if (!test_bit(__IGC_RESETTING, &adapter->state))
-		igc_update_stats(adapter);
+	if ((q_vector->rx.ring && (adapter->rx_itr_setting & 3)) ||
+	    (!q_vector->rx.ring && (adapter->tx_itr_setting & 3))) {
+		if (adapter->num_q_vectors == 1)
+			igc_set_itr(q_vector);
+		else
+			igc_update_ring_itr(q_vector);
+	}
 
-	/* only return the current stats */
-	return &netdev->stats;
+	if (!test_bit(__IGC_DOWN, &adapter->state)) {
+		if (adapter->msix_entries)
+			wr32(IGC_EIMS, q_vector->eims_value);
+		else
+			igc_irq_enable(adapter);
+	}
 }
 
-static netdev_features_t igc_fix_features(struct net_device *netdev,
-					  netdev_features_t features)
+static void igc_add_ring(struct igc_ring *ring,
+			 struct igc_ring_container *head)
 {
-	/* Since there is no support for separate Rx/Tx vlan accel
-	 * enable/disable make sure Tx flag is always in same state as Rx.
-	 */
-	if (features & NETIF_F_HW_VLAN_CTAG_RX)
-		features |= NETIF_F_HW_VLAN_CTAG_TX;
-	else
-		features &= ~NETIF_F_HW_VLAN_CTAG_TX;
-
-	return features;
+	head->ring = ring;
+	head->count++;
 }
 
-static int igc_set_features(struct net_device *netdev,
-			    netdev_features_t features)
+/**
+ * igc_cache_ring_register - Descriptor ring to register mapping
+ * @adapter: board private structure to initialize
+ *
+ * Once we know the feature-set enabled for the device, we'll cache
+ * the register offset the descriptor ring is assigned to.
+ */
+static void igc_cache_ring_register(struct igc_adapter *adapter)
 {
-	netdev_features_t changed = netdev->features ^ features;
-	struct igc_adapter *adapter = netdev_priv(netdev);
-
-	/* Add VLAN support */
-	if (!(changed & (NETIF_F_RXALL | NETIF_F_NTUPLE)))
-		return 0;
-
-	if (!(features & NETIF_F_NTUPLE)) {
-		struct hlist_node *node2;
-		struct igc_nfc_filter *rule;
+	int i = 0, j = 0;
 
-		spin_lock(&adapter->nfc_lock);
-		hlist_for_each_entry_safe(rule, node2,
-					  &adapter->nfc_filter_list, nfc_node) {
-			igc_erase_filter(adapter, rule);
-			hlist_del(&rule->nfc_node);
-			kfree(rule);
-		}
-		spin_unlock(&adapter->nfc_lock);
-		adapter->nfc_filter_count = 0;
+	switch (adapter->hw.mac.type) {
+	case igc_i225:
+	/* Fall through */
+	default:
+		for (; i < adapter->num_rx_queues; i++)
+			adapter->rx_ring[i]->reg_idx = i;
+		for (; j < adapter->num_tx_queues; j++)
+			adapter->tx_ring[j]->reg_idx = j;
+		break;
 	}
-
-	netdev->features = features;
-
-	if (netif_running(netdev))
-		igc_reinit_locked(adapter);
-	else
-		igc_reset(adapter);
-
-	return 1;
 }
 
-static netdev_features_t
-igc_features_check(struct sk_buff *skb, struct net_device *dev,
-		   netdev_features_t features)
+/**
+ * igc_poll - NAPI Rx polling callback
+ * @napi: napi polling structure
+ * @budget: count of how many packets we should handle
+ */
+static int igc_poll(struct napi_struct *napi, int budget)
 {
-	unsigned int network_hdr_len, mac_hdr_len;
+	struct igc_q_vector *q_vector = container_of(napi,
+						     struct igc_q_vector,
+						     napi);
+	bool clean_complete = true;
+	int work_done = 0;
 
-	/* Make certain the headers can be described by a context descriptor */
-	mac_hdr_len = skb_network_header(skb) - skb->data;
-	if (unlikely(mac_hdr_len > IGC_MAX_MAC_HDR_LEN))
-		return features & ~(NETIF_F_HW_CSUM |
-				    NETIF_F_SCTP_CRC |
-				    NETIF_F_HW_VLAN_CTAG_TX |
-				    NETIF_F_TSO |
-				    NETIF_F_TSO6);
+	if (q_vector->tx.ring)
+		clean_complete = igc_clean_tx_irq(q_vector, budget);
 
-	network_hdr_len = skb_checksum_start(skb) - skb_network_header(skb);
-	if (unlikely(network_hdr_len >  IGC_MAX_NETWORK_HDR_LEN))
-		return features & ~(NETIF_F_HW_CSUM |
-				    NETIF_F_SCTP_CRC |
-				    NETIF_F_TSO |
-				    NETIF_F_TSO6);
+	if (q_vector->rx.ring) {
+		int cleaned = igc_clean_rx_irq(q_vector, budget);
 
-	/* We can only support IPv4 TSO in tunnels if we can mangle the
-	 * inner IP ID field, so strip TSO if MANGLEID is not supported.
+		work_done += cleaned;
+		if (cleaned >= budget)
+			clean_complete = false;
+	}
+
+	/* If all work not completed, return budget and keep polling */
+	if (!clean_complete)
+		return budget;
+
+	/* Exit the polling mode, but don't re-enable interrupts if stack might
+	 * poll us due to busy-polling
 	 */
-	if (skb->encapsulation && !(features & NETIF_F_TSO_MANGLEID))
-		features &= ~NETIF_F_TSO;
+	if (likely(napi_complete_done(napi, work_done)))
+		igc_ring_irq_enable(q_vector);
 
-	return features;
+	return min(work_done, budget - 1);
 }
 
-/* Add a MAC filter for 'addr' directing matching traffic to 'queue',
- * 'flags' is used to indicate what kind of match is made, match is by
- * default for the destination address, if matching by source address
- * is desired the flag IGC_MAC_STATE_SRC_ADDR can be used.
+/**
+ * igc_alloc_q_vector - Allocate memory for a single interrupt vector
+ * @adapter: board private structure to initialize
+ * @v_count: q_vectors allocated on adapter, used for ring interleaving
+ * @v_idx: index of vector in adapter struct
+ * @txr_count: total number of Tx rings to allocate
+ * @txr_idx: index of first Tx ring to allocate
+ * @rxr_count: total number of Rx rings to allocate
+ * @rxr_idx: index of first Rx ring to allocate
+ *
+ * We allocate one q_vector.  If allocation fails we return -ENOMEM.
  */
-static int igc_add_mac_filter_flags(struct igc_adapter *adapter,
-				    const u8 *addr, const u8 queue,
-				    const u8 flags)
+static int igc_alloc_q_vector(struct igc_adapter *adapter,
+			      unsigned int v_count, unsigned int v_idx,
+			      unsigned int txr_count, unsigned int txr_idx,
+			      unsigned int rxr_count, unsigned int rxr_idx)
 {
-	struct igc_hw *hw = &adapter->hw;
-	int rar_entries = hw->mac.rar_entry_count;
-	int i;
+	struct igc_q_vector *q_vector;
+	struct igc_ring *ring;
+	int ring_count;
 
-	if (is_zero_ether_addr(addr))
-		return -EINVAL;
+	/* igc only supports 1 Tx and/or 1 Rx queue per vector */
+	if (txr_count > 1 || rxr_count > 1)
+		return -ENOMEM;
 
-	/* Search for the first empty entry in the MAC table.
-	 * Do not touch entries at the end of the table reserved for the VF MAC
-	 * addresses.
-	 */
-	for (i = 0; i < rar_entries; i++) {
-		if (!igc_mac_entry_can_be_used(&adapter->mac_table[i],
-					       addr, flags))
-			continue;
+	ring_count = txr_count + rxr_count;
 
-		ether_addr_copy(adapter->mac_table[i].addr, addr);
-		adapter->mac_table[i].queue = queue;
-		adapter->mac_table[i].state |= IGC_MAC_STATE_IN_USE | flags;
+	/* allocate q_vector and rings */
+	q_vector = adapter->q_vector[v_idx];
+	if (!q_vector)
+		q_vector = kzalloc(struct_size(q_vector, ring, ring_count),
+				   GFP_KERNEL);
+	else
+		memset(q_vector, 0, struct_size(q_vector, ring, ring_count));
+	if (!q_vector)
+		return -ENOMEM;
 
-		igc_rar_set_index(adapter, i);
-		return i;
-	}
+	/* initialize NAPI */
+	netif_napi_add(adapter->netdev, &q_vector->napi,
+		       igc_poll, 64);
 
-	return -ENOSPC;
-}
+	/* tie q_vector and adapter together */
+	adapter->q_vector[v_idx] = q_vector;
+	q_vector->adapter = adapter;
 
-int igc_add_mac_steering_filter(struct igc_adapter *adapter,
-				const u8 *addr, u8 queue, u8 flags)
-{
-	return igc_add_mac_filter_flags(adapter, addr, queue,
-					IGC_MAC_STATE_QUEUE_STEERING | flags);
-}
+	/* initialize work limits */
+	q_vector->tx.work_limit = adapter->tx_work_limit;
 
-/* Remove a MAC filter for 'addr' directing matching traffic to
- * 'queue', 'flags' is used to indicate what kind of match need to be
- * removed, match is by default for the destination address, if
- * matching by source address is to be removed the flag
- * IGC_MAC_STATE_SRC_ADDR can be used.
- */
-static int igc_del_mac_filter_flags(struct igc_adapter *adapter,
-				    const u8 *addr, const u8 queue,
-				    const u8 flags)
-{
-	struct igc_hw *hw = &adapter->hw;
-	int rar_entries = hw->mac.rar_entry_count;
-	int i;
+	/* initialize ITR configuration */
+	q_vector->itr_register = adapter->io_addr + IGC_EITR(0);
+	q_vector->itr_val = IGC_START_ITR;
 
-	if (is_zero_ether_addr(addr))
-		return -EINVAL;
+	/* initialize pointer to rings */
+	ring = q_vector->ring;
 
-	/* Search for matching entry in the MAC table based on given address
-	 * and queue. Do not touch entries at the end of the table reserved
-	 * for the VF MAC addresses.
-	 */
-	for (i = 0; i < rar_entries; i++) {
-		if (!(adapter->mac_table[i].state & IGC_MAC_STATE_IN_USE))
-			continue;
-		if ((adapter->mac_table[i].state & flags) != flags)
-			continue;
-		if (adapter->mac_table[i].queue != queue)
-			continue;
-		if (!ether_addr_equal(adapter->mac_table[i].addr, addr))
-			continue;
+	/* initialize ITR */
+	if (rxr_count) {
+		/* rx or rx/tx vector */
+		if (!adapter->rx_itr_setting || adapter->rx_itr_setting > 3)
+			q_vector->itr_val = adapter->rx_itr_setting;
+	} else {
+		/* tx only vector */
+		if (!adapter->tx_itr_setting || adapter->tx_itr_setting > 3)
+			q_vector->itr_val = adapter->tx_itr_setting;
+	}
 
-		/* When a filter for the default address is "deleted",
-		 * we return it to its initial configuration
-		 */
-		if (adapter->mac_table[i].state & IGC_MAC_STATE_DEFAULT) {
-			adapter->mac_table[i].state =
-				IGC_MAC_STATE_DEFAULT | IGC_MAC_STATE_IN_USE;
-		} else {
-			adapter->mac_table[i].state = 0;
-			adapter->mac_table[i].queue = 0;
-			memset(adapter->mac_table[i].addr, 0, ETH_ALEN);
-		}
+	if (txr_count) {
+		/* assign generic ring traits */
+		ring->dev = &adapter->pdev->dev;
+		ring->netdev = adapter->netdev;
 
-		igc_rar_set_index(adapter, i);
-		return 0;
+		/* configure backlink on ring */
+		ring->q_vector = q_vector;
+
+		/* update q_vector Tx values */
+		igc_add_ring(ring, &q_vector->tx);
+
+		/* apply Tx specific ring traits */
+		ring->count = adapter->tx_ring_count;
+		ring->queue_index = txr_idx;
+
+		/* assign ring to adapter */
+		adapter->tx_ring[txr_idx] = ring;
+
+		/* push pointer to next ring */
+		ring++;
 	}
 
-	return -ENOENT;
-}
+	if (rxr_count) {
+		/* assign generic ring traits */
+		ring->dev = &adapter->pdev->dev;
+		ring->netdev = adapter->netdev;
 
-int igc_del_mac_steering_filter(struct igc_adapter *adapter,
-				const u8 *addr, u8 queue, u8 flags)
-{
-	return igc_del_mac_filter_flags(adapter, addr, queue,
-					IGC_MAC_STATE_QUEUE_STEERING | flags);
+		/* configure backlink on ring */
+		ring->q_vector = q_vector;
+
+		/* update q_vector Rx values */
+		igc_add_ring(ring, &q_vector->rx);
+
+		/* apply Rx specific ring traits */
+		ring->count = adapter->rx_ring_count;
+		ring->queue_index = rxr_idx;
+
+		/* assign ring to adapter */
+		adapter->rx_ring[rxr_idx] = ring;
+	}
+
+	return 0;
 }
 
 /**
- * igc_msix_other - msix other interrupt handler
- * @irq: interrupt number
- * @data: pointer to a q_vector
+ * igc_alloc_q_vectors - Allocate memory for interrupt vectors
+ * @adapter: board private structure to initialize
+ *
+ * We allocate one q_vector per queue interrupt.  If allocation fails we
+ * return -ENOMEM.
  */
-static irqreturn_t igc_msix_other(int irq, void *data)
+static int igc_alloc_q_vectors(struct igc_adapter *adapter)
 {
-	struct igc_adapter *adapter = data;
-	struct igc_hw *hw = &adapter->hw;
-	u32 icr = rd32(IGC_ICR);
+	int rxr_remaining = adapter->num_rx_queues;
+	int txr_remaining = adapter->num_tx_queues;
+	int rxr_idx = 0, txr_idx = 0, v_idx = 0;
+	int q_vectors = adapter->num_q_vectors;
+	int err;
 
-	/* reading ICR causes bit 31 of EICR to be cleared */
-	if (icr & IGC_ICR_DRSTA)
-		schedule_work(&adapter->reset_task);
+	if (q_vectors >= (rxr_remaining + txr_remaining)) {
+		for (; rxr_remaining; v_idx++) {
+			err = igc_alloc_q_vector(adapter, q_vectors, v_idx,
+						 0, 0, 1, rxr_idx);
 
-	if (icr & IGC_ICR_DOUTSYNC) {
-		/* HW is reporting DMA is out of sync */
-		adapter->stats.doosync++;
-	}
+			if (err)
+				goto err_out;
 
-	if (icr & IGC_ICR_LSC) {
-		hw->mac.get_link_status = 1;
-		/* guard against interrupt when we're going down */
-		if (!test_bit(__IGC_DOWN, &adapter->state))
-			mod_timer(&adapter->watchdog_timer, jiffies + 1);
+			/* update counts and index */
+			rxr_remaining--;
+			rxr_idx++;
+		}
 	}
 
-	wr32(IGC_EIMS, adapter->eims_other);
-
-	return IRQ_HANDLED;
-}
-
-static void igc_write_itr(struct igc_q_vector *q_vector)
-{
-	u32 itr_val = q_vector->itr_val & IGC_QVECTOR_MASK;
-
-	if (!q_vector->set_itr)
-		return;
+	for (; v_idx < q_vectors; v_idx++) {
+		int rqpv = DIV_ROUND_UP(rxr_remaining, q_vectors - v_idx);
+		int tqpv = DIV_ROUND_UP(txr_remaining, q_vectors - v_idx);
 
-	if (!itr_val)
-		itr_val = IGC_ITR_VAL_MASK;
+		err = igc_alloc_q_vector(adapter, q_vectors, v_idx,
+					 tqpv, txr_idx, rqpv, rxr_idx);
 
-	itr_val |= IGC_EITR_CNT_IGNR;
+		if (err)
+			goto err_out;
 
-	writel(itr_val, q_vector->itr_register);
-	q_vector->set_itr = 0;
-}
+		/* update counts and index */
+		rxr_remaining -= rqpv;
+		txr_remaining -= tqpv;
+		rxr_idx++;
+		txr_idx++;
+	}
 
-static irqreturn_t igc_msix_ring(int irq, void *data)
-{
-	struct igc_q_vector *q_vector = data;
+	return 0;
 
-	/* Write the ITR value calculated from the previous interrupt. */
-	igc_write_itr(q_vector);
+err_out:
+	adapter->num_tx_queues = 0;
+	adapter->num_rx_queues = 0;
+	adapter->num_q_vectors = 0;
 
-	napi_schedule(&q_vector->napi);
+	while (v_idx--)
+		igc_free_q_vector(adapter, v_idx);
 
-	return IRQ_HANDLED;
+	return -ENOMEM;
 }
 
 /**
- * igc_request_msix - Initialize MSI-X interrupts
+ * igc_init_interrupt_scheme - initialize interrupts, allocate queues/vectors
  * @adapter: Pointer to adapter structure
+ * @msix: boolean for MSI-X capability
  *
- * igc_request_msix allocates MSI-X vectors and requests interrupts from the
- * kernel.
+ * This function initializes the interrupts and allocates all of the queues.
  */
-static int igc_request_msix(struct igc_adapter *adapter)
+static int igc_init_interrupt_scheme(struct igc_adapter *adapter, bool msix)
 {
-	int i = 0, err = 0, vector = 0, free_vector = 0;
-	struct net_device *netdev = adapter->netdev;
+	struct pci_dev *pdev = adapter->pdev;
+	int err = 0;
 
-	err = request_irq(adapter->msix_entries[vector].vector,
-			  &igc_msix_other, 0, netdev->name, adapter);
-	if (err)
-		goto err_out;
+	igc_set_interrupt_capability(adapter, msix);
 
-	for (i = 0; i < adapter->num_q_vectors; i++) {
-		struct igc_q_vector *q_vector = adapter->q_vector[i];
+	err = igc_alloc_q_vectors(adapter);
+	if (err) {
+		dev_err(&pdev->dev, "Unable to allocate memory for vectors\n");
+		goto err_alloc_q_vectors;
+	}
 
-		vector++;
-
-		q_vector->itr_register = adapter->io_addr + IGC_EITR(vector);
-
-		if (q_vector->rx.ring && q_vector->tx.ring)
-			sprintf(q_vector->name, "%s-TxRx-%u", netdev->name,
-				q_vector->rx.ring->queue_index);
-		else if (q_vector->tx.ring)
-			sprintf(q_vector->name, "%s-tx-%u", netdev->name,
-				q_vector->tx.ring->queue_index);
-		else if (q_vector->rx.ring)
-			sprintf(q_vector->name, "%s-rx-%u", netdev->name,
-				q_vector->rx.ring->queue_index);
-		else
-			sprintf(q_vector->name, "%s-unused", netdev->name);
-
-		err = request_irq(adapter->msix_entries[vector].vector,
-				  igc_msix_ring, 0, q_vector->name,
-				  q_vector);
-		if (err)
-			goto err_free;
-	}
+	igc_cache_ring_register(adapter);
 
-	igc_configure_msix(adapter);
 	return 0;
 
-err_free:
-	/* free already assigned IRQs */
-	free_irq(adapter->msix_entries[free_vector++].vector, adapter);
-
-	vector--;
-	for (i = 0; i < vector; i++) {
-		free_irq(adapter->msix_entries[free_vector++].vector,
-			 adapter->q_vector[i]);
-	}
-err_out:
+err_alloc_q_vectors:
+	igc_reset_interrupt_capability(adapter);
 	return err;
 }
 
 /**
- * igc_reset_q_vector - Reset config for interrupt vector
+ * igc_sw_init - Initialize general software structures (struct igc_adapter)
  * @adapter: board private structure to initialize
- * @v_idx: Index of vector to be reset
  *
- * If NAPI is enabled it will delete any references to the
- * NAPI struct. This is preparation for igc_free_q_vector.
+ * igc_sw_init initializes the Adapter private data structure.
+ * Fields are initialized based on PCI device information and
+ * OS network device settings (MTU size).
  */
-static void igc_reset_q_vector(struct igc_adapter *adapter, int v_idx)
+static int igc_sw_init(struct igc_adapter *adapter)
 {
-	struct igc_q_vector *q_vector = adapter->q_vector[v_idx];
-
-	/* if we're coming from igc_set_interrupt_capability, the vectors are
-	 * not yet allocated
-	 */
-	if (!q_vector)
-		return;
-
-	if (q_vector->tx.ring)
-		adapter->tx_ring[q_vector->tx.ring->queue_index] = NULL;
+	struct net_device *netdev = adapter->netdev;
+	struct pci_dev *pdev = adapter->pdev;
+	struct igc_hw *hw = &adapter->hw;
 
-	if (q_vector->rx.ring)
-		adapter->rx_ring[q_vector->rx.ring->queue_index] = NULL;
+	int size = sizeof(struct igc_mac_addr) * hw->mac.rar_entry_count;
 
-	netif_napi_del(&q_vector->napi);
-}
+	pci_read_config_word(pdev, PCI_COMMAND, &hw->bus.pci_cmd_word);
 
-static void igc_reset_interrupt_capability(struct igc_adapter *adapter)
-{
-	int v_idx = adapter->num_q_vectors;
+	/* set default ring sizes */
+	adapter->tx_ring_count = IGC_DEFAULT_TXD;
+	adapter->rx_ring_count = IGC_DEFAULT_RXD;
 
-	if (adapter->msix_entries) {
-		pci_disable_msix(adapter->pdev);
-		kfree(adapter->msix_entries);
-		adapter->msix_entries = NULL;
-	} else if (adapter->flags & IGC_FLAG_HAS_MSI) {
-		pci_disable_msi(adapter->pdev);
-	}
+	/* set default ITR values */
+	adapter->rx_itr_setting = IGC_DEFAULT_ITR;
+	adapter->tx_itr_setting = IGC_DEFAULT_ITR;
 
-	while (v_idx--)
-		igc_reset_q_vector(adapter, v_idx);
-}
+	/* set default work limits */
+	adapter->tx_work_limit = IGC_DEFAULT_TX_WORK;
 
-/**
- * igc_set_interrupt_capability - set MSI or MSI-X if supported
- * @adapter: Pointer to adapter structure
- * @msix: boolean value for MSI-X capability
- *
- * Attempt to configure interrupts using the best available
- * capabilities of the hardware and kernel.
- */
-static void igc_set_interrupt_capability(struct igc_adapter *adapter,
-					 bool msix)
-{
-	int numvecs, i;
-	int err;
+	/* adjust max frame to be at least the size of a standard frame */
+	adapter->max_frame_size = netdev->mtu + ETH_HLEN + ETH_FCS_LEN +
+				VLAN_HLEN;
+	adapter->min_frame_size = ETH_ZLEN + ETH_FCS_LEN;
 
-	if (!msix)
-		goto msi_only;
+	spin_lock_init(&adapter->nfc_lock);
+	spin_lock_init(&adapter->stats64_lock);
+	/* Assume MSI-X interrupts, will be checked during IRQ allocation */
 	adapter->flags |= IGC_FLAG_HAS_MSIX;
 
-	/* Number of supported queues. */
-	adapter->num_rx_queues = adapter->rss_queues;
+	adapter->mac_table = kzalloc(size, GFP_ATOMIC);
+	if (!adapter->mac_table)
+		return -ENOMEM;
 
-	adapter->num_tx_queues = adapter->rss_queues;
+	igc_init_queue_configuration(adapter);
 
-	/* start with one vector for every Rx queue */
-	numvecs = adapter->num_rx_queues;
+	/* This call may decrease the number of queues */
+	if (igc_init_interrupt_scheme(adapter, true)) {
+		dev_err(&pdev->dev, "Unable to allocate memory for queues\n");
+		return -ENOMEM;
+	}
 
-	/* if Tx handler is separate add 1 for every Tx queue */
-	if (!(adapter->flags & IGC_FLAG_QUEUE_PAIRS))
-		numvecs += adapter->num_tx_queues;
+	/* Explicitly disable IRQ since the NIC can be in any state. */
+	igc_irq_disable(adapter);
 
-	/* store the number of vectors reserved for queues */
-	adapter->num_q_vectors = numvecs;
+	set_bit(__IGC_DOWN, &adapter->state);
 
-	/* add 1 vector for link status interrupts */
-	numvecs++;
+	return 0;
+}
 
-	adapter->msix_entries = kcalloc(numvecs, sizeof(struct msix_entry),
-					GFP_KERNEL);
+/**
+ * igc_up - Open the interface and prepare it to handle traffic
+ * @adapter: board private structure
+ */
+void igc_up(struct igc_adapter *adapter)
+{
+	struct igc_hw *hw = &adapter->hw;
+	int i = 0;
 
-	if (!adapter->msix_entries)
-		return;
+	/* hardware has been reset, we need to reload some things */
+	igc_configure(adapter);
 
-	/* populate entry values */
-	for (i = 0; i < numvecs; i++)
-		adapter->msix_entries[i].entry = i;
+	clear_bit(__IGC_DOWN, &adapter->state);
 
-	err = pci_enable_msix_range(adapter->pdev,
-				    adapter->msix_entries,
-				    numvecs,
-				    numvecs);
-	if (err > 0)
-		return;
+	for (i = 0; i < adapter->num_q_vectors; i++)
+		napi_enable(&adapter->q_vector[i]->napi);
 
-	kfree(adapter->msix_entries);
-	adapter->msix_entries = NULL;
+	if (adapter->msix_entries)
+		igc_configure_msix(adapter);
+	else
+		igc_assign_vector(adapter->q_vector[0], 0);
 
-	igc_reset_interrupt_capability(adapter);
+	/* Clear any pending interrupts. */
+	rd32(IGC_ICR);
+	igc_irq_enable(adapter);
 
-msi_only:
-	adapter->flags &= ~IGC_FLAG_HAS_MSIX;
+	netif_tx_start_all_queues(adapter->netdev);
 
-	adapter->rss_queues = 1;
-	adapter->flags |= IGC_FLAG_QUEUE_PAIRS;
-	adapter->num_rx_queues = 1;
-	adapter->num_tx_queues = 1;
-	adapter->num_q_vectors = 1;
-	if (!pci_enable_msi(adapter->pdev))
-		adapter->flags |= IGC_FLAG_HAS_MSI;
+	/* start the watchdog. */
+	hw->mac.get_link_status = 1;
+	schedule_work(&adapter->watchdog_task);
 }
 
 /**
- * igc_free_q_vector - Free memory allocated for specific interrupt vector
- * @adapter: board private structure to initialize
- * @v_idx: Index of vector to be freed
- *
- * This function frees the memory allocated to the q_vector.
+ * igc_update_stats - Update the board statistics counters
+ * @adapter: board private structure
  */
-static void igc_free_q_vector(struct igc_adapter *adapter, int v_idx)
+void igc_update_stats(struct igc_adapter *adapter)
 {
-	struct igc_q_vector *q_vector = adapter->q_vector[v_idx];
-
-	adapter->q_vector[v_idx] = NULL;
+	struct rtnl_link_stats64 *net_stats = &adapter->stats64;
+	struct pci_dev *pdev = adapter->pdev;
+	struct igc_hw *hw = &adapter->hw;
+	u64 _bytes, _packets;
+	u64 bytes, packets;
+	unsigned int start;
+	u32 mpc;
+	int i;
 
-	/* igc_get_stats64() might access the rings on this vector,
-	 * we must wait a grace period before freeing it.
+	/* Prevent stats update while adapter is being reset, or if the pci
+	 * connection is down.
 	 */
-	if (q_vector)
-		kfree_rcu(q_vector, rcu);
-}
+	if (adapter->link_speed == 0)
+		return;
+	if (pci_channel_offline(pdev))
+		return;
 
-/**
- * igc_free_q_vectors - Free memory allocated for interrupt vectors
- * @adapter: board private structure to initialize
- *
- * This function frees the memory allocated to the q_vectors.  In addition if
- * NAPI is enabled it will delete any references to the NAPI struct prior
- * to freeing the q_vector.
- */
-static void igc_free_q_vectors(struct igc_adapter *adapter)
-{
-	int v_idx = adapter->num_q_vectors;
+	packets = 0;
+	bytes = 0;
 
-	adapter->num_tx_queues = 0;
-	adapter->num_rx_queues = 0;
-	adapter->num_q_vectors = 0;
+	rcu_read_lock();
+	for (i = 0; i < adapter->num_rx_queues; i++) {
+		struct igc_ring *ring = adapter->rx_ring[i];
+		u32 rqdpc = rd32(IGC_RQDPC(i));
 
-	while (v_idx--) {
-		igc_reset_q_vector(adapter, v_idx);
-		igc_free_q_vector(adapter, v_idx);
-	}
-}
+		if (hw->mac.type >= igc_i225)
+			wr32(IGC_RQDPC(i), 0);
 
-/**
- * igc_clear_interrupt_scheme - reset the device to a state of no interrupts
- * @adapter: Pointer to adapter structure
- *
- * This function resets the device so that it has 0 rx queues, tx queues, and
- * MSI-X interrupts allocated.
- */
-static void igc_clear_interrupt_scheme(struct igc_adapter *adapter)
-{
-	igc_free_q_vectors(adapter);
-	igc_reset_interrupt_capability(adapter);
-}
+		if (rqdpc) {
+			ring->rx_stats.drops += rqdpc;
+			net_stats->rx_fifo_errors += rqdpc;
+		}
 
-/* Need to wait a few seconds after link up to get diagnostic information from
- * the phy
- */
-static void igc_update_phy_info(struct timer_list *t)
+		do {
+			start = u64_stats_fetch_begin_irq(&ring->rx_syncp);
+			_bytes = ring->rx_stats.bytes;
+			_packets = ring->rx_stats.packets;
+		} while (u64_stats_fetch_retry_irq(&ring->rx_syncp, start));
+		bytes += _bytes;
+		packets += _packets;
+	}
+
+	net_stats->rx_bytes = bytes;
+	net_stats->rx_packets = packets;
+
+	packets = 0;
+	bytes = 0;
+	for (i = 0; i < adapter->num_tx_queues; i++) {
+		struct igc_ring *ring = adapter->tx_ring[i];
+
+		do {
+			start = u64_stats_fetch_begin_irq(&ring->tx_syncp);
+			_bytes = ring->tx_stats.bytes;
+			_packets = ring->tx_stats.packets;
+		} while (u64_stats_fetch_retry_irq(&ring->tx_syncp, start));
+		bytes += _bytes;
+		packets += _packets;
+	}
+	net_stats->tx_bytes = bytes;
+	net_stats->tx_packets = packets;
+	rcu_read_unlock();
+
+	/* read stats registers */
+	adapter->stats.crcerrs += rd32(IGC_CRCERRS);
+	adapter->stats.gprc += rd32(IGC_GPRC);
+	adapter->stats.gorc += rd32(IGC_GORCL);
+	rd32(IGC_GORCH); /* clear GORCL */
+	adapter->stats.bprc += rd32(IGC_BPRC);
+	adapter->stats.mprc += rd32(IGC_MPRC);
+	adapter->stats.roc += rd32(IGC_ROC);
+
+	adapter->stats.prc64 += rd32(IGC_PRC64);
+	adapter->stats.prc127 += rd32(IGC_PRC127);
+	adapter->stats.prc255 += rd32(IGC_PRC255);
+	adapter->stats.prc511 += rd32(IGC_PRC511);
+	adapter->stats.prc1023 += rd32(IGC_PRC1023);
+	adapter->stats.prc1522 += rd32(IGC_PRC1522);
+	adapter->stats.symerrs += rd32(IGC_SYMERRS);
+	adapter->stats.sec += rd32(IGC_SEC);
+
+	mpc = rd32(IGC_MPC);
+	adapter->stats.mpc += mpc;
+	net_stats->rx_fifo_errors += mpc;
+	adapter->stats.scc += rd32(IGC_SCC);
+	adapter->stats.ecol += rd32(IGC_ECOL);
+	adapter->stats.mcc += rd32(IGC_MCC);
+	adapter->stats.latecol += rd32(IGC_LATECOL);
+	adapter->stats.dc += rd32(IGC_DC);
+	adapter->stats.rlec += rd32(IGC_RLEC);
+	adapter->stats.xonrxc += rd32(IGC_XONRXC);
+	adapter->stats.xontxc += rd32(IGC_XONTXC);
+	adapter->stats.xoffrxc += rd32(IGC_XOFFRXC);
+	adapter->stats.xofftxc += rd32(IGC_XOFFTXC);
+	adapter->stats.fcruc += rd32(IGC_FCRUC);
+	adapter->stats.gptc += rd32(IGC_GPTC);
+	adapter->stats.gotc += rd32(IGC_GOTCL);
+	rd32(IGC_GOTCH); /* clear GOTCL */
+	adapter->stats.rnbc += rd32(IGC_RNBC);
+	adapter->stats.ruc += rd32(IGC_RUC);
+	adapter->stats.rfc += rd32(IGC_RFC);
+	adapter->stats.rjc += rd32(IGC_RJC);
+	adapter->stats.tor += rd32(IGC_TORH);
+	adapter->stats.tot += rd32(IGC_TOTH);
+	adapter->stats.tpr += rd32(IGC_TPR);
+
+	adapter->stats.ptc64 += rd32(IGC_PTC64);
+	adapter->stats.ptc127 += rd32(IGC_PTC127);
+	adapter->stats.ptc255 += rd32(IGC_PTC255);
+	adapter->stats.ptc511 += rd32(IGC_PTC511);
+	adapter->stats.ptc1023 += rd32(IGC_PTC1023);
+	adapter->stats.ptc1522 += rd32(IGC_PTC1522);
+
+	adapter->stats.mptc += rd32(IGC_MPTC);
+	adapter->stats.bptc += rd32(IGC_BPTC);
+
+	adapter->stats.tpt += rd32(IGC_TPT);
+	adapter->stats.colc += rd32(IGC_COLC);
+
+	adapter->stats.algnerrc += rd32(IGC_ALGNERRC);
+
+	adapter->stats.tsctc += rd32(IGC_TSCTC);
+	adapter->stats.tsctfc += rd32(IGC_TSCTFC);
+
+	adapter->stats.iac += rd32(IGC_IAC);
+	adapter->stats.icrxoc += rd32(IGC_ICRXOC);
+	adapter->stats.icrxptc += rd32(IGC_ICRXPTC);
+	adapter->stats.icrxatc += rd32(IGC_ICRXATC);
+	adapter->stats.ictxptc += rd32(IGC_ICTXPTC);
+	adapter->stats.ictxatc += rd32(IGC_ICTXATC);
+	adapter->stats.ictxqec += rd32(IGC_ICTXQEC);
+	adapter->stats.ictxqmtc += rd32(IGC_ICTXQMTC);
+	adapter->stats.icrxdmtc += rd32(IGC_ICRXDMTC);
+
+	/* Fill out the OS statistics structure */
+	net_stats->multicast = adapter->stats.mprc;
+	net_stats->collisions = adapter->stats.colc;
+
+	/* Rx Errors */
+
+	/* RLEC on some newer hardware can be incorrect so build
+	 * our own version based on RUC and ROC
+	 */
+	net_stats->rx_errors = adapter->stats.rxerrc +
+		adapter->stats.crcerrs + adapter->stats.algnerrc +
+		adapter->stats.ruc + adapter->stats.roc +
+		adapter->stats.cexterr;
+	net_stats->rx_length_errors = adapter->stats.ruc +
+				      adapter->stats.roc;
+	net_stats->rx_crc_errors = adapter->stats.crcerrs;
+	net_stats->rx_frame_errors = adapter->stats.algnerrc;
+	net_stats->rx_missed_errors = adapter->stats.mpc;
+
+	/* Tx Errors */
+	net_stats->tx_errors = adapter->stats.ecol +
+			       adapter->stats.latecol;
+	net_stats->tx_aborted_errors = adapter->stats.ecol;
+	net_stats->tx_window_errors = adapter->stats.latecol;
+	net_stats->tx_carrier_errors = adapter->stats.tncrs;
+
+	/* Tx Dropped needs to be maintained elsewhere */
+
+	/* Management Stats */
+	adapter->stats.mgptc += rd32(IGC_MGTPTC);
+	adapter->stats.mgprc += rd32(IGC_MGTPRC);
+	adapter->stats.mgpdc += rd32(IGC_MGTPDC);
+}
+
+static void igc_nfc_filter_exit(struct igc_adapter *adapter)
 {
-	struct igc_adapter *adapter = from_timer(adapter, t, phy_info_timer);
+	struct igc_nfc_filter *rule;
 
-	igc_get_phy_info(&adapter->hw);
+	spin_lock(&adapter->nfc_lock);
+
+	hlist_for_each_entry(rule, &adapter->nfc_filter_list, nfc_node)
+		igc_erase_filter(adapter, rule);
+
+	hlist_for_each_entry(rule, &adapter->cls_flower_list, nfc_node)
+		igc_erase_filter(adapter, rule);
+
+	spin_unlock(&adapter->nfc_lock);
 }
 
 /**
- * igc_has_link - check shared code for link and determine up/down
- * @adapter: pointer to driver private info
+ * igc_down - Close the interface
+ * @adapter: board private structure
  */
-bool igc_has_link(struct igc_adapter *adapter)
+void igc_down(struct igc_adapter *adapter)
 {
+	struct net_device *netdev = adapter->netdev;
 	struct igc_hw *hw = &adapter->hw;
-	bool link_active = false;
+	u32 tctl, rctl;
+	int i = 0;
 
-	/* get_link_status is set on LSC (link status) interrupt or
-	 * rx sequence error interrupt.  get_link_status will stay
-	 * false until the igc_check_for_link establishes link
-	 * for copper adapters ONLY
-	 */
-	switch (hw->phy.media_type) {
-	case igc_media_type_copper:
-		if (!hw->mac.get_link_status)
-			return true;
-		hw->mac.ops.check_for_link(hw);
-		link_active = !hw->mac.get_link_status;
-		break;
-	default:
-	case igc_media_type_unknown:
-		break;
-	}
+	set_bit(__IGC_DOWN, &adapter->state);
 
-	if (hw->mac.type == igc_i225 &&
-	    hw->phy.id == I225_I_PHY_ID) {
-		if (!netif_carrier_ok(adapter->netdev)) {
-			adapter->flags &= ~IGC_FLAG_NEED_LINK_UPDATE;
-		} else if (!(adapter->flags & IGC_FLAG_NEED_LINK_UPDATE)) {
-			adapter->flags |= IGC_FLAG_NEED_LINK_UPDATE;
-			adapter->link_check_timeout = jiffies;
+	/* disable receives in the hardware */
+	rctl = rd32(IGC_RCTL);
+	wr32(IGC_RCTL, rctl & ~IGC_RCTL_EN);
+	/* flush and sleep below */
+
+	igc_nfc_filter_exit(adapter);
+
+	/* set trans_start so we don't get spurious watchdogs during reset */
+	netif_trans_update(netdev);
+
+	netif_carrier_off(netdev);
+	netif_tx_stop_all_queues(netdev);
+
+	/* disable transmits in the hardware */
+	tctl = rd32(IGC_TCTL);
+	tctl &= ~IGC_TCTL_EN;
+	wr32(IGC_TCTL, tctl);
+	/* flush both disables and wait for them to finish */
+	wrfl();
+	usleep_range(10000, 20000);
+
+	igc_irq_disable(adapter);
+
+	adapter->flags &= ~IGC_FLAG_NEED_LINK_UPDATE;
+
+	for (i = 0; i < adapter->num_q_vectors; i++) {
+		if (adapter->q_vector[i]) {
+			napi_synchronize(&adapter->q_vector[i]->napi);
+			napi_disable(&adapter->q_vector[i]->napi);
 		}
 	}
 
-	return link_active;
+	del_timer_sync(&adapter->watchdog_timer);
+	del_timer_sync(&adapter->phy_info_timer);
+
+	/* record the stats before reset*/
+	spin_lock(&adapter->stats64_lock);
+	igc_update_stats(adapter);
+	spin_unlock(&adapter->stats64_lock);
+
+	adapter->link_speed = 0;
+	adapter->link_duplex = 0;
+
+	if (!pci_channel_offline(adapter->pdev))
+		igc_reset(adapter);
+
+	/* clear VLAN promisc flag so VFTA will be updated if necessary */
+	adapter->flags &= ~IGC_FLAG_VLAN_PROMISC;
+
+	igc_clean_all_tx_rings(adapter);
+	igc_clean_all_rx_rings(adapter);
 }
 
-/**
- * igc_watchdog - Timer Call-back
- * @t: timer for the watchdog
- */
-static void igc_watchdog(struct timer_list *t)
+void igc_reinit_locked(struct igc_adapter *adapter)
 {
-	struct igc_adapter *adapter = from_timer(adapter, t, watchdog_timer);
-	/* Do the rest outside of interrupt context */
-	schedule_work(&adapter->watchdog_task);
+	WARN_ON(in_interrupt());
+	while (test_and_set_bit(__IGC_RESETTING, &adapter->state))
+		usleep_range(1000, 2000);
+	igc_down(adapter);
+	igc_up(adapter);
+	clear_bit(__IGC_RESETTING, &adapter->state);
 }
 
-static void igc_watchdog_task(struct work_struct *work)
+static void igc_reset_task(struct work_struct *work)
 {
-	struct igc_adapter *adapter = container_of(work,
-						   struct igc_adapter,
-						   watchdog_task);
-	struct net_device *netdev = adapter->netdev;
-	struct igc_hw *hw = &adapter->hw;
-	struct igc_phy_info *phy = &hw->phy;
-	u16 phy_data, retry_count = 20;
-	u32 connsw;
-	u32 link;
-	int i;
+	struct igc_adapter *adapter;
 
-	link = igc_has_link(adapter);
+	adapter = container_of(work, struct igc_adapter, reset_task);
 
-	if (adapter->flags & IGC_FLAG_NEED_LINK_UPDATE) {
-		if (time_after(jiffies, (adapter->link_check_timeout + HZ)))
-			adapter->flags &= ~IGC_FLAG_NEED_LINK_UPDATE;
-		else
-			link = false;
-	}
+	netdev_err(adapter->netdev, "Reset adapter\n");
+	igc_reinit_locked(adapter);
+}
 
-	/* Force link down if we have fiber to swap to */
-	if (adapter->flags & IGC_FLAG_MAS_ENABLE) {
-		if (hw->phy.media_type == igc_media_type_copper) {
-			connsw = rd32(IGC_CONNSW);
-			if (!(connsw & IGC_CONNSW_AUTOSENSE_EN))
-				link = 0;
-		}
-	}
-	if (link) {
-		if (!netif_carrier_ok(netdev)) {
-			u32 ctrl;
+/**
+ * igc_change_mtu - Change the Maximum Transfer Unit
+ * @netdev: network interface device structure
+ * @new_mtu: new value for maximum frame size
+ *
+ * Returns 0 on success, negative on failure
+ */
+static int igc_change_mtu(struct net_device *netdev, int new_mtu)
+{
+	int max_frame = new_mtu + ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN;
+	struct igc_adapter *adapter = netdev_priv(netdev);
 
-			hw->mac.ops.get_speed_and_duplex(hw,
-							 &adapter->link_speed,
-							 &adapter->link_duplex);
+	/* adjust max frame to be at least the size of a standard frame */
+	if (max_frame < (ETH_FRAME_LEN + ETH_FCS_LEN))
+		max_frame = ETH_FRAME_LEN + ETH_FCS_LEN;
 
-			ctrl = rd32(IGC_CTRL);
-			/* Link status message must follow this format */
-			netdev_info(netdev,
-				    "igc: %s NIC Link is Up %d Mbps %s Duplex, Flow Control: %s\n",
-				    netdev->name,
-				    adapter->link_speed,
-				    adapter->link_duplex == FULL_DUPLEX ?
-				    "Full" : "Half",
-				    (ctrl & IGC_CTRL_TFCE) &&
-				    (ctrl & IGC_CTRL_RFCE) ? "RX/TX" :
-				    (ctrl & IGC_CTRL_RFCE) ?  "RX" :
-				    (ctrl & IGC_CTRL_TFCE) ?  "TX" : "None");
+	while (test_and_set_bit(__IGC_RESETTING, &adapter->state))
+		usleep_range(1000, 2000);
 
-			/* check if SmartSpeed worked */
-			igc_check_downshift(hw);
-			if (phy->speed_downgraded)
-				netdev_warn(netdev, "Link Speed was downgraded by SmartSpeed\n");
+	/* igc_down has a dependency on max_frame_size */
+	adapter->max_frame_size = max_frame;
 
-			/* adjust timeout factor according to speed/duplex */
-			adapter->tx_timeout_factor = 1;
-			switch (adapter->link_speed) {
-			case SPEED_10:
-				adapter->tx_timeout_factor = 14;
-				break;
-			case SPEED_100:
-				/* maybe add some timeout factor ? */
-				break;
-			}
+	if (netif_running(netdev))
+		igc_down(adapter);
 
-			if (adapter->link_speed != SPEED_1000)
-				goto no_wait;
+	netdev_dbg(netdev, "changing MTU from %d to %d\n",
+		   netdev->mtu, new_mtu);
+	netdev->mtu = new_mtu;
 
-			/* wait for Remote receiver status OK */
-retry_read_status:
-			if (!igc_read_phy_reg(hw, PHY_1000T_STATUS,
-					      &phy_data)) {
-				if (!(phy_data & SR_1000T_REMOTE_RX_STATUS) &&
-				    retry_count) {
-					msleep(100);
-					retry_count--;
-					goto retry_read_status;
-				} else if (!retry_count) {
-					dev_err(&adapter->pdev->dev, "exceed max 2 second\n");
-				}
-			} else {
-				dev_err(&adapter->pdev->dev, "read 1000Base-T Status Reg\n");
-			}
-no_wait:
-			netif_carrier_on(netdev);
+	if (netif_running(netdev))
+		igc_up(adapter);
+	else
+		igc_reset(adapter);
 
-			/* link state has changed, schedule phy info update */
-			if (!test_bit(__IGC_DOWN, &adapter->state))
-				mod_timer(&adapter->phy_info_timer,
-					  round_jiffies(jiffies + 2 * HZ));
-		}
-	} else {
-		if (netif_carrier_ok(netdev)) {
-			adapter->link_speed = 0;
-			adapter->link_duplex = 0;
+	clear_bit(__IGC_RESETTING, &adapter->state);
 
-			/* Links status message must follow this format */
-			netdev_info(netdev, "igc: %s NIC Link is Down\n",
-				    netdev->name);
-			netif_carrier_off(netdev);
+	return 0;
+}
 
-			/* link state has changed, schedule phy info update */
-			if (!test_bit(__IGC_DOWN, &adapter->state))
-				mod_timer(&adapter->phy_info_timer,
-					  round_jiffies(jiffies + 2 * HZ));
+/**
+ * igc_get_stats - Get System Network Statistics
+ * @netdev: network interface device structure
+ *
+ * Returns the address of the device statistics structure.
+ * The statistics are updated here and also from the timer callback.
+ */
+static struct net_device_stats *igc_get_stats(struct net_device *netdev)
+{
+	struct igc_adapter *adapter = netdev_priv(netdev);
 
-			/* link is down, time to check for alternate media */
-			if (adapter->flags & IGC_FLAG_MAS_ENABLE) {
-				if (adapter->flags & IGC_FLAG_MEDIA_RESET) {
-					schedule_work(&adapter->reset_task);
-					/* return immediately */
-					return;
-				}
-			}
+	if (!test_bit(__IGC_RESETTING, &adapter->state))
+		igc_update_stats(adapter);
 
-		/* also check for alternate media here */
-		} else if (!netif_carrier_ok(netdev) &&
-			   (adapter->flags & IGC_FLAG_MAS_ENABLE)) {
-			if (adapter->flags & IGC_FLAG_MEDIA_RESET) {
-				schedule_work(&adapter->reset_task);
-				/* return immediately */
-				return;
-			}
-		}
-	}
+	/* only return the current stats */
+	return &netdev->stats;
+}
+
+static netdev_features_t igc_fix_features(struct net_device *netdev,
+					  netdev_features_t features)
+{
+	/* Since there is no support for separate Rx/Tx vlan accel
+	 * enable/disable make sure Tx flag is always in same state as Rx.
+	 */
+	if (features & NETIF_F_HW_VLAN_CTAG_RX)
+		features |= NETIF_F_HW_VLAN_CTAG_TX;
+	else
+		features &= ~NETIF_F_HW_VLAN_CTAG_TX;
+
+	return features;
+}
+
+static int igc_set_features(struct net_device *netdev,
+			    netdev_features_t features)
+{
+	netdev_features_t changed = netdev->features ^ features;
+	struct igc_adapter *adapter = netdev_priv(netdev);
 
-	spin_lock(&adapter->stats64_lock);
-	igc_update_stats(adapter);
-	spin_unlock(&adapter->stats64_lock);
+	/* Add VLAN support */
+	if (!(changed & (NETIF_F_RXALL | NETIF_F_NTUPLE)))
+		return 0;
 
-	for (i = 0; i < adapter->num_tx_queues; i++) {
-		struct igc_ring *tx_ring = adapter->tx_ring[i];
+	if (!(features & NETIF_F_NTUPLE)) {
+		struct hlist_node *node2;
+		struct igc_nfc_filter *rule;
 
-		if (!netif_carrier_ok(netdev)) {
-			/* We've lost link, so the controller stops DMA,
-			 * but we've got queued Tx work that's never going
-			 * to get done, so reset controller to flush Tx.
-			 * (Do the reset outside of interrupt context).
-			 */
-			if (igc_desc_unused(tx_ring) + 1 < tx_ring->count) {
-				adapter->tx_timeout_count++;
-				schedule_work(&adapter->reset_task);
-				/* return immediately since reset is imminent */
-				return;
-			}
+		spin_lock(&adapter->nfc_lock);
+		hlist_for_each_entry_safe(rule, node2,
+					  &adapter->nfc_filter_list, nfc_node) {
+			igc_erase_filter(adapter, rule);
+			hlist_del(&rule->nfc_node);
+			kfree(rule);
 		}
-
-		/* Force detection of hung controller every watchdog period */
-		set_bit(IGC_RING_FLAG_TX_DETECT_HANG, &tx_ring->flags);
+		spin_unlock(&adapter->nfc_lock);
+		adapter->nfc_filter_count = 0;
 	}
 
-	/* Cause software interrupt to ensure Rx ring is cleaned */
-	if (adapter->flags & IGC_FLAG_HAS_MSIX) {
-		u32 eics = 0;
+	netdev->features = features;
 
-		for (i = 0; i < adapter->num_q_vectors; i++)
-			eics |= adapter->q_vector[i]->eims_value;
-		wr32(IGC_EICS, eics);
-	} else {
-		wr32(IGC_ICS, IGC_ICS_RXDMT0);
-	}
+	if (netif_running(netdev))
+		igc_reinit_locked(adapter);
+	else
+		igc_reset(adapter);
 
-	/* Reset the timer */
-	if (!test_bit(__IGC_DOWN, &adapter->state)) {
-		if (adapter->flags & IGC_FLAG_NEED_LINK_UPDATE)
-			mod_timer(&adapter->watchdog_timer,
-				  round_jiffies(jiffies +  HZ));
-		else
-			mod_timer(&adapter->watchdog_timer,
-				  round_jiffies(jiffies + 2 * HZ));
-	}
+	return 1;
 }
 
-/**
- * igc_update_ring_itr - update the dynamic ITR value based on packet size
- * @q_vector: pointer to q_vector
- *
- * Stores a new ITR value based on strictly on packet size.  This
- * algorithm is less sophisticated than that used in igc_update_itr,
- * due to the difficulty of synchronizing statistics across multiple
- * receive rings.  The divisors and thresholds used by this function
- * were determined based on theoretical maximum wire speed and testing
- * data, in order to minimize response time while increasing bulk
- * throughput.
- * NOTE: This function is called only when operating in a multiqueue
- * receive environment.
- */
-static void igc_update_ring_itr(struct igc_q_vector *q_vector)
+static netdev_features_t
+igc_features_check(struct sk_buff *skb, struct net_device *dev,
+		   netdev_features_t features)
 {
-	struct igc_adapter *adapter = q_vector->adapter;
-	int new_val = q_vector->itr_val;
-	int avg_wire_size = 0;
-	unsigned int packets;
+	unsigned int network_hdr_len, mac_hdr_len;
 
-	/* For non-gigabit speeds, just fix the interrupt rate at 4000
-	 * ints/sec - ITR timer value of 120 ticks.
-	 */
-	switch (adapter->link_speed) {
-	case SPEED_10:
-	case SPEED_100:
-		new_val = IGC_4K_ITR;
-		goto set_itr_val;
-	default:
-		break;
-	}
+	/* Make certain the headers can be described by a context descriptor */
+	mac_hdr_len = skb_network_header(skb) - skb->data;
+	if (unlikely(mac_hdr_len > IGC_MAX_MAC_HDR_LEN))
+		return features & ~(NETIF_F_HW_CSUM |
+				    NETIF_F_SCTP_CRC |
+				    NETIF_F_HW_VLAN_CTAG_TX |
+				    NETIF_F_TSO |
+				    NETIF_F_TSO6);
 
-	packets = q_vector->rx.total_packets;
-	if (packets)
-		avg_wire_size = q_vector->rx.total_bytes / packets;
+	network_hdr_len = skb_checksum_start(skb) - skb_network_header(skb);
+	if (unlikely(network_hdr_len >  IGC_MAX_NETWORK_HDR_LEN))
+		return features & ~(NETIF_F_HW_CSUM |
+				    NETIF_F_SCTP_CRC |
+				    NETIF_F_TSO |
+				    NETIF_F_TSO6);
 
-	packets = q_vector->tx.total_packets;
-	if (packets)
-		avg_wire_size = max_t(u32, avg_wire_size,
-				      q_vector->tx.total_bytes / packets);
+	/* We can only support IPv4 TSO in tunnels if we can mangle the
+	 * inner IP ID field, so strip TSO if MANGLEID is not supported.
+	 */
+	if (skb->encapsulation && !(features & NETIF_F_TSO_MANGLEID))
+		features &= ~NETIF_F_TSO;
 
-	/* if avg_wire_size isn't set no work was done */
-	if (!avg_wire_size)
-		goto clear_counts;
+	return features;
+}
 
-	/* Add 24 bytes to size to account for CRC, preamble, and gap */
-	avg_wire_size += 24;
+/* Add a MAC filter for 'addr' directing matching traffic to 'queue',
+ * 'flags' is used to indicate what kind of match is made, match is by
+ * default for the destination address, if matching by source address
+ * is desired the flag IGC_MAC_STATE_SRC_ADDR can be used.
+ */
+static int igc_add_mac_filter_flags(struct igc_adapter *adapter,
+				    const u8 *addr, const u8 queue,
+				    const u8 flags)
+{
+	struct igc_hw *hw = &adapter->hw;
+	int rar_entries = hw->mac.rar_entry_count;
+	int i;
 
-	/* Don't starve jumbo frames */
-	avg_wire_size = min(avg_wire_size, 3000);
+	if (is_zero_ether_addr(addr))
+		return -EINVAL;
 
-	/* Give a little boost to mid-size frames */
-	if (avg_wire_size > 300 && avg_wire_size < 1200)
-		new_val = avg_wire_size / 3;
-	else
-		new_val = avg_wire_size / 2;
+	/* Search for the first empty entry in the MAC table.
+	 * Do not touch entries at the end of the table reserved for the VF MAC
+	 * addresses.
+	 */
+	for (i = 0; i < rar_entries; i++) {
+		if (!igc_mac_entry_can_be_used(&adapter->mac_table[i],
+					       addr, flags))
+			continue;
 
-	/* conservative mode (itr 3) eliminates the lowest_latency setting */
-	if (new_val < IGC_20K_ITR &&
-	    ((q_vector->rx.ring && adapter->rx_itr_setting == 3) ||
-	    (!q_vector->rx.ring && adapter->tx_itr_setting == 3)))
-		new_val = IGC_20K_ITR;
+		ether_addr_copy(adapter->mac_table[i].addr, addr);
+		adapter->mac_table[i].queue = queue;
+		adapter->mac_table[i].state |= IGC_MAC_STATE_IN_USE | flags;
 
-set_itr_val:
-	if (new_val != q_vector->itr_val) {
-		q_vector->itr_val = new_val;
-		q_vector->set_itr = 1;
+		igc_rar_set_index(adapter, i);
+		return i;
 	}
-clear_counts:
-	q_vector->rx.total_bytes = 0;
-	q_vector->rx.total_packets = 0;
-	q_vector->tx.total_bytes = 0;
-	q_vector->tx.total_packets = 0;
+
+	return -ENOSPC;
 }
 
-/**
- * igc_update_itr - update the dynamic ITR value based on statistics
- * @q_vector: pointer to q_vector
- * @ring_container: ring info to update the itr for
- *
- * Stores a new ITR value based on packets and byte
- * counts during the last interrupt.  The advantage of per interrupt
- * computation is faster updates and more accurate ITR for the current
- * traffic pattern.  Constants in this function were computed
- * based on theoretical maximum wire speed and thresholds were set based
- * on testing data as well as attempting to minimize response time
- * while increasing bulk throughput.
- * NOTE: These calculations are only valid when operating in a single-
- * queue environment.
+int igc_add_mac_steering_filter(struct igc_adapter *adapter,
+				const u8 *addr, u8 queue, u8 flags)
+{
+	return igc_add_mac_filter_flags(adapter, addr, queue,
+					IGC_MAC_STATE_QUEUE_STEERING | flags);
+}
+
+/* Remove a MAC filter for 'addr' directing matching traffic to
+ * 'queue', 'flags' is used to indicate what kind of match need to be
+ * removed, match is by default for the destination address, if
+ * matching by source address is to be removed the flag
+ * IGC_MAC_STATE_SRC_ADDR can be used.
  */
-static void igc_update_itr(struct igc_q_vector *q_vector,
-			   struct igc_ring_container *ring_container)
+static int igc_del_mac_filter_flags(struct igc_adapter *adapter,
+				    const u8 *addr, const u8 queue,
+				    const u8 flags)
 {
-	unsigned int packets = ring_container->total_packets;
-	unsigned int bytes = ring_container->total_bytes;
-	u8 itrval = ring_container->itr;
+	struct igc_hw *hw = &adapter->hw;
+	int rar_entries = hw->mac.rar_entry_count;
+	int i;
 
-	/* no packets, exit with status unchanged */
-	if (packets == 0)
-		return;
+	if (is_zero_ether_addr(addr))
+		return -EINVAL;
 
-	switch (itrval) {
-	case lowest_latency:
-		/* handle TSO and jumbo frames */
-		if (bytes / packets > 8000)
-			itrval = bulk_latency;
-		else if ((packets < 5) && (bytes > 512))
-			itrval = low_latency;
-		break;
-	case low_latency:  /* 50 usec aka 20000 ints/s */
-		if (bytes > 10000) {
-			/* this if handles the TSO accounting */
-			if (bytes / packets > 8000)
-				itrval = bulk_latency;
-			else if ((packets < 10) || ((bytes / packets) > 1200))
-				itrval = bulk_latency;
-			else if ((packets > 35))
-				itrval = lowest_latency;
-		} else if (bytes / packets > 2000) {
-			itrval = bulk_latency;
-		} else if (packets <= 2 && bytes < 512) {
-			itrval = lowest_latency;
-		}
-		break;
-	case bulk_latency: /* 250 usec aka 4000 ints/s */
-		if (bytes > 25000) {
-			if (packets > 35)
-				itrval = low_latency;
-		} else if (bytes < 1500) {
-			itrval = low_latency;
+	/* Search for matching entry in the MAC table based on given address
+	 * and queue. Do not touch entries at the end of the table reserved
+	 * for the VF MAC addresses.
+	 */
+	for (i = 0; i < rar_entries; i++) {
+		if (!(adapter->mac_table[i].state & IGC_MAC_STATE_IN_USE))
+			continue;
+		if ((adapter->mac_table[i].state & flags) != flags)
+			continue;
+		if (adapter->mac_table[i].queue != queue)
+			continue;
+		if (!ether_addr_equal(adapter->mac_table[i].addr, addr))
+			continue;
+
+		/* When a filter for the default address is "deleted",
+		 * we return it to its initial configuration
+		 */
+		if (adapter->mac_table[i].state & IGC_MAC_STATE_DEFAULT) {
+			adapter->mac_table[i].state =
+				IGC_MAC_STATE_DEFAULT | IGC_MAC_STATE_IN_USE;
+		} else {
+			adapter->mac_table[i].state = 0;
+			adapter->mac_table[i].queue = 0;
+			memset(adapter->mac_table[i].addr, 0, ETH_ALEN);
 		}
-		break;
+
+		igc_rar_set_index(adapter, i);
+		return 0;
 	}
 
-	/* clear work counters since we have the values we need */
-	ring_container->total_bytes = 0;
-	ring_container->total_packets = 0;
+	return -ENOENT;
+}
 
-	/* write updated itr to ring container */
-	ring_container->itr = itrval;
+int igc_del_mac_steering_filter(struct igc_adapter *adapter,
+				const u8 *addr, u8 queue, u8 flags)
+{
+	return igc_del_mac_filter_flags(adapter, addr, queue,
+					IGC_MAC_STATE_QUEUE_STEERING | flags);
 }
 
 /**
- * igc_intr_msi - Interrupt Handler
+ * igc_msix_other - msix other interrupt handler
  * @irq: interrupt number
- * @data: pointer to a network interface device structure
+ * @data: pointer to a q_vector
  */
-static irqreturn_t igc_intr_msi(int irq, void *data)
+static irqreturn_t igc_msix_other(int irq, void *data)
 {
 	struct igc_adapter *adapter = data;
-	struct igc_q_vector *q_vector = adapter->q_vector[0];
 	struct igc_hw *hw = &adapter->hw;
-	/* read ICR disables interrupts using IAM */
 	u32 icr = rd32(IGC_ICR);
 
-	igc_write_itr(q_vector);
-
+	/* reading ICR causes bit 31 of EICR to be cleared */
 	if (icr & IGC_ICR_DRSTA)
 		schedule_work(&adapter->reset_task);
 
@@ -3586,408 +3655,428 @@ static irqreturn_t igc_intr_msi(int irq, void *data)
 		adapter->stats.doosync++;
 	}
 
-	if (icr & (IGC_ICR_RXSEQ | IGC_ICR_LSC)) {
+	if (icr & IGC_ICR_LSC) {
 		hw->mac.get_link_status = 1;
+		/* guard against interrupt when we're going down */
 		if (!test_bit(__IGC_DOWN, &adapter->state))
 			mod_timer(&adapter->watchdog_timer, jiffies + 1);
 	}
 
-	napi_schedule(&q_vector->napi);
+	wr32(IGC_EIMS, adapter->eims_other);
 
 	return IRQ_HANDLED;
 }
 
-/**
- * igc_intr - Legacy Interrupt Handler
- * @irq: interrupt number
- * @data: pointer to a network interface device structure
- */
-static irqreturn_t igc_intr(int irq, void *data)
+static void igc_write_itr(struct igc_q_vector *q_vector)
 {
-	struct igc_adapter *adapter = data;
-	struct igc_q_vector *q_vector = adapter->q_vector[0];
-	struct igc_hw *hw = &adapter->hw;
-	/* Interrupt Auto-Mask...upon reading ICR, interrupts are masked.  No
-	 * need for the IMC write
-	 */
-	u32 icr = rd32(IGC_ICR);
+	u32 itr_val = q_vector->itr_val & IGC_QVECTOR_MASK;
 
-	/* IMS will not auto-mask if INT_ASSERTED is not set, and if it is
-	 * not set, then the adapter didn't send an interrupt
-	 */
-	if (!(icr & IGC_ICR_INT_ASSERTED))
-		return IRQ_NONE;
+	if (!q_vector->set_itr)
+		return;
 
-	igc_write_itr(q_vector);
+	if (!itr_val)
+		itr_val = IGC_ITR_VAL_MASK;
 
-	if (icr & IGC_ICR_DRSTA)
-		schedule_work(&adapter->reset_task);
+	itr_val |= IGC_EITR_CNT_IGNR;
 
-	if (icr & IGC_ICR_DOUTSYNC) {
-		/* HW is reporting DMA is out of sync */
-		adapter->stats.doosync++;
-	}
+	writel(itr_val, q_vector->itr_register);
+	q_vector->set_itr = 0;
+}
 
-	if (icr & (IGC_ICR_RXSEQ | IGC_ICR_LSC)) {
-		hw->mac.get_link_status = 1;
-		/* guard against interrupt when we're going down */
-		if (!test_bit(__IGC_DOWN, &adapter->state))
-			mod_timer(&adapter->watchdog_timer, jiffies + 1);
-	}
+static irqreturn_t igc_msix_ring(int irq, void *data)
+{
+	struct igc_q_vector *q_vector = data;
+
+	/* Write the ITR value calculated from the previous interrupt. */
+	igc_write_itr(q_vector);
 
 	napi_schedule(&q_vector->napi);
 
 	return IRQ_HANDLED;
 }
 
-static void igc_set_itr(struct igc_q_vector *q_vector)
+/**
+ * igc_request_msix - Initialize MSI-X interrupts
+ * @adapter: Pointer to adapter structure
+ *
+ * igc_request_msix allocates MSI-X vectors and requests interrupts from the
+ * kernel.
+ */
+static int igc_request_msix(struct igc_adapter *adapter)
 {
-	struct igc_adapter *adapter = q_vector->adapter;
-	u32 new_itr = q_vector->itr_val;
-	u8 current_itr = 0;
+	int i = 0, err = 0, vector = 0, free_vector = 0;
+	struct net_device *netdev = adapter->netdev;
 
-	/* for non-gigabit speeds, just fix the interrupt rate at 4000 */
-	switch (adapter->link_speed) {
-	case SPEED_10:
-	case SPEED_100:
-		current_itr = 0;
-		new_itr = IGC_4K_ITR;
-		goto set_itr_now;
-	default:
-		break;
-	}
+	err = request_irq(adapter->msix_entries[vector].vector,
+			  &igc_msix_other, 0, netdev->name, adapter);
+	if (err)
+		goto err_out;
 
-	igc_update_itr(q_vector, &q_vector->tx);
-	igc_update_itr(q_vector, &q_vector->rx);
+	for (i = 0; i < adapter->num_q_vectors; i++) {
+		struct igc_q_vector *q_vector = adapter->q_vector[i];
 
-	current_itr = max(q_vector->rx.itr, q_vector->tx.itr);
+		vector++;
 
-	/* conservative mode (itr 3) eliminates the lowest_latency setting */
-	if (current_itr == lowest_latency &&
-	    ((q_vector->rx.ring && adapter->rx_itr_setting == 3) ||
-	    (!q_vector->rx.ring && adapter->tx_itr_setting == 3)))
-		current_itr = low_latency;
+		q_vector->itr_register = adapter->io_addr + IGC_EITR(vector);
 
-	switch (current_itr) {
-	/* counts and packets in update_itr are dependent on these numbers */
-	case lowest_latency:
-		new_itr = IGC_70K_ITR; /* 70,000 ints/sec */
-		break;
-	case low_latency:
-		new_itr = IGC_20K_ITR; /* 20,000 ints/sec */
-		break;
-	case bulk_latency:
-		new_itr = IGC_4K_ITR;  /* 4,000 ints/sec */
-		break;
-	default:
-		break;
-	}
+		if (q_vector->rx.ring && q_vector->tx.ring)
+			sprintf(q_vector->name, "%s-TxRx-%u", netdev->name,
+				q_vector->rx.ring->queue_index);
+		else if (q_vector->tx.ring)
+			sprintf(q_vector->name, "%s-tx-%u", netdev->name,
+				q_vector->tx.ring->queue_index);
+		else if (q_vector->rx.ring)
+			sprintf(q_vector->name, "%s-rx-%u", netdev->name,
+				q_vector->rx.ring->queue_index);
+		else
+			sprintf(q_vector->name, "%s-unused", netdev->name);
 
-set_itr_now:
-	if (new_itr != q_vector->itr_val) {
-		/* this attempts to bias the interrupt rate towards Bulk
-		 * by adding intermediate steps when interrupt rate is
-		 * increasing
-		 */
-		new_itr = new_itr > q_vector->itr_val ?
-			  max((new_itr * q_vector->itr_val) /
-			  (new_itr + (q_vector->itr_val >> 2)),
-			  new_itr) : new_itr;
-		/* Don't write the value here; it resets the adapter's
-		 * internal timer, and causes us to delay far longer than
-		 * we should between interrupts.  Instead, we write the ITR
-		 * value at the beginning of the next interrupt so the timing
-		 * ends up being correct.
-		 */
-		q_vector->itr_val = new_itr;
-		q_vector->set_itr = 1;
+		err = request_irq(adapter->msix_entries[vector].vector,
+				  igc_msix_ring, 0, q_vector->name,
+				  q_vector);
+		if (err)
+			goto err_free;
 	}
-}
 
-static void igc_ring_irq_enable(struct igc_q_vector *q_vector)
-{
-	struct igc_adapter *adapter = q_vector->adapter;
-	struct igc_hw *hw = &adapter->hw;
+	igc_configure_msix(adapter);
+	return 0;
 
-	if ((q_vector->rx.ring && (adapter->rx_itr_setting & 3)) ||
-	    (!q_vector->rx.ring && (adapter->tx_itr_setting & 3))) {
-		if (adapter->num_q_vectors == 1)
-			igc_set_itr(q_vector);
-		else
-			igc_update_ring_itr(q_vector);
-	}
+err_free:
+	/* free already assigned IRQs */
+	free_irq(adapter->msix_entries[free_vector++].vector, adapter);
 
-	if (!test_bit(__IGC_DOWN, &adapter->state)) {
-		if (adapter->msix_entries)
-			wr32(IGC_EIMS, q_vector->eims_value);
-		else
-			igc_irq_enable(adapter);
+	vector--;
+	for (i = 0; i < vector; i++) {
+		free_irq(adapter->msix_entries[free_vector++].vector,
+			 adapter->q_vector[i]);
 	}
+err_out:
+	return err;
 }
 
 /**
- * igc_poll - NAPI Rx polling callback
- * @napi: napi polling structure
- * @budget: count of how many packets we should handle
+ * igc_clear_interrupt_scheme - reset the device to a state of no interrupts
+ * @adapter: Pointer to adapter structure
+ *
+ * This function resets the device so that it has 0 rx queues, tx queues, and
+ * MSI-X interrupts allocated.
  */
-static int igc_poll(struct napi_struct *napi, int budget)
+static void igc_clear_interrupt_scheme(struct igc_adapter *adapter)
 {
-	struct igc_q_vector *q_vector = container_of(napi,
-						     struct igc_q_vector,
-						     napi);
-	bool clean_complete = true;
-	int work_done = 0;
-
-	if (q_vector->tx.ring)
-		clean_complete = igc_clean_tx_irq(q_vector, budget);
-
-	if (q_vector->rx.ring) {
-		int cleaned = igc_clean_rx_irq(q_vector, budget);
-
-		work_done += cleaned;
-		if (cleaned >= budget)
-			clean_complete = false;
-	}
-
-	/* If all work not completed, return budget and keep polling */
-	if (!clean_complete)
-		return budget;
+	igc_free_q_vectors(adapter);
+	igc_reset_interrupt_capability(adapter);
+}
 
-	/* Exit the polling mode, but don't re-enable interrupts if stack might
-	 * poll us due to busy-polling
-	 */
-	if (likely(napi_complete_done(napi, work_done)))
-		igc_ring_irq_enable(q_vector);
+/* Need to wait a few seconds after link up to get diagnostic information from
+ * the phy
+ */
+static void igc_update_phy_info(struct timer_list *t)
+{
+	struct igc_adapter *adapter = from_timer(adapter, t, phy_info_timer);
 
-	return min(work_done, budget - 1);
+	igc_get_phy_info(&adapter->hw);
 }
 
-static void igc_add_ring(struct igc_ring *ring,
-			 struct igc_ring_container *head)
+/**
+ * igc_has_link - check shared code for link and determine up/down
+ * @adapter: pointer to driver private info
+ */
+bool igc_has_link(struct igc_adapter *adapter)
 {
-	head->ring = ring;
-	head->count++;
+	struct igc_hw *hw = &adapter->hw;
+	bool link_active = false;
+
+	/* get_link_status is set on LSC (link status) interrupt or
+	 * rx sequence error interrupt.  get_link_status will stay
+	 * false until the igc_check_for_link establishes link
+	 * for copper adapters ONLY
+	 */
+	switch (hw->phy.media_type) {
+	case igc_media_type_copper:
+		if (!hw->mac.get_link_status)
+			return true;
+		hw->mac.ops.check_for_link(hw);
+		link_active = !hw->mac.get_link_status;
+		break;
+	default:
+	case igc_media_type_unknown:
+		break;
+	}
+
+	if (hw->mac.type == igc_i225 &&
+	    hw->phy.id == I225_I_PHY_ID) {
+		if (!netif_carrier_ok(adapter->netdev)) {
+			adapter->flags &= ~IGC_FLAG_NEED_LINK_UPDATE;
+		} else if (!(adapter->flags & IGC_FLAG_NEED_LINK_UPDATE)) {
+			adapter->flags |= IGC_FLAG_NEED_LINK_UPDATE;
+			adapter->link_check_timeout = jiffies;
+		}
+	}
+
+	return link_active;
 }
 
 /**
- * igc_alloc_q_vector - Allocate memory for a single interrupt vector
- * @adapter: board private structure to initialize
- * @v_count: q_vectors allocated on adapter, used for ring interleaving
- * @v_idx: index of vector in adapter struct
- * @txr_count: total number of Tx rings to allocate
- * @txr_idx: index of first Tx ring to allocate
- * @rxr_count: total number of Rx rings to allocate
- * @rxr_idx: index of first Rx ring to allocate
- *
- * We allocate one q_vector.  If allocation fails we return -ENOMEM.
+ * igc_watchdog - Timer Call-back
+ * @t: timer for the watchdog
  */
-static int igc_alloc_q_vector(struct igc_adapter *adapter,
-			      unsigned int v_count, unsigned int v_idx,
-			      unsigned int txr_count, unsigned int txr_idx,
-			      unsigned int rxr_count, unsigned int rxr_idx)
+static void igc_watchdog(struct timer_list *t)
 {
-	struct igc_q_vector *q_vector;
-	struct igc_ring *ring;
-	int ring_count;
+	struct igc_adapter *adapter = from_timer(adapter, t, watchdog_timer);
+	/* Do the rest outside of interrupt context */
+	schedule_work(&adapter->watchdog_task);
+}
 
-	/* igc only supports 1 Tx and/or 1 Rx queue per vector */
-	if (txr_count > 1 || rxr_count > 1)
-		return -ENOMEM;
+static void igc_watchdog_task(struct work_struct *work)
+{
+	struct igc_adapter *adapter = container_of(work,
+						   struct igc_adapter,
+						   watchdog_task);
+	struct net_device *netdev = adapter->netdev;
+	struct igc_hw *hw = &adapter->hw;
+	struct igc_phy_info *phy = &hw->phy;
+	u16 phy_data, retry_count = 20;
+	u32 connsw;
+	u32 link;
+	int i;
 
-	ring_count = txr_count + rxr_count;
+	link = igc_has_link(adapter);
 
-	/* allocate q_vector and rings */
-	q_vector = adapter->q_vector[v_idx];
-	if (!q_vector)
-		q_vector = kzalloc(struct_size(q_vector, ring, ring_count),
-				   GFP_KERNEL);
-	else
-		memset(q_vector, 0, struct_size(q_vector, ring, ring_count));
-	if (!q_vector)
-		return -ENOMEM;
+	if (adapter->flags & IGC_FLAG_NEED_LINK_UPDATE) {
+		if (time_after(jiffies, (adapter->link_check_timeout + HZ)))
+			adapter->flags &= ~IGC_FLAG_NEED_LINK_UPDATE;
+		else
+			link = false;
+	}
 
-	/* initialize NAPI */
-	netif_napi_add(adapter->netdev, &q_vector->napi,
-		       igc_poll, 64);
+	/* Force link down if we have fiber to swap to */
+	if (adapter->flags & IGC_FLAG_MAS_ENABLE) {
+		if (hw->phy.media_type == igc_media_type_copper) {
+			connsw = rd32(IGC_CONNSW);
+			if (!(connsw & IGC_CONNSW_AUTOSENSE_EN))
+				link = 0;
+		}
+	}
+	if (link) {
+		if (!netif_carrier_ok(netdev)) {
+			u32 ctrl;
 
-	/* tie q_vector and adapter together */
-	adapter->q_vector[v_idx] = q_vector;
-	q_vector->adapter = adapter;
+			hw->mac.ops.get_speed_and_duplex(hw,
+							 &adapter->link_speed,
+							 &adapter->link_duplex);
 
-	/* initialize work limits */
-	q_vector->tx.work_limit = adapter->tx_work_limit;
+			ctrl = rd32(IGC_CTRL);
+			/* Link status message must follow this format */
+			netdev_info(netdev,
+				    "igc: %s NIC Link is Up %d Mbps %s Duplex, Flow Control: %s\n",
+				    netdev->name,
+				    adapter->link_speed,
+				    adapter->link_duplex == FULL_DUPLEX ?
+				    "Full" : "Half",
+				    (ctrl & IGC_CTRL_TFCE) &&
+				    (ctrl & IGC_CTRL_RFCE) ? "RX/TX" :
+				    (ctrl & IGC_CTRL_RFCE) ?  "RX" :
+				    (ctrl & IGC_CTRL_TFCE) ?  "TX" : "None");
 
-	/* initialize ITR configuration */
-	q_vector->itr_register = adapter->io_addr + IGC_EITR(0);
-	q_vector->itr_val = IGC_START_ITR;
+			/* check if SmartSpeed worked */
+			igc_check_downshift(hw);
+			if (phy->speed_downgraded)
+				netdev_warn(netdev, "Link Speed was downgraded by SmartSpeed\n");
 
-	/* initialize pointer to rings */
-	ring = q_vector->ring;
+			/* adjust timeout factor according to speed/duplex */
+			adapter->tx_timeout_factor = 1;
+			switch (adapter->link_speed) {
+			case SPEED_10:
+				adapter->tx_timeout_factor = 14;
+				break;
+			case SPEED_100:
+				/* maybe add some timeout factor ? */
+				break;
+			}
 
-	/* initialize ITR */
-	if (rxr_count) {
-		/* rx or rx/tx vector */
-		if (!adapter->rx_itr_setting || adapter->rx_itr_setting > 3)
-			q_vector->itr_val = adapter->rx_itr_setting;
-	} else {
-		/* tx only vector */
-		if (!adapter->tx_itr_setting || adapter->tx_itr_setting > 3)
-			q_vector->itr_val = adapter->tx_itr_setting;
-	}
+			if (adapter->link_speed != SPEED_1000)
+				goto no_wait;
 
-	if (txr_count) {
-		/* assign generic ring traits */
-		ring->dev = &adapter->pdev->dev;
-		ring->netdev = adapter->netdev;
+			/* wait for Remote receiver status OK */
+retry_read_status:
+			if (!igc_read_phy_reg(hw, PHY_1000T_STATUS,
+					      &phy_data)) {
+				if (!(phy_data & SR_1000T_REMOTE_RX_STATUS) &&
+				    retry_count) {
+					msleep(100);
+					retry_count--;
+					goto retry_read_status;
+				} else if (!retry_count) {
+					dev_err(&adapter->pdev->dev, "exceed max 2 second\n");
+				}
+			} else {
+				dev_err(&adapter->pdev->dev, "read 1000Base-T Status Reg\n");
+			}
+no_wait:
+			netif_carrier_on(netdev);
 
-		/* configure backlink on ring */
-		ring->q_vector = q_vector;
+			/* link state has changed, schedule phy info update */
+			if (!test_bit(__IGC_DOWN, &adapter->state))
+				mod_timer(&adapter->phy_info_timer,
+					  round_jiffies(jiffies + 2 * HZ));
+		}
+	} else {
+		if (netif_carrier_ok(netdev)) {
+			adapter->link_speed = 0;
+			adapter->link_duplex = 0;
 
-		/* update q_vector Tx values */
-		igc_add_ring(ring, &q_vector->tx);
+			/* Links status message must follow this format */
+			netdev_info(netdev, "igc: %s NIC Link is Down\n",
+				    netdev->name);
+			netif_carrier_off(netdev);
 
-		/* apply Tx specific ring traits */
-		ring->count = adapter->tx_ring_count;
-		ring->queue_index = txr_idx;
+			/* link state has changed, schedule phy info update */
+			if (!test_bit(__IGC_DOWN, &adapter->state))
+				mod_timer(&adapter->phy_info_timer,
+					  round_jiffies(jiffies + 2 * HZ));
 
-		/* assign ring to adapter */
-		adapter->tx_ring[txr_idx] = ring;
+			/* link is down, time to check for alternate media */
+			if (adapter->flags & IGC_FLAG_MAS_ENABLE) {
+				if (adapter->flags & IGC_FLAG_MEDIA_RESET) {
+					schedule_work(&adapter->reset_task);
+					/* return immediately */
+					return;
+				}
+			}
 
-		/* push pointer to next ring */
-		ring++;
+		/* also check for alternate media here */
+		} else if (!netif_carrier_ok(netdev) &&
+			   (adapter->flags & IGC_FLAG_MAS_ENABLE)) {
+			if (adapter->flags & IGC_FLAG_MEDIA_RESET) {
+				schedule_work(&adapter->reset_task);
+				/* return immediately */
+				return;
+			}
+		}
 	}
 
-	if (rxr_count) {
-		/* assign generic ring traits */
-		ring->dev = &adapter->pdev->dev;
-		ring->netdev = adapter->netdev;
+	spin_lock(&adapter->stats64_lock);
+	igc_update_stats(adapter);
+	spin_unlock(&adapter->stats64_lock);
 
-		/* configure backlink on ring */
-		ring->q_vector = q_vector;
+	for (i = 0; i < adapter->num_tx_queues; i++) {
+		struct igc_ring *tx_ring = adapter->tx_ring[i];
+
+		if (!netif_carrier_ok(netdev)) {
+			/* We've lost link, so the controller stops DMA,
+			 * but we've got queued Tx work that's never going
+			 * to get done, so reset controller to flush Tx.
+			 * (Do the reset outside of interrupt context).
+			 */
+			if (igc_desc_unused(tx_ring) + 1 < tx_ring->count) {
+				adapter->tx_timeout_count++;
+				schedule_work(&adapter->reset_task);
+				/* return immediately since reset is imminent */
+				return;
+			}
+		}
 
-		/* update q_vector Rx values */
-		igc_add_ring(ring, &q_vector->rx);
+		/* Force detection of hung controller every watchdog period */
+		set_bit(IGC_RING_FLAG_TX_DETECT_HANG, &tx_ring->flags);
+	}
 
-		/* apply Rx specific ring traits */
-		ring->count = adapter->rx_ring_count;
-		ring->queue_index = rxr_idx;
+	/* Cause software interrupt to ensure Rx ring is cleaned */
+	if (adapter->flags & IGC_FLAG_HAS_MSIX) {
+		u32 eics = 0;
 
-		/* assign ring to adapter */
-		adapter->rx_ring[rxr_idx] = ring;
+		for (i = 0; i < adapter->num_q_vectors; i++)
+			eics |= adapter->q_vector[i]->eims_value;
+		wr32(IGC_EICS, eics);
+	} else {
+		wr32(IGC_ICS, IGC_ICS_RXDMT0);
 	}
 
-	return 0;
+	/* Reset the timer */
+	if (!test_bit(__IGC_DOWN, &adapter->state)) {
+		if (adapter->flags & IGC_FLAG_NEED_LINK_UPDATE)
+			mod_timer(&adapter->watchdog_timer,
+				  round_jiffies(jiffies +  HZ));
+		else
+			mod_timer(&adapter->watchdog_timer,
+				  round_jiffies(jiffies + 2 * HZ));
+	}
 }
 
 /**
- * igc_alloc_q_vectors - Allocate memory for interrupt vectors
- * @adapter: board private structure to initialize
- *
- * We allocate one q_vector per queue interrupt.  If allocation fails we
- * return -ENOMEM.
+ * igc_intr_msi - Interrupt Handler
+ * @irq: interrupt number
+ * @data: pointer to a network interface device structure
  */
-static int igc_alloc_q_vectors(struct igc_adapter *adapter)
+static irqreturn_t igc_intr_msi(int irq, void *data)
 {
-	int rxr_remaining = adapter->num_rx_queues;
-	int txr_remaining = adapter->num_tx_queues;
-	int rxr_idx = 0, txr_idx = 0, v_idx = 0;
-	int q_vectors = adapter->num_q_vectors;
-	int err;
+	struct igc_adapter *adapter = data;
+	struct igc_q_vector *q_vector = adapter->q_vector[0];
+	struct igc_hw *hw = &adapter->hw;
+	/* read ICR disables interrupts using IAM */
+	u32 icr = rd32(IGC_ICR);
 
-	if (q_vectors >= (rxr_remaining + txr_remaining)) {
-		for (; rxr_remaining; v_idx++) {
-			err = igc_alloc_q_vector(adapter, q_vectors, v_idx,
-						 0, 0, 1, rxr_idx);
+	igc_write_itr(q_vector);
 
-			if (err)
-				goto err_out;
+	if (icr & IGC_ICR_DRSTA)
+		schedule_work(&adapter->reset_task);
 
-			/* update counts and index */
-			rxr_remaining--;
-			rxr_idx++;
-		}
+	if (icr & IGC_ICR_DOUTSYNC) {
+		/* HW is reporting DMA is out of sync */
+		adapter->stats.doosync++;
 	}
 
-	for (; v_idx < q_vectors; v_idx++) {
-		int rqpv = DIV_ROUND_UP(rxr_remaining, q_vectors - v_idx);
-		int tqpv = DIV_ROUND_UP(txr_remaining, q_vectors - v_idx);
-
-		err = igc_alloc_q_vector(adapter, q_vectors, v_idx,
-					 tqpv, txr_idx, rqpv, rxr_idx);
-
-		if (err)
-			goto err_out;
-
-		/* update counts and index */
-		rxr_remaining -= rqpv;
-		txr_remaining -= tqpv;
-		rxr_idx++;
-		txr_idx++;
+	if (icr & (IGC_ICR_RXSEQ | IGC_ICR_LSC)) {
+		hw->mac.get_link_status = 1;
+		if (!test_bit(__IGC_DOWN, &adapter->state))
+			mod_timer(&adapter->watchdog_timer, jiffies + 1);
 	}
 
-	return 0;
-
-err_out:
-	adapter->num_tx_queues = 0;
-	adapter->num_rx_queues = 0;
-	adapter->num_q_vectors = 0;
-
-	while (v_idx--)
-		igc_free_q_vector(adapter, v_idx);
+	napi_schedule(&q_vector->napi);
 
-	return -ENOMEM;
+	return IRQ_HANDLED;
 }
 
 /**
- * igc_cache_ring_register - Descriptor ring to register mapping
- * @adapter: board private structure to initialize
- *
- * Once we know the feature-set enabled for the device, we'll cache
- * the register offset the descriptor ring is assigned to.
+ * igc_intr - Legacy Interrupt Handler
+ * @irq: interrupt number
+ * @data: pointer to a network interface device structure
  */
-static void igc_cache_ring_register(struct igc_adapter *adapter)
+static irqreturn_t igc_intr(int irq, void *data)
 {
-	int i = 0, j = 0;
+	struct igc_adapter *adapter = data;
+	struct igc_q_vector *q_vector = adapter->q_vector[0];
+	struct igc_hw *hw = &adapter->hw;
+	/* Interrupt Auto-Mask...upon reading ICR, interrupts are masked.  No
+	 * need for the IMC write
+	 */
+	u32 icr = rd32(IGC_ICR);
 
-	switch (adapter->hw.mac.type) {
-	case igc_i225:
-	/* Fall through */
-	default:
-		for (; i < adapter->num_rx_queues; i++)
-			adapter->rx_ring[i]->reg_idx = i;
-		for (; j < adapter->num_tx_queues; j++)
-			adapter->tx_ring[j]->reg_idx = j;
-		break;
-	}
-}
+	/* IMS will not auto-mask if INT_ASSERTED is not set, and if it is
+	 * not set, then the adapter didn't send an interrupt
+	 */
+	if (!(icr & IGC_ICR_INT_ASSERTED))
+		return IRQ_NONE;
 
-/**
- * igc_init_interrupt_scheme - initialize interrupts, allocate queues/vectors
- * @adapter: Pointer to adapter structure
- * @msix: boolean for MSI-X capability
- *
- * This function initializes the interrupts and allocates all of the queues.
- */
-static int igc_init_interrupt_scheme(struct igc_adapter *adapter, bool msix)
-{
-	struct pci_dev *pdev = adapter->pdev;
-	int err = 0;
+	igc_write_itr(q_vector);
 
-	igc_set_interrupt_capability(adapter, msix);
+	if (icr & IGC_ICR_DRSTA)
+		schedule_work(&adapter->reset_task);
 
-	err = igc_alloc_q_vectors(adapter);
-	if (err) {
-		dev_err(&pdev->dev, "Unable to allocate memory for vectors\n");
-		goto err_alloc_q_vectors;
+	if (icr & IGC_ICR_DOUTSYNC) {
+		/* HW is reporting DMA is out of sync */
+		adapter->stats.doosync++;
 	}
 
-	igc_cache_ring_register(adapter);
+	if (icr & (IGC_ICR_RXSEQ | IGC_ICR_LSC)) {
+		hw->mac.get_link_status = 1;
+		/* guard against interrupt when we're going down */
+		if (!test_bit(__IGC_DOWN, &adapter->state))
+			mod_timer(&adapter->watchdog_timer, jiffies + 1);
+	}
 
-	return 0;
+	napi_schedule(&q_vector->napi);
 
-err_alloc_q_vectors:
-	igc_reset_interrupt_capability(adapter);
-	return err;
+	return IRQ_HANDLED;
 }
 
 static void igc_free_irq(struct igc_adapter *adapter)
@@ -4776,98 +4865,6 @@ static struct pci_driver igc_driver = {
 	.shutdown = igc_shutdown,
 };
 
-void igc_set_flag_queue_pairs(struct igc_adapter *adapter,
-			      const u32 max_rss_queues)
-{
-	/* Determine if we need to pair queues. */
-	/* If rss_queues > half of max_rss_queues, pair the queues in
-	 * order to conserve interrupts due to limited supply.
-	 */
-	if (adapter->rss_queues > (max_rss_queues / 2))
-		adapter->flags |= IGC_FLAG_QUEUE_PAIRS;
-	else
-		adapter->flags &= ~IGC_FLAG_QUEUE_PAIRS;
-}
-
-unsigned int igc_get_max_rss_queues(struct igc_adapter *adapter)
-{
-	unsigned int max_rss_queues;
-
-	/* Determine the maximum number of RSS queues supported. */
-	max_rss_queues = IGC_MAX_RX_QUEUES;
-
-	return max_rss_queues;
-}
-
-static void igc_init_queue_configuration(struct igc_adapter *adapter)
-{
-	u32 max_rss_queues;
-
-	max_rss_queues = igc_get_max_rss_queues(adapter);
-	adapter->rss_queues = min_t(u32, max_rss_queues, num_online_cpus());
-
-	igc_set_flag_queue_pairs(adapter, max_rss_queues);
-}
-
-/**
- * igc_sw_init - Initialize general software structures (struct igc_adapter)
- * @adapter: board private structure to initialize
- *
- * igc_sw_init initializes the Adapter private data structure.
- * Fields are initialized based on PCI device information and
- * OS network device settings (MTU size).
- */
-static int igc_sw_init(struct igc_adapter *adapter)
-{
-	struct net_device *netdev = adapter->netdev;
-	struct pci_dev *pdev = adapter->pdev;
-	struct igc_hw *hw = &adapter->hw;
-
-	int size = sizeof(struct igc_mac_addr) * hw->mac.rar_entry_count;
-
-	pci_read_config_word(pdev, PCI_COMMAND, &hw->bus.pci_cmd_word);
-
-	/* set default ring sizes */
-	adapter->tx_ring_count = IGC_DEFAULT_TXD;
-	adapter->rx_ring_count = IGC_DEFAULT_RXD;
-
-	/* set default ITR values */
-	adapter->rx_itr_setting = IGC_DEFAULT_ITR;
-	adapter->tx_itr_setting = IGC_DEFAULT_ITR;
-
-	/* set default work limits */
-	adapter->tx_work_limit = IGC_DEFAULT_TX_WORK;
-
-	/* adjust max frame to be at least the size of a standard frame */
-	adapter->max_frame_size = netdev->mtu + ETH_HLEN + ETH_FCS_LEN +
-				VLAN_HLEN;
-	adapter->min_frame_size = ETH_ZLEN + ETH_FCS_LEN;
-
-	spin_lock_init(&adapter->nfc_lock);
-	spin_lock_init(&adapter->stats64_lock);
-	/* Assume MSI-X interrupts, will be checked during IRQ allocation */
-	adapter->flags |= IGC_FLAG_HAS_MSIX;
-
-	adapter->mac_table = kzalloc(size, GFP_ATOMIC);
-	if (!adapter->mac_table)
-		return -ENOMEM;
-
-	igc_init_queue_configuration(adapter);
-
-	/* This call may decrease the number of queues */
-	if (igc_init_interrupt_scheme(adapter, true)) {
-		dev_err(&pdev->dev, "Unable to allocate memory for queues\n");
-		return -ENOMEM;
-	}
-
-	/* Explicitly disable IRQ since the NIC can be in any state. */
-	igc_irq_disable(adapter);
-
-	set_bit(__IGC_DOWN, &adapter->state);
-
-	return 0;
-}
-
 /**
  * igc_reinit_queues - return error
  * @adapter: pointer to adapter structure
-- 
cgit v1.2.3-59-g8ed1b


From 9a69a4c8802adf642bc4a13d471b5a86b44ed434 Mon Sep 17 00:00:00 2001
From: "Jason A. Donenfeld" <Jason@zx2c4.com>
Date: Thu, 2 Jan 2020 17:47:49 +0100
Subject: wireguard: selftests: remove ancient kernel compatibility code

Quite a bit of the test suite was designed to work with ancient kernels.
Thankfully we no longer have to deal with this. This commit updates
things that we can finally update and removes things that we can finally
remove, to avoid the build-up of the last several years as a result of
having to support ancient kernels. We can finally rely on suppress_
prefixlength being available. On the build side of things, the no-PIE
hack is no longer required, and we can bump some of the tools, repair
our m68k and i686-kvm support, and get better coverage of the static
branches used in the crypto lib and in udp_tunnel.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/wireguard/netns.sh         | 11 ++-
 tools/testing/selftests/wireguard/qemu/Makefile    | 82 +++++++++++-----------
 .../selftests/wireguard/qemu/arch/m68k.config      |  2 +-
 tools/testing/selftests/wireguard/qemu/init.c      |  1 +
 .../testing/selftests/wireguard/qemu/kernel.config |  2 +
 5 files changed, 50 insertions(+), 48 deletions(-)

diff --git a/tools/testing/selftests/wireguard/netns.sh b/tools/testing/selftests/wireguard/netns.sh
index e7310d9390f7..d5c85c7494f2 100755
--- a/tools/testing/selftests/wireguard/netns.sh
+++ b/tools/testing/selftests/wireguard/netns.sh
@@ -37,7 +37,7 @@ n2() { pretty 2 "$*"; maybe_exec ip netns exec $netns2 "$@"; }
 ip0() { pretty 0 "ip $*"; ip -n $netns0 "$@"; }
 ip1() { pretty 1 "ip $*"; ip -n $netns1 "$@"; }
 ip2() { pretty 2 "ip $*"; ip -n $netns2 "$@"; }
-sleep() { read -t "$1" -N 0 || true; }
+sleep() { read -t "$1" -N 1 || true; }
 waitiperf() { pretty "${1//*-}" "wait for iperf:5201"; while [[ $(ss -N "$1" -tlp 'sport = 5201') != *iperf3* ]]; do sleep 0.1; done; }
 waitncatudp() { pretty "${1//*-}" "wait for udp:1111"; while [[ $(ss -N "$1" -ulp 'sport = 1111') != *ncat* ]]; do sleep 0.1; done; }
 waitncattcp() { pretty "${1//*-}" "wait for tcp:1111"; while [[ $(ss -N "$1" -tlp 'sport = 1111') != *ncat* ]]; do sleep 0.1; done; }
@@ -294,12 +294,9 @@ ip1 -6 rule add table main suppress_prefixlength 0
 ip1 -4 route add default dev wg0 table 51820
 ip1 -4 rule add not fwmark 51820 table 51820
 ip1 -4 rule add table main suppress_prefixlength 0
-# suppress_prefixlength only got added in 3.12, and we want to support 3.10+.
-if [[ $(ip1 -4 rule show all) == *suppress_prefixlength* ]]; then
-	# Flood the pings instead of sending just one, to trigger routing table reference counting bugs.
-	n1 ping -W 1 -c 100 -f 192.168.99.7
-	n1 ping -W 1 -c 100 -f abab::1111
-fi
+# Flood the pings instead of sending just one, to trigger routing table reference counting bugs.
+n1 ping -W 1 -c 100 -f 192.168.99.7
+n1 ping -W 1 -c 100 -f abab::1111
 
 n0 iptables -t nat -F
 ip0 link del vethrc
diff --git a/tools/testing/selftests/wireguard/qemu/Makefile b/tools/testing/selftests/wireguard/qemu/Makefile
index 6d51bf78eeff..f10aa3590adc 100644
--- a/tools/testing/selftests/wireguard/qemu/Makefile
+++ b/tools/testing/selftests/wireguard/qemu/Makefile
@@ -5,6 +5,7 @@
 PWD := $(shell pwd)
 
 CHOST := $(shell gcc -dumpmachine)
+HOST_ARCH := $(firstword $(subst -, ,$(CHOST)))
 ifneq (,$(ARCH))
 CBUILD := $(subst -gcc,,$(lastword $(subst /, ,$(firstword $(wildcard $(foreach bindir,$(subst :, ,$(PATH)),$(bindir)/$(ARCH)-*-gcc))))))
 ifeq (,$(CBUILD))
@@ -37,19 +38,19 @@ endef
 define file_download =
 $(DISTFILES_PATH)/$(1):
 	mkdir -p $(DISTFILES_PATH)
-	flock -x $$@.lock -c '[ -f $$@ ] && exit 0; wget -O $$@.tmp $(MIRROR)$(1) || wget -t inf --retry-on-http-error=404 -O $$@.tmp $(2)$(1) || rm -f $$@.tmp'
+	flock -x $$@.lock -c '[ -f $$@ ] && exit 0; wget -O $$@.tmp $(MIRROR)$(1) || wget -O $$@.tmp $(2)$(1) || rm -f $$@.tmp'
 	if echo "$(3)  $$@.tmp" | sha256sum -c -; then mv $$@.tmp $$@; else rm -f $$@.tmp; exit 71; fi
 endef
 
-$(eval $(call tar_download,MUSL,musl,1.1.20,.tar.gz,https://www.musl-libc.org/releases/,44be8771d0e6c6b5f82dd15662eb2957c9a3173a19a8b49966ac0542bbd40d61))
+$(eval $(call tar_download,MUSL,musl,1.1.24,.tar.gz,https://www.musl-libc.org/releases/,1370c9a812b2cf2a7d92802510cca0058cc37e66a7bedd70051f0a34015022a3))
 $(eval $(call tar_download,LIBMNL,libmnl,1.0.4,.tar.bz2,https://www.netfilter.org/projects/libmnl/files/,171f89699f286a5854b72b91d06e8f8e3683064c5901fb09d954a9ab6f551f81))
-$(eval $(call tar_download,IPERF,iperf,3.1.7,.tar.gz,http://downloads.es.net/pub/iperf/,a4ef73406fe92250602b8da2ae89ec53211f805df97a1d1d629db5a14043734f))
+$(eval $(call tar_download,IPERF,iperf,3.7,.tar.gz,https://downloads.es.net/pub/iperf/,d846040224317caf2f75c843d309a950a7db23f9b44b94688ccbe557d6d1710c))
 $(eval $(call tar_download,BASH,bash,5.0,.tar.gz,https://ftp.gnu.org/gnu/bash/,b4a80f2ac66170b2913efbfb9f2594f1f76c7b1afd11f799e22035d63077fb4d))
-$(eval $(call tar_download,IPROUTE2,iproute2,5.1.0,.tar.gz,https://www.kernel.org/pub/linux/utils/net/iproute2/,9b43707d6075ecdca14803ca8ce0c8553848c49fa1586d12fd508d66577243f2))
-$(eval $(call tar_download,IPTABLES,iptables,1.6.1,.tar.bz2,https://www.netfilter.org/projects/iptables/files/,0fc2d7bd5d7be11311726466789d4c65fb4c8e096c9182b56ce97440864f0cf5))
-$(eval $(call tar_download,NMAP,nmap,7.60,.tar.bz2,https://nmap.org/dist/,a8796ecc4fa6c38aad6139d9515dc8113023a82e9d787e5a5fb5fa1b05516f21))
-$(eval $(call tar_download,IPUTILS,iputils,s20161105,.tar.gz,https://github.com/iputils/iputils/archive/s20161105.tar.gz/#,f813092f03d17294fd23544b129b95cdb87fe19f7970a51908a6b88509acad8a))
-$(eval $(call tar_download,WIREGUARD_TOOLS,WireGuard,0.0.20191212,.tar.xz,https://git.zx2c4.com/WireGuard/snapshot/,b0d718380f7a8822b2f12d75e462fa4eafa3a77871002981f367cd4fe2a1b071))
+$(eval $(call tar_download,IPROUTE2,iproute2,5.4.0,.tar.xz,https://www.kernel.org/pub/linux/utils/net/iproute2/,fe97aa60a0d4c5ac830be18937e18dc3400ca713a33a89ad896ff1e3d46086ae))
+$(eval $(call tar_download,IPTABLES,iptables,1.8.4,.tar.bz2,https://www.netfilter.org/projects/iptables/files/,993a3a5490a544c2cbf2ef15cf7e7ed21af1845baf228318d5c36ef8827e157c))
+$(eval $(call tar_download,NMAP,nmap,7.80,.tar.bz2,https://nmap.org/dist/,fcfa5a0e42099e12e4bf7a68ebe6fde05553383a682e816a7ec9256ab4773faa))
+$(eval $(call tar_download,IPUTILS,iputils,s20190709,.tar.gz,https://github.com/iputils/iputils/archive/s20190709.tar.gz/#,a15720dd741d7538dd2645f9f516d193636ae4300ff7dbc8bfca757bf166490a))
+$(eval $(call tar_download,WIREGUARD_TOOLS,wireguard-tools,1.0.20191226,.tar.xz,https://git.zx2c4.com/wireguard-tools/snapshot/,aa8af0fdc9872d369d8c890a84dbc2a2466b55795dccd5b47721b2d97644b04f))
 
 KERNEL_BUILD_PATH := $(BUILD_PATH)/kernel$(if $(findstring yes,$(DEBUG_KERNEL)),-debug)
 rwildcard=$(foreach d,$(wildcard $1*),$(call rwildcard,$d/,$2) $(filter $(subst *,%,$2),$d))
@@ -59,23 +60,21 @@ export CFLAGS ?= -O3 -pipe
 export LDFLAGS ?=
 export CPPFLAGS := -I$(BUILD_PATH)/include
 
-ifeq ($(CHOST),$(CBUILD))
+ifeq ($(HOST_ARCH),$(ARCH))
 CROSS_COMPILE_FLAG := --host=$(CHOST)
-NOPIE_GCC := gcc -fno-PIE
 CFLAGS += -march=native
 STRIP := strip
 else
 $(info Cross compilation: building for $(CBUILD) using $(CHOST))
 CROSS_COMPILE_FLAG := --build=$(CBUILD) --host=$(CHOST)
 export CROSS_COMPILE=$(CBUILD)-
-NOPIE_GCC := $(CBUILD)-gcc -fno-PIE
 STRIP := $(CBUILD)-strip
 endif
 ifeq ($(ARCH),aarch64)
 QEMU_ARCH := aarch64
 KERNEL_ARCH := arm64
 KERNEL_BZIMAGE := $(KERNEL_BUILD_PATH)/arch/arm64/boot/Image
-ifeq ($(CHOST),$(CBUILD))
+ifeq ($(HOST_ARCH),$(ARCH))
 QEMU_MACHINE := -cpu host -machine virt,gic_version=host,accel=kvm
 else
 QEMU_MACHINE := -cpu cortex-a53 -machine virt
@@ -85,7 +84,7 @@ else ifeq ($(ARCH),aarch64_be)
 QEMU_ARCH := aarch64
 KERNEL_ARCH := arm64
 KERNEL_BZIMAGE := $(KERNEL_BUILD_PATH)/arch/arm64/boot/Image
-ifeq ($(CHOST),$(CBUILD))
+ifeq ($(HOST_ARCH),$(ARCH))
 QEMU_MACHINE := -cpu host -machine virt,gic_version=host,accel=kvm
 else
 QEMU_MACHINE := -cpu cortex-a53 -machine virt
@@ -95,7 +94,7 @@ else ifeq ($(ARCH),arm)
 QEMU_ARCH := arm
 KERNEL_ARCH := arm
 KERNEL_BZIMAGE := $(KERNEL_BUILD_PATH)/arch/arm/boot/zImage
-ifeq ($(CHOST),$(CBUILD))
+ifeq ($(HOST_ARCH),$(ARCH))
 QEMU_MACHINE := -cpu host -machine virt,gic_version=host,accel=kvm
 else
 QEMU_MACHINE := -cpu cortex-a15 -machine virt
@@ -105,7 +104,7 @@ else ifeq ($(ARCH),armeb)
 QEMU_ARCH := arm
 KERNEL_ARCH := arm
 KERNEL_BZIMAGE := $(KERNEL_BUILD_PATH)/arch/arm/boot/zImage
-ifeq ($(CHOST),$(CBUILD))
+ifeq ($(HOST_ARCH),$(ARCH))
 QEMU_MACHINE := -cpu host -machine virt,gic_version=host,accel=kvm
 else
 QEMU_MACHINE := -cpu cortex-a15 -machine virt
@@ -116,7 +115,7 @@ else ifeq ($(ARCH),x86_64)
 QEMU_ARCH := x86_64
 KERNEL_ARCH := x86_64
 KERNEL_BZIMAGE := $(KERNEL_BUILD_PATH)/arch/x86/boot/bzImage
-ifeq ($(CHOST),$(CBUILD))
+ifeq ($(HOST_ARCH),$(ARCH))
 QEMU_MACHINE := -cpu host -machine q35,accel=kvm
 else
 QEMU_MACHINE := -cpu Skylake-Server -machine q35
@@ -126,7 +125,7 @@ else ifeq ($(ARCH),i686)
 QEMU_ARCH := i386
 KERNEL_ARCH := x86
 KERNEL_BZIMAGE := $(KERNEL_BUILD_PATH)/arch/x86/boot/bzImage
-ifeq ($(subst i686,x86_64,$(CBUILD)),$(CHOST))
+ifeq ($(subst x86_64,i686,$(HOST_ARCH)),$(ARCH))
 QEMU_MACHINE := -cpu host -machine q35,accel=kvm
 else
 QEMU_MACHINE := -cpu coreduo -machine q35
@@ -136,7 +135,7 @@ else ifeq ($(ARCH),mips64)
 QEMU_ARCH := mips64
 KERNEL_ARCH := mips
 KERNEL_BZIMAGE := $(KERNEL_BUILD_PATH)/vmlinux
-ifeq ($(CHOST),$(CBUILD))
+ifeq ($(HOST_ARCH),$(ARCH))
 QEMU_MACHINE := -cpu host -machine malta,accel=kvm
 CFLAGS += -EB
 else
@@ -147,7 +146,7 @@ else ifeq ($(ARCH),mips64el)
 QEMU_ARCH := mips64el
 KERNEL_ARCH := mips
 KERNEL_BZIMAGE := $(KERNEL_BUILD_PATH)/vmlinux
-ifeq ($(CHOST),$(CBUILD))
+ifeq ($(HOST_ARCH),$(ARCH))
 QEMU_MACHINE := -cpu host -machine malta,accel=kvm
 CFLAGS += -EL
 else
@@ -158,7 +157,7 @@ else ifeq ($(ARCH),mips)
 QEMU_ARCH := mips
 KERNEL_ARCH := mips
 KERNEL_BZIMAGE := $(KERNEL_BUILD_PATH)/vmlinux
-ifeq ($(CHOST),$(CBUILD))
+ifeq ($(HOST_ARCH),$(ARCH))
 QEMU_MACHINE := -cpu host -machine malta,accel=kvm
 CFLAGS += -EB
 else
@@ -169,7 +168,7 @@ else ifeq ($(ARCH),mipsel)
 QEMU_ARCH := mipsel
 KERNEL_ARCH := mips
 KERNEL_BZIMAGE := $(KERNEL_BUILD_PATH)/vmlinux
-ifeq ($(CHOST),$(CBUILD))
+ifeq ($(HOST_ARCH),$(ARCH))
 QEMU_MACHINE := -cpu host -machine malta,accel=kvm
 CFLAGS += -EL
 else
@@ -180,7 +179,7 @@ else ifeq ($(ARCH),powerpc64le)
 QEMU_ARCH := ppc64
 KERNEL_ARCH := powerpc
 KERNEL_BZIMAGE := $(KERNEL_BUILD_PATH)/vmlinux
-ifeq ($(CHOST),$(CBUILD))
+ifeq ($(HOST_ARCH),$(ARCH))
 QEMU_MACHINE := -cpu host,accel=kvm -machine pseries
 else
 QEMU_MACHINE := -machine pseries
@@ -190,7 +189,7 @@ else ifeq ($(ARCH),powerpc)
 QEMU_ARCH := ppc
 KERNEL_ARCH := powerpc
 KERNEL_BZIMAGE := $(KERNEL_BUILD_PATH)/arch/powerpc/boot/uImage
-ifeq ($(CHOST),$(CBUILD))
+ifeq ($(HOST_ARCH),$(ARCH))
 QEMU_MACHINE := -cpu host,accel=kvm -machine ppce500
 else
 QEMU_MACHINE := -machine ppce500
@@ -200,10 +199,11 @@ else ifeq ($(ARCH),m68k)
 QEMU_ARCH := m68k
 KERNEL_ARCH := m68k
 KERNEL_BZIMAGE := $(KERNEL_BUILD_PATH)/vmlinux
-ifeq ($(CHOST),$(CBUILD))
-QEMU_MACHINE := -cpu host,accel=kvm -machine q800
+KERNEL_CMDLINE := $(shell sed -n 's/CONFIG_CMDLINE=\(.*\)/\1/p' arch/m68k.config)
+ifeq ($(HOST_ARCH),$(ARCH))
+QEMU_MACHINE := -cpu host,accel=kvm -machine q800 -smp 1 -append $(KERNEL_CMDLINE)
 else
-QEMU_MACHINE := -machine q800
+QEMU_MACHINE := -machine q800 -smp 1 -append $(KERNEL_CMDLINE)
 endif
 else
 $(error I only build: x86_64, i686, arm, armeb, aarch64, aarch64_be, mips, mipsel, mips64, mips64el, powerpc64le, powerpc, m68k)
@@ -238,14 +238,14 @@ $(BUILD_PATH)/init-cpio-spec.txt:
 	echo "nod /dev/console 644 0 0 c 5 1" >> $@
 	echo "dir /bin 755 0 0" >> $@
 	echo "file /bin/iperf3 $(IPERF_PATH)/src/iperf3 755 0 0" >> $@
-	echo "file /bin/wg $(WIREGUARD_TOOLS_PATH)/src/tools/wg 755 0 0" >> $@
+	echo "file /bin/wg $(WIREGUARD_TOOLS_PATH)/src/wg 755 0 0" >> $@
 	echo "file /bin/bash $(BASH_PATH)/bash 755 0 0" >> $@
 	echo "file /bin/ip $(IPROUTE2_PATH)/ip/ip 755 0 0" >> $@
 	echo "file /bin/ss $(IPROUTE2_PATH)/misc/ss 755 0 0" >> $@
 	echo "file /bin/ping $(IPUTILS_PATH)/ping 755 0 0" >> $@
 	echo "file /bin/ncat $(NMAP_PATH)/ncat/ncat 755 0 0" >> $@
-	echo "file /bin/xtables-multi $(IPTABLES_PATH)/iptables/xtables-multi 755 0 0" >> $@
-	echo "slink /bin/iptables xtables-multi 777 0 0" >> $@
+	echo "file /bin/xtables-legacy-multi $(IPTABLES_PATH)/iptables/xtables-legacy-multi 755 0 0" >> $@
+	echo "slink /bin/iptables xtables-legacy-multi 777 0 0" >> $@
 	echo "slink /bin/ping6 ping 777 0 0" >> $@
 	echo "dir /lib 755 0 0" >> $@
 	echo "file /lib/libc.so $(MUSL_PATH)/lib/libc.so 755 0 0" >> $@
@@ -260,8 +260,8 @@ $(KERNEL_BUILD_PATH)/.config: kernel.config arch/$(ARCH).config
 	cd $(KERNEL_BUILD_PATH) && ARCH=$(KERNEL_ARCH) $(KERNEL_PATH)/scripts/kconfig/merge_config.sh -n $(KERNEL_BUILD_PATH)/.config $(KERNEL_BUILD_PATH)/minimal.config
 	$(if $(findstring yes,$(DEBUG_KERNEL)),cp debug.config $(KERNEL_BUILD_PATH) && cd $(KERNEL_BUILD_PATH) && ARCH=$(KERNEL_ARCH) $(KERNEL_PATH)/scripts/kconfig/merge_config.sh -n $(KERNEL_BUILD_PATH)/.config debug.config,)
 
-$(KERNEL_BZIMAGE): $(KERNEL_BUILD_PATH)/.config $(BUILD_PATH)/init-cpio-spec.txt $(MUSL_PATH)/lib/libc.so $(IPERF_PATH)/src/iperf3 $(IPUTILS_PATH)/ping $(BASH_PATH)/bash $(IPROUTE2_PATH)/misc/ss $(IPROUTE2_PATH)/ip/ip $(IPTABLES_PATH)/iptables/xtables-multi $(NMAP_PATH)/ncat/ncat $(WIREGUARD_TOOLS_PATH)/src/tools/wg $(BUILD_PATH)/init ../netns.sh $(WIREGUARD_SOURCES)
-	$(MAKE) -C $(KERNEL_PATH) O=$(KERNEL_BUILD_PATH) ARCH=$(KERNEL_ARCH) CROSS_COMPILE=$(CROSS_COMPILE) CC="$(NOPIE_GCC)"
+$(KERNEL_BZIMAGE): $(KERNEL_BUILD_PATH)/.config $(BUILD_PATH)/init-cpio-spec.txt $(MUSL_PATH)/lib/libc.so $(IPERF_PATH)/src/iperf3 $(IPUTILS_PATH)/ping $(BASH_PATH)/bash $(IPROUTE2_PATH)/misc/ss $(IPROUTE2_PATH)/ip/ip $(IPTABLES_PATH)/iptables/xtables-legacy-multi $(NMAP_PATH)/ncat/ncat $(WIREGUARD_TOOLS_PATH)/src/wg $(BUILD_PATH)/init ../netns.sh $(WIREGUARD_SOURCES)
+	$(MAKE) -C $(KERNEL_PATH) O=$(KERNEL_BUILD_PATH) ARCH=$(KERNEL_ARCH) CROSS_COMPILE=$(CROSS_COMPILE)
 
 $(BUILD_PATH)/include/linux/.installed: | $(KERNEL_BUILD_PATH)/.config
 	$(MAKE) -C $(KERNEL_PATH) O=$(KERNEL_BUILD_PATH) INSTALL_HDR_PATH=$(BUILD_PATH) ARCH=$(KERNEL_ARCH) CROSS_COMPILE=$(CROSS_COMPILE) headers_install
@@ -280,7 +280,7 @@ $(BUILD_PATH)/include/.installed: $(MUSL_PATH)/lib/libc.so
 
 $(MUSL_CC): $(MUSL_PATH)/lib/libc.so
 	sh $(MUSL_PATH)/tools/musl-gcc.specs.sh $(BUILD_PATH)/include $(MUSL_PATH)/lib /lib/ld-linux.so.1 > $(BUILD_PATH)/musl-gcc.specs
-	printf '#!/bin/sh\nexec "$(REAL_CC)" --specs="$(BUILD_PATH)/musl-gcc.specs" -fno-stack-protector -no-pie "$$@"\n' > $(BUILD_PATH)/musl-gcc
+	printf '#!/bin/sh\nexec "$(REAL_CC)" --specs="$(BUILD_PATH)/musl-gcc.specs" "$$@"\n' > $(BUILD_PATH)/musl-gcc
 	chmod +x $(BUILD_PATH)/musl-gcc
 
 $(IPERF_PATH)/.installed: $(IPERF_TAR)
@@ -291,7 +291,7 @@ $(IPERF_PATH)/.installed: $(IPERF_TAR)
 	touch $@
 
 $(IPERF_PATH)/src/iperf3: | $(IPERF_PATH)/.installed $(USERSPACE_DEPS)
-	cd $(IPERF_PATH) && CFLAGS="$(CFLAGS) -D_GNU_SOURCE" ./configure --prefix=/ $(CROSS_COMPILE_FLAG) --enable-static --disable-shared
+	cd $(IPERF_PATH) && CFLAGS="$(CFLAGS) -D_GNU_SOURCE" ./configure --prefix=/ $(CROSS_COMPILE_FLAG) --enable-static --disable-shared --with-openssl=no
 	$(MAKE) -C $(IPERF_PATH)
 	$(STRIP) -s $@
 
@@ -308,8 +308,8 @@ $(WIREGUARD_TOOLS_PATH)/.installed: $(WIREGUARD_TOOLS_TAR)
 	flock -s $<.lock tar -C $(BUILD_PATH) -xf $<
 	touch $@
 
-$(WIREGUARD_TOOLS_PATH)/src/tools/wg: | $(WIREGUARD_TOOLS_PATH)/.installed $(LIBMNL_PATH)/src/.libs/libmnl.a $(USERSPACE_DEPS)
-	LDFLAGS="$(LDFLAGS) -L$(LIBMNL_PATH)/src/.libs" $(MAKE) -C $(WIREGUARD_TOOLS_PATH)/src/tools LIBMNL_CFLAGS="-I$(LIBMNL_PATH)/include" LIBMNL_LDLIBS="-lmnl" wg
+$(WIREGUARD_TOOLS_PATH)/src/wg: | $(WIREGUARD_TOOLS_PATH)/.installed $(LIBMNL_PATH)/src/.libs/libmnl.a $(USERSPACE_DEPS)
+	LDFLAGS="$(LDFLAGS) -L$(LIBMNL_PATH)/src/.libs" $(MAKE) -C $(WIREGUARD_TOOLS_PATH)/src LIBMNL_CFLAGS="-I$(LIBMNL_PATH)/include" LIBMNL_LDLIBS="-lmnl" wg
 	$(STRIP) -s $@
 
 $(BUILD_PATH)/init: init.c | $(USERSPACE_DEPS)
@@ -323,7 +323,8 @@ $(IPUTILS_PATH)/.installed: $(IPUTILS_TAR)
 	touch $@
 
 $(IPUTILS_PATH)/ping: | $(IPUTILS_PATH)/.installed $(USERSPACE_DEPS)
-	$(MAKE) -C $(IPUTILS_PATH) USE_CAP=no USE_IDN=no USE_NETTLE=no USE_CRYPTO=no ping
+	sed -i /atexit/d $(IPUTILS_PATH)/ping.c
+	cd $(IPUTILS_PATH) && $(CC) $(CFLAGS) -std=c99 -o $@ ping.c ping_common.c ping6_common.c iputils_common.c -D_GNU_SOURCE -D'IPUTILS_VERSION(f)=f' -lresolv $(LDFLAGS)
 	$(STRIP) -s $@
 
 $(BASH_PATH)/.installed: $(BASH_TAR)
@@ -357,7 +358,7 @@ $(IPTABLES_PATH)/.installed: $(IPTABLES_TAR)
 	sed -i -e "/nfnetlink=[01]/s:=[01]:=0:" -e "/nfconntrack=[01]/s:=[01]:=0:" $(IPTABLES_PATH)/configure
 	touch $@
 
-$(IPTABLES_PATH)/iptables/xtables-multi: | $(IPTABLES_PATH)/.installed $(LIBMNL_PATH)/src/.libs/libmnl.a $(USERSPACE_DEPS)
+$(IPTABLES_PATH)/iptables/xtables-legacy-multi: | $(IPTABLES_PATH)/.installed $(LIBMNL_PATH)/src/.libs/libmnl.a $(USERSPACE_DEPS)
 	cd $(IPTABLES_PATH) && PKG_CONFIG_LIBDIR="$(LIBMNL_PATH)" ./configure --prefix=/ $(CROSS_COMPILE_FLAG) --enable-static --disable-shared --disable-nftables --disable-bpf-compiler --disable-nfsynproxy --disable-libipq --with-kernel=$(BUILD_PATH)/include
 	$(MAKE) -C $(IPTABLES_PATH)
 	$(STRIP) -s $@
@@ -368,8 +369,9 @@ $(NMAP_PATH)/.installed: $(NMAP_TAR)
 	touch $@
 
 $(NMAP_PATH)/ncat/ncat: | $(NMAP_PATH)/.installed $(USERSPACE_DEPS)
-	cd $(NMAP_PATH) && ./configure --prefix=/ $(CROSS_COMPILE_FLAG) --enable-static --disable-shared --without-ndiff --without-zenmap --without-nping --with-libpcap=included --with-libpcre=included --with-libdnet=included --without-liblua --with-liblinear=included --without-nmap-update --without-openssl --with-pcap=linux
-	$(MAKE) -C $(NMAP_PATH) build-ncat
+	cd $(NMAP_PATH) && ./configure --prefix=/ $(CROSS_COMPILE_FLAG) --enable-static --disable-shared --without-ndiff --without-zenmap --without-nping --with-libpcap=included --with-libpcre=included --with-libdnet=included --without-liblua --with-liblinear=included --without-nmap-update --without-openssl --with-pcap=linux --without-libssh
+	$(MAKE) -C $(NMAP_PATH)/libpcap
+	$(MAKE) -C $(NMAP_PATH)/ncat
 	$(STRIP) -s $@
 
 clean:
@@ -379,7 +381,7 @@ distclean: clean
 	rm -rf $(DISTFILES_PATH)
 
 menuconfig: $(KERNEL_BUILD_PATH)/.config
-	$(MAKE) -C $(KERNEL_PATH) O=$(KERNEL_BUILD_PATH) ARCH=$(KERNEL_ARCH) CROSS_COMPILE=$(CROSS_COMPILE) CC="$(NOPIE_GCC)" menuconfig
+	$(MAKE) -C $(KERNEL_PATH) O=$(KERNEL_BUILD_PATH) ARCH=$(KERNEL_ARCH) CROSS_COMPILE=$(CROSS_COMPILE) menuconfig
 
 .PHONY: qemu build clean distclean menuconfig
 .DELETE_ON_ERROR:
diff --git a/tools/testing/selftests/wireguard/qemu/arch/m68k.config b/tools/testing/selftests/wireguard/qemu/arch/m68k.config
index 5381ea10896c..62a15bdb877e 100644
--- a/tools/testing/selftests/wireguard/qemu/arch/m68k.config
+++ b/tools/testing/selftests/wireguard/qemu/arch/m68k.config
@@ -1,9 +1,9 @@
 CONFIG_MMU=y
+CONFIG_M68KCLASSIC=y
 CONFIG_M68040=y
 CONFIG_MAC=y
 CONFIG_SERIAL_PMACZILOG=y
 CONFIG_SERIAL_PMACZILOG_TTYS=y
 CONFIG_SERIAL_PMACZILOG_CONSOLE=y
-CONFIG_CMDLINE_BOOL=y
 CONFIG_CMDLINE="console=ttyS0 wg.success=ttyS1"
 CONFIG_FRAME_WARN=1024
diff --git a/tools/testing/selftests/wireguard/qemu/init.c b/tools/testing/selftests/wireguard/qemu/init.c
index 51e5ddedee88..90bc9813cadc 100644
--- a/tools/testing/selftests/wireguard/qemu/init.c
+++ b/tools/testing/selftests/wireguard/qemu/init.c
@@ -21,6 +21,7 @@
 #include <sys/reboot.h>
 #include <sys/utsname.h>
 #include <sys/sendfile.h>
+#include <sys/sysmacros.h>
 #include <linux/random.h>
 #include <linux/version.h>
 
diff --git a/tools/testing/selftests/wireguard/qemu/kernel.config b/tools/testing/selftests/wireguard/qemu/kernel.config
index 9cca30206014..af9323a0b6e0 100644
--- a/tools/testing/selftests/wireguard/qemu/kernel.config
+++ b/tools/testing/selftests/wireguard/qemu/kernel.config
@@ -39,6 +39,7 @@ CONFIG_PRINTK=y
 CONFIG_KALLSYMS=y
 CONFIG_BUG=y
 CONFIG_CC_OPTIMIZE_FOR_PERFORMANCE=y
+CONFIG_JUMP_LABEL=y
 CONFIG_EMBEDDED=n
 CONFIG_BASE_FULL=y
 CONFIG_FUTEX=y
@@ -55,6 +56,7 @@ CONFIG_NO_HZ_IDLE=y
 CONFIG_NO_HZ_FULL=n
 CONFIG_HZ_PERIODIC=n
 CONFIG_HIGH_RES_TIMERS=y
+CONFIG_COMPAT_32BIT_TIME=y
 CONFIG_ARCH_RANDOM=y
 CONFIG_FILE_LOCKING=y
 CONFIG_POSIX_TIMERS=y
-- 
cgit v1.2.3-59-g8ed1b


From 04d2ea92a18417619182cbb79063f154892b0150 Mon Sep 17 00:00:00 2001
From: "Jason A. Donenfeld" <Jason@zx2c4.com>
Date: Thu, 2 Jan 2020 17:47:50 +0100
Subject: wireguard: queueing: do not account for pfmemalloc when clearing skb
 header

Before 8b7008620b84 ("net: Don't copy pfmemalloc flag in __copy_skb_
header()"), the pfmemalloc flag used to be between headers_start and
headers_end, which is a region we clear when preparing the packet for
encryption/decryption. This is a parameter we certainly want to
preserve, which is why 8b7008620b84 moved it out of there. The code here
was written in a world before 8b7008620b84, though, where we had to
manually account for it. This commit brings things up to speed.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/wireguard/queueing.h | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/drivers/net/wireguard/queueing.h b/drivers/net/wireguard/queueing.h
index e49a464238fd..fecb559cbdb6 100644
--- a/drivers/net/wireguard/queueing.h
+++ b/drivers/net/wireguard/queueing.h
@@ -83,13 +83,10 @@ static inline __be16 wg_skb_examine_untrusted_ip_hdr(struct sk_buff *skb)
 
 static inline void wg_reset_packet(struct sk_buff *skb)
 {
-	const int pfmemalloc = skb->pfmemalloc;
-
 	skb_scrub_packet(skb, true);
 	memset(&skb->headers_start, 0,
 	       offsetof(struct sk_buff, headers_end) -
 		       offsetof(struct sk_buff, headers_start));
-	skb->pfmemalloc = pfmemalloc;
 	skb->queue_mapping = 0;
 	skb->nohdr = 0;
 	skb->peeked = 0;
-- 
cgit v1.2.3-59-g8ed1b


From 736775d06bac60d7a353e405398b48b2bd8b1e54 Mon Sep 17 00:00:00 2001
From: "Jason A. Donenfeld" <Jason@zx2c4.com>
Date: Thu, 2 Jan 2020 17:47:51 +0100
Subject: wireguard: socket: mark skbs as not on list when receiving via gro

Certain drivers will pass gro skbs to udp, at which point the udp driver
simply iterates through them and passes them off to encap_rcv, which is
where we pick up. At the moment, we're not attempting to coalesce these
into bundles, but we also don't want to wind up having cascaded lists of
skbs treated separately. The right behavior here, then, is to just mark
each incoming one as not on a list. This can be seen in practice, for
example, with Qualcomm's rmnet_perf driver.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Tested-by: Yaroslav Furman <yaro330@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/wireguard/socket.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wireguard/socket.c b/drivers/net/wireguard/socket.c
index c46256d0d81c..262f3b5c819d 100644
--- a/drivers/net/wireguard/socket.c
+++ b/drivers/net/wireguard/socket.c
@@ -333,6 +333,7 @@ static int wg_receive(struct sock *sk, struct sk_buff *skb)
 	wg = sk->sk_user_data;
 	if (unlikely(!wg))
 		goto err;
+	skb_mark_not_on_list(skb);
 	wg_packet_receive(wg, skb);
 	return 0;
 
-- 
cgit v1.2.3-59-g8ed1b


From aa1d54c65d11ab32e72f8122709e9dff4049ce93 Mon Sep 17 00:00:00 2001
From: Florian Fainelli <f.fainelli@gmail.com>
Date: Thu, 2 Jan 2020 15:34:45 -0800
Subject: net: dsa: vsc73xx: Remove dependency on CONFIG_OF

There is no build time dependency on CONFIG_OF, but we do need to make
sure we gate the initialization of the gpio_chip::of_node member with a
proper check on CONFIG_OF_GPIO. This enables the driver to build on
platforms that do not have CONFIG_OF enabled.

Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/Kconfig                | 3 ---
 drivers/net/dsa/vitesse-vsc73xx-core.c | 2 ++
 2 files changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/net/dsa/Kconfig b/drivers/net/dsa/Kconfig
index cbd74a72d0a1..2d38dbc9dd8c 100644
--- a/drivers/net/dsa/Kconfig
+++ b/drivers/net/dsa/Kconfig
@@ -105,7 +105,6 @@ config NET_DSA_SMSC_LAN9303_MDIO
 
 config NET_DSA_VITESSE_VSC73XX
 	tristate
-	depends on OF
 	depends on NET_DSA
 	select FIXED_PHY
 	select VITESSE_PHY
@@ -116,7 +115,6 @@ config NET_DSA_VITESSE_VSC73XX
 
 config NET_DSA_VITESSE_VSC73XX_SPI
 	tristate "Vitesse VSC7385/7388/7395/7398 SPI mode support"
-	depends on OF
 	depends on NET_DSA
 	depends on SPI
 	select NET_DSA_VITESSE_VSC73XX
@@ -126,7 +124,6 @@ config NET_DSA_VITESSE_VSC73XX_SPI
 
 config NET_DSA_VITESSE_VSC73XX_PLATFORM
 	tristate "Vitesse VSC7385/7388/7395/7398 Platform mode support"
-	depends on OF
 	depends on NET_DSA
 	depends on HAS_IOMEM
 	select NET_DSA_VITESSE_VSC73XX
diff --git a/drivers/net/dsa/vitesse-vsc73xx-core.c b/drivers/net/dsa/vitesse-vsc73xx-core.c
index 42c1574d45f2..69fc0110ce04 100644
--- a/drivers/net/dsa/vitesse-vsc73xx-core.c
+++ b/drivers/net/dsa/vitesse-vsc73xx-core.c
@@ -1111,7 +1111,9 @@ static int vsc73xx_gpio_probe(struct vsc73xx *vsc)
 	vsc->gc.ngpio = 4;
 	vsc->gc.owner = THIS_MODULE;
 	vsc->gc.parent = vsc->dev;
+#if IS_ENABLED(CONFIG_OF_GPIO)
 	vsc->gc.of_node = vsc->dev->of_node;
+#endif
 	vsc->gc.base = -1;
 	vsc->gc.get = vsc73xx_gpio_get;
 	vsc->gc.set = vsc73xx_gpio_set;
-- 
cgit v1.2.3-59-g8ed1b


From b605c9abc4d2e79eed1332be14a79960a1e9f503 Mon Sep 17 00:00:00 2001
From: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Date: Thu, 2 Jan 2020 17:03:18 -0800
Subject: net: phylink: switch to using fwnode_gpiod_get_index()

Instead of fwnode_get_named_gpiod() that I plan to hide away, let's use
the new fwnode_gpiod_get_index() that mimics gpiod_get_index(), but
works with arbitrary firmware node.

Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/phylink.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index ba9468cc8e13..3f7fa634134a 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -186,8 +186,8 @@ static int phylink_parse_fixedlink(struct phylink *pl,
 			pl->link_config.pause |= MLO_PAUSE_ASYM;
 
 		if (ret == 0) {
-			desc = fwnode_get_named_gpiod(fixed_node, "link-gpios",
-						      0, GPIOD_IN, "?");
+			desc = fwnode_gpiod_get_index(fixed_node, "link", 0,
+						      GPIOD_IN, "?");
 
 			if (!IS_ERR(desc))
 				pl->link_gpio = desc;
-- 
cgit v1.2.3-59-g8ed1b


From d266f19f3ae7fbcaf92229639b78d2110ae44f33 Mon Sep 17 00:00:00 2001
From: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Date: Thu, 2 Jan 2020 17:03:19 -0800
Subject: net: phy: fixed_phy: fix use-after-free when checking link GPIO

If we fail to locate GPIO for any reason other than deferral or
not-found-GPIO, we try to print device tree node info, however if might
be freed already as we called of_node_put() on it.

Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/fixed_phy.c | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

diff --git a/drivers/net/phy/fixed_phy.c b/drivers/net/phy/fixed_phy.c
index 7c5265fd2b94..4190f9ed5313 100644
--- a/drivers/net/phy/fixed_phy.c
+++ b/drivers/net/phy/fixed_phy.c
@@ -212,16 +212,13 @@ static struct gpio_desc *fixed_phy_get_gpiod(struct device_node *np)
 	 */
 	gpiod = gpiod_get_from_of_node(fixed_link_node, "link-gpios", 0,
 				       GPIOD_IN, "mdio");
-	of_node_put(fixed_link_node);
-	if (IS_ERR(gpiod)) {
-		if (PTR_ERR(gpiod) == -EPROBE_DEFER)
-			return gpiod;
-
+	if (IS_ERR(gpiod) && PTR_ERR(gpiod) != -EPROBE_DEFER) {
 		if (PTR_ERR(gpiod) != -ENOENT)
 			pr_err("error getting GPIO for fixed link %pOF, proceed without\n",
 			       fixed_link_node);
 		gpiod = NULL;
 	}
+	of_node_put(fixed_link_node);
 
 	return gpiod;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 5ffcc85814f27e04c1c0180bf7a3b7b76bb776c8 Mon Sep 17 00:00:00 2001
From: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Date: Thu, 2 Jan 2020 17:03:20 -0800
Subject: net: phy: fixed_phy: switch to using fwnode_gpiod_get_index

gpiod_get_from_of_node() is being retired in favor of
[devm_]fwnode_gpiod_get_index(), that behaves similar to
[devm_]gpiod_get_index(), but can work with arbitrary firmware node. It
will also be able to support secondary software nodes.

Let's switch this driver over.

Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/fixed_phy.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/phy/fixed_phy.c b/drivers/net/phy/fixed_phy.c
index 4190f9ed5313..4a3d34f40cb9 100644
--- a/drivers/net/phy/fixed_phy.c
+++ b/drivers/net/phy/fixed_phy.c
@@ -210,8 +210,8 @@ static struct gpio_desc *fixed_phy_get_gpiod(struct device_node *np)
 	 * Linux device associated with it, we simply have obtain
 	 * the GPIO descriptor from the device tree like this.
 	 */
-	gpiod = gpiod_get_from_of_node(fixed_link_node, "link-gpios", 0,
-				       GPIOD_IN, "mdio");
+	gpiod = fwnode_gpiod_get_index(of_fwnode_handle(fixed_link_node),
+				       "link", 0, GPIOD_IN, "mdio");
 	if (IS_ERR(gpiod) && PTR_ERR(gpiod) != -EPROBE_DEFER) {
 		if (PTR_ERR(gpiod) != -ENOENT)
 			pr_err("error getting GPIO for fixed link %pOF, proceed without\n",
-- 
cgit v1.2.3-59-g8ed1b


From 14a65084f9310ba6a4017c365f9c9820b099dde5 Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzk@kernel.org>
Date: Fri, 3 Jan 2020 18:11:30 +0100
Subject: net: ethernet: sxgbe: Rename Samsung to lowercase

Fix up inconsistent usage of upper and lowercase letters in "Samsung"
name.

"SAMSUNG" is not an abbreviation but a regular trademarked name.
Therefore it should be written with lowercase letters starting with
capital letter.

Although advertisement materials usually use uppercase "SAMSUNG", the
lowercase version is used in all legal aspects (e.g. on Wikipedia and in
privacy/legal statements on
https://www.samsung.com/semiconductor/privacy-global/).

Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c | 2 +-
 include/linux/sxgbe_platform.h                  | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c b/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c
index cd6e0de48248..7d3a1c0df09c 100644
--- a/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c
+++ b/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c
@@ -2296,7 +2296,7 @@ __setup("sxgbeeth=", sxgbe_cmdline_opt);
 
 
 
-MODULE_DESCRIPTION("SAMSUNG 10G/2.5G/1G Ethernet PLATFORM driver");
+MODULE_DESCRIPTION("Samsung 10G/2.5G/1G Ethernet PLATFORM driver");
 
 MODULE_PARM_DESC(debug, "Message Level (-1: default, 0: no output, 16: all)");
 MODULE_PARM_DESC(eee_timer, "EEE-LPI Default LS timer value");
diff --git a/include/linux/sxgbe_platform.h b/include/linux/sxgbe_platform.h
index 85ec745767bd..966146f7267a 100644
--- a/include/linux/sxgbe_platform.h
+++ b/include/linux/sxgbe_platform.h
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0-only */
 /*
- * 10G controller driver for Samsung EXYNOS SoCs
+ * 10G controller driver for Samsung Exynos SoCs
  *
  * Copyright (C) 2013 Samsung Electronics Co., Ltd.
  *		http://www.samsung.com
-- 
cgit v1.2.3-59-g8ed1b


From 3d462ce2a11784e278d4f425bf5c4cadeab2af3a Mon Sep 17 00:00:00 2001
From: Shannon Nelson <snelson@pensando.io>
Date: Fri, 3 Jan 2020 09:55:07 -0800
Subject: ionic: ionic_if bits for sr-iov support

Adds new AdminQ calls and their related structs for
supporting PF controls on VFs:
    CMD_OPCODE_VF_GETATTR
    CMD_OPCODE_VF_SETATTR

Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/pensando/ionic/ionic_if.h | 97 ++++++++++++++++++++++++++
 1 file changed, 97 insertions(+)

diff --git a/drivers/net/ethernet/pensando/ionic/ionic_if.h b/drivers/net/ethernet/pensando/ionic/ionic_if.h
index 39317cdfa6cf..f131adad96e3 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_if.h
+++ b/drivers/net/ethernet/pensando/ionic/ionic_if.h
@@ -51,6 +51,10 @@ enum ionic_cmd_opcode {
 	IONIC_CMD_RDMA_CREATE_CQ		= 52,
 	IONIC_CMD_RDMA_CREATE_ADMINQ		= 53,
 
+	/* SR/IOV commands */
+	IONIC_CMD_VF_GETATTR			= 60,
+	IONIC_CMD_VF_SETATTR			= 61,
+
 	/* QoS commands */
 	IONIC_CMD_QOS_CLASS_IDENTIFY		= 240,
 	IONIC_CMD_QOS_CLASS_INIT		= 241,
@@ -1639,6 +1643,93 @@ enum ionic_qos_sched_type {
 	IONIC_QOS_SCHED_TYPE_DWRR	= 1,	/* Deficit weighted round-robin */
 };
 
+enum ionic_vf_attr {
+	IONIC_VF_ATTR_SPOOFCHK	= 1,
+	IONIC_VF_ATTR_TRUST	= 2,
+	IONIC_VF_ATTR_MAC	= 3,
+	IONIC_VF_ATTR_LINKSTATE	= 4,
+	IONIC_VF_ATTR_VLAN	= 5,
+	IONIC_VF_ATTR_RATE	= 6,
+	IONIC_VF_ATTR_STATSADDR	= 7,
+};
+
+/**
+ * VF link status
+ */
+enum ionic_vf_link_status {
+	IONIC_VF_LINK_STATUS_AUTO = 0,	/* link state of the uplink */
+	IONIC_VF_LINK_STATUS_UP   = 1,	/* link is always up */
+	IONIC_VF_LINK_STATUS_DOWN = 2,	/* link is always down */
+};
+
+/**
+ * struct ionic_vf_setattr_cmd - Set VF attributes on the NIC
+ * @opcode:     Opcode
+ * @index:      VF index
+ * @attr:       Attribute type (enum ionic_vf_attr)
+ *	macaddr		mac address
+ *	vlanid		vlan ID
+ *	maxrate		max Tx rate in Mbps
+ *	spoofchk	enable address spoof checking
+ *	trust		enable VF trust
+ *	linkstate	set link up or down
+ *	stats_pa	set DMA address for VF stats
+ */
+struct ionic_vf_setattr_cmd {
+	u8     opcode;
+	u8     attr;
+	__le16 vf_index;
+	union {
+		u8     macaddr[6];
+		__le16 vlanid;
+		__le32 maxrate;
+		u8     spoofchk;
+		u8     trust;
+		u8     linkstate;
+		__le64 stats_pa;
+		u8     pad[60];
+	};
+};
+
+struct ionic_vf_setattr_comp {
+	u8     status;
+	u8     attr;
+	__le16 vf_index;
+	__le16 comp_index;
+	u8     rsvd[9];
+	u8     color;
+};
+
+/**
+ * struct ionic_vf_getattr_cmd - Get VF attributes from the NIC
+ * @opcode:     Opcode
+ * @index:      VF index
+ * @attr:       Attribute type (enum ionic_vf_attr)
+ */
+struct ionic_vf_getattr_cmd {
+	u8     opcode;
+	u8     attr;
+	__le16 vf_index;
+	u8     rsvd[60];
+};
+
+struct ionic_vf_getattr_comp {
+	u8     status;
+	u8     attr;
+	__le16 vf_index;
+	union {
+		u8     macaddr[6];
+		__le16 vlanid;
+		__le32 maxrate;
+		u8     spoofchk;
+		u8     trust;
+		u8     linkstate;
+		__le64 stats_pa;
+		u8     pad[11];
+	};
+	u8     color;
+};
+
 /**
  * union ionic_qos_config - Qos configuration structure
  * @flags:		Configuration flags
@@ -2289,6 +2380,9 @@ union ionic_dev_cmd {
 	struct ionic_port_getattr_cmd port_getattr;
 	struct ionic_port_setattr_cmd port_setattr;
 
+	struct ionic_vf_setattr_cmd vf_setattr;
+	struct ionic_vf_getattr_cmd vf_getattr;
+
 	struct ionic_lif_identify_cmd lif_identify;
 	struct ionic_lif_init_cmd lif_init;
 	struct ionic_lif_reset_cmd lif_reset;
@@ -2318,6 +2412,9 @@ union ionic_dev_cmd_comp {
 	struct ionic_port_getattr_comp port_getattr;
 	struct ionic_port_setattr_comp port_setattr;
 
+	struct ionic_vf_setattr_comp vf_setattr;
+	struct ionic_vf_getattr_comp vf_getattr;
+
 	struct ionic_lif_identify_comp lif_identify;
 	struct ionic_lif_init_comp lif_init;
 	ionic_lif_reset_comp lif_reset;
-- 
cgit v1.2.3-59-g8ed1b


From fbb39807e9ae0f60e2f6a0a628aadab73554fbbe Mon Sep 17 00:00:00 2001
From: Shannon Nelson <snelson@pensando.io>
Date: Fri, 3 Jan 2020 09:55:08 -0800
Subject: ionic: support sr-iov operations

Add the netdev ops for managing VFs.  Since most of the
management work happens in the NIC firmware, the driver becomes
mostly a pass-through for the network stack commands that want
to control and configure the VFs.

We also tweak ionic_station_set() a little to allow for
the VFs that start off with a zero'd mac address.

Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/pensando/ionic/ionic.h        |  17 +-
 .../net/ethernet/pensando/ionic/ionic_bus_pci.c    | 113 ++++++++++
 drivers/net/ethernet/pensando/ionic/ionic_dev.c    |  58 +++++
 drivers/net/ethernet/pensando/ionic/ionic_dev.h    |   7 +
 drivers/net/ethernet/pensando/ionic/ionic_lif.c    | 247 ++++++++++++++++++++-
 drivers/net/ethernet/pensando/ionic/ionic_main.c   |   4 +
 6 files changed, 438 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ethernet/pensando/ionic/ionic.h b/drivers/net/ethernet/pensando/ionic/ionic.h
index 98e102af7756..ca0b21911ad6 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic.h
+++ b/drivers/net/ethernet/pensando/ionic/ionic.h
@@ -12,7 +12,7 @@ struct ionic_lif;
 
 #define IONIC_DRV_NAME		"ionic"
 #define IONIC_DRV_DESCRIPTION	"Pensando Ethernet NIC Driver"
-#define IONIC_DRV_VERSION	"0.18.0-k"
+#define IONIC_DRV_VERSION	"0.20.0-k"
 
 #define PCI_VENDOR_ID_PENSANDO			0x1dd8
 
@@ -25,6 +25,18 @@ struct ionic_lif;
 
 #define DEVCMD_TIMEOUT  10
 
+struct ionic_vf {
+	u16	 index;
+	u8	 macaddr[6];
+	__le32	 maxrate;
+	__le16	 vlanid;
+	u8	 spoofchk;
+	u8	 trusted;
+	u8	 linkstate;
+	dma_addr_t       stats_pa;
+	struct ionic_lif_stats stats;
+};
+
 struct ionic {
 	struct pci_dev *pdev;
 	struct device *dev;
@@ -46,6 +58,9 @@ struct ionic {
 	DECLARE_BITMAP(intrs, IONIC_INTR_CTRL_REGS_MAX);
 	struct work_struct nb_work;
 	struct notifier_block nb;
+	struct rw_semaphore vf_op_lock;	/* lock for VF operations */
+	struct ionic_vf *vfs;
+	int num_vfs;
 	struct timer_list watchdog_timer;
 	int watchdog_period;
 };
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c b/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c
index 9a9ab8cb2cb3..448d7b23b2f7 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c
@@ -104,10 +104,112 @@ void ionic_bus_unmap_dbpage(struct ionic *ionic, void __iomem *page)
 	iounmap(page);
 }
 
+static void ionic_vf_dealloc_locked(struct ionic *ionic)
+{
+	struct ionic_vf *v;
+	dma_addr_t dma = 0;
+	int i;
+
+	if (!ionic->vfs)
+		return;
+
+	for (i = ionic->num_vfs - 1; i >= 0; i--) {
+		v = &ionic->vfs[i];
+
+		if (v->stats_pa) {
+			(void)ionic_set_vf_config(ionic, i,
+						  IONIC_VF_ATTR_STATSADDR,
+						  (u8 *)&dma);
+			dma_unmap_single(ionic->dev, v->stats_pa,
+					 sizeof(v->stats), DMA_FROM_DEVICE);
+			v->stats_pa = 0;
+		}
+	}
+
+	kfree(ionic->vfs);
+	ionic->vfs = NULL;
+	ionic->num_vfs = 0;
+}
+
+static void ionic_vf_dealloc(struct ionic *ionic)
+{
+	down_write(&ionic->vf_op_lock);
+	ionic_vf_dealloc_locked(ionic);
+	up_write(&ionic->vf_op_lock);
+}
+
+static int ionic_vf_alloc(struct ionic *ionic, int num_vfs)
+{
+	struct ionic_vf *v;
+	int err = 0;
+	int i;
+
+	down_write(&ionic->vf_op_lock);
+
+	ionic->vfs = kcalloc(num_vfs, sizeof(struct ionic_vf), GFP_KERNEL);
+	if (!ionic->vfs) {
+		err = -ENOMEM;
+		goto out;
+	}
+
+	for (i = 0; i < num_vfs; i++) {
+		v = &ionic->vfs[i];
+		v->stats_pa = dma_map_single(ionic->dev, &v->stats,
+					     sizeof(v->stats), DMA_FROM_DEVICE);
+		if (dma_mapping_error(ionic->dev, v->stats_pa)) {
+			v->stats_pa = 0;
+			err = -ENODEV;
+			goto out;
+		}
+
+		/* ignore failures from older FW, we just won't get stats */
+		(void)ionic_set_vf_config(ionic, i, IONIC_VF_ATTR_STATSADDR,
+					  (u8 *)&v->stats_pa);
+		ionic->num_vfs++;
+	}
+
+out:
+	if (err)
+		ionic_vf_dealloc_locked(ionic);
+	up_write(&ionic->vf_op_lock);
+	return err;
+}
+
+static int ionic_sriov_configure(struct pci_dev *pdev, int num_vfs)
+{
+	struct ionic *ionic = pci_get_drvdata(pdev);
+	struct device *dev = ionic->dev;
+	int ret = 0;
+
+	if (num_vfs > 0) {
+		ret = pci_enable_sriov(pdev, num_vfs);
+		if (ret) {
+			dev_err(dev, "Cannot enable SRIOV: %d\n", ret);
+			goto out;
+		}
+
+		ret = ionic_vf_alloc(ionic, num_vfs);
+		if (ret) {
+			dev_err(dev, "Cannot alloc VFs: %d\n", ret);
+			pci_disable_sriov(pdev);
+			goto out;
+		}
+
+		ret = num_vfs;
+	} else {
+		pci_disable_sriov(pdev);
+		ionic_vf_dealloc(ionic);
+	}
+
+out:
+	return ret;
+}
+
 static int ionic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
 	struct device *dev = &pdev->dev;
 	struct ionic *ionic;
+	int num_vfs;
 	int err;
 
 	ionic = ionic_devlink_alloc(dev);
@@ -206,6 +308,15 @@ static int ionic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto err_out_free_lifs;
 	}
 
+	init_rwsem(&ionic->vf_op_lock);
+	num_vfs = pci_num_vf(pdev);
+	if (num_vfs) {
+		dev_info(dev, "%d VFs found already enabled\n", num_vfs);
+		err = ionic_vf_alloc(ionic, num_vfs);
+		if (err)
+			dev_err(dev, "Cannot enable existing VFs: %d\n", err);
+	}
+
 	err = ionic_lifs_register(ionic);
 	if (err) {
 		dev_err(dev, "Cannot register LIFs: %d, aborting\n", err);
@@ -223,6 +334,7 @@ static int ionic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 err_out_deregister_lifs:
 	ionic_lifs_unregister(ionic);
 err_out_deinit_lifs:
+	ionic_vf_dealloc(ionic);
 	ionic_lifs_deinit(ionic);
 err_out_free_lifs:
 	ionic_lifs_free(ionic);
@@ -279,6 +391,7 @@ static struct pci_driver ionic_driver = {
 	.id_table = ionic_id_table,
 	.probe = ionic_probe,
 	.remove = ionic_remove,
+	.sriov_configure = ionic_sriov_configure,
 };
 
 int ionic_bus_register_driver(void)
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_dev.c b/drivers/net/ethernet/pensando/ionic/ionic_dev.c
index 5f9d2ec70446..87f82f36812f 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_dev.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_dev.c
@@ -286,6 +286,64 @@ void ionic_dev_cmd_port_pause(struct ionic_dev *idev, u8 pause_type)
 	ionic_dev_cmd_go(idev, &cmd);
 }
 
+/* VF commands */
+int ionic_set_vf_config(struct ionic *ionic, int vf, u8 attr, u8 *data)
+{
+	union ionic_dev_cmd cmd = {
+		.vf_setattr.opcode = IONIC_CMD_VF_SETATTR,
+		.vf_setattr.attr = attr,
+		.vf_setattr.vf_index = vf,
+	};
+	int err;
+
+	switch (attr) {
+	case IONIC_VF_ATTR_SPOOFCHK:
+		cmd.vf_setattr.spoofchk = *data;
+		dev_dbg(ionic->dev, "%s: vf %d spoof %d\n",
+			__func__, vf, *data);
+		break;
+	case IONIC_VF_ATTR_TRUST:
+		cmd.vf_setattr.trust = *data;
+		dev_dbg(ionic->dev, "%s: vf %d trust %d\n",
+			__func__, vf, *data);
+		break;
+	case IONIC_VF_ATTR_LINKSTATE:
+		cmd.vf_setattr.linkstate = *data;
+		dev_dbg(ionic->dev, "%s: vf %d linkstate %d\n",
+			__func__, vf, *data);
+		break;
+	case IONIC_VF_ATTR_MAC:
+		ether_addr_copy(cmd.vf_setattr.macaddr, data);
+		dev_dbg(ionic->dev, "%s: vf %d macaddr %pM\n",
+			__func__, vf, data);
+		break;
+	case IONIC_VF_ATTR_VLAN:
+		cmd.vf_setattr.vlanid = cpu_to_le16(*(u16 *)data);
+		dev_dbg(ionic->dev, "%s: vf %d vlan %d\n",
+			__func__, vf, *(u16 *)data);
+		break;
+	case IONIC_VF_ATTR_RATE:
+		cmd.vf_setattr.maxrate = cpu_to_le32(*(u32 *)data);
+		dev_dbg(ionic->dev, "%s: vf %d maxrate %d\n",
+			__func__, vf, *(u32 *)data);
+		break;
+	case IONIC_VF_ATTR_STATSADDR:
+		cmd.vf_setattr.stats_pa = cpu_to_le64(*(u64 *)data);
+		dev_dbg(ionic->dev, "%s: vf %d stats_pa 0x%08llx\n",
+			__func__, vf, *(u64 *)data);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	mutex_lock(&ionic->dev_cmd_lock);
+	ionic_dev_cmd_go(&ionic->idev, &cmd);
+	err = ionic_dev_cmd_wait(ionic, DEVCMD_TIMEOUT);
+	mutex_unlock(&ionic->dev_cmd_lock);
+
+	return err;
+}
+
 /* LIF commands */
 void ionic_dev_cmd_lif_identify(struct ionic_dev *idev, u8 type, u8 ver)
 {
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_dev.h b/drivers/net/ethernet/pensando/ionic/ionic_dev.h
index 4665c5dc5324..7838e342c4fd 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_dev.h
+++ b/drivers/net/ethernet/pensando/ionic/ionic_dev.h
@@ -113,6 +113,12 @@ static_assert(sizeof(struct ionic_rxq_desc) == 16);
 static_assert(sizeof(struct ionic_rxq_sg_desc) == 128);
 static_assert(sizeof(struct ionic_rxq_comp) == 16);
 
+/* SR/IOV */
+static_assert(sizeof(struct ionic_vf_setattr_cmd) == 64);
+static_assert(sizeof(struct ionic_vf_setattr_comp) == 16);
+static_assert(sizeof(struct ionic_vf_getattr_cmd) == 64);
+static_assert(sizeof(struct ionic_vf_getattr_comp) == 16);
+
 struct ionic_devinfo {
 	u8 asic_type;
 	u8 asic_rev;
@@ -275,6 +281,7 @@ void ionic_dev_cmd_port_autoneg(struct ionic_dev *idev, u8 an_enable);
 void ionic_dev_cmd_port_fec(struct ionic_dev *idev, u8 fec_type);
 void ionic_dev_cmd_port_pause(struct ionic_dev *idev, u8 pause_type);
 
+int ionic_set_vf_config(struct ionic *ionic, int vf, u8 attr, u8 *data);
 void ionic_dev_cmd_lif_identify(struct ionic_dev *idev, u8 type, u8 ver);
 void ionic_dev_cmd_lif_init(struct ionic_dev *idev, u16 lif_index,
 			    dma_addr_t addr);
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.c b/drivers/net/ethernet/pensando/ionic/ionic_lif.c
index a76108a9c7db..191271f6260d 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_lif.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.c
@@ -1619,6 +1619,227 @@ int ionic_stop(struct net_device *netdev)
 	return err;
 }
 
+static int ionic_get_vf_config(struct net_device *netdev,
+			       int vf, struct ifla_vf_info *ivf)
+{
+	struct ionic_lif *lif = netdev_priv(netdev);
+	struct ionic *ionic = lif->ionic;
+	int ret = 0;
+
+	down_read(&ionic->vf_op_lock);
+
+	if (vf >= pci_num_vf(ionic->pdev) || !ionic->vfs) {
+		ret = -EINVAL;
+	} else {
+		ivf->vf           = vf;
+		ivf->vlan         = ionic->vfs[vf].vlanid;
+		ivf->qos	  = 0;
+		ivf->spoofchk     = ionic->vfs[vf].spoofchk;
+		ivf->linkstate    = ionic->vfs[vf].linkstate;
+		ivf->max_tx_rate  = ionic->vfs[vf].maxrate;
+		ivf->trusted      = ionic->vfs[vf].trusted;
+		ether_addr_copy(ivf->mac, ionic->vfs[vf].macaddr);
+	}
+
+	up_read(&ionic->vf_op_lock);
+	return ret;
+}
+
+static int ionic_get_vf_stats(struct net_device *netdev, int vf,
+			      struct ifla_vf_stats *vf_stats)
+{
+	struct ionic_lif *lif = netdev_priv(netdev);
+	struct ionic *ionic = lif->ionic;
+	struct ionic_lif_stats *vs;
+	int ret = 0;
+
+	down_read(&ionic->vf_op_lock);
+
+	if (vf >= pci_num_vf(ionic->pdev) || !ionic->vfs) {
+		ret = -EINVAL;
+	} else {
+		memset(vf_stats, 0, sizeof(*vf_stats));
+		vs = &ionic->vfs[vf].stats;
+
+		vf_stats->rx_packets = le64_to_cpu(vs->rx_ucast_packets);
+		vf_stats->tx_packets = le64_to_cpu(vs->tx_ucast_packets);
+		vf_stats->rx_bytes   = le64_to_cpu(vs->rx_ucast_bytes);
+		vf_stats->tx_bytes   = le64_to_cpu(vs->tx_ucast_bytes);
+		vf_stats->broadcast  = le64_to_cpu(vs->rx_bcast_packets);
+		vf_stats->multicast  = le64_to_cpu(vs->rx_mcast_packets);
+		vf_stats->rx_dropped = le64_to_cpu(vs->rx_ucast_drop_packets) +
+				       le64_to_cpu(vs->rx_mcast_drop_packets) +
+				       le64_to_cpu(vs->rx_bcast_drop_packets);
+		vf_stats->tx_dropped = le64_to_cpu(vs->tx_ucast_drop_packets) +
+				       le64_to_cpu(vs->tx_mcast_drop_packets) +
+				       le64_to_cpu(vs->tx_bcast_drop_packets);
+	}
+
+	up_read(&ionic->vf_op_lock);
+	return ret;
+}
+
+static int ionic_set_vf_mac(struct net_device *netdev, int vf, u8 *mac)
+{
+	struct ionic_lif *lif = netdev_priv(netdev);
+	struct ionic *ionic = lif->ionic;
+	int ret;
+
+	if (!(is_zero_ether_addr(mac) || is_valid_ether_addr(mac)))
+		return -EINVAL;
+
+	down_read(&ionic->vf_op_lock);
+
+	if (vf >= pci_num_vf(ionic->pdev) || !ionic->vfs) {
+		ret = -EINVAL;
+	} else {
+		ret = ionic_set_vf_config(ionic, vf, IONIC_VF_ATTR_MAC, mac);
+		if (!ret)
+			ether_addr_copy(ionic->vfs[vf].macaddr, mac);
+	}
+
+	up_read(&ionic->vf_op_lock);
+	return ret;
+}
+
+static int ionic_set_vf_vlan(struct net_device *netdev, int vf, u16 vlan,
+			     u8 qos, __be16 proto)
+{
+	struct ionic_lif *lif = netdev_priv(netdev);
+	struct ionic *ionic = lif->ionic;
+	int ret;
+
+	/* until someday when we support qos */
+	if (qos)
+		return -EINVAL;
+
+	if (vlan > 4095)
+		return -EINVAL;
+
+	if (proto != htons(ETH_P_8021Q))
+		return -EPROTONOSUPPORT;
+
+	down_read(&ionic->vf_op_lock);
+
+	if (vf >= pci_num_vf(ionic->pdev) || !ionic->vfs) {
+		ret = -EINVAL;
+	} else {
+		ret = ionic_set_vf_config(ionic, vf,
+					  IONIC_VF_ATTR_VLAN, (u8 *)&vlan);
+		if (!ret)
+			ionic->vfs[vf].vlanid = vlan;
+	}
+
+	up_read(&ionic->vf_op_lock);
+	return ret;
+}
+
+static int ionic_set_vf_rate(struct net_device *netdev, int vf,
+			     int tx_min, int tx_max)
+{
+	struct ionic_lif *lif = netdev_priv(netdev);
+	struct ionic *ionic = lif->ionic;
+	int ret;
+
+	/* setting the min just seems silly */
+	if (tx_min)
+		return -EINVAL;
+
+	down_write(&ionic->vf_op_lock);
+
+	if (vf >= pci_num_vf(ionic->pdev) || !ionic->vfs) {
+		ret = -EINVAL;
+	} else {
+		ret = ionic_set_vf_config(ionic, vf,
+					  IONIC_VF_ATTR_RATE, (u8 *)&tx_max);
+		if (!ret)
+			lif->ionic->vfs[vf].maxrate = tx_max;
+	}
+
+	up_write(&ionic->vf_op_lock);
+	return ret;
+}
+
+static int ionic_set_vf_spoofchk(struct net_device *netdev, int vf, bool set)
+{
+	struct ionic_lif *lif = netdev_priv(netdev);
+	struct ionic *ionic = lif->ionic;
+	u8 data = set;  /* convert to u8 for config */
+	int ret;
+
+	down_write(&ionic->vf_op_lock);
+
+	if (vf >= pci_num_vf(ionic->pdev) || !ionic->vfs) {
+		ret = -EINVAL;
+	} else {
+		ret = ionic_set_vf_config(ionic, vf,
+					  IONIC_VF_ATTR_SPOOFCHK, &data);
+		if (!ret)
+			ionic->vfs[vf].spoofchk = data;
+	}
+
+	up_write(&ionic->vf_op_lock);
+	return ret;
+}
+
+static int ionic_set_vf_trust(struct net_device *netdev, int vf, bool set)
+{
+	struct ionic_lif *lif = netdev_priv(netdev);
+	struct ionic *ionic = lif->ionic;
+	u8 data = set;  /* convert to u8 for config */
+	int ret;
+
+	down_write(&ionic->vf_op_lock);
+
+	if (vf >= pci_num_vf(ionic->pdev) || !ionic->vfs) {
+		ret = -EINVAL;
+	} else {
+		ret = ionic_set_vf_config(ionic, vf,
+					  IONIC_VF_ATTR_TRUST, &data);
+		if (!ret)
+			ionic->vfs[vf].trusted = data;
+	}
+
+	up_write(&ionic->vf_op_lock);
+	return ret;
+}
+
+static int ionic_set_vf_link_state(struct net_device *netdev, int vf, int set)
+{
+	struct ionic_lif *lif = netdev_priv(netdev);
+	struct ionic *ionic = lif->ionic;
+	u8 data;
+	int ret;
+
+	switch (set) {
+	case IFLA_VF_LINK_STATE_ENABLE:
+		data = IONIC_VF_LINK_STATUS_UP;
+		break;
+	case IFLA_VF_LINK_STATE_DISABLE:
+		data = IONIC_VF_LINK_STATUS_DOWN;
+		break;
+	case IFLA_VF_LINK_STATE_AUTO:
+		data = IONIC_VF_LINK_STATUS_AUTO;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	down_write(&ionic->vf_op_lock);
+
+	if (vf >= pci_num_vf(ionic->pdev) || !ionic->vfs) {
+		ret = -EINVAL;
+	} else {
+		ret = ionic_set_vf_config(ionic, vf,
+					  IONIC_VF_ATTR_LINKSTATE, &data);
+		if (!ret)
+			ionic->vfs[vf].linkstate = set;
+	}
+
+	up_write(&ionic->vf_op_lock);
+	return ret;
+}
+
 static const struct net_device_ops ionic_netdev_ops = {
 	.ndo_open               = ionic_open,
 	.ndo_stop               = ionic_stop,
@@ -1632,6 +1853,14 @@ static const struct net_device_ops ionic_netdev_ops = {
 	.ndo_change_mtu         = ionic_change_mtu,
 	.ndo_vlan_rx_add_vid    = ionic_vlan_rx_add_vid,
 	.ndo_vlan_rx_kill_vid   = ionic_vlan_rx_kill_vid,
+	.ndo_set_vf_vlan	= ionic_set_vf_vlan,
+	.ndo_set_vf_trust	= ionic_set_vf_trust,
+	.ndo_set_vf_mac		= ionic_set_vf_mac,
+	.ndo_set_vf_rate	= ionic_set_vf_rate,
+	.ndo_set_vf_spoofchk	= ionic_set_vf_spoofchk,
+	.ndo_get_vf_config	= ionic_get_vf_config,
+	.ndo_set_vf_link_state	= ionic_set_vf_link_state,
+	.ndo_get_vf_stats       = ionic_get_vf_stats,
 };
 
 int ionic_reset_queues(struct ionic_lif *lif)
@@ -1965,18 +2194,22 @@ static int ionic_station_set(struct ionic_lif *lif)
 	if (err)
 		return err;
 
+	if (is_zero_ether_addr(ctx.comp.lif_getattr.mac))
+		return 0;
+
 	memcpy(addr.sa_data, ctx.comp.lif_getattr.mac, netdev->addr_len);
 	addr.sa_family = AF_INET;
 	err = eth_prepare_mac_addr_change(netdev, &addr);
-	if (err)
-		return err;
-
-	if (!is_zero_ether_addr(netdev->dev_addr)) {
-		netdev_dbg(lif->netdev, "deleting station MAC addr %pM\n",
-			   netdev->dev_addr);
-		ionic_lif_addr(lif, netdev->dev_addr, false);
+	if (err) {
+		netdev_warn(lif->netdev, "ignoring bad MAC addr from NIC %pM\n",
+			    addr.sa_data);
+		return 0;
 	}
 
+	netdev_dbg(lif->netdev, "deleting station MAC addr %pM\n",
+		   netdev->dev_addr);
+	ionic_lif_addr(lif, netdev->dev_addr, false);
+
 	eth_commit_mac_addr_change(netdev, &addr);
 	netdev_dbg(lif->netdev, "adding station MAC addr %pM\n",
 		   netdev->dev_addr);
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_main.c b/drivers/net/ethernet/pensando/ionic/ionic_main.c
index 3590ea7fd88a..837b85f2fed9 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_main.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_main.c
@@ -165,6 +165,10 @@ static const char *ionic_opcode_to_str(enum ionic_cmd_opcode opcode)
 		return "IONIC_CMD_FW_DOWNLOAD";
 	case IONIC_CMD_FW_CONTROL:
 		return "IONIC_CMD_FW_CONTROL";
+	case IONIC_CMD_VF_GETATTR:
+		return "IONIC_CMD_VF_GETATTR";
+	case IONIC_CMD_VF_SETATTR:
+		return "IONIC_CMD_VF_SETATTR";
 	default:
 		return "DEVCMD_UNKNOWN";
 	}
-- 
cgit v1.2.3-59-g8ed1b


From c114574ebfdf42f826776f717c8056a00fa94881 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Fri, 3 Jan 2020 20:43:17 +0000
Subject: net: phy: add PHY_INTERFACE_MODE_10GBASER

Recent discussion has revealed that the use of PHY_INTERFACE_MODE_10GKR
is incorrect. Add a 10GBASE-R definition, document both the -R and -KR
versions, and the fact that 10GKR was used incorrectly.

Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/phy.rst | 18 ++++++++++++++++++
 include/linux/phy.h              | 12 ++++++++----
 2 files changed, 26 insertions(+), 4 deletions(-)

diff --git a/Documentation/networking/phy.rst b/Documentation/networking/phy.rst
index e0a7c7af6525..1e4735cc0553 100644
--- a/Documentation/networking/phy.rst
+++ b/Documentation/networking/phy.rst
@@ -267,6 +267,24 @@ Some of the interface modes are described below:
     duplex, pause or other settings.  This is dependent on the MAC and/or
     PHY behaviour.
 
+``PHY_INTERFACE_MODE_10GBASER``
+    This is the IEEE 802.3 Clause 49 defined 10GBASE-R protocol used with
+    various different mediums. Please refer to the IEEE standard for a
+    definition of this.
+
+    Note: 10GBASE-R is just one protocol that can be used with XFI and SFI.
+    XFI and SFI permit multiple protocols over a single SERDES lane, and
+    also defines the electrical characteristics of the signals with a host
+    compliance board plugged into the host XFP/SFP connector. Therefore,
+    XFI and SFI are not PHY interface types in their own right.
+
+``PHY_INTERFACE_MODE_10GKR``
+    This is the IEEE 802.3 Clause 49 defined 10GBASE-R with Clause 73
+    autonegotiation. Please refer to the IEEE standard for further
+    information.
+
+    Note: due to legacy usage, some 10GBASE-R usage incorrectly makes
+    use of this definition.
 
 Pause frames / flow control
 ===========================
diff --git a/include/linux/phy.h b/include/linux/phy.h
index 30e599c454db..5932bb8e9c35 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -100,9 +100,11 @@ typedef enum {
 	PHY_INTERFACE_MODE_2500BASEX,
 	PHY_INTERFACE_MODE_RXAUI,
 	PHY_INTERFACE_MODE_XAUI,
-	/* 10GBASE-KR, XFI, SFI - single lane 10G Serdes */
-	PHY_INTERFACE_MODE_10GKR,
+	/* 10GBASE-R, XFI, SFI - single lane 10G Serdes */
+	PHY_INTERFACE_MODE_10GBASER,
 	PHY_INTERFACE_MODE_USXGMII,
+	/* 10GBASE-KR - with Clause 73 AN */
+	PHY_INTERFACE_MODE_10GKR,
 	PHY_INTERFACE_MODE_MAX,
 } phy_interface_t;
 
@@ -176,10 +178,12 @@ static inline const char *phy_modes(phy_interface_t interface)
 		return "rxaui";
 	case PHY_INTERFACE_MODE_XAUI:
 		return "xaui";
-	case PHY_INTERFACE_MODE_10GKR:
-		return "10gbase-kr";
+	case PHY_INTERFACE_MODE_10GBASER:
+		return "10gbase-r";
 	case PHY_INTERFACE_MODE_USXGMII:
 		return "usxgmii";
+	case PHY_INTERFACE_MODE_10GKR:
+		return "10gbase-kr";
 	default:
 		return "unknown";
 	}
-- 
cgit v1.2.3-59-g8ed1b


From e0f909bc3a242296da9ccff78277f26d4883a79d Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Fri, 3 Jan 2020 20:43:23 +0000
Subject: net: switch to using PHY_INTERFACE_MODE_10GBASER rather than 10GKR

Switch network drivers, phy drivers, and SFP/phylink over to use the
more correct 10GBASE-R, rather than 10GBASE-KR. 10GBASE-KR is backplane
ethernet, which is 10GBASE-R with autonegotiation on top, which our
current usage on the affected platforms does not have.

The only remaining user of PHY_INTERFACE_MODE_10GKR is the Aquantia
PHY, which has a separate mode for 10GBASE-KR.

For Marvell mvpp2, we detect 10GBASE-KR, and rewrite it to 10GBASE-R
for compatibility with existing DT - this is the only network driver
at present that makes use of PHY_INTERFACE_MODE_10GKR.

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c | 19 ++++++++++++++-----
 drivers/net/phy/aquantia_main.c                 |  7 +++++--
 drivers/net/phy/bcm84881.c                      |  4 ++--
 drivers/net/phy/marvell10g.c                    | 11 ++++++-----
 drivers/net/phy/phylink.c                       |  1 +
 drivers/net/phy/sfp-bus.c                       |  2 +-
 drivers/phy/marvell/phy-mvebu-cp110-comphy.c    | 20 ++++++++++----------
 7 files changed, 39 insertions(+), 25 deletions(-)

diff --git a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
index cd32c7196a7f..72133cbe55d4 100644
--- a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
+++ b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
@@ -1114,7 +1114,7 @@ mvpp2_shared_interrupt_mask_unmask(struct mvpp2_port *port, bool mask)
 /* Port configuration routines */
 static bool mvpp2_is_xlg(phy_interface_t interface)
 {
-	return interface == PHY_INTERFACE_MODE_10GKR ||
+	return interface == PHY_INTERFACE_MODE_10GBASER ||
 	       interface == PHY_INTERFACE_MODE_XAUI;
 }
 
@@ -1200,7 +1200,7 @@ static int mvpp22_gop_init(struct mvpp2_port *port)
 	case PHY_INTERFACE_MODE_2500BASEX:
 		mvpp22_gop_init_sgmii(port);
 		break;
-	case PHY_INTERFACE_MODE_10GKR:
+	case PHY_INTERFACE_MODE_10GBASER:
 		if (port->gop_id != 0)
 			goto invalid_conf;
 		mvpp22_gop_init_10gkr(port);
@@ -1649,7 +1649,7 @@ static void mvpp22_pcs_reset_deassert(struct mvpp2_port *port)
 	xpcs = priv->iface_base + MVPP22_XPCS_BASE(port->gop_id);
 
 	switch (port->phy_interface) {
-	case PHY_INTERFACE_MODE_10GKR:
+	case PHY_INTERFACE_MODE_10GBASER:
 		val = readl(mpcs + MVPP22_MPCS_CLK_RESET);
 		val |= MAC_CLK_RESET_MAC | MAC_CLK_RESET_SD_RX |
 		       MAC_CLK_RESET_SD_TX;
@@ -4758,7 +4758,7 @@ static void mvpp2_phylink_validate(struct phylink_config *config,
 
 	/* Invalid combinations */
 	switch (state->interface) {
-	case PHY_INTERFACE_MODE_10GKR:
+	case PHY_INTERFACE_MODE_10GBASER:
 	case PHY_INTERFACE_MODE_XAUI:
 		if (port->gop_id != 0)
 			goto empty_set;
@@ -4780,7 +4780,7 @@ static void mvpp2_phylink_validate(struct phylink_config *config,
 	phylink_set(mask, Asym_Pause);
 
 	switch (state->interface) {
-	case PHY_INTERFACE_MODE_10GKR:
+	case PHY_INTERFACE_MODE_10GBASER:
 	case PHY_INTERFACE_MODE_XAUI:
 	case PHY_INTERFACE_MODE_NA:
 		if (port->gop_id == 0) {
@@ -5247,6 +5247,15 @@ static int mvpp2_port_probe(struct platform_device *pdev,
 		goto err_free_netdev;
 	}
 
+	/*
+	 * Rewrite 10GBASE-KR to 10GBASE-R for compatibility with existing DT.
+	 * Existing usage of 10GBASE-KR is not correct; no backplane
+	 * negotiation is done, and this driver does not actually support
+	 * 10GBASE-KR.
+	 */
+	if (phy_mode == PHY_INTERFACE_MODE_10GKR)
+		phy_mode = PHY_INTERFACE_MODE_10GBASER;
+
 	if (port_node) {
 		comphy = devm_of_phy_get(&pdev->dev, port_node, NULL);
 		if (IS_ERR(comphy)) {
diff --git a/drivers/net/phy/aquantia_main.c b/drivers/net/phy/aquantia_main.c
index 975789d9349d..31927b2c7d5a 100644
--- a/drivers/net/phy/aquantia_main.c
+++ b/drivers/net/phy/aquantia_main.c
@@ -358,9 +358,11 @@ static int aqr107_read_status(struct phy_device *phydev)
 
 	switch (FIELD_GET(MDIO_PHYXS_VEND_IF_STATUS_TYPE_MASK, val)) {
 	case MDIO_PHYXS_VEND_IF_STATUS_TYPE_KR:
-	case MDIO_PHYXS_VEND_IF_STATUS_TYPE_XFI:
 		phydev->interface = PHY_INTERFACE_MODE_10GKR;
 		break;
+	case MDIO_PHYXS_VEND_IF_STATUS_TYPE_XFI:
+		phydev->interface = PHY_INTERFACE_MODE_10GBASER;
+		break;
 	case MDIO_PHYXS_VEND_IF_STATUS_TYPE_USXGMII:
 		phydev->interface = PHY_INTERFACE_MODE_USXGMII;
 		break;
@@ -493,7 +495,8 @@ static int aqr107_config_init(struct phy_device *phydev)
 	    phydev->interface != PHY_INTERFACE_MODE_2500BASEX &&
 	    phydev->interface != PHY_INTERFACE_MODE_XGMII &&
 	    phydev->interface != PHY_INTERFACE_MODE_USXGMII &&
-	    phydev->interface != PHY_INTERFACE_MODE_10GKR)
+	    phydev->interface != PHY_INTERFACE_MODE_10GKR &&
+	    phydev->interface != PHY_INTERFACE_MODE_10GBASER)
 		return -ENODEV;
 
 	WARN(phydev->interface == PHY_INTERFACE_MODE_XGMII,
diff --git a/drivers/net/phy/bcm84881.c b/drivers/net/phy/bcm84881.c
index db59911b9b3c..14d55a77eb28 100644
--- a/drivers/net/phy/bcm84881.c
+++ b/drivers/net/phy/bcm84881.c
@@ -53,7 +53,7 @@ static int bcm84881_config_init(struct phy_device *phydev)
 	switch (phydev->interface) {
 	case PHY_INTERFACE_MODE_SGMII:
 	case PHY_INTERFACE_MODE_2500BASEX:
-	case PHY_INTERFACE_MODE_10GKR:
+	case PHY_INTERFACE_MODE_10GBASER:
 		break;
 	default:
 		return -ENODEV;
@@ -218,7 +218,7 @@ static int bcm84881_read_status(struct phy_device *phydev)
 	if (mode == 1 || mode == 2)
 		phydev->interface = PHY_INTERFACE_MODE_SGMII;
 	else if (mode == 3)
-		phydev->interface = PHY_INTERFACE_MODE_10GKR;
+		phydev->interface = PHY_INTERFACE_MODE_10GBASER;
 	else if (mode == 4)
 		phydev->interface = PHY_INTERFACE_MODE_2500BASEX;
 	switch (mode & 7) {
diff --git a/drivers/net/phy/marvell10g.c b/drivers/net/phy/marvell10g.c
index 512f27b0b5cd..64c9f3bba2cd 100644
--- a/drivers/net/phy/marvell10g.c
+++ b/drivers/net/phy/marvell10g.c
@@ -216,7 +216,7 @@ static int mv3310_sfp_insert(void *upstream, const struct sfp_eeprom_id *id)
 	sfp_parse_support(phydev->sfp_bus, id, support);
 	iface = sfp_select_interface(phydev->sfp_bus, support);
 
-	if (iface != PHY_INTERFACE_MODE_10GKR) {
+	if (iface != PHY_INTERFACE_MODE_10GBASER) {
 		dev_err(&phydev->mdio.dev, "incompatible SFP module inserted\n");
 		return -EINVAL;
 	}
@@ -304,7 +304,7 @@ static int mv3310_config_init(struct phy_device *phydev)
 	    phydev->interface != PHY_INTERFACE_MODE_2500BASEX &&
 	    phydev->interface != PHY_INTERFACE_MODE_XAUI &&
 	    phydev->interface != PHY_INTERFACE_MODE_RXAUI &&
-	    phydev->interface != PHY_INTERFACE_MODE_10GKR)
+	    phydev->interface != PHY_INTERFACE_MODE_10GBASER)
 		return -ENODEV;
 
 	return 0;
@@ -386,16 +386,17 @@ static void mv3310_update_interface(struct phy_device *phydev)
 {
 	if ((phydev->interface == PHY_INTERFACE_MODE_SGMII ||
 	     phydev->interface == PHY_INTERFACE_MODE_2500BASEX ||
-	     phydev->interface == PHY_INTERFACE_MODE_10GKR) && phydev->link) {
+	     phydev->interface == PHY_INTERFACE_MODE_10GBASER) &&
+	    phydev->link) {
 		/* The PHY automatically switches its serdes interface (and
-		 * active PHYXS instance) between Cisco SGMII, 10GBase-KR and
+		 * active PHYXS instance) between Cisco SGMII, 10GBase-R and
 		 * 2500BaseX modes according to the speed.  Florian suggests
 		 * setting phydev->interface to communicate this to the MAC.
 		 * Only do this if we are already in one of the above modes.
 		 */
 		switch (phydev->speed) {
 		case SPEED_10000:
-			phydev->interface = PHY_INTERFACE_MODE_10GKR;
+			phydev->interface = PHY_INTERFACE_MODE_10GBASER;
 			break;
 		case SPEED_2500:
 			phydev->interface = PHY_INTERFACE_MODE_2500BASEX;
diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index 3f7fa634134a..1edca9725370 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -298,6 +298,7 @@ static int phylink_parse_mode(struct phylink *pl, struct fwnode_handle *fwnode)
 			break;
 
 		case PHY_INTERFACE_MODE_10GKR:
+		case PHY_INTERFACE_MODE_10GBASER:
 			phylink_set(pl->supported, 10baseT_Half);
 			phylink_set(pl->supported, 10baseT_Full);
 			phylink_set(pl->supported, 100baseT_Half);
diff --git a/drivers/net/phy/sfp-bus.c b/drivers/net/phy/sfp-bus.c
index 06e6429b8b71..d949ea7b4f8c 100644
--- a/drivers/net/phy/sfp-bus.c
+++ b/drivers/net/phy/sfp-bus.c
@@ -373,7 +373,7 @@ phy_interface_t sfp_select_interface(struct sfp_bus *bus,
 	    phylink_test(link_modes, 10000baseLRM_Full) ||
 	    phylink_test(link_modes, 10000baseER_Full) ||
 	    phylink_test(link_modes, 10000baseT_Full))
-		return PHY_INTERFACE_MODE_10GKR;
+		return PHY_INTERFACE_MODE_10GBASER;
 
 	if (phylink_test(link_modes, 2500baseX_Full))
 		return PHY_INTERFACE_MODE_2500BASEX;
diff --git a/drivers/phy/marvell/phy-mvebu-cp110-comphy.c b/drivers/phy/marvell/phy-mvebu-cp110-comphy.c
index e3b87c94aaf6..e41367f36ee1 100644
--- a/drivers/phy/marvell/phy-mvebu-cp110-comphy.c
+++ b/drivers/phy/marvell/phy-mvebu-cp110-comphy.c
@@ -221,7 +221,7 @@ static const struct mvebu_comphy_conf mvebu_comphy_cp110_modes[] = {
 	ETH_CONF(2, 0, PHY_INTERFACE_MODE_SGMII, 0x1, COMPHY_FW_MODE_SGMII),
 	ETH_CONF(2, 0, PHY_INTERFACE_MODE_2500BASEX, 0x1, COMPHY_FW_MODE_HS_SGMII),
 	ETH_CONF(2, 0, PHY_INTERFACE_MODE_RXAUI, 0x1, COMPHY_FW_MODE_RXAUI),
-	ETH_CONF(2, 0, PHY_INTERFACE_MODE_10GKR, 0x1, COMPHY_FW_MODE_XFI),
+	ETH_CONF(2, 0, PHY_INTERFACE_MODE_10GBASER, 0x1, COMPHY_FW_MODE_XFI),
 	GEN_CONF(2, 0, PHY_MODE_USB_HOST_SS, COMPHY_FW_MODE_USB3H),
 	GEN_CONF(2, 0, PHY_MODE_SATA, COMPHY_FW_MODE_SATA),
 	GEN_CONF(2, 0, PHY_MODE_PCIE, COMPHY_FW_MODE_PCIE),
@@ -235,14 +235,14 @@ static const struct mvebu_comphy_conf mvebu_comphy_cp110_modes[] = {
 	/* lane 4 */
 	ETH_CONF(4, 0, PHY_INTERFACE_MODE_SGMII, 0x2, COMPHY_FW_MODE_SGMII),
 	ETH_CONF(4, 0, PHY_INTERFACE_MODE_2500BASEX, 0x2, COMPHY_FW_MODE_HS_SGMII),
-	ETH_CONF(4, 0, PHY_INTERFACE_MODE_10GKR, 0x2, COMPHY_FW_MODE_XFI),
+	ETH_CONF(4, 0, PHY_INTERFACE_MODE_10GBASER, 0x2, COMPHY_FW_MODE_XFI),
 	ETH_CONF(4, 0, PHY_INTERFACE_MODE_RXAUI, 0x2, COMPHY_FW_MODE_RXAUI),
 	GEN_CONF(4, 0, PHY_MODE_USB_DEVICE_SS, COMPHY_FW_MODE_USB3D),
 	GEN_CONF(4, 1, PHY_MODE_USB_HOST_SS, COMPHY_FW_MODE_USB3H),
 	GEN_CONF(4, 1, PHY_MODE_PCIE, COMPHY_FW_MODE_PCIE),
 	ETH_CONF(4, 1, PHY_INTERFACE_MODE_SGMII, 0x1, COMPHY_FW_MODE_SGMII),
 	ETH_CONF(4, 1, PHY_INTERFACE_MODE_2500BASEX, -1, COMPHY_FW_MODE_HS_SGMII),
-	ETH_CONF(4, 1, PHY_INTERFACE_MODE_10GKR, -1, COMPHY_FW_MODE_XFI),
+	ETH_CONF(4, 1, PHY_INTERFACE_MODE_10GBASER, -1, COMPHY_FW_MODE_XFI),
 	/* lane 5 */
 	ETH_CONF(5, 1, PHY_INTERFACE_MODE_RXAUI, 0x2, COMPHY_FW_MODE_RXAUI),
 	GEN_CONF(5, 1, PHY_MODE_SATA, COMPHY_FW_MODE_SATA),
@@ -342,7 +342,7 @@ static int mvebu_comphy_ethernet_init_reset(struct mvebu_comphy_lane *lane)
 		 MVEBU_COMPHY_SERDES_CFG0_RXAUI_MODE);
 
 	switch (lane->submode) {
-	case PHY_INTERFACE_MODE_10GKR:
+	case PHY_INTERFACE_MODE_10GBASER:
 		val |= MVEBU_COMPHY_SERDES_CFG0_GEN_RX(0xe) |
 		       MVEBU_COMPHY_SERDES_CFG0_GEN_TX(0xe);
 		break;
@@ -417,7 +417,7 @@ static int mvebu_comphy_ethernet_init_reset(struct mvebu_comphy_lane *lane)
 	/* refclk selection */
 	val = readl(priv->base + MVEBU_COMPHY_MISC_CTRL0(lane->id));
 	val &= ~MVEBU_COMPHY_MISC_CTRL0_REFCLK_SEL;
-	if (lane->submode == PHY_INTERFACE_MODE_10GKR)
+	if (lane->submode == PHY_INTERFACE_MODE_10GBASER)
 		val |= MVEBU_COMPHY_MISC_CTRL0_ICP_FORCE;
 	writel(val, priv->base + MVEBU_COMPHY_MISC_CTRL0(lane->id));
 
@@ -564,7 +564,7 @@ static int mvebu_comphy_set_mode_rxaui(struct phy *phy)
 	return mvebu_comphy_init_plls(lane);
 }
 
-static int mvebu_comphy_set_mode_10gkr(struct phy *phy)
+static int mvebu_comphy_set_mode_10gbaser(struct phy *phy)
 {
 	struct mvebu_comphy_lane *lane = phy_get_drvdata(phy);
 	struct mvebu_comphy_priv *priv = lane->priv;
@@ -735,8 +735,8 @@ static int mvebu_comphy_power_on_legacy(struct phy *phy)
 	case PHY_INTERFACE_MODE_RXAUI:
 		ret = mvebu_comphy_set_mode_rxaui(phy);
 		break;
-	case PHY_INTERFACE_MODE_10GKR:
-		ret = mvebu_comphy_set_mode_10gkr(phy);
+	case PHY_INTERFACE_MODE_10GBASER:
+		ret = mvebu_comphy_set_mode_10gbaser(phy);
 		break;
 	default:
 		return -ENOTSUPP;
@@ -782,8 +782,8 @@ static int mvebu_comphy_power_on(struct phy *phy)
 				lane->id);
 			fw_speed = COMPHY_FW_SPEED_3125;
 			break;
-		case PHY_INTERFACE_MODE_10GKR:
-			dev_dbg(priv->dev, "set lane %d to 10G-KR mode\n",
+		case PHY_INTERFACE_MODE_10GBASER:
+			dev_dbg(priv->dev, "set lane %d to 10GBASE-R mode\n",
 				lane->id);
 			fw_speed = COMPHY_FW_SPEED_103125;
 			break;
-- 
cgit v1.2.3-59-g8ed1b


From 0a51826c6e05c5b6cc423b376b81c311e9e485b0 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <olteanv@gmail.com>
Date: Sat, 4 Jan 2020 02:37:09 +0200
Subject: net: dsa: sja1105: Always send through management routes in slot 0

I finally found out how the 4 management route slots are supposed to
be used, but.. it's not worth it.

The description from the comment I've just deleted in this commit is
still true: when more than 1 management slot is active at the same time,
the switch will match frames incoming [from the CPU port] on the lowest
numbered management slot that matches the frame's DMAC.

My issue was that one was not supposed to statically assign each port a
slot. Yes, there are 4 slots and also 4 non-CPU ports, but that is a
mere coincidence.

Instead, the switch can be used like this: every management frame gets a
slot at the right of the most recently assigned slot:

Send mgmt frame 1 through S0:    S0 x  x  x
Send mgmt frame 2 through S1:    S0 S1 x  x
Send mgmt frame 3 through S2:    S0 S1 S2 x
Send mgmt frame 4 through S3:    S0 S1 S2 S3

The difference compared to the old usage is that the transmission of
frames 1-4 doesn't need to wait until the completion of the management
route. It is safe to use a slot to the right of the most recently used
one, because by protocol nobody will program a slot to your left and
"steal" your route towards the correct egress port.

So there is a potential throughput benefit here.

But mgmt frame 5 has no more free slot to use, so it has to wait until
_all_ of S0, S1, S2, S3 are full, in order to use S0 again.

And that's actually exactly the problem: I was looking for something
that would bring more predictable transmission latency, but this is
exactly the opposite: 3 out of 4 frames would be transmitted quicker,
but the 4th would draw the short straw and have a worse worst-case
latency than before.

Useless.

Things are made even worse by PTP TX timestamping, which is something I
won't go deeply into here. Suffice to say that the fact there is a
driver-level lock on the SPI bus offsets any potential throughput gains
that parallelism might bring.

So there's no going back to the multi-slot scheme, remove the
"mgmt_slot" variable from sja1105_port and the dummy static assignment
made at probe time.

While passing by, also remove the assignment to casc_port altogether.
Don't pretend that we support cascaded setups.

Signed-off-by: Vladimir Oltean <olteanv@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/sja1105/sja1105_main.c | 26 +-------------------------
 include/linux/dsa/sja1105.h            |  1 -
 2 files changed, 1 insertion(+), 26 deletions(-)

diff --git a/drivers/net/dsa/sja1105/sja1105_main.c b/drivers/net/dsa/sja1105/sja1105_main.c
index 1da5ac111499..79dd965227bc 100644
--- a/drivers/net/dsa/sja1105/sja1105_main.c
+++ b/drivers/net/dsa/sja1105/sja1105_main.c
@@ -426,14 +426,6 @@ static int sja1105_init_general_params(struct sja1105_private *priv)
 		.tpid2 = ETH_P_SJA1105,
 	};
 	struct sja1105_table *table;
-	int i, k = 0;
-
-	for (i = 0; i < SJA1105_NUM_PORTS; i++) {
-		if (dsa_is_dsa_port(priv->ds, i))
-			default_general_params.casc_port = i;
-		else if (dsa_is_user_port(priv->ds, i))
-			priv->ports[i].mgmt_slot = k++;
-	}
 
 	table = &priv->static_config.tables[BLK_IDX_GENERAL_PARAMS];
 
@@ -1827,30 +1819,14 @@ static netdev_tx_t sja1105_port_deferred_xmit(struct dsa_switch *ds, int port,
 					      struct sk_buff *skb)
 {
 	struct sja1105_private *priv = ds->priv;
-	struct sja1105_port *sp = &priv->ports[port];
-	int slot = sp->mgmt_slot;
 	struct sk_buff *clone;
 
-	/* The tragic fact about the switch having 4x2 slots for installing
-	 * management routes is that all of them except one are actually
-	 * useless.
-	 * If 2 slots are simultaneously configured for two BPDUs sent to the
-	 * same (multicast) DMAC but on different egress ports, the switch
-	 * would confuse them and redirect first frame it receives on the CPU
-	 * port towards the port configured on the numerically first slot
-	 * (therefore wrong port), then second received frame on second slot
-	 * (also wrong port).
-	 * So for all practical purposes, there needs to be a lock that
-	 * prevents that from happening. The slot used here is utterly useless
-	 * (could have simply been 0 just as fine), but we are doing it
-	 * nonetheless, in case a smarter idea ever comes up in the future.
-	 */
 	mutex_lock(&priv->mgmt_lock);
 
 	/* The clone, if there, was made by dsa_skb_tx_timestamp */
 	clone = DSA_SKB_CB(skb)->clone;
 
-	sja1105_mgmt_xmit(ds, port, slot, skb, !!clone);
+	sja1105_mgmt_xmit(ds, port, 0, skb, !!clone);
 
 	if (!clone)
 		goto out;
diff --git a/include/linux/dsa/sja1105.h b/include/linux/dsa/sja1105.h
index c0b6a603ea8c..317e05b2584b 100644
--- a/include/linux/dsa/sja1105.h
+++ b/include/linux/dsa/sja1105.h
@@ -56,7 +56,6 @@ struct sja1105_port {
 	struct sja1105_tagger_data *data;
 	struct dsa_port *dp;
 	bool hwts_tx_en;
-	int mgmt_slot;
 };
 
 #endif /* _NET_DSA_SJA1105_H */
-- 
cgit v1.2.3-59-g8ed1b


From a68578c20a9667463ee3000402b21644ea62d753 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <olteanv@gmail.com>
Date: Sat, 4 Jan 2020 02:37:10 +0200
Subject: net: dsa: Make deferred_xmit private to sja1105

There are 3 things that are wrong with the DSA deferred xmit mechanism:

1. Its introduction has made the DSA hotpath ever so slightly more
   inefficient for everybody, since DSA_SKB_CB(skb)->deferred_xmit needs
   to be initialized to false for every transmitted frame, in order to
   figure out whether the driver requested deferral or not (a very rare
   occasion, rare even for the only driver that does use this mechanism:
   sja1105). That was necessary to avoid kfree_skb from freeing the skb.

2. Because L2 PTP is a link-local protocol like STP, it requires
   management routes and deferred xmit with this switch. But as opposed
   to STP, the deferred work mechanism needs to schedule the packet
   rather quickly for the TX timstamp to be collected in time and sent
   to user space. But there is no provision for controlling the
   scheduling priority of this deferred xmit workqueue. Too bad this is
   a rather specific requirement for a feature that nobody else uses
   (more below).

3. Perhaps most importantly, it makes the DSA core adhere a bit too
   much to the NXP company-wide policy "Innovate Where It Doesn't
   Matter". The sja1105 is probably the only DSA switch that requires
   some frames sent from the CPU to be routed to the slave port via an
   out-of-band configuration (register write) rather than in-band (DSA
   tag). And there are indeed very good reasons to not want to do that:
   if that out-of-band register is at the other end of a slow bus such
   as SPI, then you limit that Ethernet flow's throughput to effectively
   the throughput of the SPI bus. So hardware vendors should definitely
   not be encouraged to design this way. We do _not_ want more
   widespread use of this mechanism.

Luckily we have a solution for each of the 3 issues:

For 1, we can just remove that variable in the skb->cb and counteract
the effect of kfree_skb with skb_get, much to the same effect. The
advantage, of course, being that anybody who doesn't use deferred xmit
doesn't need to do any extra operation in the hotpath.

For 2, we can create a kernel thread for each port's deferred xmit work.
If the user switch ports are named swp0, swp1, swp2, the kernel threads
will be named swp0_xmit, swp1_xmit, swp2_xmit (there appears to be a 15
character length limit on kernel thread names). With this, the user can
change the scheduling priority with chrt $(pidof swp2_xmit).

For 3, we can actually move the entire implementation to the sja1105
driver.

So this patch deletes the generic implementation from the DSA core and
adds a new one, more adequate to the requirements of PTP TX
timestamping, in sja1105_main.c.

Suggested-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Vladimir Oltean <olteanv@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/sja1105/sja1105_main.c | 96 ++++++++++++++++++++++++++--------
 include/linux/dsa/sja1105.h            |  3 ++
 include/net/dsa.h                      |  9 ----
 net/dsa/dsa_priv.h                     |  2 -
 net/dsa/slave.c                        | 37 +------------
 net/dsa/tag_sja1105.c                  | 15 +++++-
 6 files changed, 93 insertions(+), 69 deletions(-)

diff --git a/drivers/net/dsa/sja1105/sja1105_main.c b/drivers/net/dsa/sja1105/sja1105_main.c
index 79dd965227bc..61795833c8f5 100644
--- a/drivers/net/dsa/sja1105/sja1105_main.c
+++ b/drivers/net/dsa/sja1105/sja1105_main.c
@@ -1732,6 +1732,16 @@ static int sja1105_setup(struct dsa_switch *ds)
 static void sja1105_teardown(struct dsa_switch *ds)
 {
 	struct sja1105_private *priv = ds->priv;
+	int port;
+
+	for (port = 0; port < SJA1105_NUM_PORTS; port++) {
+		struct sja1105_port *sp = &priv->ports[port];
+
+		if (!dsa_is_user_port(ds, port))
+			continue;
+
+		kthread_destroy_worker(sp->xmit_worker);
+	}
 
 	sja1105_tas_teardown(ds);
 	sja1105_ptp_clock_unregister(ds);
@@ -1753,6 +1763,18 @@ static int sja1105_port_enable(struct dsa_switch *ds, int port,
 	return 0;
 }
 
+static void sja1105_port_disable(struct dsa_switch *ds, int port)
+{
+	struct sja1105_private *priv = ds->priv;
+	struct sja1105_port *sp = &priv->ports[port];
+
+	if (!dsa_is_user_port(ds, port))
+		return;
+
+	kthread_cancel_work_sync(&sp->xmit_work);
+	skb_queue_purge(&sp->xmit_queue);
+}
+
 static int sja1105_mgmt_xmit(struct dsa_switch *ds, int port, int slot,
 			     struct sk_buff *skb, bool takets)
 {
@@ -1811,31 +1833,36 @@ static int sja1105_mgmt_xmit(struct dsa_switch *ds, int port, int slot,
 	return NETDEV_TX_OK;
 }
 
+#define work_to_port(work) \
+		container_of((work), struct sja1105_port, xmit_work)
+#define tagger_to_sja1105(t) \
+		container_of((t), struct sja1105_private, tagger_data)
+
 /* Deferred work is unfortunately necessary because setting up the management
  * route cannot be done from atomit context (SPI transfer takes a sleepable
  * lock on the bus)
  */
-static netdev_tx_t sja1105_port_deferred_xmit(struct dsa_switch *ds, int port,
-					      struct sk_buff *skb)
+static void sja1105_port_deferred_xmit(struct kthread_work *work)
 {
-	struct sja1105_private *priv = ds->priv;
-	struct sk_buff *clone;
-
-	mutex_lock(&priv->mgmt_lock);
+	struct sja1105_port *sp = work_to_port(work);
+	struct sja1105_tagger_data *tagger_data = sp->data;
+	struct sja1105_private *priv = tagger_to_sja1105(tagger_data);
+	int port = sp - priv->ports;
+	struct sk_buff *skb;
 
-	/* The clone, if there, was made by dsa_skb_tx_timestamp */
-	clone = DSA_SKB_CB(skb)->clone;
+	while ((skb = skb_dequeue(&sp->xmit_queue)) != NULL) {
+		struct sk_buff *clone = DSA_SKB_CB(skb)->clone;
 
-	sja1105_mgmt_xmit(ds, port, 0, skb, !!clone);
+		mutex_lock(&priv->mgmt_lock);
 
-	if (!clone)
-		goto out;
+		sja1105_mgmt_xmit(priv->ds, port, 0, skb, !!clone);
 
-	sja1105_ptp_txtstamp_skb(ds, port, clone);
+		/* The clone, if there, was made by dsa_skb_tx_timestamp */
+		if (clone)
+			sja1105_ptp_txtstamp_skb(priv->ds, port, clone);
 
-out:
-	mutex_unlock(&priv->mgmt_lock);
-	return NETDEV_TX_OK;
+		mutex_unlock(&priv->mgmt_lock);
+	}
 }
 
 /* The MAXAGE setting belongs to the L2 Forwarding Parameters table,
@@ -1966,6 +1993,7 @@ static const struct dsa_switch_ops sja1105_switch_ops = {
 	.get_sset_count		= sja1105_get_sset_count,
 	.get_ts_info		= sja1105_get_ts_info,
 	.port_enable		= sja1105_port_enable,
+	.port_disable		= sja1105_port_disable,
 	.port_fdb_dump		= sja1105_fdb_dump,
 	.port_fdb_add		= sja1105_fdb_add,
 	.port_fdb_del		= sja1105_fdb_del,
@@ -1979,7 +2007,6 @@ static const struct dsa_switch_ops sja1105_switch_ops = {
 	.port_mdb_prepare	= sja1105_mdb_prepare,
 	.port_mdb_add		= sja1105_mdb_add,
 	.port_mdb_del		= sja1105_mdb_del,
-	.port_deferred_xmit	= sja1105_port_deferred_xmit,
 	.port_hwtstamp_get	= sja1105_hwtstamp_get,
 	.port_hwtstamp_set	= sja1105_hwtstamp_set,
 	.port_rxtstamp		= sja1105_port_rxtstamp,
@@ -2031,7 +2058,7 @@ static int sja1105_probe(struct spi_device *spi)
 	struct device *dev = &spi->dev;
 	struct sja1105_private *priv;
 	struct dsa_switch *ds;
-	int rc, i;
+	int rc, port;
 
 	if (!dev->of_node) {
 		dev_err(dev, "No DTS bindings for SJA1105 driver\n");
@@ -2096,15 +2123,42 @@ static int sja1105_probe(struct spi_device *spi)
 		return rc;
 
 	/* Connections between dsa_port and sja1105_port */
-	for (i = 0; i < SJA1105_NUM_PORTS; i++) {
-		struct sja1105_port *sp = &priv->ports[i];
+	for (port = 0; port < SJA1105_NUM_PORTS; port++) {
+		struct sja1105_port *sp = &priv->ports[port];
+		struct dsa_port *dp = dsa_to_port(ds, port);
+		struct net_device *slave;
+
+		if (!dsa_is_user_port(ds, port))
+			continue;
 
-		dsa_to_port(ds, i)->priv = sp;
-		sp->dp = dsa_to_port(ds, i);
+		dp->priv = sp;
+		sp->dp = dp;
 		sp->data = tagger_data;
+		slave = dp->slave;
+		kthread_init_work(&sp->xmit_work, sja1105_port_deferred_xmit);
+		sp->xmit_worker = kthread_create_worker(0, "%s_xmit",
+							slave->name);
+		if (IS_ERR(sp->xmit_worker)) {
+			rc = PTR_ERR(sp->xmit_worker);
+			dev_err(ds->dev,
+				"failed to create deferred xmit thread: %d\n",
+				rc);
+			goto out;
+		}
+		skb_queue_head_init(&sp->xmit_queue);
 	}
 
 	return 0;
+out:
+	while (port-- > 0) {
+		struct sja1105_port *sp = &priv->ports[port];
+
+		if (!dsa_is_user_port(ds, port))
+			continue;
+
+		kthread_destroy_worker(sp->xmit_worker);
+	}
+	return rc;
 }
 
 static int sja1105_remove(struct spi_device *spi)
diff --git a/include/linux/dsa/sja1105.h b/include/linux/dsa/sja1105.h
index 317e05b2584b..fa5735c353cd 100644
--- a/include/linux/dsa/sja1105.h
+++ b/include/linux/dsa/sja1105.h
@@ -53,6 +53,9 @@ struct sja1105_skb_cb {
 	((struct sja1105_skb_cb *)DSA_SKB_CB_PRIV(skb))
 
 struct sja1105_port {
+	struct kthread_worker *xmit_worker;
+	struct kthread_work xmit_work;
+	struct sk_buff_head xmit_queue;
 	struct sja1105_tagger_data *data;
 	struct dsa_port *dp;
 	bool hwts_tx_en;
diff --git a/include/net/dsa.h b/include/net/dsa.h
index da5578db228e..23b1c58656d4 100644
--- a/include/net/dsa.h
+++ b/include/net/dsa.h
@@ -90,7 +90,6 @@ struct dsa_device_ops {
 
 struct dsa_skb_cb {
 	struct sk_buff *clone;
-	bool deferred_xmit;
 };
 
 struct __dsa_skb_cb {
@@ -192,9 +191,6 @@ struct dsa_port {
 	struct phylink		*pl;
 	struct phylink_config	pl_config;
 
-	struct work_struct	xmit_work;
-	struct sk_buff_head	xmit_queue;
-
 	struct list_head list;
 
 	/*
@@ -564,11 +560,6 @@ struct dsa_switch_ops {
 	bool	(*port_rxtstamp)(struct dsa_switch *ds, int port,
 				 struct sk_buff *skb, unsigned int type);
 
-	/*
-	 * Deferred frame Tx
-	 */
-	netdev_tx_t (*port_deferred_xmit)(struct dsa_switch *ds, int port,
-					  struct sk_buff *skb);
 	/* Devlink parameters */
 	int	(*devlink_param_get)(struct dsa_switch *ds, u32 id,
 				     struct devlink_param_gset_ctx *ctx);
diff --git a/net/dsa/dsa_priv.h b/net/dsa/dsa_priv.h
index 09ea2fd78c74..8a162605b861 100644
--- a/net/dsa/dsa_priv.h
+++ b/net/dsa/dsa_priv.h
@@ -162,8 +162,6 @@ int dsa_slave_resume(struct net_device *slave_dev);
 int dsa_slave_register_notifier(void);
 void dsa_slave_unregister_notifier(void);
 
-void *dsa_defer_xmit(struct sk_buff *skb, struct net_device *dev);
-
 static inline struct dsa_port *dsa_slave_to_port(const struct net_device *dev)
 {
 	struct dsa_slave_priv *p = netdev_priv(dev);
diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index 78ffc87dc25e..c1828bdc79dc 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -116,9 +116,6 @@ static int dsa_slave_close(struct net_device *dev)
 	struct net_device *master = dsa_slave_to_master(dev);
 	struct dsa_port *dp = dsa_slave_to_port(dev);
 
-	cancel_work_sync(&dp->xmit_work);
-	skb_queue_purge(&dp->xmit_queue);
-
 	phylink_stop(dp->pl);
 
 	dsa_port_disable(dp);
@@ -518,7 +515,6 @@ static netdev_tx_t dsa_slave_xmit(struct sk_buff *skb, struct net_device *dev)
 	s->tx_bytes += skb->len;
 	u64_stats_update_end(&s->syncp);
 
-	DSA_SKB_CB(skb)->deferred_xmit = false;
 	DSA_SKB_CB(skb)->clone = NULL;
 
 	/* Identify PTP protocol packets, clone them, and pass them to the
@@ -531,39 +527,13 @@ static netdev_tx_t dsa_slave_xmit(struct sk_buff *skb, struct net_device *dev)
 	 */
 	nskb = p->xmit(skb, dev);
 	if (!nskb) {
-		if (!DSA_SKB_CB(skb)->deferred_xmit)
-			kfree_skb(skb);
+		kfree_skb(skb);
 		return NETDEV_TX_OK;
 	}
 
 	return dsa_enqueue_skb(nskb, dev);
 }
 
-void *dsa_defer_xmit(struct sk_buff *skb, struct net_device *dev)
-{
-	struct dsa_port *dp = dsa_slave_to_port(dev);
-
-	DSA_SKB_CB(skb)->deferred_xmit = true;
-
-	skb_queue_tail(&dp->xmit_queue, skb);
-	schedule_work(&dp->xmit_work);
-	return NULL;
-}
-EXPORT_SYMBOL_GPL(dsa_defer_xmit);
-
-static void dsa_port_xmit_work(struct work_struct *work)
-{
-	struct dsa_port *dp = container_of(work, struct dsa_port, xmit_work);
-	struct dsa_switch *ds = dp->ds;
-	struct sk_buff *skb;
-
-	if (unlikely(!ds->ops->port_deferred_xmit))
-		return;
-
-	while ((skb = skb_dequeue(&dp->xmit_queue)) != NULL)
-		ds->ops->port_deferred_xmit(ds, dp->index, skb);
-}
-
 /* ethtool operations *******************************************************/
 
 static void dsa_slave_get_drvinfo(struct net_device *dev,
@@ -1367,9 +1337,6 @@ int dsa_slave_suspend(struct net_device *slave_dev)
 	if (!netif_running(slave_dev))
 		return 0;
 
-	cancel_work_sync(&dp->xmit_work);
-	skb_queue_purge(&dp->xmit_queue);
-
 	netif_device_detach(slave_dev);
 
 	rtnl_lock();
@@ -1455,8 +1422,6 @@ int dsa_slave_create(struct dsa_port *port)
 	}
 	p->dp = port;
 	INIT_LIST_HEAD(&p->mall_tc_list);
-	INIT_WORK(&port->xmit_work, dsa_port_xmit_work);
-	skb_queue_head_init(&port->xmit_queue);
 	p->xmit = cpu_dp->tag_ops->xmit;
 	port->slave = slave_dev;
 
diff --git a/net/dsa/tag_sja1105.c b/net/dsa/tag_sja1105.c
index 63ef2a14c934..7c2b84393cc6 100644
--- a/net/dsa/tag_sja1105.c
+++ b/net/dsa/tag_sja1105.c
@@ -83,6 +83,19 @@ static bool sja1105_filter(const struct sk_buff *skb, struct net_device *dev)
 	return false;
 }
 
+/* Calls sja1105_port_deferred_xmit in sja1105_main.c */
+static struct sk_buff *sja1105_defer_xmit(struct sja1105_port *sp,
+					  struct sk_buff *skb)
+{
+	/* Increase refcount so the kfree_skb in dsa_slave_xmit
+	 * won't really free the packet.
+	 */
+	skb_queue_tail(&sp->xmit_queue, skb_get(skb));
+	kthread_queue_work(sp->xmit_worker, &sp->xmit_work);
+
+	return NULL;
+}
+
 static struct sk_buff *sja1105_xmit(struct sk_buff *skb,
 				    struct net_device *netdev)
 {
@@ -97,7 +110,7 @@ static struct sk_buff *sja1105_xmit(struct sk_buff *skb,
 	 * is the .port_deferred_xmit driver callback.
 	 */
 	if (unlikely(sja1105_is_link_local(skb)))
-		return dsa_defer_xmit(skb, netdev);
+		return sja1105_defer_xmit(dp->priv, skb);
 
 	/* If we are under a vlan_filtering bridge, IP termination on
 	 * switch ports based on 802.1Q tags is simply too brittle to
-- 
cgit v1.2.3-59-g8ed1b


From 2821d50fc0c45e45bc10781d4dd332e21e7fb980 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <olteanv@gmail.com>
Date: Sat, 4 Jan 2020 02:37:11 +0200
Subject: net: dsa: tag_sja1105: Slightly improve the Xmas tree in sja1105_xmit

This is a cosmetic patch that makes the dp, tx_vid, queue_mapping and
pcp local variable definitions a bit closer in length, so they don't
look like an eyesore as much.

The 'ds' variable is not used otherwise, except for ds->dp.

Signed-off-by: Vladimir Oltean <olteanv@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/dsa/tag_sja1105.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/net/dsa/tag_sja1105.c b/net/dsa/tag_sja1105.c
index 7c2b84393cc6..5366ea430349 100644
--- a/net/dsa/tag_sja1105.c
+++ b/net/dsa/tag_sja1105.c
@@ -100,8 +100,7 @@ static struct sk_buff *sja1105_xmit(struct sk_buff *skb,
 				    struct net_device *netdev)
 {
 	struct dsa_port *dp = dsa_slave_to_port(netdev);
-	struct dsa_switch *ds = dp->ds;
-	u16 tx_vid = dsa_8021q_tx_vid(ds, dp->index);
+	u16 tx_vid = dsa_8021q_tx_vid(dp->ds, dp->index);
 	u16 queue_mapping = skb_get_queue_mapping(skb);
 	u8 pcp = netdev_txq_to_tc(netdev, queue_mapping);
 
-- 
cgit v1.2.3-59-g8ed1b


From 6c930994503d9b5bd34b1329427dd7d3d6d37cd4 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Mon, 6 Jan 2020 03:34:09 +0200
Subject: mii: Add helpers for parsing SGMII auto-negotiation

Typically a MAC PCS auto-configures itself after it receives the
negotiated copper-side link settings from the PHY, but some MAC devices
are more special and need manual interpretation of the SGMII AN result.

In other cases, the PCS exposes the entire tx_config_reg base page as it
is transmitted on the wire during auto-negotiation, so it makes sense to
be able to decode the equivalent lp_advertised bit mask from the raw u16
(of course, "lp" considering the PCS to be the local PHY).

Therefore, add the bit definitions for the SGMII registers 4 and 5
(local device ability, link partner ability), as well as a link_mode
conversion helper that can be used to feed the AN results into
phy_resolve_aneg_linkmode.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/mii.h      | 50 ++++++++++++++++++++++++++++++++++++++++++++++++
 include/uapi/linux/mii.h | 12 ++++++++++++
 2 files changed, 62 insertions(+)

diff --git a/include/linux/mii.h b/include/linux/mii.h
index 4ce8901a1af6..18c6208f56fc 100644
--- a/include/linux/mii.h
+++ b/include/linux/mii.h
@@ -372,6 +372,56 @@ static inline u32 mii_lpa_to_ethtool_lpa_x(u32 lpa)
 	return result | mii_adv_to_ethtool_adv_x(lpa);
 }
 
+/**
+ * mii_lpa_mod_linkmode_adv_sgmii
+ * @lp_advertising: pointer to destination link mode.
+ * @lpa: value of the MII_LPA register
+ *
+ * A small helper function that translates MII_LPA bits to
+ * linkmode advertisement settings for SGMII.
+ * Leaves other bits unchanged.
+ */
+static inline void
+mii_lpa_mod_linkmode_lpa_sgmii(unsigned long *lp_advertising, u32 lpa)
+{
+	u32 speed_duplex = lpa & LPA_SGMII_DPX_SPD_MASK;
+
+	linkmode_mod_bit(ETHTOOL_LINK_MODE_1000baseT_Half_BIT, lp_advertising,
+			 speed_duplex == LPA_SGMII_1000HALF);
+
+	linkmode_mod_bit(ETHTOOL_LINK_MODE_1000baseT_Full_BIT, lp_advertising,
+			 speed_duplex == LPA_SGMII_1000FULL);
+
+	linkmode_mod_bit(ETHTOOL_LINK_MODE_100baseT_Half_BIT, lp_advertising,
+			 speed_duplex == LPA_SGMII_100HALF);
+
+	linkmode_mod_bit(ETHTOOL_LINK_MODE_100baseT_Full_BIT, lp_advertising,
+			 speed_duplex == LPA_SGMII_100FULL);
+
+	linkmode_mod_bit(ETHTOOL_LINK_MODE_10baseT_Half_BIT, lp_advertising,
+			 speed_duplex == LPA_SGMII_10HALF);
+
+	linkmode_mod_bit(ETHTOOL_LINK_MODE_10baseT_Full_BIT, lp_advertising,
+			 speed_duplex == LPA_SGMII_10FULL);
+}
+
+/**
+ * mii_lpa_to_linkmode_adv_sgmii
+ * @advertising: pointer to destination link mode.
+ * @lpa: value of the MII_LPA register
+ *
+ * A small helper function that translates MII_ADVERTISE bits
+ * to linkmode advertisement settings when in SGMII mode.
+ * Clears the old value of advertising.
+ */
+static inline void mii_lpa_to_linkmode_lpa_sgmii(unsigned long *lp_advertising,
+						 u32 lpa)
+{
+	linkmode_zero(lp_advertising);
+
+	mii_lpa_mod_linkmode_lpa_sgmii(lp_advertising, lpa);
+}
+
 /**
  * mii_adv_mod_linkmode_adv_t
  * @advertising:pointer to destination link mode.
diff --git a/include/uapi/linux/mii.h b/include/uapi/linux/mii.h
index 51b48e4be1f2..0b9c3beda345 100644
--- a/include/uapi/linux/mii.h
+++ b/include/uapi/linux/mii.h
@@ -131,6 +131,18 @@
 #define NWAYTEST_LOOPBACK	0x0100	/* Enable loopback for N-way   */
 #define NWAYTEST_RESV2		0xfe00	/* Unused...                   */
 
+/* MAC and PHY tx_config_Reg[15:0] for SGMII in-band auto-negotiation.*/
+#define ADVERTISE_SGMII		0x0001	/* MAC can do SGMII            */
+#define LPA_SGMII		0x0001	/* PHY can do SGMII            */
+#define LPA_SGMII_DPX_SPD_MASK	0x1C00	/* SGMII duplex and speed bits */
+#define LPA_SGMII_10HALF	0x0000	/* Can do 10mbps half-duplex   */
+#define LPA_SGMII_10FULL	0x1000	/* Can do 10mbps full-duplex   */
+#define LPA_SGMII_100HALF	0x0400	/* Can do 100mbps half-duplex  */
+#define LPA_SGMII_100FULL	0x1400	/* Can do 100mbps full-duplex  */
+#define LPA_SGMII_1000HALF	0x0800	/* Can do 1000mbps half-duplex */
+#define LPA_SGMII_1000FULL	0x1800	/* Can do 1000mbps full-duplex */
+#define LPA_SGMII_LINK		0x8000	/* PHY link with copper-side partner */
+
 /* 1000BASE-T Control register */
 #define ADVERTISE_1000FULL	0x0200  /* Advertise 1000BASE-T full duplex */
 #define ADVERTISE_1000HALF	0x0100  /* Advertise 1000BASE-T half duplex */
-- 
cgit v1.2.3-59-g8ed1b


From 3a68ba6fbab786a50f380894c832bf36ba7335f3 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Mon, 6 Jan 2020 03:34:10 +0200
Subject: net: phylink: make QSGMII a valid PHY mode for in-band AN

QSGMII is a SerDes protocol clocked at 5 Gbaud (4 times higher than
SGMII which is clocked at 1.25 Gbaud), with the same 8b/10b encoding and
some extra symbols for synchronization. Logically it offers 4 SGMII
interfaces multiplexed onto the same physical lanes. Each MAC PCS has
its own in-band AN process with the system side of the QSGMII PHY, which
is identical to the regular SGMII AN process.

So allow QSGMII as a valid in-band AN mode, since it is no different
from software perspective from regular SGMII.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/phylink.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index 1edca9725370..88686e0f9ae1 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -281,6 +281,7 @@ static int phylink_parse_mode(struct phylink *pl, struct fwnode_handle *fwnode)
 
 		switch (pl->link_config.interface) {
 		case PHY_INTERFACE_MODE_SGMII:
+		case PHY_INTERFACE_MODE_QSGMII:
 			phylink_set(pl->supported, 10baseT_Half);
 			phylink_set(pl->supported, 10baseT_Full);
 			phylink_set(pl->supported, 100baseT_Half);
-- 
cgit v1.2.3-59-g8ed1b


From 1511ed0a0167f523a84b4e727372a5d2ce1b6c2f Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Mon, 6 Jan 2020 03:34:11 +0200
Subject: net: phylink: add support for polling MAC PCS

Some MAC PCS blocks are unable to provide interrupts when their status
changes. As we already have support in phylink for polling status, use
this to provide a hook for MACs to enable polling mode.

The patch idea was picked up from Russell King's suggestion on the macb
phylink patch thread here [0] but the implementation was changed.
Instead of introducing a new phylink_start_poll() function, which would
make the implementation cumbersome for common PHYLINK implementations
for multiple types of devices, like DSA, just add a boolean property to
the phylink_config structure, which is just as backwards-compatible.

https://lkml.org/lkml/2019/12/16/603

Suggested-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/sfp-phylink.rst | 3 ++-
 drivers/net/phy/phylink.c                | 3 ++-
 include/linux/phylink.h                  | 2 ++
 3 files changed, 6 insertions(+), 2 deletions(-)

diff --git a/Documentation/networking/sfp-phylink.rst b/Documentation/networking/sfp-phylink.rst
index a5e00a159d21..d753a309f9d1 100644
--- a/Documentation/networking/sfp-phylink.rst
+++ b/Documentation/networking/sfp-phylink.rst
@@ -251,7 +251,8 @@ this documentation.
 	phylink_mac_change(priv->phylink, link_is_up);
 
     where ``link_is_up`` is true if the link is currently up or false
-    otherwise.
+    otherwise. If a MAC is unable to provide these interrupts, then
+    it should set ``priv->phylink_config.pcs_poll = true;`` in step 9.
 
 11. Verify that the driver does not call::
 
diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index 88686e0f9ae1..af914a8842bd 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -1022,7 +1022,8 @@ void phylink_start(struct phylink *pl)
 		if (irq <= 0)
 			mod_timer(&pl->link_poll, jiffies + HZ);
 	}
-	if (pl->cfg_link_an_mode == MLO_AN_FIXED && pl->get_fixed_state)
+	if ((pl->cfg_link_an_mode == MLO_AN_FIXED && pl->get_fixed_state) ||
+	    pl->config->pcs_poll)
 		mod_timer(&pl->link_poll, jiffies + HZ);
 	if (pl->phydev)
 		phy_start(pl->phydev);
diff --git a/include/linux/phylink.h b/include/linux/phylink.h
index fed5488e3c75..523209e70947 100644
--- a/include/linux/phylink.h
+++ b/include/linux/phylink.h
@@ -63,10 +63,12 @@ enum phylink_op_type {
  * struct phylink_config - PHYLINK configuration structure
  * @dev: a pointer to a struct device associated with the MAC
  * @type: operation type of PHYLINK instance
+ * @pcs_poll: MAC PCS cannot provide link change interrupt
  */
 struct phylink_config {
 	struct device *dev;
 	enum phylink_op_type type;
+	bool pcs_poll;
 };
 
 /**
-- 
cgit v1.2.3-59-g8ed1b


From 787cac3f5a650fd3184a41c5a27a2fe9ded833aa Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Mon, 6 Jan 2020 03:34:12 +0200
Subject: net: dsa: Pass pcs_poll flag from driver to PHYLINK

The DSA drivers that implement .phylink_mac_link_state should normally
register an interrupt for the PCS, from which they should call
phylink_mac_change(). However not all switches implement this, and those
who don't should set this flag in dsa_switch in the .setup callback, so
that PHYLINK will poll for a few ms until the in-band AN link timer
expires and the PCS state settles.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/dsa.h | 5 +++++
 net/dsa/port.c    | 1 +
 2 files changed, 6 insertions(+)

diff --git a/include/net/dsa.h b/include/net/dsa.h
index 23b1c58656d4..0c39fed8cd99 100644
--- a/include/net/dsa.h
+++ b/include/net/dsa.h
@@ -279,6 +279,11 @@ struct dsa_switch {
 	 */
 	bool			vlan_filtering;
 
+	/* MAC PCS does not provide link state change interrupt, and requires
+	 * polling. Flag passed on to PHYLINK.
+	 */
+	bool			pcs_poll;
+
 	size_t num_ports;
 };
 
diff --git a/net/dsa/port.c b/net/dsa/port.c
index ffb5601f7ed6..774facb8d547 100644
--- a/net/dsa/port.c
+++ b/net/dsa/port.c
@@ -599,6 +599,7 @@ static int dsa_port_phylink_register(struct dsa_port *dp)
 
 	dp->pl_config.dev = ds->dev;
 	dp->pl_config.type = PHYLINK_DEV;
+	dp->pl_config.pcs_poll = ds->pcs_poll;
 
 	dp->pl = phylink_create(&dp->pl_config, of_fwnode_handle(port_dn),
 				mode, &dsa_port_phylink_mac_ops);
-- 
cgit v1.2.3-59-g8ed1b


From 6517798dd3432a0002109809bf74e4fcf9bb0c7d Mon Sep 17 00:00:00 2001
From: Claudiu Manoil <claudiu.manoil@nxp.com>
Date: Mon, 6 Jan 2020 03:34:13 +0200
Subject: enetc: Make MDIO accessors more generic and export to
 include/linux/fsl

Within the LS1028A SoC, the register map for the ENETC MDIO controller
is instantiated a few times: for the central (external) MDIO controller,
for the internal bus of each standalone ENETC port, and for the internal
bus of the Felix switch.

Refactoring is needed to support multiple MDIO buses from multiple
drivers. The enetc_hw structure is made an opaque type and a smaller
enetc_mdio_priv is created.

'mdio_base' - MDIO registers base address - is being parameterized, to
be able to work with different MDIO register bases.

The ENETC MDIO bus operations are exported from the fsl-enetc-mdio
kernel object, the same that registers the central MDIO controller (the
dedicated PF). The ENETC main driver has been changed to select it, and
use its exported helpers to further register its private MDIO bus. The
DSA Felix driver will do the same.

Signed-off-by: Claudiu Manoil <claudiu.manoil@nxp.com>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/freescale/enetc/Kconfig       |   1 +
 drivers/net/ethernet/freescale/enetc/Makefile      |   2 +-
 drivers/net/ethernet/freescale/enetc/enetc_hw.h    |   1 +
 drivers/net/ethernet/freescale/enetc/enetc_mdio.c  | 110 +++++++++------------
 drivers/net/ethernet/freescale/enetc/enetc_mdio.h  |  12 ---
 .../net/ethernet/freescale/enetc/enetc_pci_mdio.c  |  43 ++++----
 drivers/net/ethernet/freescale/enetc/enetc_pf.c    |  47 +++++++++
 drivers/net/ethernet/freescale/enetc/enetc_pf.h    |   4 -
 include/linux/fsl/enetc_mdio.h                     |  55 +++++++++++
 9 files changed, 176 insertions(+), 99 deletions(-)
 delete mode 100644 drivers/net/ethernet/freescale/enetc/enetc_mdio.h
 create mode 100644 include/linux/fsl/enetc_mdio.h

diff --git a/drivers/net/ethernet/freescale/enetc/Kconfig b/drivers/net/ethernet/freescale/enetc/Kconfig
index edad4ca46327..fe942de19597 100644
--- a/drivers/net/ethernet/freescale/enetc/Kconfig
+++ b/drivers/net/ethernet/freescale/enetc/Kconfig
@@ -2,6 +2,7 @@
 config FSL_ENETC
 	tristate "ENETC PF driver"
 	depends on PCI && PCI_MSI && (ARCH_LAYERSCAPE || COMPILE_TEST)
+	select FSL_ENETC_MDIO
 	select PHYLIB
 	help
 	  This driver supports NXP ENETC gigabit ethernet controller PCIe
diff --git a/drivers/net/ethernet/freescale/enetc/Makefile b/drivers/net/ethernet/freescale/enetc/Makefile
index d0db33e5b6b7..74f7ac253b8b 100644
--- a/drivers/net/ethernet/freescale/enetc/Makefile
+++ b/drivers/net/ethernet/freescale/enetc/Makefile
@@ -3,7 +3,7 @@
 common-objs := enetc.o enetc_cbdr.o enetc_ethtool.o
 
 obj-$(CONFIG_FSL_ENETC) += fsl-enetc.o
-fsl-enetc-y := enetc_pf.o enetc_mdio.o $(common-objs)
+fsl-enetc-y := enetc_pf.o $(common-objs)
 fsl-enetc-$(CONFIG_PCI_IOV) += enetc_msg.o
 fsl-enetc-$(CONFIG_FSL_ENETC_QOS) += enetc_qos.o
 
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_hw.h b/drivers/net/ethernet/freescale/enetc/enetc_hw.h
index 8375cd886dba..62554f28ce07 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc_hw.h
+++ b/drivers/net/ethernet/freescale/enetc/enetc_hw.h
@@ -200,6 +200,7 @@ enum enetc_bdr_type {TX, RX};
 #define ENETC_PFPMR		0x1900
 #define ENETC_PFPMR_PMACE	BIT(1)
 #define ENETC_PFPMR_MWLM	BIT(0)
+#define ENETC_EMDIO_BASE	0x1c00
 #define ENETC_PSIUMHFR0(n, err)	(((err) ? 0x1d08 : 0x1d00) + (n) * 0x10)
 #define ENETC_PSIUMHFR1(n)	(0x1d04 + (n) * 0x10)
 #define ENETC_PSIMMHFR0(n, err)	(((err) ? 0x1d00 : 0x1d08) + (n) * 0x10)
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_mdio.c b/drivers/net/ethernet/freescale/enetc/enetc_mdio.c
index 149883c8f0b8..18c68e048d43 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc_mdio.c
+++ b/drivers/net/ethernet/freescale/enetc/enetc_mdio.c
@@ -1,24 +1,35 @@
 // SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
 /* Copyright 2019 NXP */
 
+#include <linux/fsl/enetc_mdio.h>
 #include <linux/mdio.h>
 #include <linux/of_mdio.h>
 #include <linux/iopoll.h>
 #include <linux/of.h>
 
-#include "enetc_mdio.h"
+#include "enetc_pf.h"
 
-#define	ENETC_MDIO_REG_OFFSET	0x1c00
 #define	ENETC_MDIO_CFG	0x0	/* MDIO configuration and status */
 #define	ENETC_MDIO_CTL	0x4	/* MDIO control */
 #define	ENETC_MDIO_DATA	0x8	/* MDIO data */
 #define	ENETC_MDIO_ADDR	0xc	/* MDIO address */
 
-#define enetc_mdio_rd(hw, off) \
-	enetc_port_rd(hw, ENETC_##off + ENETC_MDIO_REG_OFFSET)
-#define enetc_mdio_wr(hw, off, val) \
-	enetc_port_wr(hw, ENETC_##off + ENETC_MDIO_REG_OFFSET, val)
-#define enetc_mdio_rd_reg(off)	enetc_mdio_rd(hw, off)
+static inline u32 _enetc_mdio_rd(struct enetc_mdio_priv *mdio_priv, int off)
+{
+	return enetc_port_rd(mdio_priv->hw, mdio_priv->mdio_base + off);
+}
+
+static inline void _enetc_mdio_wr(struct enetc_mdio_priv *mdio_priv, int off,
+				  u32 val)
+{
+	enetc_port_wr(mdio_priv->hw, mdio_priv->mdio_base + off, val);
+}
+
+#define enetc_mdio_rd(mdio_priv, off) \
+	_enetc_mdio_rd(mdio_priv, ENETC_##off)
+#define enetc_mdio_wr(mdio_priv, off, val) \
+	_enetc_mdio_wr(mdio_priv, ENETC_##off, val)
+#define enetc_mdio_rd_reg(off)	enetc_mdio_rd(mdio_priv, off)
 
 #define ENETC_MDC_DIV		258
 
@@ -35,7 +46,7 @@
 #define MDIO_DATA(x)		((x) & 0xffff)
 
 #define TIMEOUT	1000
-static int enetc_mdio_wait_complete(struct enetc_hw *hw)
+static int enetc_mdio_wait_complete(struct enetc_mdio_priv *mdio_priv)
 {
 	u32 val;
 
@@ -46,7 +57,6 @@ static int enetc_mdio_wait_complete(struct enetc_hw *hw)
 int enetc_mdio_write(struct mii_bus *bus, int phy_id, int regnum, u16 value)
 {
 	struct enetc_mdio_priv *mdio_priv = bus->priv;
-	struct enetc_hw *hw = mdio_priv->hw;
 	u32 mdio_ctl, mdio_cfg;
 	u16 dev_addr;
 	int ret;
@@ -61,39 +71,39 @@ int enetc_mdio_write(struct mii_bus *bus, int phy_id, int regnum, u16 value)
 		mdio_cfg &= ~MDIO_CFG_ENC45;
 	}
 
-	enetc_mdio_wr(hw, MDIO_CFG, mdio_cfg);
+	enetc_mdio_wr(mdio_priv, MDIO_CFG, mdio_cfg);
 
-	ret = enetc_mdio_wait_complete(hw);
+	ret = enetc_mdio_wait_complete(mdio_priv);
 	if (ret)
 		return ret;
 
 	/* set port and dev addr */
 	mdio_ctl = MDIO_CTL_PORT_ADDR(phy_id) | MDIO_CTL_DEV_ADDR(dev_addr);
-	enetc_mdio_wr(hw, MDIO_CTL, mdio_ctl);
+	enetc_mdio_wr(mdio_priv, MDIO_CTL, mdio_ctl);
 
 	/* set the register address */
 	if (regnum & MII_ADDR_C45) {
-		enetc_mdio_wr(hw, MDIO_ADDR, regnum & 0xffff);
+		enetc_mdio_wr(mdio_priv, MDIO_ADDR, regnum & 0xffff);
 
-		ret = enetc_mdio_wait_complete(hw);
+		ret = enetc_mdio_wait_complete(mdio_priv);
 		if (ret)
 			return ret;
 	}
 
 	/* write the value */
-	enetc_mdio_wr(hw, MDIO_DATA, MDIO_DATA(value));
+	enetc_mdio_wr(mdio_priv, MDIO_DATA, MDIO_DATA(value));
 
-	ret = enetc_mdio_wait_complete(hw);
+	ret = enetc_mdio_wait_complete(mdio_priv);
 	if (ret)
 		return ret;
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(enetc_mdio_write);
 
 int enetc_mdio_read(struct mii_bus *bus, int phy_id, int regnum)
 {
 	struct enetc_mdio_priv *mdio_priv = bus->priv;
-	struct enetc_hw *hw = mdio_priv->hw;
 	u32 mdio_ctl, mdio_cfg;
 	u16 dev_addr, value;
 	int ret;
@@ -107,86 +117,56 @@ int enetc_mdio_read(struct mii_bus *bus, int phy_id, int regnum)
 		mdio_cfg &= ~MDIO_CFG_ENC45;
 	}
 
-	enetc_mdio_wr(hw, MDIO_CFG, mdio_cfg);
+	enetc_mdio_wr(mdio_priv, MDIO_CFG, mdio_cfg);
 
-	ret = enetc_mdio_wait_complete(hw);
+	ret = enetc_mdio_wait_complete(mdio_priv);
 	if (ret)
 		return ret;
 
 	/* set port and device addr */
 	mdio_ctl = MDIO_CTL_PORT_ADDR(phy_id) | MDIO_CTL_DEV_ADDR(dev_addr);
-	enetc_mdio_wr(hw, MDIO_CTL, mdio_ctl);
+	enetc_mdio_wr(mdio_priv, MDIO_CTL, mdio_ctl);
 
 	/* set the register address */
 	if (regnum & MII_ADDR_C45) {
-		enetc_mdio_wr(hw, MDIO_ADDR, regnum & 0xffff);
+		enetc_mdio_wr(mdio_priv, MDIO_ADDR, regnum & 0xffff);
 
-		ret = enetc_mdio_wait_complete(hw);
+		ret = enetc_mdio_wait_complete(mdio_priv);
 		if (ret)
 			return ret;
 	}
 
 	/* initiate the read */
-	enetc_mdio_wr(hw, MDIO_CTL, mdio_ctl | MDIO_CTL_READ);
+	enetc_mdio_wr(mdio_priv, MDIO_CTL, mdio_ctl | MDIO_CTL_READ);
 
-	ret = enetc_mdio_wait_complete(hw);
+	ret = enetc_mdio_wait_complete(mdio_priv);
 	if (ret)
 		return ret;
 
 	/* return all Fs if nothing was there */
-	if (enetc_mdio_rd(hw, MDIO_CFG) & MDIO_CFG_RD_ER) {
+	if (enetc_mdio_rd(mdio_priv, MDIO_CFG) & MDIO_CFG_RD_ER) {
 		dev_dbg(&bus->dev,
 			"Error while reading PHY%d reg at %d.%hhu\n",
 			phy_id, dev_addr, regnum);
 		return 0xffff;
 	}
 
-	value = enetc_mdio_rd(hw, MDIO_DATA) & 0xffff;
+	value = enetc_mdio_rd(mdio_priv, MDIO_DATA) & 0xffff;
 
 	return value;
 }
+EXPORT_SYMBOL_GPL(enetc_mdio_read);
 
-int enetc_mdio_probe(struct enetc_pf *pf)
+struct enetc_hw *enetc_hw_alloc(struct device *dev, void __iomem *port_regs)
 {
-	struct device *dev = &pf->si->pdev->dev;
-	struct enetc_mdio_priv *mdio_priv;
-	struct device_node *np;
-	struct mii_bus *bus;
-	int err;
-
-	bus = devm_mdiobus_alloc_size(dev, sizeof(*mdio_priv));
-	if (!bus)
-		return -ENOMEM;
-
-	bus->name = "Freescale ENETC MDIO Bus";
-	bus->read = enetc_mdio_read;
-	bus->write = enetc_mdio_write;
-	bus->parent = dev;
-	mdio_priv = bus->priv;
-	mdio_priv->hw = &pf->si->hw;
-	snprintf(bus->id, MII_BUS_ID_SIZE, "%s", dev_name(dev));
-
-	np = of_get_child_by_name(dev->of_node, "mdio");
-	if (!np) {
-		dev_err(dev, "MDIO node missing\n");
-		return -EINVAL;
-	}
-
-	err = of_mdiobus_register(bus, np);
-	if (err) {
-		of_node_put(np);
-		dev_err(dev, "cannot register MDIO bus\n");
-		return err;
-	}
+	struct enetc_hw *hw;
 
-	of_node_put(np);
-	pf->mdio = bus;
+	hw = devm_kzalloc(dev, sizeof(*hw), GFP_KERNEL);
+	if (!hw)
+		return ERR_PTR(-ENOMEM);
 
-	return 0;
-}
+	hw->port = port_regs;
 
-void enetc_mdio_remove(struct enetc_pf *pf)
-{
-	if (pf->mdio)
-		mdiobus_unregister(pf->mdio);
+	return hw;
 }
+EXPORT_SYMBOL_GPL(enetc_hw_alloc);
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_mdio.h b/drivers/net/ethernet/freescale/enetc/enetc_mdio.h
deleted file mode 100644
index 60c9a3889824..000000000000
--- a/drivers/net/ethernet/freescale/enetc/enetc_mdio.h
+++ /dev/null
@@ -1,12 +0,0 @@
-/* SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) */
-/* Copyright 2019 NXP */
-
-#include <linux/phy.h>
-#include "enetc_pf.h"
-
-struct enetc_mdio_priv {
-	struct enetc_hw *hw;
-};
-
-int enetc_mdio_write(struct mii_bus *bus, int phy_id, int regnum, u16 value);
-int enetc_mdio_read(struct mii_bus *bus, int phy_id, int regnum);
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_pci_mdio.c b/drivers/net/ethernet/freescale/enetc/enetc_pci_mdio.c
index fbd41ce01f06..87c0e969da40 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc_pci_mdio.c
+++ b/drivers/net/ethernet/freescale/enetc/enetc_pci_mdio.c
@@ -1,7 +1,8 @@
 // SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
 /* Copyright 2019 NXP */
+#include <linux/fsl/enetc_mdio.h>
 #include <linux/of_mdio.h>
-#include "enetc_mdio.h"
+#include "enetc_pf.h"
 
 #define ENETC_MDIO_DEV_ID	0xee01
 #define ENETC_MDIO_DEV_NAME	"FSL PCIe IE Central MDIO"
@@ -13,17 +14,29 @@ static int enetc_pci_mdio_probe(struct pci_dev *pdev,
 {
 	struct enetc_mdio_priv *mdio_priv;
 	struct device *dev = &pdev->dev;
+	void __iomem *port_regs;
 	struct enetc_hw *hw;
 	struct mii_bus *bus;
 	int err;
 
-	hw = devm_kzalloc(dev, sizeof(*hw), GFP_KERNEL);
-	if (!hw)
-		return -ENOMEM;
+	port_regs = pci_iomap(pdev, 0, 0);
+	if (!port_regs) {
+		dev_err(dev, "iomap failed\n");
+		err = -ENXIO;
+		goto err_ioremap;
+	}
+
+	hw = enetc_hw_alloc(dev, port_regs);
+	if (IS_ERR(enetc_hw_alloc)) {
+		err = PTR_ERR(hw);
+		goto err_hw_alloc;
+	}
 
 	bus = devm_mdiobus_alloc_size(dev, sizeof(*mdio_priv));
-	if (!bus)
-		return -ENOMEM;
+	if (!bus) {
+		err = -ENOMEM;
+		goto err_mdiobus_alloc;
+	}
 
 	bus->name = ENETC_MDIO_BUS_NAME;
 	bus->read = enetc_mdio_read;
@@ -31,13 +44,14 @@ static int enetc_pci_mdio_probe(struct pci_dev *pdev,
 	bus->parent = dev;
 	mdio_priv = bus->priv;
 	mdio_priv->hw = hw;
+	mdio_priv->mdio_base = ENETC_EMDIO_BASE;
 	snprintf(bus->id, MII_BUS_ID_SIZE, "%s", dev_name(dev));
 
 	pcie_flr(pdev);
 	err = pci_enable_device_mem(pdev);
 	if (err) {
 		dev_err(dev, "device enable failed\n");
-		return err;
+		goto err_pci_enable;
 	}
 
 	err = pci_request_region(pdev, 0, KBUILD_MODNAME);
@@ -46,13 +60,6 @@ static int enetc_pci_mdio_probe(struct pci_dev *pdev,
 		goto err_pci_mem_reg;
 	}
 
-	hw->port = pci_iomap(pdev, 0, 0);
-	if (!hw->port) {
-		err = -ENXIO;
-		dev_err(dev, "iomap failed\n");
-		goto err_ioremap;
-	}
-
 	err = of_mdiobus_register(bus, dev->of_node);
 	if (err)
 		goto err_mdiobus_reg;
@@ -62,12 +69,14 @@ static int enetc_pci_mdio_probe(struct pci_dev *pdev,
 	return 0;
 
 err_mdiobus_reg:
-	iounmap(mdio_priv->hw->port);
-err_ioremap:
 	pci_release_mem_regions(pdev);
 err_pci_mem_reg:
 	pci_disable_device(pdev);
-
+err_pci_enable:
+err_mdiobus_alloc:
+	iounmap(port_regs);
+err_hw_alloc:
+err_ioremap:
 	return err;
 }
 
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_pf.c b/drivers/net/ethernet/freescale/enetc/enetc_pf.c
index e7482d483b28..fc0d7d99e9a1 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc_pf.c
+++ b/drivers/net/ethernet/freescale/enetc/enetc_pf.c
@@ -2,6 +2,7 @@
 /* Copyright 2017-2019 NXP */
 
 #include <linux/module.h>
+#include <linux/fsl/enetc_mdio.h>
 #include <linux/of_mdio.h>
 #include <linux/of_net.h>
 #include "enetc_pf.h"
@@ -749,6 +750,52 @@ static void enetc_pf_netdev_setup(struct enetc_si *si, struct net_device *ndev,
 	enetc_get_primary_mac_addr(&si->hw, ndev->dev_addr);
 }
 
+static int enetc_mdio_probe(struct enetc_pf *pf)
+{
+	struct device *dev = &pf->si->pdev->dev;
+	struct enetc_mdio_priv *mdio_priv;
+	struct device_node *np;
+	struct mii_bus *bus;
+	int err;
+
+	bus = devm_mdiobus_alloc_size(dev, sizeof(*mdio_priv));
+	if (!bus)
+		return -ENOMEM;
+
+	bus->name = "Freescale ENETC MDIO Bus";
+	bus->read = enetc_mdio_read;
+	bus->write = enetc_mdio_write;
+	bus->parent = dev;
+	mdio_priv = bus->priv;
+	mdio_priv->hw = &pf->si->hw;
+	mdio_priv->mdio_base = ENETC_EMDIO_BASE;
+	snprintf(bus->id, MII_BUS_ID_SIZE, "%s", dev_name(dev));
+
+	np = of_get_child_by_name(dev->of_node, "mdio");
+	if (!np) {
+		dev_err(dev, "MDIO node missing\n");
+		return -EINVAL;
+	}
+
+	err = of_mdiobus_register(bus, np);
+	if (err) {
+		of_node_put(np);
+		dev_err(dev, "cannot register MDIO bus\n");
+		return err;
+	}
+
+	of_node_put(np);
+	pf->mdio = bus;
+
+	return 0;
+}
+
+static void enetc_mdio_remove(struct enetc_pf *pf)
+{
+	if (pf->mdio)
+		mdiobus_unregister(pf->mdio);
+}
+
 static int enetc_of_get_phy(struct enetc_ndev_priv *priv)
 {
 	struct enetc_pf *pf = enetc_si_priv(priv->si);
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_pf.h b/drivers/net/ethernet/freescale/enetc/enetc_pf.h
index 10dd1b53bb08..59e65a6f6c3e 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc_pf.h
+++ b/drivers/net/ethernet/freescale/enetc/enetc_pf.h
@@ -49,7 +49,3 @@ struct enetc_pf {
 int enetc_msg_psi_init(struct enetc_pf *pf);
 void enetc_msg_psi_free(struct enetc_pf *pf);
 void enetc_msg_handle_rxmsg(struct enetc_pf *pf, int mbox_id, u16 *status);
-
-/* MDIO */
-int enetc_mdio_probe(struct enetc_pf *pf);
-void enetc_mdio_remove(struct enetc_pf *pf);
diff --git a/include/linux/fsl/enetc_mdio.h b/include/linux/fsl/enetc_mdio.h
new file mode 100644
index 000000000000..4875dd38af7e
--- /dev/null
+++ b/include/linux/fsl/enetc_mdio.h
@@ -0,0 +1,55 @@
+/* SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) */
+/* Copyright 2019 NXP */
+
+#ifndef _FSL_ENETC_MDIO_H_
+#define _FSL_ENETC_MDIO_H_
+
+#include <linux/phy.h>
+
+/* PCS registers */
+#define ENETC_PCS_LINK_TIMER1			0x12
+#define ENETC_PCS_LINK_TIMER1_VAL		0x06a0
+#define ENETC_PCS_LINK_TIMER2			0x13
+#define ENETC_PCS_LINK_TIMER2_VAL		0x0003
+#define ENETC_PCS_IF_MODE			0x14
+#define ENETC_PCS_IF_MODE_SGMII_EN		BIT(0)
+#define ENETC_PCS_IF_MODE_USE_SGMII_AN		BIT(1)
+#define ENETC_PCS_IF_MODE_SGMII_SPEED(x)	(((x) << 2) & GENMASK(3, 2))
+
+/* Not a mistake, the SerDes PLL needs to be set at 3.125 GHz by Reset
+ * Configuration Word (RCW, outside Linux control) for 2.5G SGMII mode. The PCS
+ * still thinks it's at gigabit.
+ */
+enum enetc_pcs_speed {
+	ENETC_PCS_SPEED_10	= 0,
+	ENETC_PCS_SPEED_100	= 1,
+	ENETC_PCS_SPEED_1000	= 2,
+	ENETC_PCS_SPEED_2500	= 2,
+};
+
+struct enetc_hw;
+
+struct enetc_mdio_priv {
+	struct enetc_hw *hw;
+	int mdio_base;
+};
+
+#if IS_REACHABLE(CONFIG_FSL_ENETC_MDIO)
+
+int enetc_mdio_read(struct mii_bus *bus, int phy_id, int regnum);
+int enetc_mdio_write(struct mii_bus *bus, int phy_id, int regnum, u16 value);
+struct enetc_hw *enetc_hw_alloc(struct device *dev, void __iomem *port_regs);
+
+#else
+
+static inline int enetc_mdio_read(struct mii_bus *bus, int phy_id, int regnum)
+{ return -EINVAL; }
+static inline int enetc_mdio_write(struct mii_bus *bus, int phy_id, int regnum,
+				   u16 value)
+{ return -EINVAL; }
+struct enetc_hw *enetc_hw_alloc(struct device *dev, void __iomem *port_regs)
+{ return ERR_PTR(-EINVAL); }
+
+#endif
+
+#endif
-- 
cgit v1.2.3-59-g8ed1b


From d79d30327f7095e660801271f6740c0aeae0741b Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Mon, 6 Jan 2020 03:34:14 +0200
Subject: enetc: Set MDIO_CFG_HOLD to the recommended value of 2

This increases the MDIO hold time to 5 enet_clk cycles from the previous
value of 0. This is actually the out-of-reset value, that the driver was
previously overwriting with 0. Zero worked for the external MDIO, but
breaks communication with the internal MDIO buses on which the PCS of
ENETC SI's and Felix switch are found.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/freescale/enetc/enetc_mdio.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/freescale/enetc/enetc_mdio.c b/drivers/net/ethernet/freescale/enetc/enetc_mdio.c
index 18c68e048d43..48c32a171afa 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc_mdio.c
+++ b/drivers/net/ethernet/freescale/enetc/enetc_mdio.c
@@ -31,15 +31,19 @@ static inline void _enetc_mdio_wr(struct enetc_mdio_priv *mdio_priv, int off,
 	_enetc_mdio_wr(mdio_priv, ENETC_##off, val)
 #define enetc_mdio_rd_reg(off)	enetc_mdio_rd(mdio_priv, off)
 
-#define ENETC_MDC_DIV		258
-
 #define MDIO_CFG_CLKDIV(x)	((((x) >> 1) & 0xff) << 8)
 #define MDIO_CFG_BSY		BIT(0)
 #define MDIO_CFG_RD_ER		BIT(1)
+#define MDIO_CFG_HOLD(x)	(((x) << 2) & GENMASK(4, 2))
 #define MDIO_CFG_ENC45		BIT(6)
  /* external MDIO only - driven on neg MDC edge */
 #define MDIO_CFG_NEG		BIT(23)
 
+#define ENETC_EMDIO_CFG \
+	(MDIO_CFG_HOLD(2) | \
+	 MDIO_CFG_CLKDIV(258) | \
+	 MDIO_CFG_NEG)
+
 #define MDIO_CTL_DEV_ADDR(x)	((x) & 0x1f)
 #define MDIO_CTL_PORT_ADDR(x)	(((x) & 0x1f) << 5)
 #define MDIO_CTL_READ		BIT(15)
@@ -61,7 +65,7 @@ int enetc_mdio_write(struct mii_bus *bus, int phy_id, int regnum, u16 value)
 	u16 dev_addr;
 	int ret;
 
-	mdio_cfg = MDIO_CFG_CLKDIV(ENETC_MDC_DIV) | MDIO_CFG_NEG;
+	mdio_cfg = ENETC_EMDIO_CFG;
 	if (regnum & MII_ADDR_C45) {
 		dev_addr = (regnum >> 16) & 0x1f;
 		mdio_cfg |= MDIO_CFG_ENC45;
@@ -108,7 +112,7 @@ int enetc_mdio_read(struct mii_bus *bus, int phy_id, int regnum)
 	u16 dev_addr, value;
 	int ret;
 
-	mdio_cfg = MDIO_CFG_CLKDIV(ENETC_MDC_DIV) | MDIO_CFG_NEG;
+	mdio_cfg = ENETC_EMDIO_CFG;
 	if (regnum & MII_ADDR_C45) {
 		dev_addr = (regnum >> 16) & 0x1f;
 		mdio_cfg |= MDIO_CFG_ENC45;
-- 
cgit v1.2.3-59-g8ed1b


From ee50d07c9fc8155b5a3c6c29eae1459a12cf2fb4 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Mon, 6 Jan 2020 03:34:15 +0200
Subject: net: mscc: ocelot: make phy_mode a member of the common struct
 ocelot_port

The Ocelot switchdev driver and the Felix DSA one need it for different
reasons. Felix (or at least the VSC9959 instantiation in NXP LS1028A) is
integrated with the traditional NXP Layerscape PCS design which does not
support runtime configuration of SerDes protocol. So it needs to
pre-validate the phy-mode from the device tree and prevent PHYLINK from
attempting to change it. For this, it needs to cache it in a private
variable.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mscc/ocelot.c       | 7 ++++---
 drivers/net/ethernet/mscc/ocelot.h       | 1 -
 drivers/net/ethernet/mscc/ocelot_board.c | 4 ++--
 include/soc/mscc/ocelot.h                | 2 ++
 4 files changed, 8 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/mscc/ocelot.c b/drivers/net/ethernet/mscc/ocelot.c
index 985b46d7e3d1..86d543ab1ab9 100644
--- a/drivers/net/ethernet/mscc/ocelot.c
+++ b/drivers/net/ethernet/mscc/ocelot.c
@@ -500,13 +500,14 @@ EXPORT_SYMBOL(ocelot_port_enable);
 static int ocelot_port_open(struct net_device *dev)
 {
 	struct ocelot_port_private *priv = netdev_priv(dev);
-	struct ocelot *ocelot = priv->port.ocelot;
+	struct ocelot_port *ocelot_port = &priv->port;
+	struct ocelot *ocelot = ocelot_port->ocelot;
 	int port = priv->chip_port;
 	int err;
 
 	if (priv->serdes) {
 		err = phy_set_mode_ext(priv->serdes, PHY_MODE_ETHERNET,
-				       priv->phy_mode);
+				       ocelot_port->phy_mode);
 		if (err) {
 			netdev_err(dev, "Could not set mode of SerDes\n");
 			return err;
@@ -514,7 +515,7 @@ static int ocelot_port_open(struct net_device *dev)
 	}
 
 	err = phy_connect_direct(dev, priv->phy, &ocelot_port_adjust_link,
-				 priv->phy_mode);
+				 ocelot_port->phy_mode);
 	if (err) {
 		netdev_err(dev, "Could not attach to PHY\n");
 		return err;
diff --git a/drivers/net/ethernet/mscc/ocelot.h b/drivers/net/ethernet/mscc/ocelot.h
index c259114c48fd..7b77d44ed7cf 100644
--- a/drivers/net/ethernet/mscc/ocelot.h
+++ b/drivers/net/ethernet/mscc/ocelot.h
@@ -68,7 +68,6 @@ struct ocelot_port_private {
 
 	u8 vlan_aware;
 
-	phy_interface_t phy_mode;
 	struct phy *serdes;
 
 	struct ocelot_port_tc tc;
diff --git a/drivers/net/ethernet/mscc/ocelot_board.c b/drivers/net/ethernet/mscc/ocelot_board.c
index 2da8eee27e98..b38820849faa 100644
--- a/drivers/net/ethernet/mscc/ocelot_board.c
+++ b/drivers/net/ethernet/mscc/ocelot_board.c
@@ -402,9 +402,9 @@ static int mscc_ocelot_probe(struct platform_device *pdev)
 
 		of_get_phy_mode(portnp, &phy_mode);
 
-		priv->phy_mode = phy_mode;
+		ocelot_port->phy_mode = phy_mode;
 
-		switch (priv->phy_mode) {
+		switch (ocelot_port->phy_mode) {
 		case PHY_INTERFACE_MODE_NA:
 			continue;
 		case PHY_INTERFACE_MODE_SGMII:
diff --git a/include/soc/mscc/ocelot.h b/include/soc/mscc/ocelot.h
index 64cbbbe74a36..068f96b1a83e 100644
--- a/include/soc/mscc/ocelot.h
+++ b/include/soc/mscc/ocelot.h
@@ -420,6 +420,8 @@ struct ocelot_port {
 	u8				ptp_cmd;
 	struct sk_buff_head		tx_skbs;
 	u8				ts_id;
+
+	phy_interface_t			phy_mode;
 };
 
 struct ocelot {
-- 
cgit v1.2.3-59-g8ed1b


From 964ee5c82b770c2d8a5ccefeee3384c1061ce3ae Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Mon, 6 Jan 2020 03:34:16 +0200
Subject: net: mscc: ocelot: export ANA, DEV and QSYS registers to
 include/soc/mscc

Since the Felix DSA driver is implementing its own PHYLINK instance due
to SoC differences, it needs access to the few registers that are
common, mainly for flow control.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mscc/ocelot.h      |   6 +-
 drivers/net/ethernet/mscc/ocelot_ana.h  | 625 --------------------------------
 drivers/net/ethernet/mscc/ocelot_dev.h  | 275 --------------
 drivers/net/ethernet/mscc/ocelot_qsys.h | 270 --------------
 include/soc/mscc/ocelot_ana.h           | 625 ++++++++++++++++++++++++++++++++
 include/soc/mscc/ocelot_dev.h           | 275 ++++++++++++++
 include/soc/mscc/ocelot_qsys.h          | 270 ++++++++++++++
 7 files changed, 1173 insertions(+), 1173 deletions(-)
 delete mode 100644 drivers/net/ethernet/mscc/ocelot_ana.h
 delete mode 100644 drivers/net/ethernet/mscc/ocelot_dev.h
 delete mode 100644 drivers/net/ethernet/mscc/ocelot_qsys.h
 create mode 100644 include/soc/mscc/ocelot_ana.h
 create mode 100644 include/soc/mscc/ocelot_dev.h
 create mode 100644 include/soc/mscc/ocelot_qsys.h

diff --git a/drivers/net/ethernet/mscc/ocelot.h b/drivers/net/ethernet/mscc/ocelot.h
index 7b77d44ed7cf..04372ba72fec 100644
--- a/drivers/net/ethernet/mscc/ocelot.h
+++ b/drivers/net/ethernet/mscc/ocelot.h
@@ -18,11 +18,11 @@
 #include <linux/ptp_clock_kernel.h>
 #include <linux/regmap.h>
 
+#include <soc/mscc/ocelot_qsys.h>
 #include <soc/mscc/ocelot_sys.h>
+#include <soc/mscc/ocelot_dev.h>
+#include <soc/mscc/ocelot_ana.h>
 #include <soc/mscc/ocelot.h>
-#include "ocelot_ana.h"
-#include "ocelot_dev.h"
-#include "ocelot_qsys.h"
 #include "ocelot_rew.h"
 #include "ocelot_qs.h"
 #include "ocelot_tc.h"
diff --git a/drivers/net/ethernet/mscc/ocelot_ana.h b/drivers/net/ethernet/mscc/ocelot_ana.h
deleted file mode 100644
index 841c6ec22b64..000000000000
--- a/drivers/net/ethernet/mscc/ocelot_ana.h
+++ /dev/null
@@ -1,625 +0,0 @@
-/* SPDX-License-Identifier: (GPL-2.0 OR MIT) */
-/*
- * Microsemi Ocelot Switch driver
- *
- * Copyright (c) 2017 Microsemi Corporation
- */
-
-#ifndef _MSCC_OCELOT_ANA_H_
-#define _MSCC_OCELOT_ANA_H_
-
-#define ANA_ANAGEFIL_B_DOM_EN                             BIT(22)
-#define ANA_ANAGEFIL_B_DOM_VAL                            BIT(21)
-#define ANA_ANAGEFIL_AGE_LOCKED                           BIT(20)
-#define ANA_ANAGEFIL_PID_EN                               BIT(19)
-#define ANA_ANAGEFIL_PID_VAL(x)                           (((x) << 14) & GENMASK(18, 14))
-#define ANA_ANAGEFIL_PID_VAL_M                            GENMASK(18, 14)
-#define ANA_ANAGEFIL_PID_VAL_X(x)                         (((x) & GENMASK(18, 14)) >> 14)
-#define ANA_ANAGEFIL_VID_EN                               BIT(13)
-#define ANA_ANAGEFIL_VID_VAL(x)                           ((x) & GENMASK(12, 0))
-#define ANA_ANAGEFIL_VID_VAL_M                            GENMASK(12, 0)
-
-#define ANA_STORMLIMIT_CFG_RSZ                            0x4
-
-#define ANA_STORMLIMIT_CFG_STORM_RATE(x)                  (((x) << 3) & GENMASK(6, 3))
-#define ANA_STORMLIMIT_CFG_STORM_RATE_M                   GENMASK(6, 3)
-#define ANA_STORMLIMIT_CFG_STORM_RATE_X(x)                (((x) & GENMASK(6, 3)) >> 3)
-#define ANA_STORMLIMIT_CFG_STORM_UNIT                     BIT(2)
-#define ANA_STORMLIMIT_CFG_STORM_MODE(x)                  ((x) & GENMASK(1, 0))
-#define ANA_STORMLIMIT_CFG_STORM_MODE_M                   GENMASK(1, 0)
-
-#define ANA_AUTOAGE_AGE_FAST                              BIT(21)
-#define ANA_AUTOAGE_AGE_PERIOD(x)                         (((x) << 1) & GENMASK(20, 1))
-#define ANA_AUTOAGE_AGE_PERIOD_M                          GENMASK(20, 1)
-#define ANA_AUTOAGE_AGE_PERIOD_X(x)                       (((x) & GENMASK(20, 1)) >> 1)
-#define ANA_AUTOAGE_AUTOAGE_LOCKED                        BIT(0)
-
-#define ANA_MACTOPTIONS_REDUCED_TABLE                     BIT(1)
-#define ANA_MACTOPTIONS_SHADOW                            BIT(0)
-
-#define ANA_AGENCTRL_FID_MASK(x)                          (((x) << 12) & GENMASK(23, 12))
-#define ANA_AGENCTRL_FID_MASK_M                           GENMASK(23, 12)
-#define ANA_AGENCTRL_FID_MASK_X(x)                        (((x) & GENMASK(23, 12)) >> 12)
-#define ANA_AGENCTRL_IGNORE_DMAC_FLAGS                    BIT(11)
-#define ANA_AGENCTRL_IGNORE_SMAC_FLAGS                    BIT(10)
-#define ANA_AGENCTRL_FLOOD_SPECIAL                        BIT(9)
-#define ANA_AGENCTRL_FLOOD_IGNORE_VLAN                    BIT(8)
-#define ANA_AGENCTRL_MIRROR_CPU                           BIT(7)
-#define ANA_AGENCTRL_LEARN_CPU_COPY                       BIT(6)
-#define ANA_AGENCTRL_LEARN_FWD_KILL                       BIT(5)
-#define ANA_AGENCTRL_LEARN_IGNORE_VLAN                    BIT(4)
-#define ANA_AGENCTRL_CPU_CPU_KILL_ENA                     BIT(3)
-#define ANA_AGENCTRL_GREEN_COUNT_MODE                     BIT(2)
-#define ANA_AGENCTRL_YELLOW_COUNT_MODE                    BIT(1)
-#define ANA_AGENCTRL_RED_COUNT_MODE                       BIT(0)
-
-#define ANA_FLOODING_RSZ                                  0x4
-
-#define ANA_FLOODING_FLD_UNICAST(x)                       (((x) << 12) & GENMASK(17, 12))
-#define ANA_FLOODING_FLD_UNICAST_M                        GENMASK(17, 12)
-#define ANA_FLOODING_FLD_UNICAST_X(x)                     (((x) & GENMASK(17, 12)) >> 12)
-#define ANA_FLOODING_FLD_BROADCAST(x)                     (((x) << 6) & GENMASK(11, 6))
-#define ANA_FLOODING_FLD_BROADCAST_M                      GENMASK(11, 6)
-#define ANA_FLOODING_FLD_BROADCAST_X(x)                   (((x) & GENMASK(11, 6)) >> 6)
-#define ANA_FLOODING_FLD_MULTICAST(x)                     ((x) & GENMASK(5, 0))
-#define ANA_FLOODING_FLD_MULTICAST_M                      GENMASK(5, 0)
-
-#define ANA_FLOODING_IPMC_FLD_MC4_CTRL(x)                 (((x) << 18) & GENMASK(23, 18))
-#define ANA_FLOODING_IPMC_FLD_MC4_CTRL_M                  GENMASK(23, 18)
-#define ANA_FLOODING_IPMC_FLD_MC4_CTRL_X(x)               (((x) & GENMASK(23, 18)) >> 18)
-#define ANA_FLOODING_IPMC_FLD_MC4_DATA(x)                 (((x) << 12) & GENMASK(17, 12))
-#define ANA_FLOODING_IPMC_FLD_MC4_DATA_M                  GENMASK(17, 12)
-#define ANA_FLOODING_IPMC_FLD_MC4_DATA_X(x)               (((x) & GENMASK(17, 12)) >> 12)
-#define ANA_FLOODING_IPMC_FLD_MC6_CTRL(x)                 (((x) << 6) & GENMASK(11, 6))
-#define ANA_FLOODING_IPMC_FLD_MC6_CTRL_M                  GENMASK(11, 6)
-#define ANA_FLOODING_IPMC_FLD_MC6_CTRL_X(x)               (((x) & GENMASK(11, 6)) >> 6)
-#define ANA_FLOODING_IPMC_FLD_MC6_DATA(x)                 ((x) & GENMASK(5, 0))
-#define ANA_FLOODING_IPMC_FLD_MC6_DATA_M                  GENMASK(5, 0)
-
-#define ANA_SFLOW_CFG_RSZ                                 0x4
-
-#define ANA_SFLOW_CFG_SF_RATE(x)                          (((x) << 2) & GENMASK(13, 2))
-#define ANA_SFLOW_CFG_SF_RATE_M                           GENMASK(13, 2)
-#define ANA_SFLOW_CFG_SF_RATE_X(x)                        (((x) & GENMASK(13, 2)) >> 2)
-#define ANA_SFLOW_CFG_SF_SAMPLE_RX                        BIT(1)
-#define ANA_SFLOW_CFG_SF_SAMPLE_TX                        BIT(0)
-
-#define ANA_PORT_MODE_RSZ                                 0x4
-
-#define ANA_PORT_MODE_REDTAG_PARSE_CFG                    BIT(3)
-#define ANA_PORT_MODE_VLAN_PARSE_CFG(x)                   (((x) << 1) & GENMASK(2, 1))
-#define ANA_PORT_MODE_VLAN_PARSE_CFG_M                    GENMASK(2, 1)
-#define ANA_PORT_MODE_VLAN_PARSE_CFG_X(x)                 (((x) & GENMASK(2, 1)) >> 1)
-#define ANA_PORT_MODE_L3_PARSE_CFG                        BIT(0)
-
-#define ANA_CUT_THRU_CFG_RSZ                              0x4
-
-#define ANA_PGID_PGID_RSZ                                 0x4
-
-#define ANA_PGID_PGID_PGID(x)                             ((x) & GENMASK(11, 0))
-#define ANA_PGID_PGID_PGID_M                              GENMASK(11, 0)
-#define ANA_PGID_PGID_CPUQ_DST_PGID(x)                    (((x) << 27) & GENMASK(29, 27))
-#define ANA_PGID_PGID_CPUQ_DST_PGID_M                     GENMASK(29, 27)
-#define ANA_PGID_PGID_CPUQ_DST_PGID_X(x)                  (((x) & GENMASK(29, 27)) >> 27)
-
-#define ANA_TABLES_MACHDATA_VID(x)                        (((x) << 16) & GENMASK(28, 16))
-#define ANA_TABLES_MACHDATA_VID_M                         GENMASK(28, 16)
-#define ANA_TABLES_MACHDATA_VID_X(x)                      (((x) & GENMASK(28, 16)) >> 16)
-#define ANA_TABLES_MACHDATA_MACHDATA(x)                   ((x) & GENMASK(15, 0))
-#define ANA_TABLES_MACHDATA_MACHDATA_M                    GENMASK(15, 0)
-
-#define ANA_TABLES_STREAMDATA_SSID_VALID                  BIT(16)
-#define ANA_TABLES_STREAMDATA_SSID(x)                     (((x) << 9) & GENMASK(15, 9))
-#define ANA_TABLES_STREAMDATA_SSID_M                      GENMASK(15, 9)
-#define ANA_TABLES_STREAMDATA_SSID_X(x)                   (((x) & GENMASK(15, 9)) >> 9)
-#define ANA_TABLES_STREAMDATA_SFID_VALID                  BIT(8)
-#define ANA_TABLES_STREAMDATA_SFID(x)                     ((x) & GENMASK(7, 0))
-#define ANA_TABLES_STREAMDATA_SFID_M                      GENMASK(7, 0)
-
-#define ANA_TABLES_MACACCESS_MAC_CPU_COPY                 BIT(15)
-#define ANA_TABLES_MACACCESS_SRC_KILL                     BIT(14)
-#define ANA_TABLES_MACACCESS_IGNORE_VLAN                  BIT(13)
-#define ANA_TABLES_MACACCESS_AGED_FLAG                    BIT(12)
-#define ANA_TABLES_MACACCESS_VALID                        BIT(11)
-#define ANA_TABLES_MACACCESS_ENTRYTYPE(x)                 (((x) << 9) & GENMASK(10, 9))
-#define ANA_TABLES_MACACCESS_ENTRYTYPE_M                  GENMASK(10, 9)
-#define ANA_TABLES_MACACCESS_ENTRYTYPE_X(x)               (((x) & GENMASK(10, 9)) >> 9)
-#define ANA_TABLES_MACACCESS_DEST_IDX(x)                  (((x) << 3) & GENMASK(8, 3))
-#define ANA_TABLES_MACACCESS_DEST_IDX_M                   GENMASK(8, 3)
-#define ANA_TABLES_MACACCESS_DEST_IDX_X(x)                (((x) & GENMASK(8, 3)) >> 3)
-#define ANA_TABLES_MACACCESS_MAC_TABLE_CMD(x)             ((x) & GENMASK(2, 0))
-#define ANA_TABLES_MACACCESS_MAC_TABLE_CMD_M              GENMASK(2, 0)
-#define MACACCESS_CMD_IDLE                     0
-#define MACACCESS_CMD_LEARN                    1
-#define MACACCESS_CMD_FORGET                   2
-#define MACACCESS_CMD_AGE                      3
-#define MACACCESS_CMD_GET_NEXT                 4
-#define MACACCESS_CMD_INIT                     5
-#define MACACCESS_CMD_READ                     6
-#define MACACCESS_CMD_WRITE                    7
-
-#define ANA_TABLES_VLANACCESS_VLAN_PORT_MASK(x)           (((x) << 2) & GENMASK(13, 2))
-#define ANA_TABLES_VLANACCESS_VLAN_PORT_MASK_M            GENMASK(13, 2)
-#define ANA_TABLES_VLANACCESS_VLAN_PORT_MASK_X(x)         (((x) & GENMASK(13, 2)) >> 2)
-#define ANA_TABLES_VLANACCESS_VLAN_TBL_CMD(x)             ((x) & GENMASK(1, 0))
-#define ANA_TABLES_VLANACCESS_VLAN_TBL_CMD_M              GENMASK(1, 0)
-#define ANA_TABLES_VLANACCESS_CMD_IDLE                    0x0
-#define ANA_TABLES_VLANACCESS_CMD_WRITE                   0x2
-#define ANA_TABLES_VLANACCESS_CMD_INIT                    0x3
-
-#define ANA_TABLES_VLANTIDX_VLAN_SEC_FWD_ENA              BIT(17)
-#define ANA_TABLES_VLANTIDX_VLAN_FLOOD_DIS                BIT(16)
-#define ANA_TABLES_VLANTIDX_VLAN_PRIV_VLAN                BIT(15)
-#define ANA_TABLES_VLANTIDX_VLAN_LEARN_DISABLED           BIT(14)
-#define ANA_TABLES_VLANTIDX_VLAN_MIRROR                   BIT(13)
-#define ANA_TABLES_VLANTIDX_VLAN_SRC_CHK                  BIT(12)
-#define ANA_TABLES_VLANTIDX_V_INDEX(x)                    ((x) & GENMASK(11, 0))
-#define ANA_TABLES_VLANTIDX_V_INDEX_M                     GENMASK(11, 0)
-
-#define ANA_TABLES_ISDXACCESS_ISDX_PORT_MASK(x)           (((x) << 2) & GENMASK(8, 2))
-#define ANA_TABLES_ISDXACCESS_ISDX_PORT_MASK_M            GENMASK(8, 2)
-#define ANA_TABLES_ISDXACCESS_ISDX_PORT_MASK_X(x)         (((x) & GENMASK(8, 2)) >> 2)
-#define ANA_TABLES_ISDXACCESS_ISDX_TBL_CMD(x)             ((x) & GENMASK(1, 0))
-#define ANA_TABLES_ISDXACCESS_ISDX_TBL_CMD_M              GENMASK(1, 0)
-
-#define ANA_TABLES_ISDXTIDX_ISDX_SDLBI(x)                 (((x) << 21) & GENMASK(28, 21))
-#define ANA_TABLES_ISDXTIDX_ISDX_SDLBI_M                  GENMASK(28, 21)
-#define ANA_TABLES_ISDXTIDX_ISDX_SDLBI_X(x)               (((x) & GENMASK(28, 21)) >> 21)
-#define ANA_TABLES_ISDXTIDX_ISDX_MSTI(x)                  (((x) << 15) & GENMASK(20, 15))
-#define ANA_TABLES_ISDXTIDX_ISDX_MSTI_M                   GENMASK(20, 15)
-#define ANA_TABLES_ISDXTIDX_ISDX_MSTI_X(x)                (((x) & GENMASK(20, 15)) >> 15)
-#define ANA_TABLES_ISDXTIDX_ISDX_ES0_KEY_ENA              BIT(14)
-#define ANA_TABLES_ISDXTIDX_ISDX_FORCE_ENA                BIT(10)
-#define ANA_TABLES_ISDXTIDX_ISDX_INDEX(x)                 ((x) & GENMASK(7, 0))
-#define ANA_TABLES_ISDXTIDX_ISDX_INDEX_M                  GENMASK(7, 0)
-
-#define ANA_TABLES_ENTRYLIM_RSZ                           0x4
-
-#define ANA_TABLES_ENTRYLIM_ENTRYLIM(x)                   (((x) << 14) & GENMASK(17, 14))
-#define ANA_TABLES_ENTRYLIM_ENTRYLIM_M                    GENMASK(17, 14)
-#define ANA_TABLES_ENTRYLIM_ENTRYLIM_X(x)                 (((x) & GENMASK(17, 14)) >> 14)
-#define ANA_TABLES_ENTRYLIM_ENTRYSTAT(x)                  ((x) & GENMASK(13, 0))
-#define ANA_TABLES_ENTRYLIM_ENTRYSTAT_M                   GENMASK(13, 0)
-
-#define ANA_TABLES_STREAMACCESS_GEN_REC_SEQ_NUM(x)        (((x) << 4) & GENMASK(31, 4))
-#define ANA_TABLES_STREAMACCESS_GEN_REC_SEQ_NUM_M         GENMASK(31, 4)
-#define ANA_TABLES_STREAMACCESS_GEN_REC_SEQ_NUM_X(x)      (((x) & GENMASK(31, 4)) >> 4)
-#define ANA_TABLES_STREAMACCESS_SEQ_GEN_REC_ENA           BIT(3)
-#define ANA_TABLES_STREAMACCESS_GEN_REC_TYPE              BIT(2)
-#define ANA_TABLES_STREAMACCESS_STREAM_TBL_CMD(x)         ((x) & GENMASK(1, 0))
-#define ANA_TABLES_STREAMACCESS_STREAM_TBL_CMD_M          GENMASK(1, 0)
-
-#define ANA_TABLES_STREAMTIDX_SEQ_GEN_ERR_STATUS(x)       (((x) << 30) & GENMASK(31, 30))
-#define ANA_TABLES_STREAMTIDX_SEQ_GEN_ERR_STATUS_M        GENMASK(31, 30)
-#define ANA_TABLES_STREAMTIDX_SEQ_GEN_ERR_STATUS_X(x)     (((x) & GENMASK(31, 30)) >> 30)
-#define ANA_TABLES_STREAMTIDX_S_INDEX(x)                  (((x) << 16) & GENMASK(22, 16))
-#define ANA_TABLES_STREAMTIDX_S_INDEX_M                   GENMASK(22, 16)
-#define ANA_TABLES_STREAMTIDX_S_INDEX_X(x)                (((x) & GENMASK(22, 16)) >> 16)
-#define ANA_TABLES_STREAMTIDX_FORCE_SF_BEHAVIOUR          BIT(14)
-#define ANA_TABLES_STREAMTIDX_SEQ_HISTORY_LEN(x)          (((x) << 8) & GENMASK(13, 8))
-#define ANA_TABLES_STREAMTIDX_SEQ_HISTORY_LEN_M           GENMASK(13, 8)
-#define ANA_TABLES_STREAMTIDX_SEQ_HISTORY_LEN_X(x)        (((x) & GENMASK(13, 8)) >> 8)
-#define ANA_TABLES_STREAMTIDX_RESET_ON_ROGUE              BIT(7)
-#define ANA_TABLES_STREAMTIDX_REDTAG_POP                  BIT(6)
-#define ANA_TABLES_STREAMTIDX_STREAM_SPLIT                BIT(5)
-#define ANA_TABLES_STREAMTIDX_SEQ_SPACE_LOG2(x)           ((x) & GENMASK(4, 0))
-#define ANA_TABLES_STREAMTIDX_SEQ_SPACE_LOG2_M            GENMASK(4, 0)
-
-#define ANA_TABLES_SEQ_MASK_SPLIT_MASK(x)                 (((x) << 16) & GENMASK(22, 16))
-#define ANA_TABLES_SEQ_MASK_SPLIT_MASK_M                  GENMASK(22, 16)
-#define ANA_TABLES_SEQ_MASK_SPLIT_MASK_X(x)               (((x) & GENMASK(22, 16)) >> 16)
-#define ANA_TABLES_SEQ_MASK_INPUT_PORT_MASK(x)            ((x) & GENMASK(6, 0))
-#define ANA_TABLES_SEQ_MASK_INPUT_PORT_MASK_M             GENMASK(6, 0)
-
-#define ANA_TABLES_SFID_MASK_IGR_PORT_MASK(x)             (((x) << 1) & GENMASK(7, 1))
-#define ANA_TABLES_SFID_MASK_IGR_PORT_MASK_M              GENMASK(7, 1)
-#define ANA_TABLES_SFID_MASK_IGR_PORT_MASK_X(x)           (((x) & GENMASK(7, 1)) >> 1)
-#define ANA_TABLES_SFID_MASK_IGR_SRCPORT_MATCH_ENA        BIT(0)
-
-#define ANA_TABLES_SFIDACCESS_IGR_PRIO_MATCH_ENA          BIT(22)
-#define ANA_TABLES_SFIDACCESS_IGR_PRIO(x)                 (((x) << 19) & GENMASK(21, 19))
-#define ANA_TABLES_SFIDACCESS_IGR_PRIO_M                  GENMASK(21, 19)
-#define ANA_TABLES_SFIDACCESS_IGR_PRIO_X(x)               (((x) & GENMASK(21, 19)) >> 19)
-#define ANA_TABLES_SFIDACCESS_FORCE_BLOCK                 BIT(18)
-#define ANA_TABLES_SFIDACCESS_MAX_SDU_LEN(x)              (((x) << 2) & GENMASK(17, 2))
-#define ANA_TABLES_SFIDACCESS_MAX_SDU_LEN_M               GENMASK(17, 2)
-#define ANA_TABLES_SFIDACCESS_MAX_SDU_LEN_X(x)            (((x) & GENMASK(17, 2)) >> 2)
-#define ANA_TABLES_SFIDACCESS_SFID_TBL_CMD(x)             ((x) & GENMASK(1, 0))
-#define ANA_TABLES_SFIDACCESS_SFID_TBL_CMD_M              GENMASK(1, 0)
-
-#define ANA_TABLES_SFIDTIDX_SGID_VALID                    BIT(26)
-#define ANA_TABLES_SFIDTIDX_SGID(x)                       (((x) << 18) & GENMASK(25, 18))
-#define ANA_TABLES_SFIDTIDX_SGID_M                        GENMASK(25, 18)
-#define ANA_TABLES_SFIDTIDX_SGID_X(x)                     (((x) & GENMASK(25, 18)) >> 18)
-#define ANA_TABLES_SFIDTIDX_POL_ENA                       BIT(17)
-#define ANA_TABLES_SFIDTIDX_POL_IDX(x)                    (((x) << 8) & GENMASK(16, 8))
-#define ANA_TABLES_SFIDTIDX_POL_IDX_M                     GENMASK(16, 8)
-#define ANA_TABLES_SFIDTIDX_POL_IDX_X(x)                  (((x) & GENMASK(16, 8)) >> 8)
-#define ANA_TABLES_SFIDTIDX_SFID_INDEX(x)                 ((x) & GENMASK(7, 0))
-#define ANA_TABLES_SFIDTIDX_SFID_INDEX_M                  GENMASK(7, 0)
-
-#define ANA_MSTI_STATE_RSZ                                0x4
-
-#define ANA_OAM_UPM_LM_CNT_RSZ                            0x4
-
-#define ANA_SG_ACCESS_CTRL_SGID(x)                        ((x) & GENMASK(7, 0))
-#define ANA_SG_ACCESS_CTRL_SGID_M                         GENMASK(7, 0)
-#define ANA_SG_ACCESS_CTRL_CONFIG_CHANGE                  BIT(28)
-
-#define ANA_SG_CONFIG_REG_3_BASE_TIME_SEC_MSB(x)          ((x) & GENMASK(15, 0))
-#define ANA_SG_CONFIG_REG_3_BASE_TIME_SEC_MSB_M           GENMASK(15, 0)
-#define ANA_SG_CONFIG_REG_3_LIST_LENGTH(x)                (((x) << 16) & GENMASK(18, 16))
-#define ANA_SG_CONFIG_REG_3_LIST_LENGTH_M                 GENMASK(18, 16)
-#define ANA_SG_CONFIG_REG_3_LIST_LENGTH_X(x)              (((x) & GENMASK(18, 16)) >> 16)
-#define ANA_SG_CONFIG_REG_3_GATE_ENABLE                   BIT(20)
-#define ANA_SG_CONFIG_REG_3_INIT_IPS(x)                   (((x) << 24) & GENMASK(27, 24))
-#define ANA_SG_CONFIG_REG_3_INIT_IPS_M                    GENMASK(27, 24)
-#define ANA_SG_CONFIG_REG_3_INIT_IPS_X(x)                 (((x) & GENMASK(27, 24)) >> 24)
-#define ANA_SG_CONFIG_REG_3_INIT_GATE_STATE               BIT(28)
-
-#define ANA_SG_GCL_GS_CONFIG_RSZ                          0x4
-
-#define ANA_SG_GCL_GS_CONFIG_IPS(x)                       ((x) & GENMASK(3, 0))
-#define ANA_SG_GCL_GS_CONFIG_IPS_M                        GENMASK(3, 0)
-#define ANA_SG_GCL_GS_CONFIG_GATE_STATE                   BIT(4)
-
-#define ANA_SG_GCL_TI_CONFIG_RSZ                          0x4
-
-#define ANA_SG_STATUS_REG_3_CFG_CHG_TIME_SEC_MSB(x)       ((x) & GENMASK(15, 0))
-#define ANA_SG_STATUS_REG_3_CFG_CHG_TIME_SEC_MSB_M        GENMASK(15, 0)
-#define ANA_SG_STATUS_REG_3_GATE_STATE                    BIT(16)
-#define ANA_SG_STATUS_REG_3_IPS(x)                        (((x) << 20) & GENMASK(23, 20))
-#define ANA_SG_STATUS_REG_3_IPS_M                         GENMASK(23, 20)
-#define ANA_SG_STATUS_REG_3_IPS_X(x)                      (((x) & GENMASK(23, 20)) >> 20)
-#define ANA_SG_STATUS_REG_3_CONFIG_PENDING                BIT(24)
-
-#define ANA_PORT_VLAN_CFG_GSZ                             0x100
-
-#define ANA_PORT_VLAN_CFG_VLAN_VID_AS_ISDX                BIT(21)
-#define ANA_PORT_VLAN_CFG_VLAN_AWARE_ENA                  BIT(20)
-#define ANA_PORT_VLAN_CFG_VLAN_POP_CNT(x)                 (((x) << 18) & GENMASK(19, 18))
-#define ANA_PORT_VLAN_CFG_VLAN_POP_CNT_M                  GENMASK(19, 18)
-#define ANA_PORT_VLAN_CFG_VLAN_POP_CNT_X(x)               (((x) & GENMASK(19, 18)) >> 18)
-#define ANA_PORT_VLAN_CFG_VLAN_INNER_TAG_ENA              BIT(17)
-#define ANA_PORT_VLAN_CFG_VLAN_TAG_TYPE                   BIT(16)
-#define ANA_PORT_VLAN_CFG_VLAN_DEI                        BIT(15)
-#define ANA_PORT_VLAN_CFG_VLAN_PCP(x)                     (((x) << 12) & GENMASK(14, 12))
-#define ANA_PORT_VLAN_CFG_VLAN_PCP_M                      GENMASK(14, 12)
-#define ANA_PORT_VLAN_CFG_VLAN_PCP_X(x)                   (((x) & GENMASK(14, 12)) >> 12)
-#define ANA_PORT_VLAN_CFG_VLAN_VID(x)                     ((x) & GENMASK(11, 0))
-#define ANA_PORT_VLAN_CFG_VLAN_VID_M                      GENMASK(11, 0)
-
-#define ANA_PORT_DROP_CFG_GSZ                             0x100
-
-#define ANA_PORT_DROP_CFG_DROP_UNTAGGED_ENA               BIT(6)
-#define ANA_PORT_DROP_CFG_DROP_S_TAGGED_ENA               BIT(5)
-#define ANA_PORT_DROP_CFG_DROP_C_TAGGED_ENA               BIT(4)
-#define ANA_PORT_DROP_CFG_DROP_PRIO_S_TAGGED_ENA          BIT(3)
-#define ANA_PORT_DROP_CFG_DROP_PRIO_C_TAGGED_ENA          BIT(2)
-#define ANA_PORT_DROP_CFG_DROP_NULL_MAC_ENA               BIT(1)
-#define ANA_PORT_DROP_CFG_DROP_MC_SMAC_ENA                BIT(0)
-
-#define ANA_PORT_QOS_CFG_GSZ                              0x100
-
-#define ANA_PORT_QOS_CFG_DP_DEFAULT_VAL                   BIT(8)
-#define ANA_PORT_QOS_CFG_QOS_DEFAULT_VAL(x)               (((x) << 5) & GENMASK(7, 5))
-#define ANA_PORT_QOS_CFG_QOS_DEFAULT_VAL_M                GENMASK(7, 5)
-#define ANA_PORT_QOS_CFG_QOS_DEFAULT_VAL_X(x)             (((x) & GENMASK(7, 5)) >> 5)
-#define ANA_PORT_QOS_CFG_QOS_DSCP_ENA                     BIT(4)
-#define ANA_PORT_QOS_CFG_QOS_PCP_ENA                      BIT(3)
-#define ANA_PORT_QOS_CFG_DSCP_TRANSLATE_ENA               BIT(2)
-#define ANA_PORT_QOS_CFG_DSCP_REWR_CFG(x)                 ((x) & GENMASK(1, 0))
-#define ANA_PORT_QOS_CFG_DSCP_REWR_CFG_M                  GENMASK(1, 0)
-
-#define ANA_PORT_VCAP_CFG_GSZ                             0x100
-
-#define ANA_PORT_VCAP_CFG_S1_ENA                          BIT(14)
-#define ANA_PORT_VCAP_CFG_S1_DMAC_DIP_ENA(x)              (((x) << 11) & GENMASK(13, 11))
-#define ANA_PORT_VCAP_CFG_S1_DMAC_DIP_ENA_M               GENMASK(13, 11)
-#define ANA_PORT_VCAP_CFG_S1_DMAC_DIP_ENA_X(x)            (((x) & GENMASK(13, 11)) >> 11)
-#define ANA_PORT_VCAP_CFG_S1_VLAN_INNER_TAG_ENA(x)        (((x) << 8) & GENMASK(10, 8))
-#define ANA_PORT_VCAP_CFG_S1_VLAN_INNER_TAG_ENA_M         GENMASK(10, 8)
-#define ANA_PORT_VCAP_CFG_S1_VLAN_INNER_TAG_ENA_X(x)      (((x) & GENMASK(10, 8)) >> 8)
-#define ANA_PORT_VCAP_CFG_PAG_VAL(x)                      ((x) & GENMASK(7, 0))
-#define ANA_PORT_VCAP_CFG_PAG_VAL_M                       GENMASK(7, 0)
-
-#define ANA_PORT_VCAP_S1_KEY_CFG_GSZ                      0x100
-#define ANA_PORT_VCAP_S1_KEY_CFG_RSZ                      0x4
-
-#define ANA_PORT_VCAP_S1_KEY_CFG_S1_KEY_IP6_CFG(x)        (((x) << 4) & GENMASK(6, 4))
-#define ANA_PORT_VCAP_S1_KEY_CFG_S1_KEY_IP6_CFG_M         GENMASK(6, 4)
-#define ANA_PORT_VCAP_S1_KEY_CFG_S1_KEY_IP6_CFG_X(x)      (((x) & GENMASK(6, 4)) >> 4)
-#define ANA_PORT_VCAP_S1_KEY_CFG_S1_KEY_IP4_CFG(x)        (((x) << 2) & GENMASK(3, 2))
-#define ANA_PORT_VCAP_S1_KEY_CFG_S1_KEY_IP4_CFG_M         GENMASK(3, 2)
-#define ANA_PORT_VCAP_S1_KEY_CFG_S1_KEY_IP4_CFG_X(x)      (((x) & GENMASK(3, 2)) >> 2)
-#define ANA_PORT_VCAP_S1_KEY_CFG_S1_KEY_OTHER_CFG(x)      ((x) & GENMASK(1, 0))
-#define ANA_PORT_VCAP_S1_KEY_CFG_S1_KEY_OTHER_CFG_M       GENMASK(1, 0)
-
-#define ANA_PORT_VCAP_S2_CFG_GSZ                          0x100
-
-#define ANA_PORT_VCAP_S2_CFG_S2_UDP_PAYLOAD_ENA(x)        (((x) << 17) & GENMASK(18, 17))
-#define ANA_PORT_VCAP_S2_CFG_S2_UDP_PAYLOAD_ENA_M         GENMASK(18, 17)
-#define ANA_PORT_VCAP_S2_CFG_S2_UDP_PAYLOAD_ENA_X(x)      (((x) & GENMASK(18, 17)) >> 17)
-#define ANA_PORT_VCAP_S2_CFG_S2_ETYPE_PAYLOAD_ENA(x)      (((x) << 15) & GENMASK(16, 15))
-#define ANA_PORT_VCAP_S2_CFG_S2_ETYPE_PAYLOAD_ENA_M       GENMASK(16, 15)
-#define ANA_PORT_VCAP_S2_CFG_S2_ETYPE_PAYLOAD_ENA_X(x)    (((x) & GENMASK(16, 15)) >> 15)
-#define ANA_PORT_VCAP_S2_CFG_S2_ENA                       BIT(14)
-#define ANA_PORT_VCAP_S2_CFG_S2_SNAP_DIS(x)               (((x) << 12) & GENMASK(13, 12))
-#define ANA_PORT_VCAP_S2_CFG_S2_SNAP_DIS_M                GENMASK(13, 12)
-#define ANA_PORT_VCAP_S2_CFG_S2_SNAP_DIS_X(x)             (((x) & GENMASK(13, 12)) >> 12)
-#define ANA_PORT_VCAP_S2_CFG_S2_ARP_DIS(x)                (((x) << 10) & GENMASK(11, 10))
-#define ANA_PORT_VCAP_S2_CFG_S2_ARP_DIS_M                 GENMASK(11, 10)
-#define ANA_PORT_VCAP_S2_CFG_S2_ARP_DIS_X(x)              (((x) & GENMASK(11, 10)) >> 10)
-#define ANA_PORT_VCAP_S2_CFG_S2_IP_TCPUDP_DIS(x)          (((x) << 8) & GENMASK(9, 8))
-#define ANA_PORT_VCAP_S2_CFG_S2_IP_TCPUDP_DIS_M           GENMASK(9, 8)
-#define ANA_PORT_VCAP_S2_CFG_S2_IP_TCPUDP_DIS_X(x)        (((x) & GENMASK(9, 8)) >> 8)
-#define ANA_PORT_VCAP_S2_CFG_S2_IP_OTHER_DIS(x)           (((x) << 6) & GENMASK(7, 6))
-#define ANA_PORT_VCAP_S2_CFG_S2_IP_OTHER_DIS_M            GENMASK(7, 6)
-#define ANA_PORT_VCAP_S2_CFG_S2_IP_OTHER_DIS_X(x)         (((x) & GENMASK(7, 6)) >> 6)
-#define ANA_PORT_VCAP_S2_CFG_S2_IP6_CFG(x)                (((x) << 2) & GENMASK(5, 2))
-#define ANA_PORT_VCAP_S2_CFG_S2_IP6_CFG_M                 GENMASK(5, 2)
-#define ANA_PORT_VCAP_S2_CFG_S2_IP6_CFG_X(x)              (((x) & GENMASK(5, 2)) >> 2)
-#define ANA_PORT_VCAP_S2_CFG_S2_OAM_DIS(x)                ((x) & GENMASK(1, 0))
-#define ANA_PORT_VCAP_S2_CFG_S2_OAM_DIS_M                 GENMASK(1, 0)
-
-#define ANA_PORT_PCP_DEI_MAP_GSZ                          0x100
-#define ANA_PORT_PCP_DEI_MAP_RSZ                          0x4
-
-#define ANA_PORT_PCP_DEI_MAP_DP_PCP_DEI_VAL               BIT(3)
-#define ANA_PORT_PCP_DEI_MAP_QOS_PCP_DEI_VAL(x)           ((x) & GENMASK(2, 0))
-#define ANA_PORT_PCP_DEI_MAP_QOS_PCP_DEI_VAL_M            GENMASK(2, 0)
-
-#define ANA_PORT_CPU_FWD_CFG_GSZ                          0x100
-
-#define ANA_PORT_CPU_FWD_CFG_CPU_VRAP_REDIR_ENA           BIT(7)
-#define ANA_PORT_CPU_FWD_CFG_CPU_MLD_REDIR_ENA            BIT(6)
-#define ANA_PORT_CPU_FWD_CFG_CPU_IGMP_REDIR_ENA           BIT(5)
-#define ANA_PORT_CPU_FWD_CFG_CPU_IPMC_CTRL_COPY_ENA       BIT(4)
-#define ANA_PORT_CPU_FWD_CFG_CPU_SRC_COPY_ENA             BIT(3)
-#define ANA_PORT_CPU_FWD_CFG_CPU_ALLBRIDGE_DROP_ENA       BIT(2)
-#define ANA_PORT_CPU_FWD_CFG_CPU_ALLBRIDGE_REDIR_ENA      BIT(1)
-#define ANA_PORT_CPU_FWD_CFG_CPU_OAM_ENA                  BIT(0)
-
-#define ANA_PORT_CPU_FWD_BPDU_CFG_GSZ                     0x100
-
-#define ANA_PORT_CPU_FWD_BPDU_CFG_BPDU_DROP_ENA(x)        (((x) << 16) & GENMASK(31, 16))
-#define ANA_PORT_CPU_FWD_BPDU_CFG_BPDU_DROP_ENA_M         GENMASK(31, 16)
-#define ANA_PORT_CPU_FWD_BPDU_CFG_BPDU_DROP_ENA_X(x)      (((x) & GENMASK(31, 16)) >> 16)
-#define ANA_PORT_CPU_FWD_BPDU_CFG_BPDU_REDIR_ENA(x)       ((x) & GENMASK(15, 0))
-#define ANA_PORT_CPU_FWD_BPDU_CFG_BPDU_REDIR_ENA_M        GENMASK(15, 0)
-
-#define ANA_PORT_CPU_FWD_GARP_CFG_GSZ                     0x100
-
-#define ANA_PORT_CPU_FWD_GARP_CFG_GARP_DROP_ENA(x)        (((x) << 16) & GENMASK(31, 16))
-#define ANA_PORT_CPU_FWD_GARP_CFG_GARP_DROP_ENA_M         GENMASK(31, 16)
-#define ANA_PORT_CPU_FWD_GARP_CFG_GARP_DROP_ENA_X(x)      (((x) & GENMASK(31, 16)) >> 16)
-#define ANA_PORT_CPU_FWD_GARP_CFG_GARP_REDIR_ENA(x)       ((x) & GENMASK(15, 0))
-#define ANA_PORT_CPU_FWD_GARP_CFG_GARP_REDIR_ENA_M        GENMASK(15, 0)
-
-#define ANA_PORT_CPU_FWD_CCM_CFG_GSZ                      0x100
-
-#define ANA_PORT_CPU_FWD_CCM_CFG_CCM_DROP_ENA(x)          (((x) << 16) & GENMASK(31, 16))
-#define ANA_PORT_CPU_FWD_CCM_CFG_CCM_DROP_ENA_M           GENMASK(31, 16)
-#define ANA_PORT_CPU_FWD_CCM_CFG_CCM_DROP_ENA_X(x)        (((x) & GENMASK(31, 16)) >> 16)
-#define ANA_PORT_CPU_FWD_CCM_CFG_CCM_REDIR_ENA(x)         ((x) & GENMASK(15, 0))
-#define ANA_PORT_CPU_FWD_CCM_CFG_CCM_REDIR_ENA_M          GENMASK(15, 0)
-
-#define ANA_PORT_PORT_CFG_GSZ                             0x100
-
-#define ANA_PORT_PORT_CFG_SRC_MIRROR_ENA                  BIT(15)
-#define ANA_PORT_PORT_CFG_LIMIT_DROP                      BIT(14)
-#define ANA_PORT_PORT_CFG_LIMIT_CPU                       BIT(13)
-#define ANA_PORT_PORT_CFG_LOCKED_PORTMOVE_DROP            BIT(12)
-#define ANA_PORT_PORT_CFG_LOCKED_PORTMOVE_CPU             BIT(11)
-#define ANA_PORT_PORT_CFG_LEARNDROP                       BIT(10)
-#define ANA_PORT_PORT_CFG_LEARNCPU                        BIT(9)
-#define ANA_PORT_PORT_CFG_LEARNAUTO                       BIT(8)
-#define ANA_PORT_PORT_CFG_LEARN_ENA                       BIT(7)
-#define ANA_PORT_PORT_CFG_RECV_ENA                        BIT(6)
-#define ANA_PORT_PORT_CFG_PORTID_VAL(x)                   (((x) << 2) & GENMASK(5, 2))
-#define ANA_PORT_PORT_CFG_PORTID_VAL_M                    GENMASK(5, 2)
-#define ANA_PORT_PORT_CFG_PORTID_VAL_X(x)                 (((x) & GENMASK(5, 2)) >> 2)
-#define ANA_PORT_PORT_CFG_USE_B_DOM_TBL                   BIT(1)
-#define ANA_PORT_PORT_CFG_LSR_MODE                        BIT(0)
-
-#define ANA_PORT_POL_CFG_GSZ                              0x100
-
-#define ANA_PORT_POL_CFG_POL_CPU_REDIR_8021               BIT(19)
-#define ANA_PORT_POL_CFG_POL_CPU_REDIR_IP                 BIT(18)
-#define ANA_PORT_POL_CFG_PORT_POL_ENA                     BIT(17)
-#define ANA_PORT_POL_CFG_QUEUE_POL_ENA(x)                 (((x) << 9) & GENMASK(16, 9))
-#define ANA_PORT_POL_CFG_QUEUE_POL_ENA_M                  GENMASK(16, 9)
-#define ANA_PORT_POL_CFG_QUEUE_POL_ENA_X(x)               (((x) & GENMASK(16, 9)) >> 9)
-#define ANA_PORT_POL_CFG_POL_ORDER(x)                     ((x) & GENMASK(8, 0))
-#define ANA_PORT_POL_CFG_POL_ORDER_M                      GENMASK(8, 0)
-
-#define ANA_PORT_PTP_CFG_GSZ                              0x100
-
-#define ANA_PORT_PTP_CFG_PTP_BACKPLANE_MODE               BIT(0)
-
-#define ANA_PORT_PTP_DLY1_CFG_GSZ                         0x100
-
-#define ANA_PORT_PTP_DLY2_CFG_GSZ                         0x100
-
-#define ANA_PORT_SFID_CFG_GSZ                             0x100
-#define ANA_PORT_SFID_CFG_RSZ                             0x4
-
-#define ANA_PORT_SFID_CFG_SFID_VALID                      BIT(8)
-#define ANA_PORT_SFID_CFG_SFID(x)                         ((x) & GENMASK(7, 0))
-#define ANA_PORT_SFID_CFG_SFID_M                          GENMASK(7, 0)
-
-#define ANA_PFC_PFC_CFG_GSZ                               0x40
-
-#define ANA_PFC_PFC_CFG_RX_PFC_ENA(x)                     (((x) << 2) & GENMASK(9, 2))
-#define ANA_PFC_PFC_CFG_RX_PFC_ENA_M                      GENMASK(9, 2)
-#define ANA_PFC_PFC_CFG_RX_PFC_ENA_X(x)                   (((x) & GENMASK(9, 2)) >> 2)
-#define ANA_PFC_PFC_CFG_FC_LINK_SPEED(x)                  ((x) & GENMASK(1, 0))
-#define ANA_PFC_PFC_CFG_FC_LINK_SPEED_M                   GENMASK(1, 0)
-
-#define ANA_PFC_PFC_TIMER_GSZ                             0x40
-#define ANA_PFC_PFC_TIMER_RSZ                             0x4
-
-#define ANA_IPT_OAM_MEP_CFG_GSZ                           0x8
-
-#define ANA_IPT_OAM_MEP_CFG_MEP_IDX_P(x)                  (((x) << 6) & GENMASK(10, 6))
-#define ANA_IPT_OAM_MEP_CFG_MEP_IDX_P_M                   GENMASK(10, 6)
-#define ANA_IPT_OAM_MEP_CFG_MEP_IDX_P_X(x)                (((x) & GENMASK(10, 6)) >> 6)
-#define ANA_IPT_OAM_MEP_CFG_MEP_IDX(x)                    (((x) << 1) & GENMASK(5, 1))
-#define ANA_IPT_OAM_MEP_CFG_MEP_IDX_M                     GENMASK(5, 1)
-#define ANA_IPT_OAM_MEP_CFG_MEP_IDX_X(x)                  (((x) & GENMASK(5, 1)) >> 1)
-#define ANA_IPT_OAM_MEP_CFG_MEP_IDX_ENA                   BIT(0)
-
-#define ANA_IPT_IPT_GSZ                                   0x8
-
-#define ANA_IPT_IPT_IPT_CFG(x)                            (((x) << 15) & GENMASK(16, 15))
-#define ANA_IPT_IPT_IPT_CFG_M                             GENMASK(16, 15)
-#define ANA_IPT_IPT_IPT_CFG_X(x)                          (((x) & GENMASK(16, 15)) >> 15)
-#define ANA_IPT_IPT_ISDX_P(x)                             (((x) << 7) & GENMASK(14, 7))
-#define ANA_IPT_IPT_ISDX_P_M                              GENMASK(14, 7)
-#define ANA_IPT_IPT_ISDX_P_X(x)                           (((x) & GENMASK(14, 7)) >> 7)
-#define ANA_IPT_IPT_PPT_IDX(x)                            ((x) & GENMASK(6, 0))
-#define ANA_IPT_IPT_PPT_IDX_M                             GENMASK(6, 0)
-
-#define ANA_PPT_PPT_RSZ                                   0x4
-
-#define ANA_FID_MAP_FID_MAP_RSZ                           0x4
-
-#define ANA_FID_MAP_FID_MAP_FID_C_VAL(x)                  (((x) << 6) & GENMASK(11, 6))
-#define ANA_FID_MAP_FID_MAP_FID_C_VAL_M                   GENMASK(11, 6)
-#define ANA_FID_MAP_FID_MAP_FID_C_VAL_X(x)                (((x) & GENMASK(11, 6)) >> 6)
-#define ANA_FID_MAP_FID_MAP_FID_B_VAL(x)                  ((x) & GENMASK(5, 0))
-#define ANA_FID_MAP_FID_MAP_FID_B_VAL_M                   GENMASK(5, 0)
-
-#define ANA_AGGR_CFG_AC_RND_ENA                           BIT(7)
-#define ANA_AGGR_CFG_AC_DMAC_ENA                          BIT(6)
-#define ANA_AGGR_CFG_AC_SMAC_ENA                          BIT(5)
-#define ANA_AGGR_CFG_AC_IP6_FLOW_LBL_ENA                  BIT(4)
-#define ANA_AGGR_CFG_AC_IP6_TCPUDP_ENA                    BIT(3)
-#define ANA_AGGR_CFG_AC_IP4_SIPDIP_ENA                    BIT(2)
-#define ANA_AGGR_CFG_AC_IP4_TCPUDP_ENA                    BIT(1)
-#define ANA_AGGR_CFG_AC_ISDX_ENA                          BIT(0)
-
-#define ANA_CPUQ_CFG_CPUQ_MLD(x)                          (((x) << 27) & GENMASK(29, 27))
-#define ANA_CPUQ_CFG_CPUQ_MLD_M                           GENMASK(29, 27)
-#define ANA_CPUQ_CFG_CPUQ_MLD_X(x)                        (((x) & GENMASK(29, 27)) >> 27)
-#define ANA_CPUQ_CFG_CPUQ_IGMP(x)                         (((x) << 24) & GENMASK(26, 24))
-#define ANA_CPUQ_CFG_CPUQ_IGMP_M                          GENMASK(26, 24)
-#define ANA_CPUQ_CFG_CPUQ_IGMP_X(x)                       (((x) & GENMASK(26, 24)) >> 24)
-#define ANA_CPUQ_CFG_CPUQ_IPMC_CTRL(x)                    (((x) << 21) & GENMASK(23, 21))
-#define ANA_CPUQ_CFG_CPUQ_IPMC_CTRL_M                     GENMASK(23, 21)
-#define ANA_CPUQ_CFG_CPUQ_IPMC_CTRL_X(x)                  (((x) & GENMASK(23, 21)) >> 21)
-#define ANA_CPUQ_CFG_CPUQ_ALLBRIDGE(x)                    (((x) << 18) & GENMASK(20, 18))
-#define ANA_CPUQ_CFG_CPUQ_ALLBRIDGE_M                     GENMASK(20, 18)
-#define ANA_CPUQ_CFG_CPUQ_ALLBRIDGE_X(x)                  (((x) & GENMASK(20, 18)) >> 18)
-#define ANA_CPUQ_CFG_CPUQ_LOCKED_PORTMOVE(x)              (((x) << 15) & GENMASK(17, 15))
-#define ANA_CPUQ_CFG_CPUQ_LOCKED_PORTMOVE_M               GENMASK(17, 15)
-#define ANA_CPUQ_CFG_CPUQ_LOCKED_PORTMOVE_X(x)            (((x) & GENMASK(17, 15)) >> 15)
-#define ANA_CPUQ_CFG_CPUQ_SRC_COPY(x)                     (((x) << 12) & GENMASK(14, 12))
-#define ANA_CPUQ_CFG_CPUQ_SRC_COPY_M                      GENMASK(14, 12)
-#define ANA_CPUQ_CFG_CPUQ_SRC_COPY_X(x)                   (((x) & GENMASK(14, 12)) >> 12)
-#define ANA_CPUQ_CFG_CPUQ_MAC_COPY(x)                     (((x) << 9) & GENMASK(11, 9))
-#define ANA_CPUQ_CFG_CPUQ_MAC_COPY_M                      GENMASK(11, 9)
-#define ANA_CPUQ_CFG_CPUQ_MAC_COPY_X(x)                   (((x) & GENMASK(11, 9)) >> 9)
-#define ANA_CPUQ_CFG_CPUQ_LRN(x)                          (((x) << 6) & GENMASK(8, 6))
-#define ANA_CPUQ_CFG_CPUQ_LRN_M                           GENMASK(8, 6)
-#define ANA_CPUQ_CFG_CPUQ_LRN_X(x)                        (((x) & GENMASK(8, 6)) >> 6)
-#define ANA_CPUQ_CFG_CPUQ_MIRROR(x)                       (((x) << 3) & GENMASK(5, 3))
-#define ANA_CPUQ_CFG_CPUQ_MIRROR_M                        GENMASK(5, 3)
-#define ANA_CPUQ_CFG_CPUQ_MIRROR_X(x)                     (((x) & GENMASK(5, 3)) >> 3)
-#define ANA_CPUQ_CFG_CPUQ_SFLOW(x)                        ((x) & GENMASK(2, 0))
-#define ANA_CPUQ_CFG_CPUQ_SFLOW_M                         GENMASK(2, 0)
-
-#define ANA_CPUQ_8021_CFG_RSZ                             0x4
-
-#define ANA_CPUQ_8021_CFG_CPUQ_BPDU_VAL(x)                (((x) << 6) & GENMASK(8, 6))
-#define ANA_CPUQ_8021_CFG_CPUQ_BPDU_VAL_M                 GENMASK(8, 6)
-#define ANA_CPUQ_8021_CFG_CPUQ_BPDU_VAL_X(x)              (((x) & GENMASK(8, 6)) >> 6)
-#define ANA_CPUQ_8021_CFG_CPUQ_GARP_VAL(x)                (((x) << 3) & GENMASK(5, 3))
-#define ANA_CPUQ_8021_CFG_CPUQ_GARP_VAL_M                 GENMASK(5, 3)
-#define ANA_CPUQ_8021_CFG_CPUQ_GARP_VAL_X(x)              (((x) & GENMASK(5, 3)) >> 3)
-#define ANA_CPUQ_8021_CFG_CPUQ_CCM_VAL(x)                 ((x) & GENMASK(2, 0))
-#define ANA_CPUQ_8021_CFG_CPUQ_CCM_VAL_M                  GENMASK(2, 0)
-
-#define ANA_DSCP_CFG_RSZ                                  0x4
-
-#define ANA_DSCP_CFG_DP_DSCP_VAL                          BIT(11)
-#define ANA_DSCP_CFG_QOS_DSCP_VAL(x)                      (((x) << 8) & GENMASK(10, 8))
-#define ANA_DSCP_CFG_QOS_DSCP_VAL_M                       GENMASK(10, 8)
-#define ANA_DSCP_CFG_QOS_DSCP_VAL_X(x)                    (((x) & GENMASK(10, 8)) >> 8)
-#define ANA_DSCP_CFG_DSCP_TRANSLATE_VAL(x)                (((x) << 2) & GENMASK(7, 2))
-#define ANA_DSCP_CFG_DSCP_TRANSLATE_VAL_M                 GENMASK(7, 2)
-#define ANA_DSCP_CFG_DSCP_TRANSLATE_VAL_X(x)              (((x) & GENMASK(7, 2)) >> 2)
-#define ANA_DSCP_CFG_DSCP_TRUST_ENA                       BIT(1)
-#define ANA_DSCP_CFG_DSCP_REWR_ENA                        BIT(0)
-
-#define ANA_DSCP_REWR_CFG_RSZ                             0x4
-
-#define ANA_VCAP_RNG_TYPE_CFG_RSZ                         0x4
-
-#define ANA_VCAP_RNG_VAL_CFG_RSZ                          0x4
-
-#define ANA_VCAP_RNG_VAL_CFG_VCAP_RNG_MIN_VAL(x)          (((x) << 16) & GENMASK(31, 16))
-#define ANA_VCAP_RNG_VAL_CFG_VCAP_RNG_MIN_VAL_M           GENMASK(31, 16)
-#define ANA_VCAP_RNG_VAL_CFG_VCAP_RNG_MIN_VAL_X(x)        (((x) & GENMASK(31, 16)) >> 16)
-#define ANA_VCAP_RNG_VAL_CFG_VCAP_RNG_MAX_VAL(x)          ((x) & GENMASK(15, 0))
-#define ANA_VCAP_RNG_VAL_CFG_VCAP_RNG_MAX_VAL_M           GENMASK(15, 0)
-
-#define ANA_VRAP_CFG_VRAP_VLAN_AWARE_ENA                  BIT(12)
-#define ANA_VRAP_CFG_VRAP_VID(x)                          ((x) & GENMASK(11, 0))
-#define ANA_VRAP_CFG_VRAP_VID_M                           GENMASK(11, 0)
-
-#define ANA_DISCARD_CFG_DROP_TAGGING_ISDX0                BIT(3)
-#define ANA_DISCARD_CFG_DROP_CTRLPROT_ISDX0               BIT(2)
-#define ANA_DISCARD_CFG_DROP_TAGGING_S2_ENA               BIT(1)
-#define ANA_DISCARD_CFG_DROP_CTRLPROT_S2_ENA              BIT(0)
-
-#define ANA_FID_CFG_VID_MC_ENA                            BIT(0)
-
-#define ANA_POL_PIR_CFG_GSZ                               0x20
-
-#define ANA_POL_PIR_CFG_PIR_RATE(x)                       (((x) << 6) & GENMASK(20, 6))
-#define ANA_POL_PIR_CFG_PIR_RATE_M                        GENMASK(20, 6)
-#define ANA_POL_PIR_CFG_PIR_RATE_X(x)                     (((x) & GENMASK(20, 6)) >> 6)
-#define ANA_POL_PIR_CFG_PIR_BURST(x)                      ((x) & GENMASK(5, 0))
-#define ANA_POL_PIR_CFG_PIR_BURST_M                       GENMASK(5, 0)
-
-#define ANA_POL_CIR_CFG_GSZ                               0x20
-
-#define ANA_POL_CIR_CFG_CIR_RATE(x)                       (((x) << 6) & GENMASK(20, 6))
-#define ANA_POL_CIR_CFG_CIR_RATE_M                        GENMASK(20, 6)
-#define ANA_POL_CIR_CFG_CIR_RATE_X(x)                     (((x) & GENMASK(20, 6)) >> 6)
-#define ANA_POL_CIR_CFG_CIR_BURST(x)                      ((x) & GENMASK(5, 0))
-#define ANA_POL_CIR_CFG_CIR_BURST_M                       GENMASK(5, 0)
-
-#define ANA_POL_MODE_CFG_GSZ                              0x20
-
-#define ANA_POL_MODE_CFG_IPG_SIZE(x)                      (((x) << 5) & GENMASK(9, 5))
-#define ANA_POL_MODE_CFG_IPG_SIZE_M                       GENMASK(9, 5)
-#define ANA_POL_MODE_CFG_IPG_SIZE_X(x)                    (((x) & GENMASK(9, 5)) >> 5)
-#define ANA_POL_MODE_CFG_FRM_MODE(x)                      (((x) << 3) & GENMASK(4, 3))
-#define ANA_POL_MODE_CFG_FRM_MODE_M                       GENMASK(4, 3)
-#define ANA_POL_MODE_CFG_FRM_MODE_X(x)                    (((x) & GENMASK(4, 3)) >> 3)
-#define ANA_POL_MODE_CFG_DLB_COUPLED                      BIT(2)
-#define ANA_POL_MODE_CFG_CIR_ENA                          BIT(1)
-#define ANA_POL_MODE_CFG_OVERSHOOT_ENA                    BIT(0)
-
-#define ANA_POL_PIR_STATE_GSZ                             0x20
-
-#define ANA_POL_CIR_STATE_GSZ                             0x20
-
-#define ANA_POL_STATE_GSZ                                 0x20
-
-#define ANA_POL_FLOWC_RSZ                                 0x4
-
-#define ANA_POL_FLOWC_POL_FLOWC                           BIT(0)
-
-#define ANA_POL_HYST_POL_FC_HYST(x)                       (((x) << 4) & GENMASK(9, 4))
-#define ANA_POL_HYST_POL_FC_HYST_M                        GENMASK(9, 4)
-#define ANA_POL_HYST_POL_FC_HYST_X(x)                     (((x) & GENMASK(9, 4)) >> 4)
-#define ANA_POL_HYST_POL_STOP_HYST(x)                     ((x) & GENMASK(3, 0))
-#define ANA_POL_HYST_POL_STOP_HYST_M                      GENMASK(3, 0)
-
-#define ANA_POL_MISC_CFG_POL_CLOSE_ALL                    BIT(1)
-#define ANA_POL_MISC_CFG_POL_LEAK_DIS                     BIT(0)
-
-#endif
diff --git a/drivers/net/ethernet/mscc/ocelot_dev.h b/drivers/net/ethernet/mscc/ocelot_dev.h
deleted file mode 100644
index 0a50d53bbd3f..000000000000
--- a/drivers/net/ethernet/mscc/ocelot_dev.h
+++ /dev/null
@@ -1,275 +0,0 @@
-/* SPDX-License-Identifier: (GPL-2.0 OR MIT) */
-/*
- * Microsemi Ocelot Switch driver
- *
- * Copyright (c) 2017 Microsemi Corporation
- */
-
-#ifndef _MSCC_OCELOT_DEV_H_
-#define _MSCC_OCELOT_DEV_H_
-
-#define DEV_CLOCK_CFG                                     0x0
-
-#define DEV_CLOCK_CFG_MAC_TX_RST                          BIT(7)
-#define DEV_CLOCK_CFG_MAC_RX_RST                          BIT(6)
-#define DEV_CLOCK_CFG_PCS_TX_RST                          BIT(5)
-#define DEV_CLOCK_CFG_PCS_RX_RST                          BIT(4)
-#define DEV_CLOCK_CFG_PORT_RST                            BIT(3)
-#define DEV_CLOCK_CFG_PHY_RST                             BIT(2)
-#define DEV_CLOCK_CFG_LINK_SPEED(x)                       ((x) & GENMASK(1, 0))
-#define DEV_CLOCK_CFG_LINK_SPEED_M                        GENMASK(1, 0)
-
-#define DEV_PORT_MISC                                     0x4
-
-#define DEV_PORT_MISC_FWD_ERROR_ENA                       BIT(4)
-#define DEV_PORT_MISC_FWD_PAUSE_ENA                       BIT(3)
-#define DEV_PORT_MISC_FWD_CTRL_ENA                        BIT(2)
-#define DEV_PORT_MISC_DEV_LOOP_ENA                        BIT(1)
-#define DEV_PORT_MISC_HDX_FAST_DIS                        BIT(0)
-
-#define DEV_EVENTS                                        0x8
-
-#define DEV_EEE_CFG                                       0xc
-
-#define DEV_EEE_CFG_EEE_ENA                               BIT(22)
-#define DEV_EEE_CFG_EEE_TIMER_AGE(x)                      (((x) << 15) & GENMASK(21, 15))
-#define DEV_EEE_CFG_EEE_TIMER_AGE_M                       GENMASK(21, 15)
-#define DEV_EEE_CFG_EEE_TIMER_AGE_X(x)                    (((x) & GENMASK(21, 15)) >> 15)
-#define DEV_EEE_CFG_EEE_TIMER_WAKEUP(x)                   (((x) << 8) & GENMASK(14, 8))
-#define DEV_EEE_CFG_EEE_TIMER_WAKEUP_M                    GENMASK(14, 8)
-#define DEV_EEE_CFG_EEE_TIMER_WAKEUP_X(x)                 (((x) & GENMASK(14, 8)) >> 8)
-#define DEV_EEE_CFG_EEE_TIMER_HOLDOFF(x)                  (((x) << 1) & GENMASK(7, 1))
-#define DEV_EEE_CFG_EEE_TIMER_HOLDOFF_M                   GENMASK(7, 1)
-#define DEV_EEE_CFG_EEE_TIMER_HOLDOFF_X(x)                (((x) & GENMASK(7, 1)) >> 1)
-#define DEV_EEE_CFG_PORT_LPI                              BIT(0)
-
-#define DEV_RX_PATH_DELAY                                 0x10
-
-#define DEV_TX_PATH_DELAY                                 0x14
-
-#define DEV_PTP_PREDICT_CFG                               0x18
-
-#define DEV_PTP_PREDICT_CFG_PTP_PHY_PREDICT_CFG(x)        (((x) << 4) & GENMASK(11, 4))
-#define DEV_PTP_PREDICT_CFG_PTP_PHY_PREDICT_CFG_M         GENMASK(11, 4)
-#define DEV_PTP_PREDICT_CFG_PTP_PHY_PREDICT_CFG_X(x)      (((x) & GENMASK(11, 4)) >> 4)
-#define DEV_PTP_PREDICT_CFG_PTP_PHASE_PREDICT_CFG(x)      ((x) & GENMASK(3, 0))
-#define DEV_PTP_PREDICT_CFG_PTP_PHASE_PREDICT_CFG_M       GENMASK(3, 0)
-
-#define DEV_MAC_ENA_CFG                                   0x1c
-
-#define DEV_MAC_ENA_CFG_RX_ENA                            BIT(4)
-#define DEV_MAC_ENA_CFG_TX_ENA                            BIT(0)
-
-#define DEV_MAC_MODE_CFG                                  0x20
-
-#define DEV_MAC_MODE_CFG_FC_WORD_SYNC_ENA                 BIT(8)
-#define DEV_MAC_MODE_CFG_GIGA_MODE_ENA                    BIT(4)
-#define DEV_MAC_MODE_CFG_FDX_ENA                          BIT(0)
-
-#define DEV_MAC_MAXLEN_CFG                                0x24
-
-#define DEV_MAC_TAGS_CFG                                  0x28
-
-#define DEV_MAC_TAGS_CFG_TAG_ID(x)                        (((x) << 16) & GENMASK(31, 16))
-#define DEV_MAC_TAGS_CFG_TAG_ID_M                         GENMASK(31, 16)
-#define DEV_MAC_TAGS_CFG_TAG_ID_X(x)                      (((x) & GENMASK(31, 16)) >> 16)
-#define DEV_MAC_TAGS_CFG_VLAN_LEN_AWR_ENA                 BIT(2)
-#define DEV_MAC_TAGS_CFG_PB_ENA                           BIT(1)
-#define DEV_MAC_TAGS_CFG_VLAN_AWR_ENA                     BIT(0)
-
-#define DEV_MAC_ADV_CHK_CFG                               0x2c
-
-#define DEV_MAC_ADV_CHK_CFG_LEN_DROP_ENA                  BIT(0)
-
-#define DEV_MAC_IFG_CFG                                   0x30
-
-#define DEV_MAC_IFG_CFG_RESTORE_OLD_IPG_CHECK             BIT(17)
-#define DEV_MAC_IFG_CFG_REDUCED_TX_IFG                    BIT(16)
-#define DEV_MAC_IFG_CFG_TX_IFG(x)                         (((x) << 8) & GENMASK(12, 8))
-#define DEV_MAC_IFG_CFG_TX_IFG_M                          GENMASK(12, 8)
-#define DEV_MAC_IFG_CFG_TX_IFG_X(x)                       (((x) & GENMASK(12, 8)) >> 8)
-#define DEV_MAC_IFG_CFG_RX_IFG2(x)                        (((x) << 4) & GENMASK(7, 4))
-#define DEV_MAC_IFG_CFG_RX_IFG2_M                         GENMASK(7, 4)
-#define DEV_MAC_IFG_CFG_RX_IFG2_X(x)                      (((x) & GENMASK(7, 4)) >> 4)
-#define DEV_MAC_IFG_CFG_RX_IFG1(x)                        ((x) & GENMASK(3, 0))
-#define DEV_MAC_IFG_CFG_RX_IFG1_M                         GENMASK(3, 0)
-
-#define DEV_MAC_HDX_CFG                                   0x34
-
-#define DEV_MAC_HDX_CFG_BYPASS_COL_SYNC                   BIT(26)
-#define DEV_MAC_HDX_CFG_OB_ENA                            BIT(25)
-#define DEV_MAC_HDX_CFG_WEXC_DIS                          BIT(24)
-#define DEV_MAC_HDX_CFG_SEED(x)                           (((x) << 16) & GENMASK(23, 16))
-#define DEV_MAC_HDX_CFG_SEED_M                            GENMASK(23, 16)
-#define DEV_MAC_HDX_CFG_SEED_X(x)                         (((x) & GENMASK(23, 16)) >> 16)
-#define DEV_MAC_HDX_CFG_SEED_LOAD                         BIT(12)
-#define DEV_MAC_HDX_CFG_RETRY_AFTER_EXC_COL_ENA           BIT(8)
-#define DEV_MAC_HDX_CFG_LATE_COL_POS(x)                   ((x) & GENMASK(6, 0))
-#define DEV_MAC_HDX_CFG_LATE_COL_POS_M                    GENMASK(6, 0)
-
-#define DEV_MAC_DBG_CFG                                   0x38
-
-#define DEV_MAC_DBG_CFG_TBI_MODE                          BIT(4)
-#define DEV_MAC_DBG_CFG_IFG_CRS_EXT_CHK_ENA               BIT(0)
-
-#define DEV_MAC_FC_MAC_LOW_CFG                            0x3c
-
-#define DEV_MAC_FC_MAC_HIGH_CFG                           0x40
-
-#define DEV_MAC_STICKY                                    0x44
-
-#define DEV_MAC_STICKY_RX_IPG_SHRINK_STICKY               BIT(9)
-#define DEV_MAC_STICKY_RX_PREAM_SHRINK_STICKY             BIT(8)
-#define DEV_MAC_STICKY_RX_CARRIER_EXT_STICKY              BIT(7)
-#define DEV_MAC_STICKY_RX_CARRIER_EXT_ERR_STICKY          BIT(6)
-#define DEV_MAC_STICKY_RX_JUNK_STICKY                     BIT(5)
-#define DEV_MAC_STICKY_TX_RETRANSMIT_STICKY               BIT(4)
-#define DEV_MAC_STICKY_TX_JAM_STICKY                      BIT(3)
-#define DEV_MAC_STICKY_TX_FIFO_OFLW_STICKY                BIT(2)
-#define DEV_MAC_STICKY_TX_FRM_LEN_OVR_STICKY              BIT(1)
-#define DEV_MAC_STICKY_TX_ABORT_STICKY                    BIT(0)
-
-#define PCS1G_CFG                                         0x48
-
-#define PCS1G_CFG_LINK_STATUS_TYPE                        BIT(4)
-#define PCS1G_CFG_AN_LINK_CTRL_ENA                        BIT(1)
-#define PCS1G_CFG_PCS_ENA                                 BIT(0)
-
-#define PCS1G_MODE_CFG                                    0x4c
-
-#define PCS1G_MODE_CFG_UNIDIR_MODE_ENA                    BIT(4)
-#define PCS1G_MODE_CFG_SGMII_MODE_ENA                     BIT(0)
-
-#define PCS1G_SD_CFG                                      0x50
-
-#define PCS1G_SD_CFG_SD_SEL                               BIT(8)
-#define PCS1G_SD_CFG_SD_POL                               BIT(4)
-#define PCS1G_SD_CFG_SD_ENA                               BIT(0)
-
-#define PCS1G_ANEG_CFG                                    0x54
-
-#define PCS1G_ANEG_CFG_ADV_ABILITY(x)                     (((x) << 16) & GENMASK(31, 16))
-#define PCS1G_ANEG_CFG_ADV_ABILITY_M                      GENMASK(31, 16)
-#define PCS1G_ANEG_CFG_ADV_ABILITY_X(x)                   (((x) & GENMASK(31, 16)) >> 16)
-#define PCS1G_ANEG_CFG_SW_RESOLVE_ENA                     BIT(8)
-#define PCS1G_ANEG_CFG_ANEG_RESTART_ONE_SHOT              BIT(1)
-#define PCS1G_ANEG_CFG_ANEG_ENA                           BIT(0)
-
-#define PCS1G_ANEG_NP_CFG                                 0x58
-
-#define PCS1G_ANEG_NP_CFG_NP_TX(x)                        (((x) << 16) & GENMASK(31, 16))
-#define PCS1G_ANEG_NP_CFG_NP_TX_M                         GENMASK(31, 16)
-#define PCS1G_ANEG_NP_CFG_NP_TX_X(x)                      (((x) & GENMASK(31, 16)) >> 16)
-#define PCS1G_ANEG_NP_CFG_NP_LOADED_ONE_SHOT              BIT(0)
-
-#define PCS1G_LB_CFG                                      0x5c
-
-#define PCS1G_LB_CFG_RA_ENA                               BIT(4)
-#define PCS1G_LB_CFG_GMII_PHY_LB_ENA                      BIT(1)
-#define PCS1G_LB_CFG_TBI_HOST_LB_ENA                      BIT(0)
-
-#define PCS1G_DBG_CFG                                     0x60
-
-#define PCS1G_DBG_CFG_UDLT                                BIT(0)
-
-#define PCS1G_CDET_CFG                                    0x64
-
-#define PCS1G_CDET_CFG_CDET_ENA                           BIT(0)
-
-#define PCS1G_ANEG_STATUS                                 0x68
-
-#define PCS1G_ANEG_STATUS_LP_ADV_ABILITY(x)               (((x) << 16) & GENMASK(31, 16))
-#define PCS1G_ANEG_STATUS_LP_ADV_ABILITY_M                GENMASK(31, 16)
-#define PCS1G_ANEG_STATUS_LP_ADV_ABILITY_X(x)             (((x) & GENMASK(31, 16)) >> 16)
-#define PCS1G_ANEG_STATUS_PR                              BIT(4)
-#define PCS1G_ANEG_STATUS_PAGE_RX_STICKY                  BIT(3)
-#define PCS1G_ANEG_STATUS_ANEG_COMPLETE                   BIT(0)
-
-#define PCS1G_ANEG_NP_STATUS                              0x6c
-
-#define PCS1G_LINK_STATUS                                 0x70
-
-#define PCS1G_LINK_STATUS_DELAY_VAR(x)                    (((x) << 12) & GENMASK(15, 12))
-#define PCS1G_LINK_STATUS_DELAY_VAR_M                     GENMASK(15, 12)
-#define PCS1G_LINK_STATUS_DELAY_VAR_X(x)                  (((x) & GENMASK(15, 12)) >> 12)
-#define PCS1G_LINK_STATUS_SIGNAL_DETECT                   BIT(8)
-#define PCS1G_LINK_STATUS_LINK_STATUS                     BIT(4)
-#define PCS1G_LINK_STATUS_SYNC_STATUS                     BIT(0)
-
-#define PCS1G_LINK_DOWN_CNT                               0x74
-
-#define PCS1G_STICKY                                      0x78
-
-#define PCS1G_STICKY_LINK_DOWN_STICKY                     BIT(4)
-#define PCS1G_STICKY_OUT_OF_SYNC_STICKY                   BIT(0)
-
-#define PCS1G_DEBUG_STATUS                                0x7c
-
-#define PCS1G_LPI_CFG                                     0x80
-
-#define PCS1G_LPI_CFG_QSGMII_MS_SEL                       BIT(20)
-#define PCS1G_LPI_CFG_RX_LPI_OUT_DIS                      BIT(17)
-#define PCS1G_LPI_CFG_LPI_TESTMODE                        BIT(16)
-#define PCS1G_LPI_CFG_LPI_RX_WTIM(x)                      (((x) << 4) & GENMASK(5, 4))
-#define PCS1G_LPI_CFG_LPI_RX_WTIM_M                       GENMASK(5, 4)
-#define PCS1G_LPI_CFG_LPI_RX_WTIM_X(x)                    (((x) & GENMASK(5, 4)) >> 4)
-#define PCS1G_LPI_CFG_TX_ASSERT_LPIDLE                    BIT(0)
-
-#define PCS1G_LPI_WAKE_ERROR_CNT                          0x84
-
-#define PCS1G_LPI_STATUS                                  0x88
-
-#define PCS1G_LPI_STATUS_RX_LPI_FAIL                      BIT(16)
-#define PCS1G_LPI_STATUS_RX_LPI_EVENT_STICKY              BIT(12)
-#define PCS1G_LPI_STATUS_RX_QUIET                         BIT(9)
-#define PCS1G_LPI_STATUS_RX_LPI_MODE                      BIT(8)
-#define PCS1G_LPI_STATUS_TX_LPI_EVENT_STICKY              BIT(4)
-#define PCS1G_LPI_STATUS_TX_QUIET                         BIT(1)
-#define PCS1G_LPI_STATUS_TX_LPI_MODE                      BIT(0)
-
-#define PCS1G_TSTPAT_MODE_CFG                             0x8c
-
-#define PCS1G_TSTPAT_STATUS                               0x90
-
-#define PCS1G_TSTPAT_STATUS_JTP_ERR_CNT(x)                (((x) << 8) & GENMASK(15, 8))
-#define PCS1G_TSTPAT_STATUS_JTP_ERR_CNT_M                 GENMASK(15, 8)
-#define PCS1G_TSTPAT_STATUS_JTP_ERR_CNT_X(x)              (((x) & GENMASK(15, 8)) >> 8)
-#define PCS1G_TSTPAT_STATUS_JTP_ERR                       BIT(4)
-#define PCS1G_TSTPAT_STATUS_JTP_LOCK                      BIT(0)
-
-#define DEV_PCS_FX100_CFG                                 0x94
-
-#define DEV_PCS_FX100_CFG_SD_SEL                          BIT(26)
-#define DEV_PCS_FX100_CFG_SD_POL                          BIT(25)
-#define DEV_PCS_FX100_CFG_SD_ENA                          BIT(24)
-#define DEV_PCS_FX100_CFG_LOOPBACK_ENA                    BIT(20)
-#define DEV_PCS_FX100_CFG_SWAP_MII_ENA                    BIT(16)
-#define DEV_PCS_FX100_CFG_RXBITSEL(x)                     (((x) << 12) & GENMASK(15, 12))
-#define DEV_PCS_FX100_CFG_RXBITSEL_M                      GENMASK(15, 12)
-#define DEV_PCS_FX100_CFG_RXBITSEL_X(x)                   (((x) & GENMASK(15, 12)) >> 12)
-#define DEV_PCS_FX100_CFG_SIGDET_CFG(x)                   (((x) << 9) & GENMASK(10, 9))
-#define DEV_PCS_FX100_CFG_SIGDET_CFG_M                    GENMASK(10, 9)
-#define DEV_PCS_FX100_CFG_SIGDET_CFG_X(x)                 (((x) & GENMASK(10, 9)) >> 9)
-#define DEV_PCS_FX100_CFG_LINKHYST_TM_ENA                 BIT(8)
-#define DEV_PCS_FX100_CFG_LINKHYSTTIMER(x)                (((x) << 4) & GENMASK(7, 4))
-#define DEV_PCS_FX100_CFG_LINKHYSTTIMER_M                 GENMASK(7, 4)
-#define DEV_PCS_FX100_CFG_LINKHYSTTIMER_X(x)              (((x) & GENMASK(7, 4)) >> 4)
-#define DEV_PCS_FX100_CFG_UNIDIR_MODE_ENA                 BIT(3)
-#define DEV_PCS_FX100_CFG_FEFCHK_ENA                      BIT(2)
-#define DEV_PCS_FX100_CFG_FEFGEN_ENA                      BIT(1)
-#define DEV_PCS_FX100_CFG_PCS_ENA                         BIT(0)
-
-#define DEV_PCS_FX100_STATUS                              0x98
-
-#define DEV_PCS_FX100_STATUS_EDGE_POS_PTP(x)              (((x) << 8) & GENMASK(11, 8))
-#define DEV_PCS_FX100_STATUS_EDGE_POS_PTP_M               GENMASK(11, 8)
-#define DEV_PCS_FX100_STATUS_EDGE_POS_PTP_X(x)            (((x) & GENMASK(11, 8)) >> 8)
-#define DEV_PCS_FX100_STATUS_PCS_ERROR_STICKY             BIT(7)
-#define DEV_PCS_FX100_STATUS_FEF_FOUND_STICKY             BIT(6)
-#define DEV_PCS_FX100_STATUS_SSD_ERROR_STICKY             BIT(5)
-#define DEV_PCS_FX100_STATUS_SYNC_LOST_STICKY             BIT(4)
-#define DEV_PCS_FX100_STATUS_FEF_STATUS                   BIT(2)
-#define DEV_PCS_FX100_STATUS_SIGNAL_DETECT                BIT(1)
-#define DEV_PCS_FX100_STATUS_SYNC_STATUS                  BIT(0)
-
-#endif
diff --git a/drivers/net/ethernet/mscc/ocelot_qsys.h b/drivers/net/ethernet/mscc/ocelot_qsys.h
deleted file mode 100644
index d8c63aa761be..000000000000
--- a/drivers/net/ethernet/mscc/ocelot_qsys.h
+++ /dev/null
@@ -1,270 +0,0 @@
-/* SPDX-License-Identifier: (GPL-2.0 OR MIT) */
-/*
- * Microsemi Ocelot Switch driver
- *
- * Copyright (c) 2017 Microsemi Corporation
- */
-
-#ifndef _MSCC_OCELOT_QSYS_H_
-#define _MSCC_OCELOT_QSYS_H_
-
-#define QSYS_PORT_MODE_RSZ                                0x4
-
-#define QSYS_PORT_MODE_DEQUEUE_DIS                        BIT(1)
-#define QSYS_PORT_MODE_DEQUEUE_LATE                       BIT(0)
-
-#define QSYS_SWITCH_PORT_MODE_RSZ                         0x4
-
-#define QSYS_SWITCH_PORT_MODE_PORT_ENA                    BIT(14)
-#define QSYS_SWITCH_PORT_MODE_SCH_NEXT_CFG(x)             (((x) << 11) & GENMASK(13, 11))
-#define QSYS_SWITCH_PORT_MODE_SCH_NEXT_CFG_M              GENMASK(13, 11)
-#define QSYS_SWITCH_PORT_MODE_SCH_NEXT_CFG_X(x)           (((x) & GENMASK(13, 11)) >> 11)
-#define QSYS_SWITCH_PORT_MODE_YEL_RSRVD                   BIT(10)
-#define QSYS_SWITCH_PORT_MODE_INGRESS_DROP_MODE           BIT(9)
-#define QSYS_SWITCH_PORT_MODE_TX_PFC_ENA(x)               (((x) << 1) & GENMASK(8, 1))
-#define QSYS_SWITCH_PORT_MODE_TX_PFC_ENA_M                GENMASK(8, 1)
-#define QSYS_SWITCH_PORT_MODE_TX_PFC_ENA_X(x)             (((x) & GENMASK(8, 1)) >> 1)
-#define QSYS_SWITCH_PORT_MODE_TX_PFC_MODE                 BIT(0)
-
-#define QSYS_STAT_CNT_CFG_TX_GREEN_CNT_MODE               BIT(5)
-#define QSYS_STAT_CNT_CFG_TX_YELLOW_CNT_MODE              BIT(4)
-#define QSYS_STAT_CNT_CFG_DROP_GREEN_CNT_MODE             BIT(3)
-#define QSYS_STAT_CNT_CFG_DROP_YELLOW_CNT_MODE            BIT(2)
-#define QSYS_STAT_CNT_CFG_DROP_COUNT_ONCE                 BIT(1)
-#define QSYS_STAT_CNT_CFG_DROP_COUNT_EGRESS               BIT(0)
-
-#define QSYS_EEE_CFG_RSZ                                  0x4
-
-#define QSYS_EEE_THRES_EEE_HIGH_BYTES(x)                  (((x) << 8) & GENMASK(15, 8))
-#define QSYS_EEE_THRES_EEE_HIGH_BYTES_M                   GENMASK(15, 8)
-#define QSYS_EEE_THRES_EEE_HIGH_BYTES_X(x)                (((x) & GENMASK(15, 8)) >> 8)
-#define QSYS_EEE_THRES_EEE_HIGH_FRAMES(x)                 ((x) & GENMASK(7, 0))
-#define QSYS_EEE_THRES_EEE_HIGH_FRAMES_M                  GENMASK(7, 0)
-
-#define QSYS_SW_STATUS_RSZ                                0x4
-
-#define QSYS_EXT_CPU_CFG_EXT_CPU_PORT(x)                  (((x) << 8) & GENMASK(12, 8))
-#define QSYS_EXT_CPU_CFG_EXT_CPU_PORT_M                   GENMASK(12, 8)
-#define QSYS_EXT_CPU_CFG_EXT_CPU_PORT_X(x)                (((x) & GENMASK(12, 8)) >> 8)
-#define QSYS_EXT_CPU_CFG_EXT_CPUQ_MSK(x)                  ((x) & GENMASK(7, 0))
-#define QSYS_EXT_CPU_CFG_EXT_CPUQ_MSK_M                   GENMASK(7, 0)
-
-#define QSYS_QMAP_GSZ                                     0x4
-
-#define QSYS_QMAP_SE_BASE(x)                              (((x) << 5) & GENMASK(12, 5))
-#define QSYS_QMAP_SE_BASE_M                               GENMASK(12, 5)
-#define QSYS_QMAP_SE_BASE_X(x)                            (((x) & GENMASK(12, 5)) >> 5)
-#define QSYS_QMAP_SE_IDX_SEL(x)                           (((x) << 2) & GENMASK(4, 2))
-#define QSYS_QMAP_SE_IDX_SEL_M                            GENMASK(4, 2)
-#define QSYS_QMAP_SE_IDX_SEL_X(x)                         (((x) & GENMASK(4, 2)) >> 2)
-#define QSYS_QMAP_SE_INP_SEL(x)                           ((x) & GENMASK(1, 0))
-#define QSYS_QMAP_SE_INP_SEL_M                            GENMASK(1, 0)
-
-#define QSYS_ISDX_SGRP_GSZ                                0x4
-
-#define QSYS_TIMED_FRAME_ENTRY_GSZ                        0x4
-
-#define QSYS_TFRM_MISC_TIMED_CANCEL_SLOT(x)               (((x) << 9) & GENMASK(18, 9))
-#define QSYS_TFRM_MISC_TIMED_CANCEL_SLOT_M                GENMASK(18, 9)
-#define QSYS_TFRM_MISC_TIMED_CANCEL_SLOT_X(x)             (((x) & GENMASK(18, 9)) >> 9)
-#define QSYS_TFRM_MISC_TIMED_CANCEL_1SHOT                 BIT(8)
-#define QSYS_TFRM_MISC_TIMED_SLOT_MODE_MC                 BIT(7)
-#define QSYS_TFRM_MISC_TIMED_ENTRY_FAST_CNT(x)            ((x) & GENMASK(6, 0))
-#define QSYS_TFRM_MISC_TIMED_ENTRY_FAST_CNT_M             GENMASK(6, 0)
-
-#define QSYS_RED_PROFILE_RSZ                              0x4
-
-#define QSYS_RED_PROFILE_WM_RED_LOW(x)                    (((x) << 8) & GENMASK(15, 8))
-#define QSYS_RED_PROFILE_WM_RED_LOW_M                     GENMASK(15, 8)
-#define QSYS_RED_PROFILE_WM_RED_LOW_X(x)                  (((x) & GENMASK(15, 8)) >> 8)
-#define QSYS_RED_PROFILE_WM_RED_HIGH(x)                   ((x) & GENMASK(7, 0))
-#define QSYS_RED_PROFILE_WM_RED_HIGH_M                    GENMASK(7, 0)
-
-#define QSYS_RES_CFG_GSZ                                  0x8
-
-#define QSYS_RES_STAT_GSZ                                 0x8
-
-#define QSYS_RES_STAT_INUSE(x)                            (((x) << 12) & GENMASK(23, 12))
-#define QSYS_RES_STAT_INUSE_M                             GENMASK(23, 12)
-#define QSYS_RES_STAT_INUSE_X(x)                          (((x) & GENMASK(23, 12)) >> 12)
-#define QSYS_RES_STAT_MAXUSE(x)                           ((x) & GENMASK(11, 0))
-#define QSYS_RES_STAT_MAXUSE_M                            GENMASK(11, 0)
-
-#define QSYS_EVENTS_CORE_EV_FDC(x)                        (((x) << 2) & GENMASK(4, 2))
-#define QSYS_EVENTS_CORE_EV_FDC_M                         GENMASK(4, 2)
-#define QSYS_EVENTS_CORE_EV_FDC_X(x)                      (((x) & GENMASK(4, 2)) >> 2)
-#define QSYS_EVENTS_CORE_EV_FRD(x)                        ((x) & GENMASK(1, 0))
-#define QSYS_EVENTS_CORE_EV_FRD_M                         GENMASK(1, 0)
-
-#define QSYS_QMAXSDU_CFG_0_RSZ                            0x4
-
-#define QSYS_QMAXSDU_CFG_1_RSZ                            0x4
-
-#define QSYS_QMAXSDU_CFG_2_RSZ                            0x4
-
-#define QSYS_QMAXSDU_CFG_3_RSZ                            0x4
-
-#define QSYS_QMAXSDU_CFG_4_RSZ                            0x4
-
-#define QSYS_QMAXSDU_CFG_5_RSZ                            0x4
-
-#define QSYS_QMAXSDU_CFG_6_RSZ                            0x4
-
-#define QSYS_QMAXSDU_CFG_7_RSZ                            0x4
-
-#define QSYS_PREEMPTION_CFG_RSZ                           0x4
-
-#define QSYS_PREEMPTION_CFG_P_QUEUES(x)                   ((x) & GENMASK(7, 0))
-#define QSYS_PREEMPTION_CFG_P_QUEUES_M                    GENMASK(7, 0)
-#define QSYS_PREEMPTION_CFG_MM_ADD_FRAG_SIZE(x)           (((x) << 8) & GENMASK(9, 8))
-#define QSYS_PREEMPTION_CFG_MM_ADD_FRAG_SIZE_M            GENMASK(9, 8)
-#define QSYS_PREEMPTION_CFG_MM_ADD_FRAG_SIZE_X(x)         (((x) & GENMASK(9, 8)) >> 8)
-#define QSYS_PREEMPTION_CFG_STRICT_IPG(x)                 (((x) << 12) & GENMASK(13, 12))
-#define QSYS_PREEMPTION_CFG_STRICT_IPG_M                  GENMASK(13, 12)
-#define QSYS_PREEMPTION_CFG_STRICT_IPG_X(x)               (((x) & GENMASK(13, 12)) >> 12)
-#define QSYS_PREEMPTION_CFG_HOLD_ADVANCE(x)               (((x) << 16) & GENMASK(31, 16))
-#define QSYS_PREEMPTION_CFG_HOLD_ADVANCE_M                GENMASK(31, 16)
-#define QSYS_PREEMPTION_CFG_HOLD_ADVANCE_X(x)             (((x) & GENMASK(31, 16)) >> 16)
-
-#define QSYS_CIR_CFG_GSZ                                  0x80
-
-#define QSYS_CIR_CFG_CIR_RATE(x)                          (((x) << 6) & GENMASK(20, 6))
-#define QSYS_CIR_CFG_CIR_RATE_M                           GENMASK(20, 6)
-#define QSYS_CIR_CFG_CIR_RATE_X(x)                        (((x) & GENMASK(20, 6)) >> 6)
-#define QSYS_CIR_CFG_CIR_BURST(x)                         ((x) & GENMASK(5, 0))
-#define QSYS_CIR_CFG_CIR_BURST_M                          GENMASK(5, 0)
-
-#define QSYS_EIR_CFG_GSZ                                  0x80
-
-#define QSYS_EIR_CFG_EIR_RATE(x)                          (((x) << 7) & GENMASK(21, 7))
-#define QSYS_EIR_CFG_EIR_RATE_M                           GENMASK(21, 7)
-#define QSYS_EIR_CFG_EIR_RATE_X(x)                        (((x) & GENMASK(21, 7)) >> 7)
-#define QSYS_EIR_CFG_EIR_BURST(x)                         (((x) << 1) & GENMASK(6, 1))
-#define QSYS_EIR_CFG_EIR_BURST_M                          GENMASK(6, 1)
-#define QSYS_EIR_CFG_EIR_BURST_X(x)                       (((x) & GENMASK(6, 1)) >> 1)
-#define QSYS_EIR_CFG_EIR_MARK_ENA                         BIT(0)
-
-#define QSYS_SE_CFG_GSZ                                   0x80
-
-#define QSYS_SE_CFG_SE_DWRR_CNT(x)                        (((x) << 6) & GENMASK(9, 6))
-#define QSYS_SE_CFG_SE_DWRR_CNT_M                         GENMASK(9, 6)
-#define QSYS_SE_CFG_SE_DWRR_CNT_X(x)                      (((x) & GENMASK(9, 6)) >> 6)
-#define QSYS_SE_CFG_SE_RR_ENA                             BIT(5)
-#define QSYS_SE_CFG_SE_AVB_ENA                            BIT(4)
-#define QSYS_SE_CFG_SE_FRM_MODE(x)                        (((x) << 2) & GENMASK(3, 2))
-#define QSYS_SE_CFG_SE_FRM_MODE_M                         GENMASK(3, 2)
-#define QSYS_SE_CFG_SE_FRM_MODE_X(x)                      (((x) & GENMASK(3, 2)) >> 2)
-#define QSYS_SE_CFG_SE_EXC_ENA                            BIT(1)
-#define QSYS_SE_CFG_SE_EXC_FWD                            BIT(0)
-
-#define QSYS_SE_DWRR_CFG_GSZ                              0x80
-#define QSYS_SE_DWRR_CFG_RSZ                              0x4
-
-#define QSYS_SE_CONNECT_GSZ                               0x80
-
-#define QSYS_SE_CONNECT_SE_OUTP_IDX(x)                    (((x) << 17) & GENMASK(24, 17))
-#define QSYS_SE_CONNECT_SE_OUTP_IDX_M                     GENMASK(24, 17)
-#define QSYS_SE_CONNECT_SE_OUTP_IDX_X(x)                  (((x) & GENMASK(24, 17)) >> 17)
-#define QSYS_SE_CONNECT_SE_INP_IDX(x)                     (((x) << 9) & GENMASK(16, 9))
-#define QSYS_SE_CONNECT_SE_INP_IDX_M                      GENMASK(16, 9)
-#define QSYS_SE_CONNECT_SE_INP_IDX_X(x)                   (((x) & GENMASK(16, 9)) >> 9)
-#define QSYS_SE_CONNECT_SE_OUTP_CON(x)                    (((x) << 5) & GENMASK(8, 5))
-#define QSYS_SE_CONNECT_SE_OUTP_CON_M                     GENMASK(8, 5)
-#define QSYS_SE_CONNECT_SE_OUTP_CON_X(x)                  (((x) & GENMASK(8, 5)) >> 5)
-#define QSYS_SE_CONNECT_SE_INP_CNT(x)                     (((x) << 1) & GENMASK(4, 1))
-#define QSYS_SE_CONNECT_SE_INP_CNT_M                      GENMASK(4, 1)
-#define QSYS_SE_CONNECT_SE_INP_CNT_X(x)                   (((x) & GENMASK(4, 1)) >> 1)
-#define QSYS_SE_CONNECT_SE_TERMINAL                       BIT(0)
-
-#define QSYS_SE_DLB_SENSE_GSZ                             0x80
-
-#define QSYS_SE_DLB_SENSE_SE_DLB_PRIO(x)                  (((x) << 11) & GENMASK(13, 11))
-#define QSYS_SE_DLB_SENSE_SE_DLB_PRIO_M                   GENMASK(13, 11)
-#define QSYS_SE_DLB_SENSE_SE_DLB_PRIO_X(x)                (((x) & GENMASK(13, 11)) >> 11)
-#define QSYS_SE_DLB_SENSE_SE_DLB_SPORT(x)                 (((x) << 7) & GENMASK(10, 7))
-#define QSYS_SE_DLB_SENSE_SE_DLB_SPORT_M                  GENMASK(10, 7)
-#define QSYS_SE_DLB_SENSE_SE_DLB_SPORT_X(x)               (((x) & GENMASK(10, 7)) >> 7)
-#define QSYS_SE_DLB_SENSE_SE_DLB_DPORT(x)                 (((x) << 3) & GENMASK(6, 3))
-#define QSYS_SE_DLB_SENSE_SE_DLB_DPORT_M                  GENMASK(6, 3)
-#define QSYS_SE_DLB_SENSE_SE_DLB_DPORT_X(x)               (((x) & GENMASK(6, 3)) >> 3)
-#define QSYS_SE_DLB_SENSE_SE_DLB_PRIO_ENA                 BIT(2)
-#define QSYS_SE_DLB_SENSE_SE_DLB_SPORT_ENA                BIT(1)
-#define QSYS_SE_DLB_SENSE_SE_DLB_DPORT_ENA                BIT(0)
-
-#define QSYS_CIR_STATE_GSZ                                0x80
-
-#define QSYS_CIR_STATE_CIR_LVL(x)                         (((x) << 4) & GENMASK(25, 4))
-#define QSYS_CIR_STATE_CIR_LVL_M                          GENMASK(25, 4)
-#define QSYS_CIR_STATE_CIR_LVL_X(x)                       (((x) & GENMASK(25, 4)) >> 4)
-#define QSYS_CIR_STATE_SHP_TIME(x)                        ((x) & GENMASK(3, 0))
-#define QSYS_CIR_STATE_SHP_TIME_M                         GENMASK(3, 0)
-
-#define QSYS_EIR_STATE_GSZ                                0x80
-
-#define QSYS_SE_STATE_GSZ                                 0x80
-
-#define QSYS_SE_STATE_SE_OUTP_LVL(x)                      (((x) << 1) & GENMASK(2, 1))
-#define QSYS_SE_STATE_SE_OUTP_LVL_M                       GENMASK(2, 1)
-#define QSYS_SE_STATE_SE_OUTP_LVL_X(x)                    (((x) & GENMASK(2, 1)) >> 1)
-#define QSYS_SE_STATE_SE_WAS_YEL                          BIT(0)
-
-#define QSYS_HSCH_MISC_CFG_SE_CONNECT_VLD                 BIT(8)
-#define QSYS_HSCH_MISC_CFG_FRM_ADJ(x)                     (((x) << 3) & GENMASK(7, 3))
-#define QSYS_HSCH_MISC_CFG_FRM_ADJ_M                      GENMASK(7, 3)
-#define QSYS_HSCH_MISC_CFG_FRM_ADJ_X(x)                   (((x) & GENMASK(7, 3)) >> 3)
-#define QSYS_HSCH_MISC_CFG_LEAK_DIS                       BIT(2)
-#define QSYS_HSCH_MISC_CFG_QSHP_EXC_ENA                   BIT(1)
-#define QSYS_HSCH_MISC_CFG_PFC_BYP_UPD                    BIT(0)
-
-#define QSYS_TAG_CONFIG_RSZ                               0x4
-
-#define QSYS_TAG_CONFIG_ENABLE                            BIT(0)
-#define QSYS_TAG_CONFIG_LINK_SPEED(x)                     (((x) << 4) & GENMASK(5, 4))
-#define QSYS_TAG_CONFIG_LINK_SPEED_M                      GENMASK(5, 4)
-#define QSYS_TAG_CONFIG_LINK_SPEED_X(x)                   (((x) & GENMASK(5, 4)) >> 4)
-#define QSYS_TAG_CONFIG_INIT_GATE_STATE(x)                (((x) << 8) & GENMASK(15, 8))
-#define QSYS_TAG_CONFIG_INIT_GATE_STATE_M                 GENMASK(15, 8)
-#define QSYS_TAG_CONFIG_INIT_GATE_STATE_X(x)              (((x) & GENMASK(15, 8)) >> 8)
-#define QSYS_TAG_CONFIG_SCH_TRAFFIC_QUEUES(x)             (((x) << 16) & GENMASK(23, 16))
-#define QSYS_TAG_CONFIG_SCH_TRAFFIC_QUEUES_M              GENMASK(23, 16)
-#define QSYS_TAG_CONFIG_SCH_TRAFFIC_QUEUES_X(x)           (((x) & GENMASK(23, 16)) >> 16)
-
-#define QSYS_TAS_PARAM_CFG_CTRL_PORT_NUM(x)               ((x) & GENMASK(7, 0))
-#define QSYS_TAS_PARAM_CFG_CTRL_PORT_NUM_M                GENMASK(7, 0)
-#define QSYS_TAS_PARAM_CFG_CTRL_ALWAYS_GUARD_BAND_SCH_Q   BIT(8)
-#define QSYS_TAS_PARAM_CFG_CTRL_CONFIG_CHANGE             BIT(16)
-
-#define QSYS_PORT_MAX_SDU_RSZ                             0x4
-
-#define QSYS_PARAM_CFG_REG_3_BASE_TIME_SEC_MSB(x)         ((x) & GENMASK(15, 0))
-#define QSYS_PARAM_CFG_REG_3_BASE_TIME_SEC_MSB_M          GENMASK(15, 0)
-#define QSYS_PARAM_CFG_REG_3_LIST_LENGTH(x)               (((x) << 16) & GENMASK(31, 16))
-#define QSYS_PARAM_CFG_REG_3_LIST_LENGTH_M                GENMASK(31, 16)
-#define QSYS_PARAM_CFG_REG_3_LIST_LENGTH_X(x)             (((x) & GENMASK(31, 16)) >> 16)
-
-#define QSYS_GCL_CFG_REG_1_GCL_ENTRY_NUM(x)               ((x) & GENMASK(5, 0))
-#define QSYS_GCL_CFG_REG_1_GCL_ENTRY_NUM_M                GENMASK(5, 0)
-#define QSYS_GCL_CFG_REG_1_GATE_STATE(x)                  (((x) << 8) & GENMASK(15, 8))
-#define QSYS_GCL_CFG_REG_1_GATE_STATE_M                   GENMASK(15, 8)
-#define QSYS_GCL_CFG_REG_1_GATE_STATE_X(x)                (((x) & GENMASK(15, 8)) >> 8)
-
-#define QSYS_PARAM_STATUS_REG_3_BASE_TIME_SEC_MSB(x)      ((x) & GENMASK(15, 0))
-#define QSYS_PARAM_STATUS_REG_3_BASE_TIME_SEC_MSB_M       GENMASK(15, 0)
-#define QSYS_PARAM_STATUS_REG_3_LIST_LENGTH(x)            (((x) << 16) & GENMASK(31, 16))
-#define QSYS_PARAM_STATUS_REG_3_LIST_LENGTH_M             GENMASK(31, 16)
-#define QSYS_PARAM_STATUS_REG_3_LIST_LENGTH_X(x)          (((x) & GENMASK(31, 16)) >> 16)
-
-#define QSYS_PARAM_STATUS_REG_8_CFG_CHG_TIME_SEC_MSB(x)   ((x) & GENMASK(15, 0))
-#define QSYS_PARAM_STATUS_REG_8_CFG_CHG_TIME_SEC_MSB_M    GENMASK(15, 0)
-#define QSYS_PARAM_STATUS_REG_8_OPER_GATE_STATE(x)        (((x) << 16) & GENMASK(23, 16))
-#define QSYS_PARAM_STATUS_REG_8_OPER_GATE_STATE_M         GENMASK(23, 16)
-#define QSYS_PARAM_STATUS_REG_8_OPER_GATE_STATE_X(x)      (((x) & GENMASK(23, 16)) >> 16)
-#define QSYS_PARAM_STATUS_REG_8_CONFIG_PENDING            BIT(24)
-
-#define QSYS_GCL_STATUS_REG_1_GCL_ENTRY_NUM(x)            ((x) & GENMASK(5, 0))
-#define QSYS_GCL_STATUS_REG_1_GCL_ENTRY_NUM_M             GENMASK(5, 0)
-#define QSYS_GCL_STATUS_REG_1_GATE_STATE(x)               (((x) << 8) & GENMASK(15, 8))
-#define QSYS_GCL_STATUS_REG_1_GATE_STATE_M                GENMASK(15, 8)
-#define QSYS_GCL_STATUS_REG_1_GATE_STATE_X(x)             (((x) & GENMASK(15, 8)) >> 8)
-
-#endif
diff --git a/include/soc/mscc/ocelot_ana.h b/include/soc/mscc/ocelot_ana.h
new file mode 100644
index 000000000000..841c6ec22b64
--- /dev/null
+++ b/include/soc/mscc/ocelot_ana.h
@@ -0,0 +1,625 @@
+/* SPDX-License-Identifier: (GPL-2.0 OR MIT) */
+/*
+ * Microsemi Ocelot Switch driver
+ *
+ * Copyright (c) 2017 Microsemi Corporation
+ */
+
+#ifndef _MSCC_OCELOT_ANA_H_
+#define _MSCC_OCELOT_ANA_H_
+
+#define ANA_ANAGEFIL_B_DOM_EN                             BIT(22)
+#define ANA_ANAGEFIL_B_DOM_VAL                            BIT(21)
+#define ANA_ANAGEFIL_AGE_LOCKED                           BIT(20)
+#define ANA_ANAGEFIL_PID_EN                               BIT(19)
+#define ANA_ANAGEFIL_PID_VAL(x)                           (((x) << 14) & GENMASK(18, 14))
+#define ANA_ANAGEFIL_PID_VAL_M                            GENMASK(18, 14)
+#define ANA_ANAGEFIL_PID_VAL_X(x)                         (((x) & GENMASK(18, 14)) >> 14)
+#define ANA_ANAGEFIL_VID_EN                               BIT(13)
+#define ANA_ANAGEFIL_VID_VAL(x)                           ((x) & GENMASK(12, 0))
+#define ANA_ANAGEFIL_VID_VAL_M                            GENMASK(12, 0)
+
+#define ANA_STORMLIMIT_CFG_RSZ                            0x4
+
+#define ANA_STORMLIMIT_CFG_STORM_RATE(x)                  (((x) << 3) & GENMASK(6, 3))
+#define ANA_STORMLIMIT_CFG_STORM_RATE_M                   GENMASK(6, 3)
+#define ANA_STORMLIMIT_CFG_STORM_RATE_X(x)                (((x) & GENMASK(6, 3)) >> 3)
+#define ANA_STORMLIMIT_CFG_STORM_UNIT                     BIT(2)
+#define ANA_STORMLIMIT_CFG_STORM_MODE(x)                  ((x) & GENMASK(1, 0))
+#define ANA_STORMLIMIT_CFG_STORM_MODE_M                   GENMASK(1, 0)
+
+#define ANA_AUTOAGE_AGE_FAST                              BIT(21)
+#define ANA_AUTOAGE_AGE_PERIOD(x)                         (((x) << 1) & GENMASK(20, 1))
+#define ANA_AUTOAGE_AGE_PERIOD_M                          GENMASK(20, 1)
+#define ANA_AUTOAGE_AGE_PERIOD_X(x)                       (((x) & GENMASK(20, 1)) >> 1)
+#define ANA_AUTOAGE_AUTOAGE_LOCKED                        BIT(0)
+
+#define ANA_MACTOPTIONS_REDUCED_TABLE                     BIT(1)
+#define ANA_MACTOPTIONS_SHADOW                            BIT(0)
+
+#define ANA_AGENCTRL_FID_MASK(x)                          (((x) << 12) & GENMASK(23, 12))
+#define ANA_AGENCTRL_FID_MASK_M                           GENMASK(23, 12)
+#define ANA_AGENCTRL_FID_MASK_X(x)                        (((x) & GENMASK(23, 12)) >> 12)
+#define ANA_AGENCTRL_IGNORE_DMAC_FLAGS                    BIT(11)
+#define ANA_AGENCTRL_IGNORE_SMAC_FLAGS                    BIT(10)
+#define ANA_AGENCTRL_FLOOD_SPECIAL                        BIT(9)
+#define ANA_AGENCTRL_FLOOD_IGNORE_VLAN                    BIT(8)
+#define ANA_AGENCTRL_MIRROR_CPU                           BIT(7)
+#define ANA_AGENCTRL_LEARN_CPU_COPY                       BIT(6)
+#define ANA_AGENCTRL_LEARN_FWD_KILL                       BIT(5)
+#define ANA_AGENCTRL_LEARN_IGNORE_VLAN                    BIT(4)
+#define ANA_AGENCTRL_CPU_CPU_KILL_ENA                     BIT(3)
+#define ANA_AGENCTRL_GREEN_COUNT_MODE                     BIT(2)
+#define ANA_AGENCTRL_YELLOW_COUNT_MODE                    BIT(1)
+#define ANA_AGENCTRL_RED_COUNT_MODE                       BIT(0)
+
+#define ANA_FLOODING_RSZ                                  0x4
+
+#define ANA_FLOODING_FLD_UNICAST(x)                       (((x) << 12) & GENMASK(17, 12))
+#define ANA_FLOODING_FLD_UNICAST_M                        GENMASK(17, 12)
+#define ANA_FLOODING_FLD_UNICAST_X(x)                     (((x) & GENMASK(17, 12)) >> 12)
+#define ANA_FLOODING_FLD_BROADCAST(x)                     (((x) << 6) & GENMASK(11, 6))
+#define ANA_FLOODING_FLD_BROADCAST_M                      GENMASK(11, 6)
+#define ANA_FLOODING_FLD_BROADCAST_X(x)                   (((x) & GENMASK(11, 6)) >> 6)
+#define ANA_FLOODING_FLD_MULTICAST(x)                     ((x) & GENMASK(5, 0))
+#define ANA_FLOODING_FLD_MULTICAST_M                      GENMASK(5, 0)
+
+#define ANA_FLOODING_IPMC_FLD_MC4_CTRL(x)                 (((x) << 18) & GENMASK(23, 18))
+#define ANA_FLOODING_IPMC_FLD_MC4_CTRL_M                  GENMASK(23, 18)
+#define ANA_FLOODING_IPMC_FLD_MC4_CTRL_X(x)               (((x) & GENMASK(23, 18)) >> 18)
+#define ANA_FLOODING_IPMC_FLD_MC4_DATA(x)                 (((x) << 12) & GENMASK(17, 12))
+#define ANA_FLOODING_IPMC_FLD_MC4_DATA_M                  GENMASK(17, 12)
+#define ANA_FLOODING_IPMC_FLD_MC4_DATA_X(x)               (((x) & GENMASK(17, 12)) >> 12)
+#define ANA_FLOODING_IPMC_FLD_MC6_CTRL(x)                 (((x) << 6) & GENMASK(11, 6))
+#define ANA_FLOODING_IPMC_FLD_MC6_CTRL_M                  GENMASK(11, 6)
+#define ANA_FLOODING_IPMC_FLD_MC6_CTRL_X(x)               (((x) & GENMASK(11, 6)) >> 6)
+#define ANA_FLOODING_IPMC_FLD_MC6_DATA(x)                 ((x) & GENMASK(5, 0))
+#define ANA_FLOODING_IPMC_FLD_MC6_DATA_M                  GENMASK(5, 0)
+
+#define ANA_SFLOW_CFG_RSZ                                 0x4
+
+#define ANA_SFLOW_CFG_SF_RATE(x)                          (((x) << 2) & GENMASK(13, 2))
+#define ANA_SFLOW_CFG_SF_RATE_M                           GENMASK(13, 2)
+#define ANA_SFLOW_CFG_SF_RATE_X(x)                        (((x) & GENMASK(13, 2)) >> 2)
+#define ANA_SFLOW_CFG_SF_SAMPLE_RX                        BIT(1)
+#define ANA_SFLOW_CFG_SF_SAMPLE_TX                        BIT(0)
+
+#define ANA_PORT_MODE_RSZ                                 0x4
+
+#define ANA_PORT_MODE_REDTAG_PARSE_CFG                    BIT(3)
+#define ANA_PORT_MODE_VLAN_PARSE_CFG(x)                   (((x) << 1) & GENMASK(2, 1))
+#define ANA_PORT_MODE_VLAN_PARSE_CFG_M                    GENMASK(2, 1)
+#define ANA_PORT_MODE_VLAN_PARSE_CFG_X(x)                 (((x) & GENMASK(2, 1)) >> 1)
+#define ANA_PORT_MODE_L3_PARSE_CFG                        BIT(0)
+
+#define ANA_CUT_THRU_CFG_RSZ                              0x4
+
+#define ANA_PGID_PGID_RSZ                                 0x4
+
+#define ANA_PGID_PGID_PGID(x)                             ((x) & GENMASK(11, 0))
+#define ANA_PGID_PGID_PGID_M                              GENMASK(11, 0)
+#define ANA_PGID_PGID_CPUQ_DST_PGID(x)                    (((x) << 27) & GENMASK(29, 27))
+#define ANA_PGID_PGID_CPUQ_DST_PGID_M                     GENMASK(29, 27)
+#define ANA_PGID_PGID_CPUQ_DST_PGID_X(x)                  (((x) & GENMASK(29, 27)) >> 27)
+
+#define ANA_TABLES_MACHDATA_VID(x)                        (((x) << 16) & GENMASK(28, 16))
+#define ANA_TABLES_MACHDATA_VID_M                         GENMASK(28, 16)
+#define ANA_TABLES_MACHDATA_VID_X(x)                      (((x) & GENMASK(28, 16)) >> 16)
+#define ANA_TABLES_MACHDATA_MACHDATA(x)                   ((x) & GENMASK(15, 0))
+#define ANA_TABLES_MACHDATA_MACHDATA_M                    GENMASK(15, 0)
+
+#define ANA_TABLES_STREAMDATA_SSID_VALID                  BIT(16)
+#define ANA_TABLES_STREAMDATA_SSID(x)                     (((x) << 9) & GENMASK(15, 9))
+#define ANA_TABLES_STREAMDATA_SSID_M                      GENMASK(15, 9)
+#define ANA_TABLES_STREAMDATA_SSID_X(x)                   (((x) & GENMASK(15, 9)) >> 9)
+#define ANA_TABLES_STREAMDATA_SFID_VALID                  BIT(8)
+#define ANA_TABLES_STREAMDATA_SFID(x)                     ((x) & GENMASK(7, 0))
+#define ANA_TABLES_STREAMDATA_SFID_M                      GENMASK(7, 0)
+
+#define ANA_TABLES_MACACCESS_MAC_CPU_COPY                 BIT(15)
+#define ANA_TABLES_MACACCESS_SRC_KILL                     BIT(14)
+#define ANA_TABLES_MACACCESS_IGNORE_VLAN                  BIT(13)
+#define ANA_TABLES_MACACCESS_AGED_FLAG                    BIT(12)
+#define ANA_TABLES_MACACCESS_VALID                        BIT(11)
+#define ANA_TABLES_MACACCESS_ENTRYTYPE(x)                 (((x) << 9) & GENMASK(10, 9))
+#define ANA_TABLES_MACACCESS_ENTRYTYPE_M                  GENMASK(10, 9)
+#define ANA_TABLES_MACACCESS_ENTRYTYPE_X(x)               (((x) & GENMASK(10, 9)) >> 9)
+#define ANA_TABLES_MACACCESS_DEST_IDX(x)                  (((x) << 3) & GENMASK(8, 3))
+#define ANA_TABLES_MACACCESS_DEST_IDX_M                   GENMASK(8, 3)
+#define ANA_TABLES_MACACCESS_DEST_IDX_X(x)                (((x) & GENMASK(8, 3)) >> 3)
+#define ANA_TABLES_MACACCESS_MAC_TABLE_CMD(x)             ((x) & GENMASK(2, 0))
+#define ANA_TABLES_MACACCESS_MAC_TABLE_CMD_M              GENMASK(2, 0)
+#define MACACCESS_CMD_IDLE                     0
+#define MACACCESS_CMD_LEARN                    1
+#define MACACCESS_CMD_FORGET                   2
+#define MACACCESS_CMD_AGE                      3
+#define MACACCESS_CMD_GET_NEXT                 4
+#define MACACCESS_CMD_INIT                     5
+#define MACACCESS_CMD_READ                     6
+#define MACACCESS_CMD_WRITE                    7
+
+#define ANA_TABLES_VLANACCESS_VLAN_PORT_MASK(x)           (((x) << 2) & GENMASK(13, 2))
+#define ANA_TABLES_VLANACCESS_VLAN_PORT_MASK_M            GENMASK(13, 2)
+#define ANA_TABLES_VLANACCESS_VLAN_PORT_MASK_X(x)         (((x) & GENMASK(13, 2)) >> 2)
+#define ANA_TABLES_VLANACCESS_VLAN_TBL_CMD(x)             ((x) & GENMASK(1, 0))
+#define ANA_TABLES_VLANACCESS_VLAN_TBL_CMD_M              GENMASK(1, 0)
+#define ANA_TABLES_VLANACCESS_CMD_IDLE                    0x0
+#define ANA_TABLES_VLANACCESS_CMD_WRITE                   0x2
+#define ANA_TABLES_VLANACCESS_CMD_INIT                    0x3
+
+#define ANA_TABLES_VLANTIDX_VLAN_SEC_FWD_ENA              BIT(17)
+#define ANA_TABLES_VLANTIDX_VLAN_FLOOD_DIS                BIT(16)
+#define ANA_TABLES_VLANTIDX_VLAN_PRIV_VLAN                BIT(15)
+#define ANA_TABLES_VLANTIDX_VLAN_LEARN_DISABLED           BIT(14)
+#define ANA_TABLES_VLANTIDX_VLAN_MIRROR                   BIT(13)
+#define ANA_TABLES_VLANTIDX_VLAN_SRC_CHK                  BIT(12)
+#define ANA_TABLES_VLANTIDX_V_INDEX(x)                    ((x) & GENMASK(11, 0))
+#define ANA_TABLES_VLANTIDX_V_INDEX_M                     GENMASK(11, 0)
+
+#define ANA_TABLES_ISDXACCESS_ISDX_PORT_MASK(x)           (((x) << 2) & GENMASK(8, 2))
+#define ANA_TABLES_ISDXACCESS_ISDX_PORT_MASK_M            GENMASK(8, 2)
+#define ANA_TABLES_ISDXACCESS_ISDX_PORT_MASK_X(x)         (((x) & GENMASK(8, 2)) >> 2)
+#define ANA_TABLES_ISDXACCESS_ISDX_TBL_CMD(x)             ((x) & GENMASK(1, 0))
+#define ANA_TABLES_ISDXACCESS_ISDX_TBL_CMD_M              GENMASK(1, 0)
+
+#define ANA_TABLES_ISDXTIDX_ISDX_SDLBI(x)                 (((x) << 21) & GENMASK(28, 21))
+#define ANA_TABLES_ISDXTIDX_ISDX_SDLBI_M                  GENMASK(28, 21)
+#define ANA_TABLES_ISDXTIDX_ISDX_SDLBI_X(x)               (((x) & GENMASK(28, 21)) >> 21)
+#define ANA_TABLES_ISDXTIDX_ISDX_MSTI(x)                  (((x) << 15) & GENMASK(20, 15))
+#define ANA_TABLES_ISDXTIDX_ISDX_MSTI_M                   GENMASK(20, 15)
+#define ANA_TABLES_ISDXTIDX_ISDX_MSTI_X(x)                (((x) & GENMASK(20, 15)) >> 15)
+#define ANA_TABLES_ISDXTIDX_ISDX_ES0_KEY_ENA              BIT(14)
+#define ANA_TABLES_ISDXTIDX_ISDX_FORCE_ENA                BIT(10)
+#define ANA_TABLES_ISDXTIDX_ISDX_INDEX(x)                 ((x) & GENMASK(7, 0))
+#define ANA_TABLES_ISDXTIDX_ISDX_INDEX_M                  GENMASK(7, 0)
+
+#define ANA_TABLES_ENTRYLIM_RSZ                           0x4
+
+#define ANA_TABLES_ENTRYLIM_ENTRYLIM(x)                   (((x) << 14) & GENMASK(17, 14))
+#define ANA_TABLES_ENTRYLIM_ENTRYLIM_M                    GENMASK(17, 14)
+#define ANA_TABLES_ENTRYLIM_ENTRYLIM_X(x)                 (((x) & GENMASK(17, 14)) >> 14)
+#define ANA_TABLES_ENTRYLIM_ENTRYSTAT(x)                  ((x) & GENMASK(13, 0))
+#define ANA_TABLES_ENTRYLIM_ENTRYSTAT_M                   GENMASK(13, 0)
+
+#define ANA_TABLES_STREAMACCESS_GEN_REC_SEQ_NUM(x)        (((x) << 4) & GENMASK(31, 4))
+#define ANA_TABLES_STREAMACCESS_GEN_REC_SEQ_NUM_M         GENMASK(31, 4)
+#define ANA_TABLES_STREAMACCESS_GEN_REC_SEQ_NUM_X(x)      (((x) & GENMASK(31, 4)) >> 4)
+#define ANA_TABLES_STREAMACCESS_SEQ_GEN_REC_ENA           BIT(3)
+#define ANA_TABLES_STREAMACCESS_GEN_REC_TYPE              BIT(2)
+#define ANA_TABLES_STREAMACCESS_STREAM_TBL_CMD(x)         ((x) & GENMASK(1, 0))
+#define ANA_TABLES_STREAMACCESS_STREAM_TBL_CMD_M          GENMASK(1, 0)
+
+#define ANA_TABLES_STREAMTIDX_SEQ_GEN_ERR_STATUS(x)       (((x) << 30) & GENMASK(31, 30))
+#define ANA_TABLES_STREAMTIDX_SEQ_GEN_ERR_STATUS_M        GENMASK(31, 30)
+#define ANA_TABLES_STREAMTIDX_SEQ_GEN_ERR_STATUS_X(x)     (((x) & GENMASK(31, 30)) >> 30)
+#define ANA_TABLES_STREAMTIDX_S_INDEX(x)                  (((x) << 16) & GENMASK(22, 16))
+#define ANA_TABLES_STREAMTIDX_S_INDEX_M                   GENMASK(22, 16)
+#define ANA_TABLES_STREAMTIDX_S_INDEX_X(x)                (((x) & GENMASK(22, 16)) >> 16)
+#define ANA_TABLES_STREAMTIDX_FORCE_SF_BEHAVIOUR          BIT(14)
+#define ANA_TABLES_STREAMTIDX_SEQ_HISTORY_LEN(x)          (((x) << 8) & GENMASK(13, 8))
+#define ANA_TABLES_STREAMTIDX_SEQ_HISTORY_LEN_M           GENMASK(13, 8)
+#define ANA_TABLES_STREAMTIDX_SEQ_HISTORY_LEN_X(x)        (((x) & GENMASK(13, 8)) >> 8)
+#define ANA_TABLES_STREAMTIDX_RESET_ON_ROGUE              BIT(7)
+#define ANA_TABLES_STREAMTIDX_REDTAG_POP                  BIT(6)
+#define ANA_TABLES_STREAMTIDX_STREAM_SPLIT                BIT(5)
+#define ANA_TABLES_STREAMTIDX_SEQ_SPACE_LOG2(x)           ((x) & GENMASK(4, 0))
+#define ANA_TABLES_STREAMTIDX_SEQ_SPACE_LOG2_M            GENMASK(4, 0)
+
+#define ANA_TABLES_SEQ_MASK_SPLIT_MASK(x)                 (((x) << 16) & GENMASK(22, 16))
+#define ANA_TABLES_SEQ_MASK_SPLIT_MASK_M                  GENMASK(22, 16)
+#define ANA_TABLES_SEQ_MASK_SPLIT_MASK_X(x)               (((x) & GENMASK(22, 16)) >> 16)
+#define ANA_TABLES_SEQ_MASK_INPUT_PORT_MASK(x)            ((x) & GENMASK(6, 0))
+#define ANA_TABLES_SEQ_MASK_INPUT_PORT_MASK_M             GENMASK(6, 0)
+
+#define ANA_TABLES_SFID_MASK_IGR_PORT_MASK(x)             (((x) << 1) & GENMASK(7, 1))
+#define ANA_TABLES_SFID_MASK_IGR_PORT_MASK_M              GENMASK(7, 1)
+#define ANA_TABLES_SFID_MASK_IGR_PORT_MASK_X(x)           (((x) & GENMASK(7, 1)) >> 1)
+#define ANA_TABLES_SFID_MASK_IGR_SRCPORT_MATCH_ENA        BIT(0)
+
+#define ANA_TABLES_SFIDACCESS_IGR_PRIO_MATCH_ENA          BIT(22)
+#define ANA_TABLES_SFIDACCESS_IGR_PRIO(x)                 (((x) << 19) & GENMASK(21, 19))
+#define ANA_TABLES_SFIDACCESS_IGR_PRIO_M                  GENMASK(21, 19)
+#define ANA_TABLES_SFIDACCESS_IGR_PRIO_X(x)               (((x) & GENMASK(21, 19)) >> 19)
+#define ANA_TABLES_SFIDACCESS_FORCE_BLOCK                 BIT(18)
+#define ANA_TABLES_SFIDACCESS_MAX_SDU_LEN(x)              (((x) << 2) & GENMASK(17, 2))
+#define ANA_TABLES_SFIDACCESS_MAX_SDU_LEN_M               GENMASK(17, 2)
+#define ANA_TABLES_SFIDACCESS_MAX_SDU_LEN_X(x)            (((x) & GENMASK(17, 2)) >> 2)
+#define ANA_TABLES_SFIDACCESS_SFID_TBL_CMD(x)             ((x) & GENMASK(1, 0))
+#define ANA_TABLES_SFIDACCESS_SFID_TBL_CMD_M              GENMASK(1, 0)
+
+#define ANA_TABLES_SFIDTIDX_SGID_VALID                    BIT(26)
+#define ANA_TABLES_SFIDTIDX_SGID(x)                       (((x) << 18) & GENMASK(25, 18))
+#define ANA_TABLES_SFIDTIDX_SGID_M                        GENMASK(25, 18)
+#define ANA_TABLES_SFIDTIDX_SGID_X(x)                     (((x) & GENMASK(25, 18)) >> 18)
+#define ANA_TABLES_SFIDTIDX_POL_ENA                       BIT(17)
+#define ANA_TABLES_SFIDTIDX_POL_IDX(x)                    (((x) << 8) & GENMASK(16, 8))
+#define ANA_TABLES_SFIDTIDX_POL_IDX_M                     GENMASK(16, 8)
+#define ANA_TABLES_SFIDTIDX_POL_IDX_X(x)                  (((x) & GENMASK(16, 8)) >> 8)
+#define ANA_TABLES_SFIDTIDX_SFID_INDEX(x)                 ((x) & GENMASK(7, 0))
+#define ANA_TABLES_SFIDTIDX_SFID_INDEX_M                  GENMASK(7, 0)
+
+#define ANA_MSTI_STATE_RSZ                                0x4
+
+#define ANA_OAM_UPM_LM_CNT_RSZ                            0x4
+
+#define ANA_SG_ACCESS_CTRL_SGID(x)                        ((x) & GENMASK(7, 0))
+#define ANA_SG_ACCESS_CTRL_SGID_M                         GENMASK(7, 0)
+#define ANA_SG_ACCESS_CTRL_CONFIG_CHANGE                  BIT(28)
+
+#define ANA_SG_CONFIG_REG_3_BASE_TIME_SEC_MSB(x)          ((x) & GENMASK(15, 0))
+#define ANA_SG_CONFIG_REG_3_BASE_TIME_SEC_MSB_M           GENMASK(15, 0)
+#define ANA_SG_CONFIG_REG_3_LIST_LENGTH(x)                (((x) << 16) & GENMASK(18, 16))
+#define ANA_SG_CONFIG_REG_3_LIST_LENGTH_M                 GENMASK(18, 16)
+#define ANA_SG_CONFIG_REG_3_LIST_LENGTH_X(x)              (((x) & GENMASK(18, 16)) >> 16)
+#define ANA_SG_CONFIG_REG_3_GATE_ENABLE                   BIT(20)
+#define ANA_SG_CONFIG_REG_3_INIT_IPS(x)                   (((x) << 24) & GENMASK(27, 24))
+#define ANA_SG_CONFIG_REG_3_INIT_IPS_M                    GENMASK(27, 24)
+#define ANA_SG_CONFIG_REG_3_INIT_IPS_X(x)                 (((x) & GENMASK(27, 24)) >> 24)
+#define ANA_SG_CONFIG_REG_3_INIT_GATE_STATE               BIT(28)
+
+#define ANA_SG_GCL_GS_CONFIG_RSZ                          0x4
+
+#define ANA_SG_GCL_GS_CONFIG_IPS(x)                       ((x) & GENMASK(3, 0))
+#define ANA_SG_GCL_GS_CONFIG_IPS_M                        GENMASK(3, 0)
+#define ANA_SG_GCL_GS_CONFIG_GATE_STATE                   BIT(4)
+
+#define ANA_SG_GCL_TI_CONFIG_RSZ                          0x4
+
+#define ANA_SG_STATUS_REG_3_CFG_CHG_TIME_SEC_MSB(x)       ((x) & GENMASK(15, 0))
+#define ANA_SG_STATUS_REG_3_CFG_CHG_TIME_SEC_MSB_M        GENMASK(15, 0)
+#define ANA_SG_STATUS_REG_3_GATE_STATE                    BIT(16)
+#define ANA_SG_STATUS_REG_3_IPS(x)                        (((x) << 20) & GENMASK(23, 20))
+#define ANA_SG_STATUS_REG_3_IPS_M                         GENMASK(23, 20)
+#define ANA_SG_STATUS_REG_3_IPS_X(x)                      (((x) & GENMASK(23, 20)) >> 20)
+#define ANA_SG_STATUS_REG_3_CONFIG_PENDING                BIT(24)
+
+#define ANA_PORT_VLAN_CFG_GSZ                             0x100
+
+#define ANA_PORT_VLAN_CFG_VLAN_VID_AS_ISDX                BIT(21)
+#define ANA_PORT_VLAN_CFG_VLAN_AWARE_ENA                  BIT(20)
+#define ANA_PORT_VLAN_CFG_VLAN_POP_CNT(x)                 (((x) << 18) & GENMASK(19, 18))
+#define ANA_PORT_VLAN_CFG_VLAN_POP_CNT_M                  GENMASK(19, 18)
+#define ANA_PORT_VLAN_CFG_VLAN_POP_CNT_X(x)               (((x) & GENMASK(19, 18)) >> 18)
+#define ANA_PORT_VLAN_CFG_VLAN_INNER_TAG_ENA              BIT(17)
+#define ANA_PORT_VLAN_CFG_VLAN_TAG_TYPE                   BIT(16)
+#define ANA_PORT_VLAN_CFG_VLAN_DEI                        BIT(15)
+#define ANA_PORT_VLAN_CFG_VLAN_PCP(x)                     (((x) << 12) & GENMASK(14, 12))
+#define ANA_PORT_VLAN_CFG_VLAN_PCP_M                      GENMASK(14, 12)
+#define ANA_PORT_VLAN_CFG_VLAN_PCP_X(x)                   (((x) & GENMASK(14, 12)) >> 12)
+#define ANA_PORT_VLAN_CFG_VLAN_VID(x)                     ((x) & GENMASK(11, 0))
+#define ANA_PORT_VLAN_CFG_VLAN_VID_M                      GENMASK(11, 0)
+
+#define ANA_PORT_DROP_CFG_GSZ                             0x100
+
+#define ANA_PORT_DROP_CFG_DROP_UNTAGGED_ENA               BIT(6)
+#define ANA_PORT_DROP_CFG_DROP_S_TAGGED_ENA               BIT(5)
+#define ANA_PORT_DROP_CFG_DROP_C_TAGGED_ENA               BIT(4)
+#define ANA_PORT_DROP_CFG_DROP_PRIO_S_TAGGED_ENA          BIT(3)
+#define ANA_PORT_DROP_CFG_DROP_PRIO_C_TAGGED_ENA          BIT(2)
+#define ANA_PORT_DROP_CFG_DROP_NULL_MAC_ENA               BIT(1)
+#define ANA_PORT_DROP_CFG_DROP_MC_SMAC_ENA                BIT(0)
+
+#define ANA_PORT_QOS_CFG_GSZ                              0x100
+
+#define ANA_PORT_QOS_CFG_DP_DEFAULT_VAL                   BIT(8)
+#define ANA_PORT_QOS_CFG_QOS_DEFAULT_VAL(x)               (((x) << 5) & GENMASK(7, 5))
+#define ANA_PORT_QOS_CFG_QOS_DEFAULT_VAL_M                GENMASK(7, 5)
+#define ANA_PORT_QOS_CFG_QOS_DEFAULT_VAL_X(x)             (((x) & GENMASK(7, 5)) >> 5)
+#define ANA_PORT_QOS_CFG_QOS_DSCP_ENA                     BIT(4)
+#define ANA_PORT_QOS_CFG_QOS_PCP_ENA                      BIT(3)
+#define ANA_PORT_QOS_CFG_DSCP_TRANSLATE_ENA               BIT(2)
+#define ANA_PORT_QOS_CFG_DSCP_REWR_CFG(x)                 ((x) & GENMASK(1, 0))
+#define ANA_PORT_QOS_CFG_DSCP_REWR_CFG_M                  GENMASK(1, 0)
+
+#define ANA_PORT_VCAP_CFG_GSZ                             0x100
+
+#define ANA_PORT_VCAP_CFG_S1_ENA                          BIT(14)
+#define ANA_PORT_VCAP_CFG_S1_DMAC_DIP_ENA(x)              (((x) << 11) & GENMASK(13, 11))
+#define ANA_PORT_VCAP_CFG_S1_DMAC_DIP_ENA_M               GENMASK(13, 11)
+#define ANA_PORT_VCAP_CFG_S1_DMAC_DIP_ENA_X(x)            (((x) & GENMASK(13, 11)) >> 11)
+#define ANA_PORT_VCAP_CFG_S1_VLAN_INNER_TAG_ENA(x)        (((x) << 8) & GENMASK(10, 8))
+#define ANA_PORT_VCAP_CFG_S1_VLAN_INNER_TAG_ENA_M         GENMASK(10, 8)
+#define ANA_PORT_VCAP_CFG_S1_VLAN_INNER_TAG_ENA_X(x)      (((x) & GENMASK(10, 8)) >> 8)
+#define ANA_PORT_VCAP_CFG_PAG_VAL(x)                      ((x) & GENMASK(7, 0))
+#define ANA_PORT_VCAP_CFG_PAG_VAL_M                       GENMASK(7, 0)
+
+#define ANA_PORT_VCAP_S1_KEY_CFG_GSZ                      0x100
+#define ANA_PORT_VCAP_S1_KEY_CFG_RSZ                      0x4
+
+#define ANA_PORT_VCAP_S1_KEY_CFG_S1_KEY_IP6_CFG(x)        (((x) << 4) & GENMASK(6, 4))
+#define ANA_PORT_VCAP_S1_KEY_CFG_S1_KEY_IP6_CFG_M         GENMASK(6, 4)
+#define ANA_PORT_VCAP_S1_KEY_CFG_S1_KEY_IP6_CFG_X(x)      (((x) & GENMASK(6, 4)) >> 4)
+#define ANA_PORT_VCAP_S1_KEY_CFG_S1_KEY_IP4_CFG(x)        (((x) << 2) & GENMASK(3, 2))
+#define ANA_PORT_VCAP_S1_KEY_CFG_S1_KEY_IP4_CFG_M         GENMASK(3, 2)
+#define ANA_PORT_VCAP_S1_KEY_CFG_S1_KEY_IP4_CFG_X(x)      (((x) & GENMASK(3, 2)) >> 2)
+#define ANA_PORT_VCAP_S1_KEY_CFG_S1_KEY_OTHER_CFG(x)      ((x) & GENMASK(1, 0))
+#define ANA_PORT_VCAP_S1_KEY_CFG_S1_KEY_OTHER_CFG_M       GENMASK(1, 0)
+
+#define ANA_PORT_VCAP_S2_CFG_GSZ                          0x100
+
+#define ANA_PORT_VCAP_S2_CFG_S2_UDP_PAYLOAD_ENA(x)        (((x) << 17) & GENMASK(18, 17))
+#define ANA_PORT_VCAP_S2_CFG_S2_UDP_PAYLOAD_ENA_M         GENMASK(18, 17)
+#define ANA_PORT_VCAP_S2_CFG_S2_UDP_PAYLOAD_ENA_X(x)      (((x) & GENMASK(18, 17)) >> 17)
+#define ANA_PORT_VCAP_S2_CFG_S2_ETYPE_PAYLOAD_ENA(x)      (((x) << 15) & GENMASK(16, 15))
+#define ANA_PORT_VCAP_S2_CFG_S2_ETYPE_PAYLOAD_ENA_M       GENMASK(16, 15)
+#define ANA_PORT_VCAP_S2_CFG_S2_ETYPE_PAYLOAD_ENA_X(x)    (((x) & GENMASK(16, 15)) >> 15)
+#define ANA_PORT_VCAP_S2_CFG_S2_ENA                       BIT(14)
+#define ANA_PORT_VCAP_S2_CFG_S2_SNAP_DIS(x)               (((x) << 12) & GENMASK(13, 12))
+#define ANA_PORT_VCAP_S2_CFG_S2_SNAP_DIS_M                GENMASK(13, 12)
+#define ANA_PORT_VCAP_S2_CFG_S2_SNAP_DIS_X(x)             (((x) & GENMASK(13, 12)) >> 12)
+#define ANA_PORT_VCAP_S2_CFG_S2_ARP_DIS(x)                (((x) << 10) & GENMASK(11, 10))
+#define ANA_PORT_VCAP_S2_CFG_S2_ARP_DIS_M                 GENMASK(11, 10)
+#define ANA_PORT_VCAP_S2_CFG_S2_ARP_DIS_X(x)              (((x) & GENMASK(11, 10)) >> 10)
+#define ANA_PORT_VCAP_S2_CFG_S2_IP_TCPUDP_DIS(x)          (((x) << 8) & GENMASK(9, 8))
+#define ANA_PORT_VCAP_S2_CFG_S2_IP_TCPUDP_DIS_M           GENMASK(9, 8)
+#define ANA_PORT_VCAP_S2_CFG_S2_IP_TCPUDP_DIS_X(x)        (((x) & GENMASK(9, 8)) >> 8)
+#define ANA_PORT_VCAP_S2_CFG_S2_IP_OTHER_DIS(x)           (((x) << 6) & GENMASK(7, 6))
+#define ANA_PORT_VCAP_S2_CFG_S2_IP_OTHER_DIS_M            GENMASK(7, 6)
+#define ANA_PORT_VCAP_S2_CFG_S2_IP_OTHER_DIS_X(x)         (((x) & GENMASK(7, 6)) >> 6)
+#define ANA_PORT_VCAP_S2_CFG_S2_IP6_CFG(x)                (((x) << 2) & GENMASK(5, 2))
+#define ANA_PORT_VCAP_S2_CFG_S2_IP6_CFG_M                 GENMASK(5, 2)
+#define ANA_PORT_VCAP_S2_CFG_S2_IP6_CFG_X(x)              (((x) & GENMASK(5, 2)) >> 2)
+#define ANA_PORT_VCAP_S2_CFG_S2_OAM_DIS(x)                ((x) & GENMASK(1, 0))
+#define ANA_PORT_VCAP_S2_CFG_S2_OAM_DIS_M                 GENMASK(1, 0)
+
+#define ANA_PORT_PCP_DEI_MAP_GSZ                          0x100
+#define ANA_PORT_PCP_DEI_MAP_RSZ                          0x4
+
+#define ANA_PORT_PCP_DEI_MAP_DP_PCP_DEI_VAL               BIT(3)
+#define ANA_PORT_PCP_DEI_MAP_QOS_PCP_DEI_VAL(x)           ((x) & GENMASK(2, 0))
+#define ANA_PORT_PCP_DEI_MAP_QOS_PCP_DEI_VAL_M            GENMASK(2, 0)
+
+#define ANA_PORT_CPU_FWD_CFG_GSZ                          0x100
+
+#define ANA_PORT_CPU_FWD_CFG_CPU_VRAP_REDIR_ENA           BIT(7)
+#define ANA_PORT_CPU_FWD_CFG_CPU_MLD_REDIR_ENA            BIT(6)
+#define ANA_PORT_CPU_FWD_CFG_CPU_IGMP_REDIR_ENA           BIT(5)
+#define ANA_PORT_CPU_FWD_CFG_CPU_IPMC_CTRL_COPY_ENA       BIT(4)
+#define ANA_PORT_CPU_FWD_CFG_CPU_SRC_COPY_ENA             BIT(3)
+#define ANA_PORT_CPU_FWD_CFG_CPU_ALLBRIDGE_DROP_ENA       BIT(2)
+#define ANA_PORT_CPU_FWD_CFG_CPU_ALLBRIDGE_REDIR_ENA      BIT(1)
+#define ANA_PORT_CPU_FWD_CFG_CPU_OAM_ENA                  BIT(0)
+
+#define ANA_PORT_CPU_FWD_BPDU_CFG_GSZ                     0x100
+
+#define ANA_PORT_CPU_FWD_BPDU_CFG_BPDU_DROP_ENA(x)        (((x) << 16) & GENMASK(31, 16))
+#define ANA_PORT_CPU_FWD_BPDU_CFG_BPDU_DROP_ENA_M         GENMASK(31, 16)
+#define ANA_PORT_CPU_FWD_BPDU_CFG_BPDU_DROP_ENA_X(x)      (((x) & GENMASK(31, 16)) >> 16)
+#define ANA_PORT_CPU_FWD_BPDU_CFG_BPDU_REDIR_ENA(x)       ((x) & GENMASK(15, 0))
+#define ANA_PORT_CPU_FWD_BPDU_CFG_BPDU_REDIR_ENA_M        GENMASK(15, 0)
+
+#define ANA_PORT_CPU_FWD_GARP_CFG_GSZ                     0x100
+
+#define ANA_PORT_CPU_FWD_GARP_CFG_GARP_DROP_ENA(x)        (((x) << 16) & GENMASK(31, 16))
+#define ANA_PORT_CPU_FWD_GARP_CFG_GARP_DROP_ENA_M         GENMASK(31, 16)
+#define ANA_PORT_CPU_FWD_GARP_CFG_GARP_DROP_ENA_X(x)      (((x) & GENMASK(31, 16)) >> 16)
+#define ANA_PORT_CPU_FWD_GARP_CFG_GARP_REDIR_ENA(x)       ((x) & GENMASK(15, 0))
+#define ANA_PORT_CPU_FWD_GARP_CFG_GARP_REDIR_ENA_M        GENMASK(15, 0)
+
+#define ANA_PORT_CPU_FWD_CCM_CFG_GSZ                      0x100
+
+#define ANA_PORT_CPU_FWD_CCM_CFG_CCM_DROP_ENA(x)          (((x) << 16) & GENMASK(31, 16))
+#define ANA_PORT_CPU_FWD_CCM_CFG_CCM_DROP_ENA_M           GENMASK(31, 16)
+#define ANA_PORT_CPU_FWD_CCM_CFG_CCM_DROP_ENA_X(x)        (((x) & GENMASK(31, 16)) >> 16)
+#define ANA_PORT_CPU_FWD_CCM_CFG_CCM_REDIR_ENA(x)         ((x) & GENMASK(15, 0))
+#define ANA_PORT_CPU_FWD_CCM_CFG_CCM_REDIR_ENA_M          GENMASK(15, 0)
+
+#define ANA_PORT_PORT_CFG_GSZ                             0x100
+
+#define ANA_PORT_PORT_CFG_SRC_MIRROR_ENA                  BIT(15)
+#define ANA_PORT_PORT_CFG_LIMIT_DROP                      BIT(14)
+#define ANA_PORT_PORT_CFG_LIMIT_CPU                       BIT(13)
+#define ANA_PORT_PORT_CFG_LOCKED_PORTMOVE_DROP            BIT(12)
+#define ANA_PORT_PORT_CFG_LOCKED_PORTMOVE_CPU             BIT(11)
+#define ANA_PORT_PORT_CFG_LEARNDROP                       BIT(10)
+#define ANA_PORT_PORT_CFG_LEARNCPU                        BIT(9)
+#define ANA_PORT_PORT_CFG_LEARNAUTO                       BIT(8)
+#define ANA_PORT_PORT_CFG_LEARN_ENA                       BIT(7)
+#define ANA_PORT_PORT_CFG_RECV_ENA                        BIT(6)
+#define ANA_PORT_PORT_CFG_PORTID_VAL(x)                   (((x) << 2) & GENMASK(5, 2))
+#define ANA_PORT_PORT_CFG_PORTID_VAL_M                    GENMASK(5, 2)
+#define ANA_PORT_PORT_CFG_PORTID_VAL_X(x)                 (((x) & GENMASK(5, 2)) >> 2)
+#define ANA_PORT_PORT_CFG_USE_B_DOM_TBL                   BIT(1)
+#define ANA_PORT_PORT_CFG_LSR_MODE                        BIT(0)
+
+#define ANA_PORT_POL_CFG_GSZ                              0x100
+
+#define ANA_PORT_POL_CFG_POL_CPU_REDIR_8021               BIT(19)
+#define ANA_PORT_POL_CFG_POL_CPU_REDIR_IP                 BIT(18)
+#define ANA_PORT_POL_CFG_PORT_POL_ENA                     BIT(17)
+#define ANA_PORT_POL_CFG_QUEUE_POL_ENA(x)                 (((x) << 9) & GENMASK(16, 9))
+#define ANA_PORT_POL_CFG_QUEUE_POL_ENA_M                  GENMASK(16, 9)
+#define ANA_PORT_POL_CFG_QUEUE_POL_ENA_X(x)               (((x) & GENMASK(16, 9)) >> 9)
+#define ANA_PORT_POL_CFG_POL_ORDER(x)                     ((x) & GENMASK(8, 0))
+#define ANA_PORT_POL_CFG_POL_ORDER_M                      GENMASK(8, 0)
+
+#define ANA_PORT_PTP_CFG_GSZ                              0x100
+
+#define ANA_PORT_PTP_CFG_PTP_BACKPLANE_MODE               BIT(0)
+
+#define ANA_PORT_PTP_DLY1_CFG_GSZ                         0x100
+
+#define ANA_PORT_PTP_DLY2_CFG_GSZ                         0x100
+
+#define ANA_PORT_SFID_CFG_GSZ                             0x100
+#define ANA_PORT_SFID_CFG_RSZ                             0x4
+
+#define ANA_PORT_SFID_CFG_SFID_VALID                      BIT(8)
+#define ANA_PORT_SFID_CFG_SFID(x)                         ((x) & GENMASK(7, 0))
+#define ANA_PORT_SFID_CFG_SFID_M                          GENMASK(7, 0)
+
+#define ANA_PFC_PFC_CFG_GSZ                               0x40
+
+#define ANA_PFC_PFC_CFG_RX_PFC_ENA(x)                     (((x) << 2) & GENMASK(9, 2))
+#define ANA_PFC_PFC_CFG_RX_PFC_ENA_M                      GENMASK(9, 2)
+#define ANA_PFC_PFC_CFG_RX_PFC_ENA_X(x)                   (((x) & GENMASK(9, 2)) >> 2)
+#define ANA_PFC_PFC_CFG_FC_LINK_SPEED(x)                  ((x) & GENMASK(1, 0))
+#define ANA_PFC_PFC_CFG_FC_LINK_SPEED_M                   GENMASK(1, 0)
+
+#define ANA_PFC_PFC_TIMER_GSZ                             0x40
+#define ANA_PFC_PFC_TIMER_RSZ                             0x4
+
+#define ANA_IPT_OAM_MEP_CFG_GSZ                           0x8
+
+#define ANA_IPT_OAM_MEP_CFG_MEP_IDX_P(x)                  (((x) << 6) & GENMASK(10, 6))
+#define ANA_IPT_OAM_MEP_CFG_MEP_IDX_P_M                   GENMASK(10, 6)
+#define ANA_IPT_OAM_MEP_CFG_MEP_IDX_P_X(x)                (((x) & GENMASK(10, 6)) >> 6)
+#define ANA_IPT_OAM_MEP_CFG_MEP_IDX(x)                    (((x) << 1) & GENMASK(5, 1))
+#define ANA_IPT_OAM_MEP_CFG_MEP_IDX_M                     GENMASK(5, 1)
+#define ANA_IPT_OAM_MEP_CFG_MEP_IDX_X(x)                  (((x) & GENMASK(5, 1)) >> 1)
+#define ANA_IPT_OAM_MEP_CFG_MEP_IDX_ENA                   BIT(0)
+
+#define ANA_IPT_IPT_GSZ                                   0x8
+
+#define ANA_IPT_IPT_IPT_CFG(x)                            (((x) << 15) & GENMASK(16, 15))
+#define ANA_IPT_IPT_IPT_CFG_M                             GENMASK(16, 15)
+#define ANA_IPT_IPT_IPT_CFG_X(x)                          (((x) & GENMASK(16, 15)) >> 15)
+#define ANA_IPT_IPT_ISDX_P(x)                             (((x) << 7) & GENMASK(14, 7))
+#define ANA_IPT_IPT_ISDX_P_M                              GENMASK(14, 7)
+#define ANA_IPT_IPT_ISDX_P_X(x)                           (((x) & GENMASK(14, 7)) >> 7)
+#define ANA_IPT_IPT_PPT_IDX(x)                            ((x) & GENMASK(6, 0))
+#define ANA_IPT_IPT_PPT_IDX_M                             GENMASK(6, 0)
+
+#define ANA_PPT_PPT_RSZ                                   0x4
+
+#define ANA_FID_MAP_FID_MAP_RSZ                           0x4
+
+#define ANA_FID_MAP_FID_MAP_FID_C_VAL(x)                  (((x) << 6) & GENMASK(11, 6))
+#define ANA_FID_MAP_FID_MAP_FID_C_VAL_M                   GENMASK(11, 6)
+#define ANA_FID_MAP_FID_MAP_FID_C_VAL_X(x)                (((x) & GENMASK(11, 6)) >> 6)
+#define ANA_FID_MAP_FID_MAP_FID_B_VAL(x)                  ((x) & GENMASK(5, 0))
+#define ANA_FID_MAP_FID_MAP_FID_B_VAL_M                   GENMASK(5, 0)
+
+#define ANA_AGGR_CFG_AC_RND_ENA                           BIT(7)
+#define ANA_AGGR_CFG_AC_DMAC_ENA                          BIT(6)
+#define ANA_AGGR_CFG_AC_SMAC_ENA                          BIT(5)
+#define ANA_AGGR_CFG_AC_IP6_FLOW_LBL_ENA                  BIT(4)
+#define ANA_AGGR_CFG_AC_IP6_TCPUDP_ENA                    BIT(3)
+#define ANA_AGGR_CFG_AC_IP4_SIPDIP_ENA                    BIT(2)
+#define ANA_AGGR_CFG_AC_IP4_TCPUDP_ENA                    BIT(1)
+#define ANA_AGGR_CFG_AC_ISDX_ENA                          BIT(0)
+
+#define ANA_CPUQ_CFG_CPUQ_MLD(x)                          (((x) << 27) & GENMASK(29, 27))
+#define ANA_CPUQ_CFG_CPUQ_MLD_M                           GENMASK(29, 27)
+#define ANA_CPUQ_CFG_CPUQ_MLD_X(x)                        (((x) & GENMASK(29, 27)) >> 27)
+#define ANA_CPUQ_CFG_CPUQ_IGMP(x)                         (((x) << 24) & GENMASK(26, 24))
+#define ANA_CPUQ_CFG_CPUQ_IGMP_M                          GENMASK(26, 24)
+#define ANA_CPUQ_CFG_CPUQ_IGMP_X(x)                       (((x) & GENMASK(26, 24)) >> 24)
+#define ANA_CPUQ_CFG_CPUQ_IPMC_CTRL(x)                    (((x) << 21) & GENMASK(23, 21))
+#define ANA_CPUQ_CFG_CPUQ_IPMC_CTRL_M                     GENMASK(23, 21)
+#define ANA_CPUQ_CFG_CPUQ_IPMC_CTRL_X(x)                  (((x) & GENMASK(23, 21)) >> 21)
+#define ANA_CPUQ_CFG_CPUQ_ALLBRIDGE(x)                    (((x) << 18) & GENMASK(20, 18))
+#define ANA_CPUQ_CFG_CPUQ_ALLBRIDGE_M                     GENMASK(20, 18)
+#define ANA_CPUQ_CFG_CPUQ_ALLBRIDGE_X(x)                  (((x) & GENMASK(20, 18)) >> 18)
+#define ANA_CPUQ_CFG_CPUQ_LOCKED_PORTMOVE(x)              (((x) << 15) & GENMASK(17, 15))
+#define ANA_CPUQ_CFG_CPUQ_LOCKED_PORTMOVE_M               GENMASK(17, 15)
+#define ANA_CPUQ_CFG_CPUQ_LOCKED_PORTMOVE_X(x)            (((x) & GENMASK(17, 15)) >> 15)
+#define ANA_CPUQ_CFG_CPUQ_SRC_COPY(x)                     (((x) << 12) & GENMASK(14, 12))
+#define ANA_CPUQ_CFG_CPUQ_SRC_COPY_M                      GENMASK(14, 12)
+#define ANA_CPUQ_CFG_CPUQ_SRC_COPY_X(x)                   (((x) & GENMASK(14, 12)) >> 12)
+#define ANA_CPUQ_CFG_CPUQ_MAC_COPY(x)                     (((x) << 9) & GENMASK(11, 9))
+#define ANA_CPUQ_CFG_CPUQ_MAC_COPY_M                      GENMASK(11, 9)
+#define ANA_CPUQ_CFG_CPUQ_MAC_COPY_X(x)                   (((x) & GENMASK(11, 9)) >> 9)
+#define ANA_CPUQ_CFG_CPUQ_LRN(x)                          (((x) << 6) & GENMASK(8, 6))
+#define ANA_CPUQ_CFG_CPUQ_LRN_M                           GENMASK(8, 6)
+#define ANA_CPUQ_CFG_CPUQ_LRN_X(x)                        (((x) & GENMASK(8, 6)) >> 6)
+#define ANA_CPUQ_CFG_CPUQ_MIRROR(x)                       (((x) << 3) & GENMASK(5, 3))
+#define ANA_CPUQ_CFG_CPUQ_MIRROR_M                        GENMASK(5, 3)
+#define ANA_CPUQ_CFG_CPUQ_MIRROR_X(x)                     (((x) & GENMASK(5, 3)) >> 3)
+#define ANA_CPUQ_CFG_CPUQ_SFLOW(x)                        ((x) & GENMASK(2, 0))
+#define ANA_CPUQ_CFG_CPUQ_SFLOW_M                         GENMASK(2, 0)
+
+#define ANA_CPUQ_8021_CFG_RSZ                             0x4
+
+#define ANA_CPUQ_8021_CFG_CPUQ_BPDU_VAL(x)                (((x) << 6) & GENMASK(8, 6))
+#define ANA_CPUQ_8021_CFG_CPUQ_BPDU_VAL_M                 GENMASK(8, 6)
+#define ANA_CPUQ_8021_CFG_CPUQ_BPDU_VAL_X(x)              (((x) & GENMASK(8, 6)) >> 6)
+#define ANA_CPUQ_8021_CFG_CPUQ_GARP_VAL(x)                (((x) << 3) & GENMASK(5, 3))
+#define ANA_CPUQ_8021_CFG_CPUQ_GARP_VAL_M                 GENMASK(5, 3)
+#define ANA_CPUQ_8021_CFG_CPUQ_GARP_VAL_X(x)              (((x) & GENMASK(5, 3)) >> 3)
+#define ANA_CPUQ_8021_CFG_CPUQ_CCM_VAL(x)                 ((x) & GENMASK(2, 0))
+#define ANA_CPUQ_8021_CFG_CPUQ_CCM_VAL_M                  GENMASK(2, 0)
+
+#define ANA_DSCP_CFG_RSZ                                  0x4
+
+#define ANA_DSCP_CFG_DP_DSCP_VAL                          BIT(11)
+#define ANA_DSCP_CFG_QOS_DSCP_VAL(x)                      (((x) << 8) & GENMASK(10, 8))
+#define ANA_DSCP_CFG_QOS_DSCP_VAL_M                       GENMASK(10, 8)
+#define ANA_DSCP_CFG_QOS_DSCP_VAL_X(x)                    (((x) & GENMASK(10, 8)) >> 8)
+#define ANA_DSCP_CFG_DSCP_TRANSLATE_VAL(x)                (((x) << 2) & GENMASK(7, 2))
+#define ANA_DSCP_CFG_DSCP_TRANSLATE_VAL_M                 GENMASK(7, 2)
+#define ANA_DSCP_CFG_DSCP_TRANSLATE_VAL_X(x)              (((x) & GENMASK(7, 2)) >> 2)
+#define ANA_DSCP_CFG_DSCP_TRUST_ENA                       BIT(1)
+#define ANA_DSCP_CFG_DSCP_REWR_ENA                        BIT(0)
+
+#define ANA_DSCP_REWR_CFG_RSZ                             0x4
+
+#define ANA_VCAP_RNG_TYPE_CFG_RSZ                         0x4
+
+#define ANA_VCAP_RNG_VAL_CFG_RSZ                          0x4
+
+#define ANA_VCAP_RNG_VAL_CFG_VCAP_RNG_MIN_VAL(x)          (((x) << 16) & GENMASK(31, 16))
+#define ANA_VCAP_RNG_VAL_CFG_VCAP_RNG_MIN_VAL_M           GENMASK(31, 16)
+#define ANA_VCAP_RNG_VAL_CFG_VCAP_RNG_MIN_VAL_X(x)        (((x) & GENMASK(31, 16)) >> 16)
+#define ANA_VCAP_RNG_VAL_CFG_VCAP_RNG_MAX_VAL(x)          ((x) & GENMASK(15, 0))
+#define ANA_VCAP_RNG_VAL_CFG_VCAP_RNG_MAX_VAL_M           GENMASK(15, 0)
+
+#define ANA_VRAP_CFG_VRAP_VLAN_AWARE_ENA                  BIT(12)
+#define ANA_VRAP_CFG_VRAP_VID(x)                          ((x) & GENMASK(11, 0))
+#define ANA_VRAP_CFG_VRAP_VID_M                           GENMASK(11, 0)
+
+#define ANA_DISCARD_CFG_DROP_TAGGING_ISDX0                BIT(3)
+#define ANA_DISCARD_CFG_DROP_CTRLPROT_ISDX0               BIT(2)
+#define ANA_DISCARD_CFG_DROP_TAGGING_S2_ENA               BIT(1)
+#define ANA_DISCARD_CFG_DROP_CTRLPROT_S2_ENA              BIT(0)
+
+#define ANA_FID_CFG_VID_MC_ENA                            BIT(0)
+
+#define ANA_POL_PIR_CFG_GSZ                               0x20
+
+#define ANA_POL_PIR_CFG_PIR_RATE(x)                       (((x) << 6) & GENMASK(20, 6))
+#define ANA_POL_PIR_CFG_PIR_RATE_M                        GENMASK(20, 6)
+#define ANA_POL_PIR_CFG_PIR_RATE_X(x)                     (((x) & GENMASK(20, 6)) >> 6)
+#define ANA_POL_PIR_CFG_PIR_BURST(x)                      ((x) & GENMASK(5, 0))
+#define ANA_POL_PIR_CFG_PIR_BURST_M                       GENMASK(5, 0)
+
+#define ANA_POL_CIR_CFG_GSZ                               0x20
+
+#define ANA_POL_CIR_CFG_CIR_RATE(x)                       (((x) << 6) & GENMASK(20, 6))
+#define ANA_POL_CIR_CFG_CIR_RATE_M                        GENMASK(20, 6)
+#define ANA_POL_CIR_CFG_CIR_RATE_X(x)                     (((x) & GENMASK(20, 6)) >> 6)
+#define ANA_POL_CIR_CFG_CIR_BURST(x)                      ((x) & GENMASK(5, 0))
+#define ANA_POL_CIR_CFG_CIR_BURST_M                       GENMASK(5, 0)
+
+#define ANA_POL_MODE_CFG_GSZ                              0x20
+
+#define ANA_POL_MODE_CFG_IPG_SIZE(x)                      (((x) << 5) & GENMASK(9, 5))
+#define ANA_POL_MODE_CFG_IPG_SIZE_M                       GENMASK(9, 5)
+#define ANA_POL_MODE_CFG_IPG_SIZE_X(x)                    (((x) & GENMASK(9, 5)) >> 5)
+#define ANA_POL_MODE_CFG_FRM_MODE(x)                      (((x) << 3) & GENMASK(4, 3))
+#define ANA_POL_MODE_CFG_FRM_MODE_M                       GENMASK(4, 3)
+#define ANA_POL_MODE_CFG_FRM_MODE_X(x)                    (((x) & GENMASK(4, 3)) >> 3)
+#define ANA_POL_MODE_CFG_DLB_COUPLED                      BIT(2)
+#define ANA_POL_MODE_CFG_CIR_ENA                          BIT(1)
+#define ANA_POL_MODE_CFG_OVERSHOOT_ENA                    BIT(0)
+
+#define ANA_POL_PIR_STATE_GSZ                             0x20
+
+#define ANA_POL_CIR_STATE_GSZ                             0x20
+
+#define ANA_POL_STATE_GSZ                                 0x20
+
+#define ANA_POL_FLOWC_RSZ                                 0x4
+
+#define ANA_POL_FLOWC_POL_FLOWC                           BIT(0)
+
+#define ANA_POL_HYST_POL_FC_HYST(x)                       (((x) << 4) & GENMASK(9, 4))
+#define ANA_POL_HYST_POL_FC_HYST_M                        GENMASK(9, 4)
+#define ANA_POL_HYST_POL_FC_HYST_X(x)                     (((x) & GENMASK(9, 4)) >> 4)
+#define ANA_POL_HYST_POL_STOP_HYST(x)                     ((x) & GENMASK(3, 0))
+#define ANA_POL_HYST_POL_STOP_HYST_M                      GENMASK(3, 0)
+
+#define ANA_POL_MISC_CFG_POL_CLOSE_ALL                    BIT(1)
+#define ANA_POL_MISC_CFG_POL_LEAK_DIS                     BIT(0)
+
+#endif
diff --git a/include/soc/mscc/ocelot_dev.h b/include/soc/mscc/ocelot_dev.h
new file mode 100644
index 000000000000..0a50d53bbd3f
--- /dev/null
+++ b/include/soc/mscc/ocelot_dev.h
@@ -0,0 +1,275 @@
+/* SPDX-License-Identifier: (GPL-2.0 OR MIT) */
+/*
+ * Microsemi Ocelot Switch driver
+ *
+ * Copyright (c) 2017 Microsemi Corporation
+ */
+
+#ifndef _MSCC_OCELOT_DEV_H_
+#define _MSCC_OCELOT_DEV_H_
+
+#define DEV_CLOCK_CFG                                     0x0
+
+#define DEV_CLOCK_CFG_MAC_TX_RST                          BIT(7)
+#define DEV_CLOCK_CFG_MAC_RX_RST                          BIT(6)
+#define DEV_CLOCK_CFG_PCS_TX_RST                          BIT(5)
+#define DEV_CLOCK_CFG_PCS_RX_RST                          BIT(4)
+#define DEV_CLOCK_CFG_PORT_RST                            BIT(3)
+#define DEV_CLOCK_CFG_PHY_RST                             BIT(2)
+#define DEV_CLOCK_CFG_LINK_SPEED(x)                       ((x) & GENMASK(1, 0))
+#define DEV_CLOCK_CFG_LINK_SPEED_M                        GENMASK(1, 0)
+
+#define DEV_PORT_MISC                                     0x4
+
+#define DEV_PORT_MISC_FWD_ERROR_ENA                       BIT(4)
+#define DEV_PORT_MISC_FWD_PAUSE_ENA                       BIT(3)
+#define DEV_PORT_MISC_FWD_CTRL_ENA                        BIT(2)
+#define DEV_PORT_MISC_DEV_LOOP_ENA                        BIT(1)
+#define DEV_PORT_MISC_HDX_FAST_DIS                        BIT(0)
+
+#define DEV_EVENTS                                        0x8
+
+#define DEV_EEE_CFG                                       0xc
+
+#define DEV_EEE_CFG_EEE_ENA                               BIT(22)
+#define DEV_EEE_CFG_EEE_TIMER_AGE(x)                      (((x) << 15) & GENMASK(21, 15))
+#define DEV_EEE_CFG_EEE_TIMER_AGE_M                       GENMASK(21, 15)
+#define DEV_EEE_CFG_EEE_TIMER_AGE_X(x)                    (((x) & GENMASK(21, 15)) >> 15)
+#define DEV_EEE_CFG_EEE_TIMER_WAKEUP(x)                   (((x) << 8) & GENMASK(14, 8))
+#define DEV_EEE_CFG_EEE_TIMER_WAKEUP_M                    GENMASK(14, 8)
+#define DEV_EEE_CFG_EEE_TIMER_WAKEUP_X(x)                 (((x) & GENMASK(14, 8)) >> 8)
+#define DEV_EEE_CFG_EEE_TIMER_HOLDOFF(x)                  (((x) << 1) & GENMASK(7, 1))
+#define DEV_EEE_CFG_EEE_TIMER_HOLDOFF_M                   GENMASK(7, 1)
+#define DEV_EEE_CFG_EEE_TIMER_HOLDOFF_X(x)                (((x) & GENMASK(7, 1)) >> 1)
+#define DEV_EEE_CFG_PORT_LPI                              BIT(0)
+
+#define DEV_RX_PATH_DELAY                                 0x10
+
+#define DEV_TX_PATH_DELAY                                 0x14
+
+#define DEV_PTP_PREDICT_CFG                               0x18
+
+#define DEV_PTP_PREDICT_CFG_PTP_PHY_PREDICT_CFG(x)        (((x) << 4) & GENMASK(11, 4))
+#define DEV_PTP_PREDICT_CFG_PTP_PHY_PREDICT_CFG_M         GENMASK(11, 4)
+#define DEV_PTP_PREDICT_CFG_PTP_PHY_PREDICT_CFG_X(x)      (((x) & GENMASK(11, 4)) >> 4)
+#define DEV_PTP_PREDICT_CFG_PTP_PHASE_PREDICT_CFG(x)      ((x) & GENMASK(3, 0))
+#define DEV_PTP_PREDICT_CFG_PTP_PHASE_PREDICT_CFG_M       GENMASK(3, 0)
+
+#define DEV_MAC_ENA_CFG                                   0x1c
+
+#define DEV_MAC_ENA_CFG_RX_ENA                            BIT(4)
+#define DEV_MAC_ENA_CFG_TX_ENA                            BIT(0)
+
+#define DEV_MAC_MODE_CFG                                  0x20
+
+#define DEV_MAC_MODE_CFG_FC_WORD_SYNC_ENA                 BIT(8)
+#define DEV_MAC_MODE_CFG_GIGA_MODE_ENA                    BIT(4)
+#define DEV_MAC_MODE_CFG_FDX_ENA                          BIT(0)
+
+#define DEV_MAC_MAXLEN_CFG                                0x24
+
+#define DEV_MAC_TAGS_CFG                                  0x28
+
+#define DEV_MAC_TAGS_CFG_TAG_ID(x)                        (((x) << 16) & GENMASK(31, 16))
+#define DEV_MAC_TAGS_CFG_TAG_ID_M                         GENMASK(31, 16)
+#define DEV_MAC_TAGS_CFG_TAG_ID_X(x)                      (((x) & GENMASK(31, 16)) >> 16)
+#define DEV_MAC_TAGS_CFG_VLAN_LEN_AWR_ENA                 BIT(2)
+#define DEV_MAC_TAGS_CFG_PB_ENA                           BIT(1)
+#define DEV_MAC_TAGS_CFG_VLAN_AWR_ENA                     BIT(0)
+
+#define DEV_MAC_ADV_CHK_CFG                               0x2c
+
+#define DEV_MAC_ADV_CHK_CFG_LEN_DROP_ENA                  BIT(0)
+
+#define DEV_MAC_IFG_CFG                                   0x30
+
+#define DEV_MAC_IFG_CFG_RESTORE_OLD_IPG_CHECK             BIT(17)
+#define DEV_MAC_IFG_CFG_REDUCED_TX_IFG                    BIT(16)
+#define DEV_MAC_IFG_CFG_TX_IFG(x)                         (((x) << 8) & GENMASK(12, 8))
+#define DEV_MAC_IFG_CFG_TX_IFG_M                          GENMASK(12, 8)
+#define DEV_MAC_IFG_CFG_TX_IFG_X(x)                       (((x) & GENMASK(12, 8)) >> 8)
+#define DEV_MAC_IFG_CFG_RX_IFG2(x)                        (((x) << 4) & GENMASK(7, 4))
+#define DEV_MAC_IFG_CFG_RX_IFG2_M                         GENMASK(7, 4)
+#define DEV_MAC_IFG_CFG_RX_IFG2_X(x)                      (((x) & GENMASK(7, 4)) >> 4)
+#define DEV_MAC_IFG_CFG_RX_IFG1(x)                        ((x) & GENMASK(3, 0))
+#define DEV_MAC_IFG_CFG_RX_IFG1_M                         GENMASK(3, 0)
+
+#define DEV_MAC_HDX_CFG                                   0x34
+
+#define DEV_MAC_HDX_CFG_BYPASS_COL_SYNC                   BIT(26)
+#define DEV_MAC_HDX_CFG_OB_ENA                            BIT(25)
+#define DEV_MAC_HDX_CFG_WEXC_DIS                          BIT(24)
+#define DEV_MAC_HDX_CFG_SEED(x)                           (((x) << 16) & GENMASK(23, 16))
+#define DEV_MAC_HDX_CFG_SEED_M                            GENMASK(23, 16)
+#define DEV_MAC_HDX_CFG_SEED_X(x)                         (((x) & GENMASK(23, 16)) >> 16)
+#define DEV_MAC_HDX_CFG_SEED_LOAD                         BIT(12)
+#define DEV_MAC_HDX_CFG_RETRY_AFTER_EXC_COL_ENA           BIT(8)
+#define DEV_MAC_HDX_CFG_LATE_COL_POS(x)                   ((x) & GENMASK(6, 0))
+#define DEV_MAC_HDX_CFG_LATE_COL_POS_M                    GENMASK(6, 0)
+
+#define DEV_MAC_DBG_CFG                                   0x38
+
+#define DEV_MAC_DBG_CFG_TBI_MODE                          BIT(4)
+#define DEV_MAC_DBG_CFG_IFG_CRS_EXT_CHK_ENA               BIT(0)
+
+#define DEV_MAC_FC_MAC_LOW_CFG                            0x3c
+
+#define DEV_MAC_FC_MAC_HIGH_CFG                           0x40
+
+#define DEV_MAC_STICKY                                    0x44
+
+#define DEV_MAC_STICKY_RX_IPG_SHRINK_STICKY               BIT(9)
+#define DEV_MAC_STICKY_RX_PREAM_SHRINK_STICKY             BIT(8)
+#define DEV_MAC_STICKY_RX_CARRIER_EXT_STICKY              BIT(7)
+#define DEV_MAC_STICKY_RX_CARRIER_EXT_ERR_STICKY          BIT(6)
+#define DEV_MAC_STICKY_RX_JUNK_STICKY                     BIT(5)
+#define DEV_MAC_STICKY_TX_RETRANSMIT_STICKY               BIT(4)
+#define DEV_MAC_STICKY_TX_JAM_STICKY                      BIT(3)
+#define DEV_MAC_STICKY_TX_FIFO_OFLW_STICKY                BIT(2)
+#define DEV_MAC_STICKY_TX_FRM_LEN_OVR_STICKY              BIT(1)
+#define DEV_MAC_STICKY_TX_ABORT_STICKY                    BIT(0)
+
+#define PCS1G_CFG                                         0x48
+
+#define PCS1G_CFG_LINK_STATUS_TYPE                        BIT(4)
+#define PCS1G_CFG_AN_LINK_CTRL_ENA                        BIT(1)
+#define PCS1G_CFG_PCS_ENA                                 BIT(0)
+
+#define PCS1G_MODE_CFG                                    0x4c
+
+#define PCS1G_MODE_CFG_UNIDIR_MODE_ENA                    BIT(4)
+#define PCS1G_MODE_CFG_SGMII_MODE_ENA                     BIT(0)
+
+#define PCS1G_SD_CFG                                      0x50
+
+#define PCS1G_SD_CFG_SD_SEL                               BIT(8)
+#define PCS1G_SD_CFG_SD_POL                               BIT(4)
+#define PCS1G_SD_CFG_SD_ENA                               BIT(0)
+
+#define PCS1G_ANEG_CFG                                    0x54
+
+#define PCS1G_ANEG_CFG_ADV_ABILITY(x)                     (((x) << 16) & GENMASK(31, 16))
+#define PCS1G_ANEG_CFG_ADV_ABILITY_M                      GENMASK(31, 16)
+#define PCS1G_ANEG_CFG_ADV_ABILITY_X(x)                   (((x) & GENMASK(31, 16)) >> 16)
+#define PCS1G_ANEG_CFG_SW_RESOLVE_ENA                     BIT(8)
+#define PCS1G_ANEG_CFG_ANEG_RESTART_ONE_SHOT              BIT(1)
+#define PCS1G_ANEG_CFG_ANEG_ENA                           BIT(0)
+
+#define PCS1G_ANEG_NP_CFG                                 0x58
+
+#define PCS1G_ANEG_NP_CFG_NP_TX(x)                        (((x) << 16) & GENMASK(31, 16))
+#define PCS1G_ANEG_NP_CFG_NP_TX_M                         GENMASK(31, 16)
+#define PCS1G_ANEG_NP_CFG_NP_TX_X(x)                      (((x) & GENMASK(31, 16)) >> 16)
+#define PCS1G_ANEG_NP_CFG_NP_LOADED_ONE_SHOT              BIT(0)
+
+#define PCS1G_LB_CFG                                      0x5c
+
+#define PCS1G_LB_CFG_RA_ENA                               BIT(4)
+#define PCS1G_LB_CFG_GMII_PHY_LB_ENA                      BIT(1)
+#define PCS1G_LB_CFG_TBI_HOST_LB_ENA                      BIT(0)
+
+#define PCS1G_DBG_CFG                                     0x60
+
+#define PCS1G_DBG_CFG_UDLT                                BIT(0)
+
+#define PCS1G_CDET_CFG                                    0x64
+
+#define PCS1G_CDET_CFG_CDET_ENA                           BIT(0)
+
+#define PCS1G_ANEG_STATUS                                 0x68
+
+#define PCS1G_ANEG_STATUS_LP_ADV_ABILITY(x)               (((x) << 16) & GENMASK(31, 16))
+#define PCS1G_ANEG_STATUS_LP_ADV_ABILITY_M                GENMASK(31, 16)
+#define PCS1G_ANEG_STATUS_LP_ADV_ABILITY_X(x)             (((x) & GENMASK(31, 16)) >> 16)
+#define PCS1G_ANEG_STATUS_PR                              BIT(4)
+#define PCS1G_ANEG_STATUS_PAGE_RX_STICKY                  BIT(3)
+#define PCS1G_ANEG_STATUS_ANEG_COMPLETE                   BIT(0)
+
+#define PCS1G_ANEG_NP_STATUS                              0x6c
+
+#define PCS1G_LINK_STATUS                                 0x70
+
+#define PCS1G_LINK_STATUS_DELAY_VAR(x)                    (((x) << 12) & GENMASK(15, 12))
+#define PCS1G_LINK_STATUS_DELAY_VAR_M                     GENMASK(15, 12)
+#define PCS1G_LINK_STATUS_DELAY_VAR_X(x)                  (((x) & GENMASK(15, 12)) >> 12)
+#define PCS1G_LINK_STATUS_SIGNAL_DETECT                   BIT(8)
+#define PCS1G_LINK_STATUS_LINK_STATUS                     BIT(4)
+#define PCS1G_LINK_STATUS_SYNC_STATUS                     BIT(0)
+
+#define PCS1G_LINK_DOWN_CNT                               0x74
+
+#define PCS1G_STICKY                                      0x78
+
+#define PCS1G_STICKY_LINK_DOWN_STICKY                     BIT(4)
+#define PCS1G_STICKY_OUT_OF_SYNC_STICKY                   BIT(0)
+
+#define PCS1G_DEBUG_STATUS                                0x7c
+
+#define PCS1G_LPI_CFG                                     0x80
+
+#define PCS1G_LPI_CFG_QSGMII_MS_SEL                       BIT(20)
+#define PCS1G_LPI_CFG_RX_LPI_OUT_DIS                      BIT(17)
+#define PCS1G_LPI_CFG_LPI_TESTMODE                        BIT(16)
+#define PCS1G_LPI_CFG_LPI_RX_WTIM(x)                      (((x) << 4) & GENMASK(5, 4))
+#define PCS1G_LPI_CFG_LPI_RX_WTIM_M                       GENMASK(5, 4)
+#define PCS1G_LPI_CFG_LPI_RX_WTIM_X(x)                    (((x) & GENMASK(5, 4)) >> 4)
+#define PCS1G_LPI_CFG_TX_ASSERT_LPIDLE                    BIT(0)
+
+#define PCS1G_LPI_WAKE_ERROR_CNT                          0x84
+
+#define PCS1G_LPI_STATUS                                  0x88
+
+#define PCS1G_LPI_STATUS_RX_LPI_FAIL                      BIT(16)
+#define PCS1G_LPI_STATUS_RX_LPI_EVENT_STICKY              BIT(12)
+#define PCS1G_LPI_STATUS_RX_QUIET                         BIT(9)
+#define PCS1G_LPI_STATUS_RX_LPI_MODE                      BIT(8)
+#define PCS1G_LPI_STATUS_TX_LPI_EVENT_STICKY              BIT(4)
+#define PCS1G_LPI_STATUS_TX_QUIET                         BIT(1)
+#define PCS1G_LPI_STATUS_TX_LPI_MODE                      BIT(0)
+
+#define PCS1G_TSTPAT_MODE_CFG                             0x8c
+
+#define PCS1G_TSTPAT_STATUS                               0x90
+
+#define PCS1G_TSTPAT_STATUS_JTP_ERR_CNT(x)                (((x) << 8) & GENMASK(15, 8))
+#define PCS1G_TSTPAT_STATUS_JTP_ERR_CNT_M                 GENMASK(15, 8)
+#define PCS1G_TSTPAT_STATUS_JTP_ERR_CNT_X(x)              (((x) & GENMASK(15, 8)) >> 8)
+#define PCS1G_TSTPAT_STATUS_JTP_ERR                       BIT(4)
+#define PCS1G_TSTPAT_STATUS_JTP_LOCK                      BIT(0)
+
+#define DEV_PCS_FX100_CFG                                 0x94
+
+#define DEV_PCS_FX100_CFG_SD_SEL                          BIT(26)
+#define DEV_PCS_FX100_CFG_SD_POL                          BIT(25)
+#define DEV_PCS_FX100_CFG_SD_ENA                          BIT(24)
+#define DEV_PCS_FX100_CFG_LOOPBACK_ENA                    BIT(20)
+#define DEV_PCS_FX100_CFG_SWAP_MII_ENA                    BIT(16)
+#define DEV_PCS_FX100_CFG_RXBITSEL(x)                     (((x) << 12) & GENMASK(15, 12))
+#define DEV_PCS_FX100_CFG_RXBITSEL_M                      GENMASK(15, 12)
+#define DEV_PCS_FX100_CFG_RXBITSEL_X(x)                   (((x) & GENMASK(15, 12)) >> 12)
+#define DEV_PCS_FX100_CFG_SIGDET_CFG(x)                   (((x) << 9) & GENMASK(10, 9))
+#define DEV_PCS_FX100_CFG_SIGDET_CFG_M                    GENMASK(10, 9)
+#define DEV_PCS_FX100_CFG_SIGDET_CFG_X(x)                 (((x) & GENMASK(10, 9)) >> 9)
+#define DEV_PCS_FX100_CFG_LINKHYST_TM_ENA                 BIT(8)
+#define DEV_PCS_FX100_CFG_LINKHYSTTIMER(x)                (((x) << 4) & GENMASK(7, 4))
+#define DEV_PCS_FX100_CFG_LINKHYSTTIMER_M                 GENMASK(7, 4)
+#define DEV_PCS_FX100_CFG_LINKHYSTTIMER_X(x)              (((x) & GENMASK(7, 4)) >> 4)
+#define DEV_PCS_FX100_CFG_UNIDIR_MODE_ENA                 BIT(3)
+#define DEV_PCS_FX100_CFG_FEFCHK_ENA                      BIT(2)
+#define DEV_PCS_FX100_CFG_FEFGEN_ENA                      BIT(1)
+#define DEV_PCS_FX100_CFG_PCS_ENA                         BIT(0)
+
+#define DEV_PCS_FX100_STATUS                              0x98
+
+#define DEV_PCS_FX100_STATUS_EDGE_POS_PTP(x)              (((x) << 8) & GENMASK(11, 8))
+#define DEV_PCS_FX100_STATUS_EDGE_POS_PTP_M               GENMASK(11, 8)
+#define DEV_PCS_FX100_STATUS_EDGE_POS_PTP_X(x)            (((x) & GENMASK(11, 8)) >> 8)
+#define DEV_PCS_FX100_STATUS_PCS_ERROR_STICKY             BIT(7)
+#define DEV_PCS_FX100_STATUS_FEF_FOUND_STICKY             BIT(6)
+#define DEV_PCS_FX100_STATUS_SSD_ERROR_STICKY             BIT(5)
+#define DEV_PCS_FX100_STATUS_SYNC_LOST_STICKY             BIT(4)
+#define DEV_PCS_FX100_STATUS_FEF_STATUS                   BIT(2)
+#define DEV_PCS_FX100_STATUS_SIGNAL_DETECT                BIT(1)
+#define DEV_PCS_FX100_STATUS_SYNC_STATUS                  BIT(0)
+
+#endif
diff --git a/include/soc/mscc/ocelot_qsys.h b/include/soc/mscc/ocelot_qsys.h
new file mode 100644
index 000000000000..d8c63aa761be
--- /dev/null
+++ b/include/soc/mscc/ocelot_qsys.h
@@ -0,0 +1,270 @@
+/* SPDX-License-Identifier: (GPL-2.0 OR MIT) */
+/*
+ * Microsemi Ocelot Switch driver
+ *
+ * Copyright (c) 2017 Microsemi Corporation
+ */
+
+#ifndef _MSCC_OCELOT_QSYS_H_
+#define _MSCC_OCELOT_QSYS_H_
+
+#define QSYS_PORT_MODE_RSZ                                0x4
+
+#define QSYS_PORT_MODE_DEQUEUE_DIS                        BIT(1)
+#define QSYS_PORT_MODE_DEQUEUE_LATE                       BIT(0)
+
+#define QSYS_SWITCH_PORT_MODE_RSZ                         0x4
+
+#define QSYS_SWITCH_PORT_MODE_PORT_ENA                    BIT(14)
+#define QSYS_SWITCH_PORT_MODE_SCH_NEXT_CFG(x)             (((x) << 11) & GENMASK(13, 11))
+#define QSYS_SWITCH_PORT_MODE_SCH_NEXT_CFG_M              GENMASK(13, 11)
+#define QSYS_SWITCH_PORT_MODE_SCH_NEXT_CFG_X(x)           (((x) & GENMASK(13, 11)) >> 11)
+#define QSYS_SWITCH_PORT_MODE_YEL_RSRVD                   BIT(10)
+#define QSYS_SWITCH_PORT_MODE_INGRESS_DROP_MODE           BIT(9)
+#define QSYS_SWITCH_PORT_MODE_TX_PFC_ENA(x)               (((x) << 1) & GENMASK(8, 1))
+#define QSYS_SWITCH_PORT_MODE_TX_PFC_ENA_M                GENMASK(8, 1)
+#define QSYS_SWITCH_PORT_MODE_TX_PFC_ENA_X(x)             (((x) & GENMASK(8, 1)) >> 1)
+#define QSYS_SWITCH_PORT_MODE_TX_PFC_MODE                 BIT(0)
+
+#define QSYS_STAT_CNT_CFG_TX_GREEN_CNT_MODE               BIT(5)
+#define QSYS_STAT_CNT_CFG_TX_YELLOW_CNT_MODE              BIT(4)
+#define QSYS_STAT_CNT_CFG_DROP_GREEN_CNT_MODE             BIT(3)
+#define QSYS_STAT_CNT_CFG_DROP_YELLOW_CNT_MODE            BIT(2)
+#define QSYS_STAT_CNT_CFG_DROP_COUNT_ONCE                 BIT(1)
+#define QSYS_STAT_CNT_CFG_DROP_COUNT_EGRESS               BIT(0)
+
+#define QSYS_EEE_CFG_RSZ                                  0x4
+
+#define QSYS_EEE_THRES_EEE_HIGH_BYTES(x)                  (((x) << 8) & GENMASK(15, 8))
+#define QSYS_EEE_THRES_EEE_HIGH_BYTES_M                   GENMASK(15, 8)
+#define QSYS_EEE_THRES_EEE_HIGH_BYTES_X(x)                (((x) & GENMASK(15, 8)) >> 8)
+#define QSYS_EEE_THRES_EEE_HIGH_FRAMES(x)                 ((x) & GENMASK(7, 0))
+#define QSYS_EEE_THRES_EEE_HIGH_FRAMES_M                  GENMASK(7, 0)
+
+#define QSYS_SW_STATUS_RSZ                                0x4
+
+#define QSYS_EXT_CPU_CFG_EXT_CPU_PORT(x)                  (((x) << 8) & GENMASK(12, 8))
+#define QSYS_EXT_CPU_CFG_EXT_CPU_PORT_M                   GENMASK(12, 8)
+#define QSYS_EXT_CPU_CFG_EXT_CPU_PORT_X(x)                (((x) & GENMASK(12, 8)) >> 8)
+#define QSYS_EXT_CPU_CFG_EXT_CPUQ_MSK(x)                  ((x) & GENMASK(7, 0))
+#define QSYS_EXT_CPU_CFG_EXT_CPUQ_MSK_M                   GENMASK(7, 0)
+
+#define QSYS_QMAP_GSZ                                     0x4
+
+#define QSYS_QMAP_SE_BASE(x)                              (((x) << 5) & GENMASK(12, 5))
+#define QSYS_QMAP_SE_BASE_M                               GENMASK(12, 5)
+#define QSYS_QMAP_SE_BASE_X(x)                            (((x) & GENMASK(12, 5)) >> 5)
+#define QSYS_QMAP_SE_IDX_SEL(x)                           (((x) << 2) & GENMASK(4, 2))
+#define QSYS_QMAP_SE_IDX_SEL_M                            GENMASK(4, 2)
+#define QSYS_QMAP_SE_IDX_SEL_X(x)                         (((x) & GENMASK(4, 2)) >> 2)
+#define QSYS_QMAP_SE_INP_SEL(x)                           ((x) & GENMASK(1, 0))
+#define QSYS_QMAP_SE_INP_SEL_M                            GENMASK(1, 0)
+
+#define QSYS_ISDX_SGRP_GSZ                                0x4
+
+#define QSYS_TIMED_FRAME_ENTRY_GSZ                        0x4
+
+#define QSYS_TFRM_MISC_TIMED_CANCEL_SLOT(x)               (((x) << 9) & GENMASK(18, 9))
+#define QSYS_TFRM_MISC_TIMED_CANCEL_SLOT_M                GENMASK(18, 9)
+#define QSYS_TFRM_MISC_TIMED_CANCEL_SLOT_X(x)             (((x) & GENMASK(18, 9)) >> 9)
+#define QSYS_TFRM_MISC_TIMED_CANCEL_1SHOT                 BIT(8)
+#define QSYS_TFRM_MISC_TIMED_SLOT_MODE_MC                 BIT(7)
+#define QSYS_TFRM_MISC_TIMED_ENTRY_FAST_CNT(x)            ((x) & GENMASK(6, 0))
+#define QSYS_TFRM_MISC_TIMED_ENTRY_FAST_CNT_M             GENMASK(6, 0)
+
+#define QSYS_RED_PROFILE_RSZ                              0x4
+
+#define QSYS_RED_PROFILE_WM_RED_LOW(x)                    (((x) << 8) & GENMASK(15, 8))
+#define QSYS_RED_PROFILE_WM_RED_LOW_M                     GENMASK(15, 8)
+#define QSYS_RED_PROFILE_WM_RED_LOW_X(x)                  (((x) & GENMASK(15, 8)) >> 8)
+#define QSYS_RED_PROFILE_WM_RED_HIGH(x)                   ((x) & GENMASK(7, 0))
+#define QSYS_RED_PROFILE_WM_RED_HIGH_M                    GENMASK(7, 0)
+
+#define QSYS_RES_CFG_GSZ                                  0x8
+
+#define QSYS_RES_STAT_GSZ                                 0x8
+
+#define QSYS_RES_STAT_INUSE(x)                            (((x) << 12) & GENMASK(23, 12))
+#define QSYS_RES_STAT_INUSE_M                             GENMASK(23, 12)
+#define QSYS_RES_STAT_INUSE_X(x)                          (((x) & GENMASK(23, 12)) >> 12)
+#define QSYS_RES_STAT_MAXUSE(x)                           ((x) & GENMASK(11, 0))
+#define QSYS_RES_STAT_MAXUSE_M                            GENMASK(11, 0)
+
+#define QSYS_EVENTS_CORE_EV_FDC(x)                        (((x) << 2) & GENMASK(4, 2))
+#define QSYS_EVENTS_CORE_EV_FDC_M                         GENMASK(4, 2)
+#define QSYS_EVENTS_CORE_EV_FDC_X(x)                      (((x) & GENMASK(4, 2)) >> 2)
+#define QSYS_EVENTS_CORE_EV_FRD(x)                        ((x) & GENMASK(1, 0))
+#define QSYS_EVENTS_CORE_EV_FRD_M                         GENMASK(1, 0)
+
+#define QSYS_QMAXSDU_CFG_0_RSZ                            0x4
+
+#define QSYS_QMAXSDU_CFG_1_RSZ                            0x4
+
+#define QSYS_QMAXSDU_CFG_2_RSZ                            0x4
+
+#define QSYS_QMAXSDU_CFG_3_RSZ                            0x4
+
+#define QSYS_QMAXSDU_CFG_4_RSZ                            0x4
+
+#define QSYS_QMAXSDU_CFG_5_RSZ                            0x4
+
+#define QSYS_QMAXSDU_CFG_6_RSZ                            0x4
+
+#define QSYS_QMAXSDU_CFG_7_RSZ                            0x4
+
+#define QSYS_PREEMPTION_CFG_RSZ                           0x4
+
+#define QSYS_PREEMPTION_CFG_P_QUEUES(x)                   ((x) & GENMASK(7, 0))
+#define QSYS_PREEMPTION_CFG_P_QUEUES_M                    GENMASK(7, 0)
+#define QSYS_PREEMPTION_CFG_MM_ADD_FRAG_SIZE(x)           (((x) << 8) & GENMASK(9, 8))
+#define QSYS_PREEMPTION_CFG_MM_ADD_FRAG_SIZE_M            GENMASK(9, 8)
+#define QSYS_PREEMPTION_CFG_MM_ADD_FRAG_SIZE_X(x)         (((x) & GENMASK(9, 8)) >> 8)
+#define QSYS_PREEMPTION_CFG_STRICT_IPG(x)                 (((x) << 12) & GENMASK(13, 12))
+#define QSYS_PREEMPTION_CFG_STRICT_IPG_M                  GENMASK(13, 12)
+#define QSYS_PREEMPTION_CFG_STRICT_IPG_X(x)               (((x) & GENMASK(13, 12)) >> 12)
+#define QSYS_PREEMPTION_CFG_HOLD_ADVANCE(x)               (((x) << 16) & GENMASK(31, 16))
+#define QSYS_PREEMPTION_CFG_HOLD_ADVANCE_M                GENMASK(31, 16)
+#define QSYS_PREEMPTION_CFG_HOLD_ADVANCE_X(x)             (((x) & GENMASK(31, 16)) >> 16)
+
+#define QSYS_CIR_CFG_GSZ                                  0x80
+
+#define QSYS_CIR_CFG_CIR_RATE(x)                          (((x) << 6) & GENMASK(20, 6))
+#define QSYS_CIR_CFG_CIR_RATE_M                           GENMASK(20, 6)
+#define QSYS_CIR_CFG_CIR_RATE_X(x)                        (((x) & GENMASK(20, 6)) >> 6)
+#define QSYS_CIR_CFG_CIR_BURST(x)                         ((x) & GENMASK(5, 0))
+#define QSYS_CIR_CFG_CIR_BURST_M                          GENMASK(5, 0)
+
+#define QSYS_EIR_CFG_GSZ                                  0x80
+
+#define QSYS_EIR_CFG_EIR_RATE(x)                          (((x) << 7) & GENMASK(21, 7))
+#define QSYS_EIR_CFG_EIR_RATE_M                           GENMASK(21, 7)
+#define QSYS_EIR_CFG_EIR_RATE_X(x)                        (((x) & GENMASK(21, 7)) >> 7)
+#define QSYS_EIR_CFG_EIR_BURST(x)                         (((x) << 1) & GENMASK(6, 1))
+#define QSYS_EIR_CFG_EIR_BURST_M                          GENMASK(6, 1)
+#define QSYS_EIR_CFG_EIR_BURST_X(x)                       (((x) & GENMASK(6, 1)) >> 1)
+#define QSYS_EIR_CFG_EIR_MARK_ENA                         BIT(0)
+
+#define QSYS_SE_CFG_GSZ                                   0x80
+
+#define QSYS_SE_CFG_SE_DWRR_CNT(x)                        (((x) << 6) & GENMASK(9, 6))
+#define QSYS_SE_CFG_SE_DWRR_CNT_M                         GENMASK(9, 6)
+#define QSYS_SE_CFG_SE_DWRR_CNT_X(x)                      (((x) & GENMASK(9, 6)) >> 6)
+#define QSYS_SE_CFG_SE_RR_ENA                             BIT(5)
+#define QSYS_SE_CFG_SE_AVB_ENA                            BIT(4)
+#define QSYS_SE_CFG_SE_FRM_MODE(x)                        (((x) << 2) & GENMASK(3, 2))
+#define QSYS_SE_CFG_SE_FRM_MODE_M                         GENMASK(3, 2)
+#define QSYS_SE_CFG_SE_FRM_MODE_X(x)                      (((x) & GENMASK(3, 2)) >> 2)
+#define QSYS_SE_CFG_SE_EXC_ENA                            BIT(1)
+#define QSYS_SE_CFG_SE_EXC_FWD                            BIT(0)
+
+#define QSYS_SE_DWRR_CFG_GSZ                              0x80
+#define QSYS_SE_DWRR_CFG_RSZ                              0x4
+
+#define QSYS_SE_CONNECT_GSZ                               0x80
+
+#define QSYS_SE_CONNECT_SE_OUTP_IDX(x)                    (((x) << 17) & GENMASK(24, 17))
+#define QSYS_SE_CONNECT_SE_OUTP_IDX_M                     GENMASK(24, 17)
+#define QSYS_SE_CONNECT_SE_OUTP_IDX_X(x)                  (((x) & GENMASK(24, 17)) >> 17)
+#define QSYS_SE_CONNECT_SE_INP_IDX(x)                     (((x) << 9) & GENMASK(16, 9))
+#define QSYS_SE_CONNECT_SE_INP_IDX_M                      GENMASK(16, 9)
+#define QSYS_SE_CONNECT_SE_INP_IDX_X(x)                   (((x) & GENMASK(16, 9)) >> 9)
+#define QSYS_SE_CONNECT_SE_OUTP_CON(x)                    (((x) << 5) & GENMASK(8, 5))
+#define QSYS_SE_CONNECT_SE_OUTP_CON_M                     GENMASK(8, 5)
+#define QSYS_SE_CONNECT_SE_OUTP_CON_X(x)                  (((x) & GENMASK(8, 5)) >> 5)
+#define QSYS_SE_CONNECT_SE_INP_CNT(x)                     (((x) << 1) & GENMASK(4, 1))
+#define QSYS_SE_CONNECT_SE_INP_CNT_M                      GENMASK(4, 1)
+#define QSYS_SE_CONNECT_SE_INP_CNT_X(x)                   (((x) & GENMASK(4, 1)) >> 1)
+#define QSYS_SE_CONNECT_SE_TERMINAL                       BIT(0)
+
+#define QSYS_SE_DLB_SENSE_GSZ                             0x80
+
+#define QSYS_SE_DLB_SENSE_SE_DLB_PRIO(x)                  (((x) << 11) & GENMASK(13, 11))
+#define QSYS_SE_DLB_SENSE_SE_DLB_PRIO_M                   GENMASK(13, 11)
+#define QSYS_SE_DLB_SENSE_SE_DLB_PRIO_X(x)                (((x) & GENMASK(13, 11)) >> 11)
+#define QSYS_SE_DLB_SENSE_SE_DLB_SPORT(x)                 (((x) << 7) & GENMASK(10, 7))
+#define QSYS_SE_DLB_SENSE_SE_DLB_SPORT_M                  GENMASK(10, 7)
+#define QSYS_SE_DLB_SENSE_SE_DLB_SPORT_X(x)               (((x) & GENMASK(10, 7)) >> 7)
+#define QSYS_SE_DLB_SENSE_SE_DLB_DPORT(x)                 (((x) << 3) & GENMASK(6, 3))
+#define QSYS_SE_DLB_SENSE_SE_DLB_DPORT_M                  GENMASK(6, 3)
+#define QSYS_SE_DLB_SENSE_SE_DLB_DPORT_X(x)               (((x) & GENMASK(6, 3)) >> 3)
+#define QSYS_SE_DLB_SENSE_SE_DLB_PRIO_ENA                 BIT(2)
+#define QSYS_SE_DLB_SENSE_SE_DLB_SPORT_ENA                BIT(1)
+#define QSYS_SE_DLB_SENSE_SE_DLB_DPORT_ENA                BIT(0)
+
+#define QSYS_CIR_STATE_GSZ                                0x80
+
+#define QSYS_CIR_STATE_CIR_LVL(x)                         (((x) << 4) & GENMASK(25, 4))
+#define QSYS_CIR_STATE_CIR_LVL_M                          GENMASK(25, 4)
+#define QSYS_CIR_STATE_CIR_LVL_X(x)                       (((x) & GENMASK(25, 4)) >> 4)
+#define QSYS_CIR_STATE_SHP_TIME(x)                        ((x) & GENMASK(3, 0))
+#define QSYS_CIR_STATE_SHP_TIME_M                         GENMASK(3, 0)
+
+#define QSYS_EIR_STATE_GSZ                                0x80
+
+#define QSYS_SE_STATE_GSZ                                 0x80
+
+#define QSYS_SE_STATE_SE_OUTP_LVL(x)                      (((x) << 1) & GENMASK(2, 1))
+#define QSYS_SE_STATE_SE_OUTP_LVL_M                       GENMASK(2, 1)
+#define QSYS_SE_STATE_SE_OUTP_LVL_X(x)                    (((x) & GENMASK(2, 1)) >> 1)
+#define QSYS_SE_STATE_SE_WAS_YEL                          BIT(0)
+
+#define QSYS_HSCH_MISC_CFG_SE_CONNECT_VLD                 BIT(8)
+#define QSYS_HSCH_MISC_CFG_FRM_ADJ(x)                     (((x) << 3) & GENMASK(7, 3))
+#define QSYS_HSCH_MISC_CFG_FRM_ADJ_M                      GENMASK(7, 3)
+#define QSYS_HSCH_MISC_CFG_FRM_ADJ_X(x)                   (((x) & GENMASK(7, 3)) >> 3)
+#define QSYS_HSCH_MISC_CFG_LEAK_DIS                       BIT(2)
+#define QSYS_HSCH_MISC_CFG_QSHP_EXC_ENA                   BIT(1)
+#define QSYS_HSCH_MISC_CFG_PFC_BYP_UPD                    BIT(0)
+
+#define QSYS_TAG_CONFIG_RSZ                               0x4
+
+#define QSYS_TAG_CONFIG_ENABLE                            BIT(0)
+#define QSYS_TAG_CONFIG_LINK_SPEED(x)                     (((x) << 4) & GENMASK(5, 4))
+#define QSYS_TAG_CONFIG_LINK_SPEED_M                      GENMASK(5, 4)
+#define QSYS_TAG_CONFIG_LINK_SPEED_X(x)                   (((x) & GENMASK(5, 4)) >> 4)
+#define QSYS_TAG_CONFIG_INIT_GATE_STATE(x)                (((x) << 8) & GENMASK(15, 8))
+#define QSYS_TAG_CONFIG_INIT_GATE_STATE_M                 GENMASK(15, 8)
+#define QSYS_TAG_CONFIG_INIT_GATE_STATE_X(x)              (((x) & GENMASK(15, 8)) >> 8)
+#define QSYS_TAG_CONFIG_SCH_TRAFFIC_QUEUES(x)             (((x) << 16) & GENMASK(23, 16))
+#define QSYS_TAG_CONFIG_SCH_TRAFFIC_QUEUES_M              GENMASK(23, 16)
+#define QSYS_TAG_CONFIG_SCH_TRAFFIC_QUEUES_X(x)           (((x) & GENMASK(23, 16)) >> 16)
+
+#define QSYS_TAS_PARAM_CFG_CTRL_PORT_NUM(x)               ((x) & GENMASK(7, 0))
+#define QSYS_TAS_PARAM_CFG_CTRL_PORT_NUM_M                GENMASK(7, 0)
+#define QSYS_TAS_PARAM_CFG_CTRL_ALWAYS_GUARD_BAND_SCH_Q   BIT(8)
+#define QSYS_TAS_PARAM_CFG_CTRL_CONFIG_CHANGE             BIT(16)
+
+#define QSYS_PORT_MAX_SDU_RSZ                             0x4
+
+#define QSYS_PARAM_CFG_REG_3_BASE_TIME_SEC_MSB(x)         ((x) & GENMASK(15, 0))
+#define QSYS_PARAM_CFG_REG_3_BASE_TIME_SEC_MSB_M          GENMASK(15, 0)
+#define QSYS_PARAM_CFG_REG_3_LIST_LENGTH(x)               (((x) << 16) & GENMASK(31, 16))
+#define QSYS_PARAM_CFG_REG_3_LIST_LENGTH_M                GENMASK(31, 16)
+#define QSYS_PARAM_CFG_REG_3_LIST_LENGTH_X(x)             (((x) & GENMASK(31, 16)) >> 16)
+
+#define QSYS_GCL_CFG_REG_1_GCL_ENTRY_NUM(x)               ((x) & GENMASK(5, 0))
+#define QSYS_GCL_CFG_REG_1_GCL_ENTRY_NUM_M                GENMASK(5, 0)
+#define QSYS_GCL_CFG_REG_1_GATE_STATE(x)                  (((x) << 8) & GENMASK(15, 8))
+#define QSYS_GCL_CFG_REG_1_GATE_STATE_M                   GENMASK(15, 8)
+#define QSYS_GCL_CFG_REG_1_GATE_STATE_X(x)                (((x) & GENMASK(15, 8)) >> 8)
+
+#define QSYS_PARAM_STATUS_REG_3_BASE_TIME_SEC_MSB(x)      ((x) & GENMASK(15, 0))
+#define QSYS_PARAM_STATUS_REG_3_BASE_TIME_SEC_MSB_M       GENMASK(15, 0)
+#define QSYS_PARAM_STATUS_REG_3_LIST_LENGTH(x)            (((x) << 16) & GENMASK(31, 16))
+#define QSYS_PARAM_STATUS_REG_3_LIST_LENGTH_M             GENMASK(31, 16)
+#define QSYS_PARAM_STATUS_REG_3_LIST_LENGTH_X(x)          (((x) & GENMASK(31, 16)) >> 16)
+
+#define QSYS_PARAM_STATUS_REG_8_CFG_CHG_TIME_SEC_MSB(x)   ((x) & GENMASK(15, 0))
+#define QSYS_PARAM_STATUS_REG_8_CFG_CHG_TIME_SEC_MSB_M    GENMASK(15, 0)
+#define QSYS_PARAM_STATUS_REG_8_OPER_GATE_STATE(x)        (((x) << 16) & GENMASK(23, 16))
+#define QSYS_PARAM_STATUS_REG_8_OPER_GATE_STATE_M         GENMASK(23, 16)
+#define QSYS_PARAM_STATUS_REG_8_OPER_GATE_STATE_X(x)      (((x) & GENMASK(23, 16)) >> 16)
+#define QSYS_PARAM_STATUS_REG_8_CONFIG_PENDING            BIT(24)
+
+#define QSYS_GCL_STATUS_REG_1_GCL_ENTRY_NUM(x)            ((x) & GENMASK(5, 0))
+#define QSYS_GCL_STATUS_REG_1_GCL_ENTRY_NUM_M             GENMASK(5, 0)
+#define QSYS_GCL_STATUS_REG_1_GATE_STATE(x)               (((x) << 8) & GENMASK(15, 8))
+#define QSYS_GCL_STATUS_REG_1_GATE_STATE_M                GENMASK(15, 8)
+#define QSYS_GCL_STATUS_REG_1_GATE_STATE_X(x)             (((x) & GENMASK(15, 8)) >> 8)
+
+#endif
-- 
cgit v1.2.3-59-g8ed1b


From bdeced75b13f8a0fc8e32b70e517a8dbb7d51738 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Mon, 6 Jan 2020 03:34:17 +0200
Subject: net: dsa: felix: Add PCS operations for PHYLINK

Layerscape SoCs traditionally expose the SerDes configuration/status for
Ethernet protocols (PCS for SGMII/USXGMII/10GBase-R etc etc) in a register
format that is compatible with clause 22 or clause 45 (depending on
SerDes protocol). Each MAC has its own internal MDIO bus on which there
is one or more of these PCS's, responding to commands at a configurable
PHY address. The per-port internal MDIO bus (which is just for PCSs) is
totally separate and has nothing to do with the dedicated external MDIO
controller (which is just for PHYs), but the register map for the MDIO
controller is the same.

The VSC9959 (Felix) switch instantiated in the LS1028A is integrated
in hardware with the ENETC PCS of its DSA master, and reuses its MDIO
controller driver, so Felix has been made to depend on it in Kconfig.

 +------------------------------------------------------------------------+
 |                   +--------+ GMII (typically disabled via RCW)         |
 | ENETC PCI         |  ENETC |--------------------------+                |
 | Root Complex      | port 3 |-----------------------+  |                |
 | Integrated        +--------+                       |  |                |
 | Endpoint                                           |  |                |
 |                   +--------+ 2.5G GMII             |  |                |
 |                   |  ENETC |--------------+        |  |                |
 |                   | port 2 |-----------+  |        |  |                |
 |                   +--------+           |  |        |  |                |
 |                                     +--------+  +--------+             |
 |                                     |  Felix |  |  Felix |             |
 |                                     | port 4 |  | port 5 |             |
 |                                     +--------+  +--------+             |
 |                                                                        |
 | +--------+  +--------+  +--------+  +--------+  +--------+  +--------+ |
 | |  ENETC |  |  ENETC |  |  Felix |  |  Felix |  |  Felix |  |  Felix | |
 | | port 0 |  | port 1 |  | port 0 |  | port 1 |  | port 2 |  | port 3 | |
 +------------------------------------------------------------------------+
 |    ||||  SerDes |          ||||        ||||        ||||        ||||    |
 | +--------+block |       +--------------------------------------------+ |
 | |  ENETC |      |       |       ENETC port 2 internal MDIO bus       | |
 | | port 0 |      |       |  PCS         PCS          PCS        PCS   | |
 | |   PCS  |      |       |   0           1            2          3    | |
 +-----------------|------------------------------------------------------+
        v          v           v           v            v          v
     SGMII/      RGMII    QSGMII/QSXGMII/4xSGMII/4x1000Base-X/4x2500Base-X
    USXGMII/   (bypasses
  1000Base-X/   SerDes)
  2500Base-X

In the LS1028A SoC described above, the VSC9959 Felix switch is PF5 of
the ENETC root complex, and has 2 BARs:
- BAR 4: the switch's effective registers
- BAR 0: the MDIO controller register map lended from ENETC port 2
         (PF2), for accessing its associated PCS's.

This explanation is necessary because the patch does some renaming
"pci_bar" -> "switch_pci_bar" for clarity, which would otherwise appear
a bit obtuse.

The fact that the internal MDIO bus is "borrowed" is relevant because
the register map is found in PF5 (the switch) but it triggers an access
fault if PF2 (the ENETC DSA master) is not enabled. This is not treated
in any way (and I don't think it can be treated).

All of this is so SoC-specific, that it was contained as much as
possible in the platform-integration file felix_vsc9959.c.

We need to parse and pre-validate the device tree because of 2 reasons:
- The PHY mode (SerDes protocol) cannot change at runtime due to SoC
  design.
- There is a circular dependency in that we need to know what clause the
  PCS speaks in order to find it on the internal MDIO bus. But the
  clause of the PCS depends on what phy-mode it is configured for.

The goal of this patch is to make steps towards removing the bootloader
dependency for SGMII PCS pre-configuration, as well as to add support
for monitoring the in-band SGMII AN between the PCS and the system-side
link partner (PHY or other MAC).

In practice the bootloader dependency is not completely removed. U-Boot
pre-programs the PHY address at which each PCS can be found on the
internal MDIO bus (MDEV_PORT). This is needed because the PCS of each
port has the same out-of-reset PHY address of zero. The SerDes register
for changing MDEV_PORT is pretty deep in the SoC (outside the addresses
of the ENETC PCI BARs) and therefore inaccessible to us from here.

Felix VSC9959 and Ocelot VSC7514 are integrated very differently in
their respective SoCs, and for that reason Felix does not use the Ocelot
core library for PHYLINK. On one hand we don't want to impose the
fixed phy-mode limitation to Ocelot, and on the other hand Felix doesn't
need to force the MAC link speed the way Ocelot does, since the MAC is
connected to the PCS through a fixed GMII, and the PCS is the one who
does the rate adaptation at lower link speeds, which the MAC does not
even need to know about. In fact changing the GMII speed for Felix
irrecoverably breaks transmission through that port until a reset.

The pair with ENETC port 3 and Felix port 5 is optional and doesn't
support tagging. When we enable it, swp5 is a regular slave port, albeit
an internal one. The trouble is that it doesn't work, and that is
because the DSA PHYLIB adaptation layer doesn't treat fixed-link slave
ports. So that is yet another reason for wanting to convert Felix to the
native PHYLINK API.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/ocelot/Kconfig         |   2 +
 drivers/net/dsa/ocelot/felix.c         | 265 ++++++++++++++++-
 drivers/net/dsa/ocelot/felix.h         |  16 +-
 drivers/net/dsa/ocelot/felix_vsc9959.c | 501 ++++++++++++++++++++++++++++++++-
 4 files changed, 767 insertions(+), 17 deletions(-)

diff --git a/drivers/net/dsa/ocelot/Kconfig b/drivers/net/dsa/ocelot/Kconfig
index 6f9804093150..a5b7cca03d09 100644
--- a/drivers/net/dsa/ocelot/Kconfig
+++ b/drivers/net/dsa/ocelot/Kconfig
@@ -3,8 +3,10 @@ config NET_DSA_MSCC_FELIX
 	tristate "Ocelot / Felix Ethernet switch support"
 	depends on NET_DSA && PCI
 	depends on NET_VENDOR_MICROSEMI
+	depends on NET_VENDOR_FREESCALE
 	select MSCC_OCELOT_SWITCH
 	select NET_DSA_TAG_OCELOT
+	select FSL_ENETC_MDIO
 	help
 	  This driver supports the VSC9959 network switch, which is a member of
 	  the Vitesse / Microsemi / Microchip Ocelot family of switching cores.
diff --git a/drivers/net/dsa/ocelot/felix.c b/drivers/net/dsa/ocelot/felix.c
index b7f92464815d..f072dd75cea2 100644
--- a/drivers/net/dsa/ocelot/felix.c
+++ b/drivers/net/dsa/ocelot/felix.c
@@ -2,9 +2,14 @@
 /* Copyright 2019 NXP Semiconductors
  */
 #include <uapi/linux/if_bridge.h>
+#include <soc/mscc/ocelot_qsys.h>
+#include <soc/mscc/ocelot_sys.h>
+#include <soc/mscc/ocelot_dev.h>
+#include <soc/mscc/ocelot_ana.h>
 #include <soc/mscc/ocelot.h>
 #include <linux/packing.h>
 #include <linux/module.h>
+#include <linux/of_net.h>
 #include <linux/pci.h>
 #include <linux/of.h>
 #include <net/dsa.h>
@@ -26,14 +31,6 @@ static int felix_set_ageing_time(struct dsa_switch *ds,
 	return 0;
 }
 
-static void felix_adjust_link(struct dsa_switch *ds, int port,
-			      struct phy_device *phydev)
-{
-	struct ocelot *ocelot = ds->priv;
-
-	ocelot_adjust_link(ocelot, port, phydev);
-}
-
 static int felix_fdb_dump(struct dsa_switch *ds, int port,
 			  dsa_fdb_dump_cb_t *cb, void *data)
 {
@@ -155,6 +152,138 @@ static void felix_port_disable(struct dsa_switch *ds, int port)
 	return ocelot_port_disable(ocelot, port);
 }
 
+static void felix_phylink_validate(struct dsa_switch *ds, int port,
+				   unsigned long *supported,
+				   struct phylink_link_state *state)
+{
+	struct ocelot *ocelot = ds->priv;
+	struct ocelot_port *ocelot_port = ocelot->ports[port];
+	__ETHTOOL_DECLARE_LINK_MODE_MASK(mask) = { 0, };
+
+	if (state->interface != PHY_INTERFACE_MODE_NA &&
+	    state->interface != ocelot_port->phy_mode) {
+		bitmap_zero(supported, __ETHTOOL_LINK_MODE_MASK_NBITS);
+		return;
+	}
+
+	/* No half-duplex. */
+	phylink_set_port_modes(mask);
+	phylink_set(mask, Autoneg);
+	phylink_set(mask, Pause);
+	phylink_set(mask, Asym_Pause);
+	if (state->interface != PHY_INTERFACE_MODE_2500BASEX) {
+		phylink_set(mask, 10baseT_Full);
+		phylink_set(mask, 100baseT_Full);
+		phylink_set(mask, 1000baseT_Full);
+	}
+	/* The internal ports that run at 2.5G are overclocked GMII */
+	if (state->interface == PHY_INTERFACE_MODE_GMII ||
+	    state->interface == PHY_INTERFACE_MODE_2500BASEX ||
+	    state->interface == PHY_INTERFACE_MODE_USXGMII) {
+		phylink_set(mask, 2500baseT_Full);
+		phylink_set(mask, 2500baseX_Full);
+	}
+
+	bitmap_and(supported, supported, mask,
+		   __ETHTOOL_LINK_MODE_MASK_NBITS);
+	bitmap_and(state->advertising, state->advertising, mask,
+		   __ETHTOOL_LINK_MODE_MASK_NBITS);
+}
+
+static int felix_phylink_mac_pcs_get_state(struct dsa_switch *ds, int port,
+					   struct phylink_link_state *state)
+{
+	struct ocelot *ocelot = ds->priv;
+	struct felix *felix = ocelot_to_felix(ocelot);
+
+	if (felix->info->pcs_link_state)
+		felix->info->pcs_link_state(ocelot, port, state);
+
+	return 0;
+}
+
+static void felix_phylink_mac_config(struct dsa_switch *ds, int port,
+				     unsigned int link_an_mode,
+				     const struct phylink_link_state *state)
+{
+	struct ocelot *ocelot = ds->priv;
+	struct ocelot_port *ocelot_port = ocelot->ports[port];
+	struct felix *felix = ocelot_to_felix(ocelot);
+	u32 mac_fc_cfg;
+
+	/* Take port out of reset by clearing the MAC_TX_RST, MAC_RX_RST and
+	 * PORT_RST bits in CLOCK_CFG
+	 */
+	ocelot_port_writel(ocelot_port, DEV_CLOCK_CFG_LINK_SPEED(state->speed),
+			   DEV_CLOCK_CFG);
+
+	/* Flow control. Link speed is only used here to evaluate the time
+	 * specification in incoming pause frames.
+	 */
+	mac_fc_cfg = SYS_MAC_FC_CFG_FC_LINK_SPEED(state->speed);
+	if (state->pause & MLO_PAUSE_RX)
+		mac_fc_cfg |= SYS_MAC_FC_CFG_RX_FC_ENA;
+	if (state->pause & MLO_PAUSE_TX)
+		mac_fc_cfg |= SYS_MAC_FC_CFG_TX_FC_ENA |
+			      SYS_MAC_FC_CFG_PAUSE_VAL_CFG(0xffff) |
+			      SYS_MAC_FC_CFG_FC_LATENCY_CFG(0x7) |
+			      SYS_MAC_FC_CFG_ZERO_PAUSE_ENA;
+	ocelot_write_rix(ocelot, mac_fc_cfg, SYS_MAC_FC_CFG, port);
+
+	ocelot_write_rix(ocelot, 0, ANA_POL_FLOWC, port);
+
+	if (felix->info->pcs_init)
+		felix->info->pcs_init(ocelot, port, link_an_mode, state);
+}
+
+static void felix_phylink_mac_an_restart(struct dsa_switch *ds, int port)
+{
+	struct ocelot *ocelot = ds->priv;
+	struct felix *felix = ocelot_to_felix(ocelot);
+
+	if (felix->info->pcs_an_restart)
+		felix->info->pcs_an_restart(ocelot, port);
+}
+
+static void felix_phylink_mac_link_down(struct dsa_switch *ds, int port,
+					unsigned int link_an_mode,
+					phy_interface_t interface)
+{
+	struct ocelot *ocelot = ds->priv;
+	struct ocelot_port *ocelot_port = ocelot->ports[port];
+
+	ocelot_port_writel(ocelot_port, 0, DEV_MAC_ENA_CFG);
+	ocelot_rmw_rix(ocelot, 0, QSYS_SWITCH_PORT_MODE_PORT_ENA,
+		       QSYS_SWITCH_PORT_MODE, port);
+}
+
+static void felix_phylink_mac_link_up(struct dsa_switch *ds, int port,
+				      unsigned int link_an_mode,
+				      phy_interface_t interface,
+				      struct phy_device *phydev)
+{
+	struct ocelot *ocelot = ds->priv;
+	struct ocelot_port *ocelot_port = ocelot->ports[port];
+
+	/* Enable MAC module */
+	ocelot_port_writel(ocelot_port, DEV_MAC_ENA_CFG_RX_ENA |
+			   DEV_MAC_ENA_CFG_TX_ENA, DEV_MAC_ENA_CFG);
+
+	/* Enable receiving frames on the port, and activate auto-learning of
+	 * MAC addresses.
+	 */
+	ocelot_write_gix(ocelot, ANA_PORT_PORT_CFG_LEARNAUTO |
+			 ANA_PORT_PORT_CFG_RECV_ENA |
+			 ANA_PORT_PORT_CFG_PORTID_VAL(port),
+			 ANA_PORT_PORT_CFG, port);
+
+	/* Core: Enable port for frame transfer */
+	ocelot_write_rix(ocelot, QSYS_SWITCH_PORT_MODE_INGRESS_DROP_MODE |
+			 QSYS_SWITCH_PORT_MODE_SCH_NEXT_CFG(1) |
+			 QSYS_SWITCH_PORT_MODE_PORT_ENA,
+			 QSYS_SWITCH_PORT_MODE, port);
+}
+
 static void felix_get_strings(struct dsa_switch *ds, int port,
 			      u32 stringset, u8 *data)
 {
@@ -185,10 +314,76 @@ static int felix_get_ts_info(struct dsa_switch *ds, int port,
 	return ocelot_get_ts_info(ocelot, port, info);
 }
 
+static int felix_parse_ports_node(struct felix *felix,
+				  struct device_node *ports_node,
+				  phy_interface_t *port_phy_modes)
+{
+	struct ocelot *ocelot = &felix->ocelot;
+	struct device *dev = felix->ocelot.dev;
+	struct device_node *child;
+
+	for_each_child_of_node(ports_node, child) {
+		phy_interface_t phy_mode;
+		u32 port;
+		int err;
+
+		/* Get switch port number from DT */
+		if (of_property_read_u32(child, "reg", &port) < 0) {
+			dev_err(dev, "Port number not defined in device tree "
+				"(property \"reg\")\n");
+			of_node_put(child);
+			return -ENODEV;
+		}
+
+		/* Get PHY mode from DT */
+		err = of_get_phy_mode(child, &phy_mode);
+		if (err) {
+			dev_err(dev, "Failed to read phy-mode or "
+				"phy-interface-type property for port %d\n",
+				port);
+			of_node_put(child);
+			return -ENODEV;
+		}
+
+		err = felix->info->prevalidate_phy_mode(ocelot, port, phy_mode);
+		if (err < 0) {
+			dev_err(dev, "Unsupported PHY mode %s on port %d\n",
+				phy_modes(phy_mode), port);
+			return err;
+		}
+
+		port_phy_modes[port] = phy_mode;
+	}
+
+	return 0;
+}
+
+static int felix_parse_dt(struct felix *felix, phy_interface_t *port_phy_modes)
+{
+	struct device *dev = felix->ocelot.dev;
+	struct device_node *switch_node;
+	struct device_node *ports_node;
+	int err;
+
+	switch_node = dev->of_node;
+
+	ports_node = of_get_child_by_name(switch_node, "ports");
+	if (!ports_node) {
+		dev_err(dev, "Incorrect bindings: absent \"ports\" node\n");
+		return -ENODEV;
+	}
+
+	err = felix_parse_ports_node(felix, ports_node, port_phy_modes);
+	of_node_put(ports_node);
+
+	return err;
+}
+
 static int felix_init_structs(struct felix *felix, int num_phys_ports)
 {
 	struct ocelot *ocelot = &felix->ocelot;
-	resource_size_t base;
+	phy_interface_t *port_phy_modes;
+	resource_size_t switch_base;
 	int port, i, err;
 
 	ocelot->num_phys_ports = num_phys_ports;
@@ -203,7 +398,19 @@ static int felix_init_structs(struct felix *felix, int num_phys_ports)
 	ocelot->shared_queue_sz	= felix->info->shared_queue_sz;
 	ocelot->ops		= felix->info->ops;
 
-	base = pci_resource_start(felix->pdev, felix->info->pci_bar);
+	port_phy_modes = kcalloc(num_phys_ports, sizeof(phy_interface_t),
+				 GFP_KERNEL);
+	if (!port_phy_modes)
+		return -ENOMEM;
+
+	err = felix_parse_dt(felix, port_phy_modes);
+	if (err) {
+		kfree(port_phy_modes);
+		return err;
+	}
+
+	switch_base = pci_resource_start(felix->pdev,
+					 felix->info->switch_pci_bar);
 
 	for (i = 0; i < TARGET_MAX; i++) {
 		struct regmap *target;
@@ -214,13 +421,14 @@ static int felix_init_structs(struct felix *felix, int num_phys_ports)
 
 		res = &felix->info->target_io_res[i];
 		res->flags = IORESOURCE_MEM;
-		res->start += base;
-		res->end += base;
+		res->start += switch_base;
+		res->end += switch_base;
 
 		target = ocelot_regmap_init(ocelot, res);
 		if (IS_ERR(target)) {
 			dev_err(ocelot->dev,
 				"Failed to map device memory space\n");
+			kfree(port_phy_modes);
 			return PTR_ERR(target);
 		}
 
@@ -230,6 +438,7 @@ static int felix_init_structs(struct felix *felix, int num_phys_ports)
 	err = ocelot_regfields_init(ocelot, felix->info->regfields);
 	if (err) {
 		dev_err(ocelot->dev, "failed to init reg fields map\n");
+		kfree(port_phy_modes);
 		return err;
 	}
 
@@ -244,26 +453,37 @@ static int felix_init_structs(struct felix *felix, int num_phys_ports)
 		if (!ocelot_port) {
 			dev_err(ocelot->dev,
 				"failed to allocate port memory\n");
+			kfree(port_phy_modes);
 			return -ENOMEM;
 		}
 
 		res = &felix->info->port_io_res[port];
 		res->flags = IORESOURCE_MEM;
-		res->start += base;
-		res->end += base;
+		res->start += switch_base;
+		res->end += switch_base;
 
 		port_regs = devm_ioremap_resource(ocelot->dev, res);
 		if (IS_ERR(port_regs)) {
 			dev_err(ocelot->dev,
 				"failed to map registers for port %d\n", port);
+			kfree(port_phy_modes);
 			return PTR_ERR(port_regs);
 		}
 
+		ocelot_port->phy_mode = port_phy_modes[port];
 		ocelot_port->ocelot = ocelot;
 		ocelot_port->regs = port_regs;
 		ocelot->ports[port] = ocelot_port;
 	}
 
+	kfree(port_phy_modes);
+
+	if (felix->info->mdio_bus_alloc) {
+		err = felix->info->mdio_bus_alloc(ocelot);
+		if (err < 0)
+			return err;
+	}
+
 	return 0;
 }
 
@@ -293,12 +513,22 @@ static int felix_setup(struct dsa_switch *ds)
 					    OCELOT_TAG_PREFIX_LONG);
 	}
 
+	/* It looks like the MAC/PCS interrupt register - PM0_IEVENT (0x8040)
+	 * isn't instantiated for the Felix PF.
+	 * In-band AN may take a few ms to complete, so we need to poll.
+	 */
+	ds->pcs_poll = true;
+
 	return 0;
 }
 
 static void felix_teardown(struct dsa_switch *ds)
 {
 	struct ocelot *ocelot = ds->priv;
+	struct felix *felix = ocelot_to_felix(ocelot);
+
+	if (felix->info->mdio_bus_free)
+		felix->info->mdio_bus_free(ocelot);
 
 	/* stop workqueue thread */
 	ocelot_deinit(ocelot);
@@ -369,7 +599,12 @@ static const struct dsa_switch_ops felix_switch_ops = {
 	.get_ethtool_stats	= felix_get_ethtool_stats,
 	.get_sset_count		= felix_get_sset_count,
 	.get_ts_info		= felix_get_ts_info,
-	.adjust_link		= felix_adjust_link,
+	.phylink_validate	= felix_phylink_validate,
+	.phylink_mac_link_state	= felix_phylink_mac_pcs_get_state,
+	.phylink_mac_config	= felix_phylink_mac_config,
+	.phylink_mac_an_restart	= felix_phylink_mac_an_restart,
+	.phylink_mac_link_down	= felix_phylink_mac_link_down,
+	.phylink_mac_link_up	= felix_phylink_mac_link_up,
 	.port_enable		= felix_port_enable,
 	.port_disable		= felix_port_disable,
 	.port_fdb_dump		= felix_fdb_dump,
diff --git a/drivers/net/dsa/ocelot/felix.h b/drivers/net/dsa/ocelot/felix.h
index 204296e51d0c..3a7580015b62 100644
--- a/drivers/net/dsa/ocelot/felix.h
+++ b/drivers/net/dsa/ocelot/felix.h
@@ -10,6 +10,7 @@
 struct felix_info {
 	struct resource			*target_io_res;
 	struct resource			*port_io_res;
+	struct resource			*imdio_res;
 	const struct reg_field		*regfields;
 	const u32 *const		*map;
 	const struct ocelot_ops		*ops;
@@ -17,7 +18,18 @@ struct felix_info {
 	const struct ocelot_stat_layout	*stats_layout;
 	unsigned int			num_stats;
 	int				num_ports;
-	int				pci_bar;
+	int				switch_pci_bar;
+	int				imdio_pci_bar;
+	int	(*mdio_bus_alloc)(struct ocelot *ocelot);
+	void	(*mdio_bus_free)(struct ocelot *ocelot);
+	void	(*pcs_init)(struct ocelot *ocelot, int port,
+			    unsigned int link_an_mode,
+			    const struct phylink_link_state *state);
+	void	(*pcs_an_restart)(struct ocelot *ocelot, int port);
+	void	(*pcs_link_state)(struct ocelot *ocelot, int port,
+				  struct phylink_link_state *state);
+	int	(*prevalidate_phy_mode)(struct ocelot *ocelot, int port,
+					phy_interface_t phy_mode);
 };
 
 extern struct felix_info		felix_info_vsc9959;
@@ -32,6 +44,8 @@ struct felix {
 	struct pci_dev			*pdev;
 	struct felix_info		*info;
 	struct ocelot			ocelot;
+	struct mii_bus			*imdio;
+	struct phy_device		**pcs;
 };
 
 #endif
diff --git a/drivers/net/dsa/ocelot/felix_vsc9959.c b/drivers/net/dsa/ocelot/felix_vsc9959.c
index b9758b0d18c7..03482616faa7 100644
--- a/drivers/net/dsa/ocelot/felix_vsc9959.c
+++ b/drivers/net/dsa/ocelot/felix_vsc9959.c
@@ -2,12 +2,33 @@
 /* Copyright 2017 Microsemi Corporation
  * Copyright 2018-2019 NXP Semiconductors
  */
+#include <linux/fsl/enetc_mdio.h>
 #include <soc/mscc/ocelot_sys.h>
 #include <soc/mscc/ocelot.h>
 #include <linux/iopoll.h>
 #include <linux/pci.h>
 #include "felix.h"
 
+/* TODO: should find a better place for these */
+#define USXGMII_BMCR_RESET		BIT(15)
+#define USXGMII_BMCR_AN_EN		BIT(12)
+#define USXGMII_BMCR_RST_AN		BIT(9)
+#define USXGMII_BMSR_LNKS(status)	(((status) & GENMASK(2, 2)) >> 2)
+#define USXGMII_BMSR_AN_CMPL(status)	(((status) & GENMASK(5, 5)) >> 5)
+#define USXGMII_ADVERTISE_LNKS(x)	(((x) << 15) & BIT(15))
+#define USXGMII_ADVERTISE_FDX		BIT(12)
+#define USXGMII_ADVERTISE_SPEED(x)	(((x) << 9) & GENMASK(11, 9))
+#define USXGMII_LPA_LNKS(lpa)		((lpa) >> 15)
+#define USXGMII_LPA_DUPLEX(lpa)		(((lpa) & GENMASK(12, 12)) >> 12)
+#define USXGMII_LPA_SPEED(lpa)		(((lpa) & GENMASK(11, 9)) >> 9)
+
+enum usxgmii_speed {
+	USXGMII_SPEED_10	= 0,
+	USXGMII_SPEED_100	= 1,
+	USXGMII_SPEED_1000	= 2,
+	USXGMII_SPEED_2500	= 4,
+};
+
 static const u32 vsc9959_ana_regmap[] = {
 	REG(ANA_ADVLEARN,			0x0089a0),
 	REG(ANA_VLANMASK,			0x0089a4),
@@ -386,6 +407,15 @@ static struct resource vsc9959_port_io_res[] = {
 	},
 };
 
+/* Port MAC 0 Internal MDIO bus through which the SerDes acting as an
+ * SGMII/QSGMII MAC PCS can be found.
+ */
+static struct resource vsc9959_imdio_res = {
+	.start		= 0x8030,
+	.end		= 0x8040,
+	.name		= "imdio",
+};
+
 static const struct reg_field vsc9959_regfields[] = {
 	[ANA_ADVLEARN_VLAN_CHK] = REG_FIELD(ANA_ADVLEARN, 6, 6),
 	[ANA_ADVLEARN_LEARN_MIRROR] = REG_FIELD(ANA_ADVLEARN, 0, 5),
@@ -565,13 +595,475 @@ static int vsc9959_reset(struct ocelot *ocelot)
 	return 0;
 }
 
+static void vsc9959_pcs_an_restart_sgmii(struct phy_device *pcs)
+{
+	phy_set_bits(pcs, MII_BMCR, BMCR_ANRESTART);
+}
+
+static void vsc9959_pcs_an_restart_usxgmii(struct phy_device *pcs)
+{
+	phy_write_mmd(pcs, MDIO_MMD_VEND2, MII_BMCR,
+		      USXGMII_BMCR_RESET |
+		      USXGMII_BMCR_AN_EN |
+		      USXGMII_BMCR_RST_AN);
+}
+
+static void vsc9959_pcs_an_restart(struct ocelot *ocelot, int port)
+{
+	struct felix *felix = ocelot_to_felix(ocelot);
+	struct phy_device *pcs = felix->pcs[port];
+
+	if (!pcs)
+		return;
+
+	switch (pcs->interface) {
+	case PHY_INTERFACE_MODE_SGMII:
+	case PHY_INTERFACE_MODE_QSGMII:
+		vsc9959_pcs_an_restart_sgmii(pcs);
+		break;
+	case PHY_INTERFACE_MODE_USXGMII:
+		vsc9959_pcs_an_restart_usxgmii(pcs);
+		break;
+	default:
+		dev_err(ocelot->dev, "Invalid PCS interface type %s\n",
+			phy_modes(pcs->interface));
+		break;
+	}
+}
+
+/* We enable SGMII AN only when the PHY has managed = "in-band-status" in the
+ * device tree. If we are in MLO_AN_PHY mode, we program directly state->speed
+ * into the PCS, which is retrieved out-of-band over MDIO. This also has the
+ * benefit of working with SGMII fixed-links, like downstream switches, where
+ * both link partners attempt to operate as AN slaves and therefore AN never
+ * completes.  But it also has the disadvantage that some PHY chips don't pass
+ * traffic if SGMII AN is enabled but not completed (acknowledged by us), so
+ * setting MLO_AN_INBAND is actually required for those.
+ */
+static void vsc9959_pcs_init_sgmii(struct phy_device *pcs,
+				   unsigned int link_an_mode,
+				   const struct phylink_link_state *state)
+{
+	if (link_an_mode == MLO_AN_INBAND) {
+		/* SGMII spec requires tx_config_Reg[15:0] to be exactly 0x4001
+		 * for the MAC PCS in order to acknowledge the AN.
+		 */
+		phy_write(pcs, MII_ADVERTISE, ADVERTISE_SGMII |
+					      ADVERTISE_LPACK);
+
+		phy_write(pcs, ENETC_PCS_IF_MODE,
+			  ENETC_PCS_IF_MODE_SGMII_EN |
+			  ENETC_PCS_IF_MODE_USE_SGMII_AN);
+
+		/* Adjust link timer for SGMII */
+		phy_write(pcs, ENETC_PCS_LINK_TIMER1,
+			  ENETC_PCS_LINK_TIMER1_VAL);
+		phy_write(pcs, ENETC_PCS_LINK_TIMER2,
+			  ENETC_PCS_LINK_TIMER2_VAL);
+
+		phy_write(pcs, MII_BMCR, BMCR_ANRESTART | BMCR_ANENABLE);
+	} else {
+		int speed;
+
+		if (state->duplex == DUPLEX_HALF) {
+			phydev_err(pcs, "Half duplex not supported\n");
+			return;
+		}
+		switch (state->speed) {
+		case SPEED_1000:
+			speed = ENETC_PCS_SPEED_1000;
+			break;
+		case SPEED_100:
+			speed = ENETC_PCS_SPEED_100;
+			break;
+		case SPEED_10:
+			speed = ENETC_PCS_SPEED_10;
+			break;
+		case SPEED_UNKNOWN:
+			/* Silently don't do anything */
+			return;
+		default:
+			phydev_err(pcs, "Invalid PCS speed %d\n", state->speed);
+			return;
+		}
+
+		phy_write(pcs, ENETC_PCS_IF_MODE,
+			  ENETC_PCS_IF_MODE_SGMII_EN |
+			  ENETC_PCS_IF_MODE_SGMII_SPEED(speed));
+
+		/* Yes, not a mistake: speed is given by IF_MODE. */
+		phy_write(pcs, MII_BMCR, BMCR_RESET |
+					 BMCR_SPEED1000 |
+					 BMCR_FULLDPLX);
+	}
+}
+
+/* 2500Base-X is SerDes protocol 7 on Felix and 6 on ENETC. It is a SerDes lane
+ * clocked at 3.125 GHz which encodes symbols with 8b/10b and does not have
+ * auto-negotiation of any link parameters. Electrically it is compatible with
+ * a single lane of XAUI.
+ * The hardware reference manual wants to call this mode SGMII, but it isn't
+ * really, since the fundamental features of SGMII:
+ * - Downgrading the link speed by duplicating symbols
+ * - Auto-negotiation
+ * are not there.
+ * The speed is configured at 1000 in the IF_MODE and BMCR MDIO registers
+ * because the clock frequency is actually given by a PLL configured in the
+ * Reset Configuration Word (RCW).
+ * Since there is no difference between fixed speed SGMII w/o AN and 802.3z w/o
+ * AN, we call this PHY interface type 2500Base-X. In case a PHY negotiates a
+ * lower link speed on line side, the system-side interface remains fixed at
+ * 2500 Mbps and we do rate adaptation through pause frames.
+ */
+static void vsc9959_pcs_init_2500basex(struct phy_device *pcs,
+				       unsigned int link_an_mode,
+				       const struct phylink_link_state *state)
+{
+	if (link_an_mode == MLO_AN_INBAND) {
+		phydev_err(pcs, "AN not supported on 3.125GHz SerDes lane\n");
+		return;
+	}
+
+	phy_write(pcs, ENETC_PCS_IF_MODE,
+		  ENETC_PCS_IF_MODE_SGMII_EN |
+		  ENETC_PCS_IF_MODE_SGMII_SPEED(ENETC_PCS_SPEED_2500));
+
+	phy_write(pcs, MII_BMCR, BMCR_SPEED1000 |
+				 BMCR_FULLDPLX |
+				 BMCR_RESET);
+}
+
+static void vsc9959_pcs_init_usxgmii(struct phy_device *pcs,
+				     unsigned int link_an_mode,
+				     const struct phylink_link_state *state)
+{
+	if (link_an_mode != MLO_AN_INBAND) {
+		phydev_err(pcs, "USXGMII only supports in-band AN for now\n");
+		return;
+	}
+
+	/* Configure device ability for the USXGMII Replicator */
+	phy_write_mmd(pcs, MDIO_MMD_VEND2, MII_ADVERTISE,
+		      USXGMII_ADVERTISE_SPEED(USXGMII_SPEED_2500) |
+		      USXGMII_ADVERTISE_LNKS(1) |
+		      ADVERTISE_SGMII |
+		      ADVERTISE_LPACK |
+		      USXGMII_ADVERTISE_FDX);
+}
+
+static void vsc9959_pcs_init(struct ocelot *ocelot, int port,
+			     unsigned int link_an_mode,
+			     const struct phylink_link_state *state)
+{
+	struct felix *felix = ocelot_to_felix(ocelot);
+	struct phy_device *pcs = felix->pcs[port];
+
+	if (!pcs)
+		return;
+
+	/* The PCS does not implement the BMSR register fully, so capability
+	 * detection via genphy_read_abilities does not work. Since we can get
+	 * the PHY config word from the LPA register though, there is still
+	 * value in using the generic phy_resolve_aneg_linkmode function. So
+	 * populate the supported and advertising link modes manually here.
+	 */
+	linkmode_set_bit_array(phy_basic_ports_array,
+			       ARRAY_SIZE(phy_basic_ports_array),
+			       pcs->supported);
+	linkmode_set_bit(ETHTOOL_LINK_MODE_10baseT_Full_BIT, pcs->supported);
+	linkmode_set_bit(ETHTOOL_LINK_MODE_100baseT_Full_BIT, pcs->supported);
+	linkmode_set_bit(ETHTOOL_LINK_MODE_1000baseT_Full_BIT, pcs->supported);
+	if (pcs->interface == PHY_INTERFACE_MODE_2500BASEX ||
+	    pcs->interface == PHY_INTERFACE_MODE_USXGMII)
+		linkmode_set_bit(ETHTOOL_LINK_MODE_2500baseX_Full_BIT,
+				 pcs->supported);
+	if (pcs->interface != PHY_INTERFACE_MODE_2500BASEX)
+		linkmode_set_bit(ETHTOOL_LINK_MODE_Autoneg_BIT,
+				 pcs->supported);
+	phy_advertise_supported(pcs);
+
+	switch (pcs->interface) {
+	case PHY_INTERFACE_MODE_SGMII:
+	case PHY_INTERFACE_MODE_QSGMII:
+		vsc9959_pcs_init_sgmii(pcs, link_an_mode, state);
+		break;
+	case PHY_INTERFACE_MODE_2500BASEX:
+		vsc9959_pcs_init_2500basex(pcs, link_an_mode, state);
+		break;
+	case PHY_INTERFACE_MODE_USXGMII:
+		vsc9959_pcs_init_usxgmii(pcs, link_an_mode, state);
+		break;
+	default:
+		dev_err(ocelot->dev, "Unsupported link mode %s\n",
+			phy_modes(pcs->interface));
+	}
+}
+
+static void vsc9959_pcs_link_state_resolve(struct phy_device *pcs,
+					   struct phylink_link_state *state)
+{
+	state->an_complete = pcs->autoneg_complete;
+	state->an_enabled = pcs->autoneg;
+	state->link = pcs->link;
+	state->duplex = pcs->duplex;
+	state->speed = pcs->speed;
+	/* SGMII AN does not negotiate flow control, but that's ok,
+	 * since phylink already knows that, and does:
+	 *	link_state.pause |= pl->phy_state.pause;
+	 */
+	state->pause = MLO_PAUSE_NONE;
+
+	phydev_dbg(pcs,
+		   "mode=%s/%s/%s adv=%*pb lpa=%*pb link=%u an_enabled=%u an_complete=%u\n",
+		   phy_modes(pcs->interface),
+		   phy_speed_to_str(pcs->speed),
+		   phy_duplex_to_str(pcs->duplex),
+		   __ETHTOOL_LINK_MODE_MASK_NBITS, pcs->advertising,
+		   __ETHTOOL_LINK_MODE_MASK_NBITS, pcs->lp_advertising,
+		   pcs->link, pcs->autoneg, pcs->autoneg_complete);
+}
+
+static void vsc9959_pcs_link_state_sgmii(struct phy_device *pcs,
+					 struct phylink_link_state *state)
+{
+	int err;
+
+	err = genphy_update_link(pcs);
+	if (err < 0)
+		return;
+
+	if (pcs->autoneg_complete) {
+		u16 lpa = phy_read(pcs, MII_LPA);
+
+		mii_lpa_to_linkmode_lpa_sgmii(pcs->lp_advertising, lpa);
+
+		phy_resolve_aneg_linkmode(pcs);
+	}
+}
+
+static void vsc9959_pcs_link_state_2500basex(struct phy_device *pcs,
+					     struct phylink_link_state *state)
+{
+	int err;
+
+	err = genphy_update_link(pcs);
+	if (err < 0)
+		return;
+
+	pcs->speed = SPEED_2500;
+	pcs->asym_pause = true;
+	pcs->pause = true;
+}
+
+static void vsc9959_pcs_link_state_usxgmii(struct phy_device *pcs,
+					   struct phylink_link_state *state)
+{
+	int status, lpa;
+
+	status = phy_read_mmd(pcs, MDIO_MMD_VEND2, MII_BMSR);
+	if (status < 0)
+		return;
+
+	pcs->autoneg = true;
+	pcs->autoneg_complete = USXGMII_BMSR_AN_CMPL(status);
+	pcs->link = USXGMII_BMSR_LNKS(status);
+
+	if (!pcs->link || !pcs->autoneg_complete)
+		return;
+
+	lpa = phy_read_mmd(pcs, MDIO_MMD_VEND2, MII_LPA);
+	if (lpa < 0)
+		return;
+
+	switch (USXGMII_LPA_SPEED(lpa)) {
+	case USXGMII_SPEED_10:
+		pcs->speed = SPEED_10;
+		break;
+	case USXGMII_SPEED_100:
+		pcs->speed = SPEED_100;
+		break;
+	case USXGMII_SPEED_1000:
+		pcs->speed = SPEED_1000;
+		break;
+	case USXGMII_SPEED_2500:
+		pcs->speed = SPEED_2500;
+		break;
+	default:
+		break;
+	}
+
+	pcs->link = USXGMII_LPA_LNKS(lpa);
+	if (USXGMII_LPA_DUPLEX(lpa))
+		pcs->duplex = DUPLEX_FULL;
+	else
+		pcs->duplex = DUPLEX_HALF;
+}
+
+static void vsc9959_pcs_link_state(struct ocelot *ocelot, int port,
+				   struct phylink_link_state *state)
+{
+	struct felix *felix = ocelot_to_felix(ocelot);
+	struct phy_device *pcs = felix->pcs[port];
+
+	if (!pcs)
+		return;
+
+	pcs->speed = SPEED_UNKNOWN;
+	pcs->duplex = DUPLEX_UNKNOWN;
+	pcs->pause = 0;
+	pcs->asym_pause = 0;
+
+	switch (pcs->interface) {
+	case PHY_INTERFACE_MODE_SGMII:
+	case PHY_INTERFACE_MODE_QSGMII:
+		vsc9959_pcs_link_state_sgmii(pcs, state);
+		break;
+	case PHY_INTERFACE_MODE_2500BASEX:
+		vsc9959_pcs_link_state_2500basex(pcs, state);
+		break;
+	case PHY_INTERFACE_MODE_USXGMII:
+		vsc9959_pcs_link_state_usxgmii(pcs, state);
+		break;
+	default:
+		return;
+	}
+
+	vsc9959_pcs_link_state_resolve(pcs, state);
+}
+
+static int vsc9959_prevalidate_phy_mode(struct ocelot *ocelot, int port,
+					phy_interface_t phy_mode)
+{
+	switch (phy_mode) {
+	case PHY_INTERFACE_MODE_GMII:
+		/* Only supported on internal to-CPU ports */
+		if (port != 4 && port != 5)
+			return -ENOTSUPP;
+		return 0;
+	case PHY_INTERFACE_MODE_SGMII:
+	case PHY_INTERFACE_MODE_QSGMII:
+	case PHY_INTERFACE_MODE_USXGMII:
+	case PHY_INTERFACE_MODE_2500BASEX:
+		/* Not supported on internal to-CPU ports */
+		if (port == 4 || port == 5)
+			return -ENOTSUPP;
+		return 0;
+	default:
+		return -ENOTSUPP;
+	}
+}
+
 static const struct ocelot_ops vsc9959_ops = {
 	.reset			= vsc9959_reset,
 };
 
+static int vsc9959_mdio_bus_alloc(struct ocelot *ocelot)
+{
+	struct felix *felix = ocelot_to_felix(ocelot);
+	struct enetc_mdio_priv *mdio_priv;
+	struct device *dev = ocelot->dev;
+	resource_size_t imdio_base;
+	void __iomem *imdio_regs;
+	struct resource *res;
+	struct enetc_hw *hw;
+	struct mii_bus *bus;
+	int port;
+	int rc;
+
+	felix->pcs = devm_kcalloc(dev, felix->info->num_ports,
+				  sizeof(struct phy_device *),
+				  GFP_KERNEL);
+	if (!felix->pcs) {
+		dev_err(dev, "failed to allocate array for PCS PHYs\n");
+		return -ENOMEM;
+	}
+
+	imdio_base = pci_resource_start(felix->pdev,
+					felix->info->imdio_pci_bar);
+
+	res = felix->info->imdio_res;
+	res->flags = IORESOURCE_MEM;
+	res->start += imdio_base;
+	res->end += imdio_base;
+
+	imdio_regs = devm_ioremap_resource(dev, res);
+	if (IS_ERR(imdio_regs)) {
+		dev_err(dev, "failed to map internal MDIO registers\n");
+		return PTR_ERR(imdio_regs);
+	}
+
+	hw = enetc_hw_alloc(dev, imdio_regs);
+	if (IS_ERR(hw)) {
+		dev_err(dev, "failed to allocate ENETC HW structure\n");
+		return PTR_ERR(hw);
+	}
+
+	bus = devm_mdiobus_alloc_size(dev, sizeof(*mdio_priv));
+	if (!bus)
+		return -ENOMEM;
+
+	bus->name = "VSC9959 internal MDIO bus";
+	bus->read = enetc_mdio_read;
+	bus->write = enetc_mdio_write;
+	bus->parent = dev;
+	mdio_priv = bus->priv;
+	mdio_priv->hw = hw;
+	/* This gets added to imdio_regs, which already maps addresses
+	 * starting with the proper offset.
+	 */
+	mdio_priv->mdio_base = 0;
+	snprintf(bus->id, MII_BUS_ID_SIZE, "%s-imdio", dev_name(dev));
+
+	/* Needed in order to initialize the bus mutex lock */
+	rc = mdiobus_register(bus);
+	if (rc < 0) {
+		dev_err(dev, "failed to register MDIO bus\n");
+		return rc;
+	}
+
+	felix->imdio = bus;
+
+	for (port = 0; port < felix->info->num_ports; port++) {
+		struct ocelot_port *ocelot_port = ocelot->ports[port];
+		struct phy_device *pcs;
+		bool is_c45 = false;
+
+		if (ocelot_port->phy_mode == PHY_INTERFACE_MODE_USXGMII)
+			is_c45 = true;
+
+		pcs = get_phy_device(felix->imdio, port, is_c45);
+		if (IS_ERR(pcs))
+			continue;
+
+		pcs->interface = ocelot_port->phy_mode;
+		felix->pcs[port] = pcs;
+
+		dev_info(dev, "Found PCS at internal MDIO address %d\n", port);
+	}
+
+	return 0;
+}
+
+static void vsc9959_mdio_bus_free(struct ocelot *ocelot)
+{
+	struct felix *felix = ocelot_to_felix(ocelot);
+	int port;
+
+	for (port = 0; port < ocelot->num_phys_ports; port++) {
+		struct phy_device *pcs = felix->pcs[port];
+
+		if (!pcs)
+			continue;
+
+		put_device(&pcs->mdio.dev);
+	}
+	mdiobus_unregister(felix->imdio);
+}
+
 struct felix_info felix_info_vsc9959 = {
 	.target_io_res		= vsc9959_target_io_res,
 	.port_io_res		= vsc9959_port_io_res,
+	.imdio_res		= &vsc9959_imdio_res,
 	.regfields		= vsc9959_regfields,
 	.map			= vsc9959_regmap,
 	.ops			= &vsc9959_ops,
@@ -579,5 +1071,12 @@ struct felix_info felix_info_vsc9959 = {
 	.num_stats		= ARRAY_SIZE(vsc9959_stats_layout),
 	.shared_queue_sz	= 128 * 1024,
 	.num_ports		= 6,
-	.pci_bar		= 4,
+	.switch_pci_bar		= 4,
+	.imdio_pci_bar		= 0,
+	.mdio_bus_alloc		= vsc9959_mdio_bus_alloc,
+	.mdio_bus_free		= vsc9959_mdio_bus_free,
+	.pcs_init		= vsc9959_pcs_init,
+	.pcs_an_restart		= vsc9959_pcs_an_restart,
+	.pcs_link_state		= vsc9959_pcs_link_state,
+	.prevalidate_phy_mode	= vsc9959_prevalidate_phy_mode,
 };
-- 
cgit v1.2.3-59-g8ed1b


From 19220f35b3708dc069135046061fbe7366d5cb6e Mon Sep 17 00:00:00 2001
From: Rocky Liao <rjliao@codeaurora.org>
Date: Mon, 6 Jan 2020 14:07:44 +0800
Subject: Bluetooth: btusb: Add support for 04ca:3021 QCA_ROME device

USB "VendorID:04ca ProductID:3021" is a new QCA ROME USB
Bluetooth device, this patch will support firmware downloading for it.

T:  Bus=02 Lev=02 Prnt=02 Port=05 Cnt=01 Dev#=  3 Spd=12   MxCh= 0
D:  Ver= 2.01 Cls=e0(wlcon) Sub=01 Prot=01 MxPS=64 #Cfgs=  1
P:  Vendor=04ca ProdID=3021 Rev= 0.01
C:* #Ifs= 2 Cfg#= 1 Atr=e0 MxPwr=100mA
I:* If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=81(I) Atr=03(Int.) MxPS=  16 Ivl=1ms
E:  Ad=82(I) Atr=02(Bulk) MxPS=  64 Ivl=0ms
E:  Ad=02(O) Atr=02(Bulk) MxPS=  64 Ivl=0ms
I:* If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=83(I) Atr=01(Isoc) MxPS=   0 Ivl=1ms
E:  Ad=03(O) Atr=01(Isoc) MxPS=   0 Ivl=1ms
I:  If#= 1 Alt= 1 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=83(I) Atr=01(Isoc) MxPS=   9 Ivl=1ms
E:  Ad=03(O) Atr=01(Isoc) MxPS=   9 Ivl=1ms
I:  If#= 1 Alt= 2 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=83(I) Atr=01(Isoc) MxPS=  17 Ivl=1ms
E:  Ad=03(O) Atr=01(Isoc) MxPS=  17 Ivl=1ms
I:  If#= 1 Alt= 3 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=83(I) Atr=01(Isoc) MxPS=  25 Ivl=1ms
E:  Ad=03(O) Atr=01(Isoc) MxPS=  25 Ivl=1ms
I:  If#= 1 Alt= 4 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=83(I) Atr=01(Isoc) MxPS=  33 Ivl=1ms
E:  Ad=03(O) Atr=01(Isoc) MxPS=  33 Ivl=1ms
I:  If#= 1 Alt= 5 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=83(I) Atr=01(Isoc) MxPS=  49 Ivl=1ms
E:  Ad=03(O) Atr=01(Isoc) MxPS=  49 Ivl=1ms

Signed-off-by: Rocky Liao <rjliao@codeaurora.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 drivers/bluetooth/btusb.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 0eaeca0a64fb..f5924f3e8b8d 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -266,6 +266,7 @@ static const struct usb_device_id blacklist_table[] = {
 	{ USB_DEVICE(0x04ca, 0x3015), .driver_info = BTUSB_QCA_ROME },
 	{ USB_DEVICE(0x04ca, 0x3016), .driver_info = BTUSB_QCA_ROME },
 	{ USB_DEVICE(0x04ca, 0x301a), .driver_info = BTUSB_QCA_ROME },
+	{ USB_DEVICE(0x04ca, 0x3021), .driver_info = BTUSB_QCA_ROME },
 	{ USB_DEVICE(0x13d3, 0x3491), .driver_info = BTUSB_QCA_ROME },
 	{ USB_DEVICE(0x13d3, 0x3496), .driver_info = BTUSB_QCA_ROME },
 	{ USB_DEVICE(0x13d3, 0x3501), .driver_info = BTUSB_QCA_ROME },
-- 
cgit v1.2.3-59-g8ed1b


From 698a89541c223b247a3b08cd0b16c97344445d96 Mon Sep 17 00:00:00 2001
From: Yunsheng Lin <linyunsheng@huawei.com>
Date: Sat, 4 Jan 2020 10:49:24 +0800
Subject: net: hns3: add trace event support for HNS3 driver

This adds trace support for HNS3 driver. It also declares
some events which could be used to trace the events when a
TX/RX BD is processed, and other events which are related to
the processing of sk_buff, such as TSO, GRO.

Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/Makefile     |   2 +
 drivers/net/ethernet/hisilicon/hns3/hns3_enet.c  |  32 +++++-
 drivers/net/ethernet/hisilicon/hns3/hns3_enet.h  |   1 +
 drivers/net/ethernet/hisilicon/hns3/hns3_trace.h | 139 +++++++++++++++++++++++
 4 files changed, 172 insertions(+), 2 deletions(-)
 create mode 100644 drivers/net/ethernet/hisilicon/hns3/hns3_trace.h

diff --git a/drivers/net/ethernet/hisilicon/hns3/Makefile b/drivers/net/ethernet/hisilicon/hns3/Makefile
index d01bf536eb86..7aa2fac76c5e 100644
--- a/drivers/net/ethernet/hisilicon/hns3/Makefile
+++ b/drivers/net/ethernet/hisilicon/hns3/Makefile
@@ -3,6 +3,8 @@
 # Makefile for the HISILICON network device drivers.
 #
 
+ccflags-y += -I$(srctree)/$(src)
+
 obj-$(CONFIG_HNS3) += hns3pf/
 obj-$(CONFIG_HNS3) += hns3vf/
 
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
index aee5facc89b5..9a0694ae54c2 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
@@ -24,6 +24,12 @@
 
 #include "hnae3.h"
 #include "hns3_enet.h"
+/* All hns3 tracepoints are defined by the include below, which
+ * must be included exactly once across the whole kernel with
+ * CREATE_TRACE_POINTS defined
+ */
+#define CREATE_TRACE_POINTS
+#include "hns3_trace.h"
 
 #define hns3_set_field(origin, shift, val)	((origin) |= ((val) << (shift)))
 #define hns3_tx_bd_count(S)	DIV_ROUND_UP(S, HNS3_MAX_BD_SIZE)
@@ -734,6 +740,8 @@ static int hns3_set_tso(struct sk_buff *skb, u32 *paylen,
 	/* get MSS for TSO */
 	*mss = skb_shinfo(skb)->gso_size;
 
+	trace_hns3_tso(skb);
+
 	return 0;
 }
 
@@ -1138,6 +1146,7 @@ static int hns3_fill_desc(struct hns3_enet_ring *ring, void *priv,
 		desc->tx.bdtp_fe_sc_vld_ra_ri =
 			cpu_to_le16(BIT(HNS3_TXD_VLD_B));
 
+		trace_hns3_tx_desc(ring, ring->next_to_use);
 		ring_ptr_move_fw(ring, next_to_use);
 		return HNS3_LIKELY_BD_NUM;
 	}
@@ -1161,6 +1170,7 @@ static int hns3_fill_desc(struct hns3_enet_ring *ring, void *priv,
 		desc->tx.bdtp_fe_sc_vld_ra_ri =
 				cpu_to_le16(BIT(HNS3_TXD_VLD_B));
 
+		trace_hns3_tx_desc(ring, ring->next_to_use);
 		/* move ring pointer to next */
 		ring_ptr_move_fw(ring, next_to_use);
 
@@ -1286,6 +1296,14 @@ static bool hns3_skb_need_linearized(struct sk_buff *skb, unsigned int *bd_size,
 	return false;
 }
 
+void hns3_shinfo_pack(struct skb_shared_info *shinfo, __u32 *size)
+{
+	int i = 0;
+
+	for (i = 0; i < MAX_SKB_FRAGS; i++)
+		size[i] = skb_frag_size(&shinfo->frags[i]);
+}
+
 static int hns3_nic_maybe_stop_tx(struct hns3_enet_ring *ring,
 				  struct net_device *netdev,
 				  struct sk_buff *skb)
@@ -1297,8 +1315,10 @@ static int hns3_nic_maybe_stop_tx(struct hns3_enet_ring *ring,
 	bd_num = hns3_tx_bd_num(skb, bd_size);
 	if (unlikely(bd_num > HNS3_MAX_NON_TSO_BD_NUM)) {
 		if (bd_num <= HNS3_MAX_TSO_BD_NUM && skb_is_gso(skb) &&
-		    !hns3_skb_need_linearized(skb, bd_size, bd_num))
+		    !hns3_skb_need_linearized(skb, bd_size, bd_num)) {
+			trace_hns3_over_8bd(skb);
 			goto out;
+		}
 
 		if (__skb_linearize(skb))
 			return -ENOMEM;
@@ -1306,8 +1326,10 @@ static int hns3_nic_maybe_stop_tx(struct hns3_enet_ring *ring,
 		bd_num = hns3_tx_bd_count(skb->len);
 		if ((skb_is_gso(skb) && bd_num > HNS3_MAX_TSO_BD_NUM) ||
 		    (!skb_is_gso(skb) &&
-		     bd_num > HNS3_MAX_NON_TSO_BD_NUM))
+		     bd_num > HNS3_MAX_NON_TSO_BD_NUM)) {
+			trace_hns3_over_8bd(skb);
 			return -ENOMEM;
+		}
 
 		u64_stats_update_begin(&ring->syncp);
 		ring->stats.tx_copy++;
@@ -1448,6 +1470,7 @@ out:
 					(ring->desc_num - 1);
 	ring->desc[pre_ntu].tx.bdtp_fe_sc_vld_ra_ri |=
 				cpu_to_le16(BIT(HNS3_TXD_FE_B));
+	trace_hns3_tx_desc(ring, pre_ntu);
 
 	/* Complete translate all packets */
 	dev_queue = netdev_get_tx_queue(netdev, ring->queue_index);
@@ -2700,6 +2723,9 @@ static int hns3_gro_complete(struct sk_buff *skb, u32 l234info)
 	skb->csum_start = (unsigned char *)th - skb->head;
 	skb->csum_offset = offsetof(struct tcphdr, check);
 	skb->ip_summed = CHECKSUM_PARTIAL;
+
+	trace_hns3_gro(skb);
+
 	return 0;
 }
 
@@ -2836,6 +2862,7 @@ static int hns3_alloc_skb(struct hns3_enet_ring *ring, unsigned int length,
 		return -ENOMEM;
 	}
 
+	trace_hns3_rx_desc(ring);
 	prefetchw(skb->data);
 
 	ring->pending_buf = 1;
@@ -2910,6 +2937,7 @@ static int hns3_add_frag(struct hns3_enet_ring *ring)
 		}
 
 		hns3_nic_reuse_page(skb, ring->frag_num++, ring, 0, desc_cb);
+		trace_hns3_rx_desc(ring);
 		ring_ptr_move_fw(ring, next_to_clean);
 		ring->pending_buf++;
 	} while (!(bd_base_info & BIT(HNS3_RXD_FE_B)));
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.h b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.h
index 9d47abd5c37c..abefd7a179f7 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.h
@@ -673,4 +673,5 @@ void hns3_dbg_init(struct hnae3_handle *handle);
 void hns3_dbg_uninit(struct hnae3_handle *handle);
 void hns3_dbg_register_debugfs(const char *debugfs_dir_name);
 void hns3_dbg_unregister_debugfs(void);
+void hns3_shinfo_pack(struct skb_shared_info *shinfo, __u32 *size);
 #endif
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_trace.h b/drivers/net/ethernet/hisilicon/hns3/hns3_trace.h
new file mode 100644
index 000000000000..7bddcca148a5
--- /dev/null
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_trace.h
@@ -0,0 +1,139 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/* Copyright (c) 2018-2019 Hisilicon Limited. */
+
+/* This must be outside ifdef _HNS3_TRACE_H */
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM hns3
+
+#if !defined(_HNS3_TRACE_H_) || defined(TRACE_HEADER_MULTI_READ)
+#define _HNS3_TRACE_H_
+
+#include <linux/tracepoint.h>
+
+#define DESC_NR		(sizeof(struct hns3_desc) / sizeof(u32))
+
+DECLARE_EVENT_CLASS(hns3_skb_template,
+	TP_PROTO(struct sk_buff *skb),
+	TP_ARGS(skb),
+
+	TP_STRUCT__entry(
+		__field(unsigned int, headlen)
+		__field(unsigned int, len)
+		__field(__u8, nr_frags)
+		__field(__u8, ip_summed)
+		__field(unsigned int, hdr_len)
+		__field(unsigned short, gso_size)
+		__field(unsigned short, gso_segs)
+		__field(unsigned int, gso_type)
+		__field(bool, fraglist)
+		__array(__u32, size, MAX_SKB_FRAGS)
+	),
+
+	TP_fast_assign(
+		__entry->headlen = skb_headlen(skb);
+		__entry->len = skb->len;
+		__entry->nr_frags = skb_shinfo(skb)->nr_frags;
+		__entry->gso_size = skb_shinfo(skb)->gso_size;
+		__entry->gso_segs = skb_shinfo(skb)->gso_segs;
+		__entry->gso_type = skb_shinfo(skb)->gso_type;
+		__entry->hdr_len = skb->encapsulation ?
+		skb_inner_transport_offset(skb) + inner_tcp_hdrlen(skb) :
+		skb_transport_offset(skb) + tcp_hdrlen(skb);
+		__entry->ip_summed = skb->ip_summed;
+		__entry->fraglist = skb_has_frag_list(skb);
+		hns3_shinfo_pack(skb_shinfo(skb), __entry->size);
+	),
+
+	TP_printk(
+		"len: %u, %u, %u, cs: %u, gso: %u, %u, %x, frag(%d %u): %s",
+		__entry->headlen, __entry->len, __entry->hdr_len,
+		__entry->ip_summed, __entry->gso_size, __entry->gso_segs,
+		__entry->gso_type, __entry->fraglist, __entry->nr_frags,
+		__print_array(__entry->size, MAX_SKB_FRAGS, sizeof(__u32))
+	)
+);
+
+DEFINE_EVENT(hns3_skb_template, hns3_over_8bd,
+	TP_PROTO(struct sk_buff *skb),
+	TP_ARGS(skb));
+
+DEFINE_EVENT(hns3_skb_template, hns3_gro,
+	TP_PROTO(struct sk_buff *skb),
+	TP_ARGS(skb));
+
+DEFINE_EVENT(hns3_skb_template, hns3_tso,
+	TP_PROTO(struct sk_buff *skb),
+	TP_ARGS(skb));
+
+TRACE_EVENT(hns3_tx_desc,
+	TP_PROTO(struct hns3_enet_ring *ring, int cur_ntu),
+	TP_ARGS(ring, cur_ntu),
+
+	TP_STRUCT__entry(
+		__field(int, index)
+		__field(int, ntu)
+		__field(int, ntc)
+		__field(dma_addr_t, desc_dma)
+		__array(u32, desc, DESC_NR)
+		__string(devname, ring->tqp->handle->kinfo.netdev->name)
+	),
+
+	TP_fast_assign(
+		__entry->index = ring->tqp->tqp_index;
+		__entry->ntu = ring->next_to_use;
+		__entry->ntc = ring->next_to_clean;
+		__entry->desc_dma = ring->desc_dma_addr,
+		memcpy(__entry->desc, &ring->desc[cur_ntu],
+		       sizeof(struct hns3_desc));
+		__assign_str(devname, ring->tqp->handle->kinfo.netdev->name);
+	),
+
+	TP_printk(
+		"%s-%d-%d/%d desc(%pad): %s",
+		__get_str(devname), __entry->index, __entry->ntu,
+		__entry->ntc, &__entry->desc_dma,
+		__print_array(__entry->desc, DESC_NR, sizeof(u32))
+	)
+);
+
+TRACE_EVENT(hns3_rx_desc,
+	TP_PROTO(struct hns3_enet_ring *ring),
+	TP_ARGS(ring),
+
+	TP_STRUCT__entry(
+		__field(int, index)
+		__field(int, ntu)
+		__field(int, ntc)
+		__field(dma_addr_t, desc_dma)
+		__field(dma_addr_t, buf_dma)
+		__array(u32, desc, DESC_NR)
+		__string(devname, ring->tqp->handle->kinfo.netdev->name)
+	),
+
+	TP_fast_assign(
+		__entry->index = ring->tqp->tqp_index;
+		__entry->ntu = ring->next_to_use;
+		__entry->ntc = ring->next_to_clean;
+		__entry->desc_dma = ring->desc_dma_addr;
+		__entry->buf_dma = ring->desc_cb[ring->next_to_clean].dma;
+		memcpy(__entry->desc, &ring->desc[ring->next_to_clean],
+		       sizeof(struct hns3_desc));
+		__assign_str(devname, ring->tqp->handle->kinfo.netdev->name);
+	),
+
+	TP_printk(
+		"%s-%d-%d/%d desc(%pad) buf(%pad): %s",
+		__get_str(devname), __entry->index, __entry->ntu,
+		__entry->ntc, &__entry->desc_dma, &__entry->buf_dma,
+		__print_array(__entry->desc, DESC_NR, sizeof(u32))
+	)
+);
+
+#endif /* _HNS3_TRACE_H_ */
+
+/* This must be outside ifdef _HNS3_TRACE_H */
+#undef TRACE_INCLUDE_PATH
+#define TRACE_INCLUDE_PATH .
+#undef TRACE_INCLUDE_FILE
+#define TRACE_INCLUDE_FILE hns3_trace
+#include <trace/define_trace.h>
-- 
cgit v1.2.3-59-g8ed1b


From 08a100689d4baf296d6898c687ea8d005da8d234 Mon Sep 17 00:00:00 2001
From: Yonglong Liu <liuyonglong@huawei.com>
Date: Sat, 4 Jan 2020 10:49:25 +0800
Subject: net: hns3: re-organize vector handle

To prevent loss user's IRQ affinity configuration when DOWN,
this patch moves out release/request operation of the vector
handle from net DOWN/UP, just do it when vector resource changes.

Signed-off-by: Yonglong Liu <liuyonglong@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hns3_enet.c | 100 ++++++++++++------------
 1 file changed, 52 insertions(+), 48 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
index 9a0694ae54c2..01bad673443d 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
@@ -161,6 +161,8 @@ static int hns3_nic_init_irq(struct hns3_nic_priv *priv)
 			return ret;
 		}
 
+		disable_irq(tqp_vectors->vector_irq);
+
 		irq_set_affinity_hint(tqp_vectors->vector_irq,
 				      &tqp_vectors->affinity_mask);
 
@@ -179,6 +181,7 @@ static void hns3_mask_vector_irq(struct hns3_enet_tqp_vector *tqp_vector,
 static void hns3_vector_enable(struct hns3_enet_tqp_vector *tqp_vector)
 {
 	napi_enable(&tqp_vector->napi);
+	enable_irq(tqp_vector->vector_irq);
 
 	/* enable vector */
 	hns3_mask_vector_irq(tqp_vector, 1);
@@ -378,18 +381,6 @@ static int hns3_nic_net_up(struct net_device *netdev)
 	if (ret)
 		return ret;
 
-	/* the device can work without cpu rmap, only aRFS needs it */
-	ret = hns3_set_rx_cpu_rmap(netdev);
-	if (ret)
-		netdev_warn(netdev, "set rx cpu rmap fail, ret=%d!\n", ret);
-
-	/* get irq resource for all vectors */
-	ret = hns3_nic_init_irq(priv);
-	if (ret) {
-		netdev_err(netdev, "init irq failed! ret=%d\n", ret);
-		goto free_rmap;
-	}
-
 	clear_bit(HNS3_NIC_STATE_DOWN, &priv->state);
 
 	/* enable the vectors */
@@ -402,22 +393,15 @@ static int hns3_nic_net_up(struct net_device *netdev)
 
 	/* start the ae_dev */
 	ret = h->ae_algo->ops->start ? h->ae_algo->ops->start(h) : 0;
-	if (ret)
-		goto out_start_err;
-
-	return 0;
-
-out_start_err:
-	set_bit(HNS3_NIC_STATE_DOWN, &priv->state);
-	while (j--)
-		hns3_tqp_disable(h->kinfo.tqp[j]);
+	if (ret) {
+		set_bit(HNS3_NIC_STATE_DOWN, &priv->state);
+		while (j--)
+			hns3_tqp_disable(h->kinfo.tqp[j]);
 
-	for (j = i - 1; j >= 0; j--)
-		hns3_vector_disable(&priv->tqp_vector[j]);
+		for (j = i - 1; j >= 0; j--)
+			hns3_vector_disable(&priv->tqp_vector[j]);
+	}
 
-	hns3_nic_uninit_irq(priv);
-free_rmap:
-	hns3_free_rx_cpu_rmap(netdev);
 	return ret;
 }
 
@@ -514,11 +498,6 @@ static void hns3_nic_net_down(struct net_device *netdev)
 	if (ops->stop)
 		ops->stop(priv->ae_handle);
 
-	hns3_free_rx_cpu_rmap(netdev);
-
-	/* free irq resources */
-	hns3_nic_uninit_irq(priv);
-
 	/* delay ring buffer clearing to hns3_reset_notify_uninit_enet
 	 * during reset process, because driver may not be able
 	 * to disable the ring through firmware when downing the netdev.
@@ -3642,19 +3621,13 @@ static void hns3_nic_uninit_vector_data(struct hns3_nic_priv *priv)
 
 		hns3_free_vector_ring_chain(tqp_vector, &vector_ring_chain);
 
-		if (tqp_vector->irq_init_flag == HNS3_VECTOR_INITED) {
-			irq_set_affinity_hint(tqp_vector->vector_irq, NULL);
-			free_irq(tqp_vector->vector_irq, tqp_vector);
-			tqp_vector->irq_init_flag = HNS3_VECTOR_NOT_INITED;
-		}
-
 		hns3_clear_ring_group(&tqp_vector->rx_group);
 		hns3_clear_ring_group(&tqp_vector->tx_group);
 		netif_napi_del(&priv->tqp_vector[i].napi);
 	}
 }
 
-static int hns3_nic_dealloc_vector_data(struct hns3_nic_priv *priv)
+static void hns3_nic_dealloc_vector_data(struct hns3_nic_priv *priv)
 {
 	struct hnae3_handle *h = priv->ae_handle;
 	struct pci_dev *pdev = h->pdev;
@@ -3666,11 +3639,10 @@ static int hns3_nic_dealloc_vector_data(struct hns3_nic_priv *priv)
 		tqp_vector = &priv->tqp_vector[i];
 		ret = h->ae_algo->ops->put_vector(h, tqp_vector->vector_irq);
 		if (ret)
-			return ret;
+			return;
 	}
 
 	devm_kfree(&pdev->dev, priv->tqp_vector);
-	return 0;
 }
 
 static void hns3_ring_get_cfg(struct hnae3_queue *q, struct hns3_nic_priv *priv,
@@ -4069,6 +4041,18 @@ static int hns3_client_init(struct hnae3_handle *handle)
 		goto out_reg_netdev_fail;
 	}
 
+	/* the device can work without cpu rmap, only aRFS needs it */
+	ret = hns3_set_rx_cpu_rmap(netdev);
+	if (ret)
+		dev_warn(priv->dev, "set rx cpu rmap fail, ret=%d\n", ret);
+
+	ret = hns3_nic_init_irq(priv);
+	if (ret) {
+		dev_err(priv->dev, "init irq failed! ret=%d\n", ret);
+		hns3_free_rx_cpu_rmap(netdev);
+		goto out_init_irq_fail;
+	}
+
 	ret = hns3_client_start(handle);
 	if (ret) {
 		dev_err(priv->dev, "hns3_client_start fail! ret=%d\n", ret);
@@ -4090,6 +4074,9 @@ static int hns3_client_init(struct hnae3_handle *handle)
 	return ret;
 
 out_client_start:
+	hns3_free_rx_cpu_rmap(netdev);
+	hns3_nic_uninit_irq(priv);
+out_init_irq_fail:
 	unregister_netdev(netdev);
 out_reg_netdev_fail:
 	hns3_uninit_phy(netdev);
@@ -4127,15 +4114,17 @@ static void hns3_client_uninit(struct hnae3_handle *handle, bool reset)
 		goto out_netdev_free;
 	}
 
+	hns3_free_rx_cpu_rmap(netdev);
+
+	hns3_nic_uninit_irq(priv);
+
 	hns3_del_all_fd_rules(netdev, true);
 
 	hns3_clear_all_ring(handle, true);
 
 	hns3_nic_uninit_vector_data(priv);
 
-	ret = hns3_nic_dealloc_vector_data(priv);
-	if (ret)
-		netdev_err(netdev, "dealloc vector error\n");
+	hns3_nic_dealloc_vector_data(priv);
 
 	ret = hns3_uninit_all_ring(priv);
 	if (ret)
@@ -4462,17 +4451,32 @@ static int hns3_reset_notify_init_enet(struct hnae3_handle *handle)
 	if (ret)
 		goto err_uninit_vector;
 
+	/* the device can work without cpu rmap, only aRFS needs it */
+	ret = hns3_set_rx_cpu_rmap(netdev);
+	if (ret)
+		dev_warn(priv->dev, "set rx cpu rmap fail, ret=%d\n", ret);
+
+	ret = hns3_nic_init_irq(priv);
+	if (ret) {
+		dev_err(priv->dev, "init irq failed! ret=%d\n", ret);
+		hns3_free_rx_cpu_rmap(netdev);
+		goto err_init_irq_fail;
+	}
+
 	ret = hns3_client_start(handle);
 	if (ret) {
 		dev_err(priv->dev, "hns3_client_start fail! ret=%d\n", ret);
-		goto err_uninit_ring;
+		goto err_client_start_fail;
 	}
 
 	set_bit(HNS3_NIC_STATE_INITED, &priv->state);
 
 	return ret;
 
-err_uninit_ring:
+err_client_start_fail:
+	hns3_free_rx_cpu_rmap(netdev);
+	hns3_nic_uninit_irq(priv);
+err_init_irq_fail:
 	hns3_uninit_all_ring(priv);
 err_uninit_vector:
 	hns3_nic_uninit_vector_data(priv);
@@ -4522,6 +4526,8 @@ static int hns3_reset_notify_uninit_enet(struct hnae3_handle *handle)
 		return 0;
 	}
 
+	hns3_free_rx_cpu_rmap(netdev);
+	hns3_nic_uninit_irq(priv);
 	hns3_clear_all_ring(handle, true);
 	hns3_reset_tx_queue(priv->ae_handle);
 
@@ -4529,9 +4535,7 @@ static int hns3_reset_notify_uninit_enet(struct hnae3_handle *handle)
 
 	hns3_store_coal(priv);
 
-	ret = hns3_nic_dealloc_vector_data(priv);
-	if (ret)
-		netdev_err(netdev, "dealloc vector error\n");
+	hns3_nic_dealloc_vector_data(priv);
 
 	ret = hns3_uninit_all_ring(priv);
 	if (ret)
-- 
cgit v1.2.3-59-g8ed1b


From 5bffde62a18098a4268e56c5dea175e208fa0cf4 Mon Sep 17 00:00:00 2001
From: Yonglong Liu <liuyonglong@huawei.com>
Date: Sat, 4 Jan 2020 10:49:26 +0800
Subject: net: hns3: modify the IRQ name of TQP vector

When rename the net devices, the IRQ number can not be
fetched by the net device name, because the driver request
the IRQ resources only when the vector resource changed, and
the rename operation did not change the vector resources,
so the IRQ name keeps the previous net device name.
So this patch modifies the name of the TQP IRQ as
"pci driver name"-"pci name"-"TxRx"-"index".

Signed-off-by: Yonglong Liu <liuyonglong@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hns3_enet.c | 21 ++++++++++++---------
 1 file changed, 12 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
index 01bad673443d..e240d99f7ca8 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
@@ -133,18 +133,21 @@ static int hns3_nic_init_irq(struct hns3_nic_priv *priv)
 			continue;
 
 		if (tqp_vectors->tx_group.ring && tqp_vectors->rx_group.ring) {
-			snprintf(tqp_vectors->name, HNAE3_INT_NAME_LEN - 1,
-				 "%s-%s-%d", priv->netdev->name, "TxRx",
-				 txrx_int_idx++);
+			snprintf(tqp_vectors->name, HNAE3_INT_NAME_LEN,
+				 "%s-%s-%s-%d", hns3_driver_name,
+				 pci_name(priv->ae_handle->pdev),
+				 "TxRx", txrx_int_idx++);
 			txrx_int_idx++;
 		} else if (tqp_vectors->rx_group.ring) {
-			snprintf(tqp_vectors->name, HNAE3_INT_NAME_LEN - 1,
-				 "%s-%s-%d", priv->netdev->name, "Rx",
-				 rx_int_idx++);
+			snprintf(tqp_vectors->name, HNAE3_INT_NAME_LEN,
+				 "%s-%s-%s-%d", hns3_driver_name,
+				 pci_name(priv->ae_handle->pdev),
+				 "Rx", rx_int_idx++);
 		} else if (tqp_vectors->tx_group.ring) {
-			snprintf(tqp_vectors->name, HNAE3_INT_NAME_LEN - 1,
-				 "%s-%s-%d", priv->netdev->name, "Tx",
-				 tx_int_idx++);
+			snprintf(tqp_vectors->name, HNAE3_INT_NAME_LEN,
+				 "%s-%s-%s-%d", hns3_driver_name,
+				 pci_name(priv->ae_handle->pdev),
+				 "Tx", tx_int_idx++);
 		} else {
 			/* Skip this unused q_vector */
 			continue;
-- 
cgit v1.2.3-59-g8ed1b


From 7ab2b53e4657ee0289878b09f9748a35b0e6010b Mon Sep 17 00:00:00 2001
From: Yonglong Liu <liuyonglong@huawei.com>
Date: Sat, 4 Jan 2020 10:49:27 +0800
Subject: net: hns3: modify an unsuitable log in hclge_map_ring_to_vector()

When the returned vector_id less than 0, the message should print
out the vector who is getting vector index fail.

So this patch replaces vector_id with vector, and re-format the
message.

Signed-off-by: Yonglong Liu <liuyonglong@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
index 5e3743929e1c..b69612736956 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
@@ -4698,7 +4698,7 @@ static int hclge_map_ring_to_vector(struct hnae3_handle *handle, int vector,
 	vector_id = hclge_get_vector_index(hdev, vector);
 	if (vector_id < 0) {
 		dev_err(&hdev->pdev->dev,
-			"Get vector index fail. vector_id =%d\n", vector_id);
+			"failed to get vector index. vector=%d\n", vector);
 		return vector_id;
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From f97c4d823fb7d210afa53c6980dd8b8d0deee916 Mon Sep 17 00:00:00 2001
From: Yonglong Liu <liuyonglong@huawei.com>
Date: Sat, 4 Jan 2020 10:49:28 +0800
Subject: net: hns3: modify the IRQ name of misc vectors

The misc IRQ of all the devices have the same name, so it's
hard to find the right misc IRQ of the device.

This patch modifies the misc IRQ names as "hclge/hclgevf"-misc-
"pci name". And now the IRQ name is not related to net device
name anymore, so change the HNAE3_INT_NAME_LEN to 32 bytes, and
that is enough.

Signed-off-by: Yonglong Liu <liuyonglong@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hnae3.h               | 3 +--
 drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.h    | 2 ++
 drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c   | 4 +++-
 drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c | 4 +++-
 drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h | 1 +
 5 files changed, 10 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hnae3.h b/drivers/net/ethernet/hisilicon/hns3/hnae3.h
index 3b5e2d7251e7..ed97bd6c03e6 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hnae3.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hnae3.h
@@ -575,8 +575,7 @@ struct hnae3_ae_algo {
 	const struct pci_device_id *pdev_id_table;
 };
 
-#define HNAE3_INT_NAME_EXT_LEN    32	 /* Max extra information length */
-#define HNAE3_INT_NAME_LEN        (IFNAMSIZ + HNAE3_INT_NAME_EXT_LEN)
+#define HNAE3_INT_NAME_LEN        32
 #define HNAE3_ITR_COUNTDOWN_START 100
 
 struct hnae3_tc_info {
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.h
index d97da67f07a1..96498d9b4754 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.h
@@ -6,6 +6,7 @@
 #include <linux/types.h>
 #include <linux/io.h>
 #include <linux/etherdevice.h>
+#include "hnae3.h"
 
 #define HCLGE_CMDQ_TX_TIMEOUT		30000
 #define HCLGE_DESC_DATA_LEN		6
@@ -63,6 +64,7 @@ enum hclge_cmd_status {
 struct hclge_misc_vector {
 	u8 __iomem *addr;
 	int vector_irq;
+	char name[HNAE3_INT_NAME_LEN];
 };
 
 struct hclge_cmq {
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
index b69612736956..4a15510a4bab 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
@@ -3177,8 +3177,10 @@ static int hclge_misc_irq_init(struct hclge_dev *hdev)
 	hclge_get_misc_vector(hdev);
 
 	/* this would be explicitly freed in the end */
+	snprintf(hdev->misc_vector.name, HNAE3_INT_NAME_LEN, "%s-misc-%s",
+		 HCLGE_NAME, pci_name(hdev->pdev));
 	ret = request_irq(hdev->misc_vector.vector_irq, hclge_misc_irq_handle,
-			  0, "hclge_misc", hdev);
+			  0, hdev->misc_vector.name, hdev);
 	if (ret) {
 		hclge_free_vector(hdev, 0);
 		dev_err(&hdev->pdev->dev, "request misc irq(%d) fail\n",
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
index c33b8027f801..cfa797ef9906 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
@@ -2376,8 +2376,10 @@ static int hclgevf_misc_irq_init(struct hclgevf_dev *hdev)
 
 	hclgevf_get_misc_vector(hdev);
 
+	snprintf(hdev->misc_vector.name, HNAE3_INT_NAME_LEN, "%s-misc-%s",
+		 HCLGEVF_NAME, pci_name(hdev->pdev));
 	ret = request_irq(hdev->misc_vector.vector_irq, hclgevf_misc_irq_handle,
-			  0, "hclgevf_cmd", hdev);
+			  0, hdev->misc_vector.name, hdev);
 	if (ret) {
 		dev_err(&hdev->pdev->dev, "VF failed to request misc irq(%d)\n",
 			hdev->misc_vector.vector_irq);
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h
index 003114f6db6c..2cbc7dfe2f91 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h
@@ -221,6 +221,7 @@ struct hclgevf_rss_cfg {
 struct hclgevf_misc_vector {
 	u8 __iomem *addr;
 	int vector_irq;
+	char name[HNAE3_INT_NAME_LEN];
 };
 
 struct hclgevf_rst_stats {
-- 
cgit v1.2.3-59-g8ed1b


From 2af8cb6126f36ffad848d1b0af714c09eb801b9d Mon Sep 17 00:00:00 2001
From: Guojia Liao <liaoguojia@huawei.com>
Date: Sat, 4 Jan 2020 10:49:29 +0800
Subject: net: hns3: add protection when get SFP speed as 0

In some case, the MAC speed get from hardware maybe 0, it should
not be set to mac->speed.

Signed-off-by: Guojia Liao <liaoguojia@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
index 4a15510a4bab..8bddda7f9092 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
@@ -2835,6 +2835,12 @@ static int hclge_get_sfp_info(struct hclge_dev *hdev, struct hclge_mac *mac)
 		return ret;
 	}
 
+	/* In some case, mac speed get from IMP may be 0, it shouldn't be
+	 * set to mac->speed.
+	 */
+	if (!le32_to_cpu(resp->speed))
+		return 0;
+
 	mac->speed = le32_to_cpu(resp->speed);
 	/* if resp->speed_ability is 0, it means it's an old version
 	 * firmware, do not update these params
-- 
cgit v1.2.3-59-g8ed1b


From 7061867b59e0e5231b64655356c67ec31f44bc01 Mon Sep 17 00:00:00 2001
From: Huazhong Tan <tanhuazhong@huawei.com>
Date: Sat, 4 Jan 2020 10:49:30 +0800
Subject: net: hns3: replace an unsuitable variable type in
 hclge_inform_reset_assert_to_vf()

In hclge_inform_reset_assert_to_vf(), variable reset_type(enum type)
will be copied into msg_data whose size is 2 bytes. Currently, hip08
is a little-endian machine, so the lower two bytes of reset_type will
be copied to msg_data. But when running on a big-endian machine,
msg_data will have a wrong value(the higher two bytes of reset_type).

So this patch modifies the type of reset_type to u16, and adds a
build check in case enum hnae3_reset_type has value larger than
U16_MAX.

Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hnae3.h            | 1 +
 drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c | 4 +++-
 2 files changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hnae3.h b/drivers/net/ethernet/hisilicon/hns3/hnae3.h
index ed97bd6c03e6..6b131ab36c7e 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hnae3.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hnae3.h
@@ -164,6 +164,7 @@ enum hnae3_reset_type {
 	HNAE3_IMP_RESET,
 	HNAE3_UNKNOWN_RESET,
 	HNAE3_NONE_RESET,
+	HNAE3_MAX_RESET,
 };
 
 enum hnae3_flr_state {
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c
index f905dd3386b3..a3c0822191a9 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c
@@ -86,10 +86,12 @@ static int hclge_send_mbx_msg(struct hclge_vport *vport, u8 *msg, u16 msg_len,
 int hclge_inform_reset_assert_to_vf(struct hclge_vport *vport)
 {
 	struct hclge_dev *hdev = vport->back;
-	enum hnae3_reset_type reset_type;
+	u16 reset_type;
 	u8 msg_data[2];
 	u8 dest_vfid;
 
+	BUILD_BUG_ON(HNAE3_MAX_RESET > U16_MAX);
+
 	dest_vfid = (u8)vport->vport_id;
 
 	if (hdev->reset_type == HNAE3_FUNC_RESET)
-- 
cgit v1.2.3-59-g8ed1b


From 7f39febf2e08a227e9696d2ebac2637d9ed98efc Mon Sep 17 00:00:00 2001
From: Huazhong Tan <tanhuazhong@huawei.com>
Date: Sat, 4 Jan 2020 10:49:31 +0800
Subject: net: hns3: modify an unsuitable reset level for hardware error

According to hardware user manual, when hardware reports error
'roc_pkt_without_key_port', the driver should assert function
reset to do the recovery.

So this patch uses HNAE3_FUNC_RESET to replace HNAE3_GLOBAL_RESET.

Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c
index dc66b4e13377..f8127d72d645 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c
@@ -505,7 +505,7 @@ static const struct hclge_hw_error hclge_ssu_mem_ecc_err_int[] = {
 
 static const struct hclge_hw_error hclge_ssu_port_based_err_int[] = {
 	{ .int_msk = BIT(0), .msg = "roc_pkt_without_key_port",
-	  .reset_level = HNAE3_GLOBAL_RESET },
+	  .reset_level = HNAE3_FUNC_RESET },
 	{ .int_msk = BIT(1), .msg = "tpu_pkt_without_key_port",
 	  .reset_level = HNAE3_GLOBAL_RESET },
 	{ .int_msk = BIT(2), .msg = "igu_pkt_without_key_port",
@@ -599,7 +599,7 @@ static const struct hclge_hw_error hclge_ssu_ets_tcg_int[] = {
 
 static const struct hclge_hw_error hclge_ssu_port_based_pf_int[] = {
 	{ .int_msk = BIT(0), .msg = "roc_pkt_without_key_port",
-	  .reset_level = HNAE3_GLOBAL_RESET },
+	  .reset_level = HNAE3_FUNC_RESET },
 	{ .int_msk = BIT(9), .msg = "low_water_line_err_port",
 	  .reset_level = HNAE3_NONE_RESET },
 	{ .int_msk = BIT(10), .msg = "hi_water_line_err_port",
-- 
cgit v1.2.3-59-g8ed1b


From 5f2958052c5820d0795616fcc299097390cf92a8 Mon Sep 17 00:00:00 2001
From: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Date: Mon, 2 Dec 2019 15:19:49 -0800
Subject: igc: Add basic skeleton for PTP

This allows the creation of the /dev/ptpX device for i225, and reading
and writing the time.

Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/Makefile      |   2 +-
 drivers/net/ethernet/intel/igc/igc.h         |  26 ++
 drivers/net/ethernet/intel/igc/igc_defines.h |  12 +
 drivers/net/ethernet/intel/igc/igc_main.c    |  27 +++
 drivers/net/ethernet/intel/igc/igc_ptp.c     | 349 +++++++++++++++++++++++++++
 drivers/net/ethernet/intel/igc/igc_regs.h    |  24 ++
 6 files changed, 439 insertions(+), 1 deletion(-)
 create mode 100644 drivers/net/ethernet/intel/igc/igc_ptp.c

diff --git a/drivers/net/ethernet/intel/igc/Makefile b/drivers/net/ethernet/intel/igc/Makefile
index 88c6f88baac5..49fb1e1965cd 100644
--- a/drivers/net/ethernet/intel/igc/Makefile
+++ b/drivers/net/ethernet/intel/igc/Makefile
@@ -8,4 +8,4 @@
 obj-$(CONFIG_IGC) += igc.o
 
 igc-objs := igc_main.o igc_mac.o igc_i225.o igc_base.o igc_nvm.o igc_phy.o \
-igc_ethtool.o
+igc_ethtool.o igc_ptp.o
diff --git a/drivers/net/ethernet/intel/igc/igc.h b/drivers/net/ethernet/intel/igc/igc.h
index 612fe9ec81a4..7586f237747a 100644
--- a/drivers/net/ethernet/intel/igc/igc.h
+++ b/drivers/net/ethernet/intel/igc/igc.h
@@ -10,6 +10,9 @@
 #include <linux/vmalloc.h>
 #include <linux/ethtool.h>
 #include <linux/sctp.h>
+#include <linux/ptp_clock_kernel.h>
+#include <linux/timecounter.h>
+#include <linux/net_tstamp.h>
 
 #include "igc_hw.h"
 
@@ -45,11 +48,15 @@ extern char igc_driver_version[];
 #define IGC_REGS_LEN			740
 #define IGC_RETA_SIZE			128
 
+/* flags controlling PTP/1588 function */
+#define IGC_PTP_ENABLED		BIT(0)
+
 /* Interrupt defines */
 #define IGC_START_ITR			648 /* ~6000 ints/sec */
 #define IGC_FLAG_HAS_MSI		BIT(0)
 #define IGC_FLAG_QUEUE_PAIRS		BIT(3)
 #define IGC_FLAG_DMAC			BIT(4)
+#define IGC_FLAG_PTP			BIT(8)
 #define IGC_FLAG_NEED_LINK_UPDATE	BIT(9)
 #define IGC_FLAG_MEDIA_RESET		BIT(10)
 #define IGC_FLAG_MAS_ENABLE		BIT(12)
@@ -432,6 +439,20 @@ struct igc_adapter {
 
 	unsigned long link_check_timeout;
 	struct igc_info ei;
+
+	struct ptp_clock *ptp_clock;
+	struct ptp_clock_info ptp_caps;
+	struct work_struct ptp_tx_work;
+	struct sk_buff *ptp_tx_skb;
+	struct hwtstamp_config tstamp_config;
+	unsigned long ptp_tx_start;
+	unsigned long last_rx_ptp_check;
+	unsigned long last_rx_timestamp;
+	unsigned int ptp_flags;
+	/* System time value lock */
+	spinlock_t tmreg_lock;
+	struct cyclecounter cc;
+	struct timecounter tc;
 };
 
 /* igc_desc_unused - calculate if we have unused descriptors */
@@ -515,6 +536,11 @@ int igc_add_filter(struct igc_adapter *adapter,
 int igc_erase_filter(struct igc_adapter *adapter,
 		     struct igc_nfc_filter *input);
 
+void igc_ptp_init(struct igc_adapter *adapter);
+void igc_ptp_reset(struct igc_adapter *adapter);
+void igc_ptp_stop(struct igc_adapter *adapter);
+int igc_ptp_set_ts_config(struct net_device *netdev, struct ifreq *ifr);
+int igc_ptp_get_ts_config(struct net_device *netdev, struct ifreq *ifr);
 #define igc_rx_pg_size(_ring) (PAGE_SIZE << igc_rx_pg_order(_ring))
 
 #define IGC_TXD_DCMD	(IGC_ADVTXD_DCMD_EOP | IGC_ADVTXD_DCMD_RS)
diff --git a/drivers/net/ethernet/intel/igc/igc_defines.h b/drivers/net/ethernet/intel/igc/igc_defines.h
index 50dffd5db606..12f9127e40b7 100644
--- a/drivers/net/ethernet/intel/igc/igc_defines.h
+++ b/drivers/net/ethernet/intel/igc/igc_defines.h
@@ -218,6 +218,7 @@
 #define IGC_ICR_RXDMT0		BIT(4)	/* Rx desc min. threshold (0) */
 #define IGC_ICR_RXO		BIT(6)	/* Rx overrun */
 #define IGC_ICR_RXT0		BIT(7)	/* Rx timer intr (ring 0) */
+#define IGC_ICR_TS		BIT(19)	/* Time Sync Interrupt */
 #define IGC_ICR_DRSTA		BIT(30)	/* Device Reset Asserted */
 
 /* If this bit asserted, the driver should claim the interrupt */
@@ -240,6 +241,7 @@
 #define IGC_IMS_DRSTA		IGC_ICR_DRSTA	/* Device Reset Asserted */
 #define IGC_IMS_RXT0		IGC_ICR_RXT0	/* Rx timer intr */
 #define IGC_IMS_RXDMT0		IGC_ICR_RXDMT0	/* Rx desc min. threshold */
+#define IGC_IMS_TS		IGC_ICR_TS	/* Time Sync Interrupt */
 
 #define IGC_QVECTOR_MASK	0x7FFC		/* Q-vector mask */
 #define IGC_ITR_VAL_MASK	0x04		/* ITR value mask */
@@ -355,6 +357,16 @@
 #define I225_RXPBSIZE_DEFAULT	0x000000A2 /* RXPBSIZE default */
 #define I225_TXPBSIZE_DEFAULT	0x04000014 /* TXPBSIZE default */
 
+/* Time Sync Interrupt Causes */
+#define IGC_TSICR_SYS_WRAP	BIT(0) /* SYSTIM Wrap around. */
+#define IGC_TSICR_TXTS		BIT(1) /* Transmit Timestamp. */
+#define IGC_TSICR_TT0		BIT(3) /* Target Time 0 Trigger. */
+#define IGC_TSICR_TT1		BIT(4) /* Target Time 1 Trigger. */
+#define IGC_TSICR_AUTT0		BIT(5) /* Auxiliary Timestamp 0 Taken. */
+#define IGC_TSICR_AUTT1		BIT(6) /* Auxiliary Timestamp 1 Taken. */
+
+#define IGC_TSICR_INTERRUPTS	IGC_TSICR_TXTS
+
 /* Receive Checksum Control */
 #define IGC_RXCSUM_CRCOFL	0x00000800   /* CRC32 offload enable */
 #define IGC_RXCSUM_PCSD		0x00002000   /* packet checksum disabled */
diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 24fbbc928d66..f242205792e4 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -102,6 +102,9 @@ void igc_reset(struct igc_adapter *adapter)
 	if (!netif_running(adapter->netdev))
 		igc_power_down_link(adapter);
 
+	/* Re-enable PTP, where applicable. */
+	igc_ptp_reset(adapter);
+
 	igc_get_phy_info(hw);
 }
 
@@ -4277,6 +4280,24 @@ static int igc_close(struct net_device *netdev)
 	return 0;
 }
 
+/**
+ * igc_ioctl - Access the hwtstamp interface
+ * @netdev: network interface device structure
+ * @ifreq: interface request data
+ * @cmd: ioctl command
+ **/
+static int igc_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
+{
+	switch (cmd) {
+	case SIOCGHWTSTAMP:
+		return igc_ptp_get_ts_config(netdev, ifr);
+	case SIOCSHWTSTAMP:
+		return igc_ptp_set_ts_config(netdev, ifr);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
 static const struct net_device_ops igc_netdev_ops = {
 	.ndo_open		= igc_open,
 	.ndo_stop		= igc_close,
@@ -4288,6 +4309,7 @@ static const struct net_device_ops igc_netdev_ops = {
 	.ndo_fix_features	= igc_fix_features,
 	.ndo_set_features	= igc_set_features,
 	.ndo_features_check	= igc_features_check,
+	.ndo_do_ioctl		= igc_ioctl,
 };
 
 /* PCIe configuration access */
@@ -4588,6 +4610,9 @@ static int igc_probe(struct pci_dev *pdev,
 	 /* carrier off reporting is important to ethtool even BEFORE open */
 	netif_carrier_off(netdev);
 
+	/* do hw tstamp init after resetting */
+	igc_ptp_init(adapter);
+
 	/* Check if Media Autosense is enabled */
 	adapter->ei = *ei;
 
@@ -4629,6 +4654,8 @@ static void igc_remove(struct pci_dev *pdev)
 	struct net_device *netdev = pci_get_drvdata(pdev);
 	struct igc_adapter *adapter = netdev_priv(netdev);
 
+	igc_ptp_stop(adapter);
+
 	set_bit(__IGC_DOWN, &adapter->state);
 
 	del_timer_sync(&adapter->watchdog_timer);
diff --git a/drivers/net/ethernet/intel/igc/igc_ptp.c b/drivers/net/ethernet/intel/igc/igc_ptp.c
new file mode 100644
index 000000000000..bab384880f96
--- /dev/null
+++ b/drivers/net/ethernet/intel/igc/igc_ptp.c
@@ -0,0 +1,349 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c)  2019 Intel Corporation */
+
+#include "igc.h"
+
+#include <linux/module.h>
+#include <linux/device.h>
+#include <linux/pci.h>
+#include <linux/ptp_classify.h>
+#include <linux/clocksource.h>
+
+#define INCVALUE_MASK		0x7fffffff
+#define ISGN			0x80000000
+
+#define IGC_SYSTIM_OVERFLOW_PERIOD	(HZ * 60 * 9)
+#define IGC_PTP_TX_TIMEOUT		(HZ * 15)
+
+/* SYSTIM read access for I225 */
+static void igc_ptp_read_i225(struct igc_adapter *adapter,
+			      struct timespec64 *ts)
+{
+	struct igc_hw *hw = &adapter->hw;
+	u32 sec, nsec;
+
+	/* The timestamp latches on lowest register read. For I210/I211, the
+	 * lowest register is SYSTIMR. Since we only need to provide nanosecond
+	 * resolution, we can ignore it.
+	 */
+	rd32(IGC_SYSTIMR);
+	nsec = rd32(IGC_SYSTIML);
+	sec = rd32(IGC_SYSTIMH);
+
+	ts->tv_sec = sec;
+	ts->tv_nsec = nsec;
+}
+
+static void igc_ptp_write_i225(struct igc_adapter *adapter,
+			       const struct timespec64 *ts)
+{
+	struct igc_hw *hw = &adapter->hw;
+
+	/* Writing the SYSTIMR register is not necessary as it only
+	 * provides sub-nanosecond resolution.
+	 */
+	wr32(IGC_SYSTIML, ts->tv_nsec);
+	wr32(IGC_SYSTIMH, ts->tv_sec);
+}
+
+static int igc_ptp_adjfine_i225(struct ptp_clock_info *ptp, long scaled_ppm)
+{
+	struct igc_adapter *igc = container_of(ptp, struct igc_adapter,
+					       ptp_caps);
+	struct igc_hw *hw = &igc->hw;
+	int neg_adj = 0;
+	u64 rate;
+	u32 inca;
+
+	if (scaled_ppm < 0) {
+		neg_adj = 1;
+		scaled_ppm = -scaled_ppm;
+	}
+	rate = scaled_ppm;
+	rate <<= 14;
+	rate = div_u64(rate, 78125);
+
+	inca = rate & INCVALUE_MASK;
+	if (neg_adj)
+		inca |= ISGN;
+
+	wr32(IGC_TIMINCA, inca);
+
+	return 0;
+}
+
+static int igc_ptp_adjtime_i225(struct ptp_clock_info *ptp, s64 delta)
+{
+	struct igc_adapter *igc = container_of(ptp, struct igc_adapter,
+					       ptp_caps);
+	struct timespec64 now, then = ns_to_timespec64(delta);
+	unsigned long flags;
+
+	spin_lock_irqsave(&igc->tmreg_lock, flags);
+
+	igc_ptp_read_i225(igc, &now);
+	now = timespec64_add(now, then);
+	igc_ptp_write_i225(igc, (const struct timespec64 *)&now);
+
+	spin_unlock_irqrestore(&igc->tmreg_lock, flags);
+
+	return 0;
+}
+
+static int igc_ptp_gettimex64_i225(struct ptp_clock_info *ptp,
+				   struct timespec64 *ts,
+				   struct ptp_system_timestamp *sts)
+{
+	struct igc_adapter *igc = container_of(ptp, struct igc_adapter,
+					       ptp_caps);
+	struct igc_hw *hw = &igc->hw;
+	unsigned long flags;
+
+	spin_lock_irqsave(&igc->tmreg_lock, flags);
+
+	ptp_read_system_prets(sts);
+	rd32(IGC_SYSTIMR);
+	ptp_read_system_postts(sts);
+	ts->tv_nsec = rd32(IGC_SYSTIML);
+	ts->tv_sec = rd32(IGC_SYSTIMH);
+
+	spin_unlock_irqrestore(&igc->tmreg_lock, flags);
+
+	return 0;
+}
+
+static int igc_ptp_settime_i225(struct ptp_clock_info *ptp,
+				const struct timespec64 *ts)
+{
+	struct igc_adapter *igc = container_of(ptp, struct igc_adapter,
+					       ptp_caps);
+	unsigned long flags;
+
+	spin_lock_irqsave(&igc->tmreg_lock, flags);
+
+	igc_ptp_write_i225(igc, ts);
+
+	spin_unlock_irqrestore(&igc->tmreg_lock, flags);
+
+	return 0;
+}
+
+static int igc_ptp_feature_enable_i225(struct ptp_clock_info *ptp,
+				       struct ptp_clock_request *rq, int on)
+{
+	return -EOPNOTSUPP;
+}
+
+static int igc_ptp_set_timestamp_mode(struct igc_adapter *adapter,
+				      struct hwtstamp_config *config)
+{
+	return 0;
+}
+
+void igc_ptp_tx_hang(struct igc_adapter *adapter)
+{
+	bool timeout = time_is_before_jiffies(adapter->ptp_tx_start +
+					      IGC_PTP_TX_TIMEOUT);
+	struct igc_hw *hw = &adapter->hw;
+
+	if (!adapter->ptp_tx_skb)
+		return;
+
+	if (!test_bit(__IGC_PTP_TX_IN_PROGRESS, &adapter->state))
+		return;
+
+	/* If we haven't received a timestamp within the timeout, it is
+	 * reasonable to assume that it will never occur, so we can unlock the
+	 * timestamp bit when this occurs.
+	 */
+	if (timeout) {
+		cancel_work_sync(&adapter->ptp_tx_work);
+		dev_kfree_skb_any(adapter->ptp_tx_skb);
+		adapter->ptp_tx_skb = NULL;
+		clear_bit_unlock(__IGC_PTP_TX_IN_PROGRESS, &adapter->state);
+		adapter->tx_hwtstamp_timeouts++;
+		/* Clear the Tx valid bit in TSYNCTXCTL register to enable
+		 * interrupt
+		 */
+		rd32(IGC_TXSTMPH);
+		dev_warn(&adapter->pdev->dev, "clearing Tx timestamp hang\n");
+	}
+}
+
+void igc_ptp_tx_work(struct work_struct *work)
+{
+}
+
+/**
+ * igc_ptp_set_ts_config - set hardware time stamping config
+ * @netdev: network interface device structure
+ * @ifreq: interface request data
+ *
+ **/
+int igc_ptp_set_ts_config(struct net_device *netdev, struct ifreq *ifr)
+{
+	struct igc_adapter *adapter = netdev_priv(netdev);
+	struct hwtstamp_config config;
+	int err;
+
+	if (copy_from_user(&config, ifr->ifr_data, sizeof(config)))
+		return -EFAULT;
+
+	err = igc_ptp_set_timestamp_mode(adapter, &config);
+	if (err)
+		return err;
+
+	/* save these settings for future reference */
+	memcpy(&adapter->tstamp_config, &config,
+	       sizeof(adapter->tstamp_config));
+
+	return copy_to_user(ifr->ifr_data, &config, sizeof(config)) ?
+		-EFAULT : 0;
+}
+
+/**
+ * igc_ptp_get_ts_config - get hardware time stamping config
+ * @netdev: network interface device structure
+ * @ifreq: interface request data
+ *
+ * Get the hwtstamp_config settings to return to the user. Rather than attempt
+ * to deconstruct the settings from the registers, just return a shadow copy
+ * of the last known settings.
+ **/
+int igc_ptp_get_ts_config(struct net_device *netdev, struct ifreq *ifr)
+{
+	struct igc_adapter *adapter = netdev_priv(netdev);
+	struct hwtstamp_config *config = &adapter->tstamp_config;
+
+	return copy_to_user(ifr->ifr_data, config, sizeof(*config)) ?
+		-EFAULT : 0;
+}
+
+/**
+ * igc_ptp_init - Initialize PTP functionality
+ * @adapter: Board private structure
+ *
+ * This function is called at device probe to initialize the PTP
+ * functionality.
+ */
+void igc_ptp_init(struct igc_adapter *adapter)
+{
+	struct net_device *netdev = adapter->netdev;
+	struct igc_hw *hw = &adapter->hw;
+
+	switch (hw->mac.type) {
+	case igc_i225:
+		snprintf(adapter->ptp_caps.name, 16, "%pm", netdev->dev_addr);
+		adapter->ptp_caps.owner = THIS_MODULE;
+		adapter->ptp_caps.max_adj = 62499999;
+		adapter->ptp_caps.adjfine = igc_ptp_adjfine_i225;
+		adapter->ptp_caps.adjtime = igc_ptp_adjtime_i225;
+		adapter->ptp_caps.gettimex64 = igc_ptp_gettimex64_i225;
+		adapter->ptp_caps.settime64 = igc_ptp_settime_i225;
+		adapter->ptp_caps.enable = igc_ptp_feature_enable_i225;
+		break;
+	default:
+		adapter->ptp_clock = NULL;
+		return;
+	}
+
+	spin_lock_init(&adapter->tmreg_lock);
+	INIT_WORK(&adapter->ptp_tx_work, igc_ptp_tx_work);
+
+	adapter->tstamp_config.rx_filter = HWTSTAMP_FILTER_NONE;
+	adapter->tstamp_config.tx_type = HWTSTAMP_TX_OFF;
+
+	igc_ptp_reset(adapter);
+
+	adapter->ptp_clock = ptp_clock_register(&adapter->ptp_caps,
+						&adapter->pdev->dev);
+	if (IS_ERR(adapter->ptp_clock)) {
+		adapter->ptp_clock = NULL;
+		dev_err(&adapter->pdev->dev, "ptp_clock_register failed\n");
+	} else if (adapter->ptp_clock) {
+		dev_info(&adapter->pdev->dev, "added PHC on %s\n",
+			 adapter->netdev->name);
+		adapter->ptp_flags |= IGC_PTP_ENABLED;
+	}
+}
+
+/**
+ * igc_ptp_suspend - Disable PTP work items and prepare for suspend
+ * @adapter: Board private structure
+ *
+ * This function stops the overflow check work and PTP Tx timestamp work, and
+ * will prepare the device for OS suspend.
+ */
+void igc_ptp_suspend(struct igc_adapter *adapter)
+{
+	if (!(adapter->ptp_flags & IGC_PTP_ENABLED))
+		return;
+
+	cancel_work_sync(&adapter->ptp_tx_work);
+	if (adapter->ptp_tx_skb) {
+		dev_kfree_skb_any(adapter->ptp_tx_skb);
+		adapter->ptp_tx_skb = NULL;
+		clear_bit_unlock(__IGC_PTP_TX_IN_PROGRESS, &adapter->state);
+	}
+}
+
+/**
+ * igc_ptp_stop - Disable PTP device and stop the overflow check.
+ * @adapter: Board private structure.
+ *
+ * This function stops the PTP support and cancels the delayed work.
+ **/
+void igc_ptp_stop(struct igc_adapter *adapter)
+{
+	igc_ptp_suspend(adapter);
+
+	if (adapter->ptp_clock) {
+		ptp_clock_unregister(adapter->ptp_clock);
+		dev_info(&adapter->pdev->dev, "removed PHC on %s\n",
+			 adapter->netdev->name);
+		adapter->ptp_flags &= ~IGC_PTP_ENABLED;
+	}
+}
+
+/**
+ * igc_ptp_reset - Re-enable the adapter for PTP following a reset.
+ * @adapter: Board private structure.
+ *
+ * This function handles the reset work required to re-enable the PTP device.
+ **/
+void igc_ptp_reset(struct igc_adapter *adapter)
+{
+	struct igc_hw *hw = &adapter->hw;
+	unsigned long flags;
+
+	/* reset the tstamp_config */
+	igc_ptp_set_timestamp_mode(adapter, &adapter->tstamp_config);
+
+	spin_lock_irqsave(&adapter->tmreg_lock, flags);
+
+	switch (adapter->hw.mac.type) {
+	case igc_i225:
+		wr32(IGC_TSAUXC, 0x0);
+		wr32(IGC_TSSDP, 0x0);
+		wr32(IGC_TSIM, IGC_TSICR_INTERRUPTS);
+		wr32(IGC_IMS, IGC_IMS_TS);
+		break;
+	default:
+		/* No work to do. */
+		goto out;
+	}
+
+	/* Re-initialize the timer. */
+	if (hw->mac.type == igc_i225) {
+		struct timespec64 ts64 = ktime_to_timespec64(ktime_get_real());
+
+		igc_ptp_write_i225(adapter, &ts64);
+	} else {
+		timecounter_init(&adapter->tc, &adapter->cc,
+				 ktime_to_ns(ktime_get_real()));
+	}
+out:
+	spin_unlock_irqrestore(&adapter->tmreg_lock, flags);
+
+	wrfl();
+}
diff --git a/drivers/net/ethernet/intel/igc/igc_regs.h b/drivers/net/ethernet/intel/igc/igc_regs.h
index 93a9139f08c5..b5c72430a18b 100644
--- a/drivers/net/ethernet/intel/igc/igc_regs.h
+++ b/drivers/net/ethernet/intel/igc/igc_regs.h
@@ -209,6 +209,30 @@
 #define IGC_LENERRS	0x04138  /* Length Errors Count */
 #define IGC_HRMPC	0x0A018  /* Header Redirection Missed Packet Count */
 
+/* Time sync registers */
+#define IGC_TSICR	0x0B66C  /* Time Sync Interrupt Cause */
+#define IGC_TSIM	0x0B674  /* Time Sync Interrupt Mask Register */
+#define IGC_TSAUXC	0x0B640  /* Timesync Auxiliary Control register */
+#define IGC_TSYNCRXCTL	0x0B620  /* Rx Time Sync Control register - RW */
+#define IGC_TSYNCTXCTL	0x0B614  /* Tx Time Sync Control register - RW */
+#define IGC_TSYNCRXCFG	0x05F50  /* Time Sync Rx Configuration - RW */
+#define IGC_TSSDP	0x0003C  /* Time Sync SDP Configuration Register - RW */
+
+#define IGC_IMIR(_i)	(0x05A80 + ((_i) * 4))  /* Immediate Interrupt */
+#define IGC_IMIREXT(_i)	(0x05AA0 + ((_i) * 4))  /* Immediate INTR Ext*/
+
+#define IGC_FTQF(_n)	(0x059E0 + (4 * (_n)))  /* 5-tuple Queue Fltr */
+/* System Time Registers */
+#define IGC_SYSTIML	0x0B600  /* System time register Low - RO */
+#define IGC_SYSTIMH	0x0B604  /* System time register High - RO */
+#define IGC_SYSTIMR	0x0B6F8  /* System time register Residue */
+#define IGC_TIMINCA	0x0B608  /* Increment attributes register - RW */
+
+#define IGC_RXSTMPL	0x0B624  /* Rx timestamp Low - RO */
+#define IGC_RXSTMPH	0x0B628  /* Rx timestamp High - RO */
+#define IGC_TXSTMPL	0x0B618  /* Tx timestamp value Low - RO */
+#define IGC_TXSTMPH	0x0B61C  /* Tx timestamp value High - RO */
+
 /* Management registers */
 #define IGC_MANC	0x05820  /* Management Control - RW */
 
-- 
cgit v1.2.3-59-g8ed1b


From 62b0fb099ce2559293913d5bacf0c3aef08324e1 Mon Sep 17 00:00:00 2001
From: Amit Cohen <amitc@mellanox.com>
Date: Sun, 5 Jan 2020 18:20:50 +0200
Subject: mlxsw: spectrum: Disable SIP_CLASS_E check in hardware pipeline

The check drops packets if they need to be routed and their source IP is
from class E, i.e., belongs to 240.0.0.0/4 address range, but different
from 255.255.255.255.

Disable the check since the kernel forwards such packets and does not
drop them.

Signed-off-by: Amit Cohen <amitc@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum.c | 2 ++
 drivers/net/ethernet/mellanox/mlxsw/trap.h     | 1 +
 2 files changed, 3 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index fe02ffb1e3a8..c595060aa0ad 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -4545,6 +4545,8 @@ static const struct mlxsw_listener mlxsw_sp_listener[] = {
 	MLXSW_SP_RXL_MARK(DECAP_ECN0, TRAP_TO_CPU, ROUTER_EXP, false),
 	MLXSW_SP_RXL_MARK(IPV4_VRRP, TRAP_TO_CPU, VRRP, false),
 	MLXSW_SP_RXL_MARK(IPV6_VRRP, TRAP_TO_CPU, VRRP, false),
+	MLXSW_SP_RXL_NO_MARK(DISCARD_ING_ROUTER_SIP_CLASS_E, FORWARD,
+			     ROUTER_EXP, false),
 	/* PKT Sample trap */
 	MLXSW_RXL(mlxsw_sp_rx_listener_sample_func, PKT_SAMPLE, MIRROR_TO_CPU,
 		  false, SP_IP2ME, DISCARD),
diff --git a/drivers/net/ethernet/mellanox/mlxsw/trap.h b/drivers/net/ethernet/mellanox/mlxsw/trap.h
index 0c1c142bb6b0..e63cb82050c6 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/trap.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/trap.h
@@ -80,6 +80,7 @@ enum {
 	MLXSW_TRAP_ID_DISCARD_ING_ROUTER_UC_DIP_MC_DMAC = 0x161,
 	MLXSW_TRAP_ID_DISCARD_ING_ROUTER_DIP_LB = 0x162,
 	MLXSW_TRAP_ID_DISCARD_ING_ROUTER_SIP_MC = 0x163,
+	MLXSW_TRAP_ID_DISCARD_ING_ROUTER_SIP_CLASS_E = 0x164,
 	MLXSW_TRAP_ID_DISCARD_ING_ROUTER_SIP_LB = 0x165,
 	MLXSW_TRAP_ID_DISCARD_ING_ROUTER_CORRUPTED_IP_HDR = 0x167,
 	MLXSW_TRAP_ID_DISCARD_ING_ROUTER_IPV4_SIP_BC = 0x16A,
-- 
cgit v1.2.3-59-g8ed1b


From 383dbf70d99e31c4dacf0eca8411581b6ec0635d Mon Sep 17 00:00:00 2001
From: Amit Cohen <amitc@mellanox.com>
Date: Sun, 5 Jan 2020 18:20:51 +0200
Subject: selftests: forwarding: router: Add test case for source IP in class E
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add test case to check that packets are not dropped when they need to be
routed and their source IP in class E, (i.e., 240.0.0.0 –
255.255.255.254).

Signed-off-by: Amit Cohen <amitc@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/net/forwarding/router.sh | 38 +++++++++++++++++++++++-
 1 file changed, 37 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/net/forwarding/router.sh b/tools/testing/selftests/net/forwarding/router.sh
index a75cb51cc5bd..6ad652ad7e73 100755
--- a/tools/testing/selftests/net/forwarding/router.sh
+++ b/tools/testing/selftests/net/forwarding/router.sh
@@ -1,9 +1,15 @@
 #!/bin/bash
 # SPDX-License-Identifier: GPL-2.0
 
-ALL_TESTS="ping_ipv4 ping_ipv6"
+ALL_TESTS="
+	ping_ipv4
+	ping_ipv6
+	sip_in_class_e
+"
+
 NUM_NETIFS=4
 source lib.sh
+source tc_common.sh
 
 h1_create()
 {
@@ -64,6 +70,8 @@ router_create()
 	ip link set dev $rp1 up
 	ip link set dev $rp2 up
 
+	tc qdisc add dev $rp2 clsact
+
 	ip address add 192.0.2.1/24 dev $rp1
 	ip address add 2001:db8:1::1/64 dev $rp1
 
@@ -79,6 +87,8 @@ router_destroy()
 	ip address del 2001:db8:1::1/64 dev $rp1
 	ip address del 192.0.2.1/24 dev $rp1
 
+	tc qdisc del dev $rp2 clsact
+
 	ip link set dev $rp2 down
 	ip link set dev $rp1 down
 }
@@ -91,6 +101,8 @@ setup_prepare()
 	rp2=${NETIFS[p3]}
 	h2=${NETIFS[p4]}
 
+	rp1mac=$(mac_get $rp1)
+
 	vrf_prepare
 
 	h1_create
@@ -125,6 +137,30 @@ ping_ipv6()
 	ping6_test $h1 2001:db8:2::2
 }
 
+sip_in_class_e()
+{
+	RET=0
+
+	# Disable rpfilter to prevent packets to be dropped because of it.
+	sysctl_set net.ipv4.conf.all.rp_filter 0
+	sysctl_set net.ipv4.conf.$rp1.rp_filter 0
+
+	tc filter add dev $rp2 egress protocol ip pref 1 handle 101 \
+		flower src_ip 240.0.0.1 ip_proto udp action pass
+
+	$MZ $h1 -t udp "sp=54321,dp=12345" -c 5 -d 1msec \
+		-A 240.0.0.1 -b $rp1mac -B 198.51.100.2 -q
+
+	tc_check_packets "dev $rp2 egress" 101 5
+	check_err $? "Packets were dropped"
+
+	log_test "Source IP in class E"
+
+	tc filter del dev $rp2 egress protocol ip pref 1 handle 101 flower
+	sysctl_restore net.ipv4.conf.$rp1.rp_filter
+	sysctl_restore net.ipv4.conf.all.rp_filter
+}
+
 trap cleanup EXIT
 
 setup_prepare
-- 
cgit v1.2.3-59-g8ed1b


From 359ec56679efde731c09b2f1f9c93f65dd8e29df Mon Sep 17 00:00:00 2001
From: Amit Cohen <amitc@mellanox.com>
Date: Sun, 5 Jan 2020 18:20:52 +0200
Subject: mlxsw: spectrum: Disable MC_DMAC check in hardware pipeline

The check drops packets if they need to be routed and their multicast
MAC mismatched to their multicast destination IP.

For IPV4:
DMAC is mismatched if it is different from {01-00-5E-0 (25 bits),
DIP[22:0]}

For IPV6:
DMAC is mismatched if it is different from {33-33-0 (16 bits),
DIP[31:0]}

Disable the check since the kernel forwards such packets and does not
drop them.

Signed-off-by: Amit Cohen <amitc@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum.c | 2 ++
 drivers/net/ethernet/mellanox/mlxsw/trap.h     | 1 +
 2 files changed, 3 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index c595060aa0ad..114122b52166 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -4547,6 +4547,8 @@ static const struct mlxsw_listener mlxsw_sp_listener[] = {
 	MLXSW_SP_RXL_MARK(IPV6_VRRP, TRAP_TO_CPU, VRRP, false),
 	MLXSW_SP_RXL_NO_MARK(DISCARD_ING_ROUTER_SIP_CLASS_E, FORWARD,
 			     ROUTER_EXP, false),
+	MLXSW_SP_RXL_NO_MARK(DISCARD_ING_ROUTER_MC_DMAC, FORWARD,
+			     ROUTER_EXP, false),
 	/* PKT Sample trap */
 	MLXSW_RXL(mlxsw_sp_rx_listener_sample_func, PKT_SAMPLE, MIRROR_TO_CPU,
 		  false, SP_IP2ME, DISCARD),
diff --git a/drivers/net/ethernet/mellanox/mlxsw/trap.h b/drivers/net/ethernet/mellanox/mlxsw/trap.h
index e63cb82050c6..7ccec9e28cdd 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/trap.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/trap.h
@@ -83,6 +83,7 @@ enum {
 	MLXSW_TRAP_ID_DISCARD_ING_ROUTER_SIP_CLASS_E = 0x164,
 	MLXSW_TRAP_ID_DISCARD_ING_ROUTER_SIP_LB = 0x165,
 	MLXSW_TRAP_ID_DISCARD_ING_ROUTER_CORRUPTED_IP_HDR = 0x167,
+	MLXSW_TRAP_ID_DISCARD_ING_ROUTER_MC_DMAC = 0x168,
 	MLXSW_TRAP_ID_DISCARD_ING_ROUTER_IPV4_SIP_BC = 0x16A,
 	MLXSW_TRAP_ID_DISCARD_ING_ROUTER_IPV4_DIP_LOCAL_NET = 0x16B,
 	MLXSW_TRAP_ID_DISCARD_ROUTER_LPM4 = 0x17B,
-- 
cgit v1.2.3-59-g8ed1b


From 59b3a4f3481e89c4936b4baffea9a26a6846b70c Mon Sep 17 00:00:00 2001
From: Amit Cohen <amitc@mellanox.com>
Date: Sun, 5 Jan 2020 18:20:53 +0200
Subject: selftests: forwarding: router: Add test case for multicast
 destination MAC mismatch

Add test case to check that packets are not dropped when they need to be
routed and their multicast MAC mismatched to their multicast destination
IP.

i.e., destination IP is multicast and
	* for IPV4: DMAC !=  {01-00-5E-0 (25 bits), DIP[22:0]}
	* for IPV6: DMAC !=  {33-33-0 (16 bits), DIP[31:0]}

Signed-off-by: Amit Cohen <amitc@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/net/forwarding/router.sh | 82 ++++++++++++++++++++++++
 1 file changed, 82 insertions(+)

diff --git a/tools/testing/selftests/net/forwarding/router.sh b/tools/testing/selftests/net/forwarding/router.sh
index 6ad652ad7e73..e1f4d6145326 100755
--- a/tools/testing/selftests/net/forwarding/router.sh
+++ b/tools/testing/selftests/net/forwarding/router.sh
@@ -5,12 +5,17 @@ ALL_TESTS="
 	ping_ipv4
 	ping_ipv6
 	sip_in_class_e
+	mc_mac_mismatch
 "
 
 NUM_NETIFS=4
 source lib.sh
 source tc_common.sh
 
+require_command $MCD
+require_command $MC_CLI
+table_name=selftests
+
 h1_create()
 {
 	vrf_create "vrf-h1"
@@ -93,6 +98,25 @@ router_destroy()
 	ip link set dev $rp1 down
 }
 
+start_mcd()
+{
+	SMCROUTEDIR="$(mktemp -d)"
+
+	for ((i = 1; i <= $NUM_NETIFS; ++i)); do
+		echo "phyint ${NETIFS[p$i]} enable" >> \
+			$SMCROUTEDIR/$table_name.conf
+	done
+
+	$MCD -N -I $table_name -f $SMCROUTEDIR/$table_name.conf \
+		-P $SMCROUTEDIR/$table_name.pid
+}
+
+kill_mcd()
+{
+	pkill $MCD
+	rm -rf $SMCROUTEDIR
+}
+
 setup_prepare()
 {
 	h1=${NETIFS[p1]}
@@ -103,6 +127,8 @@ setup_prepare()
 
 	rp1mac=$(mac_get $rp1)
 
+	start_mcd
+
 	vrf_prepare
 
 	h1_create
@@ -125,6 +151,8 @@ cleanup()
 	h1_destroy
 
 	vrf_cleanup
+
+	kill_mcd
 }
 
 ping_ipv4()
@@ -161,6 +189,60 @@ sip_in_class_e()
 	sysctl_restore net.ipv4.conf.all.rp_filter
 }
 
+create_mcast_sg()
+{
+	local if_name=$1; shift
+	local s_addr=$1; shift
+	local mcast=$1; shift
+	local dest_ifs=${@}
+
+	$MC_CLI -I $table_name add $if_name $s_addr $mcast $dest_ifs
+}
+
+delete_mcast_sg()
+{
+	local if_name=$1; shift
+	local s_addr=$1; shift
+	local mcast=$1; shift
+	local dest_ifs=${@}
+
+	$MC_CLI -I $table_name remove $if_name $s_addr $mcast $dest_ifs
+}
+
+__mc_mac_mismatch()
+{
+	local desc=$1; shift
+	local proto=$1; shift
+	local sip=$1; shift
+	local dip=$1; shift
+	local flags=${1:-""}; shift
+	local dmac=01:02:03:04:05:06
+
+	RET=0
+
+	tc filter add dev $rp2 egress protocol $proto pref 1 handle 101 \
+		flower dst_ip $dip action pass
+
+	create_mcast_sg $rp1 $sip $dip $rp2
+
+	$MZ $flags $h1 -t udp "sp=54321,dp=12345" -c 5 -d 1msec -b $dmac \
+		-B $dip -q
+
+	tc_check_packets "dev $rp2 egress" 101 5
+	check_err $? "Packets were dropped"
+
+	log_test "Multicast MAC mismatch: $desc"
+
+	delete_mcast_sg $rp1 $sip $dip $rp2
+	tc filter del dev $rp2 egress protocol $proto pref 1 handle 101 flower
+}
+
+mc_mac_mismatch()
+{
+	__mc_mac_mismatch "IPv4" "ip" 192.0.2.2 225.1.2.3
+	__mc_mac_mismatch "IPv6" "ipv6" 2001:db8:1::2 ff0e::3 "-6"
+}
+
 trap cleanup EXIT
 
 setup_prepare
-- 
cgit v1.2.3-59-g8ed1b


From e317b0f77eb67b2ef1663a35c2ff58e58736f71f Mon Sep 17 00:00:00 2001
From: Amit Cohen <amitc@mellanox.com>
Date: Sun, 5 Jan 2020 18:20:54 +0200
Subject: mlxsw: spectrum: Disable SIP_DIP check in hardware pipeline

The check drops packets if they need to be routed and their source IP
equals to their destination IP.

Disable the check since the kernel forwards such packets and does not
drop them.

Signed-off-by: Amit Cohen <amitc@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum.c | 2 ++
 drivers/net/ethernet/mellanox/mlxsw/trap.h     | 1 +
 2 files changed, 3 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index 114122b52166..0214252cc6b4 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -4549,6 +4549,8 @@ static const struct mlxsw_listener mlxsw_sp_listener[] = {
 			     ROUTER_EXP, false),
 	MLXSW_SP_RXL_NO_MARK(DISCARD_ING_ROUTER_MC_DMAC, FORWARD,
 			     ROUTER_EXP, false),
+	MLXSW_SP_RXL_NO_MARK(DISCARD_ING_ROUTER_SIP_DIP, FORWARD,
+			     ROUTER_EXP, false),
 	/* PKT Sample trap */
 	MLXSW_RXL(mlxsw_sp_rx_listener_sample_func, PKT_SAMPLE, MIRROR_TO_CPU,
 		  false, SP_IP2ME, DISCARD),
diff --git a/drivers/net/ethernet/mellanox/mlxsw/trap.h b/drivers/net/ethernet/mellanox/mlxsw/trap.h
index 7ccec9e28cdd..02c68820eae5 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/trap.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/trap.h
@@ -84,6 +84,7 @@ enum {
 	MLXSW_TRAP_ID_DISCARD_ING_ROUTER_SIP_LB = 0x165,
 	MLXSW_TRAP_ID_DISCARD_ING_ROUTER_CORRUPTED_IP_HDR = 0x167,
 	MLXSW_TRAP_ID_DISCARD_ING_ROUTER_MC_DMAC = 0x168,
+	MLXSW_TRAP_ID_DISCARD_ING_ROUTER_SIP_DIP = 0x169,
 	MLXSW_TRAP_ID_DISCARD_ING_ROUTER_IPV4_SIP_BC = 0x16A,
 	MLXSW_TRAP_ID_DISCARD_ING_ROUTER_IPV4_DIP_LOCAL_NET = 0x16B,
 	MLXSW_TRAP_ID_DISCARD_ROUTER_LPM4 = 0x17B,
-- 
cgit v1.2.3-59-g8ed1b


From 6e734f863c872a67f1b21129e762263958d17744 Mon Sep 17 00:00:00 2001
From: Amit Cohen <amitc@mellanox.com>
Date: Sun, 5 Jan 2020 18:20:55 +0200
Subject: selftests: forwarding: router: Add test case for source IP equals
 destination IP

Add test case to check that packets are not dropped when they need to be
routed and their source IP equals to their destination IP.

Signed-off-by: Amit Cohen <amitc@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/net/forwarding/router.sh | 44 ++++++++++++++++++++++++
 1 file changed, 44 insertions(+)

diff --git a/tools/testing/selftests/net/forwarding/router.sh b/tools/testing/selftests/net/forwarding/router.sh
index e1f4d6145326..c894f7240417 100755
--- a/tools/testing/selftests/net/forwarding/router.sh
+++ b/tools/testing/selftests/net/forwarding/router.sh
@@ -6,6 +6,8 @@ ALL_TESTS="
 	ping_ipv6
 	sip_in_class_e
 	mc_mac_mismatch
+	ipv4_sip_equal_dip
+	ipv6_sip_equal_dip
 "
 
 NUM_NETIFS=4
@@ -243,6 +245,48 @@ mc_mac_mismatch()
 	__mc_mac_mismatch "IPv6" "ipv6" 2001:db8:1::2 ff0e::3 "-6"
 }
 
+ipv4_sip_equal_dip()
+{
+	RET=0
+
+	# Disable rpfilter to prevent packets to be dropped because of it.
+	sysctl_set net.ipv4.conf.all.rp_filter 0
+	sysctl_set net.ipv4.conf.$rp1.rp_filter 0
+
+	tc filter add dev $rp2 egress protocol ip pref 1 handle 101 \
+		flower src_ip 198.51.100.2  action pass
+
+	$MZ $h1 -t udp "sp=54321,dp=12345" -c 5 -d 1msec \
+		-A 198.51.100.2 -b $rp1mac -B 198.51.100.2 -q
+
+	tc_check_packets "dev $rp2 egress" 101 5
+	check_err $? "Packets were dropped"
+
+	log_test "Source IP is equal to destination IP: IPv4"
+
+	tc filter del dev $rp2 egress protocol ip pref 1 handle 101 flower
+	sysctl_restore net.ipv4.conf.$rp1.rp_filter
+	sysctl_restore net.ipv4.conf.all.rp_filter
+}
+
+ipv6_sip_equal_dip()
+{
+	RET=0
+
+	tc filter add dev $rp2 egress protocol ipv6 pref 1 handle 101 \
+		flower src_ip 2001:db8:2::2 action pass
+
+	$MZ -6 $h1 -t udp "sp=54321,dp=12345" -c 5 -d 1msec \
+		-A 2001:db8:2::2 -b $rp1mac -B 2001:db8:2::2 -q
+
+	tc_check_packets "dev $rp2 egress" 101 5
+	check_err $? "Packets were dropped"
+
+	log_test "Source IP is equal to destination IP: IPv6"
+
+	tc filter del dev $rp2 egress protocol ipv6 pref 1 handle 101 flower
+}
+
 trap cleanup EXIT
 
 setup_prepare
-- 
cgit v1.2.3-59-g8ed1b


From ca360db4b82537d0cbc7a54018b2096c757c4828 Mon Sep 17 00:00:00 2001
From: Amit Cohen <amitc@mellanox.com>
Date: Sun, 5 Jan 2020 18:20:56 +0200
Subject: mlxsw: spectrum: Disable DIP_LINK_LOCAL check in hardware pipeline

The check drops packets if they need to be routed and their destination
IP is link-local, i.e., belongs to 169.254.0.0/16 address range.

Disable the check since the kernel forwards such packets and does not
drop them.

Signed-off-by: Amit Cohen <amitc@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum.c | 2 ++
 drivers/net/ethernet/mellanox/mlxsw/trap.h     | 1 +
 2 files changed, 3 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index 0214252cc6b4..431c3765b545 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -4551,6 +4551,8 @@ static const struct mlxsw_listener mlxsw_sp_listener[] = {
 			     ROUTER_EXP, false),
 	MLXSW_SP_RXL_NO_MARK(DISCARD_ING_ROUTER_SIP_DIP, FORWARD,
 			     ROUTER_EXP, false),
+	MLXSW_SP_RXL_NO_MARK(DISCARD_ING_ROUTER_DIP_LINK_LOCAL, FORWARD,
+			     ROUTER_EXP, false),
 	/* PKT Sample trap */
 	MLXSW_RXL(mlxsw_sp_rx_listener_sample_func, PKT_SAMPLE, MIRROR_TO_CPU,
 		  false, SP_IP2ME, DISCARD),
diff --git a/drivers/net/ethernet/mellanox/mlxsw/trap.h b/drivers/net/ethernet/mellanox/mlxsw/trap.h
index 02c68820eae5..3d2331be05d8 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/trap.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/trap.h
@@ -87,6 +87,7 @@ enum {
 	MLXSW_TRAP_ID_DISCARD_ING_ROUTER_SIP_DIP = 0x169,
 	MLXSW_TRAP_ID_DISCARD_ING_ROUTER_IPV4_SIP_BC = 0x16A,
 	MLXSW_TRAP_ID_DISCARD_ING_ROUTER_IPV4_DIP_LOCAL_NET = 0x16B,
+	MLXSW_TRAP_ID_DISCARD_ING_ROUTER_DIP_LINK_LOCAL = 0x16C,
 	MLXSW_TRAP_ID_DISCARD_ROUTER_LPM4 = 0x17B,
 	MLXSW_TRAP_ID_DISCARD_ROUTER_LPM6 = 0x17C,
 	MLXSW_TRAP_ID_DISCARD_IPV6_MC_DIP_RESERVED_SCOPE = 0x1B0,
-- 
cgit v1.2.3-59-g8ed1b


From ef11ffa21b75c97780f345615e46e5e985406513 Mon Sep 17 00:00:00 2001
From: Amit Cohen <amitc@mellanox.com>
Date: Sun, 5 Jan 2020 18:20:57 +0200
Subject: selftests: forwarding: router: Add test case for destination IP
 link-local

Add test case to check that packets are not dropped when they need to be
routed and their destination is link-local, i.e., 169.254.0.0/16.

Signed-off-by: Amit Cohen <amitc@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/net/forwarding/router.sh | 25 ++++++++++++++++++++++++
 1 file changed, 25 insertions(+)

diff --git a/tools/testing/selftests/net/forwarding/router.sh b/tools/testing/selftests/net/forwarding/router.sh
index c894f7240417..057f91b05098 100755
--- a/tools/testing/selftests/net/forwarding/router.sh
+++ b/tools/testing/selftests/net/forwarding/router.sh
@@ -8,6 +8,7 @@ ALL_TESTS="
 	mc_mac_mismatch
 	ipv4_sip_equal_dip
 	ipv6_sip_equal_dip
+	ipv4_dip_link_local
 "
 
 NUM_NETIFS=4
@@ -287,6 +288,30 @@ ipv6_sip_equal_dip()
 	tc filter del dev $rp2 egress protocol ipv6 pref 1 handle 101 flower
 }
 
+ipv4_dip_link_local()
+{
+	local dip=169.254.1.1
+
+	RET=0
+
+	tc filter add dev $rp2 egress protocol ip pref 1 handle 101 \
+		flower dst_ip $dip action pass
+
+	ip neigh add 169.254.1.1 lladdr 00:11:22:33:44:55 dev $rp2
+	ip route add 169.254.1.0/24 dev $rp2
+
+	$MZ $h1 -t udp "sp=54321,dp=12345" -c 5 -d 1msec -b $rp1mac -B $dip -q
+
+	tc_check_packets "dev $rp2 egress" 101 5
+	check_err $? "Packets were dropped"
+
+	log_test "IPv4 destination IP is link-local"
+
+	ip route del 169.254.1.0/24 dev $rp2
+	ip neigh del 169.254.1.1 lladdr 00:11:22:33:44:55 dev $rp2
+	tc filter del dev $rp2 egress protocol ip pref 1 handle 101 flower
+}
+
 trap cleanup EXIT
 
 setup_prepare
-- 
cgit v1.2.3-59-g8ed1b


From b289ba5e07105548b8219695e5443d807a825eb8 Mon Sep 17 00:00:00 2001
From: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Date: Sun, 5 Jan 2020 18:36:07 +0100
Subject: gtp: simplify error handling code in 'gtp_encap_enable()'

'gtp_encap_disable_sock(sk)' handles the case where sk is NULL, so there
is no need to test it before calling the function.

This saves a few line of code.

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/gtp.c | 9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)

diff --git a/drivers/net/gtp.c b/drivers/net/gtp.c
index fca471e27f39..dea3978b223c 100644
--- a/drivers/net/gtp.c
+++ b/drivers/net/gtp.c
@@ -851,8 +851,7 @@ static int gtp_encap_enable(struct gtp_dev *gtp, struct nlattr *data[])
 
 		sk1u = gtp_encap_enable_socket(fd1, UDP_ENCAP_GTP1U, gtp);
 		if (IS_ERR(sk1u)) {
-			if (sk0)
-				gtp_encap_disable_sock(sk0);
+			gtp_encap_disable_sock(sk0);
 			return PTR_ERR(sk1u);
 		}
 	}
@@ -860,10 +859,8 @@ static int gtp_encap_enable(struct gtp_dev *gtp, struct nlattr *data[])
 	if (data[IFLA_GTP_ROLE]) {
 		role = nla_get_u32(data[IFLA_GTP_ROLE]);
 		if (role > GTP_ROLE_SGSN) {
-			if (sk0)
-				gtp_encap_disable_sock(sk0);
-			if (sk1u)
-				gtp_encap_disable_sock(sk1u);
+			gtp_encap_disable_sock(sk0);
+			gtp_encap_disable_sock(sk1u);
 			return -EINVAL;
 		}
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 17aa23ee72342efdc3b214f582399dd22478887f Mon Sep 17 00:00:00 2001
From: David Ahern <dsahern@gmail.com>
Date: Sun, 5 Jan 2020 20:02:05 -0800
Subject: fcnal-test: Fix vrf argument in local tcp tests

The recent MD5 tests added duplicate configuration in the default VRF.
This change exposed a bug in existing tests designed to verify no
connection when client and server are not in the same domain. The
server should be running bound to the vrf device with the client run
in the default VRF (the -2 option is meant for validating connection
data). Fix the option for both tests.

While technically this is a bug in previous releases, the tests are
properly failing since the default VRF does not have any routing
configuration so there really is no need to backport to prior releases.

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/net/fcnal-test.sh | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/tools/testing/selftests/net/fcnal-test.sh b/tools/testing/selftests/net/fcnal-test.sh
index 15cb8ab68c7c..fb5c55dd6df8 100755
--- a/tools/testing/selftests/net/fcnal-test.sh
+++ b/tools/testing/selftests/net/fcnal-test.sh
@@ -1165,8 +1165,8 @@ ipv4_tcp_vrf()
 	for a in ${NSA_IP} ${VRF_IP}
 	do
 		log_start
-		show_hint "Should fail 'No route to host' since client is not bound to VRF"
-		run_cmd nettest -s -2 ${VRF} &
+		show_hint "Should fail 'Connection refused' since client is not bound to VRF"
+		run_cmd nettest -s -d ${VRF} &
 		sleep 1
 		run_cmd nettest -r ${a}
 		log_test_addr ${a} $? 1 "Global server, local connection"
@@ -2663,8 +2663,8 @@ ipv6_tcp_vrf()
 	for a in ${NSA_IP6} ${VRF_IP6}
 	do
 		log_start
-		show_hint "Fails 'No route to host' since client is not in VRF"
-		run_cmd nettest -6 -s -2 ${VRF} &
+		show_hint "Fails 'Connection refused' since client is not in VRF"
+		run_cmd nettest -6 -s -d ${VRF} &
 		sleep 1
 		run_cmd nettest -6 -r ${a}
 		log_test_addr ${a} $? 1 "Global server, local connection"
-- 
cgit v1.2.3-59-g8ed1b


From a69faa09102bd2138bac5eea9d27dcf196118832 Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Mon, 6 Jan 2020 07:39:31 +0100
Subject: wil6210: get rid of begin() and complete() ethtool_ops

The wil6210 driver locks a mutex in begin() ethtool_ops callback and
unlocks it in complete() so that all ethtool requests are serialized. This
is not going to work correctly with netlink interface; e.g. when ioctl
triggers a netlink notification, netlink code would call begin() again
while the mutex taken by ioctl code is still held by the same task.

Let's get rid of the begin() and complete() callbacks and move the mutex
locking into the remaining ethtool_ops handlers except get_drvinfo which
only copies strings that are not changing so that there is no need for
serialization.

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/wireless/ath/wil6210/ethtool.c | 43 +++++++++++-------------------
 1 file changed, 16 insertions(+), 27 deletions(-)

diff --git a/drivers/net/wireless/ath/wil6210/ethtool.c b/drivers/net/wireless/ath/wil6210/ethtool.c
index 912c4eaf017b..fef10886ca4a 100644
--- a/drivers/net/wireless/ath/wil6210/ethtool.c
+++ b/drivers/net/wireless/ath/wil6210/ethtool.c
@@ -11,26 +11,6 @@
 
 #include "wil6210.h"
 
-static int wil_ethtoolops_begin(struct net_device *ndev)
-{
-	struct wil6210_priv *wil = ndev_to_wil(ndev);
-
-	mutex_lock(&wil->mutex);
-
-	wil_dbg_misc(wil, "ethtoolops_begin\n");
-
-	return 0;
-}
-
-static void wil_ethtoolops_complete(struct net_device *ndev)
-{
-	struct wil6210_priv *wil = ndev_to_wil(ndev);
-
-	wil_dbg_misc(wil, "ethtoolops_complete\n");
-
-	mutex_unlock(&wil->mutex);
-}
-
 static int wil_ethtoolops_get_coalesce(struct net_device *ndev,
 				       struct ethtool_coalesce *cp)
 {
@@ -39,11 +19,12 @@ static int wil_ethtoolops_get_coalesce(struct net_device *ndev,
 	u32 rx_itr_en, rx_itr_val = 0;
 	int ret;
 
+	mutex_lock(&wil->mutex);
 	wil_dbg_misc(wil, "ethtoolops_get_coalesce\n");
 
 	ret = wil_pm_runtime_get(wil);
 	if (ret < 0)
-		return ret;
+		goto out;
 
 	tx_itr_en = wil_r(wil, RGF_DMA_ITR_TX_CNT_CTL);
 	if (tx_itr_en & BIT_DMA_ITR_TX_CNT_CTL_EN)
@@ -57,7 +38,11 @@ static int wil_ethtoolops_get_coalesce(struct net_device *ndev,
 
 	cp->tx_coalesce_usecs = tx_itr_val;
 	cp->rx_coalesce_usecs = rx_itr_val;
-	return 0;
+	ret = 0;
+
+out:
+	mutex_unlock(&wil->mutex);
+	return ret;
 }
 
 static int wil_ethtoolops_set_coalesce(struct net_device *ndev,
@@ -67,12 +52,14 @@ static int wil_ethtoolops_set_coalesce(struct net_device *ndev,
 	struct wireless_dev *wdev = ndev->ieee80211_ptr;
 	int ret;
 
+	mutex_lock(&wil->mutex);
 	wil_dbg_misc(wil, "ethtoolops_set_coalesce: rx %d usec, tx %d usec\n",
 		     cp->rx_coalesce_usecs, cp->tx_coalesce_usecs);
 
 	if (wdev->iftype == NL80211_IFTYPE_MONITOR) {
 		wil_dbg_misc(wil, "No IRQ coalescing in monitor mode\n");
-		return -EINVAL;
+		ret = -EINVAL;
+		goto out;
 	}
 
 	/* only @rx_coalesce_usecs and @tx_coalesce_usecs supported,
@@ -88,24 +75,26 @@ static int wil_ethtoolops_set_coalesce(struct net_device *ndev,
 
 	ret = wil_pm_runtime_get(wil);
 	if (ret < 0)
-		return ret;
+		goto out;
 
 	wil->txrx_ops.configure_interrupt_moderation(wil);
 
 	wil_pm_runtime_put(wil);
+	ret = 0;
 
-	return 0;
+out:
+	mutex_unlock(&wil->mutex);
+	return ret;
 
 out_bad:
 	wil_dbg_misc(wil, "Unsupported coalescing params. Raw command:\n");
 	print_hex_dump_debug("DBG[MISC] coal ", DUMP_PREFIX_OFFSET, 16, 4,
 			     cp, sizeof(*cp), false);
+	mutex_unlock(&wil->mutex);
 	return -EINVAL;
 }
 
 static const struct ethtool_ops wil_ethtool_ops = {
-	.begin		= wil_ethtoolops_begin,
-	.complete	= wil_ethtoolops_complete,
 	.get_drvinfo	= cfg80211_get_drvinfo,
 	.get_coalesce	= wil_ethtoolops_get_coalesce,
 	.set_coalesce	= wil_ethtoolops_set_coalesce,
-- 
cgit v1.2.3-59-g8ed1b


From 71f711a4f1dd1f6ebc26870ef17ae79581f37ae7 Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Mon, 6 Jan 2020 07:39:36 +0100
Subject: via-velocity: allow nesting of ethtool_ops begin() and complete()

Unlike most networking drivers using begin() and complete() ethtool_ops
callbacks to resume a device which is down and suspend it again when done,
via-velocity does not use standard refcounted infrastructure but sets
device sleep state directly.

With the introduction of netlink ethtool interface, we may have nested
begin-complete blocks so that inner complete() would put the device back to
sleep for the rest of the outer block.

To avoid rewriting an old and not very actively developed driver, just add
a nesting counter and only perform resume and suspend on the outermost
level.

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/via/via-velocity.c | 14 ++++++++++----
 drivers/net/ethernet/via/via-velocity.h |  1 +
 2 files changed, 11 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/via/via-velocity.c b/drivers/net/ethernet/via/via-velocity.c
index 346e44115c4e..4b556b74541a 100644
--- a/drivers/net/ethernet/via/via-velocity.c
+++ b/drivers/net/ethernet/via/via-velocity.c
@@ -3257,12 +3257,16 @@ static struct platform_driver velocity_platform_driver = {
  *	@dev: network device
  *
  *	Called before an ethtool operation. We need to make sure the
- *	chip is out of D3 state before we poke at it.
+ *	chip is out of D3 state before we poke at it. In case of ethtool
+ *	ops nesting, only wake the device up in the outermost block.
  */
 static int velocity_ethtool_up(struct net_device *dev)
 {
 	struct velocity_info *vptr = netdev_priv(dev);
-	if (!netif_running(dev))
+
+	if (vptr->ethtool_ops_nesting == U32_MAX)
+		return -EBUSY;
+	if (!vptr->ethtool_ops_nesting++ && !netif_running(dev))
 		velocity_set_power_state(vptr, PCI_D0);
 	return 0;
 }
@@ -3272,12 +3276,14 @@ static int velocity_ethtool_up(struct net_device *dev)
  *	@dev: network device
  *
  *	Called after an ethtool operation. Restore the chip back to D3
- *	state if it isn't running.
+ *	state if it isn't running. In case of ethtool ops nesting, only
+ *	put the device to sleep in the outermost block.
  */
 static void velocity_ethtool_down(struct net_device *dev)
 {
 	struct velocity_info *vptr = netdev_priv(dev);
-	if (!netif_running(dev))
+
+	if (!--vptr->ethtool_ops_nesting && !netif_running(dev))
 		velocity_set_power_state(vptr, PCI_D3hot);
 }
 
diff --git a/drivers/net/ethernet/via/via-velocity.h b/drivers/net/ethernet/via/via-velocity.h
index cdfe7809e3c1..f196e71d2c04 100644
--- a/drivers/net/ethernet/via/via-velocity.h
+++ b/drivers/net/ethernet/via/via-velocity.h
@@ -1483,6 +1483,7 @@ struct velocity_info {
 	struct velocity_context context;
 
 	u32 ticks;
+	u32 ethtool_ops_nesting;
 
 	u8 rev_id;
 
-- 
cgit v1.2.3-59-g8ed1b


From 4ac0ac847f05bc49efb6ea9e8770f1dfec31f540 Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Mon, 6 Jan 2020 07:39:41 +0100
Subject: epic100: allow nesting of ethtool_ops begin() and complete()

Unlike most networking drivers using begin() and complete() ethtool_ops
callbacks to resume a device which is down and suspend it again when done,
epic100 does not use standard refcounted infrastructure but sets device
sleep state directly.

With the introduction of netlink ethtool interface, we may have nested
begin-complete blocks so that inner complete() would put the device back to
sleep for the rest of the outer block.

To avoid rewriting an old and not very actively developed driver, just add
a nesting counter and only perform resume and suspend on the outermost
level.

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/smsc/epic100.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/smsc/epic100.c b/drivers/net/ethernet/smsc/epic100.c
index 912760e8514c..61ddee0c2a2e 100644
--- a/drivers/net/ethernet/smsc/epic100.c
+++ b/drivers/net/ethernet/smsc/epic100.c
@@ -280,6 +280,7 @@ struct epic_private {
 	signed char phys[4];				/* MII device addresses. */
 	u16 advertising;					/* NWay media advertisement */
 	int mii_phy_cnt;
+	u32 ethtool_ops_nesting;
 	struct mii_if_info mii;
 	unsigned int tx_full:1;				/* The Tx queue is full. */
 	unsigned int default_port:4;		/* Last dev->if_port value. */
@@ -1435,8 +1436,10 @@ static int ethtool_begin(struct net_device *dev)
 	struct epic_private *ep = netdev_priv(dev);
 	void __iomem *ioaddr = ep->ioaddr;
 
+	if (ep->ethtool_ops_nesting == U32_MAX)
+		return -EBUSY;
 	/* power-up, if interface is down */
-	if (!netif_running(dev)) {
+	if (!ep->ethtool_ops_nesting++ && !netif_running(dev)) {
 		ew32(GENCTL, 0x0200);
 		ew32(NVCTL, (er32(NVCTL) & ~0x003c) | 0x4800);
 	}
@@ -1449,7 +1452,7 @@ static void ethtool_complete(struct net_device *dev)
 	void __iomem *ioaddr = ep->ioaddr;
 
 	/* power-down, if interface is down */
-	if (!netif_running(dev)) {
+	if (!--ep->ethtool_ops_nesting && !netif_running(dev)) {
 		ew32(GENCTL, 0x0008);
 		ew32(NVCTL, (er32(NVCTL) & ~0x483c) | 0x0000);
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 81b055205e8ba2d400c8fa5845ba540a4a880a3a Mon Sep 17 00:00:00 2001
From: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Date: Mon, 2 Dec 2019 15:19:50 -0800
Subject: igc: Add support for RX timestamping

This adds support for timestamping received packets.

It is based on the i210, as many features of i225 work the same way.
The main difference from i210 is that i225 has support for choosing
the timer register to use when timestamping packets. Right now, we
only support using timer 0. The other difference is that i225 stores
two timestamps in the receive descriptor, right now, we only retrieve
one.

Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc.h         |  17 ++
 drivers/net/ethernet/intel/igc/igc_defines.h |  38 ++++
 drivers/net/ethernet/intel/igc/igc_main.c    |  10 +
 drivers/net/ethernet/intel/igc/igc_ptp.c     | 272 +++++++++++++++++++++++++++
 drivers/net/ethernet/intel/igc/igc_regs.h    |   3 +
 5 files changed, 340 insertions(+)

diff --git a/drivers/net/ethernet/intel/igc/igc.h b/drivers/net/ethernet/intel/igc/igc.h
index 7586f237747a..9b0d1cbf941f 100644
--- a/drivers/net/ethernet/intel/igc/igc.h
+++ b/drivers/net/ethernet/intel/igc/igc.h
@@ -107,6 +107,20 @@ extern char igc_driver_version[];
 #define AUTO_ALL_MODES		0
 #define IGC_RX_HDR_LEN			IGC_RXBUFFER_256
 
+/* Transmit and receive latency (for PTP timestamps) */
+/* FIXME: These values were estimated using the ones that i210 has as
+ * basis, they seem to provide good numbers with ptp4l/phc2sys, but we
+ * need to confirm them.
+ */
+#define IGC_I225_TX_LATENCY_10		9542
+#define IGC_I225_TX_LATENCY_100		1024
+#define IGC_I225_TX_LATENCY_1000	178
+#define IGC_I225_TX_LATENCY_2500	64
+#define IGC_I225_RX_LATENCY_10		20662
+#define IGC_I225_RX_LATENCY_100		2213
+#define IGC_I225_RX_LATENCY_1000	448
+#define IGC_I225_RX_LATENCY_2500	160
+
 /* RX and TX descriptor control thresholds.
  * PTHRESH - MAC will consider prefetch if it has fewer than this number of
  *           descriptors available in its onboard memory.
@@ -539,6 +553,9 @@ int igc_erase_filter(struct igc_adapter *adapter,
 void igc_ptp_init(struct igc_adapter *adapter);
 void igc_ptp_reset(struct igc_adapter *adapter);
 void igc_ptp_stop(struct igc_adapter *adapter);
+void igc_ptp_rx_rgtstamp(struct igc_q_vector *q_vector, struct sk_buff *skb);
+void igc_ptp_rx_pktstamp(struct igc_q_vector *q_vector, void *va,
+			 struct sk_buff *skb);
 int igc_ptp_set_ts_config(struct net_device *netdev, struct ifreq *ifr);
 int igc_ptp_get_ts_config(struct net_device *netdev, struct ifreq *ifr);
 #define igc_rx_pg_size(_ring) (PAGE_SIZE << igc_rx_pg_order(_ring))
diff --git a/drivers/net/ethernet/intel/igc/igc_defines.h b/drivers/net/ethernet/intel/igc/igc_defines.h
index 12f9127e40b7..9dede618362f 100644
--- a/drivers/net/ethernet/intel/igc/igc_defines.h
+++ b/drivers/net/ethernet/intel/igc/igc_defines.h
@@ -314,12 +314,21 @@
 #define IGC_RCTL_RDMTS_HALF	0x00000000 /* Rx desc min thresh size */
 #define IGC_RCTL_BAM		0x00008000 /* broadcast enable */
 
+/* Split Replication Receive Control */
+#define IGC_SRRCTL_TIMESTAMP		0x40000000
+#define IGC_SRRCTL_TIMER1SEL(timer)	(((timer) & 0x3) << 14)
+#define IGC_SRRCTL_TIMER0SEL(timer)	(((timer) & 0x3) << 17)
+
 /* Receive Descriptor bit definitions */
 #define IGC_RXD_STAT_EOP	0x02	/* End of Packet */
 #define IGC_RXD_STAT_IXSM	0x04	/* Ignore checksum */
 #define IGC_RXD_STAT_UDPCS	0x10	/* UDP xsum calculated */
 #define IGC_RXD_STAT_TCPCS	0x20	/* TCP xsum calculated */
 
+/* Advanced Receive Descriptor bit definitions */
+#define IGC_RXDADV_STAT_TSIP	0x08000 /* timestamp in packet */
+#define IGC_RXDADV_STAT_TS	0x10000 /* Pkt was time stamped */
+
 #define IGC_RXDEXT_STATERR_CE		0x01000000
 #define IGC_RXDEXT_STATERR_SE		0x02000000
 #define IGC_RXDEXT_STATERR_SEQ		0x04000000
@@ -356,6 +365,7 @@
 
 #define I225_RXPBSIZE_DEFAULT	0x000000A2 /* RXPBSIZE default */
 #define I225_TXPBSIZE_DEFAULT	0x04000014 /* TXPBSIZE default */
+#define IGC_RXPBS_CFG_TS_EN	0x80000000 /* Timestamp in Rx buffer */
 
 /* Time Sync Interrupt Causes */
 #define IGC_TSICR_SYS_WRAP	BIT(0) /* SYSTIM Wrap around. */
@@ -367,6 +377,34 @@
 
 #define IGC_TSICR_INTERRUPTS	IGC_TSICR_TXTS
 
+/* PTP Queue Filter */
+#define IGC_ETQF_1588		BIT(30)
+
+#define IGC_FTQF_VF_BP		0x00008000
+#define IGC_FTQF_1588_TIME_STAMP	0x08000000
+#define IGC_FTQF_MASK			0xF0000000
+#define IGC_FTQF_MASK_PROTO_BP	0x10000000
+
+/* Time Sync Receive Control bit definitions */
+#define IGC_TSYNCRXCTL_VALID		0x00000001  /* Rx timestamp valid */
+#define IGC_TSYNCRXCTL_TYPE_MASK	0x0000000E  /* Rx type mask */
+#define IGC_TSYNCRXCTL_TYPE_L2_V2	0x00
+#define IGC_TSYNCRXCTL_TYPE_L4_V1	0x02
+#define IGC_TSYNCRXCTL_TYPE_L2_L4_V2	0x04
+#define IGC_TSYNCRXCTL_TYPE_ALL		0x08
+#define IGC_TSYNCRXCTL_TYPE_EVENT_V2	0x0A
+#define IGC_TSYNCRXCTL_ENABLED		0x00000010  /* enable Rx timestamping */
+#define IGC_TSYNCRXCTL_SYSCFI		0x00000020  /* Sys clock frequency */
+
+/* Time Sync Receive Configuration */
+#define IGC_TSYNCRXCFG_PTP_V1_CTRLT_MASK	0x000000FF
+#define IGC_TSYNCRXCFG_PTP_V1_SYNC_MESSAGE	0x00
+#define IGC_TSYNCRXCFG_PTP_V1_DELAY_REQ_MESSAGE	0x01
+
+/* Immediate Interrupt Receive Extended */
+#define IGC_IMIREXT_CTRL_BP	0x00080000  /* Bypass check of ctrl bits */
+#define IGC_IMIREXT_SIZE_BP	0x00001000  /* Packet size bypass */
+
 /* Receive Checksum Control */
 #define IGC_RXCSUM_CRCOFL	0x00000800   /* CRC32 offload enable */
 #define IGC_RXCSUM_PCSD		0x00002000   /* packet checksum disabled */
diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index f242205792e4..504dbb74ca75 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -1299,6 +1299,10 @@ static void igc_process_skb_fields(struct igc_ring *rx_ring,
 
 	igc_rx_checksum(rx_ring, rx_desc, skb);
 
+	if (igc_test_staterr(rx_desc, IGC_RXDADV_STAT_TS) &&
+	    !igc_test_staterr(rx_desc, IGC_RXDADV_STAT_TSIP))
+		igc_ptp_rx_rgtstamp(rx_ring->q_vector, skb);
+
 	skb_record_rx_queue(skb, rx_ring->queue_index);
 
 	skb->protocol = eth_type_trans(skb, rx_ring->netdev);
@@ -1418,6 +1422,12 @@ static struct sk_buff *igc_construct_skb(struct igc_ring *rx_ring,
 	if (unlikely(!skb))
 		return NULL;
 
+	if (unlikely(igc_test_staterr(rx_desc, IGC_RXDADV_STAT_TSIP))) {
+		igc_ptp_rx_pktstamp(rx_ring->q_vector, va, skb);
+		va += IGC_TS_HDR_LEN;
+		size -= IGC_TS_HDR_LEN;
+	}
+
 	/* Determine available headroom for copy */
 	headlen = size;
 	if (headlen > IGC_RX_HDR_LEN)
diff --git a/drivers/net/ethernet/intel/igc/igc_ptp.c b/drivers/net/ethernet/intel/igc/igc_ptp.c
index bab384880f96..093ffc9312df 100644
--- a/drivers/net/ethernet/intel/igc/igc_ptp.c
+++ b/drivers/net/ethernet/intel/igc/igc_ptp.c
@@ -134,9 +134,281 @@ static int igc_ptp_feature_enable_i225(struct ptp_clock_info *ptp,
 	return -EOPNOTSUPP;
 }
 
+/**
+ * igc_ptp_systim_to_hwtstamp - convert system time value to HW timestamp
+ * @adapter: board private structure
+ * @hwtstamps: timestamp structure to update
+ * @systim: unsigned 64bit system time value
+ *
+ * We need to convert the system time value stored in the RX/TXSTMP registers
+ * into a hwtstamp which can be used by the upper level timestamping functions.
+ **/
+static void igc_ptp_systim_to_hwtstamp(struct igc_adapter *adapter,
+				       struct skb_shared_hwtstamps *hwtstamps,
+				       u64 systim)
+{
+	switch (adapter->hw.mac.type) {
+	case igc_i225:
+		memset(hwtstamps, 0, sizeof(*hwtstamps));
+		/* Upper 32 bits contain s, lower 32 bits contain ns. */
+		hwtstamps->hwtstamp = ktime_set(systim >> 32,
+						systim & 0xFFFFFFFF);
+		break;
+	default:
+		break;
+	}
+}
+
+/**
+ * igc_ptp_rx_pktstamp - retrieve Rx per packet timestamp
+ * @q_vector: Pointer to interrupt specific structure
+ * @va: Pointer to address containing Rx buffer
+ * @skb: Buffer containing timestamp and packet
+ *
+ * This function is meant to retrieve the first timestamp from the
+ * first buffer of an incoming frame. The value is stored in little
+ * endian format starting on byte 0. There's a second timestamp
+ * starting on byte 8.
+ **/
+void igc_ptp_rx_pktstamp(struct igc_q_vector *q_vector, void *va,
+			 struct sk_buff *skb)
+{
+	struct igc_adapter *adapter = q_vector->adapter;
+	__le64 *regval = (__le64 *)va;
+	int adjust = 0;
+
+	/* The timestamp is recorded in little endian format.
+	 * DWORD: | 0          | 1           | 2          | 3
+	 * Field: | Timer0 Low | Timer0 High | Timer1 Low | Timer1 High
+	 */
+	igc_ptp_systim_to_hwtstamp(adapter, skb_hwtstamps(skb),
+				   le64_to_cpu(regval[0]));
+
+	/* adjust timestamp for the RX latency based on link speed */
+	if (adapter->hw.mac.type == igc_i225) {
+		switch (adapter->link_speed) {
+		case SPEED_10:
+			adjust = IGC_I225_RX_LATENCY_10;
+			break;
+		case SPEED_100:
+			adjust = IGC_I225_RX_LATENCY_100;
+			break;
+		case SPEED_1000:
+			adjust = IGC_I225_RX_LATENCY_1000;
+			break;
+		case SPEED_2500:
+			adjust = IGC_I225_RX_LATENCY_2500;
+			break;
+		}
+	}
+	skb_hwtstamps(skb)->hwtstamp =
+		ktime_sub_ns(skb_hwtstamps(skb)->hwtstamp, adjust);
+}
+
+/**
+ * igc_ptp_rx_rgtstamp - retrieve Rx timestamp stored in register
+ * @q_vector: Pointer to interrupt specific structure
+ * @skb: Buffer containing timestamp and packet
+ *
+ * This function is meant to retrieve a timestamp from the internal registers
+ * of the adapter and store it in the skb.
+ */
+void igc_ptp_rx_rgtstamp(struct igc_q_vector *q_vector,
+			 struct sk_buff *skb)
+{
+	struct igc_adapter *adapter = q_vector->adapter;
+	struct igc_hw *hw = &adapter->hw;
+	u64 regval;
+
+	/* If this bit is set, then the RX registers contain the time
+	 * stamp. No other packet will be time stamped until we read
+	 * these registers, so read the registers to make them
+	 * available again. Because only one packet can be time
+	 * stamped at a time, we know that the register values must
+	 * belong to this one here and therefore we don't need to
+	 * compare any of the additional attributes stored for it.
+	 *
+	 * If nothing went wrong, then it should have a shared
+	 * tx_flags that we can turn into a skb_shared_hwtstamps.
+	 */
+	if (!(rd32(IGC_TSYNCRXCTL) & IGC_TSYNCRXCTL_VALID))
+		return;
+
+	regval = rd32(IGC_RXSTMPL);
+	regval |= (u64)rd32(IGC_RXSTMPH) << 32;
+
+	igc_ptp_systim_to_hwtstamp(adapter, skb_hwtstamps(skb), regval);
+
+	/* Update the last_rx_timestamp timer in order to enable watchdog check
+	 * for error case of latched timestamp on a dropped packet.
+	 */
+	adapter->last_rx_timestamp = jiffies;
+}
+
+/**
+ * igc_ptp_enable_tstamp_rxqueue - Enable RX timestamp for a queue
+ * @rx_ring: Pointer to RX queue
+ * @timer: Index for timer
+ *
+ * This function enables RX timestamping for a queue, and selects
+ * which 1588 timer will provide the timestamp.
+ */
+static void igc_ptp_enable_tstamp_rxqueue(struct igc_adapter *adapter,
+					  struct igc_ring *rx_ring, u8 timer)
+{
+	struct igc_hw *hw = &adapter->hw;
+	int reg_idx = rx_ring->reg_idx;
+	u32 srrctl = rd32(IGC_SRRCTL(reg_idx));
+
+	srrctl |= IGC_SRRCTL_TIMESTAMP;
+	srrctl |= IGC_SRRCTL_TIMER1SEL(timer);
+	srrctl |= IGC_SRRCTL_TIMER0SEL(timer);
+
+	wr32(IGC_SRRCTL(reg_idx), srrctl);
+}
+
+static void igc_ptp_enable_tstamp_all_rxqueues(struct igc_adapter *adapter,
+					       u8 timer)
+{
+	int i;
+
+	for (i = 0; i < adapter->num_rx_queues; i++) {
+		struct igc_ring *ring = adapter->rx_ring[i];
+
+		igc_ptp_enable_tstamp_rxqueue(adapter, ring, timer);
+	}
+}
+
+/**
+ * igc_ptp_set_timestamp_mode - setup hardware for timestamping
+ * @adapter: networking device structure
+ * @config: hwtstamp configuration
+ *
+ * Incoming time stamping has to be configured via the hardware
+ * filters. Not all combinations are supported, in particular event
+ * type has to be specified. Matching the kind of event packet is
+ * not supported, with the exception of "all V2 events regardless of
+ * level 2 or 4".
+ *
+ */
 static int igc_ptp_set_timestamp_mode(struct igc_adapter *adapter,
 				      struct hwtstamp_config *config)
 {
+	u32 tsync_rx_ctl = IGC_TSYNCRXCTL_ENABLED;
+	struct igc_hw *hw = &adapter->hw;
+	u32 tsync_rx_cfg = 0;
+	bool is_l4 = false;
+	bool is_l2 = false;
+	u32 regval;
+
+	/* reserved for future extensions */
+	if (config->flags)
+		return -EINVAL;
+
+	switch (config->rx_filter) {
+	case HWTSTAMP_FILTER_NONE:
+		tsync_rx_ctl = 0;
+		break;
+	case HWTSTAMP_FILTER_PTP_V1_L4_SYNC:
+		tsync_rx_ctl |= IGC_TSYNCRXCTL_TYPE_L4_V1;
+		tsync_rx_cfg = IGC_TSYNCRXCFG_PTP_V1_SYNC_MESSAGE;
+		is_l4 = true;
+		break;
+	case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ:
+		tsync_rx_ctl |= IGC_TSYNCRXCTL_TYPE_L4_V1;
+		tsync_rx_cfg = IGC_TSYNCRXCFG_PTP_V1_DELAY_REQ_MESSAGE;
+		is_l4 = true;
+		break;
+	case HWTSTAMP_FILTER_PTP_V2_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_L2_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_L4_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_L2_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_L4_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ:
+	case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ:
+	case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ:
+		tsync_rx_ctl |= IGC_TSYNCRXCTL_TYPE_EVENT_V2;
+		config->rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT;
+		is_l2 = true;
+		is_l4 = true;
+		break;
+	case HWTSTAMP_FILTER_PTP_V1_L4_EVENT:
+	case HWTSTAMP_FILTER_NTP_ALL:
+	case HWTSTAMP_FILTER_ALL:
+		tsync_rx_ctl |= IGC_TSYNCRXCTL_TYPE_ALL;
+		config->rx_filter = HWTSTAMP_FILTER_ALL;
+		break;
+		/* fall through */
+	default:
+		config->rx_filter = HWTSTAMP_FILTER_NONE;
+		return -ERANGE;
+	}
+
+	/* Per-packet timestamping only works if all packets are
+	 * timestamped, so enable timestamping in all packets as long
+	 * as one Rx filter was configured.
+	 */
+	if (tsync_rx_ctl) {
+		tsync_rx_ctl = IGC_TSYNCRXCTL_ENABLED;
+		tsync_rx_ctl |= IGC_TSYNCRXCTL_TYPE_ALL;
+		config->rx_filter = HWTSTAMP_FILTER_ALL;
+		is_l2 = true;
+		is_l4 = true;
+
+		if (hw->mac.type == igc_i225) {
+			regval = rd32(IGC_RXPBS);
+			regval |= IGC_RXPBS_CFG_TS_EN;
+			wr32(IGC_RXPBS, regval);
+
+			/* FIXME: For now, only support retrieving RX
+			 * timestamps from timer 0
+			 */
+			igc_ptp_enable_tstamp_all_rxqueues(adapter, 0);
+		}
+	}
+
+	/* enable/disable RX */
+	regval = rd32(IGC_TSYNCRXCTL);
+	regval &= ~(IGC_TSYNCRXCTL_ENABLED | IGC_TSYNCRXCTL_TYPE_MASK);
+	regval |= tsync_rx_ctl;
+	wr32(IGC_TSYNCRXCTL, regval);
+
+	/* define which PTP packets are time stamped */
+	wr32(IGC_TSYNCRXCFG, tsync_rx_cfg);
+
+	/* define ethertype filter for timestamped packets */
+	if (is_l2)
+		wr32(IGC_ETQF(3),
+		     (IGC_ETQF_FILTER_ENABLE | /* enable filter */
+		     IGC_ETQF_1588 | /* enable timestamping */
+		     ETH_P_1588)); /* 1588 eth protocol type */
+	else
+		wr32(IGC_ETQF(3), 0);
+
+	/* L4 Queue Filter[3]: filter by destination port and protocol */
+	if (is_l4) {
+		u32 ftqf = (IPPROTO_UDP /* UDP */
+			    | IGC_FTQF_VF_BP /* VF not compared */
+			    | IGC_FTQF_1588_TIME_STAMP /* Enable Timestamp */
+			    | IGC_FTQF_MASK); /* mask all inputs */
+		ftqf &= ~IGC_FTQF_MASK_PROTO_BP; /* enable protocol check */
+
+		wr32(IGC_IMIR(3), htons(PTP_EV_PORT));
+		wr32(IGC_IMIREXT(3),
+		     (IGC_IMIREXT_SIZE_BP | IGC_IMIREXT_CTRL_BP));
+		wr32(IGC_FTQF(3), ftqf);
+	} else {
+		wr32(IGC_FTQF(3), IGC_FTQF_MASK);
+	}
+	wrfl();
+
+	/* clear TX/RX time stamp registers, just to be sure */
+	regval = rd32(IGC_TXSTMPL);
+	regval = rd32(IGC_TXSTMPH);
+	regval = rd32(IGC_RXSTMPL);
+	regval = rd32(IGC_RXSTMPH);
+
 	return 0;
 }
 
diff --git a/drivers/net/ethernet/intel/igc/igc_regs.h b/drivers/net/ethernet/intel/igc/igc_regs.h
index b5c72430a18b..c82111051898 100644
--- a/drivers/net/ethernet/intel/igc/igc_regs.h
+++ b/drivers/net/ethernet/intel/igc/igc_regs.h
@@ -222,6 +222,9 @@
 #define IGC_IMIREXT(_i)	(0x05AA0 + ((_i) * 4))  /* Immediate INTR Ext*/
 
 #define IGC_FTQF(_n)	(0x059E0 + (4 * (_n)))  /* 5-tuple Queue Fltr */
+
+#define IGC_RXPBS	0x02404  /* Rx Packet Buffer Size - RW */
+
 /* System Time Registers */
 #define IGC_SYSTIML	0x0B600  /* System time register Low - RO */
 #define IGC_SYSTIMH	0x0B604  /* System time register High - RO */
-- 
cgit v1.2.3-59-g8ed1b


From 2c344ae24501558b14341247abef41e0b420c090 Mon Sep 17 00:00:00 2001
From: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Date: Mon, 2 Dec 2019 15:19:51 -0800
Subject: igc: Add support for TX timestamping

This adds support for timestamping packets being transmitted.

Based on the code from i210. The basic differences is that i225 has 4
registers to store the transmit timestamps (i210 has one). Right now,
we only support retrieving from one register, support for using the
other registers will be added later.

Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc.h         |  2 +
 drivers/net/ethernet/intel/igc/igc_defines.h | 14 +++++
 drivers/net/ethernet/intel/igc/igc_main.c    | 50 +++++++++++++++
 drivers/net/ethernet/intel/igc/igc_ptp.c     | 92 ++++++++++++++++++++++++++++
 4 files changed, 158 insertions(+)

diff --git a/drivers/net/ethernet/intel/igc/igc.h b/drivers/net/ethernet/intel/igc/igc.h
index 9b0d1cbf941f..52066bdbbad0 100644
--- a/drivers/net/ethernet/intel/igc/igc.h
+++ b/drivers/net/ethernet/intel/igc/igc.h
@@ -558,6 +558,8 @@ void igc_ptp_rx_pktstamp(struct igc_q_vector *q_vector, void *va,
 			 struct sk_buff *skb);
 int igc_ptp_set_ts_config(struct net_device *netdev, struct ifreq *ifr);
 int igc_ptp_get_ts_config(struct net_device *netdev, struct ifreq *ifr);
+void igc_ptp_tx_hang(struct igc_adapter *adapter);
+
 #define igc_rx_pg_size(_ring) (PAGE_SIZE << igc_rx_pg_order(_ring))
 
 #define IGC_TXD_DCMD	(IGC_ADVTXD_DCMD_EOP | IGC_ADVTXD_DCMD_RS)
diff --git a/drivers/net/ethernet/intel/igc/igc_defines.h b/drivers/net/ethernet/intel/igc/igc_defines.h
index 9dede618362f..586fa14098eb 100644
--- a/drivers/net/ethernet/intel/igc/igc_defines.h
+++ b/drivers/net/ethernet/intel/igc/igc_defines.h
@@ -401,10 +401,24 @@
 #define IGC_TSYNCRXCFG_PTP_V1_SYNC_MESSAGE	0x00
 #define IGC_TSYNCRXCFG_PTP_V1_DELAY_REQ_MESSAGE	0x01
 
+/* Immediate Interrupt Receive */
+#define IGC_IMIR_CLEAR_MASK	0xF001FFFF /* IMIR Reg Clear Mask */
+#define IGC_IMIR_PORT_BYPASS	0x20000 /* IMIR Port Bypass Bit */
+#define IGC_IMIR_PRIORITY_SHIFT	29 /* IMIR Priority Shift */
+#define IGC_IMIREXT_CLEAR_MASK	0x7FFFF /* IMIREXT Reg Clear Mask */
+
 /* Immediate Interrupt Receive Extended */
 #define IGC_IMIREXT_CTRL_BP	0x00080000  /* Bypass check of ctrl bits */
 #define IGC_IMIREXT_SIZE_BP	0x00001000  /* Packet size bypass */
 
+/* Time Sync Transmit Control bit definitions */
+#define IGC_TSYNCTXCTL_VALID			0x00000001  /* Tx timestamp valid */
+#define IGC_TSYNCTXCTL_ENABLED			0x00000010  /* enable Tx timestamping */
+#define IGC_TSYNCTXCTL_MAX_ALLOWED_DLY_MASK	0x0000F000  /* max delay */
+#define IGC_TSYNCTXCTL_SYNC_COMP_ERR		0x20000000  /* sync err */
+#define IGC_TSYNCTXCTL_SYNC_COMP		0x40000000  /* sync complete */
+#define IGC_TSYNCTXCTL_START_SYNC		0x80000000  /* initiate sync */
+
 /* Receive Checksum Control */
 #define IGC_RXCSUM_CRCOFL	0x00000800   /* CRC32 offload enable */
 #define IGC_RXCSUM_PCSD		0x00002000   /* packet checksum disabled */
diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 504dbb74ca75..c359f3d9fb25 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -987,6 +987,11 @@ static inline int igc_maybe_stop_tx(struct igc_ring *tx_ring, const u16 size)
 	return __igc_maybe_stop_tx(tx_ring, size);
 }
 
+#define IGC_SET_FLAG(_input, _flag, _result) \
+	(((_flag) <= (_result)) ?				\
+	 ((u32)((_input) & (_flag)) * ((_result) / (_flag))) :	\
+	 ((u32)((_input) & (_flag)) / ((_flag) / (_result))))
+
 static u32 igc_tx_cmd_type(struct sk_buff *skb, u32 tx_flags)
 {
 	/* set type for advanced descriptor with frame checksum insertion */
@@ -994,6 +999,10 @@ static u32 igc_tx_cmd_type(struct sk_buff *skb, u32 tx_flags)
 		       IGC_ADVTXD_DCMD_DEXT |
 		       IGC_ADVTXD_DCMD_IFCS;
 
+	/* set timestamp bit if present */
+	cmd_type |= IGC_SET_FLAG(tx_flags, IGC_TX_FLAGS_TSTAMP,
+				 (IGC_ADVTXD_MAC_TSTAMP));
+
 	return cmd_type;
 }
 
@@ -1192,6 +1201,26 @@ static netdev_tx_t igc_xmit_frame_ring(struct sk_buff *skb,
 	first->bytecount = skb->len;
 	first->gso_segs = 1;
 
+	if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)) {
+		struct igc_adapter *adapter = netdev_priv(tx_ring->netdev);
+
+		/* FIXME: add support for retrieving timestamps from
+		 * the other timer registers before skipping the
+		 * timestamping request.
+		 */
+		if (adapter->tstamp_config.tx_type == HWTSTAMP_TX_ON &&
+		    !test_and_set_bit_lock(__IGC_PTP_TX_IN_PROGRESS,
+					   &adapter->state)) {
+			skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
+			tx_flags |= IGC_TX_FLAGS_TSTAMP;
+
+			adapter->ptp_tx_skb = skb_get(skb);
+			adapter->ptp_tx_start = jiffies;
+		} else {
+			adapter->tx_hwtstamp_skipped++;
+		}
+	}
+
 	/* record initial flags and protocol */
 	first->tx_flags = tx_flags;
 	first->protocol = protocol;
@@ -3648,6 +3677,22 @@ int igc_del_mac_steering_filter(struct igc_adapter *adapter,
 					IGC_MAC_STATE_QUEUE_STEERING | flags);
 }
 
+static void igc_tsync_interrupt(struct igc_adapter *adapter)
+{
+	struct igc_hw *hw = &adapter->hw;
+	u32 tsicr = rd32(IGC_TSICR);
+	u32 ack = 0;
+
+	if (tsicr & IGC_TSICR_TXTS) {
+		/* retrieve hardware timestamp */
+		schedule_work(&adapter->ptp_tx_work);
+		ack |= IGC_TSICR_TXTS;
+	}
+
+	/* acknowledge the interrupts */
+	wr32(IGC_TSICR, ack);
+}
+
 /**
  * igc_msix_other - msix other interrupt handler
  * @irq: interrupt number
@@ -3675,6 +3720,9 @@ static irqreturn_t igc_msix_other(int irq, void *data)
 			mod_timer(&adapter->watchdog_timer, jiffies + 1);
 	}
 
+	if (icr & IGC_ICR_TS)
+		igc_tsync_interrupt(adapter);
+
 	wr32(IGC_EIMS, adapter->eims_other);
 
 	return IRQ_HANDLED;
@@ -4004,6 +4052,8 @@ no_wait:
 		wr32(IGC_ICS, IGC_ICS_RXDMT0);
 	}
 
+	igc_ptp_tx_hang(adapter);
+
 	/* Reset the timer */
 	if (!test_bit(__IGC_DOWN, &adapter->state)) {
 		if (adapter->flags & IGC_FLAG_NEED_LINK_UPDATE)
diff --git a/drivers/net/ethernet/intel/igc/igc_ptp.c b/drivers/net/ethernet/intel/igc/igc_ptp.c
index 093ffc9312df..79ffb833aa80 100644
--- a/drivers/net/ethernet/intel/igc/igc_ptp.c
+++ b/drivers/net/ethernet/intel/igc/igc_ptp.c
@@ -284,6 +284,12 @@ static void igc_ptp_enable_tstamp_all_rxqueues(struct igc_adapter *adapter,
  * @adapter: networking device structure
  * @config: hwtstamp configuration
  *
+ * Outgoing time stamping can be enabled and disabled. Play nice and
+ * disable it when requested, although it shouldn't case any overhead
+ * when no packet needs it. At most one packet in the queue may be
+ * marked for time stamping, otherwise it would be impossible to tell
+ * for sure to which packet the hardware time stamp belongs.
+ *
  * Incoming time stamping has to be configured via the hardware
  * filters. Not all combinations are supported, in particular event
  * type has to be specified. Matching the kind of event packet is
@@ -294,6 +300,7 @@ static void igc_ptp_enable_tstamp_all_rxqueues(struct igc_adapter *adapter,
 static int igc_ptp_set_timestamp_mode(struct igc_adapter *adapter,
 				      struct hwtstamp_config *config)
 {
+	u32 tsync_tx_ctl = IGC_TSYNCTXCTL_ENABLED;
 	u32 tsync_rx_ctl = IGC_TSYNCRXCTL_ENABLED;
 	struct igc_hw *hw = &adapter->hw;
 	u32 tsync_rx_cfg = 0;
@@ -305,6 +312,15 @@ static int igc_ptp_set_timestamp_mode(struct igc_adapter *adapter,
 	if (config->flags)
 		return -EINVAL;
 
+	switch (config->tx_type) {
+	case HWTSTAMP_TX_OFF:
+		tsync_tx_ctl = 0;
+	case HWTSTAMP_TX_ON:
+		break;
+	default:
+		return -ERANGE;
+	}
+
 	switch (config->rx_filter) {
 	case HWTSTAMP_FILTER_NONE:
 		tsync_rx_ctl = 0;
@@ -368,6 +384,15 @@ static int igc_ptp_set_timestamp_mode(struct igc_adapter *adapter,
 		}
 	}
 
+	if (tsync_tx_ctl)
+		tsync_tx_ctl = IGC_TSYNCTXCTL_ENABLED;
+
+	/* enable/disable TX */
+	regval = rd32(IGC_TSYNCTXCTL);
+	regval &= ~IGC_TSYNCTXCTL_ENABLED;
+	regval |= tsync_tx_ctl;
+	wr32(IGC_TSYNCTXCTL, regval);
+
 	/* enable/disable RX */
 	regval = rd32(IGC_TSYNCRXCTL);
 	regval &= ~(IGC_TSYNCRXCTL_ENABLED | IGC_TSYNCRXCTL_TYPE_MASK);
@@ -442,8 +467,75 @@ void igc_ptp_tx_hang(struct igc_adapter *adapter)
 	}
 }
 
+/**
+ * igc_ptp_tx_hwtstamp - utility function which checks for TX time stamp
+ * @adapter: Board private structure
+ *
+ * If we were asked to do hardware stamping and such a time stamp is
+ * available, then it must have been for this skb here because we only
+ * allow only one such packet into the queue.
+ */
+static void igc_ptp_tx_hwtstamp(struct igc_adapter *adapter)
+{
+	struct sk_buff *skb = adapter->ptp_tx_skb;
+	struct skb_shared_hwtstamps shhwtstamps;
+	struct igc_hw *hw = &adapter->hw;
+	u64 regval;
+
+	regval = rd32(IGC_TXSTMPL);
+	regval |= (u64)rd32(IGC_TXSTMPH) << 32;
+	igc_ptp_systim_to_hwtstamp(adapter, &shhwtstamps, regval);
+
+	/* Clear the lock early before calling skb_tstamp_tx so that
+	 * applications are not woken up before the lock bit is clear. We use
+	 * a copy of the skb pointer to ensure other threads can't change it
+	 * while we're notifying the stack.
+	 */
+	adapter->ptp_tx_skb = NULL;
+	clear_bit_unlock(__IGC_PTP_TX_IN_PROGRESS, &adapter->state);
+
+	/* Notify the stack and free the skb after we've unlocked */
+	skb_tstamp_tx(skb, &shhwtstamps);
+	dev_kfree_skb_any(skb);
+}
+
+/**
+ * igc_ptp_tx_work
+ * @work: pointer to work struct
+ *
+ * This work function polls the TSYNCTXCTL valid bit to determine when a
+ * timestamp has been taken for the current stored skb.
+ */
 void igc_ptp_tx_work(struct work_struct *work)
 {
+	struct igc_adapter *adapter = container_of(work, struct igc_adapter,
+						   ptp_tx_work);
+	struct igc_hw *hw = &adapter->hw;
+	u32 tsynctxctl;
+
+	if (!adapter->ptp_tx_skb)
+		return;
+
+	if (time_is_before_jiffies(adapter->ptp_tx_start +
+				   IGC_PTP_TX_TIMEOUT)) {
+		dev_kfree_skb_any(adapter->ptp_tx_skb);
+		adapter->ptp_tx_skb = NULL;
+		clear_bit_unlock(__IGC_PTP_TX_IN_PROGRESS, &adapter->state);
+		adapter->tx_hwtstamp_timeouts++;
+		/* Clear the tx valid bit in TSYNCTXCTL register to enable
+		 * interrupt
+		 */
+		rd32(IGC_TXSTMPH);
+		dev_warn(&adapter->pdev->dev, "clearing Tx timestamp hang\n");
+		return;
+	}
+
+	tsynctxctl = rd32(IGC_TSYNCTXCTL);
+	if (tsynctxctl & IGC_TSYNCTXCTL_VALID)
+		igc_ptp_tx_hwtstamp(adapter);
+	else
+		/* reschedule to check later */
+		schedule_work(&adapter->ptp_tx_work);
 }
 
 /**
-- 
cgit v1.2.3-59-g8ed1b


From 60dbede0c4f3d7b1a1f56d1f91bbca3c5a73f7f5 Mon Sep 17 00:00:00 2001
From: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Date: Mon, 2 Dec 2019 15:19:52 -0800
Subject: igc: Add support for ethtool GET_TS_INFO command

This command allows igc to report what types of timestamping are
supported. ptp4l uses this to detect if the hardware supports
timestamping.

Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_ethtool.c | 34 ++++++++++++++++++++++++++++
 1 file changed, 34 insertions(+)

diff --git a/drivers/net/ethernet/intel/igc/igc_ethtool.c b/drivers/net/ethernet/intel/igc/igc_ethtool.c
index 455c1cdceb6e..ee07011e13e9 100644
--- a/drivers/net/ethernet/intel/igc/igc_ethtool.c
+++ b/drivers/net/ethernet/intel/igc/igc_ethtool.c
@@ -1600,6 +1600,39 @@ static int igc_set_channels(struct net_device *netdev,
 	return 0;
 }
 
+static int igc_get_ts_info(struct net_device *dev,
+			   struct ethtool_ts_info *info)
+{
+	struct igc_adapter *adapter = netdev_priv(dev);
+
+	if (adapter->ptp_clock)
+		info->phc_index = ptp_clock_index(adapter->ptp_clock);
+	else
+		info->phc_index = -1;
+
+	switch (adapter->hw.mac.type) {
+	case igc_i225:
+		info->so_timestamping =
+			SOF_TIMESTAMPING_TX_SOFTWARE |
+			SOF_TIMESTAMPING_RX_SOFTWARE |
+			SOF_TIMESTAMPING_SOFTWARE |
+			SOF_TIMESTAMPING_TX_HARDWARE |
+			SOF_TIMESTAMPING_RX_HARDWARE |
+			SOF_TIMESTAMPING_RAW_HARDWARE;
+
+		info->tx_types =
+			BIT(HWTSTAMP_TX_OFF) |
+			BIT(HWTSTAMP_TX_ON);
+
+		info->rx_filters = BIT(HWTSTAMP_FILTER_NONE);
+		info->rx_filters |= BIT(HWTSTAMP_FILTER_ALL);
+
+		return 0;
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
 static u32 igc_get_priv_flags(struct net_device *netdev)
 {
 	struct igc_adapter *adapter = netdev_priv(netdev);
@@ -1847,6 +1880,7 @@ static const struct ethtool_ops igc_ethtool_ops = {
 	.get_rxfh_indir_size	= igc_get_rxfh_indir_size,
 	.get_rxfh		= igc_get_rxfh,
 	.set_rxfh		= igc_set_rxfh,
+	.get_ts_info		= igc_get_ts_info,
 	.get_channels		= igc_get_channels,
 	.set_channels		= igc_set_channels,
 	.get_priv_flags		= igc_get_priv_flags,
-- 
cgit v1.2.3-59-g8ed1b


From a299df3524eabc1def8e93bf005b07ea396ff2bd Mon Sep 17 00:00:00 2001
From: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Date: Mon, 2 Dec 2019 15:19:53 -0800
Subject: igc: Use Start of Packet signal from PHY for timestamping

For better accuracy, i225 is able to do timestamping using the Start of
Packet signal from the PHY.

Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_defines.h | 2 ++
 drivers/net/ethernet/intel/igc/igc_ptp.c     | 5 ++++-
 2 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_defines.h b/drivers/net/ethernet/intel/igc/igc_defines.h
index 586fa14098eb..2121fc34e300 100644
--- a/drivers/net/ethernet/intel/igc/igc_defines.h
+++ b/drivers/net/ethernet/intel/igc/igc_defines.h
@@ -395,6 +395,7 @@
 #define IGC_TSYNCRXCTL_TYPE_EVENT_V2	0x0A
 #define IGC_TSYNCRXCTL_ENABLED		0x00000010  /* enable Rx timestamping */
 #define IGC_TSYNCRXCTL_SYSCFI		0x00000020  /* Sys clock frequency */
+#define IGC_TSYNCRXCTL_RXSYNSIG		0x00000400  /* Sample RX tstamp in PHY sop */
 
 /* Time Sync Receive Configuration */
 #define IGC_TSYNCRXCFG_PTP_V1_CTRLT_MASK	0x000000FF
@@ -418,6 +419,7 @@
 #define IGC_TSYNCTXCTL_SYNC_COMP_ERR		0x20000000  /* sync err */
 #define IGC_TSYNCTXCTL_SYNC_COMP		0x40000000  /* sync complete */
 #define IGC_TSYNCTXCTL_START_SYNC		0x80000000  /* initiate sync */
+#define IGC_TSYNCTXCTL_TXSYNSIG			0x00000020  /* Sample TX tstamp in PHY sop */
 
 /* Receive Checksum Control */
 #define IGC_RXCSUM_CRCOFL	0x00000800   /* CRC32 offload enable */
diff --git a/drivers/net/ethernet/intel/igc/igc_ptp.c b/drivers/net/ethernet/intel/igc/igc_ptp.c
index 79ffb833aa80..693506587198 100644
--- a/drivers/net/ethernet/intel/igc/igc_ptp.c
+++ b/drivers/net/ethernet/intel/igc/igc_ptp.c
@@ -368,6 +368,7 @@ static int igc_ptp_set_timestamp_mode(struct igc_adapter *adapter,
 	if (tsync_rx_ctl) {
 		tsync_rx_ctl = IGC_TSYNCRXCTL_ENABLED;
 		tsync_rx_ctl |= IGC_TSYNCRXCTL_TYPE_ALL;
+		tsync_rx_ctl |= IGC_TSYNCRXCTL_RXSYNSIG;
 		config->rx_filter = HWTSTAMP_FILTER_ALL;
 		is_l2 = true;
 		is_l4 = true;
@@ -384,8 +385,10 @@ static int igc_ptp_set_timestamp_mode(struct igc_adapter *adapter,
 		}
 	}
 
-	if (tsync_tx_ctl)
+	if (tsync_tx_ctl) {
 		tsync_tx_ctl = IGC_TSYNCTXCTL_ENABLED;
+		tsync_tx_ctl |= IGC_TSYNCTXCTL_TXSYNSIG;
+	}
 
 	/* enable/disable TX */
 	regval = rd32(IGC_TSYNCTXCTL);
-- 
cgit v1.2.3-59-g8ed1b


From 3095383a8ab4cff7d5a891a5293d064b0b4e4874 Mon Sep 17 00:00:00 2001
From: Andrew Lunn <andrew@lunn.ch>
Date: Mon, 6 Jan 2020 17:13:48 +0100
Subject: net: dsa: mv88e6xxx: Unique IRQ name

Dynamically generate a unique switch interrupt name, based on the
device name.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/mv88e6xxx/chip.c | 5 ++++-
 drivers/net/dsa/mv88e6xxx/chip.h | 1 +
 2 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c
index 5eeeb6566196..e21e460c11f7 100644
--- a/drivers/net/dsa/mv88e6xxx/chip.c
+++ b/drivers/net/dsa/mv88e6xxx/chip.c
@@ -340,11 +340,14 @@ static int mv88e6xxx_g1_irq_setup(struct mv88e6xxx_chip *chip)
 	 */
 	irq_set_lockdep_class(chip->irq, &lock_key, &request_key);
 
+	snprintf(chip->irq_name, sizeof(chip->irq_name),
+		 "mv88e6xxx-%s", dev_name(chip->dev));
+
 	mv88e6xxx_reg_unlock(chip);
 	err = request_threaded_irq(chip->irq, NULL,
 				   mv88e6xxx_g1_irq_thread_fn,
 				   IRQF_ONESHOT | IRQF_SHARED,
-				   dev_name(chip->dev), chip);
+				   chip->irq_name, chip);
 	mv88e6xxx_reg_lock(chip);
 	if (err)
 		mv88e6xxx_g1_irq_free_common(chip);
diff --git a/drivers/net/dsa/mv88e6xxx/chip.h b/drivers/net/dsa/mv88e6xxx/chip.h
index 8a8e38bfb161..03aec58513ab 100644
--- a/drivers/net/dsa/mv88e6xxx/chip.h
+++ b/drivers/net/dsa/mv88e6xxx/chip.h
@@ -292,6 +292,7 @@ struct mv88e6xxx_chip {
 	struct mv88e6xxx_irq g1_irq;
 	struct mv88e6xxx_irq g2_irq;
 	int irq;
+	char irq_name[32];
 	int device_irq;
 	int watchdog_irq;
 
-- 
cgit v1.2.3-59-g8ed1b


From e6f2f6b824a3f9af4a719545f691c303d510b26e Mon Sep 17 00:00:00 2001
From: Andrew Lunn <andrew@lunn.ch>
Date: Mon, 6 Jan 2020 17:13:49 +0100
Subject: net: dsa: mv88e6xxx: Unique SERDES interrupt names

Dynamically generate a unique SERDES interrupt name, based on the
device name and the port the SERDES is for. For example:

 95:          3  mv88e6xxx-g2   9 Edge      mv88e6xxx-0.2:00-serdes-9
 96:          0  mv88e6xxx-g2  10 Edge      mv88e6xxx-0.2:00-serdes-10

The 0.2:00 indicates the switch and -9 indicates port 9.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/mv88e6xxx/chip.c | 6 +++++-
 drivers/net/dsa/mv88e6xxx/chip.h | 1 +
 2 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c
index e21e460c11f7..99816ca9e5e4 100644
--- a/drivers/net/dsa/mv88e6xxx/chip.c
+++ b/drivers/net/dsa/mv88e6xxx/chip.c
@@ -2306,10 +2306,14 @@ static int mv88e6xxx_serdes_irq_request(struct mv88e6xxx_chip *chip, int port,
 	if (!irq)
 		return 0;
 
+	snprintf(dev_id->serdes_irq_name, sizeof(dev_id->serdes_irq_name),
+		 "mv88e6xxx-%s-serdes-%d", dev_name(chip->dev), port);
+
 	/* Requesting the IRQ will trigger IRQ callbacks, so release the lock */
 	mv88e6xxx_reg_unlock(chip);
 	err = request_threaded_irq(irq, NULL, mv88e6xxx_serdes_irq_thread_fn,
-				   IRQF_ONESHOT, "mv88e6xxx-serdes", dev_id);
+				   IRQF_ONESHOT, dev_id->serdes_irq_name,
+				   dev_id);
 	mv88e6xxx_reg_lock(chip);
 	if (err)
 		return err;
diff --git a/drivers/net/dsa/mv88e6xxx/chip.h b/drivers/net/dsa/mv88e6xxx/chip.h
index 03aec58513ab..b7613ef600d2 100644
--- a/drivers/net/dsa/mv88e6xxx/chip.h
+++ b/drivers/net/dsa/mv88e6xxx/chip.h
@@ -236,6 +236,7 @@ struct mv88e6xxx_port {
 	bool mirror_ingress;
 	bool mirror_egress;
 	unsigned int serdes_irq;
+	char serdes_irq_name[32];
 };
 
 struct mv88e6xxx_chip {
-- 
cgit v1.2.3-59-g8ed1b


From 8b4db28914b5b49e296884739d9710aedb2c368e Mon Sep 17 00:00:00 2001
From: Andrew Lunn <andrew@lunn.ch>
Date: Mon, 6 Jan 2020 17:13:50 +0100
Subject: net: dsa: mv88e6xxx: Unique watchdog IRQ name

Dynamically generate a unique watchdog interrupt name, based on the
device name.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/mv88e6xxx/chip.h    | 1 +
 drivers/net/dsa/mv88e6xxx/global2.c | 5 ++++-
 2 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/net/dsa/mv88e6xxx/chip.h b/drivers/net/dsa/mv88e6xxx/chip.h
index b7613ef600d2..3558c677e1d3 100644
--- a/drivers/net/dsa/mv88e6xxx/chip.h
+++ b/drivers/net/dsa/mv88e6xxx/chip.h
@@ -296,6 +296,7 @@ struct mv88e6xxx_chip {
 	char irq_name[32];
 	int device_irq;
 	int watchdog_irq;
+	char watchdog_irq_name[32];
 
 	int atu_prob_irq;
 	int vtu_prob_irq;
diff --git a/drivers/net/dsa/mv88e6xxx/global2.c b/drivers/net/dsa/mv88e6xxx/global2.c
index 87bfe7c8c9cd..617174e94faa 100644
--- a/drivers/net/dsa/mv88e6xxx/global2.c
+++ b/drivers/net/dsa/mv88e6xxx/global2.c
@@ -948,10 +948,13 @@ static int mv88e6xxx_g2_watchdog_setup(struct mv88e6xxx_chip *chip)
 	if (chip->watchdog_irq < 0)
 		return chip->watchdog_irq;
 
+	snprintf(chip->watchdog_irq_name, sizeof(chip->watchdog_irq_name),
+		 "mv88e6xxx-%s-watchdog", dev_name(chip->dev));
+
 	err = request_threaded_irq(chip->watchdog_irq, NULL,
 				   mv88e6xxx_g2_watchdog_thread_fn,
 				   IRQF_ONESHOT | IRQF_TRIGGER_FALLING,
-				   "mv88e6xxx-watchdog", chip);
+				   chip->watchdog_irq_name, chip);
 	if (err)
 		return err;
 
-- 
cgit v1.2.3-59-g8ed1b


From 06acd1148bb35a0c4144fec0a97685974dc89e61 Mon Sep 17 00:00:00 2001
From: Andrew Lunn <andrew@lunn.ch>
Date: Mon, 6 Jan 2020 17:13:51 +0100
Subject: net: dsa: mv88e6xxx: Unique g2 IRQ name

Dynamically generate a unique g2 interrupt name, based on the
device name.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/mv88e6xxx/chip.h    | 1 +
 drivers/net/dsa/mv88e6xxx/global2.c | 5 ++++-
 2 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/net/dsa/mv88e6xxx/chip.h b/drivers/net/dsa/mv88e6xxx/chip.h
index 3558c677e1d3..47fd909fccc2 100644
--- a/drivers/net/dsa/mv88e6xxx/chip.h
+++ b/drivers/net/dsa/mv88e6xxx/chip.h
@@ -295,6 +295,7 @@ struct mv88e6xxx_chip {
 	int irq;
 	char irq_name[32];
 	int device_irq;
+	char device_irq_name[32];
 	int watchdog_irq;
 	char watchdog_irq_name[32];
 
diff --git a/drivers/net/dsa/mv88e6xxx/global2.c b/drivers/net/dsa/mv88e6xxx/global2.c
index 617174e94faa..01503014b1ee 100644
--- a/drivers/net/dsa/mv88e6xxx/global2.c
+++ b/drivers/net/dsa/mv88e6xxx/global2.c
@@ -1117,9 +1117,12 @@ int mv88e6xxx_g2_irq_setup(struct mv88e6xxx_chip *chip)
 		goto out;
 	}
 
+	snprintf(chip->device_irq_name, sizeof(chip->device_irq_name),
+		 "mv88e6xxx-%s-g2", dev_name(chip->dev));
+
 	err = request_threaded_irq(chip->device_irq, NULL,
 				   mv88e6xxx_g2_irq_thread_fn,
-				   IRQF_ONESHOT, "mv88e6xxx-g2", chip);
+				   IRQF_ONESHOT, chip->device_irq_name, chip);
 	if (err)
 		goto out;
 
-- 
cgit v1.2.3-59-g8ed1b


From 8ddf0b5693f64a0f71ee4b20332ab4cdc390fc76 Mon Sep 17 00:00:00 2001
From: Andrew Lunn <andrew@lunn.ch>
Date: Mon, 6 Jan 2020 17:13:52 +0100
Subject: net: dsa: mv88e6xxx: Unique ATU and VTU IRQ names

Dynamically generate a unique interrupt name for the VTU and ATU,
based on the device name.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/mv88e6xxx/chip.h        | 2 ++
 drivers/net/dsa/mv88e6xxx/global1_atu.c | 5 ++++-
 drivers/net/dsa/mv88e6xxx/global1_vtu.c | 5 ++++-
 3 files changed, 10 insertions(+), 2 deletions(-)

diff --git a/drivers/net/dsa/mv88e6xxx/chip.h b/drivers/net/dsa/mv88e6xxx/chip.h
index 47fd909fccc2..f332cb4b2fbf 100644
--- a/drivers/net/dsa/mv88e6xxx/chip.h
+++ b/drivers/net/dsa/mv88e6xxx/chip.h
@@ -300,7 +300,9 @@ struct mv88e6xxx_chip {
 	char watchdog_irq_name[32];
 
 	int atu_prob_irq;
+	char atu_prob_irq_name[32];
 	int vtu_prob_irq;
+	char vtu_prob_irq_name[32];
 	struct kthread_worker *kworker;
 	struct kthread_delayed_work irq_poll_work;
 
diff --git a/drivers/net/dsa/mv88e6xxx/global1_atu.c b/drivers/net/dsa/mv88e6xxx/global1_atu.c
index bdcd25560dd2..bac9a8a68e50 100644
--- a/drivers/net/dsa/mv88e6xxx/global1_atu.c
+++ b/drivers/net/dsa/mv88e6xxx/global1_atu.c
@@ -425,9 +425,12 @@ int mv88e6xxx_g1_atu_prob_irq_setup(struct mv88e6xxx_chip *chip)
 	if (chip->atu_prob_irq < 0)
 		return chip->atu_prob_irq;
 
+	snprintf(chip->atu_prob_irq_name, sizeof(chip->atu_prob_irq_name),
+		 "mv88e6xxx-%s-g1-atu-prob", dev_name(chip->dev));
+
 	err = request_threaded_irq(chip->atu_prob_irq, NULL,
 				   mv88e6xxx_g1_atu_prob_irq_thread_fn,
-				   IRQF_ONESHOT, "mv88e6xxx-g1-atu-prob",
+				   IRQF_ONESHOT, chip->atu_prob_irq_name,
 				   chip);
 	if (err)
 		irq_dispose_mapping(chip->atu_prob_irq);
diff --git a/drivers/net/dsa/mv88e6xxx/global1_vtu.c b/drivers/net/dsa/mv88e6xxx/global1_vtu.c
index 33056a609e96..48390b7b18ad 100644
--- a/drivers/net/dsa/mv88e6xxx/global1_vtu.c
+++ b/drivers/net/dsa/mv88e6xxx/global1_vtu.c
@@ -631,9 +631,12 @@ int mv88e6xxx_g1_vtu_prob_irq_setup(struct mv88e6xxx_chip *chip)
 	if (chip->vtu_prob_irq < 0)
 		return chip->vtu_prob_irq;
 
+	snprintf(chip->vtu_prob_irq_name, sizeof(chip->vtu_prob_irq_name),
+		 "mv88e6xxx-%s-g1-vtu-prob", dev_name(chip->dev));
+
 	err = request_threaded_irq(chip->vtu_prob_irq, NULL,
 				   mv88e6xxx_g1_vtu_prob_irq_thread_fn,
-				   IRQF_ONESHOT, "mv88e6xxx-g1-vtu-prob",
+				   IRQF_ONESHOT, chip->vtu_prob_irq_name,
 				   chip);
 	if (err)
 		irq_dispose_mapping(chip->vtu_prob_irq);
-- 
cgit v1.2.3-59-g8ed1b


From 8007880a2ca97c34e7ccd1fcf12daf854b792544 Mon Sep 17 00:00:00 2001
From: Zhu Yanjun <zyjzyj2000@gmail.com>
Date: Sat, 14 Dec 2019 10:51:17 +0200
Subject: net/mlx5: limit the function in local scope

The function mlx5_buf_alloc_node is only used by the function in the
local scope. So it is appropriate to limit this function in the local
scope.

Signed-off-by: Zhu Yanjun <zyjzyj2000@gmail.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/alloc.c | 4 ++--
 include/linux/mlx5/driver.h                     | 2 --
 2 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/alloc.c b/drivers/net/ethernet/mellanox/mlx5/core/alloc.c
index 549f962cd86e..42198e64a7f4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/alloc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/alloc.c
@@ -71,8 +71,8 @@ static void *mlx5_dma_zalloc_coherent_node(struct mlx5_core_dev *dev,
 	return cpu_handle;
 }
 
-int mlx5_buf_alloc_node(struct mlx5_core_dev *dev, int size,
-			struct mlx5_frag_buf *buf, int node)
+static int mlx5_buf_alloc_node(struct mlx5_core_dev *dev, int size,
+			       struct mlx5_frag_buf *buf, int node)
 {
 	dma_addr_t t;
 
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index 27200dea0297..59cff380f41a 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -928,8 +928,6 @@ void mlx5_start_health_poll(struct mlx5_core_dev *dev);
 void mlx5_stop_health_poll(struct mlx5_core_dev *dev, bool disable_health);
 void mlx5_drain_health_wq(struct mlx5_core_dev *dev);
 void mlx5_trigger_health_work(struct mlx5_core_dev *dev);
-int mlx5_buf_alloc_node(struct mlx5_core_dev *dev, int size,
-			struct mlx5_frag_buf *buf, int node);
 int mlx5_buf_alloc(struct mlx5_core_dev *dev,
 		   int size, struct mlx5_frag_buf *buf);
 void mlx5_buf_free(struct mlx5_core_dev *dev, struct mlx5_frag_buf *buf);
-- 
cgit v1.2.3-59-g8ed1b


From 42ae1a5c76691928ed217c7e40269db27f5225e9 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Sat, 4 Jan 2020 22:51:44 +0100
Subject: mlx5: work around high stack usage with gcc

In some configurations, gcc tries too hard to optimize this code:

drivers/net/ethernet/mellanox/mlx5/core/en_stats.c: In function 'mlx5e_grp_sw_update_stats':
drivers/net/ethernet/mellanox/mlx5/core/en_stats.c:302:1: error: the frame size of 1336 bytes is larger than 1024 bytes [-Werror=frame-larger-than=]

As was stated in the bug report, the reason is that gcc runs into a corner
case in the register allocator that is rather hard to fix in a good way.

As there is an easy way to work around it, just add a comment and the
barrier that stops gcc from trying to overoptimize the function.

Link: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92657
Cc: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_stats.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
index 9f09253f9f46..a05158472ed1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
@@ -297,6 +297,9 @@ static void mlx5e_grp_sw_update_stats(struct mlx5e_priv *priv)
 			s->tx_tls_drop_bypass_req   += sq_stats->tls_drop_bypass_req;
 #endif
 			s->tx_cqes		+= sq_stats->cqes;
+
+			/* https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92657 */
+			barrier();
 		}
 	}
 }
-- 
cgit v1.2.3-59-g8ed1b


From 15fc92ec3ad4cfb34218e6b6c38c1355938fe49e Mon Sep 17 00:00:00 2001
From: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Date: Tue, 10 Dec 2019 22:49:42 +0800
Subject: net/mlx5e: Support accept action on nic table

In one case, we may forward packets from one vport
to others, but only one packets flow will be accepted,
which destination ip was assign to VF.

+-----+     +-----+            +-----+
| VFn |     | VF1 |            | VF0 | accept
+--+--+     +--+--+  hairpin   +--^--+
   |           | <--------------- |
   |           |                  |
+--+-----------v-+             +--+-------------+
|   eswitch PF1  |             |   eswitch PF0  |
+----------------+             +----------------+

tc filter add dev $PF0 protocol all parent ffff: prio 1 handle 1 \
	flower skip_sw action mirred egress redirect dev $VF0_REP
tc filter add dev $VF0 protocol ip  parent ffff: prio 1 handle 1 \
	flower skip_sw dst_ip $VF0_IP action pass
tc filter add dev $VF0 protocol all parent ffff: prio 2 handle 2 \
	flower skip_sw action mirred egress redirect dev $VF1

Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 9b32a9c0f497..e8f2d0e4913d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -2842,6 +2842,10 @@ static int parse_tc_nic_actions(struct mlx5e_priv *priv,
 
 	flow_action_for_each(i, act, flow_action) {
 		switch (act->id) {
+		case FLOW_ACTION_ACCEPT:
+			action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
+				  MLX5_FLOW_CONTEXT_ACTION_COUNT;
+			break;
 		case FLOW_ACTION_DROP:
 			action |= MLX5_FLOW_CONTEXT_ACTION_DROP;
 			if (MLX5_CAP_FLOWTABLE(priv->mdev,
-- 
cgit v1.2.3-59-g8ed1b


From 57c7fce14b1ad512a42abe33cb721a2ea3520d4b Mon Sep 17 00:00:00 2001
From: Fan Li <fanl@mellanox.com>
Date: Mon, 16 Dec 2019 14:46:15 +0200
Subject: net/mlx5: Increase the max number of channels to 128

Currently the max number of channels is limited to 64, which is half of
the indirection table size to allow some flexibility. But on servers
with more than 64 cores, users may want to utilize more queues.

This patch increases the advertised max number of channels to 128 by
changing the ratio between channels and indirection table slots to 1:1.
At the same time, the driver still enable no more than 64 channels at
loading. Users can change it by ethtool afterwards.

Signed-off-by: Fan Li <fanl@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h          |  6 +++---
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c     | 12 +++++++-----
 drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c |  4 ++--
 3 files changed, 12 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 9c8427698238..fc80b59db9a8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -135,7 +135,7 @@ struct page_pool;
 #define MLX5E_LOG_INDIR_RQT_SIZE       0x7
 #define MLX5E_INDIR_RQT_SIZE           BIT(MLX5E_LOG_INDIR_RQT_SIZE)
 #define MLX5E_MIN_NUM_CHANNELS         0x1
-#define MLX5E_MAX_NUM_CHANNELS         (MLX5E_INDIR_RQT_SIZE >> 1)
+#define MLX5E_MAX_NUM_CHANNELS         MLX5E_INDIR_RQT_SIZE
 #define MLX5E_MAX_NUM_SQS              (MLX5E_MAX_NUM_CHANNELS * MLX5E_MAX_NUM_TC)
 #define MLX5E_TX_CQ_POLL_BUDGET        128
 #define MLX5E_TX_XSK_POLL_BUDGET       64
@@ -1175,11 +1175,11 @@ int mlx5e_attach_netdev(struct mlx5e_priv *priv);
 void mlx5e_detach_netdev(struct mlx5e_priv *priv);
 void mlx5e_destroy_netdev(struct mlx5e_priv *priv);
 void mlx5e_set_netdev_mtu_boundaries(struct mlx5e_priv *priv);
-void mlx5e_build_nic_params(struct mlx5_core_dev *mdev,
+void mlx5e_build_nic_params(struct mlx5e_priv *priv,
 			    struct mlx5e_xsk *xsk,
 			    struct mlx5e_rss_params *rss_params,
 			    struct mlx5e_params *params,
-			    u16 max_channels, u16 mtu);
+			    u16 mtu);
 void mlx5e_build_rq_params(struct mlx5_core_dev *mdev,
 			   struct mlx5e_params *params);
 void mlx5e_build_rss_params(struct mlx5e_rss_params *rss_params,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 319b39f25592..78737fd42616 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -4739,17 +4739,19 @@ void mlx5e_build_rss_params(struct mlx5e_rss_params *rss_params,
 			tirc_default_config[tt].rx_hash_fields;
 }
 
-void mlx5e_build_nic_params(struct mlx5_core_dev *mdev,
+void mlx5e_build_nic_params(struct mlx5e_priv *priv,
 			    struct mlx5e_xsk *xsk,
 			    struct mlx5e_rss_params *rss_params,
 			    struct mlx5e_params *params,
-			    u16 max_channels, u16 mtu)
+			    u16 mtu)
 {
+	struct mlx5_core_dev *mdev = priv->mdev;
 	u8 rx_cq_period_mode;
 
 	params->sw_mtu = mtu;
 	params->hard_mtu = MLX5E_ETH_HARD_MTU;
-	params->num_channels = max_channels;
+	params->num_channels = min_t(unsigned int, MLX5E_MAX_NUM_CHANNELS / 2,
+				     priv->max_nch);
 	params->num_tc       = 1;
 
 	/* SQ */
@@ -4986,8 +4988,8 @@ static int mlx5e_nic_init(struct mlx5_core_dev *mdev,
 	if (err)
 		return err;
 
-	mlx5e_build_nic_params(mdev, &priv->xsk, rss, &priv->channels.params,
-			       priv->max_nch, netdev->mtu);
+	mlx5e_build_nic_params(priv, &priv->xsk, rss, &priv->channels.params,
+			       netdev->mtu);
 
 	mlx5e_timestamp_init(priv);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c
index 3ed8ab2d703d..7c87f523e370 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c
@@ -87,8 +87,8 @@ int mlx5i_init(struct mlx5_core_dev *mdev,
 	mlx5e_set_netdev_mtu_boundaries(priv);
 	netdev->mtu = netdev->max_mtu;
 
-	mlx5e_build_nic_params(mdev, NULL, &priv->rss_params, &priv->channels.params,
-			       priv->max_nch, netdev->mtu);
+	mlx5e_build_nic_params(priv, NULL, &priv->rss_params, &priv->channels.params,
+			       netdev->mtu);
 	mlx5i_build_nic_params(mdev, &priv->channels.params);
 
 	mlx5e_timestamp_init(priv);
-- 
cgit v1.2.3-59-g8ed1b


From 7396ae3d1cfe05be42e5f6b1883c9ed594cc42ba Mon Sep 17 00:00:00 2001
From: Parav Pandit <parav@mellanox.com>
Date: Thu, 5 Dec 2019 16:06:20 -0600
Subject: net/mlx5: Reduce No CQ found log level from warn to debug

In below sequence, a EQE entry arrives for a CQ which is on the path of
being destroyed.

           cpu-0               cpu-1
           ------              -----
mlx5_core_destroy_cq()      mlx5_eq_comp_int()
  mlx5_eq_del_cq()          [..]
    radix_tree_delete()     [..]
  [..]                         mlx5_eq_cq_get() /* Didn't find CQ is
                                                 * a valid case.
                                                 */
  /* destroy CQ in hw */
  mlx5_cmd_exec()

This is still a valid scenario and correct delete CQ sequence, as
mirror of the CQ create sequence.
Hence, suppress the non harmful debug message from warn to debug level.
Keep the debug log message rate limited because user application can
trigger it repeatedly.

Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/eq.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
index 580c71cb9dfa..2c716abc0f27 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
@@ -156,7 +156,8 @@ static int mlx5_eq_comp_int(struct notifier_block *nb,
 			cq->comp(cq, eqe);
 			mlx5_cq_put(cq);
 		} else {
-			mlx5_core_warn(eq->dev, "Completion event for bogus CQ 0x%x\n", cqn);
+			dev_dbg_ratelimited(eq->dev->device,
+					    "Completion event for bogus CQ 0x%x\n", cqn);
 		}
 
 		++eq->cons_index;
-- 
cgit v1.2.3-59-g8ed1b


From 3ed879965cc4ea13fe0908468b653c4ff2cb1309 Mon Sep 17 00:00:00 2001
From: Parav Pandit <parav@mellanox.com>
Date: Fri, 6 Dec 2019 15:13:41 -0600
Subject: net/mlx5: Use async EQ setup cleanup helpers for multiple EQs

Use helper routines to setup and teardown multiple EQs and reuse the
code in setup, cleanup and error unwinding flows.

Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/eq.c | 114 ++++++++++++---------------
 1 file changed, 49 insertions(+), 65 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
index 2c716abc0f27..cccea3a8eddd 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
@@ -564,6 +564,39 @@ static void gather_async_events_mask(struct mlx5_core_dev *dev, u64 mask[4])
 		gather_user_async_events(dev, mask);
 }
 
+static int
+setup_async_eq(struct mlx5_core_dev *dev, struct mlx5_eq_async *eq,
+	       struct mlx5_eq_param *param, const char *name)
+{
+	int err;
+
+	eq->irq_nb.notifier_call = mlx5_eq_async_int;
+
+	err = create_async_eq(dev, &eq->core, param);
+	if (err) {
+		mlx5_core_warn(dev, "failed to create %s EQ %d\n", name, err);
+		return err;
+	}
+	err = mlx5_eq_enable(dev, &eq->core, &eq->irq_nb);
+	if (err) {
+		mlx5_core_warn(dev, "failed to enable %s EQ %d\n", name, err);
+		destroy_async_eq(dev, &eq->core);
+	}
+	return err;
+}
+
+static void cleanup_async_eq(struct mlx5_core_dev *dev,
+			     struct mlx5_eq_async *eq, const char *name)
+{
+	int err;
+
+	mlx5_eq_disable(dev, &eq->core, &eq->irq_nb);
+	err = destroy_async_eq(dev, &eq->core);
+	if (err)
+		mlx5_core_err(dev, "failed to destroy %s eq, err(%d)\n",
+			      name, err);
+}
+
 static int create_async_eqs(struct mlx5_core_dev *dev)
 {
 	struct mlx5_eq_table *table = dev->priv.eq_table;
@@ -573,77 +606,45 @@ static int create_async_eqs(struct mlx5_core_dev *dev)
 	MLX5_NB_INIT(&table->cq_err_nb, cq_err_event_notifier, CQ_ERROR);
 	mlx5_eq_notifier_register(dev, &table->cq_err_nb);
 
-	table->cmd_eq.irq_nb.notifier_call = mlx5_eq_async_int;
 	param = (struct mlx5_eq_param) {
 		.irq_index = 0,
 		.nent = MLX5_NUM_CMD_EQE,
+		.mask[0] = 1ull << MLX5_EVENT_TYPE_CMD,
 	};
-
-	param.mask[0] = 1ull << MLX5_EVENT_TYPE_CMD;
-	err = create_async_eq(dev, &table->cmd_eq.core, &param);
-	if (err) {
-		mlx5_core_warn(dev, "failed to create cmd EQ %d\n", err);
-		goto err0;
-	}
-	err = mlx5_eq_enable(dev, &table->cmd_eq.core, &table->cmd_eq.irq_nb);
-	if (err) {
-		mlx5_core_warn(dev, "failed to enable cmd EQ %d\n", err);
+	err = setup_async_eq(dev, &table->cmd_eq, &param, "cmd");
+	if (err)
 		goto err1;
-	}
+
 	mlx5_cmd_use_events(dev);
 
-	table->async_eq.irq_nb.notifier_call = mlx5_eq_async_int;
 	param = (struct mlx5_eq_param) {
 		.irq_index = 0,
 		.nent = MLX5_NUM_ASYNC_EQE,
 	};
 
 	gather_async_events_mask(dev, param.mask);
-	err = create_async_eq(dev, &table->async_eq.core, &param);
-	if (err) {
-		mlx5_core_warn(dev, "failed to create async EQ %d\n", err);
+	err = setup_async_eq(dev, &table->async_eq, &param, "async");
+	if (err)
 		goto err2;
-	}
-	err = mlx5_eq_enable(dev, &table->async_eq.core,
-			     &table->async_eq.irq_nb);
-	if (err) {
-		mlx5_core_warn(dev, "failed to enable async EQ %d\n", err);
-		goto err3;
-	}
 
-	table->pages_eq.irq_nb.notifier_call = mlx5_eq_async_int;
 	param = (struct mlx5_eq_param) {
 		.irq_index = 0,
 		.nent = /* TODO: sriov max_vf + */ 1,
+		.mask[0] = 1ull << MLX5_EVENT_TYPE_PAGE_REQUEST,
 	};
 
-	param.mask[0] = 1ull << MLX5_EVENT_TYPE_PAGE_REQUEST;
-	err = create_async_eq(dev, &table->pages_eq.core, &param);
-	if (err) {
-		mlx5_core_warn(dev, "failed to create pages EQ %d\n", err);
-		goto err4;
-	}
-	err = mlx5_eq_enable(dev, &table->pages_eq.core,
-			     &table->pages_eq.irq_nb);
-	if (err) {
-		mlx5_core_warn(dev, "failed to enable pages EQ %d\n", err);
-		goto err5;
-	}
+	err = setup_async_eq(dev, &table->pages_eq, &param, "pages");
+	if (err)
+		goto err3;
 
-	return err;
+	return 0;
 
-err5:
-	destroy_async_eq(dev, &table->pages_eq.core);
-err4:
-	mlx5_eq_disable(dev, &table->async_eq.core, &table->async_eq.irq_nb);
 err3:
-	destroy_async_eq(dev, &table->async_eq.core);
+	cleanup_async_eq(dev, &table->async_eq, "async");
 err2:
 	mlx5_cmd_use_polling(dev);
-	mlx5_eq_disable(dev, &table->cmd_eq.core, &table->cmd_eq.irq_nb);
+	cleanup_async_eq(dev, &table->cmd_eq, "cmd");
 err1:
-	destroy_async_eq(dev, &table->cmd_eq.core);
-err0:
 	mlx5_eq_notifier_unregister(dev, &table->cq_err_nb);
 	return err;
 }
@@ -651,28 +652,11 @@ err0:
 static void destroy_async_eqs(struct mlx5_core_dev *dev)
 {
 	struct mlx5_eq_table *table = dev->priv.eq_table;
-	int err;
-
-	mlx5_eq_disable(dev, &table->pages_eq.core, &table->pages_eq.irq_nb);
-	err = destroy_async_eq(dev, &table->pages_eq.core);
-	if (err)
-		mlx5_core_err(dev, "failed to destroy pages eq, err(%d)\n",
-			      err);
-
-	mlx5_eq_disable(dev, &table->async_eq.core, &table->async_eq.irq_nb);
-	err = destroy_async_eq(dev, &table->async_eq.core);
-	if (err)
-		mlx5_core_err(dev, "failed to destroy async eq, err(%d)\n",
-			      err);
 
+	cleanup_async_eq(dev, &table->pages_eq, "pages");
+	cleanup_async_eq(dev, &table->async_eq, "async");
 	mlx5_cmd_use_polling(dev);
-
-	mlx5_eq_disable(dev, &table->cmd_eq.core, &table->cmd_eq.irq_nb);
-	err = destroy_async_eq(dev, &table->cmd_eq.core);
-	if (err)
-		mlx5_core_err(dev, "failed to destroy command eq, err(%d)\n",
-			      err);
-
+	cleanup_async_eq(dev, &table->cmd_eq, "cmd");
 	mlx5_eq_notifier_unregister(dev, &table->cq_err_nb);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From cc78dbd7686acf0a9729a187fdfcb8d2d3a80671 Mon Sep 17 00:00:00 2001
From: Alex Vesker <valex@mellanox.com>
Date: Tue, 24 Dec 2019 22:52:03 +0200
Subject: net/mlx5: DR, Use attributes struct for FW flow table creation

Instead of using multiple variables use a simple struct. The
number of passed argument was too high after adding the encap
decap support bits arguments used for multiple destination reformat.

Signed-off-by: Alex Vesker <valex@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../ethernet/mellanox/mlx5/core/steering/dr_cmd.c  | 39 +++++++++++-----------
 .../ethernet/mellanox/mlx5/core/steering/dr_fw.c   | 12 +++++--
 .../mellanox/mlx5/core/steering/dr_table.c         | 16 +++++----
 .../mellanox/mlx5/core/steering/dr_types.h         | 18 ++++++----
 4 files changed, 50 insertions(+), 35 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c
index 41662c4e2664..ec35b297dcab 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c
@@ -320,12 +320,7 @@ int mlx5dr_cmd_destroy_flow_group(struct mlx5_core_dev *mdev,
 }
 
 int mlx5dr_cmd_create_flow_table(struct mlx5_core_dev *mdev,
-				 u32 table_type,
-				 u64 icm_addr_rx,
-				 u64 icm_addr_tx,
-				 u8 level,
-				 bool sw_owner,
-				 bool term_tbl,
+				 struct mlx5dr_cmd_create_flow_table_attr *attr,
 				 u64 *fdb_rx_icm_addr,
 				 u32 *table_id)
 {
@@ -335,37 +330,43 @@ int mlx5dr_cmd_create_flow_table(struct mlx5_core_dev *mdev,
 	int err;
 
 	MLX5_SET(create_flow_table_in, in, opcode, MLX5_CMD_OP_CREATE_FLOW_TABLE);
-	MLX5_SET(create_flow_table_in, in, table_type, table_type);
+	MLX5_SET(create_flow_table_in, in, table_type, attr->table_type);
 
 	ft_mdev = MLX5_ADDR_OF(create_flow_table_in, in, flow_table_context);
-	MLX5_SET(flow_table_context, ft_mdev, termination_table, term_tbl);
-	MLX5_SET(flow_table_context, ft_mdev, sw_owner, sw_owner);
-	MLX5_SET(flow_table_context, ft_mdev, level, level);
+	MLX5_SET(flow_table_context, ft_mdev, termination_table, attr->term_tbl);
+	MLX5_SET(flow_table_context, ft_mdev, sw_owner, attr->sw_owner);
+	MLX5_SET(flow_table_context, ft_mdev, level, attr->level);
 
-	if (sw_owner) {
+	if (attr->sw_owner) {
 		/* icm_addr_0 used for FDB RX / NIC TX / NIC_RX
 		 * icm_addr_1 used for FDB TX
 		 */
-		if (table_type == MLX5_FLOW_TABLE_TYPE_NIC_RX) {
+		if (attr->table_type == MLX5_FLOW_TABLE_TYPE_NIC_RX) {
 			MLX5_SET64(flow_table_context, ft_mdev,
-				   sw_owner_icm_root_0, icm_addr_rx);
-		} else if (table_type == MLX5_FLOW_TABLE_TYPE_NIC_TX) {
+				   sw_owner_icm_root_0, attr->icm_addr_rx);
+		} else if (attr->table_type == MLX5_FLOW_TABLE_TYPE_NIC_TX) {
 			MLX5_SET64(flow_table_context, ft_mdev,
-				   sw_owner_icm_root_0, icm_addr_tx);
-		} else if (table_type == MLX5_FLOW_TABLE_TYPE_FDB) {
+				   sw_owner_icm_root_0, attr->icm_addr_tx);
+		} else if (attr->table_type == MLX5_FLOW_TABLE_TYPE_FDB) {
 			MLX5_SET64(flow_table_context, ft_mdev,
-				   sw_owner_icm_root_0, icm_addr_rx);
+				   sw_owner_icm_root_0, attr->icm_addr_rx);
 			MLX5_SET64(flow_table_context, ft_mdev,
-				   sw_owner_icm_root_1, icm_addr_tx);
+				   sw_owner_icm_root_1, attr->icm_addr_tx);
 		}
 	}
 
+	MLX5_SET(create_flow_table_in, in, flow_table_context.decap_en,
+		 attr->decap_en);
+	MLX5_SET(create_flow_table_in, in, flow_table_context.reformat_en,
+		 attr->reformat_en);
+
 	err = mlx5_cmd_exec(mdev, in, sizeof(in), out, sizeof(out));
 	if (err)
 		return err;
 
 	*table_id = MLX5_GET(create_flow_table_out, out, table_id);
-	if (!sw_owner && table_type == MLX5_FLOW_TABLE_TYPE_FDB)
+	if (!attr->sw_owner && attr->table_type == MLX5_FLOW_TABLE_TYPE_FDB &&
+	    fdb_rx_icm_addr)
 		*fdb_rx_icm_addr =
 		(u64)MLX5_GET(create_flow_table_out, out, icm_address_31_0) |
 		(u64)MLX5_GET(create_flow_table_out, out, icm_address_39_32) << 32 |
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_fw.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_fw.c
index 60ef6e6171e3..9d2180cb095f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_fw.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_fw.c
@@ -7,6 +7,7 @@
 struct mlx5dr_fw_recalc_cs_ft *
 mlx5dr_fw_create_recalc_cs_ft(struct mlx5dr_domain *dmn, u32 vport_num)
 {
+	struct mlx5dr_cmd_create_flow_table_attr ft_attr = {};
 	struct mlx5dr_fw_recalc_cs_ft *recalc_cs_ft;
 	u32 table_id, group_id, modify_hdr_id;
 	u64 rx_icm_addr, modify_ttl_action;
@@ -16,9 +17,14 @@ mlx5dr_fw_create_recalc_cs_ft(struct mlx5dr_domain *dmn, u32 vport_num)
 	if (!recalc_cs_ft)
 		return NULL;
 
-	ret = mlx5dr_cmd_create_flow_table(dmn->mdev, MLX5_FLOW_TABLE_TYPE_FDB,
-					   0, 0, dmn->info.caps.max_ft_level - 1,
-					   false, true, &rx_icm_addr, &table_id);
+	ft_attr.table_type = MLX5_FLOW_TABLE_TYPE_FDB;
+	ft_attr.level = dmn->info.caps.max_ft_level - 1;
+	ft_attr.term_tbl = true;
+
+	ret = mlx5dr_cmd_create_flow_table(dmn->mdev,
+					   &ft_attr,
+					   &rx_icm_addr,
+					   &table_id);
 	if (ret) {
 		mlx5dr_err(dmn, "Failed creating TTL W/A FW flow table %d\n", ret);
 		goto free_ttl_tbl;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c
index e178d8d3dbc9..7a4e6a43bdea 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c
@@ -211,6 +211,7 @@ static int dr_table_destroy_sw_owned_tbl(struct mlx5dr_table *tbl)
 
 static int dr_table_create_sw_owned_tbl(struct mlx5dr_table *tbl)
 {
+	struct mlx5dr_cmd_create_flow_table_attr ft_attr = {};
 	u64 icm_addr_rx = 0;
 	u64 icm_addr_tx = 0;
 	int ret;
@@ -221,13 +222,14 @@ static int dr_table_create_sw_owned_tbl(struct mlx5dr_table *tbl)
 	if (tbl->tx.s_anchor)
 		icm_addr_tx = tbl->tx.s_anchor->chunk->icm_addr;
 
-	ret = mlx5dr_cmd_create_flow_table(tbl->dmn->mdev,
-					   tbl->table_type,
-					   icm_addr_rx,
-					   icm_addr_tx,
-					   tbl->dmn->info.caps.max_ft_level - 1,
-					   true, false, NULL,
-					   &tbl->table_id);
+	ft_attr.table_type = tbl->table_type;
+	ft_attr.icm_addr_rx = icm_addr_rx;
+	ft_attr.icm_addr_tx = icm_addr_tx;
+	ft_attr.level = tbl->dmn->info.caps.max_ft_level - 1;
+	ft_attr.sw_owner = true;
+
+	ret = mlx5dr_cmd_create_flow_table(tbl->dmn->mdev, &ft_attr,
+					   NULL, &tbl->table_id);
 
 	return ret;
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
index 290fe61c33d0..bc82b76cf04e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
@@ -867,6 +867,17 @@ struct mlx5dr_cmd_query_flow_table_details {
 	u64 sw_owner_icm_root_0;
 };
 
+struct mlx5dr_cmd_create_flow_table_attr {
+	u32 table_type;
+	u64 icm_addr_rx;
+	u64 icm_addr_tx;
+	u8 level;
+	bool sw_owner;
+	bool term_tbl;
+	bool decap_en;
+	bool reformat_en;
+};
+
 /* internal API functions */
 int mlx5dr_cmd_query_device(struct mlx5_core_dev *mdev,
 			    struct mlx5dr_cmd_caps *caps);
@@ -904,12 +915,7 @@ int mlx5dr_cmd_destroy_flow_group(struct mlx5_core_dev *mdev,
 				  u32 table_id,
 				  u32 group_id);
 int mlx5dr_cmd_create_flow_table(struct mlx5_core_dev *mdev,
-				 u32 table_type,
-				 u64 icm_addr_rx,
-				 u64 icm_addr_tx,
-				 u8 level,
-				 bool sw_owner,
-				 bool term_tbl,
+				 struct mlx5dr_cmd_create_flow_table_attr *attr,
 				 u64 *fdb_rx_icm_addr,
 				 u32 *table_id);
 int mlx5dr_cmd_destroy_flow_table(struct mlx5_core_dev *mdev,
-- 
cgit v1.2.3-59-g8ed1b


From 6de03d2dcb0ec53e5862ae5844563bbbccde74bb Mon Sep 17 00:00:00 2001
From: Erez Shitrit <erezsh@mellanox.com>
Date: Thu, 2 Jan 2020 15:54:35 +0200
Subject: net/mlx5: DR, Create FTE entry in the FW from SW-steering

Implement the FW command to setup a FTE (Flow Table Entry) into the FW
managed flow tables.

Signed-off-by: Erez Shitrit <erezsh@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../ethernet/mellanox/mlx5/core/steering/dr_cmd.c  | 205 +++++++++++++++++++++
 .../mellanox/mlx5/core/steering/dr_types.h         |  37 ++++
 2 files changed, 242 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c
index ec35b297dcab..461b39376daf 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c
@@ -479,3 +479,208 @@ int mlx5dr_cmd_query_gid(struct mlx5_core_dev *mdev, u8 vhca_port_num,
 
 	return 0;
 }
+
+static int mlx5dr_cmd_set_extended_dest(struct mlx5_core_dev *dev,
+					struct mlx5dr_cmd_fte_info *fte,
+					bool *extended_dest)
+{
+	int fw_log_max_fdb_encap_uplink = MLX5_CAP_ESW(dev, log_max_fdb_encap_uplink);
+	int num_fwd_destinations = 0;
+	int num_encap = 0;
+	int i;
+
+	*extended_dest = false;
+	if (!(fte->action.action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST))
+		return 0;
+	for (i = 0; i < fte->dests_size; i++) {
+		if (fte->dest_arr[i].type == MLX5_FLOW_DESTINATION_TYPE_COUNTER)
+			continue;
+		if (fte->dest_arr[i].type == MLX5_FLOW_DESTINATION_TYPE_VPORT &&
+		    fte->dest_arr[i].vport.flags & MLX5_FLOW_DEST_VPORT_REFORMAT_ID)
+			num_encap++;
+		num_fwd_destinations++;
+	}
+
+	if (num_fwd_destinations > 1 && num_encap > 0)
+		*extended_dest = true;
+
+	if (*extended_dest && !fw_log_max_fdb_encap_uplink) {
+		mlx5_core_warn(dev, "FW does not support extended destination");
+		return -EOPNOTSUPP;
+	}
+	if (num_encap > (1 << fw_log_max_fdb_encap_uplink)) {
+		mlx5_core_warn(dev, "FW does not support more than %d encaps",
+			       1 << fw_log_max_fdb_encap_uplink);
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+int mlx5dr_cmd_set_fte(struct mlx5_core_dev *dev,
+		       int opmod, int modify_mask,
+		       struct mlx5dr_cmd_ft_info *ft,
+		       u32 group_id,
+		       struct mlx5dr_cmd_fte_info *fte)
+{
+	u32 out[MLX5_ST_SZ_DW(set_fte_out)] = {};
+	void *in_flow_context, *vlan;
+	bool extended_dest = false;
+	void *in_match_value;
+	unsigned int inlen;
+	int dst_cnt_size;
+	void *in_dests;
+	u32 *in;
+	int err;
+	int i;
+
+	if (mlx5dr_cmd_set_extended_dest(dev, fte, &extended_dest))
+		return -EOPNOTSUPP;
+
+	if (!extended_dest)
+		dst_cnt_size = MLX5_ST_SZ_BYTES(dest_format_struct);
+	else
+		dst_cnt_size = MLX5_ST_SZ_BYTES(extended_dest_format);
+
+	inlen = MLX5_ST_SZ_BYTES(set_fte_in) + fte->dests_size * dst_cnt_size;
+	in = kvzalloc(inlen, GFP_KERNEL);
+	if (!in)
+		return -ENOMEM;
+
+	MLX5_SET(set_fte_in, in, opcode, MLX5_CMD_OP_SET_FLOW_TABLE_ENTRY);
+	MLX5_SET(set_fte_in, in, op_mod, opmod);
+	MLX5_SET(set_fte_in, in, modify_enable_mask, modify_mask);
+	MLX5_SET(set_fte_in, in, table_type, ft->type);
+	MLX5_SET(set_fte_in, in, table_id, ft->id);
+	MLX5_SET(set_fte_in, in, flow_index, fte->index);
+	if (ft->vport) {
+		MLX5_SET(set_fte_in, in, vport_number, ft->vport);
+		MLX5_SET(set_fte_in, in, other_vport, 1);
+	}
+
+	in_flow_context = MLX5_ADDR_OF(set_fte_in, in, flow_context);
+	MLX5_SET(flow_context, in_flow_context, group_id, group_id);
+
+	MLX5_SET(flow_context, in_flow_context, flow_tag,
+		 fte->flow_context.flow_tag);
+	MLX5_SET(flow_context, in_flow_context, flow_source,
+		 fte->flow_context.flow_source);
+
+	MLX5_SET(flow_context, in_flow_context, extended_destination,
+		 extended_dest);
+	if (extended_dest) {
+		u32 action;
+
+		action = fte->action.action &
+			~MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT;
+		MLX5_SET(flow_context, in_flow_context, action, action);
+	} else {
+		MLX5_SET(flow_context, in_flow_context, action,
+			 fte->action.action);
+		if (fte->action.pkt_reformat)
+			MLX5_SET(flow_context, in_flow_context, packet_reformat_id,
+				 fte->action.pkt_reformat->id);
+	}
+	if (fte->action.modify_hdr)
+		MLX5_SET(flow_context, in_flow_context, modify_header_id,
+			 fte->action.modify_hdr->id);
+
+	vlan = MLX5_ADDR_OF(flow_context, in_flow_context, push_vlan);
+
+	MLX5_SET(vlan, vlan, ethtype, fte->action.vlan[0].ethtype);
+	MLX5_SET(vlan, vlan, vid, fte->action.vlan[0].vid);
+	MLX5_SET(vlan, vlan, prio, fte->action.vlan[0].prio);
+
+	vlan = MLX5_ADDR_OF(flow_context, in_flow_context, push_vlan_2);
+
+	MLX5_SET(vlan, vlan, ethtype, fte->action.vlan[1].ethtype);
+	MLX5_SET(vlan, vlan, vid, fte->action.vlan[1].vid);
+	MLX5_SET(vlan, vlan, prio, fte->action.vlan[1].prio);
+
+	in_match_value = MLX5_ADDR_OF(flow_context, in_flow_context,
+				      match_value);
+	memcpy(in_match_value, fte->val, sizeof(u32) * MLX5_ST_SZ_DW_MATCH_PARAM);
+
+	in_dests = MLX5_ADDR_OF(flow_context, in_flow_context, destination);
+	if (fte->action.action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) {
+		int list_size = 0;
+
+		for (i = 0; i < fte->dests_size; i++) {
+			unsigned int id, type = fte->dest_arr[i].type;
+
+			if (type == MLX5_FLOW_DESTINATION_TYPE_COUNTER)
+				continue;
+
+			switch (type) {
+			case MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE_NUM:
+				id = fte->dest_arr[i].ft_num;
+				type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
+				break;
+			case MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE:
+				id = fte->dest_arr[i].ft_id;
+				break;
+			case MLX5_FLOW_DESTINATION_TYPE_VPORT:
+				id = fte->dest_arr[i].vport.num;
+				MLX5_SET(dest_format_struct, in_dests,
+					 destination_eswitch_owner_vhca_id_valid,
+					 !!(fte->dest_arr[i].vport.flags &
+					    MLX5_FLOW_DEST_VPORT_VHCA_ID));
+				MLX5_SET(dest_format_struct, in_dests,
+					 destination_eswitch_owner_vhca_id,
+					 fte->dest_arr[i].vport.vhca_id);
+				if (extended_dest && (fte->dest_arr[i].vport.flags &
+						    MLX5_FLOW_DEST_VPORT_REFORMAT_ID)) {
+					MLX5_SET(dest_format_struct, in_dests,
+						 packet_reformat,
+						 !!(fte->dest_arr[i].vport.flags &
+						    MLX5_FLOW_DEST_VPORT_REFORMAT_ID));
+					MLX5_SET(extended_dest_format, in_dests,
+						 packet_reformat_id,
+						 fte->dest_arr[i].vport.reformat_id);
+				}
+				break;
+			default:
+				id = fte->dest_arr[i].tir_num;
+			}
+
+			MLX5_SET(dest_format_struct, in_dests, destination_type,
+				 type);
+			MLX5_SET(dest_format_struct, in_dests, destination_id, id);
+			in_dests += dst_cnt_size;
+			list_size++;
+		}
+
+		MLX5_SET(flow_context, in_flow_context, destination_list_size,
+			 list_size);
+	}
+
+	if (fte->action.action & MLX5_FLOW_CONTEXT_ACTION_COUNT) {
+		int max_list_size = BIT(MLX5_CAP_FLOWTABLE_TYPE(dev,
+					log_max_flow_counter,
+					ft->type));
+		int list_size = 0;
+
+		for (i = 0; i < fte->dests_size; i++) {
+			if (fte->dest_arr[i].type !=
+			    MLX5_FLOW_DESTINATION_TYPE_COUNTER)
+				continue;
+
+			MLX5_SET(flow_counter_list, in_dests, flow_counter_id,
+				 fte->dest_arr[i].counter_id);
+			in_dests += dst_cnt_size;
+			list_size++;
+		}
+		if (list_size > max_list_size) {
+			err = -EINVAL;
+			goto err_out;
+		}
+
+		MLX5_SET(flow_context, in_flow_context, flow_counter_list_size,
+			 list_size);
+	}
+
+	err = mlx5_cmd_exec(dev, in, inlen, out, sizeof(out));
+err_out:
+	kvfree(in);
+	return err;
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
index bc82b76cf04e..09ac9aadad1a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
@@ -1057,6 +1057,43 @@ int mlx5dr_send_postsend_formatted_htbl(struct mlx5dr_domain *dmn,
 int mlx5dr_send_postsend_action(struct mlx5dr_domain *dmn,
 				struct mlx5dr_action *action);
 
+struct mlx5dr_cmd_ft_info {
+	u32 id;
+	u16 vport;
+	enum fs_flow_table_type type;
+};
+
+struct mlx5dr_cmd_flow_destination_hw_info {
+	enum mlx5_flow_destination_type type;
+	union {
+		u32 tir_num;
+		u32 ft_num;
+		u32 ft_id;
+		u32 counter_id;
+		struct {
+			u16 num;
+			u16 vhca_id;
+			u32 reformat_id;
+			u8 flags;
+		} vport;
+	};
+};
+
+struct mlx5dr_cmd_fte_info {
+	u32 dests_size;
+	u32 index;
+	struct mlx5_flow_context flow_context;
+	u32 *val;
+	struct mlx5_flow_act action;
+	struct mlx5dr_cmd_flow_destination_hw_info *dest_arr;
+};
+
+int mlx5dr_cmd_set_fte(struct mlx5_core_dev *dev,
+		       int opmod, int modify_mask,
+		       struct mlx5dr_cmd_ft_info *ft,
+		       u32 group_id,
+		       struct mlx5dr_cmd_fte_info *fte);
+
 struct mlx5dr_fw_recalc_cs_ft {
 	u64 rx_icm_addr;
 	u32 table_id;
-- 
cgit v1.2.3-59-g8ed1b


From 34583beea4b741589793ce3ad5436b0a7c6d0cdb Mon Sep 17 00:00:00 2001
From: Erez Shitrit <erezsh@mellanox.com>
Date: Tue, 24 Dec 2019 22:54:15 +0200
Subject: net/mlx5: DR, Create multi-destination table for SW-steering use

Currently SW steering doesn't have the means to access HW iterators to
support multi-destination (FTEs) flow table entries.

In order to support multi-destination FTEs for port-mirroring, SW
steering will create a dedicated multi-destination FW managed flow table
and FTEs via direct FW commands that we introduced in the previous patch.

Signed-off-by: Erez Shitrit <erezsh@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../ethernet/mellanox/mlx5/core/steering/dr_fw.c   | 67 ++++++++++++++++++++++
 .../mellanox/mlx5/core/steering/dr_types.h         |  8 +++
 2 files changed, 75 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_fw.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_fw.c
index 9d2180cb095f..1fbcd012bb85 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_fw.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_fw.c
@@ -97,3 +97,70 @@ void mlx5dr_fw_destroy_recalc_cs_ft(struct mlx5dr_domain *dmn,
 
 	kfree(recalc_cs_ft);
 }
+
+int mlx5dr_fw_create_md_tbl(struct mlx5dr_domain *dmn,
+			    struct mlx5dr_cmd_flow_destination_hw_info *dest,
+			    int num_dest,
+			    bool reformat_req,
+			    u32 *tbl_id,
+			    u32 *group_id)
+{
+	struct mlx5dr_cmd_create_flow_table_attr ft_attr = {};
+	struct mlx5dr_cmd_fte_info fte_info = {};
+	u32 val[MLX5_ST_SZ_DW_MATCH_PARAM] = {};
+	struct mlx5dr_cmd_ft_info ft_info = {};
+	int ret;
+
+	ft_attr.table_type = MLX5_FLOW_TABLE_TYPE_FDB;
+	ft_attr.level = dmn->info.caps.max_ft_level - 2;
+	ft_attr.reformat_en = reformat_req;
+	ft_attr.decap_en = reformat_req;
+
+	ret = mlx5dr_cmd_create_flow_table(dmn->mdev, &ft_attr, NULL, tbl_id);
+	if (ret) {
+		mlx5dr_err(dmn, "Failed creating multi dest FW flow table %d\n", ret);
+		return ret;
+	}
+
+	ret = mlx5dr_cmd_create_empty_flow_group(dmn->mdev,
+						 MLX5_FLOW_TABLE_TYPE_FDB,
+						 *tbl_id, group_id);
+	if (ret) {
+		mlx5dr_err(dmn, "Failed creating multi dest FW flow group %d\n", ret);
+		goto free_flow_table;
+	}
+
+	ft_info.id = *tbl_id;
+	ft_info.type = FS_FT_FDB;
+	fte_info.action.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
+	fte_info.dests_size = num_dest;
+	fte_info.val = val;
+	fte_info.dest_arr = dest;
+
+	ret = mlx5dr_cmd_set_fte(dmn->mdev, 0, 0, &ft_info, *group_id, &fte_info);
+	if (ret) {
+		mlx5dr_err(dmn, "Failed setting fte into table %d\n", ret);
+		goto free_flow_group;
+	}
+
+	return 0;
+
+free_flow_group:
+	mlx5dr_cmd_destroy_flow_group(dmn->mdev, MLX5_FLOW_TABLE_TYPE_FDB,
+				      *tbl_id, *group_id);
+free_flow_table:
+	mlx5dr_cmd_destroy_flow_table(dmn->mdev, *tbl_id,
+				      MLX5_FLOW_TABLE_TYPE_FDB);
+	return ret;
+}
+
+void mlx5dr_fw_destroy_md_tbl(struct mlx5dr_domain *dmn,
+			      u32 tbl_id, u32 group_id)
+{
+	mlx5dr_cmd_del_flow_table_entry(dmn->mdev, FS_FT_FDB, tbl_id);
+	mlx5dr_cmd_destroy_flow_group(dmn->mdev,
+				      MLX5_FLOW_TABLE_TYPE_FDB,
+				      tbl_id, group_id);
+	mlx5dr_cmd_destroy_flow_table(dmn->mdev, tbl_id,
+				      MLX5_FLOW_TABLE_TYPE_FDB);
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
index 09ac9aadad1a..c37226a14311 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
@@ -1108,4 +1108,12 @@ void mlx5dr_fw_destroy_recalc_cs_ft(struct mlx5dr_domain *dmn,
 int mlx5dr_domain_cache_get_recalc_cs_ft_addr(struct mlx5dr_domain *dmn,
 					      u32 vport_num,
 					      u64 *rx_icm_addr);
+int mlx5dr_fw_create_md_tbl(struct mlx5dr_domain *dmn,
+			    struct mlx5dr_cmd_flow_destination_hw_info *dest,
+			    int num_dest,
+			    bool reformat_req,
+			    u32 *tbl_id,
+			    u32 *group_id);
+void mlx5dr_fw_destroy_md_tbl(struct mlx5dr_domain *dmn, u32 tbl_id,
+			      u32 group_id);
 #endif  /* _DR_TYPES_H_ */
-- 
cgit v1.2.3-59-g8ed1b


From 988fd6b32d07400bd66259603ac9e08c33c02a05 Mon Sep 17 00:00:00 2001
From: Erez Shitrit <erezsh@mellanox.com>
Date: Thu, 2 Jan 2020 16:55:45 +0200
Subject: net/mlx5: DR, Pass table flags at creation to lower layer

We need to have the flow-table flags when creation sw-steering tables,
this parameter exists in the layer between fs_core to sw_steering, this
patch gives it to the creation function.

Signed-off-by: Erez Shitrit <erezsh@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c | 7 ++++++-
 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h | 1 +
 drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c    | 2 +-
 drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h   | 4 ++--
 4 files changed, 10 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c
index 7a4e6a43bdea..14ce2d7dbb66 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c
@@ -211,6 +211,8 @@ static int dr_table_destroy_sw_owned_tbl(struct mlx5dr_table *tbl)
 
 static int dr_table_create_sw_owned_tbl(struct mlx5dr_table *tbl)
 {
+	bool en_encap = !!(tbl->flags & MLX5_FLOW_TABLE_TUNNEL_EN_REFORMAT);
+	bool en_decap = !!(tbl->flags & MLX5_FLOW_TABLE_TUNNEL_EN_DECAP);
 	struct mlx5dr_cmd_create_flow_table_attr ft_attr = {};
 	u64 icm_addr_rx = 0;
 	u64 icm_addr_tx = 0;
@@ -227,6 +229,8 @@ static int dr_table_create_sw_owned_tbl(struct mlx5dr_table *tbl)
 	ft_attr.icm_addr_tx = icm_addr_tx;
 	ft_attr.level = tbl->dmn->info.caps.max_ft_level - 1;
 	ft_attr.sw_owner = true;
+	ft_attr.decap_en = en_decap;
+	ft_attr.reformat_en = en_encap;
 
 	ret = mlx5dr_cmd_create_flow_table(tbl->dmn->mdev, &ft_attr,
 					   NULL, &tbl->table_id);
@@ -234,7 +238,7 @@ static int dr_table_create_sw_owned_tbl(struct mlx5dr_table *tbl)
 	return ret;
 }
 
-struct mlx5dr_table *mlx5dr_table_create(struct mlx5dr_domain *dmn, u32 level)
+struct mlx5dr_table *mlx5dr_table_create(struct mlx5dr_domain *dmn, u32 level, u32 flags)
 {
 	struct mlx5dr_table *tbl;
 	int ret;
@@ -247,6 +251,7 @@ struct mlx5dr_table *mlx5dr_table_create(struct mlx5dr_domain *dmn, u32 level)
 
 	tbl->dmn = dmn;
 	tbl->level = level;
+	tbl->flags = flags;
 	refcount_set(&tbl->refcount, 1);
 
 	ret = dr_table_init(tbl);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
index c37226a14311..de6bfa655326 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
@@ -679,6 +679,7 @@ struct mlx5dr_table {
 	u32 level;
 	u32 table_type;
 	u32 table_id;
+	u32 flags;
 	struct list_head matcher_list;
 	struct mlx5dr_action *miss_action;
 	refcount_t refcount;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c
index 3d587d0bdbbe..8ed0f087b1e0 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c
@@ -74,7 +74,7 @@ static int mlx5_cmd_dr_create_flow_table(struct mlx5_flow_root_namespace *ns,
 								    next_ft);
 
 	tbl = mlx5dr_table_create(ns->fs_dr_domain.dr_domain,
-				  ft->level);
+				  ft->level, ft->flags);
 	if (!tbl) {
 		mlx5_core_err(ns->dev, "Failed creating dr flow_table\n");
 		return -EINVAL;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h
index adda9cbfba45..fb3ac697df1b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h
@@ -46,7 +46,7 @@ void mlx5dr_domain_set_peer(struct mlx5dr_domain *dmn,
 			    struct mlx5dr_domain *peer_dmn);
 
 struct mlx5dr_table *
-mlx5dr_table_create(struct mlx5dr_domain *domain, u32 level);
+mlx5dr_table_create(struct mlx5dr_domain *domain, u32 level, u32 flags);
 
 int mlx5dr_table_destroy(struct mlx5dr_table *table);
 
@@ -131,7 +131,7 @@ mlx5dr_domain_set_peer(struct mlx5dr_domain *dmn,
 		       struct mlx5dr_domain *peer_dmn) { }
 
 static inline struct mlx5dr_table *
-mlx5dr_table_create(struct mlx5dr_domain *domain, u32 level) { return NULL; }
+mlx5dr_table_create(struct mlx5dr_domain *domain, u32 level, u32 flags) { return NULL; }
 
 static inline int
 mlx5dr_table_destroy(struct mlx5dr_table *table) { return 0; }
-- 
cgit v1.2.3-59-g8ed1b


From aec292ee6f290920fc77907dfe1b94baa600484d Mon Sep 17 00:00:00 2001
From: Alex Vesker <valex@mellanox.com>
Date: Sun, 15 Dec 2019 21:15:15 +0200
Subject: net/mlx5: DR, Align dest FT action creation to API

Function prefix was changed to be similar to other action APIs.
In order to support other FW tables the mlx5_flow_table struct was
replaced with table id and type.

Signed-off-by: Alex Vesker <valex@mellanox.com>
Reviewed-by: Erez Shitrit <erezsh@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../mellanox/mlx5/core/steering/dr_action.c         | 21 +++++++++++++--------
 .../ethernet/mellanox/mlx5/core/steering/dr_types.h |  5 +++--
 .../ethernet/mellanox/mlx5/core/steering/fs_dr.c    |  6 +++---
 .../ethernet/mellanox/mlx5/core/steering/mlx5dr.h   |  8 ++++----
 4 files changed, 23 insertions(+), 17 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
index 004c56c2fc0c..3e2318761c8c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
@@ -690,9 +690,9 @@ int mlx5dr_actions_build_ste_arr(struct mlx5dr_matcher *matcher,
 
 				/* get the relevant addresses */
 				if (!action->dest_tbl.fw_tbl.rx_icm_addr) {
-					ret = mlx5dr_cmd_query_flow_table(action->dest_tbl.fw_tbl.mdev,
-									  action->dest_tbl.fw_tbl.ft->type,
-									  action->dest_tbl.fw_tbl.ft->id,
+					ret = mlx5dr_cmd_query_flow_table(dmn->mdev,
+									  action->dest_tbl.fw_tbl.type,
+									  action->dest_tbl.fw_tbl.id,
 									  &output);
 					if (!ret) {
 						action->dest_tbl.fw_tbl.tx_icm_addr =
@@ -982,8 +982,8 @@ dec_ref:
 }
 
 struct mlx5dr_action *
-mlx5dr_create_action_dest_flow_fw_table(struct mlx5_flow_table *ft,
-					struct mlx5_core_dev *mdev)
+mlx5dr_action_create_dest_flow_fw_table(struct mlx5dr_domain *dmn,
+					struct mlx5_flow_table *ft)
 {
 	struct mlx5dr_action *action;
 
@@ -992,8 +992,11 @@ mlx5dr_create_action_dest_flow_fw_table(struct mlx5_flow_table *ft,
 		return NULL;
 
 	action->dest_tbl.is_fw_tbl = 1;
-	action->dest_tbl.fw_tbl.ft = ft;
-	action->dest_tbl.fw_tbl.mdev = mdev;
+	action->dest_tbl.fw_tbl.type = ft->type;
+	action->dest_tbl.fw_tbl.id = ft->id;
+	action->dest_tbl.fw_tbl.dmn = dmn;
+
+	refcount_inc(&dmn->refcount);
 
 	return action;
 }
@@ -1559,7 +1562,9 @@ int mlx5dr_action_destroy(struct mlx5dr_action *action)
 
 	switch (action->action_type) {
 	case DR_ACTION_TYP_FT:
-		if (!action->dest_tbl.is_fw_tbl)
+		if (action->dest_tbl.is_fw_tbl)
+			refcount_dec(&action->dest_tbl.fw_tbl.dmn->refcount);
+		else
 			refcount_dec(&action->dest_tbl.tbl->refcount);
 		break;
 	case DR_ACTION_TYP_TNL_L2_TO_L2:
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
index de6bfa655326..27f1d931bf9f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
@@ -743,10 +743,11 @@ struct mlx5dr_action {
 			union {
 				struct mlx5dr_table *tbl;
 				struct {
-					struct mlx5_flow_table *ft;
+					struct mlx5dr_domain *dmn;
+					u32 id;
+					enum fs_flow_table_type type;
 					u64 rx_icm_addr;
 					u64 tx_icm_addr;
-					struct mlx5_core_dev *mdev;
 				} fw_tbl;
 			};
 		} dest_tbl;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c
index 8ed0f087b1e0..e51262ec77bb 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c
@@ -184,13 +184,13 @@ static struct mlx5dr_action *create_vport_action(struct mlx5dr_domain *domain,
 					       dest_attr->vport.vhca_id);
 }
 
-static struct mlx5dr_action *create_ft_action(struct mlx5_core_dev *dev,
+static struct mlx5dr_action *create_ft_action(struct mlx5dr_domain *domain,
 					      struct mlx5_flow_rule *dst)
 {
 	struct mlx5_flow_table *dest_ft = dst->dest_attr.ft;
 
 	if (mlx5_dr_is_fw_table(dest_ft->flags))
-		return mlx5dr_create_action_dest_flow_fw_table(dest_ft, dev);
+		return mlx5dr_action_create_dest_flow_fw_table(domain, dest_ft);
 	return mlx5dr_action_create_dest_table(dest_ft->fs_dr_table.dr_table);
 }
 
@@ -373,7 +373,7 @@ static int mlx5_cmd_dr_create_fte(struct mlx5_flow_root_namespace *ns,
 				actions[num_actions++] = tmp_action;
 				break;
 			case MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE:
-				tmp_action = create_ft_action(dev, dst);
+				tmp_action = create_ft_action(domain, dst);
 				if (!tmp_action) {
 					err = -ENOMEM;
 					goto free_actions;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h
index fb3ac697df1b..932362d89c66 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h
@@ -75,8 +75,8 @@ struct mlx5dr_action *
 mlx5dr_action_create_dest_table(struct mlx5dr_table *table);
 
 struct mlx5dr_action *
-mlx5dr_create_action_dest_flow_fw_table(struct mlx5_flow_table *ft,
-					struct mlx5_core_dev *mdev);
+mlx5dr_action_create_dest_flow_fw_table(struct mlx5dr_domain *domain,
+					struct mlx5_flow_table *ft);
 
 struct mlx5dr_action *
 mlx5dr_action_create_dest_vport(struct mlx5dr_domain *domain,
@@ -165,8 +165,8 @@ static inline struct mlx5dr_action *
 mlx5dr_action_create_dest_table(struct mlx5dr_table *table) { return NULL; }
 
 static inline struct mlx5dr_action *
-mlx5dr_create_action_dest_flow_fw_table(struct mlx5_flow_table *ft,
-					struct mlx5_core_dev *mdev) { return NULL; }
+mlx5dr_action_create_dest_flow_fw_table(struct mlx5dr_domain *domain,
+					struct mlx5_flow_table *ft) { return NULL; }
 
 static inline struct mlx5dr_action *
 mlx5dr_action_create_dest_vport(struct mlx5dr_domain *domain,
-- 
cgit v1.2.3-59-g8ed1b


From b8853c969f406bfb682740b7646ee8b4623ee955 Mon Sep 17 00:00:00 2001
From: Alex Vesker <valex@mellanox.com>
Date: Sun, 15 Dec 2019 21:27:54 +0200
Subject: net/mlx5: DR, Add support for multiple destination table action

A multiple destination table action allows HW packet duplication
to multiple destinations, this is useful for multicast or mirroring
traffic for debug. Duplicating is done using a FW flow table with
multiple destinations.

The new action creation function, mlx5dr_action_create_mult_dest_tbl
will allow creating a single table to iterate over several dr actions.

Signed-off-by: Alex Vesker <valex@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../mellanox/mlx5/core/steering/dr_action.c        | 114 +++++++++++++++++++++
 .../mellanox/mlx5/core/steering/dr_types.h         |   3 +
 .../ethernet/mellanox/mlx5/core/steering/mlx5dr.h  |  15 +++
 3 files changed, 132 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
index 3e2318761c8c..9359eed10889 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
@@ -981,6 +981,104 @@ dec_ref:
 	return NULL;
 }
 
+struct mlx5dr_action *
+mlx5dr_action_create_mult_dest_tbl(struct mlx5dr_domain *dmn,
+				   struct mlx5dr_action_dest *dests,
+				   u32 num_of_dests)
+{
+	struct mlx5dr_cmd_flow_destination_hw_info *hw_dests;
+	struct mlx5dr_action **ref_actions;
+	struct mlx5dr_action *action;
+	bool reformat_req = false;
+	u32 num_of_ref = 0;
+	int ret;
+	int i;
+
+	if (dmn->type != MLX5DR_DOMAIN_TYPE_FDB) {
+		mlx5dr_err(dmn, "Multiple destination support is for FDB only\n");
+		return NULL;
+	}
+
+	hw_dests = kzalloc(sizeof(*hw_dests) * num_of_dests, GFP_KERNEL);
+	if (!hw_dests)
+		return NULL;
+
+	ref_actions = kzalloc(sizeof(*ref_actions) * num_of_dests * 2, GFP_KERNEL);
+	if (!ref_actions)
+		goto free_hw_dests;
+
+	for (i = 0; i < num_of_dests; i++) {
+		struct mlx5dr_action *reformat_action = dests[i].reformat;
+		struct mlx5dr_action *dest_action = dests[i].dest;
+
+		ref_actions[num_of_ref++] = dest_action;
+
+		switch (dest_action->action_type) {
+		case DR_ACTION_TYP_VPORT:
+			hw_dests[i].vport.flags = MLX5_FLOW_DEST_VPORT_VHCA_ID;
+			hw_dests[i].type = MLX5_FLOW_DESTINATION_TYPE_VPORT;
+			hw_dests[i].vport.num = dest_action->vport.caps->num;
+			hw_dests[i].vport.vhca_id = dest_action->vport.caps->vhca_gvmi;
+			if (reformat_action) {
+				reformat_req = true;
+				hw_dests[i].vport.reformat_id =
+					reformat_action->reformat.reformat_id;
+				ref_actions[num_of_ref++] = reformat_action;
+				hw_dests[i].vport.flags |= MLX5_FLOW_DEST_VPORT_REFORMAT_ID;
+			}
+			break;
+
+		case DR_ACTION_TYP_FT:
+			hw_dests[i].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
+			if (dest_action->dest_tbl.is_fw_tbl)
+				hw_dests[i].ft_id = dest_action->dest_tbl.fw_tbl.id;
+			else
+				hw_dests[i].ft_id = dest_action->dest_tbl.tbl->table_id;
+			break;
+
+		default:
+			mlx5dr_dbg(dmn, "Invalid multiple destinations action\n");
+			goto free_ref_actions;
+		}
+	}
+
+	action = dr_action_create_generic(DR_ACTION_TYP_FT);
+	if (!action)
+		goto free_ref_actions;
+
+	ret = mlx5dr_fw_create_md_tbl(dmn,
+				      hw_dests,
+				      num_of_dests,
+				      reformat_req,
+				      &action->dest_tbl.fw_tbl.id,
+				      &action->dest_tbl.fw_tbl.group_id);
+	if (ret)
+		goto free_action;
+
+	refcount_inc(&dmn->refcount);
+
+	for (i = 0; i < num_of_ref; i++)
+		refcount_inc(&ref_actions[i]->refcount);
+
+	action->dest_tbl.is_fw_tbl = true;
+	action->dest_tbl.fw_tbl.dmn = dmn;
+	action->dest_tbl.fw_tbl.type = FS_FT_FDB;
+	action->dest_tbl.fw_tbl.ref_actions = ref_actions;
+	action->dest_tbl.fw_tbl.num_of_ref_actions = num_of_ref;
+
+	kfree(hw_dests);
+
+	return action;
+
+free_action:
+	kfree(action);
+free_ref_actions:
+	kfree(ref_actions);
+free_hw_dests:
+	kfree(hw_dests);
+	return NULL;
+}
+
 struct mlx5dr_action *
 mlx5dr_action_create_dest_flow_fw_table(struct mlx5dr_domain *dmn,
 					struct mlx5_flow_table *ft)
@@ -1566,6 +1664,22 @@ int mlx5dr_action_destroy(struct mlx5dr_action *action)
 			refcount_dec(&action->dest_tbl.fw_tbl.dmn->refcount);
 		else
 			refcount_dec(&action->dest_tbl.tbl->refcount);
+
+		if (action->dest_tbl.is_fw_tbl &&
+		    action->dest_tbl.fw_tbl.num_of_ref_actions) {
+			struct mlx5dr_action **ref_actions;
+			int i;
+
+			ref_actions = action->dest_tbl.fw_tbl.ref_actions;
+			for (i = 0; i < action->dest_tbl.fw_tbl.num_of_ref_actions; i++)
+				refcount_dec(&ref_actions[i]->refcount);
+
+			kfree(ref_actions);
+
+			mlx5dr_fw_destroy_md_tbl(action->dest_tbl.fw_tbl.dmn,
+						 action->dest_tbl.fw_tbl.id,
+						 action->dest_tbl.fw_tbl.group_id);
+		}
 		break;
 	case DR_ACTION_TYP_TNL_L2_TO_L2:
 		refcount_dec(&action->reformat.dmn->refcount);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
index 27f1d931bf9f..0fc52d634e10 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
@@ -745,9 +745,12 @@ struct mlx5dr_action {
 				struct {
 					struct mlx5dr_domain *dmn;
 					u32 id;
+					u32 group_id;
 					enum fs_flow_table_type type;
 					u64 rx_icm_addr;
 					u64 tx_icm_addr;
+					struct mlx5dr_action **ref_actions;
+					u32 num_of_ref_actions;
 				} fw_tbl;
 			};
 		} dest_tbl;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h
index 932362d89c66..e1edc9c247b7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h
@@ -33,6 +33,11 @@ struct mlx5dr_match_parameters {
 	u64 *match_buf; /* Device spec format */
 };
 
+struct mlx5dr_action_dest {
+	struct mlx5dr_action *dest;
+	struct mlx5dr_action *reformat;
+};
+
 #ifdef CONFIG_MLX5_SW_STEERING
 
 struct mlx5dr_domain *
@@ -83,6 +88,11 @@ mlx5dr_action_create_dest_vport(struct mlx5dr_domain *domain,
 				u32 vport, u8 vhca_id_valid,
 				u16 vhca_id);
 
+struct mlx5dr_action *
+mlx5dr_action_create_mult_dest_tbl(struct mlx5dr_domain *dmn,
+				   struct mlx5dr_action_dest *dests,
+				   u32 num_of_dests);
+
 struct mlx5dr_action *mlx5dr_action_create_drop(void);
 
 struct mlx5dr_action *mlx5dr_action_create_tag(u32 tag_value);
@@ -173,6 +183,11 @@ mlx5dr_action_create_dest_vport(struct mlx5dr_domain *domain,
 				u32 vport, u8 vhca_id_valid,
 				u16 vhca_id) { return NULL; }
 
+static inline struct mlx5dr_action *
+mlx5dr_action_create_mult_dest_tbl(struct mlx5dr_domain *dmn,
+				   struct mlx5dr_action_dest *dests,
+				   u32 num_of_dests)  { return NULL; }
+
 static inline struct mlx5dr_action *
 mlx5dr_action_create_drop(void) { return NULL; }
 
-- 
cgit v1.2.3-59-g8ed1b


From 7ee3f6d2486e25d96d2309da4d53fe10a58e2b63 Mon Sep 17 00:00:00 2001
From: Alex Vesker <valex@mellanox.com>
Date: Sun, 15 Dec 2019 21:50:32 +0200
Subject: net/mlx5: DR, Create multiple destination action from dr_create_fte

Until now it was possible to pass a packet to a single destination such
as vport or flow table. With the new support if multiple vports or multiple
tables are provided as destinations, fs_dr will create a multiple
destination table action, this action should replace other destination
actions provided to mlx5dr_create_rule.
Each vport destination can be provided with a reformat actions which
will be done before forwarding the packet to the vport.

Signed-off-by: Alex Vesker <valex@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../ethernet/mellanox/mlx5/core/steering/fs_dr.c   | 88 ++++++++++++++++++----
 1 file changed, 72 insertions(+), 16 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c
index e51262ec77bb..b43275cde8bf 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c
@@ -206,6 +206,12 @@ static struct mlx5dr_action *create_action_push_vlan(struct mlx5dr_domain *domai
 	return mlx5dr_action_create_push_vlan(domain, htonl(vlan_hdr));
 }
 
+static bool contain_vport_reformat_action(struct mlx5_flow_rule *dst)
+{
+	return dst->dest_attr.type == MLX5_FLOW_DESTINATION_TYPE_VPORT &&
+		dst->dest_attr.vport.flags & MLX5_FLOW_DEST_VPORT_REFORMAT_ID;
+}
+
 #define MLX5_FLOW_CONTEXT_ACTION_MAX  20
 static int mlx5_cmd_dr_create_fte(struct mlx5_flow_root_namespace *ns,
 				  struct mlx5_flow_table *ft,
@@ -213,7 +219,7 @@ static int mlx5_cmd_dr_create_fte(struct mlx5_flow_root_namespace *ns,
 				  struct fs_fte *fte)
 {
 	struct mlx5dr_domain *domain = ns->fs_dr_domain.dr_domain;
-	struct mlx5dr_action *term_action = NULL;
+	struct mlx5dr_action_dest *term_actions;
 	struct mlx5dr_match_parameters params;
 	struct mlx5_core_dev *dev = ns->dev;
 	struct mlx5dr_action **fs_dr_actions;
@@ -223,6 +229,7 @@ static int mlx5_cmd_dr_create_fte(struct mlx5_flow_root_namespace *ns,
 	struct mlx5dr_rule *rule;
 	struct mlx5_flow_rule *dst;
 	int fs_dr_num_actions = 0;
+	int num_term_actions = 0;
 	int num_actions = 0;
 	size_t match_sz;
 	int err = 0;
@@ -233,18 +240,38 @@ static int mlx5_cmd_dr_create_fte(struct mlx5_flow_root_namespace *ns,
 
 	actions = kcalloc(MLX5_FLOW_CONTEXT_ACTION_MAX, sizeof(*actions),
 			  GFP_KERNEL);
-	if (!actions)
-		return -ENOMEM;
+	if (!actions) {
+		err = -ENOMEM;
+		goto out_err;
+	}
 
 	fs_dr_actions = kcalloc(MLX5_FLOW_CONTEXT_ACTION_MAX,
 				sizeof(*fs_dr_actions), GFP_KERNEL);
 	if (!fs_dr_actions) {
-		kfree(actions);
-		return -ENOMEM;
+		err = -ENOMEM;
+		goto free_actions_alloc;
+	}
+
+	term_actions = kcalloc(MLX5_FLOW_CONTEXT_ACTION_MAX,
+			       sizeof(*term_actions), GFP_KERNEL);
+	if (!term_actions) {
+		err = -ENOMEM;
+		goto free_fs_dr_actions_alloc;
 	}
 
 	match_sz = sizeof(fte->val);
 
+	/* Drop reformat action bit if destination vport set with reformat */
+	if (fte->action.action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) {
+		list_for_each_entry(dst, &fte->node.children, node.list) {
+			if (!contain_vport_reformat_action(dst))
+				continue;
+
+			fte->action.action &= ~MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT;
+			break;
+		}
+	}
+
 	/* The order of the actions are must to be keep, only the following
 	 * order is supported by SW steering:
 	 * TX: push vlan -> modify header -> encap
@@ -335,7 +362,7 @@ static int mlx5_cmd_dr_create_fte(struct mlx5_flow_root_namespace *ns,
 			goto free_actions;
 		}
 		fs_dr_actions[fs_dr_num_actions++] = tmp_action;
-		term_action = tmp_action;
+		term_actions[num_term_actions++].dest = tmp_action;
 	}
 
 	if (fte->flow_context.flow_tag) {
@@ -354,7 +381,8 @@ static int mlx5_cmd_dr_create_fte(struct mlx5_flow_root_namespace *ns,
 			enum mlx5_flow_destination_type type = dst->dest_attr.type;
 			u32 id;
 
-			if (num_actions == MLX5_FLOW_CONTEXT_ACTION_MAX) {
+			if (num_actions == MLX5_FLOW_CONTEXT_ACTION_MAX ||
+			    num_term_actions >= MLX5_FLOW_CONTEXT_ACTION_MAX) {
 				err = -ENOSPC;
 				goto free_actions;
 			}
@@ -379,7 +407,7 @@ static int mlx5_cmd_dr_create_fte(struct mlx5_flow_root_namespace *ns,
 					goto free_actions;
 				}
 				fs_dr_actions[fs_dr_num_actions++] = tmp_action;
-				term_action = tmp_action;
+				term_actions[num_term_actions++].dest = tmp_action;
 				break;
 			case MLX5_FLOW_DESTINATION_TYPE_VPORT:
 				tmp_action = create_vport_action(domain, dst);
@@ -388,7 +416,14 @@ static int mlx5_cmd_dr_create_fte(struct mlx5_flow_root_namespace *ns,
 					goto free_actions;
 				}
 				fs_dr_actions[fs_dr_num_actions++] = tmp_action;
-				term_action = tmp_action;
+				term_actions[num_term_actions].dest = tmp_action;
+
+				if (dst->dest_attr.vport.flags &
+				    MLX5_FLOW_DEST_VPORT_REFORMAT_ID)
+					term_actions[num_term_actions].reformat =
+						dst->dest_attr.vport.pkt_reformat->action.dr_action;
+
+				num_term_actions++;
 				break;
 			default:
 				err = -EOPNOTSUPP;
@@ -399,9 +434,22 @@ static int mlx5_cmd_dr_create_fte(struct mlx5_flow_root_namespace *ns,
 
 	params.match_sz = match_sz;
 	params.match_buf = (u64 *)fte->val;
-
-	if (term_action)
-		actions[num_actions++] = term_action;
+	if (num_term_actions == 1) {
+		if (term_actions->reformat)
+			actions[num_actions++] = term_actions->reformat;
+
+		actions[num_actions++] = term_actions->dest;
+	} else if (num_term_actions > 1) {
+		tmp_action = mlx5dr_action_create_mult_dest_tbl(domain,
+								term_actions,
+								num_term_actions);
+		if (!tmp_action) {
+			err = -EOPNOTSUPP;
+			goto free_actions;
+		}
+		fs_dr_actions[fs_dr_num_actions++] = tmp_action;
+		actions[num_actions++] = tmp_action;
+	}
 
 	rule = mlx5dr_rule_create(group->fs_dr_matcher.dr_matcher,
 				  &params,
@@ -412,7 +460,9 @@ static int mlx5_cmd_dr_create_fte(struct mlx5_flow_root_namespace *ns,
 		goto free_actions;
 	}
 
+	kfree(term_actions);
 	kfree(actions);
+
 	fte->fs_dr_rule.dr_rule = rule;
 	fte->fs_dr_rule.num_actions = fs_dr_num_actions;
 	fte->fs_dr_rule.dr_actions = fs_dr_actions;
@@ -420,13 +470,18 @@ static int mlx5_cmd_dr_create_fte(struct mlx5_flow_root_namespace *ns,
 	return 0;
 
 free_actions:
-	for (i = 0; i < fs_dr_num_actions; i++)
+	/* Free in reverse order to handle action dependencies */
+	for (i = fs_dr_num_actions - 1; i >= 0; i--)
 		if (!IS_ERR_OR_NULL(fs_dr_actions[i]))
 			mlx5dr_action_destroy(fs_dr_actions[i]);
 
-	mlx5_core_err(dev, "Failed to create dr rule err(%d)\n", err);
-	kfree(actions);
+	kfree(term_actions);
+free_fs_dr_actions_alloc:
 	kfree(fs_dr_actions);
+free_actions_alloc:
+	kfree(actions);
+out_err:
+	mlx5_core_err(dev, "Failed to create dr rule err(%d)\n", err);
 	return err;
 }
 
@@ -533,7 +588,8 @@ static int mlx5_cmd_dr_delete_fte(struct mlx5_flow_root_namespace *ns,
 	if (err)
 		return err;
 
-	for (i = 0; i < rule->num_actions; i++)
+	/* Free in reverse order to handle action dependencies */
+	for (i = rule->num_actions - 1; i >= 0; i--)
 		if (!IS_ERR_OR_NULL(rule->dr_actions[i]))
 			mlx5dr_action_destroy(rule->dr_actions[i]);
 
-- 
cgit v1.2.3-59-g8ed1b


From 3daca28f15f5e70ae54859abd66dcdb6be31ba91 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <snelson@pensando.io>
Date: Mon, 6 Jan 2020 19:43:46 -0800
Subject: ionic: drop use of subdevice tags

The subdevice concept is not being used in the driver, so
drop the references to it.

Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/pensando/ionic/ionic.h | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/net/ethernet/pensando/ionic/ionic.h b/drivers/net/ethernet/pensando/ionic/ionic.h
index ca0b21911ad6..bb106a32f416 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic.h
+++ b/drivers/net/ethernet/pensando/ionic/ionic.h
@@ -19,10 +19,6 @@ struct ionic_lif;
 #define PCI_DEVICE_ID_PENSANDO_IONIC_ETH_PF	0x1002
 #define PCI_DEVICE_ID_PENSANDO_IONIC_ETH_VF	0x1003
 
-#define IONIC_SUBDEV_ID_NAPLES_25	0x4000
-#define IONIC_SUBDEV_ID_NAPLES_100_4	0x4001
-#define IONIC_SUBDEV_ID_NAPLES_100_8	0x4002
-
 #define DEVCMD_TIMEOUT  10
 
 struct ionic_vf {
-- 
cgit v1.2.3-59-g8ed1b


From 24cfa8c76278d48fe521a363b74e36861b4462d6 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <snelson@pensando.io>
Date: Mon, 6 Jan 2020 19:43:47 -0800
Subject: ionic: add Rx dropped packet counter

Add a counter for packets dropped by the driver, typically
for bad size or a receive error seen by the device.

Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/pensando/ionic/ionic_lif.h   |  1 +
 drivers/net/ethernet/pensando/ionic/ionic_stats.c |  1 +
 drivers/net/ethernet/pensando/ionic/ionic_txrx.c  | 12 +++++++++---
 3 files changed, 11 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.h b/drivers/net/ethernet/pensando/ionic/ionic_lif.h
index a55fd1f8c31b..9c5a7dd45f9d 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_lif.h
+++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.h
@@ -37,6 +37,7 @@ struct ionic_rx_stats {
 	u64 csum_complete;
 	u64 csum_error;
 	u64 buffers_posted;
+	u64 dropped;
 };
 
 #define IONIC_QCQ_F_INITED		BIT(0)
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_stats.c b/drivers/net/ethernet/pensando/ionic/ionic_stats.c
index 03916b6d47f2..a1e9796a660a 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_stats.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_stats.c
@@ -39,6 +39,7 @@ static const struct ionic_stat_desc ionic_rx_stats_desc[] = {
 	IONIC_RX_STAT_DESC(csum_none),
 	IONIC_RX_STAT_DESC(csum_complete),
 	IONIC_RX_STAT_DESC(csum_error),
+	IONIC_RX_STAT_DESC(dropped),
 };
 
 static const struct ionic_stat_desc ionic_txq_stats_desc[] = {
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c
index 97e79949b359..a009bbe9f9be 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c
@@ -152,12 +152,16 @@ static void ionic_rx_clean(struct ionic_queue *q, struct ionic_desc_info *desc_i
 	stats = q_to_rx_stats(q);
 	netdev = q->lif->netdev;
 
-	if (comp->status)
+	if (comp->status) {
+		stats->dropped++;
 		return;
+	}
 
 	/* no packet processing while resetting */
-	if (unlikely(test_bit(IONIC_LIF_QUEUE_RESET, q->lif->state)))
+	if (unlikely(test_bit(IONIC_LIF_QUEUE_RESET, q->lif->state))) {
+		stats->dropped++;
 		return;
+	}
 
 	stats->pkts++;
 	stats->bytes += le16_to_cpu(comp->len);
@@ -167,8 +171,10 @@ static void ionic_rx_clean(struct ionic_queue *q, struct ionic_desc_info *desc_i
 	else
 		skb = ionic_rx_frags(q, desc_info, cq_info);
 
-	if (unlikely(!skb))
+	if (unlikely(!skb)) {
+		stats->dropped++;
 		return;
+	}
 
 	skb_record_rx_queue(skb, q->index);
 
-- 
cgit v1.2.3-59-g8ed1b


From c37d6e3f25cf95cdefaa51eaad699382f34cdd0c Mon Sep 17 00:00:00 2001
From: Shannon Nelson <snelson@pensando.io>
Date: Mon, 6 Jan 2020 19:43:48 -0800
Subject: ionic: restrict received packets to mtu size

Make sure the NIC drops packets that are larger than the
specified MTU.

The front end of the NIC will accept packets larger than MTU and
will copy all the data it can to fill up the driver's posted
buffers - if the buffers are not long enough the packet will
then get dropped.  With the Rx SG buffers allocagted as full
pages, we are currently setting up more space than MTU size
available and end up receiving some packets that are larger
than MTU, up to the size of buffers posted.  To be sure the
NIC doesn't waste our time with oversized packets we need to
lie a little in the SG descriptor about how long is the last
SG element.

At dealloc time, we know the allocation was a page, so the
deallocation doesn't care about what length we put in the
descriptor.

Signed-off-by: Shannon Nelson <snelson@pensando.io>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/pensando/ionic/ionic_txrx.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c
index a009bbe9f9be..e452f4242ba0 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c
@@ -343,6 +343,8 @@ void ionic_rx_fill(struct ionic_queue *q)
 	struct ionic_rxq_sg_desc *sg_desc;
 	struct ionic_rxq_sg_elem *sg_elem;
 	struct ionic_rxq_desc *desc;
+	unsigned int remain_len;
+	unsigned int seg_len;
 	unsigned int nfrags;
 	bool ring_doorbell;
 	unsigned int i, j;
@@ -352,6 +354,7 @@ void ionic_rx_fill(struct ionic_queue *q)
 	nfrags = round_up(len, PAGE_SIZE) / PAGE_SIZE;
 
 	for (i = ionic_q_space_avail(q); i; i--) {
+		remain_len = len;
 		desc_info = q->head;
 		desc = desc_info->desc;
 		sg_desc = desc_info->sg_desc;
@@ -375,7 +378,9 @@ void ionic_rx_fill(struct ionic_queue *q)
 			return;
 		}
 		desc->addr = cpu_to_le64(page_info->dma_addr);
-		desc->len = cpu_to_le16(PAGE_SIZE);
+		seg_len = min_t(unsigned int, PAGE_SIZE, len);
+		desc->len = cpu_to_le16(seg_len);
+		remain_len -= seg_len;
 		page_info++;
 
 		/* fill sg descriptors - pages[1..n] */
@@ -391,7 +396,9 @@ void ionic_rx_fill(struct ionic_queue *q)
 				return;
 			}
 			sg_elem->addr = cpu_to_le64(page_info->dma_addr);
-			sg_elem->len = cpu_to_le16(PAGE_SIZE);
+			seg_len = min_t(unsigned int, PAGE_SIZE, remain_len);
+			sg_elem->len = cpu_to_le16(seg_len);
+			remain_len -= seg_len;
 			page_info++;
 		}
 
-- 
cgit v1.2.3-59-g8ed1b


From 6be1a5ce1b39b43294a8eda9ba5254792decb15e Mon Sep 17 00:00:00 2001
From: Shannon Nelson <snelson@pensando.io>
Date: Mon, 6 Jan 2020 19:43:49 -0800
Subject: ionic: clear compiler warning on hb use before set

Build checks have pointed out that 'hb' can theoretically
be used before set, so let's initialize it and get rid
of the compiler complaint.

Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/pensando/ionic/ionic_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/pensando/ionic/ionic_main.c b/drivers/net/ethernet/pensando/ionic/ionic_main.c
index 837b85f2fed9..a8e3fb73b465 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_main.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_main.c
@@ -330,9 +330,9 @@ int ionic_dev_cmd_wait(struct ionic *ionic, unsigned long max_seconds)
 	unsigned long max_wait;
 	unsigned long duration;
 	int opcode;
+	int hb = 0;
 	int done;
 	int err;
-	int hb;
 
 	WARN_ON(in_interrupt());
 
-- 
cgit v1.2.3-59-g8ed1b


From e64dec834ed2aaa2e32d823cba2dd65ac53f9dc3 Mon Sep 17 00:00:00 2001
From: Chen Zhou <chenzhou10@huawei.com>
Date: Tue, 7 Jan 2020 17:28:55 +0800
Subject: net: ch9200: use __func__ in debug message

Use __func__ to print the function name instead of hard coded string.

Signed-off-by: Chen Zhou <chenzhou10@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/usb/ch9200.c | 22 +++++++++++-----------
 1 file changed, 11 insertions(+), 11 deletions(-)

diff --git a/drivers/net/usb/ch9200.c b/drivers/net/usb/ch9200.c
index 9df3c1ffff35..3c2dc7474d4a 100644
--- a/drivers/net/usb/ch9200.c
+++ b/drivers/net/usb/ch9200.c
@@ -111,8 +111,8 @@ static int control_read(struct usbnet *dev,
 		request_type = (USB_DIR_IN | USB_TYPE_VENDOR |
 				USB_RECIP_DEVICE);
 
-	netdev_dbg(dev->net, "Control_read() index=0x%02x size=%d\n",
-		   index, size);
+	netdev_dbg(dev->net, "%s() index=0x%02x size=%d\n",
+		   __func__, index, size);
 
 	buf = kmalloc(size, GFP_KERNEL);
 	if (!buf) {
@@ -151,8 +151,8 @@ static int control_write(struct usbnet *dev, unsigned char request,
 		request_type = (USB_DIR_OUT | USB_TYPE_VENDOR |
 				USB_RECIP_DEVICE);
 
-	netdev_dbg(dev->net, "Control_write() index=0x%02x size=%d\n",
-		   index, size);
+	netdev_dbg(dev->net, "%s() index=0x%02x size=%d\n",
+		   __func__, index, size);
 
 	if (data) {
 		buf = kmemdup(data, size, GFP_KERNEL);
@@ -181,8 +181,8 @@ static int ch9200_mdio_read(struct net_device *netdev, int phy_id, int loc)
 	struct usbnet *dev = netdev_priv(netdev);
 	unsigned char buff[2];
 
-	netdev_dbg(netdev, "ch9200_mdio_read phy_id:%02x loc:%02x\n",
-		   phy_id, loc);
+	netdev_dbg(netdev, "%s phy_id:%02x loc:%02x\n",
+		   __func__, phy_id, loc);
 
 	if (phy_id != 0)
 		return -ENODEV;
@@ -199,8 +199,8 @@ static void ch9200_mdio_write(struct net_device *netdev,
 	struct usbnet *dev = netdev_priv(netdev);
 	unsigned char buff[2];
 
-	netdev_dbg(netdev, "ch9200_mdio_write() phy_id=%02x loc:%02x\n",
-		   phy_id, loc);
+	netdev_dbg(netdev, "%s() phy_id=%02x loc:%02x\n",
+		   __func__, phy_id, loc);
 
 	if (phy_id != 0)
 		return;
@@ -219,8 +219,8 @@ static int ch9200_link_reset(struct usbnet *dev)
 	mii_check_media(&dev->mii, 1, 1);
 	mii_ethtool_gset(&dev->mii, &ecmd);
 
-	netdev_dbg(dev->net, "link_reset() speed:%d duplex:%d\n",
-		   ecmd.speed, ecmd.duplex);
+	netdev_dbg(dev->net, "%s() speed:%d duplex:%d\n",
+		   __func__, ecmd.speed, ecmd.duplex);
 
 	return 0;
 }
@@ -309,7 +309,7 @@ static int get_mac_address(struct usbnet *dev, unsigned char *data)
 	unsigned char mac_addr[0x06];
 	int rd_mac_len = 0;
 
-	netdev_dbg(dev->net, "get_mac_address:\n\tusbnet VID:%0x PID:%0x\n",
+	netdev_dbg(dev->net, "%s:\n\tusbnet VID:%0x PID:%0x\n", __func__,
 		   le16_to_cpu(dev->udev->descriptor.idVendor),
 		   le16_to_cpu(dev->udev->descriptor.idProduct));
 
-- 
cgit v1.2.3-59-g8ed1b


From 195234b8856e590624d2a15ff0258366fc81cfe2 Mon Sep 17 00:00:00 2001
From: Chen Zhou <chenzhou10@huawei.com>
Date: Tue, 7 Jan 2020 17:28:56 +0800
Subject: net: ch9200: remove unnecessary return

The return is not needed, remove it.

Signed-off-by: Chen Zhou <chenzhou10@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/usb/ch9200.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/usb/ch9200.c b/drivers/net/usb/ch9200.c
index 3c2dc7474d4a..d7f3b70d5477 100644
--- a/drivers/net/usb/ch9200.c
+++ b/drivers/net/usb/ch9200.c
@@ -130,8 +130,6 @@ static int control_read(struct usbnet *dev,
 		err = -EINVAL;
 	kfree(buf);
 
-	return err;
-
 err_out:
 	return err;
 }
-- 
cgit v1.2.3-59-g8ed1b


From c68d724826cd049f6408fa52bb0130641457aba1 Mon Sep 17 00:00:00 2001
From: Chen Zhou <chenzhou10@huawei.com>
Date: Tue, 7 Jan 2020 17:33:46 +0800
Subject: drivers: net: cisco_hdlc: use __func__ in debug message

Use __func__ to print the function name instead of hard coded string.
BTW, replace printk(KERN_DEBUG, ...) with netdev_dbg.

Signed-off-by: Chen Zhou <chenzhou10@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/wan/hdlc_cisco.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wan/hdlc_cisco.c b/drivers/net/wan/hdlc_cisco.c
index a030f5aa6b95..d8cba3625c18 100644
--- a/drivers/net/wan/hdlc_cisco.c
+++ b/drivers/net/wan/hdlc_cisco.c
@@ -75,7 +75,7 @@ static int cisco_hard_header(struct sk_buff *skb, struct net_device *dev,
 {
 	struct hdlc_header *data;
 #ifdef DEBUG_HARD_HEADER
-	printk(KERN_DEBUG "%s: cisco_hard_header called\n", dev->name);
+	netdev_dbg(dev, "%s called\n", __func__);
 #endif
 
 	skb_push(skb, sizeof(struct hdlc_header));
@@ -101,7 +101,7 @@ static void cisco_keepalive_send(struct net_device *dev, u32 type,
 	skb = dev_alloc_skb(sizeof(struct hdlc_header) +
 			    sizeof(struct cisco_packet));
 	if (!skb) {
-		netdev_warn(dev, "Memory squeeze on cisco_keepalive_send()\n");
+		netdev_warn(dev, "Memory squeeze on %s()\n", __func__);
 		return;
 	}
 	skb_reserve(skb, 4);
-- 
cgit v1.2.3-59-g8ed1b


From 1501125460faae8128519e90b84ed69fd15668fb Mon Sep 17 00:00:00 2001
From: Jose Abreu <Jose.Abreu@synopsys.com>
Date: Tue, 7 Jan 2020 11:37:18 +0100
Subject: MAINTAINERS: Add stmmac Ethernet driver documentation entry

Add the missing entry for the file that documents the stmicro Ethernet
driver stmmac.

Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 MAINTAINERS | 1 +
 1 file changed, 1 insertion(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index 34de47832883..66a2e5e07117 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -15777,6 +15777,7 @@ M:	Jose Abreu <joabreu@synopsys.com>
 L:	netdev@vger.kernel.org
 W:	http://www.stlinux.com
 S:	Supported
+F:	Documentation/networking/device_drivers/stmicro/
 F:	drivers/net/ethernet/stmicro/stmmac/
 
 SUN3/3X
-- 
cgit v1.2.3-59-g8ed1b


From 2ffebffbe7dc8fc966174eedee8fa76cb6143820 Mon Sep 17 00:00:00 2001
From: Jose Abreu <Jose.Abreu@synopsys.com>
Date: Tue, 7 Jan 2020 11:37:19 +0100
Subject: Documentation: networking: Convert stmmac documentation to RST format

Convert the documentation of the driver to RST format and delete the old
txt and old information that no longer applies.

Also, add some new information.

Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../networking/device_drivers/stmicro/stmmac.rst   | 697 +++++++++++++++++++++
 .../networking/device_drivers/stmicro/stmmac.txt   | 401 ------------
 2 files changed, 697 insertions(+), 401 deletions(-)
 create mode 100644 Documentation/networking/device_drivers/stmicro/stmmac.rst
 delete mode 100644 Documentation/networking/device_drivers/stmicro/stmmac.txt

diff --git a/Documentation/networking/device_drivers/stmicro/stmmac.rst b/Documentation/networking/device_drivers/stmicro/stmmac.rst
new file mode 100644
index 000000000000..c34bab3d2df0
--- /dev/null
+++ b/Documentation/networking/device_drivers/stmicro/stmmac.rst
@@ -0,0 +1,697 @@
+.. SPDX-License-Identifier: GPL-2.0+
+
+==============================================================
+Linux Driver for the Synopsys(R) Ethernet Controllers "stmmac"
+==============================================================
+
+Authors: Giuseppe Cavallaro <peppe.cavallaro@st.com>,
+Alexandre Torgue <alexandre.torgue@st.com>, Jose Abreu <joabreu@synopsys.com>
+
+Contents
+========
+
+- In This Release
+- Feature List
+- Kernel Configuration
+- Command Line Parameters
+- Driver Information and Notes
+- Debug Information
+- Support
+
+In This Release
+===============
+
+This file describes the stmmac Linux Driver for all the Synopsys(R) Ethernet
+Controllers.
+
+Currently, this network device driver is for all STi embedded MAC/GMAC
+(i.e. 7xxx/5xxx SoCs), SPEAr (arm), Loongson1B (mips) and XILINX XC2V3000
+FF1152AMT0221 D1215994A VIRTEX FPGA board. The Synopsys Ethernet QoS 5.0 IPK
+is also supported.
+
+DesignWare(R) Cores Ethernet MAC 10/100/1000 Universal version 3.70a
+(and older) and DesignWare(R) Cores Ethernet Quality-of-Service version 4.0
+(and upper) have been used for developing this driver as well as
+DesignWare(R) Cores XGMAC - 10G Ethernet MAC.
+
+This driver supports both the platform bus and PCI.
+
+This driver includes support for the following Synopsys(R) DesignWare(R)
+Cores Ethernet Controllers and corresponding minimum and maximum versions:
+
++-------------------------------+--------------+--------------+--------------+
+| Controller Name               | Min. Version | Max. Version | Abbrev. Name |
++===============================+==============+==============+==============+
+| Ethernet MAC Universal        | N/A          | 3.73a        | GMAC         |
++-------------------------------+--------------+--------------+--------------+
+| Ethernet Quality-of-Service   | 4.00a        | N/A          | GMAC4+       |
++-------------------------------+--------------+--------------+--------------+
+| XGMAC - 10G Ethernet MAC      | 2.10a        | N/A          | XGMAC2+      |
++-------------------------------+--------------+--------------+--------------+
+
+For questions related to hardware requirements, refer to the documentation
+supplied with your Ethernet adapter. All hardware requirements listed apply
+to use with Linux.
+
+Feature List
+============
+
+The following features are available in this driver:
+ - GMII/MII/RGMII/SGMII/RMII/XGMII Interface
+ - Half-Duplex / Full-Duplex Operation
+ - Energy Efficient Ethernet (EEE)
+ - IEEE 802.3x PAUSE Packets (Flow Control)
+ - RMON/MIB Counters
+ - IEEE 1588 Timestamping (PTP)
+ - Pulse-Per-Second Output (PPS)
+ - MDIO Clause 22 / Clause 45 Interface
+ - MAC Loopback
+ - ARP Offloading
+ - Automatic CRC / PAD Insertion and Checking
+ - Checksum Offload for Received and Transmitted Packets
+ - Standard or Jumbo Ethernet Packets
+ - Source Address Insertion / Replacement
+ - VLAN TAG Insertion / Replacement / Deletion / Filtering (HASH and PERFECT)
+ - Programmable TX and RX Watchdog and Coalesce Settings
+ - Destination Address Filtering (PERFECT)
+ - HASH Filtering (Multicast)
+ - Layer 3 / Layer 4 Filtering
+ - Remote Wake-Up Detection
+ - Receive Side Scaling (RSS)
+ - Frame Preemption for TX and RX
+ - Programmable Burst Length, Threshold, Queue Size
+ - Multiple Queues (up to 8)
+ - Multiple Scheduling Algorithms (TX: WRR, DWRR, WFQ, SP, CBS, EST, TBS;
+   RX: WRR, SP)
+ - Flexible RX Parser
+ - TCP / UDP Segmentation Offload (TSO, USO)
+ - Split Header (SPH)
+ - Safety Features (ECC Protection, Data Parity Protection)
+ - Selftests using Ethtool
+
+Kernel Configuration
+====================
+
+The kernel configuration option is ``CONFIG_STMMAC_ETH``:
+ - ``CONFIG_STMMAC_PLATFORM``: is to enable the platform driver.
+ - ``CONFIG_STMMAC_PCI``: is to enable the pci driver.
+
+Command Line Parameters
+=======================
+
+If the driver is built as a module the following optional parameters are used
+by entering them on the command line with the modprobe command using this
+syntax (e.g. for PCI module)::
+
+    modprobe stmmac_pci [<option>=<VAL1>,<VAL2>,...]
+
+Driver parameters can be also passed in command line by using::
+
+    stmmaceth=watchdog:100,chain_mode=1
+
+The default value for each parameter is generally the recommended setting,
+unless otherwise noted.
+
+watchdog
+--------
+:Valid Range: 5000-None
+:Default Value: 5000
+
+This parameter overrides the transmit timeout in milliseconds.
+
+debug
+-----
+:Valid Range: 0-16 (0=none,...,16=all)
+:Default Value: 0
+
+This parameter adjusts the level of debug messages displayed in the system
+logs.
+
+phyaddr
+-------
+:Valid Range: 0-31
+:Default Value: -1
+
+This parameter overrides the physical address of the PHY device.
+
+flow_ctrl
+---------
+:Valid Range: 0-3 (0=off,1=rx,2=tx,3=rx/tx)
+:Default Value: 3
+
+This parameter changes the default Flow Control ability.
+
+pause
+-----
+:Valid Range: 0-65535
+:Default Value: 65535
+
+This parameter changes the default Flow Control Pause time.
+
+tc
+--
+:Valid Range: 64-256
+:Default Value: 64
+
+This parameter changes the default HW FIFO Threshold control value.
+
+buf_sz
+------
+:Valid Range: 1536-16384
+:Default Value: 1536
+
+This parameter changes the default RX DMA packet buffer size.
+
+eee_timer
+---------
+:Valid Range: 0-None
+:Default Value: 1000
+
+This parameter changes the default LPI TX Expiration time in milliseconds.
+
+chain_mode
+----------
+:Valid Range: 0-1 (0=off,1=on)
+:Default Value: 0
+
+This parameter changes the default mode of operation from Ring Mode to
+Chain Mode.
+
+Driver Information and Notes
+============================
+
+Transmit Process
+----------------
+
+The xmit method is invoked when the kernel needs to transmit a packet; it sets
+the descriptors in the ring and informs the DMA engine that there is a packet
+ready to be transmitted.
+
+By default, the driver sets the ``NETIF_F_SG`` bit in the features field of
+the ``net_device`` structure, enabling the scatter-gather feature. This is
+true on chips and configurations where the checksum can be done in hardware.
+
+Once the controller has finished transmitting the packet, timer will be
+scheduled to release the transmit resources.
+
+Receive Process
+---------------
+
+When one or more packets are received, an interrupt happens. The interrupts
+are not queued, so the driver has to scan all the descriptors in the ring
+during the receive process.
+
+This is based on NAPI, so the interrupt handler signals only if there is work
+to be done, and it exits. Then the poll method will be scheduled at some
+future point.
+
+The incoming packets are stored, by the DMA, in a list of pre-allocated socket
+buffers in order to avoid the memcpy (zero-copy).
+
+Interrupt Mitigation
+--------------------
+
+The driver is able to mitigate the number of its DMA interrupts using NAPI for
+the reception on chips older than the 3.50. New chips have an HW RX Watchdog
+used for this mitigation.
+
+Mitigation parameters can be tuned by ethtool.
+
+WoL
+---
+
+Wake up on Lan feature through Magic and Unicast frames are supported for the
+GMAC, GMAC4/5 and XGMAC core.
+
+DMA Descriptors
+---------------
+
+Driver handles both normal and alternate descriptors. The latter has been only
+tested on DesignWare(R) Cores Ethernet MAC Universal version 3.41a and later.
+
+stmmac supports DMA descriptor to operate both in dual buffer (RING) and
+linked-list(CHAINED) mode. In RING each descriptor points to two data buffer
+pointers whereas in CHAINED mode they point to only one data buffer pointer.
+RING mode is the default.
+
+In CHAINED mode each descriptor will have pointer to next descriptor in the
+list, hence creating the explicit chaining in the descriptor itself, whereas
+such explicit chaining is not possible in RING mode.
+
+Extended Descriptors
+--------------------
+
+The extended descriptors give us information about the Ethernet payload when
+it is carrying PTP packets or TCP/UDP/ICMP over IP. These are not available on
+GMAC Synopsys(R) chips older than the 3.50. At probe time the driver will
+decide if these can be actually used. This support also is mandatory for PTPv2
+because the extra descriptors are used for saving the hardware timestamps and
+Extended Status.
+
+Ethtool Support
+---------------
+
+Ethtool is supported. For example, driver statistics (including RMON),
+internal errors can be taken using::
+
+    ethtool -S ethX
+
+Ethtool selftests are also supported. This allows to do some early sanity
+checks to the HW using MAC and PHY loopback mechanisms::
+
+    ethtool -t ethX
+
+Jumbo and Segmentation Offloading
+---------------------------------
+
+Jumbo frames are supported and tested for the GMAC. The GSO has been also
+added but it's performed in software. LRO is not supported.
+
+TSO Support
+-----------
+
+TSO (TCP Segmentation Offload) feature is supported by GMAC > 4.x and XGMAC
+chip family. When a packet is sent through TCP protocol, the TCP stack ensures
+that the SKB provided to the low level driver (stmmac in our case) matches
+with the maximum frame len (IP header + TCP header + payload <= 1500 bytes
+(for MTU set to 1500)). It means that if an application using TCP want to send
+a packet which will have a length (after adding headers) > 1514 the packet
+will be split in several TCP packets: The data payload is split and headers
+(TCP/IP ..) are added. It is done by software.
+
+When TSO is enabled, the TCP stack doesn't care about the maximum frame length
+and provide SKB packet to stmmac as it is. The GMAC IP will have to perform
+the segmentation by it self to match with maximum frame length.
+
+This feature can be enabled in device tree through ``snps,tso`` entry.
+
+Energy Efficient Ethernet
+-------------------------
+
+Energy Efficient Ethernet (EEE) enables IEEE 802.3 MAC sublayer along with a
+family of Physical layer to operate in the Low Power Idle (LPI) mode. The EEE
+mode supports the IEEE 802.3 MAC operation at 100Mbps, 1000Mbps and 1Gbps.
+
+The LPI mode allows power saving by switching off parts of the communication
+device functionality when there is no data to be transmitted & received.
+The system on both the side of the link can disable some functionalities and
+save power during the period of low-link utilization. The MAC controls whether
+the system should enter or exit the LPI mode and communicate this to PHY.
+
+As soon as the interface is opened, the driver verifies if the EEE can be
+supported. This is done by looking at both the DMA HW capability register and
+the PHY devices MCD registers.
+
+To enter in TX LPI mode the driver needs to have a software timer that enable
+and disable the LPI mode when there is nothing to be transmitted.
+
+Precision Time Protocol (PTP)
+-----------------------------
+
+The driver supports the IEEE 1588-2002, Precision Time Protocol (PTP), which
+enables precise synchronization of clocks in measurement and control systems
+implemented with technologies such as network communication.
+
+In addition to the basic timestamp features mentioned in IEEE 1588-2002
+Timestamps, new GMAC cores support the advanced timestamp features.
+IEEE 1588-2008 can be enabled when configuring the Kernel.
+
+SGMII/RGMII Support
+-------------------
+
+New GMAC devices provide own way to manage RGMII/SGMII. This information is
+available at run-time by looking at the HW capability register. This means
+that the stmmac can manage auto-negotiation and link status w/o using the
+PHYLIB stuff. In fact, the HW provides a subset of extended registers to
+restart the ANE, verify Full/Half duplex mode and Speed. Thanks to these
+registers, it is possible to look at the Auto-negotiated Link Parter Ability.
+
+Physical
+--------
+
+The driver is compatible with Physical Abstraction Layer to be connected with
+PHY and GPHY devices.
+
+Platform Information
+--------------------
+
+Several information can be passed through the platform and device-tree.
+
+::
+
+    struct plat_stmmacenet_data {
+
+1) Bus identifier::
+
+        int bus_id;
+
+2) PHY Physical Address. If set to -1 the driver will pick the first PHY it
+finds::
+
+        int phy_addr;
+
+3) PHY Device Interface::
+
+        int interface;
+
+4) Specific platform fields for the MDIO bus::
+
+        struct stmmac_mdio_bus_data *mdio_bus_data;
+
+5) Internal DMA parameters::
+
+        struct stmmac_dma_cfg *dma_cfg;
+
+6) Fixed CSR Clock Range selection::
+
+        int clk_csr;
+
+7) HW uses the GMAC core::
+
+        int has_gmac;
+
+8) If set the MAC will use Enhanced Descriptors::
+
+        int enh_desc;
+
+9) Core is able to perform TX Checksum and/or RX Checksum in HW::
+
+        int tx_coe;
+        int rx_coe;
+
+11) Some HWs are not able to perform the csum in HW for over-sized frames due
+to limited buffer sizes. Setting this flag the csum will be done in SW on
+JUMBO frames::
+
+        int bugged_jumbo;
+
+12) Core has the embedded power module::
+
+        int pmt;
+
+13) Force DMA to use the Store and Forward mode or Threshold mode::
+
+        int force_sf_dma_mode;
+        int force_thresh_dma_mode;
+
+15) Force to disable the RX Watchdog feature and switch to NAPI mode::
+
+        int riwt_off;
+
+16) Limit the maximum operating speed and MTU::
+
+        int max_speed;
+        int maxmtu;
+
+18) Number of Multicast/Unicast filters::
+
+        int multicast_filter_bins;
+        int unicast_filter_entries;
+
+20) Limit the maximum TX and RX FIFO size::
+
+        int tx_fifo_size;
+        int rx_fifo_size;
+
+21) Use the specified number of TX and RX Queues::
+
+        u32 rx_queues_to_use;
+        u32 tx_queues_to_use;
+
+22) Use the specified TX and RX scheduling algorithm::
+
+        u8 rx_sched_algorithm;
+        u8 tx_sched_algorithm;
+
+23) Internal TX and RX Queue parameters::
+
+        struct stmmac_rxq_cfg rx_queues_cfg[MTL_MAX_RX_QUEUES];
+        struct stmmac_txq_cfg tx_queues_cfg[MTL_MAX_TX_QUEUES];
+
+24) This callback is used for modifying some syscfg registers (on ST SoCs)
+according to the link speed negotiated by the physical layer::
+
+        void (*fix_mac_speed)(void *priv, unsigned int speed);
+
+25) Callbacks used for calling a custom initialization; This is sometimes
+necessary on some platforms (e.g. ST boxes) where the HW needs to have set
+some PIO lines or system cfg registers. init/exit callbacks should not use
+or modify platform data::
+
+        int (*init)(struct platform_device *pdev, void *priv);
+        void (*exit)(struct platform_device *pdev, void *priv);
+
+26) Perform HW setup of the bus. For example, on some ST platforms this field
+is used to configure the AMBA bridge to generate more efficient STBus traffic::
+
+        struct mac_device_info *(*setup)(void *priv);
+        void *bsp_priv;
+
+27) Internal clocks and rates::
+
+        struct clk *stmmac_clk;
+        struct clk *pclk;
+        struct clk *clk_ptp_ref;
+        unsigned int clk_ptp_rate;
+        unsigned int clk_ref_rate;
+        s32 ptp_max_adj;
+
+28) Main reset::
+
+        struct reset_control *stmmac_rst;
+
+29) AXI Internal Parameters::
+
+        struct stmmac_axi *axi;
+
+30) HW uses GMAC>4 cores::
+
+        int has_gmac4;
+
+31) HW is sun8i based::
+
+        bool has_sun8i;
+
+32) Enables TSO feature::
+
+        bool tso_en;
+
+33) Enables Receive Side Scaling (RSS) feature::
+
+        int rss_en;
+
+34) MAC Port selection::
+
+        int mac_port_sel_speed;
+
+35) Enables TX LPI Clock Gating::
+
+        bool en_tx_lpi_clockgating;
+
+36) HW uses XGMAC>2.10 cores::
+
+        int has_xgmac;
+
+::
+
+    }
+
+For MDIO bus data, we have:
+
+::
+
+    struct stmmac_mdio_bus_data {
+
+1) PHY mask passed when MDIO bus is registered::
+
+        unsigned int phy_mask;
+
+2) List of IRQs, one per PHY::
+
+        int *irqs;
+
+3) If IRQs is NULL, use this for probed PHY::
+
+        int probed_phy_irq;
+
+4) Set to true if PHY needs reset::
+
+        bool needs_reset;
+
+::
+
+    }
+
+For DMA engine configuration, we have:
+
+::
+
+    struct stmmac_dma_cfg {
+
+1) Programmable Burst Length (TX and RX)::
+
+        int pbl;
+
+2) If set, DMA TX / RX will use this value rather than pbl::
+
+        int txpbl;
+        int rxpbl;
+
+3) Enable 8xPBL::
+
+        bool pblx8;
+
+4) Enable Fixed or Mixed burst::
+
+        int fixed_burst;
+        int mixed_burst;
+
+5) Enable Address Aligned Beats::
+
+        bool aal;
+
+6) Enable Enhanced Addressing (> 32 bits)::
+
+        bool eame;
+
+::
+
+    }
+
+For DMA AXI parameters, we have:
+
+::
+
+    struct stmmac_axi {
+
+1) Enable AXI LPI::
+
+        bool axi_lpi_en;
+        bool axi_xit_frm;
+
+2) Set AXI Write / Read maximum outstanding requests::
+
+        u32 axi_wr_osr_lmt;
+        u32 axi_rd_osr_lmt;
+
+3) Set AXI 4KB bursts::
+
+        bool axi_kbbe;
+
+4) Set AXI maximum burst length map::
+
+        u32 axi_blen[AXI_BLEN];
+
+5) Set AXI Fixed burst / mixed burst::
+
+        bool axi_fb;
+        bool axi_mb;
+
+6) Set AXI rebuild incrx mode::
+
+        bool axi_rb;
+
+::
+
+    }
+
+For the RX Queues configuration, we have:
+
+::
+
+    struct stmmac_rxq_cfg {
+
+1) Mode to use (DCB or AVB)::
+
+        u8 mode_to_use;
+
+2) DMA channel to use::
+
+        u32 chan;
+
+3) Packet routing, if applicable::
+
+        u8 pkt_route;
+
+4) Use priority routing, and priority to route::
+
+        bool use_prio;
+        u32 prio;
+
+::
+
+    }
+
+For the TX Queues configuration, we have:
+
+::
+
+    struct stmmac_txq_cfg {
+
+1) Queue weight in scheduler::
+
+        u32 weight;
+
+2) Mode to use (DCB or AVB)::
+
+        u8 mode_to_use;
+
+3) Credit Base Shaper Parameters::
+
+        u32 send_slope;
+        u32 idle_slope;
+        u32 high_credit;
+        u32 low_credit;
+
+4) Use priority scheduling, and priority::
+
+        bool use_prio;
+        u32 prio;
+
+::
+
+    }
+
+Device Tree Information
+-----------------------
+
+Please refer to the following document:
+Documentation/devicetree/bindings/net/snps,dwmac.yaml
+
+HW Capabilities
+---------------
+
+Note that, starting from new chips, where it is available the HW capability
+register, many configurations are discovered at run-time for example to
+understand if EEE, HW csum, PTP, enhanced descriptor etc are actually
+available. As strategy adopted in this driver, the information from the HW
+capability register can replace what has been passed from the platform.
+
+Debug Information
+=================
+
+The driver exports many information i.e. internal statistics, debug
+information, MAC and DMA registers etc.
+
+These can be read in several ways depending on the type of the information
+actually needed.
+
+For example a user can be use the ethtool support to get statistics: e.g.
+using: ``ethtool -S ethX`` (that shows the Management counters (MMC) if
+supported) or sees the MAC/DMA registers: e.g. using: ``ethtool -d ethX``
+
+Compiling the Kernel with ``CONFIG_DEBUG_FS`` the driver will export the
+following debugfs entries:
+
+ - ``descriptors_status``: To show the DMA TX/RX descriptor rings
+ - ``dma_cap``: To show the HW Capabilities
+
+Developer can also use the ``debug`` module parameter to get further debug
+information (please see: NETIF Msg Level).
+
+Support
+=======
+
+If an issue is identified with the released source code on a supported kernel
+with a supported adapter, email the specific information related to the
+issue to netdev@vger.kernel.org
diff --git a/Documentation/networking/device_drivers/stmicro/stmmac.txt b/Documentation/networking/device_drivers/stmicro/stmmac.txt
deleted file mode 100644
index 1ae979fd90d2..000000000000
--- a/Documentation/networking/device_drivers/stmicro/stmmac.txt
+++ /dev/null
@@ -1,401 +0,0 @@
-       STMicroelectronics 10/100/1000 Synopsys Ethernet driver
-
-Copyright (C) 2007-2015  STMicroelectronics Ltd
-Author: Giuseppe Cavallaro <peppe.cavallaro@st.com>
-
-This is the driver for the MAC 10/100/1000 on-chip Ethernet controllers
-(Synopsys IP blocks).
-
-Currently this network device driver is for all STi embedded MAC/GMAC
-(i.e. 7xxx/5xxx SoCs), SPEAr (arm), Loongson1B (mips) and XLINX XC2V3000
-FF1152AMT0221 D1215994A VIRTEX FPGA board.
-
-DWC Ether MAC 10/100/1000 Universal version 3.70a (and older) and DWC Ether
-MAC 10/100 Universal version 4.0 have been used for developing this driver.
-
-This driver supports both the platform bus and PCI.
-
-Please, for more information also visit: www.stlinux.com
-
-1) Kernel Configuration
-The kernel configuration option is STMMAC_ETH:
- Device Drivers ---> Network device support ---> Ethernet (1000 Mbit) --->
- STMicroelectronics 10/100/1000 Ethernet driver (STMMAC_ETH)
-
-CONFIG_STMMAC_PLATFORM: is to enable the platform driver.
-CONFIG_STMMAC_PCI: is to enable the pci driver.
-
-2) Driver parameters list:
-	debug: message level (0: no output, 16: all);
-	phyaddr: to manually provide the physical address to the PHY device;
-	buf_sz: DMA buffer size;
-	tc: control the HW FIFO threshold;
-	watchdog: transmit timeout (in milliseconds);
-	flow_ctrl: Flow control ability [on/off];
-	pause: Flow Control Pause Time;
-	eee_timer: tx EEE timer;
-	chain_mode: select chain mode instead of ring.
-
-3) Command line options
-Driver parameters can be also passed in command line by using:
-	stmmaceth=watchdog:100,chain_mode=1
-
-4) Driver information and notes
-
-4.1) Transmit process
-The xmit method is invoked when the kernel needs to transmit a packet; it sets
-the descriptors in the ring and informs the DMA engine, that there is a packet
-ready to be transmitted.
-By default, the driver sets the NETIF_F_SG bit in the features field of the
-net_device structure, enabling the scatter-gather feature. This is true on
-chips and configurations where the checksum can be done in hardware.
-Once the controller has finished transmitting the packet, timer will be
-scheduled to release the transmit resources.
-
-4.2) Receive process
-When one or more packets are received, an interrupt happens. The interrupts
-are not queued, so the driver has to scan all the descriptors in the ring during
-the receive process.
-This is based on NAPI, so the interrupt handler signals only if there is work
-to be done, and it exits.
-Then the poll method will be scheduled at some future point.
-The incoming packets are stored, by the DMA, in a list of pre-allocated socket
-buffers in order to avoid the memcpy (zero-copy).
-
-4.3) Interrupt mitigation
-The driver is able to mitigate the number of its DMA interrupts
-using NAPI for the reception on chips older than the 3.50.
-New chips have an HW RX-Watchdog used for this mitigation.
-Mitigation parameters can be tuned by ethtool.
-
-4.4) WOL
-Wake up on Lan feature through Magic and Unicast frames are supported for the
-GMAC core.
-
-4.5) DMA descriptors
-Driver handles both normal and alternate descriptors. The latter has been only
-tested on DWC Ether MAC 10/100/1000 Universal version 3.41a and later.
-
-STMMAC supports DMA descriptor to operate both in dual buffer (RING)
-and linked-list(CHAINED) mode. In RING each descriptor points to two
-data buffer pointers whereas in CHAINED mode they point to only one data
-buffer pointer. RING mode is the default.
-
-In CHAINED mode each descriptor will have pointer to next descriptor in
-the list, hence creating the explicit chaining in the descriptor itself,
-whereas such explicit chaining is not possible in RING mode.
-
-4.5.1) Extended descriptors
-The extended descriptors give us information about the Ethernet payload
-when it is carrying PTP packets or TCP/UDP/ICMP over IP.
-These are not available on GMAC Synopsys chips older than the 3.50.
-At probe time the driver will decide if these can be actually used.
-This support also is mandatory for PTPv2 because the extra descriptors
-are used for saving the hardware timestamps and Extended Status.
-
-4.6) Ethtool support
-Ethtool is supported.
-
-For example, driver statistics (including RMON), internal errors can be taken
-using:
-  # ethtool -S ethX
-command
-
-4.7) Jumbo and Segmentation Offloading
-Jumbo frames are supported and tested for the GMAC.
-The GSO has been also added but it's performed in software.
-LRO is not supported.
-
-4.8) Physical
-The driver is compatible with Physical Abstraction Layer to be connected with
-PHY and GPHY devices.
-
-4.9) Platform information
-Several information can be passed through the platform and device-tree.
-
-struct plat_stmmacenet_data {
-	char *phy_bus_name;
-	int bus_id;
-	int phy_addr;
-	int interface;
-	struct stmmac_mdio_bus_data *mdio_bus_data;
-	struct stmmac_dma_cfg *dma_cfg;
-	int clk_csr;
-	int has_gmac;
-	int enh_desc;
-	int tx_coe;
-	int rx_coe;
-	int bugged_jumbo;
-	int pmt;
-	int force_sf_dma_mode;
-	int force_thresh_dma_mode;
-	int riwt_off;
-	int max_speed;
-	int maxmtu;
-	void (*fix_mac_speed)(void *priv, unsigned int speed);
-	void (*bus_setup)(void __iomem *ioaddr);
-	int (*init)(struct platform_device *pdev, void *priv);
-	void (*exit)(struct platform_device *pdev, void *priv);
-	void *bsp_priv;
-	int has_gmac4;
-	bool tso_en;
-};
-
-Where:
- o phy_bus_name: phy bus name to attach to the stmmac.
- o bus_id: bus identifier.
- o phy_addr: the physical address can be passed from the platform.
-	    If it is set to -1 the driver will automatically
-	    detect it at run-time by probing all the 32 addresses.
- o interface: PHY device's interface.
- o mdio_bus_data: specific platform fields for the MDIO bus.
- o dma_cfg: internal DMA parameters
-   o pbl: the Programmable Burst Length is maximum number of beats to
-       be transferred in one DMA transaction.
-       GMAC also enables the 4xPBL by default. (8xPBL for GMAC 3.50 and newer)
-   o txpbl/rxpbl: GMAC and newer supports independent DMA pbl for tx/rx.
-   o pblx8: Enable 8xPBL (4xPBL for core rev < 3.50). Enabled by default.
-   o fixed_burst/mixed_burst/aal
- o clk_csr: fixed CSR Clock range selection.
- o has_gmac: uses the GMAC core.
- o enh_desc: if sets the MAC will use the enhanced descriptor structure.
- o tx_coe: core is able to perform the tx csum in HW.
- o rx_coe: the supports three check sum offloading engine types:
-	   type_1, type_2 (full csum) and no RX coe.
- o bugged_jumbo: some HWs are not able to perform the csum in HW for
-		over-sized frames due to limited buffer sizes.
-		Setting this flag the csum will be done in SW on
-		JUMBO frames.
- o pmt: core has the embedded power module (optional).
- o force_sf_dma_mode: force DMA to use the Store and Forward mode
-		     instead of the Threshold.
- o force_thresh_dma_mode: force DMA to use the Threshold mode other than
-		     the Store and Forward mode.
- o riwt_off: force to disable the RX watchdog feature and switch to NAPI mode.
- o fix_mac_speed: this callback is used for modifying some syscfg registers
-		 (on ST SoCs) according to the link speed negotiated by the
-		 physical layer .
- o bus_setup: perform HW setup of the bus. For example, on some ST platforms
-	     this field is used to configure the AMBA  bridge to generate more
-	     efficient STBus traffic.
- o init/exit: callbacks used for calling a custom initialization;
-	     this is sometime necessary on some platforms (e.g. ST boxes)
-	     where the HW needs to have set some PIO lines or system cfg
-	     registers.  init/exit callbacks should not use or modify
-	     platform data.
- o bsp_priv: another private pointer.
- o has_gmac4: uses GMAC4 core.
- o tso_en: Enables TSO (TCP Segmentation Offload) feature.
-
-For MDIO bus The we have:
-
- struct stmmac_mdio_bus_data {
-	int (*phy_reset)(void *priv);
-	unsigned int phy_mask;
-	int *irqs;
-	int probed_phy_irq;
- };
-
-Where:
- o phy_reset: hook to reset the phy device attached to the bus.
- o phy_mask: phy mask passed when register the MDIO bus within the driver.
- o irqs: list of IRQs, one per PHY.
- o probed_phy_irq: if irqs is NULL, use this for probed PHY.
-
-For DMA engine we have the following internal fields that should be
-tuned according to the HW capabilities.
-
-struct stmmac_dma_cfg {
-	int pbl;
-	int txpbl;
-	int rxpbl;
-	bool pblx8;
-	int fixed_burst;
-	int mixed_burst;
-	bool aal;
-};
-
-Where:
- o pbl: Programmable Burst Length (tx and rx)
- o txpbl: Transmit Programmable Burst Length. Only for GMAC and newer.
-	 If set, DMA tx will use this value rather than pbl.
- o rxpbl: Receive Programmable Burst Length. Only for GMAC and newer.
-	 If set, DMA rx will use this value rather than pbl.
- o pblx8: Enable 8xPBL (4xPBL for core rev < 3.50). Enabled by default.
- o fixed_burst: program the DMA to use the fixed burst mode
- o mixed_burst: program the DMA to use the mixed burst mode
- o aal: Address-Aligned Beats
-
----
-
-Below an example how the structures above are using on ST platforms.
-
- static struct plat_stmmacenet_data stxYYY_ethernet_platform_data = {
-	.has_gmac = 0,
-	.enh_desc = 0,
-	.fix_mac_speed = stxYYY_ethernet_fix_mac_speed,
-				|
-				|-> to write an internal syscfg
-				|   on this platform when the
-				|   link speed changes from 10 to
-				|   100 and viceversa
-	.init = &stmmac_claim_resource,
-				|
-				|-> On ST SoC this calls own "PAD"
-				|   manager framework to claim
-				|   all the resources necessary
-				|   (GPIO ...). The .custom_cfg field
-				|   is used to pass a custom config.
-};
-
-Below the usage of the stmmac_mdio_bus_data: on this SoC, in fact,
-there are two MAC cores: one MAC is for MDIO Bus/PHY emulation
-with fixed_link support.
-
-static struct stmmac_mdio_bus_data stmmac1_mdio_bus = {
-	.phy_reset = phy_reset;
-		|
-		|-> function to provide the phy_reset on this board
-	.phy_mask = 0,
-};
-
-static struct fixed_phy_status stmmac0_fixed_phy_status = {
-	.link = 1,
-	.speed = 100,
-	.duplex = 1,
-};
-
-During the board's device_init we can configure the first
-MAC for fixed_link by calling:
-  fixed_phy_add(PHY_POLL, 1, &stmmac0_fixed_phy_status);
-and the second one, with a real PHY device attached to the bus,
-by using the stmmac_mdio_bus_data structure (to provide the id, the
-reset procedure etc).
-
-Note that, starting from new chips, where it is available the HW capability
-register, many configurations are discovered at run-time for example to
-understand if EEE, HW csum, PTP, enhanced descriptor etc are actually
-available. As strategy adopted in this driver, the information from the HW
-capability register can replace what has been passed from the platform.
-
-4.10) Device-tree support.
-
-Please see the following document:
-	Documentation/devicetree/bindings/net/stmmac.txt
-
-4.11) This is a summary of the content of some relevant files:
- o stmmac_main.c: implements the main network device driver;
- o stmmac_mdio.c: provides MDIO functions;
- o stmmac_pci: this is the PCI driver;
- o stmmac_platform.c: this the platform driver (OF supported);
- o stmmac_ethtool.c: implements the ethtool support;
- o stmmac.h: private driver structure;
- o common.h: common definitions and VFTs;
- o mmc_core.c/mmc.h: Management MAC Counters;
- o stmmac_hwtstamp.c: HW timestamp support for PTP;
- o stmmac_ptp.c: PTP 1588 clock;
- o stmmac_pcs.h: Physical Coding Sublayer common implementation;
- o dwmac-<XXX>.c: these are for the platform glue-logic file; e.g. dwmac-sti.c
-   for STMicroelectronics SoCs.
-
-- GMAC 3.x
- o descs.h: descriptor structure definitions;
- o dwmac1000_core.c: dwmac GiGa core functions;
- o dwmac1000_dma.c: dma functions for the GMAC chip;
- o dwmac1000.h: specific header file for the dwmac GiGa;
- o dwmac100_core: dwmac 100 core code;
- o dwmac100_dma.c: dma functions for the dwmac 100 chip;
- o dwmac1000.h: specific header file for the MAC;
- o dwmac_lib.c: generic DMA functions;
- o enh_desc.c: functions for handling enhanced descriptors;
- o norm_desc.c: functions for handling normal descriptors;
- o chain_mode.c/ring_mode.c:: functions to manage RING/CHAINED modes;
-
-- GMAC4.x generation
- o dwmac4_core.c: dwmac GMAC4.x core functions;
- o dwmac4_desc.c: functions for handling GMAC4.x descriptors;
- o dwmac4_descs.h: descriptor definitions;
- o dwmac4_dma.c: dma functions for the GMAC4.x chip;
- o dwmac4_dma.h: dma definitions for the GMAC4.x chip;
- o dwmac4.h: core definitions for the GMAC4.x chip;
- o dwmac4_lib.c: generic GMAC4.x functions;
-
-4.12) TSO support (GMAC4.x)
-
-TSO (Tcp Segmentation Offload) feature is supported by GMAC 4.x chip family.
-When a packet is sent through TCP protocol, the TCP stack ensures that
-the SKB provided to the low level driver (stmmac in our case) matches with
-the maximum frame len (IP header + TCP header + payload <= 1500 bytes (for
-MTU set to 1500)). It means that if an application using TCP want to send a
-packet which will have a length (after adding headers) > 1514 the packet
-will be split in several TCP packets: The data payload is split and headers
-(TCP/IP ..) are added. It is done by software.
-
-When TSO is enabled, the TCP stack doesn't care about the maximum frame
-length and provide SKB packet to stmmac as it is. The GMAC IP will have to
-perform the segmentation by it self to match with maximum frame length.
-
-This feature can be enabled in device tree through "snps,tso" entry.
-
-5) Debug Information
-
-The driver exports many information i.e. internal statistics,
-debug information, MAC and DMA registers etc.
-
-These can be read in several ways depending on the
-type of the information actually needed.
-
-For example a user can be use the ethtool support
-to get statistics: e.g. using: ethtool -S ethX
-(that shows the Management counters (MMC) if supported)
-or sees the MAC/DMA registers: e.g. using: ethtool -d ethX
-
-Compiling the Kernel with CONFIG_DEBUG_FS the driver will export the following
-debugfs entries:
-
-/sys/kernel/debug/stmmaceth/descriptors_status
-  To show the DMA TX/RX descriptor rings
-
-Developer can also use the "debug" module parameter to get further debug
-information (please see: NETIF Msg Level).
-
-6) Energy Efficient Ethernet
-
-Energy Efficient Ethernet(EEE) enables IEEE 802.3 MAC sublayer along
-with a family of Physical layer to operate in the Low power Idle(LPI)
-mode. The EEE mode supports the IEEE 802.3 MAC operation at 100Mbps,
-1000Mbps & 10Gbps.
-
-The LPI mode allows power saving by switching off parts of the
-communication device functionality when there is no data to be
-transmitted & received. The system on both the side of the link can
-disable some functionalities & save power during the period of low-link
-utilization. The MAC controls whether the system should enter or exit
-the LPI mode & communicate this to PHY.
-
-As soon as the interface is opened, the driver verifies if the EEE can
-be supported. This is done by looking at both the DMA HW capability
-register and the PHY devices MCD registers.
-To enter in Tx LPI mode the driver needs to have a software timer
-that enable and disable the LPI mode when there is nothing to be
-transmitted.
-
-7) Precision Time Protocol (PTP)
-The driver supports the IEEE 1588-2002, Precision Time Protocol (PTP),
-which enables precise synchronization of clocks in measurement and
-control systems implemented with technologies such as network
-communication.
-
-In addition to the basic timestamp features mentioned in IEEE 1588-2002
-Timestamps, new GMAC cores support the advanced timestamp features.
-IEEE 1588-2008 that can be enabled when configure the Kernel.
-
-8) SGMII/RGMII support
-New GMAC devices provide own way to manage RGMII/SGMII.
-This information is available at run-time by looking at the
-HW capability register. This means that the stmmac can manage
-auto-negotiation and link status w/o using the PHYLIB stuff.
-In fact, the HW provides a subset of extended registers to
-restart the ANE, verify Full/Half duplex mode and Speed.
-Thanks to these registers, it is possible to look at the
-Auto-negotiated Link Parter Ability.
-- 
cgit v1.2.3-59-g8ed1b


From b053b28e93da37e6646aabd36a8ee19121fc53b0 Mon Sep 17 00:00:00 2001
From: Jose Abreu <Jose.Abreu@synopsys.com>
Date: Tue, 7 Jan 2020 11:37:20 +0100
Subject: Documentation: networking: Add stmmac to device drivers list

Add the stmmac RST file to the index.

Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/device_drivers/index.rst | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/networking/device_drivers/index.rst b/Documentation/networking/device_drivers/index.rst
index c1f7f75e5fd9..4bc6ff29976a 100644
--- a/Documentation/networking/device_drivers/index.rst
+++ b/Documentation/networking/device_drivers/index.rst
@@ -25,6 +25,7 @@ Contents:
    mellanox/mlx5
    netronome/nfp
    pensando/ionic
+   stmicro/stmmac
 
 .. only::  subproject and html
 
-- 
cgit v1.2.3-59-g8ed1b


From 0d6e5bfc9c5bc852d240cb35b0d850cdb660c4c9 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@oracle.com>
Date: Tue, 7 Jan 2020 16:11:43 +0300
Subject: enetc: Fix an off by one in enetc_setup_tc_txtime()

The priv->tx_ring[] has 16 elements but only priv->num_tx_rings are
set up, the rest are NULL.  This ">" comparison should be ">=" to avoid
a potential crash.

Fixes: 0d08c9ec7d6e ("enetc: add support time specific departure base on the qos etf")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/freescale/enetc/enetc_qos.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/freescale/enetc/enetc_qos.c b/drivers/net/ethernet/freescale/enetc/enetc_qos.c
index e910aaf0f5ec..00382b7c5bd8 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc_qos.c
+++ b/drivers/net/ethernet/freescale/enetc/enetc_qos.c
@@ -315,7 +315,7 @@ int enetc_setup_tc_txtime(struct net_device *ndev, void *type_data)
 
 	tc = qopt->queue;
 
-	if (tc < 0 || tc > priv->num_tx_rings)
+	if (tc < 0 || tc >= priv->num_tx_rings)
 		return -EINVAL;
 
 	/* Do not support TXSTART and TX CSUM offload simutaniously */
-- 
cgit v1.2.3-59-g8ed1b


From 4addbcb387c9519b320a9411cad68f0c01e9ed4b Mon Sep 17 00:00:00 2001
From: YueHaibing <yuehaibing@huawei.com>
Date: Tue, 7 Jan 2020 22:14:54 +0800
Subject: enetc: Fix inconsistent IS_ERR and PTR_ERR

The proper pointer to be passed as argument is hw
Detected using Coccinelle.

Fixes: 6517798dd343 ("enetc: Make MDIO accessors more generic and export to include/linux/fsl")
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/freescale/enetc/enetc_pci_mdio.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/freescale/enetc/enetc_pci_mdio.c b/drivers/net/ethernet/freescale/enetc/enetc_pci_mdio.c
index 87c0e969da40..ebc635f8a4cc 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc_pci_mdio.c
+++ b/drivers/net/ethernet/freescale/enetc/enetc_pci_mdio.c
@@ -27,7 +27,7 @@ static int enetc_pci_mdio_probe(struct pci_dev *pdev,
 	}
 
 	hw = enetc_hw_alloc(dev, port_regs);
-	if (IS_ERR(enetc_hw_alloc)) {
+	if (IS_ERR(hw)) {
 		err = PTR_ERR(hw);
 		goto err_hw_alloc;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 1ece2fbe9b427d379455f18a874bcd3ab86a2419 Mon Sep 17 00:00:00 2001
From: Vincent Cheng <vincent.cheng.xh@renesas.com>
Date: Tue, 7 Jan 2020 09:47:57 -0500
Subject: ptp: clockmatrix: Rework clockmatrix version information.

Simplify and fix the version information displayed by the driver.
The new info better relects what is needed to support the hardware.

Prev:
Version: 4.8.0, Pipeline 22169 0x4001, Rev 0, Bond 5, CSR 311, IRQ 2

New:
Version: 4.8.0, Id: 0x4001  Hw Rev: 5  OTP Config Select: 15

- Remove pipeline, CSR and IRQ because version x.y.z already incorporates
  this information.
- Remove bond number because it is not used.
- Remove rev number because register was not implemented, always 0
- Add HW Rev ID register to replace rev number
- Add OTP config select to show the user configuration chosen by
  the configurable GPIO pins on start-up

Signed-off-by: Vincent Cheng <vincent.cheng.xh@renesas.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/ptp/idt8a340_reg.h    |  2 ++
 drivers/ptp/ptp_clockmatrix.c | 77 ++++++++-----------------------------------
 2 files changed, 15 insertions(+), 64 deletions(-)

diff --git a/drivers/ptp/idt8a340_reg.h b/drivers/ptp/idt8a340_reg.h
index 9263bc33b8f4..69eedda9f731 100644
--- a/drivers/ptp/idt8a340_reg.h
+++ b/drivers/ptp/idt8a340_reg.h
@@ -77,6 +77,8 @@
 #define JTAG_DEVICE_ID                    0x001c
 #define PRODUCT_ID                        0x001e
 
+#define OTP_SCSR_CONFIG_SELECT            0x0022
+
 #define STATUS                            0xc03c
 #define USER_GPIO0_TO_7_STATUS            0x008a
 #define USER_GPIO8_TO_15_STATUS           0x008b
diff --git a/drivers/ptp/ptp_clockmatrix.c b/drivers/ptp/ptp_clockmatrix.c
index e85836715f0b..032e112c3dd9 100644
--- a/drivers/ptp/ptp_clockmatrix.c
+++ b/drivers/ptp/ptp_clockmatrix.c
@@ -405,6 +405,7 @@ static int _idtcm_set_dpll_tod(struct idtcm_channel *channel,
 	if (wr_trig == HW_TOD_WR_TRIG_SEL_MSB) {
 
 		if (idtcm->calculate_overhead_flag) {
+			/* Assumption: I2C @ 400KHz */
 			total_overhead_ns =  ktime_to_ns(ktime_get_raw()
 							 - idtcm->start_time)
 					     + idtcm->tod_write_overhead_ns
@@ -596,44 +597,7 @@ static int idtcm_state_machine_reset(struct idtcm *idtcm)
 
 static int idtcm_read_hw_rev_id(struct idtcm *idtcm, u8 *hw_rev_id)
 {
-	return idtcm_read(idtcm,
-			  GENERAL_STATUS,
-			  HW_REV_ID,
-			  hw_rev_id,
-			  sizeof(u8));
-}
-
-static int idtcm_read_bond_id(struct idtcm *idtcm, u8 *bond_id)
-{
-	return idtcm_read(idtcm,
-			  GENERAL_STATUS,
-			  BOND_ID,
-			  bond_id,
-			  sizeof(u8));
-}
-
-static int idtcm_read_hw_csr_id(struct idtcm *idtcm, u16 *hw_csr_id)
-{
-	int err;
-	u8 buf[2] = {0};
-
-	err = idtcm_read(idtcm, GENERAL_STATUS, HW_CSR_ID, buf, sizeof(buf));
-
-	*hw_csr_id = (buf[1] << 8) | buf[0];
-
-	return err;
-}
-
-static int idtcm_read_hw_irq_id(struct idtcm *idtcm, u16 *hw_irq_id)
-{
-	int err;
-	u8 buf[2] = {0};
-
-	err = idtcm_read(idtcm, GENERAL_STATUS, HW_IRQ_ID, buf, sizeof(buf));
-
-	*hw_irq_id = (buf[1] << 8) | buf[0];
-
-	return err;
+	return idtcm_read(idtcm, HW_REVISION, REV_ID, hw_rev_id, sizeof(u8));
 }
 
 static int idtcm_read_product_id(struct idtcm *idtcm, u16 *product_id)
@@ -674,20 +638,11 @@ static int idtcm_read_hotfix_release(struct idtcm *idtcm, u8 *hotfix)
 			  sizeof(u8));
 }
 
-static int idtcm_read_pipeline(struct idtcm *idtcm, u32 *pipeline)
+static int idtcm_read_otp_scsr_config_select(struct idtcm *idtcm,
+					     u8 *config_select)
 {
-	int err;
-	u8 buf[4] = {0};
-
-	err = idtcm_read(idtcm,
-			 GENERAL_STATUS,
-			 PIPELINE_ID,
-			 &buf[0],
-			 sizeof(buf));
-
-	*pipeline = (buf[3] << 24) | (buf[2] << 16) | (buf[1] << 8) | buf[0];
-
-	return err;
+	return idtcm_read(idtcm, GENERAL_STATUS, OTP_SCSR_CONFIG_SELECT,
+			  config_select, sizeof(u8));
 }
 
 static int process_pll_mask(struct idtcm *idtcm, u32 addr, u8 val, u8 *mask)
@@ -1078,28 +1033,22 @@ static void idtcm_display_version_info(struct idtcm *idtcm)
 	u8 major;
 	u8 minor;
 	u8 hotfix;
-	u32 pipeline;
 	u16 product_id;
-	u16 csr_id;
-	u16 irq_id;
 	u8 hw_rev_id;
-	u8 bond_id;
+	u8 config_select;
+	char *fmt = "%d.%d.%d, Id: 0x%04x  HW Rev: %d  OTP Config Select: %d\n";
 
 	idtcm_read_major_release(idtcm, &major);
 	idtcm_read_minor_release(idtcm, &minor);
 	idtcm_read_hotfix_release(idtcm, &hotfix);
-	idtcm_read_pipeline(idtcm, &pipeline);
 
 	idtcm_read_product_id(idtcm, &product_id);
 	idtcm_read_hw_rev_id(idtcm, &hw_rev_id);
-	idtcm_read_bond_id(idtcm, &bond_id);
-	idtcm_read_hw_csr_id(idtcm, &csr_id);
-	idtcm_read_hw_irq_id(idtcm, &irq_id);
-
-	dev_info(&idtcm->client->dev, "Version:  %d.%d.%d, Pipeline %u\t"
-		 "0x%04x, Rev %d, Bond %d, CSR %d, IRQ %d\n",
-		 major, minor, hotfix, pipeline,
-		 product_id, hw_rev_id, bond_id, csr_id, irq_id);
+
+	idtcm_read_otp_scsr_config_select(idtcm, &config_select);
+
+	dev_info(&idtcm->client->dev, fmt, major, minor, hotfix,
+		 product_id, hw_rev_id, config_select);
 }
 
 static const struct ptp_clock_info idtcm_caps = {
-- 
cgit v1.2.3-59-g8ed1b


From 53ebeca24a8749f539f1957c5de07fdb031b09ed Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.king@canonical.com>
Date: Tue, 7 Jan 2020 15:24:15 +0000
Subject: net/rose: remove redundant assignment to variable failed

The variable failed is being assigned a value that is never read, the
following goto statement jumps to the end of the function and variable
failed is not referenced at all.  Remove the redundant assignment.

Addresses-Coverity: ("Unused value")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/rose/rose_route.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/net/rose/rose_route.c b/net/rose/rose_route.c
index c53307623236..5277631fa14c 100644
--- a/net/rose/rose_route.c
+++ b/net/rose/rose_route.c
@@ -696,7 +696,6 @@ struct rose_neigh *rose_get_neigh(rose_address *addr, unsigned char *cause,
 				for (i = 0; i < node->count; i++) {
 					if (!rose_ftimer_running(node->neighbour[i])) {
 						res = node->neighbour[i];
-						failed = 0;
 						goto out;
 					}
 					failed = 1;
-- 
cgit v1.2.3-59-g8ed1b


From 5ff051200308ab6f4c64c0fe52187bf4a1234dac Mon Sep 17 00:00:00 2001
From: Michal Rostecki <mrostecki@opensuse.org>
Date: Wed, 8 Jan 2020 17:23:52 +0100
Subject: libbpf: Add probe for large INSN limit

Introduce a new probe which checks whether kernel has large maximum
program size which was increased in the following commit:

c04c0d2b968a ("bpf: increase complexity limit and maximum program size")

Based on the similar check in Cilium[0], authored by Daniel Borkmann.

  [0] https://github.com/cilium/cilium/commit/657d0f585afd26232cfa5d4e70b6f64d2ea91596

Co-authored-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Michal Rostecki <mrostecki@opensuse.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com>
Link: https://lore.kernel.org/bpf/20200108162428.25014-2-mrostecki@opensuse.org
---
 tools/lib/bpf/libbpf.h        |  1 +
 tools/lib/bpf/libbpf.map      |  1 +
 tools/lib/bpf/libbpf_probes.c | 21 +++++++++++++++++++++
 3 files changed, 23 insertions(+)

diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index fe592ef48f1b..26bf539f1b3c 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -521,6 +521,7 @@ LIBBPF_API bool bpf_probe_prog_type(enum bpf_prog_type prog_type,
 LIBBPF_API bool bpf_probe_map_type(enum bpf_map_type map_type, __u32 ifindex);
 LIBBPF_API bool bpf_probe_helper(enum bpf_func_id id,
 				 enum bpf_prog_type prog_type, __u32 ifindex);
+LIBBPF_API bool bpf_probe_large_insn_limit(__u32 ifindex);
 
 /*
  * Get bpf_prog_info in continuous memory
diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
index e9713a574243..b300d74c921a 100644
--- a/tools/lib/bpf/libbpf.map
+++ b/tools/lib/bpf/libbpf.map
@@ -219,6 +219,7 @@ LIBBPF_0.0.7 {
 		bpf_object__detach_skeleton;
 		bpf_object__load_skeleton;
 		bpf_object__open_skeleton;
+		bpf_probe_large_insn_limit;
 		bpf_prog_attach_xattr;
 		bpf_program__attach;
 		bpf_program__name;
diff --git a/tools/lib/bpf/libbpf_probes.c b/tools/lib/bpf/libbpf_probes.c
index a9eb8b322671..221e6ad97012 100644
--- a/tools/lib/bpf/libbpf_probes.c
+++ b/tools/lib/bpf/libbpf_probes.c
@@ -321,3 +321,24 @@ bool bpf_probe_helper(enum bpf_func_id id, enum bpf_prog_type prog_type,
 
 	return res;
 }
+
+/*
+ * Probe for availability of kernel commit (5.3):
+ *
+ * c04c0d2b968a ("bpf: increase complexity limit and maximum program size")
+ */
+bool bpf_probe_large_insn_limit(__u32 ifindex)
+{
+	struct bpf_insn insns[BPF_MAXINSNS + 1];
+	int i;
+
+	for (i = 0; i < BPF_MAXINSNS; i++)
+		insns[i] = BPF_MOV64_IMM(BPF_REG_0, 1);
+	insns[BPF_MAXINSNS] = BPF_EXIT_INSN();
+
+	errno = 0;
+	probe_load(BPF_PROG_TYPE_SCHED_CLS, insns, ARRAY_SIZE(insns), NULL, 0,
+		   ifindex);
+
+	return errno != E2BIG && errno != EINVAL;
+}
-- 
cgit v1.2.3-59-g8ed1b


From 2faef64aa6b3c94c940c3a62ed22f3af4980223a Mon Sep 17 00:00:00 2001
From: Michal Rostecki <mrostecki@opensuse.org>
Date: Wed, 8 Jan 2020 17:23:53 +0100
Subject: bpftool: Add misc section and probe for large INSN limit

Introduce a new probe section (misc) for probes not related to concrete
map types, program types, functions or kernel configuration. Introduce a
probe for large INSN limit as the first one in that section.

Example outputs:

  # bpftool feature probe
  [...]
  Scanning miscellaneous eBPF features...
  Large program size limit is available

  # bpftool feature probe macros
  [...]
  /*** eBPF misc features ***/
  #define HAVE_HAVE_LARGE_INSN_LIMIT

  # bpftool feature probe -j | jq '.["misc"]'
  {
    "have_large_insn_limit": true
  }

Signed-off-by: Michal Rostecki <mrostecki@opensuse.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com>
Link: https://lore.kernel.org/bpf/20200108162428.25014-3-mrostecki@opensuse.org
---
 tools/bpf/bpftool/feature.c | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/tools/bpf/bpftool/feature.c b/tools/bpf/bpftool/feature.c
index 03bdc5b3ac49..94735d968c34 100644
--- a/tools/bpf/bpftool/feature.c
+++ b/tools/bpf/bpftool/feature.c
@@ -572,6 +572,18 @@ probe_helpers_for_progtype(enum bpf_prog_type prog_type, bool supported_type,
 		printf("\n");
 }
 
+static void
+probe_large_insn_limit(const char *define_prefix, __u32 ifindex)
+{
+	bool res;
+
+	res = bpf_probe_large_insn_limit(ifindex);
+	print_bool_feature("have_large_insn_limit",
+			   "Large program size limit",
+			   "HAVE_LARGE_INSN_LIMIT",
+			   res, define_prefix);
+}
+
 static int do_probe(int argc, char **argv)
 {
 	enum probe_component target = COMPONENT_UNSPEC;
@@ -724,6 +736,12 @@ static int do_probe(int argc, char **argv)
 		probe_helpers_for_progtype(i, supported_types[i],
 					   define_prefix, ifindex);
 
+	print_end_then_start_section("misc",
+				     "Scanning miscellaneous eBPF features...",
+				     "/*** eBPF misc features ***/",
+				     define_prefix);
+	probe_large_insn_limit(define_prefix, ifindex);
+
 exit_close_json:
 	if (json_output) {
 		/* End current "section" of probes */
-- 
cgit v1.2.3-59-g8ed1b


From e102774588b3ac0d221ed2d03a5153e056f1354f Mon Sep 17 00:00:00 2001
From: yu kuai <yukuai3@huawei.com>
Date: Mon, 6 Jan 2020 22:18:45 +0800
Subject: net: 3com: 3c59x: remove set but not used variable 'mii_reg1'
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fixes gcc '-Wunused-but-set-variable' warning:

drivers/net/ethernet/3com/3c59x.c: In function ‘vortex_up’:
drivers/net/ethernet/3com/3c59x.c:1551:9: warning: variable
‘mii_reg1’ set but not used [-Wunused-but-set-variable]

It is never used, and so can be removed.

Signed-off-by: yu kuai <yukuai3@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/3com/3c59x.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/3com/3c59x.c b/drivers/net/ethernet/3com/3c59x.c
index fc046797c0ea..a2b7f7ab8170 100644
--- a/drivers/net/ethernet/3com/3c59x.c
+++ b/drivers/net/ethernet/3com/3c59x.c
@@ -1548,7 +1548,7 @@ vortex_up(struct net_device *dev)
 	struct vortex_private *vp = netdev_priv(dev);
 	void __iomem *ioaddr = vp->ioaddr;
 	unsigned int config;
-	int i, mii_reg1, mii_reg5, err = 0;
+	int i, mii_reg5, err = 0;
 
 	if (VORTEX_PCI(vp)) {
 		pci_set_power_state(VORTEX_PCI(vp), PCI_D0);	/* Go active */
@@ -1605,7 +1605,7 @@ vortex_up(struct net_device *dev)
 	window_write32(vp, config, 3, Wn3_Config);
 
 	if (dev->if_port == XCVR_MII || dev->if_port == XCVR_NWAY) {
-		mii_reg1 = mdio_read(dev, vp->phys[0], MII_BMSR);
+		mdio_read(dev, vp->phys[0], MII_BMSR);
 		mii_reg5 = mdio_read(dev, vp->phys[0], MII_LPA);
 		vp->partner_flow_ctrl = ((mii_reg5 & 0x0400) != 0);
 		vp->mii.full_duplex = vp->full_duplex;
-- 
cgit v1.2.3-59-g8ed1b


From 788d10c02f7ed1c971445da5e2addfc70f620311 Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.king@canonical.com>
Date: Tue, 7 Jan 2020 18:00:13 +0000
Subject: Bluetooth: remove redundant assignment to variable icid

Variable icid is being rc is assigned with a value that is never
read. The assignment is redundant and can be removed.

Addresses-Coverity: ("Unused value")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 net/bluetooth/l2cap_core.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c
index 1bca608e0170..195459a1e53e 100644
--- a/net/bluetooth/l2cap_core.c
+++ b/net/bluetooth/l2cap_core.c
@@ -5081,7 +5081,6 @@ static inline int l2cap_move_channel_req(struct l2cap_conn *conn,
 	chan->move_role = L2CAP_MOVE_ROLE_RESPONDER;
 	l2cap_move_setup(chan);
 	chan->move_id = req->dest_amp_id;
-	icid = chan->dcid;
 
 	if (req->dest_amp_id == AMP_ID_BREDR) {
 		/* Moving to BR/EDR */
-- 
cgit v1.2.3-59-g8ed1b


From 87c2a2a7d7631d0caa956b73f9f20010b5f55681 Mon Sep 17 00:00:00 2001
From: Wei Yongjun <weiyongjun1@huawei.com>
Date: Wed, 8 Jan 2020 03:59:31 +0000
Subject: Bluetooth: hci_qca: Use vfree() instead of kfree()

Use vfree() instead of kfree() to free vmalloc()
allocated data.

Fixes: d841502c79e3 ("Bluetooth: hci_qca: Collect controller memory dump during SSR")
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Reviewed-by: Balakrishna Godavarthi <bgodavar@codeaurora.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 drivers/bluetooth/hci_qca.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c
index 9392cc7f9908..a17260641283 100644
--- a/drivers/bluetooth/hci_qca.c
+++ b/drivers/bluetooth/hci_qca.c
@@ -529,7 +529,7 @@ static void hci_memdump_timeout(struct timer_list *t)
 	bt_dev_err(hu->hdev, "clearing allocated memory due to memdump timeout");
 	/* Inject hw error event to reset the device and driver. */
 	hci_reset_dev(hu->hdev);
-	kfree(memdump_buf);
+	vfree(memdump_buf);
 	kfree(qca_memdump);
 	qca->memdump_state = QCA_MEMDUMP_TIMEOUT;
 	del_timer(&qca->memdump_timer);
@@ -1437,7 +1437,7 @@ static void qca_wait_for_dump_collection(struct hci_dev *hdev)
 		bt_dev_err(hu->hdev, "Clearing the buffers due to timeout");
 		if (qca_memdump)
 			memdump_buf = qca_memdump->memdump_buf_tail;
-		kfree(memdump_buf);
+		vfree(memdump_buf);
 		kfree(qca_memdump);
 		qca->memdump_state = QCA_MEMDUMP_TIMEOUT;
 		del_timer(&qca->memdump_timer);
-- 
cgit v1.2.3-59-g8ed1b


From 56b084ed6eaa0a6353395deffb601dbd6570a1a5 Mon Sep 17 00:00:00 2001
From: YueHaibing <yuehaibing@huawei.com>
Date: Wed, 8 Jan 2020 01:54:31 +0000
Subject: Bluetooth: hci_qca: Remove set but not used variable 'opcode'

Fixes gcc '-Wunused-but-set-variable' warning:

drivers/bluetooth/hci_qca.c: In function 'qca_controller_memdump':
drivers/bluetooth/hci_qca.c:980:6: warning:
 variable 'opcode' set but not used [-Wunused-but-set-variable]

It is never used since commit d841502c79e3 ("Bluetooth: hci_qca: Collect
controller memory dump during SSR"), so remove it.

Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 drivers/bluetooth/hci_qca.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c
index a17260641283..82e4cd4b6663 100644
--- a/drivers/bluetooth/hci_qca.c
+++ b/drivers/bluetooth/hci_qca.c
@@ -977,7 +977,7 @@ static void qca_controller_memdump(struct work_struct *work)
 	struct qca_dump_size *dump;
 	char *memdump_buf;
 	char nullBuff[QCA_DUMP_PACKET_SIZE] = { 0 };
-	u16 opcode, seq_no;
+	u16 seq_no;
 	u32 dump_size;
 
 	while ((skb = skb_dequeue(&qca->rx_memdump_q))) {
@@ -993,7 +993,6 @@ static void qca_controller_memdump(struct work_struct *work)
 
 		qca->memdump_state = QCA_MEMDUMP_COLLECTING;
 		cmd_hdr = (void *) skb->data;
-		opcode = __le16_to_cpu(cmd_hdr->opcode);
 		seq_no = __le16_to_cpu(cmd_hdr->seq_no);
 		skb_pull(skb, sizeof(struct qca_memdump_event_hdr));
 
-- 
cgit v1.2.3-59-g8ed1b


From 7c7b58d46b7636ff3cfb01cc633dda852fda70c8 Mon Sep 17 00:00:00 2001
From: Vijay Khemka <vijaykhemka@fb.com>
Date: Tue, 7 Jan 2020 11:30:33 -0800
Subject: net/ncsi: Send device address as source address

After receiving device mac address from device, send this as
a source address for further commands instead of broadcast
address.

This will help in multi host NIC cards.

Signed-off-by: Vijay Khemka <vijaykhemka@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ncsi/ncsi-cmd.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/net/ncsi/ncsi-cmd.c b/net/ncsi/ncsi-cmd.c
index 0187e65176c0..ba9ae482141b 100644
--- a/net/ncsi/ncsi-cmd.c
+++ b/net/ncsi/ncsi-cmd.c
@@ -369,7 +369,15 @@ int ncsi_xmit_cmd(struct ncsi_cmd_arg *nca)
 	eh = skb_push(nr->cmd, sizeof(*eh));
 	eh->h_proto = htons(ETH_P_NCSI);
 	eth_broadcast_addr(eh->h_dest);
-	eth_broadcast_addr(eh->h_source);
+
+	/* If mac address received from device then use it for
+	 * source address as unicast address else use broadcast
+	 * address as source address
+	 */
+	if (nca->ndp->gma_flag == 1)
+		memcpy(eh->h_source, nca->ndp->ndev.dev->dev_addr, ETH_ALEN);
+	else
+		eth_broadcast_addr(eh->h_source);
 
 	/* Start the timer for the request that might not have
 	 * corresponding response. Given NCSI is an internal
-- 
cgit v1.2.3-59-g8ed1b


From b9ae51273655a72a12fba730843fd72fb132735a Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Tue, 7 Jan 2020 21:03:39 +0100
Subject: hsr: fix dummy hsr_debugfs_rename() declaration

The hsr_debugfs_rename prototype got an extra 'void' that needs to
be removed again:

In file included from /git/arm-soc/net/hsr/hsr_main.c:12:
net/hsr/hsr_main.h:194:20: error: two or more data types in declaration specifiers
 static inline void void hsr_debugfs_rename(struct net_device *dev)

Fixes: 4c2d5e33dcd3 ("hsr: rename debugfs file when interface name is changed")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/hsr/hsr_main.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/hsr/hsr_main.h b/net/hsr/hsr_main.h
index d40de84a637f..754d84b217f0 100644
--- a/net/hsr/hsr_main.h
+++ b/net/hsr/hsr_main.h
@@ -191,7 +191,7 @@ void hsr_debugfs_term(struct hsr_priv *priv);
 void hsr_debugfs_create_root(void);
 void hsr_debugfs_remove_root(void);
 #else
-static inline void void hsr_debugfs_rename(struct net_device *dev)
+static inline void hsr_debugfs_rename(struct net_device *dev)
 {
 }
 static inline void hsr_debugfs_init(struct hsr_priv *priv,
-- 
cgit v1.2.3-59-g8ed1b


From e1253f391042e34de66faf7d499595c00f99a076 Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Wed, 8 Jan 2020 16:10:32 +0000
Subject: sfc: add new headers in preparation for code split

New headers contain prototypes of functions that will be common between
ef10 and upcoming driver.
Removed static modifier from the affected functions.
Some function prototypes were removed from existing headers.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/ef10.c             |  2 +
 drivers/net/ethernet/sfc/efx.c              | 91 +++++++++++++----------------
 drivers/net/ethernet/sfc/efx.h              | 32 ----------
 drivers/net/ethernet/sfc/efx_channels.h     | 55 +++++++++++++++++
 drivers/net/ethernet/sfc/efx_common.h       | 52 +++++++++++++++++
 drivers/net/ethernet/sfc/ethtool.c          |  3 +
 drivers/net/ethernet/sfc/farch.c            |  1 +
 drivers/net/ethernet/sfc/mcdi.h             |  1 -
 drivers/net/ethernet/sfc/mcdi_functions.h   | 30 ++++++++++
 drivers/net/ethernet/sfc/mcdi_port.c        | 50 ++++++----------
 drivers/net/ethernet/sfc/mcdi_port_common.h | 53 +++++++++++++++++
 drivers/net/ethernet/sfc/net_driver.h       | 13 ++++-
 drivers/net/ethernet/sfc/nic.h              |  6 ++
 drivers/net/ethernet/sfc/rx.c               | 22 +++----
 drivers/net/ethernet/sfc/rx_common.h        | 42 +++++++++++++
 drivers/net/ethernet/sfc/selftest.c         |  2 +
 drivers/net/ethernet/sfc/siena.c            |  1 +
 drivers/net/ethernet/sfc/tx.c               | 19 +++---
 drivers/net/ethernet/sfc/tx_common.h        | 31 ++++++++++
 19 files changed, 365 insertions(+), 141 deletions(-)
 create mode 100644 drivers/net/ethernet/sfc/efx_channels.h
 create mode 100644 drivers/net/ethernet/sfc/efx_common.h
 create mode 100644 drivers/net/ethernet/sfc/mcdi_functions.h
 create mode 100644 drivers/net/ethernet/sfc/mcdi_port_common.h
 create mode 100644 drivers/net/ethernet/sfc/rx_common.h
 create mode 100644 drivers/net/ethernet/sfc/tx_common.h

diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c
index 4d9bbccc6f89..de6e6cd4469b 100644
--- a/drivers/net/ethernet/sfc/ef10.c
+++ b/drivers/net/ethernet/sfc/ef10.c
@@ -9,10 +9,12 @@
 #include "io.h"
 #include "mcdi.h"
 #include "mcdi_pcol.h"
+#include "mcdi_port_common.h"
 #include "nic.h"
 #include "workarounds.h"
 #include "selftest.h"
 #include "ef10_sriov.h"
+#include "rx_common.h"
 #include <linux/in.h>
 #include <linux/jhash.h>
 #include <linux/wait.h>
diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index 033907e6fdb0..c5bcdfcfee87 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -23,6 +23,10 @@
 #include <net/gre.h>
 #include <net/udp_tunnel.h>
 #include "efx.h"
+#include "efx_common.h"
+#include "efx_channels.h"
+#include "rx_common.h"
+#include "tx_common.h"
 #include "nic.h"
 #include "io.h"
 #include "selftest.h"
@@ -214,18 +218,8 @@ MODULE_PARM_DESC(debug, "Bitmapped debugging message enable value");
  *
  *************************************************************************/
 
-static int efx_soft_enable_interrupts(struct efx_nic *efx);
-static void efx_soft_disable_interrupts(struct efx_nic *efx);
-static void efx_remove_channel(struct efx_channel *channel);
-static void efx_remove_channels(struct efx_nic *efx);
 static const struct efx_channel_type efx_default_channel_type;
 static void efx_remove_port(struct efx_nic *efx);
-static void efx_init_napi_channel(struct efx_channel *channel);
-static void efx_fini_napi(struct efx_nic *efx);
-static void efx_fini_napi_channel(struct efx_channel *channel);
-static void efx_fini_struct(struct efx_nic *efx);
-static void efx_start_all(struct efx_nic *efx);
-static void efx_stop_all(struct efx_nic *efx);
 static int efx_xdp_setup_prog(struct efx_nic *efx, struct bpf_prog *prog);
 static int efx_xdp(struct net_device *dev, struct netdev_bpf *xdp);
 static int efx_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **xdpfs,
@@ -239,7 +233,7 @@ static int efx_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **xdpfs,
 			ASSERT_RTNL();			\
 	} while (0)
 
-static int efx_check_disabled(struct efx_nic *efx)
+int efx_check_disabled(struct efx_nic *efx)
 {
 	if (efx->state == STATE_DISABLED || efx->state == STATE_RECOVERY) {
 		netif_err(efx, drv, efx->net_dev,
@@ -375,7 +369,7 @@ static int efx_poll(struct napi_struct *napi, int budget)
  * is reset, the memory buffer will be reused; this guards against
  * errors during channel reset and also simplifies interrupt handling.
  */
-static int efx_probe_eventq(struct efx_channel *channel)
+int efx_probe_eventq(struct efx_channel *channel)
 {
 	struct efx_nic *efx = channel->efx;
 	unsigned long entries;
@@ -393,7 +387,7 @@ static int efx_probe_eventq(struct efx_channel *channel)
 }
 
 /* Prepare channel's event queue */
-static int efx_init_eventq(struct efx_channel *channel)
+int efx_init_eventq(struct efx_channel *channel)
 {
 	struct efx_nic *efx = channel->efx;
 	int rc;
@@ -436,7 +430,7 @@ void efx_stop_eventq(struct efx_channel *channel)
 	channel->enabled = false;
 }
 
-static void efx_fini_eventq(struct efx_channel *channel)
+void efx_fini_eventq(struct efx_channel *channel)
 {
 	if (!channel->eventq_init)
 		return;
@@ -448,7 +442,7 @@ static void efx_fini_eventq(struct efx_channel *channel)
 	channel->eventq_init = false;
 }
 
-static void efx_remove_eventq(struct efx_channel *channel)
+void efx_remove_eventq(struct efx_channel *channel)
 {
 	netif_dbg(channel->efx, drv, channel->efx->net_dev,
 		  "chan %d remove event queue\n", channel->channel);
@@ -463,7 +457,7 @@ static void efx_remove_eventq(struct efx_channel *channel)
  *************************************************************************/
 
 /* Allocate and initialise a channel structure. */
-static struct efx_channel *
+struct efx_channel *
 efx_alloc_channel(struct efx_nic *efx, int i, struct efx_channel *old_channel)
 {
 	struct efx_channel *channel;
@@ -500,8 +494,7 @@ efx_alloc_channel(struct efx_nic *efx, int i, struct efx_channel *old_channel)
 /* Allocate and initialise a channel structure, copying parameters
  * (but not resources) from an old channel structure.
  */
-static struct efx_channel *
-efx_copy_channel(const struct efx_channel *old_channel)
+struct efx_channel *efx_copy_channel(const struct efx_channel *old_channel)
 {
 	struct efx_channel *channel;
 	struct efx_rx_queue *rx_queue;
@@ -577,8 +570,7 @@ fail:
 	return rc;
 }
 
-static void
-efx_get_channel_name(struct efx_channel *channel, char *buf, size_t len)
+void efx_get_channel_name(struct efx_channel *channel, char *buf, size_t len)
 {
 	struct efx_nic *efx = channel->efx;
 	const char *type;
@@ -601,7 +593,7 @@ efx_get_channel_name(struct efx_channel *channel, char *buf, size_t len)
 	snprintf(buf, len, "%s%s-%d", efx->name, type, number);
 }
 
-static void efx_set_channel_names(struct efx_nic *efx)
+void efx_set_channel_names(struct efx_nic *efx)
 {
 	struct efx_channel *channel;
 
@@ -611,7 +603,7 @@ static void efx_set_channel_names(struct efx_nic *efx)
 					sizeof(efx->msi_context[0].name));
 }
 
-static int efx_probe_channels(struct efx_nic *efx)
+int efx_probe_channels(struct efx_nic *efx)
 {
 	struct efx_channel *channel;
 	int rc;
@@ -788,7 +780,7 @@ static void efx_stop_datapath(struct efx_nic *efx)
 	efx->xdp_rxq_info_failed = false;
 }
 
-static void efx_remove_channel(struct efx_channel *channel)
+void efx_remove_channel(struct efx_channel *channel)
 {
 	struct efx_tx_queue *tx_queue;
 	struct efx_rx_queue *rx_queue;
@@ -804,7 +796,7 @@ static void efx_remove_channel(struct efx_channel *channel)
 	channel->type->post_remove(channel);
 }
 
-static void efx_remove_channels(struct efx_nic *efx)
+void efx_remove_channels(struct efx_nic *efx)
 {
 	struct efx_channel *channel;
 
@@ -814,8 +806,7 @@ static void efx_remove_channels(struct efx_nic *efx)
 	kfree(efx->xdp_tx_queues);
 }
 
-int
-efx_realloc_channels(struct efx_nic *efx, u32 rxq_entries, u32 txq_entries)
+int efx_realloc_channels(struct efx_nic *efx, u32 rxq_entries, u32 txq_entries)
 {
 	struct efx_channel *other_channel[EFX_MAX_CHANNELS], *channel;
 	u32 old_rxq_entries, old_txq_entries;
@@ -929,7 +920,7 @@ void efx_schedule_slow_fill(struct efx_rx_queue *rx_queue)
 	mod_timer(&rx_queue->slow_fill, jiffies + msecs_to_jiffies(10));
 }
 
-static bool efx_default_channel_want_txqs(struct efx_channel *channel)
+bool efx_default_channel_want_txqs(struct efx_channel *channel)
 {
 	return channel->channel - channel->efx->tx_channel_offset <
 		channel->efx->n_tx_channels;
@@ -1292,7 +1283,7 @@ static void efx_dissociate(struct efx_nic *efx)
 }
 
 /* This configures the PCI device to enable I/O and DMA. */
-static int efx_init_io(struct efx_nic *efx)
+int efx_init_io(struct efx_nic *efx)
 {
 	struct pci_dev *pci_dev = efx->pci_dev;
 	dma_addr_t dma_mask = efx->type->max_dma_mask;
@@ -1363,7 +1354,7 @@ static int efx_init_io(struct efx_nic *efx)
 	return rc;
 }
 
-static void efx_fini_io(struct efx_nic *efx)
+void efx_fini_io(struct efx_nic *efx)
 {
 	int bar;
 
@@ -1545,7 +1536,7 @@ static int efx_allocate_msix_channels(struct efx_nic *efx,
 /* Probe the number and type of interrupts we are able to obtain, and
  * the resulting numbers of channels and RX queues.
  */
-static int efx_probe_interrupts(struct efx_nic *efx)
+int efx_probe_interrupts(struct efx_nic *efx)
 {
 	unsigned int extra_channels = 0;
 	unsigned int rss_spread;
@@ -1657,7 +1648,7 @@ static int efx_probe_interrupts(struct efx_nic *efx)
 }
 
 #if defined(CONFIG_SMP)
-static void efx_set_interrupt_affinity(struct efx_nic *efx)
+void efx_set_interrupt_affinity(struct efx_nic *efx)
 {
 	struct efx_channel *channel;
 	unsigned int cpu;
@@ -1669,7 +1660,7 @@ static void efx_set_interrupt_affinity(struct efx_nic *efx)
 	}
 }
 
-static void efx_clear_interrupt_affinity(struct efx_nic *efx)
+void efx_clear_interrupt_affinity(struct efx_nic *efx)
 {
 	struct efx_channel *channel;
 
@@ -1677,18 +1668,16 @@ static void efx_clear_interrupt_affinity(struct efx_nic *efx)
 		irq_set_affinity_hint(channel->irq, NULL);
 }
 #else
-static void
-efx_set_interrupt_affinity(struct efx_nic *efx __attribute__ ((unused)))
+void efx_set_interrupt_affinity(struct efx_nic *efx __attribute__ ((unused)))
 {
 }
 
-static void
-efx_clear_interrupt_affinity(struct efx_nic *efx __attribute__ ((unused)))
+void efx_clear_interrupt_affinity(struct efx_nic *efx __attribute__ ((unused)))
 {
 }
 #endif /* CONFIG_SMP */
 
-static int efx_soft_enable_interrupts(struct efx_nic *efx)
+int efx_soft_enable_interrupts(struct efx_nic *efx)
 {
 	struct efx_channel *channel, *end_channel;
 	int rc;
@@ -1723,7 +1712,7 @@ fail:
 	return rc;
 }
 
-static void efx_soft_disable_interrupts(struct efx_nic *efx)
+void efx_soft_disable_interrupts(struct efx_nic *efx)
 {
 	struct efx_channel *channel;
 
@@ -1751,7 +1740,7 @@ static void efx_soft_disable_interrupts(struct efx_nic *efx)
 	efx_mcdi_flush_async(efx);
 }
 
-static int efx_enable_interrupts(struct efx_nic *efx)
+int efx_enable_interrupts(struct efx_nic *efx)
 {
 	struct efx_channel *channel, *end_channel;
 	int rc;
@@ -1793,7 +1782,7 @@ fail:
 	return rc;
 }
 
-static void efx_disable_interrupts(struct efx_nic *efx)
+void efx_disable_interrupts(struct efx_nic *efx)
 {
 	struct efx_channel *channel;
 
@@ -1807,7 +1796,7 @@ static void efx_disable_interrupts(struct efx_nic *efx)
 	efx->type->irq_disable_non_ev(efx);
 }
 
-static void efx_remove_interrupts(struct efx_nic *efx)
+void efx_remove_interrupts(struct efx_nic *efx)
 {
 	struct efx_channel *channel;
 
@@ -1821,7 +1810,7 @@ static void efx_remove_interrupts(struct efx_nic *efx)
 	efx->legacy_irq = 0;
 }
 
-static int efx_set_channels(struct efx_nic *efx)
+int efx_set_channels(struct efx_nic *efx)
 {
 	struct efx_channel *channel;
 	struct efx_tx_queue *tx_queue;
@@ -2074,7 +2063,7 @@ static int efx_probe_all(struct efx_nic *efx)
  * is safe to call multiple times, so long as the NIC is not disabled.
  * Requires the RTNL lock.
  */
-static void efx_start_all(struct efx_nic *efx)
+void efx_start_all(struct efx_nic *efx)
 {
 	EFX_ASSERT_RESET_SERIALISED(efx);
 	BUG_ON(efx->state == STATE_DISABLED);
@@ -2113,7 +2102,7 @@ static void efx_start_all(struct efx_nic *efx)
  * times with the NIC in almost any state, but interrupts should be
  * enabled.  Requires the RTNL lock.
  */
-static void efx_stop_all(struct efx_nic *efx)
+void efx_stop_all(struct efx_nic *efx)
 {
 	EFX_ASSERT_RESET_SERIALISED(efx);
 
@@ -2298,7 +2287,7 @@ static int efx_ioctl(struct net_device *net_dev, struct ifreq *ifr, int cmd)
  *
  **************************************************************************/
 
-static void efx_init_napi_channel(struct efx_channel *channel)
+void efx_init_napi_channel(struct efx_channel *channel)
 {
 	struct efx_nic *efx = channel->efx;
 
@@ -2307,7 +2296,7 @@ static void efx_init_napi_channel(struct efx_channel *channel)
 		       efx_poll, napi_weight);
 }
 
-static void efx_init_napi(struct efx_nic *efx)
+void efx_init_napi(struct efx_nic *efx)
 {
 	struct efx_channel *channel;
 
@@ -2315,7 +2304,7 @@ static void efx_init_napi(struct efx_nic *efx)
 		efx_init_napi_channel(channel);
 }
 
-static void efx_fini_napi_channel(struct efx_channel *channel)
+void efx_fini_napi_channel(struct efx_channel *channel)
 {
 	if (channel->napi_dev)
 		netif_napi_del(&channel->napi_str);
@@ -2323,7 +2312,7 @@ static void efx_fini_napi_channel(struct efx_channel *channel)
 	channel->napi_dev = NULL;
 }
 
-static void efx_fini_napi(struct efx_nic *efx)
+void efx_fini_napi(struct efx_nic *efx)
 {
 	struct efx_channel *channel;
 
@@ -3203,8 +3192,8 @@ static const struct efx_phy_operations efx_dummy_phy_operations = {
 /* This zeroes out and then fills in the invariants in a struct
  * efx_nic (including all sub-structures).
  */
-static int efx_init_struct(struct efx_nic *efx,
-			   struct pci_dev *pci_dev, struct net_device *net_dev)
+int efx_init_struct(struct efx_nic *efx, struct pci_dev *pci_dev,
+		    struct net_device *net_dev)
 {
 	int rc = -ENOMEM, i;
 
@@ -3283,7 +3272,7 @@ fail:
 	return rc;
 }
 
-static void efx_fini_struct(struct efx_nic *efx)
+void efx_fini_struct(struct efx_nic *efx)
 {
 	int i;
 
diff --git a/drivers/net/ethernet/sfc/efx.h b/drivers/net/ethernet/sfc/efx.h
index 2dd8d5002315..6ff454f2cb62 100644
--- a/drivers/net/ethernet/sfc/efx.h
+++ b/drivers/net/ethernet/sfc/efx.h
@@ -15,31 +15,19 @@ int efx_net_open(struct net_device *net_dev);
 int efx_net_stop(struct net_device *net_dev);
 
 /* TX */
-int efx_probe_tx_queue(struct efx_tx_queue *tx_queue);
-void efx_remove_tx_queue(struct efx_tx_queue *tx_queue);
-void efx_init_tx_queue(struct efx_tx_queue *tx_queue);
 void efx_init_tx_queue_core_txq(struct efx_tx_queue *tx_queue);
-void efx_fini_tx_queue(struct efx_tx_queue *tx_queue);
 netdev_tx_t efx_hard_start_xmit(struct sk_buff *skb,
 				struct net_device *net_dev);
 netdev_tx_t efx_enqueue_skb(struct efx_tx_queue *tx_queue, struct sk_buff *skb);
 void efx_xmit_done(struct efx_tx_queue *tx_queue, unsigned int index);
 int efx_setup_tc(struct net_device *net_dev, enum tc_setup_type type,
 		 void *type_data);
-unsigned int efx_tx_max_skb_descs(struct efx_nic *efx);
 extern unsigned int efx_piobuf_size;
 extern bool efx_separate_tx_channels;
 
 /* RX */
 void efx_set_default_rx_indir_table(struct efx_nic *efx,
 				    struct efx_rss_context *ctx);
-void efx_rx_config_page_split(struct efx_nic *efx);
-int efx_probe_rx_queue(struct efx_rx_queue *rx_queue);
-void efx_remove_rx_queue(struct efx_rx_queue *rx_queue);
-void efx_init_rx_queue(struct efx_rx_queue *rx_queue);
-void efx_fini_rx_queue(struct efx_rx_queue *rx_queue);
-void efx_fast_push_rx_descriptors(struct efx_rx_queue *rx_queue, bool atomic);
-void efx_rx_slow_fill(struct timer_list *t);
 void __efx_rx_packet(struct efx_channel *channel);
 void efx_rx_packet(struct efx_rx_queue *rx_queue, unsigned int index,
 		   unsigned int n_frags, unsigned int len, u16 flags);
@@ -48,7 +36,6 @@ static inline void efx_rx_flush_packet(struct efx_channel *channel)
 	if (channel->rx_pkt_n_frags)
 		__efx_rx_packet(channel);
 }
-void efx_schedule_slow_fill(struct efx_rx_queue *rx_queue);
 
 #define EFX_MAX_DMAQ_SIZE 4096UL
 #define EFX_DEFAULT_DMAQ_SIZE 1024UL
@@ -80,8 +67,6 @@ static inline bool efx_rss_enabled(struct efx_nic *efx)
 
 /* Filters */
 
-void efx_mac_reconfigure(struct efx_nic *efx);
-
 /**
  * efx_filter_insert_filter - add or replace a filter
  * @efx: NIC in which to insert the filter
@@ -221,23 +206,11 @@ static inline bool efx_rss_active(struct efx_rss_context *ctx)
 /* Channels */
 int efx_channel_dummy_op_int(struct efx_channel *channel);
 void efx_channel_dummy_op_void(struct efx_channel *channel);
-int efx_realloc_channels(struct efx_nic *efx, u32 rxq_entries, u32 txq_entries);
-
-/* Ports */
-int efx_reconfigure_port(struct efx_nic *efx);
-int __efx_reconfigure_port(struct efx_nic *efx);
 
 /* Ethtool support */
 extern const struct ethtool_ops efx_ethtool_ops;
 
-/* Reset handling */
-int efx_reset(struct efx_nic *efx, enum reset_type method);
-void efx_reset_down(struct efx_nic *efx, enum reset_type method);
-int efx_reset_up(struct efx_nic *efx, enum reset_type method, bool ok);
-int efx_try_recovery(struct efx_nic *efx);
-
 /* Global */
-void efx_schedule_reset(struct efx_nic *efx, enum reset_type type);
 unsigned int efx_usecs_to_ticks(struct efx_nic *efx, unsigned int usecs);
 unsigned int efx_ticks_to_usecs(struct efx_nic *efx, unsigned int ticks);
 int efx_init_irq_moderation(struct efx_nic *efx, unsigned int tx_usecs,
@@ -245,8 +218,6 @@ int efx_init_irq_moderation(struct efx_nic *efx, unsigned int tx_usecs,
 			    bool rx_may_override_tx);
 void efx_get_irq_moderation(struct efx_nic *efx, unsigned int *tx_usecs,
 			    unsigned int *rx_usecs, bool *rx_adaptive);
-void efx_stop_eventq(struct efx_channel *channel);
-void efx_start_eventq(struct efx_channel *channel);
 
 /* Dummy PHY ops for PHY drivers */
 int efx_port_dummy_op_int(struct efx_nic *efx);
@@ -293,9 +264,6 @@ static inline void efx_schedule_channel_irq(struct efx_channel *channel)
 	efx_schedule_channel(channel);
 }
 
-void efx_link_status_changed(struct efx_nic *efx);
-void efx_link_set_advertising(struct efx_nic *efx,
-			      const unsigned long *advertising);
 void efx_link_clear_advertising(struct efx_nic *efx);
 void efx_link_set_wanted_fc(struct efx_nic *efx, u8);
 
diff --git a/drivers/net/ethernet/sfc/efx_channels.h b/drivers/net/ethernet/sfc/efx_channels.h
new file mode 100644
index 000000000000..8d7b8c4142d7
--- /dev/null
+++ b/drivers/net/ethernet/sfc/efx_channels.h
@@ -0,0 +1,55 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/****************************************************************************
+ * Driver for Solarflare network controllers and boards
+ * Copyright 2018 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#ifndef EFX_CHANNELS_H
+#define EFX_CHANNELS_H
+
+int efx_probe_interrupts(struct efx_nic *efx);
+void efx_remove_interrupts(struct efx_nic *efx);
+int efx_soft_enable_interrupts(struct efx_nic *efx);
+void efx_soft_disable_interrupts(struct efx_nic *efx);
+int efx_enable_interrupts(struct efx_nic *efx);
+void efx_disable_interrupts(struct efx_nic *efx);
+
+void efx_set_interrupt_affinity(struct efx_nic *efx);
+void efx_clear_interrupt_affinity(struct efx_nic *efx);
+
+int efx_probe_eventq(struct efx_channel *channel);
+int efx_init_eventq(struct efx_channel *channel);
+void efx_start_eventq(struct efx_channel *channel);
+void efx_stop_eventq(struct efx_channel *channel);
+void efx_fini_eventq(struct efx_channel *channel);
+void efx_remove_eventq(struct efx_channel *channel);
+
+struct efx_channel *
+efx_alloc_channel(struct efx_nic *efx, int i, struct efx_channel *old_channel);
+int efx_realloc_channels(struct efx_nic *efx, u32 rxq_entries, u32 txq_entries);
+void efx_get_channel_name(struct efx_channel *channel, char *buf, size_t len);
+void efx_set_channel_names(struct efx_nic *efx);
+int efx_init_channels(struct efx_nic *efx);
+int efx_probe_channels(struct efx_nic *efx);
+int efx_set_channels(struct efx_nic *efx);
+bool efx_default_channel_want_txqs(struct efx_channel *channel);
+void efx_remove_channel(struct efx_channel *channel);
+void efx_remove_channels(struct efx_nic *efx);
+void efx_fini_channels(struct efx_nic *efx);
+struct efx_channel *efx_copy_channel(const struct efx_channel *old_channel);
+void efx_start_channels(struct efx_nic *efx);
+void efx_stop_channels(struct efx_nic *efx);
+
+void efx_init_napi_channel(struct efx_channel *channel);
+void efx_init_napi(struct efx_nic *efx);
+void efx_fini_napi_channel(struct efx_channel *channel);
+void efx_fini_napi(struct efx_nic *efx);
+
+int efx_channel_dummy_op_int(struct efx_channel *channel);
+void efx_channel_dummy_op_void(struct efx_channel *channel);
+
+#endif
diff --git a/drivers/net/ethernet/sfc/efx_common.h b/drivers/net/ethernet/sfc/efx_common.h
new file mode 100644
index 000000000000..cb690d01adbc
--- /dev/null
+++ b/drivers/net/ethernet/sfc/efx_common.h
@@ -0,0 +1,52 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/****************************************************************************
+ * Driver for Solarflare network controllers and boards
+ * Copyright 2018 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#ifndef EFX_COMMON_H
+#define EFX_COMMON_H
+
+int efx_init_io(struct efx_nic *efx);
+void efx_fini_io(struct efx_nic *efx);
+int efx_init_struct(struct efx_nic *efx, struct pci_dev *pci_dev,
+		    struct net_device *net_dev);
+void efx_fini_struct(struct efx_nic *efx);
+
+void efx_start_all(struct efx_nic *efx);
+void efx_stop_all(struct efx_nic *efx);
+
+int efx_create_reset_workqueue(void);
+void efx_queue_reset_work(struct efx_nic *efx);
+void efx_flush_reset_workqueue(struct efx_nic *efx);
+void efx_destroy_reset_workqueue(void);
+
+void efx_start_monitor(struct efx_nic *efx);
+
+int __efx_reconfigure_port(struct efx_nic *efx);
+int efx_reconfigure_port(struct efx_nic *efx);
+
+#define EFX_ASSERT_RESET_SERIALISED(efx)		\
+	do {						\
+		if ((efx->state == STATE_READY) ||	\
+		    (efx->state == STATE_RECOVERY) ||	\
+		    (efx->state == STATE_DISABLED))	\
+			ASSERT_RTNL();			\
+	} while (0)
+
+int efx_try_recovery(struct efx_nic *efx);
+void efx_reset_down(struct efx_nic *efx, enum reset_type method);
+int efx_reset_up(struct efx_nic *efx, enum reset_type method, bool ok);
+int efx_reset(struct efx_nic *efx, enum reset_type method);
+void efx_schedule_reset(struct efx_nic *efx, enum reset_type type);
+
+int efx_check_disabled(struct efx_nic *efx);
+
+void efx_mac_reconfigure(struct efx_nic *efx);
+void efx_link_status_changed(struct efx_nic *efx);
+
+#endif
diff --git a/drivers/net/ethernet/sfc/ethtool.c b/drivers/net/ethernet/sfc/ethtool.c
index b31032da4bcb..f83398721ad7 100644
--- a/drivers/net/ethernet/sfc/ethtool.c
+++ b/drivers/net/ethernet/sfc/ethtool.c
@@ -13,6 +13,9 @@
 #include "workarounds.h"
 #include "selftest.h"
 #include "efx.h"
+#include "efx_channels.h"
+#include "rx_common.h"
+#include "tx_common.h"
 #include "filter.h"
 #include "nic.h"
 
diff --git a/drivers/net/ethernet/sfc/farch.c b/drivers/net/ethernet/sfc/farch.c
index eedd32e2bfcb..dbbb898adddb 100644
--- a/drivers/net/ethernet/sfc/farch.c
+++ b/drivers/net/ethernet/sfc/farch.c
@@ -15,6 +15,7 @@
 #include "net_driver.h"
 #include "bitfield.h"
 #include "efx.h"
+#include "rx_common.h"
 #include "nic.h"
 #include "farch_regs.h"
 #include "sriov.h"
diff --git a/drivers/net/ethernet/sfc/mcdi.h b/drivers/net/ethernet/sfc/mcdi.h
index 9081f84a2604..65e454a062f7 100644
--- a/drivers/net/ethernet/sfc/mcdi.h
+++ b/drivers/net/ethernet/sfc/mcdi.h
@@ -346,7 +346,6 @@ int efx_mcdi_flush_rxqs(struct efx_nic *efx);
 int efx_mcdi_port_probe(struct efx_nic *efx);
 void efx_mcdi_port_remove(struct efx_nic *efx);
 int efx_mcdi_port_reconfigure(struct efx_nic *efx);
-int efx_mcdi_port_get_number(struct efx_nic *efx);
 u32 efx_mcdi_phy_get_caps(struct efx_nic *efx);
 void efx_mcdi_process_link_change(struct efx_nic *efx, efx_qword_t *ev);
 int efx_mcdi_set_mac(struct efx_nic *efx);
diff --git a/drivers/net/ethernet/sfc/mcdi_functions.h b/drivers/net/ethernet/sfc/mcdi_functions.h
new file mode 100644
index 000000000000..f0726c71698b
--- /dev/null
+++ b/drivers/net/ethernet/sfc/mcdi_functions.h
@@ -0,0 +1,30 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/****************************************************************************
+ * Driver for Solarflare network controllers and boards
+ * Copyright 2018 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+#ifndef EFX_MCDI_FUNCTIONS_H
+#define EFX_MCDI_FUNCTIONS_H
+
+int efx_mcdi_alloc_vis(struct efx_nic *efx, unsigned int min_vis,
+		       unsigned int max_vis, unsigned int *vi_base,
+		       unsigned int *allocated_vis);
+int efx_mcdi_free_vis(struct efx_nic *efx);
+
+int efx_mcdi_ev_probe(struct efx_channel *channel);
+int efx_mcdi_ev_init(struct efx_channel *channel, bool v1_cut_thru, bool v2);
+void efx_mcdi_ev_remove(struct efx_channel *channel);
+void efx_mcdi_ev_fini(struct efx_channel *channel);
+int efx_mcdi_tx_init(struct efx_tx_queue *tx_queue, bool tso_v2);
+void efx_mcdi_tx_remove(struct efx_tx_queue *tx_queue);
+void efx_mcdi_tx_fini(struct efx_tx_queue *tx_queue);
+int efx_mcdi_rx_probe(struct efx_rx_queue *rx_queue);
+int efx_mcdi_rx_init(struct efx_rx_queue *rx_queue, bool want_outer_classes);
+void efx_mcdi_rx_remove(struct efx_rx_queue *rx_queue);
+void efx_mcdi_rx_fini(struct efx_rx_queue *rx_queue);
+
+#endif
diff --git a/drivers/net/ethernet/sfc/mcdi_port.c b/drivers/net/ethernet/sfc/mcdi_port.c
index fb7cde4980ed..f19d7b8a2935 100644
--- a/drivers/net/ethernet/sfc/mcdi_port.c
+++ b/drivers/net/ethernet/sfc/mcdi_port.c
@@ -14,23 +14,9 @@
 #include "mcdi_pcol.h"
 #include "nic.h"
 #include "selftest.h"
+#include "mcdi_port_common.h"
 
-struct efx_mcdi_phy_data {
-	u32 flags;
-	u32 type;
-	u32 supported_cap;
-	u32 channel;
-	u32 port;
-	u32 stats_mask;
-	u8 name[20];
-	u32 media;
-	u32 mmd_mask;
-	u8 revision[20];
-	u32 forced_cap;
-};
-
-static int
-efx_mcdi_get_phy_cfg(struct efx_nic *efx, struct efx_mcdi_phy_data *cfg)
+int efx_mcdi_get_phy_cfg(struct efx_nic *efx, struct efx_mcdi_phy_data *cfg)
 {
 	MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_PHY_CFG_OUT_LEN);
 	size_t outlen;
@@ -70,9 +56,9 @@ fail:
 	return rc;
 }
 
-static int efx_mcdi_set_link(struct efx_nic *efx, u32 capabilities,
-			     u32 flags, u32 loopback_mode,
-			     u32 loopback_speed)
+int efx_mcdi_set_link(struct efx_nic *efx, u32 capabilities,
+		      u32 flags, u32 loopback_mode,
+		      u32 loopback_speed)
 {
 	MCDI_DECLARE_BUF(inbuf, MC_CMD_SET_LINK_IN_LEN);
 	int rc;
@@ -89,7 +75,7 @@ static int efx_mcdi_set_link(struct efx_nic *efx, u32 capabilities,
 	return rc;
 }
 
-static int efx_mcdi_loopback_modes(struct efx_nic *efx, u64 *loopback_modes)
+int efx_mcdi_loopback_modes(struct efx_nic *efx, u64 *loopback_modes)
 {
 	MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_LOOPBACK_MODES_OUT_LEN);
 	size_t outlen;
@@ -168,7 +154,7 @@ static int efx_mcdi_mdio_write(struct net_device *net_dev,
 	return 0;
 }
 
-static void mcdi_to_ethtool_linkset(u32 media, u32 cap, unsigned long *linkset)
+void mcdi_to_ethtool_linkset(u32 media, u32 cap, unsigned long *linkset)
 {
 	#define SET_BIT(name)	__set_bit(ETHTOOL_LINK_MODE_ ## name ## _BIT, \
 					  linkset)
@@ -232,7 +218,7 @@ static void mcdi_to_ethtool_linkset(u32 media, u32 cap, unsigned long *linkset)
 	#undef SET_BIT
 }
 
-static u32 ethtool_linkset_to_mcdi_cap(const unsigned long *linkset)
+u32 ethtool_linkset_to_mcdi_cap(const unsigned long *linkset)
 {
 	u32 result = 0;
 
@@ -273,7 +259,7 @@ static u32 ethtool_linkset_to_mcdi_cap(const unsigned long *linkset)
 	return result;
 }
 
-static u32 efx_get_mcdi_phy_flags(struct efx_nic *efx)
+u32 efx_get_mcdi_phy_flags(struct efx_nic *efx)
 {
 	struct efx_mcdi_phy_data *phy_cfg = efx->phy_data;
 	enum efx_phy_mode mode, supported;
@@ -301,7 +287,7 @@ static u32 efx_get_mcdi_phy_flags(struct efx_nic *efx)
 	return flags;
 }
 
-static u8 mcdi_to_ethtool_media(u32 media)
+u8 mcdi_to_ethtool_media(u32 media)
 {
 	switch (media) {
 	case MC_CMD_MEDIA_XAUI:
@@ -322,7 +308,7 @@ static u8 mcdi_to_ethtool_media(u32 media)
 	}
 }
 
-static void efx_mcdi_phy_decode_link(struct efx_nic *efx,
+void efx_mcdi_phy_decode_link(struct efx_nic *efx,
 			      struct efx_link_state *link_state,
 			      u32 speed, u32 flags, u32 fcntl)
 {
@@ -365,7 +351,7 @@ static void efx_mcdi_phy_decode_link(struct efx_nic *efx,
  * Both RS and BASER (whether AUTO or not) means use FEC if cable and link
  * partner support it, preferring RS to BASER.
  */
-static u32 ethtool_fec_caps_to_mcdi(u32 ethtool_cap)
+u32 ethtool_fec_caps_to_mcdi(u32 ethtool_cap)
 {
 	u32 ret = 0;
 
@@ -392,7 +378,7 @@ static u32 ethtool_fec_caps_to_mcdi(u32 ethtool_cap)
  * maps both of those to AUTO.  This should never matter, and it's not clear
  * what a better mapping would be anyway.
  */
-static u32 mcdi_fec_caps_to_ethtool(u32 caps, bool is_25g)
+u32 mcdi_fec_caps_to_ethtool(u32 caps, bool is_25g)
 {
 	bool rs = caps & (1 << MC_CMD_PHY_CAP_RS_FEC_LBN),
 	     rs_req = caps & (1 << MC_CMD_PHY_CAP_RS_FEC_REQUESTED_LBN),
@@ -530,7 +516,7 @@ int efx_mcdi_port_reconfigure(struct efx_nic *efx)
 /* Verify that the forced flow control settings (!EFX_FC_AUTO) are
  * supported by the link partner. Warn the user if this isn't the case
  */
-static void efx_mcdi_phy_check_fcntl(struct efx_nic *efx, u32 lpa)
+void efx_mcdi_phy_check_fcntl(struct efx_nic *efx, u32 lpa)
 {
 	struct efx_mcdi_phy_data *phy_cfg = efx->phy_data;
 	u32 rmtadv;
@@ -555,7 +541,7 @@ static void efx_mcdi_phy_check_fcntl(struct efx_nic *efx, u32 lpa)
 			  "warning: link partner doesn't support pause frames");
 }
 
-static bool efx_mcdi_phy_poll(struct efx_nic *efx)
+bool efx_mcdi_phy_poll(struct efx_nic *efx)
 {
 	struct efx_link_state old_state = efx->link_state;
 	MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_LINK_OUT_LEN);
@@ -666,8 +652,8 @@ efx_mcdi_phy_set_link_ksettings(struct efx_nic *efx,
 	return 0;
 }
 
-static int efx_mcdi_phy_get_fecparam(struct efx_nic *efx,
-				     struct ethtool_fecparam *fec)
+int efx_mcdi_phy_get_fecparam(struct efx_nic *efx,
+			      struct ethtool_fecparam *fec)
 {
 	MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_LINK_OUT_V2_LEN);
 	u32 caps, active, speed; /* MCDI format */
@@ -745,7 +731,7 @@ static int efx_mcdi_phy_set_fecparam(struct efx_nic *efx,
 	return 0;
 }
 
-static int efx_mcdi_phy_test_alive(struct efx_nic *efx)
+int efx_mcdi_phy_test_alive(struct efx_nic *efx)
 {
 	MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_PHY_STATE_OUT_LEN);
 	size_t outlen;
diff --git a/drivers/net/ethernet/sfc/mcdi_port_common.h b/drivers/net/ethernet/sfc/mcdi_port_common.h
new file mode 100644
index 000000000000..10772de94b2c
--- /dev/null
+++ b/drivers/net/ethernet/sfc/mcdi_port_common.h
@@ -0,0 +1,53 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/****************************************************************************
+ * Driver for Solarflare network controllers and boards
+ * Copyright 2018 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+#ifndef EFX_MCDI_PORT_COMMON_H
+#define EFX_MCDI_PORT_COMMON_H
+
+#include "net_driver.h"
+#include "mcdi.h"
+#include "mcdi_pcol.h"
+
+struct efx_mcdi_phy_data {
+	u32 flags;
+	u32 type;
+	u32 supported_cap;
+	u32 channel;
+	u32 port;
+	u32 stats_mask;
+	u8 name[20];
+	u32 media;
+	u32 mmd_mask;
+	u8 revision[20];
+	u32 forced_cap;
+};
+
+int efx_mcdi_get_phy_cfg(struct efx_nic *efx, struct efx_mcdi_phy_data *cfg);
+void efx_link_set_advertising(struct efx_nic *efx,
+			      const unsigned long *advertising);
+int efx_mcdi_set_link(struct efx_nic *efx, u32 capabilities,
+		      u32 flags, u32 loopback_mode, u32 loopback_speed);
+int efx_mcdi_loopback_modes(struct efx_nic *efx, u64 *loopback_modes);
+void mcdi_to_ethtool_linkset(u32 media, u32 cap, unsigned long *linkset);
+u32 ethtool_linkset_to_mcdi_cap(const unsigned long *linkset);
+u32 efx_get_mcdi_phy_flags(struct efx_nic *efx);
+u8 mcdi_to_ethtool_media(u32 media);
+void efx_mcdi_phy_decode_link(struct efx_nic *efx,
+			      struct efx_link_state *link_state,
+			      u32 speed, u32 flags, u32 fcntl);
+u32 ethtool_fec_caps_to_mcdi(u32 ethtool_cap);
+u32 mcdi_fec_caps_to_ethtool(u32 caps, bool is_25g);
+void efx_mcdi_phy_check_fcntl(struct efx_nic *efx, u32 lpa);
+bool efx_mcdi_phy_poll(struct efx_nic *efx);
+int efx_mcdi_phy_get_fecparam(struct efx_nic *efx,
+			      struct ethtool_fecparam *fec);
+int efx_mcdi_phy_test_alive(struct efx_nic *efx);
+int efx_mcdi_port_get_number(struct efx_nic *efx);
+
+#endif
diff --git a/drivers/net/ethernet/sfc/net_driver.h b/drivers/net/ethernet/sfc/net_driver.h
index 709172a6995e..52e6f11d8818 100644
--- a/drivers/net/ethernet/sfc/net_driver.h
+++ b/drivers/net/ethernet/sfc/net_driver.h
@@ -138,6 +138,8 @@ struct efx_special_buffer {
  *	freed when descriptor completes
  * @xdpf: When @flags & %EFX_TX_BUF_XDP, the XDP frame information; its @data
  *	member is the associated buffer to drop a page reference on.
+ * @option: When @flags & %EFX_TX_BUF_OPTION, an EF10-specific option
+ *	descriptor.
  * @dma_addr: DMA address of the fragment.
  * @flags: Flags for allocation and DMA mapping type
  * @len: Length of this fragment.
@@ -152,7 +154,7 @@ struct efx_tx_buffer {
 		struct xdp_frame *xdpf;
 	};
 	union {
-		efx_qword_t option;
+		efx_qword_t option;    /* EF10 */
 		dma_addr_t dma_addr;
 	};
 	unsigned short flags;
@@ -1610,6 +1612,15 @@ static inline struct efx_rx_buffer *efx_rx_buffer(struct efx_rx_queue *rx_queue,
 	return &rx_queue->buffer[index];
 }
 
+static inline struct efx_rx_buffer *
+efx_rx_buf_next(struct efx_rx_queue *rx_queue, struct efx_rx_buffer *rx_buf)
+{
+	if (unlikely(rx_buf == efx_rx_buffer(rx_queue, rx_queue->ptr_mask)))
+		return efx_rx_buffer(rx_queue, 0);
+	else
+		return rx_buf + 1;
+}
+
 /**
  * EFX_MAX_FRAME_LEN - calculate maximum frame length
  *
diff --git a/drivers/net/ethernet/sfc/nic.h b/drivers/net/ethernet/sfc/nic.h
index bf0bdb22cc64..6670fda8f35a 100644
--- a/drivers/net/ethernet/sfc/nic.h
+++ b/drivers/net/ethernet/sfc/nic.h
@@ -11,6 +11,7 @@
 #include <linux/net_tstamp.h>
 #include "net_driver.h"
 #include "efx.h"
+#include "efx_common.h"
 #include "mcdi.h"
 
 enum {
@@ -505,6 +506,9 @@ static inline void efx_nic_push_buffers(struct efx_tx_queue *tx_queue)
 	tx_queue->efx->type->tx_write(tx_queue);
 }
 
+int efx_enqueue_skb_tso(struct efx_tx_queue *tx_queue, struct sk_buff *skb,
+			bool *data_mapped);
+
 /* RX data path */
 static inline int efx_nic_probe_rx(struct efx_rx_queue *rx_queue)
 {
@@ -553,6 +557,7 @@ static inline void efx_nic_eventq_read_ack(struct efx_channel *channel)
 {
 	channel->efx->type->ev_read_ack(channel);
 }
+
 void efx_nic_event_test_start(struct efx_channel *channel);
 
 /* Falcon/Siena queue operations */
@@ -670,6 +675,7 @@ struct efx_farch_register_test {
 	unsigned address;
 	efx_oword_t mask;
 };
+
 int efx_farch_test_registers(struct efx_nic *efx,
 			     const struct efx_farch_register_test *regs,
 			     size_t n_regs);
diff --git a/drivers/net/ethernet/sfc/rx.c b/drivers/net/ethernet/sfc/rx.c
index c29bf862a94c..26b5c36237fe 100644
--- a/drivers/net/ethernet/sfc/rx.c
+++ b/drivers/net/ethernet/sfc/rx.c
@@ -21,6 +21,7 @@
 #include <linux/bpf_trace.h>
 #include "net_driver.h"
 #include "efx.h"
+#include "rx_common.h"
 #include "filter.h"
 #include "nic.h"
 #include "selftest.h"
@@ -77,15 +78,6 @@ static inline u32 efx_rx_buf_hash(struct efx_nic *efx, const u8 *eh)
 #endif
 }
 
-static inline struct efx_rx_buffer *
-efx_rx_buf_next(struct efx_rx_queue *rx_queue, struct efx_rx_buffer *rx_buf)
-{
-	if (unlikely(rx_buf == efx_rx_buffer(rx_queue, rx_queue->ptr_mask)))
-		return efx_rx_buffer(rx_queue, 0);
-	else
-		return rx_buf + 1;
-}
-
 static inline void efx_sync_rx_buffer(struct efx_nic *efx,
 				      struct efx_rx_buffer *rx_buf,
 				      unsigned int len)
@@ -152,7 +144,7 @@ static struct page *efx_reuse_page(struct efx_rx_queue *rx_queue)
  * 0 on success. If a single page can be used for multiple buffers,
  * then the page will either be inserted fully, or not at all.
  */
-static int efx_init_rx_buffers(struct efx_rx_queue *rx_queue, bool atomic)
+int efx_init_rx_buffers(struct efx_rx_queue *rx_queue, bool atomic)
 {
 	struct efx_nic *efx = rx_queue->efx;
 	struct efx_rx_buffer *rx_buf;
@@ -215,8 +207,8 @@ static int efx_init_rx_buffers(struct efx_rx_queue *rx_queue, bool atomic)
 /* Unmap a DMA-mapped page.  This function is only called for the final RX
  * buffer in a page.
  */
-static void efx_unmap_rx_buffer(struct efx_nic *efx,
-				struct efx_rx_buffer *rx_buf)
+void efx_unmap_rx_buffer(struct efx_nic *efx,
+			 struct efx_rx_buffer *rx_buf)
 {
 	struct page *page = rx_buf->page;
 
@@ -229,9 +221,9 @@ static void efx_unmap_rx_buffer(struct efx_nic *efx,
 	}
 }
 
-static void efx_free_rx_buffers(struct efx_rx_queue *rx_queue,
-				struct efx_rx_buffer *rx_buf,
-				unsigned int num_bufs)
+void efx_free_rx_buffers(struct efx_rx_queue *rx_queue,
+			 struct efx_rx_buffer *rx_buf,
+			 unsigned int num_bufs)
 {
 	do {
 		if (rx_buf->page) {
diff --git a/drivers/net/ethernet/sfc/rx_common.h b/drivers/net/ethernet/sfc/rx_common.h
new file mode 100644
index 000000000000..8b23a7accea1
--- /dev/null
+++ b/drivers/net/ethernet/sfc/rx_common.h
@@ -0,0 +1,42 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/****************************************************************************
+ * Driver for Solarflare network controllers and boards
+ * Copyright 2018 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#ifndef EFX_RX_COMMON_H
+#define EFX_RX_COMMON_H
+
+/* Preferred number of descriptors to fill at once */
+#define EFX_RX_PREFERRED_BATCH 8U
+
+/* Each packet can consume up to ceil(max_frame_len / buffer_size) buffers */
+#define EFX_RX_MAX_FRAGS DIV_ROUND_UP(EFX_MAX_FRAME_LEN(EFX_MAX_MTU), \
+				      EFX_RX_USR_BUF_SIZE)
+
+void efx_rx_slow_fill(struct timer_list *t);
+
+int efx_probe_rx_queue(struct efx_rx_queue *rx_queue);
+void efx_init_rx_queue(struct efx_rx_queue *rx_queue);
+void efx_fini_rx_queue(struct efx_rx_queue *rx_queue);
+void efx_remove_rx_queue(struct efx_rx_queue *rx_queue);
+void efx_destroy_rx_queue(struct efx_rx_queue *rx_queue);
+
+void efx_init_rx_buffer(struct efx_rx_queue *rx_queue,
+			struct page *page,
+			unsigned int page_offset,
+			u16 flags);
+void efx_unmap_rx_buffer(struct efx_nic *efx, struct efx_rx_buffer *rx_buf);
+void efx_free_rx_buffers(struct efx_rx_queue *rx_queue,
+			 struct efx_rx_buffer *rx_buf,
+			 unsigned int num_bufs);
+
+void efx_schedule_slow_fill(struct efx_rx_queue *rx_queue);
+void efx_rx_config_page_split(struct efx_nic *efx);
+void efx_fast_push_rx_descriptors(struct efx_rx_queue *rx_queue, bool atomic);
+
+#endif
diff --git a/drivers/net/ethernet/sfc/selftest.c b/drivers/net/ethernet/sfc/selftest.c
index 8474cf8ea7d3..ec5002b5ec4f 100644
--- a/drivers/net/ethernet/sfc/selftest.c
+++ b/drivers/net/ethernet/sfc/selftest.c
@@ -18,6 +18,8 @@
 #include <linux/slab.h>
 #include "net_driver.h"
 #include "efx.h"
+#include "efx_common.h"
+#include "efx_channels.h"
 #include "nic.h"
 #include "selftest.h"
 #include "workarounds.h"
diff --git a/drivers/net/ethernet/sfc/siena.c b/drivers/net/ethernet/sfc/siena.c
index 81499244a4b4..810f6fc8a937 100644
--- a/drivers/net/ethernet/sfc/siena.c
+++ b/drivers/net/ethernet/sfc/siena.c
@@ -14,6 +14,7 @@
 #include "net_driver.h"
 #include "bitfield.h"
 #include "efx.h"
+#include "efx_common.h"
 #include "nic.h"
 #include "farch_regs.h"
 #include "io.h"
diff --git a/drivers/net/ethernet/sfc/tx.c b/drivers/net/ethernet/sfc/tx.c
index 00c1c4402451..1c30354e098c 100644
--- a/drivers/net/ethernet/sfc/tx.c
+++ b/drivers/net/ethernet/sfc/tx.c
@@ -20,6 +20,7 @@
 #include "io.h"
 #include "nic.h"
 #include "tx.h"
+#include "tx_common.h"
 #include "workarounds.h"
 #include "ef10_regs.h"
 
@@ -56,10 +57,10 @@ u8 *efx_tx_get_copy_buffer_limited(struct efx_tx_queue *tx_queue,
 	return efx_tx_get_copy_buffer(tx_queue, buffer);
 }
 
-static void efx_dequeue_buffer(struct efx_tx_queue *tx_queue,
-			       struct efx_tx_buffer *buffer,
-			       unsigned int *pkts_compl,
-			       unsigned int *bytes_compl)
+void efx_dequeue_buffer(struct efx_tx_queue *tx_queue,
+			struct efx_tx_buffer *buffer,
+			unsigned int *pkts_compl,
+			unsigned int *bytes_compl)
 {
 	if (buffer->unmap_len) {
 		struct device *dma_dev = &tx_queue->efx->pci_dev->dev;
@@ -333,9 +334,9 @@ static int efx_enqueue_skb_pio(struct efx_tx_queue *tx_queue,
 }
 #endif /* EFX_USE_PIO */
 
-static struct efx_tx_buffer *efx_tx_map_chunk(struct efx_tx_queue *tx_queue,
-					      dma_addr_t dma_addr,
-					      size_t len)
+struct efx_tx_buffer *efx_tx_map_chunk(struct efx_tx_queue *tx_queue,
+				       dma_addr_t dma_addr,
+				       size_t len)
 {
 	const struct efx_nic_type *nic_type = tx_queue->efx->type;
 	struct efx_tx_buffer *buffer;
@@ -359,8 +360,8 @@ static struct efx_tx_buffer *efx_tx_map_chunk(struct efx_tx_queue *tx_queue,
 
 /* Map all data from an SKB for DMA and create descriptors on the queue.
  */
-static int efx_tx_map_data(struct efx_tx_queue *tx_queue, struct sk_buff *skb,
-			   unsigned int segment_count)
+int efx_tx_map_data(struct efx_tx_queue *tx_queue, struct sk_buff *skb,
+		    unsigned int segment_count)
 {
 	struct efx_nic *efx = tx_queue->efx;
 	struct device *dma_dev = &efx->pci_dev->dev;
diff --git a/drivers/net/ethernet/sfc/tx_common.h b/drivers/net/ethernet/sfc/tx_common.h
new file mode 100644
index 000000000000..afdfc79a8ea0
--- /dev/null
+++ b/drivers/net/ethernet/sfc/tx_common.h
@@ -0,0 +1,31 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/****************************************************************************
+ * Driver for Solarflare network controllers and boards
+ * Copyright 2018 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#ifndef EFX_TX_COMMON_H
+#define EFX_TX_COMMON_H
+
+int efx_probe_tx_queue(struct efx_tx_queue *tx_queue);
+void efx_init_tx_queue(struct efx_tx_queue *tx_queue);
+void efx_fini_tx_queue(struct efx_tx_queue *tx_queue);
+void efx_remove_tx_queue(struct efx_tx_queue *tx_queue);
+
+void efx_dequeue_buffer(struct efx_tx_queue *tx_queue,
+			struct efx_tx_buffer *buffer,
+			unsigned int *pkts_compl,
+			unsigned int *bytes_compl);
+
+struct efx_tx_buffer *efx_tx_map_chunk(struct efx_tx_queue *tx_queue,
+				       dma_addr_t dma_addr, size_t len);
+int efx_tx_map_data(struct efx_tx_queue *tx_queue, struct sk_buff *skb,
+		    unsigned int segment_count);
+
+unsigned int efx_tx_max_skb_descs(struct efx_nic *efx);
+
+#endif
-- 
cgit v1.2.3-59-g8ed1b


From b194045114c234f73a6c3831fbcde0f4161bddf0 Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Wed, 8 Jan 2020 16:10:45 +0000
Subject: sfc: further preparation for code split

Added more arguments for a couple of functions.
Also moved a function to the common header.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/efx.c        | 31 ++++++++-----------------------
 drivers/net/ethernet/sfc/efx_common.h | 15 ++++++++++++---
 2 files changed, 20 insertions(+), 26 deletions(-)

diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index c5bcdfcfee87..ce8c0db2ba4b 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -233,16 +233,6 @@ static int efx_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **xdpfs,
 			ASSERT_RTNL();			\
 	} while (0)
 
-int efx_check_disabled(struct efx_nic *efx)
-{
-	if (efx->state == STATE_DISABLED || efx->state == STATE_RECOVERY) {
-		netif_err(efx, drv, efx->net_dev,
-			  "device is disabled due to earlier errors\n");
-		return -EIO;
-	}
-	return 0;
-}
-
 /**************************************************************************
  *
  * Event queue processing
@@ -1283,17 +1273,14 @@ static void efx_dissociate(struct efx_nic *efx)
 }
 
 /* This configures the PCI device to enable I/O and DMA. */
-int efx_init_io(struct efx_nic *efx)
+int efx_init_io(struct efx_nic *efx, int bar, dma_addr_t dma_mask,
+		unsigned int mem_map_size)
 {
 	struct pci_dev *pci_dev = efx->pci_dev;
-	dma_addr_t dma_mask = efx->type->max_dma_mask;
-	unsigned int mem_map_size = efx->type->mem_map_size(efx);
-	int rc, bar;
+	int rc;
 
 	netif_dbg(efx, probe, efx->net_dev, "initialising I/O\n");
 
-	bar = efx->type->mem_bar(efx);
-
 	rc = pci_enable_device(pci_dev);
 	if (rc) {
 		netif_err(efx, probe, efx->net_dev,
@@ -1354,10 +1341,8 @@ int efx_init_io(struct efx_nic *efx)
 	return rc;
 }
 
-void efx_fini_io(struct efx_nic *efx)
+void efx_fini_io(struct efx_nic *efx, int bar)
 {
-	int bar;
-
 	netif_dbg(efx, drv, efx->net_dev, "shutting down I/O\n");
 
 	if (efx->membase) {
@@ -1366,7 +1351,6 @@ void efx_fini_io(struct efx_nic *efx)
 	}
 
 	if (efx->membase_phys) {
-		bar = efx->type->mem_bar(efx);
 		pci_release_region(efx->pci_dev, bar);
 		efx->membase_phys = 0;
 	}
@@ -3548,7 +3532,7 @@ static void efx_pci_remove(struct pci_dev *pci_dev)
 
 	efx_pci_remove_main(efx);
 
-	efx_fini_io(efx);
+	efx_fini_io(efx, efx->type->mem_bar(efx));
 	netif_dbg(efx, drv, efx->net_dev, "shutdown successful\n");
 
 	efx_fini_struct(efx);
@@ -3771,7 +3755,8 @@ static int efx_pci_probe(struct pci_dev *pci_dev,
 		efx_probe_vpd_strings(efx);
 
 	/* Set up basic I/O (BAR mappings etc) */
-	rc = efx_init_io(efx);
+	rc = efx_init_io(efx, efx->type->mem_bar(efx), efx->type->max_dma_mask,
+			 efx->type->mem_map_size(efx));
 	if (rc)
 		goto fail2;
 
@@ -3815,7 +3800,7 @@ static int efx_pci_probe(struct pci_dev *pci_dev,
 	return 0;
 
  fail3:
-	efx_fini_io(efx);
+	efx_fini_io(efx, efx->type->mem_bar(efx));
  fail2:
 	efx_fini_struct(efx);
  fail1:
diff --git a/drivers/net/ethernet/sfc/efx_common.h b/drivers/net/ethernet/sfc/efx_common.h
index cb690d01adbc..c602e5257088 100644
--- a/drivers/net/ethernet/sfc/efx_common.h
+++ b/drivers/net/ethernet/sfc/efx_common.h
@@ -11,8 +11,9 @@
 #ifndef EFX_COMMON_H
 #define EFX_COMMON_H
 
-int efx_init_io(struct efx_nic *efx);
-void efx_fini_io(struct efx_nic *efx);
+int efx_init_io(struct efx_nic *efx, int bar, dma_addr_t dma_mask,
+		unsigned int mem_map_size);
+void efx_fini_io(struct efx_nic *efx, int bar);
 int efx_init_struct(struct efx_nic *efx, struct pci_dev *pci_dev,
 		    struct net_device *net_dev);
 void efx_fini_struct(struct efx_nic *efx);
@@ -44,7 +45,15 @@ int efx_reset_up(struct efx_nic *efx, enum reset_type method, bool ok);
 int efx_reset(struct efx_nic *efx, enum reset_type method);
 void efx_schedule_reset(struct efx_nic *efx, enum reset_type type);
 
-int efx_check_disabled(struct efx_nic *efx);
+static inline int efx_check_disabled(struct efx_nic *efx)
+{
+	if (efx->state == STATE_DISABLED || efx->state == STATE_RECOVERY) {
+		netif_err(efx, drv, efx->net_dev,
+			  "device is disabled due to earlier errors\n");
+		return -EIO;
+	}
+	return 0;
+}
 
 void efx_mac_reconfigure(struct efx_nic *efx);
 void efx_link_status_changed(struct efx_nic *efx);
-- 
cgit v1.2.3-59-g8ed1b


From 82c64484024af04489bb1870bd1c6b61c603ca78 Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Wed, 8 Jan 2020 16:10:59 +0000
Subject: sfc: move reset workqueue code

Small functions doing work that will be common, related to reset
workqueue management.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/Makefile     |  7 +++--
 drivers/net/ethernet/sfc/efx.c        | 22 +++++--------
 drivers/net/ethernet/sfc/efx_common.c | 58 +++++++++++++++++++++++++++++++++++
 3 files changed, 70 insertions(+), 17 deletions(-)
 create mode 100644 drivers/net/ethernet/sfc/efx_common.c

diff --git a/drivers/net/ethernet/sfc/Makefile b/drivers/net/ethernet/sfc/Makefile
index c5c297e78d06..7022cffa31f8 100644
--- a/drivers/net/ethernet/sfc/Makefile
+++ b/drivers/net/ethernet/sfc/Makefile
@@ -1,7 +1,10 @@
 # SPDX-License-Identifier: GPL-2.0
-sfc-y			+= efx.o nic.o farch.o siena.o ef10.o tx.o rx.o \
+sfc-y			+= efx.o efx_common.o nic.o \
+			   farch.o siena.o ef10.o \
+			   tx.o rx.o \
 			   selftest.o ethtool.o ptp.o tx_tso.o \
-			   mcdi.o mcdi_port.o mcdi_mon.o
+			   mcdi.o mcdi_port.o \
+			   mcdi_mon.o
 sfc-$(CONFIG_SFC_MTD)	+= mtd.o
 sfc-$(CONFIG_SFC_SRIOV)	+= sriov.o siena_sriov.o ef10_sriov.o
 
diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index ce8c0db2ba4b..c881e35b0477 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -108,12 +108,6 @@ void efx_get_udp_tunnel_type_name(u16 type, char *buf, size_t buflen)
 		snprintf(buf, buflen, "type %d", type);
 }
 
-/* Reset workqueue. If any NIC has a hardware failure then a reset will be
- * queued onto this work queue. This is not a per-nic work queue, because
- * efx_reset_work() acquires the rtnl lock, so resets are naturally serialised.
- */
-static struct workqueue_struct *reset_workqueue;
-
 /* How often and how many times to poll for a reset while waiting for a
  * BIST that another function started to complete.
  */
@@ -3106,7 +3100,7 @@ void efx_schedule_reset(struct efx_nic *efx, enum reset_type type)
 	 * reset is scheduled. So switch back to poll'd MCDI completions. */
 	efx_mcdi_mode_poll(efx);
 
-	queue_work(reset_workqueue, &efx->reset_work);
+	efx_queue_reset_work(efx);
 }
 
 /**************************************************************************
@@ -3492,7 +3486,7 @@ static void efx_pci_remove_main(struct efx_nic *efx)
 	 * are not READY.
 	 */
 	BUG_ON(efx->state == STATE_READY);
-	cancel_work_sync(&efx->reset_work);
+	efx_flush_reset_workqueue(efx);
 
 	efx_disable_interrupts(efx);
 	efx_clear_interrupt_affinity(efx);
@@ -3878,7 +3872,7 @@ static int efx_pm_thaw(struct device *dev)
 	rtnl_unlock();
 
 	/* Reschedule any quenched resets scheduled during efx_pm_freeze() */
-	queue_work(reset_workqueue, &efx->reset_work);
+	efx_queue_reset_work(efx);
 
 	return 0;
 
@@ -4077,11 +4071,9 @@ static int __init efx_init_module(void)
 		goto err_sriov;
 #endif
 
-	reset_workqueue = create_singlethread_workqueue("sfc_reset");
-	if (!reset_workqueue) {
-		rc = -ENOMEM;
+	rc = efx_create_reset_workqueue();
+	if (rc)
 		goto err_reset;
-	}
 
 	rc = pci_register_driver(&efx_pci_driver);
 	if (rc < 0)
@@ -4090,7 +4082,7 @@ static int __init efx_init_module(void)
 	return 0;
 
  err_pci:
-	destroy_workqueue(reset_workqueue);
+	efx_destroy_reset_workqueue();
  err_reset:
 #ifdef CONFIG_SFC_SRIOV
 	efx_fini_sriov();
@@ -4106,7 +4098,7 @@ static void __exit efx_exit_module(void)
 	printk(KERN_INFO "Solarflare NET driver unloading\n");
 
 	pci_unregister_driver(&efx_pci_driver);
-	destroy_workqueue(reset_workqueue);
+	efx_destroy_reset_workqueue();
 #ifdef CONFIG_SFC_SRIOV
 	efx_fini_sriov();
 #endif
diff --git a/drivers/net/ethernet/sfc/efx_common.c b/drivers/net/ethernet/sfc/efx_common.c
new file mode 100644
index 000000000000..5cadfba37fc4
--- /dev/null
+++ b/drivers/net/ethernet/sfc/efx_common.c
@@ -0,0 +1,58 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/****************************************************************************
+ * Driver for Solarflare network controllers and boards
+ * Copyright 2018 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#include "net_driver.h"
+#include <linux/module.h>
+#include <linux/netdevice.h>
+#include "efx_common.h"
+#include "efx_channels.h"
+#include "efx.h"
+#include "mcdi.h"
+#include "selftest.h"
+#include "rx_common.h"
+#include "tx_common.h"
+#include "nic.h"
+#include "io.h"
+#include "mcdi_pcol.h"
+
+/* Reset workqueue. If any NIC has a hardware failure then a reset will be
+ * queued onto this work queue. This is not a per-nic work queue, because
+ * efx_reset_work() acquires the rtnl lock, so resets are naturally serialised.
+ */
+static struct workqueue_struct *reset_workqueue;
+
+int efx_create_reset_workqueue(void)
+{
+	reset_workqueue = create_singlethread_workqueue("sfc_reset");
+	if (!reset_workqueue) {
+		printk(KERN_ERR "Failed to create reset workqueue\n");
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+void efx_queue_reset_work(struct efx_nic *efx)
+{
+	queue_work(reset_workqueue, &efx->reset_work);
+}
+
+void efx_flush_reset_workqueue(struct efx_nic *efx)
+{
+	cancel_work_sync(&efx->reset_work);
+}
+
+void efx_destroy_reset_workqueue(void)
+{
+	if (reset_workqueue) {
+		destroy_workqueue(reset_workqueue);
+		reset_workqueue = NULL;
+	}
+}
-- 
cgit v1.2.3-59-g8ed1b


From 473f5ede416c749f25ed5eb531b2fec56d518426 Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Wed, 8 Jan 2020 16:11:13 +0000
Subject: sfc: move mac configuration and status functions

Two small functions with different purposes.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/efx.c        | 44 ----------------------------------
 drivers/net/ethernet/sfc/efx_common.c | 45 +++++++++++++++++++++++++++++++++++
 2 files changed, 45 insertions(+), 44 deletions(-)

diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index c881e35b0477..752470baf4be 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -935,40 +935,6 @@ void efx_channel_dummy_op_void(struct efx_channel *channel)
  *
  **************************************************************************/
 
-/* This ensures that the kernel is kept informed (via
- * netif_carrier_on/off) of the link status, and also maintains the
- * link status's stop on the port's TX queue.
- */
-void efx_link_status_changed(struct efx_nic *efx)
-{
-	struct efx_link_state *link_state = &efx->link_state;
-
-	/* SFC Bug 5356: A net_dev notifier is registered, so we must ensure
-	 * that no events are triggered between unregister_netdev() and the
-	 * driver unloading. A more general condition is that NETDEV_CHANGE
-	 * can only be generated between NETDEV_UP and NETDEV_DOWN */
-	if (!netif_running(efx->net_dev))
-		return;
-
-	if (link_state->up != netif_carrier_ok(efx->net_dev)) {
-		efx->n_link_state_changes++;
-
-		if (link_state->up)
-			netif_carrier_on(efx->net_dev);
-		else
-			netif_carrier_off(efx->net_dev);
-	}
-
-	/* Status message for kernel log */
-	if (link_state->up)
-		netif_info(efx, link, efx->net_dev,
-			   "link up at %uMbps %s-duplex (MTU %d)\n",
-			   link_state->speed, link_state->fd ? "full" : "half",
-			   efx->net_dev->mtu);
-	else
-		netif_info(efx, link, efx->net_dev, "link down\n");
-}
-
 void efx_link_set_advertising(struct efx_nic *efx,
 			      const unsigned long *advertising)
 {
@@ -1010,16 +976,6 @@ void efx_link_set_wanted_fc(struct efx_nic *efx, u8 wanted_fc)
 
 static void efx_fini_port(struct efx_nic *efx);
 
-/* We assume that efx->type->reconfigure_mac will always try to sync RX
- * filters and therefore needs to read-lock the filter table against freeing
- */
-void efx_mac_reconfigure(struct efx_nic *efx)
-{
-	down_read(&efx->filter_sem);
-	efx->type->reconfigure_mac(efx);
-	up_read(&efx->filter_sem);
-}
-
 /* Push loopback/power/transmit disable settings to the PHY, and reconfigure
  * the MAC appropriately. All other PHY configuration changes are pushed
  * through phy_op->set_settings(), and pushed asynchronously to the MAC
diff --git a/drivers/net/ethernet/sfc/efx_common.c b/drivers/net/ethernet/sfc/efx_common.c
index 5cadfba37fc4..be8e80c9d513 100644
--- a/drivers/net/ethernet/sfc/efx_common.c
+++ b/drivers/net/ethernet/sfc/efx_common.c
@@ -56,3 +56,48 @@ void efx_destroy_reset_workqueue(void)
 		reset_workqueue = NULL;
 	}
 }
+
+/* We assume that efx->type->reconfigure_mac will always try to sync RX
+ * filters and therefore needs to read-lock the filter table against freeing
+ */
+void efx_mac_reconfigure(struct efx_nic *efx)
+{
+	down_read(&efx->filter_sem);
+	efx->type->reconfigure_mac(efx);
+	up_read(&efx->filter_sem);
+}
+
+/* This ensures that the kernel is kept informed (via
+ * netif_carrier_on/off) of the link status, and also maintains the
+ * link status's stop on the port's TX queue.
+ */
+void efx_link_status_changed(struct efx_nic *efx)
+{
+	struct efx_link_state *link_state = &efx->link_state;
+
+	/* SFC Bug 5356: A net_dev notifier is registered, so we must ensure
+	 * that no events are triggered between unregister_netdev() and the
+	 * driver unloading. A more general condition is that NETDEV_CHANGE
+	 * can only be generated between NETDEV_UP and NETDEV_DOWN
+	 */
+	if (!netif_running(efx->net_dev))
+		return;
+
+	if (link_state->up != netif_carrier_ok(efx->net_dev)) {
+		efx->n_link_state_changes++;
+
+		if (link_state->up)
+			netif_carrier_on(efx->net_dev);
+		else
+			netif_carrier_off(efx->net_dev);
+	}
+
+	/* Status message for kernel log */
+	if (link_state->up)
+		netif_info(efx, link, efx->net_dev,
+			   "link up at %uMbps %s-duplex (MTU %d)\n",
+			   link_state->speed, link_state->fd ? "full" : "half",
+			   efx->net_dev->mtu);
+	else
+		netif_info(efx, link, efx->net_dev, "link down\n");
+}
-- 
cgit v1.2.3-59-g8ed1b


From 7ec3de426014a0fff8f2fe1a3f5a7be08e416903 Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Wed, 8 Jan 2020 16:11:29 +0000
Subject: sfc: move datapath management code

The code that manages the datapath (starting, stopping, including the
port-related bits) will be common.
Three functions have been added that contain bits from other
functions. These will be moved to their final files in later patches.
Small code styling fixes included.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/efx.c        | 310 +-------------------------------
 drivers/net/ethernet/sfc/efx_common.c | 328 ++++++++++++++++++++++++++++++++++
 2 files changed, 335 insertions(+), 303 deletions(-)

diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index 752470baf4be..988687a80abe 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -618,152 +618,6 @@ fail:
 	return rc;
 }
 
-/* Channels are shutdown and reinitialised whilst the NIC is running
- * to propagate configuration changes (mtu, checksum offload), or
- * to clear hardware error conditions
- */
-static void efx_start_datapath(struct efx_nic *efx)
-{
-	netdev_features_t old_features = efx->net_dev->features;
-	bool old_rx_scatter = efx->rx_scatter;
-	struct efx_tx_queue *tx_queue;
-	struct efx_rx_queue *rx_queue;
-	struct efx_channel *channel;
-	size_t rx_buf_len;
-
-	/* Calculate the rx buffer allocation parameters required to
-	 * support the current MTU, including padding for header
-	 * alignment and overruns.
-	 */
-	efx->rx_dma_len = (efx->rx_prefix_size +
-			   EFX_MAX_FRAME_LEN(efx->net_dev->mtu) +
-			   efx->type->rx_buffer_padding);
-	rx_buf_len = (sizeof(struct efx_rx_page_state) + XDP_PACKET_HEADROOM +
-		      efx->rx_ip_align + efx->rx_dma_len);
-	if (rx_buf_len <= PAGE_SIZE) {
-		efx->rx_scatter = efx->type->always_rx_scatter;
-		efx->rx_buffer_order = 0;
-	} else if (efx->type->can_rx_scatter) {
-		BUILD_BUG_ON(EFX_RX_USR_BUF_SIZE % L1_CACHE_BYTES);
-		BUILD_BUG_ON(sizeof(struct efx_rx_page_state) +
-			     2 * ALIGN(NET_IP_ALIGN + EFX_RX_USR_BUF_SIZE,
-				       EFX_RX_BUF_ALIGNMENT) >
-			     PAGE_SIZE);
-		efx->rx_scatter = true;
-		efx->rx_dma_len = EFX_RX_USR_BUF_SIZE;
-		efx->rx_buffer_order = 0;
-	} else {
-		efx->rx_scatter = false;
-		efx->rx_buffer_order = get_order(rx_buf_len);
-	}
-
-	efx_rx_config_page_split(efx);
-	if (efx->rx_buffer_order)
-		netif_dbg(efx, drv, efx->net_dev,
-			  "RX buf len=%u; page order=%u batch=%u\n",
-			  efx->rx_dma_len, efx->rx_buffer_order,
-			  efx->rx_pages_per_batch);
-	else
-		netif_dbg(efx, drv, efx->net_dev,
-			  "RX buf len=%u step=%u bpp=%u; page batch=%u\n",
-			  efx->rx_dma_len, efx->rx_page_buf_step,
-			  efx->rx_bufs_per_page, efx->rx_pages_per_batch);
-
-	/* Restore previously fixed features in hw_features and remove
-	 * features which are fixed now
-	 */
-	efx->net_dev->hw_features |= efx->net_dev->features;
-	efx->net_dev->hw_features &= ~efx->fixed_features;
-	efx->net_dev->features |= efx->fixed_features;
-	if (efx->net_dev->features != old_features)
-		netdev_features_change(efx->net_dev);
-
-	/* RX filters may also have scatter-enabled flags */
-	if (efx->rx_scatter != old_rx_scatter)
-		efx->type->filter_update_rx_scatter(efx);
-
-	/* We must keep at least one descriptor in a TX ring empty.
-	 * We could avoid this when the queue size does not exactly
-	 * match the hardware ring size, but it's not that important.
-	 * Therefore we stop the queue when one more skb might fill
-	 * the ring completely.  We wake it when half way back to
-	 * empty.
-	 */
-	efx->txq_stop_thresh = efx->txq_entries - efx_tx_max_skb_descs(efx);
-	efx->txq_wake_thresh = efx->txq_stop_thresh / 2;
-
-	/* Initialise the channels */
-	efx_for_each_channel(channel, efx) {
-		efx_for_each_channel_tx_queue(tx_queue, channel) {
-			efx_init_tx_queue(tx_queue);
-			atomic_inc(&efx->active_queues);
-		}
-
-		efx_for_each_channel_rx_queue(rx_queue, channel) {
-			efx_init_rx_queue(rx_queue);
-			atomic_inc(&efx->active_queues);
-			efx_stop_eventq(channel);
-			efx_fast_push_rx_descriptors(rx_queue, false);
-			efx_start_eventq(channel);
-		}
-
-		WARN_ON(channel->rx_pkt_n_frags);
-	}
-
-	efx_ptp_start_datapath(efx);
-
-	if (netif_device_present(efx->net_dev))
-		netif_tx_wake_all_queues(efx->net_dev);
-}
-
-static void efx_stop_datapath(struct efx_nic *efx)
-{
-	struct efx_channel *channel;
-	struct efx_tx_queue *tx_queue;
-	struct efx_rx_queue *rx_queue;
-	int rc;
-
-	EFX_ASSERT_RESET_SERIALISED(efx);
-	BUG_ON(efx->port_enabled);
-
-	efx_ptp_stop_datapath(efx);
-
-	/* Stop RX refill */
-	efx_for_each_channel(channel, efx) {
-		efx_for_each_channel_rx_queue(rx_queue, channel)
-			rx_queue->refill_enabled = false;
-	}
-
-	efx_for_each_channel(channel, efx) {
-		/* RX packet processing is pipelined, so wait for the
-		 * NAPI handler to complete.  At least event queue 0
-		 * might be kept active by non-data events, so don't
-		 * use napi_synchronize() but actually disable NAPI
-		 * temporarily.
-		 */
-		if (efx_channel_has_rx_queue(channel)) {
-			efx_stop_eventq(channel);
-			efx_start_eventq(channel);
-		}
-	}
-
-	rc = efx->type->fini_dmaq(efx);
-	if (rc) {
-		netif_err(efx, drv, efx->net_dev, "failed to flush queues\n");
-	} else {
-		netif_dbg(efx, drv, efx->net_dev,
-			  "successfully flushed all queues\n");
-	}
-
-	efx_for_each_channel(channel, efx) {
-		efx_for_each_channel_rx_queue(rx_queue, channel)
-			efx_fini_rx_queue(rx_queue);
-		efx_for_each_possible_channel_tx_queue(tx_queue, channel)
-			efx_fini_tx_queue(tx_queue);
-	}
-	efx->xdp_rxq_info_failed = false;
-}
-
 void efx_remove_channel(struct efx_channel *channel)
 {
 	struct efx_tx_queue *tx_queue;
@@ -976,50 +830,6 @@ void efx_link_set_wanted_fc(struct efx_nic *efx, u8 wanted_fc)
 
 static void efx_fini_port(struct efx_nic *efx);
 
-/* Push loopback/power/transmit disable settings to the PHY, and reconfigure
- * the MAC appropriately. All other PHY configuration changes are pushed
- * through phy_op->set_settings(), and pushed asynchronously to the MAC
- * through efx_monitor().
- *
- * Callers must hold the mac_lock
- */
-int __efx_reconfigure_port(struct efx_nic *efx)
-{
-	enum efx_phy_mode phy_mode;
-	int rc;
-
-	WARN_ON(!mutex_is_locked(&efx->mac_lock));
-
-	/* Disable PHY transmit in mac level loopbacks */
-	phy_mode = efx->phy_mode;
-	if (LOOPBACK_INTERNAL(efx))
-		efx->phy_mode |= PHY_MODE_TX_DISABLED;
-	else
-		efx->phy_mode &= ~PHY_MODE_TX_DISABLED;
-
-	rc = efx->type->reconfigure_port(efx);
-
-	if (rc)
-		efx->phy_mode = phy_mode;
-
-	return rc;
-}
-
-/* Reinitialise the MAC to pick up new PHY settings, even if the port is
- * disabled. */
-int efx_reconfigure_port(struct efx_nic *efx)
-{
-	int rc;
-
-	EFX_ASSERT_RESET_SERIALISED(efx);
-
-	mutex_lock(&efx->mac_lock);
-	rc = __efx_reconfigure_port(efx);
-	mutex_unlock(&efx->mac_lock);
-
-	return rc;
-}
-
 /* Asynchronous work item for changing MAC promiscuity and multicast
  * hash.  Avoid a drain/rx_ingress enable by reconfiguring the current
  * MAC directly. */
@@ -1086,44 +896,6 @@ fail1:
 	return rc;
 }
 
-static void efx_start_port(struct efx_nic *efx)
-{
-	netif_dbg(efx, ifup, efx->net_dev, "start port\n");
-	BUG_ON(efx->port_enabled);
-
-	mutex_lock(&efx->mac_lock);
-	efx->port_enabled = true;
-
-	/* Ensure MAC ingress/egress is enabled */
-	efx_mac_reconfigure(efx);
-
-	mutex_unlock(&efx->mac_lock);
-}
-
-/* Cancel work for MAC reconfiguration, periodic hardware monitoring
- * and the async self-test, wait for them to finish and prevent them
- * being scheduled again.  This doesn't cover online resets, which
- * should only be cancelled when removing the device.
- */
-static void efx_stop_port(struct efx_nic *efx)
-{
-	netif_dbg(efx, ifdown, efx->net_dev, "stop port\n");
-
-	EFX_ASSERT_RESET_SERIALISED(efx);
-
-	mutex_lock(&efx->mac_lock);
-	efx->port_enabled = false;
-	mutex_unlock(&efx->mac_lock);
-
-	/* Serialise against efx_set_multicast_list() */
-	netif_addr_lock_bh(efx->net_dev);
-	netif_addr_unlock_bh(efx->net_dev);
-
-	cancel_delayed_work_sync(&efx->monitor_work);
-	efx_selftest_async_cancel(efx);
-	cancel_work_sync(&efx->mac_work);
-}
-
 static void efx_fini_port(struct efx_nic *efx)
 {
 	netif_dbg(efx, drv, efx->net_dev, "shut down port\n");
@@ -1990,81 +1762,6 @@ static int efx_probe_all(struct efx_nic *efx)
 	return rc;
 }
 
-/* If the interface is supposed to be running but is not, start
- * the hardware and software data path, regular activity for the port
- * (MAC statistics, link polling, etc.) and schedule the port to be
- * reconfigured.  Interrupts must already be enabled.  This function
- * is safe to call multiple times, so long as the NIC is not disabled.
- * Requires the RTNL lock.
- */
-void efx_start_all(struct efx_nic *efx)
-{
-	EFX_ASSERT_RESET_SERIALISED(efx);
-	BUG_ON(efx->state == STATE_DISABLED);
-
-	/* Check that it is appropriate to restart the interface. All
-	 * of these flags are safe to read under just the rtnl lock */
-	if (efx->port_enabled || !netif_running(efx->net_dev) ||
-	    efx->reset_pending)
-		return;
-
-	efx_start_port(efx);
-	efx_start_datapath(efx);
-
-	/* Start the hardware monitor if there is one */
-	if (efx->type->monitor != NULL)
-		queue_delayed_work(efx->workqueue, &efx->monitor_work,
-				   efx_monitor_interval);
-
-	/* Link state detection is normally event-driven; we have
-	 * to poll now because we could have missed a change
-	 */
-	mutex_lock(&efx->mac_lock);
-	if (efx->phy_op->poll(efx))
-		efx_link_status_changed(efx);
-	mutex_unlock(&efx->mac_lock);
-
-	efx->type->start_stats(efx);
-	efx->type->pull_stats(efx);
-	spin_lock_bh(&efx->stats_lock);
-	efx->type->update_stats(efx, NULL, NULL);
-	spin_unlock_bh(&efx->stats_lock);
-}
-
-/* Quiesce the hardware and software data path, and regular activity
- * for the port without bringing the link down.  Safe to call multiple
- * times with the NIC in almost any state, but interrupts should be
- * enabled.  Requires the RTNL lock.
- */
-void efx_stop_all(struct efx_nic *efx)
-{
-	EFX_ASSERT_RESET_SERIALISED(efx);
-
-	/* port_enabled can be read safely under the rtnl lock */
-	if (!efx->port_enabled)
-		return;
-
-	/* update stats before we go down so we can accurately count
-	 * rx_nodesc_drops
-	 */
-	efx->type->pull_stats(efx);
-	spin_lock_bh(&efx->stats_lock);
-	efx->type->update_stats(efx, NULL, NULL);
-	spin_unlock_bh(&efx->stats_lock);
-	efx->type->stop_stats(efx);
-	efx_stop_port(efx);
-
-	/* Stop the kernel transmit interface.  This is only valid if
-	 * the device is stopped or detached; otherwise the watchdog
-	 * may fire immediately.
-	 */
-	WARN_ON(netif_running(efx->net_dev) &&
-		netif_device_present(efx->net_dev));
-	netif_tx_disable(efx->net_dev);
-
-	efx_stop_datapath(efx);
-}
-
 static void efx_remove_all(struct efx_nic *efx)
 {
 	rtnl_lock();
@@ -2188,6 +1885,13 @@ static void efx_monitor(struct work_struct *data)
 			   efx_monitor_interval);
 }
 
+void efx_start_monitor(struct efx_nic *efx)
+{
+	if (efx->type->monitor)
+		queue_delayed_work(efx->workqueue, &efx->monitor_work,
+				   efx_monitor_interval);
+}
+
 /**************************************************************************
  *
  * ioctls
diff --git a/drivers/net/ethernet/sfc/efx_common.c b/drivers/net/ethernet/sfc/efx_common.c
index be8e80c9d513..445340c903b1 100644
--- a/drivers/net/ethernet/sfc/efx_common.c
+++ b/drivers/net/ethernet/sfc/efx_common.c
@@ -101,3 +101,331 @@ void efx_link_status_changed(struct efx_nic *efx)
 	else
 		netif_info(efx, link, efx->net_dev, "link down\n");
 }
+
+/**************************************************************************
+ *
+ * Event queue processing
+ *
+ *************************************************************************/
+
+void efx_start_channels(struct efx_nic *efx)
+{
+	struct efx_tx_queue *tx_queue;
+	struct efx_rx_queue *rx_queue;
+	struct efx_channel *channel;
+
+	efx_for_each_channel(channel, efx) {
+		efx_for_each_channel_tx_queue(tx_queue, channel) {
+			efx_init_tx_queue(tx_queue);
+			atomic_inc(&efx->active_queues);
+		}
+
+		efx_for_each_channel_rx_queue(rx_queue, channel) {
+			efx_init_rx_queue(rx_queue);
+			atomic_inc(&efx->active_queues);
+			efx_stop_eventq(channel);
+			efx_fast_push_rx_descriptors(rx_queue, false);
+			efx_start_eventq(channel);
+		}
+
+		WARN_ON(channel->rx_pkt_n_frags);
+	}
+}
+
+/* Channels are shutdown and reinitialised whilst the NIC is running
+ * to propagate configuration changes (mtu, checksum offload), or
+ * to clear hardware error conditions
+ */
+static void efx_start_datapath(struct efx_nic *efx)
+{
+	netdev_features_t old_features = efx->net_dev->features;
+	bool old_rx_scatter = efx->rx_scatter;
+	size_t rx_buf_len;
+
+	/* Calculate the rx buffer allocation parameters required to
+	 * support the current MTU, including padding for header
+	 * alignment and overruns.
+	 */
+	efx->rx_dma_len = (efx->rx_prefix_size +
+			   EFX_MAX_FRAME_LEN(efx->net_dev->mtu) +
+			   efx->type->rx_buffer_padding);
+	rx_buf_len = (sizeof(struct efx_rx_page_state) + XDP_PACKET_HEADROOM +
+		      efx->rx_ip_align + efx->rx_dma_len);
+	if (rx_buf_len <= PAGE_SIZE) {
+		efx->rx_scatter = efx->type->always_rx_scatter;
+		efx->rx_buffer_order = 0;
+	} else if (efx->type->can_rx_scatter) {
+		BUILD_BUG_ON(EFX_RX_USR_BUF_SIZE % L1_CACHE_BYTES);
+		BUILD_BUG_ON(sizeof(struct efx_rx_page_state) +
+			     2 * ALIGN(NET_IP_ALIGN + EFX_RX_USR_BUF_SIZE,
+				       EFX_RX_BUF_ALIGNMENT) >
+			     PAGE_SIZE);
+		efx->rx_scatter = true;
+		efx->rx_dma_len = EFX_RX_USR_BUF_SIZE;
+		efx->rx_buffer_order = 0;
+	} else {
+		efx->rx_scatter = false;
+		efx->rx_buffer_order = get_order(rx_buf_len);
+	}
+
+	efx_rx_config_page_split(efx);
+	if (efx->rx_buffer_order)
+		netif_dbg(efx, drv, efx->net_dev,
+			  "RX buf len=%u; page order=%u batch=%u\n",
+			  efx->rx_dma_len, efx->rx_buffer_order,
+			  efx->rx_pages_per_batch);
+	else
+		netif_dbg(efx, drv, efx->net_dev,
+			  "RX buf len=%u step=%u bpp=%u; page batch=%u\n",
+			  efx->rx_dma_len, efx->rx_page_buf_step,
+			  efx->rx_bufs_per_page, efx->rx_pages_per_batch);
+
+	/* Restore previously fixed features in hw_features and remove
+	 * features which are fixed now
+	 */
+	efx->net_dev->hw_features |= efx->net_dev->features;
+	efx->net_dev->hw_features &= ~efx->fixed_features;
+	efx->net_dev->features |= efx->fixed_features;
+	if (efx->net_dev->features != old_features)
+		netdev_features_change(efx->net_dev);
+
+	/* RX filters may also have scatter-enabled flags */
+	if (efx->rx_scatter != old_rx_scatter)
+		efx->type->filter_update_rx_scatter(efx);
+
+	/* We must keep at least one descriptor in a TX ring empty.
+	 * We could avoid this when the queue size does not exactly
+	 * match the hardware ring size, but it's not that important.
+	 * Therefore we stop the queue when one more skb might fill
+	 * the ring completely.  We wake it when half way back to
+	 * empty.
+	 */
+	efx->txq_stop_thresh = efx->txq_entries - efx_tx_max_skb_descs(efx);
+	efx->txq_wake_thresh = efx->txq_stop_thresh / 2;
+
+	/* Initialise the channels */
+	efx_start_channels(efx);
+
+	efx_ptp_start_datapath(efx);
+
+	if (netif_device_present(efx->net_dev))
+		netif_tx_wake_all_queues(efx->net_dev);
+}
+
+void efx_stop_channels(struct efx_nic *efx)
+{
+	struct efx_tx_queue *tx_queue;
+	struct efx_rx_queue *rx_queue;
+	struct efx_channel *channel;
+	int rc = 0;
+
+	/* Stop RX refill */
+	efx_for_each_channel(channel, efx) {
+		efx_for_each_channel_rx_queue(rx_queue, channel)
+			rx_queue->refill_enabled = false;
+	}
+
+	efx_for_each_channel(channel, efx) {
+		/* RX packet processing is pipelined, so wait for the
+		 * NAPI handler to complete.  At least event queue 0
+		 * might be kept active by non-data events, so don't
+		 * use napi_synchronize() but actually disable NAPI
+		 * temporarily.
+		 */
+		if (efx_channel_has_rx_queue(channel)) {
+			efx_stop_eventq(channel);
+			efx_start_eventq(channel);
+		}
+	}
+
+	if (efx->type->fini_dmaq)
+		rc = efx->type->fini_dmaq(efx);
+
+	if (rc) {
+		netif_err(efx, drv, efx->net_dev, "failed to flush queues\n");
+	} else {
+		netif_dbg(efx, drv, efx->net_dev,
+			  "successfully flushed all queues\n");
+	}
+
+	efx_for_each_channel(channel, efx) {
+		efx_for_each_channel_rx_queue(rx_queue, channel)
+			efx_fini_rx_queue(rx_queue);
+		efx_for_each_possible_channel_tx_queue(tx_queue, channel)
+			efx_fini_tx_queue(tx_queue);
+	}
+	efx->xdp_rxq_info_failed = false;
+}
+
+static void efx_stop_datapath(struct efx_nic *efx)
+{
+	EFX_ASSERT_RESET_SERIALISED(efx);
+	BUG_ON(efx->port_enabled);
+
+	efx_ptp_stop_datapath(efx);
+
+	efx_stop_channels(efx);
+}
+
+/**************************************************************************
+ *
+ * Port handling
+ *
+ **************************************************************************/
+
+static void efx_start_port(struct efx_nic *efx)
+{
+	netif_dbg(efx, ifup, efx->net_dev, "start port\n");
+	BUG_ON(efx->port_enabled);
+
+	mutex_lock(&efx->mac_lock);
+	efx->port_enabled = true;
+
+	/* Ensure MAC ingress/egress is enabled */
+	efx_mac_reconfigure(efx);
+
+	mutex_unlock(&efx->mac_lock);
+}
+
+/* Cancel work for MAC reconfiguration, periodic hardware monitoring
+ * and the async self-test, wait for them to finish and prevent them
+ * being scheduled again.  This doesn't cover online resets, which
+ * should only be cancelled when removing the device.
+ */
+static void efx_stop_port(struct efx_nic *efx)
+{
+	netif_dbg(efx, ifdown, efx->net_dev, "stop port\n");
+
+	EFX_ASSERT_RESET_SERIALISED(efx);
+
+	mutex_lock(&efx->mac_lock);
+	efx->port_enabled = false;
+	mutex_unlock(&efx->mac_lock);
+
+	/* Serialise against efx_set_multicast_list() */
+	netif_addr_lock_bh(efx->net_dev);
+	netif_addr_unlock_bh(efx->net_dev);
+
+	cancel_delayed_work_sync(&efx->monitor_work);
+	efx_selftest_async_cancel(efx);
+	cancel_work_sync(&efx->mac_work);
+}
+
+/* If the interface is supposed to be running but is not, start
+ * the hardware and software data path, regular activity for the port
+ * (MAC statistics, link polling, etc.) and schedule the port to be
+ * reconfigured.  Interrupts must already be enabled.  This function
+ * is safe to call multiple times, so long as the NIC is not disabled.
+ * Requires the RTNL lock.
+ */
+void efx_start_all(struct efx_nic *efx)
+{
+	EFX_ASSERT_RESET_SERIALISED(efx);
+	BUG_ON(efx->state == STATE_DISABLED);
+
+	/* Check that it is appropriate to restart the interface. All
+	 * of these flags are safe to read under just the rtnl lock
+	 */
+	if (efx->port_enabled || !netif_running(efx->net_dev) ||
+	    efx->reset_pending)
+		return;
+
+	efx_start_port(efx);
+	efx_start_datapath(efx);
+
+	/* Start the hardware monitor if there is one */
+	efx_start_monitor(efx);
+
+	/* Link state detection is normally event-driven; we have
+	 * to poll now because we could have missed a change
+	 */
+	mutex_lock(&efx->mac_lock);
+	if (efx->phy_op->poll(efx))
+		efx_link_status_changed(efx);
+	mutex_unlock(&efx->mac_lock);
+
+	efx->type->start_stats(efx);
+	efx->type->pull_stats(efx);
+	spin_lock_bh(&efx->stats_lock);
+	efx->type->update_stats(efx, NULL, NULL);
+	spin_unlock_bh(&efx->stats_lock);
+}
+
+/* Quiesce the hardware and software data path, and regular activity
+ * for the port without bringing the link down.  Safe to call multiple
+ * times with the NIC in almost any state, but interrupts should be
+ * enabled.  Requires the RTNL lock.
+ */
+void efx_stop_all(struct efx_nic *efx)
+{
+	EFX_ASSERT_RESET_SERIALISED(efx);
+
+	/* port_enabled can be read safely under the rtnl lock */
+	if (!efx->port_enabled)
+		return;
+
+	/* update stats before we go down so we can accurately count
+	 * rx_nodesc_drops
+	 */
+	efx->type->pull_stats(efx);
+	spin_lock_bh(&efx->stats_lock);
+	efx->type->update_stats(efx, NULL, NULL);
+	spin_unlock_bh(&efx->stats_lock);
+	efx->type->stop_stats(efx);
+	efx_stop_port(efx);
+
+	/* Stop the kernel transmit interface.  This is only valid if
+	 * the device is stopped or detached; otherwise the watchdog
+	 * may fire immediately.
+	 */
+	WARN_ON(netif_running(efx->net_dev) &&
+		netif_device_present(efx->net_dev));
+	netif_tx_disable(efx->net_dev);
+
+	efx_stop_datapath(efx);
+}
+
+/* Push loopback/power/transmit disable settings to the PHY, and reconfigure
+ * the MAC appropriately. All other PHY configuration changes are pushed
+ * through phy_op->set_settings(), and pushed asynchronously to the MAC
+ * through efx_monitor().
+ *
+ * Callers must hold the mac_lock
+ */
+int __efx_reconfigure_port(struct efx_nic *efx)
+{
+	enum efx_phy_mode phy_mode;
+	int rc;
+
+	WARN_ON(!mutex_is_locked(&efx->mac_lock));
+
+	/* Disable PHY transmit in mac level loopbacks */
+	phy_mode = efx->phy_mode;
+	if (LOOPBACK_INTERNAL(efx))
+		efx->phy_mode |= PHY_MODE_TX_DISABLED;
+	else
+		efx->phy_mode &= ~PHY_MODE_TX_DISABLED;
+
+	rc = efx->type->reconfigure_port(efx);
+
+	if (rc)
+		efx->phy_mode = phy_mode;
+
+	return rc;
+}
+
+/* Reinitialise the MAC to pick up new PHY settings, even if the port is
+ * disabled.
+ */
+int efx_reconfigure_port(struct efx_nic *efx)
+{
+	int rc;
+
+	EFX_ASSERT_RESET_SERIALISED(efx);
+
+	mutex_lock(&efx->mac_lock);
+	rc = __efx_reconfigure_port(efx);
+	mutex_unlock(&efx->mac_lock);
+
+	return rc;
+}
-- 
cgit v1.2.3-59-g8ed1b


From 1eaf99fe0bb836c988b40c32d48d59aea29cd717 Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Wed, 8 Jan 2020 16:11:46 +0000
Subject: sfc: move some device reset code

The rest of the reset code will be moved later.
Small code styling fixes included.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/efx.c        | 227 --------------------------------
 drivers/net/ethernet/sfc/efx_common.c | 237 ++++++++++++++++++++++++++++++++++
 2 files changed, 237 insertions(+), 227 deletions(-)

diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index 988687a80abe..40d1e8ef672d 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -2483,184 +2483,6 @@ static void efx_unregister_netdev(struct efx_nic *efx)
  *
  **************************************************************************/
 
-/* Tears down the entire software state and most of the hardware state
- * before reset.  */
-void efx_reset_down(struct efx_nic *efx, enum reset_type method)
-{
-	EFX_ASSERT_RESET_SERIALISED(efx);
-
-	if (method == RESET_TYPE_MCDI_TIMEOUT)
-		efx->type->prepare_flr(efx);
-
-	efx_stop_all(efx);
-	efx_disable_interrupts(efx);
-
-	mutex_lock(&efx->mac_lock);
-	down_write(&efx->filter_sem);
-	mutex_lock(&efx->rss_lock);
-	if (efx->port_initialized && method != RESET_TYPE_INVISIBLE &&
-	    method != RESET_TYPE_DATAPATH)
-		efx->phy_op->fini(efx);
-	efx->type->fini(efx);
-}
-
-/* This function will always ensure that the locks acquired in
- * efx_reset_down() are released. A failure return code indicates
- * that we were unable to reinitialise the hardware, and the
- * driver should be disabled. If ok is false, then the rx and tx
- * engines are not restarted, pending a RESET_DISABLE. */
-int efx_reset_up(struct efx_nic *efx, enum reset_type method, bool ok)
-{
-	int rc;
-
-	EFX_ASSERT_RESET_SERIALISED(efx);
-
-	if (method == RESET_TYPE_MCDI_TIMEOUT)
-		efx->type->finish_flr(efx);
-
-	/* Ensure that SRAM is initialised even if we're disabling the device */
-	rc = efx->type->init(efx);
-	if (rc) {
-		netif_err(efx, drv, efx->net_dev, "failed to initialise NIC\n");
-		goto fail;
-	}
-
-	if (!ok)
-		goto fail;
-
-	if (efx->port_initialized && method != RESET_TYPE_INVISIBLE &&
-	    method != RESET_TYPE_DATAPATH) {
-		rc = efx->phy_op->init(efx);
-		if (rc)
-			goto fail;
-		rc = efx->phy_op->reconfigure(efx);
-		if (rc && rc != -EPERM)
-			netif_err(efx, drv, efx->net_dev,
-				  "could not restore PHY settings\n");
-	}
-
-	rc = efx_enable_interrupts(efx);
-	if (rc)
-		goto fail;
-
-#ifdef CONFIG_SFC_SRIOV
-	rc = efx->type->vswitching_restore(efx);
-	if (rc) /* not fatal; the PF will still work fine */
-		netif_warn(efx, probe, efx->net_dev,
-			   "failed to restore vswitching rc=%d;"
-			   " VFs may not function\n", rc);
-#endif
-
-	if (efx->type->rx_restore_rss_contexts)
-		efx->type->rx_restore_rss_contexts(efx);
-	mutex_unlock(&efx->rss_lock);
-	efx->type->filter_table_restore(efx);
-	up_write(&efx->filter_sem);
-	if (efx->type->sriov_reset)
-		efx->type->sriov_reset(efx);
-
-	mutex_unlock(&efx->mac_lock);
-
-	efx_start_all(efx);
-
-	if (efx->type->udp_tnl_push_ports)
-		efx->type->udp_tnl_push_ports(efx);
-
-	return 0;
-
-fail:
-	efx->port_initialized = false;
-
-	mutex_unlock(&efx->rss_lock);
-	up_write(&efx->filter_sem);
-	mutex_unlock(&efx->mac_lock);
-
-	return rc;
-}
-
-/* Reset the NIC using the specified method.  Note that the reset may
- * fail, in which case the card will be left in an unusable state.
- *
- * Caller must hold the rtnl_lock.
- */
-int efx_reset(struct efx_nic *efx, enum reset_type method)
-{
-	int rc, rc2;
-	bool disabled;
-
-	netif_info(efx, drv, efx->net_dev, "resetting (%s)\n",
-		   RESET_TYPE(method));
-
-	efx_device_detach_sync(efx);
-	efx_reset_down(efx, method);
-
-	rc = efx->type->reset(efx, method);
-	if (rc) {
-		netif_err(efx, drv, efx->net_dev, "failed to reset hardware\n");
-		goto out;
-	}
-
-	/* Clear flags for the scopes we covered.  We assume the NIC and
-	 * driver are now quiescent so that there is no race here.
-	 */
-	if (method < RESET_TYPE_MAX_METHOD)
-		efx->reset_pending &= -(1 << (method + 1));
-	else /* it doesn't fit into the well-ordered scope hierarchy */
-		__clear_bit(method, &efx->reset_pending);
-
-	/* Reinitialise bus-mastering, which may have been turned off before
-	 * the reset was scheduled. This is still appropriate, even in the
-	 * RESET_TYPE_DISABLE since this driver generally assumes the hardware
-	 * can respond to requests. */
-	pci_set_master(efx->pci_dev);
-
-out:
-	/* Leave device stopped if necessary */
-	disabled = rc ||
-		method == RESET_TYPE_DISABLE ||
-		method == RESET_TYPE_RECOVER_OR_DISABLE;
-	rc2 = efx_reset_up(efx, method, !disabled);
-	if (rc2) {
-		disabled = true;
-		if (!rc)
-			rc = rc2;
-	}
-
-	if (disabled) {
-		dev_close(efx->net_dev);
-		netif_err(efx, drv, efx->net_dev, "has been disabled\n");
-		efx->state = STATE_DISABLED;
-	} else {
-		netif_dbg(efx, drv, efx->net_dev, "reset complete\n");
-		efx_device_attach_if_not_resetting(efx);
-	}
-	return rc;
-}
-
-/* Try recovery mechanisms.
- * For now only EEH is supported.
- * Returns 0 if the recovery mechanisms are unsuccessful.
- * Returns a non-zero value otherwise.
- */
-int efx_try_recovery(struct efx_nic *efx)
-{
-#ifdef CONFIG_EEH
-	/* A PCI error can occur and not be seen by EEH because nothing
-	 * happens on the PCI bus. In this case the driver may fail and
-	 * schedule a 'recover or reset', leading to this recovery handler.
-	 * Manually call the eeh failure check function.
-	 */
-	struct eeh_dev *eehdev = pci_dev_to_eeh_dev(efx->pci_dev);
-	if (eeh_dev_check_failure(eehdev)) {
-		/* The EEH mechanisms will handle the error and reset the
-		 * device if necessary.
-		 */
-		return 1;
-	}
-#endif
-	return 0;
-}
-
 static void efx_wait_for_bist_end(struct efx_nic *efx)
 {
 	int i;
@@ -2714,55 +2536,6 @@ static void efx_reset_work(struct work_struct *data)
 	rtnl_unlock();
 }
 
-void efx_schedule_reset(struct efx_nic *efx, enum reset_type type)
-{
-	enum reset_type method;
-
-	if (efx->state == STATE_RECOVERY) {
-		netif_dbg(efx, drv, efx->net_dev,
-			  "recovering: skip scheduling %s reset\n",
-			  RESET_TYPE(type));
-		return;
-	}
-
-	switch (type) {
-	case RESET_TYPE_INVISIBLE:
-	case RESET_TYPE_ALL:
-	case RESET_TYPE_RECOVER_OR_ALL:
-	case RESET_TYPE_WORLD:
-	case RESET_TYPE_DISABLE:
-	case RESET_TYPE_RECOVER_OR_DISABLE:
-	case RESET_TYPE_DATAPATH:
-	case RESET_TYPE_MC_BIST:
-	case RESET_TYPE_MCDI_TIMEOUT:
-		method = type;
-		netif_dbg(efx, drv, efx->net_dev, "scheduling %s reset\n",
-			  RESET_TYPE(method));
-		break;
-	default:
-		method = efx->type->map_reset_reason(type);
-		netif_dbg(efx, drv, efx->net_dev,
-			  "scheduling %s reset for %s\n",
-			  RESET_TYPE(method), RESET_TYPE(type));
-		break;
-	}
-
-	set_bit(method, &efx->reset_pending);
-	smp_mb(); /* ensure we change reset_pending before checking state */
-
-	/* If we're not READY then just leave the flags set as the cue
-	 * to abort probing or reschedule the reset later.
-	 */
-	if (READ_ONCE(efx->state) != STATE_READY)
-		return;
-
-	/* efx_process_channel() will no longer read events once a
-	 * reset is scheduled. So switch back to poll'd MCDI completions. */
-	efx_mcdi_mode_poll(efx);
-
-	efx_queue_reset_work(efx);
-}
-
 /**************************************************************************
  *
  * List of NICs we support
diff --git a/drivers/net/ethernet/sfc/efx_common.c b/drivers/net/ethernet/sfc/efx_common.c
index 445340c903b1..813ebdda50f6 100644
--- a/drivers/net/ethernet/sfc/efx_common.c
+++ b/drivers/net/ethernet/sfc/efx_common.c
@@ -429,3 +429,240 @@ int efx_reconfigure_port(struct efx_nic *efx)
 
 	return rc;
 }
+
+/**************************************************************************
+ *
+ * Device reset and suspend
+ *
+ **************************************************************************/
+
+/* Try recovery mechanisms.
+ * For now only EEH is supported.
+ * Returns 0 if the recovery mechanisms are unsuccessful.
+ * Returns a non-zero value otherwise.
+ */
+int efx_try_recovery(struct efx_nic *efx)
+{
+#ifdef CONFIG_EEH
+	/* A PCI error can occur and not be seen by EEH because nothing
+	 * happens on the PCI bus. In this case the driver may fail and
+	 * schedule a 'recover or reset', leading to this recovery handler.
+	 * Manually call the eeh failure check function.
+	 */
+	struct eeh_dev *eehdev = pci_dev_to_eeh_dev(efx->pci_dev);
+	if (eeh_dev_check_failure(eehdev)) {
+		/* The EEH mechanisms will handle the error and reset the
+		 * device if necessary.
+		 */
+		return 1;
+	}
+#endif
+	return 0;
+}
+
+/* Tears down the entire software state and most of the hardware state
+ * before reset.
+ */
+void efx_reset_down(struct efx_nic *efx, enum reset_type method)
+{
+	EFX_ASSERT_RESET_SERIALISED(efx);
+
+	if (method == RESET_TYPE_MCDI_TIMEOUT)
+		efx->type->prepare_flr(efx);
+
+	efx_stop_all(efx);
+	efx_disable_interrupts(efx);
+
+	mutex_lock(&efx->mac_lock);
+	down_write(&efx->filter_sem);
+	mutex_lock(&efx->rss_lock);
+	if (efx->port_initialized && method != RESET_TYPE_INVISIBLE &&
+	    method != RESET_TYPE_DATAPATH)
+		efx->phy_op->fini(efx);
+	efx->type->fini(efx);
+}
+
+/* This function will always ensure that the locks acquired in
+ * efx_reset_down() are released. A failure return code indicates
+ * that we were unable to reinitialise the hardware, and the
+ * driver should be disabled. If ok is false, then the rx and tx
+ * engines are not restarted, pending a RESET_DISABLE.
+ */
+int efx_reset_up(struct efx_nic *efx, enum reset_type method, bool ok)
+{
+	int rc;
+
+	EFX_ASSERT_RESET_SERIALISED(efx);
+
+	if (method == RESET_TYPE_MCDI_TIMEOUT)
+		efx->type->finish_flr(efx);
+
+	/* Ensure that SRAM is initialised even if we're disabling the device */
+	rc = efx->type->init(efx);
+	if (rc) {
+		netif_err(efx, drv, efx->net_dev, "failed to initialise NIC\n");
+		goto fail;
+	}
+
+	if (!ok)
+		goto fail;
+
+	if (efx->port_initialized && method != RESET_TYPE_INVISIBLE &&
+	    method != RESET_TYPE_DATAPATH) {
+		rc = efx->phy_op->init(efx);
+		if (rc)
+			goto fail;
+		rc = efx->phy_op->reconfigure(efx);
+		if (rc && rc != -EPERM)
+			netif_err(efx, drv, efx->net_dev,
+				  "could not restore PHY settings\n");
+	}
+
+	rc = efx_enable_interrupts(efx);
+	if (rc)
+		goto fail;
+
+#ifdef CONFIG_SFC_SRIOV
+	rc = efx->type->vswitching_restore(efx);
+	if (rc) /* not fatal; the PF will still work fine */
+		netif_warn(efx, probe, efx->net_dev,
+			   "failed to restore vswitching rc=%d;"
+			   " VFs may not function\n", rc);
+#endif
+
+	if (efx->type->rx_restore_rss_contexts)
+		efx->type->rx_restore_rss_contexts(efx);
+	mutex_unlock(&efx->rss_lock);
+	efx->type->filter_table_restore(efx);
+	up_write(&efx->filter_sem);
+	if (efx->type->sriov_reset)
+		efx->type->sriov_reset(efx);
+
+	mutex_unlock(&efx->mac_lock);
+
+	efx_start_all(efx);
+
+	if (efx->type->udp_tnl_push_ports)
+		efx->type->udp_tnl_push_ports(efx);
+
+	return 0;
+
+fail:
+	efx->port_initialized = false;
+
+	mutex_unlock(&efx->rss_lock);
+	up_write(&efx->filter_sem);
+	mutex_unlock(&efx->mac_lock);
+
+	return rc;
+}
+
+/* Reset the NIC using the specified method.  Note that the reset may
+ * fail, in which case the card will be left in an unusable state.
+ *
+ * Caller must hold the rtnl_lock.
+ */
+int efx_reset(struct efx_nic *efx, enum reset_type method)
+{
+	bool disabled;
+	int rc, rc2;
+
+	netif_info(efx, drv, efx->net_dev, "resetting (%s)\n",
+		   RESET_TYPE(method));
+
+	efx_device_detach_sync(efx);
+	efx_reset_down(efx, method);
+
+	rc = efx->type->reset(efx, method);
+	if (rc) {
+		netif_err(efx, drv, efx->net_dev, "failed to reset hardware\n");
+		goto out;
+	}
+
+	/* Clear flags for the scopes we covered.  We assume the NIC and
+	 * driver are now quiescent so that there is no race here.
+	 */
+	if (method < RESET_TYPE_MAX_METHOD)
+		efx->reset_pending &= -(1 << (method + 1));
+	else /* it doesn't fit into the well-ordered scope hierarchy */
+		__clear_bit(method, &efx->reset_pending);
+
+	/* Reinitialise bus-mastering, which may have been turned off before
+	 * the reset was scheduled. This is still appropriate, even in the
+	 * RESET_TYPE_DISABLE since this driver generally assumes the hardware
+	 * can respond to requests.
+	 */
+	pci_set_master(efx->pci_dev);
+
+out:
+	/* Leave device stopped if necessary */
+	disabled = rc ||
+		method == RESET_TYPE_DISABLE ||
+		method == RESET_TYPE_RECOVER_OR_DISABLE;
+	rc2 = efx_reset_up(efx, method, !disabled);
+	if (rc2) {
+		disabled = true;
+		if (!rc)
+			rc = rc2;
+	}
+
+	if (disabled) {
+		dev_close(efx->net_dev);
+		netif_err(efx, drv, efx->net_dev, "has been disabled\n");
+		efx->state = STATE_DISABLED;
+	} else {
+		netif_dbg(efx, drv, efx->net_dev, "reset complete\n");
+		efx_device_attach_if_not_resetting(efx);
+	}
+	return rc;
+}
+
+void efx_schedule_reset(struct efx_nic *efx, enum reset_type type)
+{
+	enum reset_type method;
+
+	if (efx->state == STATE_RECOVERY) {
+		netif_dbg(efx, drv, efx->net_dev,
+			  "recovering: skip scheduling %s reset\n",
+			  RESET_TYPE(type));
+		return;
+	}
+
+	switch (type) {
+	case RESET_TYPE_INVISIBLE:
+	case RESET_TYPE_ALL:
+	case RESET_TYPE_RECOVER_OR_ALL:
+	case RESET_TYPE_WORLD:
+	case RESET_TYPE_DISABLE:
+	case RESET_TYPE_RECOVER_OR_DISABLE:
+	case RESET_TYPE_DATAPATH:
+	case RESET_TYPE_MC_BIST:
+	case RESET_TYPE_MCDI_TIMEOUT:
+		method = type;
+		netif_dbg(efx, drv, efx->net_dev, "scheduling %s reset\n",
+			  RESET_TYPE(method));
+		break;
+	default:
+		method = efx->type->map_reset_reason(type);
+		netif_dbg(efx, drv, efx->net_dev,
+			  "scheduling %s reset for %s\n",
+			  RESET_TYPE(method), RESET_TYPE(type));
+		break;
+	}
+
+	set_bit(method, &efx->reset_pending);
+	smp_mb(); /* ensure we change reset_pending before checking state */
+
+	/* If we're not READY then just leave the flags set as the cue
+	 * to abort probing or reschedule the reset later.
+	 */
+	if (READ_ONCE(efx->state) != STATE_READY)
+		return;
+
+	/* efx_process_channel() will no longer read events once a
+	 * reset is scheduled. So switch back to poll'd MCDI completions.
+	 */
+	efx_mcdi_mode_poll(efx);
+
+	efx_queue_reset_work(efx);
+}
-- 
cgit v1.2.3-59-g8ed1b


From f1826756b499c2f60df4aa21107da3e37f78363a Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Wed, 8 Jan 2020 16:12:04 +0000
Subject: sfc: move struct init and fini code

The hardware monitor code and the reset work queue code were also
moved, with supporting macros and parameters, because they are assigned
to function pointers in the struct.
Small code styling fixes included.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/efx.c        | 375 +------------------------------
 drivers/net/ethernet/sfc/efx_common.c | 411 ++++++++++++++++++++++++++++++++++
 2 files changed, 415 insertions(+), 371 deletions(-)

diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index 40d1e8ef672d..a1af42dde450 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -43,56 +43,6 @@
  **************************************************************************
  */
 
-/* Loopback mode names (see LOOPBACK_MODE()) */
-const unsigned int efx_loopback_mode_max = LOOPBACK_MAX;
-const char *const efx_loopback_mode_names[] = {
-	[LOOPBACK_NONE]		= "NONE",
-	[LOOPBACK_DATA]		= "DATAPATH",
-	[LOOPBACK_GMAC]		= "GMAC",
-	[LOOPBACK_XGMII]	= "XGMII",
-	[LOOPBACK_XGXS]		= "XGXS",
-	[LOOPBACK_XAUI]		= "XAUI",
-	[LOOPBACK_GMII]		= "GMII",
-	[LOOPBACK_SGMII]	= "SGMII",
-	[LOOPBACK_XGBR]		= "XGBR",
-	[LOOPBACK_XFI]		= "XFI",
-	[LOOPBACK_XAUI_FAR]	= "XAUI_FAR",
-	[LOOPBACK_GMII_FAR]	= "GMII_FAR",
-	[LOOPBACK_SGMII_FAR]	= "SGMII_FAR",
-	[LOOPBACK_XFI_FAR]	= "XFI_FAR",
-	[LOOPBACK_GPHY]		= "GPHY",
-	[LOOPBACK_PHYXS]	= "PHYXS",
-	[LOOPBACK_PCS]		= "PCS",
-	[LOOPBACK_PMAPMD]	= "PMA/PMD",
-	[LOOPBACK_XPORT]	= "XPORT",
-	[LOOPBACK_XGMII_WS]	= "XGMII_WS",
-	[LOOPBACK_XAUI_WS]	= "XAUI_WS",
-	[LOOPBACK_XAUI_WS_FAR]  = "XAUI_WS_FAR",
-	[LOOPBACK_XAUI_WS_NEAR] = "XAUI_WS_NEAR",
-	[LOOPBACK_GMII_WS]	= "GMII_WS",
-	[LOOPBACK_XFI_WS]	= "XFI_WS",
-	[LOOPBACK_XFI_WS_FAR]	= "XFI_WS_FAR",
-	[LOOPBACK_PHYXS_WS]	= "PHYXS_WS",
-};
-
-const unsigned int efx_reset_type_max = RESET_TYPE_MAX;
-const char *const efx_reset_type_names[] = {
-	[RESET_TYPE_INVISIBLE]          = "INVISIBLE",
-	[RESET_TYPE_ALL]                = "ALL",
-	[RESET_TYPE_RECOVER_OR_ALL]     = "RECOVER_OR_ALL",
-	[RESET_TYPE_WORLD]              = "WORLD",
-	[RESET_TYPE_RECOVER_OR_DISABLE] = "RECOVER_OR_DISABLE",
-	[RESET_TYPE_DATAPATH]           = "DATAPATH",
-	[RESET_TYPE_MC_BIST]		= "MC_BIST",
-	[RESET_TYPE_DISABLE]            = "DISABLE",
-	[RESET_TYPE_TX_WATCHDOG]        = "TX_WATCHDOG",
-	[RESET_TYPE_INT_ERROR]          = "INT_ERROR",
-	[RESET_TYPE_DMA_ERROR]          = "DMA_ERROR",
-	[RESET_TYPE_TX_SKIP]            = "TX_SKIP",
-	[RESET_TYPE_MC_FAILURE]         = "MC_FAILURE",
-	[RESET_TYPE_MCDI_TIMEOUT]	= "MCDI_TIMEOUT (FLR)",
-};
-
 /* UDP tunnel type names */
 static const char *const efx_udp_tunnel_type_names[] = {
 	[TUNNEL_ENCAP_UDP_PORT_ENTRY_VXLAN] = "vxlan",
@@ -108,12 +58,6 @@ void efx_get_udp_tunnel_type_name(u16 type, char *buf, size_t buflen)
 		snprintf(buf, buflen, "type %d", type);
 }
 
-/* How often and how many times to poll for a reset while waiting for a
- * BIST that another function started to complete.
- */
-#define BIST_WAIT_DELAY_MS	100
-#define BIST_WAIT_DELAY_COUNT	100
-
 /**************************************************************************
  *
  * Configurable values
@@ -138,16 +82,6 @@ MODULE_PARM_DESC(efx_separate_tx_channels,
  */
 static int napi_weight = 64;
 
-/* This is the time (in jiffies) between invocations of the hardware
- * monitor.
- * On Falcon-based NICs, this will:
- * - Check the on-board hardware monitor;
- * - Poll the link state and reconfigure the hardware as necessary.
- * On Siena-based NICs for power systems with EEH support, this will give EEH a
- * chance to start.
- */
-static unsigned int efx_monitor_interval = 1 * HZ;
-
 /* Initial interrupt moderation settings.  They can be modified after
  * module load with ethtool.
  *
@@ -830,19 +764,6 @@ void efx_link_set_wanted_fc(struct efx_nic *efx, u8 wanted_fc)
 
 static void efx_fini_port(struct efx_nic *efx);
 
-/* Asynchronous work item for changing MAC promiscuity and multicast
- * hash.  Avoid a drain/rx_ingress enable by reconfiguring the current
- * MAC directly. */
-static void efx_mac_work(struct work_struct *data)
-{
-	struct efx_nic *efx = container_of(data, struct efx_nic, mac_work);
-
-	mutex_lock(&efx->mac_lock);
-	if (efx->port_enabled)
-		efx_mac_reconfigure(efx);
-	mutex_unlock(&efx->mac_lock);
-}
-
 static int efx_probe_port(struct efx_nic *efx)
 {
 	int rc;
@@ -994,94 +915,6 @@ static void efx_dissociate(struct efx_nic *efx)
 	}
 }
 
-/* This configures the PCI device to enable I/O and DMA. */
-int efx_init_io(struct efx_nic *efx, int bar, dma_addr_t dma_mask,
-		unsigned int mem_map_size)
-{
-	struct pci_dev *pci_dev = efx->pci_dev;
-	int rc;
-
-	netif_dbg(efx, probe, efx->net_dev, "initialising I/O\n");
-
-	rc = pci_enable_device(pci_dev);
-	if (rc) {
-		netif_err(efx, probe, efx->net_dev,
-			  "failed to enable PCI device\n");
-		goto fail1;
-	}
-
-	pci_set_master(pci_dev);
-
-	/* Set the PCI DMA mask.  Try all possibilities from our genuine mask
-	 * down to 32 bits, because some architectures will allow 40 bit
-	 * masks event though they reject 46 bit masks.
-	 */
-	while (dma_mask > 0x7fffffffUL) {
-		rc = dma_set_mask_and_coherent(&pci_dev->dev, dma_mask);
-		if (rc == 0)
-			break;
-		dma_mask >>= 1;
-	}
-	if (rc) {
-		netif_err(efx, probe, efx->net_dev,
-			  "could not find a suitable DMA mask\n");
-		goto fail2;
-	}
-	netif_dbg(efx, probe, efx->net_dev,
-		  "using DMA mask %llx\n", (unsigned long long) dma_mask);
-
-	efx->membase_phys = pci_resource_start(efx->pci_dev, bar);
-	rc = pci_request_region(pci_dev, bar, "sfc");
-	if (rc) {
-		netif_err(efx, probe, efx->net_dev,
-			  "request for memory BAR failed\n");
-		rc = -EIO;
-		goto fail3;
-	}
-	efx->membase = ioremap_nocache(efx->membase_phys, mem_map_size);
-	if (!efx->membase) {
-		netif_err(efx, probe, efx->net_dev,
-			  "could not map memory BAR at %llx+%x\n",
-			  (unsigned long long)efx->membase_phys, mem_map_size);
-		rc = -ENOMEM;
-		goto fail4;
-	}
-	netif_dbg(efx, probe, efx->net_dev,
-		  "memory BAR at %llx+%x (virtual %p)\n",
-		  (unsigned long long)efx->membase_phys, mem_map_size,
-		  efx->membase);
-
-	return 0;
-
- fail4:
-	pci_release_region(efx->pci_dev, bar);
- fail3:
-	efx->membase_phys = 0;
- fail2:
-	pci_disable_device(efx->pci_dev);
- fail1:
-	return rc;
-}
-
-void efx_fini_io(struct efx_nic *efx, int bar)
-{
-	netif_dbg(efx, drv, efx->net_dev, "shutting down I/O\n");
-
-	if (efx->membase) {
-		iounmap(efx->membase);
-		efx->membase = NULL;
-	}
-
-	if (efx->membase_phys) {
-		pci_release_region(efx->pci_dev, bar);
-		efx->membase_phys = 0;
-	}
-
-	/* Don't disable bus-mastering if VFs are assigned */
-	if (!pci_vfs_assigned(efx->pci_dev))
-		pci_disable_device(efx->pci_dev);
-}
-
 void efx_set_default_rx_indir_table(struct efx_nic *efx,
 				    struct efx_rss_context *ctx)
 {
@@ -1855,43 +1688,6 @@ void efx_get_irq_moderation(struct efx_nic *efx, unsigned int *tx_usecs,
 	}
 }
 
-/**************************************************************************
- *
- * Hardware monitor
- *
- **************************************************************************/
-
-/* Run periodically off the general workqueue */
-static void efx_monitor(struct work_struct *data)
-{
-	struct efx_nic *efx = container_of(data, struct efx_nic,
-					   monitor_work.work);
-
-	netif_vdbg(efx, timer, efx->net_dev,
-		   "hardware monitor executing on CPU %d\n",
-		   raw_smp_processor_id());
-	BUG_ON(efx->type->monitor == NULL);
-
-	/* If the mac_lock is already held then it is likely a port
-	 * reconfiguration is already in place, which will likely do
-	 * most of the work of monitor() anyway. */
-	if (mutex_trylock(&efx->mac_lock)) {
-		if (efx->port_enabled)
-			efx->type->monitor(efx);
-		mutex_unlock(&efx->mac_lock);
-	}
-
-	queue_delayed_work(efx->workqueue, &efx->monitor_work,
-			   efx_monitor_interval);
-}
-
-void efx_start_monitor(struct efx_nic *efx)
-{
-	if (efx->type->monitor)
-		queue_delayed_work(efx->workqueue, &efx->monitor_work,
-				   efx_monitor_interval);
-}
-
 /**************************************************************************
  *
  * ioctls
@@ -2477,65 +2273,6 @@ static void efx_unregister_netdev(struct efx_nic *efx)
 	}
 }
 
-/**************************************************************************
- *
- * Device reset and suspend
- *
- **************************************************************************/
-
-static void efx_wait_for_bist_end(struct efx_nic *efx)
-{
-	int i;
-
-	for (i = 0; i < BIST_WAIT_DELAY_COUNT; ++i) {
-		if (efx_mcdi_poll_reboot(efx))
-			goto out;
-		msleep(BIST_WAIT_DELAY_MS);
-	}
-
-	netif_err(efx, drv, efx->net_dev, "Warning: No MC reboot after BIST mode\n");
-out:
-	/* Either way unset the BIST flag. If we found no reboot we probably
-	 * won't recover, but we should try.
-	 */
-	efx->mc_bist_for_other_fn = false;
-}
-
-/* The worker thread exists so that code that cannot sleep can
- * schedule a reset for later.
- */
-static void efx_reset_work(struct work_struct *data)
-{
-	struct efx_nic *efx = container_of(data, struct efx_nic, reset_work);
-	unsigned long pending;
-	enum reset_type method;
-
-	pending = READ_ONCE(efx->reset_pending);
-	method = fls(pending) - 1;
-
-	if (method == RESET_TYPE_MC_BIST)
-		efx_wait_for_bist_end(efx);
-
-	if ((method == RESET_TYPE_RECOVER_OR_DISABLE ||
-	     method == RESET_TYPE_RECOVER_OR_ALL) &&
-	    efx_try_recovery(efx))
-		return;
-
-	if (!pending)
-		return;
-
-	rtnl_lock();
-
-	/* We checked the state in efx_schedule_reset() but it may
-	 * have changed by now.  Now that we have the RTNL lock,
-	 * it cannot change again.
-	 */
-	if (efx->state == STATE_READY)
-		(void)efx_reset(efx, method);
-
-	rtnl_unlock();
-}
-
 /**************************************************************************
  *
  * List of NICs we support
@@ -2567,93 +2304,20 @@ static const struct pci_device_id efx_pci_table[] = {
 	{0}			/* end of list */
 };
 
-/**************************************************************************
- *
- * Dummy PHY/MAC operations
- *
- * Can be used for some unimplemented operations
- * Needed so all function pointers are valid and do not have to be tested
- * before use
- *
- **************************************************************************/
-int efx_port_dummy_op_int(struct efx_nic *efx)
-{
-	return 0;
-}
-void efx_port_dummy_op_void(struct efx_nic *efx) {}
-
-static bool efx_port_dummy_op_poll(struct efx_nic *efx)
-{
-	return false;
-}
-
-static const struct efx_phy_operations efx_dummy_phy_operations = {
-	.init		 = efx_port_dummy_op_int,
-	.reconfigure	 = efx_port_dummy_op_int,
-	.poll		 = efx_port_dummy_op_poll,
-	.fini		 = efx_port_dummy_op_void,
-};
-
 /**************************************************************************
  *
  * Data housekeeping
  *
  **************************************************************************/
 
-/* This zeroes out and then fills in the invariants in a struct
- * efx_nic (including all sub-structures).
- */
-int efx_init_struct(struct efx_nic *efx, struct pci_dev *pci_dev,
-		    struct net_device *net_dev)
+int efx_init_channels(struct efx_nic *efx)
 {
-	int rc = -ENOMEM, i;
-
-	/* Initialise common structures */
-	INIT_LIST_HEAD(&efx->node);
-	INIT_LIST_HEAD(&efx->secondary_list);
-	spin_lock_init(&efx->biu_lock);
-#ifdef CONFIG_SFC_MTD
-	INIT_LIST_HEAD(&efx->mtd_list);
-#endif
-	INIT_WORK(&efx->reset_work, efx_reset_work);
-	INIT_DELAYED_WORK(&efx->monitor_work, efx_monitor);
-	INIT_DELAYED_WORK(&efx->selftest_work, efx_selftest_async_work);
-	efx->pci_dev = pci_dev;
-	efx->msg_enable = debug;
-	efx->state = STATE_UNINIT;
-	strlcpy(efx->name, pci_name(pci_dev), sizeof(efx->name));
-
-	efx->net_dev = net_dev;
-	efx->rx_prefix_size = efx->type->rx_prefix_size;
-	efx->rx_ip_align =
-		NET_IP_ALIGN ? (efx->rx_prefix_size + NET_IP_ALIGN) % 4 : 0;
-	efx->rx_packet_hash_offset =
-		efx->type->rx_hash_offset - efx->type->rx_prefix_size;
-	efx->rx_packet_ts_offset =
-		efx->type->rx_ts_offset - efx->type->rx_prefix_size;
-	INIT_LIST_HEAD(&efx->rss_context.list);
-	mutex_init(&efx->rss_lock);
-	spin_lock_init(&efx->stats_lock);
-	efx->vi_stride = EFX_DEFAULT_VI_STRIDE;
-	efx->num_mac_stats = MC_CMD_MAC_NSTATS;
-	BUILD_BUG_ON(MC_CMD_MAC_NSTATS - 1 != MC_CMD_MAC_GENERATION_END);
-	mutex_init(&efx->mac_lock);
-#ifdef CONFIG_RFS_ACCEL
-	mutex_init(&efx->rps_mutex);
-	spin_lock_init(&efx->rps_hash_lock);
-	/* Failure to allocate is not fatal, but may degrade ARFS performance */
-	efx->rps_hash_table = kcalloc(EFX_ARFS_HASH_TABLE_SIZE,
-				      sizeof(*efx->rps_hash_table), GFP_KERNEL);
-#endif
-	efx->phy_op = &efx_dummy_phy_operations;
-	efx->mdio.dev = net_dev;
-	INIT_WORK(&efx->mac_work, efx_mac_work);
-	init_waitqueue_head(&efx->flush_wq);
+	unsigned int i;
 
 	for (i = 0; i < EFX_MAX_CHANNELS; i++) {
 		efx->channel[i] = efx_alloc_channel(efx, i, NULL);
 		if (!efx->channel[i])
-			goto fail;
+			return -ENOMEM;
 		efx->msi_context[i].efx = efx;
 		efx->msi_context[i].index = i;
 	}
@@ -2661,45 +2325,14 @@ int efx_init_struct(struct efx_nic *efx, struct pci_dev *pci_dev,
 	/* Higher numbered interrupt modes are less capable! */
 	if (WARN_ON_ONCE(efx->type->max_interrupt_mode >
 			 efx->type->min_interrupt_mode)) {
-		rc = -EIO;
-		goto fail;
+		return -EIO;
 	}
 	efx->interrupt_mode = max(efx->type->max_interrupt_mode,
 				  interrupt_mode);
 	efx->interrupt_mode = min(efx->type->min_interrupt_mode,
 				  interrupt_mode);
 
-	/* Would be good to use the net_dev name, but we're too early */
-	snprintf(efx->workqueue_name, sizeof(efx->workqueue_name), "sfc%s",
-		 pci_name(pci_dev));
-	efx->workqueue = create_singlethread_workqueue(efx->workqueue_name);
-	if (!efx->workqueue)
-		goto fail;
-
 	return 0;
-
-fail:
-	efx_fini_struct(efx);
-	return rc;
-}
-
-void efx_fini_struct(struct efx_nic *efx)
-{
-	int i;
-
-#ifdef CONFIG_RFS_ACCEL
-	kfree(efx->rps_hash_table);
-#endif
-
-	for (i = 0; i < EFX_MAX_CHANNELS; i++)
-		kfree(efx->channel[i]);
-
-	kfree(efx->vpd_sn);
-
-	if (efx->workqueue) {
-		destroy_workqueue(efx->workqueue);
-		efx->workqueue = NULL;
-	}
 }
 
 void efx_update_sw_stats(struct efx_nic *efx, u64 *stats)
diff --git a/drivers/net/ethernet/sfc/efx_common.c b/drivers/net/ethernet/sfc/efx_common.c
index 813ebdda50f6..f1c487de3f04 100644
--- a/drivers/net/ethernet/sfc/efx_common.c
+++ b/drivers/net/ethernet/sfc/efx_common.c
@@ -22,6 +22,85 @@
 #include "io.h"
 #include "mcdi_pcol.h"
 
+static unsigned int debug = (NETIF_MSG_DRV | NETIF_MSG_PROBE |
+			     NETIF_MSG_LINK | NETIF_MSG_IFDOWN |
+			     NETIF_MSG_IFUP | NETIF_MSG_RX_ERR |
+			     NETIF_MSG_TX_ERR | NETIF_MSG_HW);
+module_param(debug, uint, 0);
+MODULE_PARM_DESC(debug, "Bitmapped debugging message enable value");
+
+/* This is the time (in jiffies) between invocations of the hardware
+ * monitor.
+ * On Falcon-based NICs, this will:
+ * - Check the on-board hardware monitor;
+ * - Poll the link state and reconfigure the hardware as necessary.
+ * On Siena-based NICs for power systems with EEH support, this will give EEH a
+ * chance to start.
+ */
+static unsigned int efx_monitor_interval = 1 * HZ;
+
+/* How often and how many times to poll for a reset while waiting for a
+ * BIST that another function started to complete.
+ */
+#define BIST_WAIT_DELAY_MS	100
+#define BIST_WAIT_DELAY_COUNT	100
+
+/* Default stats update time */
+#define STATS_PERIOD_MS_DEFAULT 1000
+
+const unsigned int efx_reset_type_max = RESET_TYPE_MAX;
+const char *const efx_reset_type_names[] = {
+	[RESET_TYPE_INVISIBLE]          = "INVISIBLE",
+	[RESET_TYPE_ALL]                = "ALL",
+	[RESET_TYPE_RECOVER_OR_ALL]     = "RECOVER_OR_ALL",
+	[RESET_TYPE_WORLD]              = "WORLD",
+	[RESET_TYPE_RECOVER_OR_DISABLE] = "RECOVER_OR_DISABLE",
+	[RESET_TYPE_DATAPATH]           = "DATAPATH",
+	[RESET_TYPE_MC_BIST]		= "MC_BIST",
+	[RESET_TYPE_DISABLE]            = "DISABLE",
+	[RESET_TYPE_TX_WATCHDOG]        = "TX_WATCHDOG",
+	[RESET_TYPE_INT_ERROR]          = "INT_ERROR",
+	[RESET_TYPE_DMA_ERROR]          = "DMA_ERROR",
+	[RESET_TYPE_TX_SKIP]            = "TX_SKIP",
+	[RESET_TYPE_MC_FAILURE]         = "MC_FAILURE",
+	[RESET_TYPE_MCDI_TIMEOUT]	= "MCDI_TIMEOUT (FLR)",
+};
+
+#define RESET_TYPE(type) \
+	STRING_TABLE_LOOKUP(type, efx_reset_type)
+
+/* Loopback mode names (see LOOPBACK_MODE()) */
+const unsigned int efx_loopback_mode_max = LOOPBACK_MAX;
+const char *const efx_loopback_mode_names[] = {
+	[LOOPBACK_NONE]		= "NONE",
+	[LOOPBACK_DATA]		= "DATAPATH",
+	[LOOPBACK_GMAC]		= "GMAC",
+	[LOOPBACK_XGMII]	= "XGMII",
+	[LOOPBACK_XGXS]		= "XGXS",
+	[LOOPBACK_XAUI]		= "XAUI",
+	[LOOPBACK_GMII]		= "GMII",
+	[LOOPBACK_SGMII]	= "SGMII",
+	[LOOPBACK_XGBR]		= "XGBR",
+	[LOOPBACK_XFI]		= "XFI",
+	[LOOPBACK_XAUI_FAR]	= "XAUI_FAR",
+	[LOOPBACK_GMII_FAR]	= "GMII_FAR",
+	[LOOPBACK_SGMII_FAR]	= "SGMII_FAR",
+	[LOOPBACK_XFI_FAR]	= "XFI_FAR",
+	[LOOPBACK_GPHY]		= "GPHY",
+	[LOOPBACK_PHYXS]	= "PHYXS",
+	[LOOPBACK_PCS]		= "PCS",
+	[LOOPBACK_PMAPMD]	= "PMA/PMD",
+	[LOOPBACK_XPORT]	= "XPORT",
+	[LOOPBACK_XGMII_WS]	= "XGMII_WS",
+	[LOOPBACK_XAUI_WS]	= "XAUI_WS",
+	[LOOPBACK_XAUI_WS_FAR]  = "XAUI_WS_FAR",
+	[LOOPBACK_XAUI_WS_NEAR] = "XAUI_WS_NEAR",
+	[LOOPBACK_GMII_WS]	= "GMII_WS",
+	[LOOPBACK_XFI_WS]	= "XFI_WS",
+	[LOOPBACK_XFI_WS_FAR]	= "XFI_WS_FAR",
+	[LOOPBACK_PHYXS_WS]	= "PHYXS_WS",
+};
+
 /* Reset workqueue. If any NIC has a hardware failure then a reset will be
  * queued onto this work queue. This is not a per-nic work queue, because
  * efx_reset_work() acquires the rtnl lock, so resets are naturally serialised.
@@ -67,6 +146,20 @@ void efx_mac_reconfigure(struct efx_nic *efx)
 	up_read(&efx->filter_sem);
 }
 
+/* Asynchronous work item for changing MAC promiscuity and multicast
+ * hash.  Avoid a drain/rx_ingress enable by reconfiguring the current
+ * MAC directly.
+ */
+static void efx_mac_work(struct work_struct *data)
+{
+	struct efx_nic *efx = container_of(data, struct efx_nic, mac_work);
+
+	mutex_lock(&efx->mac_lock);
+	if (efx->port_enabled)
+		efx_mac_reconfigure(efx);
+	mutex_unlock(&efx->mac_lock);
+}
+
 /* This ensures that the kernel is kept informed (via
  * netif_carrier_on/off) of the link status, and also maintains the
  * link status's stop on the port's TX queue.
@@ -102,6 +195,43 @@ void efx_link_status_changed(struct efx_nic *efx)
 		netif_info(efx, link, efx->net_dev, "link down\n");
 }
 
+/**************************************************************************
+ *
+ * Hardware monitor
+ *
+ **************************************************************************/
+
+/* Run periodically off the general workqueue */
+static void efx_monitor(struct work_struct *data)
+{
+	struct efx_nic *efx = container_of(data, struct efx_nic,
+					   monitor_work.work);
+
+	netif_vdbg(efx, timer, efx->net_dev,
+		   "hardware monitor executing on CPU %d\n",
+		   raw_smp_processor_id());
+	BUG_ON(efx->type->monitor == NULL);
+
+	/* If the mac_lock is already held then it is likely a port
+	 * reconfiguration is already in place, which will likely do
+	 * most of the work of monitor() anyway.
+	 */
+	if (mutex_trylock(&efx->mac_lock)) {
+		if (efx->port_enabled && efx->type->monitor)
+			efx->type->monitor(efx);
+		mutex_unlock(&efx->mac_lock);
+	}
+
+	efx_start_monitor(efx);
+}
+
+void efx_start_monitor(struct efx_nic *efx)
+{
+	if (efx->type->monitor)
+		queue_delayed_work(efx->workqueue, &efx->monitor_work,
+				   efx_monitor_interval);
+}
+
 /**************************************************************************
  *
  * Event queue processing
@@ -436,6 +566,24 @@ int efx_reconfigure_port(struct efx_nic *efx)
  *
  **************************************************************************/
 
+static void efx_wait_for_bist_end(struct efx_nic *efx)
+{
+	int i;
+
+	for (i = 0; i < BIST_WAIT_DELAY_COUNT; ++i) {
+		if (efx_mcdi_poll_reboot(efx))
+			goto out;
+		msleep(BIST_WAIT_DELAY_MS);
+	}
+
+	netif_err(efx, drv, efx->net_dev, "Warning: No MC reboot after BIST mode\n");
+out:
+	/* Either way unset the BIST flag. If we found no reboot we probably
+	 * won't recover, but we should try.
+	 */
+	efx->mc_bist_for_other_fn = false;
+}
+
 /* Try recovery mechanisms.
  * For now only EEH is supported.
  * Returns 0 if the recovery mechanisms are unsuccessful.
@@ -617,6 +765,41 @@ out:
 	return rc;
 }
 
+/* The worker thread exists so that code that cannot sleep can
+ * schedule a reset for later.
+ */
+static void efx_reset_work(struct work_struct *data)
+{
+	struct efx_nic *efx = container_of(data, struct efx_nic, reset_work);
+	unsigned long pending;
+	enum reset_type method;
+
+	pending = READ_ONCE(efx->reset_pending);
+	method = fls(pending) - 1;
+
+	if (method == RESET_TYPE_MC_BIST)
+		efx_wait_for_bist_end(efx);
+
+	if ((method == RESET_TYPE_RECOVER_OR_DISABLE ||
+	     method == RESET_TYPE_RECOVER_OR_ALL) &&
+	    efx_try_recovery(efx))
+		return;
+
+	if (!pending)
+		return;
+
+	rtnl_lock();
+
+	/* We checked the state in efx_schedule_reset() but it may
+	 * have changed by now.  Now that we have the RTNL lock,
+	 * it cannot change again.
+	 */
+	if (efx->state == STATE_READY)
+		(void)efx_reset(efx, method);
+
+	rtnl_unlock();
+}
+
 void efx_schedule_reset(struct efx_nic *efx, enum reset_type type)
 {
 	enum reset_type method;
@@ -666,3 +849,231 @@ void efx_schedule_reset(struct efx_nic *efx, enum reset_type type)
 
 	efx_queue_reset_work(efx);
 }
+
+/**************************************************************************
+ *
+ * Dummy PHY/MAC operations
+ *
+ * Can be used for some unimplemented operations
+ * Needed so all function pointers are valid and do not have to be tested
+ * before use
+ *
+ **************************************************************************/
+int efx_port_dummy_op_int(struct efx_nic *efx)
+{
+	return 0;
+}
+void efx_port_dummy_op_void(struct efx_nic *efx) {}
+
+static bool efx_port_dummy_op_poll(struct efx_nic *efx)
+{
+	return false;
+}
+
+static const struct efx_phy_operations efx_dummy_phy_operations = {
+	.init		 = efx_port_dummy_op_int,
+	.reconfigure	 = efx_port_dummy_op_int,
+	.poll		 = efx_port_dummy_op_poll,
+	.fini		 = efx_port_dummy_op_void,
+};
+
+/**************************************************************************
+ *
+ * Data housekeeping
+ *
+ **************************************************************************/
+
+/* This zeroes out and then fills in the invariants in a struct
+ * efx_nic (including all sub-structures).
+ */
+int efx_init_struct(struct efx_nic *efx,
+		    struct pci_dev *pci_dev, struct net_device *net_dev)
+{
+	int rc = -ENOMEM;
+
+	/* Initialise common structures */
+	INIT_LIST_HEAD(&efx->node);
+	INIT_LIST_HEAD(&efx->secondary_list);
+	spin_lock_init(&efx->biu_lock);
+#ifdef CONFIG_SFC_MTD
+	INIT_LIST_HEAD(&efx->mtd_list);
+#endif
+	INIT_WORK(&efx->reset_work, efx_reset_work);
+	INIT_DELAYED_WORK(&efx->monitor_work, efx_monitor);
+	INIT_DELAYED_WORK(&efx->selftest_work, efx_selftest_async_work);
+	efx->pci_dev = pci_dev;
+	efx->msg_enable = debug;
+	efx->state = STATE_UNINIT;
+	strlcpy(efx->name, pci_name(pci_dev), sizeof(efx->name));
+
+	efx->net_dev = net_dev;
+	efx->rx_prefix_size = efx->type->rx_prefix_size;
+	efx->rx_ip_align =
+		NET_IP_ALIGN ? (efx->rx_prefix_size + NET_IP_ALIGN) % 4 : 0;
+	efx->rx_packet_hash_offset =
+		efx->type->rx_hash_offset - efx->type->rx_prefix_size;
+	efx->rx_packet_ts_offset =
+		efx->type->rx_ts_offset - efx->type->rx_prefix_size;
+	INIT_LIST_HEAD(&efx->rss_context.list);
+	mutex_init(&efx->rss_lock);
+	spin_lock_init(&efx->stats_lock);
+	efx->vi_stride = EFX_DEFAULT_VI_STRIDE;
+	efx->num_mac_stats = MC_CMD_MAC_NSTATS;
+	BUILD_BUG_ON(MC_CMD_MAC_NSTATS - 1 != MC_CMD_MAC_GENERATION_END);
+	mutex_init(&efx->mac_lock);
+#ifdef CONFIG_RFS_ACCEL
+	mutex_init(&efx->rps_mutex);
+	spin_lock_init(&efx->rps_hash_lock);
+	/* Failure to allocate is not fatal, but may degrade ARFS performance */
+	efx->rps_hash_table = kcalloc(EFX_ARFS_HASH_TABLE_SIZE,
+				      sizeof(*efx->rps_hash_table), GFP_KERNEL);
+#endif
+	efx->phy_op = &efx_dummy_phy_operations;
+	efx->mdio.dev = net_dev;
+	INIT_WORK(&efx->mac_work, efx_mac_work);
+	init_waitqueue_head(&efx->flush_wq);
+
+	rc = efx_init_channels(efx);
+	if (rc)
+		goto fail;
+
+	/* Would be good to use the net_dev name, but we're too early */
+	snprintf(efx->workqueue_name, sizeof(efx->workqueue_name), "sfc%s",
+		 pci_name(pci_dev));
+	efx->workqueue = create_singlethread_workqueue(efx->workqueue_name);
+	if (!efx->workqueue) {
+		rc = -ENOMEM;
+		goto fail;
+	}
+
+	return 0;
+
+fail:
+	efx_fini_struct(efx);
+	return rc;
+}
+
+void efx_fini_channels(struct efx_nic *efx)
+{
+	unsigned int i;
+
+	for (i = 0; i < EFX_MAX_CHANNELS; i++)
+		if (efx->channel[i]) {
+			kfree(efx->channel[i]);
+			efx->channel[i] = NULL;
+		}
+}
+
+void efx_fini_struct(struct efx_nic *efx)
+{
+#ifdef CONFIG_RFS_ACCEL
+	kfree(efx->rps_hash_table);
+#endif
+
+	efx_fini_channels(efx);
+
+	kfree(efx->vpd_sn);
+
+	if (efx->workqueue) {
+		destroy_workqueue(efx->workqueue);
+		efx->workqueue = NULL;
+	}
+}
+
+/* This configures the PCI device to enable I/O and DMA. */
+int efx_init_io(struct efx_nic *efx, int bar, dma_addr_t dma_mask,
+		unsigned int mem_map_size)
+{
+	struct pci_dev *pci_dev = efx->pci_dev;
+	int rc;
+
+	netif_dbg(efx, probe, efx->net_dev, "initialising I/O\n");
+
+	rc = pci_enable_device(pci_dev);
+	if (rc) {
+		netif_err(efx, probe, efx->net_dev,
+			  "failed to enable PCI device\n");
+		goto fail1;
+	}
+
+	pci_set_master(pci_dev);
+
+	/* Set the PCI DMA mask.  Try all possibilities from our
+	 * genuine mask down to 32 bits, because some architectures
+	 * (e.g. x86_64 with iommu_sac_force set) will allow 40 bit
+	 * masks event though they reject 46 bit masks.
+	 */
+	while (dma_mask > 0x7fffffffUL) {
+		rc = dma_set_mask_and_coherent(&pci_dev->dev, dma_mask);
+		if (rc == 0)
+			break;
+		dma_mask >>= 1;
+	}
+	if (rc) {
+		netif_err(efx, probe, efx->net_dev,
+			  "could not find a suitable DMA mask\n");
+		goto fail2;
+	}
+	netif_dbg(efx, probe, efx->net_dev,
+		  "using DMA mask %llx\n", (unsigned long long)dma_mask);
+
+	efx->membase_phys = pci_resource_start(efx->pci_dev, bar);
+	if (!efx->membase_phys) {
+		netif_err(efx, probe, efx->net_dev,
+			  "ERROR: No BAR%d mapping from the BIOS. "
+			  "Try pci=realloc on the kernel command line\n", bar);
+		rc = -ENODEV;
+		goto fail3;
+	}
+
+	rc = pci_request_region(pci_dev, bar, "sfc");
+	if (rc) {
+		netif_err(efx, probe, efx->net_dev,
+			  "request for memory BAR failed\n");
+		rc = -EIO;
+		goto fail3;
+	}
+
+	efx->membase = ioremap_nocache(efx->membase_phys, mem_map_size);
+	if (!efx->membase) {
+		netif_err(efx, probe, efx->net_dev,
+			  "could not map memory BAR at %llx+%x\n",
+			  (unsigned long long)efx->membase_phys, mem_map_size);
+		rc = -ENOMEM;
+		goto fail4;
+	}
+	netif_dbg(efx, probe, efx->net_dev,
+		  "memory BAR at %llx+%x (virtual %p)\n",
+		  (unsigned long long)efx->membase_phys, mem_map_size,
+		  efx->membase);
+
+	return 0;
+
+fail4:
+	pci_release_region(efx->pci_dev, bar);
+fail3:
+	efx->membase_phys = 0;
+fail2:
+	pci_disable_device(efx->pci_dev);
+fail1:
+	return rc;
+}
+
+void efx_fini_io(struct efx_nic *efx, int bar)
+{
+	netif_dbg(efx, drv, efx->net_dev, "shutting down I/O\n");
+
+	if (efx->membase) {
+		iounmap(efx->membase);
+		efx->membase = NULL;
+	}
+
+	if (efx->membase_phys) {
+		pci_release_region(efx->pci_dev, bar);
+		efx->membase_phys = 0;
+	}
+
+	/* Don't disable bus-mastering if VFs are assigned */
+	if (!pci_vfs_assigned(efx->pci_dev))
+		pci_disable_device(efx->pci_dev);
+}
-- 
cgit v1.2.3-59-g8ed1b


From 768fd2664e24141ce9c93335638f98d7f269f70f Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Wed, 8 Jan 2020 16:12:18 +0000
Subject: sfc: move some channel-related code

Just a handful of function, but also removed many 'static' identifiers
so the code builds. These will, of course, be moved.
Module parameters for IRQ moderation threshold also moved.

Small code styling fixes included.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/Makefile       |   2 +-
 drivers/net/ethernet/sfc/efx.c          | 169 ----------------------------
 drivers/net/ethernet/sfc/efx_channels.c | 189 ++++++++++++++++++++++++++++++++
 3 files changed, 190 insertions(+), 170 deletions(-)
 create mode 100644 drivers/net/ethernet/sfc/efx_channels.c

diff --git a/drivers/net/ethernet/sfc/Makefile b/drivers/net/ethernet/sfc/Makefile
index 7022cffa31f8..dacfa42beffe 100644
--- a/drivers/net/ethernet/sfc/Makefile
+++ b/drivers/net/ethernet/sfc/Makefile
@@ -1,5 +1,5 @@
 # SPDX-License-Identifier: GPL-2.0
-sfc-y			+= efx.o efx_common.o nic.o \
+sfc-y			+= efx.o efx_common.o efx_channels.o nic.o \
 			   farch.o siena.o ef10.o \
 			   tx.o rx.o \
 			   selftest.o ethtool.o ptp.o tx_tso.o \
diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index a1af42dde450..530185636afd 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -77,11 +77,6 @@ module_param(efx_separate_tx_channels, bool, 0444);
 MODULE_PARM_DESC(efx_separate_tx_channels,
 		 "Use separate channels for TX and RX");
 
-/* This is the weight assigned to each of the (per-channel) virtual
- * NAPI devices.
- */
-static int napi_weight = 64;
-
 /* Initial interrupt moderation settings.  They can be modified after
  * module load with ethtool.
  *
@@ -123,16 +118,6 @@ static bool phy_flash_cfg;
 module_param(phy_flash_cfg, bool, 0644);
 MODULE_PARM_DESC(phy_flash_cfg, "Set PHYs into reflash mode initially");
 
-static unsigned irq_adapt_low_thresh = 8000;
-module_param(irq_adapt_low_thresh, uint, 0644);
-MODULE_PARM_DESC(irq_adapt_low_thresh,
-		 "Threshold score for reducing IRQ moderation");
-
-static unsigned irq_adapt_high_thresh = 16000;
-module_param(irq_adapt_high_thresh, uint, 0644);
-MODULE_PARM_DESC(irq_adapt_high_thresh,
-		 "Threshold score for increasing IRQ moderation");
-
 static unsigned debug = (NETIF_MSG_DRV | NETIF_MSG_PROBE |
 			 NETIF_MSG_LINK | NETIF_MSG_IFDOWN |
 			 NETIF_MSG_IFUP | NETIF_MSG_RX_ERR |
@@ -167,121 +152,6 @@ static int efx_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **xdpfs,
  *
  *************************************************************************/
 
-/* Process channel's event queue
- *
- * This function is responsible for processing the event queue of a
- * single channel.  The caller must guarantee that this function will
- * never be concurrently called more than once on the same channel,
- * though different channels may be being processed concurrently.
- */
-static int efx_process_channel(struct efx_channel *channel, int budget)
-{
-	struct efx_tx_queue *tx_queue;
-	struct list_head rx_list;
-	int spent;
-
-	if (unlikely(!channel->enabled))
-		return 0;
-
-	/* Prepare the batch receive list */
-	EFX_WARN_ON_PARANOID(channel->rx_list != NULL);
-	INIT_LIST_HEAD(&rx_list);
-	channel->rx_list = &rx_list;
-
-	efx_for_each_channel_tx_queue(tx_queue, channel) {
-		tx_queue->pkts_compl = 0;
-		tx_queue->bytes_compl = 0;
-	}
-
-	spent = efx_nic_process_eventq(channel, budget);
-	if (spent && efx_channel_has_rx_queue(channel)) {
-		struct efx_rx_queue *rx_queue =
-			efx_channel_get_rx_queue(channel);
-
-		efx_rx_flush_packet(channel);
-		efx_fast_push_rx_descriptors(rx_queue, true);
-	}
-
-	/* Update BQL */
-	efx_for_each_channel_tx_queue(tx_queue, channel) {
-		if (tx_queue->bytes_compl) {
-			netdev_tx_completed_queue(tx_queue->core_txq,
-				tx_queue->pkts_compl, tx_queue->bytes_compl);
-		}
-	}
-
-	/* Receive any packets we queued up */
-	netif_receive_skb_list(channel->rx_list);
-	channel->rx_list = NULL;
-
-	return spent;
-}
-
-/* NAPI poll handler
- *
- * NAPI guarantees serialisation of polls of the same device, which
- * provides the guarantee required by efx_process_channel().
- */
-static void efx_update_irq_mod(struct efx_nic *efx, struct efx_channel *channel)
-{
-	int step = efx->irq_mod_step_us;
-
-	if (channel->irq_mod_score < irq_adapt_low_thresh) {
-		if (channel->irq_moderation_us > step) {
-			channel->irq_moderation_us -= step;
-			efx->type->push_irq_moderation(channel);
-		}
-	} else if (channel->irq_mod_score > irq_adapt_high_thresh) {
-		if (channel->irq_moderation_us <
-		    efx->irq_rx_moderation_us) {
-			channel->irq_moderation_us += step;
-			efx->type->push_irq_moderation(channel);
-		}
-	}
-
-	channel->irq_count = 0;
-	channel->irq_mod_score = 0;
-}
-
-static int efx_poll(struct napi_struct *napi, int budget)
-{
-	struct efx_channel *channel =
-		container_of(napi, struct efx_channel, napi_str);
-	struct efx_nic *efx = channel->efx;
-	int spent;
-
-	netif_vdbg(efx, intr, efx->net_dev,
-		   "channel %d NAPI poll executing on CPU %d\n",
-		   channel->channel, raw_smp_processor_id());
-
-	spent = efx_process_channel(channel, budget);
-
-	xdp_do_flush_map();
-
-	if (spent < budget) {
-		if (efx_channel_has_rx_queue(channel) &&
-		    efx->irq_rx_adaptive &&
-		    unlikely(++channel->irq_count == 1000)) {
-			efx_update_irq_mod(efx, channel);
-		}
-
-#ifdef CONFIG_RFS_ACCEL
-		/* Perhaps expire some ARFS filters */
-		mod_delayed_work(system_wq, &channel->filter_work, 0);
-#endif
-
-		/* There is no race here; although napi_disable() will
-		 * only wait for napi_complete(), this isn't a problem
-		 * since efx_nic_eventq_read_ack() will have no effect if
-		 * interrupts have already been disabled.
-		 */
-		if (napi_complete_done(napi, spent))
-			efx_nic_eventq_read_ack(channel);
-	}
-
-	return spent;
-}
-
 /* Create event queue
  * Event queue memory allocations are done only once.  If the channel
  * is reset, the memory buffer will be reused; this guards against
@@ -1715,45 +1585,6 @@ static int efx_ioctl(struct net_device *net_dev, struct ifreq *ifr, int cmd)
 	return mdio_mii_ioctl(&efx->mdio, data, cmd);
 }
 
-/**************************************************************************
- *
- * NAPI interface
- *
- **************************************************************************/
-
-void efx_init_napi_channel(struct efx_channel *channel)
-{
-	struct efx_nic *efx = channel->efx;
-
-	channel->napi_dev = efx->net_dev;
-	netif_napi_add(channel->napi_dev, &channel->napi_str,
-		       efx_poll, napi_weight);
-}
-
-void efx_init_napi(struct efx_nic *efx)
-{
-	struct efx_channel *channel;
-
-	efx_for_each_channel(channel, efx)
-		efx_init_napi_channel(channel);
-}
-
-void efx_fini_napi_channel(struct efx_channel *channel)
-{
-	if (channel->napi_dev)
-		netif_napi_del(&channel->napi_str);
-
-	channel->napi_dev = NULL;
-}
-
-void efx_fini_napi(struct efx_nic *efx)
-{
-	struct efx_channel *channel;
-
-	efx_for_each_channel(channel, efx)
-		efx_fini_napi_channel(channel);
-}
-
 /**************************************************************************
  *
  * Kernel net device interface
diff --git a/drivers/net/ethernet/sfc/efx_channels.c b/drivers/net/ethernet/sfc/efx_channels.c
new file mode 100644
index 000000000000..f288d13d4d09
--- /dev/null
+++ b/drivers/net/ethernet/sfc/efx_channels.c
@@ -0,0 +1,189 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/****************************************************************************
+ * Driver for Solarflare network controllers and boards
+ * Copyright 2018 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#include "net_driver.h"
+#include <linux/module.h>
+#include "efx_channels.h"
+#include "efx.h"
+#include "efx_common.h"
+#include "tx_common.h"
+#include "rx_common.h"
+#include "nic.h"
+#include "sriov.h"
+
+static unsigned int irq_adapt_low_thresh = 8000;
+module_param(irq_adapt_low_thresh, uint, 0644);
+MODULE_PARM_DESC(irq_adapt_low_thresh,
+		 "Threshold score for reducing IRQ moderation");
+
+static unsigned int irq_adapt_high_thresh = 16000;
+module_param(irq_adapt_high_thresh, uint, 0644);
+MODULE_PARM_DESC(irq_adapt_high_thresh,
+		 "Threshold score for increasing IRQ moderation");
+
+/* This is the weight assigned to each of the (per-channel) virtual
+ * NAPI devices.
+ */
+static int napi_weight = 64;
+
+/**************************************************************************
+ *
+ * NAPI interface
+ *
+ *************************************************************************/
+
+/* Process channel's event queue
+ *
+ * This function is responsible for processing the event queue of a
+ * single channel.  The caller must guarantee that this function will
+ * never be concurrently called more than once on the same channel,
+ * though different channels may be being processed concurrently.
+ */
+static int efx_process_channel(struct efx_channel *channel, int budget)
+{
+	struct efx_tx_queue *tx_queue;
+	struct list_head rx_list;
+	int spent;
+
+	if (unlikely(!channel->enabled))
+		return 0;
+
+	/* Prepare the batch receive list */
+	EFX_WARN_ON_PARANOID(channel->rx_list != NULL);
+	INIT_LIST_HEAD(&rx_list);
+	channel->rx_list = &rx_list;
+
+	efx_for_each_channel_tx_queue(tx_queue, channel) {
+		tx_queue->pkts_compl = 0;
+		tx_queue->bytes_compl = 0;
+	}
+
+	spent = efx_nic_process_eventq(channel, budget);
+	if (spent && efx_channel_has_rx_queue(channel)) {
+		struct efx_rx_queue *rx_queue =
+			efx_channel_get_rx_queue(channel);
+
+		efx_rx_flush_packet(channel);
+		efx_fast_push_rx_descriptors(rx_queue, true);
+	}
+
+	/* Update BQL */
+	efx_for_each_channel_tx_queue(tx_queue, channel) {
+		if (tx_queue->bytes_compl) {
+			netdev_tx_completed_queue(tx_queue->core_txq,
+						  tx_queue->pkts_compl,
+						  tx_queue->bytes_compl);
+		}
+	}
+
+	/* Receive any packets we queued up */
+	netif_receive_skb_list(channel->rx_list);
+	channel->rx_list = NULL;
+
+	return spent;
+}
+
+static void efx_update_irq_mod(struct efx_nic *efx, struct efx_channel *channel)
+{
+	int step = efx->irq_mod_step_us;
+
+	if (channel->irq_mod_score < irq_adapt_low_thresh) {
+		if (channel->irq_moderation_us > step) {
+			channel->irq_moderation_us -= step;
+			efx->type->push_irq_moderation(channel);
+		}
+	} else if (channel->irq_mod_score > irq_adapt_high_thresh) {
+		if (channel->irq_moderation_us <
+		    efx->irq_rx_moderation_us) {
+			channel->irq_moderation_us += step;
+			efx->type->push_irq_moderation(channel);
+		}
+	}
+
+	channel->irq_count = 0;
+	channel->irq_mod_score = 0;
+}
+
+/* NAPI poll handler
+ *
+ * NAPI guarantees serialisation of polls of the same device, which
+ * provides the guarantee required by efx_process_channel().
+ */
+static int efx_poll(struct napi_struct *napi, int budget)
+{
+	struct efx_channel *channel =
+		container_of(napi, struct efx_channel, napi_str);
+	struct efx_nic *efx = channel->efx;
+	int spent;
+
+	netif_vdbg(efx, intr, efx->net_dev,
+		   "channel %d NAPI poll executing on CPU %d\n",
+		   channel->channel, raw_smp_processor_id());
+
+	spent = efx_process_channel(channel, budget);
+
+	xdp_do_flush_map();
+
+	if (spent < budget) {
+		if (efx_channel_has_rx_queue(channel) &&
+		    efx->irq_rx_adaptive &&
+		    unlikely(++channel->irq_count == 1000)) {
+			efx_update_irq_mod(efx, channel);
+		}
+
+#ifdef CONFIG_RFS_ACCEL
+		/* Perhaps expire some ARFS filters */
+		mod_delayed_work(system_wq, &channel->filter_work, 0);
+#endif
+
+		/* There is no race here; although napi_disable() will
+		 * only wait for napi_complete(), this isn't a problem
+		 * since efx_nic_eventq_read_ack() will have no effect if
+		 * interrupts have already been disabled.
+		 */
+		if (napi_complete_done(napi, spent))
+			efx_nic_eventq_read_ack(channel);
+	}
+
+	return spent;
+}
+
+void efx_init_napi_channel(struct efx_channel *channel)
+{
+	struct efx_nic *efx = channel->efx;
+
+	channel->napi_dev = efx->net_dev;
+	netif_napi_add(channel->napi_dev, &channel->napi_str,
+		       efx_poll, napi_weight);
+}
+
+void efx_init_napi(struct efx_nic *efx)
+{
+	struct efx_channel *channel;
+
+	efx_for_each_channel(channel, efx)
+		efx_init_napi_channel(channel);
+}
+
+void efx_fini_napi_channel(struct efx_channel *channel)
+{
+	if (channel->napi_dev)
+		netif_napi_del(&channel->napi_str);
+
+	channel->napi_dev = NULL;
+}
+
+void efx_fini_napi(struct efx_nic *efx)
+{
+	struct efx_channel *channel;
+
+	efx_for_each_channel(channel, efx)
+		efx_fini_napi_channel(channel);
+}
-- 
cgit v1.2.3-59-g8ed1b


From e20ba5b1d13fbe558cf1f654eaef347f17a36f7a Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Wed, 8 Jan 2020 16:12:34 +0000
Subject: sfc: move channel start/stop code

Also includes interrupt enabling/disabling code.
Small code styling fixes included.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/efx.c          | 119 --------------------
 drivers/net/ethernet/sfc/efx_channels.c | 190 ++++++++++++++++++++++++++++++++
 drivers/net/ethernet/sfc/efx_common.c   |  69 ------------
 3 files changed, 190 insertions(+), 188 deletions(-)

diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index 530185636afd..8c1cf4cb9c14 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -1086,125 +1086,6 @@ void efx_clear_interrupt_affinity(struct efx_nic *efx __attribute__ ((unused)))
 }
 #endif /* CONFIG_SMP */
 
-int efx_soft_enable_interrupts(struct efx_nic *efx)
-{
-	struct efx_channel *channel, *end_channel;
-	int rc;
-
-	BUG_ON(efx->state == STATE_DISABLED);
-
-	efx->irq_soft_enabled = true;
-	smp_wmb();
-
-	efx_for_each_channel(channel, efx) {
-		if (!channel->type->keep_eventq) {
-			rc = efx_init_eventq(channel);
-			if (rc)
-				goto fail;
-		}
-		efx_start_eventq(channel);
-	}
-
-	efx_mcdi_mode_event(efx);
-
-	return 0;
-fail:
-	end_channel = channel;
-	efx_for_each_channel(channel, efx) {
-		if (channel == end_channel)
-			break;
-		efx_stop_eventq(channel);
-		if (!channel->type->keep_eventq)
-			efx_fini_eventq(channel);
-	}
-
-	return rc;
-}
-
-void efx_soft_disable_interrupts(struct efx_nic *efx)
-{
-	struct efx_channel *channel;
-
-	if (efx->state == STATE_DISABLED)
-		return;
-
-	efx_mcdi_mode_poll(efx);
-
-	efx->irq_soft_enabled = false;
-	smp_wmb();
-
-	if (efx->legacy_irq)
-		synchronize_irq(efx->legacy_irq);
-
-	efx_for_each_channel(channel, efx) {
-		if (channel->irq)
-			synchronize_irq(channel->irq);
-
-		efx_stop_eventq(channel);
-		if (!channel->type->keep_eventq)
-			efx_fini_eventq(channel);
-	}
-
-	/* Flush the asynchronous MCDI request queue */
-	efx_mcdi_flush_async(efx);
-}
-
-int efx_enable_interrupts(struct efx_nic *efx)
-{
-	struct efx_channel *channel, *end_channel;
-	int rc;
-
-	BUG_ON(efx->state == STATE_DISABLED);
-
-	if (efx->eeh_disabled_legacy_irq) {
-		enable_irq(efx->legacy_irq);
-		efx->eeh_disabled_legacy_irq = false;
-	}
-
-	efx->type->irq_enable_master(efx);
-
-	efx_for_each_channel(channel, efx) {
-		if (channel->type->keep_eventq) {
-			rc = efx_init_eventq(channel);
-			if (rc)
-				goto fail;
-		}
-	}
-
-	rc = efx_soft_enable_interrupts(efx);
-	if (rc)
-		goto fail;
-
-	return 0;
-
-fail:
-	end_channel = channel;
-	efx_for_each_channel(channel, efx) {
-		if (channel == end_channel)
-			break;
-		if (channel->type->keep_eventq)
-			efx_fini_eventq(channel);
-	}
-
-	efx->type->irq_disable_non_ev(efx);
-
-	return rc;
-}
-
-void efx_disable_interrupts(struct efx_nic *efx)
-{
-	struct efx_channel *channel;
-
-	efx_soft_disable_interrupts(efx);
-
-	efx_for_each_channel(channel, efx) {
-		if (channel->type->keep_eventq)
-			efx_fini_eventq(channel);
-	}
-
-	efx->type->irq_disable_non_ev(efx);
-}
-
 void efx_remove_interrupts(struct efx_nic *efx)
 {
 	struct efx_channel *channel;
diff --git a/drivers/net/ethernet/sfc/efx_channels.c b/drivers/net/ethernet/sfc/efx_channels.c
index f288d13d4d09..e0d183bd0e78 100644
--- a/drivers/net/ethernet/sfc/efx_channels.c
+++ b/drivers/net/ethernet/sfc/efx_channels.c
@@ -33,6 +33,196 @@ MODULE_PARM_DESC(irq_adapt_high_thresh,
  */
 static int napi_weight = 64;
 
+/*************
+ * START/STOP
+ *************/
+
+int efx_soft_enable_interrupts(struct efx_nic *efx)
+{
+	struct efx_channel *channel, *end_channel;
+	int rc;
+
+	BUG_ON(efx->state == STATE_DISABLED);
+
+	efx->irq_soft_enabled = true;
+	smp_wmb();
+
+	efx_for_each_channel(channel, efx) {
+		if (!channel->type->keep_eventq) {
+			rc = efx_init_eventq(channel);
+			if (rc)
+				goto fail;
+		}
+		efx_start_eventq(channel);
+	}
+
+	efx_mcdi_mode_event(efx);
+
+	return 0;
+fail:
+	end_channel = channel;
+	efx_for_each_channel(channel, efx) {
+		if (channel == end_channel)
+			break;
+		efx_stop_eventq(channel);
+		if (!channel->type->keep_eventq)
+			efx_fini_eventq(channel);
+	}
+
+	return rc;
+}
+
+void efx_soft_disable_interrupts(struct efx_nic *efx)
+{
+	struct efx_channel *channel;
+
+	if (efx->state == STATE_DISABLED)
+		return;
+
+	efx_mcdi_mode_poll(efx);
+
+	efx->irq_soft_enabled = false;
+	smp_wmb();
+
+	if (efx->legacy_irq)
+		synchronize_irq(efx->legacy_irq);
+
+	efx_for_each_channel(channel, efx) {
+		if (channel->irq)
+			synchronize_irq(channel->irq);
+
+		efx_stop_eventq(channel);
+		if (!channel->type->keep_eventq)
+			efx_fini_eventq(channel);
+	}
+
+	/* Flush the asynchronous MCDI request queue */
+	efx_mcdi_flush_async(efx);
+}
+
+int efx_enable_interrupts(struct efx_nic *efx)
+{
+	struct efx_channel *channel, *end_channel;
+	int rc;
+
+	/* TODO: Is this really a bug? */
+	BUG_ON(efx->state == STATE_DISABLED);
+
+	if (efx->eeh_disabled_legacy_irq) {
+		enable_irq(efx->legacy_irq);
+		efx->eeh_disabled_legacy_irq = false;
+	}
+
+	efx->type->irq_enable_master(efx);
+
+	efx_for_each_channel(channel, efx) {
+		if (channel->type->keep_eventq) {
+			rc = efx_init_eventq(channel);
+			if (rc)
+				goto fail;
+		}
+	}
+
+	rc = efx_soft_enable_interrupts(efx);
+	if (rc)
+		goto fail;
+
+	return 0;
+
+fail:
+	end_channel = channel;
+	efx_for_each_channel(channel, efx) {
+		if (channel == end_channel)
+			break;
+		if (channel->type->keep_eventq)
+			efx_fini_eventq(channel);
+	}
+
+	efx->type->irq_disable_non_ev(efx);
+
+	return rc;
+}
+
+void efx_disable_interrupts(struct efx_nic *efx)
+{
+	struct efx_channel *channel;
+
+	efx_soft_disable_interrupts(efx);
+
+	efx_for_each_channel(channel, efx) {
+		if (channel->type->keep_eventq)
+			efx_fini_eventq(channel);
+	}
+
+	efx->type->irq_disable_non_ev(efx);
+}
+
+void efx_start_channels(struct efx_nic *efx)
+{
+	struct efx_tx_queue *tx_queue;
+	struct efx_rx_queue *rx_queue;
+	struct efx_channel *channel;
+
+	efx_for_each_channel(channel, efx) {
+		efx_for_each_channel_tx_queue(tx_queue, channel) {
+			efx_init_tx_queue(tx_queue);
+			atomic_inc(&efx->active_queues);
+		}
+
+		efx_for_each_channel_rx_queue(rx_queue, channel) {
+			efx_init_rx_queue(rx_queue);
+			atomic_inc(&efx->active_queues);
+			efx_stop_eventq(channel);
+			efx_fast_push_rx_descriptors(rx_queue, false);
+			efx_start_eventq(channel);
+		}
+
+		WARN_ON(channel->rx_pkt_n_frags);
+	}
+}
+
+void efx_stop_channels(struct efx_nic *efx)
+{
+	struct efx_tx_queue *tx_queue;
+	struct efx_rx_queue *rx_queue;
+	struct efx_channel *channel;
+	int rc;
+
+	/* Stop RX refill */
+	efx_for_each_channel(channel, efx) {
+		efx_for_each_channel_rx_queue(rx_queue, channel)
+			rx_queue->refill_enabled = false;
+	}
+
+	efx_for_each_channel(channel, efx) {
+		/* RX packet processing is pipelined, so wait for the
+		 * NAPI handler to complete.  At least event queue 0
+		 * might be kept active by non-data events, so don't
+		 * use napi_synchronize() but actually disable NAPI
+		 * temporarily.
+		 */
+		if (efx_channel_has_rx_queue(channel)) {
+			efx_stop_eventq(channel);
+			efx_start_eventq(channel);
+		}
+	}
+
+	rc = efx->type->fini_dmaq(efx);
+	if (rc) {
+		netif_err(efx, drv, efx->net_dev, "failed to flush queues\n");
+	} else {
+		netif_dbg(efx, drv, efx->net_dev,
+			  "successfully flushed all queues\n");
+	}
+
+	efx_for_each_channel(channel, efx) {
+		efx_for_each_channel_rx_queue(rx_queue, channel)
+			efx_fini_rx_queue(rx_queue);
+		efx_for_each_possible_channel_tx_queue(tx_queue, channel)
+			efx_fini_tx_queue(tx_queue);
+	}
+}
+
 /**************************************************************************
  *
  * NAPI interface
diff --git a/drivers/net/ethernet/sfc/efx_common.c b/drivers/net/ethernet/sfc/efx_common.c
index f1c487de3f04..d0fc51a108bf 100644
--- a/drivers/net/ethernet/sfc/efx_common.c
+++ b/drivers/net/ethernet/sfc/efx_common.c
@@ -238,30 +238,6 @@ void efx_start_monitor(struct efx_nic *efx)
  *
  *************************************************************************/
 
-void efx_start_channels(struct efx_nic *efx)
-{
-	struct efx_tx_queue *tx_queue;
-	struct efx_rx_queue *rx_queue;
-	struct efx_channel *channel;
-
-	efx_for_each_channel(channel, efx) {
-		efx_for_each_channel_tx_queue(tx_queue, channel) {
-			efx_init_tx_queue(tx_queue);
-			atomic_inc(&efx->active_queues);
-		}
-
-		efx_for_each_channel_rx_queue(rx_queue, channel) {
-			efx_init_rx_queue(rx_queue);
-			atomic_inc(&efx->active_queues);
-			efx_stop_eventq(channel);
-			efx_fast_push_rx_descriptors(rx_queue, false);
-			efx_start_eventq(channel);
-		}
-
-		WARN_ON(channel->rx_pkt_n_frags);
-	}
-}
-
 /* Channels are shutdown and reinitialised whilst the NIC is running
  * to propagate configuration changes (mtu, checksum offload), or
  * to clear hardware error conditions
@@ -342,51 +318,6 @@ static void efx_start_datapath(struct efx_nic *efx)
 		netif_tx_wake_all_queues(efx->net_dev);
 }
 
-void efx_stop_channels(struct efx_nic *efx)
-{
-	struct efx_tx_queue *tx_queue;
-	struct efx_rx_queue *rx_queue;
-	struct efx_channel *channel;
-	int rc = 0;
-
-	/* Stop RX refill */
-	efx_for_each_channel(channel, efx) {
-		efx_for_each_channel_rx_queue(rx_queue, channel)
-			rx_queue->refill_enabled = false;
-	}
-
-	efx_for_each_channel(channel, efx) {
-		/* RX packet processing is pipelined, so wait for the
-		 * NAPI handler to complete.  At least event queue 0
-		 * might be kept active by non-data events, so don't
-		 * use napi_synchronize() but actually disable NAPI
-		 * temporarily.
-		 */
-		if (efx_channel_has_rx_queue(channel)) {
-			efx_stop_eventq(channel);
-			efx_start_eventq(channel);
-		}
-	}
-
-	if (efx->type->fini_dmaq)
-		rc = efx->type->fini_dmaq(efx);
-
-	if (rc) {
-		netif_err(efx, drv, efx->net_dev, "failed to flush queues\n");
-	} else {
-		netif_dbg(efx, drv, efx->net_dev,
-			  "successfully flushed all queues\n");
-	}
-
-	efx_for_each_channel(channel, efx) {
-		efx_for_each_channel_rx_queue(rx_queue, channel)
-			efx_fini_rx_queue(rx_queue);
-		efx_for_each_possible_channel_tx_queue(tx_queue, channel)
-			efx_fini_tx_queue(tx_queue);
-	}
-	efx->xdp_rxq_info_failed = false;
-}
-
 static void efx_stop_datapath(struct efx_nic *efx)
 {
 	EFX_ASSERT_RESET_SERIALISED(efx);
-- 
cgit v1.2.3-59-g8ed1b


From 83975485077dda755a8db7dde30c80fd9cd182e5 Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Wed, 8 Jan 2020 16:12:56 +0000
Subject: sfc: move channel alloc/removal code

Reallocation and copying code is included, as well as some housekeeping
code.
Other files have been patched up a bit to accommodate the changes.

Small code styling fixes included.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/efx.c          | 420 ------------------------------
 drivers/net/ethernet/sfc/efx.h          |   4 -
 drivers/net/ethernet/sfc/efx_channels.c | 440 ++++++++++++++++++++++++++++++++
 drivers/net/ethernet/sfc/efx_common.c   |  11 -
 drivers/net/ethernet/sfc/siena_sriov.c  |   1 +
 5 files changed, 441 insertions(+), 435 deletions(-)

diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index 8c1cf4cb9c14..955fcf714992 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -96,13 +96,6 @@ static unsigned int rx_irq_mod_usec = 60;
  */
 static unsigned int tx_irq_mod_usec = 150;
 
-/* This is the first interrupt mode to try out of:
- * 0 => MSI-X
- * 1 => MSI
- * 2 => legacy
- */
-static unsigned int interrupt_mode;
-
 /* This is the requested number of CPUs to use for Receive-Side Scaling (RSS),
  * i.e. the number of CPUs among which we may distribute simultaneous
  * interrupt handling.
@@ -244,349 +237,11 @@ void efx_remove_eventq(struct efx_channel *channel)
  *
  *************************************************************************/
 
-/* Allocate and initialise a channel structure. */
-struct efx_channel *
-efx_alloc_channel(struct efx_nic *efx, int i, struct efx_channel *old_channel)
-{
-	struct efx_channel *channel;
-	struct efx_rx_queue *rx_queue;
-	struct efx_tx_queue *tx_queue;
-	int j;
-
-	channel = kzalloc(sizeof(*channel), GFP_KERNEL);
-	if (!channel)
-		return NULL;
-
-	channel->efx = efx;
-	channel->channel = i;
-	channel->type = &efx_default_channel_type;
-
-	for (j = 0; j < EFX_TXQ_TYPES; j++) {
-		tx_queue = &channel->tx_queue[j];
-		tx_queue->efx = efx;
-		tx_queue->queue = i * EFX_TXQ_TYPES + j;
-		tx_queue->channel = channel;
-	}
-
-#ifdef CONFIG_RFS_ACCEL
-	INIT_DELAYED_WORK(&channel->filter_work, efx_filter_rfs_expire);
-#endif
-
-	rx_queue = &channel->rx_queue;
-	rx_queue->efx = efx;
-	timer_setup(&rx_queue->slow_fill, efx_rx_slow_fill, 0);
-
-	return channel;
-}
-
-/* Allocate and initialise a channel structure, copying parameters
- * (but not resources) from an old channel structure.
- */
-struct efx_channel *efx_copy_channel(const struct efx_channel *old_channel)
-{
-	struct efx_channel *channel;
-	struct efx_rx_queue *rx_queue;
-	struct efx_tx_queue *tx_queue;
-	int j;
-
-	channel = kmalloc(sizeof(*channel), GFP_KERNEL);
-	if (!channel)
-		return NULL;
-
-	*channel = *old_channel;
-
-	channel->napi_dev = NULL;
-	INIT_HLIST_NODE(&channel->napi_str.napi_hash_node);
-	channel->napi_str.napi_id = 0;
-	channel->napi_str.state = 0;
-	memset(&channel->eventq, 0, sizeof(channel->eventq));
-
-	for (j = 0; j < EFX_TXQ_TYPES; j++) {
-		tx_queue = &channel->tx_queue[j];
-		if (tx_queue->channel)
-			tx_queue->channel = channel;
-		tx_queue->buffer = NULL;
-		memset(&tx_queue->txd, 0, sizeof(tx_queue->txd));
-	}
-
-	rx_queue = &channel->rx_queue;
-	rx_queue->buffer = NULL;
-	memset(&rx_queue->rxd, 0, sizeof(rx_queue->rxd));
-	timer_setup(&rx_queue->slow_fill, efx_rx_slow_fill, 0);
-#ifdef CONFIG_RFS_ACCEL
-	INIT_DELAYED_WORK(&channel->filter_work, efx_filter_rfs_expire);
-#endif
-
-	return channel;
-}
-
-static int efx_probe_channel(struct efx_channel *channel)
-{
-	struct efx_tx_queue *tx_queue;
-	struct efx_rx_queue *rx_queue;
-	int rc;
-
-	netif_dbg(channel->efx, probe, channel->efx->net_dev,
-		  "creating channel %d\n", channel->channel);
-
-	rc = channel->type->pre_probe(channel);
-	if (rc)
-		goto fail;
-
-	rc = efx_probe_eventq(channel);
-	if (rc)
-		goto fail;
-
-	efx_for_each_channel_tx_queue(tx_queue, channel) {
-		rc = efx_probe_tx_queue(tx_queue);
-		if (rc)
-			goto fail;
-	}
-
-	efx_for_each_channel_rx_queue(rx_queue, channel) {
-		rc = efx_probe_rx_queue(rx_queue);
-		if (rc)
-			goto fail;
-	}
-
-	channel->rx_list = NULL;
-
-	return 0;
-
-fail:
-	efx_remove_channel(channel);
-	return rc;
-}
-
-void efx_get_channel_name(struct efx_channel *channel, char *buf, size_t len)
-{
-	struct efx_nic *efx = channel->efx;
-	const char *type;
-	int number;
-
-	number = channel->channel;
-
-	if (number >= efx->xdp_channel_offset &&
-	    !WARN_ON_ONCE(!efx->n_xdp_channels)) {
-		type = "-xdp";
-		number -= efx->xdp_channel_offset;
-	} else if (efx->tx_channel_offset == 0) {
-		type = "";
-	} else if (number < efx->tx_channel_offset) {
-		type = "-rx";
-	} else {
-		type = "-tx";
-		number -= efx->tx_channel_offset;
-	}
-	snprintf(buf, len, "%s%s-%d", efx->name, type, number);
-}
-
-void efx_set_channel_names(struct efx_nic *efx)
-{
-	struct efx_channel *channel;
-
-	efx_for_each_channel(channel, efx)
-		channel->type->get_name(channel,
-					efx->msi_context[channel->channel].name,
-					sizeof(efx->msi_context[0].name));
-}
-
-int efx_probe_channels(struct efx_nic *efx)
-{
-	struct efx_channel *channel;
-	int rc;
-
-	/* Restart special buffer allocation */
-	efx->next_buffer_table = 0;
-
-	/* Probe channels in reverse, so that any 'extra' channels
-	 * use the start of the buffer table. This allows the traffic
-	 * channels to be resized without moving them or wasting the
-	 * entries before them.
-	 */
-	efx_for_each_channel_rev(channel, efx) {
-		rc = efx_probe_channel(channel);
-		if (rc) {
-			netif_err(efx, probe, efx->net_dev,
-				  "failed to create channel %d\n",
-				  channel->channel);
-			goto fail;
-		}
-	}
-	efx_set_channel_names(efx);
-
-	return 0;
-
-fail:
-	efx_remove_channels(efx);
-	return rc;
-}
-
-void efx_remove_channel(struct efx_channel *channel)
-{
-	struct efx_tx_queue *tx_queue;
-	struct efx_rx_queue *rx_queue;
-
-	netif_dbg(channel->efx, drv, channel->efx->net_dev,
-		  "destroy chan %d\n", channel->channel);
-
-	efx_for_each_channel_rx_queue(rx_queue, channel)
-		efx_remove_rx_queue(rx_queue);
-	efx_for_each_possible_channel_tx_queue(tx_queue, channel)
-		efx_remove_tx_queue(tx_queue);
-	efx_remove_eventq(channel);
-	channel->type->post_remove(channel);
-}
-
-void efx_remove_channels(struct efx_nic *efx)
-{
-	struct efx_channel *channel;
-
-	efx_for_each_channel(channel, efx)
-		efx_remove_channel(channel);
-
-	kfree(efx->xdp_tx_queues);
-}
-
-int efx_realloc_channels(struct efx_nic *efx, u32 rxq_entries, u32 txq_entries)
-{
-	struct efx_channel *other_channel[EFX_MAX_CHANNELS], *channel;
-	u32 old_rxq_entries, old_txq_entries;
-	unsigned i, next_buffer_table = 0;
-	int rc, rc2;
-
-	rc = efx_check_disabled(efx);
-	if (rc)
-		return rc;
-
-	/* Not all channels should be reallocated. We must avoid
-	 * reallocating their buffer table entries.
-	 */
-	efx_for_each_channel(channel, efx) {
-		struct efx_rx_queue *rx_queue;
-		struct efx_tx_queue *tx_queue;
-
-		if (channel->type->copy)
-			continue;
-		next_buffer_table = max(next_buffer_table,
-					channel->eventq.index +
-					channel->eventq.entries);
-		efx_for_each_channel_rx_queue(rx_queue, channel)
-			next_buffer_table = max(next_buffer_table,
-						rx_queue->rxd.index +
-						rx_queue->rxd.entries);
-		efx_for_each_channel_tx_queue(tx_queue, channel)
-			next_buffer_table = max(next_buffer_table,
-						tx_queue->txd.index +
-						tx_queue->txd.entries);
-	}
-
-	efx_device_detach_sync(efx);
-	efx_stop_all(efx);
-	efx_soft_disable_interrupts(efx);
-
-	/* Clone channels (where possible) */
-	memset(other_channel, 0, sizeof(other_channel));
-	for (i = 0; i < efx->n_channels; i++) {
-		channel = efx->channel[i];
-		if (channel->type->copy)
-			channel = channel->type->copy(channel);
-		if (!channel) {
-			rc = -ENOMEM;
-			goto out;
-		}
-		other_channel[i] = channel;
-	}
-
-	/* Swap entry counts and channel pointers */
-	old_rxq_entries = efx->rxq_entries;
-	old_txq_entries = efx->txq_entries;
-	efx->rxq_entries = rxq_entries;
-	efx->txq_entries = txq_entries;
-	for (i = 0; i < efx->n_channels; i++) {
-		channel = efx->channel[i];
-		efx->channel[i] = other_channel[i];
-		other_channel[i] = channel;
-	}
-
-	/* Restart buffer table allocation */
-	efx->next_buffer_table = next_buffer_table;
-
-	for (i = 0; i < efx->n_channels; i++) {
-		channel = efx->channel[i];
-		if (!channel->type->copy)
-			continue;
-		rc = efx_probe_channel(channel);
-		if (rc)
-			goto rollback;
-		efx_init_napi_channel(efx->channel[i]);
-	}
-
-out:
-	/* Destroy unused channel structures */
-	for (i = 0; i < efx->n_channels; i++) {
-		channel = other_channel[i];
-		if (channel && channel->type->copy) {
-			efx_fini_napi_channel(channel);
-			efx_remove_channel(channel);
-			kfree(channel);
-		}
-	}
-
-	rc2 = efx_soft_enable_interrupts(efx);
-	if (rc2) {
-		rc = rc ? rc : rc2;
-		netif_err(efx, drv, efx->net_dev,
-			  "unable to restart interrupts on channel reallocation\n");
-		efx_schedule_reset(efx, RESET_TYPE_DISABLE);
-	} else {
-		efx_start_all(efx);
-		efx_device_attach_if_not_resetting(efx);
-	}
-	return rc;
-
-rollback:
-	/* Swap back */
-	efx->rxq_entries = old_rxq_entries;
-	efx->txq_entries = old_txq_entries;
-	for (i = 0; i < efx->n_channels; i++) {
-		channel = efx->channel[i];
-		efx->channel[i] = other_channel[i];
-		other_channel[i] = channel;
-	}
-	goto out;
-}
-
 void efx_schedule_slow_fill(struct efx_rx_queue *rx_queue)
 {
 	mod_timer(&rx_queue->slow_fill, jiffies + msecs_to_jiffies(10));
 }
 
-bool efx_default_channel_want_txqs(struct efx_channel *channel)
-{
-	return channel->channel - channel->efx->tx_channel_offset <
-		channel->efx->n_tx_channels;
-}
-
-static const struct efx_channel_type efx_default_channel_type = {
-	.pre_probe		= efx_channel_dummy_op_int,
-	.post_remove		= efx_channel_dummy_op_void,
-	.get_name		= efx_get_channel_name,
-	.copy			= efx_copy_channel,
-	.want_txqs		= efx_default_channel_want_txqs,
-	.keep_eventq		= false,
-	.want_pio		= true,
-};
-
-int efx_channel_dummy_op_int(struct efx_channel *channel)
-{
-	return 0;
-}
-
-void efx_channel_dummy_op_void(struct efx_channel *channel)
-{
-}
-
 /**************************************************************************
  *
  * Port handling
@@ -1100,52 +755,6 @@ void efx_remove_interrupts(struct efx_nic *efx)
 	efx->legacy_irq = 0;
 }
 
-int efx_set_channels(struct efx_nic *efx)
-{
-	struct efx_channel *channel;
-	struct efx_tx_queue *tx_queue;
-	int xdp_queue_number;
-
-	efx->tx_channel_offset =
-		efx_separate_tx_channels ?
-		efx->n_channels - efx->n_tx_channels : 0;
-
-	if (efx->xdp_tx_queue_count) {
-		EFX_WARN_ON_PARANOID(efx->xdp_tx_queues);
-
-		/* Allocate array for XDP TX queue lookup. */
-		efx->xdp_tx_queues = kcalloc(efx->xdp_tx_queue_count,
-					     sizeof(*efx->xdp_tx_queues),
-					     GFP_KERNEL);
-		if (!efx->xdp_tx_queues)
-			return -ENOMEM;
-	}
-
-	/* We need to mark which channels really have RX and TX
-	 * queues, and adjust the TX queue numbers if we have separate
-	 * RX-only and TX-only channels.
-	 */
-	xdp_queue_number = 0;
-	efx_for_each_channel(channel, efx) {
-		if (channel->channel < efx->n_rx_channels)
-			channel->rx_queue.core_index = channel->channel;
-		else
-			channel->rx_queue.core_index = -1;
-
-		efx_for_each_channel_tx_queue(tx_queue, channel) {
-			tx_queue->queue -= (efx->tx_channel_offset *
-					    EFX_TXQ_TYPES);
-
-			if (efx_channel_is_xdp_tx(channel) &&
-			    xdp_queue_number < efx->xdp_tx_queue_count) {
-				efx->xdp_tx_queues[xdp_queue_number] = tx_queue;
-				xdp_queue_number++;
-			}
-		}
-	}
-	return 0;
-}
-
 static int efx_probe_nic(struct efx_nic *efx)
 {
 	int rc;
@@ -2022,31 +1631,6 @@ static const struct pci_device_id efx_pci_table[] = {
  *
  **************************************************************************/
 
-int efx_init_channels(struct efx_nic *efx)
-{
-	unsigned int i;
-
-	for (i = 0; i < EFX_MAX_CHANNELS; i++) {
-		efx->channel[i] = efx_alloc_channel(efx, i, NULL);
-		if (!efx->channel[i])
-			return -ENOMEM;
-		efx->msi_context[i].efx = efx;
-		efx->msi_context[i].index = i;
-	}
-
-	/* Higher numbered interrupt modes are less capable! */
-	if (WARN_ON_ONCE(efx->type->max_interrupt_mode >
-			 efx->type->min_interrupt_mode)) {
-		return -EIO;
-	}
-	efx->interrupt_mode = max(efx->type->max_interrupt_mode,
-				  interrupt_mode);
-	efx->interrupt_mode = min(efx->type->min_interrupt_mode,
-				  interrupt_mode);
-
-	return 0;
-}
-
 void efx_update_sw_stats(struct efx_nic *efx, u64 *stats)
 {
 	u64 n_rx_nodesc_trunc = 0;
@@ -2829,10 +2413,6 @@ static struct pci_driver efx_pci_driver = {
  *
  *************************************************************************/
 
-module_param(interrupt_mode, uint, 0444);
-MODULE_PARM_DESC(interrupt_mode,
-		 "Interrupt mode (0=>MSIX 1=>MSI 2=>legacy)");
-
 static int __init efx_init_module(void)
 {
 	int rc;
diff --git a/drivers/net/ethernet/sfc/efx.h b/drivers/net/ethernet/sfc/efx.h
index 6ff454f2cb62..2b417e779e82 100644
--- a/drivers/net/ethernet/sfc/efx.h
+++ b/drivers/net/ethernet/sfc/efx.h
@@ -203,10 +203,6 @@ static inline bool efx_rss_active(struct efx_rss_context *ctx)
 	return ctx->context_id != EFX_EF10_RSS_CONTEXT_INVALID;
 }
 
-/* Channels */
-int efx_channel_dummy_op_int(struct efx_channel *channel);
-void efx_channel_dummy_op_void(struct efx_channel *channel);
-
 /* Ethtool support */
 extern const struct ethtool_ops efx_ethtool_ops;
 
diff --git a/drivers/net/ethernet/sfc/efx_channels.c b/drivers/net/ethernet/sfc/efx_channels.c
index e0d183bd0e78..bd5bc77a1d5a 100644
--- a/drivers/net/ethernet/sfc/efx_channels.c
+++ b/drivers/net/ethernet/sfc/efx_channels.c
@@ -18,6 +18,16 @@
 #include "nic.h"
 #include "sriov.h"
 
+/* This is the first interrupt mode to try out of:
+ * 0 => MSI-X
+ * 1 => MSI
+ * 2 => legacy
+ */
+static unsigned int interrupt_mode;
+module_param(interrupt_mode, uint, 0444);
+MODULE_PARM_DESC(interrupt_mode,
+		 "Interrupt mode (0=>MSIX 1=>MSI 2=>legacy)");
+
 static unsigned int irq_adapt_low_thresh = 8000;
 module_param(irq_adapt_low_thresh, uint, 0644);
 MODULE_PARM_DESC(irq_adapt_low_thresh,
@@ -33,6 +43,436 @@ MODULE_PARM_DESC(irq_adapt_high_thresh,
  */
 static int napi_weight = 64;
 
+/***************
+ * Housekeeping
+ ***************/
+
+int efx_channel_dummy_op_int(struct efx_channel *channel)
+{
+	return 0;
+}
+
+void efx_channel_dummy_op_void(struct efx_channel *channel)
+{
+}
+
+static const struct efx_channel_type efx_default_channel_type = {
+	.pre_probe		= efx_channel_dummy_op_int,
+	.post_remove		= efx_channel_dummy_op_void,
+	.get_name		= efx_get_channel_name,
+	.copy			= efx_copy_channel,
+	.want_txqs		= efx_default_channel_want_txqs,
+	.keep_eventq		= false,
+	.want_pio		= true,
+};
+
+/**************************************************************************
+ *
+ * Channel handling
+ *
+ *************************************************************************/
+
+/* Allocate and initialise a channel structure. */
+struct efx_channel *
+efx_alloc_channel(struct efx_nic *efx, int i, struct efx_channel *old_channel)
+{
+	struct efx_rx_queue *rx_queue;
+	struct efx_tx_queue *tx_queue;
+	struct efx_channel *channel;
+	int j;
+
+	channel = kzalloc(sizeof(*channel), GFP_KERNEL);
+	if (!channel)
+		return NULL;
+
+	channel->efx = efx;
+	channel->channel = i;
+	channel->type = &efx_default_channel_type;
+
+	for (j = 0; j < EFX_TXQ_TYPES; j++) {
+		tx_queue = &channel->tx_queue[j];
+		tx_queue->efx = efx;
+		tx_queue->queue = i * EFX_TXQ_TYPES + j;
+		tx_queue->channel = channel;
+	}
+
+#ifdef CONFIG_RFS_ACCEL
+	INIT_DELAYED_WORK(&channel->filter_work, efx_filter_rfs_expire);
+#endif
+
+	rx_queue = &channel->rx_queue;
+	rx_queue->efx = efx;
+	timer_setup(&rx_queue->slow_fill, efx_rx_slow_fill, 0);
+
+	return channel;
+}
+
+int efx_init_channels(struct efx_nic *efx)
+{
+	unsigned int i;
+
+	for (i = 0; i < EFX_MAX_CHANNELS; i++) {
+		efx->channel[i] = efx_alloc_channel(efx, i, NULL);
+		if (!efx->channel[i])
+			return -ENOMEM;
+		efx->msi_context[i].efx = efx;
+		efx->msi_context[i].index = i;
+	}
+
+	/* Higher numbered interrupt modes are less capable! */
+	if (WARN_ON_ONCE(efx->type->max_interrupt_mode >
+			 efx->type->min_interrupt_mode)) {
+		return -EIO;
+	}
+	efx->interrupt_mode = max(efx->type->max_interrupt_mode,
+				  interrupt_mode);
+	efx->interrupt_mode = min(efx->type->min_interrupt_mode,
+				  interrupt_mode);
+
+	return 0;
+}
+
+void efx_fini_channels(struct efx_nic *efx)
+{
+	unsigned int i;
+
+	for (i = 0; i < EFX_MAX_CHANNELS; i++)
+		if (efx->channel[i]) {
+			kfree(efx->channel[i]);
+			efx->channel[i] = NULL;
+		}
+}
+
+/* Allocate and initialise a channel structure, copying parameters
+ * (but not resources) from an old channel structure.
+ */
+struct efx_channel *efx_copy_channel(const struct efx_channel *old_channel)
+{
+	struct efx_rx_queue *rx_queue;
+	struct efx_tx_queue *tx_queue;
+	struct efx_channel *channel;
+	int j;
+
+	channel = kmalloc(sizeof(*channel), GFP_KERNEL);
+	if (!channel)
+		return NULL;
+
+	*channel = *old_channel;
+
+	channel->napi_dev = NULL;
+	INIT_HLIST_NODE(&channel->napi_str.napi_hash_node);
+	channel->napi_str.napi_id = 0;
+	channel->napi_str.state = 0;
+	memset(&channel->eventq, 0, sizeof(channel->eventq));
+
+	for (j = 0; j < EFX_TXQ_TYPES; j++) {
+		tx_queue = &channel->tx_queue[j];
+		if (tx_queue->channel)
+			tx_queue->channel = channel;
+		tx_queue->buffer = NULL;
+		memset(&tx_queue->txd, 0, sizeof(tx_queue->txd));
+	}
+
+	rx_queue = &channel->rx_queue;
+	rx_queue->buffer = NULL;
+	memset(&rx_queue->rxd, 0, sizeof(rx_queue->rxd));
+	timer_setup(&rx_queue->slow_fill, efx_rx_slow_fill, 0);
+#ifdef CONFIG_RFS_ACCEL
+	INIT_DELAYED_WORK(&channel->filter_work, efx_filter_rfs_expire);
+#endif
+
+	return channel;
+}
+
+static int efx_probe_channel(struct efx_channel *channel)
+{
+	struct efx_tx_queue *tx_queue;
+	struct efx_rx_queue *rx_queue;
+	int rc;
+
+	netif_dbg(channel->efx, probe, channel->efx->net_dev,
+		  "creating channel %d\n", channel->channel);
+
+	rc = channel->type->pre_probe(channel);
+	if (rc)
+		goto fail;
+
+	rc = efx_probe_eventq(channel);
+	if (rc)
+		goto fail;
+
+	efx_for_each_channel_tx_queue(tx_queue, channel) {
+		rc = efx_probe_tx_queue(tx_queue);
+		if (rc)
+			goto fail;
+	}
+
+	efx_for_each_channel_rx_queue(rx_queue, channel) {
+		rc = efx_probe_rx_queue(rx_queue);
+		if (rc)
+			goto fail;
+	}
+
+	channel->rx_list = NULL;
+
+	return 0;
+
+fail:
+	efx_remove_channel(channel);
+	return rc;
+}
+
+void efx_get_channel_name(struct efx_channel *channel, char *buf, size_t len)
+{
+	struct efx_nic *efx = channel->efx;
+	const char *type;
+	int number;
+
+	number = channel->channel;
+
+	if (number >= efx->xdp_channel_offset &&
+	    !WARN_ON_ONCE(!efx->n_xdp_channels)) {
+		type = "-xdp";
+		number -= efx->xdp_channel_offset;
+	} else if (efx->tx_channel_offset == 0) {
+		type = "";
+	} else if (number < efx->tx_channel_offset) {
+		type = "-rx";
+	} else {
+		type = "-tx";
+		number -= efx->tx_channel_offset;
+	}
+	snprintf(buf, len, "%s%s-%d", efx->name, type, number);
+}
+
+void efx_set_channel_names(struct efx_nic *efx)
+{
+	struct efx_channel *channel;
+
+	efx_for_each_channel(channel, efx)
+		channel->type->get_name(channel,
+					efx->msi_context[channel->channel].name,
+					sizeof(efx->msi_context[0].name));
+}
+
+int efx_probe_channels(struct efx_nic *efx)
+{
+	struct efx_channel *channel;
+	int rc;
+
+	/* Restart special buffer allocation */
+	efx->next_buffer_table = 0;
+
+	/* Probe channels in reverse, so that any 'extra' channels
+	 * use the start of the buffer table. This allows the traffic
+	 * channels to be resized without moving them or wasting the
+	 * entries before them.
+	 */
+	efx_for_each_channel_rev(channel, efx) {
+		rc = efx_probe_channel(channel);
+		if (rc) {
+			netif_err(efx, probe, efx->net_dev,
+				  "failed to create channel %d\n",
+				  channel->channel);
+			goto fail;
+		}
+	}
+	efx_set_channel_names(efx);
+
+	return 0;
+
+fail:
+	efx_remove_channels(efx);
+	return rc;
+}
+
+void efx_remove_channel(struct efx_channel *channel)
+{
+	struct efx_tx_queue *tx_queue;
+	struct efx_rx_queue *rx_queue;
+
+	netif_dbg(channel->efx, drv, channel->efx->net_dev,
+		  "destroy chan %d\n", channel->channel);
+
+	efx_for_each_channel_rx_queue(rx_queue, channel)
+		efx_remove_rx_queue(rx_queue);
+	efx_for_each_possible_channel_tx_queue(tx_queue, channel)
+		efx_remove_tx_queue(tx_queue);
+	efx_remove_eventq(channel);
+	channel->type->post_remove(channel);
+}
+
+void efx_remove_channels(struct efx_nic *efx)
+{
+	struct efx_channel *channel;
+
+	efx_for_each_channel(channel, efx)
+		efx_remove_channel(channel);
+
+	kfree(efx->xdp_tx_queues);
+}
+
+int efx_realloc_channels(struct efx_nic *efx, u32 rxq_entries, u32 txq_entries)
+{
+	struct efx_channel *other_channel[EFX_MAX_CHANNELS], *channel;
+	unsigned int i, next_buffer_table = 0;
+	u32 old_rxq_entries, old_txq_entries;
+	int rc, rc2;
+
+	rc = efx_check_disabled(efx);
+	if (rc)
+		return rc;
+
+	/* Not all channels should be reallocated. We must avoid
+	 * reallocating their buffer table entries.
+	 */
+	efx_for_each_channel(channel, efx) {
+		struct efx_rx_queue *rx_queue;
+		struct efx_tx_queue *tx_queue;
+
+		if (channel->type->copy)
+			continue;
+		next_buffer_table = max(next_buffer_table,
+					channel->eventq.index +
+					channel->eventq.entries);
+		efx_for_each_channel_rx_queue(rx_queue, channel)
+			next_buffer_table = max(next_buffer_table,
+						rx_queue->rxd.index +
+						rx_queue->rxd.entries);
+		efx_for_each_channel_tx_queue(tx_queue, channel)
+			next_buffer_table = max(next_buffer_table,
+						tx_queue->txd.index +
+						tx_queue->txd.entries);
+	}
+
+	efx_device_detach_sync(efx);
+	efx_stop_all(efx);
+	efx_soft_disable_interrupts(efx);
+
+	/* Clone channels (where possible) */
+	memset(other_channel, 0, sizeof(other_channel));
+	for (i = 0; i < efx->n_channels; i++) {
+		channel = efx->channel[i];
+		if (channel->type->copy)
+			channel = channel->type->copy(channel);
+		if (!channel) {
+			rc = -ENOMEM;
+			goto out;
+		}
+		other_channel[i] = channel;
+	}
+
+	/* Swap entry counts and channel pointers */
+	old_rxq_entries = efx->rxq_entries;
+	old_txq_entries = efx->txq_entries;
+	efx->rxq_entries = rxq_entries;
+	efx->txq_entries = txq_entries;
+	for (i = 0; i < efx->n_channels; i++) {
+		channel = efx->channel[i];
+		efx->channel[i] = other_channel[i];
+		other_channel[i] = channel;
+	}
+
+	/* Restart buffer table allocation */
+	efx->next_buffer_table = next_buffer_table;
+
+	for (i = 0; i < efx->n_channels; i++) {
+		channel = efx->channel[i];
+		if (!channel->type->copy)
+			continue;
+		rc = efx_probe_channel(channel);
+		if (rc)
+			goto rollback;
+		efx_init_napi_channel(efx->channel[i]);
+	}
+
+out:
+	/* Destroy unused channel structures */
+	for (i = 0; i < efx->n_channels; i++) {
+		channel = other_channel[i];
+		if (channel && channel->type->copy) {
+			efx_fini_napi_channel(channel);
+			efx_remove_channel(channel);
+			kfree(channel);
+		}
+	}
+
+	rc2 = efx_soft_enable_interrupts(efx);
+	if (rc2) {
+		rc = rc ? rc : rc2;
+		netif_err(efx, drv, efx->net_dev,
+			  "unable to restart interrupts on channel reallocation\n");
+		efx_schedule_reset(efx, RESET_TYPE_DISABLE);
+	} else {
+		efx_start_all(efx);
+		efx_device_attach_if_not_resetting(efx);
+	}
+	return rc;
+
+rollback:
+	/* Swap back */
+	efx->rxq_entries = old_rxq_entries;
+	efx->txq_entries = old_txq_entries;
+	for (i = 0; i < efx->n_channels; i++) {
+		channel = efx->channel[i];
+		efx->channel[i] = other_channel[i];
+		other_channel[i] = channel;
+	}
+	goto out;
+}
+
+int efx_set_channels(struct efx_nic *efx)
+{
+	struct efx_channel *channel;
+	struct efx_tx_queue *tx_queue;
+	int xdp_queue_number;
+
+	efx->tx_channel_offset =
+		efx_separate_tx_channels ?
+		efx->n_channels - efx->n_tx_channels : 0;
+
+	if (efx->xdp_tx_queue_count) {
+		EFX_WARN_ON_PARANOID(efx->xdp_tx_queues);
+
+		/* Allocate array for XDP TX queue lookup. */
+		efx->xdp_tx_queues = kcalloc(efx->xdp_tx_queue_count,
+					     sizeof(*efx->xdp_tx_queues),
+					     GFP_KERNEL);
+		if (!efx->xdp_tx_queues)
+			return -ENOMEM;
+	}
+
+	/* We need to mark which channels really have RX and TX
+	 * queues, and adjust the TX queue numbers if we have separate
+	 * RX-only and TX-only channels.
+	 */
+	xdp_queue_number = 0;
+	efx_for_each_channel(channel, efx) {
+		if (channel->channel < efx->n_rx_channels)
+			channel->rx_queue.core_index = channel->channel;
+		else
+			channel->rx_queue.core_index = -1;
+
+		efx_for_each_channel_tx_queue(tx_queue, channel) {
+			tx_queue->queue -= (efx->tx_channel_offset *
+					    EFX_TXQ_TYPES);
+
+			if (efx_channel_is_xdp_tx(channel) &&
+			    xdp_queue_number < efx->xdp_tx_queue_count) {
+				efx->xdp_tx_queues[xdp_queue_number] = tx_queue;
+				xdp_queue_number++;
+			}
+		}
+	}
+	return 0;
+}
+
+bool efx_default_channel_want_txqs(struct efx_channel *channel)
+{
+	return channel->channel - channel->efx->tx_channel_offset <
+		channel->efx->n_tx_channels;
+}
+
 /*************
  * START/STOP
  *************/
diff --git a/drivers/net/ethernet/sfc/efx_common.c b/drivers/net/ethernet/sfc/efx_common.c
index d0fc51a108bf..fe74c66c8ec6 100644
--- a/drivers/net/ethernet/sfc/efx_common.c
+++ b/drivers/net/ethernet/sfc/efx_common.c
@@ -884,17 +884,6 @@ fail:
 	return rc;
 }
 
-void efx_fini_channels(struct efx_nic *efx)
-{
-	unsigned int i;
-
-	for (i = 0; i < EFX_MAX_CHANNELS; i++)
-		if (efx->channel[i]) {
-			kfree(efx->channel[i]);
-			efx->channel[i] = NULL;
-		}
-}
-
 void efx_fini_struct(struct efx_nic *efx)
 {
 #ifdef CONFIG_RFS_ACCEL
diff --git a/drivers/net/ethernet/sfc/siena_sriov.c b/drivers/net/ethernet/sfc/siena_sriov.c
index dfbdf05dcf79..83dcfcae3d4b 100644
--- a/drivers/net/ethernet/sfc/siena_sriov.c
+++ b/drivers/net/ethernet/sfc/siena_sriov.c
@@ -7,6 +7,7 @@
 #include <linux/module.h>
 #include "net_driver.h"
 #include "efx.h"
+#include "efx_channels.h"
 #include "nic.h"
 #include "io.h"
 #include "mcdi.h"
-- 
cgit v1.2.3-59-g8ed1b


From 37c45a4e337d2ef8f327e66ee9839c56545bdb55 Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Wed, 8 Jan 2020 16:13:10 +0000
Subject: sfc: move channel interrupt management code

Small code styling fixes included.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/efx.c          | 316 -------------------------------
 drivers/net/ethernet/sfc/efx_channels.c | 322 ++++++++++++++++++++++++++++++++
 2 files changed, 322 insertions(+), 316 deletions(-)

diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index 955fcf714992..8ac299373ee5 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -96,17 +96,6 @@ static unsigned int rx_irq_mod_usec = 60;
  */
 static unsigned int tx_irq_mod_usec = 150;
 
-/* This is the requested number of CPUs to use for Receive-Side Scaling (RSS),
- * i.e. the number of CPUs among which we may distribute simultaneous
- * interrupt handling.
- *
- * Cards without MSI-X will only target one CPU via legacy or MSI interrupt.
- * The default (0) means to assign an interrupt to each core.
- */
-static unsigned int rss_cpus;
-module_param(rss_cpus, uint, 0444);
-MODULE_PARM_DESC(rss_cpus, "Number of CPUs to use for Receive-Side Scaling");
-
 static bool phy_flash_cfg;
 module_param(phy_flash_cfg, bool, 0644);
 MODULE_PARM_DESC(phy_flash_cfg, "Set PHYs into reflash mode initially");
@@ -450,311 +439,6 @@ void efx_set_default_rx_indir_table(struct efx_nic *efx,
 			ethtool_rxfh_indir_default(i, efx->rss_spread);
 }
 
-static unsigned int efx_wanted_parallelism(struct efx_nic *efx)
-{
-	cpumask_var_t thread_mask;
-	unsigned int count;
-	int cpu;
-
-	if (rss_cpus) {
-		count = rss_cpus;
-	} else {
-		if (unlikely(!zalloc_cpumask_var(&thread_mask, GFP_KERNEL))) {
-			netif_warn(efx, probe, efx->net_dev,
-				   "RSS disabled due to allocation failure\n");
-			return 1;
-		}
-
-		count = 0;
-		for_each_online_cpu(cpu) {
-			if (!cpumask_test_cpu(cpu, thread_mask)) {
-				++count;
-				cpumask_or(thread_mask, thread_mask,
-					   topology_sibling_cpumask(cpu));
-			}
-		}
-
-		free_cpumask_var(thread_mask);
-	}
-
-	if (count > EFX_MAX_RX_QUEUES) {
-		netif_cond_dbg(efx, probe, efx->net_dev, !rss_cpus, warn,
-			       "Reducing number of rx queues from %u to %u.\n",
-			       count, EFX_MAX_RX_QUEUES);
-		count = EFX_MAX_RX_QUEUES;
-	}
-
-	/* If RSS is requested for the PF *and* VFs then we can't write RSS
-	 * table entries that are inaccessible to VFs
-	 */
-#ifdef CONFIG_SFC_SRIOV
-	if (efx->type->sriov_wanted) {
-		if (efx->type->sriov_wanted(efx) && efx_vf_size(efx) > 1 &&
-		    count > efx_vf_size(efx)) {
-			netif_warn(efx, probe, efx->net_dev,
-				   "Reducing number of RSS channels from %u to %u for "
-				   "VF support. Increase vf-msix-limit to use more "
-				   "channels on the PF.\n",
-				   count, efx_vf_size(efx));
-			count = efx_vf_size(efx);
-		}
-	}
-#endif
-
-	return count;
-}
-
-static int efx_allocate_msix_channels(struct efx_nic *efx,
-				      unsigned int max_channels,
-				      unsigned int extra_channels,
-				      unsigned int parallelism)
-{
-	unsigned int n_channels = parallelism;
-	int vec_count;
-	int n_xdp_tx;
-	int n_xdp_ev;
-
-	if (efx_separate_tx_channels)
-		n_channels *= 2;
-	n_channels += extra_channels;
-
-	/* To allow XDP transmit to happen from arbitrary NAPI contexts
-	 * we allocate a TX queue per CPU. We share event queues across
-	 * multiple tx queues, assuming tx and ev queues are both
-	 * maximum size.
-	 */
-
-	n_xdp_tx = num_possible_cpus();
-	n_xdp_ev = DIV_ROUND_UP(n_xdp_tx, EFX_TXQ_TYPES);
-
-	vec_count = pci_msix_vec_count(efx->pci_dev);
-	if (vec_count < 0)
-		return vec_count;
-
-	max_channels = min_t(unsigned int, vec_count, max_channels);
-
-	/* Check resources.
-	 * We need a channel per event queue, plus a VI per tx queue.
-	 * This may be more pessimistic than it needs to be.
-	 */
-	if (n_channels + n_xdp_ev > max_channels) {
-		netif_err(efx, drv, efx->net_dev,
-			  "Insufficient resources for %d XDP event queues (%d other channels, max %d)\n",
-			  n_xdp_ev, n_channels, max_channels);
-		efx->n_xdp_channels = 0;
-		efx->xdp_tx_per_channel = 0;
-		efx->xdp_tx_queue_count = 0;
-	} else {
-		efx->n_xdp_channels = n_xdp_ev;
-		efx->xdp_tx_per_channel = EFX_TXQ_TYPES;
-		efx->xdp_tx_queue_count = n_xdp_tx;
-		n_channels += n_xdp_ev;
-		netif_dbg(efx, drv, efx->net_dev,
-			  "Allocating %d TX and %d event queues for XDP\n",
-			  n_xdp_tx, n_xdp_ev);
-	}
-
-	if (vec_count < n_channels) {
-		netif_err(efx, drv, efx->net_dev,
-			  "WARNING: Insufficient MSI-X vectors available (%d < %u).\n",
-			  vec_count, n_channels);
-		netif_err(efx, drv, efx->net_dev,
-			  "WARNING: Performance may be reduced.\n");
-		n_channels = vec_count;
-	}
-
-	n_channels = min(n_channels, max_channels);
-
-	efx->n_channels = n_channels;
-
-	/* Ignore XDP tx channels when creating rx channels. */
-	n_channels -= efx->n_xdp_channels;
-
-	if (efx_separate_tx_channels) {
-		efx->n_tx_channels =
-			min(max(n_channels / 2, 1U),
-			    efx->max_tx_channels);
-		efx->tx_channel_offset =
-			n_channels - efx->n_tx_channels;
-		efx->n_rx_channels =
-			max(n_channels -
-			    efx->n_tx_channels, 1U);
-	} else {
-		efx->n_tx_channels = min(n_channels, efx->max_tx_channels);
-		efx->tx_channel_offset = 0;
-		efx->n_rx_channels = n_channels;
-	}
-
-	efx->n_rx_channels = min(efx->n_rx_channels, parallelism);
-	efx->n_tx_channels = min(efx->n_tx_channels, parallelism);
-
-	efx->xdp_channel_offset = n_channels;
-
-	netif_dbg(efx, drv, efx->net_dev,
-		  "Allocating %u RX channels\n",
-		  efx->n_rx_channels);
-
-	return efx->n_channels;
-}
-
-/* Probe the number and type of interrupts we are able to obtain, and
- * the resulting numbers of channels and RX queues.
- */
-int efx_probe_interrupts(struct efx_nic *efx)
-{
-	unsigned int extra_channels = 0;
-	unsigned int rss_spread;
-	unsigned int i, j;
-	int rc;
-
-	for (i = 0; i < EFX_MAX_EXTRA_CHANNELS; i++)
-		if (efx->extra_channel_type[i])
-			++extra_channels;
-
-	if (efx->interrupt_mode == EFX_INT_MODE_MSIX) {
-		unsigned int parallelism = efx_wanted_parallelism(efx);
-		struct msix_entry xentries[EFX_MAX_CHANNELS];
-		unsigned int n_channels;
-
-		rc = efx_allocate_msix_channels(efx, efx->max_channels,
-						extra_channels, parallelism);
-		if (rc >= 0) {
-			n_channels = rc;
-			for (i = 0; i < n_channels; i++)
-				xentries[i].entry = i;
-			rc = pci_enable_msix_range(efx->pci_dev, xentries, 1,
-						   n_channels);
-		}
-		if (rc < 0) {
-			/* Fall back to single channel MSI */
-			netif_err(efx, drv, efx->net_dev,
-				  "could not enable MSI-X\n");
-			if (efx->type->min_interrupt_mode >= EFX_INT_MODE_MSI)
-				efx->interrupt_mode = EFX_INT_MODE_MSI;
-			else
-				return rc;
-		} else if (rc < n_channels) {
-			netif_err(efx, drv, efx->net_dev,
-				  "WARNING: Insufficient MSI-X vectors"
-				  " available (%d < %u).\n", rc, n_channels);
-			netif_err(efx, drv, efx->net_dev,
-				  "WARNING: Performance may be reduced.\n");
-			n_channels = rc;
-		}
-
-		if (rc > 0) {
-			for (i = 0; i < efx->n_channels; i++)
-				efx_get_channel(efx, i)->irq =
-					xentries[i].vector;
-		}
-	}
-
-	/* Try single interrupt MSI */
-	if (efx->interrupt_mode == EFX_INT_MODE_MSI) {
-		efx->n_channels = 1;
-		efx->n_rx_channels = 1;
-		efx->n_tx_channels = 1;
-		efx->n_xdp_channels = 0;
-		efx->xdp_channel_offset = efx->n_channels;
-		rc = pci_enable_msi(efx->pci_dev);
-		if (rc == 0) {
-			efx_get_channel(efx, 0)->irq = efx->pci_dev->irq;
-		} else {
-			netif_err(efx, drv, efx->net_dev,
-				  "could not enable MSI\n");
-			if (efx->type->min_interrupt_mode >= EFX_INT_MODE_LEGACY)
-				efx->interrupt_mode = EFX_INT_MODE_LEGACY;
-			else
-				return rc;
-		}
-	}
-
-	/* Assume legacy interrupts */
-	if (efx->interrupt_mode == EFX_INT_MODE_LEGACY) {
-		efx->n_channels = 1 + (efx_separate_tx_channels ? 1 : 0);
-		efx->n_rx_channels = 1;
-		efx->n_tx_channels = 1;
-		efx->n_xdp_channels = 0;
-		efx->xdp_channel_offset = efx->n_channels;
-		efx->legacy_irq = efx->pci_dev->irq;
-	}
-
-	/* Assign extra channels if possible, before XDP channels */
-	efx->n_extra_tx_channels = 0;
-	j = efx->xdp_channel_offset;
-	for (i = 0; i < EFX_MAX_EXTRA_CHANNELS; i++) {
-		if (!efx->extra_channel_type[i])
-			continue;
-		if (j <= efx->tx_channel_offset + efx->n_tx_channels) {
-			efx->extra_channel_type[i]->handle_no_channel(efx);
-		} else {
-			--j;
-			efx_get_channel(efx, j)->type =
-				efx->extra_channel_type[i];
-			if (efx_channel_has_tx_queues(efx_get_channel(efx, j)))
-				efx->n_extra_tx_channels++;
-		}
-	}
-
-	rss_spread = efx->n_rx_channels;
-	/* RSS might be usable on VFs even if it is disabled on the PF */
-#ifdef CONFIG_SFC_SRIOV
-	if (efx->type->sriov_wanted) {
-		efx->rss_spread = ((rss_spread > 1 ||
-				    !efx->type->sriov_wanted(efx)) ?
-				   rss_spread : efx_vf_size(efx));
-		return 0;
-	}
-#endif
-	efx->rss_spread = rss_spread;
-
-	return 0;
-}
-
-#if defined(CONFIG_SMP)
-void efx_set_interrupt_affinity(struct efx_nic *efx)
-{
-	struct efx_channel *channel;
-	unsigned int cpu;
-
-	efx_for_each_channel(channel, efx) {
-		cpu = cpumask_local_spread(channel->channel,
-					   pcibus_to_node(efx->pci_dev->bus));
-		irq_set_affinity_hint(channel->irq, cpumask_of(cpu));
-	}
-}
-
-void efx_clear_interrupt_affinity(struct efx_nic *efx)
-{
-	struct efx_channel *channel;
-
-	efx_for_each_channel(channel, efx)
-		irq_set_affinity_hint(channel->irq, NULL);
-}
-#else
-void efx_set_interrupt_affinity(struct efx_nic *efx __attribute__ ((unused)))
-{
-}
-
-void efx_clear_interrupt_affinity(struct efx_nic *efx __attribute__ ((unused)))
-{
-}
-#endif /* CONFIG_SMP */
-
-void efx_remove_interrupts(struct efx_nic *efx)
-{
-	struct efx_channel *channel;
-
-	/* Remove MSI/MSI-X interrupts */
-	efx_for_each_channel(channel, efx)
-		channel->irq = 0;
-	pci_disable_msi(efx->pci_dev);
-	pci_disable_msix(efx->pci_dev);
-
-	/* Remove legacy interrupt */
-	efx->legacy_irq = 0;
-}
-
 static int efx_probe_nic(struct efx_nic *efx)
 {
 	int rc;
diff --git a/drivers/net/ethernet/sfc/efx_channels.c b/drivers/net/ethernet/sfc/efx_channels.c
index bd5bc77a1d5a..65006af28210 100644
--- a/drivers/net/ethernet/sfc/efx_channels.c
+++ b/drivers/net/ethernet/sfc/efx_channels.c
@@ -28,6 +28,17 @@ module_param(interrupt_mode, uint, 0444);
 MODULE_PARM_DESC(interrupt_mode,
 		 "Interrupt mode (0=>MSIX 1=>MSI 2=>legacy)");
 
+/* This is the requested number of CPUs to use for Receive-Side Scaling (RSS),
+ * i.e. the number of CPUs among which we may distribute simultaneous
+ * interrupt handling.
+ *
+ * Cards without MSI-X will only target one CPU via legacy or MSI interrupt.
+ * The default (0) means to assign an interrupt to each core.
+ */
+static unsigned int rss_cpus;
+module_param(rss_cpus, uint, 0444);
+MODULE_PARM_DESC(rss_cpus, "Number of CPUs to use for Receive-Side Scaling");
+
 static unsigned int irq_adapt_low_thresh = 8000;
 module_param(irq_adapt_low_thresh, uint, 0644);
 MODULE_PARM_DESC(irq_adapt_low_thresh,
@@ -66,6 +77,317 @@ static const struct efx_channel_type efx_default_channel_type = {
 	.want_pio		= true,
 };
 
+/*************
+ * INTERRUPTS
+ *************/
+
+static unsigned int efx_wanted_parallelism(struct efx_nic *efx)
+{
+	cpumask_var_t thread_mask;
+	unsigned int count;
+	int cpu;
+
+	if (rss_cpus) {
+		count = rss_cpus;
+	} else {
+		if (unlikely(!zalloc_cpumask_var(&thread_mask, GFP_KERNEL))) {
+			netif_warn(efx, probe, efx->net_dev,
+				   "RSS disabled due to allocation failure\n");
+			return 1;
+		}
+
+		count = 0;
+		for_each_online_cpu(cpu) {
+			if (!cpumask_test_cpu(cpu, thread_mask)) {
+				++count;
+				cpumask_or(thread_mask, thread_mask,
+					   topology_sibling_cpumask(cpu));
+			}
+		}
+
+		free_cpumask_var(thread_mask);
+	}
+
+	if (count > EFX_MAX_RX_QUEUES) {
+		netif_cond_dbg(efx, probe, efx->net_dev, !rss_cpus, warn,
+			       "Reducing number of rx queues from %u to %u.\n",
+			       count, EFX_MAX_RX_QUEUES);
+		count = EFX_MAX_RX_QUEUES;
+	}
+
+	/* If RSS is requested for the PF *and* VFs then we can't write RSS
+	 * table entries that are inaccessible to VFs
+	 */
+#ifdef CONFIG_SFC_SRIOV
+	if (efx->type->sriov_wanted) {
+		if (efx->type->sriov_wanted(efx) && efx_vf_size(efx) > 1 &&
+		    count > efx_vf_size(efx)) {
+			netif_warn(efx, probe, efx->net_dev,
+				   "Reducing number of RSS channels from %u to %u for "
+				   "VF support. Increase vf-msix-limit to use more "
+				   "channels on the PF.\n",
+				   count, efx_vf_size(efx));
+			count = efx_vf_size(efx);
+		}
+	}
+#endif
+
+	return count;
+}
+
+static int efx_allocate_msix_channels(struct efx_nic *efx,
+				      unsigned int max_channels,
+				      unsigned int extra_channels,
+				      unsigned int parallelism)
+{
+	unsigned int n_channels = parallelism;
+	int vec_count;
+	int n_xdp_tx;
+	int n_xdp_ev;
+
+	if (efx_separate_tx_channels)
+		n_channels *= 2;
+	n_channels += extra_channels;
+
+	/* To allow XDP transmit to happen from arbitrary NAPI contexts
+	 * we allocate a TX queue per CPU. We share event queues across
+	 * multiple tx queues, assuming tx and ev queues are both
+	 * maximum size.
+	 */
+
+	n_xdp_tx = num_possible_cpus();
+	n_xdp_ev = DIV_ROUND_UP(n_xdp_tx, EFX_TXQ_TYPES);
+
+	vec_count = pci_msix_vec_count(efx->pci_dev);
+	if (vec_count < 0)
+		return vec_count;
+
+	max_channels = min_t(unsigned int, vec_count, max_channels);
+
+	/* Check resources.
+	 * We need a channel per event queue, plus a VI per tx queue.
+	 * This may be more pessimistic than it needs to be.
+	 */
+	if (n_channels + n_xdp_ev > max_channels) {
+		netif_err(efx, drv, efx->net_dev,
+			  "Insufficient resources for %d XDP event queues (%d other channels, max %d)\n",
+			  n_xdp_ev, n_channels, max_channels);
+		efx->n_xdp_channels = 0;
+		efx->xdp_tx_per_channel = 0;
+		efx->xdp_tx_queue_count = 0;
+	} else {
+		efx->n_xdp_channels = n_xdp_ev;
+		efx->xdp_tx_per_channel = EFX_TXQ_TYPES;
+		efx->xdp_tx_queue_count = n_xdp_tx;
+		n_channels += n_xdp_ev;
+		netif_dbg(efx, drv, efx->net_dev,
+			  "Allocating %d TX and %d event queues for XDP\n",
+			  n_xdp_tx, n_xdp_ev);
+	}
+
+	if (vec_count < n_channels) {
+		netif_err(efx, drv, efx->net_dev,
+			  "WARNING: Insufficient MSI-X vectors available (%d < %u).\n",
+			  vec_count, n_channels);
+		netif_err(efx, drv, efx->net_dev,
+			  "WARNING: Performance may be reduced.\n");
+		n_channels = vec_count;
+	}
+
+	n_channels = min(n_channels, max_channels);
+
+	efx->n_channels = n_channels;
+
+	/* Ignore XDP tx channels when creating rx channels. */
+	n_channels -= efx->n_xdp_channels;
+
+	if (efx_separate_tx_channels) {
+		efx->n_tx_channels =
+			min(max(n_channels / 2, 1U),
+			    efx->max_tx_channels);
+		efx->tx_channel_offset =
+			n_channels - efx->n_tx_channels;
+		efx->n_rx_channels =
+			max(n_channels -
+			    efx->n_tx_channels, 1U);
+	} else {
+		efx->n_tx_channels = min(n_channels, efx->max_tx_channels);
+		efx->tx_channel_offset = 0;
+		efx->n_rx_channels = n_channels;
+	}
+
+	efx->n_rx_channels = min(efx->n_rx_channels, parallelism);
+	efx->n_tx_channels = min(efx->n_tx_channels, parallelism);
+
+	efx->xdp_channel_offset = n_channels;
+
+	netif_dbg(efx, drv, efx->net_dev,
+		  "Allocating %u RX channels\n",
+		  efx->n_rx_channels);
+
+	return efx->n_channels;
+}
+
+/* Probe the number and type of interrupts we are able to obtain, and
+ * the resulting numbers of channels and RX queues.
+ */
+int efx_probe_interrupts(struct efx_nic *efx)
+{
+	unsigned int extra_channels = 0;
+	unsigned int rss_spread;
+	unsigned int i, j;
+	int rc;
+
+	for (i = 0; i < EFX_MAX_EXTRA_CHANNELS; i++)
+		if (efx->extra_channel_type[i])
+			++extra_channels;
+
+	if (efx->interrupt_mode == EFX_INT_MODE_MSIX) {
+		unsigned int parallelism = efx_wanted_parallelism(efx);
+		struct msix_entry xentries[EFX_MAX_CHANNELS];
+		unsigned int n_channels;
+
+		rc = efx_allocate_msix_channels(efx, efx->max_channels,
+						extra_channels, parallelism);
+		if (rc >= 0) {
+			n_channels = rc;
+			for (i = 0; i < n_channels; i++)
+				xentries[i].entry = i;
+			rc = pci_enable_msix_range(efx->pci_dev, xentries, 1,
+						   n_channels);
+		}
+		if (rc < 0) {
+			/* Fall back to single channel MSI */
+			netif_err(efx, drv, efx->net_dev,
+				  "could not enable MSI-X\n");
+			if (efx->type->min_interrupt_mode >= EFX_INT_MODE_MSI)
+				efx->interrupt_mode = EFX_INT_MODE_MSI;
+			else
+				return rc;
+		} else if (rc < n_channels) {
+			netif_err(efx, drv, efx->net_dev,
+				  "WARNING: Insufficient MSI-X vectors"
+				  " available (%d < %u).\n", rc, n_channels);
+			netif_err(efx, drv, efx->net_dev,
+				  "WARNING: Performance may be reduced.\n");
+			n_channels = rc;
+		}
+
+		if (rc > 0) {
+			for (i = 0; i < efx->n_channels; i++)
+				efx_get_channel(efx, i)->irq =
+					xentries[i].vector;
+		}
+	}
+
+	/* Try single interrupt MSI */
+	if (efx->interrupt_mode == EFX_INT_MODE_MSI) {
+		efx->n_channels = 1;
+		efx->n_rx_channels = 1;
+		efx->n_tx_channels = 1;
+		efx->n_xdp_channels = 0;
+		efx->xdp_channel_offset = efx->n_channels;
+		rc = pci_enable_msi(efx->pci_dev);
+		if (rc == 0) {
+			efx_get_channel(efx, 0)->irq = efx->pci_dev->irq;
+		} else {
+			netif_err(efx, drv, efx->net_dev,
+				  "could not enable MSI\n");
+			if (efx->type->min_interrupt_mode >= EFX_INT_MODE_LEGACY)
+				efx->interrupt_mode = EFX_INT_MODE_LEGACY;
+			else
+				return rc;
+		}
+	}
+
+	/* Assume legacy interrupts */
+	if (efx->interrupt_mode == EFX_INT_MODE_LEGACY) {
+		efx->n_channels = 1 + (efx_separate_tx_channels ? 1 : 0);
+		efx->n_rx_channels = 1;
+		efx->n_tx_channels = 1;
+		efx->n_xdp_channels = 0;
+		efx->xdp_channel_offset = efx->n_channels;
+		efx->legacy_irq = efx->pci_dev->irq;
+	}
+
+	/* Assign extra channels if possible, before XDP channels */
+	efx->n_extra_tx_channels = 0;
+	j = efx->xdp_channel_offset;
+	for (i = 0; i < EFX_MAX_EXTRA_CHANNELS; i++) {
+		if (!efx->extra_channel_type[i])
+			continue;
+		if (j <= efx->tx_channel_offset + efx->n_tx_channels) {
+			efx->extra_channel_type[i]->handle_no_channel(efx);
+		} else {
+			--j;
+			efx_get_channel(efx, j)->type =
+				efx->extra_channel_type[i];
+			if (efx_channel_has_tx_queues(efx_get_channel(efx, j)))
+				efx->n_extra_tx_channels++;
+		}
+	}
+
+	rss_spread = efx->n_rx_channels;
+	/* RSS might be usable on VFs even if it is disabled on the PF */
+#ifdef CONFIG_SFC_SRIOV
+	if (efx->type->sriov_wanted) {
+		efx->rss_spread = ((rss_spread > 1 ||
+				    !efx->type->sriov_wanted(efx)) ?
+				   rss_spread : efx_vf_size(efx));
+		return 0;
+	}
+#endif
+	efx->rss_spread = rss_spread;
+
+	return 0;
+}
+
+#if defined(CONFIG_SMP)
+void efx_set_interrupt_affinity(struct efx_nic *efx)
+{
+	struct efx_channel *channel;
+	unsigned int cpu;
+
+	efx_for_each_channel(channel, efx) {
+		cpu = cpumask_local_spread(channel->channel,
+					   pcibus_to_node(efx->pci_dev->bus));
+		irq_set_affinity_hint(channel->irq, cpumask_of(cpu));
+	}
+}
+
+void efx_clear_interrupt_affinity(struct efx_nic *efx)
+{
+	struct efx_channel *channel;
+
+	efx_for_each_channel(channel, efx)
+		irq_set_affinity_hint(channel->irq, NULL);
+}
+#else
+void
+efx_set_interrupt_affinity(struct efx_nic *efx __attribute__ ((unused)))
+{
+}
+
+void
+efx_clear_interrupt_affinity(struct efx_nic *efx __attribute__ ((unused)))
+{
+}
+#endif /* CONFIG_SMP */
+
+void efx_remove_interrupts(struct efx_nic *efx)
+{
+	struct efx_channel *channel;
+
+	/* Remove MSI/MSI-X interrupts */
+	efx_for_each_channel(channel, efx)
+		channel->irq = 0;
+	pci_disable_msi(efx->pci_dev);
+	pci_disable_msix(efx->pci_dev);
+
+	/* Remove legacy interrupt */
+	efx->legacy_irq = 0;
+}
+
 /**************************************************************************
  *
  * Channel handling
-- 
cgit v1.2.3-59-g8ed1b


From 5f99925632b482a80998779c703a3bf89829bb5d Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Wed, 8 Jan 2020 16:13:28 +0000
Subject: sfc: move event queue management code

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/efx.c          | 92 ---------------------------------
 drivers/net/ethernet/sfc/efx_channels.c | 91 ++++++++++++++++++++++++++++++++
 2 files changed, 91 insertions(+), 92 deletions(-)

diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index 8ac299373ee5..4bce5c739974 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -128,98 +128,6 @@ static int efx_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **xdpfs,
 			ASSERT_RTNL();			\
 	} while (0)
 
-/**************************************************************************
- *
- * Event queue processing
- *
- *************************************************************************/
-
-/* Create event queue
- * Event queue memory allocations are done only once.  If the channel
- * is reset, the memory buffer will be reused; this guards against
- * errors during channel reset and also simplifies interrupt handling.
- */
-int efx_probe_eventq(struct efx_channel *channel)
-{
-	struct efx_nic *efx = channel->efx;
-	unsigned long entries;
-
-	netif_dbg(efx, probe, efx->net_dev,
-		  "chan %d create event queue\n", channel->channel);
-
-	/* Build an event queue with room for one event per tx and rx buffer,
-	 * plus some extra for link state events and MCDI completions. */
-	entries = roundup_pow_of_two(efx->rxq_entries + efx->txq_entries + 128);
-	EFX_WARN_ON_PARANOID(entries > EFX_MAX_EVQ_SIZE);
-	channel->eventq_mask = max(entries, EFX_MIN_EVQ_SIZE) - 1;
-
-	return efx_nic_probe_eventq(channel);
-}
-
-/* Prepare channel's event queue */
-int efx_init_eventq(struct efx_channel *channel)
-{
-	struct efx_nic *efx = channel->efx;
-	int rc;
-
-	EFX_WARN_ON_PARANOID(channel->eventq_init);
-
-	netif_dbg(efx, drv, efx->net_dev,
-		  "chan %d init event queue\n", channel->channel);
-
-	rc = efx_nic_init_eventq(channel);
-	if (rc == 0) {
-		efx->type->push_irq_moderation(channel);
-		channel->eventq_read_ptr = 0;
-		channel->eventq_init = true;
-	}
-	return rc;
-}
-
-/* Enable event queue processing and NAPI */
-void efx_start_eventq(struct efx_channel *channel)
-{
-	netif_dbg(channel->efx, ifup, channel->efx->net_dev,
-		  "chan %d start event queue\n", channel->channel);
-
-	/* Make sure the NAPI handler sees the enabled flag set */
-	channel->enabled = true;
-	smp_wmb();
-
-	napi_enable(&channel->napi_str);
-	efx_nic_eventq_read_ack(channel);
-}
-
-/* Disable event queue processing and NAPI */
-void efx_stop_eventq(struct efx_channel *channel)
-{
-	if (!channel->enabled)
-		return;
-
-	napi_disable(&channel->napi_str);
-	channel->enabled = false;
-}
-
-void efx_fini_eventq(struct efx_channel *channel)
-{
-	if (!channel->eventq_init)
-		return;
-
-	netif_dbg(channel->efx, drv, channel->efx->net_dev,
-		  "chan %d fini event queue\n", channel->channel);
-
-	efx_nic_fini_eventq(channel);
-	channel->eventq_init = false;
-}
-
-void efx_remove_eventq(struct efx_channel *channel)
-{
-	netif_dbg(channel->efx, drv, channel->efx->net_dev,
-		  "chan %d remove event queue\n", channel->channel);
-
-	efx_nic_remove_eventq(channel);
-}
-
 /**************************************************************************
  *
  * Channel handling
diff --git a/drivers/net/ethernet/sfc/efx_channels.c b/drivers/net/ethernet/sfc/efx_channels.c
index 65006af28210..21bd71daf5a0 100644
--- a/drivers/net/ethernet/sfc/efx_channels.c
+++ b/drivers/net/ethernet/sfc/efx_channels.c
@@ -388,6 +388,97 @@ void efx_remove_interrupts(struct efx_nic *efx)
 	efx->legacy_irq = 0;
 }
 
+/***************
+ * EVENT QUEUES
+ ***************/
+
+/* Create event queue
+ * Event queue memory allocations are done only once.  If the channel
+ * is reset, the memory buffer will be reused; this guards against
+ * errors during channel reset and also simplifies interrupt handling.
+ */
+int efx_probe_eventq(struct efx_channel *channel)
+{
+	struct efx_nic *efx = channel->efx;
+	unsigned long entries;
+
+	netif_dbg(efx, probe, efx->net_dev,
+		  "chan %d create event queue\n", channel->channel);
+
+	/* Build an event queue with room for one event per tx and rx buffer,
+	 * plus some extra for link state events and MCDI completions.
+	 */
+	entries = roundup_pow_of_two(efx->rxq_entries + efx->txq_entries + 128);
+	EFX_WARN_ON_PARANOID(entries > EFX_MAX_EVQ_SIZE);
+	channel->eventq_mask = max(entries, EFX_MIN_EVQ_SIZE) - 1;
+
+	return efx_nic_probe_eventq(channel);
+}
+
+/* Prepare channel's event queue */
+int efx_init_eventq(struct efx_channel *channel)
+{
+	struct efx_nic *efx = channel->efx;
+	int rc;
+
+	EFX_WARN_ON_PARANOID(channel->eventq_init);
+
+	netif_dbg(efx, drv, efx->net_dev,
+		  "chan %d init event queue\n", channel->channel);
+
+	rc = efx_nic_init_eventq(channel);
+	if (rc == 0) {
+		efx->type->push_irq_moderation(channel);
+		channel->eventq_read_ptr = 0;
+		channel->eventq_init = true;
+	}
+	return rc;
+}
+
+/* Enable event queue processing and NAPI */
+void efx_start_eventq(struct efx_channel *channel)
+{
+	netif_dbg(channel->efx, ifup, channel->efx->net_dev,
+		  "chan %d start event queue\n", channel->channel);
+
+	/* Make sure the NAPI handler sees the enabled flag set */
+	channel->enabled = true;
+	smp_wmb();
+
+	napi_enable(&channel->napi_str);
+	efx_nic_eventq_read_ack(channel);
+}
+
+/* Disable event queue processing and NAPI */
+void efx_stop_eventq(struct efx_channel *channel)
+{
+	if (!channel->enabled)
+		return;
+
+	napi_disable(&channel->napi_str);
+	channel->enabled = false;
+}
+
+void efx_fini_eventq(struct efx_channel *channel)
+{
+	if (!channel->eventq_init)
+		return;
+
+	netif_dbg(channel->efx, drv, channel->efx->net_dev,
+		  "chan %d fini event queue\n", channel->channel);
+
+	efx_nic_fini_eventq(channel);
+	channel->eventq_init = false;
+}
+
+void efx_remove_eventq(struct efx_channel *channel)
+{
+	netif_dbg(channel->efx, drv, channel->efx->net_dev,
+		  "chan %d remove event queue\n", channel->channel);
+
+	efx_nic_remove_eventq(channel);
+}
+
 /**************************************************************************
  *
  * Channel handling
-- 
cgit v1.2.3-59-g8ed1b


From 1751cc365f58ea69714be1c8ba718236354cd719 Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Wed, 8 Jan 2020 16:13:43 +0000
Subject: sfc: move common rx code

The moved code deals with managing rx buffers and queues.
A tiny bit of refactoring was required in other files to stitch the
code together.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/Makefile    |   2 +-
 drivers/net/ethernet/sfc/ef10.c      |   1 +
 drivers/net/ethernet/sfc/efx.c       |  11 -
 drivers/net/ethernet/sfc/efx.h       |   3 +
 drivers/net/ethernet/sfc/rx.c        | 366 +---------------------------------
 drivers/net/ethernet/sfc/rx_common.c | 375 +++++++++++++++++++++++++++++++++++
 6 files changed, 383 insertions(+), 375 deletions(-)
 create mode 100644 drivers/net/ethernet/sfc/rx_common.c

diff --git a/drivers/net/ethernet/sfc/Makefile b/drivers/net/ethernet/sfc/Makefile
index dacfa42beffe..03c63317a869 100644
--- a/drivers/net/ethernet/sfc/Makefile
+++ b/drivers/net/ethernet/sfc/Makefile
@@ -1,7 +1,7 @@
 # SPDX-License-Identifier: GPL-2.0
 sfc-y			+= efx.o efx_common.o efx_channels.o nic.o \
 			   farch.o siena.o ef10.o \
-			   tx.o rx.o \
+			   tx.o rx.o rx_common.o \
 			   selftest.o ethtool.o ptp.o tx_tso.o \
 			   mcdi.o mcdi_port.o \
 			   mcdi_mon.o
diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c
index de6e6cd4469b..d752ed34672d 100644
--- a/drivers/net/ethernet/sfc/ef10.c
+++ b/drivers/net/ethernet/sfc/ef10.c
@@ -5,6 +5,7 @@
  */
 
 #include "net_driver.h"
+#include "rx_common.h"
 #include "ef10_regs.h"
 #include "io.h"
 #include "mcdi.h"
diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index 4bce5c739974..655424e83d97 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -128,17 +128,6 @@ static int efx_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **xdpfs,
 			ASSERT_RTNL();			\
 	} while (0)
 
-/**************************************************************************
- *
- * Channel handling
- *
- *************************************************************************/
-
-void efx_schedule_slow_fill(struct efx_rx_queue *rx_queue)
-{
-	mod_timer(&rx_queue->slow_fill, jiffies + msecs_to_jiffies(10));
-}
-
 /**************************************************************************
  *
  * Port handling
diff --git a/drivers/net/ethernet/sfc/efx.h b/drivers/net/ethernet/sfc/efx.h
index 2b417e779e82..3920f29b2fed 100644
--- a/drivers/net/ethernet/sfc/efx.h
+++ b/drivers/net/ethernet/sfc/efx.h
@@ -37,6 +37,9 @@ static inline void efx_rx_flush_packet(struct efx_channel *channel)
 		__efx_rx_packet(channel);
 }
 
+void efx_init_rx_recycle_ring(struct efx_rx_queue *rx_queue);
+struct page *efx_reuse_page(struct efx_rx_queue *rx_queue);
+
 #define EFX_MAX_DMAQ_SIZE 4096UL
 #define EFX_DEFAULT_DMAQ_SIZE 1024UL
 #define EFX_MIN_DMAQ_SIZE 512UL
diff --git a/drivers/net/ethernet/sfc/rx.c b/drivers/net/ethernet/sfc/rx.c
index 26b5c36237fe..0e04ed7f6382 100644
--- a/drivers/net/ethernet/sfc/rx.c
+++ b/drivers/net/ethernet/sfc/rx.c
@@ -43,23 +43,10 @@
 /* Size of buffer allocated for skb header area. */
 #define EFX_SKB_HEADERS  128u
 
-/* This is the percentage fill level below which new RX descriptors
- * will be added to the RX descriptor ring.
- */
-static unsigned int rx_refill_threshold;
-
 /* Each packet can consume up to ceil(max_frame_len / buffer_size) buffers */
 #define EFX_RX_MAX_FRAGS DIV_ROUND_UP(EFX_MAX_FRAME_LEN(EFX_MAX_MTU), \
 				      EFX_RX_USR_BUF_SIZE)
 
-/*
- * RX maximum head room required.
- *
- * This must be at least 1 to prevent overflow, plus one packet-worth
- * to allow pipelined receives.
- */
-#define EFX_RXD_HEAD_ROOM (1 + EFX_RX_MAX_FRAGS)
-
 static inline u8 *efx_rx_buf_va(struct efx_rx_buffer *buf)
 {
 	return page_address(buf->page) + buf->page_offset;
@@ -86,22 +73,8 @@ static inline void efx_sync_rx_buffer(struct efx_nic *efx,
 				DMA_FROM_DEVICE);
 }
 
-void efx_rx_config_page_split(struct efx_nic *efx)
-{
-	efx->rx_page_buf_step = ALIGN(efx->rx_dma_len + efx->rx_ip_align +
-				      XDP_PACKET_HEADROOM,
-				      EFX_RX_BUF_ALIGNMENT);
-	efx->rx_bufs_per_page = efx->rx_buffer_order ? 1 :
-		((PAGE_SIZE - sizeof(struct efx_rx_page_state)) /
-		efx->rx_page_buf_step);
-	efx->rx_buffer_truesize = (PAGE_SIZE << efx->rx_buffer_order) /
-		efx->rx_bufs_per_page;
-	efx->rx_pages_per_batch = DIV_ROUND_UP(EFX_RX_PREFERRED_BATCH,
-					       efx->rx_bufs_per_page);
-}
-
 /* Check the RX page recycle ring for a page that can be reused. */
-static struct page *efx_reuse_page(struct efx_rx_queue *rx_queue)
+struct page *efx_reuse_page(struct efx_rx_queue *rx_queue)
 {
 	struct efx_nic *efx = rx_queue->efx;
 	struct page *page;
@@ -134,106 +107,6 @@ static struct page *efx_reuse_page(struct efx_rx_queue *rx_queue)
 	return NULL;
 }
 
-/**
- * efx_init_rx_buffers - create EFX_RX_BATCH page-based RX buffers
- *
- * @rx_queue:		Efx RX queue
- *
- * This allocates a batch of pages, maps them for DMA, and populates
- * struct efx_rx_buffers for each one. Return a negative error code or
- * 0 on success. If a single page can be used for multiple buffers,
- * then the page will either be inserted fully, or not at all.
- */
-int efx_init_rx_buffers(struct efx_rx_queue *rx_queue, bool atomic)
-{
-	struct efx_nic *efx = rx_queue->efx;
-	struct efx_rx_buffer *rx_buf;
-	struct page *page;
-	unsigned int page_offset;
-	struct efx_rx_page_state *state;
-	dma_addr_t dma_addr;
-	unsigned index, count;
-
-	count = 0;
-	do {
-		page = efx_reuse_page(rx_queue);
-		if (page == NULL) {
-			page = alloc_pages(__GFP_COMP |
-					   (atomic ? GFP_ATOMIC : GFP_KERNEL),
-					   efx->rx_buffer_order);
-			if (unlikely(page == NULL))
-				return -ENOMEM;
-			dma_addr =
-				dma_map_page(&efx->pci_dev->dev, page, 0,
-					     PAGE_SIZE << efx->rx_buffer_order,
-					     DMA_FROM_DEVICE);
-			if (unlikely(dma_mapping_error(&efx->pci_dev->dev,
-						       dma_addr))) {
-				__free_pages(page, efx->rx_buffer_order);
-				return -EIO;
-			}
-			state = page_address(page);
-			state->dma_addr = dma_addr;
-		} else {
-			state = page_address(page);
-			dma_addr = state->dma_addr;
-		}
-
-		dma_addr += sizeof(struct efx_rx_page_state);
-		page_offset = sizeof(struct efx_rx_page_state);
-
-		do {
-			index = rx_queue->added_count & rx_queue->ptr_mask;
-			rx_buf = efx_rx_buffer(rx_queue, index);
-			rx_buf->dma_addr = dma_addr + efx->rx_ip_align +
-					   XDP_PACKET_HEADROOM;
-			rx_buf->page = page;
-			rx_buf->page_offset = page_offset + efx->rx_ip_align +
-					      XDP_PACKET_HEADROOM;
-			rx_buf->len = efx->rx_dma_len;
-			rx_buf->flags = 0;
-			++rx_queue->added_count;
-			get_page(page);
-			dma_addr += efx->rx_page_buf_step;
-			page_offset += efx->rx_page_buf_step;
-		} while (page_offset + efx->rx_page_buf_step <= PAGE_SIZE);
-
-		rx_buf->flags = EFX_RX_BUF_LAST_IN_PAGE;
-	} while (++count < efx->rx_pages_per_batch);
-
-	return 0;
-}
-
-/* Unmap a DMA-mapped page.  This function is only called for the final RX
- * buffer in a page.
- */
-void efx_unmap_rx_buffer(struct efx_nic *efx,
-			 struct efx_rx_buffer *rx_buf)
-{
-	struct page *page = rx_buf->page;
-
-	if (page) {
-		struct efx_rx_page_state *state = page_address(page);
-		dma_unmap_page(&efx->pci_dev->dev,
-			       state->dma_addr,
-			       PAGE_SIZE << efx->rx_buffer_order,
-			       DMA_FROM_DEVICE);
-	}
-}
-
-void efx_free_rx_buffers(struct efx_rx_queue *rx_queue,
-			 struct efx_rx_buffer *rx_buf,
-			 unsigned int num_bufs)
-{
-	do {
-		if (rx_buf->page) {
-			put_page(rx_buf->page);
-			rx_buf->page = NULL;
-		}
-		rx_buf = efx_rx_buf_next(rx_queue, rx_buf);
-	} while (--num_bufs);
-}
-
 /* Attempt to recycle the page if there is an RX recycle ring; the page can
  * only be added if this is the final RX buffer, to prevent pages being used in
  * the descriptor ring and appearing in the recycle ring simultaneously.
@@ -270,21 +143,6 @@ static void efx_recycle_rx_page(struct efx_channel *channel,
 	put_page(rx_buf->page);
 }
 
-static void efx_fini_rx_buffer(struct efx_rx_queue *rx_queue,
-			       struct efx_rx_buffer *rx_buf)
-{
-	/* Release the page reference we hold for the buffer. */
-	if (rx_buf->page)
-		put_page(rx_buf->page);
-
-	/* If this is the last buffer in a page, unmap and free it. */
-	if (rx_buf->flags & EFX_RX_BUF_LAST_IN_PAGE) {
-		efx_unmap_rx_buffer(rx_queue->efx, rx_buf);
-		efx_free_rx_buffers(rx_queue, rx_buf, 1);
-	}
-	rx_buf->page = NULL;
-}
-
 /* Recycle the pages that are used by buffers that have just been received. */
 static void efx_recycle_rx_pages(struct efx_channel *channel,
 				 struct efx_rx_buffer *rx_buf,
@@ -309,78 +167,6 @@ static void efx_discard_rx_packet(struct efx_channel *channel,
 	efx_free_rx_buffers(rx_queue, rx_buf, n_frags);
 }
 
-/**
- * efx_fast_push_rx_descriptors - push new RX descriptors quickly
- * @rx_queue:		RX descriptor queue
- *
- * This will aim to fill the RX descriptor queue up to
- * @rx_queue->@max_fill. If there is insufficient atomic
- * memory to do so, a slow fill will be scheduled.
- *
- * The caller must provide serialisation (none is used here). In practise,
- * this means this function must run from the NAPI handler, or be called
- * when NAPI is disabled.
- */
-void efx_fast_push_rx_descriptors(struct efx_rx_queue *rx_queue, bool atomic)
-{
-	struct efx_nic *efx = rx_queue->efx;
-	unsigned int fill_level, batch_size;
-	int space, rc = 0;
-
-	if (!rx_queue->refill_enabled)
-		return;
-
-	/* Calculate current fill level, and exit if we don't need to fill */
-	fill_level = (rx_queue->added_count - rx_queue->removed_count);
-	EFX_WARN_ON_ONCE_PARANOID(fill_level > rx_queue->efx->rxq_entries);
-	if (fill_level >= rx_queue->fast_fill_trigger)
-		goto out;
-
-	/* Record minimum fill level */
-	if (unlikely(fill_level < rx_queue->min_fill)) {
-		if (fill_level)
-			rx_queue->min_fill = fill_level;
-	}
-
-	batch_size = efx->rx_pages_per_batch * efx->rx_bufs_per_page;
-	space = rx_queue->max_fill - fill_level;
-	EFX_WARN_ON_ONCE_PARANOID(space < batch_size);
-
-	netif_vdbg(rx_queue->efx, rx_status, rx_queue->efx->net_dev,
-		   "RX queue %d fast-filling descriptor ring from"
-		   " level %d to level %d\n",
-		   efx_rx_queue_index(rx_queue), fill_level,
-		   rx_queue->max_fill);
-
-
-	do {
-		rc = efx_init_rx_buffers(rx_queue, atomic);
-		if (unlikely(rc)) {
-			/* Ensure that we don't leave the rx queue empty */
-			efx_schedule_slow_fill(rx_queue);
-			goto out;
-		}
-	} while ((space -= batch_size) >= batch_size);
-
-	netif_vdbg(rx_queue->efx, rx_status, rx_queue->efx->net_dev,
-		   "RX queue %d fast-filled descriptor ring "
-		   "to level %d\n", efx_rx_queue_index(rx_queue),
-		   rx_queue->added_count - rx_queue->removed_count);
-
- out:
-	if (rx_queue->notified_count != rx_queue->added_count)
-		efx_nic_notify_rx_desc(rx_queue);
-}
-
-void efx_rx_slow_fill(struct timer_list *t)
-{
-	struct efx_rx_queue *rx_queue = from_timer(rx_queue, t, slow_fill);
-
-	/* Post an event to cause NAPI to run and refill the queue */
-	efx_nic_generate_fill_event(rx_queue);
-	++rx_queue->slow_fill_count;
-}
-
 static void efx_rx_packet__check_len(struct efx_rx_queue *rx_queue,
 				     struct efx_rx_buffer *rx_buf,
 				     int len)
@@ -797,41 +583,10 @@ out:
 	channel->rx_pkt_n_frags = 0;
 }
 
-int efx_probe_rx_queue(struct efx_rx_queue *rx_queue)
-{
-	struct efx_nic *efx = rx_queue->efx;
-	unsigned int entries;
-	int rc;
-
-	/* Create the smallest power-of-two aligned ring */
-	entries = max(roundup_pow_of_two(efx->rxq_entries), EFX_MIN_DMAQ_SIZE);
-	EFX_WARN_ON_PARANOID(entries > EFX_MAX_DMAQ_SIZE);
-	rx_queue->ptr_mask = entries - 1;
-
-	netif_dbg(efx, probe, efx->net_dev,
-		  "creating RX queue %d size %#x mask %#x\n",
-		  efx_rx_queue_index(rx_queue), efx->rxq_entries,
-		  rx_queue->ptr_mask);
-
-	/* Allocate RX buffers */
-	rx_queue->buffer = kcalloc(entries, sizeof(*rx_queue->buffer),
-				   GFP_KERNEL);
-	if (!rx_queue->buffer)
-		return -ENOMEM;
-
-	rc = efx_nic_probe_rx(rx_queue);
-	if (rc) {
-		kfree(rx_queue->buffer);
-		rx_queue->buffer = NULL;
-	}
-
-	return rc;
-}
-
-static void efx_init_rx_recycle_ring(struct efx_nic *efx,
-				     struct efx_rx_queue *rx_queue)
+void efx_init_rx_recycle_ring(struct efx_rx_queue *rx_queue)
 {
 	unsigned int bufs_in_recycle_ring, page_ring_size;
+	struct efx_nic *efx = rx_queue->efx;
 
 	/* Set the RX recycle ring size */
 #ifdef CONFIG_PPC64
@@ -850,121 +605,6 @@ static void efx_init_rx_recycle_ring(struct efx_nic *efx,
 	rx_queue->page_ptr_mask = page_ring_size - 1;
 }
 
-void efx_init_rx_queue(struct efx_rx_queue *rx_queue)
-{
-	struct efx_nic *efx = rx_queue->efx;
-	unsigned int max_fill, trigger, max_trigger;
-	int rc = 0;
-
-	netif_dbg(rx_queue->efx, drv, rx_queue->efx->net_dev,
-		  "initialising RX queue %d\n", efx_rx_queue_index(rx_queue));
-
-	/* Initialise ptr fields */
-	rx_queue->added_count = 0;
-	rx_queue->notified_count = 0;
-	rx_queue->removed_count = 0;
-	rx_queue->min_fill = -1U;
-	efx_init_rx_recycle_ring(efx, rx_queue);
-
-	rx_queue->page_remove = 0;
-	rx_queue->page_add = rx_queue->page_ptr_mask + 1;
-	rx_queue->page_recycle_count = 0;
-	rx_queue->page_recycle_failed = 0;
-	rx_queue->page_recycle_full = 0;
-
-	/* Initialise limit fields */
-	max_fill = efx->rxq_entries - EFX_RXD_HEAD_ROOM;
-	max_trigger =
-		max_fill - efx->rx_pages_per_batch * efx->rx_bufs_per_page;
-	if (rx_refill_threshold != 0) {
-		trigger = max_fill * min(rx_refill_threshold, 100U) / 100U;
-		if (trigger > max_trigger)
-			trigger = max_trigger;
-	} else {
-		trigger = max_trigger;
-	}
-
-	rx_queue->max_fill = max_fill;
-	rx_queue->fast_fill_trigger = trigger;
-	rx_queue->refill_enabled = true;
-
-	/* Initialise XDP queue information */
-	rc = xdp_rxq_info_reg(&rx_queue->xdp_rxq_info, efx->net_dev,
-			      rx_queue->core_index);
-
-	if (rc) {
-		netif_err(efx, rx_err, efx->net_dev,
-			  "Failure to initialise XDP queue information rc=%d\n",
-			  rc);
-		efx->xdp_rxq_info_failed = true;
-	} else {
-		rx_queue->xdp_rxq_info_valid = true;
-	}
-
-	/* Set up RX descriptor ring */
-	efx_nic_init_rx(rx_queue);
-}
-
-void efx_fini_rx_queue(struct efx_rx_queue *rx_queue)
-{
-	int i;
-	struct efx_nic *efx = rx_queue->efx;
-	struct efx_rx_buffer *rx_buf;
-
-	netif_dbg(rx_queue->efx, drv, rx_queue->efx->net_dev,
-		  "shutting down RX queue %d\n", efx_rx_queue_index(rx_queue));
-
-	del_timer_sync(&rx_queue->slow_fill);
-
-	/* Release RX buffers from the current read ptr to the write ptr */
-	if (rx_queue->buffer) {
-		for (i = rx_queue->removed_count; i < rx_queue->added_count;
-		     i++) {
-			unsigned index = i & rx_queue->ptr_mask;
-			rx_buf = efx_rx_buffer(rx_queue, index);
-			efx_fini_rx_buffer(rx_queue, rx_buf);
-		}
-	}
-
-	/* Unmap and release the pages in the recycle ring. Remove the ring. */
-	for (i = 0; i <= rx_queue->page_ptr_mask; i++) {
-		struct page *page = rx_queue->page_ring[i];
-		struct efx_rx_page_state *state;
-
-		if (page == NULL)
-			continue;
-
-		state = page_address(page);
-		dma_unmap_page(&efx->pci_dev->dev, state->dma_addr,
-			       PAGE_SIZE << efx->rx_buffer_order,
-			       DMA_FROM_DEVICE);
-		put_page(page);
-	}
-	kfree(rx_queue->page_ring);
-	rx_queue->page_ring = NULL;
-
-	if (rx_queue->xdp_rxq_info_valid)
-		xdp_rxq_info_unreg(&rx_queue->xdp_rxq_info);
-
-	rx_queue->xdp_rxq_info_valid = false;
-}
-
-void efx_remove_rx_queue(struct efx_rx_queue *rx_queue)
-{
-	netif_dbg(rx_queue->efx, drv, rx_queue->efx->net_dev,
-		  "destroying RX queue %d\n", efx_rx_queue_index(rx_queue));
-
-	efx_nic_remove_rx(rx_queue);
-
-	kfree(rx_queue->buffer);
-	rx_queue->buffer = NULL;
-}
-
-
-module_param(rx_refill_threshold, uint, 0444);
-MODULE_PARM_DESC(rx_refill_threshold,
-		 "RX descriptor ring refill threshold (%)");
-
 #ifdef CONFIG_RFS_ACCEL
 
 static void efx_filter_rfs_work(struct work_struct *data)
diff --git a/drivers/net/ethernet/sfc/rx_common.c b/drivers/net/ethernet/sfc/rx_common.c
new file mode 100644
index 000000000000..f4b5c3d828f6
--- /dev/null
+++ b/drivers/net/ethernet/sfc/rx_common.c
@@ -0,0 +1,375 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/****************************************************************************
+ * Driver for Solarflare network controllers and boards
+ * Copyright 2018 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#include "net_driver.h"
+#include <linux/module.h>
+#include "efx.h"
+#include "nic.h"
+#include "rx_common.h"
+
+/* This is the percentage fill level below which new RX descriptors
+ * will be added to the RX descriptor ring.
+ */
+static unsigned int rx_refill_threshold;
+module_param(rx_refill_threshold, uint, 0444);
+MODULE_PARM_DESC(rx_refill_threshold,
+		 "RX descriptor ring refill threshold (%)");
+
+/* RX maximum head room required.
+ *
+ * This must be at least 1 to prevent overflow, plus one packet-worth
+ * to allow pipelined receives.
+ */
+#define EFX_RXD_HEAD_ROOM (1 + EFX_RX_MAX_FRAGS)
+
+static void efx_fini_rx_buffer(struct efx_rx_queue *rx_queue,
+			       struct efx_rx_buffer *rx_buf)
+{
+	/* Release the page reference we hold for the buffer. */
+	if (rx_buf->page)
+		put_page(rx_buf->page);
+
+	/* If this is the last buffer in a page, unmap and free it. */
+	if (rx_buf->flags & EFX_RX_BUF_LAST_IN_PAGE) {
+		efx_unmap_rx_buffer(rx_queue->efx, rx_buf);
+		efx_free_rx_buffers(rx_queue, rx_buf, 1);
+	}
+	rx_buf->page = NULL;
+}
+
+int efx_probe_rx_queue(struct efx_rx_queue *rx_queue)
+{
+	struct efx_nic *efx = rx_queue->efx;
+	unsigned int entries;
+	int rc;
+
+	/* Create the smallest power-of-two aligned ring */
+	entries = max(roundup_pow_of_two(efx->rxq_entries), EFX_MIN_DMAQ_SIZE);
+	EFX_WARN_ON_PARANOID(entries > EFX_MAX_DMAQ_SIZE);
+	rx_queue->ptr_mask = entries - 1;
+
+	netif_dbg(efx, probe, efx->net_dev,
+		  "creating RX queue %d size %#x mask %#x\n",
+		  efx_rx_queue_index(rx_queue), efx->rxq_entries,
+		  rx_queue->ptr_mask);
+
+	/* Allocate RX buffers */
+	rx_queue->buffer = kcalloc(entries, sizeof(*rx_queue->buffer),
+				   GFP_KERNEL);
+	if (!rx_queue->buffer)
+		return -ENOMEM;
+
+	rc = efx_nic_probe_rx(rx_queue);
+	if (rc) {
+		kfree(rx_queue->buffer);
+		rx_queue->buffer = NULL;
+	}
+
+	return rc;
+}
+
+void efx_init_rx_queue(struct efx_rx_queue *rx_queue)
+{
+	unsigned int max_fill, trigger, max_trigger;
+	struct efx_nic *efx = rx_queue->efx;
+	int rc = 0;
+
+	netif_dbg(rx_queue->efx, drv, rx_queue->efx->net_dev,
+		  "initialising RX queue %d\n", efx_rx_queue_index(rx_queue));
+
+	/* Initialise ptr fields */
+	rx_queue->added_count = 0;
+	rx_queue->notified_count = 0;
+	rx_queue->removed_count = 0;
+	rx_queue->min_fill = -1U;
+	efx_init_rx_recycle_ring(rx_queue);
+
+	rx_queue->page_remove = 0;
+	rx_queue->page_add = rx_queue->page_ptr_mask + 1;
+	rx_queue->page_recycle_count = 0;
+	rx_queue->page_recycle_failed = 0;
+	rx_queue->page_recycle_full = 0;
+
+	/* Initialise limit fields */
+	max_fill = efx->rxq_entries - EFX_RXD_HEAD_ROOM;
+	max_trigger =
+		max_fill - efx->rx_pages_per_batch * efx->rx_bufs_per_page;
+	if (rx_refill_threshold != 0) {
+		trigger = max_fill * min(rx_refill_threshold, 100U) / 100U;
+		if (trigger > max_trigger)
+			trigger = max_trigger;
+	} else {
+		trigger = max_trigger;
+	}
+
+	rx_queue->max_fill = max_fill;
+	rx_queue->fast_fill_trigger = trigger;
+	rx_queue->refill_enabled = true;
+
+	/* Initialise XDP queue information */
+	rc = xdp_rxq_info_reg(&rx_queue->xdp_rxq_info, efx->net_dev,
+			      rx_queue->core_index);
+
+	if (rc) {
+		netif_err(efx, rx_err, efx->net_dev,
+			  "Failure to initialise XDP queue information rc=%d\n",
+			  rc);
+		efx->xdp_rxq_info_failed = true;
+	} else {
+		rx_queue->xdp_rxq_info_valid = true;
+	}
+
+	/* Set up RX descriptor ring */
+	efx_nic_init_rx(rx_queue);
+}
+
+void efx_fini_rx_queue(struct efx_rx_queue *rx_queue)
+{
+	struct efx_nic *efx = rx_queue->efx;
+	struct efx_rx_buffer *rx_buf;
+	int i;
+
+	netif_dbg(rx_queue->efx, drv, rx_queue->efx->net_dev,
+		  "shutting down RX queue %d\n", efx_rx_queue_index(rx_queue));
+
+	del_timer_sync(&rx_queue->slow_fill);
+
+	/* Release RX buffers from the current read ptr to the write ptr */
+	if (rx_queue->buffer) {
+		for (i = rx_queue->removed_count; i < rx_queue->added_count;
+		     i++) {
+			unsigned int index = i & rx_queue->ptr_mask;
+
+			rx_buf = efx_rx_buffer(rx_queue, index);
+			efx_fini_rx_buffer(rx_queue, rx_buf);
+		}
+	}
+
+	/* Unmap and release the pages in the recycle ring. Remove the ring. */
+	for (i = 0; i <= rx_queue->page_ptr_mask; i++) {
+		struct page *page = rx_queue->page_ring[i];
+		struct efx_rx_page_state *state;
+
+		if (page == NULL)
+			continue;
+
+		state = page_address(page);
+		dma_unmap_page(&efx->pci_dev->dev, state->dma_addr,
+			       PAGE_SIZE << efx->rx_buffer_order,
+			       DMA_FROM_DEVICE);
+		put_page(page);
+	}
+	kfree(rx_queue->page_ring);
+	rx_queue->page_ring = NULL;
+
+	if (rx_queue->xdp_rxq_info_valid)
+		xdp_rxq_info_unreg(&rx_queue->xdp_rxq_info);
+
+	rx_queue->xdp_rxq_info_valid = false;
+}
+
+void efx_remove_rx_queue(struct efx_rx_queue *rx_queue)
+{
+	netif_dbg(rx_queue->efx, drv, rx_queue->efx->net_dev,
+		  "destroying RX queue %d\n", efx_rx_queue_index(rx_queue));
+
+	efx_nic_remove_rx(rx_queue);
+
+	kfree(rx_queue->buffer);
+	rx_queue->buffer = NULL;
+}
+
+/* Unmap a DMA-mapped page.  This function is only called for the final RX
+ * buffer in a page.
+ */
+void efx_unmap_rx_buffer(struct efx_nic *efx,
+			 struct efx_rx_buffer *rx_buf)
+{
+	struct page *page = rx_buf->page;
+
+	if (page) {
+		struct efx_rx_page_state *state = page_address(page);
+
+		dma_unmap_page(&efx->pci_dev->dev,
+			       state->dma_addr,
+			       PAGE_SIZE << efx->rx_buffer_order,
+			       DMA_FROM_DEVICE);
+	}
+}
+
+void efx_free_rx_buffers(struct efx_rx_queue *rx_queue,
+			 struct efx_rx_buffer *rx_buf,
+			 unsigned int num_bufs)
+{
+	do {
+		if (rx_buf->page) {
+			put_page(rx_buf->page);
+			rx_buf->page = NULL;
+		}
+		rx_buf = efx_rx_buf_next(rx_queue, rx_buf);
+	} while (--num_bufs);
+}
+
+void efx_rx_slow_fill(struct timer_list *t)
+{
+	struct efx_rx_queue *rx_queue = from_timer(rx_queue, t, slow_fill);
+
+	/* Post an event to cause NAPI to run and refill the queue */
+	efx_nic_generate_fill_event(rx_queue);
+	++rx_queue->slow_fill_count;
+}
+
+void efx_schedule_slow_fill(struct efx_rx_queue *rx_queue)
+{
+	mod_timer(&rx_queue->slow_fill, jiffies + msecs_to_jiffies(10));
+}
+
+/* efx_init_rx_buffers - create EFX_RX_BATCH page-based RX buffers
+ *
+ * @rx_queue:		Efx RX queue
+ *
+ * This allocates a batch of pages, maps them for DMA, and populates
+ * struct efx_rx_buffers for each one. Return a negative error code or
+ * 0 on success. If a single page can be used for multiple buffers,
+ * then the page will either be inserted fully, or not at all.
+ */
+static int efx_init_rx_buffers(struct efx_rx_queue *rx_queue, bool atomic)
+{
+	unsigned int page_offset, index, count;
+	struct efx_nic *efx = rx_queue->efx;
+	struct efx_rx_page_state *state;
+	struct efx_rx_buffer *rx_buf;
+	dma_addr_t dma_addr;
+	struct page *page;
+
+	count = 0;
+	do {
+		page = efx_reuse_page(rx_queue);
+		if (page == NULL) {
+			page = alloc_pages(__GFP_COMP |
+					   (atomic ? GFP_ATOMIC : GFP_KERNEL),
+					   efx->rx_buffer_order);
+			if (unlikely(page == NULL))
+				return -ENOMEM;
+			dma_addr =
+				dma_map_page(&efx->pci_dev->dev, page, 0,
+					     PAGE_SIZE << efx->rx_buffer_order,
+					     DMA_FROM_DEVICE);
+			if (unlikely(dma_mapping_error(&efx->pci_dev->dev,
+						       dma_addr))) {
+				__free_pages(page, efx->rx_buffer_order);
+				return -EIO;
+			}
+			state = page_address(page);
+			state->dma_addr = dma_addr;
+		} else {
+			state = page_address(page);
+			dma_addr = state->dma_addr;
+		}
+
+		dma_addr += sizeof(struct efx_rx_page_state);
+		page_offset = sizeof(struct efx_rx_page_state);
+
+		do {
+			index = rx_queue->added_count & rx_queue->ptr_mask;
+			rx_buf = efx_rx_buffer(rx_queue, index);
+			rx_buf->dma_addr = dma_addr + efx->rx_ip_align +
+					   XDP_PACKET_HEADROOM;
+			rx_buf->page = page;
+			rx_buf->page_offset = page_offset + efx->rx_ip_align +
+					      XDP_PACKET_HEADROOM;
+			rx_buf->len = efx->rx_dma_len;
+			rx_buf->flags = 0;
+			++rx_queue->added_count;
+			get_page(page);
+			dma_addr += efx->rx_page_buf_step;
+			page_offset += efx->rx_page_buf_step;
+		} while (page_offset + efx->rx_page_buf_step <= PAGE_SIZE);
+
+		rx_buf->flags = EFX_RX_BUF_LAST_IN_PAGE;
+	} while (++count < efx->rx_pages_per_batch);
+
+	return 0;
+}
+
+void efx_rx_config_page_split(struct efx_nic *efx)
+{
+	efx->rx_page_buf_step = ALIGN(efx->rx_dma_len + efx->rx_ip_align +
+				      XDP_PACKET_HEADROOM,
+				      EFX_RX_BUF_ALIGNMENT);
+	efx->rx_bufs_per_page = efx->rx_buffer_order ? 1 :
+		((PAGE_SIZE - sizeof(struct efx_rx_page_state)) /
+		efx->rx_page_buf_step);
+	efx->rx_buffer_truesize = (PAGE_SIZE << efx->rx_buffer_order) /
+		efx->rx_bufs_per_page;
+	efx->rx_pages_per_batch = DIV_ROUND_UP(EFX_RX_PREFERRED_BATCH,
+					       efx->rx_bufs_per_page);
+}
+
+/* efx_fast_push_rx_descriptors - push new RX descriptors quickly
+ * @rx_queue:		RX descriptor queue
+ *
+ * This will aim to fill the RX descriptor queue up to
+ * @rx_queue->@max_fill. If there is insufficient atomic
+ * memory to do so, a slow fill will be scheduled.
+ *
+ * The caller must provide serialisation (none is used here). In practise,
+ * this means this function must run from the NAPI handler, or be called
+ * when NAPI is disabled.
+ */
+void efx_fast_push_rx_descriptors(struct efx_rx_queue *rx_queue, bool atomic)
+{
+	struct efx_nic *efx = rx_queue->efx;
+	unsigned int fill_level, batch_size;
+	int space, rc = 0;
+
+	if (!rx_queue->refill_enabled)
+		return;
+
+	/* Calculate current fill level, and exit if we don't need to fill */
+	fill_level = (rx_queue->added_count - rx_queue->removed_count);
+	EFX_WARN_ON_ONCE_PARANOID(fill_level > rx_queue->efx->rxq_entries);
+	if (fill_level >= rx_queue->fast_fill_trigger)
+		goto out;
+
+	/* Record minimum fill level */
+	if (unlikely(fill_level < rx_queue->min_fill)) {
+		if (fill_level)
+			rx_queue->min_fill = fill_level;
+	}
+
+	batch_size = efx->rx_pages_per_batch * efx->rx_bufs_per_page;
+	space = rx_queue->max_fill - fill_level;
+	EFX_WARN_ON_ONCE_PARANOID(space < batch_size);
+
+	netif_vdbg(rx_queue->efx, rx_status, rx_queue->efx->net_dev,
+		   "RX queue %d fast-filling descriptor ring from"
+		   " level %d to level %d\n",
+		   efx_rx_queue_index(rx_queue), fill_level,
+		   rx_queue->max_fill);
+
+	do {
+		rc = efx_init_rx_buffers(rx_queue, atomic);
+		if (unlikely(rc)) {
+			/* Ensure that we don't leave the rx queue empty */
+			efx_schedule_slow_fill(rx_queue);
+			goto out;
+		}
+	} while ((space -= batch_size) >= batch_size);
+
+	netif_vdbg(rx_queue->efx, rx_status, rx_queue->efx->net_dev,
+		   "RX queue %d fast-filled descriptor ring "
+		   "to level %d\n", efx_rx_queue_index(rx_queue),
+		   rx_queue->added_count - rx_queue->removed_count);
+
+ out:
+	if (rx_queue->notified_count != rx_queue->added_count)
+		efx_nic_notify_rx_desc(rx_queue);
+}
-- 
cgit v1.2.3-59-g8ed1b


From 17d3b21c7ba82cd88855a798b931bb3ac1388cba Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Wed, 8 Jan 2020 16:14:00 +0000
Subject: sfc: move common tx code

Once again, a tiny bit of refactoring was required to stitch the code
together (i.e. adding headers). The moved code deals with managing tx
queues and mappings.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/Makefile    |   2 +-
 drivers/net/ethernet/sfc/tx.c        | 295 ---------------------------------
 drivers/net/ethernet/sfc/tx_common.c | 310 +++++++++++++++++++++++++++++++++++
 3 files changed, 311 insertions(+), 296 deletions(-)
 create mode 100644 drivers/net/ethernet/sfc/tx_common.c

diff --git a/drivers/net/ethernet/sfc/Makefile b/drivers/net/ethernet/sfc/Makefile
index 03c63317a869..40a54df34647 100644
--- a/drivers/net/ethernet/sfc/Makefile
+++ b/drivers/net/ethernet/sfc/Makefile
@@ -1,7 +1,7 @@
 # SPDX-License-Identifier: GPL-2.0
 sfc-y			+= efx.o efx_common.o efx_channels.o nic.o \
 			   farch.o siena.o ef10.o \
-			   tx.o rx.o rx_common.o \
+			   tx.o tx_common.o rx.o rx_common.o \
 			   selftest.o ethtool.o ptp.o tx_tso.o \
 			   mcdi.o mcdi_port.o \
 			   mcdi_mon.o
diff --git a/drivers/net/ethernet/sfc/tx.c b/drivers/net/ethernet/sfc/tx.c
index 1c30354e098c..c9599fcab0b4 100644
--- a/drivers/net/ethernet/sfc/tx.c
+++ b/drivers/net/ethernet/sfc/tx.c
@@ -57,72 +57,6 @@ u8 *efx_tx_get_copy_buffer_limited(struct efx_tx_queue *tx_queue,
 	return efx_tx_get_copy_buffer(tx_queue, buffer);
 }
 
-void efx_dequeue_buffer(struct efx_tx_queue *tx_queue,
-			struct efx_tx_buffer *buffer,
-			unsigned int *pkts_compl,
-			unsigned int *bytes_compl)
-{
-	if (buffer->unmap_len) {
-		struct device *dma_dev = &tx_queue->efx->pci_dev->dev;
-		dma_addr_t unmap_addr = buffer->dma_addr - buffer->dma_offset;
-		if (buffer->flags & EFX_TX_BUF_MAP_SINGLE)
-			dma_unmap_single(dma_dev, unmap_addr, buffer->unmap_len,
-					 DMA_TO_DEVICE);
-		else
-			dma_unmap_page(dma_dev, unmap_addr, buffer->unmap_len,
-				       DMA_TO_DEVICE);
-		buffer->unmap_len = 0;
-	}
-
-	if (buffer->flags & EFX_TX_BUF_SKB) {
-		struct sk_buff *skb = (struct sk_buff *)buffer->skb;
-
-		EFX_WARN_ON_PARANOID(!pkts_compl || !bytes_compl);
-		(*pkts_compl)++;
-		(*bytes_compl) += skb->len;
-		if (tx_queue->timestamping &&
-		    (tx_queue->completed_timestamp_major ||
-		     tx_queue->completed_timestamp_minor)) {
-			struct skb_shared_hwtstamps hwtstamp;
-
-			hwtstamp.hwtstamp =
-				efx_ptp_nic_to_kernel_time(tx_queue);
-			skb_tstamp_tx(skb, &hwtstamp);
-
-			tx_queue->completed_timestamp_major = 0;
-			tx_queue->completed_timestamp_minor = 0;
-		}
-		dev_consume_skb_any((struct sk_buff *)buffer->skb);
-		netif_vdbg(tx_queue->efx, tx_done, tx_queue->efx->net_dev,
-			   "TX queue %d transmission id %x complete\n",
-			   tx_queue->queue, tx_queue->read_count);
-	} else if (buffer->flags & EFX_TX_BUF_XDP) {
-		xdp_return_frame_rx_napi(buffer->xdpf);
-	}
-
-	buffer->len = 0;
-	buffer->flags = 0;
-}
-
-unsigned int efx_tx_max_skb_descs(struct efx_nic *efx)
-{
-	/* Header and payload descriptor for each output segment, plus
-	 * one for every input fragment boundary within a segment
-	 */
-	unsigned int max_descs = EFX_TSO_MAX_SEGS * 2 + MAX_SKB_FRAGS;
-
-	/* Possibly one more per segment for option descriptors */
-	if (efx_nic_rev(efx) >= EFX_REV_HUNT_A0)
-		max_descs += EFX_TSO_MAX_SEGS;
-
-	/* Possibly more for PCIe page boundaries within input fragments */
-	if (PAGE_SIZE > EFX_PAGE_SIZE)
-		max_descs += max_t(unsigned int, MAX_SKB_FRAGS,
-				   DIV_ROUND_UP(GSO_MAX_SIZE, EFX_PAGE_SIZE));
-
-	return max_descs;
-}
-
 static void efx_tx_maybe_stop_queue(struct efx_tx_queue *txq1)
 {
 	/* We need to consider both queues that the net core sees as one */
@@ -334,107 +268,6 @@ static int efx_enqueue_skb_pio(struct efx_tx_queue *tx_queue,
 }
 #endif /* EFX_USE_PIO */
 
-struct efx_tx_buffer *efx_tx_map_chunk(struct efx_tx_queue *tx_queue,
-				       dma_addr_t dma_addr,
-				       size_t len)
-{
-	const struct efx_nic_type *nic_type = tx_queue->efx->type;
-	struct efx_tx_buffer *buffer;
-	unsigned int dma_len;
-
-	/* Map the fragment taking account of NIC-dependent DMA limits. */
-	do {
-		buffer = efx_tx_queue_get_insert_buffer(tx_queue);
-		dma_len = nic_type->tx_limit_len(tx_queue, dma_addr, len);
-
-		buffer->len = dma_len;
-		buffer->dma_addr = dma_addr;
-		buffer->flags = EFX_TX_BUF_CONT;
-		len -= dma_len;
-		dma_addr += dma_len;
-		++tx_queue->insert_count;
-	} while (len);
-
-	return buffer;
-}
-
-/* Map all data from an SKB for DMA and create descriptors on the queue.
- */
-int efx_tx_map_data(struct efx_tx_queue *tx_queue, struct sk_buff *skb,
-		    unsigned int segment_count)
-{
-	struct efx_nic *efx = tx_queue->efx;
-	struct device *dma_dev = &efx->pci_dev->dev;
-	unsigned int frag_index, nr_frags;
-	dma_addr_t dma_addr, unmap_addr;
-	unsigned short dma_flags;
-	size_t len, unmap_len;
-
-	nr_frags = skb_shinfo(skb)->nr_frags;
-	frag_index = 0;
-
-	/* Map header data. */
-	len = skb_headlen(skb);
-	dma_addr = dma_map_single(dma_dev, skb->data, len, DMA_TO_DEVICE);
-	dma_flags = EFX_TX_BUF_MAP_SINGLE;
-	unmap_len = len;
-	unmap_addr = dma_addr;
-
-	if (unlikely(dma_mapping_error(dma_dev, dma_addr)))
-		return -EIO;
-
-	if (segment_count) {
-		/* For TSO we need to put the header in to a separate
-		 * descriptor. Map this separately if necessary.
-		 */
-		size_t header_len = skb_transport_header(skb) - skb->data +
-				(tcp_hdr(skb)->doff << 2u);
-
-		if (header_len != len) {
-			tx_queue->tso_long_headers++;
-			efx_tx_map_chunk(tx_queue, dma_addr, header_len);
-			len -= header_len;
-			dma_addr += header_len;
-		}
-	}
-
-	/* Add descriptors for each fragment. */
-	do {
-		struct efx_tx_buffer *buffer;
-		skb_frag_t *fragment;
-
-		buffer = efx_tx_map_chunk(tx_queue, dma_addr, len);
-
-		/* The final descriptor for a fragment is responsible for
-		 * unmapping the whole fragment.
-		 */
-		buffer->flags = EFX_TX_BUF_CONT | dma_flags;
-		buffer->unmap_len = unmap_len;
-		buffer->dma_offset = buffer->dma_addr - unmap_addr;
-
-		if (frag_index >= nr_frags) {
-			/* Store SKB details with the final buffer for
-			 * the completion.
-			 */
-			buffer->skb = skb;
-			buffer->flags = EFX_TX_BUF_SKB | dma_flags;
-			return 0;
-		}
-
-		/* Move on to the next fragment. */
-		fragment = &skb_shinfo(skb)->frags[frag_index++];
-		len = skb_frag_size(fragment);
-		dma_addr = skb_frag_dma_map(dma_dev, fragment,
-				0, len, DMA_TO_DEVICE);
-		dma_flags = 0;
-		unmap_len = len;
-		unmap_addr = dma_addr;
-
-		if (unlikely(dma_mapping_error(dma_dev, dma_addr)))
-			return -EIO;
-	} while (1);
-}
-
 /* Remove buffers put into a tx_queue for the current packet.
  * None of the buffers must have an skb attached.
  */
@@ -877,131 +710,3 @@ void efx_xmit_done(struct efx_tx_queue *tx_queue, unsigned int index)
 		}
 	}
 }
-
-static unsigned int efx_tx_cb_page_count(struct efx_tx_queue *tx_queue)
-{
-	return DIV_ROUND_UP(tx_queue->ptr_mask + 1, PAGE_SIZE >> EFX_TX_CB_ORDER);
-}
-
-int efx_probe_tx_queue(struct efx_tx_queue *tx_queue)
-{
-	struct efx_nic *efx = tx_queue->efx;
-	unsigned int entries;
-	int rc;
-
-	/* Create the smallest power-of-two aligned ring */
-	entries = max(roundup_pow_of_two(efx->txq_entries), EFX_MIN_DMAQ_SIZE);
-	EFX_WARN_ON_PARANOID(entries > EFX_MAX_DMAQ_SIZE);
-	tx_queue->ptr_mask = entries - 1;
-
-	netif_dbg(efx, probe, efx->net_dev,
-		  "creating TX queue %d size %#x mask %#x\n",
-		  tx_queue->queue, efx->txq_entries, tx_queue->ptr_mask);
-
-	/* Allocate software ring */
-	tx_queue->buffer = kcalloc(entries, sizeof(*tx_queue->buffer),
-				   GFP_KERNEL);
-	if (!tx_queue->buffer)
-		return -ENOMEM;
-
-	tx_queue->cb_page = kcalloc(efx_tx_cb_page_count(tx_queue),
-				    sizeof(tx_queue->cb_page[0]), GFP_KERNEL);
-	if (!tx_queue->cb_page) {
-		rc = -ENOMEM;
-		goto fail1;
-	}
-
-	/* Allocate hardware ring */
-	rc = efx_nic_probe_tx(tx_queue);
-	if (rc)
-		goto fail2;
-
-	return 0;
-
-fail2:
-	kfree(tx_queue->cb_page);
-	tx_queue->cb_page = NULL;
-fail1:
-	kfree(tx_queue->buffer);
-	tx_queue->buffer = NULL;
-	return rc;
-}
-
-void efx_init_tx_queue(struct efx_tx_queue *tx_queue)
-{
-	struct efx_nic *efx = tx_queue->efx;
-
-	netif_dbg(efx, drv, efx->net_dev,
-		  "initialising TX queue %d\n", tx_queue->queue);
-
-	tx_queue->insert_count = 0;
-	tx_queue->write_count = 0;
-	tx_queue->packet_write_count = 0;
-	tx_queue->old_write_count = 0;
-	tx_queue->read_count = 0;
-	tx_queue->old_read_count = 0;
-	tx_queue->empty_read_count = 0 | EFX_EMPTY_COUNT_VALID;
-	tx_queue->xmit_more_available = false;
-	tx_queue->timestamping = (efx_ptp_use_mac_tx_timestamps(efx) &&
-				  tx_queue->channel == efx_ptp_channel(efx));
-	tx_queue->completed_desc_ptr = tx_queue->ptr_mask;
-	tx_queue->completed_timestamp_major = 0;
-	tx_queue->completed_timestamp_minor = 0;
-
-	tx_queue->xdp_tx = efx_channel_is_xdp_tx(tx_queue->channel);
-
-	/* Set up default function pointers. These may get replaced by
-	 * efx_nic_init_tx() based off NIC/queue capabilities.
-	 */
-	tx_queue->handle_tso = efx_enqueue_skb_tso;
-
-	/* Set up TX descriptor ring */
-	efx_nic_init_tx(tx_queue);
-
-	tx_queue->initialised = true;
-}
-
-void efx_fini_tx_queue(struct efx_tx_queue *tx_queue)
-{
-	struct efx_tx_buffer *buffer;
-
-	netif_dbg(tx_queue->efx, drv, tx_queue->efx->net_dev,
-		  "shutting down TX queue %d\n", tx_queue->queue);
-
-	if (!tx_queue->buffer)
-		return;
-
-	/* Free any buffers left in the ring */
-	while (tx_queue->read_count != tx_queue->write_count) {
-		unsigned int pkts_compl = 0, bytes_compl = 0;
-		buffer = &tx_queue->buffer[tx_queue->read_count & tx_queue->ptr_mask];
-		efx_dequeue_buffer(tx_queue, buffer, &pkts_compl, &bytes_compl);
-
-		++tx_queue->read_count;
-	}
-	tx_queue->xmit_more_available = false;
-	netdev_tx_reset_queue(tx_queue->core_txq);
-}
-
-void efx_remove_tx_queue(struct efx_tx_queue *tx_queue)
-{
-	int i;
-
-	if (!tx_queue->buffer)
-		return;
-
-	netif_dbg(tx_queue->efx, drv, tx_queue->efx->net_dev,
-		  "destroying TX queue %d\n", tx_queue->queue);
-	efx_nic_remove_tx(tx_queue);
-
-	if (tx_queue->cb_page) {
-		for (i = 0; i < efx_tx_cb_page_count(tx_queue); i++)
-			efx_nic_free_buffer(tx_queue->efx,
-					    &tx_queue->cb_page[i]);
-		kfree(tx_queue->cb_page);
-		tx_queue->cb_page = NULL;
-	}
-
-	kfree(tx_queue->buffer);
-	tx_queue->buffer = NULL;
-}
diff --git a/drivers/net/ethernet/sfc/tx_common.c b/drivers/net/ethernet/sfc/tx_common.c
new file mode 100644
index 000000000000..e29ade21c4b9
--- /dev/null
+++ b/drivers/net/ethernet/sfc/tx_common.c
@@ -0,0 +1,310 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/****************************************************************************
+ * Driver for Solarflare network controllers and boards
+ * Copyright 2018 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#include "net_driver.h"
+#include "efx.h"
+#include "nic.h"
+#include "tx_common.h"
+
+static unsigned int efx_tx_cb_page_count(struct efx_tx_queue *tx_queue)
+{
+	return DIV_ROUND_UP(tx_queue->ptr_mask + 1,
+			    PAGE_SIZE >> EFX_TX_CB_ORDER);
+}
+
+int efx_probe_tx_queue(struct efx_tx_queue *tx_queue)
+{
+	struct efx_nic *efx = tx_queue->efx;
+	unsigned int entries;
+	int rc;
+
+	/* Create the smallest power-of-two aligned ring */
+	entries = max(roundup_pow_of_two(efx->txq_entries), EFX_MIN_DMAQ_SIZE);
+	EFX_WARN_ON_PARANOID(entries > EFX_MAX_DMAQ_SIZE);
+	tx_queue->ptr_mask = entries - 1;
+
+	netif_dbg(efx, probe, efx->net_dev,
+		  "creating TX queue %d size %#x mask %#x\n",
+		  tx_queue->queue, efx->txq_entries, tx_queue->ptr_mask);
+
+	/* Allocate software ring */
+	tx_queue->buffer = kcalloc(entries, sizeof(*tx_queue->buffer),
+				   GFP_KERNEL);
+	if (!tx_queue->buffer)
+		return -ENOMEM;
+
+	tx_queue->cb_page = kcalloc(efx_tx_cb_page_count(tx_queue),
+				    sizeof(tx_queue->cb_page[0]), GFP_KERNEL);
+	if (!tx_queue->cb_page) {
+		rc = -ENOMEM;
+		goto fail1;
+	}
+
+	/* Allocate hardware ring */
+	rc = efx_nic_probe_tx(tx_queue);
+	if (rc)
+		goto fail2;
+
+	return 0;
+
+fail2:
+	kfree(tx_queue->cb_page);
+	tx_queue->cb_page = NULL;
+fail1:
+	kfree(tx_queue->buffer);
+	tx_queue->buffer = NULL;
+	return rc;
+}
+
+void efx_init_tx_queue(struct efx_tx_queue *tx_queue)
+{
+	struct efx_nic *efx = tx_queue->efx;
+
+	netif_dbg(efx, drv, efx->net_dev,
+		  "initialising TX queue %d\n", tx_queue->queue);
+
+	tx_queue->insert_count = 0;
+	tx_queue->write_count = 0;
+	tx_queue->packet_write_count = 0;
+	tx_queue->old_write_count = 0;
+	tx_queue->read_count = 0;
+	tx_queue->old_read_count = 0;
+	tx_queue->empty_read_count = 0 | EFX_EMPTY_COUNT_VALID;
+	tx_queue->xmit_more_available = false;
+	tx_queue->timestamping = (efx_ptp_use_mac_tx_timestamps(efx) &&
+				  tx_queue->channel == efx_ptp_channel(efx));
+	tx_queue->completed_desc_ptr = tx_queue->ptr_mask;
+	tx_queue->completed_timestamp_major = 0;
+	tx_queue->completed_timestamp_minor = 0;
+
+	tx_queue->xdp_tx = efx_channel_is_xdp_tx(tx_queue->channel);
+
+	/* Set up default function pointers. These may get replaced by
+	 * efx_nic_init_tx() based off NIC/queue capabilities.
+	 */
+	tx_queue->handle_tso = efx_enqueue_skb_tso;
+
+	/* Set up TX descriptor ring */
+	efx_nic_init_tx(tx_queue);
+
+	tx_queue->initialised = true;
+}
+
+void efx_fini_tx_queue(struct efx_tx_queue *tx_queue)
+{
+	struct efx_tx_buffer *buffer;
+
+	netif_dbg(tx_queue->efx, drv, tx_queue->efx->net_dev,
+		  "shutting down TX queue %d\n", tx_queue->queue);
+
+	if (!tx_queue->buffer)
+		return;
+
+	/* Free any buffers left in the ring */
+	while (tx_queue->read_count != tx_queue->write_count) {
+		unsigned int pkts_compl = 0, bytes_compl = 0;
+
+		buffer = &tx_queue->buffer[tx_queue->read_count & tx_queue->ptr_mask];
+		efx_dequeue_buffer(tx_queue, buffer, &pkts_compl, &bytes_compl);
+
+		++tx_queue->read_count;
+	}
+	tx_queue->xmit_more_available = false;
+	netdev_tx_reset_queue(tx_queue->core_txq);
+}
+
+void efx_remove_tx_queue(struct efx_tx_queue *tx_queue)
+{
+	int i;
+
+	if (!tx_queue->buffer)
+		return;
+
+	netif_dbg(tx_queue->efx, drv, tx_queue->efx->net_dev,
+		  "destroying TX queue %d\n", tx_queue->queue);
+	efx_nic_remove_tx(tx_queue);
+
+	if (tx_queue->cb_page) {
+		for (i = 0; i < efx_tx_cb_page_count(tx_queue); i++)
+			efx_nic_free_buffer(tx_queue->efx,
+					    &tx_queue->cb_page[i]);
+		kfree(tx_queue->cb_page);
+		tx_queue->cb_page = NULL;
+	}
+
+	kfree(tx_queue->buffer);
+	tx_queue->buffer = NULL;
+}
+
+void efx_dequeue_buffer(struct efx_tx_queue *tx_queue,
+			struct efx_tx_buffer *buffer,
+			unsigned int *pkts_compl,
+			unsigned int *bytes_compl)
+{
+	if (buffer->unmap_len) {
+		struct device *dma_dev = &tx_queue->efx->pci_dev->dev;
+		dma_addr_t unmap_addr = buffer->dma_addr - buffer->dma_offset;
+
+		if (buffer->flags & EFX_TX_BUF_MAP_SINGLE)
+			dma_unmap_single(dma_dev, unmap_addr, buffer->unmap_len,
+					 DMA_TO_DEVICE);
+		else
+			dma_unmap_page(dma_dev, unmap_addr, buffer->unmap_len,
+				       DMA_TO_DEVICE);
+		buffer->unmap_len = 0;
+	}
+
+	if (buffer->flags & EFX_TX_BUF_SKB) {
+		struct sk_buff *skb = (struct sk_buff *)buffer->skb;
+
+		EFX_WARN_ON_PARANOID(!pkts_compl || !bytes_compl);
+		(*pkts_compl)++;
+		(*bytes_compl) += skb->len;
+		if (tx_queue->timestamping &&
+		    (tx_queue->completed_timestamp_major ||
+		     tx_queue->completed_timestamp_minor)) {
+			struct skb_shared_hwtstamps hwtstamp;
+
+			hwtstamp.hwtstamp =
+				efx_ptp_nic_to_kernel_time(tx_queue);
+			skb_tstamp_tx(skb, &hwtstamp);
+
+			tx_queue->completed_timestamp_major = 0;
+			tx_queue->completed_timestamp_minor = 0;
+		}
+		dev_consume_skb_any((struct sk_buff *)buffer->skb);
+		netif_vdbg(tx_queue->efx, tx_done, tx_queue->efx->net_dev,
+			   "TX queue %d transmission id %x complete\n",
+			   tx_queue->queue, tx_queue->read_count);
+	} else if (buffer->flags & EFX_TX_BUF_XDP) {
+		xdp_return_frame_rx_napi(buffer->xdpf);
+	}
+
+	buffer->len = 0;
+	buffer->flags = 0;
+}
+
+struct efx_tx_buffer *efx_tx_map_chunk(struct efx_tx_queue *tx_queue,
+				       dma_addr_t dma_addr, size_t len)
+{
+	const struct efx_nic_type *nic_type = tx_queue->efx->type;
+	struct efx_tx_buffer *buffer;
+	unsigned int dma_len;
+
+	/* Map the fragment taking account of NIC-dependent DMA limits. */
+	do {
+		buffer = efx_tx_queue_get_insert_buffer(tx_queue);
+		dma_len = nic_type->tx_limit_len(tx_queue, dma_addr, len);
+
+		buffer->len = dma_len;
+		buffer->dma_addr = dma_addr;
+		buffer->flags = EFX_TX_BUF_CONT;
+		len -= dma_len;
+		dma_addr += dma_len;
+		++tx_queue->insert_count;
+	} while (len);
+
+	return buffer;
+}
+
+/* Map all data from an SKB for DMA and create descriptors on the queue. */
+int efx_tx_map_data(struct efx_tx_queue *tx_queue, struct sk_buff *skb,
+		    unsigned int segment_count)
+{
+	struct efx_nic *efx = tx_queue->efx;
+	struct device *dma_dev = &efx->pci_dev->dev;
+	unsigned int frag_index, nr_frags;
+	dma_addr_t dma_addr, unmap_addr;
+	unsigned short dma_flags;
+	size_t len, unmap_len;
+
+	nr_frags = skb_shinfo(skb)->nr_frags;
+	frag_index = 0;
+
+	/* Map header data. */
+	len = skb_headlen(skb);
+	dma_addr = dma_map_single(dma_dev, skb->data, len, DMA_TO_DEVICE);
+	dma_flags = EFX_TX_BUF_MAP_SINGLE;
+	unmap_len = len;
+	unmap_addr = dma_addr;
+
+	if (unlikely(dma_mapping_error(dma_dev, dma_addr)))
+		return -EIO;
+
+	if (segment_count) {
+		/* For TSO we need to put the header in to a separate
+		 * descriptor. Map this separately if necessary.
+		 */
+		size_t header_len = skb_transport_header(skb) - skb->data +
+				(tcp_hdr(skb)->doff << 2u);
+
+		if (header_len != len) {
+			tx_queue->tso_long_headers++;
+			efx_tx_map_chunk(tx_queue, dma_addr, header_len);
+			len -= header_len;
+			dma_addr += header_len;
+		}
+	}
+
+	/* Add descriptors for each fragment. */
+	do {
+		struct efx_tx_buffer *buffer;
+		skb_frag_t *fragment;
+
+		buffer = efx_tx_map_chunk(tx_queue, dma_addr, len);
+
+		/* The final descriptor for a fragment is responsible for
+		 * unmapping the whole fragment.
+		 */
+		buffer->flags = EFX_TX_BUF_CONT | dma_flags;
+		buffer->unmap_len = unmap_len;
+		buffer->dma_offset = buffer->dma_addr - unmap_addr;
+
+		if (frag_index >= nr_frags) {
+			/* Store SKB details with the final buffer for
+			 * the completion.
+			 */
+			buffer->skb = skb;
+			buffer->flags = EFX_TX_BUF_SKB | dma_flags;
+			return 0;
+		}
+
+		/* Move on to the next fragment. */
+		fragment = &skb_shinfo(skb)->frags[frag_index++];
+		len = skb_frag_size(fragment);
+		dma_addr = skb_frag_dma_map(dma_dev, fragment, 0, len,
+					    DMA_TO_DEVICE);
+		dma_flags = 0;
+		unmap_len = len;
+		unmap_addr = dma_addr;
+
+		if (unlikely(dma_mapping_error(dma_dev, dma_addr)))
+			return -EIO;
+	} while (1);
+}
+
+unsigned int efx_tx_max_skb_descs(struct efx_nic *efx)
+{
+	/* Header and payload descriptor for each output segment, plus
+	 * one for every input fragment boundary within a segment
+	 */
+	unsigned int max_descs = EFX_TSO_MAX_SEGS * 2 + MAX_SKB_FRAGS;
+
+	/* Possibly one more per segment for option descriptors */
+	if (efx_nic_rev(efx) >= EFX_REV_HUNT_A0)
+		max_descs += EFX_TSO_MAX_SEGS;
+
+	/* Possibly more for PCIe page boundaries within input fragments */
+	if (PAGE_SIZE > EFX_PAGE_SIZE)
+		max_descs += max_t(unsigned int, MAX_SKB_FRAGS,
+				   DIV_ROUND_UP(GSO_MAX_SIZE, EFX_PAGE_SIZE));
+
+	return max_descs;
+}
-- 
cgit v1.2.3-59-g8ed1b


From 542d3065f2b1a60a0cfc259c9a36faa470761d78 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Wed, 8 Jan 2020 22:44:43 +0100
Subject: socket: fix unused-function warning

When procfs is disabled, the fdinfo code causes a harmless
warning:

net/socket.c:1000:13: error: 'sock_show_fdinfo' defined but not used [-Werror=unused-function]
 static void sock_show_fdinfo(struct seq_file *m, struct file *f)

Move the function definition up so we can use a single #ifdef
around it.

Fixes: b4653342b151 ("net: Allow to show socket-specific information in /proc/[pid]/fdinfo/[fd]")
Suggested-by: Al Viro <viro@zeniv.linux.org.uk>
Acked-by: Kirill Tkhai <ktkhai@virtuozzo.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/socket.c | 23 ++++++++++++-----------
 1 file changed, 12 insertions(+), 11 deletions(-)

diff --git a/net/socket.c b/net/socket.c
index 51bf34995bcb..b79a05de7c6e 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -128,7 +128,18 @@ static ssize_t sock_sendpage(struct file *file, struct page *page,
 static ssize_t sock_splice_read(struct file *file, loff_t *ppos,
 				struct pipe_inode_info *pipe, size_t len,
 				unsigned int flags);
-static void sock_show_fdinfo(struct seq_file *m, struct file *f);
+
+#ifdef CONFIG_PROC_FS
+static void sock_show_fdinfo(struct seq_file *m, struct file *f)
+{
+	struct socket *sock = f->private_data;
+
+	if (sock->ops->show_fdinfo)
+		sock->ops->show_fdinfo(m, sock);
+}
+#else
+#define sock_show_fdinfo NULL
+#endif
 
 /*
  *	Socket files have a set of 'special' operations as well as the generic file ones. These don't appear
@@ -151,9 +162,7 @@ static const struct file_operations socket_file_ops = {
 	.sendpage =	sock_sendpage,
 	.splice_write = generic_splice_sendpage,
 	.splice_read =	sock_splice_read,
-#ifdef CONFIG_PROC_FS
 	.show_fdinfo =	sock_show_fdinfo,
-#endif
 };
 
 /*
@@ -997,14 +1006,6 @@ static ssize_t sock_write_iter(struct kiocb *iocb, struct iov_iter *from)
 	return res;
 }
 
-static void sock_show_fdinfo(struct seq_file *m, struct file *f)
-{
-	struct socket *sock = f->private_data;
-
-	if (sock->ops->show_fdinfo)
-		sock->ops->show_fdinfo(m, sock);
-}
-
 /*
  * Atomic setting of ioctl hooks to avoid race
  * with module unload.
-- 
cgit v1.2.3-59-g8ed1b


From dcfea72e79b0aa7a057c8f6024169d86a1bbc84b Mon Sep 17 00:00:00 2001
From: "Jason A. Donenfeld" <Jason@zx2c4.com>
Date: Wed, 8 Jan 2020 16:59:02 -0500
Subject: net: introduce skb_list_walk_safe for skb segment walking

As part of the continual effort to remove direct usage of skb->next and
skb->prev, this patch adds a helper for iterating through the
singly-linked variant of skb lists, which are used for lists of GSO
packet. The name "skb_list_..." has been chosen to match the existing
function, "kfree_skb_list, which also operates on these singly-linked
lists, and the "..._walk_safe" part is the same idiom as elsewhere in
the kernel.

This patch removes the helper from wireguard and puts it into
linux/skbuff.h, while making it a bit more robust for general usage. In
particular, parenthesis are added around the macro argument usage, and it
now accounts for trying to iterate through an already-null skb pointer,
which will simply run the iteration zero times. This latter enhancement
means it can be used to replace both do { ... } while and while (...)
open-coded idioms.

This should take care of these three possible usages, which match all
current methods of iterations.

skb_list_walk_safe(segs, skb, next) { ... }
skb_list_walk_safe(skb, skb, next) { ... }
skb_list_walk_safe(segs, skb, segs) { ... }

Gcc appears to generate efficient code for each of these.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/wireguard/device.h | 8 --------
 include/linux/skbuff.h         | 5 +++++
 2 files changed, 5 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireguard/device.h b/drivers/net/wireguard/device.h
index c91f3051c5c7..b15a8be9d816 100644
--- a/drivers/net/wireguard/device.h
+++ b/drivers/net/wireguard/device.h
@@ -62,12 +62,4 @@ struct wg_device {
 int wg_device_init(void);
 void wg_device_uninit(void);
 
-/* Later after the dust settles, this can be moved into include/linux/skbuff.h,
- * where virtually all code that deals with GSO segs can benefit, around ~30
- * drivers as of writing.
- */
-#define skb_list_walk_safe(first, skb, next)                                   \
-	for (skb = first, next = skb->next; skb;                               \
-	     skb = next, next = skb ? skb->next : NULL)
-
 #endif /* _WG_DEVICE_H */
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index e9133bcf0544..64e5b1be9ff5 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -1478,6 +1478,11 @@ static inline void skb_mark_not_on_list(struct sk_buff *skb)
 	skb->next = NULL;
 }
 
+/* Iterate through singly-linked GSO fragments of an skb. */
+#define skb_list_walk_safe(first, skb, next)                                   \
+	for ((skb) = (first), (next) = (skb) ? (skb)->next : NULL; (skb);      \
+	     (skb) = (next), (next) = (skb) ? (skb)->next : NULL)
+
 static inline void skb_list_del_init(struct sk_buff *skb)
 {
 	__list_del_entry(&skb->list);
-- 
cgit v1.2.3-59-g8ed1b


From 5643a552d31242e5adf3dbefba26c90b1bce2826 Mon Sep 17 00:00:00 2001
From: "Jason A. Donenfeld" <Jason@zx2c4.com>
Date: Wed, 8 Jan 2020 16:59:03 -0500
Subject: net: tap: use skb_list_walk_safe helper for gso segments

This is a straight-forward conversion case for the new function, and
while we're at it, we can remove a null write to skb->next by replacing
it with skb_mark_not_on_list.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/tap.c | 14 ++++++--------
 1 file changed, 6 insertions(+), 8 deletions(-)

diff --git a/drivers/net/tap.c b/drivers/net/tap.c
index a6d63665ad03..1f4bdd94407a 100644
--- a/drivers/net/tap.c
+++ b/drivers/net/tap.c
@@ -341,6 +341,7 @@ rx_handler_result_t tap_handle_frame(struct sk_buff **pskb)
 		features |= tap->tap_features;
 	if (netif_needs_gso(skb, features)) {
 		struct sk_buff *segs = __skb_gso_segment(skb, features, false);
+		struct sk_buff *next;
 
 		if (IS_ERR(segs))
 			goto drop;
@@ -352,16 +353,13 @@ rx_handler_result_t tap_handle_frame(struct sk_buff **pskb)
 		}
 
 		consume_skb(skb);
-		while (segs) {
-			struct sk_buff *nskb = segs->next;
-
-			segs->next = NULL;
-			if (ptr_ring_produce(&q->ring, segs)) {
-				kfree_skb(segs);
-				kfree_skb_list(nskb);
+		skb_list_walk_safe(segs, skb, next) {
+			skb_mark_not_on_list(skb);
+			if (ptr_ring_produce(&q->ring, skb)) {
+				kfree_skb(skb);
+				kfree_skb_list(next);
 				break;
 			}
-			segs = nskb;
 		}
 	} else {
 		/* If we receive a partial checksum and the tap side
-- 
cgit v1.2.3-59-g8ed1b


From 1d7a7438d78f2da56b6b3855be462cf0fc809e56 Mon Sep 17 00:00:00 2001
From: "Jason A. Donenfeld" <Jason@zx2c4.com>
Date: Wed, 8 Jan 2020 16:59:04 -0500
Subject: net: r8152: use skb_list_walk_safe helper for gso segments

This is a straight-forward conversion case for the new function, and
while we're at it, we can remove a null write to skb->next by replacing
it with skb_mark_not_on_list.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/usb/r8152.c | 12 +++++-------
 1 file changed, 5 insertions(+), 7 deletions(-)

diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c
index 9ec1da429514..fe22a582373b 100644
--- a/drivers/net/usb/r8152.c
+++ b/drivers/net/usb/r8152.c
@@ -1897,8 +1897,8 @@ static void r8152_csum_workaround(struct r8152 *tp, struct sk_buff *skb,
 {
 	if (skb_shinfo(skb)->gso_size) {
 		netdev_features_t features = tp->netdev->features;
+		struct sk_buff *segs, *seg, *next;
 		struct sk_buff_head seg_list;
-		struct sk_buff *segs, *nskb;
 
 		features &= ~(NETIF_F_SG | NETIF_F_IPV6_CSUM | NETIF_F_TSO6);
 		segs = skb_gso_segment(skb, features);
@@ -1907,12 +1907,10 @@ static void r8152_csum_workaround(struct r8152 *tp, struct sk_buff *skb,
 
 		__skb_queue_head_init(&seg_list);
 
-		do {
-			nskb = segs;
-			segs = segs->next;
-			nskb->next = NULL;
-			__skb_queue_tail(&seg_list, nskb);
-		} while (segs);
+		skb_list_walk_safe(segs, seg, next) {
+			skb_mark_not_on_list(seg);
+			__skb_queue_tail(&seg_list, seg);
+		}
 
 		skb_queue_splice(&seg_list, list);
 		dev_kfree_skb(skb);
-- 
cgit v1.2.3-59-g8ed1b


From 9f0722380f75a0873c9d70286df8354c6895f983 Mon Sep 17 00:00:00 2001
From: "Jason A. Donenfeld" <Jason@zx2c4.com>
Date: Wed, 8 Jan 2020 16:59:05 -0500
Subject: net: tg3: use skb_list_walk_safe helper for gso segments

This is a straight-forward conversion case for the new function, and
while we're at it, we can remove a null write to skb->next by replacing
it with skb_mark_not_on_list.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/tg3.c | 12 +++++-------
 1 file changed, 5 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c
index 460b4992914a..88466255bf66 100644
--- a/drivers/net/ethernet/broadcom/tg3.c
+++ b/drivers/net/ethernet/broadcom/tg3.c
@@ -7874,8 +7874,8 @@ static netdev_tx_t tg3_start_xmit(struct sk_buff *, struct net_device *);
 static int tg3_tso_bug(struct tg3 *tp, struct tg3_napi *tnapi,
 		       struct netdev_queue *txq, struct sk_buff *skb)
 {
-	struct sk_buff *segs, *nskb;
 	u32 frag_cnt_est = skb_shinfo(skb)->gso_segs * 3;
+	struct sk_buff *segs, *seg, *next;
 
 	/* Estimate the number of fragments in the worst case */
 	if (unlikely(tg3_tx_avail(tnapi) <= frag_cnt_est)) {
@@ -7898,12 +7898,10 @@ static int tg3_tso_bug(struct tg3 *tp, struct tg3_napi *tnapi,
 	if (IS_ERR(segs) || !segs)
 		goto tg3_tso_bug_end;
 
-	do {
-		nskb = segs;
-		segs = segs->next;
-		nskb->next = NULL;
-		tg3_start_xmit(nskb, tp->dev);
-	} while (segs);
+	skb_list_walk_safe(segs, seg, next) {
+		skb_mark_not_on_list(seg);
+		tg3_start_xmit(seg, tp->dev);
+	}
 
 tg3_tso_bug_end:
 	dev_consume_skb_any(skb);
-- 
cgit v1.2.3-59-g8ed1b


From 90919f14507b7a3661e58b8d17f8444a63f5cafa Mon Sep 17 00:00:00 2001
From: "Jason A. Donenfeld" <Jason@zx2c4.com>
Date: Wed, 8 Jan 2020 16:59:06 -0500
Subject: net: sunvnet: use skb_list_walk_safe helper for gso segments

This is a straight-forward conversion case for the new function, and
while we're at it, we can remove a null write to skb->next by replacing
it with skb_mark_not_on_list.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sun/sunvnet_common.c | 9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/sun/sunvnet_common.c b/drivers/net/ethernet/sun/sunvnet_common.c
index a601a306f9a5..c23ce838ff63 100644
--- a/drivers/net/ethernet/sun/sunvnet_common.c
+++ b/drivers/net/ethernet/sun/sunvnet_common.c
@@ -1223,7 +1223,7 @@ vnet_handle_offloads(struct vnet_port *port, struct sk_buff *skb,
 {
 	struct net_device *dev = VNET_PORT_TO_NET_DEVICE(port);
 	struct vio_dring_state *dr = &port->vio.drings[VIO_DRIVER_TX_RING];
-	struct sk_buff *segs;
+	struct sk_buff *segs, *curr, *next;
 	int maclen, datalen;
 	int status;
 	int gso_size, gso_type, gso_segs;
@@ -1282,11 +1282,8 @@ vnet_handle_offloads(struct vnet_port *port, struct sk_buff *skb,
 	skb_reset_mac_header(skb);
 
 	status = 0;
-	while (segs) {
-		struct sk_buff *curr = segs;
-
-		segs = segs->next;
-		curr->next = NULL;
+	skb_list_walk_safe(segs, curr, next) {
+		skb_mark_not_on_list(curr);
 		if (port->tso && curr->len > dev->mtu) {
 			skb_shinfo(curr)->gso_size = gso_size;
 			skb_shinfo(curr)->gso_type = gso_type;
-- 
cgit v1.2.3-59-g8ed1b


From 69b4ed5cbff54b8279f53a65da111e8d0cc01405 Mon Sep 17 00:00:00 2001
From: "Jason A. Donenfeld" <Jason@zx2c4.com>
Date: Wed, 8 Jan 2020 16:59:07 -0500
Subject: net: sfc: use skb_list_walk_safe helper for gso segments

This is a straight-forward conversion case for the new function, and
while we're at it, we can remove a null write to skb->next by replacing
it with skb_mark_not_on_list.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/tx.c | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/sfc/tx.c b/drivers/net/ethernet/sfc/tx.c
index c9599fcab0b4..e4f0f5c94589 100644
--- a/drivers/net/ethernet/sfc/tx.c
+++ b/drivers/net/ethernet/sfc/tx.c
@@ -307,12 +307,9 @@ static int efx_tx_tso_fallback(struct efx_tx_queue *tx_queue,
 	dev_consume_skb_any(skb);
 	skb = segments;
 
-	while (skb) {
-		next = skb->next;
-		skb->next = NULL;
-
+	skb_list_walk_safe(skb, skb, next) {
+		skb_mark_not_on_list(skb);
 		efx_enqueue_skb(tx_queue, skb);
-		skb = next;
 	}
 
 	return 0;
-- 
cgit v1.2.3-59-g8ed1b


From 536577f36ff7afd0a0aff4ebef62db8afe6fc904 Mon Sep 17 00:00:00 2001
From: "Jason A. Donenfeld" <Jason@zx2c4.com>
Date: Wed, 8 Jan 2020 16:59:08 -0500
Subject: net: myri10ge: use skb_list_walk_safe helper for gso segments

This is a straight-forward conversion case for the new function, and
while we're at it, we can remove a null write to skb->next by replacing
it with skb_mark_not_on_list.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/myricom/myri10ge/myri10ge.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
index c979f38a2e0c..2ee0d0be113a 100644
--- a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
+++ b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
@@ -2892,7 +2892,7 @@ drop:
 static netdev_tx_t myri10ge_sw_tso(struct sk_buff *skb,
 					 struct net_device *dev)
 {
-	struct sk_buff *segs, *curr;
+	struct sk_buff *segs, *curr, *next;
 	struct myri10ge_priv *mgp = netdev_priv(dev);
 	struct myri10ge_slice_state *ss;
 	netdev_tx_t status;
@@ -2901,10 +2901,8 @@ static netdev_tx_t myri10ge_sw_tso(struct sk_buff *skb,
 	if (IS_ERR(segs))
 		goto drop;
 
-	while (segs) {
-		curr = segs;
-		segs = segs->next;
-		curr->next = NULL;
+	skb_list_walk_safe(segs, curr, next) {
+		skb_mark_not_on_list(curr);
 		status = myri10ge_xmit(curr, dev);
 		if (status != 0) {
 			dev_kfree_skb_any(curr);
-- 
cgit v1.2.3-59-g8ed1b


From 66de4b179f163ea106f292778563b404958e4f42 Mon Sep 17 00:00:00 2001
From: "Jason A. Donenfeld" <Jason@zx2c4.com>
Date: Wed, 8 Jan 2020 16:59:09 -0500
Subject: net: iwlwifi: use skb_list_walk_safe helper for gso segments

This is a straight-forward conversion case for the new function, and
while we're at it, we can remove a null write to skb->next by replacing
it with skb_mark_not_on_list.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/wireless/intel/iwlwifi/mvm/tx.c | 9 ++-------
 1 file changed, 2 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index dc5c02fbc65a..6a241d37a057 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -847,10 +847,7 @@ iwl_mvm_tx_tso_segment(struct sk_buff *skb, unsigned int num_subframes,
 	else if (next)
 		consume_skb(skb);
 
-	while (next) {
-		tmp = next;
-		next = tmp->next;
-
+	skb_list_walk_safe(next, tmp, next) {
 		memcpy(tmp->cb, cb, sizeof(tmp->cb));
 		/*
 		 * Compute the length of all the data added for the A-MSDU.
@@ -880,9 +877,7 @@ iwl_mvm_tx_tso_segment(struct sk_buff *skb, unsigned int num_subframes,
 			skb_shinfo(tmp)->gso_size = 0;
 		}
 
-		tmp->prev = NULL;
-		tmp->next = NULL;
-
+		skb_mark_not_on_list(tmp);
 		__skb_queue_tail(mpdus_skb, tmp);
 		i++;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 6181e5cb752e5de9f56fbcee3f0206a2c51f1478 Mon Sep 17 00:00:00 2001
From: Vikas Gupta <vikas.gupta@broadcom.com>
Date: Thu, 2 Jan 2020 21:18:09 +0530
Subject: devlink: add support for reporter recovery completion

It is possible that a reporter recovery completion do not finish
successfully when recovery is triggered via
devlink_health_reporter_recover as recovery could be processed in
different context. In such scenario an error is returned by driver when
recover hook is invoked and successful recovery completion is
intimated later.
Expose devlink recover done API to update recovery stats.

Signed-off-by: Vikas Gupta <vikas.gupta@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/devlink.h |  2 ++
 net/core/devlink.c    | 11 +++++++++--
 2 files changed, 11 insertions(+), 2 deletions(-)

diff --git a/include/net/devlink.h b/include/net/devlink.h
index 47f87b2fcf63..453f45cc1519 100644
--- a/include/net/devlink.h
+++ b/include/net/devlink.h
@@ -1000,6 +1000,8 @@ int devlink_health_report(struct devlink_health_reporter *reporter,
 void
 devlink_health_reporter_state_update(struct devlink_health_reporter *reporter,
 				     enum devlink_health_reporter_state state);
+void
+devlink_health_reporter_recovery_done(struct devlink_health_reporter *reporter);
 
 bool devlink_is_reload_failed(const struct devlink *devlink);
 
diff --git a/net/core/devlink.c b/net/core/devlink.c
index 4c63c9a4c09e..e686ae67cd96 100644
--- a/net/core/devlink.c
+++ b/net/core/devlink.c
@@ -4860,6 +4860,14 @@ devlink_health_reporter_state_update(struct devlink_health_reporter *reporter,
 }
 EXPORT_SYMBOL_GPL(devlink_health_reporter_state_update);
 
+void
+devlink_health_reporter_recovery_done(struct devlink_health_reporter *reporter)
+{
+	reporter->recovery_count++;
+	reporter->last_recovery_ts = jiffies;
+}
+EXPORT_SYMBOL_GPL(devlink_health_reporter_recovery_done);
+
 static int
 devlink_health_reporter_recover(struct devlink_health_reporter *reporter,
 				void *priv_ctx, struct netlink_ext_ack *extack)
@@ -4876,9 +4884,8 @@ devlink_health_reporter_recover(struct devlink_health_reporter *reporter,
 	if (err)
 		return err;
 
-	reporter->recovery_count++;
+	devlink_health_reporter_recovery_done(reporter);
 	reporter->health_state = DEVLINK_HEALTH_REPORTER_STATE_HEALTHY;
-	reporter->last_recovery_ts = jiffies;
 
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 97ff3bd37face9bc1bc824cc08241fc1f860ff46 Mon Sep 17 00:00:00 2001
From: Vikas Gupta <vikas.gupta@broadcom.com>
Date: Thu, 2 Jan 2020 21:18:10 +0530
Subject: devlink: add devink notification when reporter update health state

add a devlink notification when reporter update the health
state.

Signed-off-by: Vikas Gupta <vikas.gupta@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/devlink.c | 59 ++++++++++++++++++++++++++++++++++++++----------------
 1 file changed, 42 insertions(+), 17 deletions(-)

diff --git a/net/core/devlink.c b/net/core/devlink.c
index e686ae67cd96..d30aa47052aa 100644
--- a/net/core/devlink.c
+++ b/net/core/devlink.c
@@ -4843,23 +4843,6 @@ devlink_health_reporter_destroy(struct devlink_health_reporter *reporter)
 }
 EXPORT_SYMBOL_GPL(devlink_health_reporter_destroy);
 
-void
-devlink_health_reporter_state_update(struct devlink_health_reporter *reporter,
-				     enum devlink_health_reporter_state state)
-{
-	if (WARN_ON(state != DEVLINK_HEALTH_REPORTER_STATE_HEALTHY &&
-		    state != DEVLINK_HEALTH_REPORTER_STATE_ERROR))
-		return;
-
-	if (reporter->health_state == state)
-		return;
-
-	reporter->health_state = state;
-	trace_devlink_health_reporter_state_update(reporter->devlink,
-						   reporter->ops->name, state);
-}
-EXPORT_SYMBOL_GPL(devlink_health_reporter_state_update);
-
 void
 devlink_health_reporter_recovery_done(struct devlink_health_reporter *reporter)
 {
@@ -5097,6 +5080,48 @@ genlmsg_cancel:
 	return -EMSGSIZE;
 }
 
+static void devlink_recover_notify(struct devlink_health_reporter *reporter,
+				   enum devlink_command cmd)
+{
+	struct sk_buff *msg;
+	int err;
+
+	WARN_ON(cmd != DEVLINK_CMD_HEALTH_REPORTER_RECOVER);
+
+	msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
+	if (!msg)
+		return;
+
+	err = devlink_nl_health_reporter_fill(msg, reporter->devlink,
+					      reporter, cmd, 0, 0, 0);
+	if (err) {
+		nlmsg_free(msg);
+		return;
+	}
+
+	genlmsg_multicast_netns(&devlink_nl_family,
+				devlink_net(reporter->devlink),
+				msg, 0, DEVLINK_MCGRP_CONFIG, GFP_KERNEL);
+}
+
+void
+devlink_health_reporter_state_update(struct devlink_health_reporter *reporter,
+				     enum devlink_health_reporter_state state)
+{
+	if (WARN_ON(state != DEVLINK_HEALTH_REPORTER_STATE_HEALTHY &&
+		    state != DEVLINK_HEALTH_REPORTER_STATE_ERROR))
+		return;
+
+	if (reporter->health_state == state)
+		return;
+
+	reporter->health_state = state;
+	trace_devlink_health_reporter_state_update(reporter->devlink,
+						   reporter->ops->name, state);
+	devlink_recover_notify(reporter, DEVLINK_CMD_HEALTH_REPORTER_RECOVER);
+}
+EXPORT_SYMBOL_GPL(devlink_health_reporter_state_update);
+
 static int devlink_nl_cmd_health_reporter_get_doit(struct sk_buff *skb,
 						   struct genl_info *info)
 {
-- 
cgit v1.2.3-59-g8ed1b


From 737d7a6c55964955604b9de398dac3791bab5a64 Mon Sep 17 00:00:00 2001
From: Vikas Gupta <vikas.gupta@broadcom.com>
Date: Thu, 2 Jan 2020 21:18:11 +0530
Subject: bnxt_en: Call recovery done after reset is successfully done

Return EINPROGRESS to devlink health reporter recover as we are not yet
done and call devlink_health_reporter_recovery_done once reset is
successfully completed from workqueue context.

Signed-off-by: Vikas Gupta <vikas.gupta@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c         |  1 +
 drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c | 14 ++++++++++++--
 drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.h |  1 +
 3 files changed, 14 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 7b0fe19d9ecb..39d4309b17fb 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -10822,6 +10822,7 @@ static void bnxt_fw_reset_task(struct work_struct *work)
 		smp_mb__before_atomic();
 		clear_bit(BNXT_STATE_IN_FW_RESET, &bp->state);
 		bnxt_ulp_start(bp, rc);
+		bnxt_dl_health_recovery_done(bp);
 		bnxt_dl_health_status_update(bp, true);
 		rtnl_unlock();
 		break;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c
index 3eedd4477218..0c3d224637b9 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c
@@ -89,7 +89,7 @@ static int bnxt_fw_reset_recover(struct devlink_health_reporter *reporter,
 		return -EOPNOTSUPP;
 
 	bnxt_fw_reset(bp);
-	return 0;
+	return -EINPROGRESS;
 }
 
 static const
@@ -116,7 +116,7 @@ static int bnxt_fw_fatal_recover(struct devlink_health_reporter *reporter,
 	else if (event == BNXT_FW_EXCEPTION_SP_EVENT)
 		bnxt_fw_exception(bp);
 
-	return 0;
+	return -EINPROGRESS;
 }
 
 static const
@@ -262,6 +262,16 @@ void bnxt_dl_health_status_update(struct bnxt *bp, bool healthy)
 	health->fatal = false;
 }
 
+void bnxt_dl_health_recovery_done(struct bnxt *bp)
+{
+	struct bnxt_fw_health *hlth = bp->fw_health;
+
+	if (hlth->fatal)
+		devlink_health_reporter_recovery_done(hlth->fw_fatal_reporter);
+	else
+		devlink_health_reporter_recovery_done(hlth->fw_reset_reporter);
+}
+
 static const struct devlink_ops bnxt_dl_ops = {
 #ifdef CONFIG_BNXT_SRIOV
 	.eswitch_mode_set = bnxt_dl_eswitch_mode_set,
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.h
index 6db6c3dac472..08aaa4441c78 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.h
@@ -58,6 +58,7 @@ struct bnxt_dl_nvm_param {
 
 void bnxt_devlink_health_report(struct bnxt *bp, unsigned long event);
 void bnxt_dl_health_status_update(struct bnxt *bp, bool healthy);
+void bnxt_dl_health_recovery_done(struct bnxt *bp);
 void bnxt_dl_fw_reporters_create(struct bnxt *bp);
 void bnxt_dl_fw_reporters_destroy(struct bnxt *bp, bool all);
 int bnxt_dl_register(struct bnxt *bp);
-- 
cgit v1.2.3-59-g8ed1b


From 4d776482ecc689bdd68627985ac4cb5a6f325953 Mon Sep 17 00:00:00 2001
From: Florian Fainelli <f.fainelli@gmail.com>
Date: Tue, 7 Jan 2020 21:06:05 -0800
Subject: net: dsa: Get information about stacked DSA protocol

It is possible to stack multiple DSA switches in a way that they are not
part of the tree (disjoint) but the DSA master of a switch is a DSA
slave of another. When that happens switch drivers may have to know this
is the case so as to determine whether their tagging protocol has a
remove chance of working.

This is useful for specific switch drivers such as b53 where devices
have been known to be stacked in the wild without the Broadcom tag
protocol supporting that feature. This allows b53 to continue supporting
those devices by forcing the disabling of Broadcom tags on the outermost
switches if necessary.

The get_tag_protocol() function is therefore updated to gain an
additional enum dsa_tag_protocol argument which denotes the current
tagging protocol used by the DSA master we are attached to, else
DSA_TAG_PROTO_NONE for the top of the dsa_switch_tree.

Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/b53/b53_common.c       | 22 ++++++++++++++--------
 drivers/net/dsa/b53/b53_priv.h         |  4 +++-
 drivers/net/dsa/dsa_loop.c             |  3 ++-
 drivers/net/dsa/lan9303-core.c         |  3 ++-
 drivers/net/dsa/lantiq_gswip.c         |  3 ++-
 drivers/net/dsa/microchip/ksz8795.c    |  3 ++-
 drivers/net/dsa/microchip/ksz9477.c    |  3 ++-
 drivers/net/dsa/mt7530.c               |  3 ++-
 drivers/net/dsa/mv88e6060.c            |  3 ++-
 drivers/net/dsa/mv88e6xxx/chip.c       |  3 ++-
 drivers/net/dsa/ocelot/felix.c         |  3 ++-
 drivers/net/dsa/qca/ar9331.c           |  3 ++-
 drivers/net/dsa/qca8k.c                |  3 ++-
 drivers/net/dsa/rtl8366rb.c            |  3 ++-
 drivers/net/dsa/sja1105/sja1105_main.c |  3 ++-
 drivers/net/dsa/vitesse-vsc73xx-core.c |  3 ++-
 include/net/dsa.h                      |  3 ++-
 net/dsa/dsa2.c                         | 31 +++++++++++++++++++++++++++++--
 net/dsa/dsa_priv.h                     |  1 +
 net/dsa/slave.c                        |  4 +---
 20 files changed, 78 insertions(+), 29 deletions(-)

diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c
index edacacfc9365..2b530a31ef0f 100644
--- a/drivers/net/dsa/b53/b53_common.c
+++ b/drivers/net/dsa/b53/b53_common.c
@@ -573,9 +573,8 @@ EXPORT_SYMBOL(b53_disable_port);
 
 void b53_brcm_hdr_setup(struct dsa_switch *ds, int port)
 {
-	bool tag_en = !(ds->ops->get_tag_protocol(ds, port) ==
-			 DSA_TAG_PROTO_NONE);
 	struct b53_device *dev = ds->priv;
+	bool tag_en = !(dev->tag_protocol == DSA_TAG_PROTO_NONE);
 	u8 hdr_ctl, val;
 	u16 reg;
 
@@ -1876,7 +1875,8 @@ static bool b53_can_enable_brcm_tags(struct dsa_switch *ds, int port)
 	return ret;
 }
 
-enum dsa_tag_protocol b53_get_tag_protocol(struct dsa_switch *ds, int port)
+enum dsa_tag_protocol b53_get_tag_protocol(struct dsa_switch *ds, int port,
+					   enum dsa_tag_protocol mprot)
 {
 	struct b53_device *dev = ds->priv;
 
@@ -1886,16 +1886,22 @@ enum dsa_tag_protocol b53_get_tag_protocol(struct dsa_switch *ds, int port)
 	 * misses on multicast addresses (TBD).
 	 */
 	if (is5325(dev) || is5365(dev) || is539x(dev) || is531x5(dev) ||
-	    !b53_can_enable_brcm_tags(ds, port))
-		return DSA_TAG_PROTO_NONE;
+	    !b53_can_enable_brcm_tags(ds, port)) {
+		dev->tag_protocol = DSA_TAG_PROTO_NONE;
+		goto out;
+	}
 
 	/* Broadcom BCM58xx chips have a flow accelerator on Port 8
 	 * which requires us to use the prepended Broadcom tag type
 	 */
-	if (dev->chip_id == BCM58XX_DEVICE_ID && port == B53_CPU_PORT)
-		return DSA_TAG_PROTO_BRCM_PREPEND;
+	if (dev->chip_id == BCM58XX_DEVICE_ID && port == B53_CPU_PORT) {
+		dev->tag_protocol = DSA_TAG_PROTO_BRCM_PREPEND;
+		goto out;
+	}
 
-	return DSA_TAG_PROTO_BRCM;
+	dev->tag_protocol = DSA_TAG_PROTO_BRCM;
+out:
+	return dev->tag_protocol;
 }
 EXPORT_SYMBOL(b53_get_tag_protocol);
 
diff --git a/drivers/net/dsa/b53/b53_priv.h b/drivers/net/dsa/b53/b53_priv.h
index 1877acf05081..3c30f3a7eb29 100644
--- a/drivers/net/dsa/b53/b53_priv.h
+++ b/drivers/net/dsa/b53/b53_priv.h
@@ -118,6 +118,7 @@ struct b53_device {
 	u8 jumbo_size_reg;
 	int reset_gpio;
 	u8 num_arl_entries;
+	enum dsa_tag_protocol tag_protocol;
 
 	/* used ports mask */
 	u16 enabled_ports;
@@ -359,7 +360,8 @@ int b53_mdb_del(struct dsa_switch *ds, int port,
 		const struct switchdev_obj_port_mdb *mdb);
 int b53_mirror_add(struct dsa_switch *ds, int port,
 		   struct dsa_mall_mirror_tc_entry *mirror, bool ingress);
-enum dsa_tag_protocol b53_get_tag_protocol(struct dsa_switch *ds, int port);
+enum dsa_tag_protocol b53_get_tag_protocol(struct dsa_switch *ds, int port,
+					   enum dsa_tag_protocol mprot);
 void b53_mirror_del(struct dsa_switch *ds, int port,
 		    struct dsa_mall_mirror_tc_entry *mirror);
 int b53_enable_port(struct dsa_switch *ds, int port, struct phy_device *phy);
diff --git a/drivers/net/dsa/dsa_loop.c b/drivers/net/dsa/dsa_loop.c
index c8d7ef27fd72..fdcb70b9f0e4 100644
--- a/drivers/net/dsa/dsa_loop.c
+++ b/drivers/net/dsa/dsa_loop.c
@@ -61,7 +61,8 @@ struct dsa_loop_priv {
 static struct phy_device *phydevs[PHY_MAX_ADDR];
 
 static enum dsa_tag_protocol dsa_loop_get_protocol(struct dsa_switch *ds,
-						   int port)
+						   int port,
+						   enum dsa_tag_protocol mp)
 {
 	dev_dbg(ds->dev, "%s: port: %d\n", __func__, port);
 
diff --git a/drivers/net/dsa/lan9303-core.c b/drivers/net/dsa/lan9303-core.c
index e3c333a8f45d..cc17a44dd3a8 100644
--- a/drivers/net/dsa/lan9303-core.c
+++ b/drivers/net/dsa/lan9303-core.c
@@ -883,7 +883,8 @@ static int lan9303_check_device(struct lan9303 *chip)
 /* ---------------------------- DSA -----------------------------------*/
 
 static enum dsa_tag_protocol lan9303_get_tag_protocol(struct dsa_switch *ds,
-						      int port)
+						      int port,
+						      enum dsa_tag_protocol mp)
 {
 	return DSA_TAG_PROTO_LAN9303;
 }
diff --git a/drivers/net/dsa/lantiq_gswip.c b/drivers/net/dsa/lantiq_gswip.c
index 955324968b74..0369c22fe3e1 100644
--- a/drivers/net/dsa/lantiq_gswip.c
+++ b/drivers/net/dsa/lantiq_gswip.c
@@ -841,7 +841,8 @@ static int gswip_setup(struct dsa_switch *ds)
 }
 
 static enum dsa_tag_protocol gswip_get_tag_protocol(struct dsa_switch *ds,
-						    int port)
+						    int port,
+						    enum dsa_tag_protocol mp)
 {
 	return DSA_TAG_PROTO_GSWIP;
 }
diff --git a/drivers/net/dsa/microchip/ksz8795.c b/drivers/net/dsa/microchip/ksz8795.c
index 24a5e99f7fd5..47d65b77caf7 100644
--- a/drivers/net/dsa/microchip/ksz8795.c
+++ b/drivers/net/dsa/microchip/ksz8795.c
@@ -645,7 +645,8 @@ static void ksz8795_w_phy(struct ksz_device *dev, u16 phy, u16 reg, u16 val)
 }
 
 static enum dsa_tag_protocol ksz8795_get_tag_protocol(struct dsa_switch *ds,
-						      int port)
+						      int port,
+						      enum dsa_tag_protocol mp)
 {
 	return DSA_TAG_PROTO_KSZ8795;
 }
diff --git a/drivers/net/dsa/microchip/ksz9477.c b/drivers/net/dsa/microchip/ksz9477.c
index 50ffc63d6231..9a51b8a4de5d 100644
--- a/drivers/net/dsa/microchip/ksz9477.c
+++ b/drivers/net/dsa/microchip/ksz9477.c
@@ -295,7 +295,8 @@ static void ksz9477_port_init_cnt(struct ksz_device *dev, int port)
 }
 
 static enum dsa_tag_protocol ksz9477_get_tag_protocol(struct dsa_switch *ds,
-						      int port)
+						      int port,
+						      enum dsa_tag_protocol mp)
 {
 	enum dsa_tag_protocol proto = DSA_TAG_PROTO_KSZ9477;
 	struct ksz_device *dev = ds->priv;
diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c
index ed1ec10ec62b..022466ca1c19 100644
--- a/drivers/net/dsa/mt7530.c
+++ b/drivers/net/dsa/mt7530.c
@@ -1223,7 +1223,8 @@ mt7530_port_vlan_del(struct dsa_switch *ds, int port,
 }
 
 static enum dsa_tag_protocol
-mtk_get_tag_protocol(struct dsa_switch *ds, int port)
+mtk_get_tag_protocol(struct dsa_switch *ds, int port,
+		     enum dsa_tag_protocol mp)
 {
 	struct mt7530_priv *priv = ds->priv;
 
diff --git a/drivers/net/dsa/mv88e6060.c b/drivers/net/dsa/mv88e6060.c
index a5a37f47b320..24b8219fd607 100644
--- a/drivers/net/dsa/mv88e6060.c
+++ b/drivers/net/dsa/mv88e6060.c
@@ -43,7 +43,8 @@ static const char *mv88e6060_get_name(struct mii_bus *bus, int sw_addr)
 }
 
 static enum dsa_tag_protocol mv88e6060_get_tag_protocol(struct dsa_switch *ds,
-							int port)
+							int port,
+							enum dsa_tag_protocol m)
 {
 	return DSA_TAG_PROTO_TRAILER;
 }
diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c
index 99816ca9e5e4..04ef4d00f134 100644
--- a/drivers/net/dsa/mv88e6xxx/chip.c
+++ b/drivers/net/dsa/mv88e6xxx/chip.c
@@ -5217,7 +5217,8 @@ static struct mv88e6xxx_chip *mv88e6xxx_alloc_chip(struct device *dev)
 }
 
 static enum dsa_tag_protocol mv88e6xxx_get_tag_protocol(struct dsa_switch *ds,
-							int port)
+							int port,
+							enum dsa_tag_protocol m)
 {
 	struct mv88e6xxx_chip *chip = ds->priv;
 
diff --git a/drivers/net/dsa/ocelot/felix.c b/drivers/net/dsa/ocelot/felix.c
index f072dd75cea2..feccb6201660 100644
--- a/drivers/net/dsa/ocelot/felix.c
+++ b/drivers/net/dsa/ocelot/felix.c
@@ -16,7 +16,8 @@
 #include "felix.h"
 
 static enum dsa_tag_protocol felix_get_tag_protocol(struct dsa_switch *ds,
-						    int port)
+						    int port,
+						    enum dsa_tag_protocol mp)
 {
 	return DSA_TAG_PROTO_OCELOT;
 }
diff --git a/drivers/net/dsa/qca/ar9331.c b/drivers/net/dsa/qca/ar9331.c
index da3bece75e21..de25f99e995a 100644
--- a/drivers/net/dsa/qca/ar9331.c
+++ b/drivers/net/dsa/qca/ar9331.c
@@ -347,7 +347,8 @@ static void ar9331_sw_port_disable(struct dsa_switch *ds, int port)
 }
 
 static enum dsa_tag_protocol ar9331_sw_get_tag_protocol(struct dsa_switch *ds,
-							int port)
+							int port,
+							enum dsa_tag_protocol m)
 {
 	return DSA_TAG_PROTO_AR9331;
 }
diff --git a/drivers/net/dsa/qca8k.c b/drivers/net/dsa/qca8k.c
index e548289df31e..9f4205b4439b 100644
--- a/drivers/net/dsa/qca8k.c
+++ b/drivers/net/dsa/qca8k.c
@@ -1017,7 +1017,8 @@ qca8k_port_fdb_dump(struct dsa_switch *ds, int port,
 }
 
 static enum dsa_tag_protocol
-qca8k_get_tag_protocol(struct dsa_switch *ds, int port)
+qca8k_get_tag_protocol(struct dsa_switch *ds, int port,
+		       enum dsa_tag_protocol mp)
 {
 	return DSA_TAG_PROTO_QCA;
 }
diff --git a/drivers/net/dsa/rtl8366rb.c b/drivers/net/dsa/rtl8366rb.c
index f5cc8b0a7c74..fd1977590cb4 100644
--- a/drivers/net/dsa/rtl8366rb.c
+++ b/drivers/net/dsa/rtl8366rb.c
@@ -964,7 +964,8 @@ static int rtl8366rb_setup(struct dsa_switch *ds)
 }
 
 static enum dsa_tag_protocol rtl8366_get_tag_protocol(struct dsa_switch *ds,
-						      int port)
+						      int port,
+						      enum dsa_tag_protocol mp)
 {
 	/* For now, the RTL switches are handled without any custom tags.
 	 *
diff --git a/drivers/net/dsa/sja1105/sja1105_main.c b/drivers/net/dsa/sja1105/sja1105_main.c
index 61795833c8f5..784e6b8166a0 100644
--- a/drivers/net/dsa/sja1105/sja1105_main.c
+++ b/drivers/net/dsa/sja1105/sja1105_main.c
@@ -1534,7 +1534,8 @@ static int sja1105_setup_8021q_tagging(struct dsa_switch *ds, bool enabled)
 }
 
 static enum dsa_tag_protocol
-sja1105_get_tag_protocol(struct dsa_switch *ds, int port)
+sja1105_get_tag_protocol(struct dsa_switch *ds, int port,
+			 enum dsa_tag_protocol mp)
 {
 	return DSA_TAG_PROTO_SJA1105;
 }
diff --git a/drivers/net/dsa/vitesse-vsc73xx-core.c b/drivers/net/dsa/vitesse-vsc73xx-core.c
index 69fc0110ce04..6e21a2a5cf01 100644
--- a/drivers/net/dsa/vitesse-vsc73xx-core.c
+++ b/drivers/net/dsa/vitesse-vsc73xx-core.c
@@ -542,7 +542,8 @@ static int vsc73xx_phy_write(struct dsa_switch *ds, int phy, int regnum,
 }
 
 static enum dsa_tag_protocol vsc73xx_get_tag_protocol(struct dsa_switch *ds,
-						      int port)
+						      int port,
+						      enum dsa_tag_protocol mp)
 {
 	/* The switch internally uses a 8 byte header with length,
 	 * source port, tag, LPA and priority. This is supposedly
diff --git a/include/net/dsa.h b/include/net/dsa.h
index 0c39fed8cd99..63495e3443ac 100644
--- a/include/net/dsa.h
+++ b/include/net/dsa.h
@@ -380,7 +380,8 @@ typedef int dsa_fdb_dump_cb_t(const unsigned char *addr, u16 vid,
 			      bool is_static, void *data);
 struct dsa_switch_ops {
 	enum dsa_tag_protocol (*get_tag_protocol)(struct dsa_switch *ds,
-						  int port);
+						  int port,
+						  enum dsa_tag_protocol mprot);
 
 	int	(*setup)(struct dsa_switch *ds);
 	void	(*teardown)(struct dsa_switch *ds);
diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c
index c66abbed4daf..c6d81f2baf4e 100644
--- a/net/dsa/dsa2.c
+++ b/net/dsa/dsa2.c
@@ -614,6 +614,32 @@ static int dsa_port_parse_dsa(struct dsa_port *dp)
 	return 0;
 }
 
+static enum dsa_tag_protocol dsa_get_tag_protocol(struct dsa_port *dp,
+						  struct net_device *master)
+{
+	enum dsa_tag_protocol tag_protocol = DSA_TAG_PROTO_NONE;
+	struct dsa_switch *mds, *ds = dp->ds;
+	unsigned int mdp_upstream;
+	struct dsa_port *mdp;
+
+	/* It is possible to stack DSA switches onto one another when that
+	 * happens the switch driver may want to know if its tagging protocol
+	 * is going to work in such a configuration.
+	 */
+	if (dsa_slave_dev_check(master)) {
+		mdp = dsa_slave_to_port(master);
+		mds = mdp->ds;
+		mdp_upstream = dsa_upstream_port(mds, mdp->index);
+		tag_protocol = mds->ops->get_tag_protocol(mds, mdp_upstream,
+							  DSA_TAG_PROTO_NONE);
+	}
+
+	/* If the master device is not itself a DSA slave in a disjoint DSA
+	 * tree, then return immediately.
+	 */
+	return ds->ops->get_tag_protocol(ds, dp->index, tag_protocol);
+}
+
 static int dsa_port_parse_cpu(struct dsa_port *dp, struct net_device *master)
 {
 	struct dsa_switch *ds = dp->ds;
@@ -621,20 +647,21 @@ static int dsa_port_parse_cpu(struct dsa_port *dp, struct net_device *master)
 	const struct dsa_device_ops *tag_ops;
 	enum dsa_tag_protocol tag_protocol;
 
-	tag_protocol = ds->ops->get_tag_protocol(ds, dp->index);
+	tag_protocol = dsa_get_tag_protocol(dp, master);
 	tag_ops = dsa_tag_driver_get(tag_protocol);
 	if (IS_ERR(tag_ops)) {
 		if (PTR_ERR(tag_ops) == -ENOPROTOOPT)
 			return -EPROBE_DEFER;
 		dev_warn(ds->dev, "No tagger for this switch\n");
+		dp->master = NULL;
 		return PTR_ERR(tag_ops);
 	}
 
+	dp->master = master;
 	dp->type = DSA_PORT_TYPE_CPU;
 	dp->filter = tag_ops->filter;
 	dp->rcv = tag_ops->rcv;
 	dp->tag_ops = tag_ops;
-	dp->master = master;
 	dp->dst = dst;
 
 	return 0;
diff --git a/net/dsa/dsa_priv.h b/net/dsa/dsa_priv.h
index 8a162605b861..a7662e7a691d 100644
--- a/net/dsa/dsa_priv.h
+++ b/net/dsa/dsa_priv.h
@@ -157,6 +157,7 @@ extern const struct dsa_device_ops notag_netdev_ops;
 void dsa_slave_mii_bus_init(struct dsa_switch *ds);
 int dsa_slave_create(struct dsa_port *dp);
 void dsa_slave_destroy(struct net_device *slave_dev);
+bool dsa_slave_dev_check(const struct net_device *dev);
 int dsa_slave_suspend(struct net_device *slave_dev);
 int dsa_slave_resume(struct net_device *slave_dev);
 int dsa_slave_register_notifier(void);
diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index c1828bdc79dc..088c886e609e 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -22,8 +22,6 @@
 
 #include "dsa_priv.h"
 
-static bool dsa_slave_dev_check(const struct net_device *dev);
-
 /* slave mii_bus handling ***************************************************/
 static int dsa_slave_phy_read(struct mii_bus *bus, int addr, int reg)
 {
@@ -1473,7 +1471,7 @@ void dsa_slave_destroy(struct net_device *slave_dev)
 	free_netdev(slave_dev);
 }
 
-static bool dsa_slave_dev_check(const struct net_device *dev)
+bool dsa_slave_dev_check(const struct net_device *dev)
 {
 	return dev->netdev_ops == &dsa_slave_netdev_ops;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 8fab459e69abfd04a66d76423d18ba853fced4ab Mon Sep 17 00:00:00 2001
From: Florian Fainelli <f.fainelli@gmail.com>
Date: Tue, 7 Jan 2020 21:06:06 -0800
Subject: net: dsa: b53: Enable Broadcom tags for 531x5/539x families

The BCM531x5 and BCM539x families require that the IMP port be enabled
within the management page and that management mode (SM_SW_FWD_MODE) be
turned on. Once this is done, everything works as expected, including
multicast with standalone DSA devices or bridge devices.

Because such switches are frequencly cascaded with other internal
Broadcom switches on which we want to enable Broadcom tags, update
b53_can_enable_brcm_tags() to check the kind of DSA master tagging
protocol being used, if it is one of the two supported Broadcom tagging
protocols, force DSA_TAG_PROTO_NONE.

Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/b53/b53_common.c | 46 ++++++++++++++++++++++++++++++++--------
 1 file changed, 37 insertions(+), 9 deletions(-)

diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c
index 2b530a31ef0f..060497512159 100644
--- a/drivers/net/dsa/b53/b53_common.c
+++ b/drivers/net/dsa/b53/b53_common.c
@@ -371,8 +371,6 @@ static void b53_enable_vlan(struct b53_device *dev, bool enable,
 		b53_read8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL5, &vc5);
 	}
 
-	mgmt &= ~SM_SW_FWD_MODE;
-
 	if (enable) {
 		vc0 |= VC0_VLAN_EN | VC0_VID_CHK_EN | VC0_VID_HASH_VID;
 		vc1 |= VC1_RX_MCST_UNTAG_EN | VC1_RX_MCST_FWD_EN;
@@ -594,6 +592,22 @@ void b53_brcm_hdr_setup(struct dsa_switch *ds, int port)
 		break;
 	}
 
+	/* Enable management mode if tagging is requested */
+	b53_read8(dev, B53_CTRL_PAGE, B53_SWITCH_MODE, &hdr_ctl);
+	if (tag_en)
+		hdr_ctl |= SM_SW_FWD_MODE;
+	else
+		hdr_ctl &= ~SM_SW_FWD_MODE;
+	b53_write8(dev, B53_CTRL_PAGE, B53_SWITCH_MODE, hdr_ctl);
+
+	/* Configure the appropriate IMP port */
+	b53_read8(dev, B53_MGMT_PAGE, B53_GLOBAL_CONFIG, &hdr_ctl);
+	if (port == 8)
+		hdr_ctl |= GC_FRM_MGMT_PORT_MII;
+	else if (port == 5)
+		hdr_ctl |= GC_FRM_MGMT_PORT_M;
+	b53_write8(dev, B53_MGMT_PAGE, B53_GLOBAL_CONFIG, hdr_ctl);
+
 	/* Enable Broadcom tags for IMP port */
 	b53_read8(dev, B53_MGMT_PAGE, B53_BRCM_HDR, &hdr_ctl);
 	if (tag_en)
@@ -1865,13 +1879,29 @@ static bool b53_possible_cpu_port(struct dsa_switch *ds, int port)
 	return false;
 }
 
-static bool b53_can_enable_brcm_tags(struct dsa_switch *ds, int port)
+static bool b53_can_enable_brcm_tags(struct dsa_switch *ds, int port,
+				     enum dsa_tag_protocol tag_protocol)
 {
 	bool ret = b53_possible_cpu_port(ds, port);
 
-	if (!ret)
+	if (!ret) {
 		dev_warn(ds->dev, "Port %d is not Broadcom tag capable\n",
 			 port);
+		return ret;
+	}
+
+	switch (tag_protocol) {
+	case DSA_TAG_PROTO_BRCM:
+	case DSA_TAG_PROTO_BRCM_PREPEND:
+		dev_warn(ds->dev,
+			 "Port %d is stacked to Broadcom tag switch\n", port);
+		ret = false;
+		break;
+	default:
+		ret = true;
+		break;
+	}
+
 	return ret;
 }
 
@@ -1881,12 +1911,10 @@ enum dsa_tag_protocol b53_get_tag_protocol(struct dsa_switch *ds, int port,
 	struct b53_device *dev = ds->priv;
 
 	/* Older models (5325, 5365) support a different tag format that we do
-	 * not support in net/dsa/tag_brcm.c yet. 539x and 531x5 require managed
-	 * mode to be turned on which means we need to specifically manage ARL
-	 * misses on multicast addresses (TBD).
+	 * not support in net/dsa/tag_brcm.c yet.
 	 */
-	if (is5325(dev) || is5365(dev) || is539x(dev) || is531x5(dev) ||
-	    !b53_can_enable_brcm_tags(ds, port)) {
+	if (is5325(dev) || is5365(dev) ||
+	    !b53_can_enable_brcm_tags(ds, port, mprot)) {
 		dev->tag_protocol = DSA_TAG_PROTO_NONE;
 		goto out;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From a6dd04807ce4d7b71920d996e6b84fa3733a1bfd Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@oracle.com>
Date: Wed, 8 Jan 2020 08:39:48 +0300
Subject: ethtool: fix a memory leak in ethnl_default_start()

If ethnl_default_parse() fails then we need to free a couple
memory allocations before returning.

Fixes: 728480f12442 ("ethtool: default handlers for GET requests")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Michal Kubecek <mkubecek@suse.cz>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ethtool/netlink.c | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index 4ca96c7b86b3..5d16436498ac 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -472,8 +472,8 @@ static int ethnl_default_start(struct netlink_callback *cb)
 		return -ENOMEM;
 	reply_data = kmalloc(ops->reply_data_size, GFP_KERNEL);
 	if (!reply_data) {
-		kfree(req_info);
-		return -ENOMEM;
+		ret = -ENOMEM;
+		goto free_req_info;
 	}
 
 	ret = ethnl_default_parse(req_info, cb->nlh, sock_net(cb->skb->sk), ops,
@@ -487,7 +487,7 @@ static int ethnl_default_start(struct netlink_callback *cb)
 		req_info->dev = NULL;
 	}
 	if (ret < 0)
-		return ret;
+		goto free_reply_data;
 
 	ctx->ops = ops;
 	ctx->req_info = req_info;
@@ -496,6 +496,13 @@ static int ethnl_default_start(struct netlink_callback *cb)
 	ctx->pos_idx = 0;
 
 	return 0;
+
+free_reply_data:
+	kfree(reply_data);
+free_req_info:
+	kfree(req_info);
+
+	return ret;
 }
 
 /* default ->done() handler for GET requests */
-- 
cgit v1.2.3-59-g8ed1b


From d97772dbd7727655aa894f1610e983e73678212e Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@oracle.com>
Date: Wed, 8 Jan 2020 08:41:25 +0300
Subject: ethtool: fix ->reply_size() error handling

The "ret < 0" comparison is never true because "ret" is still zero.

Fixes: 728480f12442 ("ethtool: default handlers for GET requests")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Michal Kubecek <mkubecek@suse.cz>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ethtool/netlink.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index 5d16436498ac..86b79f9bc08d 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -319,9 +319,10 @@ static int ethnl_default_doit(struct sk_buff *skb, struct genl_info *info)
 	rtnl_unlock();
 	if (ret < 0)
 		goto err_cleanup;
-	reply_len = ops->reply_size(req_info, reply_data);
+	ret = ops->reply_size(req_info, reply_data);
 	if (ret < 0)
 		goto err_cleanup;
+	reply_len = ret;
 	ret = -ENOMEM;
 	rskb = ethnl_reply_init(reply_len, req_info->dev, ops->reply_cmd,
 				ops->hdr_attr, info, &reply_payload);
@@ -555,9 +556,10 @@ static void ethnl_default_notify(struct net_device *dev, unsigned int cmd,
 	ret = ops->prepare_data(req_info, reply_data, NULL);
 	if (ret < 0)
 		goto err_cleanup;
-	reply_len = ops->reply_size(req_info, reply_data);
+	ret = ops->reply_size(req_info, reply_data);
 	if (ret < 0)
 		goto err_cleanup;
+	reply_len = ret;
 	ret = -ENOMEM;
 	skb = genlmsg_new(reply_len, GFP_KERNEL);
 	if (!skb)
-- 
cgit v1.2.3-59-g8ed1b


From ac9c41d5a053e71777e8c2191681358499eff04b Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@oracle.com>
Date: Wed, 8 Jan 2020 08:42:36 +0300
Subject: ethtool: potential NULL dereference in strset_prepare_data()

Smatch complains that the NULL checking isn't done consistently:

    net/ethtool/strset.c:253 strset_prepare_data()
    error: we previously assumed 'dev' could be null (see line 233)

It looks like there is a missing return on this path.

Fixes: 71921690f974 ("ethtool: provide string sets with STRSET_GET request")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Michal Kubecek <mkubecek@suse.cz>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ethtool/strset.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/net/ethtool/strset.c b/net/ethtool/strset.c
index 9f2243329015..82a059c13c1c 100644
--- a/net/ethtool/strset.c
+++ b/net/ethtool/strset.c
@@ -239,6 +239,7 @@ static int strset_prepare_data(const struct ethnl_req_info *req_base,
 				return -EINVAL;
 			}
 		}
+		return 0;
 	}
 
 	ret = ethnl_ops_begin(dev);
-- 
cgit v1.2.3-59-g8ed1b


From 09e6533536c9f0e8b1dd8f4658a9274f83f95da4 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Wed, 8 Jan 2020 07:36:08 +0100
Subject: r8169: add constant EnAnaPLL

Use constant EnAnaPLL for bit 14 as in vendor driver. The vendor
driver sets this bit for chip version 02 only, but I'm not aware of
any issues, so better leave it as it is.
In addition remove the useless debug message.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 0161d839fa87..9c61ce2947c2 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -492,6 +492,7 @@ enum rtl_register_content {
 	/* CPlusCmd p.31 */
 	EnableBist	= (1 << 15),	// 8168 8101
 	Mac_dbgo_oe	= (1 << 14),	// 8168 8101
+	EnAnaPLL	= (1 << 14),	// 8169
 	Normal_mode	= (1 << 13),	// unused
 	Force_half_dup	= (1 << 12),	// 8168 8101
 	Force_rxflow_en	= (1 << 11),	// 8168 8101
@@ -5212,11 +5213,8 @@ static void rtl_hw_start_8169(struct rtl8169_private *tp)
 	tp->cp_cmd |= PCIMulRW;
 
 	if (tp->mac_version == RTL_GIGA_MAC_VER_02 ||
-	    tp->mac_version == RTL_GIGA_MAC_VER_03) {
-		netif_dbg(tp, drv, tp->dev,
-			  "Set MAC Reg C+CR Offset 0xe0. Bit 3 and Bit 14 MUST be 1\n");
-		tp->cp_cmd |= (1 << 14);
-	}
+	    tp->mac_version == RTL_GIGA_MAC_VER_03)
+		tp->cp_cmd |= EnAnaPLL;
 
 	RTL_W16(tp, CPlusCmd, tp->cp_cmd);
 
-- 
cgit v1.2.3-59-g8ed1b


From 112463ddbe82f8d1f98bfb7f943e1ed3c08bc56a Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Wed, 8 Jan 2020 13:48:38 +0100
Subject: net: dsa: felix: fix link error

When the enetc driver is disabled, the mdio support fails to
get built:

drivers/net/dsa/ocelot/felix_vsc9959.o: In function `vsc9959_mdio_bus_alloc':
felix_vsc9959.c:(.text+0x19c): undefined reference to `enetc_hw_alloc'
felix_vsc9959.c:(.text+0x1d1): undefined reference to `enetc_mdio_read'
felix_vsc9959.c:(.text+0x1d8): undefined reference to `enetc_mdio_write'

Change the Makefile to enter the subdirectory for this as well.

Fixes: bdeced75b13f ("net: dsa: felix: Add PCS operations for PHYLINK")
Fixes: 6517798dd343 ("enetc: Make MDIO accessors more generic and export to include/linux/fsl")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/freescale/Makefile | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/freescale/Makefile b/drivers/net/ethernet/freescale/Makefile
index 6a93293d31e0..67c436400352 100644
--- a/drivers/net/ethernet/freescale/Makefile
+++ b/drivers/net/ethernet/freescale/Makefile
@@ -25,4 +25,5 @@ obj-$(CONFIG_FSL_DPAA_ETH) += dpaa/
 obj-$(CONFIG_FSL_DPAA2_ETH) += dpaa2/
 
 obj-$(CONFIG_FSL_ENETC) += enetc/
+obj-$(CONFIG_FSL_ENETC_MDIO) += enetc/
 obj-$(CONFIG_FSL_ENETC_VF) += enetc/
-- 
cgit v1.2.3-59-g8ed1b


From ea217e874aef6aa523331a9b5a8d49223460cf3e Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Wed, 8 Jan 2020 14:15:15 +0100
Subject: netronome: fix ipv6 link error

When the driver is built-in but ipv6 is a module, the flower
support produces a link error:

drivers/net/ethernet/netronome/nfp/flower/tunnel_conf.o: In function `nfp_tunnel_keep_alive_v6':
tunnel_conf.c:(.text+0x2aa8): undefined reference to `nd_tbl'

Add a Kconfig dependency to avoid that configuration.

Fixes: 9ea9bfa12240 ("nfp: flower: support ipv6 tunnel keep-alive messages from fw")
Acked-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/netronome/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/netronome/Kconfig b/drivers/net/ethernet/netronome/Kconfig
index bac5be4d4f43..a3f68a718813 100644
--- a/drivers/net/ethernet/netronome/Kconfig
+++ b/drivers/net/ethernet/netronome/Kconfig
@@ -31,6 +31,7 @@ config NFP_APP_FLOWER
 	bool "NFP4000/NFP6000 TC Flower offload support"
 	depends on NFP
 	depends on NET_SWITCHDEV
+	depends on IPV6!=m || NFP=m
 	default y
 	---help---
 	  Enable driver support for TC Flower offload on NFP4000 and NFP6000.
-- 
cgit v1.2.3-59-g8ed1b


From 4a4a52d49d11f5c4a0df8b9806c8c5563801f753 Mon Sep 17 00:00:00 2001
From: yuehaibing <yuehaibing@huawei.com>
Date: Wed, 8 Jan 2020 22:08:22 +0800
Subject: vmxnet3: Remove always false conditional statement

param->rx_mini_pending is __u32 variable, it will never
be less than zero.

Signed-off-by: yuehaibing <yuehaibing@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/vmxnet3/vmxnet3_ethtool.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/net/vmxnet3/vmxnet3_ethtool.c b/drivers/net/vmxnet3/vmxnet3_ethtool.c
index 0a38c76688ab..1e4b9ba70983 100644
--- a/drivers/net/vmxnet3/vmxnet3_ethtool.c
+++ b/drivers/net/vmxnet3/vmxnet3_ethtool.c
@@ -555,10 +555,8 @@ vmxnet3_set_ringparam(struct net_device *netdev,
 	}
 
 	if (VMXNET3_VERSION_GE_3(adapter)) {
-		if (param->rx_mini_pending < 0 ||
-		    param->rx_mini_pending > VMXNET3_RXDATA_DESC_MAX_SIZE) {
+		if (param->rx_mini_pending > VMXNET3_RXDATA_DESC_MAX_SIZE)
 			return -EINVAL;
-		}
 	} else if (param->rx_mini_pending != 0) {
 		return -EINVAL;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From e43002242a47e8d7b2f9446f54de982f09e7dbd1 Mon Sep 17 00:00:00 2001
From: Stanislav Fomichev <sdf@google.com>
Date: Wed, 8 Jan 2020 11:21:32 -0800
Subject: selftests/bpf: Restore original comm in test_overhead

test_overhead changes task comm in order to estimate BPF trampoline
overhead but never sets the comm back to the original one.
We have the tests (like core_reloc.c) that have 'test_progs'
as hard-coded expected comm, so let's try to preserve the
original comm.

Currently, everything works because the order of execution is:
first core_recloc, then test_overhead; but let's make it a bit
future-proof.

Other related changes: use 'test_overhead' as new comm instead of
'test' to make it easy to debug and drop '\n' at the end.

Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Petar Penkov <ppenkov@google.com>
Acked-by: Song Liu <songliubraving@fb.com>
Link: https://lore.kernel.org/bpf/20200108192132.189221-1-sdf@google.com
---
 tools/testing/selftests/bpf/prog_tests/test_overhead.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/test_overhead.c b/tools/testing/selftests/bpf/prog_tests/test_overhead.c
index c32aa28bd93f..465b371a561d 100644
--- a/tools/testing/selftests/bpf/prog_tests/test_overhead.c
+++ b/tools/testing/selftests/bpf/prog_tests/test_overhead.c
@@ -2,6 +2,7 @@
 /* Copyright (c) 2019 Facebook */
 #define _GNU_SOURCE
 #include <sched.h>
+#include <sys/prctl.h>
 #include <test_progs.h>
 
 #define MAX_CNT 100000
@@ -17,7 +18,7 @@ static __u64 time_get_ns(void)
 static int test_task_rename(const char *prog)
 {
 	int i, fd, duration = 0, err;
-	char buf[] = "test\n";
+	char buf[] = "test_overhead";
 	__u64 start_time;
 
 	fd = open("/proc/self/comm", O_WRONLY|O_TRUNC);
@@ -66,6 +67,10 @@ void test_test_overhead(void)
 	struct bpf_object *obj;
 	struct bpf_link *link;
 	int err, duration = 0;
+	char comm[16] = {};
+
+	if (CHECK_FAIL(prctl(PR_GET_NAME, comm, 0L, 0L, 0L)))
+		return;
 
 	obj = bpf_object__open_file("./test_overhead.o", NULL);
 	if (CHECK(IS_ERR(obj), "obj_open_file", "err %ld\n", PTR_ERR(obj)))
@@ -138,5 +143,6 @@ void test_test_overhead(void)
 	test_run("fexit");
 	bpf_link__destroy(link);
 cleanup:
+	prctl(PR_SET_NAME, comm, 0L, 0L, 0L);
 	bpf_object__close(obj);
 }
-- 
cgit v1.2.3-59-g8ed1b


From 65726b5b7efae718391752ae8bb85b9843fd83f4 Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <kafai@fb.com>
Date: Wed, 8 Jan 2020 16:34:54 -0800
Subject: bpf: Save PTR_TO_BTF_ID register state when spilling to stack

This patch makes the verifier save the PTR_TO_BTF_ID register state when
spilling to the stack.

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20200109003454.3854870-1-kafai@fb.com
---
 kernel/bpf/verifier.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 6f63ae7a370c..d433d70022fd 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -1916,6 +1916,7 @@ static bool is_spillable_regtype(enum bpf_reg_type type)
 	case PTR_TO_TCP_SOCK:
 	case PTR_TO_TCP_SOCK_OR_NULL:
 	case PTR_TO_XDP_SOCK:
+	case PTR_TO_BTF_ID:
 		return true;
 	default:
 		return false;
-- 
cgit v1.2.3-59-g8ed1b


From 275517ff452a535da5eef25b1c22e53fc50b0a12 Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <kafai@fb.com>
Date: Wed, 8 Jan 2020 16:34:56 -0800
Subject: bpf: Avoid storing modifier to info->btf_id

info->btf_id expects the btf_id of a struct, so it should
store the final result after skipping modifiers (if any).

It also takes this chanace to add a missing newline in one of the
bpf_log() messages.

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20200109003456.3855176-1-kafai@fb.com
---
 kernel/bpf/btf.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index ed2075884724..497ecf62d79d 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -3697,7 +3697,6 @@ bool btf_ctx_access(int off, int size, enum bpf_access_type type,
 
 	/* this is a pointer to another type */
 	info->reg_type = PTR_TO_BTF_ID;
-	info->btf_id = t->type;
 
 	if (tgt_prog) {
 		ret = btf_translate_to_vmlinux(log, btf, t, tgt_prog->type);
@@ -3708,10 +3707,14 @@ bool btf_ctx_access(int off, int size, enum bpf_access_type type,
 			return false;
 		}
 	}
+
+	info->btf_id = t->type;
 	t = btf_type_by_id(btf, t->type);
 	/* skip modifiers */
-	while (btf_type_is_modifier(t))
+	while (btf_type_is_modifier(t)) {
+		info->btf_id = t->type;
 		t = btf_type_by_id(btf, t->type);
+	}
 	if (!btf_type_is_struct(t)) {
 		bpf_log(log,
 			"func '%s' arg%d type %s is not a struct\n",
@@ -3737,7 +3740,7 @@ int btf_struct_access(struct bpf_verifier_log *log,
 again:
 	tname = __btf_name_by_offset(btf_vmlinux, t->name_off);
 	if (!btf_type_is_struct(t)) {
-		bpf_log(log, "Type '%s' is not a struct", tname);
+		bpf_log(log, "Type '%s' is not a struct\n", tname);
 		return -EINVAL;
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From 218b3f65f9081f5e1bffe6de5f0f4b22c935410b Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <kafai@fb.com>
Date: Wed, 8 Jan 2020 16:34:59 -0800
Subject: bpf: Add enum support to btf_ctx_access()

It allows bpf prog (e.g. tracing) to attach
to a kernel function that takes enum argument.

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20200109003459.3855366-1-kafai@fb.com
---
 kernel/bpf/btf.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 497ecf62d79d..6a5ccb748a72 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -3677,7 +3677,7 @@ bool btf_ctx_access(int off, int size, enum bpf_access_type type,
 	/* skip modifiers */
 	while (btf_type_is_modifier(t))
 		t = btf_type_by_id(btf, t->type);
-	if (btf_type_is_int(t))
+	if (btf_type_is_int(t) || btf_type_is_enum(t))
 		/* accessing a scalar */
 		return true;
 	if (!btf_type_is_ptr(t)) {
-- 
cgit v1.2.3-59-g8ed1b


From 976aba002fcb4b1baa71e4b0854f3d4ae48c1d4d Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <kafai@fb.com>
Date: Wed, 8 Jan 2020 16:35:01 -0800
Subject: bpf: Support bitfield read access in btf_struct_access

This patch allows bitfield access as a scalar.

It checks "off + size > t->size" to avoid accessing bitfield
end up accessing beyond the struct.  This check is done
outside of the loop since it is applicable to all access.

It also takes this chance to break early on the "off < moff" case.

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20200109003501.3855427-1-kafai@fb.com
---
 kernel/bpf/btf.c | 44 +++++++++++++++++++++++++++++++++++++++-----
 1 file changed, 39 insertions(+), 5 deletions(-)

diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 6a5ccb748a72..48bbde2e1c1e 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -3744,19 +3744,53 @@ again:
 		return -EINVAL;
 	}
 
-	for_each_member(i, t, member) {
-		if (btf_member_bitfield_size(t, member))
-			/* bitfields are not supported yet */
-			continue;
+	if (off + size > t->size) {
+		bpf_log(log, "access beyond struct %s at off %u size %u\n",
+			tname, off, size);
+		return -EACCES;
+	}
 
+	for_each_member(i, t, member) {
 		/* offset of the field in bytes */
 		moff = btf_member_bit_offset(t, member) / 8;
 		if (off + size <= moff)
 			/* won't find anything, field is already too far */
 			break;
+
+		if (btf_member_bitfield_size(t, member)) {
+			u32 end_bit = btf_member_bit_offset(t, member) +
+				btf_member_bitfield_size(t, member);
+
+			/* off <= moff instead of off == moff because clang
+			 * does not generate a BTF member for anonymous
+			 * bitfield like the ":16" here:
+			 * struct {
+			 *	int :16;
+			 *	int x:8;
+			 * };
+			 */
+			if (off <= moff &&
+			    BITS_ROUNDUP_BYTES(end_bit) <= off + size)
+				return SCALAR_VALUE;
+
+			/* off may be accessing a following member
+			 *
+			 * or
+			 *
+			 * Doing partial access at either end of this
+			 * bitfield.  Continue on this case also to
+			 * treat it as not accessing this bitfield
+			 * and eventually error out as field not
+			 * found to keep it simple.
+			 * It could be relaxed if there was a legit
+			 * partial access case later.
+			 */
+			continue;
+		}
+
 		/* In case of "off" is pointing to holes of a struct */
 		if (off < moff)
-			continue;
+			break;
 
 		/* type of the field */
 		mtype = btf_type_by_id(btf_vmlinux, member->type);
-- 
cgit v1.2.3-59-g8ed1b


From 27ae7997a66174cb8afd6a75b3989f5e0c1b9e5a Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <kafai@fb.com>
Date: Wed, 8 Jan 2020 16:35:03 -0800
Subject: bpf: Introduce BPF_PROG_TYPE_STRUCT_OPS

This patch allows the kernel's struct ops (i.e. func ptr) to be
implemented in BPF.  The first use case in this series is the
"struct tcp_congestion_ops" which will be introduced in a
latter patch.

This patch introduces a new prog type BPF_PROG_TYPE_STRUCT_OPS.
The BPF_PROG_TYPE_STRUCT_OPS prog is verified against a particular
func ptr of a kernel struct.  The attr->attach_btf_id is the btf id
of a kernel struct.  The attr->expected_attach_type is the member
"index" of that kernel struct.  The first member of a struct starts
with member index 0.  That will avoid ambiguity when a kernel struct
has multiple func ptrs with the same func signature.

For example, a BPF_PROG_TYPE_STRUCT_OPS prog is written
to implement the "init" func ptr of the "struct tcp_congestion_ops".
The attr->attach_btf_id is the btf id of the "struct tcp_congestion_ops"
of the _running_ kernel.  The attr->expected_attach_type is 3.

The ctx of BPF_PROG_TYPE_STRUCT_OPS is an array of u64 args saved
by arch_prepare_bpf_trampoline that will be done in the next
patch when introducing BPF_MAP_TYPE_STRUCT_OPS.

"struct bpf_struct_ops" is introduced as a common interface for the kernel
struct that supports BPF_PROG_TYPE_STRUCT_OPS prog.  The supporting kernel
struct will need to implement an instance of the "struct bpf_struct_ops".

The supporting kernel struct also needs to implement a bpf_verifier_ops.
During BPF_PROG_LOAD, bpf_struct_ops_find() will find the right
bpf_verifier_ops by searching the attr->attach_btf_id.

A new "btf_struct_access" is also added to the bpf_verifier_ops such
that the supporting kernel struct can optionally provide its own specific
check on accessing the func arg (e.g. provide limited write access).

After btf_vmlinux is parsed, the new bpf_struct_ops_init() is called
to initialize some values (e.g. the btf id of the supporting kernel
struct) and it can only be done once the btf_vmlinux is available.

The R0 checks at BPF_EXIT is excluded for the BPF_PROG_TYPE_STRUCT_OPS prog
if the return type of the prog->aux->attach_func_proto is "void".

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20200109003503.3855825-1-kafai@fb.com
---
 include/linux/bpf.h               |  30 +++++++++
 include/linux/bpf_types.h         |   4 ++
 include/linux/btf.h               |  34 ++++++++++
 include/uapi/linux/bpf.h          |   1 +
 kernel/bpf/Makefile               |   3 +
 kernel/bpf/bpf_struct_ops.c       | 121 ++++++++++++++++++++++++++++++++++
 kernel/bpf/bpf_struct_ops_types.h |   4 ++
 kernel/bpf/btf.c                  |  88 +++++++++++++++++--------
 kernel/bpf/syscall.c              |  17 +++--
 kernel/bpf/verifier.c             | 134 +++++++++++++++++++++++++++++---------
 10 files changed, 373 insertions(+), 63 deletions(-)
 create mode 100644 kernel/bpf/bpf_struct_ops.c
 create mode 100644 kernel/bpf/bpf_struct_ops_types.h

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index b14e51d56a82..50f3b20ae284 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -349,6 +349,10 @@ struct bpf_verifier_ops {
 				  const struct bpf_insn *src,
 				  struct bpf_insn *dst,
 				  struct bpf_prog *prog, u32 *target_size);
+	int (*btf_struct_access)(struct bpf_verifier_log *log,
+				 const struct btf_type *t, int off, int size,
+				 enum bpf_access_type atype,
+				 u32 *next_btf_id);
 };
 
 struct bpf_prog_offload_ops {
@@ -668,6 +672,32 @@ struct bpf_array_aux {
 	struct work_struct work;
 };
 
+struct btf_type;
+struct btf_member;
+
+#define BPF_STRUCT_OPS_MAX_NR_MEMBERS 64
+struct bpf_struct_ops {
+	const struct bpf_verifier_ops *verifier_ops;
+	int (*init)(struct btf *btf);
+	int (*check_member)(const struct btf_type *t,
+			    const struct btf_member *member);
+	const struct btf_type *type;
+	const char *name;
+	struct btf_func_model func_models[BPF_STRUCT_OPS_MAX_NR_MEMBERS];
+	u32 type_id;
+};
+
+#if defined(CONFIG_BPF_JIT) && defined(CONFIG_BPF_SYSCALL)
+const struct bpf_struct_ops *bpf_struct_ops_find(u32 type_id);
+void bpf_struct_ops_init(struct btf *btf);
+#else
+static inline const struct bpf_struct_ops *bpf_struct_ops_find(u32 type_id)
+{
+	return NULL;
+}
+static inline void bpf_struct_ops_init(struct btf *btf) { }
+#endif
+
 struct bpf_array {
 	struct bpf_map map;
 	u32 elem_size;
diff --git a/include/linux/bpf_types.h b/include/linux/bpf_types.h
index 93740b3614d7..fadd243ffa2d 100644
--- a/include/linux/bpf_types.h
+++ b/include/linux/bpf_types.h
@@ -65,6 +65,10 @@ BPF_PROG_TYPE(BPF_PROG_TYPE_LIRC_MODE2, lirc_mode2,
 BPF_PROG_TYPE(BPF_PROG_TYPE_SK_REUSEPORT, sk_reuseport,
 	      struct sk_reuseport_md, struct sk_reuseport_kern)
 #endif
+#if defined(CONFIG_BPF_JIT)
+BPF_PROG_TYPE(BPF_PROG_TYPE_STRUCT_OPS, bpf_struct_ops,
+	      void *, void *)
+#endif
 
 BPF_MAP_TYPE(BPF_MAP_TYPE_ARRAY, array_map_ops)
 BPF_MAP_TYPE(BPF_MAP_TYPE_PERCPU_ARRAY, percpu_array_map_ops)
diff --git a/include/linux/btf.h b/include/linux/btf.h
index 79d4abc2556a..f74a09a7120b 100644
--- a/include/linux/btf.h
+++ b/include/linux/btf.h
@@ -53,6 +53,18 @@ bool btf_member_is_reg_int(const struct btf *btf, const struct btf_type *s,
 			   u32 expected_offset, u32 expected_size);
 int btf_find_spin_lock(const struct btf *btf, const struct btf_type *t);
 bool btf_type_is_void(const struct btf_type *t);
+s32 btf_find_by_name_kind(const struct btf *btf, const char *name, u8 kind);
+const struct btf_type *btf_type_skip_modifiers(const struct btf *btf,
+					       u32 id, u32 *res_id);
+const struct btf_type *btf_type_resolve_ptr(const struct btf *btf,
+					    u32 id, u32 *res_id);
+const struct btf_type *btf_type_resolve_func_ptr(const struct btf *btf,
+						 u32 id, u32 *res_id);
+
+#define for_each_member(i, struct_type, member)			\
+	for (i = 0, member = btf_type_member(struct_type);	\
+	     i < btf_type_vlen(struct_type);			\
+	     i++, member++)
 
 static inline bool btf_type_is_ptr(const struct btf_type *t)
 {
@@ -84,6 +96,28 @@ static inline bool btf_type_is_func_proto(const struct btf_type *t)
 	return BTF_INFO_KIND(t->info) == BTF_KIND_FUNC_PROTO;
 }
 
+static inline u16 btf_type_vlen(const struct btf_type *t)
+{
+	return BTF_INFO_VLEN(t->info);
+}
+
+static inline bool btf_type_kflag(const struct btf_type *t)
+{
+	return BTF_INFO_KFLAG(t->info);
+}
+
+static inline u32 btf_member_bitfield_size(const struct btf_type *struct_type,
+					   const struct btf_member *member)
+{
+	return btf_type_kflag(struct_type) ? BTF_MEMBER_BITFIELD_SIZE(member->offset)
+					   : 0;
+}
+
+static inline const struct btf_member *btf_type_member(const struct btf_type *t)
+{
+	return (const struct btf_member *)(t + 1);
+}
+
 #ifdef CONFIG_BPF_SYSCALL
 const struct btf_type *btf_type_by_id(const struct btf *btf, u32 type_id);
 const char *btf_name_by_offset(const struct btf *btf, u32 offset);
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 7df436da542d..c1eeb3e0e116 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -174,6 +174,7 @@ enum bpf_prog_type {
 	BPF_PROG_TYPE_RAW_TRACEPOINT_WRITABLE,
 	BPF_PROG_TYPE_CGROUP_SOCKOPT,
 	BPF_PROG_TYPE_TRACING,
+	BPF_PROG_TYPE_STRUCT_OPS,
 };
 
 enum bpf_attach_type {
diff --git a/kernel/bpf/Makefile b/kernel/bpf/Makefile
index d4f330351f87..046ce5d98033 100644
--- a/kernel/bpf/Makefile
+++ b/kernel/bpf/Makefile
@@ -27,3 +27,6 @@ endif
 ifeq ($(CONFIG_SYSFS),y)
 obj-$(CONFIG_DEBUG_INFO_BTF) += sysfs_btf.o
 endif
+ifeq ($(CONFIG_BPF_JIT),y)
+obj-$(CONFIG_BPF_SYSCALL) += bpf_struct_ops.o
+endif
diff --git a/kernel/bpf/bpf_struct_ops.c b/kernel/bpf/bpf_struct_ops.c
new file mode 100644
index 000000000000..2ea68fe34c33
--- /dev/null
+++ b/kernel/bpf/bpf_struct_ops.c
@@ -0,0 +1,121 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright (c) 2019 Facebook */
+
+#include <linux/bpf.h>
+#include <linux/bpf_verifier.h>
+#include <linux/btf.h>
+#include <linux/filter.h>
+#include <linux/slab.h>
+#include <linux/numa.h>
+#include <linux/seq_file.h>
+#include <linux/refcount.h>
+
+#define BPF_STRUCT_OPS_TYPE(_name)				\
+extern struct bpf_struct_ops bpf_##_name;
+#include "bpf_struct_ops_types.h"
+#undef BPF_STRUCT_OPS_TYPE
+
+enum {
+#define BPF_STRUCT_OPS_TYPE(_name) BPF_STRUCT_OPS_TYPE_##_name,
+#include "bpf_struct_ops_types.h"
+#undef BPF_STRUCT_OPS_TYPE
+	__NR_BPF_STRUCT_OPS_TYPE,
+};
+
+static struct bpf_struct_ops * const bpf_struct_ops[] = {
+#define BPF_STRUCT_OPS_TYPE(_name)				\
+	[BPF_STRUCT_OPS_TYPE_##_name] = &bpf_##_name,
+#include "bpf_struct_ops_types.h"
+#undef BPF_STRUCT_OPS_TYPE
+};
+
+const struct bpf_verifier_ops bpf_struct_ops_verifier_ops = {
+};
+
+const struct bpf_prog_ops bpf_struct_ops_prog_ops = {
+};
+
+void bpf_struct_ops_init(struct btf *btf)
+{
+	const struct btf_member *member;
+	struct bpf_struct_ops *st_ops;
+	struct bpf_verifier_log log = {};
+	const struct btf_type *t;
+	const char *mname;
+	s32 type_id;
+	u32 i, j;
+
+	for (i = 0; i < ARRAY_SIZE(bpf_struct_ops); i++) {
+		st_ops = bpf_struct_ops[i];
+
+		type_id = btf_find_by_name_kind(btf, st_ops->name,
+						BTF_KIND_STRUCT);
+		if (type_id < 0) {
+			pr_warn("Cannot find struct %s in btf_vmlinux\n",
+				st_ops->name);
+			continue;
+		}
+		t = btf_type_by_id(btf, type_id);
+		if (btf_type_vlen(t) > BPF_STRUCT_OPS_MAX_NR_MEMBERS) {
+			pr_warn("Cannot support #%u members in struct %s\n",
+				btf_type_vlen(t), st_ops->name);
+			continue;
+		}
+
+		for_each_member(j, t, member) {
+			const struct btf_type *func_proto;
+
+			mname = btf_name_by_offset(btf, member->name_off);
+			if (!*mname) {
+				pr_warn("anon member in struct %s is not supported\n",
+					st_ops->name);
+				break;
+			}
+
+			if (btf_member_bitfield_size(t, member)) {
+				pr_warn("bit field member %s in struct %s is not supported\n",
+					mname, st_ops->name);
+				break;
+			}
+
+			func_proto = btf_type_resolve_func_ptr(btf,
+							       member->type,
+							       NULL);
+			if (func_proto &&
+			    btf_distill_func_proto(&log, btf,
+						   func_proto, mname,
+						   &st_ops->func_models[j])) {
+				pr_warn("Error in parsing func ptr %s in struct %s\n",
+					mname, st_ops->name);
+				break;
+			}
+		}
+
+		if (j == btf_type_vlen(t)) {
+			if (st_ops->init(btf)) {
+				pr_warn("Error in init bpf_struct_ops %s\n",
+					st_ops->name);
+			} else {
+				st_ops->type_id = type_id;
+				st_ops->type = t;
+			}
+		}
+	}
+}
+
+extern struct btf *btf_vmlinux;
+
+const struct bpf_struct_ops *bpf_struct_ops_find(u32 type_id)
+{
+	unsigned int i;
+
+	if (!type_id || !btf_vmlinux)
+		return NULL;
+
+	for (i = 0; i < ARRAY_SIZE(bpf_struct_ops); i++) {
+		if (bpf_struct_ops[i]->type_id == type_id)
+			return bpf_struct_ops[i];
+	}
+
+	return NULL;
+}
diff --git a/kernel/bpf/bpf_struct_ops_types.h b/kernel/bpf/bpf_struct_ops_types.h
new file mode 100644
index 000000000000..7bb13ff49ec2
--- /dev/null
+++ b/kernel/bpf/bpf_struct_ops_types.h
@@ -0,0 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* internal file - do not include directly */
+
+/* To be filled in a later patch */
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 48bbde2e1c1e..12af4a1bb1a4 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -180,11 +180,6 @@
  */
 #define BTF_MAX_SIZE (16 * 1024 * 1024)
 
-#define for_each_member(i, struct_type, member)			\
-	for (i = 0, member = btf_type_member(struct_type);	\
-	     i < btf_type_vlen(struct_type);			\
-	     i++, member++)
-
 #define for_each_member_from(i, from, struct_type, member)		\
 	for (i = from, member = btf_type_member(struct_type) + from;	\
 	     i < btf_type_vlen(struct_type);				\
@@ -382,6 +377,65 @@ static bool btf_type_is_datasec(const struct btf_type *t)
 	return BTF_INFO_KIND(t->info) == BTF_KIND_DATASEC;
 }
 
+s32 btf_find_by_name_kind(const struct btf *btf, const char *name, u8 kind)
+{
+	const struct btf_type *t;
+	const char *tname;
+	u32 i;
+
+	for (i = 1; i <= btf->nr_types; i++) {
+		t = btf->types[i];
+		if (BTF_INFO_KIND(t->info) != kind)
+			continue;
+
+		tname = btf_name_by_offset(btf, t->name_off);
+		if (!strcmp(tname, name))
+			return i;
+	}
+
+	return -ENOENT;
+}
+
+const struct btf_type *btf_type_skip_modifiers(const struct btf *btf,
+					       u32 id, u32 *res_id)
+{
+	const struct btf_type *t = btf_type_by_id(btf, id);
+
+	while (btf_type_is_modifier(t)) {
+		id = t->type;
+		t = btf_type_by_id(btf, t->type);
+	}
+
+	if (res_id)
+		*res_id = id;
+
+	return t;
+}
+
+const struct btf_type *btf_type_resolve_ptr(const struct btf *btf,
+					    u32 id, u32 *res_id)
+{
+	const struct btf_type *t;
+
+	t = btf_type_skip_modifiers(btf, id, NULL);
+	if (!btf_type_is_ptr(t))
+		return NULL;
+
+	return btf_type_skip_modifiers(btf, t->type, res_id);
+}
+
+const struct btf_type *btf_type_resolve_func_ptr(const struct btf *btf,
+						 u32 id, u32 *res_id)
+{
+	const struct btf_type *ptype;
+
+	ptype = btf_type_resolve_ptr(btf, id, res_id);
+	if (ptype && btf_type_is_func_proto(ptype))
+		return ptype;
+
+	return NULL;
+}
+
 /* Types that act only as a source, not sink or intermediate
  * type when resolving.
  */
@@ -446,16 +500,6 @@ static const char *btf_int_encoding_str(u8 encoding)
 		return "UNKN";
 }
 
-static u16 btf_type_vlen(const struct btf_type *t)
-{
-	return BTF_INFO_VLEN(t->info);
-}
-
-static bool btf_type_kflag(const struct btf_type *t)
-{
-	return BTF_INFO_KFLAG(t->info);
-}
-
 static u32 btf_member_bit_offset(const struct btf_type *struct_type,
 			     const struct btf_member *member)
 {
@@ -463,13 +507,6 @@ static u32 btf_member_bit_offset(const struct btf_type *struct_type,
 					   : member->offset;
 }
 
-static u32 btf_member_bitfield_size(const struct btf_type *struct_type,
-				    const struct btf_member *member)
-{
-	return btf_type_kflag(struct_type) ? BTF_MEMBER_BITFIELD_SIZE(member->offset)
-					   : 0;
-}
-
 static u32 btf_type_int(const struct btf_type *t)
 {
 	return *(u32 *)(t + 1);
@@ -480,11 +517,6 @@ static const struct btf_array *btf_type_array(const struct btf_type *t)
 	return (const struct btf_array *)(t + 1);
 }
 
-static const struct btf_member *btf_type_member(const struct btf_type *t)
-{
-	return (const struct btf_member *)(t + 1);
-}
-
 static const struct btf_enum *btf_type_enum(const struct btf_type *t)
 {
 	return (const struct btf_enum *)(t + 1);
@@ -3605,6 +3637,8 @@ struct btf *btf_parse_vmlinux(void)
 		goto errout;
 	}
 
+	bpf_struct_ops_init(btf);
+
 	btf_verifier_env_free(env);
 	refcount_set(&btf->refcnt, 1);
 	return btf;
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 81ee8595dfee..03a02ef4c496 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -1672,17 +1672,22 @@ bpf_prog_load_check_attach(enum bpf_prog_type prog_type,
 			   enum bpf_attach_type expected_attach_type,
 			   u32 btf_id, u32 prog_fd)
 {
-	switch (prog_type) {
-	case BPF_PROG_TYPE_TRACING:
+	if (btf_id) {
 		if (btf_id > BTF_MAX_TYPE)
 			return -EINVAL;
-		break;
-	default:
-		if (btf_id || prog_fd)
+
+		switch (prog_type) {
+		case BPF_PROG_TYPE_TRACING:
+		case BPF_PROG_TYPE_STRUCT_OPS:
+			break;
+		default:
 			return -EINVAL;
-		break;
+		}
 	}
 
+	if (prog_fd && prog_type != BPF_PROG_TYPE_TRACING)
+		return -EINVAL;
+
 	switch (prog_type) {
 	case BPF_PROG_TYPE_CGROUP_SOCK:
 		switch (expected_attach_type) {
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index d433d70022fd..586ed3c94b80 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -2859,11 +2859,6 @@ static int check_ptr_to_btf_access(struct bpf_verifier_env *env,
 	u32 btf_id;
 	int ret;
 
-	if (atype != BPF_READ) {
-		verbose(env, "only read is supported\n");
-		return -EACCES;
-	}
-
 	if (off < 0) {
 		verbose(env,
 			"R%d is ptr_%s invalid negative access: off=%d\n",
@@ -2880,17 +2875,32 @@ static int check_ptr_to_btf_access(struct bpf_verifier_env *env,
 		return -EACCES;
 	}
 
-	ret = btf_struct_access(&env->log, t, off, size, atype, &btf_id);
+	if (env->ops->btf_struct_access) {
+		ret = env->ops->btf_struct_access(&env->log, t, off, size,
+						  atype, &btf_id);
+	} else {
+		if (atype != BPF_READ) {
+			verbose(env, "only read is supported\n");
+			return -EACCES;
+		}
+
+		ret = btf_struct_access(&env->log, t, off, size, atype,
+					&btf_id);
+	}
+
 	if (ret < 0)
 		return ret;
 
-	if (ret == SCALAR_VALUE) {
-		mark_reg_unknown(env, regs, value_regno);
-		return 0;
+	if (atype == BPF_READ) {
+		if (ret == SCALAR_VALUE) {
+			mark_reg_unknown(env, regs, value_regno);
+			return 0;
+		}
+		mark_reg_known_zero(env, regs, value_regno);
+		regs[value_regno].type = PTR_TO_BTF_ID;
+		regs[value_regno].btf_id = btf_id;
 	}
-	mark_reg_known_zero(env, regs, value_regno);
-	regs[value_regno].type = PTR_TO_BTF_ID;
-	regs[value_regno].btf_id = btf_id;
+
 	return 0;
 }
 
@@ -6349,8 +6359,30 @@ static int check_ld_abs(struct bpf_verifier_env *env, struct bpf_insn *insn)
 static int check_return_code(struct bpf_verifier_env *env)
 {
 	struct tnum enforce_attach_type_range = tnum_unknown;
+	const struct bpf_prog *prog = env->prog;
 	struct bpf_reg_state *reg;
 	struct tnum range = tnum_range(0, 1);
+	int err;
+
+	/* The struct_ops func-ptr's return type could be "void" */
+	if (env->prog->type == BPF_PROG_TYPE_STRUCT_OPS &&
+	    !prog->aux->attach_func_proto->type)
+		return 0;
+
+	/* eBPF calling convetion is such that R0 is used
+	 * to return the value from eBPF program.
+	 * Make sure that it's readable at this time
+	 * of bpf_exit, which means that program wrote
+	 * something into it earlier
+	 */
+	err = check_reg_arg(env, BPF_REG_0, SRC_OP);
+	if (err)
+		return err;
+
+	if (is_pointer_value(env, BPF_REG_0)) {
+		verbose(env, "R0 leaks addr as return value\n");
+		return -EACCES;
+	}
 
 	switch (env->prog->type) {
 	case BPF_PROG_TYPE_CGROUP_SOCK_ADDR:
@@ -8016,21 +8048,6 @@ static int do_check(struct bpf_verifier_env *env)
 				if (err)
 					return err;
 
-				/* eBPF calling convetion is such that R0 is used
-				 * to return the value from eBPF program.
-				 * Make sure that it's readable at this time
-				 * of bpf_exit, which means that program wrote
-				 * something into it earlier
-				 */
-				err = check_reg_arg(env, BPF_REG_0, SRC_OP);
-				if (err)
-					return err;
-
-				if (is_pointer_value(env, BPF_REG_0)) {
-					verbose(env, "R0 leaks addr as return value\n");
-					return -EACCES;
-				}
-
 				err = check_return_code(env);
 				if (err)
 					return err;
@@ -8829,12 +8846,14 @@ static int convert_ctx_accesses(struct bpf_verifier_env *env)
 			convert_ctx_access = bpf_xdp_sock_convert_ctx_access;
 			break;
 		case PTR_TO_BTF_ID:
-			if (type == BPF_WRITE) {
+			if (type == BPF_READ) {
+				insn->code = BPF_LDX | BPF_PROBE_MEM |
+					BPF_SIZE((insn)->code);
+				env->prog->aux->num_exentries++;
+			} else if (env->prog->type != BPF_PROG_TYPE_STRUCT_OPS) {
 				verbose(env, "Writes through BTF pointers are not allowed\n");
 				return -EINVAL;
 			}
-			insn->code = BPF_LDX | BPF_PROBE_MEM | BPF_SIZE((insn)->code);
-			env->prog->aux->num_exentries++;
 			continue;
 		default:
 			continue;
@@ -9502,6 +9521,58 @@ static void print_verification_stats(struct bpf_verifier_env *env)
 		env->peak_states, env->longest_mark_read_walk);
 }
 
+static int check_struct_ops_btf_id(struct bpf_verifier_env *env)
+{
+	const struct btf_type *t, *func_proto;
+	const struct bpf_struct_ops *st_ops;
+	const struct btf_member *member;
+	struct bpf_prog *prog = env->prog;
+	u32 btf_id, member_idx;
+	const char *mname;
+
+	btf_id = prog->aux->attach_btf_id;
+	st_ops = bpf_struct_ops_find(btf_id);
+	if (!st_ops) {
+		verbose(env, "attach_btf_id %u is not a supported struct\n",
+			btf_id);
+		return -ENOTSUPP;
+	}
+
+	t = st_ops->type;
+	member_idx = prog->expected_attach_type;
+	if (member_idx >= btf_type_vlen(t)) {
+		verbose(env, "attach to invalid member idx %u of struct %s\n",
+			member_idx, st_ops->name);
+		return -EINVAL;
+	}
+
+	member = &btf_type_member(t)[member_idx];
+	mname = btf_name_by_offset(btf_vmlinux, member->name_off);
+	func_proto = btf_type_resolve_func_ptr(btf_vmlinux, member->type,
+					       NULL);
+	if (!func_proto) {
+		verbose(env, "attach to invalid member %s(@idx %u) of struct %s\n",
+			mname, member_idx, st_ops->name);
+		return -EINVAL;
+	}
+
+	if (st_ops->check_member) {
+		int err = st_ops->check_member(t, member);
+
+		if (err) {
+			verbose(env, "attach to unsupported member %s of struct %s\n",
+				mname, st_ops->name);
+			return err;
+		}
+	}
+
+	prog->aux->attach_func_proto = func_proto;
+	prog->aux->attach_func_name = mname;
+	env->ops = st_ops->verifier_ops;
+
+	return 0;
+}
+
 static int check_attach_btf_id(struct bpf_verifier_env *env)
 {
 	struct bpf_prog *prog = env->prog;
@@ -9517,6 +9588,9 @@ static int check_attach_btf_id(struct bpf_verifier_env *env)
 	long addr;
 	u64 key;
 
+	if (prog->type == BPF_PROG_TYPE_STRUCT_OPS)
+		return check_struct_ops_btf_id(env);
+
 	if (prog->type != BPF_PROG_TYPE_TRACING)
 		return 0;
 
-- 
cgit v1.2.3-59-g8ed1b


From 85d33df357b634649ddbe0a20fd2d0fc5732c3cb Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <kafai@fb.com>
Date: Wed, 8 Jan 2020 16:35:05 -0800
Subject: bpf: Introduce BPF_MAP_TYPE_STRUCT_OPS

The patch introduces BPF_MAP_TYPE_STRUCT_OPS.  The map value
is a kernel struct with its func ptr implemented in bpf prog.
This new map is the interface to register/unregister/introspect
a bpf implemented kernel struct.

The kernel struct is actually embedded inside another new struct
(or called the "value" struct in the code).  For example,
"struct tcp_congestion_ops" is embbeded in:
struct bpf_struct_ops_tcp_congestion_ops {
	refcount_t refcnt;
	enum bpf_struct_ops_state state;
	struct tcp_congestion_ops data;  /* <-- kernel subsystem struct here */
}
The map value is "struct bpf_struct_ops_tcp_congestion_ops".
The "bpftool map dump" will then be able to show the
state ("inuse"/"tobefree") and the number of subsystem's refcnt (e.g.
number of tcp_sock in the tcp_congestion_ops case).  This "value" struct
is created automatically by a macro.  Having a separate "value" struct
will also make extending "struct bpf_struct_ops_XYZ" easier (e.g. adding
"void (*init)(void)" to "struct bpf_struct_ops_XYZ" to do some
initialization works before registering the struct_ops to the kernel
subsystem).  The libbpf will take care of finding and populating the
"struct bpf_struct_ops_XYZ" from "struct XYZ".

Register a struct_ops to a kernel subsystem:
1. Load all needed BPF_PROG_TYPE_STRUCT_OPS prog(s)
2. Create a BPF_MAP_TYPE_STRUCT_OPS with attr->btf_vmlinux_value_type_id
   set to the btf id "struct bpf_struct_ops_tcp_congestion_ops" of the
   running kernel.
   Instead of reusing the attr->btf_value_type_id,
   btf_vmlinux_value_type_id s added such that attr->btf_fd can still be
   used as the "user" btf which could store other useful sysadmin/debug
   info that may be introduced in the furture,
   e.g. creation-date/compiler-details/map-creator...etc.
3. Create a "struct bpf_struct_ops_tcp_congestion_ops" object as described
   in the running kernel btf.  Populate the value of this object.
   The function ptr should be populated with the prog fds.
4. Call BPF_MAP_UPDATE with the object created in (3) as
   the map value.  The key is always "0".

During BPF_MAP_UPDATE, the code that saves the kernel-func-ptr's
args as an array of u64 is generated.  BPF_MAP_UPDATE also allows
the specific struct_ops to do some final checks in "st_ops->init_member()"
(e.g. ensure all mandatory func ptrs are implemented).
If everything looks good, it will register this kernel struct
to the kernel subsystem.  The map will not allow further update
from this point.

Unregister a struct_ops from the kernel subsystem:
BPF_MAP_DELETE with key "0".

Introspect a struct_ops:
BPF_MAP_LOOKUP_ELEM with key "0".  The map value returned will
have the prog _id_ populated as the func ptr.

The map value state (enum bpf_struct_ops_state) will transit from:
INIT (map created) =>
INUSE (map updated, i.e. reg) =>
TOBEFREE (map value deleted, i.e. unreg)

The kernel subsystem needs to call bpf_struct_ops_get() and
bpf_struct_ops_put() to manage the "refcnt" in the
"struct bpf_struct_ops_XYZ".  This patch uses a separate refcnt
for the purose of tracking the subsystem usage.  Another approach
is to reuse the map->refcnt and then "show" (i.e. during map_lookup)
the subsystem's usage by doing map->refcnt - map->usercnt to filter out
the map-fd/pinned-map usage.  However, that will also tie down the
future semantics of map->refcnt and map->usercnt.

The very first subsystem's refcnt (during reg()) holds one
count to map->refcnt.  When the very last subsystem's refcnt
is gone, it will also release the map->refcnt.  All bpf_prog will be
freed when the map->refcnt reaches 0 (i.e. during map_free()).

Here is how the bpftool map command will look like:
[root@arch-fb-vm1 bpf]# bpftool map show
6: struct_ops  name dctcp  flags 0x0
	key 4B  value 256B  max_entries 1  memlock 4096B
	btf_id 6
[root@arch-fb-vm1 bpf]# bpftool map dump id 6
[{
        "value": {
            "refcnt": {
                "refs": {
                    "counter": 1
                }
            },
            "state": 1,
            "data": {
                "list": {
                    "next": 0,
                    "prev": 0
                },
                "key": 0,
                "flags": 2,
                "init": 24,
                "release": 0,
                "ssthresh": 25,
                "cong_avoid": 30,
                "set_state": 27,
                "cwnd_event": 28,
                "in_ack_event": 26,
                "undo_cwnd": 29,
                "pkts_acked": 0,
                "min_tso_segs": 0,
                "sndbuf_expand": 0,
                "cong_control": 0,
                "get_info": 0,
                "name": [98,112,102,95,100,99,116,99,112,0,0,0,0,0,0,0
                ],
                "owner": 0
            }
        }
    }
]

Misc Notes:
* bpf_struct_ops_map_sys_lookup_elem() is added for syscall lookup.
  It does an inplace update on "*value" instead returning a pointer
  to syscall.c.  Otherwise, it needs a separate copy of "zero" value
  for the BPF_STRUCT_OPS_STATE_INIT to avoid races.

* The bpf_struct_ops_map_delete_elem() is also called without
  preempt_disable() from map_delete_elem().  It is because
  the "->unreg()" may requires sleepable context, e.g.
  the "tcp_unregister_congestion_control()".

* "const" is added to some of the existing "struct btf_func_model *"
  function arg to avoid a compiler warning caused by this patch.

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20200109003505.3855919-1-kafai@fb.com
---
 arch/x86/net/bpf_jit_comp.c |  18 +-
 include/linux/bpf.h         |  49 ++++-
 include/linux/bpf_types.h   |   3 +
 include/linux/btf.h         |  13 ++
 include/uapi/linux/bpf.h    |   7 +-
 kernel/bpf/bpf_struct_ops.c | 511 +++++++++++++++++++++++++++++++++++++++++++-
 kernel/bpf/btf.c            |  20 +-
 kernel/bpf/map_in_map.c     |   3 +-
 kernel/bpf/syscall.c        |  52 +++--
 kernel/bpf/trampoline.c     |   8 +-
 kernel/bpf/verifier.c       |   5 +
 11 files changed, 642 insertions(+), 47 deletions(-)

diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c
index 4c8a2d1f8470..9ba08e9abc09 100644
--- a/arch/x86/net/bpf_jit_comp.c
+++ b/arch/x86/net/bpf_jit_comp.c
@@ -1328,7 +1328,7 @@ emit_jmp:
 	return proglen;
 }
 
-static void save_regs(struct btf_func_model *m, u8 **prog, int nr_args,
+static void save_regs(const struct btf_func_model *m, u8 **prog, int nr_args,
 		      int stack_size)
 {
 	int i;
@@ -1344,7 +1344,7 @@ static void save_regs(struct btf_func_model *m, u8 **prog, int nr_args,
 			 -(stack_size - i * 8));
 }
 
-static void restore_regs(struct btf_func_model *m, u8 **prog, int nr_args,
+static void restore_regs(const struct btf_func_model *m, u8 **prog, int nr_args,
 			 int stack_size)
 {
 	int i;
@@ -1361,7 +1361,7 @@ static void restore_regs(struct btf_func_model *m, u8 **prog, int nr_args,
 			 -(stack_size - i * 8));
 }
 
-static int invoke_bpf(struct btf_func_model *m, u8 **pprog,
+static int invoke_bpf(const struct btf_func_model *m, u8 **pprog,
 		      struct bpf_prog **progs, int prog_cnt, int stack_size)
 {
 	u8 *prog = *pprog;
@@ -1456,7 +1456,8 @@ static int invoke_bpf(struct btf_func_model *m, u8 **pprog,
  * add rsp, 8                      // skip eth_type_trans's frame
  * ret                             // return to its caller
  */
-int arch_prepare_bpf_trampoline(void *image, struct btf_func_model *m, u32 flags,
+int arch_prepare_bpf_trampoline(void *image, void *image_end,
+				const struct btf_func_model *m, u32 flags,
 				struct bpf_prog **fentry_progs, int fentry_cnt,
 				struct bpf_prog **fexit_progs, int fexit_cnt,
 				void *orig_call)
@@ -1523,13 +1524,10 @@ int arch_prepare_bpf_trampoline(void *image, struct btf_func_model *m, u32 flags
 		/* skip our return address and return to parent */
 		EMIT4(0x48, 0x83, 0xC4, 8); /* add rsp, 8 */
 	EMIT1(0xC3); /* ret */
-	/* One half of the page has active running trampoline.
-	 * Another half is an area for next trampoline.
-	 * Make sure the trampoline generation logic doesn't overflow.
-	 */
-	if (WARN_ON_ONCE(prog - (u8 *)image > PAGE_SIZE / 2 - BPF_INSN_SAFETY))
+	/* Make sure the trampoline generation logic doesn't overflow */
+	if (WARN_ON_ONCE(prog > (u8 *)image_end - BPF_INSN_SAFETY))
 		return -EFAULT;
-	return 0;
+	return prog - (u8 *)image;
 }
 
 static int emit_cond_near_jump(u8 **pprog, void *func, void *ip, u8 jmp_cond)
diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 50f3b20ae284..a7bfe8a388c6 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -17,6 +17,7 @@
 #include <linux/u64_stats_sync.h>
 #include <linux/refcount.h>
 #include <linux/mutex.h>
+#include <linux/module.h>
 
 struct bpf_verifier_env;
 struct bpf_verifier_log;
@@ -106,6 +107,7 @@ struct bpf_map {
 	struct btf *btf;
 	struct bpf_map_memory memory;
 	char name[BPF_OBJ_NAME_LEN];
+	u32 btf_vmlinux_value_type_id;
 	bool unpriv_array;
 	bool frozen; /* write-once; write-protected by freeze_mutex */
 	/* 22 bytes hole */
@@ -183,7 +185,8 @@ static inline bool bpf_map_offload_neutral(const struct bpf_map *map)
 
 static inline bool bpf_map_support_seq_show(const struct bpf_map *map)
 {
-	return map->btf && map->ops->map_seq_show_elem;
+	return (map->btf_value_type_id || map->btf_vmlinux_value_type_id) &&
+		map->ops->map_seq_show_elem;
 }
 
 int map_check_no_btf(const struct bpf_map *map,
@@ -441,7 +444,8 @@ struct btf_func_model {
  *      fentry = a set of program to run before calling original function
  *      fexit = a set of program to run after original function
  */
-int arch_prepare_bpf_trampoline(void *image, struct btf_func_model *m, u32 flags,
+int arch_prepare_bpf_trampoline(void *image, void *image_end,
+				const struct btf_func_model *m, u32 flags,
 				struct bpf_prog **fentry_progs, int fentry_cnt,
 				struct bpf_prog **fexit_progs, int fexit_cnt,
 				void *orig_call);
@@ -672,6 +676,7 @@ struct bpf_array_aux {
 	struct work_struct work;
 };
 
+struct bpf_struct_ops_value;
 struct btf_type;
 struct btf_member;
 
@@ -681,21 +686,61 @@ struct bpf_struct_ops {
 	int (*init)(struct btf *btf);
 	int (*check_member)(const struct btf_type *t,
 			    const struct btf_member *member);
+	int (*init_member)(const struct btf_type *t,
+			   const struct btf_member *member,
+			   void *kdata, const void *udata);
+	int (*reg)(void *kdata);
+	void (*unreg)(void *kdata);
 	const struct btf_type *type;
+	const struct btf_type *value_type;
 	const char *name;
 	struct btf_func_model func_models[BPF_STRUCT_OPS_MAX_NR_MEMBERS];
 	u32 type_id;
+	u32 value_id;
 };
 
 #if defined(CONFIG_BPF_JIT) && defined(CONFIG_BPF_SYSCALL)
+#define BPF_MODULE_OWNER ((void *)((0xeB9FUL << 2) + POISON_POINTER_DELTA))
 const struct bpf_struct_ops *bpf_struct_ops_find(u32 type_id);
 void bpf_struct_ops_init(struct btf *btf);
+bool bpf_struct_ops_get(const void *kdata);
+void bpf_struct_ops_put(const void *kdata);
+int bpf_struct_ops_map_sys_lookup_elem(struct bpf_map *map, void *key,
+				       void *value);
+static inline bool bpf_try_module_get(const void *data, struct module *owner)
+{
+	if (owner == BPF_MODULE_OWNER)
+		return bpf_struct_ops_get(data);
+	else
+		return try_module_get(owner);
+}
+static inline void bpf_module_put(const void *data, struct module *owner)
+{
+	if (owner == BPF_MODULE_OWNER)
+		bpf_struct_ops_put(data);
+	else
+		module_put(owner);
+}
 #else
 static inline const struct bpf_struct_ops *bpf_struct_ops_find(u32 type_id)
 {
 	return NULL;
 }
 static inline void bpf_struct_ops_init(struct btf *btf) { }
+static inline bool bpf_try_module_get(const void *data, struct module *owner)
+{
+	return try_module_get(owner);
+}
+static inline void bpf_module_put(const void *data, struct module *owner)
+{
+	module_put(owner);
+}
+static inline int bpf_struct_ops_map_sys_lookup_elem(struct bpf_map *map,
+						     void *key,
+						     void *value)
+{
+	return -EINVAL;
+}
 #endif
 
 struct bpf_array {
diff --git a/include/linux/bpf_types.h b/include/linux/bpf_types.h
index fadd243ffa2d..9f326e6ef885 100644
--- a/include/linux/bpf_types.h
+++ b/include/linux/bpf_types.h
@@ -109,3 +109,6 @@ BPF_MAP_TYPE(BPF_MAP_TYPE_REUSEPORT_SOCKARRAY, reuseport_array_ops)
 #endif
 BPF_MAP_TYPE(BPF_MAP_TYPE_QUEUE, queue_map_ops)
 BPF_MAP_TYPE(BPF_MAP_TYPE_STACK, stack_map_ops)
+#if defined(CONFIG_BPF_JIT)
+BPF_MAP_TYPE(BPF_MAP_TYPE_STRUCT_OPS, bpf_struct_ops_map_ops)
+#endif
diff --git a/include/linux/btf.h b/include/linux/btf.h
index f74a09a7120b..881e9b76ef49 100644
--- a/include/linux/btf.h
+++ b/include/linux/btf.h
@@ -7,6 +7,8 @@
 #include <linux/types.h>
 #include <uapi/linux/btf.h>
 
+#define BTF_TYPE_EMIT(type) ((void)(type *)0)
+
 struct btf;
 struct btf_member;
 struct btf_type;
@@ -60,6 +62,10 @@ const struct btf_type *btf_type_resolve_ptr(const struct btf *btf,
 					    u32 id, u32 *res_id);
 const struct btf_type *btf_type_resolve_func_ptr(const struct btf *btf,
 						 u32 id, u32 *res_id);
+const struct btf_type *
+btf_resolve_size(const struct btf *btf, const struct btf_type *type,
+		 u32 *type_size, const struct btf_type **elem_type,
+		 u32 *total_nelems);
 
 #define for_each_member(i, struct_type, member)			\
 	for (i = 0, member = btf_type_member(struct_type);	\
@@ -106,6 +112,13 @@ static inline bool btf_type_kflag(const struct btf_type *t)
 	return BTF_INFO_KFLAG(t->info);
 }
 
+static inline u32 btf_member_bit_offset(const struct btf_type *struct_type,
+					const struct btf_member *member)
+{
+	return btf_type_kflag(struct_type) ? BTF_MEMBER_BIT_OFFSET(member->offset)
+					   : member->offset;
+}
+
 static inline u32 btf_member_bitfield_size(const struct btf_type *struct_type,
 					   const struct btf_member *member)
 {
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index c1eeb3e0e116..38059880963e 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -136,6 +136,7 @@ enum bpf_map_type {
 	BPF_MAP_TYPE_STACK,
 	BPF_MAP_TYPE_SK_STORAGE,
 	BPF_MAP_TYPE_DEVMAP_HASH,
+	BPF_MAP_TYPE_STRUCT_OPS,
 };
 
 /* Note that tracing related programs such as
@@ -398,6 +399,10 @@ union bpf_attr {
 		__u32	btf_fd;		/* fd pointing to a BTF type data */
 		__u32	btf_key_type_id;	/* BTF type_id of the key */
 		__u32	btf_value_type_id;	/* BTF type_id of the value */
+		__u32	btf_vmlinux_value_type_id;/* BTF type_id of a kernel-
+						   * struct stored as the
+						   * map value
+						   */
 	};
 
 	struct { /* anonymous struct used by BPF_MAP_*_ELEM commands */
@@ -3350,7 +3355,7 @@ struct bpf_map_info {
 	__u32 map_flags;
 	char  name[BPF_OBJ_NAME_LEN];
 	__u32 ifindex;
-	__u32 :32;
+	__u32 btf_vmlinux_value_type_id;
 	__u64 netns_dev;
 	__u64 netns_ino;
 	__u32 btf_id;
diff --git a/kernel/bpf/bpf_struct_ops.c b/kernel/bpf/bpf_struct_ops.c
index 2ea68fe34c33..ddf48f49914b 100644
--- a/kernel/bpf/bpf_struct_ops.c
+++ b/kernel/bpf/bpf_struct_ops.c
@@ -9,9 +9,68 @@
 #include <linux/numa.h>
 #include <linux/seq_file.h>
 #include <linux/refcount.h>
+#include <linux/mutex.h>
 
+enum bpf_struct_ops_state {
+	BPF_STRUCT_OPS_STATE_INIT,
+	BPF_STRUCT_OPS_STATE_INUSE,
+	BPF_STRUCT_OPS_STATE_TOBEFREE,
+};
+
+#define BPF_STRUCT_OPS_COMMON_VALUE			\
+	refcount_t refcnt;				\
+	enum bpf_struct_ops_state state
+
+struct bpf_struct_ops_value {
+	BPF_STRUCT_OPS_COMMON_VALUE;
+	char data[0] ____cacheline_aligned_in_smp;
+};
+
+struct bpf_struct_ops_map {
+	struct bpf_map map;
+	const struct bpf_struct_ops *st_ops;
+	/* protect map_update */
+	struct mutex lock;
+	/* progs has all the bpf_prog that is populated
+	 * to the func ptr of the kernel's struct
+	 * (in kvalue.data).
+	 */
+	struct bpf_prog **progs;
+	/* image is a page that has all the trampolines
+	 * that stores the func args before calling the bpf_prog.
+	 * A PAGE_SIZE "image" is enough to store all trampoline for
+	 * "progs[]".
+	 */
+	void *image;
+	/* uvalue->data stores the kernel struct
+	 * (e.g. tcp_congestion_ops) that is more useful
+	 * to userspace than the kvalue.  For example,
+	 * the bpf_prog's id is stored instead of the kernel
+	 * address of a func ptr.
+	 */
+	struct bpf_struct_ops_value *uvalue;
+	/* kvalue.data stores the actual kernel's struct
+	 * (e.g. tcp_congestion_ops) that will be
+	 * registered to the kernel subsystem.
+	 */
+	struct bpf_struct_ops_value kvalue;
+};
+
+#define VALUE_PREFIX "bpf_struct_ops_"
+#define VALUE_PREFIX_LEN (sizeof(VALUE_PREFIX) - 1)
+
+/* bpf_struct_ops_##_name (e.g. bpf_struct_ops_tcp_congestion_ops) is
+ * the map's value exposed to the userspace and its btf-type-id is
+ * stored at the map->btf_vmlinux_value_type_id.
+ *
+ */
 #define BPF_STRUCT_OPS_TYPE(_name)				\
-extern struct bpf_struct_ops bpf_##_name;
+extern struct bpf_struct_ops bpf_##_name;			\
+								\
+struct bpf_struct_ops_##_name {						\
+	BPF_STRUCT_OPS_COMMON_VALUE;				\
+	struct _name data ____cacheline_aligned_in_smp;		\
+};
 #include "bpf_struct_ops_types.h"
 #undef BPF_STRUCT_OPS_TYPE
 
@@ -35,19 +94,50 @@ const struct bpf_verifier_ops bpf_struct_ops_verifier_ops = {
 const struct bpf_prog_ops bpf_struct_ops_prog_ops = {
 };
 
+static const struct btf_type *module_type;
+
 void bpf_struct_ops_init(struct btf *btf)
 {
+	s32 type_id, value_id, module_id;
 	const struct btf_member *member;
 	struct bpf_struct_ops *st_ops;
 	struct bpf_verifier_log log = {};
 	const struct btf_type *t;
+	char value_name[128];
 	const char *mname;
-	s32 type_id;
 	u32 i, j;
 
+	/* Ensure BTF type is emitted for "struct bpf_struct_ops_##_name" */
+#define BPF_STRUCT_OPS_TYPE(_name) BTF_TYPE_EMIT(struct bpf_struct_ops_##_name);
+#include "bpf_struct_ops_types.h"
+#undef BPF_STRUCT_OPS_TYPE
+
+	module_id = btf_find_by_name_kind(btf, "module", BTF_KIND_STRUCT);
+	if (module_id < 0) {
+		pr_warn("Cannot find struct module in btf_vmlinux\n");
+		return;
+	}
+	module_type = btf_type_by_id(btf, module_id);
+
 	for (i = 0; i < ARRAY_SIZE(bpf_struct_ops); i++) {
 		st_ops = bpf_struct_ops[i];
 
+		if (strlen(st_ops->name) + VALUE_PREFIX_LEN >=
+		    sizeof(value_name)) {
+			pr_warn("struct_ops name %s is too long\n",
+				st_ops->name);
+			continue;
+		}
+		sprintf(value_name, "%s%s", VALUE_PREFIX, st_ops->name);
+
+		value_id = btf_find_by_name_kind(btf, value_name,
+						 BTF_KIND_STRUCT);
+		if (value_id < 0) {
+			pr_warn("Cannot find struct %s in btf_vmlinux\n",
+				value_name);
+			continue;
+		}
+
 		type_id = btf_find_by_name_kind(btf, st_ops->name,
 						BTF_KIND_STRUCT);
 		if (type_id < 0) {
@@ -98,6 +188,9 @@ void bpf_struct_ops_init(struct btf *btf)
 			} else {
 				st_ops->type_id = type_id;
 				st_ops->type = t;
+				st_ops->value_id = value_id;
+				st_ops->value_type = btf_type_by_id(btf,
+								    value_id);
 			}
 		}
 	}
@@ -105,6 +198,22 @@ void bpf_struct_ops_init(struct btf *btf)
 
 extern struct btf *btf_vmlinux;
 
+static const struct bpf_struct_ops *
+bpf_struct_ops_find_value(u32 value_id)
+{
+	unsigned int i;
+
+	if (!value_id || !btf_vmlinux)
+		return NULL;
+
+	for (i = 0; i < ARRAY_SIZE(bpf_struct_ops); i++) {
+		if (bpf_struct_ops[i]->value_id == value_id)
+			return bpf_struct_ops[i];
+	}
+
+	return NULL;
+}
+
 const struct bpf_struct_ops *bpf_struct_ops_find(u32 type_id)
 {
 	unsigned int i;
@@ -119,3 +228,401 @@ const struct bpf_struct_ops *bpf_struct_ops_find(u32 type_id)
 
 	return NULL;
 }
+
+static int bpf_struct_ops_map_get_next_key(struct bpf_map *map, void *key,
+					   void *next_key)
+{
+	if (key && *(u32 *)key == 0)
+		return -ENOENT;
+
+	*(u32 *)next_key = 0;
+	return 0;
+}
+
+int bpf_struct_ops_map_sys_lookup_elem(struct bpf_map *map, void *key,
+				       void *value)
+{
+	struct bpf_struct_ops_map *st_map = (struct bpf_struct_ops_map *)map;
+	struct bpf_struct_ops_value *uvalue, *kvalue;
+	enum bpf_struct_ops_state state;
+
+	if (unlikely(*(u32 *)key != 0))
+		return -ENOENT;
+
+	kvalue = &st_map->kvalue;
+	/* Pair with smp_store_release() during map_update */
+	state = smp_load_acquire(&kvalue->state);
+	if (state == BPF_STRUCT_OPS_STATE_INIT) {
+		memset(value, 0, map->value_size);
+		return 0;
+	}
+
+	/* No lock is needed.  state and refcnt do not need
+	 * to be updated together under atomic context.
+	 */
+	uvalue = (struct bpf_struct_ops_value *)value;
+	memcpy(uvalue, st_map->uvalue, map->value_size);
+	uvalue->state = state;
+	refcount_set(&uvalue->refcnt, refcount_read(&kvalue->refcnt));
+
+	return 0;
+}
+
+static void *bpf_struct_ops_map_lookup_elem(struct bpf_map *map, void *key)
+{
+	return ERR_PTR(-EINVAL);
+}
+
+static void bpf_struct_ops_map_put_progs(struct bpf_struct_ops_map *st_map)
+{
+	const struct btf_type *t = st_map->st_ops->type;
+	u32 i;
+
+	for (i = 0; i < btf_type_vlen(t); i++) {
+		if (st_map->progs[i]) {
+			bpf_prog_put(st_map->progs[i]);
+			st_map->progs[i] = NULL;
+		}
+	}
+}
+
+static int check_zero_holes(const struct btf_type *t, void *data)
+{
+	const struct btf_member *member;
+	u32 i, moff, msize, prev_mend = 0;
+	const struct btf_type *mtype;
+
+	for_each_member(i, t, member) {
+		moff = btf_member_bit_offset(t, member) / 8;
+		if (moff > prev_mend &&
+		    memchr_inv(data + prev_mend, 0, moff - prev_mend))
+			return -EINVAL;
+
+		mtype = btf_type_by_id(btf_vmlinux, member->type);
+		mtype = btf_resolve_size(btf_vmlinux, mtype, &msize,
+					 NULL, NULL);
+		if (IS_ERR(mtype))
+			return PTR_ERR(mtype);
+		prev_mend = moff + msize;
+	}
+
+	if (t->size > prev_mend &&
+	    memchr_inv(data + prev_mend, 0, t->size - prev_mend))
+		return -EINVAL;
+
+	return 0;
+}
+
+static int bpf_struct_ops_map_update_elem(struct bpf_map *map, void *key,
+					  void *value, u64 flags)
+{
+	struct bpf_struct_ops_map *st_map = (struct bpf_struct_ops_map *)map;
+	const struct bpf_struct_ops *st_ops = st_map->st_ops;
+	struct bpf_struct_ops_value *uvalue, *kvalue;
+	const struct btf_member *member;
+	const struct btf_type *t = st_ops->type;
+	void *udata, *kdata;
+	int prog_fd, err = 0;
+	void *image;
+	u32 i;
+
+	if (flags)
+		return -EINVAL;
+
+	if (*(u32 *)key != 0)
+		return -E2BIG;
+
+	err = check_zero_holes(st_ops->value_type, value);
+	if (err)
+		return err;
+
+	uvalue = (struct bpf_struct_ops_value *)value;
+	err = check_zero_holes(t, uvalue->data);
+	if (err)
+		return err;
+
+	if (uvalue->state || refcount_read(&uvalue->refcnt))
+		return -EINVAL;
+
+	uvalue = (struct bpf_struct_ops_value *)st_map->uvalue;
+	kvalue = (struct bpf_struct_ops_value *)&st_map->kvalue;
+
+	mutex_lock(&st_map->lock);
+
+	if (kvalue->state != BPF_STRUCT_OPS_STATE_INIT) {
+		err = -EBUSY;
+		goto unlock;
+	}
+
+	memcpy(uvalue, value, map->value_size);
+
+	udata = &uvalue->data;
+	kdata = &kvalue->data;
+	image = st_map->image;
+
+	for_each_member(i, t, member) {
+		const struct btf_type *mtype, *ptype;
+		struct bpf_prog *prog;
+		u32 moff;
+
+		moff = btf_member_bit_offset(t, member) / 8;
+		ptype = btf_type_resolve_ptr(btf_vmlinux, member->type, NULL);
+		if (ptype == module_type) {
+			if (*(void **)(udata + moff))
+				goto reset_unlock;
+			*(void **)(kdata + moff) = BPF_MODULE_OWNER;
+			continue;
+		}
+
+		err = st_ops->init_member(t, member, kdata, udata);
+		if (err < 0)
+			goto reset_unlock;
+
+		/* The ->init_member() has handled this member */
+		if (err > 0)
+			continue;
+
+		/* If st_ops->init_member does not handle it,
+		 * we will only handle func ptrs and zero-ed members
+		 * here.  Reject everything else.
+		 */
+
+		/* All non func ptr member must be 0 */
+		if (!ptype || !btf_type_is_func_proto(ptype)) {
+			u32 msize;
+
+			mtype = btf_type_by_id(btf_vmlinux, member->type);
+			mtype = btf_resolve_size(btf_vmlinux, mtype, &msize,
+						 NULL, NULL);
+			if (IS_ERR(mtype)) {
+				err = PTR_ERR(mtype);
+				goto reset_unlock;
+			}
+
+			if (memchr_inv(udata + moff, 0, msize)) {
+				err = -EINVAL;
+				goto reset_unlock;
+			}
+
+			continue;
+		}
+
+		prog_fd = (int)(*(unsigned long *)(udata + moff));
+		/* Similar check as the attr->attach_prog_fd */
+		if (!prog_fd)
+			continue;
+
+		prog = bpf_prog_get(prog_fd);
+		if (IS_ERR(prog)) {
+			err = PTR_ERR(prog);
+			goto reset_unlock;
+		}
+		st_map->progs[i] = prog;
+
+		if (prog->type != BPF_PROG_TYPE_STRUCT_OPS ||
+		    prog->aux->attach_btf_id != st_ops->type_id ||
+		    prog->expected_attach_type != i) {
+			err = -EINVAL;
+			goto reset_unlock;
+		}
+
+		err = arch_prepare_bpf_trampoline(image,
+						  st_map->image + PAGE_SIZE,
+						  &st_ops->func_models[i], 0,
+						  &prog, 1, NULL, 0, NULL);
+		if (err < 0)
+			goto reset_unlock;
+
+		*(void **)(kdata + moff) = image;
+		image += err;
+
+		/* put prog_id to udata */
+		*(unsigned long *)(udata + moff) = prog->aux->id;
+	}
+
+	refcount_set(&kvalue->refcnt, 1);
+	bpf_map_inc(map);
+
+	set_memory_ro((long)st_map->image, 1);
+	set_memory_x((long)st_map->image, 1);
+	err = st_ops->reg(kdata);
+	if (likely(!err)) {
+		/* Pair with smp_load_acquire() during lookup_elem().
+		 * It ensures the above udata updates (e.g. prog->aux->id)
+		 * can be seen once BPF_STRUCT_OPS_STATE_INUSE is set.
+		 */
+		smp_store_release(&kvalue->state, BPF_STRUCT_OPS_STATE_INUSE);
+		goto unlock;
+	}
+
+	/* Error during st_ops->reg().  It is very unlikely since
+	 * the above init_member() should have caught it earlier
+	 * before reg().  The only possibility is if there was a race
+	 * in registering the struct_ops (under the same name) to
+	 * a sub-system through different struct_ops's maps.
+	 */
+	set_memory_nx((long)st_map->image, 1);
+	set_memory_rw((long)st_map->image, 1);
+	bpf_map_put(map);
+
+reset_unlock:
+	bpf_struct_ops_map_put_progs(st_map);
+	memset(uvalue, 0, map->value_size);
+	memset(kvalue, 0, map->value_size);
+unlock:
+	mutex_unlock(&st_map->lock);
+	return err;
+}
+
+static int bpf_struct_ops_map_delete_elem(struct bpf_map *map, void *key)
+{
+	enum bpf_struct_ops_state prev_state;
+	struct bpf_struct_ops_map *st_map;
+
+	st_map = (struct bpf_struct_ops_map *)map;
+	prev_state = cmpxchg(&st_map->kvalue.state,
+			     BPF_STRUCT_OPS_STATE_INUSE,
+			     BPF_STRUCT_OPS_STATE_TOBEFREE);
+	if (prev_state == BPF_STRUCT_OPS_STATE_INUSE) {
+		st_map->st_ops->unreg(&st_map->kvalue.data);
+		if (refcount_dec_and_test(&st_map->kvalue.refcnt))
+			bpf_map_put(map);
+	}
+
+	return 0;
+}
+
+static void bpf_struct_ops_map_seq_show_elem(struct bpf_map *map, void *key,
+					     struct seq_file *m)
+{
+	void *value;
+
+	value = bpf_struct_ops_map_lookup_elem(map, key);
+	if (!value)
+		return;
+
+	btf_type_seq_show(btf_vmlinux, map->btf_vmlinux_value_type_id,
+			  value, m);
+	seq_puts(m, "\n");
+}
+
+static void bpf_struct_ops_map_free(struct bpf_map *map)
+{
+	struct bpf_struct_ops_map *st_map = (struct bpf_struct_ops_map *)map;
+
+	if (st_map->progs)
+		bpf_struct_ops_map_put_progs(st_map);
+	bpf_map_area_free(st_map->progs);
+	bpf_jit_free_exec(st_map->image);
+	bpf_map_area_free(st_map->uvalue);
+	bpf_map_area_free(st_map);
+}
+
+static int bpf_struct_ops_map_alloc_check(union bpf_attr *attr)
+{
+	if (attr->key_size != sizeof(unsigned int) || attr->max_entries != 1 ||
+	    attr->map_flags || !attr->btf_vmlinux_value_type_id)
+		return -EINVAL;
+	return 0;
+}
+
+static struct bpf_map *bpf_struct_ops_map_alloc(union bpf_attr *attr)
+{
+	const struct bpf_struct_ops *st_ops;
+	size_t map_total_size, st_map_size;
+	struct bpf_struct_ops_map *st_map;
+	const struct btf_type *t, *vt;
+	struct bpf_map_memory mem;
+	struct bpf_map *map;
+	int err;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return ERR_PTR(-EPERM);
+
+	st_ops = bpf_struct_ops_find_value(attr->btf_vmlinux_value_type_id);
+	if (!st_ops)
+		return ERR_PTR(-ENOTSUPP);
+
+	vt = st_ops->value_type;
+	if (attr->value_size != vt->size)
+		return ERR_PTR(-EINVAL);
+
+	t = st_ops->type;
+
+	st_map_size = sizeof(*st_map) +
+		/* kvalue stores the
+		 * struct bpf_struct_ops_tcp_congestions_ops
+		 */
+		(vt->size - sizeof(struct bpf_struct_ops_value));
+	map_total_size = st_map_size +
+		/* uvalue */
+		sizeof(vt->size) +
+		/* struct bpf_progs **progs */
+		 btf_type_vlen(t) * sizeof(struct bpf_prog *);
+	err = bpf_map_charge_init(&mem, map_total_size);
+	if (err < 0)
+		return ERR_PTR(err);
+
+	st_map = bpf_map_area_alloc(st_map_size, NUMA_NO_NODE);
+	if (!st_map) {
+		bpf_map_charge_finish(&mem);
+		return ERR_PTR(-ENOMEM);
+	}
+	st_map->st_ops = st_ops;
+	map = &st_map->map;
+
+	st_map->uvalue = bpf_map_area_alloc(vt->size, NUMA_NO_NODE);
+	st_map->progs =
+		bpf_map_area_alloc(btf_type_vlen(t) * sizeof(struct bpf_prog *),
+				   NUMA_NO_NODE);
+	st_map->image = bpf_jit_alloc_exec(PAGE_SIZE);
+	if (!st_map->uvalue || !st_map->progs || !st_map->image) {
+		bpf_struct_ops_map_free(map);
+		bpf_map_charge_finish(&mem);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	mutex_init(&st_map->lock);
+	set_vm_flush_reset_perms(st_map->image);
+	bpf_map_init_from_attr(map, attr);
+	bpf_map_charge_move(&map->memory, &mem);
+
+	return map;
+}
+
+const struct bpf_map_ops bpf_struct_ops_map_ops = {
+	.map_alloc_check = bpf_struct_ops_map_alloc_check,
+	.map_alloc = bpf_struct_ops_map_alloc,
+	.map_free = bpf_struct_ops_map_free,
+	.map_get_next_key = bpf_struct_ops_map_get_next_key,
+	.map_lookup_elem = bpf_struct_ops_map_lookup_elem,
+	.map_delete_elem = bpf_struct_ops_map_delete_elem,
+	.map_update_elem = bpf_struct_ops_map_update_elem,
+	.map_seq_show_elem = bpf_struct_ops_map_seq_show_elem,
+};
+
+/* "const void *" because some subsystem is
+ * passing a const (e.g. const struct tcp_congestion_ops *)
+ */
+bool bpf_struct_ops_get(const void *kdata)
+{
+	struct bpf_struct_ops_value *kvalue;
+
+	kvalue = container_of(kdata, struct bpf_struct_ops_value, data);
+
+	return refcount_inc_not_zero(&kvalue->refcnt);
+}
+
+void bpf_struct_ops_put(const void *kdata)
+{
+	struct bpf_struct_ops_value *kvalue;
+
+	kvalue = container_of(kdata, struct bpf_struct_ops_value, data);
+	if (refcount_dec_and_test(&kvalue->refcnt)) {
+		struct bpf_struct_ops_map *st_map;
+
+		st_map = container_of(kvalue, struct bpf_struct_ops_map,
+				      kvalue);
+		bpf_map_put(&st_map->map);
+	}
+}
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 12af4a1bb1a4..81d9cf75cacd 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -500,13 +500,6 @@ static const char *btf_int_encoding_str(u8 encoding)
 		return "UNKN";
 }
 
-static u32 btf_member_bit_offset(const struct btf_type *struct_type,
-			     const struct btf_member *member)
-{
-	return btf_type_kflag(struct_type) ? BTF_MEMBER_BIT_OFFSET(member->offset)
-					   : member->offset;
-}
-
 static u32 btf_type_int(const struct btf_type *t)
 {
 	return *(u32 *)(t + 1);
@@ -1089,7 +1082,7 @@ static const struct resolve_vertex *env_stack_peak(struct btf_verifier_env *env)
  * *elem_type: same as return type ("struct X")
  * *total_nelems: 1
  */
-static const struct btf_type *
+const struct btf_type *
 btf_resolve_size(const struct btf *btf, const struct btf_type *type,
 		 u32 *type_size, const struct btf_type **elem_type,
 		 u32 *total_nelems)
@@ -1143,8 +1136,10 @@ resolved:
 		return ERR_PTR(-EINVAL);
 
 	*type_size = nelems * size;
-	*total_nelems = nelems;
-	*elem_type = type;
+	if (total_nelems)
+		*total_nelems = nelems;
+	if (elem_type)
+		*elem_type = type;
 
 	return array_type ? : type;
 }
@@ -1858,7 +1853,10 @@ static void btf_modifier_seq_show(const struct btf *btf,
 				  u32 type_id, void *data,
 				  u8 bits_offset, struct seq_file *m)
 {
-	t = btf_type_id_resolve(btf, &type_id);
+	if (btf->resolved_ids)
+		t = btf_type_id_resolve(btf, &type_id);
+	else
+		t = btf_type_skip_modifiers(btf, type_id, NULL);
 
 	btf_type_ops(t)->seq_show(btf, t, type_id, data, bits_offset, m);
 }
diff --git a/kernel/bpf/map_in_map.c b/kernel/bpf/map_in_map.c
index 5e9366b33f0f..b3c48d1533cb 100644
--- a/kernel/bpf/map_in_map.c
+++ b/kernel/bpf/map_in_map.c
@@ -22,7 +22,8 @@ struct bpf_map *bpf_map_meta_alloc(int inner_map_ufd)
 	 */
 	if (inner_map->map_type == BPF_MAP_TYPE_PROG_ARRAY ||
 	    inner_map->map_type == BPF_MAP_TYPE_CGROUP_STORAGE ||
-	    inner_map->map_type == BPF_MAP_TYPE_PERCPU_CGROUP_STORAGE) {
+	    inner_map->map_type == BPF_MAP_TYPE_PERCPU_CGROUP_STORAGE ||
+	    inner_map->map_type == BPF_MAP_TYPE_STRUCT_OPS) {
 		fdput(f);
 		return ERR_PTR(-ENOTSUPP);
 	}
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 03a02ef4c496..f9db72a96ec0 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -628,7 +628,7 @@ static int map_check_btf(struct bpf_map *map, const struct btf *btf,
 	return ret;
 }
 
-#define BPF_MAP_CREATE_LAST_FIELD btf_value_type_id
+#define BPF_MAP_CREATE_LAST_FIELD btf_vmlinux_value_type_id
 /* called via syscall */
 static int map_create(union bpf_attr *attr)
 {
@@ -642,6 +642,14 @@ static int map_create(union bpf_attr *attr)
 	if (err)
 		return -EINVAL;
 
+	if (attr->btf_vmlinux_value_type_id) {
+		if (attr->map_type != BPF_MAP_TYPE_STRUCT_OPS ||
+		    attr->btf_key_type_id || attr->btf_value_type_id)
+			return -EINVAL;
+	} else if (attr->btf_key_type_id && !attr->btf_value_type_id) {
+		return -EINVAL;
+	}
+
 	f_flags = bpf_get_file_flag(attr->map_flags);
 	if (f_flags < 0)
 		return f_flags;
@@ -664,32 +672,35 @@ static int map_create(union bpf_attr *attr)
 	atomic64_set(&map->usercnt, 1);
 	mutex_init(&map->freeze_mutex);
 
-	if (attr->btf_key_type_id || attr->btf_value_type_id) {
+	map->spin_lock_off = -EINVAL;
+	if (attr->btf_key_type_id || attr->btf_value_type_id ||
+	    /* Even the map's value is a kernel's struct,
+	     * the bpf_prog.o must have BTF to begin with
+	     * to figure out the corresponding kernel's
+	     * counter part.  Thus, attr->btf_fd has
+	     * to be valid also.
+	     */
+	    attr->btf_vmlinux_value_type_id) {
 		struct btf *btf;
 
-		if (!attr->btf_value_type_id) {
-			err = -EINVAL;
-			goto free_map;
-		}
-
 		btf = btf_get_by_fd(attr->btf_fd);
 		if (IS_ERR(btf)) {
 			err = PTR_ERR(btf);
 			goto free_map;
 		}
+		map->btf = btf;
 
-		err = map_check_btf(map, btf, attr->btf_key_type_id,
-				    attr->btf_value_type_id);
-		if (err) {
-			btf_put(btf);
-			goto free_map;
+		if (attr->btf_value_type_id) {
+			err = map_check_btf(map, btf, attr->btf_key_type_id,
+					    attr->btf_value_type_id);
+			if (err)
+				goto free_map;
 		}
 
-		map->btf = btf;
 		map->btf_key_type_id = attr->btf_key_type_id;
 		map->btf_value_type_id = attr->btf_value_type_id;
-	} else {
-		map->spin_lock_off = -EINVAL;
+		map->btf_vmlinux_value_type_id =
+			attr->btf_vmlinux_value_type_id;
 	}
 
 	err = security_bpf_map_alloc(map);
@@ -888,6 +899,9 @@ static int map_lookup_elem(union bpf_attr *attr)
 	} else if (map->map_type == BPF_MAP_TYPE_QUEUE ||
 		   map->map_type == BPF_MAP_TYPE_STACK) {
 		err = map->ops->map_peek_elem(map, value);
+	} else if (map->map_type == BPF_MAP_TYPE_STRUCT_OPS) {
+		/* struct_ops map requires directly updating "value" */
+		err = bpf_struct_ops_map_sys_lookup_elem(map, key, value);
 	} else {
 		rcu_read_lock();
 		if (map->ops->map_lookup_elem_sys_only)
@@ -1003,7 +1017,8 @@ static int map_update_elem(union bpf_attr *attr)
 		goto out;
 	} else if (map->map_type == BPF_MAP_TYPE_CPUMAP ||
 		   map->map_type == BPF_MAP_TYPE_SOCKHASH ||
-		   map->map_type == BPF_MAP_TYPE_SOCKMAP) {
+		   map->map_type == BPF_MAP_TYPE_SOCKMAP ||
+		   map->map_type == BPF_MAP_TYPE_STRUCT_OPS) {
 		err = map->ops->map_update_elem(map, key, value, attr->flags);
 		goto out;
 	} else if (IS_FD_PROG_ARRAY(map)) {
@@ -1092,7 +1107,9 @@ static int map_delete_elem(union bpf_attr *attr)
 	if (bpf_map_is_dev_bound(map)) {
 		err = bpf_map_offload_delete_elem(map, key);
 		goto out;
-	} else if (IS_FD_PROG_ARRAY(map)) {
+	} else if (IS_FD_PROG_ARRAY(map) ||
+		   map->map_type == BPF_MAP_TYPE_STRUCT_OPS) {
+		/* These maps require sleepable context */
 		err = map->ops->map_delete_elem(map, key);
 		goto out;
 	}
@@ -2822,6 +2839,7 @@ static int bpf_map_get_info_by_fd(struct bpf_map *map,
 		info.btf_key_type_id = map->btf_key_type_id;
 		info.btf_value_type_id = map->btf_value_type_id;
 	}
+	info.btf_vmlinux_value_type_id = map->btf_vmlinux_value_type_id;
 
 	if (bpf_map_is_dev_bound(map)) {
 		err = bpf_map_offload_info_fill(&info, map);
diff --git a/kernel/bpf/trampoline.c b/kernel/bpf/trampoline.c
index 505f4e4b31d2..79a04417050d 100644
--- a/kernel/bpf/trampoline.c
+++ b/kernel/bpf/trampoline.c
@@ -160,11 +160,12 @@ static int bpf_trampoline_update(struct bpf_trampoline *tr)
 	if (fexit_cnt)
 		flags = BPF_TRAMP_F_CALL_ORIG | BPF_TRAMP_F_SKIP_FRAME;
 
-	err = arch_prepare_bpf_trampoline(new_image, &tr->func.model, flags,
+	err = arch_prepare_bpf_trampoline(new_image, new_image + PAGE_SIZE / 2,
+					  &tr->func.model, flags,
 					  fentry, fentry_cnt,
 					  fexit, fexit_cnt,
 					  tr->func.addr);
-	if (err)
+	if (err < 0)
 		goto out;
 
 	if (tr->selector)
@@ -296,7 +297,8 @@ void notrace __bpf_prog_exit(struct bpf_prog *prog, u64 start)
 }
 
 int __weak
-arch_prepare_bpf_trampoline(void *image, struct btf_func_model *m, u32 flags,
+arch_prepare_bpf_trampoline(void *image, void *image_end,
+			    const struct btf_func_model *m, u32 flags,
 			    struct bpf_prog **fentry_progs, int fentry_cnt,
 			    struct bpf_prog **fexit_progs, int fexit_cnt,
 			    void *orig_call)
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 586ed3c94b80..f5af759a8a5f 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -8155,6 +8155,11 @@ static int check_map_prog_compatibility(struct bpf_verifier_env *env,
 		return -EINVAL;
 	}
 
+	if (map->map_type == BPF_MAP_TYPE_STRUCT_OPS) {
+		verbose(env, "bpf_struct_ops map cannot be used in prog\n");
+		return -EINVAL;
+	}
+
 	return 0;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 0baf26b0fcd74bbfcef53c5d5e8bad2b99c8d0d2 Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <kafai@fb.com>
Date: Wed, 8 Jan 2020 16:35:08 -0800
Subject: bpf: tcp: Support tcp_congestion_ops in bpf

This patch makes "struct tcp_congestion_ops" to be the first user
of BPF STRUCT_OPS.  It allows implementing a tcp_congestion_ops
in bpf.

The BPF implemented tcp_congestion_ops can be used like
regular kernel tcp-cc through sysctl and setsockopt.  e.g.
[root@arch-fb-vm1 bpf]# sysctl -a | egrep congestion
net.ipv4.tcp_allowed_congestion_control = reno cubic bpf_cubic
net.ipv4.tcp_available_congestion_control = reno bic cubic bpf_cubic
net.ipv4.tcp_congestion_control = bpf_cubic

There has been attempt to move the TCP CC to the user space
(e.g. CCP in TCP).   The common arguments are faster turn around,
get away from long-tail kernel versions in production...etc,
which are legit points.

BPF has been the continuous effort to join both kernel and
userspace upsides together (e.g. XDP to gain the performance
advantage without bypassing the kernel).  The recent BPF
advancements (in particular BTF-aware verifier, BPF trampoline,
BPF CO-RE...) made implementing kernel struct ops (e.g. tcp cc)
possible in BPF.  It allows a faster turnaround for testing algorithm
in the production while leveraging the existing (and continue growing)
BPF feature/framework instead of building one specifically for
userspace TCP CC.

This patch allows write access to a few fields in tcp-sock
(in bpf_tcp_ca_btf_struct_access()).

The optional "get_info" is unsupported now.  It can be added
later.  One possible way is to output the info with a btf-id
to describe the content.

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20200109003508.3856115-1-kafai@fb.com
---
 include/linux/filter.h            |   2 +
 include/net/tcp.h                 |   2 +
 kernel/bpf/bpf_struct_ops_types.h |   7 +-
 net/core/filter.c                 |   2 +-
 net/ipv4/Makefile                 |   4 +
 net/ipv4/bpf_tcp_ca.c             | 230 ++++++++++++++++++++++++++++++++++++++
 net/ipv4/tcp_cong.c               |  16 +--
 net/ipv4/tcp_ipv4.c               |   6 +-
 net/ipv4/tcp_minisocks.c          |   4 +-
 net/ipv4/tcp_output.c             |   4 +-
 10 files changed, 261 insertions(+), 16 deletions(-)
 create mode 100644 net/ipv4/bpf_tcp_ca.c

diff --git a/include/linux/filter.h b/include/linux/filter.h
index 70e6dd960bca..a366a0b64a57 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h
@@ -843,6 +843,8 @@ int bpf_prog_create(struct bpf_prog **pfp, struct sock_fprog_kern *fprog);
 int bpf_prog_create_from_user(struct bpf_prog **pfp, struct sock_fprog *fprog,
 			      bpf_aux_classic_check_t trans, bool save_orig);
 void bpf_prog_destroy(struct bpf_prog *fp);
+const struct bpf_func_proto *
+bpf_base_func_proto(enum bpf_func_id func_id);
 
 int sk_attach_filter(struct sock_fprog *fprog, struct sock *sk);
 int sk_attach_bpf(u32 ufd, struct sock *sk);
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 7df37e2fddca..9dd975be7fdf 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1007,6 +1007,7 @@ enum tcp_ca_ack_event_flags {
 #define TCP_CONG_NON_RESTRICTED 0x1
 /* Requires ECN/ECT set on all packets */
 #define TCP_CONG_NEEDS_ECN	0x2
+#define TCP_CONG_MASK	(TCP_CONG_NON_RESTRICTED | TCP_CONG_NEEDS_ECN)
 
 union tcp_cc_info;
 
@@ -1101,6 +1102,7 @@ u32 tcp_reno_undo_cwnd(struct sock *sk);
 void tcp_reno_cong_avoid(struct sock *sk, u32 ack, u32 acked);
 extern struct tcp_congestion_ops tcp_reno;
 
+struct tcp_congestion_ops *tcp_ca_find(const char *name);
 struct tcp_congestion_ops *tcp_ca_find_key(u32 key);
 u32 tcp_ca_get_key_by_name(struct net *net, const char *name, bool *ecn_ca);
 #ifdef CONFIG_INET
diff --git a/kernel/bpf/bpf_struct_ops_types.h b/kernel/bpf/bpf_struct_ops_types.h
index 7bb13ff49ec2..066d83ea1c99 100644
--- a/kernel/bpf/bpf_struct_ops_types.h
+++ b/kernel/bpf/bpf_struct_ops_types.h
@@ -1,4 +1,9 @@
 /* SPDX-License-Identifier: GPL-2.0 */
 /* internal file - do not include directly */
 
-/* To be filled in a later patch */
+#ifdef CONFIG_BPF_JIT
+#ifdef CONFIG_INET
+#include <net/tcp.h>
+BPF_STRUCT_OPS_TYPE(tcp_congestion_ops)
+#endif
+#endif
diff --git a/net/core/filter.c b/net/core/filter.c
index 42fd17c48c5f..a702761ef369 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -5935,7 +5935,7 @@ bool bpf_helper_changes_pkt_data(void *func)
 	return false;
 }
 
-static const struct bpf_func_proto *
+const struct bpf_func_proto *
 bpf_base_func_proto(enum bpf_func_id func_id)
 {
 	switch (func_id) {
diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile
index d57ecfaf89d4..9d97bace13c8 100644
--- a/net/ipv4/Makefile
+++ b/net/ipv4/Makefile
@@ -65,3 +65,7 @@ obj-$(CONFIG_NETLABEL) += cipso_ipv4.o
 
 obj-$(CONFIG_XFRM) += xfrm4_policy.o xfrm4_state.o xfrm4_input.o \
 		      xfrm4_output.o xfrm4_protocol.o
+
+ifeq ($(CONFIG_BPF_JIT),y)
+obj-$(CONFIG_BPF_SYSCALL) += bpf_tcp_ca.o
+endif
diff --git a/net/ipv4/bpf_tcp_ca.c b/net/ipv4/bpf_tcp_ca.c
new file mode 100644
index 000000000000..9c7745b958d7
--- /dev/null
+++ b/net/ipv4/bpf_tcp_ca.c
@@ -0,0 +1,230 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2019 Facebook  */
+
+#include <linux/types.h>
+#include <linux/bpf_verifier.h>
+#include <linux/bpf.h>
+#include <linux/btf.h>
+#include <linux/filter.h>
+#include <net/tcp.h>
+
+static u32 optional_ops[] = {
+	offsetof(struct tcp_congestion_ops, init),
+	offsetof(struct tcp_congestion_ops, release),
+	offsetof(struct tcp_congestion_ops, set_state),
+	offsetof(struct tcp_congestion_ops, cwnd_event),
+	offsetof(struct tcp_congestion_ops, in_ack_event),
+	offsetof(struct tcp_congestion_ops, pkts_acked),
+	offsetof(struct tcp_congestion_ops, min_tso_segs),
+	offsetof(struct tcp_congestion_ops, sndbuf_expand),
+	offsetof(struct tcp_congestion_ops, cong_control),
+};
+
+static u32 unsupported_ops[] = {
+	offsetof(struct tcp_congestion_ops, get_info),
+};
+
+static const struct btf_type *tcp_sock_type;
+static u32 tcp_sock_id, sock_id;
+
+static int bpf_tcp_ca_init(struct btf *btf)
+{
+	s32 type_id;
+
+	type_id = btf_find_by_name_kind(btf, "sock", BTF_KIND_STRUCT);
+	if (type_id < 0)
+		return -EINVAL;
+	sock_id = type_id;
+
+	type_id = btf_find_by_name_kind(btf, "tcp_sock", BTF_KIND_STRUCT);
+	if (type_id < 0)
+		return -EINVAL;
+	tcp_sock_id = type_id;
+	tcp_sock_type = btf_type_by_id(btf, tcp_sock_id);
+
+	return 0;
+}
+
+static bool is_optional(u32 member_offset)
+{
+	unsigned int i;
+
+	for (i = 0; i < ARRAY_SIZE(optional_ops); i++) {
+		if (member_offset == optional_ops[i])
+			return true;
+	}
+
+	return false;
+}
+
+static bool is_unsupported(u32 member_offset)
+{
+	unsigned int i;
+
+	for (i = 0; i < ARRAY_SIZE(unsupported_ops); i++) {
+		if (member_offset == unsupported_ops[i])
+			return true;
+	}
+
+	return false;
+}
+
+extern struct btf *btf_vmlinux;
+
+static bool bpf_tcp_ca_is_valid_access(int off, int size,
+				       enum bpf_access_type type,
+				       const struct bpf_prog *prog,
+				       struct bpf_insn_access_aux *info)
+{
+	if (off < 0 || off >= sizeof(__u64) * MAX_BPF_FUNC_ARGS)
+		return false;
+	if (type != BPF_READ)
+		return false;
+	if (off % size != 0)
+		return false;
+
+	if (!btf_ctx_access(off, size, type, prog, info))
+		return false;
+
+	if (info->reg_type == PTR_TO_BTF_ID && info->btf_id == sock_id)
+		/* promote it to tcp_sock */
+		info->btf_id = tcp_sock_id;
+
+	return true;
+}
+
+static int bpf_tcp_ca_btf_struct_access(struct bpf_verifier_log *log,
+					const struct btf_type *t, int off,
+					int size, enum bpf_access_type atype,
+					u32 *next_btf_id)
+{
+	size_t end;
+
+	if (atype == BPF_READ)
+		return btf_struct_access(log, t, off, size, atype, next_btf_id);
+
+	if (t != tcp_sock_type) {
+		bpf_log(log, "only read is supported\n");
+		return -EACCES;
+	}
+
+	switch (off) {
+	case bpf_ctx_range(struct inet_connection_sock, icsk_ca_priv):
+		end = offsetofend(struct inet_connection_sock, icsk_ca_priv);
+		break;
+	case offsetof(struct inet_connection_sock, icsk_ack.pending):
+		end = offsetofend(struct inet_connection_sock,
+				  icsk_ack.pending);
+		break;
+	case offsetof(struct tcp_sock, snd_cwnd):
+		end = offsetofend(struct tcp_sock, snd_cwnd);
+		break;
+	case offsetof(struct tcp_sock, snd_cwnd_cnt):
+		end = offsetofend(struct tcp_sock, snd_cwnd_cnt);
+		break;
+	case offsetof(struct tcp_sock, snd_ssthresh):
+		end = offsetofend(struct tcp_sock, snd_ssthresh);
+		break;
+	case offsetof(struct tcp_sock, ecn_flags):
+		end = offsetofend(struct tcp_sock, ecn_flags);
+		break;
+	default:
+		bpf_log(log, "no write support to tcp_sock at off %d\n", off);
+		return -EACCES;
+	}
+
+	if (off + size > end) {
+		bpf_log(log,
+			"write access at off %d with size %d beyond the member of tcp_sock ended at %zu\n",
+			off, size, end);
+		return -EACCES;
+	}
+
+	return NOT_INIT;
+}
+
+static const struct bpf_func_proto *
+bpf_tcp_ca_get_func_proto(enum bpf_func_id func_id,
+			  const struct bpf_prog *prog)
+{
+	return bpf_base_func_proto(func_id);
+}
+
+static const struct bpf_verifier_ops bpf_tcp_ca_verifier_ops = {
+	.get_func_proto		= bpf_tcp_ca_get_func_proto,
+	.is_valid_access	= bpf_tcp_ca_is_valid_access,
+	.btf_struct_access	= bpf_tcp_ca_btf_struct_access,
+};
+
+static int bpf_tcp_ca_init_member(const struct btf_type *t,
+				  const struct btf_member *member,
+				  void *kdata, const void *udata)
+{
+	const struct tcp_congestion_ops *utcp_ca;
+	struct tcp_congestion_ops *tcp_ca;
+	size_t tcp_ca_name_len;
+	int prog_fd;
+	u32 moff;
+
+	utcp_ca = (const struct tcp_congestion_ops *)udata;
+	tcp_ca = (struct tcp_congestion_ops *)kdata;
+
+	moff = btf_member_bit_offset(t, member) / 8;
+	switch (moff) {
+	case offsetof(struct tcp_congestion_ops, flags):
+		if (utcp_ca->flags & ~TCP_CONG_MASK)
+			return -EINVAL;
+		tcp_ca->flags = utcp_ca->flags;
+		return 1;
+	case offsetof(struct tcp_congestion_ops, name):
+		tcp_ca_name_len = strnlen(utcp_ca->name, sizeof(utcp_ca->name));
+		if (!tcp_ca_name_len ||
+		    tcp_ca_name_len == sizeof(utcp_ca->name))
+			return -EINVAL;
+		if (tcp_ca_find(utcp_ca->name))
+			return -EEXIST;
+		memcpy(tcp_ca->name, utcp_ca->name, sizeof(tcp_ca->name));
+		return 1;
+	}
+
+	if (!btf_type_resolve_func_ptr(btf_vmlinux, member->type, NULL))
+		return 0;
+
+	/* Ensure bpf_prog is provided for compulsory func ptr */
+	prog_fd = (int)(*(unsigned long *)(udata + moff));
+	if (!prog_fd && !is_optional(moff) && !is_unsupported(moff))
+		return -EINVAL;
+
+	return 0;
+}
+
+static int bpf_tcp_ca_check_member(const struct btf_type *t,
+				   const struct btf_member *member)
+{
+	if (is_unsupported(btf_member_bit_offset(t, member) / 8))
+		return -ENOTSUPP;
+	return 0;
+}
+
+static int bpf_tcp_ca_reg(void *kdata)
+{
+	return tcp_register_congestion_control(kdata);
+}
+
+static void bpf_tcp_ca_unreg(void *kdata)
+{
+	tcp_unregister_congestion_control(kdata);
+}
+
+/* Avoid sparse warning.  It is only used in bpf_struct_ops.c. */
+extern struct bpf_struct_ops bpf_tcp_congestion_ops;
+
+struct bpf_struct_ops bpf_tcp_congestion_ops = {
+	.verifier_ops = &bpf_tcp_ca_verifier_ops,
+	.reg = bpf_tcp_ca_reg,
+	.unreg = bpf_tcp_ca_unreg,
+	.check_member = bpf_tcp_ca_check_member,
+	.init_member = bpf_tcp_ca_init_member,
+	.init = bpf_tcp_ca_init,
+	.name = "tcp_congestion_ops",
+};
diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c
index 3737ec096650..3172e31987be 100644
--- a/net/ipv4/tcp_cong.c
+++ b/net/ipv4/tcp_cong.c
@@ -21,7 +21,7 @@ static DEFINE_SPINLOCK(tcp_cong_list_lock);
 static LIST_HEAD(tcp_cong_list);
 
 /* Simple linear search, don't expect many entries! */
-static struct tcp_congestion_ops *tcp_ca_find(const char *name)
+struct tcp_congestion_ops *tcp_ca_find(const char *name)
 {
 	struct tcp_congestion_ops *e;
 
@@ -162,7 +162,7 @@ void tcp_assign_congestion_control(struct sock *sk)
 
 	rcu_read_lock();
 	ca = rcu_dereference(net->ipv4.tcp_congestion_control);
-	if (unlikely(!try_module_get(ca->owner)))
+	if (unlikely(!bpf_try_module_get(ca, ca->owner)))
 		ca = &tcp_reno;
 	icsk->icsk_ca_ops = ca;
 	rcu_read_unlock();
@@ -208,7 +208,7 @@ void tcp_cleanup_congestion_control(struct sock *sk)
 
 	if (icsk->icsk_ca_ops->release)
 		icsk->icsk_ca_ops->release(sk);
-	module_put(icsk->icsk_ca_ops->owner);
+	bpf_module_put(icsk->icsk_ca_ops, icsk->icsk_ca_ops->owner);
 }
 
 /* Used by sysctl to change default congestion control */
@@ -222,12 +222,12 @@ int tcp_set_default_congestion_control(struct net *net, const char *name)
 	ca = tcp_ca_find_autoload(net, name);
 	if (!ca) {
 		ret = -ENOENT;
-	} else if (!try_module_get(ca->owner)) {
+	} else if (!bpf_try_module_get(ca, ca->owner)) {
 		ret = -EBUSY;
 	} else {
 		prev = xchg(&net->ipv4.tcp_congestion_control, ca);
 		if (prev)
-			module_put(prev->owner);
+			bpf_module_put(prev, prev->owner);
 
 		ca->flags |= TCP_CONG_NON_RESTRICTED;
 		ret = 0;
@@ -366,19 +366,19 @@ int tcp_set_congestion_control(struct sock *sk, const char *name, bool load,
 	} else if (!load) {
 		const struct tcp_congestion_ops *old_ca = icsk->icsk_ca_ops;
 
-		if (try_module_get(ca->owner)) {
+		if (bpf_try_module_get(ca, ca->owner)) {
 			if (reinit) {
 				tcp_reinit_congestion_control(sk, ca);
 			} else {
 				icsk->icsk_ca_ops = ca;
-				module_put(old_ca->owner);
+				bpf_module_put(old_ca, old_ca->owner);
 			}
 		} else {
 			err = -EBUSY;
 		}
 	} else if (!((ca->flags & TCP_CONG_NON_RESTRICTED) || cap_net_admin)) {
 		err = -EPERM;
-	} else if (!try_module_get(ca->owner)) {
+	} else if (!bpf_try_module_get(ca, ca->owner)) {
 		err = -EBUSY;
 	} else {
 		tcp_reinit_congestion_control(sk, ca);
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 4adac9c75343..317ccca548a2 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2678,7 +2678,8 @@ static void __net_exit tcp_sk_exit(struct net *net)
 	int cpu;
 
 	if (net->ipv4.tcp_congestion_control)
-		module_put(net->ipv4.tcp_congestion_control->owner);
+		bpf_module_put(net->ipv4.tcp_congestion_control,
+			       net->ipv4.tcp_congestion_control->owner);
 
 	for_each_possible_cpu(cpu)
 		inet_ctl_sock_destroy(*per_cpu_ptr(net->ipv4.tcp_sk, cpu));
@@ -2785,7 +2786,8 @@ static int __net_init tcp_sk_init(struct net *net)
 
 	/* Reno is always built in */
 	if (!net_eq(net, &init_net) &&
-	    try_module_get(init_net.ipv4.tcp_congestion_control->owner))
+	    bpf_try_module_get(init_net.ipv4.tcp_congestion_control,
+			       init_net.ipv4.tcp_congestion_control->owner))
 		net->ipv4.tcp_congestion_control = init_net.ipv4.tcp_congestion_control;
 	else
 		net->ipv4.tcp_congestion_control = &tcp_reno;
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index c802bc80c400..ad3b56d9fa71 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -414,7 +414,7 @@ void tcp_ca_openreq_child(struct sock *sk, const struct dst_entry *dst)
 
 		rcu_read_lock();
 		ca = tcp_ca_find_key(ca_key);
-		if (likely(ca && try_module_get(ca->owner))) {
+		if (likely(ca && bpf_try_module_get(ca, ca->owner))) {
 			icsk->icsk_ca_dst_locked = tcp_ca_dst_locked(dst);
 			icsk->icsk_ca_ops = ca;
 			ca_got_dst = true;
@@ -425,7 +425,7 @@ void tcp_ca_openreq_child(struct sock *sk, const struct dst_entry *dst)
 	/* If no valid choice made yet, assign current system default ca. */
 	if (!ca_got_dst &&
 	    (!icsk->icsk_ca_setsockopt ||
-	     !try_module_get(icsk->icsk_ca_ops->owner)))
+	     !bpf_try_module_get(icsk->icsk_ca_ops, icsk->icsk_ca_ops->owner)))
 		tcp_assign_congestion_control(sk);
 
 	tcp_set_ca_state(sk, TCP_CA_Open);
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 58c92a7d671c..377cfab422df 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -3368,8 +3368,8 @@ static void tcp_ca_dst_init(struct sock *sk, const struct dst_entry *dst)
 
 	rcu_read_lock();
 	ca = tcp_ca_find_key(ca_key);
-	if (likely(ca && try_module_get(ca->owner))) {
-		module_put(icsk->icsk_ca_ops->owner);
+	if (likely(ca && bpf_try_module_get(ca, ca->owner))) {
+		bpf_module_put(icsk->icsk_ca_ops, icsk->icsk_ca_ops->owner);
 		icsk->icsk_ca_dst_locked = tcp_ca_dst_locked(dst);
 		icsk->icsk_ca_ops = ca;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 206057fe020ac5c037d5e2dd6562a9bd216ec765 Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <kafai@fb.com>
Date: Wed, 8 Jan 2020 16:45:51 -0800
Subject: bpf: Add BPF_FUNC_tcp_send_ack helper

Add a helper to send out a tcp-ack.  It will be used in the later
bpf_dctcp implementation that requires to send out an ack
when the CE state changed.

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20200109004551.3900448-1-kafai@fb.com
---
 include/uapi/linux/bpf.h | 11 ++++++++++-
 net/ipv4/bpf_tcp_ca.c    | 24 +++++++++++++++++++++++-
 2 files changed, 33 insertions(+), 2 deletions(-)

diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 38059880963e..2d6a2e572f56 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -2837,6 +2837,14 @@ union bpf_attr {
  * 	Return
  * 		On success, the strictly positive length of the string,	including
  * 		the trailing NUL character. On error, a negative value.
+ *
+ * int bpf_tcp_send_ack(void *tp, u32 rcv_nxt)
+ *	Description
+ *		Send out a tcp-ack. *tp* is the in-kernel struct tcp_sock.
+ *		*rcv_nxt* is the ack_seq to be sent out.
+ *	Return
+ *		0 on success, or a negative error in case of failure.
+ *
  */
 #define __BPF_FUNC_MAPPER(FN)		\
 	FN(unspec),			\
@@ -2954,7 +2962,8 @@ union bpf_attr {
 	FN(probe_read_user),		\
 	FN(probe_read_kernel),		\
 	FN(probe_read_user_str),	\
-	FN(probe_read_kernel_str),
+	FN(probe_read_kernel_str),	\
+	FN(tcp_send_ack),
 
 /* integer value in 'imm' field of BPF_CALL instruction selects which helper
  * function eBPF program intends to call
diff --git a/net/ipv4/bpf_tcp_ca.c b/net/ipv4/bpf_tcp_ca.c
index 9c7745b958d7..574972bc7299 100644
--- a/net/ipv4/bpf_tcp_ca.c
+++ b/net/ipv4/bpf_tcp_ca.c
@@ -143,11 +143,33 @@ static int bpf_tcp_ca_btf_struct_access(struct bpf_verifier_log *log,
 	return NOT_INIT;
 }
 
+BPF_CALL_2(bpf_tcp_send_ack, struct tcp_sock *, tp, u32, rcv_nxt)
+{
+	/* bpf_tcp_ca prog cannot have NULL tp */
+	__tcp_send_ack((struct sock *)tp, rcv_nxt);
+	return 0;
+}
+
+static const struct bpf_func_proto bpf_tcp_send_ack_proto = {
+	.func		= bpf_tcp_send_ack,
+	.gpl_only	= false,
+	/* In case we want to report error later */
+	.ret_type	= RET_INTEGER,
+	.arg1_type	= ARG_PTR_TO_BTF_ID,
+	.arg2_type	= ARG_ANYTHING,
+	.btf_id		= &tcp_sock_id,
+};
+
 static const struct bpf_func_proto *
 bpf_tcp_ca_get_func_proto(enum bpf_func_id func_id,
 			  const struct bpf_prog *prog)
 {
-	return bpf_base_func_proto(func_id);
+	switch (func_id) {
+	case BPF_FUNC_tcp_send_ack:
+		return &bpf_tcp_send_ack_proto;
+	default:
+		return bpf_base_func_proto(func_id);
+	}
 }
 
 static const struct bpf_verifier_ops bpf_tcp_ca_verifier_ops = {
-- 
cgit v1.2.3-59-g8ed1b


From 17328d618c17e0c8deab892ac227efcd55957e4d Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <kafai@fb.com>
Date: Wed, 8 Jan 2020 16:35:12 -0800
Subject: bpf: Synch uapi bpf.h to tools/

This patch sync uapi bpf.h to tools/

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200109003512.3856559-1-kafai@fb.com
---
 tools/include/uapi/linux/bpf.h | 19 +++++++++++++++++--
 1 file changed, 17 insertions(+), 2 deletions(-)

diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index 7df436da542d..2d6a2e572f56 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -136,6 +136,7 @@ enum bpf_map_type {
 	BPF_MAP_TYPE_STACK,
 	BPF_MAP_TYPE_SK_STORAGE,
 	BPF_MAP_TYPE_DEVMAP_HASH,
+	BPF_MAP_TYPE_STRUCT_OPS,
 };
 
 /* Note that tracing related programs such as
@@ -174,6 +175,7 @@ enum bpf_prog_type {
 	BPF_PROG_TYPE_RAW_TRACEPOINT_WRITABLE,
 	BPF_PROG_TYPE_CGROUP_SOCKOPT,
 	BPF_PROG_TYPE_TRACING,
+	BPF_PROG_TYPE_STRUCT_OPS,
 };
 
 enum bpf_attach_type {
@@ -397,6 +399,10 @@ union bpf_attr {
 		__u32	btf_fd;		/* fd pointing to a BTF type data */
 		__u32	btf_key_type_id;	/* BTF type_id of the key */
 		__u32	btf_value_type_id;	/* BTF type_id of the value */
+		__u32	btf_vmlinux_value_type_id;/* BTF type_id of a kernel-
+						   * struct stored as the
+						   * map value
+						   */
 	};
 
 	struct { /* anonymous struct used by BPF_MAP_*_ELEM commands */
@@ -2831,6 +2837,14 @@ union bpf_attr {
  * 	Return
  * 		On success, the strictly positive length of the string,	including
  * 		the trailing NUL character. On error, a negative value.
+ *
+ * int bpf_tcp_send_ack(void *tp, u32 rcv_nxt)
+ *	Description
+ *		Send out a tcp-ack. *tp* is the in-kernel struct tcp_sock.
+ *		*rcv_nxt* is the ack_seq to be sent out.
+ *	Return
+ *		0 on success, or a negative error in case of failure.
+ *
  */
 #define __BPF_FUNC_MAPPER(FN)		\
 	FN(unspec),			\
@@ -2948,7 +2962,8 @@ union bpf_attr {
 	FN(probe_read_user),		\
 	FN(probe_read_kernel),		\
 	FN(probe_read_user_str),	\
-	FN(probe_read_kernel_str),
+	FN(probe_read_kernel_str),	\
+	FN(tcp_send_ack),
 
 /* integer value in 'imm' field of BPF_CALL instruction selects which helper
  * function eBPF program intends to call
@@ -3349,7 +3364,7 @@ struct bpf_map_info {
 	__u32 map_flags;
 	char  name[BPF_OBJ_NAME_LEN];
 	__u32 ifindex;
-	__u32 :32;
+	__u32 btf_vmlinux_value_type_id;
 	__u64 netns_dev;
 	__u64 netns_ino;
 	__u32 btf_id;
-- 
cgit v1.2.3-59-g8ed1b


From 590a0088825016ca7ec53f1aef7e84e1211778d8 Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <kafai@fb.com>
Date: Wed, 8 Jan 2020 16:35:14 -0800
Subject: bpf: libbpf: Add STRUCT_OPS support

This patch adds BPF STRUCT_OPS support to libbpf.

The only sec_name convention is SEC(".struct_ops") to identify the
struct_ops implemented in BPF,
e.g. To implement a tcp_congestion_ops:

SEC(".struct_ops")
struct tcp_congestion_ops dctcp = {
	.init           = (void *)dctcp_init,  /* <-- a bpf_prog */
	/* ... some more func prts ... */
	.name           = "bpf_dctcp",
};

Each struct_ops is defined as a global variable under SEC(".struct_ops")
as above.  libbpf creates a map for each variable and the variable name
is the map's name.  Multiple struct_ops is supported under
SEC(".struct_ops").

In the bpf_object__open phase, libbpf will look for the SEC(".struct_ops")
section and find out what is the btf-type the struct_ops is
implementing.  Note that the btf-type here is referring to
a type in the bpf_prog.o's btf.  A "struct bpf_map" is added
by bpf_object__add_map() as other maps do.  It will then
collect (through SHT_REL) where are the bpf progs that the
func ptrs are referring to.  No btf_vmlinux is needed in
the open phase.

In the bpf_object__load phase, the map-fields, which depend
on the btf_vmlinux, are initialized (in bpf_map__init_kern_struct_ops()).
It will also set the prog->type, prog->attach_btf_id, and
prog->expected_attach_type.  Thus, the prog's properties do
not rely on its section name.
[ Currently, the bpf_prog's btf-type ==> btf_vmlinux's btf-type matching
  process is as simple as: member-name match + btf-kind match + size match.
  If these matching conditions fail, libbpf will reject.
  The current targeting support is "struct tcp_congestion_ops" which
  most of its members are function pointers.
  The member ordering of the bpf_prog's btf-type can be different from
  the btf_vmlinux's btf-type. ]

Then, all obj->maps are created as usual (in bpf_object__create_maps()).

Once the maps are created and prog's properties are all set,
the libbpf will proceed to load all the progs.

bpf_map__attach_struct_ops() is added to register a struct_ops
map to a kernel subsystem.

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200109003514.3856730-1-kafai@fb.com
---
 tools/lib/bpf/bpf.c           |  10 +-
 tools/lib/bpf/bpf.h           |   5 +-
 tools/lib/bpf/libbpf.c        | 649 +++++++++++++++++++++++++++++++++++++++++-
 tools/lib/bpf/libbpf.h        |   5 +-
 tools/lib/bpf/libbpf.map      |   3 +
 tools/lib/bpf/libbpf_probes.c |   2 +
 6 files changed, 661 insertions(+), 13 deletions(-)

diff --git a/tools/lib/bpf/bpf.c b/tools/lib/bpf/bpf.c
index a787d53699c8..b0ecbe9ef2d4 100644
--- a/tools/lib/bpf/bpf.c
+++ b/tools/lib/bpf/bpf.c
@@ -95,7 +95,11 @@ int bpf_create_map_xattr(const struct bpf_create_map_attr *create_attr)
 	attr.btf_key_type_id = create_attr->btf_key_type_id;
 	attr.btf_value_type_id = create_attr->btf_value_type_id;
 	attr.map_ifindex = create_attr->map_ifindex;
-	attr.inner_map_fd = create_attr->inner_map_fd;
+	if (attr.map_type == BPF_MAP_TYPE_STRUCT_OPS)
+		attr.btf_vmlinux_value_type_id =
+			create_attr->btf_vmlinux_value_type_id;
+	else
+		attr.inner_map_fd = create_attr->inner_map_fd;
 
 	return sys_bpf(BPF_MAP_CREATE, &attr, sizeof(attr));
 }
@@ -228,7 +232,9 @@ int bpf_load_program_xattr(const struct bpf_load_program_attr *load_attr,
 	memset(&attr, 0, sizeof(attr));
 	attr.prog_type = load_attr->prog_type;
 	attr.expected_attach_type = load_attr->expected_attach_type;
-	if (attr.prog_type == BPF_PROG_TYPE_TRACING) {
+	if (attr.prog_type == BPF_PROG_TYPE_STRUCT_OPS) {
+		attr.attach_btf_id = load_attr->attach_btf_id;
+	} else if (attr.prog_type == BPF_PROG_TYPE_TRACING) {
 		attr.attach_btf_id = load_attr->attach_btf_id;
 		attr.attach_prog_fd = load_attr->attach_prog_fd;
 	} else {
diff --git a/tools/lib/bpf/bpf.h b/tools/lib/bpf/bpf.h
index f0ab8519986e..56341d117e5b 100644
--- a/tools/lib/bpf/bpf.h
+++ b/tools/lib/bpf/bpf.h
@@ -46,7 +46,10 @@ struct bpf_create_map_attr {
 	__u32 btf_key_type_id;
 	__u32 btf_value_type_id;
 	__u32 map_ifindex;
-	__u32 inner_map_fd;
+	union {
+		__u32 inner_map_fd;
+		__u32 btf_vmlinux_value_type_id;
+	};
 };
 
 LIBBPF_API int
diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 7513165b104f..35a4422ef655 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -70,6 +70,13 @@
 
 #define __printf(a, b)	__attribute__((format(printf, a, b)))
 
+static struct btf *bpf_find_kernel_btf(void);
+static struct bpf_map *bpf_object__add_map(struct bpf_object *obj);
+static struct bpf_program *bpf_object__find_prog_by_idx(struct bpf_object *obj,
+							int idx);
+static const struct btf_type *
+skip_mods_and_typedefs(const struct btf *btf, __u32 id, __u32 *res_id);
+
 static int __base_pr(enum libbpf_print_level level, const char *format,
 		     va_list args)
 {
@@ -229,10 +236,32 @@ struct bpf_program {
 	__u32 prog_flags;
 };
 
+struct bpf_struct_ops {
+	const char *tname;
+	const struct btf_type *type;
+	struct bpf_program **progs;
+	__u32 *kern_func_off;
+	/* e.g. struct tcp_congestion_ops in bpf_prog's btf format */
+	void *data;
+	/* e.g. struct bpf_struct_ops_tcp_congestion_ops in
+	 *      btf_vmlinux's format.
+	 * struct bpf_struct_ops_tcp_congestion_ops {
+	 *	[... some other kernel fields ...]
+	 *	struct tcp_congestion_ops data;
+	 * }
+	 * kern_vdata-size == sizeof(struct bpf_struct_ops_tcp_congestion_ops)
+	 * bpf_map__init_kern_struct_ops() will populate the "kern_vdata"
+	 * from "data".
+	 */
+	void *kern_vdata;
+	__u32 type_id;
+};
+
 #define DATA_SEC ".data"
 #define BSS_SEC ".bss"
 #define RODATA_SEC ".rodata"
 #define KCONFIG_SEC ".kconfig"
+#define STRUCT_OPS_SEC ".struct_ops"
 
 enum libbpf_map_type {
 	LIBBPF_MAP_UNSPEC,
@@ -259,10 +288,12 @@ struct bpf_map {
 	struct bpf_map_def def;
 	__u32 btf_key_type_id;
 	__u32 btf_value_type_id;
+	__u32 btf_vmlinux_value_type_id;
 	void *priv;
 	bpf_map_clear_priv_t clear_priv;
 	enum libbpf_map_type libbpf_type;
 	void *mmaped;
+	struct bpf_struct_ops *st_ops;
 	char *pin_path;
 	bool pinned;
 	bool reused;
@@ -326,6 +357,7 @@ struct bpf_object {
 		Elf_Data *data;
 		Elf_Data *rodata;
 		Elf_Data *bss;
+		Elf_Data *st_ops_data;
 		size_t strtabidx;
 		struct {
 			GElf_Shdr shdr;
@@ -339,6 +371,7 @@ struct bpf_object {
 		int data_shndx;
 		int rodata_shndx;
 		int bss_shndx;
+		int st_ops_shndx;
 	} efile;
 	/*
 	 * All loaded bpf_object is linked in a list, which is
@@ -566,6 +599,359 @@ static __u32 get_kernel_version(void)
 	return KERNEL_VERSION(major, minor, patch);
 }
 
+static const struct btf_member *
+find_member_by_offset(const struct btf_type *t, __u32 bit_offset)
+{
+	struct btf_member *m;
+	int i;
+
+	for (i = 0, m = btf_members(t); i < btf_vlen(t); i++, m++) {
+		if (btf_member_bit_offset(t, i) == bit_offset)
+			return m;
+	}
+
+	return NULL;
+}
+
+static const struct btf_member *
+find_member_by_name(const struct btf *btf, const struct btf_type *t,
+		    const char *name)
+{
+	struct btf_member *m;
+	int i;
+
+	for (i = 0, m = btf_members(t); i < btf_vlen(t); i++, m++) {
+		if (!strcmp(btf__name_by_offset(btf, m->name_off), name))
+			return m;
+	}
+
+	return NULL;
+}
+
+#define STRUCT_OPS_VALUE_PREFIX "bpf_struct_ops_"
+#define STRUCT_OPS_VALUE_PREFIX_LEN (sizeof(STRUCT_OPS_VALUE_PREFIX) - 1)
+
+static int
+find_struct_ops_kern_types(const struct btf *btf, const char *tname,
+			   const struct btf_type **type, __u32 *type_id,
+			   const struct btf_type **vtype, __u32 *vtype_id,
+			   const struct btf_member **data_member)
+{
+	const struct btf_type *kern_type, *kern_vtype;
+	const struct btf_member *kern_data_member;
+	__s32 kern_vtype_id, kern_type_id;
+	char vtname[128] = STRUCT_OPS_VALUE_PREFIX;
+	__u32 i;
+
+	kern_type_id = btf__find_by_name_kind(btf, tname, BTF_KIND_STRUCT);
+	if (kern_type_id < 0) {
+		pr_warn("struct_ops init_kern: struct %s is not found in kernel BTF\n",
+			tname);
+		return kern_type_id;
+	}
+	kern_type = btf__type_by_id(btf, kern_type_id);
+
+	/* Find the corresponding "map_value" type that will be used
+	 * in map_update(BPF_MAP_TYPE_STRUCT_OPS).  For example,
+	 * find "struct bpf_struct_ops_tcp_congestion_ops" from the
+	 * btf_vmlinux.
+	 */
+	strncat(vtname + STRUCT_OPS_VALUE_PREFIX_LEN, tname,
+		sizeof(vtname) - STRUCT_OPS_VALUE_PREFIX_LEN - 1);
+	kern_vtype_id = btf__find_by_name_kind(btf, vtname,
+					       BTF_KIND_STRUCT);
+	if (kern_vtype_id < 0) {
+		pr_warn("struct_ops init_kern: struct %s is not found in kernel BTF\n",
+			vtname);
+		return kern_vtype_id;
+	}
+	kern_vtype = btf__type_by_id(btf, kern_vtype_id);
+
+	/* Find "struct tcp_congestion_ops" from
+	 * struct bpf_struct_ops_tcp_congestion_ops {
+	 *	[ ... ]
+	 *	struct tcp_congestion_ops data;
+	 * }
+	 */
+	kern_data_member = btf_members(kern_vtype);
+	for (i = 0; i < btf_vlen(kern_vtype); i++, kern_data_member++) {
+		if (kern_data_member->type == kern_type_id)
+			break;
+	}
+	if (i == btf_vlen(kern_vtype)) {
+		pr_warn("struct_ops init_kern: struct %s data is not found in struct %s\n",
+			tname, vtname);
+		return -EINVAL;
+	}
+
+	*type = kern_type;
+	*type_id = kern_type_id;
+	*vtype = kern_vtype;
+	*vtype_id = kern_vtype_id;
+	*data_member = kern_data_member;
+
+	return 0;
+}
+
+static bool bpf_map__is_struct_ops(const struct bpf_map *map)
+{
+	return map->def.type == BPF_MAP_TYPE_STRUCT_OPS;
+}
+
+/* Init the map's fields that depend on kern_btf */
+static int bpf_map__init_kern_struct_ops(struct bpf_map *map,
+					 const struct btf *btf,
+					 const struct btf *kern_btf)
+{
+	const struct btf_member *member, *kern_member, *kern_data_member;
+	const struct btf_type *type, *kern_type, *kern_vtype;
+	__u32 i, kern_type_id, kern_vtype_id, kern_data_off;
+	struct bpf_struct_ops *st_ops;
+	void *data, *kern_data;
+	const char *tname;
+	int err;
+
+	st_ops = map->st_ops;
+	type = st_ops->type;
+	tname = st_ops->tname;
+	err = find_struct_ops_kern_types(kern_btf, tname,
+					 &kern_type, &kern_type_id,
+					 &kern_vtype, &kern_vtype_id,
+					 &kern_data_member);
+	if (err)
+		return err;
+
+	pr_debug("struct_ops init_kern %s: type_id:%u kern_type_id:%u kern_vtype_id:%u\n",
+		 map->name, st_ops->type_id, kern_type_id, kern_vtype_id);
+
+	map->def.value_size = kern_vtype->size;
+	map->btf_vmlinux_value_type_id = kern_vtype_id;
+
+	st_ops->kern_vdata = calloc(1, kern_vtype->size);
+	if (!st_ops->kern_vdata)
+		return -ENOMEM;
+
+	data = st_ops->data;
+	kern_data_off = kern_data_member->offset / 8;
+	kern_data = st_ops->kern_vdata + kern_data_off;
+
+	member = btf_members(type);
+	for (i = 0; i < btf_vlen(type); i++, member++) {
+		const struct btf_type *mtype, *kern_mtype;
+		__u32 mtype_id, kern_mtype_id;
+		void *mdata, *kern_mdata;
+		__s64 msize, kern_msize;
+		__u32 moff, kern_moff;
+		__u32 kern_member_idx;
+		const char *mname;
+
+		mname = btf__name_by_offset(btf, member->name_off);
+		kern_member = find_member_by_name(kern_btf, kern_type, mname);
+		if (!kern_member) {
+			pr_warn("struct_ops init_kern %s: Cannot find member %s in kernel BTF\n",
+				map->name, mname);
+			return -ENOTSUP;
+		}
+
+		kern_member_idx = kern_member - btf_members(kern_type);
+		if (btf_member_bitfield_size(type, i) ||
+		    btf_member_bitfield_size(kern_type, kern_member_idx)) {
+			pr_warn("struct_ops init_kern %s: bitfield %s is not supported\n",
+				map->name, mname);
+			return -ENOTSUP;
+		}
+
+		moff = member->offset / 8;
+		kern_moff = kern_member->offset / 8;
+
+		mdata = data + moff;
+		kern_mdata = kern_data + kern_moff;
+
+		mtype = skip_mods_and_typedefs(btf, member->type, &mtype_id);
+		kern_mtype = skip_mods_and_typedefs(kern_btf, kern_member->type,
+						    &kern_mtype_id);
+		if (BTF_INFO_KIND(mtype->info) !=
+		    BTF_INFO_KIND(kern_mtype->info)) {
+			pr_warn("struct_ops init_kern %s: Unmatched member type %s %u != %u(kernel)\n",
+				map->name, mname, BTF_INFO_KIND(mtype->info),
+				BTF_INFO_KIND(kern_mtype->info));
+			return -ENOTSUP;
+		}
+
+		if (btf_is_ptr(mtype)) {
+			struct bpf_program *prog;
+
+			mtype = skip_mods_and_typedefs(btf, mtype->type, &mtype_id);
+			kern_mtype = skip_mods_and_typedefs(kern_btf,
+							    kern_mtype->type,
+							    &kern_mtype_id);
+			if (!btf_is_func_proto(mtype) ||
+			    !btf_is_func_proto(kern_mtype)) {
+				pr_warn("struct_ops init_kern %s: non func ptr %s is not supported\n",
+					map->name, mname);
+				return -ENOTSUP;
+			}
+
+			prog = st_ops->progs[i];
+			if (!prog) {
+				pr_debug("struct_ops init_kern %s: func ptr %s is not set\n",
+					 map->name, mname);
+				continue;
+			}
+
+			prog->attach_btf_id = kern_type_id;
+			prog->expected_attach_type = kern_member_idx;
+
+			st_ops->kern_func_off[i] = kern_data_off + kern_moff;
+
+			pr_debug("struct_ops init_kern %s: func ptr %s is set to prog %s from data(+%u) to kern_data(+%u)\n",
+				 map->name, mname, prog->name, moff,
+				 kern_moff);
+
+			continue;
+		}
+
+		msize = btf__resolve_size(btf, mtype_id);
+		kern_msize = btf__resolve_size(kern_btf, kern_mtype_id);
+		if (msize < 0 || kern_msize < 0 || msize != kern_msize) {
+			pr_warn("struct_ops init_kern %s: Error in size of member %s: %zd != %zd(kernel)\n",
+				map->name, mname, (ssize_t)msize,
+				(ssize_t)kern_msize);
+			return -ENOTSUP;
+		}
+
+		pr_debug("struct_ops init_kern %s: copy %s %u bytes from data(+%u) to kern_data(+%u)\n",
+			 map->name, mname, (unsigned int)msize,
+			 moff, kern_moff);
+		memcpy(kern_mdata, mdata, msize);
+	}
+
+	return 0;
+}
+
+static int bpf_object__init_kern_struct_ops_maps(struct bpf_object *obj)
+{
+	struct btf *kern_btf = NULL;
+	struct bpf_map *map;
+	size_t i;
+	int err;
+
+	for (i = 0; i < obj->nr_maps; i++) {
+		map = &obj->maps[i];
+
+		if (!bpf_map__is_struct_ops(map))
+			continue;
+
+		if (!kern_btf) {
+			kern_btf = bpf_find_kernel_btf();
+			if (IS_ERR(kern_btf))
+				return PTR_ERR(kern_btf);
+		}
+
+		err = bpf_map__init_kern_struct_ops(map, obj->btf, kern_btf);
+		if (err) {
+			btf__free(kern_btf);
+			return err;
+		}
+	}
+
+	btf__free(kern_btf);
+	return 0;
+}
+
+static int bpf_object__init_struct_ops_maps(struct bpf_object *obj)
+{
+	const struct btf_type *type, *datasec;
+	const struct btf_var_secinfo *vsi;
+	struct bpf_struct_ops *st_ops;
+	const char *tname, *var_name;
+	__s32 type_id, datasec_id;
+	const struct btf *btf;
+	struct bpf_map *map;
+	__u32 i;
+
+	if (obj->efile.st_ops_shndx == -1)
+		return 0;
+
+	btf = obj->btf;
+	datasec_id = btf__find_by_name_kind(btf, STRUCT_OPS_SEC,
+					    BTF_KIND_DATASEC);
+	if (datasec_id < 0) {
+		pr_warn("struct_ops init: DATASEC %s not found\n",
+			STRUCT_OPS_SEC);
+		return -EINVAL;
+	}
+
+	datasec = btf__type_by_id(btf, datasec_id);
+	vsi = btf_var_secinfos(datasec);
+	for (i = 0; i < btf_vlen(datasec); i++, vsi++) {
+		type = btf__type_by_id(obj->btf, vsi->type);
+		var_name = btf__name_by_offset(obj->btf, type->name_off);
+
+		type_id = btf__resolve_type(obj->btf, vsi->type);
+		if (type_id < 0) {
+			pr_warn("struct_ops init: Cannot resolve var type_id %u in DATASEC %s\n",
+				vsi->type, STRUCT_OPS_SEC);
+			return -EINVAL;
+		}
+
+		type = btf__type_by_id(obj->btf, type_id);
+		tname = btf__name_by_offset(obj->btf, type->name_off);
+		if (!tname[0]) {
+			pr_warn("struct_ops init: anonymous type is not supported\n");
+			return -ENOTSUP;
+		}
+		if (!btf_is_struct(type)) {
+			pr_warn("struct_ops init: %s is not a struct\n", tname);
+			return -EINVAL;
+		}
+
+		map = bpf_object__add_map(obj);
+		if (IS_ERR(map))
+			return PTR_ERR(map);
+
+		map->sec_idx = obj->efile.st_ops_shndx;
+		map->sec_offset = vsi->offset;
+		map->name = strdup(var_name);
+		if (!map->name)
+			return -ENOMEM;
+
+		map->def.type = BPF_MAP_TYPE_STRUCT_OPS;
+		map->def.key_size = sizeof(int);
+		map->def.value_size = type->size;
+		map->def.max_entries = 1;
+
+		map->st_ops = calloc(1, sizeof(*map->st_ops));
+		if (!map->st_ops)
+			return -ENOMEM;
+		st_ops = map->st_ops;
+		st_ops->data = malloc(type->size);
+		st_ops->progs = calloc(btf_vlen(type), sizeof(*st_ops->progs));
+		st_ops->kern_func_off = malloc(btf_vlen(type) *
+					       sizeof(*st_ops->kern_func_off));
+		if (!st_ops->data || !st_ops->progs || !st_ops->kern_func_off)
+			return -ENOMEM;
+
+		if (vsi->offset + type->size > obj->efile.st_ops_data->d_size) {
+			pr_warn("struct_ops init: var %s is beyond the end of DATASEC %s\n",
+				var_name, STRUCT_OPS_SEC);
+			return -EINVAL;
+		}
+
+		memcpy(st_ops->data,
+		       obj->efile.st_ops_data->d_buf + vsi->offset,
+		       type->size);
+		st_ops->tname = tname;
+		st_ops->type = type;
+		st_ops->type_id = type_id;
+
+		pr_debug("struct_ops init: struct %s(type_id=%u) %s found at offset %u\n",
+			 tname, type_id, var_name, vsi->offset);
+	}
+
+	return 0;
+}
+
 static struct bpf_object *bpf_object__new(const char *path,
 					  const void *obj_buf,
 					  size_t obj_buf_sz,
@@ -607,6 +993,7 @@ static struct bpf_object *bpf_object__new(const char *path,
 	obj->efile.data_shndx = -1;
 	obj->efile.rodata_shndx = -1;
 	obj->efile.bss_shndx = -1;
+	obj->efile.st_ops_shndx = -1;
 	obj->kconfig_map_idx = -1;
 
 	obj->kern_version = get_kernel_version();
@@ -630,6 +1017,7 @@ static void bpf_object__elf_finish(struct bpf_object *obj)
 	obj->efile.data = NULL;
 	obj->efile.rodata = NULL;
 	obj->efile.bss = NULL;
+	obj->efile.st_ops_data = NULL;
 
 	zfree(&obj->efile.reloc_sects);
 	obj->efile.nr_reloc_sects = 0;
@@ -815,6 +1203,9 @@ int bpf_object__section_size(const struct bpf_object *obj, const char *name,
 	} else if (!strcmp(name, RODATA_SEC)) {
 		if (obj->efile.rodata)
 			*size = obj->efile.rodata->d_size;
+	} else if (!strcmp(name, STRUCT_OPS_SEC)) {
+		if (obj->efile.st_ops_data)
+			*size = obj->efile.st_ops_data->d_size;
 	} else {
 		ret = bpf_object_search_section_size(obj, name, &d_size);
 		if (!ret)
@@ -1440,6 +1831,20 @@ skip_mods_and_typedefs(const struct btf *btf, __u32 id, __u32 *res_id)
 	return t;
 }
 
+static const struct btf_type *
+resolve_func_ptr(const struct btf *btf, __u32 id, __u32 *res_id)
+{
+	const struct btf_type *t;
+
+	t = skip_mods_and_typedefs(btf, id, NULL);
+	if (!btf_is_ptr(t))
+		return NULL;
+
+	t = skip_mods_and_typedefs(btf, t->type, res_id);
+
+	return btf_is_func_proto(t) ? t : NULL;
+}
+
 /*
  * Fetch integer attribute of BTF map definition. Such attributes are
  * represented using a pointer to an array, in which dimensionality of array
@@ -1787,6 +2192,7 @@ static int bpf_object__init_maps(struct bpf_object *obj,
 	err = err ?: bpf_object__init_user_btf_maps(obj, strict, pin_root_path);
 	err = err ?: bpf_object__init_global_data_maps(obj);
 	err = err ?: bpf_object__init_kconfig_map(obj);
+	err = err ?: bpf_object__init_struct_ops_maps(obj);
 	if (err)
 		return err;
 
@@ -1889,7 +2295,8 @@ static void bpf_object__sanitize_btf_ext(struct bpf_object *obj)
 static bool bpf_object__is_btf_mandatory(const struct bpf_object *obj)
 {
 	return obj->efile.btf_maps_shndx >= 0 ||
-	       obj->nr_extern > 0;
+		obj->efile.st_ops_shndx >= 0 ||
+		obj->nr_extern > 0;
 }
 
 static int bpf_object__init_btf(struct bpf_object *obj,
@@ -2088,6 +2495,9 @@ static int bpf_object__elf_collect(struct bpf_object *obj)
 			} else if (strcmp(name, RODATA_SEC) == 0) {
 				obj->efile.rodata = data;
 				obj->efile.rodata_shndx = idx;
+			} else if (strcmp(name, STRUCT_OPS_SEC) == 0) {
+				obj->efile.st_ops_data = data;
+				obj->efile.st_ops_shndx = idx;
 			} else {
 				pr_debug("skip section(%d) %s\n", idx, name);
 			}
@@ -2097,7 +2507,8 @@ static int bpf_object__elf_collect(struct bpf_object *obj)
 			int sec = sh.sh_info; /* points to other section */
 
 			/* Only do relo for section with exec instructions */
-			if (!section_have_execinstr(obj, sec)) {
+			if (!section_have_execinstr(obj, sec) &&
+			    strcmp(name, ".rel" STRUCT_OPS_SEC)) {
 				pr_debug("skip relo %s(%d) for section(%d)\n",
 					 name, idx, sec);
 				continue;
@@ -2599,8 +3010,12 @@ static int bpf_map_find_btf_info(struct bpf_object *obj, struct bpf_map *map)
 	__u32 key_type_id = 0, value_type_id = 0;
 	int ret;
 
-	/* if it's BTF-defined map, we don't need to search for type IDs */
-	if (map->sec_idx == obj->efile.btf_maps_shndx)
+	/* if it's BTF-defined map, we don't need to search for type IDs.
+	 * For struct_ops map, it does not need btf_key_type_id and
+	 * btf_value_type_id.
+	 */
+	if (map->sec_idx == obj->efile.btf_maps_shndx ||
+	    bpf_map__is_struct_ops(map))
 		return 0;
 
 	if (!bpf_map__is_internal(map)) {
@@ -3025,6 +3440,9 @@ bpf_object__create_maps(struct bpf_object *obj)
 		if (bpf_map_type__is_map_in_map(def->type) &&
 		    map->inner_map_fd >= 0)
 			create_attr.inner_map_fd = map->inner_map_fd;
+		if (bpf_map__is_struct_ops(map))
+			create_attr.btf_vmlinux_value_type_id =
+				map->btf_vmlinux_value_type_id;
 
 		if (obj->btf && !bpf_map_find_btf_info(obj, map)) {
 			create_attr.btf_fd = btf__fd(obj->btf);
@@ -3899,7 +4317,7 @@ cleanup:
  * Probe few well-known locations for vmlinux kernel image and try to load BTF
  * data out of it to use for target BTF.
  */
-static struct btf *bpf_core_find_kernel_btf(void)
+static struct btf *bpf_find_kernel_btf(void)
 {
 	struct {
 		const char *path_fmt;
@@ -4180,7 +4598,7 @@ bpf_core_reloc_fields(struct bpf_object *obj, const char *targ_btf_path)
 	if (targ_btf_path)
 		targ_btf = btf__parse_elf(targ_btf_path, NULL);
 	else
-		targ_btf = bpf_core_find_kernel_btf();
+		targ_btf = bpf_find_kernel_btf();
 	if (IS_ERR(targ_btf)) {
 		pr_warn("failed to get target BTF: %ld\n", PTR_ERR(targ_btf));
 		return PTR_ERR(targ_btf);
@@ -4379,6 +4797,10 @@ bpf_object__relocate(struct bpf_object *obj, const char *targ_btf_path)
 	return 0;
 }
 
+static int bpf_object__collect_struct_ops_map_reloc(struct bpf_object *obj,
+						    GElf_Shdr *shdr,
+						    Elf_Data *data);
+
 static int bpf_object__collect_reloc(struct bpf_object *obj)
 {
 	int i, err;
@@ -4399,6 +4821,15 @@ static int bpf_object__collect_reloc(struct bpf_object *obj)
 			return -LIBBPF_ERRNO__INTERNAL;
 		}
 
+		if (idx == obj->efile.st_ops_shndx) {
+			err = bpf_object__collect_struct_ops_map_reloc(obj,
+								       shdr,
+								       data);
+			if (err)
+				return err;
+			continue;
+		}
+
 		prog = bpf_object__find_prog_by_idx(obj, idx);
 		if (!prog) {
 			pr_warn("relocation failed: no section(%d)\n", idx);
@@ -4433,7 +4864,9 @@ load_program(struct bpf_program *prog, struct bpf_insn *insns, int insns_cnt,
 	load_attr.insns = insns;
 	load_attr.insns_cnt = insns_cnt;
 	load_attr.license = license;
-	if (prog->type == BPF_PROG_TYPE_TRACING) {
+	if (prog->type == BPF_PROG_TYPE_STRUCT_OPS) {
+		load_attr.attach_btf_id = prog->attach_btf_id;
+	} else if (prog->type == BPF_PROG_TYPE_TRACING) {
 		load_attr.attach_prog_fd = prog->attach_prog_fd;
 		load_attr.attach_btf_id = prog->attach_btf_id;
 	} else {
@@ -4679,6 +5112,9 @@ __bpf_object__open(const char *path, const void *obj_buf, size_t obj_buf_sz,
 		enum bpf_prog_type prog_type;
 		enum bpf_attach_type attach_type;
 
+		if (prog->type != BPF_PROG_TYPE_UNSPEC)
+			continue;
+
 		err = libbpf_prog_type_by_name(prog->section_name, &prog_type,
 					       &attach_type);
 		if (err == -ESRCH)
@@ -4774,8 +5210,11 @@ int bpf_object__unload(struct bpf_object *obj)
 	if (!obj)
 		return -EINVAL;
 
-	for (i = 0; i < obj->nr_maps; i++)
+	for (i = 0; i < obj->nr_maps; i++) {
 		zclose(obj->maps[i].fd);
+		if (obj->maps[i].st_ops)
+			zfree(&obj->maps[i].st_ops->kern_vdata);
+	}
 
 	for (i = 0; i < obj->nr_programs; i++)
 		bpf_program__unload(&obj->programs[i]);
@@ -4891,6 +5330,7 @@ int bpf_object__load_xattr(struct bpf_object_load_attr *attr)
 	err = err ? : bpf_object__resolve_externs(obj, obj->kconfig);
 	err = err ? : bpf_object__sanitize_and_load_btf(obj);
 	err = err ? : bpf_object__sanitize_maps(obj);
+	err = err ? : bpf_object__init_kern_struct_ops_maps(obj);
 	err = err ? : bpf_object__create_maps(obj);
 	err = err ? : bpf_object__relocate(obj, attr->target_btf_path);
 	err = err ? : bpf_object__load_progs(obj, attr->log_level);
@@ -5478,6 +5918,13 @@ void bpf_object__close(struct bpf_object *obj)
 			map->mmaped = NULL;
 		}
 
+		if (map->st_ops) {
+			zfree(&map->st_ops->data);
+			zfree(&map->st_ops->progs);
+			zfree(&map->st_ops->kern_func_off);
+			zfree(&map->st_ops);
+		}
+
 		zfree(&map->name);
 		zfree(&map->pin_path);
 	}
@@ -5746,6 +6193,7 @@ BPF_PROG_TYPE_FNS(raw_tracepoint, BPF_PROG_TYPE_RAW_TRACEPOINT);
 BPF_PROG_TYPE_FNS(xdp, BPF_PROG_TYPE_XDP);
 BPF_PROG_TYPE_FNS(perf_event, BPF_PROG_TYPE_PERF_EVENT);
 BPF_PROG_TYPE_FNS(tracing, BPF_PROG_TYPE_TRACING);
+BPF_PROG_TYPE_FNS(struct_ops, BPF_PROG_TYPE_STRUCT_OPS);
 
 enum bpf_attach_type
 bpf_program__get_expected_attach_type(struct bpf_program *prog)
@@ -5899,6 +6347,7 @@ static const struct bpf_sec_def section_defs[] = {
 						BPF_CGROUP_GETSOCKOPT),
 	BPF_EAPROG_SEC("cgroup/setsockopt",	BPF_PROG_TYPE_CGROUP_SOCKOPT,
 						BPF_CGROUP_SETSOCKOPT),
+	BPF_PROG_SEC("struct_ops",		BPF_PROG_TYPE_STRUCT_OPS),
 };
 
 #undef BPF_PROG_SEC_IMPL
@@ -5975,11 +6424,141 @@ int libbpf_prog_type_by_name(const char *name, enum bpf_prog_type *prog_type,
 	return -ESRCH;
 }
 
+static struct bpf_map *find_struct_ops_map_by_offset(struct bpf_object *obj,
+						     size_t offset)
+{
+	struct bpf_map *map;
+	size_t i;
+
+	for (i = 0; i < obj->nr_maps; i++) {
+		map = &obj->maps[i];
+		if (!bpf_map__is_struct_ops(map))
+			continue;
+		if (map->sec_offset <= offset &&
+		    offset - map->sec_offset < map->def.value_size)
+			return map;
+	}
+
+	return NULL;
+}
+
+/* Collect the reloc from ELF and populate the st_ops->progs[] */
+static int bpf_object__collect_struct_ops_map_reloc(struct bpf_object *obj,
+						    GElf_Shdr *shdr,
+						    Elf_Data *data)
+{
+	const struct btf_member *member;
+	struct bpf_struct_ops *st_ops;
+	struct bpf_program *prog;
+	unsigned int shdr_idx;
+	const struct btf *btf;
+	struct bpf_map *map;
+	Elf_Data *symbols;
+	unsigned int moff;
+	const char *name;
+	u32 member_idx;
+	GElf_Sym sym;
+	GElf_Rel rel;
+	int i, nrels;
+
+	symbols = obj->efile.symbols;
+	btf = obj->btf;
+	nrels = shdr->sh_size / shdr->sh_entsize;
+	for (i = 0; i < nrels; i++) {
+		if (!gelf_getrel(data, i, &rel)) {
+			pr_warn("struct_ops reloc: failed to get %d reloc\n", i);
+			return -LIBBPF_ERRNO__FORMAT;
+		}
+
+		if (!gelf_getsym(symbols, GELF_R_SYM(rel.r_info), &sym)) {
+			pr_warn("struct_ops reloc: symbol %zx not found\n",
+				(size_t)GELF_R_SYM(rel.r_info));
+			return -LIBBPF_ERRNO__FORMAT;
+		}
+
+		name = elf_strptr(obj->efile.elf, obj->efile.strtabidx,
+				  sym.st_name) ? : "<?>";
+		map = find_struct_ops_map_by_offset(obj, rel.r_offset);
+		if (!map) {
+			pr_warn("struct_ops reloc: cannot find map at rel.r_offset %zu\n",
+				(size_t)rel.r_offset);
+			return -EINVAL;
+		}
+
+		moff = rel.r_offset - map->sec_offset;
+		shdr_idx = sym.st_shndx;
+		st_ops = map->st_ops;
+		pr_debug("struct_ops reloc %s: for %lld value %lld shdr_idx %u rel.r_offset %zu map->sec_offset %zu name %d (\'%s\')\n",
+			 map->name,
+			 (long long)(rel.r_info >> 32),
+			 (long long)sym.st_value,
+			 shdr_idx, (size_t)rel.r_offset,
+			 map->sec_offset, sym.st_name, name);
+
+		if (shdr_idx >= SHN_LORESERVE) {
+			pr_warn("struct_ops reloc %s: rel.r_offset %zu shdr_idx %u unsupported non-static function\n",
+				map->name, (size_t)rel.r_offset, shdr_idx);
+			return -LIBBPF_ERRNO__RELOC;
+		}
+
+		member = find_member_by_offset(st_ops->type, moff * 8);
+		if (!member) {
+			pr_warn("struct_ops reloc %s: cannot find member at moff %u\n",
+				map->name, moff);
+			return -EINVAL;
+		}
+		member_idx = member - btf_members(st_ops->type);
+		name = btf__name_by_offset(btf, member->name_off);
+
+		if (!resolve_func_ptr(btf, member->type, NULL)) {
+			pr_warn("struct_ops reloc %s: cannot relocate non func ptr %s\n",
+				map->name, name);
+			return -EINVAL;
+		}
+
+		prog = bpf_object__find_prog_by_idx(obj, shdr_idx);
+		if (!prog) {
+			pr_warn("struct_ops reloc %s: cannot find prog at shdr_idx %u to relocate func ptr %s\n",
+				map->name, shdr_idx, name);
+			return -EINVAL;
+		}
+
+		if (prog->type == BPF_PROG_TYPE_UNSPEC) {
+			const struct bpf_sec_def *sec_def;
+
+			sec_def = find_sec_def(prog->section_name);
+			if (sec_def &&
+			    sec_def->prog_type != BPF_PROG_TYPE_STRUCT_OPS) {
+				/* for pr_warn */
+				prog->type = sec_def->prog_type;
+				goto invalid_prog;
+			}
+
+			prog->type = BPF_PROG_TYPE_STRUCT_OPS;
+			prog->attach_btf_id = st_ops->type_id;
+			prog->expected_attach_type = member_idx;
+		} else if (prog->type != BPF_PROG_TYPE_STRUCT_OPS ||
+			   prog->attach_btf_id != st_ops->type_id ||
+			   prog->expected_attach_type != member_idx) {
+			goto invalid_prog;
+		}
+		st_ops->progs[member_idx] = prog;
+	}
+
+	return 0;
+
+invalid_prog:
+	pr_warn("struct_ops reloc %s: cannot use prog %s in sec %s with type %u attach_btf_id %u expected_attach_type %u for func ptr %s\n",
+		map->name, prog->name, prog->section_name, prog->type,
+		prog->attach_btf_id, prog->expected_attach_type, name);
+	return -EINVAL;
+}
+
 #define BTF_PREFIX "btf_trace_"
 int libbpf_find_vmlinux_btf_id(const char *name,
 			       enum bpf_attach_type attach_type)
 {
-	struct btf *btf = bpf_core_find_kernel_btf();
+	struct btf *btf = bpf_find_kernel_btf();
 	char raw_tp_btf[128] = BTF_PREFIX;
 	char *dst = raw_tp_btf + sizeof(BTF_PREFIX) - 1;
 	const char *btf_name;
@@ -6805,6 +7384,58 @@ struct bpf_link *bpf_program__attach(struct bpf_program *prog)
 	return sec_def->attach_fn(sec_def, prog);
 }
 
+static int bpf_link__detach_struct_ops(struct bpf_link *link)
+{
+	struct bpf_link_fd *l = (void *)link;
+	__u32 zero = 0;
+
+	if (bpf_map_delete_elem(l->fd, &zero))
+		return -errno;
+
+	return 0;
+}
+
+struct bpf_link *bpf_map__attach_struct_ops(struct bpf_map *map)
+{
+	struct bpf_struct_ops *st_ops;
+	struct bpf_link_fd *link;
+	__u32 i, zero = 0;
+	int err;
+
+	if (!bpf_map__is_struct_ops(map) || map->fd == -1)
+		return ERR_PTR(-EINVAL);
+
+	link = calloc(1, sizeof(*link));
+	if (!link)
+		return ERR_PTR(-EINVAL);
+
+	st_ops = map->st_ops;
+	for (i = 0; i < btf_vlen(st_ops->type); i++) {
+		struct bpf_program *prog = st_ops->progs[i];
+		void *kern_data;
+		int prog_fd;
+
+		if (!prog)
+			continue;
+
+		prog_fd = bpf_program__fd(prog);
+		kern_data = st_ops->kern_vdata + st_ops->kern_func_off[i];
+		*(unsigned long *)kern_data = prog_fd;
+	}
+
+	err = bpf_map_update_elem(map->fd, &zero, st_ops->kern_vdata, 0);
+	if (err) {
+		err = -errno;
+		free(link);
+		return ERR_PTR(err);
+	}
+
+	link->link.detach = bpf_link__detach_struct_ops;
+	link->fd = map->fd;
+
+	return (struct bpf_link *)link;
+}
+
 enum bpf_perf_event_ret
 bpf_perf_event_read_simple(void *mmap_mem, size_t mmap_size, size_t page_size,
 			   void **copy_mem, size_t *copy_size,
diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index 26bf539f1b3c..01639f9a1062 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -239,6 +239,8 @@ bpf_program__attach_raw_tracepoint(struct bpf_program *prog,
 
 LIBBPF_API struct bpf_link *
 bpf_program__attach_trace(struct bpf_program *prog);
+struct bpf_map;
+LIBBPF_API struct bpf_link *bpf_map__attach_struct_ops(struct bpf_map *map);
 struct bpf_insn;
 
 /*
@@ -315,6 +317,7 @@ LIBBPF_API int bpf_program__set_sched_act(struct bpf_program *prog);
 LIBBPF_API int bpf_program__set_xdp(struct bpf_program *prog);
 LIBBPF_API int bpf_program__set_perf_event(struct bpf_program *prog);
 LIBBPF_API int bpf_program__set_tracing(struct bpf_program *prog);
+LIBBPF_API int bpf_program__set_struct_ops(struct bpf_program *prog);
 
 LIBBPF_API enum bpf_prog_type bpf_program__get_type(struct bpf_program *prog);
 LIBBPF_API void bpf_program__set_type(struct bpf_program *prog,
@@ -335,6 +338,7 @@ LIBBPF_API bool bpf_program__is_sched_act(const struct bpf_program *prog);
 LIBBPF_API bool bpf_program__is_xdp(const struct bpf_program *prog);
 LIBBPF_API bool bpf_program__is_perf_event(const struct bpf_program *prog);
 LIBBPF_API bool bpf_program__is_tracing(const struct bpf_program *prog);
+LIBBPF_API bool bpf_program__is_struct_ops(const struct bpf_program *prog);
 
 /*
  * No need for __attribute__((packed)), all members of 'bpf_map_def'
@@ -354,7 +358,6 @@ struct bpf_map_def {
  * The 'struct bpf_map' in include/linux/bpf.h is internal to the kernel,
  * so no need to worry about a name clash.
  */
-struct bpf_map;
 LIBBPF_API struct bpf_map *
 bpf_object__find_map_by_name(const struct bpf_object *obj, const char *name);
 
diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
index b300d74c921a..a19f04e6e3d9 100644
--- a/tools/lib/bpf/libbpf.map
+++ b/tools/lib/bpf/libbpf.map
@@ -213,6 +213,7 @@ LIBBPF_0.0.7 {
 	global:
 		btf_dump__emit_type_decl;
 		bpf_link__disconnect;
+		bpf_map__attach_struct_ops;
 		bpf_object__find_program_by_name;
 		bpf_object__attach_skeleton;
 		bpf_object__destroy_skeleton;
@@ -223,5 +224,7 @@ LIBBPF_0.0.7 {
 		bpf_prog_attach_xattr;
 		bpf_program__attach;
 		bpf_program__name;
+		bpf_program__is_struct_ops;
+		bpf_program__set_struct_ops;
 		btf__align_of;
 } LIBBPF_0.0.6;
diff --git a/tools/lib/bpf/libbpf_probes.c b/tools/lib/bpf/libbpf_probes.c
index 221e6ad97012..320697f8e4c7 100644
--- a/tools/lib/bpf/libbpf_probes.c
+++ b/tools/lib/bpf/libbpf_probes.c
@@ -103,6 +103,7 @@ probe_load(enum bpf_prog_type prog_type, const struct bpf_insn *insns,
 	case BPF_PROG_TYPE_CGROUP_SYSCTL:
 	case BPF_PROG_TYPE_CGROUP_SOCKOPT:
 	case BPF_PROG_TYPE_TRACING:
+	case BPF_PROG_TYPE_STRUCT_OPS:
 	default:
 		break;
 	}
@@ -251,6 +252,7 @@ bool bpf_probe_map_type(enum bpf_map_type map_type, __u32 ifindex)
 	case BPF_MAP_TYPE_XSKMAP:
 	case BPF_MAP_TYPE_SOCKHASH:
 	case BPF_MAP_TYPE_REUSEPORT_SOCKARRAY:
+	case BPF_MAP_TYPE_STRUCT_OPS:
 	default:
 		break;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 09903869f69f37fd7a465183545b5739c6274654 Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <kafai@fb.com>
Date: Wed, 8 Jan 2020 16:35:17 -0800
Subject: bpf: Add bpf_dctcp example

This patch adds a bpf_dctcp example.  It currently does not do
no-ECN fallback but the same could be done through the cgrp2-bpf.

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200109003517.3856825-1-kafai@fb.com
---
 tools/testing/selftests/bpf/bpf_tcp_helpers.h      | 228 +++++++++++++++++++++
 .../testing/selftests/bpf/prog_tests/bpf_tcp_ca.c  | 187 +++++++++++++++++
 tools/testing/selftests/bpf/progs/bpf_dctcp.c      | 210 +++++++++++++++++++
 3 files changed, 625 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/bpf_tcp_helpers.h
 create mode 100644 tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
 create mode 100644 tools/testing/selftests/bpf/progs/bpf_dctcp.c

diff --git a/tools/testing/selftests/bpf/bpf_tcp_helpers.h b/tools/testing/selftests/bpf/bpf_tcp_helpers.h
new file mode 100644
index 000000000000..7da3e8db3adf
--- /dev/null
+++ b/tools/testing/selftests/bpf/bpf_tcp_helpers.h
@@ -0,0 +1,228 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __BPF_TCP_HELPERS_H
+#define __BPF_TCP_HELPERS_H
+
+#include <stdbool.h>
+#include <linux/types.h>
+#include <bpf_helpers.h>
+#include <bpf_core_read.h>
+#include "bpf_trace_helpers.h"
+
+/* "struct_ops/" is only a convention.  not a requirement. */
+#define BPF_TCP_OPS_0(fname, ret_type, ...) BPF_TRACE_x(0, "struct_ops/"#fname, fname, ret_type, __VA_ARGS__)
+#define BPF_TCP_OPS_1(fname, ret_type, ...) BPF_TRACE_x(1, "struct_ops/"#fname, fname, ret_type, __VA_ARGS__)
+#define BPF_TCP_OPS_2(fname, ret_type, ...) BPF_TRACE_x(2, "struct_ops/"#fname, fname, ret_type, __VA_ARGS__)
+#define BPF_TCP_OPS_3(fname, ret_type, ...) BPF_TRACE_x(3, "struct_ops/"#fname, fname, ret_type, __VA_ARGS__)
+#define BPF_TCP_OPS_4(fname, ret_type, ...) BPF_TRACE_x(4, "struct_ops/"#fname, fname, ret_type, __VA_ARGS__)
+#define BPF_TCP_OPS_5(fname, ret_type, ...) BPF_TRACE_x(5, "struct_ops/"#fname, fname, ret_type, __VA_ARGS__)
+
+struct sock_common {
+	unsigned char	skc_state;
+} __attribute__((preserve_access_index));
+
+struct sock {
+	struct sock_common	__sk_common;
+} __attribute__((preserve_access_index));
+
+struct inet_sock {
+	struct sock		sk;
+} __attribute__((preserve_access_index));
+
+struct inet_connection_sock {
+	struct inet_sock	  icsk_inet;
+	__u8			  icsk_ca_state:6,
+				  icsk_ca_setsockopt:1,
+				  icsk_ca_dst_locked:1;
+	struct {
+		__u8		  pending;
+	} icsk_ack;
+	__u64			  icsk_ca_priv[104 / sizeof(__u64)];
+} __attribute__((preserve_access_index));
+
+struct tcp_sock {
+	struct inet_connection_sock	inet_conn;
+
+	__u32	rcv_nxt;
+	__u32	snd_nxt;
+	__u32	snd_una;
+	__u8	ecn_flags;
+	__u32	delivered;
+	__u32	delivered_ce;
+	__u32	snd_cwnd;
+	__u32	snd_cwnd_cnt;
+	__u32	snd_cwnd_clamp;
+	__u32	snd_ssthresh;
+	__u8	syn_data:1,	/* SYN includes data */
+		syn_fastopen:1,	/* SYN includes Fast Open option */
+		syn_fastopen_exp:1,/* SYN includes Fast Open exp. option */
+		syn_fastopen_ch:1, /* Active TFO re-enabling probe */
+		syn_data_acked:1,/* data in SYN is acked by SYN-ACK */
+		save_syn:1,	/* Save headers of SYN packet */
+		is_cwnd_limited:1,/* forward progress limited by snd_cwnd? */
+		syn_smc:1;	/* SYN includes SMC */
+	__u32	max_packets_out;
+	__u32	lsndtime;
+	__u32	prior_cwnd;
+} __attribute__((preserve_access_index));
+
+static __always_inline struct inet_connection_sock *inet_csk(const struct sock *sk)
+{
+	return (struct inet_connection_sock *)sk;
+}
+
+static __always_inline void *inet_csk_ca(const struct sock *sk)
+{
+	return (void *)inet_csk(sk)->icsk_ca_priv;
+}
+
+static __always_inline struct tcp_sock *tcp_sk(const struct sock *sk)
+{
+	return (struct tcp_sock *)sk;
+}
+
+static __always_inline bool before(__u32 seq1, __u32 seq2)
+{
+	return (__s32)(seq1-seq2) < 0;
+}
+#define after(seq2, seq1) 	before(seq1, seq2)
+
+#define	TCP_ECN_OK		1
+#define	TCP_ECN_QUEUE_CWR	2
+#define	TCP_ECN_DEMAND_CWR	4
+#define	TCP_ECN_SEEN		8
+
+enum inet_csk_ack_state_t {
+	ICSK_ACK_SCHED	= 1,
+	ICSK_ACK_TIMER  = 2,
+	ICSK_ACK_PUSHED = 4,
+	ICSK_ACK_PUSHED2 = 8,
+	ICSK_ACK_NOW = 16	/* Send the next ACK immediately (once) */
+};
+
+enum tcp_ca_event {
+	CA_EVENT_TX_START = 0,
+	CA_EVENT_CWND_RESTART = 1,
+	CA_EVENT_COMPLETE_CWR = 2,
+	CA_EVENT_LOSS = 3,
+	CA_EVENT_ECN_NO_CE = 4,
+	CA_EVENT_ECN_IS_CE = 5,
+};
+
+enum tcp_ca_state {
+	TCP_CA_Open = 0,
+	TCP_CA_Disorder = 1,
+	TCP_CA_CWR = 2,
+	TCP_CA_Recovery = 3,
+	TCP_CA_Loss = 4
+};
+
+struct ack_sample {
+	__u32 pkts_acked;
+	__s32 rtt_us;
+	__u32 in_flight;
+} __attribute__((preserve_access_index));
+
+struct rate_sample {
+	__u64  prior_mstamp; /* starting timestamp for interval */
+	__u32  prior_delivered;	/* tp->delivered at "prior_mstamp" */
+	__s32  delivered;		/* number of packets delivered over interval */
+	long interval_us;	/* time for tp->delivered to incr "delivered" */
+	__u32 snd_interval_us;	/* snd interval for delivered packets */
+	__u32 rcv_interval_us;	/* rcv interval for delivered packets */
+	long rtt_us;		/* RTT of last (S)ACKed packet (or -1) */
+	int  losses;		/* number of packets marked lost upon ACK */
+	__u32  acked_sacked;	/* number of packets newly (S)ACKed upon ACK */
+	__u32  prior_in_flight;	/* in flight before this ACK */
+	bool is_app_limited;	/* is sample from packet with bubble in pipe? */
+	bool is_retrans;	/* is sample from retransmission? */
+	bool is_ack_delayed;	/* is this (likely) a delayed ACK? */
+} __attribute__((preserve_access_index));
+
+#define TCP_CA_NAME_MAX		16
+#define TCP_CONG_NEEDS_ECN	0x2
+
+struct tcp_congestion_ops {
+	char name[TCP_CA_NAME_MAX];
+	__u32 flags;
+
+	/* initialize private data (optional) */
+	void (*init)(struct sock *sk);
+	/* cleanup private data  (optional) */
+	void (*release)(struct sock *sk);
+
+	/* return slow start threshold (required) */
+	__u32 (*ssthresh)(struct sock *sk);
+	/* do new cwnd calculation (required) */
+	void (*cong_avoid)(struct sock *sk, __u32 ack, __u32 acked);
+	/* call before changing ca_state (optional) */
+	void (*set_state)(struct sock *sk, __u8 new_state);
+	/* call when cwnd event occurs (optional) */
+	void (*cwnd_event)(struct sock *sk, enum tcp_ca_event ev);
+	/* call when ack arrives (optional) */
+	void (*in_ack_event)(struct sock *sk, __u32 flags);
+	/* new value of cwnd after loss (required) */
+	__u32  (*undo_cwnd)(struct sock *sk);
+	/* hook for packet ack accounting (optional) */
+	void (*pkts_acked)(struct sock *sk, const struct ack_sample *sample);
+	/* override sysctl_tcp_min_tso_segs */
+	__u32 (*min_tso_segs)(struct sock *sk);
+	/* returns the multiplier used in tcp_sndbuf_expand (optional) */
+	__u32 (*sndbuf_expand)(struct sock *sk);
+	/* call when packets are delivered to update cwnd and pacing rate,
+	 * after all the ca_state processing. (optional)
+	 */
+	void (*cong_control)(struct sock *sk, const struct rate_sample *rs);
+};
+
+#define min(a, b) ((a) < (b) ? (a) : (b))
+#define max(a, b) ((a) > (b) ? (a) : (b))
+#define min_not_zero(x, y) ({			\
+	typeof(x) __x = (x);			\
+	typeof(y) __y = (y);			\
+	__x == 0 ? __y : ((__y == 0) ? __x : min(__x, __y)); })
+
+static __always_inline __u32 tcp_slow_start(struct tcp_sock *tp, __u32 acked)
+{
+	__u32 cwnd = min(tp->snd_cwnd + acked, tp->snd_ssthresh);
+
+	acked -= cwnd - tp->snd_cwnd;
+	tp->snd_cwnd = min(cwnd, tp->snd_cwnd_clamp);
+
+	return acked;
+}
+
+static __always_inline bool tcp_in_slow_start(const struct tcp_sock *tp)
+{
+	return tp->snd_cwnd < tp->snd_ssthresh;
+}
+
+static __always_inline bool tcp_is_cwnd_limited(const struct sock *sk)
+{
+	const struct tcp_sock *tp = tcp_sk(sk);
+
+	/* If in slow start, ensure cwnd grows to twice what was ACKed. */
+	if (tcp_in_slow_start(tp))
+		return tp->snd_cwnd < 2 * tp->max_packets_out;
+
+	return !!BPF_CORE_READ_BITFIELD(tp, is_cwnd_limited);
+}
+
+static __always_inline void tcp_cong_avoid_ai(struct tcp_sock *tp, __u32 w, __u32 acked)
+{
+	/* If credits accumulated at a higher w, apply them gently now. */
+	if (tp->snd_cwnd_cnt >= w) {
+		tp->snd_cwnd_cnt = 0;
+		tp->snd_cwnd++;
+	}
+
+	tp->snd_cwnd_cnt += acked;
+	if (tp->snd_cwnd_cnt >= w) {
+		__u32 delta = tp->snd_cwnd_cnt / w;
+
+		tp->snd_cwnd_cnt -= delta * w;
+		tp->snd_cwnd += delta;
+	}
+	tp->snd_cwnd = min(tp->snd_cwnd, tp->snd_cwnd_clamp);
+}
+
+#endif
diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
new file mode 100644
index 000000000000..517318f05b1d
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
@@ -0,0 +1,187 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2019 Facebook */
+
+#include <linux/err.h>
+#include <test_progs.h>
+#include "bpf_dctcp.skel.h"
+
+#define min(a, b) ((a) < (b) ? (a) : (b))
+
+static const unsigned int total_bytes = 10 * 1024 * 1024;
+static const struct timeval timeo_sec = { .tv_sec = 10 };
+static const size_t timeo_optlen = sizeof(timeo_sec);
+static int stop, duration;
+
+static int settimeo(int fd)
+{
+	int err;
+
+	err = setsockopt(fd, SOL_SOCKET, SO_RCVTIMEO, &timeo_sec,
+			 timeo_optlen);
+	if (CHECK(err == -1, "setsockopt(fd, SO_RCVTIMEO)", "errno:%d\n",
+		  errno))
+		return -1;
+
+	err = setsockopt(fd, SOL_SOCKET, SO_SNDTIMEO, &timeo_sec,
+			 timeo_optlen);
+	if (CHECK(err == -1, "setsockopt(fd, SO_SNDTIMEO)", "errno:%d\n",
+		  errno))
+		return -1;
+
+	return 0;
+}
+
+static int settcpca(int fd, const char *tcp_ca)
+{
+	int err;
+
+	err = setsockopt(fd, IPPROTO_TCP, TCP_CONGESTION, tcp_ca, strlen(tcp_ca));
+	if (CHECK(err == -1, "setsockopt(fd, TCP_CONGESTION)", "errno:%d\n",
+		  errno))
+		return -1;
+
+	return 0;
+}
+
+static void *server(void *arg)
+{
+	int lfd = (int)(long)arg, err = 0, fd;
+	ssize_t nr_sent = 0, bytes = 0;
+	char batch[1500];
+
+	fd = accept(lfd, NULL, NULL);
+	while (fd == -1) {
+		if (errno == EINTR)
+			continue;
+		err = -errno;
+		goto done;
+	}
+
+	if (settimeo(fd)) {
+		err = -errno;
+		goto done;
+	}
+
+	while (bytes < total_bytes && !READ_ONCE(stop)) {
+		nr_sent = send(fd, &batch,
+			       min(total_bytes - bytes, sizeof(batch)), 0);
+		if (nr_sent == -1 && errno == EINTR)
+			continue;
+		if (nr_sent == -1) {
+			err = -errno;
+			break;
+		}
+		bytes += nr_sent;
+	}
+
+	CHECK(bytes != total_bytes, "send", "%zd != %u nr_sent:%zd errno:%d\n",
+	      bytes, total_bytes, nr_sent, errno);
+
+done:
+	if (fd != -1)
+		close(fd);
+	if (err) {
+		WRITE_ONCE(stop, 1);
+		return ERR_PTR(err);
+	}
+	return NULL;
+}
+
+static void do_test(const char *tcp_ca)
+{
+	struct sockaddr_in6 sa6 = {};
+	ssize_t nr_recv = 0, bytes = 0;
+	int lfd = -1, fd = -1;
+	pthread_t srv_thread;
+	socklen_t addrlen = sizeof(sa6);
+	void *thread_ret;
+	char batch[1500];
+	int err;
+
+	WRITE_ONCE(stop, 0);
+
+	lfd = socket(AF_INET6, SOCK_STREAM, 0);
+	if (CHECK(lfd == -1, "socket", "errno:%d\n", errno))
+		return;
+	fd = socket(AF_INET6, SOCK_STREAM, 0);
+	if (CHECK(fd == -1, "socket", "errno:%d\n", errno)) {
+		close(lfd);
+		return;
+	}
+
+	if (settcpca(lfd, tcp_ca) || settcpca(fd, tcp_ca) ||
+	    settimeo(lfd) || settimeo(fd))
+		goto done;
+
+	/* bind, listen and start server thread to accept */
+	sa6.sin6_family = AF_INET6;
+	sa6.sin6_addr = in6addr_loopback;
+	err = bind(lfd, (struct sockaddr *)&sa6, addrlen);
+	if (CHECK(err == -1, "bind", "errno:%d\n", errno))
+		goto done;
+	err = getsockname(lfd, (struct sockaddr *)&sa6, &addrlen);
+	if (CHECK(err == -1, "getsockname", "errno:%d\n", errno))
+		goto done;
+	err = listen(lfd, 1);
+	if (CHECK(err == -1, "listen", "errno:%d\n", errno))
+		goto done;
+	err = pthread_create(&srv_thread, NULL, server, (void *)(long)lfd);
+	if (CHECK(err != 0, "pthread_create", "err:%d\n", err))
+		goto done;
+
+	/* connect to server */
+	err = connect(fd, (struct sockaddr *)&sa6, addrlen);
+	if (CHECK(err == -1, "connect", "errno:%d\n", errno))
+		goto wait_thread;
+
+	/* recv total_bytes */
+	while (bytes < total_bytes && !READ_ONCE(stop)) {
+		nr_recv = recv(fd, &batch,
+			       min(total_bytes - bytes, sizeof(batch)), 0);
+		if (nr_recv == -1 && errno == EINTR)
+			continue;
+		if (nr_recv == -1)
+			break;
+		bytes += nr_recv;
+	}
+
+	CHECK(bytes != total_bytes, "recv", "%zd != %u nr_recv:%zd errno:%d\n",
+	      bytes, total_bytes, nr_recv, errno);
+
+wait_thread:
+	WRITE_ONCE(stop, 1);
+	pthread_join(srv_thread, &thread_ret);
+	CHECK(IS_ERR(thread_ret), "pthread_join", "thread_ret:%ld",
+	      PTR_ERR(thread_ret));
+done:
+	close(lfd);
+	close(fd);
+}
+
+static void test_dctcp(void)
+{
+	struct bpf_dctcp *dctcp_skel;
+	struct bpf_link *link;
+
+	dctcp_skel = bpf_dctcp__open_and_load();
+	if (CHECK(!dctcp_skel, "bpf_dctcp__open_and_load", "failed\n"))
+		return;
+
+	link = bpf_map__attach_struct_ops(dctcp_skel->maps.dctcp);
+	if (CHECK(IS_ERR(link), "bpf_map__attach_struct_ops", "err:%ld\n",
+		  PTR_ERR(link))) {
+		bpf_dctcp__destroy(dctcp_skel);
+		return;
+	}
+
+	do_test("bpf_dctcp");
+
+	bpf_link__destroy(link);
+	bpf_dctcp__destroy(dctcp_skel);
+}
+
+void test_bpf_tcp_ca(void)
+{
+	if (test__start_subtest("dctcp"))
+		test_dctcp();
+}
diff --git a/tools/testing/selftests/bpf/progs/bpf_dctcp.c b/tools/testing/selftests/bpf/progs/bpf_dctcp.c
new file mode 100644
index 000000000000..5f9b613663e5
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/bpf_dctcp.c
@@ -0,0 +1,210 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2019 Facebook */
+
+/* WARNING: This implemenation is not necessarily the same
+ * as the tcp_dctcp.c.  The purpose is mainly for testing
+ * the kernel BPF logic.
+ */
+
+#include <linux/bpf.h>
+#include <linux/types.h>
+#include "bpf_tcp_helpers.h"
+
+char _license[] SEC("license") = "GPL";
+
+#define DCTCP_MAX_ALPHA	1024U
+
+struct dctcp {
+	__u32 old_delivered;
+	__u32 old_delivered_ce;
+	__u32 prior_rcv_nxt;
+	__u32 dctcp_alpha;
+	__u32 next_seq;
+	__u32 ce_state;
+	__u32 loss_cwnd;
+};
+
+static unsigned int dctcp_shift_g = 4; /* g = 1/2^4 */
+static unsigned int dctcp_alpha_on_init = DCTCP_MAX_ALPHA;
+
+static __always_inline void dctcp_reset(const struct tcp_sock *tp,
+					struct dctcp *ca)
+{
+	ca->next_seq = tp->snd_nxt;
+
+	ca->old_delivered = tp->delivered;
+	ca->old_delivered_ce = tp->delivered_ce;
+}
+
+BPF_TCP_OPS_1(dctcp_init, void, struct sock *, sk)
+{
+	const struct tcp_sock *tp = tcp_sk(sk);
+	struct dctcp *ca = inet_csk_ca(sk);
+
+	ca->prior_rcv_nxt = tp->rcv_nxt;
+	ca->dctcp_alpha = min(dctcp_alpha_on_init, DCTCP_MAX_ALPHA);
+	ca->loss_cwnd = 0;
+	ca->ce_state = 0;
+
+	dctcp_reset(tp, ca);
+}
+
+BPF_TCP_OPS_1(dctcp_ssthresh, __u32, struct sock *, sk)
+{
+	struct dctcp *ca = inet_csk_ca(sk);
+	struct tcp_sock *tp = tcp_sk(sk);
+
+	ca->loss_cwnd = tp->snd_cwnd;
+	return max(tp->snd_cwnd - ((tp->snd_cwnd * ca->dctcp_alpha) >> 11U), 2U);
+}
+
+BPF_TCP_OPS_2(dctcp_update_alpha, void,
+	      struct sock *, sk, __u32, flags)
+{
+	const struct tcp_sock *tp = tcp_sk(sk);
+	struct dctcp *ca = inet_csk_ca(sk);
+
+	/* Expired RTT */
+	if (!before(tp->snd_una, ca->next_seq)) {
+		__u32 delivered_ce = tp->delivered_ce - ca->old_delivered_ce;
+		__u32 alpha = ca->dctcp_alpha;
+
+		/* alpha = (1 - g) * alpha + g * F */
+
+		alpha -= min_not_zero(alpha, alpha >> dctcp_shift_g);
+		if (delivered_ce) {
+			__u32 delivered = tp->delivered - ca->old_delivered;
+
+			/* If dctcp_shift_g == 1, a 32bit value would overflow
+			 * after 8 M packets.
+			 */
+			delivered_ce <<= (10 - dctcp_shift_g);
+			delivered_ce /= max(1U, delivered);
+
+			alpha = min(alpha + delivered_ce, DCTCP_MAX_ALPHA);
+		}
+		ca->dctcp_alpha = alpha;
+		dctcp_reset(tp, ca);
+	}
+}
+
+static __always_inline void dctcp_react_to_loss(struct sock *sk)
+{
+	struct dctcp *ca = inet_csk_ca(sk);
+	struct tcp_sock *tp = tcp_sk(sk);
+
+	ca->loss_cwnd = tp->snd_cwnd;
+	tp->snd_ssthresh = max(tp->snd_cwnd >> 1U, 2U);
+}
+
+BPF_TCP_OPS_2(dctcp_state, void, struct sock *, sk, __u8, new_state)
+{
+	if (new_state == TCP_CA_Recovery &&
+	    new_state != BPF_CORE_READ_BITFIELD(inet_csk(sk), icsk_ca_state))
+		dctcp_react_to_loss(sk);
+	/* We handle RTO in dctcp_cwnd_event to ensure that we perform only
+	 * one loss-adjustment per RTT.
+	 */
+}
+
+static __always_inline void dctcp_ece_ack_cwr(struct sock *sk, __u32 ce_state)
+{
+	struct tcp_sock *tp = tcp_sk(sk);
+
+	if (ce_state == 1)
+		tp->ecn_flags |= TCP_ECN_DEMAND_CWR;
+	else
+		tp->ecn_flags &= ~TCP_ECN_DEMAND_CWR;
+}
+
+/* Minimal DCTP CE state machine:
+ *
+ * S:	0 <- last pkt was non-CE
+ *	1 <- last pkt was CE
+ */
+static __always_inline
+void dctcp_ece_ack_update(struct sock *sk, enum tcp_ca_event evt,
+			  __u32 *prior_rcv_nxt, __u32 *ce_state)
+{
+	__u32 new_ce_state = (evt == CA_EVENT_ECN_IS_CE) ? 1 : 0;
+
+	if (*ce_state != new_ce_state) {
+		/* CE state has changed, force an immediate ACK to
+		 * reflect the new CE state. If an ACK was delayed,
+		 * send that first to reflect the prior CE state.
+		 */
+		if (inet_csk(sk)->icsk_ack.pending & ICSK_ACK_TIMER) {
+			dctcp_ece_ack_cwr(sk, *ce_state);
+			bpf_tcp_send_ack(sk, *prior_rcv_nxt);
+		}
+		inet_csk(sk)->icsk_ack.pending |= ICSK_ACK_NOW;
+	}
+	*prior_rcv_nxt = tcp_sk(sk)->rcv_nxt;
+	*ce_state = new_ce_state;
+	dctcp_ece_ack_cwr(sk, new_ce_state);
+}
+
+BPF_TCP_OPS_2(dctcp_cwnd_event, void,
+	      struct sock *, sk, enum tcp_ca_event, ev)
+{
+	struct dctcp *ca = inet_csk_ca(sk);
+
+	switch (ev) {
+	case CA_EVENT_ECN_IS_CE:
+	case CA_EVENT_ECN_NO_CE:
+		dctcp_ece_ack_update(sk, ev, &ca->prior_rcv_nxt, &ca->ce_state);
+		break;
+	case CA_EVENT_LOSS:
+		dctcp_react_to_loss(sk);
+		break;
+	default:
+		/* Don't care for the rest. */
+		break;
+	}
+}
+
+BPF_TCP_OPS_1(dctcp_cwnd_undo, __u32, struct sock *, sk)
+{
+	const struct dctcp *ca = inet_csk_ca(sk);
+
+	return max(tcp_sk(sk)->snd_cwnd, ca->loss_cwnd);
+}
+
+BPF_TCP_OPS_3(tcp_reno_cong_avoid, void,
+	      struct sock *, sk, __u32, ack, __u32, acked)
+{
+	struct tcp_sock *tp = tcp_sk(sk);
+
+	if (!tcp_is_cwnd_limited(sk))
+		return;
+
+	/* In "safe" area, increase. */
+	if (tcp_in_slow_start(tp)) {
+		acked = tcp_slow_start(tp, acked);
+		if (!acked)
+			return;
+	}
+	/* In dangerous area, increase slowly. */
+	tcp_cong_avoid_ai(tp, tp->snd_cwnd, acked);
+}
+
+SEC(".struct_ops")
+struct tcp_congestion_ops dctcp_nouse = {
+	.init		= (void *)dctcp_init,
+	.set_state	= (void *)dctcp_state,
+	.flags		= TCP_CONG_NEEDS_ECN,
+	.name		= "bpf_dctcp_nouse",
+};
+
+SEC(".struct_ops")
+struct tcp_congestion_ops dctcp = {
+	.init		= (void *)dctcp_init,
+	.in_ack_event   = (void *)dctcp_update_alpha,
+	.cwnd_event	= (void *)dctcp_cwnd_event,
+	.ssthresh	= (void *)dctcp_ssthresh,
+	.cong_avoid	= (void *)tcp_reno_cong_avoid,
+	.undo_cwnd	= (void *)dctcp_cwnd_undo,
+	.set_state	= (void *)dctcp_state,
+	.flags		= TCP_CONG_NEEDS_ECN,
+	.name		= "bpf_dctcp",
+};
-- 
cgit v1.2.3-59-g8ed1b


From f5bfcd953d811dbb8913de36b96b38da6bb62135 Mon Sep 17 00:00:00 2001
From: Andrey Ignatov <rdna@fb.com>
Date: Tue, 7 Jan 2020 17:40:06 -0800
Subject: bpf: Document BPF_F_QUERY_EFFECTIVE flag

Document BPF_F_QUERY_EFFECTIVE flag, mostly to clarify how it affects
attach_flags what may not be obvious and what may lead to confision.

Specifically attach_flags is returned only for target_fd but if programs
are inherited from an ancestor cgroup then returned attach_flags for
current cgroup may be confusing. For example, two effective programs of
same attach_type can be returned but w/o BPF_F_ALLOW_MULTI in
attach_flags.

Simple repro:
  # bpftool c s /sys/fs/cgroup/path/to/task
  ID       AttachType      AttachFlags     Name
  # bpftool c s /sys/fs/cgroup/path/to/task effective
  ID       AttachType      AttachFlags     Name
  95043    ingress                         tw_ipt_ingress
  95048    ingress                         tw_ingress

Signed-off-by: Andrey Ignatov <rdna@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Song Liu <songliubraving@fb.com>
Link: https://lore.kernel.org/bpf/20200108014006.938363-1-rdna@fb.com
---
 include/uapi/linux/bpf.h       | 7 ++++++-
 tools/include/uapi/linux/bpf.h | 7 ++++++-
 2 files changed, 12 insertions(+), 2 deletions(-)

diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 2d6a2e572f56..52966e758fe5 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -359,7 +359,12 @@ enum bpf_attach_type {
 /* Enable memory-mapping BPF map */
 #define BPF_F_MMAPABLE		(1U << 10)
 
-/* flags for BPF_PROG_QUERY */
+/* Flags for BPF_PROG_QUERY. */
+
+/* Query effective (directly attached + inherited from ancestor cgroups)
+ * programs that will be executed for events within a cgroup.
+ * attach_flags with this flag are returned only for directly attached programs.
+ */
 #define BPF_F_QUERY_EFFECTIVE	(1U << 0)
 
 enum bpf_stack_build_id_status {
diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index 2d6a2e572f56..52966e758fe5 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -359,7 +359,12 @@ enum bpf_attach_type {
 /* Enable memory-mapping BPF map */
 #define BPF_F_MMAPABLE		(1U << 10)
 
-/* flags for BPF_PROG_QUERY */
+/* Flags for BPF_PROG_QUERY. */
+
+/* Query effective (directly attached + inherited from ancestor cgroups)
+ * programs that will be executed for events within a cgroup.
+ * attach_flags with this flag are returned only for directly attached programs.
+ */
 #define BPF_F_QUERY_EFFECTIVE	(1U << 0)
 
 enum bpf_stack_build_id_status {
-- 
cgit v1.2.3-59-g8ed1b


From 43ad352dc8207015972bfc91f5d23ea4307f8ff1 Mon Sep 17 00:00:00 2001
From: Milind Parab <mparab@cadence.com>
Date: Thu, 9 Jan 2020 08:36:46 +0000
Subject: net: macb: add support for C45 MDIO read/write

This patch modify MDIO read/write functions to support
communication with C45 PHY.

Signed-off-by: Milind Parab <mparab@cadence.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/cadence/macb.h      | 15 +++++---
 drivers/net/ethernet/cadence/macb_main.c | 61 ++++++++++++++++++++++++++------
 2 files changed, 61 insertions(+), 15 deletions(-)

diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h
index 19fe4f4867c7..dbf7070fcdba 100644
--- a/drivers/net/ethernet/cadence/macb.h
+++ b/drivers/net/ethernet/cadence/macb.h
@@ -630,10 +630,17 @@
 #define GEM_CLK_DIV96				5
 
 /* Constants for MAN register */
-#define MACB_MAN_SOF				1
-#define MACB_MAN_WRITE				1
-#define MACB_MAN_READ				2
-#define MACB_MAN_CODE				2
+#define MACB_MAN_C22_SOF			1
+#define MACB_MAN_C22_WRITE			1
+#define MACB_MAN_C22_READ			2
+#define MACB_MAN_C22_CODE			2
+
+#define MACB_MAN_C45_SOF			0
+#define MACB_MAN_C45_ADDR			0
+#define MACB_MAN_C45_WRITE			1
+#define MACB_MAN_C45_POST_READ_INCR		2
+#define MACB_MAN_C45_READ			3
+#define MACB_MAN_C45_CODE			2
 
 /* Capability mask bits */
 #define MACB_CAPS_ISR_CLEAR_ON_WRITE		0x00000001
diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c
index c5ee363ca5dc..09a88f353eef 100644
--- a/drivers/net/ethernet/cadence/macb_main.c
+++ b/drivers/net/ethernet/cadence/macb_main.c
@@ -337,11 +337,30 @@ static int macb_mdio_read(struct mii_bus *bus, int mii_id, int regnum)
 	if (status < 0)
 		goto mdio_read_exit;
 
-	macb_writel(bp, MAN, (MACB_BF(SOF, MACB_MAN_SOF)
-			      | MACB_BF(RW, MACB_MAN_READ)
-			      | MACB_BF(PHYA, mii_id)
-			      | MACB_BF(REGA, regnum)
-			      | MACB_BF(CODE, MACB_MAN_CODE)));
+	if (regnum & MII_ADDR_C45) {
+		macb_writel(bp, MAN, (MACB_BF(SOF, MACB_MAN_C45_SOF)
+			    | MACB_BF(RW, MACB_MAN_C45_ADDR)
+			    | MACB_BF(PHYA, mii_id)
+			    | MACB_BF(REGA, (regnum >> 16) & 0x1F)
+			    | MACB_BF(DATA, regnum & 0xFFFF)
+			    | MACB_BF(CODE, MACB_MAN_C45_CODE)));
+
+		status = macb_mdio_wait_for_idle(bp);
+		if (status < 0)
+			goto mdio_read_exit;
+
+		macb_writel(bp, MAN, (MACB_BF(SOF, MACB_MAN_C45_SOF)
+			    | MACB_BF(RW, MACB_MAN_C45_READ)
+			    | MACB_BF(PHYA, mii_id)
+			    | MACB_BF(REGA, (regnum >> 16) & 0x1F)
+			    | MACB_BF(CODE, MACB_MAN_C45_CODE)));
+	} else {
+		macb_writel(bp, MAN, (MACB_BF(SOF, MACB_MAN_C22_SOF)
+				| MACB_BF(RW, MACB_MAN_C22_READ)
+				| MACB_BF(PHYA, mii_id)
+				| MACB_BF(REGA, regnum)
+				| MACB_BF(CODE, MACB_MAN_C22_CODE)));
+	}
 
 	status = macb_mdio_wait_for_idle(bp);
 	if (status < 0)
@@ -370,12 +389,32 @@ static int macb_mdio_write(struct mii_bus *bus, int mii_id, int regnum,
 	if (status < 0)
 		goto mdio_write_exit;
 
-	macb_writel(bp, MAN, (MACB_BF(SOF, MACB_MAN_SOF)
-			      | MACB_BF(RW, MACB_MAN_WRITE)
-			      | MACB_BF(PHYA, mii_id)
-			      | MACB_BF(REGA, regnum)
-			      | MACB_BF(CODE, MACB_MAN_CODE)
-			      | MACB_BF(DATA, value)));
+	if (regnum & MII_ADDR_C45) {
+		macb_writel(bp, MAN, (MACB_BF(SOF, MACB_MAN_C45_SOF)
+			    | MACB_BF(RW, MACB_MAN_C45_ADDR)
+			    | MACB_BF(PHYA, mii_id)
+			    | MACB_BF(REGA, (regnum >> 16) & 0x1F)
+			    | MACB_BF(DATA, regnum & 0xFFFF)
+			    | MACB_BF(CODE, MACB_MAN_C45_CODE)));
+
+		status = macb_mdio_wait_for_idle(bp);
+		if (status < 0)
+			goto mdio_write_exit;
+
+		macb_writel(bp, MAN, (MACB_BF(SOF, MACB_MAN_C45_SOF)
+			    | MACB_BF(RW, MACB_MAN_C45_WRITE)
+			    | MACB_BF(PHYA, mii_id)
+			    | MACB_BF(REGA, (regnum >> 16) & 0x1F)
+			    | MACB_BF(CODE, MACB_MAN_C45_CODE)
+			    | MACB_BF(DATA, value)));
+	} else {
+		macb_writel(bp, MAN, (MACB_BF(SOF, MACB_MAN_C22_SOF)
+				| MACB_BF(RW, MACB_MAN_C22_WRITE)
+				| MACB_BF(PHYA, mii_id)
+				| MACB_BF(REGA, regnum)
+				| MACB_BF(CODE, MACB_MAN_C22_CODE)
+				| MACB_BF(DATA, value)));
+	}
 
 	status = macb_mdio_wait_for_idle(bp);
 	if (status < 0)
-- 
cgit v1.2.3-59-g8ed1b


From a2ba4b971a69b1d96b41953ac7fa17bdcd823baa Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Thu, 9 Jan 2020 15:43:44 +0000
Subject: sfc: move some port link state/caps code

The moved code handles MCDI port link state and capabilities.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/Makefile           |   2 +-
 drivers/net/ethernet/sfc/efx.c              |  15 ---
 drivers/net/ethernet/sfc/mcdi_port.c        | 180 +------------------------
 drivers/net/ethernet/sfc/mcdi_port_common.c | 202 ++++++++++++++++++++++++++++
 4 files changed, 204 insertions(+), 195 deletions(-)
 create mode 100644 drivers/net/ethernet/sfc/mcdi_port_common.c

diff --git a/drivers/net/ethernet/sfc/Makefile b/drivers/net/ethernet/sfc/Makefile
index 40a54df34647..f6d4b2b2dd83 100644
--- a/drivers/net/ethernet/sfc/Makefile
+++ b/drivers/net/ethernet/sfc/Makefile
@@ -3,7 +3,7 @@ sfc-y			+= efx.o efx_common.o efx_channels.o nic.o \
 			   farch.o siena.o ef10.o \
 			   tx.o tx_common.o rx.o rx_common.o \
 			   selftest.o ethtool.o ptp.o tx_tso.o \
-			   mcdi.o mcdi_port.o \
+			   mcdi.o mcdi_port.o mcdi_port_common.o \
 			   mcdi_mon.o
 sfc-$(CONFIG_SFC_MTD)	+= mtd.o
 sfc-$(CONFIG_SFC_SRIOV)	+= sriov.o siena_sriov.o ef10_sriov.o
diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index 655424e83d97..33b6ed03b3ea 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -134,21 +134,6 @@ static int efx_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **xdpfs,
  *
  **************************************************************************/
 
-void efx_link_set_advertising(struct efx_nic *efx,
-			      const unsigned long *advertising)
-{
-	memcpy(efx->link_advertising, advertising,
-	       sizeof(__ETHTOOL_DECLARE_LINK_MODE_MASK()));
-
-	efx->link_advertising[0] |= ADVERTISED_Autoneg;
-	if (advertising[0] & ADVERTISED_Pause)
-		efx->wanted_fc |= (EFX_FC_TX | EFX_FC_RX);
-	else
-		efx->wanted_fc &= ~(EFX_FC_TX | EFX_FC_RX);
-	if (advertising[0] & ADVERTISED_Asym_Pause)
-		efx->wanted_fc ^= EFX_FC_TX;
-}
-
 /* Equivalent to efx_link_set_advertising with all-zeroes, except does not
  * force the Autoneg bit on.
  */
diff --git a/drivers/net/ethernet/sfc/mcdi_port.c b/drivers/net/ethernet/sfc/mcdi_port.c
index f19d7b8a2935..109c86dd130d 100644
--- a/drivers/net/ethernet/sfc/mcdi_port.c
+++ b/drivers/net/ethernet/sfc/mcdi_port.c
@@ -16,91 +16,6 @@
 #include "selftest.h"
 #include "mcdi_port_common.h"
 
-int efx_mcdi_get_phy_cfg(struct efx_nic *efx, struct efx_mcdi_phy_data *cfg)
-{
-	MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_PHY_CFG_OUT_LEN);
-	size_t outlen;
-	int rc;
-
-	BUILD_BUG_ON(MC_CMD_GET_PHY_CFG_IN_LEN != 0);
-	BUILD_BUG_ON(MC_CMD_GET_PHY_CFG_OUT_NAME_LEN != sizeof(cfg->name));
-
-	rc = efx_mcdi_rpc(efx, MC_CMD_GET_PHY_CFG, NULL, 0,
-			  outbuf, sizeof(outbuf), &outlen);
-	if (rc)
-		goto fail;
-
-	if (outlen < MC_CMD_GET_PHY_CFG_OUT_LEN) {
-		rc = -EIO;
-		goto fail;
-	}
-
-	cfg->flags = MCDI_DWORD(outbuf, GET_PHY_CFG_OUT_FLAGS);
-	cfg->type = MCDI_DWORD(outbuf, GET_PHY_CFG_OUT_TYPE);
-	cfg->supported_cap =
-		MCDI_DWORD(outbuf, GET_PHY_CFG_OUT_SUPPORTED_CAP);
-	cfg->channel = MCDI_DWORD(outbuf, GET_PHY_CFG_OUT_CHANNEL);
-	cfg->port = MCDI_DWORD(outbuf, GET_PHY_CFG_OUT_PRT);
-	cfg->stats_mask = MCDI_DWORD(outbuf, GET_PHY_CFG_OUT_STATS_MASK);
-	memcpy(cfg->name, MCDI_PTR(outbuf, GET_PHY_CFG_OUT_NAME),
-	       sizeof(cfg->name));
-	cfg->media = MCDI_DWORD(outbuf, GET_PHY_CFG_OUT_MEDIA_TYPE);
-	cfg->mmd_mask = MCDI_DWORD(outbuf, GET_PHY_CFG_OUT_MMD_MASK);
-	memcpy(cfg->revision, MCDI_PTR(outbuf, GET_PHY_CFG_OUT_REVISION),
-	       sizeof(cfg->revision));
-
-	return 0;
-
-fail:
-	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
-	return rc;
-}
-
-int efx_mcdi_set_link(struct efx_nic *efx, u32 capabilities,
-		      u32 flags, u32 loopback_mode,
-		      u32 loopback_speed)
-{
-	MCDI_DECLARE_BUF(inbuf, MC_CMD_SET_LINK_IN_LEN);
-	int rc;
-
-	BUILD_BUG_ON(MC_CMD_SET_LINK_OUT_LEN != 0);
-
-	MCDI_SET_DWORD(inbuf, SET_LINK_IN_CAP, capabilities);
-	MCDI_SET_DWORD(inbuf, SET_LINK_IN_FLAGS, flags);
-	MCDI_SET_DWORD(inbuf, SET_LINK_IN_LOOPBACK_MODE, loopback_mode);
-	MCDI_SET_DWORD(inbuf, SET_LINK_IN_LOOPBACK_SPEED, loopback_speed);
-
-	rc = efx_mcdi_rpc(efx, MC_CMD_SET_LINK, inbuf, sizeof(inbuf),
-			  NULL, 0, NULL);
-	return rc;
-}
-
-int efx_mcdi_loopback_modes(struct efx_nic *efx, u64 *loopback_modes)
-{
-	MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_LOOPBACK_MODES_OUT_LEN);
-	size_t outlen;
-	int rc;
-
-	rc = efx_mcdi_rpc(efx, MC_CMD_GET_LOOPBACK_MODES, NULL, 0,
-			  outbuf, sizeof(outbuf), &outlen);
-	if (rc)
-		goto fail;
-
-	if (outlen < (MC_CMD_GET_LOOPBACK_MODES_OUT_SUGGESTED_OFST +
-		      MC_CMD_GET_LOOPBACK_MODES_OUT_SUGGESTED_LEN)) {
-		rc = -EIO;
-		goto fail;
-	}
-
-	*loopback_modes = MCDI_QWORD(outbuf, GET_LOOPBACK_MODES_OUT_SUGGESTED);
-
-	return 0;
-
-fail:
-	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
-	return rc;
-}
-
 static int efx_mcdi_mdio_read(struct net_device *net_dev,
 			      int prtad, int devad, u16 addr)
 {
@@ -154,70 +69,6 @@ static int efx_mcdi_mdio_write(struct net_device *net_dev,
 	return 0;
 }
 
-void mcdi_to_ethtool_linkset(u32 media, u32 cap, unsigned long *linkset)
-{
-	#define SET_BIT(name)	__set_bit(ETHTOOL_LINK_MODE_ ## name ## _BIT, \
-					  linkset)
-
-	bitmap_zero(linkset, __ETHTOOL_LINK_MODE_MASK_NBITS);
-	switch (media) {
-	case MC_CMD_MEDIA_KX4:
-		SET_BIT(Backplane);
-		if (cap & (1 << MC_CMD_PHY_CAP_1000FDX_LBN))
-			SET_BIT(1000baseKX_Full);
-		if (cap & (1 << MC_CMD_PHY_CAP_10000FDX_LBN))
-			SET_BIT(10000baseKX4_Full);
-		if (cap & (1 << MC_CMD_PHY_CAP_40000FDX_LBN))
-			SET_BIT(40000baseKR4_Full);
-		break;
-
-	case MC_CMD_MEDIA_XFP:
-	case MC_CMD_MEDIA_SFP_PLUS:
-	case MC_CMD_MEDIA_QSFP_PLUS:
-		SET_BIT(FIBRE);
-		if (cap & (1 << MC_CMD_PHY_CAP_1000FDX_LBN))
-			SET_BIT(1000baseT_Full);
-		if (cap & (1 << MC_CMD_PHY_CAP_10000FDX_LBN))
-			SET_BIT(10000baseT_Full);
-		if (cap & (1 << MC_CMD_PHY_CAP_40000FDX_LBN))
-			SET_BIT(40000baseCR4_Full);
-		if (cap & (1 << MC_CMD_PHY_CAP_100000FDX_LBN))
-			SET_BIT(100000baseCR4_Full);
-		if (cap & (1 << MC_CMD_PHY_CAP_25000FDX_LBN))
-			SET_BIT(25000baseCR_Full);
-		if (cap & (1 << MC_CMD_PHY_CAP_50000FDX_LBN))
-			SET_BIT(50000baseCR2_Full);
-		break;
-
-	case MC_CMD_MEDIA_BASE_T:
-		SET_BIT(TP);
-		if (cap & (1 << MC_CMD_PHY_CAP_10HDX_LBN))
-			SET_BIT(10baseT_Half);
-		if (cap & (1 << MC_CMD_PHY_CAP_10FDX_LBN))
-			SET_BIT(10baseT_Full);
-		if (cap & (1 << MC_CMD_PHY_CAP_100HDX_LBN))
-			SET_BIT(100baseT_Half);
-		if (cap & (1 << MC_CMD_PHY_CAP_100FDX_LBN))
-			SET_BIT(100baseT_Full);
-		if (cap & (1 << MC_CMD_PHY_CAP_1000HDX_LBN))
-			SET_BIT(1000baseT_Half);
-		if (cap & (1 << MC_CMD_PHY_CAP_1000FDX_LBN))
-			SET_BIT(1000baseT_Full);
-		if (cap & (1 << MC_CMD_PHY_CAP_10000FDX_LBN))
-			SET_BIT(10000baseT_Full);
-		break;
-	}
-
-	if (cap & (1 << MC_CMD_PHY_CAP_PAUSE_LBN))
-		SET_BIT(Pause);
-	if (cap & (1 << MC_CMD_PHY_CAP_ASYM_LBN))
-		SET_BIT(Asym_Pause);
-	if (cap & (1 << MC_CMD_PHY_CAP_AN_LBN))
-		SET_BIT(Autoneg);
-
-	#undef SET_BIT
-}
-
 u32 ethtool_linkset_to_mcdi_cap(const unsigned long *linkset)
 {
 	u32 result = 0;
@@ -308,34 +159,6 @@ u8 mcdi_to_ethtool_media(u32 media)
 	}
 }
 
-void efx_mcdi_phy_decode_link(struct efx_nic *efx,
-			      struct efx_link_state *link_state,
-			      u32 speed, u32 flags, u32 fcntl)
-{
-	switch (fcntl) {
-	case MC_CMD_FCNTL_AUTO:
-		WARN_ON(1);	/* This is not a link mode */
-		link_state->fc = EFX_FC_AUTO | EFX_FC_TX | EFX_FC_RX;
-		break;
-	case MC_CMD_FCNTL_BIDIR:
-		link_state->fc = EFX_FC_TX | EFX_FC_RX;
-		break;
-	case MC_CMD_FCNTL_RESPOND:
-		link_state->fc = EFX_FC_RX;
-		break;
-	default:
-		WARN_ON(1);
-		/* Fall through */
-	case MC_CMD_FCNTL_OFF:
-		link_state->fc = 0;
-		break;
-	}
-
-	link_state->up = !!(flags & (1 << MC_CMD_GET_LINK_OUT_LINK_UP_LBN));
-	link_state->fd = !!(flags & (1 << MC_CMD_GET_LINK_OUT_FULL_DUPLEX_LBN));
-	link_state->speed = speed;
-}
-
 /* The semantics of the ethtool FEC mode bitmask are not well defined,
  * particularly the meaning of combinations of bits.  Which means we get to
  * define our own semantics, as follows:
@@ -652,8 +475,7 @@ efx_mcdi_phy_set_link_ksettings(struct efx_nic *efx,
 	return 0;
 }
 
-int efx_mcdi_phy_get_fecparam(struct efx_nic *efx,
-			      struct ethtool_fecparam *fec)
+int efx_mcdi_phy_get_fecparam(struct efx_nic *efx, struct ethtool_fecparam *fec)
 {
 	MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_LINK_OUT_V2_LEN);
 	u32 caps, active, speed; /* MCDI format */
diff --git a/drivers/net/ethernet/sfc/mcdi_port_common.c b/drivers/net/ethernet/sfc/mcdi_port_common.c
new file mode 100644
index 000000000000..0ba7b5a47d99
--- /dev/null
+++ b/drivers/net/ethernet/sfc/mcdi_port_common.c
@@ -0,0 +1,202 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/****************************************************************************
+ * Driver for Solarflare network controllers and boards
+ * Copyright 2018 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#include "mcdi_port_common.h"
+
+int efx_mcdi_get_phy_cfg(struct efx_nic *efx, struct efx_mcdi_phy_data *cfg)
+{
+	MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_PHY_CFG_OUT_LEN);
+	size_t outlen;
+	int rc;
+
+	BUILD_BUG_ON(MC_CMD_GET_PHY_CFG_IN_LEN != 0);
+	BUILD_BUG_ON(MC_CMD_GET_PHY_CFG_OUT_NAME_LEN != sizeof(cfg->name));
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_GET_PHY_CFG, NULL, 0,
+			  outbuf, sizeof(outbuf), &outlen);
+	if (rc)
+		goto fail;
+
+	if (outlen < MC_CMD_GET_PHY_CFG_OUT_LEN) {
+		rc = -EIO;
+		goto fail;
+	}
+
+	cfg->flags = MCDI_DWORD(outbuf, GET_PHY_CFG_OUT_FLAGS);
+	cfg->type = MCDI_DWORD(outbuf, GET_PHY_CFG_OUT_TYPE);
+	cfg->supported_cap =
+		MCDI_DWORD(outbuf, GET_PHY_CFG_OUT_SUPPORTED_CAP);
+	cfg->channel = MCDI_DWORD(outbuf, GET_PHY_CFG_OUT_CHANNEL);
+	cfg->port = MCDI_DWORD(outbuf, GET_PHY_CFG_OUT_PRT);
+	cfg->stats_mask = MCDI_DWORD(outbuf, GET_PHY_CFG_OUT_STATS_MASK);
+	memcpy(cfg->name, MCDI_PTR(outbuf, GET_PHY_CFG_OUT_NAME),
+	       sizeof(cfg->name));
+	cfg->media = MCDI_DWORD(outbuf, GET_PHY_CFG_OUT_MEDIA_TYPE);
+	cfg->mmd_mask = MCDI_DWORD(outbuf, GET_PHY_CFG_OUT_MMD_MASK);
+	memcpy(cfg->revision, MCDI_PTR(outbuf, GET_PHY_CFG_OUT_REVISION),
+	       sizeof(cfg->revision));
+
+	return 0;
+
+fail:
+	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+	return rc;
+}
+
+void efx_link_set_advertising(struct efx_nic *efx,
+			      const unsigned long *advertising)
+{
+	memcpy(efx->link_advertising, advertising,
+	       sizeof(__ETHTOOL_DECLARE_LINK_MODE_MASK()));
+
+	efx->link_advertising[0] |= ADVERTISED_Autoneg;
+	if (advertising[0] & ADVERTISED_Pause)
+		efx->wanted_fc |= (EFX_FC_TX | EFX_FC_RX);
+	else
+		efx->wanted_fc &= ~(EFX_FC_TX | EFX_FC_RX);
+	if (advertising[0] & ADVERTISED_Asym_Pause)
+		efx->wanted_fc ^= EFX_FC_TX;
+}
+
+int efx_mcdi_set_link(struct efx_nic *efx, u32 capabilities,
+		      u32 flags, u32 loopback_mode, u32 loopback_speed)
+{
+	MCDI_DECLARE_BUF(inbuf, MC_CMD_SET_LINK_IN_LEN);
+	int rc;
+
+	BUILD_BUG_ON(MC_CMD_SET_LINK_OUT_LEN != 0);
+
+	MCDI_SET_DWORD(inbuf, SET_LINK_IN_CAP, capabilities);
+	MCDI_SET_DWORD(inbuf, SET_LINK_IN_FLAGS, flags);
+	MCDI_SET_DWORD(inbuf, SET_LINK_IN_LOOPBACK_MODE, loopback_mode);
+	MCDI_SET_DWORD(inbuf, SET_LINK_IN_LOOPBACK_SPEED, loopback_speed);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_SET_LINK, inbuf, sizeof(inbuf),
+			  NULL, 0, NULL);
+	return rc;
+}
+
+int efx_mcdi_loopback_modes(struct efx_nic *efx, u64 *loopback_modes)
+{
+	MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_LOOPBACK_MODES_OUT_LEN);
+	size_t outlen;
+	int rc;
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_GET_LOOPBACK_MODES, NULL, 0,
+			  outbuf, sizeof(outbuf), &outlen);
+	if (rc)
+		goto fail;
+
+	if (outlen < (MC_CMD_GET_LOOPBACK_MODES_OUT_SUGGESTED_OFST +
+		      MC_CMD_GET_LOOPBACK_MODES_OUT_SUGGESTED_LEN)) {
+		rc = -EIO;
+		goto fail;
+	}
+
+	*loopback_modes = MCDI_QWORD(outbuf, GET_LOOPBACK_MODES_OUT_SUGGESTED);
+
+	return 0;
+
+fail:
+	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+	return rc;
+}
+
+void mcdi_to_ethtool_linkset(u32 media, u32 cap, unsigned long *linkset)
+{
+	#define SET_BIT(name)	__set_bit(ETHTOOL_LINK_MODE_ ## name ## _BIT, \
+					  linkset)
+
+	bitmap_zero(linkset, __ETHTOOL_LINK_MODE_MASK_NBITS);
+	switch (media) {
+	case MC_CMD_MEDIA_KX4:
+		SET_BIT(Backplane);
+		if (cap & (1 << MC_CMD_PHY_CAP_1000FDX_LBN))
+			SET_BIT(1000baseKX_Full);
+		if (cap & (1 << MC_CMD_PHY_CAP_10000FDX_LBN))
+			SET_BIT(10000baseKX4_Full);
+		if (cap & (1 << MC_CMD_PHY_CAP_40000FDX_LBN))
+			SET_BIT(40000baseKR4_Full);
+		break;
+
+	case MC_CMD_MEDIA_XFP:
+	case MC_CMD_MEDIA_SFP_PLUS:
+	case MC_CMD_MEDIA_QSFP_PLUS:
+		SET_BIT(FIBRE);
+		if (cap & (1 << MC_CMD_PHY_CAP_1000FDX_LBN))
+			SET_BIT(1000baseT_Full);
+		if (cap & (1 << MC_CMD_PHY_CAP_10000FDX_LBN))
+			SET_BIT(10000baseT_Full);
+		if (cap & (1 << MC_CMD_PHY_CAP_40000FDX_LBN))
+			SET_BIT(40000baseCR4_Full);
+		if (cap & (1 << MC_CMD_PHY_CAP_100000FDX_LBN))
+			SET_BIT(100000baseCR4_Full);
+		if (cap & (1 << MC_CMD_PHY_CAP_25000FDX_LBN))
+			SET_BIT(25000baseCR_Full);
+		if (cap & (1 << MC_CMD_PHY_CAP_50000FDX_LBN))
+			SET_BIT(50000baseCR2_Full);
+		break;
+
+	case MC_CMD_MEDIA_BASE_T:
+		SET_BIT(TP);
+		if (cap & (1 << MC_CMD_PHY_CAP_10HDX_LBN))
+			SET_BIT(10baseT_Half);
+		if (cap & (1 << MC_CMD_PHY_CAP_10FDX_LBN))
+			SET_BIT(10baseT_Full);
+		if (cap & (1 << MC_CMD_PHY_CAP_100HDX_LBN))
+			SET_BIT(100baseT_Half);
+		if (cap & (1 << MC_CMD_PHY_CAP_100FDX_LBN))
+			SET_BIT(100baseT_Full);
+		if (cap & (1 << MC_CMD_PHY_CAP_1000HDX_LBN))
+			SET_BIT(1000baseT_Half);
+		if (cap & (1 << MC_CMD_PHY_CAP_1000FDX_LBN))
+			SET_BIT(1000baseT_Full);
+		if (cap & (1 << MC_CMD_PHY_CAP_10000FDX_LBN))
+			SET_BIT(10000baseT_Full);
+		break;
+	}
+
+	if (cap & (1 << MC_CMD_PHY_CAP_PAUSE_LBN))
+		SET_BIT(Pause);
+	if (cap & (1 << MC_CMD_PHY_CAP_ASYM_LBN))
+		SET_BIT(Asym_Pause);
+	if (cap & (1 << MC_CMD_PHY_CAP_AN_LBN))
+		SET_BIT(Autoneg);
+
+	#undef SET_BIT
+}
+
+void efx_mcdi_phy_decode_link(struct efx_nic *efx,
+			      struct efx_link_state *link_state,
+			      u32 speed, u32 flags, u32 fcntl)
+{
+	switch (fcntl) {
+	case MC_CMD_FCNTL_AUTO:
+		WARN_ON(1);	/* This is not a link mode */
+		link_state->fc = EFX_FC_AUTO | EFX_FC_TX | EFX_FC_RX;
+		break;
+	case MC_CMD_FCNTL_BIDIR:
+		link_state->fc = EFX_FC_TX | EFX_FC_RX;
+		break;
+	case MC_CMD_FCNTL_RESPOND:
+		link_state->fc = EFX_FC_RX;
+		break;
+	default:
+		WARN_ON(1);
+		/* Fall through */
+	case MC_CMD_FCNTL_OFF:
+		link_state->fc = 0;
+		break;
+	}
+
+	link_state->up = !!(flags & (1 << MC_CMD_GET_LINK_OUT_LINK_UP_LBN));
+	link_state->fd = !!(flags & (1 << MC_CMD_GET_LINK_OUT_FULL_DUPLEX_LBN));
+	link_state->speed = speed;
+}
-- 
cgit v1.2.3-59-g8ed1b


From 1a8d88a8b71b9f51ae3c6a3178ed9531f1c76ef2 Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Thu, 9 Jan 2020 15:43:59 +0000
Subject: sfc: move some MCDI port utility functions

They just convert between different sets of flags/registers.
Some block comments were adjusted.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/mcdi_port.c        | 148 ----------------------------
 drivers/net/ethernet/sfc/mcdi_port_common.c | 148 ++++++++++++++++++++++++++++
 2 files changed, 148 insertions(+), 148 deletions(-)

diff --git a/drivers/net/ethernet/sfc/mcdi_port.c b/drivers/net/ethernet/sfc/mcdi_port.c
index 109c86dd130d..d811753782dc 100644
--- a/drivers/net/ethernet/sfc/mcdi_port.c
+++ b/drivers/net/ethernet/sfc/mcdi_port.c
@@ -69,154 +69,6 @@ static int efx_mcdi_mdio_write(struct net_device *net_dev,
 	return 0;
 }
 
-u32 ethtool_linkset_to_mcdi_cap(const unsigned long *linkset)
-{
-	u32 result = 0;
-
-	#define TEST_BIT(name)	test_bit(ETHTOOL_LINK_MODE_ ## name ## _BIT, \
-					 linkset)
-
-	if (TEST_BIT(10baseT_Half))
-		result |= (1 << MC_CMD_PHY_CAP_10HDX_LBN);
-	if (TEST_BIT(10baseT_Full))
-		result |= (1 << MC_CMD_PHY_CAP_10FDX_LBN);
-	if (TEST_BIT(100baseT_Half))
-		result |= (1 << MC_CMD_PHY_CAP_100HDX_LBN);
-	if (TEST_BIT(100baseT_Full))
-		result |= (1 << MC_CMD_PHY_CAP_100FDX_LBN);
-	if (TEST_BIT(1000baseT_Half))
-		result |= (1 << MC_CMD_PHY_CAP_1000HDX_LBN);
-	if (TEST_BIT(1000baseT_Full) || TEST_BIT(1000baseKX_Full))
-		result |= (1 << MC_CMD_PHY_CAP_1000FDX_LBN);
-	if (TEST_BIT(10000baseT_Full) || TEST_BIT(10000baseKX4_Full))
-		result |= (1 << MC_CMD_PHY_CAP_10000FDX_LBN);
-	if (TEST_BIT(40000baseCR4_Full) || TEST_BIT(40000baseKR4_Full))
-		result |= (1 << MC_CMD_PHY_CAP_40000FDX_LBN);
-	if (TEST_BIT(100000baseCR4_Full))
-		result |= (1 << MC_CMD_PHY_CAP_100000FDX_LBN);
-	if (TEST_BIT(25000baseCR_Full))
-		result |= (1 << MC_CMD_PHY_CAP_25000FDX_LBN);
-	if (TEST_BIT(50000baseCR2_Full))
-		result |= (1 << MC_CMD_PHY_CAP_50000FDX_LBN);
-	if (TEST_BIT(Pause))
-		result |= (1 << MC_CMD_PHY_CAP_PAUSE_LBN);
-	if (TEST_BIT(Asym_Pause))
-		result |= (1 << MC_CMD_PHY_CAP_ASYM_LBN);
-	if (TEST_BIT(Autoneg))
-		result |= (1 << MC_CMD_PHY_CAP_AN_LBN);
-
-	#undef TEST_BIT
-
-	return result;
-}
-
-u32 efx_get_mcdi_phy_flags(struct efx_nic *efx)
-{
-	struct efx_mcdi_phy_data *phy_cfg = efx->phy_data;
-	enum efx_phy_mode mode, supported;
-	u32 flags;
-
-	/* TODO: Advertise the capabilities supported by this PHY */
-	supported = 0;
-	if (phy_cfg->flags & (1 << MC_CMD_GET_PHY_CFG_OUT_TXDIS_LBN))
-		supported |= PHY_MODE_TX_DISABLED;
-	if (phy_cfg->flags & (1 << MC_CMD_GET_PHY_CFG_OUT_LOWPOWER_LBN))
-		supported |= PHY_MODE_LOW_POWER;
-	if (phy_cfg->flags & (1 << MC_CMD_GET_PHY_CFG_OUT_POWEROFF_LBN))
-		supported |= PHY_MODE_OFF;
-
-	mode = efx->phy_mode & supported;
-
-	flags = 0;
-	if (mode & PHY_MODE_TX_DISABLED)
-		flags |= (1 << MC_CMD_SET_LINK_IN_TXDIS_LBN);
-	if (mode & PHY_MODE_LOW_POWER)
-		flags |= (1 << MC_CMD_SET_LINK_IN_LOWPOWER_LBN);
-	if (mode & PHY_MODE_OFF)
-		flags |= (1 << MC_CMD_SET_LINK_IN_POWEROFF_LBN);
-
-	return flags;
-}
-
-u8 mcdi_to_ethtool_media(u32 media)
-{
-	switch (media) {
-	case MC_CMD_MEDIA_XAUI:
-	case MC_CMD_MEDIA_CX4:
-	case MC_CMD_MEDIA_KX4:
-		return PORT_OTHER;
-
-	case MC_CMD_MEDIA_XFP:
-	case MC_CMD_MEDIA_SFP_PLUS:
-	case MC_CMD_MEDIA_QSFP_PLUS:
-		return PORT_FIBRE;
-
-	case MC_CMD_MEDIA_BASE_T:
-		return PORT_TP;
-
-	default:
-		return PORT_OTHER;
-	}
-}
-
-/* The semantics of the ethtool FEC mode bitmask are not well defined,
- * particularly the meaning of combinations of bits.  Which means we get to
- * define our own semantics, as follows:
- * OFF overrides any other bits, and means "disable all FEC" (with the
- * exception of 25G KR4/CR4, where it is not possible to reject it if AN
- * partner requests it).
- * AUTO on its own means use cable requirements and link partner autoneg with
- * fw-default preferences for the cable type.
- * AUTO and either RS or BASER means use the specified FEC type if cable and
- * link partner support it, otherwise autoneg/fw-default.
- * RS or BASER alone means use the specified FEC type if cable and link partner
- * support it and either requests it, otherwise no FEC.
- * Both RS and BASER (whether AUTO or not) means use FEC if cable and link
- * partner support it, preferring RS to BASER.
- */
-u32 ethtool_fec_caps_to_mcdi(u32 ethtool_cap)
-{
-	u32 ret = 0;
-
-	if (ethtool_cap & ETHTOOL_FEC_OFF)
-		return 0;
-
-	if (ethtool_cap & ETHTOOL_FEC_AUTO)
-		ret |= (1 << MC_CMD_PHY_CAP_BASER_FEC_LBN) |
-		       (1 << MC_CMD_PHY_CAP_25G_BASER_FEC_LBN) |
-		       (1 << MC_CMD_PHY_CAP_RS_FEC_LBN);
-	if (ethtool_cap & ETHTOOL_FEC_RS)
-		ret |= (1 << MC_CMD_PHY_CAP_RS_FEC_LBN) |
-		       (1 << MC_CMD_PHY_CAP_RS_FEC_REQUESTED_LBN);
-	if (ethtool_cap & ETHTOOL_FEC_BASER)
-		ret |= (1 << MC_CMD_PHY_CAP_BASER_FEC_LBN) |
-		       (1 << MC_CMD_PHY_CAP_25G_BASER_FEC_LBN) |
-		       (1 << MC_CMD_PHY_CAP_BASER_FEC_REQUESTED_LBN) |
-		       (1 << MC_CMD_PHY_CAP_25G_BASER_FEC_REQUESTED_LBN);
-	return ret;
-}
-
-/* Invert ethtool_fec_caps_to_mcdi.  There are two combinations that function
- * can never produce, (baser xor rs) and neither req; the implementation below
- * maps both of those to AUTO.  This should never matter, and it's not clear
- * what a better mapping would be anyway.
- */
-u32 mcdi_fec_caps_to_ethtool(u32 caps, bool is_25g)
-{
-	bool rs = caps & (1 << MC_CMD_PHY_CAP_RS_FEC_LBN),
-	     rs_req = caps & (1 << MC_CMD_PHY_CAP_RS_FEC_REQUESTED_LBN),
-	     baser = is_25g ? caps & (1 << MC_CMD_PHY_CAP_25G_BASER_FEC_LBN)
-			    : caps & (1 << MC_CMD_PHY_CAP_BASER_FEC_LBN),
-	     baser_req = is_25g ? caps & (1 << MC_CMD_PHY_CAP_25G_BASER_FEC_REQUESTED_LBN)
-				: caps & (1 << MC_CMD_PHY_CAP_BASER_FEC_REQUESTED_LBN);
-
-	if (!baser && !rs)
-		return ETHTOOL_FEC_OFF;
-	return (rs_req ? ETHTOOL_FEC_RS : 0) |
-	       (baser_req ? ETHTOOL_FEC_BASER : 0) |
-	       (baser == baser_req && rs == rs_req ? 0 : ETHTOOL_FEC_AUTO);
-}
-
 static int efx_mcdi_phy_probe(struct efx_nic *efx)
 {
 	struct efx_mcdi_phy_data *phy_data;
diff --git a/drivers/net/ethernet/sfc/mcdi_port_common.c b/drivers/net/ethernet/sfc/mcdi_port_common.c
index 0ba7b5a47d99..9763d59f7681 100644
--- a/drivers/net/ethernet/sfc/mcdi_port_common.c
+++ b/drivers/net/ethernet/sfc/mcdi_port_common.c
@@ -173,6 +173,96 @@ void mcdi_to_ethtool_linkset(u32 media, u32 cap, unsigned long *linkset)
 	#undef SET_BIT
 }
 
+u32 ethtool_linkset_to_mcdi_cap(const unsigned long *linkset)
+{
+	u32 result = 0;
+
+	#define TEST_BIT(name)	test_bit(ETHTOOL_LINK_MODE_ ## name ## _BIT, \
+					 linkset)
+
+	if (TEST_BIT(10baseT_Half))
+		result |= (1 << MC_CMD_PHY_CAP_10HDX_LBN);
+	if (TEST_BIT(10baseT_Full))
+		result |= (1 << MC_CMD_PHY_CAP_10FDX_LBN);
+	if (TEST_BIT(100baseT_Half))
+		result |= (1 << MC_CMD_PHY_CAP_100HDX_LBN);
+	if (TEST_BIT(100baseT_Full))
+		result |= (1 << MC_CMD_PHY_CAP_100FDX_LBN);
+	if (TEST_BIT(1000baseT_Half))
+		result |= (1 << MC_CMD_PHY_CAP_1000HDX_LBN);
+	if (TEST_BIT(1000baseT_Full) || TEST_BIT(1000baseKX_Full))
+		result |= (1 << MC_CMD_PHY_CAP_1000FDX_LBN);
+	if (TEST_BIT(10000baseT_Full) || TEST_BIT(10000baseKX4_Full))
+		result |= (1 << MC_CMD_PHY_CAP_10000FDX_LBN);
+	if (TEST_BIT(40000baseCR4_Full) || TEST_BIT(40000baseKR4_Full))
+		result |= (1 << MC_CMD_PHY_CAP_40000FDX_LBN);
+	if (TEST_BIT(100000baseCR4_Full))
+		result |= (1 << MC_CMD_PHY_CAP_100000FDX_LBN);
+	if (TEST_BIT(25000baseCR_Full))
+		result |= (1 << MC_CMD_PHY_CAP_25000FDX_LBN);
+	if (TEST_BIT(50000baseCR2_Full))
+		result |= (1 << MC_CMD_PHY_CAP_50000FDX_LBN);
+	if (TEST_BIT(Pause))
+		result |= (1 << MC_CMD_PHY_CAP_PAUSE_LBN);
+	if (TEST_BIT(Asym_Pause))
+		result |= (1 << MC_CMD_PHY_CAP_ASYM_LBN);
+	if (TEST_BIT(Autoneg))
+		result |= (1 << MC_CMD_PHY_CAP_AN_LBN);
+
+	#undef TEST_BIT
+
+	return result;
+}
+
+u32 efx_get_mcdi_phy_flags(struct efx_nic *efx)
+{
+	struct efx_mcdi_phy_data *phy_cfg = efx->phy_data;
+	enum efx_phy_mode mode, supported;
+	u32 flags;
+
+	/* TODO: Advertise the capabilities supported by this PHY */
+	supported = 0;
+	if (phy_cfg->flags & (1 << MC_CMD_GET_PHY_CFG_OUT_TXDIS_LBN))
+		supported |= PHY_MODE_TX_DISABLED;
+	if (phy_cfg->flags & (1 << MC_CMD_GET_PHY_CFG_OUT_LOWPOWER_LBN))
+		supported |= PHY_MODE_LOW_POWER;
+	if (phy_cfg->flags & (1 << MC_CMD_GET_PHY_CFG_OUT_POWEROFF_LBN))
+		supported |= PHY_MODE_OFF;
+
+	mode = efx->phy_mode & supported;
+
+	flags = 0;
+	if (mode & PHY_MODE_TX_DISABLED)
+		flags |= (1 << MC_CMD_SET_LINK_IN_TXDIS_LBN);
+	if (mode & PHY_MODE_LOW_POWER)
+		flags |= (1 << MC_CMD_SET_LINK_IN_LOWPOWER_LBN);
+	if (mode & PHY_MODE_OFF)
+		flags |= (1 << MC_CMD_SET_LINK_IN_POWEROFF_LBN);
+
+	return flags;
+}
+
+u8 mcdi_to_ethtool_media(u32 media)
+{
+	switch (media) {
+	case MC_CMD_MEDIA_XAUI:
+	case MC_CMD_MEDIA_CX4:
+	case MC_CMD_MEDIA_KX4:
+		return PORT_OTHER;
+
+	case MC_CMD_MEDIA_XFP:
+	case MC_CMD_MEDIA_SFP_PLUS:
+	case MC_CMD_MEDIA_QSFP_PLUS:
+		return PORT_FIBRE;
+
+	case MC_CMD_MEDIA_BASE_T:
+		return PORT_TP;
+
+	default:
+		return PORT_OTHER;
+	}
+}
+
 void efx_mcdi_phy_decode_link(struct efx_nic *efx,
 			      struct efx_link_state *link_state,
 			      u32 speed, u32 flags, u32 fcntl)
@@ -200,3 +290,61 @@ void efx_mcdi_phy_decode_link(struct efx_nic *efx,
 	link_state->fd = !!(flags & (1 << MC_CMD_GET_LINK_OUT_FULL_DUPLEX_LBN));
 	link_state->speed = speed;
 }
+
+/* The semantics of the ethtool FEC mode bitmask are not well defined,
+ * particularly the meaning of combinations of bits.  Which means we get to
+ * define our own semantics, as follows:
+ * OFF overrides any other bits, and means "disable all FEC" (with the
+ * exception of 25G KR4/CR4, where it is not possible to reject it if AN
+ * partner requests it).
+ * AUTO on its own means use cable requirements and link partner autoneg with
+ * fw-default preferences for the cable type.
+ * AUTO and either RS or BASER means use the specified FEC type if cable and
+ * link partner support it, otherwise autoneg/fw-default.
+ * RS or BASER alone means use the specified FEC type if cable and link partner
+ * support it and either requests it, otherwise no FEC.
+ * Both RS and BASER (whether AUTO or not) means use FEC if cable and link
+ * partner support it, preferring RS to BASER.
+ */
+u32 ethtool_fec_caps_to_mcdi(u32 ethtool_cap)
+{
+	u32 ret = 0;
+
+	if (ethtool_cap & ETHTOOL_FEC_OFF)
+		return 0;
+
+	if (ethtool_cap & ETHTOOL_FEC_AUTO)
+		ret |= (1 << MC_CMD_PHY_CAP_BASER_FEC_LBN) |
+		       (1 << MC_CMD_PHY_CAP_25G_BASER_FEC_LBN) |
+		       (1 << MC_CMD_PHY_CAP_RS_FEC_LBN);
+	if (ethtool_cap & ETHTOOL_FEC_RS)
+		ret |= (1 << MC_CMD_PHY_CAP_RS_FEC_LBN) |
+		       (1 << MC_CMD_PHY_CAP_RS_FEC_REQUESTED_LBN);
+	if (ethtool_cap & ETHTOOL_FEC_BASER)
+		ret |= (1 << MC_CMD_PHY_CAP_BASER_FEC_LBN) |
+		       (1 << MC_CMD_PHY_CAP_25G_BASER_FEC_LBN) |
+		       (1 << MC_CMD_PHY_CAP_BASER_FEC_REQUESTED_LBN) |
+		       (1 << MC_CMD_PHY_CAP_25G_BASER_FEC_REQUESTED_LBN);
+	return ret;
+}
+
+/* Invert ethtool_fec_caps_to_mcdi.  There are two combinations that function
+ * can never produce, (baser xor rs) and neither req; the implementation below
+ * maps both of those to AUTO.  This should never matter, and it's not clear
+ * what a better mapping would be anyway.
+ */
+u32 mcdi_fec_caps_to_ethtool(u32 caps, bool is_25g)
+{
+	bool rs = caps & (1 << MC_CMD_PHY_CAP_RS_FEC_LBN),
+	     rs_req = caps & (1 << MC_CMD_PHY_CAP_RS_FEC_REQUESTED_LBN),
+	     baser = is_25g ? caps & (1 << MC_CMD_PHY_CAP_25G_BASER_FEC_LBN)
+			    : caps & (1 << MC_CMD_PHY_CAP_BASER_FEC_LBN),
+	     baser_req = is_25g ? caps & (1 << MC_CMD_PHY_CAP_25G_BASER_FEC_REQUESTED_LBN)
+				: caps & (1 << MC_CMD_PHY_CAP_BASER_FEC_REQUESTED_LBN);
+
+	if (!baser && !rs)
+		return ETHTOOL_FEC_OFF;
+	return (rs_req ? ETHTOOL_FEC_RS : 0) |
+	       (baser_req ? ETHTOOL_FEC_BASER : 0) |
+	       (baser == baser_req && rs == rs_req ? 0 : ETHTOOL_FEC_AUTO);
+}
-- 
cgit v1.2.3-59-g8ed1b


From 1cf0f76adac1615f7e65bb4388ca3c1677720662 Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Thu, 9 Jan 2020 15:44:12 +0000
Subject: sfc: move more MCDI port code

Various functions dealing with flow control, forward error correction,
polling, port number, and PHY testing.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/mcdi_port.c        | 138 ---------------------------
 drivers/net/ethernet/sfc/mcdi_port_common.c | 139 ++++++++++++++++++++++++++++
 2 files changed, 139 insertions(+), 138 deletions(-)

diff --git a/drivers/net/ethernet/sfc/mcdi_port.c b/drivers/net/ethernet/sfc/mcdi_port.c
index d811753782dc..0db6068a668d 100644
--- a/drivers/net/ethernet/sfc/mcdi_port.c
+++ b/drivers/net/ethernet/sfc/mcdi_port.c
@@ -188,58 +188,6 @@ int efx_mcdi_port_reconfigure(struct efx_nic *efx)
 				 efx->loopback_mode, 0);
 }
 
-/* Verify that the forced flow control settings (!EFX_FC_AUTO) are
- * supported by the link partner. Warn the user if this isn't the case
- */
-void efx_mcdi_phy_check_fcntl(struct efx_nic *efx, u32 lpa)
-{
-	struct efx_mcdi_phy_data *phy_cfg = efx->phy_data;
-	u32 rmtadv;
-
-	/* The link partner capabilities are only relevant if the
-	 * link supports flow control autonegotiation */
-	if (~phy_cfg->supported_cap & (1 << MC_CMD_PHY_CAP_AN_LBN))
-		return;
-
-	/* If flow control autoneg is supported and enabled, then fine */
-	if (efx->wanted_fc & EFX_FC_AUTO)
-		return;
-
-	rmtadv = 0;
-	if (lpa & (1 << MC_CMD_PHY_CAP_PAUSE_LBN))
-		rmtadv |= ADVERTISED_Pause;
-	if (lpa & (1 << MC_CMD_PHY_CAP_ASYM_LBN))
-		rmtadv |=  ADVERTISED_Asym_Pause;
-
-	if ((efx->wanted_fc & EFX_FC_TX) && rmtadv == ADVERTISED_Asym_Pause)
-		netif_err(efx, link, efx->net_dev,
-			  "warning: link partner doesn't support pause frames");
-}
-
-bool efx_mcdi_phy_poll(struct efx_nic *efx)
-{
-	struct efx_link_state old_state = efx->link_state;
-	MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_LINK_OUT_LEN);
-	int rc;
-
-	WARN_ON(!mutex_is_locked(&efx->mac_lock));
-
-	BUILD_BUG_ON(MC_CMD_GET_LINK_IN_LEN != 0);
-
-	rc = efx_mcdi_rpc(efx, MC_CMD_GET_LINK, NULL, 0,
-			  outbuf, sizeof(outbuf), NULL);
-	if (rc)
-		efx->link_state.up = false;
-	else
-		efx_mcdi_phy_decode_link(
-			efx, &efx->link_state,
-			MCDI_DWORD(outbuf, GET_LINK_OUT_LINK_SPEED),
-			MCDI_DWORD(outbuf, GET_LINK_OUT_FLAGS),
-			MCDI_DWORD(outbuf, GET_LINK_OUT_FCNTL));
-
-	return !efx_link_state_equal(&efx->link_state, &old_state);
-}
-
 static void efx_mcdi_phy_remove(struct efx_nic *efx)
 {
 	struct efx_mcdi_phy_data *phy_data = efx->phy_data;
@@ -327,57 +275,6 @@ efx_mcdi_phy_set_link_ksettings(struct efx_nic *efx,
 	return 0;
 }
 
-int efx_mcdi_phy_get_fecparam(struct efx_nic *efx, struct ethtool_fecparam *fec)
-{
-	MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_LINK_OUT_V2_LEN);
-	u32 caps, active, speed; /* MCDI format */
-	bool is_25g = false;
-	size_t outlen;
-	int rc;
-
-	BUILD_BUG_ON(MC_CMD_GET_LINK_IN_LEN != 0);
-	rc = efx_mcdi_rpc(efx, MC_CMD_GET_LINK, NULL, 0,
-			  outbuf, sizeof(outbuf), &outlen);
-	if (rc)
-		return rc;
-	if (outlen < MC_CMD_GET_LINK_OUT_V2_LEN)
-		return -EOPNOTSUPP;
-
-	/* behaviour for 25G/50G links depends on 25G BASER bit */
-	speed = MCDI_DWORD(outbuf, GET_LINK_OUT_V2_LINK_SPEED);
-	is_25g = speed == 25000 || speed == 50000;
-
-	caps = MCDI_DWORD(outbuf, GET_LINK_OUT_V2_CAP);
-	fec->fec = mcdi_fec_caps_to_ethtool(caps, is_25g);
-	/* BASER is never supported on 100G */
-	if (speed == 100000)
-		fec->fec &= ~ETHTOOL_FEC_BASER;
-
-	active = MCDI_DWORD(outbuf, GET_LINK_OUT_V2_FEC_TYPE);
-	switch (active) {
-	case MC_CMD_FEC_NONE:
-		fec->active_fec = ETHTOOL_FEC_OFF;
-		break;
-	case MC_CMD_FEC_BASER:
-		fec->active_fec = ETHTOOL_FEC_BASER;
-		break;
-	case MC_CMD_FEC_RS:
-		fec->active_fec = ETHTOOL_FEC_RS;
-		break;
-	default:
-		netif_warn(efx, hw, efx->net_dev,
-			   "Firmware reports unrecognised FEC_TYPE %u\n",
-			   active);
-		/* We don't know what firmware has picked.  AUTO is as good a
-		 * "can't happen" value as any other.
-		 */
-		fec->active_fec = ETHTOOL_FEC_AUTO;
-		break;
-	}
-
-	return 0;
-}
-
 static int efx_mcdi_phy_set_fecparam(struct efx_nic *efx,
 				     const struct ethtool_fecparam *fec)
 {
@@ -405,27 +302,6 @@ static int efx_mcdi_phy_set_fecparam(struct efx_nic *efx,
 	return 0;
 }
 
-int efx_mcdi_phy_test_alive(struct efx_nic *efx)
-{
-	MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_PHY_STATE_OUT_LEN);
-	size_t outlen;
-	int rc;
-
-	BUILD_BUG_ON(MC_CMD_GET_PHY_STATE_IN_LEN != 0);
-
-	rc = efx_mcdi_rpc(efx, MC_CMD_GET_PHY_STATE, NULL, 0,
-			  outbuf, sizeof(outbuf), &outlen);
-	if (rc)
-		return rc;
-
-	if (outlen < MC_CMD_GET_PHY_STATE_OUT_LEN)
-		return -EIO;
-	if (MCDI_DWORD(outbuf, GET_PHY_STATE_OUT_STATE) != MC_CMD_PHY_STATE_OK)
-		return -EINVAL;
-
-	return 0;
-}
-
 static const char *const mcdi_sft9001_cable_diag_names[] = {
 	"cable.pairA.length",
 	"cable.pairB.length",
@@ -1008,17 +884,3 @@ void efx_mcdi_port_remove(struct efx_nic *efx)
 	efx->phy_op->remove(efx);
 	efx_nic_free_buffer(efx, &efx->stats_buffer);
 }
-
-/* Get physical port number (EF10 only; on Siena it is same as PF number) */
-int efx_mcdi_port_get_number(struct efx_nic *efx)
-{
-	MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_PORT_ASSIGNMENT_OUT_LEN);
-	int rc;
-
-	rc = efx_mcdi_rpc(efx, MC_CMD_GET_PORT_ASSIGNMENT, NULL, 0,
-			  outbuf, sizeof(outbuf), NULL);
-	if (rc)
-		return rc;
-
-	return MCDI_DWORD(outbuf, GET_PORT_ASSIGNMENT_OUT_PORT);
-}
diff --git a/drivers/net/ethernet/sfc/mcdi_port_common.c b/drivers/net/ethernet/sfc/mcdi_port_common.c
index 9763d59f7681..3b1c559246b4 100644
--- a/drivers/net/ethernet/sfc/mcdi_port_common.c
+++ b/drivers/net/ethernet/sfc/mcdi_port_common.c
@@ -348,3 +348,142 @@ u32 mcdi_fec_caps_to_ethtool(u32 caps, bool is_25g)
 	       (baser_req ? ETHTOOL_FEC_BASER : 0) |
 	       (baser == baser_req && rs == rs_req ? 0 : ETHTOOL_FEC_AUTO);
 }
+
+/* Verify that the forced flow control settings (!EFX_FC_AUTO) are
+ * supported by the link partner. Warn the user if this isn't the case
+ */
+void efx_mcdi_phy_check_fcntl(struct efx_nic *efx, u32 lpa)
+{
+	struct efx_mcdi_phy_data *phy_cfg = efx->phy_data;
+	u32 rmtadv;
+
+	/* The link partner capabilities are only relevant if the
+	 * link supports flow control autonegotiation
+	 */
+	if (~phy_cfg->supported_cap & (1 << MC_CMD_PHY_CAP_AN_LBN))
+		return;
+
+	/* If flow control autoneg is supported and enabled, then fine */
+	if (efx->wanted_fc & EFX_FC_AUTO)
+		return;
+
+	rmtadv = 0;
+	if (lpa & (1 << MC_CMD_PHY_CAP_PAUSE_LBN))
+		rmtadv |= ADVERTISED_Pause;
+	if (lpa & (1 << MC_CMD_PHY_CAP_ASYM_LBN))
+		rmtadv |=  ADVERTISED_Asym_Pause;
+
+	if ((efx->wanted_fc & EFX_FC_TX) && rmtadv == ADVERTISED_Asym_Pause)
+		netif_err(efx, link, efx->net_dev,
+			  "warning: link partner doesn't support pause frames");
+}
+
+bool efx_mcdi_phy_poll(struct efx_nic *efx)
+{
+	struct efx_link_state old_state = efx->link_state;
+	MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_LINK_OUT_LEN);
+	int rc;
+
+	WARN_ON(!mutex_is_locked(&efx->mac_lock));
+
+	BUILD_BUG_ON(MC_CMD_GET_LINK_IN_LEN != 0);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_GET_LINK, NULL, 0,
+			  outbuf, sizeof(outbuf), NULL);
+	if (rc)
+		efx->link_state.up = false;
+	else
+		efx_mcdi_phy_decode_link(
+			efx, &efx->link_state,
+			MCDI_DWORD(outbuf, GET_LINK_OUT_LINK_SPEED),
+			MCDI_DWORD(outbuf, GET_LINK_OUT_FLAGS),
+			MCDI_DWORD(outbuf, GET_LINK_OUT_FCNTL));
+
+	return !efx_link_state_equal(&efx->link_state, &old_state);
+}
+
+int efx_mcdi_phy_get_fecparam(struct efx_nic *efx, struct ethtool_fecparam *fec)
+{
+	MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_LINK_OUT_V2_LEN);
+	u32 caps, active, speed; /* MCDI format */
+	bool is_25g = false;
+	size_t outlen;
+	int rc;
+
+	BUILD_BUG_ON(MC_CMD_GET_LINK_IN_LEN != 0);
+	rc = efx_mcdi_rpc(efx, MC_CMD_GET_LINK, NULL, 0,
+			  outbuf, sizeof(outbuf), &outlen);
+	if (rc)
+		return rc;
+	if (outlen < MC_CMD_GET_LINK_OUT_V2_LEN)
+		return -EOPNOTSUPP;
+
+	/* behaviour for 25G/50G links depends on 25G BASER bit */
+	speed = MCDI_DWORD(outbuf, GET_LINK_OUT_V2_LINK_SPEED);
+	is_25g = speed == 25000 || speed == 50000;
+
+	caps = MCDI_DWORD(outbuf, GET_LINK_OUT_V2_CAP);
+	fec->fec = mcdi_fec_caps_to_ethtool(caps, is_25g);
+	/* BASER is never supported on 100G */
+	if (speed == 100000)
+		fec->fec &= ~ETHTOOL_FEC_BASER;
+
+	active = MCDI_DWORD(outbuf, GET_LINK_OUT_V2_FEC_TYPE);
+	switch (active) {
+	case MC_CMD_FEC_NONE:
+		fec->active_fec = ETHTOOL_FEC_OFF;
+		break;
+	case MC_CMD_FEC_BASER:
+		fec->active_fec = ETHTOOL_FEC_BASER;
+		break;
+	case MC_CMD_FEC_RS:
+		fec->active_fec = ETHTOOL_FEC_RS;
+		break;
+	default:
+		netif_warn(efx, hw, efx->net_dev,
+			   "Firmware reports unrecognised FEC_TYPE %u\n",
+			   active);
+		/* We don't know what firmware has picked.  AUTO is as good a
+		 * "can't happen" value as any other.
+		 */
+		fec->active_fec = ETHTOOL_FEC_AUTO;
+		break;
+	}
+
+	return 0;
+}
+
+int efx_mcdi_phy_test_alive(struct efx_nic *efx)
+{
+	MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_PHY_STATE_OUT_LEN);
+	size_t outlen;
+	int rc;
+
+	BUILD_BUG_ON(MC_CMD_GET_PHY_STATE_IN_LEN != 0);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_GET_PHY_STATE, NULL, 0,
+			  outbuf, sizeof(outbuf), &outlen);
+	if (rc)
+		return rc;
+
+	if (outlen < MC_CMD_GET_PHY_STATE_OUT_LEN)
+		return -EIO;
+	if (MCDI_DWORD(outbuf, GET_PHY_STATE_OUT_STATE) != MC_CMD_PHY_STATE_OK)
+		return -EINVAL;
+
+	return 0;
+}
+
+/* Get physical port number (EF10 only; on Siena it is same as PF number) */
+int efx_mcdi_port_get_number(struct efx_nic *efx)
+{
+	MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_PORT_ASSIGNMENT_OUT_LEN);
+	int rc;
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_GET_PORT_ASSIGNMENT, NULL, 0,
+			  outbuf, sizeof(outbuf), NULL);
+	if (rc)
+		return rc;
+
+	return MCDI_DWORD(outbuf, GET_PORT_ASSIGNMENT_OUT_PORT);
+}
-- 
cgit v1.2.3-59-g8ed1b


From 37a5f9dceb15ff56e5240bbc2f6c63b00aab768a Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Thu, 9 Jan 2020 15:44:27 +0000
Subject: sfc: move MCDI VI alloc/free code

One function was renamed here, the other contains code extracted from
another.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/Makefile         |  2 +-
 drivers/net/ethernet/sfc/ef10.c           | 45 +++--------------------
 drivers/net/ethernet/sfc/mcdi_functions.c | 61 +++++++++++++++++++++++++++++++
 3 files changed, 68 insertions(+), 40 deletions(-)
 create mode 100644 drivers/net/ethernet/sfc/mcdi_functions.c

diff --git a/drivers/net/ethernet/sfc/Makefile b/drivers/net/ethernet/sfc/Makefile
index f6d4b2b2dd83..ffa84b3ff8e2 100644
--- a/drivers/net/ethernet/sfc/Makefile
+++ b/drivers/net/ethernet/sfc/Makefile
@@ -4,7 +4,7 @@ sfc-y			+= efx.o efx_common.o efx_channels.o nic.o \
 			   tx.o tx_common.o rx.o rx_common.o \
 			   selftest.o ethtool.o ptp.o tx_tso.o \
 			   mcdi.o mcdi_port.o mcdi_port_common.o \
-			   mcdi_mon.o
+			   mcdi_functions.o mcdi_mon.o
 sfc-$(CONFIG_SFC_MTD)	+= mtd.o
 sfc-$(CONFIG_SFC_SRIOV)	+= sriov.o siena_sriov.o ef10_sriov.o
 
diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c
index d752ed34672d..f7df5ee801ef 100644
--- a/drivers/net/ethernet/sfc/ef10.c
+++ b/drivers/net/ethernet/sfc/ef10.c
@@ -11,6 +11,7 @@
 #include "mcdi.h"
 #include "mcdi_pcol.h"
 #include "mcdi_port_common.h"
+#include "mcdi_functions.h"
 #include "nic.h"
 #include "workarounds.h"
 #include "selftest.h"
@@ -834,22 +835,6 @@ fail1:
 	return rc;
 }
 
-static int efx_ef10_free_vis(struct efx_nic *efx)
-{
-	MCDI_DECLARE_BUF_ERR(outbuf);
-	size_t outlen;
-	int rc = efx_mcdi_rpc_quiet(efx, MC_CMD_FREE_VIS, NULL, 0,
-				    outbuf, sizeof(outbuf), &outlen);
-
-	/* -EALREADY means nothing to free, so ignore */
-	if (rc == -EALREADY)
-		rc = 0;
-	if (rc)
-		efx_mcdi_display_error(efx, MC_CMD_FREE_VIS, 0, outbuf, outlen,
-				       rc);
-	return rc;
-}
-
 #ifdef EFX_USE_PIO
 
 static void efx_ef10_free_piobufs(struct efx_nic *efx)
@@ -1092,7 +1077,7 @@ static void efx_ef10_remove(struct efx_nic *efx)
 	if (nic_data->wc_membase)
 		iounmap(nic_data->wc_membase);
 
-	rc = efx_ef10_free_vis(efx);
+	rc = efx_mcdi_free_vis(efx);
 	WARN_ON(rc != 0);
 
 	if (!nic_data->must_restore_piobufs)
@@ -1263,28 +1248,10 @@ static int efx_ef10_probe_vf(struct efx_nic *efx __attribute__ ((unused)))
 static int efx_ef10_alloc_vis(struct efx_nic *efx,
 			      unsigned int min_vis, unsigned int max_vis)
 {
-	MCDI_DECLARE_BUF(inbuf, MC_CMD_ALLOC_VIS_IN_LEN);
-	MCDI_DECLARE_BUF(outbuf, MC_CMD_ALLOC_VIS_OUT_LEN);
 	struct efx_ef10_nic_data *nic_data = efx->nic_data;
-	size_t outlen;
-	int rc;
-
-	MCDI_SET_DWORD(inbuf, ALLOC_VIS_IN_MIN_VI_COUNT, min_vis);
-	MCDI_SET_DWORD(inbuf, ALLOC_VIS_IN_MAX_VI_COUNT, max_vis);
-	rc = efx_mcdi_rpc(efx, MC_CMD_ALLOC_VIS, inbuf, sizeof(inbuf),
-			  outbuf, sizeof(outbuf), &outlen);
-	if (rc != 0)
-		return rc;
 
-	if (outlen < MC_CMD_ALLOC_VIS_OUT_LEN)
-		return -EIO;
-
-	netif_dbg(efx, drv, efx->net_dev, "base VI is A0x%03x\n",
-		  MCDI_DWORD(outbuf, ALLOC_VIS_OUT_VI_BASE));
-
-	nic_data->vi_base = MCDI_DWORD(outbuf, ALLOC_VIS_OUT_VI_BASE);
-	nic_data->n_allocated_vis = MCDI_DWORD(outbuf, ALLOC_VIS_OUT_VI_COUNT);
-	return 0;
+	return efx_mcdi_alloc_vis(efx, min_vis, max_vis, &nic_data->vi_base,
+				  &nic_data->n_allocated_vis);
 }
 
 /* Note that the failure path of this function does not free
@@ -1366,7 +1333,7 @@ static int efx_ef10_dimension_resources(struct efx_nic *efx)
 	}
 
 	/* In case the last attached driver failed to free VIs, do it now */
-	rc = efx_ef10_free_vis(efx);
+	rc = efx_mcdi_free_vis(efx);
 	if (rc != 0)
 		return rc;
 
@@ -1387,7 +1354,7 @@ static int efx_ef10_dimension_resources(struct efx_nic *efx)
 		efx->max_tx_channels =
 			nic_data->n_allocated_vis / EFX_TXQ_TYPES;
 
-		efx_ef10_free_vis(efx);
+		efx_mcdi_free_vis(efx);
 		return -EAGAIN;
 	}
 
diff --git a/drivers/net/ethernet/sfc/mcdi_functions.c b/drivers/net/ethernet/sfc/mcdi_functions.c
new file mode 100644
index 000000000000..65a4689337db
--- /dev/null
+++ b/drivers/net/ethernet/sfc/mcdi_functions.c
@@ -0,0 +1,61 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/****************************************************************************
+ * Driver for Solarflare network controllers and boards
+ * Copyright 2019 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#include "net_driver.h"
+#include "efx.h"
+#include "nic.h"
+#include "mcdi_functions.h"
+#include "mcdi.h"
+#include "mcdi_pcol.h"
+
+int efx_mcdi_free_vis(struct efx_nic *efx)
+{
+	MCDI_DECLARE_BUF_ERR(outbuf);
+	size_t outlen;
+	int rc = efx_mcdi_rpc_quiet(efx, MC_CMD_FREE_VIS, NULL, 0,
+				    outbuf, sizeof(outbuf), &outlen);
+
+	/* -EALREADY means nothing to free, so ignore */
+	if (rc == -EALREADY)
+		rc = 0;
+	if (rc)
+		efx_mcdi_display_error(efx, MC_CMD_FREE_VIS, 0, outbuf, outlen,
+				       rc);
+	return rc;
+}
+
+int efx_mcdi_alloc_vis(struct efx_nic *efx, unsigned int min_vis,
+		       unsigned int max_vis, unsigned int *vi_base,
+		       unsigned int *allocated_vis)
+{
+	MCDI_DECLARE_BUF(outbuf, MC_CMD_ALLOC_VIS_OUT_LEN);
+	MCDI_DECLARE_BUF(inbuf, MC_CMD_ALLOC_VIS_IN_LEN);
+	size_t outlen;
+	int rc;
+
+	MCDI_SET_DWORD(inbuf, ALLOC_VIS_IN_MIN_VI_COUNT, min_vis);
+	MCDI_SET_DWORD(inbuf, ALLOC_VIS_IN_MAX_VI_COUNT, max_vis);
+	rc = efx_mcdi_rpc(efx, MC_CMD_ALLOC_VIS, inbuf, sizeof(inbuf),
+			  outbuf, sizeof(outbuf), &outlen);
+	if (rc != 0)
+		return rc;
+
+	if (outlen < MC_CMD_ALLOC_VIS_OUT_LEN)
+		return -EIO;
+
+	netif_dbg(efx, drv, efx->net_dev, "base VI is A0x%03x\n",
+		  MCDI_DWORD(outbuf, ALLOC_VIS_OUT_VI_BASE));
+
+	if (vi_base)
+		*vi_base = MCDI_DWORD(outbuf, ALLOC_VIS_OUT_VI_BASE);
+	if (allocated_vis)
+		*allocated_vis = MCDI_DWORD(outbuf, ALLOC_VIS_OUT_VI_COUNT);
+	return 0;
+}
-- 
cgit v1.2.3-59-g8ed1b


From 4438b587fe4bcddb64ecc44f8f1f3ddce834d645 Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Thu, 9 Jan 2020 15:44:43 +0000
Subject: sfc: move MCDI event queue management code

A function was split, the others were renamed.

Code style fixes included.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/ef10.c           | 117 ++++--------------------------
 drivers/net/ethernet/sfc/mcdi_functions.c | 103 ++++++++++++++++++++++++++
 2 files changed, 116 insertions(+), 104 deletions(-)

diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c
index f7df5ee801ef..e21ee5ccea6a 100644
--- a/drivers/net/ethernet/sfc/ef10.c
+++ b/drivers/net/ethernet/sfc/ef10.c
@@ -3199,106 +3199,20 @@ efx_ef10_rx_defer_refill_complete(struct efx_nic *efx, unsigned long cookie,
 	/* nothing to do */
 }
 
-static int efx_ef10_ev_probe(struct efx_channel *channel)
-{
-	return efx_nic_alloc_buffer(channel->efx, &channel->eventq.buf,
-				    (channel->eventq_mask + 1) *
-				    sizeof(efx_qword_t),
-				    GFP_KERNEL);
-}
-
-static void efx_ef10_ev_fini(struct efx_channel *channel)
-{
-	MCDI_DECLARE_BUF(inbuf, MC_CMD_FINI_EVQ_IN_LEN);
-	MCDI_DECLARE_BUF_ERR(outbuf);
-	struct efx_nic *efx = channel->efx;
-	size_t outlen;
-	int rc;
-
-	MCDI_SET_DWORD(inbuf, FINI_EVQ_IN_INSTANCE, channel->channel);
-
-	rc = efx_mcdi_rpc_quiet(efx, MC_CMD_FINI_EVQ, inbuf, sizeof(inbuf),
-			  outbuf, sizeof(outbuf), &outlen);
-
-	if (rc && rc != -EALREADY)
-		goto fail;
-
-	return;
-
-fail:
-	efx_mcdi_display_error(efx, MC_CMD_FINI_EVQ, MC_CMD_FINI_EVQ_IN_LEN,
-			       outbuf, outlen, rc);
-}
-
 static int efx_ef10_ev_init(struct efx_channel *channel)
 {
-	MCDI_DECLARE_BUF(inbuf,
-			 MC_CMD_INIT_EVQ_V2_IN_LEN(EFX_MAX_EVQ_SIZE * 8 /
-						   EFX_BUF_SIZE));
-	MCDI_DECLARE_BUF(outbuf, MC_CMD_INIT_EVQ_V2_OUT_LEN);
-	size_t entries = channel->eventq.buf.len / EFX_BUF_SIZE;
 	struct efx_nic *efx = channel->efx;
 	struct efx_ef10_nic_data *nic_data;
-	size_t inlen, outlen;
 	unsigned int enabled, implemented;
-	dma_addr_t dma_addr;
+	bool use_v2, cut_thru;
 	int rc;
-	int i;
 
 	nic_data = efx->nic_data;
-
-	/* Fill event queue with all ones (i.e. empty events) */
-	memset(channel->eventq.buf.addr, 0xff, channel->eventq.buf.len);
-
-	MCDI_SET_DWORD(inbuf, INIT_EVQ_IN_SIZE, channel->eventq_mask + 1);
-	MCDI_SET_DWORD(inbuf, INIT_EVQ_IN_INSTANCE, channel->channel);
-	/* INIT_EVQ expects index in vector table, not absolute */
-	MCDI_SET_DWORD(inbuf, INIT_EVQ_IN_IRQ_NUM, channel->channel);
-	MCDI_SET_DWORD(inbuf, INIT_EVQ_IN_TMR_MODE,
-		       MC_CMD_INIT_EVQ_IN_TMR_MODE_DIS);
-	MCDI_SET_DWORD(inbuf, INIT_EVQ_IN_TMR_LOAD, 0);
-	MCDI_SET_DWORD(inbuf, INIT_EVQ_IN_TMR_RELOAD, 0);
-	MCDI_SET_DWORD(inbuf, INIT_EVQ_IN_COUNT_MODE,
-		       MC_CMD_INIT_EVQ_IN_COUNT_MODE_DIS);
-	MCDI_SET_DWORD(inbuf, INIT_EVQ_IN_COUNT_THRSHLD, 0);
-
-	if (nic_data->datapath_caps2 &
-	    1 << MC_CMD_GET_CAPABILITIES_V2_OUT_INIT_EVQ_V2_LBN) {
-		/* Use the new generic approach to specifying event queue
-		 * configuration, requesting lower latency or higher throughput.
-		 * The options that actually get used appear in the output.
-		 */
-		MCDI_POPULATE_DWORD_2(inbuf, INIT_EVQ_V2_IN_FLAGS,
-				      INIT_EVQ_V2_IN_FLAG_INTERRUPTING, 1,
-				      INIT_EVQ_V2_IN_FLAG_TYPE,
-				      MC_CMD_INIT_EVQ_V2_IN_FLAG_TYPE_AUTO);
-	} else {
-		bool cut_thru = !(nic_data->datapath_caps &
-			1 << MC_CMD_GET_CAPABILITIES_OUT_RX_BATCHING_LBN);
-
-		MCDI_POPULATE_DWORD_4(inbuf, INIT_EVQ_IN_FLAGS,
-				      INIT_EVQ_IN_FLAG_INTERRUPTING, 1,
-				      INIT_EVQ_IN_FLAG_RX_MERGE, 1,
-				      INIT_EVQ_IN_FLAG_TX_MERGE, 1,
-				      INIT_EVQ_IN_FLAG_CUT_THRU, cut_thru);
-	}
-
-	dma_addr = channel->eventq.buf.dma_addr;
-	for (i = 0; i < entries; ++i) {
-		MCDI_SET_ARRAY_QWORD(inbuf, INIT_EVQ_IN_DMA_ADDR, i, dma_addr);
-		dma_addr += EFX_BUF_SIZE;
-	}
-
-	inlen = MC_CMD_INIT_EVQ_IN_LEN(entries);
-
-	rc = efx_mcdi_rpc(efx, MC_CMD_INIT_EVQ, inbuf, inlen,
-			  outbuf, sizeof(outbuf), &outlen);
-
-	if (outlen >= MC_CMD_INIT_EVQ_V2_OUT_LEN)
-		netif_dbg(efx, drv, efx->net_dev,
-			  "Channel %d using event queue flags %08x\n",
-			  channel->channel,
-			  MCDI_DWORD(outbuf, INIT_EVQ_V2_OUT_FLAGS));
+	use_v2 = nic_data->datapath_caps2 &
+			    1 << MC_CMD_GET_CAPABILITIES_V2_OUT_INIT_EVQ_V2_LBN;
+	cut_thru = !(nic_data->datapath_caps &
+			      1 << MC_CMD_GET_CAPABILITIES_OUT_RX_BATCHING_LBN);
+	rc = efx_mcdi_ev_init(channel, cut_thru, use_v2);
 
 	/* IRQ return is ignored */
 	if (channel->channel || rc)
@@ -3356,15 +3270,10 @@ static int efx_ef10_ev_init(struct efx_channel *channel)
 		return 0;
 
 fail:
-	efx_ef10_ev_fini(channel);
+	efx_mcdi_ev_fini(channel);
 	return rc;
 }
 
-static void efx_ef10_ev_remove(struct efx_channel *channel)
-{
-	efx_nic_free_buffer(channel->efx, &channel->eventq.buf);
-}
-
 static void efx_ef10_handle_rx_wrong_queue(struct efx_rx_queue *rx_queue,
 					   unsigned int rx_queue_label)
 {
@@ -6630,10 +6539,10 @@ const struct efx_nic_type efx_hunt_a0_vf_nic_type = {
 	.rx_remove = efx_ef10_rx_remove,
 	.rx_write = efx_ef10_rx_write,
 	.rx_defer_refill = efx_ef10_rx_defer_refill,
-	.ev_probe = efx_ef10_ev_probe,
+	.ev_probe = efx_mcdi_ev_probe,
 	.ev_init = efx_ef10_ev_init,
-	.ev_fini = efx_ef10_ev_fini,
-	.ev_remove = efx_ef10_ev_remove,
+	.ev_fini = efx_mcdi_ev_fini,
+	.ev_remove = efx_mcdi_ev_remove,
 	.ev_process = efx_ef10_ev_process,
 	.ev_read_ack = efx_ef10_ev_read_ack,
 	.ev_test_generate = efx_ef10_ev_test_generate,
@@ -6742,10 +6651,10 @@ const struct efx_nic_type efx_hunt_a0_nic_type = {
 	.rx_remove = efx_ef10_rx_remove,
 	.rx_write = efx_ef10_rx_write,
 	.rx_defer_refill = efx_ef10_rx_defer_refill,
-	.ev_probe = efx_ef10_ev_probe,
+	.ev_probe = efx_mcdi_ev_probe,
 	.ev_init = efx_ef10_ev_init,
-	.ev_fini = efx_ef10_ev_fini,
-	.ev_remove = efx_ef10_ev_remove,
+	.ev_fini = efx_mcdi_ev_fini,
+	.ev_remove = efx_mcdi_ev_remove,
 	.ev_process = efx_ef10_ev_process,
 	.ev_read_ack = efx_ef10_ev_read_ack,
 	.ev_test_generate = efx_ef10_ev_test_generate,
diff --git a/drivers/net/ethernet/sfc/mcdi_functions.c b/drivers/net/ethernet/sfc/mcdi_functions.c
index 65a4689337db..a1d37aec6013 100644
--- a/drivers/net/ethernet/sfc/mcdi_functions.c
+++ b/drivers/net/ethernet/sfc/mcdi_functions.c
@@ -59,3 +59,106 @@ int efx_mcdi_alloc_vis(struct efx_nic *efx, unsigned int min_vis,
 		*allocated_vis = MCDI_DWORD(outbuf, ALLOC_VIS_OUT_VI_COUNT);
 	return 0;
 }
+
+int efx_mcdi_ev_probe(struct efx_channel *channel)
+{
+	return efx_nic_alloc_buffer(channel->efx, &channel->eventq.buf,
+				    (channel->eventq_mask + 1) *
+				    sizeof(efx_qword_t),
+				    GFP_KERNEL);
+}
+
+int efx_mcdi_ev_init(struct efx_channel *channel, bool v1_cut_thru, bool v2)
+{
+	MCDI_DECLARE_BUF(inbuf,
+			 MC_CMD_INIT_EVQ_V2_IN_LEN(EFX_MAX_EVQ_SIZE * 8 /
+						   EFX_BUF_SIZE));
+	MCDI_DECLARE_BUF(outbuf, MC_CMD_INIT_EVQ_V2_OUT_LEN);
+	size_t entries = channel->eventq.buf.len / EFX_BUF_SIZE;
+	struct efx_nic *efx = channel->efx;
+	struct efx_ef10_nic_data *nic_data;
+	size_t inlen, outlen;
+	dma_addr_t dma_addr;
+	int rc, i;
+
+	nic_data = efx->nic_data;
+
+	/* Fill event queue with all ones (i.e. empty events) */
+	memset(channel->eventq.buf.addr, 0xff, channel->eventq.buf.len);
+
+	MCDI_SET_DWORD(inbuf, INIT_EVQ_IN_SIZE, channel->eventq_mask + 1);
+	MCDI_SET_DWORD(inbuf, INIT_EVQ_IN_INSTANCE, channel->channel);
+	/* INIT_EVQ expects index in vector table, not absolute */
+	MCDI_SET_DWORD(inbuf, INIT_EVQ_IN_IRQ_NUM, channel->channel);
+	MCDI_SET_DWORD(inbuf, INIT_EVQ_IN_TMR_MODE,
+		       MC_CMD_INIT_EVQ_IN_TMR_MODE_DIS);
+	MCDI_SET_DWORD(inbuf, INIT_EVQ_IN_TMR_LOAD, 0);
+	MCDI_SET_DWORD(inbuf, INIT_EVQ_IN_TMR_RELOAD, 0);
+	MCDI_SET_DWORD(inbuf, INIT_EVQ_IN_COUNT_MODE,
+		       MC_CMD_INIT_EVQ_IN_COUNT_MODE_DIS);
+	MCDI_SET_DWORD(inbuf, INIT_EVQ_IN_COUNT_THRSHLD, 0);
+
+	if (v2) {
+		/* Use the new generic approach to specifying event queue
+		 * configuration, requesting lower latency or higher throughput.
+		 * The options that actually get used appear in the output.
+		 */
+		MCDI_POPULATE_DWORD_2(inbuf, INIT_EVQ_V2_IN_FLAGS,
+				      INIT_EVQ_V2_IN_FLAG_INTERRUPTING, 1,
+				      INIT_EVQ_V2_IN_FLAG_TYPE,
+				      MC_CMD_INIT_EVQ_V2_IN_FLAG_TYPE_AUTO);
+	} else {
+		MCDI_POPULATE_DWORD_4(inbuf, INIT_EVQ_IN_FLAGS,
+				      INIT_EVQ_IN_FLAG_INTERRUPTING, 1,
+				      INIT_EVQ_IN_FLAG_RX_MERGE, 1,
+				      INIT_EVQ_IN_FLAG_TX_MERGE, 1,
+				      INIT_EVQ_IN_FLAG_CUT_THRU, v1_cut_thru);
+	}
+
+	dma_addr = channel->eventq.buf.dma_addr;
+	for (i = 0; i < entries; ++i) {
+		MCDI_SET_ARRAY_QWORD(inbuf, INIT_EVQ_IN_DMA_ADDR, i, dma_addr);
+		dma_addr += EFX_BUF_SIZE;
+	}
+
+	inlen = MC_CMD_INIT_EVQ_IN_LEN(entries);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_INIT_EVQ, inbuf, inlen,
+			  outbuf, sizeof(outbuf), &outlen);
+
+	if (outlen >= MC_CMD_INIT_EVQ_V2_OUT_LEN)
+		netif_dbg(efx, drv, efx->net_dev,
+			  "Channel %d using event queue flags %08x\n",
+			  channel->channel,
+			  MCDI_DWORD(outbuf, INIT_EVQ_V2_OUT_FLAGS));
+
+	return rc;
+}
+
+void efx_mcdi_ev_remove(struct efx_channel *channel)
+{
+	efx_nic_free_buffer(channel->efx, &channel->eventq.buf);
+}
+
+void efx_mcdi_ev_fini(struct efx_channel *channel)
+{
+	MCDI_DECLARE_BUF(inbuf, MC_CMD_FINI_EVQ_IN_LEN);
+	MCDI_DECLARE_BUF_ERR(outbuf);
+	struct efx_nic *efx = channel->efx;
+	size_t outlen;
+	int rc;
+
+	MCDI_SET_DWORD(inbuf, FINI_EVQ_IN_INSTANCE, channel->channel);
+
+	rc = efx_mcdi_rpc_quiet(efx, MC_CMD_FINI_EVQ, inbuf, sizeof(inbuf),
+				outbuf, sizeof(outbuf), &outlen);
+
+	if (rc && rc != -EALREADY)
+		goto fail;
+
+	return;
+
+fail:
+	efx_mcdi_display_error(efx, MC_CMD_FINI_EVQ, MC_CMD_FINI_EVQ_IN_LEN,
+			       outbuf, outlen, rc);
+}
-- 
cgit v1.2.3-59-g8ed1b


From 8ee4c907725cb09e1872dd4203e0a4266dd7e637 Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Thu, 9 Jan 2020 15:44:59 +0000
Subject: sfc: move MCDI transmit queue management code

A function was split, the others were renamed.

Code style fixes included.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/ef10.c           |  94 +++-------------------------
 drivers/net/ethernet/sfc/mcdi_functions.c | 100 ++++++++++++++++++++++++++++++
 2 files changed, 109 insertions(+), 85 deletions(-)

diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c
index e21ee5ccea6a..aeb6c2059eb1 100644
--- a/drivers/net/ethernet/sfc/ef10.c
+++ b/drivers/net/ethernet/sfc/ef10.c
@@ -2378,20 +2378,15 @@ static u32 efx_ef10_tso_versions(struct efx_nic *efx)
 
 static void efx_ef10_tx_init(struct efx_tx_queue *tx_queue)
 {
-	MCDI_DECLARE_BUF(inbuf, MC_CMD_INIT_TXQ_IN_LEN(EFX_MAX_DMAQ_SIZE * 8 /
-						       EFX_BUF_SIZE));
 	bool csum_offload = tx_queue->queue & EFX_TXQ_TYPE_OFFLOAD;
-	size_t entries = tx_queue->txd.buf.len / EFX_BUF_SIZE;
 	struct efx_channel *channel = tx_queue->channel;
 	struct efx_nic *efx = tx_queue->efx;
-	struct efx_ef10_nic_data *nic_data = efx->nic_data;
+	struct efx_ef10_nic_data *nic_data;
 	bool tso_v2 = false;
-	size_t inlen;
-	dma_addr_t dma_addr;
 	efx_qword_t *txd;
 	int rc;
-	int i;
-	BUILD_BUG_ON(MC_CMD_INIT_TXQ_OUT_LEN != 0);
+
+	nic_data = efx->nic_data;
 
 	/* Only attempt to enable TX timestamping if we have the license for it,
 	 * otherwise TXQ init will fail
@@ -2418,51 +2413,9 @@ static void efx_ef10_tx_init(struct efx_tx_queue *tx_queue)
 				channel->channel);
 	}
 
-	MCDI_SET_DWORD(inbuf, INIT_TXQ_IN_SIZE, tx_queue->ptr_mask + 1);
-	MCDI_SET_DWORD(inbuf, INIT_TXQ_IN_TARGET_EVQ, channel->channel);
-	MCDI_SET_DWORD(inbuf, INIT_TXQ_IN_LABEL, tx_queue->queue);
-	MCDI_SET_DWORD(inbuf, INIT_TXQ_IN_INSTANCE, tx_queue->queue);
-	MCDI_SET_DWORD(inbuf, INIT_TXQ_IN_OWNER_ID, 0);
-	MCDI_SET_DWORD(inbuf, INIT_TXQ_IN_PORT_ID, nic_data->vport_id);
-
-	dma_addr = tx_queue->txd.buf.dma_addr;
-
-	netif_dbg(efx, hw, efx->net_dev, "pushing TXQ %d. %zu entries (%llx)\n",
-		  tx_queue->queue, entries, (u64)dma_addr);
-
-	for (i = 0; i < entries; ++i) {
-		MCDI_SET_ARRAY_QWORD(inbuf, INIT_TXQ_IN_DMA_ADDR, i, dma_addr);
-		dma_addr += EFX_BUF_SIZE;
-	}
-
-	inlen = MC_CMD_INIT_TXQ_IN_LEN(entries);
-
-	do {
-		MCDI_POPULATE_DWORD_4(inbuf, INIT_TXQ_IN_FLAGS,
-				/* This flag was removed from mcdi_pcol.h for
-				 * the non-_EXT version of INIT_TXQ.  However,
-				 * firmware still honours it.
-				 */
-				INIT_TXQ_EXT_IN_FLAG_TSOV2_EN, tso_v2,
-				INIT_TXQ_IN_FLAG_IP_CSUM_DIS, !csum_offload,
-				INIT_TXQ_IN_FLAG_TCP_CSUM_DIS, !csum_offload,
-				INIT_TXQ_EXT_IN_FLAG_TIMESTAMP,
-						tx_queue->timestamping);
-
-		rc = efx_mcdi_rpc_quiet(efx, MC_CMD_INIT_TXQ, inbuf, inlen,
-					NULL, 0, NULL);
-		if (rc == -ENOSPC && tso_v2) {
-			/* Retry without TSOv2 if we're short on contexts. */
-			tso_v2 = false;
-			netif_warn(efx, probe, efx->net_dev,
-				   "TSOv2 context not available to segment in hardware. TCP performance may be reduced.\n");
-		} else if (rc) {
-			efx_mcdi_display_error(efx, MC_CMD_INIT_TXQ,
-					       MC_CMD_INIT_TXQ_EXT_IN_LEN,
-					       NULL, 0, rc);
-			goto fail;
-		}
-	} while (rc);
+	rc = efx_mcdi_tx_init(tx_queue, tso_v2);
+	if (rc)
+		goto fail;
 
 	/* A previous user of this TX queue might have set us up the
 	 * bomb by writing a descriptor to the TX push collector but
@@ -2500,35 +2453,6 @@ fail:
 		    tx_queue->queue);
 }
 
-static void efx_ef10_tx_fini(struct efx_tx_queue *tx_queue)
-{
-	MCDI_DECLARE_BUF(inbuf, MC_CMD_FINI_TXQ_IN_LEN);
-	MCDI_DECLARE_BUF_ERR(outbuf);
-	struct efx_nic *efx = tx_queue->efx;
-	size_t outlen;
-	int rc;
-
-	MCDI_SET_DWORD(inbuf, FINI_TXQ_IN_INSTANCE,
-		       tx_queue->queue);
-
-	rc = efx_mcdi_rpc_quiet(efx, MC_CMD_FINI_TXQ, inbuf, sizeof(inbuf),
-			  outbuf, sizeof(outbuf), &outlen);
-
-	if (rc && rc != -EALREADY)
-		goto fail;
-
-	return;
-
-fail:
-	efx_mcdi_display_error(efx, MC_CMD_FINI_TXQ, MC_CMD_FINI_TXQ_IN_LEN,
-			       outbuf, outlen, rc);
-}
-
-static void efx_ef10_tx_remove(struct efx_tx_queue *tx_queue)
-{
-	efx_nic_free_buffer(tx_queue->efx, &tx_queue->txd.buf);
-}
-
 /* This writes to the TX_DESC_WPTR; write pointer for TX descriptor ring */
 static inline void efx_ef10_notify_tx_desc(struct efx_tx_queue *tx_queue)
 {
@@ -3857,7 +3781,7 @@ static int efx_ef10_fini_dmaq(struct efx_nic *efx)
 			efx_for_each_channel_rx_queue(rx_queue, channel)
 				efx_ef10_rx_fini(rx_queue);
 			efx_for_each_channel_tx_queue(tx_queue, channel)
-				efx_ef10_tx_fini(tx_queue);
+				efx_mcdi_tx_fini(tx_queue);
 		}
 
 		wait_event_timeout(efx->flush_wq,
@@ -6529,7 +6453,7 @@ const struct efx_nic_type efx_hunt_a0_vf_nic_type = {
 	.irq_handle_legacy = efx_ef10_legacy_interrupt,
 	.tx_probe = efx_ef10_tx_probe,
 	.tx_init = efx_ef10_tx_init,
-	.tx_remove = efx_ef10_tx_remove,
+	.tx_remove = efx_mcdi_tx_remove,
 	.tx_write = efx_ef10_tx_write,
 	.tx_limit_len = efx_ef10_tx_limit_len,
 	.rx_push_rss_config = efx_ef10_vf_rx_push_rss_config,
@@ -6638,7 +6562,7 @@ const struct efx_nic_type efx_hunt_a0_nic_type = {
 	.irq_handle_legacy = efx_ef10_legacy_interrupt,
 	.tx_probe = efx_ef10_tx_probe,
 	.tx_init = efx_ef10_tx_init,
-	.tx_remove = efx_ef10_tx_remove,
+	.tx_remove = efx_mcdi_tx_remove,
 	.tx_write = efx_ef10_tx_write,
 	.tx_limit_len = efx_ef10_tx_limit_len,
 	.rx_push_rss_config = efx_ef10_pf_rx_push_rss_config,
diff --git a/drivers/net/ethernet/sfc/mcdi_functions.c b/drivers/net/ethernet/sfc/mcdi_functions.c
index a1d37aec6013..9dc1395e5f68 100644
--- a/drivers/net/ethernet/sfc/mcdi_functions.c
+++ b/drivers/net/ethernet/sfc/mcdi_functions.c
@@ -162,3 +162,103 @@ fail:
 	efx_mcdi_display_error(efx, MC_CMD_FINI_EVQ, MC_CMD_FINI_EVQ_IN_LEN,
 			       outbuf, outlen, rc);
 }
+
+int efx_mcdi_tx_init(struct efx_tx_queue *tx_queue, bool tso_v2)
+{
+	MCDI_DECLARE_BUF(inbuf, MC_CMD_INIT_TXQ_IN_LEN(EFX_MAX_DMAQ_SIZE * 8 /
+						       EFX_BUF_SIZE));
+	bool csum_offload = tx_queue->queue & EFX_TXQ_TYPE_OFFLOAD;
+	size_t entries = tx_queue->txd.buf.len / EFX_BUF_SIZE;
+	struct efx_channel *channel = tx_queue->channel;
+	struct efx_nic *efx = tx_queue->efx;
+	struct efx_ef10_nic_data *nic_data;
+	dma_addr_t dma_addr;
+	size_t inlen;
+	int rc, i;
+
+	BUILD_BUG_ON(MC_CMD_INIT_TXQ_OUT_LEN != 0);
+
+	nic_data = efx->nic_data;
+
+	MCDI_SET_DWORD(inbuf, INIT_TXQ_IN_SIZE, tx_queue->ptr_mask + 1);
+	MCDI_SET_DWORD(inbuf, INIT_TXQ_IN_TARGET_EVQ, channel->channel);
+	MCDI_SET_DWORD(inbuf, INIT_TXQ_IN_LABEL, tx_queue->queue);
+	MCDI_SET_DWORD(inbuf, INIT_TXQ_IN_INSTANCE, tx_queue->queue);
+	MCDI_SET_DWORD(inbuf, INIT_TXQ_IN_OWNER_ID, 0);
+	MCDI_SET_DWORD(inbuf, INIT_TXQ_IN_PORT_ID, nic_data->vport_id);
+
+	dma_addr = tx_queue->txd.buf.dma_addr;
+
+	netif_dbg(efx, hw, efx->net_dev, "pushing TXQ %d. %zu entries (%llx)\n",
+		  tx_queue->queue, entries, (u64)dma_addr);
+
+	for (i = 0; i < entries; ++i) {
+		MCDI_SET_ARRAY_QWORD(inbuf, INIT_TXQ_IN_DMA_ADDR, i, dma_addr);
+		dma_addr += EFX_BUF_SIZE;
+	}
+
+	inlen = MC_CMD_INIT_TXQ_IN_LEN(entries);
+
+	do {
+		MCDI_POPULATE_DWORD_4(inbuf, INIT_TXQ_IN_FLAGS,
+				/* This flag was removed from mcdi_pcol.h for
+				 * the non-_EXT version of INIT_TXQ.  However,
+				 * firmware still honours it.
+				 */
+				INIT_TXQ_EXT_IN_FLAG_TSOV2_EN, tso_v2,
+				INIT_TXQ_IN_FLAG_IP_CSUM_DIS, !csum_offload,
+				INIT_TXQ_IN_FLAG_TCP_CSUM_DIS, !csum_offload,
+				INIT_TXQ_EXT_IN_FLAG_TIMESTAMP,
+						tx_queue->timestamping);
+
+		rc = efx_mcdi_rpc_quiet(efx, MC_CMD_INIT_TXQ, inbuf, inlen,
+					NULL, 0, NULL);
+		if (rc == -ENOSPC && tso_v2) {
+			/* Retry without TSOv2 if we're short on contexts. */
+			tso_v2 = false;
+			netif_warn(efx, probe, efx->net_dev,
+				   "TSOv2 context not available to segment in "
+				   "hardware. TCP performance may be reduced.\n"
+				   );
+		} else if (rc) {
+			efx_mcdi_display_error(efx, MC_CMD_INIT_TXQ,
+					       MC_CMD_INIT_TXQ_EXT_IN_LEN,
+					       NULL, 0, rc);
+			goto fail;
+		}
+	} while (rc);
+
+	return 0;
+
+fail:
+	return rc;
+}
+
+void efx_mcdi_tx_remove(struct efx_tx_queue *tx_queue)
+{
+	efx_nic_free_buffer(tx_queue->efx, &tx_queue->txd.buf);
+}
+
+void efx_mcdi_tx_fini(struct efx_tx_queue *tx_queue)
+{
+	MCDI_DECLARE_BUF(inbuf, MC_CMD_FINI_TXQ_IN_LEN);
+	MCDI_DECLARE_BUF_ERR(outbuf);
+	struct efx_nic *efx = tx_queue->efx;
+	size_t outlen;
+	int rc;
+
+	MCDI_SET_DWORD(inbuf, FINI_TXQ_IN_INSTANCE,
+		       tx_queue->queue);
+
+	rc = efx_mcdi_rpc_quiet(efx, MC_CMD_FINI_TXQ, inbuf, sizeof(inbuf),
+				outbuf, sizeof(outbuf), &outlen);
+
+	if (rc && rc != -EALREADY)
+		goto fail;
+
+	return;
+
+fail:
+	efx_mcdi_display_error(efx, MC_CMD_FINI_TXQ, MC_CMD_FINI_TXQ_IN_LEN,
+			       outbuf, outlen, rc);
+}
-- 
cgit v1.2.3-59-g8ed1b


From 8da92642050df4d466cacd04bbb1d6232b070022 Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Thu, 9 Jan 2020 15:45:15 +0000
Subject: sfc: move MCDI receive queue management code

One function's prototype was changed in the header.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/ef10.c           | 99 +++----------------------------
 drivers/net/ethernet/sfc/mcdi_functions.c | 85 ++++++++++++++++++++++++++
 drivers/net/ethernet/sfc/mcdi_functions.h |  2 +-
 3 files changed, 93 insertions(+), 93 deletions(-)

diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c
index aeb6c2059eb1..dc037dd927f8 100644
--- a/drivers/net/ethernet/sfc/ef10.c
+++ b/drivers/net/ethernet/sfc/ef10.c
@@ -2967,91 +2967,6 @@ static int efx_ef10_vf_rx_push_rss_config(struct efx_nic *efx, bool user,
 	return efx_ef10_rx_push_shared_rss_config(efx, NULL);
 }
 
-static int efx_ef10_rx_probe(struct efx_rx_queue *rx_queue)
-{
-	return efx_nic_alloc_buffer(rx_queue->efx, &rx_queue->rxd.buf,
-				    (rx_queue->ptr_mask + 1) *
-				    sizeof(efx_qword_t),
-				    GFP_KERNEL);
-}
-
-static void efx_ef10_rx_init(struct efx_rx_queue *rx_queue)
-{
-	MCDI_DECLARE_BUF(inbuf,
-			 MC_CMD_INIT_RXQ_IN_LEN(EFX_MAX_DMAQ_SIZE * 8 /
-						EFX_BUF_SIZE));
-	struct efx_channel *channel = efx_rx_queue_channel(rx_queue);
-	size_t entries = rx_queue->rxd.buf.len / EFX_BUF_SIZE;
-	struct efx_nic *efx = rx_queue->efx;
-	struct efx_ef10_nic_data *nic_data = efx->nic_data;
-	size_t inlen;
-	dma_addr_t dma_addr;
-	int rc;
-	int i;
-	BUILD_BUG_ON(MC_CMD_INIT_RXQ_OUT_LEN != 0);
-
-	rx_queue->scatter_n = 0;
-	rx_queue->scatter_len = 0;
-
-	MCDI_SET_DWORD(inbuf, INIT_RXQ_IN_SIZE, rx_queue->ptr_mask + 1);
-	MCDI_SET_DWORD(inbuf, INIT_RXQ_IN_TARGET_EVQ, channel->channel);
-	MCDI_SET_DWORD(inbuf, INIT_RXQ_IN_LABEL, efx_rx_queue_index(rx_queue));
-	MCDI_SET_DWORD(inbuf, INIT_RXQ_IN_INSTANCE,
-		       efx_rx_queue_index(rx_queue));
-	MCDI_POPULATE_DWORD_2(inbuf, INIT_RXQ_IN_FLAGS,
-			      INIT_RXQ_IN_FLAG_PREFIX, 1,
-			      INIT_RXQ_IN_FLAG_TIMESTAMP, 1);
-	MCDI_SET_DWORD(inbuf, INIT_RXQ_IN_OWNER_ID, 0);
-	MCDI_SET_DWORD(inbuf, INIT_RXQ_IN_PORT_ID, nic_data->vport_id);
-
-	dma_addr = rx_queue->rxd.buf.dma_addr;
-
-	netif_dbg(efx, hw, efx->net_dev, "pushing RXQ %d. %zu entries (%llx)\n",
-		  efx_rx_queue_index(rx_queue), entries, (u64)dma_addr);
-
-	for (i = 0; i < entries; ++i) {
-		MCDI_SET_ARRAY_QWORD(inbuf, INIT_RXQ_IN_DMA_ADDR, i, dma_addr);
-		dma_addr += EFX_BUF_SIZE;
-	}
-
-	inlen = MC_CMD_INIT_RXQ_IN_LEN(entries);
-
-	rc = efx_mcdi_rpc(efx, MC_CMD_INIT_RXQ, inbuf, inlen,
-			  NULL, 0, NULL);
-	if (rc)
-		netdev_WARN(efx->net_dev, "failed to initialise RXQ %d\n",
-			    efx_rx_queue_index(rx_queue));
-}
-
-static void efx_ef10_rx_fini(struct efx_rx_queue *rx_queue)
-{
-	MCDI_DECLARE_BUF(inbuf, MC_CMD_FINI_RXQ_IN_LEN);
-	MCDI_DECLARE_BUF_ERR(outbuf);
-	struct efx_nic *efx = rx_queue->efx;
-	size_t outlen;
-	int rc;
-
-	MCDI_SET_DWORD(inbuf, FINI_RXQ_IN_INSTANCE,
-		       efx_rx_queue_index(rx_queue));
-
-	rc = efx_mcdi_rpc_quiet(efx, MC_CMD_FINI_RXQ, inbuf, sizeof(inbuf),
-			  outbuf, sizeof(outbuf), &outlen);
-
-	if (rc && rc != -EALREADY)
-		goto fail;
-
-	return;
-
-fail:
-	efx_mcdi_display_error(efx, MC_CMD_FINI_RXQ, MC_CMD_FINI_RXQ_IN_LEN,
-			       outbuf, outlen, rc);
-}
-
-static void efx_ef10_rx_remove(struct efx_rx_queue *rx_queue)
-{
-	efx_nic_free_buffer(rx_queue->efx, &rx_queue->rxd.buf);
-}
-
 /* This creates an entry in the RX descriptor queue */
 static inline void
 efx_ef10_build_rx_desc(struct efx_rx_queue *rx_queue, unsigned int index)
@@ -3779,7 +3694,7 @@ static int efx_ef10_fini_dmaq(struct efx_nic *efx)
 	if (efx->state != STATE_RECOVERY) {
 		efx_for_each_channel(channel, efx) {
 			efx_for_each_channel_rx_queue(rx_queue, channel)
-				efx_ef10_rx_fini(rx_queue);
+				efx_mcdi_rx_fini(rx_queue);
 			efx_for_each_channel_tx_queue(tx_queue, channel)
 				efx_mcdi_tx_fini(tx_queue);
 		}
@@ -6458,9 +6373,9 @@ const struct efx_nic_type efx_hunt_a0_vf_nic_type = {
 	.tx_limit_len = efx_ef10_tx_limit_len,
 	.rx_push_rss_config = efx_ef10_vf_rx_push_rss_config,
 	.rx_pull_rss_config = efx_ef10_rx_pull_rss_config,
-	.rx_probe = efx_ef10_rx_probe,
-	.rx_init = efx_ef10_rx_init,
-	.rx_remove = efx_ef10_rx_remove,
+	.rx_probe = efx_mcdi_rx_probe,
+	.rx_init = efx_mcdi_rx_init,
+	.rx_remove = efx_mcdi_rx_remove,
 	.rx_write = efx_ef10_rx_write,
 	.rx_defer_refill = efx_ef10_rx_defer_refill,
 	.ev_probe = efx_mcdi_ev_probe,
@@ -6570,9 +6485,9 @@ const struct efx_nic_type efx_hunt_a0_nic_type = {
 	.rx_push_rss_context_config = efx_ef10_rx_push_rss_context_config,
 	.rx_pull_rss_context_config = efx_ef10_rx_pull_rss_context_config,
 	.rx_restore_rss_contexts = efx_ef10_rx_restore_rss_contexts,
-	.rx_probe = efx_ef10_rx_probe,
-	.rx_init = efx_ef10_rx_init,
-	.rx_remove = efx_ef10_rx_remove,
+	.rx_probe = efx_mcdi_rx_probe,
+	.rx_init = efx_mcdi_rx_init,
+	.rx_remove = efx_mcdi_rx_remove,
 	.rx_write = efx_ef10_rx_write,
 	.rx_defer_refill = efx_ef10_rx_defer_refill,
 	.ev_probe = efx_mcdi_ev_probe,
diff --git a/drivers/net/ethernet/sfc/mcdi_functions.c b/drivers/net/ethernet/sfc/mcdi_functions.c
index 9dc1395e5f68..f022e2b9e975 100644
--- a/drivers/net/ethernet/sfc/mcdi_functions.c
+++ b/drivers/net/ethernet/sfc/mcdi_functions.c
@@ -262,3 +262,88 @@ fail:
 	efx_mcdi_display_error(efx, MC_CMD_FINI_TXQ, MC_CMD_FINI_TXQ_IN_LEN,
 			       outbuf, outlen, rc);
 }
+
+int efx_mcdi_rx_probe(struct efx_rx_queue *rx_queue)
+{
+	return efx_nic_alloc_buffer(rx_queue->efx, &rx_queue->rxd.buf,
+				    (rx_queue->ptr_mask + 1) *
+				    sizeof(efx_qword_t),
+				    GFP_KERNEL);
+}
+
+void efx_mcdi_rx_init(struct efx_rx_queue *rx_queue)
+{
+	MCDI_DECLARE_BUF(inbuf,
+			 MC_CMD_INIT_RXQ_IN_LEN(EFX_MAX_DMAQ_SIZE * 8 /
+						EFX_BUF_SIZE));
+	struct efx_channel *channel = efx_rx_queue_channel(rx_queue);
+	size_t entries = rx_queue->rxd.buf.len / EFX_BUF_SIZE;
+	struct efx_nic *efx = rx_queue->efx;
+	struct efx_ef10_nic_data *nic_data = efx->nic_data;
+	dma_addr_t dma_addr;
+	size_t inlen;
+	int rc;
+	int i;
+	BUILD_BUG_ON(MC_CMD_INIT_RXQ_OUT_LEN != 0);
+
+	rx_queue->scatter_n = 0;
+	rx_queue->scatter_len = 0;
+
+	MCDI_SET_DWORD(inbuf, INIT_RXQ_IN_SIZE, rx_queue->ptr_mask + 1);
+	MCDI_SET_DWORD(inbuf, INIT_RXQ_IN_TARGET_EVQ, channel->channel);
+	MCDI_SET_DWORD(inbuf, INIT_RXQ_IN_LABEL, efx_rx_queue_index(rx_queue));
+	MCDI_SET_DWORD(inbuf, INIT_RXQ_IN_INSTANCE,
+		       efx_rx_queue_index(rx_queue));
+	MCDI_POPULATE_DWORD_2(inbuf, INIT_RXQ_IN_FLAGS,
+			      INIT_RXQ_IN_FLAG_PREFIX, 1,
+			      INIT_RXQ_IN_FLAG_TIMESTAMP, 1);
+	MCDI_SET_DWORD(inbuf, INIT_RXQ_IN_OWNER_ID, 0);
+	MCDI_SET_DWORD(inbuf, INIT_RXQ_IN_PORT_ID, nic_data->vport_id);
+
+	dma_addr = rx_queue->rxd.buf.dma_addr;
+
+	netif_dbg(efx, hw, efx->net_dev, "pushing RXQ %d. %zu entries (%llx)\n",
+		  efx_rx_queue_index(rx_queue), entries, (u64)dma_addr);
+
+	for (i = 0; i < entries; ++i) {
+		MCDI_SET_ARRAY_QWORD(inbuf, INIT_RXQ_IN_DMA_ADDR, i, dma_addr);
+		dma_addr += EFX_BUF_SIZE;
+	}
+
+	inlen = MC_CMD_INIT_RXQ_IN_LEN(entries);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_INIT_RXQ, inbuf, inlen,
+			  NULL, 0, NULL);
+	if (rc)
+		netdev_WARN(efx->net_dev, "failed to initialise RXQ %d\n",
+			    efx_rx_queue_index(rx_queue));
+}
+
+void efx_mcdi_rx_remove(struct efx_rx_queue *rx_queue)
+{
+	efx_nic_free_buffer(rx_queue->efx, &rx_queue->rxd.buf);
+}
+
+void efx_mcdi_rx_fini(struct efx_rx_queue *rx_queue)
+{
+	MCDI_DECLARE_BUF(inbuf, MC_CMD_FINI_RXQ_IN_LEN);
+	MCDI_DECLARE_BUF_ERR(outbuf);
+	struct efx_nic *efx = rx_queue->efx;
+	size_t outlen;
+	int rc;
+
+	MCDI_SET_DWORD(inbuf, FINI_RXQ_IN_INSTANCE,
+		       efx_rx_queue_index(rx_queue));
+
+	rc = efx_mcdi_rpc_quiet(efx, MC_CMD_FINI_RXQ, inbuf, sizeof(inbuf),
+				outbuf, sizeof(outbuf), &outlen);
+
+	if (rc && rc != -EALREADY)
+		goto fail;
+
+	return;
+
+fail:
+	efx_mcdi_display_error(efx, MC_CMD_FINI_RXQ, MC_CMD_FINI_RXQ_IN_LEN,
+			       outbuf, outlen, rc);
+}
diff --git a/drivers/net/ethernet/sfc/mcdi_functions.h b/drivers/net/ethernet/sfc/mcdi_functions.h
index f0726c71698b..3c9e760238e7 100644
--- a/drivers/net/ethernet/sfc/mcdi_functions.h
+++ b/drivers/net/ethernet/sfc/mcdi_functions.h
@@ -23,7 +23,7 @@ int efx_mcdi_tx_init(struct efx_tx_queue *tx_queue, bool tso_v2);
 void efx_mcdi_tx_remove(struct efx_tx_queue *tx_queue);
 void efx_mcdi_tx_fini(struct efx_tx_queue *tx_queue);
 int efx_mcdi_rx_probe(struct efx_rx_queue *rx_queue);
-int efx_mcdi_rx_init(struct efx_rx_queue *rx_queue, bool want_outer_classes);
+void efx_mcdi_rx_init(struct efx_rx_queue *rx_queue);
 void efx_mcdi_rx_remove(struct efx_rx_queue *rx_queue);
 void efx_mcdi_rx_fini(struct efx_rx_queue *rx_queue);
 
-- 
cgit v1.2.3-59-g8ed1b


From b5775b476ef7b28a3895335a2b8533e10ba36b16 Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Thu, 9 Jan 2020 15:45:29 +0000
Subject: sfc: conditioned some functionality

Before calling certain function pointers, check that they are non-NULL.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/efx_channels.c |  6 +++--
 drivers/net/ethernet/sfc/efx_common.c   | 47 ++++++++++++++++++++-------------
 2 files changed, 32 insertions(+), 21 deletions(-)

diff --git a/drivers/net/ethernet/sfc/efx_channels.c b/drivers/net/ethernet/sfc/efx_channels.c
index 21bd71daf5a0..aeb5e8aa2f2a 100644
--- a/drivers/net/ethernet/sfc/efx_channels.c
+++ b/drivers/net/ethernet/sfc/efx_channels.c
@@ -1039,7 +1039,7 @@ void efx_stop_channels(struct efx_nic *efx)
 	struct efx_tx_queue *tx_queue;
 	struct efx_rx_queue *rx_queue;
 	struct efx_channel *channel;
-	int rc;
+	int rc = 0;
 
 	/* Stop RX refill */
 	efx_for_each_channel(channel, efx) {
@@ -1060,7 +1060,9 @@ void efx_stop_channels(struct efx_nic *efx)
 		}
 	}
 
-	rc = efx->type->fini_dmaq(efx);
+	if (efx->type->fini_dmaq)
+		rc = efx->type->fini_dmaq(efx);
+
 	if (rc) {
 		netif_err(efx, drv, efx->net_dev, "failed to flush queues\n");
 	} else {
diff --git a/drivers/net/ethernet/sfc/efx_common.c b/drivers/net/ethernet/sfc/efx_common.c
index fe74c66c8ec6..3add2b577503 100644
--- a/drivers/net/ethernet/sfc/efx_common.c
+++ b/drivers/net/ethernet/sfc/efx_common.c
@@ -141,9 +141,11 @@ void efx_destroy_reset_workqueue(void)
  */
 void efx_mac_reconfigure(struct efx_nic *efx)
 {
-	down_read(&efx->filter_sem);
-	efx->type->reconfigure_mac(efx);
-	up_read(&efx->filter_sem);
+	if (efx->type->reconfigure_mac) {
+		down_read(&efx->filter_sem);
+		efx->type->reconfigure_mac(efx);
+		up_read(&efx->filter_sem);
+	}
 }
 
 /* Asynchronous work item for changing MAC promiscuity and multicast
@@ -296,7 +298,8 @@ static void efx_start_datapath(struct efx_nic *efx)
 		netdev_features_change(efx->net_dev);
 
 	/* RX filters may also have scatter-enabled flags */
-	if (efx->rx_scatter != old_rx_scatter)
+	if ((efx->rx_scatter != old_rx_scatter) &&
+	    efx->type->filter_update_rx_scatter)
 		efx->type->filter_update_rx_scatter(efx);
 
 	/* We must keep at least one descriptor in a TX ring empty.
@@ -405,11 +408,13 @@ void efx_start_all(struct efx_nic *efx)
 		efx_link_status_changed(efx);
 	mutex_unlock(&efx->mac_lock);
 
-	efx->type->start_stats(efx);
-	efx->type->pull_stats(efx);
-	spin_lock_bh(&efx->stats_lock);
-	efx->type->update_stats(efx, NULL, NULL);
-	spin_unlock_bh(&efx->stats_lock);
+	if (efx->type->start_stats) {
+		efx->type->start_stats(efx);
+		efx->type->pull_stats(efx);
+		spin_lock_bh(&efx->stats_lock);
+		efx->type->update_stats(efx, NULL, NULL);
+		spin_unlock_bh(&efx->stats_lock);
+	}
 }
 
 /* Quiesce the hardware and software data path, and regular activity
@@ -425,14 +430,17 @@ void efx_stop_all(struct efx_nic *efx)
 	if (!efx->port_enabled)
 		return;
 
-	/* update stats before we go down so we can accurately count
-	 * rx_nodesc_drops
-	 */
-	efx->type->pull_stats(efx);
-	spin_lock_bh(&efx->stats_lock);
-	efx->type->update_stats(efx, NULL, NULL);
-	spin_unlock_bh(&efx->stats_lock);
-	efx->type->stop_stats(efx);
+	if (efx->type->update_stats) {
+		/* update stats before we go down so we can accurately count
+		 * rx_nodesc_drops
+		 */
+		efx->type->pull_stats(efx);
+		spin_lock_bh(&efx->stats_lock);
+		efx->type->update_stats(efx, NULL, NULL);
+		spin_unlock_bh(&efx->stats_lock);
+		efx->type->stop_stats(efx);
+	}
+
 	efx_stop_port(efx);
 
 	/* Stop the kernel transmit interface.  This is only valid if
@@ -456,7 +464,7 @@ void efx_stop_all(struct efx_nic *efx)
 int __efx_reconfigure_port(struct efx_nic *efx)
 {
 	enum efx_phy_mode phy_mode;
-	int rc;
+	int rc = 0;
 
 	WARN_ON(!mutex_is_locked(&efx->mac_lock));
 
@@ -467,7 +475,8 @@ int __efx_reconfigure_port(struct efx_nic *efx)
 	else
 		efx->phy_mode &= ~PHY_MODE_TX_DISABLED;
 
-	rc = efx->type->reconfigure_port(efx);
+	if (efx->type->reconfigure_port)
+		rc = efx->type->reconfigure_port(efx);
 
 	if (rc)
 		efx->phy_mode = phy_mode;
-- 
cgit v1.2.3-59-g8ed1b


From b69f7a3e53ff4bc966e56b62af955fdfa10ffa7d Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Thu, 9 Jan 2020 15:45:45 +0000
Subject: sfc: move MCDI logging device attribute

A few bits were extracted from other functions.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/efx.c        | 40 +++--------------------------------
 drivers/net/ethernet/sfc/efx_common.c | 39 ++++++++++++++++++++++++++++++++++
 drivers/net/ethernet/sfc/efx_common.h |  8 +++++++
 3 files changed, 50 insertions(+), 37 deletions(-)

diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index 33b6ed03b3ea..4affae76b03c 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -1036,28 +1036,6 @@ show_phy_type(struct device *dev, struct device_attribute *attr, char *buf)
 }
 static DEVICE_ATTR(phy_type, 0444, show_phy_type, NULL);
 
-#ifdef CONFIG_SFC_MCDI_LOGGING
-static ssize_t show_mcdi_log(struct device *dev, struct device_attribute *attr,
-			     char *buf)
-{
-	struct efx_nic *efx = dev_get_drvdata(dev);
-	struct efx_mcdi_iface *mcdi = efx_mcdi(efx);
-
-	return scnprintf(buf, PAGE_SIZE, "%d\n", mcdi->logging_enabled);
-}
-static ssize_t set_mcdi_log(struct device *dev, struct device_attribute *attr,
-			    const char *buf, size_t count)
-{
-	struct efx_nic *efx = dev_get_drvdata(dev);
-	struct efx_mcdi_iface *mcdi = efx_mcdi(efx);
-	bool enable = count > 0 && *buf != '0';
-
-	mcdi->logging_enabled = enable;
-	return count;
-}
-static DEVICE_ATTR(mcdi_logging, 0644, show_mcdi_log, set_mcdi_log);
-#endif
-
 static int efx_register_netdev(struct efx_nic *efx)
 {
 	struct net_device *net_dev = efx->net_dev;
@@ -1117,21 +1095,11 @@ static int efx_register_netdev(struct efx_nic *efx)
 			  "failed to init net dev attributes\n");
 		goto fail_registered;
 	}
-#ifdef CONFIG_SFC_MCDI_LOGGING
-	rc = device_create_file(&efx->pci_dev->dev, &dev_attr_mcdi_logging);
-	if (rc) {
-		netif_err(efx, drv, efx->net_dev,
-			  "failed to init net dev attributes\n");
-		goto fail_attr_mcdi_logging;
-	}
-#endif
+
+	efx_init_mcdi_logging(efx);
 
 	return 0;
 
-#ifdef CONFIG_SFC_MCDI_LOGGING
-fail_attr_mcdi_logging:
-	device_remove_file(&efx->pci_dev->dev, &dev_attr_phy_type);
-#endif
 fail_registered:
 	rtnl_lock();
 	efx_dissociate(efx);
@@ -1152,9 +1120,7 @@ static void efx_unregister_netdev(struct efx_nic *efx)
 
 	if (efx_dev_registered(efx)) {
 		strlcpy(efx->name, pci_name(efx->pci_dev), sizeof(efx->name));
-#ifdef CONFIG_SFC_MCDI_LOGGING
-		device_remove_file(&efx->pci_dev->dev, &dev_attr_mcdi_logging);
-#endif
+		efx_fini_mcdi_logging(efx);
 		device_remove_file(&efx->pci_dev->dev, &dev_attr_phy_type);
 		unregister_netdev(efx->net_dev);
 	}
diff --git a/drivers/net/ethernet/sfc/efx_common.c b/drivers/net/ethernet/sfc/efx_common.c
index 3add2b577503..9cd7c1550c08 100644
--- a/drivers/net/ethernet/sfc/efx_common.c
+++ b/drivers/net/ethernet/sfc/efx_common.c
@@ -1006,3 +1006,42 @@ void efx_fini_io(struct efx_nic *efx, int bar)
 	if (!pci_vfs_assigned(efx->pci_dev))
 		pci_disable_device(efx->pci_dev);
 }
+
+#ifdef CONFIG_SFC_MCDI_LOGGING
+static ssize_t show_mcdi_log(struct device *dev, struct device_attribute *attr,
+			     char *buf)
+{
+	struct efx_nic *efx = dev_get_drvdata(dev);
+	struct efx_mcdi_iface *mcdi = efx_mcdi(efx);
+
+	return scnprintf(buf, PAGE_SIZE, "%d\n", mcdi->logging_enabled);
+}
+
+static ssize_t set_mcdi_log(struct device *dev, struct device_attribute *attr,
+			    const char *buf, size_t count)
+{
+	struct efx_nic *efx = dev_get_drvdata(dev);
+	struct efx_mcdi_iface *mcdi = efx_mcdi(efx);
+	bool enable = count > 0 && *buf != '0';
+
+	mcdi->logging_enabled = enable;
+	return count;
+}
+
+static DEVICE_ATTR(mcdi_logging, 0644, show_mcdi_log, set_mcdi_log);
+
+void efx_init_mcdi_logging(struct efx_nic *efx)
+{
+	int rc = device_create_file(&efx->pci_dev->dev, &dev_attr_mcdi_logging);
+
+	if (rc) {
+		netif_warn(efx, drv, efx->net_dev,
+			   "failed to init net dev attributes\n");
+	}
+}
+
+void efx_fini_mcdi_logging(struct efx_nic *efx)
+{
+	device_remove_file(&efx->pci_dev->dev, &dev_attr_mcdi_logging);
+}
+#endif
diff --git a/drivers/net/ethernet/sfc/efx_common.h b/drivers/net/ethernet/sfc/efx_common.h
index c602e5257088..e03404d1dc0a 100644
--- a/drivers/net/ethernet/sfc/efx_common.h
+++ b/drivers/net/ethernet/sfc/efx_common.h
@@ -55,6 +55,14 @@ static inline int efx_check_disabled(struct efx_nic *efx)
 	return 0;
 }
 
+#ifdef CONFIG_SFC_MCDI_LOGGING
+void efx_init_mcdi_logging(struct efx_nic *efx);
+void efx_fini_mcdi_logging(struct efx_nic *efx);
+#else
+static inline void efx_init_mcdi_logging(struct efx_nic *efx) {}
+static inline void efx_fini_mcdi_logging(struct efx_nic *efx) {}
+#endif
+
 void efx_mac_reconfigure(struct efx_nic *efx);
 void efx_link_status_changed(struct efx_nic *efx);
 
-- 
cgit v1.2.3-59-g8ed1b


From dde8010be0a0ed9823cc1ddf75ef8448910821a9 Mon Sep 17 00:00:00 2001
From: Changqi Du <d.changqi@gmail.com>
Date: Wed, 8 Jan 2020 23:40:19 -0500
Subject: Bluetooth: btbcm : Fix warning about missing blank lines after
 declarations

This patches fixes two warnings of checkpatch.pl, both of the type
WARNING: Missing a blank line after declarations

Signed-off-by: Changqi Du <d.changqi@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 drivers/bluetooth/btbcm.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/bluetooth/btbcm.c b/drivers/bluetooth/btbcm.c
index 0795a49edfae..1f498f358f60 100644
--- a/drivers/bluetooth/btbcm.c
+++ b/drivers/bluetooth/btbcm.c
@@ -36,6 +36,7 @@ int btbcm_check_bdaddr(struct hci_dev *hdev)
 			     HCI_INIT_TIMEOUT);
 	if (IS_ERR(skb)) {
 		int err = PTR_ERR(skb);
+
 		bt_dev_err(hdev, "BCM: Reading device address failed (%d)", err);
 		return err;
 	}
@@ -223,6 +224,7 @@ static int btbcm_reset(struct hci_dev *hdev)
 	skb = __hci_cmd_sync(hdev, HCI_OP_RESET, 0, NULL, HCI_INIT_TIMEOUT);
 	if (IS_ERR(skb)) {
 		int err = PTR_ERR(skb);
+
 		bt_dev_err(hdev, "BCM: Reset failed (%d)", err);
 		return err;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 22895f8a0ef2d6b66eaa20eae2653c45410f9e51 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Thu, 9 Jan 2020 21:27:21 +0200
Subject: mlxsw: spectrum: Update firmware version to xx.2000.2714

The version adds support for 2x50 Gb/s port split option on SN3800
systems.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index 431c3765b545..38d16042e7a8 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -49,7 +49,7 @@
 
 #define MLXSW_SP1_FWREV_MAJOR 13
 #define MLXSW_SP1_FWREV_MINOR 2000
-#define MLXSW_SP1_FWREV_SUBMINOR 2308
+#define MLXSW_SP1_FWREV_SUBMINOR 2714
 #define MLXSW_SP1_FWREV_CAN_RESET_MINOR 1702
 
 static const struct mlxsw_fw_rev mlxsw_sp1_fw_rev = {
@@ -66,7 +66,7 @@ static const struct mlxsw_fw_rev mlxsw_sp1_fw_rev = {
 
 #define MLXSW_SP2_FWREV_MAJOR 29
 #define MLXSW_SP2_FWREV_MINOR 2000
-#define MLXSW_SP2_FWREV_SUBMINOR 2308
+#define MLXSW_SP2_FWREV_SUBMINOR 2714
 
 static const struct mlxsw_fw_rev mlxsw_sp2_fw_rev = {
 	.major = MLXSW_SP2_FWREV_MAJOR,
-- 
cgit v1.2.3-59-g8ed1b


From 4dab08df3b08ec236f9a159393741626546d1f95 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Thu, 9 Jan 2020 21:27:22 +0200
Subject: mlxsw: spectrum: Only require minimum firmware version

Currently, the driver ensures that the firmware version found on the
device matches the branch of the required version.

Remove this limitation so that the driver will accept the required
version or a newer version, from any branch.

This will allow us to reduce the frequency in which we need to update
the required version. New firmware versions that include necessary bug
fixes will be able to work with the driver, even if they are not from
the required branch.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum.c | 11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index 38d16042e7a8..2f0b516ee03f 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -45,8 +45,6 @@
 #include "spectrum_ptp.h"
 #include "../mlxfw/mlxfw.h"
 
-#define MLXSW_SP_FWREV_MINOR_TO_BRANCH(minor) ((minor) / 100)
-
 #define MLXSW_SP1_FWREV_MAJOR 13
 #define MLXSW_SP1_FWREV_MINOR 2000
 #define MLXSW_SP1_FWREV_SUBMINOR 2714
@@ -423,13 +421,12 @@ static int mlxsw_sp_fw_rev_validate(struct mlxsw_sp *mlxsw_sp)
 		     rev->major, req_rev->major);
 		return -EINVAL;
 	}
-	if (MLXSW_SP_FWREV_MINOR_TO_BRANCH(rev->minor) ==
-	    MLXSW_SP_FWREV_MINOR_TO_BRANCH(req_rev->minor) &&
-	    mlxsw_core_fw_rev_minor_subminor_validate(rev, req_rev))
+	if (mlxsw_core_fw_rev_minor_subminor_validate(rev, req_rev))
 		return 0;
 
-	dev_info(mlxsw_sp->bus_info->dev, "The firmware version %d.%d.%d is incompatible with the driver\n",
-		 rev->major, rev->minor, rev->subminor);
+	dev_err(mlxsw_sp->bus_info->dev, "The firmware version %d.%d.%d is incompatible with the driver (required >= %d.%d.%d)\n",
+		rev->major, rev->minor, rev->subminor, req_rev->major,
+		req_rev->minor, req_rev->subminor);
 	dev_info(mlxsw_sp->bus_info->dev, "Flashing firmware using file %s\n",
 		 fw_filename);
 
-- 
cgit v1.2.3-59-g8ed1b


From 2cab57e0653f684dc8bd0c1495c5ee32a211af82 Mon Sep 17 00:00:00 2001
From: Dejin Zheng <zhengdejin5@gmail.com>
Date: Wed, 8 Jan 2020 21:56:49 +0800
Subject: net: stmmac: pci: remove the duplicate code of set phy_mask

All members of mdio_bus_data are cleared to 0 when it was obtained
by devm_kzalloc(). so It doesn't need to set phy_mask as 0 again.

Signed-off-by: Dejin Zheng <zhengdejin5@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c
index 8237dbc3e991..40f171d310d7 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c
@@ -65,7 +65,6 @@ static void common_default_data(struct plat_stmmacenet_data *plat)
 	plat->force_sf_dma_mode = 1;
 
 	plat->mdio_bus_data->needs_reset = true;
-	plat->mdio_bus_data->phy_mask = 0;
 
 	/* Set default value for multicast hash bins */
 	plat->multicast_filter_bins = HASH_TABLE_SIZE;
@@ -154,8 +153,6 @@ static int intel_mgbe_common_data(struct pci_dev *pdev,
 	plat->tx_queues_cfg[6].weight = 0x0F;
 	plat->tx_queues_cfg[7].weight = 0x10;
 
-	plat->mdio_bus_data->phy_mask = 0;
-
 	plat->dma_cfg->pbl = 32;
 	plat->dma_cfg->pblx8 = true;
 	plat->dma_cfg->fixed_burst = 0;
@@ -386,8 +383,6 @@ static int snps_gmac5_default_data(struct pci_dev *pdev,
 	plat->tso_en = 1;
 	plat->pmt = 1;
 
-	plat->mdio_bus_data->phy_mask = 0;
-
 	/* Set default value for multicast hash bins */
 	plat->multicast_filter_bins = HASH_TABLE_SIZE;
 
-- 
cgit v1.2.3-59-g8ed1b


From 5e0fcc16e5c563fd8f16738efec974f81e0a5ea0 Mon Sep 17 00:00:00 2001
From: Vijay Khemka <vijaykhemka@fb.com>
Date: Wed, 8 Jan 2020 15:43:40 -0800
Subject: net/ncsi: Support for multi host mellanox card

Multi host Mellanox cards require MAC affinity to be set
before receiving any config commands. All config commands
should also have unicast address for source address in
command header.

Adding GMA and SMAF(Set Mac Affinity) for Mellanox card
and call these in channel probe state machine if it is
defined in device tree.

Signed-off-by: Vijay Khemka <vijaykhemka@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ncsi/internal.h    | 20 +++++++++++++++
 net/ncsi/ncsi-manage.c | 69 ++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 89 insertions(+)

diff --git a/net/ncsi/internal.h b/net/ncsi/internal.h
index ad3fd7f1da75..e37102546be6 100644
--- a/net/ncsi/internal.h
+++ b/net/ncsi/internal.h
@@ -64,6 +64,17 @@ enum {
 	NCSI_MODE_MAX
 };
 
+/* Supported media status bits for Mellanox Mac affinity command.
+ * Bit (0-2) for different protocol support; Bit 1 for RBT support,
+ * bit 1 for SMBUS support and bit 2 for PCIE support. Bit (3-5)
+ * for different protocol availability. Bit 4 for RBT, bit 4 for
+ * SMBUS and bit 5 for PCIE.
+ */
+enum {
+	MLX_MC_RBT_SUPPORT  = 0x01, /* MC supports RBT         */
+	MLX_MC_RBT_AVL      = 0x08, /* RBT medium is available */
+};
+
 /* OEM Vendor Manufacture ID */
 #define NCSI_OEM_MFR_MLX_ID             0x8119
 #define NCSI_OEM_MFR_BCM_ID             0x113d
@@ -72,9 +83,15 @@ enum {
 /* Mellanox specific OEM Command */
 #define NCSI_OEM_MLX_CMD_GMA            0x00   /* CMD ID for Get MAC */
 #define NCSI_OEM_MLX_CMD_GMA_PARAM      0x1b   /* Parameter for GMA  */
+#define NCSI_OEM_MLX_CMD_SMAF           0x01   /* CMD ID for Set MC Affinity */
+#define NCSI_OEM_MLX_CMD_SMAF_PARAM     0x07   /* Parameter for SMAF         */
 /* OEM Command payload lengths*/
 #define NCSI_OEM_BCM_CMD_GMA_LEN        12
 #define NCSI_OEM_MLX_CMD_GMA_LEN        8
+#define NCSI_OEM_MLX_CMD_SMAF_LEN        60
+/* Offset in OEM request */
+#define MLX_SMAF_MAC_ADDR_OFFSET         8     /* Offset for MAC in SMAF    */
+#define MLX_SMAF_MED_SUPPORT_OFFSET      14    /* Offset for medium in SMAF */
 /* Mac address offset in OEM response */
 #define BCM_MAC_ADDR_OFFSET             28
 #define MLX_MAC_ADDR_OFFSET             8
@@ -251,6 +268,8 @@ enum {
 	ncsi_dev_state_probe_deselect	= 0x0201,
 	ncsi_dev_state_probe_package,
 	ncsi_dev_state_probe_channel,
+	ncsi_dev_state_probe_mlx_gma,
+	ncsi_dev_state_probe_mlx_smaf,
 	ncsi_dev_state_probe_cis,
 	ncsi_dev_state_probe_gvi,
 	ncsi_dev_state_probe_gc,
@@ -311,6 +330,7 @@ struct ncsi_dev_priv {
 	struct list_head    vlan_vids;       /* List of active VLAN IDs */
 
 	bool                multi_package;   /* Enable multiple packages   */
+	bool                mlx_multi_host;  /* Enable multi host Mellanox */
 	u32                 package_whitelist; /* Packages to configure    */
 };
 
diff --git a/net/ncsi/ncsi-manage.c b/net/ncsi/ncsi-manage.c
index e20b81514029..1f387be7827b 100644
--- a/net/ncsi/ncsi-manage.c
+++ b/net/ncsi/ncsi-manage.c
@@ -8,6 +8,8 @@
 #include <linux/init.h>
 #include <linux/netdevice.h>
 #include <linux/skbuff.h>
+#include <linux/of.h>
+#include <linux/platform_device.h>
 
 #include <net/ncsi.h>
 #include <net/net_namespace.h>
@@ -730,6 +732,34 @@ static int ncsi_oem_gma_handler_mlx(struct ncsi_cmd_arg *nca)
 	return ret;
 }
 
+static int ncsi_oem_smaf_mlx(struct ncsi_cmd_arg *nca)
+{
+	union {
+		u8 data_u8[NCSI_OEM_MLX_CMD_SMAF_LEN];
+		u32 data_u32[NCSI_OEM_MLX_CMD_SMAF_LEN / sizeof(u32)];
+	} u;
+	int ret = 0;
+
+	memset(&u, 0, sizeof(u));
+	u.data_u32[0] = ntohl(NCSI_OEM_MFR_MLX_ID);
+	u.data_u8[5] = NCSI_OEM_MLX_CMD_SMAF;
+	u.data_u8[6] = NCSI_OEM_MLX_CMD_SMAF_PARAM;
+	memcpy(&u.data_u8[MLX_SMAF_MAC_ADDR_OFFSET],
+	       nca->ndp->ndev.dev->dev_addr,	ETH_ALEN);
+	u.data_u8[MLX_SMAF_MED_SUPPORT_OFFSET] =
+		(MLX_MC_RBT_AVL | MLX_MC_RBT_SUPPORT);
+
+	nca->payload = NCSI_OEM_MLX_CMD_SMAF_LEN;
+	nca->data = u.data_u8;
+
+	ret = ncsi_xmit_cmd(nca);
+	if (ret)
+		netdev_err(nca->ndp->ndev.dev,
+			   "NCSI: Failed to transmit cmd 0x%x during probe\n",
+			   nca->type);
+	return ret;
+}
+
 /* OEM Command handlers initialization */
 static struct ncsi_oem_gma_handler {
 	unsigned int	mfr_id;
@@ -1310,8 +1340,38 @@ static void ncsi_probe_channel(struct ncsi_dev_priv *ndp)
 			break;
 		}
 		nd->state = ncsi_dev_state_probe_cis;
+		if (IS_ENABLED(CONFIG_NCSI_OEM_CMD_GET_MAC) &&
+		    ndp->mlx_multi_host)
+			nd->state = ncsi_dev_state_probe_mlx_gma;
+
 		schedule_work(&ndp->work);
 		break;
+#if IS_ENABLED(CONFIG_NCSI_OEM_CMD_GET_MAC)
+	case ncsi_dev_state_probe_mlx_gma:
+		ndp->pending_req_num = 1;
+
+		nca.type = NCSI_PKT_CMD_OEM;
+		nca.package = ndp->active_package->id;
+		nca.channel = 0;
+		ret = ncsi_oem_gma_handler_mlx(&nca);
+		if (ret)
+			goto error;
+
+		nd->state = ncsi_dev_state_probe_mlx_smaf;
+		break;
+	case ncsi_dev_state_probe_mlx_smaf:
+		ndp->pending_req_num = 1;
+
+		nca.type = NCSI_PKT_CMD_OEM;
+		nca.package = ndp->active_package->id;
+		nca.channel = 0;
+		ret = ncsi_oem_smaf_mlx(&nca);
+		if (ret)
+			goto error;
+
+		nd->state = ncsi_dev_state_probe_cis;
+		break;
+#endif /* CONFIG_NCSI_OEM_CMD_GET_MAC */
 	case ncsi_dev_state_probe_cis:
 		ndp->pending_req_num = NCSI_RESERVED_CHANNEL;
 
@@ -1621,6 +1681,8 @@ struct ncsi_dev *ncsi_register_dev(struct net_device *dev,
 {
 	struct ncsi_dev_priv *ndp;
 	struct ncsi_dev *nd;
+	struct platform_device *pdev;
+	struct device_node *np;
 	unsigned long flags;
 	int i;
 
@@ -1667,6 +1729,13 @@ struct ncsi_dev *ncsi_register_dev(struct net_device *dev,
 	/* Set up generic netlink interface */
 	ncsi_init_netlink(dev);
 
+	pdev = to_platform_device(dev->dev.parent);
+	if (pdev) {
+		np = pdev->dev.of_node;
+		if (np && of_get_property(np, "mlx,multi-host", NULL))
+			ndp->mlx_multi_host = true;
+	}
+
 	return nd;
 }
 EXPORT_SYMBOL_GPL(ncsi_register_dev);
-- 
cgit v1.2.3-59-g8ed1b


From 6b3acfc3cc3d54a99cc5148960edfc38c94a93f2 Mon Sep 17 00:00:00 2001
From: Li RongQing <lirongqing@baidu.com>
Date: Thu, 9 Jan 2020 08:59:56 +0800
Subject: flow_dissector: fix document for skb_flow_get_icmp_tci

using correct input parameter name to fix the below warning:

net/core/flow_dissector.c:242: warning: Function parameter or member 'thoff' not described in 'skb_flow_get_icmp_tci'
net/core/flow_dissector.c:242: warning: Excess function parameter 'toff' description in 'skb_flow_get_icmp_tci'

Signed-off-by: Li RongQing <lirongqing@baidu.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/flow_dissector.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c
index 2dbbb030fbed..f560b4902060 100644
--- a/net/core/flow_dissector.c
+++ b/net/core/flow_dissector.c
@@ -233,7 +233,7 @@ static bool icmp_has_id(u8 type)
  * @skb: sk_buff to extract from
  * @key_icmp: struct flow_dissector_key_icmp to fill
  * @data: raw buffer pointer to the packet
- * @toff: offset to extract at
+ * @thoff: offset to extract at
  * @hlen: packet header length
  */
 void skb_flow_get_icmp_tci(const struct sk_buff *skb,
-- 
cgit v1.2.3-59-g8ed1b


From e9cdced78dc20c1592c1fb98ed064943007a46c5 Mon Sep 17 00:00:00 2001
From: Mat Martineau <mathew.j.martineau@linux.intel.com>
Date: Thu, 9 Jan 2020 07:59:14 -0800
Subject: net: Make sock protocol value checks more specific

SK_PROTOCOL_MAX is only used in two places, for DECNet and AX.25. The
limits have more to do with the those protocol definitions than they do
with the data type of sk_protocol, so remove SK_PROTOCOL_MAX and use
U8_MAX directly.

Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sock.h     | 1 -
 net/ax25/af_ax25.c     | 2 +-
 net/decnet/af_decnet.c | 2 +-
 3 files changed, 2 insertions(+), 3 deletions(-)

diff --git a/include/net/sock.h b/include/net/sock.h
index 8dff68b4c316..091e55428415 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -458,7 +458,6 @@ struct sock {
 				sk_userlocks : 4,
 				sk_protocol  : 8,
 				sk_type      : 16;
-#define SK_PROTOCOL_MAX U8_MAX
 	u16			sk_gso_max_segs;
 	u8			sk_pacing_shift;
 	unsigned long	        sk_lingertime;
diff --git a/net/ax25/af_ax25.c b/net/ax25/af_ax25.c
index 324306d6fde0..ff57ea89c27e 100644
--- a/net/ax25/af_ax25.c
+++ b/net/ax25/af_ax25.c
@@ -808,7 +808,7 @@ static int ax25_create(struct net *net, struct socket *sock, int protocol,
 	struct sock *sk;
 	ax25_cb *ax25;
 
-	if (protocol < 0 || protocol > SK_PROTOCOL_MAX)
+	if (protocol < 0 || protocol > U8_MAX)
 		return -EINVAL;
 
 	if (!net_eq(net, &init_net))
diff --git a/net/decnet/af_decnet.c b/net/decnet/af_decnet.c
index e19a92a62e14..0a46ea3bddd5 100644
--- a/net/decnet/af_decnet.c
+++ b/net/decnet/af_decnet.c
@@ -670,7 +670,7 @@ static int dn_create(struct net *net, struct socket *sock, int protocol,
 {
 	struct sock *sk;
 
-	if (protocol < 0 || protocol > SK_PROTOCOL_MAX)
+	if (protocol < 0 || protocol > U8_MAX)
 		return -EINVAL;
 
 	if (!net_eq(net, &init_net))
-- 
cgit v1.2.3-59-g8ed1b


From bf9765145b856fa2e238a5b8a54453795ba30ad6 Mon Sep 17 00:00:00 2001
From: Mat Martineau <mathew.j.martineau@linux.intel.com>
Date: Thu, 9 Jan 2020 07:59:15 -0800
Subject: sock: Make sk_protocol a 16-bit value

Match the 16-bit width of skbuff->protocol. Fills an 8-bit hole so
sizeof(struct sock) does not change.

Also take care of BPF field access for sk_type/sk_protocol. Both of them
are now outside the bitfield, so we can use load instructions without
further shifting/masking.

v5 -> v6:
 - update eBPF accessors, too (Intel's kbuild test robot)
v2 -> v3:
 - keep 'sk_type' 2 bytes aligned (Eric)
v1 -> v2:
 - preserve sk_pacing_shift as bit field (Eric)

Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: bpf@vger.kernel.org
Co-developed-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Co-developed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sock.h          | 25 ++++---------------
 include/trace/events/sock.h |  2 +-
 net/core/filter.c           | 60 +++++++++++++++++----------------------------
 3 files changed, 28 insertions(+), 59 deletions(-)

diff --git a/include/net/sock.h b/include/net/sock.h
index 091e55428415..8766f9bc3e70 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -436,30 +436,15 @@ struct sock {
 	 * Because of non atomicity rules, all
 	 * changes are protected by socket lock.
 	 */
-	unsigned int		__sk_flags_offset[0];
-#ifdef __BIG_ENDIAN_BITFIELD
-#define SK_FL_PROTO_SHIFT  16
-#define SK_FL_PROTO_MASK   0x00ff0000
-
-#define SK_FL_TYPE_SHIFT   0
-#define SK_FL_TYPE_MASK    0x0000ffff
-#else
-#define SK_FL_PROTO_SHIFT  8
-#define SK_FL_PROTO_MASK   0x0000ff00
-
-#define SK_FL_TYPE_SHIFT   16
-#define SK_FL_TYPE_MASK    0xffff0000
-#endif
-
-	unsigned int		sk_padding : 1,
+	u8			sk_padding : 1,
 				sk_kern_sock : 1,
 				sk_no_check_tx : 1,
 				sk_no_check_rx : 1,
-				sk_userlocks : 4,
-				sk_protocol  : 8,
-				sk_type      : 16;
-	u16			sk_gso_max_segs;
+				sk_userlocks : 4;
 	u8			sk_pacing_shift;
+	u16			sk_type;
+	u16			sk_protocol;
+	u16			sk_gso_max_segs;
 	unsigned long	        sk_lingertime;
 	struct proto		*sk_prot_creator;
 	rwlock_t		sk_callback_lock;
diff --git a/include/trace/events/sock.h b/include/trace/events/sock.h
index 51fe9f6719eb..3ff12b90048d 100644
--- a/include/trace/events/sock.h
+++ b/include/trace/events/sock.h
@@ -147,7 +147,7 @@ TRACE_EVENT(inet_sock_set_state,
 		__field(__u16, sport)
 		__field(__u16, dport)
 		__field(__u16, family)
-		__field(__u8, protocol)
+		__field(__u16, protocol)
 		__array(__u8, saddr, 4)
 		__array(__u8, daddr, 4)
 		__array(__u8, saddr_v6, 16)
diff --git a/net/core/filter.c b/net/core/filter.c
index 42fd17c48c5f..ef01c5599501 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -7607,21 +7607,21 @@ u32 bpf_sock_convert_ctx_access(enum bpf_access_type type,
 		break;
 
 	case offsetof(struct bpf_sock, type):
-		BUILD_BUG_ON(HWEIGHT32(SK_FL_TYPE_MASK) != BITS_PER_BYTE * 2);
-		*insn++ = BPF_LDX_MEM(BPF_W, si->dst_reg, si->src_reg,
-				      offsetof(struct sock, __sk_flags_offset));
-		*insn++ = BPF_ALU32_IMM(BPF_AND, si->dst_reg, SK_FL_TYPE_MASK);
-		*insn++ = BPF_ALU32_IMM(BPF_RSH, si->dst_reg, SK_FL_TYPE_SHIFT);
-		*target_size = 2;
+		*insn++ = BPF_LDX_MEM(
+			BPF_FIELD_SIZEOF(struct sock, sk_type),
+			si->dst_reg, si->src_reg,
+			bpf_target_off(struct sock, sk_type,
+				       sizeof_field(struct sock, sk_type),
+				       target_size));
 		break;
 
 	case offsetof(struct bpf_sock, protocol):
-		BUILD_BUG_ON(HWEIGHT32(SK_FL_PROTO_MASK) != BITS_PER_BYTE);
-		*insn++ = BPF_LDX_MEM(BPF_W, si->dst_reg, si->src_reg,
-				      offsetof(struct sock, __sk_flags_offset));
-		*insn++ = BPF_ALU32_IMM(BPF_AND, si->dst_reg, SK_FL_PROTO_MASK);
-		*insn++ = BPF_ALU32_IMM(BPF_RSH, si->dst_reg, SK_FL_PROTO_SHIFT);
-		*target_size = 1;
+		*insn++ = BPF_LDX_MEM(
+			BPF_FIELD_SIZEOF(struct sock, sk_protocol),
+			si->dst_reg, si->src_reg,
+			bpf_target_off(struct sock, sk_protocol,
+				       sizeof_field(struct sock, sk_protocol),
+				       target_size));
 		break;
 
 	case offsetof(struct bpf_sock, src_ip4):
@@ -7903,20 +7903,13 @@ static u32 sock_addr_convert_ctx_access(enum bpf_access_type type,
 		break;
 
 	case offsetof(struct bpf_sock_addr, type):
-		SOCK_ADDR_LOAD_NESTED_FIELD_SIZE_OFF(
-			struct bpf_sock_addr_kern, struct sock, sk,
-			__sk_flags_offset, BPF_W, 0);
-		*insn++ = BPF_ALU32_IMM(BPF_AND, si->dst_reg, SK_FL_TYPE_MASK);
-		*insn++ = BPF_ALU32_IMM(BPF_RSH, si->dst_reg, SK_FL_TYPE_SHIFT);
+		SOCK_ADDR_LOAD_NESTED_FIELD(struct bpf_sock_addr_kern,
+					    struct sock, sk, sk_type);
 		break;
 
 	case offsetof(struct bpf_sock_addr, protocol):
-		SOCK_ADDR_LOAD_NESTED_FIELD_SIZE_OFF(
-			struct bpf_sock_addr_kern, struct sock, sk,
-			__sk_flags_offset, BPF_W, 0);
-		*insn++ = BPF_ALU32_IMM(BPF_AND, si->dst_reg, SK_FL_PROTO_MASK);
-		*insn++ = BPF_ALU32_IMM(BPF_RSH, si->dst_reg,
-					SK_FL_PROTO_SHIFT);
+		SOCK_ADDR_LOAD_NESTED_FIELD(struct bpf_sock_addr_kern,
+					    struct sock, sk, sk_protocol);
 		break;
 
 	case offsetof(struct bpf_sock_addr, msg_src_ip4):
@@ -8835,11 +8828,11 @@ sk_reuseport_is_valid_access(int off, int size,
 				    skb,				\
 				    SKB_FIELD)
 
-#define SK_REUSEPORT_LOAD_SK_FIELD_SIZE_OFF(SK_FIELD, BPF_SIZE, EXTRA_OFF) \
-	SOCK_ADDR_LOAD_NESTED_FIELD_SIZE_OFF(struct sk_reuseport_kern,	\
-					     struct sock,		\
-					     sk,			\
-					     SK_FIELD, BPF_SIZE, EXTRA_OFF)
+#define SK_REUSEPORT_LOAD_SK_FIELD(SK_FIELD)				\
+	SOCK_ADDR_LOAD_NESTED_FIELD(struct sk_reuseport_kern,		\
+				    struct sock,			\
+				    sk,					\
+				    SK_FIELD)
 
 static u32 sk_reuseport_convert_ctx_access(enum bpf_access_type type,
 					   const struct bpf_insn *si,
@@ -8863,16 +8856,7 @@ static u32 sk_reuseport_convert_ctx_access(enum bpf_access_type type,
 		break;
 
 	case offsetof(struct sk_reuseport_md, ip_protocol):
-		BUILD_BUG_ON(HWEIGHT32(SK_FL_PROTO_MASK) != BITS_PER_BYTE);
-		SK_REUSEPORT_LOAD_SK_FIELD_SIZE_OFF(__sk_flags_offset,
-						    BPF_W, 0);
-		*insn++ = BPF_ALU32_IMM(BPF_AND, si->dst_reg, SK_FL_PROTO_MASK);
-		*insn++ = BPF_ALU32_IMM(BPF_RSH, si->dst_reg,
-					SK_FL_PROTO_SHIFT);
-		/* SK_FL_PROTO_MASK and SK_FL_PROTO_SHIFT are endian
-		 * aware.  No further narrowing or masking is needed.
-		 */
-		*target_size = 1;
+		SK_REUSEPORT_LOAD_SK_FIELD(sk_protocol);
 		break;
 
 	case offsetof(struct sk_reuseport_md, data_end):
-- 
cgit v1.2.3-59-g8ed1b


From faf391c3826cd29feae02078ca2022d2f912f7cc Mon Sep 17 00:00:00 2001
From: Mat Martineau <mathew.j.martineau@linux.intel.com>
Date: Thu, 9 Jan 2020 07:59:16 -0800
Subject: tcp: Define IPPROTO_MPTCP

To open a MPTCP socket with socket(AF_INET, SOCK_STREAM, IPPROTO_MPTCP),
IPPROTO_MPTCP needs a value that differs from IPPROTO_TCP. The existing
IPPROTO numbers mostly map directly to IANA-specified protocol numbers.
MPTCP does not have a protocol number allocated because MPTCP packets
use the TCP protocol number. Use private number not used OTA.

Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/trace/events/sock.h   | 3 ++-
 include/uapi/linux/in.h       | 2 ++
 tools/include/uapi/linux/in.h | 2 ++
 3 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/include/trace/events/sock.h b/include/trace/events/sock.h
index 3ff12b90048d..a966d4b5ab37 100644
--- a/include/trace/events/sock.h
+++ b/include/trace/events/sock.h
@@ -19,7 +19,8 @@
 #define inet_protocol_names		\
 		EM(IPPROTO_TCP)			\
 		EM(IPPROTO_DCCP)		\
-		EMe(IPPROTO_SCTP)
+		EM(IPPROTO_SCTP)		\
+		EMe(IPPROTO_MPTCP)
 
 #define tcp_state_names			\
 		EM(TCP_ESTABLISHED)		\
diff --git a/include/uapi/linux/in.h b/include/uapi/linux/in.h
index e7ad9d350a28..1521073b6348 100644
--- a/include/uapi/linux/in.h
+++ b/include/uapi/linux/in.h
@@ -76,6 +76,8 @@ enum {
 #define IPPROTO_MPLS		IPPROTO_MPLS
   IPPROTO_RAW = 255,		/* Raw IP packets			*/
 #define IPPROTO_RAW		IPPROTO_RAW
+  IPPROTO_MPTCP = 262,		/* Multipath TCP connection		*/
+#define IPPROTO_MPTCP		IPPROTO_MPTCP
   IPPROTO_MAX
 };
 #endif
diff --git a/tools/include/uapi/linux/in.h b/tools/include/uapi/linux/in.h
index e7ad9d350a28..1521073b6348 100644
--- a/tools/include/uapi/linux/in.h
+++ b/tools/include/uapi/linux/in.h
@@ -76,6 +76,8 @@ enum {
 #define IPPROTO_MPLS		IPPROTO_MPLS
   IPPROTO_RAW = 255,		/* Raw IP packets			*/
 #define IPPROTO_RAW		IPPROTO_RAW
+  IPPROTO_MPTCP = 262,		/* Multipath TCP connection		*/
+#define IPPROTO_MPTCP		IPPROTO_MPTCP
   IPPROTO_MAX
 };
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From c74a39c861aeaf6b789b7abdbb3256f54c9fb365 Mon Sep 17 00:00:00 2001
From: Mat Martineau <mathew.j.martineau@linux.intel.com>
Date: Thu, 9 Jan 2020 07:59:17 -0800
Subject: tcp: Add MPTCP option number

TCP option 30 is allocated for MPTCP by the IANA.

Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 7df37e2fddca..85f1d7ff6e8b 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -182,6 +182,7 @@ void tcp_time_wait(struct sock *sk, int state, int timeo);
 #define TCPOPT_SACK             5       /* SACK Block */
 #define TCPOPT_TIMESTAMP	8	/* Better RTT estimations/PAWS */
 #define TCPOPT_MD5SIG		19	/* MD5 Signature (RFC2385) */
+#define TCPOPT_MPTCP		30	/* Multipath TCP (RFC6824) */
 #define TCPOPT_FASTOPEN		34	/* Fast open (RFC7413) */
 #define TCPOPT_EXP		254	/* Experimental */
 /* Magic number to be after the option value for sharing TCP
-- 
cgit v1.2.3-59-g8ed1b


From 1323059301c8f36d933876233516245d882346a6 Mon Sep 17 00:00:00 2001
From: Mat Martineau <mathew.j.martineau@linux.intel.com>
Date: Thu, 9 Jan 2020 07:59:18 -0800
Subject: tcp, ulp: Add clone operation to tcp_ulp_ops

If ULP is used on a listening socket, icsk_ulp_ops and icsk_ulp_data are
copied when the listener is cloned. Sometimes the clone is immediately
deleted, which will invoke the release op on the clone and likely
corrupt the listening socket's icsk_ulp_data.

The clone operation is invoked immediately after the clone is copied and
gives the ULP type an opportunity to set up the clone socket and its
icsk_ulp_data.

The MPTCP ULP clone will silently fallback to plain TCP on allocation
failure, so 'clone()' does not need to return an error code.

v6 -> v7:
 - move and rename ulp clone helper to make it inline-friendly
v5 -> v6:
 - clarified MPTCP clone usage in commit message

Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h               |  3 +++
 net/ipv4/inet_connection_sock.c | 14 ++++++++++++++
 2 files changed, 17 insertions(+)

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 85f1d7ff6e8b..ac52633e7061 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -2154,6 +2154,9 @@ struct tcp_ulp_ops {
 	/* diagnostic */
 	int (*get_info)(const struct sock *sk, struct sk_buff *skb);
 	size_t (*get_info_size)(const struct sock *sk);
+	/* clone ulp */
+	void (*clone)(const struct request_sock *req, struct sock *newsk,
+		      const gfp_t priority);
 
 	char		name[TCP_ULP_NAME_MAX];
 	struct module	*owner;
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 18c0d5bffe12..6a691fd04398 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -770,6 +770,18 @@ void inet_csk_reqsk_queue_hash_add(struct sock *sk, struct request_sock *req,
 }
 EXPORT_SYMBOL_GPL(inet_csk_reqsk_queue_hash_add);
 
+static void inet_clone_ulp(const struct request_sock *req, struct sock *newsk,
+			   const gfp_t priority)
+{
+	struct inet_connection_sock *icsk = inet_csk(newsk);
+
+	if (!icsk->icsk_ulp_ops)
+		return;
+
+	if (icsk->icsk_ulp_ops->clone)
+		icsk->icsk_ulp_ops->clone(req, newsk, priority);
+}
+
 /**
  *	inet_csk_clone_lock - clone an inet socket, and lock its clone
  *	@sk: the socket to clone
@@ -810,6 +822,8 @@ struct sock *inet_csk_clone_lock(const struct sock *sk,
 		/* Deinitialize accept_queue to trap illegal accesses. */
 		memset(&newicsk->icsk_accept_queue, 0, sizeof(newicsk->icsk_accept_queue));
 
+		inet_clone_ulp(req, newsk, priority);
+
 		security_inet_csk_clone(newsk, req);
 	}
 	return newsk;
-- 
cgit v1.2.3-59-g8ed1b


From 3ee17bc78e0f3fdeff9890993e8f3a9f5145163b Mon Sep 17 00:00:00 2001
From: Mat Martineau <mathew.j.martineau@linux.intel.com>
Date: Thu, 9 Jan 2020 07:59:19 -0800
Subject: mptcp: Add MPTCP to skb extensions

Add enum value for MPTCP and update config dependencies

v5 -> v6:
 - fixed '__unused' field size

Co-developed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Co-developed-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 MAINTAINERS            | 10 ++++++++++
 include/linux/skbuff.h |  3 +++
 include/net/mptcp.h    | 28 ++++++++++++++++++++++++++++
 net/core/skbuff.c      |  7 +++++++
 4 files changed, 48 insertions(+)
 create mode 100644 include/net/mptcp.h

diff --git a/MAINTAINERS b/MAINTAINERS
index 218089f38e11..9dcb9cab5705 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -11573,6 +11573,16 @@ F:	net/ipv6/calipso.c
 F:	net/netfilter/xt_CONNSECMARK.c
 F:	net/netfilter/xt_SECMARK.c
 
+NETWORKING [MPTCP]
+M:	Mat Martineau <mathew.j.martineau@linux.intel.com>
+M:	Matthieu Baerts <matthieu.baerts@tessares.net>
+L:	netdev@vger.kernel.org
+L:	mptcp@lists.01.org
+W:	https://github.com/multipath-tcp/mptcp_net-next/wiki
+B:	https://github.com/multipath-tcp/mptcp_net-next/issues
+S:	Maintained
+F:	include/net/mptcp.h
+
 NETWORKING [TCP]
 M:	Eric Dumazet <edumazet@google.com>
 L:	netdev@vger.kernel.org
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 64e5b1be9ff5..f5c27600b410 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -4096,6 +4096,9 @@ enum skb_ext_id {
 #endif
 #if IS_ENABLED(CONFIG_NET_TC_SKB_EXT)
 	TC_SKB_EXT,
+#endif
+#if IS_ENABLED(CONFIG_MPTCP)
+	SKB_EXT_MPTCP,
 #endif
 	SKB_EXT_NUM, /* must be last */
 };
diff --git a/include/net/mptcp.h b/include/net/mptcp.h
new file mode 100644
index 000000000000..326043c29c0a
--- /dev/null
+++ b/include/net/mptcp.h
@@ -0,0 +1,28 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Multipath TCP
+ *
+ * Copyright (c) 2017 - 2019, Intel Corporation.
+ */
+
+#ifndef __NET_MPTCP_H
+#define __NET_MPTCP_H
+
+#include <linux/types.h>
+
+/* MPTCP sk_buff extension data */
+struct mptcp_ext {
+	u64		data_ack;
+	u64		data_seq;
+	u32		subflow_seq;
+	u16		data_len;
+	u8		use_map:1,
+			dsn64:1,
+			data_fin:1,
+			use_ack:1,
+			ack64:1,
+			__unused:3;
+	/* one byte hole */
+};
+
+#endif /* __NET_MPTCP_H */
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 44b0894d8ae1..a4106da23c34 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -68,6 +68,7 @@
 #include <net/ip6_checksum.h>
 #include <net/xfrm.h>
 #include <net/mpls.h>
+#include <net/mptcp.h>
 
 #include <linux/uaccess.h>
 #include <trace/events/skb.h>
@@ -4109,6 +4110,9 @@ static const u8 skb_ext_type_len[] = {
 #if IS_ENABLED(CONFIG_NET_TC_SKB_EXT)
 	[TC_SKB_EXT] = SKB_EXT_CHUNKSIZEOF(struct tc_skb_ext),
 #endif
+#if IS_ENABLED(CONFIG_MPTCP)
+	[SKB_EXT_MPTCP] = SKB_EXT_CHUNKSIZEOF(struct mptcp_ext),
+#endif
 };
 
 static __always_inline unsigned int skb_ext_total_length(void)
@@ -4122,6 +4126,9 @@ static __always_inline unsigned int skb_ext_total_length(void)
 #endif
 #if IS_ENABLED(CONFIG_NET_TC_SKB_EXT)
 		skb_ext_type_len[TC_SKB_EXT] +
+#endif
+#if IS_ENABLED(CONFIG_MPTCP)
+		skb_ext_type_len[SKB_EXT_MPTCP] +
 #endif
 		0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 85712484110df308215077be6ee21c4e57d7dec2 Mon Sep 17 00:00:00 2001
From: Mat Martineau <mathew.j.martineau@linux.intel.com>
Date: Thu, 9 Jan 2020 07:59:20 -0800
Subject: tcp: coalesce/collapse must respect MPTCP extensions

Coalesce and collapse of packets carrying MPTCP extensions is allowed
when the newer packet has no extension or the extensions carried by both
packets are equal.

This allows merging of TSO packet trains and even cross-TSO packets, and
does not require any additional action when moving data into existing
SKBs.

v3 -> v4:
 - allow collapsing, under mptcp_skb_can_collapse() constraint

v5 -> v6:
 - clarify MPTCP skb extensions must always be cleared at allocation
   time

Co-developed-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/mptcp.h   | 57 +++++++++++++++++++++++++++++++++++++++++++++++++++
 include/net/tcp.h     |  8 ++++++++
 net/ipv4/tcp_input.c  | 11 +++++++---
 net/ipv4/tcp_output.c |  2 +-
 4 files changed, 74 insertions(+), 4 deletions(-)

diff --git a/include/net/mptcp.h b/include/net/mptcp.h
index 326043c29c0a..0573ae75c3db 100644
--- a/include/net/mptcp.h
+++ b/include/net/mptcp.h
@@ -8,6 +8,7 @@
 #ifndef __NET_MPTCP_H
 #define __NET_MPTCP_H
 
+#include <linux/skbuff.h>
 #include <linux/types.h>
 
 /* MPTCP sk_buff extension data */
@@ -25,4 +26,60 @@ struct mptcp_ext {
 	/* one byte hole */
 };
 
+#ifdef CONFIG_MPTCP
+
+/* move the skb extension owership, with the assumption that 'to' is
+ * newly allocated
+ */
+static inline void mptcp_skb_ext_move(struct sk_buff *to,
+				      struct sk_buff *from)
+{
+	if (!skb_ext_exist(from, SKB_EXT_MPTCP))
+		return;
+
+	if (WARN_ON_ONCE(to->active_extensions))
+		skb_ext_put(to);
+
+	to->active_extensions = from->active_extensions;
+	to->extensions = from->extensions;
+	from->active_extensions = 0;
+}
+
+static inline bool mptcp_ext_matches(const struct mptcp_ext *to_ext,
+				     const struct mptcp_ext *from_ext)
+{
+	/* MPTCP always clears the ext when adding it to the skb, so
+	 * holes do not bother us here
+	 */
+	return !from_ext ||
+	       (to_ext && from_ext &&
+	        !memcmp(from_ext, to_ext, sizeof(struct mptcp_ext)));
+}
+
+/* check if skbs can be collapsed.
+ * MPTCP collapse is allowed if neither @to or @from carry an mptcp data
+ * mapping, or if the extension of @to is the same as @from.
+ * Collapsing is not possible if @to lacks an extension, but @from carries one.
+ */
+static inline bool mptcp_skb_can_collapse(const struct sk_buff *to,
+					  const struct sk_buff *from)
+{
+	return mptcp_ext_matches(skb_ext_find(to, SKB_EXT_MPTCP),
+				 skb_ext_find(from, SKB_EXT_MPTCP));
+}
+
+#else
+
+static inline void mptcp_skb_ext_move(struct sk_buff *to,
+				      const struct sk_buff *from)
+{
+}
+
+static inline bool mptcp_skb_can_collapse(const struct sk_buff *to,
+					  const struct sk_buff *from)
+{
+	return true;
+}
+
+#endif /* CONFIG_MPTCP */
 #endif /* __NET_MPTCP_H */
diff --git a/include/net/tcp.h b/include/net/tcp.h
index ac52633e7061..13bc83fab454 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -39,6 +39,7 @@
 #include <net/tcp_states.h>
 #include <net/inet_ecn.h>
 #include <net/dst.h>
+#include <net/mptcp.h>
 
 #include <linux/seq_file.h>
 #include <linux/memcontrol.h>
@@ -978,6 +979,13 @@ static inline bool tcp_skb_can_collapse_to(const struct sk_buff *skb)
 	return likely(!TCP_SKB_CB(skb)->eor);
 }
 
+static inline bool tcp_skb_can_collapse(const struct sk_buff *to,
+					const struct sk_buff *from)
+{
+	return likely(tcp_skb_can_collapse_to(to) &&
+		      mptcp_skb_can_collapse(to, from));
+}
+
 /* Events passed to congestion control interface */
 enum tcp_ca_event {
 	CA_EVENT_TX_START,	/* first transmit when no packets in flight */
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 062b696a5fa1..2914fdf1d543 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -1422,7 +1422,7 @@ static struct sk_buff *tcp_shift_skb_data(struct sock *sk, struct sk_buff *skb,
 	if ((TCP_SKB_CB(prev)->sacked & TCPCB_TAGBITS) != TCPCB_SACKED_ACKED)
 		goto fallback;
 
-	if (!tcp_skb_can_collapse_to(prev))
+	if (!tcp_skb_can_collapse(prev, skb))
 		goto fallback;
 
 	in_sack = !after(start_seq, TCP_SKB_CB(skb)->seq) &&
@@ -4423,6 +4423,9 @@ static bool tcp_try_coalesce(struct sock *sk,
 	if (TCP_SKB_CB(from)->seq != TCP_SKB_CB(to)->end_seq)
 		return false;
 
+	if (!mptcp_skb_can_collapse(to, from))
+		return false;
+
 #ifdef CONFIG_TLS_DEVICE
 	if (from->decrypted != to->decrypted)
 		return false;
@@ -4932,7 +4935,7 @@ restart:
 		/* The first skb to collapse is:
 		 * - not SYN/FIN and
 		 * - bloated or contains data before "start" or
-		 *   overlaps to the next one.
+		 *   overlaps to the next one and mptcp allow collapsing.
 		 */
 		if (!(TCP_SKB_CB(skb)->tcp_flags & (TCPHDR_SYN | TCPHDR_FIN)) &&
 		    (tcp_win_from_space(sk, skb->truesize) > skb->len ||
@@ -4941,7 +4944,7 @@ restart:
 			break;
 		}
 
-		if (n && n != tail &&
+		if (n && n != tail && mptcp_skb_can_collapse(skb, n) &&
 		    TCP_SKB_CB(skb)->end_seq != TCP_SKB_CB(n)->seq) {
 			end_of_skbs = false;
 			break;
@@ -4974,6 +4977,7 @@ restart:
 		else
 			__skb_queue_tail(&tmp, nskb); /* defer rbtree insertion */
 		skb_set_owner_r(nskb, sk);
+		mptcp_skb_ext_move(nskb, skb);
 
 		/* Copy data, releasing collapsed skbs. */
 		while (copy > 0) {
@@ -4993,6 +4997,7 @@ restart:
 				skb = tcp_collapse_one(sk, skb, list, root);
 				if (!skb ||
 				    skb == tail ||
+				    !mptcp_skb_can_collapse(nskb, skb) ||
 				    (TCP_SKB_CB(skb)->tcp_flags & (TCPHDR_SYN | TCPHDR_FIN)))
 					goto end;
 #ifdef CONFIG_TLS_DEVICE
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 58c92a7d671c..3ce7fe1c4076 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -2865,7 +2865,7 @@ static void tcp_retrans_try_collapse(struct sock *sk, struct sk_buff *to,
 		if (!tcp_can_collapse(sk, skb))
 			break;
 
-		if (!tcp_skb_can_collapse_to(to))
+		if (!tcp_skb_can_collapse(to, skb))
 			break;
 
 		space -= skb->len;
-- 
cgit v1.2.3-59-g8ed1b


From 35b2c32116091ef87a15c604cac363da8322a288 Mon Sep 17 00:00:00 2001
From: Mat Martineau <mathew.j.martineau@linux.intel.com>
Date: Thu, 9 Jan 2020 07:59:21 -0800
Subject: tcp: Export TCP functions and ops struct

MPTCP will make use of tcp_send_mss() and tcp_push() when sending
data to specific TCP subflows.

tcp_request_sock_ipvX_ops and ipvX_specific will be referenced
during TCP subflow creation.

Co-developed-by: Peter Krystad <peter.krystad@linux.intel.com>
Signed-off-by: Peter Krystad <peter.krystad@linux.intel.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h   | 8 ++++++++
 net/ipv4/tcp.c      | 6 +++---
 net/ipv4/tcp_ipv4.c | 2 +-
 net/ipv6/tcp_ipv6.c | 6 +++---
 4 files changed, 15 insertions(+), 7 deletions(-)

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 13bc83fab454..5e4133d09b9d 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -330,6 +330,9 @@ int tcp_sendpage_locked(struct sock *sk, struct page *page, int offset,
 			size_t size, int flags);
 ssize_t do_tcp_sendpages(struct sock *sk, struct page *page, int offset,
 		 size_t size, int flags);
+int tcp_send_mss(struct sock *sk, int *size_goal, int flags);
+void tcp_push(struct sock *sk, int flags, int mss_now, int nonagle,
+	      int size_goal);
 void tcp_release_cb(struct sock *sk);
 void tcp_wfree(struct sk_buff *skb);
 void tcp_write_timer_handler(struct sock *sk);
@@ -2011,6 +2014,11 @@ struct tcp_request_sock_ops {
 			   enum tcp_synack_type synack_type);
 };
 
+extern const struct tcp_request_sock_ops tcp_request_sock_ipv4_ops;
+#if IS_ENABLED(CONFIG_IPV6)
+extern const struct tcp_request_sock_ops tcp_request_sock_ipv6_ops;
+#endif
+
 #ifdef CONFIG_SYN_COOKIES
 static inline __u32 cookie_init_sequence(const struct tcp_request_sock_ops *ops,
 					 const struct sock *sk, struct sk_buff *skb,
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index f09fbc85b108..6711a97de3ce 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -690,8 +690,8 @@ static bool tcp_should_autocork(struct sock *sk, struct sk_buff *skb,
 	       refcount_read(&sk->sk_wmem_alloc) > skb->truesize;
 }
 
-static void tcp_push(struct sock *sk, int flags, int mss_now,
-		     int nonagle, int size_goal)
+void tcp_push(struct sock *sk, int flags, int mss_now,
+	      int nonagle, int size_goal)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct sk_buff *skb;
@@ -925,7 +925,7 @@ static unsigned int tcp_xmit_size_goal(struct sock *sk, u32 mss_now,
 	return max(size_goal, mss_now);
 }
 
-static int tcp_send_mss(struct sock *sk, int *size_goal, int flags)
+int tcp_send_mss(struct sock *sk, int *size_goal, int flags)
 {
 	int mss_now;
 
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 4adac9c75343..fedb537839ec 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1426,7 +1426,7 @@ struct request_sock_ops tcp_request_sock_ops __read_mostly = {
 	.syn_ack_timeout =	tcp_syn_ack_timeout,
 };
 
-static const struct tcp_request_sock_ops tcp_request_sock_ipv4_ops = {
+const struct tcp_request_sock_ops tcp_request_sock_ipv4_ops = {
 	.mss_clamp	=	TCP_MSS_DEFAULT,
 #ifdef CONFIG_TCP_MD5SIG
 	.req_md5_lookup	=	tcp_v4_md5_lookup,
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 95e4e1e95db2..5b5260103b65 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -75,7 +75,7 @@ static void	tcp_v6_reqsk_send_ack(const struct sock *sk, struct sk_buff *skb,
 static int	tcp_v6_do_rcv(struct sock *sk, struct sk_buff *skb);
 
 static const struct inet_connection_sock_af_ops ipv6_mapped;
-static const struct inet_connection_sock_af_ops ipv6_specific;
+const struct inet_connection_sock_af_ops ipv6_specific;
 #ifdef CONFIG_TCP_MD5SIG
 static const struct tcp_sock_af_ops tcp_sock_ipv6_specific;
 static const struct tcp_sock_af_ops tcp_sock_ipv6_mapped_specific;
@@ -819,7 +819,7 @@ struct request_sock_ops tcp6_request_sock_ops __read_mostly = {
 	.syn_ack_timeout =	tcp_syn_ack_timeout,
 };
 
-static const struct tcp_request_sock_ops tcp_request_sock_ipv6_ops = {
+const struct tcp_request_sock_ops tcp_request_sock_ipv6_ops = {
 	.mss_clamp	=	IPV6_MIN_MTU - sizeof(struct tcphdr) -
 				sizeof(struct ipv6hdr),
 #ifdef CONFIG_TCP_MD5SIG
@@ -1794,7 +1794,7 @@ static struct timewait_sock_ops tcp6_timewait_sock_ops = {
 	.twsk_destructor = tcp_twsk_destructor,
 };
 
-static const struct inet_connection_sock_af_ops ipv6_specific = {
+const struct inet_connection_sock_af_ops ipv6_specific = {
 	.queue_xmit	   = inet6_csk_xmit,
 	.send_check	   = tcp_v6_send_check,
 	.rebuild_header	   = inet6_sk_rebuild_header,
-- 
cgit v1.2.3-59-g8ed1b


From 9cfcca2389d7e07647ee69950f46ab5e6dfe03ac Mon Sep 17 00:00:00 2001
From: Mat Martineau <mathew.j.martineau@linux.intel.com>
Date: Thu, 9 Jan 2020 07:59:22 -0800
Subject: tcp: Check for filled TCP option space before SACK

Update the SACK check to work with zero option space available, a case
that's possible with MPTCP but not MD5+TS. Maintained only one
conditional branch for insufficient SACK space.

v1 -> v2:
- Moves the check inside the SACK branch by taking recent SACK fix:

    9424e2e7ad93 (tcp: md5: fix potential overestimation of TCP option space)

  in to account, but modifies it to work in MPTCP scenarios beyond the
  MD5+TS corner case.

Co-developed-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/tcp_output.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 3ce7fe1c4076..05109d0c675b 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -754,13 +754,17 @@ static unsigned int tcp_established_options(struct sock *sk, struct sk_buff *skb
 	eff_sacks = tp->rx_opt.num_sacks + tp->rx_opt.dsack;
 	if (unlikely(eff_sacks)) {
 		const unsigned int remaining = MAX_TCP_OPTION_SPACE - size;
+		if (unlikely(remaining < TCPOLEN_SACK_BASE_ALIGNED +
+					 TCPOLEN_SACK_PERBLOCK))
+			return size;
+
 		opts->num_sack_blocks =
 			min_t(unsigned int, eff_sacks,
 			      (remaining - TCPOLEN_SACK_BASE_ALIGNED) /
 			      TCPOLEN_SACK_PERBLOCK);
-		if (likely(opts->num_sack_blocks))
-			size += TCPOLEN_SACK_BASE_ALIGNED +
-				opts->num_sack_blocks * TCPOLEN_SACK_PERBLOCK;
+
+		size += TCPOLEN_SACK_BASE_ALIGNED +
+			opts->num_sack_blocks * TCPOLEN_SACK_PERBLOCK;
 	}
 
 	return size;
-- 
cgit v1.2.3-59-g8ed1b


From e66b2f31a068dd67172008459678821a79e4ea24 Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Thu, 9 Jan 2020 07:59:23 -0800
Subject: tcp: clean ext on tx recycle

Otherwise we will find stray/unexpected/old extensions value on next
iteration.

On tcp_write_xmit() we can end-up splitting an already queued skb in two
parts, via tso_fragment(). The newly created skb can be allocated via
the tx cache and an upper layer will not be aware of it, so that upper
layer cannot set the ext properly.

Resetting the ext on recycle ensures that stale data is not propagated
in to packet headers or elsewhere.

An alternative would be add an additional hook in tso_fragment() or in
sk_stream_alloc_skb() to init the ext for upper layers that need it.

Co-developed-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sock.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/include/net/sock.h b/include/net/sock.h
index 8766f9bc3e70..432ff73d20f3 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -1464,6 +1464,7 @@ static inline void sk_wmem_free_skb(struct sock *sk, struct sk_buff *skb)
 	sk_mem_uncharge(sk, skb->truesize);
 	if (static_branch_unlikely(&tcp_tx_skb_cache_key) &&
 	    !sk->sk_tx_skb_cache && !skb_cloned(skb)) {
+		skb_ext_reset(skb);
 		skb_zcopy_clear(skb, true);
 		sk->sk_tx_skb_cache = skb;
 		return;
-- 
cgit v1.2.3-59-g8ed1b


From 8b69a803814bb8b14155ea60df83f6d57527e69e Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Thu, 9 Jan 2020 07:59:24 -0800
Subject: skb: add helpers to allocate ext independently from sk_buff

Currently we can allocate the extension only after the skb,
this change allows the user to do the opposite, will simplify
allocation failure handling from MPTCP.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h |  3 +++
 net/core/skbuff.c      | 35 +++++++++++++++++++++++++++++++++--
 2 files changed, 36 insertions(+), 2 deletions(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index f5c27600b410..016b3c4ab99a 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -4120,6 +4120,9 @@ struct skb_ext {
 	char data[0] __aligned(8);
 };
 
+struct skb_ext *__skb_ext_alloc(void);
+void *__skb_ext_set(struct sk_buff *skb, enum skb_ext_id id,
+		    struct skb_ext *ext);
 void *skb_ext_add(struct sk_buff *skb, enum skb_ext_id id);
 void __skb_ext_del(struct sk_buff *skb, enum skb_ext_id id);
 void __skb_ext_put(struct skb_ext *ext);
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index a4106da23c34..48a7029529c9 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -5987,7 +5987,14 @@ static void *skb_ext_get_ptr(struct skb_ext *ext, enum skb_ext_id id)
 	return (void *)ext + (ext->offset[id] * SKB_EXT_ALIGN_VALUE);
 }
 
-static struct skb_ext *skb_ext_alloc(void)
+/**
+ * __skb_ext_alloc - allocate a new skb extensions storage
+ *
+ * Returns the newly allocated pointer. The pointer can later attached to a
+ * skb via __skb_ext_set().
+ * Note: caller must handle the skb_ext as an opaque data.
+ */
+struct skb_ext *__skb_ext_alloc(void)
 {
 	struct skb_ext *new = kmem_cache_alloc(skbuff_ext_cache, GFP_ATOMIC);
 
@@ -6027,6 +6034,30 @@ static struct skb_ext *skb_ext_maybe_cow(struct skb_ext *old,
 	return new;
 }
 
+/**
+ * __skb_ext_set - attach the specified extension storage to this skb
+ * @skb: buffer
+ * @id: extension id
+ * @ext: extension storage previously allocated via __skb_ext_alloc()
+ *
+ * Existing extensions, if any, are cleared.
+ *
+ * Returns the pointer to the extension.
+ */
+void *__skb_ext_set(struct sk_buff *skb, enum skb_ext_id id,
+		    struct skb_ext *ext)
+{
+	unsigned int newlen, newoff = SKB_EXT_CHUNKSIZEOF(*ext);
+
+	skb_ext_put(skb);
+	newlen = newoff + skb_ext_type_len[id];
+	ext->chunks = newlen;
+	ext->offset[id] = newoff;
+	skb->extensions = ext;
+	skb->active_extensions = 1 << id;
+	return skb_ext_get_ptr(ext, id);
+}
+
 /**
  * skb_ext_add - allocate space for given extension, COW if needed
  * @skb: buffer
@@ -6060,7 +6091,7 @@ void *skb_ext_add(struct sk_buff *skb, enum skb_ext_id id)
 	} else {
 		newoff = SKB_EXT_CHUNKSIZEOF(*new);
 
-		new = skb_ext_alloc();
+		new = __skb_ext_alloc();
 		if (!new)
 			return NULL;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 492ab0205f500ebd20b4f0c9e95a1ba007b0c542 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Thu, 9 Jan 2020 19:42:46 -0800
Subject: libbpf: Make bpf_map order and indices stable

Currently, libbpf re-sorts bpf_map structs after all the maps are added and
initialized, which might change their relative order and invalidate any
bpf_map pointer or index taken before that. This is inconvenient and
error-prone. For instance, it can cause .kconfig map index to point to a wrong
map.

Furthermore, libbpf itself doesn't rely on any specific ordering of bpf_maps,
so it's just an unnecessary complication right now. This patch drops sorting
of maps and makes their relative positions fixed. If efficient index is ever
needed, it's better to have a separate array of pointers as a search index,
instead of reordering bpf_map struct in-place. This will be less error-prone
and will allow multiple independent orderings, if necessary (e.g., either by
section index or by name).

Fixes: 166750bc1dd2 ("libbpf: Support libbpf-provided extern variables")
Reported-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200110034247.1220142-1-andriin@fb.com
---
 tools/lib/bpf/libbpf.c | 14 --------------
 1 file changed, 14 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 35a4422ef655..ee2620b2aa55 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -1123,16 +1123,6 @@ bpf_object__init_kversion(struct bpf_object *obj, void *data, size_t size)
 	return 0;
 }
 
-static int compare_bpf_map(const void *_a, const void *_b)
-{
-	const struct bpf_map *a = _a;
-	const struct bpf_map *b = _b;
-
-	if (a->sec_idx != b->sec_idx)
-		return a->sec_idx - b->sec_idx;
-	return a->sec_offset - b->sec_offset;
-}
-
 static bool bpf_map_type__is_map_in_map(enum bpf_map_type type)
 {
 	if (type == BPF_MAP_TYPE_ARRAY_OF_MAPS ||
@@ -2196,10 +2186,6 @@ static int bpf_object__init_maps(struct bpf_object *obj,
 	if (err)
 		return err;
 
-	if (obj->nr_maps) {
-		qsort(obj->maps, obj->nr_maps, sizeof(obj->maps[0]),
-		      compare_bpf_map);
-	}
 	return 0;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 2031af28a42f18cc65ec3aa85b4e5cfdf5ee0fe8 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Thu, 9 Jan 2020 21:17:14 -0800
Subject: libbpf,selftests/bpf: Fix clean targets

Libbpf's clean target should clean out generated files in $(OUTPUT) directory
and not make assumption that $(OUTPUT) directory is current working directory.

Selftest's Makefile should delegate cleaning of libbpf-generated files to
libbpf's Makefile. This ensures more robust clean up.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200110051716.1591485-2-andriin@fb.com
---
 tools/lib/bpf/Makefile               | 9 +++++----
 tools/testing/selftests/bpf/Makefile | 5 +++--
 2 files changed, 8 insertions(+), 6 deletions(-)

diff --git a/tools/lib/bpf/Makefile b/tools/lib/bpf/Makefile
index d87830e7ea63..db2afccde757 100644
--- a/tools/lib/bpf/Makefile
+++ b/tools/lib/bpf/Makefile
@@ -273,10 +273,11 @@ config-clean:
 	$(Q)$(MAKE) -C $(srctree)/tools/build/feature/ clean >/dev/null
 
 clean:
-	$(call QUIET_CLEAN, libbpf) $(RM) -rf $(CMD_TARGETS) \
-		*.o *~ *.a *.so *.so.$(LIBBPF_MAJOR_VERSION) .*.d .*.cmd \
-		*.pc LIBBPF-CFLAGS $(BPF_HELPER_DEFS) \
-		$(SHARED_OBJDIR) $(STATIC_OBJDIR)
+	$(call QUIET_CLEAN, libbpf) $(RM) -rf $(CMD_TARGETS)		     \
+		*~ .*.d .*.cmd LIBBPF-CFLAGS $(BPF_HELPER_DEFS)		     \
+		$(SHARED_OBJDIR) $(STATIC_OBJDIR)			     \
+		$(addprefix $(OUTPUT),					     \
+			    *.o *.a *.so *.so.$(LIBBPF_MAJOR_VERSION) *.pc)
 	$(call QUIET_CLEAN, core-gen) $(RM) $(OUTPUT)FEATURE-DUMP.libbpf
 
 
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index f1f949cd8ed9..cb9f18e4b98b 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -93,6 +93,7 @@ OVERRIDE_TARGETS := 1
 override define CLEAN
 	$(call msg,    CLEAN)
 	$(RM) -r $(TEST_GEN_PROGS) $(TEST_GEN_PROGS_EXTENDED) $(TEST_GEN_FILES) $(EXTRA_CLEAN)
+	$(MAKE) -C $(BPFDIR) OUTPUT=$(OUTPUT)/ clean
 endef
 
 include ../lib.mk
@@ -377,5 +378,5 @@ $(OUTPUT)/test_cpp: test_cpp.cpp $(OUTPUT)/test_core_extern.skel.h $(BPFOBJ)
 
 EXTRA_CLEAN := $(TEST_CUSTOM_PROGS)					\
 	prog_tests/tests.h map_tests/tests.h verifier/tests.h		\
-	feature $(OUTPUT)/*.o $(OUTPUT)/no_alu32 $(OUTPUT)/bpf_gcc	\
-	tools *.skel.h
+	feature								\
+	$(addprefix $(OUTPUT)/,*.o *.skel.h no_alu32 bpf_gcc tools)
-- 
cgit v1.2.3-59-g8ed1b


From 6910d7d3867a8e65eaecb18a9527ecc9eb2e5edf Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Thu, 9 Jan 2020 21:17:15 -0800
Subject: selftests/bpf: Ensure bpf_helper_defs.h are taken from selftests dir

Reorder includes search path to ensure $(OUTPUT) and $(CURDIR) go before
libbpf's directory. Also fix bpf_helpers.h to include bpf_helper_defs.h in
such a way as to leverage includes search path. This allows selftests to not
use libbpf's local and potentially stale bpf_helper_defs.h. It's important
because selftests/bpf's Makefile only re-generates bpf_helper_defs.h in
seltests' output directory, not the one in libbpf's directory.

Also force regeneration of bpf_helper_defs.h when libbpf.a is updated to
reduce staleness.

Fixes: fa633a0f8919 ("libbpf: Fix build on read-only filesystems")
Reported-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200110051716.1591485-3-andriin@fb.com
---
 tools/lib/bpf/bpf_helpers.h          |  2 +-
 tools/testing/selftests/bpf/Makefile | 12 ++++++------
 2 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/tools/lib/bpf/bpf_helpers.h b/tools/lib/bpf/bpf_helpers.h
index f69cc208778a..050bb7bf5be6 100644
--- a/tools/lib/bpf/bpf_helpers.h
+++ b/tools/lib/bpf/bpf_helpers.h
@@ -2,7 +2,7 @@
 #ifndef __BPF_HELPERS__
 #define __BPF_HELPERS__
 
-#include "bpf_helper_defs.h"
+#include <bpf_helper_defs.h>
 
 #define __uint(name, val) int (*name)[val]
 #define __type(name, val) typeof(val) *name
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index cb9f18e4b98b..c0a18994db87 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -20,8 +20,8 @@ CLANG		?= clang
 LLC		?= llc
 LLVM_OBJCOPY	?= llvm-objcopy
 BPF_GCC		?= $(shell command -v bpf-gcc;)
-CFLAGS += -g -Wall -O2 $(GENFLAGS) -I$(APIDIR) -I$(LIBDIR) -I$(BPFDIR)	\
-	  -I$(GENDIR) -I$(TOOLSINCDIR) -I$(CURDIR)			\
+CFLAGS += -g -Wall -O2 $(GENFLAGS) -I$(CURDIR) -I$(APIDIR) -I$(LIBDIR)  \
+	  -I$(BPFDIR) -I$(GENDIR) -I$(TOOLSINCDIR)			\
 	  -Dbpf_prog_load=bpf_prog_test_load				\
 	  -Dbpf_load_program=bpf_test_load_program
 LDLIBS += -lcap -lelf -lz -lrt -lpthread
@@ -153,7 +153,7 @@ $(BPFOBJ): force
 	$(MAKE) -C $(BPFDIR) OUTPUT=$(OUTPUT)/
 
 BPF_HELPERS := $(OUTPUT)/bpf_helper_defs.h $(wildcard $(BPFDIR)/bpf_*.h)
-$(OUTPUT)/bpf_helper_defs.h:
+$(OUTPUT)/bpf_helper_defs.h: $(BPFOBJ)
 	$(MAKE) -C $(BPFDIR) OUTPUT=$(OUTPUT)/ $(OUTPUT)/bpf_helper_defs.h
 
 # Get Clang's default includes on this system, as opposed to those seen by
@@ -174,8 +174,8 @@ MENDIAN=$(if $(IS_LITTLE_ENDIAN),-mlittle-endian,-mbig-endian)
 
 CLANG_SYS_INCLUDES = $(call get_sys_includes,$(CLANG))
 BPF_CFLAGS = -g -D__TARGET_ARCH_$(SRCARCH) $(MENDIAN) 			\
-	     -I. -I./include/uapi -I$(APIDIR)				\
-	     -I$(BPFDIR) -I$(abspath $(OUTPUT)/../usr/include)
+	     -I$(OUTPUT) -I$(CURDIR) -I$(CURDIR)/include/uapi		\
+	     -I$(APIDIR) -I$(BPFDIR) -I$(abspath $(OUTPUT)/../usr/include)
 
 CLANG_CFLAGS = $(CLANG_SYS_INCLUDES) \
 	       -Wno-compare-distinct-pointer-types
@@ -329,7 +329,7 @@ TRUNNER_EXTRA_SOURCES := test_progs.c cgroup_helpers.c trace_helpers.c	\
 TRUNNER_EXTRA_FILES := $(OUTPUT)/urandom_read				\
 		       $(wildcard progs/btf_dump_test_case_*.c)
 TRUNNER_BPF_BUILD_RULE := CLANG_BPF_BUILD_RULE
-TRUNNER_BPF_CFLAGS := -I. -I$(OUTPUT) $(BPF_CFLAGS) $(CLANG_CFLAGS)
+TRUNNER_BPF_CFLAGS := $(BPF_CFLAGS) $(CLANG_CFLAGS)
 TRUNNER_BPF_LDFLAGS := -mattr=+alu32
 $(eval $(call DEFINE_TEST_RUNNER,test_progs))
 
-- 
cgit v1.2.3-59-g8ed1b


From 965b9fee28d062c79a5fdaa3dfbe04c5129ee63d Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Thu, 9 Jan 2020 21:17:16 -0800
Subject: selftests/bpf: Further clean up Makefile output

Further clean up Makefile output:
- hide "entering directory" messages;
- silvence sub-Make command echoing;
- succinct MKDIR messages.

Also remove few test binaries that are not produced anymore from .gitignore.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200110051716.1591485-4-andriin@fb.com
---
 tools/testing/selftests/bpf/.gitignore |  2 --
 tools/testing/selftests/bpf/Makefile   | 10 +++++++---
 2 files changed, 7 insertions(+), 5 deletions(-)

diff --git a/tools/testing/selftests/bpf/.gitignore b/tools/testing/selftests/bpf/.gitignore
index 301ac12d5d69..1d14e3ab70be 100644
--- a/tools/testing/selftests/bpf/.gitignore
+++ b/tools/testing/selftests/bpf/.gitignore
@@ -22,11 +22,9 @@ get_cgroup_id_user
 test_skb_cgroup_id_user
 test_socket_cookie
 test_cgroup_storage
-test_select_reuseport
 test_flow_dissector
 flow_dissector_load
 test_netcnt
-test_section_names
 test_tcpnotify_user
 test_libbpf
 test_tcp_check_syncookie_user
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index c0a18994db87..c28e67548f45 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -83,9 +83,12 @@ TEST_CUSTOM_PROGS = urandom_read
 # $3 - target (assumed to be file); only file name will be emitted;
 # $4 - optional extra arg, emitted as-is, if provided.
 ifeq ($(V),1)
+Q =
 msg =
 else
+Q = @
 msg = @$(info $(1)$(if $(2), [$(2)]) $(notdir $(3)))$(if $(4), $(4))
+MAKEFLAGS += --no-print-directory
 endif
 
 # override lib.mk's default rules
@@ -147,14 +150,14 @@ DEFAULT_BPFTOOL := $(OUTPUT)/tools/usr/local/sbin/bpftool
 BPFTOOL ?= $(DEFAULT_BPFTOOL)
 
 $(DEFAULT_BPFTOOL): force
-	$(MAKE) -C $(BPFTOOLDIR) DESTDIR=$(OUTPUT)/tools install
+	$(Q)$(MAKE) -C $(BPFTOOLDIR) DESTDIR=$(OUTPUT)/tools install
 
 $(BPFOBJ): force
-	$(MAKE) -C $(BPFDIR) OUTPUT=$(OUTPUT)/
+	$(Q)$(MAKE) -C $(BPFDIR) OUTPUT=$(OUTPUT)/
 
 BPF_HELPERS := $(OUTPUT)/bpf_helper_defs.h $(wildcard $(BPFDIR)/bpf_*.h)
 $(OUTPUT)/bpf_helper_defs.h: $(BPFOBJ)
-	$(MAKE) -C $(BPFDIR) OUTPUT=$(OUTPUT)/ $(OUTPUT)/bpf_helper_defs.h
+	$(Q)$(MAKE) -C $(BPFDIR) OUTPUT=$(OUTPUT)/ $(OUTPUT)/bpf_helper_defs.h
 
 # Get Clang's default includes on this system, as opposed to those seen by
 # '-target bpf'. This fixes "missing" files on some architectures/distros,
@@ -253,6 +256,7 @@ define DEFINE_TEST_RUNNER_RULES
 ifeq ($($(TRUNNER_OUTPUT)-dir),)
 $(TRUNNER_OUTPUT)-dir := y
 $(TRUNNER_OUTPUT):
+	$$(call msg,      MKDIR,,$$@)
 	mkdir -p $$@
 endif
 
-- 
cgit v1.2.3-59-g8ed1b


From 2d3eb67f64ec317bbfe340cfcc8325d40a6ff317 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Thu, 9 Jan 2020 22:41:19 -0800
Subject: libbpf: Sanitize global functions

In case the kernel doesn't support BTF_FUNC_GLOBAL sanitize BTF produced by the
compiler for global functions.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Song Liu <songliubraving@fb.com>
Link: https://lore.kernel.org/bpf/20200110064124.1760511-2-ast@kernel.org
---
 tools/include/uapi/linux/btf.h |  6 ++++++
 tools/lib/bpf/libbpf.c         | 35 ++++++++++++++++++++++++++++++++++-
 2 files changed, 40 insertions(+), 1 deletion(-)

diff --git a/tools/include/uapi/linux/btf.h b/tools/include/uapi/linux/btf.h
index 1a2898c482ee..5a667107ad2c 100644
--- a/tools/include/uapi/linux/btf.h
+++ b/tools/include/uapi/linux/btf.h
@@ -146,6 +146,12 @@ enum {
 	BTF_VAR_GLOBAL_EXTERN = 2,
 };
 
+enum btf_func_linkage {
+	BTF_FUNC_STATIC = 0,
+	BTF_FUNC_GLOBAL = 1,
+	BTF_FUNC_EXTERN = 2,
+};
+
 /* BTF_KIND_VAR is followed by a single "struct btf_var" to describe
  * additional information related to the variable such as its linkage.
  */
diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index ee2620b2aa55..3afd780b0f06 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -173,6 +173,8 @@ struct bpf_capabilities {
 	__u32 btf_datasec:1;
 	/* BPF_F_MMAPABLE is supported for arrays */
 	__u32 array_mmap:1;
+	/* BTF_FUNC_GLOBAL is supported */
+	__u32 btf_func_global:1;
 };
 
 enum reloc_type {
@@ -2209,13 +2211,14 @@ static bool section_have_execinstr(struct bpf_object *obj, int idx)
 
 static void bpf_object__sanitize_btf(struct bpf_object *obj)
 {
+	bool has_func_global = obj->caps.btf_func_global;
 	bool has_datasec = obj->caps.btf_datasec;
 	bool has_func = obj->caps.btf_func;
 	struct btf *btf = obj->btf;
 	struct btf_type *t;
 	int i, j, vlen;
 
-	if (!obj->btf || (has_func && has_datasec))
+	if (!obj->btf || (has_func && has_datasec && has_func_global))
 		return;
 
 	for (i = 1; i <= btf__get_nr_types(btf); i++) {
@@ -2263,6 +2266,9 @@ static void bpf_object__sanitize_btf(struct bpf_object *obj)
 		} else if (!has_func && btf_is_func(t)) {
 			/* replace FUNC with TYPEDEF */
 			t->info = BTF_INFO_ENC(BTF_KIND_TYPEDEF, 0, 0);
+		} else if (!has_func_global && btf_is_func(t)) {
+			/* replace BTF_FUNC_GLOBAL with BTF_FUNC_STATIC */
+			t->info = BTF_INFO_ENC(BTF_KIND_FUNC, 0, 0);
 		}
 	}
 }
@@ -3205,6 +3211,32 @@ static int bpf_object__probe_btf_func(struct bpf_object *obj)
 	return 0;
 }
 
+static int bpf_object__probe_btf_func_global(struct bpf_object *obj)
+{
+	static const char strs[] = "\0int\0x\0a";
+	/* static void x(int a) {} */
+	__u32 types[] = {
+		/* int */
+		BTF_TYPE_INT_ENC(1, BTF_INT_SIGNED, 0, 32, 4),  /* [1] */
+		/* FUNC_PROTO */                                /* [2] */
+		BTF_TYPE_ENC(0, BTF_INFO_ENC(BTF_KIND_FUNC_PROTO, 0, 1), 0),
+		BTF_PARAM_ENC(7, 1),
+		/* FUNC x BTF_FUNC_GLOBAL */                    /* [3] */
+		BTF_TYPE_ENC(5, BTF_INFO_ENC(BTF_KIND_FUNC, 0, BTF_FUNC_GLOBAL), 2),
+	};
+	int btf_fd;
+
+	btf_fd = libbpf__load_raw_btf((char *)types, sizeof(types),
+				      strs, sizeof(strs));
+	if (btf_fd >= 0) {
+		obj->caps.btf_func_global = 1;
+		close(btf_fd);
+		return 1;
+	}
+
+	return 0;
+}
+
 static int bpf_object__probe_btf_datasec(struct bpf_object *obj)
 {
 	static const char strs[] = "\0x\0.data";
@@ -3260,6 +3292,7 @@ bpf_object__probe_caps(struct bpf_object *obj)
 		bpf_object__probe_name,
 		bpf_object__probe_global_data,
 		bpf_object__probe_btf_func,
+		bpf_object__probe_btf_func_global,
 		bpf_object__probe_btf_datasec,
 		bpf_object__probe_array_mmap,
 	};
-- 
cgit v1.2.3-59-g8ed1b


From 51c39bb1d5d105a02e29aa7960f0a395086e6342 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Thu, 9 Jan 2020 22:41:20 -0800
Subject: bpf: Introduce function-by-function verification

New llvm and old llvm with libbpf help produce BTF that distinguish global and
static functions. Unlike arguments of static function the arguments of global
functions cannot be removed or optimized away by llvm. The compiler has to use
exactly the arguments specified in a function prototype. The argument type
information allows the verifier validate each global function independently.
For now only supported argument types are pointer to context and scalars. In
the future pointers to structures, sizes, pointer to packet data can be
supported as well. Consider the following example:

static int f1(int ...)
{
  ...
}

int f3(int b);

int f2(int a)
{
  f1(a) + f3(a);
}

int f3(int b)
{
  ...
}

int main(...)
{
  f1(...) + f2(...) + f3(...);
}

The verifier will start its safety checks from the first global function f2().
It will recursively descend into f1() because it's static. Then it will check
that arguments match for the f3() invocation inside f2(). It will not descend
into f3(). It will finish f2() that has to be successfully verified for all
possible values of 'a'. Then it will proceed with f3(). That function also has
to be safe for all possible values of 'b'. Then it will start subprog 0 (which
is main() function). It will recursively descend into f1() and will skip full
check of f2() and f3(), since they are global. The order of processing global
functions doesn't affect safety, since all global functions must be proven safe
based on their arguments only.

Such function by function verification can drastically improve speed of the
verification and reduce complexity.

Note that the stack limit of 512 still applies to the call chain regardless whether
functions were static or global. The nested level of 8 also still applies. The
same recursion prevention checks are in place as well.

The type information and static/global kind is preserved after the verification
hence in the above example global function f2() and f3() can be replaced later
by equivalent functions with the same types that are loaded and verified later
without affecting safety of this main() program. Such replacement (re-linking)
of global functions is a subject of future patches.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Song Liu <songliubraving@fb.com>
Link: https://lore.kernel.org/bpf/20200110064124.1760511-3-ast@kernel.org
---
 include/linux/bpf.h          |   7 +-
 include/linux/bpf_verifier.h |  10 +-
 include/uapi/linux/btf.h     |   6 ++
 kernel/bpf/btf.c             | 175 ++++++++++++++++++++++++------
 kernel/bpf/verifier.c        | 252 ++++++++++++++++++++++++++++++++++---------
 5 files changed, 366 insertions(+), 84 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index a7bfe8a388c6..aed2bc39d72b 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -566,6 +566,7 @@ static inline void bpf_dispatcher_change_prog(struct bpf_dispatcher *d,
 #endif
 
 struct bpf_func_info_aux {
+	u16 linkage;
 	bool unreliable;
 };
 
@@ -1081,7 +1082,11 @@ int btf_distill_func_proto(struct bpf_verifier_log *log,
 			   const char *func_name,
 			   struct btf_func_model *m);
 
-int btf_check_func_arg_match(struct bpf_verifier_env *env, int subprog);
+struct bpf_reg_state;
+int btf_check_func_arg_match(struct bpf_verifier_env *env, int subprog,
+			     struct bpf_reg_state *regs);
+int btf_prepare_func_args(struct bpf_verifier_env *env, int subprog,
+			  struct bpf_reg_state *reg);
 
 struct bpf_prog *bpf_prog_by_id(u32 id);
 
diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h
index 26e40de9ef55..5406e6e96585 100644
--- a/include/linux/bpf_verifier.h
+++ b/include/linux/bpf_verifier.h
@@ -304,11 +304,13 @@ struct bpf_insn_aux_data {
 	u64 map_key_state; /* constant (32 bit) key tracking for maps */
 	int ctx_field_size; /* the ctx field size for load insn, maybe 0 */
 	int sanitize_stack_off; /* stack slot to be cleared */
-	bool seen; /* this insn was processed by the verifier */
+	u32 seen; /* this insn was processed by the verifier at env->pass_cnt */
 	bool zext_dst; /* this insn zero extends dst reg */
 	u8 alu_state; /* used in combination with alu_limit */
-	bool prune_point;
+
+	/* below fields are initialized once */
 	unsigned int orig_idx; /* original instruction index */
+	bool prune_point;
 };
 
 #define MAX_USED_MAPS 64 /* max number of maps accessed by one eBPF program */
@@ -379,6 +381,7 @@ struct bpf_verifier_env {
 		int *insn_stack;
 		int cur_stack;
 	} cfg;
+	u32 pass_cnt; /* number of times do_check() was called */
 	u32 subprog_cnt;
 	/* number of instructions analyzed by the verifier */
 	u32 prev_insn_processed, insn_processed;
@@ -428,4 +431,7 @@ bpf_prog_offload_replace_insn(struct bpf_verifier_env *env, u32 off,
 void
 bpf_prog_offload_remove_insns(struct bpf_verifier_env *env, u32 off, u32 cnt);
 
+int check_ctx_reg(struct bpf_verifier_env *env,
+		  const struct bpf_reg_state *reg, int regno);
+
 #endif /* _LINUX_BPF_VERIFIER_H */
diff --git a/include/uapi/linux/btf.h b/include/uapi/linux/btf.h
index 1a2898c482ee..5a667107ad2c 100644
--- a/include/uapi/linux/btf.h
+++ b/include/uapi/linux/btf.h
@@ -146,6 +146,12 @@ enum {
 	BTF_VAR_GLOBAL_EXTERN = 2,
 };
 
+enum btf_func_linkage {
+	BTF_FUNC_STATIC = 0,
+	BTF_FUNC_GLOBAL = 1,
+	BTF_FUNC_EXTERN = 2,
+};
+
 /* BTF_KIND_VAR is followed by a single "struct btf_var" to describe
  * additional information related to the variable such as its linkage.
  */
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 81d9cf75cacd..832b5d7fd892 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -2651,8 +2651,8 @@ static s32 btf_func_check_meta(struct btf_verifier_env *env,
 		return -EINVAL;
 	}
 
-	if (btf_type_vlen(t)) {
-		btf_verifier_log_type(env, t, "vlen != 0");
+	if (btf_type_vlen(t) > BTF_FUNC_GLOBAL) {
+		btf_verifier_log_type(env, t, "Invalid func linkage");
 		return -EINVAL;
 	}
 
@@ -3506,7 +3506,8 @@ static u8 bpf_ctx_convert_map[] = {
 
 static const struct btf_member *
 btf_get_prog_ctx_type(struct bpf_verifier_log *log, struct btf *btf,
-		      const struct btf_type *t, enum bpf_prog_type prog_type)
+		      const struct btf_type *t, enum bpf_prog_type prog_type,
+		      int arg)
 {
 	const struct btf_type *conv_struct;
 	const struct btf_type *ctx_struct;
@@ -3527,12 +3528,13 @@ btf_get_prog_ctx_type(struct bpf_verifier_log *log, struct btf *btf,
 		 * is not supported yet.
 		 * BPF_PROG_TYPE_RAW_TRACEPOINT is fine.
 		 */
-		bpf_log(log, "BPF program ctx type is not a struct\n");
+		if (log->level & BPF_LOG_LEVEL)
+			bpf_log(log, "arg#%d type is not a struct\n", arg);
 		return NULL;
 	}
 	tname = btf_name_by_offset(btf, t->name_off);
 	if (!tname) {
-		bpf_log(log, "BPF program ctx struct doesn't have a name\n");
+		bpf_log(log, "arg#%d struct doesn't have a name\n", arg);
 		return NULL;
 	}
 	/* prog_type is valid bpf program type. No need for bounds check. */
@@ -3565,11 +3567,12 @@ btf_get_prog_ctx_type(struct bpf_verifier_log *log, struct btf *btf,
 static int btf_translate_to_vmlinux(struct bpf_verifier_log *log,
 				     struct btf *btf,
 				     const struct btf_type *t,
-				     enum bpf_prog_type prog_type)
+				     enum bpf_prog_type prog_type,
+				     int arg)
 {
 	const struct btf_member *prog_ctx_type, *kern_ctx_type;
 
-	prog_ctx_type = btf_get_prog_ctx_type(log, btf, t, prog_type);
+	prog_ctx_type = btf_get_prog_ctx_type(log, btf, t, prog_type, arg);
 	if (!prog_ctx_type)
 		return -ENOENT;
 	kern_ctx_type = prog_ctx_type + 1;
@@ -3731,7 +3734,7 @@ bool btf_ctx_access(int off, int size, enum bpf_access_type type,
 	info->reg_type = PTR_TO_BTF_ID;
 
 	if (tgt_prog) {
-		ret = btf_translate_to_vmlinux(log, btf, t, tgt_prog->type);
+		ret = btf_translate_to_vmlinux(log, btf, t, tgt_prog->type, arg);
 		if (ret > 0) {
 			info->btf_id = ret;
 			return true;
@@ -4112,11 +4115,16 @@ int btf_distill_func_proto(struct bpf_verifier_log *log,
 	return 0;
 }
 
-int btf_check_func_arg_match(struct bpf_verifier_env *env, int subprog)
+/* Compare BTF of a function with given bpf_reg_state.
+ * Returns:
+ * EFAULT - there is a verifier bug. Abort verification.
+ * EINVAL - there is a type mismatch or BTF is not available.
+ * 0 - BTF matches with what bpf_reg_state expects.
+ * Only PTR_TO_CTX and SCALAR_VALUE states are recognized.
+ */
+int btf_check_func_arg_match(struct bpf_verifier_env *env, int subprog,
+			     struct bpf_reg_state *reg)
 {
-	struct bpf_verifier_state *st = env->cur_state;
-	struct bpf_func_state *func = st->frame[st->curframe];
-	struct bpf_reg_state *reg = func->regs;
 	struct bpf_verifier_log *log = &env->log;
 	struct bpf_prog *prog = env->prog;
 	struct btf *btf = prog->aux->btf;
@@ -4126,27 +4134,30 @@ int btf_check_func_arg_match(struct bpf_verifier_env *env, int subprog)
 	const char *tname;
 
 	if (!prog->aux->func_info)
-		return 0;
+		return -EINVAL;
 
 	btf_id = prog->aux->func_info[subprog].type_id;
 	if (!btf_id)
-		return 0;
+		return -EFAULT;
 
 	if (prog->aux->func_info_aux[subprog].unreliable)
-		return 0;
+		return -EINVAL;
 
 	t = btf_type_by_id(btf, btf_id);
 	if (!t || !btf_type_is_func(t)) {
-		bpf_log(log, "BTF of subprog %d doesn't point to KIND_FUNC\n",
+		/* These checks were already done by the verifier while loading
+		 * struct bpf_func_info
+		 */
+		bpf_log(log, "BTF of func#%d doesn't point to KIND_FUNC\n",
 			subprog);
-		return -EINVAL;
+		return -EFAULT;
 	}
 	tname = btf_name_by_offset(btf, t->name_off);
 
 	t = btf_type_by_id(btf, t->type);
 	if (!t || !btf_type_is_func_proto(t)) {
-		bpf_log(log, "Invalid type of func %s\n", tname);
-		return -EINVAL;
+		bpf_log(log, "Invalid BTF of func %s\n", tname);
+		return -EFAULT;
 	}
 	args = (const struct btf_param *)(t + 1);
 	nargs = btf_type_vlen(t);
@@ -4172,25 +4183,127 @@ int btf_check_func_arg_match(struct bpf_verifier_env *env, int subprog)
 				bpf_log(log, "R%d is not a pointer\n", i + 1);
 				goto out;
 			}
-			/* If program is passing PTR_TO_CTX into subprogram
-			 * check that BTF type matches.
+			/* If function expects ctx type in BTF check that caller
+			 * is passing PTR_TO_CTX.
 			 */
-			if (reg[i + 1].type == PTR_TO_CTX &&
-			    !btf_get_prog_ctx_type(log, btf, t, prog->type))
-				goto out;
-			/* All other pointers are ok */
-			continue;
+			if (btf_get_prog_ctx_type(log, btf, t, prog->type, i)) {
+				if (reg[i + 1].type != PTR_TO_CTX) {
+					bpf_log(log,
+						"arg#%d expected pointer to ctx, but got %s\n",
+						i, btf_kind_str[BTF_INFO_KIND(t->info)]);
+					goto out;
+				}
+				if (check_ctx_reg(env, &reg[i + 1], i + 1))
+					goto out;
+				continue;
+			}
 		}
-		bpf_log(log, "Unrecognized argument type %s\n",
-			btf_kind_str[BTF_INFO_KIND(t->info)]);
+		bpf_log(log, "Unrecognized arg#%d type %s\n",
+			i, btf_kind_str[BTF_INFO_KIND(t->info)]);
 		goto out;
 	}
 	return 0;
 out:
-	/* LLVM optimizations can remove arguments from static functions. */
-	bpf_log(log,
-		"Type info disagrees with actual arguments due to compiler optimizations\n");
+	/* Compiler optimizations can remove arguments from static functions
+	 * or mismatched type can be passed into a global function.
+	 * In such cases mark the function as unreliable from BTF point of view.
+	 */
 	prog->aux->func_info_aux[subprog].unreliable = true;
+	return -EINVAL;
+}
+
+/* Convert BTF of a function into bpf_reg_state if possible
+ * Returns:
+ * EFAULT - there is a verifier bug. Abort verification.
+ * EINVAL - cannot convert BTF.
+ * 0 - Successfully converted BTF into bpf_reg_state
+ * (either PTR_TO_CTX or SCALAR_VALUE).
+ */
+int btf_prepare_func_args(struct bpf_verifier_env *env, int subprog,
+			  struct bpf_reg_state *reg)
+{
+	struct bpf_verifier_log *log = &env->log;
+	struct bpf_prog *prog = env->prog;
+	struct btf *btf = prog->aux->btf;
+	const struct btf_param *args;
+	const struct btf_type *t;
+	u32 i, nargs, btf_id;
+	const char *tname;
+
+	if (!prog->aux->func_info ||
+	    prog->aux->func_info_aux[subprog].linkage != BTF_FUNC_GLOBAL) {
+		bpf_log(log, "Verifier bug\n");
+		return -EFAULT;
+	}
+
+	btf_id = prog->aux->func_info[subprog].type_id;
+	if (!btf_id) {
+		bpf_log(log, "Global functions need valid BTF\n");
+		return -EFAULT;
+	}
+
+	t = btf_type_by_id(btf, btf_id);
+	if (!t || !btf_type_is_func(t)) {
+		/* These checks were already done by the verifier while loading
+		 * struct bpf_func_info
+		 */
+		bpf_log(log, "BTF of func#%d doesn't point to KIND_FUNC\n",
+			subprog);
+		return -EFAULT;
+	}
+	tname = btf_name_by_offset(btf, t->name_off);
+
+	if (log->level & BPF_LOG_LEVEL)
+		bpf_log(log, "Validating %s() func#%d...\n",
+			tname, subprog);
+
+	if (prog->aux->func_info_aux[subprog].unreliable) {
+		bpf_log(log, "Verifier bug in function %s()\n", tname);
+		return -EFAULT;
+	}
+
+	t = btf_type_by_id(btf, t->type);
+	if (!t || !btf_type_is_func_proto(t)) {
+		bpf_log(log, "Invalid type of function %s()\n", tname);
+		return -EFAULT;
+	}
+	args = (const struct btf_param *)(t + 1);
+	nargs = btf_type_vlen(t);
+	if (nargs > 5) {
+		bpf_log(log, "Global function %s() with %d > 5 args. Buggy compiler.\n",
+			tname, nargs);
+		return -EINVAL;
+	}
+	/* check that function returns int */
+	t = btf_type_by_id(btf, t->type);
+	while (btf_type_is_modifier(t))
+		t = btf_type_by_id(btf, t->type);
+	if (!btf_type_is_int(t) && !btf_type_is_enum(t)) {
+		bpf_log(log,
+			"Global function %s() doesn't return scalar. Only those are supported.\n",
+			tname);
+		return -EINVAL;
+	}
+	/* Convert BTF function arguments into verifier types.
+	 * Only PTR_TO_CTX and SCALAR are supported atm.
+	 */
+	for (i = 0; i < nargs; i++) {
+		t = btf_type_by_id(btf, args[i].type);
+		while (btf_type_is_modifier(t))
+			t = btf_type_by_id(btf, t->type);
+		if (btf_type_is_int(t) || btf_type_is_enum(t)) {
+			reg[i + 1].type = SCALAR_VALUE;
+			continue;
+		}
+		if (btf_type_is_ptr(t) &&
+		    btf_get_prog_ctx_type(log, btf, t, prog->type, i)) {
+			reg[i + 1].type = PTR_TO_CTX;
+			continue;
+		}
+		bpf_log(log, "Arg#%d type %s in %s() is not supported yet.\n",
+			i, btf_kind_str[BTF_INFO_KIND(t->info)], tname);
+		return -EINVAL;
+	}
 	return 0;
 }
 
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index f5af759a8a5f..ca17dccc17ba 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -1122,10 +1122,6 @@ static void init_reg_state(struct bpf_verifier_env *env,
 	regs[BPF_REG_FP].type = PTR_TO_STACK;
 	mark_reg_known_zero(env, regs, BPF_REG_FP);
 	regs[BPF_REG_FP].frameno = state->frameno;
-
-	/* 1st arg to a function */
-	regs[BPF_REG_1].type = PTR_TO_CTX;
-	mark_reg_known_zero(env, regs, BPF_REG_1);
 }
 
 #define BPF_MAIN_FUNC (-1)
@@ -2739,8 +2735,8 @@ static int get_callee_stack_depth(struct bpf_verifier_env *env,
 }
 #endif
 
-static int check_ctx_reg(struct bpf_verifier_env *env,
-			 const struct bpf_reg_state *reg, int regno)
+int check_ctx_reg(struct bpf_verifier_env *env,
+		  const struct bpf_reg_state *reg, int regno)
 {
 	/* Access to ctx or passing it to a helper is only allowed in
 	 * its original, unmodified form.
@@ -3956,12 +3952,26 @@ static int release_reference(struct bpf_verifier_env *env,
 	return 0;
 }
 
+static void clear_caller_saved_regs(struct bpf_verifier_env *env,
+				    struct bpf_reg_state *regs)
+{
+	int i;
+
+	/* after the call registers r0 - r5 were scratched */
+	for (i = 0; i < CALLER_SAVED_REGS; i++) {
+		mark_reg_not_init(env, regs, caller_saved[i]);
+		check_reg_arg(env, caller_saved[i], DST_OP_NO_MARK);
+	}
+}
+
 static int check_func_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 			   int *insn_idx)
 {
 	struct bpf_verifier_state *state = env->cur_state;
+	struct bpf_func_info_aux *func_info_aux;
 	struct bpf_func_state *caller, *callee;
 	int i, err, subprog, target_insn;
+	bool is_global = false;
 
 	if (state->curframe + 1 >= MAX_CALL_FRAMES) {
 		verbose(env, "the call stack of %d frames is too deep\n",
@@ -3984,6 +3994,32 @@ static int check_func_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 		return -EFAULT;
 	}
 
+	func_info_aux = env->prog->aux->func_info_aux;
+	if (func_info_aux)
+		is_global = func_info_aux[subprog].linkage == BTF_FUNC_GLOBAL;
+	err = btf_check_func_arg_match(env, subprog, caller->regs);
+	if (err == -EFAULT)
+		return err;
+	if (is_global) {
+		if (err) {
+			verbose(env, "Caller passes invalid args into func#%d\n",
+				subprog);
+			return err;
+		} else {
+			if (env->log.level & BPF_LOG_LEVEL)
+				verbose(env,
+					"Func#%d is global and valid. Skipping.\n",
+					subprog);
+			clear_caller_saved_regs(env, caller->regs);
+
+			/* All global functions return SCALAR_VALUE */
+			mark_reg_unknown(env, caller->regs, BPF_REG_0);
+
+			/* continue with next insn after call */
+			return 0;
+		}
+	}
+
 	callee = kzalloc(sizeof(*callee), GFP_KERNEL);
 	if (!callee)
 		return -ENOMEM;
@@ -4010,18 +4046,11 @@ static int check_func_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 	for (i = BPF_REG_1; i <= BPF_REG_5; i++)
 		callee->regs[i] = caller->regs[i];
 
-	/* after the call registers r0 - r5 were scratched */
-	for (i = 0; i < CALLER_SAVED_REGS; i++) {
-		mark_reg_not_init(env, caller->regs, caller_saved[i]);
-		check_reg_arg(env, caller_saved[i], DST_OP_NO_MARK);
-	}
+	clear_caller_saved_regs(env, caller->regs);
 
 	/* only increment it after check_reg_arg() finished */
 	state->curframe++;
 
-	if (btf_check_func_arg_match(env, subprog))
-		return -EINVAL;
-
 	/* and go analyze first insn of the callee */
 	*insn_idx = target_insn;
 
@@ -6771,12 +6800,13 @@ static int check_btf_func(struct bpf_verifier_env *env,
 
 		/* check type_id */
 		type = btf_type_by_id(btf, krecord[i].type_id);
-		if (!type || BTF_INFO_KIND(type->info) != BTF_KIND_FUNC) {
+		if (!type || !btf_type_is_func(type)) {
 			verbose(env, "invalid type id %d in func info",
 				krecord[i].type_id);
 			ret = -EINVAL;
 			goto err_free;
 		}
+		info_aux[i].linkage = BTF_INFO_VLEN(type->info);
 		prev_offset = krecord[i].insn_off;
 		urecord += urec_size;
 	}
@@ -7756,35 +7786,13 @@ static bool reg_type_mismatch(enum bpf_reg_type src, enum bpf_reg_type prev)
 
 static int do_check(struct bpf_verifier_env *env)
 {
-	struct bpf_verifier_state *state;
+	struct bpf_verifier_state *state = env->cur_state;
 	struct bpf_insn *insns = env->prog->insnsi;
 	struct bpf_reg_state *regs;
 	int insn_cnt = env->prog->len;
 	bool do_print_state = false;
 	int prev_insn_idx = -1;
 
-	env->prev_linfo = NULL;
-
-	state = kzalloc(sizeof(struct bpf_verifier_state), GFP_KERNEL);
-	if (!state)
-		return -ENOMEM;
-	state->curframe = 0;
-	state->speculative = false;
-	state->branches = 1;
-	state->frame[0] = kzalloc(sizeof(struct bpf_func_state), GFP_KERNEL);
-	if (!state->frame[0]) {
-		kfree(state);
-		return -ENOMEM;
-	}
-	env->cur_state = state;
-	init_func_state(env, state->frame[0],
-			BPF_MAIN_FUNC /* callsite */,
-			0 /* frameno */,
-			0 /* subprogno, zero == main subprog */);
-
-	if (btf_check_func_arg_match(env, 0))
-		return -EINVAL;
-
 	for (;;) {
 		struct bpf_insn *insn;
 		u8 class;
@@ -7862,7 +7870,7 @@ static int do_check(struct bpf_verifier_env *env)
 		}
 
 		regs = cur_regs(env);
-		env->insn_aux_data[env->insn_idx].seen = true;
+		env->insn_aux_data[env->insn_idx].seen = env->pass_cnt;
 		prev_insn_idx = env->insn_idx;
 
 		if (class == BPF_ALU || class == BPF_ALU64) {
@@ -8082,7 +8090,7 @@ process_bpf_exit:
 					return err;
 
 				env->insn_idx++;
-				env->insn_aux_data[env->insn_idx].seen = true;
+				env->insn_aux_data[env->insn_idx].seen = env->pass_cnt;
 			} else {
 				verbose(env, "invalid BPF_LD mode\n");
 				return -EINVAL;
@@ -8095,7 +8103,6 @@ process_bpf_exit:
 		env->insn_idx++;
 	}
 
-	env->prog->aux->stack_depth = env->subprog_info[0].stack_depth;
 	return 0;
 }
 
@@ -8372,7 +8379,7 @@ static int adjust_insn_aux_data(struct bpf_verifier_env *env,
 	memcpy(new_data + off + cnt - 1, old_data + off,
 	       sizeof(struct bpf_insn_aux_data) * (prog_len - off - cnt + 1));
 	for (i = off; i < off + cnt - 1; i++) {
-		new_data[i].seen = true;
+		new_data[i].seen = env->pass_cnt;
 		new_data[i].zext_dst = insn_has_def32(env, insn + i);
 	}
 	env->insn_aux_data = new_data;
@@ -9484,6 +9491,7 @@ static void free_states(struct bpf_verifier_env *env)
 		kfree(sl);
 		sl = sln;
 	}
+	env->free_list = NULL;
 
 	if (!env->explored_states)
 		return;
@@ -9497,11 +9505,159 @@ static void free_states(struct bpf_verifier_env *env)
 			kfree(sl);
 			sl = sln;
 		}
+		env->explored_states[i] = NULL;
 	}
+}
 
-	kvfree(env->explored_states);
+/* The verifier is using insn_aux_data[] to store temporary data during
+ * verification and to store information for passes that run after the
+ * verification like dead code sanitization. do_check_common() for subprogram N
+ * may analyze many other subprograms. sanitize_insn_aux_data() clears all
+ * temporary data after do_check_common() finds that subprogram N cannot be
+ * verified independently. pass_cnt counts the number of times
+ * do_check_common() was run and insn->aux->seen tells the pass number
+ * insn_aux_data was touched. These variables are compared to clear temporary
+ * data from failed pass. For testing and experiments do_check_common() can be
+ * run multiple times even when prior attempt to verify is unsuccessful.
+ */
+static void sanitize_insn_aux_data(struct bpf_verifier_env *env)
+{
+	struct bpf_insn *insn = env->prog->insnsi;
+	struct bpf_insn_aux_data *aux;
+	int i, class;
+
+	for (i = 0; i < env->prog->len; i++) {
+		class = BPF_CLASS(insn[i].code);
+		if (class != BPF_LDX && class != BPF_STX)
+			continue;
+		aux = &env->insn_aux_data[i];
+		if (aux->seen != env->pass_cnt)
+			continue;
+		memset(aux, 0, offsetof(typeof(*aux), orig_idx));
+	}
 }
 
+static int do_check_common(struct bpf_verifier_env *env, int subprog)
+{
+	struct bpf_verifier_state *state;
+	struct bpf_reg_state *regs;
+	int ret, i;
+
+	env->prev_linfo = NULL;
+	env->pass_cnt++;
+
+	state = kzalloc(sizeof(struct bpf_verifier_state), GFP_KERNEL);
+	if (!state)
+		return -ENOMEM;
+	state->curframe = 0;
+	state->speculative = false;
+	state->branches = 1;
+	state->frame[0] = kzalloc(sizeof(struct bpf_func_state), GFP_KERNEL);
+	if (!state->frame[0]) {
+		kfree(state);
+		return -ENOMEM;
+	}
+	env->cur_state = state;
+	init_func_state(env, state->frame[0],
+			BPF_MAIN_FUNC /* callsite */,
+			0 /* frameno */,
+			subprog);
+
+	regs = state->frame[state->curframe]->regs;
+	if (subprog) {
+		ret = btf_prepare_func_args(env, subprog, regs);
+		if (ret)
+			goto out;
+		for (i = BPF_REG_1; i <= BPF_REG_5; i++) {
+			if (regs[i].type == PTR_TO_CTX)
+				mark_reg_known_zero(env, regs, i);
+			else if (regs[i].type == SCALAR_VALUE)
+				mark_reg_unknown(env, regs, i);
+		}
+	} else {
+		/* 1st arg to a function */
+		regs[BPF_REG_1].type = PTR_TO_CTX;
+		mark_reg_known_zero(env, regs, BPF_REG_1);
+		ret = btf_check_func_arg_match(env, subprog, regs);
+		if (ret == -EFAULT)
+			/* unlikely verifier bug. abort.
+			 * ret == 0 and ret < 0 are sadly acceptable for
+			 * main() function due to backward compatibility.
+			 * Like socket filter program may be written as:
+			 * int bpf_prog(struct pt_regs *ctx)
+			 * and never dereference that ctx in the program.
+			 * 'struct pt_regs' is a type mismatch for socket
+			 * filter that should be using 'struct __sk_buff'.
+			 */
+			goto out;
+	}
+
+	ret = do_check(env);
+out:
+	free_verifier_state(env->cur_state, true);
+	env->cur_state = NULL;
+	while (!pop_stack(env, NULL, NULL));
+	free_states(env);
+	if (ret)
+		/* clean aux data in case subprog was rejected */
+		sanitize_insn_aux_data(env);
+	return ret;
+}
+
+/* Verify all global functions in a BPF program one by one based on their BTF.
+ * All global functions must pass verification. Otherwise the whole program is rejected.
+ * Consider:
+ * int bar(int);
+ * int foo(int f)
+ * {
+ *    return bar(f);
+ * }
+ * int bar(int b)
+ * {
+ *    ...
+ * }
+ * foo() will be verified first for R1=any_scalar_value. During verification it
+ * will be assumed that bar() already verified successfully and call to bar()
+ * from foo() will be checked for type match only. Later bar() will be verified
+ * independently to check that it's safe for R1=any_scalar_value.
+ */
+static int do_check_subprogs(struct bpf_verifier_env *env)
+{
+	struct bpf_prog_aux *aux = env->prog->aux;
+	int i, ret;
+
+	if (!aux->func_info)
+		return 0;
+
+	for (i = 1; i < env->subprog_cnt; i++) {
+		if (aux->func_info_aux[i].linkage != BTF_FUNC_GLOBAL)
+			continue;
+		env->insn_idx = env->subprog_info[i].start;
+		WARN_ON_ONCE(env->insn_idx == 0);
+		ret = do_check_common(env, i);
+		if (ret) {
+			return ret;
+		} else if (env->log.level & BPF_LOG_LEVEL) {
+			verbose(env,
+				"Func#%d is safe for any args that match its prototype\n",
+				i);
+		}
+	}
+	return 0;
+}
+
+static int do_check_main(struct bpf_verifier_env *env)
+{
+	int ret;
+
+	env->insn_idx = 0;
+	ret = do_check_common(env, 0);
+	if (!ret)
+		env->prog->aux->stack_depth = env->subprog_info[0].stack_depth;
+	return ret;
+}
+
+
 static void print_verification_stats(struct bpf_verifier_env *env)
 {
 	int i;
@@ -9849,18 +10005,14 @@ int bpf_check(struct bpf_prog **prog, union bpf_attr *attr,
 	if (ret < 0)
 		goto skip_full_check;
 
-	ret = do_check(env);
-	if (env->cur_state) {
-		free_verifier_state(env->cur_state, true);
-		env->cur_state = NULL;
-	}
+	ret = do_check_subprogs(env);
+	ret = ret ?: do_check_main(env);
 
 	if (ret == 0 && bpf_prog_is_dev_bound(env->prog->aux))
 		ret = bpf_prog_offload_finalize(env);
 
 skip_full_check:
-	while (!pop_stack(env, NULL, NULL));
-	free_states(env);
+	kvfree(env->explored_states);
 
 	if (ret == 0)
 		ret = check_max_stack_depth(env);
-- 
cgit v1.2.3-59-g8ed1b


From 7608e4db6ded8e965dd213baddedba9602959e7d Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Thu, 9 Jan 2020 22:41:21 -0800
Subject: selftests/bpf: Add fexit-to-skb test for global funcs

Add simple fexit prog type to skb prog type test when subprogram is a global
function.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Song Liu <songliubraving@fb.com>
Link: https://lore.kernel.org/bpf/20200110064124.1760511-4-ast@kernel.org
---
 .../selftests/bpf/prog_tests/fexit_bpf2bpf.c       |  1 +
 tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c  | 15 ++++++++++++
 .../testing/selftests/bpf/progs/test_pkt_access.c  | 28 ++++++++++++++++++++++
 3 files changed, 44 insertions(+)

diff --git a/tools/testing/selftests/bpf/prog_tests/fexit_bpf2bpf.c b/tools/testing/selftests/bpf/prog_tests/fexit_bpf2bpf.c
index b426bf2f97e4..7d3740d38965 100644
--- a/tools/testing/selftests/bpf/prog_tests/fexit_bpf2bpf.c
+++ b/tools/testing/selftests/bpf/prog_tests/fexit_bpf2bpf.c
@@ -98,6 +98,7 @@ static void test_target_yes_callees(void)
 		"fexit/test_pkt_access",
 		"fexit/test_pkt_access_subprog1",
 		"fexit/test_pkt_access_subprog2",
+		"fexit/test_pkt_access_subprog3",
 	};
 	test_fexit_bpf2bpf_common("./fexit_bpf2bpf.o",
 				  "./test_pkt_access.o",
diff --git a/tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c b/tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c
index 2d211ee98a1c..81d7b4aaf79e 100644
--- a/tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c
+++ b/tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c
@@ -79,4 +79,19 @@ int test_subprog2(struct args_subprog2 *ctx)
 	test_result_subprog2 = 1;
 	return 0;
 }
+
+__u64 test_result_subprog3 = 0;
+BPF_TRACE_3("fexit/test_pkt_access_subprog3", test_subprog3,
+	    int, val, struct sk_buff *, skb, int, ret)
+{
+	int len;
+
+	__builtin_preserve_access_index(({
+		len = skb->len;
+	}));
+	if (len != 74 || ret != 74 * val || val != 3)
+		return 0;
+	test_result_subprog3 = 1;
+	return 0;
+}
 char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/test_pkt_access.c b/tools/testing/selftests/bpf/progs/test_pkt_access.c
index 3a7b4b607ed3..b77cebf71e66 100644
--- a/tools/testing/selftests/bpf/progs/test_pkt_access.c
+++ b/tools/testing/selftests/bpf/progs/test_pkt_access.c
@@ -47,6 +47,32 @@ int test_pkt_access_subprog2(int val, volatile struct __sk_buff *skb)
 	return skb->len * val;
 }
 
+#define MAX_STACK (512 - 2 * 32)
+
+__attribute__ ((noinline))
+int get_skb_len(struct __sk_buff *skb)
+{
+	volatile char buf[MAX_STACK] = {};
+
+	return skb->len;
+}
+
+int get_skb_ifindex(int, struct __sk_buff *skb, int);
+
+__attribute__ ((noinline))
+int test_pkt_access_subprog3(int val, struct __sk_buff *skb)
+{
+	return get_skb_len(skb) * get_skb_ifindex(val, skb, 1);
+}
+
+__attribute__ ((noinline))
+int get_skb_ifindex(int val, struct __sk_buff *skb, int var)
+{
+	volatile char buf[MAX_STACK] = {};
+
+	return skb->ifindex * val * var;
+}
+
 SEC("classifier/test_pkt_access")
 int test_pkt_access(struct __sk_buff *skb)
 {
@@ -82,6 +108,8 @@ int test_pkt_access(struct __sk_buff *skb)
 		return TC_ACT_SHOT;
 	if (test_pkt_access_subprog2(2, skb) != skb->len * 2)
 		return TC_ACT_SHOT;
+	if (test_pkt_access_subprog3(3, skb) != skb->len * 3 * skb->ifindex)
+		return TC_ACT_SHOT;
 	if (tcp) {
 		if (((void *)(tcp) + 20) > data_end || proto != 6)
 			return TC_ACT_SHOT;
-- 
cgit v1.2.3-59-g8ed1b


From 6db2d81a46f99edb5690cf0b03bbe3b7f29dfd7b Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Thu, 9 Jan 2020 22:41:22 -0800
Subject: selftests/bpf: Add a test for a large global function

test results:
pyperf50 with always_inlined the same function five times: processed 46378 insns
pyperf50 with global function: processed 6102 insns

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Song Liu <songliubraving@fb.com>
Link: https://lore.kernel.org/bpf/20200110064124.1760511-5-ast@kernel.org
---
 tools/testing/selftests/bpf/prog_tests/bpf_verif_scale.c | 2 ++
 tools/testing/selftests/bpf/progs/pyperf.h               | 9 +++++++--
 tools/testing/selftests/bpf/progs/pyperf_global.c        | 5 +++++
 3 files changed, 14 insertions(+), 2 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/pyperf_global.c

diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_verif_scale.c b/tools/testing/selftests/bpf/prog_tests/bpf_verif_scale.c
index 9486c13af6b2..e9f2f12ba06b 100644
--- a/tools/testing/selftests/bpf/prog_tests/bpf_verif_scale.c
+++ b/tools/testing/selftests/bpf/prog_tests/bpf_verif_scale.c
@@ -48,6 +48,8 @@ void test_bpf_verif_scale(void)
 		{ "test_verif_scale2.o", BPF_PROG_TYPE_SCHED_CLS },
 		{ "test_verif_scale3.o", BPF_PROG_TYPE_SCHED_CLS },
 
+		{ "pyperf_global.o", BPF_PROG_TYPE_RAW_TRACEPOINT },
+
 		/* full unroll by llvm */
 		{ "pyperf50.o", BPF_PROG_TYPE_RAW_TRACEPOINT },
 		{ "pyperf100.o", BPF_PROG_TYPE_RAW_TRACEPOINT },
diff --git a/tools/testing/selftests/bpf/progs/pyperf.h b/tools/testing/selftests/bpf/progs/pyperf.h
index 71d383cc9b85..e186899954e9 100644
--- a/tools/testing/selftests/bpf/progs/pyperf.h
+++ b/tools/testing/selftests/bpf/progs/pyperf.h
@@ -154,7 +154,12 @@ struct {
 	__uint(value_size, sizeof(long long) * 127);
 } stackmap SEC(".maps");
 
-static __always_inline int __on_event(struct pt_regs *ctx)
+#ifdef GLOBAL_FUNC
+__attribute__((noinline))
+#else
+static __always_inline
+#endif
+int __on_event(struct bpf_raw_tracepoint_args *ctx)
 {
 	uint64_t pid_tgid = bpf_get_current_pid_tgid();
 	pid_t pid = (pid_t)(pid_tgid >> 32);
@@ -254,7 +259,7 @@ static __always_inline int __on_event(struct pt_regs *ctx)
 }
 
 SEC("raw_tracepoint/kfree_skb")
-int on_event(struct pt_regs* ctx)
+int on_event(struct bpf_raw_tracepoint_args* ctx)
 {
 	int i, ret = 0;
 	ret |= __on_event(ctx);
diff --git a/tools/testing/selftests/bpf/progs/pyperf_global.c b/tools/testing/selftests/bpf/progs/pyperf_global.c
new file mode 100644
index 000000000000..079e78a7562b
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/pyperf_global.c
@@ -0,0 +1,5 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2020 Facebook */
+#define STACK_MAX_LEN 50
+#define GLOBAL_FUNC
+#include "pyperf.h"
-- 
cgit v1.2.3-59-g8ed1b


From e528d1c0127accbc2be48f57dd67ab89b32fd815 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Thu, 9 Jan 2020 22:41:23 -0800
Subject: selftests/bpf: Modify a test to check global functions

Make two static functions in test_xdp_noinline.c global:
before: processed 2790 insns
after: processed 2598 insns

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Song Liu <songliubraving@fb.com>
Link: https://lore.kernel.org/bpf/20200110064124.1760511-6-ast@kernel.org
---
 tools/testing/selftests/bpf/progs/test_xdp_noinline.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/test_xdp_noinline.c b/tools/testing/selftests/bpf/progs/test_xdp_noinline.c
index e88d7b9d65ab..f95bc1a17667 100644
--- a/tools/testing/selftests/bpf/progs/test_xdp_noinline.c
+++ b/tools/testing/selftests/bpf/progs/test_xdp_noinline.c
@@ -86,7 +86,7 @@ u32 jhash(const void *key, u32 length, u32 initval)
 	return c;
 }
 
-static __attribute__ ((noinline))
+__attribute__ ((noinline))
 u32 __jhash_nwords(u32 a, u32 b, u32 c, u32 initval)
 {
 	a += initval;
@@ -96,7 +96,7 @@ u32 __jhash_nwords(u32 a, u32 b, u32 c, u32 initval)
 	return c;
 }
 
-static __attribute__ ((noinline))
+__attribute__ ((noinline))
 u32 jhash_2words(u32 a, u32 b, u32 initval)
 {
 	return __jhash_nwords(a, b, 0, initval + JHASH_INITVAL + (2 << 2));
-- 
cgit v1.2.3-59-g8ed1b


From 360301a6c21be87fe881546bd5f22eccf7a165c5 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Thu, 9 Jan 2020 22:41:24 -0800
Subject: selftests/bpf: Add unit tests for global functions

test_global_func[12] - check 512 stack limit.
test_global_func[34] - check 8 frame call chain limit.
test_global_func5    - check that non-ctx pointer cannot be passed into
                       a function that expects context.
test_global_func6    - check that ctx pointer is unmodified.
test_global_func7    - check that global function returns scalar.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Song Liu <songliubraving@fb.com>
Link: https://lore.kernel.org/bpf/20200110064124.1760511-7-ast@kernel.org
---
 .../selftests/bpf/prog_tests/test_global_funcs.c   | 82 ++++++++++++++++++++++
 .../selftests/bpf/progs/test_global_func1.c        | 45 ++++++++++++
 .../selftests/bpf/progs/test_global_func2.c        |  4 ++
 .../selftests/bpf/progs/test_global_func3.c        | 65 +++++++++++++++++
 .../selftests/bpf/progs/test_global_func4.c        |  4 ++
 .../selftests/bpf/progs/test_global_func5.c        | 31 ++++++++
 .../selftests/bpf/progs/test_global_func6.c        | 31 ++++++++
 .../selftests/bpf/progs/test_global_func7.c        | 18 +++++
 8 files changed, 280 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/test_global_funcs.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_global_func1.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_global_func2.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_global_func3.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_global_func4.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_global_func5.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_global_func6.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_global_func7.c

diff --git a/tools/testing/selftests/bpf/prog_tests/test_global_funcs.c b/tools/testing/selftests/bpf/prog_tests/test_global_funcs.c
new file mode 100644
index 000000000000..25b068591e9a
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/test_global_funcs.c
@@ -0,0 +1,82 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2020 Facebook */
+#include <test_progs.h>
+
+const char *err_str;
+bool found;
+
+static int libbpf_debug_print(enum libbpf_print_level level,
+			      const char *format, va_list args)
+{
+	char *log_buf;
+
+	if (level != LIBBPF_WARN ||
+	    strcmp(format, "libbpf: \n%s\n")) {
+		vprintf(format, args);
+		return 0;
+	}
+
+	log_buf = va_arg(args, char *);
+	if (!log_buf)
+		goto out;
+	if (strstr(log_buf, err_str) == 0)
+		found = true;
+out:
+	printf(format, log_buf);
+	return 0;
+}
+
+extern int extra_prog_load_log_flags;
+
+static int check_load(const char *file)
+{
+	struct bpf_prog_load_attr attr;
+	struct bpf_object *obj = NULL;
+	int err, prog_fd;
+
+	memset(&attr, 0, sizeof(struct bpf_prog_load_attr));
+	attr.file = file;
+	attr.prog_type = BPF_PROG_TYPE_UNSPEC;
+	attr.log_level = extra_prog_load_log_flags;
+	attr.prog_flags = BPF_F_TEST_RND_HI32;
+	found = false;
+	err = bpf_prog_load_xattr(&attr, &obj, &prog_fd);
+	bpf_object__close(obj);
+	return err;
+}
+
+struct test_def {
+	const char *file;
+	const char *err_str;
+};
+
+void test_test_global_funcs(void)
+{
+	struct test_def tests[] = {
+		{ "test_global_func1.o", "combined stack size of 4 calls is 544" },
+		{ "test_global_func2.o" },
+		{ "test_global_func3.o" , "the call stack of 8 frames" },
+		{ "test_global_func4.o" },
+		{ "test_global_func5.o" , "expected pointer to ctx, but got PTR" },
+		{ "test_global_func6.o" , "modified ctx ptr R2" },
+		{ "test_global_func7.o" , "foo() doesn't return scalar" },
+	};
+	libbpf_print_fn_t old_print_fn = NULL;
+	int err, i, duration = 0;
+
+	old_print_fn = libbpf_set_print(libbpf_debug_print);
+
+	for (i = 0; i < ARRAY_SIZE(tests); i++) {
+		const struct test_def *test = &tests[i];
+
+		if (!test__start_subtest(test->file))
+			continue;
+
+		err_str = test->err_str;
+		err = check_load(test->file);
+		CHECK_FAIL(!!err ^ !!err_str);
+		if (err_str)
+			CHECK(found, "", "expected string '%s'", err_str);
+	}
+	libbpf_set_print(old_print_fn);
+}
diff --git a/tools/testing/selftests/bpf/progs/test_global_func1.c b/tools/testing/selftests/bpf/progs/test_global_func1.c
new file mode 100644
index 000000000000..97d57d6e244e
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_global_func1.c
@@ -0,0 +1,45 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright (c) 2020 Facebook */
+#include <stddef.h>
+#include <linux/bpf.h>
+#include "bpf_helpers.h"
+
+#ifndef MAX_STACK
+#define MAX_STACK (512 - 3 * 32 + 8)
+#endif
+
+static __attribute__ ((noinline))
+int f0(int var, struct __sk_buff *skb)
+{
+	return skb->len;
+}
+
+__attribute__ ((noinline))
+int f1(struct __sk_buff *skb)
+{
+	volatile char buf[MAX_STACK] = {};
+
+	return f0(0, skb) + skb->len;
+}
+
+int f3(int, struct __sk_buff *skb, int);
+
+__attribute__ ((noinline))
+int f2(int val, struct __sk_buff *skb)
+{
+	return f1(skb) + f3(val, skb, 1);
+}
+
+__attribute__ ((noinline))
+int f3(int val, struct __sk_buff *skb, int var)
+{
+	volatile char buf[MAX_STACK] = {};
+
+	return skb->ifindex * val * var;
+}
+
+SEC("classifier/test")
+int test_cls(struct __sk_buff *skb)
+{
+	return f0(1, skb) + f1(skb) + f2(2, skb) + f3(3, skb, 4);
+}
diff --git a/tools/testing/selftests/bpf/progs/test_global_func2.c b/tools/testing/selftests/bpf/progs/test_global_func2.c
new file mode 100644
index 000000000000..2c18d82923a2
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_global_func2.c
@@ -0,0 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright (c) 2020 Facebook */
+#define MAX_STACK (512 - 3 * 32)
+#include "test_global_func1.c"
diff --git a/tools/testing/selftests/bpf/progs/test_global_func3.c b/tools/testing/selftests/bpf/progs/test_global_func3.c
new file mode 100644
index 000000000000..514ecf9f51b0
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_global_func3.c
@@ -0,0 +1,65 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright (c) 2020 Facebook */
+#include <stddef.h>
+#include <linux/bpf.h>
+#include "bpf_helpers.h"
+
+__attribute__ ((noinline))
+int f1(struct __sk_buff *skb)
+{
+	return skb->len;
+}
+
+__attribute__ ((noinline))
+int f2(int val, struct __sk_buff *skb)
+{
+	return f1(skb) + val;
+}
+
+__attribute__ ((noinline))
+int f3(int val, struct __sk_buff *skb, int var)
+{
+	return f2(var, skb) + val;
+}
+
+__attribute__ ((noinline))
+int f4(struct __sk_buff *skb)
+{
+	return f3(1, skb, 2);
+}
+
+__attribute__ ((noinline))
+int f5(struct __sk_buff *skb)
+{
+	return f4(skb);
+}
+
+__attribute__ ((noinline))
+int f6(struct __sk_buff *skb)
+{
+	return f5(skb);
+}
+
+__attribute__ ((noinline))
+int f7(struct __sk_buff *skb)
+{
+	return f6(skb);
+}
+
+#ifndef NO_FN8
+__attribute__ ((noinline))
+int f8(struct __sk_buff *skb)
+{
+	return f7(skb);
+}
+#endif
+
+SEC("classifier/test")
+int test_cls(struct __sk_buff *skb)
+{
+#ifndef NO_FN8
+	return f8(skb);
+#else
+	return f7(skb);
+#endif
+}
diff --git a/tools/testing/selftests/bpf/progs/test_global_func4.c b/tools/testing/selftests/bpf/progs/test_global_func4.c
new file mode 100644
index 000000000000..610f75edf276
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_global_func4.c
@@ -0,0 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright (c) 2020 Facebook */
+#define NO_FN8
+#include "test_global_func3.c"
diff --git a/tools/testing/selftests/bpf/progs/test_global_func5.c b/tools/testing/selftests/bpf/progs/test_global_func5.c
new file mode 100644
index 000000000000..86787c03cea8
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_global_func5.c
@@ -0,0 +1,31 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright (c) 2020 Facebook */
+#include <stddef.h>
+#include <linux/bpf.h>
+#include "bpf_helpers.h"
+
+__attribute__ ((noinline))
+int f1(struct __sk_buff *skb)
+{
+	return skb->len;
+}
+
+int f3(int, struct __sk_buff *skb);
+
+__attribute__ ((noinline))
+int f2(int val, struct __sk_buff *skb)
+{
+	return f1(skb) + f3(val, (void *)&val); /* type mismatch */
+}
+
+__attribute__ ((noinline))
+int f3(int val, struct __sk_buff *skb)
+{
+	return skb->ifindex * val;
+}
+
+SEC("classifier/test")
+int test_cls(struct __sk_buff *skb)
+{
+	return f1(skb) + f2(2, skb) + f3(3, skb);
+}
diff --git a/tools/testing/selftests/bpf/progs/test_global_func6.c b/tools/testing/selftests/bpf/progs/test_global_func6.c
new file mode 100644
index 000000000000..e215fb3e6f02
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_global_func6.c
@@ -0,0 +1,31 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright (c) 2020 Facebook */
+#include <stddef.h>
+#include <linux/bpf.h>
+#include "bpf_helpers.h"
+
+__attribute__ ((noinline))
+int f1(struct __sk_buff *skb)
+{
+	return skb->len;
+}
+
+int f3(int, struct __sk_buff *skb);
+
+__attribute__ ((noinline))
+int f2(int val, struct __sk_buff *skb)
+{
+	return f1(skb) + f3(val, skb + 1); /* type mismatch */
+}
+
+__attribute__ ((noinline))
+int f3(int val, struct __sk_buff *skb)
+{
+	return skb->ifindex * val;
+}
+
+SEC("classifier/test")
+int test_cls(struct __sk_buff *skb)
+{
+	return f1(skb) + f2(2, skb) + f3(3, skb);
+}
diff --git a/tools/testing/selftests/bpf/progs/test_global_func7.c b/tools/testing/selftests/bpf/progs/test_global_func7.c
new file mode 100644
index 000000000000..ff98d93916fd
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_global_func7.c
@@ -0,0 +1,18 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright (c) 2020 Facebook */
+#include <stddef.h>
+#include <linux/bpf.h>
+#include "bpf_helpers.h"
+
+__attribute__ ((noinline))
+void foo(struct __sk_buff *skb)
+{
+	skb->tc_index = 0;
+}
+
+SEC("classifier/test")
+int test_cls(struct __sk_buff *skb)
+{
+	foo(skb);
+	return 0;
+}
-- 
cgit v1.2.3-59-g8ed1b


From 90fbca5952436e7817910b33eb4464ddd77a8964 Mon Sep 17 00:00:00 2001
From: Yishai Hadas <yishaih@mellanox.com>
Date: Thu, 12 Dec 2019 13:09:24 +0200
Subject: net/mlx5: Add Virtio Emulation related device capabilities

Add Virtio Emulation related fields to the device capabilities.

It includes a general bit to indicate whether Virtio Emulation is
supported and the capabilities structure itself.

Signed-off-by: Yishai Hadas <yishaih@mellanox.com>
Reviewed-by: Shahaf Shuler <shahafs@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
---
 include/linux/mlx5/mlx5_ifc.h | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 5d54fccf87fc..c6abaf4f1c55 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -87,6 +87,7 @@ enum {
 enum {
 	MLX5_GENERAL_OBJ_TYPES_CAP_SW_ICM = (1ULL << MLX5_OBJ_TYPE_SW_ICM),
 	MLX5_GENERAL_OBJ_TYPES_CAP_GENEVE_TLV_OPT = (1ULL << 11),
+	MLX5_GENERAL_OBJ_TYPES_CAP_VIRTIO_NET_Q = (1ULL << 13),
 };
 
 enum {
@@ -953,6 +954,19 @@ struct mlx5_ifc_device_event_cap_bits {
 	u8         user_unaffiliated_events[4][0x40];
 };
 
+struct mlx5_ifc_device_virtio_emulation_cap_bits {
+	u8         reserved_at_0[0x20];
+
+	u8         reserved_at_20[0x13];
+	u8         log_doorbell_stride[0x5];
+	u8         reserved_at_38[0x3];
+	u8         log_doorbell_bar_size[0x5];
+
+	u8         doorbell_bar_offset[0x40];
+
+	u8         reserved_at_80[0x780];
+};
+
 enum {
 	MLX5_ATOMIC_CAPS_ATOMIC_SIZE_QP_1_BYTE     = 0x0,
 	MLX5_ATOMIC_CAPS_ATOMIC_SIZE_QP_2_BYTES    = 0x2,
@@ -2751,6 +2765,7 @@ union mlx5_ifc_hca_cap_union_bits {
 	struct mlx5_ifc_fpga_cap_bits fpga_cap;
 	struct mlx5_ifc_tls_cap_bits tls_cap;
 	struct mlx5_ifc_device_mem_cap_bits device_mem_cap;
+	struct mlx5_ifc_device_virtio_emulation_cap_bits virtio_emulation_cap;
 	u8         reserved_at_0[0x8000];
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From ca1992c62cadb6c8e1e1b47e197b550f3cd89b76 Mon Sep 17 00:00:00 2001
From: Yishai Hadas <yishaih@mellanox.com>
Date: Thu, 12 Dec 2019 13:09:25 +0200
Subject: net/mlx5: Expose vDPA emulation device capabilities

Expose vDPA emulation device capabilities from the core layer.
It includes reading the capabilities from the firmware and exposing
helper functions to access the data.

Signed-off-by: Yishai Hadas <yishaih@mellanox.com>
Reviewed-by: Shahaf Shuler <shahafs@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/fw.c | 7 +++++++
 include/linux/mlx5/device.h                  | 9 +++++++++
 2 files changed, 16 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw.c b/drivers/net/ethernet/mellanox/mlx5/core/fw.c
index a19790dee7b2..c375edfe528c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fw.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fw.c
@@ -245,6 +245,13 @@ int mlx5_query_hca_caps(struct mlx5_core_dev *dev)
 			return err;
 	}
 
+	if (MLX5_CAP_GEN_64(dev, general_obj_types) &
+		MLX5_GENERAL_OBJ_TYPES_CAP_VIRTIO_NET_Q) {
+		err = mlx5_core_get_caps(dev, MLX5_CAP_VDPA_EMULATION);
+		if (err)
+			return err;
+	}
+
 	return 0;
 }
 
diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h
index cc1c230f10ee..1a1c53f0262d 100644
--- a/include/linux/mlx5/device.h
+++ b/include/linux/mlx5/device.h
@@ -1105,6 +1105,7 @@ enum mlx5_cap_type {
 	MLX5_CAP_DEV_MEM,
 	MLX5_CAP_RESERVED_16,
 	MLX5_CAP_TLS,
+	MLX5_CAP_VDPA_EMULATION = 0x13,
 	MLX5_CAP_DEV_EVENT = 0x14,
 	/* NUM OF CAP Types */
 	MLX5_CAP_NUM
@@ -1297,6 +1298,14 @@ enum mlx5_qcam_feature_groups {
 #define MLX5_CAP_DEV_EVENT(mdev, cap)\
 	MLX5_ADDR_OF(device_event_cap, (mdev)->caps.hca_cur[MLX5_CAP_DEV_EVENT], cap)
 
+#define MLX5_CAP_DEV_VDPA_EMULATION(mdev, cap)\
+	MLX5_GET(device_virtio_emulation_cap, \
+		(mdev)->caps.hca_cur[MLX5_CAP_VDPA_EMULATION], cap)
+
+#define MLX5_CAP64_DEV_VDPA_EMULATION(mdev, cap)\
+	MLX5_GET64(device_virtio_emulation_cap, \
+		(mdev)->caps.hca_cur[MLX5_CAP_VDPA_EMULATION], cap)
+
 enum {
 	MLX5_CMD_STAT_OK			= 0x0,
 	MLX5_CMD_STAT_INT_ERR			= 0x1,
-- 
cgit v1.2.3-59-g8ed1b


From 1d1a3bcffe360a56fd8cc287ed74d4c3066daf42 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Fri, 10 Jan 2020 10:19:16 -0800
Subject: libbpf: Poison kernel-only integer types

It's been a recurring issue with types like u32 slipping into libbpf source
code accidentally. This is not detected during builds inside kernel source
tree, but becomes a compilation error in libbpf's Github repo. Libbpf is
supposed to use only __{s,u}{8,16,32,64} typedefs, so poison {s,u}{8,16,32,64}
explicitly in every .c file. Doing that in a bit more centralized way, e.g.,
inside libbpf_internal.h breaks selftests, which are both using kernel u32 and
libbpf_internal.h.

This patch also fixes a new u32 occurence in libbpf.c, added recently.

Fixes: 590a00888250 ("bpf: libbpf: Add STRUCT_OPS support")
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20200110181916.271446-1-andriin@fb.com
---
 tools/lib/bpf/bpf.c            | 3 +++
 tools/lib/bpf/bpf_prog_linfo.c | 3 +++
 tools/lib/bpf/btf.c            | 3 +++
 tools/lib/bpf/btf_dump.c       | 3 +++
 tools/lib/bpf/hashmap.c        | 3 +++
 tools/lib/bpf/libbpf.c         | 5 ++++-
 tools/lib/bpf/libbpf_errno.c   | 3 +++
 tools/lib/bpf/libbpf_probes.c  | 3 +++
 tools/lib/bpf/netlink.c        | 3 +++
 tools/lib/bpf/nlattr.c         | 3 +++
 tools/lib/bpf/str_error.c      | 3 +++
 tools/lib/bpf/xsk.c            | 3 +++
 12 files changed, 37 insertions(+), 1 deletion(-)

diff --git a/tools/lib/bpf/bpf.c b/tools/lib/bpf/bpf.c
index b0ecbe9ef2d4..500afe478e94 100644
--- a/tools/lib/bpf/bpf.c
+++ b/tools/lib/bpf/bpf.c
@@ -32,6 +32,9 @@
 #include "libbpf.h"
 #include "libbpf_internal.h"
 
+/* make sure libbpf doesn't use kernel-only integer typedefs */
+#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
+
 /*
  * When building perf, unistd.h is overridden. __NR_bpf is
  * required to be defined explicitly.
diff --git a/tools/lib/bpf/bpf_prog_linfo.c b/tools/lib/bpf/bpf_prog_linfo.c
index 3ed1a27b5f7c..bafca49cb1e6 100644
--- a/tools/lib/bpf/bpf_prog_linfo.c
+++ b/tools/lib/bpf/bpf_prog_linfo.c
@@ -8,6 +8,9 @@
 #include "libbpf.h"
 #include "libbpf_internal.h"
 
+/* make sure libbpf doesn't use kernel-only integer typedefs */
+#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
+
 struct bpf_prog_linfo {
 	void *raw_linfo;
 	void *raw_jited_linfo;
diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
index 5f04f56e1eb6..cfeb6a44480b 100644
--- a/tools/lib/bpf/btf.c
+++ b/tools/lib/bpf/btf.c
@@ -17,6 +17,9 @@
 #include "libbpf_internal.h"
 #include "hashmap.h"
 
+/* make sure libbpf doesn't use kernel-only integer typedefs */
+#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
+
 #define BTF_MAX_NR_TYPES 0x7fffffff
 #define BTF_MAX_STR_OFFSET 0x7fffffff
 
diff --git a/tools/lib/bpf/btf_dump.c b/tools/lib/bpf/btf_dump.c
index e95f7710f210..885acebd4396 100644
--- a/tools/lib/bpf/btf_dump.c
+++ b/tools/lib/bpf/btf_dump.c
@@ -18,6 +18,9 @@
 #include "libbpf.h"
 #include "libbpf_internal.h"
 
+/* make sure libbpf doesn't use kernel-only integer typedefs */
+#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
+
 static const char PREFIXES[] = "\t\t\t\t\t\t\t\t\t\t\t\t\t";
 static const size_t PREFIX_CNT = sizeof(PREFIXES) - 1;
 
diff --git a/tools/lib/bpf/hashmap.c b/tools/lib/bpf/hashmap.c
index 6122272943e6..54c30c802070 100644
--- a/tools/lib/bpf/hashmap.c
+++ b/tools/lib/bpf/hashmap.c
@@ -12,6 +12,9 @@
 #include <linux/err.h>
 #include "hashmap.h"
 
+/* make sure libbpf doesn't use kernel-only integer typedefs */
+#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
+
 /* start with 4 buckets */
 #define HASHMAP_MIN_CAP_BITS 2
 
diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 3afd780b0f06..0c229f00a67e 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -55,6 +55,9 @@
 #include "libbpf_internal.h"
 #include "hashmap.h"
 
+/* make sure libbpf doesn't use kernel-only integer typedefs */
+#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
+
 #ifndef EM_BPF
 #define EM_BPF 247
 #endif
@@ -6475,7 +6478,7 @@ static int bpf_object__collect_struct_ops_map_reloc(struct bpf_object *obj,
 	Elf_Data *symbols;
 	unsigned int moff;
 	const char *name;
-	u32 member_idx;
+	__u32 member_idx;
 	GElf_Sym sym;
 	GElf_Rel rel;
 	int i, nrels;
diff --git a/tools/lib/bpf/libbpf_errno.c b/tools/lib/bpf/libbpf_errno.c
index 4343e40588c6..0afb51f7a919 100644
--- a/tools/lib/bpf/libbpf_errno.c
+++ b/tools/lib/bpf/libbpf_errno.c
@@ -13,6 +13,9 @@
 
 #include "libbpf.h"
 
+/* make sure libbpf doesn't use kernel-only integer typedefs */
+#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
+
 #define ERRNO_OFFSET(e)		((e) - __LIBBPF_ERRNO__START)
 #define ERRCODE_OFFSET(c)	ERRNO_OFFSET(LIBBPF_ERRNO__##c)
 #define NR_ERRNO	(__LIBBPF_ERRNO__END - __LIBBPF_ERRNO__START)
diff --git a/tools/lib/bpf/libbpf_probes.c b/tools/lib/bpf/libbpf_probes.c
index 320697f8e4c7..8cc992bc532a 100644
--- a/tools/lib/bpf/libbpf_probes.c
+++ b/tools/lib/bpf/libbpf_probes.c
@@ -17,6 +17,9 @@
 #include "libbpf.h"
 #include "libbpf_internal.h"
 
+/* make sure libbpf doesn't use kernel-only integer typedefs */
+#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
+
 static bool grep(const char *buffer, const char *pattern)
 {
 	return !!strstr(buffer, pattern);
diff --git a/tools/lib/bpf/netlink.c b/tools/lib/bpf/netlink.c
index 5065c1aa1061..431bd25c6cdb 100644
--- a/tools/lib/bpf/netlink.c
+++ b/tools/lib/bpf/netlink.c
@@ -15,6 +15,9 @@
 #include "libbpf_internal.h"
 #include "nlattr.h"
 
+/* make sure libbpf doesn't use kernel-only integer typedefs */
+#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
+
 #ifndef SOL_NETLINK
 #define SOL_NETLINK 270
 #endif
diff --git a/tools/lib/bpf/nlattr.c b/tools/lib/bpf/nlattr.c
index 8db44bbfc66d..0ad41dfea8eb 100644
--- a/tools/lib/bpf/nlattr.c
+++ b/tools/lib/bpf/nlattr.c
@@ -13,6 +13,9 @@
 #include <string.h>
 #include <stdio.h>
 
+/* make sure libbpf doesn't use kernel-only integer typedefs */
+#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
+
 static uint16_t nla_attr_minlen[LIBBPF_NLA_TYPE_MAX+1] = {
 	[LIBBPF_NLA_U8]		= sizeof(uint8_t),
 	[LIBBPF_NLA_U16]	= sizeof(uint16_t),
diff --git a/tools/lib/bpf/str_error.c b/tools/lib/bpf/str_error.c
index b8064eedc177..146da01979c7 100644
--- a/tools/lib/bpf/str_error.c
+++ b/tools/lib/bpf/str_error.c
@@ -4,6 +4,9 @@
 #include <stdio.h>
 #include "str_error.h"
 
+/* make sure libbpf doesn't use kernel-only integer typedefs */
+#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
+
 /*
  * Wrapper to allow for building in non-GNU systems such as Alpine Linux's musl
  * libc, while checking strerror_r() return to avoid having to check this in
diff --git a/tools/lib/bpf/xsk.c b/tools/lib/bpf/xsk.c
index 8e0ffa800a71..9807903f121e 100644
--- a/tools/lib/bpf/xsk.c
+++ b/tools/lib/bpf/xsk.c
@@ -32,6 +32,9 @@
 #include "libbpf_internal.h"
 #include "xsk.h"
 
+/* make sure libbpf doesn't use kernel-only integer typedefs */
+#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
+
 #ifndef SOL_XDP
  #define SOL_XDP 283
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From 86de7ced410f647812c16be26a261be0a591188b Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Fri, 10 Jan 2020 13:26:34 +0000
Subject: sfc: refactor selftest work init code

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/efx_common.c | 2 +-
 drivers/net/ethernet/sfc/selftest.c   | 7 ++++++-
 drivers/net/ethernet/sfc/selftest.h   | 2 +-
 3 files changed, 8 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/sfc/efx_common.c b/drivers/net/ethernet/sfc/efx_common.c
index 9cd7c1550c08..33b523b6d0a8 100644
--- a/drivers/net/ethernet/sfc/efx_common.c
+++ b/drivers/net/ethernet/sfc/efx_common.c
@@ -840,7 +840,7 @@ int efx_init_struct(struct efx_nic *efx,
 #endif
 	INIT_WORK(&efx->reset_work, efx_reset_work);
 	INIT_DELAYED_WORK(&efx->monitor_work, efx_monitor);
-	INIT_DELAYED_WORK(&efx->selftest_work, efx_selftest_async_work);
+	efx_selftest_async_init(efx);
 	efx->pci_dev = pci_dev;
 	efx->msg_enable = debug;
 	efx->state = STATE_UNINIT;
diff --git a/drivers/net/ethernet/sfc/selftest.c b/drivers/net/ethernet/sfc/selftest.c
index ec5002b5ec4f..1ae369022d7d 100644
--- a/drivers/net/ethernet/sfc/selftest.c
+++ b/drivers/net/ethernet/sfc/selftest.c
@@ -785,7 +785,7 @@ void efx_selftest_async_cancel(struct efx_nic *efx)
 	cancel_delayed_work_sync(&efx->selftest_work);
 }
 
-void efx_selftest_async_work(struct work_struct *data)
+static void efx_selftest_async_work(struct work_struct *data)
 {
 	struct efx_nic *efx = container_of(data, struct efx_nic,
 					   selftest_work.work);
@@ -804,3 +804,8 @@ void efx_selftest_async_work(struct work_struct *data)
 				  channel->channel, cpu);
 	}
 }
+
+void efx_selftest_async_init(struct efx_nic *efx)
+{
+	INIT_DELAYED_WORK(&efx->selftest_work, efx_selftest_async_work);
+}
diff --git a/drivers/net/ethernet/sfc/selftest.h b/drivers/net/ethernet/sfc/selftest.h
index a3553816d92c..ca88ebb4f6b1 100644
--- a/drivers/net/ethernet/sfc/selftest.h
+++ b/drivers/net/ethernet/sfc/selftest.h
@@ -45,8 +45,8 @@ void efx_loopback_rx_packet(struct efx_nic *efx, const char *buf_ptr,
 			    int pkt_len);
 int efx_selftest(struct efx_nic *efx, struct efx_self_tests *tests,
 		 unsigned flags);
+void efx_selftest_async_init(struct efx_nic *efx);
 void efx_selftest_async_start(struct efx_nic *efx);
 void efx_selftest_async_cancel(struct efx_nic *efx);
-void efx_selftest_async_work(struct work_struct *data);
 
 #endif /* EFX_SELFTEST_H */
-- 
cgit v1.2.3-59-g8ed1b


From b8cd94992f1758d1574f5871335fdaaf726c0944 Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Fri, 10 Jan 2020 13:26:46 +0000
Subject: sfc: move more tx code

The code that handles transmission finalization will also be common.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/tx.c        | 77 ------------------------------------
 drivers/net/ethernet/sfc/tx_common.c | 76 +++++++++++++++++++++++++++++++++++
 drivers/net/ethernet/sfc/tx_common.h |  2 +
 3 files changed, 78 insertions(+), 77 deletions(-)

diff --git a/drivers/net/ethernet/sfc/tx.c b/drivers/net/ethernet/sfc/tx.c
index e4f0f5c94589..696e12c1d9e7 100644
--- a/drivers/net/ethernet/sfc/tx.c
+++ b/drivers/net/ethernet/sfc/tx.c
@@ -518,41 +518,6 @@ int efx_xdp_tx_buffers(struct efx_nic *efx, int n, struct xdp_frame **xdpfs,
 	return i;
 }
 
-/* Remove packets from the TX queue
- *
- * This removes packets from the TX queue, up to and including the
- * specified index.
- */
-static void efx_dequeue_buffers(struct efx_tx_queue *tx_queue,
-				unsigned int index,
-				unsigned int *pkts_compl,
-				unsigned int *bytes_compl)
-{
-	struct efx_nic *efx = tx_queue->efx;
-	unsigned int stop_index, read_ptr;
-
-	stop_index = (index + 1) & tx_queue->ptr_mask;
-	read_ptr = tx_queue->read_count & tx_queue->ptr_mask;
-
-	while (read_ptr != stop_index) {
-		struct efx_tx_buffer *buffer = &tx_queue->buffer[read_ptr];
-
-		if (!(buffer->flags & EFX_TX_BUF_OPTION) &&
-		    unlikely(buffer->len == 0)) {
-			netif_err(efx, tx_err, efx->net_dev,
-				  "TX queue %d spurious TX completion id %x\n",
-				  tx_queue->queue, read_ptr);
-			efx_schedule_reset(efx, RESET_TYPE_TX_SKIP);
-			return;
-		}
-
-		efx_dequeue_buffer(tx_queue, buffer, pkts_compl, bytes_compl);
-
-		++tx_queue->read_count;
-		read_ptr = tx_queue->read_count & tx_queue->ptr_mask;
-	}
-}
-
 /* Initiate a packet transmission.  We use one channel per CPU
  * (sharing when we have more CPUs than channels).  On Falcon, the TX
  * completion events will be directed back to the CPU that transmitted
@@ -665,45 +630,3 @@ int efx_setup_tc(struct net_device *net_dev, enum tc_setup_type type,
 	net_dev->num_tc = num_tc;
 	return 0;
 }
-
-void efx_xmit_done(struct efx_tx_queue *tx_queue, unsigned int index)
-{
-	unsigned fill_level;
-	struct efx_nic *efx = tx_queue->efx;
-	struct efx_tx_queue *txq2;
-	unsigned int pkts_compl = 0, bytes_compl = 0;
-
-	EFX_WARN_ON_ONCE_PARANOID(index > tx_queue->ptr_mask);
-
-	efx_dequeue_buffers(tx_queue, index, &pkts_compl, &bytes_compl);
-	tx_queue->pkts_compl += pkts_compl;
-	tx_queue->bytes_compl += bytes_compl;
-
-	if (pkts_compl > 1)
-		++tx_queue->merge_events;
-
-	/* See if we need to restart the netif queue.  This memory
-	 * barrier ensures that we write read_count (inside
-	 * efx_dequeue_buffers()) before reading the queue status.
-	 */
-	smp_mb();
-	if (unlikely(netif_tx_queue_stopped(tx_queue->core_txq)) &&
-	    likely(efx->port_enabled) &&
-	    likely(netif_device_present(efx->net_dev))) {
-		txq2 = efx_tx_queue_partner(tx_queue);
-		fill_level = max(tx_queue->insert_count - tx_queue->read_count,
-				 txq2->insert_count - txq2->read_count);
-		if (fill_level <= efx->txq_wake_thresh)
-			netif_tx_wake_queue(tx_queue->core_txq);
-	}
-
-	/* Check whether the hardware queue is now empty */
-	if ((int)(tx_queue->read_count - tx_queue->old_write_count) >= 0) {
-		tx_queue->old_write_count = READ_ONCE(tx_queue->write_count);
-		if (tx_queue->read_count == tx_queue->old_write_count) {
-			smp_mb();
-			tx_queue->empty_read_count =
-				tx_queue->read_count | EFX_EMPTY_COUNT_VALID;
-		}
-	}
-}
diff --git a/drivers/net/ethernet/sfc/tx_common.c b/drivers/net/ethernet/sfc/tx_common.c
index e29ade21c4b9..0ce699ecae5a 100644
--- a/drivers/net/ethernet/sfc/tx_common.c
+++ b/drivers/net/ethernet/sfc/tx_common.c
@@ -191,6 +191,82 @@ void efx_dequeue_buffer(struct efx_tx_queue *tx_queue,
 	buffer->flags = 0;
 }
 
+/* Remove packets from the TX queue
+ *
+ * This removes packets from the TX queue, up to and including the
+ * specified index.
+ */
+static void efx_dequeue_buffers(struct efx_tx_queue *tx_queue,
+				unsigned int index,
+				unsigned int *pkts_compl,
+				unsigned int *bytes_compl)
+{
+	struct efx_nic *efx = tx_queue->efx;
+	unsigned int stop_index, read_ptr;
+
+	stop_index = (index + 1) & tx_queue->ptr_mask;
+	read_ptr = tx_queue->read_count & tx_queue->ptr_mask;
+
+	while (read_ptr != stop_index) {
+		struct efx_tx_buffer *buffer = &tx_queue->buffer[read_ptr];
+
+		if (!(buffer->flags & EFX_TX_BUF_OPTION) &&
+		    unlikely(buffer->len == 0)) {
+			netif_err(efx, tx_err, efx->net_dev,
+				  "TX queue %d spurious TX completion id %x\n",
+				  tx_queue->queue, read_ptr);
+			efx_schedule_reset(efx, RESET_TYPE_TX_SKIP);
+			return;
+		}
+
+		efx_dequeue_buffer(tx_queue, buffer, pkts_compl, bytes_compl);
+
+		++tx_queue->read_count;
+		read_ptr = tx_queue->read_count & tx_queue->ptr_mask;
+	}
+}
+
+void efx_xmit_done(struct efx_tx_queue *tx_queue, unsigned int index)
+{
+	unsigned int fill_level, pkts_compl = 0, bytes_compl = 0;
+	struct efx_nic *efx = tx_queue->efx;
+	struct efx_tx_queue *txq2;
+
+	EFX_WARN_ON_ONCE_PARANOID(index > tx_queue->ptr_mask);
+
+	efx_dequeue_buffers(tx_queue, index, &pkts_compl, &bytes_compl);
+	tx_queue->pkts_compl += pkts_compl;
+	tx_queue->bytes_compl += bytes_compl;
+
+	if (pkts_compl > 1)
+		++tx_queue->merge_events;
+
+	/* See if we need to restart the netif queue.  This memory
+	 * barrier ensures that we write read_count (inside
+	 * efx_dequeue_buffers()) before reading the queue status.
+	 */
+	smp_mb();
+	if (unlikely(netif_tx_queue_stopped(tx_queue->core_txq)) &&
+	    likely(efx->port_enabled) &&
+	    likely(netif_device_present(efx->net_dev))) {
+		txq2 = efx_tx_queue_partner(tx_queue);
+		fill_level = max(tx_queue->insert_count - tx_queue->read_count,
+				 txq2->insert_count - txq2->read_count);
+		if (fill_level <= efx->txq_wake_thresh)
+			netif_tx_wake_queue(tx_queue->core_txq);
+	}
+
+	/* Check whether the hardware queue is now empty */
+	if ((int)(tx_queue->read_count - tx_queue->old_write_count) >= 0) {
+		tx_queue->old_write_count = READ_ONCE(tx_queue->write_count);
+		if (tx_queue->read_count == tx_queue->old_write_count) {
+			smp_mb();
+			tx_queue->empty_read_count =
+				tx_queue->read_count | EFX_EMPTY_COUNT_VALID;
+		}
+	}
+}
+
 struct efx_tx_buffer *efx_tx_map_chunk(struct efx_tx_queue *tx_queue,
 				       dma_addr_t dma_addr, size_t len)
 {
diff --git a/drivers/net/ethernet/sfc/tx_common.h b/drivers/net/ethernet/sfc/tx_common.h
index afdfc79a8ea0..58add94ab500 100644
--- a/drivers/net/ethernet/sfc/tx_common.h
+++ b/drivers/net/ethernet/sfc/tx_common.h
@@ -21,6 +21,8 @@ void efx_dequeue_buffer(struct efx_tx_queue *tx_queue,
 			unsigned int *pkts_compl,
 			unsigned int *bytes_compl);
 
+void efx_xmit_done(struct efx_tx_queue *tx_queue, unsigned int index);
+
 struct efx_tx_buffer *efx_tx_map_chunk(struct efx_tx_queue *tx_queue,
 				       dma_addr_t dma_addr, size_t len);
 int efx_tx_map_data(struct efx_tx_queue *tx_queue, struct sk_buff *skb,
-- 
cgit v1.2.3-59-g8ed1b


From 3d95b884392f0a050fb9ca3c535981a1a5334616 Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Fri, 10 Jan 2020 13:27:03 +0000
Subject: sfc: move more rx code

Page recycling code and GRO packet receipt code were moved.

One function contains code extracted from another.

Code style fixes included.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/efx.h       |   3 -
 drivers/net/ethernet/sfc/rx.c        | 188 -------------------------------
 drivers/net/ethernet/sfc/rx_common.c | 212 ++++++++++++++++++++++++++++++++---
 drivers/net/ethernet/sfc/rx_common.h |  29 +++++
 4 files changed, 224 insertions(+), 208 deletions(-)

diff --git a/drivers/net/ethernet/sfc/efx.h b/drivers/net/ethernet/sfc/efx.h
index 3920f29b2fed..2b417e779e82 100644
--- a/drivers/net/ethernet/sfc/efx.h
+++ b/drivers/net/ethernet/sfc/efx.h
@@ -37,9 +37,6 @@ static inline void efx_rx_flush_packet(struct efx_channel *channel)
 		__efx_rx_packet(channel);
 }
 
-void efx_init_rx_recycle_ring(struct efx_rx_queue *rx_queue);
-struct page *efx_reuse_page(struct efx_rx_queue *rx_queue);
-
 #define EFX_MAX_DMAQ_SIZE 4096UL
 #define EFX_DEFAULT_DMAQ_SIZE 1024UL
 #define EFX_MIN_DMAQ_SIZE 512UL
diff --git a/drivers/net/ethernet/sfc/rx.c b/drivers/net/ethernet/sfc/rx.c
index 0e04ed7f6382..18001195410f 100644
--- a/drivers/net/ethernet/sfc/rx.c
+++ b/drivers/net/ethernet/sfc/rx.c
@@ -33,13 +33,6 @@
 /* Maximum rx prefix used by any architecture. */
 #define EFX_MAX_RX_PREFIX_SIZE 16
 
-/* Number of RX buffers to recycle pages for.  When creating the RX page recycle
- * ring, this number is divided by the number of buffers per page to calculate
- * the number of pages to store in the RX page recycle ring.
- */
-#define EFX_RECYCLE_RING_SIZE_IOMMU 4096
-#define EFX_RECYCLE_RING_SIZE_NOIOMMU (2 * EFX_RX_PREFERRED_BATCH)
-
 /* Size of buffer allocated for skb header area. */
 #define EFX_SKB_HEADERS  128u
 
@@ -47,24 +40,6 @@
 #define EFX_RX_MAX_FRAGS DIV_ROUND_UP(EFX_MAX_FRAME_LEN(EFX_MAX_MTU), \
 				      EFX_RX_USR_BUF_SIZE)
 
-static inline u8 *efx_rx_buf_va(struct efx_rx_buffer *buf)
-{
-	return page_address(buf->page) + buf->page_offset;
-}
-
-static inline u32 efx_rx_buf_hash(struct efx_nic *efx, const u8 *eh)
-{
-#if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS)
-	return __le32_to_cpup((const __le32 *)(eh + efx->rx_packet_hash_offset));
-#else
-	const u8 *data = eh + efx->rx_packet_hash_offset;
-	return (u32)data[0]	  |
-	       (u32)data[1] << 8  |
-	       (u32)data[2] << 16 |
-	       (u32)data[3] << 24;
-#endif
-}
-
 static inline void efx_sync_rx_buffer(struct efx_nic *efx,
 				      struct efx_rx_buffer *rx_buf,
 				      unsigned int len)
@@ -73,100 +48,6 @@ static inline void efx_sync_rx_buffer(struct efx_nic *efx,
 				DMA_FROM_DEVICE);
 }
 
-/* Check the RX page recycle ring for a page that can be reused. */
-struct page *efx_reuse_page(struct efx_rx_queue *rx_queue)
-{
-	struct efx_nic *efx = rx_queue->efx;
-	struct page *page;
-	struct efx_rx_page_state *state;
-	unsigned index;
-
-	index = rx_queue->page_remove & rx_queue->page_ptr_mask;
-	page = rx_queue->page_ring[index];
-	if (page == NULL)
-		return NULL;
-
-	rx_queue->page_ring[index] = NULL;
-	/* page_remove cannot exceed page_add. */
-	if (rx_queue->page_remove != rx_queue->page_add)
-		++rx_queue->page_remove;
-
-	/* If page_count is 1 then we hold the only reference to this page. */
-	if (page_count(page) == 1) {
-		++rx_queue->page_recycle_count;
-		return page;
-	} else {
-		state = page_address(page);
-		dma_unmap_page(&efx->pci_dev->dev, state->dma_addr,
-			       PAGE_SIZE << efx->rx_buffer_order,
-			       DMA_FROM_DEVICE);
-		put_page(page);
-		++rx_queue->page_recycle_failed;
-	}
-
-	return NULL;
-}
-
-/* Attempt to recycle the page if there is an RX recycle ring; the page can
- * only be added if this is the final RX buffer, to prevent pages being used in
- * the descriptor ring and appearing in the recycle ring simultaneously.
- */
-static void efx_recycle_rx_page(struct efx_channel *channel,
-				struct efx_rx_buffer *rx_buf)
-{
-	struct page *page = rx_buf->page;
-	struct efx_rx_queue *rx_queue = efx_channel_get_rx_queue(channel);
-	struct efx_nic *efx = rx_queue->efx;
-	unsigned index;
-
-	/* Only recycle the page after processing the final buffer. */
-	if (!(rx_buf->flags & EFX_RX_BUF_LAST_IN_PAGE))
-		return;
-
-	index = rx_queue->page_add & rx_queue->page_ptr_mask;
-	if (rx_queue->page_ring[index] == NULL) {
-		unsigned read_index = rx_queue->page_remove &
-			rx_queue->page_ptr_mask;
-
-		/* The next slot in the recycle ring is available, but
-		 * increment page_remove if the read pointer currently
-		 * points here.
-		 */
-		if (read_index == index)
-			++rx_queue->page_remove;
-		rx_queue->page_ring[index] = page;
-		++rx_queue->page_add;
-		return;
-	}
-	++rx_queue->page_recycle_full;
-	efx_unmap_rx_buffer(efx, rx_buf);
-	put_page(rx_buf->page);
-}
-
-/* Recycle the pages that are used by buffers that have just been received. */
-static void efx_recycle_rx_pages(struct efx_channel *channel,
-				 struct efx_rx_buffer *rx_buf,
-				 unsigned int n_frags)
-{
-	struct efx_rx_queue *rx_queue = efx_channel_get_rx_queue(channel);
-
-	do {
-		efx_recycle_rx_page(channel, rx_buf);
-		rx_buf = efx_rx_buf_next(rx_queue, rx_buf);
-	} while (--n_frags);
-}
-
-static void efx_discard_rx_packet(struct efx_channel *channel,
-				  struct efx_rx_buffer *rx_buf,
-				  unsigned int n_frags)
-{
-	struct efx_rx_queue *rx_queue = efx_channel_get_rx_queue(channel);
-
-	efx_recycle_rx_pages(channel, rx_buf, n_frags);
-
-	efx_free_rx_buffers(rx_queue, rx_buf, n_frags);
-}
-
 static void efx_rx_packet__check_len(struct efx_rx_queue *rx_queue,
 				     struct efx_rx_buffer *rx_buf,
 				     int len)
@@ -190,53 +71,6 @@ static void efx_rx_packet__check_len(struct efx_rx_queue *rx_queue,
 	efx_rx_queue_channel(rx_queue)->n_rx_overlength++;
 }
 
-/* Pass a received packet up through GRO.  GRO can handle pages
- * regardless of checksum state and skbs with a good checksum.
- */
-static void
-efx_rx_packet_gro(struct efx_channel *channel, struct efx_rx_buffer *rx_buf,
-		  unsigned int n_frags, u8 *eh)
-{
-	struct napi_struct *napi = &channel->napi_str;
-	struct efx_nic *efx = channel->efx;
-	struct sk_buff *skb;
-
-	skb = napi_get_frags(napi);
-	if (unlikely(!skb)) {
-		struct efx_rx_queue *rx_queue;
-
-		rx_queue = efx_channel_get_rx_queue(channel);
-		efx_free_rx_buffers(rx_queue, rx_buf, n_frags);
-		return;
-	}
-
-	if (efx->net_dev->features & NETIF_F_RXHASH)
-		skb_set_hash(skb, efx_rx_buf_hash(efx, eh),
-			     PKT_HASH_TYPE_L3);
-	skb->ip_summed = ((rx_buf->flags & EFX_RX_PKT_CSUMMED) ?
-			  CHECKSUM_UNNECESSARY : CHECKSUM_NONE);
-	skb->csum_level = !!(rx_buf->flags & EFX_RX_PKT_CSUM_LEVEL);
-
-	for (;;) {
-		skb_fill_page_desc(skb, skb_shinfo(skb)->nr_frags,
-				   rx_buf->page, rx_buf->page_offset,
-				   rx_buf->len);
-		rx_buf->page = NULL;
-		skb->len += rx_buf->len;
-		if (skb_shinfo(skb)->nr_frags == n_frags)
-			break;
-
-		rx_buf = efx_rx_buf_next(&channel->rx_queue, rx_buf);
-	}
-
-	skb->data_len = skb->len;
-	skb->truesize += n_frags * efx->rx_buffer_truesize;
-
-	skb_record_rx_queue(skb, channel->rx_queue.core_index);
-
-	napi_gro_frags(napi);
-}
-
 /* Allocate and construct an SKB around page fragments */
 static struct sk_buff *efx_rx_mk_skb(struct efx_channel *channel,
 				     struct efx_rx_buffer *rx_buf,
@@ -583,28 +417,6 @@ out:
 	channel->rx_pkt_n_frags = 0;
 }
 
-void efx_init_rx_recycle_ring(struct efx_rx_queue *rx_queue)
-{
-	unsigned int bufs_in_recycle_ring, page_ring_size;
-	struct efx_nic *efx = rx_queue->efx;
-
-	/* Set the RX recycle ring size */
-#ifdef CONFIG_PPC64
-	bufs_in_recycle_ring = EFX_RECYCLE_RING_SIZE_IOMMU;
-#else
-	if (iommu_present(&pci_bus_type))
-		bufs_in_recycle_ring = EFX_RECYCLE_RING_SIZE_IOMMU;
-	else
-		bufs_in_recycle_ring = EFX_RECYCLE_RING_SIZE_NOIOMMU;
-#endif /* CONFIG_PPC64 */
-
-	page_ring_size = roundup_pow_of_two(bufs_in_recycle_ring /
-					    efx->rx_bufs_per_page);
-	rx_queue->page_ring = kcalloc(page_ring_size,
-				      sizeof(*rx_queue->page_ring), GFP_KERNEL);
-	rx_queue->page_ptr_mask = page_ring_size - 1;
-}
-
 #ifdef CONFIG_RFS_ACCEL
 
 static void efx_filter_rfs_work(struct work_struct *data)
diff --git a/drivers/net/ethernet/sfc/rx_common.c b/drivers/net/ethernet/sfc/rx_common.c
index f4b5c3d828f6..9c545d873d91 100644
--- a/drivers/net/ethernet/sfc/rx_common.c
+++ b/drivers/net/ethernet/sfc/rx_common.c
@@ -10,6 +10,7 @@
 
 #include "net_driver.h"
 #include <linux/module.h>
+#include <linux/iommu.h>
 #include "efx.h"
 #include "nic.h"
 #include "rx_common.h"
@@ -22,6 +23,13 @@ module_param(rx_refill_threshold, uint, 0444);
 MODULE_PARM_DESC(rx_refill_threshold,
 		 "RX descriptor ring refill threshold (%)");
 
+/* Number of RX buffers to recycle pages for.  When creating the RX page recycle
+ * ring, this number is divided by the number of buffers per page to calculate
+ * the number of pages to store in the RX page recycle ring.
+ */
+#define EFX_RECYCLE_RING_SIZE_IOMMU 4096
+#define EFX_RECYCLE_RING_SIZE_NOIOMMU (2 * EFX_RX_PREFERRED_BATCH)
+
 /* RX maximum head room required.
  *
  * This must be at least 1 to prevent overflow, plus one packet-worth
@@ -29,6 +37,145 @@ MODULE_PARM_DESC(rx_refill_threshold,
  */
 #define EFX_RXD_HEAD_ROOM (1 + EFX_RX_MAX_FRAGS)
 
+/* Check the RX page recycle ring for a page that can be reused. */
+static struct page *efx_reuse_page(struct efx_rx_queue *rx_queue)
+{
+	struct efx_nic *efx = rx_queue->efx;
+	struct efx_rx_page_state *state;
+	unsigned int index;
+	struct page *page;
+
+	index = rx_queue->page_remove & rx_queue->page_ptr_mask;
+	page = rx_queue->page_ring[index];
+	if (page == NULL)
+		return NULL;
+
+	rx_queue->page_ring[index] = NULL;
+	/* page_remove cannot exceed page_add. */
+	if (rx_queue->page_remove != rx_queue->page_add)
+		++rx_queue->page_remove;
+
+	/* If page_count is 1 then we hold the only reference to this page. */
+	if (page_count(page) == 1) {
+		++rx_queue->page_recycle_count;
+		return page;
+	} else {
+		state = page_address(page);
+		dma_unmap_page(&efx->pci_dev->dev, state->dma_addr,
+			       PAGE_SIZE << efx->rx_buffer_order,
+			       DMA_FROM_DEVICE);
+		put_page(page);
+		++rx_queue->page_recycle_failed;
+	}
+
+	return NULL;
+}
+
+/* Attempt to recycle the page if there is an RX recycle ring; the page can
+ * only be added if this is the final RX buffer, to prevent pages being used in
+ * the descriptor ring and appearing in the recycle ring simultaneously.
+ */
+static void efx_recycle_rx_page(struct efx_channel *channel,
+				struct efx_rx_buffer *rx_buf)
+{
+	struct efx_rx_queue *rx_queue = efx_channel_get_rx_queue(channel);
+	struct efx_nic *efx = rx_queue->efx;
+	struct page *page = rx_buf->page;
+	unsigned int index;
+
+	/* Only recycle the page after processing the final buffer. */
+	if (!(rx_buf->flags & EFX_RX_BUF_LAST_IN_PAGE))
+		return;
+
+	index = rx_queue->page_add & rx_queue->page_ptr_mask;
+	if (rx_queue->page_ring[index] == NULL) {
+		unsigned int read_index = rx_queue->page_remove &
+			rx_queue->page_ptr_mask;
+
+		/* The next slot in the recycle ring is available, but
+		 * increment page_remove if the read pointer currently
+		 * points here.
+		 */
+		if (read_index == index)
+			++rx_queue->page_remove;
+		rx_queue->page_ring[index] = page;
+		++rx_queue->page_add;
+		return;
+	}
+	++rx_queue->page_recycle_full;
+	efx_unmap_rx_buffer(efx, rx_buf);
+	put_page(rx_buf->page);
+}
+
+/* Recycle the pages that are used by buffers that have just been received. */
+void efx_recycle_rx_pages(struct efx_channel *channel,
+			  struct efx_rx_buffer *rx_buf,
+			  unsigned int n_frags)
+{
+	struct efx_rx_queue *rx_queue = efx_channel_get_rx_queue(channel);
+
+	do {
+		efx_recycle_rx_page(channel, rx_buf);
+		rx_buf = efx_rx_buf_next(rx_queue, rx_buf);
+	} while (--n_frags);
+}
+
+void efx_discard_rx_packet(struct efx_channel *channel,
+			   struct efx_rx_buffer *rx_buf,
+			   unsigned int n_frags)
+{
+	struct efx_rx_queue *rx_queue = efx_channel_get_rx_queue(channel);
+
+	efx_recycle_rx_pages(channel, rx_buf, n_frags);
+
+	efx_free_rx_buffers(rx_queue, rx_buf, n_frags);
+}
+
+static void efx_init_rx_recycle_ring(struct efx_rx_queue *rx_queue)
+{
+	unsigned int bufs_in_recycle_ring, page_ring_size;
+	struct efx_nic *efx = rx_queue->efx;
+
+	/* Set the RX recycle ring size */
+#ifdef CONFIG_PPC64
+	bufs_in_recycle_ring = EFX_RECYCLE_RING_SIZE_IOMMU;
+#else
+	if (iommu_present(&pci_bus_type))
+		bufs_in_recycle_ring = EFX_RECYCLE_RING_SIZE_IOMMU;
+	else
+		bufs_in_recycle_ring = EFX_RECYCLE_RING_SIZE_NOIOMMU;
+#endif /* CONFIG_PPC64 */
+
+	page_ring_size = roundup_pow_of_two(bufs_in_recycle_ring /
+					    efx->rx_bufs_per_page);
+	rx_queue->page_ring = kcalloc(page_ring_size,
+				      sizeof(*rx_queue->page_ring), GFP_KERNEL);
+	rx_queue->page_ptr_mask = page_ring_size - 1;
+}
+
+static void efx_fini_rx_recycle_ring(struct efx_rx_queue *rx_queue)
+{
+	struct efx_nic *efx = rx_queue->efx;
+	int i;
+
+	/* Unmap and release the pages in the recycle ring. Remove the ring. */
+	for (i = 0; i <= rx_queue->page_ptr_mask; i++) {
+		struct page *page = rx_queue->page_ring[i];
+		struct efx_rx_page_state *state;
+
+		if (page == NULL)
+			continue;
+
+		state = page_address(page);
+		dma_unmap_page(&efx->pci_dev->dev, state->dma_addr,
+			       PAGE_SIZE << efx->rx_buffer_order,
+			       DMA_FROM_DEVICE);
+		put_page(page);
+	}
+	kfree(rx_queue->page_ring);
+	rx_queue->page_ring = NULL;
+}
+
 static void efx_fini_rx_buffer(struct efx_rx_queue *rx_queue,
 			       struct efx_rx_buffer *rx_buf)
 {
@@ -132,7 +279,6 @@ void efx_init_rx_queue(struct efx_rx_queue *rx_queue)
 
 void efx_fini_rx_queue(struct efx_rx_queue *rx_queue)
 {
-	struct efx_nic *efx = rx_queue->efx;
 	struct efx_rx_buffer *rx_buf;
 	int i;
 
@@ -152,22 +298,7 @@ void efx_fini_rx_queue(struct efx_rx_queue *rx_queue)
 		}
 	}
 
-	/* Unmap and release the pages in the recycle ring. Remove the ring. */
-	for (i = 0; i <= rx_queue->page_ptr_mask; i++) {
-		struct page *page = rx_queue->page_ring[i];
-		struct efx_rx_page_state *state;
-
-		if (page == NULL)
-			continue;
-
-		state = page_address(page);
-		dma_unmap_page(&efx->pci_dev->dev, state->dma_addr,
-			       PAGE_SIZE << efx->rx_buffer_order,
-			       DMA_FROM_DEVICE);
-		put_page(page);
-	}
-	kfree(rx_queue->page_ring);
-	rx_queue->page_ring = NULL;
+	efx_fini_rx_recycle_ring(rx_queue);
 
 	if (rx_queue->xdp_rxq_info_valid)
 		xdp_rxq_info_unreg(&rx_queue->xdp_rxq_info);
@@ -373,3 +504,50 @@ void efx_fast_push_rx_descriptors(struct efx_rx_queue *rx_queue, bool atomic)
 	if (rx_queue->notified_count != rx_queue->added_count)
 		efx_nic_notify_rx_desc(rx_queue);
 }
+
+/* Pass a received packet up through GRO.  GRO can handle pages
+ * regardless of checksum state and skbs with a good checksum.
+ */
+void
+efx_rx_packet_gro(struct efx_channel *channel, struct efx_rx_buffer *rx_buf,
+		  unsigned int n_frags, u8 *eh)
+{
+	struct napi_struct *napi = &channel->napi_str;
+	struct efx_nic *efx = channel->efx;
+	struct sk_buff *skb;
+
+	skb = napi_get_frags(napi);
+	if (unlikely(!skb)) {
+		struct efx_rx_queue *rx_queue;
+
+		rx_queue = efx_channel_get_rx_queue(channel);
+		efx_free_rx_buffers(rx_queue, rx_buf, n_frags);
+		return;
+	}
+
+	if (efx->net_dev->features & NETIF_F_RXHASH)
+		skb_set_hash(skb, efx_rx_buf_hash(efx, eh),
+			     PKT_HASH_TYPE_L3);
+	skb->ip_summed = ((rx_buf->flags & EFX_RX_PKT_CSUMMED) ?
+			  CHECKSUM_UNNECESSARY : CHECKSUM_NONE);
+	skb->csum_level = !!(rx_buf->flags & EFX_RX_PKT_CSUM_LEVEL);
+
+	for (;;) {
+		skb_fill_page_desc(skb, skb_shinfo(skb)->nr_frags,
+				   rx_buf->page, rx_buf->page_offset,
+				   rx_buf->len);
+		rx_buf->page = NULL;
+		skb->len += rx_buf->len;
+		if (skb_shinfo(skb)->nr_frags == n_frags)
+			break;
+
+		rx_buf = efx_rx_buf_next(&channel->rx_queue, rx_buf);
+	}
+
+	skb->data_len = skb->len;
+	skb->truesize += n_frags * efx->rx_buffer_truesize;
+
+	skb_record_rx_queue(skb, channel->rx_queue.core_index);
+
+	napi_gro_frags(napi);
+}
diff --git a/drivers/net/ethernet/sfc/rx_common.h b/drivers/net/ethernet/sfc/rx_common.h
index 8b23a7accea1..aa6aeafbc439 100644
--- a/drivers/net/ethernet/sfc/rx_common.h
+++ b/drivers/net/ethernet/sfc/rx_common.h
@@ -18,8 +18,34 @@
 #define EFX_RX_MAX_FRAGS DIV_ROUND_UP(EFX_MAX_FRAME_LEN(EFX_MAX_MTU), \
 				      EFX_RX_USR_BUF_SIZE)
 
+static inline u8 *efx_rx_buf_va(struct efx_rx_buffer *buf)
+{
+	return page_address(buf->page) + buf->page_offset;
+}
+
+static inline u32 efx_rx_buf_hash(struct efx_nic *efx, const u8 *eh)
+{
+#if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS)
+	return __le32_to_cpup((const __le32 *)(eh + efx->rx_packet_hash_offset));
+#else
+	const u8 *data = eh + efx->rx_packet_hash_offset;
+
+	return (u32)data[0]	  |
+	       (u32)data[1] << 8  |
+	       (u32)data[2] << 16 |
+	       (u32)data[3] << 24;
+#endif
+}
+
 void efx_rx_slow_fill(struct timer_list *t);
 
+void efx_recycle_rx_pages(struct efx_channel *channel,
+			  struct efx_rx_buffer *rx_buf,
+			  unsigned int n_frags);
+void efx_discard_rx_packet(struct efx_channel *channel,
+			   struct efx_rx_buffer *rx_buf,
+			   unsigned int n_frags);
+
 int efx_probe_rx_queue(struct efx_rx_queue *rx_queue);
 void efx_init_rx_queue(struct efx_rx_queue *rx_queue);
 void efx_fini_rx_queue(struct efx_rx_queue *rx_queue);
@@ -39,4 +65,7 @@ void efx_schedule_slow_fill(struct efx_rx_queue *rx_queue);
 void efx_rx_config_page_split(struct efx_nic *efx);
 void efx_fast_push_rx_descriptors(struct efx_rx_queue *rx_queue, bool atomic);
 
+void
+efx_rx_packet_gro(struct efx_channel *channel, struct efx_rx_buffer *rx_buf,
+		  unsigned int n_frags, u8 *eh);
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From 190c736a8088b6dbf62b78192e6a2401eb58177b Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Fri, 10 Jan 2020 13:27:17 +0000
Subject: sfc: move various functions

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/ef10.c             | 21 ++------------
 drivers/net/ethernet/sfc/efx.c              | 11 -------
 drivers/net/ethernet/sfc/efx_common.c       | 10 +++++++
 drivers/net/ethernet/sfc/efx_common.h       |  2 ++
 drivers/net/ethernet/sfc/mcdi.h             |  2 --
 drivers/net/ethernet/sfc/mcdi_functions.c   | 23 +++++++++++++++
 drivers/net/ethernet/sfc/mcdi_functions.h   |  1 +
 drivers/net/ethernet/sfc/mcdi_port.c        | 45 -----------------------------
 drivers/net/ethernet/sfc/mcdi_port_common.c | 45 +++++++++++++++++++++++++++++
 drivers/net/ethernet/sfc/mcdi_port_common.h |  3 ++
 drivers/net/ethernet/sfc/siena.c            |  1 +
 11 files changed, 88 insertions(+), 76 deletions(-)

diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c
index dc037dd927f8..6853b5cb3ac8 100644
--- a/drivers/net/ethernet/sfc/ef10.c
+++ b/drivers/net/ethernet/sfc/ef10.c
@@ -277,24 +277,9 @@ static int efx_ef10_init_datapath_caps(struct efx_nic *efx)
 		u8 vi_window_mode = MCDI_BYTE(outbuf,
 				GET_CAPABILITIES_V3_OUT_VI_WINDOW_MODE);
 
-		switch (vi_window_mode) {
-		case MC_CMD_GET_CAPABILITIES_V3_OUT_VI_WINDOW_MODE_8K:
-			efx->vi_stride = 8192;
-			break;
-		case MC_CMD_GET_CAPABILITIES_V3_OUT_VI_WINDOW_MODE_16K:
-			efx->vi_stride = 16384;
-			break;
-		case MC_CMD_GET_CAPABILITIES_V3_OUT_VI_WINDOW_MODE_64K:
-			efx->vi_stride = 65536;
-			break;
-		default:
-			netif_err(efx, probe, efx->net_dev,
-				  "Unrecognised VI window mode %d\n",
-				  vi_window_mode);
-			return -EIO;
-		}
-		netif_dbg(efx, probe, efx->net_dev, "vi_stride = %u\n",
-			  efx->vi_stride);
+		rc = efx_mcdi_window_mode_to_stride(efx, vi_window_mode);
+		if (rc)
+			return rc;
 	} else {
 		/* keep default VI stride */
 		netif_dbg(efx, probe, efx->net_dev,
diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index 4affae76b03c..7ad97090ab52 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -692,17 +692,6 @@ int efx_net_stop(struct net_device *net_dev)
 	return 0;
 }
 
-/* Context: process, dev_base_lock or RTNL held, non-blocking. */
-static void efx_net_stats(struct net_device *net_dev,
-			  struct rtnl_link_stats64 *stats)
-{
-	struct efx_nic *efx = netdev_priv(net_dev);
-
-	spin_lock_bh(&efx->stats_lock);
-	efx->type->update_stats(efx, NULL, stats);
-	spin_unlock_bh(&efx->stats_lock);
-}
-
 /* Context: netif_tx_lock held, BHs disabled. */
 static void efx_watchdog(struct net_device *net_dev, unsigned int txqueue)
 {
diff --git a/drivers/net/ethernet/sfc/efx_common.c b/drivers/net/ethernet/sfc/efx_common.c
index 33b523b6d0a8..0875507efd0a 100644
--- a/drivers/net/ethernet/sfc/efx_common.c
+++ b/drivers/net/ethernet/sfc/efx_common.c
@@ -454,6 +454,16 @@ void efx_stop_all(struct efx_nic *efx)
 	efx_stop_datapath(efx);
 }
 
+/* Context: process, dev_base_lock or RTNL held, non-blocking. */
+void efx_net_stats(struct net_device *net_dev, struct rtnl_link_stats64 *stats)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+
+	spin_lock_bh(&efx->stats_lock);
+	efx->type->update_stats(efx, NULL, stats);
+	spin_unlock_bh(&efx->stats_lock);
+}
+
 /* Push loopback/power/transmit disable settings to the PHY, and reconfigure
  * the MAC appropriately. All other PHY configuration changes are pushed
  * through phy_op->set_settings(), and pushed asynchronously to the MAC
diff --git a/drivers/net/ethernet/sfc/efx_common.h b/drivers/net/ethernet/sfc/efx_common.h
index e03404d1dc0a..32a23ec9b104 100644
--- a/drivers/net/ethernet/sfc/efx_common.h
+++ b/drivers/net/ethernet/sfc/efx_common.h
@@ -21,6 +21,8 @@ void efx_fini_struct(struct efx_nic *efx);
 void efx_start_all(struct efx_nic *efx);
 void efx_stop_all(struct efx_nic *efx);
 
+void efx_net_stats(struct net_device *net_dev, struct rtnl_link_stats64 *stats);
+
 int efx_create_reset_workqueue(void);
 void efx_queue_reset_work(struct efx_nic *efx);
 void efx_flush_reset_workqueue(struct efx_nic *efx);
diff --git a/drivers/net/ethernet/sfc/mcdi.h b/drivers/net/ethernet/sfc/mcdi.h
index 65e454a062f7..54a45010b576 100644
--- a/drivers/net/ethernet/sfc/mcdi.h
+++ b/drivers/net/ethernet/sfc/mcdi.h
@@ -348,8 +348,6 @@ void efx_mcdi_port_remove(struct efx_nic *efx);
 int efx_mcdi_port_reconfigure(struct efx_nic *efx);
 u32 efx_mcdi_phy_get_caps(struct efx_nic *efx);
 void efx_mcdi_process_link_change(struct efx_nic *efx, efx_qword_t *ev);
-int efx_mcdi_set_mac(struct efx_nic *efx);
-#define EFX_MC_STATS_GENERATION_INVALID ((__force __le64)(-1))
 void efx_mcdi_mac_start_stats(struct efx_nic *efx);
 void efx_mcdi_mac_stop_stats(struct efx_nic *efx);
 void efx_mcdi_mac_pull_stats(struct efx_nic *efx);
diff --git a/drivers/net/ethernet/sfc/mcdi_functions.c b/drivers/net/ethernet/sfc/mcdi_functions.c
index f022e2b9e975..0f571864df22 100644
--- a/drivers/net/ethernet/sfc/mcdi_functions.c
+++ b/drivers/net/ethernet/sfc/mcdi_functions.c
@@ -347,3 +347,26 @@ fail:
 	efx_mcdi_display_error(efx, MC_CMD_FINI_RXQ, MC_CMD_FINI_RXQ_IN_LEN,
 			       outbuf, outlen, rc);
 }
+
+int efx_mcdi_window_mode_to_stride(struct efx_nic *efx, u8 vi_window_mode)
+{
+	switch (vi_window_mode) {
+	case MC_CMD_GET_CAPABILITIES_V3_OUT_VI_WINDOW_MODE_8K:
+		efx->vi_stride = 8192;
+		break;
+	case MC_CMD_GET_CAPABILITIES_V3_OUT_VI_WINDOW_MODE_16K:
+		efx->vi_stride = 16384;
+		break;
+	case MC_CMD_GET_CAPABILITIES_V3_OUT_VI_WINDOW_MODE_64K:
+		efx->vi_stride = 65536;
+		break;
+	default:
+		netif_err(efx, probe, efx->net_dev,
+			  "Unrecognised VI window mode %d\n",
+			  vi_window_mode);
+		return -EIO;
+	}
+	netif_dbg(efx, probe, efx->net_dev, "vi_stride = %u\n",
+		  efx->vi_stride);
+	return 0;
+}
diff --git a/drivers/net/ethernet/sfc/mcdi_functions.h b/drivers/net/ethernet/sfc/mcdi_functions.h
index 3c9e760238e7..0396294bfdcd 100644
--- a/drivers/net/ethernet/sfc/mcdi_functions.h
+++ b/drivers/net/ethernet/sfc/mcdi_functions.h
@@ -26,5 +26,6 @@ int efx_mcdi_rx_probe(struct efx_rx_queue *rx_queue);
 void efx_mcdi_rx_init(struct efx_rx_queue *rx_queue);
 void efx_mcdi_rx_remove(struct efx_rx_queue *rx_queue);
 void efx_mcdi_rx_fini(struct efx_rx_queue *rx_queue);
+int efx_mcdi_window_mode_to_stride(struct efx_nic *efx, u8 vi_window_mode);
 
 #endif
diff --git a/drivers/net/ethernet/sfc/mcdi_port.c b/drivers/net/ethernet/sfc/mcdi_port.c
index 0db6068a668d..a5785205c32b 100644
--- a/drivers/net/ethernet/sfc/mcdi_port.c
+++ b/drivers/net/ethernet/sfc/mcdi_port.c
@@ -708,51 +708,6 @@ void efx_mcdi_process_link_change(struct efx_nic *efx, efx_qword_t *ev)
 	efx_link_status_changed(efx);
 }
 
-int efx_mcdi_set_mac(struct efx_nic *efx)
-{
-	u32 fcntl;
-	MCDI_DECLARE_BUF(cmdbytes, MC_CMD_SET_MAC_IN_LEN);
-
-	BUILD_BUG_ON(MC_CMD_SET_MAC_OUT_LEN != 0);
-
-	/* This has no effect on EF10 */
-	ether_addr_copy(MCDI_PTR(cmdbytes, SET_MAC_IN_ADDR),
-			efx->net_dev->dev_addr);
-
-	MCDI_SET_DWORD(cmdbytes, SET_MAC_IN_MTU,
-			EFX_MAX_FRAME_LEN(efx->net_dev->mtu));
-	MCDI_SET_DWORD(cmdbytes, SET_MAC_IN_DRAIN, 0);
-
-	/* Set simple MAC filter for Siena */
-	MCDI_POPULATE_DWORD_1(cmdbytes, SET_MAC_IN_REJECT,
-			      SET_MAC_IN_REJECT_UNCST, efx->unicast_filter);
-
-	MCDI_POPULATE_DWORD_1(cmdbytes, SET_MAC_IN_FLAGS,
-			      SET_MAC_IN_FLAG_INCLUDE_FCS,
-			      !!(efx->net_dev->features & NETIF_F_RXFCS));
-
-	switch (efx->wanted_fc) {
-	case EFX_FC_RX | EFX_FC_TX:
-		fcntl = MC_CMD_FCNTL_BIDIR;
-		break;
-	case EFX_FC_RX:
-		fcntl = MC_CMD_FCNTL_RESPOND;
-		break;
-	default:
-		fcntl = MC_CMD_FCNTL_OFF;
-		break;
-	}
-	if (efx->wanted_fc & EFX_FC_AUTO)
-		fcntl = MC_CMD_FCNTL_AUTO;
-	if (efx->fc_disable)
-		fcntl = MC_CMD_FCNTL_OFF;
-
-	MCDI_SET_DWORD(cmdbytes, SET_MAC_IN_FCNTL, fcntl);
-
-	return efx_mcdi_rpc(efx, MC_CMD_SET_MAC, cmdbytes, sizeof(cmdbytes),
-			    NULL, 0, NULL);
-}
-
 bool efx_mcdi_mac_check_fault(struct efx_nic *efx)
 {
 	MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_LINK_OUT_LEN);
diff --git a/drivers/net/ethernet/sfc/mcdi_port_common.c b/drivers/net/ethernet/sfc/mcdi_port_common.c
index 3b1c559246b4..0f3237fc68af 100644
--- a/drivers/net/ethernet/sfc/mcdi_port_common.c
+++ b/drivers/net/ethernet/sfc/mcdi_port_common.c
@@ -474,6 +474,51 @@ int efx_mcdi_phy_test_alive(struct efx_nic *efx)
 	return 0;
 }
 
+int efx_mcdi_set_mac(struct efx_nic *efx)
+{
+	u32 fcntl;
+	MCDI_DECLARE_BUF(cmdbytes, MC_CMD_SET_MAC_IN_LEN);
+
+	BUILD_BUG_ON(MC_CMD_SET_MAC_OUT_LEN != 0);
+
+	/* This has no effect on EF10 */
+	ether_addr_copy(MCDI_PTR(cmdbytes, SET_MAC_IN_ADDR),
+			efx->net_dev->dev_addr);
+
+	MCDI_SET_DWORD(cmdbytes, SET_MAC_IN_MTU,
+		       EFX_MAX_FRAME_LEN(efx->net_dev->mtu));
+	MCDI_SET_DWORD(cmdbytes, SET_MAC_IN_DRAIN, 0);
+
+	/* Set simple MAC filter for Siena */
+	MCDI_POPULATE_DWORD_1(cmdbytes, SET_MAC_IN_REJECT,
+			      SET_MAC_IN_REJECT_UNCST, efx->unicast_filter);
+
+	MCDI_POPULATE_DWORD_1(cmdbytes, SET_MAC_IN_FLAGS,
+			      SET_MAC_IN_FLAG_INCLUDE_FCS,
+			      !!(efx->net_dev->features & NETIF_F_RXFCS));
+
+	switch (efx->wanted_fc) {
+	case EFX_FC_RX | EFX_FC_TX:
+		fcntl = MC_CMD_FCNTL_BIDIR;
+		break;
+	case EFX_FC_RX:
+		fcntl = MC_CMD_FCNTL_RESPOND;
+		break;
+	default:
+		fcntl = MC_CMD_FCNTL_OFF;
+		break;
+	}
+	if (efx->wanted_fc & EFX_FC_AUTO)
+		fcntl = MC_CMD_FCNTL_AUTO;
+	if (efx->fc_disable)
+		fcntl = MC_CMD_FCNTL_OFF;
+
+	MCDI_SET_DWORD(cmdbytes, SET_MAC_IN_FCNTL, fcntl);
+
+	return efx_mcdi_rpc(efx, MC_CMD_SET_MAC, cmdbytes, sizeof(cmdbytes),
+			    NULL, 0, NULL);
+}
+
 /* Get physical port number (EF10 only; on Siena it is same as PF number) */
 int efx_mcdi_port_get_number(struct efx_nic *efx)
 {
diff --git a/drivers/net/ethernet/sfc/mcdi_port_common.h b/drivers/net/ethernet/sfc/mcdi_port_common.h
index 10772de94b2c..6b08a2b7eddf 100644
--- a/drivers/net/ethernet/sfc/mcdi_port_common.h
+++ b/drivers/net/ethernet/sfc/mcdi_port_common.h
@@ -28,6 +28,8 @@ struct efx_mcdi_phy_data {
 	u32 forced_cap;
 };
 
+#define EFX_MC_STATS_GENERATION_INVALID ((__force __le64)(-1))
+
 int efx_mcdi_get_phy_cfg(struct efx_nic *efx, struct efx_mcdi_phy_data *cfg);
 void efx_link_set_advertising(struct efx_nic *efx,
 			      const unsigned long *advertising);
@@ -48,6 +50,7 @@ bool efx_mcdi_phy_poll(struct efx_nic *efx);
 int efx_mcdi_phy_get_fecparam(struct efx_nic *efx,
 			      struct ethtool_fecparam *fec);
 int efx_mcdi_phy_test_alive(struct efx_nic *efx);
+int efx_mcdi_set_mac(struct efx_nic *efx);
 int efx_mcdi_port_get_number(struct efx_nic *efx);
 
 #endif
diff --git a/drivers/net/ethernet/sfc/siena.c b/drivers/net/ethernet/sfc/siena.c
index 810f6fc8a937..baa464161626 100644
--- a/drivers/net/ethernet/sfc/siena.c
+++ b/drivers/net/ethernet/sfc/siena.c
@@ -21,6 +21,7 @@
 #include "workarounds.h"
 #include "mcdi.h"
 #include "mcdi_pcol.h"
+#include "mcdi_port_common.h"
 #include "selftest.h"
 #include "siena_sriov.h"
 
-- 
cgit v1.2.3-59-g8ed1b


From 3653954da498fc80c5da3642ae3f8d8168d7f820 Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Fri, 10 Jan 2020 13:27:32 +0000
Subject: sfc: move some ethtool code

Various ethtool entry points are moved.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/Makefile         |   4 +-
 drivers/net/ethernet/sfc/ethtool.c        | 441 +----------------------------
 drivers/net/ethernet/sfc/ethtool_common.c | 456 ++++++++++++++++++++++++++++++
 drivers/net/ethernet/sfc/ethtool_common.h |  30 ++
 4 files changed, 489 insertions(+), 442 deletions(-)
 create mode 100644 drivers/net/ethernet/sfc/ethtool_common.c
 create mode 100644 drivers/net/ethernet/sfc/ethtool_common.h

diff --git a/drivers/net/ethernet/sfc/Makefile b/drivers/net/ethernet/sfc/Makefile
index ffa84b3ff8e2..890fd65caa2d 100644
--- a/drivers/net/ethernet/sfc/Makefile
+++ b/drivers/net/ethernet/sfc/Makefile
@@ -1,8 +1,8 @@
 # SPDX-License-Identifier: GPL-2.0
 sfc-y			+= efx.o efx_common.o efx_channels.o nic.o \
 			   farch.o siena.o ef10.o \
-			   tx.o tx_common.o rx.o rx_common.o \
-			   selftest.o ethtool.o ptp.o tx_tso.o \
+			   tx.o tx_common.o tx_tso.o rx.o rx_common.o \
+			   selftest.o ethtool.o ethtool_common.o ptp.o \
 			   mcdi.o mcdi_port.o mcdi_port_common.o \
 			   mcdi_functions.o mcdi_mon.o
 sfc-$(CONFIG_SFC_MTD)	+= mtd.o
diff --git a/drivers/net/ethernet/sfc/ethtool.c b/drivers/net/ethernet/sfc/ethtool.c
index f83398721ad7..e50901e4b2ac 100644
--- a/drivers/net/ethernet/sfc/ethtool.c
+++ b/drivers/net/ethernet/sfc/ethtool.c
@@ -16,92 +16,10 @@
 #include "efx_channels.h"
 #include "rx_common.h"
 #include "tx_common.h"
+#include "ethtool_common.h"
 #include "filter.h"
 #include "nic.h"
 
-struct efx_sw_stat_desc {
-	const char *name;
-	enum {
-		EFX_ETHTOOL_STAT_SOURCE_nic,
-		EFX_ETHTOOL_STAT_SOURCE_channel,
-		EFX_ETHTOOL_STAT_SOURCE_tx_queue
-	} source;
-	unsigned offset;
-	u64(*get_stat) (void *field); /* Reader function */
-};
-
-/* Initialiser for a struct efx_sw_stat_desc with type-checking */
-#define EFX_ETHTOOL_STAT(stat_name, source_name, field, field_type, \
-				get_stat_function) {			\
-	.name = #stat_name,						\
-	.source = EFX_ETHTOOL_STAT_SOURCE_##source_name,		\
-	.offset = ((((field_type *) 0) ==				\
-		      &((struct efx_##source_name *)0)->field) ?	\
-		    offsetof(struct efx_##source_name, field) :		\
-		    offsetof(struct efx_##source_name, field)),		\
-	.get_stat = get_stat_function,					\
-}
-
-static u64 efx_get_uint_stat(void *field)
-{
-	return *(unsigned int *)field;
-}
-
-static u64 efx_get_atomic_stat(void *field)
-{
-	return atomic_read((atomic_t *) field);
-}
-
-#define EFX_ETHTOOL_ATOMIC_NIC_ERROR_STAT(field)		\
-	EFX_ETHTOOL_STAT(field, nic, field,			\
-			 atomic_t, efx_get_atomic_stat)
-
-#define EFX_ETHTOOL_UINT_CHANNEL_STAT(field)			\
-	EFX_ETHTOOL_STAT(field, channel, n_##field,		\
-			 unsigned int, efx_get_uint_stat)
-#define EFX_ETHTOOL_UINT_CHANNEL_STAT_NO_N(field)		\
-	EFX_ETHTOOL_STAT(field, channel, field,			\
-			 unsigned int, efx_get_uint_stat)
-
-#define EFX_ETHTOOL_UINT_TXQ_STAT(field)			\
-	EFX_ETHTOOL_STAT(tx_##field, tx_queue, field,		\
-			 unsigned int, efx_get_uint_stat)
-
-static const struct efx_sw_stat_desc efx_sw_stat_desc[] = {
-	EFX_ETHTOOL_UINT_TXQ_STAT(merge_events),
-	EFX_ETHTOOL_UINT_TXQ_STAT(tso_bursts),
-	EFX_ETHTOOL_UINT_TXQ_STAT(tso_long_headers),
-	EFX_ETHTOOL_UINT_TXQ_STAT(tso_packets),
-	EFX_ETHTOOL_UINT_TXQ_STAT(tso_fallbacks),
-	EFX_ETHTOOL_UINT_TXQ_STAT(pushes),
-	EFX_ETHTOOL_UINT_TXQ_STAT(pio_packets),
-	EFX_ETHTOOL_UINT_TXQ_STAT(cb_packets),
-	EFX_ETHTOOL_ATOMIC_NIC_ERROR_STAT(rx_reset),
-	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_tobe_disc),
-	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_ip_hdr_chksum_err),
-	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_tcp_udp_chksum_err),
-	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_inner_ip_hdr_chksum_err),
-	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_inner_tcp_udp_chksum_err),
-	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_outer_ip_hdr_chksum_err),
-	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_outer_tcp_udp_chksum_err),
-	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_eth_crc_err),
-	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_mcast_mismatch),
-	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_frm_trunc),
-	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_merge_events),
-	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_merge_packets),
-	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_xdp_drops),
-	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_xdp_bad_drops),
-	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_xdp_tx),
-	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_xdp_redirect),
-#ifdef CONFIG_RFS_ACCEL
-	EFX_ETHTOOL_UINT_CHANNEL_STAT_NO_N(rfs_filter_count),
-	EFX_ETHTOOL_UINT_CHANNEL_STAT(rfs_succeeded),
-	EFX_ETHTOOL_UINT_CHANNEL_STAT(rfs_failed),
-#endif
-};
-
-#define EFX_ETHTOOL_SW_STAT_COUNT ARRAY_SIZE(efx_sw_stat_desc)
-
 #define EFX_ETHTOOL_EEPROM_MAGIC 0xEFAB
 
 /**************************************************************************
@@ -188,18 +106,6 @@ efx_ethtool_set_link_ksettings(struct net_device *net_dev,
 	return rc;
 }
 
-static void efx_ethtool_get_drvinfo(struct net_device *net_dev,
-				    struct ethtool_drvinfo *info)
-{
-	struct efx_nic *efx = netdev_priv(net_dev);
-
-	strlcpy(info->driver, KBUILD_MODNAME, sizeof(info->driver));
-	strlcpy(info->version, EFX_DRIVER_VERSION, sizeof(info->version));
-	efx_mcdi_print_fwver(efx, info->fw_version,
-			     sizeof(info->fw_version));
-	strlcpy(info->bus_info, pci_name(efx->pci_dev), sizeof(info->bus_info));
-}
-
 static int efx_ethtool_get_regs_len(struct net_device *net_dev)
 {
 	return efx_nic_get_regs_len(netdev_priv(net_dev));
@@ -214,341 +120,6 @@ static void efx_ethtool_get_regs(struct net_device *net_dev,
 	efx_nic_get_regs(efx, buf);
 }
 
-static u32 efx_ethtool_get_msglevel(struct net_device *net_dev)
-{
-	struct efx_nic *efx = netdev_priv(net_dev);
-	return efx->msg_enable;
-}
-
-static void efx_ethtool_set_msglevel(struct net_device *net_dev, u32 msg_enable)
-{
-	struct efx_nic *efx = netdev_priv(net_dev);
-	efx->msg_enable = msg_enable;
-}
-
-/**
- * efx_fill_test - fill in an individual self-test entry
- * @test_index:		Index of the test
- * @strings:		Ethtool strings, or %NULL
- * @data:		Ethtool test results, or %NULL
- * @test:		Pointer to test result (used only if data != %NULL)
- * @unit_format:	Unit name format (e.g. "chan\%d")
- * @unit_id:		Unit id (e.g. 0 for "chan0")
- * @test_format:	Test name format (e.g. "loopback.\%s.tx.sent")
- * @test_id:		Test id (e.g. "PHYXS" for "loopback.PHYXS.tx_sent")
- *
- * Fill in an individual self-test entry.
- */
-static void efx_fill_test(unsigned int test_index, u8 *strings, u64 *data,
-			  int *test, const char *unit_format, int unit_id,
-			  const char *test_format, const char *test_id)
-{
-	char unit_str[ETH_GSTRING_LEN], test_str[ETH_GSTRING_LEN];
-
-	/* Fill data value, if applicable */
-	if (data)
-		data[test_index] = *test;
-
-	/* Fill string, if applicable */
-	if (strings) {
-		if (strchr(unit_format, '%'))
-			snprintf(unit_str, sizeof(unit_str),
-				 unit_format, unit_id);
-		else
-			strcpy(unit_str, unit_format);
-		snprintf(test_str, sizeof(test_str), test_format, test_id);
-		snprintf(strings + test_index * ETH_GSTRING_LEN,
-			 ETH_GSTRING_LEN,
-			 "%-6s %-24s", unit_str, test_str);
-	}
-}
-
-#define EFX_CHANNEL_NAME(_channel) "chan%d", _channel->channel
-#define EFX_TX_QUEUE_NAME(_tx_queue) "txq%d", _tx_queue->queue
-#define EFX_RX_QUEUE_NAME(_rx_queue) "rxq%d", _rx_queue->queue
-#define EFX_LOOPBACK_NAME(_mode, _counter)			\
-	"loopback.%s." _counter, STRING_TABLE_LOOKUP(_mode, efx_loopback_mode)
-
-/**
- * efx_fill_loopback_test - fill in a block of loopback self-test entries
- * @efx:		Efx NIC
- * @lb_tests:		Efx loopback self-test results structure
- * @mode:		Loopback test mode
- * @test_index:		Starting index of the test
- * @strings:		Ethtool strings, or %NULL
- * @data:		Ethtool test results, or %NULL
- *
- * Fill in a block of loopback self-test entries.  Return new test
- * index.
- */
-static int efx_fill_loopback_test(struct efx_nic *efx,
-				  struct efx_loopback_self_tests *lb_tests,
-				  enum efx_loopback_mode mode,
-				  unsigned int test_index,
-				  u8 *strings, u64 *data)
-{
-	struct efx_channel *channel =
-		efx_get_channel(efx, efx->tx_channel_offset);
-	struct efx_tx_queue *tx_queue;
-
-	efx_for_each_channel_tx_queue(tx_queue, channel) {
-		efx_fill_test(test_index++, strings, data,
-			      &lb_tests->tx_sent[tx_queue->queue],
-			      EFX_TX_QUEUE_NAME(tx_queue),
-			      EFX_LOOPBACK_NAME(mode, "tx_sent"));
-		efx_fill_test(test_index++, strings, data,
-			      &lb_tests->tx_done[tx_queue->queue],
-			      EFX_TX_QUEUE_NAME(tx_queue),
-			      EFX_LOOPBACK_NAME(mode, "tx_done"));
-	}
-	efx_fill_test(test_index++, strings, data,
-		      &lb_tests->rx_good,
-		      "rx", 0,
-		      EFX_LOOPBACK_NAME(mode, "rx_good"));
-	efx_fill_test(test_index++, strings, data,
-		      &lb_tests->rx_bad,
-		      "rx", 0,
-		      EFX_LOOPBACK_NAME(mode, "rx_bad"));
-
-	return test_index;
-}
-
-/**
- * efx_ethtool_fill_self_tests - get self-test details
- * @efx:		Efx NIC
- * @tests:		Efx self-test results structure, or %NULL
- * @strings:		Ethtool strings, or %NULL
- * @data:		Ethtool test results, or %NULL
- *
- * Get self-test number of strings, strings, and/or test results.
- * Return number of strings (== number of test results).
- *
- * The reason for merging these three functions is to make sure that
- * they can never be inconsistent.
- */
-static int efx_ethtool_fill_self_tests(struct efx_nic *efx,
-				       struct efx_self_tests *tests,
-				       u8 *strings, u64 *data)
-{
-	struct efx_channel *channel;
-	unsigned int n = 0, i;
-	enum efx_loopback_mode mode;
-
-	efx_fill_test(n++, strings, data, &tests->phy_alive,
-		      "phy", 0, "alive", NULL);
-	efx_fill_test(n++, strings, data, &tests->nvram,
-		      "core", 0, "nvram", NULL);
-	efx_fill_test(n++, strings, data, &tests->interrupt,
-		      "core", 0, "interrupt", NULL);
-
-	/* Event queues */
-	efx_for_each_channel(channel, efx) {
-		efx_fill_test(n++, strings, data,
-			      &tests->eventq_dma[channel->channel],
-			      EFX_CHANNEL_NAME(channel),
-			      "eventq.dma", NULL);
-		efx_fill_test(n++, strings, data,
-			      &tests->eventq_int[channel->channel],
-			      EFX_CHANNEL_NAME(channel),
-			      "eventq.int", NULL);
-	}
-
-	efx_fill_test(n++, strings, data, &tests->memory,
-		      "core", 0, "memory", NULL);
-	efx_fill_test(n++, strings, data, &tests->registers,
-		      "core", 0, "registers", NULL);
-
-	if (efx->phy_op->run_tests != NULL) {
-		EFX_WARN_ON_PARANOID(efx->phy_op->test_name == NULL);
-
-		for (i = 0; true; ++i) {
-			const char *name;
-
-			EFX_WARN_ON_PARANOID(i >= EFX_MAX_PHY_TESTS);
-			name = efx->phy_op->test_name(efx, i);
-			if (name == NULL)
-				break;
-
-			efx_fill_test(n++, strings, data, &tests->phy_ext[i],
-				      "phy", 0, name, NULL);
-		}
-	}
-
-	/* Loopback tests */
-	for (mode = LOOPBACK_NONE; mode <= LOOPBACK_TEST_MAX; mode++) {
-		if (!(efx->loopback_modes & (1 << mode)))
-			continue;
-		n = efx_fill_loopback_test(efx,
-					   &tests->loopback[mode], mode, n,
-					   strings, data);
-	}
-
-	return n;
-}
-
-static size_t efx_describe_per_queue_stats(struct efx_nic *efx, u8 *strings)
-{
-	size_t n_stats = 0;
-	struct efx_channel *channel;
-
-	efx_for_each_channel(channel, efx) {
-		if (efx_channel_has_tx_queues(channel)) {
-			n_stats++;
-			if (strings != NULL) {
-				snprintf(strings, ETH_GSTRING_LEN,
-					 "tx-%u.tx_packets",
-					 channel->tx_queue[0].queue /
-					 EFX_TXQ_TYPES);
-
-				strings += ETH_GSTRING_LEN;
-			}
-		}
-	}
-	efx_for_each_channel(channel, efx) {
-		if (efx_channel_has_rx_queue(channel)) {
-			n_stats++;
-			if (strings != NULL) {
-				snprintf(strings, ETH_GSTRING_LEN,
-					 "rx-%d.rx_packets", channel->channel);
-				strings += ETH_GSTRING_LEN;
-			}
-		}
-	}
-	if (efx->xdp_tx_queue_count && efx->xdp_tx_queues) {
-		unsigned short xdp;
-
-		for (xdp = 0; xdp < efx->xdp_tx_queue_count; xdp++) {
-			n_stats++;
-			if (strings) {
-				snprintf(strings, ETH_GSTRING_LEN,
-					 "tx-xdp-cpu-%hu.tx_packets", xdp);
-				strings += ETH_GSTRING_LEN;
-			}
-		}
-	}
-
-	return n_stats;
-}
-
-static int efx_ethtool_get_sset_count(struct net_device *net_dev,
-				      int string_set)
-{
-	struct efx_nic *efx = netdev_priv(net_dev);
-
-	switch (string_set) {
-	case ETH_SS_STATS:
-		return efx->type->describe_stats(efx, NULL) +
-		       EFX_ETHTOOL_SW_STAT_COUNT +
-		       efx_describe_per_queue_stats(efx, NULL) +
-		       efx_ptp_describe_stats(efx, NULL);
-	case ETH_SS_TEST:
-		return efx_ethtool_fill_self_tests(efx, NULL, NULL, NULL);
-	default:
-		return -EINVAL;
-	}
-}
-
-static void efx_ethtool_get_strings(struct net_device *net_dev,
-				    u32 string_set, u8 *strings)
-{
-	struct efx_nic *efx = netdev_priv(net_dev);
-	int i;
-
-	switch (string_set) {
-	case ETH_SS_STATS:
-		strings += (efx->type->describe_stats(efx, strings) *
-			    ETH_GSTRING_LEN);
-		for (i = 0; i < EFX_ETHTOOL_SW_STAT_COUNT; i++)
-			strlcpy(strings + i * ETH_GSTRING_LEN,
-				efx_sw_stat_desc[i].name, ETH_GSTRING_LEN);
-		strings += EFX_ETHTOOL_SW_STAT_COUNT * ETH_GSTRING_LEN;
-		strings += (efx_describe_per_queue_stats(efx, strings) *
-			    ETH_GSTRING_LEN);
-		efx_ptp_describe_stats(efx, strings);
-		break;
-	case ETH_SS_TEST:
-		efx_ethtool_fill_self_tests(efx, NULL, strings, NULL);
-		break;
-	default:
-		/* No other string sets */
-		break;
-	}
-}
-
-static void efx_ethtool_get_stats(struct net_device *net_dev,
-				  struct ethtool_stats *stats,
-				  u64 *data)
-{
-	struct efx_nic *efx = netdev_priv(net_dev);
-	const struct efx_sw_stat_desc *stat;
-	struct efx_channel *channel;
-	struct efx_tx_queue *tx_queue;
-	struct efx_rx_queue *rx_queue;
-	int i;
-
-	spin_lock_bh(&efx->stats_lock);
-
-	/* Get NIC statistics */
-	data += efx->type->update_stats(efx, data, NULL);
-
-	/* Get software statistics */
-	for (i = 0; i < EFX_ETHTOOL_SW_STAT_COUNT; i++) {
-		stat = &efx_sw_stat_desc[i];
-		switch (stat->source) {
-		case EFX_ETHTOOL_STAT_SOURCE_nic:
-			data[i] = stat->get_stat((void *)efx + stat->offset);
-			break;
-		case EFX_ETHTOOL_STAT_SOURCE_channel:
-			data[i] = 0;
-			efx_for_each_channel(channel, efx)
-				data[i] += stat->get_stat((void *)channel +
-							  stat->offset);
-			break;
-		case EFX_ETHTOOL_STAT_SOURCE_tx_queue:
-			data[i] = 0;
-			efx_for_each_channel(channel, efx) {
-				efx_for_each_channel_tx_queue(tx_queue, channel)
-					data[i] +=
-						stat->get_stat((void *)tx_queue
-							       + stat->offset);
-			}
-			break;
-		}
-	}
-	data += EFX_ETHTOOL_SW_STAT_COUNT;
-
-	spin_unlock_bh(&efx->stats_lock);
-
-	efx_for_each_channel(channel, efx) {
-		if (efx_channel_has_tx_queues(channel)) {
-			*data = 0;
-			efx_for_each_channel_tx_queue(tx_queue, channel) {
-				*data += tx_queue->tx_packets;
-			}
-			data++;
-		}
-	}
-	efx_for_each_channel(channel, efx) {
-		if (efx_channel_has_rx_queue(channel)) {
-			*data = 0;
-			efx_for_each_channel_rx_queue(rx_queue, channel) {
-				*data += rx_queue->rx_packets;
-			}
-			data++;
-		}
-	}
-	if (efx->xdp_tx_queue_count && efx->xdp_tx_queues) {
-		int xdp;
-
-		for (xdp = 0; xdp < efx->xdp_tx_queue_count; xdp++) {
-			data[0] = efx->xdp_tx_queues[xdp]->tx_packets;
-			data++;
-		}
-	}
-
-	efx_ptp_update_stats(efx, data);
-}
-
 static void efx_ethtool_self_test(struct net_device *net_dev,
 				  struct ethtool_test *test, u64 *data)
 {
@@ -790,16 +361,6 @@ out:
 	return rc;
 }
 
-static void efx_ethtool_get_pauseparam(struct net_device *net_dev,
-				       struct ethtool_pauseparam *pause)
-{
-	struct efx_nic *efx = netdev_priv(net_dev);
-
-	pause->rx_pause = !!(efx->wanted_fc & EFX_FC_RX);
-	pause->tx_pause = !!(efx->wanted_fc & EFX_FC_TX);
-	pause->autoneg = !!(efx->wanted_fc & EFX_FC_AUTO);
-}
-
 static void efx_ethtool_get_wol(struct net_device *net_dev,
 				struct ethtool_wolinfo *wol)
 {
diff --git a/drivers/net/ethernet/sfc/ethtool_common.c b/drivers/net/ethernet/sfc/ethtool_common.c
new file mode 100644
index 000000000000..3d7f75cc5cf0
--- /dev/null
+++ b/drivers/net/ethernet/sfc/ethtool_common.c
@@ -0,0 +1,456 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/****************************************************************************
+ * Driver for Solarflare network controllers and boards
+ * Copyright 2019 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+#include <linux/module.h>
+#include <linux/netdevice.h>
+#include "net_driver.h"
+#include "mcdi.h"
+#include "nic.h"
+#include "selftest.h"
+#include "ethtool_common.h"
+
+struct efx_sw_stat_desc {
+	const char *name;
+	enum {
+		EFX_ETHTOOL_STAT_SOURCE_nic,
+		EFX_ETHTOOL_STAT_SOURCE_channel,
+		EFX_ETHTOOL_STAT_SOURCE_tx_queue
+	} source;
+	unsigned int offset;
+	u64 (*get_stat)(void *field); /* Reader function */
+};
+
+/* Initialiser for a struct efx_sw_stat_desc with type-checking */
+#define EFX_ETHTOOL_STAT(stat_name, source_name, field, field_type, \
+				get_stat_function) {			\
+	.name = #stat_name,						\
+	.source = EFX_ETHTOOL_STAT_SOURCE_##source_name,		\
+	.offset = ((((field_type *) 0) ==				\
+		      &((struct efx_##source_name *)0)->field) ?	\
+		    offsetof(struct efx_##source_name, field) :		\
+		    offsetof(struct efx_##source_name, field)),		\
+	.get_stat = get_stat_function,					\
+}
+
+static u64 efx_get_uint_stat(void *field)
+{
+	return *(unsigned int *)field;
+}
+
+static u64 efx_get_atomic_stat(void *field)
+{
+	return atomic_read((atomic_t *) field);
+}
+
+#define EFX_ETHTOOL_ATOMIC_NIC_ERROR_STAT(field)		\
+	EFX_ETHTOOL_STAT(field, nic, field,			\
+			 atomic_t, efx_get_atomic_stat)
+
+#define EFX_ETHTOOL_UINT_CHANNEL_STAT(field)			\
+	EFX_ETHTOOL_STAT(field, channel, n_##field,		\
+			 unsigned int, efx_get_uint_stat)
+#define EFX_ETHTOOL_UINT_CHANNEL_STAT_NO_N(field)		\
+	EFX_ETHTOOL_STAT(field, channel, field,			\
+			 unsigned int, efx_get_uint_stat)
+
+#define EFX_ETHTOOL_UINT_TXQ_STAT(field)			\
+	EFX_ETHTOOL_STAT(tx_##field, tx_queue, field,		\
+			 unsigned int, efx_get_uint_stat)
+
+static const struct efx_sw_stat_desc efx_sw_stat_desc[] = {
+	EFX_ETHTOOL_UINT_TXQ_STAT(merge_events),
+	EFX_ETHTOOL_UINT_TXQ_STAT(tso_bursts),
+	EFX_ETHTOOL_UINT_TXQ_STAT(tso_long_headers),
+	EFX_ETHTOOL_UINT_TXQ_STAT(tso_packets),
+	EFX_ETHTOOL_UINT_TXQ_STAT(tso_fallbacks),
+	EFX_ETHTOOL_UINT_TXQ_STAT(pushes),
+	EFX_ETHTOOL_UINT_TXQ_STAT(pio_packets),
+	EFX_ETHTOOL_UINT_TXQ_STAT(cb_packets),
+	EFX_ETHTOOL_ATOMIC_NIC_ERROR_STAT(rx_reset),
+	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_tobe_disc),
+	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_ip_hdr_chksum_err),
+	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_tcp_udp_chksum_err),
+	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_inner_ip_hdr_chksum_err),
+	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_inner_tcp_udp_chksum_err),
+	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_outer_ip_hdr_chksum_err),
+	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_outer_tcp_udp_chksum_err),
+	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_eth_crc_err),
+	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_mcast_mismatch),
+	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_frm_trunc),
+	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_merge_events),
+	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_merge_packets),
+	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_xdp_drops),
+	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_xdp_bad_drops),
+	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_xdp_tx),
+	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_xdp_redirect),
+#ifdef CONFIG_RFS_ACCEL
+	EFX_ETHTOOL_UINT_CHANNEL_STAT_NO_N(rfs_filter_count),
+	EFX_ETHTOOL_UINT_CHANNEL_STAT(rfs_succeeded),
+	EFX_ETHTOOL_UINT_CHANNEL_STAT(rfs_failed),
+#endif
+};
+
+#define EFX_ETHTOOL_SW_STAT_COUNT ARRAY_SIZE(efx_sw_stat_desc)
+
+void efx_ethtool_get_drvinfo(struct net_device *net_dev,
+			     struct ethtool_drvinfo *info)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+
+	strlcpy(info->driver, KBUILD_MODNAME, sizeof(info->driver));
+	strlcpy(info->version, EFX_DRIVER_VERSION, sizeof(info->version));
+	efx_mcdi_print_fwver(efx, info->fw_version,
+			     sizeof(info->fw_version));
+	strlcpy(info->bus_info, pci_name(efx->pci_dev), sizeof(info->bus_info));
+}
+
+u32 efx_ethtool_get_msglevel(struct net_device *net_dev)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+
+	return efx->msg_enable;
+}
+
+void efx_ethtool_set_msglevel(struct net_device *net_dev, u32 msg_enable)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+
+	efx->msg_enable = msg_enable;
+}
+
+void efx_ethtool_get_pauseparam(struct net_device *net_dev,
+				struct ethtool_pauseparam *pause)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+
+	pause->rx_pause = !!(efx->wanted_fc & EFX_FC_RX);
+	pause->tx_pause = !!(efx->wanted_fc & EFX_FC_TX);
+	pause->autoneg = !!(efx->wanted_fc & EFX_FC_AUTO);
+}
+
+/**
+ * efx_fill_test - fill in an individual self-test entry
+ * @test_index:		Index of the test
+ * @strings:		Ethtool strings, or %NULL
+ * @data:		Ethtool test results, or %NULL
+ * @test:		Pointer to test result (used only if data != %NULL)
+ * @unit_format:	Unit name format (e.g. "chan\%d")
+ * @unit_id:		Unit id (e.g. 0 for "chan0")
+ * @test_format:	Test name format (e.g. "loopback.\%s.tx.sent")
+ * @test_id:		Test id (e.g. "PHYXS" for "loopback.PHYXS.tx_sent")
+ *
+ * Fill in an individual self-test entry.
+ */
+void efx_fill_test(unsigned int test_index, u8 *strings, u64 *data,
+		   int *test, const char *unit_format, int unit_id,
+		   const char *test_format, const char *test_id)
+{
+	char unit_str[ETH_GSTRING_LEN], test_str[ETH_GSTRING_LEN];
+
+	/* Fill data value, if applicable */
+	if (data)
+		data[test_index] = *test;
+
+	/* Fill string, if applicable */
+	if (strings) {
+		if (strchr(unit_format, '%'))
+			snprintf(unit_str, sizeof(unit_str),
+				 unit_format, unit_id);
+		else
+			strcpy(unit_str, unit_format);
+		snprintf(test_str, sizeof(test_str), test_format, test_id);
+		snprintf(strings + test_index * ETH_GSTRING_LEN,
+			 ETH_GSTRING_LEN,
+			 "%-6s %-24s", unit_str, test_str);
+	}
+}
+
+#define EFX_CHANNEL_NAME(_channel) "chan%d", _channel->channel
+#define EFX_TX_QUEUE_NAME(_tx_queue) "txq%d", _tx_queue->queue
+#define EFX_RX_QUEUE_NAME(_rx_queue) "rxq%d", _rx_queue->queue
+#define EFX_LOOPBACK_NAME(_mode, _counter)			\
+	"loopback.%s." _counter, STRING_TABLE_LOOKUP(_mode, efx_loopback_mode)
+
+/**
+ * efx_fill_loopback_test - fill in a block of loopback self-test entries
+ * @efx:		Efx NIC
+ * @lb_tests:		Efx loopback self-test results structure
+ * @mode:		Loopback test mode
+ * @test_index:		Starting index of the test
+ * @strings:		Ethtool strings, or %NULL
+ * @data:		Ethtool test results, or %NULL
+ *
+ * Fill in a block of loopback self-test entries.  Return new test
+ * index.
+ */
+int efx_fill_loopback_test(struct efx_nic *efx,
+			   struct efx_loopback_self_tests *lb_tests,
+			   enum efx_loopback_mode mode,
+			   unsigned int test_index, u8 *strings, u64 *data)
+{
+	struct efx_channel *channel =
+		efx_get_channel(efx, efx->tx_channel_offset);
+	struct efx_tx_queue *tx_queue;
+
+	efx_for_each_channel_tx_queue(tx_queue, channel) {
+		efx_fill_test(test_index++, strings, data,
+			      &lb_tests->tx_sent[tx_queue->queue],
+			      EFX_TX_QUEUE_NAME(tx_queue),
+			      EFX_LOOPBACK_NAME(mode, "tx_sent"));
+		efx_fill_test(test_index++, strings, data,
+			      &lb_tests->tx_done[tx_queue->queue],
+			      EFX_TX_QUEUE_NAME(tx_queue),
+			      EFX_LOOPBACK_NAME(mode, "tx_done"));
+	}
+	efx_fill_test(test_index++, strings, data,
+		      &lb_tests->rx_good,
+		      "rx", 0,
+		      EFX_LOOPBACK_NAME(mode, "rx_good"));
+	efx_fill_test(test_index++, strings, data,
+		      &lb_tests->rx_bad,
+		      "rx", 0,
+		      EFX_LOOPBACK_NAME(mode, "rx_bad"));
+
+	return test_index;
+}
+
+/**
+ * efx_ethtool_fill_self_tests - get self-test details
+ * @efx:		Efx NIC
+ * @tests:		Efx self-test results structure, or %NULL
+ * @strings:		Ethtool strings, or %NULL
+ * @data:		Ethtool test results, or %NULL
+ *
+ * Get self-test number of strings, strings, and/or test results.
+ * Return number of strings (== number of test results).
+ *
+ * The reason for merging these three functions is to make sure that
+ * they can never be inconsistent.
+ */
+int efx_ethtool_fill_self_tests(struct efx_nic *efx,
+				struct efx_self_tests *tests,
+				u8 *strings, u64 *data)
+{
+	struct efx_channel *channel;
+	unsigned int n = 0, i;
+	enum efx_loopback_mode mode;
+
+	efx_fill_test(n++, strings, data, &tests->phy_alive,
+		      "phy", 0, "alive", NULL);
+	efx_fill_test(n++, strings, data, &tests->nvram,
+		      "core", 0, "nvram", NULL);
+	efx_fill_test(n++, strings, data, &tests->interrupt,
+		      "core", 0, "interrupt", NULL);
+
+	/* Event queues */
+	efx_for_each_channel(channel, efx) {
+		efx_fill_test(n++, strings, data,
+			      &tests->eventq_dma[channel->channel],
+			      EFX_CHANNEL_NAME(channel),
+			      "eventq.dma", NULL);
+		efx_fill_test(n++, strings, data,
+			      &tests->eventq_int[channel->channel],
+			      EFX_CHANNEL_NAME(channel),
+			      "eventq.int", NULL);
+	}
+
+	efx_fill_test(n++, strings, data, &tests->memory,
+		      "core", 0, "memory", NULL);
+	efx_fill_test(n++, strings, data, &tests->registers,
+		      "core", 0, "registers", NULL);
+
+	if (efx->phy_op->run_tests != NULL) {
+		EFX_WARN_ON_PARANOID(efx->phy_op->test_name == NULL);
+
+		for (i = 0; true; ++i) {
+			const char *name;
+
+			EFX_WARN_ON_PARANOID(i >= EFX_MAX_PHY_TESTS);
+			name = efx->phy_op->test_name(efx, i);
+			if (name == NULL)
+				break;
+
+			efx_fill_test(n++, strings, data, &tests->phy_ext[i],
+				      "phy", 0, name, NULL);
+		}
+	}
+
+	/* Loopback tests */
+	for (mode = LOOPBACK_NONE; mode <= LOOPBACK_TEST_MAX; mode++) {
+		if (!(efx->loopback_modes & (1 << mode)))
+			continue;
+		n = efx_fill_loopback_test(efx,
+					   &tests->loopback[mode], mode, n,
+					   strings, data);
+	}
+
+	return n;
+}
+
+size_t efx_describe_per_queue_stats(struct efx_nic *efx, u8 *strings)
+{
+	size_t n_stats = 0;
+	struct efx_channel *channel;
+
+	efx_for_each_channel(channel, efx) {
+		if (efx_channel_has_tx_queues(channel)) {
+			n_stats++;
+			if (strings != NULL) {
+				snprintf(strings, ETH_GSTRING_LEN,
+					 "tx-%u.tx_packets",
+					 channel->tx_queue[0].queue /
+					 EFX_TXQ_TYPES);
+
+				strings += ETH_GSTRING_LEN;
+			}
+		}
+	}
+	efx_for_each_channel(channel, efx) {
+		if (efx_channel_has_rx_queue(channel)) {
+			n_stats++;
+			if (strings != NULL) {
+				snprintf(strings, ETH_GSTRING_LEN,
+					 "rx-%d.rx_packets", channel->channel);
+				strings += ETH_GSTRING_LEN;
+			}
+		}
+	}
+	if (efx->xdp_tx_queue_count && efx->xdp_tx_queues) {
+		unsigned short xdp;
+
+		for (xdp = 0; xdp < efx->xdp_tx_queue_count; xdp++) {
+			n_stats++;
+			if (strings) {
+				snprintf(strings, ETH_GSTRING_LEN,
+					 "tx-xdp-cpu-%hu.tx_packets", xdp);
+				strings += ETH_GSTRING_LEN;
+			}
+		}
+	}
+
+	return n_stats;
+}
+
+int efx_ethtool_get_sset_count(struct net_device *net_dev, int string_set)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+
+	switch (string_set) {
+	case ETH_SS_STATS:
+		return efx->type->describe_stats(efx, NULL) +
+		       EFX_ETHTOOL_SW_STAT_COUNT +
+		       efx_describe_per_queue_stats(efx, NULL) +
+		       efx_ptp_describe_stats(efx, NULL);
+	case ETH_SS_TEST:
+		return efx_ethtool_fill_self_tests(efx, NULL, NULL, NULL);
+	default:
+		return -EINVAL;
+	}
+}
+
+void efx_ethtool_get_strings(struct net_device *net_dev,
+			     u32 string_set, u8 *strings)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	int i;
+
+	switch (string_set) {
+	case ETH_SS_STATS:
+		strings += (efx->type->describe_stats(efx, strings) *
+			    ETH_GSTRING_LEN);
+		for (i = 0; i < EFX_ETHTOOL_SW_STAT_COUNT; i++)
+			strlcpy(strings + i * ETH_GSTRING_LEN,
+				efx_sw_stat_desc[i].name, ETH_GSTRING_LEN);
+		strings += EFX_ETHTOOL_SW_STAT_COUNT * ETH_GSTRING_LEN;
+		strings += (efx_describe_per_queue_stats(efx, strings) *
+			    ETH_GSTRING_LEN);
+		efx_ptp_describe_stats(efx, strings);
+		break;
+	case ETH_SS_TEST:
+		efx_ethtool_fill_self_tests(efx, NULL, strings, NULL);
+		break;
+	default:
+		/* No other string sets */
+		break;
+	}
+}
+
+void efx_ethtool_get_stats(struct net_device *net_dev,
+			   struct ethtool_stats *stats,
+			   u64 *data)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	const struct efx_sw_stat_desc *stat;
+	struct efx_channel *channel;
+	struct efx_tx_queue *tx_queue;
+	struct efx_rx_queue *rx_queue;
+	int i;
+
+	spin_lock_bh(&efx->stats_lock);
+
+	/* Get NIC statistics */
+	data += efx->type->update_stats(efx, data, NULL);
+
+	/* Get software statistics */
+	for (i = 0; i < EFX_ETHTOOL_SW_STAT_COUNT; i++) {
+		stat = &efx_sw_stat_desc[i];
+		switch (stat->source) {
+		case EFX_ETHTOOL_STAT_SOURCE_nic:
+			data[i] = stat->get_stat((void *)efx + stat->offset);
+			break;
+		case EFX_ETHTOOL_STAT_SOURCE_channel:
+			data[i] = 0;
+			efx_for_each_channel(channel, efx)
+				data[i] += stat->get_stat((void *)channel +
+							  stat->offset);
+			break;
+		case EFX_ETHTOOL_STAT_SOURCE_tx_queue:
+			data[i] = 0;
+			efx_for_each_channel(channel, efx) {
+				efx_for_each_channel_tx_queue(tx_queue, channel)
+					data[i] +=
+						stat->get_stat((void *)tx_queue
+							       + stat->offset);
+			}
+			break;
+		}
+	}
+	data += EFX_ETHTOOL_SW_STAT_COUNT;
+
+	spin_unlock_bh(&efx->stats_lock);
+
+	efx_for_each_channel(channel, efx) {
+		if (efx_channel_has_tx_queues(channel)) {
+			*data = 0;
+			efx_for_each_channel_tx_queue(tx_queue, channel) {
+				*data += tx_queue->tx_packets;
+			}
+			data++;
+		}
+	}
+	efx_for_each_channel(channel, efx) {
+		if (efx_channel_has_rx_queue(channel)) {
+			*data = 0;
+			efx_for_each_channel_rx_queue(rx_queue, channel) {
+				*data += rx_queue->rx_packets;
+			}
+			data++;
+		}
+	}
+	if (efx->xdp_tx_queue_count && efx->xdp_tx_queues) {
+		int xdp;
+
+		for (xdp = 0; xdp < efx->xdp_tx_queue_count; xdp++) {
+			data[0] = efx->xdp_tx_queues[xdp]->tx_packets;
+			data++;
+		}
+	}
+
+	efx_ptp_update_stats(efx, data);
+}
diff --git a/drivers/net/ethernet/sfc/ethtool_common.h b/drivers/net/ethernet/sfc/ethtool_common.h
new file mode 100644
index 000000000000..fa624313f330
--- /dev/null
+++ b/drivers/net/ethernet/sfc/ethtool_common.h
@@ -0,0 +1,30 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/****************************************************************************
+ * Driver for Solarflare network controllers and boards
+ * Copyright 2019 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#ifndef EFX_ETHTOOL_COMMON_H
+#define EFX_ETHTOOL_COMMON_H
+
+void efx_ethtool_get_drvinfo(struct net_device *net_dev,
+			     struct ethtool_drvinfo *info);
+u32 efx_ethtool_get_msglevel(struct net_device *net_dev);
+void efx_ethtool_set_msglevel(struct net_device *net_dev, u32 msg_enable);
+void efx_ethtool_get_pauseparam(struct net_device *net_dev,
+				struct ethtool_pauseparam *pause);
+int efx_ethtool_fill_self_tests(struct efx_nic *efx,
+				struct efx_self_tests *tests,
+				u8 *strings, u64 *data);
+int efx_ethtool_get_sset_count(struct net_device *net_dev, int string_set);
+void efx_ethtool_get_strings(struct net_device *net_dev, u32 string_set,
+			     u8 *strings);
+void efx_ethtool_get_stats(struct net_device *net_dev,
+			   struct ethtool_stats *stats __attribute__ ((unused)),
+			   u64 *data);
+
+#endif
-- 
cgit v1.2.3-59-g8ed1b


From 88f7df35ceac7bdd4db33ceefcb1fe0b90112cda Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Fri, 10 Jan 2020 13:27:50 +0000
Subject: sfc: move a couple more functions

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/mcdi_port.c        | 33 ----------------------------
 drivers/net/ethernet/sfc/mcdi_port_common.c | 34 +++++++++++++++++++++++++++++
 drivers/net/ethernet/sfc/mcdi_port_common.h |  1 +
 drivers/net/ethernet/sfc/tx.c               | 18 ---------------
 drivers/net/ethernet/sfc/tx_common.c        | 18 +++++++++++++++
 drivers/net/ethernet/sfc/tx_common.h        |  3 +++
 6 files changed, 56 insertions(+), 51 deletions(-)

diff --git a/drivers/net/ethernet/sfc/mcdi_port.c b/drivers/net/ethernet/sfc/mcdi_port.c
index a5785205c32b..ab5227b13ae6 100644
--- a/drivers/net/ethernet/sfc/mcdi_port.c
+++ b/drivers/net/ethernet/sfc/mcdi_port.c
@@ -675,39 +675,6 @@ u32 efx_mcdi_phy_get_caps(struct efx_nic *efx)
 	return phy_data->supported_cap;
 }
 
-static unsigned int efx_mcdi_event_link_speed[] = {
-	[MCDI_EVENT_LINKCHANGE_SPEED_100M] = 100,
-	[MCDI_EVENT_LINKCHANGE_SPEED_1G] = 1000,
-	[MCDI_EVENT_LINKCHANGE_SPEED_10G] = 10000,
-	[MCDI_EVENT_LINKCHANGE_SPEED_40G] = 40000,
-	[MCDI_EVENT_LINKCHANGE_SPEED_25G] = 25000,
-	[MCDI_EVENT_LINKCHANGE_SPEED_50G] = 50000,
-	[MCDI_EVENT_LINKCHANGE_SPEED_100G] = 100000,
-};
-
-void efx_mcdi_process_link_change(struct efx_nic *efx, efx_qword_t *ev)
-{
-	u32 flags, fcntl, speed, lpa;
-
-	speed = EFX_QWORD_FIELD(*ev, MCDI_EVENT_LINKCHANGE_SPEED);
-	EFX_WARN_ON_PARANOID(speed >= ARRAY_SIZE(efx_mcdi_event_link_speed));
-	speed = efx_mcdi_event_link_speed[speed];
-
-	flags = EFX_QWORD_FIELD(*ev, MCDI_EVENT_LINKCHANGE_LINK_FLAGS);
-	fcntl = EFX_QWORD_FIELD(*ev, MCDI_EVENT_LINKCHANGE_FCNTL);
-	lpa = EFX_QWORD_FIELD(*ev, MCDI_EVENT_LINKCHANGE_LP_CAP);
-
-	/* efx->link_state is only modified by efx_mcdi_phy_get_link(),
-	 * which is only run after flushing the event queues. Therefore, it
-	 * is safe to modify the link state outside of the mac_lock here.
-	 */
-	efx_mcdi_phy_decode_link(efx, &efx->link_state, speed, flags, fcntl);
-
-	efx_mcdi_phy_check_fcntl(efx, lpa);
-
-	efx_link_status_changed(efx);
-}
-
 bool efx_mcdi_mac_check_fault(struct efx_nic *efx)
 {
 	MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_LINK_OUT_LEN);
diff --git a/drivers/net/ethernet/sfc/mcdi_port_common.c b/drivers/net/ethernet/sfc/mcdi_port_common.c
index 0f3237fc68af..a6a072ba46d3 100644
--- a/drivers/net/ethernet/sfc/mcdi_port_common.c
+++ b/drivers/net/ethernet/sfc/mcdi_port_common.c
@@ -9,6 +9,7 @@
  */
 
 #include "mcdi_port_common.h"
+#include "efx_common.h"
 
 int efx_mcdi_get_phy_cfg(struct efx_nic *efx, struct efx_mcdi_phy_data *cfg)
 {
@@ -532,3 +533,36 @@ int efx_mcdi_port_get_number(struct efx_nic *efx)
 
 	return MCDI_DWORD(outbuf, GET_PORT_ASSIGNMENT_OUT_PORT);
 }
+
+static unsigned int efx_mcdi_event_link_speed[] = {
+	[MCDI_EVENT_LINKCHANGE_SPEED_100M] = 100,
+	[MCDI_EVENT_LINKCHANGE_SPEED_1G] = 1000,
+	[MCDI_EVENT_LINKCHANGE_SPEED_10G] = 10000,
+	[MCDI_EVENT_LINKCHANGE_SPEED_40G] = 40000,
+	[MCDI_EVENT_LINKCHANGE_SPEED_25G] = 25000,
+	[MCDI_EVENT_LINKCHANGE_SPEED_50G] = 50000,
+	[MCDI_EVENT_LINKCHANGE_SPEED_100G] = 100000,
+};
+
+void efx_mcdi_process_link_change(struct efx_nic *efx, efx_qword_t *ev)
+{
+	u32 flags, fcntl, speed, lpa;
+
+	speed = EFX_QWORD_FIELD(*ev, MCDI_EVENT_LINKCHANGE_SPEED);
+	EFX_WARN_ON_PARANOID(speed >= ARRAY_SIZE(efx_mcdi_event_link_speed));
+	speed = efx_mcdi_event_link_speed[speed];
+
+	flags = EFX_QWORD_FIELD(*ev, MCDI_EVENT_LINKCHANGE_LINK_FLAGS);
+	fcntl = EFX_QWORD_FIELD(*ev, MCDI_EVENT_LINKCHANGE_FCNTL);
+	lpa = EFX_QWORD_FIELD(*ev, MCDI_EVENT_LINKCHANGE_LP_CAP);
+
+	/* efx->link_state is only modified by efx_mcdi_phy_get_link(),
+	 * which is only run after flushing the event queues. Therefore, it
+	 * is safe to modify the link state outside of the mac_lock here.
+	 */
+	efx_mcdi_phy_decode_link(efx, &efx->link_state, speed, flags, fcntl);
+
+	efx_mcdi_phy_check_fcntl(efx, lpa);
+
+	efx_link_status_changed(efx);
+}
diff --git a/drivers/net/ethernet/sfc/mcdi_port_common.h b/drivers/net/ethernet/sfc/mcdi_port_common.h
index 6b08a2b7eddf..b16f11265269 100644
--- a/drivers/net/ethernet/sfc/mcdi_port_common.h
+++ b/drivers/net/ethernet/sfc/mcdi_port_common.h
@@ -52,5 +52,6 @@ int efx_mcdi_phy_get_fecparam(struct efx_nic *efx,
 int efx_mcdi_phy_test_alive(struct efx_nic *efx);
 int efx_mcdi_set_mac(struct efx_nic *efx);
 int efx_mcdi_port_get_number(struct efx_nic *efx);
+void efx_mcdi_process_link_change(struct efx_nic *efx, efx_qword_t *ev);
 
 #endif
diff --git a/drivers/net/ethernet/sfc/tx.c b/drivers/net/ethernet/sfc/tx.c
index 696e12c1d9e7..04d7f41d7ed9 100644
--- a/drivers/net/ethernet/sfc/tx.c
+++ b/drivers/net/ethernet/sfc/tx.c
@@ -268,24 +268,6 @@ static int efx_enqueue_skb_pio(struct efx_tx_queue *tx_queue,
 }
 #endif /* EFX_USE_PIO */
 
-/* Remove buffers put into a tx_queue for the current packet.
- * None of the buffers must have an skb attached.
- */
-static void efx_enqueue_unwind(struct efx_tx_queue *tx_queue,
-			       unsigned int insert_count)
-{
-	struct efx_tx_buffer *buffer;
-	unsigned int bytes_compl = 0;
-	unsigned int pkts_compl = 0;
-
-	/* Work backwards until we hit the original insert pointer value */
-	while (tx_queue->insert_count != insert_count) {
-		--tx_queue->insert_count;
-		buffer = __efx_tx_queue_get_insert_buffer(tx_queue);
-		efx_dequeue_buffer(tx_queue, buffer, &pkts_compl, &bytes_compl);
-	}
-}
-
 /*
  * Fallback to software TSO.
  *
diff --git a/drivers/net/ethernet/sfc/tx_common.c b/drivers/net/ethernet/sfc/tx_common.c
index 0ce699ecae5a..b1571e9789d0 100644
--- a/drivers/net/ethernet/sfc/tx_common.c
+++ b/drivers/net/ethernet/sfc/tx_common.c
@@ -267,6 +267,24 @@ void efx_xmit_done(struct efx_tx_queue *tx_queue, unsigned int index)
 	}
 }
 
+/* Remove buffers put into a tx_queue for the current packet.
+ * None of the buffers must have an skb attached.
+ */
+void efx_enqueue_unwind(struct efx_tx_queue *tx_queue,
+			unsigned int insert_count)
+{
+	struct efx_tx_buffer *buffer;
+	unsigned int bytes_compl = 0;
+	unsigned int pkts_compl = 0;
+
+	/* Work backwards until we hit the original insert pointer value */
+	while (tx_queue->insert_count != insert_count) {
+		--tx_queue->insert_count;
+		buffer = __efx_tx_queue_get_insert_buffer(tx_queue);
+		efx_dequeue_buffer(tx_queue, buffer, &pkts_compl, &bytes_compl);
+	}
+}
+
 struct efx_tx_buffer *efx_tx_map_chunk(struct efx_tx_queue *tx_queue,
 				       dma_addr_t dma_addr, size_t len)
 {
diff --git a/drivers/net/ethernet/sfc/tx_common.h b/drivers/net/ethernet/sfc/tx_common.h
index 58add94ab500..f92f1fe3a87f 100644
--- a/drivers/net/ethernet/sfc/tx_common.h
+++ b/drivers/net/ethernet/sfc/tx_common.h
@@ -23,6 +23,9 @@ void efx_dequeue_buffer(struct efx_tx_queue *tx_queue,
 
 void efx_xmit_done(struct efx_tx_queue *tx_queue, unsigned int index);
 
+void efx_enqueue_unwind(struct efx_tx_queue *tx_queue,
+			unsigned int insert_count);
+
 struct efx_tx_buffer *efx_tx_map_chunk(struct efx_tx_queue *tx_queue,
 				       dma_addr_t dma_addr, size_t len);
 int efx_tx_map_data(struct efx_tx_queue *tx_queue, struct sk_buff *skb,
-- 
cgit v1.2.3-59-g8ed1b


From 960f16275bdf87d7bef59e5d446d0a2913b5f24a Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Fri, 10 Jan 2020 13:28:07 +0000
Subject: sfc: move RSS code

Style fixes included.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/efx.c       | 66 +-----------------------------------
 drivers/net/ethernet/sfc/efx.h       |  5 ---
 drivers/net/ethernet/sfc/rx_common.c | 65 +++++++++++++++++++++++++++++++++++
 drivers/net/ethernet/sfc/rx_common.h |  6 ++++
 4 files changed, 72 insertions(+), 70 deletions(-)

diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index 7ad97090ab52..a997f3d720ab 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -27,6 +27,7 @@
 #include "efx_channels.h"
 #include "rx_common.h"
 #include "tx_common.h"
+#include "rx_common.h"
 #include "nic.h"
 #include "io.h"
 #include "selftest.h"
@@ -311,16 +312,6 @@ static void efx_dissociate(struct efx_nic *efx)
 	}
 }
 
-void efx_set_default_rx_indir_table(struct efx_nic *efx,
-				    struct efx_rss_context *ctx)
-{
-	size_t i;
-
-	for (i = 0; i < ARRAY_SIZE(ctx->rx_indir_table); i++)
-		ctx->rx_indir_table[i] =
-			ethtool_rxfh_indir_default(i, efx->rss_spread);
-}
-
 static int efx_probe_nic(struct efx_nic *efx)
 {
 	int rc;
@@ -1299,61 +1290,6 @@ void efx_rps_hash_del(struct efx_nic *efx, const struct efx_filter_spec *spec)
 }
 #endif
 
-/* RSS contexts.  We're using linked lists and crappy O(n) algorithms, because
- * (a) this is an infrequent control-plane operation and (b) n is small (max 64)
- */
-struct efx_rss_context *efx_alloc_rss_context_entry(struct efx_nic *efx)
-{
-	struct list_head *head = &efx->rss_context.list;
-	struct efx_rss_context *ctx, *new;
-	u32 id = 1; /* Don't use zero, that refers to the master RSS context */
-
-	WARN_ON(!mutex_is_locked(&efx->rss_lock));
-
-	/* Search for first gap in the numbering */
-	list_for_each_entry(ctx, head, list) {
-		if (ctx->user_id != id)
-			break;
-		id++;
-		/* Check for wrap.  If this happens, we have nearly 2^32
-		 * allocated RSS contexts, which seems unlikely.
-		 */
-		if (WARN_ON_ONCE(!id))
-			return NULL;
-	}
-
-	/* Create the new entry */
-	new = kmalloc(sizeof(struct efx_rss_context), GFP_KERNEL);
-	if (!new)
-		return NULL;
-	new->context_id = EFX_EF10_RSS_CONTEXT_INVALID;
-	new->rx_hash_udp_4tuple = false;
-
-	/* Insert the new entry into the gap */
-	new->user_id = id;
-	list_add_tail(&new->list, &ctx->list);
-	return new;
-}
-
-struct efx_rss_context *efx_find_rss_context_entry(struct efx_nic *efx, u32 id)
-{
-	struct list_head *head = &efx->rss_context.list;
-	struct efx_rss_context *ctx;
-
-	WARN_ON(!mutex_is_locked(&efx->rss_lock));
-
-	list_for_each_entry(ctx, head, list)
-		if (ctx->user_id == id)
-			return ctx;
-	return NULL;
-}
-
-void efx_free_rss_context_entry(struct efx_rss_context *ctx)
-{
-	list_del(&ctx->list);
-	kfree(ctx);
-}
-
 /**************************************************************************
  *
  * PCI interface
diff --git a/drivers/net/ethernet/sfc/efx.h b/drivers/net/ethernet/sfc/efx.h
index 2b417e779e82..4db56b356f11 100644
--- a/drivers/net/ethernet/sfc/efx.h
+++ b/drivers/net/ethernet/sfc/efx.h
@@ -26,8 +26,6 @@ extern unsigned int efx_piobuf_size;
 extern bool efx_separate_tx_channels;
 
 /* RX */
-void efx_set_default_rx_indir_table(struct efx_nic *efx,
-				    struct efx_rss_context *ctx);
 void __efx_rx_packet(struct efx_channel *channel);
 void efx_rx_packet(struct efx_rx_queue *rx_queue, unsigned int index,
 		   unsigned int n_frags, unsigned int len, u16 flags);
@@ -195,9 +193,6 @@ void efx_rps_hash_del(struct efx_nic *efx, const struct efx_filter_spec *spec);
 #endif
 
 /* RSS contexts */
-struct efx_rss_context *efx_alloc_rss_context_entry(struct efx_nic *efx);
-struct efx_rss_context *efx_find_rss_context_entry(struct efx_nic *efx, u32 id);
-void efx_free_rss_context_entry(struct efx_rss_context *ctx);
 static inline bool efx_rss_active(struct efx_rss_context *ctx)
 {
 	return ctx->context_id != EFX_EF10_RSS_CONTEXT_INVALID;
diff --git a/drivers/net/ethernet/sfc/rx_common.c b/drivers/net/ethernet/sfc/rx_common.c
index 9c545d873d91..cdf21e2d5610 100644
--- a/drivers/net/ethernet/sfc/rx_common.c
+++ b/drivers/net/ethernet/sfc/rx_common.c
@@ -551,3 +551,68 @@ efx_rx_packet_gro(struct efx_channel *channel, struct efx_rx_buffer *rx_buf,
 
 	napi_gro_frags(napi);
 }
+
+/* RSS contexts.  We're using linked lists and crappy O(n) algorithms, because
+ * (a) this is an infrequent control-plane operation and (b) n is small (max 64)
+ */
+struct efx_rss_context *efx_alloc_rss_context_entry(struct efx_nic *efx)
+{
+	struct list_head *head = &efx->rss_context.list;
+	struct efx_rss_context *ctx, *new;
+	u32 id = 1; /* Don't use zero, that refers to the master RSS context */
+
+	WARN_ON(!mutex_is_locked(&efx->rss_lock));
+
+	/* Search for first gap in the numbering */
+	list_for_each_entry(ctx, head, list) {
+		if (ctx->user_id != id)
+			break;
+		id++;
+		/* Check for wrap.  If this happens, we have nearly 2^32
+		 * allocated RSS contexts, which seems unlikely.
+		 */
+		if (WARN_ON_ONCE(!id))
+			return NULL;
+	}
+
+	/* Create the new entry */
+	new = kmalloc(sizeof(*new), GFP_KERNEL);
+	if (!new)
+		return NULL;
+	new->context_id = EFX_EF10_RSS_CONTEXT_INVALID;
+	new->rx_hash_udp_4tuple = false;
+
+	/* Insert the new entry into the gap */
+	new->user_id = id;
+	list_add_tail(&new->list, &ctx->list);
+	return new;
+}
+
+struct efx_rss_context *efx_find_rss_context_entry(struct efx_nic *efx, u32 id)
+{
+	struct list_head *head = &efx->rss_context.list;
+	struct efx_rss_context *ctx;
+
+	WARN_ON(!mutex_is_locked(&efx->rss_lock));
+
+	list_for_each_entry(ctx, head, list)
+		if (ctx->user_id == id)
+			return ctx;
+	return NULL;
+}
+
+void efx_free_rss_context_entry(struct efx_rss_context *ctx)
+{
+	list_del(&ctx->list);
+	kfree(ctx);
+}
+
+void efx_set_default_rx_indir_table(struct efx_nic *efx,
+				    struct efx_rss_context *ctx)
+{
+	size_t i;
+
+	for (i = 0; i < ARRAY_SIZE(ctx->rx_indir_table); i++)
+		ctx->rx_indir_table[i] =
+			ethtool_rxfh_indir_default(i, efx->rss_spread);
+}
diff --git a/drivers/net/ethernet/sfc/rx_common.h b/drivers/net/ethernet/sfc/rx_common.h
index aa6aeafbc439..4d2e7d38e260 100644
--- a/drivers/net/ethernet/sfc/rx_common.h
+++ b/drivers/net/ethernet/sfc/rx_common.h
@@ -68,4 +68,10 @@ void efx_fast_push_rx_descriptors(struct efx_rx_queue *rx_queue, bool atomic);
 void
 efx_rx_packet_gro(struct efx_channel *channel, struct efx_rx_buffer *rx_buf,
 		  unsigned int n_frags, u8 *eh);
+
+struct efx_rss_context *efx_alloc_rss_context_entry(struct efx_nic *efx);
+struct efx_rss_context *efx_find_rss_context_entry(struct efx_nic *efx, u32 id);
+void efx_free_rss_context_entry(struct efx_rss_context *ctx);
+void efx_set_default_rx_indir_table(struct efx_nic *efx,
+				    struct efx_rss_context *ctx);
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From 314823108c9dc3a54d041d9b7251934859c853f1 Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Fri, 10 Jan 2020 13:28:23 +0000
Subject: sfc: move yet more functions

Functions are not related.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/ef10.c           | 20 +-------------
 drivers/net/ethernet/sfc/efx.c            | 45 -------------------------------
 drivers/net/ethernet/sfc/efx_common.c     | 45 +++++++++++++++++++++++++++++++
 drivers/net/ethernet/sfc/efx_common.h     |  2 ++
 drivers/net/ethernet/sfc/mcdi_functions.c | 17 ++++++++++++
 drivers/net/ethernet/sfc/mcdi_functions.h |  1 +
 6 files changed, 66 insertions(+), 64 deletions(-)

diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c
index 6853b5cb3ac8..06045e181c8f 100644
--- a/drivers/net/ethernet/sfc/ef10.c
+++ b/drivers/net/ethernet/sfc/ef10.c
@@ -189,24 +189,6 @@ static bool efx_ef10_is_vf(struct efx_nic *efx)
 	return efx->type->is_vf;
 }
 
-static int efx_ef10_get_pf_index(struct efx_nic *efx)
-{
-	MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_FUNCTION_INFO_OUT_LEN);
-	struct efx_ef10_nic_data *nic_data = efx->nic_data;
-	size_t outlen;
-	int rc;
-
-	rc = efx_mcdi_rpc(efx, MC_CMD_GET_FUNCTION_INFO, NULL, 0, outbuf,
-			  sizeof(outbuf), &outlen);
-	if (rc)
-		return rc;
-	if (outlen < sizeof(outbuf))
-		return -EIO;
-
-	nic_data->pf_index = MCDI_DWORD(outbuf, GET_FUNCTION_INFO_OUT_PF);
-	return 0;
-}
-
 #ifdef CONFIG_SFC_SRIOV
 static int efx_ef10_get_vf_index(struct efx_nic *efx)
 {
@@ -714,7 +696,7 @@ static int efx_ef10_probe(struct efx_nic *efx)
 	if (rc)
 		goto fail4;
 
-	rc = efx_ef10_get_pf_index(efx);
+	rc = efx_get_pf_index(efx, &nic_data->pf_index);
 	if (rc)
 		goto fail5;
 
diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index a997f3d720ab..87b784c207bd 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -695,51 +695,6 @@ static void efx_watchdog(struct net_device *net_dev, unsigned int txqueue)
 	efx_schedule_reset(efx, RESET_TYPE_TX_WATCHDOG);
 }
 
-static unsigned int efx_xdp_max_mtu(struct efx_nic *efx)
-{
-	/* The maximum MTU that we can fit in a single page, allowing for
-	 * framing, overhead and XDP headroom.
-	 */
-	int overhead = EFX_MAX_FRAME_LEN(0) + sizeof(struct efx_rx_page_state) +
-		       efx->rx_prefix_size + efx->type->rx_buffer_padding +
-		       efx->rx_ip_align + XDP_PACKET_HEADROOM;
-
-	return PAGE_SIZE - overhead;
-}
-
-/* Context: process, rtnl_lock() held. */
-static int efx_change_mtu(struct net_device *net_dev, int new_mtu)
-{
-	struct efx_nic *efx = netdev_priv(net_dev);
-	int rc;
-
-	rc = efx_check_disabled(efx);
-	if (rc)
-		return rc;
-
-	if (rtnl_dereference(efx->xdp_prog) &&
-	    new_mtu > efx_xdp_max_mtu(efx)) {
-		netif_err(efx, drv, efx->net_dev,
-			  "Requested MTU of %d too big for XDP (max: %d)\n",
-			  new_mtu, efx_xdp_max_mtu(efx));
-		return -EINVAL;
-	}
-
-	netif_dbg(efx, drv, efx->net_dev, "changing MTU to %d\n", new_mtu);
-
-	efx_device_detach_sync(efx);
-	efx_stop_all(efx);
-
-	mutex_lock(&efx->mac_lock);
-	net_dev->mtu = new_mtu;
-	efx_mac_reconfigure(efx);
-	mutex_unlock(&efx->mac_lock);
-
-	efx_start_all(efx);
-	efx_device_attach_if_not_resetting(efx);
-	return 0;
-}
-
 static int efx_set_mac_address(struct net_device *net_dev, void *data)
 {
 	struct efx_nic *efx = netdev_priv(net_dev);
diff --git a/drivers/net/ethernet/sfc/efx_common.c b/drivers/net/ethernet/sfc/efx_common.c
index 0875507efd0a..ab0ce62f81c1 100644
--- a/drivers/net/ethernet/sfc/efx_common.c
+++ b/drivers/net/ethernet/sfc/efx_common.c
@@ -197,6 +197,51 @@ void efx_link_status_changed(struct efx_nic *efx)
 		netif_info(efx, link, efx->net_dev, "link down\n");
 }
 
+unsigned int efx_xdp_max_mtu(struct efx_nic *efx)
+{
+	/* The maximum MTU that we can fit in a single page, allowing for
+	 * framing, overhead and XDP headroom.
+	 */
+	int overhead = EFX_MAX_FRAME_LEN(0) + sizeof(struct efx_rx_page_state) +
+		       efx->rx_prefix_size + efx->type->rx_buffer_padding +
+		       efx->rx_ip_align + XDP_PACKET_HEADROOM;
+
+	return PAGE_SIZE - overhead;
+}
+
+/* Context: process, rtnl_lock() held. */
+int efx_change_mtu(struct net_device *net_dev, int new_mtu)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	int rc;
+
+	rc = efx_check_disabled(efx);
+	if (rc)
+		return rc;
+
+	if (rtnl_dereference(efx->xdp_prog) &&
+	    new_mtu > efx_xdp_max_mtu(efx)) {
+		netif_err(efx, drv, efx->net_dev,
+			  "Requested MTU of %d too big for XDP (max: %d)\n",
+			  new_mtu, efx_xdp_max_mtu(efx));
+		return -EINVAL;
+	}
+
+	netif_dbg(efx, drv, efx->net_dev, "changing MTU to %d\n", new_mtu);
+
+	efx_device_detach_sync(efx);
+	efx_stop_all(efx);
+
+	mutex_lock(&efx->mac_lock);
+	net_dev->mtu = new_mtu;
+	efx_mac_reconfigure(efx);
+	mutex_unlock(&efx->mac_lock);
+
+	efx_start_all(efx);
+	efx_device_attach_if_not_resetting(efx);
+	return 0;
+}
+
 /**************************************************************************
  *
  * Hardware monitor
diff --git a/drivers/net/ethernet/sfc/efx_common.h b/drivers/net/ethernet/sfc/efx_common.h
index 32a23ec9b104..fa2fc681e7f9 100644
--- a/drivers/net/ethernet/sfc/efx_common.h
+++ b/drivers/net/ethernet/sfc/efx_common.h
@@ -67,5 +67,7 @@ static inline void efx_fini_mcdi_logging(struct efx_nic *efx) {}
 
 void efx_mac_reconfigure(struct efx_nic *efx);
 void efx_link_status_changed(struct efx_nic *efx);
+unsigned int efx_xdp_max_mtu(struct efx_nic *efx);
+int efx_change_mtu(struct net_device *net_dev, int new_mtu);
 
 #endif
diff --git a/drivers/net/ethernet/sfc/mcdi_functions.c b/drivers/net/ethernet/sfc/mcdi_functions.c
index 0f571864df22..f5dad8bffc3d 100644
--- a/drivers/net/ethernet/sfc/mcdi_functions.c
+++ b/drivers/net/ethernet/sfc/mcdi_functions.c
@@ -370,3 +370,20 @@ int efx_mcdi_window_mode_to_stride(struct efx_nic *efx, u8 vi_window_mode)
 		  efx->vi_stride);
 	return 0;
 }
+
+int efx_get_pf_index(struct efx_nic *efx, unsigned int *pf_index)
+{
+	MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_FUNCTION_INFO_OUT_LEN);
+	size_t outlen;
+	int rc;
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_GET_FUNCTION_INFO, NULL, 0, outbuf,
+			  sizeof(outbuf), &outlen);
+	if (rc)
+		return rc;
+	if (outlen < sizeof(outbuf))
+		return -EIO;
+
+	*pf_index = MCDI_DWORD(outbuf, GET_FUNCTION_INFO_OUT_PF);
+	return 0;
+}
diff --git a/drivers/net/ethernet/sfc/mcdi_functions.h b/drivers/net/ethernet/sfc/mcdi_functions.h
index 0396294bfdcd..ca4a5ac1a66b 100644
--- a/drivers/net/ethernet/sfc/mcdi_functions.h
+++ b/drivers/net/ethernet/sfc/mcdi_functions.h
@@ -27,5 +27,6 @@ void efx_mcdi_rx_init(struct efx_rx_queue *rx_queue);
 void efx_mcdi_rx_remove(struct efx_rx_queue *rx_queue);
 void efx_mcdi_rx_fini(struct efx_rx_queue *rx_queue);
 int efx_mcdi_window_mode_to_stride(struct efx_nic *efx, u8 vi_window_mode);
+int efx_get_pf_index(struct efx_nic *efx, unsigned int *pf_index);
 
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From f7226e0f9b64b0454d1a5a32f2553aa677fc5400 Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Fri, 10 Jan 2020 13:28:45 +0000
Subject: sfc: move RPS code

Includes a couple of filtering functions and also renames a constant.

Style fixes included.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/ef10.c       |  28 ++--
 drivers/net/ethernet/sfc/efx.c        | 200 -----------------------------
 drivers/net/ethernet/sfc/efx.h        |  24 +---
 drivers/net/ethernet/sfc/ethtool.c    |   2 +-
 drivers/net/ethernet/sfc/net_driver.h |   6 +-
 drivers/net/ethernet/sfc/rx.c         |  34 -----
 drivers/net/ethernet/sfc/rx_common.c  | 235 +++++++++++++++++++++++++++++++++-
 drivers/net/ethernet/sfc/rx_common.h  |  20 +++
 8 files changed, 273 insertions(+), 276 deletions(-)

diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c
index 06045e181c8f..82063212cac4 100644
--- a/drivers/net/ethernet/sfc/ef10.c
+++ b/drivers/net/ethernet/sfc/ef10.c
@@ -660,7 +660,7 @@ static int efx_ef10_probe(struct efx_nic *efx)
 	}
 	nic_data->warm_boot_count = rc;
 
-	efx->rss_context.context_id = EFX_EF10_RSS_CONTEXT_INVALID;
+	efx->rss_context.context_id = EFX_MCDI_RSS_CONTEXT_INVALID;
 
 	nic_data->vport_id = EVB_PORT_ID_ASSIGNED;
 
@@ -1440,7 +1440,7 @@ static void efx_ef10_reset_mc_allocations(struct efx_nic *efx)
 	nic_data->must_restore_filters = true;
 	nic_data->must_restore_piobufs = true;
 	efx_ef10_forget_old_piobufs(efx);
-	efx->rss_context.context_id = EFX_EF10_RSS_CONTEXT_INVALID;
+	efx->rss_context.context_id = EFX_MCDI_RSS_CONTEXT_INVALID;
 
 	/* Driver-created vswitches and vports must be re-created */
 	nic_data->must_probe_vswitching = true;
@@ -2598,7 +2598,7 @@ static int efx_ef10_alloc_rss_context(struct efx_nic *efx, bool exclusive,
 				    EFX_EF10_MAX_SHARED_RSS_CONTEXT_SIZE);
 
 	if (!exclusive && rss_spread == 1) {
-		ctx->context_id = EFX_EF10_RSS_CONTEXT_INVALID;
+		ctx->context_id = EFX_MCDI_RSS_CONTEXT_INVALID;
 		if (context_size)
 			*context_size = 1;
 		return 0;
@@ -2685,11 +2685,11 @@ static void efx_ef10_rx_free_indir_table(struct efx_nic *efx)
 {
 	int rc;
 
-	if (efx->rss_context.context_id != EFX_EF10_RSS_CONTEXT_INVALID) {
+	if (efx->rss_context.context_id != EFX_MCDI_RSS_CONTEXT_INVALID) {
 		rc = efx_ef10_free_rss_context(efx, efx->rss_context.context_id);
 		WARN_ON(rc != 0);
 	}
-	efx->rss_context.context_id = EFX_EF10_RSS_CONTEXT_INVALID;
+	efx->rss_context.context_id = EFX_MCDI_RSS_CONTEXT_INVALID;
 }
 
 static int efx_ef10_rx_push_shared_rss_config(struct efx_nic *efx,
@@ -2715,7 +2715,7 @@ static int efx_ef10_rx_push_exclusive_rss_config(struct efx_nic *efx,
 	struct efx_ef10_nic_data *nic_data = efx->nic_data;
 	int rc;
 
-	if (efx->rss_context.context_id == EFX_EF10_RSS_CONTEXT_INVALID ||
+	if (efx->rss_context.context_id == EFX_MCDI_RSS_CONTEXT_INVALID ||
 	    !nic_data->rx_rss_context_exclusive) {
 		rc = efx_ef10_alloc_rss_context(efx, true, &efx->rss_context,
 						NULL);
@@ -2731,7 +2731,7 @@ static int efx_ef10_rx_push_exclusive_rss_config(struct efx_nic *efx,
 		goto fail2;
 
 	if (efx->rss_context.context_id != old_rx_rss_context &&
-	    old_rx_rss_context != EFX_EF10_RSS_CONTEXT_INVALID)
+	    old_rx_rss_context != EFX_MCDI_RSS_CONTEXT_INVALID)
 		WARN_ON(efx_ef10_free_rss_context(efx, old_rx_rss_context) != 0);
 	nic_data->rx_rss_context_exclusive = true;
 	if (rx_indir_table != efx->rss_context.rx_indir_table)
@@ -2762,7 +2762,7 @@ static int efx_ef10_rx_push_rss_context_config(struct efx_nic *efx,
 
 	WARN_ON(!mutex_is_locked(&efx->rss_lock));
 
-	if (ctx->context_id == EFX_EF10_RSS_CONTEXT_INVALID) {
+	if (ctx->context_id == EFX_MCDI_RSS_CONTEXT_INVALID) {
 		rc = efx_ef10_alloc_rss_context(efx, true, ctx, NULL);
 		if (rc)
 			return rc;
@@ -2797,7 +2797,7 @@ static int efx_ef10_rx_pull_rss_context_config(struct efx_nic *efx,
 	BUILD_BUG_ON(MC_CMD_RSS_CONTEXT_GET_TABLE_IN_LEN !=
 		     MC_CMD_RSS_CONTEXT_GET_KEY_IN_LEN);
 
-	if (ctx->context_id == EFX_EF10_RSS_CONTEXT_INVALID)
+	if (ctx->context_id == EFX_MCDI_RSS_CONTEXT_INVALID)
 		return -ENOENT;
 
 	MCDI_SET_DWORD(inbuf, RSS_CONTEXT_GET_TABLE_IN_RSS_CONTEXT_ID,
@@ -2858,7 +2858,7 @@ static void efx_ef10_rx_restore_rss_contexts(struct efx_nic *efx)
 
 	list_for_each_entry(ctx, &efx->rss_context.list, list) {
 		/* previous NIC RSS context is gone */
-		ctx->context_id = EFX_EF10_RSS_CONTEXT_INVALID;
+		ctx->context_id = EFX_MCDI_RSS_CONTEXT_INVALID;
 		/* so try to allocate a new one */
 		rc = efx_ef10_rx_push_rss_context_config(efx, ctx,
 							 ctx->rx_indir_table,
@@ -2929,7 +2929,7 @@ static int efx_ef10_vf_rx_push_rss_config(struct efx_nic *efx, bool user,
 {
 	if (user)
 		return -EOPNOTSUPP;
-	if (efx->rss_context.context_id != EFX_EF10_RSS_CONTEXT_INVALID)
+	if (efx->rss_context.context_id != EFX_MCDI_RSS_CONTEXT_INVALID)
 		return 0;
 	return efx_ef10_rx_push_shared_rss_config(efx, NULL);
 }
@@ -3850,7 +3850,7 @@ static void efx_ef10_filter_push_prep(struct efx_nic *efx,
 		 */
 		if (WARN_ON_ONCE(!ctx))
 			flags &= ~EFX_FILTER_FLAG_RX_RSS;
-		else if (WARN_ON_ONCE(ctx->context_id == EFX_EF10_RSS_CONTEXT_INVALID))
+		else if (WARN_ON_ONCE(ctx->context_id == EFX_MCDI_RSS_CONTEXT_INVALID))
 			flags &= ~EFX_FILTER_FLAG_RX_RSS;
 	}
 
@@ -4029,7 +4029,7 @@ static s32 efx_ef10_filter_insert_locked(struct efx_nic *efx,
 			rc = -ENOENT;
 			goto out_unlock;
 		}
-		if (ctx->context_id == EFX_EF10_RSS_CONTEXT_INVALID) {
+		if (ctx->context_id == EFX_MCDI_RSS_CONTEXT_INVALID) {
 			rc = -EOPNOTSUPP;
 			goto out_unlock;
 		}
@@ -4770,7 +4770,7 @@ static void efx_ef10_filter_table_restore(struct efx_nic *efx)
 				invalid_filters++;
 				goto not_restored;
 			}
-			if (ctx->context_id == EFX_EF10_RSS_CONTEXT_INVALID) {
+			if (ctx->context_id == EFX_MCDI_RSS_CONTEXT_INVALID) {
 				netif_warn(efx, drv, efx->net_dev,
 					   "Warning: unable to restore a filter with RSS context %u as it was not created.\n",
 					   spec->rss_context);
diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index 87b784c207bd..110e485e6624 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -384,70 +384,6 @@ static void efx_remove_nic(struct efx_nic *efx)
 	efx->type->remove(efx);
 }
 
-static int efx_probe_filters(struct efx_nic *efx)
-{
-	int rc;
-
-	init_rwsem(&efx->filter_sem);
-	mutex_lock(&efx->mac_lock);
-	down_write(&efx->filter_sem);
-	rc = efx->type->filter_table_probe(efx);
-	if (rc)
-		goto out_unlock;
-
-#ifdef CONFIG_RFS_ACCEL
-	if (efx->type->offload_features & NETIF_F_NTUPLE) {
-		struct efx_channel *channel;
-		int i, success = 1;
-
-		efx_for_each_channel(channel, efx) {
-			channel->rps_flow_id =
-				kcalloc(efx->type->max_rx_ip_filters,
-					sizeof(*channel->rps_flow_id),
-					GFP_KERNEL);
-			if (!channel->rps_flow_id)
-				success = 0;
-			else
-				for (i = 0;
-				     i < efx->type->max_rx_ip_filters;
-				     ++i)
-					channel->rps_flow_id[i] =
-						RPS_FLOW_ID_INVALID;
-			channel->rfs_expire_index = 0;
-			channel->rfs_filter_count = 0;
-		}
-
-		if (!success) {
-			efx_for_each_channel(channel, efx)
-				kfree(channel->rps_flow_id);
-			efx->type->filter_table_remove(efx);
-			rc = -ENOMEM;
-			goto out_unlock;
-		}
-	}
-#endif
-out_unlock:
-	up_write(&efx->filter_sem);
-	mutex_unlock(&efx->mac_lock);
-	return rc;
-}
-
-static void efx_remove_filters(struct efx_nic *efx)
-{
-#ifdef CONFIG_RFS_ACCEL
-	struct efx_channel *channel;
-
-	efx_for_each_channel(channel, efx) {
-		cancel_delayed_work_sync(&channel->filter_work);
-		kfree(channel->rps_flow_id);
-	}
-#endif
-	down_write(&efx->filter_sem);
-	efx->type->filter_table_remove(efx);
-	up_write(&efx->filter_sem);
-}
-
-
 /**************************************************************************
  *
  * NIC startup/shutdown
@@ -1109,142 +1045,6 @@ void efx_update_sw_stats(struct efx_nic *efx, u64 *stats)
 	stats[GENERIC_STAT_rx_noskb_drops] = atomic_read(&efx->n_rx_noskb_drops);
 }
 
-bool efx_filter_spec_equal(const struct efx_filter_spec *left,
-			   const struct efx_filter_spec *right)
-{
-	if ((left->match_flags ^ right->match_flags) |
-	    ((left->flags ^ right->flags) &
-	     (EFX_FILTER_FLAG_RX | EFX_FILTER_FLAG_TX)))
-		return false;
-
-	return memcmp(&left->outer_vid, &right->outer_vid,
-		      sizeof(struct efx_filter_spec) -
-		      offsetof(struct efx_filter_spec, outer_vid)) == 0;
-}
-
-u32 efx_filter_spec_hash(const struct efx_filter_spec *spec)
-{
-	BUILD_BUG_ON(offsetof(struct efx_filter_spec, outer_vid) & 3);
-	return jhash2((const u32 *)&spec->outer_vid,
-		      (sizeof(struct efx_filter_spec) -
-		       offsetof(struct efx_filter_spec, outer_vid)) / 4,
-		      0);
-}
-
-#ifdef CONFIG_RFS_ACCEL
-bool efx_rps_check_rule(struct efx_arfs_rule *rule, unsigned int filter_idx,
-			bool *force)
-{
-	if (rule->filter_id == EFX_ARFS_FILTER_ID_PENDING) {
-		/* ARFS is currently updating this entry, leave it */
-		return false;
-	}
-	if (rule->filter_id == EFX_ARFS_FILTER_ID_ERROR) {
-		/* ARFS tried and failed to update this, so it's probably out
-		 * of date.  Remove the filter and the ARFS rule entry.
-		 */
-		rule->filter_id = EFX_ARFS_FILTER_ID_REMOVING;
-		*force = true;
-		return true;
-	} else if (WARN_ON(rule->filter_id != filter_idx)) { /* can't happen */
-		/* ARFS has moved on, so old filter is not needed.  Since we did
-		 * not mark the rule with EFX_ARFS_FILTER_ID_REMOVING, it will
-		 * not be removed by efx_rps_hash_del() subsequently.
-		 */
-		*force = true;
-		return true;
-	}
-	/* Remove it iff ARFS wants to. */
-	return true;
-}
-
-static
-struct hlist_head *efx_rps_hash_bucket(struct efx_nic *efx,
-				       const struct efx_filter_spec *spec)
-{
-	u32 hash = efx_filter_spec_hash(spec);
-
-	lockdep_assert_held(&efx->rps_hash_lock);
-	if (!efx->rps_hash_table)
-		return NULL;
-	return &efx->rps_hash_table[hash % EFX_ARFS_HASH_TABLE_SIZE];
-}
-
-struct efx_arfs_rule *efx_rps_hash_find(struct efx_nic *efx,
-					const struct efx_filter_spec *spec)
-{
-	struct efx_arfs_rule *rule;
-	struct hlist_head *head;
-	struct hlist_node *node;
-
-	head = efx_rps_hash_bucket(efx, spec);
-	if (!head)
-		return NULL;
-	hlist_for_each(node, head) {
-		rule = container_of(node, struct efx_arfs_rule, node);
-		if (efx_filter_spec_equal(spec, &rule->spec))
-			return rule;
-	}
-	return NULL;
-}
-
-struct efx_arfs_rule *efx_rps_hash_add(struct efx_nic *efx,
-				       const struct efx_filter_spec *spec,
-				       bool *new)
-{
-	struct efx_arfs_rule *rule;
-	struct hlist_head *head;
-	struct hlist_node *node;
-
-	head = efx_rps_hash_bucket(efx, spec);
-	if (!head)
-		return NULL;
-	hlist_for_each(node, head) {
-		rule = container_of(node, struct efx_arfs_rule, node);
-		if (efx_filter_spec_equal(spec, &rule->spec)) {
-			*new = false;
-			return rule;
-		}
-	}
-	rule = kmalloc(sizeof(*rule), GFP_ATOMIC);
-	*new = true;
-	if (rule) {
-		memcpy(&rule->spec, spec, sizeof(rule->spec));
-		hlist_add_head(&rule->node, head);
-	}
-	return rule;
-}
-
-void efx_rps_hash_del(struct efx_nic *efx, const struct efx_filter_spec *spec)
-{
-	struct efx_arfs_rule *rule;
-	struct hlist_head *head;
-	struct hlist_node *node;
-
-	head = efx_rps_hash_bucket(efx, spec);
-	if (WARN_ON(!head))
-		return;
-	hlist_for_each(node, head) {
-		rule = container_of(node, struct efx_arfs_rule, node);
-		if (efx_filter_spec_equal(spec, &rule->spec)) {
-			/* Someone already reused the entry.  We know that if
-			 * this check doesn't fire (i.e. filter_id == REMOVING)
-			 * then the REMOVING mark was put there by our caller,
-			 * because caller is holding a lock on filter table and
-			 * only holders of that lock set REMOVING.
-			 */
-			if (rule->filter_id != EFX_ARFS_FILTER_ID_REMOVING)
-				return;
-			hlist_del(node);
-			kfree(rule);
-			return;
-		}
-	}
-	/* We didn't find it. */
-	WARN_ON(1);
-}
-#endif
-
 /**************************************************************************
  *
  * PCI interface
diff --git a/drivers/net/ethernet/sfc/efx.h b/drivers/net/ethernet/sfc/efx.h
index 4db56b356f11..f1bdb04efbe4 100644
--- a/drivers/net/ethernet/sfc/efx.h
+++ b/drivers/net/ethernet/sfc/efx.h
@@ -169,33 +169,11 @@ static inline void efx_filter_rfs_expire(struct work_struct *data)
 static inline void efx_filter_rfs_expire(struct work_struct *data) {}
 #define efx_filter_rfs_enabled() 0
 #endif
-bool efx_filter_is_mc_recipient(const struct efx_filter_spec *spec);
-
-bool efx_filter_spec_equal(const struct efx_filter_spec *left,
-			   const struct efx_filter_spec *right);
-u32 efx_filter_spec_hash(const struct efx_filter_spec *spec);
-
-#ifdef CONFIG_RFS_ACCEL
-bool efx_rps_check_rule(struct efx_arfs_rule *rule, unsigned int filter_idx,
-			bool *force);
-
-struct efx_arfs_rule *efx_rps_hash_find(struct efx_nic *efx,
-					const struct efx_filter_spec *spec);
-
-/* @new is written to indicate if entry was newly added (true) or if an old
- * entry was found and returned (false).
- */
-struct efx_arfs_rule *efx_rps_hash_add(struct efx_nic *efx,
-				       const struct efx_filter_spec *spec,
-				       bool *new);
-
-void efx_rps_hash_del(struct efx_nic *efx, const struct efx_filter_spec *spec);
-#endif
 
 /* RSS contexts */
 static inline bool efx_rss_active(struct efx_rss_context *ctx)
 {
-	return ctx->context_id != EFX_EF10_RSS_CONTEXT_INVALID;
+	return ctx->context_id != EFX_MCDI_RSS_CONTEXT_INVALID;
 }
 
 /* Ethtool support */
diff --git a/drivers/net/ethernet/sfc/ethtool.c b/drivers/net/ethernet/sfc/ethtool.c
index e50901e4b2ac..993b5769525b 100644
--- a/drivers/net/ethernet/sfc/ethtool.c
+++ b/drivers/net/ethernet/sfc/ethtool.c
@@ -1020,7 +1020,7 @@ static int efx_ethtool_set_rxfh_context(struct net_device *net_dev,
 			rc = -ENOMEM;
 			goto out_unlock;
 		}
-		ctx->context_id = EFX_EF10_RSS_CONTEXT_INVALID;
+		ctx->context_id = EFX_MCDI_RSS_CONTEXT_INVALID;
 		/* Initialise indir table and key to defaults */
 		efx_set_default_rx_indir_table(efx, ctx);
 		netdev_rss_key_fill(ctx->rx_hash_key, sizeof(ctx->rx_hash_key));
diff --git a/drivers/net/ethernet/sfc/net_driver.h b/drivers/net/ethernet/sfc/net_driver.h
index 52e6f11d8818..9f9886f222c8 100644
--- a/drivers/net/ethernet/sfc/net_driver.h
+++ b/drivers/net/ethernet/sfc/net_driver.h
@@ -744,13 +744,13 @@ union efx_multicast_hash {
 struct vfdi_status;
 
 /* The reserved RSS context value */
-#define EFX_EF10_RSS_CONTEXT_INVALID	0xffffffff
+#define EFX_MCDI_RSS_CONTEXT_INVALID	0xffffffff
 /**
  * struct efx_rss_context - A user-defined RSS context for filtering
  * @list: node of linked list on which this struct is stored
  * @context_id: the RSS_CONTEXT_ID returned by MC firmware, or
- *	%EFX_EF10_RSS_CONTEXT_INVALID if this context is not present on the NIC.
- *	For Siena, 0 if RSS is active, else %EFX_EF10_RSS_CONTEXT_INVALID.
+ *	%EFX_MCDI_RSS_CONTEXT_INVALID if this context is not present on the NIC.
+ *	For Siena, 0 if RSS is active, else %EFX_MCDI_RSS_CONTEXT_INVALID.
  * @user_id: the rss_context ID exposed to userspace over ethtool.
  * @rx_hash_udp_4tuple: UDP 4-tuple hashing enabled
  * @rx_hash_key: Toeplitz hash key for this RSS context
diff --git a/drivers/net/ethernet/sfc/rx.c b/drivers/net/ethernet/sfc/rx.c
index 18001195410f..a2042f16babc 100644
--- a/drivers/net/ethernet/sfc/rx.c
+++ b/drivers/net/ethernet/sfc/rx.c
@@ -650,37 +650,3 @@ bool __efx_filter_rfs_expire(struct efx_channel *channel, unsigned int quota)
 }
 
 #endif /* CONFIG_RFS_ACCEL */
-
-/**
- * efx_filter_is_mc_recipient - test whether spec is a multicast recipient
- * @spec: Specification to test
- *
- * Return: %true if the specification is a non-drop RX filter that
- * matches a local MAC address I/G bit value of 1 or matches a local
- * IPv4 or IPv6 address value in the respective multicast address
- * range.  Otherwise %false.
- */
-bool efx_filter_is_mc_recipient(const struct efx_filter_spec *spec)
-{
-	if (!(spec->flags & EFX_FILTER_FLAG_RX) ||
-	    spec->dmaq_id == EFX_FILTER_RX_DMAQ_ID_DROP)
-		return false;
-
-	if (spec->match_flags &
-	    (EFX_FILTER_MATCH_LOC_MAC | EFX_FILTER_MATCH_LOC_MAC_IG) &&
-	    is_multicast_ether_addr(spec->loc_mac))
-		return true;
-
-	if ((spec->match_flags &
-	     (EFX_FILTER_MATCH_ETHER_TYPE | EFX_FILTER_MATCH_LOC_HOST)) ==
-	    (EFX_FILTER_MATCH_ETHER_TYPE | EFX_FILTER_MATCH_LOC_HOST)) {
-		if (spec->ether_type == htons(ETH_P_IP) &&
-		    ipv4_is_multicast(spec->loc_host[0]))
-			return true;
-		if (spec->ether_type == htons(ETH_P_IPV6) &&
-		    ((const u8 *)spec->loc_host)[0] == 0xff)
-			return true;
-	}
-
-	return false;
-}
diff --git a/drivers/net/ethernet/sfc/rx_common.c b/drivers/net/ethernet/sfc/rx_common.c
index cdf21e2d5610..ee8beb87bdc1 100644
--- a/drivers/net/ethernet/sfc/rx_common.c
+++ b/drivers/net/ethernet/sfc/rx_common.c
@@ -579,7 +579,7 @@ struct efx_rss_context *efx_alloc_rss_context_entry(struct efx_nic *efx)
 	new = kmalloc(sizeof(*new), GFP_KERNEL);
 	if (!new)
 		return NULL;
-	new->context_id = EFX_EF10_RSS_CONTEXT_INVALID;
+	new->context_id = EFX_MCDI_RSS_CONTEXT_INVALID;
 	new->rx_hash_udp_4tuple = false;
 
 	/* Insert the new entry into the gap */
@@ -616,3 +616,236 @@ void efx_set_default_rx_indir_table(struct efx_nic *efx,
 		ctx->rx_indir_table[i] =
 			ethtool_rxfh_indir_default(i, efx->rss_spread);
 }
+
+/**
+ * efx_filter_is_mc_recipient - test whether spec is a multicast recipient
+ * @spec: Specification to test
+ *
+ * Return: %true if the specification is a non-drop RX filter that
+ * matches a local MAC address I/G bit value of 1 or matches a local
+ * IPv4 or IPv6 address value in the respective multicast address
+ * range.  Otherwise %false.
+ */
+bool efx_filter_is_mc_recipient(const struct efx_filter_spec *spec)
+{
+	if (!(spec->flags & EFX_FILTER_FLAG_RX) ||
+	    spec->dmaq_id == EFX_FILTER_RX_DMAQ_ID_DROP)
+		return false;
+
+	if (spec->match_flags &
+	    (EFX_FILTER_MATCH_LOC_MAC | EFX_FILTER_MATCH_LOC_MAC_IG) &&
+	    is_multicast_ether_addr(spec->loc_mac))
+		return true;
+
+	if ((spec->match_flags &
+	     (EFX_FILTER_MATCH_ETHER_TYPE | EFX_FILTER_MATCH_LOC_HOST)) ==
+	    (EFX_FILTER_MATCH_ETHER_TYPE | EFX_FILTER_MATCH_LOC_HOST)) {
+		if (spec->ether_type == htons(ETH_P_IP) &&
+		    ipv4_is_multicast(spec->loc_host[0]))
+			return true;
+		if (spec->ether_type == htons(ETH_P_IPV6) &&
+		    ((const u8 *)spec->loc_host)[0] == 0xff)
+			return true;
+	}
+
+	return false;
+}
+
+bool efx_filter_spec_equal(const struct efx_filter_spec *left,
+			   const struct efx_filter_spec *right)
+{
+	if ((left->match_flags ^ right->match_flags) |
+	    ((left->flags ^ right->flags) &
+	     (EFX_FILTER_FLAG_RX | EFX_FILTER_FLAG_TX)))
+		return false;
+
+	return memcmp(&left->outer_vid, &right->outer_vid,
+		      sizeof(struct efx_filter_spec) -
+		      offsetof(struct efx_filter_spec, outer_vid)) == 0;
+}
+
+u32 efx_filter_spec_hash(const struct efx_filter_spec *spec)
+{
+	BUILD_BUG_ON(offsetof(struct efx_filter_spec, outer_vid) & 3);
+	return jhash2((const u32 *)&spec->outer_vid,
+		      (sizeof(struct efx_filter_spec) -
+		       offsetof(struct efx_filter_spec, outer_vid)) / 4,
+		      0);
+}
+
+#ifdef CONFIG_RFS_ACCEL
+bool efx_rps_check_rule(struct efx_arfs_rule *rule, unsigned int filter_idx,
+			bool *force)
+{
+	if (rule->filter_id == EFX_ARFS_FILTER_ID_PENDING) {
+		/* ARFS is currently updating this entry, leave it */
+		return false;
+	}
+	if (rule->filter_id == EFX_ARFS_FILTER_ID_ERROR) {
+		/* ARFS tried and failed to update this, so it's probably out
+		 * of date.  Remove the filter and the ARFS rule entry.
+		 */
+		rule->filter_id = EFX_ARFS_FILTER_ID_REMOVING;
+		*force = true;
+		return true;
+	} else if (WARN_ON(rule->filter_id != filter_idx)) { /* can't happen */
+		/* ARFS has moved on, so old filter is not needed.  Since we did
+		 * not mark the rule with EFX_ARFS_FILTER_ID_REMOVING, it will
+		 * not be removed by efx_rps_hash_del() subsequently.
+		 */
+		*force = true;
+		return true;
+	}
+	/* Remove it iff ARFS wants to. */
+	return true;
+}
+
+static
+struct hlist_head *efx_rps_hash_bucket(struct efx_nic *efx,
+				       const struct efx_filter_spec *spec)
+{
+	u32 hash = efx_filter_spec_hash(spec);
+
+	lockdep_assert_held(&efx->rps_hash_lock);
+	if (!efx->rps_hash_table)
+		return NULL;
+	return &efx->rps_hash_table[hash % EFX_ARFS_HASH_TABLE_SIZE];
+}
+
+struct efx_arfs_rule *efx_rps_hash_find(struct efx_nic *efx,
+					const struct efx_filter_spec *spec)
+{
+	struct efx_arfs_rule *rule;
+	struct hlist_head *head;
+	struct hlist_node *node;
+
+	head = efx_rps_hash_bucket(efx, spec);
+	if (!head)
+		return NULL;
+	hlist_for_each(node, head) {
+		rule = container_of(node, struct efx_arfs_rule, node);
+		if (efx_filter_spec_equal(spec, &rule->spec))
+			return rule;
+	}
+	return NULL;
+}
+
+struct efx_arfs_rule *efx_rps_hash_add(struct efx_nic *efx,
+				       const struct efx_filter_spec *spec,
+				       bool *new)
+{
+	struct efx_arfs_rule *rule;
+	struct hlist_head *head;
+	struct hlist_node *node;
+
+	head = efx_rps_hash_bucket(efx, spec);
+	if (!head)
+		return NULL;
+	hlist_for_each(node, head) {
+		rule = container_of(node, struct efx_arfs_rule, node);
+		if (efx_filter_spec_equal(spec, &rule->spec)) {
+			*new = false;
+			return rule;
+		}
+	}
+	rule = kmalloc(sizeof(*rule), GFP_ATOMIC);
+	*new = true;
+	if (rule) {
+		memcpy(&rule->spec, spec, sizeof(rule->spec));
+		hlist_add_head(&rule->node, head);
+	}
+	return rule;
+}
+
+void efx_rps_hash_del(struct efx_nic *efx, const struct efx_filter_spec *spec)
+{
+	struct efx_arfs_rule *rule;
+	struct hlist_head *head;
+	struct hlist_node *node;
+
+	head = efx_rps_hash_bucket(efx, spec);
+	if (WARN_ON(!head))
+		return;
+	hlist_for_each(node, head) {
+		rule = container_of(node, struct efx_arfs_rule, node);
+		if (efx_filter_spec_equal(spec, &rule->spec)) {
+			/* Someone already reused the entry.  We know that if
+			 * this check doesn't fire (i.e. filter_id == REMOVING)
+			 * then the REMOVING mark was put there by our caller,
+			 * because caller is holding a lock on filter table and
+			 * only holders of that lock set REMOVING.
+			 */
+			if (rule->filter_id != EFX_ARFS_FILTER_ID_REMOVING)
+				return;
+			hlist_del(node);
+			kfree(rule);
+			return;
+		}
+	}
+	/* We didn't find it. */
+	WARN_ON(1);
+}
+#endif
+
+int efx_probe_filters(struct efx_nic *efx)
+{
+	int rc;
+
+	init_rwsem(&efx->filter_sem);
+	mutex_lock(&efx->mac_lock);
+	down_write(&efx->filter_sem);
+	rc = efx->type->filter_table_probe(efx);
+	if (rc)
+		goto out_unlock;
+
+#ifdef CONFIG_RFS_ACCEL
+	if (efx->type->offload_features & NETIF_F_NTUPLE) {
+		struct efx_channel *channel;
+		int i, success = 1;
+
+		efx_for_each_channel(channel, efx) {
+			channel->rps_flow_id =
+				kcalloc(efx->type->max_rx_ip_filters,
+					sizeof(*channel->rps_flow_id),
+					GFP_KERNEL);
+			if (!channel->rps_flow_id)
+				success = 0;
+			else
+				for (i = 0;
+				     i < efx->type->max_rx_ip_filters;
+				     ++i)
+					channel->rps_flow_id[i] =
+						RPS_FLOW_ID_INVALID;
+			channel->rfs_expire_index = 0;
+			channel->rfs_filter_count = 0;
+		}
+
+		if (!success) {
+			efx_for_each_channel(channel, efx)
+				kfree(channel->rps_flow_id);
+			efx->type->filter_table_remove(efx);
+			rc = -ENOMEM;
+			goto out_unlock;
+		}
+	}
+#endif
+out_unlock:
+	up_write(&efx->filter_sem);
+	mutex_unlock(&efx->mac_lock);
+	return rc;
+}
+
+void efx_remove_filters(struct efx_nic *efx)
+{
+#ifdef CONFIG_RFS_ACCEL
+	struct efx_channel *channel;
+
+	efx_for_each_channel(channel, efx) {
+		cancel_delayed_work_sync(&channel->filter_work);
+		kfree(channel->rps_flow_id);
+	}
+#endif
+	down_write(&efx->filter_sem);
+	efx->type->filter_table_remove(efx);
+	up_write(&efx->filter_sem);
+}
diff --git a/drivers/net/ethernet/sfc/rx_common.h b/drivers/net/ethernet/sfc/rx_common.h
index 4d2e7d38e260..c41f12a89477 100644
--- a/drivers/net/ethernet/sfc/rx_common.h
+++ b/drivers/net/ethernet/sfc/rx_common.h
@@ -74,4 +74,24 @@ struct efx_rss_context *efx_find_rss_context_entry(struct efx_nic *efx, u32 id);
 void efx_free_rss_context_entry(struct efx_rss_context *ctx);
 void efx_set_default_rx_indir_table(struct efx_nic *efx,
 				    struct efx_rss_context *ctx);
+
+bool efx_filter_is_mc_recipient(const struct efx_filter_spec *spec);
+bool efx_filter_spec_equal(const struct efx_filter_spec *left,
+			   const struct efx_filter_spec *right);
+u32 efx_filter_spec_hash(const struct efx_filter_spec *spec);
+
+#ifdef CONFIG_RFS_ACCEL
+bool efx_rps_check_rule(struct efx_arfs_rule *rule, unsigned int filter_idx,
+			bool *force);
+struct efx_arfs_rule *efx_rps_hash_find(struct efx_nic *efx,
+					const struct efx_filter_spec *spec);
+struct efx_arfs_rule *efx_rps_hash_add(struct efx_nic *efx,
+				       const struct efx_filter_spec *spec,
+				       bool *new);
+void efx_rps_hash_del(struct efx_nic *efx, const struct efx_filter_spec *spec);
+#endif
+
+int efx_probe_filters(struct efx_nic *efx);
+void efx_remove_filters(struct efx_nic *efx);
+
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From 8eea7c7f3f2fcc8d580df331890090a023ef3f6b Mon Sep 17 00:00:00 2001
From: Jose Abreu <Jose.Abreu@synopsys.com>
Date: Fri, 10 Jan 2020 16:13:34 +0100
Subject: net: stmmac: xgmac: Fix missing return

If FPE is supposed to be disabled we need to return after disabling it.

Fixes: f0e56c8d8f7d ("net: stmmac: xgmac3+: Add support for Frame Preemption")
Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c
index 307105e8dea0..2af3ac5409b7 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c
@@ -1421,6 +1421,7 @@ static void dwxgmac3_fpe_configure(void __iomem *ioaddr, u32 num_txq,
 		value &= ~XGMAC_EFPE;
 
 		writel(value, ioaddr + XGMAC_FPE_CTRL_STS);
+		return;
 	}
 
 	value = readl(ioaddr + XGMAC_RXQ_CTRL1);
-- 
cgit v1.2.3-59-g8ed1b


From e735def0186770d9a2904bee6ee82a4e6dd99578 Mon Sep 17 00:00:00 2001
From: Jose Abreu <Jose.Abreu@synopsys.com>
Date: Fri, 10 Jan 2020 16:13:35 +0100
Subject: net: stmmac: gmac5+: Fix missing return

If FPE is supposed to be disabled we need to return after disabling it.

Fixes: 7c7282746883 ("net: stmmac: gmac5+: Add support for Frame Preemption")
Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac5.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac5.c b/drivers/net/ethernet/stmicro/stmmac/dwmac5.c
index 5d4a3c2458ea..494c859b4ade 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac5.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac5.c
@@ -657,6 +657,7 @@ void dwmac5_fpe_configure(void __iomem *ioaddr, u32 num_txq, u32 num_rxq,
 		value &= ~EFPE;
 
 		writel(value, ioaddr + MAC_FPE_CTRL_STS);
+		return;
 	}
 
 	value = readl(ioaddr + GMAC_RXQ_CTRL1);
-- 
cgit v1.2.3-59-g8ed1b


From ac065870d92824ee0bc275ab71fd8901dbde5055 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Fri, 10 Jan 2020 13:16:34 -0800
Subject: selftests/bpf: Add BPF_PROG, BPF_KPROBE, and BPF_KRETPROBE macros

Streamline BPF_TRACE_x macro by moving out return type and section attribute
definition out of macro itself. That makes those function look in source code
similar to other BPF programs. Additionally, simplify its usage by determining
number of arguments automatically (so just single BPF_TRACE vs a family of
BPF_TRACE_1, BPF_TRACE_2, etc). Also, allow more natural function argument
syntax without commas inbetween argument type and name.

Given this helper is useful not only for tracing tp_btf/fenty/fexit programs,
but could be used for LSM programs and others following the same pattern,
rename BPF_TRACE macro into more generic BPF_PROG. Existing BPF_TRACE_x
usages in selftests are converted to new BPF_PROG macro.

Following the same pattern, define BPF_KPROBE and BPF_KRETPROBE macros for
nicer usage of kprobe/kretprobe arguments, respectively. BPF_KRETPROBE, adopts
same convention used by fexit programs, that last defined argument is probed
function's return result.

v4->v5:
- fix test_overhead test (__set_task_comm is void) (Alexei);

v3->v4:
- rebased and fixed one more BPF_TRACE_x occurence (Alexei);

v2->v3:
- rename to shorter and as generic BPF_PROG (Alexei);

v1->v2:
- verified GCC handles pragmas as expected;
- added descriptions to macros;
- converted new STRUCT_OPS selftest to BPF_HANDLER (worked as expected);
- added original context as 'ctx' parameter, for cases where it has to be
  passed into BPF helpers. This might cause an accidental naming collision,
  unfortunately, but at least it's easy to work around. Fortunately, this
  situation produces quite legible compilation error:

progs/bpf_dctcp.c:46:6: error: redefinition of 'ctx' with a different type: 'int' vs 'unsigned long long *'
        int ctx = 123;
            ^
progs/bpf_dctcp.c:42:6: note: previous definition is here
void BPF_HANDLER(dctcp_init, struct sock *sk)
     ^
./bpf_trace_helpers.h:58:32: note: expanded from macro 'BPF_HANDLER'
____##name(unsigned long long *ctx, ##args)

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20200110211634.1614739-1-andriin@fb.com
---
 tools/testing/selftests/bpf/bpf_tcp_helpers.h      |   9 --
 tools/testing/selftests/bpf/bpf_trace_helpers.h    | 166 ++++++++++++++-------
 tools/testing/selftests/bpf/progs/bpf_dctcp.c      |  26 ++--
 tools/testing/selftests/bpf/progs/fentry_test.c    |  21 +--
 tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c  |  12 +-
 .../selftests/bpf/progs/fexit_bpf2bpf_simple.c     |   5 +-
 tools/testing/selftests/bpf/progs/fexit_test.c     |  23 +--
 tools/testing/selftests/bpf/progs/kfree_skb.c      |  16 +-
 tools/testing/selftests/bpf/progs/test_overhead.c  |  30 ++--
 .../testing/selftests/bpf/progs/test_perf_buffer.c |   3 +-
 .../testing/selftests/bpf/progs/test_probe_user.c  |   3 +-
 11 files changed, 193 insertions(+), 121 deletions(-)

diff --git a/tools/testing/selftests/bpf/bpf_tcp_helpers.h b/tools/testing/selftests/bpf/bpf_tcp_helpers.h
index 7da3e8db3adf..1052b9bb4320 100644
--- a/tools/testing/selftests/bpf/bpf_tcp_helpers.h
+++ b/tools/testing/selftests/bpf/bpf_tcp_helpers.h
@@ -6,15 +6,6 @@
 #include <linux/types.h>
 #include <bpf_helpers.h>
 #include <bpf_core_read.h>
-#include "bpf_trace_helpers.h"
-
-/* "struct_ops/" is only a convention.  not a requirement. */
-#define BPF_TCP_OPS_0(fname, ret_type, ...) BPF_TRACE_x(0, "struct_ops/"#fname, fname, ret_type, __VA_ARGS__)
-#define BPF_TCP_OPS_1(fname, ret_type, ...) BPF_TRACE_x(1, "struct_ops/"#fname, fname, ret_type, __VA_ARGS__)
-#define BPF_TCP_OPS_2(fname, ret_type, ...) BPF_TRACE_x(2, "struct_ops/"#fname, fname, ret_type, __VA_ARGS__)
-#define BPF_TCP_OPS_3(fname, ret_type, ...) BPF_TRACE_x(3, "struct_ops/"#fname, fname, ret_type, __VA_ARGS__)
-#define BPF_TCP_OPS_4(fname, ret_type, ...) BPF_TRACE_x(4, "struct_ops/"#fname, fname, ret_type, __VA_ARGS__)
-#define BPF_TCP_OPS_5(fname, ret_type, ...) BPF_TRACE_x(5, "struct_ops/"#fname, fname, ret_type, __VA_ARGS__)
 
 struct sock_common {
 	unsigned char	skc_state;
diff --git a/tools/testing/selftests/bpf/bpf_trace_helpers.h b/tools/testing/selftests/bpf/bpf_trace_helpers.h
index c76a214a53b0..1719cabc97b7 100644
--- a/tools/testing/selftests/bpf/bpf_trace_helpers.h
+++ b/tools/testing/selftests/bpf/bpf_trace_helpers.h
@@ -1,58 +1,120 @@
-/* SPDX-License-Identifier: GPL-2.0 */
+/* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */
 #ifndef __BPF_TRACE_HELPERS_H
 #define __BPF_TRACE_HELPERS_H
 
-#include "bpf_helpers.h"
-
-#define __BPF_MAP_0(i, m, v, ...) v
-#define __BPF_MAP_1(i, m, v, t, a, ...) m(t, a, ctx[i])
-#define __BPF_MAP_2(i, m, v, t, a, ...) m(t, a, ctx[i]), __BPF_MAP_1(i+1, m, v, __VA_ARGS__)
-#define __BPF_MAP_3(i, m, v, t, a, ...) m(t, a, ctx[i]), __BPF_MAP_2(i+1, m, v, __VA_ARGS__)
-#define __BPF_MAP_4(i, m, v, t, a, ...) m(t, a, ctx[i]), __BPF_MAP_3(i+1, m, v, __VA_ARGS__)
-#define __BPF_MAP_5(i, m, v, t, a, ...) m(t, a, ctx[i]), __BPF_MAP_4(i+1, m, v, __VA_ARGS__)
-#define __BPF_MAP_6(i, m, v, t, a, ...) m(t, a, ctx[i]), __BPF_MAP_5(i+1, m, v, __VA_ARGS__)
-#define __BPF_MAP_7(i, m, v, t, a, ...) m(t, a, ctx[i]), __BPF_MAP_6(i+1, m, v, __VA_ARGS__)
-#define __BPF_MAP_8(i, m, v, t, a, ...) m(t, a, ctx[i]), __BPF_MAP_7(i+1, m, v, __VA_ARGS__)
-#define __BPF_MAP_9(i, m, v, t, a, ...) m(t, a, ctx[i]), __BPF_MAP_8(i+1, m, v, __VA_ARGS__)
-#define __BPF_MAP_10(i, m, v, t, a, ...) m(t, a, ctx[i]), __BPF_MAP_9(i+1, m, v, __VA_ARGS__)
-#define __BPF_MAP_11(i, m, v, t, a, ...) m(t, a, ctx[i]), __BPF_MAP_10(i+1, m, v, __VA_ARGS__)
-#define __BPF_MAP_12(i, m, v, t, a, ...) m(t, a, ctx[i]), __BPF_MAP_11(i+1, m, v, __VA_ARGS__)
-#define __BPF_MAP(n, ...) __BPF_MAP_##n(0, __VA_ARGS__)
-
-/* BPF sizeof(void *) is always 8, so no need to cast to long first
- * for ptr to avoid compiler warning.
+#include <bpf_helpers.h>
+
+#define ___bpf_concat(a, b) a ## b
+#define ___bpf_apply(fn, n) ___bpf_concat(fn, n)
+#define ___bpf_nth(_, _1, _2, _3, _4, _5, _6, _7, _8, _9, _a, _b, _c, N, ...) N
+#define ___bpf_narg(...) \
+	___bpf_nth(_, ##__VA_ARGS__, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0)
+#define ___bpf_empty(...) \
+	___bpf_nth(_, ##__VA_ARGS__, N, N, N, N, N, N, N, N, N, N, 0)
+
+#define ___bpf_ctx_cast0() ctx
+#define ___bpf_ctx_cast1(x) ___bpf_ctx_cast0(), (void *)ctx[0]
+#define ___bpf_ctx_cast2(x, args...) ___bpf_ctx_cast1(args), (void *)ctx[1]
+#define ___bpf_ctx_cast3(x, args...) ___bpf_ctx_cast2(args), (void *)ctx[2]
+#define ___bpf_ctx_cast4(x, args...) ___bpf_ctx_cast3(args), (void *)ctx[3]
+#define ___bpf_ctx_cast5(x, args...) ___bpf_ctx_cast4(args), (void *)ctx[4]
+#define ___bpf_ctx_cast6(x, args...) ___bpf_ctx_cast5(args), (void *)ctx[5]
+#define ___bpf_ctx_cast7(x, args...) ___bpf_ctx_cast6(args), (void *)ctx[6]
+#define ___bpf_ctx_cast8(x, args...) ___bpf_ctx_cast7(args), (void *)ctx[7]
+#define ___bpf_ctx_cast9(x, args...) ___bpf_ctx_cast8(args), (void *)ctx[8]
+#define ___bpf_ctx_cast10(x, args...) ___bpf_ctx_cast9(args), (void *)ctx[9]
+#define ___bpf_ctx_cast11(x, args...) ___bpf_ctx_cast10(args), (void *)ctx[10]
+#define ___bpf_ctx_cast12(x, args...) ___bpf_ctx_cast11(args), (void *)ctx[11]
+#define ___bpf_ctx_cast(args...) \
+	___bpf_apply(___bpf_ctx_cast, ___bpf_narg(args))(args)
+
+/*
+ * BPF_PROG is a convenience wrapper for generic tp_btf/fentry/fexit and
+ * similar kinds of BPF programs, that accept input arguments as a single
+ * pointer to untyped u64 array, where each u64 can actually be a typed
+ * pointer or integer of different size. Instead of requring user to write
+ * manual casts and work with array elements by index, BPF_PROG macro
+ * allows user to declare a list of named and typed input arguments in the
+ * same syntax as for normal C function. All the casting is hidden and
+ * performed transparently, while user code can just assume working with
+ * function arguments of specified type and name.
+ *
+ * Original raw context argument is preserved as well as 'ctx' argument.
+ * This is useful when using BPF helpers that expect original context
+ * as one of the parameters (e.g., for bpf_perf_event_output()).
  */
-#define __BPF_CAST(t, a, ctx) (t) ctx
-#define __BPF_V void
-#define __BPF_N
-
-#define __BPF_DECL_ARGS(t, a, ctx) t a
-
-#define BPF_TRACE_x(x, sec_name, fname, ret_type, ...)			\
-static __always_inline ret_type						\
-____##fname(__BPF_MAP(x, __BPF_DECL_ARGS, __BPF_V, __VA_ARGS__));	\
-									\
-SEC(sec_name)								\
-ret_type fname(__u64 *ctx)						\
-{									\
-	return ____##fname(__BPF_MAP(x, __BPF_CAST, __BPF_N, __VA_ARGS__));\
-}									\
-									\
-static __always_inline							\
-ret_type ____##fname(__BPF_MAP(x, __BPF_DECL_ARGS, __BPF_V, __VA_ARGS__))
-
-#define BPF_TRACE_0(sec, fname, ...)  BPF_TRACE_x(0, sec, fname, int, __VA_ARGS__)
-#define BPF_TRACE_1(sec, fname, ...)  BPF_TRACE_x(1, sec, fname, int, __VA_ARGS__)
-#define BPF_TRACE_2(sec, fname, ...)  BPF_TRACE_x(2, sec, fname, int, __VA_ARGS__)
-#define BPF_TRACE_3(sec, fname, ...)  BPF_TRACE_x(3, sec, fname, int, __VA_ARGS__)
-#define BPF_TRACE_4(sec, fname, ...)  BPF_TRACE_x(4, sec, fname, int, __VA_ARGS__)
-#define BPF_TRACE_5(sec, fname, ...)  BPF_TRACE_x(5, sec, fname, int, __VA_ARGS__)
-#define BPF_TRACE_6(sec, fname, ...)  BPF_TRACE_x(6, sec, fname, int, __VA_ARGS__)
-#define BPF_TRACE_7(sec, fname, ...)  BPF_TRACE_x(7, sec, fname, int, __VA_ARGS__)
-#define BPF_TRACE_8(sec, fname, ...)  BPF_TRACE_x(8, sec, fname, int, __VA_ARGS__)
-#define BPF_TRACE_9(sec, fname, ...)  BPF_TRACE_x(9, sec, fname, int, __VA_ARGS__)
-#define BPF_TRACE_10(sec, fname, ...)  BPF_TRACE_x(10, sec, fname, int, __VA_ARGS__)
-#define BPF_TRACE_11(sec, fname, ...)  BPF_TRACE_x(11, sec, fname, int, __VA_ARGS__)
-#define BPF_TRACE_12(sec, fname, ...)  BPF_TRACE_x(12, sec, fname, int, __VA_ARGS__)
+#define BPF_PROG(name, args...)						    \
+name(unsigned long long *ctx);						    \
+static __always_inline typeof(name(0))					    \
+____##name(unsigned long long *ctx, ##args);				    \
+typeof(name(0)) name(unsigned long long *ctx)				    \
+{									    \
+	_Pragma("GCC diagnostic push")					    \
+	_Pragma("GCC diagnostic ignored \"-Wint-conversion\"")		    \
+	return ____##name(___bpf_ctx_cast(args));			    \
+	_Pragma("GCC diagnostic pop")					    \
+}									    \
+static __always_inline typeof(name(0))					    \
+____##name(unsigned long long *ctx, ##args)
+
+struct pt_regs;
+
+#define ___bpf_kprobe_args0() ctx
+#define ___bpf_kprobe_args1(x) \
+	___bpf_kprobe_args0(), (void *)PT_REGS_PARM1(ctx)
+#define ___bpf_kprobe_args2(x, args...) \
+	___bpf_kprobe_args1(args), (void *)PT_REGS_PARM2(ctx)
+#define ___bpf_kprobe_args3(x, args...) \
+	___bpf_kprobe_args2(args), (void *)PT_REGS_PARM3(ctx)
+#define ___bpf_kprobe_args4(x, args...) \
+	___bpf_kprobe_args3(args), (void *)PT_REGS_PARM4(ctx)
+#define ___bpf_kprobe_args5(x, args...) \
+	___bpf_kprobe_args4(args), (void *)PT_REGS_PARM5(ctx)
+#define ___bpf_kprobe_args(args...) \
+	___bpf_apply(___bpf_kprobe_args, ___bpf_narg(args))(args)
 
+/*
+ * BPF_KPROBE serves the same purpose for kprobes as BPF_PROG for
+ * tp_btf/fentry/fexit BPF programs. It hides the underlying platform-specific
+ * low-level way of getting kprobe input arguments from struct pt_regs, and
+ * provides a familiar typed and named function arguments syntax and
+ * semantics of accessing kprobe input paremeters.
+ *
+ * Original struct pt_regs* context is preserved as 'ctx' argument. This might
+ * be necessary when using BPF helpers like bpf_perf_event_output().
+ */
+#define BPF_KPROBE(name, args...)					    \
+name(struct pt_regs *ctx);						    \
+static __always_inline typeof(name(0)) ____##name(struct pt_regs *ctx, ##args);\
+typeof(name(0)) name(struct pt_regs *ctx)				    \
+{									    \
+	_Pragma("GCC diagnostic push")					    \
+	_Pragma("GCC diagnostic ignored \"-Wint-conversion\"")		    \
+	return ____##name(___bpf_kprobe_args(args));			    \
+	_Pragma("GCC diagnostic pop")					    \
+}									    \
+static __always_inline typeof(name(0)) ____##name(struct pt_regs *ctx, ##args)
+
+#define ___bpf_kretprobe_args0() ctx
+#define ___bpf_kretprobe_argsN(x, args...) \
+	___bpf_kprobe_args(args), (void *)PT_REGS_RET(ctx)
+#define ___bpf_kretprobe_args(args...) \
+	___bpf_apply(___bpf_kretprobe_args, ___bpf_empty(args))(args)
+
+/*
+ * BPF_KRETPROBE is similar to BPF_KPROBE, except, in addition to listing all
+ * input kprobe arguments, one last extra argument has to be specified, which
+ * captures kprobe return value.
+ */
+#define BPF_KRETPROBE(name, args...)					    \
+name(struct pt_regs *ctx);						    \
+static __always_inline typeof(name(0)) ____##name(struct pt_regs *ctx, ##args);\
+typeof(name(0)) name(struct pt_regs *ctx)				    \
+{									    \
+	_Pragma("GCC diagnostic push")					    \
+	_Pragma("GCC diagnostic ignored \"-Wint-conversion\"")		    \
+	return ____##name(___bpf_kretprobe_args(args));			    \
+	_Pragma("GCC diagnostic pop")					    \
+}									    \
+static __always_inline typeof(name(0)) ____##name(struct pt_regs *ctx, ##args)
 #endif
diff --git a/tools/testing/selftests/bpf/progs/bpf_dctcp.c b/tools/testing/selftests/bpf/progs/bpf_dctcp.c
index 5f9b613663e5..19650f43c192 100644
--- a/tools/testing/selftests/bpf/progs/bpf_dctcp.c
+++ b/tools/testing/selftests/bpf/progs/bpf_dctcp.c
@@ -8,6 +8,8 @@
 
 #include <linux/bpf.h>
 #include <linux/types.h>
+#include <bpf_helpers.h>
+#include <bpf_trace_helpers.h>
 #include "bpf_tcp_helpers.h"
 
 char _license[] SEC("license") = "GPL";
@@ -36,7 +38,8 @@ static __always_inline void dctcp_reset(const struct tcp_sock *tp,
 	ca->old_delivered_ce = tp->delivered_ce;
 }
 
-BPF_TCP_OPS_1(dctcp_init, void, struct sock *, sk)
+SEC("struct_ops/dctcp_init")
+void BPF_PROG(dctcp_init, struct sock *sk)
 {
 	const struct tcp_sock *tp = tcp_sk(sk);
 	struct dctcp *ca = inet_csk_ca(sk);
@@ -49,7 +52,8 @@ BPF_TCP_OPS_1(dctcp_init, void, struct sock *, sk)
 	dctcp_reset(tp, ca);
 }
 
-BPF_TCP_OPS_1(dctcp_ssthresh, __u32, struct sock *, sk)
+SEC("struct_ops/dctcp_ssthresh")
+__u32 BPF_PROG(dctcp_ssthresh, struct sock *sk)
 {
 	struct dctcp *ca = inet_csk_ca(sk);
 	struct tcp_sock *tp = tcp_sk(sk);
@@ -58,8 +62,8 @@ BPF_TCP_OPS_1(dctcp_ssthresh, __u32, struct sock *, sk)
 	return max(tp->snd_cwnd - ((tp->snd_cwnd * ca->dctcp_alpha) >> 11U), 2U);
 }
 
-BPF_TCP_OPS_2(dctcp_update_alpha, void,
-	      struct sock *, sk, __u32, flags)
+SEC("struct_ops/dctcp_update_alpha")
+void BPF_PROG(dctcp_update_alpha, struct sock *sk, __u32 flags)
 {
 	const struct tcp_sock *tp = tcp_sk(sk);
 	struct dctcp *ca = inet_csk_ca(sk);
@@ -97,7 +101,8 @@ static __always_inline void dctcp_react_to_loss(struct sock *sk)
 	tp->snd_ssthresh = max(tp->snd_cwnd >> 1U, 2U);
 }
 
-BPF_TCP_OPS_2(dctcp_state, void, struct sock *, sk, __u8, new_state)
+SEC("struct_ops/dctcp_state")
+void BPF_PROG(dctcp_state, struct sock *sk, __u8 new_state)
 {
 	if (new_state == TCP_CA_Recovery &&
 	    new_state != BPF_CORE_READ_BITFIELD(inet_csk(sk), icsk_ca_state))
@@ -144,8 +149,8 @@ void dctcp_ece_ack_update(struct sock *sk, enum tcp_ca_event evt,
 	dctcp_ece_ack_cwr(sk, new_ce_state);
 }
 
-BPF_TCP_OPS_2(dctcp_cwnd_event, void,
-	      struct sock *, sk, enum tcp_ca_event, ev)
+SEC("struct_ops/dctcp_cwnd_event")
+void BPF_PROG(dctcp_cwnd_event, struct sock *sk, enum tcp_ca_event ev)
 {
 	struct dctcp *ca = inet_csk_ca(sk);
 
@@ -163,15 +168,16 @@ BPF_TCP_OPS_2(dctcp_cwnd_event, void,
 	}
 }
 
-BPF_TCP_OPS_1(dctcp_cwnd_undo, __u32, struct sock *, sk)
+SEC("struct_ops/dctcp_cwnd_undo")
+__u32 BPF_PROG(dctcp_cwnd_undo, struct sock *sk)
 {
 	const struct dctcp *ca = inet_csk_ca(sk);
 
 	return max(tcp_sk(sk)->snd_cwnd, ca->loss_cwnd);
 }
 
-BPF_TCP_OPS_3(tcp_reno_cong_avoid, void,
-	      struct sock *, sk, __u32, ack, __u32, acked)
+SEC("struct_ops/tcp_reno_cong_avoid")
+void BPF_PROG(tcp_reno_cong_avoid, struct sock *sk, __u32 ack, __u32 acked)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 
diff --git a/tools/testing/selftests/bpf/progs/fentry_test.c b/tools/testing/selftests/bpf/progs/fentry_test.c
index 615f7c6bca77..1ae1779a8e2e 100644
--- a/tools/testing/selftests/bpf/progs/fentry_test.c
+++ b/tools/testing/selftests/bpf/progs/fentry_test.c
@@ -7,37 +7,40 @@
 char _license[] SEC("license") = "GPL";
 
 __u64 test1_result = 0;
-BPF_TRACE_1("fentry/bpf_fentry_test1", test1, int, a)
+SEC("fentry/bpf_fentry_test1")
+int BPF_PROG(test1, int a)
 {
 	test1_result = a == 1;
 	return 0;
 }
 
 __u64 test2_result = 0;
-BPF_TRACE_2("fentry/bpf_fentry_test2", test2, int, a, __u64, b)
+SEC("fentry/bpf_fentry_test2")
+int BPF_PROG(test2, int a, __u64 b)
 {
 	test2_result = a == 2 && b == 3;
 	return 0;
 }
 
 __u64 test3_result = 0;
-BPF_TRACE_3("fentry/bpf_fentry_test3", test3, char, a, int, b, __u64, c)
+SEC("fentry/bpf_fentry_test3")
+int BPF_PROG(test3, char a, int b, __u64 c)
 {
 	test3_result = a == 4 && b == 5 && c == 6;
 	return 0;
 }
 
 __u64 test4_result = 0;
-BPF_TRACE_4("fentry/bpf_fentry_test4", test4,
-	    void *, a, char, b, int, c, __u64, d)
+SEC("fentry/bpf_fentry_test4")
+int BPF_PROG(test4, void *a, char b, int c, __u64 d)
 {
 	test4_result = a == (void *)7 && b == 8 && c == 9 && d == 10;
 	return 0;
 }
 
 __u64 test5_result = 0;
-BPF_TRACE_5("fentry/bpf_fentry_test5", test5,
-	    __u64, a, void *, b, short, c, int, d, __u64, e)
+SEC("fentry/bpf_fentry_test5")
+int BPF_PROG(test5, __u64 a, void *b, short c, int d, __u64 e)
 {
 	test5_result = a == 11 && b == (void *)12 && c == 13 && d == 14 &&
 		e == 15;
@@ -45,8 +48,8 @@ BPF_TRACE_5("fentry/bpf_fentry_test5", test5,
 }
 
 __u64 test6_result = 0;
-BPF_TRACE_6("fentry/bpf_fentry_test6", test6,
-	    __u64, a, void *, b, short, c, int, d, void *, e, __u64, f)
+SEC("fentry/bpf_fentry_test6")
+int BPF_PROG(test6, __u64 a, void *b, short c, int d, void * e, __u64 f)
 {
 	test6_result = a == 16 && b == (void *)17 && c == 18 && d == 19 &&
 		e == (void *)20 && f == 21;
diff --git a/tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c b/tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c
index 81d7b4aaf79e..7c17ee159378 100644
--- a/tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c
+++ b/tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c
@@ -9,8 +9,8 @@ struct sk_buff {
 };
 
 __u64 test_result = 0;
-BPF_TRACE_2("fexit/test_pkt_access", test_main,
-	    struct sk_buff *, skb, int, ret)
+SEC("fexit/test_pkt_access")
+int BPF_PROG(test_main, struct sk_buff *skb, int ret)
 {
 	int len;
 
@@ -24,8 +24,8 @@ BPF_TRACE_2("fexit/test_pkt_access", test_main,
 }
 
 __u64 test_result_subprog1 = 0;
-BPF_TRACE_2("fexit/test_pkt_access_subprog1", test_subprog1,
-	    struct sk_buff *, skb, int, ret)
+SEC("fexit/test_pkt_access_subprog1")
+int BPF_PROG(test_subprog1, struct sk_buff *skb, int ret)
 {
 	int len;
 
@@ -81,8 +81,8 @@ int test_subprog2(struct args_subprog2 *ctx)
 }
 
 __u64 test_result_subprog3 = 0;
-BPF_TRACE_3("fexit/test_pkt_access_subprog3", test_subprog3,
-	    int, val, struct sk_buff *, skb, int, ret)
+SEC("fexit/test_pkt_access_subprog3")
+int BPF_PROG(test_subprog3, int val, struct sk_buff *skb, int ret)
 {
 	int len;
 
diff --git a/tools/testing/selftests/bpf/progs/fexit_bpf2bpf_simple.c b/tools/testing/selftests/bpf/progs/fexit_bpf2bpf_simple.c
index ebc0ab7f0f5c..ac527b10dd80 100644
--- a/tools/testing/selftests/bpf/progs/fexit_bpf2bpf_simple.c
+++ b/tools/testing/selftests/bpf/progs/fexit_bpf2bpf_simple.c
@@ -9,8 +9,9 @@ struct sk_buff {
 };
 
 __u64 test_result = 0;
-BPF_TRACE_2("fexit/test_pkt_md_access", test_main2,
-	    struct sk_buff *, skb, int, ret)
+
+SEC("fexit/test_pkt_md_access")
+int BPF_PROG(test_main2, struct sk_buff *skb, int ret)
 {
 	int len;
 
diff --git a/tools/testing/selftests/bpf/progs/fexit_test.c b/tools/testing/selftests/bpf/progs/fexit_test.c
index 86db0d60fb6e..f0dd412ca128 100644
--- a/tools/testing/selftests/bpf/progs/fexit_test.c
+++ b/tools/testing/selftests/bpf/progs/fexit_test.c
@@ -7,39 +7,41 @@
 char _license[] SEC("license") = "GPL";
 
 __u64 test1_result = 0;
-BPF_TRACE_2("fexit/bpf_fentry_test1", test1, int, a, int, ret)
+SEC("fexit/bpf_fentry_test1")
+int BPF_PROG(test1, int a, int ret)
 {
 	test1_result = a == 1 && ret == 2;
 	return 0;
 }
 
 __u64 test2_result = 0;
-BPF_TRACE_3("fexit/bpf_fentry_test2", test2, int, a, __u64, b, int, ret)
+SEC("fexit/bpf_fentry_test2")
+int BPF_PROG(test2, int a, __u64 b, int ret)
 {
 	test2_result = a == 2 && b == 3 && ret == 5;
 	return 0;
 }
 
 __u64 test3_result = 0;
-BPF_TRACE_4("fexit/bpf_fentry_test3", test3, char, a, int, b, __u64, c, int, ret)
+SEC("fexit/bpf_fentry_test3")
+int BPF_PROG(test3, char a, int b, __u64 c, int ret)
 {
 	test3_result = a == 4 && b == 5 && c == 6 && ret == 15;
 	return 0;
 }
 
 __u64 test4_result = 0;
-BPF_TRACE_5("fexit/bpf_fentry_test4", test4,
-	    void *, a, char, b, int, c, __u64, d, int, ret)
+SEC("fexit/bpf_fentry_test4")
+int BPF_PROG(test4, void *a, char b, int c, __u64 d, int ret)
 {
-
 	test4_result = a == (void *)7 && b == 8 && c == 9 && d == 10 &&
 		ret == 34;
 	return 0;
 }
 
 __u64 test5_result = 0;
-BPF_TRACE_6("fexit/bpf_fentry_test5", test5,
-	    __u64, a, void *, b, short, c, int, d, __u64, e, int, ret)
+SEC("fexit/bpf_fentry_test5")
+int BPF_PROG(test5, __u64 a, void *b, short c, int d, __u64 e, int ret)
 {
 	test5_result = a == 11 && b == (void *)12 && c == 13 && d == 14 &&
 		e == 15 && ret == 65;
@@ -47,9 +49,8 @@ BPF_TRACE_6("fexit/bpf_fentry_test5", test5,
 }
 
 __u64 test6_result = 0;
-BPF_TRACE_7("fexit/bpf_fentry_test6", test6,
-	    __u64, a, void *, b, short, c, int, d, void *, e, __u64, f,
-	    int, ret)
+SEC("fexit/bpf_fentry_test6")
+int BPF_PROG(test6, __u64 a, void *b, short c, int d, void *e, __u64 f, int ret)
 {
 	test6_result = a == 16 && b == (void *)17 && c == 18 && d == 19 &&
 		e == (void *)20 && f == 21 && ret == 111;
diff --git a/tools/testing/selftests/bpf/progs/kfree_skb.c b/tools/testing/selftests/bpf/progs/kfree_skb.c
index 974d6f3bb319..046fae868f93 100644
--- a/tools/testing/selftests/bpf/progs/kfree_skb.c
+++ b/tools/testing/selftests/bpf/progs/kfree_skb.c
@@ -57,8 +57,8 @@ struct meta {
 /* TRACE_EVENT(kfree_skb,
  *         TP_PROTO(struct sk_buff *skb, void *location),
  */
-BPF_TRACE_2("tp_btf/kfree_skb", trace_kfree_skb,
-	    struct sk_buff *, skb, void *, location)
+SEC("tp_btf/kfree_skb")
+int BPF_PROG(trace_kfree_skb, struct sk_buff *skb, void *location)
 {
 	struct net_device *dev;
 	struct callback_head *ptr;
@@ -114,9 +114,9 @@ static volatile struct {
 	bool fexit_test_ok;
 } result;
 
-BPF_TRACE_3("fentry/eth_type_trans", fentry_eth_type_trans,
-	    struct sk_buff *, skb, struct net_device *, dev,
-	    unsigned short, protocol)
+SEC("fentry/eth_type_trans")
+int BPF_PROG(fentry_eth_type_trans, struct sk_buff *skb, struct net_device *dev,
+	     unsigned short protocol)
 {
 	int len, ifindex;
 
@@ -132,9 +132,9 @@ BPF_TRACE_3("fentry/eth_type_trans", fentry_eth_type_trans,
 	return 0;
 }
 
-BPF_TRACE_3("fexit/eth_type_trans", fexit_eth_type_trans,
-	    struct sk_buff *, skb, struct net_device *, dev,
-	    unsigned short, protocol)
+SEC("fexit/eth_type_trans")
+int BPF_PROG(fexit_eth_type_trans, struct sk_buff *skb, struct net_device *dev,
+	     unsigned short protocol)
 {
 	int len, ifindex;
 
diff --git a/tools/testing/selftests/bpf/progs/test_overhead.c b/tools/testing/selftests/bpf/progs/test_overhead.c
index 96c0124a04ba..48748297b860 100644
--- a/tools/testing/selftests/bpf/progs/test_overhead.c
+++ b/tools/testing/selftests/bpf/progs/test_overhead.c
@@ -1,39 +1,45 @@
 // SPDX-License-Identifier: GPL-2.0
 /* Copyright (c) 2019 Facebook */
+#include <stdbool.h>
+#include <stddef.h>
 #include <linux/bpf.h>
+#include <linux/ptrace.h>
 #include "bpf_helpers.h"
 #include "bpf_tracing.h"
 #include "bpf_trace_helpers.h"
 
+struct task_struct;
+
 SEC("kprobe/__set_task_comm")
-int prog1(struct pt_regs *ctx)
+int BPF_KPROBE(prog1, struct task_struct *tsk, const char *buf, bool exec)
 {
-	return 0;
+	return !tsk;
 }
 
 SEC("kretprobe/__set_task_comm")
-int prog2(struct pt_regs *ctx)
+int BPF_KRETPROBE(prog2,
+		  struct task_struct *tsk, const char *buf, bool exec,
+		  int ret)
 {
-	return 0;
+	return !PT_REGS_PARM1(ctx) && ret;
 }
 
 SEC("raw_tp/task_rename")
 int prog3(struct bpf_raw_tracepoint_args *ctx)
 {
-	return 0;
+	return !ctx->args[0];
 }
 
-struct task_struct;
-BPF_TRACE_3("fentry/__set_task_comm", prog4,
-	    struct task_struct *, tsk, const char *, buf, __u8, exec)
+SEC("fentry/__set_task_comm")
+int BPF_PROG(prog4, struct task_struct *tsk, const char *buf, bool exec)
 {
-	return 0;
+	return !tsk;
 }
 
-BPF_TRACE_3("fexit/__set_task_comm", prog5,
-	    struct task_struct *, tsk, const char *, buf, __u8, exec)
+SEC("fexit/__set_task_comm")
+int BPF_PROG(prog5, struct task_struct *tsk, const char *buf, bool exec)
 {
-	return 0;
+	return !tsk;
 }
 
 char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/test_perf_buffer.c b/tools/testing/selftests/bpf/progs/test_perf_buffer.c
index 07c09ca5546a..1fdc999031ac 100644
--- a/tools/testing/selftests/bpf/progs/test_perf_buffer.c
+++ b/tools/testing/selftests/bpf/progs/test_perf_buffer.c
@@ -4,6 +4,7 @@
 #include <linux/ptrace.h>
 #include <linux/bpf.h>
 #include "bpf_helpers.h"
+#include "bpf_trace_helpers.h"
 
 struct {
 	__uint(type, BPF_MAP_TYPE_PERF_EVENT_ARRAY);
@@ -12,7 +13,7 @@ struct {
 } perf_buf_map SEC(".maps");
 
 SEC("kprobe/sys_nanosleep")
-int handle_sys_nanosleep_entry(struct pt_regs *ctx)
+int BPF_KPROBE(handle_sys_nanosleep_entry)
 {
 	int cpu = bpf_get_smp_processor_id();
 
diff --git a/tools/testing/selftests/bpf/progs/test_probe_user.c b/tools/testing/selftests/bpf/progs/test_probe_user.c
index 1871e2ece0c4..5b570969e5c5 100644
--- a/tools/testing/selftests/bpf/progs/test_probe_user.c
+++ b/tools/testing/selftests/bpf/progs/test_probe_user.c
@@ -7,11 +7,12 @@
 
 #include "bpf_helpers.h"
 #include "bpf_tracing.h"
+#include "bpf_trace_helpers.h"
 
 static struct sockaddr_in old;
 
 SEC("kprobe/__sys_connect")
-int handle_sys_connect(struct pt_regs *ctx)
+int BPF_KPROBE(handle_sys_connect)
 {
 	void *ptr = (void *)PT_REGS_PARM2(ctx);
 	struct sockaddr_in new;
-- 
cgit v1.2.3-59-g8ed1b


From becd837eebc529e0b95f5994c52f93df3baed039 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Thu, 9 Jan 2020 20:25:31 +0100
Subject: r8169: prepare for exporting rtl_hw_phy_config

In preperation of factoring out the PHY configuration to a separate
source file this patch:
- avoids accessing rtl8169_private internals by passing the phy_device
  and mac_version as separate parameters
- renames rtl_hw_phy_config to r8169_hw_phy_config to avoid namespace
  clashes with other drivers for Realtek hardware

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 217 ++++++++++++++++--------------
 1 file changed, 117 insertions(+), 100 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 9c61ce2947c2..df3df5e70153 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -693,6 +693,8 @@ struct rtl8169_private {
 };
 
 typedef void (*rtl_generic_fct)(struct rtl8169_private *tp);
+typedef void (*rtl_phy_cfg_fct)(struct rtl8169_private *tp,
+				struct phy_device *phydev);
 
 MODULE_AUTHOR("Realtek and the Linux r8169 crew <netdev@vger.kernel.org>");
 MODULE_DESCRIPTION("RealTek RTL-8169 Gigabit Ethernet driver");
@@ -2349,7 +2351,8 @@ static void rtl8125_config_eee_phy(struct rtl8169_private *tp)
 	phy_modify_paged(phydev, 0xa6d, 0x14, 0x0010, 0x0000);
 }
 
-static void rtl8169s_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8169s_hw_phy_config(struct rtl8169_private *tp,
+				   struct phy_device *phydev)
 {
 	static const struct phy_reg phy_reg_init[] = {
 		{ 0x1f, 0x0001 },
@@ -2416,9 +2419,10 @@ static void rtl8169s_hw_phy_config(struct rtl8169_private *tp)
 	rtl_writephy_batch(tp, phy_reg_init);
 }
 
-static void rtl8169sb_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8169sb_hw_phy_config(struct rtl8169_private *tp,
+				    struct phy_device *phydev)
 {
-	phy_write_paged(tp->phydev, 0x0002, 0x01, 0x90d0);
+	phy_write_paged(phydev, 0x0002, 0x01, 0x90d0);
 }
 
 static void rtl8169scd_hw_phy_config_quirk(struct rtl8169_private *tp)
@@ -2432,7 +2436,8 @@ static void rtl8169scd_hw_phy_config_quirk(struct rtl8169_private *tp)
 	phy_write_paged(tp->phydev, 0x0001, 0x10, 0xf01b);
 }
 
-static void rtl8169scd_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8169scd_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
 {
 	static const struct phy_reg phy_reg_init[] = {
 		{ 0x1f, 0x0001 },
@@ -2479,7 +2484,8 @@ static void rtl8169scd_hw_phy_config(struct rtl8169_private *tp)
 	rtl8169scd_hw_phy_config_quirk(tp);
 }
 
-static void rtl8169sce_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8169sce_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
 {
 	static const struct phy_reg phy_reg_init[] = {
 		{ 0x1f, 0x0001 },
@@ -2532,7 +2538,8 @@ static void rtl8169sce_hw_phy_config(struct rtl8169_private *tp)
 	rtl_writephy_batch(tp, phy_reg_init);
 }
 
-static void rtl8168bb_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8168bb_hw_phy_config(struct rtl8169_private *tp,
+				    struct phy_device *phydev)
 {
 	rtl_writephy(tp, 0x1f, 0x0001);
 	rtl_patchphy(tp, 0x16, 1 << 0);
@@ -2540,25 +2547,29 @@ static void rtl8168bb_hw_phy_config(struct rtl8169_private *tp)
 	rtl_writephy(tp, 0x1f, 0x0000);
 }
 
-static void rtl8168bef_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8168bef_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
 {
-	phy_write_paged(tp->phydev, 0x0001, 0x10, 0xf41b);
+	phy_write_paged(phydev, 0x0001, 0x10, 0xf41b);
 }
 
-static void rtl8168cp_1_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8168cp_1_hw_phy_config(struct rtl8169_private *tp,
+				      struct phy_device *phydev)
 {
-	phy_write(tp->phydev, 0x1d, 0x0f00);
-	phy_write_paged(tp->phydev, 0x0002, 0x0c, 0x1ec8);
+	phy_write(phydev, 0x1d, 0x0f00);
+	phy_write_paged(phydev, 0x0002, 0x0c, 0x1ec8);
 }
 
-static void rtl8168cp_2_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8168cp_2_hw_phy_config(struct rtl8169_private *tp,
+				      struct phy_device *phydev)
 {
-	phy_set_bits(tp->phydev, 0x14, BIT(5));
-	phy_set_bits(tp->phydev, 0x0d, BIT(5));
-	phy_write_paged(tp->phydev, 0x0001, 0x1d, 0x3d98);
+	phy_set_bits(phydev, 0x14, BIT(5));
+	phy_set_bits(phydev, 0x0d, BIT(5));
+	phy_write_paged(phydev, 0x0001, 0x1d, 0x3d98);
 }
 
-static void rtl8168c_1_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8168c_1_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
 {
 	static const struct phy_reg phy_reg_init[] = {
 		{ 0x1f, 0x0001 },
@@ -2587,7 +2598,8 @@ static void rtl8168c_1_hw_phy_config(struct rtl8169_private *tp)
 	rtl_writephy(tp, 0x1f, 0x0000);
 }
 
-static void rtl8168c_2_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8168c_2_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
 {
 	static const struct phy_reg phy_reg_init[] = {
 		{ 0x1f, 0x0001 },
@@ -2615,7 +2627,8 @@ static void rtl8168c_2_hw_phy_config(struct rtl8169_private *tp)
 	rtl_writephy(tp, 0x1f, 0x0000);
 }
 
-static void rtl8168c_3_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8168c_3_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
 {
 	static const struct phy_reg phy_reg_init[] = {
 		{ 0x1f, 0x0001 },
@@ -2702,7 +2715,8 @@ static void rtl8168d_apply_firmware_cond(struct rtl8169_private *tp, u16 val)
 		rtl_apply_firmware(tp);
 }
 
-static void rtl8168d_1_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8168d_1_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
 {
 	rtl_writephy_batch(tp, rtl8168d_1_phy_reg_init_0);
 
@@ -2735,8 +2749,8 @@ static void rtl8168d_1_hw_phy_config(struct rtl8169_private *tp)
 				rtl_writephy(tp, 0x0d, val | set[i]);
 		}
 	} else {
-		phy_write_paged(tp->phydev, 0x0002, 0x05, 0x6662);
-		r8168d_phy_param(tp->phydev, 0x8330, 0xffff, 0x6662);
+		phy_write_paged(phydev, 0x0002, 0x05, 0x6662);
+		r8168d_phy_param(phydev, 0x8330, 0xffff, 0x6662);
 	}
 
 	/* RSET couple improve */
@@ -2753,7 +2767,8 @@ static void rtl8168d_1_hw_phy_config(struct rtl8169_private *tp)
 	rtl8168d_apply_firmware_cond(tp, 0xbf00);
 }
 
-static void rtl8168d_2_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8168d_2_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
 {
 	rtl_writephy_batch(tp, rtl8168d_1_phy_reg_init_0);
 
@@ -2777,8 +2792,8 @@ static void rtl8168d_2_hw_phy_config(struct rtl8169_private *tp)
 				rtl_writephy(tp, 0x0d, val | set[i]);
 		}
 	} else {
-		phy_write_paged(tp->phydev, 0x0002, 0x05, 0x2642);
-		r8168d_phy_param(tp->phydev, 0x8330, 0xffff, 0x2642);
+		phy_write_paged(phydev, 0x0002, 0x05, 0x2642);
+		r8168d_phy_param(phydev, 0x8330, 0xffff, 0x2642);
 	}
 
 	/* Fine tune PLL performance */
@@ -2794,7 +2809,8 @@ static void rtl8168d_2_hw_phy_config(struct rtl8169_private *tp)
 	rtl8168d_apply_firmware_cond(tp, 0xb300);
 }
 
-static void rtl8168d_3_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8168d_3_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
 {
 	static const struct phy_reg phy_reg_init[] = {
 		{ 0x1f, 0x0002 },
@@ -2849,17 +2865,19 @@ static void rtl8168d_3_hw_phy_config(struct rtl8169_private *tp)
 
 	rtl_writephy_batch(tp, phy_reg_init);
 
-	r8168d_modify_extpage(tp->phydev, 0x0023, 0x16, 0xffff, 0x0000);
+	r8168d_modify_extpage(phydev, 0x0023, 0x16, 0xffff, 0x0000);
 }
 
-static void rtl8168d_4_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8168d_4_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
 {
-	phy_write_paged(tp->phydev, 0x0001, 0x17, 0x0cc0);
-	r8168d_modify_extpage(tp->phydev, 0x002d, 0x18, 0xffff, 0x0040);
-	phy_set_bits(tp->phydev, 0x0d, BIT(5));
+	phy_write_paged(phydev, 0x0001, 0x17, 0x0cc0);
+	r8168d_modify_extpage(phydev, 0x002d, 0x18, 0xffff, 0x0040);
+	phy_set_bits(phydev, 0x0d, BIT(5));
 }
 
-static void rtl8168e_1_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8168e_1_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
 {
 	static const struct phy_reg phy_reg_init[] = {
 		/* Channel estimation fine tune */
@@ -2871,7 +2889,6 @@ static void rtl8168e_1_hw_phy_config(struct rtl8169_private *tp)
 		{ 0x14, 0x6420 },
 		{ 0x1f, 0x0000 },
 	};
-	struct phy_device *phydev = tp->phydev;
 
 	rtl_apply_firmware(tp);
 
@@ -2918,10 +2935,9 @@ static void rtl_rar_exgmac_set(struct rtl8169_private *tp, u8 *addr)
 	rtl_eri_write(tp, 0xf4, ERIAR_MASK_1111, w[1] | (w[2] << 16));
 }
 
-static void rtl8168e_2_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8168e_2_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
 {
-	struct phy_device *phydev = tp->phydev;
-
 	rtl_apply_firmware(tp);
 
 	/* Enable Delay cap */
@@ -2963,10 +2979,9 @@ static void rtl8168e_2_hw_phy_config(struct rtl8169_private *tp)
 	rtl_writephy(tp, 0x1f, 0x0000);
 }
 
-static void rtl8168f_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8168f_hw_phy_config(struct rtl8169_private *tp,
+				   struct phy_device *phydev)
 {
-	struct phy_device *phydev = tp->phydev;
-
 	/* For 4-corner performance improve */
 	r8168d_phy_param(phydev, 0x8b80, 0x0000, 0x0006);
 
@@ -2980,10 +2995,9 @@ static void rtl8168f_hw_phy_config(struct rtl8169_private *tp)
 	rtl8168f_config_eee_phy(tp);
 }
 
-static void rtl8168f_1_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8168f_1_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
 {
-	struct phy_device *phydev = tp->phydev;
-
 	rtl_apply_firmware(tp);
 
 	/* Channel estimation fine tune */
@@ -3003,26 +3017,26 @@ static void rtl8168f_1_hw_phy_config(struct rtl8169_private *tp)
 	/* Disable hiimpedance detection (RTCT) */
 	phy_write_paged(phydev, 0x0003, 0x01, 0x328a);
 
-	rtl8168f_hw_phy_config(tp);
+	rtl8168f_hw_phy_config(tp, phydev);
 
 	/* Improve 2-pair detection performance */
 	r8168d_phy_param(phydev, 0x8b85, 0x0000, 0x4000);
 }
 
-static void rtl8168f_2_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8168f_2_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
 {
 	rtl_apply_firmware(tp);
 
-	rtl8168f_hw_phy_config(tp);
+	rtl8168f_hw_phy_config(tp, phydev);
 }
 
-static void rtl8411_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8411_hw_phy_config(struct rtl8169_private *tp,
+				  struct phy_device *phydev)
 {
-	struct phy_device *phydev = tp->phydev;
-
 	rtl_apply_firmware(tp);
 
-	rtl8168f_hw_phy_config(tp);
+	rtl8168f_hw_phy_config(tp, phydev);
 
 	/* Improve 2-pair detection performance */
 	r8168d_phy_param(phydev, 0x8b85, 0x0000, 0x4000);
@@ -3078,36 +3092,37 @@ static void rtl8168g_phy_adjust_10m_aldps(struct rtl8169_private *tp)
 	phy_modify_paged(phydev, 0x0a43, 0x10, 0x0000, 0x1003);
 }
 
-static void rtl8168g_1_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8168g_1_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
 {
 	int ret;
 
 	rtl_apply_firmware(tp);
 
-	ret = phy_read_paged(tp->phydev, 0x0a46, 0x10);
+	ret = phy_read_paged(phydev, 0x0a46, 0x10);
 	if (ret & BIT(8))
-		phy_modify_paged(tp->phydev, 0x0bcc, 0x12, BIT(15), 0);
+		phy_modify_paged(phydev, 0x0bcc, 0x12, BIT(15), 0);
 	else
-		phy_modify_paged(tp->phydev, 0x0bcc, 0x12, 0, BIT(15));
+		phy_modify_paged(phydev, 0x0bcc, 0x12, 0, BIT(15));
 
-	ret = phy_read_paged(tp->phydev, 0x0a46, 0x13);
+	ret = phy_read_paged(phydev, 0x0a46, 0x13);
 	if (ret & BIT(8))
-		phy_modify_paged(tp->phydev, 0x0c41, 0x15, 0, BIT(1));
+		phy_modify_paged(phydev, 0x0c41, 0x15, 0, BIT(1));
 	else
-		phy_modify_paged(tp->phydev, 0x0c41, 0x15, BIT(1), 0);
+		phy_modify_paged(phydev, 0x0c41, 0x15, BIT(1), 0);
 
 	/* Enable PHY auto speed down */
-	phy_modify_paged(tp->phydev, 0x0a44, 0x11, 0, BIT(3) | BIT(2));
+	phy_modify_paged(phydev, 0x0a44, 0x11, 0, BIT(3) | BIT(2));
 
 	rtl8168g_phy_adjust_10m_aldps(tp);
 
 	/* EEE auto-fallback function */
-	phy_modify_paged(tp->phydev, 0x0a4b, 0x11, 0, BIT(2));
+	phy_modify_paged(phydev, 0x0a4b, 0x11, 0, BIT(2));
 
 	/* Enable UC LPF tune function */
-	r8168g_phy_param(tp->phydev, 0x8012, 0x0000, 0x8000);
+	r8168g_phy_param(phydev, 0x8012, 0x0000, 0x8000);
 
-	phy_modify_paged(tp->phydev, 0x0c42, 0x11, BIT(13), BIT(14));
+	phy_modify_paged(phydev, 0x0c42, 0x11, BIT(13), BIT(14));
 
 	/* Improve SWR Efficiency */
 	rtl_writephy(tp, 0x1f, 0x0bcd);
@@ -3125,15 +3140,16 @@ static void rtl8168g_1_hw_phy_config(struct rtl8169_private *tp)
 	rtl8168g_config_eee_phy(tp);
 }
 
-static void rtl8168g_2_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8168g_2_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
 {
 	rtl_apply_firmware(tp);
 	rtl8168g_config_eee_phy(tp);
 }
 
-static void rtl8168h_1_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8168h_1_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
 {
-	struct phy_device *phydev = tp->phydev;
 	u16 dout_tapbin;
 	u32 data;
 
@@ -3177,10 +3193,10 @@ static void rtl8168h_1_hw_phy_config(struct rtl8169_private *tp)
 	phy_modify_paged(phydev, 0x0a42, 0x16, 0x0000, 0x0002);
 
 	/* enable GPHY 10M */
-	phy_modify_paged(tp->phydev, 0x0a44, 0x11, 0, BIT(11));
+	phy_modify_paged(phydev, 0x0a44, 0x11, 0, BIT(11));
 
 	/* SAR ADC performance */
-	phy_modify_paged(tp->phydev, 0x0bca, 0x17, BIT(12) | BIT(13), BIT(14));
+	phy_modify_paged(phydev, 0x0bca, 0x17, BIT(12) | BIT(13), BIT(14));
 
 	r8168g_phy_param(phydev, 0x803f, 0x3000, 0x0000);
 	r8168g_phy_param(phydev, 0x8047, 0x3000, 0x0000);
@@ -3191,7 +3207,7 @@ static void rtl8168h_1_hw_phy_config(struct rtl8169_private *tp)
 	r8168g_phy_param(phydev, 0x806f, 0x3000, 0x0000);
 
 	/* disable phy pfm mode */
-	phy_modify_paged(tp->phydev, 0x0a44, 0x11, BIT(7), 0);
+	phy_modify_paged(phydev, 0x0a44, 0x11, BIT(7), 0);
 
 	rtl8168g_disable_aldps(tp);
 	rtl8168h_config_eee_phy(tp);
@@ -3213,9 +3229,9 @@ static u16 rtl8168h_2_get_adc_bias_ioffset(struct rtl8169_private *tp)
 	return ioffset;
 }
 
-static void rtl8168h_2_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8168h_2_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
 {
-	struct phy_device *phydev = tp->phydev;
 	u16 ioffset, rlen;
 	u32 data;
 
@@ -3229,7 +3245,7 @@ static void rtl8168h_2_hw_phy_config(struct rtl8169_private *tp)
 	phy_modify_paged(phydev, 0x0a42, 0x16, 0x0000, 0x0002);
 
 	/* enable GPHY 10M */
-	phy_modify_paged(tp->phydev, 0x0a44, 0x11, 0, BIT(11));
+	phy_modify_paged(phydev, 0x0a44, 0x11, 0, BIT(11));
 
 	ioffset = rtl8168h_2_get_adc_bias_ioffset(tp);
 	if (ioffset != 0xffff)
@@ -3251,10 +3267,9 @@ static void rtl8168h_2_hw_phy_config(struct rtl8169_private *tp)
 	rtl8168g_config_eee_phy(tp);
 }
 
-static void rtl8168ep_1_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8168ep_1_hw_phy_config(struct rtl8169_private *tp,
+				      struct phy_device *phydev)
 {
-	struct phy_device *phydev = tp->phydev;
-
 	/* Enable PHY auto speed down */
 	phy_modify_paged(phydev, 0x0a44, 0x11, 0, BIT(3) | BIT(2));
 
@@ -3273,17 +3288,16 @@ static void rtl8168ep_1_hw_phy_config(struct rtl8169_private *tp)
 	rtl8168g_config_eee_phy(tp);
 }
 
-static void rtl8168ep_2_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8168ep_2_hw_phy_config(struct rtl8169_private *tp,
+				      struct phy_device *phydev)
 {
-	struct phy_device *phydev = tp->phydev;
-
 	rtl8168g_phy_adjust_10m_aldps(tp);
 
 	/* Enable UC LPF tune function */
 	r8168g_phy_param(phydev, 0x8012, 0x0000, 0x8000);
 
 	/* Set rg_sel_sdm_rate */
-	phy_modify_paged(tp->phydev, 0x0c42, 0x11, BIT(13), BIT(14));
+	phy_modify_paged(phydev, 0x0c42, 0x11, BIT(13), BIT(14));
 
 	/* Channel estimation parameters */
 	r8168g_phy_param(phydev, 0x80f3, 0xff00, 0x8b00);
@@ -3324,10 +3338,9 @@ static void rtl8168ep_2_hw_phy_config(struct rtl8169_private *tp)
 	rtl8168g_config_eee_phy(tp);
 }
 
-static void rtl8117_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8117_hw_phy_config(struct rtl8169_private *tp,
+				  struct phy_device *phydev)
 {
-	struct phy_device *phydev = tp->phydev;
-
 	/* CHN EST parameters adjust - fnet */
 	r8168g_phy_param(phydev, 0x808e, 0xff00, 0x4800);
 	r8168g_phy_param(phydev, 0x8090, 0xff00, 0xcc00);
@@ -3355,7 +3368,7 @@ static void rtl8117_hw_phy_config(struct rtl8169_private *tp)
 	r8168g_phy_param(phydev, 0x8011, 0x0000, 0x0800);
 
 	/* enable GPHY 10M */
-	phy_modify_paged(tp->phydev, 0x0a44, 0x11, 0, BIT(11));
+	phy_modify_paged(phydev, 0x0a44, 0x11, 0, BIT(11));
 
 	r8168g_phy_param(phydev, 0x8016, 0x0000, 0x0400);
 
@@ -3363,7 +3376,8 @@ static void rtl8117_hw_phy_config(struct rtl8169_private *tp)
 	rtl8168h_config_eee_phy(tp);
 }
 
-static void rtl8102e_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8102e_hw_phy_config(struct rtl8169_private *tp,
+				   struct phy_device *phydev)
 {
 	static const struct phy_reg phy_reg_init[] = {
 		{ 0x1f, 0x0003 },
@@ -3380,23 +3394,25 @@ static void rtl8102e_hw_phy_config(struct rtl8169_private *tp)
 	rtl_writephy_batch(tp, phy_reg_init);
 }
 
-static void rtl8105e_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8105e_hw_phy_config(struct rtl8169_private *tp,
+				   struct phy_device *phydev)
 {
 	/* Disable ALDPS before ram code */
-	phy_write(tp->phydev, 0x18, 0x0310);
+	phy_write(phydev, 0x18, 0x0310);
 	msleep(100);
 
 	rtl_apply_firmware(tp);
 
-	phy_write_paged(tp->phydev, 0x0005, 0x1a, 0x0000);
-	phy_write_paged(tp->phydev, 0x0004, 0x1c, 0x0000);
-	phy_write_paged(tp->phydev, 0x0001, 0x15, 0x7701);
+	phy_write_paged(phydev, 0x0005, 0x1a, 0x0000);
+	phy_write_paged(phydev, 0x0004, 0x1c, 0x0000);
+	phy_write_paged(phydev, 0x0001, 0x15, 0x7701);
 }
 
-static void rtl8402_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8402_hw_phy_config(struct rtl8169_private *tp,
+				  struct phy_device *phydev)
 {
 	/* Disable ALDPS before setting firmware */
-	phy_write(tp->phydev, 0x18, 0x0310);
+	phy_write(phydev, 0x18, 0x0310);
 	msleep(20);
 
 	rtl_apply_firmware(tp);
@@ -3409,7 +3425,8 @@ static void rtl8402_hw_phy_config(struct rtl8169_private *tp)
 	rtl_writephy(tp, 0x1f, 0x0000);
 }
 
-static void rtl8106e_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8106e_hw_phy_config(struct rtl8169_private *tp,
+				   struct phy_device *phydev)
 {
 	static const struct phy_reg phy_reg_init[] = {
 		{ 0x1f, 0x0004 },
@@ -3419,7 +3436,7 @@ static void rtl8106e_hw_phy_config(struct rtl8169_private *tp)
 	};
 
 	/* Disable ALDPS before ram code */
-	phy_write(tp->phydev, 0x18, 0x0310);
+	phy_write(phydev, 0x18, 0x0310);
 	msleep(100);
 
 	rtl_apply_firmware(tp);
@@ -3430,10 +3447,9 @@ static void rtl8106e_hw_phy_config(struct rtl8169_private *tp)
 	rtl_eri_write(tp, 0x1d0, ERIAR_MASK_0011, 0x0000);
 }
 
-static void rtl8125_1_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8125_1_hw_phy_config(struct rtl8169_private *tp,
+				    struct phy_device *phydev)
 {
-	struct phy_device *phydev = tp->phydev;
-
 	phy_modify_paged(phydev, 0xad4, 0x10, 0x03ff, 0x0084);
 	phy_modify_paged(phydev, 0xad4, 0x17, 0x0000, 0x0010);
 	phy_modify_paged(phydev, 0xad1, 0x13, 0x03ff, 0x0006);
@@ -3469,9 +3485,9 @@ static void rtl8125_1_hw_phy_config(struct rtl8169_private *tp)
 	rtl8125_config_eee_phy(tp);
 }
 
-static void rtl8125_2_hw_phy_config(struct rtl8169_private *tp)
+static void rtl8125_2_hw_phy_config(struct rtl8169_private *tp,
+				    struct phy_device *phydev)
 {
-	struct phy_device *phydev = tp->phydev;
 	int i;
 
 	phy_modify_paged(phydev, 0xad4, 0x17, 0x0000, 0x0010);
@@ -3534,9 +3550,11 @@ static void rtl8125_2_hw_phy_config(struct rtl8169_private *tp)
 	rtl8125_config_eee_phy(tp);
 }
 
-static void rtl_hw_phy_config(struct net_device *dev)
+static void r8169_hw_phy_config(struct rtl8169_private *tp,
+				struct phy_device *phydev,
+				enum mac_version ver)
 {
-	static const rtl_generic_fct phy_configs[] = {
+	static const rtl_phy_cfg_fct phy_configs[] = {
 		/* PCI devices. */
 		[RTL_GIGA_MAC_VER_02] = rtl8169s_hw_phy_config,
 		[RTL_GIGA_MAC_VER_03] = rtl8169s_hw_phy_config,
@@ -3593,10 +3611,9 @@ static void rtl_hw_phy_config(struct net_device *dev)
 		[RTL_GIGA_MAC_VER_60] = rtl8125_1_hw_phy_config,
 		[RTL_GIGA_MAC_VER_61] = rtl8125_2_hw_phy_config,
 	};
-	struct rtl8169_private *tp = netdev_priv(dev);
 
-	if (phy_configs[tp->mac_version])
-		phy_configs[tp->mac_version](tp);
+	if (phy_configs[ver])
+		phy_configs[ver](tp, phydev);
 }
 
 static void rtl_schedule_task(struct rtl8169_private *tp, enum rtl_flag flag)
@@ -3607,7 +3624,7 @@ static void rtl_schedule_task(struct rtl8169_private *tp, enum rtl_flag flag)
 
 static void rtl8169_init_phy(struct net_device *dev, struct rtl8169_private *tp)
 {
-	rtl_hw_phy_config(dev);
+	r8169_hw_phy_config(tp, tp->phydev, tp->mac_version);
 
 	if (tp->mac_version <= RTL_GIGA_MAC_VER_06) {
 		pci_write_config_byte(tp->pci_dev, PCI_LATENCY_TIMER, 0x40);
-- 
cgit v1.2.3-59-g8ed1b


From b5aed0b3c292a259a6b32fe3ab828a9854b9facb Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Thu, 9 Jan 2020 20:26:19 +0100
Subject: r8169: remove not needed debug print in rtl8169_init_phy

Remove a useless debug statement. This also allows to remove the
net_device parameter from rtl8169_init_phy().

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index df3df5e70153..7277d39f5b0a 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -3622,15 +3622,14 @@ static void rtl_schedule_task(struct rtl8169_private *tp, enum rtl_flag flag)
 		schedule_work(&tp->wk.work);
 }
 
-static void rtl8169_init_phy(struct net_device *dev, struct rtl8169_private *tp)
+static void rtl8169_init_phy(struct rtl8169_private *tp)
 {
 	r8169_hw_phy_config(tp, tp->phydev, tp->mac_version);
 
 	if (tp->mac_version <= RTL_GIGA_MAC_VER_06) {
 		pci_write_config_byte(tp->pci_dev, PCI_LATENCY_TIMER, 0x40);
 		pci_write_config_byte(tp->pci_dev, PCI_CACHE_LINE_SIZE, 0x08);
-		netif_dbg(tp, drv, dev,
-			  "Set MAC Reg C+CR Offset 0x82h = 0x01h\n");
+		/* set undocumented MAC Reg C+CR Offset 0x82h */
 		RTL_W8(tp, 0x82, 0x01);
 	}
 
@@ -6245,7 +6244,7 @@ static int rtl_open(struct net_device *dev)
 
 	napi_enable(&tp->napi);
 
-	rtl8169_init_phy(dev, tp);
+	rtl8169_init_phy(tp);
 
 	rtl_pll_power_up(tp);
 
@@ -6376,7 +6375,7 @@ static void __rtl8169_resume(struct net_device *dev)
 	netif_device_attach(dev);
 
 	rtl_pll_power_up(tp);
-	rtl8169_init_phy(dev, tp);
+	rtl8169_init_phy(tp);
 
 	phy_start(tp->phydev);
 
-- 
cgit v1.2.3-59-g8ed1b


From fc712387b10cd3a06b5bc9e3d0379fbd3eb9e2c5 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Thu, 9 Jan 2020 20:26:59 +0100
Subject: r8169: move RTL8169scd Gigabyte PHY quirk

In preparation of factoring out rtl8169scd_hw_phy_config() move this
quirk to rtl8169_init_phy().

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 18 +++++-------------
 1 file changed, 5 insertions(+), 13 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 7277d39f5b0a..3514de25d6c1 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -2425,17 +2425,6 @@ static void rtl8169sb_hw_phy_config(struct rtl8169_private *tp,
 	phy_write_paged(phydev, 0x0002, 0x01, 0x90d0);
 }
 
-static void rtl8169scd_hw_phy_config_quirk(struct rtl8169_private *tp)
-{
-	struct pci_dev *pdev = tp->pci_dev;
-
-	if ((pdev->subsystem_vendor != PCI_VENDOR_ID_GIGABYTE) ||
-	    (pdev->subsystem_device != 0xe000))
-		return;
-
-	phy_write_paged(tp->phydev, 0x0001, 0x10, 0xf01b);
-}
-
 static void rtl8169scd_hw_phy_config(struct rtl8169_private *tp,
 				     struct phy_device *phydev)
 {
@@ -2480,8 +2469,6 @@ static void rtl8169scd_hw_phy_config(struct rtl8169_private *tp,
 	};
 
 	rtl_writephy_batch(tp, phy_reg_init);
-
-	rtl8169scd_hw_phy_config_quirk(tp);
 }
 
 static void rtl8169sce_hw_phy_config(struct rtl8169_private *tp,
@@ -3633,6 +3620,11 @@ static void rtl8169_init_phy(struct rtl8169_private *tp)
 		RTL_W8(tp, 0x82, 0x01);
 	}
 
+	if (tp->mac_version == RTL_GIGA_MAC_VER_05 &&
+	    tp->pci_dev->subsystem_vendor == PCI_VENDOR_ID_GIGABYTE &&
+	    tp->pci_dev->subsystem_device == 0xe000)
+		phy_write_paged(tp->phydev, 0x0001, 0x10, 0xf01b);
+
 	/* We may have called phy_speed_down before */
 	phy_speed_up(tp->phydev);
 
-- 
cgit v1.2.3-59-g8ed1b


From 8cecc8f0ae2e9b12f34f08a2e18d436aa7f779bd Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Thu, 9 Jan 2020 20:27:33 +0100
Subject: r8169: change argument type of EEE PHY functions

These functions use only the phy_device member of rtl8169_private,
so we can pass the phy_device as parameter directly.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 42 +++++++++++++------------------
 1 file changed, 18 insertions(+), 24 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 3514de25d6c1..dcb6fedf8380 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -2318,34 +2318,28 @@ static void rtl8125_config_eee_mac(struct rtl8169_private *tp)
 	r8168_mac_ocp_modify(tp, 0xeb62, 0, BIT(2) | BIT(1));
 }
 
-static void rtl8168f_config_eee_phy(struct rtl8169_private *tp)
+static void rtl8168f_config_eee_phy(struct phy_device *phydev)
 {
-	struct phy_device *phydev = tp->phydev;
-
 	r8168d_modify_extpage(phydev, 0x0020, 0x15, 0, BIT(8));
 	r8168d_phy_param(phydev, 0x8b85, 0, BIT(13));
 }
 
-static void rtl8168g_config_eee_phy(struct rtl8169_private *tp)
+static void rtl8168g_config_eee_phy(struct phy_device *phydev)
 {
-	phy_modify_paged(tp->phydev, 0x0a43, 0x11, 0, BIT(4));
+	phy_modify_paged(phydev, 0x0a43, 0x11, 0, BIT(4));
 }
 
-static void rtl8168h_config_eee_phy(struct rtl8169_private *tp)
+static void rtl8168h_config_eee_phy(struct phy_device *phydev)
 {
-	struct phy_device *phydev = tp->phydev;
-
-	rtl8168g_config_eee_phy(tp);
+	rtl8168g_config_eee_phy(phydev);
 
 	phy_modify_paged(phydev, 0xa4a, 0x11, 0x0000, 0x0200);
 	phy_modify_paged(phydev, 0xa42, 0x14, 0x0000, 0x0080);
 }
 
-static void rtl8125_config_eee_phy(struct rtl8169_private *tp)
+static void rtl8125_config_eee_phy(struct phy_device *phydev)
 {
-	struct phy_device *phydev = tp->phydev;
-
-	rtl8168h_config_eee_phy(tp);
+	rtl8168h_config_eee_phy(phydev);
 
 	phy_modify_paged(phydev, 0xa6d, 0x12, 0x0001, 0x0000);
 	phy_modify_paged(phydev, 0xa6d, 0x14, 0x0010, 0x0000);
@@ -2954,7 +2948,7 @@ static void rtl8168e_2_hw_phy_config(struct rtl8169_private *tp,
 	/* Improve 2-pair detection performance */
 	r8168d_phy_param(phydev, 0x8b85, 0x0000, 0x4000);
 
-	rtl8168f_config_eee_phy(tp);
+	rtl8168f_config_eee_phy(phydev);
 
 	/* Green feature */
 	rtl_writephy(tp, 0x1f, 0x0003);
@@ -2979,7 +2973,7 @@ static void rtl8168f_hw_phy_config(struct rtl8169_private *tp,
 	/* Improve 10M EEE waveform */
 	r8168d_phy_param(phydev, 0x8b86, 0x0000, 0x0001);
 
-	rtl8168f_config_eee_phy(tp);
+	rtl8168f_config_eee_phy(phydev);
 }
 
 static void rtl8168f_1_hw_phy_config(struct rtl8169_private *tp,
@@ -3124,14 +3118,14 @@ static void rtl8168g_1_hw_phy_config(struct rtl8169_private *tp,
 	rtl_writephy(tp, 0x1f, 0x0000);
 
 	rtl8168g_disable_aldps(tp);
-	rtl8168g_config_eee_phy(tp);
+	rtl8168g_config_eee_phy(phydev);
 }
 
 static void rtl8168g_2_hw_phy_config(struct rtl8169_private *tp,
 				     struct phy_device *phydev)
 {
 	rtl_apply_firmware(tp);
-	rtl8168g_config_eee_phy(tp);
+	rtl8168g_config_eee_phy(phydev);
 }
 
 static void rtl8168h_1_hw_phy_config(struct rtl8169_private *tp,
@@ -3197,7 +3191,7 @@ static void rtl8168h_1_hw_phy_config(struct rtl8169_private *tp,
 	phy_modify_paged(phydev, 0x0a44, 0x11, BIT(7), 0);
 
 	rtl8168g_disable_aldps(tp);
-	rtl8168h_config_eee_phy(tp);
+	rtl8168h_config_eee_phy(phydev);
 }
 
 static u16 rtl8168h_2_get_adc_bias_ioffset(struct rtl8169_private *tp)
@@ -3251,7 +3245,7 @@ static void rtl8168h_2_hw_phy_config(struct rtl8169_private *tp,
 	phy_modify_paged(phydev, 0x0a44, 0x11, BIT(7), 0);
 
 	rtl8168g_disable_aldps(tp);
-	rtl8168g_config_eee_phy(tp);
+	rtl8168g_config_eee_phy(phydev);
 }
 
 static void rtl8168ep_1_hw_phy_config(struct rtl8169_private *tp,
@@ -3272,7 +3266,7 @@ static void rtl8168ep_1_hw_phy_config(struct rtl8169_private *tp,
 	phy_modify_paged(phydev, 0x0c42, 0x11, BIT(13), BIT(14));
 
 	rtl8168g_disable_aldps(tp);
-	rtl8168g_config_eee_phy(tp);
+	rtl8168g_config_eee_phy(phydev);
 }
 
 static void rtl8168ep_2_hw_phy_config(struct rtl8169_private *tp,
@@ -3322,7 +3316,7 @@ static void rtl8168ep_2_hw_phy_config(struct rtl8169_private *tp,
 	rtl_writephy(tp, 0x1f, 0x0000);
 
 	rtl8168g_disable_aldps(tp);
-	rtl8168g_config_eee_phy(tp);
+	rtl8168g_config_eee_phy(phydev);
 }
 
 static void rtl8117_hw_phy_config(struct rtl8169_private *tp,
@@ -3360,7 +3354,7 @@ static void rtl8117_hw_phy_config(struct rtl8169_private *tp,
 	r8168g_phy_param(phydev, 0x8016, 0x0000, 0x0400);
 
 	rtl8168g_disable_aldps(tp);
-	rtl8168h_config_eee_phy(tp);
+	rtl8168h_config_eee_phy(phydev);
 }
 
 static void rtl8102e_hw_phy_config(struct rtl8169_private *tp,
@@ -3469,7 +3463,7 @@ static void rtl8125_1_hw_phy_config(struct rtl8169_private *tp,
 	phy_modify_paged(phydev, 0xa5c, 0x10, 0x0400, 0x0000);
 	phy_modify_paged(phydev, 0xa44, 0x11, 0x0000, 0x0800);
 
-	rtl8125_config_eee_phy(tp);
+	rtl8125_config_eee_phy(phydev);
 }
 
 static void rtl8125_2_hw_phy_config(struct rtl8169_private *tp,
@@ -3534,7 +3528,7 @@ static void rtl8125_2_hw_phy_config(struct rtl8169_private *tp,
 	phy_modify_paged(phydev, 0xa86, 0x15, 0x0001, 0x0000);
 	phy_modify_paged(phydev, 0xa44, 0x11, 0x0000, 0x0800);
 
-	rtl8125_config_eee_phy(tp);
+	rtl8125_config_eee_phy(phydev);
 }
 
 static void r8169_hw_phy_config(struct rtl8169_private *tp,
-- 
cgit v1.2.3-59-g8ed1b


From 4c867e36b89e820d6ddd901cc75b943a1e511be4 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Thu, 9 Jan 2020 20:28:08 +0100
Subject: r8169: change argument type of RTL8168g-specific PHY config functions

These functions use only the phy_device member of rtl8169_private,
so we can pass the phy_device as parameter directly.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 26 ++++++++++++--------------
 1 file changed, 12 insertions(+), 14 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index dcb6fedf8380..94bad2b09293 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -3058,15 +3058,13 @@ static void rtl8411_hw_phy_config(struct rtl8169_private *tp,
 	rtl_writephy(tp, 0x1f, 0x0000);
 }
 
-static void rtl8168g_disable_aldps(struct rtl8169_private *tp)
+static void rtl8168g_disable_aldps(struct phy_device *phydev)
 {
-	phy_modify_paged(tp->phydev, 0x0a43, 0x10, BIT(2), 0);
+	phy_modify_paged(phydev, 0x0a43, 0x10, BIT(2), 0);
 }
 
-static void rtl8168g_phy_adjust_10m_aldps(struct rtl8169_private *tp)
+static void rtl8168g_phy_adjust_10m_aldps(struct phy_device *phydev)
 {
-	struct phy_device *phydev = tp->phydev;
-
 	phy_modify_paged(phydev, 0x0bcc, 0x14, BIT(8), 0);
 	phy_modify_paged(phydev, 0x0a44, 0x11, 0, BIT(7) | BIT(6));
 	r8168g_phy_param(phydev, 0x8084, 0x6000, 0x0000);
@@ -3095,7 +3093,7 @@ static void rtl8168g_1_hw_phy_config(struct rtl8169_private *tp,
 	/* Enable PHY auto speed down */
 	phy_modify_paged(phydev, 0x0a44, 0x11, 0, BIT(3) | BIT(2));
 
-	rtl8168g_phy_adjust_10m_aldps(tp);
+	rtl8168g_phy_adjust_10m_aldps(phydev);
 
 	/* EEE auto-fallback function */
 	phy_modify_paged(phydev, 0x0a4b, 0x11, 0, BIT(2));
@@ -3117,7 +3115,7 @@ static void rtl8168g_1_hw_phy_config(struct rtl8169_private *tp,
 	rtl_writephy(tp, 0x14, 0x1065);
 	rtl_writephy(tp, 0x1f, 0x0000);
 
-	rtl8168g_disable_aldps(tp);
+	rtl8168g_disable_aldps(phydev);
 	rtl8168g_config_eee_phy(phydev);
 }
 
@@ -3190,7 +3188,7 @@ static void rtl8168h_1_hw_phy_config(struct rtl8169_private *tp,
 	/* disable phy pfm mode */
 	phy_modify_paged(phydev, 0x0a44, 0x11, BIT(7), 0);
 
-	rtl8168g_disable_aldps(tp);
+	rtl8168g_disable_aldps(phydev);
 	rtl8168h_config_eee_phy(phydev);
 }
 
@@ -3244,7 +3242,7 @@ static void rtl8168h_2_hw_phy_config(struct rtl8169_private *tp,
 	/* disable phy pfm mode */
 	phy_modify_paged(phydev, 0x0a44, 0x11, BIT(7), 0);
 
-	rtl8168g_disable_aldps(tp);
+	rtl8168g_disable_aldps(phydev);
 	rtl8168g_config_eee_phy(phydev);
 }
 
@@ -3254,7 +3252,7 @@ static void rtl8168ep_1_hw_phy_config(struct rtl8169_private *tp,
 	/* Enable PHY auto speed down */
 	phy_modify_paged(phydev, 0x0a44, 0x11, 0, BIT(3) | BIT(2));
 
-	rtl8168g_phy_adjust_10m_aldps(tp);
+	rtl8168g_phy_adjust_10m_aldps(phydev);
 
 	/* Enable EEE auto-fallback function */
 	phy_modify_paged(phydev, 0x0a4b, 0x11, 0, BIT(2));
@@ -3265,14 +3263,14 @@ static void rtl8168ep_1_hw_phy_config(struct rtl8169_private *tp,
 	/* set rg_sel_sdm_rate */
 	phy_modify_paged(phydev, 0x0c42, 0x11, BIT(13), BIT(14));
 
-	rtl8168g_disable_aldps(tp);
+	rtl8168g_disable_aldps(phydev);
 	rtl8168g_config_eee_phy(phydev);
 }
 
 static void rtl8168ep_2_hw_phy_config(struct rtl8169_private *tp,
 				      struct phy_device *phydev)
 {
-	rtl8168g_phy_adjust_10m_aldps(tp);
+	rtl8168g_phy_adjust_10m_aldps(phydev);
 
 	/* Enable UC LPF tune function */
 	r8168g_phy_param(phydev, 0x8012, 0x0000, 0x8000);
@@ -3315,7 +3313,7 @@ static void rtl8168ep_2_hw_phy_config(struct rtl8169_private *tp,
 	rtl_writephy(tp, 0x14, 0x1065);
 	rtl_writephy(tp, 0x1f, 0x0000);
 
-	rtl8168g_disable_aldps(tp);
+	rtl8168g_disable_aldps(phydev);
 	rtl8168g_config_eee_phy(phydev);
 }
 
@@ -3353,7 +3351,7 @@ static void rtl8117_hw_phy_config(struct rtl8169_private *tp,
 
 	r8168g_phy_param(phydev, 0x8016, 0x0000, 0x0400);
 
-	rtl8168g_disable_aldps(tp);
+	rtl8168g_disable_aldps(phydev);
 	rtl8168h_config_eee_phy(phydev);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 8aa9a5d3f6af583ef6ba1f8ac66142c59ef5223a Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Thu, 9 Jan 2020 20:28:45 +0100
Subject: r8169: switch to phylib functions in rtl_writephy_batch

Switch rtl_writephy_batch() to phylib functions, as a result we can
avoid passing a rtl8169_private parameter.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 39 +++++++++++++++++--------------
 1 file changed, 21 insertions(+), 18 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 94bad2b09293..dccc5a1d316f 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -2276,16 +2276,20 @@ struct phy_reg {
 	u16 val;
 };
 
-static void __rtl_writephy_batch(struct rtl8169_private *tp,
+static void __rtl_writephy_batch(struct phy_device *phydev,
 				 const struct phy_reg *regs, int len)
 {
+	phy_lock_mdio_bus(phydev);
+
 	while (len-- > 0) {
-		rtl_writephy(tp, regs->reg, regs->val);
+		__phy_write(phydev, regs->reg, regs->val);
 		regs++;
 	}
+
+	phy_unlock_mdio_bus(phydev);
 }
 
-#define rtl_writephy_batch(tp, a) __rtl_writephy_batch(tp, a, ARRAY_SIZE(a))
+#define rtl_writephy_batch(p, a) __rtl_writephy_batch(p, a, ARRAY_SIZE(a))
 
 static void rtl_release_firmware(struct rtl8169_private *tp)
 {
@@ -2410,7 +2414,7 @@ static void rtl8169s_hw_phy_config(struct rtl8169_private *tp,
 		{ 0x00, 0x9200 }
 	};
 
-	rtl_writephy_batch(tp, phy_reg_init);
+	rtl_writephy_batch(phydev, phy_reg_init);
 }
 
 static void rtl8169sb_hw_phy_config(struct rtl8169_private *tp,
@@ -2462,7 +2466,7 @@ static void rtl8169scd_hw_phy_config(struct rtl8169_private *tp,
 		{ 0x1f, 0x0000 }
 	};
 
-	rtl_writephy_batch(tp, phy_reg_init);
+	rtl_writephy_batch(phydev, phy_reg_init);
 }
 
 static void rtl8169sce_hw_phy_config(struct rtl8169_private *tp,
@@ -2516,7 +2520,7 @@ static void rtl8169sce_hw_phy_config(struct rtl8169_private *tp,
 		{ 0x1f, 0x0000 }
 	};
 
-	rtl_writephy_batch(tp, phy_reg_init);
+	rtl_writephy_batch(phydev, phy_reg_init);
 }
 
 static void rtl8168bb_hw_phy_config(struct rtl8169_private *tp,
@@ -2572,7 +2576,7 @@ static void rtl8168c_1_hw_phy_config(struct rtl8169_private *tp,
 		{ 0x09, 0x0000 }
 	};
 
-	rtl_writephy_batch(tp, phy_reg_init);
+	rtl_writephy_batch(phydev, phy_reg_init);
 
 	rtl_patchphy(tp, 0x14, 1 << 5);
 	rtl_patchphy(tp, 0x0d, 1 << 5);
@@ -2600,7 +2604,7 @@ static void rtl8168c_2_hw_phy_config(struct rtl8169_private *tp,
 		{ 0x1f, 0x0000 }
 	};
 
-	rtl_writephy_batch(tp, phy_reg_init);
+	rtl_writephy_batch(phydev, phy_reg_init);
 
 	rtl_patchphy(tp, 0x16, 1 << 0);
 	rtl_patchphy(tp, 0x14, 1 << 5);
@@ -2623,7 +2627,7 @@ static void rtl8168c_3_hw_phy_config(struct rtl8169_private *tp,
 		{ 0x1f, 0x0000 }
 	};
 
-	rtl_writephy_batch(tp, phy_reg_init);
+	rtl_writephy_batch(phydev, phy_reg_init);
 
 	rtl_patchphy(tp, 0x16, 1 << 0);
 	rtl_patchphy(tp, 0x14, 1 << 5);
@@ -2699,7 +2703,7 @@ static void rtl8168d_apply_firmware_cond(struct rtl8169_private *tp, u16 val)
 static void rtl8168d_1_hw_phy_config(struct rtl8169_private *tp,
 				     struct phy_device *phydev)
 {
-	rtl_writephy_batch(tp, rtl8168d_1_phy_reg_init_0);
+	rtl_writephy_batch(phydev, rtl8168d_1_phy_reg_init_0);
 
 	/*
 	 * Rx Error Issue
@@ -2712,7 +2716,7 @@ static void rtl8168d_1_hw_phy_config(struct rtl8169_private *tp,
 	if (rtl8168d_efuse_read(tp, 0x01) == 0xb1) {
 		int val;
 
-		rtl_writephy_batch(tp, rtl8168d_1_phy_reg_init_1);
+		rtl_writephy_batch(phydev, rtl8168d_1_phy_reg_init_1);
 
 		val = rtl_readphy(tp, 0x0d);
 
@@ -2751,12 +2755,12 @@ static void rtl8168d_1_hw_phy_config(struct rtl8169_private *tp,
 static void rtl8168d_2_hw_phy_config(struct rtl8169_private *tp,
 				     struct phy_device *phydev)
 {
-	rtl_writephy_batch(tp, rtl8168d_1_phy_reg_init_0);
+	rtl_writephy_batch(phydev, rtl8168d_1_phy_reg_init_0);
 
 	if (rtl8168d_efuse_read(tp, 0x01) == 0xb1) {
 		int val;
 
-		rtl_writephy_batch(tp, rtl8168d_1_phy_reg_init_1);
+		rtl_writephy_batch(phydev, rtl8168d_1_phy_reg_init_1);
 
 		val = rtl_readphy(tp, 0x0d);
 		if ((val & 0x00ff) != 0x006c) {
@@ -2844,8 +2848,7 @@ static void rtl8168d_3_hw_phy_config(struct rtl8169_private *tp,
 		{ 0x1f, 0x0000 },
 	};
 
-	rtl_writephy_batch(tp, phy_reg_init);
-
+	rtl_writephy_batch(phydev, phy_reg_init);
 	r8168d_modify_extpage(phydev, 0x0023, 0x16, 0xffff, 0x0000);
 }
 
@@ -2876,7 +2879,7 @@ static void rtl8168e_1_hw_phy_config(struct rtl8169_private *tp,
 	/* Enable Delay cap */
 	r8168d_phy_param(phydev, 0x8b80, 0xffff, 0xc896);
 
-	rtl_writephy_batch(tp, phy_reg_init);
+	rtl_writephy_batch(phydev, phy_reg_init);
 
 	/* Update PFM & 10M TX idle timer */
 	r8168d_modify_extpage(phydev, 0x002f, 0x15, 0xffff, 0x1919);
@@ -3370,7 +3373,7 @@ static void rtl8102e_hw_phy_config(struct rtl8169_private *tp,
 	rtl_patchphy(tp, 0x19, 1 << 13);
 	rtl_patchphy(tp, 0x10, 1 << 15);
 
-	rtl_writephy_batch(tp, phy_reg_init);
+	rtl_writephy_batch(phydev, phy_reg_init);
 }
 
 static void rtl8105e_hw_phy_config(struct rtl8169_private *tp,
@@ -3421,7 +3424,7 @@ static void rtl8106e_hw_phy_config(struct rtl8169_private *tp,
 	rtl_apply_firmware(tp);
 
 	rtl_eri_write(tp, 0x1b0, ERIAR_MASK_0011, 0x0000);
-	rtl_writephy_batch(tp, phy_reg_init);
+	rtl_writephy_batch(phydev, phy_reg_init);
 
 	rtl_eri_write(tp, 0x1d0, ERIAR_MASK_0011, 0x0000);
 }
-- 
cgit v1.2.3-59-g8ed1b


From 8d46f6207a8948d179b107afb4dcd03c4b96c37f Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Thu, 9 Jan 2020 20:29:21 +0100
Subject: r8169: move setting ERI register 0x1d0 for RTL8106

Writing this ERI register is a MAC setting, so move it to
rtl_hw_start_8106().

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index dccc5a1d316f..d157c971c28e 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -3425,8 +3425,6 @@ static void rtl8106e_hw_phy_config(struct rtl8169_private *tp,
 
 	rtl_eri_write(tp, 0x1b0, ERIAR_MASK_0011, 0x0000);
 	rtl_writephy_batch(phydev, phy_reg_init);
-
-	rtl_eri_write(tp, 0x1d0, ERIAR_MASK_0011, 0x0000);
 }
 
 static void rtl8125_1_hw_phy_config(struct rtl8169_private *tp,
@@ -4999,6 +4997,8 @@ static void rtl_hw_start_8106(struct rtl8169_private *tp)
 	RTL_W8(tp, MCU, RTL_R8(tp, MCU) | EN_NDP | EN_OOB_RESET);
 	RTL_W8(tp, DLLPR, RTL_R8(tp, DLLPR) & ~PFM_EN);
 
+	rtl_eri_write(tp, 0x1d0, ERIAR_MASK_0011, 0x0000);
+
 	rtl_pcie_state_l2l3_disable(tp);
 	rtl_hw_aspm_clkreq_enable(tp, true);
 }
-- 
cgit v1.2.3-59-g8ed1b


From 6d7a631e27ad526ea2e47082d5d84167d0fdc984 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Thu, 9 Jan 2020 20:30:00 +0100
Subject: r8169: move disabling MAC EEE for RTL8402/RTL8106e

Move configuring EEE on MAC side out of the PHY configuration.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index d157c971c28e..3fb3f2ac6eab 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -3400,7 +3400,6 @@ static void rtl8402_hw_phy_config(struct rtl8169_private *tp,
 	rtl_apply_firmware(tp);
 
 	/* EEE setting */
-	rtl_eri_write(tp, 0x1b0, ERIAR_MASK_0011, 0x0000);
 	rtl_writephy(tp, 0x1f, 0x0004);
 	rtl_writephy(tp, 0x10, 0x401f);
 	rtl_writephy(tp, 0x19, 0x7030);
@@ -3423,7 +3422,6 @@ static void rtl8106e_hw_phy_config(struct rtl8169_private *tp,
 
 	rtl_apply_firmware(tp);
 
-	rtl_eri_write(tp, 0x1b0, ERIAR_MASK_0011, 0x0000);
 	rtl_writephy_batch(phydev, phy_reg_init);
 }
 
@@ -4983,6 +4981,9 @@ static void rtl_hw_start_8402(struct rtl8169_private *tp)
 	rtl_eri_write(tp, 0xb8, ERIAR_MASK_0011, 0x0000);
 	rtl_w0w1_eri(tp, 0x0d4, ERIAR_MASK_0011, 0x0e00, 0xff00);
 
+	/* disable EEE */
+	rtl_eri_write(tp, 0x1b0, ERIAR_MASK_0011, 0x0000);
+
 	rtl_pcie_state_l2l3_disable(tp);
 }
 
@@ -4999,6 +5000,9 @@ static void rtl_hw_start_8106(struct rtl8169_private *tp)
 
 	rtl_eri_write(tp, 0x1d0, ERIAR_MASK_0011, 0x0000);
 
+	/* disable EEE */
+	rtl_eri_write(tp, 0x1b0, ERIAR_MASK_0011, 0x0000);
+
 	rtl_pcie_state_l2l3_disable(tp);
 	rtl_hw_aspm_clkreq_enable(tp, true);
 }
-- 
cgit v1.2.3-59-g8ed1b


From 11b3bedeb043db88bfea2ce41bd3453b25ee9d45 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Thu, 9 Jan 2020 20:30:37 +0100
Subject: r8169: replace rtl_patchphy

Replace rtl_patchphy with phylib functions.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 42 ++++++++++++-------------------
 1 file changed, 16 insertions(+), 26 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 3fb3f2ac6eab..9765f49e7da9 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -1081,11 +1081,6 @@ static int rtl_readphy(struct rtl8169_private *tp, int location)
 	}
 }
 
-static void rtl_patchphy(struct rtl8169_private *tp, int reg_addr, int value)
-{
-	rtl_writephy(tp, reg_addr, rtl_readphy(tp, reg_addr) | value);
-}
-
 static void rtl_w0w1_phy(struct rtl8169_private *tp, int reg_addr, int p, int m)
 {
 	int val;
@@ -2527,7 +2522,7 @@ static void rtl8168bb_hw_phy_config(struct rtl8169_private *tp,
 				    struct phy_device *phydev)
 {
 	rtl_writephy(tp, 0x1f, 0x0001);
-	rtl_patchphy(tp, 0x16, 1 << 0);
+	phy_set_bits(phydev, 0x16, BIT(0));
 	rtl_writephy(tp, 0x10, 0xf41b);
 	rtl_writephy(tp, 0x1f, 0x0000);
 }
@@ -2578,9 +2573,8 @@ static void rtl8168c_1_hw_phy_config(struct rtl8169_private *tp,
 
 	rtl_writephy_batch(phydev, phy_reg_init);
 
-	rtl_patchphy(tp, 0x14, 1 << 5);
-	rtl_patchphy(tp, 0x0d, 1 << 5);
-	rtl_writephy(tp, 0x1f, 0x0000);
+	phy_set_bits(phydev, 0x14, BIT(5));
+	phy_set_bits(phydev, 0x0d, BIT(5));
 }
 
 static void rtl8168c_2_hw_phy_config(struct rtl8169_private *tp,
@@ -2606,10 +2600,9 @@ static void rtl8168c_2_hw_phy_config(struct rtl8169_private *tp,
 
 	rtl_writephy_batch(phydev, phy_reg_init);
 
-	rtl_patchphy(tp, 0x16, 1 << 0);
-	rtl_patchphy(tp, 0x14, 1 << 5);
-	rtl_patchphy(tp, 0x0d, 1 << 5);
-	rtl_writephy(tp, 0x1f, 0x0000);
+	phy_set_bits(phydev, 0x16, BIT(0));
+	phy_set_bits(phydev, 0x14, BIT(5));
+	phy_set_bits(phydev, 0x0d, BIT(5));
 }
 
 static void rtl8168c_3_hw_phy_config(struct rtl8169_private *tp,
@@ -2629,10 +2622,9 @@ static void rtl8168c_3_hw_phy_config(struct rtl8169_private *tp,
 
 	rtl_writephy_batch(phydev, phy_reg_init);
 
-	rtl_patchphy(tp, 0x16, 1 << 0);
-	rtl_patchphy(tp, 0x14, 1 << 5);
-	rtl_patchphy(tp, 0x0d, 1 << 5);
-	rtl_writephy(tp, 0x1f, 0x0000);
+	phy_set_bits(phydev, 0x16, BIT(0));
+	phy_set_bits(phydev, 0x14, BIT(5));
+	phy_set_bits(phydev, 0x0d, BIT(5));
 }
 
 static const struct phy_reg rtl8168d_1_phy_reg_init_0[] = {
@@ -2740,8 +2732,8 @@ static void rtl8168d_1_hw_phy_config(struct rtl8169_private *tp,
 
 	/* RSET couple improve */
 	rtl_writephy(tp, 0x1f, 0x0002);
-	rtl_patchphy(tp, 0x0d, 0x0300);
-	rtl_patchphy(tp, 0x0f, 0x0010);
+	phy_set_bits(phydev, 0x0d, 0x0300);
+	phy_set_bits(phydev, 0x0f, 0x0010);
 
 	/* Fine tune PLL performance */
 	rtl_writephy(tp, 0x1f, 0x0002);
@@ -2785,11 +2777,10 @@ static void rtl8168d_2_hw_phy_config(struct rtl8169_private *tp,
 	rtl_writephy(tp, 0x1f, 0x0002);
 	rtl_w0w1_phy(tp, 0x02, 0x0100, 0x0600);
 	rtl_w0w1_phy(tp, 0x03, 0x0000, 0xe000);
+	rtl_writephy(tp, 0x1f, 0x0000);
 
 	/* Switching regulator Slew rate */
-	rtl_writephy(tp, 0x1f, 0x0002);
-	rtl_patchphy(tp, 0x0f, 0x0017);
-	rtl_writephy(tp, 0x1f, 0x0000);
+	phy_modify_paged(phydev, 0x0002, 0x0f, 0x0000, 0x0017);
 
 	rtl8168d_apply_firmware_cond(tp, 0xb300);
 }
@@ -3368,10 +3359,9 @@ static void rtl8102e_hw_phy_config(struct rtl8169_private *tp,
 		{ 0x1f, 0x0000 }
 	};
 
-	rtl_writephy(tp, 0x1f, 0x0000);
-	rtl_patchphy(tp, 0x11, 1 << 12);
-	rtl_patchphy(tp, 0x19, 1 << 13);
-	rtl_patchphy(tp, 0x10, 1 << 15);
+	phy_set_bits(phydev, 0x11, BIT(12));
+	phy_set_bits(phydev, 0x19, BIT(13));
+	phy_set_bits(phydev, 0x10, BIT(15));
 
 	rtl_writephy_batch(phydev, phy_reg_init);
 }
-- 
cgit v1.2.3-59-g8ed1b


From c4d76995f520d2ce55f38d6f844d55daf411dd9c Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Thu, 9 Jan 2020 20:31:10 +0100
Subject: r8169: replace rtl_w0w1_phy

Replace rtl_w0w1_phy with phylib functions.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 34 +++++++++++--------------------
 1 file changed, 12 insertions(+), 22 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 9765f49e7da9..457c8cdec08f 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -1081,14 +1081,6 @@ static int rtl_readphy(struct rtl8169_private *tp, int location)
 	}
 }
 
-static void rtl_w0w1_phy(struct rtl8169_private *tp, int reg_addr, int p, int m)
-{
-	int val;
-
-	val = rtl_readphy(tp, reg_addr);
-	rtl_writephy(tp, reg_addr, (val & ~m) | p);
-}
-
 static void r8168d_modify_extpage(struct phy_device *phydev, int extpage,
 				  int reg, u16 mask, u16 val)
 {
@@ -2702,8 +2694,8 @@ static void rtl8168d_1_hw_phy_config(struct rtl8169_private *tp,
 	 * Fine Tune Switching regulator parameter
 	 */
 	rtl_writephy(tp, 0x1f, 0x0002);
-	rtl_w0w1_phy(tp, 0x0b, 0x0010, 0x00ef);
-	rtl_w0w1_phy(tp, 0x0c, 0xa200, 0x5d00);
+	phy_modify(phydev, 0x0b, 0x00ef, 0x0010);
+	phy_modify(phydev, 0x0c, 0x5d00, 0xa200);
 
 	if (rtl8168d_efuse_read(tp, 0x01) == 0xb1) {
 		int val;
@@ -2737,8 +2729,8 @@ static void rtl8168d_1_hw_phy_config(struct rtl8169_private *tp,
 
 	/* Fine tune PLL performance */
 	rtl_writephy(tp, 0x1f, 0x0002);
-	rtl_w0w1_phy(tp, 0x02, 0x0100, 0x0600);
-	rtl_w0w1_phy(tp, 0x03, 0x0000, 0xe000);
+	phy_modify(phydev, 0x02, 0x0600, 0x0100);
+	phy_clear_bits(phydev, 0x03, 0xe000);
 	rtl_writephy(tp, 0x1f, 0x0000);
 
 	rtl8168d_apply_firmware_cond(tp, 0xbf00);
@@ -2775,8 +2767,8 @@ static void rtl8168d_2_hw_phy_config(struct rtl8169_private *tp,
 
 	/* Fine tune PLL performance */
 	rtl_writephy(tp, 0x1f, 0x0002);
-	rtl_w0w1_phy(tp, 0x02, 0x0100, 0x0600);
-	rtl_w0w1_phy(tp, 0x03, 0x0000, 0xe000);
+	phy_modify(phydev, 0x02, 0x0600, 0x0100);
+	phy_clear_bits(phydev, 0x03, 0xe000);
 	rtl_writephy(tp, 0x1f, 0x0000);
 
 	/* Switching regulator Slew rate */
@@ -2929,7 +2921,7 @@ static void rtl8168e_2_hw_phy_config(struct rtl8169_private *tp,
 	/* For 4-corner performance improve */
 	rtl_writephy(tp, 0x1f, 0x0005);
 	rtl_writephy(tp, 0x05, 0x8b80);
-	rtl_w0w1_phy(tp, 0x17, 0x0006, 0x0000);
+	phy_set_bits(phydev, 0x17, 0x0006);
 	rtl_writephy(tp, 0x1f, 0x0000);
 
 	/* PHY auto speed down */
@@ -2946,12 +2938,10 @@ static void rtl8168e_2_hw_phy_config(struct rtl8169_private *tp,
 
 	/* Green feature */
 	rtl_writephy(tp, 0x1f, 0x0003);
-	rtl_w0w1_phy(tp, 0x19, 0x0001, 0x0000);
-	rtl_w0w1_phy(tp, 0x10, 0x0400, 0x0000);
-	rtl_writephy(tp, 0x1f, 0x0000);
-	rtl_writephy(tp, 0x1f, 0x0005);
-	rtl_w0w1_phy(tp, 0x01, 0x0100, 0x0000);
+	phy_set_bits(phydev, 0x19, BIT(0));
+	phy_set_bits(phydev, 0x10, BIT(10));
 	rtl_writephy(tp, 0x1f, 0x0000);
+	phy_modify_paged(phydev, 0x0005, 0x01, 0, BIT(8));
 }
 
 static void rtl8168f_hw_phy_config(struct rtl8169_private *tp,
@@ -3047,8 +3037,8 @@ static void rtl8411_hw_phy_config(struct rtl8169_private *tp,
 
 	/* Green feature */
 	rtl_writephy(tp, 0x1f, 0x0003);
-	rtl_w0w1_phy(tp, 0x19, 0x0000, 0x0001);
-	rtl_w0w1_phy(tp, 0x10, 0x0000, 0x0400);
+	phy_clear_bits(phydev, 0x19, BIT(0));
+	phy_clear_bits(phydev, 0x10, BIT(10));
 	rtl_writephy(tp, 0x1f, 0x0000);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 3ab077d2f4057a0b827d6533d6bbdd11cf31a284 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Thu, 9 Jan 2020 20:31:47 +0100
Subject: r8169: use phy_read/write instead of rtl_readphy/writephy

Replace rtl_writephy and rtl_readphy with the respective phylib
functions.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 107 +++++++++++++++---------------
 1 file changed, 53 insertions(+), 54 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 457c8cdec08f..595659c2b76e 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -2513,10 +2513,10 @@ static void rtl8169sce_hw_phy_config(struct rtl8169_private *tp,
 static void rtl8168bb_hw_phy_config(struct rtl8169_private *tp,
 				    struct phy_device *phydev)
 {
-	rtl_writephy(tp, 0x1f, 0x0001);
+	phy_write(phydev, 0x1f, 0x0001);
 	phy_set_bits(phydev, 0x16, BIT(0));
-	rtl_writephy(tp, 0x10, 0xf41b);
-	rtl_writephy(tp, 0x1f, 0x0000);
+	phy_write(phydev, 0x10, 0xf41b);
+	phy_write(phydev, 0x1f, 0x0000);
 }
 
 static void rtl8168bef_hw_phy_config(struct rtl8169_private *tp,
@@ -2671,15 +2671,16 @@ static const struct phy_reg rtl8168d_1_phy_reg_init_1[] = {
 
 static void rtl8168d_apply_firmware_cond(struct rtl8169_private *tp, u16 val)
 {
+	struct phy_device *phydev = tp->phydev;
 	u16 reg_val;
 
-	rtl_writephy(tp, 0x1f, 0x0005);
-	rtl_writephy(tp, 0x05, 0x001b);
-	reg_val = rtl_readphy(tp, 0x06);
-	rtl_writephy(tp, 0x1f, 0x0000);
+	phy_write(phydev, 0x1f, 0x0005);
+	phy_write(phydev, 0x05, 0x001b);
+	reg_val = phy_read(phydev, 0x06);
+	phy_write(phydev, 0x1f, 0x0000);
 
 	if (reg_val != val)
-		netif_warn(tp, hw, tp->dev, "chipset not ready for firmware\n");
+		phydev_warn(phydev, "chipset not ready for firmware\n");
 	else
 		rtl_apply_firmware(tp);
 }
@@ -2693,7 +2694,7 @@ static void rtl8168d_1_hw_phy_config(struct rtl8169_private *tp,
 	 * Rx Error Issue
 	 * Fine Tune Switching regulator parameter
 	 */
-	rtl_writephy(tp, 0x1f, 0x0002);
+	phy_write(phydev, 0x1f, 0x0002);
 	phy_modify(phydev, 0x0b, 0x00ef, 0x0010);
 	phy_modify(phydev, 0x0c, 0x5d00, 0xa200);
 
@@ -2702,7 +2703,7 @@ static void rtl8168d_1_hw_phy_config(struct rtl8169_private *tp,
 
 		rtl_writephy_batch(phydev, rtl8168d_1_phy_reg_init_1);
 
-		val = rtl_readphy(tp, 0x0d);
+		val = phy_read(phydev, 0x0d);
 
 		if ((val & 0x00ff) != 0x006c) {
 			static const u32 set[] = {
@@ -2711,11 +2712,11 @@ static void rtl8168d_1_hw_phy_config(struct rtl8169_private *tp,
 			};
 			int i;
 
-			rtl_writephy(tp, 0x1f, 0x0002);
+			phy_write(phydev, 0x1f, 0x0002);
 
 			val &= 0xff00;
 			for (i = 0; i < ARRAY_SIZE(set); i++)
-				rtl_writephy(tp, 0x0d, val | set[i]);
+				phy_write(phydev, 0x0d, val | set[i]);
 		}
 	} else {
 		phy_write_paged(phydev, 0x0002, 0x05, 0x6662);
@@ -2723,15 +2724,15 @@ static void rtl8168d_1_hw_phy_config(struct rtl8169_private *tp,
 	}
 
 	/* RSET couple improve */
-	rtl_writephy(tp, 0x1f, 0x0002);
+	phy_write(phydev, 0x1f, 0x0002);
 	phy_set_bits(phydev, 0x0d, 0x0300);
 	phy_set_bits(phydev, 0x0f, 0x0010);
 
 	/* Fine tune PLL performance */
-	rtl_writephy(tp, 0x1f, 0x0002);
+	phy_write(phydev, 0x1f, 0x0002);
 	phy_modify(phydev, 0x02, 0x0600, 0x0100);
 	phy_clear_bits(phydev, 0x03, 0xe000);
-	rtl_writephy(tp, 0x1f, 0x0000);
+	phy_write(phydev, 0x1f, 0x0000);
 
 	rtl8168d_apply_firmware_cond(tp, 0xbf00);
 }
@@ -2746,7 +2747,7 @@ static void rtl8168d_2_hw_phy_config(struct rtl8169_private *tp,
 
 		rtl_writephy_batch(phydev, rtl8168d_1_phy_reg_init_1);
 
-		val = rtl_readphy(tp, 0x0d);
+		val = phy_read(phydev, 0x0d);
 		if ((val & 0x00ff) != 0x006c) {
 			static const u32 set[] = {
 				0x0065, 0x0066, 0x0067, 0x0068,
@@ -2754,11 +2755,11 @@ static void rtl8168d_2_hw_phy_config(struct rtl8169_private *tp,
 			};
 			int i;
 
-			rtl_writephy(tp, 0x1f, 0x0002);
+			phy_write(phydev, 0x1f, 0x0002);
 
 			val &= 0xff00;
 			for (i = 0; i < ARRAY_SIZE(set); i++)
-				rtl_writephy(tp, 0x0d, val | set[i]);
+				phy_write(phydev, 0x0d, val | set[i]);
 		}
 	} else {
 		phy_write_paged(phydev, 0x0002, 0x05, 0x2642);
@@ -2766,10 +2767,10 @@ static void rtl8168d_2_hw_phy_config(struct rtl8169_private *tp,
 	}
 
 	/* Fine tune PLL performance */
-	rtl_writephy(tp, 0x1f, 0x0002);
+	phy_write(phydev, 0x1f, 0x0002);
 	phy_modify(phydev, 0x02, 0x0600, 0x0100);
 	phy_clear_bits(phydev, 0x03, 0xe000);
-	rtl_writephy(tp, 0x1f, 0x0000);
+	phy_write(phydev, 0x1f, 0x0000);
 
 	/* Switching regulator Slew rate */
 	phy_modify_paged(phydev, 0x0002, 0x0f, 0x0000, 0x0017);
@@ -2919,10 +2920,10 @@ static void rtl8168e_2_hw_phy_config(struct rtl8169_private *tp,
 	r8168d_phy_param(phydev, 0x8b76, 0xffff, 0x8000);
 
 	/* For 4-corner performance improve */
-	rtl_writephy(tp, 0x1f, 0x0005);
-	rtl_writephy(tp, 0x05, 0x8b80);
+	phy_write(phydev, 0x1f, 0x0005);
+	phy_write(phydev, 0x05, 0x8b80);
 	phy_set_bits(phydev, 0x17, 0x0006);
-	rtl_writephy(tp, 0x1f, 0x0000);
+	phy_write(phydev, 0x1f, 0x0000);
 
 	/* PHY auto speed down */
 	r8168d_modify_extpage(phydev, 0x002d, 0x18, 0x0000, 0x0010);
@@ -2937,10 +2938,10 @@ static void rtl8168e_2_hw_phy_config(struct rtl8169_private *tp,
 	rtl8168f_config_eee_phy(phydev);
 
 	/* Green feature */
-	rtl_writephy(tp, 0x1f, 0x0003);
+	phy_write(phydev, 0x1f, 0x0003);
 	phy_set_bits(phydev, 0x19, BIT(0));
 	phy_set_bits(phydev, 0x10, BIT(10));
-	rtl_writephy(tp, 0x1f, 0x0000);
+	phy_write(phydev, 0x1f, 0x0000);
 	phy_modify_paged(phydev, 0x0005, 0x01, 0, BIT(8));
 }
 
@@ -3036,10 +3037,10 @@ static void rtl8411_hw_phy_config(struct rtl8169_private *tp,
 	r8168d_phy_param(phydev, 0x8b85, 0x0000, 0x8000);
 
 	/* Green feature */
-	rtl_writephy(tp, 0x1f, 0x0003);
+	phy_write(phydev, 0x1f, 0x0003);
 	phy_clear_bits(phydev, 0x19, BIT(0));
 	phy_clear_bits(phydev, 0x10, BIT(10));
-	rtl_writephy(tp, 0x1f, 0x0000);
+	phy_write(phydev, 0x1f, 0x0000);
 }
 
 static void rtl8168g_disable_aldps(struct phy_device *phydev)
@@ -3088,16 +3089,16 @@ static void rtl8168g_1_hw_phy_config(struct rtl8169_private *tp,
 	phy_modify_paged(phydev, 0x0c42, 0x11, BIT(13), BIT(14));
 
 	/* Improve SWR Efficiency */
-	rtl_writephy(tp, 0x1f, 0x0bcd);
-	rtl_writephy(tp, 0x14, 0x5065);
-	rtl_writephy(tp, 0x14, 0xd065);
-	rtl_writephy(tp, 0x1f, 0x0bc8);
-	rtl_writephy(tp, 0x11, 0x5655);
-	rtl_writephy(tp, 0x1f, 0x0bcd);
-	rtl_writephy(tp, 0x14, 0x1065);
-	rtl_writephy(tp, 0x14, 0x9065);
-	rtl_writephy(tp, 0x14, 0x1065);
-	rtl_writephy(tp, 0x1f, 0x0000);
+	phy_write(phydev, 0x1f, 0x0bcd);
+	phy_write(phydev, 0x14, 0x5065);
+	phy_write(phydev, 0x14, 0xd065);
+	phy_write(phydev, 0x1f, 0x0bc8);
+	phy_write(phydev, 0x11, 0x5655);
+	phy_write(phydev, 0x1f, 0x0bcd);
+	phy_write(phydev, 0x14, 0x1065);
+	phy_write(phydev, 0x14, 0x9065);
+	phy_write(phydev, 0x14, 0x1065);
+	phy_write(phydev, 0x1f, 0x0000);
 
 	rtl8168g_disable_aldps(phydev);
 	rtl8168g_config_eee_phy(phydev);
@@ -3286,16 +3287,16 @@ static void rtl8168ep_2_hw_phy_config(struct rtl8169_private *tp,
 	r8168g_phy_param(phydev, 0x80d7, 0xff00, 0x8400);
 
 	/* Force PWM-mode */
-	rtl_writephy(tp, 0x1f, 0x0bcd);
-	rtl_writephy(tp, 0x14, 0x5065);
-	rtl_writephy(tp, 0x14, 0xd065);
-	rtl_writephy(tp, 0x1f, 0x0bc8);
-	rtl_writephy(tp, 0x12, 0x00ed);
-	rtl_writephy(tp, 0x1f, 0x0bcd);
-	rtl_writephy(tp, 0x14, 0x1065);
-	rtl_writephy(tp, 0x14, 0x9065);
-	rtl_writephy(tp, 0x14, 0x1065);
-	rtl_writephy(tp, 0x1f, 0x0000);
+	phy_write(phydev, 0x1f, 0x0bcd);
+	phy_write(phydev, 0x14, 0x5065);
+	phy_write(phydev, 0x14, 0xd065);
+	phy_write(phydev, 0x1f, 0x0bc8);
+	phy_write(phydev, 0x12, 0x00ed);
+	phy_write(phydev, 0x1f, 0x0bcd);
+	phy_write(phydev, 0x14, 0x1065);
+	phy_write(phydev, 0x14, 0x9065);
+	phy_write(phydev, 0x14, 0x1065);
+	phy_write(phydev, 0x1f, 0x0000);
 
 	rtl8168g_disable_aldps(phydev);
 	rtl8168g_config_eee_phy(phydev);
@@ -3380,10 +3381,10 @@ static void rtl8402_hw_phy_config(struct rtl8169_private *tp,
 	rtl_apply_firmware(tp);
 
 	/* EEE setting */
-	rtl_writephy(tp, 0x1f, 0x0004);
-	rtl_writephy(tp, 0x10, 0x401f);
-	rtl_writephy(tp, 0x19, 0x7030);
-	rtl_writephy(tp, 0x1f, 0x0000);
+	phy_write(phydev, 0x1f, 0x0004);
+	phy_write(phydev, 0x10, 0x401f);
+	phy_write(phydev, 0x19, 0x7030);
+	phy_write(phydev, 0x1f, 0x0000);
 }
 
 static void rtl8106e_hw_phy_config(struct rtl8169_private *tp,
@@ -4680,9 +4681,7 @@ static void rtl_hw_start_8168h_1(struct rtl8169_private *tp)
 
 	rtl_pcie_state_l2l3_disable(tp);
 
-	rtl_writephy(tp, 0x1f, 0x0c42);
-	rg_saw_cnt = (rtl_readphy(tp, 0x13) & 0x3fff);
-	rtl_writephy(tp, 0x1f, 0x0000);
+	rg_saw_cnt = phy_read_paged(tp->phydev, 0x0c42, 0x13) & 0x3fff;
 	if (rg_saw_cnt > 0) {
 		u16 sw_cnt_1ms_ini;
 
-- 
cgit v1.2.3-59-g8ed1b


From 00b522d1ebaffa66f17203f2b610a0aa4d16a900 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Thu, 9 Jan 2020 20:32:36 +0100
Subject: r8169: add phydev argument to rtl8168d_apply_firmware_cond

Pass the phy_device as parameter to rtl8168d_apply_firmware_cond(),
this avoids having to access rtl8169_private internals.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 595659c2b76e..e022f555153d 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -2669,9 +2669,10 @@ static const struct phy_reg rtl8168d_1_phy_reg_init_1[] = {
 	{ 0x1f, 0x0002 }
 };
 
-static void rtl8168d_apply_firmware_cond(struct rtl8169_private *tp, u16 val)
+static void rtl8168d_apply_firmware_cond(struct rtl8169_private *tp,
+					 struct phy_device *phydev,
+					 u16 val)
 {
-	struct phy_device *phydev = tp->phydev;
 	u16 reg_val;
 
 	phy_write(phydev, 0x1f, 0x0005);
@@ -2734,7 +2735,7 @@ static void rtl8168d_1_hw_phy_config(struct rtl8169_private *tp,
 	phy_clear_bits(phydev, 0x03, 0xe000);
 	phy_write(phydev, 0x1f, 0x0000);
 
-	rtl8168d_apply_firmware_cond(tp, 0xbf00);
+	rtl8168d_apply_firmware_cond(tp, phydev, 0xbf00);
 }
 
 static void rtl8168d_2_hw_phy_config(struct rtl8169_private *tp,
@@ -2775,7 +2776,7 @@ static void rtl8168d_2_hw_phy_config(struct rtl8169_private *tp,
 	/* Switching regulator Slew rate */
 	phy_modify_paged(phydev, 0x0002, 0x0f, 0x0000, 0x0017);
 
-	rtl8168d_apply_firmware_cond(tp, 0xb300);
+	rtl8168d_apply_firmware_cond(tp, phydev, 0xb300);
 }
 
 static void rtl8168d_3_hw_phy_config(struct rtl8169_private *tp,
-- 
cgit v1.2.3-59-g8ed1b


From 1c5be5e91d780bd2aaa2bdb1fe02265bb2a6ab36 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Thu, 9 Jan 2020 20:33:13 +0100
Subject: r8169: rename rtl_apply_firmware

Rename rtl_apply_firmware() to r8169_apply_firmware() before exporting
it to avoid namespace clashes with other drivers for Realtek hardware.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 32 +++++++++++++++----------------
 1 file changed, 16 insertions(+), 16 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index e022f555153d..f243f36952fa 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -2287,7 +2287,7 @@ static void rtl_release_firmware(struct rtl8169_private *tp)
 	}
 }
 
-static void rtl_apply_firmware(struct rtl8169_private *tp)
+static void r8169_apply_firmware(struct rtl8169_private *tp)
 {
 	/* TODO: release firmware if rtl_fw_write_firmware signals failure. */
 	if (tp->rtl_fw)
@@ -2683,7 +2683,7 @@ static void rtl8168d_apply_firmware_cond(struct rtl8169_private *tp,
 	if (reg_val != val)
 		phydev_warn(phydev, "chipset not ready for firmware\n");
 	else
-		rtl_apply_firmware(tp);
+		r8169_apply_firmware(tp);
 }
 
 static void rtl8168d_1_hw_phy_config(struct rtl8169_private *tp,
@@ -2859,7 +2859,7 @@ static void rtl8168e_1_hw_phy_config(struct rtl8169_private *tp,
 		{ 0x1f, 0x0000 },
 	};
 
-	rtl_apply_firmware(tp);
+	r8169_apply_firmware(tp);
 
 	/* Enable Delay cap */
 	r8168d_phy_param(phydev, 0x8b80, 0xffff, 0xc896);
@@ -2907,7 +2907,7 @@ static void rtl_rar_exgmac_set(struct rtl8169_private *tp, u8 *addr)
 static void rtl8168e_2_hw_phy_config(struct rtl8169_private *tp,
 				     struct phy_device *phydev)
 {
-	rtl_apply_firmware(tp);
+	r8169_apply_firmware(tp);
 
 	/* Enable Delay cap */
 	r8168d_modify_extpage(phydev, 0x00ac, 0x18, 0xffff, 0x0006);
@@ -2965,7 +2965,7 @@ static void rtl8168f_hw_phy_config(struct rtl8169_private *tp,
 static void rtl8168f_1_hw_phy_config(struct rtl8169_private *tp,
 				     struct phy_device *phydev)
 {
-	rtl_apply_firmware(tp);
+	r8169_apply_firmware(tp);
 
 	/* Channel estimation fine tune */
 	phy_write_paged(phydev, 0x0003, 0x09, 0xa20f);
@@ -2993,7 +2993,7 @@ static void rtl8168f_1_hw_phy_config(struct rtl8169_private *tp,
 static void rtl8168f_2_hw_phy_config(struct rtl8169_private *tp,
 				     struct phy_device *phydev)
 {
-	rtl_apply_firmware(tp);
+	r8169_apply_firmware(tp);
 
 	rtl8168f_hw_phy_config(tp, phydev);
 }
@@ -3001,7 +3001,7 @@ static void rtl8168f_2_hw_phy_config(struct rtl8169_private *tp,
 static void rtl8411_hw_phy_config(struct rtl8169_private *tp,
 				  struct phy_device *phydev)
 {
-	rtl_apply_firmware(tp);
+	r8169_apply_firmware(tp);
 
 	rtl8168f_hw_phy_config(tp, phydev);
 
@@ -3062,7 +3062,7 @@ static void rtl8168g_1_hw_phy_config(struct rtl8169_private *tp,
 {
 	int ret;
 
-	rtl_apply_firmware(tp);
+	r8169_apply_firmware(tp);
 
 	ret = phy_read_paged(phydev, 0x0a46, 0x10);
 	if (ret & BIT(8))
@@ -3108,7 +3108,7 @@ static void rtl8168g_1_hw_phy_config(struct rtl8169_private *tp,
 static void rtl8168g_2_hw_phy_config(struct rtl8169_private *tp,
 				     struct phy_device *phydev)
 {
-	rtl_apply_firmware(tp);
+	r8169_apply_firmware(tp);
 	rtl8168g_config_eee_phy(phydev);
 }
 
@@ -3118,7 +3118,7 @@ static void rtl8168h_1_hw_phy_config(struct rtl8169_private *tp,
 	u16 dout_tapbin;
 	u32 data;
 
-	rtl_apply_firmware(tp);
+	r8169_apply_firmware(tp);
 
 	/* CHN EST parameters adjust - giga master */
 	r8168g_phy_param(phydev, 0x809b, 0xf800, 0x8000);
@@ -3200,7 +3200,7 @@ static void rtl8168h_2_hw_phy_config(struct rtl8169_private *tp,
 	u16 ioffset, rlen;
 	u32 data;
 
-	rtl_apply_firmware(tp);
+	r8169_apply_firmware(tp);
 
 	/* CHIN EST parameter update */
 	r8168g_phy_param(phydev, 0x808a, 0x003f, 0x000a);
@@ -3365,7 +3365,7 @@ static void rtl8105e_hw_phy_config(struct rtl8169_private *tp,
 	phy_write(phydev, 0x18, 0x0310);
 	msleep(100);
 
-	rtl_apply_firmware(tp);
+	r8169_apply_firmware(tp);
 
 	phy_write_paged(phydev, 0x0005, 0x1a, 0x0000);
 	phy_write_paged(phydev, 0x0004, 0x1c, 0x0000);
@@ -3379,7 +3379,7 @@ static void rtl8402_hw_phy_config(struct rtl8169_private *tp,
 	phy_write(phydev, 0x18, 0x0310);
 	msleep(20);
 
-	rtl_apply_firmware(tp);
+	r8169_apply_firmware(tp);
 
 	/* EEE setting */
 	phy_write(phydev, 0x1f, 0x0004);
@@ -3402,7 +3402,7 @@ static void rtl8106e_hw_phy_config(struct rtl8169_private *tp,
 	phy_write(phydev, 0x18, 0x0310);
 	msleep(100);
 
-	rtl_apply_firmware(tp);
+	r8169_apply_firmware(tp);
 
 	rtl_writephy_batch(phydev, phy_reg_init);
 }
@@ -3494,7 +3494,7 @@ static void rtl8125_2_hw_phy_config(struct rtl8169_private *tp,
 	r8168g_phy_param(phydev, 0x8257, 0xffff, 0x020F);
 	r8168g_phy_param(phydev, 0x80ea, 0xffff, 0x7843);
 
-	rtl_apply_firmware(tp);
+	r8169_apply_firmware(tp);
 
 	phy_modify_paged(phydev, 0xd06, 0x14, 0x0000, 0x2000);
 
@@ -4857,7 +4857,7 @@ static void rtl_hw_start_8117(struct rtl8169_private *tp)
 	r8168_mac_ocp_write(tp, 0xc09e, 0x0000);
 
 	/* firmware is for MAC only */
-	rtl_apply_firmware(tp);
+	r8169_apply_firmware(tp);
 
 	rtl_hw_aspm_clkreq_enable(tp, true);
 }
-- 
cgit v1.2.3-59-g8ed1b


From 2992bdfa4ad272251f664cb1e5cbd93e5ea25f04 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Thu, 9 Jan 2020 20:34:05 +0100
Subject: r8169: add r8169.h

In preparation of factoring out PHY configuration to a separate source
file move commonly used definitions to new header file r8169.h.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169.h      | 78 +++++++++++++++++++++++++++++++
 drivers/net/ethernet/realtek/r8169_main.c | 70 +++------------------------
 2 files changed, 84 insertions(+), 64 deletions(-)
 create mode 100644 drivers/net/ethernet/realtek/r8169.h

diff --git a/drivers/net/ethernet/realtek/r8169.h b/drivers/net/ethernet/realtek/r8169.h
new file mode 100644
index 000000000000..22a6a057b11e
--- /dev/null
+++ b/drivers/net/ethernet/realtek/r8169.h
@@ -0,0 +1,78 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* r8169.h: RealTek 8169/8168/8101 ethernet driver.
+ *
+ * Copyright (c) 2002 ShuChen <shuchen@realtek.com.tw>
+ * Copyright (c) 2003 - 2007 Francois Romieu <romieu@fr.zoreil.com>
+ * Copyright (c) a lot of people too. Please respect their work.
+ *
+ * See MAINTAINERS file for support contact information.
+ */
+
+#include <linux/types.h>
+#include <linux/phy.h>
+
+enum mac_version {
+	/* support for ancient RTL_GIGA_MAC_VER_01 has been removed */
+	RTL_GIGA_MAC_VER_02,
+	RTL_GIGA_MAC_VER_03,
+	RTL_GIGA_MAC_VER_04,
+	RTL_GIGA_MAC_VER_05,
+	RTL_GIGA_MAC_VER_06,
+	RTL_GIGA_MAC_VER_07,
+	RTL_GIGA_MAC_VER_08,
+	RTL_GIGA_MAC_VER_09,
+	RTL_GIGA_MAC_VER_10,
+	RTL_GIGA_MAC_VER_11,
+	RTL_GIGA_MAC_VER_12,
+	RTL_GIGA_MAC_VER_13,
+	RTL_GIGA_MAC_VER_14,
+	RTL_GIGA_MAC_VER_15,
+	RTL_GIGA_MAC_VER_16,
+	RTL_GIGA_MAC_VER_17,
+	RTL_GIGA_MAC_VER_18,
+	RTL_GIGA_MAC_VER_19,
+	RTL_GIGA_MAC_VER_20,
+	RTL_GIGA_MAC_VER_21,
+	RTL_GIGA_MAC_VER_22,
+	RTL_GIGA_MAC_VER_23,
+	RTL_GIGA_MAC_VER_24,
+	RTL_GIGA_MAC_VER_25,
+	RTL_GIGA_MAC_VER_26,
+	RTL_GIGA_MAC_VER_27,
+	RTL_GIGA_MAC_VER_28,
+	RTL_GIGA_MAC_VER_29,
+	RTL_GIGA_MAC_VER_30,
+	RTL_GIGA_MAC_VER_31,
+	RTL_GIGA_MAC_VER_32,
+	RTL_GIGA_MAC_VER_33,
+	RTL_GIGA_MAC_VER_34,
+	RTL_GIGA_MAC_VER_35,
+	RTL_GIGA_MAC_VER_36,
+	RTL_GIGA_MAC_VER_37,
+	RTL_GIGA_MAC_VER_38,
+	RTL_GIGA_MAC_VER_39,
+	RTL_GIGA_MAC_VER_40,
+	RTL_GIGA_MAC_VER_41,
+	RTL_GIGA_MAC_VER_42,
+	RTL_GIGA_MAC_VER_43,
+	RTL_GIGA_MAC_VER_44,
+	RTL_GIGA_MAC_VER_45,
+	RTL_GIGA_MAC_VER_46,
+	RTL_GIGA_MAC_VER_47,
+	RTL_GIGA_MAC_VER_48,
+	RTL_GIGA_MAC_VER_49,
+	RTL_GIGA_MAC_VER_50,
+	RTL_GIGA_MAC_VER_51,
+	RTL_GIGA_MAC_VER_52,
+	RTL_GIGA_MAC_VER_60,
+	RTL_GIGA_MAC_VER_61,
+	RTL_GIGA_MAC_NONE
+};
+
+struct rtl8169_private;
+
+void r8169_apply_firmware(struct rtl8169_private *tp);
+u16 rtl8168h_2_get_adc_bias_ioffset(struct rtl8169_private *tp);
+u8 rtl8168d_efuse_read(struct rtl8169_private *tp, int reg_addr);
+void r8169_hw_phy_config(struct rtl8169_private *tp, struct phy_device *phydev,
+			 enum mac_version ver);
diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index f243f36952fa..7bbc9794d2b0 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -31,6 +31,7 @@
 #include <linux/ipv6.h>
 #include <net/ip6_checksum.h>
 
+#include "r8169.h"
 #include "r8169_firmware.h"
 
 #define MODULENAME "r8169"
@@ -84,64 +85,6 @@
 #define RTL_R16(tp, reg)		readw(tp->mmio_addr + (reg))
 #define RTL_R32(tp, reg)		readl(tp->mmio_addr + (reg))
 
-enum mac_version {
-	/* support for ancient RTL_GIGA_MAC_VER_01 has been removed */
-	RTL_GIGA_MAC_VER_02,
-	RTL_GIGA_MAC_VER_03,
-	RTL_GIGA_MAC_VER_04,
-	RTL_GIGA_MAC_VER_05,
-	RTL_GIGA_MAC_VER_06,
-	RTL_GIGA_MAC_VER_07,
-	RTL_GIGA_MAC_VER_08,
-	RTL_GIGA_MAC_VER_09,
-	RTL_GIGA_MAC_VER_10,
-	RTL_GIGA_MAC_VER_11,
-	RTL_GIGA_MAC_VER_12,
-	RTL_GIGA_MAC_VER_13,
-	RTL_GIGA_MAC_VER_14,
-	RTL_GIGA_MAC_VER_15,
-	RTL_GIGA_MAC_VER_16,
-	RTL_GIGA_MAC_VER_17,
-	RTL_GIGA_MAC_VER_18,
-	RTL_GIGA_MAC_VER_19,
-	RTL_GIGA_MAC_VER_20,
-	RTL_GIGA_MAC_VER_21,
-	RTL_GIGA_MAC_VER_22,
-	RTL_GIGA_MAC_VER_23,
-	RTL_GIGA_MAC_VER_24,
-	RTL_GIGA_MAC_VER_25,
-	RTL_GIGA_MAC_VER_26,
-	RTL_GIGA_MAC_VER_27,
-	RTL_GIGA_MAC_VER_28,
-	RTL_GIGA_MAC_VER_29,
-	RTL_GIGA_MAC_VER_30,
-	RTL_GIGA_MAC_VER_31,
-	RTL_GIGA_MAC_VER_32,
-	RTL_GIGA_MAC_VER_33,
-	RTL_GIGA_MAC_VER_34,
-	RTL_GIGA_MAC_VER_35,
-	RTL_GIGA_MAC_VER_36,
-	RTL_GIGA_MAC_VER_37,
-	RTL_GIGA_MAC_VER_38,
-	RTL_GIGA_MAC_VER_39,
-	RTL_GIGA_MAC_VER_40,
-	RTL_GIGA_MAC_VER_41,
-	RTL_GIGA_MAC_VER_42,
-	RTL_GIGA_MAC_VER_43,
-	RTL_GIGA_MAC_VER_44,
-	RTL_GIGA_MAC_VER_45,
-	RTL_GIGA_MAC_VER_46,
-	RTL_GIGA_MAC_VER_47,
-	RTL_GIGA_MAC_VER_48,
-	RTL_GIGA_MAC_VER_49,
-	RTL_GIGA_MAC_VER_50,
-	RTL_GIGA_MAC_VER_51,
-	RTL_GIGA_MAC_VER_52,
-	RTL_GIGA_MAC_VER_60,
-	RTL_GIGA_MAC_VER_61,
-	RTL_GIGA_MAC_NONE
-};
-
 #define JUMBO_1K	ETH_DATA_LEN
 #define JUMBO_4K	(4*1024 - ETH_HLEN - 2)
 #define JUMBO_6K	(6*1024 - ETH_HLEN - 2)
@@ -1362,7 +1305,7 @@ DECLARE_RTL_COND(rtl_efusear_cond)
 	return RTL_R32(tp, EFUSEAR) & EFUSEAR_FLAG;
 }
 
-static u8 rtl8168d_efuse_read(struct rtl8169_private *tp, int reg_addr)
+u8 rtl8168d_efuse_read(struct rtl8169_private *tp, int reg_addr)
 {
 	RTL_W32(tp, EFUSEAR, (reg_addr & EFUSEAR_REG_MASK) << EFUSEAR_REG_SHIFT);
 
@@ -2287,7 +2230,7 @@ static void rtl_release_firmware(struct rtl8169_private *tp)
 	}
 }
 
-static void r8169_apply_firmware(struct rtl8169_private *tp)
+void r8169_apply_firmware(struct rtl8169_private *tp)
 {
 	/* TODO: release firmware if rtl_fw_write_firmware signals failure. */
 	if (tp->rtl_fw)
@@ -3178,7 +3121,7 @@ static void rtl8168h_1_hw_phy_config(struct rtl8169_private *tp,
 	rtl8168h_config_eee_phy(phydev);
 }
 
-static u16 rtl8168h_2_get_adc_bias_ioffset(struct rtl8169_private *tp)
+u16 rtl8168h_2_get_adc_bias_ioffset(struct rtl8169_private *tp)
 {
 	u16 data1, data2, ioffset;
 
@@ -3510,9 +3453,8 @@ static void rtl8125_2_hw_phy_config(struct rtl8169_private *tp,
 	rtl8125_config_eee_phy(phydev);
 }
 
-static void r8169_hw_phy_config(struct rtl8169_private *tp,
-				struct phy_device *phydev,
-				enum mac_version ver)
+void r8169_hw_phy_config(struct rtl8169_private *tp, struct phy_device *phydev,
+			 enum mac_version ver)
 {
 	static const rtl_phy_cfg_fct phy_configs[] = {
 		/* PCI devices. */
-- 
cgit v1.2.3-59-g8ed1b


From 814c2bfd4f5f8cc84fead00b86b49a8cfc8f8e08 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Thu, 9 Jan 2020 20:34:55 +0100
Subject: r8169: factor out PHY configuration to r8169_phy_config.c

Move chip-specific PHY configurations to separate source file
r8169_phy_config.c. This improves maintainability of the driver.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/Makefile           |    2 +-
 drivers/net/ethernet/realtek/r8169_main.c       | 1291 ----------------------
 drivers/net/ethernet/realtek/r8169_phy_config.c | 1307 +++++++++++++++++++++++
 3 files changed, 1308 insertions(+), 1292 deletions(-)
 create mode 100644 drivers/net/ethernet/realtek/r8169_phy_config.c

diff --git a/drivers/net/ethernet/realtek/Makefile b/drivers/net/ethernet/realtek/Makefile
index d5304bad2372..2e1d78b106b0 100644
--- a/drivers/net/ethernet/realtek/Makefile
+++ b/drivers/net/ethernet/realtek/Makefile
@@ -6,5 +6,5 @@
 obj-$(CONFIG_8139CP) += 8139cp.o
 obj-$(CONFIG_8139TOO) += 8139too.o
 obj-$(CONFIG_ATP) += atp.o
-r8169-objs += r8169_main.o r8169_firmware.o
+r8169-objs += r8169_main.o r8169_firmware.o r8169_phy_config.o
 obj-$(CONFIG_R8169) += r8169.o
diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 7bbc9794d2b0..6d699df7d27f 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -636,8 +636,6 @@ struct rtl8169_private {
 };
 
 typedef void (*rtl_generic_fct)(struct rtl8169_private *tp);
-typedef void (*rtl_phy_cfg_fct)(struct rtl8169_private *tp,
-				struct phy_device *phydev);
 
 MODULE_AUTHOR("Realtek and the Linux r8169 crew <netdev@vger.kernel.org>");
 MODULE_DESCRIPTION("RealTek RTL-8169 Gigabit Ethernet driver");
@@ -1024,39 +1022,6 @@ static int rtl_readphy(struct rtl8169_private *tp, int location)
 	}
 }
 
-static void r8168d_modify_extpage(struct phy_device *phydev, int extpage,
-				  int reg, u16 mask, u16 val)
-{
-	int oldpage = phy_select_page(phydev, 0x0007);
-
-	__phy_write(phydev, 0x1e, extpage);
-	__phy_modify(phydev, reg, mask, val);
-
-	phy_restore_page(phydev, oldpage, 0);
-}
-
-static void r8168d_phy_param(struct phy_device *phydev, u16 parm,
-			     u16 mask, u16 val)
-{
-	int oldpage = phy_select_page(phydev, 0x0005);
-
-	__phy_write(phydev, 0x05, parm);
-	__phy_modify(phydev, 0x06, mask, val);
-
-	phy_restore_page(phydev, oldpage, 0);
-}
-
-static void r8168g_phy_param(struct phy_device *phydev, u16 parm,
-			     u16 mask, u16 val)
-{
-	int oldpage = phy_select_page(phydev, 0x0a43);
-
-	__phy_write(phydev, 0x13, parm);
-	__phy_modify(phydev, 0x14, mask, val);
-
-	phy_restore_page(phydev, oldpage, 0);
-}
-
 DECLARE_RTL_COND(rtl_ephyar_cond)
 {
 	return RTL_R32(tp, EPHYAR) & EPHYAR_FLAG;
@@ -2201,26 +2166,6 @@ static void rtl8169_get_mac_version(struct rtl8169_private *tp)
 	}
 }
 
-struct phy_reg {
-	u16 reg;
-	u16 val;
-};
-
-static void __rtl_writephy_batch(struct phy_device *phydev,
-				 const struct phy_reg *regs, int len)
-{
-	phy_lock_mdio_bus(phydev);
-
-	while (len-- > 0) {
-		__phy_write(phydev, regs->reg, regs->val);
-		regs++;
-	}
-
-	phy_unlock_mdio_bus(phydev);
-}
-
-#define rtl_writephy_batch(p, a) __rtl_writephy_batch(p, a, ARRAY_SIZE(a))
-
 static void rtl_release_firmware(struct rtl8169_private *tp)
 {
 	if (tp->rtl_fw) {
@@ -2252,587 +2197,6 @@ static void rtl8125_config_eee_mac(struct rtl8169_private *tp)
 	r8168_mac_ocp_modify(tp, 0xeb62, 0, BIT(2) | BIT(1));
 }
 
-static void rtl8168f_config_eee_phy(struct phy_device *phydev)
-{
-	r8168d_modify_extpage(phydev, 0x0020, 0x15, 0, BIT(8));
-	r8168d_phy_param(phydev, 0x8b85, 0, BIT(13));
-}
-
-static void rtl8168g_config_eee_phy(struct phy_device *phydev)
-{
-	phy_modify_paged(phydev, 0x0a43, 0x11, 0, BIT(4));
-}
-
-static void rtl8168h_config_eee_phy(struct phy_device *phydev)
-{
-	rtl8168g_config_eee_phy(phydev);
-
-	phy_modify_paged(phydev, 0xa4a, 0x11, 0x0000, 0x0200);
-	phy_modify_paged(phydev, 0xa42, 0x14, 0x0000, 0x0080);
-}
-
-static void rtl8125_config_eee_phy(struct phy_device *phydev)
-{
-	rtl8168h_config_eee_phy(phydev);
-
-	phy_modify_paged(phydev, 0xa6d, 0x12, 0x0001, 0x0000);
-	phy_modify_paged(phydev, 0xa6d, 0x14, 0x0010, 0x0000);
-}
-
-static void rtl8169s_hw_phy_config(struct rtl8169_private *tp,
-				   struct phy_device *phydev)
-{
-	static const struct phy_reg phy_reg_init[] = {
-		{ 0x1f, 0x0001 },
-		{ 0x06, 0x006e },
-		{ 0x08, 0x0708 },
-		{ 0x15, 0x4000 },
-		{ 0x18, 0x65c7 },
-
-		{ 0x1f, 0x0001 },
-		{ 0x03, 0x00a1 },
-		{ 0x02, 0x0008 },
-		{ 0x01, 0x0120 },
-		{ 0x00, 0x1000 },
-		{ 0x04, 0x0800 },
-		{ 0x04, 0x0000 },
-
-		{ 0x03, 0xff41 },
-		{ 0x02, 0xdf60 },
-		{ 0x01, 0x0140 },
-		{ 0x00, 0x0077 },
-		{ 0x04, 0x7800 },
-		{ 0x04, 0x7000 },
-
-		{ 0x03, 0x802f },
-		{ 0x02, 0x4f02 },
-		{ 0x01, 0x0409 },
-		{ 0x00, 0xf0f9 },
-		{ 0x04, 0x9800 },
-		{ 0x04, 0x9000 },
-
-		{ 0x03, 0xdf01 },
-		{ 0x02, 0xdf20 },
-		{ 0x01, 0xff95 },
-		{ 0x00, 0xba00 },
-		{ 0x04, 0xa800 },
-		{ 0x04, 0xa000 },
-
-		{ 0x03, 0xff41 },
-		{ 0x02, 0xdf20 },
-		{ 0x01, 0x0140 },
-		{ 0x00, 0x00bb },
-		{ 0x04, 0xb800 },
-		{ 0x04, 0xb000 },
-
-		{ 0x03, 0xdf41 },
-		{ 0x02, 0xdc60 },
-		{ 0x01, 0x6340 },
-		{ 0x00, 0x007d },
-		{ 0x04, 0xd800 },
-		{ 0x04, 0xd000 },
-
-		{ 0x03, 0xdf01 },
-		{ 0x02, 0xdf20 },
-		{ 0x01, 0x100a },
-		{ 0x00, 0xa0ff },
-		{ 0x04, 0xf800 },
-		{ 0x04, 0xf000 },
-
-		{ 0x1f, 0x0000 },
-		{ 0x0b, 0x0000 },
-		{ 0x00, 0x9200 }
-	};
-
-	rtl_writephy_batch(phydev, phy_reg_init);
-}
-
-static void rtl8169sb_hw_phy_config(struct rtl8169_private *tp,
-				    struct phy_device *phydev)
-{
-	phy_write_paged(phydev, 0x0002, 0x01, 0x90d0);
-}
-
-static void rtl8169scd_hw_phy_config(struct rtl8169_private *tp,
-				     struct phy_device *phydev)
-{
-	static const struct phy_reg phy_reg_init[] = {
-		{ 0x1f, 0x0001 },
-		{ 0x04, 0x0000 },
-		{ 0x03, 0x00a1 },
-		{ 0x02, 0x0008 },
-		{ 0x01, 0x0120 },
-		{ 0x00, 0x1000 },
-		{ 0x04, 0x0800 },
-		{ 0x04, 0x9000 },
-		{ 0x03, 0x802f },
-		{ 0x02, 0x4f02 },
-		{ 0x01, 0x0409 },
-		{ 0x00, 0xf099 },
-		{ 0x04, 0x9800 },
-		{ 0x04, 0xa000 },
-		{ 0x03, 0xdf01 },
-		{ 0x02, 0xdf20 },
-		{ 0x01, 0xff95 },
-		{ 0x00, 0xba00 },
-		{ 0x04, 0xa800 },
-		{ 0x04, 0xf000 },
-		{ 0x03, 0xdf01 },
-		{ 0x02, 0xdf20 },
-		{ 0x01, 0x101a },
-		{ 0x00, 0xa0ff },
-		{ 0x04, 0xf800 },
-		{ 0x04, 0x0000 },
-		{ 0x1f, 0x0000 },
-
-		{ 0x1f, 0x0001 },
-		{ 0x10, 0xf41b },
-		{ 0x14, 0xfb54 },
-		{ 0x18, 0xf5c7 },
-		{ 0x1f, 0x0000 },
-
-		{ 0x1f, 0x0001 },
-		{ 0x17, 0x0cc0 },
-		{ 0x1f, 0x0000 }
-	};
-
-	rtl_writephy_batch(phydev, phy_reg_init);
-}
-
-static void rtl8169sce_hw_phy_config(struct rtl8169_private *tp,
-				     struct phy_device *phydev)
-{
-	static const struct phy_reg phy_reg_init[] = {
-		{ 0x1f, 0x0001 },
-		{ 0x04, 0x0000 },
-		{ 0x03, 0x00a1 },
-		{ 0x02, 0x0008 },
-		{ 0x01, 0x0120 },
-		{ 0x00, 0x1000 },
-		{ 0x04, 0x0800 },
-		{ 0x04, 0x9000 },
-		{ 0x03, 0x802f },
-		{ 0x02, 0x4f02 },
-		{ 0x01, 0x0409 },
-		{ 0x00, 0xf099 },
-		{ 0x04, 0x9800 },
-		{ 0x04, 0xa000 },
-		{ 0x03, 0xdf01 },
-		{ 0x02, 0xdf20 },
-		{ 0x01, 0xff95 },
-		{ 0x00, 0xba00 },
-		{ 0x04, 0xa800 },
-		{ 0x04, 0xf000 },
-		{ 0x03, 0xdf01 },
-		{ 0x02, 0xdf20 },
-		{ 0x01, 0x101a },
-		{ 0x00, 0xa0ff },
-		{ 0x04, 0xf800 },
-		{ 0x04, 0x0000 },
-		{ 0x1f, 0x0000 },
-
-		{ 0x1f, 0x0001 },
-		{ 0x0b, 0x8480 },
-		{ 0x1f, 0x0000 },
-
-		{ 0x1f, 0x0001 },
-		{ 0x18, 0x67c7 },
-		{ 0x04, 0x2000 },
-		{ 0x03, 0x002f },
-		{ 0x02, 0x4360 },
-		{ 0x01, 0x0109 },
-		{ 0x00, 0x3022 },
-		{ 0x04, 0x2800 },
-		{ 0x1f, 0x0000 },
-
-		{ 0x1f, 0x0001 },
-		{ 0x17, 0x0cc0 },
-		{ 0x1f, 0x0000 }
-	};
-
-	rtl_writephy_batch(phydev, phy_reg_init);
-}
-
-static void rtl8168bb_hw_phy_config(struct rtl8169_private *tp,
-				    struct phy_device *phydev)
-{
-	phy_write(phydev, 0x1f, 0x0001);
-	phy_set_bits(phydev, 0x16, BIT(0));
-	phy_write(phydev, 0x10, 0xf41b);
-	phy_write(phydev, 0x1f, 0x0000);
-}
-
-static void rtl8168bef_hw_phy_config(struct rtl8169_private *tp,
-				     struct phy_device *phydev)
-{
-	phy_write_paged(phydev, 0x0001, 0x10, 0xf41b);
-}
-
-static void rtl8168cp_1_hw_phy_config(struct rtl8169_private *tp,
-				      struct phy_device *phydev)
-{
-	phy_write(phydev, 0x1d, 0x0f00);
-	phy_write_paged(phydev, 0x0002, 0x0c, 0x1ec8);
-}
-
-static void rtl8168cp_2_hw_phy_config(struct rtl8169_private *tp,
-				      struct phy_device *phydev)
-{
-	phy_set_bits(phydev, 0x14, BIT(5));
-	phy_set_bits(phydev, 0x0d, BIT(5));
-	phy_write_paged(phydev, 0x0001, 0x1d, 0x3d98);
-}
-
-static void rtl8168c_1_hw_phy_config(struct rtl8169_private *tp,
-				     struct phy_device *phydev)
-{
-	static const struct phy_reg phy_reg_init[] = {
-		{ 0x1f, 0x0001 },
-		{ 0x12, 0x2300 },
-		{ 0x1f, 0x0002 },
-		{ 0x00, 0x88d4 },
-		{ 0x01, 0x82b1 },
-		{ 0x03, 0x7002 },
-		{ 0x08, 0x9e30 },
-		{ 0x09, 0x01f0 },
-		{ 0x0a, 0x5500 },
-		{ 0x0c, 0x00c8 },
-		{ 0x1f, 0x0003 },
-		{ 0x12, 0xc096 },
-		{ 0x16, 0x000a },
-		{ 0x1f, 0x0000 },
-		{ 0x1f, 0x0000 },
-		{ 0x09, 0x2000 },
-		{ 0x09, 0x0000 }
-	};
-
-	rtl_writephy_batch(phydev, phy_reg_init);
-
-	phy_set_bits(phydev, 0x14, BIT(5));
-	phy_set_bits(phydev, 0x0d, BIT(5));
-}
-
-static void rtl8168c_2_hw_phy_config(struct rtl8169_private *tp,
-				     struct phy_device *phydev)
-{
-	static const struct phy_reg phy_reg_init[] = {
-		{ 0x1f, 0x0001 },
-		{ 0x12, 0x2300 },
-		{ 0x03, 0x802f },
-		{ 0x02, 0x4f02 },
-		{ 0x01, 0x0409 },
-		{ 0x00, 0xf099 },
-		{ 0x04, 0x9800 },
-		{ 0x04, 0x9000 },
-		{ 0x1d, 0x3d98 },
-		{ 0x1f, 0x0002 },
-		{ 0x0c, 0x7eb8 },
-		{ 0x06, 0x0761 },
-		{ 0x1f, 0x0003 },
-		{ 0x16, 0x0f0a },
-		{ 0x1f, 0x0000 }
-	};
-
-	rtl_writephy_batch(phydev, phy_reg_init);
-
-	phy_set_bits(phydev, 0x16, BIT(0));
-	phy_set_bits(phydev, 0x14, BIT(5));
-	phy_set_bits(phydev, 0x0d, BIT(5));
-}
-
-static void rtl8168c_3_hw_phy_config(struct rtl8169_private *tp,
-				     struct phy_device *phydev)
-{
-	static const struct phy_reg phy_reg_init[] = {
-		{ 0x1f, 0x0001 },
-		{ 0x12, 0x2300 },
-		{ 0x1d, 0x3d98 },
-		{ 0x1f, 0x0002 },
-		{ 0x0c, 0x7eb8 },
-		{ 0x06, 0x5461 },
-		{ 0x1f, 0x0003 },
-		{ 0x16, 0x0f0a },
-		{ 0x1f, 0x0000 }
-	};
-
-	rtl_writephy_batch(phydev, phy_reg_init);
-
-	phy_set_bits(phydev, 0x16, BIT(0));
-	phy_set_bits(phydev, 0x14, BIT(5));
-	phy_set_bits(phydev, 0x0d, BIT(5));
-}
-
-static const struct phy_reg rtl8168d_1_phy_reg_init_0[] = {
-	/* Channel Estimation */
-	{ 0x1f, 0x0001 },
-	{ 0x06, 0x4064 },
-	{ 0x07, 0x2863 },
-	{ 0x08, 0x059c },
-	{ 0x09, 0x26b4 },
-	{ 0x0a, 0x6a19 },
-	{ 0x0b, 0xdcc8 },
-	{ 0x10, 0xf06d },
-	{ 0x14, 0x7f68 },
-	{ 0x18, 0x7fd9 },
-	{ 0x1c, 0xf0ff },
-	{ 0x1d, 0x3d9c },
-	{ 0x1f, 0x0003 },
-	{ 0x12, 0xf49f },
-	{ 0x13, 0x070b },
-	{ 0x1a, 0x05ad },
-	{ 0x14, 0x94c0 },
-
-	/*
-	 * Tx Error Issue
-	 * Enhance line driver power
-	 */
-	{ 0x1f, 0x0002 },
-	{ 0x06, 0x5561 },
-	{ 0x1f, 0x0005 },
-	{ 0x05, 0x8332 },
-	{ 0x06, 0x5561 },
-
-	/*
-	 * Can not link to 1Gbps with bad cable
-	 * Decrease SNR threshold form 21.07dB to 19.04dB
-	 */
-	{ 0x1f, 0x0001 },
-	{ 0x17, 0x0cc0 },
-
-	{ 0x1f, 0x0000 },
-	{ 0x0d, 0xf880 }
-};
-
-static const struct phy_reg rtl8168d_1_phy_reg_init_1[] = {
-	{ 0x1f, 0x0002 },
-	{ 0x05, 0x669a },
-	{ 0x1f, 0x0005 },
-	{ 0x05, 0x8330 },
-	{ 0x06, 0x669a },
-	{ 0x1f, 0x0002 }
-};
-
-static void rtl8168d_apply_firmware_cond(struct rtl8169_private *tp,
-					 struct phy_device *phydev,
-					 u16 val)
-{
-	u16 reg_val;
-
-	phy_write(phydev, 0x1f, 0x0005);
-	phy_write(phydev, 0x05, 0x001b);
-	reg_val = phy_read(phydev, 0x06);
-	phy_write(phydev, 0x1f, 0x0000);
-
-	if (reg_val != val)
-		phydev_warn(phydev, "chipset not ready for firmware\n");
-	else
-		r8169_apply_firmware(tp);
-}
-
-static void rtl8168d_1_hw_phy_config(struct rtl8169_private *tp,
-				     struct phy_device *phydev)
-{
-	rtl_writephy_batch(phydev, rtl8168d_1_phy_reg_init_0);
-
-	/*
-	 * Rx Error Issue
-	 * Fine Tune Switching regulator parameter
-	 */
-	phy_write(phydev, 0x1f, 0x0002);
-	phy_modify(phydev, 0x0b, 0x00ef, 0x0010);
-	phy_modify(phydev, 0x0c, 0x5d00, 0xa200);
-
-	if (rtl8168d_efuse_read(tp, 0x01) == 0xb1) {
-		int val;
-
-		rtl_writephy_batch(phydev, rtl8168d_1_phy_reg_init_1);
-
-		val = phy_read(phydev, 0x0d);
-
-		if ((val & 0x00ff) != 0x006c) {
-			static const u32 set[] = {
-				0x0065, 0x0066, 0x0067, 0x0068,
-				0x0069, 0x006a, 0x006b, 0x006c
-			};
-			int i;
-
-			phy_write(phydev, 0x1f, 0x0002);
-
-			val &= 0xff00;
-			for (i = 0; i < ARRAY_SIZE(set); i++)
-				phy_write(phydev, 0x0d, val | set[i]);
-		}
-	} else {
-		phy_write_paged(phydev, 0x0002, 0x05, 0x6662);
-		r8168d_phy_param(phydev, 0x8330, 0xffff, 0x6662);
-	}
-
-	/* RSET couple improve */
-	phy_write(phydev, 0x1f, 0x0002);
-	phy_set_bits(phydev, 0x0d, 0x0300);
-	phy_set_bits(phydev, 0x0f, 0x0010);
-
-	/* Fine tune PLL performance */
-	phy_write(phydev, 0x1f, 0x0002);
-	phy_modify(phydev, 0x02, 0x0600, 0x0100);
-	phy_clear_bits(phydev, 0x03, 0xe000);
-	phy_write(phydev, 0x1f, 0x0000);
-
-	rtl8168d_apply_firmware_cond(tp, phydev, 0xbf00);
-}
-
-static void rtl8168d_2_hw_phy_config(struct rtl8169_private *tp,
-				     struct phy_device *phydev)
-{
-	rtl_writephy_batch(phydev, rtl8168d_1_phy_reg_init_0);
-
-	if (rtl8168d_efuse_read(tp, 0x01) == 0xb1) {
-		int val;
-
-		rtl_writephy_batch(phydev, rtl8168d_1_phy_reg_init_1);
-
-		val = phy_read(phydev, 0x0d);
-		if ((val & 0x00ff) != 0x006c) {
-			static const u32 set[] = {
-				0x0065, 0x0066, 0x0067, 0x0068,
-				0x0069, 0x006a, 0x006b, 0x006c
-			};
-			int i;
-
-			phy_write(phydev, 0x1f, 0x0002);
-
-			val &= 0xff00;
-			for (i = 0; i < ARRAY_SIZE(set); i++)
-				phy_write(phydev, 0x0d, val | set[i]);
-		}
-	} else {
-		phy_write_paged(phydev, 0x0002, 0x05, 0x2642);
-		r8168d_phy_param(phydev, 0x8330, 0xffff, 0x2642);
-	}
-
-	/* Fine tune PLL performance */
-	phy_write(phydev, 0x1f, 0x0002);
-	phy_modify(phydev, 0x02, 0x0600, 0x0100);
-	phy_clear_bits(phydev, 0x03, 0xe000);
-	phy_write(phydev, 0x1f, 0x0000);
-
-	/* Switching regulator Slew rate */
-	phy_modify_paged(phydev, 0x0002, 0x0f, 0x0000, 0x0017);
-
-	rtl8168d_apply_firmware_cond(tp, phydev, 0xb300);
-}
-
-static void rtl8168d_3_hw_phy_config(struct rtl8169_private *tp,
-				     struct phy_device *phydev)
-{
-	static const struct phy_reg phy_reg_init[] = {
-		{ 0x1f, 0x0002 },
-		{ 0x10, 0x0008 },
-		{ 0x0d, 0x006c },
-
-		{ 0x1f, 0x0000 },
-		{ 0x0d, 0xf880 },
-
-		{ 0x1f, 0x0001 },
-		{ 0x17, 0x0cc0 },
-
-		{ 0x1f, 0x0001 },
-		{ 0x0b, 0xa4d8 },
-		{ 0x09, 0x281c },
-		{ 0x07, 0x2883 },
-		{ 0x0a, 0x6b35 },
-		{ 0x1d, 0x3da4 },
-		{ 0x1c, 0xeffd },
-		{ 0x14, 0x7f52 },
-		{ 0x18, 0x7fc6 },
-		{ 0x08, 0x0601 },
-		{ 0x06, 0x4063 },
-		{ 0x10, 0xf074 },
-		{ 0x1f, 0x0003 },
-		{ 0x13, 0x0789 },
-		{ 0x12, 0xf4bd },
-		{ 0x1a, 0x04fd },
-		{ 0x14, 0x84b0 },
-		{ 0x1f, 0x0000 },
-		{ 0x00, 0x9200 },
-
-		{ 0x1f, 0x0005 },
-		{ 0x01, 0x0340 },
-		{ 0x1f, 0x0001 },
-		{ 0x04, 0x4000 },
-		{ 0x03, 0x1d21 },
-		{ 0x02, 0x0c32 },
-		{ 0x01, 0x0200 },
-		{ 0x00, 0x5554 },
-		{ 0x04, 0x4800 },
-		{ 0x04, 0x4000 },
-		{ 0x04, 0xf000 },
-		{ 0x03, 0xdf01 },
-		{ 0x02, 0xdf20 },
-		{ 0x01, 0x101a },
-		{ 0x00, 0xa0ff },
-		{ 0x04, 0xf800 },
-		{ 0x04, 0xf000 },
-		{ 0x1f, 0x0000 },
-	};
-
-	rtl_writephy_batch(phydev, phy_reg_init);
-	r8168d_modify_extpage(phydev, 0x0023, 0x16, 0xffff, 0x0000);
-}
-
-static void rtl8168d_4_hw_phy_config(struct rtl8169_private *tp,
-				     struct phy_device *phydev)
-{
-	phy_write_paged(phydev, 0x0001, 0x17, 0x0cc0);
-	r8168d_modify_extpage(phydev, 0x002d, 0x18, 0xffff, 0x0040);
-	phy_set_bits(phydev, 0x0d, BIT(5));
-}
-
-static void rtl8168e_1_hw_phy_config(struct rtl8169_private *tp,
-				     struct phy_device *phydev)
-{
-	static const struct phy_reg phy_reg_init[] = {
-		/* Channel estimation fine tune */
-		{ 0x1f, 0x0001 },
-		{ 0x0b, 0x6c20 },
-		{ 0x07, 0x2872 },
-		{ 0x1c, 0xefff },
-		{ 0x1f, 0x0003 },
-		{ 0x14, 0x6420 },
-		{ 0x1f, 0x0000 },
-	};
-
-	r8169_apply_firmware(tp);
-
-	/* Enable Delay cap */
-	r8168d_phy_param(phydev, 0x8b80, 0xffff, 0xc896);
-
-	rtl_writephy_batch(phydev, phy_reg_init);
-
-	/* Update PFM & 10M TX idle timer */
-	r8168d_modify_extpage(phydev, 0x002f, 0x15, 0xffff, 0x1919);
-
-	r8168d_modify_extpage(phydev, 0x00ac, 0x18, 0xffff, 0x0006);
-
-	/* DCO enable for 10M IDLE Power */
-	r8168d_modify_extpage(phydev, 0x0023, 0x17, 0x0000, 0x0006);
-
-	/* For impedance matching */
-	phy_modify_paged(phydev, 0x0002, 0x08, 0x7f00, 0x8000);
-
-	/* PHY auto speed down */
-	r8168d_modify_extpage(phydev, 0x002d, 0x18, 0x0000, 0x0050);
-	phy_set_bits(phydev, 0x14, BIT(15));
-
-	r8168d_phy_param(phydev, 0x8b86, 0x0000, 0x0001);
-	r8168d_phy_param(phydev, 0x8b85, 0x2000, 0x0000);
-
-	r8168d_modify_extpage(phydev, 0x0020, 0x15, 0x1100, 0x0000);
-	phy_write_paged(phydev, 0x0006, 0x00, 0x5a00);
-
-	phy_write_mmd(phydev, MDIO_MMD_AN, MDIO_AN_EEE_ADV, 0x0000);
-}
-
 static void rtl_rar_exgmac_set(struct rtl8169_private *tp, u8 *addr)
 {
 	const u16 w[] = {
@@ -2847,280 +2211,6 @@ static void rtl_rar_exgmac_set(struct rtl8169_private *tp, u8 *addr)
 	rtl_eri_write(tp, 0xf4, ERIAR_MASK_1111, w[1] | (w[2] << 16));
 }
 
-static void rtl8168e_2_hw_phy_config(struct rtl8169_private *tp,
-				     struct phy_device *phydev)
-{
-	r8169_apply_firmware(tp);
-
-	/* Enable Delay cap */
-	r8168d_modify_extpage(phydev, 0x00ac, 0x18, 0xffff, 0x0006);
-
-	/* Channel estimation fine tune */
-	phy_write_paged(phydev, 0x0003, 0x09, 0xa20f);
-
-	/* Green Setting */
-	r8168d_phy_param(phydev, 0x8b5b, 0xffff, 0x9222);
-	r8168d_phy_param(phydev, 0x8b6d, 0xffff, 0x8000);
-	r8168d_phy_param(phydev, 0x8b76, 0xffff, 0x8000);
-
-	/* For 4-corner performance improve */
-	phy_write(phydev, 0x1f, 0x0005);
-	phy_write(phydev, 0x05, 0x8b80);
-	phy_set_bits(phydev, 0x17, 0x0006);
-	phy_write(phydev, 0x1f, 0x0000);
-
-	/* PHY auto speed down */
-	r8168d_modify_extpage(phydev, 0x002d, 0x18, 0x0000, 0x0010);
-	phy_set_bits(phydev, 0x14, BIT(15));
-
-	/* improve 10M EEE waveform */
-	r8168d_phy_param(phydev, 0x8b86, 0x0000, 0x0001);
-
-	/* Improve 2-pair detection performance */
-	r8168d_phy_param(phydev, 0x8b85, 0x0000, 0x4000);
-
-	rtl8168f_config_eee_phy(phydev);
-
-	/* Green feature */
-	phy_write(phydev, 0x1f, 0x0003);
-	phy_set_bits(phydev, 0x19, BIT(0));
-	phy_set_bits(phydev, 0x10, BIT(10));
-	phy_write(phydev, 0x1f, 0x0000);
-	phy_modify_paged(phydev, 0x0005, 0x01, 0, BIT(8));
-}
-
-static void rtl8168f_hw_phy_config(struct rtl8169_private *tp,
-				   struct phy_device *phydev)
-{
-	/* For 4-corner performance improve */
-	r8168d_phy_param(phydev, 0x8b80, 0x0000, 0x0006);
-
-	/* PHY auto speed down */
-	r8168d_modify_extpage(phydev, 0x002d, 0x18, 0x0000, 0x0010);
-	phy_set_bits(phydev, 0x14, BIT(15));
-
-	/* Improve 10M EEE waveform */
-	r8168d_phy_param(phydev, 0x8b86, 0x0000, 0x0001);
-
-	rtl8168f_config_eee_phy(phydev);
-}
-
-static void rtl8168f_1_hw_phy_config(struct rtl8169_private *tp,
-				     struct phy_device *phydev)
-{
-	r8169_apply_firmware(tp);
-
-	/* Channel estimation fine tune */
-	phy_write_paged(phydev, 0x0003, 0x09, 0xa20f);
-
-	/* Modify green table for giga & fnet */
-	r8168d_phy_param(phydev, 0x8b55, 0xffff, 0x0000);
-	r8168d_phy_param(phydev, 0x8b5e, 0xffff, 0x0000);
-	r8168d_phy_param(phydev, 0x8b67, 0xffff, 0x0000);
-	r8168d_phy_param(phydev, 0x8b70, 0xffff, 0x0000);
-	r8168d_modify_extpage(phydev, 0x0078, 0x17, 0xffff, 0x0000);
-	r8168d_modify_extpage(phydev, 0x0078, 0x19, 0xffff, 0x00fb);
-
-	/* Modify green table for 10M */
-	r8168d_phy_param(phydev, 0x8b79, 0xffff, 0xaa00);
-
-	/* Disable hiimpedance detection (RTCT) */
-	phy_write_paged(phydev, 0x0003, 0x01, 0x328a);
-
-	rtl8168f_hw_phy_config(tp, phydev);
-
-	/* Improve 2-pair detection performance */
-	r8168d_phy_param(phydev, 0x8b85, 0x0000, 0x4000);
-}
-
-static void rtl8168f_2_hw_phy_config(struct rtl8169_private *tp,
-				     struct phy_device *phydev)
-{
-	r8169_apply_firmware(tp);
-
-	rtl8168f_hw_phy_config(tp, phydev);
-}
-
-static void rtl8411_hw_phy_config(struct rtl8169_private *tp,
-				  struct phy_device *phydev)
-{
-	r8169_apply_firmware(tp);
-
-	rtl8168f_hw_phy_config(tp, phydev);
-
-	/* Improve 2-pair detection performance */
-	r8168d_phy_param(phydev, 0x8b85, 0x0000, 0x4000);
-
-	/* Channel estimation fine tune */
-	phy_write_paged(phydev, 0x0003, 0x09, 0xa20f);
-
-	/* Modify green table for giga & fnet */
-	r8168d_phy_param(phydev, 0x8b55, 0xffff, 0x0000);
-	r8168d_phy_param(phydev, 0x8b5e, 0xffff, 0x0000);
-	r8168d_phy_param(phydev, 0x8b67, 0xffff, 0x0000);
-	r8168d_phy_param(phydev, 0x8b70, 0xffff, 0x0000);
-	r8168d_modify_extpage(phydev, 0x0078, 0x17, 0xffff, 0x0000);
-	r8168d_modify_extpage(phydev, 0x0078, 0x19, 0xffff, 0x00aa);
-
-	/* Modify green table for 10M */
-	r8168d_phy_param(phydev, 0x8b79, 0xffff, 0xaa00);
-
-	/* Disable hiimpedance detection (RTCT) */
-	phy_write_paged(phydev, 0x0003, 0x01, 0x328a);
-
-	/* Modify green table for giga */
-	r8168d_phy_param(phydev, 0x8b54, 0x0800, 0x0000);
-	r8168d_phy_param(phydev, 0x8b5d, 0x0800, 0x0000);
-	r8168d_phy_param(phydev, 0x8a7c, 0x0100, 0x0000);
-	r8168d_phy_param(phydev, 0x8a7f, 0x0000, 0x0100);
-	r8168d_phy_param(phydev, 0x8a82, 0x0100, 0x0000);
-	r8168d_phy_param(phydev, 0x8a85, 0x0100, 0x0000);
-	r8168d_phy_param(phydev, 0x8a88, 0x0100, 0x0000);
-
-	/* uc same-seed solution */
-	r8168d_phy_param(phydev, 0x8b85, 0x0000, 0x8000);
-
-	/* Green feature */
-	phy_write(phydev, 0x1f, 0x0003);
-	phy_clear_bits(phydev, 0x19, BIT(0));
-	phy_clear_bits(phydev, 0x10, BIT(10));
-	phy_write(phydev, 0x1f, 0x0000);
-}
-
-static void rtl8168g_disable_aldps(struct phy_device *phydev)
-{
-	phy_modify_paged(phydev, 0x0a43, 0x10, BIT(2), 0);
-}
-
-static void rtl8168g_phy_adjust_10m_aldps(struct phy_device *phydev)
-{
-	phy_modify_paged(phydev, 0x0bcc, 0x14, BIT(8), 0);
-	phy_modify_paged(phydev, 0x0a44, 0x11, 0, BIT(7) | BIT(6));
-	r8168g_phy_param(phydev, 0x8084, 0x6000, 0x0000);
-	phy_modify_paged(phydev, 0x0a43, 0x10, 0x0000, 0x1003);
-}
-
-static void rtl8168g_1_hw_phy_config(struct rtl8169_private *tp,
-				     struct phy_device *phydev)
-{
-	int ret;
-
-	r8169_apply_firmware(tp);
-
-	ret = phy_read_paged(phydev, 0x0a46, 0x10);
-	if (ret & BIT(8))
-		phy_modify_paged(phydev, 0x0bcc, 0x12, BIT(15), 0);
-	else
-		phy_modify_paged(phydev, 0x0bcc, 0x12, 0, BIT(15));
-
-	ret = phy_read_paged(phydev, 0x0a46, 0x13);
-	if (ret & BIT(8))
-		phy_modify_paged(phydev, 0x0c41, 0x15, 0, BIT(1));
-	else
-		phy_modify_paged(phydev, 0x0c41, 0x15, BIT(1), 0);
-
-	/* Enable PHY auto speed down */
-	phy_modify_paged(phydev, 0x0a44, 0x11, 0, BIT(3) | BIT(2));
-
-	rtl8168g_phy_adjust_10m_aldps(phydev);
-
-	/* EEE auto-fallback function */
-	phy_modify_paged(phydev, 0x0a4b, 0x11, 0, BIT(2));
-
-	/* Enable UC LPF tune function */
-	r8168g_phy_param(phydev, 0x8012, 0x0000, 0x8000);
-
-	phy_modify_paged(phydev, 0x0c42, 0x11, BIT(13), BIT(14));
-
-	/* Improve SWR Efficiency */
-	phy_write(phydev, 0x1f, 0x0bcd);
-	phy_write(phydev, 0x14, 0x5065);
-	phy_write(phydev, 0x14, 0xd065);
-	phy_write(phydev, 0x1f, 0x0bc8);
-	phy_write(phydev, 0x11, 0x5655);
-	phy_write(phydev, 0x1f, 0x0bcd);
-	phy_write(phydev, 0x14, 0x1065);
-	phy_write(phydev, 0x14, 0x9065);
-	phy_write(phydev, 0x14, 0x1065);
-	phy_write(phydev, 0x1f, 0x0000);
-
-	rtl8168g_disable_aldps(phydev);
-	rtl8168g_config_eee_phy(phydev);
-}
-
-static void rtl8168g_2_hw_phy_config(struct rtl8169_private *tp,
-				     struct phy_device *phydev)
-{
-	r8169_apply_firmware(tp);
-	rtl8168g_config_eee_phy(phydev);
-}
-
-static void rtl8168h_1_hw_phy_config(struct rtl8169_private *tp,
-				     struct phy_device *phydev)
-{
-	u16 dout_tapbin;
-	u32 data;
-
-	r8169_apply_firmware(tp);
-
-	/* CHN EST parameters adjust - giga master */
-	r8168g_phy_param(phydev, 0x809b, 0xf800, 0x8000);
-	r8168g_phy_param(phydev, 0x80a2, 0xff00, 0x8000);
-	r8168g_phy_param(phydev, 0x80a4, 0xff00, 0x8500);
-	r8168g_phy_param(phydev, 0x809c, 0xff00, 0xbd00);
-
-	/* CHN EST parameters adjust - giga slave */
-	r8168g_phy_param(phydev, 0x80ad, 0xf800, 0x7000);
-	r8168g_phy_param(phydev, 0x80b4, 0xff00, 0x5000);
-	r8168g_phy_param(phydev, 0x80ac, 0xff00, 0x4000);
-
-	/* CHN EST parameters adjust - fnet */
-	r8168g_phy_param(phydev, 0x808e, 0xff00, 0x1200);
-	r8168g_phy_param(phydev, 0x8090, 0xff00, 0xe500);
-	r8168g_phy_param(phydev, 0x8092, 0xff00, 0x9f00);
-
-	/* enable R-tune & PGA-retune function */
-	dout_tapbin = 0;
-	data = phy_read_paged(phydev, 0x0a46, 0x13);
-	data &= 3;
-	data <<= 2;
-	dout_tapbin |= data;
-	data = phy_read_paged(phydev, 0x0a46, 0x12);
-	data &= 0xc000;
-	data >>= 14;
-	dout_tapbin |= data;
-	dout_tapbin = ~(dout_tapbin^0x08);
-	dout_tapbin <<= 12;
-	dout_tapbin &= 0xf000;
-
-	r8168g_phy_param(phydev, 0x827a, 0xf000, dout_tapbin);
-	r8168g_phy_param(phydev, 0x827b, 0xf000, dout_tapbin);
-	r8168g_phy_param(phydev, 0x827c, 0xf000, dout_tapbin);
-	r8168g_phy_param(phydev, 0x827d, 0xf000, dout_tapbin);
-	r8168g_phy_param(phydev, 0x0811, 0x0000, 0x0800);
-	phy_modify_paged(phydev, 0x0a42, 0x16, 0x0000, 0x0002);
-
-	/* enable GPHY 10M */
-	phy_modify_paged(phydev, 0x0a44, 0x11, 0, BIT(11));
-
-	/* SAR ADC performance */
-	phy_modify_paged(phydev, 0x0bca, 0x17, BIT(12) | BIT(13), BIT(14));
-
-	r8168g_phy_param(phydev, 0x803f, 0x3000, 0x0000);
-	r8168g_phy_param(phydev, 0x8047, 0x3000, 0x0000);
-	r8168g_phy_param(phydev, 0x804f, 0x3000, 0x0000);
-	r8168g_phy_param(phydev, 0x8057, 0x3000, 0x0000);
-	r8168g_phy_param(phydev, 0x805f, 0x3000, 0x0000);
-	r8168g_phy_param(phydev, 0x8067, 0x3000, 0x0000);
-	r8168g_phy_param(phydev, 0x806f, 0x3000, 0x0000);
-
-	/* disable phy pfm mode */
-	phy_modify_paged(phydev, 0x0a44, 0x11, BIT(7), 0);
-
-	rtl8168g_disable_aldps(phydev);
-	rtl8168h_config_eee_phy(phydev);
-}
-
 u16 rtl8168h_2_get_adc_bias_ioffset(struct rtl8169_private *tp)
 {
 	u16 data1, data2, ioffset;
@@ -3137,387 +2227,6 @@ u16 rtl8168h_2_get_adc_bias_ioffset(struct rtl8169_private *tp)
 	return ioffset;
 }
 
-static void rtl8168h_2_hw_phy_config(struct rtl8169_private *tp,
-				     struct phy_device *phydev)
-{
-	u16 ioffset, rlen;
-	u32 data;
-
-	r8169_apply_firmware(tp);
-
-	/* CHIN EST parameter update */
-	r8168g_phy_param(phydev, 0x808a, 0x003f, 0x000a);
-
-	/* enable R-tune & PGA-retune function */
-	r8168g_phy_param(phydev, 0x0811, 0x0000, 0x0800);
-	phy_modify_paged(phydev, 0x0a42, 0x16, 0x0000, 0x0002);
-
-	/* enable GPHY 10M */
-	phy_modify_paged(phydev, 0x0a44, 0x11, 0, BIT(11));
-
-	ioffset = rtl8168h_2_get_adc_bias_ioffset(tp);
-	if (ioffset != 0xffff)
-		phy_write_paged(phydev, 0x0bcf, 0x16, ioffset);
-
-	/* Modify rlen (TX LPF corner frequency) level */
-	data = phy_read_paged(phydev, 0x0bcd, 0x16);
-	data &= 0x000f;
-	rlen = 0;
-	if (data > 3)
-		rlen = data - 3;
-	data = rlen | (rlen<<4) | (rlen<<8) | (rlen<<12);
-	phy_write_paged(phydev, 0x0bcd, 0x17, data);
-
-	/* disable phy pfm mode */
-	phy_modify_paged(phydev, 0x0a44, 0x11, BIT(7), 0);
-
-	rtl8168g_disable_aldps(phydev);
-	rtl8168g_config_eee_phy(phydev);
-}
-
-static void rtl8168ep_1_hw_phy_config(struct rtl8169_private *tp,
-				      struct phy_device *phydev)
-{
-	/* Enable PHY auto speed down */
-	phy_modify_paged(phydev, 0x0a44, 0x11, 0, BIT(3) | BIT(2));
-
-	rtl8168g_phy_adjust_10m_aldps(phydev);
-
-	/* Enable EEE auto-fallback function */
-	phy_modify_paged(phydev, 0x0a4b, 0x11, 0, BIT(2));
-
-	/* Enable UC LPF tune function */
-	r8168g_phy_param(phydev, 0x8012, 0x0000, 0x8000);
-
-	/* set rg_sel_sdm_rate */
-	phy_modify_paged(phydev, 0x0c42, 0x11, BIT(13), BIT(14));
-
-	rtl8168g_disable_aldps(phydev);
-	rtl8168g_config_eee_phy(phydev);
-}
-
-static void rtl8168ep_2_hw_phy_config(struct rtl8169_private *tp,
-				      struct phy_device *phydev)
-{
-	rtl8168g_phy_adjust_10m_aldps(phydev);
-
-	/* Enable UC LPF tune function */
-	r8168g_phy_param(phydev, 0x8012, 0x0000, 0x8000);
-
-	/* Set rg_sel_sdm_rate */
-	phy_modify_paged(phydev, 0x0c42, 0x11, BIT(13), BIT(14));
-
-	/* Channel estimation parameters */
-	r8168g_phy_param(phydev, 0x80f3, 0xff00, 0x8b00);
-	r8168g_phy_param(phydev, 0x80f0, 0xff00, 0x3a00);
-	r8168g_phy_param(phydev, 0x80ef, 0xff00, 0x0500);
-	r8168g_phy_param(phydev, 0x80f6, 0xff00, 0x6e00);
-	r8168g_phy_param(phydev, 0x80ec, 0xff00, 0x6800);
-	r8168g_phy_param(phydev, 0x80ed, 0xff00, 0x7c00);
-	r8168g_phy_param(phydev, 0x80f2, 0xff00, 0xf400);
-	r8168g_phy_param(phydev, 0x80f4, 0xff00, 0x8500);
-	r8168g_phy_param(phydev, 0x8110, 0xff00, 0xa800);
-	r8168g_phy_param(phydev, 0x810f, 0xff00, 0x1d00);
-	r8168g_phy_param(phydev, 0x8111, 0xff00, 0xf500);
-	r8168g_phy_param(phydev, 0x8113, 0xff00, 0x6100);
-	r8168g_phy_param(phydev, 0x8115, 0xff00, 0x9200);
-	r8168g_phy_param(phydev, 0x810e, 0xff00, 0x0400);
-	r8168g_phy_param(phydev, 0x810c, 0xff00, 0x7c00);
-	r8168g_phy_param(phydev, 0x810b, 0xff00, 0x5a00);
-	r8168g_phy_param(phydev, 0x80d1, 0xff00, 0xff00);
-	r8168g_phy_param(phydev, 0x80cd, 0xff00, 0x9e00);
-	r8168g_phy_param(phydev, 0x80d3, 0xff00, 0x0e00);
-	r8168g_phy_param(phydev, 0x80d5, 0xff00, 0xca00);
-	r8168g_phy_param(phydev, 0x80d7, 0xff00, 0x8400);
-
-	/* Force PWM-mode */
-	phy_write(phydev, 0x1f, 0x0bcd);
-	phy_write(phydev, 0x14, 0x5065);
-	phy_write(phydev, 0x14, 0xd065);
-	phy_write(phydev, 0x1f, 0x0bc8);
-	phy_write(phydev, 0x12, 0x00ed);
-	phy_write(phydev, 0x1f, 0x0bcd);
-	phy_write(phydev, 0x14, 0x1065);
-	phy_write(phydev, 0x14, 0x9065);
-	phy_write(phydev, 0x14, 0x1065);
-	phy_write(phydev, 0x1f, 0x0000);
-
-	rtl8168g_disable_aldps(phydev);
-	rtl8168g_config_eee_phy(phydev);
-}
-
-static void rtl8117_hw_phy_config(struct rtl8169_private *tp,
-				  struct phy_device *phydev)
-{
-	/* CHN EST parameters adjust - fnet */
-	r8168g_phy_param(phydev, 0x808e, 0xff00, 0x4800);
-	r8168g_phy_param(phydev, 0x8090, 0xff00, 0xcc00);
-	r8168g_phy_param(phydev, 0x8092, 0xff00, 0xb000);
-
-	r8168g_phy_param(phydev, 0x8088, 0xff00, 0x6000);
-	r8168g_phy_param(phydev, 0x808b, 0x3f00, 0x0b00);
-	r8168g_phy_param(phydev, 0x808d, 0x1f00, 0x0600);
-	r8168g_phy_param(phydev, 0x808c, 0xff00, 0xb000);
-	r8168g_phy_param(phydev, 0x80a0, 0xff00, 0x2800);
-	r8168g_phy_param(phydev, 0x80a2, 0xff00, 0x5000);
-	r8168g_phy_param(phydev, 0x809b, 0xf800, 0xb000);
-	r8168g_phy_param(phydev, 0x809a, 0xff00, 0x4b00);
-	r8168g_phy_param(phydev, 0x809d, 0x3f00, 0x0800);
-	r8168g_phy_param(phydev, 0x80a1, 0xff00, 0x7000);
-	r8168g_phy_param(phydev, 0x809f, 0x1f00, 0x0300);
-	r8168g_phy_param(phydev, 0x809e, 0xff00, 0x8800);
-	r8168g_phy_param(phydev, 0x80b2, 0xff00, 0x2200);
-	r8168g_phy_param(phydev, 0x80ad, 0xf800, 0x9800);
-	r8168g_phy_param(phydev, 0x80af, 0x3f00, 0x0800);
-	r8168g_phy_param(phydev, 0x80b3, 0xff00, 0x6f00);
-	r8168g_phy_param(phydev, 0x80b1, 0x1f00, 0x0300);
-	r8168g_phy_param(phydev, 0x80b0, 0xff00, 0x9300);
-
-	r8168g_phy_param(phydev, 0x8011, 0x0000, 0x0800);
-
-	/* enable GPHY 10M */
-	phy_modify_paged(phydev, 0x0a44, 0x11, 0, BIT(11));
-
-	r8168g_phy_param(phydev, 0x8016, 0x0000, 0x0400);
-
-	rtl8168g_disable_aldps(phydev);
-	rtl8168h_config_eee_phy(phydev);
-}
-
-static void rtl8102e_hw_phy_config(struct rtl8169_private *tp,
-				   struct phy_device *phydev)
-{
-	static const struct phy_reg phy_reg_init[] = {
-		{ 0x1f, 0x0003 },
-		{ 0x08, 0x441d },
-		{ 0x01, 0x9100 },
-		{ 0x1f, 0x0000 }
-	};
-
-	phy_set_bits(phydev, 0x11, BIT(12));
-	phy_set_bits(phydev, 0x19, BIT(13));
-	phy_set_bits(phydev, 0x10, BIT(15));
-
-	rtl_writephy_batch(phydev, phy_reg_init);
-}
-
-static void rtl8105e_hw_phy_config(struct rtl8169_private *tp,
-				   struct phy_device *phydev)
-{
-	/* Disable ALDPS before ram code */
-	phy_write(phydev, 0x18, 0x0310);
-	msleep(100);
-
-	r8169_apply_firmware(tp);
-
-	phy_write_paged(phydev, 0x0005, 0x1a, 0x0000);
-	phy_write_paged(phydev, 0x0004, 0x1c, 0x0000);
-	phy_write_paged(phydev, 0x0001, 0x15, 0x7701);
-}
-
-static void rtl8402_hw_phy_config(struct rtl8169_private *tp,
-				  struct phy_device *phydev)
-{
-	/* Disable ALDPS before setting firmware */
-	phy_write(phydev, 0x18, 0x0310);
-	msleep(20);
-
-	r8169_apply_firmware(tp);
-
-	/* EEE setting */
-	phy_write(phydev, 0x1f, 0x0004);
-	phy_write(phydev, 0x10, 0x401f);
-	phy_write(phydev, 0x19, 0x7030);
-	phy_write(phydev, 0x1f, 0x0000);
-}
-
-static void rtl8106e_hw_phy_config(struct rtl8169_private *tp,
-				   struct phy_device *phydev)
-{
-	static const struct phy_reg phy_reg_init[] = {
-		{ 0x1f, 0x0004 },
-		{ 0x10, 0xc07f },
-		{ 0x19, 0x7030 },
-		{ 0x1f, 0x0000 }
-	};
-
-	/* Disable ALDPS before ram code */
-	phy_write(phydev, 0x18, 0x0310);
-	msleep(100);
-
-	r8169_apply_firmware(tp);
-
-	rtl_writephy_batch(phydev, phy_reg_init);
-}
-
-static void rtl8125_1_hw_phy_config(struct rtl8169_private *tp,
-				    struct phy_device *phydev)
-{
-	phy_modify_paged(phydev, 0xad4, 0x10, 0x03ff, 0x0084);
-	phy_modify_paged(phydev, 0xad4, 0x17, 0x0000, 0x0010);
-	phy_modify_paged(phydev, 0xad1, 0x13, 0x03ff, 0x0006);
-	phy_modify_paged(phydev, 0xad3, 0x11, 0x003f, 0x0006);
-	phy_modify_paged(phydev, 0xac0, 0x14, 0x0000, 0x1100);
-	phy_modify_paged(phydev, 0xac8, 0x15, 0xf000, 0x7000);
-	phy_modify_paged(phydev, 0xad1, 0x14, 0x0000, 0x0400);
-	phy_modify_paged(phydev, 0xad1, 0x15, 0x0000, 0x03ff);
-	phy_modify_paged(phydev, 0xad1, 0x16, 0x0000, 0x03ff);
-
-	r8168g_phy_param(phydev, 0x80ea, 0xff00, 0xc400);
-	r8168g_phy_param(phydev, 0x80eb, 0x0700, 0x0300);
-	r8168g_phy_param(phydev, 0x80f8, 0xff00, 0x1c00);
-	r8168g_phy_param(phydev, 0x80f1, 0xff00, 0x3000);
-	r8168g_phy_param(phydev, 0x80fe, 0xff00, 0xa500);
-	r8168g_phy_param(phydev, 0x8102, 0xff00, 0x5000);
-	r8168g_phy_param(phydev, 0x8105, 0xff00, 0x3300);
-	r8168g_phy_param(phydev, 0x8100, 0xff00, 0x7000);
-	r8168g_phy_param(phydev, 0x8104, 0xff00, 0xf000);
-	r8168g_phy_param(phydev, 0x8106, 0xff00, 0x6500);
-	r8168g_phy_param(phydev, 0x80dc, 0xff00, 0xed00);
-	r8168g_phy_param(phydev, 0x80df, 0x0000, 0x0100);
-	r8168g_phy_param(phydev, 0x80e1, 0x0100, 0x0000);
-
-	phy_modify_paged(phydev, 0xbf0, 0x13, 0x003f, 0x0038);
-	r8168g_phy_param(phydev, 0x819f, 0xffff, 0xd0b6);
-
-	phy_write_paged(phydev, 0xbc3, 0x12, 0x5555);
-	phy_modify_paged(phydev, 0xbf0, 0x15, 0x0e00, 0x0a00);
-	phy_modify_paged(phydev, 0xa5c, 0x10, 0x0400, 0x0000);
-	phy_modify_paged(phydev, 0xa44, 0x11, 0x0000, 0x0800);
-
-	rtl8125_config_eee_phy(phydev);
-}
-
-static void rtl8125_2_hw_phy_config(struct rtl8169_private *tp,
-				    struct phy_device *phydev)
-{
-	int i;
-
-	phy_modify_paged(phydev, 0xad4, 0x17, 0x0000, 0x0010);
-	phy_modify_paged(phydev, 0xad1, 0x13, 0x03ff, 0x03ff);
-	phy_modify_paged(phydev, 0xad3, 0x11, 0x003f, 0x0006);
-	phy_modify_paged(phydev, 0xac0, 0x14, 0x1100, 0x0000);
-	phy_modify_paged(phydev, 0xacc, 0x10, 0x0003, 0x0002);
-	phy_modify_paged(phydev, 0xad4, 0x10, 0x00e7, 0x0044);
-	phy_modify_paged(phydev, 0xac1, 0x12, 0x0080, 0x0000);
-	phy_modify_paged(phydev, 0xac8, 0x10, 0x0300, 0x0000);
-	phy_modify_paged(phydev, 0xac5, 0x17, 0x0007, 0x0002);
-	phy_write_paged(phydev, 0xad4, 0x16, 0x00a8);
-	phy_write_paged(phydev, 0xac5, 0x16, 0x01ff);
-	phy_modify_paged(phydev, 0xac8, 0x15, 0x00f0, 0x0030);
-
-	phy_write(phydev, 0x1f, 0x0b87);
-	phy_write(phydev, 0x16, 0x80a2);
-	phy_write(phydev, 0x17, 0x0153);
-	phy_write(phydev, 0x16, 0x809c);
-	phy_write(phydev, 0x17, 0x0153);
-	phy_write(phydev, 0x1f, 0x0000);
-
-	phy_write(phydev, 0x1f, 0x0a43);
-	phy_write(phydev, 0x13, 0x81B3);
-	phy_write(phydev, 0x14, 0x0043);
-	phy_write(phydev, 0x14, 0x00A7);
-	phy_write(phydev, 0x14, 0x00D6);
-	phy_write(phydev, 0x14, 0x00EC);
-	phy_write(phydev, 0x14, 0x00F6);
-	phy_write(phydev, 0x14, 0x00FB);
-	phy_write(phydev, 0x14, 0x00FD);
-	phy_write(phydev, 0x14, 0x00FF);
-	phy_write(phydev, 0x14, 0x00BB);
-	phy_write(phydev, 0x14, 0x0058);
-	phy_write(phydev, 0x14, 0x0029);
-	phy_write(phydev, 0x14, 0x0013);
-	phy_write(phydev, 0x14, 0x0009);
-	phy_write(phydev, 0x14, 0x0004);
-	phy_write(phydev, 0x14, 0x0002);
-	for (i = 0; i < 25; i++)
-		phy_write(phydev, 0x14, 0x0000);
-	phy_write(phydev, 0x1f, 0x0000);
-
-	r8168g_phy_param(phydev, 0x8257, 0xffff, 0x020F);
-	r8168g_phy_param(phydev, 0x80ea, 0xffff, 0x7843);
-
-	r8169_apply_firmware(tp);
-
-	phy_modify_paged(phydev, 0xd06, 0x14, 0x0000, 0x2000);
-
-	r8168g_phy_param(phydev, 0x81a2, 0x0000, 0x0100);
-
-	phy_modify_paged(phydev, 0xb54, 0x16, 0xff00, 0xdb00);
-	phy_modify_paged(phydev, 0xa45, 0x12, 0x0001, 0x0000);
-	phy_modify_paged(phydev, 0xa5d, 0x12, 0x0000, 0x0020);
-	phy_modify_paged(phydev, 0xad4, 0x17, 0x0010, 0x0000);
-	phy_modify_paged(phydev, 0xa86, 0x15, 0x0001, 0x0000);
-	phy_modify_paged(phydev, 0xa44, 0x11, 0x0000, 0x0800);
-
-	rtl8125_config_eee_phy(phydev);
-}
-
-void r8169_hw_phy_config(struct rtl8169_private *tp, struct phy_device *phydev,
-			 enum mac_version ver)
-{
-	static const rtl_phy_cfg_fct phy_configs[] = {
-		/* PCI devices. */
-		[RTL_GIGA_MAC_VER_02] = rtl8169s_hw_phy_config,
-		[RTL_GIGA_MAC_VER_03] = rtl8169s_hw_phy_config,
-		[RTL_GIGA_MAC_VER_04] = rtl8169sb_hw_phy_config,
-		[RTL_GIGA_MAC_VER_05] = rtl8169scd_hw_phy_config,
-		[RTL_GIGA_MAC_VER_06] = rtl8169sce_hw_phy_config,
-		/* PCI-E devices. */
-		[RTL_GIGA_MAC_VER_07] = rtl8102e_hw_phy_config,
-		[RTL_GIGA_MAC_VER_08] = rtl8102e_hw_phy_config,
-		[RTL_GIGA_MAC_VER_09] = rtl8102e_hw_phy_config,
-		[RTL_GIGA_MAC_VER_10] = NULL,
-		[RTL_GIGA_MAC_VER_11] = rtl8168bb_hw_phy_config,
-		[RTL_GIGA_MAC_VER_12] = rtl8168bef_hw_phy_config,
-		[RTL_GIGA_MAC_VER_13] = NULL,
-		[RTL_GIGA_MAC_VER_14] = NULL,
-		[RTL_GIGA_MAC_VER_15] = NULL,
-		[RTL_GIGA_MAC_VER_16] = NULL,
-		[RTL_GIGA_MAC_VER_17] = rtl8168bef_hw_phy_config,
-		[RTL_GIGA_MAC_VER_18] = rtl8168cp_1_hw_phy_config,
-		[RTL_GIGA_MAC_VER_19] = rtl8168c_1_hw_phy_config,
-		[RTL_GIGA_MAC_VER_20] = rtl8168c_2_hw_phy_config,
-		[RTL_GIGA_MAC_VER_21] = rtl8168c_3_hw_phy_config,
-		[RTL_GIGA_MAC_VER_22] = rtl8168c_3_hw_phy_config,
-		[RTL_GIGA_MAC_VER_23] = rtl8168cp_2_hw_phy_config,
-		[RTL_GIGA_MAC_VER_24] = rtl8168cp_2_hw_phy_config,
-		[RTL_GIGA_MAC_VER_25] = rtl8168d_1_hw_phy_config,
-		[RTL_GIGA_MAC_VER_26] = rtl8168d_2_hw_phy_config,
-		[RTL_GIGA_MAC_VER_27] = rtl8168d_3_hw_phy_config,
-		[RTL_GIGA_MAC_VER_28] = rtl8168d_4_hw_phy_config,
-		[RTL_GIGA_MAC_VER_29] = rtl8105e_hw_phy_config,
-		[RTL_GIGA_MAC_VER_30] = rtl8105e_hw_phy_config,
-		[RTL_GIGA_MAC_VER_31] = NULL,
-		[RTL_GIGA_MAC_VER_32] = rtl8168e_1_hw_phy_config,
-		[RTL_GIGA_MAC_VER_33] = rtl8168e_1_hw_phy_config,
-		[RTL_GIGA_MAC_VER_34] = rtl8168e_2_hw_phy_config,
-		[RTL_GIGA_MAC_VER_35] = rtl8168f_1_hw_phy_config,
-		[RTL_GIGA_MAC_VER_36] = rtl8168f_2_hw_phy_config,
-		[RTL_GIGA_MAC_VER_37] = rtl8402_hw_phy_config,
-		[RTL_GIGA_MAC_VER_38] = rtl8411_hw_phy_config,
-		[RTL_GIGA_MAC_VER_39] = rtl8106e_hw_phy_config,
-		[RTL_GIGA_MAC_VER_40] = rtl8168g_1_hw_phy_config,
-		[RTL_GIGA_MAC_VER_41] = NULL,
-		[RTL_GIGA_MAC_VER_42] = rtl8168g_2_hw_phy_config,
-		[RTL_GIGA_MAC_VER_43] = rtl8168g_2_hw_phy_config,
-		[RTL_GIGA_MAC_VER_44] = rtl8168g_2_hw_phy_config,
-		[RTL_GIGA_MAC_VER_45] = rtl8168h_1_hw_phy_config,
-		[RTL_GIGA_MAC_VER_46] = rtl8168h_2_hw_phy_config,
-		[RTL_GIGA_MAC_VER_47] = rtl8168h_1_hw_phy_config,
-		[RTL_GIGA_MAC_VER_48] = rtl8168h_2_hw_phy_config,
-		[RTL_GIGA_MAC_VER_49] = rtl8168ep_1_hw_phy_config,
-		[RTL_GIGA_MAC_VER_50] = rtl8168ep_2_hw_phy_config,
-		[RTL_GIGA_MAC_VER_51] = rtl8168ep_2_hw_phy_config,
-		[RTL_GIGA_MAC_VER_52] = rtl8117_hw_phy_config,
-		[RTL_GIGA_MAC_VER_60] = rtl8125_1_hw_phy_config,
-		[RTL_GIGA_MAC_VER_61] = rtl8125_2_hw_phy_config,
-	};
-
-	if (phy_configs[ver])
-		phy_configs[ver](tp, phydev);
-}
-
 static void rtl_schedule_task(struct rtl8169_private *tp, enum rtl_flag flag)
 {
 	if (!test_and_set_bit(flag, tp->wk.flags))
diff --git a/drivers/net/ethernet/realtek/r8169_phy_config.c b/drivers/net/ethernet/realtek/r8169_phy_config.c
new file mode 100644
index 000000000000..e367e77c773b
--- /dev/null
+++ b/drivers/net/ethernet/realtek/r8169_phy_config.c
@@ -0,0 +1,1307 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * r8169_phy_config.c: RealTek 8169/8168/8101 ethernet driver.
+ *
+ * Copyright (c) 2002 ShuChen <shuchen@realtek.com.tw>
+ * Copyright (c) 2003 - 2007 Francois Romieu <romieu@fr.zoreil.com>
+ * Copyright (c) a lot of people too. Please respect their work.
+ *
+ * See MAINTAINERS file for support contact information.
+ */
+
+#include <linux/delay.h>
+#include <linux/phy.h>
+
+#include "r8169.h"
+
+typedef void (*rtl_phy_cfg_fct)(struct rtl8169_private *tp,
+				struct phy_device *phydev);
+
+static void r8168d_modify_extpage(struct phy_device *phydev, int extpage,
+				  int reg, u16 mask, u16 val)
+{
+	int oldpage = phy_select_page(phydev, 0x0007);
+
+	__phy_write(phydev, 0x1e, extpage);
+	__phy_modify(phydev, reg, mask, val);
+
+	phy_restore_page(phydev, oldpage, 0);
+}
+
+static void r8168d_phy_param(struct phy_device *phydev, u16 parm,
+			     u16 mask, u16 val)
+{
+	int oldpage = phy_select_page(phydev, 0x0005);
+
+	__phy_write(phydev, 0x05, parm);
+	__phy_modify(phydev, 0x06, mask, val);
+
+	phy_restore_page(phydev, oldpage, 0);
+}
+
+static void r8168g_phy_param(struct phy_device *phydev, u16 parm,
+			     u16 mask, u16 val)
+{
+	int oldpage = phy_select_page(phydev, 0x0a43);
+
+	__phy_write(phydev, 0x13, parm);
+	__phy_modify(phydev, 0x14, mask, val);
+
+	phy_restore_page(phydev, oldpage, 0);
+}
+
+struct phy_reg {
+	u16 reg;
+	u16 val;
+};
+
+static void __rtl_writephy_batch(struct phy_device *phydev,
+				 const struct phy_reg *regs, int len)
+{
+	phy_lock_mdio_bus(phydev);
+
+	while (len-- > 0) {
+		__phy_write(phydev, regs->reg, regs->val);
+		regs++;
+	}
+
+	phy_unlock_mdio_bus(phydev);
+}
+
+#define rtl_writephy_batch(p, a) __rtl_writephy_batch(p, a, ARRAY_SIZE(a))
+
+static void rtl8168f_config_eee_phy(struct phy_device *phydev)
+{
+	r8168d_modify_extpage(phydev, 0x0020, 0x15, 0, BIT(8));
+	r8168d_phy_param(phydev, 0x8b85, 0, BIT(13));
+}
+
+static void rtl8168g_config_eee_phy(struct phy_device *phydev)
+{
+	phy_modify_paged(phydev, 0x0a43, 0x11, 0, BIT(4));
+}
+
+static void rtl8168h_config_eee_phy(struct phy_device *phydev)
+{
+	rtl8168g_config_eee_phy(phydev);
+
+	phy_modify_paged(phydev, 0xa4a, 0x11, 0x0000, 0x0200);
+	phy_modify_paged(phydev, 0xa42, 0x14, 0x0000, 0x0080);
+}
+
+static void rtl8125_config_eee_phy(struct phy_device *phydev)
+{
+	rtl8168h_config_eee_phy(phydev);
+
+	phy_modify_paged(phydev, 0xa6d, 0x12, 0x0001, 0x0000);
+	phy_modify_paged(phydev, 0xa6d, 0x14, 0x0010, 0x0000);
+}
+
+static void rtl8169s_hw_phy_config(struct rtl8169_private *tp,
+				   struct phy_device *phydev)
+{
+	static const struct phy_reg phy_reg_init[] = {
+		{ 0x1f, 0x0001 },
+		{ 0x06, 0x006e },
+		{ 0x08, 0x0708 },
+		{ 0x15, 0x4000 },
+		{ 0x18, 0x65c7 },
+
+		{ 0x1f, 0x0001 },
+		{ 0x03, 0x00a1 },
+		{ 0x02, 0x0008 },
+		{ 0x01, 0x0120 },
+		{ 0x00, 0x1000 },
+		{ 0x04, 0x0800 },
+		{ 0x04, 0x0000 },
+
+		{ 0x03, 0xff41 },
+		{ 0x02, 0xdf60 },
+		{ 0x01, 0x0140 },
+		{ 0x00, 0x0077 },
+		{ 0x04, 0x7800 },
+		{ 0x04, 0x7000 },
+
+		{ 0x03, 0x802f },
+		{ 0x02, 0x4f02 },
+		{ 0x01, 0x0409 },
+		{ 0x00, 0xf0f9 },
+		{ 0x04, 0x9800 },
+		{ 0x04, 0x9000 },
+
+		{ 0x03, 0xdf01 },
+		{ 0x02, 0xdf20 },
+		{ 0x01, 0xff95 },
+		{ 0x00, 0xba00 },
+		{ 0x04, 0xa800 },
+		{ 0x04, 0xa000 },
+
+		{ 0x03, 0xff41 },
+		{ 0x02, 0xdf20 },
+		{ 0x01, 0x0140 },
+		{ 0x00, 0x00bb },
+		{ 0x04, 0xb800 },
+		{ 0x04, 0xb000 },
+
+		{ 0x03, 0xdf41 },
+		{ 0x02, 0xdc60 },
+		{ 0x01, 0x6340 },
+		{ 0x00, 0x007d },
+		{ 0x04, 0xd800 },
+		{ 0x04, 0xd000 },
+
+		{ 0x03, 0xdf01 },
+		{ 0x02, 0xdf20 },
+		{ 0x01, 0x100a },
+		{ 0x00, 0xa0ff },
+		{ 0x04, 0xf800 },
+		{ 0x04, 0xf000 },
+
+		{ 0x1f, 0x0000 },
+		{ 0x0b, 0x0000 },
+		{ 0x00, 0x9200 }
+	};
+
+	rtl_writephy_batch(phydev, phy_reg_init);
+}
+
+static void rtl8169sb_hw_phy_config(struct rtl8169_private *tp,
+				    struct phy_device *phydev)
+{
+	phy_write_paged(phydev, 0x0002, 0x01, 0x90d0);
+}
+
+static void rtl8169scd_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
+{
+	static const struct phy_reg phy_reg_init[] = {
+		{ 0x1f, 0x0001 },
+		{ 0x04, 0x0000 },
+		{ 0x03, 0x00a1 },
+		{ 0x02, 0x0008 },
+		{ 0x01, 0x0120 },
+		{ 0x00, 0x1000 },
+		{ 0x04, 0x0800 },
+		{ 0x04, 0x9000 },
+		{ 0x03, 0x802f },
+		{ 0x02, 0x4f02 },
+		{ 0x01, 0x0409 },
+		{ 0x00, 0xf099 },
+		{ 0x04, 0x9800 },
+		{ 0x04, 0xa000 },
+		{ 0x03, 0xdf01 },
+		{ 0x02, 0xdf20 },
+		{ 0x01, 0xff95 },
+		{ 0x00, 0xba00 },
+		{ 0x04, 0xa800 },
+		{ 0x04, 0xf000 },
+		{ 0x03, 0xdf01 },
+		{ 0x02, 0xdf20 },
+		{ 0x01, 0x101a },
+		{ 0x00, 0xa0ff },
+		{ 0x04, 0xf800 },
+		{ 0x04, 0x0000 },
+		{ 0x1f, 0x0000 },
+
+		{ 0x1f, 0x0001 },
+		{ 0x10, 0xf41b },
+		{ 0x14, 0xfb54 },
+		{ 0x18, 0xf5c7 },
+		{ 0x1f, 0x0000 },
+
+		{ 0x1f, 0x0001 },
+		{ 0x17, 0x0cc0 },
+		{ 0x1f, 0x0000 }
+	};
+
+	rtl_writephy_batch(phydev, phy_reg_init);
+}
+
+static void rtl8169sce_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
+{
+	static const struct phy_reg phy_reg_init[] = {
+		{ 0x1f, 0x0001 },
+		{ 0x04, 0x0000 },
+		{ 0x03, 0x00a1 },
+		{ 0x02, 0x0008 },
+		{ 0x01, 0x0120 },
+		{ 0x00, 0x1000 },
+		{ 0x04, 0x0800 },
+		{ 0x04, 0x9000 },
+		{ 0x03, 0x802f },
+		{ 0x02, 0x4f02 },
+		{ 0x01, 0x0409 },
+		{ 0x00, 0xf099 },
+		{ 0x04, 0x9800 },
+		{ 0x04, 0xa000 },
+		{ 0x03, 0xdf01 },
+		{ 0x02, 0xdf20 },
+		{ 0x01, 0xff95 },
+		{ 0x00, 0xba00 },
+		{ 0x04, 0xa800 },
+		{ 0x04, 0xf000 },
+		{ 0x03, 0xdf01 },
+		{ 0x02, 0xdf20 },
+		{ 0x01, 0x101a },
+		{ 0x00, 0xa0ff },
+		{ 0x04, 0xf800 },
+		{ 0x04, 0x0000 },
+		{ 0x1f, 0x0000 },
+
+		{ 0x1f, 0x0001 },
+		{ 0x0b, 0x8480 },
+		{ 0x1f, 0x0000 },
+
+		{ 0x1f, 0x0001 },
+		{ 0x18, 0x67c7 },
+		{ 0x04, 0x2000 },
+		{ 0x03, 0x002f },
+		{ 0x02, 0x4360 },
+		{ 0x01, 0x0109 },
+		{ 0x00, 0x3022 },
+		{ 0x04, 0x2800 },
+		{ 0x1f, 0x0000 },
+
+		{ 0x1f, 0x0001 },
+		{ 0x17, 0x0cc0 },
+		{ 0x1f, 0x0000 }
+	};
+
+	rtl_writephy_batch(phydev, phy_reg_init);
+}
+
+static void rtl8168bb_hw_phy_config(struct rtl8169_private *tp,
+				    struct phy_device *phydev)
+{
+	phy_write(phydev, 0x1f, 0x0001);
+	phy_set_bits(phydev, 0x16, BIT(0));
+	phy_write(phydev, 0x10, 0xf41b);
+	phy_write(phydev, 0x1f, 0x0000);
+}
+
+static void rtl8168bef_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
+{
+	phy_write_paged(phydev, 0x0001, 0x10, 0xf41b);
+}
+
+static void rtl8168cp_1_hw_phy_config(struct rtl8169_private *tp,
+				      struct phy_device *phydev)
+{
+	phy_write(phydev, 0x1d, 0x0f00);
+	phy_write_paged(phydev, 0x0002, 0x0c, 0x1ec8);
+}
+
+static void rtl8168cp_2_hw_phy_config(struct rtl8169_private *tp,
+				      struct phy_device *phydev)
+{
+	phy_set_bits(phydev, 0x14, BIT(5));
+	phy_set_bits(phydev, 0x0d, BIT(5));
+	phy_write_paged(phydev, 0x0001, 0x1d, 0x3d98);
+}
+
+static void rtl8168c_1_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
+{
+	static const struct phy_reg phy_reg_init[] = {
+		{ 0x1f, 0x0001 },
+		{ 0x12, 0x2300 },
+		{ 0x1f, 0x0002 },
+		{ 0x00, 0x88d4 },
+		{ 0x01, 0x82b1 },
+		{ 0x03, 0x7002 },
+		{ 0x08, 0x9e30 },
+		{ 0x09, 0x01f0 },
+		{ 0x0a, 0x5500 },
+		{ 0x0c, 0x00c8 },
+		{ 0x1f, 0x0003 },
+		{ 0x12, 0xc096 },
+		{ 0x16, 0x000a },
+		{ 0x1f, 0x0000 },
+		{ 0x1f, 0x0000 },
+		{ 0x09, 0x2000 },
+		{ 0x09, 0x0000 }
+	};
+
+	rtl_writephy_batch(phydev, phy_reg_init);
+
+	phy_set_bits(phydev, 0x14, BIT(5));
+	phy_set_bits(phydev, 0x0d, BIT(5));
+}
+
+static void rtl8168c_2_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
+{
+	static const struct phy_reg phy_reg_init[] = {
+		{ 0x1f, 0x0001 },
+		{ 0x12, 0x2300 },
+		{ 0x03, 0x802f },
+		{ 0x02, 0x4f02 },
+		{ 0x01, 0x0409 },
+		{ 0x00, 0xf099 },
+		{ 0x04, 0x9800 },
+		{ 0x04, 0x9000 },
+		{ 0x1d, 0x3d98 },
+		{ 0x1f, 0x0002 },
+		{ 0x0c, 0x7eb8 },
+		{ 0x06, 0x0761 },
+		{ 0x1f, 0x0003 },
+		{ 0x16, 0x0f0a },
+		{ 0x1f, 0x0000 }
+	};
+
+	rtl_writephy_batch(phydev, phy_reg_init);
+
+	phy_set_bits(phydev, 0x16, BIT(0));
+	phy_set_bits(phydev, 0x14, BIT(5));
+	phy_set_bits(phydev, 0x0d, BIT(5));
+}
+
+static void rtl8168c_3_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
+{
+	static const struct phy_reg phy_reg_init[] = {
+		{ 0x1f, 0x0001 },
+		{ 0x12, 0x2300 },
+		{ 0x1d, 0x3d98 },
+		{ 0x1f, 0x0002 },
+		{ 0x0c, 0x7eb8 },
+		{ 0x06, 0x5461 },
+		{ 0x1f, 0x0003 },
+		{ 0x16, 0x0f0a },
+		{ 0x1f, 0x0000 }
+	};
+
+	rtl_writephy_batch(phydev, phy_reg_init);
+
+	phy_set_bits(phydev, 0x16, BIT(0));
+	phy_set_bits(phydev, 0x14, BIT(5));
+	phy_set_bits(phydev, 0x0d, BIT(5));
+}
+
+static const struct phy_reg rtl8168d_1_phy_reg_init_0[] = {
+	/* Channel Estimation */
+	{ 0x1f, 0x0001 },
+	{ 0x06, 0x4064 },
+	{ 0x07, 0x2863 },
+	{ 0x08, 0x059c },
+	{ 0x09, 0x26b4 },
+	{ 0x0a, 0x6a19 },
+	{ 0x0b, 0xdcc8 },
+	{ 0x10, 0xf06d },
+	{ 0x14, 0x7f68 },
+	{ 0x18, 0x7fd9 },
+	{ 0x1c, 0xf0ff },
+	{ 0x1d, 0x3d9c },
+	{ 0x1f, 0x0003 },
+	{ 0x12, 0xf49f },
+	{ 0x13, 0x070b },
+	{ 0x1a, 0x05ad },
+	{ 0x14, 0x94c0 },
+
+	/*
+	 * Tx Error Issue
+	 * Enhance line driver power
+	 */
+	{ 0x1f, 0x0002 },
+	{ 0x06, 0x5561 },
+	{ 0x1f, 0x0005 },
+	{ 0x05, 0x8332 },
+	{ 0x06, 0x5561 },
+
+	/*
+	 * Can not link to 1Gbps with bad cable
+	 * Decrease SNR threshold form 21.07dB to 19.04dB
+	 */
+	{ 0x1f, 0x0001 },
+	{ 0x17, 0x0cc0 },
+
+	{ 0x1f, 0x0000 },
+	{ 0x0d, 0xf880 }
+};
+
+static const struct phy_reg rtl8168d_1_phy_reg_init_1[] = {
+	{ 0x1f, 0x0002 },
+	{ 0x05, 0x669a },
+	{ 0x1f, 0x0005 },
+	{ 0x05, 0x8330 },
+	{ 0x06, 0x669a },
+	{ 0x1f, 0x0002 }
+};
+
+static void rtl8168d_apply_firmware_cond(struct rtl8169_private *tp,
+					 struct phy_device *phydev,
+					 u16 val)
+{
+	u16 reg_val;
+
+	phy_write(phydev, 0x1f, 0x0005);
+	phy_write(phydev, 0x05, 0x001b);
+	reg_val = phy_read(phydev, 0x06);
+	phy_write(phydev, 0x1f, 0x0000);
+
+	if (reg_val != val)
+		phydev_warn(phydev, "chipset not ready for firmware\n");
+	else
+		r8169_apply_firmware(tp);
+}
+
+static void rtl8168d_1_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
+{
+	rtl_writephy_batch(phydev, rtl8168d_1_phy_reg_init_0);
+
+	/*
+	 * Rx Error Issue
+	 * Fine Tune Switching regulator parameter
+	 */
+	phy_write(phydev, 0x1f, 0x0002);
+	phy_modify(phydev, 0x0b, 0x00ef, 0x0010);
+	phy_modify(phydev, 0x0c, 0x5d00, 0xa200);
+
+	if (rtl8168d_efuse_read(tp, 0x01) == 0xb1) {
+		int val;
+
+		rtl_writephy_batch(phydev, rtl8168d_1_phy_reg_init_1);
+
+		val = phy_read(phydev, 0x0d);
+
+		if ((val & 0x00ff) != 0x006c) {
+			static const u32 set[] = {
+				0x0065, 0x0066, 0x0067, 0x0068,
+				0x0069, 0x006a, 0x006b, 0x006c
+			};
+			int i;
+
+			phy_write(phydev, 0x1f, 0x0002);
+
+			val &= 0xff00;
+			for (i = 0; i < ARRAY_SIZE(set); i++)
+				phy_write(phydev, 0x0d, val | set[i]);
+		}
+	} else {
+		phy_write_paged(phydev, 0x0002, 0x05, 0x6662);
+		r8168d_phy_param(phydev, 0x8330, 0xffff, 0x6662);
+	}
+
+	/* RSET couple improve */
+	phy_write(phydev, 0x1f, 0x0002);
+	phy_set_bits(phydev, 0x0d, 0x0300);
+	phy_set_bits(phydev, 0x0f, 0x0010);
+
+	/* Fine tune PLL performance */
+	phy_write(phydev, 0x1f, 0x0002);
+	phy_modify(phydev, 0x02, 0x0600, 0x0100);
+	phy_clear_bits(phydev, 0x03, 0xe000);
+	phy_write(phydev, 0x1f, 0x0000);
+
+	rtl8168d_apply_firmware_cond(tp, phydev, 0xbf00);
+}
+
+static void rtl8168d_2_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
+{
+	rtl_writephy_batch(phydev, rtl8168d_1_phy_reg_init_0);
+
+	if (rtl8168d_efuse_read(tp, 0x01) == 0xb1) {
+		int val;
+
+		rtl_writephy_batch(phydev, rtl8168d_1_phy_reg_init_1);
+
+		val = phy_read(phydev, 0x0d);
+		if ((val & 0x00ff) != 0x006c) {
+			static const u32 set[] = {
+				0x0065, 0x0066, 0x0067, 0x0068,
+				0x0069, 0x006a, 0x006b, 0x006c
+			};
+			int i;
+
+			phy_write(phydev, 0x1f, 0x0002);
+
+			val &= 0xff00;
+			for (i = 0; i < ARRAY_SIZE(set); i++)
+				phy_write(phydev, 0x0d, val | set[i]);
+		}
+	} else {
+		phy_write_paged(phydev, 0x0002, 0x05, 0x2642);
+		r8168d_phy_param(phydev, 0x8330, 0xffff, 0x2642);
+	}
+
+	/* Fine tune PLL performance */
+	phy_write(phydev, 0x1f, 0x0002);
+	phy_modify(phydev, 0x02, 0x0600, 0x0100);
+	phy_clear_bits(phydev, 0x03, 0xe000);
+	phy_write(phydev, 0x1f, 0x0000);
+
+	/* Switching regulator Slew rate */
+	phy_modify_paged(phydev, 0x0002, 0x0f, 0x0000, 0x0017);
+
+	rtl8168d_apply_firmware_cond(tp, phydev, 0xb300);
+}
+
+static void rtl8168d_3_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
+{
+	static const struct phy_reg phy_reg_init[] = {
+		{ 0x1f, 0x0002 },
+		{ 0x10, 0x0008 },
+		{ 0x0d, 0x006c },
+
+		{ 0x1f, 0x0000 },
+		{ 0x0d, 0xf880 },
+
+		{ 0x1f, 0x0001 },
+		{ 0x17, 0x0cc0 },
+
+		{ 0x1f, 0x0001 },
+		{ 0x0b, 0xa4d8 },
+		{ 0x09, 0x281c },
+		{ 0x07, 0x2883 },
+		{ 0x0a, 0x6b35 },
+		{ 0x1d, 0x3da4 },
+		{ 0x1c, 0xeffd },
+		{ 0x14, 0x7f52 },
+		{ 0x18, 0x7fc6 },
+		{ 0x08, 0x0601 },
+		{ 0x06, 0x4063 },
+		{ 0x10, 0xf074 },
+		{ 0x1f, 0x0003 },
+		{ 0x13, 0x0789 },
+		{ 0x12, 0xf4bd },
+		{ 0x1a, 0x04fd },
+		{ 0x14, 0x84b0 },
+		{ 0x1f, 0x0000 },
+		{ 0x00, 0x9200 },
+
+		{ 0x1f, 0x0005 },
+		{ 0x01, 0x0340 },
+		{ 0x1f, 0x0001 },
+		{ 0x04, 0x4000 },
+		{ 0x03, 0x1d21 },
+		{ 0x02, 0x0c32 },
+		{ 0x01, 0x0200 },
+		{ 0x00, 0x5554 },
+		{ 0x04, 0x4800 },
+		{ 0x04, 0x4000 },
+		{ 0x04, 0xf000 },
+		{ 0x03, 0xdf01 },
+		{ 0x02, 0xdf20 },
+		{ 0x01, 0x101a },
+		{ 0x00, 0xa0ff },
+		{ 0x04, 0xf800 },
+		{ 0x04, 0xf000 },
+		{ 0x1f, 0x0000 },
+	};
+
+	rtl_writephy_batch(phydev, phy_reg_init);
+	r8168d_modify_extpage(phydev, 0x0023, 0x16, 0xffff, 0x0000);
+}
+
+static void rtl8168d_4_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
+{
+	phy_write_paged(phydev, 0x0001, 0x17, 0x0cc0);
+	r8168d_modify_extpage(phydev, 0x002d, 0x18, 0xffff, 0x0040);
+	phy_set_bits(phydev, 0x0d, BIT(5));
+}
+
+static void rtl8168e_1_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
+{
+	static const struct phy_reg phy_reg_init[] = {
+		/* Channel estimation fine tune */
+		{ 0x1f, 0x0001 },
+		{ 0x0b, 0x6c20 },
+		{ 0x07, 0x2872 },
+		{ 0x1c, 0xefff },
+		{ 0x1f, 0x0003 },
+		{ 0x14, 0x6420 },
+		{ 0x1f, 0x0000 },
+	};
+
+	r8169_apply_firmware(tp);
+
+	/* Enable Delay cap */
+	r8168d_phy_param(phydev, 0x8b80, 0xffff, 0xc896);
+
+	rtl_writephy_batch(phydev, phy_reg_init);
+
+	/* Update PFM & 10M TX idle timer */
+	r8168d_modify_extpage(phydev, 0x002f, 0x15, 0xffff, 0x1919);
+
+	r8168d_modify_extpage(phydev, 0x00ac, 0x18, 0xffff, 0x0006);
+
+	/* DCO enable for 10M IDLE Power */
+	r8168d_modify_extpage(phydev, 0x0023, 0x17, 0x0000, 0x0006);
+
+	/* For impedance matching */
+	phy_modify_paged(phydev, 0x0002, 0x08, 0x7f00, 0x8000);
+
+	/* PHY auto speed down */
+	r8168d_modify_extpage(phydev, 0x002d, 0x18, 0x0000, 0x0050);
+	phy_set_bits(phydev, 0x14, BIT(15));
+
+	r8168d_phy_param(phydev, 0x8b86, 0x0000, 0x0001);
+	r8168d_phy_param(phydev, 0x8b85, 0x2000, 0x0000);
+
+	r8168d_modify_extpage(phydev, 0x0020, 0x15, 0x1100, 0x0000);
+	phy_write_paged(phydev, 0x0006, 0x00, 0x5a00);
+
+	phy_write_mmd(phydev, MDIO_MMD_AN, MDIO_AN_EEE_ADV, 0x0000);
+}
+
+static void rtl8168e_2_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
+{
+	r8169_apply_firmware(tp);
+
+	/* Enable Delay cap */
+	r8168d_modify_extpage(phydev, 0x00ac, 0x18, 0xffff, 0x0006);
+
+	/* Channel estimation fine tune */
+	phy_write_paged(phydev, 0x0003, 0x09, 0xa20f);
+
+	/* Green Setting */
+	r8168d_phy_param(phydev, 0x8b5b, 0xffff, 0x9222);
+	r8168d_phy_param(phydev, 0x8b6d, 0xffff, 0x8000);
+	r8168d_phy_param(phydev, 0x8b76, 0xffff, 0x8000);
+
+	/* For 4-corner performance improve */
+	phy_write(phydev, 0x1f, 0x0005);
+	phy_write(phydev, 0x05, 0x8b80);
+	phy_set_bits(phydev, 0x17, 0x0006);
+	phy_write(phydev, 0x1f, 0x0000);
+
+	/* PHY auto speed down */
+	r8168d_modify_extpage(phydev, 0x002d, 0x18, 0x0000, 0x0010);
+	phy_set_bits(phydev, 0x14, BIT(15));
+
+	/* improve 10M EEE waveform */
+	r8168d_phy_param(phydev, 0x8b86, 0x0000, 0x0001);
+
+	/* Improve 2-pair detection performance */
+	r8168d_phy_param(phydev, 0x8b85, 0x0000, 0x4000);
+
+	rtl8168f_config_eee_phy(phydev);
+
+	/* Green feature */
+	phy_write(phydev, 0x1f, 0x0003);
+	phy_set_bits(phydev, 0x19, BIT(0));
+	phy_set_bits(phydev, 0x10, BIT(10));
+	phy_write(phydev, 0x1f, 0x0000);
+	phy_modify_paged(phydev, 0x0005, 0x01, 0, BIT(8));
+}
+
+static void rtl8168f_hw_phy_config(struct rtl8169_private *tp,
+				   struct phy_device *phydev)
+{
+	/* For 4-corner performance improve */
+	r8168d_phy_param(phydev, 0x8b80, 0x0000, 0x0006);
+
+	/* PHY auto speed down */
+	r8168d_modify_extpage(phydev, 0x002d, 0x18, 0x0000, 0x0010);
+	phy_set_bits(phydev, 0x14, BIT(15));
+
+	/* Improve 10M EEE waveform */
+	r8168d_phy_param(phydev, 0x8b86, 0x0000, 0x0001);
+
+	rtl8168f_config_eee_phy(phydev);
+}
+
+static void rtl8168f_1_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
+{
+	r8169_apply_firmware(tp);
+
+	/* Channel estimation fine tune */
+	phy_write_paged(phydev, 0x0003, 0x09, 0xa20f);
+
+	/* Modify green table for giga & fnet */
+	r8168d_phy_param(phydev, 0x8b55, 0xffff, 0x0000);
+	r8168d_phy_param(phydev, 0x8b5e, 0xffff, 0x0000);
+	r8168d_phy_param(phydev, 0x8b67, 0xffff, 0x0000);
+	r8168d_phy_param(phydev, 0x8b70, 0xffff, 0x0000);
+	r8168d_modify_extpage(phydev, 0x0078, 0x17, 0xffff, 0x0000);
+	r8168d_modify_extpage(phydev, 0x0078, 0x19, 0xffff, 0x00fb);
+
+	/* Modify green table for 10M */
+	r8168d_phy_param(phydev, 0x8b79, 0xffff, 0xaa00);
+
+	/* Disable hiimpedance detection (RTCT) */
+	phy_write_paged(phydev, 0x0003, 0x01, 0x328a);
+
+	rtl8168f_hw_phy_config(tp, phydev);
+
+	/* Improve 2-pair detection performance */
+	r8168d_phy_param(phydev, 0x8b85, 0x0000, 0x4000);
+}
+
+static void rtl8168f_2_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
+{
+	r8169_apply_firmware(tp);
+
+	rtl8168f_hw_phy_config(tp, phydev);
+}
+
+static void rtl8411_hw_phy_config(struct rtl8169_private *tp,
+				  struct phy_device *phydev)
+{
+	r8169_apply_firmware(tp);
+
+	rtl8168f_hw_phy_config(tp, phydev);
+
+	/* Improve 2-pair detection performance */
+	r8168d_phy_param(phydev, 0x8b85, 0x0000, 0x4000);
+
+	/* Channel estimation fine tune */
+	phy_write_paged(phydev, 0x0003, 0x09, 0xa20f);
+
+	/* Modify green table for giga & fnet */
+	r8168d_phy_param(phydev, 0x8b55, 0xffff, 0x0000);
+	r8168d_phy_param(phydev, 0x8b5e, 0xffff, 0x0000);
+	r8168d_phy_param(phydev, 0x8b67, 0xffff, 0x0000);
+	r8168d_phy_param(phydev, 0x8b70, 0xffff, 0x0000);
+	r8168d_modify_extpage(phydev, 0x0078, 0x17, 0xffff, 0x0000);
+	r8168d_modify_extpage(phydev, 0x0078, 0x19, 0xffff, 0x00aa);
+
+	/* Modify green table for 10M */
+	r8168d_phy_param(phydev, 0x8b79, 0xffff, 0xaa00);
+
+	/* Disable hiimpedance detection (RTCT) */
+	phy_write_paged(phydev, 0x0003, 0x01, 0x328a);
+
+	/* Modify green table for giga */
+	r8168d_phy_param(phydev, 0x8b54, 0x0800, 0x0000);
+	r8168d_phy_param(phydev, 0x8b5d, 0x0800, 0x0000);
+	r8168d_phy_param(phydev, 0x8a7c, 0x0100, 0x0000);
+	r8168d_phy_param(phydev, 0x8a7f, 0x0000, 0x0100);
+	r8168d_phy_param(phydev, 0x8a82, 0x0100, 0x0000);
+	r8168d_phy_param(phydev, 0x8a85, 0x0100, 0x0000);
+	r8168d_phy_param(phydev, 0x8a88, 0x0100, 0x0000);
+
+	/* uc same-seed solution */
+	r8168d_phy_param(phydev, 0x8b85, 0x0000, 0x8000);
+
+	/* Green feature */
+	phy_write(phydev, 0x1f, 0x0003);
+	phy_clear_bits(phydev, 0x19, BIT(0));
+	phy_clear_bits(phydev, 0x10, BIT(10));
+	phy_write(phydev, 0x1f, 0x0000);
+}
+
+static void rtl8168g_disable_aldps(struct phy_device *phydev)
+{
+	phy_modify_paged(phydev, 0x0a43, 0x10, BIT(2), 0);
+}
+
+static void rtl8168g_phy_adjust_10m_aldps(struct phy_device *phydev)
+{
+	phy_modify_paged(phydev, 0x0bcc, 0x14, BIT(8), 0);
+	phy_modify_paged(phydev, 0x0a44, 0x11, 0, BIT(7) | BIT(6));
+	r8168g_phy_param(phydev, 0x8084, 0x6000, 0x0000);
+	phy_modify_paged(phydev, 0x0a43, 0x10, 0x0000, 0x1003);
+}
+
+static void rtl8168g_1_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
+{
+	int ret;
+
+	r8169_apply_firmware(tp);
+
+	ret = phy_read_paged(phydev, 0x0a46, 0x10);
+	if (ret & BIT(8))
+		phy_modify_paged(phydev, 0x0bcc, 0x12, BIT(15), 0);
+	else
+		phy_modify_paged(phydev, 0x0bcc, 0x12, 0, BIT(15));
+
+	ret = phy_read_paged(phydev, 0x0a46, 0x13);
+	if (ret & BIT(8))
+		phy_modify_paged(phydev, 0x0c41, 0x15, 0, BIT(1));
+	else
+		phy_modify_paged(phydev, 0x0c41, 0x15, BIT(1), 0);
+
+	/* Enable PHY auto speed down */
+	phy_modify_paged(phydev, 0x0a44, 0x11, 0, BIT(3) | BIT(2));
+
+	rtl8168g_phy_adjust_10m_aldps(phydev);
+
+	/* EEE auto-fallback function */
+	phy_modify_paged(phydev, 0x0a4b, 0x11, 0, BIT(2));
+
+	/* Enable UC LPF tune function */
+	r8168g_phy_param(phydev, 0x8012, 0x0000, 0x8000);
+
+	phy_modify_paged(phydev, 0x0c42, 0x11, BIT(13), BIT(14));
+
+	/* Improve SWR Efficiency */
+	phy_write(phydev, 0x1f, 0x0bcd);
+	phy_write(phydev, 0x14, 0x5065);
+	phy_write(phydev, 0x14, 0xd065);
+	phy_write(phydev, 0x1f, 0x0bc8);
+	phy_write(phydev, 0x11, 0x5655);
+	phy_write(phydev, 0x1f, 0x0bcd);
+	phy_write(phydev, 0x14, 0x1065);
+	phy_write(phydev, 0x14, 0x9065);
+	phy_write(phydev, 0x14, 0x1065);
+	phy_write(phydev, 0x1f, 0x0000);
+
+	rtl8168g_disable_aldps(phydev);
+	rtl8168g_config_eee_phy(phydev);
+}
+
+static void rtl8168g_2_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
+{
+	r8169_apply_firmware(tp);
+	rtl8168g_config_eee_phy(phydev);
+}
+
+static void rtl8168h_1_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
+{
+	u16 dout_tapbin;
+	u32 data;
+
+	r8169_apply_firmware(tp);
+
+	/* CHN EST parameters adjust - giga master */
+	r8168g_phy_param(phydev, 0x809b, 0xf800, 0x8000);
+	r8168g_phy_param(phydev, 0x80a2, 0xff00, 0x8000);
+	r8168g_phy_param(phydev, 0x80a4, 0xff00, 0x8500);
+	r8168g_phy_param(phydev, 0x809c, 0xff00, 0xbd00);
+
+	/* CHN EST parameters adjust - giga slave */
+	r8168g_phy_param(phydev, 0x80ad, 0xf800, 0x7000);
+	r8168g_phy_param(phydev, 0x80b4, 0xff00, 0x5000);
+	r8168g_phy_param(phydev, 0x80ac, 0xff00, 0x4000);
+
+	/* CHN EST parameters adjust - fnet */
+	r8168g_phy_param(phydev, 0x808e, 0xff00, 0x1200);
+	r8168g_phy_param(phydev, 0x8090, 0xff00, 0xe500);
+	r8168g_phy_param(phydev, 0x8092, 0xff00, 0x9f00);
+
+	/* enable R-tune & PGA-retune function */
+	dout_tapbin = 0;
+	data = phy_read_paged(phydev, 0x0a46, 0x13);
+	data &= 3;
+	data <<= 2;
+	dout_tapbin |= data;
+	data = phy_read_paged(phydev, 0x0a46, 0x12);
+	data &= 0xc000;
+	data >>= 14;
+	dout_tapbin |= data;
+	dout_tapbin = ~(dout_tapbin ^ 0x08);
+	dout_tapbin <<= 12;
+	dout_tapbin &= 0xf000;
+
+	r8168g_phy_param(phydev, 0x827a, 0xf000, dout_tapbin);
+	r8168g_phy_param(phydev, 0x827b, 0xf000, dout_tapbin);
+	r8168g_phy_param(phydev, 0x827c, 0xf000, dout_tapbin);
+	r8168g_phy_param(phydev, 0x827d, 0xf000, dout_tapbin);
+	r8168g_phy_param(phydev, 0x0811, 0x0000, 0x0800);
+	phy_modify_paged(phydev, 0x0a42, 0x16, 0x0000, 0x0002);
+
+	/* enable GPHY 10M */
+	phy_modify_paged(phydev, 0x0a44, 0x11, 0, BIT(11));
+
+	/* SAR ADC performance */
+	phy_modify_paged(phydev, 0x0bca, 0x17, BIT(12) | BIT(13), BIT(14));
+
+	r8168g_phy_param(phydev, 0x803f, 0x3000, 0x0000);
+	r8168g_phy_param(phydev, 0x8047, 0x3000, 0x0000);
+	r8168g_phy_param(phydev, 0x804f, 0x3000, 0x0000);
+	r8168g_phy_param(phydev, 0x8057, 0x3000, 0x0000);
+	r8168g_phy_param(phydev, 0x805f, 0x3000, 0x0000);
+	r8168g_phy_param(phydev, 0x8067, 0x3000, 0x0000);
+	r8168g_phy_param(phydev, 0x806f, 0x3000, 0x0000);
+
+	/* disable phy pfm mode */
+	phy_modify_paged(phydev, 0x0a44, 0x11, BIT(7), 0);
+
+	rtl8168g_disable_aldps(phydev);
+	rtl8168h_config_eee_phy(phydev);
+}
+
+static void rtl8168h_2_hw_phy_config(struct rtl8169_private *tp,
+				     struct phy_device *phydev)
+{
+	u16 ioffset, rlen;
+	u32 data;
+
+	r8169_apply_firmware(tp);
+
+	/* CHIN EST parameter update */
+	r8168g_phy_param(phydev, 0x808a, 0x003f, 0x000a);
+
+	/* enable R-tune & PGA-retune function */
+	r8168g_phy_param(phydev, 0x0811, 0x0000, 0x0800);
+	phy_modify_paged(phydev, 0x0a42, 0x16, 0x0000, 0x0002);
+
+	/* enable GPHY 10M */
+	phy_modify_paged(phydev, 0x0a44, 0x11, 0, BIT(11));
+
+	ioffset = rtl8168h_2_get_adc_bias_ioffset(tp);
+	if (ioffset != 0xffff)
+		phy_write_paged(phydev, 0x0bcf, 0x16, ioffset);
+
+	/* Modify rlen (TX LPF corner frequency) level */
+	data = phy_read_paged(phydev, 0x0bcd, 0x16);
+	data &= 0x000f;
+	rlen = 0;
+	if (data > 3)
+		rlen = data - 3;
+	data = rlen | (rlen << 4) | (rlen << 8) | (rlen << 12);
+	phy_write_paged(phydev, 0x0bcd, 0x17, data);
+
+	/* disable phy pfm mode */
+	phy_modify_paged(phydev, 0x0a44, 0x11, BIT(7), 0);
+
+	rtl8168g_disable_aldps(phydev);
+	rtl8168g_config_eee_phy(phydev);
+}
+
+static void rtl8168ep_1_hw_phy_config(struct rtl8169_private *tp,
+				      struct phy_device *phydev)
+{
+	/* Enable PHY auto speed down */
+	phy_modify_paged(phydev, 0x0a44, 0x11, 0, BIT(3) | BIT(2));
+
+	rtl8168g_phy_adjust_10m_aldps(phydev);
+
+	/* Enable EEE auto-fallback function */
+	phy_modify_paged(phydev, 0x0a4b, 0x11, 0, BIT(2));
+
+	/* Enable UC LPF tune function */
+	r8168g_phy_param(phydev, 0x8012, 0x0000, 0x8000);
+
+	/* set rg_sel_sdm_rate */
+	phy_modify_paged(phydev, 0x0c42, 0x11, BIT(13), BIT(14));
+
+	rtl8168g_disable_aldps(phydev);
+	rtl8168g_config_eee_phy(phydev);
+}
+
+static void rtl8168ep_2_hw_phy_config(struct rtl8169_private *tp,
+				      struct phy_device *phydev)
+{
+	rtl8168g_phy_adjust_10m_aldps(phydev);
+
+	/* Enable UC LPF tune function */
+	r8168g_phy_param(phydev, 0x8012, 0x0000, 0x8000);
+
+	/* Set rg_sel_sdm_rate */
+	phy_modify_paged(phydev, 0x0c42, 0x11, BIT(13), BIT(14));
+
+	/* Channel estimation parameters */
+	r8168g_phy_param(phydev, 0x80f3, 0xff00, 0x8b00);
+	r8168g_phy_param(phydev, 0x80f0, 0xff00, 0x3a00);
+	r8168g_phy_param(phydev, 0x80ef, 0xff00, 0x0500);
+	r8168g_phy_param(phydev, 0x80f6, 0xff00, 0x6e00);
+	r8168g_phy_param(phydev, 0x80ec, 0xff00, 0x6800);
+	r8168g_phy_param(phydev, 0x80ed, 0xff00, 0x7c00);
+	r8168g_phy_param(phydev, 0x80f2, 0xff00, 0xf400);
+	r8168g_phy_param(phydev, 0x80f4, 0xff00, 0x8500);
+	r8168g_phy_param(phydev, 0x8110, 0xff00, 0xa800);
+	r8168g_phy_param(phydev, 0x810f, 0xff00, 0x1d00);
+	r8168g_phy_param(phydev, 0x8111, 0xff00, 0xf500);
+	r8168g_phy_param(phydev, 0x8113, 0xff00, 0x6100);
+	r8168g_phy_param(phydev, 0x8115, 0xff00, 0x9200);
+	r8168g_phy_param(phydev, 0x810e, 0xff00, 0x0400);
+	r8168g_phy_param(phydev, 0x810c, 0xff00, 0x7c00);
+	r8168g_phy_param(phydev, 0x810b, 0xff00, 0x5a00);
+	r8168g_phy_param(phydev, 0x80d1, 0xff00, 0xff00);
+	r8168g_phy_param(phydev, 0x80cd, 0xff00, 0x9e00);
+	r8168g_phy_param(phydev, 0x80d3, 0xff00, 0x0e00);
+	r8168g_phy_param(phydev, 0x80d5, 0xff00, 0xca00);
+	r8168g_phy_param(phydev, 0x80d7, 0xff00, 0x8400);
+
+	/* Force PWM-mode */
+	phy_write(phydev, 0x1f, 0x0bcd);
+	phy_write(phydev, 0x14, 0x5065);
+	phy_write(phydev, 0x14, 0xd065);
+	phy_write(phydev, 0x1f, 0x0bc8);
+	phy_write(phydev, 0x12, 0x00ed);
+	phy_write(phydev, 0x1f, 0x0bcd);
+	phy_write(phydev, 0x14, 0x1065);
+	phy_write(phydev, 0x14, 0x9065);
+	phy_write(phydev, 0x14, 0x1065);
+	phy_write(phydev, 0x1f, 0x0000);
+
+	rtl8168g_disable_aldps(phydev);
+	rtl8168g_config_eee_phy(phydev);
+}
+
+static void rtl8117_hw_phy_config(struct rtl8169_private *tp,
+				  struct phy_device *phydev)
+{
+	/* CHN EST parameters adjust - fnet */
+	r8168g_phy_param(phydev, 0x808e, 0xff00, 0x4800);
+	r8168g_phy_param(phydev, 0x8090, 0xff00, 0xcc00);
+	r8168g_phy_param(phydev, 0x8092, 0xff00, 0xb000);
+
+	r8168g_phy_param(phydev, 0x8088, 0xff00, 0x6000);
+	r8168g_phy_param(phydev, 0x808b, 0x3f00, 0x0b00);
+	r8168g_phy_param(phydev, 0x808d, 0x1f00, 0x0600);
+	r8168g_phy_param(phydev, 0x808c, 0xff00, 0xb000);
+	r8168g_phy_param(phydev, 0x80a0, 0xff00, 0x2800);
+	r8168g_phy_param(phydev, 0x80a2, 0xff00, 0x5000);
+	r8168g_phy_param(phydev, 0x809b, 0xf800, 0xb000);
+	r8168g_phy_param(phydev, 0x809a, 0xff00, 0x4b00);
+	r8168g_phy_param(phydev, 0x809d, 0x3f00, 0x0800);
+	r8168g_phy_param(phydev, 0x80a1, 0xff00, 0x7000);
+	r8168g_phy_param(phydev, 0x809f, 0x1f00, 0x0300);
+	r8168g_phy_param(phydev, 0x809e, 0xff00, 0x8800);
+	r8168g_phy_param(phydev, 0x80b2, 0xff00, 0x2200);
+	r8168g_phy_param(phydev, 0x80ad, 0xf800, 0x9800);
+	r8168g_phy_param(phydev, 0x80af, 0x3f00, 0x0800);
+	r8168g_phy_param(phydev, 0x80b3, 0xff00, 0x6f00);
+	r8168g_phy_param(phydev, 0x80b1, 0x1f00, 0x0300);
+	r8168g_phy_param(phydev, 0x80b0, 0xff00, 0x9300);
+
+	r8168g_phy_param(phydev, 0x8011, 0x0000, 0x0800);
+
+	/* enable GPHY 10M */
+	phy_modify_paged(phydev, 0x0a44, 0x11, 0, BIT(11));
+
+	r8168g_phy_param(phydev, 0x8016, 0x0000, 0x0400);
+
+	rtl8168g_disable_aldps(phydev);
+	rtl8168h_config_eee_phy(phydev);
+}
+
+static void rtl8102e_hw_phy_config(struct rtl8169_private *tp,
+				   struct phy_device *phydev)
+{
+	static const struct phy_reg phy_reg_init[] = {
+		{ 0x1f, 0x0003 },
+		{ 0x08, 0x441d },
+		{ 0x01, 0x9100 },
+		{ 0x1f, 0x0000 }
+	};
+
+	phy_set_bits(phydev, 0x11, BIT(12));
+	phy_set_bits(phydev, 0x19, BIT(13));
+	phy_set_bits(phydev, 0x10, BIT(15));
+
+	rtl_writephy_batch(phydev, phy_reg_init);
+}
+
+static void rtl8105e_hw_phy_config(struct rtl8169_private *tp,
+				   struct phy_device *phydev)
+{
+	/* Disable ALDPS before ram code */
+	phy_write(phydev, 0x18, 0x0310);
+	msleep(100);
+
+	r8169_apply_firmware(tp);
+
+	phy_write_paged(phydev, 0x0005, 0x1a, 0x0000);
+	phy_write_paged(phydev, 0x0004, 0x1c, 0x0000);
+	phy_write_paged(phydev, 0x0001, 0x15, 0x7701);
+}
+
+static void rtl8402_hw_phy_config(struct rtl8169_private *tp,
+				  struct phy_device *phydev)
+{
+	/* Disable ALDPS before setting firmware */
+	phy_write(phydev, 0x18, 0x0310);
+	msleep(20);
+
+	r8169_apply_firmware(tp);
+
+	/* EEE setting */
+	phy_write(phydev, 0x1f, 0x0004);
+	phy_write(phydev, 0x10, 0x401f);
+	phy_write(phydev, 0x19, 0x7030);
+	phy_write(phydev, 0x1f, 0x0000);
+}
+
+static void rtl8106e_hw_phy_config(struct rtl8169_private *tp,
+				   struct phy_device *phydev)
+{
+	static const struct phy_reg phy_reg_init[] = {
+		{ 0x1f, 0x0004 },
+		{ 0x10, 0xc07f },
+		{ 0x19, 0x7030 },
+		{ 0x1f, 0x0000 }
+	};
+
+	/* Disable ALDPS before ram code */
+	phy_write(phydev, 0x18, 0x0310);
+	msleep(100);
+
+	r8169_apply_firmware(tp);
+
+	rtl_writephy_batch(phydev, phy_reg_init);
+}
+
+static void rtl8125_1_hw_phy_config(struct rtl8169_private *tp,
+				    struct phy_device *phydev)
+{
+	phy_modify_paged(phydev, 0xad4, 0x10, 0x03ff, 0x0084);
+	phy_modify_paged(phydev, 0xad4, 0x17, 0x0000, 0x0010);
+	phy_modify_paged(phydev, 0xad1, 0x13, 0x03ff, 0x0006);
+	phy_modify_paged(phydev, 0xad3, 0x11, 0x003f, 0x0006);
+	phy_modify_paged(phydev, 0xac0, 0x14, 0x0000, 0x1100);
+	phy_modify_paged(phydev, 0xac8, 0x15, 0xf000, 0x7000);
+	phy_modify_paged(phydev, 0xad1, 0x14, 0x0000, 0x0400);
+	phy_modify_paged(phydev, 0xad1, 0x15, 0x0000, 0x03ff);
+	phy_modify_paged(phydev, 0xad1, 0x16, 0x0000, 0x03ff);
+
+	r8168g_phy_param(phydev, 0x80ea, 0xff00, 0xc400);
+	r8168g_phy_param(phydev, 0x80eb, 0x0700, 0x0300);
+	r8168g_phy_param(phydev, 0x80f8, 0xff00, 0x1c00);
+	r8168g_phy_param(phydev, 0x80f1, 0xff00, 0x3000);
+	r8168g_phy_param(phydev, 0x80fe, 0xff00, 0xa500);
+	r8168g_phy_param(phydev, 0x8102, 0xff00, 0x5000);
+	r8168g_phy_param(phydev, 0x8105, 0xff00, 0x3300);
+	r8168g_phy_param(phydev, 0x8100, 0xff00, 0x7000);
+	r8168g_phy_param(phydev, 0x8104, 0xff00, 0xf000);
+	r8168g_phy_param(phydev, 0x8106, 0xff00, 0x6500);
+	r8168g_phy_param(phydev, 0x80dc, 0xff00, 0xed00);
+	r8168g_phy_param(phydev, 0x80df, 0x0000, 0x0100);
+	r8168g_phy_param(phydev, 0x80e1, 0x0100, 0x0000);
+
+	phy_modify_paged(phydev, 0xbf0, 0x13, 0x003f, 0x0038);
+	r8168g_phy_param(phydev, 0x819f, 0xffff, 0xd0b6);
+
+	phy_write_paged(phydev, 0xbc3, 0x12, 0x5555);
+	phy_modify_paged(phydev, 0xbf0, 0x15, 0x0e00, 0x0a00);
+	phy_modify_paged(phydev, 0xa5c, 0x10, 0x0400, 0x0000);
+	phy_modify_paged(phydev, 0xa44, 0x11, 0x0000, 0x0800);
+
+	rtl8125_config_eee_phy(phydev);
+}
+
+static void rtl8125_2_hw_phy_config(struct rtl8169_private *tp,
+				    struct phy_device *phydev)
+{
+	int i;
+
+	phy_modify_paged(phydev, 0xad4, 0x17, 0x0000, 0x0010);
+	phy_modify_paged(phydev, 0xad1, 0x13, 0x03ff, 0x03ff);
+	phy_modify_paged(phydev, 0xad3, 0x11, 0x003f, 0x0006);
+	phy_modify_paged(phydev, 0xac0, 0x14, 0x1100, 0x0000);
+	phy_modify_paged(phydev, 0xacc, 0x10, 0x0003, 0x0002);
+	phy_modify_paged(phydev, 0xad4, 0x10, 0x00e7, 0x0044);
+	phy_modify_paged(phydev, 0xac1, 0x12, 0x0080, 0x0000);
+	phy_modify_paged(phydev, 0xac8, 0x10, 0x0300, 0x0000);
+	phy_modify_paged(phydev, 0xac5, 0x17, 0x0007, 0x0002);
+	phy_write_paged(phydev, 0xad4, 0x16, 0x00a8);
+	phy_write_paged(phydev, 0xac5, 0x16, 0x01ff);
+	phy_modify_paged(phydev, 0xac8, 0x15, 0x00f0, 0x0030);
+
+	phy_write(phydev, 0x1f, 0x0b87);
+	phy_write(phydev, 0x16, 0x80a2);
+	phy_write(phydev, 0x17, 0x0153);
+	phy_write(phydev, 0x16, 0x809c);
+	phy_write(phydev, 0x17, 0x0153);
+	phy_write(phydev, 0x1f, 0x0000);
+
+	phy_write(phydev, 0x1f, 0x0a43);
+	phy_write(phydev, 0x13, 0x81B3);
+	phy_write(phydev, 0x14, 0x0043);
+	phy_write(phydev, 0x14, 0x00A7);
+	phy_write(phydev, 0x14, 0x00D6);
+	phy_write(phydev, 0x14, 0x00EC);
+	phy_write(phydev, 0x14, 0x00F6);
+	phy_write(phydev, 0x14, 0x00FB);
+	phy_write(phydev, 0x14, 0x00FD);
+	phy_write(phydev, 0x14, 0x00FF);
+	phy_write(phydev, 0x14, 0x00BB);
+	phy_write(phydev, 0x14, 0x0058);
+	phy_write(phydev, 0x14, 0x0029);
+	phy_write(phydev, 0x14, 0x0013);
+	phy_write(phydev, 0x14, 0x0009);
+	phy_write(phydev, 0x14, 0x0004);
+	phy_write(phydev, 0x14, 0x0002);
+	for (i = 0; i < 25; i++)
+		phy_write(phydev, 0x14, 0x0000);
+	phy_write(phydev, 0x1f, 0x0000);
+
+	r8168g_phy_param(phydev, 0x8257, 0xffff, 0x020F);
+	r8168g_phy_param(phydev, 0x80ea, 0xffff, 0x7843);
+
+	r8169_apply_firmware(tp);
+
+	phy_modify_paged(phydev, 0xd06, 0x14, 0x0000, 0x2000);
+
+	r8168g_phy_param(phydev, 0x81a2, 0x0000, 0x0100);
+
+	phy_modify_paged(phydev, 0xb54, 0x16, 0xff00, 0xdb00);
+	phy_modify_paged(phydev, 0xa45, 0x12, 0x0001, 0x0000);
+	phy_modify_paged(phydev, 0xa5d, 0x12, 0x0000, 0x0020);
+	phy_modify_paged(phydev, 0xad4, 0x17, 0x0010, 0x0000);
+	phy_modify_paged(phydev, 0xa86, 0x15, 0x0001, 0x0000);
+	phy_modify_paged(phydev, 0xa44, 0x11, 0x0000, 0x0800);
+
+	rtl8125_config_eee_phy(phydev);
+}
+
+void r8169_hw_phy_config(struct rtl8169_private *tp, struct phy_device *phydev,
+			 enum mac_version ver)
+{
+	static const rtl_phy_cfg_fct phy_configs[] = {
+		/* PCI devices. */
+		[RTL_GIGA_MAC_VER_02] = rtl8169s_hw_phy_config,
+		[RTL_GIGA_MAC_VER_03] = rtl8169s_hw_phy_config,
+		[RTL_GIGA_MAC_VER_04] = rtl8169sb_hw_phy_config,
+		[RTL_GIGA_MAC_VER_05] = rtl8169scd_hw_phy_config,
+		[RTL_GIGA_MAC_VER_06] = rtl8169sce_hw_phy_config,
+		/* PCI-E devices. */
+		[RTL_GIGA_MAC_VER_07] = rtl8102e_hw_phy_config,
+		[RTL_GIGA_MAC_VER_08] = rtl8102e_hw_phy_config,
+		[RTL_GIGA_MAC_VER_09] = rtl8102e_hw_phy_config,
+		[RTL_GIGA_MAC_VER_10] = NULL,
+		[RTL_GIGA_MAC_VER_11] = rtl8168bb_hw_phy_config,
+		[RTL_GIGA_MAC_VER_12] = rtl8168bef_hw_phy_config,
+		[RTL_GIGA_MAC_VER_13] = NULL,
+		[RTL_GIGA_MAC_VER_14] = NULL,
+		[RTL_GIGA_MAC_VER_15] = NULL,
+		[RTL_GIGA_MAC_VER_16] = NULL,
+		[RTL_GIGA_MAC_VER_17] = rtl8168bef_hw_phy_config,
+		[RTL_GIGA_MAC_VER_18] = rtl8168cp_1_hw_phy_config,
+		[RTL_GIGA_MAC_VER_19] = rtl8168c_1_hw_phy_config,
+		[RTL_GIGA_MAC_VER_20] = rtl8168c_2_hw_phy_config,
+		[RTL_GIGA_MAC_VER_21] = rtl8168c_3_hw_phy_config,
+		[RTL_GIGA_MAC_VER_22] = rtl8168c_3_hw_phy_config,
+		[RTL_GIGA_MAC_VER_23] = rtl8168cp_2_hw_phy_config,
+		[RTL_GIGA_MAC_VER_24] = rtl8168cp_2_hw_phy_config,
+		[RTL_GIGA_MAC_VER_25] = rtl8168d_1_hw_phy_config,
+		[RTL_GIGA_MAC_VER_26] = rtl8168d_2_hw_phy_config,
+		[RTL_GIGA_MAC_VER_27] = rtl8168d_3_hw_phy_config,
+		[RTL_GIGA_MAC_VER_28] = rtl8168d_4_hw_phy_config,
+		[RTL_GIGA_MAC_VER_29] = rtl8105e_hw_phy_config,
+		[RTL_GIGA_MAC_VER_30] = rtl8105e_hw_phy_config,
+		[RTL_GIGA_MAC_VER_31] = NULL,
+		[RTL_GIGA_MAC_VER_32] = rtl8168e_1_hw_phy_config,
+		[RTL_GIGA_MAC_VER_33] = rtl8168e_1_hw_phy_config,
+		[RTL_GIGA_MAC_VER_34] = rtl8168e_2_hw_phy_config,
+		[RTL_GIGA_MAC_VER_35] = rtl8168f_1_hw_phy_config,
+		[RTL_GIGA_MAC_VER_36] = rtl8168f_2_hw_phy_config,
+		[RTL_GIGA_MAC_VER_37] = rtl8402_hw_phy_config,
+		[RTL_GIGA_MAC_VER_38] = rtl8411_hw_phy_config,
+		[RTL_GIGA_MAC_VER_39] = rtl8106e_hw_phy_config,
+		[RTL_GIGA_MAC_VER_40] = rtl8168g_1_hw_phy_config,
+		[RTL_GIGA_MAC_VER_41] = NULL,
+		[RTL_GIGA_MAC_VER_42] = rtl8168g_2_hw_phy_config,
+		[RTL_GIGA_MAC_VER_43] = rtl8168g_2_hw_phy_config,
+		[RTL_GIGA_MAC_VER_44] = rtl8168g_2_hw_phy_config,
+		[RTL_GIGA_MAC_VER_45] = rtl8168h_1_hw_phy_config,
+		[RTL_GIGA_MAC_VER_46] = rtl8168h_2_hw_phy_config,
+		[RTL_GIGA_MAC_VER_47] = rtl8168h_1_hw_phy_config,
+		[RTL_GIGA_MAC_VER_48] = rtl8168h_2_hw_phy_config,
+		[RTL_GIGA_MAC_VER_49] = rtl8168ep_1_hw_phy_config,
+		[RTL_GIGA_MAC_VER_50] = rtl8168ep_2_hw_phy_config,
+		[RTL_GIGA_MAC_VER_51] = rtl8168ep_2_hw_phy_config,
+		[RTL_GIGA_MAC_VER_52] = rtl8117_hw_phy_config,
+		[RTL_GIGA_MAC_VER_60] = rtl8125_1_hw_phy_config,
+		[RTL_GIGA_MAC_VER_61] = rtl8125_2_hw_phy_config,
+	};
+
+	if (phy_configs[ver])
+		phy_configs[ver](tp, phydev);
+}
-- 
cgit v1.2.3-59-g8ed1b


From 468672b24fbc1c018e192dcc90e887bc9a9b2595 Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Thu, 9 Jan 2020 14:46:09 -0800
Subject: devlink: add macro for "fw.psid"

The "fw.psid" devlink info version is documented in devlink-info.rst,
and used by one driver. However, there is no associated macro for this
firmware version like there is for others. Add one now.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/devlink.h | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/include/net/devlink.h b/include/net/devlink.h
index 453f45cc1519..4e80d9acdb86 100644
--- a/include/net/devlink.h
+++ b/include/net/devlink.h
@@ -485,6 +485,8 @@ enum devlink_param_generic_id {
 #define DEVLINK_INFO_VERSION_GENERIC_FW_UNDI	"fw.undi"
 /* NCSI support/handler version */
 #define DEVLINK_INFO_VERSION_GENERIC_FW_NCSI	"fw.ncsi"
+/* FW parameter set id */
+#define DEVLINK_INFO_VERSION_GENERIC_FW_PSID	"fw.psid"
 
 struct devlink_region;
 struct devlink_info_req;
-- 
cgit v1.2.3-59-g8ed1b


From f4bdd7103652fab5ac8b0ed75fa5cbc515b50b8b Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Thu, 9 Jan 2020 14:46:10 -0800
Subject: devlink: move devlink documentation to subfolder

Combine the documentation for devlink into a subfolder, and provide an
index.rst file that can be used to generally describe devlink.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/devlink-health.txt        |  86 -------
 Documentation/networking/devlink-info-versions.rst |  64 -----
 Documentation/networking/devlink-params-bnxt.txt   |  18 --
 Documentation/networking/devlink-params-mlx5.txt   |  17 --
 Documentation/networking/devlink-params-mlxsw.txt  |  10 -
 .../networking/devlink-params-mv88e6xxx.txt        |   7 -
 Documentation/networking/devlink-params-nfp.txt    |   5 -
 .../networking/devlink-params-ti-cpsw-switch.txt   |  10 -
 Documentation/networking/devlink-params.txt        |  71 ------
 .../networking/devlink-trap-netdevsim.rst          |  20 --
 Documentation/networking/devlink-trap.rst          | 270 ---------------------
 .../networking/devlink/devlink-health.txt          |  86 +++++++
 .../networking/devlink/devlink-info-versions.rst   |  64 +++++
 .../networking/devlink/devlink-params-bnxt.txt     |  18 ++
 .../networking/devlink/devlink-params-mlx5.txt     |  17 ++
 .../networking/devlink/devlink-params-mlxsw.txt    |  10 +
 .../devlink/devlink-params-mv88e6xxx.txt           |   7 +
 .../networking/devlink/devlink-params-nfp.txt      |   5 +
 .../devlink/devlink-params-ti-cpsw-switch.txt      |  10 +
 .../networking/devlink/devlink-params.txt          |  71 ++++++
 .../networking/devlink/devlink-trap-netdevsim.rst  |  20 ++
 Documentation/networking/devlink/devlink-trap.rst  | 270 +++++++++++++++++++++
 Documentation/networking/devlink/index.rst         |  14 ++
 Documentation/networking/index.rst                 |   4 +-
 MAINTAINERS                                        |   1 +
 drivers/net/netdevsim/dev.c                        |   2 +-
 include/net/devlink.h                              |   4 +-
 27 files changed, 597 insertions(+), 584 deletions(-)
 delete mode 100644 Documentation/networking/devlink-health.txt
 delete mode 100644 Documentation/networking/devlink-info-versions.rst
 delete mode 100644 Documentation/networking/devlink-params-bnxt.txt
 delete mode 100644 Documentation/networking/devlink-params-mlx5.txt
 delete mode 100644 Documentation/networking/devlink-params-mlxsw.txt
 delete mode 100644 Documentation/networking/devlink-params-mv88e6xxx.txt
 delete mode 100644 Documentation/networking/devlink-params-nfp.txt
 delete mode 100644 Documentation/networking/devlink-params-ti-cpsw-switch.txt
 delete mode 100644 Documentation/networking/devlink-params.txt
 delete mode 100644 Documentation/networking/devlink-trap-netdevsim.rst
 delete mode 100644 Documentation/networking/devlink-trap.rst
 create mode 100644 Documentation/networking/devlink/devlink-health.txt
 create mode 100644 Documentation/networking/devlink/devlink-info-versions.rst
 create mode 100644 Documentation/networking/devlink/devlink-params-bnxt.txt
 create mode 100644 Documentation/networking/devlink/devlink-params-mlx5.txt
 create mode 100644 Documentation/networking/devlink/devlink-params-mlxsw.txt
 create mode 100644 Documentation/networking/devlink/devlink-params-mv88e6xxx.txt
 create mode 100644 Documentation/networking/devlink/devlink-params-nfp.txt
 create mode 100644 Documentation/networking/devlink/devlink-params-ti-cpsw-switch.txt
 create mode 100644 Documentation/networking/devlink/devlink-params.txt
 create mode 100644 Documentation/networking/devlink/devlink-trap-netdevsim.rst
 create mode 100644 Documentation/networking/devlink/devlink-trap.rst
 create mode 100644 Documentation/networking/devlink/index.rst

diff --git a/Documentation/networking/devlink-health.txt b/Documentation/networking/devlink-health.txt
deleted file mode 100644
index 1db3fbea0831..000000000000
--- a/Documentation/networking/devlink-health.txt
+++ /dev/null
@@ -1,86 +0,0 @@
-The health mechanism is targeted for Real Time Alerting, in order to know when
-something bad had happened to a PCI device
-- Provide alert debug information
-- Self healing
-- If problem needs vendor support, provide a way to gather all needed debugging
-  information.
-
-The main idea is to unify and centralize driver health reports in the
-generic devlink instance and allow the user to set different
-attributes of the health reporting and recovery procedures.
-
-The devlink health reporter:
-Device driver creates a "health reporter" per each error/health type.
-Error/Health type can be a known/generic (eg pci error, fw error, rx/tx error)
-or unknown (driver specific).
-For each registered health reporter a driver can issue error/health reports
-asynchronously. All health reports handling is done by devlink.
-Device driver can provide specific callbacks for each "health reporter", e.g.
- - Recovery procedures
- - Diagnostics and object dump procedures
- - OOB initial parameters
-Different parts of the driver can register different types of health reporters
-with different handlers.
-
-Once an error is reported, devlink health will do the following actions:
-  * A log is being send to the kernel trace events buffer
-  * Health status and statistics are being updated for the reporter instance
-  * Object dump is being taken and saved at the reporter instance (as long as
-    there is no other dump which is already stored)
-  * Auto recovery attempt is being done. Depends on:
-    - Auto-recovery configuration
-    - Grace period vs. time passed since last recover
-
-The user interface:
-User can access/change each reporter's parameters and driver specific callbacks
-via devlink, e.g per error type (per health reporter)
- - Configure reporter's generic parameters (like: disable/enable auto recovery)
- - Invoke recovery procedure
- - Run diagnostics
- - Object dump
-
-The devlink health interface (via netlink):
-DEVLINK_CMD_HEALTH_REPORTER_GET
-  Retrieves status and configuration info per DEV and reporter.
-DEVLINK_CMD_HEALTH_REPORTER_SET
-  Allows reporter-related configuration setting.
-DEVLINK_CMD_HEALTH_REPORTER_RECOVER
-  Triggers a reporter's recovery procedure.
-DEVLINK_CMD_HEALTH_REPORTER_DIAGNOSE
-  Retrieves diagnostics data from a reporter on a device.
-DEVLINK_CMD_HEALTH_REPORTER_DUMP_GET
-  Retrieves the last stored dump. Devlink health
-  saves a single dump. If an dump is not already stored by the devlink
-  for this reporter, devlink generates a new dump.
-  dump output is defined by the reporter.
-DEVLINK_CMD_HEALTH_REPORTER_DUMP_CLEAR
-  Clears the last saved dump file for the specified reporter.
-
-
-                                               netlink
-                                      +--------------------------+
-                                      |                          |
-                                      |            +             |
-                                      |            |             |
-                                      +--------------------------+
-                                                   |request for ops
-                                                   |(diagnose,
- mlx5_core                             devlink     |recover,
-                                                   |dump)
-+--------+                            +--------------------------+
-|        |                            |    reporter|             |
-|        |                            |  +---------v----------+  |
-|        |   ops execution            |  |                    |  |
-|     <----------------------------------+                    |  |
-|        |                            |  |                    |  |
-|        |                            |  + ^------------------+  |
-|        |                            |    | request for ops     |
-|        |                            |    | (recover, dump)     |
-|        |                            |    |                     |
-|        |                            |  +-+------------------+  |
-|        |     health report          |  | health handler     |  |
-|        +------------------------------->                    |  |
-|        |                            |  +--------------------+  |
-|        |     health reporter create |                          |
-|        +---------------------------->                          |
-+--------+                            +--------------------------+
diff --git a/Documentation/networking/devlink-info-versions.rst b/Documentation/networking/devlink-info-versions.rst
deleted file mode 100644
index 4914f581b1fd..000000000000
--- a/Documentation/networking/devlink-info-versions.rst
+++ /dev/null
@@ -1,64 +0,0 @@
-.. SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
-
-=====================
-Devlink info versions
-=====================
-
-board.id
-========
-
-Unique identifier of the board design.
-
-board.rev
-=========
-
-Board design revision.
-
-asic.id
-=======
-
-ASIC design identifier.
-
-asic.rev
-========
-
-ASIC design revision.
-
-board.manufacture
-=================
-
-An identifier of the company or the facility which produced the part.
-
-fw
-==
-
-Overall firmware version, often representing the collection of
-fw.mgmt, fw.app, etc.
-
-fw.mgmt
-=======
-
-Control unit firmware version. This firmware is responsible for house
-keeping tasks, PHY control etc. but not the packet-by-packet data path
-operation.
-
-fw.app
-======
-
-Data path microcode controlling high-speed packet processing.
-
-fw.undi
-=======
-
-UNDI software, may include the UEFI driver, firmware or both.
-
-fw.ncsi
-=======
-
-Version of the software responsible for supporting/handling the
-Network Controller Sideband Interface.
-
-fw.psid
-=======
-
-Unique identifier of the firmware parameter set.
diff --git a/Documentation/networking/devlink-params-bnxt.txt b/Documentation/networking/devlink-params-bnxt.txt
deleted file mode 100644
index 481aa303d5b4..000000000000
--- a/Documentation/networking/devlink-params-bnxt.txt
+++ /dev/null
@@ -1,18 +0,0 @@
-enable_sriov		[DEVICE, GENERIC]
-			Configuration mode: Permanent
-
-ignore_ari		[DEVICE, GENERIC]
-			Configuration mode: Permanent
-
-msix_vec_per_pf_max	[DEVICE, GENERIC]
-			Configuration mode: Permanent
-
-msix_vec_per_pf_min	[DEVICE, GENERIC]
-			Configuration mode: Permanent
-
-gre_ver_check		[DEVICE, DRIVER-SPECIFIC]
-			Generic Routing Encapsulation (GRE) version check will
-			be enabled in the device. If disabled, device skips
-			version checking for incoming packets.
-			Type: Boolean
-			Configuration mode: Permanent
diff --git a/Documentation/networking/devlink-params-mlx5.txt b/Documentation/networking/devlink-params-mlx5.txt
deleted file mode 100644
index 5071467118bd..000000000000
--- a/Documentation/networking/devlink-params-mlx5.txt
+++ /dev/null
@@ -1,17 +0,0 @@
-flow_steering_mode	[DEVICE, DRIVER-SPECIFIC]
-			Controls the flow steering mode of the driver.
-			Two modes are supported:
-			1. 'dmfs' - Device managed flow steering.
-			2. 'smfs  - Software/Driver managed flow steering.
-			In DMFS mode, the HW steering entities are created and
-			managed through the Firmware.
-			In SMFS mode, the HW steering entities are created and
-			managed though by the driver directly into Hardware
-			without firmware intervention.
-			Type: String
-			Configuration mode: runtime
-
-enable_roce		[DEVICE, GENERIC]
-			Enable handling of RoCE traffic in the device.
-			Defaultly enabled.
-			Configuration mode: driverinit
diff --git a/Documentation/networking/devlink-params-mlxsw.txt b/Documentation/networking/devlink-params-mlxsw.txt
deleted file mode 100644
index c63ea9fc7009..000000000000
--- a/Documentation/networking/devlink-params-mlxsw.txt
+++ /dev/null
@@ -1,10 +0,0 @@
-fw_load_policy		[DEVICE, GENERIC]
-			Configuration mode: driverinit
-
-acl_region_rehash_interval	[DEVICE, DRIVER-SPECIFIC]
-			Sets an interval for periodic ACL region rehashes.
-			The value is in milliseconds, minimal value is "3000".
-			Value "0" disables the periodic work.
-			The first rehash will be run right after value is set.
-			Type: u32
-			Configuration mode: runtime
diff --git a/Documentation/networking/devlink-params-mv88e6xxx.txt b/Documentation/networking/devlink-params-mv88e6xxx.txt
deleted file mode 100644
index 21c4b3556ef2..000000000000
--- a/Documentation/networking/devlink-params-mv88e6xxx.txt
+++ /dev/null
@@ -1,7 +0,0 @@
-ATU_hash		[DEVICE, DRIVER-SPECIFIC]
-			Select one of four possible hashing algorithms for
-			MAC addresses in the Address Translation Unit.
-			A value of 3 seems to work better than the default of
-			1 when many MAC addresses have the same OUI.
-			Configuration mode: runtime
-			Type: u8. 0-3 valid.
diff --git a/Documentation/networking/devlink-params-nfp.txt b/Documentation/networking/devlink-params-nfp.txt
deleted file mode 100644
index 43e4d4034865..000000000000
--- a/Documentation/networking/devlink-params-nfp.txt
+++ /dev/null
@@ -1,5 +0,0 @@
-fw_load_policy		[DEVICE, GENERIC]
-			Configuration mode: permanent
-
-reset_dev_on_drv_probe	[DEVICE, GENERIC]
-			Configuration mode: permanent
diff --git a/Documentation/networking/devlink-params-ti-cpsw-switch.txt b/Documentation/networking/devlink-params-ti-cpsw-switch.txt
deleted file mode 100644
index 4037458499f7..000000000000
--- a/Documentation/networking/devlink-params-ti-cpsw-switch.txt
+++ /dev/null
@@ -1,10 +0,0 @@
-ale_bypass	[DEVICE, DRIVER-SPECIFIC]
-		Allows to enable ALE_CONTROL(4).BYPASS mode for debug purposes.
-		All packets will be sent to the Host port only if enabled.
-		Type: bool
-		Configuration mode: runtime
-
-switch_mode	[DEVICE, DRIVER-SPECIFIC]
-		Enable switch mode
-		Type: bool
-		Configuration mode: runtime
diff --git a/Documentation/networking/devlink-params.txt b/Documentation/networking/devlink-params.txt
deleted file mode 100644
index 04e234e9acc9..000000000000
--- a/Documentation/networking/devlink-params.txt
+++ /dev/null
@@ -1,71 +0,0 @@
-Devlink configuration parameters
-================================
-Following is the list of configuration parameters via devlink interface.
-Each parameter can be generic or driver specific and are device level
-parameters.
-
-Note that the driver-specific files should contain the generic params
-they support to, with supported config modes.
-
-Each parameter can be set in different configuration modes:
-	runtime		- set while driver is running, no reset required.
-	driverinit	- applied while driver initializes, requires restart
-			driver by devlink reload command.
-	permanent	- written to device's non-volatile memory, hard reset
-			required.
-
-Following is the list of parameters:
-====================================
-enable_sriov		[DEVICE, GENERIC]
-			Enable Single Root I/O Virtualisation (SRIOV) in
-			the device.
-			Type: Boolean
-
-ignore_ari		[DEVICE, GENERIC]
-			Ignore Alternative Routing-ID Interpretation (ARI)
-			capability. If enabled, adapter will ignore ARI
-			capability even when platforms has the support
-			enabled and creates same number of partitions when
-			platform does not support ARI.
-			Type: Boolean
-
-msix_vec_per_pf_max	[DEVICE, GENERIC]
-			Provides the maximum number of MSIX interrupts that
-			a device can create. Value is same across all
-			physical functions (PFs) in the device.
-			Type: u32
-
-msix_vec_per_pf_min	[DEVICE, GENERIC]
-			Provides the minimum number of MSIX interrupts required
-			for the device initialization. Value is same across all
-			physical functions (PFs) in the device.
-			Type: u32
-
-fw_load_policy		[DEVICE, GENERIC]
-			Controls the device's firmware loading policy.
-			Valid values:
-			* DEVLINK_PARAM_FW_LOAD_POLICY_VALUE_DRIVER (0)
-			  Load firmware version preferred by the driver.
-			* DEVLINK_PARAM_FW_LOAD_POLICY_VALUE_FLASH (1)
-			  Load firmware currently stored in flash.
-			* DEVLINK_PARAM_FW_LOAD_POLICY_VALUE_DISK (2)
-			  Load firmware currently available on host's disk.
-			Type: u8
-
-reset_dev_on_drv_probe	[DEVICE, GENERIC]
-			Controls the device's reset policy on driver probe.
-			Valid values:
-			* DEVLINK_PARAM_RESET_DEV_ON_DRV_PROBE_VALUE_UNKNOWN (0)
-			  Unknown or invalid value.
-			* DEVLINK_PARAM_RESET_DEV_ON_DRV_PROBE_VALUE_ALWAYS (1)
-			  Always reset device on driver probe.
-			* DEVLINK_PARAM_RESET_DEV_ON_DRV_PROBE_VALUE_NEVER (2)
-			  Never reset device on driver probe.
-			* DEVLINK_PARAM_RESET_DEV_ON_DRV_PROBE_VALUE_DISK (3)
-			  Reset only if device firmware can be found in the
-			  filesystem.
-			Type: u8
-
-enable_roce		[DEVICE, GENERIC]
-			Enable handling of RoCE traffic in the device.
-			Type: Boolean
diff --git a/Documentation/networking/devlink-trap-netdevsim.rst b/Documentation/networking/devlink-trap-netdevsim.rst
deleted file mode 100644
index b721c9415473..000000000000
--- a/Documentation/networking/devlink-trap-netdevsim.rst
+++ /dev/null
@@ -1,20 +0,0 @@
-.. SPDX-License-Identifier: GPL-2.0
-
-======================
-Devlink Trap netdevsim
-======================
-
-Driver-specific Traps
-=====================
-
-.. list-table:: List of Driver-specific Traps Registered by ``netdevsim``
-   :widths: 5 5 90
-
-   * - Name
-     - Type
-     - Description
-   * - ``fid_miss``
-     - ``exception``
-     - When a packet enters the device it is classified to a filtering
-       indentifier (FID) based on the ingress port and VLAN. This trap is used
-       to trap packets for which a FID could not be found
diff --git a/Documentation/networking/devlink-trap.rst b/Documentation/networking/devlink-trap.rst
deleted file mode 100644
index 03311849bfb1..000000000000
--- a/Documentation/networking/devlink-trap.rst
+++ /dev/null
@@ -1,270 +0,0 @@
-.. SPDX-License-Identifier: GPL-2.0
-
-============
-Devlink Trap
-============
-
-Background
-==========
-
-Devices capable of offloading the kernel's datapath and perform functions such
-as bridging and routing must also be able to send specific packets to the
-kernel (i.e., the CPU) for processing.
-
-For example, a device acting as a multicast-aware bridge must be able to send
-IGMP membership reports to the kernel for processing by the bridge module.
-Without processing such packets, the bridge module could never populate its
-MDB.
-
-As another example, consider a device acting as router which has received an IP
-packet with a TTL of 1. Upon routing the packet the device must send it to the
-kernel so that it will route it as well and generate an ICMP Time Exceeded
-error datagram. Without letting the kernel route such packets itself, utilities
-such as ``traceroute`` could never work.
-
-The fundamental ability of sending certain packets to the kernel for processing
-is called "packet trapping".
-
-Overview
-========
-
-The ``devlink-trap`` mechanism allows capable device drivers to register their
-supported packet traps with ``devlink`` and report trapped packets to
-``devlink`` for further analysis.
-
-Upon receiving trapped packets, ``devlink`` will perform a per-trap packets and
-bytes accounting and potentially report the packet to user space via a netlink
-event along with all the provided metadata (e.g., trap reason, timestamp, input
-port). This is especially useful for drop traps (see :ref:`Trap-Types`)
-as it allows users to obtain further visibility into packet drops that would
-otherwise be invisible.
-
-The following diagram provides a general overview of ``devlink-trap``::
-
-                                    Netlink event: Packet w/ metadata
-                                                   Or a summary of recent drops
-                                  ^
-                                  |
-         Userspace                |
-        +---------------------------------------------------+
-         Kernel                   |
-                                  |
-                          +-------+--------+
-                          |                |
-                          |  drop_monitor  |
-                          |                |
-                          +-------^--------+
-                                  |
-                                  |
-                                  |
-                             +----+----+
-                             |         |      Kernel's Rx path
-                             | devlink |      (non-drop traps)
-                             |         |
-                             +----^----+      ^
-                                  |           |
-                                  +-----------+
-                                  |
-                          +-------+-------+
-                          |               |
-                          | Device driver |
-                          |               |
-                          +-------^-------+
-         Kernel                   |
-        +---------------------------------------------------+
-         Hardware                 |
-                                  | Trapped packet
-                                  |
-                               +--+---+
-                               |      |
-                               | ASIC |
-                               |      |
-                               +------+
-
-.. _Trap-Types:
-
-Trap Types
-==========
-
-The ``devlink-trap`` mechanism supports the following packet trap types:
-
-  * ``drop``: Trapped packets were dropped by the underlying device. Packets
-    are only processed by ``devlink`` and not injected to the kernel's Rx path.
-    The trap action (see :ref:`Trap-Actions`) can be changed.
-  * ``exception``: Trapped packets were not forwarded as intended by the
-    underlying device due to an exception (e.g., TTL error, missing neighbour
-    entry) and trapped to the control plane for resolution. Packets are
-    processed by ``devlink`` and injected to the kernel's Rx path. Changing the
-    action of such traps is not allowed, as it can easily break the control
-    plane.
-
-.. _Trap-Actions:
-
-Trap Actions
-============
-
-The ``devlink-trap`` mechanism supports the following packet trap actions:
-
-  * ``trap``: The sole copy of the packet is sent to the CPU.
-  * ``drop``: The packet is dropped by the underlying device and a copy is not
-    sent to the CPU.
-
-Generic Packet Traps
-====================
-
-Generic packet traps are used to describe traps that trap well-defined packets
-or packets that are trapped due to well-defined conditions (e.g., TTL error).
-Such traps can be shared by multiple device drivers and their description must
-be added to the following table:
-
-.. list-table:: List of Generic Packet Traps
-   :widths: 5 5 90
-
-   * - Name
-     - Type
-     - Description
-   * - ``source_mac_is_multicast``
-     - ``drop``
-     - Traps incoming packets that the device decided to drop because of a
-       multicast source MAC
-   * - ``vlan_tag_mismatch``
-     - ``drop``
-     - Traps incoming packets that the device decided to drop in case of VLAN
-       tag mismatch: The ingress bridge port is not configured with a PVID and
-       the packet is untagged or prio-tagged
-   * - ``ingress_vlan_filter``
-     - ``drop``
-     - Traps incoming packets that the device decided to drop in case they are
-       tagged with a VLAN that is not configured on the ingress bridge port
-   * - ``ingress_spanning_tree_filter``
-     - ``drop``
-     - Traps incoming packets that the device decided to drop in case the STP
-       state of the ingress bridge port is not "forwarding"
-   * - ``port_list_is_empty``
-     - ``drop``
-     - Traps packets that the device decided to drop in case they need to be
-       flooded (e.g., unknown unicast, unregistered multicast) and there are
-       no ports the packets should be flooded to
-   * - ``port_loopback_filter``
-     - ``drop``
-     - Traps packets that the device decided to drop in case after layer 2
-       forwarding the only port from which they should be transmitted through
-       is the port from which they were received
-   * - ``blackhole_route``
-     - ``drop``
-     - Traps packets that the device decided to drop in case they hit a
-       blackhole route
-   * - ``ttl_value_is_too_small``
-     - ``exception``
-     - Traps unicast packets that should be forwarded by the device whose TTL
-       was decremented to 0 or less
-   * - ``tail_drop``
-     - ``drop``
-     - Traps packets that the device decided to drop because they could not be
-       enqueued to a transmission queue which is full
-   * - ``non_ip``
-     - ``drop``
-     - Traps packets that the device decided to drop because they need to
-       undergo a layer 3 lookup, but are not IP or MPLS packets
-   * - ``uc_dip_over_mc_dmac``
-     - ``drop``
-     - Traps packets that the device decided to drop because they need to be
-       routed and they have a unicast destination IP and a multicast destination
-       MAC
-   * - ``dip_is_loopback_address``
-     - ``drop``
-     - Traps packets that the device decided to drop because they need to be
-       routed and their destination IP is the loopback address (i.e., 127.0.0.0/8
-       and ::1/128)
-   * - ``sip_is_mc``
-     - ``drop``
-     - Traps packets that the device decided to drop because they need to be
-       routed and their source IP is multicast (i.e., 224.0.0.0/8 and ff::/8)
-   * - ``sip_is_loopback_address``
-     - ``drop``
-     - Traps packets that the device decided to drop because they need to be
-       routed and their source IP is the loopback address (i.e., 127.0.0.0/8 and ::1/128)
-   * - ``ip_header_corrupted``
-     - ``drop``
-     - Traps packets that the device decided to drop because they need to be
-       routed and their IP header is corrupted: wrong checksum, wrong IP version
-       or too short Internet Header Length (IHL)
-   * - ``ipv4_sip_is_limited_bc``
-     - ``drop``
-     - Traps packets that the device decided to drop because they need to be
-       routed and their source IP is limited broadcast (i.e., 255.255.255.255/32)
-   * - ``ipv6_mc_dip_reserved_scope``
-     - ``drop``
-     - Traps IPv6 packets that the device decided to drop because they need to
-       be routed and their IPv6 multicast destination IP has a reserved scope
-       (i.e., ffx0::/16)
-   * - ``ipv6_mc_dip_interface_local_scope``
-     - ``drop``
-     - Traps IPv6 packets that the device decided to drop because they need to
-       be routed and their IPv6 multicast destination IP has an interface-local scope
-       (i.e., ffx1::/16)
-   * - ``mtu_value_is_too_small``
-     - ``exception``
-     - Traps packets that should have been routed by the device, but were bigger
-       than the MTU of the egress interface
-   * - ``unresolved_neigh``
-     - ``exception``
-     - Traps packets that did not have a matching IP neighbour after routing
-   * - ``mc_reverse_path_forwarding``
-     - ``exception``
-     - Traps multicast IP packets that failed reverse-path forwarding (RPF)
-       check during multicast routing
-   * - ``reject_route``
-     - ``exception``
-     - Traps packets that hit reject routes (i.e., "unreachable", "prohibit")
-   * - ``ipv4_lpm_miss``
-     - ``exception``
-     - Traps unicast IPv4 packets that did not match any route
-   * - ``ipv6_lpm_miss``
-     - ``exception``
-     - Traps unicast IPv6 packets that did not match any route
-
-Driver-specific Packet Traps
-============================
-
-Device drivers can register driver-specific packet traps, but these must be
-clearly documented. Such traps can correspond to device-specific exceptions and
-help debug packet drops caused by these exceptions. The following list includes
-links to the description of driver-specific traps registered by various device
-drivers:
-
-  * :doc:`devlink-trap-netdevsim`
-
-Generic Packet Trap Groups
-==========================
-
-Generic packet trap groups are used to aggregate logically related packet
-traps. These groups allow the user to batch operations such as setting the trap
-action of all member traps. In addition, ``devlink-trap`` can report aggregated
-per-group packets and bytes statistics, in case per-trap statistics are too
-narrow. The description of these groups must be added to the following table:
-
-.. list-table:: List of Generic Packet Trap Groups
-   :widths: 10 90
-
-   * - Name
-     - Description
-   * - ``l2_drops``
-     - Contains packet traps for packets that were dropped by the device during
-       layer 2 forwarding (i.e., bridge)
-   * - ``l3_drops``
-     - Contains packet traps for packets that were dropped by the device or hit
-       an exception (e.g., TTL error) during layer 3 forwarding
-   * - ``buffer_drops``
-     - Contains packet traps for packets that were dropped by the device due to
-       an enqueue decision
-
-Testing
-=======
-
-See ``tools/testing/selftests/drivers/net/netdevsim/devlink_trap.sh`` for a
-test covering the core infrastructure. Test cases should be added for any new
-functionality.
-
-Device drivers should focus their tests on device-specific functionality, such
-as the triggering of supported packet traps.
diff --git a/Documentation/networking/devlink/devlink-health.txt b/Documentation/networking/devlink/devlink-health.txt
new file mode 100644
index 000000000000..1db3fbea0831
--- /dev/null
+++ b/Documentation/networking/devlink/devlink-health.txt
@@ -0,0 +1,86 @@
+The health mechanism is targeted for Real Time Alerting, in order to know when
+something bad had happened to a PCI device
+- Provide alert debug information
+- Self healing
+- If problem needs vendor support, provide a way to gather all needed debugging
+  information.
+
+The main idea is to unify and centralize driver health reports in the
+generic devlink instance and allow the user to set different
+attributes of the health reporting and recovery procedures.
+
+The devlink health reporter:
+Device driver creates a "health reporter" per each error/health type.
+Error/Health type can be a known/generic (eg pci error, fw error, rx/tx error)
+or unknown (driver specific).
+For each registered health reporter a driver can issue error/health reports
+asynchronously. All health reports handling is done by devlink.
+Device driver can provide specific callbacks for each "health reporter", e.g.
+ - Recovery procedures
+ - Diagnostics and object dump procedures
+ - OOB initial parameters
+Different parts of the driver can register different types of health reporters
+with different handlers.
+
+Once an error is reported, devlink health will do the following actions:
+  * A log is being send to the kernel trace events buffer
+  * Health status and statistics are being updated for the reporter instance
+  * Object dump is being taken and saved at the reporter instance (as long as
+    there is no other dump which is already stored)
+  * Auto recovery attempt is being done. Depends on:
+    - Auto-recovery configuration
+    - Grace period vs. time passed since last recover
+
+The user interface:
+User can access/change each reporter's parameters and driver specific callbacks
+via devlink, e.g per error type (per health reporter)
+ - Configure reporter's generic parameters (like: disable/enable auto recovery)
+ - Invoke recovery procedure
+ - Run diagnostics
+ - Object dump
+
+The devlink health interface (via netlink):
+DEVLINK_CMD_HEALTH_REPORTER_GET
+  Retrieves status and configuration info per DEV and reporter.
+DEVLINK_CMD_HEALTH_REPORTER_SET
+  Allows reporter-related configuration setting.
+DEVLINK_CMD_HEALTH_REPORTER_RECOVER
+  Triggers a reporter's recovery procedure.
+DEVLINK_CMD_HEALTH_REPORTER_DIAGNOSE
+  Retrieves diagnostics data from a reporter on a device.
+DEVLINK_CMD_HEALTH_REPORTER_DUMP_GET
+  Retrieves the last stored dump. Devlink health
+  saves a single dump. If an dump is not already stored by the devlink
+  for this reporter, devlink generates a new dump.
+  dump output is defined by the reporter.
+DEVLINK_CMD_HEALTH_REPORTER_DUMP_CLEAR
+  Clears the last saved dump file for the specified reporter.
+
+
+                                               netlink
+                                      +--------------------------+
+                                      |                          |
+                                      |            +             |
+                                      |            |             |
+                                      +--------------------------+
+                                                   |request for ops
+                                                   |(diagnose,
+ mlx5_core                             devlink     |recover,
+                                                   |dump)
++--------+                            +--------------------------+
+|        |                            |    reporter|             |
+|        |                            |  +---------v----------+  |
+|        |   ops execution            |  |                    |  |
+|     <----------------------------------+                    |  |
+|        |                            |  |                    |  |
+|        |                            |  + ^------------------+  |
+|        |                            |    | request for ops     |
+|        |                            |    | (recover, dump)     |
+|        |                            |    |                     |
+|        |                            |  +-+------------------+  |
+|        |     health report          |  | health handler     |  |
+|        +------------------------------->                    |  |
+|        |                            |  +--------------------+  |
+|        |     health reporter create |                          |
+|        +---------------------------->                          |
++--------+                            +--------------------------+
diff --git a/Documentation/networking/devlink/devlink-info-versions.rst b/Documentation/networking/devlink/devlink-info-versions.rst
new file mode 100644
index 000000000000..4914f581b1fd
--- /dev/null
+++ b/Documentation/networking/devlink/devlink-info-versions.rst
@@ -0,0 +1,64 @@
+.. SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+
+=====================
+Devlink info versions
+=====================
+
+board.id
+========
+
+Unique identifier of the board design.
+
+board.rev
+=========
+
+Board design revision.
+
+asic.id
+=======
+
+ASIC design identifier.
+
+asic.rev
+========
+
+ASIC design revision.
+
+board.manufacture
+=================
+
+An identifier of the company or the facility which produced the part.
+
+fw
+==
+
+Overall firmware version, often representing the collection of
+fw.mgmt, fw.app, etc.
+
+fw.mgmt
+=======
+
+Control unit firmware version. This firmware is responsible for house
+keeping tasks, PHY control etc. but not the packet-by-packet data path
+operation.
+
+fw.app
+======
+
+Data path microcode controlling high-speed packet processing.
+
+fw.undi
+=======
+
+UNDI software, may include the UEFI driver, firmware or both.
+
+fw.ncsi
+=======
+
+Version of the software responsible for supporting/handling the
+Network Controller Sideband Interface.
+
+fw.psid
+=======
+
+Unique identifier of the firmware parameter set.
diff --git a/Documentation/networking/devlink/devlink-params-bnxt.txt b/Documentation/networking/devlink/devlink-params-bnxt.txt
new file mode 100644
index 000000000000..481aa303d5b4
--- /dev/null
+++ b/Documentation/networking/devlink/devlink-params-bnxt.txt
@@ -0,0 +1,18 @@
+enable_sriov		[DEVICE, GENERIC]
+			Configuration mode: Permanent
+
+ignore_ari		[DEVICE, GENERIC]
+			Configuration mode: Permanent
+
+msix_vec_per_pf_max	[DEVICE, GENERIC]
+			Configuration mode: Permanent
+
+msix_vec_per_pf_min	[DEVICE, GENERIC]
+			Configuration mode: Permanent
+
+gre_ver_check		[DEVICE, DRIVER-SPECIFIC]
+			Generic Routing Encapsulation (GRE) version check will
+			be enabled in the device. If disabled, device skips
+			version checking for incoming packets.
+			Type: Boolean
+			Configuration mode: Permanent
diff --git a/Documentation/networking/devlink/devlink-params-mlx5.txt b/Documentation/networking/devlink/devlink-params-mlx5.txt
new file mode 100644
index 000000000000..5071467118bd
--- /dev/null
+++ b/Documentation/networking/devlink/devlink-params-mlx5.txt
@@ -0,0 +1,17 @@
+flow_steering_mode	[DEVICE, DRIVER-SPECIFIC]
+			Controls the flow steering mode of the driver.
+			Two modes are supported:
+			1. 'dmfs' - Device managed flow steering.
+			2. 'smfs  - Software/Driver managed flow steering.
+			In DMFS mode, the HW steering entities are created and
+			managed through the Firmware.
+			In SMFS mode, the HW steering entities are created and
+			managed though by the driver directly into Hardware
+			without firmware intervention.
+			Type: String
+			Configuration mode: runtime
+
+enable_roce		[DEVICE, GENERIC]
+			Enable handling of RoCE traffic in the device.
+			Defaultly enabled.
+			Configuration mode: driverinit
diff --git a/Documentation/networking/devlink/devlink-params-mlxsw.txt b/Documentation/networking/devlink/devlink-params-mlxsw.txt
new file mode 100644
index 000000000000..c63ea9fc7009
--- /dev/null
+++ b/Documentation/networking/devlink/devlink-params-mlxsw.txt
@@ -0,0 +1,10 @@
+fw_load_policy		[DEVICE, GENERIC]
+			Configuration mode: driverinit
+
+acl_region_rehash_interval	[DEVICE, DRIVER-SPECIFIC]
+			Sets an interval for periodic ACL region rehashes.
+			The value is in milliseconds, minimal value is "3000".
+			Value "0" disables the periodic work.
+			The first rehash will be run right after value is set.
+			Type: u32
+			Configuration mode: runtime
diff --git a/Documentation/networking/devlink/devlink-params-mv88e6xxx.txt b/Documentation/networking/devlink/devlink-params-mv88e6xxx.txt
new file mode 100644
index 000000000000..21c4b3556ef2
--- /dev/null
+++ b/Documentation/networking/devlink/devlink-params-mv88e6xxx.txt
@@ -0,0 +1,7 @@
+ATU_hash		[DEVICE, DRIVER-SPECIFIC]
+			Select one of four possible hashing algorithms for
+			MAC addresses in the Address Translation Unit.
+			A value of 3 seems to work better than the default of
+			1 when many MAC addresses have the same OUI.
+			Configuration mode: runtime
+			Type: u8. 0-3 valid.
diff --git a/Documentation/networking/devlink/devlink-params-nfp.txt b/Documentation/networking/devlink/devlink-params-nfp.txt
new file mode 100644
index 000000000000..43e4d4034865
--- /dev/null
+++ b/Documentation/networking/devlink/devlink-params-nfp.txt
@@ -0,0 +1,5 @@
+fw_load_policy		[DEVICE, GENERIC]
+			Configuration mode: permanent
+
+reset_dev_on_drv_probe	[DEVICE, GENERIC]
+			Configuration mode: permanent
diff --git a/Documentation/networking/devlink/devlink-params-ti-cpsw-switch.txt b/Documentation/networking/devlink/devlink-params-ti-cpsw-switch.txt
new file mode 100644
index 000000000000..4037458499f7
--- /dev/null
+++ b/Documentation/networking/devlink/devlink-params-ti-cpsw-switch.txt
@@ -0,0 +1,10 @@
+ale_bypass	[DEVICE, DRIVER-SPECIFIC]
+		Allows to enable ALE_CONTROL(4).BYPASS mode for debug purposes.
+		All packets will be sent to the Host port only if enabled.
+		Type: bool
+		Configuration mode: runtime
+
+switch_mode	[DEVICE, DRIVER-SPECIFIC]
+		Enable switch mode
+		Type: bool
+		Configuration mode: runtime
diff --git a/Documentation/networking/devlink/devlink-params.txt b/Documentation/networking/devlink/devlink-params.txt
new file mode 100644
index 000000000000..04e234e9acc9
--- /dev/null
+++ b/Documentation/networking/devlink/devlink-params.txt
@@ -0,0 +1,71 @@
+Devlink configuration parameters
+================================
+Following is the list of configuration parameters via devlink interface.
+Each parameter can be generic or driver specific and are device level
+parameters.
+
+Note that the driver-specific files should contain the generic params
+they support to, with supported config modes.
+
+Each parameter can be set in different configuration modes:
+	runtime		- set while driver is running, no reset required.
+	driverinit	- applied while driver initializes, requires restart
+			driver by devlink reload command.
+	permanent	- written to device's non-volatile memory, hard reset
+			required.
+
+Following is the list of parameters:
+====================================
+enable_sriov		[DEVICE, GENERIC]
+			Enable Single Root I/O Virtualisation (SRIOV) in
+			the device.
+			Type: Boolean
+
+ignore_ari		[DEVICE, GENERIC]
+			Ignore Alternative Routing-ID Interpretation (ARI)
+			capability. If enabled, adapter will ignore ARI
+			capability even when platforms has the support
+			enabled and creates same number of partitions when
+			platform does not support ARI.
+			Type: Boolean
+
+msix_vec_per_pf_max	[DEVICE, GENERIC]
+			Provides the maximum number of MSIX interrupts that
+			a device can create. Value is same across all
+			physical functions (PFs) in the device.
+			Type: u32
+
+msix_vec_per_pf_min	[DEVICE, GENERIC]
+			Provides the minimum number of MSIX interrupts required
+			for the device initialization. Value is same across all
+			physical functions (PFs) in the device.
+			Type: u32
+
+fw_load_policy		[DEVICE, GENERIC]
+			Controls the device's firmware loading policy.
+			Valid values:
+			* DEVLINK_PARAM_FW_LOAD_POLICY_VALUE_DRIVER (0)
+			  Load firmware version preferred by the driver.
+			* DEVLINK_PARAM_FW_LOAD_POLICY_VALUE_FLASH (1)
+			  Load firmware currently stored in flash.
+			* DEVLINK_PARAM_FW_LOAD_POLICY_VALUE_DISK (2)
+			  Load firmware currently available on host's disk.
+			Type: u8
+
+reset_dev_on_drv_probe	[DEVICE, GENERIC]
+			Controls the device's reset policy on driver probe.
+			Valid values:
+			* DEVLINK_PARAM_RESET_DEV_ON_DRV_PROBE_VALUE_UNKNOWN (0)
+			  Unknown or invalid value.
+			* DEVLINK_PARAM_RESET_DEV_ON_DRV_PROBE_VALUE_ALWAYS (1)
+			  Always reset device on driver probe.
+			* DEVLINK_PARAM_RESET_DEV_ON_DRV_PROBE_VALUE_NEVER (2)
+			  Never reset device on driver probe.
+			* DEVLINK_PARAM_RESET_DEV_ON_DRV_PROBE_VALUE_DISK (3)
+			  Reset only if device firmware can be found in the
+			  filesystem.
+			Type: u8
+
+enable_roce		[DEVICE, GENERIC]
+			Enable handling of RoCE traffic in the device.
+			Type: Boolean
diff --git a/Documentation/networking/devlink/devlink-trap-netdevsim.rst b/Documentation/networking/devlink/devlink-trap-netdevsim.rst
new file mode 100644
index 000000000000..b721c9415473
--- /dev/null
+++ b/Documentation/networking/devlink/devlink-trap-netdevsim.rst
@@ -0,0 +1,20 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+======================
+Devlink Trap netdevsim
+======================
+
+Driver-specific Traps
+=====================
+
+.. list-table:: List of Driver-specific Traps Registered by ``netdevsim``
+   :widths: 5 5 90
+
+   * - Name
+     - Type
+     - Description
+   * - ``fid_miss``
+     - ``exception``
+     - When a packet enters the device it is classified to a filtering
+       indentifier (FID) based on the ingress port and VLAN. This trap is used
+       to trap packets for which a FID could not be found
diff --git a/Documentation/networking/devlink/devlink-trap.rst b/Documentation/networking/devlink/devlink-trap.rst
new file mode 100644
index 000000000000..03311849bfb1
--- /dev/null
+++ b/Documentation/networking/devlink/devlink-trap.rst
@@ -0,0 +1,270 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+============
+Devlink Trap
+============
+
+Background
+==========
+
+Devices capable of offloading the kernel's datapath and perform functions such
+as bridging and routing must also be able to send specific packets to the
+kernel (i.e., the CPU) for processing.
+
+For example, a device acting as a multicast-aware bridge must be able to send
+IGMP membership reports to the kernel for processing by the bridge module.
+Without processing such packets, the bridge module could never populate its
+MDB.
+
+As another example, consider a device acting as router which has received an IP
+packet with a TTL of 1. Upon routing the packet the device must send it to the
+kernel so that it will route it as well and generate an ICMP Time Exceeded
+error datagram. Without letting the kernel route such packets itself, utilities
+such as ``traceroute`` could never work.
+
+The fundamental ability of sending certain packets to the kernel for processing
+is called "packet trapping".
+
+Overview
+========
+
+The ``devlink-trap`` mechanism allows capable device drivers to register their
+supported packet traps with ``devlink`` and report trapped packets to
+``devlink`` for further analysis.
+
+Upon receiving trapped packets, ``devlink`` will perform a per-trap packets and
+bytes accounting and potentially report the packet to user space via a netlink
+event along with all the provided metadata (e.g., trap reason, timestamp, input
+port). This is especially useful for drop traps (see :ref:`Trap-Types`)
+as it allows users to obtain further visibility into packet drops that would
+otherwise be invisible.
+
+The following diagram provides a general overview of ``devlink-trap``::
+
+                                    Netlink event: Packet w/ metadata
+                                                   Or a summary of recent drops
+                                  ^
+                                  |
+         Userspace                |
+        +---------------------------------------------------+
+         Kernel                   |
+                                  |
+                          +-------+--------+
+                          |                |
+                          |  drop_monitor  |
+                          |                |
+                          +-------^--------+
+                                  |
+                                  |
+                                  |
+                             +----+----+
+                             |         |      Kernel's Rx path
+                             | devlink |      (non-drop traps)
+                             |         |
+                             +----^----+      ^
+                                  |           |
+                                  +-----------+
+                                  |
+                          +-------+-------+
+                          |               |
+                          | Device driver |
+                          |               |
+                          +-------^-------+
+         Kernel                   |
+        +---------------------------------------------------+
+         Hardware                 |
+                                  | Trapped packet
+                                  |
+                               +--+---+
+                               |      |
+                               | ASIC |
+                               |      |
+                               +------+
+
+.. _Trap-Types:
+
+Trap Types
+==========
+
+The ``devlink-trap`` mechanism supports the following packet trap types:
+
+  * ``drop``: Trapped packets were dropped by the underlying device. Packets
+    are only processed by ``devlink`` and not injected to the kernel's Rx path.
+    The trap action (see :ref:`Trap-Actions`) can be changed.
+  * ``exception``: Trapped packets were not forwarded as intended by the
+    underlying device due to an exception (e.g., TTL error, missing neighbour
+    entry) and trapped to the control plane for resolution. Packets are
+    processed by ``devlink`` and injected to the kernel's Rx path. Changing the
+    action of such traps is not allowed, as it can easily break the control
+    plane.
+
+.. _Trap-Actions:
+
+Trap Actions
+============
+
+The ``devlink-trap`` mechanism supports the following packet trap actions:
+
+  * ``trap``: The sole copy of the packet is sent to the CPU.
+  * ``drop``: The packet is dropped by the underlying device and a copy is not
+    sent to the CPU.
+
+Generic Packet Traps
+====================
+
+Generic packet traps are used to describe traps that trap well-defined packets
+or packets that are trapped due to well-defined conditions (e.g., TTL error).
+Such traps can be shared by multiple device drivers and their description must
+be added to the following table:
+
+.. list-table:: List of Generic Packet Traps
+   :widths: 5 5 90
+
+   * - Name
+     - Type
+     - Description
+   * - ``source_mac_is_multicast``
+     - ``drop``
+     - Traps incoming packets that the device decided to drop because of a
+       multicast source MAC
+   * - ``vlan_tag_mismatch``
+     - ``drop``
+     - Traps incoming packets that the device decided to drop in case of VLAN
+       tag mismatch: The ingress bridge port is not configured with a PVID and
+       the packet is untagged or prio-tagged
+   * - ``ingress_vlan_filter``
+     - ``drop``
+     - Traps incoming packets that the device decided to drop in case they are
+       tagged with a VLAN that is not configured on the ingress bridge port
+   * - ``ingress_spanning_tree_filter``
+     - ``drop``
+     - Traps incoming packets that the device decided to drop in case the STP
+       state of the ingress bridge port is not "forwarding"
+   * - ``port_list_is_empty``
+     - ``drop``
+     - Traps packets that the device decided to drop in case they need to be
+       flooded (e.g., unknown unicast, unregistered multicast) and there are
+       no ports the packets should be flooded to
+   * - ``port_loopback_filter``
+     - ``drop``
+     - Traps packets that the device decided to drop in case after layer 2
+       forwarding the only port from which they should be transmitted through
+       is the port from which they were received
+   * - ``blackhole_route``
+     - ``drop``
+     - Traps packets that the device decided to drop in case they hit a
+       blackhole route
+   * - ``ttl_value_is_too_small``
+     - ``exception``
+     - Traps unicast packets that should be forwarded by the device whose TTL
+       was decremented to 0 or less
+   * - ``tail_drop``
+     - ``drop``
+     - Traps packets that the device decided to drop because they could not be
+       enqueued to a transmission queue which is full
+   * - ``non_ip``
+     - ``drop``
+     - Traps packets that the device decided to drop because they need to
+       undergo a layer 3 lookup, but are not IP or MPLS packets
+   * - ``uc_dip_over_mc_dmac``
+     - ``drop``
+     - Traps packets that the device decided to drop because they need to be
+       routed and they have a unicast destination IP and a multicast destination
+       MAC
+   * - ``dip_is_loopback_address``
+     - ``drop``
+     - Traps packets that the device decided to drop because they need to be
+       routed and their destination IP is the loopback address (i.e., 127.0.0.0/8
+       and ::1/128)
+   * - ``sip_is_mc``
+     - ``drop``
+     - Traps packets that the device decided to drop because they need to be
+       routed and their source IP is multicast (i.e., 224.0.0.0/8 and ff::/8)
+   * - ``sip_is_loopback_address``
+     - ``drop``
+     - Traps packets that the device decided to drop because they need to be
+       routed and their source IP is the loopback address (i.e., 127.0.0.0/8 and ::1/128)
+   * - ``ip_header_corrupted``
+     - ``drop``
+     - Traps packets that the device decided to drop because they need to be
+       routed and their IP header is corrupted: wrong checksum, wrong IP version
+       or too short Internet Header Length (IHL)
+   * - ``ipv4_sip_is_limited_bc``
+     - ``drop``
+     - Traps packets that the device decided to drop because they need to be
+       routed and their source IP is limited broadcast (i.e., 255.255.255.255/32)
+   * - ``ipv6_mc_dip_reserved_scope``
+     - ``drop``
+     - Traps IPv6 packets that the device decided to drop because they need to
+       be routed and their IPv6 multicast destination IP has a reserved scope
+       (i.e., ffx0::/16)
+   * - ``ipv6_mc_dip_interface_local_scope``
+     - ``drop``
+     - Traps IPv6 packets that the device decided to drop because they need to
+       be routed and their IPv6 multicast destination IP has an interface-local scope
+       (i.e., ffx1::/16)
+   * - ``mtu_value_is_too_small``
+     - ``exception``
+     - Traps packets that should have been routed by the device, but were bigger
+       than the MTU of the egress interface
+   * - ``unresolved_neigh``
+     - ``exception``
+     - Traps packets that did not have a matching IP neighbour after routing
+   * - ``mc_reverse_path_forwarding``
+     - ``exception``
+     - Traps multicast IP packets that failed reverse-path forwarding (RPF)
+       check during multicast routing
+   * - ``reject_route``
+     - ``exception``
+     - Traps packets that hit reject routes (i.e., "unreachable", "prohibit")
+   * - ``ipv4_lpm_miss``
+     - ``exception``
+     - Traps unicast IPv4 packets that did not match any route
+   * - ``ipv6_lpm_miss``
+     - ``exception``
+     - Traps unicast IPv6 packets that did not match any route
+
+Driver-specific Packet Traps
+============================
+
+Device drivers can register driver-specific packet traps, but these must be
+clearly documented. Such traps can correspond to device-specific exceptions and
+help debug packet drops caused by these exceptions. The following list includes
+links to the description of driver-specific traps registered by various device
+drivers:
+
+  * :doc:`devlink-trap-netdevsim`
+
+Generic Packet Trap Groups
+==========================
+
+Generic packet trap groups are used to aggregate logically related packet
+traps. These groups allow the user to batch operations such as setting the trap
+action of all member traps. In addition, ``devlink-trap`` can report aggregated
+per-group packets and bytes statistics, in case per-trap statistics are too
+narrow. The description of these groups must be added to the following table:
+
+.. list-table:: List of Generic Packet Trap Groups
+   :widths: 10 90
+
+   * - Name
+     - Description
+   * - ``l2_drops``
+     - Contains packet traps for packets that were dropped by the device during
+       layer 2 forwarding (i.e., bridge)
+   * - ``l3_drops``
+     - Contains packet traps for packets that were dropped by the device or hit
+       an exception (e.g., TTL error) during layer 3 forwarding
+   * - ``buffer_drops``
+     - Contains packet traps for packets that were dropped by the device due to
+       an enqueue decision
+
+Testing
+=======
+
+See ``tools/testing/selftests/drivers/net/netdevsim/devlink_trap.sh`` for a
+test covering the core infrastructure. Test cases should be added for any new
+functionality.
+
+Device drivers should focus their tests on device-specific functionality, such
+as the triggering of supported packet traps.
diff --git a/Documentation/networking/devlink/index.rst b/Documentation/networking/devlink/index.rst
new file mode 100644
index 000000000000..1252c2a1b680
--- /dev/null
+++ b/Documentation/networking/devlink/index.rst
@@ -0,0 +1,14 @@
+Linux Devlink Documentation
+===========================
+
+devlink is an API to expose device information and resources not directly
+related to any device class, such as chip-wide/switch-ASIC-wide configuration.
+
+Contents:
+
+.. toctree::
+   :maxdepth: 1
+
+   devlink-info-versions
+   devlink-trap
+   devlink-trap-netdevsim
diff --git a/Documentation/networking/index.rst b/Documentation/networking/index.rst
index bee73be7af93..d07d9855dcd3 100644
--- a/Documentation/networking/index.rst
+++ b/Documentation/networking/index.rst
@@ -13,9 +13,7 @@ Contents:
    can_ucan_protocol
    device_drivers/index
    dsa/index
-   devlink-info-versions
-   devlink-trap
-   devlink-trap-netdevsim
+   devlink/index
    ethtool-netlink
    ieee802154
    j1939
diff --git a/MAINTAINERS b/MAINTAINERS
index 9dcb9cab5705..d0ea00e3e8b1 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -4848,6 +4848,7 @@ S:	Supported
 F:	net/core/devlink.c
 F:	include/net/devlink.h
 F:	include/uapi/linux/devlink.h
+F:	Documentation/networking/devlink
 
 DIALOG SEMICONDUCTOR DRIVERS
 M:	Support Opensource <support.opensource@diasemi.com>
diff --git a/drivers/net/netdevsim/dev.c b/drivers/net/netdevsim/dev.c
index 059711edfc61..6663f79fe5d1 100644
--- a/drivers/net/netdevsim/dev.c
+++ b/drivers/net/netdevsim/dev.c
@@ -270,7 +270,7 @@ struct nsim_trap_data {
 };
 
 /* All driver-specific traps must be documented in
- * Documentation/networking/devlink-trap-netdevsim.rst
+ * Documentation/networking/devlink/devlink-trap-netdevsim.rst
  */
 enum {
 	NSIM_TRAP_ID_BASE = DEVLINK_TRAP_GENERIC_ID_MAX,
diff --git a/include/net/devlink.h b/include/net/devlink.h
index 4e80d9acdb86..a6856f1d5d1f 100644
--- a/include/net/devlink.h
+++ b/include/net/devlink.h
@@ -564,7 +564,7 @@ struct devlink_trap {
 };
 
 /* All traps must be documented in
- * Documentation/networking/devlink-trap.rst
+ * Documentation/networking/devlink/devlink-trap.rst
  */
 enum devlink_trap_generic_id {
 	DEVLINK_TRAP_GENERIC_ID_SMAC_MC,
@@ -598,7 +598,7 @@ enum devlink_trap_generic_id {
 };
 
 /* All trap groups must be documented in
- * Documentation/networking/devlink-trap.rst
+ * Documentation/networking/devlink/devlink-trap.rst
  */
 enum devlink_trap_group_generic_id {
 	DEVLINK_TRAP_GROUP_GENERIC_ID_L2_DROPS,
-- 
cgit v1.2.3-59-g8ed1b


From f7555fd19903cb64ad9887875a47094bdb0ab0c2 Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Thu, 9 Jan 2020 14:46:11 -0800
Subject: devlink: convert devlink-health.txt to rst format

Update the devlink-health documentation to use the newer
ReStructuredText format.

Note that it's unclear what OOB stood for, and it has been left as-is
without a proper first-use expansion of the acronym.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../networking/devlink/devlink-health.rst          | 114 +++++++++++++++++++++
 .../networking/devlink/devlink-health.txt          |  86 ----------------
 Documentation/networking/devlink/index.rst         |   1 +
 3 files changed, 115 insertions(+), 86 deletions(-)
 create mode 100644 Documentation/networking/devlink/devlink-health.rst
 delete mode 100644 Documentation/networking/devlink/devlink-health.txt

diff --git a/Documentation/networking/devlink/devlink-health.rst b/Documentation/networking/devlink/devlink-health.rst
new file mode 100644
index 000000000000..0c99b11f05f9
--- /dev/null
+++ b/Documentation/networking/devlink/devlink-health.rst
@@ -0,0 +1,114 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+==============
+Devlink Health
+==============
+
+Background
+==========
+
+The ``devlink`` health mechanism is targeted for Real Time Alerting, in
+order to know when something bad happened to a PCI device.
+
+  * Provide alert debug information.
+  * Self healing.
+  * If problem needs vendor support, provide a way to gather all needed
+    debugging information.
+
+Overview
+========
+
+The main idea is to unify and centralize driver health reports in the
+generic ``devlink`` instance and allow the user to set different
+attributes of the health reporting and recovery procedures.
+
+The ``devlink`` health reporter:
+Device driver creates a "health reporter" per each error/health type.
+Error/Health type can be a known/generic (eg pci error, fw error, rx/tx error)
+or unknown (driver specific).
+For each registered health reporter a driver can issue error/health reports
+asynchronously. All health reports handling is done by ``devlink``.
+Device driver can provide specific callbacks for each "health reporter", e.g.:
+
+  * Recovery procedures
+  * Diagnostics procedures
+  * Object dump procedures
+  * OOB initial parameters
+
+Different parts of the driver can register different types of health reporters
+with different handlers.
+
+Actions
+=======
+
+Once an error is reported, devlink health will perform the following actions:
+
+  * A log is being send to the kernel trace events buffer
+  * Health status and statistics are being updated for the reporter instance
+  * Object dump is being taken and saved at the reporter instance (as long as
+    there is no other dump which is already stored)
+  * Auto recovery attempt is being done. Depends on:
+    - Auto-recovery configuration
+    - Grace period vs. time passed since last recover
+
+User Interface
+==============
+
+User can access/change each reporter's parameters and driver specific callbacks
+via ``devlink``, e.g per error type (per health reporter):
+
+  * Configure reporter's generic parameters (like: disable/enable auto recovery)
+  * Invoke recovery procedure
+  * Run diagnostics
+  * Object dump
+
+.. list-table:: List of devlink health interfaces
+   :widths: 10 90
+
+   * - Name
+     - Description
+   * - ``DEVLINK_CMD_HEALTH_REPORTER_GET``
+     - Retrieves status and configuration info per DEV and reporter.
+   * - ``DEVLINK_CMD_HEALTH_REPORTER_SET``
+     - Allows reporter-related configuration setting.
+   * - ``DEVLINK_CMD_HEALTH_REPORTER_RECOVER``
+     - Triggers a reporter's recovery procedure.
+   * - ``DEVLINK_CMD_HEALTH_REPORTER_DIAGNOSE``
+     - Retrieves diagnostics data from a reporter on a device.
+   * - ``DEVLINK_CMD_HEALTH_REPORTER_DUMP_GET``
+     - Retrieves the last stored dump. Devlink health
+       saves a single dump. If an dump is not already stored by the devlink
+       for this reporter, devlink generates a new dump.
+       dump output is defined by the reporter.
+   * - ``DEVLINK_CMD_HEALTH_REPORTER_DUMP_CLEAR``
+     - Clears the last saved dump file for the specified reporter.
+
+The following diagram provides a general overview of ``devlink-health``::
+
+                                                   netlink
+                                          +--------------------------+
+                                          |                          |
+                                          |            +             |
+                                          |            |             |
+                                          +--------------------------+
+                                                       |request for ops
+                                                       |(diagnose,
+     mlx5_core                             devlink     |recover,
+                                                       |dump)
+    +--------+                            +--------------------------+
+    |        |                            |    reporter|             |
+    |        |                            |  +---------v----------+  |
+    |        |   ops execution            |  |                    |  |
+    |     <----------------------------------+                    |  |
+    |        |                            |  |                    |  |
+    |        |                            |  + ^------------------+  |
+    |        |                            |    | request for ops     |
+    |        |                            |    | (recover, dump)     |
+    |        |                            |    |                     |
+    |        |                            |  +-+------------------+  |
+    |        |     health report          |  | health handler     |  |
+    |        +------------------------------->                    |  |
+    |        |                            |  +--------------------+  |
+    |        |     health reporter create |                          |
+    |        +---------------------------->                          |
+    +--------+                            +--------------------------+
diff --git a/Documentation/networking/devlink/devlink-health.txt b/Documentation/networking/devlink/devlink-health.txt
deleted file mode 100644
index 1db3fbea0831..000000000000
--- a/Documentation/networking/devlink/devlink-health.txt
+++ /dev/null
@@ -1,86 +0,0 @@
-The health mechanism is targeted for Real Time Alerting, in order to know when
-something bad had happened to a PCI device
-- Provide alert debug information
-- Self healing
-- If problem needs vendor support, provide a way to gather all needed debugging
-  information.
-
-The main idea is to unify and centralize driver health reports in the
-generic devlink instance and allow the user to set different
-attributes of the health reporting and recovery procedures.
-
-The devlink health reporter:
-Device driver creates a "health reporter" per each error/health type.
-Error/Health type can be a known/generic (eg pci error, fw error, rx/tx error)
-or unknown (driver specific).
-For each registered health reporter a driver can issue error/health reports
-asynchronously. All health reports handling is done by devlink.
-Device driver can provide specific callbacks for each "health reporter", e.g.
- - Recovery procedures
- - Diagnostics and object dump procedures
- - OOB initial parameters
-Different parts of the driver can register different types of health reporters
-with different handlers.
-
-Once an error is reported, devlink health will do the following actions:
-  * A log is being send to the kernel trace events buffer
-  * Health status and statistics are being updated for the reporter instance
-  * Object dump is being taken and saved at the reporter instance (as long as
-    there is no other dump which is already stored)
-  * Auto recovery attempt is being done. Depends on:
-    - Auto-recovery configuration
-    - Grace period vs. time passed since last recover
-
-The user interface:
-User can access/change each reporter's parameters and driver specific callbacks
-via devlink, e.g per error type (per health reporter)
- - Configure reporter's generic parameters (like: disable/enable auto recovery)
- - Invoke recovery procedure
- - Run diagnostics
- - Object dump
-
-The devlink health interface (via netlink):
-DEVLINK_CMD_HEALTH_REPORTER_GET
-  Retrieves status and configuration info per DEV and reporter.
-DEVLINK_CMD_HEALTH_REPORTER_SET
-  Allows reporter-related configuration setting.
-DEVLINK_CMD_HEALTH_REPORTER_RECOVER
-  Triggers a reporter's recovery procedure.
-DEVLINK_CMD_HEALTH_REPORTER_DIAGNOSE
-  Retrieves diagnostics data from a reporter on a device.
-DEVLINK_CMD_HEALTH_REPORTER_DUMP_GET
-  Retrieves the last stored dump. Devlink health
-  saves a single dump. If an dump is not already stored by the devlink
-  for this reporter, devlink generates a new dump.
-  dump output is defined by the reporter.
-DEVLINK_CMD_HEALTH_REPORTER_DUMP_CLEAR
-  Clears the last saved dump file for the specified reporter.
-
-
-                                               netlink
-                                      +--------------------------+
-                                      |                          |
-                                      |            +             |
-                                      |            |             |
-                                      +--------------------------+
-                                                   |request for ops
-                                                   |(diagnose,
- mlx5_core                             devlink     |recover,
-                                                   |dump)
-+--------+                            +--------------------------+
-|        |                            |    reporter|             |
-|        |                            |  +---------v----------+  |
-|        |   ops execution            |  |                    |  |
-|     <----------------------------------+                    |  |
-|        |                            |  |                    |  |
-|        |                            |  + ^------------------+  |
-|        |                            |    | request for ops     |
-|        |                            |    | (recover, dump)     |
-|        |                            |    |                     |
-|        |                            |  +-+------------------+  |
-|        |     health report          |  | health handler     |  |
-|        +------------------------------->                    |  |
-|        |                            |  +--------------------+  |
-|        |     health reporter create |                          |
-|        +---------------------------->                          |
-+--------+                            +--------------------------+
diff --git a/Documentation/networking/devlink/index.rst b/Documentation/networking/devlink/index.rst
index 1252c2a1b680..71d9d0e859eb 100644
--- a/Documentation/networking/devlink/index.rst
+++ b/Documentation/networking/devlink/index.rst
@@ -9,6 +9,7 @@ Contents:
 .. toctree::
    :maxdepth: 1
 
+   devlink-health
    devlink-info-versions
    devlink-trap
    devlink-trap-netdevsim
-- 
cgit v1.2.3-59-g8ed1b


From 2dfaaae521d0af16da287b0c22683fd2a31ace81 Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Thu, 9 Jan 2020 14:46:12 -0800
Subject: devlink: rename devlink-info-versions.rst and add a header

Rename the devlink-info-versions.rst file to a plain devlink-info.rst
file. Add additional paragraphs explaining what devlink-info is for,
and the expectation that drivers use the generic names where plausible.

Note that drivers which use non-standard info version names ought to
document these in a driver-specific info-versions.rst file.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../networking/devlink/devlink-info-versions.rst   | 64 ---------------
 Documentation/networking/devlink/devlink-info.rst  | 94 ++++++++++++++++++++++
 Documentation/networking/devlink/index.rst         |  2 +-
 3 files changed, 95 insertions(+), 65 deletions(-)
 delete mode 100644 Documentation/networking/devlink/devlink-info-versions.rst
 create mode 100644 Documentation/networking/devlink/devlink-info.rst

diff --git a/Documentation/networking/devlink/devlink-info-versions.rst b/Documentation/networking/devlink/devlink-info-versions.rst
deleted file mode 100644
index 4914f581b1fd..000000000000
--- a/Documentation/networking/devlink/devlink-info-versions.rst
+++ /dev/null
@@ -1,64 +0,0 @@
-.. SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
-
-=====================
-Devlink info versions
-=====================
-
-board.id
-========
-
-Unique identifier of the board design.
-
-board.rev
-=========
-
-Board design revision.
-
-asic.id
-=======
-
-ASIC design identifier.
-
-asic.rev
-========
-
-ASIC design revision.
-
-board.manufacture
-=================
-
-An identifier of the company or the facility which produced the part.
-
-fw
-==
-
-Overall firmware version, often representing the collection of
-fw.mgmt, fw.app, etc.
-
-fw.mgmt
-=======
-
-Control unit firmware version. This firmware is responsible for house
-keeping tasks, PHY control etc. but not the packet-by-packet data path
-operation.
-
-fw.app
-======
-
-Data path microcode controlling high-speed packet processing.
-
-fw.undi
-=======
-
-UNDI software, may include the UEFI driver, firmware or both.
-
-fw.ncsi
-=======
-
-Version of the software responsible for supporting/handling the
-Network Controller Sideband Interface.
-
-fw.psid
-=======
-
-Unique identifier of the firmware parameter set.
diff --git a/Documentation/networking/devlink/devlink-info.rst b/Documentation/networking/devlink/devlink-info.rst
new file mode 100644
index 000000000000..0385f15028b1
--- /dev/null
+++ b/Documentation/networking/devlink/devlink-info.rst
@@ -0,0 +1,94 @@
+.. SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+
+============
+Devlink Info
+============
+
+The ``devlink-info`` mechanism enables device drivers to report device
+information in a generic fashion. It is extensible, and enables exporting
+even device or driver specific information.
+
+devlink supports representing the following types of versions
+
+.. list-table:: List of version types
+   :widths: 5 95
+
+   * - Type
+     - Description
+   * - ``fixed``
+     - Represents fixed versions, which cannot change. For example,
+       component identifiers or the board version reported in the PCI VPD.
+   * - ``running``
+     - Represents the version of the currently running component. For
+       example the running version of firmware. These versions generally
+       only update after a reboot.
+   * - ``stored``
+     - Represents the version of a component as stored, such as after a
+       flash update. Stored values should update to reflect changes in the
+       flash even if a reboot has not yet occurred.
+
+Generic Versions
+================
+
+It is expected that drivers use the following generic names for exporting
+version information. Other information may be exposed using driver-specific
+names, but these should be documented in the driver-specific file.
+
+board.id
+--------
+
+Unique identifier of the board design.
+
+board.rev
+---------
+
+Board design revision.
+
+asic.id
+-------
+
+ASIC design identifier.
+
+asic.rev
+--------
+
+ASIC design revision.
+
+board.manufacture
+-----------------
+
+An identifier of the company or the facility which produced the part.
+
+fw
+--
+
+Overall firmware version, often representing the collection of
+fw.mgmt, fw.app, etc.
+
+fw.mgmt
+-------
+
+Control unit firmware version. This firmware is responsible for house
+keeping tasks, PHY control etc. but not the packet-by-packet data path
+operation.
+
+fw.app
+------
+
+Data path microcode controlling high-speed packet processing.
+
+fw.undi
+-------
+
+UNDI software, may include the UEFI driver, firmware or both.
+
+fw.ncsi
+-------
+
+Version of the software responsible for supporting/handling the
+Network Controller Sideband Interface.
+
+fw.psid
+-------
+
+Unique identifier of the firmware parameter set.
diff --git a/Documentation/networking/devlink/index.rst b/Documentation/networking/devlink/index.rst
index 71d9d0e859eb..687bb6e51fca 100644
--- a/Documentation/networking/devlink/index.rst
+++ b/Documentation/networking/devlink/index.rst
@@ -10,6 +10,6 @@ Contents:
    :maxdepth: 1
 
    devlink-health
-   devlink-info-versions
+   devlink-info
    devlink-trap
    devlink-trap-netdevsim
-- 
cgit v1.2.3-59-g8ed1b


From 8559dc33a31df343315a3deffda066886a097da5 Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Thu, 9 Jan 2020 14:46:13 -0800
Subject: devlink: convert devlink-params.txt to reStructuredText

Convert the generic parameters descriptions into the reStructuredText
format.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../networking/devlink/devlink-params.rst          | 90 ++++++++++++++++++++++
 .../networking/devlink/devlink-params.txt          | 71 -----------------
 Documentation/networking/devlink/index.rst         |  1 +
 3 files changed, 91 insertions(+), 71 deletions(-)
 create mode 100644 Documentation/networking/devlink/devlink-params.rst
 delete mode 100644 Documentation/networking/devlink/devlink-params.txt

diff --git a/Documentation/networking/devlink/devlink-params.rst b/Documentation/networking/devlink/devlink-params.rst
new file mode 100644
index 000000000000..b51b91f807f7
--- /dev/null
+++ b/Documentation/networking/devlink/devlink-params.rst
@@ -0,0 +1,90 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+==============
+Devlink Params
+==============
+
+``devlink`` provides capability for a driver to expose device parameters for low
+level device functionality. Since devlink can operate at the device-wide
+level, it can be used to provide configuration that may affect multiple
+ports on a single device.
+
+This document describes a number of generic parameters that are supported
+across multiple drivers. Each driver is also free to add their own
+parameters. Each driver must document the specific parameters they support,
+whether generic or not.
+
+Configuration modes
+===================
+
+Parameters may be set in different configuration modes.
+
+.. list-table:: Possible configuration modes
+   :widths: 5 90
+
+   * - Name
+     - Description
+   * - ``runtime``
+     - set while the driver is running, and takes effect immediately. No
+       reset is required.
+   * - ``driverinit``
+     - applied while the driver initializes. Requires the user to restart
+       the driver using the ``devlink`` reload command.
+   * - ``permanent``
+     - written to the device's non-volatile memory. A hard reset is required
+       for it to take effect.
+
+Generic configuration parameters
+================================
+The following is a list of generic configuration parameters that drivers may
+add. Use of generic parameters is preferred over each driver creating their
+own name.
+
+.. list-table:: List of generic parameters
+   :widths: 5 5 90
+
+   * - Name
+     - Type
+     - Description
+   * - ``enable_sriov``
+     - Boolean
+     - Enable Single Root I/O Virtualization (SRIOV) in the device.
+   * - ``ignore_ari``
+     - Boolean
+     - Ignore Alternative Routing-ID Interpretation (ARI) capability. If
+       enabled, the adapter will ignore ARI capability even when the
+       platform has support enabled. The device will create the same number
+       of partitions as when the platform does not support ARI.
+   * - ``msix_vec_per_pf_max``
+     - u32
+     - Provides the maximum number of MSI-X interrupts that a device can
+       create. Value is the same across all physical functions (PFs) in the
+       device.
+   * - ``msix_vec_per_pf_min``
+     - u32
+     - Provides the minimum number of MSI-X interrupts required for the
+       device to initialize. Value is the same across all physical functions
+       (PFs) in the device.
+   * - ``fw_load_policy``
+     - u8
+     - Control the device's firmware loading policy.
+        - ``DEVLINK_PARAM_FW_LOAD_POLICY_VALUE_DRIVER`` (0)
+          Load firmware version preferred by the driver.
+        - ``DEVLINK_PARAM_FW_LOAD_POLICY_VALUE_FLASH`` (1)
+          Load firmware currently stored in flash.
+        - ``DEVLINK_PARAM_FW_LOAD_POLICY_VALUE_DISK`` (2)
+          Load firmware currently available on host's disk.
+   * - ``reset_dev_on_drv_probe``
+     - u8
+     - Controls the device's reset policy on driver probe.
+        - ``DEVLINK_PARAM_RESET_DEV_ON_DRV_PROBE_VALUE_UNKNOWN`` (0)
+          Unknown or invalid value.
+        - ``DEVLINK_PARAM_RESET_DEV_ON_DRV_PROBE_VALUE_ALWAYS`` (1)
+          Always reset device on driver probe.
+        - ``DEVLINK_PARAM_RESET_DEV_ON_DRV_PROBE_VALUE_NEVER`` (2)
+          Never reset device on driver probe.
+        - ``DEVLINK_PARAM_RESET_DEV_ON_DRV_PROBE_VALUE_DISK`` (3)
+          Reset the device only if firmware can be found in the filesystem.
+   * - ``enable_roce``
+     - Boolean
+     - Enable handling of RoCE traffic in the device.
diff --git a/Documentation/networking/devlink/devlink-params.txt b/Documentation/networking/devlink/devlink-params.txt
deleted file mode 100644
index 04e234e9acc9..000000000000
--- a/Documentation/networking/devlink/devlink-params.txt
+++ /dev/null
@@ -1,71 +0,0 @@
-Devlink configuration parameters
-================================
-Following is the list of configuration parameters via devlink interface.
-Each parameter can be generic or driver specific and are device level
-parameters.
-
-Note that the driver-specific files should contain the generic params
-they support to, with supported config modes.
-
-Each parameter can be set in different configuration modes:
-	runtime		- set while driver is running, no reset required.
-	driverinit	- applied while driver initializes, requires restart
-			driver by devlink reload command.
-	permanent	- written to device's non-volatile memory, hard reset
-			required.
-
-Following is the list of parameters:
-====================================
-enable_sriov		[DEVICE, GENERIC]
-			Enable Single Root I/O Virtualisation (SRIOV) in
-			the device.
-			Type: Boolean
-
-ignore_ari		[DEVICE, GENERIC]
-			Ignore Alternative Routing-ID Interpretation (ARI)
-			capability. If enabled, adapter will ignore ARI
-			capability even when platforms has the support
-			enabled and creates same number of partitions when
-			platform does not support ARI.
-			Type: Boolean
-
-msix_vec_per_pf_max	[DEVICE, GENERIC]
-			Provides the maximum number of MSIX interrupts that
-			a device can create. Value is same across all
-			physical functions (PFs) in the device.
-			Type: u32
-
-msix_vec_per_pf_min	[DEVICE, GENERIC]
-			Provides the minimum number of MSIX interrupts required
-			for the device initialization. Value is same across all
-			physical functions (PFs) in the device.
-			Type: u32
-
-fw_load_policy		[DEVICE, GENERIC]
-			Controls the device's firmware loading policy.
-			Valid values:
-			* DEVLINK_PARAM_FW_LOAD_POLICY_VALUE_DRIVER (0)
-			  Load firmware version preferred by the driver.
-			* DEVLINK_PARAM_FW_LOAD_POLICY_VALUE_FLASH (1)
-			  Load firmware currently stored in flash.
-			* DEVLINK_PARAM_FW_LOAD_POLICY_VALUE_DISK (2)
-			  Load firmware currently available on host's disk.
-			Type: u8
-
-reset_dev_on_drv_probe	[DEVICE, GENERIC]
-			Controls the device's reset policy on driver probe.
-			Valid values:
-			* DEVLINK_PARAM_RESET_DEV_ON_DRV_PROBE_VALUE_UNKNOWN (0)
-			  Unknown or invalid value.
-			* DEVLINK_PARAM_RESET_DEV_ON_DRV_PROBE_VALUE_ALWAYS (1)
-			  Always reset device on driver probe.
-			* DEVLINK_PARAM_RESET_DEV_ON_DRV_PROBE_VALUE_NEVER (2)
-			  Never reset device on driver probe.
-			* DEVLINK_PARAM_RESET_DEV_ON_DRV_PROBE_VALUE_DISK (3)
-			  Reset only if device firmware can be found in the
-			  filesystem.
-			Type: u8
-
-enable_roce		[DEVICE, GENERIC]
-			Enable handling of RoCE traffic in the device.
-			Type: Boolean
diff --git a/Documentation/networking/devlink/index.rst b/Documentation/networking/devlink/index.rst
index 687bb6e51fca..ae3f20919d22 100644
--- a/Documentation/networking/devlink/index.rst
+++ b/Documentation/networking/devlink/index.rst
@@ -11,5 +11,6 @@ Contents:
 
    devlink-health
    devlink-info
+   devlink-params
    devlink-trap
    devlink-trap-netdevsim
-- 
cgit v1.2.3-59-g8ed1b


From d56a0c47a9555c35933c4f7a1f94e55bc809de41 Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Thu, 9 Jan 2020 14:46:14 -0800
Subject: devlink: add documentation for generic devlink parameters

A few generic devlink parameters have been added, but never documented.
Fix that now.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/devlink/devlink-params.rst | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/Documentation/networking/devlink/devlink-params.rst b/Documentation/networking/devlink/devlink-params.rst
index b51b91f807f7..7106c79e29be 100644
--- a/Documentation/networking/devlink/devlink-params.rst
+++ b/Documentation/networking/devlink/devlink-params.rst
@@ -88,3 +88,14 @@ own name.
    * - ``enable_roce``
      - Boolean
      - Enable handling of RoCE traffic in the device.
+   * - ``internal_err_reset``
+     - Boolean
+     - When enabled, the device driver will reset the device on internal
+       errors.
+   * - ``max_macs``
+     - u32
+     - Specifies the maximum number of MAC addresses per ethernet port of
+       this device.
+   * - ``region_snapshot_enable``
+     - Boolean
+     - Enable capture of ``devlink-region`` snapshots.
-- 
cgit v1.2.3-59-g8ed1b


From 5948f85bbcb3874746702ad8e9ca1bf787005d47 Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Thu, 9 Jan 2020 14:46:15 -0800
Subject: devlink: mention reloading in devlink-params.rst

Mention that drivers must support devlink-reload in order for driverinit
parameters to function properly

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/devlink/devlink-params.rst | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/Documentation/networking/devlink/devlink-params.rst b/Documentation/networking/devlink/devlink-params.rst
index 7106c79e29be..da2f85c0fa21 100644
--- a/Documentation/networking/devlink/devlink-params.rst
+++ b/Documentation/networking/devlink/devlink-params.rst
@@ -34,6 +34,13 @@ Parameters may be set in different configuration modes.
      - written to the device's non-volatile memory. A hard reset is required
        for it to take effect.
 
+Reloading
+---------
+
+In order for ``driverinit`` parameters to take effect, the driver must
+support reloading via the ``devlink-reload`` command. This command will
+request a reload of the device driver.
+
 Generic configuration parameters
 ================================
 The following is a list of generic configuration parameters that drivers may
-- 
cgit v1.2.3-59-g8ed1b


From 6c39e015f87fb46f68532d063d4fd337d094fdeb Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Thu, 9 Jan 2020 14:46:16 -0800
Subject: devlink: convert driver-specific files to reStructuredText

Several drivers document what parameters they support in
a devlink-params-*.txt file. This file is supposed to contain both the
list of generic parameters implemented by the driver, as well as a list
of driver-specific parameters and their descriptions.

It would also be good if the driver documentation included other
driver-specific implementations, such as info versions, devlink
regions, and so forth.

Convert all of these documentation files to reStructuredText, and rename
them to just the driver name. Future changes will include other
driver-specific implementations. Each file will contain a table for the
generic parameters implemented, as well as a separate table for the
driver-specific parameters.

Future sections such as for devlink info versions will be added to these
files. This avoids creating additional devlink-<feature>-<driver> files
for each devlink feature, reducing clutter in the documentation folder.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Cc: Tariq Toukan <tariqt@mellanox.com>
Cc: Saeed Mahameed <saeedm@mellanox.com>
Cc: Leon Romanovsky <leonro@mellanox.com>
Cc: Michael Chan <michael.chan@broadcom.com>
Cc: Andrew Lunn <andrew@lunn.ch>
Cc: Vivien Didelot <vivien.didelot@gmail.com>
Cc: Jiri Pirko <jiri@mellanox.com>
Cc: Ido Schimmel <idosch@mellanox.com>
Cc: Jakub Kicinski <jakub.kicinski@netronome.com>
Cc: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../device_drivers/ti/cpsw_switchdev.txt           |  2 +-
 Documentation/networking/devlink/bnxt.rst          | 41 ++++++++++++++++++++++
 .../networking/devlink/devlink-params-bnxt.txt     | 18 ----------
 .../networking/devlink/devlink-params-mlx5.txt     | 17 ---------
 .../networking/devlink/devlink-params-mlxsw.txt    | 10 ------
 .../devlink/devlink-params-mv88e6xxx.txt           |  7 ----
 .../networking/devlink/devlink-params-nfp.txt      |  5 ---
 .../devlink/devlink-params-ti-cpsw-switch.txt      | 10 ------
 Documentation/networking/devlink/index.rst         | 22 +++++++++++-
 Documentation/networking/devlink/mlx5.rst          | 41 ++++++++++++++++++++++
 Documentation/networking/devlink/mlxsw.rst         | 38 ++++++++++++++++++++
 Documentation/networking/devlink/mv88e6xxx.rst     | 28 +++++++++++++++
 Documentation/networking/devlink/nfp.rst           | 20 +++++++++++
 .../networking/devlink/ti-cpsw-switch.rst          | 31 ++++++++++++++++
 MAINTAINERS                                        |  2 +-
 15 files changed, 222 insertions(+), 70 deletions(-)
 create mode 100644 Documentation/networking/devlink/bnxt.rst
 delete mode 100644 Documentation/networking/devlink/devlink-params-bnxt.txt
 delete mode 100644 Documentation/networking/devlink/devlink-params-mlx5.txt
 delete mode 100644 Documentation/networking/devlink/devlink-params-mlxsw.txt
 delete mode 100644 Documentation/networking/devlink/devlink-params-mv88e6xxx.txt
 delete mode 100644 Documentation/networking/devlink/devlink-params-nfp.txt
 delete mode 100644 Documentation/networking/devlink/devlink-params-ti-cpsw-switch.txt
 create mode 100644 Documentation/networking/devlink/mlx5.rst
 create mode 100644 Documentation/networking/devlink/mlxsw.rst
 create mode 100644 Documentation/networking/devlink/mv88e6xxx.rst
 create mode 100644 Documentation/networking/devlink/nfp.rst
 create mode 100644 Documentation/networking/devlink/ti-cpsw-switch.rst

diff --git a/Documentation/networking/device_drivers/ti/cpsw_switchdev.txt b/Documentation/networking/device_drivers/ti/cpsw_switchdev.txt
index 5c8cee17fca9..12855ab268b8 100644
--- a/Documentation/networking/device_drivers/ti/cpsw_switchdev.txt
+++ b/Documentation/networking/device_drivers/ti/cpsw_switchdev.txt
@@ -39,7 +39,7 @@ but without enabling "switch" mode, or to different bridges.
 
 Devlink configuration parameters
 ====================
-See Documentation/networking/devlink-params-ti-cpsw-switch.txt
+See Documentation/networking/devlink/ti-cpsw-switch.rst
 
 ====================
 # Bridging in dual mac mode
diff --git a/Documentation/networking/devlink/bnxt.rst b/Documentation/networking/devlink/bnxt.rst
new file mode 100644
index 000000000000..79e746d22a53
--- /dev/null
+++ b/Documentation/networking/devlink/bnxt.rst
@@ -0,0 +1,41 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+====================
+bnxt devlink support
+====================
+
+This document describes the devlink features implemented by the ``bnxt``
+device driver.
+
+Parameters
+==========
+
+.. list-table:: Generic parameters implemented
+
+   * - Name
+     - Mode
+   * - ``enable_sriov``
+     - Permanent
+   * - ``ignore_ari``
+     - Permanent
+   * - ``msix_vec_per_pf_max``
+     - Permanent
+   * - ``msix_vec_per_pf_min``
+     - Permanent
+
+The ``bnxt`` driver also implements the following driver-specific
+parameters.
+
+.. list-table:: Driver-specific parameters implemented
+   :widths: 5 5 5 85
+
+   * - Name
+     - Type
+     - Mode
+     - Description
+   * - ``gre_ver_check``
+     - Boolean
+     - Permanent
+     - Generic Routing Encapsulation (GRE) version check will be enabled in
+       the device. If disabled, the device will skip the version check for
+       incoming packets.
diff --git a/Documentation/networking/devlink/devlink-params-bnxt.txt b/Documentation/networking/devlink/devlink-params-bnxt.txt
deleted file mode 100644
index 481aa303d5b4..000000000000
--- a/Documentation/networking/devlink/devlink-params-bnxt.txt
+++ /dev/null
@@ -1,18 +0,0 @@
-enable_sriov		[DEVICE, GENERIC]
-			Configuration mode: Permanent
-
-ignore_ari		[DEVICE, GENERIC]
-			Configuration mode: Permanent
-
-msix_vec_per_pf_max	[DEVICE, GENERIC]
-			Configuration mode: Permanent
-
-msix_vec_per_pf_min	[DEVICE, GENERIC]
-			Configuration mode: Permanent
-
-gre_ver_check		[DEVICE, DRIVER-SPECIFIC]
-			Generic Routing Encapsulation (GRE) version check will
-			be enabled in the device. If disabled, device skips
-			version checking for incoming packets.
-			Type: Boolean
-			Configuration mode: Permanent
diff --git a/Documentation/networking/devlink/devlink-params-mlx5.txt b/Documentation/networking/devlink/devlink-params-mlx5.txt
deleted file mode 100644
index 5071467118bd..000000000000
--- a/Documentation/networking/devlink/devlink-params-mlx5.txt
+++ /dev/null
@@ -1,17 +0,0 @@
-flow_steering_mode	[DEVICE, DRIVER-SPECIFIC]
-			Controls the flow steering mode of the driver.
-			Two modes are supported:
-			1. 'dmfs' - Device managed flow steering.
-			2. 'smfs  - Software/Driver managed flow steering.
-			In DMFS mode, the HW steering entities are created and
-			managed through the Firmware.
-			In SMFS mode, the HW steering entities are created and
-			managed though by the driver directly into Hardware
-			without firmware intervention.
-			Type: String
-			Configuration mode: runtime
-
-enable_roce		[DEVICE, GENERIC]
-			Enable handling of RoCE traffic in the device.
-			Defaultly enabled.
-			Configuration mode: driverinit
diff --git a/Documentation/networking/devlink/devlink-params-mlxsw.txt b/Documentation/networking/devlink/devlink-params-mlxsw.txt
deleted file mode 100644
index c63ea9fc7009..000000000000
--- a/Documentation/networking/devlink/devlink-params-mlxsw.txt
+++ /dev/null
@@ -1,10 +0,0 @@
-fw_load_policy		[DEVICE, GENERIC]
-			Configuration mode: driverinit
-
-acl_region_rehash_interval	[DEVICE, DRIVER-SPECIFIC]
-			Sets an interval for periodic ACL region rehashes.
-			The value is in milliseconds, minimal value is "3000".
-			Value "0" disables the periodic work.
-			The first rehash will be run right after value is set.
-			Type: u32
-			Configuration mode: runtime
diff --git a/Documentation/networking/devlink/devlink-params-mv88e6xxx.txt b/Documentation/networking/devlink/devlink-params-mv88e6xxx.txt
deleted file mode 100644
index 21c4b3556ef2..000000000000
--- a/Documentation/networking/devlink/devlink-params-mv88e6xxx.txt
+++ /dev/null
@@ -1,7 +0,0 @@
-ATU_hash		[DEVICE, DRIVER-SPECIFIC]
-			Select one of four possible hashing algorithms for
-			MAC addresses in the Address Translation Unit.
-			A value of 3 seems to work better than the default of
-			1 when many MAC addresses have the same OUI.
-			Configuration mode: runtime
-			Type: u8. 0-3 valid.
diff --git a/Documentation/networking/devlink/devlink-params-nfp.txt b/Documentation/networking/devlink/devlink-params-nfp.txt
deleted file mode 100644
index 43e4d4034865..000000000000
--- a/Documentation/networking/devlink/devlink-params-nfp.txt
+++ /dev/null
@@ -1,5 +0,0 @@
-fw_load_policy		[DEVICE, GENERIC]
-			Configuration mode: permanent
-
-reset_dev_on_drv_probe	[DEVICE, GENERIC]
-			Configuration mode: permanent
diff --git a/Documentation/networking/devlink/devlink-params-ti-cpsw-switch.txt b/Documentation/networking/devlink/devlink-params-ti-cpsw-switch.txt
deleted file mode 100644
index 4037458499f7..000000000000
--- a/Documentation/networking/devlink/devlink-params-ti-cpsw-switch.txt
+++ /dev/null
@@ -1,10 +0,0 @@
-ale_bypass	[DEVICE, DRIVER-SPECIFIC]
-		Allows to enable ALE_CONTROL(4).BYPASS mode for debug purposes.
-		All packets will be sent to the Host port only if enabled.
-		Type: bool
-		Configuration mode: runtime
-
-switch_mode	[DEVICE, DRIVER-SPECIFIC]
-		Enable switch mode
-		Type: bool
-		Configuration mode: runtime
diff --git a/Documentation/networking/devlink/index.rst b/Documentation/networking/devlink/index.rst
index ae3f20919d22..4035cbefda7c 100644
--- a/Documentation/networking/devlink/index.rst
+++ b/Documentation/networking/devlink/index.rst
@@ -4,7 +4,11 @@ Linux Devlink Documentation
 devlink is an API to expose device information and resources not directly
 related to any device class, such as chip-wide/switch-ASIC-wide configuration.
 
-Contents:
+Interface documentation
+-----------------------
+
+The following pages describe various interfaces available through devlink in
+general.
 
 .. toctree::
    :maxdepth: 1
@@ -14,3 +18,19 @@ Contents:
    devlink-params
    devlink-trap
    devlink-trap-netdevsim
+
+Driver-specific documentation
+-----------------------------
+
+Each driver that implements ``devlink`` is expected to document what
+parameters, info versions, and other features it supports.
+
+.. toctree::
+   :maxdepth: 1
+
+   bnxt
+   mlx5
+   mlxsw
+   mv88e6xxx
+   nfp
+   ti-cpsw-switch
diff --git a/Documentation/networking/devlink/mlx5.rst b/Documentation/networking/devlink/mlx5.rst
new file mode 100644
index 000000000000..cf2e19665e5c
--- /dev/null
+++ b/Documentation/networking/devlink/mlx5.rst
@@ -0,0 +1,41 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+====================
+mlx5 devlink support
+====================
+
+This document describes the devlink features implemented by the ``mlx5``
+device driver.
+
+Parameters
+==========
+
+.. list-table:: Generic parameters implemented
+
+   * - Name
+     - Mode
+   * - ``enable_roce``
+     - driverinit
+
+The ``mlx5`` driver also implements the following driver-specific
+parameters.
+
+.. list-table:: Driver-specific parameters implemented
+   :widths: 5 5 5 85
+
+   * - Name
+     - Type
+     - Mode
+     - Description
+   * - ``flow_steering_mode``
+     - string
+     - runtime
+     - Controls the flow steering mode of the driver
+
+       * ``dmfs`` Device managed flow steering. In DMFS mode, the HW
+         steering entities are created and managed through firmware.
+       * ``smfs`` Software managed flow steering. In SMFS mode, the HW
+         steering entities are created and manage through the driver without
+         firmware intervention.
+
+The ``mlx5`` driver supports reloading via ``DEVLINK_CMD_RELOAD``
diff --git a/Documentation/networking/devlink/mlxsw.rst b/Documentation/networking/devlink/mlxsw.rst
new file mode 100644
index 000000000000..dfb245f0c17d
--- /dev/null
+++ b/Documentation/networking/devlink/mlxsw.rst
@@ -0,0 +1,38 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+=====================
+mlxsw devlink support
+=====================
+
+This document describes the devlink features implemented by the ``mlxsw``
+device driver.
+
+Parameters
+==========
+
+.. list-table:: Generic parameters implemented
+
+   * - Name
+     - Mode
+   * - ``fw_load_policy``
+     - driverinit
+
+The ``mlxsw`` driver also implements the following driver-specific
+parameters.
+
+.. list-table:: Driver-specific parameters implemented
+   :widths: 5 5 5 85
+
+   * - Name
+     - Type
+     - Mode
+     - Description
+   * - ``acl_region_rehash_interval``
+     - u32
+     - runtime
+     - Sets an interval for periodic ACL region rehashes. The value is
+       specified in milliseconds, with a minimum of ``3000``. The value of
+       ``0`` disables periodic work entirely. The first rehash will be run
+       immediately after the value is set.
+
+The ``mlxsw`` driver supports reloading via ``DEVLINK_CMD_RELOAD``
diff --git a/Documentation/networking/devlink/mv88e6xxx.rst b/Documentation/networking/devlink/mv88e6xxx.rst
new file mode 100644
index 000000000000..c621212a47a1
--- /dev/null
+++ b/Documentation/networking/devlink/mv88e6xxx.rst
@@ -0,0 +1,28 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+=========================
+mv88e6xxx devlink support
+=========================
+
+This document describes the devlink features implemented by the ``mv88e6xxx``
+device driver.
+
+Parameters
+==========
+
+The ``mv88e6xxx`` driver implements the following driver-specific parameters.
+
+.. list-table:: Driver-specific parameters implemented
+   :widths: 5 5 5 85
+
+   * - Name
+     - Type
+     - Mode
+     - Description
+   * - ``ATU_hash``
+     - u8
+     - runtime
+     - Select one of four possible hashing algorithms for MAC addresses in
+       the Address Translation Unit. A value of 3 may work better than the
+       default of 1 when many MAC addresses have the same OUI. Only the
+       values 0 to 3 are valid for this parameter.
diff --git a/Documentation/networking/devlink/nfp.rst b/Documentation/networking/devlink/nfp.rst
new file mode 100644
index 000000000000..bbb3c6b2280c
--- /dev/null
+++ b/Documentation/networking/devlink/nfp.rst
@@ -0,0 +1,20 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+===================
+nfp devlink support
+===================
+
+This document describes the devlink features implemented by the ``nfp``
+device driver.
+
+Parameters
+==========
+
+.. list-table:: Generic parameters implemented
+
+   * - Name
+     - Mode
+   * - ``fw_load_policy``
+     - permanent
+   * - ``reset_dev_on_drv_probe``
+     - permanent
diff --git a/Documentation/networking/devlink/ti-cpsw-switch.rst b/Documentation/networking/devlink/ti-cpsw-switch.rst
new file mode 100644
index 000000000000..dc399e32abaa
--- /dev/null
+++ b/Documentation/networking/devlink/ti-cpsw-switch.rst
@@ -0,0 +1,31 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+==============================
+ti-cpsw-switch devlink support
+==============================
+
+This document describes the devlink features implemented by the ``ti-cpsw-switch``
+device driver.
+
+Parameters
+==========
+
+The ``ti-cpsw-switch`` driver implements the following driver-specific
+parameters.
+
+.. list-table:: Driver-specific parameters implemented
+   :widths: 5 5 5 85
+
+   * - Name
+     - Type
+     - Mode
+     - Description
+   * - ``ale_bypass``
+     - Boolean
+     - runtime
+     - Enables ALE_CONTROL(4).BYPASS mode for debugging purposes. In this
+       mode, all packets will be sent to the host port only.
+   * - ``switch_mode``
+     - Boolean
+     - runtime
+     - Enable switch mode
diff --git a/MAINTAINERS b/MAINTAINERS
index d0ea00e3e8b1..2549f10eb0b1 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -9890,7 +9890,7 @@ S:	Maintained
 F:	drivers/net/dsa/mv88e6xxx/
 F:	include/linux/platform_data/mv88e6xxx.h
 F:	Documentation/devicetree/bindings/net/dsa/marvell.txt
-F:	Documentation/networking/devlink-params-mv88e6xxx.txt
+F:	Documentation/networking/devlink/mv88e6xxx.rst
 
 MARVELL ARMADA DRM SUPPORT
 M:	Russell King <linux@armlinux.org.uk>
-- 
cgit v1.2.3-59-g8ed1b


From d4255d75856fc14faaa2760b687bc998af1cd8e9 Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Thu, 9 Jan 2020 14:46:17 -0800
Subject: devlink: document info versions for each driver

Add the set of info versions reported by each device driver, including
a description of what the version represents, and what modes (fixed,
running, stored) it reports.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Cc: Tariq Toukan <tariqt@mellanox.com>
Cc: Saeed Mahameed <saeedm@mellanox.com>
Cc: Leon Romanovsky <leonro@mellanox.com>
Cc: Michael Chan <michael.chan@broadcom.com>
Cc: Jiri Pirko <jiri@mellanox.com>
Cc: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/devlink/mlx5.rst  | 18 ++++++++++++
 Documentation/networking/devlink/mlxsw.rst | 21 ++++++++++++++
 Documentation/networking/devlink/nfp.rst   | 45 ++++++++++++++++++++++++++++++
 3 files changed, 84 insertions(+)

diff --git a/Documentation/networking/devlink/mlx5.rst b/Documentation/networking/devlink/mlx5.rst
index cf2e19665e5c..629a6e69c036 100644
--- a/Documentation/networking/devlink/mlx5.rst
+++ b/Documentation/networking/devlink/mlx5.rst
@@ -39,3 +39,21 @@ parameters.
          firmware intervention.
 
 The ``mlx5`` driver supports reloading via ``DEVLINK_CMD_RELOAD``
+
+Info versions
+=============
+
+The ``mlx5`` driver reports the following versions
+
+.. list-table:: devlink info versions implemented
+   :widths: 5 5 90
+
+   * - Name
+     - Type
+     - Description
+   * - ``fw.psid``
+     - fixed
+     - Used to represent the board id of the device.
+   * - ``fw.version``
+     - stored, running
+     - Three digit major.minor.subminor firmware version number.
diff --git a/Documentation/networking/devlink/mlxsw.rst b/Documentation/networking/devlink/mlxsw.rst
index dfb245f0c17d..ccba9769c651 100644
--- a/Documentation/networking/devlink/mlxsw.rst
+++ b/Documentation/networking/devlink/mlxsw.rst
@@ -36,3 +36,24 @@ parameters.
        immediately after the value is set.
 
 The ``mlxsw`` driver supports reloading via ``DEVLINK_CMD_RELOAD``
+
+Info versions
+=============
+
+The ``mlx5`` driver reports the following versions
+
+.. list-table:: devlink info versions implemented
+   :widths: 5 5 90
+
+   * - Name
+     - Type
+     - Description
+   * - ``hw.revision``
+     - fixed
+     - The hardware revision for this board
+   * - ``fw.psid``
+     - fixed
+     - Firmware PSID
+   * - ``fw.version``
+     - running
+     - Three digit firmware version
diff --git a/Documentation/networking/devlink/nfp.rst b/Documentation/networking/devlink/nfp.rst
index bbb3c6b2280c..a1717db0dfcc 100644
--- a/Documentation/networking/devlink/nfp.rst
+++ b/Documentation/networking/devlink/nfp.rst
@@ -18,3 +18,48 @@ Parameters
      - permanent
    * - ``reset_dev_on_drv_probe``
      - permanent
+
+Info versions
+=============
+
+The ``nfp`` driver reports the following versions
+
+.. list-table:: devlink info versions implemented
+   :widths: 5 5 90
+
+   * - Name
+     - Type
+     - Description
+   * - ``board.id``
+     - fixed
+     - Part number identifying the board design
+   * - ``board.rev``
+     - fixed
+     - Revision of the board design
+   * - ``board.manufacture``
+     - fixed
+     - Vendor of the board design
+   * - ``board.model``
+     - fixed
+     - Model name of the board design
+   * - ``fw.bundle_id``
+     - stored, running
+     - Firmware bundle id
+   * - ``fw.mgmt``
+     - stored, running
+     - Version of the management firmware
+   * - ``fw.cpld``
+     - stored, running
+     - The CPLD firmware component version
+   * - ``fw.app``
+     - stored, running
+     - The APP firmware component version
+   * - ``fw.undi``
+     - stored, running
+     - The UNDI firmware component version
+   * - ``fw.ncsi``
+     - stored, running
+     - The NSCI firmware component version
+   * - ``chip.init``
+     - stored, running
+     - The CFGR firmware component version
-- 
cgit v1.2.3-59-g8ed1b


From ee7a711f32a786819f2240290b88c8f5fc0cca96 Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Thu, 9 Jan 2020 14:46:18 -0800
Subject: devlink: add parameter documentation for the mlx4 driver

The mlx5 and mlxsw drivers have driver-specific documentation for the
devlink features they support. No such file was added for mlx4.

Add a file to document the mlx4 devlink support. Initially it contains
only the devlink parameters.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Cc: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/devlink/index.rst |  1 +
 Documentation/networking/devlink/mlx4.rst  | 43 ++++++++++++++++++++++++++++++
 2 files changed, 44 insertions(+)
 create mode 100644 Documentation/networking/devlink/mlx4.rst

diff --git a/Documentation/networking/devlink/index.rst b/Documentation/networking/devlink/index.rst
index 4035cbefda7c..0cbafef607d8 100644
--- a/Documentation/networking/devlink/index.rst
+++ b/Documentation/networking/devlink/index.rst
@@ -29,6 +29,7 @@ parameters, info versions, and other features it supports.
    :maxdepth: 1
 
    bnxt
+   mlx4
    mlx5
    mlxsw
    mv88e6xxx
diff --git a/Documentation/networking/devlink/mlx4.rst b/Documentation/networking/devlink/mlx4.rst
new file mode 100644
index 000000000000..4fa5c2b51c52
--- /dev/null
+++ b/Documentation/networking/devlink/mlx4.rst
@@ -0,0 +1,43 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+====================
+mlx4 devlink support
+====================
+
+This document describes the devlink features implemented by the ``mlx4``
+device driver.
+
+Parameters
+==========
+
+.. list-table:: Generic parameters implemented
+
+   * - Name
+     - Mode
+   * - ``internal_err_reset``
+     - driverinit, runtime
+   * - ``max_macs``
+     - driverinit
+   * - ``region_snapshot_enable``
+     - driverinit, runtime
+
+The ``mlx4`` driver also implements the following driver-specific
+parameters.
+
+.. list-table:: Driver-specific parameters implemented
+   :widths: 5 5 5 85
+
+   * - Name
+     - Type
+     - Mode
+     - Description
+   * - ``enable_64b_cqe_eqe``
+     - Boolean
+     - driverinit
+     - Enable 64 byte CQEs/EQEs, if the FW supports it.
+   * - ``enable_4k_uar``
+     - Boolean
+     - driverinit
+     - Enable using the 4k UAR.
+
+The ``mlx4`` driver supports reloading via ``DEVLINK_CMD_RELOAD``
-- 
cgit v1.2.3-59-g8ed1b


From 0f261c3ca09e112ac629588c926ec22091b17c42 Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Thu, 9 Jan 2020 14:46:19 -0800
Subject: devlink: add a driver-specific file for the qed driver

The qed driver recently added devlink support with a single devlink
parameter. Add a driver-specific file to document the devlink features
that the qed driver supports.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Cc: Ariel Elior <aelior@marvell.com>
Cc: GR-everest-linux-l2@marvell.com
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/devlink/index.rst |  1 +
 Documentation/networking/devlink/qed.rst   | 26 ++++++++++++++++++++++++++
 2 files changed, 27 insertions(+)
 create mode 100644 Documentation/networking/devlink/qed.rst

diff --git a/Documentation/networking/devlink/index.rst b/Documentation/networking/devlink/index.rst
index 0cbafef607d8..2007e257fd8a 100644
--- a/Documentation/networking/devlink/index.rst
+++ b/Documentation/networking/devlink/index.rst
@@ -34,4 +34,5 @@ parameters, info versions, and other features it supports.
    mlxsw
    mv88e6xxx
    nfp
+   qed
    ti-cpsw-switch
diff --git a/Documentation/networking/devlink/qed.rst b/Documentation/networking/devlink/qed.rst
new file mode 100644
index 000000000000..e7e17acf1eca
--- /dev/null
+++ b/Documentation/networking/devlink/qed.rst
@@ -0,0 +1,26 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+===================
+qed devlink support
+===================
+
+This document describes the devlink features implemented by the ``qed`` core
+device driver.
+
+Parameters
+==========
+
+The ``qed`` driver implements the following driver-specific parameters.
+
+.. list-table:: Driver-specific parameters implemented
+   :widths: 5 5 5 85
+
+   * - Name
+     - Type
+     - Mode
+     - Description
+   * - ``iwarp_cmt``
+     - Boolean
+     - runtime
+     - Enable iWARP functionality for 100g devices. Notee that this impacts
+       L2 performance, and is therefor not enabled by default.
-- 
cgit v1.2.3-59-g8ed1b


From 0b0f945f54584621f4f91da701909f8e78c9eb69 Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Thu, 9 Jan 2020 14:46:20 -0800
Subject: devlink: add a file documenting devlink regions

Also document the regions created by the mlx4 driver. This is currently
the only in-tree driver that creates devlink region snapshots.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Cc: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../networking/devlink/devlink-region.rst          | 54 ++++++++++++++++++++++
 Documentation/networking/devlink/index.rst         |  1 +
 Documentation/networking/devlink/mlx4.rst          | 13 ++++++
 3 files changed, 68 insertions(+)
 create mode 100644 Documentation/networking/devlink/devlink-region.rst

diff --git a/Documentation/networking/devlink/devlink-region.rst b/Documentation/networking/devlink/devlink-region.rst
new file mode 100644
index 000000000000..efa11b9a20da
--- /dev/null
+++ b/Documentation/networking/devlink/devlink-region.rst
@@ -0,0 +1,54 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+==============
+Devlink Region
+==============
+
+``devlink`` regions enable access to driver defined address regions using
+devlink.
+
+Each device can create and register its own supported address regions. The
+region can then be accessed via the devlink region interface.
+
+Region snapshots are collected by the driver, and can be accessed via read
+or dump commands. This allows future analysis on the created snapshots.
+
+The major benefit to creating a region is to provide access to internal
+address regions that are otherwise inaccessible to the user. They can be
+used to provide an additional way to debug complex error states.
+
+example usage
+-------------
+
+.. code:: shell
+
+    $ devlink region help
+    $ devlink region show [ DEV/REGION ]
+    $ devlink region del DEV/REGION snapshot SNAPSHOT_ID
+    $ devlink region dump DEV/REGION [ snapshot SNAPSHOT_ID ]
+    $ devlink region read DEV/REGION [ snapshot SNAPSHOT_ID ]
+            address ADDRESS length length
+
+    # Show all of the exposed regions with region sizes:
+    $ devlink region show
+    pci/0000:00:05.0/cr-space: size 1048576 snapshot [1 2]
+    pci/0000:00:05.0/fw-health: size 64 snapshot [1 2]
+
+    # Delete a snapshot using:
+    $ devlink region del pci/0000:00:05.0/cr-space snapshot 1
+
+    # Dump a snapshot:
+    $ devlink region dump pci/0000:00:05.0/fw-health snapshot 1
+    0000000000000000 0014 95dc 0014 9514 0035 1670 0034 db30
+    0000000000000010 0000 0000 ffff ff04 0029 8c00 0028 8cc8
+    0000000000000020 0016 0bb8 0016 1720 0000 0000 c00f 3ffc
+    0000000000000030 bada cce5 bada cce5 bada cce5 bada cce5
+
+    # Read a specific part of a snapshot:
+    $ devlink region read pci/0000:00:05.0/fw-health snapshot 1 address 0
+            length 16
+    0000000000000000 0014 95dc 0014 9514 0035 1670 0034 db30
+
+As regions are likely very device or driver specific, no generic regions are
+defined. See the driver-specific documentation files for information on the
+specific regions a driver supports.
diff --git a/Documentation/networking/devlink/index.rst b/Documentation/networking/devlink/index.rst
index 2007e257fd8a..3d351beedb0a 100644
--- a/Documentation/networking/devlink/index.rst
+++ b/Documentation/networking/devlink/index.rst
@@ -16,6 +16,7 @@ general.
    devlink-health
    devlink-info
    devlink-params
+   devlink-region
    devlink-trap
    devlink-trap-netdevsim
 
diff --git a/Documentation/networking/devlink/mlx4.rst b/Documentation/networking/devlink/mlx4.rst
index 4fa5c2b51c52..7b2d17ea5471 100644
--- a/Documentation/networking/devlink/mlx4.rst
+++ b/Documentation/networking/devlink/mlx4.rst
@@ -41,3 +41,16 @@ parameters.
      - Enable using the 4k UAR.
 
 The ``mlx4`` driver supports reloading via ``DEVLINK_CMD_RELOAD``
+
+Regions
+=======
+
+The ``mlx4`` driver supports dumping the firmware PCI crspace and health
+buffer during a critical firmware issue.
+
+In case a firmware command times out, firmware getting stuck, or a non zero
+value on the catastrophic buffer, a snapshot will be taken by the driver.
+
+The ``cr-space`` region will contain the firmware PCI crspace contents. The
+``fw-health`` region will contain the device firmware's health buffer.
+Snapshots for both of these regions are taken on the same event triggers.
-- 
cgit v1.2.3-59-g8ed1b


From cf4cb322d75253e38a41ec77d493a56c7a6648ab Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Thu, 9 Jan 2020 14:46:21 -0800
Subject: devlink: add documentation for ionic device driver

The IONIC device driver allocates a devlink and reports versions. Add
documentation for this driver.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Cc: Shannon Nelson <snelson@pensando.io>
Acked-by: Shannon Nelson <snelson@pensandi.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/devlink/index.rst |  1 +
 Documentation/networking/devlink/ionic.rst | 29 +++++++++++++++++++++++++++++
 2 files changed, 30 insertions(+)
 create mode 100644 Documentation/networking/devlink/ionic.rst

diff --git a/Documentation/networking/devlink/index.rst b/Documentation/networking/devlink/index.rst
index 3d351beedb0a..ce0ee563d83a 100644
--- a/Documentation/networking/devlink/index.rst
+++ b/Documentation/networking/devlink/index.rst
@@ -30,6 +30,7 @@ parameters, info versions, and other features it supports.
    :maxdepth: 1
 
    bnxt
+   ionic
    mlx4
    mlx5
    mlxsw
diff --git a/Documentation/networking/devlink/ionic.rst b/Documentation/networking/devlink/ionic.rst
new file mode 100644
index 000000000000..48da9c92d584
--- /dev/null
+++ b/Documentation/networking/devlink/ionic.rst
@@ -0,0 +1,29 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+=====================
+ionic devlink support
+=====================
+
+This document describes the devlink features implemented by the ``ionic``
+device driver.
+
+Info versions
+=============
+
+The ``ionic`` driver reports the following versions
+
+.. list-table:: devlink info versions implemented
+   :widths: 5 5 90
+
+   * - Name
+     - Type
+     - Description
+   * - ``fw``
+     - running
+     - Version of firmware running on the device
+   * - ``asic.id``
+     - fixed
+     - The ASIC type for this device
+   * - ``asic.rev``
+     - fixed
+     - The revision of the ASIC for this device
-- 
cgit v1.2.3-59-g8ed1b


From 04e4272cb3ae2bbb7a5d672918092366f48c9f52 Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Thu, 9 Jan 2020 14:46:22 -0800
Subject: devlink: rename and expand devlink-trap-netdevsim.rst

Rename the trap-specific netdevimsim.rst file, and expand it to include
documentation of all the devlink features currently implemented by the
netdevsim driver code.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Cc: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../networking/devlink/devlink-trap-netdevsim.rst  | 20 ------
 Documentation/networking/devlink/devlink-trap.rst  |  2 +-
 Documentation/networking/devlink/index.rst         |  2 +-
 Documentation/networking/devlink/netdevsim.rst     | 72 ++++++++++++++++++++++
 drivers/net/netdevsim/dev.c                        |  2 +-
 5 files changed, 75 insertions(+), 23 deletions(-)
 delete mode 100644 Documentation/networking/devlink/devlink-trap-netdevsim.rst
 create mode 100644 Documentation/networking/devlink/netdevsim.rst

diff --git a/Documentation/networking/devlink/devlink-trap-netdevsim.rst b/Documentation/networking/devlink/devlink-trap-netdevsim.rst
deleted file mode 100644
index b721c9415473..000000000000
--- a/Documentation/networking/devlink/devlink-trap-netdevsim.rst
+++ /dev/null
@@ -1,20 +0,0 @@
-.. SPDX-License-Identifier: GPL-2.0
-
-======================
-Devlink Trap netdevsim
-======================
-
-Driver-specific Traps
-=====================
-
-.. list-table:: List of Driver-specific Traps Registered by ``netdevsim``
-   :widths: 5 5 90
-
-   * - Name
-     - Type
-     - Description
-   * - ``fid_miss``
-     - ``exception``
-     - When a packet enters the device it is classified to a filtering
-       indentifier (FID) based on the ingress port and VLAN. This trap is used
-       to trap packets for which a FID could not be found
diff --git a/Documentation/networking/devlink/devlink-trap.rst b/Documentation/networking/devlink/devlink-trap.rst
index 03311849bfb1..cbaa750de37d 100644
--- a/Documentation/networking/devlink/devlink-trap.rst
+++ b/Documentation/networking/devlink/devlink-trap.rst
@@ -233,7 +233,7 @@ help debug packet drops caused by these exceptions. The following list includes
 links to the description of driver-specific traps registered by various device
 drivers:
 
-  * :doc:`devlink-trap-netdevsim`
+  * :doc:`netdevsim`
 
 Generic Packet Trap Groups
 ==========================
diff --git a/Documentation/networking/devlink/index.rst b/Documentation/networking/devlink/index.rst
index ce0ee563d83a..2417d56e27cc 100644
--- a/Documentation/networking/devlink/index.rst
+++ b/Documentation/networking/devlink/index.rst
@@ -18,7 +18,6 @@ general.
    devlink-params
    devlink-region
    devlink-trap
-   devlink-trap-netdevsim
 
 Driver-specific documentation
 -----------------------------
@@ -35,6 +34,7 @@ parameters, info versions, and other features it supports.
    mlx5
    mlxsw
    mv88e6xxx
+   netdevsim
    nfp
    qed
    ti-cpsw-switch
diff --git a/Documentation/networking/devlink/netdevsim.rst b/Documentation/networking/devlink/netdevsim.rst
new file mode 100644
index 000000000000..2a266b7e7b38
--- /dev/null
+++ b/Documentation/networking/devlink/netdevsim.rst
@@ -0,0 +1,72 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+=========================
+netdevsim devlink support
+=========================
+
+This document describes the ``devlink`` features supported by the
+``netdevsim`` device driver.
+
+Parameters
+==========
+
+.. list-table:: Generic parameters implemented
+
+   * - Name
+     - Mode
+   * - ``max_macs``
+     - driverinit
+
+The ``netdevsim`` driver also implements the following driver-specific
+parameters.
+
+.. list-table:: Driver-specific parameters implemented
+   :widths: 5 5 5 85
+
+   * - Name
+     - Type
+     - Mode
+     - Description
+   * - ``test1``
+     - Boolean
+     - driverinit
+     - Test parameter used to show how a driver-specific devlink parameter
+       can be implemented.
+
+The ``netdevsim`` driver supports reloading via ``DEVLINK_CMD_RELOAD``
+
+Regions
+=======
+
+The ``netdevsim`` driver exposes a ``dummy`` region as an example of how the
+devlink-region interfaces work. A snapshot is taken whenever the
+``take_snapshot`` debugfs file is written to.
+
+Resources
+=========
+
+The ``netdevsim`` driver exposes resources to control the number of FIB
+entries and FIB rule entries that the driver will allow.
+
+.. code:: shell
+
+    $ devlink resource set netdevsim/netdevsim0 path /IPv4/fib size 96
+    $ devlink resource set netdevsim/netdevsim0 path /IPv4/fib-rules size 16
+    $ devlink resource set netdevsim/netdevsim0 path /IPv6/fib size 64
+    $ devlink resource set netdevsim/netdevsim0 path /IPv6/fib-rules size 16
+    $ devlink dev reload netdevsim/netdevsim0
+
+Driver-specific Traps
+=====================
+
+.. list-table:: List of Driver-specific Traps Registered by ``netdevsim``
+   :widths: 5 5 90
+
+   * - Name
+     - Type
+     - Description
+   * - ``fid_miss``
+     - ``exception``
+     - When a packet enters the device it is classified to a filtering
+       indentifier (FID) based on the ingress port and VLAN. This trap is used
+       to trap packets for which a FID could not be found
diff --git a/drivers/net/netdevsim/dev.c b/drivers/net/netdevsim/dev.c
index 6663f79fe5d1..c572960bb54e 100644
--- a/drivers/net/netdevsim/dev.c
+++ b/drivers/net/netdevsim/dev.c
@@ -270,7 +270,7 @@ struct nsim_trap_data {
 };
 
 /* All driver-specific traps must be documented in
- * Documentation/networking/devlink/devlink-trap-netdevsim.rst
+ * Documentation/networking/devlink/netdevsim.rst
  */
 enum {
 	NSIM_TRAP_ID_BASE = DEVLINK_TRAP_GENERIC_ID_MAX,
-- 
cgit v1.2.3-59-g8ed1b


From a1af8e9a8029969632ff9f6b5ddc326c59d8c06d Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Thu, 9 Jan 2020 14:46:23 -0800
Subject: devlink: add a devlink-resource.rst documentation file

Take the little bit of documentation for resources from various commit
messages and combine it into a new devlink-resource.rst file.

This could probably be expanded on even further by someone with more
knowledge of how the devlink resources work.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../networking/devlink/devlink-resource.rst        | 62 ++++++++++++++++++++++
 Documentation/networking/devlink/index.rst         |  1 +
 2 files changed, 63 insertions(+)
 create mode 100644 Documentation/networking/devlink/devlink-resource.rst

diff --git a/Documentation/networking/devlink/devlink-resource.rst b/Documentation/networking/devlink/devlink-resource.rst
new file mode 100644
index 000000000000..93e92d2f0752
--- /dev/null
+++ b/Documentation/networking/devlink/devlink-resource.rst
@@ -0,0 +1,62 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+================
+Devlink Resource
+================
+
+``devlink`` provides the ability for drivers to register resources, which
+can allow administrators to see the device restrictions for a given
+resource, as well as how much of the given resource is currently
+in use. Additionally, these resources can optionally have configurable size.
+This could enable the administrator to limit the number of resources that
+are used.
+
+For example, the ``netdevsim`` driver enables ``/IPv4/fib`` and
+``/IPv4/fib-rules`` as resources to limit the number of IPv4 FIB entries and
+rules for a given device.
+
+Resource Ids
+============
+
+Each resource is represented by an id, and contains information about its
+current size and related sub resources. To access a sub resource, you
+specify the path of the resource. For example ``/IPv4/fib`` is the id for
+the ``fib`` sub-resource under the ``IPv4`` resource.
+
+example usage
+-------------
+
+The resources exposed by the driver can be observed, for example:
+
+.. code:: shell
+
+    $devlink resource show pci/0000:03:00.0
+    pci/0000:03:00.0:
+      name kvd size 245760 unit entry
+        resources:
+          name linear size 98304 occ 0 unit entry size_min 0 size_max 147456 size_gran 128
+          name hash_double size 60416 unit entry size_min 32768 size_max 180224 size_gran 128
+          name hash_single size 87040 unit entry size_min 65536 size_max 212992 size_gran 128
+
+Some resource's size can be changed. Examples:
+
+.. code:: shell
+
+    $devlink resource set pci/0000:03:00.0 path /kvd/hash_single size 73088
+    $devlink resource set pci/0000:03:00.0 path /kvd/hash_double size 74368
+
+The changes do not apply immediately, this can be validated by the 'size_new'
+attribute, which represents the pending change in size. For example:
+
+.. code:: shell
+
+    $devlink resource show pci/0000:03:00.0
+    pci/0000:03:00.0:
+      name kvd size 245760 unit entry size_valid false
+      resources:
+        name linear size 98304 size_new 147456 occ 0 unit entry size_min 0 size_max 147456 size_gran 128
+        name hash_double size 60416 unit entry size_min 32768 size_max 180224 size_gran 128
+        name hash_single size 87040 unit entry size_min 65536 size_max 212992 size_gran 128
+
+Note that changes in resource size may require a device reload to properly
+take effect.
diff --git a/Documentation/networking/devlink/index.rst b/Documentation/networking/devlink/index.rst
index 2417d56e27cc..10b51d863a5c 100644
--- a/Documentation/networking/devlink/index.rst
+++ b/Documentation/networking/devlink/index.rst
@@ -17,6 +17,7 @@ general.
    devlink-info
    devlink-params
    devlink-region
+   devlink-resource
    devlink-trap
 
 Driver-specific documentation
-- 
cgit v1.2.3-59-g8ed1b


From b7ec1eca71efe21d08fb1d05caa8ab06b22fe2f7 Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Thu, 9 Jan 2020 14:46:24 -0800
Subject: devlink: introduce devlink-dpipe.rst documentation file

Primarily based on the DPIPE netdev conference paper, introduce a new
file to document the dpipe interface.

This likely needs further improvement, but is at least a good overall
start.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/devlink/devlink-dpipe.rst | 252 +++++++++++++++++++++
 Documentation/networking/devlink/index.rst         |   1 +
 2 files changed, 253 insertions(+)
 create mode 100644 Documentation/networking/devlink/devlink-dpipe.rst

diff --git a/Documentation/networking/devlink/devlink-dpipe.rst b/Documentation/networking/devlink/devlink-dpipe.rst
new file mode 100644
index 000000000000..468fe1001b74
--- /dev/null
+++ b/Documentation/networking/devlink/devlink-dpipe.rst
@@ -0,0 +1,252 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+=============
+Devlink DPIPE
+=============
+
+Background
+==========
+
+While performing the hardware offloading process, much of the hardware
+specifics cannot be presented. These details are useful for debugging, and
+``devlink-dpipe`` provides a standardized way to provide visibility into the
+offloading process.
+
+For example, the routing longest prefix match (LPM) algorithm used by the
+Linux kernel may differ from the hardware implementation. The pipeline debug
+API (DPIPE) is aimed at providing the user visibility into the ASIC's
+pipeline in a generic way.
+
+The hardware offload process is expected to be done in a way that the user
+should not be able to distinguish between the hardware vs. software
+implementation. In this process, hardware specifics are neglected. In
+reality those details can have lots of meaning and should be exposed in some
+standard way.
+
+This problem is made even more complex when one wishes to offload the
+control path of the whole networking stack to a switch ASIC. Due to
+differences in the hardware and software models some processes cannot be
+represented correctly.
+
+One example is the kernel's LPM algorithm which in many cases differs
+greatly to the hardware implementation. The configuration API is the same,
+but one cannot rely on the Forward Information Base (FIB) to look like the
+Level Path Compression trie (LPC-trie) in hardware.
+
+In many situations trying to analyze systems failure solely based on the
+kernel's dump may not be enough. By combining this data with complementary
+information about the underlying hardware, this debugging can be made
+easier; additionally, the information can be useful when debugging
+performance issues.
+
+Overview
+========
+
+The ``devlink-dpipe`` interface closes this gap. The hardware's pipeline is
+modeled as a graph of match/action tables. Each table represents a specific
+hardware block. This model is not new, first being used by the P4 language.
+
+Traditionally it has been used as an alternative model for hardware
+configuration, but the ``devlink-dpipe`` interface uses it for visibility
+purposes as a standard complementary tool. The system's view from
+``devlink-dpipe`` should change according to the changes done by the
+standard configuration tools.
+
+For example, it’s quiet common to  implement Access Control Lists (ACL)
+using Ternary Content Addressable Memory (TCAM). The TCAM memory can be
+divided into TCAM regions. Complex TC filters can have multiple rules with
+different priorities and different lookup keys. On the other hand hardware
+TCAM regions have a predefined lookup key. Offloading the TC filter rules
+using TCAM engine can result in multiple TCAM regions being interconnected
+in a chain (which may affect the data path latency). In response to a new TC
+filter new tables should be created describing those regions.
+
+Model
+=====
+
+The ``DPIPE`` model introduces several objects:
+
+  * headers
+  * tables
+  * entries
+
+A ``header`` describes packet formats and provides names for fields within
+the packet. A ``table`` describes hardware blocks. An ``entry`` describes
+the actual content of a specific table.
+
+The hardware pipeline is not port specific, but rather describes the whole
+ASIC. Thus it is tied to the top of the ``devlink`` infrastructure.
+
+Drivers can register and unregister tables at run time, in order to support
+dynamic behavior. This dynamic behavior is mandatory for describing hardware
+blocks like TCAM regions which can be allocated and freed dynamically.
+
+``devlink-dpipe`` generally is not intended for configuration. The exception
+is hardware counting for a specific table.
+
+The following commands are used to obtain the ``dpipe`` objects from
+userspace:
+
+  * ``table_get``: Receive a table's description.
+  * ``headers_get``: Receive a device's supported headers.
+  * ``entries_get``: Receive a table's current entries.
+  * ``counters_set``: Enable or disable counters on a table.
+
+Table
+-----
+
+The driver should implement the following operations for each table:
+
+  * ``matches_dump``: Dump the supported matches.
+  * ``actions_dump``: Dump the supported actions.
+  * ``entries_dump``: Dump the actual content of the table.
+  * ``counters_set_update``: Synchronize hardware with counters enabled or
+    disabled.
+
+Header/Field
+------------
+
+In a similar way to P4 headers and fields are used to describe a table's
+behavior. There is a slight difference between the standard protocol headers
+and specific ASIC metadata. The protocol headers should be declared in the
+``devlink`` core API. On the other hand ASIC meta data is driver specific
+and should be defined in the driver. Additionally, each driver-specific
+devlink documentation file should document the driver-specific ``dpipe``
+headers it implements. The headers and fields are identified by enumeration.
+
+In order to provide further visibility some ASIC metadata fields could be
+mapped to kernel objects. For example, internal router interface indexes can
+be directly mapped to the net device ifindex. FIB table indexes used by
+different Virtual Routing and Forwarding (VRF) tables can be mapped to
+internal routing table indexes.
+
+Match
+-----
+
+Matches are kept primitive and close to hardware operation. Match types like
+LPM are not supported due to the fact that this is exactly a process we wish
+to describe in full detail. Example of matches:
+
+  * ``field_exact``: Exact match on a specific field.
+  * ``field_exact_mask``: Exact match on a specific field after masking.
+  * ``field_range``: Match on a specific range.
+
+The id's of the header and the field should be specified in order to
+identify the specific field. Furthermore, the header index should be
+specified in order to distinguish multiple headers of the same type in a
+packet (tunneling).
+
+Action
+------
+
+Similar to match, the actions are kept primitive and close to hardware
+operation. For example:
+
+  * ``field_modify``: Modify the field value.
+  * ``field_inc``: Increment the field value.
+  * ``push_header``: Add a header.
+  * ``pop_header``: Remove a header.
+
+Entry
+-----
+
+Entries of a specific table can be dumped on demand. Each eentry is
+identified with an index and its properties are described by a list of
+match/action values and specific counter. By dumping the tables content the
+interactions between tables can be resolved.
+
+Abstraction Example
+===================
+
+The following is an example of the abstraction model of the L3 part of
+Mellanox Spectrum ASIC. The blocks are described in the order they appear in
+the pipeline. The table sizes in the following examples are not real
+hardware sizes and are provided for demonstration purposes.
+
+LPM
+---
+
+The LPM algorithm can be implemented as a list of hash tables. Each hash
+table contains routes with the same prefix length. The root of the list is
+/32, and in case of a miss the hardware will continue to the next hash
+table. The depth of the search will affect the data path latency.
+
+In case of a hit the entry contains information about the next stage of the
+pipeline which resolves the MAC address. The next stage can be either local
+host table for directly connected routes, or adjacency table for next-hops.
+The ``meta.lpm_prefix`` field is used to connect two LPM tables.
+
+.. code::
+
+    table lpm_prefix_16 {
+      size: 4096,
+      counters_enabled: true,
+      match: { meta.vr_id: exact,
+               ipv4.dst_addr: exact_mask,
+               ipv6.dst_addr: exact_mask,
+               meta.lpm_prefix: exact },
+      action: { meta.adj_index: set,
+                meta.adj_group_size: set,
+                meta.rif_port: set,
+                meta.lpm_prefix: set },
+    }
+
+Local Host
+----------
+
+In the case of local routes the LPM lookup already resolves the egress
+router interface (RIF), yet the exact MAC address is not known. The local
+host table is a hash table combining the output interface id with
+destination IP address as a key. The result is the MAC address.
+
+.. code::
+
+    table local_host {
+      size: 4096,
+      counters_enabled: true,
+      match: { meta.rif_port: exact,
+               ipv4.dst_addr: exact},
+      action: { ethernet.daddr: set }
+    }
+
+Adjacency
+---------
+
+In case of remote routes this table does the ECMP. The LPM lookup results in
+ECMP group size and index that serves as a global offset into this table.
+Concurrently a hash of the packet is generated. Based on the ECMP group size
+and the packet's hash a local offset is generated. Multiple LPM entries can
+point to the same adjacency group.
+
+.. code::
+
+    table adjacency {
+      size: 4096,
+      counters_enabled: true,
+      match: { meta.adj_index: exact,
+               meta.adj_group_size: exact,
+               meta.packet_hash_index: exact },
+      action: { ethernet.daddr: set,
+                meta.erif: set }
+    }
+
+ERIF
+----
+
+In case the egress RIF and destination MAC have been resolved by previous
+tables this table does multiple operations like TTL decrease and MTU check.
+Then the decision of forward/drop is taken and the port L3 statistics are
+updated based on the packet's type (broadcast, unicast, multicast).
+
+.. code::
+
+    table erif {
+      size: 800,
+      counters_enabled: true,
+      match: { meta.rif_port: exact,
+               meta.is_l3_unicast: exact,
+               meta.is_l3_broadcast: exact,
+               meta.is_l3_multicast, exact },
+      action: { meta.l3_drop: set,
+                meta.l3_forward: set }
+    }
diff --git a/Documentation/networking/devlink/index.rst b/Documentation/networking/devlink/index.rst
index 10b51d863a5c..087ff54d53fc 100644
--- a/Documentation/networking/devlink/index.rst
+++ b/Documentation/networking/devlink/index.rst
@@ -13,6 +13,7 @@ general.
 .. toctree::
    :maxdepth: 1
 
+   devlink-dpipe
    devlink-health
    devlink-info
    devlink-params
-- 
cgit v1.2.3-59-g8ed1b


From 9cd3e2c64f889a48b0d14bda22e867722b55497e Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Thu, 9 Jan 2020 14:46:25 -0800
Subject: devlink: document region snapshot triggering from userspace

Now that devlink regions can be triggered via
DEVLINK_CMD_REGION_TRIGGER, document this in the devlink-region.rst
file.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/devlink/devlink-region.rst | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/Documentation/networking/devlink/devlink-region.rst b/Documentation/networking/devlink/devlink-region.rst
index efa11b9a20da..1a7683e7acb2 100644
--- a/Documentation/networking/devlink/devlink-region.rst
+++ b/Documentation/networking/devlink/devlink-region.rst
@@ -12,10 +12,13 @@ region can then be accessed via the devlink region interface.
 
 Region snapshots are collected by the driver, and can be accessed via read
 or dump commands. This allows future analysis on the created snapshots.
+Regions may optionally support triggering snapshots on demand.
 
 The major benefit to creating a region is to provide access to internal
-address regions that are otherwise inaccessible to the user. They can be
-used to provide an additional way to debug complex error states.
+address regions that are otherwise inaccessible to the user.
+
+Regions may also be used to provide an additional way to debug complex error
+states, but see also :doc:`devlink-health`
 
 example usage
 -------------
@@ -37,6 +40,9 @@ example usage
     # Delete a snapshot using:
     $ devlink region del pci/0000:00:05.0/cr-space snapshot 1
 
+    # Trigger (request) a snapshot be taken:
+    $ devlink region trigger pci/0000:00:05.0/cr-space
+
     # Dump a snapshot:
     $ devlink region dump pci/0000:00:05.0/fw-health snapshot 1
     0000000000000000 0014 95dc 0014 9514 0035 1670 0034 db30
-- 
cgit v1.2.3-59-g8ed1b


From 3071c51783b39d6a676d02a9256c3b3f87804285 Mon Sep 17 00:00:00 2001
From: Jonathan Lemon <jonathan.lemon@gmail.com>
Date: Thu, 9 Jan 2020 11:35:42 -0800
Subject: bnxt: Detach page from page pool before sending up the stack

When running in XDP mode, pages come from the page pool, and should
be freed back to the same pool or specifically detached.  Currently,
when the driver re-initializes, the page pool destruction is delayed
forever since it thinks there are oustanding pages.

Fixes: 322b87ca55f2 ("bnxt_en: add page_pool support")
Signed-off-by: Jonathan Lemon <jonathan.lemon@gmail.com>
Reviewed-by: Andy Gospodarek <gospo@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 39d4309b17fb..33eb8cd6551e 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -944,6 +944,7 @@ static struct sk_buff *bnxt_rx_page_skb(struct bnxt *bp,
 	dma_addr -= bp->rx_dma_offset;
 	dma_unmap_page_attrs(&bp->pdev->dev, dma_addr, PAGE_SIZE, bp->rx_dir,
 			     DMA_ATTR_WEAK_ORDERING);
+	page_pool_release_page(rxr->page_pool, page);
 
 	if (unlikely(!payload))
 		payload = eth_get_headlen(bp->dev, data_ptr, len);
-- 
cgit v1.2.3-59-g8ed1b


From de4ff8341f59d9893b9c1bd3b26d2f6c9d9a2dd2 Mon Sep 17 00:00:00 2001
From: YueHaibing <yuehaibing@huawei.com>
Date: Fri, 10 Jan 2020 01:35:17 +0000
Subject: sfc: remove duplicated include from ef10.c

Remove duplicated include.

Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Acked-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/ef10.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c
index 82063212cac4..4997f61de3d6 100644
--- a/drivers/net/ethernet/sfc/ef10.c
+++ b/drivers/net/ethernet/sfc/ef10.c
@@ -16,7 +16,6 @@
 #include "workarounds.h"
 #include "selftest.h"
 #include "ef10_sriov.h"
-#include "rx_common.h"
 #include <linux/in.h>
 #include <linux/jhash.h>
 #include <linux/wait.h>
-- 
cgit v1.2.3-59-g8ed1b


From 6821af8b7885870ed0dcb299e4b3cd38695538bc Mon Sep 17 00:00:00 2001
From: YueHaibing <yuehaibing@huawei.com>
Date: Fri, 10 Jan 2020 06:09:08 +0000
Subject: sfc: remove set but not used variable 'nic_data'

Fixes gcc '-Wunused-but-set-variable' warning:

drivers/net/ethernet/sfc/mcdi_functions.c: In function 'efx_mcdi_ev_init':
drivers/net/ethernet/sfc/mcdi_functions.c:79:28: warning:
 variable 'nic_data' set but not used [-Wunused-but-set-variable]

commit 4438b587fe4b ("sfc: move MCDI event queue management code")
introduces this unused variable.

Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/mcdi_functions.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/drivers/net/ethernet/sfc/mcdi_functions.c b/drivers/net/ethernet/sfc/mcdi_functions.c
index f5dad8bffc3d..dcfe78b0fa5a 100644
--- a/drivers/net/ethernet/sfc/mcdi_functions.c
+++ b/drivers/net/ethernet/sfc/mcdi_functions.c
@@ -76,13 +76,10 @@ int efx_mcdi_ev_init(struct efx_channel *channel, bool v1_cut_thru, bool v2)
 	MCDI_DECLARE_BUF(outbuf, MC_CMD_INIT_EVQ_V2_OUT_LEN);
 	size_t entries = channel->eventq.buf.len / EFX_BUF_SIZE;
 	struct efx_nic *efx = channel->efx;
-	struct efx_ef10_nic_data *nic_data;
 	size_t inlen, outlen;
 	dma_addr_t dma_addr;
 	int rc, i;
 
-	nic_data = efx->nic_data;
-
 	/* Fill event queue with all ones (i.e. empty events) */
 	memset(channel->eventq.buf.addr, 0xff, channel->eventq.buf.len);
 
-- 
cgit v1.2.3-59-g8ed1b


From d4fa06562ae2bb0379a4c6503aca6806a1aeb15e Mon Sep 17 00:00:00 2001
From: Huazhong Tan <tanhuazhong@huawei.com>
Date: Sat, 11 Jan 2020 16:33:47 +0800
Subject: net: hns3: split hclge_reset() into preparing and rebuilding part

hclge_reset() is a little bloated, and the process of PF FLR will
be separated from the reset task later. So this patch splits
hclge_reset() into hclge_reset_prepare() and hclge_reset_rebuild(),
then FLR can also reuse these two functions.

BTW, since hclge_clear_reset_cause() and hclge_reset_prepare_up()
will not affect the device, so in hclge_reset_rebuild(), these
functions are called without rtnl_lock.

Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../ethernet/hisilicon/hns3/hns3pf/hclge_main.c    | 59 +++++++++++++---------
 1 file changed, 34 insertions(+), 25 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
index 8bddda7f9092..9cbc0b620be0 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
@@ -3782,10 +3782,9 @@ static int hclge_reset_stack(struct hclge_dev *hdev)
 	return hclge_notify_client(hdev, HNAE3_RESTORE_CLIENT);
 }
 
-static void hclge_reset(struct hclge_dev *hdev)
+static int hclge_reset_prepare(struct hclge_dev *hdev)
 {
 	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev);
-	enum hnae3_reset_type reset_level;
 	int ret;
 
 	/* Initialize ae_dev reset status as well, in case enet layer wants to
@@ -3796,45 +3795,45 @@ static void hclge_reset(struct hclge_dev *hdev)
 	/* perform reset of the stack & ae device for a client */
 	ret = hclge_notify_roce_client(hdev, HNAE3_DOWN_CLIENT);
 	if (ret)
-		goto err_reset;
+		return ret;
 
 	ret = hclge_reset_prepare_down(hdev);
 	if (ret)
-		goto err_reset;
+		return ret;
 
 	rtnl_lock();
 	ret = hclge_notify_client(hdev, HNAE3_DOWN_CLIENT);
-	if (ret)
-		goto err_reset_lock;
-
 	rtnl_unlock();
-
-	ret = hclge_reset_prepare_wait(hdev);
 	if (ret)
-		goto err_reset;
+		return ret;
 
-	if (hclge_reset_wait(hdev))
-		goto err_reset;
+	return hclge_reset_prepare_wait(hdev);
+}
+
+static int hclge_reset_rebuild(struct hclge_dev *hdev)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev);
+	enum hnae3_reset_type reset_level;
+	int ret;
 
 	hdev->rst_stats.hw_reset_done_cnt++;
 
 	ret = hclge_notify_roce_client(hdev, HNAE3_UNINIT_CLIENT);
 	if (ret)
-		goto err_reset;
+		return ret;
 
 	rtnl_lock();
-
 	ret = hclge_reset_stack(hdev);
+	rtnl_unlock();
 	if (ret)
-		goto err_reset_lock;
+		return ret;
 
 	hclge_clear_reset_cause(hdev);
 
 	ret = hclge_reset_prepare_up(hdev);
 	if (ret)
-		goto err_reset_lock;
+		return ret;
 
-	rtnl_unlock();
 
 	ret = hclge_notify_roce_client(hdev, HNAE3_INIT_CLIENT);
 	/* ignore RoCE notify error if it fails HCLGE_RESET_MAX_FAIL_CNT - 1
@@ -3842,19 +3841,17 @@ static void hclge_reset(struct hclge_dev *hdev)
 	 */
 	if (ret &&
 	    hdev->rst_stats.reset_fail_cnt < HCLGE_RESET_MAX_FAIL_CNT - 1)
-		goto err_reset;
+		return ret;
 
 	rtnl_lock();
-
 	ret = hclge_notify_client(hdev, HNAE3_UP_CLIENT);
-	if (ret)
-		goto err_reset_lock;
-
 	rtnl_unlock();
+	if (ret)
+		return ret;
 
 	ret = hclge_notify_roce_client(hdev, HNAE3_UP_CLIENT);
 	if (ret)
-		goto err_reset;
+		return ret;
 
 	hdev->last_reset_time = jiffies;
 	hdev->rst_stats.reset_fail_cnt = 0;
@@ -3871,10 +3868,22 @@ static void hclge_reset(struct hclge_dev *hdev)
 	if (reset_level != HNAE3_NONE_RESET)
 		set_bit(reset_level, &hdev->reset_request);
 
+	return 0;
+}
+
+static void hclge_reset(struct hclge_dev *hdev)
+{
+	if (hclge_reset_prepare(hdev))
+		goto err_reset;
+
+	if (hclge_reset_wait(hdev))
+		goto err_reset;
+
+	if (hclge_reset_rebuild(hdev))
+		goto err_reset;
+
 	return;
 
-err_reset_lock:
-	rtnl_unlock();
 err_reset:
 	if (hclge_reset_err_handle(hdev))
 		hclge_reset_task_schedule(hdev);
-- 
cgit v1.2.3-59-g8ed1b


From 1cc9bc6e586782921f2bc31fd9f121e8fe2de100 Mon Sep 17 00:00:00 2001
From: Huazhong Tan <tanhuazhong@huawei.com>
Date: Sat, 11 Jan 2020 16:33:48 +0800
Subject: net: hns3: split hclgevf_reset() into preparing and rebuilding part

hclgevf_reset() is a little bloated, and the process of VF FLR will
be separated from the reset task later. So this patch splits
hclgevf_reset() into hclgevf_reset_prepare() and hclge_reset_rebuild(),
then FLR can also reuse these two functions. Also moves HNAE3_UP_CLIENT
into hclgevf_reset_stack().

Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c  | 81 +++++++++++-----------
 1 file changed, 39 insertions(+), 42 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
index cfa797ef9906..02b44d3a5ef1 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
@@ -1523,7 +1523,8 @@ static int hclgevf_reset_stack(struct hclgevf_dev *hdev)
 	/* clear handshake status with IMP */
 	hclgevf_reset_handshake(hdev, false);
 
-	return 0;
+	/* bring up the nic to enable TX/RX again */
+	return hclgevf_notify_client(hdev, HNAE3_UP_CLIENT);
 }
 
 static int hclgevf_reset_prepare_wait(struct hclgevf_dev *hdev)
@@ -1603,7 +1604,7 @@ static void hclgevf_reset_err_handle(struct hclgevf_dev *hdev)
 	}
 }
 
-static int hclgevf_reset(struct hclgevf_dev *hdev)
+static int hclgevf_reset_prepare(struct hclgevf_dev *hdev)
 {
 	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev);
 	int ret;
@@ -1613,62 +1614,64 @@ static int hclgevf_reset(struct hclgevf_dev *hdev)
 	 */
 	ae_dev->reset_type = hdev->reset_type;
 	hdev->rst_stats.rst_cnt++;
-	rtnl_lock();
 
+	rtnl_lock();
 	/* bring down the nic to stop any ongoing TX/RX */
 	ret = hclgevf_notify_client(hdev, HNAE3_DOWN_CLIENT);
-	if (ret)
-		goto err_reset_lock;
-
 	rtnl_unlock();
-
-	ret = hclgevf_reset_prepare_wait(hdev);
 	if (ret)
-		goto err_reset;
+		return ret;
 
-	/* check if VF could successfully fetch the hardware reset completion
-	 * status from the hardware
-	 */
-	ret = hclgevf_reset_wait(hdev);
-	if (ret) {
-		/* can't do much in this situation, will disable VF */
-		dev_err(&hdev->pdev->dev,
-			"VF failed(=%d) to fetch H/W reset completion status\n",
-			ret);
-		goto err_reset;
-	}
+	return hclgevf_reset_prepare_wait(hdev);
+}
+
+static int hclgevf_reset_rebuild(struct hclgevf_dev *hdev)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev);
+	int ret;
 
 	hdev->rst_stats.hw_rst_done_cnt++;
 
 	rtnl_lock();
-
 	/* now, re-initialize the nic client and ae device */
 	ret = hclgevf_reset_stack(hdev);
+	rtnl_unlock();
 	if (ret) {
 		dev_err(&hdev->pdev->dev, "failed to reset VF stack\n");
-		goto err_reset_lock;
+		return ret;
 	}
 
-	/* bring up the nic to enable TX/RX again */
-	ret = hclgevf_notify_client(hdev, HNAE3_UP_CLIENT);
-	if (ret)
-		goto err_reset_lock;
-
-	rtnl_unlock();
-
 	hdev->last_reset_time = jiffies;
 	ae_dev->reset_type = HNAE3_NONE_RESET;
 	hdev->rst_stats.rst_done_cnt++;
 	hdev->rst_stats.rst_fail_cnt = 0;
 	clear_bit(HCLGEVF_STATE_RST_FAIL, &hdev->state);
 
-	return ret;
-err_reset_lock:
-	rtnl_unlock();
+	return 0;
+}
+
+static void hclgevf_reset(struct hclgevf_dev *hdev)
+{
+	if (hclgevf_reset_prepare(hdev))
+		goto err_reset;
+
+	/* check if VF could successfully fetch the hardware reset completion
+	 * status from the hardware
+	 */
+	if (hclgevf_reset_wait(hdev)) {
+		/* can't do much in this situation, will disable VF */
+		dev_err(&hdev->pdev->dev,
+			"failed to fetch H/W reset completion status\n");
+		goto err_reset;
+	}
+
+	if (hclgevf_reset_rebuild(hdev))
+		goto err_reset;
+
+	return;
+
 err_reset:
 	hclgevf_reset_err_handle(hdev);
-
-	return ret;
 }
 
 static enum hnae3_reset_type hclgevf_get_reset_level(struct hclgevf_dev *hdev,
@@ -1802,8 +1805,6 @@ static void hclgevf_reset_service_task(struct hclgevf_dev *hdev)
 {
 #define	HCLGEVF_MAX_RESET_ATTEMPTS_CNT	3
 
-	int ret;
-
 	if (!test_and_clear_bit(HCLGEVF_STATE_RST_SERVICE_SCHED, &hdev->state))
 		return;
 
@@ -1822,12 +1823,8 @@ static void hclgevf_reset_service_task(struct hclgevf_dev *hdev)
 		hdev->last_reset_time = jiffies;
 		while ((hdev->reset_type =
 			hclgevf_get_reset_level(hdev, &hdev->reset_pending))
-		       != HNAE3_NONE_RESET) {
-			ret = hclgevf_reset(hdev);
-			if (ret)
-				dev_err(&hdev->pdev->dev,
-					"VF stack reset failed %d.\n", ret);
-		}
+		       != HNAE3_NONE_RESET)
+			hclgevf_reset(hdev);
 	} else if (test_and_clear_bit(HCLGEVF_RESET_REQUESTED,
 				      &hdev->reset_state)) {
 		/* we could be here when either of below happens:
-- 
cgit v1.2.3-59-g8ed1b


From 8627bdedc435865b015cedeb26e3791dec5b3a41 Mon Sep 17 00:00:00 2001
From: Huazhong Tan <tanhuazhong@huawei.com>
Date: Sat, 11 Jan 2020 16:33:49 +0800
Subject: net: hns3: refactor the precedure of PF FLR

Currently, the actual work of PF FLR is handled in the reset task,
which is asynchronous. So in some case, if the preparing and
rebuilding are not done, then the PF FLR will trigger some problems,
for example, makes hardware go into chaos.

So this patch separates the process of PF FLR from reset task, and
adds a semaphore to serialize this reset and others.

When FLR's preparing fails, if there has other higher level reset
pending or failing times less than the HCLGE_FLR_RETRY_CNT, this
preparing should be retried, otherwise PF and its VF may get into
wrong state.

BTW, while the hardware reports misc interrupt during pcie_flr(),
the driver can not receive this interrupt anymore, so disable it
when hclge_flr_prepare() return, and re-enable it when enter
hclge_flr_done().

Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../ethernet/hisilicon/hns3/hns3pf/hclge_main.c    | 85 +++++++++++-----------
 .../ethernet/hisilicon/hns3/hns3pf/hclge_main.h    |  1 +
 2 files changed, 43 insertions(+), 43 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
index 9cbc0b620be0..89621b587603 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
@@ -3277,8 +3277,6 @@ static int hclge_reset_wait(struct hclge_dev *hdev)
 		reg = HCLGE_FUN_RST_ING;
 		reg_bit = HCLGE_FUN_RST_ING_B;
 		break;
-	case HNAE3_FLR_RESET:
-		break;
 	default:
 		dev_err(&hdev->pdev->dev,
 			"Wait for unsupported reset type: %d\n",
@@ -3286,20 +3284,6 @@ static int hclge_reset_wait(struct hclge_dev *hdev)
 		return -EINVAL;
 	}
 
-	if (hdev->reset_type == HNAE3_FLR_RESET) {
-		while (!test_bit(HNAE3_FLR_DONE, &hdev->flr_state) &&
-		       cnt++ < HCLGE_RESET_WAIT_CNT)
-			msleep(HCLGE_RESET_WATI_MS);
-
-		if (!test_bit(HNAE3_FLR_DONE, &hdev->flr_state)) {
-			dev_err(&hdev->pdev->dev,
-				"flr wait timeout: %u\n", cnt);
-			return -EBUSY;
-		}
-
-		return 0;
-	}
-
 	val = hclge_read_dev(&hdev->hw, reg);
 	while (hnae3_get_bit(val, reg_bit) && cnt < HCLGE_RESET_WAIT_CNT) {
 		msleep(HCLGE_RESET_WATI_MS);
@@ -3490,12 +3474,6 @@ static void hclge_do_reset(struct hclge_dev *hdev)
 		set_bit(HNAE3_FUNC_RESET, &hdev->reset_pending);
 		hclge_reset_task_schedule(hdev);
 		break;
-	case HNAE3_FLR_RESET:
-		dev_info(&pdev->dev, "FLR requested\n");
-		/* schedule again to check later */
-		set_bit(HNAE3_FLR_RESET, &hdev->reset_pending);
-		hclge_reset_task_schedule(hdev);
-		break;
 	default:
 		dev_warn(&pdev->dev,
 			 "Unsupported reset type: %d\n", hdev->reset_type);
@@ -3650,10 +3628,6 @@ static int hclge_reset_prepare_wait(struct hclge_dev *hdev)
 		ret = hclge_func_reset_sync_vf(hdev);
 		if (ret)
 			return ret;
-
-		set_bit(HCLGE_STATE_CMD_DISABLE, &hdev->state);
-		set_bit(HNAE3_FLR_DOWN, &hdev->flr_state);
-		hdev->rst_stats.flr_rst_cnt++;
 		break;
 	case HNAE3_IMP_RESET:
 		hclge_handle_imp_error(hdev);
@@ -3989,12 +3963,13 @@ static void hclge_reset_service_task(struct hclge_dev *hdev)
 	if (!test_and_clear_bit(HCLGE_STATE_RST_SERVICE_SCHED, &hdev->state))
 		return;
 
-	if (test_and_set_bit(HCLGE_STATE_RST_HANDLING, &hdev->state))
-		return;
+	down(&hdev->reset_sem);
+	set_bit(HCLGE_STATE_RST_HANDLING, &hdev->state);
 
 	hclge_reset_subtask(hdev);
 
 	clear_bit(HCLGE_STATE_RST_HANDLING, &hdev->state);
+	up(&hdev->reset_sem);
 }
 
 static void hclge_update_vport_alive(struct hclge_dev *hdev)
@@ -9341,30 +9316,53 @@ static void hclge_state_uninit(struct hclge_dev *hdev)
 
 static void hclge_flr_prepare(struct hnae3_ae_dev *ae_dev)
 {
-#define HCLGE_FLR_WAIT_MS	100
-#define HCLGE_FLR_WAIT_CNT	50
-	struct hclge_dev *hdev = ae_dev->priv;
-	int cnt = 0;
+#define HCLGE_FLR_RETRY_WAIT_MS	500
+#define HCLGE_FLR_RETRY_CNT	5
 
-	clear_bit(HNAE3_FLR_DOWN, &hdev->flr_state);
-	clear_bit(HNAE3_FLR_DONE, &hdev->flr_state);
-	set_bit(HNAE3_FLR_RESET, &hdev->default_reset_request);
-	hclge_reset_event(hdev->pdev, NULL);
+	struct hclge_dev *hdev = ae_dev->priv;
+	int retry_cnt = 0;
+	int ret;
 
-	while (!test_bit(HNAE3_FLR_DOWN, &hdev->flr_state) &&
-	       cnt++ < HCLGE_FLR_WAIT_CNT)
-		msleep(HCLGE_FLR_WAIT_MS);
+retry:
+	down(&hdev->reset_sem);
+	set_bit(HCLGE_STATE_RST_HANDLING, &hdev->state);
+	hdev->reset_type = HNAE3_FLR_RESET;
+	ret = hclge_reset_prepare(hdev);
+	if (ret) {
+		dev_err(&hdev->pdev->dev, "fail to prepare FLR, ret=%d\n",
+			ret);
+		if (hdev->reset_pending ||
+		    retry_cnt++ < HCLGE_FLR_RETRY_CNT) {
+			dev_err(&hdev->pdev->dev,
+				"reset_pending:0x%lx, retry_cnt:%d\n",
+				hdev->reset_pending, retry_cnt);
+			clear_bit(HCLGE_STATE_RST_HANDLING, &hdev->state);
+			up(&hdev->reset_sem);
+			msleep(HCLGE_FLR_RETRY_WAIT_MS);
+			goto retry;
+		}
+	}
 
-	if (!test_bit(HNAE3_FLR_DOWN, &hdev->flr_state))
-		dev_err(&hdev->pdev->dev,
-			"flr wait down timeout: %d\n", cnt);
+	/* disable misc vector before FLR done */
+	hclge_enable_vector(&hdev->misc_vector, false);
+	set_bit(HCLGE_STATE_CMD_DISABLE, &hdev->state);
+	hdev->rst_stats.flr_rst_cnt++;
 }
 
 static void hclge_flr_done(struct hnae3_ae_dev *ae_dev)
 {
 	struct hclge_dev *hdev = ae_dev->priv;
+	int ret;
+
+	hclge_enable_vector(&hdev->misc_vector, true);
 
-	set_bit(HNAE3_FLR_DONE, &hdev->flr_state);
+	ret = hclge_reset_rebuild(hdev);
+	if (ret)
+		dev_err(&hdev->pdev->dev, "fail to rebuild, ret=%d\n", ret);
+
+	hdev->reset_type = HNAE3_NONE_RESET;
+	clear_bit(HCLGE_STATE_RST_HANDLING, &hdev->state);
+	up(&hdev->reset_sem);
 }
 
 static void hclge_clear_resetting_state(struct hclge_dev *hdev)
@@ -9407,6 +9405,7 @@ static int hclge_init_ae_dev(struct hnae3_ae_dev *ae_dev)
 
 	mutex_init(&hdev->vport_lock);
 	spin_lock_init(&hdev->fd_rule_lock);
+	sema_init(&hdev->reset_sem, 1);
 
 	ret = hclge_pci_init(hdev);
 	if (ret) {
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h
index 4e5cfda16eb7..1c1d6b358897 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h
@@ -720,6 +720,7 @@ struct hclge_dev {
 	unsigned long reset_request;	/* reset has been requested */
 	unsigned long reset_pending;	/* client rst is pending to be served */
 	struct hclge_rst_stats rst_stats;
+	struct semaphore reset_sem;	/* protect reset process */
 	u32 fw_version;
 	u16 num_vmdq_vport;		/* Num vmdq vport this PF has set up */
 	u16 num_tqps;			/* Num task queue pairs of this PF */
-- 
cgit v1.2.3-59-g8ed1b


From f28368bb4542dbc3c5b6e818211a68cbca6f4d75 Mon Sep 17 00:00:00 2001
From: Huazhong Tan <tanhuazhong@huawei.com>
Date: Sat, 11 Jan 2020 16:33:50 +0800
Subject: net: hns3: refactor the procedure of VF FLR

Currently, the actual work of VF FLR is handled in the reset task,
which is asynchronous. So in some case, if the preparing and
rebuilding are not done, then the VF FLR will trigger some problems,
for example, makes hardware go into chaos.

So this patch separates the process of VF FLR from reset task, and
adds a semaphore to serialize this reset and others.

When FLR's preparing fails, if there has other higher level reset
pending or failing times less than the HCLGE_FLR_RETRY_CNT, this
preparing should be retried, otherwise it will get into a wrong state.

BTW, while the hardware reports misc interrupt during pcie_flr(),
the driver can not receive this interrupt anymore, so disable it
when hclgevf_flr_prepare() return, and re-enable it when enter
hclgevf_flr_done().

Avoid declaring internal function hclgevf_enable_vector(), this patch
also moves its definition forward, and removes unused enum
hnae3_flr_state.

Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hnae3.h        |   5 -
 .../ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c  | 114 ++++++++++-----------
 .../ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h  |   1 +
 3 files changed, 55 insertions(+), 65 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hnae3.h b/drivers/net/ethernet/hisilicon/hns3/hnae3.h
index 6b131ab36c7e..a3e4081b84ba 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hnae3.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hnae3.h
@@ -167,11 +167,6 @@ enum hnae3_reset_type {
 	HNAE3_MAX_RESET,
 };
 
-enum hnae3_flr_state {
-	HNAE3_FLR_DOWN,
-	HNAE3_FLR_DONE,
-};
-
 enum hnae3_port_base_vlan_state {
 	HNAE3_PORT_BASE_VLAN_DISABLE,
 	HNAE3_PORT_BASE_VLAN_ENABLE,
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
index 02b44d3a5ef1..b26b8ad170d0 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
@@ -1411,32 +1411,6 @@ static int hclgevf_notify_client(struct hclgevf_dev *hdev,
 	return ret;
 }
 
-static void hclgevf_flr_done(struct hnae3_ae_dev *ae_dev)
-{
-	struct hclgevf_dev *hdev = ae_dev->priv;
-
-	set_bit(HNAE3_FLR_DONE, &hdev->flr_state);
-}
-
-static int hclgevf_flr_poll_timeout(struct hclgevf_dev *hdev,
-				    unsigned long delay_us,
-				    unsigned long wait_cnt)
-{
-	unsigned long cnt = 0;
-
-	while (!test_bit(HNAE3_FLR_DONE, &hdev->flr_state) &&
-	       cnt++ < wait_cnt)
-		usleep_range(delay_us, delay_us * 2);
-
-	if (!test_bit(HNAE3_FLR_DONE, &hdev->flr_state)) {
-		dev_err(&hdev->pdev->dev,
-			"flr wait timeout\n");
-		return -ETIMEDOUT;
-	}
-
-	return 0;
-}
-
 static int hclgevf_reset_wait(struct hclgevf_dev *hdev)
 {
 #define HCLGEVF_RESET_WAIT_US	20000
@@ -1447,11 +1421,7 @@ static int hclgevf_reset_wait(struct hclgevf_dev *hdev)
 	u32 val;
 	int ret;
 
-	if (hdev->reset_type == HNAE3_FLR_RESET)
-		return hclgevf_flr_poll_timeout(hdev,
-						HCLGEVF_RESET_WAIT_US,
-						HCLGEVF_RESET_WAIT_CNT);
-	else if (hdev->reset_type == HNAE3_VF_RESET)
+	if (hdev->reset_type == HNAE3_VF_RESET)
 		ret = readl_poll_timeout(hdev->hw.io_base +
 					 HCLGEVF_VF_RST_ING, val,
 					 !(val & HCLGEVF_VF_RST_ING_BIT),
@@ -1533,18 +1503,10 @@ static int hclgevf_reset_prepare_wait(struct hclgevf_dev *hdev)
 
 	int ret = 0;
 
-	switch (hdev->reset_type) {
-	case HNAE3_VF_FUNC_RESET:
+	if (hdev->reset_type == HNAE3_VF_FUNC_RESET) {
 		ret = hclgevf_send_mbx_msg(hdev, HCLGE_MBX_RESET, 0, NULL,
 					   0, true, NULL, sizeof(u8));
 		hdev->rst_stats.vf_func_rst_cnt++;
-		break;
-	case HNAE3_FLR_RESET:
-		set_bit(HNAE3_FLR_DOWN, &hdev->flr_state);
-		hdev->rst_stats.flr_rst_cnt++;
-		break;
-	default:
-		break;
 	}
 
 	set_bit(HCLGEVF_STATE_CMD_DISABLE, &hdev->state);
@@ -1734,25 +1696,60 @@ static void hclgevf_set_def_reset_request(struct hnae3_ae_dev *ae_dev,
 	set_bit(rst_type, &hdev->default_reset_request);
 }
 
+static void hclgevf_enable_vector(struct hclgevf_misc_vector *vector, bool en)
+{
+	writel(en ? 1 : 0, vector->addr);
+}
+
 static void hclgevf_flr_prepare(struct hnae3_ae_dev *ae_dev)
 {
-#define HCLGEVF_FLR_WAIT_MS	100
-#define HCLGEVF_FLR_WAIT_CNT	50
+#define HCLGEVF_FLR_RETRY_WAIT_MS	500
+#define HCLGEVF_FLR_RETRY_CNT		5
+
+	struct hclgevf_dev *hdev = ae_dev->priv;
+	int retry_cnt = 0;
+	int ret;
+
+retry:
+	down(&hdev->reset_sem);
+	set_bit(HCLGEVF_STATE_RST_HANDLING, &hdev->state);
+	hdev->reset_type = HNAE3_FLR_RESET;
+	ret = hclgevf_reset_prepare(hdev);
+	if (ret) {
+		dev_err(&hdev->pdev->dev, "fail to prepare FLR, ret=%d\n",
+			ret);
+		if (hdev->reset_pending ||
+		    retry_cnt++ < HCLGEVF_FLR_RETRY_CNT) {
+			dev_err(&hdev->pdev->dev,
+				"reset_pending:0x%lx, retry_cnt:%d\n",
+				hdev->reset_pending, retry_cnt);
+			clear_bit(HCLGEVF_STATE_RST_HANDLING, &hdev->state);
+			up(&hdev->reset_sem);
+			msleep(HCLGEVF_FLR_RETRY_WAIT_MS);
+			goto retry;
+		}
+	}
+
+	/* disable misc vector before FLR done */
+	hclgevf_enable_vector(&hdev->misc_vector, false);
+	hdev->rst_stats.flr_rst_cnt++;
+}
+
+static void hclgevf_flr_done(struct hnae3_ae_dev *ae_dev)
+{
 	struct hclgevf_dev *hdev = ae_dev->priv;
-	int cnt = 0;
+	int ret;
 
-	clear_bit(HNAE3_FLR_DOWN, &hdev->flr_state);
-	clear_bit(HNAE3_FLR_DONE, &hdev->flr_state);
-	set_bit(HNAE3_FLR_RESET, &hdev->default_reset_request);
-	hclgevf_reset_event(hdev->pdev, NULL);
+	hclgevf_enable_vector(&hdev->misc_vector, true);
 
-	while (!test_bit(HNAE3_FLR_DOWN, &hdev->flr_state) &&
-	       cnt++ < HCLGEVF_FLR_WAIT_CNT)
-		msleep(HCLGEVF_FLR_WAIT_MS);
+	ret = hclgevf_reset_rebuild(hdev);
+	if (ret)
+		dev_warn(&hdev->pdev->dev, "fail to rebuild, ret=%d\n",
+			 ret);
 
-	if (!test_bit(HNAE3_FLR_DOWN, &hdev->flr_state))
-		dev_err(&hdev->pdev->dev,
-			"flr wait down timeout: %d\n", cnt);
+	hdev->reset_type = HNAE3_NONE_RESET;
+	clear_bit(HCLGEVF_STATE_RST_HANDLING, &hdev->state);
+	up(&hdev->reset_sem);
 }
 
 static u32 hclgevf_get_fw_version(struct hnae3_handle *handle)
@@ -1808,8 +1805,8 @@ static void hclgevf_reset_service_task(struct hclgevf_dev *hdev)
 	if (!test_and_clear_bit(HCLGEVF_STATE_RST_SERVICE_SCHED, &hdev->state))
 		return;
 
-	if (test_and_set_bit(HCLGEVF_STATE_RST_HANDLING, &hdev->state))
-		return;
+	down(&hdev->reset_sem);
+	set_bit(HCLGEVF_STATE_RST_HANDLING, &hdev->state);
 
 	if (test_and_clear_bit(HCLGEVF_RESET_PENDING,
 			       &hdev->reset_state)) {
@@ -1866,6 +1863,7 @@ static void hclgevf_reset_service_task(struct hclgevf_dev *hdev)
 	}
 
 	clear_bit(HCLGEVF_STATE_RST_HANDLING, &hdev->state);
+	up(&hdev->reset_sem);
 }
 
 static void hclgevf_mailbox_service_task(struct hclgevf_dev *hdev)
@@ -2009,11 +2007,6 @@ static enum hclgevf_evt_cause hclgevf_check_evt_cause(struct hclgevf_dev *hdev,
 	return HCLGEVF_VECTOR0_EVENT_OTHER;
 }
 
-static void hclgevf_enable_vector(struct hclgevf_misc_vector *vector, bool en)
-{
-	writel(en ? 1 : 0, vector->addr);
-}
-
 static irqreturn_t hclgevf_misc_irq_handle(int irq, void *data)
 {
 	enum hclgevf_evt_cause event_cause;
@@ -2288,6 +2281,7 @@ static void hclgevf_state_init(struct hclgevf_dev *hdev)
 	INIT_DELAYED_WORK(&hdev->service_task, hclgevf_service_task);
 
 	mutex_init(&hdev->mbx_resp.mbx_mutex);
+	sema_init(&hdev->reset_sem, 1);
 
 	/* bring the device down */
 	set_bit(HCLGEVF_STATE_DOWN, &hdev->state);
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h
index 2cbc7dfe2f91..fee8d97f323c 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h
@@ -253,6 +253,7 @@ struct hclgevf_dev {
 	unsigned long reset_state;	/* requested, pending */
 	struct hclgevf_rst_stats rst_stats;
 	u32 reset_attempts;
+	struct semaphore reset_sem;	/* protect reset process */
 
 	u32 fw_version;
 	u16 num_tqps;		/* num task queue pairs of this PF */
-- 
cgit v1.2.3-59-g8ed1b


From 5bb784e973cf35f9883b10bb7b1588616c7ed6b1 Mon Sep 17 00:00:00 2001
From: Huazhong Tan <tanhuazhong@huawei.com>
Date: Sat, 11 Jan 2020 16:33:51 +0800
Subject: net: hns3: enlarge HCLGE_RESET_WAIT_CNT

When the load of firmware is high, its reset task may takes
more time(which will be as long as 35 seconds). So this
patch modifies HCLGE_RESET_WAIT_CNT to match the firmware's.

Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
index 89621b587603..d81c97608726 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
@@ -3260,7 +3260,8 @@ static int hclge_notify_roce_client(struct hclge_dev *hdev,
 static int hclge_reset_wait(struct hclge_dev *hdev)
 {
 #define HCLGE_RESET_WATI_MS	100
-#define HCLGE_RESET_WAIT_CNT	200
+#define HCLGE_RESET_WAIT_CNT	350
+
 	u32 val, reg, reg_bit;
 	u32 cnt = 0;
 
-- 
cgit v1.2.3-59-g8ed1b


From c3106cac2a80381e860adbd416e166f778b8806a Mon Sep 17 00:00:00 2001
From: Huazhong Tan <tanhuazhong@huawei.com>
Date: Sat, 11 Jan 2020 16:33:52 +0800
Subject: net: hns3: modify hclge_func_reset_sync_vf()'s return type to void

When synchronizes with VFs fail before PF function reset,
PF driver should go on its function reset, otherwise it
can not run normally anymore. So, hclge_func_reset_sync_vf()
should not affect the processing of PF reset, this patch
modifies its return type to void.

Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../ethernet/hisilicon/hns3/hns3pf/hclge_main.c    | 29 +++++++---------------
 1 file changed, 9 insertions(+), 20 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
index d81c97608726..21ecd797d450 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
@@ -3362,7 +3362,7 @@ static void hclge_mailbox_service_task(struct hclge_dev *hdev)
 	clear_bit(HCLGE_STATE_MBX_HANDLING, &hdev->state);
 }
 
-static int hclge_func_reset_sync_vf(struct hclge_dev *hdev)
+static void hclge_func_reset_sync_vf(struct hclge_dev *hdev)
 {
 	struct hclge_pf_rst_sync_cmd *req;
 	struct hclge_desc desc;
@@ -3382,20 +3382,19 @@ static int hclge_func_reset_sync_vf(struct hclge_dev *hdev)
 		 */
 		if (ret == -EOPNOTSUPP) {
 			msleep(HCLGE_RESET_SYNC_TIME);
-			return 0;
+			return;
 		} else if (ret) {
-			dev_err(&hdev->pdev->dev, "sync with VF fail %d!\n",
-				ret);
-			return ret;
+			dev_warn(&hdev->pdev->dev, "sync with VF fail %d!\n",
+				 ret);
+			return;
 		} else if (req->all_vf_ready) {
-			return 0;
+			return;
 		}
 		msleep(HCLGE_PF_RESET_SYNC_TIME);
 		hclge_cmd_reuse_desc(&desc, true);
 	} while (cnt++ < HCLGE_PF_RESET_SYNC_CNT);
 
-	dev_err(&hdev->pdev->dev, "sync with VF timeout!\n");
-	return -ETIME;
+	dev_warn(&hdev->pdev->dev, "sync with VF timeout!\n");
 }
 
 void hclge_report_hw_error(struct hclge_dev *hdev,
@@ -3600,12 +3599,7 @@ static int hclge_reset_prepare_wait(struct hclge_dev *hdev)
 
 	switch (hdev->reset_type) {
 	case HNAE3_FUNC_RESET:
-		/* to confirm whether all running VF is ready
-		 * before request PF reset
-		 */
-		ret = hclge_func_reset_sync_vf(hdev);
-		if (ret)
-			return ret;
+		hclge_func_reset_sync_vf(hdev);
 
 		ret = hclge_func_reset_cmd(hdev, 0);
 		if (ret) {
@@ -3623,12 +3617,7 @@ static int hclge_reset_prepare_wait(struct hclge_dev *hdev)
 		hdev->rst_stats.pf_rst_cnt++;
 		break;
 	case HNAE3_FLR_RESET:
-		/* to confirm whether all running VF is ready
-		 * before request PF reset
-		 */
-		ret = hclge_func_reset_sync_vf(hdev);
-		if (ret)
-			return ret;
+		hclge_func_reset_sync_vf(hdev);
 		break;
 	case HNAE3_IMP_RESET:
 		hclge_handle_imp_error(hdev);
-- 
cgit v1.2.3-59-g8ed1b


From c7554dcd6dbff3d1b295062649df4bce410bf50e Mon Sep 17 00:00:00 2001
From: Huazhong Tan <tanhuazhong@huawei.com>
Date: Sat, 11 Jan 2020 16:33:53 +0800
Subject: net: hns3: refactor the notification scheme of PF reset

hclge_reset_prepare_down() is only used to inform VF that PF is
going to do function reset, then using hclge_func_reset_sync_vf()
in hclge_reset_prepare_wait() to query whether VF is ready before
asserting PF function reset. To make the code more readable,
this patch uses a new function hclge_function_reset_notify_vf()
to do this job.

Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../ethernet/hisilicon/hns3/hns3pf/hclge_main.c    | 42 ++++++++++------------
 1 file changed, 19 insertions(+), 23 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
index 21ecd797d450..76e8aa40e1bc 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
@@ -3562,23 +3562,6 @@ static void hclge_clear_reset_cause(struct hclge_dev *hdev)
 	hclge_enable_vector(&hdev->misc_vector, true);
 }
 
-static int hclge_reset_prepare_down(struct hclge_dev *hdev)
-{
-	int ret = 0;
-
-	switch (hdev->reset_type) {
-	case HNAE3_FUNC_RESET:
-		/* fall through */
-	case HNAE3_FLR_RESET:
-		ret = hclge_set_all_vf_rst(hdev, true);
-		break;
-	default:
-		break;
-	}
-
-	return ret;
-}
-
 static void hclge_reset_handshake(struct hclge_dev *hdev, bool enable)
 {
 	u32 reg_val;
@@ -3592,6 +3575,19 @@ static void hclge_reset_handshake(struct hclge_dev *hdev, bool enable)
 	hclge_write_dev(&hdev->hw, HCLGE_NIC_CSQ_DEPTH_REG, reg_val);
 }
 
+static int hclge_func_reset_notify_vf(struct hclge_dev *hdev)
+{
+	int ret;
+
+	ret = hclge_set_all_vf_rst(hdev, true);
+	if (ret)
+		return ret;
+
+	hclge_func_reset_sync_vf(hdev);
+
+	return 0;
+}
+
 static int hclge_reset_prepare_wait(struct hclge_dev *hdev)
 {
 	u32 reg_val;
@@ -3599,7 +3595,9 @@ static int hclge_reset_prepare_wait(struct hclge_dev *hdev)
 
 	switch (hdev->reset_type) {
 	case HNAE3_FUNC_RESET:
-		hclge_func_reset_sync_vf(hdev);
+		ret = hclge_func_reset_notify_vf(hdev);
+		if (ret)
+			return ret;
 
 		ret = hclge_func_reset_cmd(hdev, 0);
 		if (ret) {
@@ -3617,7 +3615,9 @@ static int hclge_reset_prepare_wait(struct hclge_dev *hdev)
 		hdev->rst_stats.pf_rst_cnt++;
 		break;
 	case HNAE3_FLR_RESET:
-		hclge_func_reset_sync_vf(hdev);
+		ret = hclge_func_reset_notify_vf(hdev);
+		if (ret)
+			return ret;
 		break;
 	case HNAE3_IMP_RESET:
 		hclge_handle_imp_error(hdev);
@@ -3761,10 +3761,6 @@ static int hclge_reset_prepare(struct hclge_dev *hdev)
 	if (ret)
 		return ret;
 
-	ret = hclge_reset_prepare_down(hdev);
-	if (ret)
-		return ret;
-
 	rtnl_lock();
 	ret = hclge_notify_client(hdev, HNAE3_DOWN_CLIENT);
 	rtnl_unlock();
-- 
cgit v1.2.3-59-g8ed1b


From a442c2c3850dc308ab972f3d10d1077e2c8fd035 Mon Sep 17 00:00:00 2001
From: Jonathan Lemon <jonathan.lemon@gmail.com>
Date: Thu, 9 Jan 2020 11:23:17 -0800
Subject: mlx4: Bump up MAX_MSIX from 64 to 128

On modern hardware with a large number of cpus and using XDP,
the current MSIX limit is insufficient.  Bump the limit in
order to allow more queues.

Signed-off-by: Jonathan Lemon <jonathan.lemon@gmail.com>
Reviewed-by: Jack Wang <jinpu.wang@cloud.ionos.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/mlx4/device.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index 36e412c3d657..20372de0b587 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -47,7 +47,7 @@
 #define DEFAULT_UAR_PAGE_SHIFT  12
 
 #define MAX_MSIX_P_PORT		17
-#define MAX_MSIX		64
+#define MAX_MSIX		128
 #define MIN_MSIX_P_PORT		5
 #define MLX4_IS_LEGACY_EQ_MODE(dev_cap) ((dev_cap).num_comp_vectors < \
 					 (dev_cap).num_ports * MIN_MSIX_P_PORT)
-- 
cgit v1.2.3-59-g8ed1b


From 504c28c853ec5c626900b914b5833daf0581a344 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Sun, 12 Jan 2020 13:04:42 +0100
Subject: wan: ixp4xx_hss: fix compile-testing on 64-bit

Change the driver to use portable integer types to avoid
warnings during compile testing:

drivers/net/wan/ixp4xx_hss.c:863:21: error: cast to 'u32 *' (aka 'unsigned int *') from smaller integer type 'int' [-Werror,-Wint-to-pointer-cast]
        memcpy_swab32(mem, (u32 *)((int)skb->data & ~3), bytes / 4);
                           ^
drivers/net/wan/ixp4xx_hss.c:979:12: error: incompatible pointer types passing 'u32 *' (aka 'unsigned int *') to parameter of type 'dma_addr_t *' (aka 'unsigned long long *') [-Werror,-Wincompatible-pointer-types]
                                              &port->desc_tab_phys)))
                                              ^~~~~~~~~~~~~~~~~~~~
include/linux/dmapool.h:27:20: note: passing argument to parameter 'handle' here
                     dma_addr_t *handle);
                                 ^

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/wan/ixp4xx_hss.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wan/ixp4xx_hss.c b/drivers/net/wan/ixp4xx_hss.c
index ea6ee6a608ce..e7619cec978a 100644
--- a/drivers/net/wan/ixp4xx_hss.c
+++ b/drivers/net/wan/ixp4xx_hss.c
@@ -258,7 +258,7 @@ struct port {
 	struct hss_plat_info *plat;
 	buffer_t *rx_buff_tab[RX_DESCS], *tx_buff_tab[TX_DESCS];
 	struct desc *desc_tab;	/* coherent */
-	u32 desc_tab_phys;
+	dma_addr_t desc_tab_phys;
 	unsigned int id;
 	unsigned int clock_type, clock_rate, loopback;
 	unsigned int initialized, carrier;
@@ -858,7 +858,7 @@ static int hss_hdlc_xmit(struct sk_buff *skb, struct net_device *dev)
 		dev->stats.tx_dropped++;
 		return NETDEV_TX_OK;
 	}
-	memcpy_swab32(mem, (u32 *)((int)skb->data & ~3), bytes / 4);
+	memcpy_swab32(mem, (u32 *)((uintptr_t)skb->data & ~3), bytes / 4);
 	dev_kfree_skb(skb);
 #endif
 
-- 
cgit v1.2.3-59-g8ed1b


From c74f16b6034401b17bb1cf549871186a8ece5f92 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Sun, 12 Jan 2020 13:04:43 +0100
Subject: wan: ixp4xx_hss: prepare compile testing

The ixp4xx_hss driver needs the platform data definition and the
system clock rate to be compiled. Move both into a new platform_data
header file.

This is a prerequisite for compile testing, but turning on compile
testing requires further patches to isolate the SoC headers.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 arch/arm/mach-ixp4xx/goramo_mlr.c            |  4 ++++
 arch/arm/mach-ixp4xx/include/mach/platform.h |  9 -------
 drivers/net/wan/Kconfig                      |  3 ++-
 drivers/net/wan/ixp4xx_hss.c                 | 35 ++++++++++++++++------------
 include/linux/platform_data/wan_ixp4xx_hss.h | 17 ++++++++++++++
 5 files changed, 43 insertions(+), 25 deletions(-)
 create mode 100644 include/linux/platform_data/wan_ixp4xx_hss.h

diff --git a/arch/arm/mach-ixp4xx/goramo_mlr.c b/arch/arm/mach-ixp4xx/goramo_mlr.c
index a0e0b6b7dc5c..93b7afeee142 100644
--- a/arch/arm/mach-ixp4xx/goramo_mlr.c
+++ b/arch/arm/mach-ixp4xx/goramo_mlr.c
@@ -11,6 +11,7 @@
 #include <linux/irq.h>
 #include <linux/kernel.h>
 #include <linux/pci.h>
+#include <linux/platform_data/wan_ixp4xx_hss.h>
 #include <linux/serial_8250.h>
 #include <asm/mach-types.h>
 #include <asm/mach/arch.h>
@@ -405,6 +406,9 @@ static void __init gmlr_init(void)
 	if (hw_bits & CFG_HW_HAS_HSS1)
 		device_tab[devices++] = &device_hss_tab[1]; /* max index 5 */
 
+	hss_plat[0].timer_freq = ixp4xx_timer_freq;
+	hss_plat[1].timer_freq = ixp4xx_timer_freq;
+
 	gpio_request(GPIO_SCL, "SCL/clock");
 	gpio_request(GPIO_SDA, "SDA/data");
 	gpio_request(GPIO_STR, "strobe");
diff --git a/arch/arm/mach-ixp4xx/include/mach/platform.h b/arch/arm/mach-ixp4xx/include/mach/platform.h
index 342acbe20f7c..04ef8025accc 100644
--- a/arch/arm/mach-ixp4xx/include/mach/platform.h
+++ b/arch/arm/mach-ixp4xx/include/mach/platform.h
@@ -104,15 +104,6 @@ struct eth_plat_info {
 	u8 hwaddr[6];
 };
 
-/* Information about built-in HSS (synchronous serial) interfaces */
-struct hss_plat_info {
-	int (*set_clock)(int port, unsigned int clock_type);
-	int (*open)(int port, void *pdev,
-		    void (*set_carrier_cb)(void *pdev, int carrier));
-	void (*close)(int port, void *pdev);
-	u8 txreadyq;
-};
-
 /*
  * Frequency of clock used for primary clocksource
  */
diff --git a/drivers/net/wan/Kconfig b/drivers/net/wan/Kconfig
index dd1a147f2971..4530840e15ef 100644
--- a/drivers/net/wan/Kconfig
+++ b/drivers/net/wan/Kconfig
@@ -315,7 +315,8 @@ config DSCC4_PCI_RST
 
 config IXP4XX_HSS
 	tristate "Intel IXP4xx HSS (synchronous serial port) support"
-	depends on HDLC && ARM && ARCH_IXP4XX && IXP4XX_NPE && IXP4XX_QMGR
+	depends on HDLC && IXP4XX_NPE && IXP4XX_QMGR
+	depends on ARCH_IXP4XX
 	help
 	  Say Y here if you want to use built-in HSS ports
 	  on IXP4xx processor.
diff --git a/drivers/net/wan/ixp4xx_hss.c b/drivers/net/wan/ixp4xx_hss.c
index e7619cec978a..7c5cf77e9ef1 100644
--- a/drivers/net/wan/ixp4xx_hss.c
+++ b/drivers/net/wan/ixp4xx_hss.c
@@ -17,6 +17,7 @@
 #include <linux/io.h>
 #include <linux/kernel.h>
 #include <linux/platform_device.h>
+#include <linux/platform_data/wan_ixp4xx_hss.h>
 #include <linux/poll.h>
 #include <linux/slab.h>
 #include <linux/soc/ixp4xx/npe.h>
@@ -1182,14 +1183,14 @@ static int hss_hdlc_attach(struct net_device *dev, unsigned short encoding,
 	}
 }
 
-static u32 check_clock(u32 rate, u32 a, u32 b, u32 c,
+static u32 check_clock(u32 timer_freq, u32 rate, u32 a, u32 b, u32 c,
 		       u32 *best, u32 *best_diff, u32 *reg)
 {
 	/* a is 10-bit, b is 10-bit, c is 12-bit */
 	u64 new_rate;
 	u32 new_diff;
 
-	new_rate = ixp4xx_timer_freq * (u64)(c + 1);
+	new_rate = timer_freq * (u64)(c + 1);
 	do_div(new_rate, a * (c + 1) + b + 1);
 	new_diff = abs((u32)new_rate - rate);
 
@@ -1201,40 +1202,43 @@ static u32 check_clock(u32 rate, u32 a, u32 b, u32 c,
 	return new_diff;
 }
 
-static void find_best_clock(u32 rate, u32 *best, u32 *reg)
+static void find_best_clock(u32 timer_freq, u32 rate, u32 *best, u32 *reg)
 {
 	u32 a, b, diff = 0xFFFFFFFF;
 
-	a = ixp4xx_timer_freq / rate;
+	a = timer_freq / rate;
 
 	if (a > 0x3FF) { /* 10-bit value - we can go as slow as ca. 65 kb/s */
-		check_clock(rate, 0x3FF, 1, 1, best, &diff, reg);
+		check_clock(timer_freq, rate, 0x3FF, 1, 1, best, &diff, reg);
 		return;
 	}
 	if (a == 0) { /* > 66.666 MHz */
 		a = 1; /* minimum divider is 1 (a = 0, b = 1, c = 1) */
-		rate = ixp4xx_timer_freq;
+		rate = timer_freq;
 	}
 
-	if (rate * a == ixp4xx_timer_freq) { /* don't divide by 0 later */
-		check_clock(rate, a - 1, 1, 1, best, &diff, reg);
+	if (rate * a == timer_freq) { /* don't divide by 0 later */
+		check_clock(timer_freq, rate, a - 1, 1, 1, best, &diff, reg);
 		return;
 	}
 
 	for (b = 0; b < 0x400; b++) {
 		u64 c = (b + 1) * (u64)rate;
-		do_div(c, ixp4xx_timer_freq - rate * a);
+		do_div(c, timer_freq - rate * a);
 		c--;
 		if (c >= 0xFFF) { /* 12-bit - no need to check more 'b's */
 			if (b == 0 && /* also try a bit higher rate */
-			    !check_clock(rate, a - 1, 1, 1, best, &diff, reg))
+			    !check_clock(timer_freq, rate, a - 1, 1, 1, best,
+					 &diff, reg))
 				return;
-			check_clock(rate, a, b, 0xFFF, best, &diff, reg);
+			check_clock(timer_freq, rate, a, b, 0xFFF, best,
+				    &diff, reg);
 			return;
 		}
-		if (!check_clock(rate, a, b, c, best, &diff, reg))
+		if (!check_clock(timer_freq, rate, a, b, c, best, &diff, reg))
 			return;
-		if (!check_clock(rate, a, b, c + 1, best, &diff, reg))
+		if (!check_clock(timer_freq, rate, a, b, c + 1, best, &diff,
+				 reg))
 			return;
 	}
 }
@@ -1285,8 +1289,9 @@ static int hss_hdlc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 
 		port->clock_type = clk; /* Update settings */
 		if (clk == CLOCK_INT)
-			find_best_clock(new_line.clock_rate, &port->clock_rate,
-					&port->clock_reg);
+			find_best_clock(port->plat->timer_freq,
+					new_line.clock_rate,
+					&port->clock_rate, &port->clock_reg);
 		else {
 			port->clock_rate = 0;
 			port->clock_reg = CLK42X_SPEED_2048KHZ;
diff --git a/include/linux/platform_data/wan_ixp4xx_hss.h b/include/linux/platform_data/wan_ixp4xx_hss.h
new file mode 100644
index 000000000000..d525a0feb9e1
--- /dev/null
+++ b/include/linux/platform_data/wan_ixp4xx_hss.h
@@ -0,0 +1,17 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __PLATFORM_DATA_WAN_IXP4XX_HSS_H
+#define __PLATFORM_DATA_WAN_IXP4XX_HSS_H
+
+#include <linux/types.h>
+
+/* Information about built-in HSS (synchronous serial) interfaces */
+struct hss_plat_info {
+	int (*set_clock)(int port, unsigned int clock_type);
+	int (*open)(int port, void *pdev,
+		    void (*set_carrier_cb)(void *pdev, int carrier));
+	void (*close)(int port, void *pdev);
+	u8 txreadyq;
+	u32 timer_freq;
+};
+
+#endif
-- 
cgit v1.2.3-59-g8ed1b


From 2785543fbf0498eb749adee3d129ba8ac139c729 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Sun, 12 Jan 2020 13:04:44 +0100
Subject: ptp: ixp46x: move adjacent to ethernet driver

The ixp46x ptp driver has a somewhat unusual setup, where the ptp
driver and the ethernet driver are in different directories but
access the same registers that are defined a platform specific
header file.

Moving everything into drivers/net/ makes it look more like most
other ptp drivers and allows compile-testing this driver on
other targets.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 arch/arm/mach-ixp4xx/include/mach/ixp46x_ts.h |  68 ------
 drivers/net/ethernet/xscale/Kconfig           |  14 ++
 drivers/net/ethernet/xscale/Makefile          |   3 +-
 drivers/net/ethernet/xscale/ixp4xx_eth.c      |   3 +-
 drivers/net/ethernet/xscale/ptp_ixp46x.c      | 329 ++++++++++++++++++++++++++
 drivers/net/ethernet/xscale/ptp_ixp46x.h      |  68 ++++++
 drivers/ptp/Kconfig                           |  14 --
 drivers/ptp/Makefile                          |   3 +-
 drivers/ptp/ptp_ixp46x.c                      | 328 -------------------------
 9 files changed, 416 insertions(+), 414 deletions(-)
 delete mode 100644 arch/arm/mach-ixp4xx/include/mach/ixp46x_ts.h
 create mode 100644 drivers/net/ethernet/xscale/ptp_ixp46x.c
 create mode 100644 drivers/net/ethernet/xscale/ptp_ixp46x.h
 delete mode 100644 drivers/ptp/ptp_ixp46x.c

diff --git a/arch/arm/mach-ixp4xx/include/mach/ixp46x_ts.h b/arch/arm/mach-ixp4xx/include/mach/ixp46x_ts.h
deleted file mode 100644
index d792130e27b0..000000000000
--- a/arch/arm/mach-ixp4xx/include/mach/ixp46x_ts.h
+++ /dev/null
@@ -1,68 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0-or-later */
-/*
- * PTP 1588 clock using the IXP46X
- *
- * Copyright (C) 2010 OMICRON electronics GmbH
- */
-
-#ifndef _IXP46X_TS_H_
-#define _IXP46X_TS_H_
-
-#define DEFAULT_ADDEND 0xF0000029
-#define TICKS_NS_SHIFT 4
-
-struct ixp46x_channel_ctl {
-	u32 ch_control;  /* 0x40 Time Synchronization Channel Control */
-	u32 ch_event;    /* 0x44 Time Synchronization Channel Event */
-	u32 tx_snap_lo;  /* 0x48 Transmit Snapshot Low Register */
-	u32 tx_snap_hi;  /* 0x4C Transmit Snapshot High Register */
-	u32 rx_snap_lo;  /* 0x50 Receive Snapshot Low Register */
-	u32 rx_snap_hi;  /* 0x54 Receive Snapshot High Register */
-	u32 src_uuid_lo; /* 0x58 Source UUID0 Low Register */
-	u32 src_uuid_hi; /* 0x5C Sequence Identifier/Source UUID0 High */
-};
-
-struct ixp46x_ts_regs {
-	u32 control;     /* 0x00 Time Sync Control Register */
-	u32 event;       /* 0x04 Time Sync Event Register */
-	u32 addend;      /* 0x08 Time Sync Addend Register */
-	u32 accum;       /* 0x0C Time Sync Accumulator Register */
-	u32 test;        /* 0x10 Time Sync Test Register */
-	u32 unused;      /* 0x14 */
-	u32 rsystime_lo; /* 0x18 RawSystemTime_Low Register */
-	u32 rsystime_hi; /* 0x1C RawSystemTime_High Register */
-	u32 systime_lo;  /* 0x20 SystemTime_Low Register */
-	u32 systime_hi;  /* 0x24 SystemTime_High Register */
-	u32 trgt_lo;     /* 0x28 TargetTime_Low Register */
-	u32 trgt_hi;     /* 0x2C TargetTime_High Register */
-	u32 asms_lo;     /* 0x30 Auxiliary Slave Mode Snapshot Low  */
-	u32 asms_hi;     /* 0x34 Auxiliary Slave Mode Snapshot High */
-	u32 amms_lo;     /* 0x38 Auxiliary Master Mode Snapshot Low */
-	u32 amms_hi;     /* 0x3C Auxiliary Master Mode Snapshot High */
-
-	struct ixp46x_channel_ctl channel[3];
-};
-
-/* 0x00 Time Sync Control Register Bits */
-#define TSCR_AMM (1<<3)
-#define TSCR_ASM (1<<2)
-#define TSCR_TTM (1<<1)
-#define TSCR_RST (1<<0)
-
-/* 0x04 Time Sync Event Register Bits */
-#define TSER_SNM (1<<3)
-#define TSER_SNS (1<<2)
-#define TTIPEND  (1<<1)
-
-/* 0x40 Time Synchronization Channel Control Register Bits */
-#define MASTER_MODE   (1<<0)
-#define TIMESTAMP_ALL (1<<1)
-
-/* 0x44 Time Synchronization Channel Event Register Bits */
-#define TX_SNAPSHOT_LOCKED (1<<0)
-#define RX_SNAPSHOT_LOCKED (1<<1)
-
-/* The ptp_ixp46x module will set this variable */
-extern int ixp46x_phc_index;
-
-#endif
diff --git a/drivers/net/ethernet/xscale/Kconfig b/drivers/net/ethernet/xscale/Kconfig
index cd0a8f46e7c6..98aa7b8ddb06 100644
--- a/drivers/net/ethernet/xscale/Kconfig
+++ b/drivers/net/ethernet/xscale/Kconfig
@@ -27,4 +27,18 @@ config IXP4XX_ETH
 	  Say Y here if you want to use built-in Ethernet ports
 	  on IXP4xx processor.
 
+config PTP_1588_CLOCK_IXP46X
+	tristate "Intel IXP46x as PTP clock"
+	depends on IXP4XX_ETH
+	depends on PTP_1588_CLOCK
+	default y
+	help
+	  This driver adds support for using the IXP46X as a PTP
+	  clock. This clock is only useful if your PTP programs are
+	  getting hardware time stamps on the PTP Ethernet packets
+	  using the SO_TIMESTAMPING API.
+
+	  To compile this driver as a module, choose M here: the module
+	  will be called ptp_ixp46x.
+
 endif # NET_VENDOR_XSCALE
diff --git a/drivers/net/ethernet/xscale/Makefile b/drivers/net/ethernet/xscale/Makefile
index 794a519d07b3..607f91b1e878 100644
--- a/drivers/net/ethernet/xscale/Makefile
+++ b/drivers/net/ethernet/xscale/Makefile
@@ -3,4 +3,5 @@
 # Makefile for the Intel XScale IXP device drivers.
 #
 
-obj-$(CONFIG_IXP4XX_ETH) += ixp4xx_eth.o
+obj-$(CONFIG_IXP4XX_ETH)		+= ixp4xx_eth.o
+obj-$(CONFIG_PTP_1588_CLOCK_IXP46X)	+= ptp_ixp46x.o
diff --git a/drivers/net/ethernet/xscale/ixp4xx_eth.c b/drivers/net/ethernet/xscale/ixp4xx_eth.c
index 6fc04ffb22c2..0075ecdb21f4 100644
--- a/drivers/net/ethernet/xscale/ixp4xx_eth.c
+++ b/drivers/net/ethernet/xscale/ixp4xx_eth.c
@@ -33,10 +33,11 @@
 #include <linux/ptp_classify.h>
 #include <linux/slab.h>
 #include <linux/module.h>
-#include <mach/ixp46x_ts.h>
 #include <linux/soc/ixp4xx/npe.h>
 #include <linux/soc/ixp4xx/qmgr.h>
 
+#include "ixp46x_ts.h"
+
 #define DEBUG_DESC		0
 #define DEBUG_RX		0
 #define DEBUG_TX		0
diff --git a/drivers/net/ethernet/xscale/ptp_ixp46x.c b/drivers/net/ethernet/xscale/ptp_ixp46x.c
new file mode 100644
index 000000000000..9ecc395239e9
--- /dev/null
+++ b/drivers/net/ethernet/xscale/ptp_ixp46x.c
@@ -0,0 +1,329 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * PTP 1588 clock using the IXP46X
+ *
+ * Copyright (C) 2010 OMICRON electronics GmbH
+ */
+#include <linux/device.h>
+#include <linux/err.h>
+#include <linux/gpio.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/irq.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+
+#include <linux/ptp_clock_kernel.h>
+
+#include "ixp46x_ts.h"
+
+#define DRIVER		"ptp_ixp46x"
+#define N_EXT_TS	2
+#define MASTER_GPIO	8
+#define MASTER_IRQ	25
+#define SLAVE_GPIO	7
+#define SLAVE_IRQ	24
+
+struct ixp_clock {
+	struct ixp46x_ts_regs *regs;
+	struct ptp_clock *ptp_clock;
+	struct ptp_clock_info caps;
+	int exts0_enabled;
+	int exts1_enabled;
+};
+
+DEFINE_SPINLOCK(register_lock);
+
+/*
+ * Register access functions
+ */
+
+static u64 ixp_systime_read(struct ixp46x_ts_regs *regs)
+{
+	u64 ns;
+	u32 lo, hi;
+
+	lo = __raw_readl(&regs->systime_lo);
+	hi = __raw_readl(&regs->systime_hi);
+
+	ns = ((u64) hi) << 32;
+	ns |= lo;
+	ns <<= TICKS_NS_SHIFT;
+
+	return ns;
+}
+
+static void ixp_systime_write(struct ixp46x_ts_regs *regs, u64 ns)
+{
+	u32 hi, lo;
+
+	ns >>= TICKS_NS_SHIFT;
+	hi = ns >> 32;
+	lo = ns & 0xffffffff;
+
+	__raw_writel(lo, &regs->systime_lo);
+	__raw_writel(hi, &regs->systime_hi);
+}
+
+/*
+ * Interrupt service routine
+ */
+
+static irqreturn_t isr(int irq, void *priv)
+{
+	struct ixp_clock *ixp_clock = priv;
+	struct ixp46x_ts_regs *regs = ixp_clock->regs;
+	struct ptp_clock_event event;
+	u32 ack = 0, lo, hi, val;
+
+	val = __raw_readl(&regs->event);
+
+	if (val & TSER_SNS) {
+		ack |= TSER_SNS;
+		if (ixp_clock->exts0_enabled) {
+			hi = __raw_readl(&regs->asms_hi);
+			lo = __raw_readl(&regs->asms_lo);
+			event.type = PTP_CLOCK_EXTTS;
+			event.index = 0;
+			event.timestamp = ((u64) hi) << 32;
+			event.timestamp |= lo;
+			event.timestamp <<= TICKS_NS_SHIFT;
+			ptp_clock_event(ixp_clock->ptp_clock, &event);
+		}
+	}
+
+	if (val & TSER_SNM) {
+		ack |= TSER_SNM;
+		if (ixp_clock->exts1_enabled) {
+			hi = __raw_readl(&regs->amms_hi);
+			lo = __raw_readl(&regs->amms_lo);
+			event.type = PTP_CLOCK_EXTTS;
+			event.index = 1;
+			event.timestamp = ((u64) hi) << 32;
+			event.timestamp |= lo;
+			event.timestamp <<= TICKS_NS_SHIFT;
+			ptp_clock_event(ixp_clock->ptp_clock, &event);
+		}
+	}
+
+	if (val & TTIPEND)
+		ack |= TTIPEND; /* this bit seems to be always set */
+
+	if (ack) {
+		__raw_writel(ack, &regs->event);
+		return IRQ_HANDLED;
+	} else
+		return IRQ_NONE;
+}
+
+/*
+ * PTP clock operations
+ */
+
+static int ptp_ixp_adjfreq(struct ptp_clock_info *ptp, s32 ppb)
+{
+	u64 adj;
+	u32 diff, addend;
+	int neg_adj = 0;
+	struct ixp_clock *ixp_clock = container_of(ptp, struct ixp_clock, caps);
+	struct ixp46x_ts_regs *regs = ixp_clock->regs;
+
+	if (ppb < 0) {
+		neg_adj = 1;
+		ppb = -ppb;
+	}
+	addend = DEFAULT_ADDEND;
+	adj = addend;
+	adj *= ppb;
+	diff = div_u64(adj, 1000000000ULL);
+
+	addend = neg_adj ? addend - diff : addend + diff;
+
+	__raw_writel(addend, &regs->addend);
+
+	return 0;
+}
+
+static int ptp_ixp_adjtime(struct ptp_clock_info *ptp, s64 delta)
+{
+	s64 now;
+	unsigned long flags;
+	struct ixp_clock *ixp_clock = container_of(ptp, struct ixp_clock, caps);
+	struct ixp46x_ts_regs *regs = ixp_clock->regs;
+
+	spin_lock_irqsave(&register_lock, flags);
+
+	now = ixp_systime_read(regs);
+	now += delta;
+	ixp_systime_write(regs, now);
+
+	spin_unlock_irqrestore(&register_lock, flags);
+
+	return 0;
+}
+
+static int ptp_ixp_gettime(struct ptp_clock_info *ptp, struct timespec64 *ts)
+{
+	u64 ns;
+	unsigned long flags;
+	struct ixp_clock *ixp_clock = container_of(ptp, struct ixp_clock, caps);
+	struct ixp46x_ts_regs *regs = ixp_clock->regs;
+
+	spin_lock_irqsave(&register_lock, flags);
+
+	ns = ixp_systime_read(regs);
+
+	spin_unlock_irqrestore(&register_lock, flags);
+
+	*ts = ns_to_timespec64(ns);
+	return 0;
+}
+
+static int ptp_ixp_settime(struct ptp_clock_info *ptp,
+			   const struct timespec64 *ts)
+{
+	u64 ns;
+	unsigned long flags;
+	struct ixp_clock *ixp_clock = container_of(ptp, struct ixp_clock, caps);
+	struct ixp46x_ts_regs *regs = ixp_clock->regs;
+
+	ns = timespec64_to_ns(ts);
+
+	spin_lock_irqsave(&register_lock, flags);
+
+	ixp_systime_write(regs, ns);
+
+	spin_unlock_irqrestore(&register_lock, flags);
+
+	return 0;
+}
+
+static int ptp_ixp_enable(struct ptp_clock_info *ptp,
+			  struct ptp_clock_request *rq, int on)
+{
+	struct ixp_clock *ixp_clock = container_of(ptp, struct ixp_clock, caps);
+
+	switch (rq->type) {
+	case PTP_CLK_REQ_EXTTS:
+		switch (rq->extts.index) {
+		case 0:
+			ixp_clock->exts0_enabled = on ? 1 : 0;
+			break;
+		case 1:
+			ixp_clock->exts1_enabled = on ? 1 : 0;
+			break;
+		default:
+			return -EINVAL;
+		}
+		return 0;
+	default:
+		break;
+	}
+
+	return -EOPNOTSUPP;
+}
+
+static const struct ptp_clock_info ptp_ixp_caps = {
+	.owner		= THIS_MODULE,
+	.name		= "IXP46X timer",
+	.max_adj	= 66666655,
+	.n_ext_ts	= N_EXT_TS,
+	.n_pins		= 0,
+	.pps		= 0,
+	.adjfreq	= ptp_ixp_adjfreq,
+	.adjtime	= ptp_ixp_adjtime,
+	.gettime64	= ptp_ixp_gettime,
+	.settime64	= ptp_ixp_settime,
+	.enable		= ptp_ixp_enable,
+};
+
+/* module operations */
+
+static struct ixp_clock ixp_clock;
+
+static int setup_interrupt(int gpio)
+{
+	int irq;
+	int err;
+
+	err = gpio_request(gpio, "ixp4-ptp");
+	if (err)
+		return err;
+
+	err = gpio_direction_input(gpio);
+	if (err)
+		return err;
+
+	irq = gpio_to_irq(gpio);
+	if (irq < 0)
+		return irq;
+
+	err = irq_set_irq_type(irq, IRQF_TRIGGER_FALLING);
+	if (err) {
+		pr_err("cannot set trigger type for irq %d\n", irq);
+		return err;
+	}
+
+	err = request_irq(irq, isr, 0, DRIVER, &ixp_clock);
+	if (err) {
+		pr_err("request_irq failed for irq %d\n", irq);
+		return err;
+	}
+
+	return irq;
+}
+
+static void __exit ptp_ixp_exit(void)
+{
+	free_irq(MASTER_IRQ, &ixp_clock);
+	free_irq(SLAVE_IRQ, &ixp_clock);
+	ixp46x_phc_index = -1;
+	ptp_clock_unregister(ixp_clock.ptp_clock);
+}
+
+static int __init ptp_ixp_init(void)
+{
+	if (!cpu_is_ixp46x())
+		return -ENODEV;
+
+	ixp_clock.regs =
+		(struct ixp46x_ts_regs __iomem *) IXP4XX_TIMESYNC_BASE_VIRT;
+
+	ixp_clock.caps = ptp_ixp_caps;
+
+	ixp_clock.ptp_clock = ptp_clock_register(&ixp_clock.caps, NULL);
+
+	if (IS_ERR(ixp_clock.ptp_clock))
+		return PTR_ERR(ixp_clock.ptp_clock);
+
+	ixp46x_phc_index = ptp_clock_index(ixp_clock.ptp_clock);
+
+	__raw_writel(DEFAULT_ADDEND, &ixp_clock.regs->addend);
+	__raw_writel(1, &ixp_clock.regs->trgt_lo);
+	__raw_writel(0, &ixp_clock.regs->trgt_hi);
+	__raw_writel(TTIPEND, &ixp_clock.regs->event);
+
+	if (MASTER_IRQ != setup_interrupt(MASTER_GPIO)) {
+		pr_err("failed to setup gpio %d as irq\n", MASTER_GPIO);
+		goto no_master;
+	}
+	if (SLAVE_IRQ != setup_interrupt(SLAVE_GPIO)) {
+		pr_err("failed to setup gpio %d as irq\n", SLAVE_GPIO);
+		goto no_slave;
+	}
+
+	return 0;
+no_slave:
+	free_irq(MASTER_IRQ, &ixp_clock);
+no_master:
+	ptp_clock_unregister(ixp_clock.ptp_clock);
+	return -ENODEV;
+}
+
+module_init(ptp_ixp_init);
+module_exit(ptp_ixp_exit);
+
+MODULE_AUTHOR("Richard Cochran <richardcochran@gmail.com>");
+MODULE_DESCRIPTION("PTP clock using the IXP46X timer");
+MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/xscale/ptp_ixp46x.h b/drivers/net/ethernet/xscale/ptp_ixp46x.h
new file mode 100644
index 000000000000..d792130e27b0
--- /dev/null
+++ b/drivers/net/ethernet/xscale/ptp_ixp46x.h
@@ -0,0 +1,68 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * PTP 1588 clock using the IXP46X
+ *
+ * Copyright (C) 2010 OMICRON electronics GmbH
+ */
+
+#ifndef _IXP46X_TS_H_
+#define _IXP46X_TS_H_
+
+#define DEFAULT_ADDEND 0xF0000029
+#define TICKS_NS_SHIFT 4
+
+struct ixp46x_channel_ctl {
+	u32 ch_control;  /* 0x40 Time Synchronization Channel Control */
+	u32 ch_event;    /* 0x44 Time Synchronization Channel Event */
+	u32 tx_snap_lo;  /* 0x48 Transmit Snapshot Low Register */
+	u32 tx_snap_hi;  /* 0x4C Transmit Snapshot High Register */
+	u32 rx_snap_lo;  /* 0x50 Receive Snapshot Low Register */
+	u32 rx_snap_hi;  /* 0x54 Receive Snapshot High Register */
+	u32 src_uuid_lo; /* 0x58 Source UUID0 Low Register */
+	u32 src_uuid_hi; /* 0x5C Sequence Identifier/Source UUID0 High */
+};
+
+struct ixp46x_ts_regs {
+	u32 control;     /* 0x00 Time Sync Control Register */
+	u32 event;       /* 0x04 Time Sync Event Register */
+	u32 addend;      /* 0x08 Time Sync Addend Register */
+	u32 accum;       /* 0x0C Time Sync Accumulator Register */
+	u32 test;        /* 0x10 Time Sync Test Register */
+	u32 unused;      /* 0x14 */
+	u32 rsystime_lo; /* 0x18 RawSystemTime_Low Register */
+	u32 rsystime_hi; /* 0x1C RawSystemTime_High Register */
+	u32 systime_lo;  /* 0x20 SystemTime_Low Register */
+	u32 systime_hi;  /* 0x24 SystemTime_High Register */
+	u32 trgt_lo;     /* 0x28 TargetTime_Low Register */
+	u32 trgt_hi;     /* 0x2C TargetTime_High Register */
+	u32 asms_lo;     /* 0x30 Auxiliary Slave Mode Snapshot Low  */
+	u32 asms_hi;     /* 0x34 Auxiliary Slave Mode Snapshot High */
+	u32 amms_lo;     /* 0x38 Auxiliary Master Mode Snapshot Low */
+	u32 amms_hi;     /* 0x3C Auxiliary Master Mode Snapshot High */
+
+	struct ixp46x_channel_ctl channel[3];
+};
+
+/* 0x00 Time Sync Control Register Bits */
+#define TSCR_AMM (1<<3)
+#define TSCR_ASM (1<<2)
+#define TSCR_TTM (1<<1)
+#define TSCR_RST (1<<0)
+
+/* 0x04 Time Sync Event Register Bits */
+#define TSER_SNM (1<<3)
+#define TSER_SNS (1<<2)
+#define TTIPEND  (1<<1)
+
+/* 0x40 Time Synchronization Channel Control Register Bits */
+#define MASTER_MODE   (1<<0)
+#define TIMESTAMP_ALL (1<<1)
+
+/* 0x44 Time Synchronization Channel Event Register Bits */
+#define TX_SNAPSHOT_LOCKED (1<<0)
+#define RX_SNAPSHOT_LOCKED (1<<1)
+
+/* The ptp_ixp46x module will set this variable */
+extern int ixp46x_phc_index;
+
+#endif
diff --git a/drivers/ptp/Kconfig b/drivers/ptp/Kconfig
index c382158f587d..475c60dccaa4 100644
--- a/drivers/ptp/Kconfig
+++ b/drivers/ptp/Kconfig
@@ -56,20 +56,6 @@ config PTP_1588_CLOCK_QORIQ
 	  To compile this driver as a module, choose M here: the module
 	  will be called ptp-qoriq.
 
-config PTP_1588_CLOCK_IXP46X
-	tristate "Intel IXP46x as PTP clock"
-	depends on IXP4XX_ETH
-	depends on PTP_1588_CLOCK
-	default y
-	help
-	  This driver adds support for using the IXP46X as a PTP
-	  clock. This clock is only useful if your PTP programs are
-	  getting hardware time stamps on the PTP Ethernet packets
-	  using the SO_TIMESTAMPING API.
-
-	  To compile this driver as a module, choose M here: the module
-	  will be called ptp_ixp46x.
-
 comment "Enable PHYLIB and NETWORK_PHY_TIMESTAMPING to see the additional clocks."
 	depends on PHYLIB=n || NETWORK_PHY_TIMESTAMPING=n
 
diff --git a/drivers/ptp/Makefile b/drivers/ptp/Makefile
index 3fb91bebbaf7..8c830336f178 100644
--- a/drivers/ptp/Makefile
+++ b/drivers/ptp/Makefile
@@ -7,10 +7,9 @@ ptp-y					:= ptp_clock.o ptp_chardev.o ptp_sysfs.o
 obj-$(CONFIG_PTP_1588_CLOCK)		+= ptp.o
 obj-$(CONFIG_PTP_1588_CLOCK_DTE)	+= ptp_dte.o
 obj-$(CONFIG_PTP_1588_CLOCK_INES)	+= ptp_ines.o
-obj-$(CONFIG_PTP_1588_CLOCK_IXP46X)	+= ptp_ixp46x.o
 obj-$(CONFIG_PTP_1588_CLOCK_PCH)	+= ptp_pch.o
 obj-$(CONFIG_PTP_1588_CLOCK_KVM)	+= ptp_kvm.o
 obj-$(CONFIG_PTP_1588_CLOCK_QORIQ)	+= ptp-qoriq.o
 ptp-qoriq-y				+= ptp_qoriq.o
 ptp-qoriq-$(CONFIG_DEBUG_FS)		+= ptp_qoriq_debugfs.o
-obj-$(CONFIG_PTP_1588_CLOCK_IDTCM)	+= ptp_clockmatrix.o
\ No newline at end of file
+obj-$(CONFIG_PTP_1588_CLOCK_IDTCM)	+= ptp_clockmatrix.o
diff --git a/drivers/ptp/ptp_ixp46x.c b/drivers/ptp/ptp_ixp46x.c
deleted file mode 100644
index 67028484e9a0..000000000000
--- a/drivers/ptp/ptp_ixp46x.c
+++ /dev/null
@@ -1,328 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0-or-later
-/*
- * PTP 1588 clock using the IXP46X
- *
- * Copyright (C) 2010 OMICRON electronics GmbH
- */
-#include <linux/device.h>
-#include <linux/err.h>
-#include <linux/gpio.h>
-#include <linux/init.h>
-#include <linux/interrupt.h>
-#include <linux/io.h>
-#include <linux/irq.h>
-#include <linux/kernel.h>
-#include <linux/module.h>
-
-#include <linux/ptp_clock_kernel.h>
-#include <mach/ixp46x_ts.h>
-
-#define DRIVER		"ptp_ixp46x"
-#define N_EXT_TS	2
-#define MASTER_GPIO	8
-#define MASTER_IRQ	25
-#define SLAVE_GPIO	7
-#define SLAVE_IRQ	24
-
-struct ixp_clock {
-	struct ixp46x_ts_regs *regs;
-	struct ptp_clock *ptp_clock;
-	struct ptp_clock_info caps;
-	int exts0_enabled;
-	int exts1_enabled;
-};
-
-DEFINE_SPINLOCK(register_lock);
-
-/*
- * Register access functions
- */
-
-static u64 ixp_systime_read(struct ixp46x_ts_regs *regs)
-{
-	u64 ns;
-	u32 lo, hi;
-
-	lo = __raw_readl(&regs->systime_lo);
-	hi = __raw_readl(&regs->systime_hi);
-
-	ns = ((u64) hi) << 32;
-	ns |= lo;
-	ns <<= TICKS_NS_SHIFT;
-
-	return ns;
-}
-
-static void ixp_systime_write(struct ixp46x_ts_regs *regs, u64 ns)
-{
-	u32 hi, lo;
-
-	ns >>= TICKS_NS_SHIFT;
-	hi = ns >> 32;
-	lo = ns & 0xffffffff;
-
-	__raw_writel(lo, &regs->systime_lo);
-	__raw_writel(hi, &regs->systime_hi);
-}
-
-/*
- * Interrupt service routine
- */
-
-static irqreturn_t isr(int irq, void *priv)
-{
-	struct ixp_clock *ixp_clock = priv;
-	struct ixp46x_ts_regs *regs = ixp_clock->regs;
-	struct ptp_clock_event event;
-	u32 ack = 0, lo, hi, val;
-
-	val = __raw_readl(&regs->event);
-
-	if (val & TSER_SNS) {
-		ack |= TSER_SNS;
-		if (ixp_clock->exts0_enabled) {
-			hi = __raw_readl(&regs->asms_hi);
-			lo = __raw_readl(&regs->asms_lo);
-			event.type = PTP_CLOCK_EXTTS;
-			event.index = 0;
-			event.timestamp = ((u64) hi) << 32;
-			event.timestamp |= lo;
-			event.timestamp <<= TICKS_NS_SHIFT;
-			ptp_clock_event(ixp_clock->ptp_clock, &event);
-		}
-	}
-
-	if (val & TSER_SNM) {
-		ack |= TSER_SNM;
-		if (ixp_clock->exts1_enabled) {
-			hi = __raw_readl(&regs->amms_hi);
-			lo = __raw_readl(&regs->amms_lo);
-			event.type = PTP_CLOCK_EXTTS;
-			event.index = 1;
-			event.timestamp = ((u64) hi) << 32;
-			event.timestamp |= lo;
-			event.timestamp <<= TICKS_NS_SHIFT;
-			ptp_clock_event(ixp_clock->ptp_clock, &event);
-		}
-	}
-
-	if (val & TTIPEND)
-		ack |= TTIPEND; /* this bit seems to be always set */
-
-	if (ack) {
-		__raw_writel(ack, &regs->event);
-		return IRQ_HANDLED;
-	} else
-		return IRQ_NONE;
-}
-
-/*
- * PTP clock operations
- */
-
-static int ptp_ixp_adjfreq(struct ptp_clock_info *ptp, s32 ppb)
-{
-	u64 adj;
-	u32 diff, addend;
-	int neg_adj = 0;
-	struct ixp_clock *ixp_clock = container_of(ptp, struct ixp_clock, caps);
-	struct ixp46x_ts_regs *regs = ixp_clock->regs;
-
-	if (ppb < 0) {
-		neg_adj = 1;
-		ppb = -ppb;
-	}
-	addend = DEFAULT_ADDEND;
-	adj = addend;
-	adj *= ppb;
-	diff = div_u64(adj, 1000000000ULL);
-
-	addend = neg_adj ? addend - diff : addend + diff;
-
-	__raw_writel(addend, &regs->addend);
-
-	return 0;
-}
-
-static int ptp_ixp_adjtime(struct ptp_clock_info *ptp, s64 delta)
-{
-	s64 now;
-	unsigned long flags;
-	struct ixp_clock *ixp_clock = container_of(ptp, struct ixp_clock, caps);
-	struct ixp46x_ts_regs *regs = ixp_clock->regs;
-
-	spin_lock_irqsave(&register_lock, flags);
-
-	now = ixp_systime_read(regs);
-	now += delta;
-	ixp_systime_write(regs, now);
-
-	spin_unlock_irqrestore(&register_lock, flags);
-
-	return 0;
-}
-
-static int ptp_ixp_gettime(struct ptp_clock_info *ptp, struct timespec64 *ts)
-{
-	u64 ns;
-	unsigned long flags;
-	struct ixp_clock *ixp_clock = container_of(ptp, struct ixp_clock, caps);
-	struct ixp46x_ts_regs *regs = ixp_clock->regs;
-
-	spin_lock_irqsave(&register_lock, flags);
-
-	ns = ixp_systime_read(regs);
-
-	spin_unlock_irqrestore(&register_lock, flags);
-
-	*ts = ns_to_timespec64(ns);
-	return 0;
-}
-
-static int ptp_ixp_settime(struct ptp_clock_info *ptp,
-			   const struct timespec64 *ts)
-{
-	u64 ns;
-	unsigned long flags;
-	struct ixp_clock *ixp_clock = container_of(ptp, struct ixp_clock, caps);
-	struct ixp46x_ts_regs *regs = ixp_clock->regs;
-
-	ns = timespec64_to_ns(ts);
-
-	spin_lock_irqsave(&register_lock, flags);
-
-	ixp_systime_write(regs, ns);
-
-	spin_unlock_irqrestore(&register_lock, flags);
-
-	return 0;
-}
-
-static int ptp_ixp_enable(struct ptp_clock_info *ptp,
-			  struct ptp_clock_request *rq, int on)
-{
-	struct ixp_clock *ixp_clock = container_of(ptp, struct ixp_clock, caps);
-
-	switch (rq->type) {
-	case PTP_CLK_REQ_EXTTS:
-		switch (rq->extts.index) {
-		case 0:
-			ixp_clock->exts0_enabled = on ? 1 : 0;
-			break;
-		case 1:
-			ixp_clock->exts1_enabled = on ? 1 : 0;
-			break;
-		default:
-			return -EINVAL;
-		}
-		return 0;
-	default:
-		break;
-	}
-
-	return -EOPNOTSUPP;
-}
-
-static const struct ptp_clock_info ptp_ixp_caps = {
-	.owner		= THIS_MODULE,
-	.name		= "IXP46X timer",
-	.max_adj	= 66666655,
-	.n_ext_ts	= N_EXT_TS,
-	.n_pins		= 0,
-	.pps		= 0,
-	.adjfreq	= ptp_ixp_adjfreq,
-	.adjtime	= ptp_ixp_adjtime,
-	.gettime64	= ptp_ixp_gettime,
-	.settime64	= ptp_ixp_settime,
-	.enable		= ptp_ixp_enable,
-};
-
-/* module operations */
-
-static struct ixp_clock ixp_clock;
-
-static int setup_interrupt(int gpio)
-{
-	int irq;
-	int err;
-
-	err = gpio_request(gpio, "ixp4-ptp");
-	if (err)
-		return err;
-
-	err = gpio_direction_input(gpio);
-	if (err)
-		return err;
-
-	irq = gpio_to_irq(gpio);
-	if (irq < 0)
-		return irq;
-
-	err = irq_set_irq_type(irq, IRQF_TRIGGER_FALLING);
-	if (err) {
-		pr_err("cannot set trigger type for irq %d\n", irq);
-		return err;
-	}
-
-	err = request_irq(irq, isr, 0, DRIVER, &ixp_clock);
-	if (err) {
-		pr_err("request_irq failed for irq %d\n", irq);
-		return err;
-	}
-
-	return irq;
-}
-
-static void __exit ptp_ixp_exit(void)
-{
-	free_irq(MASTER_IRQ, &ixp_clock);
-	free_irq(SLAVE_IRQ, &ixp_clock);
-	ixp46x_phc_index = -1;
-	ptp_clock_unregister(ixp_clock.ptp_clock);
-}
-
-static int __init ptp_ixp_init(void)
-{
-	if (!cpu_is_ixp46x())
-		return -ENODEV;
-
-	ixp_clock.regs =
-		(struct ixp46x_ts_regs __iomem *) IXP4XX_TIMESYNC_BASE_VIRT;
-
-	ixp_clock.caps = ptp_ixp_caps;
-
-	ixp_clock.ptp_clock = ptp_clock_register(&ixp_clock.caps, NULL);
-
-	if (IS_ERR(ixp_clock.ptp_clock))
-		return PTR_ERR(ixp_clock.ptp_clock);
-
-	ixp46x_phc_index = ptp_clock_index(ixp_clock.ptp_clock);
-
-	__raw_writel(DEFAULT_ADDEND, &ixp_clock.regs->addend);
-	__raw_writel(1, &ixp_clock.regs->trgt_lo);
-	__raw_writel(0, &ixp_clock.regs->trgt_hi);
-	__raw_writel(TTIPEND, &ixp_clock.regs->event);
-
-	if (MASTER_IRQ != setup_interrupt(MASTER_GPIO)) {
-		pr_err("failed to setup gpio %d as irq\n", MASTER_GPIO);
-		goto no_master;
-	}
-	if (SLAVE_IRQ != setup_interrupt(SLAVE_GPIO)) {
-		pr_err("failed to setup gpio %d as irq\n", SLAVE_GPIO);
-		goto no_slave;
-	}
-
-	return 0;
-no_slave:
-	free_irq(MASTER_IRQ, &ixp_clock);
-no_master:
-	ptp_clock_unregister(ixp_clock.ptp_clock);
-	return -ENODEV;
-}
-
-module_init(ptp_ixp_init);
-module_exit(ptp_ixp_exit);
-
-MODULE_AUTHOR("Richard Cochran <richardcochran@gmail.com>");
-MODULE_DESCRIPTION("PTP clock using the IXP46X timer");
-MODULE_LICENSE("GPL");
-- 
cgit v1.2.3-59-g8ed1b


From a41a5b26d29fb0123cd3290dca453857cd8c0c66 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Sun, 12 Jan 2020 13:04:45 +0100
Subject: ixp4xx_eth: move platform_data definition

The platform data is needed to compile the driver as standalone,
so move it to a global location along with similar files.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 arch/arm/mach-ixp4xx/include/mach/platform.h | 13 +-----
 drivers/net/ethernet/xscale/ixp46x_ts.h      | 68 ++++++++++++++++++++++++++++
 drivers/net/ethernet/xscale/ixp4xx_eth.c     |  1 +
 drivers/net/ethernet/xscale/ptp_ixp46x.h     | 68 ----------------------------
 include/linux/platform_data/eth_ixp4xx.h     | 19 ++++++++
 5 files changed, 89 insertions(+), 80 deletions(-)
 create mode 100644 drivers/net/ethernet/xscale/ixp46x_ts.h
 delete mode 100644 drivers/net/ethernet/xscale/ptp_ixp46x.h
 create mode 100644 include/linux/platform_data/eth_ixp4xx.h

diff --git a/arch/arm/mach-ixp4xx/include/mach/platform.h b/arch/arm/mach-ixp4xx/include/mach/platform.h
index 04ef8025accc..6d403fe0bf52 100644
--- a/arch/arm/mach-ixp4xx/include/mach/platform.h
+++ b/arch/arm/mach-ixp4xx/include/mach/platform.h
@@ -15,6 +15,7 @@
 #ifndef __ASSEMBLY__
 
 #include <linux/reboot.h>
+#include <linux/platform_data/eth_ixp4xx.h>
 
 #include <asm/types.h>
 
@@ -92,18 +93,6 @@ struct ixp4xx_pata_data {
 	void __iomem	*cs1;
 };
 
-#define IXP4XX_ETH_NPEA		0x00
-#define IXP4XX_ETH_NPEB		0x10
-#define IXP4XX_ETH_NPEC		0x20
-
-/* Information about built-in Ethernet MAC interfaces */
-struct eth_plat_info {
-	u8 phy;		/* MII PHY ID, 0 - 31 */
-	u8 rxq;		/* configurable, currently 0 - 31 only */
-	u8 txreadyq;
-	u8 hwaddr[6];
-};
-
 /*
  * Frequency of clock used for primary clocksource
  */
diff --git a/drivers/net/ethernet/xscale/ixp46x_ts.h b/drivers/net/ethernet/xscale/ixp46x_ts.h
new file mode 100644
index 000000000000..d792130e27b0
--- /dev/null
+++ b/drivers/net/ethernet/xscale/ixp46x_ts.h
@@ -0,0 +1,68 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * PTP 1588 clock using the IXP46X
+ *
+ * Copyright (C) 2010 OMICRON electronics GmbH
+ */
+
+#ifndef _IXP46X_TS_H_
+#define _IXP46X_TS_H_
+
+#define DEFAULT_ADDEND 0xF0000029
+#define TICKS_NS_SHIFT 4
+
+struct ixp46x_channel_ctl {
+	u32 ch_control;  /* 0x40 Time Synchronization Channel Control */
+	u32 ch_event;    /* 0x44 Time Synchronization Channel Event */
+	u32 tx_snap_lo;  /* 0x48 Transmit Snapshot Low Register */
+	u32 tx_snap_hi;  /* 0x4C Transmit Snapshot High Register */
+	u32 rx_snap_lo;  /* 0x50 Receive Snapshot Low Register */
+	u32 rx_snap_hi;  /* 0x54 Receive Snapshot High Register */
+	u32 src_uuid_lo; /* 0x58 Source UUID0 Low Register */
+	u32 src_uuid_hi; /* 0x5C Sequence Identifier/Source UUID0 High */
+};
+
+struct ixp46x_ts_regs {
+	u32 control;     /* 0x00 Time Sync Control Register */
+	u32 event;       /* 0x04 Time Sync Event Register */
+	u32 addend;      /* 0x08 Time Sync Addend Register */
+	u32 accum;       /* 0x0C Time Sync Accumulator Register */
+	u32 test;        /* 0x10 Time Sync Test Register */
+	u32 unused;      /* 0x14 */
+	u32 rsystime_lo; /* 0x18 RawSystemTime_Low Register */
+	u32 rsystime_hi; /* 0x1C RawSystemTime_High Register */
+	u32 systime_lo;  /* 0x20 SystemTime_Low Register */
+	u32 systime_hi;  /* 0x24 SystemTime_High Register */
+	u32 trgt_lo;     /* 0x28 TargetTime_Low Register */
+	u32 trgt_hi;     /* 0x2C TargetTime_High Register */
+	u32 asms_lo;     /* 0x30 Auxiliary Slave Mode Snapshot Low  */
+	u32 asms_hi;     /* 0x34 Auxiliary Slave Mode Snapshot High */
+	u32 amms_lo;     /* 0x38 Auxiliary Master Mode Snapshot Low */
+	u32 amms_hi;     /* 0x3C Auxiliary Master Mode Snapshot High */
+
+	struct ixp46x_channel_ctl channel[3];
+};
+
+/* 0x00 Time Sync Control Register Bits */
+#define TSCR_AMM (1<<3)
+#define TSCR_ASM (1<<2)
+#define TSCR_TTM (1<<1)
+#define TSCR_RST (1<<0)
+
+/* 0x04 Time Sync Event Register Bits */
+#define TSER_SNM (1<<3)
+#define TSER_SNS (1<<2)
+#define TTIPEND  (1<<1)
+
+/* 0x40 Time Synchronization Channel Control Register Bits */
+#define MASTER_MODE   (1<<0)
+#define TIMESTAMP_ALL (1<<1)
+
+/* 0x44 Time Synchronization Channel Event Register Bits */
+#define TX_SNAPSHOT_LOCKED (1<<0)
+#define RX_SNAPSHOT_LOCKED (1<<1)
+
+/* The ptp_ixp46x module will set this variable */
+extern int ixp46x_phc_index;
+
+#endif
diff --git a/drivers/net/ethernet/xscale/ixp4xx_eth.c b/drivers/net/ethernet/xscale/ixp4xx_eth.c
index 0075ecdb21f4..e811bf0d23cb 100644
--- a/drivers/net/ethernet/xscale/ixp4xx_eth.c
+++ b/drivers/net/ethernet/xscale/ixp4xx_eth.c
@@ -29,6 +29,7 @@
 #include <linux/net_tstamp.h>
 #include <linux/of.h>
 #include <linux/phy.h>
+#include <linux/platform_data/eth_ixp4xx.h>
 #include <linux/platform_device.h>
 #include <linux/ptp_classify.h>
 #include <linux/slab.h>
diff --git a/drivers/net/ethernet/xscale/ptp_ixp46x.h b/drivers/net/ethernet/xscale/ptp_ixp46x.h
deleted file mode 100644
index d792130e27b0..000000000000
--- a/drivers/net/ethernet/xscale/ptp_ixp46x.h
+++ /dev/null
@@ -1,68 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0-or-later */
-/*
- * PTP 1588 clock using the IXP46X
- *
- * Copyright (C) 2010 OMICRON electronics GmbH
- */
-
-#ifndef _IXP46X_TS_H_
-#define _IXP46X_TS_H_
-
-#define DEFAULT_ADDEND 0xF0000029
-#define TICKS_NS_SHIFT 4
-
-struct ixp46x_channel_ctl {
-	u32 ch_control;  /* 0x40 Time Synchronization Channel Control */
-	u32 ch_event;    /* 0x44 Time Synchronization Channel Event */
-	u32 tx_snap_lo;  /* 0x48 Transmit Snapshot Low Register */
-	u32 tx_snap_hi;  /* 0x4C Transmit Snapshot High Register */
-	u32 rx_snap_lo;  /* 0x50 Receive Snapshot Low Register */
-	u32 rx_snap_hi;  /* 0x54 Receive Snapshot High Register */
-	u32 src_uuid_lo; /* 0x58 Source UUID0 Low Register */
-	u32 src_uuid_hi; /* 0x5C Sequence Identifier/Source UUID0 High */
-};
-
-struct ixp46x_ts_regs {
-	u32 control;     /* 0x00 Time Sync Control Register */
-	u32 event;       /* 0x04 Time Sync Event Register */
-	u32 addend;      /* 0x08 Time Sync Addend Register */
-	u32 accum;       /* 0x0C Time Sync Accumulator Register */
-	u32 test;        /* 0x10 Time Sync Test Register */
-	u32 unused;      /* 0x14 */
-	u32 rsystime_lo; /* 0x18 RawSystemTime_Low Register */
-	u32 rsystime_hi; /* 0x1C RawSystemTime_High Register */
-	u32 systime_lo;  /* 0x20 SystemTime_Low Register */
-	u32 systime_hi;  /* 0x24 SystemTime_High Register */
-	u32 trgt_lo;     /* 0x28 TargetTime_Low Register */
-	u32 trgt_hi;     /* 0x2C TargetTime_High Register */
-	u32 asms_lo;     /* 0x30 Auxiliary Slave Mode Snapshot Low  */
-	u32 asms_hi;     /* 0x34 Auxiliary Slave Mode Snapshot High */
-	u32 amms_lo;     /* 0x38 Auxiliary Master Mode Snapshot Low */
-	u32 amms_hi;     /* 0x3C Auxiliary Master Mode Snapshot High */
-
-	struct ixp46x_channel_ctl channel[3];
-};
-
-/* 0x00 Time Sync Control Register Bits */
-#define TSCR_AMM (1<<3)
-#define TSCR_ASM (1<<2)
-#define TSCR_TTM (1<<1)
-#define TSCR_RST (1<<0)
-
-/* 0x04 Time Sync Event Register Bits */
-#define TSER_SNM (1<<3)
-#define TSER_SNS (1<<2)
-#define TTIPEND  (1<<1)
-
-/* 0x40 Time Synchronization Channel Control Register Bits */
-#define MASTER_MODE   (1<<0)
-#define TIMESTAMP_ALL (1<<1)
-
-/* 0x44 Time Synchronization Channel Event Register Bits */
-#define TX_SNAPSHOT_LOCKED (1<<0)
-#define RX_SNAPSHOT_LOCKED (1<<1)
-
-/* The ptp_ixp46x module will set this variable */
-extern int ixp46x_phc_index;
-
-#endif
diff --git a/include/linux/platform_data/eth_ixp4xx.h b/include/linux/platform_data/eth_ixp4xx.h
new file mode 100644
index 000000000000..6f652ea0c6ae
--- /dev/null
+++ b/include/linux/platform_data/eth_ixp4xx.h
@@ -0,0 +1,19 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __PLATFORM_DATA_ETH_IXP4XX
+#define __PLATFORM_DATA_ETH_IXP4XX
+
+#include <linux/types.h>
+
+#define IXP4XX_ETH_NPEA		0x00
+#define IXP4XX_ETH_NPEB		0x10
+#define IXP4XX_ETH_NPEC		0x20
+
+/* Information about built-in Ethernet MAC interfaces */
+struct eth_plat_info {
+	u8 phy;		/* MII PHY ID, 0 - 31 */
+	u8 rxq;		/* configurable, currently 0 - 31 only */
+	u8 txreadyq;
+	u8 hwaddr[6];
+};
+
+#endif
-- 
cgit v1.2.3-59-g8ed1b


From c83db9ef5640548631707e8b4a7bcddc115fdbae Mon Sep 17 00:00:00 2001
From: Linus Walleij <linus.walleij@linaro.org>
Date: Sun, 12 Jan 2020 13:04:46 +0100
Subject: net: ethernet: ixp4xx: Standard module init

The IXP4xx driver was initializing the MDIO bus before even
probing, in the callbacks supposed to be used for setting up
the module itself, and with the side effect of trying to
register the MDIO bus as soon as this module was loaded or
compiled into the kernel whether the device was discovered
or not.

This does not work with multiplatform environments.

To get rid of this: set up the MDIO bus from the probe()
callback and remove it in the remove() callback. Rename
the probe() and remove() calls to reflect the most common
conventions.

Since there is a bit of checking for the ethernet feature
to be present in the MDIO registering function, making the
whole module not even be registered if we can't find an
MDIO bus, we need something similar: register the MDIO
bus when the corresponding ethernet is probed, and
return -EPROBE_DEFER on the other interfaces until this
happens. If no MDIO bus is present on any of the
registered interfaces we will eventually bail out.

None of the platforms I've seen has e.g. MDIO on EthB
and only uses EthC, there is always a Ethernet hardware
on the NPE (B, C) that has the MDIO bus, we just might
have to wait for it.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/xscale/ixp4xx_eth.c | 96 +++++++++++++++-----------------
 1 file changed, 44 insertions(+), 52 deletions(-)

diff --git a/drivers/net/ethernet/xscale/ixp4xx_eth.c b/drivers/net/ethernet/xscale/ixp4xx_eth.c
index e811bf0d23cb..799ffebba491 100644
--- a/drivers/net/ethernet/xscale/ixp4xx_eth.c
+++ b/drivers/net/ethernet/xscale/ixp4xx_eth.c
@@ -519,25 +519,14 @@ static int ixp4xx_mdio_write(struct mii_bus *bus, int phy_id, int location,
 	return ret;
 }
 
-static int ixp4xx_mdio_register(void)
+static int ixp4xx_mdio_register(struct eth_regs __iomem *regs)
 {
 	int err;
 
 	if (!(mdio_bus = mdiobus_alloc()))
 		return -ENOMEM;
 
-	if (cpu_is_ixp43x()) {
-		/* IXP43x lacks NPE-B and uses NPE-C for MII PHY access */
-		if (!(ixp4xx_read_feature_bits() & IXP4XX_FEATURE_NPEC_ETH))
-			return -ENODEV;
-		mdio_regs = (struct eth_regs __iomem *)IXP4XX_EthC_BASE_VIRT;
-	} else {
-		/* All MII PHY accesses use NPE-B Ethernet registers */
-		if (!(ixp4xx_read_feature_bits() & IXP4XX_FEATURE_NPEB_ETH0))
-			return -ENODEV;
-		mdio_regs = (struct eth_regs __iomem *)IXP4XX_EthB_BASE_VIRT;
-	}
-
+	mdio_regs = regs;
 	__raw_writel(DEFAULT_CORE_CNTRL, &mdio_regs->core_control);
 	spin_lock_init(&mdio_lock);
 	mdio_bus->name = "IXP4xx MII Bus";
@@ -1376,7 +1365,7 @@ static const struct net_device_ops ixp4xx_netdev_ops = {
 	.ndo_validate_addr = eth_validate_addr,
 };
 
-static int eth_init_one(struct platform_device *pdev)
+static int ixp4xx_eth_probe(struct platform_device *pdev)
 {
 	struct port *port;
 	struct net_device *dev;
@@ -1386,7 +1375,7 @@ static int eth_init_one(struct platform_device *pdev)
 	char phy_id[MII_BUS_ID_SIZE + 3];
 	int err;
 
-	if (!(dev = alloc_etherdev(sizeof(struct port))))
+	if (!(dev = devm_alloc_etherdev(&pdev->dev, sizeof(struct port))))
 		return -ENOMEM;
 
 	SET_NETDEV_DEV(dev, &pdev->dev);
@@ -1396,20 +1385,51 @@ static int eth_init_one(struct platform_device *pdev)
 
 	switch (port->id) {
 	case IXP4XX_ETH_NPEA:
+		/* If the MDIO bus is not up yet, defer probe */
+		if (!mdio_bus)
+			return -EPROBE_DEFER;
 		port->regs = (struct eth_regs __iomem *)IXP4XX_EthA_BASE_VIRT;
 		regs_phys  = IXP4XX_EthA_BASE_PHYS;
 		break;
 	case IXP4XX_ETH_NPEB:
+		/*
+		 * On all except IXP43x, NPE-B is used for the MDIO bus.
+		 * If there is no NPE-B in the feature set, bail out, else
+		 * register the MDIO bus.
+		 */
+		if (!cpu_is_ixp43x()) {
+			if (!(ixp4xx_read_feature_bits() &
+			      IXP4XX_FEATURE_NPEB_ETH0))
+				return -ENODEV;
+			/* Else register the MDIO bus on NPE-B */
+			if ((err = ixp4xx_mdio_register(IXP4XX_EthC_BASE_VIRT)))
+				return err;
+		}
+		if (!mdio_bus)
+			return -EPROBE_DEFER;
 		port->regs = (struct eth_regs __iomem *)IXP4XX_EthB_BASE_VIRT;
 		regs_phys  = IXP4XX_EthB_BASE_PHYS;
 		break;
 	case IXP4XX_ETH_NPEC:
+		/*
+		 * IXP43x lacks NPE-B and uses NPE-C for the MDIO bus access,
+		 * of there is no NPE-C, no bus, nothing works, so bail out.
+		 */
+		if (cpu_is_ixp43x()) {
+			if (!(ixp4xx_read_feature_bits() &
+			      IXP4XX_FEATURE_NPEC_ETH))
+				return -ENODEV;
+			/* Else register the MDIO bus on NPE-C */
+			if ((err = ixp4xx_mdio_register(IXP4XX_EthC_BASE_VIRT)))
+				return err;
+		}
+		if (!mdio_bus)
+			return -EPROBE_DEFER;
 		port->regs = (struct eth_regs __iomem *)IXP4XX_EthC_BASE_VIRT;
 		regs_phys  = IXP4XX_EthC_BASE_PHYS;
 		break;
 	default:
-		err = -ENODEV;
-		goto err_free;
+		return -ENODEV;
 	}
 
 	dev->netdev_ops = &ixp4xx_netdev_ops;
@@ -1418,10 +1438,8 @@ static int eth_init_one(struct platform_device *pdev)
 
 	netif_napi_add(dev, &port->napi, eth_poll, NAPI_WEIGHT);
 
-	if (!(port->npe = npe_request(NPE_ID(port->id)))) {
-		err = -EIO;
-		goto err_free;
-	}
+	if (!(port->npe = npe_request(NPE_ID(port->id))))
+		return -EIO;
 
 	port->mem_res = request_mem_region(regs_phys, REGS_SIZE, dev->name);
 	if (!port->mem_res) {
@@ -1467,12 +1485,10 @@ err_free_mem:
 	release_resource(port->mem_res);
 err_npe_rel:
 	npe_release(port->npe);
-err_free:
-	free_netdev(dev);
 	return err;
 }
 
-static int eth_remove_one(struct platform_device *pdev)
+static int ixp4xx_eth_remove(struct platform_device *pdev)
 {
 	struct net_device *dev = platform_get_drvdata(pdev);
 	struct phy_device *phydev = dev->phydev;
@@ -1480,45 +1496,21 @@ static int eth_remove_one(struct platform_device *pdev)
 
 	unregister_netdev(dev);
 	phy_disconnect(phydev);
+	ixp4xx_mdio_remove();
 	npe_port_tab[NPE_ID(port->id)] = NULL;
 	npe_release(port->npe);
 	release_resource(port->mem_res);
-	free_netdev(dev);
 	return 0;
 }
 
 static struct platform_driver ixp4xx_eth_driver = {
 	.driver.name	= DRV_NAME,
-	.probe		= eth_init_one,
-	.remove		= eth_remove_one,
+	.probe		= ixp4xx_eth_probe,
+	.remove		= ixp4xx_eth_remove,
 };
-
-static int __init eth_init_module(void)
-{
-	int err;
-
-	/*
-	 * FIXME: we bail out on device tree boot but this really needs
-	 * to be fixed in a nicer way: this registers the MDIO bus before
-	 * even matching the driver infrastructure, we should only probe
-	 * detected hardware.
-	 */
-	if (of_have_populated_dt())
-		return -ENODEV;
-	if ((err = ixp4xx_mdio_register()))
-		return err;
-	return platform_driver_register(&ixp4xx_eth_driver);
-}
-
-static void __exit eth_cleanup_module(void)
-{
-	platform_driver_unregister(&ixp4xx_eth_driver);
-	ixp4xx_mdio_remove();
-}
+module_platform_driver(ixp4xx_eth_driver);
 
 MODULE_AUTHOR("Krzysztof Halasa");
 MODULE_DESCRIPTION("Intel IXP4xx Ethernet driver");
 MODULE_LICENSE("GPL v2");
 MODULE_ALIAS("platform:ixp4xx_eth");
-module_init(eth_init_module);
-module_exit(eth_cleanup_module);
-- 
cgit v1.2.3-59-g8ed1b


From d813d7e57026ac7638273d00c9915f2ea83a8fce Mon Sep 17 00:00:00 2001
From: Linus Walleij <linus.walleij@linaro.org>
Date: Sun, 12 Jan 2020 13:04:47 +0100
Subject: net: ethernet: ixp4xx: Use distinct local variable

Use "ndev" for the struct net_device and "dev" for the
struct device in probe() and remove(). Add the local
"dev" pointer for later use in refactoring.

Take this opportunity to fix inverse christmas tree
coding style.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/xscale/ixp4xx_eth.c | 47 +++++++++++++++++---------------
 1 file changed, 25 insertions(+), 22 deletions(-)

diff --git a/drivers/net/ethernet/xscale/ixp4xx_eth.c b/drivers/net/ethernet/xscale/ixp4xx_eth.c
index 799ffebba491..05ab8426bb8d 100644
--- a/drivers/net/ethernet/xscale/ixp4xx_eth.c
+++ b/drivers/net/ethernet/xscale/ixp4xx_eth.c
@@ -1367,20 +1367,23 @@ static const struct net_device_ops ixp4xx_netdev_ops = {
 
 static int ixp4xx_eth_probe(struct platform_device *pdev)
 {
-	struct port *port;
-	struct net_device *dev;
-	struct eth_plat_info *plat = dev_get_platdata(&pdev->dev);
+	char phy_id[MII_BUS_ID_SIZE + 3];
 	struct phy_device *phydev = NULL;
+	struct device *dev = &pdev->dev;
+	struct eth_plat_info *plat;
+	struct net_device *ndev;
+	struct port *port;
 	u32 regs_phys;
-	char phy_id[MII_BUS_ID_SIZE + 3];
 	int err;
 
-	if (!(dev = devm_alloc_etherdev(&pdev->dev, sizeof(struct port))))
+	plat = dev_get_platdata(dev);
+
+	if (!(ndev = devm_alloc_etherdev(dev, sizeof(struct port))))
 		return -ENOMEM;
 
-	SET_NETDEV_DEV(dev, &pdev->dev);
-	port = netdev_priv(dev);
-	port->netdev = dev;
+	SET_NETDEV_DEV(ndev, dev);
+	port = netdev_priv(ndev);
+	port->netdev = ndev;
 	port->id = pdev->id;
 
 	switch (port->id) {
@@ -1432,16 +1435,16 @@ static int ixp4xx_eth_probe(struct platform_device *pdev)
 		return -ENODEV;
 	}
 
-	dev->netdev_ops = &ixp4xx_netdev_ops;
-	dev->ethtool_ops = &ixp4xx_ethtool_ops;
-	dev->tx_queue_len = 100;
+	ndev->netdev_ops = &ixp4xx_netdev_ops;
+	ndev->ethtool_ops = &ixp4xx_ethtool_ops;
+	ndev->tx_queue_len = 100;
 
-	netif_napi_add(dev, &port->napi, eth_poll, NAPI_WEIGHT);
+	netif_napi_add(ndev, &port->napi, eth_poll, NAPI_WEIGHT);
 
 	if (!(port->npe = npe_request(NPE_ID(port->id))))
 		return -EIO;
 
-	port->mem_res = request_mem_region(regs_phys, REGS_SIZE, dev->name);
+	port->mem_res = request_mem_region(regs_phys, REGS_SIZE, ndev->name);
 	if (!port->mem_res) {
 		err = -EBUSY;
 		goto err_npe_rel;
@@ -1449,9 +1452,9 @@ static int ixp4xx_eth_probe(struct platform_device *pdev)
 
 	port->plat = plat;
 	npe_port_tab[NPE_ID(port->id)] = port;
-	memcpy(dev->dev_addr, plat->hwaddr, ETH_ALEN);
+	memcpy(ndev->dev_addr, plat->hwaddr, ETH_ALEN);
 
-	platform_set_drvdata(pdev, dev);
+	platform_set_drvdata(pdev, ndev);
 
 	__raw_writel(DEFAULT_CORE_CNTRL | CORE_RESET,
 		     &port->regs->core_control);
@@ -1461,7 +1464,7 @@ static int ixp4xx_eth_probe(struct platform_device *pdev)
 
 	snprintf(phy_id, MII_BUS_ID_SIZE + 3, PHY_ID_FMT,
 		mdio_bus->id, plat->phy);
-	phydev = phy_connect(dev, phy_id, &ixp4xx_adjust_link,
+	phydev = phy_connect(ndev, phy_id, &ixp4xx_adjust_link,
 			     PHY_INTERFACE_MODE_MII);
 	if (IS_ERR(phydev)) {
 		err = PTR_ERR(phydev);
@@ -1470,10 +1473,10 @@ static int ixp4xx_eth_probe(struct platform_device *pdev)
 
 	phydev->irq = PHY_POLL;
 
-	if ((err = register_netdev(dev)))
+	if ((err = register_netdev(ndev)))
 		goto err_phy_dis;
 
-	printk(KERN_INFO "%s: MII PHY %i on %s\n", dev->name, plat->phy,
+	printk(KERN_INFO "%s: MII PHY %i on %s\n", ndev->name, plat->phy,
 	       npe_name(port->npe));
 
 	return 0;
@@ -1490,11 +1493,11 @@ err_npe_rel:
 
 static int ixp4xx_eth_remove(struct platform_device *pdev)
 {
-	struct net_device *dev = platform_get_drvdata(pdev);
-	struct phy_device *phydev = dev->phydev;
-	struct port *port = netdev_priv(dev);
+	struct net_device *ndev = platform_get_drvdata(pdev);
+	struct phy_device *phydev = ndev->phydev;
+	struct port *port = netdev_priv(ndev);
 
-	unregister_netdev(dev);
+	unregister_netdev(ndev);
 	phy_disconnect(phydev);
 	ixp4xx_mdio_remove();
 	npe_port_tab[NPE_ID(port->id)] = NULL;
-- 
cgit v1.2.3-59-g8ed1b


From 3cb5b0ee72b8b9ff0827b1edb40b6c4b00d5ab2e Mon Sep 17 00:00:00 2001
From: Linus Walleij <linus.walleij@linaro.org>
Date: Sun, 12 Jan 2020 13:04:48 +0100
Subject: net: ehernet: ixp4xx: Use netdev_* messages

Simplify and correct a bunch of messages using printk
directly to use the netdev_* macros. I have not changed
all of them, just the low-hanging fruit.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/xscale/ixp4xx_eth.c | 52 ++++++++++++++------------------
 1 file changed, 23 insertions(+), 29 deletions(-)

diff --git a/drivers/net/ethernet/xscale/ixp4xx_eth.c b/drivers/net/ethernet/xscale/ixp4xx_eth.c
index 05ab8426bb8d..f7edf8b38dea 100644
--- a/drivers/net/ethernet/xscale/ixp4xx_eth.c
+++ b/drivers/net/ethernet/xscale/ixp4xx_eth.c
@@ -572,8 +572,8 @@ static void ixp4xx_adjust_link(struct net_device *dev)
 		__raw_writel(DEFAULT_TX_CNTRL0 | TX_CNTRL0_HALFDUPLEX,
 			     &port->regs->tx_control[0]);
 
-	printk(KERN_INFO "%s: link up, speed %u Mb/s, %s duplex\n",
-	       dev->name, port->speed, port->duplex ? "full" : "half");
+	netdev_info(dev, "%s: link up, speed %u Mb/s, %s duplex\n",
+		    dev->name, port->speed, port->duplex ? "full" : "half");
 }
 
 
@@ -583,7 +583,7 @@ static inline void debug_pkt(struct net_device *dev, const char *func,
 #if DEBUG_PKT_BYTES
 	int i;
 
-	printk(KERN_DEBUG "%s: %s(%i) ", dev->name, func, len);
+	netdev_debug(dev, "%s(%i) ", func, len);
 	for (i = 0; i < len; i++) {
 		if (i >= DEBUG_PKT_BYTES)
 			break;
@@ -674,7 +674,7 @@ static int eth_poll(struct napi_struct *napi, int budget)
 	int received = 0;
 
 #if DEBUG_RX
-	printk(KERN_DEBUG "%s: eth_poll\n", dev->name);
+	netdev_debug(dev, "eth_poll\n");
 #endif
 
 	while (received < budget) {
@@ -688,23 +688,20 @@ static int eth_poll(struct napi_struct *napi, int budget)
 
 		if ((n = queue_get_desc(rxq, port, 0)) < 0) {
 #if DEBUG_RX
-			printk(KERN_DEBUG "%s: eth_poll napi_complete\n",
-			       dev->name);
+			netdev_debug(dev, "eth_poll napi_complete\n");
 #endif
 			napi_complete(napi);
 			qmgr_enable_irq(rxq);
 			if (!qmgr_stat_below_low_watermark(rxq) &&
 			    napi_reschedule(napi)) { /* not empty again */
 #if DEBUG_RX
-				printk(KERN_DEBUG "%s: eth_poll napi_reschedule succeeded\n",
-				       dev->name);
+				netdev_debug(dev, "eth_poll napi_reschedule succeeded\n");
 #endif
 				qmgr_disable_irq(rxq);
 				continue;
 			}
 #if DEBUG_RX
-			printk(KERN_DEBUG "%s: eth_poll all done\n",
-			       dev->name);
+			netdev_debug(dev, "eth_poll all done\n");
 #endif
 			return received; /* all work done */
 		}
@@ -769,7 +766,7 @@ static int eth_poll(struct napi_struct *napi, int budget)
 	}
 
 #if DEBUG_RX
-	printk(KERN_DEBUG "eth_poll(): end, not all work done\n");
+	netdev_debug(dev, "eth_poll(): end, not all work done\n");
 #endif
 	return received;		/* not all work done */
 }
@@ -833,7 +830,7 @@ static int eth_xmit(struct sk_buff *skb, struct net_device *dev)
 	struct desc *desc;
 
 #if DEBUG_TX
-	printk(KERN_DEBUG "%s: eth_xmit\n", dev->name);
+	netdev_debug(dev, "eth_xmit\n");
 #endif
 
 	if (unlikely(skb->len > MAX_MRU)) {
@@ -888,22 +885,21 @@ static int eth_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	if (qmgr_stat_below_low_watermark(txreadyq)) { /* empty */
 #if DEBUG_TX
-		printk(KERN_DEBUG "%s: eth_xmit queue full\n", dev->name);
+		netdev_debug(dev, "eth_xmit queue full\n");
 #endif
 		netif_stop_queue(dev);
 		/* we could miss TX ready interrupt */
 		/* really empty in fact */
 		if (!qmgr_stat_below_low_watermark(txreadyq)) {
 #if DEBUG_TX
-			printk(KERN_DEBUG "%s: eth_xmit ready again\n",
-			       dev->name);
+			netdev_debug(dev, "eth_xmit ready again\n");
 #endif
 			netif_wake_queue(dev);
 		}
 	}
 
 #if DEBUG_TX
-	printk(KERN_DEBUG "%s: eth_xmit end\n", dev->name);
+	netdev_debug(dev, "eth_xmit end\n");
 #endif
 
 	ixp_tx_timestamp(port, skb);
@@ -1177,8 +1173,7 @@ static int eth_open(struct net_device *dev)
 			return err;
 
 		if (npe_recv_message(npe, &msg, "ETH_GET_STATUS")) {
-			printk(KERN_ERR "%s: %s not responding\n", dev->name,
-			       npe_name(npe));
+			netdev_err(dev, "%s not responding\n", npe_name(npe));
 			return -EIO;
 		}
 		port->firmware[0] = msg.byte4;
@@ -1290,7 +1285,7 @@ static int eth_close(struct net_device *dev)
 	msg.eth_id = port->id;
 	msg.byte3 = 1;
 	if (npe_send_recv_message(port->npe, &msg, "ETH_ENABLE_LOOPBACK"))
-		printk(KERN_CRIT "%s: unable to enable loopback\n", dev->name);
+		netdev_crit(dev, "unable to enable loopback\n");
 
 	i = 0;
 	do {			/* drain RX buffers */
@@ -1314,11 +1309,11 @@ static int eth_close(struct net_device *dev)
 	} while (++i < MAX_CLOSE_WAIT);
 
 	if (buffs)
-		printk(KERN_CRIT "%s: unable to drain RX queue, %i buffer(s)"
-		       " left in NPE\n", dev->name, buffs);
+		netdev_crit(dev, "unable to drain RX queue, %i buffer(s)"
+			    " left in NPE\n", buffs);
 #if DEBUG_CLOSE
 	if (!buffs)
-		printk(KERN_DEBUG "Draining RX queue took %i cycles\n", i);
+		netdev_debug(dev, "draining RX queue took %i cycles\n", i);
 #endif
 
 	buffs = TX_DESCS;
@@ -1334,17 +1329,16 @@ static int eth_close(struct net_device *dev)
 	} while (++i < MAX_CLOSE_WAIT);
 
 	if (buffs)
-		printk(KERN_CRIT "%s: unable to drain TX queue, %i buffer(s) "
-		       "left in NPE\n", dev->name, buffs);
+		netdev_crit(dev, "unable to drain TX queue, %i buffer(s) "
+			    "left in NPE\n", buffs);
 #if DEBUG_CLOSE
 	if (!buffs)
-		printk(KERN_DEBUG "Draining TX queues took %i cycles\n", i);
+		netdev_debug(dev, "draining TX queues took %i cycles\n", i);
 #endif
 
 	msg.byte3 = 0;
 	if (npe_send_recv_message(port->npe, &msg, "ETH_DISABLE_LOOPBACK"))
-		printk(KERN_CRIT "%s: unable to disable loopback\n",
-		       dev->name);
+		netdev_crit(dev, "unable to disable loopback\n");
 
 	phy_stop(dev->phydev);
 
@@ -1476,8 +1470,8 @@ static int ixp4xx_eth_probe(struct platform_device *pdev)
 	if ((err = register_netdev(ndev)))
 		goto err_phy_dis;
 
-	printk(KERN_INFO "%s: MII PHY %i on %s\n", ndev->name, plat->phy,
-	       npe_name(port->npe));
+	netdev_info(ndev, "%s: MII PHY %i on %s\n", ndev->name, plat->phy,
+		    npe_name(port->npe));
 
 	return 0;
 
-- 
cgit v1.2.3-59-g8ed1b


From f458ac479777c627c9b92ab640afec3bfa150660 Mon Sep 17 00:00:00 2001
From: Linus Walleij <linus.walleij@linaro.org>
Date: Sun, 12 Jan 2020 13:04:49 +0100
Subject: ARM/net: ixp4xx: Pass ethernet physical base as resource

In order to probe this ethernet interface from the device tree
all physical MMIO regions must be passed as resources. Begin
this rewrite by first passing the port base address as a
resource for all platforms using this driver, remap it in
the driver and avoid using any reference of the statically
mapped virtual address in the driver.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 arch/arm/mach-ixp4xx/fsg-setup.c         | 20 ++++++++++++++++++++
 arch/arm/mach-ixp4xx/goramo_mlr.c        | 20 ++++++++++++++++++++
 arch/arm/mach-ixp4xx/ixdp425-setup.c     | 20 ++++++++++++++++++++
 arch/arm/mach-ixp4xx/nas100d-setup.c     | 10 ++++++++++
 arch/arm/mach-ixp4xx/nslu2-setup.c       | 10 ++++++++++
 arch/arm/mach-ixp4xx/omixp-setup.c       | 20 ++++++++++++++++++++
 arch/arm/mach-ixp4xx/vulcan-setup.c      | 20 ++++++++++++++++++++
 drivers/net/ethernet/xscale/ixp4xx_eth.c | 20 +++++++++++---------
 8 files changed, 131 insertions(+), 9 deletions(-)

diff --git a/arch/arm/mach-ixp4xx/fsg-setup.c b/arch/arm/mach-ixp4xx/fsg-setup.c
index 648932d8d7a8..507ee3878769 100644
--- a/arch/arm/mach-ixp4xx/fsg-setup.c
+++ b/arch/arm/mach-ixp4xx/fsg-setup.c
@@ -132,6 +132,22 @@ static struct platform_device fsg_leds = {
 };
 
 /* Built-in 10/100 Ethernet MAC interfaces */
+static struct resource fsg_eth_npeb_resources[] = {
+	{
+		.start		= IXP4XX_EthB_BASE_PHYS,
+		.end		= IXP4XX_EthB_BASE_PHYS + 0x0fff,
+		.flags		= IORESOURCE_MEM,
+	},
+};
+
+static struct resource fsg_eth_npec_resources[] = {
+	{
+		.start		= IXP4XX_EthC_BASE_PHYS,
+		.end		= IXP4XX_EthC_BASE_PHYS + 0x0fff,
+		.flags		= IORESOURCE_MEM,
+	},
+};
+
 static struct eth_plat_info fsg_plat_eth[] = {
 	{
 		.phy		= 5,
@@ -151,12 +167,16 @@ static struct platform_device fsg_eth[] = {
 		.dev = {
 			.platform_data	= fsg_plat_eth,
 		},
+		.num_resources	= ARRAY_SIZE(fsg_eth_npeb_resources),
+		.resource	= fsg_eth_npeb_resources,
 	}, {
 		.name			= "ixp4xx_eth",
 		.id			= IXP4XX_ETH_NPEC,
 		.dev = {
 			.platform_data	= fsg_plat_eth + 1,
 		},
+		.num_resources	= ARRAY_SIZE(fsg_eth_npec_resources),
+		.resource	= fsg_eth_npec_resources,
 	}
 };
 
diff --git a/arch/arm/mach-ixp4xx/goramo_mlr.c b/arch/arm/mach-ixp4xx/goramo_mlr.c
index 93b7afeee142..07b50dfcc489 100644
--- a/arch/arm/mach-ixp4xx/goramo_mlr.c
+++ b/arch/arm/mach-ixp4xx/goramo_mlr.c
@@ -273,6 +273,22 @@ static struct platform_device device_uarts = {
 
 
 /* Built-in 10/100 Ethernet MAC interfaces */
+static struct resource eth_npeb_resources[] = {
+	{
+		.start		= IXP4XX_EthB_BASE_PHYS,
+		.end		= IXP4XX_EthB_BASE_PHYS + 0x0fff,
+		.flags		= IORESOURCE_MEM,
+	},
+};
+
+static struct resource eth_npec_resources[] = {
+	{
+		.start		= IXP4XX_EthC_BASE_PHYS,
+		.end		= IXP4XX_EthC_BASE_PHYS + 0x0fff,
+		.flags		= IORESOURCE_MEM,
+	},
+};
+
 static struct eth_plat_info eth_plat[] = {
 	{
 		.phy		= 0,
@@ -290,10 +306,14 @@ static struct platform_device device_eth_tab[] = {
 		.name			= "ixp4xx_eth",
 		.id			= IXP4XX_ETH_NPEB,
 		.dev.platform_data	= eth_plat,
+		.num_resources		= ARRAY_SIZE(eth_npeb_resources),
+		.resource		= eth_npeb_resources,
 	}, {
 		.name			= "ixp4xx_eth",
 		.id			= IXP4XX_ETH_NPEC,
 		.dev.platform_data	= eth_plat + 1,
+		.num_resources		= ARRAY_SIZE(eth_npec_resources),
+		.resource		= eth_npec_resources,
 	}
 };
 
diff --git a/arch/arm/mach-ixp4xx/ixdp425-setup.c b/arch/arm/mach-ixp4xx/ixdp425-setup.c
index 6f0f7ed18ea8..45d5b720ded6 100644
--- a/arch/arm/mach-ixp4xx/ixdp425-setup.c
+++ b/arch/arm/mach-ixp4xx/ixdp425-setup.c
@@ -187,6 +187,22 @@ static struct platform_device ixdp425_uart = {
 };
 
 /* Built-in 10/100 Ethernet MAC interfaces */
+static struct resource ixp425_npeb_resources[] = {
+	{
+		.start		= IXP4XX_EthB_BASE_PHYS,
+		.end		= IXP4XX_EthB_BASE_PHYS + 0x0fff,
+		.flags		= IORESOURCE_MEM,
+	},
+};
+
+static struct resource ixp425_npec_resources[] = {
+	{
+		.start		= IXP4XX_EthC_BASE_PHYS,
+		.end		= IXP4XX_EthC_BASE_PHYS + 0x0fff,
+		.flags		= IORESOURCE_MEM,
+	},
+};
+
 static struct eth_plat_info ixdp425_plat_eth[] = {
 	{
 		.phy		= 0,
@@ -204,10 +220,14 @@ static struct platform_device ixdp425_eth[] = {
 		.name			= "ixp4xx_eth",
 		.id			= IXP4XX_ETH_NPEB,
 		.dev.platform_data	= ixdp425_plat_eth,
+		.num_resources		= ARRAY_SIZE(ixp425_npeb_resources),
+		.resource		= ixp425_npeb_resources,
 	}, {
 		.name			= "ixp4xx_eth",
 		.id			= IXP4XX_ETH_NPEC,
 		.dev.platform_data	= ixdp425_plat_eth + 1,
+		.num_resources		= ARRAY_SIZE(ixp425_npec_resources),
+		.resource		= ixp425_npec_resources,
 	}
 };
 
diff --git a/arch/arm/mach-ixp4xx/nas100d-setup.c b/arch/arm/mach-ixp4xx/nas100d-setup.c
index c142cfa8c5d6..6959ad2e3aec 100644
--- a/arch/arm/mach-ixp4xx/nas100d-setup.c
+++ b/arch/arm/mach-ixp4xx/nas100d-setup.c
@@ -165,6 +165,14 @@ static struct platform_device nas100d_uart = {
 };
 
 /* Built-in 10/100 Ethernet MAC interfaces */
+static struct resource nas100d_eth_resources[] = {
+	{
+		.start		= IXP4XX_EthB_BASE_PHYS,
+		.end		= IXP4XX_EthB_BASE_PHYS + 0x0fff,
+		.flags		= IORESOURCE_MEM,
+	},
+};
+
 static struct eth_plat_info nas100d_plat_eth[] = {
 	{
 		.phy		= 0,
@@ -178,6 +186,8 @@ static struct platform_device nas100d_eth[] = {
 		.name			= "ixp4xx_eth",
 		.id			= IXP4XX_ETH_NPEB,
 		.dev.platform_data	= nas100d_plat_eth,
+		.num_resources		= ARRAY_SIZE(nas100d_eth_resources),
+		.resource		= nas100d_eth_resources,
 	}
 };
 
diff --git a/arch/arm/mach-ixp4xx/nslu2-setup.c b/arch/arm/mach-ixp4xx/nslu2-setup.c
index ee1877fcfafe..a428bb918703 100644
--- a/arch/arm/mach-ixp4xx/nslu2-setup.c
+++ b/arch/arm/mach-ixp4xx/nslu2-setup.c
@@ -185,6 +185,14 @@ static struct platform_device nslu2_uart = {
 };
 
 /* Built-in 10/100 Ethernet MAC interfaces */
+static struct resource nslu2_eth_resources[] = {
+	{
+		.start		= IXP4XX_EthB_BASE_PHYS,
+		.end		= IXP4XX_EthB_BASE_PHYS + 0x0fff,
+		.flags		= IORESOURCE_MEM,
+	},
+};
+
 static struct eth_plat_info nslu2_plat_eth[] = {
 	{
 		.phy		= 1,
@@ -198,6 +206,8 @@ static struct platform_device nslu2_eth[] = {
 		.name			= "ixp4xx_eth",
 		.id			= IXP4XX_ETH_NPEB,
 		.dev.platform_data	= nslu2_plat_eth,
+		.num_resources		= ARRAY_SIZE(nslu2_eth_resources),
+		.resource		= nslu2_eth_resources,
 	}
 };
 
diff --git a/arch/arm/mach-ixp4xx/omixp-setup.c b/arch/arm/mach-ixp4xx/omixp-setup.c
index 6ed5a9aed600..8f2b8c473d7a 100644
--- a/arch/arm/mach-ixp4xx/omixp-setup.c
+++ b/arch/arm/mach-ixp4xx/omixp-setup.c
@@ -170,6 +170,22 @@ static struct platform_device mic256_leds = {
 };
 
 /* Built-in 10/100 Ethernet MAC interfaces */
+static struct resource ixp425_npeb_resources[] = {
+	{
+		.start		= IXP4XX_EthB_BASE_PHYS,
+		.end		= IXP4XX_EthB_BASE_PHYS + 0x0fff,
+		.flags		= IORESOURCE_MEM,
+	},
+};
+
+static struct resource ixp425_npec_resources[] = {
+	{
+		.start		= IXP4XX_EthC_BASE_PHYS,
+		.end		= IXP4XX_EthC_BASE_PHYS + 0x0fff,
+		.flags		= IORESOURCE_MEM,
+	},
+};
+
 static struct eth_plat_info ixdp425_plat_eth[] = {
 	{
 		.phy		= 0,
@@ -187,10 +203,14 @@ static struct platform_device ixdp425_eth[] = {
 		.name			= "ixp4xx_eth",
 		.id			= IXP4XX_ETH_NPEB,
 		.dev.platform_data	= ixdp425_plat_eth,
+		.num_resources		= ARRAY_SIZE(ixp425_npeb_resources),
+		.resource		= ixp425_npeb_resources,
 	}, {
 		.name			= "ixp4xx_eth",
 		.id			= IXP4XX_ETH_NPEC,
 		.dev.platform_data	= ixdp425_plat_eth + 1,
+		.num_resources		= ARRAY_SIZE(ixp425_npec_resources),
+		.resource		= ixp425_npec_resources,
 	},
 };
 
diff --git a/arch/arm/mach-ixp4xx/vulcan-setup.c b/arch/arm/mach-ixp4xx/vulcan-setup.c
index d2ebb7c675a8..e506d2af98ad 100644
--- a/arch/arm/mach-ixp4xx/vulcan-setup.c
+++ b/arch/arm/mach-ixp4xx/vulcan-setup.c
@@ -124,6 +124,22 @@ static struct platform_device vulcan_uart = {
 	.num_resources		= ARRAY_SIZE(vulcan_uart_resources),
 };
 
+static struct resource vulcan_npeb_resources[] = {
+	{
+		.start		= IXP4XX_EthB_BASE_PHYS,
+		.end		= IXP4XX_EthB_BASE_PHYS + 0x0fff,
+		.flags		= IORESOURCE_MEM,
+	},
+};
+
+static struct resource vulcan_npec_resources[] = {
+	{
+		.start		= IXP4XX_EthC_BASE_PHYS,
+		.end		= IXP4XX_EthC_BASE_PHYS + 0x0fff,
+		.flags		= IORESOURCE_MEM,
+	},
+};
+
 static struct eth_plat_info vulcan_plat_eth[] = {
 	[0] = {
 		.phy		= 0,
@@ -144,6 +160,8 @@ static struct platform_device vulcan_eth[] = {
 		.dev = {
 			.platform_data	= &vulcan_plat_eth[0],
 		},
+		.num_resources		= ARRAY_SIZE(vulcan_npeb_resources),
+		.resource		= vulcan_npeb_resources,
 	},
 	[1] = {
 		.name			= "ixp4xx_eth",
@@ -151,6 +169,8 @@ static struct platform_device vulcan_eth[] = {
 		.dev = {
 			.platform_data	= &vulcan_plat_eth[1],
 		},
+		.num_resources		= ARRAY_SIZE(vulcan_npec_resources),
+		.resource		= vulcan_npec_resources,
 	},
 };
 
diff --git a/drivers/net/ethernet/xscale/ixp4xx_eth.c b/drivers/net/ethernet/xscale/ixp4xx_eth.c
index f7edf8b38dea..ee45215c4ba4 100644
--- a/drivers/net/ethernet/xscale/ixp4xx_eth.c
+++ b/drivers/net/ethernet/xscale/ixp4xx_eth.c
@@ -1365,9 +1365,10 @@ static int ixp4xx_eth_probe(struct platform_device *pdev)
 	struct phy_device *phydev = NULL;
 	struct device *dev = &pdev->dev;
 	struct eth_plat_info *plat;
+	resource_size_t regs_phys;
 	struct net_device *ndev;
+	struct resource *res;
 	struct port *port;
-	u32 regs_phys;
 	int err;
 
 	plat = dev_get_platdata(dev);
@@ -1380,13 +1381,18 @@ static int ixp4xx_eth_probe(struct platform_device *pdev)
 	port->netdev = ndev;
 	port->id = pdev->id;
 
+	/* Get the port resource and remap */
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!res)
+		return -ENODEV;
+	regs_phys = res->start;
+	port->regs = devm_ioremap_resource(dev, res);
+
 	switch (port->id) {
 	case IXP4XX_ETH_NPEA:
 		/* If the MDIO bus is not up yet, defer probe */
 		if (!mdio_bus)
 			return -EPROBE_DEFER;
-		port->regs = (struct eth_regs __iomem *)IXP4XX_EthA_BASE_VIRT;
-		regs_phys  = IXP4XX_EthA_BASE_PHYS;
 		break;
 	case IXP4XX_ETH_NPEB:
 		/*
@@ -1399,13 +1405,11 @@ static int ixp4xx_eth_probe(struct platform_device *pdev)
 			      IXP4XX_FEATURE_NPEB_ETH0))
 				return -ENODEV;
 			/* Else register the MDIO bus on NPE-B */
-			if ((err = ixp4xx_mdio_register(IXP4XX_EthC_BASE_VIRT)))
+			if ((err = ixp4xx_mdio_register(port->regs)))
 				return err;
 		}
 		if (!mdio_bus)
 			return -EPROBE_DEFER;
-		port->regs = (struct eth_regs __iomem *)IXP4XX_EthB_BASE_VIRT;
-		regs_phys  = IXP4XX_EthB_BASE_PHYS;
 		break;
 	case IXP4XX_ETH_NPEC:
 		/*
@@ -1417,13 +1421,11 @@ static int ixp4xx_eth_probe(struct platform_device *pdev)
 			      IXP4XX_FEATURE_NPEC_ETH))
 				return -ENODEV;
 			/* Else register the MDIO bus on NPE-C */
-			if ((err = ixp4xx_mdio_register(IXP4XX_EthC_BASE_VIRT)))
+			if ((err = ixp4xx_mdio_register(port->regs)))
 				return err;
 		}
 		if (!mdio_bus)
 			return -EPROBE_DEFER;
-		port->regs = (struct eth_regs __iomem *)IXP4XX_EthC_BASE_VIRT;
-		regs_phys  = IXP4XX_EthC_BASE_PHYS;
 		break;
 	default:
 		return -ENODEV;
-- 
cgit v1.2.3-59-g8ed1b


From e45d0fad4a5f8c2d69084dbda98635adcd3f8478 Mon Sep 17 00:00:00 2001
From: Linus Walleij <linus.walleij@linaro.org>
Date: Sun, 12 Jan 2020 13:04:50 +0100
Subject: net: ethernet: ixp4xx: Use parent dev for DMA pool

Use the netdevice struct device .parent field when calling
dma_pool_create(): the .dma_coherent_mask and .dma_mask
pertains to the bus device on the hardware (platform)
bus in this case, not the struct device inside the network
device. This makes the pool allocation work.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/xscale/ixp4xx_eth.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/xscale/ixp4xx_eth.c b/drivers/net/ethernet/xscale/ixp4xx_eth.c
index ee45215c4ba4..269596c15133 100644
--- a/drivers/net/ethernet/xscale/ixp4xx_eth.c
+++ b/drivers/net/ethernet/xscale/ixp4xx_eth.c
@@ -1086,7 +1086,7 @@ static int init_queues(struct port *port)
 	int i;
 
 	if (!ports_open) {
-		dma_pool = dma_pool_create(DRV_NAME, &port->netdev->dev,
+		dma_pool = dma_pool_create(DRV_NAME, port->netdev->dev.parent,
 					   POOL_ALLOC_SIZE, 32, 0);
 		if (!dma_pool)
 			return -ENOMEM;
-- 
cgit v1.2.3-59-g8ed1b


From 2412643a9b02702aab68595e0fa0797ae5ce6176 Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert@linux-m68k.org>
Date: Sun, 12 Jan 2020 17:32:11 +0100
Subject: net: amd: a2065: Kill Sun LANCE relics

Remove unused fields, copied from the Sun LANCE driver eons ago.

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/amd/a2065.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/net/ethernet/amd/a2065.c b/drivers/net/ethernet/amd/a2065.c
index a3faf4feb204..0b1affee87e2 100644
--- a/drivers/net/ethernet/amd/a2065.c
+++ b/drivers/net/ethernet/amd/a2065.c
@@ -118,10 +118,6 @@ struct lance_private {
 	int auto_select;	      /* cable-selection by carrier */
 	unsigned short busmaster_regval;
 
-#ifdef CONFIG_SUNLANCE
-	struct Linux_SBus_DMA *ledma; /* if set this points to ledma and arch=4m */
-	int burst_sizes;	      /* ledma SBus burst sizes */
-#endif
 	struct timer_list         multicast_timer;
 	struct net_device	  *dev;
 };
-- 
cgit v1.2.3-59-g8ed1b


From e07c5f2e4e911d933c8daa3c6f6be063ee0e5c2d Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert@linux-m68k.org>
Date: Sun, 12 Jan 2020 17:33:54 +0100
Subject: net: amd: a2065: Use print_hex_dump_debug() helper

Use the print_hex_dump_debug() helper, instead of open-coding the same
operations.

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/amd/a2065.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/amd/a2065.c b/drivers/net/ethernet/amd/a2065.c
index 0b1affee87e2..2f808dbc8b0e 100644
--- a/drivers/net/ethernet/amd/a2065.c
+++ b/drivers/net/ethernet/amd/a2065.c
@@ -547,11 +547,10 @@ static netdev_tx_t lance_start_xmit(struct sk_buff *skb,
 	if (!lance_tx_buffs_avail(lp))
 		goto out_free;
 
-#ifdef DEBUG
 	/* dump the packet */
-	print_hex_dump(KERN_DEBUG, "skb->data: ", DUMP_PREFIX_NONE,
-		       16, 1, skb->data, 64, true);
-#endif
+	print_hex_dump_debug("skb->data: ", DUMP_PREFIX_NONE, 16, 1, skb->data,
+			     64, true);
+
 	entry = lp->tx_new & lp->tx_ring_mod_mask;
 	ib->btx_ring[entry].length = (-skblen) | 0xf000;
 	ib->btx_ring[entry].misc = 0;
-- 
cgit v1.2.3-59-g8ed1b


From 428cd5232482866d6775a7c6a03dcef78fb0d10e Mon Sep 17 00:00:00 2001
From: Zhang Xiaoxu <zhangxiaoxu5@huawei.com>
Date: Mon, 13 Jan 2020 19:24:11 +0800
Subject: sfc/ethtool_common: Make some function to static

Fix sparse warning:

drivers/net/ethernet/sfc/ethtool_common.c
  warning: symbol 'efx_fill_test' was not declared. Should it be static?
  warning: symbol 'efx_fill_loopback_test' was not declared.
           Should it be static?
  warning: symbol 'efx_describe_per_queue_stats' was not declared.
           Should it be static?

Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Zhang Xiaoxu <zhangxiaoxu5@huawei.com>
Reviewed-by: Martin Habets <mhabets@solarflare.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/sfc/ethtool_common.c | 17 +++++++++--------
 1 file changed, 9 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ethernet/sfc/ethtool_common.c b/drivers/net/ethernet/sfc/ethtool_common.c
index 3d7f75cc5cf0..b8d281ab6c7a 100644
--- a/drivers/net/ethernet/sfc/ethtool_common.c
+++ b/drivers/net/ethernet/sfc/ethtool_common.c
@@ -147,9 +147,9 @@ void efx_ethtool_get_pauseparam(struct net_device *net_dev,
  *
  * Fill in an individual self-test entry.
  */
-void efx_fill_test(unsigned int test_index, u8 *strings, u64 *data,
-		   int *test, const char *unit_format, int unit_id,
-		   const char *test_format, const char *test_id)
+static void efx_fill_test(unsigned int test_index, u8 *strings, u64 *data,
+			  int *test, const char *unit_format, int unit_id,
+			  const char *test_format, const char *test_id)
 {
 	char unit_str[ETH_GSTRING_LEN], test_str[ETH_GSTRING_LEN];
 
@@ -189,10 +189,11 @@ void efx_fill_test(unsigned int test_index, u8 *strings, u64 *data,
  * Fill in a block of loopback self-test entries.  Return new test
  * index.
  */
-int efx_fill_loopback_test(struct efx_nic *efx,
-			   struct efx_loopback_self_tests *lb_tests,
-			   enum efx_loopback_mode mode,
-			   unsigned int test_index, u8 *strings, u64 *data)
+static int efx_fill_loopback_test(struct efx_nic *efx,
+				  struct efx_loopback_self_tests *lb_tests,
+				  enum efx_loopback_mode mode,
+				  unsigned int test_index,
+				  u8 *strings, u64 *data)
 {
 	struct efx_channel *channel =
 		efx_get_channel(efx, efx->tx_channel_offset);
@@ -293,7 +294,7 @@ int efx_ethtool_fill_self_tests(struct efx_nic *efx,
 	return n;
 }
 
-size_t efx_describe_per_queue_stats(struct efx_nic *efx, u8 *strings)
+static size_t efx_describe_per_queue_stats(struct efx_nic *efx, u8 *strings)
 {
 	size_t n_stats = 0;
 	struct efx_channel *channel;
-- 
cgit v1.2.3-59-g8ed1b


From 533420a4151e28e6a38a830cb876afddb5587d7d Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Sun, 12 Jan 2020 23:31:38 -0800
Subject: tools: Sync uapi/linux/if_link.h

Sync uapi/linux/if_link.h into tools to avoid out of sync warnings during
libbpf build.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200113073143.1779940-2-andriin@fb.com
---
 tools/include/uapi/linux/if_link.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tools/include/uapi/linux/if_link.h b/tools/include/uapi/linux/if_link.h
index 8aec8769d944..1d69f637c5d6 100644
--- a/tools/include/uapi/linux/if_link.h
+++ b/tools/include/uapi/linux/if_link.h
@@ -169,6 +169,7 @@ enum {
 	IFLA_MAX_MTU,
 	IFLA_PROP_LIST,
 	IFLA_ALT_IFNAME, /* Alternative ifname */
+	IFLA_PERM_ADDRESS,
 	__IFLA_MAX
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From 292e1d73b125d7a3fd7ff382557e003ece3c0d65 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Sun, 12 Jan 2020 23:31:39 -0800
Subject: libbpf: Clean up bpf_helper_defs.h generation output

bpf_helpers_doc.py script, used to generate bpf_helper_defs.h, unconditionally
emits one informational message to stderr. Remove it and preserve stderr to
contain only relevant errors. Also make sure script invocations command is
muted by default in libbpf's Makefile.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200113073143.1779940-3-andriin@fb.com
---
 scripts/bpf_helpers_doc.py | 2 --
 tools/lib/bpf/Makefile     | 2 +-
 2 files changed, 1 insertion(+), 3 deletions(-)

diff --git a/scripts/bpf_helpers_doc.py b/scripts/bpf_helpers_doc.py
index 7548569e8076..90baf7d70911 100755
--- a/scripts/bpf_helpers_doc.py
+++ b/scripts/bpf_helpers_doc.py
@@ -158,8 +158,6 @@ class HeaderParser(object):
                 break
 
         self.reader.close()
-        print('Parsed description of %d helper function(s)' % len(self.helpers),
-              file=sys.stderr)
 
 ###############################################################################
 
diff --git a/tools/lib/bpf/Makefile b/tools/lib/bpf/Makefile
index db2afccde757..aee7f1a83c77 100644
--- a/tools/lib/bpf/Makefile
+++ b/tools/lib/bpf/Makefile
@@ -183,7 +183,7 @@ $(BPF_IN_STATIC): force elfdep zdep bpfdep $(BPF_HELPER_DEFS)
 	$(Q)$(MAKE) $(build)=libbpf OUTPUT=$(STATIC_OBJDIR)
 
 $(BPF_HELPER_DEFS): $(srctree)/tools/include/uapi/linux/bpf.h
-	$(Q)$(srctree)/scripts/bpf_helpers_doc.py --header 		\
+	$(QUIET_GEN)$(srctree)/scripts/bpf_helpers_doc.py --header \
 		--file $(srctree)/tools/include/uapi/linux/bpf.h > $(BPF_HELPER_DEFS)
 
 $(OUTPUT)libbpf.so: $(OUTPUT)libbpf.so.$(LIBBPF_VERSION)
-- 
cgit v1.2.3-59-g8ed1b


From 2cc51d34d93c9cf76128da6535a2a0a38c9faddb Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Sun, 12 Jan 2020 23:31:40 -0800
Subject: selftests/bpf: Conform selftests/bpf Makefile output to libbpf and
 bpftool

Bring selftest/bpf's Makefile output to the same format used by libbpf and
bpftool: 2 spaces of padding on the left + 8-character left-aligned build step
identifier.

Also, hide feature detection output by default. Can be enabled back by setting
V=1.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200113073143.1779940-4-andriin@fb.com
---
 tools/testing/selftests/bpf/Makefile | 47 +++++++++++++++++++-----------------
 1 file changed, 25 insertions(+), 22 deletions(-)

diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index c28e67548f45..bf9f7e415e95 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -87,14 +87,15 @@ Q =
 msg =
 else
 Q = @
-msg = @$(info $(1)$(if $(2), [$(2)]) $(notdir $(3)))$(if $(4), $(4))
+msg = @printf '  %-8s%s %s%s\n' "$(1)" "$(if $(2), [$(2)])" "$(notdir $(3))" "$(if $(4), $(4))";
 MAKEFLAGS += --no-print-directory
+submake_extras := feature_display=0
 endif
 
 # override lib.mk's default rules
 OVERRIDE_TARGETS := 1
 override define CLEAN
-	$(call msg,    CLEAN)
+	$(call msg,CLEAN)
 	$(RM) -r $(TEST_GEN_PROGS) $(TEST_GEN_PROGS_EXTENDED) $(TEST_GEN_FILES) $(EXTRA_CLEAN)
 	$(MAKE) -C $(BPFDIR) OUTPUT=$(OUTPUT)/ clean
 endef
@@ -112,15 +113,15 @@ $(notdir $(TEST_GEN_PROGS)						\
 	 $(TEST_CUSTOM_PROGS)): %: $(OUTPUT)/% ;
 
 $(OUTPUT)/%:%.c
-	$(call msg,     BINARY,,$@)
+	$(call msg,BINARY,,$@)
 	$(LINK.c) $^ $(LDLIBS) -o $@
 
 $(OUTPUT)/urandom_read: urandom_read.c
-	$(call msg,     BINARY,,$@)
+	$(call msg,BINARY,,$@)
 	$(CC) -o $@ $< -Wl,--build-id
 
 $(OUTPUT)/test_stub.o: test_stub.c
-	$(call msg,         CC,,$@)
+	$(call msg,CC,,$@)
 	$(CC) -c $(CFLAGS) -o $@ $<
 
 BPFOBJ := $(OUTPUT)/libbpf.a
@@ -146,18 +147,20 @@ $(OUTPUT)/test_sysctl: cgroup_helpers.c
 # force a rebuild of BPFOBJ when its dependencies are updated
 force:
 
-DEFAULT_BPFTOOL := $(OUTPUT)/tools/usr/local/sbin/bpftool
+DEFAULT_BPFTOOL := $(OUTPUT)/tools/sbin/bpftool
 BPFTOOL ?= $(DEFAULT_BPFTOOL)
 
 $(DEFAULT_BPFTOOL): force
-	$(Q)$(MAKE) -C $(BPFTOOLDIR) DESTDIR=$(OUTPUT)/tools install
+	$(Q)$(MAKE) $(submake_extras)  -C $(BPFTOOLDIR)			      \
+		    prefix= DESTDIR=$(OUTPUT)/tools/ install
 
 $(BPFOBJ): force
-	$(Q)$(MAKE) -C $(BPFDIR) OUTPUT=$(OUTPUT)/
+	$(Q)$(MAKE) $(submake_extras) -C $(BPFDIR) OUTPUT=$(OUTPUT)/
 
 BPF_HELPERS := $(OUTPUT)/bpf_helper_defs.h $(wildcard $(BPFDIR)/bpf_*.h)
 $(OUTPUT)/bpf_helper_defs.h: $(BPFOBJ)
-	$(Q)$(MAKE) -C $(BPFDIR) OUTPUT=$(OUTPUT)/ $(OUTPUT)/bpf_helper_defs.h
+	$(Q)$(MAKE) $(submake_extras) -C $(BPFDIR)			      \
+		    OUTPUT=$(OUTPUT)/ $(OUTPUT)/bpf_helper_defs.h
 
 # Get Clang's default includes on this system, as opposed to those seen by
 # '-target bpf'. This fixes "missing" files on some architectures/distros,
@@ -194,28 +197,28 @@ $(OUTPUT)/flow_dissector_load.o: flow_dissector_load.h
 # $3 - CFLAGS
 # $4 - LDFLAGS
 define CLANG_BPF_BUILD_RULE
-	$(call msg,  CLANG-LLC,$(TRUNNER_BINARY),$2)
+	$(call msg,CLNG-LLC,$(TRUNNER_BINARY),$2)
 	($(CLANG) $3 -O2 -target bpf -emit-llvm				\
 		-c $1 -o - || echo "BPF obj compilation failed") | 	\
 	$(LLC) -mattr=dwarfris -march=bpf -mcpu=probe $4 -filetype=obj -o $2
 endef
 # Similar to CLANG_BPF_BUILD_RULE, but with disabled alu32
 define CLANG_NOALU32_BPF_BUILD_RULE
-	$(call msg,  CLANG-LLC,$(TRUNNER_BINARY),$2)
+	$(call msg,CLNG-LLC,$(TRUNNER_BINARY),$2)
 	($(CLANG) $3 -O2 -target bpf -emit-llvm				\
 		-c $1 -o - || echo "BPF obj compilation failed") | 	\
 	$(LLC) -march=bpf -mcpu=v2 $4 -filetype=obj -o $2
 endef
 # Similar to CLANG_BPF_BUILD_RULE, but using native Clang and bpf LLC
 define CLANG_NATIVE_BPF_BUILD_RULE
-	$(call msg,  CLANG-BPF,$(TRUNNER_BINARY),$2)
+	$(call msg,CLNG-BPF,$(TRUNNER_BINARY),$2)
 	($(CLANG) $3 -O2 -emit-llvm					\
 		-c $1 -o - || echo "BPF obj compilation failed") | 	\
 	$(LLC) -march=bpf -mcpu=probe $4 -filetype=obj -o $2
 endef
 # Build BPF object using GCC
 define GCC_BPF_BUILD_RULE
-	$(call msg,    GCC-BPF,$(TRUNNER_BINARY),$2)
+	$(call msg,GCC-BPF,$(TRUNNER_BINARY),$2)
 	$(BPF_GCC) $3 $4 -O2 -c $1 -o $2
 endef
 
@@ -256,7 +259,7 @@ define DEFINE_TEST_RUNNER_RULES
 ifeq ($($(TRUNNER_OUTPUT)-dir),)
 $(TRUNNER_OUTPUT)-dir := y
 $(TRUNNER_OUTPUT):
-	$$(call msg,      MKDIR,,$$@)
+	$$(call msg,MKDIR,,$$@)
 	mkdir -p $$@
 endif
 
@@ -275,7 +278,7 @@ $(TRUNNER_BPF_OBJS): $(TRUNNER_OUTPUT)/%.o:				\
 $(TRUNNER_BPF_SKELS): $(TRUNNER_OUTPUT)/%.skel.h:			\
 		      $(TRUNNER_OUTPUT)/%.o				\
 		      | $(BPFTOOL) $(TRUNNER_OUTPUT)
-	$$(call msg,   GEN-SKEL,$(TRUNNER_BINARY),$$@)
+	$$(call msg,GEN-SKEL,$(TRUNNER_BINARY),$$@)
 	$$(BPFTOOL) gen skeleton $$< > $$@
 endif
 
@@ -283,7 +286,7 @@ endif
 ifeq ($($(TRUNNER_TESTS_DIR)-tests-hdr),)
 $(TRUNNER_TESTS_DIR)-tests-hdr := y
 $(TRUNNER_TESTS_HDR): $(TRUNNER_TESTS_DIR)/*.c
-	$$(call msg,   TEST-HDR,$(TRUNNER_BINARY),$$@)
+	$$(call msg,TEST-HDR,$(TRUNNER_BINARY),$$@)
 	$$(shell ( cd $(TRUNNER_TESTS_DIR);				\
 		  echo '/* Generated header, do not edit */';		\
 		  ls *.c 2> /dev/null |					\
@@ -299,7 +302,7 @@ $(TRUNNER_TEST_OBJS): $(TRUNNER_OUTPUT)/%.test.o:			\
 		      $(TRUNNER_BPF_OBJS)				\
 		      $(TRUNNER_BPF_SKELS)				\
 		      $$(BPFOBJ) | $(TRUNNER_OUTPUT)
-	$$(call msg,   TEST-OBJ,$(TRUNNER_BINARY),$$@)
+	$$(call msg,TEST-OBJ,$(TRUNNER_BINARY),$$@)
 	cd $$(@D) && $$(CC) $$(CFLAGS) -c $(CURDIR)/$$< $$(LDLIBS) -o $$(@F)
 
 $(TRUNNER_EXTRA_OBJS): $(TRUNNER_OUTPUT)/%.o:				\
@@ -307,20 +310,20 @@ $(TRUNNER_EXTRA_OBJS): $(TRUNNER_OUTPUT)/%.o:				\
 		       $(TRUNNER_EXTRA_HDRS)				\
 		       $(TRUNNER_TESTS_HDR)				\
 		       $$(BPFOBJ) | $(TRUNNER_OUTPUT)
-	$$(call msg,  EXTRA-OBJ,$(TRUNNER_BINARY),$$@)
+	$$(call msg,EXT-OBJ,$(TRUNNER_BINARY),$$@)
 	$$(CC) $$(CFLAGS) -c $$< $$(LDLIBS) -o $$@
 
 # only copy extra resources if in flavored build
 $(TRUNNER_BINARY)-extras: $(TRUNNER_EXTRA_FILES) | $(TRUNNER_OUTPUT)
 ifneq ($2,)
-	$$(call msg,  EXTRAS-CP,$(TRUNNER_BINARY),$(TRUNNER_EXTRA_FILES))
+	$$(call msg,EXT-COPY,$(TRUNNER_BINARY),$(TRUNNER_EXTRA_FILES))
 	cp -a $$^ $(TRUNNER_OUTPUT)/
 endif
 
 $(OUTPUT)/$(TRUNNER_BINARY): $(TRUNNER_TEST_OBJS)			\
 			     $(TRUNNER_EXTRA_OBJS) $$(BPFOBJ)		\
 			     | $(TRUNNER_BINARY)-extras
-	$$(call msg,     BINARY,,$$@)
+	$$(call msg,BINARY,,$$@)
 	$$(CC) $$(CFLAGS) $$(filter %.a %.o,$$^) $$(LDLIBS) -o $$@
 
 endef
@@ -372,12 +375,12 @@ verifier/tests.h: verifier/*.c
 		  echo '#endif' \
 		) > verifier/tests.h)
 $(OUTPUT)/test_verifier: test_verifier.c verifier/tests.h $(BPFOBJ) | $(OUTPUT)
-	$(call msg,     BINARY,,$@)
+	$(call msg,BINARY,,$@)
 	$(CC) $(CFLAGS) $(filter %.a %.o %.c,$^) $(LDLIBS) -o $@
 
 # Make sure we are able to include and link libbpf against c++.
 $(OUTPUT)/test_cpp: test_cpp.cpp $(OUTPUT)/test_core_extern.skel.h $(BPFOBJ)
-	$(call msg,        CXX,,$@)
+	$(call msg,CXX,,$@)
 	$(CXX) $(CFLAGS) $^ $(LDLIBS) -o $@
 
 EXTRA_CLEAN := $(TEST_CUSTOM_PROGS)					\
-- 
cgit v1.2.3-59-g8ed1b


From 1cf5b23988ea0086a252a5c8b005b075f1e9b030 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Sun, 12 Jan 2020 23:31:41 -0800
Subject: bpftool: Apply preserve_access_index attribute to all types in BTF
 dump

This patch makes structs and unions, emitted through BTF dump, automatically
CO-RE-relocatable (unless disabled with `#define BPF_NO_PRESERVE_ACCESS_INDEX`,
specified before including generated header file).

This effectivaly turns usual bpf_probe_read() call into equivalent of
bpf_core_read(), by automatically applying builtin_preserve_access_index to
any field accesses of types in generated C types header.

This is especially useful for tp_btf/fentry/fexit BPF program types. They
allow direct memory access, so BPF C code just uses straightfoward a->b->c
access pattern to read data from kernel. But without kernel structs marked as
CO-RE relocatable through preserve_access_index attribute, one has to enclose
all the data reads into a special __builtin_preserve_access_index code block,
like so:

__builtin_preserve_access_index(({
    x = p->pid; /* where p is struct task_struct *, for example */
}));

This is very inconvenient and obscures the logic quite a bit. By marking all
auto-generated types with preserve_access_index attribute the above code is
reduced to just a clean and natural `x = p->pid;`.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200113073143.1779940-5-andriin@fb.com
---
 tools/bpf/bpftool/btf.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/tools/bpf/bpftool/btf.c b/tools/bpf/bpftool/btf.c
index e5bc97b71ceb..60c75be0666d 100644
--- a/tools/bpf/bpftool/btf.c
+++ b/tools/bpf/bpftool/btf.c
@@ -370,6 +370,10 @@ static int dump_btf_c(const struct btf *btf,
 	if (IS_ERR(d))
 		return PTR_ERR(d);
 
+	printf("#ifndef BPF_NO_PRESERVE_ACCESS_INDEX\n");
+	printf("#pragma clang attribute push (__attribute__((preserve_access_index)), apply_to = record)\n");
+	printf("#endif\n\n");
+
 	if (root_type_cnt) {
 		for (i = 0; i < root_type_cnt; i++) {
 			err = btf_dump__dump_type(d, root_type_ids[i]);
@@ -386,6 +390,10 @@ static int dump_btf_c(const struct btf *btf,
 		}
 	}
 
+	printf("#ifndef BPF_NO_PRESERVE_ACCESS_INDEX\n");
+	printf("#pragma clang attribute pop\n");
+	printf("#endif\n");
+
 done:
 	btf_dump__free(d);
 	return err;
-- 
cgit v1.2.3-59-g8ed1b


From 9c01546d26d28cf57341d8380ba900bf68e26b18 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Sun, 12 Jan 2020 23:31:42 -0800
Subject: tools/bpf: Add runqslower tool to tools/bpf

Convert one of BCC tools (runqslower [0]) to BPF CO-RE + libbpf. It matches
its BCC-based counterpart 1-to-1, supporting all the same parameters and
functionality.

runqslower tool utilizes BPF skeleton, auto-generated from BPF object file,
as well as memory-mapped interface to global (read-only, in this case) data.
Its Makefile also ensures auto-generation of "relocatable" vmlinux.h, which is
necessary for BTF-typed raw tracepoints with direct memory access.

  [0] https://github.com/iovisor/bcc/blob/11bf5d02c895df9646c117c713082eb192825293/tools/runqslower.py

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200113073143.1779940-6-andriin@fb.com
---
 tools/bpf/Makefile                    |  20 +++-
 tools/bpf/runqslower/.gitignore       |   1 +
 tools/bpf/runqslower/Makefile         |  80 +++++++++++++++
 tools/bpf/runqslower/runqslower.bpf.c | 100 ++++++++++++++++++
 tools/bpf/runqslower/runqslower.c     | 187 ++++++++++++++++++++++++++++++++++
 tools/bpf/runqslower/runqslower.h     |  13 +++
 6 files changed, 396 insertions(+), 5 deletions(-)
 create mode 100644 tools/bpf/runqslower/.gitignore
 create mode 100644 tools/bpf/runqslower/Makefile
 create mode 100644 tools/bpf/runqslower/runqslower.bpf.c
 create mode 100644 tools/bpf/runqslower/runqslower.c
 create mode 100644 tools/bpf/runqslower/runqslower.h

diff --git a/tools/bpf/Makefile b/tools/bpf/Makefile
index 5535650800ab..f897eeeb0b4f 100644
--- a/tools/bpf/Makefile
+++ b/tools/bpf/Makefile
@@ -38,7 +38,7 @@ FEATURE_TESTS = libbfd disassembler-four-args
 FEATURE_DISPLAY = libbfd disassembler-four-args
 
 check_feat := 1
-NON_CHECK_FEAT_TARGETS := clean bpftool_clean
+NON_CHECK_FEAT_TARGETS := clean bpftool_clean runqslower_clean
 ifdef MAKECMDGOALS
 ifeq ($(filter-out $(NON_CHECK_FEAT_TARGETS),$(MAKECMDGOALS)),)
   check_feat := 0
@@ -73,7 +73,7 @@ $(OUTPUT)%.lex.o: $(OUTPUT)%.lex.c
 
 PROGS = $(OUTPUT)bpf_jit_disasm $(OUTPUT)bpf_dbg $(OUTPUT)bpf_asm
 
-all: $(PROGS) bpftool
+all: $(PROGS) bpftool runqslower
 
 $(OUTPUT)bpf_jit_disasm: CFLAGS += -DPACKAGE='bpf_jit_disasm'
 $(OUTPUT)bpf_jit_disasm: $(OUTPUT)bpf_jit_disasm.o
@@ -89,7 +89,7 @@ $(OUTPUT)bpf_exp.lex.c: $(OUTPUT)bpf_exp.yacc.c
 $(OUTPUT)bpf_exp.yacc.o: $(OUTPUT)bpf_exp.yacc.c
 $(OUTPUT)bpf_exp.lex.o: $(OUTPUT)bpf_exp.lex.c
 
-clean: bpftool_clean
+clean: bpftool_clean runqslower_clean
 	$(call QUIET_CLEAN, bpf-progs)
 	$(Q)$(RM) -r -- $(OUTPUT)*.o $(OUTPUT)bpf_jit_disasm $(OUTPUT)bpf_dbg \
 	       $(OUTPUT)bpf_asm $(OUTPUT)bpf_exp.yacc.* $(OUTPUT)bpf_exp.lex.*
@@ -97,7 +97,7 @@ clean: bpftool_clean
 	$(Q)$(RM) -- $(OUTPUT)FEATURE-DUMP.bpf
 	$(Q)$(RM) -r -- $(OUTPUT)feature
 
-install: $(PROGS) bpftool_install
+install: $(PROGS) bpftool_install runqslower_install
 	$(call QUIET_INSTALL, bpf_jit_disasm)
 	$(Q)$(INSTALL) -m 0755 -d $(DESTDIR)$(prefix)/bin
 	$(Q)$(INSTALL) $(OUTPUT)bpf_jit_disasm $(DESTDIR)$(prefix)/bin/bpf_jit_disasm
@@ -115,4 +115,14 @@ bpftool_install:
 bpftool_clean:
 	$(call descend,bpftool,clean)
 
-.PHONY: all install clean bpftool bpftool_install bpftool_clean
+runqslower:
+	$(call descend,runqslower)
+
+runqslower_install:
+	$(call descend,runqslower,install)
+
+runqslower_clean:
+	$(call descend,runqslower,clean)
+
+.PHONY: all install clean bpftool bpftool_install bpftool_clean \
+	runqslower runqslower_install runqslower_clean
diff --git a/tools/bpf/runqslower/.gitignore b/tools/bpf/runqslower/.gitignore
new file mode 100644
index 000000000000..90a456a2a72f
--- /dev/null
+++ b/tools/bpf/runqslower/.gitignore
@@ -0,0 +1 @@
+/.output
diff --git a/tools/bpf/runqslower/Makefile b/tools/bpf/runqslower/Makefile
new file mode 100644
index 000000000000..f1363ae8e473
--- /dev/null
+++ b/tools/bpf/runqslower/Makefile
@@ -0,0 +1,80 @@
+# SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause)
+OUTPUT := .output
+CLANG := clang
+LLC := llc
+LLVM_STRIP := llvm-strip
+DEFAULT_BPFTOOL := $(OUTPUT)/sbin/bpftool
+BPFTOOL ?= $(DEFAULT_BPFTOOL)
+LIBBPF_SRC := $(abspath ../../lib/bpf)
+CFLAGS := -g -Wall
+
+# Try to detect best kernel BTF source
+KERNEL_REL := $(shell uname -r)
+ifneq ("$(wildcard /sys/kenerl/btf/vmlinux)","")
+VMLINUX_BTF := /sys/kernel/btf/vmlinux
+else ifneq ("$(wildcard /boot/vmlinux-$(KERNEL_REL))","")
+VMLINUX_BTF := /boot/vmlinux-$(KERNEL_REL)
+else
+$(error "Can't detect kernel BTF, use VMLINUX_BTF to specify it explicitly")
+endif
+
+abs_out := $(abspath $(OUTPUT))
+ifeq ($(V),1)
+Q =
+msg =
+else
+Q = @
+msg = @printf '  %-8s %s%s\n' "$(1)" "$(notdir $(2))" "$(if $(3), $(3))";
+MAKEFLAGS += --no-print-directory
+submake_extras := feature_display=0
+endif
+
+.DELETE_ON_ERROR:
+
+.PHONY: all clean runqslower
+all: runqslower
+
+runqslower: $(OUTPUT)/runqslower
+
+clean:
+	$(call msg,CLEAN)
+	$(Q)rm -rf $(OUTPUT) runqslower
+
+$(OUTPUT)/runqslower: $(OUTPUT)/runqslower.o $(OUTPUT)/libbpf.a
+	$(call msg,BINARY,$@)
+	$(Q)$(CC) $(CFLAGS) -lelf -lz $^ -o $@
+
+$(OUTPUT)/runqslower.o: runqslower.h $(OUTPUT)/runqslower.skel.h	      \
+			$(OUTPUT)/runqslower.bpf.o
+
+$(OUTPUT)/runqslower.bpf.o: $(OUTPUT)/vmlinux.h runqslower.h
+
+$(OUTPUT)/%.skel.h: $(OUTPUT)/%.bpf.o | $(BPFTOOL)
+	$(call msg,GEN-SKEL,$@)
+	$(Q)$(BPFTOOL) gen skeleton $< > $@
+
+$(OUTPUT)/%.bpf.o: %.bpf.c $(OUTPUT)/libbpf.a | $(OUTPUT)
+	$(call msg,BPF,$@)
+	$(Q)$(CLANG) -g -O2 -target bpf -I$(OUTPUT) -I$(LIBBPF_SRC)	      \
+		 -c $(filter %.c,$^) -o $@ &&				      \
+	$(LLVM_STRIP) -g $@
+
+$(OUTPUT)/%.o: %.c | $(OUTPUT)
+	$(call msg,CC,$@)
+	$(Q)$(CC) $(CFLAGS) -I$(LIBBPF_SRC) -I$(OUTPUT) -c $(filter %.c,$^) -o $@
+
+$(OUTPUT):
+	$(call msg,MKDIR,$@)
+	$(Q)mkdir -p $(OUTPUT)
+
+$(OUTPUT)/vmlinux.h: $(VMLINUX_BTF) | $(OUTPUT) $(BPFTOOL)
+	$(call msg,GEN,$@)
+	$(Q)$(BPFTOOL) btf dump file $(VMLINUX_BTF) format c > $@
+
+$(OUTPUT)/libbpf.a: | $(OUTPUT)
+	$(Q)$(MAKE) $(submake_extras) -C $(LIBBPF_SRC)			       \
+		    OUTPUT=$(abs_out)/ $(abs_out)/libbpf.a
+
+$(DEFAULT_BPFTOOL):
+	$(Q)$(MAKE) $(submake_extras) -C ../bpftool			      \
+		    prefix= OUTPUT=$(abs_out)/ DESTDIR=$(abs_out) install
diff --git a/tools/bpf/runqslower/runqslower.bpf.c b/tools/bpf/runqslower/runqslower.bpf.c
new file mode 100644
index 000000000000..623cce4d37f5
--- /dev/null
+++ b/tools/bpf/runqslower/runqslower.bpf.c
@@ -0,0 +1,100 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2019 Facebook
+#include "vmlinux.h"
+#include <bpf_helpers.h>
+#include "runqslower.h"
+
+#define TASK_RUNNING 0
+
+#define BPF_F_INDEX_MASK		0xffffffffULL
+#define BPF_F_CURRENT_CPU		BPF_F_INDEX_MASK
+
+const volatile __u64 min_us = 0;
+const volatile pid_t targ_pid = 0;
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 10240);
+	__type(key, u32);
+	__type(value, u64);
+} start SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_PERF_EVENT_ARRAY);
+	__uint(key_size, sizeof(u32));
+	__uint(value_size, sizeof(u32));
+} events SEC(".maps");
+
+/* record enqueue timestamp */
+__always_inline
+static int trace_enqueue(u32 tgid, u32 pid)
+{
+	u64 ts;
+
+	if (!pid || (targ_pid && targ_pid != pid))
+		return 0;
+
+	ts = bpf_ktime_get_ns();
+	bpf_map_update_elem(&start, &pid, &ts, 0);
+	return 0;
+}
+
+SEC("tp_btf/sched_wakeup")
+int handle__sched_wakeup(u64 *ctx)
+{
+	/* TP_PROTO(struct task_struct *p) */
+	struct task_struct *p = (void *)ctx[0];
+
+	return trace_enqueue(p->tgid, p->pid);
+}
+
+SEC("tp_btf/sched_wakeup_new")
+int handle__sched_wakeup_new(u64 *ctx)
+{
+	/* TP_PROTO(struct task_struct *p) */
+	struct task_struct *p = (void *)ctx[0];
+
+	return trace_enqueue(p->tgid, p->pid);
+}
+
+SEC("tp_btf/sched_switch")
+int handle__sched_switch(u64 *ctx)
+{
+	/* TP_PROTO(bool preempt, struct task_struct *prev,
+	 *	    struct task_struct *next)
+	 */
+	struct task_struct *prev = (struct task_struct *)ctx[1];
+	struct task_struct *next = (struct task_struct *)ctx[2];
+	struct event event = {};
+	u64 *tsp, delta_us;
+	long state;
+	u32 pid;
+
+	/* ivcsw: treat like an enqueue event and store timestamp */
+	if (prev->state == TASK_RUNNING)
+		trace_enqueue(prev->tgid, prev->pid);
+
+	pid = next->pid;
+
+	/* fetch timestamp and calculate delta */
+	tsp = bpf_map_lookup_elem(&start, &pid);
+	if (!tsp)
+		return 0;   /* missed enqueue */
+
+	delta_us = (bpf_ktime_get_ns() - *tsp) / 1000;
+	if (min_us && delta_us <= min_us)
+		return 0;
+
+	event.pid = pid;
+	event.delta_us = delta_us;
+	bpf_get_current_comm(&event.task, sizeof(event.task));
+
+	/* output */
+	bpf_perf_event_output(ctx, &events, BPF_F_CURRENT_CPU,
+			      &event, sizeof(event));
+
+	bpf_map_delete_elem(&start, &pid);
+	return 0;
+}
+
+char LICENSE[] SEC("license") = "GPL";
diff --git a/tools/bpf/runqslower/runqslower.c b/tools/bpf/runqslower/runqslower.c
new file mode 100644
index 000000000000..996f0e2c560e
--- /dev/null
+++ b/tools/bpf/runqslower/runqslower.c
@@ -0,0 +1,187 @@
+// SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause)
+// Copyright (c) 2019 Facebook
+#include <argp.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/resource.h>
+#include <time.h>
+#include <libbpf.h>
+#include <bpf.h>
+#include "runqslower.h"
+#include "runqslower.skel.h"
+
+struct env {
+	pid_t pid;
+	__u64 min_us;
+	bool verbose;
+} env = {
+	.min_us = 10000,
+};
+
+const char *argp_program_version = "runqslower 0.1";
+const char *argp_program_bug_address = "<bpf@vger.kernel.org>";
+const char argp_program_doc[] =
+"runqslower    Trace long process scheduling delays.\n"
+"              For Linux, uses eBPF, BPF CO-RE, libbpf, BTF.\n"
+"\n"
+"This script traces high scheduling delays between tasks being\n"
+"ready to run and them running on CPU after that.\n"
+"\n"
+"USAGE: runqslower [-p PID] [min_us]\n"
+"\n"
+"EXAMPLES:\n"
+"    runqslower         # trace run queue latency higher than 10000 us (default)\n"
+"    runqslower 1000    # trace run queue latency higher than 1000 us\n"
+"    runqslower -p 123  # trace pid 123 only\n";
+
+static const struct argp_option opts[] = {
+	{ "pid", 'p', "PID", 0, "Process PID to trace"},
+	{ "verbose", 'v', NULL, 0, "Verbose debug output" },
+	{},
+};
+
+static error_t parse_arg(int key, char *arg, struct argp_state *state)
+{
+	static int pos_args;
+	int pid;
+	long long min_us;
+
+	switch (key) {
+	case 'v':
+		env.verbose = true;
+		break;
+	case 'p':
+		errno = 0;
+		pid = strtol(arg, NULL, 10);
+		if (errno || pid <= 0) {
+			fprintf(stderr, "Invalid PID: %s\n", arg);
+			argp_usage(state);
+		}
+		env.pid = pid;
+		break;
+	case ARGP_KEY_ARG:
+		if (pos_args++) {
+			fprintf(stderr,
+				"Unrecognized positional argument: %s\n", arg);
+			argp_usage(state);
+		}
+		errno = 0;
+		min_us = strtoll(arg, NULL, 10);
+		if (errno || min_us <= 0) {
+			fprintf(stderr, "Invalid delay (in us): %s\n", arg);
+			argp_usage(state);
+		}
+		env.min_us = min_us;
+		break;
+	default:
+		return ARGP_ERR_UNKNOWN;
+	}
+	return 0;
+}
+
+int libbpf_print_fn(enum libbpf_print_level level,
+		    const char *format, va_list args)
+{
+	if (level == LIBBPF_DEBUG && !env.verbose)
+		return 0;
+	return vfprintf(stderr, format, args);
+}
+
+static int bump_memlock_rlimit(void)
+{
+	struct rlimit rlim_new = {
+		.rlim_cur	= RLIM_INFINITY,
+		.rlim_max	= RLIM_INFINITY,
+	};
+
+	return setrlimit(RLIMIT_MEMLOCK, &rlim_new);
+}
+
+void handle_event(void *ctx, int cpu, void *data, __u32 data_sz)
+{
+	const struct event *e = data;
+	struct tm *tm;
+	char ts[32];
+	time_t t;
+
+	time(&t);
+	tm = localtime(&t);
+	strftime(ts, sizeof(ts), "%H:%M:%S", tm);
+	printf("%-8s %-16s %-6d %14llu\n", ts, e->task, e->pid, e->delta_us);
+}
+
+void handle_lost_events(void *ctx, int cpu, __u64 lost_cnt)
+{
+	printf("Lost %llu events on CPU #%d!\n", lost_cnt, cpu);
+}
+
+int main(int argc, char **argv)
+{
+	static const struct argp argp = {
+		.options = opts,
+		.parser = parse_arg,
+		.doc = argp_program_doc,
+	};
+	struct perf_buffer_opts pb_opts;
+	struct perf_buffer *pb = NULL;
+	struct runqslower_bpf *obj;
+	int err;
+
+	err = argp_parse(&argp, argc, argv, 0, NULL, NULL);
+	if (err)
+		return err;
+
+	libbpf_set_print(libbpf_print_fn);
+
+	err = bump_memlock_rlimit();
+	if (err) {
+		fprintf(stderr, "failed to increase rlimit: %d", err);
+		return 1;
+	}
+
+	obj = runqslower_bpf__open();
+	if (!obj) {
+		fprintf(stderr, "failed to open and/or load BPF object\n");
+		return 1;
+	}
+
+	/* initialize global data (filtering options) */
+	obj->rodata->targ_pid = env.pid;
+	obj->rodata->min_us = env.min_us;
+
+	err = runqslower_bpf__load(obj);
+	if (err) {
+		fprintf(stderr, "failed to load BPF object: %d\n", err);
+		goto cleanup;
+	}
+
+	err = runqslower_bpf__attach(obj);
+	if (err) {
+		fprintf(stderr, "failed to attach BPF programs\n");
+		goto cleanup;
+	}
+
+	printf("Tracing run queue latency higher than %llu us\n", env.min_us);
+	printf("%-8s %-16s %-6s %14s\n", "TIME", "COMM", "PID", "LAT(us)");
+
+	pb_opts.sample_cb = handle_event;
+	pb_opts.lost_cb = handle_lost_events;
+	pb = perf_buffer__new(bpf_map__fd(obj->maps.events), 64, &pb_opts);
+	err = libbpf_get_error(pb);
+	if (err) {
+		pb = NULL;
+		fprintf(stderr, "failed to open perf buffer: %d\n", err);
+		goto cleanup;
+	}
+
+	while ((err = perf_buffer__poll(pb, 100)) >= 0)
+		;
+	printf("Error polling perf buffer: %d\n", err);
+
+cleanup:
+	perf_buffer__free(pb);
+	runqslower_bpf__destroy(obj);
+
+	return err != 0;
+}
diff --git a/tools/bpf/runqslower/runqslower.h b/tools/bpf/runqslower/runqslower.h
new file mode 100644
index 000000000000..9db225425e5f
--- /dev/null
+++ b/tools/bpf/runqslower/runqslower.h
@@ -0,0 +1,13 @@
+/* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */
+#ifndef __RUNQSLOWER_H
+#define __RUNQSLOWER_H
+
+#define TASK_COMM_LEN 16
+
+struct event {
+	char task[TASK_COMM_LEN];
+	__u64 delta_us;
+	pid_t pid;
+};
+
+#endif /* __RUNQSLOWER_H */
-- 
cgit v1.2.3-59-g8ed1b


From 1657adccaa37e1895356658212a1847f617ebf8f Mon Sep 17 00:00:00 2001
From: Lorenzo Bianconi <lorenzo@kernel.org>
Date: Mon, 13 Jan 2020 10:28:12 +0100
Subject: net: mvneta: change page pool nid to NUMA_NO_NODE

With 'commit 44768decb7c0 ("page_pool: handle page recycle for NUMA_NO_NODE
condition")' we can safely change nid to NUMA_NO_NODE and accommodate
future NUMA aware hardware using mvneta network interface

Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/marvell/mvneta.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index 71a872d46bc4..587e35d10dc7 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -3071,7 +3071,7 @@ static int mvneta_create_page_pool(struct mvneta_port *pp,
 		.order = 0,
 		.flags = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV,
 		.pool_size = size,
-		.nid = cpu_to_node(0),
+		.nid = NUMA_NO_NODE,
 		.dev = pp->dev->dev.parent,
 		.dma_dir = xdp_prog ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE,
 		.offset = pp->rx_offset_correction,
-- 
cgit v1.2.3-59-g8ed1b


From 0eac8ce95bb386838121189b2aa2216cd070f143 Mon Sep 17 00:00:00 2001
From: Jesper Dangaard Brouer <brouer@redhat.com>
Date: Mon, 13 Jan 2020 11:22:16 +0100
Subject: ptr_ring: add include of linux/mm.h

Commit 0bf7800f1799 ("ptr_ring: try vmalloc() when kmalloc() fails")
started to use kvmalloc_array and kvfree, which are defined in mm.h,
the previous functions kcalloc and kfree, which are defined in slab.h.

Add the missing include of linux/mm.h.  This went unnoticed as other
include files happened to include mm.h.

Fixes: 0bf7800f1799 ("ptr_ring: try vmalloc() when kmalloc() fails")
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/ptr_ring.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/include/linux/ptr_ring.h b/include/linux/ptr_ring.h
index 0abe9a4fc842..417db0a79a62 100644
--- a/include/linux/ptr_ring.h
+++ b/include/linux/ptr_ring.h
@@ -23,6 +23,7 @@
 #include <linux/types.h>
 #include <linux/compiler.h>
 #include <linux/slab.h>
+#include <linux/mm.h>
 #include <asm/errno.h>
 #endif
 
-- 
cgit v1.2.3-59-g8ed1b


From ab9837b5ed51f128f898a93d873ba7198abc1140 Mon Sep 17 00:00:00 2001
From: Chen Zhou <chenzhou10@huawei.com>
Date: Mon, 13 Jan 2020 21:15:16 +0800
Subject: amd-xgbe: remove unnecessary conversion to bool

The conversion to bool is not needed, remove it.

Signed-off-by: Chen Zhou <chenzhou10@huawei.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c b/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c
index 128cd648ba99..46c3c1ca38d6 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c
@@ -1227,7 +1227,7 @@ static bool xgbe_phy_sfp_verify_eeprom(u8 cc_in, u8 *buf, unsigned int len)
 	for (cc = 0; len; buf++, len--)
 		cc += *buf;
 
-	return (cc == cc_in) ? true : false;
+	return cc == cc_in;
 }
 
 static int xgbe_phy_sfp_read_eeprom(struct xgbe_prv_data *pdata)
-- 
cgit v1.2.3-59-g8ed1b


From 579a25a854d482bc9d0f9ab0e07ba32fb66bd9e3 Mon Sep 17 00:00:00 2001
From: Jose Abreu <Jose.Abreu@synopsys.com>
Date: Mon, 13 Jan 2020 17:24:09 +0100
Subject: net: stmmac: Initial support for TBS

Adds the initial hooks for TBS support. This needs a 32 byte descriptor
in order for it to work with current HW. Adds all the logic for Enhanced
Descriptors in main core but no HW related logic for now.

Changes from v2:
- Use bitfield for TBS status / support (Jakub)
- Remove unneeded cache alignment (Jakub)
- Fix checkpatch issues

Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/descs.h       |   9 +
 drivers/net/ethernet/stmicro/stmmac/hwif.h        |   7 +
 drivers/net/ethernet/stmicro/stmmac/stmmac.h      |   5 +
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 199 +++++++++++++++-------
 include/linux/stmmac.h                            |   1 +
 5 files changed, 156 insertions(+), 65 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/descs.h b/drivers/net/ethernet/stmicro/stmmac/descs.h
index 9f0b9a9e63b3..49d6a866244f 100644
--- a/drivers/net/ethernet/stmicro/stmmac/descs.h
+++ b/drivers/net/ethernet/stmicro/stmmac/descs.h
@@ -171,6 +171,15 @@ struct dma_extended_desc {
 	__le32 des7;	/* Tx/Rx Timestamp High */
 };
 
+/* Enhanced descriptor for TBS */
+struct dma_edesc {
+	__le32 des4;
+	__le32 des5;
+	__le32 des6;
+	__le32 des7;
+	struct dma_desc basic;
+};
+
 /* Transmit checksum insertion control */
 #define	TX_CIC_FULL	3	/* Include IP header and pseudoheader */
 
diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.h b/drivers/net/ethernet/stmicro/stmmac/hwif.h
index 905a6f0edaca..71c23cbd7af8 100644
--- a/drivers/net/ethernet/stmicro/stmmac/hwif.h
+++ b/drivers/net/ethernet/stmicro/stmmac/hwif.h
@@ -29,6 +29,7 @@ struct stmmac_extra_stats;
 struct stmmac_safety_stats;
 struct dma_desc;
 struct dma_extended_desc;
+struct dma_edesc;
 
 /* Descriptors helpers */
 struct stmmac_desc_ops {
@@ -95,6 +96,7 @@ struct stmmac_desc_ops {
 	void (*set_vlan_tag)(struct dma_desc *p, u16 tag, u16 inner_tag,
 			     u32 inner_type);
 	void (*set_vlan)(struct dma_desc *p, u32 type);
+	void (*set_tbs)(struct dma_edesc *p, u32 sec, u32 nsec);
 };
 
 #define stmmac_init_rx_desc(__priv, __args...) \
@@ -157,6 +159,8 @@ struct stmmac_desc_ops {
 	stmmac_do_void_callback(__priv, desc, set_vlan_tag, __args)
 #define stmmac_set_desc_vlan(__priv, __args...) \
 	stmmac_do_void_callback(__priv, desc, set_vlan, __args)
+#define stmmac_set_desc_tbs(__priv, __args...) \
+	stmmac_do_void_callback(__priv, desc, set_tbs, __args)
 
 struct stmmac_dma_cfg;
 struct dma_features;
@@ -210,6 +214,7 @@ struct stmmac_dma_ops {
 	void (*qmode)(void __iomem *ioaddr, u32 channel, u8 qmode);
 	void (*set_bfsize)(void __iomem *ioaddr, int bfsize, u32 chan);
 	void (*enable_sph)(void __iomem *ioaddr, bool en, u32 chan);
+	int (*enable_tbs)(void __iomem *ioaddr, bool en, u32 chan);
 };
 
 #define stmmac_reset(__priv, __args...) \
@@ -268,6 +273,8 @@ struct stmmac_dma_ops {
 	stmmac_do_void_callback(__priv, dma, set_bfsize, __args)
 #define stmmac_enable_sph(__priv, __args...) \
 	stmmac_do_void_callback(__priv, dma, enable_sph, __args)
+#define stmmac_enable_tbs(__priv, __args...) \
+	stmmac_do_callback(__priv, dma, enable_tbs, __args)
 
 struct mac_device_info;
 struct net_device;
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
index f98c5eefb382..9c02fc754bf1 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
@@ -39,13 +39,18 @@ struct stmmac_tx_info {
 	bool is_jumbo;
 };
 
+#define STMMAC_TBS_AVAIL	BIT(0)
+#define STMMAC_TBS_EN		BIT(1)
+
 /* Frequently used values are kept adjacent for cache effect */
 struct stmmac_tx_queue {
 	u32 tx_count_frames;
+	int tbs;
 	struct timer_list txtimer;
 	u32 queue_index;
 	struct stmmac_priv *priv_data;
 	struct dma_extended_desc *dma_etx ____cacheline_aligned_in_smp;
+	struct dma_edesc *dma_entx;
 	struct dma_desc *dma_tx;
 	struct sk_buff **tx_skbuff;
 	struct stmmac_tx_info *tx_skbuff_dma;
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index f113138df0d9..baffb4e8d99a 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -1090,6 +1090,8 @@ static void stmmac_display_tx_rings(struct stmmac_priv *priv)
 
 		if (priv->extend_desc)
 			head_tx = (void *)tx_q->dma_etx;
+		else if (tx_q->tbs & STMMAC_TBS_AVAIL)
+			head_tx = (void *)tx_q->dma_entx;
 		else
 			head_tx = (void *)tx_q->dma_tx;
 
@@ -1163,13 +1165,19 @@ static void stmmac_clear_tx_descriptors(struct stmmac_priv *priv, u32 queue)
 	int i;
 
 	/* Clear the TX descriptors */
-	for (i = 0; i < DMA_TX_SIZE; i++)
+	for (i = 0; i < DMA_TX_SIZE; i++) {
+		int last = (i == (DMA_TX_SIZE - 1));
+		struct dma_desc *p;
+
 		if (priv->extend_desc)
-			stmmac_init_tx_desc(priv, &tx_q->dma_etx[i].basic,
-					priv->mode, (i == DMA_TX_SIZE - 1));
+			p = &tx_q->dma_etx[i].basic;
+		else if (tx_q->tbs & STMMAC_TBS_AVAIL)
+			p = &tx_q->dma_entx[i].basic;
 		else
-			stmmac_init_tx_desc(priv, &tx_q->dma_tx[i],
-					priv->mode, (i == DMA_TX_SIZE - 1));
+			p = &tx_q->dma_tx[i];
+
+		stmmac_init_tx_desc(priv, p, priv->mode, last);
+	}
 }
 
 /**
@@ -1383,7 +1391,7 @@ static int init_dma_tx_desc_rings(struct net_device *dev)
 			if (priv->extend_desc)
 				stmmac_mode_init(priv, tx_q->dma_etx,
 						tx_q->dma_tx_phy, DMA_TX_SIZE, 1);
-			else
+			else if (!(tx_q->tbs & STMMAC_TBS_AVAIL))
 				stmmac_mode_init(priv, tx_q->dma_tx,
 						tx_q->dma_tx_phy, DMA_TX_SIZE, 0);
 		}
@@ -1392,6 +1400,8 @@ static int init_dma_tx_desc_rings(struct net_device *dev)
 			struct dma_desc *p;
 			if (priv->extend_desc)
 				p = &((tx_q->dma_etx + i)->basic);
+			else if (tx_q->tbs & STMMAC_TBS_AVAIL)
+				p = &((tx_q->dma_entx + i)->basic);
 			else
 				p = tx_q->dma_tx + i;
 
@@ -1511,19 +1521,26 @@ static void free_dma_tx_desc_resources(struct stmmac_priv *priv)
 	/* Free TX queue resources */
 	for (queue = 0; queue < tx_count; queue++) {
 		struct stmmac_tx_queue *tx_q = &priv->tx_queue[queue];
+		size_t size;
+		void *addr;
 
 		/* Release the DMA TX socket buffers */
 		dma_free_tx_skbufs(priv, queue);
 
-		/* Free DMA regions of consistent memory previously allocated */
-		if (!priv->extend_desc)
-			dma_free_coherent(priv->device,
-					  DMA_TX_SIZE * sizeof(struct dma_desc),
-					  tx_q->dma_tx, tx_q->dma_tx_phy);
-		else
-			dma_free_coherent(priv->device, DMA_TX_SIZE *
-					  sizeof(struct dma_extended_desc),
-					  tx_q->dma_etx, tx_q->dma_tx_phy);
+		if (priv->extend_desc) {
+			size = sizeof(struct dma_extended_desc);
+			addr = tx_q->dma_etx;
+		} else if (tx_q->tbs & STMMAC_TBS_AVAIL) {
+			size = sizeof(struct dma_edesc);
+			addr = tx_q->dma_entx;
+		} else {
+			size = sizeof(struct dma_desc);
+			addr = tx_q->dma_tx;
+		}
+
+		size *= DMA_TX_SIZE;
+
+		dma_free_coherent(priv->device, size, addr, tx_q->dma_tx_phy);
 
 		kfree(tx_q->tx_skbuff_dma);
 		kfree(tx_q->tx_skbuff);
@@ -1616,6 +1633,8 @@ static int alloc_dma_tx_desc_resources(struct stmmac_priv *priv)
 	/* TX queues buffers and DMA */
 	for (queue = 0; queue < tx_count; queue++) {
 		struct stmmac_tx_queue *tx_q = &priv->tx_queue[queue];
+		size_t size;
+		void *addr;
 
 		tx_q->queue_index = queue;
 		tx_q->priv_data = priv;
@@ -1632,28 +1651,32 @@ static int alloc_dma_tx_desc_resources(struct stmmac_priv *priv)
 		if (!tx_q->tx_skbuff)
 			goto err_dma;
 
-		if (priv->extend_desc) {
-			tx_q->dma_etx = dma_alloc_coherent(priv->device,
-							   DMA_TX_SIZE * sizeof(struct dma_extended_desc),
-							   &tx_q->dma_tx_phy,
-							   GFP_KERNEL);
-			if (!tx_q->dma_etx)
-				goto err_dma;
-		} else {
-			tx_q->dma_tx = dma_alloc_coherent(priv->device,
-							  DMA_TX_SIZE * sizeof(struct dma_desc),
-							  &tx_q->dma_tx_phy,
-							  GFP_KERNEL);
-			if (!tx_q->dma_tx)
-				goto err_dma;
-		}
+		if (priv->extend_desc)
+			size = sizeof(struct dma_extended_desc);
+		else if (tx_q->tbs & STMMAC_TBS_AVAIL)
+			size = sizeof(struct dma_edesc);
+		else
+			size = sizeof(struct dma_desc);
+
+		size *= DMA_TX_SIZE;
+
+		addr = dma_alloc_coherent(priv->device, size,
+					  &tx_q->dma_tx_phy, GFP_KERNEL);
+		if (!addr)
+			goto err_dma;
+
+		if (priv->extend_desc)
+			tx_q->dma_etx = addr;
+		else if (tx_q->tbs & STMMAC_TBS_AVAIL)
+			tx_q->dma_entx = addr;
+		else
+			tx_q->dma_tx = addr;
 	}
 
 	return 0;
 
 err_dma:
 	free_dma_tx_desc_resources(priv);
-
 	return ret;
 }
 
@@ -1885,6 +1908,8 @@ static int stmmac_tx_clean(struct stmmac_priv *priv, int budget, u32 queue)
 
 		if (priv->extend_desc)
 			p = (struct dma_desc *)(tx_q->dma_etx + entry);
+		else if (tx_q->tbs & STMMAC_TBS_AVAIL)
+			p = &tx_q->dma_entx[entry].basic;
 		else
 			p = tx_q->dma_tx + entry;
 
@@ -1983,19 +2008,12 @@ static int stmmac_tx_clean(struct stmmac_priv *priv, int budget, u32 queue)
 static void stmmac_tx_err(struct stmmac_priv *priv, u32 chan)
 {
 	struct stmmac_tx_queue *tx_q = &priv->tx_queue[chan];
-	int i;
 
 	netif_tx_stop_queue(netdev_get_tx_queue(priv->dev, chan));
 
 	stmmac_stop_tx_dma(priv, chan);
 	dma_free_tx_skbufs(priv, chan);
-	for (i = 0; i < DMA_TX_SIZE; i++)
-		if (priv->extend_desc)
-			stmmac_init_tx_desc(priv, &tx_q->dma_etx[i].basic,
-					priv->mode, (i == DMA_TX_SIZE - 1));
-		else
-			stmmac_init_tx_desc(priv, &tx_q->dma_tx[i],
-					priv->mode, (i == DMA_TX_SIZE - 1));
+	stmmac_clear_tx_descriptors(priv, chan);
 	tx_q->dirty_tx = 0;
 	tx_q->cur_tx = 0;
 	tx_q->mss = 0;
@@ -2632,6 +2650,14 @@ static int stmmac_hw_setup(struct net_device *dev, bool init_ptp)
 	if (priv->dma_cap.vlins)
 		stmmac_enable_vlan(priv, priv->hw, STMMAC_VLAN_INSERT);
 
+	/* TBS */
+	for (chan = 0; chan < tx_cnt; chan++) {
+		struct stmmac_tx_queue *tx_q = &priv->tx_queue[chan];
+		int enable = tx_q->tbs & STMMAC_TBS_AVAIL;
+
+		stmmac_enable_tbs(priv, priv->ioaddr, enable, chan);
+	}
+
 	/* Start the ball rolling... */
 	stmmac_start_all_dma(priv);
 
@@ -2689,6 +2715,16 @@ static int stmmac_open(struct net_device *dev)
 
 	priv->rx_copybreak = STMMAC_RX_COPYBREAK;
 
+	/* Earlier check for TBS */
+	for (chan = 0; chan < priv->plat->tx_queues_to_use; chan++) {
+		struct stmmac_tx_queue *tx_q = &priv->tx_queue[chan];
+		int tbs_en = priv->plat->tx_queues_cfg[chan].tbs_en;
+
+		tx_q->tbs |= tbs_en ? STMMAC_TBS_AVAIL : 0;
+		if (stmmac_enable_tbs(priv, priv->ioaddr, tbs_en, chan))
+			tx_q->tbs &= ~STMMAC_TBS_AVAIL;
+	}
+
 	ret = alloc_dma_desc_resources(priv);
 	if (ret < 0) {
 		netdev_err(priv->dev, "%s: DMA descriptors allocation failed\n",
@@ -2837,7 +2873,11 @@ static bool stmmac_vlan_insert(struct stmmac_priv *priv, struct sk_buff *skb,
 
 	tag = skb_vlan_tag_get(skb);
 
-	p = tx_q->dma_tx + tx_q->cur_tx;
+	if (tx_q->tbs & STMMAC_TBS_AVAIL)
+		p = &tx_q->dma_entx[tx_q->cur_tx].basic;
+	else
+		p = &tx_q->dma_tx[tx_q->cur_tx];
+
 	if (stmmac_set_desc_vlan_tag(priv, p, tag, inner_tag, inner_type))
 		return false;
 
@@ -2872,7 +2912,11 @@ static void stmmac_tso_allocator(struct stmmac_priv *priv, dma_addr_t des,
 
 		tx_q->cur_tx = STMMAC_GET_ENTRY(tx_q->cur_tx, DMA_TX_SIZE);
 		WARN_ON(tx_q->tx_skbuff[tx_q->cur_tx]);
-		desc = tx_q->dma_tx + tx_q->cur_tx;
+
+		if (tx_q->tbs & STMMAC_TBS_AVAIL)
+			desc = &tx_q->dma_entx[tx_q->cur_tx].basic;
+		else
+			desc = &tx_q->dma_tx[tx_q->cur_tx];
 
 		curr_addr = des + (total_len - tmp_len);
 		if (priv->dma_cap.addr64 <= 32)
@@ -2923,13 +2967,13 @@ static netdev_tx_t stmmac_tso_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct dma_desc *desc, *first, *mss_desc = NULL;
 	struct stmmac_priv *priv = netdev_priv(dev);
+	int desc_size, tmp_pay_len = 0, first_tx;
 	int nfrags = skb_shinfo(skb)->nr_frags;
 	u32 queue = skb_get_queue_mapping(skb);
 	unsigned int first_entry, tx_packets;
-	int tmp_pay_len = 0, first_tx;
 	struct stmmac_tx_queue *tx_q;
-	u8 proto_hdr_len, hdr;
 	bool has_vlan, set_ic;
+	u8 proto_hdr_len, hdr;
 	u32 pay_len, mss;
 	dma_addr_t des;
 	int i;
@@ -2966,7 +3010,11 @@ static netdev_tx_t stmmac_tso_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	/* set new MSS value if needed */
 	if (mss != tx_q->mss) {
-		mss_desc = tx_q->dma_tx + tx_q->cur_tx;
+		if (tx_q->tbs & STMMAC_TBS_AVAIL)
+			mss_desc = &tx_q->dma_entx[tx_q->cur_tx].basic;
+		else
+			mss_desc = &tx_q->dma_tx[tx_q->cur_tx];
+
 		stmmac_set_mss(priv, mss_desc, mss);
 		tx_q->mss = mss;
 		tx_q->cur_tx = STMMAC_GET_ENTRY(tx_q->cur_tx, DMA_TX_SIZE);
@@ -2986,7 +3034,10 @@ static netdev_tx_t stmmac_tso_xmit(struct sk_buff *skb, struct net_device *dev)
 	first_entry = tx_q->cur_tx;
 	WARN_ON(tx_q->tx_skbuff[first_entry]);
 
-	desc = tx_q->dma_tx + first_entry;
+	if (tx_q->tbs & STMMAC_TBS_AVAIL)
+		desc = &tx_q->dma_entx[first_entry].basic;
+	else
+		desc = &tx_q->dma_tx[first_entry];
 	first = desc;
 
 	if (has_vlan)
@@ -3058,7 +3109,11 @@ static netdev_tx_t stmmac_tso_xmit(struct sk_buff *skb, struct net_device *dev)
 		set_ic = false;
 
 	if (set_ic) {
-		desc = &tx_q->dma_tx[tx_q->cur_tx];
+		if (tx_q->tbs & STMMAC_TBS_AVAIL)
+			desc = &tx_q->dma_entx[tx_q->cur_tx].basic;
+		else
+			desc = &tx_q->dma_tx[tx_q->cur_tx];
+
 		tx_q->tx_count_frames = 0;
 		stmmac_set_tx_ic(priv, desc);
 		priv->xstats.tx_set_ic_bit++;
@@ -3121,16 +3176,18 @@ static netdev_tx_t stmmac_tso_xmit(struct sk_buff *skb, struct net_device *dev)
 		pr_info("%s: curr=%d dirty=%d f=%d, e=%d, f_p=%p, nfrags %d\n",
 			__func__, tx_q->cur_tx, tx_q->dirty_tx, first_entry,
 			tx_q->cur_tx, first, nfrags);
-
-		stmmac_display_ring(priv, (void *)tx_q->dma_tx, DMA_TX_SIZE, 0);
-
 		pr_info(">>> frame to be transmitted: ");
 		print_pkt(skb->data, skb_headlen(skb));
 	}
 
 	netdev_tx_sent_queue(netdev_get_tx_queue(dev, queue), skb->len);
 
-	tx_q->tx_tail_addr = tx_q->dma_tx_phy + (tx_q->cur_tx * sizeof(*desc));
+	if (tx_q->tbs & STMMAC_TBS_AVAIL)
+		desc_size = sizeof(struct dma_edesc);
+	else
+		desc_size = sizeof(struct dma_desc);
+
+	tx_q->tx_tail_addr = tx_q->dma_tx_phy + (tx_q->cur_tx * desc_size);
 	stmmac_set_tx_tail_ptr(priv, priv->ioaddr, tx_q->tx_tail_addr, queue);
 	stmmac_tx_timer_arm(priv, queue);
 
@@ -3160,10 +3217,11 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
 	u32 queue = skb_get_queue_mapping(skb);
 	int nfrags = skb_shinfo(skb)->nr_frags;
 	int gso = skb_shinfo(skb)->gso_type;
+	struct dma_edesc *tbs_desc = NULL;
+	int entry, desc_size, first_tx;
 	struct dma_desc *desc, *first;
 	struct stmmac_tx_queue *tx_q;
 	bool has_vlan, set_ic;
-	int entry, first_tx;
 	dma_addr_t des;
 
 	tx_q = &priv->tx_queue[queue];
@@ -3203,6 +3261,8 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	if (likely(priv->extend_desc))
 		desc = (struct dma_desc *)(tx_q->dma_etx + entry);
+	else if (tx_q->tbs & STMMAC_TBS_AVAIL)
+		desc = &tx_q->dma_entx[entry].basic;
 	else
 		desc = tx_q->dma_tx + entry;
 
@@ -3232,6 +3292,8 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
 
 		if (likely(priv->extend_desc))
 			desc = (struct dma_desc *)(tx_q->dma_etx + entry);
+		else if (tx_q->tbs & STMMAC_TBS_AVAIL)
+			desc = &tx_q->dma_entx[entry].basic;
 		else
 			desc = tx_q->dma_tx + entry;
 
@@ -3278,6 +3340,8 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
 	if (set_ic) {
 		if (likely(priv->extend_desc))
 			desc = &tx_q->dma_etx[entry].basic;
+		else if (tx_q->tbs & STMMAC_TBS_AVAIL)
+			desc = &tx_q->dma_entx[entry].basic;
 		else
 			desc = &tx_q->dma_tx[entry];
 
@@ -3295,20 +3359,11 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
 	tx_q->cur_tx = entry;
 
 	if (netif_msg_pktdata(priv)) {
-		void *tx_head;
-
 		netdev_dbg(priv->dev,
 			   "%s: curr=%d dirty=%d f=%d, e=%d, first=%p, nfrags=%d",
 			   __func__, tx_q->cur_tx, tx_q->dirty_tx, first_entry,
 			   entry, first, nfrags);
 
-		if (priv->extend_desc)
-			tx_head = (void *)tx_q->dma_etx;
-		else
-			tx_head = (void *)tx_q->dma_tx;
-
-		stmmac_display_ring(priv, tx_head, DMA_TX_SIZE, false);
-
 		netdev_dbg(priv->dev, ">>> frame to be transmitted: ");
 		print_pkt(skb->data, skb->len);
 	}
@@ -3354,12 +3409,19 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
 
 		/* Prepare the first descriptor setting the OWN bit too */
 		stmmac_prepare_tx_desc(priv, first, 1, nopaged_len,
-				csum_insertion, priv->mode, 1, last_segment,
+				csum_insertion, priv->mode, 0, last_segment,
 				skb->len);
-	} else {
-		stmmac_set_tx_owner(priv, first);
 	}
 
+	if (tx_q->tbs & STMMAC_TBS_EN) {
+		struct timespec64 ts = ns_to_timespec64(skb->tstamp);
+
+		tbs_desc = &tx_q->dma_entx[first_entry];
+		stmmac_set_desc_tbs(priv, tbs_desc, ts.tv_sec, ts.tv_nsec);
+	}
+
+	stmmac_set_tx_owner(priv, first);
+
 	/* The own bit must be the latest setting done when prepare the
 	 * descriptor and then barrier is needed to make sure that
 	 * all is coherent before granting the DMA engine.
@@ -3370,7 +3432,14 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	stmmac_enable_dma_transmission(priv, priv->ioaddr);
 
-	tx_q->tx_tail_addr = tx_q->dma_tx_phy + (tx_q->cur_tx * sizeof(*desc));
+	if (likely(priv->extend_desc))
+		desc_size = sizeof(struct dma_extended_desc);
+	else if (tx_q->tbs & STMMAC_TBS_AVAIL)
+		desc_size = sizeof(struct dma_edesc);
+	else
+		desc_size = sizeof(struct dma_desc);
+
+	tx_q->tx_tail_addr = tx_q->dma_tx_phy + (tx_q->cur_tx * desc_size);
 	stmmac_set_tx_tail_ptr(priv, priv->ioaddr, tx_q->tx_tail_addr, queue);
 	stmmac_tx_timer_arm(priv, queue);
 
@@ -4193,7 +4262,7 @@ static int stmmac_rings_status_show(struct seq_file *seq, void *v)
 			seq_printf(seq, "Extended descriptor ring:\n");
 			sysfs_display_ring((void *)tx_q->dma_etx,
 					   DMA_TX_SIZE, 1, seq);
-		} else {
+		} else if (!(tx_q->tbs & STMMAC_TBS_AVAIL)) {
 			seq_printf(seq, "Descriptor ring:\n");
 			sysfs_display_ring((void *)tx_q->dma_tx,
 					   DMA_TX_SIZE, 0, seq);
diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h
index 0531afa9b21e..19190c609282 100644
--- a/include/linux/stmmac.h
+++ b/include/linux/stmmac.h
@@ -139,6 +139,7 @@ struct stmmac_txq_cfg {
 	u32 low_credit;
 	bool use_prio;
 	u32 prio;
+	int tbs_en;
 };
 
 struct plat_stmmacenet_data {
-- 
cgit v1.2.3-59-g8ed1b


From 430b383c737ca07e83e0fe2a8d1ecdff6f4974bd Mon Sep 17 00:00:00 2001
From: Jose Abreu <Jose.Abreu@synopsys.com>
Date: Mon, 13 Jan 2020 17:24:10 +0100
Subject: net: stmmac: tc: Add support for ETF Scheduler using TBS

Adds the support for ETF scheduler using TBS feature which is available
in XGMAC and QoS IPs.

Changes from v2:
- Fix checkpatch issues (Jakub)
- Use the TBS bitfield

Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/common.h      |  1 +
 drivers/net/ethernet/stmicro/stmmac/hwif.h        |  5 +++++
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c |  2 ++
 drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c   | 21 +++++++++++++++++++++
 4 files changed, 29 insertions(+)

diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h
index 31003b67d24f..487099092693 100644
--- a/drivers/net/ethernet/stmicro/stmmac/common.h
+++ b/drivers/net/ethernet/stmicro/stmmac/common.h
@@ -368,6 +368,7 @@ struct dma_features {
 	unsigned int estdep;
 	unsigned int estsel;
 	unsigned int fpesel;
+	unsigned int tbssel;
 };
 
 /* RX Buffer size must be multiple of 4/8/16 bytes */
diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.h b/drivers/net/ethernet/stmicro/stmmac/hwif.h
index 71c23cbd7af8..df63b0367aff 100644
--- a/drivers/net/ethernet/stmicro/stmmac/hwif.h
+++ b/drivers/net/ethernet/stmicro/stmmac/hwif.h
@@ -533,6 +533,7 @@ struct tc_cls_u32_offload;
 struct tc_cbs_qopt_offload;
 struct flow_cls_offload;
 struct tc_taprio_qopt_offload;
+struct tc_etf_qopt_offload;
 
 struct stmmac_tc_ops {
 	int (*init)(struct stmmac_priv *priv);
@@ -544,6 +545,8 @@ struct stmmac_tc_ops {
 			 struct flow_cls_offload *cls);
 	int (*setup_taprio)(struct stmmac_priv *priv,
 			    struct tc_taprio_qopt_offload *qopt);
+	int (*setup_etf)(struct stmmac_priv *priv,
+			 struct tc_etf_qopt_offload *qopt);
 };
 
 #define stmmac_tc_init(__priv, __args...) \
@@ -556,6 +559,8 @@ struct stmmac_tc_ops {
 	stmmac_do_callback(__priv, tc, setup_cls, __args)
 #define stmmac_tc_setup_taprio(__priv, __args...) \
 	stmmac_do_callback(__priv, tc, setup_taprio, __args)
+#define stmmac_tc_setup_etf(__priv, __args...) \
+	stmmac_do_callback(__priv, tc, setup_etf, __args)
 
 struct stmmac_counters;
 
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index baffb4e8d99a..43af4fc5ab8f 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -4159,6 +4159,8 @@ static int stmmac_setup_tc(struct net_device *ndev, enum tc_setup_type type,
 		return stmmac_tc_setup_cbs(priv, priv, type_data);
 	case TC_SETUP_QDISC_TAPRIO:
 		return stmmac_tc_setup_taprio(priv, priv, type_data);
+	case TC_SETUP_QDISC_ETF:
+		return stmmac_tc_setup_etf(priv, priv, type_data);
 	default:
 		return -EOPNOTSUPP;
 	}
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c
index 6c4686b77516..a4ce165af36b 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c
@@ -727,10 +727,31 @@ disable:
 	return ret;
 }
 
+static int tc_setup_etf(struct stmmac_priv *priv,
+			struct tc_etf_qopt_offload *qopt)
+{
+	if (!priv->dma_cap.tbssel)
+		return -EOPNOTSUPP;
+	if (qopt->queue >= priv->plat->tx_queues_to_use)
+		return -EINVAL;
+	if (!(priv->tx_queue[qopt->queue].tbs & STMMAC_TBS_AVAIL))
+		return -EINVAL;
+
+	if (qopt->enable)
+		priv->tx_queue[qopt->queue].tbs |= STMMAC_TBS_EN;
+	else
+		priv->tx_queue[qopt->queue].tbs &= ~STMMAC_TBS_EN;
+
+	netdev_info(priv->dev, "%s ETF for Queue %d\n",
+		    qopt->enable ? "enabled" : "disabled", qopt->queue);
+	return 0;
+}
+
 const struct stmmac_tc_ops dwmac510_tc_ops = {
 	.init = tc_init,
 	.setup_cls_u32 = tc_setup_cls_u32,
 	.setup_cbs = tc_setup_cbs,
 	.setup_cls = tc_setup_cls,
 	.setup_taprio = tc_setup_taprio,
+	.setup_etf = tc_setup_etf,
 };
-- 
cgit v1.2.3-59-g8ed1b


From 6a549b9f0dcd9a73991077da64eb3603b3e5f681 Mon Sep 17 00:00:00 2001
From: Jose Abreu <Jose.Abreu@synopsys.com>
Date: Mon, 13 Jan 2020 17:24:11 +0100
Subject: net: stmmac: xgmac: Add TBS support

Adds all the necessary HW hooks to support TBS feature in XGMAC cores.

Changes from v1:
- Remove unneeded LT shift as the IP already does this.

Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h     | 13 ++++++++++++
 .../net/ethernet/stmicro/stmmac/dwxgmac2_descs.c   |  9 ++++++++
 drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c | 24 ++++++++++++++++++++++
 3 files changed, 46 insertions(+)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h
index 64d13e50e403..6c3b8a950f58 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h
+++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h
@@ -139,6 +139,7 @@
 #define XGMAC_HWFEAT_TXQCNT		GENMASK(9, 6)
 #define XGMAC_HWFEAT_RXQCNT		GENMASK(3, 0)
 #define XGMAC_HW_FEATURE3		0x00000128
+#define XGMAC_HWFEAT_TBSSEL		BIT(27)
 #define XGMAC_HWFEAT_FPESEL		BIT(26)
 #define XGMAC_HWFEAT_ESTWID		GENMASK(24, 23)
 #define XGMAC_HWFEAT_ESTDEP		GENMASK(22, 20)
@@ -346,6 +347,13 @@
 #define XGMAC_TDPS			GENMASK(29, 0)
 #define XGMAC_RX_EDMA_CTRL		0x00003044
 #define XGMAC_RDPS			GENMASK(29, 0)
+#define XGMAC_DMA_TBS_CTRL0		0x00003054
+#define XGMAC_DMA_TBS_CTRL1		0x00003058
+#define XGMAC_DMA_TBS_CTRL2		0x0000305c
+#define XGMAC_DMA_TBS_CTRL3		0x00003060
+#define XGMAC_FTOS			GENMASK(31, 8)
+#define XGMAC_FTOV			BIT(0)
+#define XGMAC_DEF_FTOS			(XGMAC_FTOS | XGMAC_FTOV)
 #define XGMAC_DMA_SAFETY_INT_STATUS	0x00003064
 #define XGMAC_MCSIS			BIT(31)
 #define XGMAC_MSUIS			BIT(29)
@@ -360,6 +368,7 @@
 #define XGMAC_SPH			BIT(24)
 #define XGMAC_PBLx8			BIT(16)
 #define XGMAC_DMA_CH_TX_CONTROL(x)	(0x00003104 + (0x80 * (x)))
+#define XGMAC_EDSE			BIT(28)
 #define XGMAC_TxPBL			GENMASK(21, 16)
 #define XGMAC_TxPBL_SHIFT		16
 #define XGMAC_TSE			BIT(12)
@@ -404,6 +413,9 @@
 #define XGMAC_REGSIZE			((0x0000317c + (0x80 * 15)) / 4)
 
 /* Descriptors */
+#define XGMAC_TDES0_LTV			BIT(31)
+#define XGMAC_TDES0_LT			GENMASK(7, 0)
+#define XGMAC_TDES1_LT			GENMASK(31, 8)
 #define XGMAC_TDES2_IVT			GENMASK(31, 16)
 #define XGMAC_TDES2_IVT_SHIFT		16
 #define XGMAC_TDES2_IOC			BIT(31)
@@ -422,6 +434,7 @@
 #define XGMAC_TDES3_TCMSSV		BIT(26)
 #define XGMAC_TDES3_SAIC		GENMASK(25, 23)
 #define XGMAC_TDES3_SAIC_SHIFT		23
+#define XGMAC_TDES3_TBSV		BIT(24)
 #define XGMAC_TDES3_THL			GENMASK(22, 19)
 #define XGMAC_TDES3_THL_SHIFT		19
 #define XGMAC_TDES3_IVTIR		GENMASK(19, 18)
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_descs.c b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_descs.c
index bd5838ce1e8a..c3d654cfa9ef 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_descs.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_descs.c
@@ -339,6 +339,14 @@ static void dwxgmac2_set_vlan(struct dma_desc *p, u32 type)
 	p->des2 |= cpu_to_le32(type & XGMAC_TDES2_VTIR);
 }
 
+static void dwxgmac2_set_tbs(struct dma_edesc *p, u32 sec, u32 nsec)
+{
+	p->des4 = cpu_to_le32((sec & XGMAC_TDES0_LT) | XGMAC_TDES0_LTV);
+	p->des5 = cpu_to_le32(nsec & XGMAC_TDES1_LT);
+	p->des6 = 0;
+	p->des7 = 0;
+}
+
 const struct stmmac_desc_ops dwxgmac210_desc_ops = {
 	.tx_status = dwxgmac2_get_tx_status,
 	.rx_status = dwxgmac2_get_rx_status,
@@ -368,4 +376,5 @@ const struct stmmac_desc_ops dwxgmac210_desc_ops = {
 	.set_sarc = dwxgmac2_set_sarc,
 	.set_vlan_tag = dwxgmac2_set_vlan_tag,
 	.set_vlan = dwxgmac2_set_vlan,
+	.set_tbs = dwxgmac2_set_tbs,
 };
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c
index bbbfa793a367..77308c5c5d29 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c
@@ -429,6 +429,7 @@ static void dwxgmac2_get_hw_feature(void __iomem *ioaddr,
 
 	/* MAC HW feature 3 */
 	hw_cap = readl(ioaddr + XGMAC_HW_FEATURE3);
+	dma_cap->tbssel = (hw_cap & XGMAC_HWFEAT_TBSSEL) >> 27;
 	dma_cap->fpesel = (hw_cap & XGMAC_HWFEAT_FPESEL) >> 26;
 	dma_cap->estwid = (hw_cap & XGMAC_HWFEAT_ESTWID) >> 23;
 	dma_cap->estdep = (hw_cap & XGMAC_HWFEAT_ESTDEP) >> 20;
@@ -523,6 +524,28 @@ static void dwxgmac2_enable_sph(void __iomem *ioaddr, bool en, u32 chan)
 	writel(value, ioaddr + XGMAC_DMA_CH_CONTROL(chan));
 }
 
+static int dwxgmac2_enable_tbs(void __iomem *ioaddr, bool en, u32 chan)
+{
+	u32 value = readl(ioaddr + XGMAC_DMA_CH_TX_CONTROL(chan));
+
+	if (en)
+		value |= XGMAC_EDSE;
+	else
+		value &= ~XGMAC_EDSE;
+
+	writel(value, ioaddr + XGMAC_DMA_CH_TX_CONTROL(chan));
+
+	value = readl(ioaddr + XGMAC_DMA_CH_TX_CONTROL(chan)) & XGMAC_EDSE;
+	if (en && !value)
+		return -EIO;
+
+	writel(XGMAC_DEF_FTOS, ioaddr + XGMAC_DMA_TBS_CTRL0);
+	writel(XGMAC_DEF_FTOS, ioaddr + XGMAC_DMA_TBS_CTRL1);
+	writel(XGMAC_DEF_FTOS, ioaddr + XGMAC_DMA_TBS_CTRL2);
+	writel(XGMAC_DEF_FTOS, ioaddr + XGMAC_DMA_TBS_CTRL3);
+	return 0;
+}
+
 const struct stmmac_dma_ops dwxgmac210_dma_ops = {
 	.reset = dwxgmac2_dma_reset,
 	.init = dwxgmac2_dma_init,
@@ -550,4 +573,5 @@ const struct stmmac_dma_ops dwxgmac210_dma_ops = {
 	.qmode = dwxgmac2_qmode,
 	.set_bfsize = dwxgmac2_set_bfsize,
 	.enable_sph = dwxgmac2_enable_sph,
+	.enable_tbs = dwxgmac2_enable_tbs,
 };
-- 
cgit v1.2.3-59-g8ed1b


From 58ae92814008a324fc5698fa76fcd7497207fe0f Mon Sep 17 00:00:00 2001
From: Jose Abreu <Jose.Abreu@synopsys.com>
Date: Mon, 13 Jan 2020 17:24:12 +0100
Subject: net: stmmac: gmac4+: Add TBS support

Adds all the necessary HW hooks to support TBS feature in QoS cores.

Changes from v1:
- Remove unneeded LT shift as the IP already does this.

Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac4.h       |  1 +
 drivers/net/ethernet/stmicro/stmmac/dwmac4_descs.c | 10 ++++++++++
 drivers/net/ethernet/stmicro/stmmac/dwmac4_descs.h |  7 +++++++
 drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c   | 21 +++++++++++++++++++++
 drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h   |  7 +++++++
 5 files changed, 46 insertions(+)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4.h b/drivers/net/ethernet/stmicro/stmmac/dwmac4.h
index 2e6b60a476c6..af50af27550b 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac4.h
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4.h
@@ -239,6 +239,7 @@ enum power_event {
 
 /* MAC HW features3 bitmap */
 #define GMAC_HW_FEAT_ASP		GENMASK(29, 28)
+#define GMAC_HW_FEAT_TBSSEL		BIT(27)
 #define GMAC_HW_FEAT_FPESEL		BIT(26)
 #define GMAC_HW_FEAT_ESTWID		GENMASK(21, 20)
 #define GMAC_HW_FEAT_ESTDEP		GENMASK(19, 17)
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_descs.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_descs.c
index 3e14da69f378..eff82065a501 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_descs.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_descs.c
@@ -10,6 +10,7 @@
 
 #include <linux/stmmac.h>
 #include "common.h"
+#include "dwmac4.h"
 #include "dwmac4_descs.h"
 
 static int dwmac4_wrback_get_tx_status(void *data, struct stmmac_extra_stats *x,
@@ -505,6 +506,14 @@ static void dwmac4_set_sec_addr(struct dma_desc *p, dma_addr_t addr)
 	p->des3 = cpu_to_le32(upper_32_bits(addr) | RDES3_BUFFER2_VALID_ADDR);
 }
 
+static void dwmac4_set_tbs(struct dma_edesc *p, u32 sec, u32 nsec)
+{
+	p->des4 = cpu_to_le32((sec & TDES4_LT) | TDES4_LTV);
+	p->des5 = cpu_to_le32(nsec & TDES5_LT);
+	p->des6 = 0;
+	p->des7 = 0;
+}
+
 const struct stmmac_desc_ops dwmac4_desc_ops = {
 	.tx_status = dwmac4_wrback_get_tx_status,
 	.rx_status = dwmac4_wrback_get_rx_status,
@@ -534,6 +543,7 @@ const struct stmmac_desc_ops dwmac4_desc_ops = {
 	.set_vlan = dwmac4_set_vlan,
 	.get_rx_header_len = dwmac4_get_rx_header_len,
 	.set_sec_addr = dwmac4_set_sec_addr,
+	.set_tbs = dwmac4_set_tbs,
 };
 
 const struct stmmac_mode_ops dwmac4_ring_mode_ops = {
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_descs.h b/drivers/net/ethernet/stmicro/stmmac/dwmac4_descs.h
index 6d92109dc9aa..6da070ccd737 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_descs.h
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_descs.h
@@ -73,6 +73,13 @@
 #define TDES3_CONTEXT_TYPE		BIT(30)
 #define	TDES3_CONTEXT_TYPE_SHIFT	30
 
+/* TDES4 */
+#define TDES4_LTV			BIT(31)
+#define TDES4_LT			GENMASK(7, 0)
+
+/* TDES5 */
+#define TDES5_LT			GENMASK(31, 8)
+
 /* TDS3 use for both format (read and write back) */
 #define TDES3_OWN			BIT(31)
 #define TDES3_OWN_SHIFT			31
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c
index 213d44482ffa..bb29bfcd62c3 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c
@@ -404,6 +404,7 @@ static void dwmac4_get_hw_feature(void __iomem *ioaddr,
 
 	/* 5.10 Features */
 	dma_cap->asp = (hw_cap & GMAC_HW_FEAT_ASP) >> 28;
+	dma_cap->tbssel = (hw_cap & GMAC_HW_FEAT_TBSSEL) >> 27;
 	dma_cap->fpesel = (hw_cap & GMAC_HW_FEAT_FPESEL) >> 26;
 	dma_cap->estwid = (hw_cap & GMAC_HW_FEAT_ESTWID) >> 20;
 	dma_cap->estdep = (hw_cap & GMAC_HW_FEAT_ESTDEP) >> 17;
@@ -471,6 +472,25 @@ static void dwmac4_enable_sph(void __iomem *ioaddr, bool en, u32 chan)
 	writel(value, ioaddr + DMA_CHAN_CONTROL(chan));
 }
 
+static int dwmac4_enable_tbs(void __iomem *ioaddr, bool en, u32 chan)
+{
+	u32 value = readl(ioaddr + DMA_CHAN_TX_CONTROL(chan));
+
+	if (en)
+		value |= DMA_CONTROL_EDSE;
+	else
+		value &= ~DMA_CONTROL_EDSE;
+
+	writel(value, ioaddr + DMA_CHAN_TX_CONTROL(chan));
+
+	value = readl(ioaddr + DMA_CHAN_TX_CONTROL(chan)) & DMA_CONTROL_EDSE;
+	if (en && !value)
+		return -EIO;
+
+	writel(DMA_TBS_DEF_FTOS, ioaddr + DMA_TBS_CTRL);
+	return 0;
+}
+
 const struct stmmac_dma_ops dwmac4_dma_ops = {
 	.reset = dwmac4_dma_reset,
 	.init = dwmac4_dma_init,
@@ -527,4 +547,5 @@ const struct stmmac_dma_ops dwmac410_dma_ops = {
 	.qmode = dwmac4_qmode,
 	.set_bfsize = dwmac4_set_bfsize,
 	.enable_sph = dwmac4_enable_sph,
+	.enable_tbs = dwmac4_enable_tbs,
 };
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h b/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h
index bcb6d5190f3d..8391ca63d943 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h
@@ -22,6 +22,7 @@
 #define DMA_DEBUG_STATUS_1		0x00001010
 #define DMA_DEBUG_STATUS_2		0x00001014
 #define DMA_AXI_BUS_MODE		0x00001028
+#define DMA_TBS_CTRL			0x00001050
 
 /* DMA Bus Mode bitmap */
 #define DMA_BUS_MODE_SFT_RESET		BIT(0)
@@ -82,6 +83,11 @@
 
 #define DMA_AXI_BURST_LEN_MASK		0x000000FE
 
+/* DMA TBS Control */
+#define DMA_TBS_FTOS			GENMASK(31, 8)
+#define DMA_TBS_FTOV			BIT(0)
+#define DMA_TBS_DEF_FTOS		(DMA_TBS_FTOS | DMA_TBS_FTOV)
+
 /* Following DMA defines are chanels oriented */
 #define DMA_CHAN_BASE_ADDR		0x00001100
 #define DMA_CHAN_BASE_OFFSET		0x80
@@ -114,6 +120,7 @@
 #define DMA_CONTROL_MSS_MASK		GENMASK(13, 0)
 
 /* DMA Tx Channel X Control register defines */
+#define DMA_CONTROL_EDSE		BIT(28)
 #define DMA_CONTROL_TSE			BIT(12)
 #define DMA_CONTROL_OSP			BIT(4)
 #define DMA_CONTROL_ST			BIT(0)
-- 
cgit v1.2.3-59-g8ed1b


From 7eadf57290ecd2b8e01a48dfaf55126ad60e6a3b Mon Sep 17 00:00:00 2001
From: Jose Abreu <Jose.Abreu@synopsys.com>
Date: Mon, 13 Jan 2020 17:24:13 +0100
Subject: net: stmmac: pci: Enable TBS on GMAC5 IPK PCI entry

Enable TBS support on GMAC5 PCI entry for all Queues except Queue 0.

Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c
index 40f171d310d7..623521052152 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c
@@ -401,6 +401,8 @@ static int snps_gmac5_default_data(struct pci_dev *pdev,
 		plat->tx_queues_cfg[i].use_prio = false;
 		plat->tx_queues_cfg[i].mode_to_use = MTL_QUEUE_DCB;
 		plat->tx_queues_cfg[i].weight = 25;
+		if (i > 0)
+			plat->tx_queues_cfg[i].tbs_en = 1;
 	}
 
 	plat->rx_sched_algorithm = MTL_RX_ALGORITHM_SP;
-- 
cgit v1.2.3-59-g8ed1b


From 44e6547570654a447edcfff33023cf61f5fdcd01 Mon Sep 17 00:00:00 2001
From: Jose Abreu <Jose.Abreu@synopsys.com>
Date: Mon, 13 Jan 2020 17:24:14 +0100
Subject: net: stmmac: Add missing information in DebugFS capabilities file

Adds more information regarding HW Capabilities in the corresponding
DebugFS file.

Changes from v2:
- Remove the TX/RX queues in use (Jakub)

Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 43af4fc5ab8f..2738d97495e6 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -4365,6 +4365,12 @@ static int stmmac_dma_cap_show(struct seq_file *seq, void *v)
 		   priv->dma_cap.l3l4fnum);
 	seq_printf(seq, "\tARP Offloading: %s\n",
 		   priv->dma_cap.arpoffsel ? "Y" : "N");
+	seq_printf(seq, "\tEnhancements to Scheduled Traffic (EST): %s\n",
+		   priv->dma_cap.estsel ? "Y" : "N");
+	seq_printf(seq, "\tFrame Preemption (FPE): %s\n",
+		   priv->dma_cap.fpesel ? "Y" : "N");
+	seq_printf(seq, "\tTime-Based Scheduling (TBS): %s\n",
+		   priv->dma_cap.tbssel ? "Y" : "N");
 	return 0;
 }
 DEFINE_SHOW_ATTRIBUTE(stmmac_dma_cap);
-- 
cgit v1.2.3-59-g8ed1b


From 05373e31baed87c3daafdc6d30dc679f057b5cd0 Mon Sep 17 00:00:00 2001
From: Jose Abreu <Jose.Abreu@synopsys.com>
Date: Mon, 13 Jan 2020 17:24:15 +0100
Subject: net: stmmac: selftests: Switch to dev_direct_xmit()

In the upcoming commit for TBS selftest we will need to send a packet on
a specific Queue. As stmmac fallsback to netdev_pick_tx() on the select
Queue callback, we need to switch all selftests logic to
dev_direct_xmit() so that we can send the given SKB on a specific Queue.

Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../net/ethernet/stmicro/stmmac/stmmac_selftests.c | 25 ++++++----------------
 1 file changed, 6 insertions(+), 19 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c
index 13227909287c..d50ae59fe3d8 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c
@@ -339,8 +339,7 @@ static int __stmmac_test_loopback(struct stmmac_priv *priv,
 		goto cleanup;
 	}
 
-	skb_set_queue_mapping(skb, attr->queue_mapping);
-	ret = dev_queue_xmit(skb);
+	ret = dev_direct_xmit(skb, attr->queue_mapping);
 	if (ret)
 		goto cleanup;
 
@@ -926,8 +925,7 @@ static int __stmmac_test_vlanfilt(struct stmmac_priv *priv)
 			goto vlan_del;
 		}
 
-		skb_set_queue_mapping(skb, 0);
-		ret = dev_queue_xmit(skb);
+		ret = dev_direct_xmit(skb, 0);
 		if (ret)
 			goto vlan_del;
 
@@ -1018,8 +1016,7 @@ static int __stmmac_test_dvlanfilt(struct stmmac_priv *priv)
 			goto vlan_del;
 		}
 
-		skb_set_queue_mapping(skb, 0);
-		ret = dev_queue_xmit(skb);
+		ret = dev_direct_xmit(skb, 0);
 		if (ret)
 			goto vlan_del;
 
@@ -1286,8 +1283,7 @@ static int stmmac_test_vlanoff_common(struct stmmac_priv *priv, bool svlan)
 	__vlan_hwaccel_put_tag(skb, htons(proto), tpriv->vlan_id);
 	skb->protocol = htons(proto);
 
-	skb_set_queue_mapping(skb, 0);
-	ret = dev_queue_xmit(skb);
+	ret = dev_direct_xmit(skb, 0);
 	if (ret)
 		goto vlan_del;
 
@@ -1639,8 +1635,7 @@ static int stmmac_test_arpoffload(struct stmmac_priv *priv)
 	if (ret)
 		goto cleanup;
 
-	skb_set_queue_mapping(skb, 0);
-	ret = dev_queue_xmit(skb);
+	ret = dev_direct_xmit(skb, 0);
 	if (ret)
 		goto cleanup_promisc;
 
@@ -1869,7 +1864,6 @@ void stmmac_selftest_run(struct net_device *dev,
 {
 	struct stmmac_priv *priv = netdev_priv(dev);
 	int count = stmmac_selftest_get_count(priv);
-	int carrier = netif_carrier_ok(dev);
 	int i, ret;
 
 	memset(buf, 0, sizeof(*buf) * count);
@@ -1879,15 +1873,12 @@ void stmmac_selftest_run(struct net_device *dev,
 		netdev_err(priv->dev, "Only offline tests are supported\n");
 		etest->flags |= ETH_TEST_FL_FAILED;
 		return;
-	} else if (!carrier) {
+	} else if (!netif_carrier_ok(dev)) {
 		netdev_err(priv->dev, "You need valid Link to execute tests\n");
 		etest->flags |= ETH_TEST_FL_FAILED;
 		return;
 	}
 
-	/* We don't want extra traffic */
-	netif_carrier_off(dev);
-
 	/* Wait for queues drain */
 	msleep(200);
 
@@ -1942,10 +1933,6 @@ void stmmac_selftest_run(struct net_device *dev,
 			break;
 		}
 	}
-
-	/* Restart everything */
-	if (carrier)
-		netif_carrier_on(dev);
 }
 
 void stmmac_selftest_get_strings(struct stmmac_priv *priv, u8 *data)
-- 
cgit v1.2.3-59-g8ed1b


From 28c1cf73c923065d1265ea32b9c1bfc8d63370e9 Mon Sep 17 00:00:00 2001
From: Jose Abreu <Jose.Abreu@synopsys.com>
Date: Mon, 13 Jan 2020 17:24:16 +0100
Subject: net: stmmac: selftests: Add a test for TBS feature

Add a new test for TBS feature which is used in ETF scheduler. In this
test, we send a packet with a launch time specified as now + 500ms and
check if the packet was transmitted on that time frame.

Changes from v2:
- Use the TBS bitfield
- Remove debug message

Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../net/ethernet/stmicro/stmmac/stmmac_selftests.c | 71 ++++++++++++++++++++++
 1 file changed, 71 insertions(+)

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c
index d50ae59fe3d8..5a3aea628fc2 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c
@@ -14,6 +14,7 @@
 #include <linux/phy.h>
 #include <linux/udp.h>
 #include <net/pkt_cls.h>
+#include <net/pkt_sched.h>
 #include <net/tcp.h>
 #include <net/udp.h>
 #include <net/tc_act/tc_gact.h>
@@ -50,6 +51,7 @@ struct stmmac_packet_attrs {
 	u8 id;
 	int sarc;
 	u16 queue_mapping;
+	u64 timestamp;
 };
 
 static u8 stmmac_test_next_id;
@@ -208,6 +210,9 @@ static struct sk_buff *stmmac_test_get_udp_skb(struct stmmac_priv *priv,
 	skb->pkt_type = PACKET_HOST;
 	skb->dev = priv->dev;
 
+	if (attr->timestamp)
+		skb->tstamp = ns_to_ktime(attr->timestamp);
+
 	return skb;
 }
 
@@ -1723,6 +1728,68 @@ static int stmmac_test_sph(struct stmmac_priv *priv)
 	return 0;
 }
 
+static int stmmac_test_tbs(struct stmmac_priv *priv)
+{
+#define STMMAC_TBS_LT_OFFSET		(500 * 1000 * 1000) /* 500 ms*/
+	struct stmmac_packet_attrs attr = { };
+	struct tc_etf_qopt_offload qopt;
+	u64 start_time, curr_time = 0;
+	unsigned long flags;
+	int ret, i;
+
+	if (!priv->hwts_tx_en)
+		return -EOPNOTSUPP;
+
+	/* Find first TBS enabled Queue, if any */
+	for (i = 0; i < priv->plat->tx_queues_to_use; i++)
+		if (priv->tx_queue[i].tbs & STMMAC_TBS_AVAIL)
+			break;
+
+	if (i >= priv->plat->tx_queues_to_use)
+		return -EOPNOTSUPP;
+
+	qopt.enable = true;
+	qopt.queue = i;
+
+	ret = stmmac_tc_setup_etf(priv, priv, &qopt);
+	if (ret)
+		return ret;
+
+	spin_lock_irqsave(&priv->ptp_lock, flags);
+	stmmac_get_systime(priv, priv->ptpaddr, &curr_time);
+	spin_unlock_irqrestore(&priv->ptp_lock, flags);
+
+	if (!curr_time) {
+		ret = -EOPNOTSUPP;
+		goto fail_disable;
+	}
+
+	start_time = curr_time;
+	curr_time += STMMAC_TBS_LT_OFFSET;
+
+	attr.dst = priv->dev->dev_addr;
+	attr.timestamp = curr_time;
+	attr.timeout = nsecs_to_jiffies(2 * STMMAC_TBS_LT_OFFSET);
+	attr.queue_mapping = i;
+
+	ret = __stmmac_test_loopback(priv, &attr);
+	if (ret)
+		goto fail_disable;
+
+	/* Check if expected time has elapsed */
+	spin_lock_irqsave(&priv->ptp_lock, flags);
+	stmmac_get_systime(priv, priv->ptpaddr, &curr_time);
+	spin_unlock_irqrestore(&priv->ptp_lock, flags);
+
+	if ((curr_time - start_time) < STMMAC_TBS_LT_OFFSET)
+		ret = -EINVAL;
+
+fail_disable:
+	qopt.enable = false;
+	stmmac_tc_setup_etf(priv, priv, &qopt);
+	return ret;
+}
+
 #define STMMAC_LOOPBACK_NONE	0
 #define STMMAC_LOOPBACK_MAC	1
 #define STMMAC_LOOPBACK_PHY	2
@@ -1856,6 +1923,10 @@ static const struct stmmac_test {
 		.name = "Split Header               ",
 		.lb = STMMAC_LOOPBACK_PHY,
 		.fn = stmmac_test_sph,
+	}, {
+		.name = "TBS (ETF Scheduler)        ",
+		.lb = STMMAC_LOOPBACK_PHY,
+		.fn = stmmac_test_tbs,
 	},
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From 5e6d8401ade984296ae9629cb06458c2784e1882 Mon Sep 17 00:00:00 2001
From: Rocky Liao <rjliao@codeaurora.org>
Date: Mon, 13 Jan 2020 12:30:20 +0800
Subject: Bluetooth: hci_qca: Add qca_power_on() API to support both wcn399x
 and Rome power up

This patch adds a unified API qca_power_on() to support both wcn399x and
Rome power on. For wcn399x it calls the qca_wcn3990_init() to init the
regulators, and for Rome it pulls up the bt_en GPIO to power up the btsoc.
It also moves all the power up operation from hdev->open() to
hdev->setup().

Signed-off-by: Rocky Liao <rjliao@codeaurora.org>
Reviewed-by: Matthias Kaehlcke <mka@chromium.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 drivers/bluetooth/hci_qca.c | 54 +++++++++++++++++++++++++++------------------
 1 file changed, 33 insertions(+), 21 deletions(-)

diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c
index 82e4cd4b6663..992622dc1263 100644
--- a/drivers/bluetooth/hci_qca.c
+++ b/drivers/bluetooth/hci_qca.c
@@ -541,7 +541,6 @@ static int qca_open(struct hci_uart *hu)
 {
 	struct qca_serdev *qcadev;
 	struct qca_data *qca;
-	int ret;
 
 	BT_DBG("hu %p qca_open", hu);
 
@@ -582,23 +581,10 @@ static int qca_open(struct hci_uart *hu)
 	hu->priv = qca;
 
 	if (hu->serdev) {
-
 		qcadev = serdev_device_get_drvdata(hu->serdev);
-		if (!qca_is_wcn399x(qcadev->btsoc_type)) {
-			gpiod_set_value_cansleep(qcadev->bt_en, 1);
-			/* Controller needs time to bootup. */
-			msleep(150);
-		} else {
+		if (qca_is_wcn399x(qcadev->btsoc_type)) {
 			hu->init_speed = qcadev->init_speed;
 			hu->oper_speed = qcadev->oper_speed;
-			ret = qca_regulator_enable(qcadev);
-			if (ret) {
-				destroy_workqueue(qca->workqueue);
-				kfree_skb(qca->rx_skb);
-				hu->priv = NULL;
-				kfree(qca);
-				return ret;
-			}
 		}
 	}
 
@@ -1531,6 +1517,31 @@ static int qca_wcn3990_init(struct hci_uart *hu)
 	return 0;
 }
 
+static int qca_power_on(struct hci_dev *hdev)
+{
+	struct hci_uart *hu = hci_get_drvdata(hdev);
+	enum qca_btsoc_type soc_type = qca_soc_type(hu);
+	struct qca_serdev *qcadev;
+	int ret = 0;
+
+	/* Non-serdev device usually is powered by external power
+	 * and don't need additional action in driver for power on
+	 */
+	if (!hu->serdev)
+		return 0;
+
+	if (qca_is_wcn399x(soc_type)) {
+		ret = qca_wcn3990_init(hu);
+	} else {
+		qcadev = serdev_device_get_drvdata(hu->serdev);
+		gpiod_set_value_cansleep(qcadev->bt_en, 1);
+		/* Controller needs time to bootup. */
+		msleep(150);
+	}
+
+	return ret;
+}
+
 static int qca_setup(struct hci_uart *hu)
 {
 	struct hci_dev *hdev = hu->hdev;
@@ -1553,24 +1564,25 @@ static int qca_setup(struct hci_uart *hu)
 	 */
 	set_bit(HCI_QUIRK_SIMULTANEOUS_DISCOVERY, &hdev->quirks);
 
-	if (qca_is_wcn399x(soc_type)) {
-		bt_dev_info(hdev, "setting up wcn3990");
+	bt_dev_info(hdev, "setting up %s",
+		qca_is_wcn399x(soc_type) ? "wcn399x" : "ROME");
 
+	ret = qca_power_on(hdev);
+	if (ret)
+		return ret;
+
+	if (qca_is_wcn399x(soc_type)) {
 		/* Enable NON_PERSISTENT_SETUP QUIRK to ensure to execute
 		 * setup for every hci up.
 		 */
 		set_bit(HCI_QUIRK_NON_PERSISTENT_SETUP, &hdev->quirks);
 		set_bit(HCI_QUIRK_USE_BDADDR_PROPERTY, &hdev->quirks);
 		hu->hdev->shutdown = qca_power_off;
-		ret = qca_wcn3990_init(hu);
-		if (ret)
-			return ret;
 
 		ret = qca_read_soc_version(hdev, &soc_ver, soc_type);
 		if (ret)
 			return ret;
 	} else {
-		bt_dev_info(hdev, "ROME setup");
 		qca_set_speed(hu, QCA_INIT_SPEED);
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From 3a0d3092a4edbbcd62360c44a931c8feaf4b4ff1 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Sun, 12 Jan 2020 23:31:43 -0800
Subject: selftests/bpf: Build runqslower from selftests

Ensure runqslower tool is built as part of selftests to prevent it from bit
rotting.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200113073143.1779940-7-andriin@fb.com
---
 tools/bpf/runqslower/Makefile        | 2 +-
 tools/testing/selftests/bpf/Makefile | 7 ++++++-
 2 files changed, 7 insertions(+), 2 deletions(-)

diff --git a/tools/bpf/runqslower/Makefile b/tools/bpf/runqslower/Makefile
index f1363ae8e473..cff2fbcd29a8 100644
--- a/tools/bpf/runqslower/Makefile
+++ b/tools/bpf/runqslower/Makefile
@@ -10,7 +10,7 @@ CFLAGS := -g -Wall
 
 # Try to detect best kernel BTF source
 KERNEL_REL := $(shell uname -r)
-ifneq ("$(wildcard /sys/kenerl/btf/vmlinux)","")
+ifneq ("$(wildcard /sys/kernel/btf/vmlinux)","")
 VMLINUX_BTF := /sys/kernel/btf/vmlinux
 else ifneq ("$(wildcard /boot/vmlinux-$(KERNEL_REL))","")
 VMLINUX_BTF := /boot/vmlinux-$(KERNEL_REL)
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index bf9f7e415e95..246d09ffb296 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -73,7 +73,7 @@ TEST_PROGS_EXTENDED := with_addr.sh \
 # Compile but not part of 'make run_tests'
 TEST_GEN_PROGS_EXTENDED = test_sock_addr test_skb_cgroup_id_user \
 	flow_dissector_load test_flow_dissector test_tcp_check_syncookie_user \
-	test_lirc_mode2_user xdping test_cpp
+	test_lirc_mode2_user xdping test_cpp runqslower
 
 TEST_CUSTOM_PROGS = urandom_read
 
@@ -124,6 +124,11 @@ $(OUTPUT)/test_stub.o: test_stub.c
 	$(call msg,CC,,$@)
 	$(CC) -c $(CFLAGS) -o $@ $<
 
+.PHONY: $(OUTPUT)/runqslower
+$(OUTPUT)/runqslower: force
+	$(Q)$(MAKE) $(submake_extras) -C $(TOOLSDIR)/bpf/runqslower	      \
+		    OUTPUT=$(CURDIR)/tools/
+
 BPFOBJ := $(OUTPUT)/libbpf.a
 
 $(TEST_GEN_PROGS) $(TEST_GEN_PROGS_EXTENDED): $(OUTPUT)/test_stub.o $(BPFOBJ)
-- 
cgit v1.2.3-59-g8ed1b


From 3b4130418f62b0e7a4685cc2c03bb41c6cb8922d Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <kafai@fb.com>
Date: Mon, 13 Jan 2020 23:26:47 -0800
Subject: bpf: Fix seq_show for BPF_MAP_TYPE_STRUCT_OPS

Instead of using bpf_struct_ops_map_lookup_elem() which is
not implemented, bpf_struct_ops_map_seq_show_elem() should
also use bpf_struct_ops_map_sys_lookup_elem() which does
an inplace update to the value.  The change allocates
a value to pass to bpf_struct_ops_map_sys_lookup_elem().

[root@arch-fb-vm1 bpf]# cat /sys/fs/bpf/dctcp
{{{1}},BPF_STRUCT_OPS_STATE_INUSE,{{00000000df93eebc,00000000df93eebc},0,2, ...

Fixes: 85d33df357b6 ("bpf: Introduce BPF_MAP_TYPE_STRUCT_OPS")
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200114072647.3188298-1-kafai@fb.com
---
 kernel/bpf/bpf_struct_ops.c | 14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)

diff --git a/kernel/bpf/bpf_struct_ops.c b/kernel/bpf/bpf_struct_ops.c
index ddf48f49914b..8ad1c9ea26b2 100644
--- a/kernel/bpf/bpf_struct_ops.c
+++ b/kernel/bpf/bpf_struct_ops.c
@@ -496,14 +496,20 @@ static void bpf_struct_ops_map_seq_show_elem(struct bpf_map *map, void *key,
 					     struct seq_file *m)
 {
 	void *value;
+	int err;
 
-	value = bpf_struct_ops_map_lookup_elem(map, key);
+	value = kmalloc(map->value_size, GFP_USER | __GFP_NOWARN);
 	if (!value)
 		return;
 
-	btf_type_seq_show(btf_vmlinux, map->btf_vmlinux_value_type_id,
-			  value, m);
-	seq_puts(m, "\n");
+	err = bpf_struct_ops_map_sys_lookup_elem(map, key, value);
+	if (!err) {
+		btf_type_seq_show(btf_vmlinux, map->btf_vmlinux_value_type_id,
+				  value, m);
+		seq_puts(m, "\n");
+	}
+
+	kfree(value);
 }
 
 static void bpf_struct_ops_map_free(struct bpf_map *map)
-- 
cgit v1.2.3-59-g8ed1b


From e27f178793de16ca1b421f2c3f4bc3497b2ce723 Mon Sep 17 00:00:00 2001
From: Florian Fainelli <f.fainelli@gmail.com>
Date: Sun, 12 Jan 2020 09:35:38 -0800
Subject: net: phy: Added IRQ print to phylink_bringup_phy()

The information about the PHY attached to the PHYLINK instance is useful
but is missing the IRQ prints that phy_attached_info() adds.
phy_attached_info() is a bit long and it would not be possible to use
phylink_info() anyway.

Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/phy_device.c | 12 ++++++++++--
 drivers/net/phy/phylink.c    |  7 +++++--
 include/linux/phy.h          |  2 ++
 3 files changed, 17 insertions(+), 4 deletions(-)

diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c
index e5dc9f87f495..6a5056e0ae77 100644
--- a/drivers/net/phy/phy_device.c
+++ b/drivers/net/phy/phy_device.c
@@ -1107,9 +1107,8 @@ void phy_attached_info(struct phy_device *phydev)
 EXPORT_SYMBOL(phy_attached_info);
 
 #define ATTACHED_FMT "attached PHY driver [%s] (mii_bus:phy_addr=%s, irq=%s)"
-void phy_attached_print(struct phy_device *phydev, const char *fmt, ...)
+char *phy_attached_info_irq(struct phy_device *phydev)
 {
-	const char *drv_name = phydev->drv ? phydev->drv->name : "unbound";
 	char *irq_str;
 	char irq_num[8];
 
@@ -1126,6 +1125,14 @@ void phy_attached_print(struct phy_device *phydev, const char *fmt, ...)
 		break;
 	}
 
+	return kasprintf(GFP_KERNEL, "%s", irq_str);
+}
+EXPORT_SYMBOL(phy_attached_info_irq);
+
+void phy_attached_print(struct phy_device *phydev, const char *fmt, ...)
+{
+	const char *drv_name = phydev->drv ? phydev->drv->name : "unbound";
+	char *irq_str = phy_attached_info_irq(phydev);
 
 	if (!fmt) {
 		phydev_info(phydev, ATTACHED_FMT "\n",
@@ -1142,6 +1149,7 @@ void phy_attached_print(struct phy_device *phydev, const char *fmt, ...)
 		vprintk(fmt, ap);
 		va_end(ap);
 	}
+	kfree(irq_str);
 }
 EXPORT_SYMBOL(phy_attached_print);
 
diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index 8d786c99f97a..efabbfa4a6d3 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -726,6 +726,7 @@ static int phylink_bringup_phy(struct phylink *pl, struct phy_device *phy,
 {
 	struct phylink_link_state config;
 	__ETHTOOL_DECLARE_LINK_MODE_MASK(supported);
+	char *irq_str;
 	int ret;
 
 	/*
@@ -761,9 +762,11 @@ static int phylink_bringup_phy(struct phylink *pl, struct phy_device *phy,
 	phy->phylink = pl;
 	phy->phy_link_change = phylink_phy_change;
 
+	irq_str = phy_attached_info_irq(phy);
 	phylink_info(pl,
-		     "PHY [%s] driver [%s]\n", dev_name(&phy->mdio.dev),
-		     phy->drv->name);
+		     "PHY [%s] driver [%s] (irq=%s)\n",
+		     dev_name(&phy->mdio.dev), phy->drv->name, irq_str);
+	kfree(irq_str);
 
 	mutex_lock(&phy->lock);
 	mutex_lock(&pl->state_mutex);
diff --git a/include/linux/phy.h b/include/linux/phy.h
index 5932bb8e9c35..3a70b756ac1a 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -1131,6 +1131,8 @@ static inline void phy_unlock_mdio_bus(struct phy_device *phydev)
 
 void phy_attached_print(struct phy_device *phydev, const char *fmt, ...)
 	__printf(2, 3);
+char *phy_attached_info_irq(struct phy_device *phydev)
+	__malloc;
 void phy_attached_info(struct phy_device *phydev);
 
 /* Clause 22 PHY */
-- 
cgit v1.2.3-59-g8ed1b


From 621400366bdba122b37bfa5f5a64ebfd07919a4a Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert+renesas@glider.be>
Date: Mon, 13 Jan 2020 14:05:29 +0100
Subject: mdio_bus: Simplify reset handling and extend to non-DT systems

Convert mdiobus_register_reset() from open-coded DT-only optional reset
handling to reset_control_get_optional_exclusive().  This not only
simplifies the code, but also adds support for lookup-based resets on
non-DT systems.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Philipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/mdio_bus.c | 16 +++++-----------
 1 file changed, 5 insertions(+), 11 deletions(-)

diff --git a/drivers/net/phy/mdio_bus.c b/drivers/net/phy/mdio_bus.c
index 229e480179ff..8d753bb07227 100644
--- a/drivers/net/phy/mdio_bus.c
+++ b/drivers/net/phy/mdio_bus.c
@@ -59,17 +59,11 @@ static int mdiobus_register_gpiod(struct mdio_device *mdiodev)
 
 static int mdiobus_register_reset(struct mdio_device *mdiodev)
 {
-	struct reset_control *reset = NULL;
-
-	if (mdiodev->dev.of_node)
-		reset = of_reset_control_get_exclusive(mdiodev->dev.of_node,
-						       "phy");
-	if (IS_ERR(reset)) {
-		if (PTR_ERR(reset) == -ENOENT || PTR_ERR(reset) == -ENOTSUPP)
-			reset = NULL;
-		else
-			return PTR_ERR(reset);
-	}
+	struct reset_control *reset;
+
+	reset = reset_control_get_optional_exclusive(&mdiodev->dev, "phy");
+	if (IS_ERR(reset))
+		return PTR_ERR(reset);
 
 	mdiodev->reset_ctrl = reset;
 
-- 
cgit v1.2.3-59-g8ed1b


From 4905294162bda43bef65b411cf791005c7e15b28 Mon Sep 17 00:00:00 2001
From: Guillaume Nault <gnault@redhat.com>
Date: Mon, 13 Jan 2020 22:39:20 +0100
Subject: netns: Remove __peernet2id_alloc()

__peernet2id_alloc() was used for both plain lookups and for netns ID
allocations (depending the value of '*alloc'). Let's separate lookups
from allocations instead. That is, integrate the lookup code into
__peernet2id() and make peernet2id_alloc() responsible for allocating
new netns IDs when necessary.

This makes it clear that __peernet2id() doesn't modify the idr and
prepares the code for lockless lookups.

Also, mark the 'net' argument of __peernet2id() as 'const', since we're
modifying this line.

Signed-off-by: Guillaume Nault <gnault@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/net_namespace.c | 55 ++++++++++++++++++++----------------------------
 1 file changed, 23 insertions(+), 32 deletions(-)

diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c
index 39402840025e..05e07d24b45b 100644
--- a/net/core/net_namespace.c
+++ b/net/core/net_namespace.c
@@ -211,16 +211,10 @@ static int net_eq_idr(int id, void *net, void *peer)
 	return 0;
 }
 
-/* Should be called with nsid_lock held. If a new id is assigned, the bool alloc
- * is set to true, thus the caller knows that the new id must be notified via
- * rtnl.
- */
-static int __peernet2id_alloc(struct net *net, struct net *peer, bool *alloc)
+/* Should be called with nsid_lock held. */
+static int __peernet2id(const struct net *net, struct net *peer)
 {
 	int id = idr_for_each(&net->netns_ids, net_eq_idr, peer);
-	bool alloc_it = *alloc;
-
-	*alloc = false;
 
 	/* Magic value for id 0. */
 	if (id == NET_ID_ZERO)
@@ -228,23 +222,9 @@ static int __peernet2id_alloc(struct net *net, struct net *peer, bool *alloc)
 	if (id > 0)
 		return id;
 
-	if (alloc_it) {
-		id = alloc_netid(net, peer, -1);
-		*alloc = true;
-		return id >= 0 ? id : NETNSA_NSID_NOT_ASSIGNED;
-	}
-
 	return NETNSA_NSID_NOT_ASSIGNED;
 }
 
-/* should be called with nsid_lock held */
-static int __peernet2id(struct net *net, struct net *peer)
-{
-	bool no = false;
-
-	return __peernet2id_alloc(net, peer, &no);
-}
-
 static void rtnl_net_notifyid(struct net *net, int cmd, int id, u32 portid,
 			      struct nlmsghdr *nlh, gfp_t gfp);
 /* This function returns the id of a peer netns. If no id is assigned, one will
@@ -252,26 +232,37 @@ static void rtnl_net_notifyid(struct net *net, int cmd, int id, u32 portid,
  */
 int peernet2id_alloc(struct net *net, struct net *peer, gfp_t gfp)
 {
-	bool alloc = false, alive = false;
 	int id;
 
 	if (refcount_read(&net->count) == 0)
 		return NETNSA_NSID_NOT_ASSIGNED;
+
 	spin_lock_bh(&net->nsid_lock);
-	/*
-	 * When peer is obtained from RCU lists, we may race with
+	id = __peernet2id(net, peer);
+	if (id >= 0) {
+		spin_unlock_bh(&net->nsid_lock);
+		return id;
+	}
+
+	/* When peer is obtained from RCU lists, we may race with
 	 * its cleanup. Check whether it's alive, and this guarantees
 	 * we never hash a peer back to net->netns_ids, after it has
 	 * just been idr_remove()'d from there in cleanup_net().
 	 */
-	if (maybe_get_net(peer))
-		alive = alloc = true;
-	id = __peernet2id_alloc(net, peer, &alloc);
+	if (!maybe_get_net(peer)) {
+		spin_unlock_bh(&net->nsid_lock);
+		return NETNSA_NSID_NOT_ASSIGNED;
+	}
+
+	id = alloc_netid(net, peer, -1);
 	spin_unlock_bh(&net->nsid_lock);
-	if (alloc && id >= 0)
-		rtnl_net_notifyid(net, RTM_NEWNSID, id, 0, NULL, gfp);
-	if (alive)
-		put_net(peer);
+
+	put_net(peer);
+	if (id < 0)
+		return NETNSA_NSID_NOT_ASSIGNED;
+
+	rtnl_net_notifyid(net, RTM_NEWNSID, id, 0, NULL, gfp);
+
 	return id;
 }
 EXPORT_SYMBOL_GPL(peernet2id_alloc);
-- 
cgit v1.2.3-59-g8ed1b


From 2dce224f469f060b9998a5a869151ef83c08ce77 Mon Sep 17 00:00:00 2001
From: Guillaume Nault <gnault@redhat.com>
Date: Mon, 13 Jan 2020 22:39:22 +0100
Subject: netns: protect netns ID lookups with RCU

__peernet2id() can be protected by RCU as it only calls idr_for_each(),
which is RCU-safe, and never modifies the nsid table.

rtnl_net_dumpid() can also do lockless lookups. It does two nested
idr_for_each() calls on nsid tables (one direct call and one indirect
call because of rtnl_net_dumpid_one() calling __peernet2id()). The
netnsid tables are never updated. Therefore it is safe to not take the
nsid_lock and run within an RCU-critical section instead.

Signed-off-by: Guillaume Nault <gnault@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/net_namespace.c | 22 +++++++---------------
 1 file changed, 7 insertions(+), 15 deletions(-)

diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c
index 05e07d24b45b..e7a5ff4966c9 100644
--- a/net/core/net_namespace.c
+++ b/net/core/net_namespace.c
@@ -211,7 +211,7 @@ static int net_eq_idr(int id, void *net, void *peer)
 	return 0;
 }
 
-/* Should be called with nsid_lock held. */
+/* Must be called from RCU-critical section or with nsid_lock held */
 static int __peernet2id(const struct net *net, struct net *peer)
 {
 	int id = idr_for_each(&net->netns_ids, net_eq_idr, peer);
@@ -272,9 +272,10 @@ int peernet2id(struct net *net, struct net *peer)
 {
 	int id;
 
-	spin_lock_bh(&net->nsid_lock);
+	rcu_read_lock();
 	id = __peernet2id(net, peer);
-	spin_unlock_bh(&net->nsid_lock);
+	rcu_read_unlock();
+
 	return id;
 }
 EXPORT_SYMBOL(peernet2id);
@@ -941,6 +942,7 @@ struct rtnl_net_dump_cb {
 	int s_idx;
 };
 
+/* Runs in RCU-critical section. */
 static int rtnl_net_dumpid_one(int id, void *peer, void *data)
 {
 	struct rtnl_net_dump_cb *net_cb = (struct rtnl_net_dump_cb *)data;
@@ -1025,19 +1027,9 @@ static int rtnl_net_dumpid(struct sk_buff *skb, struct netlink_callback *cb)
 			goto end;
 	}
 
-	spin_lock_bh(&net_cb.tgt_net->nsid_lock);
-	if (net_cb.fillargs.add_ref &&
-	    !net_eq(net_cb.ref_net, net_cb.tgt_net) &&
-	    !spin_trylock_bh(&net_cb.ref_net->nsid_lock)) {
-		spin_unlock_bh(&net_cb.tgt_net->nsid_lock);
-		err = -EAGAIN;
-		goto end;
-	}
+	rcu_read_lock();
 	idr_for_each(&net_cb.tgt_net->netns_ids, rtnl_net_dumpid_one, &net_cb);
-	if (net_cb.fillargs.add_ref &&
-	    !net_eq(net_cb.ref_net, net_cb.tgt_net))
-		spin_unlock_bh(&net_cb.ref_net->nsid_lock);
-	spin_unlock_bh(&net_cb.tgt_net->nsid_lock);
+	rcu_read_unlock();
 
 	cb->args[0] = net_cb.idx;
 end:
-- 
cgit v1.2.3-59-g8ed1b


From 8d7e5dee972f1cde2ba96c621f1541fa36e7d4f4 Mon Sep 17 00:00:00 2001
From: Guillaume Nault <gnault@redhat.com>
Date: Mon, 13 Jan 2020 22:39:23 +0100
Subject: netns: don't disable BHs when locking "nsid_lock"

When peernet2id() had to lock "nsid_lock" before iterating through the
nsid table, we had to disable BHs, because VXLAN can call peernet2id()
from the xmit path:
  vxlan_xmit() -> vxlan_fdb_miss() -> vxlan_fdb_notify()
    -> __vxlan_fdb_notify() -> vxlan_fdb_info() -> peernet2id().

Now that peernet2id() uses RCU protection, "nsid_lock" isn't used in BH
context anymore. Therefore, we can safely use plain
spin_lock()/spin_unlock() and let BHs run when holding "nsid_lock".

Signed-off-by: Guillaume Nault <gnault@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/net_namespace.c | 22 +++++++++++-----------
 1 file changed, 11 insertions(+), 11 deletions(-)

diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c
index e7a5ff4966c9..6412c1fbfcb5 100644
--- a/net/core/net_namespace.c
+++ b/net/core/net_namespace.c
@@ -237,10 +237,10 @@ int peernet2id_alloc(struct net *net, struct net *peer, gfp_t gfp)
 	if (refcount_read(&net->count) == 0)
 		return NETNSA_NSID_NOT_ASSIGNED;
 
-	spin_lock_bh(&net->nsid_lock);
+	spin_lock(&net->nsid_lock);
 	id = __peernet2id(net, peer);
 	if (id >= 0) {
-		spin_unlock_bh(&net->nsid_lock);
+		spin_unlock(&net->nsid_lock);
 		return id;
 	}
 
@@ -250,12 +250,12 @@ int peernet2id_alloc(struct net *net, struct net *peer, gfp_t gfp)
 	 * just been idr_remove()'d from there in cleanup_net().
 	 */
 	if (!maybe_get_net(peer)) {
-		spin_unlock_bh(&net->nsid_lock);
+		spin_unlock(&net->nsid_lock);
 		return NETNSA_NSID_NOT_ASSIGNED;
 	}
 
 	id = alloc_netid(net, peer, -1);
-	spin_unlock_bh(&net->nsid_lock);
+	spin_unlock(&net->nsid_lock);
 
 	put_net(peer);
 	if (id < 0)
@@ -520,20 +520,20 @@ static void unhash_nsid(struct net *net, struct net *last)
 	for_each_net(tmp) {
 		int id;
 
-		spin_lock_bh(&tmp->nsid_lock);
+		spin_lock(&tmp->nsid_lock);
 		id = __peernet2id(tmp, net);
 		if (id >= 0)
 			idr_remove(&tmp->netns_ids, id);
-		spin_unlock_bh(&tmp->nsid_lock);
+		spin_unlock(&tmp->nsid_lock);
 		if (id >= 0)
 			rtnl_net_notifyid(tmp, RTM_DELNSID, id, 0, NULL,
 					  GFP_KERNEL);
 		if (tmp == last)
 			break;
 	}
-	spin_lock_bh(&net->nsid_lock);
+	spin_lock(&net->nsid_lock);
 	idr_destroy(&net->netns_ids);
-	spin_unlock_bh(&net->nsid_lock);
+	spin_unlock(&net->nsid_lock);
 }
 
 static LLIST_HEAD(cleanup_list);
@@ -746,9 +746,9 @@ static int rtnl_net_newid(struct sk_buff *skb, struct nlmsghdr *nlh,
 		return PTR_ERR(peer);
 	}
 
-	spin_lock_bh(&net->nsid_lock);
+	spin_lock(&net->nsid_lock);
 	if (__peernet2id(net, peer) >= 0) {
-		spin_unlock_bh(&net->nsid_lock);
+		spin_unlock(&net->nsid_lock);
 		err = -EEXIST;
 		NL_SET_BAD_ATTR(extack, nla);
 		NL_SET_ERR_MSG(extack,
@@ -757,7 +757,7 @@ static int rtnl_net_newid(struct sk_buff *skb, struct nlmsghdr *nlh,
 	}
 
 	err = alloc_netid(net, peer, nsid);
-	spin_unlock_bh(&net->nsid_lock);
+	spin_unlock(&net->nsid_lock);
 	if (err >= 0) {
 		rtnl_net_notifyid(net, RTM_NEWNSID, err, NETLINK_CB(skb).portid,
 				  nlh, GFP_KERNEL);
-- 
cgit v1.2.3-59-g8ed1b


From c0e4eadfb8daf2e9557c7450f9b237c08b404419 Mon Sep 17 00:00:00 2001
From: Antoine Tenart <antoine.tenart@bootlin.com>
Date: Mon, 13 Jan 2020 23:31:39 +0100
Subject: net: macsec: move some definitions in a dedicated header

This patch moves some structure, type and identifier definitions into a
MACsec specific header. This patch does not modify how the MACsec code
is running and only move things around. This is a preparation for the
future MACsec hardware offloading support, which will re-use those
definitions outside macsec.c.

Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/macsec.c | 164 +----------------------------------------------
 include/net/macsec.h | 177 +++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 178 insertions(+), 163 deletions(-)
 create mode 100644 include/net/macsec.h

diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c
index afd8b2a08245..a336eee018f0 100644
--- a/drivers/net/macsec.c
+++ b/drivers/net/macsec.c
@@ -16,11 +16,10 @@
 #include <net/genetlink.h>
 #include <net/sock.h>
 #include <net/gro_cells.h>
+#include <net/macsec.h>
 
 #include <uapi/linux/if_macsec.h>
 
-typedef u64 __bitwise sci_t;
-
 #define MACSEC_SCI_LEN 8
 
 /* SecTAG length = macsec_eth_header without the optional SCI */
@@ -58,8 +57,6 @@ struct macsec_eth_header {
 #define GCM_AES_IV_LEN 12
 #define DEFAULT_ICV_LEN 16
 
-#define MACSEC_NUM_AN 4 /* 2 bits for the association number */
-
 #define for_each_rxsc(secy, sc)				\
 	for (sc = rcu_dereference_bh(secy->rx_sc);	\
 	     sc;					\
@@ -77,49 +74,6 @@ struct gcm_iv {
 	__be32 pn;
 };
 
-/**
- * struct macsec_key - SA key
- * @id: user-provided key identifier
- * @tfm: crypto struct, key storage
- */
-struct macsec_key {
-	u8 id[MACSEC_KEYID_LEN];
-	struct crypto_aead *tfm;
-};
-
-struct macsec_rx_sc_stats {
-	__u64 InOctetsValidated;
-	__u64 InOctetsDecrypted;
-	__u64 InPktsUnchecked;
-	__u64 InPktsDelayed;
-	__u64 InPktsOK;
-	__u64 InPktsInvalid;
-	__u64 InPktsLate;
-	__u64 InPktsNotValid;
-	__u64 InPktsNotUsingSA;
-	__u64 InPktsUnusedSA;
-};
-
-struct macsec_rx_sa_stats {
-	__u32 InPktsOK;
-	__u32 InPktsInvalid;
-	__u32 InPktsNotValid;
-	__u32 InPktsNotUsingSA;
-	__u32 InPktsUnusedSA;
-};
-
-struct macsec_tx_sa_stats {
-	__u32 OutPktsProtected;
-	__u32 OutPktsEncrypted;
-};
-
-struct macsec_tx_sc_stats {
-	__u64 OutPktsProtected;
-	__u64 OutPktsEncrypted;
-	__u64 OutOctetsProtected;
-	__u64 OutOctetsEncrypted;
-};
-
 struct macsec_dev_stats {
 	__u64 OutPktsUntagged;
 	__u64 InPktsUntagged;
@@ -131,124 +85,8 @@ struct macsec_dev_stats {
 	__u64 InPktsOverrun;
 };
 
-/**
- * struct macsec_rx_sa - receive secure association
- * @active:
- * @next_pn: packet number expected for the next packet
- * @lock: protects next_pn manipulations
- * @key: key structure
- * @stats: per-SA stats
- */
-struct macsec_rx_sa {
-	struct macsec_key key;
-	spinlock_t lock;
-	u32 next_pn;
-	refcount_t refcnt;
-	bool active;
-	struct macsec_rx_sa_stats __percpu *stats;
-	struct macsec_rx_sc *sc;
-	struct rcu_head rcu;
-};
-
-struct pcpu_rx_sc_stats {
-	struct macsec_rx_sc_stats stats;
-	struct u64_stats_sync syncp;
-};
-
-/**
- * struct macsec_rx_sc - receive secure channel
- * @sci: secure channel identifier for this SC
- * @active: channel is active
- * @sa: array of secure associations
- * @stats: per-SC stats
- */
-struct macsec_rx_sc {
-	struct macsec_rx_sc __rcu *next;
-	sci_t sci;
-	bool active;
-	struct macsec_rx_sa __rcu *sa[MACSEC_NUM_AN];
-	struct pcpu_rx_sc_stats __percpu *stats;
-	refcount_t refcnt;
-	struct rcu_head rcu_head;
-};
-
-/**
- * struct macsec_tx_sa - transmit secure association
- * @active:
- * @next_pn: packet number to use for the next packet
- * @lock: protects next_pn manipulations
- * @key: key structure
- * @stats: per-SA stats
- */
-struct macsec_tx_sa {
-	struct macsec_key key;
-	spinlock_t lock;
-	u32 next_pn;
-	refcount_t refcnt;
-	bool active;
-	struct macsec_tx_sa_stats __percpu *stats;
-	struct rcu_head rcu;
-};
-
-struct pcpu_tx_sc_stats {
-	struct macsec_tx_sc_stats stats;
-	struct u64_stats_sync syncp;
-};
-
-/**
- * struct macsec_tx_sc - transmit secure channel
- * @active:
- * @encoding_sa: association number of the SA currently in use
- * @encrypt: encrypt packets on transmit, or authenticate only
- * @send_sci: always include the SCI in the SecTAG
- * @end_station:
- * @scb: single copy broadcast flag
- * @sa: array of secure associations
- * @stats: stats for this TXSC
- */
-struct macsec_tx_sc {
-	bool active;
-	u8 encoding_sa;
-	bool encrypt;
-	bool send_sci;
-	bool end_station;
-	bool scb;
-	struct macsec_tx_sa __rcu *sa[MACSEC_NUM_AN];
-	struct pcpu_tx_sc_stats __percpu *stats;
-};
-
 #define MACSEC_VALIDATE_DEFAULT MACSEC_VALIDATE_STRICT
 
-/**
- * struct macsec_secy - MACsec Security Entity
- * @netdev: netdevice for this SecY
- * @n_rx_sc: number of receive secure channels configured on this SecY
- * @sci: secure channel identifier used for tx
- * @key_len: length of keys used by the cipher suite
- * @icv_len: length of ICV used by the cipher suite
- * @validate_frames: validation mode
- * @operational: MAC_Operational flag
- * @protect_frames: enable protection for this SecY
- * @replay_protect: enable packet number checks on receive
- * @replay_window: size of the replay window
- * @tx_sc: transmit secure channel
- * @rx_sc: linked list of receive secure channels
- */
-struct macsec_secy {
-	struct net_device *netdev;
-	unsigned int n_rx_sc;
-	sci_t sci;
-	u16 key_len;
-	u16 icv_len;
-	enum macsec_validation_type validate_frames;
-	bool operational;
-	bool protect_frames;
-	bool replay_protect;
-	u32 replay_window;
-	struct macsec_tx_sc tx_sc;
-	struct macsec_rx_sc __rcu *rx_sc;
-};
-
 struct pcpu_secy_stats {
 	struct macsec_dev_stats stats;
 	struct u64_stats_sync syncp;
diff --git a/include/net/macsec.h b/include/net/macsec.h
new file mode 100644
index 000000000000..e7b41c1043f6
--- /dev/null
+++ b/include/net/macsec.h
@@ -0,0 +1,177 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/*
+ * MACsec netdev header, used for h/w accelerated implementations.
+ *
+ * Copyright (c) 2015 Sabrina Dubroca <sd@queasysnail.net>
+ */
+#ifndef _NET_MACSEC_H_
+#define _NET_MACSEC_H_
+
+#include <linux/u64_stats_sync.h>
+#include <uapi/linux/if_link.h>
+#include <uapi/linux/if_macsec.h>
+
+typedef u64 __bitwise sci_t;
+
+#define MACSEC_NUM_AN 4 /* 2 bits for the association number */
+
+/**
+ * struct macsec_key - SA key
+ * @id: user-provided key identifier
+ * @tfm: crypto struct, key storage
+ */
+struct macsec_key {
+	u8 id[MACSEC_KEYID_LEN];
+	struct crypto_aead *tfm;
+};
+
+struct macsec_rx_sc_stats {
+	__u64 InOctetsValidated;
+	__u64 InOctetsDecrypted;
+	__u64 InPktsUnchecked;
+	__u64 InPktsDelayed;
+	__u64 InPktsOK;
+	__u64 InPktsInvalid;
+	__u64 InPktsLate;
+	__u64 InPktsNotValid;
+	__u64 InPktsNotUsingSA;
+	__u64 InPktsUnusedSA;
+};
+
+struct macsec_rx_sa_stats {
+	__u32 InPktsOK;
+	__u32 InPktsInvalid;
+	__u32 InPktsNotValid;
+	__u32 InPktsNotUsingSA;
+	__u32 InPktsUnusedSA;
+};
+
+struct macsec_tx_sa_stats {
+	__u32 OutPktsProtected;
+	__u32 OutPktsEncrypted;
+};
+
+struct macsec_tx_sc_stats {
+	__u64 OutPktsProtected;
+	__u64 OutPktsEncrypted;
+	__u64 OutOctetsProtected;
+	__u64 OutOctetsEncrypted;
+};
+
+/**
+ * struct macsec_rx_sa - receive secure association
+ * @active:
+ * @next_pn: packet number expected for the next packet
+ * @lock: protects next_pn manipulations
+ * @key: key structure
+ * @stats: per-SA stats
+ */
+struct macsec_rx_sa {
+	struct macsec_key key;
+	spinlock_t lock;
+	u32 next_pn;
+	refcount_t refcnt;
+	bool active;
+	struct macsec_rx_sa_stats __percpu *stats;
+	struct macsec_rx_sc *sc;
+	struct rcu_head rcu;
+};
+
+struct pcpu_rx_sc_stats {
+	struct macsec_rx_sc_stats stats;
+	struct u64_stats_sync syncp;
+};
+
+struct pcpu_tx_sc_stats {
+	struct macsec_tx_sc_stats stats;
+	struct u64_stats_sync syncp;
+};
+
+/**
+ * struct macsec_rx_sc - receive secure channel
+ * @sci: secure channel identifier for this SC
+ * @active: channel is active
+ * @sa: array of secure associations
+ * @stats: per-SC stats
+ */
+struct macsec_rx_sc {
+	struct macsec_rx_sc __rcu *next;
+	sci_t sci;
+	bool active;
+	struct macsec_rx_sa __rcu *sa[MACSEC_NUM_AN];
+	struct pcpu_rx_sc_stats __percpu *stats;
+	refcount_t refcnt;
+	struct rcu_head rcu_head;
+};
+
+/**
+ * struct macsec_tx_sa - transmit secure association
+ * @active:
+ * @next_pn: packet number to use for the next packet
+ * @lock: protects next_pn manipulations
+ * @key: key structure
+ * @stats: per-SA stats
+ */
+struct macsec_tx_sa {
+	struct macsec_key key;
+	spinlock_t lock;
+	u32 next_pn;
+	refcount_t refcnt;
+	bool active;
+	struct macsec_tx_sa_stats __percpu *stats;
+	struct rcu_head rcu;
+};
+
+/**
+ * struct macsec_tx_sc - transmit secure channel
+ * @active:
+ * @encoding_sa: association number of the SA currently in use
+ * @encrypt: encrypt packets on transmit, or authenticate only
+ * @send_sci: always include the SCI in the SecTAG
+ * @end_station:
+ * @scb: single copy broadcast flag
+ * @sa: array of secure associations
+ * @stats: stats for this TXSC
+ */
+struct macsec_tx_sc {
+	bool active;
+	u8 encoding_sa;
+	bool encrypt;
+	bool send_sci;
+	bool end_station;
+	bool scb;
+	struct macsec_tx_sa __rcu *sa[MACSEC_NUM_AN];
+	struct pcpu_tx_sc_stats __percpu *stats;
+};
+
+/**
+ * struct macsec_secy - MACsec Security Entity
+ * @netdev: netdevice for this SecY
+ * @n_rx_sc: number of receive secure channels configured on this SecY
+ * @sci: secure channel identifier used for tx
+ * @key_len: length of keys used by the cipher suite
+ * @icv_len: length of ICV used by the cipher suite
+ * @validate_frames: validation mode
+ * @operational: MAC_Operational flag
+ * @protect_frames: enable protection for this SecY
+ * @replay_protect: enable packet number checks on receive
+ * @replay_window: size of the replay window
+ * @tx_sc: transmit secure channel
+ * @rx_sc: linked list of receive secure channels
+ */
+struct macsec_secy {
+	struct net_device *netdev;
+	unsigned int n_rx_sc;
+	sci_t sci;
+	u16 key_len;
+	u16 icv_len;
+	enum macsec_validation_type validate_frames;
+	bool operational;
+	bool protect_frames;
+	bool replay_protect;
+	u32 replay_window;
+	struct macsec_tx_sc tx_sc;
+	struct macsec_rx_sc __rcu *rx_sc;
+};
+
+#endif /* _NET_MACSEC_H_ */
-- 
cgit v1.2.3-59-g8ed1b


From 76564261a7db80c5f5c624e0122a28787f266bdf Mon Sep 17 00:00:00 2001
From: Antoine Tenart <antoine.tenart@bootlin.com>
Date: Mon, 13 Jan 2020 23:31:40 +0100
Subject: net: macsec: introduce the macsec_context structure

This patch introduces the macsec_context structure. It will be used
in the kernel to exchange information between the common MACsec
implementation (macsec.c) and the MACsec hardware offloading
implementations. This structure contains pointers to MACsec specific
structures which contain the actual MACsec configuration, and to the
underlying device (phydev for now).

Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/phy.h                |  2 ++
 include/net/macsec.h               | 21 +++++++++++++++++++++
 include/uapi/linux/if_link.h       |  7 +++++++
 tools/include/uapi/linux/if_link.h |  7 +++++++
 4 files changed, 37 insertions(+)

diff --git a/include/linux/phy.h b/include/linux/phy.h
index 3a70b756ac1a..be079a7bb40a 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -332,6 +332,8 @@ struct phy_c45_device_ids {
 	u32 device_ids[8];
 };
 
+struct macsec_context;
+
 /* phy_device: An instance of a PHY
  *
  * drv: Pointer to the driver for this PHY instance
diff --git a/include/net/macsec.h b/include/net/macsec.h
index e7b41c1043f6..0b98803f92ec 100644
--- a/include/net/macsec.h
+++ b/include/net/macsec.h
@@ -174,4 +174,25 @@ struct macsec_secy {
 	struct macsec_rx_sc __rcu *rx_sc;
 };
 
+/**
+ * struct macsec_context - MACsec context for hardware offloading
+ */
+struct macsec_context {
+	struct phy_device *phydev;
+	enum macsec_offload offload;
+
+	struct macsec_secy *secy;
+	struct macsec_rx_sc *rx_sc;
+	struct {
+		unsigned char assoc_num;
+		u8 key[MACSEC_KEYID_LEN];
+		union {
+			struct macsec_rx_sa *rx_sa;
+			struct macsec_tx_sa *tx_sa;
+		};
+	} sa;
+
+	u8 prepare:1;
+};
+
 #endif /* _NET_MACSEC_H_ */
diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h
index 1d69f637c5d6..024af2d1d0af 100644
--- a/include/uapi/linux/if_link.h
+++ b/include/uapi/linux/if_link.h
@@ -486,6 +486,13 @@ enum macsec_validation_type {
 	MACSEC_VALIDATE_MAX = __MACSEC_VALIDATE_END - 1,
 };
 
+enum macsec_offload {
+	MACSEC_OFFLOAD_OFF = 0,
+	MACSEC_OFFLOAD_PHY = 1,
+	__MACSEC_OFFLOAD_END,
+	MACSEC_OFFLOAD_MAX = __MACSEC_OFFLOAD_END - 1,
+};
+
 /* IPVLAN section */
 enum {
 	IFLA_IPVLAN_UNSPEC,
diff --git a/tools/include/uapi/linux/if_link.h b/tools/include/uapi/linux/if_link.h
index 8aec8769d944..42efdb84d189 100644
--- a/tools/include/uapi/linux/if_link.h
+++ b/tools/include/uapi/linux/if_link.h
@@ -485,6 +485,13 @@ enum macsec_validation_type {
 	MACSEC_VALIDATE_MAX = __MACSEC_VALIDATE_END - 1,
 };
 
+enum macsec_offload {
+	MACSEC_OFFLOAD_OFF = 0,
+	MACSEC_OFFLOAD_PHY = 1,
+	__MACSEC_OFFLOAD_END,
+	MACSEC_OFFLOAD_MAX = __MACSEC_OFFLOAD_END - 1,
+};
+
 /* IPVLAN section */
 enum {
 	IFLA_IPVLAN_UNSPEC,
-- 
cgit v1.2.3-59-g8ed1b


From 0830e20b62ad156f7df5ff5b9c4cea280ebe8fef Mon Sep 17 00:00:00 2001
From: Antoine Tenart <antoine.tenart@bootlin.com>
Date: Mon, 13 Jan 2020 23:31:41 +0100
Subject: net: macsec: introduce MACsec ops

This patch introduces MACsec ops for drivers to support offloading
MACsec operations.

Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/macsec.h | 24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)

diff --git a/include/net/macsec.h b/include/net/macsec.h
index 0b98803f92ec..16e7e5061178 100644
--- a/include/net/macsec.h
+++ b/include/net/macsec.h
@@ -195,4 +195,28 @@ struct macsec_context {
 	u8 prepare:1;
 };
 
+/**
+ * struct macsec_ops - MACsec offloading operations
+ */
+struct macsec_ops {
+	/* Device wide */
+	int (*mdo_dev_open)(struct macsec_context *ctx);
+	int (*mdo_dev_stop)(struct macsec_context *ctx);
+	/* SecY */
+	int (*mdo_add_secy)(struct macsec_context *ctx);
+	int (*mdo_upd_secy)(struct macsec_context *ctx);
+	int (*mdo_del_secy)(struct macsec_context *ctx);
+	/* Security channels */
+	int (*mdo_add_rxsc)(struct macsec_context *ctx);
+	int (*mdo_upd_rxsc)(struct macsec_context *ctx);
+	int (*mdo_del_rxsc)(struct macsec_context *ctx);
+	/* Security associations */
+	int (*mdo_add_rxsa)(struct macsec_context *ctx);
+	int (*mdo_upd_rxsa)(struct macsec_context *ctx);
+	int (*mdo_del_rxsa)(struct macsec_context *ctx);
+	int (*mdo_add_txsa)(struct macsec_context *ctx);
+	int (*mdo_upd_txsa)(struct macsec_context *ctx);
+	int (*mdo_del_txsa)(struct macsec_context *ctx);
+};
+
 #endif /* _NET_MACSEC_H_ */
-- 
cgit v1.2.3-59-g8ed1b


From 2e18135845b359f26c37df38ba56565496517c10 Mon Sep 17 00:00:00 2001
From: Antoine Tenart <antoine.tenart@bootlin.com>
Date: Mon, 13 Jan 2020 23:31:42 +0100
Subject: net: phy: add MACsec ops in phy_device

This patch adds a reference to MACsec ops in the phy_device, to allow
PHYs to support offloading MACsec operations. The phydev lock will be
held while calling those helpers.

Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/phy.h | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/include/linux/phy.h b/include/linux/phy.h
index be079a7bb40a..2929d0bc307f 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -333,6 +333,7 @@ struct phy_c45_device_ids {
 };
 
 struct macsec_context;
+struct macsec_ops;
 
 /* phy_device: An instance of a PHY
  *
@@ -356,6 +357,7 @@ struct macsec_context;
  * attached_dev: The attached enet driver's device instance ptr
  * adjust_link: Callback for the enet controller to respond to
  * changes in the link state.
+ * macsec_ops: MACsec offloading ops.
  *
  * speed, duplex, pause, supported, advertising, lp_advertising,
  * and autoneg are used like in mii_if_info
@@ -455,6 +457,11 @@ struct phy_device {
 
 	void (*phy_link_change)(struct phy_device *, bool up, bool do_carrier);
 	void (*adjust_link)(struct net_device *dev);
+
+#if IS_ENABLED(CONFIG_MACSEC)
+	/* MACsec management functions */
+	const struct macsec_ops *macsec_ops;
+#endif
 };
 #define to_phy_device(d) container_of(to_mdio_device(d), \
 				      struct phy_device, mdio)
-- 
cgit v1.2.3-59-g8ed1b


From 3cf3227a21d1fb020fe26128e60321bd2151e922 Mon Sep 17 00:00:00 2001
From: Antoine Tenart <antoine.tenart@bootlin.com>
Date: Mon, 13 Jan 2020 23:31:43 +0100
Subject: net: macsec: hardware offloading infrastructure

This patch introduces the MACsec hardware offloading infrastructure.

The main idea here is to re-use the logic and data structures of the
software MACsec implementation. This allows not to duplicate definitions
and structure storing the same kind of information. It also allows to
use a unified genlink interface for both MACsec implementations (so that
the same userspace tool, `ip macsec`, is used with the same arguments).
The MACsec offloading support cannot be disabled if an interface
supports it at the moment.

The MACsec configuration is passed to device drivers supporting it
through macsec_ops which are called from the MACsec genl helpers. Those
functions call the macsec ops of PHY and Ethernet drivers in two steps:
a preparation one, and a commit one. The first step is allowed to fail
and should be used to check if a provided configuration is compatible
with the features provided by a MACsec engine, while the second step is
not allowed to fail and should only be used to enable a given MACsec
configuration. Two extra calls are made: when a virtual MACsec interface
is created and when it is deleted, so that the hardware driver can stay
in sync.

The Rx and TX handlers are modified to take in account the special case
were the MACsec transformation happens in the hardware, whether in a PHY
or in a MAC, as the packets seen by the networking stack on both the
physical and MACsec virtual interface are exactly the same. This leads
to some limitations: the hardware and software implementations can't be
used on the same physical interface, as the policies would be impossible
to fulfill (such as strict validation of the frames). Also only a single
virtual MACsec interface can be offloaded to a physical port supporting
hardware offloading as it would be impossible to guess onto which
interface a given packet should go (for ingress traffic).

Another limitation as of now is that the counters and statistics are not
reported back from the hardware to the software MACsec implementation.
This isn't an issue when using offloaded MACsec transformations, but it
should be added in the future so that the MACsec state can be reported
to the user (which would also improve the debug).

Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/macsec.c | 453 +++++++++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 441 insertions(+), 12 deletions(-)

diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c
index a336eee018f0..36b0416381bf 100644
--- a/drivers/net/macsec.c
+++ b/drivers/net/macsec.c
@@ -11,12 +11,14 @@
 #include <linux/module.h>
 #include <crypto/aead.h>
 #include <linux/etherdevice.h>
+#include <linux/netdevice.h>
 #include <linux/rtnetlink.h>
 #include <linux/refcount.h>
 #include <net/genetlink.h>
 #include <net/sock.h>
 #include <net/gro_cells.h>
 #include <net/macsec.h>
+#include <linux/phy.h>
 
 #include <uapi/linux/if_macsec.h>
 
@@ -98,6 +100,7 @@ struct pcpu_secy_stats {
  * @real_dev: pointer to underlying netdevice
  * @stats: MACsec device stats
  * @secys: linked list of SecY's on the underlying device
+ * @offload: status of offloading on the MACsec device
  */
 struct macsec_dev {
 	struct macsec_secy secy;
@@ -105,6 +108,7 @@ struct macsec_dev {
 	struct pcpu_secy_stats __percpu *stats;
 	struct list_head secys;
 	struct gro_cells gro_cells;
+	enum macsec_offload offload;
 };
 
 /**
@@ -318,6 +322,56 @@ static void macsec_set_shortlen(struct macsec_eth_header *h, size_t data_len)
 		h->short_length = data_len;
 }
 
+/* Checks if a MACsec interface is being offloaded to an hardware engine */
+static bool macsec_is_offloaded(struct macsec_dev *macsec)
+{
+	if (macsec->offload == MACSEC_OFFLOAD_PHY)
+		return true;
+
+	return false;
+}
+
+/* Checks if underlying layers implement MACsec offloading functions. */
+static bool macsec_check_offload(enum macsec_offload offload,
+				 struct macsec_dev *macsec)
+{
+	if (!macsec || !macsec->real_dev)
+		return false;
+
+	if (offload == MACSEC_OFFLOAD_PHY)
+		return macsec->real_dev->phydev &&
+		       macsec->real_dev->phydev->macsec_ops;
+
+	return false;
+}
+
+static const struct macsec_ops *__macsec_get_ops(enum macsec_offload offload,
+						 struct macsec_dev *macsec,
+						 struct macsec_context *ctx)
+{
+	if (ctx) {
+		memset(ctx, 0, sizeof(*ctx));
+		ctx->offload = offload;
+
+		if (offload == MACSEC_OFFLOAD_PHY)
+			ctx->phydev = macsec->real_dev->phydev;
+	}
+
+	return macsec->real_dev->phydev->macsec_ops;
+}
+
+/* Returns a pointer to the MACsec ops struct if any and updates the MACsec
+ * context device reference if provided.
+ */
+static const struct macsec_ops *macsec_get_ops(struct macsec_dev *macsec,
+					       struct macsec_context *ctx)
+{
+	if (!macsec_check_offload(macsec->offload, macsec))
+		return NULL;
+
+	return __macsec_get_ops(macsec->offload, macsec, ctx);
+}
+
 /* validate MACsec packet according to IEEE 802.1AE-2006 9.12 */
 static bool macsec_validate_skb(struct sk_buff *skb, u16 icv_len)
 {
@@ -867,8 +921,10 @@ static struct macsec_rx_sc *find_rx_sc_rtnl(struct macsec_secy *secy, sci_t sci)
 	return NULL;
 }
 
-static void handle_not_macsec(struct sk_buff *skb)
+static enum rx_handler_result handle_not_macsec(struct sk_buff *skb)
 {
+	/* Deliver to the uncontrolled port by default */
+	enum rx_handler_result ret = RX_HANDLER_PASS;
 	struct macsec_rxh_data *rxd;
 	struct macsec_dev *macsec;
 
@@ -883,7 +939,8 @@ static void handle_not_macsec(struct sk_buff *skb)
 		struct sk_buff *nskb;
 		struct pcpu_secy_stats *secy_stats = this_cpu_ptr(macsec->stats);
 
-		if (macsec->secy.validate_frames == MACSEC_VALIDATE_STRICT) {
+		if (!macsec_is_offloaded(macsec) &&
+		    macsec->secy.validate_frames == MACSEC_VALIDATE_STRICT) {
 			u64_stats_update_begin(&secy_stats->syncp);
 			secy_stats->stats.InPktsNoTag++;
 			u64_stats_update_end(&secy_stats->syncp);
@@ -902,9 +959,17 @@ static void handle_not_macsec(struct sk_buff *skb)
 			secy_stats->stats.InPktsUntagged++;
 			u64_stats_update_end(&secy_stats->syncp);
 		}
+
+		if (netif_running(macsec->secy.netdev) &&
+		    macsec_is_offloaded(macsec)) {
+			ret = RX_HANDLER_EXACT;
+			goto out;
+		}
 	}
 
+out:
 	rcu_read_unlock();
+	return ret;
 }
 
 static rx_handler_result_t macsec_handle_frame(struct sk_buff **pskb)
@@ -929,12 +994,8 @@ static rx_handler_result_t macsec_handle_frame(struct sk_buff **pskb)
 		goto drop_direct;
 
 	hdr = macsec_ethhdr(skb);
-	if (hdr->eth.h_proto != htons(ETH_P_MACSEC)) {
-		handle_not_macsec(skb);
-
-		/* and deliver to the uncontrolled port */
-		return RX_HANDLER_PASS;
-	}
+	if (hdr->eth.h_proto != htons(ETH_P_MACSEC))
+		return handle_not_macsec(skb);
 
 	skb = skb_unshare(skb, GFP_ATOMIC);
 	*pskb = skb;
@@ -1440,6 +1501,40 @@ static const struct nla_policy macsec_genl_sa_policy[NUM_MACSEC_SA_ATTR] = {
 				 .len = MACSEC_MAX_KEY_LEN, },
 };
 
+/* Offloads an operation to a device driver */
+static int macsec_offload(int (* const func)(struct macsec_context *),
+			  struct macsec_context *ctx)
+{
+	int ret;
+
+	if (unlikely(!func))
+		return 0;
+
+	if (ctx->offload == MACSEC_OFFLOAD_PHY)
+		mutex_lock(&ctx->phydev->lock);
+
+	/* Phase I: prepare. The drive should fail here if there are going to be
+	 * issues in the commit phase.
+	 */
+	ctx->prepare = true;
+	ret = (*func)(ctx);
+	if (ret)
+		goto phy_unlock;
+
+	/* Phase II: commit. This step cannot fail. */
+	ctx->prepare = false;
+	ret = (*func)(ctx);
+	/* This should never happen: commit is not allowed to fail */
+	if (unlikely(ret))
+		WARN(1, "MACsec offloading commit failed (%d)\n", ret);
+
+phy_unlock:
+	if (ctx->offload == MACSEC_OFFLOAD_PHY)
+		mutex_unlock(&ctx->phydev->lock);
+
+	return ret;
+}
+
 static int parse_sa_config(struct nlattr **attrs, struct nlattr **tb_sa)
 {
 	if (!attrs[MACSEC_ATTR_SA_CONFIG])
@@ -1555,13 +1650,40 @@ static int macsec_add_rxsa(struct sk_buff *skb, struct genl_info *info)
 	if (tb_sa[MACSEC_SA_ATTR_ACTIVE])
 		rx_sa->active = !!nla_get_u8(tb_sa[MACSEC_SA_ATTR_ACTIVE]);
 
-	nla_memcpy(rx_sa->key.id, tb_sa[MACSEC_SA_ATTR_KEYID], MACSEC_KEYID_LEN);
 	rx_sa->sc = rx_sc;
+
+	/* If h/w offloading is available, propagate to the device */
+	if (macsec_is_offloaded(netdev_priv(dev))) {
+		const struct macsec_ops *ops;
+		struct macsec_context ctx;
+
+		ops = macsec_get_ops(netdev_priv(dev), &ctx);
+		if (!ops) {
+			err = -EOPNOTSUPP;
+			goto cleanup;
+		}
+
+		ctx.sa.assoc_num = assoc_num;
+		ctx.sa.rx_sa = rx_sa;
+		memcpy(ctx.sa.key, nla_data(tb_sa[MACSEC_SA_ATTR_KEY]),
+		       MACSEC_KEYID_LEN);
+
+		err = macsec_offload(ops->mdo_add_rxsa, &ctx);
+		if (err)
+			goto cleanup;
+	}
+
+	nla_memcpy(rx_sa->key.id, tb_sa[MACSEC_SA_ATTR_KEYID], MACSEC_KEYID_LEN);
 	rcu_assign_pointer(rx_sc->sa[assoc_num], rx_sa);
 
 	rtnl_unlock();
 
 	return 0;
+
+cleanup:
+	kfree(rx_sa);
+	rtnl_unlock();
+	return err;
 }
 
 static bool validate_add_rxsc(struct nlattr **attrs)
@@ -1584,6 +1706,8 @@ static int macsec_add_rxsc(struct sk_buff *skb, struct genl_info *info)
 	struct nlattr **attrs = info->attrs;
 	struct macsec_rx_sc *rx_sc;
 	struct nlattr *tb_rxsc[MACSEC_RXSC_ATTR_MAX + 1];
+	bool was_active;
+	int ret;
 
 	if (!attrs[MACSEC_ATTR_IFINDEX])
 		return -EINVAL;
@@ -1609,12 +1733,35 @@ static int macsec_add_rxsc(struct sk_buff *skb, struct genl_info *info)
 		return PTR_ERR(rx_sc);
 	}
 
+	was_active = rx_sc->active;
 	if (tb_rxsc[MACSEC_RXSC_ATTR_ACTIVE])
 		rx_sc->active = !!nla_get_u8(tb_rxsc[MACSEC_RXSC_ATTR_ACTIVE]);
 
+	if (macsec_is_offloaded(netdev_priv(dev))) {
+		const struct macsec_ops *ops;
+		struct macsec_context ctx;
+
+		ops = macsec_get_ops(netdev_priv(dev), &ctx);
+		if (!ops) {
+			ret = -EOPNOTSUPP;
+			goto cleanup;
+		}
+
+		ctx.rx_sc = rx_sc;
+
+		ret = macsec_offload(ops->mdo_add_rxsc, &ctx);
+		if (ret)
+			goto cleanup;
+	}
+
 	rtnl_unlock();
 
 	return 0;
+
+cleanup:
+	rx_sc->active = was_active;
+	rtnl_unlock();
+	return ret;
 }
 
 static bool validate_add_txsa(struct nlattr **attrs)
@@ -1651,6 +1798,7 @@ static int macsec_add_txsa(struct sk_buff *skb, struct genl_info *info)
 	struct macsec_tx_sa *tx_sa;
 	unsigned char assoc_num;
 	struct nlattr *tb_sa[MACSEC_SA_ATTR_MAX + 1];
+	bool was_operational;
 	int err;
 
 	if (!attrs[MACSEC_ATTR_IFINDEX])
@@ -1701,8 +1849,6 @@ static int macsec_add_txsa(struct sk_buff *skb, struct genl_info *info)
 		return err;
 	}
 
-	nla_memcpy(tx_sa->key.id, tb_sa[MACSEC_SA_ATTR_KEYID], MACSEC_KEYID_LEN);
-
 	spin_lock_bh(&tx_sa->lock);
 	tx_sa->next_pn = nla_get_u32(tb_sa[MACSEC_SA_ATTR_PN]);
 	spin_unlock_bh(&tx_sa->lock);
@@ -1710,14 +1856,43 @@ static int macsec_add_txsa(struct sk_buff *skb, struct genl_info *info)
 	if (tb_sa[MACSEC_SA_ATTR_ACTIVE])
 		tx_sa->active = !!nla_get_u8(tb_sa[MACSEC_SA_ATTR_ACTIVE]);
 
+	was_operational = secy->operational;
 	if (assoc_num == tx_sc->encoding_sa && tx_sa->active)
 		secy->operational = true;
 
+	/* If h/w offloading is available, propagate to the device */
+	if (macsec_is_offloaded(netdev_priv(dev))) {
+		const struct macsec_ops *ops;
+		struct macsec_context ctx;
+
+		ops = macsec_get_ops(netdev_priv(dev), &ctx);
+		if (!ops) {
+			err = -EOPNOTSUPP;
+			goto cleanup;
+		}
+
+		ctx.sa.assoc_num = assoc_num;
+		ctx.sa.tx_sa = tx_sa;
+		memcpy(ctx.sa.key, nla_data(tb_sa[MACSEC_SA_ATTR_KEY]),
+		       MACSEC_KEYID_LEN);
+
+		err = macsec_offload(ops->mdo_add_txsa, &ctx);
+		if (err)
+			goto cleanup;
+	}
+
+	nla_memcpy(tx_sa->key.id, tb_sa[MACSEC_SA_ATTR_KEYID], MACSEC_KEYID_LEN);
 	rcu_assign_pointer(tx_sc->sa[assoc_num], tx_sa);
 
 	rtnl_unlock();
 
 	return 0;
+
+cleanup:
+	secy->operational = was_operational;
+	kfree(tx_sa);
+	rtnl_unlock();
+	return err;
 }
 
 static int macsec_del_rxsa(struct sk_buff *skb, struct genl_info *info)
@@ -1730,6 +1905,7 @@ static int macsec_del_rxsa(struct sk_buff *skb, struct genl_info *info)
 	u8 assoc_num;
 	struct nlattr *tb_rxsc[MACSEC_RXSC_ATTR_MAX + 1];
 	struct nlattr *tb_sa[MACSEC_SA_ATTR_MAX + 1];
+	int ret;
 
 	if (!attrs[MACSEC_ATTR_IFINDEX])
 		return -EINVAL;
@@ -1753,12 +1929,35 @@ static int macsec_del_rxsa(struct sk_buff *skb, struct genl_info *info)
 		return -EBUSY;
 	}
 
+	/* If h/w offloading is available, propagate to the device */
+	if (macsec_is_offloaded(netdev_priv(dev))) {
+		const struct macsec_ops *ops;
+		struct macsec_context ctx;
+
+		ops = macsec_get_ops(netdev_priv(dev), &ctx);
+		if (!ops) {
+			ret = -EOPNOTSUPP;
+			goto cleanup;
+		}
+
+		ctx.sa.assoc_num = assoc_num;
+		ctx.sa.rx_sa = rx_sa;
+
+		ret = macsec_offload(ops->mdo_del_rxsa, &ctx);
+		if (ret)
+			goto cleanup;
+	}
+
 	RCU_INIT_POINTER(rx_sc->sa[assoc_num], NULL);
 	clear_rx_sa(rx_sa);
 
 	rtnl_unlock();
 
 	return 0;
+
+cleanup:
+	rtnl_unlock();
+	return ret;
 }
 
 static int macsec_del_rxsc(struct sk_buff *skb, struct genl_info *info)
@@ -1769,6 +1968,7 @@ static int macsec_del_rxsc(struct sk_buff *skb, struct genl_info *info)
 	struct macsec_rx_sc *rx_sc;
 	sci_t sci;
 	struct nlattr *tb_rxsc[MACSEC_RXSC_ATTR_MAX + 1];
+	int ret;
 
 	if (!attrs[MACSEC_ATTR_IFINDEX])
 		return -EINVAL;
@@ -1795,10 +1995,31 @@ static int macsec_del_rxsc(struct sk_buff *skb, struct genl_info *info)
 		return -ENODEV;
 	}
 
+	/* If h/w offloading is available, propagate to the device */
+	if (macsec_is_offloaded(netdev_priv(dev))) {
+		const struct macsec_ops *ops;
+		struct macsec_context ctx;
+
+		ops = macsec_get_ops(netdev_priv(dev), &ctx);
+		if (!ops) {
+			ret = -EOPNOTSUPP;
+			goto cleanup;
+		}
+
+		ctx.rx_sc = rx_sc;
+		ret = macsec_offload(ops->mdo_del_rxsc, &ctx);
+		if (ret)
+			goto cleanup;
+	}
+
 	free_rx_sc(rx_sc);
 	rtnl_unlock();
 
 	return 0;
+
+cleanup:
+	rtnl_unlock();
+	return ret;
 }
 
 static int macsec_del_txsa(struct sk_buff *skb, struct genl_info *info)
@@ -1810,6 +2031,7 @@ static int macsec_del_txsa(struct sk_buff *skb, struct genl_info *info)
 	struct macsec_tx_sa *tx_sa;
 	u8 assoc_num;
 	struct nlattr *tb_sa[MACSEC_SA_ATTR_MAX + 1];
+	int ret;
 
 	if (!attrs[MACSEC_ATTR_IFINDEX])
 		return -EINVAL;
@@ -1830,12 +2052,35 @@ static int macsec_del_txsa(struct sk_buff *skb, struct genl_info *info)
 		return -EBUSY;
 	}
 
+	/* If h/w offloading is available, propagate to the device */
+	if (macsec_is_offloaded(netdev_priv(dev))) {
+		const struct macsec_ops *ops;
+		struct macsec_context ctx;
+
+		ops = macsec_get_ops(netdev_priv(dev), &ctx);
+		if (!ops) {
+			ret = -EOPNOTSUPP;
+			goto cleanup;
+		}
+
+		ctx.sa.assoc_num = assoc_num;
+		ctx.sa.tx_sa = tx_sa;
+
+		ret = macsec_offload(ops->mdo_del_txsa, &ctx);
+		if (ret)
+			goto cleanup;
+	}
+
 	RCU_INIT_POINTER(tx_sc->sa[assoc_num], NULL);
 	clear_tx_sa(tx_sa);
 
 	rtnl_unlock();
 
 	return 0;
+
+cleanup:
+	rtnl_unlock();
+	return ret;
 }
 
 static bool validate_upd_sa(struct nlattr **attrs)
@@ -1868,6 +2113,9 @@ static int macsec_upd_txsa(struct sk_buff *skb, struct genl_info *info)
 	struct macsec_tx_sa *tx_sa;
 	u8 assoc_num;
 	struct nlattr *tb_sa[MACSEC_SA_ATTR_MAX + 1];
+	bool was_operational, was_active;
+	u32 prev_pn = 0;
+	int ret = 0;
 
 	if (!attrs[MACSEC_ATTR_IFINDEX])
 		return -EINVAL;
@@ -1888,19 +2136,52 @@ static int macsec_upd_txsa(struct sk_buff *skb, struct genl_info *info)
 
 	if (tb_sa[MACSEC_SA_ATTR_PN]) {
 		spin_lock_bh(&tx_sa->lock);
+		prev_pn = tx_sa->next_pn;
 		tx_sa->next_pn = nla_get_u32(tb_sa[MACSEC_SA_ATTR_PN]);
 		spin_unlock_bh(&tx_sa->lock);
 	}
 
+	was_active = tx_sa->active;
 	if (tb_sa[MACSEC_SA_ATTR_ACTIVE])
 		tx_sa->active = nla_get_u8(tb_sa[MACSEC_SA_ATTR_ACTIVE]);
 
+	was_operational = secy->operational;
 	if (assoc_num == tx_sc->encoding_sa)
 		secy->operational = tx_sa->active;
 
+	/* If h/w offloading is available, propagate to the device */
+	if (macsec_is_offloaded(netdev_priv(dev))) {
+		const struct macsec_ops *ops;
+		struct macsec_context ctx;
+
+		ops = macsec_get_ops(netdev_priv(dev), &ctx);
+		if (!ops) {
+			ret = -EOPNOTSUPP;
+			goto cleanup;
+		}
+
+		ctx.sa.assoc_num = assoc_num;
+		ctx.sa.tx_sa = tx_sa;
+
+		ret = macsec_offload(ops->mdo_upd_txsa, &ctx);
+		if (ret)
+			goto cleanup;
+	}
+
 	rtnl_unlock();
 
 	return 0;
+
+cleanup:
+	if (tb_sa[MACSEC_SA_ATTR_PN]) {
+		spin_lock_bh(&tx_sa->lock);
+		tx_sa->next_pn = prev_pn;
+		spin_unlock_bh(&tx_sa->lock);
+	}
+	tx_sa->active = was_active;
+	secy->operational = was_operational;
+	rtnl_unlock();
+	return ret;
 }
 
 static int macsec_upd_rxsa(struct sk_buff *skb, struct genl_info *info)
@@ -1913,6 +2194,9 @@ static int macsec_upd_rxsa(struct sk_buff *skb, struct genl_info *info)
 	u8 assoc_num;
 	struct nlattr *tb_rxsc[MACSEC_RXSC_ATTR_MAX + 1];
 	struct nlattr *tb_sa[MACSEC_SA_ATTR_MAX + 1];
+	bool was_active;
+	u32 prev_pn = 0;
+	int ret = 0;
 
 	if (!attrs[MACSEC_ATTR_IFINDEX])
 		return -EINVAL;
@@ -1936,15 +2220,46 @@ static int macsec_upd_rxsa(struct sk_buff *skb, struct genl_info *info)
 
 	if (tb_sa[MACSEC_SA_ATTR_PN]) {
 		spin_lock_bh(&rx_sa->lock);
+		prev_pn = rx_sa->next_pn;
 		rx_sa->next_pn = nla_get_u32(tb_sa[MACSEC_SA_ATTR_PN]);
 		spin_unlock_bh(&rx_sa->lock);
 	}
 
+	was_active = rx_sa->active;
 	if (tb_sa[MACSEC_SA_ATTR_ACTIVE])
 		rx_sa->active = nla_get_u8(tb_sa[MACSEC_SA_ATTR_ACTIVE]);
 
+	/* If h/w offloading is available, propagate to the device */
+	if (macsec_is_offloaded(netdev_priv(dev))) {
+		const struct macsec_ops *ops;
+		struct macsec_context ctx;
+
+		ops = macsec_get_ops(netdev_priv(dev), &ctx);
+		if (!ops) {
+			ret = -EOPNOTSUPP;
+			goto cleanup;
+		}
+
+		ctx.sa.assoc_num = assoc_num;
+		ctx.sa.rx_sa = rx_sa;
+
+		ret = macsec_offload(ops->mdo_upd_rxsa, &ctx);
+		if (ret)
+			goto cleanup;
+	}
+
 	rtnl_unlock();
 	return 0;
+
+cleanup:
+	if (tb_sa[MACSEC_SA_ATTR_PN]) {
+		spin_lock_bh(&rx_sa->lock);
+		rx_sa->next_pn = prev_pn;
+		spin_unlock_bh(&rx_sa->lock);
+	}
+	rx_sa->active = was_active;
+	rtnl_unlock();
+	return ret;
 }
 
 static int macsec_upd_rxsc(struct sk_buff *skb, struct genl_info *info)
@@ -1954,6 +2269,9 @@ static int macsec_upd_rxsc(struct sk_buff *skb, struct genl_info *info)
 	struct macsec_secy *secy;
 	struct macsec_rx_sc *rx_sc;
 	struct nlattr *tb_rxsc[MACSEC_RXSC_ATTR_MAX + 1];
+	unsigned int prev_n_rx_sc;
+	bool was_active;
+	int ret;
 
 	if (!attrs[MACSEC_ATTR_IFINDEX])
 		return -EINVAL;
@@ -1971,6 +2289,8 @@ static int macsec_upd_rxsc(struct sk_buff *skb, struct genl_info *info)
 		return PTR_ERR(rx_sc);
 	}
 
+	was_active = rx_sc->active;
+	prev_n_rx_sc = secy->n_rx_sc;
 	if (tb_rxsc[MACSEC_RXSC_ATTR_ACTIVE]) {
 		bool new = !!nla_get_u8(tb_rxsc[MACSEC_RXSC_ATTR_ACTIVE]);
 
@@ -1980,9 +2300,33 @@ static int macsec_upd_rxsc(struct sk_buff *skb, struct genl_info *info)
 		rx_sc->active = new;
 	}
 
+	/* If h/w offloading is available, propagate to the device */
+	if (macsec_is_offloaded(netdev_priv(dev))) {
+		const struct macsec_ops *ops;
+		struct macsec_context ctx;
+
+		ops = macsec_get_ops(netdev_priv(dev), &ctx);
+		if (!ops) {
+			ret = -EOPNOTSUPP;
+			goto cleanup;
+		}
+
+		ctx.rx_sc = rx_sc;
+
+		ret = macsec_offload(ops->mdo_upd_rxsc, &ctx);
+		if (ret)
+			goto cleanup;
+	}
+
 	rtnl_unlock();
 
 	return 0;
+
+cleanup:
+	secy->n_rx_sc = prev_n_rx_sc;
+	rx_sc->active = was_active;
+	rtnl_unlock();
+	return ret;
 }
 
 static int copy_tx_sa_stats(struct sk_buff *skb,
@@ -2550,6 +2894,11 @@ static netdev_tx_t macsec_start_xmit(struct sk_buff *skb,
 	struct pcpu_secy_stats *secy_stats;
 	int ret, len;
 
+	if (macsec_is_offloaded(netdev_priv(dev))) {
+		skb->dev = macsec->real_dev;
+		return dev_queue_xmit(skb);
+	}
+
 	/* 10.5 */
 	if (!secy->protect_frames) {
 		secy_stats = this_cpu_ptr(macsec->stats);
@@ -2663,6 +3012,22 @@ static int macsec_dev_open(struct net_device *dev)
 			goto clear_allmulti;
 	}
 
+	/* If h/w offloading is available, propagate to the device */
+	if (macsec_is_offloaded(macsec)) {
+		const struct macsec_ops *ops;
+		struct macsec_context ctx;
+
+		ops = macsec_get_ops(netdev_priv(dev), &ctx);
+		if (!ops) {
+			err = -EOPNOTSUPP;
+			goto clear_allmulti;
+		}
+
+		err = macsec_offload(ops->mdo_dev_open, &ctx);
+		if (err)
+			goto clear_allmulti;
+	}
+
 	if (netif_carrier_ok(real_dev))
 		netif_carrier_on(dev);
 
@@ -2683,6 +3048,16 @@ static int macsec_dev_stop(struct net_device *dev)
 
 	netif_carrier_off(dev);
 
+	/* If h/w offloading is available, propagate to the device */
+	if (macsec_is_offloaded(macsec)) {
+		const struct macsec_ops *ops;
+		struct macsec_context ctx;
+
+		ops = macsec_get_ops(macsec, &ctx);
+		if (ops)
+			macsec_offload(ops->mdo_dev_stop, &ctx);
+	}
+
 	dev_mc_unsync(real_dev, dev);
 	dev_uc_unsync(real_dev, dev);
 
@@ -2914,6 +3289,11 @@ static int macsec_changelink(struct net_device *dev, struct nlattr *tb[],
 			     struct nlattr *data[],
 			     struct netlink_ext_ack *extack)
 {
+	struct macsec_dev *macsec = macsec_priv(dev);
+	struct macsec_tx_sa tx_sc;
+	struct macsec_secy secy;
+	int ret;
+
 	if (!data)
 		return 0;
 
@@ -2923,7 +3303,41 @@ static int macsec_changelink(struct net_device *dev, struct nlattr *tb[],
 	    data[IFLA_MACSEC_PORT])
 		return -EINVAL;
 
-	return macsec_changelink_common(dev, data);
+	/* Keep a copy of unmodified secy and tx_sc, in case the offload
+	 * propagation fails, to revert macsec_changelink_common.
+	 */
+	memcpy(&secy, &macsec->secy, sizeof(secy));
+	memcpy(&tx_sc, &macsec->secy.tx_sc, sizeof(tx_sc));
+
+	ret = macsec_changelink_common(dev, data);
+	if (ret)
+		return ret;
+
+	/* If h/w offloading is available, propagate to the device */
+	if (macsec_is_offloaded(macsec)) {
+		const struct macsec_ops *ops;
+		struct macsec_context ctx;
+		int ret;
+
+		ops = macsec_get_ops(netdev_priv(dev), &ctx);
+		if (!ops) {
+			ret = -EOPNOTSUPP;
+			goto cleanup;
+		}
+
+		ctx.secy = &macsec->secy;
+		ret = macsec_offload(ops->mdo_upd_secy, &ctx);
+		if (ret)
+			goto cleanup;
+	}
+
+	return 0;
+
+cleanup:
+	memcpy(&macsec->secy.tx_sc, &tx_sc, sizeof(tx_sc));
+	memcpy(&macsec->secy, &secy, sizeof(secy));
+
+	return ret;
 }
 
 static void macsec_del_dev(struct macsec_dev *macsec)
@@ -2966,6 +3380,18 @@ static void macsec_dellink(struct net_device *dev, struct list_head *head)
 	struct net_device *real_dev = macsec->real_dev;
 	struct macsec_rxh_data *rxd = macsec_data_rtnl(real_dev);
 
+	/* If h/w offloading is available, propagate to the device */
+	if (macsec_is_offloaded(macsec)) {
+		const struct macsec_ops *ops;
+		struct macsec_context ctx;
+
+		ops = macsec_get_ops(netdev_priv(dev), &ctx);
+		if (ops) {
+			ctx.secy = &macsec->secy;
+			macsec_offload(ops->mdo_del_secy, &ctx);
+		}
+	}
+
 	macsec_common_dellink(dev, head);
 
 	if (list_empty(&rxd->secys)) {
@@ -3077,6 +3503,9 @@ static int macsec_newlink(struct net *net, struct net_device *dev,
 
 	macsec->real_dev = real_dev;
 
+	/* MACsec offloading is off by default */
+	macsec->offload = MACSEC_OFFLOAD_OFF;
+
 	if (data && data[IFLA_MACSEC_ICV_LEN])
 		icv_len = nla_get_u8(data[IFLA_MACSEC_ICV_LEN]);
 	dev->mtu = real_dev->mtu - icv_len - macsec_extra_len(true);
-- 
cgit v1.2.3-59-g8ed1b


From dcb780fb279514f268826f2e9f4df3bc75610703 Mon Sep 17 00:00:00 2001
From: Antoine Tenart <antoine.tenart@bootlin.com>
Date: Mon, 13 Jan 2020 23:31:44 +0100
Subject: net: macsec: add nla support for changing the offloading selection

MACsec offloading to underlying hardware devices is disabled by default
(the software implementation is used). This patch adds support for
changing this setting through the MACsec netlink interface. Many checks
are done when enabling offloading on a given MACsec interface as there
are limitations (it must be supported by the hardware, only a single
interface can be offloaded on a given physical device at a time, rules
can't be moved for now).

Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/macsec.c           | 145 ++++++++++++++++++++++++++++++++++++++++-
 include/uapi/linux/if_macsec.h |  11 ++++
 2 files changed, 153 insertions(+), 3 deletions(-)

diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c
index 36b0416381bf..e515919e8687 100644
--- a/drivers/net/macsec.c
+++ b/drivers/net/macsec.c
@@ -1484,6 +1484,7 @@ static const struct nla_policy macsec_genl_policy[NUM_MACSEC_ATTR] = {
 	[MACSEC_ATTR_IFINDEX] = { .type = NLA_U32 },
 	[MACSEC_ATTR_RXSC_CONFIG] = { .type = NLA_NESTED },
 	[MACSEC_ATTR_SA_CONFIG] = { .type = NLA_NESTED },
+	[MACSEC_ATTR_OFFLOAD] = { .type = NLA_NESTED },
 };
 
 static const struct nla_policy macsec_genl_rxsc_policy[NUM_MACSEC_RXSC_ATTR] = {
@@ -1501,6 +1502,10 @@ static const struct nla_policy macsec_genl_sa_policy[NUM_MACSEC_SA_ATTR] = {
 				 .len = MACSEC_MAX_KEY_LEN, },
 };
 
+static const struct nla_policy macsec_genl_offload_policy[NUM_MACSEC_OFFLOAD_ATTR] = {
+	[MACSEC_OFFLOAD_ATTR_TYPE] = { .type = NLA_U8 },
+};
+
 /* Offloads an operation to a device driver */
 static int macsec_offload(int (* const func)(struct macsec_context *),
 			  struct macsec_context *ctx)
@@ -2329,6 +2334,126 @@ cleanup:
 	return ret;
 }
 
+static bool macsec_is_configured(struct macsec_dev *macsec)
+{
+	struct macsec_secy *secy = &macsec->secy;
+	struct macsec_tx_sc *tx_sc = &secy->tx_sc;
+	int i;
+
+	if (secy->n_rx_sc > 0)
+		return true;
+
+	for (i = 0; i < MACSEC_NUM_AN; i++)
+		if (tx_sc->sa[i])
+			return true;
+
+	return false;
+}
+
+static int macsec_upd_offload(struct sk_buff *skb, struct genl_info *info)
+{
+	struct nlattr *tb_offload[MACSEC_OFFLOAD_ATTR_MAX + 1];
+	enum macsec_offload offload, prev_offload;
+	int (*func)(struct macsec_context *ctx);
+	struct nlattr **attrs = info->attrs;
+	struct net_device *dev, *loop_dev;
+	const struct macsec_ops *ops;
+	struct macsec_context ctx;
+	struct macsec_dev *macsec;
+	struct net *loop_net;
+	int ret;
+
+	if (!attrs[MACSEC_ATTR_IFINDEX])
+		return -EINVAL;
+
+	if (!attrs[MACSEC_ATTR_OFFLOAD])
+		return -EINVAL;
+
+	if (nla_parse_nested_deprecated(tb_offload, MACSEC_OFFLOAD_ATTR_MAX,
+					attrs[MACSEC_ATTR_OFFLOAD],
+					macsec_genl_offload_policy, NULL))
+		return -EINVAL;
+
+	dev = get_dev_from_nl(genl_info_net(info), attrs);
+	if (IS_ERR(dev))
+		return PTR_ERR(dev);
+	macsec = macsec_priv(dev);
+
+	offload = nla_get_u8(tb_offload[MACSEC_OFFLOAD_ATTR_TYPE]);
+	if (macsec->offload == offload)
+		return 0;
+
+	/* Check if the offloading mode is supported by the underlying layers */
+	if (offload != MACSEC_OFFLOAD_OFF &&
+	    !macsec_check_offload(offload, macsec))
+		return -EOPNOTSUPP;
+
+	if (offload == MACSEC_OFFLOAD_OFF)
+		goto skip_limitation;
+
+	/* Check the physical interface isn't offloading another interface
+	 * first.
+	 */
+	for_each_net(loop_net) {
+		for_each_netdev(loop_net, loop_dev) {
+			struct macsec_dev *priv;
+
+			if (!netif_is_macsec(loop_dev))
+				continue;
+
+			priv = macsec_priv(loop_dev);
+
+			if (priv->real_dev == macsec->real_dev &&
+			    priv->offload != MACSEC_OFFLOAD_OFF)
+				return -EBUSY;
+		}
+	}
+
+skip_limitation:
+	/* Check if the net device is busy. */
+	if (netif_running(dev))
+		return -EBUSY;
+
+	rtnl_lock();
+
+	prev_offload = macsec->offload;
+	macsec->offload = offload;
+
+	/* Check if the device already has rules configured: we do not support
+	 * rules migration.
+	 */
+	if (macsec_is_configured(macsec)) {
+		ret = -EBUSY;
+		goto rollback;
+	}
+
+	ops = __macsec_get_ops(offload == MACSEC_OFFLOAD_OFF ? prev_offload : offload,
+			       macsec, &ctx);
+	if (!ops) {
+		ret = -EOPNOTSUPP;
+		goto rollback;
+	}
+
+	if (prev_offload == MACSEC_OFFLOAD_OFF)
+		func = ops->mdo_add_secy;
+	else
+		func = ops->mdo_del_secy;
+
+	ctx.secy = &macsec->secy;
+	ret = macsec_offload(func, &ctx);
+	if (ret)
+		goto rollback;
+
+	rtnl_unlock();
+	return 0;
+
+rollback:
+	macsec->offload = prev_offload;
+
+	rtnl_unlock();
+	return ret;
+}
+
 static int copy_tx_sa_stats(struct sk_buff *skb,
 			    struct macsec_tx_sa_stats __percpu *pstats)
 {
@@ -2590,12 +2715,13 @@ static noinline_for_stack int
 dump_secy(struct macsec_secy *secy, struct net_device *dev,
 	  struct sk_buff *skb, struct netlink_callback *cb)
 {
-	struct macsec_rx_sc *rx_sc;
+	struct macsec_dev *macsec = netdev_priv(dev);
 	struct macsec_tx_sc *tx_sc = &secy->tx_sc;
 	struct nlattr *txsa_list, *rxsc_list;
-	int i, j;
-	void *hdr;
+	struct macsec_rx_sc *rx_sc;
 	struct nlattr *attr;
+	void *hdr;
+	int i, j;
 
 	hdr = genlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq,
 			  &macsec_fam, NLM_F_MULTI, MACSEC_CMD_GET_TXSC);
@@ -2607,6 +2733,13 @@ dump_secy(struct macsec_secy *secy, struct net_device *dev,
 	if (nla_put_u32(skb, MACSEC_ATTR_IFINDEX, dev->ifindex))
 		goto nla_put_failure;
 
+	attr = nla_nest_start_noflag(skb, MACSEC_ATTR_OFFLOAD);
+	if (!attr)
+		goto nla_put_failure;
+	if (nla_put_u8(skb, MACSEC_OFFLOAD_ATTR_TYPE, macsec->offload))
+		goto nla_put_failure;
+	nla_nest_end(skb, attr);
+
 	if (nla_put_secy(secy, skb))
 		goto nla_put_failure;
 
@@ -2872,6 +3005,12 @@ static const struct genl_ops macsec_genl_ops[] = {
 		.doit = macsec_upd_rxsa,
 		.flags = GENL_ADMIN_PERM,
 	},
+	{
+		.cmd = MACSEC_CMD_UPD_OFFLOAD,
+		.validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP,
+		.doit = macsec_upd_offload,
+		.flags = GENL_ADMIN_PERM,
+	},
 };
 
 static struct genl_family macsec_fam __ro_after_init = {
diff --git a/include/uapi/linux/if_macsec.h b/include/uapi/linux/if_macsec.h
index 98e4d5d7c45c..1d63c43c38cc 100644
--- a/include/uapi/linux/if_macsec.h
+++ b/include/uapi/linux/if_macsec.h
@@ -45,6 +45,7 @@ enum macsec_attrs {
 	MACSEC_ATTR_RXSC_LIST,   /* dump, nested, macsec_rxsc_attrs for each RXSC */
 	MACSEC_ATTR_TXSC_STATS,  /* dump, nested, macsec_txsc_stats_attr */
 	MACSEC_ATTR_SECY_STATS,  /* dump, nested, macsec_secy_stats_attr */
+	MACSEC_ATTR_OFFLOAD,     /* config, nested, macsec_offload_attrs */
 	__MACSEC_ATTR_END,
 	NUM_MACSEC_ATTR = __MACSEC_ATTR_END,
 	MACSEC_ATTR_MAX = __MACSEC_ATTR_END - 1,
@@ -97,6 +98,15 @@ enum macsec_sa_attrs {
 	MACSEC_SA_ATTR_MAX = __MACSEC_SA_ATTR_END - 1,
 };
 
+enum macsec_offload_attrs {
+	MACSEC_OFFLOAD_ATTR_UNSPEC,
+	MACSEC_OFFLOAD_ATTR_TYPE, /* config/dump, u8 0..2 */
+	MACSEC_OFFLOAD_ATTR_PAD,
+	__MACSEC_OFFLOAD_ATTR_END,
+	NUM_MACSEC_OFFLOAD_ATTR = __MACSEC_OFFLOAD_ATTR_END,
+	MACSEC_OFFLOAD_ATTR_MAX = __MACSEC_OFFLOAD_ATTR_END - 1,
+};
+
 enum macsec_nl_commands {
 	MACSEC_CMD_GET_TXSC,
 	MACSEC_CMD_ADD_RXSC,
@@ -108,6 +118,7 @@ enum macsec_nl_commands {
 	MACSEC_CMD_ADD_RXSA,
 	MACSEC_CMD_DEL_RXSA,
 	MACSEC_CMD_UPD_RXSA,
+	MACSEC_CMD_UPD_OFFLOAD,
 };
 
 /* u64 per-RXSC stats */
-- 
cgit v1.2.3-59-g8ed1b


From 1bbe0ecc2a1a008bcfeb7fd2d8f95c8e9a1867c6 Mon Sep 17 00:00:00 2001
From: Antoine Tenart <antoine.tenart@bootlin.com>
Date: Mon, 13 Jan 2020 23:31:45 +0100
Subject: net: phy: mscc: macsec initialization

This patch adds support for initializing the MACsec engine found within
some Microsemi PHYs. The engine is initialized in a passthrough mode and
does not modify any incoming or outgoing packet. But thanks to this it
now can be configured to perform MACsec transformations on packets,
which will be supported by a future patch.

The MACsec read and write functions are wrapped into two versions: one
called during the init phase, and the other one later on. This is
because the init functions in the Microsemi PHY driver are called while
the MDIO bus lock is taken.

Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/mscc.c           | 382 +++++++++++++++++++++++++++++++++++++++
 drivers/net/phy/mscc_fc_buffer.h |  64 +++++++
 drivers/net/phy/mscc_mac.h       | 159 ++++++++++++++++
 drivers/net/phy/mscc_macsec.h    | 260 ++++++++++++++++++++++++++
 4 files changed, 865 insertions(+)
 create mode 100644 drivers/net/phy/mscc_fc_buffer.h
 create mode 100644 drivers/net/phy/mscc_mac.h
 create mode 100644 drivers/net/phy/mscc_macsec.h

diff --git a/drivers/net/phy/mscc.c b/drivers/net/phy/mscc.c
index 50214c081164..8579a59a1336 100644
--- a/drivers/net/phy/mscc.c
+++ b/drivers/net/phy/mscc.c
@@ -18,6 +18,10 @@
 #include <linux/netdevice.h>
 #include <dt-bindings/net/mscc-phy-vsc8531.h>
 
+#include "mscc_macsec.h"
+#include "mscc_mac.h"
+#include "mscc_fc_buffer.h"
+
 enum rgmii_rx_clock_delay {
 	RGMII_RX_CLK_DELAY_0_2_NS = 0,
 	RGMII_RX_CLK_DELAY_0_8_NS = 1,
@@ -121,6 +125,26 @@ enum rgmii_rx_clock_delay {
 #define PHY_S6G_PLL_FSM_CTRL_DATA_POS	  8
 #define PHY_S6G_PLL_FSM_ENA_POS		  7
 
+#define MSCC_EXT_PAGE_MACSEC_17		  17
+#define MSCC_EXT_PAGE_MACSEC_18		  18
+
+#define MSCC_EXT_PAGE_MACSEC_19		  19
+#define MSCC_PHY_MACSEC_19_REG_ADDR(x)	  (x)
+#define MSCC_PHY_MACSEC_19_TARGET(x)	  ((x) << 12)
+#define MSCC_PHY_MACSEC_19_READ		  BIT(14)
+#define MSCC_PHY_MACSEC_19_CMD		  BIT(15)
+
+#define MSCC_EXT_PAGE_MACSEC_20		  20
+#define MSCC_PHY_MACSEC_20_TARGET(x)	  (x)
+enum macsec_bank {
+	FC_BUFFER   = 0x04,
+	HOST_MAC    = 0x05,
+	LINE_MAC    = 0x06,
+	IP_1588     = 0x0e,
+	MACSEC_INGR = 0x38,
+	MACSEC_EGR  = 0x3c,
+};
+
 #define MSCC_EXT_PAGE_ACCESS		  31
 #define MSCC_PHY_PAGE_STANDARD		  0x0000 /* Standard registers */
 #define MSCC_PHY_PAGE_EXTENDED		  0x0001 /* Extended registers */
@@ -128,6 +152,7 @@ enum rgmii_rx_clock_delay {
 #define MSCC_PHY_PAGE_EXTENDED_3	  0x0003 /* Extended reg - page 3 */
 #define MSCC_PHY_PAGE_EXTENDED_4	  0x0004 /* Extended reg - page 4 */
 #define MSCC_PHY_PAGE_CSR_CNTL		  MSCC_PHY_PAGE_EXTENDED_4
+#define MSCC_PHY_PAGE_MACSEC		  MSCC_PHY_PAGE_EXTENDED_4
 /* Extended reg - GPIO; this is a bank of registers that are shared for all PHYs
  * in the same package.
  */
@@ -1584,6 +1609,350 @@ out:
 	return ret;
 }
 
+#if IS_ENABLED(CONFIG_MACSEC)
+static u32 vsc8584_macsec_phy_read(struct phy_device *phydev,
+				   enum macsec_bank bank, u32 reg)
+{
+	u32 val, val_l = 0, val_h = 0;
+	unsigned long deadline;
+	int rc;
+
+	rc = phy_select_page(phydev, MSCC_PHY_PAGE_MACSEC);
+	if (rc < 0)
+		goto failed;
+
+	__phy_write(phydev, MSCC_EXT_PAGE_MACSEC_20,
+		    MSCC_PHY_MACSEC_20_TARGET(bank >> 2));
+
+	if (bank >> 2 == 0x1)
+		/* non-MACsec access */
+		bank &= 0x3;
+	else
+		bank = 0;
+
+	__phy_write(phydev, MSCC_EXT_PAGE_MACSEC_19,
+		    MSCC_PHY_MACSEC_19_CMD | MSCC_PHY_MACSEC_19_READ |
+		    MSCC_PHY_MACSEC_19_REG_ADDR(reg) |
+		    MSCC_PHY_MACSEC_19_TARGET(bank));
+
+	deadline = jiffies + msecs_to_jiffies(PROC_CMD_NCOMPLETED_TIMEOUT_MS);
+	do {
+		val = __phy_read(phydev, MSCC_EXT_PAGE_MACSEC_19);
+	} while (time_before(jiffies, deadline) && !(val & MSCC_PHY_MACSEC_19_CMD));
+
+	val_l = __phy_read(phydev, MSCC_EXT_PAGE_MACSEC_17);
+	val_h = __phy_read(phydev, MSCC_EXT_PAGE_MACSEC_18);
+
+failed:
+	phy_restore_page(phydev, rc, rc);
+
+	return (val_h << 16) | val_l;
+}
+
+static void vsc8584_macsec_phy_write(struct phy_device *phydev,
+				     enum macsec_bank bank, u32 reg, u32 val)
+{
+	unsigned long deadline;
+	int rc;
+
+	rc = phy_select_page(phydev, MSCC_PHY_PAGE_MACSEC);
+	if (rc < 0)
+		goto failed;
+
+	__phy_write(phydev, MSCC_EXT_PAGE_MACSEC_20,
+		    MSCC_PHY_MACSEC_20_TARGET(bank >> 2));
+
+	if ((bank >> 2 == 0x1) || (bank >> 2 == 0x3))
+		bank &= 0x3;
+	else
+		/* MACsec access */
+		bank = 0;
+
+	__phy_write(phydev, MSCC_EXT_PAGE_MACSEC_17, (u16)val);
+	__phy_write(phydev, MSCC_EXT_PAGE_MACSEC_18, (u16)(val >> 16));
+
+	__phy_write(phydev, MSCC_EXT_PAGE_MACSEC_19,
+		    MSCC_PHY_MACSEC_19_CMD | MSCC_PHY_MACSEC_19_REG_ADDR(reg) |
+		    MSCC_PHY_MACSEC_19_TARGET(bank));
+
+	deadline = jiffies + msecs_to_jiffies(PROC_CMD_NCOMPLETED_TIMEOUT_MS);
+	do {
+		val = __phy_read(phydev, MSCC_EXT_PAGE_MACSEC_19);
+	} while (time_before(jiffies, deadline) && !(val & MSCC_PHY_MACSEC_19_CMD));
+
+failed:
+	phy_restore_page(phydev, rc, rc);
+}
+
+static void vsc8584_macsec_classification(struct phy_device *phydev,
+					  enum macsec_bank bank)
+{
+	/* enable VLAN tag parsing */
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_SAM_CP_TAG,
+				 MSCC_MS_SAM_CP_TAG_PARSE_STAG |
+				 MSCC_MS_SAM_CP_TAG_PARSE_QTAG |
+				 MSCC_MS_SAM_CP_TAG_PARSE_QINQ);
+}
+
+static void vsc8584_macsec_flow_default_action(struct phy_device *phydev,
+					       enum macsec_bank bank,
+					       bool block)
+{
+	u32 port = (bank == MACSEC_INGR) ?
+		    MSCC_MS_PORT_UNCONTROLLED : MSCC_MS_PORT_COMMON;
+	u32 action = MSCC_MS_FLOW_BYPASS;
+
+	if (block)
+		action = MSCC_MS_FLOW_DROP;
+
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_SAM_NM_FLOW_NCP,
+				 /* MACsec untagged */
+				 MSCC_MS_SAM_NM_FLOW_NCP_UNTAGGED_FLOW_TYPE(action) |
+				 MSCC_MS_SAM_NM_FLOW_NCP_UNTAGGED_DROP_ACTION(MSCC_MS_ACTION_DROP) |
+				 MSCC_MS_SAM_NM_FLOW_NCP_UNTAGGED_DEST_PORT(port) |
+				 /* MACsec tagged */
+				 MSCC_MS_SAM_NM_FLOW_NCP_TAGGED_FLOW_TYPE(action) |
+				 MSCC_MS_SAM_NM_FLOW_NCP_TAGGED_DROP_ACTION(MSCC_MS_ACTION_DROP) |
+				 MSCC_MS_SAM_NM_FLOW_NCP_TAGGED_DEST_PORT(port) |
+				 /* Bad tag */
+				 MSCC_MS_SAM_NM_FLOW_NCP_BADTAG_FLOW_TYPE(action) |
+				 MSCC_MS_SAM_NM_FLOW_NCP_BADTAG_DROP_ACTION(MSCC_MS_ACTION_DROP) |
+				 MSCC_MS_SAM_NM_FLOW_NCP_BADTAG_DEST_PORT(port) |
+				 /* Kay tag */
+				 MSCC_MS_SAM_NM_FLOW_NCP_KAY_FLOW_TYPE(action) |
+				 MSCC_MS_SAM_NM_FLOW_NCP_KAY_DROP_ACTION(MSCC_MS_ACTION_DROP) |
+				 MSCC_MS_SAM_NM_FLOW_NCP_KAY_DEST_PORT(port));
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_SAM_NM_FLOW_CP,
+				 /* MACsec untagged */
+				 MSCC_MS_SAM_NM_FLOW_NCP_UNTAGGED_FLOW_TYPE(action) |
+				 MSCC_MS_SAM_NM_FLOW_CP_UNTAGGED_DROP_ACTION(MSCC_MS_ACTION_DROP) |
+				 MSCC_MS_SAM_NM_FLOW_CP_UNTAGGED_DEST_PORT(port) |
+				 /* MACsec tagged */
+				 MSCC_MS_SAM_NM_FLOW_NCP_TAGGED_FLOW_TYPE(action) |
+				 MSCC_MS_SAM_NM_FLOW_CP_TAGGED_DROP_ACTION(MSCC_MS_ACTION_DROP) |
+				 MSCC_MS_SAM_NM_FLOW_CP_TAGGED_DEST_PORT(port) |
+				 /* Bad tag */
+				 MSCC_MS_SAM_NM_FLOW_NCP_BADTAG_FLOW_TYPE(action) |
+				 MSCC_MS_SAM_NM_FLOW_CP_BADTAG_DROP_ACTION(MSCC_MS_ACTION_DROP) |
+				 MSCC_MS_SAM_NM_FLOW_CP_BADTAG_DEST_PORT(port) |
+				 /* Kay tag */
+				 MSCC_MS_SAM_NM_FLOW_NCP_KAY_FLOW_TYPE(action) |
+				 MSCC_MS_SAM_NM_FLOW_CP_KAY_DROP_ACTION(MSCC_MS_ACTION_DROP) |
+				 MSCC_MS_SAM_NM_FLOW_CP_KAY_DEST_PORT(port));
+}
+
+static void vsc8584_macsec_integrity_checks(struct phy_device *phydev,
+					    enum macsec_bank bank)
+{
+	u32 val;
+
+	if (bank != MACSEC_INGR)
+		return;
+
+	/* Set default rules to pass unmatched frames */
+	val = vsc8584_macsec_phy_read(phydev, bank,
+				      MSCC_MS_PARAMS2_IG_CC_CONTROL);
+	val |= MSCC_MS_PARAMS2_IG_CC_CONTROL_NON_MATCH_CTRL_ACT |
+	       MSCC_MS_PARAMS2_IG_CC_CONTROL_NON_MATCH_ACT;
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_PARAMS2_IG_CC_CONTROL,
+				 val);
+
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_PARAMS2_IG_CP_TAG,
+				 MSCC_MS_PARAMS2_IG_CP_TAG_PARSE_STAG |
+				 MSCC_MS_PARAMS2_IG_CP_TAG_PARSE_QTAG |
+				 MSCC_MS_PARAMS2_IG_CP_TAG_PARSE_QINQ);
+}
+
+static void vsc8584_macsec_block_init(struct phy_device *phydev,
+				      enum macsec_bank bank)
+{
+	u32 val;
+	int i;
+
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_ENA_CFG,
+				 MSCC_MS_ENA_CFG_SW_RST |
+				 MSCC_MS_ENA_CFG_MACSEC_BYPASS_ENA);
+
+	/* Set the MACsec block out of s/w reset and enable clocks */
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_ENA_CFG,
+				 MSCC_MS_ENA_CFG_CLK_ENA);
+
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_STATUS_CONTEXT_CTRL,
+				 bank == MACSEC_INGR ? 0xe5880214 : 0xe5880218);
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_MISC_CONTROL,
+				 MSCC_MS_MISC_CONTROL_MC_LATENCY_FIX(bank == MACSEC_INGR ? 57 : 40) |
+				 MSCC_MS_MISC_CONTROL_XFORM_REC_SIZE(bank == MACSEC_INGR ? 1 : 2));
+
+	/* Clear the counters */
+	val = vsc8584_macsec_phy_read(phydev, bank, MSCC_MS_COUNT_CONTROL);
+	val |= MSCC_MS_COUNT_CONTROL_AUTO_CNTR_RESET;
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_COUNT_CONTROL, val);
+
+	/* Enable octet increment mode */
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_PP_CTRL,
+				 MSCC_MS_PP_CTRL_MACSEC_OCTET_INCR_MODE);
+
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_BLOCK_CTX_UPDATE, 0x3);
+
+	val = vsc8584_macsec_phy_read(phydev, bank, MSCC_MS_COUNT_CONTROL);
+	val |= MSCC_MS_COUNT_CONTROL_RESET_ALL;
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_COUNT_CONTROL, val);
+
+	/* Set the MTU */
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_NON_VLAN_MTU_CHECK,
+				 MSCC_MS_NON_VLAN_MTU_CHECK_NV_MTU_COMPARE(32761) |
+				 MSCC_MS_NON_VLAN_MTU_CHECK_NV_MTU_COMP_DROP);
+
+	for (i = 0; i < 8; i++)
+		vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_VLAN_MTU_CHECK(i),
+					 MSCC_MS_VLAN_MTU_CHECK_MTU_COMPARE(32761) |
+					 MSCC_MS_VLAN_MTU_CHECK_MTU_COMP_DROP);
+
+	if (bank == MACSEC_EGR) {
+		val = vsc8584_macsec_phy_read(phydev, bank, MSCC_MS_INTR_CTRL_STATUS);
+		val &= ~MSCC_MS_INTR_CTRL_STATUS_INTR_ENABLE_M;
+		vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_INTR_CTRL_STATUS, val);
+
+		vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_FC_CFG,
+					 MSCC_MS_FC_CFG_FCBUF_ENA |
+					 MSCC_MS_FC_CFG_LOW_THRESH(0x1) |
+					 MSCC_MS_FC_CFG_HIGH_THRESH(0x4) |
+					 MSCC_MS_FC_CFG_LOW_BYTES_VAL(0x4) |
+					 MSCC_MS_FC_CFG_HIGH_BYTES_VAL(0x6));
+	}
+
+	vsc8584_macsec_classification(phydev, bank);
+	vsc8584_macsec_flow_default_action(phydev, bank, false);
+	vsc8584_macsec_integrity_checks(phydev, bank);
+
+	/* Enable the MACsec block */
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_ENA_CFG,
+				 MSCC_MS_ENA_CFG_CLK_ENA |
+				 MSCC_MS_ENA_CFG_MACSEC_ENA |
+				 MSCC_MS_ENA_CFG_MACSEC_SPEED_MODE(0x5));
+}
+
+static void vsc8584_macsec_mac_init(struct phy_device *phydev,
+				    enum macsec_bank bank)
+{
+	u32 val;
+	int i;
+
+	/* Clear host & line stats */
+	for (i = 0; i < 36; i++)
+		vsc8584_macsec_phy_write(phydev, bank, 0x1c + i, 0);
+
+	val = vsc8584_macsec_phy_read(phydev, bank,
+				      MSCC_MAC_PAUSE_CFG_TX_FRAME_CTRL);
+	val &= ~MSCC_MAC_PAUSE_CFG_TX_FRAME_CTRL_PAUSE_MODE_M;
+	val |= MSCC_MAC_PAUSE_CFG_TX_FRAME_CTRL_PAUSE_MODE(2) |
+	       MSCC_MAC_PAUSE_CFG_TX_FRAME_CTRL_PAUSE_VALUE(0xffff);
+	vsc8584_macsec_phy_write(phydev, bank,
+				 MSCC_MAC_PAUSE_CFG_TX_FRAME_CTRL, val);
+
+	val = vsc8584_macsec_phy_read(phydev, bank,
+				      MSCC_MAC_PAUSE_CFG_TX_FRAME_CTRL_2);
+	val |= 0xffff;
+	vsc8584_macsec_phy_write(phydev, bank,
+				 MSCC_MAC_PAUSE_CFG_TX_FRAME_CTRL_2, val);
+
+	val = vsc8584_macsec_phy_read(phydev, bank,
+				      MSCC_MAC_PAUSE_CFG_RX_FRAME_CTRL);
+	if (bank == HOST_MAC)
+		val |= MSCC_MAC_PAUSE_CFG_RX_FRAME_CTRL_PAUSE_TIMER_ENA |
+		       MSCC_MAC_PAUSE_CFG_RX_FRAME_CTRL_PAUSE_FRAME_DROP_ENA;
+	else
+		val |= MSCC_MAC_PAUSE_CFG_RX_FRAME_CTRL_PAUSE_REACT_ENA |
+		       MSCC_MAC_PAUSE_CFG_RX_FRAME_CTRL_PAUSE_FRAME_DROP_ENA |
+		       MSCC_MAC_PAUSE_CFG_RX_FRAME_CTRL_PAUSE_MODE |
+		       MSCC_MAC_PAUSE_CFG_RX_FRAME_CTRL_EARLY_PAUSE_DETECT_ENA;
+	vsc8584_macsec_phy_write(phydev, bank,
+				 MSCC_MAC_PAUSE_CFG_RX_FRAME_CTRL, val);
+
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MAC_CFG_PKTINF_CFG,
+				 MSCC_MAC_CFG_PKTINF_CFG_STRIP_FCS_ENA |
+				 MSCC_MAC_CFG_PKTINF_CFG_INSERT_FCS_ENA |
+				 MSCC_MAC_CFG_PKTINF_CFG_LPI_RELAY_ENA |
+				 MSCC_MAC_CFG_PKTINF_CFG_STRIP_PREAMBLE_ENA |
+				 MSCC_MAC_CFG_PKTINF_CFG_INSERT_PREAMBLE_ENA |
+				 (bank == HOST_MAC ?
+				  MSCC_MAC_CFG_PKTINF_CFG_ENABLE_TX_PADDING : 0));
+
+	val = vsc8584_macsec_phy_read(phydev, bank, MSCC_MAC_CFG_MODE_CFG);
+	val &= ~MSCC_MAC_CFG_MODE_CFG_DISABLE_DIC;
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MAC_CFG_MODE_CFG, val);
+
+	val = vsc8584_macsec_phy_read(phydev, bank, MSCC_MAC_CFG_MAXLEN_CFG);
+	val &= ~MSCC_MAC_CFG_MAXLEN_CFG_MAX_LEN_M;
+	val |= MSCC_MAC_CFG_MAXLEN_CFG_MAX_LEN(10240);
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MAC_CFG_MAXLEN_CFG, val);
+
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MAC_CFG_ADV_CHK_CFG,
+				 MSCC_MAC_CFG_ADV_CHK_CFG_SFD_CHK_ENA |
+				 MSCC_MAC_CFG_ADV_CHK_CFG_PRM_CHK_ENA |
+				 MSCC_MAC_CFG_ADV_CHK_CFG_OOR_ERR_ENA |
+				 MSCC_MAC_CFG_ADV_CHK_CFG_INR_ERR_ENA);
+
+	val = vsc8584_macsec_phy_read(phydev, bank, MSCC_MAC_CFG_LFS_CFG);
+	val &= ~MSCC_MAC_CFG_LFS_CFG_LFS_MODE_ENA;
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MAC_CFG_LFS_CFG, val);
+
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MAC_CFG_ENA_CFG,
+				 MSCC_MAC_CFG_ENA_CFG_RX_CLK_ENA |
+				 MSCC_MAC_CFG_ENA_CFG_TX_CLK_ENA |
+				 MSCC_MAC_CFG_ENA_CFG_RX_ENA |
+				 MSCC_MAC_CFG_ENA_CFG_TX_ENA);
+}
+
+/* Must be called with mdio_lock taken */
+static int vsc8584_macsec_init(struct phy_device *phydev)
+{
+	u32 val;
+
+	vsc8584_macsec_block_init(phydev, MACSEC_INGR);
+	vsc8584_macsec_block_init(phydev, MACSEC_EGR);
+	vsc8584_macsec_mac_init(phydev, HOST_MAC);
+	vsc8584_macsec_mac_init(phydev, LINE_MAC);
+
+	vsc8584_macsec_phy_write(phydev, FC_BUFFER,
+				 MSCC_FCBUF_FC_READ_THRESH_CFG,
+				 MSCC_FCBUF_FC_READ_THRESH_CFG_TX_THRESH(4) |
+				 MSCC_FCBUF_FC_READ_THRESH_CFG_RX_THRESH(5));
+
+	val = vsc8584_macsec_phy_read(phydev, FC_BUFFER, MSCC_FCBUF_MODE_CFG);
+	val |= MSCC_FCBUF_MODE_CFG_PAUSE_GEN_ENA |
+	       MSCC_FCBUF_MODE_CFG_RX_PPM_RATE_ADAPT_ENA |
+	       MSCC_FCBUF_MODE_CFG_TX_PPM_RATE_ADAPT_ENA;
+	vsc8584_macsec_phy_write(phydev, FC_BUFFER, MSCC_FCBUF_MODE_CFG, val);
+
+	vsc8584_macsec_phy_write(phydev, FC_BUFFER, MSCC_FCBUF_PPM_RATE_ADAPT_THRESH_CFG,
+				 MSCC_FCBUF_PPM_RATE_ADAPT_THRESH_CFG_TX_THRESH(8) |
+				 MSCC_FCBUF_PPM_RATE_ADAPT_THRESH_CFG_TX_OFFSET(9));
+
+	val = vsc8584_macsec_phy_read(phydev, FC_BUFFER,
+				      MSCC_FCBUF_TX_DATA_QUEUE_CFG);
+	val &= ~(MSCC_FCBUF_TX_DATA_QUEUE_CFG_START_M |
+		 MSCC_FCBUF_TX_DATA_QUEUE_CFG_END_M);
+	val |= MSCC_FCBUF_TX_DATA_QUEUE_CFG_START(0) |
+		MSCC_FCBUF_TX_DATA_QUEUE_CFG_END(5119);
+	vsc8584_macsec_phy_write(phydev, FC_BUFFER,
+				 MSCC_FCBUF_TX_DATA_QUEUE_CFG, val);
+
+	val = vsc8584_macsec_phy_read(phydev, FC_BUFFER, MSCC_FCBUF_ENA_CFG);
+	val |= MSCC_FCBUF_ENA_CFG_TX_ENA | MSCC_FCBUF_ENA_CFG_RX_ENA;
+	vsc8584_macsec_phy_write(phydev, FC_BUFFER, MSCC_FCBUF_ENA_CFG, val);
+
+	val = vsc8584_macsec_phy_read(phydev, IP_1588,
+				      MSCC_PROC_0_IP_1588_TOP_CFG_STAT_MODE_CTL);
+	val &= ~MSCC_PROC_0_IP_1588_TOP_CFG_STAT_MODE_CTL_PROTOCOL_MODE_M;
+	val |= MSCC_PROC_0_IP_1588_TOP_CFG_STAT_MODE_CTL_PROTOCOL_MODE(4);
+	vsc8584_macsec_phy_write(phydev, IP_1588,
+				 MSCC_PROC_0_IP_1588_TOP_CFG_STAT_MODE_CTL, val);
+
+	return 0;
+}
+#endif /* CONFIG_MACSEC */
+
 /* Check if one PHY has already done the init of the parts common to all PHYs
  * in the Quad PHY package.
  */
@@ -1733,6 +2102,19 @@ static int vsc8584_config_init(struct phy_device *phydev)
 
 	mutex_unlock(&phydev->mdio.bus->mdio_lock);
 
+#if IS_ENABLED(CONFIG_MACSEC)
+	/* MACsec */
+	switch (phydev->phy_id & phydev->drv->phy_id_mask) {
+	case PHY_ID_VSC856X:
+	case PHY_ID_VSC8575:
+	case PHY_ID_VSC8582:
+	case PHY_ID_VSC8584:
+		ret = vsc8584_macsec_init(phydev);
+		if (ret)
+			goto err;
+	}
+#endif
+
 	phy_write(phydev, MSCC_EXT_PAGE_ACCESS, MSCC_PHY_PAGE_STANDARD);
 
 	val = phy_read(phydev, MSCC_PHY_EXT_PHY_CNTL_1);
diff --git a/drivers/net/phy/mscc_fc_buffer.h b/drivers/net/phy/mscc_fc_buffer.h
new file mode 100644
index 000000000000..7e9c0e877895
--- /dev/null
+++ b/drivers/net/phy/mscc_fc_buffer.h
@@ -0,0 +1,64 @@
+/* SPDX-License-Identifier: (GPL-2.0 OR MIT) */
+/*
+ * Microsemi Ocelot Switch driver
+ *
+ * Copyright (C) 2019 Microsemi Corporation
+ */
+
+#ifndef _MSCC_OCELOT_FC_BUFFER_H_
+#define _MSCC_OCELOT_FC_BUFFER_H_
+
+#define MSCC_FCBUF_ENA_CFG					0x00
+#define MSCC_FCBUF_MODE_CFG					0x01
+#define MSCC_FCBUF_PPM_RATE_ADAPT_THRESH_CFG			0x02
+#define MSCC_FCBUF_TX_CTRL_QUEUE_CFG				0x03
+#define MSCC_FCBUF_TX_DATA_QUEUE_CFG				0x04
+#define MSCC_FCBUF_RX_DATA_QUEUE_CFG				0x05
+#define MSCC_FCBUF_TX_BUFF_XON_XOFF_THRESH_CFG			0x06
+#define MSCC_FCBUF_FC_READ_THRESH_CFG				0x07
+#define MSCC_FCBUF_TX_FRM_GAP_COMP				0x08
+
+#define MSCC_FCBUF_ENA_CFG_TX_ENA				BIT(0)
+#define MSCC_FCBUF_ENA_CFG_RX_ENA				BIT(4)
+
+#define MSCC_FCBUF_MODE_CFG_DROP_BEHAVIOUR			BIT(4)
+#define MSCC_FCBUF_MODE_CFG_PAUSE_REACT_ENA			BIT(8)
+#define MSCC_FCBUF_MODE_CFG_RX_PPM_RATE_ADAPT_ENA		BIT(12)
+#define MSCC_FCBUF_MODE_CFG_TX_PPM_RATE_ADAPT_ENA		BIT(16)
+#define MSCC_FCBUF_MODE_CFG_TX_CTRL_QUEUE_ENA			BIT(20)
+#define MSCC_FCBUF_MODE_CFG_PAUSE_GEN_ENA			BIT(24)
+#define MSCC_FCBUF_MODE_CFG_INCLUDE_PAUSE_RCVD_IN_PAUSE_GEN	BIT(28)
+
+#define MSCC_FCBUF_PPM_RATE_ADAPT_THRESH_CFG_TX_THRESH(x)	(x)
+#define MSCC_FCBUF_PPM_RATE_ADAPT_THRESH_CFG_TX_THRESH_M	GENMASK(15, 0)
+#define MSCC_FCBUF_PPM_RATE_ADAPT_THRESH_CFG_TX_OFFSET(x)	((x) << 16)
+#define MSCC_FCBUF_PPM_RATE_ADAPT_THRESH_CFG_TX_OFFSET_M	GENMASK(19, 16)
+#define MSCC_FCBUF_PPM_RATE_ADAPT_THRESH_CFG_RX_THRESH(x)	((x) << 20)
+#define MSCC_FCBUF_PPM_RATE_ADAPT_THRESH_CFG_RX_THRESH_M	GENMASK(31, 20)
+
+#define MSCC_FCBUF_TX_CTRL_QUEUE_CFG_START(x)			(x)
+#define MSCC_FCBUF_TX_CTRL_QUEUE_CFG_START_M			GENMASK(15, 0)
+#define MSCC_FCBUF_TX_CTRL_QUEUE_CFG_END(x)			((x) << 16)
+#define MSCC_FCBUF_TX_CTRL_QUEUE_CFG_END_M			GENMASK(31, 16)
+
+#define MSCC_FCBUF_TX_DATA_QUEUE_CFG_START(x)			(x)
+#define MSCC_FCBUF_TX_DATA_QUEUE_CFG_START_M			GENMASK(15, 0)
+#define MSCC_FCBUF_TX_DATA_QUEUE_CFG_END(x)			((x) << 16)
+#define MSCC_FCBUF_TX_DATA_QUEUE_CFG_END_M			GENMASK(31, 16)
+
+#define MSCC_FCBUF_RX_DATA_QUEUE_CFG_START(x)			(x)
+#define MSCC_FCBUF_RX_DATA_QUEUE_CFG_START_M			GENMASK(15, 0)
+#define MSCC_FCBUF_RX_DATA_QUEUE_CFG_END(x)			((x) << 16)
+#define MSCC_FCBUF_RX_DATA_QUEUE_CFG_END_M			GENMASK(31, 16)
+
+#define MSCC_FCBUF_TX_BUFF_XON_XOFF_THRESH_CFG_XOFF_THRESH(x)	(x)
+#define MSCC_FCBUF_TX_BUFF_XON_XOFF_THRESH_CFG_XOFF_THRESH_M	GENMASK(15, 0)
+#define MSCC_FCBUF_TX_BUFF_XON_XOFF_THRESH_CFG_XON_THRESH(x)	((x) << 16)
+#define MSCC_FCBUF_TX_BUFF_XON_XOFF_THRESH_CFG_XON_THRESH_M	GENMASK(31, 16)
+
+#define MSCC_FCBUF_FC_READ_THRESH_CFG_TX_THRESH(x)		(x)
+#define MSCC_FCBUF_FC_READ_THRESH_CFG_TX_THRESH_M		GENMASK(15, 0)
+#define MSCC_FCBUF_FC_READ_THRESH_CFG_RX_THRESH(x)		((x) << 16)
+#define MSCC_FCBUF_FC_READ_THRESH_CFG_RX_THRESH_M		GENMASK(31, 16)
+
+#endif
diff --git a/drivers/net/phy/mscc_mac.h b/drivers/net/phy/mscc_mac.h
new file mode 100644
index 000000000000..9420ee5175a6
--- /dev/null
+++ b/drivers/net/phy/mscc_mac.h
@@ -0,0 +1,159 @@
+/* SPDX-License-Identifier: (GPL-2.0 OR MIT) */
+/*
+ * Microsemi Ocelot Switch driver
+ *
+ * Copyright (c) 2017 Microsemi Corporation
+ */
+
+#ifndef _MSCC_OCELOT_LINE_MAC_H_
+#define _MSCC_OCELOT_LINE_MAC_H_
+
+#define MSCC_MAC_CFG_ENA_CFG					0x00
+#define MSCC_MAC_CFG_MODE_CFG					0x01
+#define MSCC_MAC_CFG_MAXLEN_CFG					0x02
+#define MSCC_MAC_CFG_NUM_TAGS_CFG				0x03
+#define MSCC_MAC_CFG_TAGS_CFG					0x04
+#define MSCC_MAC_CFG_ADV_CHK_CFG				0x07
+#define MSCC_MAC_CFG_LFS_CFG					0x08
+#define MSCC_MAC_CFG_LB_CFG					0x09
+#define MSCC_MAC_CFG_PKTINF_CFG					0x0a
+#define MSCC_MAC_PAUSE_CFG_TX_FRAME_CTRL			0x0b
+#define MSCC_MAC_PAUSE_CFG_TX_FRAME_CTRL_2			0x0c
+#define MSCC_MAC_PAUSE_CFG_RX_FRAME_CTRL			0x0d
+#define MSCC_MAC_PAUSE_CFG_STATE				0x0e
+#define MSCC_MAC_PAUSE_CFG_MAC_ADDRESS_LSB			0x0f
+#define MSCC_MAC_PAUSE_CFG_MAC_ADDRESS_MSB			0x10
+#define MSCC_MAC_STATUS_RX_LANE_STICKY_0			0x11
+#define MSCC_MAC_STATUS_RX_LANE_STICKY_1			0x12
+#define MSCC_MAC_STATUS_TX_MONITOR_STICKY			0x13
+#define MSCC_MAC_STATUS_TX_MONITOR_STICKY_MASK			0x14
+#define MSCC_MAC_STATUS_STICKY					0x15
+#define MSCC_MAC_STATUS_STICKY_MASK				0x16
+#define MSCC_MAC_STATS_32BIT_RX_HIH_CKSM_ERR_CNT		0x17
+#define MSCC_MAC_STATS_32BIT_RX_XGMII_PROT_ERR_CNT		0x18
+#define MSCC_MAC_STATS_32BIT_RX_SYMBOL_ERR_CNT			0x19
+#define MSCC_MAC_STATS_32BIT_RX_PAUSE_CNT			0x1a
+#define MSCC_MAC_STATS_32BIT_RX_UNSUP_OPCODE_CNT		0x1b
+#define MSCC_MAC_STATS_32BIT_RX_UC_CNT				0x1c
+#define MSCC_MAC_STATS_32BIT_RX_MC_CNT				0x1d
+#define MSCC_MAC_STATS_32BIT_RX_BC_CNT				0x1e
+#define MSCC_MAC_STATS_32BIT_RX_CRC_ERR_CNT			0x1f
+#define MSCC_MAC_STATS_32BIT_RX_UNDERSIZE_CNT			0x20
+#define MSCC_MAC_STATS_32BIT_RX_FRAGMENTS_CNT			0x21
+#define MSCC_MAC_STATS_32BIT_RX_IN_RANGE_LEN_ERR_CNT		0x22
+#define MSCC_MAC_STATS_32BIT_RX_OUT_OF_RANGE_LEN_ERR_CNT	0x23
+#define MSCC_MAC_STATS_32BIT_RX_OVERSIZE_CNT			0x24
+#define MSCC_MAC_STATS_32BIT_RX_JABBERS_CNT			0x25
+#define MSCC_MAC_STATS_32BIT_RX_SIZE64_CNT			0x26
+#define MSCC_MAC_STATS_32BIT_RX_SIZE65TO127_CNT			0x27
+#define MSCC_MAC_STATS_32BIT_RX_SIZE128TO255_CNT		0x28
+#define MSCC_MAC_STATS_32BIT_RX_SIZE256TO511_CNT		0x29
+#define MSCC_MAC_STATS_32BIT_RX_SIZE512TO1023_CNT		0x2a
+#define MSCC_MAC_STATS_32BIT_RX_SIZE1024TO1518_CNT		0x2b
+#define MSCC_MAC_STATS_32BIT_RX_SIZE1519TOMAX_CNT		0x2c
+#define MSCC_MAC_STATS_32BIT_RX_IPG_SHRINK_CNT			0x2d
+#define MSCC_MAC_STATS_32BIT_TX_PAUSE_CNT			0x2e
+#define MSCC_MAC_STATS_32BIT_TX_UC_CNT				0x2f
+#define MSCC_MAC_STATS_32BIT_TX_MC_CNT				0x30
+#define MSCC_MAC_STATS_32BIT_TX_BC_CNT				0x31
+#define MSCC_MAC_STATS_32BIT_TX_SIZE64_CNT			0x32
+#define MSCC_MAC_STATS_32BIT_TX_SIZE65TO127_CNT			0x33
+#define MSCC_MAC_STATS_32BIT_TX_SIZE128TO255_CNT		0x34
+#define MSCC_MAC_STATS_32BIT_TX_SIZE256TO511_CNT		0x35
+#define MSCC_MAC_STATS_32BIT_TX_SIZE512TO1023_CNT		0x36
+#define MSCC_MAC_STATS_32BIT_TX_SIZE1024TO1518_CNT		0x37
+#define MSCC_MAC_STATS_32BIT_TX_SIZE1519TOMAX_CNT		0x38
+#define MSCC_MAC_STATS_40BIT_RX_BAD_BYTES_CNT			0x39
+#define MSCC_MAC_STATS_40BIT_RX_BAD_BYTES_MSB_CNT		0x3a
+#define MSCC_MAC_STATS_40BIT_RX_OK_BYTES_CNT			0x3b
+#define MSCC_MAC_STATS_40BIT_RX_OK_BYTES_MSB_CNT		0x3c
+#define MSCC_MAC_STATS_40BIT_RX_IN_BYTES_CNT			0x3d
+#define MSCC_MAC_STATS_40BIT_RX_IN_BYTES_MSB_CNT		0x3e
+#define MSCC_MAC_STATS_40BIT_TX_OK_BYTES_CNT			0x3f
+#define MSCC_MAC_STATS_40BIT_TX_OK_BYTES_MSB_CNT		0x40
+#define MSCC_MAC_STATS_40BIT_TX_OUT_BYTES_CNT			0x41
+#define MSCC_MAC_STATS_40BIT_TX_OUT_BYTES_MSB_CNT		0x42
+
+#define MSCC_MAC_CFG_ENA_CFG_RX_CLK_ENA				BIT(0)
+#define MSCC_MAC_CFG_ENA_CFG_TX_CLK_ENA				BIT(4)
+#define MSCC_MAC_CFG_ENA_CFG_RX_SW_RST				BIT(8)
+#define MSCC_MAC_CFG_ENA_CFG_TX_SW_RST				BIT(12)
+#define MSCC_MAC_CFG_ENA_CFG_RX_ENA				BIT(16)
+#define MSCC_MAC_CFG_ENA_CFG_TX_ENA				BIT(20)
+
+#define MSCC_MAC_CFG_MODE_CFG_FORCE_CW_UPDATE_INTERVAL(x)	((x) << 20)
+#define MSCC_MAC_CFG_MODE_CFG_FORCE_CW_UPDATE_INTERVAL_M	GENMASK(29, 20)
+#define MSCC_MAC_CFG_MODE_CFG_FORCE_CW_UPDATE			BIT(16)
+#define MSCC_MAC_CFG_MODE_CFG_TUNNEL_PAUSE_FRAMES		BIT(14)
+#define MSCC_MAC_CFG_MODE_CFG_MAC_PREAMBLE_CFG(x)		((x) << 10)
+#define MSCC_MAC_CFG_MODE_CFG_MAC_PREAMBLE_CFG_M		GENMASK(12, 10)
+#define MSCC_MAC_CFG_MODE_CFG_MAC_IPG_CFG			BIT(6)
+#define MSCC_MAC_CFG_MODE_CFG_XGMII_GEN_MODE_ENA		BIT(4)
+#define MSCC_MAC_CFG_MODE_CFG_HIH_CRC_CHECK			BIT(2)
+#define MSCC_MAC_CFG_MODE_CFG_UNDERSIZED_FRAME_DROP_DIS		BIT(1)
+#define MSCC_MAC_CFG_MODE_CFG_DISABLE_DIC			BIT(0)
+
+#define MSCC_MAC_CFG_MAXLEN_CFG_MAX_LEN_TAG_CHK			BIT(16)
+#define MSCC_MAC_CFG_MAXLEN_CFG_MAX_LEN(x)			(x)
+#define MSCC_MAC_CFG_MAXLEN_CFG_MAX_LEN_M			GENMASK(15, 0)
+
+#define MSCC_MAC_CFG_TAGS_CFG_RSZ				0x4
+#define MSCC_MAC_CFG_TAGS_CFG_TAG_ID(x)				((x) << 16)
+#define MSCC_MAC_CFG_TAGS_CFG_TAG_ID_M				GENMASK(31, 16)
+#define MSCC_MAC_CFG_TAGS_CFG_TAG_ENA				BIT(4)
+
+#define MSCC_MAC_CFG_ADV_CHK_CFG_EXT_EOP_CHK_ENA		BIT(24)
+#define MSCC_MAC_CFG_ADV_CHK_CFG_EXT_SOP_CHK_ENA		BIT(20)
+#define MSCC_MAC_CFG_ADV_CHK_CFG_SFD_CHK_ENA			BIT(16)
+#define MSCC_MAC_CFG_ADV_CHK_CFG_PRM_SHK_CHK_DIS		BIT(12)
+#define MSCC_MAC_CFG_ADV_CHK_CFG_PRM_CHK_ENA			BIT(8)
+#define MSCC_MAC_CFG_ADV_CHK_CFG_OOR_ERR_ENA			BIT(4)
+#define MSCC_MAC_CFG_ADV_CHK_CFG_INR_ERR_ENA			BIT(0)
+
+#define MSCC_MAC_CFG_LFS_CFG_LFS_INH_TX				BIT(8)
+#define MSCC_MAC_CFG_LFS_CFG_LFS_DIS_TX				BIT(4)
+#define MSCC_MAC_CFG_LFS_CFG_LFS_UNIDIR_ENA			BIT(3)
+#define MSCC_MAC_CFG_LFS_CFG_USE_LEADING_EDGE_DETECT		BIT(2)
+#define MSCC_MAC_CFG_LFS_CFG_SPURIOUS_Q_DIS			BIT(1)
+#define MSCC_MAC_CFG_LFS_CFG_LFS_MODE_ENA			BIT(0)
+
+#define MSCC_MAC_CFG_LB_CFG_XGMII_HOST_LB_ENA			BIT(4)
+#define MSCC_MAC_CFG_LB_CFG_XGMII_PHY_LB_ENA			BIT(0)
+
+#define MSCC_MAC_CFG_PKTINF_CFG_STRIP_FCS_ENA			BIT(0)
+#define MSCC_MAC_CFG_PKTINF_CFG_INSERT_FCS_ENA			BIT(4)
+#define MSCC_MAC_CFG_PKTINF_CFG_STRIP_PREAMBLE_ENA		BIT(8)
+#define MSCC_MAC_CFG_PKTINF_CFG_INSERT_PREAMBLE_ENA		BIT(12)
+#define MSCC_MAC_CFG_PKTINF_CFG_LPI_RELAY_ENA			BIT(16)
+#define MSCC_MAC_CFG_PKTINF_CFG_LF_RELAY_ENA			BIT(20)
+#define MSCC_MAC_CFG_PKTINF_CFG_RF_RELAY_ENA			BIT(24)
+#define MSCC_MAC_CFG_PKTINF_CFG_ENABLE_TX_PADDING		BIT(25)
+#define MSCC_MAC_CFG_PKTINF_CFG_ENABLE_RX_PADDING		BIT(26)
+#define MSCC_MAC_CFG_PKTINF_CFG_ENABLE_4BYTE_PREAMBLE		BIT(27)
+#define MSCC_MAC_CFG_PKTINF_CFG_MACSEC_BYPASS_NUM_PTP_STALL_CLKS(x)	((x) << 28)
+#define MSCC_MAC_CFG_PKTINF_CFG_MACSEC_BYPASS_NUM_PTP_STALL_CLKS_M	GENMASK(30, 28)
+
+#define MSCC_MAC_PAUSE_CFG_TX_FRAME_CTRL_PAUSE_VALUE(x)		((x) << 16)
+#define MSCC_MAC_PAUSE_CFG_TX_FRAME_CTRL_PAUSE_VALUE_M		GENMASK(31, 16)
+#define MSCC_MAC_PAUSE_CFG_TX_FRAME_CTRL_WAIT_FOR_LPI_LOW	BIT(12)
+#define MSCC_MAC_PAUSE_CFG_TX_FRAME_CTRL_USE_PAUSE_STALL_ENA	BIT(8)
+#define MSCC_MAC_PAUSE_CFG_TX_FRAME_CTRL_PAUSE_REPL_MODE	BIT(4)
+#define MSCC_MAC_PAUSE_CFG_TX_FRAME_CTRL_PAUSE_FRC_FRAME	BIT(2)
+#define MSCC_MAC_PAUSE_CFG_TX_FRAME_CTRL_PAUSE_MODE(x)		(x)
+#define MSCC_MAC_PAUSE_CFG_TX_FRAME_CTRL_PAUSE_MODE_M		GENMASK(1, 0)
+
+#define MSCC_MAC_PAUSE_CFG_RX_FRAME_CTRL_EARLY_PAUSE_DETECT_ENA	BIT(16)
+#define MSCC_MAC_PAUSE_CFG_RX_FRAME_CTRL_PRE_CRC_MODE		BIT(20)
+#define MSCC_MAC_PAUSE_CFG_RX_FRAME_CTRL_PAUSE_TIMER_ENA	BIT(12)
+#define MSCC_MAC_PAUSE_CFG_RX_FRAME_CTRL_PAUSE_REACT_ENA	BIT(8)
+#define MSCC_MAC_PAUSE_CFG_RX_FRAME_CTRL_PAUSE_FRAME_DROP_ENA	BIT(4)
+#define MSCC_MAC_PAUSE_CFG_RX_FRAME_CTRL_PAUSE_MODE		BIT(0)
+
+#define MSCC_MAC_PAUSE_CFG_STATE_PAUSE_STATE			BIT(0)
+#define MSCC_MAC_PAUSE_CFG_STATE_MAC_TX_PAUSE_GEN		BIT(4)
+
+#define MSCC_PROC_0_IP_1588_TOP_CFG_STAT_MODE_CTL			0x2
+#define MSCC_PROC_0_IP_1588_TOP_CFG_STAT_MODE_CTL_PROTOCOL_MODE(x)	(x)
+#define MSCC_PROC_0_IP_1588_TOP_CFG_STAT_MODE_CTL_PROTOCOL_MODE_M	GENMASK(2, 0)
+
+#endif /* _MSCC_OCELOT_LINE_MAC_H_ */
diff --git a/drivers/net/phy/mscc_macsec.h b/drivers/net/phy/mscc_macsec.h
new file mode 100644
index 000000000000..0d108da28dad
--- /dev/null
+++ b/drivers/net/phy/mscc_macsec.h
@@ -0,0 +1,260 @@
+/* SPDX-License-Identifier: (GPL-2.0 OR MIT) */
+/*
+ * Microsemi Ocelot Switch driver
+ *
+ * Copyright (c) 2018 Microsemi Corporation
+ */
+
+#ifndef _MSCC_OCELOT_MACSEC_H_
+#define _MSCC_OCELOT_MACSEC_H_
+
+#define CONTROL_TYPE_EGRESS		0x6
+#define CONTROL_TYPE_INGRESS		0xf
+#define CONTROL_IV0			BIT(5)
+#define CONTROL_IV1			BIT(6)
+#define CONTROL_IV2			BIT(7)
+#define CONTROL_UPDATE_SEQ		BIT(13)
+#define CONTROL_IV_IN_SEQ		BIT(14)
+#define CONTROL_ENCRYPT_AUTH		BIT(15)
+#define CONTROL_KEY_IN_CTX		BIT(16)
+#define CONTROL_CRYPTO_ALG(x)		((x) << 17)
+#define     CTRYPTO_ALG_AES_CTR_128	0x5
+#define     CTRYPTO_ALG_AES_CTR_192	0x6
+#define     CTRYPTO_ALG_AES_CTR_256	0x7
+#define CONTROL_DIGEST_TYPE(x)		((x) << 21)
+#define CONTROL_AUTH_ALG(x)		((x) << 23)
+#define     AUTH_ALG_AES_GHAS		0x4
+#define CONTROL_AN(x)			((x) << 26)
+#define CONTROL_SEQ_TYPE(x)		((x) << 28)
+#define CONTROL_SEQ_MASK		BIT(30)
+#define CONTROL_CONTEXT_ID		BIT(31)
+
+enum mscc_macsec_destination_ports {
+	MSCC_MS_PORT_COMMON		= 0,
+	MSCC_MS_PORT_RSVD		= 1,
+	MSCC_MS_PORT_CONTROLLED		= 2,
+	MSCC_MS_PORT_UNCONTROLLED	= 3,
+};
+
+enum mscc_macsec_drop_actions {
+	MSCC_MS_ACTION_BYPASS_CRC	= 0,
+	MSCC_MS_ACTION_BYPASS_BAD	= 1,
+	MSCC_MS_ACTION_DROP		= 2,
+	MSCC_MS_ACTION_BYPASS		= 3,
+};
+
+enum mscc_macsec_flow_types {
+	MSCC_MS_FLOW_BYPASS		= 0,
+	MSCC_MS_FLOW_DROP		= 1,
+	MSCC_MS_FLOW_INGRESS		= 2,
+	MSCC_MS_FLOW_EGRESS		= 3,
+};
+
+enum mscc_macsec_validate_levels {
+	MSCC_MS_VALIDATE_DISABLED	= 0,
+	MSCC_MS_VALIDATE_CHECK		= 1,
+	MSCC_MS_VALIDATE_STRICT		= 2,
+};
+
+#define MSCC_MS_XFORM_REC(x, y)		(((x) << 5) + (y))
+#define MSCC_MS_ENA_CFG			0x800
+#define MSCC_MS_FC_CFG			0x804
+#define MSCC_MS_SAM_MISC_MATCH(x)	(0x1004 + ((x) << 4))
+#define MSCC_MS_SAM_MATCH_SCI_LO(x)	(0x1005 + ((x) << 4))
+#define MSCC_MS_SAM_MATCH_SCI_HI(x)	(0x1006 + ((x) << 4))
+#define MSCC_MS_SAM_MASK(x)		(0x1007 + ((x) << 4))
+#define MSCC_MS_SAM_ENTRY_SET1		0x1808
+#define MSCC_MS_SAM_ENTRY_CLEAR1	0x180c
+#define MSCC_MS_SAM_FLOW_CTRL(x)	(0x1c00 + (x))
+#define MSCC_MS_SAM_CP_TAG		0x1e40
+#define MSCC_MS_SAM_NM_FLOW_NCP		0x1e51
+#define MSCC_MS_SAM_NM_FLOW_CP		0x1e52
+#define MSCC_MS_MISC_CONTROL		0x1e5f
+#define MSCC_MS_COUNT_CONTROL		0x3204
+#define MSCC_MS_PARAMS2_IG_CC_CONTROL	0x3a10
+#define MSCC_MS_PARAMS2_IG_CP_TAG	0x3a14
+#define MSCC_MS_VLAN_MTU_CHECK(x)	(0x3c40 + (x))
+#define MSCC_MS_NON_VLAN_MTU_CHECK	0x3c48
+#define MSCC_MS_PP_CTRL			0x3c4b
+#define MSCC_MS_STATUS_CONTEXT_CTRL	0x3d02
+#define MSCC_MS_INTR_CTRL_STATUS	0x3d04
+#define MSCC_MS_BLOCK_CTX_UPDATE	0x3d0c
+
+/* MACSEC_ENA_CFG */
+#define MSCC_MS_ENA_CFG_CLK_ENA				BIT(0)
+#define MSCC_MS_ENA_CFG_SW_RST				BIT(1)
+#define MSCC_MS_ENA_CFG_MACSEC_BYPASS_ENA		BIT(8)
+#define MSCC_MS_ENA_CFG_MACSEC_ENA			BIT(9)
+#define MSCC_MS_ENA_CFG_MACSEC_SPEED_MODE(x)		((x) << 10)
+#define MSCC_MS_ENA_CFG_MACSEC_SPEED_MODE_M		GENMASK(12, 10)
+
+/* MACSEC_FC_CFG */
+#define MSCC_MS_FC_CFG_FCBUF_ENA			BIT(0)
+#define MSCC_MS_FC_CFG_USE_PKT_EXPANSION_INDICATION	BIT(1)
+#define MSCC_MS_FC_CFG_LOW_THRESH(x)			((x) << 4)
+#define MSCC_MS_FC_CFG_LOW_THRESH_M			GENMASK(7, 4)
+#define MSCC_MS_FC_CFG_HIGH_THRESH(x)			((x) << 8)
+#define MSCC_MS_FC_CFG_HIGH_THRESH_M			GENMASK(11, 8)
+#define MSCC_MS_FC_CFG_LOW_BYTES_VAL(x)			((x) << 12)
+#define MSCC_MS_FC_CFG_LOW_BYTES_VAL_M			GENMASK(14, 12)
+#define MSCC_MS_FC_CFG_HIGH_BYTES_VAL(x)		((x) << 16)
+#define MSCC_MS_FC_CFG_HIGH_BYTES_VAL_M			GENMASK(18, 16)
+
+/* MSCC_MS_SAM_MAC_SA_MATCH_HI */
+#define MSCC_MS_SAM_MAC_SA_MATCH_HI_ETYPE(x)		((x) << 16)
+#define MSCC_MS_SAM_MAC_SA_MATCH_HI_ETYPE_M		GENMASK(31, 16)
+
+/* MACSEC_SAM_MISC_MATCH */
+#define MSCC_MS_SAM_MISC_MATCH_VLAN_VALID		BIT(0)
+#define MSCC_MS_SAM_MISC_MATCH_QINQ_FOUND		BIT(1)
+#define MSCC_MS_SAM_MISC_MATCH_STAG_VALID		BIT(2)
+#define MSCC_MS_SAM_MISC_MATCH_QTAG_VALID		BIT(3)
+#define MSCC_MS_SAM_MISC_MATCH_VLAN_UP(x)		((x) << 4)
+#define MSCC_MS_SAM_MISC_MATCH_VLAN_UP_M		GENMASK(6, 4)
+#define MSCC_MS_SAM_MISC_MATCH_CONTROL_PACKET		BIT(7)
+#define MSCC_MS_SAM_MISC_MATCH_UNTAGGED			BIT(8)
+#define MSCC_MS_SAM_MISC_MATCH_TAGGED			BIT(9)
+#define MSCC_MS_SAM_MISC_MATCH_BAD_TAG			BIT(10)
+#define MSCC_MS_SAM_MISC_MATCH_KAY_TAG			BIT(11)
+#define MSCC_MS_SAM_MISC_MATCH_SOURCE_PORT(x)		((x) << 12)
+#define MSCC_MS_SAM_MISC_MATCH_SOURCE_PORT_M		GENMASK(13, 12)
+#define MSCC_MS_SAM_MISC_MATCH_PRIORITY(x)		((x) << 16)
+#define MSCC_MS_SAM_MISC_MATCH_PRIORITY_M		GENMASK(19, 16)
+#define MSCC_MS_SAM_MISC_MATCH_AN(x)			((x) << 24)
+#define MSCC_MS_SAM_MISC_MATCH_TCI(x)			((x) << 26)
+
+/* MACSEC_SAM_MASK */
+#define MSCC_MS_SAM_MASK_MAC_SA_MASK(x)			(x)
+#define MSCC_MS_SAM_MASK_MAC_SA_MASK_M			GENMASK(5, 0)
+#define MSCC_MS_SAM_MASK_MAC_DA_MASK(x)			((x) << 6)
+#define MSCC_MS_SAM_MASK_MAC_DA_MASK_M			GENMASK(11, 6)
+#define MSCC_MS_SAM_MASK_MAC_ETYPE_MASK			BIT(12)
+#define MSCC_MS_SAM_MASK_VLAN_VLD_MASK			BIT(13)
+#define MSCC_MS_SAM_MASK_QINQ_FOUND_MASK		BIT(14)
+#define MSCC_MS_SAM_MASK_STAG_VLD_MASK			BIT(15)
+#define MSCC_MS_SAM_MASK_QTAG_VLD_MASK			BIT(16)
+#define MSCC_MS_SAM_MASK_VLAN_UP_MASK			BIT(17)
+#define MSCC_MS_SAM_MASK_VLAN_ID_MASK			BIT(18)
+#define MSCC_MS_SAM_MASK_SOURCE_PORT_MASK		BIT(19)
+#define MSCC_MS_SAM_MASK_CTL_PACKET_MASK		BIT(20)
+#define MSCC_MS_SAM_MASK_VLAN_UP_INNER_MASK		BIT(21)
+#define MSCC_MS_SAM_MASK_VLAN_ID_INNER_MASK		BIT(22)
+#define MSCC_MS_SAM_MASK_SCI_MASK			BIT(23)
+#define MSCC_MS_SAM_MASK_AN_MASK(x)			((x) << 24)
+#define MSCC_MS_SAM_MASK_TCI_MASK(x)			((x) << 26)
+
+/* MACSEC_SAM_FLOW_CTRL_EGR */
+#define MSCC_MS_SAM_FLOW_CTRL_FLOW_TYPE(x)		(x)
+#define MSCC_MS_SAM_FLOW_CTRL_FLOW_TYPE_M		GENMASK(1, 0)
+#define MSCC_MS_SAM_FLOW_CTRL_DEST_PORT(x)		((x) << 2)
+#define MSCC_MS_SAM_FLOW_CTRL_DEST_PORT_M		GENMASK(3, 2)
+#define MSCC_MS_SAM_FLOW_CTRL_RESV_4			BIT(4)
+#define MSCC_MS_SAM_FLOW_CTRL_FLOW_CRYPT_AUTH		BIT(5)
+#define MSCC_MS_SAM_FLOW_CTRL_DROP_ACTION(x)		((x) << 6)
+#define MSCC_MS_SAM_FLOW_CTRL_DROP_ACTION_M		GENMASK(7, 6)
+#define MSCC_MS_SAM_FLOW_CTRL_RESV_15_TO_8(x)		((x) << 8)
+#define MSCC_MS_SAM_FLOW_CTRL_RESV_15_TO_8_M		GENMASK(15, 8)
+#define MSCC_MS_SAM_FLOW_CTRL_PROTECT_FRAME		BIT(16)
+#define MSCC_MS_SAM_FLOW_CTRL_REPLAY_PROTECT		BIT(16)
+#define MSCC_MS_SAM_FLOW_CTRL_SA_IN_USE			BIT(17)
+#define MSCC_MS_SAM_FLOW_CTRL_INCLUDE_SCI		BIT(18)
+#define MSCC_MS_SAM_FLOW_CTRL_USE_ES			BIT(19)
+#define MSCC_MS_SAM_FLOW_CTRL_USE_SCB			BIT(20)
+#define MSCC_MS_SAM_FLOW_CTRL_VALIDATE_FRAMES(x)	((x) << 19)
+#define MSCC_MS_SAM_FLOW_CTRL_TAG_BYPASS_SIZE(x)	((x) << 21)
+#define MSCC_MS_SAM_FLOW_CTRL_TAG_BYPASS_SIZE_M		GENMASK(22, 21)
+#define MSCC_MS_SAM_FLOW_CTRL_RESV_23			BIT(23)
+#define MSCC_MS_SAM_FLOW_CTRL_CONFIDENTIALITY_OFFSET(x)	((x) << 24)
+#define MSCC_MS_SAM_FLOW_CTRL_CONFIDENTIALITY_OFFSET_M	GENMASK(30, 24)
+#define MSCC_MS_SAM_FLOW_CTRL_CONF_PROTECT		BIT(31)
+
+/* MACSEC_SAM_CP_TAG */
+#define MSCC_MS_SAM_CP_TAG_MAP_TBL(x)			(x)
+#define MSCC_MS_SAM_CP_TAG_MAP_TBL_M			GENMASK(23, 0)
+#define MSCC_MS_SAM_CP_TAG_DEF_UP(x)			((x) << 24)
+#define MSCC_MS_SAM_CP_TAG_DEF_UP_M			GENMASK(26, 24)
+#define MSCC_MS_SAM_CP_TAG_STAG_UP_EN			BIT(27)
+#define MSCC_MS_SAM_CP_TAG_QTAG_UP_EN			BIT(28)
+#define MSCC_MS_SAM_CP_TAG_PARSE_QINQ			BIT(29)
+#define MSCC_MS_SAM_CP_TAG_PARSE_STAG			BIT(30)
+#define MSCC_MS_SAM_CP_TAG_PARSE_QTAG			BIT(31)
+
+/* MACSEC_SAM_NM_FLOW_NCP */
+#define MSCC_MS_SAM_NM_FLOW_NCP_UNTAGGED_FLOW_TYPE(x)	(x)
+#define MSCC_MS_SAM_NM_FLOW_NCP_UNTAGGED_DEST_PORT(x)	((x) << 2)
+#define MSCC_MS_SAM_NM_FLOW_NCP_UNTAGGED_DROP_ACTION(x)	((x) << 6)
+#define MSCC_MS_SAM_NM_FLOW_NCP_TAGGED_FLOW_TYPE(x)	((x) << 8)
+#define MSCC_MS_SAM_NM_FLOW_NCP_TAGGED_DEST_PORT(x)	((x) << 10)
+#define MSCC_MS_SAM_NM_FLOW_NCP_TAGGED_DROP_ACTION(x)	((x) << 14)
+#define MSCC_MS_SAM_NM_FLOW_NCP_BADTAG_FLOW_TYPE(x)	((x) << 16)
+#define MSCC_MS_SAM_NM_FLOW_NCP_BADTAG_DEST_PORT(x)	((x) << 18)
+#define MSCC_MS_SAM_NM_FLOW_NCP_BADTAG_DROP_ACTION(x)	((x) << 22)
+#define MSCC_MS_SAM_NM_FLOW_NCP_KAY_FLOW_TYPE(x)	((x) << 24)
+#define MSCC_MS_SAM_NM_FLOW_NCP_KAY_DEST_PORT(x)	((x) << 26)
+#define MSCC_MS_SAM_NM_FLOW_NCP_KAY_DROP_ACTION(x)	((x) << 30)
+
+/* MACSEC_SAM_NM_FLOW_CP */
+#define MSCC_MS_SAM_NM_FLOW_CP_UNTAGGED_FLOW_TYPE(x)	(x)
+#define MSCC_MS_SAM_NM_FLOW_CP_UNTAGGED_DEST_PORT(x)	((x) << 2)
+#define MSCC_MS_SAM_NM_FLOW_CP_UNTAGGED_DROP_ACTION(x)	((x) << 6)
+#define MSCC_MS_SAM_NM_FLOW_CP_TAGGED_FLOW_TYPE(x)	((x) << 8)
+#define MSCC_MS_SAM_NM_FLOW_CP_TAGGED_DEST_PORT(x)	((x) << 10)
+#define MSCC_MS_SAM_NM_FLOW_CP_TAGGED_DROP_ACTION(x)	((x) << 14)
+#define MSCC_MS_SAM_NM_FLOW_CP_BADTAG_FLOW_TYPE(x)	((x) << 16)
+#define MSCC_MS_SAM_NM_FLOW_CP_BADTAG_DEST_PORT(x)	((x) << 18)
+#define MSCC_MS_SAM_NM_FLOW_CP_BADTAG_DROP_ACTION(x)	((x) << 22)
+#define MSCC_MS_SAM_NM_FLOW_CP_KAY_FLOW_TYPE(x)		((x) << 24)
+#define MSCC_MS_SAM_NM_FLOW_CP_KAY_DEST_PORT(x)		((x) << 26)
+#define MSCC_MS_SAM_NM_FLOW_CP_KAY_DROP_ACTION(x)	((x) << 30)
+
+/* MACSEC_MISC_CONTROL */
+#define MSCC_MS_MISC_CONTROL_MC_LATENCY_FIX(x)		(x)
+#define MSCC_MS_MISC_CONTROL_MC_LATENCY_FIX_M		GENMASK(5, 0)
+#define MSCC_MS_MISC_CONTROL_STATIC_BYPASS		BIT(8)
+#define MSCC_MS_MISC_CONTROL_NM_MACSEC_EN		BIT(9)
+#define MSCC_MS_MISC_CONTROL_VALIDATE_FRAMES(x)		((x) << 10)
+#define MSCC_MS_MISC_CONTROL_VALIDATE_FRAMES_M		GENMASK(11, 10)
+#define MSCC_MS_MISC_CONTROL_XFORM_REC_SIZE(x)		((x) << 24)
+#define MSCC_MS_MISC_CONTROL_XFORM_REC_SIZE_M		GENMASK(25, 24)
+
+/* MACSEC_COUNT_CONTROL */
+#define MSCC_MS_COUNT_CONTROL_RESET_ALL			BIT(0)
+#define MSCC_MS_COUNT_CONTROL_DEBUG_ACCESS		BIT(1)
+#define MSCC_MS_COUNT_CONTROL_SATURATE_CNTRS		BIT(2)
+#define MSCC_MS_COUNT_CONTROL_AUTO_CNTR_RESET		BIT(3)
+
+/* MACSEC_PARAMS2_IG_CC_CONTROL */
+#define MSCC_MS_PARAMS2_IG_CC_CONTROL_NON_MATCH_CTRL_ACT	BIT(14)
+#define MSCC_MS_PARAMS2_IG_CC_CONTROL_NON_MATCH_ACT	BIT(15)
+
+/* MACSEC_PARAMS2_IG_CP_TAG */
+#define MSCC_MS_PARAMS2_IG_CP_TAG_MAP_TBL(x)		(x)
+#define MSCC_MS_PARAMS2_IG_CP_TAG_MAP_TBL_M		GENMASK(23, 0)
+#define MSCC_MS_PARAMS2_IG_CP_TAG_DEF_UP(x)		((x) << 24)
+#define MSCC_MS_PARAMS2_IG_CP_TAG_DEF_UP_M		GENMASK(26, 24)
+#define MSCC_MS_PARAMS2_IG_CP_TAG_STAG_UP_EN		BIT(27)
+#define MSCC_MS_PARAMS2_IG_CP_TAG_QTAG_UP_EN		BIT(28)
+#define MSCC_MS_PARAMS2_IG_CP_TAG_PARSE_QINQ		BIT(29)
+#define MSCC_MS_PARAMS2_IG_CP_TAG_PARSE_STAG		BIT(30)
+#define MSCC_MS_PARAMS2_IG_CP_TAG_PARSE_QTAG		BIT(31)
+
+/* MACSEC_VLAN_MTU_CHECK */
+#define MSCC_MS_VLAN_MTU_CHECK_MTU_COMPARE(x)		(x)
+#define MSCC_MS_VLAN_MTU_CHECK_MTU_COMPARE_M		GENMASK(14, 0)
+#define MSCC_MS_VLAN_MTU_CHECK_MTU_COMP_DROP		BIT(15)
+
+/* MACSEC_NON_VLAN_MTU_CHECK */
+#define MSCC_MS_NON_VLAN_MTU_CHECK_NV_MTU_COMPARE(x)	(x)
+#define MSCC_MS_NON_VLAN_MTU_CHECK_NV_MTU_COMPARE_M	GENMASK(14, 0)
+#define MSCC_MS_NON_VLAN_MTU_CHECK_NV_MTU_COMP_DROP	BIT(15)
+
+/* MACSEC_PP_CTRL */
+#define MSCC_MS_PP_CTRL_MACSEC_OCTET_INCR_MODE		BIT(0)
+
+/* MACSEC_INTR_CTRL_STATUS */
+#define MSCC_MS_INTR_CTRL_STATUS_INTR_CLR_STATUS(x)	(x)
+#define MSCC_MS_INTR_CTRL_STATUS_INTR_CLR_STATUS_M	GENMASK(15, 0)
+#define MSCC_MS_INTR_CTRL_STATUS_INTR_ENABLE(x)		((x) << 16)
+#define MSCC_MS_INTR_CTRL_STATUS_INTR_ENABLE_M		GENMASK(31, 16)
+
+#endif
-- 
cgit v1.2.3-59-g8ed1b


From 28c5107aa904ef9db6b023039d20b6b4c4181675 Mon Sep 17 00:00:00 2001
From: Antoine Tenart <antoine.tenart@bootlin.com>
Date: Mon, 13 Jan 2020 23:31:46 +0100
Subject: net: phy: mscc: macsec support

This patch adds MACsec offloading support to some Microsemi PHYs, to
configure flows and transformations so that matched packets can be
processed by the MACsec engine, either at egress, or at ingress.

Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/Kconfig       |   3 +
 drivers/net/phy/mscc.c        | 691 ++++++++++++++++++++++++++++++++++++++++++
 drivers/net/phy/mscc_macsec.h |   4 +
 3 files changed, 698 insertions(+)

diff --git a/drivers/net/phy/Kconfig b/drivers/net/phy/Kconfig
index 2e016271e126..ac82ff959b7c 100644
--- a/drivers/net/phy/Kconfig
+++ b/drivers/net/phy/Kconfig
@@ -437,6 +437,9 @@ config MICROCHIP_T1_PHY
 
 config MICROSEMI_PHY
 	tristate "Microsemi PHYs"
+	depends on MACSEC || MACSEC=n
+	select CRYPTO_AES
+	select CRYPTO_ECB
 	---help---
 	  Currently supports VSC8514, VSC8530, VSC8531, VSC8540 and VSC8541 PHYs
 
diff --git a/drivers/net/phy/mscc.c b/drivers/net/phy/mscc.c
index 8579a59a1336..ccf17818570f 100644
--- a/drivers/net/phy/mscc.c
+++ b/drivers/net/phy/mscc.c
@@ -18,6 +18,13 @@
 #include <linux/netdevice.h>
 #include <dt-bindings/net/mscc-phy-vsc8531.h>
 
+#include <linux/scatterlist.h>
+#include <crypto/skcipher.h>
+
+#if IS_ENABLED(CONFIG_MACSEC)
+#include <net/macsec.h>
+#endif
+
 #include "mscc_macsec.h"
 #include "mscc_mac.h"
 #include "mscc_fc_buffer.h"
@@ -436,6 +443,44 @@ static const struct vsc85xx_hw_stat vsc8584_hw_stats[] = {
 	},
 };
 
+#if IS_ENABLED(CONFIG_MACSEC)
+struct macsec_flow {
+	struct list_head list;
+	enum mscc_macsec_destination_ports port;
+	enum macsec_bank bank;
+	u32 index;
+	int assoc_num;
+	bool has_transformation;
+
+	/* Highest takes precedence [0..15] */
+	u8 priority;
+
+	u8 key[MACSEC_KEYID_LEN];
+
+	union {
+		struct macsec_rx_sa *rx_sa;
+		struct macsec_tx_sa *tx_sa;
+	};
+
+	/* Matching */
+	struct {
+		u8 sci:1;
+		u8 tagged:1;
+		u8 untagged:1;
+		u8 etype:1;
+	} match;
+
+	u16 etype;
+
+	/* Action */
+	struct {
+		u8 bypass:1;
+		u8 drop:1;
+	} action;
+
+};
+#endif
+
 struct vsc8531_private {
 	int rate_magic;
 	u16 supp_led_modes;
@@ -449,6 +494,19 @@ struct vsc8531_private {
 	 * package.
 	 */
 	unsigned int base_addr;
+
+#if IS_ENABLED(CONFIG_MACSEC)
+	/* MACsec fields:
+	 * - One SecY per device (enforced at the s/w implementation level)
+	 * - macsec_flows: list of h/w flows
+	 * - ingr_flows: bitmap of ingress flows
+	 * - egr_flows: bitmap of egress flows
+	 */
+	struct macsec_secy *secy;
+	struct list_head macsec_flows;
+	unsigned long ingr_flows;
+	unsigned long egr_flows;
+#endif
 };
 
 #ifdef CONFIG_OF_MDIO
@@ -1951,6 +2009,634 @@ static int vsc8584_macsec_init(struct phy_device *phydev)
 
 	return 0;
 }
+
+static void vsc8584_macsec_flow(struct phy_device *phydev,
+				struct macsec_flow *flow)
+{
+	struct vsc8531_private *priv = phydev->priv;
+	enum macsec_bank bank = flow->bank;
+	u32 val, match = 0, mask = 0, action = 0, idx = flow->index;
+
+	if (flow->match.tagged)
+		match |= MSCC_MS_SAM_MISC_MATCH_TAGGED;
+	if (flow->match.untagged)
+		match |= MSCC_MS_SAM_MISC_MATCH_UNTAGGED;
+
+	if (bank == MACSEC_INGR && flow->assoc_num >= 0) {
+		match |= MSCC_MS_SAM_MISC_MATCH_AN(flow->assoc_num);
+		mask |= MSCC_MS_SAM_MASK_AN_MASK(0x3);
+	}
+
+	if (bank == MACSEC_INGR && flow->match.sci && flow->rx_sa->sc->sci) {
+		match |= MSCC_MS_SAM_MISC_MATCH_TCI(BIT(3));
+		mask |= MSCC_MS_SAM_MASK_TCI_MASK(BIT(3)) |
+			MSCC_MS_SAM_MASK_SCI_MASK;
+
+		vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_SAM_MATCH_SCI_LO(idx),
+					 lower_32_bits(flow->rx_sa->sc->sci));
+		vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_SAM_MATCH_SCI_HI(idx),
+					 upper_32_bits(flow->rx_sa->sc->sci));
+	}
+
+	if (flow->match.etype) {
+		mask |= MSCC_MS_SAM_MASK_MAC_ETYPE_MASK;
+
+		vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_SAM_MAC_SA_MATCH_HI(idx),
+					 MSCC_MS_SAM_MAC_SA_MATCH_HI_ETYPE(htons(flow->etype)));
+	}
+
+	match |= MSCC_MS_SAM_MISC_MATCH_PRIORITY(flow->priority);
+
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_SAM_MISC_MATCH(idx), match);
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_SAM_MASK(idx), mask);
+
+	/* Action for matching packets */
+	if (flow->action.drop)
+		action = MSCC_MS_FLOW_DROP;
+	else if (flow->action.bypass || flow->port == MSCC_MS_PORT_UNCONTROLLED)
+		action = MSCC_MS_FLOW_BYPASS;
+	else
+		action = (bank == MACSEC_INGR) ?
+			 MSCC_MS_FLOW_INGRESS : MSCC_MS_FLOW_EGRESS;
+
+	val = MSCC_MS_SAM_FLOW_CTRL_FLOW_TYPE(action) |
+	      MSCC_MS_SAM_FLOW_CTRL_DROP_ACTION(MSCC_MS_ACTION_DROP) |
+	      MSCC_MS_SAM_FLOW_CTRL_DEST_PORT(flow->port);
+
+	if (action == MSCC_MS_FLOW_BYPASS)
+		goto write_ctrl;
+
+	if (bank == MACSEC_INGR) {
+		if (priv->secy->replay_protect)
+			val |= MSCC_MS_SAM_FLOW_CTRL_REPLAY_PROTECT;
+		if (priv->secy->validate_frames == MACSEC_VALIDATE_STRICT)
+			val |= MSCC_MS_SAM_FLOW_CTRL_VALIDATE_FRAMES(MSCC_MS_VALIDATE_STRICT);
+		else if (priv->secy->validate_frames == MACSEC_VALIDATE_CHECK)
+			val |= MSCC_MS_SAM_FLOW_CTRL_VALIDATE_FRAMES(MSCC_MS_VALIDATE_CHECK);
+	} else if (bank == MACSEC_EGR) {
+		if (priv->secy->protect_frames)
+			val |= MSCC_MS_SAM_FLOW_CTRL_PROTECT_FRAME;
+		if (priv->secy->tx_sc.encrypt)
+			val |= MSCC_MS_SAM_FLOW_CTRL_CONF_PROTECT;
+		if (priv->secy->tx_sc.send_sci)
+			val |= MSCC_MS_SAM_FLOW_CTRL_INCLUDE_SCI;
+	}
+
+write_ctrl:
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_SAM_FLOW_CTRL(idx), val);
+}
+
+static struct macsec_flow *vsc8584_macsec_find_flow(struct macsec_context *ctx,
+						    enum macsec_bank bank)
+{
+	struct vsc8531_private *priv = ctx->phydev->priv;
+	struct macsec_flow *pos, *tmp;
+
+	list_for_each_entry_safe(pos, tmp, &priv->macsec_flows, list)
+		if (pos->assoc_num == ctx->sa.assoc_num && pos->bank == bank)
+			return pos;
+
+	return ERR_PTR(-ENOENT);
+}
+
+static void vsc8584_macsec_flow_enable(struct phy_device *phydev,
+				       struct macsec_flow *flow)
+{
+	enum macsec_bank bank = flow->bank;
+	u32 val, idx = flow->index;
+
+	if ((flow->bank == MACSEC_INGR && flow->rx_sa && !flow->rx_sa->active) ||
+	    (flow->bank == MACSEC_EGR && flow->tx_sa && !flow->tx_sa->active))
+		return;
+
+	/* Enable */
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_SAM_ENTRY_SET1, BIT(idx));
+
+	/* Set in-use */
+	val = vsc8584_macsec_phy_read(phydev, bank, MSCC_MS_SAM_FLOW_CTRL(idx));
+	val |= MSCC_MS_SAM_FLOW_CTRL_SA_IN_USE;
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_SAM_FLOW_CTRL(idx), val);
+}
+
+static void vsc8584_macsec_flow_disable(struct phy_device *phydev,
+					struct macsec_flow *flow)
+{
+	enum macsec_bank bank = flow->bank;
+	u32 val, idx = flow->index;
+
+	/* Disable */
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_SAM_ENTRY_CLEAR1, BIT(idx));
+
+	/* Clear in-use */
+	val = vsc8584_macsec_phy_read(phydev, bank, MSCC_MS_SAM_FLOW_CTRL(idx));
+	val &= ~MSCC_MS_SAM_FLOW_CTRL_SA_IN_USE;
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_SAM_FLOW_CTRL(idx), val);
+}
+
+static u32 vsc8584_macsec_flow_context_id(struct macsec_flow *flow)
+{
+	if (flow->bank == MACSEC_INGR)
+		return flow->index + MSCC_MS_MAX_FLOWS;
+
+	return flow->index;
+}
+
+/* Derive the AES key to get a key for the hash autentication */
+static int vsc8584_macsec_derive_key(const u8 key[MACSEC_KEYID_LEN],
+				     u16 key_len, u8 hkey[16])
+{
+	struct crypto_skcipher *tfm = crypto_alloc_skcipher("ecb(aes)", 0, 0);
+	struct skcipher_request *req = NULL;
+	struct scatterlist src, dst;
+	DECLARE_CRYPTO_WAIT(wait);
+	u32 input[4] = {0};
+	int ret;
+
+	if (IS_ERR(tfm))
+		return PTR_ERR(tfm);
+
+	req = skcipher_request_alloc(tfm, GFP_KERNEL);
+	if (!req) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	skcipher_request_set_callback(req, CRYPTO_TFM_REQ_MAY_BACKLOG |
+				      CRYPTO_TFM_REQ_MAY_SLEEP, crypto_req_done,
+				      &wait);
+	ret = crypto_skcipher_setkey(tfm, key, key_len);
+	if (ret < 0)
+		goto out;
+
+	sg_init_one(&src, input, 16);
+	sg_init_one(&dst, hkey, 16);
+	skcipher_request_set_crypt(req, &src, &dst, 16, NULL);
+
+	ret = crypto_wait_req(crypto_skcipher_encrypt(req), &wait);
+
+out:
+	skcipher_request_free(req);
+	crypto_free_skcipher(tfm);
+	return ret;
+}
+
+static int vsc8584_macsec_transformation(struct phy_device *phydev,
+					 struct macsec_flow *flow)
+{
+	struct vsc8531_private *priv = phydev->priv;
+	enum macsec_bank bank = flow->bank;
+	int i, ret, index = flow->index;
+	u32 rec = 0, control = 0;
+	u8 hkey[16];
+	sci_t sci;
+
+	ret = vsc8584_macsec_derive_key(flow->key, priv->secy->key_len, hkey);
+	if (ret)
+		return ret;
+
+	switch (priv->secy->key_len) {
+	case 16:
+		control |= CONTROL_CRYPTO_ALG(CTRYPTO_ALG_AES_CTR_128);
+		break;
+	case 32:
+		control |= CONTROL_CRYPTO_ALG(CTRYPTO_ALG_AES_CTR_256);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	control |= (bank == MACSEC_EGR) ?
+		   (CONTROL_TYPE_EGRESS | CONTROL_AN(priv->secy->tx_sc.encoding_sa)) :
+		   (CONTROL_TYPE_INGRESS | CONTROL_SEQ_MASK);
+
+	control |= CONTROL_UPDATE_SEQ | CONTROL_ENCRYPT_AUTH | CONTROL_KEY_IN_CTX |
+		   CONTROL_IV0 | CONTROL_IV1 | CONTROL_IV_IN_SEQ |
+		   CONTROL_DIGEST_TYPE(0x2) | CONTROL_SEQ_TYPE(0x1) |
+		   CONTROL_AUTH_ALG(AUTH_ALG_AES_GHAS) | CONTROL_CONTEXT_ID;
+
+	/* Set the control word */
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_XFORM_REC(index, rec++),
+				 control);
+
+	/* Set the context ID. Must be unique. */
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_XFORM_REC(index, rec++),
+				 vsc8584_macsec_flow_context_id(flow));
+
+	/* Set the encryption/decryption key */
+	for (i = 0; i < priv->secy->key_len / sizeof(u32); i++)
+		vsc8584_macsec_phy_write(phydev, bank,
+					 MSCC_MS_XFORM_REC(index, rec++),
+					 ((u32 *)flow->key)[i]);
+
+	/* Set the authentication key */
+	for (i = 0; i < 4; i++)
+		vsc8584_macsec_phy_write(phydev, bank,
+					 MSCC_MS_XFORM_REC(index, rec++),
+					 ((u32 *)hkey)[i]);
+
+	/* Initial sequence number */
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_XFORM_REC(index, rec++),
+				 bank == MACSEC_INGR ?
+				 flow->rx_sa->next_pn : flow->tx_sa->next_pn);
+
+	if (bank == MACSEC_INGR)
+		/* Set the mask (replay window size) */
+		vsc8584_macsec_phy_write(phydev, bank,
+					 MSCC_MS_XFORM_REC(index, rec++),
+					 priv->secy->replay_window);
+
+	/* Set the input vectors */
+	sci = bank == MACSEC_INGR ? flow->rx_sa->sc->sci : priv->secy->sci;
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_XFORM_REC(index, rec++),
+				 lower_32_bits(sci));
+	vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_XFORM_REC(index, rec++),
+				 upper_32_bits(sci));
+
+	while (rec < 20)
+		vsc8584_macsec_phy_write(phydev, bank, MSCC_MS_XFORM_REC(index, rec++),
+					 0);
+
+	flow->has_transformation = true;
+	return 0;
+}
+
+static struct macsec_flow *vsc8584_macsec_alloc_flow(struct vsc8531_private *priv,
+						     enum macsec_bank bank)
+{
+	unsigned long *bitmap = bank == MACSEC_INGR ?
+				&priv->ingr_flows : &priv->egr_flows;
+	struct macsec_flow *flow;
+	int index;
+
+	index = find_first_zero_bit(bitmap, MSCC_MS_MAX_FLOWS);
+
+	if (index == MSCC_MS_MAX_FLOWS)
+		return ERR_PTR(-ENOMEM);
+
+	flow = kzalloc(sizeof(*flow), GFP_KERNEL);
+	if (!flow)
+		return ERR_PTR(-ENOMEM);
+
+	set_bit(index, bitmap);
+	flow->index = index;
+	flow->bank = bank;
+	flow->priority = 8;
+	flow->assoc_num = -1;
+
+	list_add_tail(&flow->list, &priv->macsec_flows);
+	return flow;
+}
+
+static void vsc8584_macsec_free_flow(struct vsc8531_private *priv,
+				     struct macsec_flow *flow)
+{
+	unsigned long *bitmap = flow->bank == MACSEC_INGR ?
+				&priv->ingr_flows : &priv->egr_flows;
+
+	list_del(&flow->list);
+	clear_bit(flow->index, bitmap);
+	kfree(flow);
+}
+
+static int vsc8584_macsec_add_flow(struct phy_device *phydev,
+				   struct macsec_flow *flow, bool update)
+{
+	int ret;
+
+	flow->port = MSCC_MS_PORT_CONTROLLED;
+	vsc8584_macsec_flow(phydev, flow);
+
+	if (update)
+		return 0;
+
+	ret = vsc8584_macsec_transformation(phydev, flow);
+	if (ret) {
+		vsc8584_macsec_free_flow(phydev->priv, flow);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int vsc8584_macsec_default_flows(struct phy_device *phydev)
+{
+	struct macsec_flow *flow;
+
+	/* Add a rule to let the MKA traffic go through, ingress */
+	flow = vsc8584_macsec_alloc_flow(phydev->priv, MACSEC_INGR);
+	if (IS_ERR(flow))
+		return PTR_ERR(flow);
+
+	flow->priority = 15;
+	flow->port = MSCC_MS_PORT_UNCONTROLLED;
+	flow->match.tagged = 1;
+	flow->match.untagged = 1;
+	flow->match.etype = 1;
+	flow->etype = ETH_P_PAE;
+	flow->action.bypass = 1;
+
+	vsc8584_macsec_flow(phydev, flow);
+	vsc8584_macsec_flow_enable(phydev, flow);
+
+	/* Add a rule to let the MKA traffic go through, egress */
+	flow = vsc8584_macsec_alloc_flow(phydev->priv, MACSEC_EGR);
+	if (IS_ERR(flow))
+		return PTR_ERR(flow);
+
+	flow->priority = 15;
+	flow->port = MSCC_MS_PORT_COMMON;
+	flow->match.untagged = 1;
+	flow->match.etype = 1;
+	flow->etype = ETH_P_PAE;
+	flow->action.bypass = 1;
+
+	vsc8584_macsec_flow(phydev, flow);
+	vsc8584_macsec_flow_enable(phydev, flow);
+
+	return 0;
+}
+
+static void vsc8584_macsec_del_flow(struct phy_device *phydev,
+				    struct macsec_flow *flow)
+{
+	vsc8584_macsec_flow_disable(phydev, flow);
+	vsc8584_macsec_free_flow(phydev->priv, flow);
+}
+
+static int __vsc8584_macsec_add_rxsa(struct macsec_context *ctx,
+				     struct macsec_flow *flow, bool update)
+{
+	struct phy_device *phydev = ctx->phydev;
+	struct vsc8531_private *priv = phydev->priv;
+
+	if (!flow) {
+		flow = vsc8584_macsec_alloc_flow(priv, MACSEC_INGR);
+		if (IS_ERR(flow))
+			return PTR_ERR(flow);
+
+		memcpy(flow->key, ctx->sa.key, priv->secy->key_len);
+	}
+
+	flow->assoc_num = ctx->sa.assoc_num;
+	flow->rx_sa = ctx->sa.rx_sa;
+
+	/* Always match tagged packets on ingress */
+	flow->match.tagged = 1;
+	flow->match.sci = 1;
+
+	if (priv->secy->validate_frames != MACSEC_VALIDATE_DISABLED)
+		flow->match.untagged = 1;
+
+	return vsc8584_macsec_add_flow(phydev, flow, update);
+}
+
+static int __vsc8584_macsec_add_txsa(struct macsec_context *ctx,
+				     struct macsec_flow *flow, bool update)
+{
+	struct phy_device *phydev = ctx->phydev;
+	struct vsc8531_private *priv = phydev->priv;
+
+	if (!flow) {
+		flow = vsc8584_macsec_alloc_flow(priv, MACSEC_EGR);
+		if (IS_ERR(flow))
+			return PTR_ERR(flow);
+
+		memcpy(flow->key, ctx->sa.key, priv->secy->key_len);
+	}
+
+	flow->assoc_num = ctx->sa.assoc_num;
+	flow->tx_sa = ctx->sa.tx_sa;
+
+	/* Always match untagged packets on egress */
+	flow->match.untagged = 1;
+
+	return vsc8584_macsec_add_flow(phydev, flow, update);
+}
+
+static int vsc8584_macsec_dev_open(struct macsec_context *ctx)
+{
+	struct vsc8531_private *priv = ctx->phydev->priv;
+	struct macsec_flow *flow, *tmp;
+
+	/* No operation to perform before the commit step */
+	if (ctx->prepare)
+		return 0;
+
+	list_for_each_entry_safe(flow, tmp, &priv->macsec_flows, list)
+		vsc8584_macsec_flow_enable(ctx->phydev, flow);
+
+	return 0;
+}
+
+static int vsc8584_macsec_dev_stop(struct macsec_context *ctx)
+{
+	struct vsc8531_private *priv = ctx->phydev->priv;
+	struct macsec_flow *flow, *tmp;
+
+	/* No operation to perform before the commit step */
+	if (ctx->prepare)
+		return 0;
+
+	list_for_each_entry_safe(flow, tmp, &priv->macsec_flows, list)
+		vsc8584_macsec_flow_disable(ctx->phydev, flow);
+
+	return 0;
+}
+
+static int vsc8584_macsec_add_secy(struct macsec_context *ctx)
+{
+	struct vsc8531_private *priv = ctx->phydev->priv;
+	struct macsec_secy *secy = ctx->secy;
+
+	if (ctx->prepare) {
+		if (priv->secy)
+			return -EEXIST;
+
+		return 0;
+	}
+
+	priv->secy = secy;
+
+	vsc8584_macsec_flow_default_action(ctx->phydev, MACSEC_EGR,
+					   secy->validate_frames != MACSEC_VALIDATE_DISABLED);
+	vsc8584_macsec_flow_default_action(ctx->phydev, MACSEC_INGR,
+					   secy->validate_frames != MACSEC_VALIDATE_DISABLED);
+
+	return vsc8584_macsec_default_flows(ctx->phydev);
+}
+
+static int vsc8584_macsec_del_secy(struct macsec_context *ctx)
+{
+	struct vsc8531_private *priv = ctx->phydev->priv;
+	struct macsec_flow *flow, *tmp;
+
+	/* No operation to perform before the commit step */
+	if (ctx->prepare)
+		return 0;
+
+	list_for_each_entry_safe(flow, tmp, &priv->macsec_flows, list)
+		vsc8584_macsec_del_flow(ctx->phydev, flow);
+
+	vsc8584_macsec_flow_default_action(ctx->phydev, MACSEC_EGR, false);
+	vsc8584_macsec_flow_default_action(ctx->phydev, MACSEC_INGR, false);
+
+	priv->secy = NULL;
+	return 0;
+}
+
+static int vsc8584_macsec_upd_secy(struct macsec_context *ctx)
+{
+	/* No operation to perform before the commit step */
+	if (ctx->prepare)
+		return 0;
+
+	vsc8584_macsec_del_secy(ctx);
+	return vsc8584_macsec_add_secy(ctx);
+}
+
+static int vsc8584_macsec_add_rxsc(struct macsec_context *ctx)
+{
+	/* Nothing to do */
+	return 0;
+}
+
+static int vsc8584_macsec_upd_rxsc(struct macsec_context *ctx)
+{
+	return -EOPNOTSUPP;
+}
+
+static int vsc8584_macsec_del_rxsc(struct macsec_context *ctx)
+{
+	struct vsc8531_private *priv = ctx->phydev->priv;
+	struct macsec_flow *flow, *tmp;
+
+	/* No operation to perform before the commit step */
+	if (ctx->prepare)
+		return 0;
+
+	list_for_each_entry_safe(flow, tmp, &priv->macsec_flows, list) {
+		if (flow->bank == MACSEC_INGR && flow->rx_sa &&
+		    flow->rx_sa->sc->sci == ctx->rx_sc->sci)
+			vsc8584_macsec_del_flow(ctx->phydev, flow);
+	}
+
+	return 0;
+}
+
+static int vsc8584_macsec_add_rxsa(struct macsec_context *ctx)
+{
+	struct macsec_flow *flow = NULL;
+
+	if (ctx->prepare)
+		return __vsc8584_macsec_add_rxsa(ctx, flow, false);
+
+	flow = vsc8584_macsec_find_flow(ctx, MACSEC_INGR);
+	if (IS_ERR(flow))
+		return PTR_ERR(flow);
+
+	vsc8584_macsec_flow_enable(ctx->phydev, flow);
+	return 0;
+}
+
+static int vsc8584_macsec_upd_rxsa(struct macsec_context *ctx)
+{
+	struct macsec_flow *flow;
+
+	flow = vsc8584_macsec_find_flow(ctx, MACSEC_INGR);
+	if (IS_ERR(flow))
+		return PTR_ERR(flow);
+
+	if (ctx->prepare) {
+		/* Make sure the flow is disabled before updating it */
+		vsc8584_macsec_flow_disable(ctx->phydev, flow);
+
+		return __vsc8584_macsec_add_rxsa(ctx, flow, true);
+	}
+
+	vsc8584_macsec_flow_enable(ctx->phydev, flow);
+	return 0;
+}
+
+static int vsc8584_macsec_del_rxsa(struct macsec_context *ctx)
+{
+	struct macsec_flow *flow;
+
+	flow = vsc8584_macsec_find_flow(ctx, MACSEC_INGR);
+
+	if (IS_ERR(flow))
+		return PTR_ERR(flow);
+	if (ctx->prepare)
+		return 0;
+
+	vsc8584_macsec_del_flow(ctx->phydev, flow);
+	return 0;
+}
+
+static int vsc8584_macsec_add_txsa(struct macsec_context *ctx)
+{
+	struct macsec_flow *flow = NULL;
+
+	if (ctx->prepare)
+		return __vsc8584_macsec_add_txsa(ctx, flow, false);
+
+	flow = vsc8584_macsec_find_flow(ctx, MACSEC_EGR);
+	if (IS_ERR(flow))
+		return PTR_ERR(flow);
+
+	vsc8584_macsec_flow_enable(ctx->phydev, flow);
+	return 0;
+}
+
+static int vsc8584_macsec_upd_txsa(struct macsec_context *ctx)
+{
+	struct macsec_flow *flow;
+
+	flow = vsc8584_macsec_find_flow(ctx, MACSEC_EGR);
+	if (IS_ERR(flow))
+		return PTR_ERR(flow);
+
+	if (ctx->prepare) {
+		/* Make sure the flow is disabled before updating it */
+		vsc8584_macsec_flow_disable(ctx->phydev, flow);
+
+		return __vsc8584_macsec_add_txsa(ctx, flow, true);
+	}
+
+	vsc8584_macsec_flow_enable(ctx->phydev, flow);
+	return 0;
+}
+
+static int vsc8584_macsec_del_txsa(struct macsec_context *ctx)
+{
+	struct macsec_flow *flow;
+
+	flow = vsc8584_macsec_find_flow(ctx, MACSEC_EGR);
+
+	if (IS_ERR(flow))
+		return PTR_ERR(flow);
+	if (ctx->prepare)
+		return 0;
+
+	vsc8584_macsec_del_flow(ctx->phydev, flow);
+	return 0;
+}
+
+static struct macsec_ops vsc8584_macsec_ops = {
+	.mdo_dev_open = vsc8584_macsec_dev_open,
+	.mdo_dev_stop = vsc8584_macsec_dev_stop,
+	.mdo_add_secy = vsc8584_macsec_add_secy,
+	.mdo_upd_secy = vsc8584_macsec_upd_secy,
+	.mdo_del_secy = vsc8584_macsec_del_secy,
+	.mdo_add_rxsc = vsc8584_macsec_add_rxsc,
+	.mdo_upd_rxsc = vsc8584_macsec_upd_rxsc,
+	.mdo_del_rxsc = vsc8584_macsec_del_rxsc,
+	.mdo_add_rxsa = vsc8584_macsec_add_rxsa,
+	.mdo_upd_rxsa = vsc8584_macsec_upd_rxsa,
+	.mdo_del_rxsa = vsc8584_macsec_del_rxsa,
+	.mdo_add_txsa = vsc8584_macsec_add_txsa,
+	.mdo_upd_txsa = vsc8584_macsec_upd_txsa,
+	.mdo_del_txsa = vsc8584_macsec_del_txsa,
+};
 #endif /* CONFIG_MACSEC */
 
 /* Check if one PHY has already done the init of the parts common to all PHYs
@@ -2109,6 +2795,11 @@ static int vsc8584_config_init(struct phy_device *phydev)
 	case PHY_ID_VSC8575:
 	case PHY_ID_VSC8582:
 	case PHY_ID_VSC8584:
+		INIT_LIST_HEAD(&vsc8531->macsec_flows);
+		vsc8531->secy = NULL;
+
+		phydev->macsec_ops = &vsc8584_macsec_ops;
+
 		ret = vsc8584_macsec_init(phydev);
 		if (ret)
 			goto err;
diff --git a/drivers/net/phy/mscc_macsec.h b/drivers/net/phy/mscc_macsec.h
index 0d108da28dad..9b5d0af91d20 100644
--- a/drivers/net/phy/mscc_macsec.h
+++ b/drivers/net/phy/mscc_macsec.h
@@ -8,6 +8,8 @@
 #ifndef _MSCC_OCELOT_MACSEC_H_
 #define _MSCC_OCELOT_MACSEC_H_
 
+#define MSCC_MS_MAX_FLOWS		16
+
 #define CONTROL_TYPE_EGRESS		0x6
 #define CONTROL_TYPE_INGRESS		0xf
 #define CONTROL_IV0			BIT(5)
@@ -59,6 +61,8 @@ enum mscc_macsec_validate_levels {
 #define MSCC_MS_XFORM_REC(x, y)		(((x) << 5) + (y))
 #define MSCC_MS_ENA_CFG			0x800
 #define MSCC_MS_FC_CFG			0x804
+#define MSCC_MS_SAM_MAC_SA_MATCH_LO(x)	(0x1000 + ((x) << 4))
+#define MSCC_MS_SAM_MAC_SA_MATCH_HI(x)	(0x1001 + ((x) << 4))
 #define MSCC_MS_SAM_MISC_MATCH(x)	(0x1004 + ((x) << 4))
 #define MSCC_MS_SAM_MATCH_SCI_LO(x)	(0x1005 + ((x) << 4))
 #define MSCC_MS_SAM_MATCH_SCI_HI(x)	(0x1006 + ((x) << 4))
-- 
cgit v1.2.3-59-g8ed1b


From 5c937de78b39e47ce9924fc4b863c5b727edc328 Mon Sep 17 00:00:00 2001
From: Antoine Tenart <antoine.tenart@bootlin.com>
Date: Mon, 13 Jan 2020 23:31:47 +0100
Subject: net: macsec: PN wrap callback

Allow to call macsec_pn_wrapped from hardware drivers to notify when a
PN rolls over. Some drivers might used an interrupt to implement this.

Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/macsec.c | 25 +++++++++++++++++++------
 include/net/macsec.h |  2 ++
 2 files changed, 21 insertions(+), 6 deletions(-)

diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c
index e515919e8687..45bfd99f17fa 100644
--- a/drivers/net/macsec.c
+++ b/drivers/net/macsec.c
@@ -424,6 +424,23 @@ static struct macsec_eth_header *macsec_ethhdr(struct sk_buff *skb)
 	return (struct macsec_eth_header *)skb_mac_header(skb);
 }
 
+static void __macsec_pn_wrapped(struct macsec_secy *secy,
+				struct macsec_tx_sa *tx_sa)
+{
+	pr_debug("PN wrapped, transitioning to !oper\n");
+	tx_sa->active = false;
+	if (secy->protect_frames)
+		secy->operational = false;
+}
+
+void macsec_pn_wrapped(struct macsec_secy *secy, struct macsec_tx_sa *tx_sa)
+{
+	spin_lock_bh(&tx_sa->lock);
+	__macsec_pn_wrapped(secy, tx_sa);
+	spin_unlock_bh(&tx_sa->lock);
+}
+EXPORT_SYMBOL_GPL(macsec_pn_wrapped);
+
 static u32 tx_sa_update_pn(struct macsec_tx_sa *tx_sa, struct macsec_secy *secy)
 {
 	u32 pn;
@@ -432,12 +449,8 @@ static u32 tx_sa_update_pn(struct macsec_tx_sa *tx_sa, struct macsec_secy *secy)
 	pn = tx_sa->next_pn;
 
 	tx_sa->next_pn++;
-	if (tx_sa->next_pn == 0) {
-		pr_debug("PN wrapped, transitioning to !oper\n");
-		tx_sa->active = false;
-		if (secy->protect_frames)
-			secy->operational = false;
-	}
+	if (tx_sa->next_pn == 0)
+		__macsec_pn_wrapped(secy, tx_sa);
 	spin_unlock_bh(&tx_sa->lock);
 
 	return pn;
diff --git a/include/net/macsec.h b/include/net/macsec.h
index 16e7e5061178..92e43db8b566 100644
--- a/include/net/macsec.h
+++ b/include/net/macsec.h
@@ -219,4 +219,6 @@ struct macsec_ops {
 	int (*mdo_del_txsa)(struct macsec_context *ctx);
 };
 
+void macsec_pn_wrapped(struct macsec_secy *secy, struct macsec_tx_sa *tx_sa);
+
 #endif /* _NET_MACSEC_H_ */
-- 
cgit v1.2.3-59-g8ed1b


From 781449a4ae3b381950ee9aec4d8a54e35f66ab9b Mon Sep 17 00:00:00 2001
From: Antoine Tenart <antoine.tenart@bootlin.com>
Date: Mon, 13 Jan 2020 23:31:48 +0100
Subject: net: phy: mscc: PN rollover support

This patch adds support for handling MACsec PN rollover in the mscc PHY
driver. When a flow rolls over, an interrupt is fired. This patch adds
the logic to check all flows and identify the one rolling over in the
handle_interrupt PHY helper, then disables the flow and report the event
to the MACsec core.

Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/mscc.c        | 60 ++++++++++++++++++++++++++++++++++++++++++-
 drivers/net/phy/mscc_macsec.h |  2 ++
 2 files changed, 61 insertions(+), 1 deletion(-)

diff --git a/drivers/net/phy/mscc.c b/drivers/net/phy/mscc.c
index ccf17818570f..937ac7da2789 100644
--- a/drivers/net/phy/mscc.c
+++ b/drivers/net/phy/mscc.c
@@ -80,7 +80,7 @@ enum rgmii_rx_clock_delay {
 #define MSCC_PHY_EXT_PHY_CNTL_2		  24
 
 #define MII_VSC85XX_INT_MASK		  25
-#define MII_VSC85XX_INT_MASK_MASK	  0xa000
+#define MII_VSC85XX_INT_MASK_MASK	  0xa020
 #define MII_VSC85XX_INT_MASK_WOL	  0x0040
 #define MII_VSC85XX_INT_STATUS		  26
 
@@ -207,6 +207,9 @@ enum macsec_bank {
 #define SECURE_ON_ENABLE		  0x8000
 #define SECURE_ON_PASSWD_LEN_4		  0x4000
 
+#define MSCC_PHY_EXTENDED_INT		  28
+#define MSCC_PHY_EXTENDED_INT_MS_EGR	  BIT(9)
+
 /* Extended Page 3 Registers */
 #define MSCC_PHY_SERDES_TX_VALID_CNT	  21
 #define MSCC_PHY_SERDES_TX_CRC_ERR_CNT	  22
@@ -2831,6 +2834,43 @@ err:
 	return ret;
 }
 
+static int vsc8584_handle_interrupt(struct phy_device *phydev)
+{
+#if IS_ENABLED(CONFIG_MACSEC)
+	struct vsc8531_private *priv = phydev->priv;
+	struct macsec_flow *flow, *tmp;
+	u32 cause, rec;
+
+	/* Check MACsec PN rollover */
+	cause = vsc8584_macsec_phy_read(phydev, MACSEC_EGR,
+					MSCC_MS_INTR_CTRL_STATUS);
+	cause &= MSCC_MS_INTR_CTRL_STATUS_INTR_CLR_STATUS_M;
+	if (!(cause & MACSEC_INTR_CTRL_STATUS_ROLLOVER))
+		goto skip_rollover;
+
+	rec = 6 + priv->secy->key_len / sizeof(u32);
+	list_for_each_entry_safe(flow, tmp, &priv->macsec_flows, list) {
+		u32 val;
+
+		if (flow->bank != MACSEC_EGR || !flow->has_transformation)
+			continue;
+
+		val = vsc8584_macsec_phy_read(phydev, MACSEC_EGR,
+					      MSCC_MS_XFORM_REC(flow->index, rec));
+		if (val == 0xffffffff) {
+			vsc8584_macsec_flow_disable(phydev, flow);
+			macsec_pn_wrapped(priv->secy, flow->tx_sa);
+			break;
+		}
+	}
+
+skip_rollover:
+#endif
+
+	phy_mac_interrupt(phydev);
+	return 0;
+}
+
 static int vsc85xx_config_init(struct phy_device *phydev)
 {
 	int rc, i, phy_id;
@@ -3274,6 +3314,20 @@ static int vsc85xx_config_intr(struct phy_device *phydev)
 	int rc;
 
 	if (phydev->interrupts == PHY_INTERRUPT_ENABLED) {
+#if IS_ENABLED(CONFIG_MACSEC)
+		phy_write(phydev, MSCC_EXT_PAGE_ACCESS,
+			  MSCC_PHY_PAGE_EXTENDED_2);
+		phy_write(phydev, MSCC_PHY_EXTENDED_INT,
+			  MSCC_PHY_EXTENDED_INT_MS_EGR);
+		phy_write(phydev, MSCC_EXT_PAGE_ACCESS,
+			  MSCC_PHY_PAGE_STANDARD);
+
+		vsc8584_macsec_phy_write(phydev, MACSEC_EGR,
+					 MSCC_MS_AIC_CTRL, 0xf);
+		vsc8584_macsec_phy_write(phydev, MACSEC_EGR,
+			MSCC_MS_INTR_CTRL_STATUS,
+			MSCC_MS_INTR_CTRL_STATUS_INTR_ENABLE(MACSEC_INTR_CTRL_STATUS_ROLLOVER));
+#endif
 		rc = phy_write(phydev, MII_VSC85XX_INT_MASK,
 			       MII_VSC85XX_INT_MASK_MASK);
 	} else {
@@ -3623,6 +3677,7 @@ static struct phy_driver vsc85xx_driver[] = {
 	.config_aneg    = &vsc85xx_config_aneg,
 	.aneg_done	= &genphy_aneg_done,
 	.read_status	= &vsc85xx_read_status,
+	.handle_interrupt = &vsc8584_handle_interrupt,
 	.ack_interrupt  = &vsc85xx_ack_interrupt,
 	.config_intr    = &vsc85xx_config_intr,
 	.did_interrupt  = &vsc8584_did_interrupt,
@@ -3675,6 +3730,7 @@ static struct phy_driver vsc85xx_driver[] = {
 	.config_aneg    = &vsc85xx_config_aneg,
 	.aneg_done	= &genphy_aneg_done,
 	.read_status	= &vsc85xx_read_status,
+	.handle_interrupt = &vsc8584_handle_interrupt,
 	.ack_interrupt  = &vsc85xx_ack_interrupt,
 	.config_intr    = &vsc85xx_config_intr,
 	.did_interrupt  = &vsc8584_did_interrupt,
@@ -3699,6 +3755,7 @@ static struct phy_driver vsc85xx_driver[] = {
 	.config_aneg    = &vsc85xx_config_aneg,
 	.aneg_done	= &genphy_aneg_done,
 	.read_status	= &vsc85xx_read_status,
+	.handle_interrupt = &vsc8584_handle_interrupt,
 	.ack_interrupt  = &vsc85xx_ack_interrupt,
 	.config_intr    = &vsc85xx_config_intr,
 	.did_interrupt  = &vsc8584_did_interrupt,
@@ -3723,6 +3780,7 @@ static struct phy_driver vsc85xx_driver[] = {
 	.config_aneg    = &vsc85xx_config_aneg,
 	.aneg_done	= &genphy_aneg_done,
 	.read_status	= &vsc85xx_read_status,
+	.handle_interrupt = &vsc8584_handle_interrupt,
 	.ack_interrupt  = &vsc85xx_ack_interrupt,
 	.config_intr    = &vsc85xx_config_intr,
 	.did_interrupt  = &vsc8584_did_interrupt,
diff --git a/drivers/net/phy/mscc_macsec.h b/drivers/net/phy/mscc_macsec.h
index 9b5d0af91d20..d9ab6aba7482 100644
--- a/drivers/net/phy/mscc_macsec.h
+++ b/drivers/net/phy/mscc_macsec.h
@@ -83,6 +83,7 @@ enum mscc_macsec_validate_levels {
 #define MSCC_MS_STATUS_CONTEXT_CTRL	0x3d02
 #define MSCC_MS_INTR_CTRL_STATUS	0x3d04
 #define MSCC_MS_BLOCK_CTX_UPDATE	0x3d0c
+#define MSCC_MS_AIC_CTRL		0x3e02
 
 /* MACSEC_ENA_CFG */
 #define MSCC_MS_ENA_CFG_CLK_ENA				BIT(0)
@@ -260,5 +261,6 @@ enum mscc_macsec_validate_levels {
 #define MSCC_MS_INTR_CTRL_STATUS_INTR_CLR_STATUS_M	GENMASK(15, 0)
 #define MSCC_MS_INTR_CTRL_STATUS_INTR_ENABLE(x)		((x) << 16)
 #define MSCC_MS_INTR_CTRL_STATUS_INTR_ENABLE_M		GENMASK(31, 16)
+#define MACSEC_INTR_CTRL_STATUS_ROLLOVER		BIT(5)
 
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From 0a29275b6300f39f78a87f2038bbfe5bdbaeca47 Mon Sep 17 00:00:00 2001
From: Li RongQing <lirongqing@baidu.com>
Date: Fri, 10 Jan 2020 09:04:37 +0800
Subject: bpf: Return -EBADRQC for invalid map type in __bpf_tx_xdp_map

A negative value should be returned if map->map_type is invalid
although that is impossible now, but if we run into such situation
in future, then xdpbuff could be leaked.

Daniel Borkmann suggested:

-EBADRQC should be returned to stay consistent with generic XDP
for the tracepoint output and not to be confused with -EOPNOTSUPP
from other locations like dev_map_enqueue() when ndo_xdp_xmit is
missing and such.

Suggested-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Li RongQing <lirongqing@baidu.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/1578618277-18085-1-git-send-email-lirongqing@baidu.com
---
 net/core/filter.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/core/filter.c b/net/core/filter.c
index a702761ef369..d6f58dc4e1c4 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -3522,7 +3522,7 @@ static int __bpf_tx_xdp_map(struct net_device *dev_rx, void *fwd,
 	case BPF_MAP_TYPE_XSKMAP:
 		return __xsk_map_redirect(fwd, xdp);
 	default:
-		break;
+		return -EBADRQC;
 	}
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 5eee7bd7e245914e4e050c413dfe864e31805207 Mon Sep 17 00:00:00 2001
From: "Jason A. Donenfeld" <Jason@zx2c4.com>
Date: Mon, 13 Jan 2020 18:42:26 -0500
Subject: net: skbuff: disambiguate argument and member for skb_list_walk_safe
 helper

This worked before, because we made all callers name their next pointer
"next". But in trying to be more "drop-in" ready, the silliness here is
revealed. This commit fixes the problem by making the macro argument and
the member use different names.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 016b3c4ab99a..aaf73b34f72f 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -1479,9 +1479,9 @@ static inline void skb_mark_not_on_list(struct sk_buff *skb)
 }
 
 /* Iterate through singly-linked GSO fragments of an skb. */
-#define skb_list_walk_safe(first, skb, next)                                   \
-	for ((skb) = (first), (next) = (skb) ? (skb)->next : NULL; (skb);      \
-	     (skb) = (next), (next) = (skb) ? (skb)->next : NULL)
+#define skb_list_walk_safe(first, skb, next_skb)                               \
+	for ((skb) = (first), (next_skb) = (skb) ? (skb)->next : NULL; (skb);  \
+	     (skb) = (next_skb), (next_skb) = (skb) ? (skb)->next : NULL)
 
 static inline void skb_list_del_init(struct sk_buff *skb)
 {
-- 
cgit v1.2.3-59-g8ed1b


From 1a186c14cedbf3364634f2dfa9302ed31b8fab19 Mon Sep 17 00:00:00 2001
From: "Jason A. Donenfeld" <Jason@zx2c4.com>
Date: Mon, 13 Jan 2020 18:42:27 -0500
Subject: net: udp: use skb_list_walk_safe helper for gso segments

This is a straight-forward conversion case for the new function,
iterating over the return value from udp_rcv_segment, which actually is
a wrapper around skb_gso_segment.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/udp.c | 3 +--
 net/ipv6/udp.c | 3 +--
 2 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 93a355b6b092..208da0917469 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -2104,8 +2104,7 @@ static int udp_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
 	BUILD_BUG_ON(sizeof(struct udp_skb_cb) > SKB_SGO_CB_OFFSET);
 	__skb_push(skb, -skb_mac_offset(skb));
 	segs = udp_rcv_segment(sk, skb, true);
-	for (skb = segs; skb; skb = next) {
-		next = skb->next;
+	skb_list_walk_safe(segs, skb, next) {
 		__skb_pull(skb, skb_transport_offset(skb));
 		ret = udp_queue_rcv_one_skb(sk, skb);
 		if (ret > 0)
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 9fec580c968e..5dc439a391fe 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -690,8 +690,7 @@ static int udpv6_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
 
 	__skb_push(skb, -skb_mac_offset(skb));
 	segs = udp_rcv_segment(sk, skb, false);
-	for (skb = segs; skb; skb = next) {
-		next = skb->next;
+	skb_list_walk_safe(segs, skb, next) {
 		__skb_pull(skb, skb_transport_offset(skb));
 
 		ret = udpv6_queue_rcv_one_skb(sk, skb);
-- 
cgit v1.2.3-59-g8ed1b


From c3b18e0d925439043b5887c9a6129ff2ffaf58b0 Mon Sep 17 00:00:00 2001
From: "Jason A. Donenfeld" <Jason@zx2c4.com>
Date: Mon, 13 Jan 2020 18:42:28 -0500
Subject: net: xfrm: use skb_list_walk_safe helper for gso segments

This is converts xfrm segment iteration to use the new function, keeping
the flow of the existing code as intact as possible. One case is very
straight-forward, whereas the other case has some more subtle code that
likes to peak at ->next and relink skbs. By keeping the variables the
same as before, we can upgrade this code with minimal surgery required.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/xfrm/xfrm_device.c | 15 ++++-----------
 net/xfrm/xfrm_output.c |  9 +++------
 2 files changed, 7 insertions(+), 17 deletions(-)

diff --git a/net/xfrm/xfrm_device.c b/net/xfrm/xfrm_device.c
index 189ef15acbbc..50f567a88f45 100644
--- a/net/xfrm/xfrm_device.c
+++ b/net/xfrm/xfrm_device.c
@@ -78,7 +78,7 @@ struct sk_buff *validate_xmit_xfrm(struct sk_buff *skb, netdev_features_t featur
 	int err;
 	unsigned long flags;
 	struct xfrm_state *x;
-	struct sk_buff *skb2;
+	struct sk_buff *skb2, *nskb;
 	struct softnet_data *sd;
 	netdev_features_t esp_features = features;
 	struct xfrm_offload *xo = xfrm_offload(skb);
@@ -148,11 +148,7 @@ struct sk_buff *validate_xmit_xfrm(struct sk_buff *skb, netdev_features_t featur
 		return skb;
 	}
 
-	skb2 = skb;
-
-	do {
-		struct sk_buff *nskb = skb2->next;
-
+	skb_list_walk_safe(skb, skb2, nskb) {
 		esp_features |= skb->dev->gso_partial_features;
 		skb_mark_not_on_list(skb2);
 
@@ -176,14 +172,11 @@ struct sk_buff *validate_xmit_xfrm(struct sk_buff *skb, netdev_features_t featur
 			if (!skb)
 				return NULL;
 
-			goto skip_push;
+			continue;
 		}
 
 		skb_push(skb2, skb2->data - skb_mac_header(skb2));
-
-skip_push:
-		skb2 = nskb;
-	} while (skb2);
+	}
 
 	return skb;
 }
diff --git a/net/xfrm/xfrm_output.c b/net/xfrm/xfrm_output.c
index b1db55b50ba1..fafc7aba705f 100644
--- a/net/xfrm/xfrm_output.c
+++ b/net/xfrm/xfrm_output.c
@@ -533,7 +533,7 @@ static int xfrm_output2(struct net *net, struct sock *sk, struct sk_buff *skb)
 
 static int xfrm_output_gso(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
-	struct sk_buff *segs;
+	struct sk_buff *segs, *nskb;
 
 	BUILD_BUG_ON(sizeof(*IPCB(skb)) > SKB_SGO_CB_OFFSET);
 	BUILD_BUG_ON(sizeof(*IP6CB(skb)) > SKB_SGO_CB_OFFSET);
@@ -544,8 +544,7 @@ static int xfrm_output_gso(struct net *net, struct sock *sk, struct sk_buff *skb
 	if (segs == NULL)
 		return -EINVAL;
 
-	do {
-		struct sk_buff *nskb = segs->next;
+	skb_list_walk_safe(segs, segs, nskb) {
 		int err;
 
 		skb_mark_not_on_list(segs);
@@ -555,9 +554,7 @@ static int xfrm_output_gso(struct net *net, struct sock *sk, struct sk_buff *skb
 			kfree_skb_list(nskb);
 			return err;
 		}
-
-		segs = nskb;
-	} while (segs);
+	}
 
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 2cec4448db38758832c2edad439f99584bb8fa0d Mon Sep 17 00:00:00 2001
From: "Jason A. Donenfeld" <Jason@zx2c4.com>
Date: Mon, 13 Jan 2020 18:42:29 -0500
Subject: net: openvswitch: use skb_list_walk_safe helper for gso segments

This is a straight-forward conversion case for the new function, keeping
the flow of the existing code as intact as possible.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/openvswitch/datapath.c | 11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)

diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c
index e3a37d22539c..659c2a790fe7 100644
--- a/net/openvswitch/datapath.c
+++ b/net/openvswitch/datapath.c
@@ -321,8 +321,7 @@ static int queue_gso_packets(struct datapath *dp, struct sk_buff *skb,
 	}
 
 	/* Queue all of the segments. */
-	skb = segs;
-	do {
+	skb_list_walk_safe(segs, skb, nskb) {
 		if (gso_type & SKB_GSO_UDP && skb != segs)
 			key = &later_key;
 
@@ -330,17 +329,15 @@ static int queue_gso_packets(struct datapath *dp, struct sk_buff *skb,
 		if (err)
 			break;
 
-	} while ((skb = skb->next));
+	}
 
 	/* Free all of the segments. */
-	skb = segs;
-	do {
-		nskb = skb->next;
+	skb_list_walk_safe(segs, skb, nskb) {
 		if (err)
 			kfree_skb(skb);
 		else
 			consume_skb(skb);
-	} while ((skb = nskb));
+	}
 	return err;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From b950d8a5b300ceeb2e530291cddc6b3e547291f0 Mon Sep 17 00:00:00 2001
From: "Jason A. Donenfeld" <Jason@zx2c4.com>
Date: Mon, 13 Jan 2020 18:42:30 -0500
Subject: net: sched: use skb_list_walk_safe helper for gso segments

This is a straight-forward conversion case for the new function, keeping
the flow of the existing code as intact as possible.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sched/sch_cake.c | 4 +---
 net/sched/sch_tbf.c  | 4 +---
 2 files changed, 2 insertions(+), 6 deletions(-)

diff --git a/net/sched/sch_cake.c b/net/sched/sch_cake.c
index 90ef7cc79b69..1496e87cd07b 100644
--- a/net/sched/sch_cake.c
+++ b/net/sched/sch_cake.c
@@ -1682,8 +1682,7 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch,
 		if (IS_ERR_OR_NULL(segs))
 			return qdisc_drop(skb, sch, to_free);
 
-		while (segs) {
-			nskb = segs->next;
+		skb_list_walk_safe(segs, segs, nskb) {
 			skb_mark_not_on_list(segs);
 			qdisc_skb_cb(segs)->pkt_len = segs->len;
 			cobalt_set_enqueue_time(segs, now);
@@ -1696,7 +1695,6 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch,
 			slen += segs->len;
 			q->buffer_used += segs->truesize;
 			b->packets++;
-			segs = nskb;
 		}
 
 		/* stats */
diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c
index 5f72f3f916a5..2cd94973795c 100644
--- a/net/sched/sch_tbf.c
+++ b/net/sched/sch_tbf.c
@@ -155,8 +155,7 @@ static int tbf_segment(struct sk_buff *skb, struct Qdisc *sch,
 		return qdisc_drop(skb, sch, to_free);
 
 	nb = 0;
-	while (segs) {
-		nskb = segs->next;
+	skb_list_walk_safe(segs, segs, nskb) {
 		skb_mark_not_on_list(segs);
 		qdisc_skb_cb(segs)->pkt_len = segs->len;
 		len += segs->len;
@@ -167,7 +166,6 @@ static int tbf_segment(struct sk_buff *skb, struct Qdisc *sch,
 		} else {
 			nb++;
 		}
-		segs = nskb;
 	}
 	sch->q.qlen += nb;
 	if (nb > 1)
-- 
cgit v1.2.3-59-g8ed1b


From 88bebdf5b26fe9fdeb36f2152978a3bb478b0d4d Mon Sep 17 00:00:00 2001
From: "Jason A. Donenfeld" <Jason@zx2c4.com>
Date: Mon, 13 Jan 2020 18:42:31 -0500
Subject: net: ipv4: use skb_list_walk_safe helper for gso segments

This is a straight-forward conversion case for the new function, keeping
the flow of the existing code as intact as possible.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/ip_output.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 14db1e0b8a6e..d84819893db9 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -240,8 +240,8 @@ static int ip_finish_output2(struct net *net, struct sock *sk, struct sk_buff *s
 static int ip_finish_output_gso(struct net *net, struct sock *sk,
 				struct sk_buff *skb, unsigned int mtu)
 {
+	struct sk_buff *segs, *nskb;
 	netdev_features_t features;
-	struct sk_buff *segs;
 	int ret = 0;
 
 	/* common case: seglen is <= mtu
@@ -272,8 +272,7 @@ static int ip_finish_output_gso(struct net *net, struct sock *sk,
 
 	consume_skb(skb);
 
-	do {
-		struct sk_buff *nskb = segs->next;
+	skb_list_walk_safe(segs, segs, nskb) {
 		int err;
 
 		skb_mark_not_on_list(segs);
@@ -281,8 +280,7 @@ static int ip_finish_output_gso(struct net *net, struct sock *sk,
 
 		if (err && ret == 0)
 			ret = err;
-		segs = nskb;
-	} while (segs);
+	}
 
 	return ret;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 2670ee77c99c76f46c061caf65daa2e2bfb5dcc5 Mon Sep 17 00:00:00 2001
From: "Jason A. Donenfeld" <Jason@zx2c4.com>
Date: Mon, 13 Jan 2020 18:42:32 -0500
Subject: net: netfilter: use skb_list_walk_safe helper for gso segments

This is a straight-forward conversion case for the new function, keeping
the flow of the existing code as intact as possible.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/netfilter/nfnetlink_queue.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c
index feabdfb22920..76535fd9278c 100644
--- a/net/netfilter/nfnetlink_queue.c
+++ b/net/netfilter/nfnetlink_queue.c
@@ -778,7 +778,7 @@ nfqnl_enqueue_packet(struct nf_queue_entry *entry, unsigned int queuenum)
 {
 	unsigned int queued;
 	struct nfqnl_instance *queue;
-	struct sk_buff *skb, *segs;
+	struct sk_buff *skb, *segs, *nskb;
 	int err = -ENOBUFS;
 	struct net *net = entry->state.net;
 	struct nfnl_queue_net *q = nfnl_queue_pernet(net);
@@ -815,8 +815,7 @@ nfqnl_enqueue_packet(struct nf_queue_entry *entry, unsigned int queuenum)
 		goto out_err;
 	queued = 0;
 	err = 0;
-	do {
-		struct sk_buff *nskb = segs->next;
+	skb_list_walk_safe(segs, segs, nskb) {
 		if (err == 0)
 			err = __nfqnl_enqueue_packet_gso(net, queue,
 							segs, entry);
@@ -824,8 +823,7 @@ nfqnl_enqueue_packet(struct nf_queue_entry *entry, unsigned int queuenum)
 			queued++;
 		else
 			kfree_skb(segs);
-		segs = nskb;
-	} while (segs);
+	}
 
 	if (queued) {
 		if (err) /* some segments are already queued */
-- 
cgit v1.2.3-59-g8ed1b


From 9f3ef3d7026fe0aef5c2d649e21b8c71c17d8fc2 Mon Sep 17 00:00:00 2001
From: "Jason A. Donenfeld" <Jason@zx2c4.com>
Date: Mon, 13 Jan 2020 18:42:33 -0500
Subject: net: mac80211: use skb_list_walk_safe helper for gso segments

This is a conversion case for the new function, keeping the flow of the
existing code as intact as possible. We also switch over to using
skb_mark_not_on_list instead of a null write to skb->next.

Finally, this code appeared to have a memory leak in the case where
header building fails before the last gso segment. In that case, the
remaining segments are not freed. So this commit also adds the proper
kfree_skb_list call for the remainder of the skbs.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/mac80211/tx.c | 13 +++++--------
 1 file changed, 5 insertions(+), 8 deletions(-)

diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index a8a7306a1f56..4bd1faf4f779 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -3949,18 +3949,15 @@ void __ieee80211_subif_start_xmit(struct sk_buff *skb,
 		}
 	}
 
-	next = skb;
-	while (next) {
-		skb = next;
-		next = skb->next;
-
-		skb->prev = NULL;
-		skb->next = NULL;
+	skb_list_walk_safe(skb, skb, next) {
+		skb_mark_not_on_list(skb);
 
 		skb = ieee80211_build_hdr(sdata, skb, info_flags,
 					  sta, ctrl_flags);
-		if (IS_ERR(skb))
+		if (IS_ERR(skb)) {
+			kfree_skb_list(next);
 			goto out;
+		}
 
 		ieee80211_tx_stats(dev, skb->len);
 
-- 
cgit v1.2.3-59-g8ed1b


From a4d35e77357c72acd52d6b1e7093b750b384b7cc Mon Sep 17 00:00:00 2001
From: Tian Tao <tiantao6@hisilicon.com>
Date: Tue, 14 Jan 2020 10:57:39 +0800
Subject: nfc: No need to set .owner platform_driver_register

the i2c_add_driver will set the .owner to THIS_MODULE

Signed-off-by: Tian Tao <tiantao6@hisilicon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/nfc/pn533/i2c.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/nfc/pn533/i2c.c b/drivers/nfc/pn533/i2c.c
index 7507176cca0a..0207e66cee21 100644
--- a/drivers/nfc/pn533/i2c.c
+++ b/drivers/nfc/pn533/i2c.c
@@ -274,7 +274,6 @@ MODULE_DEVICE_TABLE(i2c, pn533_i2c_id_table);
 static struct i2c_driver pn533_i2c_driver = {
 	.driver = {
 		   .name = PN533_I2C_DRIVER_NAME,
-		   .owner = THIS_MODULE,
 		   .of_match_table = of_match_ptr(of_pn533_i2c_match),
 		  },
 	.probe = pn533_i2c_probe,
-- 
cgit v1.2.3-59-g8ed1b


From 7786a1af2a6bceb07860ec720e74714004438834 Mon Sep 17 00:00:00 2001
From: Niu Xilei <niu_xilei@163.com>
Date: Tue, 14 Jan 2020 11:12:29 +0800
Subject: pktgen: Allow configuration of IPv6 source address range

Pktgen can use only one IPv6 source address from output device or src6
command setting. In pressure test we need create lots of sessions more
than 65535. So add src6_min and src6_max command to set the range.

Signed-off-by: Niu Xilei <niu_xilei@163.com>

Changes since v3:
 - function set_src_in6_addr use static instead of static inline
 - precompute min_in6_l,min_in6_h,max_in6_h,max_in6_l in setup time
Changes since v2:
 - reword subject line
Changes since v1:
 - only create IPv6 source address over least significant 64 bit range

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/pktgen.c | 98 +++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 98 insertions(+)

diff --git a/net/core/pktgen.c b/net/core/pktgen.c
index 294bfcf0ce0e..890be1b4877e 100644
--- a/net/core/pktgen.c
+++ b/net/core/pktgen.c
@@ -323,6 +323,10 @@ struct pktgen_dev {
 	struct in6_addr max_in6_daddr;
 	struct in6_addr min_in6_saddr;
 	struct in6_addr max_in6_saddr;
+	u64  max_in6_h;
+	u64  max_in6_l;
+	u64  min_in6_h;
+	u64  min_in6_l;
 
 	/* If we're doing ranges, random or incremental, then this
 	 * defines the min/max for those ranges.
@@ -1355,6 +1359,59 @@ static ssize_t pktgen_if_write(struct file *file,
 		sprintf(pg_result, "OK: dst6_max=%s", buf);
 		return count;
 	}
+	if (!strcmp(name, "src6_min")) {
+		len = strn_len(&user_buffer[i], sizeof(buf) - 1);
+		if (len < 0)
+			return len;
+
+		pkt_dev->flags |= F_IPV6;
+
+		if (copy_from_user(buf, &user_buffer[i], len))
+			return -EFAULT;
+		buf[len] = 0;
+
+		in6_pton(buf, -1, pkt_dev->min_in6_saddr.s6_addr, -1, NULL);
+		snprintf(buf, sizeof(buf), "%pI6c", &pkt_dev->min_in6_saddr);
+
+		memcpy(&pkt_dev->min_in6_h, pkt_dev->min_in6_saddr.s6_addr, 8);
+		memcpy(&pkt_dev->min_in6_l, pkt_dev->min_in6_saddr.s6_addr + 8, 8);
+		pkt_dev->min_in6_h = be64_to_cpu(pkt_dev->min_in6_h);
+		pkt_dev->min_in6_l = be64_to_cpu(pkt_dev->min_in6_l);
+
+		pkt_dev->cur_in6_saddr = pkt_dev->min_in6_saddr;
+		if (debug)
+			pr_debug("src6_min set to: %s\n", buf);
+
+		i += len;
+		sprintf(pg_result, "OK: src6_min=%s", buf);
+		return count;
+	}
+	if (!strcmp(name, "src6_max")) {
+		len = strn_len(&user_buffer[i], sizeof(buf) - 1);
+		if (len < 0)
+			return len;
+
+		pkt_dev->flags |= F_IPV6;
+
+		if (copy_from_user(buf, &user_buffer[i], len))
+			return -EFAULT;
+		buf[len] = 0;
+
+		in6_pton(buf, -1, pkt_dev->max_in6_saddr.s6_addr, -1, NULL);
+		snprintf(buf, sizeof(buf), "%pI6c", &pkt_dev->max_in6_saddr);
+
+		memcpy(&pkt_dev->max_in6_h, pkt_dev->max_in6_saddr.s6_addr, 8);
+		memcpy(&pkt_dev->max_in6_l, pkt_dev->max_in6_saddr.s6_addr + 8, 8);
+		pkt_dev->max_in6_h = be64_to_cpu(pkt_dev->max_in6_h);
+		pkt_dev->max_in6_l = be64_to_cpu(pkt_dev->max_in6_l);
+
+		if (debug)
+			pr_debug("src6_max set to: %s\n", buf);
+
+		i += len;
+		sprintf(pg_result, "OK: src6_max=%s", buf);
+		return count;
+	}
 	if (!strcmp(name, "src6")) {
 		len = strn_len(&user_buffer[i], sizeof(buf) - 1);
 		if (len < 0)
@@ -2286,6 +2343,45 @@ static void set_cur_queue_map(struct pktgen_dev *pkt_dev)
 	pkt_dev->cur_queue_map  = pkt_dev->cur_queue_map % pkt_dev->odev->real_num_tx_queues;
 }
 
+/* generate ipv6 source addr */
+static void set_src_in6_addr(struct pktgen_dev *pkt_dev)
+{
+	u64 min6, max6, rand, i;
+	struct in6_addr addr6;
+	__be64 addr_l, *t;
+
+	min6 = pkt_dev->min_in6_l;
+	max6 = pkt_dev->max_in6_l;
+
+	/* only generate source address in least significant 64 bits range
+	 * most significant 64 bits must be equal
+	 */
+	if (pkt_dev->max_in6_h != pkt_dev->min_in6_h || min6 >= max6)
+		return;
+
+	addr6 = pkt_dev->min_in6_saddr;
+	t = (__be64 *)addr6.s6_addr + 1;
+
+	if (pkt_dev->flags & F_IPSRC_RND) {
+		do {
+			prandom_bytes(&rand, sizeof(rand));
+			rand = rand % (max6 - min6) + min6;
+			addr_l = cpu_to_be64(rand);
+			memcpy(t, &addr_l, 8);
+		} while (ipv6_addr_loopback(&addr6) ||
+			 ipv6_addr_v4mapped(&addr6) ||
+			 ipv6_addr_is_multicast(&addr6));
+	} else {
+		addr6 = pkt_dev->cur_in6_saddr;
+		i = be64_to_cpu(*t);
+		if (++i > max6)
+			i = min6;
+		addr_l = cpu_to_be64(i);
+		memcpy(t, &addr_l, 8);
+	}
+	pkt_dev->cur_in6_saddr = addr6;
+}
+
 /* Increment/randomize headers according to flags and current values
  * for IP src/dest, UDP src/dst port, MAC-Addr src/dst
  */
@@ -2454,6 +2550,8 @@ static void mod_cur_headers(struct pktgen_dev *pkt_dev)
 		}
 	} else {		/* IPV6 * */
 
+		set_src_in6_addr(pkt_dev);
+
 		if (!ipv6_addr_any(&pkt_dev->min_in6_daddr)) {
 			int i;
 
-- 
cgit v1.2.3-59-g8ed1b


From cb6530b99fafea6c0636c4640bd21301d12cdbc9 Mon Sep 17 00:00:00 2001
From: Bjorn Andersson <bjorn.andersson@linaro.org>
Date: Mon, 13 Jan 2020 23:56:59 -0800
Subject: net: qrtr: Move resume-tx transmission to recvmsg

The confirm-rx bit is used to implement a per port flow control, in
order to make sure that no messages are dropped due to resource
exhaustion. Move the resume-tx transmission to recvmsg to only confirm
messages as they are consumed by the application.

Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/qrtr/qrtr.c | 60 +++++++++++++++++++++++++++++++--------------------------
 1 file changed, 33 insertions(+), 27 deletions(-)

diff --git a/net/qrtr/qrtr.c b/net/qrtr/qrtr.c
index 3d24d45be5f4..60f73b44c889 100644
--- a/net/qrtr/qrtr.c
+++ b/net/qrtr/qrtr.c
@@ -362,22 +362,11 @@ static void qrtr_port_put(struct qrtr_sock *ipc);
 static void qrtr_node_rx_work(struct work_struct *work)
 {
 	struct qrtr_node *node = container_of(work, struct qrtr_node, work);
-	struct qrtr_ctrl_pkt *pkt;
-	struct sockaddr_qrtr dst;
-	struct sockaddr_qrtr src;
 	struct sk_buff *skb;
 
 	while ((skb = skb_dequeue(&node->rx_queue)) != NULL) {
 		struct qrtr_sock *ipc;
-		struct qrtr_cb *cb;
-		int confirm;
-
-		cb = (struct qrtr_cb *)skb->cb;
-		src.sq_node = cb->src_node;
-		src.sq_port = cb->src_port;
-		dst.sq_node = cb->dst_node;
-		dst.sq_port = cb->dst_port;
-		confirm = !!cb->confirm_rx;
+		struct qrtr_cb *cb = (struct qrtr_cb *)skb->cb;
 
 		qrtr_node_assign(node, cb->src_node);
 
@@ -390,20 +379,6 @@ static void qrtr_node_rx_work(struct work_struct *work)
 
 			qrtr_port_put(ipc);
 		}
-
-		if (confirm) {
-			skb = qrtr_alloc_ctrl_packet(&pkt);
-			if (!skb)
-				break;
-
-			pkt->cmd = cpu_to_le32(QRTR_TYPE_RESUME_TX);
-			pkt->client.node = cpu_to_le32(dst.sq_node);
-			pkt->client.port = cpu_to_le32(dst.sq_port);
-
-			if (qrtr_node_enqueue(node, skb, QRTR_TYPE_RESUME_TX,
-					      &dst, &src))
-				break;
-		}
 	}
 }
 
@@ -816,6 +791,34 @@ out_node:
 	return rc;
 }
 
+static int qrtr_send_resume_tx(struct qrtr_cb *cb)
+{
+	struct sockaddr_qrtr remote = { AF_QIPCRTR, cb->src_node, cb->src_port };
+	struct sockaddr_qrtr local = { AF_QIPCRTR, cb->dst_node, cb->dst_port };
+	struct qrtr_ctrl_pkt *pkt;
+	struct qrtr_node *node;
+	struct sk_buff *skb;
+	int ret;
+
+	node = qrtr_node_lookup(remote.sq_node);
+	if (!node)
+		return -EINVAL;
+
+	skb = qrtr_alloc_ctrl_packet(&pkt);
+	if (!skb)
+		return -ENOMEM;
+
+	pkt->cmd = cpu_to_le32(QRTR_TYPE_RESUME_TX);
+	pkt->client.node = cpu_to_le32(cb->dst_node);
+	pkt->client.port = cpu_to_le32(cb->dst_port);
+
+	ret = qrtr_node_enqueue(node, skb, QRTR_TYPE_RESUME_TX, &local, &remote);
+
+	qrtr_node_release(node);
+
+	return ret;
+}
+
 static int qrtr_recvmsg(struct socket *sock, struct msghdr *msg,
 			size_t size, int flags)
 {
@@ -838,6 +841,7 @@ static int qrtr_recvmsg(struct socket *sock, struct msghdr *msg,
 		release_sock(sk);
 		return rc;
 	}
+	cb = (struct qrtr_cb *)skb->cb;
 
 	copied = skb->len;
 	if (copied > size) {
@@ -851,7 +855,6 @@ static int qrtr_recvmsg(struct socket *sock, struct msghdr *msg,
 	rc = copied;
 
 	if (addr) {
-		cb = (struct qrtr_cb *)skb->cb;
 		addr->sq_family = AF_QIPCRTR;
 		addr->sq_node = cb->src_node;
 		addr->sq_port = cb->src_port;
@@ -859,6 +862,9 @@ static int qrtr_recvmsg(struct socket *sock, struct msghdr *msg,
 	}
 
 out:
+	if (cb->confirm_rx)
+		qrtr_send_resume_tx(cb);
+
 	skb_free_datagram(sk, skb);
 	release_sock(sk);
 
-- 
cgit v1.2.3-59-g8ed1b


From 5fdeb0d372ab33b4175043a2a4a1730239a217f1 Mon Sep 17 00:00:00 2001
From: Bjorn Andersson <bjorn.andersson@linaro.org>
Date: Mon, 13 Jan 2020 23:57:00 -0800
Subject: net: qrtr: Implement outgoing flow control

In order to prevent overconsumption of resources on the remote side QRTR
implements a flow control mechanism.

The mechanism works by the sender keeping track of the number of
outstanding unconfirmed messages that has been transmitted to a
particular node/port pair.

Upon count reaching a low watermark (L) the confirm_rx bit is set in the
outgoing message and when the count reaching a high watermark (H)
transmission will be blocked upon the reception of a resume_tx message
from the remote, that resets the counter to 0.

This guarantees that there will be at most 2H - L messages in flight.
Values chosen for L and H are 5 and 10 respectively.

Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/qrtr/qrtr.c | 194 ++++++++++++++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 187 insertions(+), 7 deletions(-)

diff --git a/net/qrtr/qrtr.c b/net/qrtr/qrtr.c
index 60f73b44c889..2b7f462476c0 100644
--- a/net/qrtr/qrtr.c
+++ b/net/qrtr/qrtr.c
@@ -8,6 +8,7 @@
 #include <linux/qrtr.h>
 #include <linux/termios.h>	/* For TIOCINQ/OUTQ */
 #include <linux/numa.h>
+#include <linux/wait.h>
 
 #include <net/sock.h>
 
@@ -113,6 +114,8 @@ static DEFINE_MUTEX(qrtr_port_lock);
  * @ep: endpoint
  * @ref: reference count for node
  * @nid: node id
+ * @qrtr_tx_flow: tree of qrtr_tx_flow, keyed by node << 32 | port
+ * @qrtr_tx_lock: lock for qrtr_tx_flow inserts
  * @rx_queue: receive queue
  * @work: scheduled work struct for recv work
  * @item: list item for broadcast list
@@ -123,11 +126,29 @@ struct qrtr_node {
 	struct kref ref;
 	unsigned int nid;
 
+	struct radix_tree_root qrtr_tx_flow;
+	struct mutex qrtr_tx_lock; /* for qrtr_tx_flow */
+
 	struct sk_buff_head rx_queue;
 	struct work_struct work;
 	struct list_head item;
 };
 
+/**
+ * struct qrtr_tx_flow - tx flow control
+ * @resume_tx: waiters for a resume tx from the remote
+ * @pending: number of waiting senders
+ * @tx_failed: indicates that a message with confirm_rx flag was lost
+ */
+struct qrtr_tx_flow {
+	struct wait_queue_head resume_tx;
+	int pending;
+	int tx_failed;
+};
+
+#define QRTR_TX_FLOW_HIGH	10
+#define QRTR_TX_FLOW_LOW	5
+
 static int qrtr_local_enqueue(struct qrtr_node *node, struct sk_buff *skb,
 			      int type, struct sockaddr_qrtr *from,
 			      struct sockaddr_qrtr *to);
@@ -143,6 +164,8 @@ static int qrtr_bcast_enqueue(struct qrtr_node *node, struct sk_buff *skb,
 static void __qrtr_node_release(struct kref *kref)
 {
 	struct qrtr_node *node = container_of(kref, struct qrtr_node, ref);
+	struct radix_tree_iter iter;
+	void __rcu **slot;
 
 	if (node->nid != QRTR_EP_NID_AUTO)
 		radix_tree_delete(&qrtr_nodes, node->nid);
@@ -152,6 +175,12 @@ static void __qrtr_node_release(struct kref *kref)
 
 	cancel_work_sync(&node->work);
 	skb_queue_purge(&node->rx_queue);
+
+	/* Free tx flow counters */
+	radix_tree_for_each_slot(slot, &node->qrtr_tx_flow, &iter, 0) {
+		radix_tree_iter_delete(&node->qrtr_tx_flow, &iter, slot);
+		kfree(*slot);
+	}
 	kfree(node);
 }
 
@@ -171,6 +200,126 @@ static void qrtr_node_release(struct qrtr_node *node)
 	kref_put_mutex(&node->ref, __qrtr_node_release, &qrtr_node_lock);
 }
 
+/**
+ * qrtr_tx_resume() - reset flow control counter
+ * @node:	qrtr_node that the QRTR_TYPE_RESUME_TX packet arrived on
+ * @skb:	resume_tx packet
+ */
+static void qrtr_tx_resume(struct qrtr_node *node, struct sk_buff *skb)
+{
+	struct qrtr_ctrl_pkt *pkt = (struct qrtr_ctrl_pkt *)skb->data;
+	u64 remote_node = le32_to_cpu(pkt->client.node);
+	u32 remote_port = le32_to_cpu(pkt->client.port);
+	struct qrtr_tx_flow *flow;
+	unsigned long key;
+
+	key = remote_node << 32 | remote_port;
+
+	rcu_read_lock();
+	flow = radix_tree_lookup(&node->qrtr_tx_flow, key);
+	rcu_read_unlock();
+	if (flow) {
+		spin_lock(&flow->resume_tx.lock);
+		flow->pending = 0;
+		spin_unlock(&flow->resume_tx.lock);
+		wake_up_interruptible_all(&flow->resume_tx);
+	}
+
+	consume_skb(skb);
+}
+
+/**
+ * qrtr_tx_wait() - flow control for outgoing packets
+ * @node:	qrtr_node that the packet is to be send to
+ * @dest_node:	node id of the destination
+ * @dest_port:	port number of the destination
+ * @type:	type of message
+ *
+ * The flow control scheme is based around the low and high "watermarks". When
+ * the low watermark is passed the confirm_rx flag is set on the outgoing
+ * message, which will trigger the remote to send a control message of the type
+ * QRTR_TYPE_RESUME_TX to reset the counter. If the high watermark is hit
+ * further transmision should be paused.
+ *
+ * Return: 1 if confirm_rx should be set, 0 otherwise or errno failure
+ */
+static int qrtr_tx_wait(struct qrtr_node *node, int dest_node, int dest_port,
+			int type)
+{
+	unsigned long key = (u64)dest_node << 32 | dest_port;
+	struct qrtr_tx_flow *flow;
+	int confirm_rx = 0;
+	int ret;
+
+	/* Never set confirm_rx on non-data packets */
+	if (type != QRTR_TYPE_DATA)
+		return 0;
+
+	mutex_lock(&node->qrtr_tx_lock);
+	flow = radix_tree_lookup(&node->qrtr_tx_flow, key);
+	if (!flow) {
+		flow = kzalloc(sizeof(*flow), GFP_KERNEL);
+		if (flow) {
+			init_waitqueue_head(&flow->resume_tx);
+			radix_tree_insert(&node->qrtr_tx_flow, key, flow);
+		}
+	}
+	mutex_unlock(&node->qrtr_tx_lock);
+
+	/* Set confirm_rx if we where unable to find and allocate a flow */
+	if (!flow)
+		return 1;
+
+	spin_lock_irq(&flow->resume_tx.lock);
+	ret = wait_event_interruptible_locked_irq(flow->resume_tx,
+						  flow->pending < QRTR_TX_FLOW_HIGH ||
+						  flow->tx_failed ||
+						  !node->ep);
+	if (ret < 0) {
+		confirm_rx = ret;
+	} else if (!node->ep) {
+		confirm_rx = -EPIPE;
+	} else if (flow->tx_failed) {
+		flow->tx_failed = 0;
+		confirm_rx = 1;
+	} else {
+		flow->pending++;
+		confirm_rx = flow->pending == QRTR_TX_FLOW_LOW;
+	}
+	spin_unlock_irq(&flow->resume_tx.lock);
+
+	return confirm_rx;
+}
+
+/**
+ * qrtr_tx_flow_failed() - flag that tx of confirm_rx flagged messages failed
+ * @node:	qrtr_node that the packet is to be send to
+ * @dest_node:	node id of the destination
+ * @dest_port:	port number of the destination
+ *
+ * Signal that the transmission of a message with confirm_rx flag failed. The
+ * flow's "pending" counter will keep incrementing towards QRTR_TX_FLOW_HIGH,
+ * at which point transmission would stall forever waiting for the resume TX
+ * message associated with the dropped confirm_rx message.
+ * Work around this by marking the flow as having a failed transmission and
+ * cause the next transmission attempt to be sent with the confirm_rx.
+ */
+static void qrtr_tx_flow_failed(struct qrtr_node *node, int dest_node,
+				int dest_port)
+{
+	unsigned long key = (u64)dest_node << 32 | dest_port;
+	struct qrtr_tx_flow *flow;
+
+	rcu_read_lock();
+	flow = radix_tree_lookup(&node->qrtr_tx_flow, key);
+	rcu_read_unlock();
+	if (flow) {
+		spin_lock_irq(&flow->resume_tx.lock);
+		flow->tx_failed = 1;
+		spin_unlock_irq(&flow->resume_tx.lock);
+	}
+}
+
 /* Pass an outgoing packet socket buffer to the endpoint driver. */
 static int qrtr_node_enqueue(struct qrtr_node *node, struct sk_buff *skb,
 			     int type, struct sockaddr_qrtr *from,
@@ -179,6 +328,13 @@ static int qrtr_node_enqueue(struct qrtr_node *node, struct sk_buff *skb,
 	struct qrtr_hdr_v1 *hdr;
 	size_t len = skb->len;
 	int rc = -ENODEV;
+	int confirm_rx;
+
+	confirm_rx = qrtr_tx_wait(node, to->sq_node, to->sq_port, type);
+	if (confirm_rx < 0) {
+		kfree_skb(skb);
+		return confirm_rx;
+	}
 
 	hdr = skb_push(skb, sizeof(*hdr));
 	hdr->version = cpu_to_le32(QRTR_PROTO_VER_1);
@@ -194,7 +350,7 @@ static int qrtr_node_enqueue(struct qrtr_node *node, struct sk_buff *skb,
 	}
 
 	hdr->size = cpu_to_le32(len);
-	hdr->confirm_rx = 0;
+	hdr->confirm_rx = !!confirm_rx;
 
 	skb_put_padto(skb, ALIGN(len, 4) + sizeof(*hdr));
 
@@ -205,6 +361,11 @@ static int qrtr_node_enqueue(struct qrtr_node *node, struct sk_buff *skb,
 		kfree_skb(skb);
 	mutex_unlock(&node->ep_lock);
 
+	/* Need to ensure that a subsequent message carries the otherwise lost
+	 * confirm_rx flag if we dropped this one */
+	if (rc && confirm_rx)
+		qrtr_tx_flow_failed(node, to->sq_node, to->sq_port);
+
 	return rc;
 }
 
@@ -311,7 +472,8 @@ int qrtr_endpoint_post(struct qrtr_endpoint *ep, const void *data, size_t len)
 	if (len != ALIGN(size, 4) + hdrlen)
 		goto err;
 
-	if (cb->dst_port != QRTR_PORT_CTRL && cb->type != QRTR_TYPE_DATA)
+	if (cb->dst_port != QRTR_PORT_CTRL && cb->type != QRTR_TYPE_DATA &&
+	    cb->type != QRTR_TYPE_RESUME_TX)
 		goto err;
 
 	skb_put_data(skb, data + hdrlen, size);
@@ -370,14 +532,18 @@ static void qrtr_node_rx_work(struct work_struct *work)
 
 		qrtr_node_assign(node, cb->src_node);
 
-		ipc = qrtr_port_lookup(cb->dst_port);
-		if (!ipc) {
-			kfree_skb(skb);
+		if (cb->type == QRTR_TYPE_RESUME_TX) {
+			qrtr_tx_resume(node, skb);
 		} else {
-			if (sock_queue_rcv_skb(&ipc->sk, skb))
+			ipc = qrtr_port_lookup(cb->dst_port);
+			if (!ipc) {
 				kfree_skb(skb);
+			} else {
+				if (sock_queue_rcv_skb(&ipc->sk, skb))
+					kfree_skb(skb);
 
-			qrtr_port_put(ipc);
+				qrtr_port_put(ipc);
+			}
 		}
 	}
 }
@@ -408,6 +574,9 @@ int qrtr_endpoint_register(struct qrtr_endpoint *ep, unsigned int nid)
 	node->nid = QRTR_EP_NID_AUTO;
 	node->ep = ep;
 
+	INIT_RADIX_TREE(&node->qrtr_tx_flow, GFP_KERNEL);
+	mutex_init(&node->qrtr_tx_lock);
+
 	qrtr_node_assign(node, nid);
 
 	mutex_lock(&qrtr_node_lock);
@@ -428,8 +597,11 @@ void qrtr_endpoint_unregister(struct qrtr_endpoint *ep)
 	struct qrtr_node *node = ep->node;
 	struct sockaddr_qrtr src = {AF_QIPCRTR, node->nid, QRTR_PORT_CTRL};
 	struct sockaddr_qrtr dst = {AF_QIPCRTR, qrtr_local_nid, QRTR_PORT_CTRL};
+	struct radix_tree_iter iter;
 	struct qrtr_ctrl_pkt *pkt;
+	struct qrtr_tx_flow *flow;
 	struct sk_buff *skb;
+	void __rcu **slot;
 
 	mutex_lock(&node->ep_lock);
 	node->ep = NULL;
@@ -442,6 +614,14 @@ void qrtr_endpoint_unregister(struct qrtr_endpoint *ep)
 		qrtr_local_enqueue(NULL, skb, QRTR_TYPE_BYE, &src, &dst);
 	}
 
+	/* Wake up any transmitters waiting for resume-tx from the node */
+	mutex_lock(&node->qrtr_tx_lock);
+	radix_tree_for_each_slot(slot, &node->qrtr_tx_flow, &iter, 0) {
+		flow = *slot;
+		wake_up_interruptible_all(&flow->resume_tx);
+	}
+	mutex_unlock(&node->qrtr_tx_lock);
+
 	qrtr_node_release(node);
 	ep->node = NULL;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 0a7e0d0ef05440db03c3199e84d228db943b237f Mon Sep 17 00:00:00 2001
From: Bjorn Andersson <bjorn.andersson@linaro.org>
Date: Mon, 13 Jan 2020 23:57:01 -0800
Subject: net: qrtr: Migrate node lookup tree to spinlock

Move operations on the qrtr_nodes radix tree under a separate spinlock
and make the qrtr_nodes tree GFP_ATOMIC, to allow operation from atomic
context in a subsequent patch.

Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/qrtr/qrtr.c | 20 ++++++++++++++------
 1 file changed, 14 insertions(+), 6 deletions(-)

diff --git a/net/qrtr/qrtr.c b/net/qrtr/qrtr.c
index 2b7f462476c0..ca802798e19e 100644
--- a/net/qrtr/qrtr.c
+++ b/net/qrtr/qrtr.c
@@ -8,6 +8,7 @@
 #include <linux/qrtr.h>
 #include <linux/termios.h>	/* For TIOCINQ/OUTQ */
 #include <linux/numa.h>
+#include <linux/spinlock.h>
 #include <linux/wait.h>
 
 #include <net/sock.h>
@@ -98,10 +99,11 @@ static inline struct qrtr_sock *qrtr_sk(struct sock *sk)
 static unsigned int qrtr_local_nid = NUMA_NO_NODE;
 
 /* for node ids */
-static RADIX_TREE(qrtr_nodes, GFP_KERNEL);
+static RADIX_TREE(qrtr_nodes, GFP_ATOMIC);
+static DEFINE_SPINLOCK(qrtr_nodes_lock);
 /* broadcast list */
 static LIST_HEAD(qrtr_all_nodes);
-/* lock for qrtr_nodes, qrtr_all_nodes and node reference */
+/* lock for qrtr_all_nodes and node reference */
 static DEFINE_MUTEX(qrtr_node_lock);
 
 /* local port allocation management */
@@ -165,10 +167,13 @@ static void __qrtr_node_release(struct kref *kref)
 {
 	struct qrtr_node *node = container_of(kref, struct qrtr_node, ref);
 	struct radix_tree_iter iter;
+	unsigned long flags;
 	void __rcu **slot;
 
+	spin_lock_irqsave(&qrtr_nodes_lock, flags);
 	if (node->nid != QRTR_EP_NID_AUTO)
 		radix_tree_delete(&qrtr_nodes, node->nid);
+	spin_unlock_irqrestore(&qrtr_nodes_lock, flags);
 
 	list_del(&node->item);
 	mutex_unlock(&qrtr_node_lock);
@@ -376,11 +381,12 @@ static int qrtr_node_enqueue(struct qrtr_node *node, struct sk_buff *skb,
 static struct qrtr_node *qrtr_node_lookup(unsigned int nid)
 {
 	struct qrtr_node *node;
+	unsigned long flags;
 
-	mutex_lock(&qrtr_node_lock);
+	spin_lock_irqsave(&qrtr_nodes_lock, flags);
 	node = radix_tree_lookup(&qrtr_nodes, nid);
 	node = qrtr_node_acquire(node);
-	mutex_unlock(&qrtr_node_lock);
+	spin_unlock_irqrestore(&qrtr_nodes_lock, flags);
 
 	return node;
 }
@@ -392,13 +398,15 @@ static struct qrtr_node *qrtr_node_lookup(unsigned int nid)
  */
 static void qrtr_node_assign(struct qrtr_node *node, unsigned int nid)
 {
+	unsigned long flags;
+
 	if (node->nid != QRTR_EP_NID_AUTO || nid == QRTR_EP_NID_AUTO)
 		return;
 
-	mutex_lock(&qrtr_node_lock);
+	spin_lock_irqsave(&qrtr_nodes_lock, flags);
 	radix_tree_insert(&qrtr_nodes, nid, node);
 	node->nid = nid;
-	mutex_unlock(&qrtr_node_lock);
+	spin_unlock_irqrestore(&qrtr_nodes_lock, flags);
 }
 
 /**
-- 
cgit v1.2.3-59-g8ed1b


From f16a4b26f31f95dddb12cf3c2390906a735203ae Mon Sep 17 00:00:00 2001
From: Bjorn Andersson <bjorn.andersson@linaro.org>
Date: Mon, 13 Jan 2020 23:57:02 -0800
Subject: net: qrtr: Make qrtr_port_lookup() use RCU

The important part of qrtr_port_lookup() wrt synchronization is that the
function returns a reference counted struct qrtr_sock, or fail.

As such we need only to ensure that an decrement of the object's
refcount happens inbetween the finding of the object in the idr and
qrtr_port_lookup()'s own increment of the object.

By using RCU and putting a synchronization point after we remove the
mapping from the idr, but before it can be released we achieve this -
with the benefit of not having to hold the mutex in qrtr_port_lookup().

Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/qrtr/qrtr.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/net/qrtr/qrtr.c b/net/qrtr/qrtr.c
index ca802798e19e..0fe08f060b8c 100644
--- a/net/qrtr/qrtr.c
+++ b/net/qrtr/qrtr.c
@@ -646,11 +646,11 @@ static struct qrtr_sock *qrtr_port_lookup(int port)
 	if (port == QRTR_PORT_CTRL)
 		port = 0;
 
-	mutex_lock(&qrtr_port_lock);
+	rcu_read_lock();
 	ipc = idr_find(&qrtr_ports, port);
 	if (ipc)
 		sock_hold(&ipc->sk);
-	mutex_unlock(&qrtr_port_lock);
+	rcu_read_unlock();
 
 	return ipc;
 }
@@ -692,6 +692,10 @@ static void qrtr_port_remove(struct qrtr_sock *ipc)
 	mutex_lock(&qrtr_port_lock);
 	idr_remove(&qrtr_ports, port);
 	mutex_unlock(&qrtr_port_lock);
+
+	/* Ensure that if qrtr_port_lookup() did enter the RCU read section we
+	 * wait for it to up increment the refcount */
+	synchronize_rcu();
 }
 
 /* Assign port number to socket.
-- 
cgit v1.2.3-59-g8ed1b


From e04df98adf7d7d946aa927822ccec24680501662 Mon Sep 17 00:00:00 2001
From: Bjorn Andersson <bjorn.andersson@linaro.org>
Date: Mon, 13 Jan 2020 23:57:03 -0800
Subject: net: qrtr: Remove receive worker

Rather than enqueuing messages and scheduling a worker to deliver them
to the individual sockets we can now, thanks to the previous work, move
this directly into the endpoint callback.

This saves us a context switch per incoming message and removes the
possibility of an opportunistic suspend to happen between the message is
coming from the endpoint until it ends up in the socket's receive
buffer.

Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/qrtr/qrtr.c | 57 +++++++++++++++++----------------------------------------
 1 file changed, 17 insertions(+), 40 deletions(-)

diff --git a/net/qrtr/qrtr.c b/net/qrtr/qrtr.c
index 0fe08f060b8c..5a8e42ad1504 100644
--- a/net/qrtr/qrtr.c
+++ b/net/qrtr/qrtr.c
@@ -119,7 +119,6 @@ static DEFINE_MUTEX(qrtr_port_lock);
  * @qrtr_tx_flow: tree of qrtr_tx_flow, keyed by node << 32 | port
  * @qrtr_tx_lock: lock for qrtr_tx_flow inserts
  * @rx_queue: receive queue
- * @work: scheduled work struct for recv work
  * @item: list item for broadcast list
  */
 struct qrtr_node {
@@ -132,7 +131,6 @@ struct qrtr_node {
 	struct mutex qrtr_tx_lock; /* for qrtr_tx_flow */
 
 	struct sk_buff_head rx_queue;
-	struct work_struct work;
 	struct list_head item;
 };
 
@@ -157,6 +155,8 @@ static int qrtr_local_enqueue(struct qrtr_node *node, struct sk_buff *skb,
 static int qrtr_bcast_enqueue(struct qrtr_node *node, struct sk_buff *skb,
 			      int type, struct sockaddr_qrtr *from,
 			      struct sockaddr_qrtr *to);
+static struct qrtr_sock *qrtr_port_lookup(int port);
+static void qrtr_port_put(struct qrtr_sock *ipc);
 
 /* Release node resources and free the node.
  *
@@ -178,7 +178,6 @@ static void __qrtr_node_release(struct kref *kref)
 	list_del(&node->item);
 	mutex_unlock(&qrtr_node_lock);
 
-	cancel_work_sync(&node->work);
 	skb_queue_purge(&node->rx_queue);
 
 	/* Free tx flow counters */
@@ -422,6 +421,7 @@ int qrtr_endpoint_post(struct qrtr_endpoint *ep, const void *data, size_t len)
 	struct qrtr_node *node = ep->node;
 	const struct qrtr_hdr_v1 *v1;
 	const struct qrtr_hdr_v2 *v2;
+	struct qrtr_sock *ipc;
 	struct sk_buff *skb;
 	struct qrtr_cb *cb;
 	unsigned int size;
@@ -486,8 +486,20 @@ int qrtr_endpoint_post(struct qrtr_endpoint *ep, const void *data, size_t len)
 
 	skb_put_data(skb, data + hdrlen, size);
 
-	skb_queue_tail(&node->rx_queue, skb);
-	schedule_work(&node->work);
+	qrtr_node_assign(node, cb->src_node);
+
+	if (cb->type == QRTR_TYPE_RESUME_TX) {
+		qrtr_tx_resume(node, skb);
+	} else {
+		ipc = qrtr_port_lookup(cb->dst_port);
+		if (!ipc)
+			goto err;
+
+		if (sock_queue_rcv_skb(&ipc->sk, skb))
+			goto err;
+
+		qrtr_port_put(ipc);
+	}
 
 	return 0;
 
@@ -522,40 +534,6 @@ static struct sk_buff *qrtr_alloc_ctrl_packet(struct qrtr_ctrl_pkt **pkt)
 	return skb;
 }
 
-static struct qrtr_sock *qrtr_port_lookup(int port);
-static void qrtr_port_put(struct qrtr_sock *ipc);
-
-/* Handle and route a received packet.
- *
- * This will auto-reply with resume-tx packet as necessary.
- */
-static void qrtr_node_rx_work(struct work_struct *work)
-{
-	struct qrtr_node *node = container_of(work, struct qrtr_node, work);
-	struct sk_buff *skb;
-
-	while ((skb = skb_dequeue(&node->rx_queue)) != NULL) {
-		struct qrtr_sock *ipc;
-		struct qrtr_cb *cb = (struct qrtr_cb *)skb->cb;
-
-		qrtr_node_assign(node, cb->src_node);
-
-		if (cb->type == QRTR_TYPE_RESUME_TX) {
-			qrtr_tx_resume(node, skb);
-		} else {
-			ipc = qrtr_port_lookup(cb->dst_port);
-			if (!ipc) {
-				kfree_skb(skb);
-			} else {
-				if (sock_queue_rcv_skb(&ipc->sk, skb))
-					kfree_skb(skb);
-
-				qrtr_port_put(ipc);
-			}
-		}
-	}
-}
-
 /**
  * qrtr_endpoint_register() - register a new endpoint
  * @ep: endpoint to register
@@ -575,7 +553,6 @@ int qrtr_endpoint_register(struct qrtr_endpoint *ep, unsigned int nid)
 	if (!node)
 		return -ENOMEM;
 
-	INIT_WORK(&node->work, qrtr_node_rx_work);
 	kref_init(&node->ref);
 	mutex_init(&node->ep_lock);
 	skb_queue_head_init(&node->rx_queue);
-- 
cgit v1.2.3-59-g8ed1b


From 0fadc0a2b1c472c50a6d05b76c5f73c769bce81e Mon Sep 17 00:00:00 2001
From: Lorenzo Bianconi <lorenzo@kernel.org>
Date: Tue, 14 Jan 2020 10:24:19 +0100
Subject: net: socionext: get rid of huge dma sync in netsec_alloc_rx_data

Socionext driver can run on dma coherent and non-coherent devices.
Get rid of huge dma_sync_single_for_device in netsec_alloc_rx_data since
now the driver can let page_pool API to managed needed DMA sync

Reviewed-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/socionext/netsec.c | 43 ++++++++++++++++++++-------------
 1 file changed, 26 insertions(+), 17 deletions(-)

diff --git a/drivers/net/ethernet/socionext/netsec.c b/drivers/net/ethernet/socionext/netsec.c
index b5a9e947a4a8..6870a6ce76a6 100644
--- a/drivers/net/ethernet/socionext/netsec.c
+++ b/drivers/net/ethernet/socionext/netsec.c
@@ -243,6 +243,7 @@
 			       NET_IP_ALIGN)
 #define NETSEC_RX_BUF_NON_DATA (NETSEC_RXBUF_HEADROOM + \
 				SKB_DATA_ALIGN(sizeof(struct skb_shared_info)))
+#define NETSEC_RX_BUF_SIZE	(PAGE_SIZE - NETSEC_RX_BUF_NON_DATA)
 
 #define DESC_SZ	sizeof(struct netsec_de)
 
@@ -719,7 +720,6 @@ static void *netsec_alloc_rx_data(struct netsec_priv *priv,
 {
 
 	struct netsec_desc_ring *dring = &priv->desc_ring[NETSEC_RING_RX];
-	enum dma_data_direction dma_dir;
 	struct page *page;
 
 	page = page_pool_dev_alloc_pages(dring->page_pool);
@@ -734,9 +734,7 @@ static void *netsec_alloc_rx_data(struct netsec_priv *priv,
 	/* Make sure the incoming payload fits in the page for XDP and non-XDP
 	 * cases and reserve enough space for headroom + skb_shared_info
 	 */
-	*desc_len = PAGE_SIZE - NETSEC_RX_BUF_NON_DATA;
-	dma_dir = page_pool_get_dma_dir(dring->page_pool);
-	dma_sync_single_for_device(priv->dev, *dma_handle, *desc_len, dma_dir);
+	*desc_len = NETSEC_RX_BUF_SIZE;
 
 	return page_address(page);
 }
@@ -883,6 +881,8 @@ static u32 netsec_xdp_xmit_back(struct netsec_priv *priv, struct xdp_buff *xdp)
 static u32 netsec_run_xdp(struct netsec_priv *priv, struct bpf_prog *prog,
 			  struct xdp_buff *xdp)
 {
+	struct netsec_desc_ring *dring = &priv->desc_ring[NETSEC_RING_RX];
+	unsigned int len = xdp->data_end - xdp->data;
 	u32 ret = NETSEC_XDP_PASS;
 	int err;
 	u32 act;
@@ -896,7 +896,9 @@ static u32 netsec_run_xdp(struct netsec_priv *priv, struct bpf_prog *prog,
 	case XDP_TX:
 		ret = netsec_xdp_xmit_back(priv, xdp);
 		if (ret != NETSEC_XDP_TX)
-			xdp_return_buff(xdp);
+			__page_pool_put_page(dring->page_pool,
+					     virt_to_head_page(xdp->data),
+					     len, true);
 		break;
 	case XDP_REDIRECT:
 		err = xdp_do_redirect(priv->ndev, xdp, prog);
@@ -904,7 +906,9 @@ static u32 netsec_run_xdp(struct netsec_priv *priv, struct bpf_prog *prog,
 			ret = NETSEC_XDP_REDIR;
 		} else {
 			ret = NETSEC_XDP_CONSUMED;
-			xdp_return_buff(xdp);
+			__page_pool_put_page(dring->page_pool,
+					     virt_to_head_page(xdp->data),
+					     len, true);
 		}
 		break;
 	default:
@@ -915,7 +919,9 @@ static u32 netsec_run_xdp(struct netsec_priv *priv, struct bpf_prog *prog,
 		/* fall through -- handle aborts by dropping packet */
 	case XDP_DROP:
 		ret = NETSEC_XDP_CONSUMED;
-		xdp_return_buff(xdp);
+		__page_pool_put_page(dring->page_pool,
+				     virt_to_head_page(xdp->data),
+				     len, true);
 		break;
 	}
 
@@ -1014,7 +1020,8 @@ static int netsec_process_rx(struct netsec_priv *priv, int budget)
 			 * cache state. Since we paid the allocation cost if
 			 * building an skb fails try to put the page into cache
 			 */
-			page_pool_recycle_direct(dring->page_pool, page);
+			__page_pool_put_page(dring->page_pool, page,
+					     pkt_len, true);
 			netif_err(priv, drv, priv->ndev,
 				  "rx failed to build skb\n");
 			break;
@@ -1272,17 +1279,19 @@ static int netsec_setup_rx_dring(struct netsec_priv *priv)
 {
 	struct netsec_desc_ring *dring = &priv->desc_ring[NETSEC_RING_RX];
 	struct bpf_prog *xdp_prog = READ_ONCE(priv->xdp_prog);
-	struct page_pool_params pp_params = { 0 };
+	struct page_pool_params pp_params = {
+		.order = 0,
+		/* internal DMA mapping in page_pool */
+		.flags = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV,
+		.pool_size = DESC_NUM,
+		.nid = NUMA_NO_NODE,
+		.dev = priv->dev,
+		.dma_dir = xdp_prog ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE,
+		.offset = NETSEC_RXBUF_HEADROOM,
+		.max_len = NETSEC_RX_BUF_SIZE,
+	};
 	int i, err;
 
-	pp_params.order = 0;
-	/* internal DMA mapping in page_pool */
-	pp_params.flags = PP_FLAG_DMA_MAP;
-	pp_params.pool_size = DESC_NUM;
-	pp_params.nid = NUMA_NO_NODE;
-	pp_params.dev = priv->dev;
-	pp_params.dma_dir = xdp_prog ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE;
-
 	dring->page_pool = page_pool_create(&pp_params);
 	if (IS_ERR(dring->page_pool)) {
 		err = PTR_ERR(dring->page_pool);
-- 
cgit v1.2.3-59-g8ed1b


From 6324d0fa03bf118e9682b6f3da437deaadb61924 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Tue, 14 Jan 2020 13:23:09 +0200
Subject: ipv4: Replace route in list before notifying

Subsequent patches will add an offload / trap indication to routes which
will signal if the route is present in hardware or not.

After programming the route to the hardware, drivers will have to ask
the IPv4 code to set the flags by passing the route's key.

In the case of route replace, the new route is notified before it is
actually inserted into the FIB alias list. This can prevent simple
drivers (e.g., netdevsim) that program the route to the hardware in the
same context it is notified in from being able to set the flag.

Solve this by first inserting the new route to the list and rollback the
operation in case the route was vetoed.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/fib_trie.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index b92a42433a7d..39f56d68ec19 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -1221,23 +1221,26 @@ int fib_table_insert(struct net *net, struct fib_table *tb,
 			new_fa->tb_id = tb->tb_id;
 			new_fa->fa_default = -1;
 
+			hlist_replace_rcu(&fa->fa_list, &new_fa->fa_list);
+
 			if (fib_find_alias(&l->leaf, fa->fa_slen, 0, 0,
-					   tb->tb_id, true) == fa) {
+					   tb->tb_id, true) == new_fa) {
 				enum fib_event_type fib_event;
 
 				fib_event = FIB_EVENT_ENTRY_REPLACE;
 				err = call_fib_entry_notifiers(net, fib_event,
 							       key, plen,
 							       new_fa, extack);
-				if (err)
+				if (err) {
+					hlist_replace_rcu(&new_fa->fa_list,
+							  &fa->fa_list);
 					goto out_free_new_fa;
+				}
 			}
 
 			rtmsg_fib(RTM_NEWROUTE, htonl(key), new_fa, plen,
 				  tb->tb_id, &cfg->fc_nlinfo, nlflags);
 
-			hlist_replace_rcu(&fa->fa_list, &new_fa->fa_list);
-
 			alias_free_mem_rcu(fa);
 
 			fib_release_info(fi_drop);
-- 
cgit v1.2.3-59-g8ed1b


From 1e301fd04eaaa5b1e3c202450d86864e6714d783 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Tue, 14 Jan 2020 13:23:10 +0200
Subject: ipv4: Encapsulate function arguments in a struct

fib_dump_info() is used to prepare RTM_{NEW,DEL}ROUTE netlink messages
using the passed arguments. Currently, the function takes 11 arguments,
6 of which are attributes of the route being dumped (e.g., prefix, TOS).

The next patch will need the function to also dump to user space an
indication if the route is present in hardware or not. Instead of
passing yet another argument, change the function to take a struct
containing the different route attributes.

v2:
* Name last argument of fib_dump_info()
* Move 'struct fib_rt_info' to include/net/ip_fib.h so that it could
  later be passed to fib_alias_hw_flags_set()

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_fib.h     |  9 +++++++++
 net/ipv4/fib_lookup.h    |  5 ++---
 net/ipv4/fib_semantics.c | 26 ++++++++++++++++----------
 net/ipv4/fib_trie.c      | 14 +++++++++-----
 net/ipv4/route.c         | 12 +++++++++---
 5 files changed, 45 insertions(+), 21 deletions(-)

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index b9cba41c6d4f..0c071c820e33 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -204,6 +204,15 @@ __be32 fib_result_prefsrc(struct net *net, struct fib_result *res);
 #define FIB_RES_DEV(res)	(FIB_RES_NHC(res)->nhc_dev)
 #define FIB_RES_OIF(res)	(FIB_RES_NHC(res)->nhc_oif)
 
+struct fib_rt_info {
+	struct fib_info		*fi;
+	u32			tb_id;
+	__be32			dst;
+	int			dst_len;
+	u8			tos;
+	u8			type;
+};
+
 struct fib_entry_notifier_info {
 	struct fib_notifier_info info; /* must be first */
 	u32 dst;
diff --git a/net/ipv4/fib_lookup.h b/net/ipv4/fib_lookup.h
index a68b5e21ec51..a4b829358bfa 100644
--- a/net/ipv4/fib_lookup.h
+++ b/net/ipv4/fib_lookup.h
@@ -35,9 +35,8 @@ struct fib_info *fib_create_info(struct fib_config *cfg,
 int fib_nh_match(struct fib_config *cfg, struct fib_info *fi,
 		 struct netlink_ext_ack *extack);
 bool fib_metrics_match(struct fib_config *cfg, struct fib_info *fi);
-int fib_dump_info(struct sk_buff *skb, u32 pid, u32 seq, int event, u32 tb_id,
-		  u8 type, __be32 dst, int dst_len, u8 tos, struct fib_info *fi,
-		  unsigned int);
+int fib_dump_info(struct sk_buff *skb, u32 pid, u32 seq, int event,
+		  struct fib_rt_info *fri, unsigned int flags);
 void rtmsg_fib(int event, __be32 key, struct fib_alias *fa, int dst_len,
 	       u32 tb_id, const struct nl_info *info, unsigned int nlm_flags);
 
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index f1888c683426..3ed1349be428 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -504,6 +504,7 @@ void rtmsg_fib(int event, __be32 key, struct fib_alias *fa,
 	       int dst_len, u32 tb_id, const struct nl_info *info,
 	       unsigned int nlm_flags)
 {
+	struct fib_rt_info fri;
 	struct sk_buff *skb;
 	u32 seq = info->nlh ? info->nlh->nlmsg_seq : 0;
 	int err = -ENOBUFS;
@@ -512,9 +513,13 @@ void rtmsg_fib(int event, __be32 key, struct fib_alias *fa,
 	if (!skb)
 		goto errout;
 
-	err = fib_dump_info(skb, info->portid, seq, event, tb_id,
-			    fa->fa_type, key, dst_len,
-			    fa->fa_tos, fa->fa_info, nlm_flags);
+	fri.fi = fa->fa_info;
+	fri.tb_id = tb_id;
+	fri.dst = key;
+	fri.dst_len = dst_len;
+	fri.tos = fa->fa_tos;
+	fri.type = fa->fa_type;
+	err = fib_dump_info(skb, info->portid, seq, event, &fri, nlm_flags);
 	if (err < 0) {
 		/* -EMSGSIZE implies BUG in fib_nlmsg_size() */
 		WARN_ON(err == -EMSGSIZE);
@@ -1725,10 +1730,11 @@ static int fib_add_multipath(struct sk_buff *skb, struct fib_info *fi)
 #endif
 
 int fib_dump_info(struct sk_buff *skb, u32 portid, u32 seq, int event,
-		  u32 tb_id, u8 type, __be32 dst, int dst_len, u8 tos,
-		  struct fib_info *fi, unsigned int flags)
+		  struct fib_rt_info *fri, unsigned int flags)
 {
-	unsigned int nhs = fib_info_num_path(fi);
+	unsigned int nhs = fib_info_num_path(fri->fi);
+	struct fib_info *fi = fri->fi;
+	u32 tb_id = fri->tb_id;
 	struct nlmsghdr *nlh;
 	struct rtmsg *rtm;
 
@@ -1738,22 +1744,22 @@ int fib_dump_info(struct sk_buff *skb, u32 portid, u32 seq, int event,
 
 	rtm = nlmsg_data(nlh);
 	rtm->rtm_family = AF_INET;
-	rtm->rtm_dst_len = dst_len;
+	rtm->rtm_dst_len = fri->dst_len;
 	rtm->rtm_src_len = 0;
-	rtm->rtm_tos = tos;
+	rtm->rtm_tos = fri->tos;
 	if (tb_id < 256)
 		rtm->rtm_table = tb_id;
 	else
 		rtm->rtm_table = RT_TABLE_COMPAT;
 	if (nla_put_u32(skb, RTA_TABLE, tb_id))
 		goto nla_put_failure;
-	rtm->rtm_type = type;
+	rtm->rtm_type = fri->type;
 	rtm->rtm_flags = fi->fib_flags;
 	rtm->rtm_scope = fi->fib_scope;
 	rtm->rtm_protocol = fi->fib_protocol;
 
 	if (rtm->rtm_dst_len &&
-	    nla_put_in_addr(skb, RTA_DST, dst))
+	    nla_put_in_addr(skb, RTA_DST, fri->dst))
 		goto nla_put_failure;
 	if (fi->fib_priority &&
 	    nla_put_u32(skb, RTA_PRIORITY, fi->fib_priority))
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index 39f56d68ec19..75af3f8ae50e 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -2194,14 +2194,18 @@ static int fn_trie_dump_leaf(struct key_vector *l, struct fib_table *tb,
 
 		if (filter->dump_routes) {
 			if (!s_fa) {
+				struct fib_rt_info fri;
+
+				fri.fi = fi;
+				fri.tb_id = tb->tb_id;
+				fri.dst = xkey;
+				fri.dst_len = KEYLENGTH - fa->fa_slen;
+				fri.tos = fa->fa_tos;
+				fri.type = fa->fa_type;
 				err = fib_dump_info(skb,
 						    NETLINK_CB(cb->skb).portid,
 						    cb->nlh->nlmsg_seq,
-						    RTM_NEWROUTE,
-						    tb->tb_id, fa->fa_type,
-						    xkey,
-						    KEYLENGTH - fa->fa_slen,
-						    fa->fa_tos, fi, flags);
+						    RTM_NEWROUTE, &fri, flags);
 				if (err < 0)
 					goto stop;
 			}
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 87e979f2b74a..167a7357d12a 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -3223,16 +3223,22 @@ static int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh,
 	skb_reset_mac_header(skb);
 
 	if (rtm->rtm_flags & RTM_F_FIB_MATCH) {
+		struct fib_rt_info fri;
+
 		if (!res.fi) {
 			err = fib_props[res.type].error;
 			if (!err)
 				err = -EHOSTUNREACH;
 			goto errout_rcu;
 		}
+		fri.fi = res.fi;
+		fri.tb_id = table_id;
+		fri.dst = res.prefix;
+		fri.dst_len = res.prefixlen;
+		fri.tos = fl4.flowi4_tos;
+		fri.type = rt->rt_type;
 		err = fib_dump_info(skb, NETLINK_CB(in_skb).portid,
-				    nlh->nlmsg_seq, RTM_NEWROUTE, table_id,
-				    rt->rt_type, res.prefix, res.prefixlen,
-				    fl4.flowi4_tos, res.fi, 0);
+				    nlh->nlmsg_seq, RTM_NEWROUTE, &fri, 0);
 	} else {
 		err = rt_fill_info(net, dst, src, rt, table_id, &fl4, skb,
 				   NETLINK_CB(in_skb).portid,
-- 
cgit v1.2.3-59-g8ed1b


From 90b93f1b31f86dcde2fa3c57f1ae33d28d87a1f8 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Tue, 14 Jan 2020 13:23:11 +0200
Subject: ipv4: Add "offload" and "trap" indications to routes

When performing L3 offload, routes and nexthops are usually programmed
into two different tables in the underlying device. Therefore, the fact
that a nexthop resides in hardware does not necessarily mean that all
the associated routes also reside in hardware and vice-versa.

While the kernel can signal to user space the presence of a nexthop in
hardware (via 'RTNH_F_OFFLOAD'), it does not have a corresponding flag
for routes. In addition, the fact that a route resides in hardware does
not necessarily mean that the traffic is offloaded. For example,
unreachable routes (i.e., 'RTN_UNREACHABLE') are programmed to trap
packets to the CPU so that the kernel will be able to generate the
appropriate ICMP error packet.

This patch adds an "offload" and "trap" indications to IPv4 routes, so
that users will have better visibility into the offload process.

'struct fib_alias' is extended with two new fields that indicate if the
route resides in hardware or not and if it is offloading traffic from
the kernel or trapping packets to it. Note that the new fields are added
in the 6 bytes hole and therefore the struct still fits in a single
cache line [1].

Capable drivers are expected to invoke fib_alias_hw_flags_set() with the
route's key in order to set the flags.

The indications are dumped to user space via a new flags (i.e.,
'RTM_F_OFFLOAD' and 'RTM_F_TRAP') in the 'rtm_flags' field in the
ancillary header.

v2:
* Make use of 'struct fib_rt_info' in fib_alias_hw_flags_set()

[1]
struct fib_alias {
        struct hlist_node  fa_list;                      /*     0    16 */
        struct fib_info *          fa_info;              /*    16     8 */
        u8                         fa_tos;               /*    24     1 */
        u8                         fa_type;              /*    25     1 */
        u8                         fa_state;             /*    26     1 */
        u8                         fa_slen;              /*    27     1 */
        u32                        tb_id;                /*    28     4 */
        s16                        fa_default;           /*    32     2 */
        u8                         offload:1;            /*    34: 0  1 */
        u8                         trap:1;               /*    34: 1  1 */
        u8                         unused:6;             /*    34: 2  1 */

        /* XXX 5 bytes hole, try to pack */

        struct callback_head rcu __attribute__((__aligned__(8))); /*    40    16 */

        /* size: 56, cachelines: 1, members: 12 */
        /* sum members: 50, holes: 1, sum holes: 5 */
        /* sum bitfield members: 8 bits (1 bytes) */
        /* forced alignments: 1, forced holes: 1, sum forced holes: 5 */
        /* last cacheline: 56 bytes */
} __attribute__((__aligned__(8)));

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_fib.h           |  4 ++++
 include/uapi/linux/rtnetlink.h |  2 ++
 net/ipv4/fib_lookup.h          |  3 +++
 net/ipv4/fib_semantics.c       |  7 ++++++
 net/ipv4/fib_trie.c            | 52 ++++++++++++++++++++++++++++++++++++++++++
 net/ipv4/route.c               | 19 +++++++++++++++
 6 files changed, 87 insertions(+)

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index 0c071c820e33..6a1ae49809de 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -211,6 +211,9 @@ struct fib_rt_info {
 	int			dst_len;
 	u8			tos;
 	u8			type;
+	u8			offload:1,
+				trap:1,
+				unused:6;
 };
 
 struct fib_entry_notifier_info {
@@ -473,6 +476,7 @@ int fib_nh_common_init(struct fib_nh_common *nhc, struct nlattr *fc_encap,
 void fib_nh_common_release(struct fib_nh_common *nhc);
 
 /* Exported by fib_trie.c */
+void fib_alias_hw_flags_set(struct net *net, const struct fib_rt_info *fri);
 void fib_trie_init(void);
 struct fib_table *fib_trie_table(u32 id, struct fib_table *alias);
 
diff --git a/include/uapi/linux/rtnetlink.h b/include/uapi/linux/rtnetlink.h
index 1418a8362bb7..cd43321d20dd 100644
--- a/include/uapi/linux/rtnetlink.h
+++ b/include/uapi/linux/rtnetlink.h
@@ -309,6 +309,8 @@ enum rt_scope_t {
 #define RTM_F_PREFIX		0x800	/* Prefix addresses		*/
 #define RTM_F_LOOKUP_TABLE	0x1000	/* set rtm_table to FIB lookup result */
 #define RTM_F_FIB_MATCH	        0x2000	/* return full fib lookup match */
+#define RTM_F_OFFLOAD		0x4000	/* route is offloaded */
+#define RTM_F_TRAP		0x8000	/* route is trapping packets */
 
 /* Reserved table identifiers */
 
diff --git a/net/ipv4/fib_lookup.h b/net/ipv4/fib_lookup.h
index a4b829358bfa..c092e9a55790 100644
--- a/net/ipv4/fib_lookup.h
+++ b/net/ipv4/fib_lookup.h
@@ -16,6 +16,9 @@ struct fib_alias {
 	u8			fa_slen;
 	u32			tb_id;
 	s16			fa_default;
+	u8			offload:1,
+				trap:1,
+				unused:6;
 	struct rcu_head		rcu;
 };
 
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index 3ed1349be428..a803cdd9400a 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -519,6 +519,8 @@ void rtmsg_fib(int event, __be32 key, struct fib_alias *fa,
 	fri.dst_len = dst_len;
 	fri.tos = fa->fa_tos;
 	fri.type = fa->fa_type;
+	fri.offload = fa->offload;
+	fri.trap = fa->trap;
 	err = fib_dump_info(skb, info->portid, seq, event, &fri, nlm_flags);
 	if (err < 0) {
 		/* -EMSGSIZE implies BUG in fib_nlmsg_size() */
@@ -1801,6 +1803,11 @@ int fib_dump_info(struct sk_buff *skb, u32 portid, u32 seq, int event,
 			goto nla_put_failure;
 	}
 
+	if (fri->offload)
+		rtm->rtm_flags |= RTM_F_OFFLOAD;
+	if (fri->trap)
+		rtm->rtm_flags |= RTM_F_TRAP;
+
 	nlmsg_end(skb, nlh);
 	return 0;
 
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index 75af3f8ae50e..6ce1f2bbffd0 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -1012,6 +1012,52 @@ static struct fib_alias *fib_find_alias(struct hlist_head *fah, u8 slen,
 	return NULL;
 }
 
+static struct fib_alias *
+fib_find_matching_alias(struct net *net, const struct fib_rt_info *fri)
+{
+	u8 slen = KEYLENGTH - fri->dst_len;
+	struct key_vector *l, *tp;
+	struct fib_table *tb;
+	struct fib_alias *fa;
+	struct trie *t;
+
+	tb = fib_get_table(net, fri->tb_id);
+	if (!tb)
+		return NULL;
+
+	t = (struct trie *)tb->tb_data;
+	l = fib_find_node(t, &tp, be32_to_cpu(fri->dst));
+	if (!l)
+		return NULL;
+
+	hlist_for_each_entry_rcu(fa, &l->leaf, fa_list) {
+		if (fa->fa_slen == slen && fa->tb_id == fri->tb_id &&
+		    fa->fa_tos == fri->tos && fa->fa_info == fri->fi &&
+		    fa->fa_type == fri->type)
+			return fa;
+	}
+
+	return NULL;
+}
+
+void fib_alias_hw_flags_set(struct net *net, const struct fib_rt_info *fri)
+{
+	struct fib_alias *fa_match;
+
+	rcu_read_lock();
+
+	fa_match = fib_find_matching_alias(net, fri);
+	if (!fa_match)
+		goto out;
+
+	fa_match->offload = fri->offload;
+	fa_match->trap = fri->trap;
+
+out:
+	rcu_read_unlock();
+}
+EXPORT_SYMBOL_GPL(fib_alias_hw_flags_set);
+
 static void trie_rebalance(struct trie *t, struct key_vector *tn)
 {
 	while (!IS_TRIE(tn))
@@ -1220,6 +1266,8 @@ int fib_table_insert(struct net *net, struct fib_table *tb,
 			new_fa->fa_slen = fa->fa_slen;
 			new_fa->tb_id = tb->tb_id;
 			new_fa->fa_default = -1;
+			new_fa->offload = 0;
+			new_fa->trap = 0;
 
 			hlist_replace_rcu(&fa->fa_list, &new_fa->fa_list);
 
@@ -1278,6 +1326,8 @@ int fib_table_insert(struct net *net, struct fib_table *tb,
 	new_fa->fa_slen = slen;
 	new_fa->tb_id = tb->tb_id;
 	new_fa->fa_default = -1;
+	new_fa->offload = 0;
+	new_fa->trap = 0;
 
 	/* Insert new entry to the list. */
 	err = fib_insert_alias(t, tp, l, new_fa, fa, key);
@@ -2202,6 +2252,8 @@ static int fn_trie_dump_leaf(struct key_vector *l, struct fib_table *tb,
 				fri.dst_len = KEYLENGTH - fa->fa_slen;
 				fri.tos = fa->fa_tos;
 				fri.type = fa->fa_type;
+				fri.offload = fa->offload;
+				fri.trap = fa->trap;
 				err = fib_dump_info(skb,
 						    NETLINK_CB(cb->skb).portid,
 						    cb->nlh->nlmsg_seq,
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 167a7357d12a..2010888e68ca 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -3237,6 +3237,25 @@ static int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh,
 		fri.dst_len = res.prefixlen;
 		fri.tos = fl4.flowi4_tos;
 		fri.type = rt->rt_type;
+		fri.offload = 0;
+		fri.trap = 0;
+		if (res.fa_head) {
+			struct fib_alias *fa;
+
+			hlist_for_each_entry_rcu(fa, res.fa_head, fa_list) {
+				u8 slen = 32 - fri.dst_len;
+
+				if (fa->fa_slen == slen &&
+				    fa->tb_id == fri.tb_id &&
+				    fa->fa_tos == fri.tos &&
+				    fa->fa_info == res.fi &&
+				    fa->fa_type == fri.type) {
+					fri.offload = fa->offload;
+					fri.trap = fa->trap;
+					break;
+				}
+			}
+		}
 		err = fib_dump_info(skb, NETLINK_CB(in_skb).portid,
 				    nlh->nlmsg_seq, RTM_NEWROUTE, &fri, 0);
 	} else {
-- 
cgit v1.2.3-59-g8ed1b


From bb3c4ab93e44784c1e958bdbba7824bba40f23cd Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Tue, 14 Jan 2020 13:23:12 +0200
Subject: ipv6: Add "offload" and "trap" indications to routes

In a similar fashion to previous patch, add "offload" and "trap"
indication to IPv6 routes.

This is done by using two unused bits in 'struct fib6_info' to hold
these indications. Capable drivers are expected to set these when
processing the various in-kernel route notifications.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Acked-by: Roopa Prabhu <roopa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_fib.h | 11 ++++++++++-
 net/ipv6/route.c      |  7 +++++++
 2 files changed, 17 insertions(+), 1 deletion(-)

diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index b579faea41e9..fd60a8ac02ee 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -192,7 +192,9 @@ struct fib6_info {
 					dst_nopolicy:1,
 					dst_host:1,
 					fib6_destroying:1,
-					unused:3;
+					offload:1,
+					trap:1,
+					unused:1;
 
 	struct rcu_head			rcu;
 	struct nexthop			*nh;
@@ -329,6 +331,13 @@ static inline void fib6_info_release(struct fib6_info *f6i)
 		call_rcu(&f6i->rcu, fib6_info_destroy_rcu);
 }
 
+static inline void fib6_info_hw_flags_set(struct fib6_info *f6i, bool offload,
+					  bool trap)
+{
+	f6i->offload = offload;
+	f6i->trap = trap;
+}
+
 enum fib6_walk_state {
 #ifdef CONFIG_IPV6_SUBTREES
 	FWS_S,
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 0253b702afb7..4fbdc60b4e07 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -5576,6 +5576,13 @@ static int rt6_fill_node(struct net *net, struct sk_buff *skb,
 		expires -= jiffies;
 	}
 
+	if (!dst) {
+		if (rt->offload)
+			rtm->rtm_flags |= RTM_F_OFFLOAD;
+		if (rt->trap)
+			rtm->rtm_flags |= RTM_F_TRAP;
+	}
+
 	if (rtnl_put_cacheinfo(skb, dst, 0, expires, dst ? dst->error : 0) < 0)
 		goto nla_put_failure;
 
-- 
cgit v1.2.3-59-g8ed1b


From 8c5a5b9b9185e5f3f14c81b2d1cef7e6f837c280 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Tue, 14 Jan 2020 13:23:13 +0200
Subject: mlxsw: spectrum_router: Separate nexthop offload indication from
 route

The driver currently uses the 'RTNH_F_OFFLOAD' flag for both routes and
nexthops, which is cumbersome and unnecessary now that we have separate
flag for the route itself.

Separate the offload indication for nexthops from routes and call it
whenever the offload state within the nexthop group changes.

Note that IPv6 (unlike IPv4) does not share the same nexthop group
between different routes, whereas mlxsw does. Therefore, whenever the
offload indication within an IPv6 nexthop group changes, all the linked
routes need to be updated.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../net/ethernet/mellanox/mlxsw/spectrum_router.c  | 92 ++++++++++++++++++----
 1 file changed, 75 insertions(+), 17 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
index 7f70aa799064..cbc23485f487 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
@@ -3235,20 +3235,6 @@ mlxsw_sp_nexthop_fib_entries_update(struct mlxsw_sp *mlxsw_sp,
 	return 0;
 }
 
-static void
-mlxsw_sp_fib_entry_offload_refresh(struct mlxsw_sp_fib_entry *fib_entry,
-				   enum mlxsw_reg_ralue_op op, int err);
-
-static void
-mlxsw_sp_nexthop_fib_entries_refresh(struct mlxsw_sp_nexthop_group *nh_grp)
-{
-	enum mlxsw_reg_ralue_op op = MLXSW_REG_RALUE_OP_WRITE_WRITE;
-	struct mlxsw_sp_fib_entry *fib_entry;
-
-	list_for_each_entry(fib_entry, &nh_grp->fib_list, nexthop_group_node)
-		mlxsw_sp_fib_entry_offload_refresh(fib_entry, op, 0);
-}
-
 static void mlxsw_sp_adj_grp_size_round_up(u16 *p_adj_grp_size)
 {
 	/* Valid sizes for an adjacency group are:
@@ -3352,6 +3338,73 @@ mlxsw_sp_nexthop_group_rebalance(struct mlxsw_sp_nexthop_group *nh_grp)
 	}
 }
 
+static struct mlxsw_sp_nexthop *
+mlxsw_sp_rt6_nexthop(struct mlxsw_sp_nexthop_group *nh_grp,
+		     const struct mlxsw_sp_rt6 *mlxsw_sp_rt6);
+
+static void
+mlxsw_sp_nexthop4_group_offload_refresh(struct mlxsw_sp *mlxsw_sp,
+					struct mlxsw_sp_nexthop_group *nh_grp)
+{
+	int i;
+
+	for (i = 0; i < nh_grp->count; i++) {
+		struct mlxsw_sp_nexthop *nh = &nh_grp->nexthops[i];
+
+		if (nh->offloaded)
+			nh->key.fib_nh->fib_nh_flags |= RTNH_F_OFFLOAD;
+		else
+			nh->key.fib_nh->fib_nh_flags &= ~RTNH_F_OFFLOAD;
+	}
+}
+
+static void
+__mlxsw_sp_nexthop6_group_offload_refresh(struct mlxsw_sp_nexthop_group *nh_grp,
+					  struct mlxsw_sp_fib6_entry *fib6_entry)
+{
+	struct mlxsw_sp_rt6 *mlxsw_sp_rt6;
+
+	list_for_each_entry(mlxsw_sp_rt6, &fib6_entry->rt6_list, list) {
+		struct fib6_nh *fib6_nh = mlxsw_sp_rt6->rt->fib6_nh;
+		struct mlxsw_sp_nexthop *nh;
+
+		nh = mlxsw_sp_rt6_nexthop(nh_grp, mlxsw_sp_rt6);
+		if (nh && nh->offloaded)
+			fib6_nh->fib_nh_flags |= RTNH_F_OFFLOAD;
+		else
+			fib6_nh->fib_nh_flags &= ~RTNH_F_OFFLOAD;
+	}
+}
+
+static void
+mlxsw_sp_nexthop6_group_offload_refresh(struct mlxsw_sp *mlxsw_sp,
+					struct mlxsw_sp_nexthop_group *nh_grp)
+{
+	struct mlxsw_sp_fib6_entry *fib6_entry;
+
+	/* Unfortunately, in IPv6 the route and the nexthop are described by
+	 * the same struct, so we need to iterate over all the routes using the
+	 * nexthop group and set / clear the offload indication for them.
+	 */
+	list_for_each_entry(fib6_entry, &nh_grp->fib_list,
+			    common.nexthop_group_node)
+		__mlxsw_sp_nexthop6_group_offload_refresh(nh_grp, fib6_entry);
+}
+
+static void
+mlxsw_sp_nexthop_group_offload_refresh(struct mlxsw_sp *mlxsw_sp,
+				       struct mlxsw_sp_nexthop_group *nh_grp)
+{
+	switch (mlxsw_sp_nexthop_group_type(nh_grp)) {
+	case AF_INET:
+		mlxsw_sp_nexthop4_group_offload_refresh(mlxsw_sp, nh_grp);
+		break;
+	case AF_INET6:
+		mlxsw_sp_nexthop6_group_offload_refresh(mlxsw_sp, nh_grp);
+		break;
+	}
+}
+
 static void
 mlxsw_sp_nexthop_group_refresh(struct mlxsw_sp *mlxsw_sp,
 			       struct mlxsw_sp_nexthop_group *nh_grp)
@@ -3425,6 +3478,8 @@ mlxsw_sp_nexthop_group_refresh(struct mlxsw_sp *mlxsw_sp,
 		goto set_trap;
 	}
 
+	mlxsw_sp_nexthop_group_offload_refresh(mlxsw_sp, nh_grp);
+
 	if (!old_adj_index_valid) {
 		/* The trap was set for fib entries, so we have to call
 		 * fib entry update to unset it and use adjacency index.
@@ -3446,9 +3501,6 @@ mlxsw_sp_nexthop_group_refresh(struct mlxsw_sp *mlxsw_sp,
 		goto set_trap;
 	}
 
-	/* Offload state within the group changed, so update the flags. */
-	mlxsw_sp_nexthop_fib_entries_refresh(nh_grp);
-
 	return;
 
 set_trap:
@@ -3461,6 +3513,7 @@ set_trap:
 	err = mlxsw_sp_nexthop_fib_entries_update(mlxsw_sp, nh_grp);
 	if (err)
 		dev_warn(mlxsw_sp->bus_info->dev, "Failed to set traps for fib entries.\n");
+	mlxsw_sp_nexthop_group_offload_refresh(mlxsw_sp, nh_grp);
 	if (old_adj_index_valid)
 		mlxsw_sp_kvdl_free(mlxsw_sp, MLXSW_SP_KVDL_ENTRY_TYPE_ADJ,
 				   nh_grp->ecmp_size, nh_grp->adj_index);
@@ -5113,6 +5166,11 @@ static int mlxsw_sp_nexthop6_group_get(struct mlxsw_sp *mlxsw_sp,
 		      &nh_grp->fib_list);
 	fib6_entry->common.nh_group = nh_grp;
 
+	/* The route and the nexthop are described by the same struct, so we
+	 * need to the update the nexthop offload indication for the new route.
+	 */
+	__mlxsw_sp_nexthop6_group_offload_refresh(nh_grp, fib6_entry);
+
 	return 0;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From ee5a0448e72b0857260e04f68126b2bfa27c154e Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Tue, 14 Jan 2020 13:23:14 +0200
Subject: mlxsw: spectrum_router: Set hardware flags for routes

Previous patches added support for two hardware flags for IPv4 and IPv6
routes: 'RTM_F_OFFLOAD' and 'RTM_F_TRAP'. Both indicate the presence of
the route in hardware. The first indicates that traffic is actually
offloaded from the kernel, whereas the second indicates that packets
hitting such routes are trapped to the kernel for processing (e.g., host
routes).

Use these two flags in mlxsw. The flags are modified in two places.
Firstly, whenever a route is updated in the device's table. This
includes the addition, deletion or update of a route. For example, when
a host route is promoted to perform NVE decapsulation, its action in the
device is updated, the 'RTM_F_OFFLOAD' flag set and the 'RTM_F_TRAP'
flag cleared.

Secondly, when a route is replaced and overwritten by another route, its
flags are cleared.

v2:
* Convert to new fib_alias_hw_flags_set() interface

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../net/ethernet/mellanox/mlxsw/spectrum_router.c  | 154 ++++++++++-----------
 1 file changed, 77 insertions(+), 77 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
index cbc23485f487..a014bd8d504d 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
@@ -4096,131 +4096,128 @@ mlxsw_sp_rt6_nexthop(struct mlxsw_sp_nexthop_group *nh_grp,
 }
 
 static void
-mlxsw_sp_fib4_entry_offload_set(struct mlxsw_sp_fib_entry *fib_entry)
+mlxsw_sp_fib4_entry_hw_flags_set(struct mlxsw_sp *mlxsw_sp,
+				 struct mlxsw_sp_fib_entry *fib_entry)
 {
-	struct mlxsw_sp_nexthop_group *nh_grp = fib_entry->nh_group;
-	int i;
-
-	if (fib_entry->type == MLXSW_SP_FIB_ENTRY_TYPE_LOCAL ||
-	    fib_entry->type == MLXSW_SP_FIB_ENTRY_TYPE_BLACKHOLE ||
-	    fib_entry->type == MLXSW_SP_FIB_ENTRY_TYPE_IPIP_DECAP ||
-	    fib_entry->type == MLXSW_SP_FIB_ENTRY_TYPE_NVE_DECAP) {
-		nh_grp->nexthops->key.fib_nh->fib_nh_flags |= RTNH_F_OFFLOAD;
-		return;
-	}
-
-	for (i = 0; i < nh_grp->count; i++) {
-		struct mlxsw_sp_nexthop *nh = &nh_grp->nexthops[i];
+	struct fib_info *fi = mlxsw_sp_nexthop4_group_fi(fib_entry->nh_group);
+	u32 *p_dst = (u32 *) fib_entry->fib_node->key.addr;
+	int dst_len = fib_entry->fib_node->key.prefix_len;
+	struct mlxsw_sp_fib4_entry *fib4_entry;
+	struct fib_rt_info fri;
+	bool should_offload;
 
-		if (nh->offloaded)
-			nh->key.fib_nh->fib_nh_flags |= RTNH_F_OFFLOAD;
-		else
-			nh->key.fib_nh->fib_nh_flags &= ~RTNH_F_OFFLOAD;
-	}
+	should_offload = mlxsw_sp_fib_entry_should_offload(fib_entry);
+	fib4_entry = container_of(fib_entry, struct mlxsw_sp_fib4_entry,
+				  common);
+	fri.fi = fi;
+	fri.tb_id = fib4_entry->tb_id;
+	fri.dst = cpu_to_be32(*p_dst);
+	fri.dst_len = dst_len;
+	fri.tos = fib4_entry->tos;
+	fri.type = fib4_entry->type;
+	fri.offload = should_offload;
+	fri.trap = !should_offload;
+	fib_alias_hw_flags_set(mlxsw_sp_net(mlxsw_sp), &fri);
 }
 
 static void
-mlxsw_sp_fib4_entry_offload_unset(struct mlxsw_sp_fib_entry *fib_entry)
+mlxsw_sp_fib4_entry_hw_flags_clear(struct mlxsw_sp *mlxsw_sp,
+				   struct mlxsw_sp_fib_entry *fib_entry)
 {
-	struct mlxsw_sp_nexthop_group *nh_grp = fib_entry->nh_group;
-	int i;
-
-	if (!list_is_singular(&nh_grp->fib_list))
-		return;
-
-	for (i = 0; i < nh_grp->count; i++) {
-		struct mlxsw_sp_nexthop *nh = &nh_grp->nexthops[i];
+	struct fib_info *fi = mlxsw_sp_nexthop4_group_fi(fib_entry->nh_group);
+	u32 *p_dst = (u32 *) fib_entry->fib_node->key.addr;
+	int dst_len = fib_entry->fib_node->key.prefix_len;
+	struct mlxsw_sp_fib4_entry *fib4_entry;
+	struct fib_rt_info fri;
 
-		nh->key.fib_nh->fib_nh_flags &= ~RTNH_F_OFFLOAD;
-	}
+	fib4_entry = container_of(fib_entry, struct mlxsw_sp_fib4_entry,
+				  common);
+	fri.fi = fi;
+	fri.tb_id = fib4_entry->tb_id;
+	fri.dst = cpu_to_be32(*p_dst);
+	fri.dst_len = dst_len;
+	fri.tos = fib4_entry->tos;
+	fri.type = fib4_entry->type;
+	fri.offload = false;
+	fri.trap = false;
+	fib_alias_hw_flags_set(mlxsw_sp_net(mlxsw_sp), &fri);
 }
 
 static void
-mlxsw_sp_fib6_entry_offload_set(struct mlxsw_sp_fib_entry *fib_entry)
+mlxsw_sp_fib6_entry_hw_flags_set(struct mlxsw_sp *mlxsw_sp,
+				 struct mlxsw_sp_fib_entry *fib_entry)
 {
 	struct mlxsw_sp_fib6_entry *fib6_entry;
 	struct mlxsw_sp_rt6 *mlxsw_sp_rt6;
+	bool should_offload;
+
+	should_offload = mlxsw_sp_fib_entry_should_offload(fib_entry);
 
+	/* In IPv6 a multipath route is represented using multiple routes, so
+	 * we need to set the flags on all of them.
+	 */
 	fib6_entry = container_of(fib_entry, struct mlxsw_sp_fib6_entry,
 				  common);
-
-	if (fib_entry->type == MLXSW_SP_FIB_ENTRY_TYPE_LOCAL ||
-	    fib_entry->type == MLXSW_SP_FIB_ENTRY_TYPE_BLACKHOLE) {
-		list_first_entry(&fib6_entry->rt6_list, struct mlxsw_sp_rt6,
-				 list)->rt->fib6_nh->fib_nh_flags |= RTNH_F_OFFLOAD;
-		return;
-	}
-
-	list_for_each_entry(mlxsw_sp_rt6, &fib6_entry->rt6_list, list) {
-		struct mlxsw_sp_nexthop_group *nh_grp = fib_entry->nh_group;
-		struct fib6_nh *fib6_nh = mlxsw_sp_rt6->rt->fib6_nh;
-		struct mlxsw_sp_nexthop *nh;
-
-		nh = mlxsw_sp_rt6_nexthop(nh_grp, mlxsw_sp_rt6);
-		if (nh && nh->offloaded)
-			fib6_nh->fib_nh_flags |= RTNH_F_OFFLOAD;
-		else
-			fib6_nh->fib_nh_flags &= ~RTNH_F_OFFLOAD;
-	}
+	list_for_each_entry(mlxsw_sp_rt6, &fib6_entry->rt6_list, list)
+		fib6_info_hw_flags_set(mlxsw_sp_rt6->rt, should_offload,
+				       !should_offload);
 }
 
 static void
-mlxsw_sp_fib6_entry_offload_unset(struct mlxsw_sp_fib_entry *fib_entry)
+mlxsw_sp_fib6_entry_hw_flags_clear(struct mlxsw_sp *mlxsw_sp,
+				   struct mlxsw_sp_fib_entry *fib_entry)
 {
 	struct mlxsw_sp_fib6_entry *fib6_entry;
 	struct mlxsw_sp_rt6 *mlxsw_sp_rt6;
 
 	fib6_entry = container_of(fib_entry, struct mlxsw_sp_fib6_entry,
 				  common);
-	list_for_each_entry(mlxsw_sp_rt6, &fib6_entry->rt6_list, list) {
-		struct fib6_info *rt = mlxsw_sp_rt6->rt;
-
-		rt->fib6_nh->fib_nh_flags &= ~RTNH_F_OFFLOAD;
-	}
+	list_for_each_entry(mlxsw_sp_rt6, &fib6_entry->rt6_list, list)
+		fib6_info_hw_flags_set(mlxsw_sp_rt6->rt, false, false);
 }
 
-static void mlxsw_sp_fib_entry_offload_set(struct mlxsw_sp_fib_entry *fib_entry)
+static void
+mlxsw_sp_fib_entry_hw_flags_set(struct mlxsw_sp *mlxsw_sp,
+				struct mlxsw_sp_fib_entry *fib_entry)
 {
 	switch (fib_entry->fib_node->fib->proto) {
 	case MLXSW_SP_L3_PROTO_IPV4:
-		mlxsw_sp_fib4_entry_offload_set(fib_entry);
+		mlxsw_sp_fib4_entry_hw_flags_set(mlxsw_sp, fib_entry);
 		break;
 	case MLXSW_SP_L3_PROTO_IPV6:
-		mlxsw_sp_fib6_entry_offload_set(fib_entry);
+		mlxsw_sp_fib6_entry_hw_flags_set(mlxsw_sp, fib_entry);
 		break;
 	}
 }
 
 static void
-mlxsw_sp_fib_entry_offload_unset(struct mlxsw_sp_fib_entry *fib_entry)
+mlxsw_sp_fib_entry_hw_flags_clear(struct mlxsw_sp *mlxsw_sp,
+				  struct mlxsw_sp_fib_entry *fib_entry)
 {
 	switch (fib_entry->fib_node->fib->proto) {
 	case MLXSW_SP_L3_PROTO_IPV4:
-		mlxsw_sp_fib4_entry_offload_unset(fib_entry);
+		mlxsw_sp_fib4_entry_hw_flags_clear(mlxsw_sp, fib_entry);
 		break;
 	case MLXSW_SP_L3_PROTO_IPV6:
-		mlxsw_sp_fib6_entry_offload_unset(fib_entry);
+		mlxsw_sp_fib6_entry_hw_flags_clear(mlxsw_sp, fib_entry);
 		break;
 	}
 }
 
 static void
-mlxsw_sp_fib_entry_offload_refresh(struct mlxsw_sp_fib_entry *fib_entry,
-				   enum mlxsw_reg_ralue_op op, int err)
+mlxsw_sp_fib_entry_hw_flags_refresh(struct mlxsw_sp *mlxsw_sp,
+				    struct mlxsw_sp_fib_entry *fib_entry,
+				    enum mlxsw_reg_ralue_op op)
 {
 	switch (op) {
-	case MLXSW_REG_RALUE_OP_WRITE_DELETE:
-		return mlxsw_sp_fib_entry_offload_unset(fib_entry);
 	case MLXSW_REG_RALUE_OP_WRITE_WRITE:
-		if (err)
-			return;
-		if (mlxsw_sp_fib_entry_should_offload(fib_entry))
-			mlxsw_sp_fib_entry_offload_set(fib_entry);
-		else
-			mlxsw_sp_fib_entry_offload_unset(fib_entry);
-		return;
+		mlxsw_sp_fib_entry_hw_flags_set(mlxsw_sp, fib_entry);
+		break;
+	case MLXSW_REG_RALUE_OP_WRITE_DELETE:
+		mlxsw_sp_fib_entry_hw_flags_clear(mlxsw_sp, fib_entry);
+		break;
 	default:
-		return;
+		break;
 	}
 }
 
@@ -4447,7 +4444,10 @@ static int mlxsw_sp_fib_entry_op(struct mlxsw_sp *mlxsw_sp,
 {
 	int err = __mlxsw_sp_fib_entry_op(mlxsw_sp, fib_entry, op);
 
-	mlxsw_sp_fib_entry_offload_refresh(fib_entry, op, err);
+	if (err)
+		return err;
+
+	mlxsw_sp_fib_entry_hw_flags_refresh(mlxsw_sp, fib_entry, op);
 
 	return err;
 }
@@ -4883,7 +4883,7 @@ mlxsw_sp_router_fib4_replace(struct mlxsw_sp *mlxsw_sp,
 	if (!replaced)
 		return 0;
 
-	mlxsw_sp_fib_entry_offload_unset(replaced);
+	mlxsw_sp_fib_entry_hw_flags_clear(mlxsw_sp, replaced);
 	fib4_replaced = container_of(replaced, struct mlxsw_sp_fib4_entry,
 				     common);
 	mlxsw_sp_fib4_entry_destroy(mlxsw_sp, fib4_replaced);
@@ -5451,7 +5451,7 @@ static int mlxsw_sp_router_fib6_replace(struct mlxsw_sp *mlxsw_sp,
 	if (!replaced)
 		return 0;
 
-	mlxsw_sp_fib_entry_offload_unset(replaced);
+	mlxsw_sp_fib_entry_hw_flags_clear(mlxsw_sp, replaced);
 	fib6_replaced = container_of(replaced, struct mlxsw_sp_fib6_entry,
 				     common);
 	mlxsw_sp_fib6_entry_destroy(mlxsw_sp, fib6_replaced);
-- 
cgit v1.2.3-59-g8ed1b


From 48bb9eb47b2708bd3ded16684ddd258e53bd402c Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Tue, 14 Jan 2020 13:23:15 +0200
Subject: netdevsim: fib: Add dummy implementation for FIB offload

Implement dummy IPv4 and IPv6 FIB "offload" in the driver by storing
currently "programmed" routes in a hash table. Each route in the hash
table is marked with "trap" indication. The indication is cleared when
the route is replaced or when the netdevsim instance is deleted.

This will later allow us to test the route offload API on top of
netdevsim.

v2:
* Convert to new fib_alias_hw_flags_set() interface

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/Kconfig         |   1 +
 drivers/net/netdevsim/fib.c | 671 +++++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 663 insertions(+), 9 deletions(-)

diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index 01e2657e4c26..e65ed4d0a7ad 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -549,6 +549,7 @@ source "drivers/net/hyperv/Kconfig"
 config NETDEVSIM
 	tristate "Simulated networking device"
 	depends on DEBUG_FS
+	depends on IPV6 || IPV6=n
 	select NET_DEVLINK
 	help
 	  This driver is a developer testing tool and software model that can
diff --git a/drivers/net/netdevsim/fib.c b/drivers/net/netdevsim/fib.c
index b5df308b4e33..8f56289fc2ec 100644
--- a/drivers/net/netdevsim/fib.c
+++ b/drivers/net/netdevsim/fib.c
@@ -14,6 +14,12 @@
  * THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION.
  */
 
+#include <linux/in6.h>
+#include <linux/kernel.h>
+#include <linux/list.h>
+#include <linux/rhashtable.h>
+#include <linux/spinlock_types.h>
+#include <linux/types.h>
 #include <net/fib_notifier.h>
 #include <net/ip_fib.h>
 #include <net/ip6_fib.h>
@@ -36,6 +42,48 @@ struct nsim_fib_data {
 	struct notifier_block fib_nb;
 	struct nsim_per_fib_data ipv4;
 	struct nsim_per_fib_data ipv6;
+	struct rhashtable fib_rt_ht;
+	struct list_head fib_rt_list;
+	spinlock_t fib_lock;	/* Protects hashtable, list and accounting */
+	struct devlink *devlink;
+};
+
+struct nsim_fib_rt_key {
+	unsigned char addr[sizeof(struct in6_addr)];
+	unsigned char prefix_len;
+	int family;
+	u32 tb_id;
+};
+
+struct nsim_fib_rt {
+	struct nsim_fib_rt_key key;
+	struct rhash_head ht_node;
+	struct list_head list;	/* Member of fib_rt_list */
+};
+
+struct nsim_fib4_rt {
+	struct nsim_fib_rt common;
+	struct fib_info *fi;
+	u8 tos;
+	u8 type;
+};
+
+struct nsim_fib6_rt {
+	struct nsim_fib_rt common;
+	struct list_head nh_list;
+	unsigned int nhs;
+};
+
+struct nsim_fib6_rt_nh {
+	struct list_head list;	/* Member of nh_list */
+	struct fib6_info *rt;
+};
+
+static const struct rhashtable_params nsim_fib_rt_ht_params = {
+	.key_offset = offsetof(struct nsim_fib_rt, key),
+	.head_offset = offsetof(struct nsim_fib_rt, ht_node),
+	.key_len = sizeof(struct nsim_fib_rt_key),
+	.automatic_shrinking = true,
 };
 
 u64 nsim_fib_get_val(struct nsim_fib_data *fib_data,
@@ -144,18 +192,556 @@ static int nsim_fib_account(struct nsim_fib_entry *entry, bool add,
 	return err;
 }
 
+static void nsim_fib_rt_init(struct nsim_fib_data *data,
+			     struct nsim_fib_rt *fib_rt, const void *addr,
+			     size_t addr_len, unsigned int prefix_len,
+			     int family, u32 tb_id)
+{
+	memcpy(fib_rt->key.addr, addr, addr_len);
+	fib_rt->key.prefix_len = prefix_len;
+	fib_rt->key.family = family;
+	fib_rt->key.tb_id = tb_id;
+	list_add(&fib_rt->list, &data->fib_rt_list);
+}
+
+static void nsim_fib_rt_fini(struct nsim_fib_rt *fib_rt)
+{
+	list_del(&fib_rt->list);
+}
+
+static struct nsim_fib_rt *nsim_fib_rt_lookup(struct rhashtable *fib_rt_ht,
+					      const void *addr, size_t addr_len,
+					      unsigned int prefix_len,
+					      int family, u32 tb_id)
+{
+	struct nsim_fib_rt_key key;
+
+	memset(&key, 0, sizeof(key));
+	memcpy(key.addr, addr, addr_len);
+	key.prefix_len = prefix_len;
+	key.family = family;
+	key.tb_id = tb_id;
+
+	return rhashtable_lookup_fast(fib_rt_ht, &key, nsim_fib_rt_ht_params);
+}
+
+static struct nsim_fib4_rt *
+nsim_fib4_rt_create(struct nsim_fib_data *data,
+		    struct fib_entry_notifier_info *fen_info)
+{
+	struct nsim_fib4_rt *fib4_rt;
+
+	fib4_rt = kzalloc(sizeof(*fib4_rt), GFP_ATOMIC);
+	if (!fib4_rt)
+		return NULL;
+
+	nsim_fib_rt_init(data, &fib4_rt->common, &fen_info->dst, sizeof(u32),
+			 fen_info->dst_len, AF_INET, fen_info->tb_id);
+
+	fib4_rt->fi = fen_info->fi;
+	fib_info_hold(fib4_rt->fi);
+	fib4_rt->tos = fen_info->tos;
+	fib4_rt->type = fen_info->type;
+
+	return fib4_rt;
+}
+
+static void nsim_fib4_rt_destroy(struct nsim_fib4_rt *fib4_rt)
+{
+	fib_info_put(fib4_rt->fi);
+	nsim_fib_rt_fini(&fib4_rt->common);
+	kfree(fib4_rt);
+}
+
+static struct nsim_fib4_rt *
+nsim_fib4_rt_lookup(struct rhashtable *fib_rt_ht,
+		    const struct fib_entry_notifier_info *fen_info)
+{
+	struct nsim_fib_rt *fib_rt;
+
+	fib_rt = nsim_fib_rt_lookup(fib_rt_ht, &fen_info->dst, sizeof(u32),
+				    fen_info->dst_len, AF_INET,
+				    fen_info->tb_id);
+	if (!fib_rt)
+		return NULL;
+
+	return container_of(fib_rt, struct nsim_fib4_rt, common);
+}
+
+static void nsim_fib4_rt_hw_flags_set(struct net *net,
+				      const struct nsim_fib4_rt *fib4_rt,
+				      bool trap)
+{
+	u32 *p_dst = (u32 *) fib4_rt->common.key.addr;
+	int dst_len = fib4_rt->common.key.prefix_len;
+	struct fib_rt_info fri;
+
+	fri.fi = fib4_rt->fi;
+	fri.tb_id = fib4_rt->common.key.tb_id;
+	fri.dst = cpu_to_be32(*p_dst);
+	fri.dst_len = dst_len;
+	fri.tos = fib4_rt->tos;
+	fri.type = fib4_rt->type;
+	fri.offload = false;
+	fri.trap = trap;
+	fib_alias_hw_flags_set(net, &fri);
+}
+
+static int nsim_fib4_rt_add(struct nsim_fib_data *data,
+			    struct nsim_fib4_rt *fib4_rt,
+			    struct netlink_ext_ack *extack)
+{
+	struct net *net = devlink_net(data->devlink);
+	int err;
+
+	err = nsim_fib_account(&data->ipv4.fib, true, extack);
+	if (err)
+		return err;
+
+	err = rhashtable_insert_fast(&data->fib_rt_ht,
+				     &fib4_rt->common.ht_node,
+				     nsim_fib_rt_ht_params);
+	if (err) {
+		NL_SET_ERR_MSG_MOD(extack, "Failed to insert IPv4 route");
+		goto err_fib_dismiss;
+	}
+
+	nsim_fib4_rt_hw_flags_set(net, fib4_rt, true);
+
+	return 0;
+
+err_fib_dismiss:
+	nsim_fib_account(&data->ipv4.fib, false, extack);
+	return err;
+}
+
+static int nsim_fib4_rt_replace(struct nsim_fib_data *data,
+				struct nsim_fib4_rt *fib4_rt,
+				struct nsim_fib4_rt *fib4_rt_old,
+				struct netlink_ext_ack *extack)
+{
+	struct net *net = devlink_net(data->devlink);
+	int err;
+
+	/* We are replacing a route, so no need to change the accounting. */
+	err = rhashtable_replace_fast(&data->fib_rt_ht,
+				      &fib4_rt_old->common.ht_node,
+				      &fib4_rt->common.ht_node,
+				      nsim_fib_rt_ht_params);
+	if (err) {
+		NL_SET_ERR_MSG_MOD(extack, "Failed to replace IPv4 route");
+		return err;
+	}
+
+	nsim_fib4_rt_hw_flags_set(net, fib4_rt, true);
+
+	nsim_fib4_rt_hw_flags_set(net, fib4_rt_old, false);
+	nsim_fib4_rt_destroy(fib4_rt_old);
+
+	return 0;
+}
+
+static int nsim_fib4_rt_insert(struct nsim_fib_data *data,
+			       struct fib_entry_notifier_info *fen_info)
+{
+	struct netlink_ext_ack *extack = fen_info->info.extack;
+	struct nsim_fib4_rt *fib4_rt, *fib4_rt_old;
+	int err;
+
+	fib4_rt = nsim_fib4_rt_create(data, fen_info);
+	if (!fib4_rt)
+		return -ENOMEM;
+
+	fib4_rt_old = nsim_fib4_rt_lookup(&data->fib_rt_ht, fen_info);
+	if (!fib4_rt_old)
+		err = nsim_fib4_rt_add(data, fib4_rt, extack);
+	else
+		err = nsim_fib4_rt_replace(data, fib4_rt, fib4_rt_old, extack);
+
+	if (err)
+		nsim_fib4_rt_destroy(fib4_rt);
+
+	return err;
+}
+
+static void nsim_fib4_rt_remove(struct nsim_fib_data *data,
+				const struct fib_entry_notifier_info *fen_info)
+{
+	struct netlink_ext_ack *extack = fen_info->info.extack;
+	struct nsim_fib4_rt *fib4_rt;
+
+	fib4_rt = nsim_fib4_rt_lookup(&data->fib_rt_ht, fen_info);
+	if (WARN_ON_ONCE(!fib4_rt))
+		return;
+
+	rhashtable_remove_fast(&data->fib_rt_ht, &fib4_rt->common.ht_node,
+			       nsim_fib_rt_ht_params);
+	nsim_fib_account(&data->ipv4.fib, false, extack);
+	nsim_fib4_rt_destroy(fib4_rt);
+}
+
+static int nsim_fib4_event(struct nsim_fib_data *data,
+			   struct fib_notifier_info *info,
+			   unsigned long event)
+{
+	struct fib_entry_notifier_info *fen_info;
+	int err = 0;
+
+	fen_info = container_of(info, struct fib_entry_notifier_info, info);
+
+	if (fen_info->fi->nh) {
+		NL_SET_ERR_MSG_MOD(info->extack, "IPv4 route with nexthop objects is not supported");
+		return 0;
+	}
+
+	switch (event) {
+	case FIB_EVENT_ENTRY_REPLACE:
+		err = nsim_fib4_rt_insert(data, fen_info);
+		break;
+	case FIB_EVENT_ENTRY_DEL:
+		nsim_fib4_rt_remove(data, fen_info);
+		break;
+	default:
+		break;
+	}
+
+	return err;
+}
+
+static struct nsim_fib6_rt_nh *
+nsim_fib6_rt_nh_find(const struct nsim_fib6_rt *fib6_rt,
+		     const struct fib6_info *rt)
+{
+	struct nsim_fib6_rt_nh *fib6_rt_nh;
+
+	list_for_each_entry(fib6_rt_nh, &fib6_rt->nh_list, list) {
+		if (fib6_rt_nh->rt == rt)
+			return fib6_rt_nh;
+	}
+
+	return NULL;
+}
+
+static int nsim_fib6_rt_nh_add(struct nsim_fib6_rt *fib6_rt,
+			       struct fib6_info *rt)
+{
+	struct nsim_fib6_rt_nh *fib6_rt_nh;
+
+	fib6_rt_nh = kzalloc(sizeof(*fib6_rt_nh), GFP_ATOMIC);
+	if (!fib6_rt_nh)
+		return -ENOMEM;
+
+	fib6_info_hold(rt);
+	fib6_rt_nh->rt = rt;
+	list_add_tail(&fib6_rt_nh->list, &fib6_rt->nh_list);
+	fib6_rt->nhs++;
+
+	return 0;
+}
+
+static void nsim_fib6_rt_nh_del(struct nsim_fib6_rt *fib6_rt,
+				const struct fib6_info *rt)
+{
+	struct nsim_fib6_rt_nh *fib6_rt_nh;
+
+	fib6_rt_nh = nsim_fib6_rt_nh_find(fib6_rt, rt);
+	if (WARN_ON_ONCE(!fib6_rt_nh))
+		return;
+
+	fib6_rt->nhs--;
+	list_del(&fib6_rt_nh->list);
+#if IS_ENABLED(CONFIG_IPV6)
+	fib6_info_release(fib6_rt_nh->rt);
+#endif
+	kfree(fib6_rt_nh);
+}
+
+static struct nsim_fib6_rt *
+nsim_fib6_rt_create(struct nsim_fib_data *data,
+		    struct fib6_entry_notifier_info *fen6_info)
+{
+	struct fib6_info *iter, *rt = fen6_info->rt;
+	struct nsim_fib6_rt *fib6_rt;
+	int i = 0;
+	int err;
+
+	fib6_rt = kzalloc(sizeof(*fib6_rt), GFP_ATOMIC);
+	if (!fib6_rt)
+		return NULL;
+
+	nsim_fib_rt_init(data, &fib6_rt->common, &rt->fib6_dst.addr,
+			 sizeof(rt->fib6_dst.addr), rt->fib6_dst.plen, AF_INET6,
+			 rt->fib6_table->tb6_id);
+
+	/* We consider a multipath IPv6 route as one entry, but it can be made
+	 * up from several fib6_info structs (one for each nexthop), so we
+	 * add them all to the same list under the entry.
+	 */
+	INIT_LIST_HEAD(&fib6_rt->nh_list);
+
+	err = nsim_fib6_rt_nh_add(fib6_rt, rt);
+	if (err)
+		goto err_fib_rt_fini;
+
+	if (!fen6_info->nsiblings)
+		return fib6_rt;
+
+	list_for_each_entry(iter, &rt->fib6_siblings, fib6_siblings) {
+		if (i == fen6_info->nsiblings)
+			break;
+
+		err = nsim_fib6_rt_nh_add(fib6_rt, iter);
+		if (err)
+			goto err_fib6_rt_nh_del;
+		i++;
+	}
+	WARN_ON_ONCE(i != fen6_info->nsiblings);
+
+	return fib6_rt;
+
+err_fib6_rt_nh_del:
+	list_for_each_entry_continue_reverse(iter, &rt->fib6_siblings,
+					     fib6_siblings)
+		nsim_fib6_rt_nh_del(fib6_rt, iter);
+	nsim_fib6_rt_nh_del(fib6_rt, rt);
+err_fib_rt_fini:
+	nsim_fib_rt_fini(&fib6_rt->common);
+	kfree(fib6_rt);
+	return ERR_PTR(err);
+}
+
+static void nsim_fib6_rt_destroy(struct nsim_fib6_rt *fib6_rt)
+{
+	struct nsim_fib6_rt_nh *iter, *tmp;
+
+	list_for_each_entry_safe(iter, tmp, &fib6_rt->nh_list, list)
+		nsim_fib6_rt_nh_del(fib6_rt, iter->rt);
+	WARN_ON_ONCE(!list_empty(&fib6_rt->nh_list));
+	nsim_fib_rt_fini(&fib6_rt->common);
+	kfree(fib6_rt);
+}
+
+static struct nsim_fib6_rt *
+nsim_fib6_rt_lookup(struct rhashtable *fib_rt_ht, const struct fib6_info *rt)
+{
+	struct nsim_fib_rt *fib_rt;
+
+	fib_rt = nsim_fib_rt_lookup(fib_rt_ht, &rt->fib6_dst.addr,
+				    sizeof(rt->fib6_dst.addr),
+				    rt->fib6_dst.plen, AF_INET6,
+				    rt->fib6_table->tb6_id);
+	if (!fib_rt)
+		return NULL;
+
+	return container_of(fib_rt, struct nsim_fib6_rt, common);
+}
+
+static int nsim_fib6_rt_append(struct nsim_fib_data *data,
+			       struct fib6_entry_notifier_info *fen6_info)
+{
+	struct fib6_info *iter, *rt = fen6_info->rt;
+	struct nsim_fib6_rt *fib6_rt;
+	int i = 0;
+	int err;
+
+	fib6_rt = nsim_fib6_rt_lookup(&data->fib_rt_ht, rt);
+	if (WARN_ON_ONCE(!fib6_rt))
+		return -EINVAL;
+
+	err = nsim_fib6_rt_nh_add(fib6_rt, rt);
+	if (err)
+		return err;
+	rt->trap = true;
+
+	if (!fen6_info->nsiblings)
+		return 0;
+
+	list_for_each_entry(iter, &rt->fib6_siblings, fib6_siblings) {
+		if (i == fen6_info->nsiblings)
+			break;
+
+		err = nsim_fib6_rt_nh_add(fib6_rt, iter);
+		if (err)
+			goto err_fib6_rt_nh_del;
+		iter->trap = true;
+		i++;
+	}
+	WARN_ON_ONCE(i != fen6_info->nsiblings);
+
+	return 0;
+
+err_fib6_rt_nh_del:
+	list_for_each_entry_continue_reverse(iter, &rt->fib6_siblings,
+					     fib6_siblings) {
+		iter->trap = false;
+		nsim_fib6_rt_nh_del(fib6_rt, iter);
+	}
+	rt->trap = false;
+	nsim_fib6_rt_nh_del(fib6_rt, rt);
+	return err;
+}
+
+static void nsim_fib6_rt_hw_flags_set(const struct nsim_fib6_rt *fib6_rt,
+				      bool trap)
+{
+	struct nsim_fib6_rt_nh *fib6_rt_nh;
+
+	list_for_each_entry(fib6_rt_nh, &fib6_rt->nh_list, list)
+		fib6_info_hw_flags_set(fib6_rt_nh->rt, false, trap);
+}
+
+static int nsim_fib6_rt_add(struct nsim_fib_data *data,
+			    struct nsim_fib6_rt *fib6_rt,
+			    struct netlink_ext_ack *extack)
+{
+	int err;
+
+	err = nsim_fib_account(&data->ipv6.fib, true, extack);
+	if (err)
+		return err;
+
+	err = rhashtable_insert_fast(&data->fib_rt_ht,
+				     &fib6_rt->common.ht_node,
+				     nsim_fib_rt_ht_params);
+	if (err) {
+		NL_SET_ERR_MSG_MOD(extack, "Failed to insert IPv6 route");
+		goto err_fib_dismiss;
+	}
+
+	nsim_fib6_rt_hw_flags_set(fib6_rt, true);
+
+	return 0;
+
+err_fib_dismiss:
+	nsim_fib_account(&data->ipv6.fib, false, extack);
+	return err;
+}
+
+static int nsim_fib6_rt_replace(struct nsim_fib_data *data,
+				struct nsim_fib6_rt *fib6_rt,
+				struct nsim_fib6_rt *fib6_rt_old,
+				struct netlink_ext_ack *extack)
+{
+	int err;
+
+	/* We are replacing a route, so no need to change the accounting. */
+	err = rhashtable_replace_fast(&data->fib_rt_ht,
+				      &fib6_rt_old->common.ht_node,
+				      &fib6_rt->common.ht_node,
+				      nsim_fib_rt_ht_params);
+	if (err) {
+		NL_SET_ERR_MSG_MOD(extack, "Failed to replace IPv6 route");
+		return err;
+	}
+
+	nsim_fib6_rt_hw_flags_set(fib6_rt, true);
+
+	nsim_fib6_rt_hw_flags_set(fib6_rt_old, false);
+	nsim_fib6_rt_destroy(fib6_rt_old);
+
+	return 0;
+}
+
+static int nsim_fib6_rt_insert(struct nsim_fib_data *data,
+			       struct fib6_entry_notifier_info *fen6_info)
+{
+	struct netlink_ext_ack *extack = fen6_info->info.extack;
+	struct nsim_fib6_rt *fib6_rt, *fib6_rt_old;
+	int err;
+
+	fib6_rt = nsim_fib6_rt_create(data, fen6_info);
+	if (!fib6_rt)
+		return -ENOMEM;
+
+	fib6_rt_old = nsim_fib6_rt_lookup(&data->fib_rt_ht, fen6_info->rt);
+	if (!fib6_rt_old)
+		err = nsim_fib6_rt_add(data, fib6_rt, extack);
+	else
+		err = nsim_fib6_rt_replace(data, fib6_rt, fib6_rt_old, extack);
+
+	if (err)
+		nsim_fib6_rt_destroy(fib6_rt);
+
+	return err;
+}
+
+static void
+nsim_fib6_rt_remove(struct nsim_fib_data *data,
+		    const struct fib6_entry_notifier_info *fen6_info)
+{
+	struct netlink_ext_ack *extack = fen6_info->info.extack;
+	struct nsim_fib6_rt *fib6_rt;
+
+	/* Multipath routes are first added to the FIB trie and only then
+	 * notified. If we vetoed the addition, we will get a delete
+	 * notification for a route we do not have. Therefore, do not warn if
+	 * route was not found.
+	 */
+	fib6_rt = nsim_fib6_rt_lookup(&data->fib_rt_ht, fen6_info->rt);
+	if (!fib6_rt)
+		return;
+
+	/* If not all the nexthops are deleted, then only reduce the nexthop
+	 * group.
+	 */
+	if (fen6_info->nsiblings + 1 != fib6_rt->nhs) {
+		nsim_fib6_rt_nh_del(fib6_rt, fen6_info->rt);
+		return;
+	}
+
+	rhashtable_remove_fast(&data->fib_rt_ht, &fib6_rt->common.ht_node,
+			       nsim_fib_rt_ht_params);
+	nsim_fib_account(&data->ipv6.fib, false, extack);
+	nsim_fib6_rt_destroy(fib6_rt);
+}
+
+static int nsim_fib6_event(struct nsim_fib_data *data,
+			   struct fib_notifier_info *info,
+			   unsigned long event)
+{
+	struct fib6_entry_notifier_info *fen6_info;
+	int err = 0;
+
+	fen6_info = container_of(info, struct fib6_entry_notifier_info, info);
+
+	if (fen6_info->rt->nh) {
+		NL_SET_ERR_MSG_MOD(info->extack, "IPv6 route with nexthop objects is not supported");
+		return 0;
+	}
+
+	if (fen6_info->rt->fib6_src.plen) {
+		NL_SET_ERR_MSG_MOD(info->extack, "IPv6 source-specific route is not supported");
+		return 0;
+	}
+
+	switch (event) {
+	case FIB_EVENT_ENTRY_REPLACE:
+		err = nsim_fib6_rt_insert(data, fen6_info);
+		break;
+	case FIB_EVENT_ENTRY_APPEND:
+		err = nsim_fib6_rt_append(data, fen6_info);
+		break;
+	case FIB_EVENT_ENTRY_DEL:
+		nsim_fib6_rt_remove(data, fen6_info);
+		break;
+	default:
+		break;
+	}
+
+	return err;
+}
+
 static int nsim_fib_event(struct nsim_fib_data *data,
-			  struct fib_notifier_info *info, bool add)
+			  struct fib_notifier_info *info, unsigned long event)
 {
-	struct netlink_ext_ack *extack = info->extack;
 	int err = 0;
 
 	switch (info->family) {
 	case AF_INET:
-		err = nsim_fib_account(&data->ipv4.fib, add, extack);
+		err = nsim_fib4_event(data, info, event);
 		break;
 	case AF_INET6:
-		err = nsim_fib_account(&data->ipv6.fib, add, extack);
+		err = nsim_fib6_event(data, info, event);
 		break;
 	}
 
@@ -170,6 +756,9 @@ static int nsim_fib_event_nb(struct notifier_block *nb, unsigned long event,
 	struct fib_notifier_info *info = ptr;
 	int err = 0;
 
+	/* IPv6 routes can be added via RAs from softIRQ. */
+	spin_lock_bh(&data->fib_lock);
+
 	switch (event) {
 	case FIB_EVENT_RULE_ADD: /* fall through */
 	case FIB_EVENT_RULE_DEL:
@@ -178,23 +767,74 @@ static int nsim_fib_event_nb(struct notifier_block *nb, unsigned long event,
 		break;
 
 	case FIB_EVENT_ENTRY_REPLACE:  /* fall through */
+	case FIB_EVENT_ENTRY_APPEND:  /* fall through */
 	case FIB_EVENT_ENTRY_DEL:
-		err = nsim_fib_event(data, info, event != FIB_EVENT_ENTRY_DEL);
+		err = nsim_fib_event(data, info, event);
 		break;
 	}
 
+	spin_unlock_bh(&data->fib_lock);
+
 	return notifier_from_errno(err);
 }
 
+static void nsim_fib4_rt_free(struct nsim_fib_rt *fib_rt,
+			      struct nsim_fib_data *data)
+{
+	struct devlink *devlink = data->devlink;
+	struct nsim_fib4_rt *fib4_rt;
+
+	fib4_rt = container_of(fib_rt, struct nsim_fib4_rt, common);
+	nsim_fib4_rt_hw_flags_set(devlink_net(devlink), fib4_rt, false);
+	nsim_fib_account(&data->ipv4.fib, false, NULL);
+	nsim_fib4_rt_destroy(fib4_rt);
+}
+
+static void nsim_fib6_rt_free(struct nsim_fib_rt *fib_rt,
+			      struct nsim_fib_data *data)
+{
+	struct nsim_fib6_rt *fib6_rt;
+
+	fib6_rt = container_of(fib_rt, struct nsim_fib6_rt, common);
+	nsim_fib6_rt_hw_flags_set(fib6_rt, false);
+	nsim_fib_account(&data->ipv6.fib, false, NULL);
+	nsim_fib6_rt_destroy(fib6_rt);
+}
+
+static void nsim_fib_rt_free(void *ptr, void *arg)
+{
+	struct nsim_fib_rt *fib_rt = ptr;
+	struct nsim_fib_data *data = arg;
+
+	switch (fib_rt->key.family) {
+	case AF_INET:
+		nsim_fib4_rt_free(fib_rt, data);
+		break;
+	case AF_INET6:
+		nsim_fib6_rt_free(fib_rt, data);
+		break;
+	default:
+		WARN_ON_ONCE(1);
+	}
+}
+
 /* inconsistent dump, trying again */
 static void nsim_fib_dump_inconsistent(struct notifier_block *nb)
 {
 	struct nsim_fib_data *data = container_of(nb, struct nsim_fib_data,
 						  fib_nb);
+	struct nsim_fib_rt *fib_rt, *fib_rt_tmp;
+
+	/* The notifier block is still not registered, so we do not need to
+	 * take any locks here.
+	 */
+	list_for_each_entry_safe(fib_rt, fib_rt_tmp, &data->fib_rt_list, list) {
+		rhashtable_remove_fast(&data->fib_rt_ht, &fib_rt->ht_node,
+				       nsim_fib_rt_ht_params);
+		nsim_fib_rt_free(fib_rt, data);
+	}
 
-	data->ipv4.fib.num = 0ULL;
 	data->ipv4.rules.num = 0ULL;
-	data->ipv6.fib.num = 0ULL;
 	data->ipv6.rules.num = 0ULL;
 }
 
@@ -255,6 +895,13 @@ struct nsim_fib_data *nsim_fib_create(struct devlink *devlink,
 	data = kzalloc(sizeof(*data), GFP_KERNEL);
 	if (!data)
 		return ERR_PTR(-ENOMEM);
+	data->devlink = devlink;
+
+	spin_lock_init(&data->fib_lock);
+	INIT_LIST_HEAD(&data->fib_rt_list);
+	err = rhashtable_init(&data->fib_rt_ht, &nsim_fib_rt_ht_params);
+	if (err)
+		goto err_data_free;
 
 	nsim_fib_set_max_all(data, devlink);
 
@@ -263,7 +910,7 @@ struct nsim_fib_data *nsim_fib_create(struct devlink *devlink,
 				    nsim_fib_dump_inconsistent, extack);
 	if (err) {
 		pr_err("Failed to register fib notifier\n");
-		goto err_out;
+		goto err_rhashtable_destroy;
 	}
 
 	devlink_resource_occ_get_register(devlink,
@@ -284,7 +931,10 @@ struct nsim_fib_data *nsim_fib_create(struct devlink *devlink,
 					  data);
 	return data;
 
-err_out:
+err_rhashtable_destroy:
+	rhashtable_free_and_destroy(&data->fib_rt_ht, nsim_fib_rt_free,
+				    data);
+err_data_free:
 	kfree(data);
 	return ERR_PTR(err);
 }
@@ -300,5 +950,8 @@ void nsim_fib_destroy(struct devlink *devlink, struct nsim_fib_data *data)
 	devlink_resource_occ_get_unregister(devlink,
 					    NSIM_RESOURCE_IPV4_FIB);
 	unregister_fib_notifier(devlink_net(devlink), &data->fib_nb);
+	rhashtable_free_and_destroy(&data->fib_rt_ht, nsim_fib_rt_free,
+				    data);
+	WARN_ON_ONCE(!list_empty(&data->fib_rt_list));
 	kfree(data);
 }
-- 
cgit v1.2.3-59-g8ed1b


From c662455b64a1f0dacb2127f402bd5401b88d42d1 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Tue, 14 Jan 2020 13:23:16 +0200
Subject: selftests: forwarding: Add helpers and tests for FIB offload

Implement a set of common helpers and tests for FIB offload that can be
used by multiple drivers to check their FIB offload implementations.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../selftests/net/forwarding/fib_offload_lib.sh    | 873 +++++++++++++++++++++
 1 file changed, 873 insertions(+)
 create mode 100644 tools/testing/selftests/net/forwarding/fib_offload_lib.sh

diff --git a/tools/testing/selftests/net/forwarding/fib_offload_lib.sh b/tools/testing/selftests/net/forwarding/fib_offload_lib.sh
new file mode 100644
index 000000000000..66496659bea7
--- /dev/null
+++ b/tools/testing/selftests/net/forwarding/fib_offload_lib.sh
@@ -0,0 +1,873 @@
+# SPDX-License-Identifier: GPL-2.0
+#
+# Various helpers and tests to verify FIB offload.
+
+__fib_trap_check()
+{
+	local ns=$1; shift
+	local family=$1; shift
+	local route=$1; shift
+	local should_fail=$1; shift
+	local ret
+
+	ip -n $ns -j -p -$family route show $route \
+		| jq -e '.[]["flags"] | contains(["trap"])' &> /dev/null
+	ret=$?
+	if [[ $should_fail == "true" ]]; then
+		if [[ $ret -ne 0 ]]; then
+			return 0
+		else
+			return 1
+		fi
+	fi
+
+	return $ret
+}
+
+fib_trap_check()
+{
+	local ns=$1; shift
+	local family=$1; shift
+	local route=$1; shift
+	local should_fail=$1; shift
+
+	busywait 5000 __fib_trap_check $ns $family "$route" $should_fail
+}
+
+fib4_trap_check()
+{
+	local ns=$1; shift
+	local route=$1; shift
+	local should_fail=$1; shift
+
+	fib_trap_check $ns 4 "$route" $should_fail
+}
+
+fib6_trap_check()
+{
+	local ns=$1; shift
+	local route=$1; shift
+	local should_fail=$1; shift
+
+	fib_trap_check $ns 6 "$route" $should_fail
+}
+
+fib_ipv4_identical_routes_test()
+{
+	local ns=$1; shift
+	local i
+
+	RET=0
+
+	for i in $(seq 1 3); do
+		ip -n $ns link add name dummy$i type dummy
+		ip -n $ns link set dev dummy$i up
+	done
+
+	ip -n $ns route add 192.0.2.0/24 dev dummy1 tos 0 metric 1024
+	fib4_trap_check $ns "192.0.2.0/24 dev dummy1 tos 0 metric 1024" false
+	check_err $? "Route not in hardware when should"
+
+	ip -n $ns route append 192.0.2.0/24 dev dummy2 tos 0 metric 1024
+	fib4_trap_check $ns "192.0.2.0/24 dev dummy2 tos 0 metric 1024" true
+	check_err $? "Appended route in hardware when should not"
+
+	ip -n $ns route prepend 192.0.2.0/24 dev dummy3 tos 0 metric 1024
+	fib4_trap_check $ns "192.0.2.0/24 dev dummy3 tos 0 metric 1024" false
+	check_err $? "Prepended route not in hardware when should"
+
+	fib4_trap_check $ns "192.0.2.0/24 dev dummy1 tos 0 metric 1024" true
+	check_err $? "Route was not replaced in hardware by prepended one"
+
+	log_test "IPv4 identical routes"
+
+	for i in $(seq 1 3); do
+		ip -n $ns link del dev dummy$i
+	done
+}
+
+fib_ipv4_tos_test()
+{
+	local ns=$1; shift
+
+	RET=0
+
+	ip -n $ns link add name dummy1 type dummy
+	ip -n $ns link set dev dummy1 up
+
+	ip -n $ns route add 192.0.2.0/24 dev dummy1 tos 0 metric 1024
+	fib4_trap_check $ns "192.0.2.0/24 dev dummy1 tos 0 metric 1024" false
+	check_err $? "Route not in hardware when should"
+
+	ip -n $ns route add 192.0.2.0/24 dev dummy1 tos 2 metric 1024
+	fib4_trap_check $ns "192.0.2.0/24 dev dummy1 tos 2 metric 1024" false
+	check_err $? "Highest TOS route not in hardware when should"
+
+	fib4_trap_check $ns "192.0.2.0/24 dev dummy1 tos 0 metric 1024" true
+	check_err $? "Lowest TOS route still in hardware when should not"
+
+	ip -n $ns route add 192.0.2.0/24 dev dummy1 tos 1 metric 1024
+	fib4_trap_check $ns "192.0.2.0/24 dev dummy1 tos 1 metric 1024" true
+	check_err $? "Middle TOS route in hardware when should not"
+
+	log_test "IPv4 routes with TOS"
+
+	ip -n $ns link del dev dummy1
+}
+
+fib_ipv4_metric_test()
+{
+	local ns=$1; shift
+
+	RET=0
+
+	ip -n $ns link add name dummy1 type dummy
+	ip -n $ns link set dev dummy1 up
+
+	ip -n $ns route add 192.0.2.0/24 dev dummy1 metric 1024
+	fib4_trap_check $ns "192.0.2.0/24 dev dummy1 metric 1024" false
+	check_err $? "Route not in hardware when should"
+
+	ip -n $ns route add 192.0.2.0/24 dev dummy1 metric 1022
+	fib4_trap_check $ns "192.0.2.0/24 dev dummy1 metric 1022" false
+	check_err $? "Lowest metric route not in hardware when should"
+
+	fib4_trap_check $ns "192.0.2.0/24 dev dummy1 metric 1024" true
+	check_err $? "Highest metric route still in hardware when should not"
+
+	ip -n $ns route add 192.0.2.0/24 dev dummy1 metric 1023
+	fib4_trap_check $ns "192.0.2.0/24 dev dummy1 metric 1023" true
+	check_err $? "Middle metric route in hardware when should not"
+
+	log_test "IPv4 routes with metric"
+
+	ip -n $ns link del dev dummy1
+}
+
+fib_ipv4_replace_test()
+{
+	local ns=$1; shift
+	local i
+
+	RET=0
+
+	for i in $(seq 1 2); do
+		ip -n $ns link add name dummy$i type dummy
+		ip -n $ns link set dev dummy$i up
+	done
+
+	ip -n $ns route add 192.0.2.0/24 dev dummy1 metric 1024
+	fib4_trap_check $ns "192.0.2.0/24 dev dummy1 metric 1024" false
+	check_err $? "Route not in hardware when should"
+
+	ip -n $ns route replace 192.0.2.0/24 dev dummy2 metric 1024
+	fib4_trap_check $ns "192.0.2.0/24 dev dummy2 metric 1024" false
+	check_err $? "Replacement route not in hardware when should"
+
+	# Add a route with an higher metric and make sure that replacing it
+	# does not affect the lower metric one.
+	ip -n $ns route add 192.0.2.0/24 dev dummy1 metric 1025
+	ip -n $ns route replace 192.0.2.0/24 dev dummy2 metric 1025
+
+	fib4_trap_check $ns "192.0.2.0/24 dev dummy2 metric 1024" false
+	check_err $? "Lowest metric route not in hardware when should"
+	fib4_trap_check $ns "192.0.2.0/24 dev dummy2 metric 1025" true
+	check_err $? "Highest metric route in hardware when should not"
+
+	log_test "IPv4 route replace"
+
+	for i in $(seq 1 2); do
+		ip -n $ns link del dev dummy$i
+	done
+}
+
+fib_ipv4_delete_test()
+{
+	local ns=$1; shift
+	local metric
+
+	RET=0
+
+	ip -n $ns link add name dummy1 type dummy
+	ip -n $ns link set dev dummy1 up
+
+	# Insert multiple routes with the same prefix and length and varying
+	# metrics. Make sure that throughout delete operations the lowest
+	# metric route is the one in hardware.
+	for metric in $(seq 1024 1026); do
+		ip -n $ns route add 192.0.2.0/24 dev dummy1 metric $metric
+	done
+
+	fib4_trap_check $ns "192.0.2.0/24 dev dummy1 metric 1024" false
+	check_err $? "Route not in hardware when should"
+
+	ip -n $ns route del 192.0.2.0/24 dev dummy1 metric 1024
+	fib4_trap_check $ns "192.0.2.0/24 dev dummy1 metric 1025" false
+	check_err $? "Lowest metric route not in hardware when should"
+
+	ip -n $ns route del 192.0.2.0/24 dev dummy1 metric 1026
+	fib4_trap_check $ns "192.0.2.0/24 dev dummy1 metric 1025" false
+	check_err $? "Sole route not in hardware when should"
+
+	log_test "IPv4 route delete"
+
+	ip -n $ns link del dev dummy1
+}
+
+fib_ipv4_plen_test()
+{
+	local ns=$1; shift
+
+	RET=0
+
+	ip -n $ns link add name dummy1 type dummy
+	ip -n $ns link set dev dummy1 up
+
+	# Add two routes with the same key and different prefix length and
+	# make sure both are in hardware. It can be verfied that both are
+	# sharing the same leaf by checking the /proc/net/fib_trie
+	ip -n $ns route add 192.0.2.0/24 dev dummy1
+	ip -n $ns route add 192.0.2.0/25 dev dummy1
+
+	fib4_trap_check $ns "192.0.2.0/24 dev dummy1" false
+	check_err $? "/24 not in hardware when should"
+
+	fib4_trap_check $ns "192.0.2.0/25 dev dummy1" false
+	check_err $? "/25 not in hardware when should"
+
+	log_test "IPv4 routes with different prefix length"
+
+	ip -n $ns link del dev dummy1
+}
+
+fib_ipv4_replay_metric_test()
+{
+	local ns=$1; shift
+	local devlink_dev=$1; shift
+
+	RET=0
+
+	ip -n $ns link add name dummy1 type dummy
+	ip -n $ns link set dev dummy1 up
+
+	ip -n $ns route add 192.0.2.0/24 dev dummy1 metric 1024
+	ip -n $ns route add 192.0.2.0/24 dev dummy1 metric 1025
+
+	devlink -N $ns dev reload $devlink_dev
+
+	fib4_trap_check $ns "192.0.2.0/24 dev dummy1 metric 1024" false
+	check_err $? "Lowest metric route not in hardware when should"
+
+	fib4_trap_check $ns "192.0.2.0/24 dev dummy1 metric 1025" true
+	check_err $? "Highest metric route in hardware when should not"
+
+	log_test "IPv4 routes replay - metric"
+
+	ip -n $ns link del dev dummy1
+}
+
+fib_ipv4_replay_tos_test()
+{
+	local ns=$1; shift
+	local devlink_dev=$1; shift
+
+	RET=0
+
+	ip -n $ns link add name dummy1 type dummy
+	ip -n $ns link set dev dummy1 up
+
+	ip -n $ns route add 192.0.2.0/24 dev dummy1 tos 0
+	ip -n $ns route add 192.0.2.0/24 dev dummy1 tos 1
+
+	devlink -N $ns dev reload $devlink_dev
+
+	fib4_trap_check $ns "192.0.2.0/24 dev dummy1 tos 1" false
+	check_err $? "Highest TOS route not in hardware when should"
+
+	fib4_trap_check $ns "192.0.2.0/24 dev dummy1 tos 0" true
+	check_err $? "Lowest TOS route in hardware when should not"
+
+	log_test "IPv4 routes replay - TOS"
+
+	ip -n $ns link del dev dummy1
+}
+
+fib_ipv4_replay_plen_test()
+{
+	local ns=$1; shift
+	local devlink_dev=$1; shift
+
+	RET=0
+
+	ip -n $ns link add name dummy1 type dummy
+	ip -n $ns link set dev dummy1 up
+
+	ip -n $ns route add 192.0.2.0/24 dev dummy1
+	ip -n $ns route add 192.0.2.0/25 dev dummy1
+
+	devlink -N $ns dev reload $devlink_dev
+
+	fib4_trap_check $ns "192.0.2.0/24 dev dummy1" false
+	check_err $? "/24 not in hardware when should"
+
+	fib4_trap_check $ns "192.0.2.0/25 dev dummy1" false
+	check_err $? "/25 not in hardware when should"
+
+	log_test "IPv4 routes replay - prefix length"
+
+	ip -n $ns link del dev dummy1
+}
+
+fib_ipv4_flush_test()
+{
+	local ns=$1; shift
+	local metric
+
+	RET=0
+
+	ip -n $ns link add name dummy1 type dummy
+	ip -n $ns link set dev dummy1 up
+
+	# Exercise the routes flushing code paths by inserting various
+	# prefix routes on a netdev and then deleting it.
+	for metric in $(seq 1 20); do
+		ip -n $ns route add 192.0.2.0/24 dev dummy1 metric $metric
+	done
+
+	ip -n $ns link del dev dummy1
+
+	log_test "IPv4 routes flushing"
+}
+
+fib_ipv6_add_test()
+{
+	local ns=$1; shift
+
+	RET=0
+
+	for i in $(seq 1 2); do
+		ip -n $ns link add name dummy$i type dummy
+		ip -n $ns link set dev dummy$i up
+	done
+
+	ip -n $ns route add 2001:db8:1::/64 dev dummy1 metric 1024
+	fib6_trap_check $ns "2001:db8:1::/64 dev dummy1 metric 1024" false
+	check_err $? "Route not in hardware when should"
+
+	ip -n $ns route append 2001:db8:1::/64 dev dummy2 metric 1024
+	fib6_trap_check $ns "2001:db8:1::/64 dev dummy2 metric 1024" true
+	check_err $? "Route in hardware when should not"
+
+	fib6_trap_check $ns "2001:db8:1::/64 dev dummy1 metric 1024" false
+	check_err $? "Route not in hardware after appending route"
+
+	log_test "IPv6 single route add"
+
+	for i in $(seq 1 2); do
+		ip -n $ns link del dev dummy$i
+	done
+}
+
+fib_ipv6_metric_test()
+{
+	local ns=$1; shift
+
+	RET=0
+
+	ip -n $ns link add name dummy1 type dummy
+	ip -n $ns link set dev dummy1 up
+
+	ip -n $ns route add 2001:db8:1::/64 dev dummy1 metric 1024
+	fib6_trap_check $ns "2001:db8:1::/64 dev dummy1 metric 1024" false
+	check_err $? "Route not in hardware when should"
+
+	ip -n $ns route add 2001:db8:1::/64 dev dummy1 metric 1022
+	fib6_trap_check $ns "2001:db8:1::/64 dev dummy1 metric 1022" false
+	check_err $? "Lowest metric route not in hardware when should"
+
+	fib6_trap_check $ns "2001:db8:1::/64 dev dummy1 metric 1024" true
+	check_err $? "Highest metric route still in hardware when should not"
+
+	ip -n $ns route add 2001:db8:1::/64 dev dummy1 metric 1023
+	fib6_trap_check $ns "2001:db8:1::/64 dev dummy1 metric 1023" true
+	check_err $? "Middle metric route in hardware when should not"
+
+	log_test "IPv6 routes with metric"
+
+	ip -n $ns link del dev dummy1
+}
+
+fib_ipv6_append_single_test()
+{
+	local ns=$1; shift
+
+	# When an IPv6 multipath route is added without the 'nexthop' keyword,
+	# different code paths are taken compared to when the keyword is used.
+	# This test tries to verify the former.
+	RET=0
+
+	for i in $(seq 1 2); do
+		ip -n $ns link add name dummy$i type dummy
+		ip -n $ns link set dev dummy$i up
+		ip -n $ns address add 2001:db8:$i::1/64 dev dummy$i
+	done
+
+	ip -n $ns route add 2001:db8:10::/64 via 2001:db8:1::2 metric 1024
+	fib6_trap_check $ns "2001:db8:10::/64 metric 1024" false
+	check_err $? "Route not in hardware when should"
+
+	ip -n $ns route append 2001:db8:10::/64 via 2001:db8:2::2 metric 1024
+	fib6_trap_check $ns "2001:db8:10::/64 metric 1024" false
+	check_err $? "Route not in hardware after appending"
+
+	ip -n $ns route add 2001:db8:10::/64 via 2001:db8:1::2 metric 1025
+	fib6_trap_check $ns "2001:db8:10::/64 metric 1025" true
+	check_err $? "Route in hardware when should not"
+
+	ip -n $ns route append 2001:db8:10::/64 via 2001:db8:2::2 metric 1025
+	fib6_trap_check $ns "2001:db8:10::/64 metric 1025" true
+	check_err $? "Route in hardware when should not after appending"
+
+	fib6_trap_check $ns "2001:db8:10::/64 metric 1024" false
+	check_err $? "Lowest metric route not in hardware when should"
+
+	log_test "IPv6 append single route without 'nexthop' keyword"
+
+	for i in $(seq 1 2); do
+		ip -n $ns link del dev dummy$i
+	done
+}
+
+fib_ipv6_replace_single_test()
+{
+	local ns=$1; shift
+	local i
+
+	RET=0
+
+	for i in $(seq 1 2); do
+		ip -n $ns link add name dummy$i type dummy
+		ip -n $ns link set dev dummy$i up
+	done
+
+	ip -n $ns route add 2001:db8:1::/64 dev dummy1 metric 1024
+	fib6_trap_check $ns "2001:db8:1::/64 dev dummy1 metric 1024" false
+	check_err $? "Route not in hardware when should"
+
+	ip -n $ns route replace 2001:db8:1::/64 dev dummy2 metric 1024
+	fib6_trap_check $ns "2001:db8:1::/64 dev dummy2 metric 1024" false
+	check_err $? "Replacement route not in hardware when should"
+
+	# Add a route with an higher metric and make sure that replacing it
+	# does not affect the lower metric one.
+	ip -n $ns route add 2001:db8:1::/64 dev dummy1 metric 1025
+	ip -n $ns route replace 2001:db8:1::/64 dev dummy2 metric 1025
+
+	fib6_trap_check $ns "2001:db8:1::/64 dev dummy2 metric 1024" false
+	check_err $? "Lowest metric route not in hardware when should"
+	fib6_trap_check $ns "2001:db8:1::/64 dev dummy2 metric 1025" true
+	check_err $? "Highest metric route in hardware when should not"
+
+	log_test "IPv6 single route replace"
+
+	for i in $(seq 1 2); do
+		ip -n $ns link del dev dummy$i
+	done
+}
+
+fib_ipv6_metric_multipath_test()
+{
+	local ns=$1; shift
+
+	RET=0
+
+	for i in $(seq 1 2); do
+		ip -n $ns link add name dummy$i type dummy
+		ip -n $ns link set dev dummy$i up
+		ip -n $ns address add 2001:db8:$i::1/64 dev dummy$i
+	done
+
+	ip -n $ns route add 2001:db8:10::/64 metric 1024 \
+		nexthop via 2001:db8:1::2 dev dummy1 \
+		nexthop via 2001:db8:2::2 dev dummy2
+	fib6_trap_check $ns "2001:db8:10::/64 metric 1024" false
+	check_err $? "Route not in hardware when should"
+
+	ip -n $ns route add 2001:db8:10::/64 metric 1022 \
+		nexthop via 2001:db8:1::2 dev dummy1 \
+		nexthop via 2001:db8:2::2 dev dummy2
+	fib6_trap_check $ns "2001:db8:10::/64 metric 1022" false
+	check_err $? "Lowest metric route not in hardware when should"
+
+	ip -n $ns route add 2001:db8:10::/64 metric 1023 \
+		nexthop via 2001:db8:1::2 dev dummy1 \
+		nexthop via 2001:db8:2::2 dev dummy2
+	fib6_trap_check $ns "2001:db8:10::/64 metric 1024" true
+	check_err $? "Highest metric route still in hardware when should not"
+
+	fib6_trap_check $ns "2001:db8:10::/64 metric 1023" true
+	check_err $? "Middle metric route in hardware when should not"
+
+	log_test "IPv6 multipath routes with metric"
+
+	for i in $(seq 1 2); do
+		ip -n $ns link del dev dummy$i
+	done
+}
+
+fib_ipv6_append_multipath_test()
+{
+	local ns=$1; shift
+
+	RET=0
+
+	for i in $(seq 1 3); do
+		ip -n $ns link add name dummy$i type dummy
+		ip -n $ns link set dev dummy$i up
+		ip -n $ns address add 2001:db8:$i::1/64 dev dummy$i
+	done
+
+	ip -n $ns route add 2001:db8:10::/64 metric 1024 \
+		nexthop via 2001:db8:1::2 dev dummy1
+	fib6_trap_check $ns "2001:db8:10::/64 metric 1024" false
+	check_err $? "Route not in hardware when should"
+
+	ip -n $ns route append 2001:db8:10::/64 metric 1024 \
+		nexthop via 2001:db8:2::2 dev dummy2 \
+		nexthop via 2001:db8:3::2 dev dummy3
+	fib6_trap_check $ns "2001:db8:10::/64 metric 1024" false
+	check_err $? "Route not in hardware after appending"
+
+	ip -n $ns route add 2001:db8:10::/64 metric 1025 \
+		nexthop via 2001:db8:1::2 dev dummy1
+	fib6_trap_check $ns "2001:db8:10::/64 metric 1025" true
+	check_err $? "Route in hardware when should not"
+
+	ip -n $ns route append 2001:db8:10::/64 metric 1025 \
+		nexthop via 2001:db8:2::2 dev dummy2 \
+		nexthop via 2001:db8:3::2 dev dummy3
+	fib6_trap_check $ns "2001:db8:10::/64 metric 1025" true
+	check_err $? "Route in hardware when should not after appending"
+
+	fib6_trap_check $ns "2001:db8:10::/64 metric 1024" false
+	check_err $? "Lowest metric route not in hardware when should"
+
+	log_test "IPv6 append multipath route with 'nexthop' keyword"
+
+	for i in $(seq 1 3); do
+		ip -n $ns link del dev dummy$i
+	done
+}
+
+fib_ipv6_replace_multipath_test()
+{
+	local ns=$1; shift
+	local i
+
+	RET=0
+
+	for i in $(seq 1 3); do
+		ip -n $ns link add name dummy$i type dummy
+		ip -n $ns link set dev dummy$i up
+		ip -n $ns address add 2001:db8:$i::1/64 dev dummy$i
+	done
+
+	ip -n $ns route add 2001:db8:10::/64 metric 1024 \
+		nexthop via 2001:db8:1::2 dev dummy1 \
+		nexthop via 2001:db8:2::2 dev dummy2
+	fib6_trap_check $ns "2001:db8:10::/64 metric 1024" false
+	check_err $? "Route not in hardware when should"
+
+	ip -n $ns route replace 2001:db8:10::/64 metric 1024 \
+		nexthop via 2001:db8:1::2 dev dummy1 \
+		nexthop via 2001:db8:3::2 dev dummy3
+	fib6_trap_check $ns "2001:db8:10::/64 metric 1024" false
+	check_err $? "Replacement route not in hardware when should"
+
+	# Add a route with an higher metric and make sure that replacing it
+	# does not affect the lower metric one.
+	ip -n $ns route add 2001:db8:10::/64 metric 1025 \
+		nexthop via 2001:db8:1::2 dev dummy1 \
+		nexthop via 2001:db8:2::2 dev dummy2
+	ip -n $ns route replace 2001:db8:10::/64 metric 1025 \
+		nexthop via 2001:db8:1::2 dev dummy1 \
+		nexthop via 2001:db8:3::2 dev dummy3
+
+	fib6_trap_check $ns "2001:db8:10::/64 metric 1024" false
+	check_err $? "Lowest metric route not in hardware when should"
+	fib6_trap_check $ns "2001:db8:10::/64 metric 1025" true
+	check_err $? "Highest metric route in hardware when should not"
+
+	log_test "IPv6 multipath route replace"
+
+	for i in $(seq 1 3); do
+		ip -n $ns link del dev dummy$i
+	done
+}
+
+fib_ipv6_append_multipath_to_single_test()
+{
+	local ns=$1; shift
+
+	# Test that when the first route in the leaf is not a multipath route
+	# and we try to append a multipath route with the same metric to it, it
+	# is not notified.
+	RET=0
+
+	for i in $(seq 1 2); do
+		ip -n $ns link add name dummy$i type dummy
+		ip -n $ns link set dev dummy$i up
+		ip -n $ns address add 2001:db8:$i::1/64 dev dummy$i
+	done
+
+	ip -n $ns route add 2001:db8:10::/64 dev dummy1 metric 1024
+	fib6_trap_check $ns "2001:db8:10::/64 dev dummy1 metric 1024" false
+	check_err $? "Route not in hardware when should"
+
+	ip -n $ns route append 2001:db8:10::/64 metric 1024 \
+		nexthop via 2001:db8:2::2 dev dummy2
+	fib6_trap_check $ns "2001:db8:10::/64 dev dummy2 metric 1024" true
+	check_err $? "Route in hardware when should not"
+
+	fib6_trap_check $ns "2001:db8:10::/64 dev dummy1 metric 1024" false
+	check_err $? "Route not in hardware after append"
+
+	log_test "IPv6 append multipath route to non-multipath route"
+
+	for i in $(seq 1 2); do
+		ip -n $ns link del dev dummy$i
+	done
+}
+
+fib_ipv6_delete_single_test()
+{
+	local ns=$1; shift
+
+	# Test various deletion scenarios, where only a single route is
+	# deleted from the FIB node.
+	for i in $(seq 1 2); do
+		ip -n $ns link add name dummy$i type dummy
+		ip -n $ns link set dev dummy$i up
+		ip -n $ns address add 2001:db8:$i::1/64 dev dummy$i
+	done
+
+	# Test deletion of a single route when it is the only route in the FIB
+	# node.
+	RET=0
+
+	ip -n $ns route add 2001:db8:10::/64 dev dummy1 metric 1024
+	ip -n $ns route del 2001:db8:10::/64 dev dummy1 metric 1024
+
+	log_test "IPv6 delete sole single route"
+
+	# Test that deletion of last route does not affect the first one.
+	RET=0
+
+	ip -n $ns route add 2001:db8:10::/64 dev dummy1 metric 1024
+	ip -n $ns route add 2001:db8:10::/64 dev dummy1 metric 1025
+	ip -n $ns route del 2001:db8:10::/64 dev dummy1 metric 1025
+
+	fib6_trap_check $ns "2001:db8:10::/64 dev dummy1 metric 1024" false
+	check_err $? "Route not in hardware after deleting higher metric route"
+
+	log_test "IPv6 delete single route not in hardware"
+
+	ip -n $ns route del 2001:db8:10::/64 dev dummy1 metric 1024
+
+	# Test that first route is replaced by next single route in the FIB
+	# node.
+	RET=0
+
+	ip -n $ns route add 2001:db8:10::/64 dev dummy1 metric 1024
+	ip -n $ns route add 2001:db8:10::/64 dev dummy1 metric 1025
+	ip -n $ns route del 2001:db8:10::/64 dev dummy1 metric 1024
+
+	fib6_trap_check $ns "2001:db8:10::/64 dev dummy1 metric 1025" false
+	check_err $? "Route not in hardware after deleting lowest metric route"
+
+	log_test "IPv6 delete single route - replaced by single"
+
+	ip -n $ns route del 2001:db8:10::/64 dev dummy1 metric 1025
+
+	# Test that first route is replaced by next multipath route in the FIB
+	# node.
+	RET=0
+
+	ip -n $ns route add 2001:db8:10::/64 dev dummy1 metric 1024
+	ip -n $ns route add 2001:db8:10::/64 metric 1025 \
+		nexthop via 2001:db8:1::2 dev dummy1 \
+		nexthop via 2001:db8:2::2 dev dummy2
+	ip -n $ns route del 2001:db8:10::/64 dev dummy1 metric 1024
+
+	fib6_trap_check $ns "2001:db8:10::/64 metric 1025" false
+	check_err $? "Route not in hardware after deleting lowest metric route"
+
+	log_test "IPv6 delete single route - replaced by multipath"
+
+	ip -n $ns route del 2001:db8:10::/64 metric 1025
+
+	# Test deletion of a single nexthop from a multipath route.
+	ip -n $ns route add 2001:db8:10::/64 metric 1024 \
+		nexthop via 2001:db8:1::2 dev dummy1 \
+		nexthop via 2001:db8:2::2 dev dummy2
+	ip -n $ns route del 2001:db8:10::/64 metric 1024 \
+		nexthop via 2001:db8:1::2 dev dummy1
+
+	fib6_trap_check $ns "2001:db8:10::/64 metric 1024" false
+	check_err $? "Route not in hardware after deleting a single nexthop"
+
+	log_test "IPv6 delete single nexthop"
+
+	ip -n $ns route del 2001:db8:10::/64 metric 1024
+
+	for i in $(seq 1 2); do
+		ip -n $ns link del dev dummy$i
+	done
+}
+
+fib_ipv6_delete_multipath_test()
+{
+	local ns=$1; shift
+
+	# Test various deletion scenarios, where an entire multipath route is
+	# deleted from the FIB node.
+	for i in $(seq 1 2); do
+		ip -n $ns link add name dummy$i type dummy
+		ip -n $ns link set dev dummy$i up
+		ip -n $ns address add 2001:db8:$i::1/64 dev dummy$i
+	done
+
+	# Test deletion of a multipath route when it is the only route in the
+	# FIB node.
+	RET=0
+
+	ip -n $ns route add 2001:db8:10::/64 metric 1024 \
+		nexthop via 2001:db8:1::2 dev dummy1 \
+		nexthop via 2001:db8:2::2 dev dummy2
+	ip -n $ns route del 2001:db8:10::/64 metric 1024
+
+	log_test "IPv6 delete sole multipath route"
+
+	# Test that deletion of last route does not affect the first one.
+	RET=0
+
+	ip -n $ns route add 2001:db8:10::/64 metric 1024 \
+		nexthop via 2001:db8:1::2 dev dummy1 \
+		nexthop via 2001:db8:2::2 dev dummy2
+	ip -n $ns route add 2001:db8:10::/64 metric 1025 \
+		nexthop via 2001:db8:1::2 dev dummy1 \
+		nexthop via 2001:db8:2::2 dev dummy2
+	ip -n $ns route del 2001:db8:10::/64 metric 1025
+
+	fib6_trap_check $ns "2001:db8:10::/64 metric 1024" false
+	check_err $? "Route not in hardware after deleting higher metric route"
+
+	log_test "IPv6 delete multipath route not in hardware"
+
+	ip -n $ns route del 2001:db8:10::/64 metric 1024
+
+	# Test that first route is replaced by next single route in the FIB
+	# node.
+	RET=0
+
+	ip -n $ns route add 2001:db8:10::/64 metric 1024 \
+		nexthop via 2001:db8:1::2 dev dummy1 \
+		nexthop via 2001:db8:2::2 dev dummy2
+	ip -n $ns route add 2001:db8:10::/64 dev dummy1 metric 1025
+	ip -n $ns route del 2001:db8:10::/64 metric 1024
+
+	fib6_trap_check $ns "2001:db8:10::/64 dev dummy1 metric 1025" false
+	check_err $? "Route not in hardware after deleting lowest metric route"
+
+	log_test "IPv6 delete multipath route - replaced by single"
+
+	ip -n $ns route del 2001:db8:10::/64 dev dummy1 metric 1025
+
+	# Test that first route is replaced by next multipath route in the FIB
+	# node.
+	RET=0
+
+	ip -n $ns route add 2001:db8:10::/64 metric 1024 \
+		nexthop via 2001:db8:1::2 dev dummy1 \
+		nexthop via 2001:db8:2::2 dev dummy2
+	ip -n $ns route add 2001:db8:10::/64 metric 1025 \
+		nexthop via 2001:db8:1::2 dev dummy1 \
+		nexthop via 2001:db8:2::2 dev dummy2
+	ip -n $ns route del 2001:db8:10::/64 metric 1024
+
+	fib6_trap_check $ns "2001:db8:10::/64 metric 1025" false
+	check_err $? "Route not in hardware after deleting lowest metric route"
+
+	log_test "IPv6 delete multipath route - replaced by multipath"
+
+	ip -n $ns route del 2001:db8:10::/64 metric 1025
+
+	for i in $(seq 1 2); do
+		ip -n $ns link del dev dummy$i
+	done
+}
+
+fib_ipv6_replay_single_test()
+{
+	local ns=$1; shift
+	local devlink_dev=$1; shift
+
+	RET=0
+
+	for i in $(seq 1 2); do
+		ip -n $ns link add name dummy$i type dummy
+		ip -n $ns link set dev dummy$i up
+	done
+
+	ip -n $ns route add 2001:db8:1::/64 dev dummy1
+	ip -n $ns route append 2001:db8:1::/64 dev dummy2
+
+	devlink -N $ns dev reload $devlink_dev
+
+	fib6_trap_check $ns "2001:db8:1::/64 dev dummy1" false
+	check_err $? "First route not in hardware when should"
+
+	fib6_trap_check $ns "2001:db8:1::/64 dev dummy2" true
+	check_err $? "Second route in hardware when should not"
+
+	log_test "IPv6 routes replay - single route"
+
+	for i in $(seq 1 2); do
+		ip -n $ns link del dev dummy$i
+	done
+}
+
+fib_ipv6_replay_multipath_test()
+{
+	local ns=$1; shift
+	local devlink_dev=$1; shift
+
+	RET=0
+
+	for i in $(seq 1 2); do
+		ip -n $ns link add name dummy$i type dummy
+		ip -n $ns link set dev dummy$i up
+		ip -n $ns address add 2001:db8:$i::1/64 dev dummy$i
+	done
+
+	ip -n $ns route add 2001:db8:10::/64 metric 1024 \
+		nexthop via 2001:db8:1::2 dev dummy1 \
+		nexthop via 2001:db8:2::2 dev dummy2
+	ip -n $ns route add 2001:db8:10::/64 metric 1025 \
+		nexthop via 2001:db8:1::2 dev dummy1 \
+		nexthop via 2001:db8:2::2 dev dummy2
+
+	devlink -N $ns dev reload $devlink_dev
+
+	fib6_trap_check $ns "2001:db8:10::/64 metric 1024" false
+	check_err $? "First route not in hardware when should"
+
+	fib6_trap_check $ns "2001:db8:10::/64 metric 1025" true
+	check_err $? "Second route in hardware when should not"
+
+	log_test "IPv6 routes replay - multipath route"
+
+	for i in $(seq 1 2); do
+		ip -n $ns link del dev dummy$i
+	done
+}
-- 
cgit v1.2.3-59-g8ed1b


From ffdc5149c5e2328bb8f0e136efd4935829fe89b3 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Tue, 14 Jan 2020 13:23:17 +0200
Subject: selftests: netdevsim: Add test for FIB offload API

Test various aspects of the FIB offload API on top of the netdevsim
implementation. Both good and bad flows are tested.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../testing/selftests/drivers/net/netdevsim/fib.sh | 341 +++++++++++++++++++++
 1 file changed, 341 insertions(+)
 create mode 100755 tools/testing/selftests/drivers/net/netdevsim/fib.sh

diff --git a/tools/testing/selftests/drivers/net/netdevsim/fib.sh b/tools/testing/selftests/drivers/net/netdevsim/fib.sh
new file mode 100755
index 000000000000..2f87c3be76a9
--- /dev/null
+++ b/tools/testing/selftests/drivers/net/netdevsim/fib.sh
@@ -0,0 +1,341 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+#
+# This test is for checking the FIB offload API. It makes use of netdevsim
+# which registers a listener to the FIB notification chain.
+
+lib_dir=$(dirname $0)/../../../net/forwarding
+
+ALL_TESTS="
+	ipv4_identical_routes
+	ipv4_tos
+	ipv4_metric
+	ipv4_replace
+	ipv4_delete
+	ipv4_plen
+	ipv4_replay
+	ipv4_flush
+	ipv4_error_path
+	ipv6_add
+	ipv6_metric
+	ipv6_append_single
+	ipv6_replace_single
+	ipv6_metric_multipath
+	ipv6_append_multipath
+	ipv6_replace_multipath
+	ipv6_append_multipath_to_single
+	ipv6_delete_single
+	ipv6_delete_multipath
+	ipv6_replay_single
+	ipv6_replay_multipath
+	ipv6_error_path
+"
+NETDEVSIM_PATH=/sys/bus/netdevsim/
+DEV_ADDR=1337
+DEV=netdevsim${DEV_ADDR}
+DEVLINK_DEV=netdevsim/${DEV}
+SYSFS_NET_DIR=/sys/bus/netdevsim/devices/$DEV/net/
+NUM_NETIFS=0
+source $lib_dir/lib.sh
+source $lib_dir/devlink_lib.sh
+source $lib_dir/fib_offload_lib.sh
+
+ipv4_identical_routes()
+{
+	fib_ipv4_identical_routes_test "testns1"
+}
+
+ipv4_tos()
+{
+	fib_ipv4_tos_test "testns1"
+}
+
+ipv4_metric()
+{
+	fib_ipv4_metric_test "testns1"
+}
+
+ipv4_replace()
+{
+	fib_ipv4_replace_test "testns1"
+}
+
+ipv4_delete()
+{
+	fib_ipv4_delete_test "testns1"
+}
+
+ipv4_plen()
+{
+	fib_ipv4_plen_test "testns1"
+}
+
+ipv4_replay_metric()
+{
+	fib_ipv4_replay_metric_test "testns1" "$DEVLINK_DEV"
+}
+
+ipv4_replay_tos()
+{
+	fib_ipv4_replay_tos_test "testns1" "$DEVLINK_DEV"
+}
+
+ipv4_replay_plen()
+{
+	fib_ipv4_replay_plen_test "testns1" "$DEVLINK_DEV"
+}
+
+ipv4_replay()
+{
+	ipv4_replay_metric
+	ipv4_replay_tos
+	ipv4_replay_plen
+}
+
+ipv4_flush()
+{
+	fib_ipv4_flush_test "testns1"
+}
+
+ipv4_error_path_add()
+{
+	local lsb
+
+	RET=0
+
+	ip -n testns1 link add name dummy1 type dummy
+	ip -n testns1 link set dev dummy1 up
+
+	devlink -N testns1 resource set $DEVLINK_DEV path IPv4/fib size 10
+	devlink -N testns1 dev reload $DEVLINK_DEV
+
+	for lsb in $(seq 1 20); do
+		ip -n testns1 route add 192.0.2.${lsb}/32 dev dummy1 \
+			&> /dev/null
+	done
+
+	log_test "IPv4 error path - add"
+
+	ip -n testns1 link del dev dummy1
+}
+
+ipv4_error_path_replay()
+{
+	local lsb
+
+	RET=0
+
+	ip -n testns1 link add name dummy1 type dummy
+	ip -n testns1 link set dev dummy1 up
+
+	devlink -N testns1 resource set $DEVLINK_DEV path IPv4/fib size 100
+	devlink -N testns1 dev reload $DEVLINK_DEV
+
+	for lsb in $(seq 1 20); do
+		ip -n testns1 route add 192.0.2.${lsb}/32 dev dummy1
+	done
+
+	devlink -N testns1 resource set $DEVLINK_DEV path IPv4/fib size 10
+	devlink -N testns1 dev reload $DEVLINK_DEV &> /dev/null
+
+	log_test "IPv4 error path - replay"
+
+	ip -n testns1 link del dev dummy1
+
+	# Successfully reload after deleting all the routes.
+	devlink -N testns1 resource set $DEVLINK_DEV path IPv4/fib size 100
+	devlink -N testns1 dev reload $DEVLINK_DEV
+}
+
+ipv4_error_path()
+{
+	# Test the different error paths of the notifiers by limiting the size
+	# of the "IPv4/fib" resource.
+	ipv4_error_path_add
+	ipv4_error_path_replay
+}
+
+ipv6_add()
+{
+	fib_ipv6_add_test "testns1"
+}
+
+ipv6_metric()
+{
+	fib_ipv6_metric_test "testns1"
+}
+
+ipv6_append_single()
+{
+	fib_ipv6_append_single_test "testns1"
+}
+
+ipv6_replace_single()
+{
+	fib_ipv6_replace_single_test "testns1"
+}
+
+ipv6_metric_multipath()
+{
+	fib_ipv6_metric_multipath_test "testns1"
+}
+
+ipv6_append_multipath()
+{
+	fib_ipv6_append_multipath_test "testns1"
+}
+
+ipv6_replace_multipath()
+{
+	fib_ipv6_replace_multipath_test "testns1"
+}
+
+ipv6_append_multipath_to_single()
+{
+	fib_ipv6_append_multipath_to_single_test "testns1"
+}
+
+ipv6_delete_single()
+{
+	fib_ipv6_delete_single_test "testns1"
+}
+
+ipv6_delete_multipath()
+{
+	fib_ipv6_delete_multipath_test "testns1"
+}
+
+ipv6_replay_single()
+{
+	fib_ipv6_replay_single_test "testns1" "$DEVLINK_DEV"
+}
+
+ipv6_replay_multipath()
+{
+	fib_ipv6_replay_multipath_test "testns1" "$DEVLINK_DEV"
+}
+
+ipv6_error_path_add_single()
+{
+	local lsb
+
+	RET=0
+
+	ip -n testns1 link add name dummy1 type dummy
+	ip -n testns1 link set dev dummy1 up
+
+	devlink -N testns1 resource set $DEVLINK_DEV path IPv6/fib size 10
+	devlink -N testns1 dev reload $DEVLINK_DEV
+
+	for lsb in $(seq 1 20); do
+		ip -n testns1 route add 2001:db8:1::${lsb}/128 dev dummy1 \
+			&> /dev/null
+	done
+
+	log_test "IPv6 error path - add single"
+
+	ip -n testns1 link del dev dummy1
+}
+
+ipv6_error_path_add_multipath()
+{
+	local lsb
+
+	RET=0
+
+	for i in $(seq 1 2); do
+		ip -n testns1 link add name dummy$i type dummy
+		ip -n testns1 link set dev dummy$i up
+		ip -n testns1 address add 2001:db8:$i::1/64 dev dummy$i
+	done
+
+	devlink -N testns1 resource set $DEVLINK_DEV path IPv6/fib size 10
+	devlink -N testns1 dev reload $DEVLINK_DEV
+
+	for lsb in $(seq 1 20); do
+		ip -n testns1 route add 2001:db8:10::${lsb}/128 \
+			nexthop via 2001:db8:1::2 dev dummy1 \
+			nexthop via 2001:db8:2::2 dev dummy2 &> /dev/null
+	done
+
+	log_test "IPv6 error path - add multipath"
+
+	for i in $(seq 1 2); do
+		ip -n testns1 link del dev dummy$i
+	done
+}
+
+ipv6_error_path_replay()
+{
+	local lsb
+
+	RET=0
+
+	ip -n testns1 link add name dummy1 type dummy
+	ip -n testns1 link set dev dummy1 up
+
+	devlink -N testns1 resource set $DEVLINK_DEV path IPv6/fib size 100
+	devlink -N testns1 dev reload $DEVLINK_DEV
+
+	for lsb in $(seq 1 20); do
+		ip -n testns1 route add 2001:db8:1::${lsb}/128 dev dummy1
+	done
+
+	devlink -N testns1 resource set $DEVLINK_DEV path IPv6/fib size 10
+	devlink -N testns1 dev reload $DEVLINK_DEV &> /dev/null
+
+	log_test "IPv6 error path - replay"
+
+	ip -n testns1 link del dev dummy1
+
+	# Successfully reload after deleting all the routes.
+	devlink -N testns1 resource set $DEVLINK_DEV path IPv6/fib size 100
+	devlink -N testns1 dev reload $DEVLINK_DEV
+}
+
+ipv6_error_path()
+{
+	# Test the different error paths of the notifiers by limiting the size
+	# of the "IPv6/fib" resource.
+	ipv6_error_path_add_single
+	ipv6_error_path_add_multipath
+	ipv6_error_path_replay
+}
+
+setup_prepare()
+{
+	local netdev
+
+	modprobe netdevsim &> /dev/null
+
+	echo "$DEV_ADDR 1" > ${NETDEVSIM_PATH}/new_device
+	while [ ! -d $SYSFS_NET_DIR ] ; do :; done
+
+	ip netns add testns1
+	if [ $? -ne 0 ]; then
+		echo "Failed to add netns \"testns1\""
+		exit 1
+	fi
+
+	devlink dev reload $DEVLINK_DEV netns testns1
+	if [ $? -ne 0 ]; then
+		echo "Failed to reload into netns \"testns1\""
+		exit 1
+	fi
+}
+
+cleanup()
+{
+	pre_cleanup
+	ip netns del testns1
+	echo "$DEV_ADDR" > ${NETDEVSIM_PATH}/del_device
+	modprobe -r netdevsim &> /dev/null
+}
+
+trap cleanup EXIT
+
+setup_prepare
+
+tests_run
+
+exit $EXIT_STATUS
-- 
cgit v1.2.3-59-g8ed1b


From 212a37c225aed5d7dbe20cb1c1881bf7d35cbc54 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Tue, 14 Jan 2020 13:23:18 +0200
Subject: selftests: mlxsw: Add test for FIB offload API

The test reuses the common FIB offload tests in order to make sure that
mlxsw correctly implements FIB offload.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/drivers/net/mlxsw/fib.sh | 180 +++++++++++++++++++++++
 1 file changed, 180 insertions(+)
 create mode 100755 tools/testing/selftests/drivers/net/mlxsw/fib.sh

diff --git a/tools/testing/selftests/drivers/net/mlxsw/fib.sh b/tools/testing/selftests/drivers/net/mlxsw/fib.sh
new file mode 100755
index 000000000000..45115f81c2b1
--- /dev/null
+++ b/tools/testing/selftests/drivers/net/mlxsw/fib.sh
@@ -0,0 +1,180 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+#
+# This test is for checking the FIB offload API on top of mlxsw.
+
+lib_dir=$(dirname $0)/../../../net/forwarding
+
+ALL_TESTS="
+	ipv4_identical_routes
+	ipv4_tos
+	ipv4_metric
+	ipv4_replace
+	ipv4_delete
+	ipv4_plen
+	ipv4_replay
+	ipv4_flush
+	ipv6_add
+	ipv6_metric
+	ipv6_append_single
+	ipv6_replace_single
+	ipv6_metric_multipath
+	ipv6_append_multipath
+	ipv6_replace_multipath
+	ipv6_append_multipath_to_single
+	ipv6_delete_single
+	ipv6_delete_multipath
+	ipv6_replay_single
+	ipv6_replay_multipath
+"
+NUM_NETIFS=0
+source $lib_dir/lib.sh
+source $lib_dir/devlink_lib.sh
+source $lib_dir/fib_offload_lib.sh
+
+ipv4_identical_routes()
+{
+	fib_ipv4_identical_routes_test "testns1"
+}
+
+ipv4_tos()
+{
+	fib_ipv4_tos_test "testns1"
+}
+
+ipv4_metric()
+{
+	fib_ipv4_metric_test "testns1"
+}
+
+ipv4_replace()
+{
+	fib_ipv4_replace_test "testns1"
+}
+
+ipv4_delete()
+{
+	fib_ipv4_delete_test "testns1"
+}
+
+ipv4_plen()
+{
+	fib_ipv4_plen_test "testns1"
+}
+
+ipv4_replay_metric()
+{
+	fib_ipv4_replay_metric_test "testns1" "$DEVLINK_DEV"
+}
+
+ipv4_replay_tos()
+{
+	fib_ipv4_replay_tos_test "testns1" "$DEVLINK_DEV"
+}
+
+ipv4_replay_plen()
+{
+	fib_ipv4_replay_plen_test "testns1" "$DEVLINK_DEV"
+}
+
+ipv4_replay()
+{
+	ipv4_replay_metric
+	ipv4_replay_tos
+	ipv4_replay_plen
+}
+
+ipv4_flush()
+{
+	fib_ipv4_flush_test "testns1"
+}
+
+ipv6_add()
+{
+	fib_ipv6_add_test "testns1"
+}
+
+ipv6_metric()
+{
+	fib_ipv6_metric_test "testns1"
+}
+
+ipv6_append_single()
+{
+	fib_ipv6_append_single_test "testns1"
+}
+
+ipv6_replace_single()
+{
+	fib_ipv6_replace_single_test "testns1"
+}
+
+ipv6_metric_multipath()
+{
+	fib_ipv6_metric_multipath_test "testns1"
+}
+
+ipv6_append_multipath()
+{
+	fib_ipv6_append_multipath_test "testns1"
+}
+
+ipv6_replace_multipath()
+{
+	fib_ipv6_replace_multipath_test "testns1"
+}
+
+ipv6_append_multipath_to_single()
+{
+	fib_ipv6_append_multipath_to_single_test "testns1"
+}
+
+ipv6_delete_single()
+{
+	fib_ipv6_delete_single_test "testns1"
+}
+
+ipv6_delete_multipath()
+{
+	fib_ipv6_delete_multipath_test "testns1"
+}
+
+ipv6_replay_single()
+{
+	fib_ipv6_replay_single_test "testns1" "$DEVLINK_DEV"
+}
+
+ipv6_replay_multipath()
+{
+	fib_ipv6_replay_multipath_test "testns1" "$DEVLINK_DEV"
+}
+
+setup_prepare()
+{
+	ip netns add testns1
+	if [ $? -ne 0 ]; then
+		echo "Failed to add netns \"testns1\""
+		exit 1
+	fi
+
+	devlink dev reload $DEVLINK_DEV netns testns1
+	if [ $? -ne 0 ]; then
+		echo "Failed to reload into netns \"testns1\""
+		exit 1
+	fi
+}
+
+cleanup()
+{
+	pre_cleanup
+	devlink -N testns1 dev reload $DEVLINK_DEV netns $$
+	ip netns del testns1
+}
+
+trap cleanup EXIT
+
+setup_prepare
+
+tests_run
+
+exit $EXIT_STATUS
-- 
cgit v1.2.3-59-g8ed1b


From 5a46facbbcd454985992b5109185329aebf82a02 Mon Sep 17 00:00:00 2001
From: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Date: Tue, 14 Jan 2020 19:56:07 +0200
Subject: net: bridge: vlan: add helpers to check for vlan id/range validity

Add helpers to check if a vlan id or range are valid. The range helper
must be called when range start or end are detected.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/bridge/br_netlink.c | 13 +++----------
 net/bridge/br_private.h | 31 +++++++++++++++++++++++++++++++
 2 files changed, 34 insertions(+), 10 deletions(-)

diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c
index 60136575aea4..14100e8653e6 100644
--- a/net/bridge/br_netlink.c
+++ b/net/bridge/br_netlink.c
@@ -568,17 +568,13 @@ static int br_process_vlan_info(struct net_bridge *br,
 				bool *changed,
 				struct netlink_ext_ack *extack)
 {
-	if (!vinfo_curr->vid || vinfo_curr->vid >= VLAN_VID_MASK)
+	if (!br_vlan_valid_id(vinfo_curr->vid))
 		return -EINVAL;
 
 	if (vinfo_curr->flags & BRIDGE_VLAN_INFO_RANGE_BEGIN) {
-		/* check if we are already processing a range */
-		if (*vinfo_last)
+		if (!br_vlan_valid_range(vinfo_curr, *vinfo_last))
 			return -EINVAL;
 		*vinfo_last = vinfo_curr;
-		/* don't allow range of pvids */
-		if ((*vinfo_last)->flags & BRIDGE_VLAN_INFO_PVID)
-			return -EINVAL;
 		return 0;
 	}
 
@@ -586,10 +582,7 @@ static int br_process_vlan_info(struct net_bridge *br,
 		struct bridge_vlan_info tmp_vinfo;
 		int v, err;
 
-		if (!(vinfo_curr->flags & BRIDGE_VLAN_INFO_RANGE_END))
-			return -EINVAL;
-
-		if (vinfo_curr->vid <= (*vinfo_last)->vid)
+		if (!br_vlan_valid_range(vinfo_curr, *vinfo_last))
 			return -EINVAL;
 
 		memcpy(&tmp_vinfo, *vinfo_last,
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index f540f3bdf294..dbc0089e2c1a 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -507,6 +507,37 @@ static inline bool nbp_state_should_learn(const struct net_bridge_port *p)
 	return p->state == BR_STATE_LEARNING || p->state == BR_STATE_FORWARDING;
 }
 
+static inline bool br_vlan_valid_id(u16 vid)
+{
+	return vid > 0 && vid < VLAN_VID_MASK;
+}
+
+static inline bool br_vlan_valid_range(const struct bridge_vlan_info *cur,
+				       const struct bridge_vlan_info *last)
+{
+	/* pvid flag is not allowed in ranges */
+	if (cur->flags & BRIDGE_VLAN_INFO_PVID)
+		return false;
+
+	/* check for required range flags */
+	if (!(cur->flags & (BRIDGE_VLAN_INFO_RANGE_BEGIN |
+			    BRIDGE_VLAN_INFO_RANGE_END)))
+		return false;
+
+	/* when cur is the range end, check if:
+	 *  - it has range start flag
+	 *  - range ids are invalid (end is equal to or before start)
+	 */
+	if (last) {
+		if (cur->flags & BRIDGE_VLAN_INFO_RANGE_BEGIN)
+			return false;
+		else if (cur->vid <= last->vid)
+			return false;
+	}
+
+	return true;
+}
+
 static inline int br_opt_get(const struct net_bridge *br,
 			     enum net_bridge_opts opt)
 {
-- 
cgit v1.2.3-59-g8ed1b


From 8f4cc940a149b9fe013a191d6d8dc87aee9a204f Mon Sep 17 00:00:00 2001
From: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Date: Tue, 14 Jan 2020 19:56:08 +0200
Subject: net: bridge: netlink: add extack error messages when processing vlans

Add extack messages on vlan processing errors. We need to move the flags
missing check after the "last" check since we may have "last" set but
lack a range end flag in the next entry.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/bridge/br_netlink.c |  6 +++---
 net/bridge/br_private.h | 38 +++++++++++++++++++++++++++-----------
 2 files changed, 30 insertions(+), 14 deletions(-)

diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c
index 14100e8653e6..40942cece51a 100644
--- a/net/bridge/br_netlink.c
+++ b/net/bridge/br_netlink.c
@@ -568,11 +568,11 @@ static int br_process_vlan_info(struct net_bridge *br,
 				bool *changed,
 				struct netlink_ext_ack *extack)
 {
-	if (!br_vlan_valid_id(vinfo_curr->vid))
+	if (!br_vlan_valid_id(vinfo_curr->vid, extack))
 		return -EINVAL;
 
 	if (vinfo_curr->flags & BRIDGE_VLAN_INFO_RANGE_BEGIN) {
-		if (!br_vlan_valid_range(vinfo_curr, *vinfo_last))
+		if (!br_vlan_valid_range(vinfo_curr, *vinfo_last, extack))
 			return -EINVAL;
 		*vinfo_last = vinfo_curr;
 		return 0;
@@ -582,7 +582,7 @@ static int br_process_vlan_info(struct net_bridge *br,
 		struct bridge_vlan_info tmp_vinfo;
 		int v, err;
 
-		if (!br_vlan_valid_range(vinfo_curr, *vinfo_last))
+		if (!br_vlan_valid_range(vinfo_curr, *vinfo_last, extack))
 			return -EINVAL;
 
 		memcpy(&tmp_vinfo, *vinfo_last,
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index dbc0089e2c1a..a7dddc5d7790 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -507,32 +507,48 @@ static inline bool nbp_state_should_learn(const struct net_bridge_port *p)
 	return p->state == BR_STATE_LEARNING || p->state == BR_STATE_FORWARDING;
 }
 
-static inline bool br_vlan_valid_id(u16 vid)
+static inline bool br_vlan_valid_id(u16 vid, struct netlink_ext_ack *extack)
 {
-	return vid > 0 && vid < VLAN_VID_MASK;
+	bool ret = vid > 0 && vid < VLAN_VID_MASK;
+
+	if (!ret)
+		NL_SET_ERR_MSG_MOD(extack, "Vlan id is invalid");
+
+	return ret;
 }
 
 static inline bool br_vlan_valid_range(const struct bridge_vlan_info *cur,
-				       const struct bridge_vlan_info *last)
+				       const struct bridge_vlan_info *last,
+				       struct netlink_ext_ack *extack)
 {
 	/* pvid flag is not allowed in ranges */
-	if (cur->flags & BRIDGE_VLAN_INFO_PVID)
-		return false;
-
-	/* check for required range flags */
-	if (!(cur->flags & (BRIDGE_VLAN_INFO_RANGE_BEGIN |
-			    BRIDGE_VLAN_INFO_RANGE_END)))
+	if (cur->flags & BRIDGE_VLAN_INFO_PVID) {
+		NL_SET_ERR_MSG_MOD(extack, "Pvid isn't allowed in a range");
 		return false;
+	}
 
 	/* when cur is the range end, check if:
 	 *  - it has range start flag
 	 *  - range ids are invalid (end is equal to or before start)
 	 */
 	if (last) {
-		if (cur->flags & BRIDGE_VLAN_INFO_RANGE_BEGIN)
+		if (cur->flags & BRIDGE_VLAN_INFO_RANGE_BEGIN) {
+			NL_SET_ERR_MSG_MOD(extack, "Found a new vlan range start while processing one");
 			return false;
-		else if (cur->vid <= last->vid)
+		} else if (!(cur->flags & BRIDGE_VLAN_INFO_RANGE_END)) {
+			NL_SET_ERR_MSG_MOD(extack, "Vlan range end flag is missing");
 			return false;
+		} else if (cur->vid <= last->vid) {
+			NL_SET_ERR_MSG_MOD(extack, "End vlan id is less than or equal to start vlan id");
+			return false;
+		}
+	}
+
+	/* check for required range flags */
+	if (!(cur->flags & (BRIDGE_VLAN_INFO_RANGE_BEGIN |
+			    BRIDGE_VLAN_INFO_RANGE_END))) {
+		NL_SET_ERR_MSG_MOD(extack, "Both vlan range flags are missing");
+		return false;
 	}
 
 	return true;
-- 
cgit v1.2.3-59-g8ed1b


From 8dcea187088bce5d2f1149294ad109f022653547 Mon Sep 17 00:00:00 2001
From: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Date: Tue, 14 Jan 2020 19:56:09 +0200
Subject: net: bridge: vlan: add rtm definitions and dump support

This patch adds vlan rtm definitions:
 - NEWVLAN: to be used for creating vlans, setting options and
   notifications
 - DELVLAN: to be used for deleting vlans
 - GETVLAN: used for dumping vlan information

Dumping vlans which can span multiple messages is added now with basic
information (vid and flags). We use nlmsg_parse() to validate the header
length in order to be able to extend the message with filtering
attributes later.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/uapi/linux/if_bridge.h |  28 ++++++++
 include/uapi/linux/rtnetlink.h |   7 ++
 net/bridge/br_netlink.c        |   2 +
 net/bridge/br_private.h        |  14 ++++
 net/bridge/br_vlan.c           | 148 +++++++++++++++++++++++++++++++++++++++++
 security/selinux/nlmsgtab.c    |   5 +-
 6 files changed, 203 insertions(+), 1 deletion(-)

diff --git a/include/uapi/linux/if_bridge.h b/include/uapi/linux/if_bridge.h
index 4a58e3d7de46..4da04f77d9ee 100644
--- a/include/uapi/linux/if_bridge.h
+++ b/include/uapi/linux/if_bridge.h
@@ -165,6 +165,34 @@ struct bridge_stp_xstats {
 	__u64 tx_tcn;
 };
 
+/* Bridge vlan RTM header */
+struct br_vlan_msg {
+	__u8 family;
+	__u8 reserved1;
+	__u16 reserved2;
+	__u32 ifindex;
+};
+
+/* Bridge vlan RTM attributes
+ * [BRIDGE_VLANDB_ENTRY] = {
+ *     [BRIDGE_VLANDB_ENTRY_INFO]
+ *     ...
+ * }
+ */
+enum {
+	BRIDGE_VLANDB_UNSPEC,
+	BRIDGE_VLANDB_ENTRY,
+	__BRIDGE_VLANDB_MAX,
+};
+#define BRIDGE_VLANDB_MAX (__BRIDGE_VLANDB_MAX - 1)
+
+enum {
+	BRIDGE_VLANDB_ENTRY_UNSPEC,
+	BRIDGE_VLANDB_ENTRY_INFO,
+	__BRIDGE_VLANDB_ENTRY_MAX,
+};
+#define BRIDGE_VLANDB_ENTRY_MAX (__BRIDGE_VLANDB_ENTRY_MAX - 1)
+
 /* Bridge multicast database attributes
  * [MDBA_MDB] = {
  *     [MDBA_MDB_ENTRY] = {
diff --git a/include/uapi/linux/rtnetlink.h b/include/uapi/linux/rtnetlink.h
index cd43321d20dd..b333cff14071 100644
--- a/include/uapi/linux/rtnetlink.h
+++ b/include/uapi/linux/rtnetlink.h
@@ -171,6 +171,13 @@ enum {
 	RTM_GETLINKPROP,
 #define RTM_GETLINKPROP	RTM_GETLINKPROP
 
+	RTM_NEWVLAN = 112,
+#define RTM_NEWNVLAN	RTM_NEWVLAN
+	RTM_DELVLAN,
+#define RTM_DELVLAN	RTM_DELVLAN
+	RTM_GETVLAN,
+#define RTM_GETVLAN	RTM_GETVLAN
+
 	__RTM_MAX,
 #define RTM_MAX		(((__RTM_MAX + 3) & ~3) - 1)
 };
diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c
index 40942cece51a..75a7ecf95d7f 100644
--- a/net/bridge/br_netlink.c
+++ b/net/bridge/br_netlink.c
@@ -1657,6 +1657,7 @@ int __init br_netlink_init(void)
 	int err;
 
 	br_mdb_init();
+	br_vlan_rtnl_init();
 	rtnl_af_register(&br_af_ops);
 
 	err = rtnl_link_register(&br_link_ops);
@@ -1674,6 +1675,7 @@ out_af:
 void br_netlink_fini(void)
 {
 	br_mdb_uninit();
+	br_vlan_rtnl_uninit();
 	rtnl_af_unregister(&br_af_ops);
 	rtnl_link_unregister(&br_link_ops);
 }
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index a7dddc5d7790..1c00411ae938 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -958,6 +958,8 @@ void br_vlan_get_stats(const struct net_bridge_vlan *v,
 void br_vlan_port_event(struct net_bridge_port *p, unsigned long event);
 int br_vlan_bridge_event(struct net_device *dev, unsigned long event,
 			 void *ptr);
+void br_vlan_rtnl_init(void);
+void br_vlan_rtnl_uninit(void);
 
 static inline struct net_bridge_vlan_group *br_vlan_group(
 					const struct net_bridge *br)
@@ -1009,6 +1011,10 @@ static inline u16 br_get_pvid(const struct net_bridge_vlan_group *vg)
 	return vg->pvid;
 }
 
+static inline u16 br_vlan_flags(const struct net_bridge_vlan *v, u16 pvid)
+{
+	return v->vid == pvid ? v->flags | BRIDGE_VLAN_INFO_PVID : v->flags;
+}
 #else
 static inline bool br_allowed_ingress(const struct net_bridge *br,
 				      struct net_bridge_vlan_group *vg,
@@ -1152,6 +1158,14 @@ static inline int br_vlan_bridge_event(struct net_device *dev,
 {
 	return 0;
 }
+
+static inline void br_vlan_rtnl_init(void)
+{
+}
+
+static inline void br_vlan_rtnl_uninit(void)
+{
+}
 #endif
 
 struct nf_br_ops {
diff --git a/net/bridge/br_vlan.c b/net/bridge/br_vlan.c
index bb98984cd27d..5f2ac4f244f5 100644
--- a/net/bridge/br_vlan.c
+++ b/net/bridge/br_vlan.c
@@ -1505,3 +1505,151 @@ void br_vlan_port_event(struct net_bridge_port *p, unsigned long event)
 		break;
 	}
 }
+
+static bool br_vlan_fill_vids(struct sk_buff *skb, u16 vid, u16 flags)
+{
+	struct bridge_vlan_info info;
+	struct nlattr *nest;
+
+	nest = nla_nest_start(skb, BRIDGE_VLANDB_ENTRY);
+	if (!nest)
+		return false;
+
+	memset(&info, 0, sizeof(info));
+	info.vid = vid;
+	if (flags & BRIDGE_VLAN_INFO_UNTAGGED)
+		info.flags |= BRIDGE_VLAN_INFO_UNTAGGED;
+	if (flags & BRIDGE_VLAN_INFO_PVID)
+		info.flags |= BRIDGE_VLAN_INFO_PVID;
+
+	if (nla_put(skb, BRIDGE_VLANDB_ENTRY_INFO, sizeof(info), &info))
+		goto out_err;
+
+	nla_nest_end(skb, nest);
+
+	return true;
+
+out_err:
+	nla_nest_cancel(skb, nest);
+	return false;
+}
+
+static int br_vlan_dump_dev(const struct net_device *dev,
+			    struct sk_buff *skb,
+			    struct netlink_callback *cb)
+{
+	struct net_bridge_vlan_group *vg;
+	int idx = 0, s_idx = cb->args[1];
+	struct nlmsghdr *nlh = NULL;
+	struct net_bridge_vlan *v;
+	struct net_bridge_port *p;
+	struct br_vlan_msg *bvm;
+	struct net_bridge *br;
+	int err = 0;
+	u16 pvid;
+
+	if (!netif_is_bridge_master(dev) && !netif_is_bridge_port(dev))
+		return -EINVAL;
+
+	if (netif_is_bridge_master(dev)) {
+		br = netdev_priv(dev);
+		vg = br_vlan_group_rcu(br);
+		p = NULL;
+	} else {
+		p = br_port_get_rcu(dev);
+		if (WARN_ON(!p))
+			return -EINVAL;
+		vg = nbp_vlan_group_rcu(p);
+		br = p->br;
+	}
+
+	if (!vg)
+		return 0;
+
+	nlh = nlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq,
+			RTM_NEWVLAN, sizeof(*bvm), NLM_F_MULTI);
+	if (!nlh)
+		return -EMSGSIZE;
+	bvm = nlmsg_data(nlh);
+	memset(bvm, 0, sizeof(*bvm));
+	bvm->family = PF_BRIDGE;
+	bvm->ifindex = dev->ifindex;
+	pvid = br_get_pvid(vg);
+
+	list_for_each_entry_rcu(v, &vg->vlan_list, vlist) {
+		if (!br_vlan_should_use(v))
+			continue;
+		if (idx < s_idx)
+			goto skip;
+		if (!br_vlan_fill_vids(skb, v->vid, br_vlan_flags(v, pvid))) {
+			err = -EMSGSIZE;
+			break;
+		}
+skip:
+		idx++;
+	}
+	if (err)
+		cb->args[1] = idx;
+	else
+		cb->args[1] = 0;
+	nlmsg_end(skb, nlh);
+
+	return err;
+}
+
+static int br_vlan_rtm_dump(struct sk_buff *skb, struct netlink_callback *cb)
+{
+	int idx = 0, err = 0, s_idx = cb->args[0];
+	struct net *net = sock_net(skb->sk);
+	struct br_vlan_msg *bvm;
+	struct net_device *dev;
+
+	err = nlmsg_parse(cb->nlh, sizeof(*bvm), NULL, 0, NULL, cb->extack);
+	if (err < 0)
+		return err;
+
+	bvm = nlmsg_data(cb->nlh);
+
+	rcu_read_lock();
+	if (bvm->ifindex) {
+		dev = dev_get_by_index_rcu(net, bvm->ifindex);
+		if (!dev) {
+			err = -ENODEV;
+			goto out_err;
+		}
+		err = br_vlan_dump_dev(dev, skb, cb);
+		if (err && err != -EMSGSIZE)
+			goto out_err;
+	} else {
+		for_each_netdev_rcu(net, dev) {
+			if (idx < s_idx)
+				goto skip;
+
+			err = br_vlan_dump_dev(dev, skb, cb);
+			if (err == -EMSGSIZE)
+				break;
+skip:
+			idx++;
+		}
+	}
+	cb->args[0] = idx;
+	rcu_read_unlock();
+
+	return skb->len;
+
+out_err:
+	rcu_read_unlock();
+
+	return err;
+}
+
+void br_vlan_rtnl_init(void)
+{
+	rtnl_register_module(THIS_MODULE, PF_BRIDGE, RTM_GETVLAN, NULL,
+			     br_vlan_rtm_dump, 0);
+}
+
+void br_vlan_rtnl_uninit(void)
+{
+	rtnl_unregister(PF_BRIDGE, RTM_GETVLAN);
+}
diff --git a/security/selinux/nlmsgtab.c b/security/selinux/nlmsgtab.c
index c97fdae8f71b..b69231918686 100644
--- a/security/selinux/nlmsgtab.c
+++ b/security/selinux/nlmsgtab.c
@@ -85,6 +85,9 @@ static const struct nlmsg_perm nlmsg_route_perms[] =
 	{ RTM_GETNEXTHOP,	NETLINK_ROUTE_SOCKET__NLMSG_READ  },
 	{ RTM_NEWLINKPROP,	NETLINK_ROUTE_SOCKET__NLMSG_WRITE },
 	{ RTM_DELLINKPROP,	NETLINK_ROUTE_SOCKET__NLMSG_WRITE },
+	{ RTM_NEWVLAN,		NETLINK_ROUTE_SOCKET__NLMSG_WRITE },
+	{ RTM_DELVLAN,		NETLINK_ROUTE_SOCKET__NLMSG_WRITE },
+	{ RTM_GETVLAN,		NETLINK_ROUTE_SOCKET__NLMSG_READ  },
 };
 
 static const struct nlmsg_perm nlmsg_tcpdiag_perms[] =
@@ -168,7 +171,7 @@ int selinux_nlmsg_lookup(u16 sclass, u16 nlmsg_type, u32 *perm)
 		 * structures at the top of this file with the new mappings
 		 * before updating the BUILD_BUG_ON() macro!
 		 */
-		BUILD_BUG_ON(RTM_MAX != (RTM_NEWLINKPROP + 3));
+		BUILD_BUG_ON(RTM_MAX != (RTM_NEWVLAN + 3));
 		err = nlmsg_perm(nlmsg_type, perm, nlmsg_route_perms,
 				 sizeof(nlmsg_route_perms));
 		break;
-- 
cgit v1.2.3-59-g8ed1b


From f26b296585dca043d78f13ef67cb17fb200d5e57 Mon Sep 17 00:00:00 2001
From: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Date: Tue, 14 Jan 2020 19:56:10 +0200
Subject: net: bridge: vlan: add new rtm message support

Add initial RTM_NEWVLAN support which can only create vlans, operating
similar to the current br_afspec(). We will use it later to also change
per-vlan options. Old-style (flag-based) vlan ranges are not allowed
when using RTM messages, we will introduce vlan ranges later via a new
nested attribute which would allow us to have all the information about a
range encapsulated into a single nl attribute.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/bridge/br_netlink.c |  12 +++---
 net/bridge/br_private.h |   6 +++
 net/bridge/br_vlan.c    | 111 ++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 123 insertions(+), 6 deletions(-)

diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c
index 75a7ecf95d7f..b3da4f46dc64 100644
--- a/net/bridge/br_netlink.c
+++ b/net/bridge/br_netlink.c
@@ -561,12 +561,12 @@ static int br_vlan_info(struct net_bridge *br, struct net_bridge_port *p,
 	return err;
 }
 
-static int br_process_vlan_info(struct net_bridge *br,
-				struct net_bridge_port *p, int cmd,
-				struct bridge_vlan_info *vinfo_curr,
-				struct bridge_vlan_info **vinfo_last,
-				bool *changed,
-				struct netlink_ext_ack *extack)
+int br_process_vlan_info(struct net_bridge *br,
+			 struct net_bridge_port *p, int cmd,
+			 struct bridge_vlan_info *vinfo_curr,
+			 struct bridge_vlan_info **vinfo_last,
+			 bool *changed,
+			 struct netlink_ext_ack *extack)
 {
 	if (!br_vlan_valid_id(vinfo_curr->vid, extack))
 		return -EINVAL;
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index 1c00411ae938..ee3871dea68f 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -1237,6 +1237,12 @@ int br_setlink(struct net_device *dev, struct nlmsghdr *nlmsg, u16 flags,
 int br_dellink(struct net_device *dev, struct nlmsghdr *nlmsg, u16 flags);
 int br_getlink(struct sk_buff *skb, u32 pid, u32 seq, struct net_device *dev,
 	       u32 filter_mask, int nlflags);
+int br_process_vlan_info(struct net_bridge *br,
+			 struct net_bridge_port *p, int cmd,
+			 struct bridge_vlan_info *vinfo_curr,
+			 struct bridge_vlan_info **vinfo_last,
+			 bool *changed,
+			 struct netlink_ext_ack *extack);
 
 #ifdef CONFIG_SYSFS
 /* br_sysfs_if.c */
diff --git a/net/bridge/br_vlan.c b/net/bridge/br_vlan.c
index 5f2ac4f244f5..6da0210b01eb 100644
--- a/net/bridge/br_vlan.c
+++ b/net/bridge/br_vlan.c
@@ -1643,13 +1643,124 @@ out_err:
 	return err;
 }
 
+static const struct nla_policy br_vlan_db_policy[BRIDGE_VLANDB_ENTRY_MAX + 1] = {
+	[BRIDGE_VLANDB_ENTRY_INFO]	= { .type = NLA_EXACT_LEN,
+					    .len = sizeof(struct bridge_vlan_info) },
+};
+
+static int br_vlan_rtm_process_one(struct net_device *dev,
+				   const struct nlattr *attr,
+				   int cmd, struct netlink_ext_ack *extack)
+{
+	struct bridge_vlan_info *vinfo, *vinfo_last = NULL;
+	struct nlattr *tb[BRIDGE_VLANDB_ENTRY_MAX + 1];
+	struct net_bridge_vlan_group *vg;
+	struct net_bridge_port *p = NULL;
+	int err = 0, cmdmap = 0;
+	struct net_bridge *br;
+	bool changed = false;
+
+	if (netif_is_bridge_master(dev)) {
+		br = netdev_priv(dev);
+		vg = br_vlan_group(br);
+	} else {
+		p = br_port_get_rtnl(dev);
+		if (WARN_ON(!p))
+			return -ENODEV;
+		br = p->br;
+		vg = nbp_vlan_group(p);
+	}
+
+	if (WARN_ON(!vg))
+		return -ENODEV;
+
+	err = nla_parse_nested(tb, BRIDGE_VLANDB_ENTRY_MAX, attr,
+			       br_vlan_db_policy, extack);
+	if (err)
+		return err;
+
+	if (!tb[BRIDGE_VLANDB_ENTRY_INFO]) {
+		NL_SET_ERR_MSG_MOD(extack, "Missing vlan entry info");
+		return -EINVAL;
+	}
+
+	vinfo = nla_data(tb[BRIDGE_VLANDB_ENTRY_INFO]);
+	if (vinfo->flags & (BRIDGE_VLAN_INFO_RANGE_BEGIN |
+			    BRIDGE_VLAN_INFO_RANGE_END)) {
+		NL_SET_ERR_MSG_MOD(extack, "Old-style vlan ranges are not allowed when using RTM vlan calls");
+		return -EINVAL;
+	}
+	if (!br_vlan_valid_id(vinfo->vid, extack))
+		return -EINVAL;
+
+	switch (cmd) {
+	case RTM_NEWVLAN:
+		cmdmap = RTM_SETLINK;
+		break;
+	}
+
+	err = br_process_vlan_info(br, p, cmdmap, vinfo, &vinfo_last, &changed,
+				   extack);
+	if (changed)
+		br_ifinfo_notify(cmdmap, br, p);
+
+	return err;
+}
+
+static int br_vlan_rtm_process(struct sk_buff *skb, struct nlmsghdr *nlh,
+			       struct netlink_ext_ack *extack)
+{
+	struct net *net = sock_net(skb->sk);
+	struct br_vlan_msg *bvm;
+	struct net_device *dev;
+	struct nlattr *attr;
+	int err, vlans = 0;
+	int rem;
+
+	/* this should validate the header and check for remaining bytes */
+	err = nlmsg_parse(nlh, sizeof(*bvm), NULL, BRIDGE_VLANDB_MAX, NULL,
+			  extack);
+	if (err < 0)
+		return err;
+
+	bvm = nlmsg_data(nlh);
+	dev = __dev_get_by_index(net, bvm->ifindex);
+	if (!dev)
+		return -ENODEV;
+
+	if (!netif_is_bridge_master(dev) && !netif_is_bridge_port(dev)) {
+		NL_SET_ERR_MSG_MOD(extack, "The device is not a valid bridge or bridge port");
+		return -EINVAL;
+	}
+
+	nlmsg_for_each_attr(attr, nlh, sizeof(*bvm), rem) {
+		if (nla_type(attr) != BRIDGE_VLANDB_ENTRY)
+			continue;
+
+		vlans++;
+		err = br_vlan_rtm_process_one(dev, attr, nlh->nlmsg_type,
+					      extack);
+		if (err)
+			break;
+	}
+	if (!vlans) {
+		NL_SET_ERR_MSG_MOD(extack, "No vlans found to process");
+		err = -EINVAL;
+	}
+
+	return err;
+}
+
 void br_vlan_rtnl_init(void)
 {
 	rtnl_register_module(THIS_MODULE, PF_BRIDGE, RTM_GETVLAN, NULL,
 			     br_vlan_rtm_dump, 0);
+	rtnl_register_module(THIS_MODULE, PF_BRIDGE, RTM_NEWVLAN,
+			     br_vlan_rtm_process, NULL, 0);
 }
 
 void br_vlan_rtnl_uninit(void)
 {
 	rtnl_unregister(PF_BRIDGE, RTM_GETVLAN);
+	rtnl_unregister(PF_BRIDGE, RTM_NEWVLAN);
 }
-- 
cgit v1.2.3-59-g8ed1b


From adb3ce9bcb0f565b21e0ce0c75223c6854670b80 Mon Sep 17 00:00:00 2001
From: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Date: Tue, 14 Jan 2020 19:56:11 +0200
Subject: net: bridge: vlan: add del rtm message support

Adding RTM_DELVLAN support similar to RTM_NEWVLAN is simple, just need to
map DELVLAN to DELLINK and register the handler.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/bridge/br_vlan.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/net/bridge/br_vlan.c b/net/bridge/br_vlan.c
index 6da0210b01eb..89d5fa75c575 100644
--- a/net/bridge/br_vlan.c
+++ b/net/bridge/br_vlan.c
@@ -1697,6 +1697,9 @@ static int br_vlan_rtm_process_one(struct net_device *dev,
 	case RTM_NEWVLAN:
 		cmdmap = RTM_SETLINK;
 		break;
+	case RTM_DELVLAN:
+		cmdmap = RTM_DELLINK;
+		break;
 	}
 
 	err = br_process_vlan_info(br, p, cmdmap, vinfo, &vinfo_last, &changed,
@@ -1757,10 +1760,13 @@ void br_vlan_rtnl_init(void)
 			     br_vlan_rtm_dump, 0);
 	rtnl_register_module(THIS_MODULE, PF_BRIDGE, RTM_NEWVLAN,
 			     br_vlan_rtm_process, NULL, 0);
+	rtnl_register_module(THIS_MODULE, PF_BRIDGE, RTM_DELVLAN,
+			     br_vlan_rtm_process, NULL, 0);
 }
 
 void br_vlan_rtnl_uninit(void)
 {
 	rtnl_unregister(PF_BRIDGE, RTM_GETVLAN);
 	rtnl_unregister(PF_BRIDGE, RTM_NEWVLAN);
+	rtnl_unregister(PF_BRIDGE, RTM_DELVLAN);
 }
-- 
cgit v1.2.3-59-g8ed1b


From 0ab558795184f87b532363e262357160d25f5d64 Mon Sep 17 00:00:00 2001
From: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Date: Tue, 14 Jan 2020 19:56:12 +0200
Subject: net: bridge: vlan: add rtm range support

Add a new vlandb nl attribute - BRIDGE_VLANDB_ENTRY_RANGE which causes
RTM_NEWVLAN/DELVAN to act on a range. Dumps now automatically compress
similar vlans into ranges. This will be also used when per-vlan options
are introduced and vlans' options match, they will be put into a single
range which is encapsulated in one netlink attribute. We need to run
similar checks as br_process_vlan_info() does because these ranges will
be used for options setting and they'll be able to skip
br_process_vlan_info().

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/uapi/linux/if_bridge.h |  1 +
 net/bridge/br_vlan.c           | 86 +++++++++++++++++++++++++++++++++++-------
 2 files changed, 73 insertions(+), 14 deletions(-)

diff --git a/include/uapi/linux/if_bridge.h b/include/uapi/linux/if_bridge.h
index 4da04f77d9ee..ac38f0b674b8 100644
--- a/include/uapi/linux/if_bridge.h
+++ b/include/uapi/linux/if_bridge.h
@@ -189,6 +189,7 @@ enum {
 enum {
 	BRIDGE_VLANDB_ENTRY_UNSPEC,
 	BRIDGE_VLANDB_ENTRY_INFO,
+	BRIDGE_VLANDB_ENTRY_RANGE,
 	__BRIDGE_VLANDB_ENTRY_MAX,
 };
 #define BRIDGE_VLANDB_ENTRY_MAX (__BRIDGE_VLANDB_ENTRY_MAX - 1)
diff --git a/net/bridge/br_vlan.c b/net/bridge/br_vlan.c
index 89d5fa75c575..9d64a86f2cbd 100644
--- a/net/bridge/br_vlan.c
+++ b/net/bridge/br_vlan.c
@@ -1506,7 +1506,8 @@ void br_vlan_port_event(struct net_bridge_port *p, unsigned long event)
 	}
 }
 
-static bool br_vlan_fill_vids(struct sk_buff *skb, u16 vid, u16 flags)
+static bool br_vlan_fill_vids(struct sk_buff *skb, u16 vid, u16 vid_range,
+			      u16 flags)
 {
 	struct bridge_vlan_info info;
 	struct nlattr *nest;
@@ -1525,6 +1526,11 @@ static bool br_vlan_fill_vids(struct sk_buff *skb, u16 vid, u16 flags)
 	if (nla_put(skb, BRIDGE_VLANDB_ENTRY_INFO, sizeof(info), &info))
 		goto out_err;
 
+	if (vid_range && vid < vid_range &&
+	    !(flags & BRIDGE_VLAN_INFO_PVID) &&
+	    nla_put_u16(skb, BRIDGE_VLANDB_ENTRY_RANGE, vid_range))
+		goto out_err;
+
 	nla_nest_end(skb, nest);
 
 	return true;
@@ -1534,14 +1540,22 @@ out_err:
 	return false;
 }
 
+/* check if v_curr can enter a range ending in range_end */
+static bool br_vlan_can_enter_range(const struct net_bridge_vlan *v_curr,
+				    const struct net_bridge_vlan *range_end)
+{
+	return v_curr->vid - range_end->vid == 1 &&
+	       range_end->flags == v_curr->flags;
+}
+
 static int br_vlan_dump_dev(const struct net_device *dev,
 			    struct sk_buff *skb,
 			    struct netlink_callback *cb)
 {
+	struct net_bridge_vlan *v, *range_start = NULL, *range_end = NULL;
 	struct net_bridge_vlan_group *vg;
 	int idx = 0, s_idx = cb->args[1];
 	struct nlmsghdr *nlh = NULL;
-	struct net_bridge_vlan *v;
 	struct net_bridge_port *p;
 	struct br_vlan_msg *bvm;
 	struct net_bridge *br;
@@ -1576,22 +1590,49 @@ static int br_vlan_dump_dev(const struct net_device *dev,
 	bvm->ifindex = dev->ifindex;
 	pvid = br_get_pvid(vg);
 
+	/* idx must stay at range's beginning until it is filled in */
 	list_for_each_entry_rcu(v, &vg->vlan_list, vlist) {
 		if (!br_vlan_should_use(v))
 			continue;
-		if (idx < s_idx)
-			goto skip;
-		if (!br_vlan_fill_vids(skb, v->vid, br_vlan_flags(v, pvid))) {
-			err = -EMSGSIZE;
-			break;
+		if (idx < s_idx) {
+			idx++;
+			continue;
 		}
-skip:
-		idx++;
+
+		if (!range_start) {
+			range_start = v;
+			range_end = v;
+			continue;
+		}
+
+		if (v->vid == pvid || !br_vlan_can_enter_range(v, range_end)) {
+			u16 flags = br_vlan_flags(range_start, pvid);
+
+			if (!br_vlan_fill_vids(skb, range_start->vid,
+					       range_end->vid, flags)) {
+				err = -EMSGSIZE;
+				break;
+			}
+			/* advance number of filled vlans */
+			idx += range_end->vid - range_start->vid + 1;
+
+			range_start = v;
+		}
+		range_end = v;
 	}
-	if (err)
-		cb->args[1] = idx;
-	else
-		cb->args[1] = 0;
+
+	/* err will be 0 and range_start will be set in 3 cases here:
+	 * - first vlan (range_start == range_end)
+	 * - last vlan (range_start == range_end, not in range)
+	 * - last vlan range (range_start != range_end, in range)
+	 */
+	if (!err && range_start &&
+	    !br_vlan_fill_vids(skb, range_start->vid, range_end->vid,
+			       br_vlan_flags(range_start, pvid)))
+		err = -EMSGSIZE;
+
+	cb->args[1] = err ? idx : 0;
+
 	nlmsg_end(skb, nlh);
 
 	return err;
@@ -1646,13 +1687,14 @@ out_err:
 static const struct nla_policy br_vlan_db_policy[BRIDGE_VLANDB_ENTRY_MAX + 1] = {
 	[BRIDGE_VLANDB_ENTRY_INFO]	= { .type = NLA_EXACT_LEN,
 					    .len = sizeof(struct bridge_vlan_info) },
+	[BRIDGE_VLANDB_ENTRY_RANGE]	= { .type = NLA_U16 },
 };
 
 static int br_vlan_rtm_process_one(struct net_device *dev,
 				   const struct nlattr *attr,
 				   int cmd, struct netlink_ext_ack *extack)
 {
-	struct bridge_vlan_info *vinfo, *vinfo_last = NULL;
+	struct bridge_vlan_info *vinfo, vrange_end, *vinfo_last = NULL;
 	struct nlattr *tb[BRIDGE_VLANDB_ENTRY_MAX + 1];
 	struct net_bridge_vlan_group *vg;
 	struct net_bridge_port *p = NULL;
@@ -1683,6 +1725,7 @@ static int br_vlan_rtm_process_one(struct net_device *dev,
 		NL_SET_ERR_MSG_MOD(extack, "Missing vlan entry info");
 		return -EINVAL;
 	}
+	memset(&vrange_end, 0, sizeof(vrange_end));
 
 	vinfo = nla_data(tb[BRIDGE_VLANDB_ENTRY_INFO]);
 	if (vinfo->flags & (BRIDGE_VLAN_INFO_RANGE_BEGIN |
@@ -1693,6 +1736,21 @@ static int br_vlan_rtm_process_one(struct net_device *dev,
 	if (!br_vlan_valid_id(vinfo->vid, extack))
 		return -EINVAL;
 
+	if (tb[BRIDGE_VLANDB_ENTRY_RANGE]) {
+		vrange_end.vid = nla_get_u16(tb[BRIDGE_VLANDB_ENTRY_RANGE]);
+		/* validate user-provided flags without RANGE_BEGIN */
+		vrange_end.flags = BRIDGE_VLAN_INFO_RANGE_END | vinfo->flags;
+		vinfo->flags |= BRIDGE_VLAN_INFO_RANGE_BEGIN;
+
+		/* vinfo_last is the range start, vinfo the range end */
+		vinfo_last = vinfo;
+		vinfo = &vrange_end;
+
+		if (!br_vlan_valid_id(vinfo->vid, extack) ||
+		    !br_vlan_valid_range(vinfo, vinfo_last, extack))
+			return -EINVAL;
+	}
+
 	switch (cmd) {
 	case RTM_NEWVLAN:
 		cmdmap = RTM_SETLINK;
-- 
cgit v1.2.3-59-g8ed1b


From cf5bddb95cbe5340e486e84d46cf2f0bb324078c Mon Sep 17 00:00:00 2001
From: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Date: Tue, 14 Jan 2020 19:56:13 +0200
Subject: net: bridge: vlan: add rtnetlink group and notify support

Add a new rtnetlink group for bridge vlan notifications - RTNLGRP_BRVLAN
and add support for sending vlan notifications (both single and ranges).
No functional changes intended, the notification support will be used by
later patches.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/uapi/linux/rtnetlink.h |  2 ++
 net/bridge/br_private.h        | 11 ++++++
 net/bridge/br_vlan.c           | 79 ++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 92 insertions(+)

diff --git a/include/uapi/linux/rtnetlink.h b/include/uapi/linux/rtnetlink.h
index b333cff14071..4a8c5b745157 100644
--- a/include/uapi/linux/rtnetlink.h
+++ b/include/uapi/linux/rtnetlink.h
@@ -730,6 +730,8 @@ enum rtnetlink_groups {
 #define RTNLGRP_IPV6_MROUTE_R	RTNLGRP_IPV6_MROUTE_R
 	RTNLGRP_NEXTHOP,
 #define RTNLGRP_NEXTHOP		RTNLGRP_NEXTHOP
+	RTNLGRP_BRVLAN,
+#define RTNLGRP_BRVLAN		RTNLGRP_BRVLAN
 	__RTNLGRP_MAX
 };
 #define RTNLGRP_MAX	(__RTNLGRP_MAX - 1)
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index ee3871dea68f..ba162c8197da 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -960,6 +960,10 @@ int br_vlan_bridge_event(struct net_device *dev, unsigned long event,
 			 void *ptr);
 void br_vlan_rtnl_init(void);
 void br_vlan_rtnl_uninit(void);
+void br_vlan_notify(const struct net_bridge *br,
+		    const struct net_bridge_port *p,
+		    u16 vid, u16 vid_range,
+		    int cmd);
 
 static inline struct net_bridge_vlan_group *br_vlan_group(
 					const struct net_bridge *br)
@@ -1166,6 +1170,13 @@ static inline void br_vlan_rtnl_init(void)
 static inline void br_vlan_rtnl_uninit(void)
 {
 }
+
+static inline void br_vlan_notify(const struct net_bridge *br,
+				  const struct net_bridge_port *p,
+				  u16 vid, u16 vid_range,
+				  int cmd)
+{
+}
 #endif
 
 struct nf_br_ops {
diff --git a/net/bridge/br_vlan.c b/net/bridge/br_vlan.c
index 9d64a86f2cbd..5d52a2604547 100644
--- a/net/bridge/br_vlan.c
+++ b/net/bridge/br_vlan.c
@@ -1540,6 +1540,85 @@ out_err:
 	return false;
 }
 
+static size_t rtnl_vlan_nlmsg_size(void)
+{
+	return NLMSG_ALIGN(sizeof(struct br_vlan_msg))
+		+ nla_total_size(0) /* BRIDGE_VLANDB_ENTRY */
+		+ nla_total_size(sizeof(u16)) /* BRIDGE_VLANDB_ENTRY_RANGE */
+		+ nla_total_size(sizeof(struct bridge_vlan_info)); /* BRIDGE_VLANDB_ENTRY_INFO */
+}
+
+void br_vlan_notify(const struct net_bridge *br,
+		    const struct net_bridge_port *p,
+		    u16 vid, u16 vid_range,
+		    int cmd)
+{
+	struct net_bridge_vlan_group *vg;
+	struct net_bridge_vlan *v;
+	struct br_vlan_msg *bvm;
+	struct nlmsghdr *nlh;
+	struct sk_buff *skb;
+	int err = -ENOBUFS;
+	struct net *net;
+	u16 flags = 0;
+	int ifindex;
+
+	/* right now notifications are done only with rtnl held */
+	ASSERT_RTNL();
+
+	if (p) {
+		ifindex = p->dev->ifindex;
+		vg = nbp_vlan_group(p);
+		net = dev_net(p->dev);
+	} else {
+		ifindex = br->dev->ifindex;
+		vg = br_vlan_group(br);
+		net = dev_net(br->dev);
+	}
+
+	skb = nlmsg_new(rtnl_vlan_nlmsg_size(), GFP_KERNEL);
+	if (!skb)
+		goto out_err;
+
+	err = -EMSGSIZE;
+	nlh = nlmsg_put(skb, 0, 0, cmd, sizeof(*bvm), 0);
+	if (!nlh)
+		goto out_err;
+	bvm = nlmsg_data(nlh);
+	memset(bvm, 0, sizeof(*bvm));
+	bvm->family = AF_BRIDGE;
+	bvm->ifindex = ifindex;
+
+	switch (cmd) {
+	case RTM_NEWVLAN:
+		/* need to find the vlan due to flags/options */
+		v = br_vlan_find(vg, vid);
+		if (!v || !br_vlan_should_use(v))
+			goto out_kfree;
+
+		flags = v->flags;
+		if (br_get_pvid(vg) == v->vid)
+			flags |= BRIDGE_VLAN_INFO_PVID;
+		break;
+	case RTM_DELVLAN:
+		break;
+	default:
+		goto out_kfree;
+	}
+
+	if (!br_vlan_fill_vids(skb, vid, vid_range, flags))
+		goto out_err;
+
+	nlmsg_end(skb, nlh);
+	rtnl_notify(skb, net, 0, RTNLGRP_BRVLAN, NULL, GFP_KERNEL);
+	return;
+
+out_err:
+	rtnl_set_sk_err(net, RTNLGRP_BRVLAN, err);
+out_kfree:
+	kfree_skb(skb);
+}
+
 /* check if v_curr can enter a range ending in range_end */
 static bool br_vlan_can_enter_range(const struct net_bridge_vlan *v_curr,
 				    const struct net_bridge_vlan *range_end)
-- 
cgit v1.2.3-59-g8ed1b


From f545923b4a6bd7abedac50ea3fce76c713be1b74 Mon Sep 17 00:00:00 2001
From: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Date: Tue, 14 Jan 2020 19:56:14 +0200
Subject: net: bridge: vlan: notify on vlan add/delete/change flags

Now that we can notify, send a notification on add/del or change of flags.
Notifications are also compressed when possible to reduce their number
and relieve user-space of extra processing, due to that we have to
manually notify after each add/del in order to avoid double
notifications. We try hard to notify only about the vlans which actually
changed, thus a single command can result in multiple notifications
about disjoint ranges if there were vlans which didn't change inside.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/bridge/br_netlink.c | 34 ++++++++++++++++++++---
 net/bridge/br_private.h | 12 +++++++++
 net/bridge/br_vlan.c    | 71 ++++++++++++++++++++++++++++++++++++++-----------
 3 files changed, 99 insertions(+), 18 deletions(-)

diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c
index b3da4f46dc64..43dab4066f91 100644
--- a/net/bridge/br_netlink.c
+++ b/net/bridge/br_netlink.c
@@ -568,9 +568,14 @@ int br_process_vlan_info(struct net_bridge *br,
 			 bool *changed,
 			 struct netlink_ext_ack *extack)
 {
+	int err, rtm_cmd;
+
 	if (!br_vlan_valid_id(vinfo_curr->vid, extack))
 		return -EINVAL;
 
+	/* needed for vlan-only NEWVLAN/DELVLAN notifications */
+	rtm_cmd = br_afspec_cmd_to_rtm(cmd);
+
 	if (vinfo_curr->flags & BRIDGE_VLAN_INFO_RANGE_BEGIN) {
 		if (!br_vlan_valid_range(vinfo_curr, *vinfo_last, extack))
 			return -EINVAL;
@@ -580,7 +585,7 @@ int br_process_vlan_info(struct net_bridge *br,
 
 	if (*vinfo_last) {
 		struct bridge_vlan_info tmp_vinfo;
-		int v, err;
+		int v, v_change_start = 0;
 
 		if (!br_vlan_valid_range(vinfo_curr, *vinfo_last, extack))
 			return -EINVAL;
@@ -588,18 +593,41 @@ int br_process_vlan_info(struct net_bridge *br,
 		memcpy(&tmp_vinfo, *vinfo_last,
 		       sizeof(struct bridge_vlan_info));
 		for (v = (*vinfo_last)->vid; v <= vinfo_curr->vid; v++) {
+			bool curr_change = false;
+
 			tmp_vinfo.vid = v;
-			err = br_vlan_info(br, p, cmd, &tmp_vinfo, changed,
+			err = br_vlan_info(br, p, cmd, &tmp_vinfo, &curr_change,
 					   extack);
 			if (err)
 				break;
+			if (curr_change) {
+				*changed = curr_change;
+				if (!v_change_start)
+					v_change_start = v;
+			} else {
+				/* nothing to notify yet */
+				if (!v_change_start)
+					continue;
+				br_vlan_notify(br, p, v_change_start,
+					       v - 1, rtm_cmd);
+				v_change_start = 0;
+			}
 		}
+		/* v_change_start is set only if the last/whole range changed */
+		if (v_change_start)
+			br_vlan_notify(br, p, v_change_start,
+				       v - 1, rtm_cmd);
+
 		*vinfo_last = NULL;
 
 		return err;
 	}
 
-	return br_vlan_info(br, p, cmd, vinfo_curr, changed, extack);
+	err = br_vlan_info(br, p, cmd, vinfo_curr, changed, extack);
+	if (*changed)
+		br_vlan_notify(br, p, vinfo_curr->vid, 0, rtm_cmd);
+
+	return err;
 }
 
 static int br_afspec(struct net_bridge *br,
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index ba162c8197da..a6226ff2f0cc 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -554,6 +554,18 @@ static inline bool br_vlan_valid_range(const struct bridge_vlan_info *cur,
 	return true;
 }
 
+static inline int br_afspec_cmd_to_rtm(int cmd)
+{
+	switch (cmd) {
+	case RTM_SETLINK:
+		return RTM_NEWVLAN;
+	case RTM_DELLINK:
+		return RTM_DELVLAN;
+	}
+
+	return 0;
+}
+
 static inline int br_opt_get(const struct net_bridge *br,
 			     enum net_bridge_opts opt)
 {
diff --git a/net/bridge/br_vlan.c b/net/bridge/br_vlan.c
index 5d52a2604547..e4f7dd10c3f8 100644
--- a/net/bridge/br_vlan.c
+++ b/net/bridge/br_vlan.c
@@ -257,6 +257,10 @@ static int __vlan_add(struct net_bridge_vlan *v, u16 flags,
 					  &changed, extack);
 			if (err)
 				goto out_filt;
+
+			if (changed)
+				br_vlan_notify(br, NULL, v->vid, 0,
+					       RTM_NEWVLAN);
 		}
 
 		masterv = br_vlan_get_master(br, v->vid, extack);
@@ -380,13 +384,31 @@ static void __vlan_group_free(struct net_bridge_vlan_group *vg)
 	kfree(vg);
 }
 
-static void __vlan_flush(struct net_bridge_vlan_group *vg)
+static void __vlan_flush(const struct net_bridge *br,
+			 const struct net_bridge_port *p,
+			 struct net_bridge_vlan_group *vg)
 {
 	struct net_bridge_vlan *vlan, *tmp;
+	u16 v_start = 0, v_end = 0;
 
 	__vlan_delete_pvid(vg, vg->pvid);
-	list_for_each_entry_safe(vlan, tmp, &vg->vlan_list, vlist)
+	list_for_each_entry_safe(vlan, tmp, &vg->vlan_list, vlist) {
+		/* take care of disjoint ranges */
+		if (!v_start) {
+			v_start = vlan->vid;
+		} else if (vlan->vid - v_end != 1) {
+			/* found range end, notify and start next one */
+			br_vlan_notify(br, p, v_start, v_end, RTM_DELVLAN);
+			v_start = vlan->vid;
+		}
+		v_end = vlan->vid;
+
 		__vlan_del(vlan);
+	}
+
+	/* notify about the last/whole vlan range */
+	if (v_start)
+		br_vlan_notify(br, p, v_start, v_end, RTM_DELVLAN);
 }
 
 struct sk_buff *br_handle_vlan(struct net_bridge *br,
@@ -716,7 +738,7 @@ void br_vlan_flush(struct net_bridge *br)
 	ASSERT_RTNL();
 
 	vg = br_vlan_group(br);
-	__vlan_flush(vg);
+	__vlan_flush(br, NULL, vg);
 	RCU_INIT_POINTER(br->vlgrp, NULL);
 	synchronize_rcu();
 	__vlan_group_free(vg);
@@ -925,12 +947,15 @@ static void br_vlan_disable_default_pvid(struct net_bridge *br)
 	/* Disable default_pvid on all ports where it is still
 	 * configured.
 	 */
-	if (vlan_default_pvid(br_vlan_group(br), pvid))
-		br_vlan_delete(br, pvid);
+	if (vlan_default_pvid(br_vlan_group(br), pvid)) {
+		if (!br_vlan_delete(br, pvid))
+			br_vlan_notify(br, NULL, pvid, 0, RTM_DELVLAN);
+	}
 
 	list_for_each_entry(p, &br->port_list, list) {
-		if (vlan_default_pvid(nbp_vlan_group(p), pvid))
-			nbp_vlan_delete(p, pvid);
+		if (vlan_default_pvid(nbp_vlan_group(p), pvid) &&
+		    !nbp_vlan_delete(p, pvid))
+			br_vlan_notify(br, p, pvid, 0, RTM_DELVLAN);
 	}
 
 	br->default_pvid = 0;
@@ -972,7 +997,10 @@ int __br_vlan_set_default_pvid(struct net_bridge *br, u16 pvid,
 				  &vlchange, extack);
 		if (err)
 			goto out;
-		br_vlan_delete(br, old_pvid);
+
+		if (br_vlan_delete(br, old_pvid))
+			br_vlan_notify(br, NULL, old_pvid, 0, RTM_DELVLAN);
+		br_vlan_notify(br, NULL, pvid, 0, RTM_NEWVLAN);
 		set_bit(0, changed);
 	}
 
@@ -992,7 +1020,9 @@ int __br_vlan_set_default_pvid(struct net_bridge *br, u16 pvid,
 				   &vlchange, extack);
 		if (err)
 			goto err_port;
-		nbp_vlan_delete(p, old_pvid);
+		if (nbp_vlan_delete(p, old_pvid))
+			br_vlan_notify(br, p, old_pvid, 0, RTM_DELVLAN);
+		br_vlan_notify(p->br, p, pvid, 0, RTM_NEWVLAN);
 		set_bit(p->port_no, changed);
 	}
 
@@ -1007,22 +1037,28 @@ err_port:
 		if (!test_bit(p->port_no, changed))
 			continue;
 
-		if (old_pvid)
+		if (old_pvid) {
 			nbp_vlan_add(p, old_pvid,
 				     BRIDGE_VLAN_INFO_PVID |
 				     BRIDGE_VLAN_INFO_UNTAGGED,
 				     &vlchange, NULL);
+			br_vlan_notify(p->br, p, old_pvid, 0, RTM_NEWVLAN);
+		}
 		nbp_vlan_delete(p, pvid);
+		br_vlan_notify(br, p, pvid, 0, RTM_DELVLAN);
 	}
 
 	if (test_bit(0, changed)) {
-		if (old_pvid)
+		if (old_pvid) {
 			br_vlan_add(br, old_pvid,
 				    BRIDGE_VLAN_INFO_PVID |
 				    BRIDGE_VLAN_INFO_UNTAGGED |
 				    BRIDGE_VLAN_INFO_BRENTRY,
 				    &vlchange, NULL);
+			br_vlan_notify(br, NULL, old_pvid, 0, RTM_NEWVLAN);
+		}
 		br_vlan_delete(br, pvid);
+		br_vlan_notify(br, NULL, pvid, 0, RTM_DELVLAN);
 	}
 	goto out;
 }
@@ -1115,6 +1151,7 @@ int nbp_vlan_init(struct net_bridge_port *p, struct netlink_ext_ack *extack)
 				   &changed, extack);
 		if (ret)
 			goto err_vlan_add;
+		br_vlan_notify(p->br, p, p->br->default_pvid, 0, RTM_NEWVLAN);
 	}
 out:
 	return ret;
@@ -1196,7 +1233,7 @@ void nbp_vlan_flush(struct net_bridge_port *port)
 	ASSERT_RTNL();
 
 	vg = nbp_vlan_group(port);
-	__vlan_flush(vg);
+	__vlan_flush(port->br, port, vg);
 	RCU_INIT_POINTER(port->vlgrp, NULL);
 	synchronize_rcu();
 	__vlan_group_free(vg);
@@ -1462,8 +1499,8 @@ int br_vlan_bridge_event(struct net_device *dev, unsigned long event, void *ptr)
 {
 	struct netdev_notifier_changeupper_info *info;
 	struct net_bridge *br = netdev_priv(dev);
-	bool changed;
-	int ret = 0;
+	int vlcmd = 0, ret = 0;
+	bool changed = false;
 
 	switch (event) {
 	case NETDEV_REGISTER:
@@ -1471,9 +1508,11 @@ int br_vlan_bridge_event(struct net_device *dev, unsigned long event, void *ptr)
 				  BRIDGE_VLAN_INFO_PVID |
 				  BRIDGE_VLAN_INFO_UNTAGGED |
 				  BRIDGE_VLAN_INFO_BRENTRY, &changed, NULL);
+		vlcmd = RTM_NEWVLAN;
 		break;
 	case NETDEV_UNREGISTER:
-		br_vlan_delete(br, br->default_pvid);
+		changed = !br_vlan_delete(br, br->default_pvid);
+		vlcmd = RTM_DELVLAN;
 		break;
 	case NETDEV_CHANGEUPPER:
 		info = ptr;
@@ -1487,6 +1526,8 @@ int br_vlan_bridge_event(struct net_device *dev, unsigned long event, void *ptr)
 		br_vlan_link_state_change(dev, br);
 		break;
 	}
+	if (changed)
+		br_vlan_notify(br, NULL, br->default_pvid, 0, vlcmd);
 
 	return ret;
 }
-- 
cgit v1.2.3-59-g8ed1b


From d3a56931f9c8644a52e7a6735954c84ea895e8c3 Mon Sep 17 00:00:00 2001
From: Magnus Karlsson <magnus.karlsson@intel.com>
Date: Tue, 14 Jan 2020 10:49:25 +0100
Subject: xsk: Support allocations of large umems

When registering a umem area that is sufficiently large (>1G on an
x86), kmalloc cannot be used to allocate one of the internal data
structures, as the size requested gets too large. Use kvmalloc instead
that falls back on vmalloc if the allocation is too large for kmalloc.

Also add accounting for this structure as it is triggered by a user
space action (the XDP_UMEM_REG setsockopt) and it is by far the
largest structure of kernel allocated memory in xsk.

Reported-by: Ryan Goodfellow <rgoodfel@isi.edu>
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Jonathan Lemon <jonathan.lemon@gmail.com>
Link: https://lore.kernel.org/bpf/1578995365-7050-1-git-send-email-magnus.karlsson@intel.com
---
 net/xdp/xdp_umem.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/net/xdp/xdp_umem.c b/net/xdp/xdp_umem.c
index 3049af269fbf..f93e917e0929 100644
--- a/net/xdp/xdp_umem.c
+++ b/net/xdp/xdp_umem.c
@@ -249,7 +249,7 @@ static void xdp_umem_release(struct xdp_umem *umem)
 	xdp_umem_unmap_pages(umem);
 	xdp_umem_unpin_pages(umem);
 
-	kfree(umem->pages);
+	kvfree(umem->pages);
 	umem->pages = NULL;
 
 	xdp_umem_unaccount_pages(umem);
@@ -409,7 +409,8 @@ static int xdp_umem_reg(struct xdp_umem *umem, struct xdp_umem_reg *mr)
 	if (err)
 		goto out_account;
 
-	umem->pages = kcalloc(umem->npgs, sizeof(*umem->pages), GFP_KERNEL);
+	umem->pages = kvcalloc(umem->npgs, sizeof(*umem->pages),
+			       GFP_KERNEL_ACCOUNT);
 	if (!umem->pages) {
 		err = -ENOMEM;
 		goto out_pin;
@@ -419,7 +420,7 @@ static int xdp_umem_reg(struct xdp_umem *umem, struct xdp_umem_reg *mr)
 	if (!err)
 		return 0;
 
-	kfree(umem->pages);
+	kvfree(umem->pages);
 
 out_pin:
 	xdp_umem_unpin_pages(umem);
-- 
cgit v1.2.3-59-g8ed1b


From 8482941f09067da42f9c3362e15bfb3f3c19d610 Mon Sep 17 00:00:00 2001
From: Yonghong Song <yhs@fb.com>
Date: Tue, 14 Jan 2020 19:50:02 -0800
Subject: bpf: Add bpf_send_signal_thread() helper

Commit 8b401f9ed244 ("bpf: implement bpf_send_signal() helper")
added helper bpf_send_signal() which permits bpf program to
send a signal to the current process. The signal may be
delivered to any threads in the process.

We found a use case where sending the signal to the current
thread is more preferable.
  - A bpf program will collect the stack trace and then
    send signal to the user application.
  - The user application will add some thread specific
    information to the just collected stack trace for
    later analysis.

If bpf_send_signal() is used, user application will need
to check whether the thread receiving the signal matches
the thread collecting the stack by checking thread id.
If not, it will need to send signal to another thread
through pthread_kill().

This patch proposed a new helper bpf_send_signal_thread(),
which sends the signal to the thread corresponding to
the current kernel task. This way, user space is guaranteed that
bpf_program execution context and user space signal handling
context are the same thread.

Signed-off-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200115035002.602336-1-yhs@fb.com
---
 include/uapi/linux/bpf.h       | 19 +++++++++++++++++--
 kernel/trace/bpf_trace.c       | 27 ++++++++++++++++++++++++---
 tools/include/uapi/linux/bpf.h | 19 +++++++++++++++++--
 3 files changed, 58 insertions(+), 7 deletions(-)

diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 52966e758fe5..a88837d4dd18 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -2714,7 +2714,8 @@ union bpf_attr {
  *
  * int bpf_send_signal(u32 sig)
  *	Description
- *		Send signal *sig* to the current task.
+ *		Send signal *sig* to the process of the current task.
+ *		The signal may be delivered to any of this process's threads.
  *	Return
  *		0 on success or successfully queued.
  *
@@ -2850,6 +2851,19 @@ union bpf_attr {
  *	Return
  *		0 on success, or a negative error in case of failure.
  *
+ * int bpf_send_signal_thread(u32 sig)
+ *	Description
+ *		Send signal *sig* to the thread corresponding to the current task.
+ *	Return
+ *		0 on success or successfully queued.
+ *
+ *		**-EBUSY** if work queue under nmi is full.
+ *
+ *		**-EINVAL** if *sig* is invalid.
+ *
+ *		**-EPERM** if no permission to send the *sig*.
+ *
+ *		**-EAGAIN** if bpf program can try again.
  */
 #define __BPF_FUNC_MAPPER(FN)		\
 	FN(unspec),			\
@@ -2968,7 +2982,8 @@ union bpf_attr {
 	FN(probe_read_kernel),		\
 	FN(probe_read_user_str),	\
 	FN(probe_read_kernel_str),	\
-	FN(tcp_send_ack),
+	FN(tcp_send_ack),		\
+	FN(send_signal_thread),
 
 /* integer value in 'imm' field of BPF_CALL instruction selects which helper
  * function eBPF program intends to call
diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
index e5ef4ae9edb5..19e793aa441a 100644
--- a/kernel/trace/bpf_trace.c
+++ b/kernel/trace/bpf_trace.c
@@ -703,6 +703,7 @@ struct send_signal_irq_work {
 	struct irq_work irq_work;
 	struct task_struct *task;
 	u32 sig;
+	enum pid_type type;
 };
 
 static DEFINE_PER_CPU(struct send_signal_irq_work, send_signal_work);
@@ -712,10 +713,10 @@ static void do_bpf_send_signal(struct irq_work *entry)
 	struct send_signal_irq_work *work;
 
 	work = container_of(entry, struct send_signal_irq_work, irq_work);
-	group_send_sig_info(work->sig, SEND_SIG_PRIV, work->task, PIDTYPE_TGID);
+	group_send_sig_info(work->sig, SEND_SIG_PRIV, work->task, work->type);
 }
 
-BPF_CALL_1(bpf_send_signal, u32, sig)
+static int bpf_send_signal_common(u32 sig, enum pid_type type)
 {
 	struct send_signal_irq_work *work = NULL;
 
@@ -748,11 +749,17 @@ BPF_CALL_1(bpf_send_signal, u32, sig)
 		 */
 		work->task = current;
 		work->sig = sig;
+		work->type = type;
 		irq_work_queue(&work->irq_work);
 		return 0;
 	}
 
-	return group_send_sig_info(sig, SEND_SIG_PRIV, current, PIDTYPE_TGID);
+	return group_send_sig_info(sig, SEND_SIG_PRIV, current, type);
+}
+
+BPF_CALL_1(bpf_send_signal, u32, sig)
+{
+	return bpf_send_signal_common(sig, PIDTYPE_TGID);
 }
 
 static const struct bpf_func_proto bpf_send_signal_proto = {
@@ -762,6 +769,18 @@ static const struct bpf_func_proto bpf_send_signal_proto = {
 	.arg1_type	= ARG_ANYTHING,
 };
 
+BPF_CALL_1(bpf_send_signal_thread, u32, sig)
+{
+	return bpf_send_signal_common(sig, PIDTYPE_PID);
+}
+
+static const struct bpf_func_proto bpf_send_signal_thread_proto = {
+	.func		= bpf_send_signal_thread,
+	.gpl_only	= false,
+	.ret_type	= RET_INTEGER,
+	.arg1_type	= ARG_ANYTHING,
+};
+
 static const struct bpf_func_proto *
 tracing_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog)
 {
@@ -822,6 +841,8 @@ tracing_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog)
 #endif
 	case BPF_FUNC_send_signal:
 		return &bpf_send_signal_proto;
+	case BPF_FUNC_send_signal_thread:
+		return &bpf_send_signal_thread_proto;
 	default:
 		return NULL;
 	}
diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index 52966e758fe5..a88837d4dd18 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -2714,7 +2714,8 @@ union bpf_attr {
  *
  * int bpf_send_signal(u32 sig)
  *	Description
- *		Send signal *sig* to the current task.
+ *		Send signal *sig* to the process of the current task.
+ *		The signal may be delivered to any of this process's threads.
  *	Return
  *		0 on success or successfully queued.
  *
@@ -2850,6 +2851,19 @@ union bpf_attr {
  *	Return
  *		0 on success, or a negative error in case of failure.
  *
+ * int bpf_send_signal_thread(u32 sig)
+ *	Description
+ *		Send signal *sig* to the thread corresponding to the current task.
+ *	Return
+ *		0 on success or successfully queued.
+ *
+ *		**-EBUSY** if work queue under nmi is full.
+ *
+ *		**-EINVAL** if *sig* is invalid.
+ *
+ *		**-EPERM** if no permission to send the *sig*.
+ *
+ *		**-EAGAIN** if bpf program can try again.
  */
 #define __BPF_FUNC_MAPPER(FN)		\
 	FN(unspec),			\
@@ -2968,7 +2982,8 @@ union bpf_attr {
 	FN(probe_read_kernel),		\
 	FN(probe_read_user_str),	\
 	FN(probe_read_kernel_str),	\
-	FN(tcp_send_ack),
+	FN(tcp_send_ack),		\
+	FN(send_signal_thread),
 
 /* integer value in 'imm' field of BPF_CALL instruction selects which helper
  * function eBPF program intends to call
-- 
cgit v1.2.3-59-g8ed1b


From ab8b7f0cb3585ecb40e3305850f28658a1b8940e Mon Sep 17 00:00:00 2001
From: Yonghong Song <yhs@fb.com>
Date: Tue, 14 Jan 2020 19:50:03 -0800
Subject: tools/bpf: Add self tests for bpf_send_signal_thread()

The test_progs send_signal() is amended to test
bpf_send_signal_thread() as well.

   $ ./test_progs -n 40
   #40/1 send_signal_tracepoint:OK
   #40/2 send_signal_perf:OK
   #40/3 send_signal_nmi:OK
   #40/4 send_signal_tracepoint_thread:OK
   #40/5 send_signal_perf_thread:OK
   #40/6 send_signal_nmi_thread:OK
   #40 send_signal:OK
   Summary: 1/6 PASSED, 0 SKIPPED, 0 FAILED

Also took this opportunity to rewrite the send_signal test
using skeleton framework and array mmap to make code
simpler and more readable.

Signed-off-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200115035003.602425-1-yhs@fb.com
---
 .../testing/selftests/bpf/prog_tests/send_signal.c | 128 ++++++++-------------
 .../selftests/bpf/progs/test_send_signal_kern.c    |  51 ++++----
 2 files changed, 73 insertions(+), 106 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/send_signal.c b/tools/testing/selftests/bpf/prog_tests/send_signal.c
index b607112c64e7..d4cedd86c424 100644
--- a/tools/testing/selftests/bpf/prog_tests/send_signal.c
+++ b/tools/testing/selftests/bpf/prog_tests/send_signal.c
@@ -1,5 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <test_progs.h>
+#include "test_send_signal_kern.skel.h"
 
 static volatile int sigusr1_received = 0;
 
@@ -9,17 +10,15 @@ static void sigusr1_handler(int signum)
 }
 
 static void test_send_signal_common(struct perf_event_attr *attr,
-				    int prog_type,
+				    bool signal_thread,
 				    const char *test_name)
 {
-	int err = -1, pmu_fd, prog_fd, info_map_fd, status_map_fd;
-	const char *file = "./test_send_signal_kern.o";
-	struct bpf_object *obj = NULL;
+	struct test_send_signal_kern *skel;
 	int pipe_c2p[2], pipe_p2c[2];
-	__u32 key = 0, duration = 0;
+	int err = -1, pmu_fd = -1;
+	__u32 duration = 0;
 	char buf[256];
 	pid_t pid;
-	__u64 val;
 
 	if (CHECK(pipe(pipe_c2p), test_name,
 		  "pipe pipe_c2p error: %s\n", strerror(errno)))
@@ -73,45 +72,42 @@ static void test_send_signal_common(struct perf_event_attr *attr,
 	close(pipe_c2p[1]); /* close write */
 	close(pipe_p2c[0]); /* close read */
 
-	err = bpf_prog_load(file, prog_type, &obj, &prog_fd);
-	if (CHECK(err < 0, test_name, "bpf_prog_load error: %s\n",
-		  strerror(errno)))
-		goto prog_load_failure;
-
-	pmu_fd = syscall(__NR_perf_event_open, attr, pid, -1,
-			 -1 /* group id */, 0 /* flags */);
-	if (CHECK(pmu_fd < 0, test_name, "perf_event_open error: %s\n",
-		  strerror(errno))) {
-		err = -1;
-		goto close_prog;
-	}
-
-	err = ioctl(pmu_fd, PERF_EVENT_IOC_ENABLE, 0);
-	if (CHECK(err < 0, test_name, "ioctl perf_event_ioc_enable error: %s\n",
-		  strerror(errno)))
-		goto disable_pmu;
+	skel = test_send_signal_kern__open_and_load();
+	if (CHECK(!skel, "skel_open_and_load", "skeleton open_and_load failed\n"))
+		goto skel_open_load_failure;
 
-	err = ioctl(pmu_fd, PERF_EVENT_IOC_SET_BPF, prog_fd);
-	if (CHECK(err < 0, test_name, "ioctl perf_event_ioc_set_bpf error: %s\n",
-		  strerror(errno)))
-		goto disable_pmu;
+	/* add a delay for child thread to ramp up */
+	usleep(100);
 
-	err = -1;
-	info_map_fd = bpf_object__find_map_fd_by_name(obj, "info_map");
-	if (CHECK(info_map_fd < 0, test_name, "find map %s error\n", "info_map"))
-		goto disable_pmu;
+	if (!attr) {
+		err = test_send_signal_kern__attach(skel);
+		if (CHECK(err, "skel_attach", "skeleton attach failed\n")) {
+			err = -1;
+			goto destroy_skel;
+		}
+	} else {
+		pmu_fd = syscall(__NR_perf_event_open, attr, pid, -1,
+				 -1 /* group id */, 0 /* flags */);
+		if (CHECK(pmu_fd < 0, test_name, "perf_event_open error: %s\n",
+			strerror(errno))) {
+			err = -1;
+			goto destroy_skel;
+		}
 
-	status_map_fd = bpf_object__find_map_fd_by_name(obj, "status_map");
-	if (CHECK(status_map_fd < 0, test_name, "find map %s error\n", "status_map"))
-		goto disable_pmu;
+		skel->links.send_signal_perf =
+			bpf_program__attach_perf_event(skel->progs.send_signal_perf, pmu_fd);
+		if (CHECK(IS_ERR(skel->links.send_signal_perf), "attach_perf_event",
+			  "err %ld\n", PTR_ERR(skel->links.send_signal_perf)))
+			goto disable_pmu;
+	}
 
 	/* wait until child signal handler installed */
 	read(pipe_c2p[0], buf, 1);
 
 	/* trigger the bpf send_signal */
-	key = 0;
-	val = (((__u64)(SIGUSR1)) << 32) | pid;
-	bpf_map_update_elem(info_map_fd, &key, &val, 0);
+	skel->bss->pid = pid;
+	skel->bss->sig = SIGUSR1;
+	skel->bss->signal_thread = signal_thread;
 
 	/* notify child that bpf program can send_signal now */
 	write(pipe_p2c[1], buf, 1);
@@ -132,46 +128,20 @@ static void test_send_signal_common(struct perf_event_attr *attr,
 
 disable_pmu:
 	close(pmu_fd);
-close_prog:
-	bpf_object__close(obj);
-prog_load_failure:
+destroy_skel:
+	test_send_signal_kern__destroy(skel);
+skel_open_load_failure:
 	close(pipe_c2p[0]);
 	close(pipe_p2c[1]);
 	wait(NULL);
 }
 
-static void test_send_signal_tracepoint(void)
+static void test_send_signal_tracepoint(bool signal_thread)
 {
-	const char *id_path = "/sys/kernel/debug/tracing/events/syscalls/sys_enter_nanosleep/id";
-	struct perf_event_attr attr = {
-		.type = PERF_TYPE_TRACEPOINT,
-		.sample_type = PERF_SAMPLE_RAW | PERF_SAMPLE_CALLCHAIN,
-		.sample_period = 1,
-		.wakeup_events = 1,
-	};
-	__u32 duration = 0;
-	int bytes, efd;
-	char buf[256];
-
-	efd = open(id_path, O_RDONLY, 0);
-	if (CHECK(efd < 0, "tracepoint",
-		  "open syscalls/sys_enter_nanosleep/id failure: %s\n",
-		  strerror(errno)))
-		return;
-
-	bytes = read(efd, buf, sizeof(buf));
-	close(efd);
-	if (CHECK(bytes <= 0 || bytes >= sizeof(buf), "tracepoint",
-		  "read syscalls/sys_enter_nanosleep/id failure: %s\n",
-		  strerror(errno)))
-		return;
-
-	attr.config = strtol(buf, NULL, 0);
-
-	test_send_signal_common(&attr, BPF_PROG_TYPE_TRACEPOINT, "tracepoint");
+	test_send_signal_common(NULL, signal_thread, "tracepoint");
 }
 
-static void test_send_signal_perf(void)
+static void test_send_signal_perf(bool signal_thread)
 {
 	struct perf_event_attr attr = {
 		.sample_period = 1,
@@ -179,11 +149,10 @@ static void test_send_signal_perf(void)
 		.config = PERF_COUNT_SW_CPU_CLOCK,
 	};
 
-	test_send_signal_common(&attr, BPF_PROG_TYPE_PERF_EVENT,
-				"perf_sw_event");
+	test_send_signal_common(&attr, signal_thread, "perf_sw_event");
 }
 
-static void test_send_signal_nmi(void)
+static void test_send_signal_nmi(bool signal_thread)
 {
 	struct perf_event_attr attr = {
 		.sample_freq = 50,
@@ -210,16 +179,21 @@ static void test_send_signal_nmi(void)
 		close(pmu_fd);
 	}
 
-	test_send_signal_common(&attr, BPF_PROG_TYPE_PERF_EVENT,
-				"perf_hw_event");
+	test_send_signal_common(&attr, signal_thread, "perf_hw_event");
 }
 
 void test_send_signal(void)
 {
 	if (test__start_subtest("send_signal_tracepoint"))
-		test_send_signal_tracepoint();
+		test_send_signal_tracepoint(false);
 	if (test__start_subtest("send_signal_perf"))
-		test_send_signal_perf();
+		test_send_signal_perf(false);
 	if (test__start_subtest("send_signal_nmi"))
-		test_send_signal_nmi();
+		test_send_signal_nmi(false);
+	if (test__start_subtest("send_signal_tracepoint_thread"))
+		test_send_signal_tracepoint(true);
+	if (test__start_subtest("send_signal_perf_thread"))
+		test_send_signal_perf(true);
+	if (test__start_subtest("send_signal_nmi_thread"))
+		test_send_signal_nmi(true);
 }
diff --git a/tools/testing/selftests/bpf/progs/test_send_signal_kern.c b/tools/testing/selftests/bpf/progs/test_send_signal_kern.c
index 0e6be01157e6..726733f9f20b 100644
--- a/tools/testing/selftests/bpf/progs/test_send_signal_kern.c
+++ b/tools/testing/selftests/bpf/progs/test_send_signal_kern.c
@@ -4,44 +4,37 @@
 #include <linux/version.h>
 #include "bpf_helpers.h"
 
-struct {
-	__uint(type, BPF_MAP_TYPE_ARRAY);
-	__uint(max_entries, 1);
-	__type(key, __u32);
-	__type(value, __u64);
-} info_map SEC(".maps");
-
-struct {
-	__uint(type, BPF_MAP_TYPE_ARRAY);
-	__uint(max_entries, 1);
-	__type(key, __u32);
-	__type(value, __u64);
-} status_map SEC(".maps");
-
-SEC("send_signal_demo")
-int bpf_send_signal_test(void *ctx)
+__u32 sig = 0, pid = 0, status = 0, signal_thread = 0;
+
+static __always_inline int bpf_send_signal_test(void *ctx)
 {
-	__u64 *info_val, *status_val;
-	__u32 key = 0, pid, sig;
 	int ret;
 
-	status_val = bpf_map_lookup_elem(&status_map, &key);
-	if (!status_val || *status_val != 0)
-		return 0;
-
-	info_val = bpf_map_lookup_elem(&info_map, &key);
-	if (!info_val || *info_val == 0)
+	if (status != 0 || sig == 0 || pid == 0)
 		return 0;
 
-	sig = *info_val >> 32;
-	pid = *info_val & 0xffffFFFF;
-
 	if ((bpf_get_current_pid_tgid() >> 32) == pid) {
-		ret = bpf_send_signal(sig);
+		if (signal_thread)
+			ret = bpf_send_signal_thread(sig);
+		else
+			ret = bpf_send_signal(sig);
 		if (ret == 0)
-			*status_val = 1;
+			status = 1;
 	}
 
 	return 0;
 }
+
+SEC("tracepoint/syscalls/sys_enter_nanosleep")
+int send_signal_tp(void *ctx)
+{
+	return bpf_send_signal_test(ctx);
+}
+
+SEC("perf_event")
+int send_signal_perf(void *ctx)
+{
+	return bpf_send_signal_test(ctx);
+}
+
 char __license[] SEC("license") = "GPL";
-- 
cgit v1.2.3-59-g8ed1b


From 9173cac3b64e6785dd604f5075e6035b045a0026 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Wed, 15 Jan 2020 11:08:56 -0800
Subject: libbpf: Support .text sub-calls relocations

The LLVM patch https://reviews.llvm.org/D72197 makes LLVM emit function call
relocations within the same section. This includes a default .text section,
which contains any BPF sub-programs. This wasn't the case before and so libbpf
was able to get a way with slightly simpler handling of subprogram call
relocations.

This patch adds support for .text section relocations. It needs to ensure
correct order of relocations, so does two passes:
- first, relocate .text instructions, if there are any relocations in it;
- then process all the other programs and copy over patched .text instructions
for all sub-program calls.

v1->v2:
- break early once .text program is processed.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com>
Cc: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200115190856.2391325-1-andriin@fb.com
---
 tools/lib/bpf/libbpf.c | 29 ++++++++++++++++++++++-------
 1 file changed, 22 insertions(+), 7 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 0c229f00a67e..23868883477f 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -4692,13 +4692,7 @@ bpf_program__reloc_text(struct bpf_program *prog, struct bpf_object *obj,
 	size_t new_cnt;
 	int err;
 
-	if (prog->idx == obj->efile.text_shndx) {
-		pr_warn("relo in .text insn %d into off %d (insn #%d)\n",
-			relo->insn_idx, relo->sym_off, relo->sym_off / 8);
-		return -LIBBPF_ERRNO__RELOC;
-	}
-
-	if (prog->main_prog_cnt == 0) {
+	if (prog->idx != obj->efile.text_shndx && prog->main_prog_cnt == 0) {
 		text = bpf_object__find_prog_by_idx(obj, obj->efile.text_shndx);
 		if (!text) {
 			pr_warn("no .text section found yet relo into text exist\n");
@@ -4728,6 +4722,7 @@ bpf_program__reloc_text(struct bpf_program *prog, struct bpf_object *obj,
 			 text->insns_cnt, text->section_name,
 			 prog->section_name);
 	}
+
 	insn = &prog->insns[relo->insn_idx];
 	insn->imm += relo->sym_off / 8 + prog->main_prog_cnt - relo->insn_idx;
 	return 0;
@@ -4807,8 +4802,28 @@ bpf_object__relocate(struct bpf_object *obj, const char *targ_btf_path)
 			return err;
 		}
 	}
+	/* ensure .text is relocated first, as it's going to be copied as-is
+	 * later for sub-program calls
+	 */
+	for (i = 0; i < obj->nr_programs; i++) {
+		prog = &obj->programs[i];
+		if (prog->idx != obj->efile.text_shndx)
+			continue;
+
+		err = bpf_program__relocate(prog, obj);
+		if (err) {
+			pr_warn("failed to relocate '%s'\n", prog->section_name);
+			return err;
+		}
+		break;
+	}
+	/* now relocate everything but .text, which by now is relocated
+	 * properly, so we can copy raw sub-program instructions as is safely
+	 */
 	for (i = 0; i < obj->nr_programs; i++) {
 		prog = &obj->programs[i];
+		if (prog->idx == obj->efile.text_shndx)
+			continue;
 
 		err = bpf_program__relocate(prog, obj);
 		if (err) {
-- 
cgit v1.2.3-59-g8ed1b


From 83e4b88be1ff9671f2a2fb040a09c45cc3fb40e6 Mon Sep 17 00:00:00 2001
From: Eelco Chaudron <echaudro@redhat.com>
Date: Wed, 15 Jan 2020 13:15:39 +0000
Subject: selftests/bpf: Add a test for attaching a bpf fentry/fexit trace to
 an XDP program

Add a test that will attach a FENTRY and FEXIT program to the XDP test
program. It will also verify data from the XDP context on FENTRY and
verifies the return code on exit.

Signed-off-by: Eelco Chaudron <echaudro@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/157909410480.47481.11202505690938004673.stgit@xdp-tutorial
---
 .../testing/selftests/bpf/prog_tests/xdp_bpf2bpf.c | 65 ++++++++++++++++++++++
 .../testing/selftests/bpf/progs/test_xdp_bpf2bpf.c | 44 +++++++++++++++
 2 files changed, 109 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/xdp_bpf2bpf.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_xdp_bpf2bpf.c

diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_bpf2bpf.c b/tools/testing/selftests/bpf/prog_tests/xdp_bpf2bpf.c
new file mode 100644
index 000000000000..6b56bdc73ebc
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/xdp_bpf2bpf.c
@@ -0,0 +1,65 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <test_progs.h>
+#include <net/if.h>
+#include "test_xdp.skel.h"
+#include "test_xdp_bpf2bpf.skel.h"
+
+void test_xdp_bpf2bpf(void)
+{
+	__u32 duration = 0, retval, size;
+	char buf[128];
+	int err, pkt_fd, map_fd;
+	struct iphdr *iph = (void *)buf + sizeof(struct ethhdr);
+	struct iptnl_info value4 = {.family = AF_INET};
+	struct test_xdp *pkt_skel = NULL;
+	struct test_xdp_bpf2bpf *ftrace_skel = NULL;
+	struct vip key4 = {.protocol = 6, .family = AF_INET};
+	DECLARE_LIBBPF_OPTS(bpf_object_open_opts, opts);
+
+	/* Load XDP program to introspect */
+	pkt_skel = test_xdp__open_and_load();
+	if (CHECK(!pkt_skel, "pkt_skel_load", "test_xdp skeleton failed\n"))
+		return;
+
+	pkt_fd = bpf_program__fd(pkt_skel->progs._xdp_tx_iptunnel);
+
+	map_fd = bpf_map__fd(pkt_skel->maps.vip2tnl);
+	bpf_map_update_elem(map_fd, &key4, &value4, 0);
+
+	/* Load trace program */
+	opts.attach_prog_fd = pkt_fd,
+	ftrace_skel = test_xdp_bpf2bpf__open_opts(&opts);
+	if (CHECK(!ftrace_skel, "__open", "ftrace skeleton failed\n"))
+		goto out;
+
+	err = test_xdp_bpf2bpf__load(ftrace_skel);
+	if (CHECK(err, "__load", "ftrace skeleton failed\n"))
+		goto out;
+
+	err = test_xdp_bpf2bpf__attach(ftrace_skel);
+	if (CHECK(err, "ftrace_attach", "ftrace attach failed: %d\n", err))
+		goto out;
+
+	/* Run test program */
+	err = bpf_prog_test_run(pkt_fd, 1, &pkt_v4, sizeof(pkt_v4),
+				buf, &size, &retval, &duration);
+
+	if (CHECK(err || retval != XDP_TX || size != 74 ||
+		  iph->protocol != IPPROTO_IPIP, "ipv4",
+		  "err %d errno %d retval %d size %d\n",
+		  err, errno, retval, size))
+		goto out;
+
+	/* Verify test results */
+	if (CHECK(ftrace_skel->bss->test_result_fentry != if_nametoindex("lo"),
+		  "result", "fentry failed err %llu\n",
+		  ftrace_skel->bss->test_result_fentry))
+		goto out;
+
+	CHECK(ftrace_skel->bss->test_result_fexit != XDP_TX, "result",
+	      "fexit failed err %llu\n", ftrace_skel->bss->test_result_fexit);
+
+out:
+	test_xdp__destroy(pkt_skel);
+	test_xdp_bpf2bpf__destroy(ftrace_skel);
+}
diff --git a/tools/testing/selftests/bpf/progs/test_xdp_bpf2bpf.c b/tools/testing/selftests/bpf/progs/test_xdp_bpf2bpf.c
new file mode 100644
index 000000000000..f8f105af6743
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_xdp_bpf2bpf.c
@@ -0,0 +1,44 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/bpf.h>
+#include "bpf_helpers.h"
+#include "bpf_trace_helpers.h"
+
+struct net_device {
+	/* Structure does not need to contain all entries,
+	 * as "preserve_access_index" will use BTF to fix this...
+	 */
+	int ifindex;
+} __attribute__((preserve_access_index));
+
+struct xdp_rxq_info {
+	/* Structure does not need to contain all entries,
+	 * as "preserve_access_index" will use BTF to fix this...
+	 */
+	struct net_device *dev;
+	__u32 queue_index;
+} __attribute__((preserve_access_index));
+
+struct xdp_buff {
+	void *data;
+	void *data_end;
+	void *data_meta;
+	void *data_hard_start;
+	unsigned long handle;
+	struct xdp_rxq_info *rxq;
+} __attribute__((preserve_access_index));
+
+__u64 test_result_fentry = 0;
+SEC("fentry/_xdp_tx_iptunnel")
+int BPF_PROG(trace_on_entry, struct xdp_buff *xdp)
+{
+	test_result_fentry = xdp->rxq->dev->ifindex;
+	return 0;
+}
+
+__u64 test_result_fexit = 0;
+SEC("fexit/_xdp_tx_iptunnel")
+int BPF_PROG(trace_on_exit, struct xdp_buff *xdp, int ret)
+{
+	test_result_fexit = ret;
+	return 0;
+}
-- 
cgit v1.2.3-59-g8ed1b


From 600a87490ff9823d065fc15e86c709e707033ecc Mon Sep 17 00:00:00 2001
From: Alain Michaud <alainm@chromium.org>
Date: Tue, 7 Jan 2020 00:43:17 +0000
Subject: Bluetooth: Implementation of MGMT_OP_SET_BLOCKED_KEYS.

MGMT command is added to receive the list of blocked keys from
user-space.

The list is used to:
1) Block keys from being distributed by the device during
   the ke distribution phase of SMP.
2) Filter out any keys that were previously saved so
   they are no longer used.

Signed-off-by: Alain Michaud <alainm@chromium.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 include/net/bluetooth/hci_core.h | 10 +++++
 include/net/bluetooth/mgmt.h     | 17 ++++++++
 net/bluetooth/hci_core.c         | 85 ++++++++++++++++++++++++++++++++++++----
 net/bluetooth/hci_debugfs.c      | 17 ++++++++
 net/bluetooth/mgmt.c             | 76 +++++++++++++++++++++++++++++++++++
 net/bluetooth/smp.c              | 18 +++++++++
 6 files changed, 215 insertions(+), 8 deletions(-)

diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
index faebe3859931..89ecf0a80aa1 100644
--- a/include/net/bluetooth/hci_core.h
+++ b/include/net/bluetooth/hci_core.h
@@ -118,6 +118,13 @@ struct bt_uuid {
 	u8 svc_hint;
 };
 
+struct blocked_key {
+	struct list_head list;
+	struct rcu_head rcu;
+	u8 type;
+	u8 val[16];
+};
+
 struct smp_csrk {
 	bdaddr_t bdaddr;
 	u8 bdaddr_type;
@@ -397,6 +404,7 @@ struct hci_dev {
 	struct list_head	le_conn_params;
 	struct list_head	pend_le_conns;
 	struct list_head	pend_le_reports;
+	struct list_head	blocked_keys;
 
 	struct hci_dev_stats	stat;
 
@@ -1123,6 +1131,8 @@ struct smp_irk *hci_find_irk_by_addr(struct hci_dev *hdev, bdaddr_t *bdaddr,
 struct smp_irk *hci_add_irk(struct hci_dev *hdev, bdaddr_t *bdaddr,
 			    u8 addr_type, u8 val[16], bdaddr_t *rpa);
 void hci_remove_irk(struct hci_dev *hdev, bdaddr_t *bdaddr, u8 addr_type);
+bool hci_is_blocked_key(struct hci_dev *hdev, u8 type, u8 val[16]);
+void hci_blocked_keys_clear(struct hci_dev *hdev);
 void hci_smp_irks_clear(struct hci_dev *hdev);
 
 bool hci_bdaddr_is_paired(struct hci_dev *hdev, bdaddr_t *bdaddr, u8 type);
diff --git a/include/net/bluetooth/mgmt.h b/include/net/bluetooth/mgmt.h
index 9cee7ddc6741..a90666af05bd 100644
--- a/include/net/bluetooth/mgmt.h
+++ b/include/net/bluetooth/mgmt.h
@@ -654,6 +654,23 @@ struct mgmt_cp_set_phy_confguration {
 } __packed;
 #define MGMT_SET_PHY_CONFIGURATION_SIZE	4
 
+#define MGMT_OP_SET_BLOCKED_KEYS	0x0046
+
+#define HCI_BLOCKED_KEY_TYPE_LINKKEY	0x00
+#define HCI_BLOCKED_KEY_TYPE_LTK	0x01
+#define HCI_BLOCKED_KEY_TYPE_IRK	0x02
+
+struct mgmt_blocked_key_info {
+	__u8 type;
+	__u8 val[16];
+} __packed;
+
+struct mgmt_cp_set_blocked_keys {
+	__le16 key_count;
+	struct mgmt_blocked_key_info keys[0];
+} __packed;
+#define MGMT_OP_SET_BLOCKED_KEYS_SIZE 2
+
 #define MGMT_EV_CMD_COMPLETE		0x0001
 struct mgmt_ev_cmd_complete {
 	__le16	opcode;
diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c
index 9e19d5a3aac8..f0298db26dc3 100644
--- a/net/bluetooth/hci_core.c
+++ b/net/bluetooth/hci_core.c
@@ -2311,6 +2311,33 @@ void hci_smp_irks_clear(struct hci_dev *hdev)
 	}
 }
 
+void hci_blocked_keys_clear(struct hci_dev *hdev)
+{
+	struct blocked_key *b;
+
+	list_for_each_entry_rcu(b, &hdev->blocked_keys, list) {
+		list_del_rcu(&b->list);
+		kfree_rcu(b, rcu);
+	}
+}
+
+bool hci_is_blocked_key(struct hci_dev *hdev, u8 type, u8 val[16])
+{
+	bool blocked = false;
+	struct blocked_key *b;
+
+	rcu_read_lock();
+	list_for_each_entry(b, &hdev->blocked_keys, list) {
+		if (b->type == type && !memcmp(b->val, val, sizeof(b->val))) {
+			blocked = true;
+			break;
+		}
+	}
+
+	rcu_read_unlock();
+	return blocked;
+}
+
 struct link_key *hci_find_link_key(struct hci_dev *hdev, bdaddr_t *bdaddr)
 {
 	struct link_key *k;
@@ -2319,6 +2346,16 @@ struct link_key *hci_find_link_key(struct hci_dev *hdev, bdaddr_t *bdaddr)
 	list_for_each_entry_rcu(k, &hdev->link_keys, list) {
 		if (bacmp(bdaddr, &k->bdaddr) == 0) {
 			rcu_read_unlock();
+
+			if (hci_is_blocked_key(hdev,
+					       HCI_BLOCKED_KEY_TYPE_LINKKEY,
+					       k->val)) {
+				bt_dev_warn_ratelimited(hdev,
+							"Link key blocked for %pMR",
+							&k->bdaddr);
+				return NULL;
+			}
+
 			return k;
 		}
 	}
@@ -2387,6 +2424,15 @@ struct smp_ltk *hci_find_ltk(struct hci_dev *hdev, bdaddr_t *bdaddr,
 
 		if (smp_ltk_is_sc(k) || ltk_role(k->type) == role) {
 			rcu_read_unlock();
+
+			if (hci_is_blocked_key(hdev, HCI_BLOCKED_KEY_TYPE_LTK,
+					       k->val)) {
+				bt_dev_warn_ratelimited(hdev,
+							"LTK blocked for %pMR",
+							&k->bdaddr);
+				return NULL;
+			}
+
 			return k;
 		}
 	}
@@ -2397,31 +2443,42 @@ struct smp_ltk *hci_find_ltk(struct hci_dev *hdev, bdaddr_t *bdaddr,
 
 struct smp_irk *hci_find_irk_by_rpa(struct hci_dev *hdev, bdaddr_t *rpa)
 {
+	struct smp_irk *irk_to_return = NULL;
 	struct smp_irk *irk;
 
 	rcu_read_lock();
 	list_for_each_entry_rcu(irk, &hdev->identity_resolving_keys, list) {
 		if (!bacmp(&irk->rpa, rpa)) {
-			rcu_read_unlock();
-			return irk;
+			irk_to_return = irk;
+			goto done;
 		}
 	}
 
 	list_for_each_entry_rcu(irk, &hdev->identity_resolving_keys, list) {
 		if (smp_irk_matches(hdev, irk->val, rpa)) {
 			bacpy(&irk->rpa, rpa);
-			rcu_read_unlock();
-			return irk;
+			irk_to_return = irk;
+			goto done;
 		}
 	}
+
+done:
+	if (irk_to_return && hci_is_blocked_key(hdev, HCI_BLOCKED_KEY_TYPE_IRK,
+						irk_to_return->val)) {
+		bt_dev_warn_ratelimited(hdev, "Identity key blocked for %pMR",
+					&irk_to_return->bdaddr);
+		irk_to_return = NULL;
+	}
+
 	rcu_read_unlock();
 
-	return NULL;
+	return irk_to_return;
 }
 
 struct smp_irk *hci_find_irk_by_addr(struct hci_dev *hdev, bdaddr_t *bdaddr,
 				     u8 addr_type)
 {
+	struct smp_irk *irk_to_return = NULL;
 	struct smp_irk *irk;
 
 	/* Identity Address must be public or static random */
@@ -2432,13 +2489,23 @@ struct smp_irk *hci_find_irk_by_addr(struct hci_dev *hdev, bdaddr_t *bdaddr,
 	list_for_each_entry_rcu(irk, &hdev->identity_resolving_keys, list) {
 		if (addr_type == irk->addr_type &&
 		    bacmp(bdaddr, &irk->bdaddr) == 0) {
-			rcu_read_unlock();
-			return irk;
+			irk_to_return = irk;
+			goto done;
 		}
 	}
+
+done:
+
+	if (irk_to_return && hci_is_blocked_key(hdev, HCI_BLOCKED_KEY_TYPE_IRK,
+						irk_to_return->val)) {
+		bt_dev_warn_ratelimited(hdev, "Identity key blocked for %pMR",
+					&irk_to_return->bdaddr);
+		irk_to_return = NULL;
+	}
+
 	rcu_read_unlock();
 
-	return NULL;
+	return irk_to_return;
 }
 
 struct link_key *hci_add_link_key(struct hci_dev *hdev, struct hci_conn *conn,
@@ -3244,6 +3311,7 @@ struct hci_dev *hci_alloc_dev(void)
 	INIT_LIST_HEAD(&hdev->pend_le_reports);
 	INIT_LIST_HEAD(&hdev->conn_hash.list);
 	INIT_LIST_HEAD(&hdev->adv_instances);
+	INIT_LIST_HEAD(&hdev->blocked_keys);
 
 	INIT_WORK(&hdev->rx_work, hci_rx_work);
 	INIT_WORK(&hdev->cmd_work, hci_cmd_work);
@@ -3443,6 +3511,7 @@ void hci_unregister_dev(struct hci_dev *hdev)
 	hci_bdaddr_list_clear(&hdev->le_resolv_list);
 	hci_conn_params_clear_all(hdev);
 	hci_discovery_filter_clear(hdev);
+	hci_blocked_keys_clear(hdev);
 	hci_dev_unlock(hdev);
 
 	hci_dev_put(hdev);
diff --git a/net/bluetooth/hci_debugfs.c b/net/bluetooth/hci_debugfs.c
index 402e2cc54044..1c8100bc4e04 100644
--- a/net/bluetooth/hci_debugfs.c
+++ b/net/bluetooth/hci_debugfs.c
@@ -152,6 +152,21 @@ static int blacklist_show(struct seq_file *f, void *p)
 
 DEFINE_SHOW_ATTRIBUTE(blacklist);
 
+static int blocked_keys_show(struct seq_file *f, void *p)
+{
+	struct hci_dev *hdev = f->private;
+	struct blocked_key *key;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(key, &hdev->blocked_keys, list)
+		seq_printf(f, "%u %*phN\n", key->type, 16, key->val);
+	rcu_read_unlock();
+
+	return 0;
+}
+
+DEFINE_SHOW_ATTRIBUTE(blocked_keys);
+
 static int uuids_show(struct seq_file *f, void *p)
 {
 	struct hci_dev *hdev = f->private;
@@ -308,6 +323,8 @@ void hci_debugfs_create_common(struct hci_dev *hdev)
 			    &device_list_fops);
 	debugfs_create_file("blacklist", 0444, hdev->debugfs, hdev,
 			    &blacklist_fops);
+	debugfs_create_file("blocked_keys", 0444, hdev->debugfs, hdev,
+			    &blocked_keys_fops);
 	debugfs_create_file("uuids", 0444, hdev->debugfs, hdev, &uuids_fops);
 	debugfs_create_file("remote_oob", 0400, hdev->debugfs, hdev,
 			    &remote_oob_fops);
diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c
index acb7c6d5643f..339c762eb6fd 100644
--- a/net/bluetooth/mgmt.c
+++ b/net/bluetooth/mgmt.c
@@ -106,6 +106,7 @@ static const u16 mgmt_commands[] = {
 	MGMT_OP_START_LIMITED_DISCOVERY,
 	MGMT_OP_READ_EXT_INFO,
 	MGMT_OP_SET_APPEARANCE,
+	MGMT_OP_SET_BLOCKED_KEYS,
 };
 
 static const u16 mgmt_events[] = {
@@ -2341,6 +2342,14 @@ static int load_link_keys(struct sock *sk, struct hci_dev *hdev, void *data,
 	for (i = 0; i < key_count; i++) {
 		struct mgmt_link_key_info *key = &cp->keys[i];
 
+		if (hci_is_blocked_key(hdev,
+				       HCI_BLOCKED_KEY_TYPE_LINKKEY,
+				       key->val)) {
+			bt_dev_warn(hdev, "Skipping blocked link key for %pMR",
+				    &key->addr.bdaddr);
+			continue;
+		}
+
 		/* Always ignore debug keys and require a new pairing if
 		 * the user wants to use them.
 		 */
@@ -3531,6 +3540,55 @@ unlock:
 	return err;
 }
 
+static int set_blocked_keys(struct sock *sk, struct hci_dev *hdev, void *data,
+			    u16 len)
+{
+	int err = MGMT_STATUS_SUCCESS;
+	struct mgmt_cp_set_blocked_keys *keys = data;
+	const u16 max_key_count = ((U16_MAX - sizeof(*keys)) /
+				   sizeof(struct mgmt_blocked_key_info));
+	u16 key_count, expected_len;
+	int i;
+
+	BT_DBG("request for %s", hdev->name);
+
+	key_count = __le16_to_cpu(keys->key_count);
+	if (key_count > max_key_count) {
+		bt_dev_err(hdev, "too big key_count value %u", key_count);
+		return mgmt_cmd_status(sk, hdev->id, MGMT_OP_SET_BLOCKED_KEYS,
+				       MGMT_STATUS_INVALID_PARAMS);
+	}
+
+	expected_len = struct_size(keys, keys, key_count);
+	if (expected_len != len) {
+		bt_dev_err(hdev, "expected %u bytes, got %u bytes",
+			   expected_len, len);
+		return mgmt_cmd_status(sk, hdev->id, MGMT_OP_SET_BLOCKED_KEYS,
+				       MGMT_STATUS_INVALID_PARAMS);
+	}
+
+	hci_dev_lock(hdev);
+
+	hci_blocked_keys_clear(hdev);
+
+	for (i = 0; i < keys->key_count; ++i) {
+		struct blocked_key *b = kzalloc(sizeof(*b), GFP_KERNEL);
+
+		if (!b) {
+			err = MGMT_STATUS_NO_RESOURCES;
+			break;
+		}
+
+		b->type = keys->keys[i].type;
+		memcpy(b->val, keys->keys[i].val, sizeof(b->val));
+		list_add_rcu(&b->list, &hdev->blocked_keys);
+	}
+	hci_dev_unlock(hdev);
+
+	return mgmt_cmd_complete(sk, hdev->id, MGMT_OP_SET_BLOCKED_KEYS,
+				err, NULL, 0);
+}
+
 static void read_local_oob_data_complete(struct hci_dev *hdev, u8 status,
 				         u16 opcode, struct sk_buff *skb)
 {
@@ -5051,6 +5109,14 @@ static int load_irks(struct sock *sk, struct hci_dev *hdev, void *cp_data,
 	for (i = 0; i < irk_count; i++) {
 		struct mgmt_irk_info *irk = &cp->irks[i];
 
+		if (hci_is_blocked_key(hdev,
+				       HCI_BLOCKED_KEY_TYPE_IRK,
+				       irk->val)) {
+			bt_dev_warn(hdev, "Skipping blocked IRK for %pMR",
+				    &irk->addr.bdaddr);
+			continue;
+		}
+
 		hci_add_irk(hdev, &irk->addr.bdaddr,
 			    le_addr_type(irk->addr.type), irk->val,
 			    BDADDR_ANY);
@@ -5134,6 +5200,14 @@ static int load_long_term_keys(struct sock *sk, struct hci_dev *hdev,
 		struct mgmt_ltk_info *key = &cp->keys[i];
 		u8 type, authenticated;
 
+		if (hci_is_blocked_key(hdev,
+				       HCI_BLOCKED_KEY_TYPE_LTK,
+				       key->val)) {
+			bt_dev_warn(hdev, "Skipping blocked LTK for %pMR",
+				    &key->addr.bdaddr);
+			continue;
+		}
+
 		switch (key->type) {
 		case MGMT_LTK_UNAUTHENTICATED:
 			authenticated = 0x00;
@@ -6914,6 +6988,8 @@ static const struct hci_mgmt_handler mgmt_handlers[] = {
 	{ set_appearance,	   MGMT_SET_APPEARANCE_SIZE },
 	{ get_phy_configuration,   MGMT_GET_PHY_CONFIGURATION_SIZE },
 	{ set_phy_configuration,   MGMT_SET_PHY_CONFIGURATION_SIZE },
+	{ set_blocked_keys,	   MGMT_OP_SET_BLOCKED_KEYS_SIZE,
+						HCI_MGMT_VAR_LEN },
 };
 
 void mgmt_index_added(struct hci_dev *hdev)
diff --git a/net/bluetooth/smp.c b/net/bluetooth/smp.c
index 6b42be4b5861..4ece170c518e 100644
--- a/net/bluetooth/smp.c
+++ b/net/bluetooth/smp.c
@@ -2453,6 +2453,15 @@ static int smp_cmd_encrypt_info(struct l2cap_conn *conn, struct sk_buff *skb)
 	if (skb->len < sizeof(*rp))
 		return SMP_INVALID_PARAMS;
 
+	/* Pairing is aborted if any blocked keys are distributed */
+	if (hci_is_blocked_key(conn->hcon->hdev, HCI_BLOCKED_KEY_TYPE_LTK,
+			       rp->ltk)) {
+		bt_dev_warn_ratelimited(conn->hcon->hdev,
+					"LTK blocked for %pMR",
+					&conn->hcon->dst);
+		return SMP_INVALID_PARAMS;
+	}
+
 	SMP_ALLOW_CMD(smp, SMP_CMD_MASTER_IDENT);
 
 	skb_pull(skb, sizeof(*rp));
@@ -2509,6 +2518,15 @@ static int smp_cmd_ident_info(struct l2cap_conn *conn, struct sk_buff *skb)
 	if (skb->len < sizeof(*info))
 		return SMP_INVALID_PARAMS;
 
+	/* Pairing is aborted if any blocked keys are distributed */
+	if (hci_is_blocked_key(conn->hcon->hdev, HCI_BLOCKED_KEY_TYPE_IRK,
+			       info->irk)) {
+		bt_dev_warn_ratelimited(conn->hcon->hdev,
+					"Identity key blocked for %pMR",
+					&conn->hcon->dst);
+		return SMP_INVALID_PARAMS;
+	}
+
 	SMP_ALLOW_CMD(smp, SMP_CMD_IDENT_ADDR_INFO);
 
 	skb_pull(skb, sizeof(*info));
-- 
cgit v1.2.3-59-g8ed1b


From 4de0fc599eb936d37542f819e931ba3fd8e435ca Mon Sep 17 00:00:00 2001
From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Date: Wed, 15 Jan 2020 13:02:11 -0800
Subject: Bluetooth: Add definitions for CIS connections

These adds the HCI definitions for handling CIS connections along with
ISO data packets.

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 include/net/bluetooth/hci.h | 159 +++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 158 insertions(+), 1 deletion(-)

diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h
index 07b6ecedc6ce..6293bdd7d862 100644
--- a/include/net/bluetooth/hci.h
+++ b/include/net/bluetooth/hci.h
@@ -27,6 +27,7 @@
 
 #define HCI_MAX_ACL_SIZE	1024
 #define HCI_MAX_SCO_SIZE	255
+#define HCI_MAX_ISO_SIZE	251
 #define HCI_MAX_EVENT_SIZE	260
 #define HCI_MAX_FRAME_SIZE	(HCI_MAX_ACL_SIZE + 4)
 
@@ -303,6 +304,7 @@ enum {
 #define HCI_ACLDATA_PKT		0x02
 #define HCI_SCODATA_PKT		0x03
 #define HCI_EVENT_PKT		0x04
+#define HCI_ISODATA_PKT		0x05
 #define HCI_DIAG_PKT		0xf0
 #define HCI_VENDOR_PKT		0xff
 
@@ -352,6 +354,15 @@ enum {
 #define ACL_ACTIVE_BCAST	0x04
 #define ACL_PICO_BCAST		0x08
 
+/* ISO PB flags */
+#define ISO_START		0x00
+#define ISO_CONT		0x01
+#define ISO_SINGLE		0x02
+#define ISO_END			0x03
+
+/* ISO TS flags */
+#define ISO_TS			0x01
+
 /* Baseband links */
 #define SCO_LINK	0x00
 #define ACL_LINK	0x01
@@ -359,6 +370,7 @@ enum {
 /* Low Energy links do not have defined link type. Use invented one */
 #define LE_LINK		0x80
 #define AMP_LINK	0x81
+#define ISO_LINK	0x82
 #define INVALID_LINK	0xff
 
 /* LMP features */
@@ -440,6 +452,8 @@ enum {
 #define HCI_LE_PHY_2M			0x01
 #define HCI_LE_PHY_CODED		0x08
 #define HCI_LE_CHAN_SEL_ALG2		0x40
+#define HCI_LE_CIS_MASTER		0x10
+#define HCI_LE_CIS_SLAVE		0x20
 
 /* Connection modes */
 #define HCI_CM_ACTIVE	0x0000
@@ -1718,6 +1732,86 @@ struct hci_cp_le_set_adv_set_rand_addr {
 	bdaddr_t  bdaddr;
 } __packed;
 
+#define HCI_OP_LE_READ_BUFFER_SIZE_V2	0x2060
+struct hci_rp_le_read_buffer_size_v2 {
+	__u8    status;
+	__le16  acl_mtu;
+	__u8    acl_max_pkt;
+	__le16  iso_mtu;
+	__u8    iso_max_pkt;
+} __packed;
+
+#define HCI_OP_LE_READ_ISO_TX_SYNC		0x2061
+struct hci_cp_le_read_iso_tx_sync {
+	__le16  handle;
+} __packed;
+
+struct hci_rp_le_read_iso_tx_sync {
+	__u8    status;
+	__le16  handle;
+	__le16  seq;
+	__le32  imestamp;
+	__u8    offset[3];
+} __packed;
+
+#define HCI_OP_LE_SET_CIG_PARAMS		0x2062
+struct hci_cis_params {
+	__u8    cis_id;
+	__le16  m_sdu;
+	__le16  s_sdu;
+	__u8    m_phy;
+	__u8    s_phy;
+	__u8    m_rtn;
+	__u8    s_rtn;
+} __packed;
+
+struct hci_cp_le_set_cig_params {
+	__u8    cig_id;
+	__u8    m_interval[3];
+	__u8    s_interval[3];
+	__u8    sca;
+	__u8    packing;
+	__u8    framing;
+	__le16  m_latency;
+	__le16  s_latency;
+	__u8    num_cis;
+	struct hci_cis_params cis[0];
+} __packed;
+
+struct hci_rp_le_set_cig_params {
+	__u8    status;
+	__u8    cig_id;
+	__u8    num_handles;
+	__le16  handle[0];
+} __packed;
+
+#define HCI_OP_LE_CREATE_CIS			0x2064
+struct hci_cis {
+	__le16  cis_handle;
+	__le16  acl_handle;
+} __packed;
+
+struct hci_cp_le_create_cis {
+	__u8    num_cis;
+	struct hci_cis cis[0];
+} __packed;
+
+#define HCI_OP_LE_REMOVE_CIG			0x2065
+struct hci_cp_le_remove_cig {
+	__u8    cig_id;
+} __packed;
+
+#define HCI_OP_LE_ACCEPT_CIS			0x2066
+struct hci_cp_le_accept_cis {
+	__le16  handle;
+} __packed;
+
+#define HCI_OP_LE_REJECT_CIS			0x2067
+struct hci_cp_le_reject_cis {
+	__le16  handle;
+	__u8    reason;
+} __packed;
+
 /* ---- HCI Events ---- */
 #define HCI_EV_INQUIRY_COMPLETE		0x01
 
@@ -2189,7 +2283,7 @@ struct hci_ev_le_direct_adv_info {
 #define HCI_EV_LE_PHY_UPDATE_COMPLETE	0x0c
 struct hci_ev_le_phy_update_complete {
 	__u8  status;
-	__u16 handle;
+	__le16 handle;
 	__u8  tx_phy;
 	__u8  rx_phy;
 } __packed;
@@ -2234,6 +2328,34 @@ struct hci_evt_le_ext_adv_set_term {
 	__u8	num_evts;
 } __packed;
 
+#define HCI_EVT_LE_CIS_ESTABLISHED	0x19
+struct hci_evt_le_cis_established {
+	__u8  status;
+	__le16 handle;
+	__u8  cig_sync_delay[3];
+	__u8  cis_sync_delay[3];
+	__u8  m_latency[3];
+	__u8  s_latency[3];
+	__u8  m_phy;
+	__u8  s_phy;
+	__u8  nse;
+	__u8  m_bn;
+	__u8  s_bn;
+	__u8  m_ft;
+	__u8  s_ft;
+	__le16 m_mtu;
+	__le16 s_mtu;
+	__le16 interval;
+} __packed;
+
+#define HCI_EVT_LE_CIS_REQ		0x1a
+struct hci_evt_le_cis_req {
+	__le16 acl_handle;
+	__le16 cis_handle;
+	__u8  cig_id;
+	__u8  cis_id;
+} __packed;
+
 #define HCI_EV_VENDOR			0xff
 
 /* Internal events generated by Bluetooth stack */
@@ -2262,6 +2384,7 @@ struct hci_ev_si_security {
 #define HCI_EVENT_HDR_SIZE   2
 #define HCI_ACL_HDR_SIZE     4
 #define HCI_SCO_HDR_SIZE     3
+#define HCI_ISO_HDR_SIZE     4
 
 struct hci_command_hdr {
 	__le16	opcode;		/* OCF & OGF */
@@ -2283,6 +2406,30 @@ struct hci_sco_hdr {
 	__u8	dlen;
 } __packed;
 
+struct hci_iso_hdr {
+	__le16	handle;
+	__le16	dlen;
+	__u8	data[0];
+} __packed;
+
+/* ISO data packet status flags */
+#define HCI_ISO_STATUS_VALID	0x00
+#define HCI_ISO_STATUS_INVALID	0x01
+#define HCI_ISO_STATUS_NOP	0x02
+
+#define HCI_ISO_DATA_HDR_SIZE	4
+struct hci_iso_data_hdr {
+	__le16	sn;
+	__le16	slen;
+};
+
+#define HCI_ISO_TS_DATA_HDR_SIZE 8
+struct hci_iso_ts_data_hdr {
+	__le32	ts;
+	__le16	sn;
+	__le16	slen;
+};
+
 static inline struct hci_event_hdr *hci_event_hdr(const struct sk_buff *skb)
 {
 	return (struct hci_event_hdr *) skb->data;
@@ -2308,4 +2455,14 @@ static inline struct hci_sco_hdr *hci_sco_hdr(const struct sk_buff *skb)
 #define hci_handle(h)		(h & 0x0fff)
 #define hci_flags(h)		(h >> 12)
 
+/* ISO handle and flags pack/unpack */
+#define hci_iso_flags_pb(f)		(f & 0x0003)
+#define hci_iso_flags_ts(f)		((f >> 2) & 0x0001)
+#define hci_iso_flags_pack(pb, ts)	((pb & 0x03) | ((ts & 0x01) << 2))
+
+/* ISO data length and flags pack/unpack */
+#define hci_iso_data_len_pack(h, f)	((__u16) ((h) | ((f) << 14)))
+#define hci_iso_data_len(h)		((h) & 0x3fff)
+#define hci_iso_data_flags(h)		((h) >> 14)
+
 #endif /* __HCI_H */
-- 
cgit v1.2.3-59-g8ed1b


From f92a8cb569e39e28515e3f9c0ccaa16f874644b7 Mon Sep 17 00:00:00 2001
From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Date: Wed, 15 Jan 2020 13:02:16 -0800
Subject: Bluetooth: hci_vhci: Add support for ISO packets

This make virtual controllers to pass ISO packets around.

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 drivers/bluetooth/hci_vhci.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/bluetooth/hci_vhci.c b/drivers/bluetooth/hci_vhci.c
index 65e41c1d760f..8ab26dec5f6e 100644
--- a/drivers/bluetooth/hci_vhci.c
+++ b/drivers/bluetooth/hci_vhci.c
@@ -178,6 +178,7 @@ static inline ssize_t vhci_get_user(struct vhci_data *data,
 	case HCI_EVENT_PKT:
 	case HCI_ACLDATA_PKT:
 	case HCI_SCODATA_PKT:
+	case HCI_ISODATA_PKT:
 		if (!data->hdev) {
 			kfree_skb(skb);
 			return -ENODEV;
-- 
cgit v1.2.3-59-g8ed1b


From f9a619db7c137b7c2dec0414d8deb8ec762ae8f9 Mon Sep 17 00:00:00 2001
From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Date: Wed, 15 Jan 2020 13:02:17 -0800
Subject: Bluetooth: monitor: Add support for ISO packets

This enables passing ISO packets to the monitor socket.

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 include/net/bluetooth/hci_mon.h | 2 ++
 net/bluetooth/hci_sock.c        | 6 ++++++
 2 files changed, 8 insertions(+)

diff --git a/include/net/bluetooth/hci_mon.h b/include/net/bluetooth/hci_mon.h
index 240786b04a46..2d5fcda1bcd0 100644
--- a/include/net/bluetooth/hci_mon.h
+++ b/include/net/bluetooth/hci_mon.h
@@ -49,6 +49,8 @@ struct hci_mon_hdr {
 #define HCI_MON_CTRL_CLOSE	15
 #define HCI_MON_CTRL_COMMAND	16
 #define HCI_MON_CTRL_EVENT	17
+#define HCI_MON_ISO_TX_PKT	18
+#define HCI_MON_ISO_RX_PKT	19
 
 struct hci_mon_new_index {
 	__u8		type;
diff --git a/net/bluetooth/hci_sock.c b/net/bluetooth/hci_sock.c
index 5d0ed28c0d3a..3ae508674ef7 100644
--- a/net/bluetooth/hci_sock.c
+++ b/net/bluetooth/hci_sock.c
@@ -324,6 +324,12 @@ void hci_send_to_monitor(struct hci_dev *hdev, struct sk_buff *skb)
 		else
 			opcode = cpu_to_le16(HCI_MON_SCO_TX_PKT);
 		break;
+	case HCI_ISODATA_PKT:
+		if (bt_cb(skb)->incoming)
+			opcode = cpu_to_le16(HCI_MON_ISO_RX_PKT);
+		else
+			opcode = cpu_to_le16(HCI_MON_ISO_TX_PKT);
+		break;
 	case HCI_DIAG_PKT:
 		opcode = cpu_to_le16(HCI_MON_VENDOR_DIAG);
 		break;
-- 
cgit v1.2.3-59-g8ed1b


From 1b1d29e5149990e44634b2e681de71effd463591 Mon Sep 17 00:00:00 2001
From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Date: Wed, 15 Jan 2020 13:02:18 -0800
Subject: Bluetooth: Make use of __check_timeout on hci_sched_le

This reuse __check_timeout on hci_sched_le following the same logic
used hci_sched_acl.

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 net/bluetooth/hci_core.c | 11 +++--------
 1 file changed, 3 insertions(+), 8 deletions(-)

diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c
index f0298db26dc3..1ca7508b6ca7 100644
--- a/net/bluetooth/hci_core.c
+++ b/net/bluetooth/hci_core.c
@@ -4287,15 +4287,10 @@ static void hci_sched_le(struct hci_dev *hdev)
 	if (!hci_conn_num(hdev, LE_LINK))
 		return;
 
-	if (!hci_dev_test_flag(hdev, HCI_UNCONFIGURED)) {
-		/* LE tx timeout must be longer than maximum
-		 * link supervision timeout (40.9 seconds) */
-		if (!hdev->le_cnt && hdev->le_pkts &&
-		    time_after(jiffies, hdev->le_last_tx + HZ * 45))
-			hci_link_tx_to(hdev, LE_LINK);
-	}
-
 	cnt = hdev->le_pkts ? hdev->le_cnt : hdev->acl_cnt;
+
+	__check_timeout(hdev, cnt);
+
 	tmp = cnt;
 	while (cnt && (chan = hci_chan_sent(hdev, LE_LINK, &quote))) {
 		u32 priority = (skb_peek(&chan->data_q))->priority;
-- 
cgit v1.2.3-59-g8ed1b


From ef564119ba832f55337935038dc0a91baa7417d1 Mon Sep 17 00:00:00 2001
From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Date: Wed, 15 Jan 2020 13:02:19 -0800
Subject: Bluetooth: hci_h4: Add support for ISO packets

This enables H4 driver to properly handle ISO packets.

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 drivers/bluetooth/hci_h4.c   | 1 +
 drivers/bluetooth/hci_uart.h | 7 +++++++
 2 files changed, 8 insertions(+)

diff --git a/drivers/bluetooth/hci_h4.c b/drivers/bluetooth/hci_h4.c
index 19ba52005009..6dc1fbeb564b 100644
--- a/drivers/bluetooth/hci_h4.c
+++ b/drivers/bluetooth/hci_h4.c
@@ -103,6 +103,7 @@ static const struct h4_recv_pkt h4_recv_pkts[] = {
 	{ H4_RECV_ACL,   .recv = hci_recv_frame },
 	{ H4_RECV_SCO,   .recv = hci_recv_frame },
 	{ H4_RECV_EVENT, .recv = hci_recv_frame },
+	{ H4_RECV_ISO,   .recv = hci_recv_frame },
 };
 
 /* Recv data */
diff --git a/drivers/bluetooth/hci_uart.h b/drivers/bluetooth/hci_uart.h
index 6ab631101019..4e039d7a16f8 100644
--- a/drivers/bluetooth/hci_uart.h
+++ b/drivers/bluetooth/hci_uart.h
@@ -143,6 +143,13 @@ struct h4_recv_pkt {
 	.lsize = 1, \
 	.maxlen = HCI_MAX_EVENT_SIZE
 
+#define H4_RECV_ISO \
+	.type = HCI_ISODATA_PKT, \
+	.hlen = HCI_ISO_HDR_SIZE, \
+	.loff = 2, \
+	.lsize = 2, \
+	.maxlen = HCI_MAX_FRAME_SIZE \
+
 struct sk_buff *h4_recv_buf(struct hci_dev *hdev, struct sk_buff *skb,
 			    const unsigned char *buffer, int count,
 			    const struct h4_recv_pkt *pkts, int pkts_count);
-- 
cgit v1.2.3-59-g8ed1b


From 1cc3c10c5aea84d4b0400423449c316eed3f27df Mon Sep 17 00:00:00 2001
From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Date: Wed, 15 Jan 2020 13:02:20 -0800
Subject: Bluetooth: hci_h5: Add support for ISO packets

This enables H5 driver to properly handle ISO packets.

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 drivers/bluetooth/hci_h5.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/bluetooth/hci_h5.c b/drivers/bluetooth/hci_h5.c
index dacf297baf59..0b14547482a7 100644
--- a/drivers/bluetooth/hci_h5.c
+++ b/drivers/bluetooth/hci_h5.c
@@ -385,6 +385,7 @@ static void h5_complete_rx_pkt(struct hci_uart *hu)
 	case HCI_EVENT_PKT:
 	case HCI_ACLDATA_PKT:
 	case HCI_SCODATA_PKT:
+	case HCI_ISODATA_PKT:
 		hci_skb_pkt_type(h5->rx_skb) = H5_HDR_PKT_TYPE(hdr);
 
 		/* Remove Three-wire header */
@@ -594,6 +595,7 @@ static int h5_enqueue(struct hci_uart *hu, struct sk_buff *skb)
 		break;
 
 	case HCI_SCODATA_PKT:
+	case HCI_ISODATA_PKT:
 		skb_queue_tail(&h5->unrel, skb);
 		break;
 
@@ -636,6 +638,7 @@ static bool valid_packet_type(u8 type)
 	case HCI_ACLDATA_PKT:
 	case HCI_COMMAND_PKT:
 	case HCI_SCODATA_PKT:
+	case HCI_ISODATA_PKT:
 	case HCI_3WIRE_LINK_PKT:
 	case HCI_3WIRE_ACK_PKT:
 		return true;
-- 
cgit v1.2.3-59-g8ed1b


From 7e8aeffb112aa1d1600ce0ee44e444b63e3d41f9 Mon Sep 17 00:00:00 2001
From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Date: Wed, 15 Jan 2020 13:02:21 -0800
Subject: Bluetooth: btsdio: Check for valid packet type

Check for valid packet type before calling hci_recv_frame which is
inline with what other drivers are doing.

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 drivers/bluetooth/btsdio.c | 19 ++++++++++++++-----
 1 file changed, 14 insertions(+), 5 deletions(-)

diff --git a/drivers/bluetooth/btsdio.c b/drivers/bluetooth/btsdio.c
index fd9571d5fdac..199e8f7d426d 100644
--- a/drivers/bluetooth/btsdio.c
+++ b/drivers/bluetooth/btsdio.c
@@ -145,11 +145,20 @@ static int btsdio_rx_packet(struct btsdio_data *data)
 
 	data->hdev->stat.byte_rx += len;
 
-	hci_skb_pkt_type(skb) = hdr[3];
-
-	err = hci_recv_frame(data->hdev, skb);
-	if (err < 0)
-		return err;
+	switch (hdr[3]) {
+	case HCI_EVENT_PKT:
+	case HCI_ACLDATA_PKT:
+	case HCI_SCODATA_PKT:
+	case HCI_ISODATA_PKT:
+		hci_skb_pkt_type(skb) = hdr[3];
+		err = hci_recv_frame(data->hdev, skb);
+		if (err < 0)
+			return err;
+		break;
+	default:
+		kfree_skb(skb);
+		return -EINVAL;
+	}
 
 	sdio_writeb(data->func, 0x00, REG_PC_RRT, NULL);
 
-- 
cgit v1.2.3-59-g8ed1b


From 2fe465e69dc13f7b6bf8248d08bd16844c1ce545 Mon Sep 17 00:00:00 2001
From: Guillaume La Roque <glaroque@baylibre.com>
Date: Wed, 15 Jan 2020 11:12:42 +0100
Subject: dt-bindings: net: bluetooth: add interrupts properties

add interrupts and interrupt-names as optional properties
to support host-wakeup by interrupt properties instead of
host-wakeup-gpios.

Signed-off-by: Guillaume La Roque <glaroque@baylibre.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 Documentation/devicetree/bindings/net/broadcom-bluetooth.txt | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/Documentation/devicetree/bindings/net/broadcom-bluetooth.txt b/Documentation/devicetree/bindings/net/broadcom-bluetooth.txt
index c44a30dbe43d..dd258674633c 100644
--- a/Documentation/devicetree/bindings/net/broadcom-bluetooth.txt
+++ b/Documentation/devicetree/bindings/net/broadcom-bluetooth.txt
@@ -23,7 +23,9 @@ Optional properties:
  - max-speed: see Documentation/devicetree/bindings/serial/slave-device.txt
  - shutdown-gpios: GPIO specifier, used to enable the BT module
  - device-wakeup-gpios: GPIO specifier, used to wakeup the controller
- - host-wakeup-gpios: GPIO specifier, used to wakeup the host processor
+ - host-wakeup-gpios: GPIO specifier, used to wakeup the host processor.
+                      deprecated, replaced by interrupts and
+                      "host-wakeup" interrupt-names
  - clocks: 1 or 2 clocks as defined in clock-names below, in that order
  - clock-names: names for clock inputs, matching the clocks given
    - "extclk": deprecated, replaced by "txco"
@@ -37,7 +39,8 @@ Optional properties:
     - pcm-frame-type: short, long
     - pcm-sync-mode: slave, master
     - pcm-clock-mode: slave, master
-
+ - interrupts: must be one, used to wakeup the host processor
+ - interrupt-names: must be "host-wakeup"
 
 Example:
 
-- 
cgit v1.2.3-59-g8ed1b


From f25a96c8eb46b86744540ade61c2dbc844c23dbd Mon Sep 17 00:00:00 2001
From: Guillaume La Roque <glaroque@baylibre.com>
Date: Wed, 15 Jan 2020 11:12:43 +0100
Subject: Bluetooth: hci_bcm: enable IRQ capability from devicetree

Add support for getting IRQ directly from DT instead of relying on
converting a GPIO to IRQ. This is needed for platforms with GPIO
controllers that that do not support gpiod_to_irq().

Reviewed-by: Kevin Hilman <khilman@baylibre.com>
Signed-off-by: Guillaume La Roque <glaroque@baylibre.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 drivers/bluetooth/hci_bcm.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/bluetooth/hci_bcm.c b/drivers/bluetooth/hci_bcm.c
index 769bb4404bd1..b236cb11c0dc 100644
--- a/drivers/bluetooth/hci_bcm.c
+++ b/drivers/bluetooth/hci_bcm.c
@@ -13,6 +13,7 @@
 #include <linux/module.h>
 #include <linux/acpi.h>
 #include <linux/of.h>
+#include <linux/of_irq.h>
 #include <linux/property.h>
 #include <linux/platform_data/x86/apple.h>
 #include <linux/platform_device.h>
@@ -1151,6 +1152,8 @@ static int bcm_of_probe(struct bcm_device *bdev)
 	device_property_read_u32(bdev->dev, "max-speed", &bdev->oper_speed);
 	device_property_read_u8_array(bdev->dev, "brcm,bt-pcm-int-params",
 				      bdev->pcm_int_params, 5);
+	bdev->irq = of_irq_get_byname(bdev->dev->of_node, "host-wakeup");
+
 	return 0;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 5559904ccc0867a0ce796761681e40defe4a5f44 Mon Sep 17 00:00:00 2001
From: Rocky Liao <rjliao@codeaurora.org>
Date: Wed, 15 Jan 2020 16:55:50 +0800
Subject: Bluetooth: hci_qca: Add QCA Rome power off support to the
 qca_power_shutdown()

Current qca_power_shutdown() only supports wcn399x, this patch adds Rome
power off support to it. For Rome it just needs to pull down the bt_en
GPIO to power off it. This patch also replaces all the power off operation
in qca_close() with the unified qca_power_shutdown() call.

Signed-off-by: Rocky Liao <rjliao@codeaurora.org>
Reviewed-by: Matthias Kaehlcke <mka@chromium.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 drivers/bluetooth/hci_qca.c | 28 ++++++++++++++++------------
 1 file changed, 16 insertions(+), 12 deletions(-)

diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c
index 992622dc1263..ecb74965be10 100644
--- a/drivers/bluetooth/hci_qca.c
+++ b/drivers/bluetooth/hci_qca.c
@@ -663,7 +663,6 @@ static int qca_flush(struct hci_uart *hu)
 /* Close protocol */
 static int qca_close(struct hci_uart *hu)
 {
-	struct qca_serdev *qcadev;
 	struct qca_data *qca = hu->priv;
 
 	BT_DBG("hu %p qca close", hu);
@@ -679,14 +678,7 @@ static int qca_close(struct hci_uart *hu)
 	destroy_workqueue(qca->workqueue);
 	qca->hu = NULL;
 
-	if (hu->serdev) {
-		qcadev = serdev_device_get_drvdata(hu->serdev);
-		if (qca_is_wcn399x(qcadev->btsoc_type))
-			qca_power_shutdown(hu);
-		else
-			gpiod_set_value_cansleep(qcadev->bt_en, 0);
-
-	}
+	qca_power_shutdown(hu);
 
 	kfree_skb(qca->rx_skb);
 
@@ -1685,6 +1677,7 @@ static void qca_power_shutdown(struct hci_uart *hu)
 	struct qca_serdev *qcadev;
 	struct qca_data *qca = hu->priv;
 	unsigned long flags;
+	enum qca_btsoc_type soc_type = qca_soc_type(hu);
 
 	qcadev = serdev_device_get_drvdata(hu->serdev);
 
@@ -1697,11 +1690,22 @@ static void qca_power_shutdown(struct hci_uart *hu)
 	qca_flush(hu);
 	spin_unlock_irqrestore(&qca->hci_ibs_lock, flags);
 
-	host_set_baudrate(hu, 2400);
-	qca_send_power_pulse(hu, false);
-	qca_regulator_disable(qcadev);
 	hu->hdev->hw_error = NULL;
 	hu->hdev->cmd_timeout = NULL;
+
+	/* Non-serdev device usually is powered by external power
+	 * and don't need additional action in driver for power down
+	 */
+	if (!hu->serdev)
+		return;
+
+	if (qca_is_wcn399x(soc_type)) {
+		host_set_baudrate(hu, 2400);
+		qca_send_power_pulse(hu, false);
+		qca_regulator_disable(qcadev);
+	} else {
+		gpiod_set_value_cansleep(qcadev->bt_en, 0);
+	}
 }
 
 static int qca_power_off(struct hci_dev *hdev)
-- 
cgit v1.2.3-59-g8ed1b


From bb2500ab0270aa0d1d39614f08f4f9a2dc0df8c1 Mon Sep 17 00:00:00 2001
From: Rocky Liao <rjliao@codeaurora.org>
Date: Wed, 15 Jan 2020 16:55:51 +0800
Subject: Bluetooth: hci_qca: Retry btsoc initialize when it fails

This patch adds the retry of btsoc initialization when it fails. There are
reports that the btsoc initialization may fail on some platforms but the
repro ratio is very low. The symptoms is the firmware downloading failed
due to the UART write timed out. The failure may be caused by UART,
platform HW or the btsoc itself but it's very difficlut to root cause,
given the repro ratio is very low. Add a retry for the btsoc initialization
can work around most of the failures and make Bluetooth finally works.

Signed-off-by: Rocky Liao <rjliao@codeaurora.org>
Reviewed-by: Matthias Kaehlcke <mka@chromium.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 drivers/bluetooth/hci_qca.c | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c
index ecb74965be10..1139142e8eed 100644
--- a/drivers/bluetooth/hci_qca.c
+++ b/drivers/bluetooth/hci_qca.c
@@ -55,6 +55,9 @@
 /* Controller debug log header */
 #define QCA_DEBUG_HANDLE	0x2EDC
 
+/* max retry count when init fails */
+#define MAX_INIT_RETRIES 3
+
 /* Controller dump header */
 #define QCA_SSR_DUMP_HANDLE		0x0108
 #define QCA_DUMP_PACKET_SIZE		255
@@ -1539,6 +1542,7 @@ static int qca_setup(struct hci_uart *hu)
 	struct hci_dev *hdev = hu->hdev;
 	struct qca_data *qca = hu->priv;
 	unsigned int speed, qca_baudrate = QCA_BAUDRATE_115200;
+	unsigned int retries = 0;
 	enum qca_btsoc_type soc_type = qca_soc_type(hu);
 	const char *firmware_name = qca_get_firmware_name(hu);
 	int ret;
@@ -1559,6 +1563,7 @@ static int qca_setup(struct hci_uart *hu)
 	bt_dev_info(hdev, "setting up %s",
 		qca_is_wcn399x(soc_type) ? "wcn399x" : "ROME");
 
+retry:
 	ret = qca_power_on(hdev);
 	if (ret)
 		return ret;
@@ -1613,6 +1618,20 @@ static int qca_setup(struct hci_uart *hu)
 		 * patch/nvm-config is found, so run with original fw/config.
 		 */
 		ret = 0;
+	} else {
+		if (retries < MAX_INIT_RETRIES) {
+			qca_power_shutdown(hu);
+			if (hu->serdev) {
+				serdev_device_close(hu->serdev);
+				ret = serdev_device_open(hu->serdev);
+				if (ret) {
+					bt_dev_err(hdev, "failed to open port");
+					return ret;
+				}
+			}
+			retries++;
+			goto retry;
+		}
 	}
 
 	/* Setup bdaddr */
-- 
cgit v1.2.3-59-g8ed1b


From 15c14a3dca421f086c187155afba3222b879472d Mon Sep 17 00:00:00 2001
From: Brian Vazquez <brianvv@google.com>
Date: Wed, 15 Jan 2020 10:43:00 -0800
Subject: bpf: Add bpf_map_{value_size, update_value, map_copy_value} functions

This commit moves reusable code from map_lookup_elem and map_update_elem
to avoid code duplication in kernel/bpf/syscall.c.

Signed-off-by: Brian Vazquez <brianvv@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20200115184308.162644-2-brianvv@google.com
---
 kernel/bpf/syscall.c | 280 ++++++++++++++++++++++++++++-----------------------
 1 file changed, 152 insertions(+), 128 deletions(-)

diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index f9db72a96ec0..08b0b6e40454 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -129,6 +129,154 @@ static struct bpf_map *find_and_alloc_map(union bpf_attr *attr)
 	return map;
 }
 
+static u32 bpf_map_value_size(struct bpf_map *map)
+{
+	if (map->map_type == BPF_MAP_TYPE_PERCPU_HASH ||
+	    map->map_type == BPF_MAP_TYPE_LRU_PERCPU_HASH ||
+	    map->map_type == BPF_MAP_TYPE_PERCPU_ARRAY ||
+	    map->map_type == BPF_MAP_TYPE_PERCPU_CGROUP_STORAGE)
+		return round_up(map->value_size, 8) * num_possible_cpus();
+	else if (IS_FD_MAP(map))
+		return sizeof(u32);
+	else
+		return  map->value_size;
+}
+
+static void maybe_wait_bpf_programs(struct bpf_map *map)
+{
+	/* Wait for any running BPF programs to complete so that
+	 * userspace, when we return to it, knows that all programs
+	 * that could be running use the new map value.
+	 */
+	if (map->map_type == BPF_MAP_TYPE_HASH_OF_MAPS ||
+	    map->map_type == BPF_MAP_TYPE_ARRAY_OF_MAPS)
+		synchronize_rcu();
+}
+
+static int bpf_map_update_value(struct bpf_map *map, struct fd f, void *key,
+				void *value, __u64 flags)
+{
+	int err;
+
+	/* Need to create a kthread, thus must support schedule */
+	if (bpf_map_is_dev_bound(map)) {
+		return bpf_map_offload_update_elem(map, key, value, flags);
+	} else if (map->map_type == BPF_MAP_TYPE_CPUMAP ||
+		   map->map_type == BPF_MAP_TYPE_SOCKHASH ||
+		   map->map_type == BPF_MAP_TYPE_SOCKMAP ||
+		   map->map_type == BPF_MAP_TYPE_STRUCT_OPS) {
+		return map->ops->map_update_elem(map, key, value, flags);
+	} else if (IS_FD_PROG_ARRAY(map)) {
+		return bpf_fd_array_map_update_elem(map, f.file, key, value,
+						    flags);
+	}
+
+	/* must increment bpf_prog_active to avoid kprobe+bpf triggering from
+	 * inside bpf map update or delete otherwise deadlocks are possible
+	 */
+	preempt_disable();
+	__this_cpu_inc(bpf_prog_active);
+	if (map->map_type == BPF_MAP_TYPE_PERCPU_HASH ||
+	    map->map_type == BPF_MAP_TYPE_LRU_PERCPU_HASH) {
+		err = bpf_percpu_hash_update(map, key, value, flags);
+	} else if (map->map_type == BPF_MAP_TYPE_PERCPU_ARRAY) {
+		err = bpf_percpu_array_update(map, key, value, flags);
+	} else if (map->map_type == BPF_MAP_TYPE_PERCPU_CGROUP_STORAGE) {
+		err = bpf_percpu_cgroup_storage_update(map, key, value,
+						       flags);
+	} else if (IS_FD_ARRAY(map)) {
+		rcu_read_lock();
+		err = bpf_fd_array_map_update_elem(map, f.file, key, value,
+						   flags);
+		rcu_read_unlock();
+	} else if (map->map_type == BPF_MAP_TYPE_HASH_OF_MAPS) {
+		rcu_read_lock();
+		err = bpf_fd_htab_map_update_elem(map, f.file, key, value,
+						  flags);
+		rcu_read_unlock();
+	} else if (map->map_type == BPF_MAP_TYPE_REUSEPORT_SOCKARRAY) {
+		/* rcu_read_lock() is not needed */
+		err = bpf_fd_reuseport_array_update_elem(map, key, value,
+							 flags);
+	} else if (map->map_type == BPF_MAP_TYPE_QUEUE ||
+		   map->map_type == BPF_MAP_TYPE_STACK) {
+		err = map->ops->map_push_elem(map, value, flags);
+	} else {
+		rcu_read_lock();
+		err = map->ops->map_update_elem(map, key, value, flags);
+		rcu_read_unlock();
+	}
+	__this_cpu_dec(bpf_prog_active);
+	preempt_enable();
+	maybe_wait_bpf_programs(map);
+
+	return err;
+}
+
+static int bpf_map_copy_value(struct bpf_map *map, void *key, void *value,
+			      __u64 flags)
+{
+	void *ptr;
+	int err;
+
+	if (bpf_map_is_dev_bound(map)) {
+		err =  bpf_map_offload_lookup_elem(map, key, value);
+		return err;
+	}
+
+	preempt_disable();
+	this_cpu_inc(bpf_prog_active);
+	if (map->map_type == BPF_MAP_TYPE_PERCPU_HASH ||
+	    map->map_type == BPF_MAP_TYPE_LRU_PERCPU_HASH) {
+		err = bpf_percpu_hash_copy(map, key, value);
+	} else if (map->map_type == BPF_MAP_TYPE_PERCPU_ARRAY) {
+		err = bpf_percpu_array_copy(map, key, value);
+	} else if (map->map_type == BPF_MAP_TYPE_PERCPU_CGROUP_STORAGE) {
+		err = bpf_percpu_cgroup_storage_copy(map, key, value);
+	} else if (map->map_type == BPF_MAP_TYPE_STACK_TRACE) {
+		err = bpf_stackmap_copy(map, key, value);
+	} else if (IS_FD_ARRAY(map) || IS_FD_PROG_ARRAY(map)) {
+		err = bpf_fd_array_map_lookup_elem(map, key, value);
+	} else if (IS_FD_HASH(map)) {
+		err = bpf_fd_htab_map_lookup_elem(map, key, value);
+	} else if (map->map_type == BPF_MAP_TYPE_REUSEPORT_SOCKARRAY) {
+		err = bpf_fd_reuseport_array_lookup_elem(map, key, value);
+	} else if (map->map_type == BPF_MAP_TYPE_QUEUE ||
+		   map->map_type == BPF_MAP_TYPE_STACK) {
+		err = map->ops->map_peek_elem(map, value);
+	} else if (map->map_type == BPF_MAP_TYPE_STRUCT_OPS) {
+		/* struct_ops map requires directly updating "value" */
+		err = bpf_struct_ops_map_sys_lookup_elem(map, key, value);
+	} else {
+		rcu_read_lock();
+		if (map->ops->map_lookup_elem_sys_only)
+			ptr = map->ops->map_lookup_elem_sys_only(map, key);
+		else
+			ptr = map->ops->map_lookup_elem(map, key);
+		if (IS_ERR(ptr)) {
+			err = PTR_ERR(ptr);
+		} else if (!ptr) {
+			err = -ENOENT;
+		} else {
+			err = 0;
+			if (flags & BPF_F_LOCK)
+				/* lock 'ptr' and copy everything but lock */
+				copy_map_value_locked(map, value, ptr, true);
+			else
+				copy_map_value(map, value, ptr);
+			/* mask lock, since value wasn't zero inited */
+			check_and_init_map_lock(map, value);
+		}
+		rcu_read_unlock();
+	}
+
+	this_cpu_dec(bpf_prog_active);
+	preempt_enable();
+	maybe_wait_bpf_programs(map);
+
+	return err;
+}
+
 static void *__bpf_map_area_alloc(u64 size, int numa_node, bool mmapable)
 {
 	/* We really just want to fail instead of triggering OOM killer
@@ -827,7 +975,7 @@ static int map_lookup_elem(union bpf_attr *attr)
 	void __user *uvalue = u64_to_user_ptr(attr->value);
 	int ufd = attr->map_fd;
 	struct bpf_map *map;
-	void *key, *value, *ptr;
+	void *key, *value;
 	u32 value_size;
 	struct fd f;
 	int err;
@@ -859,75 +1007,14 @@ static int map_lookup_elem(union bpf_attr *attr)
 		goto err_put;
 	}
 
-	if (map->map_type == BPF_MAP_TYPE_PERCPU_HASH ||
-	    map->map_type == BPF_MAP_TYPE_LRU_PERCPU_HASH ||
-	    map->map_type == BPF_MAP_TYPE_PERCPU_ARRAY ||
-	    map->map_type == BPF_MAP_TYPE_PERCPU_CGROUP_STORAGE)
-		value_size = round_up(map->value_size, 8) * num_possible_cpus();
-	else if (IS_FD_MAP(map))
-		value_size = sizeof(u32);
-	else
-		value_size = map->value_size;
+	value_size = bpf_map_value_size(map);
 
 	err = -ENOMEM;
 	value = kmalloc(value_size, GFP_USER | __GFP_NOWARN);
 	if (!value)
 		goto free_key;
 
-	if (bpf_map_is_dev_bound(map)) {
-		err = bpf_map_offload_lookup_elem(map, key, value);
-		goto done;
-	}
-
-	preempt_disable();
-	this_cpu_inc(bpf_prog_active);
-	if (map->map_type == BPF_MAP_TYPE_PERCPU_HASH ||
-	    map->map_type == BPF_MAP_TYPE_LRU_PERCPU_HASH) {
-		err = bpf_percpu_hash_copy(map, key, value);
-	} else if (map->map_type == BPF_MAP_TYPE_PERCPU_ARRAY) {
-		err = bpf_percpu_array_copy(map, key, value);
-	} else if (map->map_type == BPF_MAP_TYPE_PERCPU_CGROUP_STORAGE) {
-		err = bpf_percpu_cgroup_storage_copy(map, key, value);
-	} else if (map->map_type == BPF_MAP_TYPE_STACK_TRACE) {
-		err = bpf_stackmap_copy(map, key, value);
-	} else if (IS_FD_ARRAY(map) || IS_FD_PROG_ARRAY(map)) {
-		err = bpf_fd_array_map_lookup_elem(map, key, value);
-	} else if (IS_FD_HASH(map)) {
-		err = bpf_fd_htab_map_lookup_elem(map, key, value);
-	} else if (map->map_type == BPF_MAP_TYPE_REUSEPORT_SOCKARRAY) {
-		err = bpf_fd_reuseport_array_lookup_elem(map, key, value);
-	} else if (map->map_type == BPF_MAP_TYPE_QUEUE ||
-		   map->map_type == BPF_MAP_TYPE_STACK) {
-		err = map->ops->map_peek_elem(map, value);
-	} else if (map->map_type == BPF_MAP_TYPE_STRUCT_OPS) {
-		/* struct_ops map requires directly updating "value" */
-		err = bpf_struct_ops_map_sys_lookup_elem(map, key, value);
-	} else {
-		rcu_read_lock();
-		if (map->ops->map_lookup_elem_sys_only)
-			ptr = map->ops->map_lookup_elem_sys_only(map, key);
-		else
-			ptr = map->ops->map_lookup_elem(map, key);
-		if (IS_ERR(ptr)) {
-			err = PTR_ERR(ptr);
-		} else if (!ptr) {
-			err = -ENOENT;
-		} else {
-			err = 0;
-			if (attr->flags & BPF_F_LOCK)
-				/* lock 'ptr' and copy everything but lock */
-				copy_map_value_locked(map, value, ptr, true);
-			else
-				copy_map_value(map, value, ptr);
-			/* mask lock, since value wasn't zero inited */
-			check_and_init_map_lock(map, value);
-		}
-		rcu_read_unlock();
-	}
-	this_cpu_dec(bpf_prog_active);
-	preempt_enable();
-
-done:
+	err = bpf_map_copy_value(map, key, value, attr->flags);
 	if (err)
 		goto free_value;
 
@@ -946,16 +1033,6 @@ err_put:
 	return err;
 }
 
-static void maybe_wait_bpf_programs(struct bpf_map *map)
-{
-	/* Wait for any running BPF programs to complete so that
-	 * userspace, when we return to it, knows that all programs
-	 * that could be running use the new map value.
-	 */
-	if (map->map_type == BPF_MAP_TYPE_HASH_OF_MAPS ||
-	    map->map_type == BPF_MAP_TYPE_ARRAY_OF_MAPS)
-		synchronize_rcu();
-}
 
 #define BPF_MAP_UPDATE_ELEM_LAST_FIELD flags
 
@@ -1011,61 +1088,8 @@ static int map_update_elem(union bpf_attr *attr)
 	if (copy_from_user(value, uvalue, value_size) != 0)
 		goto free_value;
 
-	/* Need to create a kthread, thus must support schedule */
-	if (bpf_map_is_dev_bound(map)) {
-		err = bpf_map_offload_update_elem(map, key, value, attr->flags);
-		goto out;
-	} else if (map->map_type == BPF_MAP_TYPE_CPUMAP ||
-		   map->map_type == BPF_MAP_TYPE_SOCKHASH ||
-		   map->map_type == BPF_MAP_TYPE_SOCKMAP ||
-		   map->map_type == BPF_MAP_TYPE_STRUCT_OPS) {
-		err = map->ops->map_update_elem(map, key, value, attr->flags);
-		goto out;
-	} else if (IS_FD_PROG_ARRAY(map)) {
-		err = bpf_fd_array_map_update_elem(map, f.file, key, value,
-						   attr->flags);
-		goto out;
-	}
+	err = bpf_map_update_value(map, f, key, value, attr->flags);
 
-	/* must increment bpf_prog_active to avoid kprobe+bpf triggering from
-	 * inside bpf map update or delete otherwise deadlocks are possible
-	 */
-	preempt_disable();
-	__this_cpu_inc(bpf_prog_active);
-	if (map->map_type == BPF_MAP_TYPE_PERCPU_HASH ||
-	    map->map_type == BPF_MAP_TYPE_LRU_PERCPU_HASH) {
-		err = bpf_percpu_hash_update(map, key, value, attr->flags);
-	} else if (map->map_type == BPF_MAP_TYPE_PERCPU_ARRAY) {
-		err = bpf_percpu_array_update(map, key, value, attr->flags);
-	} else if (map->map_type == BPF_MAP_TYPE_PERCPU_CGROUP_STORAGE) {
-		err = bpf_percpu_cgroup_storage_update(map, key, value,
-						       attr->flags);
-	} else if (IS_FD_ARRAY(map)) {
-		rcu_read_lock();
-		err = bpf_fd_array_map_update_elem(map, f.file, key, value,
-						   attr->flags);
-		rcu_read_unlock();
-	} else if (map->map_type == BPF_MAP_TYPE_HASH_OF_MAPS) {
-		rcu_read_lock();
-		err = bpf_fd_htab_map_update_elem(map, f.file, key, value,
-						  attr->flags);
-		rcu_read_unlock();
-	} else if (map->map_type == BPF_MAP_TYPE_REUSEPORT_SOCKARRAY) {
-		/* rcu_read_lock() is not needed */
-		err = bpf_fd_reuseport_array_update_elem(map, key, value,
-							 attr->flags);
-	} else if (map->map_type == BPF_MAP_TYPE_QUEUE ||
-		   map->map_type == BPF_MAP_TYPE_STACK) {
-		err = map->ops->map_push_elem(map, value, attr->flags);
-	} else {
-		rcu_read_lock();
-		err = map->ops->map_update_elem(map, key, value, attr->flags);
-		rcu_read_unlock();
-	}
-	__this_cpu_dec(bpf_prog_active);
-	preempt_enable();
-	maybe_wait_bpf_programs(map);
-out:
 free_value:
 	kfree(value);
 free_key:
-- 
cgit v1.2.3-59-g8ed1b


From cb4d03ab499d4c040f4ab6fd4389d2b49f42b5a5 Mon Sep 17 00:00:00 2001
From: Brian Vazquez <brianvv@google.com>
Date: Wed, 15 Jan 2020 10:43:01 -0800
Subject: bpf: Add generic support for lookup batch op

This commit introduces generic support for the bpf_map_lookup_batch.
This implementation can be used by almost all the bpf maps since its core
implementation is relying on the existing map_get_next_key and
map_lookup_elem. The bpf syscall subcommand introduced is:

  BPF_MAP_LOOKUP_BATCH

The UAPI attribute is:

  struct { /* struct used by BPF_MAP_*_BATCH commands */
         __aligned_u64   in_batch;       /* start batch,
                                          * NULL to start from beginning
                                          */
         __aligned_u64   out_batch;      /* output: next start batch */
         __aligned_u64   keys;
         __aligned_u64   values;
         __u32           count;          /* input/output:
                                          * input: # of key/value
                                          * elements
                                          * output: # of filled elements
                                          */
         __u32           map_fd;
         __u64           elem_flags;
         __u64           flags;
  } batch;

in_batch/out_batch are opaque values use to communicate between
user/kernel space, in_batch/out_batch must be of key_size length.

To start iterating from the beginning in_batch must be null,
count is the # of key/value elements to retrieve. Note that the 'keys'
buffer must be a buffer of key_size * count size and the 'values' buffer
must be value_size * count, where value_size must be aligned to 8 bytes
by userspace if it's dealing with percpu maps. 'count' will contain the
number of keys/values successfully retrieved. Note that 'count' is an
input/output variable and it can contain a lower value after a call.

If there's no more entries to retrieve, ENOENT will be returned. If error
is ENOENT, count might be > 0 in case it copied some values but there were
no more entries to retrieve.

Note that if the return code is an error and not -EFAULT,
count indicates the number of elements successfully processed.

Suggested-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Brian Vazquez <brianvv@google.com>
Signed-off-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200115184308.162644-3-brianvv@google.com
---
 include/linux/bpf.h      |   5 ++
 include/uapi/linux/bpf.h |  18 ++++++
 kernel/bpf/syscall.c     | 160 +++++++++++++++++++++++++++++++++++++++++++++--
 3 files changed, 179 insertions(+), 4 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index aed2bc39d72b..807744ecaa5a 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -44,6 +44,8 @@ struct bpf_map_ops {
 	int (*map_get_next_key)(struct bpf_map *map, void *key, void *next_key);
 	void (*map_release_uref)(struct bpf_map *map);
 	void *(*map_lookup_elem_sys_only)(struct bpf_map *map, void *key);
+	int (*map_lookup_batch)(struct bpf_map *map, const union bpf_attr *attr,
+				union bpf_attr __user *uattr);
 
 	/* funcs callable from userspace and from eBPF programs */
 	void *(*map_lookup_elem)(struct bpf_map *map, void *key);
@@ -982,6 +984,9 @@ void *bpf_map_area_alloc(u64 size, int numa_node);
 void *bpf_map_area_mmapable_alloc(u64 size, int numa_node);
 void bpf_map_area_free(void *base);
 void bpf_map_init_from_attr(struct bpf_map *map, union bpf_attr *attr);
+int  generic_map_lookup_batch(struct bpf_map *map,
+			      const union bpf_attr *attr,
+			      union bpf_attr __user *uattr);
 
 extern int sysctl_unprivileged_bpf_disabled;
 
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index a88837d4dd18..0721b2c6bb8c 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -107,6 +107,7 @@ enum bpf_cmd {
 	BPF_MAP_LOOKUP_AND_DELETE_ELEM,
 	BPF_MAP_FREEZE,
 	BPF_BTF_GET_NEXT_ID,
+	BPF_MAP_LOOKUP_BATCH,
 };
 
 enum bpf_map_type {
@@ -420,6 +421,23 @@ union bpf_attr {
 		__u64		flags;
 	};
 
+	struct { /* struct used by BPF_MAP_*_BATCH commands */
+		__aligned_u64	in_batch;	/* start batch,
+						 * NULL to start from beginning
+						 */
+		__aligned_u64	out_batch;	/* output: next start batch */
+		__aligned_u64	keys;
+		__aligned_u64	values;
+		__u32		count;		/* input/output:
+						 * input: # of key/value
+						 * elements
+						 * output: # of filled elements
+						 */
+		__u32		map_fd;
+		__u64		elem_flags;
+		__u64		flags;
+	} batch;
+
 	struct { /* anonymous struct used by BPF_PROG_LOAD command */
 		__u32		prog_type;	/* one of enum bpf_prog_type */
 		__u32		insn_cnt;
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 08b0b6e40454..d604ddbb1afb 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -219,10 +219,8 @@ static int bpf_map_copy_value(struct bpf_map *map, void *key, void *value,
 	void *ptr;
 	int err;
 
-	if (bpf_map_is_dev_bound(map)) {
-		err =  bpf_map_offload_lookup_elem(map, key, value);
-		return err;
-	}
+	if (bpf_map_is_dev_bound(map))
+		return bpf_map_offload_lookup_elem(map, key, value);
 
 	preempt_disable();
 	this_cpu_inc(bpf_prog_active);
@@ -1220,6 +1218,109 @@ err_put:
 	return err;
 }
 
+#define MAP_LOOKUP_RETRIES 3
+
+int generic_map_lookup_batch(struct bpf_map *map,
+				    const union bpf_attr *attr,
+				    union bpf_attr __user *uattr)
+{
+	void __user *uobatch = u64_to_user_ptr(attr->batch.out_batch);
+	void __user *ubatch = u64_to_user_ptr(attr->batch.in_batch);
+	void __user *values = u64_to_user_ptr(attr->batch.values);
+	void __user *keys = u64_to_user_ptr(attr->batch.keys);
+	void *buf, *buf_prevkey, *prev_key, *key, *value;
+	int err, retry = MAP_LOOKUP_RETRIES;
+	u32 value_size, cp, max_count;
+	bool first_key = false;
+
+	if (attr->batch.elem_flags & ~BPF_F_LOCK)
+		return -EINVAL;
+
+	if ((attr->batch.elem_flags & BPF_F_LOCK) &&
+	    !map_value_has_spin_lock(map))
+		return -EINVAL;
+
+	value_size = bpf_map_value_size(map);
+
+	max_count = attr->batch.count;
+	if (!max_count)
+		return 0;
+
+	if (put_user(0, &uattr->batch.count))
+		return -EFAULT;
+
+	buf_prevkey = kmalloc(map->key_size, GFP_USER | __GFP_NOWARN);
+	if (!buf_prevkey)
+		return -ENOMEM;
+
+	buf = kmalloc(map->key_size + value_size, GFP_USER | __GFP_NOWARN);
+	if (!buf) {
+		kvfree(buf_prevkey);
+		return -ENOMEM;
+	}
+
+	err = -EFAULT;
+	first_key = false;
+	prev_key = NULL;
+	if (ubatch && copy_from_user(buf_prevkey, ubatch, map->key_size))
+		goto free_buf;
+	key = buf;
+	value = key + map->key_size;
+	if (ubatch)
+		prev_key = buf_prevkey;
+
+	for (cp = 0; cp < max_count;) {
+		rcu_read_lock();
+		err = map->ops->map_get_next_key(map, prev_key, key);
+		rcu_read_unlock();
+		if (err)
+			break;
+		err = bpf_map_copy_value(map, key, value,
+					 attr->batch.elem_flags);
+
+		if (err == -ENOENT) {
+			if (retry) {
+				retry--;
+				continue;
+			}
+			err = -EINTR;
+			break;
+		}
+
+		if (err)
+			goto free_buf;
+
+		if (copy_to_user(keys + cp * map->key_size, key,
+				 map->key_size)) {
+			err = -EFAULT;
+			goto free_buf;
+		}
+		if (copy_to_user(values + cp * value_size, value, value_size)) {
+			err = -EFAULT;
+			goto free_buf;
+		}
+
+		if (!prev_key)
+			prev_key = buf_prevkey;
+
+		swap(prev_key, key);
+		retry = MAP_LOOKUP_RETRIES;
+		cp++;
+	}
+
+	if (err == -EFAULT)
+		goto free_buf;
+
+	if ((copy_to_user(&uattr->batch.count, &cp, sizeof(cp)) ||
+		    (cp && copy_to_user(uobatch, prev_key, map->key_size))))
+		err = -EFAULT;
+
+free_buf:
+	kfree(buf_prevkey);
+	kfree(buf);
+	return err;
+}
+
 #define BPF_MAP_LOOKUP_AND_DELETE_ELEM_LAST_FIELD value
 
 static int map_lookup_and_delete_elem(union bpf_attr *attr)
@@ -3076,6 +3177,54 @@ out:
 	return err;
 }
 
+#define BPF_MAP_BATCH_LAST_FIELD batch.flags
+
+#define BPF_DO_BATCH(fn)			\
+	do {					\
+		if (!fn) {			\
+			err = -ENOTSUPP;	\
+			goto err_put;		\
+		}				\
+		err = fn(map, attr, uattr);	\
+	} while (0)
+
+static int bpf_map_do_batch(const union bpf_attr *attr,
+			    union bpf_attr __user *uattr,
+			    int cmd)
+{
+	struct bpf_map *map;
+	int err, ufd;
+	struct fd f;
+
+	if (CHECK_ATTR(BPF_MAP_BATCH))
+		return -EINVAL;
+
+	ufd = attr->batch.map_fd;
+	f = fdget(ufd);
+	map = __bpf_map_get(f);
+	if (IS_ERR(map))
+		return PTR_ERR(map);
+
+	if (cmd == BPF_MAP_LOOKUP_BATCH &&
+	    !(map_get_sys_perms(map, f) & FMODE_CAN_READ)) {
+		err = -EPERM;
+		goto err_put;
+	}
+
+	if (cmd != BPF_MAP_LOOKUP_BATCH &&
+	    !(map_get_sys_perms(map, f) & FMODE_CAN_WRITE)) {
+		err = -EPERM;
+		goto err_put;
+	}
+
+	if (cmd == BPF_MAP_LOOKUP_BATCH)
+		BPF_DO_BATCH(map->ops->map_lookup_batch);
+
+err_put:
+	fdput(f);
+	return err;
+}
+
 SYSCALL_DEFINE3(bpf, int, cmd, union bpf_attr __user *, uattr, unsigned int, size)
 {
 	union bpf_attr attr = {};
@@ -3173,6 +3322,9 @@ SYSCALL_DEFINE3(bpf, int, cmd, union bpf_attr __user *, uattr, unsigned int, siz
 	case BPF_MAP_LOOKUP_AND_DELETE_ELEM:
 		err = map_lookup_and_delete_elem(&attr);
 		break;
+	case BPF_MAP_LOOKUP_BATCH:
+		err = bpf_map_do_batch(&attr, uattr, BPF_MAP_LOOKUP_BATCH);
+		break;
 	default:
 		err = -EINVAL;
 		break;
-- 
cgit v1.2.3-59-g8ed1b


From aa2e93b8e58e18442edfb2427446732415bc215e Mon Sep 17 00:00:00 2001
From: Brian Vazquez <brianvv@google.com>
Date: Wed, 15 Jan 2020 10:43:02 -0800
Subject: bpf: Add generic support for update and delete batch ops

This commit adds generic support for update and delete batch ops that
can be used for almost all the bpf maps. These commands share the same
UAPI attr that lookup and lookup_and_delete batch ops use and the
syscall commands are:

  BPF_MAP_UPDATE_BATCH
  BPF_MAP_DELETE_BATCH

The main difference between update/delete and lookup batch ops is that
for update/delete keys/values must be specified for userspace and
because of that, neither in_batch nor out_batch are used.

Suggested-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Brian Vazquez <brianvv@google.com>
Signed-off-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200115184308.162644-4-brianvv@google.com
---
 include/linux/bpf.h      |  10 +++++
 include/uapi/linux/bpf.h |   2 +
 kernel/bpf/syscall.c     | 115 +++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 127 insertions(+)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 807744ecaa5a..05466ad6cf1c 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -46,6 +46,10 @@ struct bpf_map_ops {
 	void *(*map_lookup_elem_sys_only)(struct bpf_map *map, void *key);
 	int (*map_lookup_batch)(struct bpf_map *map, const union bpf_attr *attr,
 				union bpf_attr __user *uattr);
+	int (*map_update_batch)(struct bpf_map *map, const union bpf_attr *attr,
+				union bpf_attr __user *uattr);
+	int (*map_delete_batch)(struct bpf_map *map, const union bpf_attr *attr,
+				union bpf_attr __user *uattr);
 
 	/* funcs callable from userspace and from eBPF programs */
 	void *(*map_lookup_elem)(struct bpf_map *map, void *key);
@@ -987,6 +991,12 @@ void bpf_map_init_from_attr(struct bpf_map *map, union bpf_attr *attr);
 int  generic_map_lookup_batch(struct bpf_map *map,
 			      const union bpf_attr *attr,
 			      union bpf_attr __user *uattr);
+int  generic_map_update_batch(struct bpf_map *map,
+			      const union bpf_attr *attr,
+			      union bpf_attr __user *uattr);
+int  generic_map_delete_batch(struct bpf_map *map,
+			      const union bpf_attr *attr,
+			      union bpf_attr __user *uattr);
 
 extern int sysctl_unprivileged_bpf_disabled;
 
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 0721b2c6bb8c..d5320b0be459 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -108,6 +108,8 @@ enum bpf_cmd {
 	BPF_MAP_FREEZE,
 	BPF_BTF_GET_NEXT_ID,
 	BPF_MAP_LOOKUP_BATCH,
+	BPF_MAP_UPDATE_BATCH,
+	BPF_MAP_DELETE_BATCH,
 };
 
 enum bpf_map_type {
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index d604ddbb1afb..ce8244d1ba99 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -1218,6 +1218,111 @@ err_put:
 	return err;
 }
 
+int generic_map_delete_batch(struct bpf_map *map,
+			     const union bpf_attr *attr,
+			     union bpf_attr __user *uattr)
+{
+	void __user *keys = u64_to_user_ptr(attr->batch.keys);
+	u32 cp, max_count;
+	int err = 0;
+	void *key;
+
+	if (attr->batch.elem_flags & ~BPF_F_LOCK)
+		return -EINVAL;
+
+	if ((attr->batch.elem_flags & BPF_F_LOCK) &&
+	    !map_value_has_spin_lock(map)) {
+		return -EINVAL;
+	}
+
+	max_count = attr->batch.count;
+	if (!max_count)
+		return 0;
+
+	for (cp = 0; cp < max_count; cp++) {
+		key = __bpf_copy_key(keys + cp * map->key_size, map->key_size);
+		if (IS_ERR(key)) {
+			err = PTR_ERR(key);
+			break;
+		}
+
+		if (bpf_map_is_dev_bound(map)) {
+			err = bpf_map_offload_delete_elem(map, key);
+			break;
+		}
+
+		preempt_disable();
+		__this_cpu_inc(bpf_prog_active);
+		rcu_read_lock();
+		err = map->ops->map_delete_elem(map, key);
+		rcu_read_unlock();
+		__this_cpu_dec(bpf_prog_active);
+		preempt_enable();
+		maybe_wait_bpf_programs(map);
+		if (err)
+			break;
+	}
+	if (copy_to_user(&uattr->batch.count, &cp, sizeof(cp)))
+		err = -EFAULT;
+	return err;
+}
+
+int generic_map_update_batch(struct bpf_map *map,
+			     const union bpf_attr *attr,
+			     union bpf_attr __user *uattr)
+{
+	void __user *values = u64_to_user_ptr(attr->batch.values);
+	void __user *keys = u64_to_user_ptr(attr->batch.keys);
+	u32 value_size, cp, max_count;
+	int ufd = attr->map_fd;
+	void *key, *value;
+	struct fd f;
+	int err = 0;
+
+	f = fdget(ufd);
+	if (attr->batch.elem_flags & ~BPF_F_LOCK)
+		return -EINVAL;
+
+	if ((attr->batch.elem_flags & BPF_F_LOCK) &&
+	    !map_value_has_spin_lock(map)) {
+		return -EINVAL;
+	}
+
+	value_size = bpf_map_value_size(map);
+
+	max_count = attr->batch.count;
+	if (!max_count)
+		return 0;
+
+	value = kmalloc(value_size, GFP_USER | __GFP_NOWARN);
+	if (!value)
+		return -ENOMEM;
+
+	for (cp = 0; cp < max_count; cp++) {
+		key = __bpf_copy_key(keys + cp * map->key_size, map->key_size);
+		if (IS_ERR(key)) {
+			err = PTR_ERR(key);
+			break;
+		}
+		err = -EFAULT;
+		if (copy_from_user(value, values + cp * value_size, value_size))
+			break;
+
+		err = bpf_map_update_value(map, f, key, value,
+					   attr->batch.elem_flags);
+
+		if (err)
+			break;
+	}
+
+	if (copy_to_user(&uattr->batch.count, &cp, sizeof(cp)))
+		err = -EFAULT;
+
+	kfree(value);
+	kfree(key);
+	return err;
+}
+
 #define MAP_LOOKUP_RETRIES 3
 
 int generic_map_lookup_batch(struct bpf_map *map,
@@ -3219,6 +3324,10 @@ static int bpf_map_do_batch(const union bpf_attr *attr,
 
 	if (cmd == BPF_MAP_LOOKUP_BATCH)
 		BPF_DO_BATCH(map->ops->map_lookup_batch);
+	else if (cmd == BPF_MAP_UPDATE_BATCH)
+		BPF_DO_BATCH(map->ops->map_update_batch);
+	else
+		BPF_DO_BATCH(map->ops->map_delete_batch);
 
 err_put:
 	fdput(f);
@@ -3325,6 +3434,12 @@ SYSCALL_DEFINE3(bpf, int, cmd, union bpf_attr __user *, uattr, unsigned int, siz
 	case BPF_MAP_LOOKUP_BATCH:
 		err = bpf_map_do_batch(&attr, uattr, BPF_MAP_LOOKUP_BATCH);
 		break;
+	case BPF_MAP_UPDATE_BATCH:
+		err = bpf_map_do_batch(&attr, uattr, BPF_MAP_UPDATE_BATCH);
+		break;
+	case BPF_MAP_DELETE_BATCH:
+		err = bpf_map_do_batch(&attr, uattr, BPF_MAP_DELETE_BATCH);
+		break;
 	default:
 		err = -EINVAL;
 		break;
-- 
cgit v1.2.3-59-g8ed1b


From c60f2d2861778de6370a4f4ca6ab1d7d4a32efae Mon Sep 17 00:00:00 2001
From: Brian Vazquez <brianvv@google.com>
Date: Wed, 15 Jan 2020 10:43:03 -0800
Subject: bpf: Add lookup and update batch ops to arraymap

This adds the generic batch ops functionality to bpf arraymap, note that
since deletion is not a valid operation for arraymap, only batch and
lookup are added.

Signed-off-by: Brian Vazquez <brianvv@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20200115184308.162644-5-brianvv@google.com
---
 kernel/bpf/arraymap.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c
index f0d19bbb9211..95d77770353c 100644
--- a/kernel/bpf/arraymap.c
+++ b/kernel/bpf/arraymap.c
@@ -503,6 +503,8 @@ const struct bpf_map_ops array_map_ops = {
 	.map_mmap = array_map_mmap,
 	.map_seq_show_elem = array_map_seq_show_elem,
 	.map_check_btf = array_map_check_btf,
+	.map_lookup_batch = generic_map_lookup_batch,
+	.map_update_batch = generic_map_update_batch,
 };
 
 const struct bpf_map_ops percpu_array_map_ops = {
-- 
cgit v1.2.3-59-g8ed1b


From 057996380a42bb64ccc04383cfa9c0ace4ea11f0 Mon Sep 17 00:00:00 2001
From: Yonghong Song <yhs@fb.com>
Date: Wed, 15 Jan 2020 10:43:04 -0800
Subject: bpf: Add batch ops to all htab bpf map

htab can't use generic batch support due some problematic behaviours
inherent to the data structre, i.e. while iterating the bpf map  a
concurrent program might delete the next entry that batch was about to
use, in that case there's no easy solution to retrieve the next entry,
the issue has been discussed multiple times (see [1] and [2]).

The only way hmap can be traversed without the problem previously
exposed is by making sure that the map is traversing entire buckets.
This commit implements those strict requirements for hmap, the
implementation follows the same interaction that generic support with
some exceptions:

 - If keys/values buffer are not big enough to traverse a bucket,
   ENOSPC will be returned.
 - out_batch contains the value of the next bucket in the iteration, not
   the next key, but this is transparent for the user since the user
   should never use out_batch for other than bpf batch syscalls.

This commits implements BPF_MAP_LOOKUP_BATCH and adds support for new
command BPF_MAP_LOOKUP_AND_DELETE_BATCH. Note that for update/delete
batch ops it is possible to use the generic implementations.

[1] https://lore.kernel.org/bpf/20190724165803.87470-1-brianvv@google.com/
[2] https://lore.kernel.org/bpf/20190906225434.3635421-1-yhs@fb.com/

Signed-off-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Brian Vazquez <brianvv@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200115184308.162644-6-brianvv@google.com
---
 include/linux/bpf.h      |   3 +
 include/uapi/linux/bpf.h |   1 +
 kernel/bpf/hashtab.c     | 264 +++++++++++++++++++++++++++++++++++++++++++++++
 kernel/bpf/syscall.c     |   9 +-
 4 files changed, 276 insertions(+), 1 deletion(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 05466ad6cf1c..3517e32149a4 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -46,6 +46,9 @@ struct bpf_map_ops {
 	void *(*map_lookup_elem_sys_only)(struct bpf_map *map, void *key);
 	int (*map_lookup_batch)(struct bpf_map *map, const union bpf_attr *attr,
 				union bpf_attr __user *uattr);
+	int (*map_lookup_and_delete_batch)(struct bpf_map *map,
+					   const union bpf_attr *attr,
+					   union bpf_attr __user *uattr);
 	int (*map_update_batch)(struct bpf_map *map, const union bpf_attr *attr,
 				union bpf_attr __user *uattr);
 	int (*map_delete_batch)(struct bpf_map *map, const union bpf_attr *attr,
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index d5320b0be459..033d90a2282d 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -108,6 +108,7 @@ enum bpf_cmd {
 	BPF_MAP_FREEZE,
 	BPF_BTF_GET_NEXT_ID,
 	BPF_MAP_LOOKUP_BATCH,
+	BPF_MAP_LOOKUP_AND_DELETE_BATCH,
 	BPF_MAP_UPDATE_BATCH,
 	BPF_MAP_DELETE_BATCH,
 };
diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c
index 22066a62c8c9..2d182c4ee9d9 100644
--- a/kernel/bpf/hashtab.c
+++ b/kernel/bpf/hashtab.c
@@ -17,6 +17,16 @@
 	(BPF_F_NO_PREALLOC | BPF_F_NO_COMMON_LRU | BPF_F_NUMA_NODE |	\
 	 BPF_F_ACCESS_MASK | BPF_F_ZERO_SEED)
 
+#define BATCH_OPS(_name)			\
+	.map_lookup_batch =			\
+	_name##_map_lookup_batch,		\
+	.map_lookup_and_delete_batch =		\
+	_name##_map_lookup_and_delete_batch,	\
+	.map_update_batch =			\
+	generic_map_update_batch,		\
+	.map_delete_batch =			\
+	generic_map_delete_batch
+
 struct bucket {
 	struct hlist_nulls_head head;
 	raw_spinlock_t lock;
@@ -1232,6 +1242,256 @@ static void htab_map_seq_show_elem(struct bpf_map *map, void *key,
 	rcu_read_unlock();
 }
 
+static int
+__htab_map_lookup_and_delete_batch(struct bpf_map *map,
+				   const union bpf_attr *attr,
+				   union bpf_attr __user *uattr,
+				   bool do_delete, bool is_lru_map,
+				   bool is_percpu)
+{
+	struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
+	u32 bucket_cnt, total, key_size, value_size, roundup_key_size;
+	void *keys = NULL, *values = NULL, *value, *dst_key, *dst_val;
+	void __user *uvalues = u64_to_user_ptr(attr->batch.values);
+	void __user *ukeys = u64_to_user_ptr(attr->batch.keys);
+	void *ubatch = u64_to_user_ptr(attr->batch.in_batch);
+	u32 batch, max_count, size, bucket_size;
+	u64 elem_map_flags, map_flags;
+	struct hlist_nulls_head *head;
+	struct hlist_nulls_node *n;
+	unsigned long flags;
+	struct htab_elem *l;
+	struct bucket *b;
+	int ret = 0;
+
+	elem_map_flags = attr->batch.elem_flags;
+	if ((elem_map_flags & ~BPF_F_LOCK) ||
+	    ((elem_map_flags & BPF_F_LOCK) && !map_value_has_spin_lock(map)))
+		return -EINVAL;
+
+	map_flags = attr->batch.flags;
+	if (map_flags)
+		return -EINVAL;
+
+	max_count = attr->batch.count;
+	if (!max_count)
+		return 0;
+
+	if (put_user(0, &uattr->batch.count))
+		return -EFAULT;
+
+	batch = 0;
+	if (ubatch && copy_from_user(&batch, ubatch, sizeof(batch)))
+		return -EFAULT;
+
+	if (batch >= htab->n_buckets)
+		return -ENOENT;
+
+	key_size = htab->map.key_size;
+	roundup_key_size = round_up(htab->map.key_size, 8);
+	value_size = htab->map.value_size;
+	size = round_up(value_size, 8);
+	if (is_percpu)
+		value_size = size * num_possible_cpus();
+	total = 0;
+	/* while experimenting with hash tables with sizes ranging from 10 to
+	 * 1000, it was observed that a bucket can have upto 5 entries.
+	 */
+	bucket_size = 5;
+
+alloc:
+	/* We cannot do copy_from_user or copy_to_user inside
+	 * the rcu_read_lock. Allocate enough space here.
+	 */
+	keys = kvmalloc(key_size * bucket_size, GFP_USER | __GFP_NOWARN);
+	values = kvmalloc(value_size * bucket_size, GFP_USER | __GFP_NOWARN);
+	if (!keys || !values) {
+		ret = -ENOMEM;
+		goto after_loop;
+	}
+
+again:
+	preempt_disable();
+	this_cpu_inc(bpf_prog_active);
+	rcu_read_lock();
+again_nocopy:
+	dst_key = keys;
+	dst_val = values;
+	b = &htab->buckets[batch];
+	head = &b->head;
+	raw_spin_lock_irqsave(&b->lock, flags);
+
+	bucket_cnt = 0;
+	hlist_nulls_for_each_entry_rcu(l, n, head, hash_node)
+		bucket_cnt++;
+
+	if (bucket_cnt > (max_count - total)) {
+		if (total == 0)
+			ret = -ENOSPC;
+		raw_spin_unlock_irqrestore(&b->lock, flags);
+		rcu_read_unlock();
+		this_cpu_dec(bpf_prog_active);
+		preempt_enable();
+		goto after_loop;
+	}
+
+	if (bucket_cnt > bucket_size) {
+		bucket_size = bucket_cnt;
+		raw_spin_unlock_irqrestore(&b->lock, flags);
+		rcu_read_unlock();
+		this_cpu_dec(bpf_prog_active);
+		preempt_enable();
+		kvfree(keys);
+		kvfree(values);
+		goto alloc;
+	}
+
+	hlist_nulls_for_each_entry_safe(l, n, head, hash_node) {
+		memcpy(dst_key, l->key, key_size);
+
+		if (is_percpu) {
+			int off = 0, cpu;
+			void __percpu *pptr;
+
+			pptr = htab_elem_get_ptr(l, map->key_size);
+			for_each_possible_cpu(cpu) {
+				bpf_long_memcpy(dst_val + off,
+						per_cpu_ptr(pptr, cpu), size);
+				off += size;
+			}
+		} else {
+			value = l->key + roundup_key_size;
+			if (elem_map_flags & BPF_F_LOCK)
+				copy_map_value_locked(map, dst_val, value,
+						      true);
+			else
+				copy_map_value(map, dst_val, value);
+			check_and_init_map_lock(map, dst_val);
+		}
+		if (do_delete) {
+			hlist_nulls_del_rcu(&l->hash_node);
+			if (is_lru_map)
+				bpf_lru_push_free(&htab->lru, &l->lru_node);
+			else
+				free_htab_elem(htab, l);
+		}
+		dst_key += key_size;
+		dst_val += value_size;
+	}
+
+	raw_spin_unlock_irqrestore(&b->lock, flags);
+	/* If we are not copying data, we can go to next bucket and avoid
+	 * unlocking the rcu.
+	 */
+	if (!bucket_cnt && (batch + 1 < htab->n_buckets)) {
+		batch++;
+		goto again_nocopy;
+	}
+
+	rcu_read_unlock();
+	this_cpu_dec(bpf_prog_active);
+	preempt_enable();
+	if (bucket_cnt && (copy_to_user(ukeys + total * key_size, keys,
+	    key_size * bucket_cnt) ||
+	    copy_to_user(uvalues + total * value_size, values,
+	    value_size * bucket_cnt))) {
+		ret = -EFAULT;
+		goto after_loop;
+	}
+
+	total += bucket_cnt;
+	batch++;
+	if (batch >= htab->n_buckets) {
+		ret = -ENOENT;
+		goto after_loop;
+	}
+	goto again;
+
+after_loop:
+	if (ret == -EFAULT)
+		goto out;
+
+	/* copy # of entries and next batch */
+	ubatch = u64_to_user_ptr(attr->batch.out_batch);
+	if (copy_to_user(ubatch, &batch, sizeof(batch)) ||
+	    put_user(total, &uattr->batch.count))
+		ret = -EFAULT;
+
+out:
+	kvfree(keys);
+	kvfree(values);
+	return ret;
+}
+
+static int
+htab_percpu_map_lookup_batch(struct bpf_map *map, const union bpf_attr *attr,
+			     union bpf_attr __user *uattr)
+{
+	return __htab_map_lookup_and_delete_batch(map, attr, uattr, false,
+						  false, true);
+}
+
+static int
+htab_percpu_map_lookup_and_delete_batch(struct bpf_map *map,
+					const union bpf_attr *attr,
+					union bpf_attr __user *uattr)
+{
+	return __htab_map_lookup_and_delete_batch(map, attr, uattr, true,
+						  false, true);
+}
+
+static int
+htab_map_lookup_batch(struct bpf_map *map, const union bpf_attr *attr,
+		      union bpf_attr __user *uattr)
+{
+	return __htab_map_lookup_and_delete_batch(map, attr, uattr, false,
+						  false, false);
+}
+
+static int
+htab_map_lookup_and_delete_batch(struct bpf_map *map,
+				 const union bpf_attr *attr,
+				 union bpf_attr __user *uattr)
+{
+	return __htab_map_lookup_and_delete_batch(map, attr, uattr, true,
+						  false, false);
+}
+
+static int
+htab_lru_percpu_map_lookup_batch(struct bpf_map *map,
+				 const union bpf_attr *attr,
+				 union bpf_attr __user *uattr)
+{
+	return __htab_map_lookup_and_delete_batch(map, attr, uattr, false,
+						  true, true);
+}
+
+static int
+htab_lru_percpu_map_lookup_and_delete_batch(struct bpf_map *map,
+					    const union bpf_attr *attr,
+					    union bpf_attr __user *uattr)
+{
+	return __htab_map_lookup_and_delete_batch(map, attr, uattr, true,
+						  true, true);
+}
+
+static int
+htab_lru_map_lookup_batch(struct bpf_map *map, const union bpf_attr *attr,
+			  union bpf_attr __user *uattr)
+{
+	return __htab_map_lookup_and_delete_batch(map, attr, uattr, false,
+						  true, false);
+}
+
+static int
+htab_lru_map_lookup_and_delete_batch(struct bpf_map *map,
+				     const union bpf_attr *attr,
+				     union bpf_attr __user *uattr)
+{
+	return __htab_map_lookup_and_delete_batch(map, attr, uattr, true,
+						  true, false);
+}
+
 const struct bpf_map_ops htab_map_ops = {
 	.map_alloc_check = htab_map_alloc_check,
 	.map_alloc = htab_map_alloc,
@@ -1242,6 +1502,7 @@ const struct bpf_map_ops htab_map_ops = {
 	.map_delete_elem = htab_map_delete_elem,
 	.map_gen_lookup = htab_map_gen_lookup,
 	.map_seq_show_elem = htab_map_seq_show_elem,
+	BATCH_OPS(htab),
 };
 
 const struct bpf_map_ops htab_lru_map_ops = {
@@ -1255,6 +1516,7 @@ const struct bpf_map_ops htab_lru_map_ops = {
 	.map_delete_elem = htab_lru_map_delete_elem,
 	.map_gen_lookup = htab_lru_map_gen_lookup,
 	.map_seq_show_elem = htab_map_seq_show_elem,
+	BATCH_OPS(htab_lru),
 };
 
 /* Called from eBPF program */
@@ -1368,6 +1630,7 @@ const struct bpf_map_ops htab_percpu_map_ops = {
 	.map_update_elem = htab_percpu_map_update_elem,
 	.map_delete_elem = htab_map_delete_elem,
 	.map_seq_show_elem = htab_percpu_map_seq_show_elem,
+	BATCH_OPS(htab_percpu),
 };
 
 const struct bpf_map_ops htab_lru_percpu_map_ops = {
@@ -1379,6 +1642,7 @@ const struct bpf_map_ops htab_lru_percpu_map_ops = {
 	.map_update_elem = htab_lru_percpu_map_update_elem,
 	.map_delete_elem = htab_lru_map_delete_elem,
 	.map_seq_show_elem = htab_percpu_map_seq_show_elem,
+	BATCH_OPS(htab_lru_percpu),
 };
 
 static int fd_htab_map_alloc_check(union bpf_attr *attr)
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index ce8244d1ba99..0d94d361379c 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -3310,7 +3310,8 @@ static int bpf_map_do_batch(const union bpf_attr *attr,
 	if (IS_ERR(map))
 		return PTR_ERR(map);
 
-	if (cmd == BPF_MAP_LOOKUP_BATCH &&
+	if ((cmd == BPF_MAP_LOOKUP_BATCH ||
+	     cmd == BPF_MAP_LOOKUP_AND_DELETE_BATCH) &&
 	    !(map_get_sys_perms(map, f) & FMODE_CAN_READ)) {
 		err = -EPERM;
 		goto err_put;
@@ -3324,6 +3325,8 @@ static int bpf_map_do_batch(const union bpf_attr *attr,
 
 	if (cmd == BPF_MAP_LOOKUP_BATCH)
 		BPF_DO_BATCH(map->ops->map_lookup_batch);
+	else if (cmd == BPF_MAP_LOOKUP_AND_DELETE_BATCH)
+		BPF_DO_BATCH(map->ops->map_lookup_and_delete_batch);
 	else if (cmd == BPF_MAP_UPDATE_BATCH)
 		BPF_DO_BATCH(map->ops->map_update_batch);
 	else
@@ -3434,6 +3437,10 @@ SYSCALL_DEFINE3(bpf, int, cmd, union bpf_attr __user *, uattr, unsigned int, siz
 	case BPF_MAP_LOOKUP_BATCH:
 		err = bpf_map_do_batch(&attr, uattr, BPF_MAP_LOOKUP_BATCH);
 		break;
+	case BPF_MAP_LOOKUP_AND_DELETE_BATCH:
+		err = bpf_map_do_batch(&attr, uattr,
+				       BPF_MAP_LOOKUP_AND_DELETE_BATCH);
+		break;
 	case BPF_MAP_UPDATE_BATCH:
 		err = bpf_map_do_batch(&attr, uattr, BPF_MAP_UPDATE_BATCH);
 		break;
-- 
cgit v1.2.3-59-g8ed1b


From a1e3a3b8ba2c381a043d0c4273837895c1d3cf9c Mon Sep 17 00:00:00 2001
From: Yonghong Song <yhs@fb.com>
Date: Wed, 15 Jan 2020 10:43:05 -0800
Subject: tools/bpf: Sync uapi header bpf.h

sync uapi header include/uapi/linux/bpf.h to
tools/include/uapi/linux/bpf.h

Signed-off-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200115184308.162644-7-brianvv@google.com
---
 tools/include/uapi/linux/bpf.h | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index a88837d4dd18..033d90a2282d 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -107,6 +107,10 @@ enum bpf_cmd {
 	BPF_MAP_LOOKUP_AND_DELETE_ELEM,
 	BPF_MAP_FREEZE,
 	BPF_BTF_GET_NEXT_ID,
+	BPF_MAP_LOOKUP_BATCH,
+	BPF_MAP_LOOKUP_AND_DELETE_BATCH,
+	BPF_MAP_UPDATE_BATCH,
+	BPF_MAP_DELETE_BATCH,
 };
 
 enum bpf_map_type {
@@ -420,6 +424,23 @@ union bpf_attr {
 		__u64		flags;
 	};
 
+	struct { /* struct used by BPF_MAP_*_BATCH commands */
+		__aligned_u64	in_batch;	/* start batch,
+						 * NULL to start from beginning
+						 */
+		__aligned_u64	out_batch;	/* output: next start batch */
+		__aligned_u64	keys;
+		__aligned_u64	values;
+		__u32		count;		/* input/output:
+						 * input: # of key/value
+						 * elements
+						 * output: # of filled elements
+						 */
+		__u32		map_fd;
+		__u64		elem_flags;
+		__u64		flags;
+	} batch;
+
 	struct { /* anonymous struct used by BPF_PROG_LOAD command */
 		__u32		prog_type;	/* one of enum bpf_prog_type */
 		__u32		insn_cnt;
-- 
cgit v1.2.3-59-g8ed1b


From 2ab3d86ea1859d1a4b830ce3551cba725620593a Mon Sep 17 00:00:00 2001
From: Yonghong Song <yhs@fb.com>
Date: Wed, 15 Jan 2020 10:43:06 -0800
Subject: libbpf: Add libbpf support to batch ops

Added four libbpf API functions to support map batch operations:
  . int bpf_map_delete_batch( ... )
  . int bpf_map_lookup_batch( ... )
  . int bpf_map_lookup_and_delete_batch( ... )
  . int bpf_map_update_batch( ... )

Signed-off-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200115184308.162644-8-brianvv@google.com
---
 tools/lib/bpf/bpf.c      | 58 ++++++++++++++++++++++++++++++++++++++++++++++++
 tools/lib/bpf/bpf.h      | 22 ++++++++++++++++++
 tools/lib/bpf/libbpf.map |  4 ++++
 3 files changed, 84 insertions(+)

diff --git a/tools/lib/bpf/bpf.c b/tools/lib/bpf/bpf.c
index 500afe478e94..317727d61214 100644
--- a/tools/lib/bpf/bpf.c
+++ b/tools/lib/bpf/bpf.c
@@ -452,6 +452,64 @@ int bpf_map_freeze(int fd)
 	return sys_bpf(BPF_MAP_FREEZE, &attr, sizeof(attr));
 }
 
+static int bpf_map_batch_common(int cmd, int fd, void  *in_batch,
+				void *out_batch, void *keys, void *values,
+				__u32 *count,
+				const struct bpf_map_batch_opts *opts)
+{
+	union bpf_attr attr = {};
+	int ret;
+
+	if (!OPTS_VALID(opts, bpf_map_batch_opts))
+		return -EINVAL;
+
+	memset(&attr, 0, sizeof(attr));
+	attr.batch.map_fd = fd;
+	attr.batch.in_batch = ptr_to_u64(in_batch);
+	attr.batch.out_batch = ptr_to_u64(out_batch);
+	attr.batch.keys = ptr_to_u64(keys);
+	attr.batch.values = ptr_to_u64(values);
+	attr.batch.count = *count;
+	attr.batch.elem_flags  = OPTS_GET(opts, elem_flags, 0);
+	attr.batch.flags = OPTS_GET(opts, flags, 0);
+
+	ret = sys_bpf(cmd, &attr, sizeof(attr));
+	*count = attr.batch.count;
+
+	return ret;
+}
+
+int bpf_map_delete_batch(int fd, void *keys, __u32 *count,
+			 const struct bpf_map_batch_opts *opts)
+{
+	return bpf_map_batch_common(BPF_MAP_DELETE_BATCH, fd, NULL,
+				    NULL, keys, NULL, count, opts);
+}
+
+int bpf_map_lookup_batch(int fd, void *in_batch, void *out_batch, void *keys,
+			 void *values, __u32 *count,
+			 const struct bpf_map_batch_opts *opts)
+{
+	return bpf_map_batch_common(BPF_MAP_LOOKUP_BATCH, fd, in_batch,
+				    out_batch, keys, values, count, opts);
+}
+
+int bpf_map_lookup_and_delete_batch(int fd, void *in_batch, void *out_batch,
+				    void *keys, void *values, __u32 *count,
+				    const struct bpf_map_batch_opts *opts)
+{
+	return bpf_map_batch_common(BPF_MAP_LOOKUP_AND_DELETE_BATCH,
+				    fd, in_batch, out_batch, keys, values,
+				    count, opts);
+}
+
+int bpf_map_update_batch(int fd, void *keys, void *values, __u32 *count,
+			 const struct bpf_map_batch_opts *opts)
+{
+	return bpf_map_batch_common(BPF_MAP_UPDATE_BATCH, fd, NULL, NULL,
+				    keys, values, count, opts);
+}
+
 int bpf_obj_pin(int fd, const char *pathname)
 {
 	union bpf_attr attr;
diff --git a/tools/lib/bpf/bpf.h b/tools/lib/bpf/bpf.h
index 56341d117e5b..b976e77316cc 100644
--- a/tools/lib/bpf/bpf.h
+++ b/tools/lib/bpf/bpf.h
@@ -127,6 +127,28 @@ LIBBPF_API int bpf_map_lookup_and_delete_elem(int fd, const void *key,
 LIBBPF_API int bpf_map_delete_elem(int fd, const void *key);
 LIBBPF_API int bpf_map_get_next_key(int fd, const void *key, void *next_key);
 LIBBPF_API int bpf_map_freeze(int fd);
+
+struct bpf_map_batch_opts {
+	size_t sz; /* size of this struct for forward/backward compatibility */
+	__u64 elem_flags;
+	__u64 flags;
+};
+#define bpf_map_batch_opts__last_field flags
+
+LIBBPF_API int bpf_map_delete_batch(int fd, void *keys,
+				    __u32 *count,
+				    const struct bpf_map_batch_opts *opts);
+LIBBPF_API int bpf_map_lookup_batch(int fd, void *in_batch, void *out_batch,
+				    void *keys, void *values, __u32 *count,
+				    const struct bpf_map_batch_opts *opts);
+LIBBPF_API int bpf_map_lookup_and_delete_batch(int fd, void *in_batch,
+					void *out_batch, void *keys,
+					void *values, __u32 *count,
+					const struct bpf_map_batch_opts *opts);
+LIBBPF_API int bpf_map_update_batch(int fd, void *keys, void *values,
+				    __u32 *count,
+				    const struct bpf_map_batch_opts *opts);
+
 LIBBPF_API int bpf_obj_pin(int fd, const char *pathname);
 LIBBPF_API int bpf_obj_get(const char *pathname);
 
diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
index a19f04e6e3d9..1902a0fc6afc 100644
--- a/tools/lib/bpf/libbpf.map
+++ b/tools/lib/bpf/libbpf.map
@@ -214,6 +214,10 @@ LIBBPF_0.0.7 {
 		btf_dump__emit_type_decl;
 		bpf_link__disconnect;
 		bpf_map__attach_struct_ops;
+		bpf_map_delete_batch;
+		bpf_map_lookup_and_delete_batch;
+		bpf_map_lookup_batch;
+		bpf_map_update_batch;
 		bpf_object__find_program_by_name;
 		bpf_object__attach_skeleton;
 		bpf_object__destroy_skeleton;
-- 
cgit v1.2.3-59-g8ed1b


From 30ff3c59137d00e083f68437665e00895cc271c7 Mon Sep 17 00:00:00 2001
From: Yonghong Song <yhs@fb.com>
Date: Wed, 15 Jan 2020 10:43:07 -0800
Subject: selftests/bpf: Add batch ops testing for htab and htab_percpu map

Tested bpf_map_lookup_batch(), bpf_map_lookup_and_delete_batch(),
bpf_map_update_batch(), and bpf_map_delete_batch() functionality.
  $ ./test_maps
    ...
      test_htab_map_batch_ops:PASS
      test_htab_percpu_map_batch_ops:PASS
    ...

Signed-off-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Brian Vazquez <brianvv@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200115184308.162644-9-brianvv@google.com
---
 .../selftests/bpf/map_tests/htab_map_batch_ops.c   | 283 +++++++++++++++++++++
 1 file changed, 283 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/map_tests/htab_map_batch_ops.c

diff --git a/tools/testing/selftests/bpf/map_tests/htab_map_batch_ops.c b/tools/testing/selftests/bpf/map_tests/htab_map_batch_ops.c
new file mode 100644
index 000000000000..976bf415fbdd
--- /dev/null
+++ b/tools/testing/selftests/bpf/map_tests/htab_map_batch_ops.c
@@ -0,0 +1,283 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2019 Facebook  */
+#include <stdio.h>
+#include <errno.h>
+#include <string.h>
+
+#include <bpf/bpf.h>
+#include <bpf/libbpf.h>
+
+#include <bpf_util.h>
+#include <test_maps.h>
+
+static void map_batch_update(int map_fd, __u32 max_entries, int *keys,
+			     void *values, bool is_pcpu)
+{
+	typedef BPF_DECLARE_PERCPU(int, value);
+	value *v = NULL;
+	int i, j, err;
+	DECLARE_LIBBPF_OPTS(bpf_map_batch_opts, opts,
+		.elem_flags = 0,
+		.flags = 0,
+	);
+
+	if (is_pcpu)
+		v = (value *)values;
+
+	for (i = 0; i < max_entries; i++) {
+		keys[i] = i + 1;
+		if (is_pcpu)
+			for (j = 0; j < bpf_num_possible_cpus(); j++)
+				bpf_percpu(v[i], j) = i + 2 + j;
+		else
+			((int *)values)[i] = i + 2;
+	}
+
+	err = bpf_map_update_batch(map_fd, keys, values, &max_entries, &opts);
+	CHECK(err, "bpf_map_update_batch()", "error:%s\n", strerror(errno));
+}
+
+static void map_batch_verify(int *visited, __u32 max_entries,
+			     int *keys, void *values, bool is_pcpu)
+{
+	typedef BPF_DECLARE_PERCPU(int, value);
+	value *v = NULL;
+	int i, j;
+
+	if (is_pcpu)
+		v = (value *)values;
+
+	memset(visited, 0, max_entries * sizeof(*visited));
+	for (i = 0; i < max_entries; i++) {
+
+		if (is_pcpu) {
+			for (j = 0; j < bpf_num_possible_cpus(); j++) {
+				CHECK(keys[i] + 1 + j != bpf_percpu(v[i], j),
+				      "key/value checking",
+				      "error: i %d j %d key %d value %d\n",
+				      i, j, keys[i], bpf_percpu(v[i],  j));
+			}
+		} else {
+			CHECK(keys[i] + 1 != ((int *)values)[i],
+			      "key/value checking",
+			      "error: i %d key %d value %d\n", i, keys[i],
+			      ((int *)values)[i]);
+		}
+
+		visited[i] = 1;
+
+	}
+	for (i = 0; i < max_entries; i++) {
+		CHECK(visited[i] != 1, "visited checking",
+		      "error: keys array at index %d missing\n", i);
+	}
+}
+
+void __test_map_lookup_and_delete_batch(bool is_pcpu)
+{
+	__u32 batch, count, total, total_success;
+	typedef BPF_DECLARE_PERCPU(int, value);
+	int map_fd, *keys, *visited, key;
+	const __u32 max_entries = 10;
+	value pcpu_values[max_entries];
+	int err, step, value_size;
+	bool nospace_err;
+	void *values;
+	struct bpf_create_map_attr xattr = {
+		.name = "hash_map",
+		.map_type = is_pcpu ? BPF_MAP_TYPE_PERCPU_HASH :
+			    BPF_MAP_TYPE_HASH,
+		.key_size = sizeof(int),
+		.value_size = sizeof(int),
+	};
+	DECLARE_LIBBPF_OPTS(bpf_map_batch_opts, opts,
+		.elem_flags = 0,
+		.flags = 0,
+	);
+
+	xattr.max_entries = max_entries;
+	map_fd = bpf_create_map_xattr(&xattr);
+	CHECK(map_fd == -1,
+	      "bpf_create_map_xattr()", "error:%s\n", strerror(errno));
+
+	value_size = is_pcpu ? sizeof(value) : sizeof(int);
+	keys = malloc(max_entries * sizeof(int));
+	if (is_pcpu)
+		values = pcpu_values;
+	else
+		values = malloc(max_entries * sizeof(int));
+	visited = malloc(max_entries * sizeof(int));
+	CHECK(!keys || !values || !visited, "malloc()",
+	      "error:%s\n", strerror(errno));
+
+	/* test 1: lookup/delete an empty hash table, -ENOENT */
+	count = max_entries;
+	err = bpf_map_lookup_and_delete_batch(map_fd, NULL, &batch, keys,
+					      values, &count, &opts);
+	CHECK((err && errno != ENOENT), "empty map",
+	      "error: %s\n", strerror(errno));
+
+	/* populate elements to the map */
+	map_batch_update(map_fd, max_entries, keys, values, is_pcpu);
+
+	/* test 2: lookup/delete with count = 0, success */
+	count = 0;
+	err = bpf_map_lookup_and_delete_batch(map_fd, NULL, &batch, keys,
+					      values, &count, &opts);
+	CHECK(err, "count = 0", "error: %s\n", strerror(errno));
+
+	/* test 3: lookup/delete with count = max_entries, success */
+	memset(keys, 0, max_entries * sizeof(*keys));
+	memset(values, 0, max_entries * value_size);
+	count = max_entries;
+	err = bpf_map_lookup_and_delete_batch(map_fd, NULL, &batch, keys,
+					      values, &count, &opts);
+	CHECK((err && errno != ENOENT), "count = max_entries",
+	       "error: %s\n", strerror(errno));
+	CHECK(count != max_entries, "count = max_entries",
+	      "count = %u, max_entries = %u\n", count, max_entries);
+	map_batch_verify(visited, max_entries, keys, values, is_pcpu);
+
+	/* bpf_map_get_next_key() should return -ENOENT for an empty map. */
+	err = bpf_map_get_next_key(map_fd, NULL, &key);
+	CHECK(!err, "bpf_map_get_next_key()", "error: %s\n", strerror(errno));
+
+	/* test 4: lookup/delete in a loop with various steps. */
+	total_success = 0;
+	for (step = 1; step < max_entries; step++) {
+		map_batch_update(map_fd, max_entries, keys, values, is_pcpu);
+		memset(keys, 0, max_entries * sizeof(*keys));
+		memset(values, 0, max_entries * value_size);
+		total = 0;
+		/* iteratively lookup/delete elements with 'step'
+		 * elements each
+		 */
+		count = step;
+		nospace_err = false;
+		while (true) {
+			err = bpf_map_lookup_batch(map_fd,
+						   total ? &batch : NULL,
+						   &batch, keys + total,
+						   values +
+						   total * value_size,
+						   &count, &opts);
+			/* It is possible that we are failing due to buffer size
+			 * not big enough. In such cases, let us just exit and
+			 * go with large steps. Not that a buffer size with
+			 * max_entries should always work.
+			 */
+			if (err && errno == ENOSPC) {
+				nospace_err = true;
+				break;
+			}
+
+			CHECK((err && errno != ENOENT), "lookup with steps",
+			      "error: %s\n", strerror(errno));
+
+			total += count;
+			if (err)
+				break;
+
+		}
+		if (nospace_err == true)
+			continue;
+
+		CHECK(total != max_entries, "lookup with steps",
+		      "total = %u, max_entries = %u\n", total, max_entries);
+		map_batch_verify(visited, max_entries, keys, values, is_pcpu);
+
+		total = 0;
+		count = step;
+		while (total < max_entries) {
+			if (max_entries - total < step)
+				count = max_entries - total;
+			err = bpf_map_delete_batch(map_fd,
+						   keys + total,
+						   &count, &opts);
+			CHECK((err && errno != ENOENT), "delete batch",
+			      "error: %s\n", strerror(errno));
+			total += count;
+			if (err)
+				break;
+		}
+		CHECK(total != max_entries, "delete with steps",
+		      "total = %u, max_entries = %u\n", total, max_entries);
+
+		/* check map is empty, errono == ENOENT */
+		err = bpf_map_get_next_key(map_fd, NULL, &key);
+		CHECK(!err || errno != ENOENT, "bpf_map_get_next_key()",
+		      "error: %s\n", strerror(errno));
+
+		/* iteratively lookup/delete elements with 'step'
+		 * elements each
+		 */
+		map_batch_update(map_fd, max_entries, keys, values, is_pcpu);
+		memset(keys, 0, max_entries * sizeof(*keys));
+		memset(values, 0, max_entries * value_size);
+		total = 0;
+		count = step;
+		nospace_err = false;
+		while (true) {
+			err = bpf_map_lookup_and_delete_batch(map_fd,
+							total ? &batch : NULL,
+							&batch, keys + total,
+							values +
+							total * value_size,
+							&count, &opts);
+			/* It is possible that we are failing due to buffer size
+			 * not big enough. In such cases, let us just exit and
+			 * go with large steps. Not that a buffer size with
+			 * max_entries should always work.
+			 */
+			if (err && errno == ENOSPC) {
+				nospace_err = true;
+				break;
+			}
+
+			CHECK((err && errno != ENOENT), "lookup with steps",
+			      "error: %s\n", strerror(errno));
+
+			total += count;
+			if (err)
+				break;
+		}
+
+		if (nospace_err == true)
+			continue;
+
+		CHECK(total != max_entries, "lookup/delete with steps",
+		      "total = %u, max_entries = %u\n", total, max_entries);
+
+		map_batch_verify(visited, max_entries, keys, values, is_pcpu);
+		err = bpf_map_get_next_key(map_fd, NULL, &key);
+		CHECK(!err, "bpf_map_get_next_key()", "error: %s\n",
+		      strerror(errno));
+
+		total_success++;
+	}
+
+	CHECK(total_success == 0, "check total_success",
+	      "unexpected failure\n");
+	free(keys);
+	free(visited);
+	if (!is_pcpu)
+		free(values);
+}
+
+void htab_map_batch_ops(void)
+{
+	__test_map_lookup_and_delete_batch(false);
+	printf("test_%s:PASS\n", __func__);
+}
+
+void htab_percpu_map_batch_ops(void)
+{
+	__test_map_lookup_and_delete_batch(true);
+	printf("test_%s:PASS\n", __func__);
+}
+
+void test_htab_map_batch_ops(void)
+{
+	htab_map_batch_ops();
+	htab_percpu_map_batch_ops();
+}
-- 
cgit v1.2.3-59-g8ed1b


From f0fac2cec2861abc6fc0c094fb2584d9a33f868d Mon Sep 17 00:00:00 2001
From: Brian Vazquez <brianvv@google.com>
Date: Wed, 15 Jan 2020 10:43:08 -0800
Subject: selftests/bpf: Add batch ops testing to array bpf map

Tested bpf_map_lookup_batch() and bpf_map_update_batch()
functionality.

  $ ./test_maps
      ...
        test_array_map_batch_ops:PASS
      ...

Signed-off-by: Brian Vazquez <brianvv@google.com>
Signed-off-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200115184308.162644-10-brianvv@google.com
---
 .../selftests/bpf/map_tests/array_map_batch_ops.c  | 129 +++++++++++++++++++++
 1 file changed, 129 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/map_tests/array_map_batch_ops.c

diff --git a/tools/testing/selftests/bpf/map_tests/array_map_batch_ops.c b/tools/testing/selftests/bpf/map_tests/array_map_batch_ops.c
new file mode 100644
index 000000000000..f0a64d8ac59a
--- /dev/null
+++ b/tools/testing/selftests/bpf/map_tests/array_map_batch_ops.c
@@ -0,0 +1,129 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <stdio.h>
+#include <errno.h>
+#include <string.h>
+
+#include <bpf/bpf.h>
+#include <bpf/libbpf.h>
+
+#include <test_maps.h>
+
+static void map_batch_update(int map_fd, __u32 max_entries, int *keys,
+			     int *values)
+{
+	int i, err;
+	DECLARE_LIBBPF_OPTS(bpf_map_batch_opts, opts,
+		.elem_flags = 0,
+		.flags = 0,
+	);
+
+	for (i = 0; i < max_entries; i++) {
+		keys[i] = i;
+		values[i] = i + 1;
+	}
+
+	err = bpf_map_update_batch(map_fd, keys, values, &max_entries, &opts);
+	CHECK(err, "bpf_map_update_batch()", "error:%s\n", strerror(errno));
+}
+
+static void map_batch_verify(int *visited, __u32 max_entries,
+			     int *keys, int *values)
+{
+	int i;
+
+	memset(visited, 0, max_entries * sizeof(*visited));
+	for (i = 0; i < max_entries; i++) {
+		CHECK(keys[i] + 1 != values[i], "key/value checking",
+		      "error: i %d key %d value %d\n", i, keys[i], values[i]);
+		visited[i] = 1;
+	}
+	for (i = 0; i < max_entries; i++) {
+		CHECK(visited[i] != 1, "visited checking",
+		      "error: keys array at index %d missing\n", i);
+	}
+}
+
+void test_array_map_batch_ops(void)
+{
+	struct bpf_create_map_attr xattr = {
+		.name = "array_map",
+		.map_type = BPF_MAP_TYPE_ARRAY,
+		.key_size = sizeof(int),
+		.value_size = sizeof(int),
+	};
+	int map_fd, *keys, *values, *visited;
+	__u32 count, total, total_success;
+	const __u32 max_entries = 10;
+	bool nospace_err;
+	__u64 batch = 0;
+	int err, step;
+	DECLARE_LIBBPF_OPTS(bpf_map_batch_opts, opts,
+		.elem_flags = 0,
+		.flags = 0,
+	);
+
+	xattr.max_entries = max_entries;
+	map_fd = bpf_create_map_xattr(&xattr);
+	CHECK(map_fd == -1,
+	      "bpf_create_map_xattr()", "error:%s\n", strerror(errno));
+
+	keys = malloc(max_entries * sizeof(int));
+	values = malloc(max_entries * sizeof(int));
+	visited = malloc(max_entries * sizeof(int));
+	CHECK(!keys || !values || !visited, "malloc()", "error:%s\n",
+	      strerror(errno));
+
+	/* populate elements to the map */
+	map_batch_update(map_fd, max_entries, keys, values);
+
+	/* test 1: lookup in a loop with various steps. */
+	total_success = 0;
+	for (step = 1; step < max_entries; step++) {
+		map_batch_update(map_fd, max_entries, keys, values);
+		map_batch_verify(visited, max_entries, keys, values);
+		memset(keys, 0, max_entries * sizeof(*keys));
+		memset(values, 0, max_entries * sizeof(*values));
+		batch = 0;
+		total = 0;
+		/* iteratively lookup/delete elements with 'step'
+		 * elements each.
+		 */
+		count = step;
+		nospace_err = false;
+		while (true) {
+			err = bpf_map_lookup_batch(map_fd,
+						total ? &batch : NULL, &batch,
+						keys + total,
+						values + total,
+						&count, &opts);
+
+			CHECK((err && errno != ENOENT), "lookup with steps",
+			      "error: %s\n", strerror(errno));
+
+			total += count;
+			if (err)
+				break;
+
+		}
+
+		if (nospace_err == true)
+			continue;
+
+		CHECK(total != max_entries, "lookup with steps",
+		      "total = %u, max_entries = %u\n", total, max_entries);
+
+		map_batch_verify(visited, max_entries, keys, values);
+
+		total_success++;
+	}
+
+	CHECK(total_success == 0, "check total_success",
+	      "unexpected failure\n");
+
+	printf("%s:PASS\n", __func__);
+
+	free(keys);
+	free(values);
+	free(visited);
+}
-- 
cgit v1.2.3-59-g8ed1b


From 43d28c612f4c02a3fdea3d906170faece8ada9e1 Mon Sep 17 00:00:00 2001
From: Ulrich Weber <ulrich.weber@gmail.com>
Date: Tue, 14 Jan 2020 15:19:43 +0100
Subject: pptp: support sockets bound to an interface

use sk_bound_dev_if for route lookup as already done
in most of the other ip_route_output_ports() calls.

Since most PPPoA providers use 10.0.0.138 as default gateway IP
this will allow connections to multiple PPTP providers with the
same IP address over different interfaces.

Signed-off-by: Ulrich Weber <ulrich.weber@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ppp/pptp.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ppp/pptp.c b/drivers/net/ppp/pptp.c
index e1fabb3e3246..acccb747aeda 100644
--- a/drivers/net/ppp/pptp.c
+++ b/drivers/net/ppp/pptp.c
@@ -155,7 +155,7 @@ static int pptp_xmit(struct ppp_channel *chan, struct sk_buff *skb)
 				   opt->dst_addr.sin_addr.s_addr,
 				   opt->src_addr.sin_addr.s_addr,
 				   0, 0, IPPROTO_GRE,
-				   RT_TOS(0), 0);
+				   RT_TOS(0), sk->sk_bound_dev_if);
 	if (IS_ERR(rt))
 		goto tx_error;
 
@@ -444,7 +444,8 @@ static int pptp_connect(struct socket *sock, struct sockaddr *uservaddr,
 				   opt->dst_addr.sin_addr.s_addr,
 				   opt->src_addr.sin_addr.s_addr,
 				   0, 0,
-				   IPPROTO_GRE, RT_CONN_FLAGS(sk), 0);
+				   IPPROTO_GRE, RT_CONN_FLAGS(sk),
+				   sk->sk_bound_dev_if);
 	if (IS_ERR(rt)) {
 		error = -EHOSTUNREACH;
 		goto end;
-- 
cgit v1.2.3-59-g8ed1b


From 1ccf6c13d9c7282466f46189e9a1a2ed7f19816a Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Tue, 14 Jan 2020 12:09:18 -0800
Subject: devlink: fix typos in qed documentation

Review of the recently added documentation file for the qed driver
noticed a couple of typos. Fix them now.

Noticed-by: Michal Kalderon <mkalderon@marvell.com>
Fixes: 0f261c3ca09e ("devlink: add a driver-specific file for the qed driver")
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/devlink/qed.rst | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/Documentation/networking/devlink/qed.rst b/Documentation/networking/devlink/qed.rst
index e7e17acf1eca..805c6f63621a 100644
--- a/Documentation/networking/devlink/qed.rst
+++ b/Documentation/networking/devlink/qed.rst
@@ -22,5 +22,5 @@ The ``qed`` driver implements the following driver-specific parameters.
    * - ``iwarp_cmt``
      - Boolean
      - runtime
-     - Enable iWARP functionality for 100g devices. Notee that this impacts
-       L2 performance, and is therefor not enabled by default.
+     - Enable iWARP functionality for 100g devices. Note that this impacts
+       L2 performance, and is therefore not enabled by default.
-- 
cgit v1.2.3-59-g8ed1b


From d7de72674af53c3839eff091899f18c221dd30d2 Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <kafai@fb.com>
Date: Wed, 15 Jan 2020 15:00:19 -0800
Subject: bpftool: Fix a leak of btf object

When testing a map has btf or not, maps_have_btf() tests it by actually
getting a btf_fd from sys_bpf(BPF_BTF_GET_FD_BY_ID). However, it
forgot to btf__free() it.

In maps_have_btf() stage, there is no need to test it by really
calling sys_bpf(BPF_BTF_GET_FD_BY_ID). Testing non zero
info.btf_id is good enough.

Also, the err_close case is unnecessary, and also causes double
close() because the calling func do_dump() will close() all fds again.

Fixes: 99f9863a0c45 ("bpftool: Match maps by name")
Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Cc: Paul Chaignon <paul.chaignon@orange.com>
Link: https://lore.kernel.org/bpf/20200115230019.1101352-1-kafai@fb.com
---
 tools/bpf/bpftool/map.c | 16 ++--------------
 1 file changed, 2 insertions(+), 14 deletions(-)

diff --git a/tools/bpf/bpftool/map.c b/tools/bpf/bpftool/map.c
index c01f76fa6876..e00e9e19d6b7 100644
--- a/tools/bpf/bpftool/map.c
+++ b/tools/bpf/bpftool/map.c
@@ -915,32 +915,20 @@ static int maps_have_btf(int *fds, int nb_fds)
 {
 	struct bpf_map_info info = {};
 	__u32 len = sizeof(info);
-	struct btf *btf = NULL;
 	int err, i;
 
 	for (i = 0; i < nb_fds; i++) {
 		err = bpf_obj_get_info_by_fd(fds[i], &info, &len);
 		if (err) {
 			p_err("can't get map info: %s", strerror(errno));
-			goto err_close;
-		}
-
-		err = btf__get_from_id(info.btf_id, &btf);
-		if (err) {
-			p_err("failed to get btf");
-			goto err_close;
+			return -1;
 		}
 
-		if (!btf)
+		if (!info.btf_id)
 			return 0;
 	}
 
 	return 1;
-
-err_close:
-	for (; i < nb_fds; i++)
-		close(fds[i]);
-	return -1;
 }
 
 static int
-- 
cgit v1.2.3-59-g8ed1b


From 188a486619e6c9d7b8531ba6c4215b31304d293f Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <kafai@fb.com>
Date: Wed, 15 Jan 2020 15:00:25 -0800
Subject: bpftool: Fix missing BTF output for json during map dump

The btf availability check is only done for plain text output.
It causes the whole BTF output went missing when json_output
is used.

This patch simplifies the logic a little by avoiding passing "int btf" to
map_dump().

For plain text output, the btf_wtr is only created when the map has
BTF (i.e. info->btf_id != 0).  The nullness of "json_writer_t *wtr"
in map_dump() alone can decide if dumping BTF output is needed.
As long as wtr is not NULL, map_dump() will print out the BTF-described
data whenever a map has BTF available (i.e. info->btf_id != 0)
regardless of json or plain-text output.

In do_dump(), the "int btf" is also renamed to "int do_plain_btf".

Fixes: 99f9863a0c45 ("bpftool: Match maps by name")
Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Cc: Paul Chaignon <paul.chaignon@orange.com>
Link: https://lore.kernel.org/bpf/20200115230025.1101828-1-kafai@fb.com
---
 tools/bpf/bpftool/map.c | 42 ++++++++++++++++++++----------------------
 1 file changed, 20 insertions(+), 22 deletions(-)

diff --git a/tools/bpf/bpftool/map.c b/tools/bpf/bpftool/map.c
index e00e9e19d6b7..45c1eda6512c 100644
--- a/tools/bpf/bpftool/map.c
+++ b/tools/bpf/bpftool/map.c
@@ -933,7 +933,7 @@ static int maps_have_btf(int *fds, int nb_fds)
 
 static int
 map_dump(int fd, struct bpf_map_info *info, json_writer_t *wtr,
-	 bool enable_btf, bool show_header)
+	 bool show_header)
 {
 	void *key, *value, *prev_key;
 	unsigned int num_elems = 0;
@@ -950,18 +950,16 @@ map_dump(int fd, struct bpf_map_info *info, json_writer_t *wtr,
 
 	prev_key = NULL;
 
-	if (enable_btf) {
-		err = btf__get_from_id(info->btf_id, &btf);
-		if (err || !btf) {
-			/* enable_btf is true only if we've already checked
-			 * that all maps have BTF information.
-			 */
-			p_err("failed to get btf");
-			goto exit_free;
+	if (wtr) {
+		if (info->btf_id) {
+			err = btf__get_from_id(info->btf_id, &btf);
+			if (err || !btf) {
+				err = err ? : -ESRCH;
+				p_err("failed to get btf");
+				goto exit_free;
+			}
 		}
-	}
 
-	if (wtr) {
 		if (show_header) {
 			jsonw_start_object(wtr);	/* map object */
 			show_map_header_json(info, wtr);
@@ -1009,7 +1007,7 @@ static int do_dump(int argc, char **argv)
 {
 	json_writer_t *wtr = NULL, *btf_wtr = NULL;
 	struct bpf_map_info info = {};
-	int nb_fds, i = 0, btf = 0;
+	int nb_fds, i = 0;
 	__u32 len = sizeof(info);
 	int *fds = NULL;
 	int err = -1;
@@ -1029,17 +1027,17 @@ static int do_dump(int argc, char **argv)
 	if (json_output) {
 		wtr = json_wtr;
 	} else {
-		btf = maps_have_btf(fds, nb_fds);
-		if (btf < 0)
+		int do_plain_btf;
+
+		do_plain_btf = maps_have_btf(fds, nb_fds);
+		if (do_plain_btf < 0)
 			goto exit_close;
-		if (btf) {
+
+		if (do_plain_btf) {
 			btf_wtr = get_btf_writer();
-			if (btf_wtr) {
-				wtr = btf_wtr;
-			} else {
+			wtr = btf_wtr;
+			if (!btf_wtr)
 				p_info("failed to create json writer for btf. falling back to plain output");
-				btf = 0;
-			}
 		}
 	}
 
@@ -1050,7 +1048,7 @@ static int do_dump(int argc, char **argv)
 			p_err("can't get map info: %s", strerror(errno));
 			break;
 		}
-		err = map_dump(fds[i], &info, wtr, btf, nb_fds > 1);
+		err = map_dump(fds[i], &info, wtr, nb_fds > 1);
 		if (!wtr && i != nb_fds - 1)
 			printf("\n");
 
@@ -1061,7 +1059,7 @@ static int do_dump(int argc, char **argv)
 	if (wtr && nb_fds > 1)
 		jsonw_end_array(wtr);	/* root array */
 
-	if (btf)
+	if (btf_wtr)
 		jsonw_destroy(&btf_wtr);
 exit_close:
 	for (; i < nb_fds; i++)
-- 
cgit v1.2.3-59-g8ed1b


From fb2426ad00b136c88d57457504d1e843159a367a Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <kafai@fb.com>
Date: Wed, 15 Jan 2020 15:00:31 -0800
Subject: libbpf: Expose bpf_find_kernel_btf as a LIBBPF_API

This patch exposes bpf_find_kernel_btf() as a LIBBPF_API.
It will be used in 'bpftool map dump' in a following patch
to dump a map with btf_vmlinux_value_type_id set.

bpf_find_kernel_btf() is renamed to libbpf_find_kernel_btf()
and moved to btf.c.  As <linux/kernel.h> is included,
some of the max/min type casting needs to be fixed.

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/20200115230031.1102305-1-kafai@fb.com
---
 tools/lib/bpf/btf.c      | 102 ++++++++++++++++++++++++++++++++++++++++++++---
 tools/lib/bpf/btf.h      |   2 +
 tools/lib/bpf/libbpf.c   |  93 ++----------------------------------------
 tools/lib/bpf/libbpf.map |   1 +
 4 files changed, 102 insertions(+), 96 deletions(-)

diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
index cfeb6a44480b..3d1c25fc97ae 100644
--- a/tools/lib/bpf/btf.c
+++ b/tools/lib/bpf/btf.c
@@ -8,6 +8,10 @@
 #include <fcntl.h>
 #include <unistd.h>
 #include <errno.h>
+#include <sys/utsname.h>
+#include <sys/param.h>
+#include <sys/stat.h>
+#include <linux/kernel.h>
 #include <linux/err.h>
 #include <linux/btf.h>
 #include <gelf.h>
@@ -20,8 +24,8 @@
 /* make sure libbpf doesn't use kernel-only integer typedefs */
 #pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
 
-#define BTF_MAX_NR_TYPES 0x7fffffff
-#define BTF_MAX_STR_OFFSET 0x7fffffff
+#define BTF_MAX_NR_TYPES 0x7fffffffU
+#define BTF_MAX_STR_OFFSET 0x7fffffffU
 
 static struct btf_type btf_void;
 
@@ -53,7 +57,7 @@ static int btf_add_type(struct btf *btf, struct btf_type *t)
 		if (btf->types_size == BTF_MAX_NR_TYPES)
 			return -E2BIG;
 
-		expand_by = max(btf->types_size >> 2, 16);
+		expand_by = max(btf->types_size >> 2, 16U);
 		new_size = min(BTF_MAX_NR_TYPES, btf->types_size + expand_by);
 
 		new_types = realloc(btf->types, sizeof(*new_types) * new_size);
@@ -289,7 +293,7 @@ int btf__align_of(const struct btf *btf, __u32 id)
 	switch (kind) {
 	case BTF_KIND_INT:
 	case BTF_KIND_ENUM:
-		return min(sizeof(void *), t->size);
+		return min(sizeof(void *), (size_t)t->size);
 	case BTF_KIND_PTR:
 		return sizeof(void *);
 	case BTF_KIND_TYPEDEF:
@@ -1401,7 +1405,7 @@ static int btf_dedup_hypot_map_add(struct btf_dedup *d,
 	if (d->hypot_cnt == d->hypot_cap) {
 		__u32 *new_list;
 
-		d->hypot_cap += max(16, d->hypot_cap / 2);
+		d->hypot_cap += max((size_t)16, d->hypot_cap / 2);
 		new_list = realloc(d->hypot_list, sizeof(__u32) * d->hypot_cap);
 		if (!new_list)
 			return -ENOMEM;
@@ -1697,7 +1701,7 @@ static int btf_dedup_strings(struct btf_dedup *d)
 		if (strs.cnt + 1 > strs.cap) {
 			struct btf_str_ptr *new_ptrs;
 
-			strs.cap += max(strs.cnt / 2, 16);
+			strs.cap += max(strs.cnt / 2, 16U);
 			new_ptrs = realloc(strs.ptrs,
 					   sizeof(strs.ptrs[0]) * strs.cap);
 			if (!new_ptrs) {
@@ -2931,3 +2935,89 @@ static int btf_dedup_remap_types(struct btf_dedup *d)
 	}
 	return 0;
 }
+
+static struct btf *btf_load_raw(const char *path)
+{
+	struct btf *btf;
+	size_t read_cnt;
+	struct stat st;
+	void *data;
+	FILE *f;
+
+	if (stat(path, &st))
+		return ERR_PTR(-errno);
+
+	data = malloc(st.st_size);
+	if (!data)
+		return ERR_PTR(-ENOMEM);
+
+	f = fopen(path, "rb");
+	if (!f) {
+		btf = ERR_PTR(-errno);
+		goto cleanup;
+	}
+
+	read_cnt = fread(data, 1, st.st_size, f);
+	fclose(f);
+	if (read_cnt < st.st_size) {
+		btf = ERR_PTR(-EBADF);
+		goto cleanup;
+	}
+
+	btf = btf__new(data, read_cnt);
+
+cleanup:
+	free(data);
+	return btf;
+}
+
+/*
+ * Probe few well-known locations for vmlinux kernel image and try to load BTF
+ * data out of it to use for target BTF.
+ */
+struct btf *libbpf_find_kernel_btf(void)
+{
+	struct {
+		const char *path_fmt;
+		bool raw_btf;
+	} locations[] = {
+		/* try canonical vmlinux BTF through sysfs first */
+		{ "/sys/kernel/btf/vmlinux", true /* raw BTF */ },
+		/* fall back to trying to find vmlinux ELF on disk otherwise */
+		{ "/boot/vmlinux-%1$s" },
+		{ "/lib/modules/%1$s/vmlinux-%1$s" },
+		{ "/lib/modules/%1$s/build/vmlinux" },
+		{ "/usr/lib/modules/%1$s/kernel/vmlinux" },
+		{ "/usr/lib/debug/boot/vmlinux-%1$s" },
+		{ "/usr/lib/debug/boot/vmlinux-%1$s.debug" },
+		{ "/usr/lib/debug/lib/modules/%1$s/vmlinux" },
+	};
+	char path[PATH_MAX + 1];
+	struct utsname buf;
+	struct btf *btf;
+	int i;
+
+	uname(&buf);
+
+	for (i = 0; i < ARRAY_SIZE(locations); i++) {
+		snprintf(path, PATH_MAX, locations[i].path_fmt, buf.release);
+
+		if (access(path, R_OK))
+			continue;
+
+		if (locations[i].raw_btf)
+			btf = btf_load_raw(path);
+		else
+			btf = btf__parse_elf(path, NULL);
+
+		pr_debug("loading kernel BTF '%s': %ld\n",
+			 path, IS_ERR(btf) ? PTR_ERR(btf) : 0);
+		if (IS_ERR(btf))
+			continue;
+
+		return btf;
+	}
+
+	pr_warn("failed to find valid kernel BTF\n");
+	return ERR_PTR(-ESRCH);
+}
diff --git a/tools/lib/bpf/btf.h b/tools/lib/bpf/btf.h
index 8d73f7f5551f..70c1b7ec2bd0 100644
--- a/tools/lib/bpf/btf.h
+++ b/tools/lib/bpf/btf.h
@@ -102,6 +102,8 @@ LIBBPF_API int btf_ext__reloc_line_info(const struct btf *btf,
 LIBBPF_API __u32 btf_ext__func_info_rec_size(const struct btf_ext *btf_ext);
 LIBBPF_API __u32 btf_ext__line_info_rec_size(const struct btf_ext *btf_ext);
 
+LIBBPF_API struct btf *libbpf_find_kernel_btf(void);
+
 struct btf_dedup_opts {
 	unsigned int dedup_table_size;
 	bool dont_resolve_fwds;
diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 23868883477f..3afaca9bce1d 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -73,7 +73,6 @@
 
 #define __printf(a, b)	__attribute__((format(printf, a, b)))
 
-static struct btf *bpf_find_kernel_btf(void);
 static struct bpf_map *bpf_object__add_map(struct bpf_object *obj);
 static struct bpf_program *bpf_object__find_prog_by_idx(struct bpf_object *obj,
 							int idx);
@@ -848,7 +847,7 @@ static int bpf_object__init_kern_struct_ops_maps(struct bpf_object *obj)
 			continue;
 
 		if (!kern_btf) {
-			kern_btf = bpf_find_kernel_btf();
+			kern_btf = libbpf_find_kernel_btf();
 			if (IS_ERR(kern_btf))
 				return PTR_ERR(kern_btf);
 		}
@@ -4300,92 +4299,6 @@ static int bpf_core_reloc_insn(struct bpf_program *prog,
 	return 0;
 }
 
-static struct btf *btf_load_raw(const char *path)
-{
-	struct btf *btf;
-	size_t read_cnt;
-	struct stat st;
-	void *data;
-	FILE *f;
-
-	if (stat(path, &st))
-		return ERR_PTR(-errno);
-
-	data = malloc(st.st_size);
-	if (!data)
-		return ERR_PTR(-ENOMEM);
-
-	f = fopen(path, "rb");
-	if (!f) {
-		btf = ERR_PTR(-errno);
-		goto cleanup;
-	}
-
-	read_cnt = fread(data, 1, st.st_size, f);
-	fclose(f);
-	if (read_cnt < st.st_size) {
-		btf = ERR_PTR(-EBADF);
-		goto cleanup;
-	}
-
-	btf = btf__new(data, read_cnt);
-
-cleanup:
-	free(data);
-	return btf;
-}
-
-/*
- * Probe few well-known locations for vmlinux kernel image and try to load BTF
- * data out of it to use for target BTF.
- */
-static struct btf *bpf_find_kernel_btf(void)
-{
-	struct {
-		const char *path_fmt;
-		bool raw_btf;
-	} locations[] = {
-		/* try canonical vmlinux BTF through sysfs first */
-		{ "/sys/kernel/btf/vmlinux", true /* raw BTF */ },
-		/* fall back to trying to find vmlinux ELF on disk otherwise */
-		{ "/boot/vmlinux-%1$s" },
-		{ "/lib/modules/%1$s/vmlinux-%1$s" },
-		{ "/lib/modules/%1$s/build/vmlinux" },
-		{ "/usr/lib/modules/%1$s/kernel/vmlinux" },
-		{ "/usr/lib/debug/boot/vmlinux-%1$s" },
-		{ "/usr/lib/debug/boot/vmlinux-%1$s.debug" },
-		{ "/usr/lib/debug/lib/modules/%1$s/vmlinux" },
-	};
-	char path[PATH_MAX + 1];
-	struct utsname buf;
-	struct btf *btf;
-	int i;
-
-	uname(&buf);
-
-	for (i = 0; i < ARRAY_SIZE(locations); i++) {
-		snprintf(path, PATH_MAX, locations[i].path_fmt, buf.release);
-
-		if (access(path, R_OK))
-			continue;
-
-		if (locations[i].raw_btf)
-			btf = btf_load_raw(path);
-		else
-			btf = btf__parse_elf(path, NULL);
-
-		pr_debug("loading kernel BTF '%s': %ld\n",
-			 path, IS_ERR(btf) ? PTR_ERR(btf) : 0);
-		if (IS_ERR(btf))
-			continue;
-
-		return btf;
-	}
-
-	pr_warn("failed to find valid kernel BTF\n");
-	return ERR_PTR(-ESRCH);
-}
-
 /* Output spec definition in the format:
  * [<type-id>] (<type-name>) + <raw-spec> => <offset>@<spec>,
  * where <spec> is a C-syntax view of recorded field access, e.g.: x.a[3].b
@@ -4620,7 +4533,7 @@ bpf_core_reloc_fields(struct bpf_object *obj, const char *targ_btf_path)
 	if (targ_btf_path)
 		targ_btf = btf__parse_elf(targ_btf_path, NULL);
 	else
-		targ_btf = bpf_find_kernel_btf();
+		targ_btf = libbpf_find_kernel_btf();
 	if (IS_ERR(targ_btf)) {
 		pr_warn("failed to get target BTF: %ld\n", PTR_ERR(targ_btf));
 		return PTR_ERR(targ_btf);
@@ -6595,7 +6508,7 @@ invalid_prog:
 int libbpf_find_vmlinux_btf_id(const char *name,
 			       enum bpf_attach_type attach_type)
 {
-	struct btf *btf = bpf_find_kernel_btf();
+	struct btf *btf = libbpf_find_kernel_btf();
 	char raw_tp_btf[128] = BTF_PREFIX;
 	char *dst = raw_tp_btf + sizeof(BTF_PREFIX) - 1;
 	const char *btf_name;
diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
index 1902a0fc6afc..64ec71ba41f1 100644
--- a/tools/lib/bpf/libbpf.map
+++ b/tools/lib/bpf/libbpf.map
@@ -231,4 +231,5 @@ LIBBPF_0.0.7 {
 		bpf_program__is_struct_ops;
 		bpf_program__set_struct_ops;
 		btf__align_of;
+		libbpf_find_kernel_btf;
 } LIBBPF_0.0.6;
-- 
cgit v1.2.3-59-g8ed1b


From 84c72ceee91c166e19b506d618dbe9df6eacce97 Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <kafai@fb.com>
Date: Wed, 15 Jan 2020 15:00:37 -0800
Subject: bpftool: Add struct_ops map name

This patch adds BPF_MAP_TYPE_STRUCT_OPS to "struct_ops" name mapping
so that "bpftool map show" can print the "struct_ops" map type
properly.

[root@arch-fb-vm1 bpf]# ~/devshare/fb-kernel/linux/tools/bpf/bpftool/bpftool map show id 8
8: struct_ops  name dctcp  flags 0x0
	key 4B  value 256B  max_entries 1  memlock 4096B
	btf_id 7

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/20200115230037.1102674-1-kafai@fb.com
---
 tools/bpf/bpftool/map.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tools/bpf/bpftool/map.c b/tools/bpf/bpftool/map.c
index 45c1eda6512c..4c5b15d736b6 100644
--- a/tools/bpf/bpftool/map.c
+++ b/tools/bpf/bpftool/map.c
@@ -48,6 +48,7 @@ const char * const map_type_name[] = {
 	[BPF_MAP_TYPE_QUEUE]			= "queue",
 	[BPF_MAP_TYPE_STACK]			= "stack",
 	[BPF_MAP_TYPE_SK_STORAGE]		= "sk_storage",
+	[BPF_MAP_TYPE_STRUCT_OPS]		= "struct_ops",
 };
 
 const size_t map_type_name_size = ARRAY_SIZE(map_type_name);
-- 
cgit v1.2.3-59-g8ed1b


From 4e1ea33292ff2ea0a82f52b2600abb4aba752094 Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <kafai@fb.com>
Date: Wed, 15 Jan 2020 15:00:44 -0800
Subject: bpftool: Support dumping a map with btf_vmlinux_value_type_id

This patch makes bpftool support dumping a map's value properly
when the map's value type is a type of the running kernel's btf.
(i.e. map_info.btf_vmlinux_value_type_id is set instead of
map_info.btf_value_type_id).  The first usecase is for the
BPF_MAP_TYPE_STRUCT_OPS.

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/20200115230044.1103008-1-kafai@fb.com
---
 tools/bpf/bpftool/map.c | 61 ++++++++++++++++++++++++++++++++++++++++---------
 1 file changed, 50 insertions(+), 11 deletions(-)

diff --git a/tools/bpf/bpftool/map.c b/tools/bpf/bpftool/map.c
index 4c5b15d736b6..86f8ab0b7e63 100644
--- a/tools/bpf/bpftool/map.c
+++ b/tools/bpf/bpftool/map.c
@@ -252,6 +252,7 @@ static int do_dump_btf(const struct btf_dumper *d,
 		       struct bpf_map_info *map_info, void *key,
 		       void *value)
 {
+	__u32 value_id;
 	int ret;
 
 	/* start of key-value pair */
@@ -265,9 +266,12 @@ static int do_dump_btf(const struct btf_dumper *d,
 			goto err_end_obj;
 	}
 
+	value_id = map_info->btf_vmlinux_value_type_id ?
+		: map_info->btf_value_type_id;
+
 	if (!map_is_per_cpu(map_info->type)) {
 		jsonw_name(d->jw, "value");
-		ret = btf_dumper_type(d, map_info->btf_value_type_id, value);
+		ret = btf_dumper_type(d, value_id, value);
 	} else {
 		unsigned int i, n, step;
 
@@ -279,8 +283,7 @@ static int do_dump_btf(const struct btf_dumper *d,
 			jsonw_start_object(d->jw);
 			jsonw_int_field(d->jw, "cpu", i);
 			jsonw_name(d->jw, "value");
-			ret = btf_dumper_type(d, map_info->btf_value_type_id,
-					      value + i * step);
+			ret = btf_dumper_type(d, value_id, value + i * step);
 			jsonw_end_object(d->jw);
 			if (ret)
 				break;
@@ -932,6 +935,44 @@ static int maps_have_btf(int *fds, int nb_fds)
 	return 1;
 }
 
+static struct btf *btf_vmlinux;
+
+static struct btf *get_map_kv_btf(const struct bpf_map_info *info)
+{
+	struct btf *btf = NULL;
+
+	if (info->btf_vmlinux_value_type_id) {
+		if (!btf_vmlinux) {
+			btf_vmlinux = libbpf_find_kernel_btf();
+			if (IS_ERR(btf_vmlinux))
+				p_err("failed to get kernel btf");
+		}
+		return btf_vmlinux;
+	} else if (info->btf_value_type_id) {
+		int err;
+
+		err = btf__get_from_id(info->btf_id, &btf);
+		if (err || !btf) {
+			p_err("failed to get btf");
+			btf = err ? ERR_PTR(err) : ERR_PTR(-ESRCH);
+		}
+	}
+
+	return btf;
+}
+
+static void free_map_kv_btf(struct btf *btf)
+{
+	if (!IS_ERR(btf) && btf != btf_vmlinux)
+		btf__free(btf);
+}
+
+static void free_btf_vmlinux(void)
+{
+	if (!IS_ERR(btf_vmlinux))
+		btf__free(btf_vmlinux);
+}
+
 static int
 map_dump(int fd, struct bpf_map_info *info, json_writer_t *wtr,
 	 bool show_header)
@@ -952,13 +993,10 @@ map_dump(int fd, struct bpf_map_info *info, json_writer_t *wtr,
 	prev_key = NULL;
 
 	if (wtr) {
-		if (info->btf_id) {
-			err = btf__get_from_id(info->btf_id, &btf);
-			if (err || !btf) {
-				err = err ? : -ESRCH;
-				p_err("failed to get btf");
-				goto exit_free;
-			}
+		btf = get_map_kv_btf(info);
+		if (IS_ERR(btf)) {
+			err = PTR_ERR(btf);
+			goto exit_free;
 		}
 
 		if (show_header) {
@@ -999,7 +1037,7 @@ exit_free:
 	free(key);
 	free(value);
 	close(fd);
-	btf__free(btf);
+	free_map_kv_btf(btf);
 
 	return err;
 }
@@ -1067,6 +1105,7 @@ exit_close:
 		close(fds[i]);
 exit_free:
 	free(fds);
+	free_btf_vmlinux();
 	return err;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From b65053cd94f46619b4aae746b98f2d8d9274540e Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Wed, 15 Jan 2020 16:55:49 -0800
Subject: selftests/bpf: Add whitelist/blacklist of test names to test_progs

Add ability to specify a list of test name substrings for selecting which
tests to run. So now -t is accepting a comma-separated list of strings,
similarly to how -n accepts a comma-separated list of test numbers.

Additionally, add ability to blacklist tests by name. Blacklist takes
precedence over whitelist. Blacklisting is important for cases where it's
known that some tests can't pass (e.g., due to perf hardware events that are
not available within VM). This is going to be used for libbpf testing in
Travis CI in its Github repo.

Example runs with just whitelist and whitelist + blacklist:

  $ sudo ./test_progs -tattach,core/existence
  #1 attach_probe:OK
  #6 cgroup_attach_autodetach:OK
  #7 cgroup_attach_multi:OK
  #8 cgroup_attach_override:OK
  #9 core_extern:OK
  #10/44 existence:OK
  #10/45 existence___minimal:OK
  #10/46 existence__err_int_sz:OK
  #10/47 existence__err_int_type:OK
  #10/48 existence__err_int_kind:OK
  #10/49 existence__err_arr_kind:OK
  #10/50 existence__err_arr_value_type:OK
  #10/51 existence__err_struct_type:OK
  #10 core_reloc:OK
  #19 flow_dissector_reattach:OK
  #60 tp_attach_query:OK
  Summary: 8/8 PASSED, 0 SKIPPED, 0 FAILED

  $ sudo ./test_progs -tattach,core/existence -bcgroup,flow/arr
  #1 attach_probe:OK
  #9 core_extern:OK
  #10/44 existence:OK
  #10/45 existence___minimal:OK
  #10/46 existence__err_int_sz:OK
  #10/47 existence__err_int_type:OK
  #10/48 existence__err_int_kind:OK
  #10/51 existence__err_struct_type:OK
  #10 core_reloc:OK
  #60 tp_attach_query:OK
  Summary: 4/6 PASSED, 0 SKIPPED, 0 FAILED

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Cc: Julia Kartseva <hex@fb.com>
Link: https://lore.kernel.org/bpf/20200116005549.3644118-1-andriin@fb.com
---
 tools/testing/selftests/bpf/test_progs.c | 83 ++++++++++++++++++++++++++++----
 tools/testing/selftests/bpf/test_progs.h |  8 ++-
 2 files changed, 80 insertions(+), 11 deletions(-)

diff --git a/tools/testing/selftests/bpf/test_progs.c b/tools/testing/selftests/bpf/test_progs.c
index 7fa7d08a8104..bab1e6f1d8f1 100644
--- a/tools/testing/selftests/bpf/test_progs.c
+++ b/tools/testing/selftests/bpf/test_progs.c
@@ -8,7 +8,7 @@
 #include <string.h>
 
 /* defined in test_progs.h */
-struct test_env env;
+struct test_env env = {};
 
 struct prog_test_def {
 	const char *test_name;
@@ -29,10 +29,19 @@ struct prog_test_def {
 
 static bool should_run(struct test_selector *sel, int num, const char *name)
 {
-	if (sel->name && sel->name[0] && !strstr(name, sel->name))
-		return false;
+	int i;
+
+	for (i = 0; i < sel->blacklist.cnt; i++) {
+		if (strstr(name, sel->blacklist.strs[i]))
+			return false;
+	}
 
-	if (!sel->num_set)
+	for (i = 0; i < sel->whitelist.cnt; i++) {
+		if (strstr(name, sel->whitelist.strs[i]))
+			return true;
+	}
+
+	if (!sel->whitelist.cnt && !sel->num_set)
 		return true;
 
 	return num < sel->num_set_len && sel->num_set[num];
@@ -334,6 +343,7 @@ const char argp_program_doc[] = "BPF selftests test runner";
 enum ARG_KEYS {
 	ARG_TEST_NUM = 'n',
 	ARG_TEST_NAME = 't',
+	ARG_TEST_NAME_BLACKLIST = 'b',
 	ARG_VERIFIER_STATS = 's',
 	ARG_VERBOSE = 'v',
 };
@@ -341,8 +351,10 @@ enum ARG_KEYS {
 static const struct argp_option opts[] = {
 	{ "num", ARG_TEST_NUM, "NUM", 0,
 	  "Run test number NUM only " },
-	{ "name", ARG_TEST_NAME, "NAME", 0,
-	  "Run tests with names containing NAME" },
+	{ "name", ARG_TEST_NAME, "NAMES", 0,
+	  "Run tests with names containing any string from NAMES list" },
+	{ "name-blacklist", ARG_TEST_NAME_BLACKLIST, "NAMES", 0,
+	  "Don't run tests with names containing any string from NAMES list" },
 	{ "verifier-stats", ARG_VERIFIER_STATS, NULL, 0,
 	  "Output verifier statistics", },
 	{ "verbose", ARG_VERBOSE, "LEVEL", OPTION_ARG_OPTIONAL,
@@ -359,6 +371,41 @@ static int libbpf_print_fn(enum libbpf_print_level level,
 	return 0;
 }
 
+static int parse_str_list(const char *s, struct str_set *set)
+{
+	char *input, *state = NULL, *next, **tmp, **strs = NULL;
+	int cnt = 0;
+
+	input = strdup(s);
+	if (!input)
+		return -ENOMEM;
+
+	set->cnt = 0;
+	set->strs = NULL;
+
+	while ((next = strtok_r(state ? NULL : input, ",", &state))) {
+		tmp = realloc(strs, sizeof(*strs) * (cnt + 1));
+		if (!tmp)
+			goto err;
+		strs = tmp;
+
+		strs[cnt] = strdup(next);
+		if (!strs[cnt])
+			goto err;
+
+		cnt++;
+	}
+
+	set->cnt = cnt;
+	set->strs = (const char **)strs;
+	free(input);
+	return 0;
+err:
+	free(strs);
+	free(input);
+	return -ENOMEM;
+}
+
 int parse_num_list(const char *s, struct test_selector *sel)
 {
 	int i, set_len = 0, num, start = 0, end = -1;
@@ -449,12 +496,24 @@ static error_t parse_arg(int key, char *arg, struct argp_state *state)
 
 		if (subtest_str) {
 			*subtest_str = '\0';
-			env->subtest_selector.name = strdup(subtest_str + 1);
-			if (!env->subtest_selector.name)
+			if (parse_str_list(subtest_str + 1,
+					   &env->subtest_selector.whitelist))
+				return -ENOMEM;
+		}
+		if (parse_str_list(arg, &env->test_selector.whitelist))
+			return -ENOMEM;
+		break;
+	}
+	case ARG_TEST_NAME_BLACKLIST: {
+		char *subtest_str = strchr(arg, '/');
+
+		if (subtest_str) {
+			*subtest_str = '\0';
+			if (parse_str_list(subtest_str + 1,
+					   &env->subtest_selector.blacklist))
 				return -ENOMEM;
 		}
-		env->test_selector.name = strdup(arg);
-		if (!env->test_selector.name)
+		if (parse_str_list(arg, &env->test_selector.blacklist))
 			return -ENOMEM;
 		break;
 	}
@@ -617,7 +676,11 @@ int main(int argc, char **argv)
 	printf("Summary: %d/%d PASSED, %d SKIPPED, %d FAILED\n",
 	       env.succ_cnt, env.sub_succ_cnt, env.skip_cnt, env.fail_cnt);
 
+	free(env.test_selector.blacklist.strs);
+	free(env.test_selector.whitelist.strs);
 	free(env.test_selector.num_set);
+	free(env.subtest_selector.blacklist.strs);
+	free(env.subtest_selector.whitelist.strs);
 	free(env.subtest_selector.num_set);
 
 	return env.fail_cnt ? EXIT_FAILURE : EXIT_SUCCESS;
diff --git a/tools/testing/selftests/bpf/test_progs.h b/tools/testing/selftests/bpf/test_progs.h
index de1fdaa4e7b4..99933a1857ca 100644
--- a/tools/testing/selftests/bpf/test_progs.h
+++ b/tools/testing/selftests/bpf/test_progs.h
@@ -46,8 +46,14 @@ enum verbosity {
 	VERBOSE_SUPER,
 };
 
+struct str_set {
+	const char **strs;
+	int cnt;
+};
+
 struct test_selector {
-	const char *name;
+	struct str_set whitelist;
+	struct str_set blacklist;
 	bool *num_set;
 	int num_set_len;
 };
-- 
cgit v1.2.3-59-g8ed1b


From ae563183b647b3bdb47e8a78a5de879adf733735 Mon Sep 17 00:00:00 2001
From: Rocky Liao <rjliao@codeaurora.org>
Date: Thu, 16 Jan 2020 11:22:54 +0800
Subject: Bluetooth: hci_qca: Enable power off/on support during hci down/up
 for QCA Rome

This patch registers hdev->shutdown() callback and also sets
HCI_QUIRK_NON_PERSISTENT_SETUP for QCA Rome. It will power-off the BT chip
during hci down and power-on/initialize the chip again during hci up. As
wcn399x already enabled this, this patch also removed the callback register
and QUIRK setting in qca_setup() for wcn399x and uniformly do this in the
probe() routine.

Signed-off-by: Rocky Liao <rjliao@codeaurora.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 drivers/bluetooth/hci_qca.c | 20 +++++++++++---------
 1 file changed, 11 insertions(+), 9 deletions(-)

diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c
index 1139142e8eed..d6e0c99ee5eb 100644
--- a/drivers/bluetooth/hci_qca.c
+++ b/drivers/bluetooth/hci_qca.c
@@ -1569,12 +1569,7 @@ retry:
 		return ret;
 
 	if (qca_is_wcn399x(soc_type)) {
-		/* Enable NON_PERSISTENT_SETUP QUIRK to ensure to execute
-		 * setup for every hci up.
-		 */
-		set_bit(HCI_QUIRK_NON_PERSISTENT_SETUP, &hdev->quirks);
 		set_bit(HCI_QUIRK_USE_BDADDR_PROPERTY, &hdev->quirks);
-		hu->hdev->shutdown = qca_power_off;
 
 		ret = qca_read_soc_version(hdev, &soc_ver, soc_type);
 		if (ret)
@@ -1813,6 +1808,7 @@ static int qca_init_regulators(struct qca_power *qca,
 static int qca_serdev_probe(struct serdev_device *serdev)
 {
 	struct qca_serdev *qcadev;
+	struct hci_dev *hdev;
 	const struct qca_vreg_data *data;
 	int err;
 
@@ -1838,7 +1834,7 @@ static int qca_serdev_probe(struct serdev_device *serdev)
 					  data->num_vregs);
 		if (err) {
 			BT_ERR("Failed to init regulators:%d", err);
-			goto out;
+			return err;
 		}
 
 		qcadev->bt_power->vregs_on = false;
@@ -1851,7 +1847,7 @@ static int qca_serdev_probe(struct serdev_device *serdev)
 		err = hci_uart_register_device(&qcadev->serdev_hu, &qca_proto);
 		if (err) {
 			BT_ERR("wcn3990 serdev registration failed");
-			goto out;
+			return err;
 		}
 	} else {
 		qcadev->btsoc_type = QCA_ROME;
@@ -1877,12 +1873,18 @@ static int qca_serdev_probe(struct serdev_device *serdev)
 			return err;
 
 		err = hci_uart_register_device(&qcadev->serdev_hu, &qca_proto);
-		if (err)
+		if (err) {
+			BT_ERR("Rome serdev registration failed");
 			clk_disable_unprepare(qcadev->susclk);
+			return err;
+		}
 	}
 
-out:	return err;
+	hdev = qcadev->serdev_hu.hdev;
+	set_bit(HCI_QUIRK_NON_PERSISTENT_SETUP, &hdev->quirks);
+	hdev->shutdown = qca_power_off;
 
+	return 0;
 }
 
 static void qca_serdev_remove(struct serdev_device *serdev)
-- 
cgit v1.2.3-59-g8ed1b


From 117717e57440d2b5e07da40c621aa4f0ba423b80 Mon Sep 17 00:00:00 2001
From: Marcel Holtmann <marcel@holtmann.org>
Date: Wed, 15 Jan 2020 21:35:32 +0100
Subject: Bluetooth: Increment management interface revision

Increment the mgmt revision due to the recently added commands.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
---
 net/bluetooth/mgmt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c
index 339c762eb6fd..0dc610faab70 100644
--- a/net/bluetooth/mgmt.c
+++ b/net/bluetooth/mgmt.c
@@ -38,7 +38,7 @@
 #include "mgmt_util.h"
 
 #define MGMT_VERSION	1
-#define MGMT_REVISION	14
+#define MGMT_REVISION	15
 
 static const u16 mgmt_commands[] = {
 	MGMT_OP_READ_INDEX_LIST,
-- 
cgit v1.2.3-59-g8ed1b


From 6bc8038035267d12df2bf78a8e1a5f07069fabb8 Mon Sep 17 00:00:00 2001
From: YueHaibing <yuehaibing@huawei.com>
Date: Thu, 16 Jan 2020 02:26:57 +0000
Subject: sfc: remove duplicated include from efx.c

Remove duplicated include.

Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/efx.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index 110e485e6624..4481f21a1f43 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -27,7 +27,6 @@
 #include "efx_channels.h"
 #include "rx_common.h"
 #include "tx_common.h"
-#include "rx_common.h"
 #include "nic.h"
 #include "io.h"
 #include "selftest.h"
-- 
cgit v1.2.3-59-g8ed1b


From c320e527e1548305f31d95ec405140b04aed25f5 Mon Sep 17 00:00:00 2001
From: Moni Shoua <monis@mellanox.com>
Date: Wed, 15 Jan 2020 14:43:31 +0200
Subject: IB: Allow calls to ib_umem_get from kernel ULPs

So far the assumption was that ib_umem_get() and ib_umem_odp_get()
are called from flows that start in UVERBS and therefore has a user
context. This assumption restricts flows that are initiated by ULPs
and need the service that ib_umem_get() provides.

This patch changes ib_umem_get() and ib_umem_odp_get() to get IB device
directly by relying on the fact that both UVERBS and ULPs sets that
field correctly.

Reviewed-by: Guy Levi <guyle@mellanox.com>
Signed-off-by: Moni Shoua <monis@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
---
 drivers/infiniband/core/umem.c                | 27 +++++++++----------------
 drivers/infiniband/core/umem_odp.c            | 29 +++++++--------------------
 drivers/infiniband/hw/bnxt_re/ib_verbs.c      | 12 ++++++-----
 drivers/infiniband/hw/cxgb4/mem.c             |  2 +-
 drivers/infiniband/hw/efa/efa_verbs.c         |  2 +-
 drivers/infiniband/hw/hns/hns_roce_cq.c       |  2 +-
 drivers/infiniband/hw/hns/hns_roce_db.c       |  3 ++-
 drivers/infiniband/hw/hns/hns_roce_mr.c       |  4 ++--
 drivers/infiniband/hw/hns/hns_roce_qp.c       |  2 +-
 drivers/infiniband/hw/hns/hns_roce_srq.c      |  5 +++--
 drivers/infiniband/hw/i40iw/i40iw_verbs.c     |  2 +-
 drivers/infiniband/hw/mlx4/cq.c               |  2 +-
 drivers/infiniband/hw/mlx4/doorbell.c         |  3 ++-
 drivers/infiniband/hw/mlx4/mr.c               |  8 ++++----
 drivers/infiniband/hw/mlx4/qp.c               |  5 +++--
 drivers/infiniband/hw/mlx4/srq.c              |  3 ++-
 drivers/infiniband/hw/mlx5/cq.c               |  6 +++---
 drivers/infiniband/hw/mlx5/devx.c             |  2 +-
 drivers/infiniband/hw/mlx5/doorbell.c         |  3 ++-
 drivers/infiniband/hw/mlx5/mr.c               | 18 ++++++++---------
 drivers/infiniband/hw/mlx5/odp.c              |  2 +-
 drivers/infiniband/hw/mlx5/qp.c               |  4 ++--
 drivers/infiniband/hw/mlx5/srq.c              |  2 +-
 drivers/infiniband/hw/mthca/mthca_provider.c  |  2 +-
 drivers/infiniband/hw/ocrdma/ocrdma_verbs.c   |  2 +-
 drivers/infiniband/hw/qedr/verbs.c            |  9 ++++-----
 drivers/infiniband/hw/vmw_pvrdma/pvrdma_cq.c  |  2 +-
 drivers/infiniband/hw/vmw_pvrdma/pvrdma_mr.c  |  2 +-
 drivers/infiniband/hw/vmw_pvrdma/pvrdma_qp.c  |  7 ++++---
 drivers/infiniband/hw/vmw_pvrdma/pvrdma_srq.c |  2 +-
 drivers/infiniband/sw/rdmavt/mr.c             |  2 +-
 drivers/infiniband/sw/rxe/rxe_mr.c            |  2 +-
 include/rdma/ib_umem.h                        |  4 ++--
 include/rdma/ib_umem_odp.h                    |  6 +++---
 34 files changed, 85 insertions(+), 103 deletions(-)

diff --git a/drivers/infiniband/core/umem.c b/drivers/infiniband/core/umem.c
index 7a3b99597ead..146f98fbf22b 100644
--- a/drivers/infiniband/core/umem.c
+++ b/drivers/infiniband/core/umem.c
@@ -181,15 +181,14 @@ EXPORT_SYMBOL(ib_umem_find_best_pgsz);
 /**
  * ib_umem_get - Pin and DMA map userspace memory.
  *
- * @udata: userspace context to pin memory for
+ * @device: IB device to connect UMEM
  * @addr: userspace virtual address to start at
  * @size: length of region to pin
  * @access: IB_ACCESS_xxx flags for memory being pinned
  */
-struct ib_umem *ib_umem_get(struct ib_udata *udata, unsigned long addr,
+struct ib_umem *ib_umem_get(struct ib_device *device, unsigned long addr,
 			    size_t size, int access)
 {
-	struct ib_ucontext *context;
 	struct ib_umem *umem;
 	struct page **page_list;
 	unsigned long lock_limit;
@@ -201,14 +200,6 @@ struct ib_umem *ib_umem_get(struct ib_udata *udata, unsigned long addr,
 	struct scatterlist *sg;
 	unsigned int gup_flags = FOLL_WRITE;
 
-	if (!udata)
-		return ERR_PTR(-EIO);
-
-	context = container_of(udata, struct uverbs_attr_bundle, driver_udata)
-			  ->context;
-	if (!context)
-		return ERR_PTR(-EIO);
-
 	/*
 	 * If the combination of the addr and size requested for this memory
 	 * region causes an integer overflow, return error.
@@ -226,7 +217,7 @@ struct ib_umem *ib_umem_get(struct ib_udata *udata, unsigned long addr,
 	umem = kzalloc(sizeof(*umem), GFP_KERNEL);
 	if (!umem)
 		return ERR_PTR(-ENOMEM);
-	umem->ibdev = context->device;
+	umem->ibdev      = device;
 	umem->length     = size;
 	umem->address    = addr;
 	umem->writable   = ib_access_writable(access);
@@ -281,7 +272,7 @@ struct ib_umem *ib_umem_get(struct ib_udata *udata, unsigned long addr,
 		npages   -= ret;
 
 		sg = ib_umem_add_sg_table(sg, page_list, ret,
-			dma_get_max_seg_size(context->device->dma_device),
+			dma_get_max_seg_size(device->dma_device),
 			&umem->sg_nents);
 
 		up_read(&mm->mmap_sem);
@@ -289,10 +280,10 @@ struct ib_umem *ib_umem_get(struct ib_udata *udata, unsigned long addr,
 
 	sg_mark_end(sg);
 
-	umem->nmap = ib_dma_map_sg(context->device,
-				  umem->sg_head.sgl,
-				  umem->sg_nents,
-				  DMA_BIDIRECTIONAL);
+	umem->nmap = ib_dma_map_sg(device,
+				   umem->sg_head.sgl,
+				   umem->sg_nents,
+				   DMA_BIDIRECTIONAL);
 
 	if (!umem->nmap) {
 		ret = -ENOMEM;
@@ -303,7 +294,7 @@ struct ib_umem *ib_umem_get(struct ib_udata *udata, unsigned long addr,
 	goto out;
 
 umem_release:
-	__ib_umem_release(context->device, umem, 0);
+	__ib_umem_release(device, umem, 0);
 vma:
 	atomic64_sub(ib_umem_num_pages(umem), &mm->pinned_vm);
 out:
diff --git a/drivers/infiniband/core/umem_odp.c b/drivers/infiniband/core/umem_odp.c
index e42d44e501fd..dac3fd2ebc26 100644
--- a/drivers/infiniband/core/umem_odp.c
+++ b/drivers/infiniband/core/umem_odp.c
@@ -110,15 +110,12 @@ out_page_list:
  * They exist only to hold the per_mm reference to help the driver create
  * children umems.
  *
- * @udata: udata from the syscall being used to create the umem
+ * @device: IB device to create UMEM
  * @access: ib_reg_mr access flags
  */
-struct ib_umem_odp *ib_umem_odp_alloc_implicit(struct ib_udata *udata,
+struct ib_umem_odp *ib_umem_odp_alloc_implicit(struct ib_device *device,
 					       int access)
 {
-	struct ib_ucontext *context =
-		container_of(udata, struct uverbs_attr_bundle, driver_udata)
-			->context;
 	struct ib_umem *umem;
 	struct ib_umem_odp *umem_odp;
 	int ret;
@@ -126,14 +123,11 @@ struct ib_umem_odp *ib_umem_odp_alloc_implicit(struct ib_udata *udata,
 	if (access & IB_ACCESS_HUGETLB)
 		return ERR_PTR(-EINVAL);
 
-	if (!context)
-		return ERR_PTR(-EIO);
-
 	umem_odp = kzalloc(sizeof(*umem_odp), GFP_KERNEL);
 	if (!umem_odp)
 		return ERR_PTR(-ENOMEM);
 	umem = &umem_odp->umem;
-	umem->ibdev = context->device;
+	umem->ibdev = device;
 	umem->writable = ib_access_writable(access);
 	umem->owning_mm = current->mm;
 	umem_odp->is_implicit_odp = 1;
@@ -201,7 +195,7 @@ EXPORT_SYMBOL(ib_umem_odp_alloc_child);
 /**
  * ib_umem_odp_get - Create a umem_odp for a userspace va
  *
- * @udata: userspace context to pin memory for
+ * @device: IB device struct to get UMEM
  * @addr: userspace virtual address to start at
  * @size: length of region to pin
  * @access: IB_ACCESS_xxx flags for memory being pinned
@@ -210,23 +204,14 @@ EXPORT_SYMBOL(ib_umem_odp_alloc_child);
  * pinning, instead, stores the mm for future page fault handling in
  * conjunction with MMU notifiers.
  */
-struct ib_umem_odp *ib_umem_odp_get(struct ib_udata *udata, unsigned long addr,
-				    size_t size, int access,
+struct ib_umem_odp *ib_umem_odp_get(struct ib_device *device,
+				    unsigned long addr, size_t size, int access,
 				    const struct mmu_interval_notifier_ops *ops)
 {
 	struct ib_umem_odp *umem_odp;
-	struct ib_ucontext *context;
 	struct mm_struct *mm;
 	int ret;
 
-	if (!udata)
-		return ERR_PTR(-EIO);
-
-	context = container_of(udata, struct uverbs_attr_bundle, driver_udata)
-			  ->context;
-	if (!context)
-		return ERR_PTR(-EIO);
-
 	if (WARN_ON_ONCE(!(access & IB_ACCESS_ON_DEMAND)))
 		return ERR_PTR(-EINVAL);
 
@@ -234,7 +219,7 @@ struct ib_umem_odp *ib_umem_odp_get(struct ib_udata *udata, unsigned long addr,
 	if (!umem_odp)
 		return ERR_PTR(-ENOMEM);
 
-	umem_odp->umem.ibdev = context->device;
+	umem_odp->umem.ibdev = device;
 	umem_odp->umem.length = size;
 	umem_odp->umem.address = addr;
 	umem_odp->umem.writable = ib_access_writable(access);
diff --git a/drivers/infiniband/hw/bnxt_re/ib_verbs.c b/drivers/infiniband/hw/bnxt_re/ib_verbs.c
index ad5112a2325f..52b6a4d85460 100644
--- a/drivers/infiniband/hw/bnxt_re/ib_verbs.c
+++ b/drivers/infiniband/hw/bnxt_re/ib_verbs.c
@@ -837,7 +837,8 @@ static int bnxt_re_init_user_qp(struct bnxt_re_dev *rdev, struct bnxt_re_pd *pd,
 		bytes += (qplib_qp->sq.max_wqe * psn_sz);
 	}
 	bytes = PAGE_ALIGN(bytes);
-	umem = ib_umem_get(udata, ureq.qpsva, bytes, IB_ACCESS_LOCAL_WRITE);
+	umem = ib_umem_get(&rdev->ibdev, ureq.qpsva, bytes,
+			   IB_ACCESS_LOCAL_WRITE);
 	if (IS_ERR(umem))
 		return PTR_ERR(umem);
 
@@ -850,7 +851,7 @@ static int bnxt_re_init_user_qp(struct bnxt_re_dev *rdev, struct bnxt_re_pd *pd,
 	if (!qp->qplib_qp.srq) {
 		bytes = (qplib_qp->rq.max_wqe * BNXT_QPLIB_MAX_RQE_ENTRY_SIZE);
 		bytes = PAGE_ALIGN(bytes);
-		umem = ib_umem_get(udata, ureq.qprva, bytes,
+		umem = ib_umem_get(&rdev->ibdev, ureq.qprva, bytes,
 				   IB_ACCESS_LOCAL_WRITE);
 		if (IS_ERR(umem))
 			goto rqfail;
@@ -1304,7 +1305,8 @@ static int bnxt_re_init_user_srq(struct bnxt_re_dev *rdev,
 
 	bytes = (qplib_srq->max_wqe * BNXT_QPLIB_MAX_RQE_ENTRY_SIZE);
 	bytes = PAGE_ALIGN(bytes);
-	umem = ib_umem_get(udata, ureq.srqva, bytes, IB_ACCESS_LOCAL_WRITE);
+	umem = ib_umem_get(&rdev->ibdev, ureq.srqva, bytes,
+			   IB_ACCESS_LOCAL_WRITE);
 	if (IS_ERR(umem))
 		return PTR_ERR(umem);
 
@@ -2545,7 +2547,7 @@ int bnxt_re_create_cq(struct ib_cq *ibcq, const struct ib_cq_init_attr *attr,
 			goto fail;
 		}
 
-		cq->umem = ib_umem_get(udata, req.cq_va,
+		cq->umem = ib_umem_get(&rdev->ibdev, req.cq_va,
 				       entries * sizeof(struct cq_base),
 				       IB_ACCESS_LOCAL_WRITE);
 		if (IS_ERR(cq->umem)) {
@@ -3514,7 +3516,7 @@ struct ib_mr *bnxt_re_reg_user_mr(struct ib_pd *ib_pd, u64 start, u64 length,
 	/* The fixed portion of the rkey is the same as the lkey */
 	mr->ib_mr.rkey = mr->qplib_mr.rkey;
 
-	umem = ib_umem_get(udata, start, length, mr_access_flags);
+	umem = ib_umem_get(&rdev->ibdev, start, length, mr_access_flags);
 	if (IS_ERR(umem)) {
 		dev_err(rdev_to_dev(rdev), "Failed to get umem");
 		rc = -EFAULT;
diff --git a/drivers/infiniband/hw/cxgb4/mem.c b/drivers/infiniband/hw/cxgb4/mem.c
index fe3a7e8561df..962dc97a8ff2 100644
--- a/drivers/infiniband/hw/cxgb4/mem.c
+++ b/drivers/infiniband/hw/cxgb4/mem.c
@@ -543,7 +543,7 @@ struct ib_mr *c4iw_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
 
 	mhp->rhp = rhp;
 
-	mhp->umem = ib_umem_get(udata, start, length, acc);
+	mhp->umem = ib_umem_get(pd->device, start, length, acc);
 	if (IS_ERR(mhp->umem))
 		goto err_free_skb;
 
diff --git a/drivers/infiniband/hw/efa/efa_verbs.c b/drivers/infiniband/hw/efa/efa_verbs.c
index 50c22575aed6..5c35aa72f515 100644
--- a/drivers/infiniband/hw/efa/efa_verbs.c
+++ b/drivers/infiniband/hw/efa/efa_verbs.c
@@ -1384,7 +1384,7 @@ struct ib_mr *efa_reg_mr(struct ib_pd *ibpd, u64 start, u64 length,
 		goto err_out;
 	}
 
-	mr->umem = ib_umem_get(udata, start, length, access_flags);
+	mr->umem = ib_umem_get(ibpd->device, start, length, access_flags);
 	if (IS_ERR(mr->umem)) {
 		err = PTR_ERR(mr->umem);
 		ibdev_dbg(&dev->ibdev,
diff --git a/drivers/infiniband/hw/hns/hns_roce_cq.c b/drivers/infiniband/hw/hns/hns_roce_cq.c
index af1d8823b3f0..a2d1e5331bf1 100644
--- a/drivers/infiniband/hw/hns/hns_roce_cq.c
+++ b/drivers/infiniband/hw/hns/hns_roce_cq.c
@@ -163,7 +163,7 @@ static int get_cq_umem(struct hns_roce_dev *hr_dev, struct hns_roce_cq *hr_cq,
 	u32 npages;
 	int ret;
 
-	*umem = ib_umem_get(udata, ucmd.buf_addr, buf->size,
+	*umem = ib_umem_get(&hr_dev->ib_dev, ucmd.buf_addr, buf->size,
 			    IB_ACCESS_LOCAL_WRITE);
 	if (IS_ERR(*umem))
 		return PTR_ERR(*umem);
diff --git a/drivers/infiniband/hw/hns/hns_roce_db.c b/drivers/infiniband/hw/hns/hns_roce_db.c
index 10af6958ab69..bff6abdccfb0 100644
--- a/drivers/infiniband/hw/hns/hns_roce_db.c
+++ b/drivers/infiniband/hw/hns/hns_roce_db.c
@@ -31,7 +31,8 @@ int hns_roce_db_map_user(struct hns_roce_ucontext *context,
 
 	refcount_set(&page->refcount, 1);
 	page->user_virt = page_addr;
-	page->umem = ib_umem_get(udata, page_addr, PAGE_SIZE, 0);
+	page->umem = ib_umem_get(context->ibucontext.device, page_addr,
+				 PAGE_SIZE, 0);
 	if (IS_ERR(page->umem)) {
 		ret = PTR_ERR(page->umem);
 		kfree(page);
diff --git a/drivers/infiniband/hw/hns/hns_roce_mr.c b/drivers/infiniband/hw/hns/hns_roce_mr.c
index 9ad19170c3f9..3ff610549c74 100644
--- a/drivers/infiniband/hw/hns/hns_roce_mr.c
+++ b/drivers/infiniband/hw/hns/hns_roce_mr.c
@@ -1145,7 +1145,7 @@ struct ib_mr *hns_roce_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
 	if (!mr)
 		return ERR_PTR(-ENOMEM);
 
-	mr->umem = ib_umem_get(udata, start, length, access_flags);
+	mr->umem = ib_umem_get(pd->device, start, length, access_flags);
 	if (IS_ERR(mr->umem)) {
 		ret = PTR_ERR(mr->umem);
 		goto err_free;
@@ -1230,7 +1230,7 @@ static int rereg_mr_trans(struct ib_mr *ibmr, int flags,
 	}
 	ib_umem_release(mr->umem);
 
-	mr->umem = ib_umem_get(udata, start, length, mr_access_flags);
+	mr->umem = ib_umem_get(ibmr->device, start, length, mr_access_flags);
 	if (IS_ERR(mr->umem)) {
 		ret = PTR_ERR(mr->umem);
 		mr->umem = NULL;
diff --git a/drivers/infiniband/hw/hns/hns_roce_qp.c b/drivers/infiniband/hw/hns/hns_roce_qp.c
index a6565b674801..eb2ee6a581aa 100644
--- a/drivers/infiniband/hw/hns/hns_roce_qp.c
+++ b/drivers/infiniband/hw/hns/hns_roce_qp.c
@@ -744,7 +744,7 @@ static int hns_roce_create_qp_common(struct hns_roce_dev *hr_dev,
 			goto err_alloc_rq_inline_buf;
 		}
 
-		hr_qp->umem = ib_umem_get(udata, ucmd.buf_addr,
+		hr_qp->umem = ib_umem_get(ib_pd->device, ucmd.buf_addr,
 					  hr_qp->buff_size, 0);
 		if (IS_ERR(hr_qp->umem)) {
 			dev_err(dev, "ib_umem_get error for create qp\n");
diff --git a/drivers/infiniband/hw/hns/hns_roce_srq.c b/drivers/infiniband/hw/hns/hns_roce_srq.c
index 7113ebfdb4f0..c6d5f06f9cde 100644
--- a/drivers/infiniband/hw/hns/hns_roce_srq.c
+++ b/drivers/infiniband/hw/hns/hns_roce_srq.c
@@ -186,7 +186,8 @@ static int create_user_srq(struct hns_roce_srq *srq, struct ib_udata *udata,
 	if (ib_copy_from_udata(&ucmd, udata, sizeof(ucmd)))
 		return -EFAULT;
 
-	srq->umem = ib_umem_get(udata, ucmd.buf_addr, srq_buf_size, 0);
+	srq->umem =
+		ib_umem_get(srq->ibsrq.device, ucmd.buf_addr, srq_buf_size, 0);
 	if (IS_ERR(srq->umem))
 		return PTR_ERR(srq->umem);
 
@@ -205,7 +206,7 @@ static int create_user_srq(struct hns_roce_srq *srq, struct ib_udata *udata,
 		goto err_user_srq_mtt;
 
 	/* config index queue BA */
-	srq->idx_que.umem = ib_umem_get(udata, ucmd.que_addr,
+	srq->idx_que.umem = ib_umem_get(srq->ibsrq.device, ucmd.que_addr,
 					srq->idx_que.buf_size, 0);
 	if (IS_ERR(srq->idx_que.umem)) {
 		dev_err(hr_dev->dev, "ib_umem_get error for index queue\n");
diff --git a/drivers/infiniband/hw/i40iw/i40iw_verbs.c b/drivers/infiniband/hw/i40iw/i40iw_verbs.c
index dbd96d029d8b..96488fb443eb 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_verbs.c
+++ b/drivers/infiniband/hw/i40iw/i40iw_verbs.c
@@ -1761,7 +1761,7 @@ static struct ib_mr *i40iw_reg_user_mr(struct ib_pd *pd,
 
 	if (length > I40IW_MAX_MR_SIZE)
 		return ERR_PTR(-EINVAL);
-	region = ib_umem_get(udata, start, length, acc);
+	region = ib_umem_get(pd->device, start, length, acc);
 	if (IS_ERR(region))
 		return (struct ib_mr *)region;
 
diff --git a/drivers/infiniband/hw/mlx4/cq.c b/drivers/infiniband/hw/mlx4/cq.c
index 306b21281fa2..a57033d4b0e5 100644
--- a/drivers/infiniband/hw/mlx4/cq.c
+++ b/drivers/infiniband/hw/mlx4/cq.c
@@ -144,7 +144,7 @@ static int mlx4_ib_get_cq_umem(struct mlx4_ib_dev *dev, struct ib_udata *udata,
 	int shift;
 	int n;
 
-	*umem = ib_umem_get(udata, buf_addr, cqe * cqe_size,
+	*umem = ib_umem_get(&dev->ib_dev, buf_addr, cqe * cqe_size,
 			    IB_ACCESS_LOCAL_WRITE);
 	if (IS_ERR(*umem))
 		return PTR_ERR(*umem);
diff --git a/drivers/infiniband/hw/mlx4/doorbell.c b/drivers/infiniband/hw/mlx4/doorbell.c
index 714f9df5bf39..d41f03ccb0e1 100644
--- a/drivers/infiniband/hw/mlx4/doorbell.c
+++ b/drivers/infiniband/hw/mlx4/doorbell.c
@@ -64,7 +64,8 @@ int mlx4_ib_db_map_user(struct ib_udata *udata, unsigned long virt,
 
 	page->user_virt = (virt & PAGE_MASK);
 	page->refcnt    = 0;
-	page->umem = ib_umem_get(udata, virt & PAGE_MASK, PAGE_SIZE, 0);
+	page->umem = ib_umem_get(context->ibucontext.device, virt & PAGE_MASK,
+				 PAGE_SIZE, 0);
 	if (IS_ERR(page->umem)) {
 		err = PTR_ERR(page->umem);
 		kfree(page);
diff --git a/drivers/infiniband/hw/mlx4/mr.c b/drivers/infiniband/hw/mlx4/mr.c
index dfa17bcdcdbc..b0121c90c561 100644
--- a/drivers/infiniband/hw/mlx4/mr.c
+++ b/drivers/infiniband/hw/mlx4/mr.c
@@ -367,7 +367,7 @@ end:
 	return block_shift;
 }
 
-static struct ib_umem *mlx4_get_umem_mr(struct ib_udata *udata, u64 start,
+static struct ib_umem *mlx4_get_umem_mr(struct ib_device *device, u64 start,
 					u64 length, int access_flags)
 {
 	/*
@@ -398,7 +398,7 @@ static struct ib_umem *mlx4_get_umem_mr(struct ib_udata *udata, u64 start,
 		up_read(&current->mm->mmap_sem);
 	}
 
-	return ib_umem_get(udata, start, length, access_flags);
+	return ib_umem_get(device, start, length, access_flags);
 }
 
 struct ib_mr *mlx4_ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
@@ -415,7 +415,7 @@ struct ib_mr *mlx4_ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
 	if (!mr)
 		return ERR_PTR(-ENOMEM);
 
-	mr->umem = mlx4_get_umem_mr(udata, start, length, access_flags);
+	mr->umem = mlx4_get_umem_mr(pd->device, start, length, access_flags);
 	if (IS_ERR(mr->umem)) {
 		err = PTR_ERR(mr->umem);
 		goto err_free;
@@ -504,7 +504,7 @@ int mlx4_ib_rereg_user_mr(struct ib_mr *mr, int flags,
 
 		mlx4_mr_rereg_mem_cleanup(dev->dev, &mmr->mmr);
 		ib_umem_release(mmr->umem);
-		mmr->umem = mlx4_get_umem_mr(udata, start, length,
+		mmr->umem = mlx4_get_umem_mr(mr->device, start, length,
 					     mr_access_flags);
 		if (IS_ERR(mmr->umem)) {
 			err = PTR_ERR(mmr->umem);
diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c
index 85f57b76e446..cb23cdb9389a 100644
--- a/drivers/infiniband/hw/mlx4/qp.c
+++ b/drivers/infiniband/hw/mlx4/qp.c
@@ -916,7 +916,7 @@ static int create_rq(struct ib_pd *pd, struct ib_qp_init_attr *init_attr,
 	qp->buf_size = (qp->rq.wqe_cnt << qp->rq.wqe_shift) +
 		       (qp->sq.wqe_cnt << qp->sq.wqe_shift);
 
-	qp->umem = ib_umem_get(udata, wq.buf_addr, qp->buf_size, 0);
+	qp->umem = ib_umem_get(pd->device, wq.buf_addr, qp->buf_size, 0);
 	if (IS_ERR(qp->umem)) {
 		err = PTR_ERR(qp->umem);
 		goto err;
@@ -1110,7 +1110,8 @@ static int create_qp_common(struct ib_pd *pd, struct ib_qp_init_attr *init_attr,
 		if (err)
 			goto err;
 
-		qp->umem = ib_umem_get(udata, ucmd.buf_addr, qp->buf_size, 0);
+		qp->umem =
+			ib_umem_get(pd->device, ucmd.buf_addr, qp->buf_size, 0);
 		if (IS_ERR(qp->umem)) {
 			err = PTR_ERR(qp->umem);
 			goto err;
diff --git a/drivers/infiniband/hw/mlx4/srq.c b/drivers/infiniband/hw/mlx4/srq.c
index 8dcf6e3d9ae2..8f9d5035142d 100644
--- a/drivers/infiniband/hw/mlx4/srq.c
+++ b/drivers/infiniband/hw/mlx4/srq.c
@@ -110,7 +110,8 @@ int mlx4_ib_create_srq(struct ib_srq *ib_srq,
 		if (ib_copy_from_udata(&ucmd, udata, sizeof(ucmd)))
 			return -EFAULT;
 
-		srq->umem = ib_umem_get(udata, ucmd.buf_addr, buf_size, 0);
+		srq->umem =
+			ib_umem_get(ib_srq->device, ucmd.buf_addr, buf_size, 0);
 		if (IS_ERR(srq->umem))
 			return PTR_ERR(srq->umem);
 
diff --git a/drivers/infiniband/hw/mlx5/cq.c b/drivers/infiniband/hw/mlx5/cq.c
index dd8d24ee8e1d..367a71bc5f4b 100644
--- a/drivers/infiniband/hw/mlx5/cq.c
+++ b/drivers/infiniband/hw/mlx5/cq.c
@@ -708,8 +708,8 @@ static int create_cq_user(struct mlx5_ib_dev *dev, struct ib_udata *udata,
 	*cqe_size = ucmd.cqe_size;
 
 	cq->buf.umem =
-		ib_umem_get(udata, ucmd.buf_addr, entries * ucmd.cqe_size,
-			    IB_ACCESS_LOCAL_WRITE);
+		ib_umem_get(&dev->ib_dev, ucmd.buf_addr,
+			    entries * ucmd.cqe_size, IB_ACCESS_LOCAL_WRITE);
 	if (IS_ERR(cq->buf.umem)) {
 		err = PTR_ERR(cq->buf.umem);
 		return err;
@@ -1108,7 +1108,7 @@ static int resize_user(struct mlx5_ib_dev *dev, struct mlx5_ib_cq *cq,
 	if (ucmd.cqe_size && SIZE_MAX / ucmd.cqe_size <= entries - 1)
 		return -EINVAL;
 
-	umem = ib_umem_get(udata, ucmd.buf_addr,
+	umem = ib_umem_get(&dev->ib_dev, ucmd.buf_addr,
 			   (size_t)ucmd.cqe_size * entries,
 			   IB_ACCESS_LOCAL_WRITE);
 	if (IS_ERR(umem)) {
diff --git a/drivers/infiniband/hw/mlx5/devx.c b/drivers/infiniband/hw/mlx5/devx.c
index 9d0a18cf9e5e..685b8ed96b4e 100644
--- a/drivers/infiniband/hw/mlx5/devx.c
+++ b/drivers/infiniband/hw/mlx5/devx.c
@@ -2134,7 +2134,7 @@ static int devx_umem_get(struct mlx5_ib_dev *dev, struct ib_ucontext *ucontext,
 	if (err)
 		return err;
 
-	obj->umem = ib_umem_get(&attrs->driver_udata, addr, size, access);
+	obj->umem = ib_umem_get(&dev->ib_dev, addr, size, access);
 	if (IS_ERR(obj->umem))
 		return PTR_ERR(obj->umem);
 
diff --git a/drivers/infiniband/hw/mlx5/doorbell.c b/drivers/infiniband/hw/mlx5/doorbell.c
index 12737c509aa2..61475b571531 100644
--- a/drivers/infiniband/hw/mlx5/doorbell.c
+++ b/drivers/infiniband/hw/mlx5/doorbell.c
@@ -64,7 +64,8 @@ int mlx5_ib_db_map_user(struct mlx5_ib_ucontext *context,
 
 	page->user_virt = (virt & PAGE_MASK);
 	page->refcnt    = 0;
-	page->umem = ib_umem_get(udata, virt & PAGE_MASK, PAGE_SIZE, 0);
+	page->umem = ib_umem_get(context->ibucontext.device, virt & PAGE_MASK,
+				 PAGE_SIZE, 0);
 	if (IS_ERR(page->umem)) {
 		err = PTR_ERR(page->umem);
 		kfree(page);
diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c
index ea8bfc3e2d8d..f79bb44b94fe 100644
--- a/drivers/infiniband/hw/mlx5/mr.c
+++ b/drivers/infiniband/hw/mlx5/mr.c
@@ -737,10 +737,9 @@ static int mr_cache_max_order(struct mlx5_ib_dev *dev)
 	return MLX5_MAX_UMR_SHIFT;
 }
 
-static int mr_umem_get(struct mlx5_ib_dev *dev, struct ib_udata *udata,
-		       u64 start, u64 length, int access_flags,
-		       struct ib_umem **umem, int *npages, int *page_shift,
-		       int *ncont, int *order)
+static int mr_umem_get(struct mlx5_ib_dev *dev, u64 start, u64 length,
+		       int access_flags, struct ib_umem **umem, int *npages,
+		       int *page_shift, int *ncont, int *order)
 {
 	struct ib_umem *u;
 
@@ -749,7 +748,7 @@ static int mr_umem_get(struct mlx5_ib_dev *dev, struct ib_udata *udata,
 	if (access_flags & IB_ACCESS_ON_DEMAND) {
 		struct ib_umem_odp *odp;
 
-		odp = ib_umem_odp_get(udata, start, length, access_flags,
+		odp = ib_umem_odp_get(&dev->ib_dev, start, length, access_flags,
 				      &mlx5_mn_ops);
 		if (IS_ERR(odp)) {
 			mlx5_ib_dbg(dev, "umem get failed (%ld)\n",
@@ -765,7 +764,7 @@ static int mr_umem_get(struct mlx5_ib_dev *dev, struct ib_udata *udata,
 		if (order)
 			*order = ilog2(roundup_pow_of_two(*ncont));
 	} else {
-		u = ib_umem_get(udata, start, length, access_flags);
+		u = ib_umem_get(&dev->ib_dev, start, length, access_flags);
 		if (IS_ERR(u)) {
 			mlx5_ib_dbg(dev, "umem get failed (%ld)\n", PTR_ERR(u));
 			return PTR_ERR(u);
@@ -1257,7 +1256,7 @@ struct ib_mr *mlx5_ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
 		return &mr->ibmr;
 	}
 
-	err = mr_umem_get(dev, udata, start, length, access_flags, &umem,
+	err = mr_umem_get(dev, start, length, access_flags, &umem,
 			  &npages, &page_shift, &ncont, &order);
 
 	if (err < 0)
@@ -1424,9 +1423,8 @@ int mlx5_ib_rereg_user_mr(struct ib_mr *ib_mr, int flags, u64 start,
 		flags |= IB_MR_REREG_TRANS;
 		ib_umem_release(mr->umem);
 		mr->umem = NULL;
-		err = mr_umem_get(dev, udata, addr, len, access_flags,
-				  &mr->umem, &npages, &page_shift, &ncont,
-				  &order);
+		err = mr_umem_get(dev, addr, len, access_flags, &mr->umem,
+				  &npages, &page_shift, &ncont, &order);
 		if (err)
 			goto err;
 	}
diff --git a/drivers/infiniband/hw/mlx5/odp.c b/drivers/infiniband/hw/mlx5/odp.c
index f924250f80c2..3b3ceb5acdd3 100644
--- a/drivers/infiniband/hw/mlx5/odp.c
+++ b/drivers/infiniband/hw/mlx5/odp.c
@@ -497,7 +497,7 @@ struct mlx5_ib_mr *mlx5_ib_alloc_implicit_mr(struct mlx5_ib_pd *pd,
 	struct mlx5_ib_mr *imr;
 	int err;
 
-	umem_odp = ib_umem_odp_alloc_implicit(udata, access_flags);
+	umem_odp = ib_umem_odp_alloc_implicit(&dev->ib_dev, access_flags);
 	if (IS_ERR(umem_odp))
 		return ERR_CAST(umem_odp);
 
diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c
index 7e51870e9e01..7f0bde313560 100644
--- a/drivers/infiniband/hw/mlx5/qp.c
+++ b/drivers/infiniband/hw/mlx5/qp.c
@@ -749,7 +749,7 @@ static int mlx5_ib_umem_get(struct mlx5_ib_dev *dev, struct ib_udata *udata,
 {
 	int err;
 
-	*umem = ib_umem_get(udata, addr, size, 0);
+	*umem = ib_umem_get(&dev->ib_dev, addr, size, 0);
 	if (IS_ERR(*umem)) {
 		mlx5_ib_dbg(dev, "umem_get failed\n");
 		return PTR_ERR(*umem);
@@ -806,7 +806,7 @@ static int create_user_rq(struct mlx5_ib_dev *dev, struct ib_pd *pd,
 	if (!ucmd->buf_addr)
 		return -EINVAL;
 
-	rwq->umem = ib_umem_get(udata, ucmd->buf_addr, rwq->buf_size, 0);
+	rwq->umem = ib_umem_get(&dev->ib_dev, ucmd->buf_addr, rwq->buf_size, 0);
 	if (IS_ERR(rwq->umem)) {
 		mlx5_ib_dbg(dev, "umem_get failed\n");
 		err = PTR_ERR(rwq->umem);
diff --git a/drivers/infiniband/hw/mlx5/srq.c b/drivers/infiniband/hw/mlx5/srq.c
index 62939df3c692..b1a8a9175040 100644
--- a/drivers/infiniband/hw/mlx5/srq.c
+++ b/drivers/infiniband/hw/mlx5/srq.c
@@ -80,7 +80,7 @@ static int create_srq_user(struct ib_pd *pd, struct mlx5_ib_srq *srq,
 
 	srq->wq_sig = !!(ucmd.flags & MLX5_SRQ_FLAG_SIGNATURE);
 
-	srq->umem = ib_umem_get(udata, ucmd.buf_addr, buf_size, 0);
+	srq->umem = ib_umem_get(pd->device, ucmd.buf_addr, buf_size, 0);
 	if (IS_ERR(srq->umem)) {
 		mlx5_ib_dbg(dev, "failed umem get, size %d\n", buf_size);
 		err = PTR_ERR(srq->umem);
diff --git a/drivers/infiniband/hw/mthca/mthca_provider.c b/drivers/infiniband/hw/mthca/mthca_provider.c
index 33002530fee7..ac19d57803b5 100644
--- a/drivers/infiniband/hw/mthca/mthca_provider.c
+++ b/drivers/infiniband/hw/mthca/mthca_provider.c
@@ -880,7 +880,7 @@ static struct ib_mr *mthca_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
 	if (!mr)
 		return ERR_PTR(-ENOMEM);
 
-	mr->umem = ib_umem_get(udata, start, length, acc);
+	mr->umem = ib_umem_get(pd->device, start, length, acc);
 	if (IS_ERR(mr->umem)) {
 		err = PTR_ERR(mr->umem);
 		goto err;
diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c
index 9bc1ca6f6f9e..d47ea675734b 100644
--- a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c
+++ b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c
@@ -869,7 +869,7 @@ struct ib_mr *ocrdma_reg_user_mr(struct ib_pd *ibpd, u64 start, u64 len,
 	mr = kzalloc(sizeof(*mr), GFP_KERNEL);
 	if (!mr)
 		return ERR_PTR(status);
-	mr->umem = ib_umem_get(udata, start, len, acc);
+	mr->umem = ib_umem_get(ibpd->device, start, len, acc);
 	if (IS_ERR(mr->umem)) {
 		status = -EFAULT;
 		goto umem_err;
diff --git a/drivers/infiniband/hw/qedr/verbs.c b/drivers/infiniband/hw/qedr/verbs.c
index 4cd292966aa9..920f35e28cfc 100644
--- a/drivers/infiniband/hw/qedr/verbs.c
+++ b/drivers/infiniband/hw/qedr/verbs.c
@@ -772,7 +772,7 @@ static inline int qedr_init_user_queue(struct ib_udata *udata,
 
 	q->buf_addr = buf_addr;
 	q->buf_len = buf_len;
-	q->umem = ib_umem_get(udata, q->buf_addr, q->buf_len, access);
+	q->umem = ib_umem_get(&dev->ibdev, q->buf_addr, q->buf_len, access);
 	if (IS_ERR(q->umem)) {
 		DP_ERR(dev, "create user queue: failed ib_umem_get, got %ld\n",
 		       PTR_ERR(q->umem));
@@ -1415,9 +1415,8 @@ static int qedr_init_srq_user_params(struct ib_udata *udata,
 	if (rc)
 		return rc;
 
-	srq->prod_umem =
-		ib_umem_get(udata, ureq->prod_pair_addr,
-			    sizeof(struct rdma_srq_producers), access);
+	srq->prod_umem = ib_umem_get(srq->ibsrq.device, ureq->prod_pair_addr,
+				     sizeof(struct rdma_srq_producers), access);
 	if (IS_ERR(srq->prod_umem)) {
 		qedr_free_pbl(srq->dev, &srq->usrq.pbl_info, srq->usrq.pbl_tbl);
 		ib_umem_release(srq->usrq.umem);
@@ -2839,7 +2838,7 @@ struct ib_mr *qedr_reg_user_mr(struct ib_pd *ibpd, u64 start, u64 len,
 
 	mr->type = QEDR_MR_USER;
 
-	mr->umem = ib_umem_get(udata, start, len, acc);
+	mr->umem = ib_umem_get(ibpd->device, start, len, acc);
 	if (IS_ERR(mr->umem)) {
 		rc = -EFAULT;
 		goto err0;
diff --git a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_cq.c b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_cq.c
index a26a4fd86bf4..4f6cc0de7ef9 100644
--- a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_cq.c
+++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_cq.c
@@ -135,7 +135,7 @@ int pvrdma_create_cq(struct ib_cq *ibcq, const struct ib_cq_init_attr *attr,
 			goto err_cq;
 		}
 
-		cq->umem = ib_umem_get(udata, ucmd.buf_addr, ucmd.buf_size,
+		cq->umem = ib_umem_get(ibdev, ucmd.buf_addr, ucmd.buf_size,
 				       IB_ACCESS_LOCAL_WRITE);
 		if (IS_ERR(cq->umem)) {
 			ret = PTR_ERR(cq->umem);
diff --git a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_mr.c b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_mr.c
index c61e665ff261..b039f1f00e05 100644
--- a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_mr.c
+++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_mr.c
@@ -126,7 +126,7 @@ struct ib_mr *pvrdma_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
 		return ERR_PTR(-EINVAL);
 	}
 
-	umem = ib_umem_get(udata, start, length, access_flags);
+	umem = ib_umem_get(pd->device, start, length, access_flags);
 	if (IS_ERR(umem)) {
 		dev_warn(&dev->pdev->dev,
 			 "could not get umem for mem region\n");
diff --git a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_qp.c b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_qp.c
index f15809c28f67..9de1281f9a3b 100644
--- a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_qp.c
+++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_qp.c
@@ -276,8 +276,9 @@ struct ib_qp *pvrdma_create_qp(struct ib_pd *pd,
 
 			if (!is_srq) {
 				/* set qp->sq.wqe_cnt, shift, buf_size.. */
-				qp->rumem = ib_umem_get(udata, ucmd.rbuf_addr,
-							ucmd.rbuf_size, 0);
+				qp->rumem =
+					ib_umem_get(pd->device, ucmd.rbuf_addr,
+						    ucmd.rbuf_size, 0);
 				if (IS_ERR(qp->rumem)) {
 					ret = PTR_ERR(qp->rumem);
 					goto err_qp;
@@ -288,7 +289,7 @@ struct ib_qp *pvrdma_create_qp(struct ib_pd *pd,
 				qp->srq = to_vsrq(init_attr->srq);
 			}
 
-			qp->sumem = ib_umem_get(udata, ucmd.sbuf_addr,
+			qp->sumem = ib_umem_get(pd->device, ucmd.sbuf_addr,
 						ucmd.sbuf_size, 0);
 			if (IS_ERR(qp->sumem)) {
 				if (!is_srq)
diff --git a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_srq.c b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_srq.c
index 98c8be71d91d..d330decfb80a 100644
--- a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_srq.c
+++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_srq.c
@@ -146,7 +146,7 @@ int pvrdma_create_srq(struct ib_srq *ibsrq, struct ib_srq_init_attr *init_attr,
 		goto err_srq;
 	}
 
-	srq->umem = ib_umem_get(udata, ucmd.buf_addr, ucmd.buf_size, 0);
+	srq->umem = ib_umem_get(ibsrq->device, ucmd.buf_addr, ucmd.buf_size, 0);
 	if (IS_ERR(srq->umem)) {
 		ret = PTR_ERR(srq->umem);
 		goto err_srq;
diff --git a/drivers/infiniband/sw/rdmavt/mr.c b/drivers/infiniband/sw/rdmavt/mr.c
index b9a76bf74857..72f6534fbb52 100644
--- a/drivers/infiniband/sw/rdmavt/mr.c
+++ b/drivers/infiniband/sw/rdmavt/mr.c
@@ -390,7 +390,7 @@ struct ib_mr *rvt_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
 	if (length == 0)
 		return ERR_PTR(-EINVAL);
 
-	umem = ib_umem_get(udata, start, length, mr_access_flags);
+	umem = ib_umem_get(pd->device, start, length, mr_access_flags);
 	if (IS_ERR(umem))
 		return (void *)umem;
 
diff --git a/drivers/infiniband/sw/rxe/rxe_mr.c b/drivers/infiniband/sw/rxe/rxe_mr.c
index 35a2baf2f364..e83c7b518bfa 100644
--- a/drivers/infiniband/sw/rxe/rxe_mr.c
+++ b/drivers/infiniband/sw/rxe/rxe_mr.c
@@ -169,7 +169,7 @@ int rxe_mem_init_user(struct rxe_pd *pd, u64 start,
 	void			*vaddr;
 	int err;
 
-	umem = ib_umem_get(udata, start, length, access);
+	umem = ib_umem_get(pd->ibpd.device, start, length, access);
 	if (IS_ERR(umem)) {
 		pr_warn("err %d from rxe_umem_get\n",
 			(int)PTR_ERR(umem));
diff --git a/include/rdma/ib_umem.h b/include/rdma/ib_umem.h
index 753f54e17e0a..e3518fd6b95b 100644
--- a/include/rdma/ib_umem.h
+++ b/include/rdma/ib_umem.h
@@ -69,7 +69,7 @@ static inline size_t ib_umem_num_pages(struct ib_umem *umem)
 
 #ifdef CONFIG_INFINIBAND_USER_MEM
 
-struct ib_umem *ib_umem_get(struct ib_udata *udata, unsigned long addr,
+struct ib_umem *ib_umem_get(struct ib_device *device, unsigned long addr,
 			    size_t size, int access);
 void ib_umem_release(struct ib_umem *umem);
 int ib_umem_page_count(struct ib_umem *umem);
@@ -83,7 +83,7 @@ unsigned long ib_umem_find_best_pgsz(struct ib_umem *umem,
 
 #include <linux/err.h>
 
-static inline struct ib_umem *ib_umem_get(struct ib_udata *udata,
+static inline struct ib_umem *ib_umem_get(struct ib_device *device,
 					  unsigned long addr, size_t size,
 					  int access)
 {
diff --git a/include/rdma/ib_umem_odp.h b/include/rdma/ib_umem_odp.h
index 81429acc8257..64314ff76612 100644
--- a/include/rdma/ib_umem_odp.h
+++ b/include/rdma/ib_umem_odp.h
@@ -114,9 +114,9 @@ static inline size_t ib_umem_odp_num_pages(struct ib_umem_odp *umem_odp)
 #ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
 
 struct ib_umem_odp *
-ib_umem_odp_get(struct ib_udata *udata, unsigned long addr, size_t size,
+ib_umem_odp_get(struct ib_device *device, unsigned long addr, size_t size,
 		int access, const struct mmu_interval_notifier_ops *ops);
-struct ib_umem_odp *ib_umem_odp_alloc_implicit(struct ib_udata *udata,
+struct ib_umem_odp *ib_umem_odp_alloc_implicit(struct ib_device *device,
 					       int access);
 struct ib_umem_odp *
 ib_umem_odp_alloc_child(struct ib_umem_odp *root_umem, unsigned long addr,
@@ -134,7 +134,7 @@ void ib_umem_odp_unmap_dma_pages(struct ib_umem_odp *umem_odp, u64 start_offset,
 #else /* CONFIG_INFINIBAND_ON_DEMAND_PAGING */
 
 static inline struct ib_umem_odp *
-ib_umem_odp_get(struct ib_udata *udata, unsigned long addr, size_t size,
+ib_umem_odp_get(struct ib_device *device, unsigned long addr, size_t size,
 		int access, const struct mmu_interval_notifier_ops *ops)
 {
 	return ERR_PTR(-EINVAL);
-- 
cgit v1.2.3-59-g8ed1b


From 33006bd4f37f7d2c3d1cf0268b4f327b5fdc2558 Mon Sep 17 00:00:00 2001
From: Moni Shoua <monis@mellanox.com>
Date: Wed, 15 Jan 2020 14:43:32 +0200
Subject: IB/core: Introduce ib_reg_user_mr

Add ib_reg_user_mr() for kernel ULPs to register user MRs.

The common use case that uses this function is a userspace application
that allocates memory for HCA access but the responsibility to register
the memory at the HCA is on an kernel ULP. This ULP that acts as an agent
for the userspace application.

This function is intended to be used without a user context so vendor
drivers need to be aware of calling reg_user_mr() device operation with
udata equal to NULL.

Among all drivers, i40iw is the only driver which relies on presence
of udata, so check udata existence for that driver.

Signed-off-by: Moni Shoua <monis@mellanox.com>
Reviewed-by: Guy Levi <guyle@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
---
 drivers/infiniband/core/verbs.c           | 30 ++++++++++++++++++++++++++++++
 drivers/infiniband/hw/efa/efa_verbs.c     |  2 +-
 drivers/infiniband/hw/i40iw/i40iw_verbs.c |  3 +++
 include/rdma/ib_verbs.h                   |  6 ++++++
 4 files changed, 40 insertions(+), 1 deletion(-)

diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c
index dd765e176cdd..7a69e4bbe877 100644
--- a/drivers/infiniband/core/verbs.c
+++ b/drivers/infiniband/core/verbs.c
@@ -1990,6 +1990,36 @@ EXPORT_SYMBOL(ib_resize_cq);
 
 /* Memory regions */
 
+struct ib_mr *ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
+			     u64 virt_addr, int access_flags)
+{
+	struct ib_mr *mr;
+
+	if (access_flags & IB_ACCESS_ON_DEMAND) {
+		if (!(pd->device->attrs.device_cap_flags &
+		      IB_DEVICE_ON_DEMAND_PAGING)) {
+			pr_debug("ODP support not available\n");
+			return ERR_PTR(-EINVAL);
+		}
+	}
+
+	mr = pd->device->ops.reg_user_mr(pd, start, length, virt_addr,
+					 access_flags, NULL);
+
+	if (IS_ERR(mr))
+		return mr;
+
+	mr->device = pd->device;
+	mr->pd = pd;
+	mr->dm = NULL;
+	atomic_inc(&pd->usecnt);
+	mr->res.type = RDMA_RESTRACK_MR;
+	rdma_restrack_kadd(&mr->res);
+
+	return mr;
+}
+EXPORT_SYMBOL(ib_reg_user_mr);
+
 int ib_dereg_mr_user(struct ib_mr *mr, struct ib_udata *udata)
 {
 	struct ib_pd *pd = mr->pd;
diff --git a/drivers/infiniband/hw/efa/efa_verbs.c b/drivers/infiniband/hw/efa/efa_verbs.c
index 5c35aa72f515..4822f5fa12be 100644
--- a/drivers/infiniband/hw/efa/efa_verbs.c
+++ b/drivers/infiniband/hw/efa/efa_verbs.c
@@ -1358,7 +1358,7 @@ struct ib_mr *efa_reg_mr(struct ib_pd *ibpd, u64 start, u64 length,
 	int inline_size;
 	int err;
 
-	if (udata->inlen &&
+	if (udata && udata->inlen &&
 	    !ib_is_udata_cleared(udata, 0, sizeof(udata->inlen))) {
 		ibdev_dbg(&dev->ibdev,
 			  "Incompatible ABI params, udata not cleared\n");
diff --git a/drivers/infiniband/hw/i40iw/i40iw_verbs.c b/drivers/infiniband/hw/i40iw/i40iw_verbs.c
index 96488fb443eb..c335de91508f 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_verbs.c
+++ b/drivers/infiniband/hw/i40iw/i40iw_verbs.c
@@ -1756,6 +1756,9 @@ static struct ib_mr *i40iw_reg_user_mr(struct ib_pd *pd,
 	int ret;
 	int pg_shift;
 
+	if (!udata)
+		return ERR_PTR(-EOPNOTSUPP);
+
 	if (iwdev->closing)
 		return ERR_PTR(-ENODEV);
 
diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
index 5608e14e3aad..170d5ec95b79 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -4153,6 +4153,12 @@ static inline void ib_dma_free_coherent(struct ib_device *dev,
 	dma_free_coherent(dev->dma_device, size, cpu_addr, dma_handle);
 }
 
+/* ib_reg_user_mr - register a memory region for virtual addresses from kernel
+ * space. This function should be called when 'current' is the owning MM.
+ */
+struct ib_mr *ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
+			     u64 virt_addr, int mr_access_flags);
+
 /**
  * ib_dereg_mr_user - Deregisters a memory region and removes it from the
  *   HCA translation table.
-- 
cgit v1.2.3-59-g8ed1b


From 87d8069f6b028793254ddd0a66df1d7b6d79b450 Mon Sep 17 00:00:00 2001
From: Moni Shoua <monis@mellanox.com>
Date: Wed, 15 Jan 2020 14:43:33 +0200
Subject: IB/core: Add interface to advise_mr for kernel users

Allow ULPs to call advise_mr, so they can control ODP regions
in the same way as user space applications.

Signed-off-by: Moni Shoua <monis@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
---
 drivers/infiniband/core/verbs.c | 11 +++++++++++
 include/rdma/ib_verbs.h         |  3 +++
 2 files changed, 14 insertions(+)

diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c
index 7a69e4bbe877..d33bdc9b01cd 100644
--- a/drivers/infiniband/core/verbs.c
+++ b/drivers/infiniband/core/verbs.c
@@ -2020,6 +2020,17 @@ struct ib_mr *ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
 }
 EXPORT_SYMBOL(ib_reg_user_mr);
 
+int ib_advise_mr(struct ib_pd *pd, enum ib_uverbs_advise_mr_advice advice,
+		 u32 flags, struct ib_sge *sg_list, u32 num_sge)
+{
+	if (!pd->device->ops.advise_mr)
+		return -EOPNOTSUPP;
+
+	return pd->device->ops.advise_mr(pd, advice, flags, sg_list, num_sge,
+					 NULL);
+}
+EXPORT_SYMBOL(ib_advise_mr);
+
 int ib_dereg_mr_user(struct ib_mr *mr, struct ib_udata *udata)
 {
 	struct ib_pd *pd = mr->pd;
diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
index 170d5ec95b79..e2cc62217cc2 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -4159,6 +4159,9 @@ static inline void ib_dma_free_coherent(struct ib_device *dev,
 struct ib_mr *ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
 			     u64 virt_addr, int mr_access_flags);
 
+/* ib_advise_mr -  give an advice about an address range in a memory region */
+int ib_advise_mr(struct ib_pd *pd, enum ib_uverbs_advise_mr_advice advice,
+		 u32 flags, struct ib_sge *sg_list, u32 num_sge);
 /**
  * ib_dereg_mr_user - Deregisters a memory region and removes it from the
  *   HCA translation table.
-- 
cgit v1.2.3-59-g8ed1b


From da9ee9d8a8745e70e481446e0bfe2d773b1c364b Mon Sep 17 00:00:00 2001
From: Moni Shoua <monis@mellanox.com>
Date: Wed, 15 Jan 2020 14:43:34 +0200
Subject: IB/mlx5: Add ODP WQE handlers for kernel QPs

One of the steps in ODP page fault handler for WQEs is to read a WQE
from a QP send queue or receive queue buffer at a specific index.

Since the implementation of this buffer is different between kernel and
user QP the implementation of the handler needs to be aware of that and
handle it in a different way.

ODP for kernel MRs is currently supported only for RDMA_READ
and RDMA_WRITE operations so change the handler to
- read a WQE from a kernel QP send queue
- fail if access to receive queue or shared receive queue is
  required for a kernel QP

Signed-off-by: Moni Shoua <monis@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
---
 drivers/infiniband/hw/mlx5/mlx5_ib.h |  12 +--
 drivers/infiniband/hw/mlx5/odp.c     |  12 +--
 drivers/infiniband/hw/mlx5/qp.c      | 163 ++++++++++++++++++++++-------------
 3 files changed, 117 insertions(+), 70 deletions(-)

diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h
index b06f32ff5748..77d495b2032d 100644
--- a/drivers/infiniband/hw/mlx5/mlx5_ib.h
+++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h
@@ -1153,12 +1153,12 @@ int mlx5_ib_post_send(struct ib_qp *ibqp, const struct ib_send_wr *wr,
 		      const struct ib_send_wr **bad_wr);
 int mlx5_ib_post_recv(struct ib_qp *ibqp, const struct ib_recv_wr *wr,
 		      const struct ib_recv_wr **bad_wr);
-int mlx5_ib_read_user_wqe_sq(struct mlx5_ib_qp *qp, int wqe_index, void *buffer,
-			     int buflen, size_t *bc);
-int mlx5_ib_read_user_wqe_rq(struct mlx5_ib_qp *qp, int wqe_index, void *buffer,
-			     int buflen, size_t *bc);
-int mlx5_ib_read_user_wqe_srq(struct mlx5_ib_srq *srq, int wqe_index,
-			      void *buffer, int buflen, size_t *bc);
+int mlx5_ib_read_wqe_sq(struct mlx5_ib_qp *qp, int wqe_index, void *buffer,
+			size_t buflen, size_t *bc);
+int mlx5_ib_read_wqe_rq(struct mlx5_ib_qp *qp, int wqe_index, void *buffer,
+			size_t buflen, size_t *bc);
+int mlx5_ib_read_wqe_srq(struct mlx5_ib_srq *srq, int wqe_index, void *buffer,
+			 size_t buflen, size_t *bc);
 int mlx5_ib_create_cq(struct ib_cq *ibcq, const struct ib_cq_init_attr *attr,
 		      struct ib_udata *udata);
 void mlx5_ib_destroy_cq(struct ib_cq *cq, struct ib_udata *udata);
diff --git a/drivers/infiniband/hw/mlx5/odp.c b/drivers/infiniband/hw/mlx5/odp.c
index 3b3ceb5acdd3..3642c6a491c2 100644
--- a/drivers/infiniband/hw/mlx5/odp.c
+++ b/drivers/infiniband/hw/mlx5/odp.c
@@ -1237,15 +1237,15 @@ static void mlx5_ib_mr_wqe_pfault_handler(struct mlx5_ib_dev *dev,
 	wqe = wqe_start;
 	qp = (res->res == MLX5_RES_QP) ? res_to_qp(res) : NULL;
 	if (qp && sq) {
-		ret = mlx5_ib_read_user_wqe_sq(qp, wqe_index, wqe, PAGE_SIZE,
-					       &bytes_copied);
+		ret = mlx5_ib_read_wqe_sq(qp, wqe_index, wqe, PAGE_SIZE,
+					  &bytes_copied);
 		if (ret)
 			goto read_user;
 		ret = mlx5_ib_mr_initiator_pfault_handler(
 			dev, pfault, qp, &wqe, &wqe_end, bytes_copied);
 	} else if (qp && !sq) {
-		ret = mlx5_ib_read_user_wqe_rq(qp, wqe_index, wqe, PAGE_SIZE,
-					       &bytes_copied);
+		ret = mlx5_ib_read_wqe_rq(qp, wqe_index, wqe, PAGE_SIZE,
+					  &bytes_copied);
 		if (ret)
 			goto read_user;
 		ret = mlx5_ib_mr_responder_pfault_handler_rq(
@@ -1253,8 +1253,8 @@ static void mlx5_ib_mr_wqe_pfault_handler(struct mlx5_ib_dev *dev,
 	} else if (!qp) {
 		struct mlx5_ib_srq *srq = res_to_srq(res);
 
-		ret = mlx5_ib_read_user_wqe_srq(srq, wqe_index, wqe, PAGE_SIZE,
-						&bytes_copied);
+		ret = mlx5_ib_read_wqe_srq(srq, wqe_index, wqe, PAGE_SIZE,
+					   &bytes_copied);
 		if (ret)
 			goto read_user;
 		ret = mlx5_ib_mr_responder_pfault_handler_srq(
diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c
index 7f0bde313560..ae7cbd9c9bca 100644
--- a/drivers/infiniband/hw/mlx5/qp.c
+++ b/drivers/infiniband/hw/mlx5/qp.c
@@ -129,14 +129,10 @@ static int is_sqp(enum ib_qp_type qp_type)
  *
  * Return: zero on success, or an error code.
  */
-static int mlx5_ib_read_user_wqe_common(struct ib_umem *umem,
-					void *buffer,
-					u32 buflen,
-					int wqe_index,
-					int wq_offset,
-					int wq_wqe_cnt,
-					int wq_wqe_shift,
-					int bcnt,
+static int mlx5_ib_read_user_wqe_common(struct ib_umem *umem, void *buffer,
+					size_t buflen, int wqe_index,
+					int wq_offset, int wq_wqe_cnt,
+					int wq_wqe_shift, int bcnt,
 					size_t *bytes_copied)
 {
 	size_t offset = wq_offset + ((wqe_index % wq_wqe_cnt) << wq_wqe_shift);
@@ -160,11 +156,43 @@ static int mlx5_ib_read_user_wqe_common(struct ib_umem *umem,
 	return 0;
 }
 
-int mlx5_ib_read_user_wqe_sq(struct mlx5_ib_qp *qp,
-			     int wqe_index,
-			     void *buffer,
-			     int buflen,
-			     size_t *bc)
+static int mlx5_ib_read_kernel_wqe_sq(struct mlx5_ib_qp *qp, int wqe_index,
+				      void *buffer, size_t buflen, size_t *bc)
+{
+	struct mlx5_wqe_ctrl_seg *ctrl;
+	size_t bytes_copied = 0;
+	size_t wqe_length;
+	void *p;
+	int ds;
+
+	wqe_index = wqe_index & qp->sq.fbc.sz_m1;
+
+	/* read the control segment first */
+	p = mlx5_frag_buf_get_wqe(&qp->sq.fbc, wqe_index);
+	ctrl = p;
+	ds = be32_to_cpu(ctrl->qpn_ds) & MLX5_WQE_CTRL_DS_MASK;
+	wqe_length = ds * MLX5_WQE_DS_UNITS;
+
+	/* read rest of WQE if it spreads over more than one stride */
+	while (bytes_copied < wqe_length) {
+		size_t copy_length =
+			min_t(size_t, buflen - bytes_copied, MLX5_SEND_WQE_BB);
+
+		if (!copy_length)
+			break;
+
+		memcpy(buffer + bytes_copied, p, copy_length);
+		bytes_copied += copy_length;
+
+		wqe_index = (wqe_index + 1) & qp->sq.fbc.sz_m1;
+		p = mlx5_frag_buf_get_wqe(&qp->sq.fbc, wqe_index);
+	}
+	*bc = bytes_copied;
+	return 0;
+}
+
+static int mlx5_ib_read_user_wqe_sq(struct mlx5_ib_qp *qp, int wqe_index,
+				    void *buffer, size_t buflen, size_t *bc)
 {
 	struct mlx5_ib_qp_base *base = &qp->trans_qp.base;
 	struct ib_umem *umem = base->ubuffer.umem;
@@ -176,18 +204,10 @@ int mlx5_ib_read_user_wqe_sq(struct mlx5_ib_qp *qp,
 	int ret;
 	int ds;
 
-	if (buflen < sizeof(*ctrl))
-		return -EINVAL;
-
 	/* at first read as much as possible */
-	ret = mlx5_ib_read_user_wqe_common(umem,
-					   buffer,
-					   buflen,
-					   wqe_index,
-					   wq->offset,
-					   wq->wqe_cnt,
-					   wq->wqe_shift,
-					   buflen,
+	ret = mlx5_ib_read_user_wqe_common(umem, buffer, buflen, wqe_index,
+					   wq->offset, wq->wqe_cnt,
+					   wq->wqe_shift, buflen,
 					   &bytes_copied);
 	if (ret)
 		return ret;
@@ -210,13 +230,9 @@ int mlx5_ib_read_user_wqe_sq(struct mlx5_ib_qp *qp,
 	 * so read the remaining bytes starting
 	 * from  wqe_index 0
 	 */
-	ret = mlx5_ib_read_user_wqe_common(umem,
-					   buffer + bytes_copied,
-					   buflen - bytes_copied,
-					   0,
-					   wq->offset,
-					   wq->wqe_cnt,
-					   wq->wqe_shift,
+	ret = mlx5_ib_read_user_wqe_common(umem, buffer + bytes_copied,
+					   buflen - bytes_copied, 0, wq->offset,
+					   wq->wqe_cnt, wq->wqe_shift,
 					   wqe_length - bytes_copied,
 					   &bytes_copied2);
 
@@ -226,11 +242,24 @@ int mlx5_ib_read_user_wqe_sq(struct mlx5_ib_qp *qp,
 	return 0;
 }
 
-int mlx5_ib_read_user_wqe_rq(struct mlx5_ib_qp *qp,
-			     int wqe_index,
-			     void *buffer,
-			     int buflen,
-			     size_t *bc)
+int mlx5_ib_read_wqe_sq(struct mlx5_ib_qp *qp, int wqe_index, void *buffer,
+			size_t buflen, size_t *bc)
+{
+	struct mlx5_ib_qp_base *base = &qp->trans_qp.base;
+	struct ib_umem *umem = base->ubuffer.umem;
+
+	if (buflen < sizeof(struct mlx5_wqe_ctrl_seg))
+		return -EINVAL;
+
+	if (!umem)
+		return mlx5_ib_read_kernel_wqe_sq(qp, wqe_index, buffer,
+						  buflen, bc);
+
+	return mlx5_ib_read_user_wqe_sq(qp, wqe_index, buffer, buflen, bc);
+}
+
+static int mlx5_ib_read_user_wqe_rq(struct mlx5_ib_qp *qp, int wqe_index,
+				    void *buffer, size_t buflen, size_t *bc)
 {
 	struct mlx5_ib_qp_base *base = &qp->trans_qp.base;
 	struct ib_umem *umem = base->ubuffer.umem;
@@ -238,14 +267,9 @@ int mlx5_ib_read_user_wqe_rq(struct mlx5_ib_qp *qp,
 	size_t bytes_copied;
 	int ret;
 
-	ret = mlx5_ib_read_user_wqe_common(umem,
-					   buffer,
-					   buflen,
-					   wqe_index,
-					   wq->offset,
-					   wq->wqe_cnt,
-					   wq->wqe_shift,
-					   buflen,
+	ret = mlx5_ib_read_user_wqe_common(umem, buffer, buflen, wqe_index,
+					   wq->offset, wq->wqe_cnt,
+					   wq->wqe_shift, buflen,
 					   &bytes_copied);
 
 	if (ret)
@@ -254,25 +278,33 @@ int mlx5_ib_read_user_wqe_rq(struct mlx5_ib_qp *qp,
 	return 0;
 }
 
-int mlx5_ib_read_user_wqe_srq(struct mlx5_ib_srq *srq,
-			      int wqe_index,
-			      void *buffer,
-			      int buflen,
-			      size_t *bc)
+int mlx5_ib_read_wqe_rq(struct mlx5_ib_qp *qp, int wqe_index, void *buffer,
+			size_t buflen, size_t *bc)
+{
+	struct mlx5_ib_qp_base *base = &qp->trans_qp.base;
+	struct ib_umem *umem = base->ubuffer.umem;
+	struct mlx5_ib_wq *wq = &qp->rq;
+	size_t wqe_size = 1 << wq->wqe_shift;
+
+	if (buflen < wqe_size)
+		return -EINVAL;
+
+	if (!umem)
+		return -EOPNOTSUPP;
+
+	return mlx5_ib_read_user_wqe_rq(qp, wqe_index, buffer, buflen, bc);
+}
+
+static int mlx5_ib_read_user_wqe_srq(struct mlx5_ib_srq *srq, int wqe_index,
+				     void *buffer, size_t buflen, size_t *bc)
 {
 	struct ib_umem *umem = srq->umem;
 	size_t bytes_copied;
 	int ret;
 
-	ret = mlx5_ib_read_user_wqe_common(umem,
-					   buffer,
-					   buflen,
-					   wqe_index,
-					   0,
-					   srq->msrq.max,
-					   srq->msrq.wqe_shift,
-					   buflen,
-					   &bytes_copied);
+	ret = mlx5_ib_read_user_wqe_common(umem, buffer, buflen, wqe_index, 0,
+					   srq->msrq.max, srq->msrq.wqe_shift,
+					   buflen, &bytes_copied);
 
 	if (ret)
 		return ret;
@@ -280,6 +312,21 @@ int mlx5_ib_read_user_wqe_srq(struct mlx5_ib_srq *srq,
 	return 0;
 }
 
+int mlx5_ib_read_wqe_srq(struct mlx5_ib_srq *srq, int wqe_index, void *buffer,
+			 size_t buflen, size_t *bc)
+{
+	struct ib_umem *umem = srq->umem;
+	size_t wqe_size = 1 << srq->msrq.wqe_shift;
+
+	if (buflen < wqe_size)
+		return -EINVAL;
+
+	if (!umem)
+		return -EOPNOTSUPP;
+
+	return mlx5_ib_read_user_wqe_srq(srq, wqe_index, buffer, buflen, bc);
+}
+
 static void mlx5_ib_qp_event(struct mlx5_core_qp *qp, int type)
 {
 	struct ib_qp *ibqp = &to_mibqp(qp)->ibqp;
-- 
cgit v1.2.3-59-g8ed1b


From 4835709176e8ccf6561abc9f5c405293e008095f Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@mellanox.com>
Date: Wed, 15 Jan 2020 14:43:35 +0200
Subject: RDMA/mlx5: Don't fake udata for kernel path

Kernel paths must not set udata and provide NULL pointer,
instead of faking zeroed udata struct.

Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
---
 drivers/infiniband/hw/mlx5/main.c | 34 ++++++++++++++++------------------
 1 file changed, 16 insertions(+), 18 deletions(-)

diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
index 997cbfe4b90c..760630c7aae7 100644
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@ -815,6 +815,7 @@ static int mlx5_ib_query_device(struct ib_device *ibdev,
 				struct ib_device_attr *props,
 				struct ib_udata *uhw)
 {
+	size_t uhw_outlen = (uhw) ? uhw->outlen : 0;
 	struct mlx5_ib_dev *dev = to_mdev(ibdev);
 	struct mlx5_core_dev *mdev = dev->mdev;
 	int err = -ENOMEM;
@@ -828,12 +829,12 @@ static int mlx5_ib_query_device(struct ib_device *ibdev,
 	u64 max_tso;
 
 	resp_len = sizeof(resp.comp_mask) + sizeof(resp.response_length);
-	if (uhw->outlen && uhw->outlen < resp_len)
+	if (uhw_outlen && uhw_outlen < resp_len)
 		return -EINVAL;
 
 	resp.response_length = resp_len;
 
-	if (uhw->inlen && !ib_is_udata_cleared(uhw, 0, uhw->inlen))
+	if (uhw && uhw->inlen && !ib_is_udata_cleared(uhw, 0, uhw->inlen))
 		return -EINVAL;
 
 	memset(props, 0, sizeof(*props));
@@ -897,7 +898,7 @@ static int mlx5_ib_query_device(struct ib_device *ibdev,
 			props->raw_packet_caps |=
 				IB_RAW_PACKET_CAP_CVLAN_STRIPPING;
 
-		if (field_avail(typeof(resp), tso_caps, uhw->outlen)) {
+		if (field_avail(typeof(resp), tso_caps, uhw_outlen)) {
 			max_tso = MLX5_CAP_ETH(mdev, max_lso_cap);
 			if (max_tso) {
 				resp.tso_caps.max_tso = 1 << max_tso;
@@ -907,7 +908,7 @@ static int mlx5_ib_query_device(struct ib_device *ibdev,
 			}
 		}
 
-		if (field_avail(typeof(resp), rss_caps, uhw->outlen)) {
+		if (field_avail(typeof(resp), rss_caps, uhw_outlen)) {
 			resp.rss_caps.rx_hash_function =
 						MLX5_RX_HASH_FUNC_TOEPLITZ;
 			resp.rss_caps.rx_hash_fields_mask =
@@ -927,9 +928,9 @@ static int mlx5_ib_query_device(struct ib_device *ibdev,
 			resp.response_length += sizeof(resp.rss_caps);
 		}
 	} else {
-		if (field_avail(typeof(resp), tso_caps, uhw->outlen))
+		if (field_avail(typeof(resp), tso_caps, uhw_outlen))
 			resp.response_length += sizeof(resp.tso_caps);
-		if (field_avail(typeof(resp), rss_caps, uhw->outlen))
+		if (field_avail(typeof(resp), rss_caps, uhw_outlen))
 			resp.response_length += sizeof(resp.rss_caps);
 	}
 
@@ -1054,7 +1055,7 @@ static int mlx5_ib_query_device(struct ib_device *ibdev,
 						MLX5_MAX_CQ_PERIOD;
 	}
 
-	if (field_avail(typeof(resp), cqe_comp_caps, uhw->outlen)) {
+	if (field_avail(typeof(resp), cqe_comp_caps, uhw_outlen)) {
 		resp.response_length += sizeof(resp.cqe_comp_caps);
 
 		if (MLX5_CAP_GEN(dev->mdev, cqe_compression)) {
@@ -1072,7 +1073,7 @@ static int mlx5_ib_query_device(struct ib_device *ibdev,
 		}
 	}
 
-	if (field_avail(typeof(resp), packet_pacing_caps, uhw->outlen) &&
+	if (field_avail(typeof(resp), packet_pacing_caps, uhw_outlen) &&
 	    raw_support) {
 		if (MLX5_CAP_QOS(mdev, packet_pacing) &&
 		    MLX5_CAP_GEN(mdev, qos)) {
@@ -1091,7 +1092,7 @@ static int mlx5_ib_query_device(struct ib_device *ibdev,
 	}
 
 	if (field_avail(typeof(resp), mlx5_ib_support_multi_pkt_send_wqes,
-			uhw->outlen)) {
+			uhw_outlen)) {
 		if (MLX5_CAP_ETH(mdev, multi_pkt_send_wqe))
 			resp.mlx5_ib_support_multi_pkt_send_wqes =
 				MLX5_IB_ALLOW_MPW;
@@ -1104,7 +1105,7 @@ static int mlx5_ib_query_device(struct ib_device *ibdev,
 			sizeof(resp.mlx5_ib_support_multi_pkt_send_wqes);
 	}
 
-	if (field_avail(typeof(resp), flags, uhw->outlen)) {
+	if (field_avail(typeof(resp), flags, uhw_outlen)) {
 		resp.response_length += sizeof(resp.flags);
 
 		if (MLX5_CAP_GEN(mdev, cqe_compression_128))
@@ -1120,8 +1121,7 @@ static int mlx5_ib_query_device(struct ib_device *ibdev,
 		resp.flags |= MLX5_IB_QUERY_DEV_RESP_FLAGS_SCAT2CQE_DCT;
 	}
 
-	if (field_avail(typeof(resp), sw_parsing_caps,
-			uhw->outlen)) {
+	if (field_avail(typeof(resp), sw_parsing_caps, uhw_outlen)) {
 		resp.response_length += sizeof(resp.sw_parsing_caps);
 		if (MLX5_CAP_ETH(mdev, swp)) {
 			resp.sw_parsing_caps.sw_parsing_offloads |=
@@ -1141,7 +1141,7 @@ static int mlx5_ib_query_device(struct ib_device *ibdev,
 		}
 	}
 
-	if (field_avail(typeof(resp), striding_rq_caps, uhw->outlen) &&
+	if (field_avail(typeof(resp), striding_rq_caps, uhw_outlen) &&
 	    raw_support) {
 		resp.response_length += sizeof(resp.striding_rq_caps);
 		if (MLX5_CAP_GEN(mdev, striding_rq)) {
@@ -1164,8 +1164,7 @@ static int mlx5_ib_query_device(struct ib_device *ibdev,
 		}
 	}
 
-	if (field_avail(typeof(resp), tunnel_offloads_caps,
-			uhw->outlen)) {
+	if (field_avail(typeof(resp), tunnel_offloads_caps, uhw_outlen)) {
 		resp.response_length += sizeof(resp.tunnel_offloads_caps);
 		if (MLX5_CAP_ETH(mdev, tunnel_stateless_vxlan))
 			resp.tunnel_offloads_caps |=
@@ -1186,7 +1185,7 @@ static int mlx5_ib_query_device(struct ib_device *ibdev,
 				MLX5_IB_TUNNELED_OFFLOADS_MPLS_UDP;
 	}
 
-	if (uhw->outlen) {
+	if (uhw_outlen) {
 		err = ib_copy_to_udata(uhw, &resp, resp.response_length);
 
 		if (err)
@@ -4771,7 +4770,6 @@ static int __get_port_caps(struct mlx5_ib_dev *dev, u8 port)
 	struct ib_device_attr *dprops = NULL;
 	struct ib_port_attr *pprops = NULL;
 	int err = -ENOMEM;
-	struct ib_udata uhw = {.inlen = 0, .outlen = 0};
 
 	pprops = kzalloc(sizeof(*pprops), GFP_KERNEL);
 	if (!pprops)
@@ -4781,7 +4779,7 @@ static int __get_port_caps(struct mlx5_ib_dev *dev, u8 port)
 	if (!dprops)
 		goto out;
 
-	err = mlx5_ib_query_device(&dev->ib_dev, dprops, &uhw);
+	err = mlx5_ib_query_device(&dev->ib_dev, dprops, NULL);
 	if (err) {
 		mlx5_ib_warn(dev, "query_device failed %d\n", err);
 		goto out;
-- 
cgit v1.2.3-59-g8ed1b


From a73a895588520a40e52f967a7b2892baf810c708 Mon Sep 17 00:00:00 2001
From: Moni Shoua <monis@mellanox.com>
Date: Wed, 15 Jan 2020 14:43:36 +0200
Subject: IB/mlx5: Mask out unsupported ODP capabilities for kernel QPs

The ODP handler for WQEs in RQ or SRQ is not implented for kernel QPs.
Therefore don't report support in these if query comes from a kernel user.

Signed-off-by: Moni Shoua <monis@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
---
 drivers/infiniband/hw/mlx5/main.c | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
index 760630c7aae7..994d59cd9daf 100644
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@ -1015,6 +1015,23 @@ static int mlx5_ib_query_device(struct ib_device *ibdev,
 		if (dev->odp_caps.general_caps & IB_ODP_SUPPORT)
 			props->device_cap_flags |= IB_DEVICE_ON_DEMAND_PAGING;
 		props->odp_caps = dev->odp_caps;
+		if (!uhw) {
+			/* ODP for kernel QPs is not implemented for receive
+			 * WQEs and SRQ WQEs
+			 */
+			props->odp_caps.per_transport_caps.rc_odp_caps &=
+				~(IB_ODP_SUPPORT_READ |
+				  IB_ODP_SUPPORT_SRQ_RECV);
+			props->odp_caps.per_transport_caps.uc_odp_caps &=
+				~(IB_ODP_SUPPORT_READ |
+				  IB_ODP_SUPPORT_SRQ_RECV);
+			props->odp_caps.per_transport_caps.ud_odp_caps &=
+				~(IB_ODP_SUPPORT_READ |
+				  IB_ODP_SUPPORT_SRQ_RECV);
+			props->odp_caps.per_transport_caps.xrc_odp_caps &=
+				~(IB_ODP_SUPPORT_READ |
+				  IB_ODP_SUPPORT_SRQ_RECV);
+		}
 	}
 
 	if (MLX5_CAP_GEN(mdev, cd))
-- 
cgit v1.2.3-59-g8ed1b


From 8ffc32485158528f870b62707077ab494ba31deb Mon Sep 17 00:00:00 2001
From: Jason Gunthorpe <jgg@mellanox.com>
Date: Wed, 15 Jan 2020 14:43:37 +0200
Subject: RDMA/mlx5: Fix handling of IOVA != user_va in ODP paths

Till recently it was not possible for userspace to specify a different
IOVA, but with the new ibv_reg_mr_iova() library call this can be done.

To compute the user_va we must compute:
  user_va = (iova - iova_start) + user_va_start

while being cautious of overflow and other math problems.

The iova is not reliably stored in the mmkey when the MR is created. Only
the cached creation path (the common one) set it, so it must also be set
when creating uncached MRs.

Fix the weird use of iova when computing the starting page index in the
MR. In the normal case, when iova == umem.address:
  iova & (~(BIT(page_shift) - 1)) ==
  ALIGN_DOWN(umem.address, odp->page_size) ==
  ib_umem_start(odp)

And when iova is different using it in math with a user_va is wrong.

Finally, do not allow an implicit ODP to be created with a non-zero IOVA
as we have no support for that.

Fixes: 7bdf65d411c1 ("IB/mlx5: Handle page faults")
Signed-off-by: Moni Shoua <monis@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
---
 drivers/infiniband/hw/mlx5/mr.c  |  2 ++
 drivers/infiniband/hw/mlx5/odp.c | 19 +++++++++++++------
 2 files changed, 15 insertions(+), 6 deletions(-)

diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c
index f79bb44b94fe..44a0ee6bd9f1 100644
--- a/drivers/infiniband/hw/mlx5/mr.c
+++ b/drivers/infiniband/hw/mlx5/mr.c
@@ -1246,6 +1246,8 @@ struct ib_mr *mlx5_ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
 
 	if (IS_ENABLED(CONFIG_INFINIBAND_ON_DEMAND_PAGING) && !start &&
 	    length == U64_MAX) {
+		if (virt_addr != start)
+			return ERR_PTR(-EINVAL);
 		if (!(access_flags & IB_ACCESS_ON_DEMAND) ||
 		    !(dev->odp_caps.general_caps & IB_ODP_SUPPORT_IMPLICIT))
 			return ERR_PTR(-EINVAL);
diff --git a/drivers/infiniband/hw/mlx5/odp.c b/drivers/infiniband/hw/mlx5/odp.c
index 3642c6a491c2..0afb0042bd53 100644
--- a/drivers/infiniband/hw/mlx5/odp.c
+++ b/drivers/infiniband/hw/mlx5/odp.c
@@ -624,11 +624,10 @@ static int pagefault_real_mr(struct mlx5_ib_mr *mr, struct ib_umem_odp *odp,
 	bool downgrade = flags & MLX5_PF_FLAGS_DOWNGRADE;
 	unsigned long current_seq;
 	u64 access_mask;
-	u64 start_idx, page_mask;
+	u64 start_idx;
 
 	page_shift = odp->page_shift;
-	page_mask = ~(BIT(page_shift) - 1);
-	start_idx = (user_va - (mr->mmkey.iova & page_mask)) >> page_shift;
+	start_idx = (user_va - ib_umem_start(odp)) >> page_shift;
 	access_mask = ODP_READ_ALLOWED_BIT;
 
 	if (odp->umem.writable && !downgrade)
@@ -767,11 +766,19 @@ static int pagefault_mr(struct mlx5_ib_mr *mr, u64 io_virt, size_t bcnt,
 {
 	struct ib_umem_odp *odp = to_ib_umem_odp(mr->umem);
 
+	if (unlikely(io_virt < mr->mmkey.iova))
+		return -EFAULT;
+
 	if (!odp->is_implicit_odp) {
-		if (unlikely(io_virt < ib_umem_start(odp) ||
-			     ib_umem_end(odp) - io_virt < bcnt))
+		u64 user_va;
+
+		if (check_add_overflow(io_virt - mr->mmkey.iova,
+				       (u64)odp->umem.address, &user_va))
+			return -EFAULT;
+		if (unlikely(user_va >= ib_umem_end(odp) ||
+			     ib_umem_end(odp) - user_va < bcnt))
 			return -EFAULT;
-		return pagefault_real_mr(mr, odp, io_virt, bcnt, bytes_mapped,
+		return pagefault_real_mr(mr, odp, user_va, bcnt, bytes_mapped,
 					 flags);
 	}
 	return pagefault_implicit_mr(mr, odp, io_virt, bcnt, bytes_mapped,
-- 
cgit v1.2.3-59-g8ed1b


From c4c86abb3f9fb3d5bc113fb996298a30ec63e07b Mon Sep 17 00:00:00 2001
From: Hans Westgaard Ry <hans.westgaard.ry@oracle.com>
Date: Wed, 15 Jan 2020 14:43:38 +0200
Subject: net/rds: Detect need of On-Demand-Paging memory registration

Add code to check if memory intended for RDMA is FS-DAX-memory. RDS
will fail with error code EOPNOTSUPP if FS-DAX-memory is detected.

Signed-off-by: Hans Westgaard Ry <hans.westgaard.ry@oracle.com>
Acked-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
---
 net/rds/rdma.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/net/rds/rdma.c b/net/rds/rdma.c
index 916f5ec373d8..eb23c38ce2b3 100644
--- a/net/rds/rdma.c
+++ b/net/rds/rdma.c
@@ -156,11 +156,13 @@ void rds_rdma_drop_keys(struct rds_sock *rs)
 static int rds_pin_pages(unsigned long user_addr, unsigned int nr_pages,
 			struct page **pages, int write)
 {
+	unsigned int gup_flags = FOLL_LONGTERM;
 	int ret;
 
-	ret = get_user_pages_fast(user_addr, nr_pages, write ? FOLL_WRITE : 0,
-				  pages);
+	if (write)
+		gup_flags |= FOLL_WRITE;
 
+	ret = get_user_pages_fast(user_addr, nr_pages, gup_flags, pages);
 	if (ret >= 0 && ret < nr_pages) {
 		while (ret--)
 			put_page(pages[ret]);
-- 
cgit v1.2.3-59-g8ed1b


From 858e284f0ec18bff2620d9a6afe764dc683f8ba1 Mon Sep 17 00:00:00 2001
From: Brian Vazquez <brianvv@google.com>
Date: Wed, 15 Jan 2020 20:59:18 -0800
Subject: libbpf: Fix unneeded extra initialization in bpf_map_batch_common

bpf_attr doesn't required to be declared with '= {}' as memset is used
in the code.

Fixes: 2ab3d86ea1859 ("libbpf: Add libbpf support to batch ops")
Reported-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Brian Vazquez <brianvv@google.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20200116045918.75597-1-brianvv@google.com
---
 tools/lib/bpf/bpf.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/lib/bpf/bpf.c b/tools/lib/bpf/bpf.c
index 317727d61214..ed42b006533c 100644
--- a/tools/lib/bpf/bpf.c
+++ b/tools/lib/bpf/bpf.c
@@ -457,7 +457,7 @@ static int bpf_map_batch_common(int cmd, int fd, void  *in_batch,
 				__u32 *count,
 				const struct bpf_map_batch_opts *opts)
 {
-	union bpf_attr attr = {};
+	union bpf_attr attr;
 	int ret;
 
 	if (!OPTS_VALID(opts, bpf_map_batch_opts))
-- 
cgit v1.2.3-59-g8ed1b


From 4a7faaf4add3108522512c16c9ee08fb2ad4525e Mon Sep 17 00:00:00 2001
From: Jeremy Sowden <jeremy@azazel.net>
Date: Wed, 1 Jan 2020 13:41:32 +0000
Subject: netfilter: nft_bitwise: correct uapi header comment.

The comment documenting how bitwise expressions work includes a table
which summarizes the mask and xor arguments combined to express the
supported boolean operations.  However, the row for OR:

 mask    xor
 0       x

is incorrect.

  dreg = (sreg & 0) ^ x

is not equivalent to:

  dreg = sreg | x

What the code actually does is:

  dreg = (sreg & ~x) ^ x

Update the documentation to match.

Signed-off-by: Jeremy Sowden <jeremy@azazel.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/uapi/linux/netfilter/nf_tables.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h
index e237ecbdcd8a..dd4611767933 100644
--- a/include/uapi/linux/netfilter/nf_tables.h
+++ b/include/uapi/linux/netfilter/nf_tables.h
@@ -501,7 +501,7 @@ enum nft_immediate_attributes {
  *
  * 		mask	xor
  * NOT:		1	1
- * OR:		0	x
+ * OR:		~x	x
  * XOR:		1	x
  * AND:		x	0
  */
-- 
cgit v1.2.3-59-g8ed1b


From 79b9b685dde1d1bf43cf84163c76953dc3781c85 Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Sun, 5 Jan 2020 22:26:38 +0100
Subject: netfilter: flowtable: fetch stats only if flow is still alive

Do not fetch statistics if flow has expired since it might not in
hardware anymore. After this update, remove the FLOW_OFFLOAD_HW_DYING
check from nf_flow_offload_stats() since this flag is never set on.

Fixes: c29f74e0df7a ("netfilter: nf_flow_table: hardware offload support")
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Acked-by: wenxu <wenxu@ucloud.cn>
---
 net/netfilter/nf_flow_table_core.c    | 5 ++---
 net/netfilter/nf_flow_table_offload.c | 3 +--
 2 files changed, 3 insertions(+), 5 deletions(-)

diff --git a/net/netfilter/nf_flow_table_core.c b/net/netfilter/nf_flow_table_core.c
index e33a73cb1f42..9e6de2bbeccb 100644
--- a/net/netfilter/nf_flow_table_core.c
+++ b/net/netfilter/nf_flow_table_core.c
@@ -348,9 +348,6 @@ static void nf_flow_offload_gc_step(struct flow_offload *flow, void *data)
 {
 	struct nf_flowtable *flow_table = data;
 
-	if (flow->flags & FLOW_OFFLOAD_HW)
-		nf_flow_offload_stats(flow_table, flow);
-
 	if (nf_flow_has_expired(flow) || nf_ct_is_dying(flow->ct) ||
 	    (flow->flags & (FLOW_OFFLOAD_DYING | FLOW_OFFLOAD_TEARDOWN))) {
 		if (flow->flags & FLOW_OFFLOAD_HW) {
@@ -361,6 +358,8 @@ static void nf_flow_offload_gc_step(struct flow_offload *flow, void *data)
 		} else {
 			flow_offload_del(flow_table, flow);
 		}
+	} else if (flow->flags & FLOW_OFFLOAD_HW) {
+		nf_flow_offload_stats(flow_table, flow);
 	}
 }
 
diff --git a/net/netfilter/nf_flow_table_offload.c b/net/netfilter/nf_flow_table_offload.c
index d06969af1085..4d1e81e2880f 100644
--- a/net/netfilter/nf_flow_table_offload.c
+++ b/net/netfilter/nf_flow_table_offload.c
@@ -784,8 +784,7 @@ void nf_flow_offload_stats(struct nf_flowtable *flowtable,
 	__s32 delta;
 
 	delta = nf_flow_timeout_delta(flow->timeout);
-	if ((delta >= (9 * NF_FLOW_TIMEOUT) / 10) ||
-	    flow->flags & FLOW_OFFLOAD_HW_DYING)
+	if ((delta >= (9 * NF_FLOW_TIMEOUT) / 10))
 		return;
 
 	offload = kzalloc(sizeof(struct flow_offload_work), GFP_ATOMIC);
-- 
cgit v1.2.3-59-g8ed1b


From a7521a60a5f3e1f58a015fedb6e69aed40455feb Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Mon, 6 Jan 2020 12:42:55 +0100
Subject: netfilter: flowtable: restrict flow dissector match on meta ingress
 device

Set on FLOW_DISSECTOR_KEY_META meta key using flow tuple ingress interface.

Fixes: c29f74e0df7a ("netfilter: nf_flow_table: hardware offload support")
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nf_flow_table_offload.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/net/netfilter/nf_flow_table_offload.c b/net/netfilter/nf_flow_table_offload.c
index 4d1e81e2880f..b879e673953f 100644
--- a/net/netfilter/nf_flow_table_offload.c
+++ b/net/netfilter/nf_flow_table_offload.c
@@ -24,6 +24,7 @@ struct flow_offload_work {
 };
 
 struct nf_flow_key {
+	struct flow_dissector_key_meta			meta;
 	struct flow_dissector_key_control		control;
 	struct flow_dissector_key_basic			basic;
 	union {
@@ -55,6 +56,7 @@ static int nf_flow_rule_match(struct nf_flow_match *match,
 	struct nf_flow_key *mask = &match->mask;
 	struct nf_flow_key *key = &match->key;
 
+	NF_FLOW_DISSECTOR(match, FLOW_DISSECTOR_KEY_META, meta);
 	NF_FLOW_DISSECTOR(match, FLOW_DISSECTOR_KEY_CONTROL, control);
 	NF_FLOW_DISSECTOR(match, FLOW_DISSECTOR_KEY_BASIC, basic);
 	NF_FLOW_DISSECTOR(match, FLOW_DISSECTOR_KEY_IPV4_ADDRS, ipv4);
@@ -62,6 +64,9 @@ static int nf_flow_rule_match(struct nf_flow_match *match,
 	NF_FLOW_DISSECTOR(match, FLOW_DISSECTOR_KEY_TCP, tcp);
 	NF_FLOW_DISSECTOR(match, FLOW_DISSECTOR_KEY_PORTS, tp);
 
+	key->meta.ingress_ifindex = tuple->iifidx;
+	mask->meta.ingress_ifindex = 0xffffffff;
+
 	switch (tuple->l3proto) {
 	case AF_INET:
 		key->control.addr_type = FLOW_DISSECTOR_KEY_IPV4_ADDRS;
@@ -105,7 +110,8 @@ static int nf_flow_rule_match(struct nf_flow_match *match,
 	key->tp.dst = tuple->dst_port;
 	mask->tp.dst = 0xffff;
 
-	match->dissector.used_keys |= BIT(FLOW_DISSECTOR_KEY_CONTROL) |
+	match->dissector.used_keys |= BIT(FLOW_DISSECTOR_KEY_META) |
+				      BIT(FLOW_DISSECTOR_KEY_CONTROL) |
 				      BIT(FLOW_DISSECTOR_KEY_BASIC) |
 				      BIT(FLOW_DISSECTOR_KEY_PORTS);
 	return 0;
-- 
cgit v1.2.3-59-g8ed1b


From 87265d842c59d93798b380afe3288a2bdc19bcd9 Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Sun, 5 Jan 2020 20:48:51 +0100
Subject: netfilter: flowtable: add nf_flow_offload_work_alloc()

Add helper function to allocate and initialize flow offload work and use
it to consolidate existing code.

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nf_flow_table_offload.c | 38 ++++++++++++++++++++---------------
 1 file changed, 22 insertions(+), 16 deletions(-)

diff --git a/net/netfilter/nf_flow_table_offload.c b/net/netfilter/nf_flow_table_offload.c
index b879e673953f..d161623107a1 100644
--- a/net/netfilter/nf_flow_table_offload.c
+++ b/net/netfilter/nf_flow_table_offload.c
@@ -748,21 +748,35 @@ static void flow_offload_queue_work(struct flow_offload_work *offload)
 	schedule_work(&nf_flow_offload_work);
 }
 
-void nf_flow_offload_add(struct nf_flowtable *flowtable,
-			 struct flow_offload *flow)
+static struct flow_offload_work *
+nf_flow_offload_work_alloc(struct nf_flowtable *flowtable,
+			   struct flow_offload *flow, unsigned int cmd)
 {
 	struct flow_offload_work *offload;
 
 	offload = kmalloc(sizeof(struct flow_offload_work), GFP_ATOMIC);
 	if (!offload)
-		return;
+		return NULL;
 
-	offload->cmd = FLOW_CLS_REPLACE;
+	offload->cmd = cmd;
 	offload->flow = flow;
 	offload->priority = flowtable->priority;
 	offload->flowtable = flowtable;
-	flow->flags |= FLOW_OFFLOAD_HW;
 
+	return offload;
+}
+
+
+void nf_flow_offload_add(struct nf_flowtable *flowtable,
+			 struct flow_offload *flow)
+{
+	struct flow_offload_work *offload;
+
+	offload = nf_flow_offload_work_alloc(flowtable, flow, FLOW_CLS_REPLACE);
+	if (!offload)
+		return;
+
+	flow->flags |= FLOW_OFFLOAD_HW;
 	flow_offload_queue_work(offload);
 }
 
@@ -771,15 +785,11 @@ void nf_flow_offload_del(struct nf_flowtable *flowtable,
 {
 	struct flow_offload_work *offload;
 
-	offload = kzalloc(sizeof(struct flow_offload_work), GFP_ATOMIC);
+	offload = nf_flow_offload_work_alloc(flowtable, flow, FLOW_CLS_DESTROY);
 	if (!offload)
 		return;
 
-	offload->cmd = FLOW_CLS_DESTROY;
-	offload->flow = flow;
-	offload->flow->flags |= FLOW_OFFLOAD_HW_DYING;
-	offload->flowtable = flowtable;
-
+	flow->flags |= FLOW_OFFLOAD_HW_DYING;
 	flow_offload_queue_work(offload);
 }
 
@@ -793,14 +803,10 @@ void nf_flow_offload_stats(struct nf_flowtable *flowtable,
 	if ((delta >= (9 * NF_FLOW_TIMEOUT) / 10))
 		return;
 
-	offload = kzalloc(sizeof(struct flow_offload_work), GFP_ATOMIC);
+	offload = nf_flow_offload_work_alloc(flowtable, flow, FLOW_CLS_STATS);
 	if (!offload)
 		return;
 
-	offload->cmd = FLOW_CLS_STATS;
-	offload->flow = flow;
-	offload->flowtable = flowtable;
-
 	flow_offload_queue_work(offload);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 445db8d09659eb27bcd5920cb91d91686f0197d0 Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Sun, 5 Jan 2020 22:00:57 +0100
Subject: netfilter: flowtable: remove dying bit, use teardown bit instead

The dying bit removes the conntrack entry if the netdev that owns this
flow is going down. Instead, use the teardown mechanism to push back the
flow to conntrack to let the classic software path decide what to do
with it.

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_flow_table.h | 5 -----
 net/netfilter/nf_flow_table_core.c    | 8 +++-----
 2 files changed, 3 insertions(+), 10 deletions(-)

diff --git a/include/net/netfilter/nf_flow_table.h b/include/net/netfilter/nf_flow_table.h
index 415b8f49d150..4ad924d5f983 100644
--- a/include/net/netfilter/nf_flow_table.h
+++ b/include/net/netfilter/nf_flow_table.h
@@ -85,7 +85,6 @@ struct flow_offload_tuple_rhash {
 
 #define FLOW_OFFLOAD_SNAT	0x1
 #define FLOW_OFFLOAD_DNAT	0x2
-#define FLOW_OFFLOAD_DYING	0x4
 #define FLOW_OFFLOAD_TEARDOWN	0x8
 #define FLOW_OFFLOAD_HW		0x10
 #define FLOW_OFFLOAD_HW_DYING	0x20
@@ -134,10 +133,6 @@ int nf_flow_table_init(struct nf_flowtable *flow_table);
 void nf_flow_table_free(struct nf_flowtable *flow_table);
 
 void flow_offload_teardown(struct flow_offload *flow);
-static inline void flow_offload_dead(struct flow_offload *flow)
-{
-	flow->flags |= FLOW_OFFLOAD_DYING;
-}
 
 int nf_flow_snat_port(const struct flow_offload *flow,
 		      struct sk_buff *skb, unsigned int thoff,
diff --git a/net/netfilter/nf_flow_table_core.c b/net/netfilter/nf_flow_table_core.c
index 9e6de2bbeccb..a9ed93a9e007 100644
--- a/net/netfilter/nf_flow_table_core.c
+++ b/net/netfilter/nf_flow_table_core.c
@@ -182,8 +182,6 @@ void flow_offload_free(struct flow_offload *flow)
 	default:
 		break;
 	}
-	if (flow->flags & FLOW_OFFLOAD_DYING)
-		nf_ct_delete(flow->ct, 0, 0);
 	nf_ct_put(flow->ct);
 	kfree_rcu(flow, rcu_head);
 }
@@ -300,7 +298,7 @@ flow_offload_lookup(struct nf_flowtable *flow_table,
 
 	dir = tuplehash->tuple.dir;
 	flow = container_of(tuplehash, struct flow_offload, tuplehash[dir]);
-	if (flow->flags & (FLOW_OFFLOAD_DYING | FLOW_OFFLOAD_TEARDOWN))
+	if (flow->flags & FLOW_OFFLOAD_TEARDOWN)
 		return NULL;
 
 	if (unlikely(nf_ct_is_dying(flow->ct)))
@@ -349,7 +347,7 @@ static void nf_flow_offload_gc_step(struct flow_offload *flow, void *data)
 	struct nf_flowtable *flow_table = data;
 
 	if (nf_flow_has_expired(flow) || nf_ct_is_dying(flow->ct) ||
-	    (flow->flags & (FLOW_OFFLOAD_DYING | FLOW_OFFLOAD_TEARDOWN))) {
+	    (flow->flags & FLOW_OFFLOAD_TEARDOWN)) {
 		if (flow->flags & FLOW_OFFLOAD_HW) {
 			if (!(flow->flags & FLOW_OFFLOAD_HW_DYING))
 				nf_flow_offload_del(flow_table, flow);
@@ -523,7 +521,7 @@ static void nf_flow_table_do_cleanup(struct flow_offload *flow, void *data)
 	if (net_eq(nf_ct_net(flow->ct), dev_net(dev)) &&
 	    (flow->tuplehash[0].tuple.iifidx == dev->ifindex ||
 	     flow->tuplehash[1].tuple.iifidx == dev->ifindex))
-		flow_offload_dead(flow);
+		flow_offload_teardown(flow);
 }
 
 static void nf_flow_table_iterate_cleanup(struct nf_flowtable *flowtable,
-- 
cgit v1.2.3-59-g8ed1b


From 355a8b13f87a8964ebe785b065f1388a1bd00c7e Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Sun, 5 Jan 2020 20:41:15 +0100
Subject: netfilter: flowtable: use atomic bitwise operations for flow flags

Originally, all flow flag bits were set on only from the workqueue. With
the introduction of the flow teardown state and hardware offload this is
no longer true. Let's be safe and use atomic bitwise operation to
operation with flow flags.

Fixes: 59c466dd68e7 ("netfilter: nf_flow_table: add a new flow state for tearing down offloading")
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_flow_table.h | 16 +++++++++-------
 net/netfilter/nf_flow_table_core.c    | 20 ++++++++++----------
 net/netfilter/nf_flow_table_ip.c      |  8 ++++----
 net/netfilter/nf_flow_table_offload.c | 20 ++++++++++----------
 4 files changed, 33 insertions(+), 31 deletions(-)

diff --git a/include/net/netfilter/nf_flow_table.h b/include/net/netfilter/nf_flow_table.h
index 4ad924d5f983..5a10e28c3e40 100644
--- a/include/net/netfilter/nf_flow_table.h
+++ b/include/net/netfilter/nf_flow_table.h
@@ -83,12 +83,14 @@ struct flow_offload_tuple_rhash {
 	struct flow_offload_tuple	tuple;
 };
 
-#define FLOW_OFFLOAD_SNAT	0x1
-#define FLOW_OFFLOAD_DNAT	0x2
-#define FLOW_OFFLOAD_TEARDOWN	0x8
-#define FLOW_OFFLOAD_HW		0x10
-#define FLOW_OFFLOAD_HW_DYING	0x20
-#define FLOW_OFFLOAD_HW_DEAD	0x40
+enum nf_flow_flags {
+	NF_FLOW_SNAT,
+	NF_FLOW_DNAT,
+	NF_FLOW_TEARDOWN,
+	NF_FLOW_HW,
+	NF_FLOW_HW_DYING,
+	NF_FLOW_HW_DEAD,
+};
 
 enum flow_offload_type {
 	NF_FLOW_OFFLOAD_UNSPEC	= 0,
@@ -98,7 +100,7 @@ enum flow_offload_type {
 struct flow_offload {
 	struct flow_offload_tuple_rhash		tuplehash[FLOW_OFFLOAD_DIR_MAX];
 	struct nf_conn				*ct;
-	u16					flags;
+	unsigned long				flags;
 	u16					type;
 	u32					timeout;
 	struct rcu_head				rcu_head;
diff --git a/net/netfilter/nf_flow_table_core.c b/net/netfilter/nf_flow_table_core.c
index a9ed93a9e007..9f134f44d139 100644
--- a/net/netfilter/nf_flow_table_core.c
+++ b/net/netfilter/nf_flow_table_core.c
@@ -61,9 +61,9 @@ struct flow_offload *flow_offload_alloc(struct nf_conn *ct)
 	flow_offload_fill_dir(flow, FLOW_OFFLOAD_DIR_REPLY);
 
 	if (ct->status & IPS_SRC_NAT)
-		flow->flags |= FLOW_OFFLOAD_SNAT;
+		__set_bit(NF_FLOW_SNAT, &flow->flags);
 	if (ct->status & IPS_DST_NAT)
-		flow->flags |= FLOW_OFFLOAD_DNAT;
+		__set_bit(NF_FLOW_DNAT, &flow->flags);
 
 	return flow;
 
@@ -269,7 +269,7 @@ static void flow_offload_del(struct nf_flowtable *flow_table,
 
 	if (nf_flow_has_expired(flow))
 		flow_offload_fixup_ct(flow->ct);
-	else if (flow->flags & FLOW_OFFLOAD_TEARDOWN)
+	else if (test_bit(NF_FLOW_TEARDOWN, &flow->flags))
 		flow_offload_fixup_ct_timeout(flow->ct);
 
 	flow_offload_free(flow);
@@ -277,7 +277,7 @@ static void flow_offload_del(struct nf_flowtable *flow_table,
 
 void flow_offload_teardown(struct flow_offload *flow)
 {
-	flow->flags |= FLOW_OFFLOAD_TEARDOWN;
+	set_bit(NF_FLOW_TEARDOWN, &flow->flags);
 
 	flow_offload_fixup_ct_state(flow->ct);
 }
@@ -298,7 +298,7 @@ flow_offload_lookup(struct nf_flowtable *flow_table,
 
 	dir = tuplehash->tuple.dir;
 	flow = container_of(tuplehash, struct flow_offload, tuplehash[dir]);
-	if (flow->flags & FLOW_OFFLOAD_TEARDOWN)
+	if (test_bit(NF_FLOW_TEARDOWN, &flow->flags))
 		return NULL;
 
 	if (unlikely(nf_ct_is_dying(flow->ct)))
@@ -347,16 +347,16 @@ static void nf_flow_offload_gc_step(struct flow_offload *flow, void *data)
 	struct nf_flowtable *flow_table = data;
 
 	if (nf_flow_has_expired(flow) || nf_ct_is_dying(flow->ct) ||
-	    (flow->flags & FLOW_OFFLOAD_TEARDOWN)) {
-		if (flow->flags & FLOW_OFFLOAD_HW) {
-			if (!(flow->flags & FLOW_OFFLOAD_HW_DYING))
+	    test_bit(NF_FLOW_TEARDOWN, &flow->flags)) {
+		if (test_bit(NF_FLOW_HW, &flow->flags)) {
+			if (!test_bit(NF_FLOW_HW_DYING, &flow->flags))
 				nf_flow_offload_del(flow_table, flow);
-			else if (flow->flags & FLOW_OFFLOAD_HW_DEAD)
+			else if (test_bit(NF_FLOW_HW_DEAD, &flow->flags))
 				flow_offload_del(flow_table, flow);
 		} else {
 			flow_offload_del(flow_table, flow);
 		}
-	} else if (flow->flags & FLOW_OFFLOAD_HW) {
+	} else if (test_bit(NF_FLOW_HW, &flow->flags)) {
 		nf_flow_offload_stats(flow_table, flow);
 	}
 }
diff --git a/net/netfilter/nf_flow_table_ip.c b/net/netfilter/nf_flow_table_ip.c
index 7ea2ddc2aa93..f4ccb5f5008b 100644
--- a/net/netfilter/nf_flow_table_ip.c
+++ b/net/netfilter/nf_flow_table_ip.c
@@ -144,11 +144,11 @@ static int nf_flow_nat_ip(const struct flow_offload *flow, struct sk_buff *skb,
 {
 	struct iphdr *iph = ip_hdr(skb);
 
-	if (flow->flags & FLOW_OFFLOAD_SNAT &&
+	if (test_bit(NF_FLOW_SNAT, &flow->flags) &&
 	    (nf_flow_snat_port(flow, skb, thoff, iph->protocol, dir) < 0 ||
 	     nf_flow_snat_ip(flow, skb, iph, thoff, dir) < 0))
 		return -1;
-	if (flow->flags & FLOW_OFFLOAD_DNAT &&
+	if (test_bit(NF_FLOW_DNAT, &flow->flags) &&
 	    (nf_flow_dnat_port(flow, skb, thoff, iph->protocol, dir) < 0 ||
 	     nf_flow_dnat_ip(flow, skb, iph, thoff, dir) < 0))
 		return -1;
@@ -414,11 +414,11 @@ static int nf_flow_nat_ipv6(const struct flow_offload *flow,
 	struct ipv6hdr *ip6h = ipv6_hdr(skb);
 	unsigned int thoff = sizeof(*ip6h);
 
-	if (flow->flags & FLOW_OFFLOAD_SNAT &&
+	if (test_bit(NF_FLOW_SNAT, &flow->flags) &&
 	    (nf_flow_snat_port(flow, skb, thoff, ip6h->nexthdr, dir) < 0 ||
 	     nf_flow_snat_ipv6(flow, skb, ip6h, thoff, dir) < 0))
 		return -1;
-	if (flow->flags & FLOW_OFFLOAD_DNAT &&
+	if (test_bit(NF_FLOW_DNAT, &flow->flags) &&
 	    (nf_flow_dnat_port(flow, skb, thoff, ip6h->nexthdr, dir) < 0 ||
 	     nf_flow_dnat_ipv6(flow, skb, ip6h, thoff, dir) < 0))
 		return -1;
diff --git a/net/netfilter/nf_flow_table_offload.c b/net/netfilter/nf_flow_table_offload.c
index d161623107a1..8a1fe391666e 100644
--- a/net/netfilter/nf_flow_table_offload.c
+++ b/net/netfilter/nf_flow_table_offload.c
@@ -450,16 +450,16 @@ int nf_flow_rule_route_ipv4(struct net *net, const struct flow_offload *flow,
 	    flow_offload_eth_dst(net, flow, dir, flow_rule) < 0)
 		return -1;
 
-	if (flow->flags & FLOW_OFFLOAD_SNAT) {
+	if (test_bit(NF_FLOW_SNAT, &flow->flags)) {
 		flow_offload_ipv4_snat(net, flow, dir, flow_rule);
 		flow_offload_port_snat(net, flow, dir, flow_rule);
 	}
-	if (flow->flags & FLOW_OFFLOAD_DNAT) {
+	if (test_bit(NF_FLOW_DNAT, &flow->flags)) {
 		flow_offload_ipv4_dnat(net, flow, dir, flow_rule);
 		flow_offload_port_dnat(net, flow, dir, flow_rule);
 	}
-	if (flow->flags & FLOW_OFFLOAD_SNAT ||
-	    flow->flags & FLOW_OFFLOAD_DNAT)
+	if (test_bit(NF_FLOW_SNAT, &flow->flags) ||
+	    test_bit(NF_FLOW_DNAT, &flow->flags))
 		flow_offload_ipv4_checksum(net, flow, flow_rule);
 
 	flow_offload_redirect(flow, dir, flow_rule);
@@ -476,11 +476,11 @@ int nf_flow_rule_route_ipv6(struct net *net, const struct flow_offload *flow,
 	    flow_offload_eth_dst(net, flow, dir, flow_rule) < 0)
 		return -1;
 
-	if (flow->flags & FLOW_OFFLOAD_SNAT) {
+	if (test_bit(NF_FLOW_SNAT, &flow->flags)) {
 		flow_offload_ipv6_snat(net, flow, dir, flow_rule);
 		flow_offload_port_snat(net, flow, dir, flow_rule);
 	}
-	if (flow->flags & FLOW_OFFLOAD_DNAT) {
+	if (test_bit(NF_FLOW_DNAT, &flow->flags)) {
 		flow_offload_ipv6_dnat(net, flow, dir, flow_rule);
 		flow_offload_port_dnat(net, flow, dir, flow_rule);
 	}
@@ -636,7 +636,7 @@ static void flow_offload_tuple_del(struct flow_offload_work *offload,
 	list_for_each_entry(block_cb, &flowtable->flow_block.cb_list, list)
 		block_cb->cb(TC_SETUP_CLSFLOWER, &cls_flow, block_cb->cb_priv);
 
-	offload->flow->flags |= FLOW_OFFLOAD_HW_DEAD;
+	set_bit(NF_FLOW_HW_DEAD, &offload->flow->flags);
 }
 
 static int flow_offload_rule_add(struct flow_offload_work *offload,
@@ -723,7 +723,7 @@ static void flow_offload_work_handler(struct work_struct *work)
 		case FLOW_CLS_REPLACE:
 			ret = flow_offload_work_add(offload);
 			if (ret < 0)
-				offload->flow->flags &= ~FLOW_OFFLOAD_HW;
+				__clear_bit(NF_FLOW_HW, &offload->flow->flags);
 			break;
 		case FLOW_CLS_DESTROY:
 			flow_offload_work_del(offload);
@@ -776,7 +776,7 @@ void nf_flow_offload_add(struct nf_flowtable *flowtable,
 	if (!offload)
 		return;
 
-	flow->flags |= FLOW_OFFLOAD_HW;
+	__set_bit(NF_FLOW_HW, &flow->flags);
 	flow_offload_queue_work(offload);
 }
 
@@ -789,7 +789,7 @@ void nf_flow_offload_del(struct nf_flowtable *flowtable,
 	if (!offload)
 		return;
 
-	flow->flags |= FLOW_OFFLOAD_HW_DYING;
+	set_bit(NF_FLOW_HW_DYING, &flow->flags);
 	flow_offload_queue_work(offload);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From a5449cdcaac5c78d62b8bea8f79158071f23da01 Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Tue, 7 Jan 2020 09:56:27 +0100
Subject: netfilter: flowtable: add nf_flowtable_hw_offload() helper function

This function checks for the NF_FLOWTABLE_HW_OFFLOAD flag, meaning that
the flowtable hardware offload is enabled.

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_flow_table.h | 5 +++++
 net/netfilter/nf_flow_table_core.c    | 2 +-
 net/netfilter/nf_flow_table_offload.c | 4 ++--
 3 files changed, 8 insertions(+), 3 deletions(-)

diff --git a/include/net/netfilter/nf_flow_table.h b/include/net/netfilter/nf_flow_table.h
index 5a10e28c3e40..9ee1eaeaab04 100644
--- a/include/net/netfilter/nf_flow_table.h
+++ b/include/net/netfilter/nf_flow_table.h
@@ -47,6 +47,11 @@ struct nf_flowtable {
 	possible_net_t			net;
 };
 
+static inline bool nf_flowtable_hw_offload(struct nf_flowtable *flowtable)
+{
+	return flowtable->flags & NF_FLOWTABLE_HW_OFFLOAD;
+}
+
 enum flow_offload_tuple_dir {
 	FLOW_OFFLOAD_DIR_ORIGINAL = IP_CT_DIR_ORIGINAL,
 	FLOW_OFFLOAD_DIR_REPLY = IP_CT_DIR_REPLY,
diff --git a/net/netfilter/nf_flow_table_core.c b/net/netfilter/nf_flow_table_core.c
index 9f134f44d139..e919bafd68d1 100644
--- a/net/netfilter/nf_flow_table_core.c
+++ b/net/netfilter/nf_flow_table_core.c
@@ -243,7 +243,7 @@ int flow_offload_add(struct nf_flowtable *flow_table, struct flow_offload *flow)
 		return err;
 	}
 
-	if (flow_table->flags & NF_FLOWTABLE_HW_OFFLOAD)
+	if (nf_flowtable_hw_offload(flow_table))
 		nf_flow_offload_add(flow_table, flow);
 
 	return 0;
diff --git a/net/netfilter/nf_flow_table_offload.c b/net/netfilter/nf_flow_table_offload.c
index 8a1fe391666e..b4c79fbb2d82 100644
--- a/net/netfilter/nf_flow_table_offload.c
+++ b/net/netfilter/nf_flow_table_offload.c
@@ -812,7 +812,7 @@ void nf_flow_offload_stats(struct nf_flowtable *flowtable,
 
 void nf_flow_table_offload_flush(struct nf_flowtable *flowtable)
 {
-	if (flowtable->flags & NF_FLOWTABLE_HW_OFFLOAD)
+	if (nf_flowtable_hw_offload(flowtable))
 		flush_work(&nf_flow_offload_work);
 }
 
@@ -849,7 +849,7 @@ int nf_flow_table_offload_setup(struct nf_flowtable *flowtable,
 	struct flow_block_offload bo = {};
 	int err;
 
-	if (!(flowtable->flags & NF_FLOWTABLE_HW_OFFLOAD))
+	if (!nf_flowtable_hw_offload(flowtable))
 		return 0;
 
 	if (!dev->netdev_ops->ndo_setup_tc)
-- 
cgit v1.2.3-59-g8ed1b


From f698fe40829b21088d323c8b0a7c626571528fc6 Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Mon, 6 Jan 2020 12:56:47 +0100
Subject: netfilter: flowtable: refresh flow if hardware offload fails

If nf_flow_offload_add() fails to add the flow to hardware, then the
NF_FLOW_HW_REFRESH flag bit is set and the flow remains in the flowtable
software path.

If flowtable hardware offload is enabled, this patch enqueues a new
request to offload this flow to hardware.

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_flow_table.h |  1 +
 net/netfilter/nf_flow_table_core.c    |  4 +++-
 net/netfilter/nf_flow_table_ip.c      | 13 +++++++++++++
 net/netfilter/nf_flow_table_offload.c | 14 +++++---------
 4 files changed, 22 insertions(+), 10 deletions(-)

diff --git a/include/net/netfilter/nf_flow_table.h b/include/net/netfilter/nf_flow_table.h
index 9ee1eaeaab04..e0f709d9d547 100644
--- a/include/net/netfilter/nf_flow_table.h
+++ b/include/net/netfilter/nf_flow_table.h
@@ -95,6 +95,7 @@ enum nf_flow_flags {
 	NF_FLOW_HW,
 	NF_FLOW_HW_DYING,
 	NF_FLOW_HW_DEAD,
+	NF_FLOW_HW_REFRESH,
 };
 
 enum flow_offload_type {
diff --git a/net/netfilter/nf_flow_table_core.c b/net/netfilter/nf_flow_table_core.c
index e919bafd68d1..7e91989a1b55 100644
--- a/net/netfilter/nf_flow_table_core.c
+++ b/net/netfilter/nf_flow_table_core.c
@@ -243,8 +243,10 @@ int flow_offload_add(struct nf_flowtable *flow_table, struct flow_offload *flow)
 		return err;
 	}
 
-	if (nf_flowtable_hw_offload(flow_table))
+	if (nf_flowtable_hw_offload(flow_table)) {
+		__set_bit(NF_FLOW_HW, &flow->flags);
 		nf_flow_offload_add(flow_table, flow);
+	}
 
 	return 0;
 }
diff --git a/net/netfilter/nf_flow_table_ip.c b/net/netfilter/nf_flow_table_ip.c
index f4ccb5f5008b..9e563fd3da0f 100644
--- a/net/netfilter/nf_flow_table_ip.c
+++ b/net/netfilter/nf_flow_table_ip.c
@@ -232,6 +232,13 @@ static unsigned int nf_flow_xmit_xfrm(struct sk_buff *skb,
 	return NF_STOLEN;
 }
 
+static bool nf_flow_offload_refresh(struct nf_flowtable *flow_table,
+				    struct flow_offload *flow)
+{
+	return nf_flowtable_hw_offload(flow_table) &&
+	       test_and_clear_bit(NF_FLOW_HW_REFRESH, &flow->flags);
+}
+
 unsigned int
 nf_flow_offload_ip_hook(void *priv, struct sk_buff *skb,
 			const struct nf_hook_state *state)
@@ -272,6 +279,9 @@ nf_flow_offload_ip_hook(void *priv, struct sk_buff *skb,
 	if (nf_flow_state_check(flow, ip_hdr(skb)->protocol, skb, thoff))
 		return NF_ACCEPT;
 
+	if (unlikely(nf_flow_offload_refresh(flow_table, flow)))
+		nf_flow_offload_add(flow_table, flow);
+
 	if (nf_flow_offload_dst_check(&rt->dst)) {
 		flow_offload_teardown(flow);
 		return NF_ACCEPT;
@@ -498,6 +508,9 @@ nf_flow_offload_ipv6_hook(void *priv, struct sk_buff *skb,
 				sizeof(*ip6h)))
 		return NF_ACCEPT;
 
+	if (unlikely(nf_flow_offload_refresh(flow_table, flow)))
+		nf_flow_offload_add(flow_table, flow);
+
 	if (nf_flow_offload_dst_check(&rt->dst)) {
 		flow_offload_teardown(flow);
 		return NF_ACCEPT;
diff --git a/net/netfilter/nf_flow_table_offload.c b/net/netfilter/nf_flow_table_offload.c
index b4c79fbb2d82..77b129f196c6 100644
--- a/net/netfilter/nf_flow_table_offload.c
+++ b/net/netfilter/nf_flow_table_offload.c
@@ -654,20 +654,20 @@ static int flow_offload_rule_add(struct flow_offload_work *offload,
 	return 0;
 }
 
-static int flow_offload_work_add(struct flow_offload_work *offload)
+static void flow_offload_work_add(struct flow_offload_work *offload)
 {
 	struct nf_flow_rule *flow_rule[FLOW_OFFLOAD_DIR_MAX];
 	int err;
 
 	err = nf_flow_offload_alloc(offload, flow_rule);
 	if (err < 0)
-		return -ENOMEM;
+		return;
 
 	err = flow_offload_rule_add(offload, flow_rule);
+	if (err < 0)
+		set_bit(NF_FLOW_HW_REFRESH, &offload->flow->flags);
 
 	nf_flow_offload_destroy(flow_rule);
-
-	return err;
 }
 
 static void flow_offload_work_del(struct flow_offload_work *offload)
@@ -712,7 +712,6 @@ static void flow_offload_work_handler(struct work_struct *work)
 {
 	struct flow_offload_work *offload, *next;
 	LIST_HEAD(offload_pending_list);
-	int ret;
 
 	spin_lock_bh(&flow_offload_pending_list_lock);
 	list_replace_init(&flow_offload_pending_list, &offload_pending_list);
@@ -721,9 +720,7 @@ static void flow_offload_work_handler(struct work_struct *work)
 	list_for_each_entry_safe(offload, next, &offload_pending_list, list) {
 		switch (offload->cmd) {
 		case FLOW_CLS_REPLACE:
-			ret = flow_offload_work_add(offload);
-			if (ret < 0)
-				__clear_bit(NF_FLOW_HW, &offload->flow->flags);
+			flow_offload_work_add(offload);
 			break;
 		case FLOW_CLS_DESTROY:
 			flow_offload_work_del(offload);
@@ -776,7 +773,6 @@ void nf_flow_offload_add(struct nf_flowtable *flowtable,
 	if (!offload)
 		return;
 
-	__set_bit(NF_FLOW_HW, &flow->flags);
 	flow_offload_queue_work(offload);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 28b3a4270c0fc064557e409111f2a678e64b6fa7 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Tue, 7 Jan 2020 12:25:10 +0100
Subject: netfilter: hashlimit: do not use indirect calls during gc

no need, just use a simple boolean to indicate we want to reap all
entries.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/xt_hashlimit.c | 22 ++++------------------
 1 file changed, 4 insertions(+), 18 deletions(-)

diff --git a/net/netfilter/xt_hashlimit.c b/net/netfilter/xt_hashlimit.c
index ced3fc8fad7c..bccd47cd7190 100644
--- a/net/netfilter/xt_hashlimit.c
+++ b/net/netfilter/xt_hashlimit.c
@@ -357,21 +357,7 @@ static int htable_create(struct net *net, struct hashlimit_cfg3 *cfg,
 	return 0;
 }
 
-static bool select_all(const struct xt_hashlimit_htable *ht,
-		       const struct dsthash_ent *he)
-{
-	return true;
-}
-
-static bool select_gc(const struct xt_hashlimit_htable *ht,
-		      const struct dsthash_ent *he)
-{
-	return time_after_eq(jiffies, he->expires);
-}
-
-static void htable_selective_cleanup(struct xt_hashlimit_htable *ht,
-			bool (*select)(const struct xt_hashlimit_htable *ht,
-				      const struct dsthash_ent *he))
+static void htable_selective_cleanup(struct xt_hashlimit_htable *ht, bool select_all)
 {
 	unsigned int i;
 
@@ -381,7 +367,7 @@ static void htable_selective_cleanup(struct xt_hashlimit_htable *ht,
 
 		spin_lock_bh(&ht->lock);
 		hlist_for_each_entry_safe(dh, n, &ht->hash[i], node) {
-			if ((*select)(ht, dh))
+			if (time_after_eq(jiffies, dh->expires) || select_all)
 				dsthash_free(ht, dh);
 		}
 		spin_unlock_bh(&ht->lock);
@@ -395,7 +381,7 @@ static void htable_gc(struct work_struct *work)
 
 	ht = container_of(work, struct xt_hashlimit_htable, gc_work.work);
 
-	htable_selective_cleanup(ht, select_gc);
+	htable_selective_cleanup(ht, false);
 
 	queue_delayed_work(system_power_efficient_wq,
 			   &ht->gc_work, msecs_to_jiffies(ht->cfg.gc_interval));
@@ -419,7 +405,7 @@ static void htable_destroy(struct xt_hashlimit_htable *hinfo)
 {
 	cancel_delayed_work_sync(&hinfo->gc_work);
 	htable_remove_proc_entry(hinfo);
-	htable_selective_cleanup(hinfo, select_all);
+	htable_selective_cleanup(hinfo, true);
 	kfree(hinfo->name);
 	vfree(hinfo);
 }
-- 
cgit v1.2.3-59-g8ed1b


From ae29045018c83495f8c5033afabbedc09f24d7c2 Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Mon, 13 Jan 2020 19:02:14 +0100
Subject: netfilter: flowtable: add nf_flow_offload_tuple() helper

Consolidate code to configure the flow_cls_offload structure into one
helper function.

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nf_flow_table_offload.c | 47 ++++++++++++++++++-----------------
 1 file changed, 24 insertions(+), 23 deletions(-)

diff --git a/net/netfilter/nf_flow_table_offload.c b/net/netfilter/nf_flow_table_offload.c
index 77b129f196c6..3cd8dc8714e3 100644
--- a/net/netfilter/nf_flow_table_offload.c
+++ b/net/netfilter/nf_flow_table_offload.c
@@ -592,23 +592,25 @@ static void nf_flow_offload_init(struct flow_cls_offload *cls_flow,
 	cls_flow->cookie = (unsigned long)tuple;
 }
 
-static int flow_offload_tuple_add(struct flow_offload_work *offload,
-				  struct nf_flow_rule *flow_rule,
-				  enum flow_offload_tuple_dir dir)
+static int nf_flow_offload_tuple(struct nf_flowtable *flowtable,
+				 struct flow_offload *flow,
+				 struct nf_flow_rule *flow_rule,
+				 enum flow_offload_tuple_dir dir,
+				 int priority, int cmd,
+				 struct list_head *block_cb_list)
 {
-	struct nf_flowtable *flowtable = offload->flowtable;
 	struct flow_cls_offload cls_flow = {};
 	struct flow_block_cb *block_cb;
 	struct netlink_ext_ack extack;
 	__be16 proto = ETH_P_ALL;
 	int err, i = 0;
 
-	nf_flow_offload_init(&cls_flow, proto, offload->priority,
-			     FLOW_CLS_REPLACE,
-			     &offload->flow->tuplehash[dir].tuple, &extack);
-	cls_flow.rule = flow_rule->rule;
+	nf_flow_offload_init(&cls_flow, proto, priority, cmd,
+			     &flow->tuplehash[dir].tuple, &extack);
+	if (cmd == FLOW_CLS_REPLACE)
+		cls_flow.rule = flow_rule->rule;
 
-	list_for_each_entry(block_cb, &flowtable->flow_block.cb_list, list) {
+	list_for_each_entry(block_cb, block_cb_list, list) {
 		err = block_cb->cb(TC_SETUP_CLSFLOWER, &cls_flow,
 				   block_cb->cb_priv);
 		if (err < 0)
@@ -620,23 +622,22 @@ static int flow_offload_tuple_add(struct flow_offload_work *offload,
 	return i;
 }
 
+static int flow_offload_tuple_add(struct flow_offload_work *offload,
+				  struct nf_flow_rule *flow_rule,
+				  enum flow_offload_tuple_dir dir)
+{
+	return nf_flow_offload_tuple(offload->flowtable, offload->flow,
+				     flow_rule, dir, offload->priority,
+				     FLOW_CLS_REPLACE,
+				     &offload->flowtable->flow_block.cb_list);
+}
+
 static void flow_offload_tuple_del(struct flow_offload_work *offload,
 				   enum flow_offload_tuple_dir dir)
 {
-	struct nf_flowtable *flowtable = offload->flowtable;
-	struct flow_cls_offload cls_flow = {};
-	struct flow_block_cb *block_cb;
-	struct netlink_ext_ack extack;
-	__be16 proto = ETH_P_ALL;
-
-	nf_flow_offload_init(&cls_flow, proto, offload->priority,
-			     FLOW_CLS_DESTROY,
-			     &offload->flow->tuplehash[dir].tuple, &extack);
-
-	list_for_each_entry(block_cb, &flowtable->flow_block.cb_list, list)
-		block_cb->cb(TC_SETUP_CLSFLOWER, &cls_flow, block_cb->cb_priv);
-
-	set_bit(NF_FLOW_HW_DEAD, &offload->flow->flags);
+	nf_flow_offload_tuple(offload->flowtable, offload->flow, NULL, dir,
+			      offload->priority, FLOW_CLS_DESTROY,
+			      &offload->flowtable->flow_block.cb_list);
 }
 
 static int flow_offload_rule_add(struct flow_offload_work *offload,
-- 
cgit v1.2.3-59-g8ed1b


From a7965d58ddab0253ddfae58bd5c7d2de46ef0f76 Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Mon, 13 Jan 2020 19:02:22 +0100
Subject: netfilter: flowtable: add nf_flow_table_offload_cmd()

Split nf_flow_table_offload_setup() in two functions to make it more
maintainable.

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nf_flow_table_offload.c | 40 ++++++++++++++++++++++++-----------
 1 file changed, 28 insertions(+), 12 deletions(-)

diff --git a/net/netfilter/nf_flow_table_offload.c b/net/netfilter/nf_flow_table_offload.c
index 3cd8dc8714e3..c8b70ffeef0c 100644
--- a/net/netfilter/nf_flow_table_offload.c
+++ b/net/netfilter/nf_flow_table_offload.c
@@ -838,12 +838,12 @@ static int nf_flow_table_block_setup(struct nf_flowtable *flowtable,
 	return err;
 }
 
-int nf_flow_table_offload_setup(struct nf_flowtable *flowtable,
-				struct net_device *dev,
-				enum flow_block_command cmd)
+static int nf_flow_table_offload_cmd(struct flow_block_offload *bo,
+				     struct nf_flowtable *flowtable,
+				     struct net_device *dev,
+				     enum flow_block_command cmd,
+				     struct netlink_ext_ack *extack)
 {
-	struct netlink_ext_ack extack = {};
-	struct flow_block_offload bo = {};
 	int err;
 
 	if (!nf_flowtable_hw_offload(flowtable))
@@ -852,14 +852,30 @@ int nf_flow_table_offload_setup(struct nf_flowtable *flowtable,
 	if (!dev->netdev_ops->ndo_setup_tc)
 		return -EOPNOTSUPP;
 
-	bo.net		= dev_net(dev);
-	bo.block	= &flowtable->flow_block;
-	bo.command	= cmd;
-	bo.binder_type	= FLOW_BLOCK_BINDER_TYPE_CLSACT_INGRESS;
-	bo.extack	= &extack;
-	INIT_LIST_HEAD(&bo.cb_list);
+	memset(bo, 0, sizeof(*bo));
+	bo->net		= dev_net(dev);
+	bo->block	= &flowtable->flow_block;
+	bo->command	= cmd;
+	bo->binder_type	= FLOW_BLOCK_BINDER_TYPE_CLSACT_INGRESS;
+	bo->extack	= extack;
+	INIT_LIST_HEAD(&bo->cb_list);
+
+	err = dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_FT, bo);
+	if (err < 0)
+		return err;
+
+	return 0;
+}
+
+int nf_flow_table_offload_setup(struct nf_flowtable *flowtable,
+				struct net_device *dev,
+				enum flow_block_command cmd)
+{
+	struct netlink_ext_ack extack = {};
+	struct flow_block_offload bo;
+	int err;
 
-	err = dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_FT, &bo);
+	err = nf_flow_table_offload_cmd(&bo, flowtable, dev, cmd, &extack);
 	if (err < 0)
 		return err;
 
-- 
cgit v1.2.3-59-g8ed1b


From fbf19ddf396b3526718a64e24ace316c64d3ed1f Mon Sep 17 00:00:00 2001
From: Jeremy Sowden <jeremy@azazel.net>
Date: Wed, 15 Jan 2020 20:05:48 +0000
Subject: netfilter: nf_tables: white-space fixes.

Indentation fixes for the parameters of a few nft functions.

Signed-off-by: Jeremy Sowden <jeremy@azazel.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nft_bitwise.c    | 4 ++--
 net/netfilter/nft_set_bitmap.c | 4 ++--
 net/netfilter/nft_set_hash.c   | 2 +-
 3 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/net/netfilter/nft_bitwise.c b/net/netfilter/nft_bitwise.c
index 10e9d50e4e19..e8ca1ec105f8 100644
--- a/net/netfilter/nft_bitwise.c
+++ b/net/netfilter/nft_bitwise.c
@@ -130,8 +130,8 @@ nla_put_failure:
 static struct nft_data zero;
 
 static int nft_bitwise_offload(struct nft_offload_ctx *ctx,
-                               struct nft_flow_rule *flow,
-                               const struct nft_expr *expr)
+			       struct nft_flow_rule *flow,
+			       const struct nft_expr *expr)
 {
 	const struct nft_bitwise *priv = nft_expr_priv(expr);
 	struct nft_offload_reg *reg = &ctx->regs[priv->dreg];
diff --git a/net/netfilter/nft_set_bitmap.c b/net/netfilter/nft_set_bitmap.c
index 087a056e34d1..87e8d9ba0c9b 100644
--- a/net/netfilter/nft_set_bitmap.c
+++ b/net/netfilter/nft_set_bitmap.c
@@ -259,8 +259,8 @@ static u64 nft_bitmap_privsize(const struct nlattr * const nla[],
 }
 
 static int nft_bitmap_init(const struct nft_set *set,
-			 const struct nft_set_desc *desc,
-			 const struct nlattr * const nla[])
+			   const struct nft_set_desc *desc,
+			   const struct nlattr * const nla[])
 {
 	struct nft_bitmap *priv = nft_set_priv(set);
 
diff --git a/net/netfilter/nft_set_hash.c b/net/netfilter/nft_set_hash.c
index b331a3c9a3a8..d350a7cd3af0 100644
--- a/net/netfilter/nft_set_hash.c
+++ b/net/netfilter/nft_set_hash.c
@@ -645,7 +645,7 @@ static bool nft_hash_estimate(const struct nft_set_desc *desc, u32 features,
 }
 
 static bool nft_hash_fast_estimate(const struct nft_set_desc *desc, u32 features,
-			      struct nft_set_estimate *est)
+				   struct nft_set_estimate *est)
 {
 	if (!desc->size)
 		return false;
-- 
cgit v1.2.3-59-g8ed1b


From 265ec7b0e8c35ea70e694da43aa8896e2614def9 Mon Sep 17 00:00:00 2001
From: Jeremy Sowden <jeremy@azazel.net>
Date: Wed, 15 Jan 2020 20:05:49 +0000
Subject: netfilter: bitwise: remove NULL comparisons from attribute checks.

In later patches, we will be adding more checks.  In order to be
consistent and prevent complaints from checkpatch.pl, replace the
existing comparisons with NULL with logical NOT operators.

Signed-off-by: Jeremy Sowden <jeremy@azazel.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nft_bitwise.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/net/netfilter/nft_bitwise.c b/net/netfilter/nft_bitwise.c
index e8ca1ec105f8..85605fb1e360 100644
--- a/net/netfilter/nft_bitwise.c
+++ b/net/netfilter/nft_bitwise.c
@@ -52,11 +52,11 @@ static int nft_bitwise_init(const struct nft_ctx *ctx,
 	u32 len;
 	int err;
 
-	if (tb[NFTA_BITWISE_SREG] == NULL ||
-	    tb[NFTA_BITWISE_DREG] == NULL ||
-	    tb[NFTA_BITWISE_LEN] == NULL ||
-	    tb[NFTA_BITWISE_MASK] == NULL ||
-	    tb[NFTA_BITWISE_XOR] == NULL)
+	if (!tb[NFTA_BITWISE_SREG] ||
+	    !tb[NFTA_BITWISE_DREG] ||
+	    !tb[NFTA_BITWISE_LEN]  ||
+	    !tb[NFTA_BITWISE_MASK] ||
+	    !tb[NFTA_BITWISE_XOR])
 		return -EINVAL;
 
 	err = nft_parse_u32_check(tb[NFTA_BITWISE_LEN], U8_MAX, &len);
-- 
cgit v1.2.3-59-g8ed1b


From 577c734a81e2f59aa4490071ebe074fc05d5540b Mon Sep 17 00:00:00 2001
From: Jeremy Sowden <jeremy@azazel.net>
Date: Wed, 15 Jan 2020 20:05:50 +0000
Subject: netfilter: bitwise: replace gotos with returns.

When dumping a bitwise expression, if any of the puts fails, we use goto
to jump to a label.  However, no clean-up is required and the only
statement at the label is a return.  Drop the goto's and return
immediately instead.

Signed-off-by: Jeremy Sowden <jeremy@azazel.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nft_bitwise.c | 13 +++++--------
 1 file changed, 5 insertions(+), 8 deletions(-)

diff --git a/net/netfilter/nft_bitwise.c b/net/netfilter/nft_bitwise.c
index 85605fb1e360..c15e9beb5243 100644
--- a/net/netfilter/nft_bitwise.c
+++ b/net/netfilter/nft_bitwise.c
@@ -107,24 +107,21 @@ static int nft_bitwise_dump(struct sk_buff *skb, const struct nft_expr *expr)
 	const struct nft_bitwise *priv = nft_expr_priv(expr);
 
 	if (nft_dump_register(skb, NFTA_BITWISE_SREG, priv->sreg))
-		goto nla_put_failure;
+		return -1;
 	if (nft_dump_register(skb, NFTA_BITWISE_DREG, priv->dreg))
-		goto nla_put_failure;
+		return -1;
 	if (nla_put_be32(skb, NFTA_BITWISE_LEN, htonl(priv->len)))
-		goto nla_put_failure;
+		return -1;
 
 	if (nft_data_dump(skb, NFTA_BITWISE_MASK, &priv->mask,
 			  NFT_DATA_VALUE, priv->len) < 0)
-		goto nla_put_failure;
+		return -1;
 
 	if (nft_data_dump(skb, NFTA_BITWISE_XOR, &priv->xor,
 			  NFT_DATA_VALUE, priv->len) < 0)
-		goto nla_put_failure;
+		return -1;
 
 	return 0;
-
-nla_put_failure:
-	return -1;
 }
 
 static struct nft_data zero;
-- 
cgit v1.2.3-59-g8ed1b


From 9d1f979986c2e29632b6a8f7a8ef8b3c7d24a48c Mon Sep 17 00:00:00 2001
From: Jeremy Sowden <jeremy@azazel.net>
Date: Wed, 15 Jan 2020 20:05:51 +0000
Subject: netfilter: bitwise: add NFTA_BITWISE_OP netlink attribute.

Add a new bitwise netlink attribute, NFTA_BITWISE_OP, which is set to a
value of a new enum, nft_bitwise_ops.  It describes the type of
operation an expression contains.  Currently, it only has one value:
NFT_BITWISE_BOOL.  More values will be added later to implement shifts.

Signed-off-by: Jeremy Sowden <jeremy@azazel.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/uapi/linux/netfilter/nf_tables.h | 12 ++++++++++++
 net/netfilter/nft_bitwise.c              | 16 ++++++++++++++++
 2 files changed, 28 insertions(+)

diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h
index dd4611767933..0cddf357281f 100644
--- a/include/uapi/linux/netfilter/nf_tables.h
+++ b/include/uapi/linux/netfilter/nf_tables.h
@@ -484,6 +484,16 @@ enum nft_immediate_attributes {
 };
 #define NFTA_IMMEDIATE_MAX	(__NFTA_IMMEDIATE_MAX - 1)
 
+/**
+ * enum nft_bitwise_ops - nf_tables bitwise operations
+ *
+ * @NFT_BITWISE_BOOL: mask-and-xor operation used to implement NOT, AND, OR and
+ *                    XOR boolean operations
+ */
+enum nft_bitwise_ops {
+	NFT_BITWISE_BOOL,
+};
+
 /**
  * enum nft_bitwise_attributes - nf_tables bitwise expression netlink attributes
  *
@@ -492,6 +502,7 @@ enum nft_immediate_attributes {
  * @NFTA_BITWISE_LEN: length of operands (NLA_U32)
  * @NFTA_BITWISE_MASK: mask value (NLA_NESTED: nft_data_attributes)
  * @NFTA_BITWISE_XOR: xor value (NLA_NESTED: nft_data_attributes)
+ * @NFTA_BITWISE_OP: type of operation (NLA_U32: nft_bitwise_ops)
  *
  * The bitwise expression performs the following operation:
  *
@@ -512,6 +523,7 @@ enum nft_bitwise_attributes {
 	NFTA_BITWISE_LEN,
 	NFTA_BITWISE_MASK,
 	NFTA_BITWISE_XOR,
+	NFTA_BITWISE_OP,
 	__NFTA_BITWISE_MAX
 };
 #define NFTA_BITWISE_MAX	(__NFTA_BITWISE_MAX - 1)
diff --git a/net/netfilter/nft_bitwise.c b/net/netfilter/nft_bitwise.c
index c15e9beb5243..6948df7b0587 100644
--- a/net/netfilter/nft_bitwise.c
+++ b/net/netfilter/nft_bitwise.c
@@ -18,6 +18,7 @@
 struct nft_bitwise {
 	enum nft_registers	sreg:8;
 	enum nft_registers	dreg:8;
+	enum nft_bitwise_ops	op:8;
 	u8			len;
 	struct nft_data		mask;
 	struct nft_data		xor;
@@ -41,6 +42,7 @@ static const struct nla_policy nft_bitwise_policy[NFTA_BITWISE_MAX + 1] = {
 	[NFTA_BITWISE_LEN]	= { .type = NLA_U32 },
 	[NFTA_BITWISE_MASK]	= { .type = NLA_NESTED },
 	[NFTA_BITWISE_XOR]	= { .type = NLA_NESTED },
+	[NFTA_BITWISE_OP]	= { .type = NLA_U32 },
 };
 
 static int nft_bitwise_init(const struct nft_ctx *ctx,
@@ -76,6 +78,18 @@ static int nft_bitwise_init(const struct nft_ctx *ctx,
 	if (err < 0)
 		return err;
 
+	if (tb[NFTA_BITWISE_OP]) {
+		priv->op = ntohl(nla_get_be32(tb[NFTA_BITWISE_OP]));
+		switch (priv->op) {
+		case NFT_BITWISE_BOOL:
+			break;
+		default:
+			return -EOPNOTSUPP;
+		}
+	} else {
+		priv->op = NFT_BITWISE_BOOL;
+	}
+
 	err = nft_data_init(NULL, &priv->mask, sizeof(priv->mask), &d1,
 			    tb[NFTA_BITWISE_MASK]);
 	if (err < 0)
@@ -112,6 +126,8 @@ static int nft_bitwise_dump(struct sk_buff *skb, const struct nft_expr *expr)
 		return -1;
 	if (nla_put_be32(skb, NFTA_BITWISE_LEN, htonl(priv->len)))
 		return -1;
+	if (nla_put_be32(skb, NFTA_BITWISE_OP, htonl(priv->op)))
+		return -1;
 
 	if (nft_data_dump(skb, NFTA_BITWISE_MASK, &priv->mask,
 			  NFT_DATA_VALUE, priv->len) < 0)
-- 
cgit v1.2.3-59-g8ed1b


From 3f8d9eb032ec76c35344a2453c4c4a0a29805e3f Mon Sep 17 00:00:00 2001
From: Jeremy Sowden <jeremy@azazel.net>
Date: Wed, 15 Jan 2020 20:05:52 +0000
Subject: netfilter: bitwise: add helper for initializing boolean operations.

Split the code specific to initializing bitwise boolean operations out
into a separate function.  A similar function will be added later for
shift operations.

Signed-off-by: Jeremy Sowden <jeremy@azazel.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nft_bitwise.c | 66 ++++++++++++++++++++++++++++-----------------
 1 file changed, 41 insertions(+), 25 deletions(-)

diff --git a/net/netfilter/nft_bitwise.c b/net/netfilter/nft_bitwise.c
index 6948df7b0587..d0cc5f753e52 100644
--- a/net/netfilter/nft_bitwise.c
+++ b/net/netfilter/nft_bitwise.c
@@ -45,20 +45,53 @@ static const struct nla_policy nft_bitwise_policy[NFTA_BITWISE_MAX + 1] = {
 	[NFTA_BITWISE_OP]	= { .type = NLA_U32 },
 };
 
+static int nft_bitwise_init_bool(struct nft_bitwise *priv,
+				 const struct nlattr *const tb[])
+{
+	struct nft_data_desc d1, d2;
+	int err;
+
+	if (!tb[NFTA_BITWISE_MASK] ||
+	    !tb[NFTA_BITWISE_XOR])
+		return -EINVAL;
+
+	err = nft_data_init(NULL, &priv->mask, sizeof(priv->mask), &d1,
+			    tb[NFTA_BITWISE_MASK]);
+	if (err < 0)
+		return err;
+	if (d1.type != NFT_DATA_VALUE || d1.len != priv->len) {
+		err = -EINVAL;
+		goto err1;
+	}
+
+	err = nft_data_init(NULL, &priv->xor, sizeof(priv->xor), &d2,
+			    tb[NFTA_BITWISE_XOR]);
+	if (err < 0)
+		goto err1;
+	if (d2.type != NFT_DATA_VALUE || d2.len != priv->len) {
+		err = -EINVAL;
+		goto err2;
+	}
+
+	return 0;
+err2:
+	nft_data_release(&priv->xor, d2.type);
+err1:
+	nft_data_release(&priv->mask, d1.type);
+	return err;
+}
+
 static int nft_bitwise_init(const struct nft_ctx *ctx,
 			    const struct nft_expr *expr,
 			    const struct nlattr * const tb[])
 {
 	struct nft_bitwise *priv = nft_expr_priv(expr);
-	struct nft_data_desc d1, d2;
 	u32 len;
 	int err;
 
 	if (!tb[NFTA_BITWISE_SREG] ||
 	    !tb[NFTA_BITWISE_DREG] ||
-	    !tb[NFTA_BITWISE_LEN]  ||
-	    !tb[NFTA_BITWISE_MASK] ||
-	    !tb[NFTA_BITWISE_XOR])
+	    !tb[NFTA_BITWISE_LEN])
 		return -EINVAL;
 
 	err = nft_parse_u32_check(tb[NFTA_BITWISE_LEN], U8_MAX, &len);
@@ -90,29 +123,12 @@ static int nft_bitwise_init(const struct nft_ctx *ctx,
 		priv->op = NFT_BITWISE_BOOL;
 	}
 
-	err = nft_data_init(NULL, &priv->mask, sizeof(priv->mask), &d1,
-			    tb[NFTA_BITWISE_MASK]);
-	if (err < 0)
-		return err;
-	if (d1.type != NFT_DATA_VALUE || d1.len != priv->len) {
-		err = -EINVAL;
-		goto err1;
-	}
-
-	err = nft_data_init(NULL, &priv->xor, sizeof(priv->xor), &d2,
-			    tb[NFTA_BITWISE_XOR]);
-	if (err < 0)
-		goto err1;
-	if (d2.type != NFT_DATA_VALUE || d2.len != priv->len) {
-		err = -EINVAL;
-		goto err2;
+	switch(priv->op) {
+	case NFT_BITWISE_BOOL:
+		err = nft_bitwise_init_bool(priv, tb);
+		break;
 	}
 
-	return 0;
-err2:
-	nft_data_release(&priv->xor, d2.type);
-err1:
-	nft_data_release(&priv->mask, d1.type);
 	return err;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 71d6ded3ac496f9a6563752917921b7db0f38a34 Mon Sep 17 00:00:00 2001
From: Jeremy Sowden <jeremy@azazel.net>
Date: Wed, 15 Jan 2020 20:05:53 +0000
Subject: netfilter: bitwise: add helper for evaluating boolean operations.

Split the code specific to evaluating bitwise boolean operations out
into a separate function.  Similar functions will be added later for
shift operations.

Signed-off-by: Jeremy Sowden <jeremy@azazel.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nft_bitwise.c | 17 ++++++++++++++---
 1 file changed, 14 insertions(+), 3 deletions(-)

diff --git a/net/netfilter/nft_bitwise.c b/net/netfilter/nft_bitwise.c
index d0cc5f753e52..88a91c63f213 100644
--- a/net/netfilter/nft_bitwise.c
+++ b/net/netfilter/nft_bitwise.c
@@ -24,16 +24,27 @@ struct nft_bitwise {
 	struct nft_data		xor;
 };
 
+static void nft_bitwise_eval_bool(u32 *dst, const u32 *src,
+				  const struct nft_bitwise *priv)
+{
+	unsigned int i;
+
+	for (i = 0; i < DIV_ROUND_UP(priv->len, 4); i++)
+		dst[i] = (src[i] & priv->mask.data[i]) ^ priv->xor.data[i];
+}
+
 void nft_bitwise_eval(const struct nft_expr *expr,
 		      struct nft_regs *regs, const struct nft_pktinfo *pkt)
 {
 	const struct nft_bitwise *priv = nft_expr_priv(expr);
 	const u32 *src = &regs->data[priv->sreg];
 	u32 *dst = &regs->data[priv->dreg];
-	unsigned int i;
 
-	for (i = 0; i < DIV_ROUND_UP(priv->len, 4); i++)
-		dst[i] = (src[i] & priv->mask.data[i]) ^ priv->xor.data[i];
+	switch (priv->op) {
+	case NFT_BITWISE_BOOL:
+		nft_bitwise_eval_bool(dst, src, priv);
+		break;
+	}
 }
 
 static const struct nla_policy nft_bitwise_policy[NFTA_BITWISE_MAX + 1] = {
-- 
cgit v1.2.3-59-g8ed1b


From 4d57ca2be146bdb272b57892ec59cd273f9ef3fc Mon Sep 17 00:00:00 2001
From: Jeremy Sowden <jeremy@azazel.net>
Date: Wed, 15 Jan 2020 20:05:54 +0000
Subject: netfilter: bitwise: add helper for dumping boolean operations.

Split the code specific to dumping bitwise boolean operations out into a
separate function.  A similar function will be added later for shift
operations.

Signed-off-by: Jeremy Sowden <jeremy@azazel.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nft_bitwise.c | 29 +++++++++++++++++++++--------
 1 file changed, 21 insertions(+), 8 deletions(-)

diff --git a/net/netfilter/nft_bitwise.c b/net/netfilter/nft_bitwise.c
index 88a91c63f213..41265134cf0b 100644
--- a/net/netfilter/nft_bitwise.c
+++ b/net/netfilter/nft_bitwise.c
@@ -143,9 +143,24 @@ static int nft_bitwise_init(const struct nft_ctx *ctx,
 	return err;
 }
 
+static int nft_bitwise_dump_bool(struct sk_buff *skb,
+				 const struct nft_bitwise *priv)
+{
+	if (nft_data_dump(skb, NFTA_BITWISE_MASK, &priv->mask,
+			  NFT_DATA_VALUE, priv->len) < 0)
+		return -1;
+
+	if (nft_data_dump(skb, NFTA_BITWISE_XOR, &priv->xor,
+			  NFT_DATA_VALUE, priv->len) < 0)
+		return -1;
+
+	return 0;
+}
+
 static int nft_bitwise_dump(struct sk_buff *skb, const struct nft_expr *expr)
 {
 	const struct nft_bitwise *priv = nft_expr_priv(expr);
+	int err = 0;
 
 	if (nft_dump_register(skb, NFTA_BITWISE_SREG, priv->sreg))
 		return -1;
@@ -156,15 +171,13 @@ static int nft_bitwise_dump(struct sk_buff *skb, const struct nft_expr *expr)
 	if (nla_put_be32(skb, NFTA_BITWISE_OP, htonl(priv->op)))
 		return -1;
 
-	if (nft_data_dump(skb, NFTA_BITWISE_MASK, &priv->mask,
-			  NFT_DATA_VALUE, priv->len) < 0)
-		return -1;
-
-	if (nft_data_dump(skb, NFTA_BITWISE_XOR, &priv->xor,
-			  NFT_DATA_VALUE, priv->len) < 0)
-		return -1;
+	switch (priv->op) {
+	case NFT_BITWISE_BOOL:
+		err = nft_bitwise_dump_bool(skb, priv);
+		break;
+	}
 
-	return 0;
+	return err;
 }
 
 static struct nft_data zero;
-- 
cgit v1.2.3-59-g8ed1b


From ed991d43634b11e759a800850ba9656b5b467a56 Mon Sep 17 00:00:00 2001
From: Jeremy Sowden <jeremy@azazel.net>
Date: Wed, 15 Jan 2020 20:05:55 +0000
Subject: netfilter: bitwise: only offload boolean operations.

Only boolean operations supports offloading, so check the type of the
operation and return an error for other types.

Signed-off-by: Jeremy Sowden <jeremy@azazel.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nft_bitwise.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/net/netfilter/nft_bitwise.c b/net/netfilter/nft_bitwise.c
index 41265134cf0b..b4619d9989ea 100644
--- a/net/netfilter/nft_bitwise.c
+++ b/net/netfilter/nft_bitwise.c
@@ -189,6 +189,9 @@ static int nft_bitwise_offload(struct nft_offload_ctx *ctx,
 	const struct nft_bitwise *priv = nft_expr_priv(expr);
 	struct nft_offload_reg *reg = &ctx->regs[priv->dreg];
 
+	if (priv->op != NFT_BITWISE_BOOL)
+		return -EOPNOTSUPP;
+
 	if (memcmp(&priv->xor, &zero, sizeof(priv->xor)) ||
 	    priv->sreg != priv->dreg || priv->len != reg->len)
 		return -EOPNOTSUPP;
-- 
cgit v1.2.3-59-g8ed1b


From 779f725e142cd987a69296c0eb7d416ee3ba57dd Mon Sep 17 00:00:00 2001
From: Jeremy Sowden <jeremy@azazel.net>
Date: Wed, 15 Jan 2020 20:05:56 +0000
Subject: netfilter: bitwise: add NFTA_BITWISE_DATA attribute.

Add a new bitwise netlink attribute that will be used by shift
operations to store the size of the shift.  It is not used by boolean
operations.

Signed-off-by: Jeremy Sowden <jeremy@azazel.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/uapi/linux/netfilter/nf_tables.h | 3 +++
 net/netfilter/nft_bitwise.c              | 5 +++++
 2 files changed, 8 insertions(+)

diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h
index 0cddf357281f..8bef0620bc4f 100644
--- a/include/uapi/linux/netfilter/nf_tables.h
+++ b/include/uapi/linux/netfilter/nf_tables.h
@@ -503,6 +503,8 @@ enum nft_bitwise_ops {
  * @NFTA_BITWISE_MASK: mask value (NLA_NESTED: nft_data_attributes)
  * @NFTA_BITWISE_XOR: xor value (NLA_NESTED: nft_data_attributes)
  * @NFTA_BITWISE_OP: type of operation (NLA_U32: nft_bitwise_ops)
+ * @NFTA_BITWISE_DATA: argument for non-boolean operations
+ *                     (NLA_NESTED: nft_data_attributes)
  *
  * The bitwise expression performs the following operation:
  *
@@ -524,6 +526,7 @@ enum nft_bitwise_attributes {
 	NFTA_BITWISE_MASK,
 	NFTA_BITWISE_XOR,
 	NFTA_BITWISE_OP,
+	NFTA_BITWISE_DATA,
 	__NFTA_BITWISE_MAX
 };
 #define NFTA_BITWISE_MAX	(__NFTA_BITWISE_MAX - 1)
diff --git a/net/netfilter/nft_bitwise.c b/net/netfilter/nft_bitwise.c
index b4619d9989ea..744008a527fb 100644
--- a/net/netfilter/nft_bitwise.c
+++ b/net/netfilter/nft_bitwise.c
@@ -22,6 +22,7 @@ struct nft_bitwise {
 	u8			len;
 	struct nft_data		mask;
 	struct nft_data		xor;
+	struct nft_data		data;
 };
 
 static void nft_bitwise_eval_bool(u32 *dst, const u32 *src,
@@ -54,6 +55,7 @@ static const struct nla_policy nft_bitwise_policy[NFTA_BITWISE_MAX + 1] = {
 	[NFTA_BITWISE_MASK]	= { .type = NLA_NESTED },
 	[NFTA_BITWISE_XOR]	= { .type = NLA_NESTED },
 	[NFTA_BITWISE_OP]	= { .type = NLA_U32 },
+	[NFTA_BITWISE_DATA]	= { .type = NLA_NESTED },
 };
 
 static int nft_bitwise_init_bool(struct nft_bitwise *priv,
@@ -62,6 +64,9 @@ static int nft_bitwise_init_bool(struct nft_bitwise *priv,
 	struct nft_data_desc d1, d2;
 	int err;
 
+	if (tb[NFTA_BITWISE_DATA])
+		return -EINVAL;
+
 	if (!tb[NFTA_BITWISE_MASK] ||
 	    !tb[NFTA_BITWISE_XOR])
 		return -EINVAL;
-- 
cgit v1.2.3-59-g8ed1b


From 567d746b55bc66d3800c9ae91d50f0c5deb2fd93 Mon Sep 17 00:00:00 2001
From: Jeremy Sowden <jeremy@azazel.net>
Date: Wed, 15 Jan 2020 20:05:57 +0000
Subject: netfilter: bitwise: add support for shifts.

Hitherto nft_bitwise has only supported boolean operations: NOT, AND, OR
and XOR.  Extend it to do shifts as well.

Signed-off-by: Jeremy Sowden <jeremy@azazel.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/uapi/linux/netfilter/nf_tables.h |  9 +++-
 net/netfilter/nft_bitwise.c              | 77 ++++++++++++++++++++++++++++++++
 2 files changed, 84 insertions(+), 2 deletions(-)

diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h
index 8bef0620bc4f..261864736b26 100644
--- a/include/uapi/linux/netfilter/nf_tables.h
+++ b/include/uapi/linux/netfilter/nf_tables.h
@@ -489,9 +489,13 @@ enum nft_immediate_attributes {
  *
  * @NFT_BITWISE_BOOL: mask-and-xor operation used to implement NOT, AND, OR and
  *                    XOR boolean operations
+ * @NFT_BITWISE_LSHIFT: left-shift operation
+ * @NFT_BITWISE_RSHIFT: right-shift operation
  */
 enum nft_bitwise_ops {
 	NFT_BITWISE_BOOL,
+	NFT_BITWISE_LSHIFT,
+	NFT_BITWISE_RSHIFT,
 };
 
 /**
@@ -506,11 +510,12 @@ enum nft_bitwise_ops {
  * @NFTA_BITWISE_DATA: argument for non-boolean operations
  *                     (NLA_NESTED: nft_data_attributes)
  *
- * The bitwise expression performs the following operation:
+ * The bitwise expression supports boolean and shift operations.  It implements
+ * the boolean operations by performing the following operation:
  *
  * dreg = (sreg & mask) ^ xor
  *
- * which allow to express all bitwise operations:
+ * with these mask and xor values:
  *
  * 		mask	xor
  * NOT:		1	1
diff --git a/net/netfilter/nft_bitwise.c b/net/netfilter/nft_bitwise.c
index 744008a527fb..0ed2281f03be 100644
--- a/net/netfilter/nft_bitwise.c
+++ b/net/netfilter/nft_bitwise.c
@@ -34,6 +34,32 @@ static void nft_bitwise_eval_bool(u32 *dst, const u32 *src,
 		dst[i] = (src[i] & priv->mask.data[i]) ^ priv->xor.data[i];
 }
 
+static void nft_bitwise_eval_lshift(u32 *dst, const u32 *src,
+				    const struct nft_bitwise *priv)
+{
+	u32 shift = priv->data.data[0];
+	unsigned int i;
+	u32 carry = 0;
+
+	for (i = DIV_ROUND_UP(priv->len, sizeof(u32)); i > 0; i--) {
+		dst[i - 1] = (src[i - 1] << shift) | carry;
+		carry = src[i - 1] >> (BITS_PER_TYPE(u32) - shift);
+	}
+}
+
+static void nft_bitwise_eval_rshift(u32 *dst, const u32 *src,
+				    const struct nft_bitwise *priv)
+{
+	u32 shift = priv->data.data[0];
+	unsigned int i;
+	u32 carry = 0;
+
+	for (i = 0; i < DIV_ROUND_UP(priv->len, sizeof(u32)); i++) {
+		dst[i] = carry | (src[i] >> shift);
+		carry = src[i] << (BITS_PER_TYPE(u32) - shift);
+	}
+}
+
 void nft_bitwise_eval(const struct nft_expr *expr,
 		      struct nft_regs *regs, const struct nft_pktinfo *pkt)
 {
@@ -45,6 +71,12 @@ void nft_bitwise_eval(const struct nft_expr *expr,
 	case NFT_BITWISE_BOOL:
 		nft_bitwise_eval_bool(dst, src, priv);
 		break;
+	case NFT_BITWISE_LSHIFT:
+		nft_bitwise_eval_lshift(dst, src, priv);
+		break;
+	case NFT_BITWISE_RSHIFT:
+		nft_bitwise_eval_rshift(dst, src, priv);
+		break;
 	}
 }
 
@@ -97,6 +129,32 @@ err1:
 	return err;
 }
 
+static int nft_bitwise_init_shift(struct nft_bitwise *priv,
+				  const struct nlattr *const tb[])
+{
+	struct nft_data_desc d;
+	int err;
+
+	if (tb[NFTA_BITWISE_MASK] ||
+	    tb[NFTA_BITWISE_XOR])
+		return -EINVAL;
+
+	if (!tb[NFTA_BITWISE_DATA])
+		return -EINVAL;
+
+	err = nft_data_init(NULL, &priv->data, sizeof(priv->data), &d,
+			    tb[NFTA_BITWISE_DATA]);
+	if (err < 0)
+		return err;
+	if (d.type != NFT_DATA_VALUE || d.len != sizeof(u32) ||
+	    priv->data.data[0] >= BITS_PER_TYPE(u32)) {
+		nft_data_release(&priv->data, d.type);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
 static int nft_bitwise_init(const struct nft_ctx *ctx,
 			    const struct nft_expr *expr,
 			    const struct nlattr * const tb[])
@@ -131,6 +189,8 @@ static int nft_bitwise_init(const struct nft_ctx *ctx,
 		priv->op = ntohl(nla_get_be32(tb[NFTA_BITWISE_OP]));
 		switch (priv->op) {
 		case NFT_BITWISE_BOOL:
+		case NFT_BITWISE_LSHIFT:
+		case NFT_BITWISE_RSHIFT:
 			break;
 		default:
 			return -EOPNOTSUPP;
@@ -143,6 +203,10 @@ static int nft_bitwise_init(const struct nft_ctx *ctx,
 	case NFT_BITWISE_BOOL:
 		err = nft_bitwise_init_bool(priv, tb);
 		break;
+	case NFT_BITWISE_LSHIFT:
+	case NFT_BITWISE_RSHIFT:
+		err = nft_bitwise_init_shift(priv, tb);
+		break;
 	}
 
 	return err;
@@ -162,6 +226,15 @@ static int nft_bitwise_dump_bool(struct sk_buff *skb,
 	return 0;
 }
 
+static int nft_bitwise_dump_shift(struct sk_buff *skb,
+				  const struct nft_bitwise *priv)
+{
+	if (nft_data_dump(skb, NFTA_BITWISE_DATA, &priv->data,
+			  NFT_DATA_VALUE, sizeof(u32)) < 0)
+		return -1;
+	return 0;
+}
+
 static int nft_bitwise_dump(struct sk_buff *skb, const struct nft_expr *expr)
 {
 	const struct nft_bitwise *priv = nft_expr_priv(expr);
@@ -180,6 +253,10 @@ static int nft_bitwise_dump(struct sk_buff *skb, const struct nft_expr *expr)
 	case NFT_BITWISE_BOOL:
 		err = nft_bitwise_dump_bool(skb, priv);
 		break;
+	case NFT_BITWISE_LSHIFT:
+	case NFT_BITWISE_RSHIFT:
+		err = nft_bitwise_dump_shift(skb, priv);
+		break;
 	}
 
 	return err;
-- 
cgit v1.2.3-59-g8ed1b


From 35697c12d7ffd31a56d3c9604066a166b75d0169 Mon Sep 17 00:00:00 2001
From: Yonghong Song <yhs@fb.com>
Date: Thu, 16 Jan 2020 09:40:04 -0800
Subject: selftests/bpf: Fix test_progs send_signal flakiness with nmi mode

Alexei observed that test_progs send_signal may fail if run
with command line "./test_progs" and the tests will pass
if just run "./test_progs -n 40".

I observed similar issue with nmi subtest failure
and added a delay 100 us in Commit ab8b7f0cb358
("tools/bpf: Add self tests for bpf_send_signal_thread()")
and the problem is gone for me. But the issue still exists
in Alexei's testing environment.

The current code uses sample_freq = 50 (50 events/second), which
may not be enough. But if the sample_freq value is larger than
sysctl kernel/perf_event_max_sample_rate, the perf_event_open
syscall will fail.

This patch changed nmi perf testing to use sample_period = 1,
which means trying to sampling every event. This seems fixing
the issue.

Fixes: ab8b7f0cb358 ("tools/bpf: Add self tests for bpf_send_signal_thread()")
Signed-off-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/20200116174004.1522812-1-yhs@fb.com
---
 tools/testing/selftests/bpf/prog_tests/send_signal.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/send_signal.c b/tools/testing/selftests/bpf/prog_tests/send_signal.c
index d4cedd86c424..504abb7bfb95 100644
--- a/tools/testing/selftests/bpf/prog_tests/send_signal.c
+++ b/tools/testing/selftests/bpf/prog_tests/send_signal.c
@@ -76,9 +76,6 @@ static void test_send_signal_common(struct perf_event_attr *attr,
 	if (CHECK(!skel, "skel_open_and_load", "skeleton open_and_load failed\n"))
 		goto skel_open_load_failure;
 
-	/* add a delay for child thread to ramp up */
-	usleep(100);
-
 	if (!attr) {
 		err = test_send_signal_kern__attach(skel);
 		if (CHECK(err, "skel_attach", "skeleton attach failed\n")) {
@@ -155,8 +152,7 @@ static void test_send_signal_perf(bool signal_thread)
 static void test_send_signal_nmi(bool signal_thread)
 {
 	struct perf_event_attr attr = {
-		.sample_freq = 50,
-		.freq = 1,
+		.sample_period = 1,
 		.type = PERF_TYPE_HARDWARE,
 		.config = PERF_COUNT_HW_CPU_CYCLES,
 	};
-- 
cgit v1.2.3-59-g8ed1b


From f397464eb7c25bda903ec8b9cf5701e72a1f7b16 Mon Sep 17 00:00:00 2001
From: Eran Ben Elisha <eranbe@mellanox.com>
Date: Mon, 7 Oct 2019 10:29:46 +0300
Subject: net/mlx5: Add structures layout for new MCAM access reg groups

MCAM has 3 access_reg_groups (0-2). Defines data structures in order to
read and parse access_reg_groups #1 and #2.

Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 include/linux/mlx5/mlx5_ifc.h | 24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)

diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index c6abaf4f1c55..43cdf9211747 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -8832,6 +8832,28 @@ struct mlx5_ifc_mcam_access_reg_bits {
 	u8         regs_31_to_0[0x20];
 };
 
+struct mlx5_ifc_mcam_access_reg_bits1 {
+	u8         regs_127_to_96[0x20];
+
+	u8         regs_95_to_64[0x20];
+
+	u8         regs_63_to_32[0x20];
+
+	u8         regs_31_to_0[0x20];
+};
+
+struct mlx5_ifc_mcam_access_reg_bits2 {
+	u8         regs_127_to_99[0x1d];
+	u8         mirc[0x1];
+	u8         regs_97_to_96[0x2];
+
+	u8         regs_95_to_64[0x20];
+
+	u8         regs_63_to_32[0x20];
+
+	u8         regs_31_to_0[0x20];
+};
+
 struct mlx5_ifc_mcam_reg_bits {
 	u8         reserved_at_0[0x8];
 	u8         feature_group[0x8];
@@ -8842,6 +8864,8 @@ struct mlx5_ifc_mcam_reg_bits {
 
 	union {
 		struct mlx5_ifc_mcam_access_reg_bits access_regs;
+		struct mlx5_ifc_mcam_access_reg_bits1 access_regs1;
+		struct mlx5_ifc_mcam_access_reg_bits2 access_regs2;
 		u8         reserved_at_0[0x80];
 	} mng_access_reg_cap_mask;
 
-- 
cgit v1.2.3-59-g8ed1b


From 932ef155117cc5caf1108bd27664dab974ba6e89 Mon Sep 17 00:00:00 2001
From: Eran Ben Elisha <eranbe@mellanox.com>
Date: Mon, 7 Oct 2019 10:31:42 +0300
Subject: net/mlx5: Read MCAM register groups 1 and 2

On load, Driver caches MCAM (Management Capabilities Mask Register)
registers. in addition to the only MCAM register group (0) the driver
already reads, here we add support for reading groups 1 and 2.

Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/fw.c | 15 +++++++++------
 include/linux/mlx5/device.h                  | 14 +++++++++++++-
 include/linux/mlx5/driver.h                  |  2 +-
 3 files changed, 23 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw.c b/drivers/net/ethernet/mellanox/mlx5/core/fw.c
index c375edfe528c..d89ff1d09119 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fw.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fw.c
@@ -131,11 +131,11 @@ static int mlx5_get_pcam_reg(struct mlx5_core_dev *dev)
 				   MLX5_PCAM_REGS_5000_TO_507F);
 }
 
-static int mlx5_get_mcam_reg(struct mlx5_core_dev *dev)
+static int mlx5_get_mcam_access_reg_group(struct mlx5_core_dev *dev,
+					  enum mlx5_mcam_reg_groups group)
 {
-	return mlx5_query_mcam_reg(dev, dev->caps.mcam,
-				   MLX5_MCAM_FEATURE_ENHANCED_FEATURES,
-				   MLX5_MCAM_REGS_FIRST_128);
+	return mlx5_query_mcam_reg(dev, dev->caps.mcam[group],
+				   MLX5_MCAM_FEATURE_ENHANCED_FEATURES, group);
 }
 
 static int mlx5_get_qcam_reg(struct mlx5_core_dev *dev)
@@ -221,8 +221,11 @@ int mlx5_query_hca_caps(struct mlx5_core_dev *dev)
 	if (MLX5_CAP_GEN(dev, pcam_reg))
 		mlx5_get_pcam_reg(dev);
 
-	if (MLX5_CAP_GEN(dev, mcam_reg))
-		mlx5_get_mcam_reg(dev);
+	if (MLX5_CAP_GEN(dev, mcam_reg)) {
+		mlx5_get_mcam_access_reg_group(dev, MLX5_MCAM_REGS_FIRST_128);
+		mlx5_get_mcam_access_reg_group(dev, MLX5_MCAM_REGS_0x9080_0x90FF);
+		mlx5_get_mcam_access_reg_group(dev, MLX5_MCAM_REGS_0x9100_0x917F);
+	}
 
 	if (MLX5_CAP_GEN(dev, qcam_reg))
 		mlx5_get_qcam_reg(dev);
diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h
index 1a1c53f0262d..0e62c3db45e5 100644
--- a/include/linux/mlx5/device.h
+++ b/include/linux/mlx5/device.h
@@ -1121,6 +1121,9 @@ enum mlx5_pcam_feature_groups {
 
 enum mlx5_mcam_reg_groups {
 	MLX5_MCAM_REGS_FIRST_128                    = 0x0,
+	MLX5_MCAM_REGS_0x9080_0x90FF                = 0x1,
+	MLX5_MCAM_REGS_0x9100_0x917F                = 0x2,
+	MLX5_MCAM_REGS_NUM                          = 0x3,
 };
 
 enum mlx5_mcam_feature_groups {
@@ -1269,7 +1272,16 @@ enum mlx5_qcam_feature_groups {
 	MLX5_GET(pcam_reg, (mdev)->caps.pcam, port_access_reg_cap_mask.regs_5000_to_507f.reg)
 
 #define MLX5_CAP_MCAM_REG(mdev, reg) \
-	MLX5_GET(mcam_reg, (mdev)->caps.mcam, mng_access_reg_cap_mask.access_regs.reg)
+	MLX5_GET(mcam_reg, (mdev)->caps.mcam[MLX5_MCAM_REGS_FIRST_128], \
+		 mng_access_reg_cap_mask.access_regs.reg)
+
+#define MLX5_CAP_MCAM_REG1(mdev, reg) \
+	MLX5_GET(mcam_reg, (mdev)->caps.mcam[MLX5_MCAM_REGS_0x9080_0x90FF], \
+		 mng_access_reg_cap_mask.access_regs1.reg)
+
+#define MLX5_CAP_MCAM_REG2(mdev, reg) \
+	MLX5_GET(mcam_reg, (mdev)->caps.mcam[MLX5_MCAM_REGS_0x9100_0x917F], \
+		 mng_access_reg_cap_mask.access_regs2.reg)
 
 #define MLX5_CAP_MCAM_FEATURE(mdev, fld) \
 	MLX5_GET(mcam_reg, (mdev)->caps.mcam, mng_feature_cap_mask.enhanced_features.fld)
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index 27200dea0297..54431256af42 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -684,7 +684,7 @@ struct mlx5_core_dev {
 		u32 hca_cur[MLX5_CAP_NUM][MLX5_UN_SZ_DW(hca_cap_union)];
 		u32 hca_max[MLX5_CAP_NUM][MLX5_UN_SZ_DW(hca_cap_union)];
 		u32 pcam[MLX5_ST_SZ_DW(pcam_reg)];
-		u32 mcam[MLX5_ST_SZ_DW(mcam_reg)];
+		u32 mcam[MLX5_MCAM_REGS_NUM][MLX5_ST_SZ_DW(mcam_reg)];
 		u32 fpga[MLX5_ST_SZ_DW(fpga_cap)];
 		u32 qcam[MLX5_ST_SZ_DW(qcam_reg)];
 		u8  embedded_cpu;
-- 
cgit v1.2.3-59-g8ed1b


From bab58ba10ecfa39c46d280d2acbca6054e1e863d Mon Sep 17 00:00:00 2001
From: Eran Ben Elisha <eranbe@mellanox.com>
Date: Mon, 7 Oct 2019 10:30:32 +0300
Subject: net/mlx5: Add structures and defines for MIRC register

Add needed structures, layouts and defines for MIRC (Management Image
Re-activation Control) register. This structure will be used for the FSM
reactivation flow in the downstream patches.

Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 include/linux/mlx5/driver.h   | 1 +
 include/linux/mlx5/mlx5_ifc.h | 8 ++++++++
 2 files changed, 9 insertions(+)

diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index 54431256af42..7848b9858587 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -145,6 +145,7 @@ enum {
 	MLX5_REG_MCC		 = 0x9062,
 	MLX5_REG_MCDA		 = 0x9063,
 	MLX5_REG_MCAM		 = 0x907f,
+	MLX5_REG_MIRC		 = 0x9162,
 };
 
 enum mlx5_qpts_trust_state {
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 43cdf9211747..a133583c3e4f 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -9471,6 +9471,13 @@ struct mlx5_ifc_mcda_reg_bits {
 	u8         data[0][0x20];
 };
 
+struct mlx5_ifc_mirc_reg_bits {
+	u8         reserved_at_0[0x18];
+	u8         status_code[0x8];
+
+	u8         reserved_at_20[0x20];
+};
+
 union mlx5_ifc_ports_control_registers_document_bits {
 	struct mlx5_ifc_bufferx_reg_bits bufferx_reg;
 	struct mlx5_ifc_eth_2819_cntrs_grp_data_layout_bits eth_2819_cntrs_grp_data_layout;
@@ -9526,6 +9533,7 @@ union mlx5_ifc_ports_control_registers_document_bits {
 	struct mlx5_ifc_mcqi_reg_bits mcqi_reg;
 	struct mlx5_ifc_mcc_reg_bits mcc_reg;
 	struct mlx5_ifc_mcda_reg_bits mcda_reg;
+	struct mlx5_ifc_mirc_reg_bits mirc_reg;
 	u8         reserved_at_0[0x60e0];
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From 609b82727f719b41b50440c4028d48d0b2e04913 Mon Sep 17 00:00:00 2001
From: Aya Levin <ayal@mellanox.com>
Date: Mon, 4 Nov 2019 14:51:55 +0200
Subject: net/mlx5: Expose resource dump register mapping

Add new register enumeration for resource dump. Add layout mapping for
resource dump: access command and response.

Signed-off-by: Aya Levin <ayal@mellanox.com>
Reviewed-by: Moshe Shemesh <moshe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 include/linux/mlx5/driver.h   |   1 +
 include/linux/mlx5/mlx5_ifc.h | 130 +++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 130 insertions(+), 1 deletion(-)

diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index 7848b9858587..c821fa4d7475 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -146,6 +146,7 @@ enum {
 	MLX5_REG_MCDA		 = 0x9063,
 	MLX5_REG_MCAM		 = 0x907f,
 	MLX5_REG_MIRC		 = 0x9162,
+	MLX5_REG_RESOURCE_DUMP   = 0xC000,
 };
 
 enum mlx5_qpts_trust_state {
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index a133583c3e4f..6fe0431e11ec 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -823,7 +823,9 @@ struct mlx5_ifc_qos_cap_bits {
 struct mlx5_ifc_debug_cap_bits {
 	u8         core_dump_general[0x1];
 	u8         core_dump_qp[0x1];
-	u8         reserved_at_2[0x1e];
+	u8         reserved_at_2[0x7];
+	u8         resource_dump[0x1];
+	u8         reserved_at_a[0x16];
 
 	u8         reserved_at_20[0x2];
 	u8         stall_detect[0x1];
@@ -1767,6 +1769,132 @@ struct mlx5_ifc_resize_field_select_bits {
 	u8         resize_field_select[0x20];
 };
 
+struct mlx5_ifc_resource_dump_bits {
+	u8         more_dump[0x1];
+	u8         inline_dump[0x1];
+	u8         reserved_at_2[0xa];
+	u8         seq_num[0x4];
+	u8         segment_type[0x10];
+
+	u8         reserved_at_20[0x10];
+	u8         vhca_id[0x10];
+
+	u8         index1[0x20];
+
+	u8         index2[0x20];
+
+	u8         num_of_obj1[0x10];
+	u8         num_of_obj2[0x10];
+
+	u8         reserved_at_a0[0x20];
+
+	u8         device_opaque[0x40];
+
+	u8         mkey[0x20];
+
+	u8         size[0x20];
+
+	u8         address[0x40];
+
+	u8         inline_data[52][0x20];
+};
+
+struct mlx5_ifc_resource_dump_menu_record_bits {
+	u8         reserved_at_0[0x4];
+	u8         num_of_obj2_supports_active[0x1];
+	u8         num_of_obj2_supports_all[0x1];
+	u8         must_have_num_of_obj2[0x1];
+	u8         support_num_of_obj2[0x1];
+	u8         num_of_obj1_supports_active[0x1];
+	u8         num_of_obj1_supports_all[0x1];
+	u8         must_have_num_of_obj1[0x1];
+	u8         support_num_of_obj1[0x1];
+	u8         must_have_index2[0x1];
+	u8         support_index2[0x1];
+	u8         must_have_index1[0x1];
+	u8         support_index1[0x1];
+	u8         segment_type[0x10];
+
+	u8         segment_name[4][0x20];
+
+	u8         index1_name[4][0x20];
+
+	u8         index2_name[4][0x20];
+};
+
+struct mlx5_ifc_resource_dump_segment_header_bits {
+	u8         length_dw[0x10];
+	u8         segment_type[0x10];
+};
+
+struct mlx5_ifc_resource_dump_command_segment_bits {
+	struct mlx5_ifc_resource_dump_segment_header_bits segment_header;
+
+	u8         segment_called[0x10];
+	u8         vhca_id[0x10];
+
+	u8         index1[0x20];
+
+	u8         index2[0x20];
+
+	u8         num_of_obj1[0x10];
+	u8         num_of_obj2[0x10];
+};
+
+struct mlx5_ifc_resource_dump_error_segment_bits {
+	struct mlx5_ifc_resource_dump_segment_header_bits segment_header;
+
+	u8         reserved_at_20[0x10];
+	u8         syndrome_id[0x10];
+
+	u8         reserved_at_40[0x40];
+
+	u8         error[8][0x20];
+};
+
+struct mlx5_ifc_resource_dump_info_segment_bits {
+	struct mlx5_ifc_resource_dump_segment_header_bits segment_header;
+
+	u8         reserved_at_20[0x18];
+	u8         dump_version[0x8];
+
+	u8         hw_version[0x20];
+
+	u8         fw_version[0x20];
+};
+
+struct mlx5_ifc_resource_dump_menu_segment_bits {
+	struct mlx5_ifc_resource_dump_segment_header_bits segment_header;
+
+	u8         reserved_at_20[0x10];
+	u8         num_of_records[0x10];
+
+	struct mlx5_ifc_resource_dump_menu_record_bits record[0];
+};
+
+struct mlx5_ifc_resource_dump_resource_segment_bits {
+	struct mlx5_ifc_resource_dump_segment_header_bits segment_header;
+
+	u8         reserved_at_20[0x20];
+
+	u8         index1[0x20];
+
+	u8         index2[0x20];
+
+	u8         payload[0][0x20];
+};
+
+struct mlx5_ifc_resource_dump_terminate_segment_bits {
+	struct mlx5_ifc_resource_dump_segment_header_bits segment_header;
+};
+
+struct mlx5_ifc_menu_resource_dump_response_bits {
+	struct mlx5_ifc_resource_dump_info_segment_bits info;
+	struct mlx5_ifc_resource_dump_command_segment_bits cmd;
+	struct mlx5_ifc_resource_dump_menu_segment_bits menu;
+	struct mlx5_ifc_resource_dump_terminate_segment_bits terminate;
+};
+
 enum {
 	MLX5_MODIFY_FIELD_SELECT_MODIFY_FIELD_SELECT_CQ_PERIOD     = 0x1,
 	MLX5_MODIFY_FIELD_SELECT_MODIFY_FIELD_SELECT_CQ_MAX_COUNT  = 0x2,
-- 
cgit v1.2.3-59-g8ed1b


From 31d8bde1c8812c9b44065dcd98e554488c6a98d2 Mon Sep 17 00:00:00 2001
From: Hamdan Igbaria <hamdani@mellanox.com>
Date: Thu, 9 Jan 2020 13:26:53 +0200
Subject: net/mlx5: Add copy header action struct layout

Add definition for copy header action, copy action is used
to copy header fields from source to destination.

Signed-off-by: Hamdan Igbaria <hamdani@mellanox.com>
Signed-off-by: Alex Vesker <valex@mellanox.com>
Reviewed-by: Alex Vesker <valex@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 include/linux/mlx5/mlx5_ifc.h | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 6fe0431e11ec..23613a6ea51c 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -5609,6 +5609,21 @@ struct mlx5_ifc_add_action_in_bits {
 	u8         data[0x20];
 };
 
+struct mlx5_ifc_copy_action_in_bits {
+	u8         action_type[0x4];
+	u8         src_field[0xc];
+	u8         reserved_at_10[0x3];
+	u8         src_offset[0x5];
+	u8         reserved_at_18[0x3];
+	u8         length[0x5];
+
+	u8         reserved_at_20[0x4];
+	u8         dst_field[0xc];
+	u8         reserved_at_30[0x3];
+	u8         dst_offset[0x5];
+	u8         reserved_at_38[0x8];
+};
+
 union mlx5_ifc_set_action_in_add_action_in_auto_bits {
 	struct mlx5_ifc_set_action_in_bits set_action_in;
 	struct mlx5_ifc_add_action_in_bits add_action_in;
@@ -5618,6 +5633,7 @@ union mlx5_ifc_set_action_in_add_action_in_auto_bits {
 enum {
 	MLX5_ACTION_TYPE_SET   = 0x1,
 	MLX5_ACTION_TYPE_ADD   = 0x2,
+	MLX5_ACTION_TYPE_COPY  = 0x3,
 };
 
 enum {
-- 
cgit v1.2.3-59-g8ed1b


From 822e114b50641d3b57d2eb30939e60d8b4758288 Mon Sep 17 00:00:00 2001
From: Paul Blakey <paulb@mellanox.com>
Date: Mon, 1 Apr 2019 13:31:32 +0300
Subject: net/mlx5: Add mlx5_ifc definitions for connection tracking support

Add the required hardware definitions to mlx5_ifc:
ignore_flow_level, registers, copy_header, and fwd_and_modify cap.

Signed-off-by: Paul Blakey <paulb@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Reviewed-by: Oz Sholomo <ozsh@mellanox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 include/linux/mlx5/mlx5_ifc.h | 24 ++++++++++++++++++++----
 1 file changed, 20 insertions(+), 4 deletions(-)

diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 23613a6ea51c..e9c165ffe3f9 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -375,8 +375,17 @@ struct mlx5_ifc_flow_table_fields_supported_bits {
 	u8	   outer_esp_spi[0x1];
 	u8         reserved_at_58[0x2];
 	u8         bth_dst_qp[0x1];
+	u8         reserved_at_5b[0x5];
 
-	u8         reserved_at_5b[0x25];
+	u8         reserved_at_60[0x18];
+	u8         metadata_reg_c_7[0x1];
+	u8         metadata_reg_c_6[0x1];
+	u8         metadata_reg_c_5[0x1];
+	u8         metadata_reg_c_4[0x1];
+	u8         metadata_reg_c_3[0x1];
+	u8         metadata_reg_c_2[0x1];
+	u8         metadata_reg_c_1[0x1];
+	u8         metadata_reg_c_0[0x1];
 };
 
 struct mlx5_ifc_flow_table_prop_layout_bits {
@@ -401,7 +410,8 @@ struct mlx5_ifc_flow_table_prop_layout_bits {
 	u8	   reformat_l3_tunnel_to_l2[0x1];
 	u8	   reformat_l2_to_l3_tunnel[0x1];
 	u8	   reformat_and_modify_action[0x1];
-	u8         reserved_at_15[0x2];
+	u8	   ignore_flow_level[0x1];
+	u8         reserved_at_16[0x1];
 	u8	   table_miss_action_domain[0x1];
 	u8         termination_table[0x1];
 	u8         reserved_at_19[0x7];
@@ -722,7 +732,9 @@ enum {
 
 struct mlx5_ifc_flow_table_eswitch_cap_bits {
 	u8      fdb_to_vport_reg_c_id[0x8];
-	u8      reserved_at_8[0xf];
+	u8      reserved_at_8[0xd];
+	u8      fdb_modify_header_fwd_to_table[0x1];
+	u8      reserved_at_16[0x1];
 	u8      flow_source[0x1];
 	u8      reserved_at_18[0x2];
 	u8      multi_fdb_encap[0x1];
@@ -4141,7 +4153,8 @@ struct mlx5_ifc_set_fte_in_bits {
 	u8         reserved_at_a0[0x8];
 	u8         table_id[0x18];
 
-	u8         reserved_at_c0[0x18];
+	u8         ignore_flow_level[0x1];
+	u8         reserved_at_c1[0x17];
 	u8         modify_enable_mask[0x8];
 
 	u8         reserved_at_e0[0x20];
@@ -5627,6 +5640,7 @@ struct mlx5_ifc_copy_action_in_bits {
 union mlx5_ifc_set_action_in_add_action_in_auto_bits {
 	struct mlx5_ifc_set_action_in_bits set_action_in;
 	struct mlx5_ifc_add_action_in_bits add_action_in;
+	struct mlx5_ifc_copy_action_in_bits copy_action_in;
 	u8         reserved_at_0[0x40];
 };
 
@@ -5669,6 +5683,8 @@ enum {
 	MLX5_ACTION_IN_FIELD_METADATA_REG_C_3  = 0x54,
 	MLX5_ACTION_IN_FIELD_METADATA_REG_C_4  = 0x55,
 	MLX5_ACTION_IN_FIELD_METADATA_REG_C_5  = 0x56,
+	MLX5_ACTION_IN_FIELD_METADATA_REG_C_6  = 0x57,
+	MLX5_ACTION_IN_FIELD_METADATA_REG_C_7  = 0x58,
 	MLX5_ACTION_IN_FIELD_OUT_TCP_SEQ_NUM   = 0x59,
 	MLX5_ACTION_IN_FIELD_OUT_TCP_ACK_NUM   = 0x5B,
 };
-- 
cgit v1.2.3-59-g8ed1b


From a58837f52d432f32995b1c00e803cc4db18762d3 Mon Sep 17 00:00:00 2001
From: Aya Levin <ayal@mellanox.com>
Date: Mon, 30 Dec 2019 14:22:57 +0200
Subject: net/mlx5e: Expose FEC feilds and related capability bit

Introduce 50G per lane FEC modes capability bit and newly supported
fields in PPLM register which allow this configuration.

Signed-off-by: Aya Levin <ayal@mellanox.com>
Reviewed-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 include/linux/mlx5/mlx5_ifc.h | 16 +++++++++++++++-
 1 file changed, 15 insertions(+), 1 deletion(-)

diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index e9c165ffe3f9..2ab4562b4851 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -8581,6 +8581,18 @@ struct mlx5_ifc_pplm_reg_bits {
 	u8	   fec_override_admin_50g[0x4];
 	u8	   fec_override_admin_25g[0x4];
 	u8	   fec_override_admin_10g_40g[0x4];
+
+	u8         fec_override_cap_400g_8x[0x10];
+	u8         fec_override_cap_200g_4x[0x10];
+
+	u8         fec_override_cap_100g_2x[0x10];
+	u8         fec_override_cap_50g_1x[0x10];
+
+	u8         fec_override_admin_400g_8x[0x10];
+	u8         fec_override_admin_200g_4x[0x10];
+
+	u8         fec_override_admin_100g_2x[0x10];
+	u8         fec_override_admin_50g_1x[0x10];
 };
 
 struct mlx5_ifc_ppcnt_reg_bits {
@@ -8907,7 +8919,9 @@ struct mlx5_ifc_mpegc_reg_bits {
 };
 
 struct mlx5_ifc_pcam_enhanced_features_bits {
-	u8         reserved_at_0[0x6d];
+	u8         reserved_at_0[0x68];
+	u8         fec_50G_per_lane_in_pplm[0x1];
+	u8         reserved_at_69[0x4];
 	u8         rx_icrc_encapsulated_counter[0x1];
 	u8	   reserved_at_6e[0x4];
 	u8         ptys_extended_ethernet[0x1];
-- 
cgit v1.2.3-59-g8ed1b


From 827a8cb2dd2b72848652b2a425bba3262808ff44 Mon Sep 17 00:00:00 2001
From: Aharon Landau <aharonl@mellanox.com>
Date: Mon, 16 Dec 2019 12:50:13 +0200
Subject: net/mlx5e: Add discard counters per priority

Add counters that count (per priority) the number of received
packets that dropped due to lack of buffers on a physical port. If
this counter is increasing, it implies that the adapter is
congested and cannot absorb the traffic coming from the network.

Signed-off-by: Aharon Landau <aharonl@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_stats.c | 1 +
 include/linux/mlx5/mlx5_ifc.h                      | 4 +++-
 2 files changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
index 9f09253f9f46..33081b7edbf3 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
@@ -1130,6 +1130,7 @@ static void mlx5e_grp_per_port_buffer_congest_update_stats(struct mlx5e_priv *pr
 static const struct counter_desc pport_per_prio_traffic_stats_desc[] = {
 	{ "rx_prio%d_bytes", PPORT_PER_PRIO_OFF(rx_octets) },
 	{ "rx_prio%d_packets", PPORT_PER_PRIO_OFF(rx_frames) },
+	{ "rx_prio%d_discards", PPORT_PER_PRIO_OFF(rx_discards) },
 	{ "tx_prio%d_bytes", PPORT_PER_PRIO_OFF(tx_octets) },
 	{ "tx_prio%d_packets", PPORT_PER_PRIO_OFF(tx_frames) },
 };
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 2ab4562b4851..ee0a34d66c7c 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -2180,7 +2180,9 @@ struct mlx5_ifc_eth_per_prio_grp_data_layout_bits {
 
 	u8         rx_pause_transition_low[0x20];
 
-	u8         reserved_at_3c0[0x40];
+	u8         rx_discards_high[0x20];
+
+	u8         rx_discards_low[0x20];
 
 	u8         device_stall_minor_watermark_cnt_high[0x20];
 
-- 
cgit v1.2.3-59-g8ed1b


From 61dc7b0141c51f5fa4aed97e49f9cf102ec51479 Mon Sep 17 00:00:00 2001
From: Paul Blakey <paulb@mellanox.com>
Date: Thu, 14 Nov 2019 16:59:58 +0200
Subject: net/mlx5: Refactor mlx5_create_auto_grouped_flow_table

Refactor mlx5_create_auto_grouped_flow_table() to use ft_attr param
which already carries the max_fte, prio and flags memebers, and is
used the same in similar mlx5_create_flow_table() function.

Signed-off-by: Paul Blakey <paulb@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Reviewed-by: Oz Shlomo <ozsh@mellanox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/infiniband/hw/mlx5/main.c                   | 10 ++++++----
 .../net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c |  9 ++++++---
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c     | 13 ++++++++-----
 drivers/net/ethernet/mellanox/mlx5/core/eswitch.c   |  7 +++++--
 .../ethernet/mellanox/mlx5/core/eswitch_offloads.c  | 13 +++++++------
 .../mellanox/mlx5/core/eswitch_offloads_termtbl.c   | 11 ++++++-----
 drivers/net/ethernet/mellanox/mlx5/core/fs_core.c   | 21 ++++++---------------
 include/linux/mlx5/fs.h                             | 16 ++++++++--------
 8 files changed, 52 insertions(+), 48 deletions(-)

diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
index 51100350b688..f3e6abfa4872 100644
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@ -3239,12 +3239,14 @@ static struct mlx5_ib_flow_prio *_get_prio(struct mlx5_flow_namespace *ns,
 					   int num_entries, int num_groups,
 					   u32 flags)
 {
+	struct mlx5_flow_table_attr ft_attr = {};
 	struct mlx5_flow_table *ft;
 
-	ft = mlx5_create_auto_grouped_flow_table(ns, priority,
-						 num_entries,
-						 num_groups,
-						 0, flags);
+	ft_attr.prio = priority;
+	ft_attr.max_fte = num_entries;
+	ft_attr.flags = flags;
+	ft_attr.autogroup.max_num_groups = num_groups;
+	ft = mlx5_create_auto_grouped_flow_table(ns, &ft_attr);
 	if (IS_ERR(ft))
 		return ERR_CAST(ft);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c
index acd946f2ddbe..3bc2ac3d53fc 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c
@@ -58,6 +58,7 @@ static struct mlx5e_ethtool_table *get_flow_table(struct mlx5e_priv *priv,
 						  struct ethtool_rx_flow_spec *fs,
 						  int num_tuples)
 {
+	struct mlx5_flow_table_attr ft_attr = {};
 	struct mlx5e_ethtool_table *eth_ft;
 	struct mlx5_flow_namespace *ns;
 	struct mlx5_flow_table *ft;
@@ -102,9 +103,11 @@ static struct mlx5e_ethtool_table *get_flow_table(struct mlx5e_priv *priv,
 	table_size = min_t(u32, BIT(MLX5_CAP_FLOWTABLE(priv->mdev,
 						       flow_table_properties_nic_receive.log_max_ft_size)),
 			   MLX5E_ETHTOOL_NUM_ENTRIES);
-	ft = mlx5_create_auto_grouped_flow_table(ns, prio,
-						 table_size,
-						 MLX5E_ETHTOOL_NUM_GROUPS, 0, 0);
+
+	ft_attr.prio = prio;
+	ft_attr.max_fte = table_size;
+	ft_attr.autogroup.max_num_groups = MLX5E_ETHTOOL_NUM_GROUPS;
+	ft = mlx5_create_auto_grouped_flow_table(ns, &ft_attr);
 	if (IS_ERR(ft))
 		return (void *)ft;
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 9b32a9c0f497..80d91e8abcfa 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -960,7 +960,8 @@ mlx5e_tc_add_nic_flow(struct mlx5e_priv *priv,
 
 	mutex_lock(&priv->fs.tc.t_lock);
 	if (IS_ERR_OR_NULL(priv->fs.tc.t)) {
-		int tc_grp_size, tc_tbl_size;
+		struct mlx5_flow_table_attr ft_attr = {};
+		int tc_grp_size, tc_tbl_size, tc_num_grps;
 		u32 max_flow_counter;
 
 		max_flow_counter = (MLX5_CAP_GEN(dev, max_flow_counter_31_16) << 16) |
@@ -970,13 +971,15 @@ mlx5e_tc_add_nic_flow(struct mlx5e_priv *priv,
 
 		tc_tbl_size = min_t(int, tc_grp_size * MLX5E_TC_TABLE_NUM_GROUPS,
 				    BIT(MLX5_CAP_FLOWTABLE_NIC_RX(dev, log_max_ft_size)));
+		tc_num_grps = MLX5E_TC_TABLE_NUM_GROUPS;
 
+		ft_attr.prio = MLX5E_TC_PRIO;
+		ft_attr.max_fte = tc_tbl_size;
+		ft_attr.level = MLX5E_TC_FT_LEVEL;
+		ft_attr.autogroup.max_num_groups = tc_num_grps;
 		priv->fs.tc.t =
 			mlx5_create_auto_grouped_flow_table(priv->fs.ns,
-							    MLX5E_TC_PRIO,
-							    tc_tbl_size,
-							    MLX5E_TC_TABLE_NUM_GROUPS,
-							    MLX5E_TC_FT_LEVEL, 0);
+							    &ft_attr);
 		if (IS_ERR(priv->fs.tc.t)) {
 			mutex_unlock(&priv->fs.tc.t_lock);
 			NL_SET_ERR_MSG_MOD(extack,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index 2c965ad0d744..05b13a1e829c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -277,6 +277,7 @@ enum {
 
 static int esw_create_legacy_vepa_table(struct mlx5_eswitch *esw)
 {
+	struct mlx5_flow_table_attr ft_attr = {};
 	struct mlx5_core_dev *dev = esw->dev;
 	struct mlx5_flow_namespace *root_ns;
 	struct mlx5_flow_table *fdb;
@@ -289,8 +290,10 @@ static int esw_create_legacy_vepa_table(struct mlx5_eswitch *esw)
 	}
 
 	/* num FTE 2, num FG 2 */
-	fdb = mlx5_create_auto_grouped_flow_table(root_ns, LEGACY_VEPA_PRIO,
-						  2, 2, 0, 0);
+	ft_attr.prio = LEGACY_VEPA_PRIO;
+	ft_attr.max_fte = 2;
+	ft_attr.autogroup.max_num_groups = 2;
+	fdb = mlx5_create_auto_grouped_flow_table(root_ns, &ft_attr);
 	if (IS_ERR(fdb)) {
 		err = PTR_ERR(fdb);
 		esw_warn(dev, "Failed to create VEPA FDB err %d\n", err);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 243a5440867e..4b0d992263b1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -904,6 +904,7 @@ create_next_size_table(struct mlx5_eswitch *esw,
 		       int level,
 		       u32 flags)
 {
+	struct mlx5_flow_table_attr ft_attr = {};
 	struct mlx5_flow_table *fdb;
 	int sz;
 
@@ -911,12 +912,12 @@ create_next_size_table(struct mlx5_eswitch *esw,
 	if (!sz)
 		return ERR_PTR(-ENOSPC);
 
-	fdb = mlx5_create_auto_grouped_flow_table(ns,
-						  table_prio,
-						  sz,
-						  ESW_OFFLOADS_NUM_GROUPS,
-						  level,
-						  flags);
+	ft_attr.max_fte = sz;
+	ft_attr.prio = table_prio;
+	ft_attr.level = level;
+	ft_attr.flags = flags;
+	ft_attr.autogroup.max_num_groups = ESW_OFFLOADS_NUM_GROUPS;
+	fdb = mlx5_create_auto_grouped_flow_table(ns, &ft_attr);
 	if (IS_ERR(fdb)) {
 		esw_warn(esw->dev, "Failed to create FDB Table err %d (table prio: %d, level: %d, size: %d)\n",
 			 (int)PTR_ERR(fdb), table_prio, level, sz);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c
index 366bda1bb1c3..dc08ed9339ab 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c
@@ -50,8 +50,8 @@ mlx5_eswitch_termtbl_create(struct mlx5_core_dev *dev,
 			    struct mlx5_flow_act *flow_act)
 {
 	static const struct mlx5_flow_spec spec = {};
+	struct mlx5_flow_table_attr ft_attr = {};
 	struct mlx5_flow_namespace *root_ns;
-	int prio, flags;
 	int err;
 
 	root_ns = mlx5_get_flow_namespace(dev, MLX5_FLOW_NAMESPACE_FDB);
@@ -63,10 +63,11 @@ mlx5_eswitch_termtbl_create(struct mlx5_core_dev *dev,
 	/* As this is the terminating action then the termination table is the
 	 * same prio as the slow path
 	 */
-	prio = FDB_SLOW_PATH;
-	flags = MLX5_FLOW_TABLE_TERMINATION;
-	tt->termtbl = mlx5_create_auto_grouped_flow_table(root_ns, prio, 1, 1,
-							  0, flags);
+	ft_attr.flags = MLX5_FLOW_TABLE_TERMINATION;
+	ft_attr.prio = FDB_SLOW_PATH;
+	ft_attr.max_fte = 1;
+	ft_attr.autogroup.max_num_groups = 1;
+	tt->termtbl = mlx5_create_auto_grouped_flow_table(root_ns, &ft_attr);
 	if (IS_ERR(tt->termtbl)) {
 		esw_warn(dev, "Failed to create termination table\n");
 		return -EOPNOTSUPP;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
index d60577484567..04ebdd6cf670 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
@@ -1111,31 +1111,22 @@ EXPORT_SYMBOL(mlx5_create_lag_demux_flow_table);
 
 struct mlx5_flow_table*
 mlx5_create_auto_grouped_flow_table(struct mlx5_flow_namespace *ns,
-				    int prio,
-				    int num_flow_table_entries,
-				    int max_num_groups,
-				    u32 level,
-				    u32 flags)
+				    struct mlx5_flow_table_attr *ft_attr)
 {
-	struct mlx5_flow_table_attr ft_attr = {};
 	struct mlx5_flow_table *ft;
 
-	if (max_num_groups > num_flow_table_entries)
+	if (ft_attr->autogroup.max_num_groups > ft_attr->max_fte)
 		return ERR_PTR(-EINVAL);
 
-	ft_attr.max_fte = num_flow_table_entries;
-	ft_attr.prio    = prio;
-	ft_attr.level   = level;
-	ft_attr.flags   = flags;
-
-	ft = mlx5_create_flow_table(ns, &ft_attr);
+	ft = mlx5_create_flow_table(ns, ft_attr);
 	if (IS_ERR(ft))
 		return ft;
 
 	ft->autogroup.active = true;
-	ft->autogroup.required_groups = max_num_groups;
+	ft->autogroup.required_groups = ft_attr->autogroup.max_num_groups;
 	/* We save place for flow groups in addition to max types */
-	ft->autogroup.group_size = ft->max_fte / (max_num_groups + 1);
+	ft->autogroup.group_size = ft->max_fte /
+				   (ft->autogroup.required_groups + 1);
 
 	return ft;
 }
diff --git a/include/linux/mlx5/fs.h b/include/linux/mlx5/fs.h
index 4e5b84e66822..a3f8b63839de 100644
--- a/include/linux/mlx5/fs.h
+++ b/include/linux/mlx5/fs.h
@@ -145,25 +145,25 @@ mlx5_get_flow_vport_acl_namespace(struct mlx5_core_dev *dev,
 				  enum mlx5_flow_namespace_type type,
 				  int vport);
 
-struct mlx5_flow_table *
-mlx5_create_auto_grouped_flow_table(struct mlx5_flow_namespace *ns,
-				    int prio,
-				    int num_flow_table_entries,
-				    int max_num_groups,
-				    u32 level,
-				    u32 flags);
-
 struct mlx5_flow_table_attr {
 	int prio;
 	int max_fte;
 	u32 level;
 	u32 flags;
+
+	struct {
+		int max_num_groups;
+	} autogroup;
 };
 
 struct mlx5_flow_table *
 mlx5_create_flow_table(struct mlx5_flow_namespace *ns,
 		       struct mlx5_flow_table_attr *ft_attr);
 
+struct mlx5_flow_table *
+mlx5_create_auto_grouped_flow_table(struct mlx5_flow_namespace *ns,
+				    struct mlx5_flow_table_attr *ft_attr);
+
 struct mlx5_flow_table *
 mlx5_create_vport_flow_table(struct mlx5_flow_namespace *ns,
 			     int prio,
-- 
cgit v1.2.3-59-g8ed1b


From 5281a0c909194c477656e89401ac11dd7b29ad2d Mon Sep 17 00:00:00 2001
From: Paul Blakey <paulb@mellanox.com>
Date: Tue, 23 Jul 2019 11:43:57 +0300
Subject: net/mlx5: fs_core: Introduce unmanaged flow tables

Currently, Most of the steering tree is statically declared ahead of time,
with steering prios instances allocated for each fdb chain to assign max
number of levels for each of them. This allows fs_core to manage the
connections and  levels of the flow tables hierarcy to prevent loops, but
restricts us with the number of supported chains and priorities.

Introduce unmananged flow tables, allowing the user to manage the flow
table connections. A unamanged table is detached from the fs_core flow
table hierarcy, and is only connected back to the hierarchy by explicit
FTEs forward actions.

This will be used together with firmware that supports ignoring the flow
table levels to increase the number of supported chains and prios.

Signed-off-by: Paul Blakey <paulb@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Reviewed-by: Oz Shlomo <ozsh@mellanox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 41 ++++++++++++++++-------
 include/linux/mlx5/fs.h                           |  2 ++
 2 files changed, 31 insertions(+), 12 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
index 51913e2cde5c..456d3739b166 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
@@ -1006,7 +1006,8 @@ static struct mlx5_flow_table *__mlx5_create_flow_table(struct mlx5_flow_namespa
 							u16 vport)
 {
 	struct mlx5_flow_root_namespace *root = find_root(&ns->node);
-	struct mlx5_flow_table *next_ft = NULL;
+	bool unmanaged = ft_attr->flags & MLX5_FLOW_TABLE_UNMANAGED;
+	struct mlx5_flow_table *next_ft;
 	struct fs_prio *fs_prio = NULL;
 	struct mlx5_flow_table *ft;
 	int log_table_sz;
@@ -1023,14 +1024,21 @@ static struct mlx5_flow_table *__mlx5_create_flow_table(struct mlx5_flow_namespa
 		err = -EINVAL;
 		goto unlock_root;
 	}
-	if (ft_attr->level >= fs_prio->num_levels) {
-		err = -ENOSPC;
-		goto unlock_root;
+	if (!unmanaged) {
+		/* The level is related to the
+		 * priority level range.
+		 */
+		if (ft_attr->level >= fs_prio->num_levels) {
+			err = -ENOSPC;
+			goto unlock_root;
+		}
+
+		ft_attr->level += fs_prio->start_level;
 	}
+
 	/* The level is related to the
 	 * priority level range.
 	 */
-	ft_attr->level += fs_prio->start_level;
 	ft = alloc_flow_table(ft_attr->level,
 			      vport,
 			      ft_attr->max_fte ? roundup_pow_of_two(ft_attr->max_fte) : 0,
@@ -1043,19 +1051,27 @@ static struct mlx5_flow_table *__mlx5_create_flow_table(struct mlx5_flow_namespa
 
 	tree_init_node(&ft->node, del_hw_flow_table, del_sw_flow_table);
 	log_table_sz = ft->max_fte ? ilog2(ft->max_fte) : 0;
-	next_ft = find_next_chained_ft(fs_prio);
+	next_ft = unmanaged ? ft_attr->next_ft :
+			      find_next_chained_ft(fs_prio);
 	ft->def_miss_action = ns->def_miss_action;
 	err = root->cmds->create_flow_table(root, ft, log_table_sz, next_ft);
 	if (err)
 		goto free_ft;
 
-	err = connect_flow_table(root->dev, ft, fs_prio);
-	if (err)
-		goto destroy_ft;
+	if (!unmanaged) {
+		err = connect_flow_table(root->dev, ft, fs_prio);
+		if (err)
+			goto destroy_ft;
+	}
+
 	ft->node.active = true;
 	down_write_ref_node(&fs_prio->node, false);
-	tree_add_node(&ft->node, &fs_prio->node);
-	list_add_flow_table(ft, fs_prio);
+	if (!unmanaged) {
+		tree_add_node(&ft->node, &fs_prio->node);
+		list_add_flow_table(ft, fs_prio);
+	} else {
+		ft->node.root = fs_prio->node.root;
+	}
 	fs_prio->num_ft++;
 	up_write_ref_node(&fs_prio->node, false);
 	mutex_unlock(&root->chain_lock);
@@ -2024,7 +2040,8 @@ int mlx5_destroy_flow_table(struct mlx5_flow_table *ft)
 	int err = 0;
 
 	mutex_lock(&root->chain_lock);
-	err = disconnect_flow_table(ft);
+	if (!(ft->flags & MLX5_FLOW_TABLE_UNMANAGED))
+		err = disconnect_flow_table(ft);
 	if (err) {
 		mutex_unlock(&root->chain_lock);
 		return err;
diff --git a/include/linux/mlx5/fs.h b/include/linux/mlx5/fs.h
index a3f8b63839de..de2c838bae90 100644
--- a/include/linux/mlx5/fs.h
+++ b/include/linux/mlx5/fs.h
@@ -48,6 +48,7 @@ enum {
 	MLX5_FLOW_TABLE_TUNNEL_EN_REFORMAT = BIT(0),
 	MLX5_FLOW_TABLE_TUNNEL_EN_DECAP = BIT(1),
 	MLX5_FLOW_TABLE_TERMINATION = BIT(2),
+	MLX5_FLOW_TABLE_UNMANAGED = BIT(3),
 };
 
 #define LEFTOVERS_RULE_NUM	 2
@@ -150,6 +151,7 @@ struct mlx5_flow_table_attr {
 	int max_fte;
 	u32 level;
 	u32 flags;
+	struct mlx5_flow_table *next_ft;
 
 	struct {
 		int max_num_groups;
-- 
cgit v1.2.3-59-g8ed1b


From ff189b43568216c6211e9e7ddd9026cb8295e744 Mon Sep 17 00:00:00 2001
From: Paul Blakey <paulb@mellanox.com>
Date: Sun, 5 Jan 2020 15:15:54 +0200
Subject: net/mlx5: Add ignore level support fwd to table rules

If user sets ignore flow level flag on a rule, that rule can point to
a flow table of any level, including those with levels equal or less
than the level of the flow table it is added on.

This with unamanged tables will be used to create a FDB chain/prio
hierarchy much larger than currently supported level range.

Signed-off-by: Paul Blakey <paulb@mellanox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c  |  3 +++
 drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 18 +++++++++++++++---
 include/linux/mlx5/fs.h                           |  1 +
 3 files changed, 19 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
index 3c816e81f8d9..b25465d9e030 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
@@ -432,6 +432,9 @@ static int mlx5_cmd_set_fte(struct mlx5_core_dev *dev,
 	MLX5_SET(set_fte_in, in, table_type, ft->type);
 	MLX5_SET(set_fte_in, in, table_id,   ft->id);
 	MLX5_SET(set_fte_in, in, flow_index, fte->index);
+	MLX5_SET(set_fte_in, in, ignore_flow_level,
+		 !!(fte->action.flags & FLOW_ACT_IGNORE_FLOW_LEVEL));
+
 	if (ft->vport) {
 		MLX5_SET(set_fte_in, in, vport_number, ft->vport);
 		MLX5_SET(set_fte_in, in, other_vport, 1);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
index 456d3739b166..355a424f4506 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
@@ -1536,18 +1536,30 @@ static bool counter_is_valid(u32 action)
 }
 
 static bool dest_is_valid(struct mlx5_flow_destination *dest,
-			  u32 action,
+			  struct mlx5_flow_act *flow_act,
 			  struct mlx5_flow_table *ft)
 {
+	bool ignore_level = flow_act->flags & FLOW_ACT_IGNORE_FLOW_LEVEL;
+	u32 action = flow_act->action;
+
 	if (dest && (dest->type == MLX5_FLOW_DESTINATION_TYPE_COUNTER))
 		return counter_is_valid(action);
 
 	if (!(action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST))
 		return true;
 
+	if (ignore_level) {
+		if (ft->type != FS_FT_FDB)
+			return false;
+
+		if (dest->type == MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE &&
+		    dest->ft->type != FS_FT_FDB)
+			return false;
+	}
+
 	if (!dest || ((dest->type ==
 	    MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE) &&
-	    (dest->ft->level <= ft->level)))
+	    (dest->ft->level <= ft->level && !ignore_level)))
 		return false;
 	return true;
 }
@@ -1777,7 +1789,7 @@ _mlx5_add_flow_rules(struct mlx5_flow_table *ft,
 		return ERR_PTR(-EINVAL);
 
 	for (i = 0; i < dest_num; i++) {
-		if (!dest_is_valid(&dest[i], flow_act->action, ft))
+		if (!dest_is_valid(&dest[i], flow_act, ft))
 			return ERR_PTR(-EINVAL);
 	}
 	nested_down_read_ref_node(&ft->node, FS_LOCK_GRANDPARENT);
diff --git a/include/linux/mlx5/fs.h b/include/linux/mlx5/fs.h
index de2c838bae90..81f393fb7d96 100644
--- a/include/linux/mlx5/fs.h
+++ b/include/linux/mlx5/fs.h
@@ -196,6 +196,7 @@ struct mlx5_fs_vlan {
 
 enum {
 	FLOW_ACT_NO_APPEND = BIT(0),
+	FLOW_ACT_IGNORE_FLOW_LEVEL = BIT(1),
 };
 
 struct mlx5_flow_act {
-- 
cgit v1.2.3-59-g8ed1b


From 79cdb0aaea8b5478db34afa1d4d5ecc808689a67 Mon Sep 17 00:00:00 2001
From: Paul Blakey <paulb@mellanox.com>
Date: Thu, 14 Nov 2019 17:02:59 +0200
Subject: net/mlx5: Allow creating autogroups with reserved entries

Exclude the last n entries for an autogrouped flow table.

Reserving entries at the end of the FT will ensure that this FG will be
the last to be evaluated. This will be used in the next patch to create
a miss group enabling custom actions on FT miss.

Signed-off-by: Paul Blakey <paulb@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Reviewed-by: Oz Shlomo <ozsh@mellanox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 26 +++++++++++++++--------
 drivers/net/ethernet/mellanox/mlx5/core/fs_core.h |  1 +
 include/linux/mlx5/fs.h                           |  1 +
 3 files changed, 19 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
index 355a424f4506..c7a16ae05fa8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
@@ -579,7 +579,9 @@ static void del_sw_flow_group(struct fs_node *node)
 
 	rhashtable_destroy(&fg->ftes_hash);
 	ida_destroy(&fg->fte_allocator);
-	if (ft->autogroup.active && fg->max_ftes == ft->autogroup.group_size)
+	if (ft->autogroup.active &&
+	    fg->max_ftes == ft->autogroup.group_size &&
+	    fg->start_index < ft->autogroup.max_fte)
 		ft->autogroup.num_groups--;
 	err = rhltable_remove(&ft->fgs_hash,
 			      &fg->hash,
@@ -1121,9 +1123,14 @@ struct mlx5_flow_table*
 mlx5_create_auto_grouped_flow_table(struct mlx5_flow_namespace *ns,
 				    struct mlx5_flow_table_attr *ft_attr)
 {
+	int num_reserved_entries = ft_attr->autogroup.num_reserved_entries;
+	int autogroups_max_fte = ft_attr->max_fte - num_reserved_entries;
+	int max_num_groups = ft_attr->autogroup.max_num_groups;
 	struct mlx5_flow_table *ft;
 
-	if (ft_attr->autogroup.max_num_groups > ft_attr->max_fte)
+	if (max_num_groups > autogroups_max_fte)
+		return ERR_PTR(-EINVAL);
+	if (num_reserved_entries > ft_attr->max_fte)
 		return ERR_PTR(-EINVAL);
 
 	ft = mlx5_create_flow_table(ns, ft_attr);
@@ -1131,10 +1138,10 @@ mlx5_create_auto_grouped_flow_table(struct mlx5_flow_namespace *ns,
 		return ft;
 
 	ft->autogroup.active = true;
-	ft->autogroup.required_groups = ft_attr->autogroup.max_num_groups;
+	ft->autogroup.required_groups = max_num_groups;
+	ft->autogroup.max_fte = autogroups_max_fte;
 	/* We save place for flow groups in addition to max types */
-	ft->autogroup.group_size = ft->max_fte /
-				   (ft->autogroup.required_groups + 1);
+	ft->autogroup.group_size = autogroups_max_fte / (max_num_groups + 1);
 
 	return ft;
 }
@@ -1156,7 +1163,7 @@ struct mlx5_flow_group *mlx5_create_flow_group(struct mlx5_flow_table *ft,
 	struct mlx5_flow_group *fg;
 	int err;
 
-	if (ft->autogroup.active)
+	if (ft->autogroup.active && start_index < ft->autogroup.max_fte)
 		return ERR_PTR(-EPERM);
 
 	down_write_ref_node(&ft->node, false);
@@ -1329,9 +1336,10 @@ static struct mlx5_flow_group *alloc_auto_flow_group(struct mlx5_flow_table  *ft
 						     const struct mlx5_flow_spec *spec)
 {
 	struct list_head *prev = &ft->node.children;
-	struct mlx5_flow_group *fg;
+	u32 max_fte = ft->autogroup.max_fte;
 	unsigned int candidate_index = 0;
 	unsigned int group_size = 0;
+	struct mlx5_flow_group *fg;
 
 	if (!ft->autogroup.active)
 		return ERR_PTR(-ENOENT);
@@ -1339,7 +1347,7 @@ static struct mlx5_flow_group *alloc_auto_flow_group(struct mlx5_flow_table  *ft
 	if (ft->autogroup.num_groups < ft->autogroup.required_groups)
 		group_size = ft->autogroup.group_size;
 
-	/*  ft->max_fte == ft->autogroup.max_types */
+	/*  max_fte == ft->autogroup.max_types */
 	if (group_size == 0)
 		group_size = 1;
 
@@ -1352,7 +1360,7 @@ static struct mlx5_flow_group *alloc_auto_flow_group(struct mlx5_flow_table  *ft
 		prev = &fg->node.list;
 	}
 
-	if (candidate_index + group_size > ft->max_fte)
+	if (candidate_index + group_size > max_fte)
 		return ERR_PTR(-ENOSPC);
 
 	fg = alloc_insert_flow_group(ft,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
index c2621b911563..be5f5e32c1e8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
@@ -164,6 +164,7 @@ struct mlx5_flow_table {
 		unsigned int		required_groups;
 		unsigned int		group_size;
 		unsigned int		num_groups;
+		unsigned int		max_fte;
 	} autogroup;
 	/* Protect fwd_rules */
 	struct mutex			lock;
diff --git a/include/linux/mlx5/fs.h b/include/linux/mlx5/fs.h
index 81f393fb7d96..4cae16016b2b 100644
--- a/include/linux/mlx5/fs.h
+++ b/include/linux/mlx5/fs.h
@@ -155,6 +155,7 @@ struct mlx5_flow_table_attr {
 
 	struct {
 		int max_num_groups;
+		int num_reserved_entries;
 	} autogroup;
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From e66cbc961cfe375f977b3bfae13abec4df8c8521 Mon Sep 17 00:00:00 2001
From: Paul Blakey <paulb@mellanox.com>
Date: Tue, 26 Nov 2019 14:13:42 +0200
Subject: net/mlx5: ft: Use getter function to get ft chain

FT chain is defined as the next chain after tc.

To prepare for next patches that will increase the number of tc
chains available at runtime, use a getter function to get this
value.

The define is still used in static fs_core allocation,
to calculate the number of chains. This static allocation
will be used if the relevant capabilities won't be available
to support dynamic chains.

Signed-off-by: Paul Blakey <paulb@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_rep.c           | 2 +-
 drivers/net/ethernet/mellanox/mlx5/core/eswitch.h          | 3 +++
 drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c | 5 +++++
 3 files changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index f175cb24bb67..d85b56452ee1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -1268,7 +1268,7 @@ static int mlx5e_rep_setup_ft_cb(enum tc_setup_type type, void *type_data,
 		 * reserved ft chain.
 		 */
 		memcpy(&cls_flower, f, sizeof(*f));
-		cls_flower.common.chain_index = FDB_FT_CHAIN;
+		cls_flower.common.chain_index = mlx5_eswitch_get_ft_chain(esw);
 		err = mlx5e_rep_setup_tc_cls_flower(priv, &cls_flower, flags);
 		memcpy(&f->stats, &cls_flower.stats, sizeof(f->stats));
 		return err;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index ffcff3ba3701..69ff3031d1c0 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -364,6 +364,9 @@ mlx5_eswitch_get_prio_range(struct mlx5_eswitch *esw);
 u32
 mlx5_eswitch_get_chain_range(struct mlx5_eswitch *esw);
 
+unsigned int
+mlx5_eswitch_get_ft_chain(struct mlx5_eswitch *esw);
+
 struct mlx5_flow_handle *
 mlx5_eswitch_create_vport_rx_rule(struct mlx5_eswitch *esw, u16 vport,
 				  struct mlx5_flow_destination *dest);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 4b0d992263b1..81bafd4b44bb 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -80,6 +80,11 @@ u32 mlx5_eswitch_get_chain_range(struct mlx5_eswitch *esw)
 	return 0;
 }
 
+u32 mlx5_eswitch_get_ft_chain(struct mlx5_eswitch *esw)
+{
+	return mlx5_eswitch_get_chain_range(esw) + 1;
+}
+
 u16 mlx5_eswitch_get_prio_range(struct mlx5_eswitch *esw)
 {
 	if (esw->fdb_table.flags & ESW_FDB_CHAINS_AND_PRIOS_SUPPORTED)
-- 
cgit v1.2.3-59-g8ed1b


From 82270e12544ee76ea9a3117a769a6d466a2e646b Mon Sep 17 00:00:00 2001
From: Paul Blakey <paulb@mellanox.com>
Date: Tue, 26 Nov 2019 14:15:00 +0200
Subject: net/mlx5: ft: Check prio and chain sanity for ft offload

Before changing the chain from original chain to ft offload chain,
make sure user doesn't actually use chains.

While here, normalize the prio range to that which we support.

Signed-off-by: Paul Blakey <paulb@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_rep.c | 27 ++++++++++++++++++------
 1 file changed, 20 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index d85b56452ee1..5d93c506ae8b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -1247,8 +1247,7 @@ static int mlx5e_rep_setup_tc_cb(enum tc_setup_type type, void *type_data,
 static int mlx5e_rep_setup_ft_cb(enum tc_setup_type type, void *type_data,
 				 void *cb_priv)
 {
-	struct flow_cls_offload *f = type_data;
-	struct flow_cls_offload cls_flower;
+	struct flow_cls_offload tmp, *f = type_data;
 	struct mlx5e_priv *priv = cb_priv;
 	struct mlx5_eswitch *esw;
 	unsigned long flags;
@@ -1261,16 +1260,30 @@ static int mlx5e_rep_setup_ft_cb(enum tc_setup_type type, void *type_data,
 
 	switch (type) {
 	case TC_SETUP_CLSFLOWER:
-		if (!mlx5_eswitch_prios_supported(esw) || f->common.chain_index)
+		memcpy(&tmp, f, sizeof(*f));
+
+		if (!mlx5_eswitch_prios_supported(esw) ||
+		    tmp.common.chain_index)
 			return -EOPNOTSUPP;
 
 		/* Re-use tc offload path by moving the ft flow to the
 		 * reserved ft chain.
+		 *
+		 * FT offload can use prio range [0, INT_MAX], so we
+		 * normalize it to range [1, mlx5_eswitch_get_prio_range(esw)]
+		 * as with tc, where prio 0 isn't supported.
+		 *
+		 * We only support chain 0 of FT offload.
 		 */
-		memcpy(&cls_flower, f, sizeof(*f));
-		cls_flower.common.chain_index = mlx5_eswitch_get_ft_chain(esw);
-		err = mlx5e_rep_setup_tc_cls_flower(priv, &cls_flower, flags);
-		memcpy(&f->stats, &cls_flower.stats, sizeof(f->stats));
+		if (tmp.common.prio >= mlx5_eswitch_get_prio_range(esw))
+			return -EOPNOTSUPP;
+		if (tmp.common.chain_index != 0)
+			return -EOPNOTSUPP;
+
+		tmp.common.chain_index = mlx5_eswitch_get_ft_chain(esw);
+		tmp.common.prio++;
+		err = mlx5e_rep_setup_tc_cls_flower(priv, &tmp, flags);
+		memcpy(&f->stats, &tmp.stats, sizeof(f->stats));
 		return err;
 	default:
 		return -EOPNOTSUPP;
-- 
cgit v1.2.3-59-g8ed1b


From 39ac237ce00968545e7298faa9e07ecb7e440fb5 Mon Sep 17 00:00:00 2001
From: Paul Blakey <paulb@mellanox.com>
Date: Wed, 8 Jan 2020 12:11:04 +0200
Subject: net/mlx5: E-Switch, Refactor chains and priorities

To support the entire chain and prio range (32bit + 16bit),
instead of a using a static array of chains/prios of limited size, create
them dynamically, and use a rhashtable to search for existing chains/prio
combinations.

This will be used in next patch to actually increase the number using
unamanged tables support and ignore flow level capability.

Signed-off-by: Paul Blakey <paulb@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Reviewed-by: Oz Shlomo <ozsh@mellanox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/Makefile   |   2 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_rep.c   |  11 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c    |  14 +-
 drivers/net/ethernet/mellanox/mlx5/core/eswitch.h  |  30 +-
 .../ethernet/mellanox/mlx5/core/eswitch_offloads.c | 303 ++----------
 .../mellanox/mlx5/core/eswitch_offloads_chains.c   | 542 +++++++++++++++++++++
 .../mellanox/mlx5/core/eswitch_offloads_chains.h   |  27 +
 7 files changed, 637 insertions(+), 292 deletions(-)
 create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.c
 create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.h

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
index a6f390fdb971..d3e06cec8317 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -42,7 +42,7 @@ mlx5_core-$(CONFIG_PCI_HYPERV_INTERFACE) += en/hv_vhca_stats.o
 # Core extra
 #
 mlx5_core-$(CONFIG_MLX5_ESWITCH)   += eswitch.o eswitch_offloads.o eswitch_offloads_termtbl.o \
-				      ecpf.o rdma.o
+				      ecpf.o rdma.o eswitch_offloads_chains.o
 mlx5_core-$(CONFIG_MLX5_MPFS)      += lib/mpfs.o
 mlx5_core-$(CONFIG_VXLAN)          += lib/vxlan.o
 mlx5_core-$(CONFIG_PTP_1588_CLOCK) += lib/clock.o
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index 5d93c506ae8b..446eb4d6c983 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -41,6 +41,7 @@
 #include <net/ipv6_stubs.h>
 
 #include "eswitch.h"
+#include "eswitch_offloads_chains.h"
 #include "en.h"
 #include "en_rep.h"
 #include "en_tc.h"
@@ -1262,25 +1263,25 @@ static int mlx5e_rep_setup_ft_cb(enum tc_setup_type type, void *type_data,
 	case TC_SETUP_CLSFLOWER:
 		memcpy(&tmp, f, sizeof(*f));
 
-		if (!mlx5_eswitch_prios_supported(esw) ||
+		if (!mlx5_esw_chains_prios_supported(esw) ||
 		    tmp.common.chain_index)
 			return -EOPNOTSUPP;
 
 		/* Re-use tc offload path by moving the ft flow to the
 		 * reserved ft chain.
 		 *
-		 * FT offload can use prio range [0, INT_MAX], so we
-		 * normalize it to range [1, mlx5_eswitch_get_prio_range(esw)]
+		 * FT offload can use prio range [0, INT_MAX], so we normalize
+		 * it to range [1, mlx5_esw_chains_get_prio_range(esw)]
 		 * as with tc, where prio 0 isn't supported.
 		 *
 		 * We only support chain 0 of FT offload.
 		 */
-		if (tmp.common.prio >= mlx5_eswitch_get_prio_range(esw))
+		if (tmp.common.prio >= mlx5_esw_chains_get_prio_range(esw))
 			return -EOPNOTSUPP;
 		if (tmp.common.chain_index != 0)
 			return -EOPNOTSUPP;
 
-		tmp.common.chain_index = mlx5_eswitch_get_ft_chain(esw);
+		tmp.common.chain_index = mlx5_esw_chains_get_ft_chain(esw);
 		tmp.common.prio++;
 		err = mlx5e_rep_setup_tc_cls_flower(priv, &tmp, flags);
 		memcpy(&f->stats, &tmp.stats, sizeof(f->stats));
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 5aafbb8d2e8e..26f559b453dc 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -51,6 +51,7 @@
 #include "en_rep.h"
 #include "en_tc.h"
 #include "eswitch.h"
+#include "eswitch_offloads_chains.h"
 #include "fs_core.h"
 #include "en/port.h"
 #include "en/tc_tun.h"
@@ -1083,7 +1084,7 @@ mlx5e_tc_offload_to_slow_path(struct mlx5_eswitch *esw,
 	memcpy(slow_attr, flow->esw_attr, sizeof(*slow_attr));
 	slow_attr->action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
 	slow_attr->split_count = 0;
-	slow_attr->dest_chain = FDB_TC_SLOW_PATH_CHAIN;
+	slow_attr->flags |= MLX5_ESW_ATTR_FLAG_SLOW_PATH;
 
 	rule = mlx5e_tc_offload_fdb_rules(esw, flow, spec, slow_attr);
 	if (!IS_ERR(rule))
@@ -1100,7 +1101,7 @@ mlx5e_tc_unoffload_from_slow_path(struct mlx5_eswitch *esw,
 	memcpy(slow_attr, flow->esw_attr, sizeof(*slow_attr));
 	slow_attr->action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
 	slow_attr->split_count = 0;
-	slow_attr->dest_chain = FDB_TC_SLOW_PATH_CHAIN;
+	slow_attr->flags |= MLX5_ESW_ATTR_FLAG_SLOW_PATH;
 	mlx5e_tc_unoffload_fdb_rules(esw, flow, slow_attr);
 	flow_flag_clear(flow, SLOW);
 }
@@ -1160,19 +1161,18 @@ mlx5e_tc_add_fdb_flow(struct mlx5e_priv *priv,
 		      struct netlink_ext_ack *extack)
 {
 	struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
-	u32 max_chain = mlx5_eswitch_get_chain_range(esw);
 	struct mlx5_esw_flow_attr *attr = flow->esw_attr;
 	struct mlx5e_tc_flow_parse_attr *parse_attr = attr->parse_attr;
-	u16 max_prio = mlx5_eswitch_get_prio_range(esw);
 	struct net_device *out_dev, *encap_dev = NULL;
 	struct mlx5_fc *counter = NULL;
 	struct mlx5e_rep_priv *rpriv;
 	struct mlx5e_priv *out_priv;
 	bool encap_valid = true;
+	u32 max_prio, max_chain;
 	int err = 0;
 	int out_index;
 
-	if (!mlx5_eswitch_prios_supported(esw) && attr->prio != 1) {
+	if (!mlx5_esw_chains_prios_supported(esw) && attr->prio != 1) {
 		NL_SET_ERR_MSG(extack, "E-switch priorities unsupported, upgrade FW");
 		return -EOPNOTSUPP;
 	}
@@ -1182,11 +1182,13 @@ mlx5e_tc_add_fdb_flow(struct mlx5e_priv *priv,
 	 * FDB_FT_CHAIN which is outside tc range.
 	 * See mlx5e_rep_setup_ft_cb().
 	 */
+	max_chain = mlx5_esw_chains_get_chain_range(esw);
 	if (!mlx5e_is_ft_flow(flow) && attr->chain > max_chain) {
 		NL_SET_ERR_MSG(extack, "Requested chain is out of supported range");
 		return -EOPNOTSUPP;
 	}
 
+	max_prio = mlx5_esw_chains_get_prio_range(esw);
 	if (attr->prio > max_prio) {
 		NL_SET_ERR_MSG(extack, "Requested priority is out of supported range");
 		return -EOPNOTSUPP;
@@ -3469,7 +3471,7 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv,
 			break;
 		case FLOW_ACTION_GOTO: {
 			u32 dest_chain = act->chain_index;
-			u32 max_chain = mlx5_eswitch_get_chain_range(esw);
+			u32 max_chain = mlx5_esw_chains_get_chain_range(esw);
 
 			if (ft_flow) {
 				NL_SET_ERR_MSG_MOD(extack, "Goto action is not supported");
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index 69ff3031d1c0..4472710ccc9c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -157,7 +157,7 @@ enum offloads_fdb_flags {
 	ESW_FDB_CHAINS_AND_PRIOS_SUPPORTED = BIT(0),
 };
 
-extern const unsigned int ESW_POOLS[4];
+struct mlx5_esw_chains_priv;
 
 struct mlx5_eswitch_fdb {
 	union {
@@ -182,14 +182,7 @@ struct mlx5_eswitch_fdb {
 			struct mlx5_flow_handle *miss_rule_multi;
 			int vlan_push_pop_refcount;
 
-			struct {
-				struct mlx5_flow_table *fdb;
-				u32 num_rules;
-			} fdb_prio[FDB_NUM_CHAINS][FDB_TC_MAX_PRIO + 1][FDB_TC_LEVELS_PER_PRIO];
-			/* Protects fdb_prio table */
-			struct mutex fdb_prio_lock;
-
-			int fdb_left[ARRAY_SIZE(ESW_POOLS)];
+			struct mlx5_esw_chains_priv *esw_chains_priv;
 		} offloads;
 	};
 	u32 flags;
@@ -355,18 +348,6 @@ mlx5_eswitch_del_fwd_rule(struct mlx5_eswitch *esw,
 			  struct mlx5_flow_handle *rule,
 			  struct mlx5_esw_flow_attr *attr);
 
-bool
-mlx5_eswitch_prios_supported(struct mlx5_eswitch *esw);
-
-u16
-mlx5_eswitch_get_prio_range(struct mlx5_eswitch *esw);
-
-u32
-mlx5_eswitch_get_chain_range(struct mlx5_eswitch *esw);
-
-unsigned int
-mlx5_eswitch_get_ft_chain(struct mlx5_eswitch *esw);
-
 struct mlx5_flow_handle *
 mlx5_eswitch_create_vport_rx_rule(struct mlx5_eswitch *esw, u16 vport,
 				  struct mlx5_flow_destination *dest);
@@ -391,6 +372,11 @@ enum {
 	MLX5_ESW_DEST_ENCAP_VALID   = BIT(1),
 };
 
+enum {
+	MLX5_ESW_ATTR_FLAG_VLAN_HANDLED  = BIT(0),
+	MLX5_ESW_ATTR_FLAG_SLOW_PATH     = BIT(1),
+};
+
 struct mlx5_esw_flow_attr {
 	struct mlx5_eswitch_rep *in_rep;
 	struct mlx5_core_dev	*in_mdev;
@@ -404,7 +390,6 @@ struct mlx5_esw_flow_attr {
 	u16	vlan_vid[MLX5_FS_VLAN_DEPTH];
 	u8	vlan_prio[MLX5_FS_VLAN_DEPTH];
 	u8	total_vlan;
-	bool	vlan_handled;
 	struct {
 		u32 flags;
 		struct mlx5_eswitch_rep *rep;
@@ -419,6 +404,7 @@ struct mlx5_esw_flow_attr {
 	u32	chain;
 	u16	prio;
 	u32	dest_chain;
+	u32	flags;
 	struct mlx5e_tc_flow_parse_attr *parse_attr;
 };
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 81bafd4b44bb..47f8729197e0 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -37,6 +37,7 @@
 #include <linux/mlx5/fs.h>
 #include "mlx5_core.h"
 #include "eswitch.h"
+#include "eswitch_offloads_chains.h"
 #include "rdma.h"
 #include "en.h"
 #include "fs_core.h"
@@ -47,10 +48,6 @@
  * one for multicast.
  */
 #define MLX5_ESW_MISS_FLOWS (2)
-
-#define fdb_prio_table(esw, chain, prio, level) \
-	(esw)->fdb_table.offloads.fdb_prio[(chain)][(prio)][(level)]
-
 #define UPLINK_REP_INDEX 0
 
 static struct mlx5_eswitch_rep *mlx5_eswitch_get_rep(struct mlx5_eswitch *esw,
@@ -62,37 +59,6 @@ static struct mlx5_eswitch_rep *mlx5_eswitch_get_rep(struct mlx5_eswitch *esw,
 	return &esw->offloads.vport_reps[idx];
 }
 
-static struct mlx5_flow_table *
-esw_get_prio_table(struct mlx5_eswitch *esw, u32 chain, u16 prio, int level);
-static void
-esw_put_prio_table(struct mlx5_eswitch *esw, u32 chain, u16 prio, int level);
-
-bool mlx5_eswitch_prios_supported(struct mlx5_eswitch *esw)
-{
-	return (!!(esw->fdb_table.flags & ESW_FDB_CHAINS_AND_PRIOS_SUPPORTED));
-}
-
-u32 mlx5_eswitch_get_chain_range(struct mlx5_eswitch *esw)
-{
-	if (esw->fdb_table.flags & ESW_FDB_CHAINS_AND_PRIOS_SUPPORTED)
-		return FDB_TC_MAX_CHAIN;
-
-	return 0;
-}
-
-u32 mlx5_eswitch_get_ft_chain(struct mlx5_eswitch *esw)
-{
-	return mlx5_eswitch_get_chain_range(esw) + 1;
-}
-
-u16 mlx5_eswitch_get_prio_range(struct mlx5_eswitch *esw)
-{
-	if (esw->fdb_table.flags & ESW_FDB_CHAINS_AND_PRIOS_SUPPORTED)
-		return FDB_TC_MAX_PRIO;
-
-	return 1;
-}
-
 static bool
 esw_check_ingress_prio_tag_enabled(const struct mlx5_eswitch *esw,
 				   const struct mlx5_vport *vport)
@@ -180,10 +146,17 @@ mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw,
 	}
 
 	if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) {
-		if (attr->dest_chain) {
-			struct mlx5_flow_table *ft;
+		struct mlx5_flow_table *ft;
 
-			ft = esw_get_prio_table(esw, attr->dest_chain, 1, 0);
+		if (attr->flags & MLX5_ESW_ATTR_FLAG_SLOW_PATH) {
+			flow_act.flags |= FLOW_ACT_IGNORE_FLOW_LEVEL;
+			dest[i].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
+			dest[i].ft = esw->fdb_table.offloads.slow_fdb;
+			i++;
+		} else if (attr->dest_chain) {
+			flow_act.flags |= FLOW_ACT_IGNORE_FLOW_LEVEL;
+			ft = mlx5_esw_chains_get_table(esw, attr->dest_chain,
+						       1, 0);
 			if (IS_ERR(ft)) {
 				rule = ERR_CAST(ft);
 				goto err_create_goto_table;
@@ -228,7 +201,8 @@ mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw,
 	if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR)
 		flow_act.modify_hdr = attr->modify_hdr;
 
-	fdb = esw_get_prio_table(esw, attr->chain, attr->prio, !!split);
+	fdb = mlx5_esw_chains_get_table(esw, attr->chain, attr->prio,
+					!!split);
 	if (IS_ERR(fdb)) {
 		rule = ERR_CAST(fdb);
 		goto err_esw_get;
@@ -247,10 +221,10 @@ mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw,
 	return rule;
 
 err_add_rule:
-	esw_put_prio_table(esw, attr->chain, attr->prio, !!split);
+	mlx5_esw_chains_put_table(esw, attr->chain, attr->prio, !!split);
 err_esw_get:
-	if (attr->dest_chain)
-		esw_put_prio_table(esw, attr->dest_chain, 1, 0);
+	if (!(attr->flags & MLX5_ESW_ATTR_FLAG_SLOW_PATH) && attr->dest_chain)
+		mlx5_esw_chains_put_table(esw, attr->dest_chain, 1, 0);
 err_create_goto_table:
 	return rule;
 }
@@ -267,13 +241,13 @@ mlx5_eswitch_add_fwd_rule(struct mlx5_eswitch *esw,
 	struct mlx5_flow_handle *rule;
 	int i;
 
-	fast_fdb = esw_get_prio_table(esw, attr->chain, attr->prio, 0);
+	fast_fdb = mlx5_esw_chains_get_table(esw, attr->chain, attr->prio, 0);
 	if (IS_ERR(fast_fdb)) {
 		rule = ERR_CAST(fast_fdb);
 		goto err_get_fast;
 	}
 
-	fwd_fdb = esw_get_prio_table(esw, attr->chain, attr->prio, 1);
+	fwd_fdb = mlx5_esw_chains_get_table(esw, attr->chain, attr->prio, 1);
 	if (IS_ERR(fwd_fdb)) {
 		rule = ERR_CAST(fwd_fdb);
 		goto err_get_fwd;
@@ -310,9 +284,9 @@ mlx5_eswitch_add_fwd_rule(struct mlx5_eswitch *esw,
 
 	return rule;
 add_err:
-	esw_put_prio_table(esw, attr->chain, attr->prio, 1);
+	mlx5_esw_chains_put_table(esw, attr->chain, attr->prio, 1);
 err_get_fwd:
-	esw_put_prio_table(esw, attr->chain, attr->prio, 0);
+	mlx5_esw_chains_put_table(esw, attr->chain, attr->prio, 0);
 err_get_fast:
 	return rule;
 }
@@ -337,12 +311,13 @@ __mlx5_eswitch_del_rule(struct mlx5_eswitch *esw,
 	atomic64_dec(&esw->offloads.num_flows);
 
 	if (fwd_rule)  {
-		esw_put_prio_table(esw, attr->chain, attr->prio, 1);
-		esw_put_prio_table(esw, attr->chain, attr->prio, 0);
+		mlx5_esw_chains_put_table(esw, attr->chain, attr->prio, 1);
+		mlx5_esw_chains_put_table(esw, attr->chain, attr->prio, 0);
 	} else {
-		esw_put_prio_table(esw, attr->chain, attr->prio, !!split);
+		mlx5_esw_chains_put_table(esw, attr->chain, attr->prio,
+					  !!split);
 		if (attr->dest_chain)
-			esw_put_prio_table(esw, attr->dest_chain, 1, 0);
+			mlx5_esw_chains_put_table(esw, attr->dest_chain, 1, 0);
 	}
 }
 
@@ -456,7 +431,7 @@ int mlx5_eswitch_add_vlan_action(struct mlx5_eswitch *esw,
 	if (err)
 		goto unlock;
 
-	attr->vlan_handled = false;
+	attr->flags &= ~MLX5_ESW_ATTR_FLAG_VLAN_HANDLED;
 
 	vport = esw_vlan_action_get_vport(attr, push, pop);
 
@@ -464,7 +439,7 @@ int mlx5_eswitch_add_vlan_action(struct mlx5_eswitch *esw,
 		/* tracks VF --> wire rules without vlan push action */
 		if (attr->dests[0].rep->vport == MLX5_VPORT_UPLINK) {
 			vport->vlan_refcount++;
-			attr->vlan_handled = true;
+			attr->flags |= MLX5_ESW_ATTR_FLAG_VLAN_HANDLED;
 		}
 
 		goto unlock;
@@ -495,7 +470,7 @@ skip_set_push:
 	}
 out:
 	if (!err)
-		attr->vlan_handled = true;
+		attr->flags |= MLX5_ESW_ATTR_FLAG_VLAN_HANDLED;
 unlock:
 	mutex_unlock(&esw->state_lock);
 	return err;
@@ -513,7 +488,7 @@ int mlx5_eswitch_del_vlan_action(struct mlx5_eswitch *esw,
 	if (mlx5_eswitch_vlan_actions_supported(esw->dev, 1))
 		return 0;
 
-	if (!attr->vlan_handled)
+	if (!(attr->flags & MLX5_ESW_ATTR_FLAG_VLAN_HANDLED))
 		return 0;
 
 	push = !!(attr->action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH);
@@ -587,8 +562,8 @@ mlx5_eswitch_add_send_to_vport_rule(struct mlx5_eswitch *esw, u16 vport,
 	dest.vport.num = vport;
 	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
 
-	flow_rule = mlx5_add_flow_rules(esw->fdb_table.offloads.slow_fdb, spec,
-					&flow_act, &dest, 1);
+	flow_rule = mlx5_add_flow_rules(esw->fdb_table.offloads.slow_fdb,
+					spec, &flow_act, &dest, 1);
 	if (IS_ERR(flow_rule))
 		esw_warn(esw->dev, "FDB: Failed to add send to vport rule err %ld\n", PTR_ERR(flow_rule));
 out:
@@ -829,8 +804,8 @@ static int esw_add_fdb_miss_rule(struct mlx5_eswitch *esw)
 	dest.vport.num = esw->manager_vport;
 	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
 
-	flow_rule = mlx5_add_flow_rules(esw->fdb_table.offloads.slow_fdb, spec,
-					&flow_act, &dest, 1);
+	flow_rule = mlx5_add_flow_rules(esw->fdb_table.offloads.slow_fdb,
+					spec, &flow_act, &dest, 1);
 	if (IS_ERR(flow_rule)) {
 		err = PTR_ERR(flow_rule);
 		esw_warn(esw->dev,  "FDB: Failed to add unicast miss flow rule err %d\n", err);
@@ -844,8 +819,8 @@ static int esw_add_fdb_miss_rule(struct mlx5_eswitch *esw)
 	dmac_v = MLX5_ADDR_OF(fte_match_param, headers_v,
 			      outer_headers.dmac_47_16);
 	dmac_v[0] = 0x01;
-	flow_rule = mlx5_add_flow_rules(esw->fdb_table.offloads.slow_fdb, spec,
-					&flow_act, &dest, 1);
+	flow_rule = mlx5_add_flow_rules(esw->fdb_table.offloads.slow_fdb,
+					spec, &flow_act, &dest, 1);
 	if (IS_ERR(flow_rule)) {
 		err = PTR_ERR(flow_rule);
 		esw_warn(esw->dev, "FDB: Failed to add multicast miss flow rule err %d\n", err);
@@ -860,175 +835,6 @@ out:
 	return err;
 }
 
-#define ESW_OFFLOADS_NUM_GROUPS  4
-
-/* Firmware currently has 4 pool of 4 sizes that it supports (ESW_POOLS),
- * and a virtual memory region of 16M (ESW_SIZE), this region is duplicated
- * for each flow table pool. We can allocate up to 16M of each pool,
- * and we keep track of how much we used via put/get_sz_to_pool.
- * Firmware doesn't report any of this for now.
- * ESW_POOL is expected to be sorted from large to small
- */
-#define ESW_SIZE (16 * 1024 * 1024)
-const unsigned int ESW_POOLS[4] = { 4 * 1024 * 1024, 1 * 1024 * 1024,
-				    64 * 1024, 4 * 1024 };
-
-static int
-get_sz_from_pool(struct mlx5_eswitch *esw)
-{
-	int sz = 0, i;
-
-	for (i = 0; i < ARRAY_SIZE(ESW_POOLS); i++) {
-		if (esw->fdb_table.offloads.fdb_left[i]) {
-			--esw->fdb_table.offloads.fdb_left[i];
-			sz = ESW_POOLS[i];
-			break;
-		}
-	}
-
-	return sz;
-}
-
-static void
-put_sz_to_pool(struct mlx5_eswitch *esw, int sz)
-{
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(ESW_POOLS); i++) {
-		if (sz >= ESW_POOLS[i]) {
-			++esw->fdb_table.offloads.fdb_left[i];
-			break;
-		}
-	}
-}
-
-static struct mlx5_flow_table *
-create_next_size_table(struct mlx5_eswitch *esw,
-		       struct mlx5_flow_namespace *ns,
-		       u16 table_prio,
-		       int level,
-		       u32 flags)
-{
-	struct mlx5_flow_table_attr ft_attr = {};
-	struct mlx5_flow_table *fdb;
-	int sz;
-
-	sz = get_sz_from_pool(esw);
-	if (!sz)
-		return ERR_PTR(-ENOSPC);
-
-	ft_attr.max_fte = sz;
-	ft_attr.prio = table_prio;
-	ft_attr.level = level;
-	ft_attr.flags = flags;
-	ft_attr.autogroup.max_num_groups = ESW_OFFLOADS_NUM_GROUPS;
-	fdb = mlx5_create_auto_grouped_flow_table(ns, &ft_attr);
-	if (IS_ERR(fdb)) {
-		esw_warn(esw->dev, "Failed to create FDB Table err %d (table prio: %d, level: %d, size: %d)\n",
-			 (int)PTR_ERR(fdb), table_prio, level, sz);
-		put_sz_to_pool(esw, sz);
-	}
-
-	return fdb;
-}
-
-static struct mlx5_flow_table *
-esw_get_prio_table(struct mlx5_eswitch *esw, u32 chain, u16 prio, int level)
-{
-	struct mlx5_core_dev *dev = esw->dev;
-	struct mlx5_flow_table *fdb = NULL;
-	struct mlx5_flow_namespace *ns;
-	int table_prio, l = 0;
-	u32 flags = 0;
-
-	if (chain == FDB_TC_SLOW_PATH_CHAIN)
-		return esw->fdb_table.offloads.slow_fdb;
-
-	mutex_lock(&esw->fdb_table.offloads.fdb_prio_lock);
-
-	fdb = fdb_prio_table(esw, chain, prio, level).fdb;
-	if (fdb) {
-		/* take ref on earlier levels as well */
-		while (level >= 0)
-			fdb_prio_table(esw, chain, prio, level--).num_rules++;
-		mutex_unlock(&esw->fdb_table.offloads.fdb_prio_lock);
-		return fdb;
-	}
-
-	ns = mlx5_get_fdb_sub_ns(dev, chain);
-	if (!ns) {
-		esw_warn(dev, "Failed to get FDB sub namespace\n");
-		mutex_unlock(&esw->fdb_table.offloads.fdb_prio_lock);
-		return ERR_PTR(-EOPNOTSUPP);
-	}
-
-	if (esw->offloads.encap != DEVLINK_ESWITCH_ENCAP_MODE_NONE)
-		flags |= (MLX5_FLOW_TABLE_TUNNEL_EN_REFORMAT |
-			  MLX5_FLOW_TABLE_TUNNEL_EN_DECAP);
-
-	table_prio = prio - 1;
-
-	/* create earlier levels for correct fs_core lookup when
-	 * connecting tables
-	 */
-	for (l = 0; l <= level; l++) {
-		if (fdb_prio_table(esw, chain, prio, l).fdb) {
-			fdb_prio_table(esw, chain, prio, l).num_rules++;
-			continue;
-		}
-
-		fdb = create_next_size_table(esw, ns, table_prio, l, flags);
-		if (IS_ERR(fdb)) {
-			l--;
-			goto err_create_fdb;
-		}
-
-		fdb_prio_table(esw, chain, prio, l).fdb = fdb;
-		fdb_prio_table(esw, chain, prio, l).num_rules = 1;
-	}
-
-	mutex_unlock(&esw->fdb_table.offloads.fdb_prio_lock);
-	return fdb;
-
-err_create_fdb:
-	mutex_unlock(&esw->fdb_table.offloads.fdb_prio_lock);
-	if (l >= 0)
-		esw_put_prio_table(esw, chain, prio, l);
-
-	return fdb;
-}
-
-static void
-esw_put_prio_table(struct mlx5_eswitch *esw, u32 chain, u16 prio, int level)
-{
-	int l;
-
-	if (chain == FDB_TC_SLOW_PATH_CHAIN)
-		return;
-
-	mutex_lock(&esw->fdb_table.offloads.fdb_prio_lock);
-
-	for (l = level; l >= 0; l--) {
-		if (--(fdb_prio_table(esw, chain, prio, l).num_rules) > 0)
-			continue;
-
-		put_sz_to_pool(esw, fdb_prio_table(esw, chain, prio, l).fdb->max_fte);
-		mlx5_destroy_flow_table(fdb_prio_table(esw, chain, prio, l).fdb);
-		fdb_prio_table(esw, chain, prio, l).fdb = NULL;
-	}
-
-	mutex_unlock(&esw->fdb_table.offloads.fdb_prio_lock);
-}
-
-static void esw_destroy_offloads_fast_fdb_tables(struct mlx5_eswitch *esw)
-{
-	/* If lazy creation isn't supported, deref the fast path tables */
-	if (!(esw->fdb_table.flags & ESW_FDB_CHAINS_AND_PRIOS_SUPPORTED)) {
-		esw_put_prio_table(esw, 0, 1, 1);
-		esw_put_prio_table(esw, 0, 1, 0);
-	}
-}
-
 #define MAX_PF_SQ 256
 #define MAX_SQ_NVPORTS 32
 
@@ -1061,16 +867,16 @@ static int esw_create_offloads_fdb_tables(struct mlx5_eswitch *esw, int nvports)
 	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
 	struct mlx5_flow_table_attr ft_attr = {};
 	struct mlx5_core_dev *dev = esw->dev;
-	u32 *flow_group_in, max_flow_counter;
 	struct mlx5_flow_namespace *root_ns;
 	struct mlx5_flow_table *fdb = NULL;
-	int table_size, ix, err = 0, i;
+	u32 flags = 0, *flow_group_in;
+	int table_size, ix, err = 0;
 	struct mlx5_flow_group *g;
-	u32 flags = 0, fdb_max;
 	void *match_criteria;
 	u8 *dmac;
 
 	esw_debug(esw->dev, "Create offloads FDB Tables\n");
+
 	flow_group_in = kvzalloc(inlen, GFP_KERNEL);
 	if (!flow_group_in)
 		return -ENOMEM;
@@ -1089,19 +895,6 @@ static int esw_create_offloads_fdb_tables(struct mlx5_eswitch *esw, int nvports)
 		goto ns_err;
 	}
 
-	max_flow_counter = (MLX5_CAP_GEN(dev, max_flow_counter_31_16) << 16) |
-			    MLX5_CAP_GEN(dev, max_flow_counter_15_0);
-	fdb_max = 1 << MLX5_CAP_ESW_FLOWTABLE_FDB(dev, log_max_ft_size);
-
-	esw_debug(dev, "Create offloads FDB table, min (max esw size(2^%d), max counters(%d), groups(%d), max flow table size(%d))\n",
-		  MLX5_CAP_ESW_FLOWTABLE_FDB(dev, log_max_ft_size),
-		  max_flow_counter, ESW_OFFLOADS_NUM_GROUPS,
-		  fdb_max);
-
-	for (i = 0; i < ARRAY_SIZE(ESW_POOLS); i++)
-		esw->fdb_table.offloads.fdb_left[i] =
-			ESW_POOLS[i] <= fdb_max ? ESW_SIZE / ESW_POOLS[i] : 0;
-
 	table_size = nvports * MAX_SQ_NVPORTS + MAX_PF_SQ +
 		MLX5_ESW_MISS_FLOWS + esw->total_vports;
 
@@ -1124,16 +917,10 @@ static int esw_create_offloads_fdb_tables(struct mlx5_eswitch *esw, int nvports)
 	}
 	esw->fdb_table.offloads.slow_fdb = fdb;
 
-	/* If lazy creation isn't supported, open the fast path tables now */
-	if (!MLX5_CAP_ESW_FLOWTABLE(esw->dev, multi_fdb_encap) &&
-	    esw->offloads.encap != DEVLINK_ESWITCH_ENCAP_MODE_NONE) {
-		esw->fdb_table.flags &= ~ESW_FDB_CHAINS_AND_PRIOS_SUPPORTED;
-		esw_warn(dev, "Lazy creation of flow tables isn't supported, ignoring priorities\n");
-		esw_get_prio_table(esw, 0, 1, 0);
-		esw_get_prio_table(esw, 0, 1, 1);
-	} else {
-		esw_debug(dev, "Lazy creation of flow tables supported, deferring table opening\n");
-		esw->fdb_table.flags |= ESW_FDB_CHAINS_AND_PRIOS_SUPPORTED;
+	err = mlx5_esw_chains_create(esw);
+	if (err) {
+		esw_warn(dev, "Failed to create fdb chains err(%d)\n", err);
+		goto fdb_chains_err;
 	}
 
 	/* create send-to-vport group */
@@ -1224,7 +1011,8 @@ miss_err:
 peer_miss_err:
 	mlx5_destroy_flow_group(esw->fdb_table.offloads.send_to_vport_grp);
 send_vport_err:
-	esw_destroy_offloads_fast_fdb_tables(esw);
+	mlx5_esw_chains_destroy(esw);
+fdb_chains_err:
 	mlx5_destroy_flow_table(esw->fdb_table.offloads.slow_fdb);
 slow_fdb_err:
 	/* Holds true only as long as DMFS is the default */
@@ -1246,8 +1034,8 @@ static void esw_destroy_offloads_fdb_tables(struct mlx5_eswitch *esw)
 	mlx5_destroy_flow_group(esw->fdb_table.offloads.peer_miss_grp);
 	mlx5_destroy_flow_group(esw->fdb_table.offloads.miss_grp);
 
+	mlx5_esw_chains_destroy(esw);
 	mlx5_destroy_flow_table(esw->fdb_table.offloads.slow_fdb);
-	esw_destroy_offloads_fast_fdb_tables(esw);
 	/* Holds true only as long as DMFS is the default */
 	mlx5_flow_namespace_set_mode(esw->fdb_table.offloads.ns,
 				     MLX5_FLOW_STEERING_MODE_DMFS);
@@ -2117,7 +1905,6 @@ static int esw_offloads_steering_init(struct mlx5_eswitch *esw)
 		total_vports = num_vfs + MLX5_SPECIAL_VPORTS(esw->dev);
 
 	memset(&esw->fdb_table.offloads, 0, sizeof(struct offloads_fdb));
-	mutex_init(&esw->fdb_table.offloads.fdb_prio_lock);
 
 	err = esw_create_uplink_offloads_acl_tables(esw);
 	if (err)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.c
new file mode 100644
index 000000000000..a694cc52d94c
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.c
@@ -0,0 +1,542 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+// Copyright (c) 2020 Mellanox Technologies.
+
+#include <linux/mlx5/driver.h>
+#include <linux/mlx5/mlx5_ifc.h>
+#include <linux/mlx5/fs.h>
+
+#include "eswitch_offloads_chains.h"
+#include "mlx5_core.h"
+#include "fs_core.h"
+#include "eswitch.h"
+#include "en.h"
+
+#define esw_chains_priv(esw) ((esw)->fdb_table.offloads.esw_chains_priv)
+#define esw_chains_lock(esw) (esw_chains_priv(esw)->lock)
+#define esw_chains_ht(esw) (esw_chains_priv(esw)->chains_ht)
+#define esw_prios_ht(esw) (esw_chains_priv(esw)->prios_ht)
+#define fdb_pool_left(esw) (esw_chains_priv(esw)->fdb_left)
+
+#define ESW_OFFLOADS_NUM_GROUPS  4
+
+/* Firmware currently has 4 pool of 4 sizes that it supports (ESW_POOLS),
+ * and a virtual memory region of 16M (ESW_SIZE), this region is duplicated
+ * for each flow table pool. We can allocate up to 16M of each pool,
+ * and we keep track of how much we used via get_next_avail_sz_from_pool.
+ * Firmware doesn't report any of this for now.
+ * ESW_POOL is expected to be sorted from large to small and match firmware
+ * pools.
+ */
+#define ESW_SIZE (16 * 1024 * 1024)
+const unsigned int ESW_POOLS[] = { 4 * 1024 * 1024,
+				   1 * 1024 * 1024,
+				   64 * 1024,
+				   4 * 1024, };
+
+struct mlx5_esw_chains_priv {
+	struct rhashtable chains_ht;
+	struct rhashtable prios_ht;
+	/* Protects above chains_ht and prios_ht */
+	struct mutex lock;
+
+	int fdb_left[ARRAY_SIZE(ESW_POOLS)];
+};
+
+struct fdb_chain {
+	struct rhash_head node;
+
+	u32 chain;
+
+	int ref;
+
+	struct mlx5_eswitch *esw;
+};
+
+struct fdb_prio_key {
+	u32 chain;
+	u32 prio;
+	u32 level;
+};
+
+struct fdb_prio {
+	struct rhash_head node;
+
+	struct fdb_prio_key key;
+
+	int ref;
+
+	struct fdb_chain *fdb_chain;
+	struct mlx5_flow_table *fdb;
+};
+
+static const struct rhashtable_params chain_params = {
+	.head_offset = offsetof(struct fdb_chain, node),
+	.key_offset = offsetof(struct fdb_chain, chain),
+	.key_len = sizeof_field(struct fdb_chain, chain),
+	.automatic_shrinking = true,
+};
+
+static const struct rhashtable_params prio_params = {
+	.head_offset = offsetof(struct fdb_prio, node),
+	.key_offset = offsetof(struct fdb_prio, key),
+	.key_len = sizeof_field(struct fdb_prio, key),
+	.automatic_shrinking = true,
+};
+
+bool mlx5_esw_chains_prios_supported(struct mlx5_eswitch *esw)
+{
+	return esw->fdb_table.flags & ESW_FDB_CHAINS_AND_PRIOS_SUPPORTED;
+}
+
+u32 mlx5_esw_chains_get_chain_range(struct mlx5_eswitch *esw)
+{
+	if (!mlx5_esw_chains_prios_supported(esw))
+		return 1;
+
+	return FDB_TC_MAX_CHAIN;
+}
+
+u32 mlx5_esw_chains_get_ft_chain(struct mlx5_eswitch *esw)
+{
+	return mlx5_esw_chains_get_chain_range(esw) + 1;
+}
+
+u32 mlx5_esw_chains_get_prio_range(struct mlx5_eswitch *esw)
+{
+	if (!mlx5_esw_chains_prios_supported(esw))
+		return 1;
+
+	return FDB_TC_MAX_PRIO;
+}
+
+static unsigned int mlx5_esw_chains_get_level_range(struct mlx5_eswitch *esw)
+{
+	return FDB_TC_LEVELS_PER_PRIO;
+}
+
+#define POOL_NEXT_SIZE 0
+static int
+mlx5_esw_chains_get_avail_sz_from_pool(struct mlx5_eswitch *esw,
+				       int desired_size)
+{
+	int i, found_i = -1;
+
+	for (i = ARRAY_SIZE(ESW_POOLS) - 1; i >= 0; i--) {
+		if (fdb_pool_left(esw)[i] && ESW_POOLS[i] > desired_size) {
+			found_i = i;
+			if (desired_size != POOL_NEXT_SIZE)
+				break;
+		}
+	}
+
+	if (found_i != -1) {
+		--fdb_pool_left(esw)[found_i];
+		return ESW_POOLS[found_i];
+	}
+
+	return 0;
+}
+
+static void
+mlx5_esw_chains_put_sz_to_pool(struct mlx5_eswitch *esw, int sz)
+{
+	int i;
+
+	for (i = ARRAY_SIZE(ESW_POOLS) - 1; i >= 0; i--) {
+		if (sz == ESW_POOLS[i]) {
+			++fdb_pool_left(esw)[i];
+			return;
+		}
+	}
+
+	WARN_ONCE(1, "Couldn't find size %d in fdb size pool", sz);
+}
+
+static void
+mlx5_esw_chains_init_sz_pool(struct mlx5_eswitch *esw)
+{
+	u32 fdb_max;
+	int i;
+
+	fdb_max = 1 << MLX5_CAP_ESW_FLOWTABLE_FDB(esw->dev, log_max_ft_size);
+
+	for (i = ARRAY_SIZE(ESW_POOLS) - 1; i >= 0; i--)
+		fdb_pool_left(esw)[i] =
+			ESW_POOLS[i] <= fdb_max ? ESW_SIZE / ESW_POOLS[i] : 0;
+}
+
+static struct mlx5_flow_table *
+mlx5_esw_chains_create_fdb_table(struct mlx5_eswitch *esw,
+				 u32 chain, u32 prio, u32 level)
+{
+	struct mlx5_flow_table_attr ft_attr = {};
+	struct mlx5_flow_namespace *ns;
+	struct mlx5_flow_table *fdb;
+	int sz;
+
+	if (esw->offloads.encap != DEVLINK_ESWITCH_ENCAP_MODE_NONE)
+		ft_attr.flags |= (MLX5_FLOW_TABLE_TUNNEL_EN_REFORMAT |
+				  MLX5_FLOW_TABLE_TUNNEL_EN_DECAP);
+
+	sz = mlx5_esw_chains_get_avail_sz_from_pool(esw, POOL_NEXT_SIZE);
+	if (!sz)
+		return ERR_PTR(-ENOSPC);
+
+	ft_attr.max_fte = sz;
+	ft_attr.level = level;
+	ft_attr.prio = prio - 1;
+	ft_attr.autogroup.max_num_groups = ESW_OFFLOADS_NUM_GROUPS;
+	ns = mlx5_get_fdb_sub_ns(esw->dev, chain);
+
+	fdb = mlx5_create_auto_grouped_flow_table(ns, &ft_attr);
+	if (IS_ERR(fdb)) {
+		esw_warn(esw->dev,
+			 "Failed to create FDB table err %d (chain: %d, prio: %d, level: %d, size: %d)\n",
+			 (int)PTR_ERR(fdb), chain, prio, level, sz);
+		mlx5_esw_chains_put_sz_to_pool(esw, sz);
+		return fdb;
+	}
+
+	return fdb;
+}
+
+static void
+mlx5_esw_chains_destroy_fdb_table(struct mlx5_eswitch *esw,
+				  struct mlx5_flow_table *fdb)
+{
+	mlx5_esw_chains_put_sz_to_pool(esw, fdb->max_fte);
+	mlx5_destroy_flow_table(fdb);
+}
+
+static struct fdb_chain *
+mlx5_esw_chains_create_fdb_chain(struct mlx5_eswitch *esw, u32 chain)
+{
+	struct fdb_chain *fdb_chain = NULL;
+	int err;
+
+	fdb_chain = kvzalloc(sizeof(*fdb_chain), GFP_KERNEL);
+	if (!fdb_chain)
+		return ERR_PTR(-ENOMEM);
+
+	fdb_chain->esw = esw;
+	fdb_chain->chain = chain;
+
+	err = rhashtable_insert_fast(&esw_chains_ht(esw), &fdb_chain->node,
+				     chain_params);
+	if (err)
+		goto err_insert;
+
+	return fdb_chain;
+
+err_insert:
+	kvfree(fdb_chain);
+	return ERR_PTR(err);
+}
+
+static void
+mlx5_esw_chains_destroy_fdb_chain(struct fdb_chain *fdb_chain)
+{
+	struct mlx5_eswitch *esw = fdb_chain->esw;
+
+	rhashtable_remove_fast(&esw_chains_ht(esw), &fdb_chain->node,
+			       chain_params);
+	kvfree(fdb_chain);
+}
+
+static struct fdb_chain *
+mlx5_esw_chains_get_fdb_chain(struct mlx5_eswitch *esw, u32 chain)
+{
+	struct fdb_chain *fdb_chain;
+
+	fdb_chain = rhashtable_lookup_fast(&esw_chains_ht(esw), &chain,
+					   chain_params);
+	if (!fdb_chain) {
+		fdb_chain = mlx5_esw_chains_create_fdb_chain(esw, chain);
+		if (IS_ERR(fdb_chain))
+			return fdb_chain;
+	}
+
+	fdb_chain->ref++;
+
+	return fdb_chain;
+}
+
+static void
+mlx5_esw_chains_put_fdb_chain(struct fdb_chain *fdb_chain)
+{
+	if (--fdb_chain->ref == 0)
+		mlx5_esw_chains_destroy_fdb_chain(fdb_chain);
+}
+
+static struct fdb_prio *
+mlx5_esw_chains_create_fdb_prio(struct mlx5_eswitch *esw,
+				u32 chain, u32 prio, u32 level)
+{
+	struct fdb_prio *fdb_prio = NULL;
+	struct fdb_chain *fdb_chain;
+	struct mlx5_flow_table *fdb;
+	int err;
+
+	fdb_chain = mlx5_esw_chains_get_fdb_chain(esw, chain);
+	if (IS_ERR(fdb_chain))
+		return ERR_CAST(fdb_chain);
+
+	fdb_prio = kvzalloc(sizeof(*fdb_prio), GFP_KERNEL);
+	if (!fdb_prio) {
+		err = -ENOMEM;
+		goto err_alloc;
+	}
+
+	fdb = mlx5_esw_chains_create_fdb_table(esw, fdb_chain->chain, prio,
+					       level);
+	if (IS_ERR(fdb)) {
+		err = PTR_ERR(fdb);
+		goto err_create;
+	}
+
+	fdb_prio->fdb_chain = fdb_chain;
+	fdb_prio->key.chain = chain;
+	fdb_prio->key.prio = prio;
+	fdb_prio->key.level = level;
+	fdb_prio->fdb = fdb;
+
+	err = rhashtable_insert_fast(&esw_prios_ht(esw), &fdb_prio->node,
+				     prio_params);
+	if (err)
+		goto err_insert;
+
+	return fdb_prio;
+
+err_insert:
+	mlx5_esw_chains_destroy_fdb_table(esw, fdb);
+err_create:
+	kvfree(fdb_prio);
+err_alloc:
+	mlx5_esw_chains_put_fdb_chain(fdb_chain);
+	return ERR_PTR(err);
+}
+
+static void
+mlx5_esw_chains_destroy_fdb_prio(struct mlx5_eswitch *esw,
+				 struct fdb_prio *fdb_prio)
+{
+	struct fdb_chain *fdb_chain = fdb_prio->fdb_chain;
+
+	rhashtable_remove_fast(&esw_prios_ht(esw), &fdb_prio->node,
+			       prio_params);
+	mlx5_esw_chains_destroy_fdb_table(esw, fdb_prio->fdb);
+	mlx5_esw_chains_put_fdb_chain(fdb_chain);
+	kvfree(fdb_prio);
+}
+
+struct mlx5_flow_table *
+mlx5_esw_chains_get_table(struct mlx5_eswitch *esw, u32 chain, u32 prio,
+			  u32 level)
+{
+	struct mlx5_flow_table *prev_fts;
+	struct fdb_prio *fdb_prio;
+	struct fdb_prio_key key;
+	int l = 0;
+
+	if ((chain > mlx5_esw_chains_get_chain_range(esw) &&
+	     chain != mlx5_esw_chains_get_ft_chain(esw)) ||
+	    prio > mlx5_esw_chains_get_prio_range(esw) ||
+	    level > mlx5_esw_chains_get_level_range(esw))
+		return ERR_PTR(-EOPNOTSUPP);
+
+	/* create earlier levels for correct fs_core lookup when
+	 * connecting tables.
+	 */
+	for (l = 0; l < level; l++) {
+		prev_fts = mlx5_esw_chains_get_table(esw, chain, prio, l);
+		if (IS_ERR(prev_fts)) {
+			fdb_prio = ERR_CAST(prev_fts);
+			goto err_get_prevs;
+		}
+	}
+
+	key.chain = chain;
+	key.prio = prio;
+	key.level = level;
+
+	mutex_lock(&esw_chains_lock(esw));
+	fdb_prio = rhashtable_lookup_fast(&esw_prios_ht(esw), &key,
+					  prio_params);
+	if (!fdb_prio) {
+		fdb_prio = mlx5_esw_chains_create_fdb_prio(esw, chain,
+							   prio, level);
+		if (IS_ERR(fdb_prio))
+			goto err_create_prio;
+	}
+
+	++fdb_prio->ref;
+	mutex_unlock(&esw_chains_lock(esw));
+
+	return fdb_prio->fdb;
+
+err_create_prio:
+	mutex_unlock(&esw_chains_lock(esw));
+err_get_prevs:
+	while (--l >= 0)
+		mlx5_esw_chains_put_table(esw, chain, prio, l);
+	return ERR_CAST(fdb_prio);
+}
+
+void
+mlx5_esw_chains_put_table(struct mlx5_eswitch *esw, u32 chain, u32 prio,
+			  u32 level)
+{
+	struct fdb_prio *fdb_prio;
+	struct fdb_prio_key key;
+
+	key.chain = chain;
+	key.prio = prio;
+	key.level = level;
+
+	mutex_lock(&esw_chains_lock(esw));
+	fdb_prio = rhashtable_lookup_fast(&esw_prios_ht(esw), &key,
+					  prio_params);
+	if (!fdb_prio)
+		goto err_get_prio;
+
+	if (--fdb_prio->ref == 0)
+		mlx5_esw_chains_destroy_fdb_prio(esw, fdb_prio);
+	mutex_unlock(&esw_chains_lock(esw));
+
+	while (level-- > 0)
+		mlx5_esw_chains_put_table(esw, chain, prio, level);
+
+	return;
+
+err_get_prio:
+	mutex_unlock(&esw_chains_lock(esw));
+	WARN_ONCE(1,
+		  "Couldn't find table: (chain: %d prio: %d level: %d)",
+		  chain, prio, level);
+}
+
+static int
+mlx5_esw_chains_init(struct mlx5_eswitch *esw)
+{
+	struct mlx5_esw_chains_priv *chains_priv;
+	struct mlx5_core_dev *dev = esw->dev;
+	u32 max_flow_counter, fdb_max;
+	int err;
+
+	chains_priv = kzalloc(sizeof(*chains_priv), GFP_KERNEL);
+	if (!chains_priv)
+		return -ENOMEM;
+	esw_chains_priv(esw) = chains_priv;
+
+	max_flow_counter = (MLX5_CAP_GEN(dev, max_flow_counter_31_16) << 16) |
+			    MLX5_CAP_GEN(dev, max_flow_counter_15_0);
+	fdb_max = 1 << MLX5_CAP_ESW_FLOWTABLE_FDB(dev, log_max_ft_size);
+
+	esw_debug(dev,
+		  "Init esw offloads chains, max counters(%d), groups(%d), max flow table size(%d)\n",
+		  max_flow_counter, ESW_OFFLOADS_NUM_GROUPS, fdb_max);
+
+	mlx5_esw_chains_init_sz_pool(esw);
+
+	if (!MLX5_CAP_ESW_FLOWTABLE(esw->dev, multi_fdb_encap) &&
+	    esw->offloads.encap != DEVLINK_ESWITCH_ENCAP_MODE_NONE) {
+		esw->fdb_table.flags &= ~ESW_FDB_CHAINS_AND_PRIOS_SUPPORTED;
+		esw_warn(dev, "Tc chains and priorities offload aren't supported, update firmware if needed\n");
+	} else {
+		esw->fdb_table.flags |= ESW_FDB_CHAINS_AND_PRIOS_SUPPORTED;
+		esw_info(dev, "Supported tc offload range - chains: %u, prios: %u\n",
+			 mlx5_esw_chains_get_chain_range(esw),
+			 mlx5_esw_chains_get_prio_range(esw));
+	}
+
+	err = rhashtable_init(&esw_chains_ht(esw), &chain_params);
+	if (err)
+		goto init_chains_ht_err;
+
+	err = rhashtable_init(&esw_prios_ht(esw), &prio_params);
+	if (err)
+		goto init_prios_ht_err;
+
+	mutex_init(&esw_chains_lock(esw));
+
+	return 0;
+
+init_prios_ht_err:
+	rhashtable_destroy(&esw_chains_ht(esw));
+init_chains_ht_err:
+	kfree(chains_priv);
+	return err;
+}
+
+static void
+mlx5_esw_chains_cleanup(struct mlx5_eswitch *esw)
+{
+	mutex_destroy(&esw_chains_lock(esw));
+	rhashtable_destroy(&esw_prios_ht(esw));
+	rhashtable_destroy(&esw_chains_ht(esw));
+
+	kfree(esw_chains_priv(esw));
+}
+
+static int
+mlx5_esw_chains_open(struct mlx5_eswitch *esw)
+{
+	struct mlx5_flow_table *ft;
+	int err;
+
+	/* Always open the root for fast path */
+	ft = mlx5_esw_chains_get_table(esw, 0, 1, 0);
+	if (IS_ERR(ft))
+		return PTR_ERR(ft);
+
+	/* Open level 1 for split rules now if prios isn't supported  */
+	if (!mlx5_esw_chains_prios_supported(esw)) {
+		ft = mlx5_esw_chains_get_table(esw, 0, 1, 1);
+
+		if (IS_ERR(ft)) {
+			err = PTR_ERR(ft);
+			goto level_1_err;
+		}
+	}
+
+	return 0;
+
+level_1_err:
+	mlx5_esw_chains_put_table(esw, 0, 1, 0);
+	return err;
+}
+
+static void
+mlx5_esw_chains_close(struct mlx5_eswitch *esw)
+{
+	if (!mlx5_esw_chains_prios_supported(esw))
+		mlx5_esw_chains_put_table(esw, 0, 1, 1);
+	mlx5_esw_chains_put_table(esw, 0, 1, 0);
+}
+
+int
+mlx5_esw_chains_create(struct mlx5_eswitch *esw)
+{
+	int err;
+
+	err = mlx5_esw_chains_init(esw);
+	if (err)
+		return err;
+
+	err = mlx5_esw_chains_open(esw);
+	if (err)
+		goto err_open;
+
+	return 0;
+
+err_open:
+	mlx5_esw_chains_cleanup(esw);
+	return err;
+}
+
+void
+mlx5_esw_chains_destroy(struct mlx5_eswitch *esw)
+{
+	mlx5_esw_chains_close(esw);
+	mlx5_esw_chains_cleanup(esw);
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.h
new file mode 100644
index 000000000000..52fadacab84d
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.h
@@ -0,0 +1,27 @@
+/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */
+/* Copyright (c) 2020 Mellanox Technologies. */
+
+#ifndef __ML5_ESW_CHAINS_H__
+#define __ML5_ESW_CHAINS_H__
+
+bool
+mlx5_esw_chains_prios_supported(struct mlx5_eswitch *esw);
+u32
+mlx5_esw_chains_get_prio_range(struct mlx5_eswitch *esw);
+u32
+mlx5_esw_chains_get_chain_range(struct mlx5_eswitch *esw);
+u32
+mlx5_esw_chains_get_ft_chain(struct mlx5_eswitch *esw);
+
+struct mlx5_flow_table *
+mlx5_esw_chains_get_table(struct mlx5_eswitch *esw, u32 chain, u32 prio,
+			  u32 level);
+void
+mlx5_esw_chains_put_table(struct mlx5_eswitch *esw, u32 chain, u32 prio,
+			  u32 level);
+
+int mlx5_esw_chains_create(struct mlx5_eswitch *esw);
+void mlx5_esw_chains_destroy(struct mlx5_eswitch *esw);
+
+#endif /* __ML5_ESW_CHAINS_H__ */
+
-- 
cgit v1.2.3-59-g8ed1b


From 278d51f24330718aefd7fe86996a6da66fd345e7 Mon Sep 17 00:00:00 2001
From: Paul Blakey <paulb@mellanox.com>
Date: Wed, 20 Nov 2019 15:06:19 +0200
Subject: net/mlx5: E-Switch, Increase number of chains and priorities

Increase the number of chains and priorities to support
the whole range available in tc.

We use unmanaged tables and ignore flow level to create more
tables than what we declared to fs_core steering, and we manage
the connections between the tables themselves.

To support that we need FW with ignore_flow_level capability.
Otherwise the old behaviour will be used, where we are limited
by the number of levels we declared (4 chains, 16 prios).

Signed-off-by: Paul Blakey <paulb@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Reviewed-by: Oz Shlomo <ozsh@mellanox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../ethernet/mellanox/mlx5/core/eswitch_offloads.c |   3 +-
 .../mellanox/mlx5/core/eswitch_offloads_chains.c   | 238 ++++++++++++++++++++-
 .../mellanox/mlx5/core/eswitch_offloads_chains.h   |   3 +
 3 files changed, 232 insertions(+), 12 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 47f8729197e0..a6d0b62ef234 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -151,7 +151,7 @@ mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw,
 		if (attr->flags & MLX5_ESW_ATTR_FLAG_SLOW_PATH) {
 			flow_act.flags |= FLOW_ACT_IGNORE_FLOW_LEVEL;
 			dest[i].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
-			dest[i].ft = esw->fdb_table.offloads.slow_fdb;
+			dest[i].ft = mlx5_esw_chains_get_tc_end_ft(esw);
 			i++;
 		} else if (attr->dest_chain) {
 			flow_act.flags |= FLOW_ACT_IGNORE_FLOW_LEVEL;
@@ -275,6 +275,7 @@ mlx5_eswitch_add_fwd_rule(struct mlx5_eswitch *esw,
 	if (attr->outer_match_level != MLX5_MATCH_NONE)
 		spec->match_criteria_enable |= MLX5_MATCH_OUTER_HEADERS;
 
+	flow_act.flags |= FLOW_ACT_IGNORE_FLOW_LEVEL;
 	rule = mlx5_add_flow_rules(fast_fdb, spec, &flow_act, dest, i);
 
 	if (IS_ERR(rule))
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.c
index a694cc52d94c..3a60eb5360bd 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.c
@@ -16,6 +16,10 @@
 #define esw_chains_ht(esw) (esw_chains_priv(esw)->chains_ht)
 #define esw_prios_ht(esw) (esw_chains_priv(esw)->prios_ht)
 #define fdb_pool_left(esw) (esw_chains_priv(esw)->fdb_left)
+#define tc_slow_fdb(esw) ((esw)->fdb_table.offloads.slow_fdb)
+#define tc_end_fdb(esw) (esw_chains_priv(esw)->tc_end_fdb)
+#define fdb_ignore_flow_level_supported(esw) \
+	(MLX5_CAP_ESW_FLOWTABLE_FDB((esw)->dev, ignore_flow_level))
 
 #define ESW_OFFLOADS_NUM_GROUPS  4
 
@@ -39,6 +43,8 @@ struct mlx5_esw_chains_priv {
 	/* Protects above chains_ht and prios_ht */
 	struct mutex lock;
 
+	struct mlx5_flow_table *tc_end_fdb;
+
 	int fdb_left[ARRAY_SIZE(ESW_POOLS)];
 };
 
@@ -50,6 +56,7 @@ struct fdb_chain {
 	int ref;
 
 	struct mlx5_eswitch *esw;
+	struct list_head prios_list;
 };
 
 struct fdb_prio_key {
@@ -60,6 +67,7 @@ struct fdb_prio_key {
 
 struct fdb_prio {
 	struct rhash_head node;
+	struct list_head list;
 
 	struct fdb_prio_key key;
 
@@ -67,6 +75,9 @@ struct fdb_prio {
 
 	struct fdb_chain *fdb_chain;
 	struct mlx5_flow_table *fdb;
+	struct mlx5_flow_table *next_fdb;
+	struct mlx5_flow_group *miss_group;
+	struct mlx5_flow_handle *miss_rule;
 };
 
 static const struct rhashtable_params chain_params = {
@@ -93,6 +104,9 @@ u32 mlx5_esw_chains_get_chain_range(struct mlx5_eswitch *esw)
 	if (!mlx5_esw_chains_prios_supported(esw))
 		return 1;
 
+	if (fdb_ignore_flow_level_supported(esw))
+		return UINT_MAX - 1;
+
 	return FDB_TC_MAX_CHAIN;
 }
 
@@ -106,11 +120,17 @@ u32 mlx5_esw_chains_get_prio_range(struct mlx5_eswitch *esw)
 	if (!mlx5_esw_chains_prios_supported(esw))
 		return 1;
 
+	if (fdb_ignore_flow_level_supported(esw))
+		return UINT_MAX;
+
 	return FDB_TC_MAX_PRIO;
 }
 
 static unsigned int mlx5_esw_chains_get_level_range(struct mlx5_eswitch *esw)
 {
+	if (fdb_ignore_flow_level_supported(esw))
+		return UINT_MAX;
+
 	return FDB_TC_LEVELS_PER_PRIO;
 }
 
@@ -181,13 +201,40 @@ mlx5_esw_chains_create_fdb_table(struct mlx5_eswitch *esw,
 	sz = mlx5_esw_chains_get_avail_sz_from_pool(esw, POOL_NEXT_SIZE);
 	if (!sz)
 		return ERR_PTR(-ENOSPC);
-
 	ft_attr.max_fte = sz;
-	ft_attr.level = level;
-	ft_attr.prio = prio - 1;
-	ft_attr.autogroup.max_num_groups = ESW_OFFLOADS_NUM_GROUPS;
-	ns = mlx5_get_fdb_sub_ns(esw->dev, chain);
 
+	/* We use tc_slow_fdb(esw) as the table's next_ft till
+	 * ignore_flow_level is allowed on FT creation and not just for FTEs.
+	 * Instead caller should add an explicit miss rule if needed.
+	 */
+	ft_attr.next_ft = tc_slow_fdb(esw);
+
+	/* The root table(chain 0, prio 1, level 0) is required to be
+	 * connected to the previous prio (FDB_BYPASS_PATH if exists).
+	 * We always create it, as a managed table, in order to align with
+	 * fs_core logic.
+	 */
+	if (!fdb_ignore_flow_level_supported(esw) ||
+	    (chain == 0 && prio == 1 && level == 0)) {
+		ft_attr.level = level;
+		ft_attr.prio = prio - 1;
+		ns = mlx5_get_fdb_sub_ns(esw->dev, chain);
+	} else {
+		ft_attr.flags |= MLX5_FLOW_TABLE_UNMANAGED;
+		ft_attr.prio = FDB_TC_OFFLOAD;
+		/* Firmware doesn't allow us to create another level 0 table,
+		 * so we create all unmanaged tables as level 1.
+		 *
+		 * To connect them, we use explicit miss rules with
+		 * ignore_flow_level. Caller is responsible to create
+		 * these rules (if needed).
+		 */
+		ft_attr.level = 1;
+		ns = mlx5_get_flow_namespace(esw->dev, MLX5_FLOW_NAMESPACE_FDB);
+	}
+
+	ft_attr.autogroup.num_reserved_entries = 2;
+	ft_attr.autogroup.max_num_groups = ESW_OFFLOADS_NUM_GROUPS;
 	fdb = mlx5_create_auto_grouped_flow_table(ns, &ft_attr);
 	if (IS_ERR(fdb)) {
 		esw_warn(esw->dev,
@@ -220,6 +267,7 @@ mlx5_esw_chains_create_fdb_chain(struct mlx5_eswitch *esw, u32 chain)
 
 	fdb_chain->esw = esw;
 	fdb_chain->chain = chain;
+	INIT_LIST_HEAD(&fdb_chain->prios_list);
 
 	err = rhashtable_insert_fast(&esw_chains_ht(esw), &fdb_chain->node,
 				     chain_params);
@@ -261,6 +309,79 @@ mlx5_esw_chains_get_fdb_chain(struct mlx5_eswitch *esw, u32 chain)
 	return fdb_chain;
 }
 
+static struct mlx5_flow_handle *
+mlx5_esw_chains_add_miss_rule(struct mlx5_flow_table *fdb,
+			      struct mlx5_flow_table *next_fdb)
+{
+	static const struct mlx5_flow_spec spec = {};
+	struct mlx5_flow_destination dest = {};
+	struct mlx5_flow_act act = {};
+
+	act.flags  = FLOW_ACT_IGNORE_FLOW_LEVEL | FLOW_ACT_NO_APPEND;
+	act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
+	dest.type  = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
+	dest.ft = next_fdb;
+
+	return mlx5_add_flow_rules(fdb, &spec, &act, &dest, 1);
+}
+
+static int
+mlx5_esw_chains_update_prio_prevs(struct fdb_prio *fdb_prio,
+				  struct mlx5_flow_table *next_fdb)
+{
+	struct mlx5_flow_handle *miss_rules[FDB_TC_LEVELS_PER_PRIO + 1] = {};
+	struct fdb_chain *fdb_chain = fdb_prio->fdb_chain;
+	struct fdb_prio *pos;
+	int n = 0, err;
+
+	if (fdb_prio->key.level)
+		return 0;
+
+	/* Iterate in reverse order until reaching the level 0 rule of
+	 * the previous priority, adding all the miss rules first, so we can
+	 * revert them if any of them fails.
+	 */
+	pos = fdb_prio;
+	list_for_each_entry_continue_reverse(pos,
+					     &fdb_chain->prios_list,
+					     list) {
+		miss_rules[n] = mlx5_esw_chains_add_miss_rule(pos->fdb,
+							      next_fdb);
+		if (IS_ERR(miss_rules[n])) {
+			err = PTR_ERR(miss_rules[n]);
+			goto err_prev_rule;
+		}
+
+		n++;
+		if (!pos->key.level)
+			break;
+	}
+
+	/* Success, delete old miss rules, and update the pointers. */
+	n = 0;
+	pos = fdb_prio;
+	list_for_each_entry_continue_reverse(pos,
+					     &fdb_chain->prios_list,
+					     list) {
+		mlx5_del_flow_rules(pos->miss_rule);
+
+		pos->miss_rule = miss_rules[n];
+		pos->next_fdb = next_fdb;
+
+		n++;
+		if (!pos->key.level)
+			break;
+	}
+
+	return 0;
+
+err_prev_rule:
+	while (--n >= 0)
+		mlx5_del_flow_rules(miss_rules[n]);
+
+	return err;
+}
+
 static void
 mlx5_esw_chains_put_fdb_chain(struct fdb_chain *fdb_chain)
 {
@@ -272,9 +393,15 @@ static struct fdb_prio *
 mlx5_esw_chains_create_fdb_prio(struct mlx5_eswitch *esw,
 				u32 chain, u32 prio, u32 level)
 {
+	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
+	struct mlx5_flow_handle *miss_rule = NULL;
+	struct mlx5_flow_group *miss_group;
 	struct fdb_prio *fdb_prio = NULL;
+	struct mlx5_flow_table *next_fdb;
 	struct fdb_chain *fdb_chain;
 	struct mlx5_flow_table *fdb;
+	struct list_head *pos;
+	u32 *flow_group_in;
 	int err;
 
 	fdb_chain = mlx5_esw_chains_get_fdb_chain(esw, chain);
@@ -282,18 +409,65 @@ mlx5_esw_chains_create_fdb_prio(struct mlx5_eswitch *esw,
 		return ERR_CAST(fdb_chain);
 
 	fdb_prio = kvzalloc(sizeof(*fdb_prio), GFP_KERNEL);
-	if (!fdb_prio) {
+	flow_group_in = kvzalloc(inlen, GFP_KERNEL);
+	if (!fdb_prio || !flow_group_in) {
 		err = -ENOMEM;
 		goto err_alloc;
 	}
 
-	fdb = mlx5_esw_chains_create_fdb_table(esw, fdb_chain->chain, prio,
-					       level);
+	/* Chain's prio list is sorted by prio and level.
+	 * And all levels of some prio point to the next prio's level 0.
+	 * Example list (prio, level):
+	 * (3,0)->(3,1)->(5,0)->(5,1)->(6,1)->(7,0)
+	 * In hardware, we will we have the following pointers:
+	 * (3,0) -> (5,0) -> (7,0) -> Slow path
+	 * (3,1) -> (5,0)
+	 * (5,1) -> (7,0)
+	 * (6,1) -> (7,0)
+	 */
+
+	/* Default miss for each chain: */
+	next_fdb = (chain == mlx5_esw_chains_get_ft_chain(esw)) ?
+		    tc_slow_fdb(esw) :
+		    tc_end_fdb(esw);
+	list_for_each(pos, &fdb_chain->prios_list) {
+		struct fdb_prio *p = list_entry(pos, struct fdb_prio, list);
+
+		/* exit on first pos that is larger */
+		if (prio < p->key.prio || (prio == p->key.prio &&
+					   level < p->key.level)) {
+			/* Get next level 0 table */
+			next_fdb = p->key.level == 0 ? p->fdb : p->next_fdb;
+			break;
+		}
+	}
+
+	fdb = mlx5_esw_chains_create_fdb_table(esw, chain, prio, level);
 	if (IS_ERR(fdb)) {
 		err = PTR_ERR(fdb);
 		goto err_create;
 	}
 
+	MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index,
+		 fdb->max_fte - 2);
+	MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index,
+		 fdb->max_fte - 1);
+	miss_group = mlx5_create_flow_group(fdb, flow_group_in);
+	if (IS_ERR(miss_group)) {
+		err = PTR_ERR(miss_group);
+		goto err_group;
+	}
+
+	/* Add miss rule to next_fdb */
+	miss_rule = mlx5_esw_chains_add_miss_rule(fdb, next_fdb);
+	if (IS_ERR(miss_rule)) {
+		err = PTR_ERR(miss_rule);
+		goto err_miss_rule;
+	}
+
+	fdb_prio->miss_group = miss_group;
+	fdb_prio->miss_rule = miss_rule;
+	fdb_prio->next_fdb = next_fdb;
 	fdb_prio->fdb_chain = fdb_chain;
 	fdb_prio->key.chain = chain;
 	fdb_prio->key.prio = prio;
@@ -305,13 +479,30 @@ mlx5_esw_chains_create_fdb_prio(struct mlx5_eswitch *esw,
 	if (err)
 		goto err_insert;
 
+	list_add(&fdb_prio->list, pos->prev);
+
+	/* Table is ready, connect it */
+	err = mlx5_esw_chains_update_prio_prevs(fdb_prio, fdb);
+	if (err)
+		goto err_update;
+
+	kvfree(flow_group_in);
 	return fdb_prio;
 
+err_update:
+	list_del(&fdb_prio->list);
+	rhashtable_remove_fast(&esw_prios_ht(esw), &fdb_prio->node,
+			       prio_params);
 err_insert:
+	mlx5_del_flow_rules(miss_rule);
+err_miss_rule:
+	mlx5_destroy_flow_group(miss_group);
+err_group:
 	mlx5_esw_chains_destroy_fdb_table(esw, fdb);
 err_create:
-	kvfree(fdb_prio);
 err_alloc:
+	kvfree(fdb_prio);
+	kvfree(flow_group_in);
 	mlx5_esw_chains_put_fdb_chain(fdb_chain);
 	return ERR_PTR(err);
 }
@@ -322,8 +513,14 @@ mlx5_esw_chains_destroy_fdb_prio(struct mlx5_eswitch *esw,
 {
 	struct fdb_chain *fdb_chain = fdb_prio->fdb_chain;
 
+	WARN_ON(mlx5_esw_chains_update_prio_prevs(fdb_prio,
+						  fdb_prio->next_fdb));
+
+	list_del(&fdb_prio->list);
 	rhashtable_remove_fast(&esw_prios_ht(esw), &fdb_prio->node,
 			       prio_params);
+	mlx5_del_flow_rules(fdb_prio->miss_rule);
+	mlx5_destroy_flow_group(fdb_prio->miss_group);
 	mlx5_esw_chains_destroy_fdb_table(esw, fdb_prio->fdb);
 	mlx5_esw_chains_put_fdb_chain(fdb_chain);
 	kvfree(fdb_prio);
@@ -415,6 +612,12 @@ err_get_prio:
 		  chain, prio, level);
 }
 
+struct mlx5_flow_table *
+mlx5_esw_chains_get_tc_end_ft(struct mlx5_eswitch *esw)
+{
+	return tc_end_fdb(esw);
+}
+
 static int
 mlx5_esw_chains_init(struct mlx5_eswitch *esw)
 {
@@ -484,11 +687,21 @@ mlx5_esw_chains_open(struct mlx5_eswitch *esw)
 	struct mlx5_flow_table *ft;
 	int err;
 
-	/* Always open the root for fast path */
-	ft = mlx5_esw_chains_get_table(esw, 0, 1, 0);
+	/* Create tc_end_fdb(esw) which is the always created ft chain */
+	ft = mlx5_esw_chains_get_table(esw, mlx5_esw_chains_get_ft_chain(esw),
+				       1, 0);
 	if (IS_ERR(ft))
 		return PTR_ERR(ft);
 
+	tc_end_fdb(esw) = ft;
+
+	/* Always open the root for fast path */
+	ft = mlx5_esw_chains_get_table(esw, 0, 1, 0);
+	if (IS_ERR(ft)) {
+		err = PTR_ERR(ft);
+		goto level_0_err;
+	}
+
 	/* Open level 1 for split rules now if prios isn't supported  */
 	if (!mlx5_esw_chains_prios_supported(esw)) {
 		ft = mlx5_esw_chains_get_table(esw, 0, 1, 1);
@@ -503,6 +716,8 @@ mlx5_esw_chains_open(struct mlx5_eswitch *esw)
 
 level_1_err:
 	mlx5_esw_chains_put_table(esw, 0, 1, 0);
+level_0_err:
+	mlx5_esw_chains_put_table(esw, mlx5_esw_chains_get_ft_chain(esw), 1, 0);
 	return err;
 }
 
@@ -512,6 +727,7 @@ mlx5_esw_chains_close(struct mlx5_eswitch *esw)
 	if (!mlx5_esw_chains_prios_supported(esw))
 		mlx5_esw_chains_put_table(esw, 0, 1, 1);
 	mlx5_esw_chains_put_table(esw, 0, 1, 0);
+	mlx5_esw_chains_put_table(esw, mlx5_esw_chains_get_ft_chain(esw), 1, 0);
 }
 
 int
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.h
index 52fadacab84d..2e13097fe348 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.h
@@ -20,6 +20,9 @@ void
 mlx5_esw_chains_put_table(struct mlx5_eswitch *esw, u32 chain, u32 prio,
 			  u32 level);
 
+struct mlx5_flow_table *
+mlx5_esw_chains_get_tc_end_ft(struct mlx5_eswitch *esw);
+
 int mlx5_esw_chains_create(struct mlx5_eswitch *esw);
 void mlx5_esw_chains_destroy(struct mlx5_eswitch *esw);
 
-- 
cgit v1.2.3-59-g8ed1b


From 20f21d98cf12b8ecd69e8defc93fae9e3b353b13 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Thu, 16 Jan 2020 16:41:02 -0800
Subject: libbpf: Revert bpf_helper_defs.h inclusion regression
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Revert bpf_helpers.h's change to include auto-generated bpf_helper_defs.h
through <> instead of "", which causes it to be searched in include path. This
can break existing applications that don't have their include path pointing
directly to where libbpf installs its headers.

There is ongoing work to make all (not just bpf_helper_defs.h) includes more
consistent across libbpf and its consumers, but this unbreaks user code as is
right now without any regressions. Selftests still behave sub-optimally
(taking bpf_helper_defs.h from libbpf's source directory, if it's present
there), which will be fixed in subsequent patches.

Fixes: 6910d7d3867a ("selftests/bpf: Ensure bpf_helper_defs.h are taken from selftests dir")
Reported-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200117004103.148068-1-andriin@fb.com
---
 tools/lib/bpf/bpf_helpers.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/lib/bpf/bpf_helpers.h b/tools/lib/bpf/bpf_helpers.h
index 050bb7bf5be6..f69cc208778a 100644
--- a/tools/lib/bpf/bpf_helpers.h
+++ b/tools/lib/bpf/bpf_helpers.h
@@ -2,7 +2,7 @@
 #ifndef __BPF_HELPERS__
 #define __BPF_HELPERS__
 
-#include <bpf_helper_defs.h>
+#include "bpf_helper_defs.h"
 
 #define __uint(name, val) int (*name)[val]
 #define __type(name, val) typeof(val) *name
-- 
cgit v1.2.3-59-g8ed1b


From 75ccae62cb8d42a619323a85c577107b8b37d797 Mon Sep 17 00:00:00 2001
From: Toke Høiland-Jørgensen <toke@redhat.com>
Date: Thu, 16 Jan 2020 16:14:44 +0100
Subject: xdp: Move devmap bulk queue into struct net_device
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Commit 96360004b862 ("xdp: Make devmap flush_list common for all map
instances"), changed devmap flushing to be a global operation instead of a
per-map operation. However, the queue structure used for bulking was still
allocated as part of the containing map.

This patch moves the devmap bulk queue into struct net_device. The
motivation for this is reusing it for the non-map variant of XDP_REDIRECT,
which will be changed in a subsequent commit.  To avoid other fields of
struct net_device moving to different cache lines, we also move a couple of
other members around.

We defer the actual allocation of the bulk queue structure until the
NETDEV_REGISTER notification devmap.c. This makes it possible to check for
ndo_xdp_xmit support before allocating the structure, which is not possible
at the time struct net_device is allocated. However, we keep the freeing in
free_netdev() to avoid adding another RCU callback on NETDEV_UNREGISTER.

Because of this change, we lose the reference back to the map that
originated the redirect, so change the tracepoint to always return 0 as the
map ID and index. Otherwise no functional change is intended with this
patch.

After this patch, the relevant part of struct net_device looks like this,
according to pahole:

	/* --- cacheline 14 boundary (896 bytes) --- */
	struct netdev_queue *      _tx __attribute__((__aligned__(64))); /*   896     8 */
	unsigned int               num_tx_queues;        /*   904     4 */
	unsigned int               real_num_tx_queues;   /*   908     4 */
	struct Qdisc *             qdisc;                /*   912     8 */
	unsigned int               tx_queue_len;         /*   920     4 */
	spinlock_t                 tx_global_lock;       /*   924     4 */
	struct xdp_dev_bulk_queue * xdp_bulkq;           /*   928     8 */
	struct xps_dev_maps *      xps_cpus_map;         /*   936     8 */
	struct xps_dev_maps *      xps_rxqs_map;         /*   944     8 */
	struct mini_Qdisc *        miniq_egress;         /*   952     8 */
	/* --- cacheline 15 boundary (960 bytes) --- */
	struct hlist_head  qdisc_hash[16];               /*   960   128 */
	/* --- cacheline 17 boundary (1088 bytes) --- */
	struct timer_list  watchdog_timer;               /*  1088    40 */

	/* XXX last struct has 4 bytes of padding */

	int                        watchdog_timeo;       /*  1128     4 */

	/* XXX 4 bytes hole, try to pack */

	struct list_head   todo_list;                    /*  1136    16 */
	/* --- cacheline 18 boundary (1152 bytes) --- */

Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Björn Töpel <bjorn.topel@intel.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/157918768397.1458396.12673224324627072349.stgit@toke.dk
---
 include/linux/netdevice.h  | 13 ++++++----
 include/trace/events/xdp.h |  2 +-
 kernel/bpf/devmap.c        | 63 ++++++++++++++++++++--------------------------
 net/core/dev.c             |  2 ++
 4 files changed, 38 insertions(+), 42 deletions(-)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 2741aa35bec6..5ec3537fbdb1 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -876,6 +876,7 @@ enum bpf_netdev_command {
 struct bpf_prog_offload_ops;
 struct netlink_ext_ack;
 struct xdp_umem;
+struct xdp_dev_bulk_queue;
 
 struct netdev_bpf {
 	enum bpf_netdev_command command;
@@ -1986,12 +1987,10 @@ struct net_device {
 	unsigned int		num_tx_queues;
 	unsigned int		real_num_tx_queues;
 	struct Qdisc		*qdisc;
-#ifdef CONFIG_NET_SCHED
-	DECLARE_HASHTABLE	(qdisc_hash, 4);
-#endif
 	unsigned int		tx_queue_len;
 	spinlock_t		tx_global_lock;
-	int			watchdog_timeo;
+
+	struct xdp_dev_bulk_queue __percpu *xdp_bulkq;
 
 #ifdef CONFIG_XPS
 	struct xps_dev_maps __rcu *xps_cpus_map;
@@ -2001,11 +2000,15 @@ struct net_device {
 	struct mini_Qdisc __rcu	*miniq_egress;
 #endif
 
+#ifdef CONFIG_NET_SCHED
+	DECLARE_HASHTABLE	(qdisc_hash, 4);
+#endif
 	/* These may be needed for future network-power-down code. */
 	struct timer_list	watchdog_timer;
+	int			watchdog_timeo;
 
-	int __percpu		*pcpu_refcnt;
 	struct list_head	todo_list;
+	int __percpu		*pcpu_refcnt;
 
 	struct list_head	link_watch_list;
 
diff --git a/include/trace/events/xdp.h b/include/trace/events/xdp.h
index a7378bcd9928..72bad13d4a3c 100644
--- a/include/trace/events/xdp.h
+++ b/include/trace/events/xdp.h
@@ -278,7 +278,7 @@ TRACE_EVENT(xdp_devmap_xmit,
 	),
 
 	TP_fast_assign(
-		__entry->map_id		= map->id;
+		__entry->map_id		= map ? map->id : 0;
 		__entry->act		= XDP_REDIRECT;
 		__entry->map_index	= map_index;
 		__entry->drops		= drops;
diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index da9c832fc5c8..030d125c3839 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -53,13 +53,11 @@
 	(BPF_F_NUMA_NODE | BPF_F_RDONLY | BPF_F_WRONLY)
 
 #define DEV_MAP_BULK_SIZE 16
-struct bpf_dtab_netdev;
-
-struct xdp_bulk_queue {
+struct xdp_dev_bulk_queue {
 	struct xdp_frame *q[DEV_MAP_BULK_SIZE];
 	struct list_head flush_node;
+	struct net_device *dev;
 	struct net_device *dev_rx;
-	struct bpf_dtab_netdev *obj;
 	unsigned int count;
 };
 
@@ -67,9 +65,8 @@ struct bpf_dtab_netdev {
 	struct net_device *dev; /* must be first member, due to tracepoint */
 	struct hlist_node index_hlist;
 	struct bpf_dtab *dtab;
-	struct xdp_bulk_queue __percpu *bulkq;
 	struct rcu_head rcu;
-	unsigned int idx; /* keep track of map index for tracepoint */
+	unsigned int idx;
 };
 
 struct bpf_dtab {
@@ -219,7 +216,6 @@ static void dev_map_free(struct bpf_map *map)
 
 			hlist_for_each_entry_safe(dev, next, head, index_hlist) {
 				hlist_del_rcu(&dev->index_hlist);
-				free_percpu(dev->bulkq);
 				dev_put(dev->dev);
 				kfree(dev);
 			}
@@ -234,7 +230,6 @@ static void dev_map_free(struct bpf_map *map)
 			if (!dev)
 				continue;
 
-			free_percpu(dev->bulkq);
 			dev_put(dev->dev);
 			kfree(dev);
 		}
@@ -320,10 +315,9 @@ static int dev_map_hash_get_next_key(struct bpf_map *map, void *key,
 	return -ENOENT;
 }
 
-static int bq_xmit_all(struct xdp_bulk_queue *bq, u32 flags)
+static int bq_xmit_all(struct xdp_dev_bulk_queue *bq, u32 flags)
 {
-	struct bpf_dtab_netdev *obj = bq->obj;
-	struct net_device *dev = obj->dev;
+	struct net_device *dev = bq->dev;
 	int sent = 0, drops = 0, err = 0;
 	int i;
 
@@ -346,8 +340,7 @@ static int bq_xmit_all(struct xdp_bulk_queue *bq, u32 flags)
 out:
 	bq->count = 0;
 
-	trace_xdp_devmap_xmit(&obj->dtab->map, obj->idx,
-			      sent, drops, bq->dev_rx, dev, err);
+	trace_xdp_devmap_xmit(NULL, 0, sent, drops, bq->dev_rx, dev, err);
 	bq->dev_rx = NULL;
 	__list_del_clearprev(&bq->flush_node);
 	return 0;
@@ -374,7 +367,7 @@ error:
 void __dev_map_flush(void)
 {
 	struct list_head *flush_list = this_cpu_ptr(&dev_map_flush_list);
-	struct xdp_bulk_queue *bq, *tmp;
+	struct xdp_dev_bulk_queue *bq, *tmp;
 
 	rcu_read_lock();
 	list_for_each_entry_safe(bq, tmp, flush_list, flush_node)
@@ -401,12 +394,12 @@ struct bpf_dtab_netdev *__dev_map_lookup_elem(struct bpf_map *map, u32 key)
 /* Runs under RCU-read-side, plus in softirq under NAPI protection.
  * Thus, safe percpu variable access.
  */
-static int bq_enqueue(struct bpf_dtab_netdev *obj, struct xdp_frame *xdpf,
+static int bq_enqueue(struct net_device *dev, struct xdp_frame *xdpf,
 		      struct net_device *dev_rx)
 
 {
 	struct list_head *flush_list = this_cpu_ptr(&dev_map_flush_list);
-	struct xdp_bulk_queue *bq = this_cpu_ptr(obj->bulkq);
+	struct xdp_dev_bulk_queue *bq = this_cpu_ptr(dev->xdp_bulkq);
 
 	if (unlikely(bq->count == DEV_MAP_BULK_SIZE))
 		bq_xmit_all(bq, 0);
@@ -444,7 +437,7 @@ int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
 	if (unlikely(!xdpf))
 		return -EOVERFLOW;
 
-	return bq_enqueue(dst, xdpf, dev_rx);
+	return bq_enqueue(dev, xdpf, dev_rx);
 }
 
 int dev_map_generic_redirect(struct bpf_dtab_netdev *dst, struct sk_buff *skb,
@@ -483,7 +476,6 @@ static void __dev_map_entry_free(struct rcu_head *rcu)
 	struct bpf_dtab_netdev *dev;
 
 	dev = container_of(rcu, struct bpf_dtab_netdev, rcu);
-	free_percpu(dev->bulkq);
 	dev_put(dev->dev);
 	kfree(dev);
 }
@@ -538,30 +530,15 @@ static struct bpf_dtab_netdev *__dev_map_alloc_node(struct net *net,
 						    u32 ifindex,
 						    unsigned int idx)
 {
-	gfp_t gfp = GFP_ATOMIC | __GFP_NOWARN;
 	struct bpf_dtab_netdev *dev;
-	struct xdp_bulk_queue *bq;
-	int cpu;
 
-	dev = kmalloc_node(sizeof(*dev), gfp, dtab->map.numa_node);
+	dev = kmalloc_node(sizeof(*dev), GFP_ATOMIC | __GFP_NOWARN,
+			   dtab->map.numa_node);
 	if (!dev)
 		return ERR_PTR(-ENOMEM);
 
-	dev->bulkq = __alloc_percpu_gfp(sizeof(*dev->bulkq),
-					sizeof(void *), gfp);
-	if (!dev->bulkq) {
-		kfree(dev);
-		return ERR_PTR(-ENOMEM);
-	}
-
-	for_each_possible_cpu(cpu) {
-		bq = per_cpu_ptr(dev->bulkq, cpu);
-		bq->obj = dev;
-	}
-
 	dev->dev = dev_get_by_index(net, ifindex);
 	if (!dev->dev) {
-		free_percpu(dev->bulkq);
 		kfree(dev);
 		return ERR_PTR(-EINVAL);
 	}
@@ -721,9 +698,23 @@ static int dev_map_notification(struct notifier_block *notifier,
 {
 	struct net_device *netdev = netdev_notifier_info_to_dev(ptr);
 	struct bpf_dtab *dtab;
-	int i;
+	int i, cpu;
 
 	switch (event) {
+	case NETDEV_REGISTER:
+		if (!netdev->netdev_ops->ndo_xdp_xmit || netdev->xdp_bulkq)
+			break;
+
+		/* will be freed in free_netdev() */
+		netdev->xdp_bulkq =
+			__alloc_percpu_gfp(sizeof(struct xdp_dev_bulk_queue),
+					   sizeof(void *), GFP_ATOMIC);
+		if (!netdev->xdp_bulkq)
+			return NOTIFY_BAD;
+
+		for_each_possible_cpu(cpu)
+			per_cpu_ptr(netdev->xdp_bulkq, cpu)->dev = netdev;
+		break;
 	case NETDEV_UNREGISTER:
 		/* This rcu_read_lock/unlock pair is needed because
 		 * dev_map_list is an RCU list AND to ensure a delete
diff --git a/net/core/dev.c b/net/core/dev.c
index d99f88c58636..e7802a41ae7f 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -9847,6 +9847,8 @@ void free_netdev(struct net_device *dev)
 
 	free_percpu(dev->pcpu_refcnt);
 	dev->pcpu_refcnt = NULL;
+	free_percpu(dev->xdp_bulkq);
+	dev->xdp_bulkq = NULL;
 
 	netdev_unregister_lockdep_key(dev);
 
-- 
cgit v1.2.3-59-g8ed1b


From 1d233886dd904edbf239eeffe435c3308ae97625 Mon Sep 17 00:00:00 2001
From: Toke Høiland-Jørgensen <toke@redhat.com>
Date: Thu, 16 Jan 2020 16:14:45 +0100
Subject: xdp: Use bulking for non-map XDP_REDIRECT and consolidate code paths
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Since the bulk queue used by XDP_REDIRECT now lives in struct net_device,
we can re-use the bulking for the non-map version of the bpf_redirect()
helper. This is a simple matter of having xdp_do_redirect_slow() queue the
frame on the bulk queue instead of sending it out with __bpf_tx_xdp().

Unfortunately we can't make the bpf_redirect() helper return an error if
the ifindex doesn't exit (as bpf_redirect_map() does), because we don't
have a reference to the network namespace of the ingress device at the time
the helper is called. So we have to leave it as-is and keep the device
lookup in xdp_do_redirect_slow().

Since this leaves less reason to have the non-map redirect code in a
separate function, so we get rid of the xdp_do_redirect_slow() function
entirely. This does lose us the tracepoint disambiguation, but fortunately
the xdp_redirect and xdp_redirect_map tracepoints use the same tracepoint
entry structures. This means both can contain a map index, so we can just
amend the tracepoint definitions so we always emit the xdp_redirect(_err)
tracepoints, but with the map ID only populated if a map is present. This
means we retire the xdp_redirect_map(_err) tracepoints entirely, but keep
the definitions around in case someone is still listening for them.

With this change, the performance of the xdp_redirect sample program goes
from 5Mpps to 8.4Mpps (a 68% increase).

Since the flush functions are no longer map-specific, rename the flush()
functions to drop _map from their names. One of the renamed functions is
the xdp_do_flush_map() callback used in all the xdp-enabled drivers. To
keep from having to update all drivers, use a #define to keep the old name
working, and only update the virtual drivers in this patch.

Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/157918768505.1458396.17518057312953572912.stgit@toke.dk
---
 drivers/net/tun.c          |   4 +-
 drivers/net/veth.c         |   2 +-
 drivers/net/virtio_net.c   |   2 +-
 include/linux/bpf.h        |  13 +++++-
 include/linux/filter.h     |  10 ++++-
 include/trace/events/xdp.h | 101 ++++++++++++++++++++-------------------------
 kernel/bpf/devmap.c        |  32 +++++++++-----
 net/core/filter.c          |  90 +++++++++-------------------------------
 8 files changed, 108 insertions(+), 146 deletions(-)

diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index 683d371e6e82..3a5a6c655dda 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -1718,7 +1718,7 @@ static struct sk_buff *tun_build_skb(struct tun_struct *tun,
 		if (err < 0)
 			goto err_xdp;
 		if (err == XDP_REDIRECT)
-			xdp_do_flush_map();
+			xdp_do_flush();
 		if (err != XDP_PASS)
 			goto out;
 
@@ -2549,7 +2549,7 @@ static int tun_sendmsg(struct socket *sock, struct msghdr *m, size_t total_len)
 		}
 
 		if (flush)
-			xdp_do_flush_map();
+			xdp_do_flush();
 
 		rcu_read_unlock();
 		local_bh_enable();
diff --git a/drivers/net/veth.c b/drivers/net/veth.c
index a552df37a347..1c89017beebb 100644
--- a/drivers/net/veth.c
+++ b/drivers/net/veth.c
@@ -769,7 +769,7 @@ static int veth_poll(struct napi_struct *napi, int budget)
 	if (xdp_xmit & VETH_XDP_TX)
 		veth_xdp_flush(rq->dev, &bq);
 	if (xdp_xmit & VETH_XDP_REDIR)
-		xdp_do_flush_map();
+		xdp_do_flush();
 	xdp_clear_return_frame_no_direct();
 
 	return done;
diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 4d7d5434cc5d..c458cd313281 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -1432,7 +1432,7 @@ static int virtnet_poll(struct napi_struct *napi, int budget)
 		virtqueue_napi_complete(napi, rq->vq, received);
 
 	if (xdp_xmit & VIRTIO_XDP_REDIR)
-		xdp_do_flush_map();
+		xdp_do_flush();
 
 	if (xdp_xmit & VIRTIO_XDP_TX) {
 		sq = virtnet_xdp_sq(vi);
diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 3517e32149a4..8e3b8f4ad183 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -1056,7 +1056,9 @@ struct sk_buff;
 
 struct bpf_dtab_netdev *__dev_map_lookup_elem(struct bpf_map *map, u32 key);
 struct bpf_dtab_netdev *__dev_map_hash_lookup_elem(struct bpf_map *map, u32 key);
-void __dev_map_flush(void);
+void __dev_flush(void);
+int dev_xdp_enqueue(struct net_device *dev, struct xdp_buff *xdp,
+		    struct net_device *dev_rx);
 int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
 		    struct net_device *dev_rx);
 int dev_map_generic_redirect(struct bpf_dtab_netdev *dst, struct sk_buff *skb,
@@ -1169,13 +1171,20 @@ static inline struct net_device  *__dev_map_hash_lookup_elem(struct bpf_map *map
 	return NULL;
 }
 
-static inline void __dev_map_flush(void)
+static inline void __dev_flush(void)
 {
 }
 
 struct xdp_buff;
 struct bpf_dtab_netdev;
 
+static inline
+int dev_xdp_enqueue(struct net_device *dev, struct xdp_buff *xdp,
+		    struct net_device *dev_rx)
+{
+	return 0;
+}
+
 static inline
 int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
 		    struct net_device *dev_rx)
diff --git a/include/linux/filter.h b/include/linux/filter.h
index a366a0b64a57..f349e2c0884c 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h
@@ -918,7 +918,7 @@ static inline int xdp_ok_fwd_dev(const struct net_device *fwd,
 	return 0;
 }
 
-/* The pair of xdp_do_redirect and xdp_do_flush_map MUST be called in the
+/* The pair of xdp_do_redirect and xdp_do_flush MUST be called in the
  * same cpu context. Further for best results no more than a single map
  * for the do_redirect/do_flush pair should be used. This limitation is
  * because we only track one map and force a flush when the map changes.
@@ -929,7 +929,13 @@ int xdp_do_generic_redirect(struct net_device *dev, struct sk_buff *skb,
 int xdp_do_redirect(struct net_device *dev,
 		    struct xdp_buff *xdp,
 		    struct bpf_prog *prog);
-void xdp_do_flush_map(void);
+void xdp_do_flush(void);
+
+/* The xdp_do_flush_map() helper has been renamed to drop the _map suffix, as
+ * it is no longer only flushing maps. Keep this define for compatibility
+ * until all drivers are updated - do not use xdp_do_flush_map() in new code!
+ */
+#define xdp_do_flush_map xdp_do_flush
 
 void bpf_warn_invalid_xdp_action(u32 act);
 
diff --git a/include/trace/events/xdp.h b/include/trace/events/xdp.h
index 72bad13d4a3c..b680973687b4 100644
--- a/include/trace/events/xdp.h
+++ b/include/trace/events/xdp.h
@@ -79,14 +79,26 @@ TRACE_EVENT(xdp_bulk_tx,
 		  __entry->sent, __entry->drops, __entry->err)
 );
 
+#ifndef __DEVMAP_OBJ_TYPE
+#define __DEVMAP_OBJ_TYPE
+struct _bpf_dtab_netdev {
+	struct net_device *dev;
+};
+#endif /* __DEVMAP_OBJ_TYPE */
+
+#define devmap_ifindex(tgt, map)				\
+	(((map->map_type == BPF_MAP_TYPE_DEVMAP ||	\
+		  map->map_type == BPF_MAP_TYPE_DEVMAP_HASH)) ? \
+	  ((struct _bpf_dtab_netdev *)tgt)->dev->ifindex : 0)
+
 DECLARE_EVENT_CLASS(xdp_redirect_template,
 
 	TP_PROTO(const struct net_device *dev,
 		 const struct bpf_prog *xdp,
-		 int to_ifindex, int err,
-		 const struct bpf_map *map, u32 map_index),
+		 const void *tgt, int err,
+		 const struct bpf_map *map, u32 index),
 
-	TP_ARGS(dev, xdp, to_ifindex, err, map, map_index),
+	TP_ARGS(dev, xdp, tgt, err, map, index),
 
 	TP_STRUCT__entry(
 		__field(int, prog_id)
@@ -103,90 +115,65 @@ DECLARE_EVENT_CLASS(xdp_redirect_template,
 		__entry->act		= XDP_REDIRECT;
 		__entry->ifindex	= dev->ifindex;
 		__entry->err		= err;
-		__entry->to_ifindex	= to_ifindex;
+		__entry->to_ifindex	= map ? devmap_ifindex(tgt, map) :
+						index;
 		__entry->map_id		= map ? map->id : 0;
-		__entry->map_index	= map_index;
+		__entry->map_index	= map ? index : 0;
 	),
 
-	TP_printk("prog_id=%d action=%s ifindex=%d to_ifindex=%d err=%d",
+	TP_printk("prog_id=%d action=%s ifindex=%d to_ifindex=%d err=%d"
+		  " map_id=%d map_index=%d",
 		  __entry->prog_id,
 		  __print_symbolic(__entry->act, __XDP_ACT_SYM_TAB),
 		  __entry->ifindex, __entry->to_ifindex,
-		  __entry->err)
+		  __entry->err, __entry->map_id, __entry->map_index)
 );
 
 DEFINE_EVENT(xdp_redirect_template, xdp_redirect,
 	TP_PROTO(const struct net_device *dev,
 		 const struct bpf_prog *xdp,
-		 int to_ifindex, int err,
-		 const struct bpf_map *map, u32 map_index),
-	TP_ARGS(dev, xdp, to_ifindex, err, map, map_index)
+		 const void *tgt, int err,
+		 const struct bpf_map *map, u32 index),
+	TP_ARGS(dev, xdp, tgt, err, map, index)
 );
 
 DEFINE_EVENT(xdp_redirect_template, xdp_redirect_err,
 	TP_PROTO(const struct net_device *dev,
 		 const struct bpf_prog *xdp,
-		 int to_ifindex, int err,
-		 const struct bpf_map *map, u32 map_index),
-	TP_ARGS(dev, xdp, to_ifindex, err, map, map_index)
+		 const void *tgt, int err,
+		 const struct bpf_map *map, u32 index),
+	TP_ARGS(dev, xdp, tgt, err, map, index)
 );
 
 #define _trace_xdp_redirect(dev, xdp, to)		\
-	 trace_xdp_redirect(dev, xdp, to, 0, NULL, 0);
+	 trace_xdp_redirect(dev, xdp, NULL, 0, NULL, to);
 
 #define _trace_xdp_redirect_err(dev, xdp, to, err)	\
-	 trace_xdp_redirect_err(dev, xdp, to, err, NULL, 0);
+	 trace_xdp_redirect_err(dev, xdp, NULL, err, NULL, to);
+
+#define _trace_xdp_redirect_map(dev, xdp, to, map, index)		\
+	 trace_xdp_redirect(dev, xdp, to, 0, map, index);
+
+#define _trace_xdp_redirect_map_err(dev, xdp, to, map, index, err)	\
+	 trace_xdp_redirect_err(dev, xdp, to, err, map, index);
 
-DEFINE_EVENT_PRINT(xdp_redirect_template, xdp_redirect_map,
+/* not used anymore, but kept around so as not to break old programs */
+DEFINE_EVENT(xdp_redirect_template, xdp_redirect_map,
 	TP_PROTO(const struct net_device *dev,
 		 const struct bpf_prog *xdp,
-		 int to_ifindex, int err,
-		 const struct bpf_map *map, u32 map_index),
-	TP_ARGS(dev, xdp, to_ifindex, err, map, map_index),
-	TP_printk("prog_id=%d action=%s ifindex=%d to_ifindex=%d err=%d"
-		  " map_id=%d map_index=%d",
-		  __entry->prog_id,
-		  __print_symbolic(__entry->act, __XDP_ACT_SYM_TAB),
-		  __entry->ifindex, __entry->to_ifindex,
-		  __entry->err,
-		  __entry->map_id, __entry->map_index)
+		 const void *tgt, int err,
+		 const struct bpf_map *map, u32 index),
+	TP_ARGS(dev, xdp, tgt, err, map, index)
 );
 
-DEFINE_EVENT_PRINT(xdp_redirect_template, xdp_redirect_map_err,
+DEFINE_EVENT(xdp_redirect_template, xdp_redirect_map_err,
 	TP_PROTO(const struct net_device *dev,
 		 const struct bpf_prog *xdp,
-		 int to_ifindex, int err,
-		 const struct bpf_map *map, u32 map_index),
-	TP_ARGS(dev, xdp, to_ifindex, err, map, map_index),
-	TP_printk("prog_id=%d action=%s ifindex=%d to_ifindex=%d err=%d"
-		  " map_id=%d map_index=%d",
-		  __entry->prog_id,
-		  __print_symbolic(__entry->act, __XDP_ACT_SYM_TAB),
-		  __entry->ifindex, __entry->to_ifindex,
-		  __entry->err,
-		  __entry->map_id, __entry->map_index)
+		 const void *tgt, int err,
+		 const struct bpf_map *map, u32 index),
+	TP_ARGS(dev, xdp, tgt, err, map, index)
 );
 
-#ifndef __DEVMAP_OBJ_TYPE
-#define __DEVMAP_OBJ_TYPE
-struct _bpf_dtab_netdev {
-	struct net_device *dev;
-};
-#endif /* __DEVMAP_OBJ_TYPE */
-
-#define devmap_ifindex(fwd, map)				\
-	((map->map_type == BPF_MAP_TYPE_DEVMAP ||		\
-	  map->map_type == BPF_MAP_TYPE_DEVMAP_HASH) ?		\
-	  ((struct _bpf_dtab_netdev *)fwd)->dev->ifindex : 0)
-
-#define _trace_xdp_redirect_map(dev, xdp, fwd, map, idx)		\
-	 trace_xdp_redirect_map(dev, xdp, devmap_ifindex(fwd, map),	\
-				0, map, idx)
-
-#define _trace_xdp_redirect_map_err(dev, xdp, fwd, map, idx, err)	\
-	 trace_xdp_redirect_map_err(dev, xdp, devmap_ifindex(fwd, map),	\
-				    err, map, idx)
-
 TRACE_EVENT(xdp_cpumap_kthread,
 
 	TP_PROTO(int map_id, unsigned int processed,  unsigned int drops,
diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index 030d125c3839..d5311009953f 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -81,7 +81,7 @@ struct bpf_dtab {
 	u32 n_buckets;
 };
 
-static DEFINE_PER_CPU(struct list_head, dev_map_flush_list);
+static DEFINE_PER_CPU(struct list_head, dev_flush_list);
 static DEFINE_SPINLOCK(dev_map_lock);
 static LIST_HEAD(dev_map_list);
 
@@ -357,16 +357,16 @@ error:
 	goto out;
 }
 
-/* __dev_map_flush is called from xdp_do_flush_map() which _must_ be signaled
+/* __dev_flush is called from xdp_do_flush() which _must_ be signaled
  * from the driver before returning from its napi->poll() routine. The poll()
  * routine is called either from busy_poll context or net_rx_action signaled
  * from NET_RX_SOFTIRQ. Either way the poll routine must complete before the
  * net device can be torn down. On devmap tear down we ensure the flush list
  * is empty before completing to ensure all flush operations have completed.
  */
-void __dev_map_flush(void)
+void __dev_flush(void)
 {
-	struct list_head *flush_list = this_cpu_ptr(&dev_map_flush_list);
+	struct list_head *flush_list = this_cpu_ptr(&dev_flush_list);
 	struct xdp_dev_bulk_queue *bq, *tmp;
 
 	rcu_read_lock();
@@ -396,9 +396,8 @@ struct bpf_dtab_netdev *__dev_map_lookup_elem(struct bpf_map *map, u32 key)
  */
 static int bq_enqueue(struct net_device *dev, struct xdp_frame *xdpf,
 		      struct net_device *dev_rx)
-
 {
-	struct list_head *flush_list = this_cpu_ptr(&dev_map_flush_list);
+	struct list_head *flush_list = this_cpu_ptr(&dev_flush_list);
 	struct xdp_dev_bulk_queue *bq = this_cpu_ptr(dev->xdp_bulkq);
 
 	if (unlikely(bq->count == DEV_MAP_BULK_SIZE))
@@ -419,10 +418,9 @@ static int bq_enqueue(struct net_device *dev, struct xdp_frame *xdpf,
 	return 0;
 }
 
-int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
-		    struct net_device *dev_rx)
+static inline int __xdp_enqueue(struct net_device *dev, struct xdp_buff *xdp,
+			       struct net_device *dev_rx)
 {
-	struct net_device *dev = dst->dev;
 	struct xdp_frame *xdpf;
 	int err;
 
@@ -440,6 +438,20 @@ int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
 	return bq_enqueue(dev, xdpf, dev_rx);
 }
 
+int dev_xdp_enqueue(struct net_device *dev, struct xdp_buff *xdp,
+		    struct net_device *dev_rx)
+{
+	return __xdp_enqueue(dev, xdp, dev_rx);
+}
+
+int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
+		    struct net_device *dev_rx)
+{
+	struct net_device *dev = dst->dev;
+
+	return __xdp_enqueue(dev, xdp, dev_rx);
+}
+
 int dev_map_generic_redirect(struct bpf_dtab_netdev *dst, struct sk_buff *skb,
 			     struct bpf_prog *xdp_prog)
 {
@@ -762,7 +774,7 @@ static int __init dev_map_init(void)
 	register_netdevice_notifier(&dev_map_notifier);
 
 	for_each_possible_cpu(cpu)
-		INIT_LIST_HEAD(&per_cpu(dev_map_flush_list, cpu));
+		INIT_LIST_HEAD(&per_cpu(dev_flush_list, cpu));
 	return 0;
 }
 
diff --git a/net/core/filter.c b/net/core/filter.c
index d6f58dc4e1c4..17de6747d9e3 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -3458,58 +3458,6 @@ static const struct bpf_func_proto bpf_xdp_adjust_meta_proto = {
 	.arg2_type	= ARG_ANYTHING,
 };
 
-static int __bpf_tx_xdp(struct net_device *dev,
-			struct bpf_map *map,
-			struct xdp_buff *xdp,
-			u32 index)
-{
-	struct xdp_frame *xdpf;
-	int err, sent;
-
-	if (!dev->netdev_ops->ndo_xdp_xmit) {
-		return -EOPNOTSUPP;
-	}
-
-	err = xdp_ok_fwd_dev(dev, xdp->data_end - xdp->data);
-	if (unlikely(err))
-		return err;
-
-	xdpf = convert_to_xdp_frame(xdp);
-	if (unlikely(!xdpf))
-		return -EOVERFLOW;
-
-	sent = dev->netdev_ops->ndo_xdp_xmit(dev, 1, &xdpf, XDP_XMIT_FLUSH);
-	if (sent <= 0)
-		return sent;
-	return 0;
-}
-
-static noinline int
-xdp_do_redirect_slow(struct net_device *dev, struct xdp_buff *xdp,
-		     struct bpf_prog *xdp_prog, struct bpf_redirect_info *ri)
-{
-	struct net_device *fwd;
-	u32 index = ri->tgt_index;
-	int err;
-
-	fwd = dev_get_by_index_rcu(dev_net(dev), index);
-	ri->tgt_index = 0;
-	if (unlikely(!fwd)) {
-		err = -EINVAL;
-		goto err;
-	}
-
-	err = __bpf_tx_xdp(fwd, NULL, xdp, 0);
-	if (unlikely(err))
-		goto err;
-
-	_trace_xdp_redirect(dev, xdp_prog, index);
-	return 0;
-err:
-	_trace_xdp_redirect_err(dev, xdp_prog, index, err);
-	return err;
-}
-
 static int __bpf_tx_xdp_map(struct net_device *dev_rx, void *fwd,
 			    struct bpf_map *map, struct xdp_buff *xdp)
 {
@@ -3527,13 +3475,13 @@ static int __bpf_tx_xdp_map(struct net_device *dev_rx, void *fwd,
 	return 0;
 }
 
-void xdp_do_flush_map(void)
+void xdp_do_flush(void)
 {
-	__dev_map_flush();
+	__dev_flush();
 	__cpu_map_flush();
 	__xsk_map_flush();
 }
-EXPORT_SYMBOL_GPL(xdp_do_flush_map);
+EXPORT_SYMBOL_GPL(xdp_do_flush);
 
 static inline void *__xdp_map_lookup_elem(struct bpf_map *map, u32 index)
 {
@@ -3568,10 +3516,11 @@ void bpf_clear_redirect_map(struct bpf_map *map)
 	}
 }
 
-static int xdp_do_redirect_map(struct net_device *dev, struct xdp_buff *xdp,
-			       struct bpf_prog *xdp_prog, struct bpf_map *map,
-			       struct bpf_redirect_info *ri)
+int xdp_do_redirect(struct net_device *dev, struct xdp_buff *xdp,
+		    struct bpf_prog *xdp_prog)
 {
+	struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info);
+	struct bpf_map *map = READ_ONCE(ri->map);
 	u32 index = ri->tgt_index;
 	void *fwd = ri->tgt_value;
 	int err;
@@ -3580,7 +3529,18 @@ static int xdp_do_redirect_map(struct net_device *dev, struct xdp_buff *xdp,
 	ri->tgt_value = NULL;
 	WRITE_ONCE(ri->map, NULL);
 
-	err = __bpf_tx_xdp_map(dev, fwd, map, xdp);
+	if (unlikely(!map)) {
+		fwd = dev_get_by_index_rcu(dev_net(dev), index);
+		if (unlikely(!fwd)) {
+			err = -EINVAL;
+			goto err;
+		}
+
+		err = dev_xdp_enqueue(fwd, xdp, dev);
+	} else {
+		err = __bpf_tx_xdp_map(dev, fwd, map, xdp);
+	}
+
 	if (unlikely(err))
 		goto err;
 
@@ -3590,18 +3550,6 @@ err:
 	_trace_xdp_redirect_map_err(dev, xdp_prog, fwd, map, index, err);
 	return err;
 }
-
-int xdp_do_redirect(struct net_device *dev, struct xdp_buff *xdp,
-		    struct bpf_prog *xdp_prog)
-{
-	struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info);
-	struct bpf_map *map = READ_ONCE(ri->map);
-
-	if (likely(map))
-		return xdp_do_redirect_map(dev, xdp, xdp_prog, map, ri);
-
-	return xdp_do_redirect_slow(dev, xdp, xdp_prog, ri);
-}
 EXPORT_SYMBOL_GPL(xdp_do_redirect);
 
 static int xdp_do_generic_redirect_map(struct net_device *dev,
-- 
cgit v1.2.3-59-g8ed1b


From 58aa94f922c1b44e0919d1814d2eab5b9e8bf945 Mon Sep 17 00:00:00 2001
From: Jesper Dangaard Brouer <brouer@redhat.com>
Date: Thu, 16 Jan 2020 16:14:46 +0100
Subject: devmap: Adjust tracepoint for map-less queue flush
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Now that we don't have a reference to a devmap when flushing the device
bulk queue, let's change the the devmap_xmit tracepoint to remote the
map_id and map_index fields entirely. Rearrange the fields so 'drops' and
'sent' stay in the same position in the tracepoint struct, to make it
possible for the xdp_monitor utility to read both the old and the new
format.

Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/157918768613.1458396.9165902403373826572.stgit@toke.dk
---
 include/trace/events/xdp.h     | 29 ++++++++++++-----------------
 kernel/bpf/devmap.c            |  2 +-
 samples/bpf/xdp_monitor_kern.c |  8 +++-----
 3 files changed, 16 insertions(+), 23 deletions(-)

diff --git a/include/trace/events/xdp.h b/include/trace/events/xdp.h
index b680973687b4..b95d65e8c628 100644
--- a/include/trace/events/xdp.h
+++ b/include/trace/events/xdp.h
@@ -246,43 +246,38 @@ TRACE_EVENT(xdp_cpumap_enqueue,
 
 TRACE_EVENT(xdp_devmap_xmit,
 
-	TP_PROTO(const struct bpf_map *map, u32 map_index,
-		 int sent, int drops,
-		 const struct net_device *from_dev,
-		 const struct net_device *to_dev, int err),
+	TP_PROTO(const struct net_device *from_dev,
+		 const struct net_device *to_dev,
+		 int sent, int drops, int err),
 
-	TP_ARGS(map, map_index, sent, drops, from_dev, to_dev, err),
+	TP_ARGS(from_dev, to_dev, sent, drops, err),
 
 	TP_STRUCT__entry(
-		__field(int, map_id)
+		__field(int, from_ifindex)
 		__field(u32, act)
-		__field(u32, map_index)
+		__field(int, to_ifindex)
 		__field(int, drops)
 		__field(int, sent)
-		__field(int, from_ifindex)
-		__field(int, to_ifindex)
 		__field(int, err)
 	),
 
 	TP_fast_assign(
-		__entry->map_id		= map ? map->id : 0;
+		__entry->from_ifindex	= from_dev->ifindex;
 		__entry->act		= XDP_REDIRECT;
-		__entry->map_index	= map_index;
+		__entry->to_ifindex	= to_dev->ifindex;
 		__entry->drops		= drops;
 		__entry->sent		= sent;
-		__entry->from_ifindex	= from_dev->ifindex;
-		__entry->to_ifindex	= to_dev->ifindex;
 		__entry->err		= err;
 	),
 
 	TP_printk("ndo_xdp_xmit"
-		  " map_id=%d map_index=%d action=%s"
+		  " from_ifindex=%d to_ifindex=%d action=%s"
 		  " sent=%d drops=%d"
-		  " from_ifindex=%d to_ifindex=%d err=%d",
-		  __entry->map_id, __entry->map_index,
+		  " err=%d",
+		  __entry->from_ifindex, __entry->to_ifindex,
 		  __print_symbolic(__entry->act, __XDP_ACT_SYM_TAB),
 		  __entry->sent, __entry->drops,
-		  __entry->from_ifindex, __entry->to_ifindex, __entry->err)
+		  __entry->err)
 );
 
 /* Expect users already include <net/xdp.h>, but not xdp_priv.h */
diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index d5311009953f..de630f980282 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -340,7 +340,7 @@ static int bq_xmit_all(struct xdp_dev_bulk_queue *bq, u32 flags)
 out:
 	bq->count = 0;
 
-	trace_xdp_devmap_xmit(NULL, 0, sent, drops, bq->dev_rx, dev, err);
+	trace_xdp_devmap_xmit(bq->dev_rx, dev, sent, drops, err);
 	bq->dev_rx = NULL;
 	__list_del_clearprev(&bq->flush_node);
 	return 0;
diff --git a/samples/bpf/xdp_monitor_kern.c b/samples/bpf/xdp_monitor_kern.c
index ad10fe700d7d..39458a44472e 100644
--- a/samples/bpf/xdp_monitor_kern.c
+++ b/samples/bpf/xdp_monitor_kern.c
@@ -222,14 +222,12 @@ struct bpf_map_def SEC("maps") devmap_xmit_cnt = {
  */
 struct devmap_xmit_ctx {
 	u64 __pad;		// First 8 bytes are not accessible by bpf code
-	int map_id;		//	offset:8;  size:4; signed:1;
+	int from_ifindex;	//	offset:8;  size:4; signed:1;
 	u32 act;		//	offset:12; size:4; signed:0;
-	u32 map_index;		//	offset:16; size:4; signed:0;
+	int to_ifindex; 	//	offset:16; size:4; signed:1;
 	int drops;		//	offset:20; size:4; signed:1;
 	int sent;		//	offset:24; size:4; signed:1;
-	int from_ifindex;	//	offset:28; size:4; signed:1;
-	int to_ifindex;		//	offset:32; size:4; signed:1;
-	int err;		//	offset:36; size:4; signed:1;
+	int err;		//	offset:28; size:4; signed:1;
 };
 
 SEC("tracepoint/xdp/xdp_devmap_xmit")
-- 
cgit v1.2.3-59-g8ed1b


From 81f2b572cf4fd5c4178fe0e2b5bb1ab096385fd8 Mon Sep 17 00:00:00 2001
From: YueHaibing <yuehaibing@huawei.com>
Date: Thu, 16 Jan 2020 22:53:00 +0800
Subject: bpf: Remove set but not used variable 'first_key'

kernel/bpf/syscall.c: In function generic_map_lookup_batch:
kernel/bpf/syscall.c:1339:7: warning: variable first_key set but not used [-Wunused-but-set-variable]

It is never used, so remove it.

Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Brian Vazquez <brianvv@google.com>
Link: https://lore.kernel.org/bpf/20200116145300.59056-1-yuehaibing@huawei.com
---
 kernel/bpf/syscall.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 0d94d361379c..c26a71460f02 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -1336,7 +1336,6 @@ int generic_map_lookup_batch(struct bpf_map *map,
 	void *buf, *buf_prevkey, *prev_key, *key, *value;
 	int err, retry = MAP_LOOKUP_RETRIES;
 	u32 value_size, cp, max_count;
-	bool first_key = false;
 
 	if (attr->batch.elem_flags & ~BPF_F_LOCK)
 		return -EINVAL;
@@ -1365,7 +1364,6 @@ int generic_map_lookup_batch(struct bpf_map *map,
 	}
 
 	err = -EFAULT;
-	first_key = false;
 	prev_key = NULL;
 	if (ubatch && copy_from_user(buf_prevkey, ubatch, map->key_size))
 		goto free_buf;
-- 
cgit v1.2.3-59-g8ed1b


From f3265971ded98a069ad699b51b8a5ab95e9e5be1 Mon Sep 17 00:00:00 2001
From: Madhuparna Bhowmik <madhuparnabhowmik04@gmail.com>
Date: Wed, 15 Jan 2020 21:25:53 +0530
Subject: net: xen-netback: hash.c: Use built-in RCU list checking

list_for_each_entry_rcu has built-in RCU and lock checking.
Pass cond argument to list_for_each_entry_rcu.

Signed-off-by: Madhuparna Bhowmik <madhuparnabhowmik04@gmail.com>
Acked-by: Wei Liu <wei.liu@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/xen-netback/hash.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/net/xen-netback/hash.c b/drivers/net/xen-netback/hash.c
index 10d580c3dea3..6b7532f7c936 100644
--- a/drivers/net/xen-netback/hash.c
+++ b/drivers/net/xen-netback/hash.c
@@ -51,7 +51,8 @@ static void xenvif_add_hash(struct xenvif *vif, const u8 *tag,
 
 	found = false;
 	oldest = NULL;
-	list_for_each_entry_rcu(entry, &vif->hash.cache.list, link) {
+	list_for_each_entry_rcu(entry, &vif->hash.cache.list, link,
+				lockdep_is_held(&vif->hash.cache.lock)) {
 		/* Make sure we don't add duplicate entries */
 		if (entry->len == len &&
 		    memcmp(entry->tag, tag, len) == 0)
@@ -102,7 +103,8 @@ static void xenvif_flush_hash(struct xenvif *vif)
 
 	spin_lock_irqsave(&vif->hash.cache.lock, flags);
 
-	list_for_each_entry_rcu(entry, &vif->hash.cache.list, link) {
+	list_for_each_entry_rcu(entry, &vif->hash.cache.list, link,
+				lockdep_is_held(&vif->hash.cache.lock)) {
 		list_del_rcu(&entry->link);
 		vif->hash.cache.count--;
 		kfree_rcu(entry, rcu);
-- 
cgit v1.2.3-59-g8ed1b


From 41cdc741048b0d04604c02aad9ec19f7d9130b70 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Wed, 15 Jan 2020 11:57:41 -0800
Subject: netdevsim: fix nsim_fib6_rt_create() error path

It seems nsim_fib6_rt_create() intent was to return
either a valid pointer or an embedded error code.

BUG: unable to handle page fault for address: fffffffffffffff4
PGD 9870067 P4D 9870067 PUD 9872067 PMD 0
Oops: 0000 [#1] PREEMPT SMP KASAN
CPU: 0 PID: 22851 Comm: syz-executor.1 Not tainted 5.5.0-rc5-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
RIP: 0010:jhash2 include/linux/jhash.h:125 [inline]
RIP: 0010:rhashtable_jhash2+0x76/0x2c0 lib/rhashtable.c:963
Code: b9 00 00 00 00 00 fc ff df 48 c1 e8 03 0f b6 14 08 4c 89 f0 83 e0 07 83 c0 03 38 d0 7c 08 84 d2 0f 85 30 02 00 00 49 8d 7e 04 <41> 8b 06 48 be 00 00 00 00 00 fc ff df 48 89 fa 48 c1 ea 03 0f b6
RSP: 0018:ffffc90016127190 EFLAGS: 00010246
RAX: 0000000000000007 RBX: 00000000dfb3ab49 RCX: dffffc0000000000
RDX: 0000000000000000 RSI: ffffffff839ba7c8 RDI: fffffffffffffff8
RBP: ffffc900161271c0 R08: ffff8880951f8640 R09: ffffed1015d0703d
R10: ffffed1015d0703c R11: ffff8880ae8381e3 R12: 00000000dfb3ab49
R13: 00000000dfb3ab49 R14: fffffffffffffff4 R15: 0000000000000007
FS:  00007f40bfbc6700(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: fffffffffffffff4 CR3: 0000000093660000 CR4: 00000000001406f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 rht_key_get_hash include/linux/rhashtable.h:133 [inline]
 rht_key_hashfn include/linux/rhashtable.h:159 [inline]
 rht_head_hashfn include/linux/rhashtable.h:174 [inline]
 __rhashtable_insert_fast.constprop.0+0xe15/0x1180 include/linux/rhashtable.h:723
 rhashtable_insert_fast include/linux/rhashtable.h:832 [inline]
 nsim_fib6_rt_add drivers/net/netdevsim/fib.c:603 [inline]
 nsim_fib6_rt_insert drivers/net/netdevsim/fib.c:658 [inline]
 nsim_fib6_event drivers/net/netdevsim/fib.c:719 [inline]
 nsim_fib_event drivers/net/netdevsim/fib.c:744 [inline]
 nsim_fib_event_nb+0x1b16/0x2600 drivers/net/netdevsim/fib.c:772
 notifier_call_chain+0xc2/0x230 kernel/notifier.c:83
 __atomic_notifier_call_chain+0xa6/0x1a0 kernel/notifier.c:173
 atomic_notifier_call_chain+0x2e/0x40 kernel/notifier.c:183
 call_fib_notifiers+0x173/0x2a0 net/core/fib_notifier.c:35
 call_fib6_notifiers+0x4b/0x60 net/ipv6/fib6_notifier.c:22
 call_fib6_entry_notifiers+0xfb/0x150 net/ipv6/ip6_fib.c:399
 fib6_add_rt2node net/ipv6/ip6_fib.c:1216 [inline]
 fib6_add+0x20cd/0x3ec0 net/ipv6/ip6_fib.c:1471
 __ip6_ins_rt+0x54/0x80 net/ipv6/route.c:1315
 ip6_ins_rt+0x96/0xd0 net/ipv6/route.c:1325
 __ipv6_dev_ac_inc+0x76f/0xb20 net/ipv6/anycast.c:324
 ipv6_sock_ac_join+0x4c1/0x790 net/ipv6/anycast.c:139
 do_ipv6_setsockopt.isra.0+0x3908/0x4290 net/ipv6/ipv6_sockglue.c:670
 ipv6_setsockopt+0xff/0x180 net/ipv6/ipv6_sockglue.c:944
 udpv6_setsockopt+0x68/0xb0 net/ipv6/udp.c:1564
 sock_common_setsockopt+0x94/0xd0 net/core/sock.c:3149
 __sys_setsockopt+0x261/0x4c0 net/socket.c:2130
 __do_sys_setsockopt net/socket.c:2146 [inline]
 __se_sys_setsockopt net/socket.c:2143 [inline]
 __x64_sys_setsockopt+0xbe/0x150 net/socket.c:2143
 do_syscall_64+0xfa/0x790 arch/x86/entry/common.c:294
 entry_SYSCALL_64_after_hwframe+0x49/0xbe
RIP: 0033:0x45aff9

Fixes: 48bb9eb47b27 ("netdevsim: fib: Add dummy implementation for FIB offload")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: syzbot <syzkaller@googlegroups.com>
Cc: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/netdevsim/fib.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/netdevsim/fib.c b/drivers/net/netdevsim/fib.c
index 8f56289fc2ec..f32d56ac3e80 100644
--- a/drivers/net/netdevsim/fib.c
+++ b/drivers/net/netdevsim/fib.c
@@ -467,7 +467,7 @@ nsim_fib6_rt_create(struct nsim_fib_data *data,
 
 	fib6_rt = kzalloc(sizeof(*fib6_rt), GFP_ATOMIC);
 	if (!fib6_rt)
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 
 	nsim_fib_rt_init(data, &fib6_rt->common, &rt->fib6_dst.addr,
 			 sizeof(rt->fib6_dst.addr), rt->fib6_dst.plen, AF_INET6,
@@ -650,8 +650,8 @@ static int nsim_fib6_rt_insert(struct nsim_fib_data *data,
 	int err;
 
 	fib6_rt = nsim_fib6_rt_create(data, fen6_info);
-	if (!fib6_rt)
-		return -ENOMEM;
+	if (IS_ERR(fib6_rt))
+		return PTR_ERR(fib6_rt);
 
 	fib6_rt_old = nsim_fib6_rt_lookup(&data->fib_rt_ht, fen6_info->rt);
 	if (!fib6_rt_old)
-- 
cgit v1.2.3-59-g8ed1b


From 080bb352fad00d04995102f681b134e3754bfb6e Mon Sep 17 00:00:00 2001
From: Florian Fainelli <f.fainelli@gmail.com>
Date: Wed, 15 Jan 2020 20:48:50 -0800
Subject: net: phy: Maintain MDIO device and bus statistics

We maintain global statistics for an entire MDIO bus, as well as broken
down, per MDIO bus address statistics. Given that it is possible for
MDIO devices such as switches to access MDIO bus addresses for which
there is not a mdio_device instance created (therefore not a a
corresponding device directory in sysfs either), we also maintain
per-address statistics under the statistics folder. The layout looks
like this:

/sys/class/mdio_bus/../statistics/
	transfers
	errrors
	writes
	reads
	transfers_<addr>
	errors_<addr>
	writes_<addr>
	reads_<addr>

When a mdio_device instance is registered, a statistics/ folder is
created with the tranfers, errors, writes and reads attributes which
point to the appropriate MDIO bus statistics structure.

Statistics are 64-bit unsigned quantities and maintained through the
u64_stats_sync.h helper functions.

Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Tested-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/ABI/testing/sysfs-bus-mdio |  63 ++++++++
 drivers/net/phy/mdio_bus.c               | 251 ++++++++++++++++++++++++++++++-
 include/linux/phy.h                      |  11 ++
 3 files changed, 323 insertions(+), 2 deletions(-)
 create mode 100644 Documentation/ABI/testing/sysfs-bus-mdio

diff --git a/Documentation/ABI/testing/sysfs-bus-mdio b/Documentation/ABI/testing/sysfs-bus-mdio
new file mode 100644
index 000000000000..da86efc7781b
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-bus-mdio
@@ -0,0 +1,63 @@
+What:          /sys/bus/mdio_bus/devices/.../statistics/
+Date:          January 2020
+KernelVersion: 5.6
+Contact:       netdev@vger.kernel.org
+Description:
+		This folder contains statistics about global and per
+		MDIO bus address statistics.
+
+What:          /sys/bus/mdio_bus/devices/.../statistics/transfers
+Date:          January 2020
+KernelVersion: 5.6
+Contact:       netdev@vger.kernel.org
+Description:
+		Total number of transfers for this MDIO bus.
+
+What:          /sys/bus/mdio_bus/devices/.../statistics/errors
+Date:          January 2020
+KernelVersion: 5.6
+Contact:       netdev@vger.kernel.org
+Description:
+		Total number of transfer errors for this MDIO bus.
+
+What:          /sys/bus/mdio_bus/devices/.../statistics/writes
+Date:          January 2020
+KernelVersion: 5.6
+Contact:       netdev@vger.kernel.org
+Description:
+		Total number of write transactions for this MDIO bus.
+
+What:          /sys/bus/mdio_bus/devices/.../statistics/reads
+Date:          January 2020
+KernelVersion: 5.6
+Contact:       netdev@vger.kernel.org
+Description:
+		Total number of read transactions for this MDIO bus.
+
+What:          /sys/bus/mdio_bus/devices/.../statistics/transfers_<addr>
+Date:          January 2020
+KernelVersion: 5.6
+Contact:       netdev@vger.kernel.org
+Description:
+		Total number of transfers for this MDIO bus address.
+
+What:          /sys/bus/mdio_bus/devices/.../statistics/errors_<addr>
+Date:          January 2020
+KernelVersion: 5.6
+Contact:       netdev@vger.kernel.org
+Description:
+		Total number of transfer errors for this MDIO bus address.
+
+What:          /sys/bus/mdio_bus/devices/.../statistics/writes_<addr>
+Date:          January 2020
+KernelVersion: 5.6
+Contact:       netdev@vger.kernel.org
+Description:
+		Total number of write transactions for this MDIO bus address.
+
+What:          /sys/bus/mdio_bus/devices/.../statistics/reads_<addr>
+Date:          January 2020
+KernelVersion: 5.6
+Contact:       netdev@vger.kernel.org
+Description:
+		Total number of read transactions for this MDIO bus address.
diff --git a/drivers/net/phy/mdio_bus.c b/drivers/net/phy/mdio_bus.c
index 8d753bb07227..9bb9f37f21dc 100644
--- a/drivers/net/phy/mdio_bus.c
+++ b/drivers/net/phy/mdio_bus.c
@@ -158,9 +158,11 @@ struct mii_bus *mdiobus_alloc_size(size_t size)
 	if (size)
 		bus->priv = (void *)bus + aligned_size;
 
-	/* Initialise the interrupts to polling */
-	for (i = 0; i < PHY_MAX_ADDR; i++)
+	/* Initialise the interrupts to polling and 64-bit seqcounts */
+	for (i = 0; i < PHY_MAX_ADDR; i++) {
 		bus->irq[i] = PHY_POLL;
+		u64_stats_init(&bus->stats[i].syncp);
+	}
 
 	return bus;
 }
@@ -249,9 +251,215 @@ static void mdiobus_release(struct device *d)
 	kfree(bus);
 }
 
+struct mdio_bus_stat_attr {
+	int addr;
+	unsigned int field_offset;
+};
+
+static u64 mdio_bus_get_stat(struct mdio_bus_stats *s, unsigned int offset)
+{
+	const char *p = (const char *)s + offset;
+	unsigned int start;
+	u64 val = 0;
+
+	do {
+		start = u64_stats_fetch_begin(&s->syncp);
+		val = u64_stats_read((const u64_stats_t *)p);
+	} while (u64_stats_fetch_retry(&s->syncp, start));
+
+	return val;
+}
+
+static u64 mdio_bus_get_global_stat(struct mii_bus *bus, unsigned int offset)
+{
+	unsigned int i;
+	u64 val = 0;
+
+	for (i = 0; i < PHY_MAX_ADDR; i++)
+		val += mdio_bus_get_stat(&bus->stats[i], offset);
+
+	return val;
+}
+
+static ssize_t mdio_bus_stat_field_show(struct device *dev,
+					struct device_attribute *attr,
+					char *buf)
+{
+	struct mii_bus *bus = to_mii_bus(dev);
+	struct mdio_bus_stat_attr *sattr;
+	struct dev_ext_attribute *eattr;
+	u64 val;
+
+	eattr = container_of(attr, struct dev_ext_attribute, attr);
+	sattr = eattr->var;
+
+	if (sattr->addr < 0)
+		val = mdio_bus_get_global_stat(bus, sattr->field_offset);
+	else
+		val = mdio_bus_get_stat(&bus->stats[sattr->addr],
+					sattr->field_offset);
+
+	return sprintf(buf, "%llu\n", val);
+}
+
+static ssize_t mdio_bus_device_stat_field_show(struct device *dev,
+					       struct device_attribute *attr,
+					       char *buf)
+{
+	struct mdio_device *mdiodev = to_mdio_device(dev);
+	struct mii_bus *bus = mdiodev->bus;
+	struct mdio_bus_stat_attr *sattr;
+	struct dev_ext_attribute *eattr;
+	int addr = mdiodev->addr;
+	u64 val;
+
+	eattr = container_of(attr, struct dev_ext_attribute, attr);
+	sattr = eattr->var;
+
+	val = mdio_bus_get_stat(&bus->stats[addr], sattr->field_offset);
+
+	return sprintf(buf, "%llu\n", val);
+}
+
+#define MDIO_BUS_STATS_ATTR_DECL(field, file)				\
+static struct dev_ext_attribute dev_attr_mdio_bus_##field = {		\
+	.attr = { .attr = { .name = file, .mode = 0444 },		\
+		     .show = mdio_bus_stat_field_show,			\
+	},								\
+	.var = &((struct mdio_bus_stat_attr) {				\
+		-1, offsetof(struct mdio_bus_stats, field)		\
+	}),								\
+};									\
+static struct dev_ext_attribute dev_attr_mdio_bus_device_##field = {	\
+	.attr = { .attr = { .name = file, .mode = 0444 },		\
+		     .show = mdio_bus_device_stat_field_show,		\
+	},								\
+	.var = &((struct mdio_bus_stat_attr) {				\
+		-1, offsetof(struct mdio_bus_stats, field)		\
+	}),								\
+};
+
+#define MDIO_BUS_STATS_ATTR(field)					\
+	MDIO_BUS_STATS_ATTR_DECL(field, __stringify(field))
+
+MDIO_BUS_STATS_ATTR(transfers);
+MDIO_BUS_STATS_ATTR(errors);
+MDIO_BUS_STATS_ATTR(writes);
+MDIO_BUS_STATS_ATTR(reads);
+
+#define MDIO_BUS_STATS_ADDR_ATTR_DECL(field, addr, file)		\
+static struct dev_ext_attribute dev_attr_mdio_bus_addr_##field##_##addr = { \
+	.attr = { .attr = { .name = file, .mode = 0444 },		\
+		     .show = mdio_bus_stat_field_show,			\
+	},								\
+	.var = &((struct mdio_bus_stat_attr) {				\
+		addr, offsetof(struct mdio_bus_stats, field)		\
+	}),								\
+}
+
+#define MDIO_BUS_STATS_ADDR_ATTR(field, addr)				\
+	MDIO_BUS_STATS_ADDR_ATTR_DECL(field, addr,			\
+				 __stringify(field) "_" __stringify(addr))
+
+#define MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(addr)			\
+	MDIO_BUS_STATS_ADDR_ATTR(transfers, addr);			\
+	MDIO_BUS_STATS_ADDR_ATTR(errors, addr);				\
+	MDIO_BUS_STATS_ADDR_ATTR(writes, addr);				\
+	MDIO_BUS_STATS_ADDR_ATTR(reads, addr)				\
+
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(0);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(1);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(2);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(3);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(4);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(5);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(6);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(7);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(8);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(9);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(10);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(11);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(12);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(13);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(14);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(15);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(16);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(17);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(18);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(19);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(20);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(21);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(22);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(23);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(24);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(25);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(26);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(27);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(28);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(29);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(30);
+MDIO_BUS_STATS_ADDR_ATTR_GROUP_DECL(31);
+
+#define MDIO_BUS_STATS_ADDR_ATTR_GROUP(addr)				\
+	&dev_attr_mdio_bus_addr_transfers_##addr.attr.attr,		\
+	&dev_attr_mdio_bus_addr_errors_##addr.attr.attr,		\
+	&dev_attr_mdio_bus_addr_writes_##addr.attr.attr,		\
+	&dev_attr_mdio_bus_addr_reads_##addr.attr.attr			\
+
+static struct attribute *mdio_bus_statistics_attrs[] = {
+	&dev_attr_mdio_bus_transfers.attr.attr,
+	&dev_attr_mdio_bus_errors.attr.attr,
+	&dev_attr_mdio_bus_writes.attr.attr,
+	&dev_attr_mdio_bus_reads.attr.attr,
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(0),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(1),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(2),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(3),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(4),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(5),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(6),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(7),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(8),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(9),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(10),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(11),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(12),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(13),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(14),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(15),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(16),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(17),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(18),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(19),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(20),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(21),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(22),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(23),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(24),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(25),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(26),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(27),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(28),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(29),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(30),
+	MDIO_BUS_STATS_ADDR_ATTR_GROUP(31),
+	NULL,
+};
+
+static const struct attribute_group mdio_bus_statistics_group = {
+	.name	= "statistics",
+	.attrs	= mdio_bus_statistics_attrs,
+};
+
+static const struct attribute_group *mdio_bus_groups[] = {
+	&mdio_bus_statistics_group,
+	NULL,
+};
+
 static struct class mdio_bus_class = {
 	.name		= "mdio_bus",
 	.dev_release	= mdiobus_release,
+	.dev_groups	= mdio_bus_groups,
 };
 
 #if IS_ENABLED(CONFIG_OF_MDIO)
@@ -530,6 +738,24 @@ struct phy_device *mdiobus_scan(struct mii_bus *bus, int addr)
 }
 EXPORT_SYMBOL(mdiobus_scan);
 
+static void mdiobus_stats_acct(struct mdio_bus_stats *stats, bool op, int ret)
+{
+	u64_stats_update_begin(&stats->syncp);
+
+	u64_stats_inc(&stats->transfers);
+	if (ret < 0) {
+		u64_stats_inc(&stats->errors);
+		goto out;
+	}
+
+	if (op)
+		u64_stats_inc(&stats->reads);
+	else
+		u64_stats_inc(&stats->writes);
+out:
+	u64_stats_update_end(&stats->syncp);
+}
+
 /**
  * __mdiobus_read - Unlocked version of the mdiobus_read function
  * @bus: the mii_bus struct
@@ -549,6 +775,7 @@ int __mdiobus_read(struct mii_bus *bus, int addr, u32 regnum)
 	retval = bus->read(bus, addr, regnum);
 
 	trace_mdio_access(bus, 1, addr, regnum, retval, retval);
+	mdiobus_stats_acct(&bus->stats[addr], true, retval);
 
 	return retval;
 }
@@ -574,6 +801,7 @@ int __mdiobus_write(struct mii_bus *bus, int addr, u32 regnum, u16 val)
 	err = bus->write(bus, addr, regnum, val);
 
 	trace_mdio_access(bus, 0, addr, regnum, val, err);
+	mdiobus_stats_acct(&bus->stats[addr], false, err);
 
 	return err;
 }
@@ -719,8 +947,27 @@ static int mdio_uevent(struct device *dev, struct kobj_uevent_env *env)
 	return 0;
 }
 
+static struct attribute *mdio_bus_device_statistics_attrs[] = {
+	&dev_attr_mdio_bus_device_transfers.attr.attr,
+	&dev_attr_mdio_bus_device_errors.attr.attr,
+	&dev_attr_mdio_bus_device_writes.attr.attr,
+	&dev_attr_mdio_bus_device_reads.attr.attr,
+	NULL,
+};
+
+static const struct attribute_group mdio_bus_device_statistics_group = {
+	.name	= "statistics",
+	.attrs	= mdio_bus_device_statistics_attrs,
+};
+
+static const struct attribute_group *mdio_bus_dev_groups[] = {
+	&mdio_bus_device_statistics_group,
+	NULL,
+};
+
 struct bus_type mdio_bus_type = {
 	.name		= "mdio_bus",
+	.dev_groups	= mdio_bus_dev_groups,
 	.match		= mdio_bus_match,
 	.uevent		= mdio_uevent,
 };
diff --git a/include/linux/phy.h b/include/linux/phy.h
index 2929d0bc307f..99a87f02667f 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -22,6 +22,7 @@
 #include <linux/timer.h>
 #include <linux/workqueue.h>
 #include <linux/mod_devicetable.h>
+#include <linux/u64_stats_sync.h>
 
 #include <linux/atomic.h>
 
@@ -212,6 +213,15 @@ struct sfp_bus;
 struct sfp_upstream_ops;
 struct sk_buff;
 
+struct mdio_bus_stats {
+	u64_stats_t transfers;
+	u64_stats_t errors;
+	u64_stats_t writes;
+	u64_stats_t reads;
+	/* Must be last, add new statistics above */
+	struct u64_stats_sync syncp;
+};
+
 /*
  * The Bus class for PHYs.  Devices which provide access to
  * PHYs should register using this structure
@@ -224,6 +234,7 @@ struct mii_bus {
 	int (*read)(struct mii_bus *bus, int addr, int regnum);
 	int (*write)(struct mii_bus *bus, int addr, int regnum, u16 val);
 	int (*reset)(struct mii_bus *bus);
+	struct mdio_bus_stats stats[PHY_MAX_ADDR];
 
 	/*
 	 * A lock to ensure that only one thing can read/write
-- 
cgit v1.2.3-59-g8ed1b


From 1f399fc797b5f78df8a993effdc0ef04304fdcea Mon Sep 17 00:00:00 2001
From: Hongbo Yao <yaohongbo@huawei.com>
Date: Thu, 16 Jan 2020 21:14:04 +0800
Subject: drivers/net: netdevsim depends on INET

If CONFIG_INET is not set and CONFIG_NETDEVSIM=y.
Building drivers/net/netdevsim/fib.o will get the following error:

drivers/net/netdevsim/fib.o: In function `nsim_fib4_rt_hw_flags_set':
fib.c:(.text+0x12b): undefined reference to `fib_alias_hw_flags_set'
drivers/net/netdevsim/fib.o: In function `nsim_fib4_rt_destroy':
fib.c:(.text+0xb11): undefined reference to `free_fib_info'

Correct the Kconfig for netdevsim.

Reported-by: Hulk Robot <hulkci@huawei.com>
Fixes: 48bb9eb47b270 ("netdevsim: fib: Add dummy implementation for FIB offload")
Signed-off-by: Hongbo Yao <yaohongbo@huawei.com>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index e65ed4d0a7ad..dee79588d2b1 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -549,6 +549,7 @@ source "drivers/net/hyperv/Kconfig"
 config NETDEVSIM
 	tristate "Simulated networking device"
 	depends on DEBUG_FS
+	depends on INET
 	depends on IPV6 || IPV6=n
 	select NET_DEVLINK
 	help
-- 
cgit v1.2.3-59-g8ed1b


From aa63b9471bcaca7392df461caafd570a565d17f0 Mon Sep 17 00:00:00 2001
From: Alexandru Ardelean <alexandru.ardelean@analog.com>
Date: Thu, 16 Jan 2020 16:15:35 +0200
Subject: net: phy: adin: const-ify static data

Some bits of static data should have been made const from the start.
This change adds the const qualifier where appropriate.

Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/adin.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/phy/adin.c b/drivers/net/phy/adin.c
index cf5a391c93e6..1dca3e883df4 100644
--- a/drivers/net/phy/adin.c
+++ b/drivers/net/phy/adin.c
@@ -145,7 +145,7 @@ struct adin_clause45_mmd_map {
 	u16 adin_regnum;
 };
 
-static struct adin_clause45_mmd_map adin_clause45_mmd_map[] = {
+static const struct adin_clause45_mmd_map adin_clause45_mmd_map[] = {
 	{ MDIO_MMD_PCS,	MDIO_PCS_EEE_ABLE,	ADIN1300_EEE_CAP_REG },
 	{ MDIO_MMD_AN,	MDIO_AN_EEE_LPABLE,	ADIN1300_EEE_LPABLE_REG },
 	{ MDIO_MMD_AN,	MDIO_AN_EEE_ADV,	ADIN1300_EEE_ADV_REG },
@@ -159,7 +159,7 @@ struct adin_hw_stat {
 	u16 reg2;
 };
 
-static struct adin_hw_stat adin_hw_stats[] = {
+static const struct adin_hw_stat adin_hw_stats[] = {
 	{ "total_frames_checked_count",		0x940A, 0x940B }, /* hi + lo */
 	{ "length_error_frames_count",		0x940C },
 	{ "alignment_error_frames_count",	0x940D },
@@ -456,7 +456,7 @@ static int adin_phy_config_intr(struct phy_device *phydev)
 static int adin_cl45_to_adin_reg(struct phy_device *phydev, int devad,
 				 u16 cl45_regnum)
 {
-	struct adin_clause45_mmd_map *m;
+	const struct adin_clause45_mmd_map *m;
 	int i;
 
 	if (devad == MDIO_MMD_VEND1)
@@ -650,7 +650,7 @@ static void adin_get_strings(struct phy_device *phydev, u8 *data)
 }
 
 static int adin_read_mmd_stat_regs(struct phy_device *phydev,
-				   struct adin_hw_stat *stat,
+				   const struct adin_hw_stat *stat,
 				   u32 *val)
 {
 	int ret;
@@ -676,7 +676,7 @@ static int adin_read_mmd_stat_regs(struct phy_device *phydev,
 
 static u64 adin_get_stat(struct phy_device *phydev, int i)
 {
-	struct adin_hw_stat *stat = &adin_hw_stats[i];
+	const struct adin_hw_stat *stat = &adin_hw_stats[i];
 	struct adin_priv *priv = phydev->priv;
 	u32 val;
 	int ret;
-- 
cgit v1.2.3-59-g8ed1b


From 90f29f0eada4d60e1f6ae537502ddb2202b9540d Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Thu, 16 Jan 2020 19:18:28 +0200
Subject: enetc: Don't print from enetc_sched_speed_set when link goes down

It is not an error to unplug a cable from the ENETC port even with TSN
offloads, so don't spam the log with link-related messages from the
tc-taprio offload subsystem, a single notification is sufficient:

[10972.351859] fsl_enetc 0000:00:00.0 eno0: Qbv PSPEED set speed link down.
[10972.360241] fsl_enetc 0000:00:00.0 eno0: Link is Down

Fixes: 2e47cb415f0a ("enetc: update TSN Qbv PSPEED set according to adjust link speed")
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/freescale/enetc/enetc_qos.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ethernet/freescale/enetc/enetc_qos.c b/drivers/net/ethernet/freescale/enetc/enetc_qos.c
index 00382b7c5bd8..0c6bf3a55a9a 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc_qos.c
+++ b/drivers/net/ethernet/freescale/enetc/enetc_qos.c
@@ -36,7 +36,6 @@ void enetc_sched_speed_set(struct net_device *ndev)
 	case SPEED_10:
 	default:
 		pspeed = ENETC_PMR_PSPEED_10M;
-		netdev_err(ndev, "Qbv PSPEED set speed link down.\n");
 	}
 
 	priv->speed = speed;
-- 
cgit v1.2.3-59-g8ed1b


From 1d0ee02b35b5173f84f97912fd7c982bd063a1f5 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Thu, 16 Jan 2020 19:59:44 +0200
Subject: Documentation: Fix typo in devlink documentation

The driver is named "mlxsw", not "mlx5".

Fixes: d4255d75856f ("devlink: document info versions for each driver")
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/devlink/mlxsw.rst | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/Documentation/networking/devlink/mlxsw.rst b/Documentation/networking/devlink/mlxsw.rst
index ccba9769c651..5f9bb0a0616a 100644
--- a/Documentation/networking/devlink/mlxsw.rst
+++ b/Documentation/networking/devlink/mlxsw.rst
@@ -40,7 +40,7 @@ The ``mlxsw`` driver supports reloading via ``DEVLINK_CMD_RELOAD``
 Info versions
 =============
 
-The ``mlx5`` driver reports the following versions
+The ``mlxsw`` driver reports the following versions
 
 .. list-table:: devlink info versions implemented
    :widths: 5 5 90
-- 
cgit v1.2.3-59-g8ed1b


From 062a33b1ad7c8df9f8f03f7b01c18e31a794a01d Mon Sep 17 00:00:00 2001
From: Alex Marginean <alexandru.marginean@nxp.com>
Date: Thu, 16 Jan 2020 20:05:06 +0200
Subject: net: dsa: felix: Set USXGMII link based on BMSR, not LPA

At least some PHYs (AQR412) don't advertise copper-side link status
during system side AN.

So remove this duplicate assignment to pcs->link and rely on the
previous one for link state: the local indication from the MAC PCS.

Fixes: bdeced75b13f ("net: dsa: felix: Add PCS operations for PHYLINK")
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/ocelot/felix_vsc9959.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/dsa/ocelot/felix_vsc9959.c b/drivers/net/dsa/ocelot/felix_vsc9959.c
index 03482616faa7..1e82b0d72058 100644
--- a/drivers/net/dsa/ocelot/felix_vsc9959.c
+++ b/drivers/net/dsa/ocelot/felix_vsc9959.c
@@ -892,7 +892,6 @@ static void vsc9959_pcs_link_state_usxgmii(struct phy_device *pcs,
 		break;
 	}
 
-	pcs->link = USXGMII_LPA_LNKS(lpa);
 	if (USXGMII_LPA_DUPLEX(lpa))
 		pcs->duplex = DUPLEX_FULL;
 	else
-- 
cgit v1.2.3-59-g8ed1b


From 8c6123e150f26eda011bc284e7aff221ea4d306b Mon Sep 17 00:00:00 2001
From: Alex Marginean <alexandru.marginean@nxp.com>
Date: Thu, 16 Jan 2020 20:09:59 +0200
Subject: net: dsa: felix: Don't restart PCS SGMII AN if not needed

Some PHYs like VSC8234 don't like it when AN restarts on their system side
and they restart line side AN too, going into an endless link up/down loop.
Don't restart PCS AN if link is up already.

Although in theory this feedback loop should be possible with the other
in-band AN modes too, for some reason it was not seen with the VSC8514
QSGMII and AQR412 USXGMII PHYs. So keep this logic only for SGMII where
the problem was found.

Fixes: bdeced75b13f ("net: dsa: felix: Add PCS operations for PHYLINK")
Suggested-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/ocelot/felix_vsc9959.c | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/drivers/net/dsa/ocelot/felix_vsc9959.c b/drivers/net/dsa/ocelot/felix_vsc9959.c
index 1e82b0d72058..2c812b481778 100644
--- a/drivers/net/dsa/ocelot/felix_vsc9959.c
+++ b/drivers/net/dsa/ocelot/felix_vsc9959.c
@@ -645,6 +645,27 @@ static void vsc9959_pcs_init_sgmii(struct phy_device *pcs,
 				   const struct phylink_link_state *state)
 {
 	if (link_an_mode == MLO_AN_INBAND) {
+		int bmsr, bmcr;
+
+		/* Some PHYs like VSC8234 don't like it when AN restarts on
+		 * their system  side and they restart line side AN too, going
+		 * into an endless link up/down loop.  Don't restart PCS AN if
+		 * link is up already.
+		 * We do check that AN is enabled just in case this is the 1st
+		 * call, PCS detects a carrier but AN is disabled from power on
+		 * or by boot loader.
+		 */
+		bmcr = phy_read(pcs, MII_BMCR);
+		if (bmcr < 0)
+			return;
+
+		bmsr = phy_read(pcs, MII_BMSR);
+		if (bmsr < 0)
+			return;
+
+		if ((bmcr & BMCR_ANENABLE) && (bmsr & BMSR_LSTATUS))
+			return;
+
 		/* SGMII spec requires tx_config_Reg[15:0] to be exactly 0x4001
 		 * for the MAC PCS in order to acknowledge the AN.
 		 */
-- 
cgit v1.2.3-59-g8ed1b


From 37fe45ad126dd5e9ade8be7ae5010fa1072abbbc Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Thu, 16 Jan 2020 20:41:53 +0200
Subject: net: dsa: felix: Don't error out on disabled ports with no phy-mode

The felix_parse_ports_node function was tested only on device trees
where all ports were enabled. Fix this check so that the driver
continues to probe only with the ports where status is not "disabled",
as expected.

Fixes: bdeced75b13f ("net: dsa: felix: Add PCS operations for PHYLINK")
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/ocelot/felix.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/dsa/ocelot/felix.c b/drivers/net/dsa/ocelot/felix.c
index feccb6201660..269cc6953d47 100644
--- a/drivers/net/dsa/ocelot/felix.c
+++ b/drivers/net/dsa/ocelot/felix.c
@@ -323,7 +323,7 @@ static int felix_parse_ports_node(struct felix *felix,
 	struct device *dev = felix->ocelot.dev;
 	struct device_node *child;
 
-	for_each_child_of_node(ports_node, child) {
+	for_each_available_child_of_node(ports_node, child) {
 		phy_interface_t phy_mode;
 		u32 port;
 		int err;
-- 
cgit v1.2.3-59-g8ed1b


From 56f200c78ce4d94680a27a1ce97a29ebeb4f23e1 Mon Sep 17 00:00:00 2001
From: Guillaume Nault <gnault@redhat.com>
Date: Thu, 16 Jan 2020 21:16:46 +0100
Subject: netns: Constify exported functions

Mark function parameters as 'const' where possible.

Signed-off-by: Guillaume Nault <gnault@redhat.com>
Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/net_namespace.h | 10 +++++-----
 net/core/net_namespace.c    |  6 +++---
 2 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index b8ceaf0cd997..854d39ef1ca3 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -347,9 +347,9 @@ static inline struct net *read_pnet(const possible_net_t *pnet)
 #endif
 
 int peernet2id_alloc(struct net *net, struct net *peer, gfp_t gfp);
-int peernet2id(struct net *net, struct net *peer);
-bool peernet_has_id(struct net *net, struct net *peer);
-struct net *get_net_ns_by_id(struct net *net, int id);
+int peernet2id(const struct net *net, struct net *peer);
+bool peernet_has_id(const struct net *net, struct net *peer);
+struct net *get_net_ns_by_id(const struct net *net, int id);
 
 struct pernet_operations {
 	struct list_head list;
@@ -427,7 +427,7 @@ static inline void unregister_net_sysctl_table(struct ctl_table_header *header)
 }
 #endif
 
-static inline int rt_genid_ipv4(struct net *net)
+static inline int rt_genid_ipv4(const struct net *net)
 {
 	return atomic_read(&net->ipv4.rt_genid);
 }
@@ -459,7 +459,7 @@ static inline void rt_genid_bump_all(struct net *net)
 	rt_genid_bump_ipv6(net);
 }
 
-static inline int fnhe_genid(struct net *net)
+static inline int fnhe_genid(const struct net *net)
 {
 	return atomic_read(&net->fnhe_genid);
 }
diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c
index 6412c1fbfcb5..757cc1d084e7 100644
--- a/net/core/net_namespace.c
+++ b/net/core/net_namespace.c
@@ -268,7 +268,7 @@ int peernet2id_alloc(struct net *net, struct net *peer, gfp_t gfp)
 EXPORT_SYMBOL_GPL(peernet2id_alloc);
 
 /* This function returns, if assigned, the id of a peer netns. */
-int peernet2id(struct net *net, struct net *peer)
+int peernet2id(const struct net *net, struct net *peer)
 {
 	int id;
 
@@ -283,12 +283,12 @@ EXPORT_SYMBOL(peernet2id);
 /* This function returns true is the peer netns has an id assigned into the
  * current netns.
  */
-bool peernet_has_id(struct net *net, struct net *peer)
+bool peernet_has_id(const struct net *net, struct net *peer)
 {
 	return peernet2id(net, peer) >= 0;
 }
 
-struct net *get_net_ns_by_id(struct net *net, int id)
+struct net *get_net_ns_by_id(const struct net *net, int id)
 {
 	struct net *peer;
 
-- 
cgit v1.2.3-59-g8ed1b


From bc0eb9a333918d2cc4f42f7f4cf09eaf029ac9de Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Thu, 16 Jan 2020 22:07:58 -0800
Subject: libbpf: Fix error handling bug in btf_dump__new

Fix missing jump to error handling in btf_dump__new, found by Coverity static
code analysis.

Fixes: 9f81654eebe8 ("libbpf: Expose BTF-to-C type declaration emitting API")
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200117060801.1311525-2-andriin@fb.com
---
 tools/lib/bpf/btf_dump.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tools/lib/bpf/btf_dump.c b/tools/lib/bpf/btf_dump.c
index 885acebd4396..bd09ed1710f1 100644
--- a/tools/lib/bpf/btf_dump.c
+++ b/tools/lib/bpf/btf_dump.c
@@ -142,6 +142,7 @@ struct btf_dump *btf_dump__new(const struct btf *btf,
 	if (IS_ERR(d->type_names)) {
 		err = PTR_ERR(d->type_names);
 		d->type_names = NULL;
+		goto err;
 	}
 	d->ident_names = hashmap__new(str_hash_fn, str_equal_fn, NULL);
 	if (IS_ERR(d->ident_names)) {
-- 
cgit v1.2.3-59-g8ed1b


From b7d7f3e185b6e643dcd49d1486b11cade8519008 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Thu, 16 Jan 2020 22:07:59 -0800
Subject: libbpf: Simplify BTF initialization logic

Current implementation of bpf_object's BTF initialization is very convoluted
and thus prone to errors. It doesn't have to be like that. This patch
simplifies it significantly.

This code also triggered static analysis issues over logically dead code due
to redundant error checks. This simplification should fix that as well.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200117060801.1311525-3-andriin@fb.com
---
 tools/lib/bpf/libbpf.c | 19 ++++++-------------
 1 file changed, 6 insertions(+), 13 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 3afaca9bce1d..3b0b88c3377d 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -2297,16 +2297,18 @@ static int bpf_object__init_btf(struct bpf_object *obj,
 				Elf_Data *btf_data,
 				Elf_Data *btf_ext_data)
 {
-	bool btf_required = bpf_object__is_btf_mandatory(obj);
-	int err = 0;
+	int err = -ENOENT;
 
 	if (btf_data) {
 		obj->btf = btf__new(btf_data->d_buf, btf_data->d_size);
 		if (IS_ERR(obj->btf)) {
+			err = PTR_ERR(obj->btf);
+			obj->btf = NULL;
 			pr_warn("Error loading ELF section %s: %d.\n",
 				BTF_ELF_SEC, err);
 			goto out;
 		}
+		err = 0;
 	}
 	if (btf_ext_data) {
 		if (!obj->btf) {
@@ -2324,18 +2326,9 @@ static int bpf_object__init_btf(struct bpf_object *obj,
 		}
 	}
 out:
-	if (err || IS_ERR(obj->btf)) {
-		if (btf_required)
-			err = err ? : PTR_ERR(obj->btf);
-		else
-			err = 0;
-		if (!IS_ERR_OR_NULL(obj->btf))
-			btf__free(obj->btf);
-		obj->btf = NULL;
-	}
-	if (btf_required && !obj->btf) {
+	if (err && bpf_object__is_btf_mandatory(obj)) {
 		pr_warn("BTF is required, but is missing or corrupted.\n");
-		return err == 0 ? -ENOENT : err;
+		return err;
 	}
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From c701917e647c6aaee5e7bbb7a2c1ca99e8552c58 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Thu, 16 Jan 2020 22:08:00 -0800
Subject: libbpf: Fix potential multiplication overflow in mmap() size
 calculation

Prevent potential overflow performed in 32-bit integers, before assigning
result to size_t. Reported by LGTM static analysis.

Fixes: eba9c5f498a1 ("libbpf: Refactor global data map initialization")
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200117060801.1311525-4-andriin@fb.com
---
 tools/lib/bpf/libbpf.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 3b0b88c3377d..fc41c3f2e983 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -1283,7 +1283,7 @@ static size_t bpf_map_mmap_sz(const struct bpf_map *map)
 	long page_sz = sysconf(_SC_PAGE_SIZE);
 	size_t map_sz;
 
-	map_sz = roundup(map->def.value_size, 8) * map->def.max_entries;
+	map_sz = (size_t)roundup(map->def.value_size, 8) * map->def.max_entries;
 	map_sz = roundup(map_sz, page_sz);
 	return map_sz;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 93d85dc5651d8337e3681441df1d5dd4184784ab Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Sun, 29 Dec 2019 16:09:09 +0200
Subject: igc: Fix typo in a comment

Fix typo in a context descriptor comment

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index c359f3d9fb25..1e27389bbda8 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -880,7 +880,7 @@ static void igc_tx_ctxtdesc(struct igc_ring *tx_ring,
 	/* set bits to identify this as an advanced context descriptor */
 	type_tucmd |= IGC_TXD_CMD_DEXT | IGC_ADVTXD_DTYP_CTXT;
 
-	/* For 82575, context index must be unique per ring. */
+	/* For i225, context index must be unique per ring. */
 	if (test_bit(IGC_RING_FLAG_TX_CTX_IDX, &tx_ring->flags))
 		mss_l4len_idx |= tx_ring->reg_idx << 4;
 
-- 
cgit v1.2.3-59-g8ed1b


From 34e9c162f09e224e94e519643970f6d7d607c8fd Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Wed, 1 Jan 2020 08:49:13 +0200
Subject: igc: Remove unused definition

Remove the unused IGC_FUNC_0 definition and make the code cleaner

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_hw.h | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_hw.h b/drivers/net/ethernet/intel/igc/igc_hw.h
index 20f710645746..be7273c4b53f 100644
--- a/drivers/net/ethernet/intel/igc/igc_hw.h
+++ b/drivers/net/ethernet/intel/igc/igc_hw.h
@@ -22,8 +22,6 @@
 #define IGC_DEV_ID_I220_V			0x15F7
 #define IGC_DEV_ID_I225_K			0x3100
 
-#define IGC_FUNC_0				0
-
 /* Function pointers for the MAC. */
 struct igc_mac_operations {
 	s32 (*check_for_link)(struct igc_hw *hw);
-- 
cgit v1.2.3-59-g8ed1b


From 0e7d4b93b0de94386be44d2a7ffa417a5c4adc1b Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Wed, 1 Jan 2020 09:14:17 +0200
Subject: igc: Add SKU for i225 device

Add support for blank NVM SKU

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_base.c | 1 +
 drivers/net/ethernet/intel/igc/igc_hw.h   | 1 +
 drivers/net/ethernet/intel/igc/igc_main.c | 1 +
 3 files changed, 3 insertions(+)

diff --git a/drivers/net/ethernet/intel/igc/igc_base.c b/drivers/net/ethernet/intel/igc/igc_base.c
index db289bcce21d..5a506440560a 100644
--- a/drivers/net/ethernet/intel/igc/igc_base.c
+++ b/drivers/net/ethernet/intel/igc/igc_base.c
@@ -212,6 +212,7 @@ static s32 igc_get_invariants_base(struct igc_hw *hw)
 	case IGC_DEV_ID_I225_I:
 	case IGC_DEV_ID_I220_V:
 	case IGC_DEV_ID_I225_K:
+	case IGC_DEV_ID_I225_BLANK_NVM:
 		mac->type = igc_i225;
 		break;
 	default:
diff --git a/drivers/net/ethernet/intel/igc/igc_hw.h b/drivers/net/ethernet/intel/igc/igc_hw.h
index be7273c4b53f..90ac0e0144d8 100644
--- a/drivers/net/ethernet/intel/igc/igc_hw.h
+++ b/drivers/net/ethernet/intel/igc/igc_hw.h
@@ -21,6 +21,7 @@
 #define IGC_DEV_ID_I225_I			0x15F8
 #define IGC_DEV_ID_I220_V			0x15F7
 #define IGC_DEV_ID_I225_K			0x3100
+#define IGC_DEV_ID_I225_BLANK_NVM		0x15FD
 
 /* Function pointers for the MAC. */
 struct igc_mac_operations {
diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 1e27389bbda8..a8f8d279f3b1 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -45,6 +45,7 @@ static const struct pci_device_id igc_pci_tbl[] = {
 	{ PCI_VDEVICE(INTEL, IGC_DEV_ID_I225_I), board_base },
 	{ PCI_VDEVICE(INTEL, IGC_DEV_ID_I220_V), board_base },
 	{ PCI_VDEVICE(INTEL, IGC_DEV_ID_I225_K), board_base },
+	{ PCI_VDEVICE(INTEL, IGC_DEV_ID_I225_BLANK_NVM), board_base },
 	/* required last entry */
 	{0, }
 };
-- 
cgit v1.2.3-59-g8ed1b


From f38b782dccabb2ada5c5b3c0dfb5ef0758a67ce0 Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Sun, 5 Jan 2020 14:20:53 +0200
Subject: igc: Add support for TSO

TCP segmentation offload allows a device to segment a single frame
into multiple frames with a data payload size specified in socket buffer.
As a result we can now send data approximately up to seven percents fast
than was previously possible on my system.

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_defines.h |   4 +
 drivers/net/ethernet/intel/igc/igc_main.c    | 113 ++++++++++++++++++++++++++-
 2 files changed, 116 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_defines.h b/drivers/net/ethernet/intel/igc/igc_defines.h
index 2121fc34e300..9e34b0969322 100644
--- a/drivers/net/ethernet/intel/igc/igc_defines.h
+++ b/drivers/net/ethernet/intel/igc/igc_defines.h
@@ -282,6 +282,10 @@
 #define IGC_TXD_STAT_TC		0x00000004 /* Tx Underrun */
 #define IGC_TXD_EXTCMD_TSTAMP	0x00000010 /* IEEE1588 Timestamp packet */
 
+/* IPSec Encrypt Enable */
+#define IGC_ADVTXD_L4LEN_SHIFT	8  /* Adv ctxt L4LEN shift */
+#define IGC_ADVTXD_MSS_SHIFT	16 /* Adv ctxt MSS shift */
+
 /* Transmit Control */
 #define IGC_TCTL_EN		0x00000002 /* enable Tx */
 #define IGC_TCTL_PSP		0x00000008 /* pad short packets */
diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index a8f8d279f3b1..d9d5425fe8d9 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -1000,6 +1000,10 @@ static u32 igc_tx_cmd_type(struct sk_buff *skb, u32 tx_flags)
 		       IGC_ADVTXD_DCMD_DEXT |
 		       IGC_ADVTXD_DCMD_IFCS;
 
+	/* set segmentation bits for TSO */
+	cmd_type |= IGC_SET_FLAG(tx_flags, IGC_TX_FLAGS_TSO,
+				 (IGC_ADVTXD_DCMD_TSE));
+
 	/* set timestamp bit if present */
 	cmd_type |= IGC_SET_FLAG(tx_flags, IGC_TX_FLAGS_TSTAMP,
 				 (IGC_ADVTXD_MAC_TSTAMP));
@@ -1171,6 +1175,100 @@ dma_error:
 	return -1;
 }
 
+static int igc_tso(struct igc_ring *tx_ring,
+		   struct igc_tx_buffer *first,
+		   u8 *hdr_len)
+{
+	u32 vlan_macip_lens, type_tucmd, mss_l4len_idx;
+	struct sk_buff *skb = first->skb;
+	union {
+		struct iphdr *v4;
+		struct ipv6hdr *v6;
+		unsigned char *hdr;
+	} ip;
+	union {
+		struct tcphdr *tcp;
+		struct udphdr *udp;
+		unsigned char *hdr;
+	} l4;
+	u32 paylen, l4_offset;
+	int err;
+
+	if (skb->ip_summed != CHECKSUM_PARTIAL)
+		return 0;
+
+	if (!skb_is_gso(skb))
+		return 0;
+
+	err = skb_cow_head(skb, 0);
+	if (err < 0)
+		return err;
+
+	ip.hdr = skb_network_header(skb);
+	l4.hdr = skb_checksum_start(skb);
+
+	/* ADV DTYP TUCMD MKRLOC/ISCSIHEDLEN */
+	type_tucmd = IGC_ADVTXD_TUCMD_L4T_TCP;
+
+	/* initialize outer IP header fields */
+	if (ip.v4->version == 4) {
+		unsigned char *csum_start = skb_checksum_start(skb);
+		unsigned char *trans_start = ip.hdr + (ip.v4->ihl * 4);
+
+		/* IP header will have to cancel out any data that
+		 * is not a part of the outer IP header
+		 */
+		ip.v4->check = csum_fold(csum_partial(trans_start,
+						      csum_start - trans_start,
+						      0));
+		type_tucmd |= IGC_ADVTXD_TUCMD_IPV4;
+
+		ip.v4->tot_len = 0;
+		first->tx_flags |= IGC_TX_FLAGS_TSO |
+				   IGC_TX_FLAGS_CSUM |
+				   IGC_TX_FLAGS_IPV4;
+	} else {
+		ip.v6->payload_len = 0;
+		first->tx_flags |= IGC_TX_FLAGS_TSO |
+				   IGC_TX_FLAGS_CSUM;
+	}
+
+	/* determine offset of inner transport header */
+	l4_offset = l4.hdr - skb->data;
+
+	/* remove payload length from inner checksum */
+	paylen = skb->len - l4_offset;
+	if (type_tucmd & IGC_ADVTXD_TUCMD_L4T_TCP) {
+		/* compute length of segmentation header */
+		*hdr_len = (l4.tcp->doff * 4) + l4_offset;
+		csum_replace_by_diff(&l4.tcp->check,
+				     (__force __wsum)htonl(paylen));
+	} else {
+		/* compute length of segmentation header */
+		*hdr_len = sizeof(*l4.udp) + l4_offset;
+		csum_replace_by_diff(&l4.udp->check,
+				     (__force __wsum)htonl(paylen));
+	}
+
+	/* update gso size and bytecount with header size */
+	first->gso_segs = skb_shinfo(skb)->gso_segs;
+	first->bytecount += (first->gso_segs - 1) * *hdr_len;
+
+	/* MSS L4LEN IDX */
+	mss_l4len_idx = (*hdr_len - l4_offset) << IGC_ADVTXD_L4LEN_SHIFT;
+	mss_l4len_idx |= skb_shinfo(skb)->gso_size << IGC_ADVTXD_MSS_SHIFT;
+
+	/* VLAN MACLEN IPLEN */
+	vlan_macip_lens = l4.hdr - ip.hdr;
+	vlan_macip_lens |= (ip.hdr - skb->data) << IGC_ADVTXD_MACLEN_SHIFT;
+	vlan_macip_lens |= first->tx_flags & IGC_TX_FLAGS_VLAN_MASK;
+
+	igc_tx_ctxtdesc(tx_ring, first, vlan_macip_lens,
+			type_tucmd, mss_l4len_idx);
+
+	return 1;
+}
+
 static netdev_tx_t igc_xmit_frame_ring(struct sk_buff *skb,
 				       struct igc_ring *tx_ring)
 {
@@ -1180,6 +1278,7 @@ static netdev_tx_t igc_xmit_frame_ring(struct sk_buff *skb,
 	u32 tx_flags = 0;
 	unsigned short f;
 	u8 hdr_len = 0;
+	int tso = 0;
 
 	/* need: 1 descriptor per page * PAGE_SIZE/IGC_MAX_DATA_PER_TXD,
 	 *	+ 1 desc for skb_headlen/IGC_MAX_DATA_PER_TXD,
@@ -1226,10 +1325,20 @@ static netdev_tx_t igc_xmit_frame_ring(struct sk_buff *skb,
 	first->tx_flags = tx_flags;
 	first->protocol = protocol;
 
-	igc_tx_csum(tx_ring, first);
+	tso = igc_tso(tx_ring, first, &hdr_len);
+	if (tso < 0)
+		goto out_drop;
+	else if (!tso)
+		igc_tx_csum(tx_ring, first);
 
 	igc_tx_map(tx_ring, first, hdr_len);
 
+	return NETDEV_TX_OK;
+
+out_drop:
+	dev_kfree_skb_any(first->skb);
+	first->skb = NULL;
+
 	return NETDEV_TX_OK;
 }
 
@@ -4589,6 +4698,8 @@ static int igc_probe(struct pci_dev *pdev,
 
 	/* Add supported features to the features list*/
 	netdev->features |= NETIF_F_SG;
+	netdev->features |= NETIF_F_TSO;
+	netdev->features |= NETIF_F_TSO6;
 	netdev->features |= NETIF_F_RXCSUM;
 	netdev->features |= NETIF_F_HW_CSUM;
 	netdev->features |= NETIF_F_SCTP_CRC;
-- 
cgit v1.2.3-59-g8ed1b


From bcb3244ccdc4cebb6ed3d47d973da7b1a03837c0 Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Wed, 8 Jan 2020 10:19:24 +0200
Subject: igc: Add PHY power management control

PHY power management control should provide a reliable and accurate
indication of PHY reset completion and decrease the delay time
after a PHY reset

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_defines.h |  1 +
 drivers/net/ethernet/intel/igc/igc_phy.c     | 16 +++++++++++++++-
 drivers/net/ethernet/intel/igc/igc_regs.h    |  1 +
 3 files changed, 17 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_defines.h b/drivers/net/ethernet/intel/igc/igc_defines.h
index 9e34b0969322..58efa7a02c68 100644
--- a/drivers/net/ethernet/intel/igc/igc_defines.h
+++ b/drivers/net/ethernet/intel/igc/igc_defines.h
@@ -464,6 +464,7 @@
 /* PHY Status Register */
 #define MII_SR_LINK_STATUS	0x0004 /* Link Status 1 = link */
 #define MII_SR_AUTONEG_COMPLETE	0x0020 /* Auto Neg Complete */
+#define IGC_PHY_RST_COMP	0x0100 /* Internal PHY reset completion */
 
 /* PHY 1000 MII Register/Bit Definitions */
 /* PHY Registers defined by IEEE */
diff --git a/drivers/net/ethernet/intel/igc/igc_phy.c b/drivers/net/ethernet/intel/igc/igc_phy.c
index f4b05af0dd2f..8e1799508edc 100644
--- a/drivers/net/ethernet/intel/igc/igc_phy.c
+++ b/drivers/net/ethernet/intel/igc/igc_phy.c
@@ -173,6 +173,7 @@ s32 igc_check_downshift(struct igc_hw *hw)
 s32 igc_phy_hw_reset(struct igc_hw *hw)
 {
 	struct igc_phy_info *phy = &hw->phy;
+	u32 phpm = 0, timeout = 10000;
 	s32  ret_val;
 	u32 ctrl;
 
@@ -186,6 +187,8 @@ s32 igc_phy_hw_reset(struct igc_hw *hw)
 	if (ret_val)
 		goto out;
 
+	phpm = rd32(IGC_I225_PHPM);
+
 	ctrl = rd32(IGC_CTRL);
 	wr32(IGC_CTRL, ctrl | IGC_CTRL_PHY_RST);
 	wrfl();
@@ -195,7 +198,18 @@ s32 igc_phy_hw_reset(struct igc_hw *hw)
 	wr32(IGC_CTRL, ctrl);
 	wrfl();
 
-	usleep_range(1500, 2000);
+	/* SW should guarantee 100us for the completion of the PHY reset */
+	usleep_range(100, 150);
+	do {
+		phpm = rd32(IGC_I225_PHPM);
+		timeout--;
+		udelay(1);
+	} while (!(phpm & IGC_PHY_RST_COMP) && timeout);
+
+	if (!timeout)
+		hw_dbg("Timeout is expired after a phy reset\n");
+
+	usleep_range(100, 150);
 
 	phy->ops.release(hw);
 
diff --git a/drivers/net/ethernet/intel/igc/igc_regs.h b/drivers/net/ethernet/intel/igc/igc_regs.h
index c82111051898..c9029b549b90 100644
--- a/drivers/net/ethernet/intel/igc/igc_regs.h
+++ b/drivers/net/ethernet/intel/igc/igc_regs.h
@@ -12,6 +12,7 @@
 #define IGC_MDIC		0x00020  /* MDI Control - RW */
 #define IGC_MDICNFG		0x00E04  /* MDC/MDIO Configuration - RW */
 #define IGC_CONNSW		0x00034  /* Copper/Fiber switch control - RW */
+#define IGC_I225_PHPM		0x00E14  /* I225 PHY Power Management */
 
 /* Internal Packet Buffer Size Registers */
 #define IGC_RXPBS		0x02404  /* Rx Packet Buffer Size - RW */
-- 
cgit v1.2.3-59-g8ed1b


From 2da259c5fe1f528dd9bd4d0430a2732de18c190b Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Thu, 19 Dec 2019 12:10:00 -0800
Subject: fm10k: use txqueue parameter in fm10k_tx_timeout

Make use of the new txqueue parameter to the .ndo_tx_timeout function.
In fm10k_tx_timeout, remove the now unnecessary loop to determine which
Tx queue is stuck. Instead, just double check the specified queue

This could be improved further to attempt resetting only the specific
queue that got stuck. However, that is a much larger refactor and has
been left as a future improvement.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/fm10k/fm10k_netdev.c | 17 ++++++++++-------
 1 file changed, 10 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c b/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c
index ba2566e2123d..0637ccadee79 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c
@@ -696,21 +696,24 @@ static netdev_tx_t fm10k_xmit_frame(struct sk_buff *skb, struct net_device *dev)
 /**
  * fm10k_tx_timeout - Respond to a Tx Hang
  * @netdev: network interface device structure
+ * @txqueue: the index of the Tx queue that timed out
  **/
 static void fm10k_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct fm10k_intfc *interface = netdev_priv(netdev);
+	struct fm10k_ring *tx_ring;
 	bool real_tx_hang = false;
-	int i;
-
-#define TX_TIMEO_LIMIT 16000
-	for (i = 0; i < interface->num_tx_queues; i++) {
-		struct fm10k_ring *tx_ring = interface->tx_ring[i];
 
-		if (check_for_tx_hang(tx_ring) && fm10k_check_tx_hang(tx_ring))
-			real_tx_hang = true;
+	if (txqueue >= interface->num_tx_queues) {
+		WARN(1, "invalid Tx queue index %d", txqueue);
+		return;
 	}
 
+	tx_ring = interface->tx_ring[txqueue];
+	if (check_for_tx_hang(tx_ring) && fm10k_check_tx_hang(tx_ring))
+		real_tx_hang = true;
+
+#define TX_TIMEO_LIMIT 16000
 	if (real_tx_hang) {
 		fm10k_tx_timeout_reset(interface);
 	} else {
-- 
cgit v1.2.3-59-g8ed1b


From 65279beec34565a3d9d4d098ea7f602ffc419695 Mon Sep 17 00:00:00 2001
From: Julio Faracco <jcfaracco@gmail.com>
Date: Wed, 18 Dec 2019 15:38:44 -0300
Subject: i40e: Removing hung_queue variable to use txqueue function parameter

The scope of function .ndo_tx_timeout was changed to include the hang
queue when a TX timeout event occurs. See commit 0290bd291cc0
("netdev: pass the stuck queue to the timeout handler") for more
details. Now, drivers don't need to identify which queue is stopped.
Drivers can simply use the queue index provided by dev_watchdog and
execute all actions needed to restore network traffic. This commit do
some cleanups into Intel i40e driver to remove a redundant loop to find
stopped queue.

Signed-off-by: Julio Faracco <jcfaracco@gmail.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_main.c | 41 ++++++++---------------------
 1 file changed, 11 insertions(+), 30 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 33912cf964eb..8c3e753bfb9d 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -307,37 +307,18 @@ static void i40e_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 	struct i40e_vsi *vsi = np->vsi;
 	struct i40e_pf *pf = vsi->back;
 	struct i40e_ring *tx_ring = NULL;
-	unsigned int i, hung_queue = 0;
+	unsigned int i;
 	u32 head, val;
 
 	pf->tx_timeout_count++;
 
-	/* find the stopped queue the same way the stack does */
-	for (i = 0; i < netdev->num_tx_queues; i++) {
-		struct netdev_queue *q;
-		unsigned long trans_start;
-
-		q = netdev_get_tx_queue(netdev, i);
-		trans_start = q->trans_start;
-		if (netif_xmit_stopped(q) &&
-		    time_after(jiffies,
-			       (trans_start + netdev->watchdog_timeo))) {
-			hung_queue = i;
-			break;
-		}
-	}
-
-	if (i == netdev->num_tx_queues) {
-		netdev_info(netdev, "tx_timeout: no netdev hung queue found\n");
-	} else {
-		/* now that we have an index, find the tx_ring struct */
-		for (i = 0; i < vsi->num_queue_pairs; i++) {
-			if (vsi->tx_rings[i] && vsi->tx_rings[i]->desc) {
-				if (hung_queue ==
-				    vsi->tx_rings[i]->queue_index) {
-					tx_ring = vsi->tx_rings[i];
-					break;
-				}
+	/* with txqueue index, find the tx_ring struct */
+	for (i = 0; i < vsi->num_queue_pairs; i++) {
+		if (vsi->tx_rings[i] && vsi->tx_rings[i]->desc) {
+			if (txqueue ==
+			    vsi->tx_rings[i]->queue_index) {
+				tx_ring = vsi->tx_rings[i];
+				break;
 			}
 		}
 	}
@@ -363,14 +344,14 @@ static void i40e_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 			val = rd32(&pf->hw, I40E_PFINT_DYN_CTL0);
 
 		netdev_info(netdev, "tx_timeout: VSI_seid: %d, Q %d, NTC: 0x%x, HWB: 0x%x, NTU: 0x%x, TAIL: 0x%x, INT: 0x%x\n",
-			    vsi->seid, hung_queue, tx_ring->next_to_clean,
+			    vsi->seid, txqueue, tx_ring->next_to_clean,
 			    head, tx_ring->next_to_use,
 			    readl(tx_ring->tail), val);
 	}
 
 	pf->tx_timeout_last_recovery = jiffies;
-	netdev_info(netdev, "tx_timeout recovery level %d, hung_queue %d\n",
-		    pf->tx_timeout_recovery_level, hung_queue);
+	netdev_info(netdev, "tx_timeout recovery level %d, txqueue %d\n",
+		    pf->tx_timeout_recovery_level, txqueue);
 
 	switch (pf->tx_timeout_recovery_level) {
 	case 1:
-- 
cgit v1.2.3-59-g8ed1b


From ed5a3f664c557e2ee5d92f75398a0fc8ab522ad5 Mon Sep 17 00:00:00 2001
From: Julio Faracco <jcfaracco@gmail.com>
Date: Wed, 18 Dec 2019 15:38:45 -0300
Subject: ice: Removing hung_queue variable to use txqueue function parameter

The scope of function .ndo_tx_timeout was changed to include the hang
queue when a TX timeout event occurs. See commit 0290bd291cc0
("netdev: pass the stuck queue to the timeout handler") for more
details. Now, drivers don't need to identify which queue is stopped.
Drivers can simply use the queue index provided by dev_watchdog and
execute all actions needed to restore network traffic. This commit do
some cleanups into Intel ice driver to remove a redundant loop to find
stopped queue.

Signed-off-by: Julio Faracco <jcfaracco@gmail.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_main.c | 41 +++++++++----------------------
 1 file changed, 11 insertions(+), 30 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index bf539483e25e..eb9d00608e9a 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -5086,36 +5086,17 @@ static void ice_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 	struct ice_ring *tx_ring = NULL;
 	struct ice_vsi *vsi = np->vsi;
 	struct ice_pf *pf = vsi->back;
-	int hung_queue = -1;
 	u32 i;
 
 	pf->tx_timeout_count++;
 
-	/* find the stopped queue the same way dev_watchdog() does */
-	for (i = 0; i < netdev->num_tx_queues; i++) {
-		unsigned long trans_start;
-		struct netdev_queue *q;
-
-		q = netdev_get_tx_queue(netdev, i);
-		trans_start = q->trans_start;
-		if (netif_xmit_stopped(q) &&
-		    time_after(jiffies,
-			       trans_start + netdev->watchdog_timeo)) {
-			hung_queue = i;
-			break;
-		}
-	}
-
-	if (i == netdev->num_tx_queues)
-		netdev_info(netdev, "tx_timeout: no netdev hung queue found\n");
-	else
-		/* now that we have an index, find the tx_ring struct */
-		for (i = 0; i < vsi->num_txq; i++)
-			if (vsi->tx_rings[i] && vsi->tx_rings[i]->desc)
-				if (hung_queue == vsi->tx_rings[i]->q_index) {
-					tx_ring = vsi->tx_rings[i];
-					break;
-				}
+	/* now that we have an index, find the tx_ring struct */
+	for (i = 0; i < vsi->num_txq; i++)
+		if (vsi->tx_rings[i] && vsi->tx_rings[i]->desc)
+			if (txqueue == vsi->tx_rings[i]->q_index) {
+				tx_ring = vsi->tx_rings[i];
+				break;
+			}
 
 	/* Reset recovery level if enough time has elapsed after last timeout.
 	 * Also ensure no new reset action happens before next timeout period.
@@ -5130,19 +5111,19 @@ static void ice_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 		struct ice_hw *hw = &pf->hw;
 		u32 head, val = 0;
 
-		head = (rd32(hw, QTX_COMM_HEAD(vsi->txq_map[hung_queue])) &
+		head = (rd32(hw, QTX_COMM_HEAD(vsi->txq_map[txqueue])) &
 			QTX_COMM_HEAD_HEAD_M) >> QTX_COMM_HEAD_HEAD_S;
 		/* Read interrupt register */
 		val = rd32(hw, GLINT_DYN_CTL(tx_ring->q_vector->reg_idx));
 
 		netdev_info(netdev, "tx_timeout: VSI_num: %d, Q %d, NTC: 0x%x, HW_HEAD: 0x%x, NTU: 0x%x, INT: 0x%x\n",
-			    vsi->vsi_num, hung_queue, tx_ring->next_to_clean,
+			    vsi->vsi_num, txqueue, tx_ring->next_to_clean,
 			    head, tx_ring->next_to_use, val);
 	}
 
 	pf->tx_timeout_last_recovery = jiffies;
-	netdev_info(netdev, "tx_timeout recovery level %d, hung_queue %d\n",
-		    pf->tx_timeout_recovery_level, hung_queue);
+	netdev_info(netdev, "tx_timeout recovery level %d, txqueue %d\n",
+		    pf->tx_timeout_recovery_level, txqueue);
 
 	switch (pf->tx_timeout_recovery_level) {
 	case 1:
-- 
cgit v1.2.3-59-g8ed1b


From 102d412a3d565a105f16e883169239dd66b680b5 Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.king@canonical.com>
Date: Thu, 19 Dec 2019 23:45:35 +0000
Subject: ice: remove redundant assignment to variable xmit_done

The variable xmit_done is being initialized with a value that is never
read and it is being updated later with a new value. The initialization
is redundant and can be removed.

Addresses-Coverity: ("Unused value")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_xsk.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_xsk.c b/drivers/net/ethernet/intel/ice/ice_xsk.c
index 0e401f116d54..149dca0012ba 100644
--- a/drivers/net/ethernet/intel/ice/ice_xsk.c
+++ b/drivers/net/ethernet/intel/ice/ice_xsk.c
@@ -1020,8 +1020,8 @@ bool ice_clean_tx_irq_zc(struct ice_ring *xdp_ring, int budget)
 	s16 ntc = xdp_ring->next_to_clean;
 	struct ice_tx_desc *tx_desc;
 	struct ice_tx_buf *tx_buf;
-	bool xmit_done = true;
 	u32 xsk_frames = 0;
+	bool xmit_done;
 
 	tx_desc = ICE_TX_DESC(xdp_ring, ntc);
 	tx_buf = &xdp_ring->tx_buf[ntc];
-- 
cgit v1.2.3-59-g8ed1b


From a6ed02cac690b635dbb938690e795812ce1e14ca Mon Sep 17 00:00:00 2001
From: KP Singh <kpsingh@google.com>
Date: Fri, 17 Jan 2020 22:28:25 +0100
Subject: libbpf: Load btf_vmlinux only once per object.

As more programs (TRACING, STRUCT_OPS, and upcoming LSM) use vmlinux
BTF information, loading the BTF vmlinux information for every program
in an object is sub-optimal. The fix was originally proposed in:

   https://lore.kernel.org/bpf/CAEf4BzZodr3LKJuM7QwD38BiEH02Cc1UbtnGpVkCJ00Mf+V_Qg@mail.gmail.com/

The btf_vmlinux is populated in the object if any of the programs in
the object requires it just before the programs are loaded and freed
after the programs finish loading.

Reported-by: Andrii Nakryiko <andrii.nakryiko@gmail.com>
Signed-off-by: KP Singh <kpsingh@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Reviewed-by: Brendan Jackman <jackmanb@chromium.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/20200117212825.11755-1-kpsingh@chromium.org
---
 tools/lib/bpf/libbpf.c | 153 ++++++++++++++++++++++++++++++++-----------------
 1 file changed, 101 insertions(+), 52 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index fc41c3f2e983..faab96a42141 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -385,6 +385,10 @@ struct bpf_object {
 	struct list_head list;
 
 	struct btf *btf;
+	/* Parse and load BTF vmlinux if any of the programs in the object need
+	 * it at load time.
+	 */
+	struct btf *btf_vmlinux;
 	struct btf_ext *btf_ext;
 
 	void *priv;
@@ -633,7 +637,8 @@ find_member_by_name(const struct btf *btf, const struct btf_type *t,
 }
 
 #define STRUCT_OPS_VALUE_PREFIX "bpf_struct_ops_"
-#define STRUCT_OPS_VALUE_PREFIX_LEN (sizeof(STRUCT_OPS_VALUE_PREFIX) - 1)
+static int find_btf_by_prefix_kind(const struct btf *btf, const char *prefix,
+				   const char *name, __u32 kind);
 
 static int
 find_struct_ops_kern_types(const struct btf *btf, const char *tname,
@@ -644,7 +649,6 @@ find_struct_ops_kern_types(const struct btf *btf, const char *tname,
 	const struct btf_type *kern_type, *kern_vtype;
 	const struct btf_member *kern_data_member;
 	__s32 kern_vtype_id, kern_type_id;
-	char vtname[128] = STRUCT_OPS_VALUE_PREFIX;
 	__u32 i;
 
 	kern_type_id = btf__find_by_name_kind(btf, tname, BTF_KIND_STRUCT);
@@ -660,13 +664,11 @@ find_struct_ops_kern_types(const struct btf *btf, const char *tname,
 	 * find "struct bpf_struct_ops_tcp_congestion_ops" from the
 	 * btf_vmlinux.
 	 */
-	strncat(vtname + STRUCT_OPS_VALUE_PREFIX_LEN, tname,
-		sizeof(vtname) - STRUCT_OPS_VALUE_PREFIX_LEN - 1);
-	kern_vtype_id = btf__find_by_name_kind(btf, vtname,
-					       BTF_KIND_STRUCT);
+	kern_vtype_id = find_btf_by_prefix_kind(btf, STRUCT_OPS_VALUE_PREFIX,
+						tname, BTF_KIND_STRUCT);
 	if (kern_vtype_id < 0) {
-		pr_warn("struct_ops init_kern: struct %s is not found in kernel BTF\n",
-			vtname);
+		pr_warn("struct_ops init_kern: struct %s%s is not found in kernel BTF\n",
+			STRUCT_OPS_VALUE_PREFIX, tname);
 		return kern_vtype_id;
 	}
 	kern_vtype = btf__type_by_id(btf, kern_vtype_id);
@@ -683,8 +685,8 @@ find_struct_ops_kern_types(const struct btf *btf, const char *tname,
 			break;
 	}
 	if (i == btf_vlen(kern_vtype)) {
-		pr_warn("struct_ops init_kern: struct %s data is not found in struct %s\n",
-			tname, vtname);
+		pr_warn("struct_ops init_kern: struct %s data is not found in struct %s%s\n",
+			tname, STRUCT_OPS_VALUE_PREFIX, tname);
 		return -EINVAL;
 	}
 
@@ -835,7 +837,6 @@ static int bpf_map__init_kern_struct_ops(struct bpf_map *map,
 
 static int bpf_object__init_kern_struct_ops_maps(struct bpf_object *obj)
 {
-	struct btf *kern_btf = NULL;
 	struct bpf_map *map;
 	size_t i;
 	int err;
@@ -846,20 +847,12 @@ static int bpf_object__init_kern_struct_ops_maps(struct bpf_object *obj)
 		if (!bpf_map__is_struct_ops(map))
 			continue;
 
-		if (!kern_btf) {
-			kern_btf = libbpf_find_kernel_btf();
-			if (IS_ERR(kern_btf))
-				return PTR_ERR(kern_btf);
-		}
-
-		err = bpf_map__init_kern_struct_ops(map, obj->btf, kern_btf);
-		if (err) {
-			btf__free(kern_btf);
+		err = bpf_map__init_kern_struct_ops(map, obj->btf,
+						    obj->btf_vmlinux);
+		if (err)
 			return err;
-		}
 	}
 
-	btf__free(kern_btf);
 	return 0;
 }
 
@@ -2357,6 +2350,41 @@ static int bpf_object__finalize_btf(struct bpf_object *obj)
 	return 0;
 }
 
+static inline bool libbpf_prog_needs_vmlinux_btf(struct bpf_program *prog)
+{
+	if (prog->type == BPF_PROG_TYPE_STRUCT_OPS)
+		return true;
+
+	/* BPF_PROG_TYPE_TRACING programs which do not attach to other programs
+	 * also need vmlinux BTF
+	 */
+	if (prog->type == BPF_PROG_TYPE_TRACING && !prog->attach_prog_fd)
+		return true;
+
+	return false;
+}
+
+static int bpf_object__load_vmlinux_btf(struct bpf_object *obj)
+{
+	struct bpf_program *prog;
+	int err;
+
+	bpf_object__for_each_program(prog, obj) {
+		if (libbpf_prog_needs_vmlinux_btf(prog)) {
+			obj->btf_vmlinux = libbpf_find_kernel_btf();
+			if (IS_ERR(obj->btf_vmlinux)) {
+				err = PTR_ERR(obj->btf_vmlinux);
+				pr_warn("Error loading vmlinux BTF: %d\n", err);
+				obj->btf_vmlinux = NULL;
+				return err;
+			}
+			return 0;
+		}
+	}
+
+	return 0;
+}
+
 static int bpf_object__sanitize_and_load_btf(struct bpf_object *obj)
 {
 	int err = 0;
@@ -4884,18 +4912,14 @@ out:
 	return ret;
 }
 
-static int libbpf_find_attach_btf_id(const char *name,
-				     enum bpf_attach_type attach_type,
-				     __u32 attach_prog_fd);
+static int libbpf_find_attach_btf_id(struct bpf_program *prog);
 
 int bpf_program__load(struct bpf_program *prog, char *license, __u32 kern_ver)
 {
 	int err = 0, fd, i, btf_id;
 
 	if (prog->type == BPF_PROG_TYPE_TRACING) {
-		btf_id = libbpf_find_attach_btf_id(prog->section_name,
-						   prog->expected_attach_type,
-						   prog->attach_prog_fd);
+		btf_id = libbpf_find_attach_btf_id(prog);
 		if (btf_id <= 0)
 			return btf_id;
 		prog->attach_btf_id = btf_id;
@@ -5273,10 +5297,15 @@ int bpf_object__load_xattr(struct bpf_object_load_attr *attr)
 	err = err ? : bpf_object__resolve_externs(obj, obj->kconfig);
 	err = err ? : bpf_object__sanitize_and_load_btf(obj);
 	err = err ? : bpf_object__sanitize_maps(obj);
+	err = err ? : bpf_object__load_vmlinux_btf(obj);
 	err = err ? : bpf_object__init_kern_struct_ops_maps(obj);
 	err = err ? : bpf_object__create_maps(obj);
 	err = err ? : bpf_object__relocate(obj, attr->target_btf_path);
 	err = err ? : bpf_object__load_progs(obj, attr->log_level);
+
+	btf__free(obj->btf_vmlinux);
+	obj->btf_vmlinux = NULL;
+
 	if (err)
 		goto out;
 
@@ -6497,34 +6526,52 @@ invalid_prog:
 	return -EINVAL;
 }
 
-#define BTF_PREFIX "btf_trace_"
+#define BTF_TRACE_PREFIX "btf_trace_"
+#define BTF_MAX_NAME_SIZE 128
+
+static int find_btf_by_prefix_kind(const struct btf *btf, const char *prefix,
+				   const char *name, __u32 kind)
+{
+	char btf_type_name[BTF_MAX_NAME_SIZE];
+	int ret;
+
+	ret = snprintf(btf_type_name, sizeof(btf_type_name),
+		       "%s%s", prefix, name);
+	/* snprintf returns the number of characters written excluding the
+	 * the terminating null. So, if >= BTF_MAX_NAME_SIZE are written, it
+	 * indicates truncation.
+	 */
+	if (ret < 0 || ret >= sizeof(btf_type_name))
+		return -ENAMETOOLONG;
+	return btf__find_by_name_kind(btf, btf_type_name, kind);
+}
+
+static inline int __find_vmlinux_btf_id(struct btf *btf, const char *name,
+					enum bpf_attach_type attach_type)
+{
+	int err;
+
+	if (attach_type == BPF_TRACE_RAW_TP)
+		err = find_btf_by_prefix_kind(btf, BTF_TRACE_PREFIX, name,
+					      BTF_KIND_TYPEDEF);
+	else
+		err = btf__find_by_name_kind(btf, name, BTF_KIND_FUNC);
+
+	return err;
+}
+
 int libbpf_find_vmlinux_btf_id(const char *name,
 			       enum bpf_attach_type attach_type)
 {
-	struct btf *btf = libbpf_find_kernel_btf();
-	char raw_tp_btf[128] = BTF_PREFIX;
-	char *dst = raw_tp_btf + sizeof(BTF_PREFIX) - 1;
-	const char *btf_name;
-	int err = -EINVAL;
-	__u32 kind;
+	struct btf *btf;
 
+	btf = libbpf_find_kernel_btf();
 	if (IS_ERR(btf)) {
 		pr_warn("vmlinux BTF is not found\n");
 		return -EINVAL;
 	}
 
-	if (attach_type == BPF_TRACE_RAW_TP) {
-		/* prepend "btf_trace_" prefix per kernel convention */
-		strncat(dst, name, sizeof(raw_tp_btf) - sizeof(BTF_PREFIX));
-		btf_name = raw_tp_btf;
-		kind = BTF_KIND_TYPEDEF;
-	} else {
-		btf_name = name;
-		kind = BTF_KIND_FUNC;
-	}
-	err = btf__find_by_name_kind(btf, btf_name, kind);
-	btf__free(btf);
-	return err;
+	return __find_vmlinux_btf_id(btf, name, attach_type);
 }
 
 static int libbpf_find_prog_btf_id(const char *name, __u32 attach_prog_fd)
@@ -6560,10 +6607,11 @@ out:
 	return err;
 }
 
-static int libbpf_find_attach_btf_id(const char *name,
-				     enum bpf_attach_type attach_type,
-				     __u32 attach_prog_fd)
+static int libbpf_find_attach_btf_id(struct bpf_program *prog)
 {
+	enum bpf_attach_type attach_type = prog->expected_attach_type;
+	__u32 attach_prog_fd = prog->attach_prog_fd;
+	const char *name = prog->section_name;
 	int i, err;
 
 	if (!name)
@@ -6578,8 +6626,9 @@ static int libbpf_find_attach_btf_id(const char *name,
 			err = libbpf_find_prog_btf_id(name + section_defs[i].len,
 						      attach_prog_fd);
 		else
-			err = libbpf_find_vmlinux_btf_id(name + section_defs[i].len,
-							 attach_type);
+			err = __find_vmlinux_btf_id(prog->obj->btf_vmlinux,
+						    name + section_defs[i].len,
+						    attach_type);
 		if (err <= 0)
 			pr_warn("%s is not found in vmlinux BTF\n", name);
 		return err;
-- 
cgit v1.2.3-59-g8ed1b


From 2eafa1746f17872483d1033b0116ec71435ea19d Mon Sep 17 00:00:00 2001
From: Hans Westgaard Ry <hans.westgaard.ry@oracle.com>
Date: Wed, 15 Jan 2020 14:43:39 +0200
Subject: net/rds: Handle ODP mr registration/unregistration

On-Demand-Paging MRs are registered using ib_reg_user_mr and
unregistered with ib_dereg_mr.

Signed-off-by: Hans Westgaard Ry <hans.westgaard.ry@oracle.com>
Acked-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
---
 net/rds/ib.c      |   7 +++
 net/rds/ib.h      |   3 +-
 net/rds/ib_mr.h   |   7 ++-
 net/rds/ib_rdma.c |  75 ++++++++++++++++++++++++++-
 net/rds/ib_send.c |  44 +++++++++++-----
 net/rds/rdma.c    | 151 ++++++++++++++++++++++++++++++++++++++++--------------
 net/rds/rds.h     |  13 ++++-
 7 files changed, 244 insertions(+), 56 deletions(-)

diff --git a/net/rds/ib.c b/net/rds/ib.c
index 3fd5f40189bd..a792d8a3872a 100644
--- a/net/rds/ib.c
+++ b/net/rds/ib.c
@@ -156,6 +156,13 @@ static void rds_ib_add_one(struct ib_device *device)
 	has_fmr = (device->ops.alloc_fmr && device->ops.dealloc_fmr &&
 		   device->ops.map_phys_fmr && device->ops.unmap_fmr);
 	rds_ibdev->use_fastreg = (has_fr && !has_fmr);
+	rds_ibdev->odp_capable =
+		!!(device->attrs.device_cap_flags &
+		   IB_DEVICE_ON_DEMAND_PAGING) &&
+		!!(device->attrs.odp_caps.per_transport_caps.rc_odp_caps &
+		   IB_ODP_SUPPORT_WRITE) &&
+		!!(device->attrs.odp_caps.per_transport_caps.rc_odp_caps &
+		   IB_ODP_SUPPORT_READ);
 
 	rds_ibdev->fmr_max_remaps = device->attrs.max_map_per_fmr?: 32;
 	rds_ibdev->max_1m_mrs = device->attrs.max_mr ?
diff --git a/net/rds/ib.h b/net/rds/ib.h
index 6e6f24753998..0296f1f7acda 100644
--- a/net/rds/ib.h
+++ b/net/rds/ib.h
@@ -247,7 +247,8 @@ struct rds_ib_device {
 	struct ib_device	*dev;
 	struct ib_pd		*pd;
 	struct dma_pool		*rid_hdrs_pool; /* RDS headers DMA pool */
-	bool                    use_fastreg;
+	u8			use_fastreg:1;
+	u8			odp_capable:1;
 
 	unsigned int		max_mrs;
 	struct rds_ib_mr_pool	*mr_1m_pool;
diff --git a/net/rds/ib_mr.h b/net/rds/ib_mr.h
index 9045a8c0edff..0c8252d7fe2b 100644
--- a/net/rds/ib_mr.h
+++ b/net/rds/ib_mr.h
@@ -67,6 +67,7 @@ struct rds_ib_frmr {
 
 /* This is stored as mr->r_trans_private. */
 struct rds_ib_mr {
+	struct delayed_work		work;
 	struct rds_ib_device		*device;
 	struct rds_ib_mr_pool		*pool;
 	struct rds_ib_connection	*ic;
@@ -81,9 +82,11 @@ struct rds_ib_mr {
 	unsigned int			sg_len;
 	int				sg_dma_len;
 
+	u8				odp:1;
 	union {
 		struct rds_ib_fmr	fmr;
 		struct rds_ib_frmr	frmr;
+		struct ib_mr		*mr;
 	} u;
 };
 
@@ -122,12 +125,14 @@ void rds6_ib_get_mr_info(struct rds_ib_device *rds_ibdev,
 void rds_ib_destroy_mr_pool(struct rds_ib_mr_pool *);
 void *rds_ib_get_mr(struct scatterlist *sg, unsigned long nents,
 		    struct rds_sock *rs, u32 *key_ret,
-		    struct rds_connection *conn);
+		    struct rds_connection *conn, u64 start, u64 length,
+		    int need_odp);
 void rds_ib_sync_mr(void *trans_private, int dir);
 void rds_ib_free_mr(void *trans_private, int invalidate);
 void rds_ib_flush_mrs(void);
 int rds_ib_mr_init(void);
 void rds_ib_mr_exit(void);
+u32 rds_ib_get_lkey(void *trans_private);
 
 void __rds_ib_teardown_mr(struct rds_ib_mr *);
 void rds_ib_teardown_mr(struct rds_ib_mr *);
diff --git a/net/rds/ib_rdma.c b/net/rds/ib_rdma.c
index c8c1e3ae8d84..ea4d7daec251 100644
--- a/net/rds/ib_rdma.c
+++ b/net/rds/ib_rdma.c
@@ -37,8 +37,15 @@
 
 #include "rds_single_path.h"
 #include "ib_mr.h"
+#include "rds.h"
 
 struct workqueue_struct *rds_ib_mr_wq;
+struct rds_ib_dereg_odp_mr {
+	struct work_struct work;
+	struct ib_mr *mr;
+};
+
+static void rds_ib_odp_mr_worker(struct work_struct *work);
 
 static struct rds_ib_device *rds_ib_get_device(__be32 ipaddr)
 {
@@ -213,6 +220,9 @@ void rds_ib_sync_mr(void *trans_private, int direction)
 	struct rds_ib_mr *ibmr = trans_private;
 	struct rds_ib_device *rds_ibdev = ibmr->device;
 
+	if (ibmr->odp)
+		return;
+
 	switch (direction) {
 	case DMA_FROM_DEVICE:
 		ib_dma_sync_sg_for_cpu(rds_ibdev->dev, ibmr->sg,
@@ -482,6 +492,16 @@ void rds_ib_free_mr(void *trans_private, int invalidate)
 
 	rdsdebug("RDS/IB: free_mr nents %u\n", ibmr->sg_len);
 
+	if (ibmr->odp) {
+		/* A MR created and marked as use_once. We use delayed work,
+		 * because there is a change that we are in interrupt and can't
+		 * call to ib_dereg_mr() directly.
+		 */
+		INIT_DELAYED_WORK(&ibmr->work, rds_ib_odp_mr_worker);
+		queue_delayed_work(rds_ib_mr_wq, &ibmr->work, 0);
+		return;
+	}
+
 	/* Return it to the pool's free list */
 	if (rds_ibdev->use_fastreg)
 		rds_ib_free_frmr_list(ibmr);
@@ -526,9 +546,17 @@ void rds_ib_flush_mrs(void)
 	up_read(&rds_ib_devices_lock);
 }
 
+u32 rds_ib_get_lkey(void *trans_private)
+{
+	struct rds_ib_mr *ibmr = trans_private;
+
+	return ibmr->u.mr->lkey;
+}
+
 void *rds_ib_get_mr(struct scatterlist *sg, unsigned long nents,
 		    struct rds_sock *rs, u32 *key_ret,
-		    struct rds_connection *conn)
+		    struct rds_connection *conn,
+		    u64 start, u64 length, int need_odp)
 {
 	struct rds_ib_device *rds_ibdev;
 	struct rds_ib_mr *ibmr = NULL;
@@ -541,6 +569,42 @@ void *rds_ib_get_mr(struct scatterlist *sg, unsigned long nents,
 		goto out;
 	}
 
+	if (need_odp == ODP_ZEROBASED || need_odp == ODP_VIRTUAL) {
+		u64 virt_addr = need_odp == ODP_ZEROBASED ? 0 : start;
+		int access_flags =
+			(IB_ACCESS_LOCAL_WRITE | IB_ACCESS_REMOTE_READ |
+			 IB_ACCESS_REMOTE_WRITE | IB_ACCESS_REMOTE_ATOMIC |
+			 IB_ACCESS_ON_DEMAND);
+		struct ib_mr *ib_mr;
+
+		if (!rds_ibdev->odp_capable) {
+			ret = -EOPNOTSUPP;
+			goto out;
+		}
+
+		ib_mr = ib_reg_user_mr(rds_ibdev->pd, start, length, virt_addr,
+				       access_flags);
+
+		if (IS_ERR(ib_mr)) {
+			rdsdebug("rds_ib_get_user_mr returned %d\n",
+				 IS_ERR(ib_mr));
+			ret = PTR_ERR(ib_mr);
+			goto out;
+		}
+		if (key_ret)
+			*key_ret = ib_mr->rkey;
+
+		ibmr = kzalloc(sizeof(*ibmr), GFP_KERNEL);
+		if (!ibmr) {
+			ib_dereg_mr(ib_mr);
+			ret = -ENOMEM;
+			goto out;
+		}
+		ibmr->u.mr = ib_mr;
+		ibmr->odp = 1;
+		return ibmr;
+	}
+
 	if (conn)
 		ic = conn->c_transport_data;
 
@@ -629,3 +693,12 @@ void rds_ib_mr_exit(void)
 {
 	destroy_workqueue(rds_ib_mr_wq);
 }
+
+static void rds_ib_odp_mr_worker(struct work_struct  *work)
+{
+	struct rds_ib_mr *ibmr;
+
+	ibmr = container_of(work, struct rds_ib_mr, work.work);
+	ib_dereg_mr(ibmr->u.mr);
+	kfree(ibmr);
+}
diff --git a/net/rds/ib_send.c b/net/rds/ib_send.c
index d1cc1d7778d8..dfe778220657 100644
--- a/net/rds/ib_send.c
+++ b/net/rds/ib_send.c
@@ -39,6 +39,7 @@
 #include "rds_single_path.h"
 #include "rds.h"
 #include "ib.h"
+#include "ib_mr.h"
 
 /*
  * Convert IB-specific error message to RDS error message and call core
@@ -635,6 +636,7 @@ int rds_ib_xmit(struct rds_connection *conn, struct rds_message *rm,
 		send->s_sge[0].addr = ic->i_send_hdrs_dma[pos];
 
 		send->s_sge[0].length = sizeof(struct rds_header);
+		send->s_sge[0].lkey = ic->i_pd->local_dma_lkey;
 
 		memcpy(ic->i_send_hdrs[pos], &rm->m_inc.i_hdr,
 		       sizeof(struct rds_header));
@@ -650,6 +652,7 @@ int rds_ib_xmit(struct rds_connection *conn, struct rds_message *rm,
 			send->s_sge[1].addr = sg_dma_address(scat);
 			send->s_sge[1].addr += rm->data.op_dmaoff;
 			send->s_sge[1].length = len;
+			send->s_sge[1].lkey = ic->i_pd->local_dma_lkey;
 
 			bytes_sent += len;
 			rm->data.op_dmaoff += len;
@@ -858,20 +861,29 @@ int rds_ib_xmit_rdma(struct rds_connection *conn, struct rm_rdma_op *op)
 	int ret;
 	int num_sge;
 	int nr_sig = 0;
+	u64 odp_addr = op->op_odp_addr;
+	u32 odp_lkey = 0;
 
 	/* map the op the first time we see it */
-	if (!op->op_mapped) {
-		op->op_count = ib_dma_map_sg(ic->i_cm_id->device,
-					     op->op_sg, op->op_nents, (op->op_write) ?
-					     DMA_TO_DEVICE : DMA_FROM_DEVICE);
-		rdsdebug("ic %p mapping op %p: %d\n", ic, op, op->op_count);
-		if (op->op_count == 0) {
-			rds_ib_stats_inc(s_ib_tx_sg_mapping_failure);
-			ret = -ENOMEM; /* XXX ? */
-			goto out;
+	if (!op->op_odp_mr) {
+		if (!op->op_mapped) {
+			op->op_count =
+				ib_dma_map_sg(ic->i_cm_id->device, op->op_sg,
+					      op->op_nents,
+					      (op->op_write) ? DMA_TO_DEVICE :
+							       DMA_FROM_DEVICE);
+			rdsdebug("ic %p mapping op %p: %d\n", ic, op,
+				 op->op_count);
+			if (op->op_count == 0) {
+				rds_ib_stats_inc(s_ib_tx_sg_mapping_failure);
+				ret = -ENOMEM; /* XXX ? */
+				goto out;
+			}
+			op->op_mapped = 1;
 		}
-
-		op->op_mapped = 1;
+	} else {
+		op->op_count = op->op_nents;
+		odp_lkey = rds_ib_get_lkey(op->op_odp_mr->r_trans_private);
 	}
 
 	/*
@@ -923,14 +935,20 @@ int rds_ib_xmit_rdma(struct rds_connection *conn, struct rm_rdma_op *op)
 		for (j = 0; j < send->s_rdma_wr.wr.num_sge &&
 		     scat != &op->op_sg[op->op_count]; j++) {
 			len = sg_dma_len(scat);
-			send->s_sge[j].addr = sg_dma_address(scat);
+			if (!op->op_odp_mr) {
+				send->s_sge[j].addr = sg_dma_address(scat);
+				send->s_sge[j].lkey = ic->i_pd->local_dma_lkey;
+			} else {
+				send->s_sge[j].addr = odp_addr;
+				send->s_sge[j].lkey = odp_lkey;
+			}
 			send->s_sge[j].length = len;
-			send->s_sge[j].lkey = ic->i_pd->local_dma_lkey;
 
 			sent += len;
 			rdsdebug("ic %p sent %d remote_addr %llu\n", ic, sent, remote_addr);
 
 			remote_addr += len;
+			odp_addr += len;
 			scat++;
 		}
 
diff --git a/net/rds/rdma.c b/net/rds/rdma.c
index eb23c38ce2b3..3341eee87bf9 100644
--- a/net/rds/rdma.c
+++ b/net/rds/rdma.c
@@ -177,13 +177,14 @@ static int __rds_rdma_map(struct rds_sock *rs, struct rds_get_mr_args *args,
 			  struct rds_conn_path *cp)
 {
 	struct rds_mr *mr = NULL, *found;
+	struct scatterlist *sg = NULL;
 	unsigned int nr_pages;
 	struct page **pages = NULL;
-	struct scatterlist *sg;
 	void *trans_private;
 	unsigned long flags;
 	rds_rdma_cookie_t cookie;
-	unsigned int nents;
+	unsigned int nents = 0;
+	int need_odp = 0;
 	long i;
 	int ret;
 
@@ -197,6 +198,21 @@ static int __rds_rdma_map(struct rds_sock *rs, struct rds_get_mr_args *args,
 		goto out;
 	}
 
+	/* If the combination of the addr and size requested for this memory
+	 * region causes an integer overflow, return error.
+	 */
+	if (((args->vec.addr + args->vec.bytes) < args->vec.addr) ||
+	    PAGE_ALIGN(args->vec.addr + args->vec.bytes) <
+		    (args->vec.addr + args->vec.bytes)) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	if (!can_do_mlock()) {
+		ret = -EPERM;
+		goto out;
+	}
+
 	nr_pages = rds_pages_in_vec(&args->vec);
 	if (nr_pages == 0) {
 		ret = -EINVAL;
@@ -250,36 +266,44 @@ static int __rds_rdma_map(struct rds_sock *rs, struct rds_get_mr_args *args,
 	 * the zero page.
 	 */
 	ret = rds_pin_pages(args->vec.addr, nr_pages, pages, 1);
-	if (ret < 0)
-		goto out;
-
-	nents = ret;
-	sg = kcalloc(nents, sizeof(*sg), GFP_KERNEL);
-	if (!sg) {
-		ret = -ENOMEM;
+	if (ret == -EOPNOTSUPP) {
+		need_odp = 1;
+	} else if (ret <= 0) {
 		goto out;
-	}
-	WARN_ON(!nents);
-	sg_init_table(sg, nents);
-
-	/* Stick all pages into the scatterlist */
-	for (i = 0 ; i < nents; i++)
-		sg_set_page(&sg[i], pages[i], PAGE_SIZE, 0);
+	} else {
+		nents = ret;
+		sg = kcalloc(nents, sizeof(*sg), GFP_KERNEL);
+		if (!sg) {
+			ret = -ENOMEM;
+			goto out;
+		}
+		WARN_ON(!nents);
+		sg_init_table(sg, nents);
 
-	rdsdebug("RDS: trans_private nents is %u\n", nents);
+		/* Stick all pages into the scatterlist */
+		for (i = 0 ; i < nents; i++)
+			sg_set_page(&sg[i], pages[i], PAGE_SIZE, 0);
 
+		rdsdebug("RDS: trans_private nents is %u\n", nents);
+	}
 	/* Obtain a transport specific MR. If this succeeds, the
 	 * s/g list is now owned by the MR.
 	 * Note that dma_map() implies that pending writes are
 	 * flushed to RAM, so no dma_sync is needed here. */
-	trans_private = rs->rs_transport->get_mr(sg, nents, rs,
-						 &mr->r_key,
-						 cp ? cp->cp_conn : NULL);
+	trans_private = rs->rs_transport->get_mr(
+		sg, nents, rs, &mr->r_key, cp ? cp->cp_conn : NULL,
+		args->vec.addr, args->vec.bytes,
+		need_odp ? ODP_ZEROBASED : ODP_NOT_NEEDED);
 
 	if (IS_ERR(trans_private)) {
-		for (i = 0 ; i < nents; i++)
-			put_page(sg_page(&sg[i]));
-		kfree(sg);
+		/* In ODP case, we don't GUP pages, so don't need
+		 * to release anything.
+		 */
+		if (!need_odp) {
+			for (i = 0 ; i < nents; i++)
+				put_page(sg_page(&sg[i]));
+			kfree(sg);
+		}
 		ret = PTR_ERR(trans_private);
 		goto out;
 	}
@@ -293,7 +317,11 @@ static int __rds_rdma_map(struct rds_sock *rs, struct rds_get_mr_args *args,
 	 * map page aligned regions. So we keep the offset, and build
 	 * a 64bit cookie containing <R_Key, offset> and pass that
 	 * around. */
-	cookie = rds_rdma_make_cookie(mr->r_key, args->vec.addr & ~PAGE_MASK);
+	if (need_odp)
+		cookie = rds_rdma_make_cookie(mr->r_key, 0);
+	else
+		cookie = rds_rdma_make_cookie(mr->r_key,
+					      args->vec.addr & ~PAGE_MASK);
 	if (cookie_ret)
 		*cookie_ret = cookie;
 
@@ -458,22 +486,26 @@ void rds_rdma_free_op(struct rm_rdma_op *ro)
 {
 	unsigned int i;
 
-	for (i = 0; i < ro->op_nents; i++) {
-		struct page *page = sg_page(&ro->op_sg[i]);
-
-		/* Mark page dirty if it was possibly modified, which
-		 * is the case for a RDMA_READ which copies from remote
-		 * to local memory */
-		if (!ro->op_write) {
-			WARN_ON(!page->mapping && irqs_disabled());
-			set_page_dirty(page);
+	if (ro->op_odp_mr) {
+		rds_mr_put(ro->op_odp_mr);
+	} else {
+		for (i = 0; i < ro->op_nents; i++) {
+			struct page *page = sg_page(&ro->op_sg[i]);
+
+			/* Mark page dirty if it was possibly modified, which
+			 * is the case for a RDMA_READ which copies from remote
+			 * to local memory
+			 */
+			if (!ro->op_write)
+				set_page_dirty(page);
+			put_page(page);
 		}
-		put_page(page);
 	}
 
 	kfree(ro->op_notifier);
 	ro->op_notifier = NULL;
 	ro->op_active = 0;
+	ro->op_odp_mr = NULL;
 }
 
 void rds_atomic_free_op(struct rm_atomic_op *ao)
@@ -583,6 +615,7 @@ int rds_cmsg_rdma_args(struct rds_sock *rs, struct rds_message *rm,
 	struct rds_iovec *iovs;
 	unsigned int i, j;
 	int ret = 0;
+	bool odp_supported = true;
 
 	if (cmsg->cmsg_len < CMSG_LEN(sizeof(struct rds_rdma_args))
 	    || rm->rdma.op_active)
@@ -604,6 +637,9 @@ int rds_cmsg_rdma_args(struct rds_sock *rs, struct rds_message *rm,
 		ret = -EINVAL;
 		goto out_ret;
 	}
+	/* odp-mr is not supported for multiple requests within one message */
+	if (args->nr_local != 1)
+		odp_supported = false;
 
 	iovs = vec->iov;
 
@@ -625,6 +661,8 @@ int rds_cmsg_rdma_args(struct rds_sock *rs, struct rds_message *rm,
 	op->op_silent = !!(args->flags & RDS_RDMA_SILENT);
 	op->op_active = 1;
 	op->op_recverr = rs->rs_recverr;
+	op->op_odp_mr = NULL;
+
 	WARN_ON(!nr_pages);
 	op->op_sg = rds_message_alloc_sgs(rm, nr_pages, &ret);
 	if (!op->op_sg)
@@ -674,10 +712,44 @@ int rds_cmsg_rdma_args(struct rds_sock *rs, struct rds_message *rm,
 		 * If it's a READ operation, we need to pin the pages for writing.
 		 */
 		ret = rds_pin_pages(iov->addr, nr, pages, !op->op_write);
-		if (ret < 0)
+		if ((!odp_supported && ret <= 0) ||
+		    (odp_supported && ret <= 0 && ret != -EOPNOTSUPP))
 			goto out_pages;
-		else
-			ret = 0;
+
+		if (ret == -EOPNOTSUPP) {
+			struct rds_mr *local_odp_mr;
+
+			if (!rs->rs_transport->get_mr) {
+				ret = -EOPNOTSUPP;
+				goto out_pages;
+			}
+			local_odp_mr =
+				kzalloc(sizeof(*local_odp_mr), GFP_KERNEL);
+			if (!local_odp_mr) {
+				ret = -ENOMEM;
+				goto out_pages;
+			}
+			RB_CLEAR_NODE(&local_odp_mr->r_rb_node);
+			refcount_set(&local_odp_mr->r_refcount, 1);
+			local_odp_mr->r_trans = rs->rs_transport;
+			local_odp_mr->r_sock = rs;
+			local_odp_mr->r_trans_private =
+				rs->rs_transport->get_mr(
+					NULL, 0, rs, &local_odp_mr->r_key, NULL,
+					iov->addr, iov->bytes, ODP_VIRTUAL);
+			if (IS_ERR(local_odp_mr->r_trans_private)) {
+				ret = IS_ERR(local_odp_mr->r_trans_private);
+				rdsdebug("get_mr ret %d %p\"", ret,
+					 local_odp_mr->r_trans_private);
+				kfree(local_odp_mr);
+				ret = -EOPNOTSUPP;
+				goto out_pages;
+			}
+			rdsdebug("Need odp; local_odp_mr %p trans_private %p\n",
+				 local_odp_mr, local_odp_mr->r_trans_private);
+			op->op_odp_mr = local_odp_mr;
+			op->op_odp_addr = iov->addr;
+		}
 
 		rdsdebug("RDS: nr_bytes %u nr %u iov->bytes %llu iov->addr %llx\n",
 			 nr_bytes, nr, iov->bytes, iov->addr);
@@ -693,6 +765,7 @@ int rds_cmsg_rdma_args(struct rds_sock *rs, struct rds_message *rm,
 					min_t(unsigned int, iov->bytes, PAGE_SIZE - offset),
 					offset);
 
+			sg_dma_len(sg) = sg->length;
 			rdsdebug("RDS: sg->offset %x sg->len %x iov->addr %llx iov->bytes %llu\n",
 			       sg->offset, sg->length, iov->addr, iov->bytes);
 
@@ -711,6 +784,7 @@ int rds_cmsg_rdma_args(struct rds_sock *rs, struct rds_message *rm,
 		goto out_pages;
 	}
 	op->op_bytes = nr_bytes;
+	ret = 0;
 
 out_pages:
 	kfree(pages);
@@ -757,7 +831,8 @@ int rds_cmsg_rdma_dest(struct rds_sock *rs, struct rds_message *rm,
 	spin_unlock_irqrestore(&rs->rs_rdma_lock, flags);
 
 	if (mr) {
-		mr->r_trans->sync_mr(mr->r_trans_private, DMA_TO_DEVICE);
+		mr->r_trans->sync_mr(mr->r_trans_private,
+				     DMA_TO_DEVICE);
 		rm->rdma.op_rdma_mr = mr;
 	}
 	return err;
diff --git a/net/rds/rds.h b/net/rds/rds.h
index 53e86911773a..e4a603523083 100644
--- a/net/rds/rds.h
+++ b/net/rds/rds.h
@@ -40,7 +40,6 @@
 #ifdef ATOMIC64_INIT
 #define KERNEL_HAS_ATOMIC64
 #endif
-
 #ifdef RDS_DEBUG
 #define rdsdebug(fmt, args...) pr_debug("%s(): " fmt, __func__ , ##args)
 #else
@@ -478,6 +477,9 @@ struct rds_message {
 			struct rds_notifier	*op_notifier;
 
 			struct rds_mr		*op_rdma_mr;
+
+			u64			op_odp_addr;
+			struct rds_mr		*op_odp_mr;
 		} rdma;
 		struct rm_data_op {
 			unsigned int		op_active:1;
@@ -573,7 +575,8 @@ struct rds_transport {
 	void (*exit)(void);
 	void *(*get_mr)(struct scatterlist *sg, unsigned long nr_sg,
 			struct rds_sock *rs, u32 *key_ret,
-			struct rds_connection *conn);
+			struct rds_connection *conn,
+			u64 start, u64 length, int need_odp);
 	void (*sync_mr)(void *trans_private, int direction);
 	void (*free_mr)(void *trans_private, int invalidate);
 	void (*flush_mrs)(void);
@@ -956,6 +959,12 @@ static inline bool rds_destroy_pending(struct rds_connection *conn)
 	       (conn->c_trans->t_unloading && conn->c_trans->t_unloading(conn));
 }
 
+enum {
+	ODP_NOT_NEEDED,
+	ODP_ZEROBASED,
+	ODP_VIRTUAL
+};
+
 /* stats.c */
 DECLARE_PER_CPU_SHARED_ALIGNED(struct rds_statistics, rds_stats);
 #define rds_stats_inc_which(which, member) do {		\
-- 
cgit v1.2.3-59-g8ed1b


From b2dfc6765e45a3154800333234e4952b5412d792 Mon Sep 17 00:00:00 2001
From: Hans Westgaard Ry <hans.westgaard.ry@oracle.com>
Date: Wed, 15 Jan 2020 14:43:40 +0200
Subject: net/rds: Use prefetch for On-Demand-Paging MR

Try prefetching pages when using On-Demand-Paging MR using
ib_advise_mr.

Signed-off-by: Hans Westgaard Ry <hans.westgaard.ry@oracle.com>
Acked-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
---
 net/rds/ib_rdma.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/net/rds/ib_rdma.c b/net/rds/ib_rdma.c
index ea4d7daec251..b34b24e237f8 100644
--- a/net/rds/ib_rdma.c
+++ b/net/rds/ib_rdma.c
@@ -575,6 +575,7 @@ void *rds_ib_get_mr(struct scatterlist *sg, unsigned long nents,
 			(IB_ACCESS_LOCAL_WRITE | IB_ACCESS_REMOTE_READ |
 			 IB_ACCESS_REMOTE_WRITE | IB_ACCESS_REMOTE_ATOMIC |
 			 IB_ACCESS_ON_DEMAND);
+		struct ib_sge sge = {};
 		struct ib_mr *ib_mr;
 
 		if (!rds_ibdev->odp_capable) {
@@ -602,6 +603,14 @@ void *rds_ib_get_mr(struct scatterlist *sg, unsigned long nents,
 		}
 		ibmr->u.mr = ib_mr;
 		ibmr->odp = 1;
+
+		sge.addr = virt_addr;
+		sge.length = length;
+		sge.lkey = ib_mr->lkey;
+
+		ib_advise_mr(rds_ibdev->pd,
+			     IB_UVERBS_ADVISE_MR_ADVICE_PREFETCH_WRITE,
+			     IB_UVERBS_ADVISE_MR_FLAG_FLUSH, &sge, 1);
 		return ibmr;
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From f3660937e17c6fcc3a9184dd7d5091110e7e9094 Mon Sep 17 00:00:00 2001
From: Alex Marginean <alexandru.marginean@nxp.com>
Date: Thu, 16 Jan 2020 20:19:32 +0200
Subject: net: dsa: felix: Handle PAUSE RX regardless of AN result

Flow control is used with 2500Base-X and AQR PHYs to do rate adaptation
between line side 100/1000 links and MAC running at 2.5G.

This is independent of the flow control configuration settled on line
side though AN.

In general, allowing the MAC to handle flow control even if not
negotiated with the link partner should not be a problem, so the patch
just enables it in all cases.

Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/ocelot/felix.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/net/dsa/ocelot/felix.c b/drivers/net/dsa/ocelot/felix.c
index 269cc6953d47..8a38290c29fd 100644
--- a/drivers/net/dsa/ocelot/felix.c
+++ b/drivers/net/dsa/ocelot/felix.c
@@ -222,8 +222,12 @@ static void felix_phylink_mac_config(struct dsa_switch *ds, int port,
 	 * specification in incoming pause frames.
 	 */
 	mac_fc_cfg = SYS_MAC_FC_CFG_FC_LINK_SPEED(state->speed);
-	if (state->pause & MLO_PAUSE_RX)
-		mac_fc_cfg |= SYS_MAC_FC_CFG_RX_FC_ENA;
+
+	/* handle Rx pause in all cases, with 2500base-X this is used for rate
+	 * adaptation.
+	 */
+	mac_fc_cfg |= SYS_MAC_FC_CFG_RX_FC_ENA;
+
 	if (state->pause & MLO_PAUSE_TX)
 		mac_fc_cfg |= SYS_MAC_FC_CFG_TX_FC_ENA |
 			      SYS_MAC_FC_CFG_PAUSE_VAL_CFG(0xffff) |
-- 
cgit v1.2.3-59-g8ed1b


From 74984a1904b3768625ef45032643f1dc2b21eb31 Mon Sep 17 00:00:00 2001
From: Alex Marginean <alexandru.marginean@nxp.com>
Date: Thu, 16 Jan 2020 20:19:33 +0200
Subject: net: dsa: felix: Allow PHY to AN 10/100/1000 with 2500 serdes link

If the serdes link is set to 2500 using interfce type 2500base-X, lower
link speeds over on the line side should still be supported.
Rate adaptation is done out of band, in our case using AQR PHYs this is
done using flow control.

Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/ocelot/felix.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/drivers/net/dsa/ocelot/felix.c b/drivers/net/dsa/ocelot/felix.c
index 8a38290c29fd..3257962c147e 100644
--- a/drivers/net/dsa/ocelot/felix.c
+++ b/drivers/net/dsa/ocelot/felix.c
@@ -172,11 +172,10 @@ static void felix_phylink_validate(struct dsa_switch *ds, int port,
 	phylink_set(mask, Autoneg);
 	phylink_set(mask, Pause);
 	phylink_set(mask, Asym_Pause);
-	if (state->interface != PHY_INTERFACE_MODE_2500BASEX) {
-		phylink_set(mask, 10baseT_Full);
-		phylink_set(mask, 100baseT_Full);
-		phylink_set(mask, 1000baseT_Full);
-	}
+	phylink_set(mask, 10baseT_Full);
+	phylink_set(mask, 100baseT_Full);
+	phylink_set(mask, 1000baseT_Full);
+
 	/* The internal ports that run at 2.5G are overclocked GMII */
 	if (state->interface == PHY_INTERFACE_MODE_GMII ||
 	    state->interface == PHY_INTERFACE_MODE_2500BASEX ||
-- 
cgit v1.2.3-59-g8ed1b


From 0c58ac1e0155fdbefb976ff2d321d1093b92fe50 Mon Sep 17 00:00:00 2001
From: Dejin Zheng <zhengdejin5@gmail.com>
Date: Fri, 17 Jan 2020 23:10:42 +0800
Subject: net: phy: adin: fix a warning about msleep

found a warning by the following command:
./scripts/checkpatch.pl -f drivers/net/phy/adin.c

WARNING: msleep < 20ms can sleep for up to 20ms; see Documentation/timers/timers-howto.rst
 #628: FILE: drivers/net/phy/adin.c:628:
+	msleep(10);

Signed-off-by: Dejin Zheng <zhengdejin5@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/adin.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/phy/adin.c b/drivers/net/phy/adin.c
index 1dca3e883df4..c7eabe4382fb 100644
--- a/drivers/net/phy/adin.c
+++ b/drivers/net/phy/adin.c
@@ -625,7 +625,7 @@ static int adin_soft_reset(struct phy_device *phydev)
 	if (rc < 0)
 		return rc;
 
-	msleep(10);
+	msleep(20);
 
 	/* If we get a read error something may be wrong */
 	rc = phy_read_mmd(phydev, MDIO_MMD_VEND1,
-- 
cgit v1.2.3-59-g8ed1b


From 5b051621609d535fac6bdcd232530cc5248e791d Mon Sep 17 00:00:00 2001
From: Amit Cohen <amitc@mellanox.com>
Date: Sun, 19 Jan 2020 15:00:46 +0200
Subject: mlxsw: Add irif and erif disabled traps

IRIF_DISABLED and ERIF_DISABLED are driver specific traps. Packets are
dropped for these reasons when they need to be routed through/from
existing router interfaces (RIF) which are disabled.

Add devlink driver-specific traps and mlxsw trap IDs used to report
these traps.

Signed-off-by: Amit Cohen <amitc@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/devlink/devlink-trap.rst  |  1 +
 Documentation/networking/devlink/mlxsw.rst         | 22 ++++++++++++++++++
 .../net/ethernet/mellanox/mlxsw/spectrum_trap.c    | 26 ++++++++++++++++++++++
 drivers/net/ethernet/mellanox/mlxsw/trap.h         |  2 ++
 4 files changed, 51 insertions(+)

diff --git a/Documentation/networking/devlink/devlink-trap.rst b/Documentation/networking/devlink/devlink-trap.rst
index cbaa750de37d..68245ea387ad 100644
--- a/Documentation/networking/devlink/devlink-trap.rst
+++ b/Documentation/networking/devlink/devlink-trap.rst
@@ -234,6 +234,7 @@ links to the description of driver-specific traps registered by various device
 drivers:
 
   * :doc:`netdevsim`
+  * :doc:`mlxsw`
 
 Generic Packet Trap Groups
 ==========================
diff --git a/Documentation/networking/devlink/mlxsw.rst b/Documentation/networking/devlink/mlxsw.rst
index 5f9bb0a0616a..cf857cb4ba8f 100644
--- a/Documentation/networking/devlink/mlxsw.rst
+++ b/Documentation/networking/devlink/mlxsw.rst
@@ -57,3 +57,25 @@ The ``mlxsw`` driver reports the following versions
    * - ``fw.version``
      - running
      - Three digit firmware version
+
+Driver-specific Traps
+=====================
+
+.. list-table:: List of Driver-specific Traps Registered by ``mlxsw``
+   :widths: 5 5 90
+
+   * - Name
+     - Type
+     - Description
+   * - ``irif_disabled``
+     - ``drop``
+     - Traps packets that the device decided to drop because they need to be
+       routed from a disabled router interface (RIF). This can happen during
+       RIF dismantle, when the RIF is first disabled before being removed
+       completely
+   * - ``erif_disabled``
+     - ``drop``
+     - Traps packets that the device decided to drop because they need to be
+       routed through a disabled router interface (RIF). This can happen during
+       RIF dismantle, when the RIF is first disabled before being removed
+       completely
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c
index e0d7c49ffae0..42013fe11131 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c
@@ -9,6 +9,20 @@
 #include "reg.h"
 #include "spectrum.h"
 
+/* All driver-specific traps must be documented in
+ * Documentation/networking/devlink/mlxsw.rst
+ */
+enum {
+	DEVLINK_MLXSW_TRAP_ID_BASE = DEVLINK_TRAP_GENERIC_ID_MAX,
+	DEVLINK_MLXSW_TRAP_ID_IRIF_DISABLED,
+	DEVLINK_MLXSW_TRAP_ID_ERIF_DISABLED,
+};
+
+#define DEVLINK_MLXSW_TRAP_NAME_IRIF_DISABLED \
+	"irif_disabled"
+#define DEVLINK_MLXSW_TRAP_NAME_ERIF_DISABLED \
+	"erif_disabled"
+
 #define MLXSW_SP_TRAP_METADATA DEVLINK_TRAP_METADATA_TYPE_F_IN_PORT
 
 static void mlxsw_sp_rx_drop_listener(struct sk_buff *skb, u8 local_port,
@@ -21,6 +35,12 @@ static void mlxsw_sp_rx_exception_listener(struct sk_buff *skb, u8 local_port,
 			     DEVLINK_TRAP_GROUP_GENERIC(_group_id),	      \
 			     MLXSW_SP_TRAP_METADATA)
 
+#define MLXSW_SP_TRAP_DRIVER_DROP(_id, _group_id)			      \
+	DEVLINK_TRAP_DRIVER(DROP, DROP, DEVLINK_MLXSW_TRAP_ID_##_id,	      \
+			    DEVLINK_MLXSW_TRAP_NAME_##_id,		      \
+			    DEVLINK_TRAP_GROUP_GENERIC(_group_id),	      \
+			    MLXSW_SP_TRAP_METADATA)
+
 #define MLXSW_SP_TRAP_EXCEPTION(_id, _group_id)		      \
 	DEVLINK_TRAP_GENERIC(EXCEPTION, TRAP, _id,			      \
 			     DEVLINK_TRAP_GROUP_GENERIC(_group_id),	      \
@@ -58,6 +78,8 @@ static struct devlink_trap mlxsw_sp_traps_arr[] = {
 	MLXSW_SP_TRAP_EXCEPTION(UNRESOLVED_NEIGH, L3_DROPS),
 	MLXSW_SP_TRAP_EXCEPTION(IPV4_LPM_UNICAST_MISS, L3_DROPS),
 	MLXSW_SP_TRAP_EXCEPTION(IPV6_LPM_UNICAST_MISS, L3_DROPS),
+	MLXSW_SP_TRAP_DRIVER_DROP(IRIF_DISABLED, L3_DROPS),
+	MLXSW_SP_TRAP_DRIVER_DROP(ERIF_DISABLED, L3_DROPS),
 };
 
 static struct mlxsw_listener mlxsw_sp_listeners_arr[] = {
@@ -90,6 +112,8 @@ static struct mlxsw_listener mlxsw_sp_listeners_arr[] = {
 			       TRAP_EXCEPTION_TO_CPU),
 	MLXSW_SP_RXL_EXCEPTION(DISCARD_ROUTER_LPM6, ROUTER_EXP,
 			       TRAP_EXCEPTION_TO_CPU),
+	MLXSW_SP_RXL_DISCARD(ROUTER_IRIF_EN, L3_DISCARDS),
+	MLXSW_SP_RXL_DISCARD(ROUTER_ERIF_EN, L3_DISCARDS),
 };
 
 /* Mapping between hardware trap and devlink trap. Multiple hardware traps can
@@ -123,6 +147,8 @@ static u16 mlxsw_sp_listener_devlink_map[] = {
 	DEVLINK_TRAP_GENERIC_ID_UNRESOLVED_NEIGH,
 	DEVLINK_TRAP_GENERIC_ID_IPV4_LPM_UNICAST_MISS,
 	DEVLINK_TRAP_GENERIC_ID_IPV6_LPM_UNICAST_MISS,
+	DEVLINK_MLXSW_TRAP_ID_IRIF_DISABLED,
+	DEVLINK_MLXSW_TRAP_ID_ERIF_DISABLED,
 };
 
 static int mlxsw_sp_rx_listener(struct mlxsw_sp *mlxsw_sp, struct sk_buff *skb,
diff --git a/drivers/net/ethernet/mellanox/mlxsw/trap.h b/drivers/net/ethernet/mellanox/mlxsw/trap.h
index 3d2331be05d8..8573cc3a0010 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/trap.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/trap.h
@@ -88,6 +88,8 @@ enum {
 	MLXSW_TRAP_ID_DISCARD_ING_ROUTER_IPV4_SIP_BC = 0x16A,
 	MLXSW_TRAP_ID_DISCARD_ING_ROUTER_IPV4_DIP_LOCAL_NET = 0x16B,
 	MLXSW_TRAP_ID_DISCARD_ING_ROUTER_DIP_LINK_LOCAL = 0x16C,
+	MLXSW_TRAP_ID_DISCARD_ROUTER_IRIF_EN = 0x178,
+	MLXSW_TRAP_ID_DISCARD_ROUTER_ERIF_EN = 0x179,
 	MLXSW_TRAP_ID_DISCARD_ROUTER_LPM4 = 0x17B,
 	MLXSW_TRAP_ID_DISCARD_ROUTER_LPM6 = 0x17C,
 	MLXSW_TRAP_ID_DISCARD_IPV6_MC_DIP_RESERVED_SCOPE = 0x1B0,
-- 
cgit v1.2.3-59-g8ed1b


From 740e87bc3db8fe299e17877b7055e9ae3c33e4e6 Mon Sep 17 00:00:00 2001
From: Amit Cohen <amitc@mellanox.com>
Date: Sun, 19 Jan 2020 15:00:47 +0200
Subject: selftests: devlink_trap_l3_drops: Add test cases of irif and erif
 disabled

Add test cases to check that packets routed through disabled RIFs and
packets routed from disabled RIFs are dropped and devlink counters
increase when the action is trap.

Signed-off-by: Amit Cohen <amitc@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../drivers/net/mlxsw/devlink_trap_l3_drops.sh     | 112 +++++++++++++++++++++
 1 file changed, 112 insertions(+)

diff --git a/tools/testing/selftests/drivers/net/mlxsw/devlink_trap_l3_drops.sh b/tools/testing/selftests/drivers/net/mlxsw/devlink_trap_l3_drops.sh
index b4efb023ae51..d88d8e47d11b 100755
--- a/tools/testing/selftests/drivers/net/mlxsw/devlink_trap_l3_drops.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/devlink_trap_l3_drops.sh
@@ -50,6 +50,8 @@ ALL_TESTS="
 	ipv6_mc_dip_reserved_scope_test
 	ipv6_mc_dip_interface_local_scope_test
 	blackhole_route_test
+	irif_disabled_test
+	erif_disabled_test
 "
 
 NUM_NETIFS=4
@@ -553,6 +555,116 @@ blackhole_route_test()
 	__blackhole_route_test "6" "2001:db8:2::/120" "ipv6" $h2_ipv6 "icmpv6"
 }
 
+irif_disabled_test()
+{
+	local trap_name="irif_disabled"
+	local group_name="l3_drops"
+	local t0_packets t0_bytes
+	local t1_packets t1_bytes
+	local mz_pid
+
+	RET=0
+
+	ping_check $trap_name
+
+	devlink_trap_action_set $trap_name "trap"
+
+	# When RIF of a physical port ("Sub-port RIF") is destroyed, we first
+	# block the STP of the {Port, VLAN} so packets cannot get into the RIF.
+	# Using bridge enables us to see this trap because when bridge is
+	# destroyed, there is a small time window that packets can go into the
+	# RIF, while it is disabled.
+	ip link add dev br0 type bridge
+	ip link set dev $rp1 master br0
+	ip address flush dev $rp1
+	__addr_add_del br0 add 192.0.2.2/24
+	ip li set dev br0 up
+
+	t0_packets=$(devlink_trap_rx_packets_get $trap_name)
+	t0_bytes=$(devlink_trap_rx_bytes_get $trap_name)
+
+	# Generate packets to h2 through br0 RIF that will be removed later
+	$MZ $h1 -t udp "sp=54321,dp=12345" -c 0 -p 100 -a own -b $rp1mac \
+		-B $h2_ipv4 -q &
+	mz_pid=$!
+
+	# Wait before removing br0 RIF to allow packets to go into the bridge.
+	sleep 1
+
+	# Flushing address will dismantle the RIF
+	ip address flush dev br0
+
+	t1_packets=$(devlink_trap_rx_packets_get $trap_name)
+	t1_bytes=$(devlink_trap_rx_bytes_get $trap_name)
+
+	if [[ $t0_packets -eq $t1_packets && $t0_bytes -eq $t1_bytes ]]; then
+		check_err 1 "Trap stats idle when packets should be trapped"
+	fi
+
+	log_test "Ingress RIF disabled"
+
+	kill $mz_pid && wait $mz_pid &> /dev/null
+	ip link set dev $rp1 nomaster
+	__addr_add_del $rp1 add 192.0.2.2/24 2001:db8:1::2/64
+	ip link del dev br0 type bridge
+	devlink_trap_action_set $trap_name "drop"
+}
+
+erif_disabled_test()
+{
+	local trap_name="erif_disabled"
+	local group_name="l3_drops"
+	local t0_packets t0_bytes
+	local t1_packets t1_bytes
+	local mz_pid
+
+	RET=0
+
+	ping_check $trap_name
+
+	devlink_trap_action_set $trap_name "trap"
+	ip link add dev br0 type bridge
+	ip add flush dev $rp1
+	ip link set dev $rp1 master br0
+	__addr_add_del br0 add 192.0.2.2/24
+	ip link set dev br0 up
+
+	t0_packets=$(devlink_trap_rx_packets_get $trap_name)
+	t0_bytes=$(devlink_trap_rx_bytes_get $trap_name)
+
+	rp2mac=$(mac_get $rp2)
+
+	# Generate packets that should go out through br0 RIF that will be
+	# removed later
+	$MZ $h2 -t udp "sp=54321,dp=12345" -c 0 -p 100 -a own -b $rp2mac \
+		-B 192.0.2.1 -q &
+	mz_pid=$!
+
+	sleep 5
+	# In order to see this trap we need a route that points to disabled RIF.
+	# When ipv6 address is flushed, there is a delay and the routes are
+	# deleted before the RIF and we cannot get state that we have route
+	# to disabled RIF.
+	# Delete IPv6 address first and then check this trap with flushing IPv4.
+	ip -6 add flush dev br0
+	ip -4 add flush dev br0
+
+	t1_packets=$(devlink_trap_rx_packets_get $trap_name)
+	t1_bytes=$(devlink_trap_rx_bytes_get $trap_name)
+
+	if [[ $t0_packets -eq $t1_packets && $t0_bytes -eq $t1_bytes ]]; then
+		check_err 1 "Trap stats idle when packets should be trapped"
+	fi
+
+	log_test "Egress RIF disabled"
+
+	kill $mz_pid && wait $mz_pid &> /dev/null
+	ip link set dev $rp1 nomaster
+	__addr_add_del $rp1 add 192.0.2.2/24 2001:db8:1::2/64
+	ip link del dev br0 type bridge
+	devlink_trap_action_set $trap_name "drop"
+}
+
 trap cleanup EXIT
 
 setup_prepare
-- 
cgit v1.2.3-59-g8ed1b


From 95f0ead8f04bec18e474594ef585f3734bd85b4c Mon Sep 17 00:00:00 2001
From: Amit Cohen <amitc@mellanox.com>
Date: Sun, 19 Jan 2020 15:00:48 +0200
Subject: devlink: Add non-routable packet trap

Add packet trap that can report packets that reached the router, but are
non-routable. For example, IGMP queries can be flooded by the device in
layer 2 and reach the router. Such packets should not be routed and
instead dropped.

Signed-off-by: Amit Cohen <amitc@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/devlink/devlink-trap.rst | 6 ++++++
 include/net/devlink.h                             | 3 +++
 net/core/devlink.c                                | 1 +
 3 files changed, 10 insertions(+)

diff --git a/Documentation/networking/devlink/devlink-trap.rst b/Documentation/networking/devlink/devlink-trap.rst
index 68245ea387ad..a62e1d6eb3e4 100644
--- a/Documentation/networking/devlink/devlink-trap.rst
+++ b/Documentation/networking/devlink/devlink-trap.rst
@@ -223,6 +223,12 @@ be added to the following table:
    * - ``ipv6_lpm_miss``
      - ``exception``
      - Traps unicast IPv6 packets that did not match any route
+   * - ``non_routable_packet``
+     - ``drop``
+     - Traps packets that the device decided to drop because they are not
+       supposed to be routed. For example, IGMP queries can be flooded by the
+       device in layer 2 and reach the router. Such packets should not be
+       routed and instead dropped
 
 Driver-specific Packet Traps
 ============================
diff --git a/include/net/devlink.h b/include/net/devlink.h
index a6856f1d5d1f..08b757753e1c 100644
--- a/include/net/devlink.h
+++ b/include/net/devlink.h
@@ -591,6 +591,7 @@ enum devlink_trap_generic_id {
 	DEVLINK_TRAP_GENERIC_ID_REJECT_ROUTE,
 	DEVLINK_TRAP_GENERIC_ID_IPV4_LPM_UNICAST_MISS,
 	DEVLINK_TRAP_GENERIC_ID_IPV6_LPM_UNICAST_MISS,
+	DEVLINK_TRAP_GENERIC_ID_NON_ROUTABLE,
 
 	/* Add new generic trap IDs above */
 	__DEVLINK_TRAP_GENERIC_ID_MAX,
@@ -659,6 +660,8 @@ enum devlink_trap_group_generic_id {
 	"ipv4_lpm_miss"
 #define DEVLINK_TRAP_GENERIC_NAME_IPV6_LPM_UNICAST_MISS \
 	"ipv6_lpm_miss"
+#define DEVLINK_TRAP_GENERIC_NAME_NON_ROUTABLE \
+	"non_routable_packet"
 
 #define DEVLINK_TRAP_GROUP_GENERIC_NAME_L2_DROPS \
 	"l2_drops"
diff --git a/net/core/devlink.c b/net/core/devlink.c
index d30aa47052aa..c10e38d724bc 100644
--- a/net/core/devlink.c
+++ b/net/core/devlink.c
@@ -7706,6 +7706,7 @@ static const struct devlink_trap devlink_trap_generic[] = {
 	DEVLINK_TRAP(REJECT_ROUTE, EXCEPTION),
 	DEVLINK_TRAP(IPV4_LPM_UNICAST_MISS, EXCEPTION),
 	DEVLINK_TRAP(IPV6_LPM_UNICAST_MISS, EXCEPTION),
+	DEVLINK_TRAP(NON_ROUTABLE, DROP),
 };
 
 #define DEVLINK_TRAP_GROUP(_id)						      \
-- 
cgit v1.2.3-59-g8ed1b


From 196442ec5f13751ed0c66d5c4bc9b58288dea430 Mon Sep 17 00:00:00 2001
From: Amit Cohen <amitc@mellanox.com>
Date: Sun, 19 Jan 2020 15:00:49 +0200
Subject: mlxsw: Add NON_ROUTABLE trap

Add a trap for packets that the device decided to drop because they are
not supposed to be routed. For example, IGMP queries can be flooded by
the device in layer 2 and reach the router. Such packets should not be
routed and instead dropped.

Signed-off-by: Amit Cohen <amitc@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c | 3 +++
 drivers/net/ethernet/mellanox/mlxsw/trap.h          | 1 +
 2 files changed, 4 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c
index 42013fe11131..8706821f5851 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c
@@ -80,6 +80,7 @@ static struct devlink_trap mlxsw_sp_traps_arr[] = {
 	MLXSW_SP_TRAP_EXCEPTION(IPV6_LPM_UNICAST_MISS, L3_DROPS),
 	MLXSW_SP_TRAP_DRIVER_DROP(IRIF_DISABLED, L3_DROPS),
 	MLXSW_SP_TRAP_DRIVER_DROP(ERIF_DISABLED, L3_DROPS),
+	MLXSW_SP_TRAP_DROP(NON_ROUTABLE, L3_DROPS),
 };
 
 static struct mlxsw_listener mlxsw_sp_listeners_arr[] = {
@@ -114,6 +115,7 @@ static struct mlxsw_listener mlxsw_sp_listeners_arr[] = {
 			       TRAP_EXCEPTION_TO_CPU),
 	MLXSW_SP_RXL_DISCARD(ROUTER_IRIF_EN, L3_DISCARDS),
 	MLXSW_SP_RXL_DISCARD(ROUTER_ERIF_EN, L3_DISCARDS),
+	MLXSW_SP_RXL_DISCARD(NON_ROUTABLE, L3_DISCARDS),
 };
 
 /* Mapping between hardware trap and devlink trap. Multiple hardware traps can
@@ -149,6 +151,7 @@ static u16 mlxsw_sp_listener_devlink_map[] = {
 	DEVLINK_TRAP_GENERIC_ID_IPV6_LPM_UNICAST_MISS,
 	DEVLINK_MLXSW_TRAP_ID_IRIF_DISABLED,
 	DEVLINK_MLXSW_TRAP_ID_ERIF_DISABLED,
+	DEVLINK_TRAP_GENERIC_ID_NON_ROUTABLE,
 };
 
 static int mlxsw_sp_rx_listener(struct mlxsw_sp *mlxsw_sp, struct sk_buff *skb,
diff --git a/drivers/net/ethernet/mellanox/mlxsw/trap.h b/drivers/net/ethernet/mellanox/mlxsw/trap.h
index 8573cc3a0010..7d07ec8d440b 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/trap.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/trap.h
@@ -67,6 +67,7 @@ enum {
 	MLXSW_TRAP_ID_NVE_ENCAP_ARP = 0xBD,
 	MLXSW_TRAP_ID_ROUTER_ALERT_IPV4 = 0xD6,
 	MLXSW_TRAP_ID_ROUTER_ALERT_IPV6 = 0xD7,
+	MLXSW_TRAP_ID_DISCARD_NON_ROUTABLE = 0x11A,
 	MLXSW_TRAP_ID_DISCARD_ROUTER2 = 0x130,
 	MLXSW_TRAP_ID_DISCARD_ROUTER3 = 0x131,
 	MLXSW_TRAP_ID_DISCARD_ING_PACKET_SMAC_MC = 0x140,
-- 
cgit v1.2.3-59-g8ed1b


From 20174900ad38cb090616add90665e2ff41c29dc9 Mon Sep 17 00:00:00 2001
From: Amit Cohen <amitc@mellanox.com>
Date: Sun, 19 Jan 2020 15:00:50 +0200
Subject: mlxsw: reg: Add Tunneling IPinIP Encapsulation ECN Mapping Register

This register performs mapping from overlay ECN to underlay ECN during
IPinIP encapsulation.

Signed-off-by: Amit Cohen <amitc@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/reg.h | 31 +++++++++++++++++++++++++++++++
 1 file changed, 31 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h
index fd59280cf979..19a84641d485 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/reg.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h
@@ -10140,6 +10140,36 @@ static inline void mlxsw_reg_tigcr_pack(char *payload, bool ttlc, u8 ttl_uc)
 	mlxsw_reg_tigcr_ttl_uc_set(payload, ttl_uc);
 }
 
+/* TIEEM - Tunneling IPinIP Encapsulation ECN Mapping Register
+ * -----------------------------------------------------------
+ * The TIEEM register maps ECN of the IP header at the ingress to the
+ * encapsulation to the ECN of the underlay network.
+ */
+#define MLXSW_REG_TIEEM_ID 0xA812
+#define MLXSW_REG_TIEEM_LEN 0x0C
+
+MLXSW_REG_DEFINE(tieem, MLXSW_REG_TIEEM_ID, MLXSW_REG_TIEEM_LEN);
+
+/* reg_tieem_overlay_ecn
+ * ECN of the IP header in the overlay network.
+ * Access: Index
+ */
+MLXSW_ITEM32(reg, tieem, overlay_ecn, 0x04, 24, 2);
+
+/* reg_tineem_underlay_ecn
+ * ECN of the IP header in the underlay network.
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, tieem, underlay_ecn, 0x04, 16, 2);
+
+static inline void mlxsw_reg_tieem_pack(char *payload, u8 overlay_ecn,
+					u8 underlay_ecn)
+{
+	MLXSW_REG_ZERO(tieem, payload);
+	mlxsw_reg_tieem_overlay_ecn_set(payload, overlay_ecn);
+	mlxsw_reg_tieem_underlay_ecn_set(payload, underlay_ecn);
+}
+
 /* SBPR - Shared Buffer Pools Register
  * -----------------------------------
  * The SBPR configures and retrieves the shared buffer pools and configuration.
@@ -10684,6 +10714,7 @@ static const struct mlxsw_reg_info *mlxsw_reg_infos[] = {
 	MLXSW_REG(tndem),
 	MLXSW_REG(tnpc),
 	MLXSW_REG(tigcr),
+	MLXSW_REG(tieem),
 	MLXSW_REG(sbpr),
 	MLXSW_REG(sbcm),
 	MLXSW_REG(sbpm),
-- 
cgit v1.2.3-59-g8ed1b


From 839607e2ec8dbe7c3e43995921dae487976e8899 Mon Sep 17 00:00:00 2001
From: Amit Cohen <amitc@mellanox.com>
Date: Sun, 19 Jan 2020 15:00:51 +0200
Subject: mlxsw: reg: Add Tunneling IPinIP Decapsulation ECN Mapping Register

This register configures the actions that are done during IPinIP
decapsulation based on the ECN bits.

Signed-off-by: Amit Cohen <amitc@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/reg.h | 57 +++++++++++++++++++++++++++++++
 1 file changed, 57 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h
index 19a84641d485..47c29776d5d9 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/reg.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h
@@ -10170,6 +10170,62 @@ static inline void mlxsw_reg_tieem_pack(char *payload, u8 overlay_ecn,
 	mlxsw_reg_tieem_underlay_ecn_set(payload, underlay_ecn);
 }
 
+/* TIDEM - Tunneling IPinIP Decapsulation ECN Mapping Register
+ * -----------------------------------------------------------
+ * The TIDEM register configures the actions that are done in the
+ * decapsulation.
+ */
+#define MLXSW_REG_TIDEM_ID 0xA813
+#define MLXSW_REG_TIDEM_LEN 0x0C
+
+MLXSW_REG_DEFINE(tidem, MLXSW_REG_TIDEM_ID, MLXSW_REG_TIDEM_LEN);
+
+/* reg_tidem_underlay_ecn
+ * ECN field of the IP header in the underlay network.
+ * Access: Index
+ */
+MLXSW_ITEM32(reg, tidem, underlay_ecn, 0x04, 24, 2);
+
+/* reg_tidem_overlay_ecn
+ * ECN field of the IP header in the overlay network.
+ * Access: Index
+ */
+MLXSW_ITEM32(reg, tidem, overlay_ecn, 0x04, 16, 2);
+
+/* reg_tidem_eip_ecn
+ * Egress IP ECN. ECN field of the IP header of the packet which goes out
+ * from the decapsulation.
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, tidem, eip_ecn, 0x04, 8, 2);
+
+/* reg_tidem_trap_en
+ * Trap enable:
+ * 0 - No trap due to decap ECN
+ * 1 - Trap enable with trap_id
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, tidem, trap_en, 0x08, 28, 4);
+
+/* reg_tidem_trap_id
+ * Trap ID. Either DECAP_ECN0 or DECAP_ECN1.
+ * Reserved when trap_en is '0'.
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, tidem, trap_id, 0x08, 0, 9);
+
+static inline void mlxsw_reg_tidem_pack(char *payload, u8 underlay_ecn,
+					u8 overlay_ecn, u8 eip_ecn,
+					bool trap_en, u16 trap_id)
+{
+	MLXSW_REG_ZERO(tidem, payload);
+	mlxsw_reg_tidem_underlay_ecn_set(payload, underlay_ecn);
+	mlxsw_reg_tidem_overlay_ecn_set(payload, overlay_ecn);
+	mlxsw_reg_tidem_eip_ecn_set(payload, eip_ecn);
+	mlxsw_reg_tidem_trap_en_set(payload, trap_en);
+	mlxsw_reg_tidem_trap_id_set(payload, trap_id);
+}
+
 /* SBPR - Shared Buffer Pools Register
  * -----------------------------------
  * The SBPR configures and retrieves the shared buffer pools and configuration.
@@ -10715,6 +10771,7 @@ static const struct mlxsw_reg_info *mlxsw_reg_infos[] = {
 	MLXSW_REG(tnpc),
 	MLXSW_REG(tigcr),
 	MLXSW_REG(tieem),
+	MLXSW_REG(tidem),
 	MLXSW_REG(sbpr),
 	MLXSW_REG(sbcm),
 	MLXSW_REG(sbpm),
-- 
cgit v1.2.3-59-g8ed1b


From 4a44ee67a7cd6603d28d71f8fe5e4e0e4dfae1e5 Mon Sep 17 00:00:00 2001
From: Amit Cohen <amitc@mellanox.com>
Date: Sun, 19 Jan 2020 15:00:52 +0200
Subject: mlxsw: Add ECN configurations with IPinIP tunnels

Initialize ECN mapping registers during router init according to
INET_ECN_encapsulate() and INET_ECN_decapsulate().

For IP-in-IP encapsulation, this is required to ensure that ECN bits in
the underlay are set in accordance with the kernel. For decapsulation,
this is required to ensure that packets with invalid ECN combination in
underlay and overlay are trapped to the kernel and not forwarded.

Signed-off-by: Amit Cohen <amitc@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../net/ethernet/mellanox/mlxsw/spectrum_ipip.c    | 60 ++++++++++++++++++++++
 .../net/ethernet/mellanox/mlxsw/spectrum_router.c  | 10 ++++
 .../net/ethernet/mellanox/mlxsw/spectrum_router.h  |  3 ++
 3 files changed, 73 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ipip.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ipip.c
index 6400cd644b7a..a8525992528f 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ipip.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ipip.c
@@ -3,8 +3,10 @@
 
 #include <net/ip_tunnels.h>
 #include <net/ip6_tunnel.h>
+#include <net/inet_ecn.h>
 
 #include "spectrum_ipip.h"
+#include "reg.h"
 
 struct ip_tunnel_parm
 mlxsw_sp_ipip_netdev_parms4(const struct net_device *ol_dev)
@@ -338,3 +340,61 @@ static const struct mlxsw_sp_ipip_ops mlxsw_sp_ipip_gre4_ops = {
 const struct mlxsw_sp_ipip_ops *mlxsw_sp_ipip_ops_arr[] = {
 	[MLXSW_SP_IPIP_TYPE_GRE4] = &mlxsw_sp_ipip_gre4_ops,
 };
+
+static int mlxsw_sp_ipip_ecn_encap_init_one(struct mlxsw_sp *mlxsw_sp,
+					    u8 inner_ecn, u8 outer_ecn)
+{
+	char tieem_pl[MLXSW_REG_TIEEM_LEN];
+
+	mlxsw_reg_tieem_pack(tieem_pl, inner_ecn, outer_ecn);
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(tieem), tieem_pl);
+}
+
+int mlxsw_sp_ipip_ecn_encap_init(struct mlxsw_sp *mlxsw_sp)
+{
+	int i;
+
+	/* Iterate over inner ECN values */
+	for (i = INET_ECN_NOT_ECT; i <= INET_ECN_CE; i++) {
+		u8 outer_ecn = INET_ECN_encapsulate(0, i);
+		int err;
+
+		err = mlxsw_sp_ipip_ecn_encap_init_one(mlxsw_sp, i, outer_ecn);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static int mlxsw_sp_ipip_ecn_decap_init_one(struct mlxsw_sp *mlxsw_sp,
+					    u8 inner_ecn, u8 outer_ecn)
+{
+	char tidem_pl[MLXSW_REG_TIDEM_LEN];
+	bool trap_en, set_ce = false;
+	u8 new_inner_ecn;
+
+	trap_en = __INET_ECN_decapsulate(outer_ecn, inner_ecn, &set_ce);
+	new_inner_ecn = set_ce ? INET_ECN_CE : inner_ecn;
+
+	mlxsw_reg_tidem_pack(tidem_pl, outer_ecn, inner_ecn, new_inner_ecn,
+			     trap_en, trap_en ? MLXSW_TRAP_ID_DECAP_ECN0 : 0);
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(tidem), tidem_pl);
+}
+
+int mlxsw_sp_ipip_ecn_decap_init(struct mlxsw_sp *mlxsw_sp)
+{
+	int i, j, err;
+
+	/* Iterate over inner ECN values */
+	for (i = INET_ECN_NOT_ECT; i <= INET_ECN_CE; i++) {
+		/* Iterate over outer ECN values */
+		for (j = INET_ECN_NOT_ECT; j <= INET_ECN_CE; j++) {
+			err = mlxsw_sp_ipip_ecn_decap_init_one(mlxsw_sp, i, j);
+			if (err)
+				return err;
+		}
+	}
+
+	return 0;
+}
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
index a014bd8d504d..ce707723f8cf 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
@@ -7791,8 +7791,18 @@ mlxsw_sp_ipip_config_tigcr(struct mlxsw_sp *mlxsw_sp)
 
 static int mlxsw_sp_ipips_init(struct mlxsw_sp *mlxsw_sp)
 {
+	int err;
+
 	mlxsw_sp->router->ipip_ops_arr = mlxsw_sp_ipip_ops_arr;
 	INIT_LIST_HEAD(&mlxsw_sp->router->ipip_list);
+
+	err = mlxsw_sp_ipip_ecn_encap_init(mlxsw_sp);
+	if (err)
+		return err;
+	err = mlxsw_sp_ipip_ecn_decap_init(mlxsw_sp);
+	if (err)
+		return err;
+
 	return mlxsw_sp_ipip_config_tigcr(mlxsw_sp);
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.h
index cc1de91e8217..c9b94f435cdd 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.h
@@ -104,4 +104,7 @@ static inline bool mlxsw_sp_l3addr_eq(const union mlxsw_sp_l3addr *addr1,
 	return !memcmp(addr1, addr2, sizeof(*addr1));
 }
 
+int mlxsw_sp_ipip_ecn_encap_init(struct mlxsw_sp *mlxsw_sp);
+int mlxsw_sp_ipip_ecn_decap_init(struct mlxsw_sp *mlxsw_sp);
+
 #endif /* _MLXSW_ROUTER_H_*/
-- 
cgit v1.2.3-59-g8ed1b


From f528dfc460ef015bb612595112508ef469d0d3b9 Mon Sep 17 00:00:00 2001
From: Amit Cohen <amitc@mellanox.com>
Date: Sun, 19 Jan 2020 15:00:53 +0200
Subject: mlxsw: spectrum_trap: Reorder cases according to enum order

Move L3_DROPS case to appear after L2_DROPS case.

Signed-off-by: Amit Cohen <amitc@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c
index 8706821f5851..5ae60eb11a2d 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c
@@ -333,8 +333,8 @@ mlxsw_sp_trap_group_policer_init(struct mlxsw_sp *mlxsw_sp,
 	u32 rate;
 
 	switch (group->id) {
-	case DEVLINK_TRAP_GROUP_GENERIC_ID_L3_DROPS:/* fall through */
-	case DEVLINK_TRAP_GROUP_GENERIC_ID_L2_DROPS:
+	case DEVLINK_TRAP_GROUP_GENERIC_ID_L2_DROPS: /* fall through */
+	case DEVLINK_TRAP_GROUP_GENERIC_ID_L3_DROPS:
 		policer_id = MLXSW_SP_DISCARD_POLICER_ID;
 		ir_units = MLXSW_REG_QPCR_IR_UNITS_M;
 		is_bytes = false;
-- 
cgit v1.2.3-59-g8ed1b


From 13c056ec7d006b11557cebd9f1803edd646d2876 Mon Sep 17 00:00:00 2001
From: Amit Cohen <amitc@mellanox.com>
Date: Sun, 19 Jan 2020 15:00:54 +0200
Subject: devlink: Add tunnel generic packet traps

Add packet traps that can report packets that were dropped during tunnel
decapsulation.

Signed-off-by: Amit Cohen <amitc@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/devlink/devlink-trap.rst | 8 ++++++++
 include/net/devlink.h                             | 6 ++++++
 net/core/devlink.c                                | 2 ++
 3 files changed, 16 insertions(+)

diff --git a/Documentation/networking/devlink/devlink-trap.rst b/Documentation/networking/devlink/devlink-trap.rst
index a62e1d6eb3e4..014f0a34c0e4 100644
--- a/Documentation/networking/devlink/devlink-trap.rst
+++ b/Documentation/networking/devlink/devlink-trap.rst
@@ -229,6 +229,11 @@ be added to the following table:
        supposed to be routed. For example, IGMP queries can be flooded by the
        device in layer 2 and reach the router. Such packets should not be
        routed and instead dropped
+   * - ``decap_error``
+     - ``exception``
+     - Traps NVE and IPinIP packets that the device decided to drop because of
+       failure during decapsulation (e.g., packet being too short, reserved
+       bits set in VXLAN header)
 
 Driver-specific Packet Traps
 ============================
@@ -265,6 +270,9 @@ narrow. The description of these groups must be added to the following table:
    * - ``buffer_drops``
      - Contains packet traps for packets that were dropped by the device due to
        an enqueue decision
+   * - ``tunnel_drops``
+     - Contains packet traps for packets that were dropped by the device during
+       tunnel encapsulation / decapsulation
 
 Testing
 =======
diff --git a/include/net/devlink.h b/include/net/devlink.h
index 08b757753e1c..455282a4b714 100644
--- a/include/net/devlink.h
+++ b/include/net/devlink.h
@@ -592,6 +592,7 @@ enum devlink_trap_generic_id {
 	DEVLINK_TRAP_GENERIC_ID_IPV4_LPM_UNICAST_MISS,
 	DEVLINK_TRAP_GENERIC_ID_IPV6_LPM_UNICAST_MISS,
 	DEVLINK_TRAP_GENERIC_ID_NON_ROUTABLE,
+	DEVLINK_TRAP_GENERIC_ID_DECAP_ERROR,
 
 	/* Add new generic trap IDs above */
 	__DEVLINK_TRAP_GENERIC_ID_MAX,
@@ -605,6 +606,7 @@ enum devlink_trap_group_generic_id {
 	DEVLINK_TRAP_GROUP_GENERIC_ID_L2_DROPS,
 	DEVLINK_TRAP_GROUP_GENERIC_ID_L3_DROPS,
 	DEVLINK_TRAP_GROUP_GENERIC_ID_BUFFER_DROPS,
+	DEVLINK_TRAP_GROUP_GENERIC_ID_TUNNEL_DROPS,
 
 	/* Add new generic trap group IDs above */
 	__DEVLINK_TRAP_GROUP_GENERIC_ID_MAX,
@@ -662,6 +664,8 @@ enum devlink_trap_group_generic_id {
 	"ipv6_lpm_miss"
 #define DEVLINK_TRAP_GENERIC_NAME_NON_ROUTABLE \
 	"non_routable_packet"
+#define DEVLINK_TRAP_GENERIC_NAME_DECAP_ERROR \
+	"decap_error"
 
 #define DEVLINK_TRAP_GROUP_GENERIC_NAME_L2_DROPS \
 	"l2_drops"
@@ -669,6 +673,8 @@ enum devlink_trap_group_generic_id {
 	"l3_drops"
 #define DEVLINK_TRAP_GROUP_GENERIC_NAME_BUFFER_DROPS \
 	"buffer_drops"
+#define DEVLINK_TRAP_GROUP_GENERIC_NAME_TUNNEL_DROPS \
+	"tunnel_drops"
 
 #define DEVLINK_TRAP_GENERIC(_type, _init_action, _id, _group, _metadata_cap) \
 	{								      \
diff --git a/net/core/devlink.c b/net/core/devlink.c
index c10e38d724bc..af85fcd9b01e 100644
--- a/net/core/devlink.c
+++ b/net/core/devlink.c
@@ -7707,6 +7707,7 @@ static const struct devlink_trap devlink_trap_generic[] = {
 	DEVLINK_TRAP(IPV4_LPM_UNICAST_MISS, EXCEPTION),
 	DEVLINK_TRAP(IPV6_LPM_UNICAST_MISS, EXCEPTION),
 	DEVLINK_TRAP(NON_ROUTABLE, DROP),
+	DEVLINK_TRAP(DECAP_ERROR, EXCEPTION),
 };
 
 #define DEVLINK_TRAP_GROUP(_id)						      \
@@ -7719,6 +7720,7 @@ static const struct devlink_trap_group devlink_trap_group_generic[] = {
 	DEVLINK_TRAP_GROUP(L2_DROPS),
 	DEVLINK_TRAP_GROUP(L3_DROPS),
 	DEVLINK_TRAP_GROUP(BUFFER_DROPS),
+	DEVLINK_TRAP_GROUP(TUNNEL_DROPS),
 };
 
 static int devlink_trap_generic_verify(const struct devlink_trap *trap)
-- 
cgit v1.2.3-59-g8ed1b


From a318bf621a6eabc45e51ed9297b4c01991bfd7e4 Mon Sep 17 00:00:00 2001
From: Amit Cohen <amitc@mellanox.com>
Date: Sun, 19 Jan 2020 15:00:55 +0200
Subject: mlxsw: Add tunnel devlink-trap support

Add the trap IDs and trap group used to report tunnel drops. Register
tunnel packet traps and associated tunnel trap group with devlink
during driver initialization.

Signed-off-by: Amit Cohen <amitc@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/reg.h           |  1 +
 drivers/net/ethernet/mellanox/mlxsw/spectrum.c      |  2 --
 drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c | 18 +++++++++++++++++-
 drivers/net/ethernet/mellanox/mlxsw/trap.h          |  1 +
 4 files changed, 19 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h
index 47c29776d5d9..0b80e75e87c3 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/reg.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h
@@ -5513,6 +5513,7 @@ enum mlxsw_reg_htgt_discard_trap_group {
 	MLXSW_REG_HTGT_DISCARD_TRAP_GROUP_BASE = MLXSW_REG_HTGT_TRAP_GROUP_MAX,
 	MLXSW_REG_HTGT_TRAP_GROUP_SP_L2_DISCARDS,
 	MLXSW_REG_HTGT_TRAP_GROUP_SP_L3_DISCARDS,
+	MLXSW_REG_HTGT_TRAP_GROUP_SP_TUNNEL_DISCARDS,
 };
 
 /* reg_htgt_trap_group
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index 2f0b516ee03f..cc30a69a6df0 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -4538,8 +4538,6 @@ static const struct mlxsw_listener mlxsw_sp_listener[] = {
 			  false),
 	MLXSW_SP_RXL_MARK(ROUTER_ALERT_IPV4, TRAP_TO_CPU, ROUTER_EXP, false),
 	MLXSW_SP_RXL_MARK(ROUTER_ALERT_IPV6, TRAP_TO_CPU, ROUTER_EXP, false),
-	MLXSW_SP_RXL_MARK(IPIP_DECAP_ERROR, TRAP_TO_CPU, ROUTER_EXP, false),
-	MLXSW_SP_RXL_MARK(DECAP_ECN0, TRAP_TO_CPU, ROUTER_EXP, false),
 	MLXSW_SP_RXL_MARK(IPV4_VRRP, TRAP_TO_CPU, VRRP, false),
 	MLXSW_SP_RXL_MARK(IPV6_VRRP, TRAP_TO_CPU, VRRP, false),
 	MLXSW_SP_RXL_NO_MARK(DISCARD_ING_ROUTER_SIP_CLASS_E, FORWARD,
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c
index 5ae60eb11a2d..b03bb3a54fc8 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c
@@ -81,6 +81,7 @@ static struct devlink_trap mlxsw_sp_traps_arr[] = {
 	MLXSW_SP_TRAP_DRIVER_DROP(IRIF_DISABLED, L3_DROPS),
 	MLXSW_SP_TRAP_DRIVER_DROP(ERIF_DISABLED, L3_DROPS),
 	MLXSW_SP_TRAP_DROP(NON_ROUTABLE, L3_DROPS),
+	MLXSW_SP_TRAP_EXCEPTION(DECAP_ERROR, TUNNEL_DROPS),
 };
 
 static struct mlxsw_listener mlxsw_sp_listeners_arr[] = {
@@ -116,6 +117,11 @@ static struct mlxsw_listener mlxsw_sp_listeners_arr[] = {
 	MLXSW_SP_RXL_DISCARD(ROUTER_IRIF_EN, L3_DISCARDS),
 	MLXSW_SP_RXL_DISCARD(ROUTER_ERIF_EN, L3_DISCARDS),
 	MLXSW_SP_RXL_DISCARD(NON_ROUTABLE, L3_DISCARDS),
+	MLXSW_SP_RXL_EXCEPTION(DECAP_ECN0, ROUTER_EXP, TRAP_EXCEPTION_TO_CPU),
+	MLXSW_SP_RXL_EXCEPTION(IPIP_DECAP_ERROR, ROUTER_EXP,
+			       TRAP_EXCEPTION_TO_CPU),
+	MLXSW_SP_RXL_EXCEPTION(DISCARD_DEC_PKT, TUNNEL_DISCARDS,
+			       TRAP_EXCEPTION_TO_CPU),
 };
 
 /* Mapping between hardware trap and devlink trap. Multiple hardware traps can
@@ -152,6 +158,9 @@ static u16 mlxsw_sp_listener_devlink_map[] = {
 	DEVLINK_MLXSW_TRAP_ID_IRIF_DISABLED,
 	DEVLINK_MLXSW_TRAP_ID_ERIF_DISABLED,
 	DEVLINK_TRAP_GENERIC_ID_NON_ROUTABLE,
+	DEVLINK_TRAP_GENERIC_ID_DECAP_ERROR,
+	DEVLINK_TRAP_GENERIC_ID_DECAP_ERROR,
+	DEVLINK_TRAP_GENERIC_ID_DECAP_ERROR,
 };
 
 static int mlxsw_sp_rx_listener(struct mlxsw_sp *mlxsw_sp, struct sk_buff *skb,
@@ -334,7 +343,8 @@ mlxsw_sp_trap_group_policer_init(struct mlxsw_sp *mlxsw_sp,
 
 	switch (group->id) {
 	case DEVLINK_TRAP_GROUP_GENERIC_ID_L2_DROPS: /* fall through */
-	case DEVLINK_TRAP_GROUP_GENERIC_ID_L3_DROPS:
+	case DEVLINK_TRAP_GROUP_GENERIC_ID_L3_DROPS: /* fall through */
+	case DEVLINK_TRAP_GROUP_GENERIC_ID_TUNNEL_DROPS:
 		policer_id = MLXSW_SP_DISCARD_POLICER_ID;
 		ir_units = MLXSW_REG_QPCR_IR_UNITS_M;
 		is_bytes = false;
@@ -371,6 +381,12 @@ __mlxsw_sp_trap_group_init(struct mlxsw_sp *mlxsw_sp,
 		priority = 0;
 		tc = 1;
 		break;
+	case DEVLINK_TRAP_GROUP_GENERIC_ID_TUNNEL_DROPS:
+		group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_TUNNEL_DISCARDS;
+		policer_id = MLXSW_SP_DISCARD_POLICER_ID;
+		priority = 0;
+		tc = 1;
+		break;
 	default:
 		return -EINVAL;
 	}
diff --git a/drivers/net/ethernet/mellanox/mlxsw/trap.h b/drivers/net/ethernet/mellanox/mlxsw/trap.h
index 7d07ec8d440b..24f92b157dfb 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/trap.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/trap.h
@@ -93,6 +93,7 @@ enum {
 	MLXSW_TRAP_ID_DISCARD_ROUTER_ERIF_EN = 0x179,
 	MLXSW_TRAP_ID_DISCARD_ROUTER_LPM4 = 0x17B,
 	MLXSW_TRAP_ID_DISCARD_ROUTER_LPM6 = 0x17C,
+	MLXSW_TRAP_ID_DISCARD_DEC_PKT = 0x188,
 	MLXSW_TRAP_ID_DISCARD_IPV6_MC_DIP_RESERVED_SCOPE = 0x1B0,
 	MLXSW_TRAP_ID_DISCARD_IPV6_MC_DIP_INTERFACE_LOCAL_SCOPE = 0x1B1,
 	MLXSW_TRAP_ID_ACL0 = 0x1C0,
-- 
cgit v1.2.3-59-g8ed1b


From ca264ef6ed7cc1f15201d298a44fb79e2c63449c Mon Sep 17 00:00:00 2001
From: Amit Cohen <amitc@mellanox.com>
Date: Sun, 19 Jan 2020 15:00:56 +0200
Subject: selftests: devlink_trap_tunnel_vxlan: Add test case for decap_error

Test that the trap is triggered under the right conditions and that
devlink counters increase.

Signed-off-by: Amit Cohen <amitc@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../drivers/net/mlxsw/devlink_trap_tunnel_vxlan.sh | 276 +++++++++++++++++++++
 1 file changed, 276 insertions(+)
 create mode 100755 tools/testing/selftests/drivers/net/mlxsw/devlink_trap_tunnel_vxlan.sh

diff --git a/tools/testing/selftests/drivers/net/mlxsw/devlink_trap_tunnel_vxlan.sh b/tools/testing/selftests/drivers/net/mlxsw/devlink_trap_tunnel_vxlan.sh
new file mode 100755
index 000000000000..a699edae8358
--- /dev/null
+++ b/tools/testing/selftests/drivers/net/mlxsw/devlink_trap_tunnel_vxlan.sh
@@ -0,0 +1,276 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+#
+# Test devlink-trap tunnel drops and exceptions functionality over mlxsw.
+# Check all traps to make sure they are triggered under the right
+# conditions.
+
+# +--------------------+
+# | H1 (vrf)           |
+# |    + $h1           |
+# |    | 192.0.2.1/28  |
+# +----|---------------+
+#      |
+# +----|----------------------------------------------------------------------+
+# | SW |                                                                      |
+# | +--|--------------------------------------------------------------------+ |
+# | |  + $swp1                   BR1 (802.1d)                               | |
+# | |                                                                       | |
+# | |  + vx1 (vxlan)                                                        | |
+# | |    local 192.0.2.17                                                   | |
+# | |    id 1000 dstport $VXPORT                                            | |
+# | +-----------------------------------------------------------------------+ |
+# |                                                                           |
+# |    + $rp1                                                                 |
+# |    | 192.0.2.17/28                                                        |
+# +----|----------------------------------------------------------------------+
+#      |
+# +----|--------------------------------------------------------+
+# |    |                                             VRF2       |
+# |    + $rp2                                                   |
+# |      192.0.2.18/28                                          |
+# |                                                             |
+# +-------------------------------------------------------------+
+
+lib_dir=$(dirname $0)/../../../net/forwarding
+
+ALL_TESTS="
+	decap_error_test
+"
+
+NUM_NETIFS=4
+source $lib_dir/lib.sh
+source $lib_dir/tc_common.sh
+source $lib_dir/devlink_lib.sh
+
+: ${VXPORT:=4789}
+export VXPORT
+
+h1_create()
+{
+	simple_if_init $h1 192.0.2.1/28
+}
+
+h1_destroy()
+{
+	simple_if_fini $h1 192.0.2.1/28
+}
+
+switch_create()
+{
+	ip link add name br1 type bridge vlan_filtering 0 mcast_snooping 0
+	# Make sure the bridge uses the MAC address of the local port and not
+	# that of the VxLAN's device.
+	ip link set dev br1 address $(mac_get $swp1)
+	ip link set dev br1 up
+
+	tc qdisc add dev $swp1 clsact
+	ip link set dev $swp1 master br1
+	ip link set dev $swp1 up
+
+	ip link add name vx1 type vxlan id 1000 local 192.0.2.17 \
+		dstport "$VXPORT" nolearning noudpcsum tos inherit ttl 100
+	ip link set dev vx1 master br1
+	ip link set dev vx1 up
+
+	ip address add dev $rp1 192.0.2.17/28
+	ip link set dev $rp1 up
+}
+
+switch_destroy()
+{
+	ip link set dev $rp1 down
+	ip address del dev $rp1 192.0.2.17/28
+
+	ip link set dev vx1 down
+	ip link set dev vx1 nomaster
+	ip link del dev vx1
+
+	ip link set dev $swp1 down
+	ip link set dev $swp1 nomaster
+	tc qdisc del dev $swp1 clsact
+
+	ip link set dev br1 down
+	ip link del dev br1
+}
+
+vrf2_create()
+{
+	simple_if_init $rp2 192.0.2.18/28
+}
+
+vrf2_destroy()
+{
+	simple_if_fini $rp2 192.0.2.18/28
+}
+
+setup_prepare()
+{
+	h1=${NETIFS[p1]}
+	swp1=${NETIFS[p2]}
+
+	rp1=${NETIFS[p3]}
+	rp2=${NETIFS[p4]}
+
+	vrf_prepare
+	forwarding_enable
+	h1_create
+	switch_create
+	vrf2_create
+}
+
+cleanup()
+{
+	pre_cleanup
+
+	vrf2_destroy
+	switch_destroy
+	h1_destroy
+	forwarding_restore
+	vrf_cleanup
+}
+
+ecn_payload_get()
+{
+	dest_mac=$(mac_get $h1)
+	p=$(:
+		)"08:"$(                      : VXLAN flags
+		)"00:00:00:"$(                : VXLAN reserved
+		)"00:03:e8:"$(                : VXLAN VNI : 1000
+		)"00:"$(                      : VXLAN reserved
+		)"$dest_mac:"$(               : ETH daddr
+		)"00:00:00:00:00:00:"$(       : ETH saddr
+		)"08:00:"$(                   : ETH type
+		)"45:"$(                      : IP version + IHL
+		)"00:"$(                      : IP TOS
+		)"00:14:"$(                   : IP total length
+		)"00:00:"$(                   : IP identification
+		)"20:00:"$(                   : IP flags + frag off
+		)"40:"$(                      : IP TTL
+		)"00:"$(                      : IP proto
+		)"D6:E5:"$(                   : IP header csum
+		)"c0:00:02:03:"$(             : IP saddr: 192.0.2.3
+		)"c0:00:02:01:"$(             : IP daddr: 192.0.2.1
+		)
+	echo $p
+}
+
+ecn_decap_test()
+{
+	local trap_name="decap_error"
+	local group_name="tunnel_drops"
+	local desc=$1; shift
+	local ecn_desc=$1; shift
+	local outer_tos=$1; shift
+	local mz_pid
+
+	RET=0
+
+	tc filter add dev $swp1 egress protocol ip pref 1 handle 101 \
+		flower src_ip 192.0.2.3 dst_ip 192.0.2.1 action pass
+
+	rp1_mac=$(mac_get $rp1)
+	payload=$(ecn_payload_get)
+
+	ip vrf exec v$rp2 $MZ $rp2 -c 0 -d 1msec -b $rp1_mac -B 192.0.2.17 \
+		-t udp sp=12345,dp=$VXPORT,tos=$outer_tos,p=$payload -q &
+	mz_pid=$!
+
+	devlink_trap_exception_test $trap_name $group_name
+
+	tc_check_packets "dev $swp1 egress" 101 0
+	check_err $? "Packets were not dropped"
+
+	log_test "$desc: Inner ECN is not ECT and outer is $ecn_desc"
+
+	kill $mz_pid && wait $mz_pid &> /dev/null
+	tc filter del dev $swp1 egress protocol ip pref 1 handle 101 flower
+}
+
+reserved_bits_payload_get()
+{
+	dest_mac=$(mac_get $h1)
+	p=$(:
+		)"08:"$(                      : VXLAN flags
+		)"01:00:00:"$(                : VXLAN reserved
+		)"00:03:e8:"$(                : VXLAN VNI : 1000
+		)"00:"$(                      : VXLAN reserved
+		)"$dest_mac:"$(               : ETH daddr
+		)"00:00:00:00:00:00:"$(       : ETH saddr
+		)"08:00:"$(                   : ETH type
+		)"45:"$(                      : IP version + IHL
+		)"00:"$(                      : IP TOS
+		)"00:14:"$(                   : IP total length
+		)"00:00:"$(                   : IP identification
+		)"20:00:"$(                   : IP flags + frag off
+		)"40:"$(                      : IP TTL
+		)"00:"$(                      : IP proto
+		)"00:00:"$(                   : IP header csum
+		)"c0:00:02:03:"$(             : IP saddr: 192.0.2.3
+		)"c0:00:02:01:"$(             : IP daddr: 192.0.2.1
+		)
+	echo $p
+}
+
+short_payload_get()
+{
+        dest_mac=$(mac_get $h1)
+        p=$(:
+		)"08:"$(                      : VXLAN flags
+		)"01:00:00:"$(                : VXLAN reserved
+		)"00:03:e8:"$(                : VXLAN VNI : 1000
+		)"00:"$(                      : VXLAN reserved
+		)
+        echo $p
+}
+
+corrupted_packet_test()
+{
+	local trap_name="decap_error"
+	local group_name="tunnel_drops"
+	local desc=$1; shift
+	local payload_get=$1; shift
+	local mz_pid
+
+	RET=0
+
+	# In case of too short packet, there is no any inner packet,
+	# so the matching will always succeed
+	tc filter add dev $swp1 egress protocol ip pref 1 handle 101 \
+		flower skip_hw src_ip 192.0.2.3 dst_ip 192.0.2.1 action pass
+
+	rp1_mac=$(mac_get $rp1)
+	payload=$($payload_get)
+	ip vrf exec v$rp2 $MZ $rp2 -c 0 -d 1msec -b $rp1_mac \
+		-B 192.0.2.17 -t udp sp=12345,dp=$VXPORT,p=$payload -q &
+	mz_pid=$!
+
+	devlink_trap_exception_test $trap_name $group_name
+
+	tc_check_packets "dev $swp1 egress" 101 0
+	check_err $? "Packets were not dropped"
+
+	log_test "$desc"
+
+	kill $mz_pid && wait $mz_pid &> /dev/null
+	tc filter del dev $swp1 egress protocol ip pref 1 handle 101 flower
+}
+
+decap_error_test()
+{
+	ecn_decap_test "Decap error" "ECT(1)" 01
+	ecn_decap_test "Decap error" "ECT(0)" 02
+	ecn_decap_test "Decap error" "CE" 03
+
+	corrupted_packet_test "Decap error: Reserved bits in use" \
+		"reserved_bits_payload_get"
+	corrupted_packet_test "Decap error: No L2 header" "short_payload_get"
+}
+
+trap cleanup EXIT
+
+setup_prepare
+setup_wait
+tests_run
+
+exit $EXIT_STATUS
-- 
cgit v1.2.3-59-g8ed1b


From 27942c700017ba84c8fcfcc921bbaf8a91caaefd Mon Sep 17 00:00:00 2001
From: Amit Cohen <amitc@mellanox.com>
Date: Sun, 19 Jan 2020 15:00:57 +0200
Subject: selftests: devlink_trap_tunnel_ipip: Add test case for decap_error

Test that the trap is triggered under the right conditions and that
devlink counters increase.

Signed-off-by: Amit Cohen <amitc@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../drivers/net/mlxsw/devlink_trap_tunnel_ipip.sh  | 265 +++++++++++++++++++++
 1 file changed, 265 insertions(+)
 create mode 100755 tools/testing/selftests/drivers/net/mlxsw/devlink_trap_tunnel_ipip.sh

diff --git a/tools/testing/selftests/drivers/net/mlxsw/devlink_trap_tunnel_ipip.sh b/tools/testing/selftests/drivers/net/mlxsw/devlink_trap_tunnel_ipip.sh
new file mode 100755
index 000000000000..039629bb92a3
--- /dev/null
+++ b/tools/testing/selftests/drivers/net/mlxsw/devlink_trap_tunnel_ipip.sh
@@ -0,0 +1,265 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+#
+# Test devlink-trap tunnel exceptions functionality over mlxsw.
+# Check all exception traps to make sure they are triggered under the right
+# conditions.
+
+# +-------------------------+
+# | H1                      |
+# |               $h1 +     |
+# |      192.0.2.1/28 |     |
+# +-------------------|-----+
+#                     |
+# +-------------------|-----+
+# | SW1               |     |
+# |              $swp1 +    |
+# |      192.0.2.2/28       |
+# |                         |
+# |  + g1a (gre)            |
+# |    loc=192.0.2.65       |
+# |    rem=192.0.2.66       |
+# |    tos=inherit          |
+# |                         |
+# |  + $rp1                 |
+# |  |  198.51.100.1/28     |
+# +--|----------------------+
+#    |
+# +--|----------------------+
+# |  |                 VRF2 |
+# | + $rp2                  |
+# |   198.51.100.2/28       |
+# +-------------------------+
+
+lib_dir=$(dirname $0)/../../../net/forwarding
+
+ALL_TESTS="
+	decap_error_test
+"
+
+NUM_NETIFS=4
+source $lib_dir/lib.sh
+source $lib_dir/tc_common.sh
+source $lib_dir/devlink_lib.sh
+
+h1_create()
+{
+	simple_if_init $h1 192.0.2.1/28
+}
+
+h1_destroy()
+{
+	simple_if_fini $h1 192.0.2.1/28
+}
+
+vrf2_create()
+{
+	simple_if_init $rp2 198.51.100.2/28
+}
+
+vrf2_destroy()
+{
+	simple_if_fini $rp2 198.51.100.2/28
+}
+
+switch_create()
+{
+	__addr_add_del $swp1 add 192.0.2.2/28
+	tc qdisc add dev $swp1 clsact
+	ip link set dev $swp1 up
+
+	tunnel_create g1 gre 192.0.2.65 192.0.2.66 tos inherit
+	__addr_add_del g1 add 192.0.2.65/32
+	ip link set dev g1 up
+
+	__addr_add_del $rp1 add 198.51.100.1/28
+	ip link set dev $rp1 up
+}
+
+switch_destroy()
+{
+	ip link set dev $rp1 down
+	__addr_add_del $rp1 del 198.51.100.1/28
+
+	ip link set dev g1 down
+	__addr_add_del g1 del 192.0.2.65/32
+	tunnel_destroy g1
+
+	ip link set dev $swp1 down
+	tc qdisc del dev $swp1 clsact
+	__addr_add_del $swp1 del 192.0.2.2/28
+}
+
+setup_prepare()
+{
+	h1=${NETIFS[p1]}
+	swp1=${NETIFS[p2]}
+
+	rp1=${NETIFS[p3]}
+	rp2=${NETIFS[p4]}
+
+	forwarding_enable
+	vrf_prepare
+	h1_create
+	switch_create
+	vrf2_create
+}
+
+cleanup()
+{
+	pre_cleanup
+
+	vrf2_destroy
+	switch_destroy
+	h1_destroy
+	vrf_cleanup
+	forwarding_restore
+}
+
+ecn_payload_get()
+{
+	p=$(:
+		)"0"$(		              : GRE flags
+	        )"0:00:"$(                    : Reserved + version
+		)"08:00:"$(		      : ETH protocol type
+		)"4"$(	                      : IP version
+		)"5:"$(                       : IHL
+		)"00:"$(                      : IP TOS
+		)"00:14:"$(                   : IP total length
+		)"00:00:"$(                   : IP identification
+		)"20:00:"$(                   : IP flags + frag off
+		)"30:"$(                      : IP TTL
+		)"01:"$(                      : IP proto
+		)"E7:E6:"$(    	              : IP header csum
+		)"C0:00:01:01:"$(             : IP saddr : 192.0.1.1
+		)"C0:00:02:01:"$(             : IP daddr : 192.0.2.1
+		)
+	echo $p
+}
+
+ecn_decap_test()
+{
+	local trap_name="decap_error"
+	local group_name="tunnel_drops"
+	local desc=$1; shift
+	local ecn_desc=$1; shift
+	local outer_tos=$1; shift
+	local mz_pid
+
+	RET=0
+
+	tc filter add dev $swp1 egress protocol ip pref 1 handle 101 \
+		flower src_ip 192.0.1.1 dst_ip 192.0.2.1 action pass
+
+	rp1_mac=$(mac_get $rp1)
+	rp2_mac=$(mac_get $rp2)
+	payload=$(ecn_payload_get)
+
+	ip vrf exec v$rp2 $MZ $rp2 -c 0 -d 1msec -a $rp2_mac -b $rp1_mac \
+		-A 192.0.2.66 -B 192.0.2.65 -t ip \
+			len=48,tos=$outer_tos,proto=47,p=$payload -q &
+
+	mz_pid=$!
+
+	devlink_trap_exception_test $trap_name $group_name
+
+	tc_check_packets "dev $swp1 egress" 101 0
+	check_err $? "Packets were not dropped"
+
+	log_test "$desc: Inner ECN is not ECT and outer is $ecn_desc"
+
+	kill $mz_pid && wait $mz_pid &> /dev/null
+	tc filter del dev $swp1 egress protocol ip pref 1 handle 101 flower
+}
+
+ipip_payload_get()
+{
+	local flags=$1; shift
+	local key=$1; shift
+
+	p=$(:
+		)"$flags"$(		      : GRE flags
+	        )"0:00:"$(                    : Reserved + version
+		)"08:00:"$(		      : ETH protocol type
+		)"$key"$( 		      : Key
+		)"4"$(	                      : IP version
+		)"5:"$(                       : IHL
+		)"00:"$(                      : IP TOS
+		)"00:14:"$(                   : IP total length
+		)"00:00:"$(                   : IP identification
+		)"20:00:"$(                   : IP flags + frag off
+		)"30:"$(                      : IP TTL
+		)"01:"$(                      : IP proto
+		)"E7:E6:"$(    	              : IP header csum
+		)"C0:00:01:01:"$(             : IP saddr : 192.0.1.1
+		)"C0:00:02:01:"$(             : IP daddr : 192.0.2.1
+		)
+	echo $p
+}
+
+no_matching_tunnel_test()
+{
+	local trap_name="decap_error"
+	local group_name="tunnel_drops"
+	local desc=$1; shift
+	local sip=$1; shift
+	local mz_pid
+
+	RET=0
+
+	tc filter add dev $swp1 egress protocol ip pref 1 handle 101 \
+		flower src_ip 192.0.1.1 dst_ip 192.0.2.1 action pass
+
+	rp1_mac=$(mac_get $rp1)
+	rp2_mac=$(mac_get $rp2)
+	payload=$(ipip_payload_get "$@")
+
+	ip vrf exec v$rp2 $MZ $rp2 -c 0 -d 1msec -a $rp2_mac -b $rp1_mac \
+		-A $sip -B 192.0.2.65 -t ip len=48,proto=47,p=$payload -q &
+	mz_pid=$!
+
+	devlink_trap_exception_test $trap_name $group_name
+
+	tc_check_packets "dev $swp1 egress" 101 0
+	check_err $? "Packets were not dropped"
+
+	log_test "$desc"
+
+	kill $mz_pid && wait $mz_pid &> /dev/null
+	tc filter del dev $swp1 egress protocol ip pref 1 handle 101 flower
+}
+
+decap_error_test()
+{
+	# Correct source IP - the remote address
+	local sip=192.0.2.66
+
+	ecn_decap_test "Decap error" "ECT(1)" 01
+	ecn_decap_test "Decap error" "ECT(0)" 02
+	ecn_decap_test "Decap error" "CE" 03
+
+	no_matching_tunnel_test "Decap error: Source IP check failed" \
+		192.0.2.68 "0"
+	no_matching_tunnel_test \
+		"Decap error: Key exists but was not expected" $sip "2" ":E9:"
+
+	# Destroy the tunnel and create new one with key
+	__addr_add_del g1 del 192.0.2.65/32
+	tunnel_destroy g1
+
+	tunnel_create g1 gre 192.0.2.65 192.0.2.66 tos inherit key 233
+	__addr_add_del g1 add 192.0.2.65/32
+
+	no_matching_tunnel_test \
+		"Decap error: Key does not exist but was expected" $sip "0"
+	no_matching_tunnel_test \
+		"Decap error: Packet has a wrong key field" $sip "2" "E8:"
+}
+
+trap cleanup EXIT
+
+setup_prepare
+setup_wait
+tests_run
+
+exit $EXIT_STATUS
-- 
cgit v1.2.3-59-g8ed1b


From c3cae4916e57d2f0364d5e7769218547fb1b7c60 Mon Sep 17 00:00:00 2001
From: Amit Cohen <amitc@mellanox.com>
Date: Sun, 19 Jan 2020 15:00:58 +0200
Subject: devlink: Add overlay source MAC is multicast trap

Add packet trap that can report NVE packets that the device decided to
drop because their overlay source MAC is multicast.

Signed-off-by: Amit Cohen <amitc@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/devlink/devlink-trap.rst | 4 ++++
 include/net/devlink.h                             | 3 +++
 net/core/devlink.c                                | 1 +
 3 files changed, 8 insertions(+)

diff --git a/Documentation/networking/devlink/devlink-trap.rst b/Documentation/networking/devlink/devlink-trap.rst
index 014f0a34c0e4..47a429bb8658 100644
--- a/Documentation/networking/devlink/devlink-trap.rst
+++ b/Documentation/networking/devlink/devlink-trap.rst
@@ -234,6 +234,10 @@ be added to the following table:
      - Traps NVE and IPinIP packets that the device decided to drop because of
        failure during decapsulation (e.g., packet being too short, reserved
        bits set in VXLAN header)
+   * - ``overlay_smac_is_mc``
+     - ``drop``
+     - Traps NVE packets that the device decided to drop because their overlay
+       source MAC is multicast
 
 Driver-specific Packet Traps
 ============================
diff --git a/include/net/devlink.h b/include/net/devlink.h
index 455282a4b714..2813fd06ee89 100644
--- a/include/net/devlink.h
+++ b/include/net/devlink.h
@@ -593,6 +593,7 @@ enum devlink_trap_generic_id {
 	DEVLINK_TRAP_GENERIC_ID_IPV6_LPM_UNICAST_MISS,
 	DEVLINK_TRAP_GENERIC_ID_NON_ROUTABLE,
 	DEVLINK_TRAP_GENERIC_ID_DECAP_ERROR,
+	DEVLINK_TRAP_GENERIC_ID_OVERLAY_SMAC_MC,
 
 	/* Add new generic trap IDs above */
 	__DEVLINK_TRAP_GENERIC_ID_MAX,
@@ -666,6 +667,8 @@ enum devlink_trap_group_generic_id {
 	"non_routable_packet"
 #define DEVLINK_TRAP_GENERIC_NAME_DECAP_ERROR \
 	"decap_error"
+#define DEVLINK_TRAP_GENERIC_NAME_OVERLAY_SMAC_MC \
+	"overlay_smac_is_mc"
 
 #define DEVLINK_TRAP_GROUP_GENERIC_NAME_L2_DROPS \
 	"l2_drops"
diff --git a/net/core/devlink.c b/net/core/devlink.c
index af85fcd9b01e..e5b19bd2cbe2 100644
--- a/net/core/devlink.c
+++ b/net/core/devlink.c
@@ -7708,6 +7708,7 @@ static const struct devlink_trap devlink_trap_generic[] = {
 	DEVLINK_TRAP(IPV6_LPM_UNICAST_MISS, EXCEPTION),
 	DEVLINK_TRAP(NON_ROUTABLE, DROP),
 	DEVLINK_TRAP(DECAP_ERROR, EXCEPTION),
+	DEVLINK_TRAP(OVERLAY_SMAC_MC, DROP),
 };
 
 #define DEVLINK_TRAP_GROUP(_id)						      \
-- 
cgit v1.2.3-59-g8ed1b


From 3aed0722f7d75240ab7ea0ee2813f6a82f53c60e Mon Sep 17 00:00:00 2001
From: Amit Cohen <amitc@mellanox.com>
Date: Sun, 19 Jan 2020 15:00:59 +0200
Subject: mlxsw: Add OVERLAY_SMAC_MC trap

Add a trap for NVE packets that the device decided to drop because their
overlay source MAC is multicast.

Signed-off-by: Amit Cohen <amitc@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c | 3 +++
 drivers/net/ethernet/mellanox/mlxsw/trap.h          | 1 +
 2 files changed, 4 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c
index b03bb3a54fc8..60205aa3f6a5 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c
@@ -82,6 +82,7 @@ static struct devlink_trap mlxsw_sp_traps_arr[] = {
 	MLXSW_SP_TRAP_DRIVER_DROP(ERIF_DISABLED, L3_DROPS),
 	MLXSW_SP_TRAP_DROP(NON_ROUTABLE, L3_DROPS),
 	MLXSW_SP_TRAP_EXCEPTION(DECAP_ERROR, TUNNEL_DROPS),
+	MLXSW_SP_TRAP_DROP(OVERLAY_SMAC_MC, TUNNEL_DROPS),
 };
 
 static struct mlxsw_listener mlxsw_sp_listeners_arr[] = {
@@ -122,6 +123,7 @@ static struct mlxsw_listener mlxsw_sp_listeners_arr[] = {
 			       TRAP_EXCEPTION_TO_CPU),
 	MLXSW_SP_RXL_EXCEPTION(DISCARD_DEC_PKT, TUNNEL_DISCARDS,
 			       TRAP_EXCEPTION_TO_CPU),
+	MLXSW_SP_RXL_DISCARD(OVERLAY_SMAC_MC, TUNNEL_DISCARDS),
 };
 
 /* Mapping between hardware trap and devlink trap. Multiple hardware traps can
@@ -161,6 +163,7 @@ static u16 mlxsw_sp_listener_devlink_map[] = {
 	DEVLINK_TRAP_GENERIC_ID_DECAP_ERROR,
 	DEVLINK_TRAP_GENERIC_ID_DECAP_ERROR,
 	DEVLINK_TRAP_GENERIC_ID_DECAP_ERROR,
+	DEVLINK_TRAP_GENERIC_ID_OVERLAY_SMAC_MC,
 };
 
 static int mlxsw_sp_rx_listener(struct mlxsw_sp *mlxsw_sp, struct sk_buff *skb,
diff --git a/drivers/net/ethernet/mellanox/mlxsw/trap.h b/drivers/net/ethernet/mellanox/mlxsw/trap.h
index 24f92b157dfb..12e1fa998d42 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/trap.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/trap.h
@@ -94,6 +94,7 @@ enum {
 	MLXSW_TRAP_ID_DISCARD_ROUTER_LPM4 = 0x17B,
 	MLXSW_TRAP_ID_DISCARD_ROUTER_LPM6 = 0x17C,
 	MLXSW_TRAP_ID_DISCARD_DEC_PKT = 0x188,
+	MLXSW_TRAP_ID_DISCARD_OVERLAY_SMAC_MC = 0x190,
 	MLXSW_TRAP_ID_DISCARD_IPV6_MC_DIP_RESERVED_SCOPE = 0x1B0,
 	MLXSW_TRAP_ID_DISCARD_IPV6_MC_DIP_INTERFACE_LOCAL_SCOPE = 0x1B1,
 	MLXSW_TRAP_ID_ACL0 = 0x1C0,
-- 
cgit v1.2.3-59-g8ed1b


From b3073dfba8ee95697ebcafab90489ee847a74433 Mon Sep 17 00:00:00 2001
From: Amit Cohen <amitc@mellanox.com>
Date: Sun, 19 Jan 2020 15:01:00 +0200
Subject: selftests: devlink_trap_tunnel_vxlan: Add test case for
 overlay_smac_is_mc

Test that the trap is triggered under the right conditions and that
devlink counters increase when action is trap.

Signed-off-by: Amit Cohen <amitc@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../drivers/net/mlxsw/devlink_trap_tunnel_vxlan.sh | 54 ++++++++++++++++++++++
 1 file changed, 54 insertions(+)

diff --git a/tools/testing/selftests/drivers/net/mlxsw/devlink_trap_tunnel_vxlan.sh b/tools/testing/selftests/drivers/net/mlxsw/devlink_trap_tunnel_vxlan.sh
index a699edae8358..fd19161dd4ec 100755
--- a/tools/testing/selftests/drivers/net/mlxsw/devlink_trap_tunnel_vxlan.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/devlink_trap_tunnel_vxlan.sh
@@ -36,6 +36,7 @@ lib_dir=$(dirname $0)/../../../net/forwarding
 
 ALL_TESTS="
 	decap_error_test
+	overlay_smac_is_mc_test
 "
 
 NUM_NETIFS=4
@@ -267,6 +268,59 @@ decap_error_test()
 	corrupted_packet_test "Decap error: No L2 header" "short_payload_get"
 }
 
+mc_smac_payload_get()
+{
+	dest_mac=$(mac_get $h1)
+	source_mac=01:02:03:04:05:06
+	p=$(:
+		)"08:"$(                      : VXLAN flags
+		)"00:00:00:"$(                : VXLAN reserved
+		)"00:03:e8:"$(                : VXLAN VNI : 1000
+		)"00:"$(                      : VXLAN reserved
+		)"$dest_mac:"$(               : ETH daddr
+		)"$source_mac:"$(             : ETH saddr
+		)"08:00:"$(                   : ETH type
+		)"45:"$(                      : IP version + IHL
+		)"00:"$(                      : IP TOS
+		)"00:14:"$(                   : IP total length
+		)"00:00:"$(                   : IP identification
+		)"20:00:"$(                   : IP flags + frag off
+		)"40:"$(                      : IP TTL
+		)"00:"$(                      : IP proto
+		)"00:00:"$(                   : IP header csum
+		)"c0:00:02:03:"$(             : IP saddr: 192.0.2.3
+		)"c0:00:02:01:"$(             : IP daddr: 192.0.2.1
+		)
+	echo $p
+}
+
+overlay_smac_is_mc_test()
+{
+	local trap_name="overlay_smac_is_mc"
+	local group_name="tunnel_drops"
+	local mz_pid
+
+	RET=0
+
+	# The matching will be checked on devlink_trap_drop_test()
+	# and the filter will be removed on devlink_trap_drop_cleanup()
+	tc filter add dev $swp1 egress protocol ip pref 1 handle 101 \
+		flower src_mac 01:02:03:04:05:06 action pass
+
+	rp1_mac=$(mac_get $rp1)
+	payload=$(mc_smac_payload_get)
+
+	ip vrf exec v$rp2 $MZ $rp2 -c 0 -d 1msec -b $rp1_mac \
+		-B 192.0.2.17 -t udp sp=12345,dp=$VXPORT,p=$payload -q &
+	mz_pid=$!
+
+	devlink_trap_drop_test $trap_name $group_name $swp1
+
+	log_test "Overlay source MAC is multicast"
+
+	devlink_trap_drop_cleanup $mz_pid $swp1 "ip"
+}
+
 trap cleanup EXIT
 
 setup_prepare
-- 
cgit v1.2.3-59-g8ed1b


From a47b9e15c32f8cd7ad5cb124a54d607b28879f79 Mon Sep 17 00:00:00 2001
From: Dejin Zheng <zhengdejin5@gmail.com>
Date: Wed, 15 Jan 2020 23:53:23 +0800
Subject: net: stmmac: modified pcs mode support for RGMII

snps databook noted that physical coding sublayer (PCS) interface
that can be used when the MAC is configured for the TBI, RTBI, or
SGMII PHY interface. we have RGMII and SGMII in a SoC and it also
has the PCS block. it needs stmmac_init_phy and stmmac_mdio_register
function for initializing phy when it used RGMII interface.

Signed-off-by: Dejin Zheng <zhengdejin5@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 17 ++++++-----------
 1 file changed, 6 insertions(+), 11 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 2738d97495e6..ff1cbfc834b0 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -388,9 +388,8 @@ bool stmmac_eee_init(struct stmmac_priv *priv)
 	/* Using PCS we cannot dial with the phy registers at this stage
 	 * so we do not support extra feature like EEE.
 	 */
-	if ((priv->hw->pcs == STMMAC_PCS_RGMII) ||
-	    (priv->hw->pcs == STMMAC_PCS_TBI) ||
-	    (priv->hw->pcs == STMMAC_PCS_RTBI))
+	if (priv->hw->pcs == STMMAC_PCS_TBI ||
+	    priv->hw->pcs == STMMAC_PCS_RTBI)
 		return false;
 
 	/* Check if MAC core supports the EEE feature. */
@@ -2687,8 +2686,7 @@ static int stmmac_open(struct net_device *dev)
 	u32 chan;
 	int ret;
 
-	if (priv->hw->pcs != STMMAC_PCS_RGMII &&
-	    priv->hw->pcs != STMMAC_PCS_TBI &&
+	if (priv->hw->pcs != STMMAC_PCS_TBI &&
 	    priv->hw->pcs != STMMAC_PCS_RTBI) {
 		ret = stmmac_init_phy(dev);
 		if (ret) {
@@ -4876,8 +4874,7 @@ int stmmac_dvr_probe(struct device *device,
 
 	stmmac_check_pcs_mode(priv);
 
-	if (priv->hw->pcs != STMMAC_PCS_RGMII  &&
-	    priv->hw->pcs != STMMAC_PCS_TBI &&
+	if (priv->hw->pcs != STMMAC_PCS_TBI &&
 	    priv->hw->pcs != STMMAC_PCS_RTBI) {
 		/* MDIO bus Registration */
 		ret = stmmac_mdio_register(ndev);
@@ -4911,8 +4908,7 @@ int stmmac_dvr_probe(struct device *device,
 error_netdev_register:
 	phylink_destroy(priv->phylink);
 error_phy_setup:
-	if (priv->hw->pcs != STMMAC_PCS_RGMII &&
-	    priv->hw->pcs != STMMAC_PCS_TBI &&
+	if (priv->hw->pcs != STMMAC_PCS_TBI &&
 	    priv->hw->pcs != STMMAC_PCS_RTBI)
 		stmmac_mdio_unregister(ndev);
 error_mdio_register:
@@ -4957,8 +4953,7 @@ int stmmac_dvr_remove(struct device *dev)
 		reset_control_assert(priv->plat->stmmac_rst);
 	clk_disable_unprepare(priv->plat->pclk);
 	clk_disable_unprepare(priv->plat->stmmac_clk);
-	if (priv->hw->pcs != STMMAC_PCS_RGMII &&
-	    priv->hw->pcs != STMMAC_PCS_TBI &&
+	if (priv->hw->pcs != STMMAC_PCS_TBI &&
 	    priv->hw->pcs != STMMAC_PCS_RTBI)
 		stmmac_mdio_unregister(ndev);
 	destroy_workqueue(priv->wq);
-- 
cgit v1.2.3-59-g8ed1b


From 6cbdcf25bdec944f85bb2ef7ebe959bdb7dc59b8 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Thu, 16 Jan 2020 19:36:56 +0200
Subject: net: phylink: Allow 2.5BASE-T, 5GBASE-T and 10GBASE-T for the 10G
 link modes

For some reason, PHYLINK does not put the copper modes for 802.3bz
(NBASE-T) and 802.3an-2006 (10GBASE-T) in the PHY's supported mask, when
the PHY-MAC connection is a 10G-capable one (10GBase-KR, 10GBase-R,
USXGMII). One possible way through which the cable side can work at the
lower speed is by having the PHY emit PAUSE frames towards the MAC. So
fix that omission.

Also include the 2500Base-X fiber mode in this list while we're at it.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/phylink.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index efabbfa4a6d3..43d6985fb2f1 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -307,6 +307,10 @@ static int phylink_parse_mode(struct phylink *pl, struct fwnode_handle *fwnode)
 			phylink_set(pl->supported, 1000baseT_Half);
 			phylink_set(pl->supported, 1000baseT_Full);
 			phylink_set(pl->supported, 1000baseX_Full);
+			phylink_set(pl->supported, 2500baseT_Full);
+			phylink_set(pl->supported, 2500baseX_Full);
+			phylink_set(pl->supported, 5000baseT_Full);
+			phylink_set(pl->supported, 10000baseT_Full);
 			phylink_set(pl->supported, 10000baseKR_Full);
 			phylink_set(pl->supported, 10000baseCR_Full);
 			phylink_set(pl->supported, 10000baseSR_Full);
-- 
cgit v1.2.3-59-g8ed1b


From 3e41d04e2b7373490977d7145fb12109f23cde9c Mon Sep 17 00:00:00 2001
From: Alex Marginean <alexandru.marginean@nxp.com>
Date: Thu, 16 Jan 2020 19:46:28 +0200
Subject: net: phy: don't crash in phy_read/_write_mmd without a PHY driver

The APIs can be used by Ethernet drivers without actually loading a PHY
driver. This may become more widespread in the future with 802.3z
compatible MAC PCS devices being locally driven by the MAC driver when
configuring for a PHY mode with in-band negotiation.

Check that drv is not NULL before reading from it.

Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/phy-core.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/phy/phy-core.c b/drivers/net/phy/phy-core.c
index 769a076514b0..a4d2d59fceca 100644
--- a/drivers/net/phy/phy-core.c
+++ b/drivers/net/phy/phy-core.c
@@ -387,7 +387,7 @@ int __phy_read_mmd(struct phy_device *phydev, int devad, u32 regnum)
 	if (regnum > (u16)~0 || devad > 32)
 		return -EINVAL;
 
-	if (phydev->drv->read_mmd) {
+	if (phydev->drv && phydev->drv->read_mmd) {
 		val = phydev->drv->read_mmd(phydev, devad, regnum);
 	} else if (phydev->is_c45) {
 		u32 addr = MII_ADDR_C45 | (devad << 16) | (regnum & 0xffff);
@@ -444,7 +444,7 @@ int __phy_write_mmd(struct phy_device *phydev, int devad, u32 regnum, u16 val)
 	if (regnum > (u16)~0 || devad > 32)
 		return -EINVAL;
 
-	if (phydev->drv->write_mmd) {
+	if (phydev->drv && phydev->drv->write_mmd) {
 		ret = phydev->drv->write_mmd(phydev, devad, regnum, val);
 	} else if (phydev->is_c45) {
 		u32 addr = MII_ADDR_C45 | (devad << 16) | (regnum & 0xffff);
-- 
cgit v1.2.3-59-g8ed1b


From 04e2246329d4a72c2ec8ebd1ebc3f7250dece2da Mon Sep 17 00:00:00 2001
From: Alex Marginean <alexandru.marginean@nxp.com>
Date: Sat, 18 Jan 2020 14:19:15 +0200
Subject: net: phylink: allow in-band AN for USXGMII

USXGMII supports passing link information in-band between PHY and MAC PCS,
add it to the list of protocols that support in-band AN mode.

Being a MAC-PHY protocol that can auto-negotiate link speeds up to 10
Gbps, we populate the initial supported mask with the entire spectrum of
link modes up to 10G that PHYLINK supports, and we let the driver reduce
that mask in its .phylink_validate method.

Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/phylink.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index 43d6985fb2f1..70b9a143db84 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -298,6 +298,7 @@ static int phylink_parse_mode(struct phylink *pl, struct fwnode_handle *fwnode)
 			phylink_set(pl->supported, 2500baseX_Full);
 			break;
 
+		case PHY_INTERFACE_MODE_USXGMII:
 		case PHY_INTERFACE_MODE_10GKR:
 		case PHY_INTERFACE_MODE_10GBASER:
 			phylink_set(pl->supported, 10baseT_Half);
-- 
cgit v1.2.3-59-g8ed1b


From 4262c38dc42e092987f41cb1695240ac7dab86a9 Mon Sep 17 00:00:00 2001
From: Andrew Lunn <andrew@lunn.ch>
Date: Sat, 18 Jan 2020 19:40:56 +0100
Subject: net: dsa: mv88e6xxx: Add SERDES stats counters to all 6390 family
 members

The SERDES statistics are valid for all members of the 6390 family,
not just the 6390 itself. Add the needed callbacks to all members of
the family.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/mv88e6xxx/chip.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c
index 04ef4d00f134..8c9289549688 100644
--- a/drivers/net/dsa/mv88e6xxx/chip.c
+++ b/drivers/net/dsa/mv88e6xxx/chip.c
@@ -3845,6 +3845,9 @@ static const struct mv88e6xxx_ops mv88e6190_ops = {
 	.serdes_irq_mapping = mv88e6390_serdes_irq_mapping,
 	.serdes_irq_enable = mv88e6390_serdes_irq_enable,
 	.serdes_irq_status = mv88e6390_serdes_irq_status,
+	.serdes_get_strings = mv88e6390_serdes_get_strings,
+	.serdes_get_stats = mv88e6390_serdes_get_stats,
+	.phylink_validate = mv88e6390_phylink_validate,
 	.gpio_ops = &mv88e6352_gpio_ops,
 	.phylink_validate = mv88e6390_phylink_validate,
 };
@@ -3896,6 +3899,9 @@ static const struct mv88e6xxx_ops mv88e6190x_ops = {
 	.serdes_irq_mapping = mv88e6390_serdes_irq_mapping,
 	.serdes_irq_enable = mv88e6390_serdes_irq_enable,
 	.serdes_irq_status = mv88e6390_serdes_irq_status,
+	.serdes_get_strings = mv88e6390_serdes_get_strings,
+	.serdes_get_stats = mv88e6390_serdes_get_stats,
+	.phylink_validate = mv88e6390_phylink_validate,
 	.gpio_ops = &mv88e6352_gpio_ops,
 	.phylink_validate = mv88e6390x_phylink_validate,
 };
@@ -3946,6 +3952,9 @@ static const struct mv88e6xxx_ops mv88e6191_ops = {
 	.serdes_irq_mapping = mv88e6390_serdes_irq_mapping,
 	.serdes_irq_enable = mv88e6390_serdes_irq_enable,
 	.serdes_irq_status = mv88e6390_serdes_irq_status,
+	.serdes_get_strings = mv88e6390_serdes_get_strings,
+	.serdes_get_stats = mv88e6390_serdes_get_stats,
+	.phylink_validate = mv88e6390_phylink_validate,
 	.avb_ops = &mv88e6390_avb_ops,
 	.ptp_ops = &mv88e6352_ptp_ops,
 	.phylink_validate = mv88e6390_phylink_validate,
@@ -4092,6 +4101,9 @@ static const struct mv88e6xxx_ops mv88e6290_ops = {
 	.serdes_irq_mapping = mv88e6390_serdes_irq_mapping,
 	.serdes_irq_enable = mv88e6390_serdes_irq_enable,
 	.serdes_irq_status = mv88e6390_serdes_irq_status,
+	.serdes_get_strings = mv88e6390_serdes_get_strings,
+	.serdes_get_stats = mv88e6390_serdes_get_stats,
+	.phylink_validate = mv88e6390_phylink_validate,
 	.gpio_ops = &mv88e6352_gpio_ops,
 	.avb_ops = &mv88e6390_avb_ops,
 	.ptp_ops = &mv88e6352_ptp_ops,
@@ -4486,6 +4498,9 @@ static const struct mv88e6xxx_ops mv88e6390x_ops = {
 	.serdes_irq_mapping = mv88e6390_serdes_irq_mapping,
 	.serdes_irq_enable = mv88e6390_serdes_irq_enable,
 	.serdes_irq_status = mv88e6390_serdes_irq_status,
+	.serdes_get_sset_count = mv88e6390_serdes_get_sset_count,
+	.serdes_get_strings = mv88e6390_serdes_get_strings,
+	.serdes_get_stats = mv88e6390_serdes_get_stats,
 	.gpio_ops = &mv88e6352_gpio_ops,
 	.avb_ops = &mv88e6390_avb_ops,
 	.ptp_ops = &mv88e6352_ptp_ops,
-- 
cgit v1.2.3-59-g8ed1b


From 2ab1d925aa4c0c179dd1eb492e8c03536972707b Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Sun, 19 Jan 2020 14:31:55 +0100
Subject: net: phy: add generic ndo_do_ioctl handler phy_do_ioctl

A number of network drivers has the same glue code to use phy_mii_ioctl
as ndo_do_ioctl handler. So let's add such a generic ndo_do_ioctl
handler to phylib.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/phy.c | 15 +++++++++++++++
 include/linux/phy.h   |  1 +
 2 files changed, 16 insertions(+)

diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c
index 541ed01496bf..da05b3480abb 100644
--- a/drivers/net/phy/phy.c
+++ b/drivers/net/phy/phy.c
@@ -432,6 +432,21 @@ int phy_mii_ioctl(struct phy_device *phydev, struct ifreq *ifr, int cmd)
 }
 EXPORT_SYMBOL(phy_mii_ioctl);
 
+/**
+ * phy_do_ioctl - generic ndo_do_ioctl implementation
+ * @dev: the net_device struct
+ * @ifr: &struct ifreq for socket ioctl's
+ * @cmd: ioctl cmd to execute
+ */
+int phy_do_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
+{
+	if (!netif_running(dev) || !dev->phydev)
+		return -ENODEV;
+
+	return phy_mii_ioctl(dev->phydev, ifr, cmd);
+}
+EXPORT_SYMBOL(phy_do_ioctl);
+
 void phy_queue_state_machine(struct phy_device *phydev, unsigned long jiffies)
 {
 	mod_delayed_work(system_power_efficient_wq, &phydev->state_queue,
diff --git a/include/linux/phy.h b/include/linux/phy.h
index 99a87f02667f..be6b3a1b03da 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -1242,6 +1242,7 @@ void phy_ethtool_ksettings_get(struct phy_device *phydev,
 int phy_ethtool_ksettings_set(struct phy_device *phydev,
 			      const struct ethtool_link_ksettings *cmd);
 int phy_mii_ioctl(struct phy_device *phydev, struct ifreq *ifr, int cmd);
+int phy_do_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd);
 void phy_request_interrupt(struct phy_device *phydev);
 void phy_free_interrupt(struct phy_device *phydev);
 void phy_print_status(struct phy_device *phydev);
-- 
cgit v1.2.3-59-g8ed1b


From df007186488e33716ccf0c01f93cc5fe1ef91847 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Sun, 19 Jan 2020 14:32:49 +0100
Subject: r8169: use generic ndo_do_ioctl handler phy_do_ioctl

Replace rtl8169_ioctl with new generic function phy_do_ioctl.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 12 +-----------
 1 file changed, 1 insertion(+), 11 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 6d699df7d27f..175f951b6547 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -2298,16 +2298,6 @@ static int rtl_set_mac_address(struct net_device *dev, void *p)
 	return 0;
 }
 
-static int rtl8169_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
-{
-	struct rtl8169_private *tp = netdev_priv(dev);
-
-	if (!netif_running(dev))
-		return -ENODEV;
-
-	return phy_mii_ioctl(tp->phydev, ifr, cmd);
-}
-
 static void rtl_wol_suspend_quirk(struct rtl8169_private *tp)
 {
 	switch (tp->mac_version) {
@@ -5168,7 +5158,7 @@ static const struct net_device_ops rtl_netdev_ops = {
 	.ndo_fix_features	= rtl8169_fix_features,
 	.ndo_set_features	= rtl8169_set_features,
 	.ndo_set_mac_address	= rtl_set_mac_address,
-	.ndo_do_ioctl		= rtl8169_ioctl,
+	.ndo_do_ioctl		= phy_do_ioctl,
 	.ndo_set_rx_mode	= rtl_set_rx_mode,
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller	= rtl8169_netpoll,
-- 
cgit v1.2.3-59-g8ed1b


From ac9cc4e25044b699eb8f0f4822ad58a2f5130abb Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@mellanox.com>
Date: Mon, 20 Jan 2020 09:52:50 +0200
Subject: mlxsw: spectrum: Push code getting port speed into a helper

Currently PTP code queries directly PTYS register for port speed from
work scheduled upon PUDE event. Since the speed needs to be used for
SPAN buffer size computation as well, push the code into a separate
helper.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum.c     | 21 +++++++++++++++++++++
 drivers/net/ethernet/mellanox/mlxsw/spectrum.h     |  1 +
 drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c | 14 ++------------
 3 files changed, 24 insertions(+), 12 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index 8639f32ec4d5..1bedf9bc0a57 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -3535,6 +3535,27 @@ mlxsw_sp_port_speed_by_width_set(struct mlxsw_sp_port *mlxsw_sp_port)
 	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ptys), ptys_pl);
 }
 
+int mlxsw_sp_port_speed_get(struct mlxsw_sp_port *mlxsw_sp_port, u32 *speed)
+{
+	const struct mlxsw_sp_port_type_speed_ops *port_type_speed_ops;
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char ptys_pl[MLXSW_REG_PTYS_LEN];
+	u32 eth_proto_oper;
+	int err;
+
+	port_type_speed_ops = mlxsw_sp->port_type_speed_ops;
+	port_type_speed_ops->reg_ptys_eth_pack(mlxsw_sp, ptys_pl,
+					       mlxsw_sp_port->local_port, 0,
+					       false);
+	err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(ptys), ptys_pl);
+	if (err)
+		return err;
+	port_type_speed_ops->reg_ptys_eth_unpack(mlxsw_sp, ptys_pl, NULL, NULL,
+						 &eth_proto_oper);
+	*speed = port_type_speed_ops->from_ptys_speed(mlxsw_sp, eth_proto_oper);
+	return 0;
+}
+
 int mlxsw_sp_port_ets_set(struct mlxsw_sp_port *mlxsw_sp_port,
 			  enum mlxsw_reg_qeec_hr hr, u8 index, u8 next_index,
 			  bool dwrr, u8 dwrr_weight)
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
index 948ef4720d40..6477b473cad7 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
@@ -471,6 +471,7 @@ extern struct notifier_block mlxsw_sp_switchdev_notifier;
 /* spectrum.c */
 void mlxsw_sp_rx_listener_no_mark_func(struct sk_buff *skb,
 				       u8 local_port, void *priv);
+int mlxsw_sp_port_speed_get(struct mlxsw_sp_port *mlxsw_sp_port, u32 *speed);
 int mlxsw_sp_port_ets_set(struct mlxsw_sp_port *mlxsw_sp_port,
 			  enum mlxsw_reg_qeec_hr hr, u8 index, u8 next_index,
 			  bool dwrr, u8 dwrr_weight);
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c
index 4aaaa4937b1a..34f7c3501b08 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c
@@ -1016,27 +1016,17 @@ mlxsw_sp1_ptp_port_shaper_set(struct mlxsw_sp_port *mlxsw_sp_port, bool enable)
 
 static int mlxsw_sp1_ptp_port_shaper_check(struct mlxsw_sp_port *mlxsw_sp_port)
 {
-	const struct mlxsw_sp_port_type_speed_ops *port_type_speed_ops;
-	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
-	char ptys_pl[MLXSW_REG_PTYS_LEN];
-	u32 eth_proto_oper, speed;
 	bool ptps = false;
 	int err, i;
+	u32 speed;
 
 	if (!mlxsw_sp1_ptp_hwtstamp_enabled(mlxsw_sp_port))
 		return mlxsw_sp1_ptp_port_shaper_set(mlxsw_sp_port, false);
 
-	port_type_speed_ops = mlxsw_sp->port_type_speed_ops;
-	port_type_speed_ops->reg_ptys_eth_pack(mlxsw_sp, ptys_pl,
-					       mlxsw_sp_port->local_port, 0,
-					       false);
-	err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(ptys), ptys_pl);
+	err = mlxsw_sp_port_speed_get(mlxsw_sp_port, &speed);
 	if (err)
 		return err;
-	port_type_speed_ops->reg_ptys_eth_unpack(mlxsw_sp, ptys_pl, NULL, NULL,
-						 &eth_proto_oper);
 
-	speed = port_type_speed_ops->from_ptys_speed(mlxsw_sp, eth_proto_oper);
 	for (i = 0; i < MLXSW_SP1_PTP_SHAPER_PARAMS_LEN; i++) {
 		if (mlxsw_sp1_ptp_shaper_params[i].ethtool_speed == speed) {
 			ptps = true;
-- 
cgit v1.2.3-59-g8ed1b


From 31c25b949813292e2f85b0e8708b767e55aba093 Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@mellanox.com>
Date: Mon, 20 Jan 2020 09:52:51 +0200
Subject: mlxsw: spectrum_span: Put buffsize update code into helper function

Avoid duplication of code that is doing buffsize update and put it into
a separate helper function.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../net/ethernet/mellanox/mlxsw/spectrum_span.c    | 37 +++++++++-------------
 1 file changed, 15 insertions(+), 22 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c
index 200d324e6d99..7917c6ea262e 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c
@@ -754,26 +754,25 @@ static int mlxsw_sp_span_mtu_to_buffsize(const struct mlxsw_sp *mlxsw_sp,
 	return mlxsw_sp_bytes_cells(mlxsw_sp, mtu * 5 / 2) + 1;
 }
 
-int mlxsw_sp_span_port_mtu_update(struct mlxsw_sp_port *port, u16 mtu)
+static int
+mlxsw_sp_span_port_buffsize_update(struct mlxsw_sp_port *mlxsw_sp_port, u16 mtu)
 {
-	struct mlxsw_sp *mlxsw_sp = port->mlxsw_sp;
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
 	char sbib_pl[MLXSW_REG_SBIB_LEN];
-	int err;
+	u32 buffsize;
 
+	buffsize = mlxsw_sp_span_mtu_to_buffsize(mlxsw_sp, mtu);
+	mlxsw_reg_sbib_pack(sbib_pl, mlxsw_sp_port->local_port, buffsize);
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sbib), sbib_pl);
+}
+
+int mlxsw_sp_span_port_mtu_update(struct mlxsw_sp_port *port, u16 mtu)
+{
 	/* If port is egress mirrored, the shared buffer size should be
 	 * updated according to the mtu value
 	 */
-	if (mlxsw_sp_span_is_egress_mirror(port)) {
-		u32 buffsize = mlxsw_sp_span_mtu_to_buffsize(mlxsw_sp, mtu);
-
-		mlxsw_reg_sbib_pack(sbib_pl, port->local_port, buffsize);
-		err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sbib), sbib_pl);
-		if (err) {
-			netdev_err(port->dev, "Could not update shared buffer for mirroring\n");
-			return err;
-		}
-	}
-
+	if (mlxsw_sp_span_is_egress_mirror(port))
+		return mlxsw_sp_span_port_buffsize_update(port, mtu);
 	return 0;
 }
 
@@ -836,15 +835,9 @@ mlxsw_sp_span_inspected_port_add(struct mlxsw_sp_port *port,
 
 	/* if it is an egress SPAN, bind a shared buffer to it */
 	if (type == MLXSW_SP_SPAN_EGRESS) {
-		u32 buffsize = mlxsw_sp_span_mtu_to_buffsize(mlxsw_sp,
-							     port->dev->mtu);
-
-		mlxsw_reg_sbib_pack(sbib_pl, port->local_port, buffsize);
-		err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sbib), sbib_pl);
-		if (err) {
-			netdev_err(port->dev, "Could not create shared buffer for mirroring\n");
+		err = mlxsw_sp_span_port_buffsize_update(port, port->dev->mtu);
+		if (err)
 			return err;
-		}
 	}
 
 	if (bind) {
-- 
cgit v1.2.3-59-g8ed1b


From ff9fdfec5f96e9e70469775da498226a1aaa8a68 Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@mellanox.com>
Date: Mon, 20 Jan 2020 09:52:52 +0200
Subject: mlxsw: spectrum: Fix SPAN egress mirroring buffer size for Spectrum-2

For SPAN egress mirroring buffer size, it is needed to use a different
formula for Spectrum and Spectrum-2. Move the buffer size computation to
ops and implement new formula for Spectrum-2.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum.c     | 34 ++++++++++++++++++++++
 drivers/net/ethernet/mellanox/mlxsw/spectrum.h     |  3 ++
 .../net/ethernet/mellanox/mlxsw/spectrum_span.c    | 16 +++++-----
 3 files changed, 46 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index 1bedf9bc0a57..f6d2c6f6889f 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -195,6 +195,10 @@ struct mlxsw_sp_ptp_ops {
 			  u64 *data, int data_index);
 };
 
+struct mlxsw_sp_span_ops {
+	u32 (*buffsize_get)(int mtu, u32 speed);
+};
+
 static int mlxsw_sp_component_query(struct mlxfw_dev *mlxfw_dev,
 				    u16 component_index, u32 *p_max_size,
 				    u8 *p_align_bits, u16 *p_max_write_size)
@@ -4914,6 +4918,33 @@ static const struct mlxsw_sp_ptp_ops mlxsw_sp2_ptp_ops = {
 	.get_stats	= mlxsw_sp2_get_stats,
 };
 
+static u32 mlxsw_sp1_span_buffsize_get(int mtu, u32 speed)
+{
+	return mtu * 5 / 2;
+}
+
+static const struct mlxsw_sp_span_ops mlxsw_sp1_span_ops = {
+	.buffsize_get = mlxsw_sp1_span_buffsize_get,
+};
+
+#define MLXSW_SP2_SPAN_EG_MIRROR_BUFFER_FACTOR 38
+
+static u32 mlxsw_sp2_span_buffsize_get(int mtu, u32 speed)
+{
+	return 3 * mtu + MLXSW_SP2_SPAN_EG_MIRROR_BUFFER_FACTOR * speed / 1000;
+}
+
+static const struct mlxsw_sp_span_ops mlxsw_sp2_span_ops = {
+	.buffsize_get = mlxsw_sp2_span_buffsize_get,
+};
+
+u32 mlxsw_sp_span_buffsize_get(struct mlxsw_sp *mlxsw_sp, int mtu, u32 speed)
+{
+	u32 buffsize = mlxsw_sp->span_ops->buffsize_get(speed, mtu);
+
+	return mlxsw_sp_bytes_cells(mlxsw_sp, buffsize) + 1;
+}
+
 static int mlxsw_sp_netdevice_event(struct notifier_block *unused,
 				    unsigned long event, void *ptr);
 
@@ -5135,6 +5166,7 @@ static int mlxsw_sp1_init(struct mlxsw_core *mlxsw_core,
 	mlxsw_sp->sb_vals = &mlxsw_sp1_sb_vals;
 	mlxsw_sp->port_type_speed_ops = &mlxsw_sp1_port_type_speed_ops;
 	mlxsw_sp->ptp_ops = &mlxsw_sp1_ptp_ops;
+	mlxsw_sp->span_ops = &mlxsw_sp1_span_ops;
 	mlxsw_sp->listeners = mlxsw_sp1_listener;
 	mlxsw_sp->listeners_count = ARRAY_SIZE(mlxsw_sp1_listener);
 
@@ -5160,6 +5192,7 @@ static int mlxsw_sp2_init(struct mlxsw_core *mlxsw_core,
 	mlxsw_sp->sb_vals = &mlxsw_sp2_sb_vals;
 	mlxsw_sp->port_type_speed_ops = &mlxsw_sp2_port_type_speed_ops;
 	mlxsw_sp->ptp_ops = &mlxsw_sp2_ptp_ops;
+	mlxsw_sp->span_ops = &mlxsw_sp2_span_ops;
 
 	return mlxsw_sp_init(mlxsw_core, mlxsw_bus_info, extack);
 }
@@ -5181,6 +5214,7 @@ static int mlxsw_sp3_init(struct mlxsw_core *mlxsw_core,
 	mlxsw_sp->sb_vals = &mlxsw_sp2_sb_vals;
 	mlxsw_sp->port_type_speed_ops = &mlxsw_sp2_port_type_speed_ops;
 	mlxsw_sp->ptp_ops = &mlxsw_sp2_ptp_ops;
+	mlxsw_sp->span_ops = &mlxsw_sp2_span_ops;
 
 	return mlxsw_sp_init(mlxsw_core, mlxsw_bus_info, extack);
 }
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
index 6477b473cad7..39c3d3ee79a6 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
@@ -140,6 +140,7 @@ struct mlxsw_sp_sb_vals;
 struct mlxsw_sp_port_type_speed_ops;
 struct mlxsw_sp_ptp_state;
 struct mlxsw_sp_ptp_ops;
+struct mlxsw_sp_span_ops;
 
 struct mlxsw_sp_port_mapping {
 	u8 module;
@@ -185,6 +186,7 @@ struct mlxsw_sp {
 	const struct mlxsw_sp_sb_vals *sb_vals;
 	const struct mlxsw_sp_port_type_speed_ops *port_type_speed_ops;
 	const struct mlxsw_sp_ptp_ops *ptp_ops;
+	const struct mlxsw_sp_span_ops *span_ops;
 	const struct mlxsw_listener *listeners;
 	size_t listeners_count;
 };
@@ -502,6 +504,7 @@ int mlxsw_sp_flow_counter_alloc(struct mlxsw_sp *mlxsw_sp,
 				unsigned int *p_counter_index);
 void mlxsw_sp_flow_counter_free(struct mlxsw_sp *mlxsw_sp,
 				unsigned int counter_index);
+u32 mlxsw_sp_span_buffsize_get(struct mlxsw_sp *mlxsw_sp, int mtu, u32 speed);
 bool mlxsw_sp_port_dev_check(const struct net_device *dev);
 struct mlxsw_sp *mlxsw_sp_lower_get(struct net_device *dev);
 struct mlxsw_sp_port *mlxsw_sp_port_dev_lower_find(struct net_device *dev);
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c
index 7917c6ea262e..88b8edf14387 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c
@@ -748,20 +748,22 @@ static bool mlxsw_sp_span_is_egress_mirror(struct mlxsw_sp_port *port)
 	return false;
 }
 
-static int mlxsw_sp_span_mtu_to_buffsize(const struct mlxsw_sp *mlxsw_sp,
-					 int mtu)
-{
-	return mlxsw_sp_bytes_cells(mlxsw_sp, mtu * 5 / 2) + 1;
-}
-
 static int
 mlxsw_sp_span_port_buffsize_update(struct mlxsw_sp_port *mlxsw_sp_port, u16 mtu)
 {
 	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
 	char sbib_pl[MLXSW_REG_SBIB_LEN];
 	u32 buffsize;
+	u32 speed;
+	int err;
+
+	err = mlxsw_sp_port_speed_get(mlxsw_sp_port, &speed);
+	if (err)
+		return err;
+	if (speed == SPEED_UNKNOWN)
+		speed = 0;
 
-	buffsize = mlxsw_sp_span_mtu_to_buffsize(mlxsw_sp, mtu);
+	buffsize = mlxsw_sp_span_buffsize_get(mlxsw_sp, speed, mtu);
 	mlxsw_reg_sbib_pack(sbib_pl, mlxsw_sp_port->local_port, buffsize);
 	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sbib), sbib_pl);
 }
-- 
cgit v1.2.3-59-g8ed1b


From 3a3e627ce0ac0fe15e0a011540a1f7c96bf8a07b Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@mellanox.com>
Date: Mon, 20 Jan 2020 09:52:53 +0200
Subject: spectrum: Add a delayed work to update SPAN buffsize according to
 speed

When PUDE event is handled and the link is up, update the port SPAN
buffer size according to the current speed.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum.c      |  4 ++++
 drivers/net/ethernet/mellanox/mlxsw/spectrum.h      |  3 +++
 drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c | 16 ++++++++++++++++
 drivers/net/ethernet/mellanox/mlxsw/spectrum_span.h |  1 +
 4 files changed, 24 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index f6d2c6f6889f..9eb3ac7669f7 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -3908,6 +3908,8 @@ static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port,
 
 	INIT_DELAYED_WORK(&mlxsw_sp_port->ptp.shaper_dw,
 			  mlxsw_sp->ptp_ops->shaper_work);
+	INIT_DELAYED_WORK(&mlxsw_sp_port->span.speed_update_dw,
+			  mlxsw_sp_span_speed_update_work);
 
 	mlxsw_sp->ports[local_port] = mlxsw_sp_port;
 	err = register_netdev(dev);
@@ -3964,6 +3966,7 @@ static void mlxsw_sp_port_remove(struct mlxsw_sp *mlxsw_sp, u8 local_port)
 	struct mlxsw_sp_port *mlxsw_sp_port = mlxsw_sp->ports[local_port];
 
 	cancel_delayed_work_sync(&mlxsw_sp_port->periodic_hw_stats.update_dw);
+	cancel_delayed_work_sync(&mlxsw_sp_port->span.speed_update_dw);
 	cancel_delayed_work_sync(&mlxsw_sp_port->ptp.shaper_dw);
 	mlxsw_sp_port_ptp_clear(mlxsw_sp_port);
 	mlxsw_core_port_clear(mlxsw_sp->core, local_port, mlxsw_sp);
@@ -4371,6 +4374,7 @@ static void mlxsw_sp_pude_event_func(const struct mlxsw_reg_info *reg,
 		netdev_info(mlxsw_sp_port->dev, "link up\n");
 		netif_carrier_on(mlxsw_sp_port->dev);
 		mlxsw_core_schedule_dw(&mlxsw_sp_port->ptp.shaper_dw, 0);
+		mlxsw_core_schedule_dw(&mlxsw_sp_port->span.speed_update_dw, 0);
 	} else {
 		netdev_info(mlxsw_sp_port->dev, "link down\n");
 		netif_carrier_off(mlxsw_sp_port->dev);
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
index 39c3d3ee79a6..5f3b74360dc8 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
@@ -294,6 +294,9 @@ struct mlxsw_sp_port {
 		struct mlxsw_sp_ptp_port_stats stats;
 	} ptp;
 	u8 split_base_local_port;
+	struct {
+		struct delayed_work speed_update_dw;
+	} span;
 };
 
 struct mlxsw_sp_port_type_speed_ops {
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c
index 88b8edf14387..0cdd7954a085 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c
@@ -778,6 +778,22 @@ int mlxsw_sp_span_port_mtu_update(struct mlxsw_sp_port *port, u16 mtu)
 	return 0;
 }
 
+void mlxsw_sp_span_speed_update_work(struct work_struct *work)
+{
+	struct delayed_work *dwork = to_delayed_work(work);
+	struct mlxsw_sp_port *mlxsw_sp_port;
+
+	mlxsw_sp_port = container_of(dwork, struct mlxsw_sp_port,
+				     span.speed_update_dw);
+
+	/* If port is egress mirrored, the shared buffer size should be
+	 * updated according to the speed value.
+	 */
+	if (mlxsw_sp_span_is_egress_mirror(mlxsw_sp_port))
+		mlxsw_sp_span_port_buffsize_update(mlxsw_sp_port,
+						   mlxsw_sp_port->dev->mtu);
+}
+
 static struct mlxsw_sp_span_inspected_port *
 mlxsw_sp_span_entry_bound_port_find(struct mlxsw_sp_span_entry *span_entry,
 				    enum mlxsw_sp_span_type type,
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.h
index 5e04252f2a11..59724335525f 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.h
@@ -74,5 +74,6 @@ void mlxsw_sp_span_entry_invalidate(struct mlxsw_sp *mlxsw_sp,
 				    struct mlxsw_sp_span_entry *span_entry);
 
 int mlxsw_sp_span_port_mtu_update(struct mlxsw_sp_port *port, u16 mtu);
+void mlxsw_sp_span_speed_update_work(struct work_struct *work);
 
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From 2e3a94aa2bfc6de95a0700f0a868c6f5db3a9592 Mon Sep 17 00:00:00 2001
From: Brian Vazquez <brianvv@google.com>
Date: Sun, 19 Jan 2020 11:40:40 -0800
Subject: bpf: Fix memory leaks in generic update/delete batch ops

Generic update/delete batch ops functions were using __bpf_copy_key
without properly freeing the memory. Handle the memory allocation and
copy_from_user separately.

Fixes: aa2e93b8e58e ("bpf: Add generic support for update and delete batch ops")
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Brian Vazquez <brianvv@google.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20200119194040.128369-1-brianvv@google.com
---
 kernel/bpf/syscall.c | 30 +++++++++++++++++++-----------
 1 file changed, 19 insertions(+), 11 deletions(-)

diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index c26a71460f02..9a840c57f6df 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -1239,12 +1239,15 @@ int generic_map_delete_batch(struct bpf_map *map,
 	if (!max_count)
 		return 0;
 
+	key = kmalloc(map->key_size, GFP_USER | __GFP_NOWARN);
+	if (!key)
+		return -ENOMEM;
+
 	for (cp = 0; cp < max_count; cp++) {
-		key = __bpf_copy_key(keys + cp * map->key_size, map->key_size);
-		if (IS_ERR(key)) {
-			err = PTR_ERR(key);
+		err = -EFAULT;
+		if (copy_from_user(key, keys + cp * map->key_size,
+				   map->key_size))
 			break;
-		}
 
 		if (bpf_map_is_dev_bound(map)) {
 			err = bpf_map_offload_delete_elem(map, key);
@@ -1264,6 +1267,8 @@ int generic_map_delete_batch(struct bpf_map *map,
 	}
 	if (copy_to_user(&uattr->batch.count, &cp, sizeof(cp)))
 		err = -EFAULT;
+
+	kfree(key);
 	return err;
 }
 
@@ -1294,18 +1299,21 @@ int generic_map_update_batch(struct bpf_map *map,
 	if (!max_count)
 		return 0;
 
+	key = kmalloc(map->key_size, GFP_USER | __GFP_NOWARN);
+	if (!key)
+		return -ENOMEM;
+
 	value = kmalloc(value_size, GFP_USER | __GFP_NOWARN);
-	if (!value)
+	if (!value) {
+		kfree(key);
 		return -ENOMEM;
+	}
 
 	for (cp = 0; cp < max_count; cp++) {
-		key = __bpf_copy_key(keys + cp * map->key_size, map->key_size);
-		if (IS_ERR(key)) {
-			err = PTR_ERR(key);
-			break;
-		}
 		err = -EFAULT;
-		if (copy_from_user(value, values + cp * value_size, value_size))
+		if (copy_from_user(key, keys + cp * map->key_size,
+		    map->key_size) ||
+		    copy_from_user(value, values + cp * value_size, value_size))
 			break;
 
 		err = bpf_map_update_value(map, f, key, value,
-- 
cgit v1.2.3-59-g8ed1b


From 396b83fa75a3deb9aa68296c090f4eb540917d9d Mon Sep 17 00:00:00 2001
From: Stanislav Fomichev <sdf@google.com>
Date: Fri, 17 Jan 2020 17:05:46 -0800
Subject: selftests/bpf: Don't check for btf fd in test_btf

After commit 0d13bfce023a ("libbpf: Don't require root for
bpf_object__open()") we no longer load BTF during bpf_object__open(),
so let's remove the expectation from test_btf that the fd is not -1.
The test currently fails.

Before:
BTF libbpf test[1] (test_btf_haskv.o): do_test_file:4152:FAIL bpf_object__btf_fd: -1
BTF libbpf test[2] (test_btf_newkv.o): do_test_file:4152:FAIL bpf_object__btf_fd: -1
BTF libbpf test[3] (test_btf_nokv.o): do_test_file:4152:FAIL bpf_object__btf_fd: -1

After:
BTF libbpf test[1] (test_btf_haskv.o): OK
BTF libbpf test[2] (test_btf_newkv.o): OK
BTF libbpf test[3] (test_btf_nokv.o): OK

Fixes: 0d13bfce023a ("libbpf: Don't require root for bpf_object__open()")
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20200118010546.74279-1-sdf@google.com
---
 tools/testing/selftests/bpf/test_btf.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/tools/testing/selftests/bpf/test_btf.c b/tools/testing/selftests/bpf/test_btf.c
index 3d617e806054..93040ca83e60 100644
--- a/tools/testing/selftests/bpf/test_btf.c
+++ b/tools/testing/selftests/bpf/test_btf.c
@@ -4148,10 +4148,6 @@ static int do_test_file(unsigned int test_num)
 	if (CHECK(IS_ERR(obj), "obj: %ld", PTR_ERR(obj)))
 		return PTR_ERR(obj);
 
-	err = bpf_object__btf_fd(obj);
-	if (CHECK(err == -1, "bpf_object__btf_fd: -1"))
-		goto done;
-
 	prog = bpf_program__next(NULL, obj);
 	if (CHECK(!prog, "Cannot find bpf_prog")) {
 		err = -1;
-- 
cgit v1.2.3-59-g8ed1b


From f1c3656c6d9c147d07d16614455aceb34932bdeb Mon Sep 17 00:00:00 2001
From: Hangbin Liu <liuhangbin@gmail.com>
Date: Fri, 17 Jan 2020 18:06:56 +0800
Subject: selftests/bpf: Skip perf hw events test if the setup disabled it

The same with commit 4e59afbbed96 ("selftests/bpf: skip nmi test when perf
hw events are disabled"), it would make more sense to skip the
test_stacktrace_build_id_nmi test if the setup (e.g. virtual machines) has
disabled hardware perf events.

Fixes: 13790d1cc72c ("bpf: add selftest for stackmap with build_id in NMI context")
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20200117100656.10359-1-liuhangbin@gmail.com
---
 tools/testing/selftests/bpf/prog_tests/stacktrace_build_id_nmi.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id_nmi.c b/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id_nmi.c
index 8974450a4bdb..f002e3090d92 100644
--- a/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id_nmi.c
+++ b/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id_nmi.c
@@ -49,8 +49,12 @@ retry:
 	pmu_fd = syscall(__NR_perf_event_open, &attr, -1 /* pid */,
 			 0 /* cpu 0 */, -1 /* group id */,
 			 0 /* flags */);
-	if (CHECK(pmu_fd < 0, "perf_event_open",
-		  "err %d errno %d. Does the test host support PERF_COUNT_HW_CPU_CYCLES?\n",
+	if (pmu_fd < 0 && errno == ENOENT) {
+		printf("%s:SKIP:no PERF_COUNT_HW_CPU_CYCLES\n", __func__);
+		test__skip();
+		goto cleanup;
+	}
+	if (CHECK(pmu_fd < 0, "perf_event_open", "err %d errno %d\n",
 		  pmu_fd, errno))
 		goto cleanup;
 
-- 
cgit v1.2.3-59-g8ed1b


From b2e5e93ae8af6a34bca536cdc4b453ab1e707b8b Mon Sep 17 00:00:00 2001
From: Toke Høiland-Jørgensen <toke@redhat.com>
Date: Mon, 20 Jan 2020 14:06:41 +0100
Subject: samples/bpf: Don't try to remove user's homedir on clean
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The 'clean' rule in the samples/bpf Makefile tries to remove backup
files (ending in ~). However, if no such files exist, it will instead try
to remove the user's home directory. While the attempt is mostly harmless,
it does lead to a somewhat scary warning like this:

rm: cannot remove '~': Is a directory

Fix this by using find instead of shell expansion to locate any actual
backup files that need to be removed.

Fixes: b62a796c109c ("samples/bpf: allow make to be run from samples/bpf/ directory")
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Link: https://lore.kernel.org/bpf/157952560126.1683545.7273054725976032511.stgit@toke.dk
---
 samples/bpf/Makefile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile
index 5b89c0370f33..f86d713a17a5 100644
--- a/samples/bpf/Makefile
+++ b/samples/bpf/Makefile
@@ -254,7 +254,7 @@ all:
 
 clean:
 	$(MAKE) -C ../../ M=$(CURDIR) clean
-	@rm -f *~
+	@find $(CURDIR) -type f -name '*~' -delete
 
 $(LIBBPF): FORCE
 # Fix up variables inherited from Kbuild that tools/ build system won't like
-- 
cgit v1.2.3-59-g8ed1b


From a835d38d41ca4f8eab5f97512446212f42f094a1 Mon Sep 17 00:00:00 2001
From: Toke Høiland-Jørgensen <toke@redhat.com>
Date: Mon, 20 Jan 2020 14:06:42 +0100
Subject: tools/bpf/runqslower: Fix override option for VMLINUX_BTF
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The runqslower tool refuses to build without a file to read vmlinux BTF
from. The build fails with an error message to override the location by
setting the VMLINUX_BTF variable if autodetection fails. However, the
Makefile doesn't actually work with that override - the error message is
still emitted.

Fix this by including the value of VMLINUX_BTF in the expansion, and only
emitting the error message if the *result* is empty. Also permit running
'make clean' even though no VMLINUX_BTF is set.

Fixes: 9c01546d26d2 ("tools/bpf: Add runqslower tool to tools/bpf")
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/157952560237.1683545.17771785178857224877.stgit@toke.dk
---
 tools/bpf/runqslower/Makefile | 19 ++++++++++---------
 1 file changed, 10 insertions(+), 9 deletions(-)

diff --git a/tools/bpf/runqslower/Makefile b/tools/bpf/runqslower/Makefile
index cff2fbcd29a8..3242ab874ac0 100644
--- a/tools/bpf/runqslower/Makefile
+++ b/tools/bpf/runqslower/Makefile
@@ -10,13 +10,9 @@ CFLAGS := -g -Wall
 
 # Try to detect best kernel BTF source
 KERNEL_REL := $(shell uname -r)
-ifneq ("$(wildcard /sys/kernel/btf/vmlinux)","")
-VMLINUX_BTF := /sys/kernel/btf/vmlinux
-else ifneq ("$(wildcard /boot/vmlinux-$(KERNEL_REL))","")
-VMLINUX_BTF := /boot/vmlinux-$(KERNEL_REL)
-else
-$(error "Can't detect kernel BTF, use VMLINUX_BTF to specify it explicitly")
-endif
+VMLINUX_BTF_PATHS := /sys/kernel/btf/vmlinux /boot/vmlinux-$(KERNEL_REL)
+VMLINUX_BTF_PATH := $(or $(VMLINUX_BTF),$(firstword			       \
+					  $(wildcard $(VMLINUX_BTF_PATHS))))
 
 abs_out := $(abspath $(OUTPUT))
 ifeq ($(V),1)
@@ -67,9 +63,14 @@ $(OUTPUT):
 	$(call msg,MKDIR,$@)
 	$(Q)mkdir -p $(OUTPUT)
 
-$(OUTPUT)/vmlinux.h: $(VMLINUX_BTF) | $(OUTPUT) $(BPFTOOL)
+$(OUTPUT)/vmlinux.h: $(VMLINUX_BTF_PATH) | $(OUTPUT) $(BPFTOOL)
 	$(call msg,GEN,$@)
-	$(Q)$(BPFTOOL) btf dump file $(VMLINUX_BTF) format c > $@
+	$(Q)if [ ! -e "$(VMLINUX_BTF_PATH)" ] ; then \
+		echo "Couldn't find kernel BTF; set VMLINUX_BTF to"	       \
+			"specify its location." >&2;			       \
+		exit 1;\
+	fi
+	$(Q)$(BPFTOOL) btf dump file $(VMLINUX_BTF_PATH) format c > $@
 
 $(OUTPUT)/libbpf.a: | $(OUTPUT)
 	$(Q)$(MAKE) $(submake_extras) -C $(LIBBPF_SRC)			       \
-- 
cgit v1.2.3-59-g8ed1b


From bfe42797bc7e962f01028436e106769ca9338b65 Mon Sep 17 00:00:00 2001
From: Toke Høiland-Jørgensen <toke@redhat.com>
Date: Mon, 20 Jan 2020 14:06:43 +0100
Subject: selftests: Pass VMLINUX_BTF to runqslower Makefile
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add a VMLINUX_BTF variable with the locally-built path when calling the
runqslower Makefile from selftests. This makes sure a simple 'make'
invocation in the selftests dir works even when there is no BTF information
for the running kernel. Do a wildcard expansion and include the same paths
for BTF for the running kernel as in the runqslower Makefile, to make it
possible to build selftests without having a vmlinux in the local tree.

Also fix the make invocation to use $(OUTPUT)/tools as the destination
directory instead of $(CURDIR)/tools.

Fixes: 3a0d3092a4ed ("selftests/bpf: Build runqslower from selftests")
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/157952560344.1683545.2723631988771664417.stgit@toke.dk
---
 tools/testing/selftests/bpf/Makefile | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index 246d09ffb296..8240282aef7f 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -124,10 +124,14 @@ $(OUTPUT)/test_stub.o: test_stub.c
 	$(call msg,CC,,$@)
 	$(CC) -c $(CFLAGS) -o $@ $<
 
+VMLINUX_BTF_PATHS := $(abspath ../../../../vmlinux)			\
+			       /sys/kernel/btf/vmlinux			\
+			       /boot/vmlinux-$(shell uname -r)
+VMLINUX_BTF:= $(firstword $(wildcard $(VMLINUX_BTF_PATHS)))
 .PHONY: $(OUTPUT)/runqslower
 $(OUTPUT)/runqslower: force
-	$(Q)$(MAKE) $(submake_extras) -C $(TOOLSDIR)/bpf/runqslower	      \
-		    OUTPUT=$(CURDIR)/tools/
+	$(Q)$(MAKE) $(submake_extras) -C $(TOOLSDIR)/bpf/runqslower	\
+		    OUTPUT=$(OUTPUT)/tools/ VMLINUX_BTF=$(VMLINUX_BTF)
 
 BPFOBJ := $(OUTPUT)/libbpf.a
 
-- 
cgit v1.2.3-59-g8ed1b


From 5b554ce518afee47d4ba4330ad8fbfd6e29e1c13 Mon Sep 17 00:00:00 2001
From: Toke Høiland-Jørgensen <toke@redhat.com>
Date: Mon, 20 Jan 2020 14:06:44 +0100
Subject: tools/runqslower: Use consistent include paths for libbpf
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix the runqslower tool to include libbpf header files with the bpf/
prefix, to be consistent with external users of the library. Also ensure
that all includes of exported libbpf header files (those that are exported
on 'make install' of the library) use bracketed includes instead of quoted.

To not break the build, keep the old include path until everything has been
changed to the new one; a subsequent patch will remove that.

Fixes: 6910d7d3867a ("selftests/bpf: Ensure bpf_helper_defs.h are taken from selftests dir")
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/157952560457.1683545.9913736511685743625.stgit@toke.dk
---
 tools/bpf/runqslower/Makefile         | 5 +++--
 tools/bpf/runqslower/runqslower.bpf.c | 2 +-
 tools/bpf/runqslower/runqslower.c     | 4 ++--
 3 files changed, 6 insertions(+), 5 deletions(-)

diff --git a/tools/bpf/runqslower/Makefile b/tools/bpf/runqslower/Makefile
index 3242ab874ac0..b7b2230f807b 100644
--- a/tools/bpf/runqslower/Makefile
+++ b/tools/bpf/runqslower/Makefile
@@ -6,6 +6,7 @@ LLVM_STRIP := llvm-strip
 DEFAULT_BPFTOOL := $(OUTPUT)/sbin/bpftool
 BPFTOOL ?= $(DEFAULT_BPFTOOL)
 LIBBPF_SRC := $(abspath ../../lib/bpf)
+INCLUDES := -I$(OUTPUT) -I$(abspath ../../lib) -I$(abspath ../../lib/bpf)
 CFLAGS := -g -Wall
 
 # Try to detect best kernel BTF source
@@ -51,13 +52,13 @@ $(OUTPUT)/%.skel.h: $(OUTPUT)/%.bpf.o | $(BPFTOOL)
 
 $(OUTPUT)/%.bpf.o: %.bpf.c $(OUTPUT)/libbpf.a | $(OUTPUT)
 	$(call msg,BPF,$@)
-	$(Q)$(CLANG) -g -O2 -target bpf -I$(OUTPUT) -I$(LIBBPF_SRC)	      \
+	$(Q)$(CLANG) -g -O2 -target bpf $(INCLUDES)			      \
 		 -c $(filter %.c,$^) -o $@ &&				      \
 	$(LLVM_STRIP) -g $@
 
 $(OUTPUT)/%.o: %.c | $(OUTPUT)
 	$(call msg,CC,$@)
-	$(Q)$(CC) $(CFLAGS) -I$(LIBBPF_SRC) -I$(OUTPUT) -c $(filter %.c,$^) -o $@
+	$(Q)$(CC) $(CFLAGS) $(INCLUDES) -c $(filter %.c,$^) -o $@
 
 $(OUTPUT):
 	$(call msg,MKDIR,$@)
diff --git a/tools/bpf/runqslower/runqslower.bpf.c b/tools/bpf/runqslower/runqslower.bpf.c
index 623cce4d37f5..48a39f72fadf 100644
--- a/tools/bpf/runqslower/runqslower.bpf.c
+++ b/tools/bpf/runqslower/runqslower.bpf.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 // Copyright (c) 2019 Facebook
 #include "vmlinux.h"
-#include <bpf_helpers.h>
+#include <bpf/bpf_helpers.h>
 #include "runqslower.h"
 
 #define TASK_RUNNING 0
diff --git a/tools/bpf/runqslower/runqslower.c b/tools/bpf/runqslower/runqslower.c
index 996f0e2c560e..d89715844952 100644
--- a/tools/bpf/runqslower/runqslower.c
+++ b/tools/bpf/runqslower/runqslower.c
@@ -6,8 +6,8 @@
 #include <string.h>
 #include <sys/resource.h>
 #include <time.h>
-#include <libbpf.h>
-#include <bpf.h>
+#include <bpf/libbpf.h>
+#include <bpf/bpf.h>
 #include "runqslower.h"
 #include "runqslower.skel.h"
 
-- 
cgit v1.2.3-59-g8ed1b


From 3e689141e64df91b009a289c0559adedfe62f511 Mon Sep 17 00:00:00 2001
From: Toke Høiland-Jørgensen <toke@redhat.com>
Date: Mon, 20 Jan 2020 14:06:45 +0100
Subject: selftests: Use consistent include paths for libbpf
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix all selftests to include libbpf header files with the bpf/ prefix, to
be consistent with external users of the library. Also ensure that all
includes of exported libbpf header files (those that are exported on 'make
install' of the library) use bracketed includes instead of quoted.

To not break the build, keep the old include path until everything has been
changed to the new one; a subsequent patch will remove that.

Fixes: 6910d7d3867a ("selftests/bpf: Ensure bpf_helper_defs.h are taken from selftests dir")
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/157952560568.1683545.9649335788846513446.stgit@toke.dk
---
 tools/testing/selftests/bpf/Makefile                                | 2 +-
 tools/testing/selftests/bpf/bpf_tcp_helpers.h                       | 4 ++--
 tools/testing/selftests/bpf/bpf_trace_helpers.h                     | 2 +-
 tools/testing/selftests/bpf/bpf_util.h                              | 2 +-
 tools/testing/selftests/bpf/prog_tests/cpu_mask.c                   | 2 +-
 tools/testing/selftests/bpf/prog_tests/perf_buffer.c                | 2 +-
 tools/testing/selftests/bpf/progs/bpf_dctcp.c                       | 4 ++--
 tools/testing/selftests/bpf/progs/bpf_flow.c                        | 4 ++--
 tools/testing/selftests/bpf/progs/connect4_prog.c                   | 4 ++--
 tools/testing/selftests/bpf/progs/connect6_prog.c                   | 4 ++--
 tools/testing/selftests/bpf/progs/dev_cgroup.c                      | 2 +-
 tools/testing/selftests/bpf/progs/fentry_test.c                     | 2 +-
 tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c                   | 2 +-
 tools/testing/selftests/bpf/progs/fexit_bpf2bpf_simple.c            | 2 +-
 tools/testing/selftests/bpf/progs/fexit_test.c                      | 2 +-
 tools/testing/selftests/bpf/progs/get_cgroup_id_kern.c              | 2 +-
 tools/testing/selftests/bpf/progs/kfree_skb.c                       | 4 ++--
 tools/testing/selftests/bpf/progs/loop1.c                           | 4 ++--
 tools/testing/selftests/bpf/progs/loop2.c                           | 4 ++--
 tools/testing/selftests/bpf/progs/loop3.c                           | 4 ++--
 tools/testing/selftests/bpf/progs/loop4.c                           | 2 +-
 tools/testing/selftests/bpf/progs/loop5.c                           | 2 +-
 tools/testing/selftests/bpf/progs/netcnt_prog.c                     | 2 +-
 tools/testing/selftests/bpf/progs/pyperf.h                          | 2 +-
 tools/testing/selftests/bpf/progs/sample_map_ret0.c                 | 2 +-
 tools/testing/selftests/bpf/progs/sendmsg4_prog.c                   | 4 ++--
 tools/testing/selftests/bpf/progs/sendmsg6_prog.c                   | 4 ++--
 tools/testing/selftests/bpf/progs/socket_cookie_prog.c              | 4 ++--
 tools/testing/selftests/bpf/progs/sockmap_parse_prog.c              | 4 ++--
 tools/testing/selftests/bpf/progs/sockmap_tcp_msg_prog.c            | 4 ++--
 tools/testing/selftests/bpf/progs/sockmap_verdict_prog.c            | 4 ++--
 tools/testing/selftests/bpf/progs/sockopt_inherit.c                 | 2 +-
 tools/testing/selftests/bpf/progs/sockopt_multi.c                   | 2 +-
 tools/testing/selftests/bpf/progs/sockopt_sk.c                      | 2 +-
 tools/testing/selftests/bpf/progs/strobemeta.h                      | 2 +-
 tools/testing/selftests/bpf/progs/tailcall1.c                       | 2 +-
 tools/testing/selftests/bpf/progs/tailcall2.c                       | 2 +-
 tools/testing/selftests/bpf/progs/tailcall3.c                       | 2 +-
 tools/testing/selftests/bpf/progs/tailcall4.c                       | 2 +-
 tools/testing/selftests/bpf/progs/tailcall5.c                       | 2 +-
 tools/testing/selftests/bpf/progs/tcp_rtt.c                         | 2 +-
 tools/testing/selftests/bpf/progs/test_adjust_tail.c                | 2 +-
 tools/testing/selftests/bpf/progs/test_attach_probe.c               | 2 +-
 tools/testing/selftests/bpf/progs/test_btf_haskv.c                  | 2 +-
 tools/testing/selftests/bpf/progs/test_btf_newkv.c                  | 2 +-
 tools/testing/selftests/bpf/progs/test_btf_nokv.c                   | 2 +-
 tools/testing/selftests/bpf/progs/test_core_extern.c                | 2 +-
 tools/testing/selftests/bpf/progs/test_core_reloc_arrays.c          | 4 ++--
 .../testing/selftests/bpf/progs/test_core_reloc_bitfields_direct.c  | 4 ++--
 .../testing/selftests/bpf/progs/test_core_reloc_bitfields_probed.c  | 4 ++--
 tools/testing/selftests/bpf/progs/test_core_reloc_existence.c       | 4 ++--
 tools/testing/selftests/bpf/progs/test_core_reloc_flavors.c         | 4 ++--
 tools/testing/selftests/bpf/progs/test_core_reloc_ints.c            | 4 ++--
 tools/testing/selftests/bpf/progs/test_core_reloc_kernel.c          | 4 ++--
 tools/testing/selftests/bpf/progs/test_core_reloc_misc.c            | 4 ++--
 tools/testing/selftests/bpf/progs/test_core_reloc_mods.c            | 4 ++--
 tools/testing/selftests/bpf/progs/test_core_reloc_nesting.c         | 4 ++--
 tools/testing/selftests/bpf/progs/test_core_reloc_primitives.c      | 4 ++--
 tools/testing/selftests/bpf/progs/test_core_reloc_ptr_as_arr.c      | 4 ++--
 tools/testing/selftests/bpf/progs/test_core_reloc_size.c            | 4 ++--
 tools/testing/selftests/bpf/progs/test_get_stack_rawtp.c            | 2 +-
 tools/testing/selftests/bpf/progs/test_global_data.c                | 2 +-
 tools/testing/selftests/bpf/progs/test_global_func1.c               | 2 +-
 tools/testing/selftests/bpf/progs/test_global_func3.c               | 2 +-
 tools/testing/selftests/bpf/progs/test_global_func5.c               | 2 +-
 tools/testing/selftests/bpf/progs/test_global_func6.c               | 2 +-
 tools/testing/selftests/bpf/progs/test_global_func7.c               | 2 +-
 tools/testing/selftests/bpf/progs/test_l4lb.c                       | 4 ++--
 tools/testing/selftests/bpf/progs/test_l4lb_noinline.c              | 4 ++--
 tools/testing/selftests/bpf/progs/test_lirc_mode2_kern.c            | 2 +-
 tools/testing/selftests/bpf/progs/test_lwt_ip_encap.c               | 4 ++--
 tools/testing/selftests/bpf/progs/test_lwt_seg6local.c              | 4 ++--
 tools/testing/selftests/bpf/progs/test_map_in_map.c                 | 2 +-
 tools/testing/selftests/bpf/progs/test_map_lock.c                   | 2 +-
 tools/testing/selftests/bpf/progs/test_mmap.c                       | 2 +-
 tools/testing/selftests/bpf/progs/test_obj_id.c                     | 2 +-
 tools/testing/selftests/bpf/progs/test_overhead.c                   | 4 ++--
 tools/testing/selftests/bpf/progs/test_perf_buffer.c                | 2 +-
 tools/testing/selftests/bpf/progs/test_pinning.c                    | 2 +-
 tools/testing/selftests/bpf/progs/test_pinning_invalid.c            | 2 +-
 tools/testing/selftests/bpf/progs/test_pkt_access.c                 | 4 ++--
 tools/testing/selftests/bpf/progs/test_pkt_md_access.c              | 2 +-
 tools/testing/selftests/bpf/progs/test_probe_user.c                 | 4 ++--
 tools/testing/selftests/bpf/progs/test_queue_stack_map.h            | 2 +-
 tools/testing/selftests/bpf/progs/test_rdonly_maps.c                | 2 +-
 tools/testing/selftests/bpf/progs/test_seg6_loop.c                  | 4 ++--
 tools/testing/selftests/bpf/progs/test_select_reuseport_kern.c      | 4 ++--
 tools/testing/selftests/bpf/progs/test_send_signal_kern.c           | 2 +-
 tools/testing/selftests/bpf/progs/test_sk_lookup_kern.c             | 4 ++--
 tools/testing/selftests/bpf/progs/test_skb_cgroup_id_kern.c         | 2 +-
 tools/testing/selftests/bpf/progs/test_skb_ctx.c                    | 2 +-
 tools/testing/selftests/bpf/progs/test_skeleton.c                   | 2 +-
 tools/testing/selftests/bpf/progs/test_sock_fields_kern.c           | 4 ++--
 tools/testing/selftests/bpf/progs/test_spin_lock.c                  | 2 +-
 tools/testing/selftests/bpf/progs/test_stacktrace_build_id.c        | 2 +-
 tools/testing/selftests/bpf/progs/test_stacktrace_map.c             | 2 +-
 tools/testing/selftests/bpf/progs/test_sysctl_loop1.c               | 2 +-
 tools/testing/selftests/bpf/progs/test_sysctl_loop2.c               | 2 +-
 tools/testing/selftests/bpf/progs/test_sysctl_prog.c                | 2 +-
 tools/testing/selftests/bpf/progs/test_tc_edt.c                     | 4 ++--
 tools/testing/selftests/bpf/progs/test_tc_tunnel.c                  | 4 ++--
 tools/testing/selftests/bpf/progs/test_tcp_check_syncookie_kern.c   | 4 ++--
 tools/testing/selftests/bpf/progs/test_tcp_estats.c                 | 2 +-
 tools/testing/selftests/bpf/progs/test_tcpbpf_kern.c                | 4 ++--
 tools/testing/selftests/bpf/progs/test_tcpnotify_kern.c             | 4 ++--
 tools/testing/selftests/bpf/progs/test_tracepoint.c                 | 2 +-
 tools/testing/selftests/bpf/progs/test_tunnel_kern.c                | 4 ++--
 tools/testing/selftests/bpf/progs/test_verif_scale1.c               | 2 +-
 tools/testing/selftests/bpf/progs/test_verif_scale2.c               | 2 +-
 tools/testing/selftests/bpf/progs/test_verif_scale3.c               | 2 +-
 tools/testing/selftests/bpf/progs/test_xdp.c                        | 4 ++--
 tools/testing/selftests/bpf/progs/test_xdp_bpf2bpf.c                | 2 +-
 tools/testing/selftests/bpf/progs/test_xdp_loop.c                   | 4 ++--
 tools/testing/selftests/bpf/progs/test_xdp_meta.c                   | 2 +-
 tools/testing/selftests/bpf/progs/test_xdp_noinline.c               | 4 ++--
 tools/testing/selftests/bpf/progs/test_xdp_redirect.c               | 2 +-
 tools/testing/selftests/bpf/progs/test_xdp_vlan.c                   | 4 ++--
 tools/testing/selftests/bpf/progs/xdp_dummy.c                       | 2 +-
 tools/testing/selftests/bpf/progs/xdp_redirect_map.c                | 2 +-
 tools/testing/selftests/bpf/progs/xdp_tx.c                          | 2 +-
 tools/testing/selftests/bpf/progs/xdping_kern.c                     | 4 ++--
 tools/testing/selftests/bpf/test_cpp.cpp                            | 6 +++---
 tools/testing/selftests/bpf/test_hashmap.c                          | 2 +-
 tools/testing/selftests/bpf/test_progs.h                            | 2 +-
 tools/testing/selftests/bpf/test_sock.c                             | 2 +-
 tools/testing/selftests/bpf/test_sockmap_kern.h                     | 4 ++--
 tools/testing/selftests/bpf/test_sysctl.c                           | 2 +-
 tools/testing/selftests/bpf/trace_helpers.h                         | 2 +-
 128 files changed, 181 insertions(+), 181 deletions(-)

diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index 8240282aef7f..2cd91e0524cf 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -190,7 +190,7 @@ MENDIAN=$(if $(IS_LITTLE_ENDIAN),-mlittle-endian,-mbig-endian)
 CLANG_SYS_INCLUDES = $(call get_sys_includes,$(CLANG))
 BPF_CFLAGS = -g -D__TARGET_ARCH_$(SRCARCH) $(MENDIAN) 			\
 	     -I$(OUTPUT) -I$(CURDIR) -I$(CURDIR)/include/uapi		\
-	     -I$(APIDIR) -I$(BPFDIR) -I$(abspath $(OUTPUT)/../usr/include)
+	     -I$(APIDIR) -I$(LIBDIR) -I$(BPFDIR) -I$(abspath $(OUTPUT)/../usr/include)
 
 CLANG_CFLAGS = $(CLANG_SYS_INCLUDES) \
 	       -Wno-compare-distinct-pointer-types
diff --git a/tools/testing/selftests/bpf/bpf_tcp_helpers.h b/tools/testing/selftests/bpf/bpf_tcp_helpers.h
index 1052b9bb4320..6fee732f0297 100644
--- a/tools/testing/selftests/bpf/bpf_tcp_helpers.h
+++ b/tools/testing/selftests/bpf/bpf_tcp_helpers.h
@@ -4,8 +4,8 @@
 
 #include <stdbool.h>
 #include <linux/types.h>
-#include <bpf_helpers.h>
-#include <bpf_core_read.h>
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_core_read.h>
 
 struct sock_common {
 	unsigned char	skc_state;
diff --git a/tools/testing/selftests/bpf/bpf_trace_helpers.h b/tools/testing/selftests/bpf/bpf_trace_helpers.h
index 1719cabc97b7..c6f1354d93fb 100644
--- a/tools/testing/selftests/bpf/bpf_trace_helpers.h
+++ b/tools/testing/selftests/bpf/bpf_trace_helpers.h
@@ -2,7 +2,7 @@
 #ifndef __BPF_TRACE_HELPERS_H
 #define __BPF_TRACE_HELPERS_H
 
-#include <bpf_helpers.h>
+#include <bpf/bpf_helpers.h>
 
 #define ___bpf_concat(a, b) a ## b
 #define ___bpf_apply(fn, n) ___bpf_concat(fn, n)
diff --git a/tools/testing/selftests/bpf/bpf_util.h b/tools/testing/selftests/bpf/bpf_util.h
index ec219f84e041..a3352a64c067 100644
--- a/tools/testing/selftests/bpf/bpf_util.h
+++ b/tools/testing/selftests/bpf/bpf_util.h
@@ -6,7 +6,7 @@
 #include <stdlib.h>
 #include <string.h>
 #include <errno.h>
-#include <libbpf.h> /* libbpf_num_possible_cpus */
+#include <bpf/libbpf.h> /* libbpf_num_possible_cpus */
 
 static inline unsigned int bpf_num_possible_cpus(void)
 {
diff --git a/tools/testing/selftests/bpf/prog_tests/cpu_mask.c b/tools/testing/selftests/bpf/prog_tests/cpu_mask.c
index 1fa1bdbaffa9..f7c7e25232be 100644
--- a/tools/testing/selftests/bpf/prog_tests/cpu_mask.c
+++ b/tools/testing/selftests/bpf/prog_tests/cpu_mask.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <test_progs.h>
 #include <bpf/btf.h>
-#include "libbpf_internal.h"
+#include "bpf/libbpf_internal.h"
 
 static int duration = 0;
 
diff --git a/tools/testing/selftests/bpf/prog_tests/perf_buffer.c b/tools/testing/selftests/bpf/prog_tests/perf_buffer.c
index cf6c87936c69..1450ea2dd4cc 100644
--- a/tools/testing/selftests/bpf/prog_tests/perf_buffer.c
+++ b/tools/testing/selftests/bpf/prog_tests/perf_buffer.c
@@ -4,7 +4,7 @@
 #include <sched.h>
 #include <sys/socket.h>
 #include <test_progs.h>
-#include "libbpf_internal.h"
+#include "bpf/libbpf_internal.h"
 
 static void on_sample(void *ctx, int cpu, void *data, __u32 size)
 {
diff --git a/tools/testing/selftests/bpf/progs/bpf_dctcp.c b/tools/testing/selftests/bpf/progs/bpf_dctcp.c
index 19650f43c192..b631fb5032d2 100644
--- a/tools/testing/selftests/bpf/progs/bpf_dctcp.c
+++ b/tools/testing/selftests/bpf/progs/bpf_dctcp.c
@@ -8,8 +8,8 @@
 
 #include <linux/bpf.h>
 #include <linux/types.h>
-#include <bpf_helpers.h>
-#include <bpf_trace_helpers.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_trace_helpers.h"
 #include "bpf_tcp_helpers.h"
 
 char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/bpf_flow.c b/tools/testing/selftests/bpf/progs/bpf_flow.c
index 040a44206f29..9941f0ba471e 100644
--- a/tools/testing/selftests/bpf/progs/bpf_flow.c
+++ b/tools/testing/selftests/bpf/progs/bpf_flow.c
@@ -16,8 +16,8 @@
 #include <sys/socket.h>
 #include <linux/if_tunnel.h>
 #include <linux/mpls.h>
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 int _version SEC("version") = 1;
 #define PROG(F) SEC(#F) int bpf_func_##F
diff --git a/tools/testing/selftests/bpf/progs/connect4_prog.c b/tools/testing/selftests/bpf/progs/connect4_prog.c
index 1fd244d35ba9..75085119c5bb 100644
--- a/tools/testing/selftests/bpf/progs/connect4_prog.c
+++ b/tools/testing/selftests/bpf/progs/connect4_prog.c
@@ -9,8 +9,8 @@
 #include <linux/in6.h>
 #include <sys/socket.h>
 
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 #define SRC_REWRITE_IP4		0x7f000004U
 #define DST_REWRITE_IP4		0x7f000001U
diff --git a/tools/testing/selftests/bpf/progs/connect6_prog.c b/tools/testing/selftests/bpf/progs/connect6_prog.c
index 26397ab7b3c7..506d0f81a375 100644
--- a/tools/testing/selftests/bpf/progs/connect6_prog.c
+++ b/tools/testing/selftests/bpf/progs/connect6_prog.c
@@ -9,8 +9,8 @@
 #include <linux/in6.h>
 #include <sys/socket.h>
 
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 #define SRC_REWRITE_IP6_0	0
 #define SRC_REWRITE_IP6_1	0
diff --git a/tools/testing/selftests/bpf/progs/dev_cgroup.c b/tools/testing/selftests/bpf/progs/dev_cgroup.c
index ce41a3475f27..8924e06bdef0 100644
--- a/tools/testing/selftests/bpf/progs/dev_cgroup.c
+++ b/tools/testing/selftests/bpf/progs/dev_cgroup.c
@@ -7,7 +7,7 @@
 
 #include <linux/bpf.h>
 #include <linux/version.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 SEC("cgroup/dev")
 int bpf_prog1(struct bpf_cgroup_dev_ctx *ctx)
diff --git a/tools/testing/selftests/bpf/progs/fentry_test.c b/tools/testing/selftests/bpf/progs/fentry_test.c
index 1ae1779a8e2e..38d3a82144ca 100644
--- a/tools/testing/selftests/bpf/progs/fentry_test.c
+++ b/tools/testing/selftests/bpf/progs/fentry_test.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 /* Copyright (c) 2019 Facebook */
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #include "bpf_trace_helpers.h"
 
 char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c b/tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c
index 7c17ee159378..ec5710767d13 100644
--- a/tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c
+++ b/tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 /* Copyright (c) 2019 Facebook */
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #include "bpf_trace_helpers.h"
 
 struct sk_buff {
diff --git a/tools/testing/selftests/bpf/progs/fexit_bpf2bpf_simple.c b/tools/testing/selftests/bpf/progs/fexit_bpf2bpf_simple.c
index ac527b10dd80..92f3fa47cf40 100644
--- a/tools/testing/selftests/bpf/progs/fexit_bpf2bpf_simple.c
+++ b/tools/testing/selftests/bpf/progs/fexit_bpf2bpf_simple.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 /* Copyright (c) 2019 Facebook */
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #include "bpf_trace_helpers.h"
 
 struct sk_buff {
diff --git a/tools/testing/selftests/bpf/progs/fexit_test.c b/tools/testing/selftests/bpf/progs/fexit_test.c
index f0dd412ca128..348109b9ea07 100644
--- a/tools/testing/selftests/bpf/progs/fexit_test.c
+++ b/tools/testing/selftests/bpf/progs/fexit_test.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 /* Copyright (c) 2019 Facebook */
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #include "bpf_trace_helpers.h"
 
 char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/get_cgroup_id_kern.c b/tools/testing/selftests/bpf/progs/get_cgroup_id_kern.c
index 16c54ade6888..6b42db2fe391 100644
--- a/tools/testing/selftests/bpf/progs/get_cgroup_id_kern.c
+++ b/tools/testing/selftests/bpf/progs/get_cgroup_id_kern.c
@@ -2,7 +2,7 @@
 // Copyright (c) 2018 Facebook
 
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 struct {
 	__uint(type, BPF_MAP_TYPE_ARRAY);
diff --git a/tools/testing/selftests/bpf/progs/kfree_skb.c b/tools/testing/selftests/bpf/progs/kfree_skb.c
index 046fae868f93..8f48a909f079 100644
--- a/tools/testing/selftests/bpf/progs/kfree_skb.c
+++ b/tools/testing/selftests/bpf/progs/kfree_skb.c
@@ -2,8 +2,8 @@
 // Copyright (c) 2019 Facebook
 #include <linux/bpf.h>
 #include <stdbool.h>
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 #include "bpf_trace_helpers.h"
 
 char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/loop1.c b/tools/testing/selftests/bpf/progs/loop1.c
index 40ac722a9da5..50e66772c046 100644
--- a/tools/testing/selftests/bpf/progs/loop1.c
+++ b/tools/testing/selftests/bpf/progs/loop1.c
@@ -6,8 +6,8 @@
 #include <stddef.h>
 #include <stdbool.h>
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
-#include "bpf_tracing.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
 
 char _license[] SEC("license") = "GPL";
 
diff --git a/tools/testing/selftests/bpf/progs/loop2.c b/tools/testing/selftests/bpf/progs/loop2.c
index bb80f29aa7f7..947bb7e988c2 100644
--- a/tools/testing/selftests/bpf/progs/loop2.c
+++ b/tools/testing/selftests/bpf/progs/loop2.c
@@ -6,8 +6,8 @@
 #include <stddef.h>
 #include <stdbool.h>
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
-#include "bpf_tracing.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
 
 char _license[] SEC("license") = "GPL";
 
diff --git a/tools/testing/selftests/bpf/progs/loop3.c b/tools/testing/selftests/bpf/progs/loop3.c
index 2b9165a7afe1..76e93b31c14b 100644
--- a/tools/testing/selftests/bpf/progs/loop3.c
+++ b/tools/testing/selftests/bpf/progs/loop3.c
@@ -6,8 +6,8 @@
 #include <stddef.h>
 #include <stdbool.h>
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
-#include "bpf_tracing.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
 
 char _license[] SEC("license") = "GPL";
 
diff --git a/tools/testing/selftests/bpf/progs/loop4.c b/tools/testing/selftests/bpf/progs/loop4.c
index 650859022771..b35337926d66 100644
--- a/tools/testing/selftests/bpf/progs/loop4.c
+++ b/tools/testing/selftests/bpf/progs/loop4.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 // Copyright (c) 2019 Facebook
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 char _license[] SEC("license") = "GPL";
 
diff --git a/tools/testing/selftests/bpf/progs/loop5.c b/tools/testing/selftests/bpf/progs/loop5.c
index 28d1d668f07c..913791923fa3 100644
--- a/tools/testing/selftests/bpf/progs/loop5.c
+++ b/tools/testing/selftests/bpf/progs/loop5.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 // Copyright (c) 2019 Facebook
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #define barrier() __asm__ __volatile__("": : :"memory")
 
 char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/netcnt_prog.c b/tools/testing/selftests/bpf/progs/netcnt_prog.c
index 38a997852cad..d071adf178bd 100644
--- a/tools/testing/selftests/bpf/progs/netcnt_prog.c
+++ b/tools/testing/selftests/bpf/progs/netcnt_prog.c
@@ -2,7 +2,7 @@
 #include <linux/bpf.h>
 #include <linux/version.h>
 
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #include "netcnt_common.h"
 
 #define MAX_BPS	(3 * 1024 * 1024)
diff --git a/tools/testing/selftests/bpf/progs/pyperf.h b/tools/testing/selftests/bpf/progs/pyperf.h
index e186899954e9..cc615b82b56e 100644
--- a/tools/testing/selftests/bpf/progs/pyperf.h
+++ b/tools/testing/selftests/bpf/progs/pyperf.h
@@ -6,7 +6,7 @@
 #include <stddef.h>
 #include <stdbool.h>
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 #define FUNCTION_NAME_LEN 64
 #define FILE_NAME_LEN 128
diff --git a/tools/testing/selftests/bpf/progs/sample_map_ret0.c b/tools/testing/selftests/bpf/progs/sample_map_ret0.c
index 0756303676ac..1612a32007b6 100644
--- a/tools/testing/selftests/bpf/progs/sample_map_ret0.c
+++ b/tools/testing/selftests/bpf/progs/sample_map_ret0.c
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause) */
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 struct bpf_map_def SEC("maps") htab = {
 	.type = BPF_MAP_TYPE_HASH,
diff --git a/tools/testing/selftests/bpf/progs/sendmsg4_prog.c b/tools/testing/selftests/bpf/progs/sendmsg4_prog.c
index a91536b1c47e..092d9da536f3 100644
--- a/tools/testing/selftests/bpf/progs/sendmsg4_prog.c
+++ b/tools/testing/selftests/bpf/progs/sendmsg4_prog.c
@@ -5,8 +5,8 @@
 #include <linux/bpf.h>
 #include <sys/socket.h>
 
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 #define SRC1_IP4		0xAC100001U /* 172.16.0.1 */
 #define SRC2_IP4		0x00000000U
diff --git a/tools/testing/selftests/bpf/progs/sendmsg6_prog.c b/tools/testing/selftests/bpf/progs/sendmsg6_prog.c
index a68062820410..255a432bc163 100644
--- a/tools/testing/selftests/bpf/progs/sendmsg6_prog.c
+++ b/tools/testing/selftests/bpf/progs/sendmsg6_prog.c
@@ -5,8 +5,8 @@
 #include <linux/bpf.h>
 #include <sys/socket.h>
 
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 #define SRC_REWRITE_IP6_0	0
 #define SRC_REWRITE_IP6_1	0
diff --git a/tools/testing/selftests/bpf/progs/socket_cookie_prog.c b/tools/testing/selftests/bpf/progs/socket_cookie_prog.c
index e4440fdd94cb..0cb5656a22b0 100644
--- a/tools/testing/selftests/bpf/progs/socket_cookie_prog.c
+++ b/tools/testing/selftests/bpf/progs/socket_cookie_prog.c
@@ -4,8 +4,8 @@
 #include <linux/bpf.h>
 #include <sys/socket.h>
 
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 struct socket_cookie {
 	__u64 cookie_key;
diff --git a/tools/testing/selftests/bpf/progs/sockmap_parse_prog.c b/tools/testing/selftests/bpf/progs/sockmap_parse_prog.c
index 9390e0244259..a5c6d5903b22 100644
--- a/tools/testing/selftests/bpf/progs/sockmap_parse_prog.c
+++ b/tools/testing/selftests/bpf/progs/sockmap_parse_prog.c
@@ -1,6 +1,6 @@
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 int _version SEC("version") = 1;
 
diff --git a/tools/testing/selftests/bpf/progs/sockmap_tcp_msg_prog.c b/tools/testing/selftests/bpf/progs/sockmap_tcp_msg_prog.c
index e80484d98a1a..fdb4bf4408fa 100644
--- a/tools/testing/selftests/bpf/progs/sockmap_tcp_msg_prog.c
+++ b/tools/testing/selftests/bpf/progs/sockmap_tcp_msg_prog.c
@@ -1,7 +1,7 @@
 #include <linux/bpf.h>
 
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 int _version SEC("version") = 1;
 
diff --git a/tools/testing/selftests/bpf/progs/sockmap_verdict_prog.c b/tools/testing/selftests/bpf/progs/sockmap_verdict_prog.c
index 433e23918a62..4797dc985064 100644
--- a/tools/testing/selftests/bpf/progs/sockmap_verdict_prog.c
+++ b/tools/testing/selftests/bpf/progs/sockmap_verdict_prog.c
@@ -1,6 +1,6 @@
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 int _version SEC("version") = 1;
 
diff --git a/tools/testing/selftests/bpf/progs/sockopt_inherit.c b/tools/testing/selftests/bpf/progs/sockopt_inherit.c
index dede0fcd6102..c6d428a8d785 100644
--- a/tools/testing/selftests/bpf/progs/sockopt_inherit.c
+++ b/tools/testing/selftests/bpf/progs/sockopt_inherit.c
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 char _license[] SEC("license") = "GPL";
 __u32 _version SEC("version") = 1;
diff --git a/tools/testing/selftests/bpf/progs/sockopt_multi.c b/tools/testing/selftests/bpf/progs/sockopt_multi.c
index 4afd2595c08e..9d8c212dde9f 100644
--- a/tools/testing/selftests/bpf/progs/sockopt_multi.c
+++ b/tools/testing/selftests/bpf/progs/sockopt_multi.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <netinet/in.h>
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 char _license[] SEC("license") = "GPL";
 __u32 _version SEC("version") = 1;
diff --git a/tools/testing/selftests/bpf/progs/sockopt_sk.c b/tools/testing/selftests/bpf/progs/sockopt_sk.c
index 1bafbb944e37..d5a5eeb5fb52 100644
--- a/tools/testing/selftests/bpf/progs/sockopt_sk.c
+++ b/tools/testing/selftests/bpf/progs/sockopt_sk.c
@@ -3,7 +3,7 @@
 #include <netinet/in.h>
 #include <netinet/tcp.h>
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 char _license[] SEC("license") = "GPL";
 __u32 _version SEC("version") = 1;
diff --git a/tools/testing/selftests/bpf/progs/strobemeta.h b/tools/testing/selftests/bpf/progs/strobemeta.h
index 4bf16e0a1b0e..ad61b722a9de 100644
--- a/tools/testing/selftests/bpf/progs/strobemeta.h
+++ b/tools/testing/selftests/bpf/progs/strobemeta.h
@@ -8,7 +8,7 @@
 #include <linux/ptrace.h>
 #include <linux/sched.h>
 #include <linux/types.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 typedef uint32_t pid_t;
 struct task_struct {};
diff --git a/tools/testing/selftests/bpf/progs/tailcall1.c b/tools/testing/selftests/bpf/progs/tailcall1.c
index 63531e1a9fa4..1f407e65ae52 100644
--- a/tools/testing/selftests/bpf/progs/tailcall1.c
+++ b/tools/testing/selftests/bpf/progs/tailcall1.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <linux/bpf.h>
 
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 struct {
 	__uint(type, BPF_MAP_TYPE_PROG_ARRAY);
diff --git a/tools/testing/selftests/bpf/progs/tailcall2.c b/tools/testing/selftests/bpf/progs/tailcall2.c
index 21c85c477210..a093e739cf0e 100644
--- a/tools/testing/selftests/bpf/progs/tailcall2.c
+++ b/tools/testing/selftests/bpf/progs/tailcall2.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <linux/bpf.h>
 
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 struct {
 	__uint(type, BPF_MAP_TYPE_PROG_ARRAY);
diff --git a/tools/testing/selftests/bpf/progs/tailcall3.c b/tools/testing/selftests/bpf/progs/tailcall3.c
index 1ecae198b8c1..cabda877cf0a 100644
--- a/tools/testing/selftests/bpf/progs/tailcall3.c
+++ b/tools/testing/selftests/bpf/progs/tailcall3.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <linux/bpf.h>
 
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 struct {
 	__uint(type, BPF_MAP_TYPE_PROG_ARRAY);
diff --git a/tools/testing/selftests/bpf/progs/tailcall4.c b/tools/testing/selftests/bpf/progs/tailcall4.c
index 499388758119..f82075b47d7d 100644
--- a/tools/testing/selftests/bpf/progs/tailcall4.c
+++ b/tools/testing/selftests/bpf/progs/tailcall4.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <linux/bpf.h>
 
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 struct {
 	__uint(type, BPF_MAP_TYPE_PROG_ARRAY);
diff --git a/tools/testing/selftests/bpf/progs/tailcall5.c b/tools/testing/selftests/bpf/progs/tailcall5.c
index 49c64eb53f19..ce5450744fd4 100644
--- a/tools/testing/selftests/bpf/progs/tailcall5.c
+++ b/tools/testing/selftests/bpf/progs/tailcall5.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <linux/bpf.h>
 
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 struct {
 	__uint(type, BPF_MAP_TYPE_PROG_ARRAY);
diff --git a/tools/testing/selftests/bpf/progs/tcp_rtt.c b/tools/testing/selftests/bpf/progs/tcp_rtt.c
index 2cf813a06b83..0cb3204ddb18 100644
--- a/tools/testing/selftests/bpf/progs/tcp_rtt.c
+++ b/tools/testing/selftests/bpf/progs/tcp_rtt.c
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 char _license[] SEC("license") = "GPL";
 __u32 _version SEC("version") = 1;
diff --git a/tools/testing/selftests/bpf/progs/test_adjust_tail.c b/tools/testing/selftests/bpf/progs/test_adjust_tail.c
index 4cd5e860c903..b7fc85769bdc 100644
--- a/tools/testing/selftests/bpf/progs/test_adjust_tail.c
+++ b/tools/testing/selftests/bpf/progs/test_adjust_tail.c
@@ -7,7 +7,7 @@
  */
 #include <linux/bpf.h>
 #include <linux/if_ether.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 int _version SEC("version") = 1;
 
diff --git a/tools/testing/selftests/bpf/progs/test_attach_probe.c b/tools/testing/selftests/bpf/progs/test_attach_probe.c
index 221b69700625..dd8fae6660ab 100644
--- a/tools/testing/selftests/bpf/progs/test_attach_probe.c
+++ b/tools/testing/selftests/bpf/progs/test_attach_probe.c
@@ -3,7 +3,7 @@
 
 #include <linux/ptrace.h>
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 int kprobe_res = 0;
 int kretprobe_res = 0;
diff --git a/tools/testing/selftests/bpf/progs/test_btf_haskv.c b/tools/testing/selftests/bpf/progs/test_btf_haskv.c
index 62ad7e22105e..88b0566da13d 100644
--- a/tools/testing/selftests/bpf/progs/test_btf_haskv.c
+++ b/tools/testing/selftests/bpf/progs/test_btf_haskv.c
@@ -1,7 +1,7 @@
 /* SPDX-License-Identifier: GPL-2.0 */
 /* Copyright (c) 2018 Facebook */
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #include "bpf_legacy.h"
 
 int _version SEC("version") = 1;
diff --git a/tools/testing/selftests/bpf/progs/test_btf_newkv.c b/tools/testing/selftests/bpf/progs/test_btf_newkv.c
index fb8d91a1dbe0..a924e53c8e9d 100644
--- a/tools/testing/selftests/bpf/progs/test_btf_newkv.c
+++ b/tools/testing/selftests/bpf/progs/test_btf_newkv.c
@@ -1,7 +1,7 @@
 /* SPDX-License-Identifier: GPL-2.0 */
 /* Copyright (c) 2018 Facebook */
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #include "bpf_legacy.h"
 
 int _version SEC("version") = 1;
diff --git a/tools/testing/selftests/bpf/progs/test_btf_nokv.c b/tools/testing/selftests/bpf/progs/test_btf_nokv.c
index 3f4422044759..983aedd1c072 100644
--- a/tools/testing/selftests/bpf/progs/test_btf_nokv.c
+++ b/tools/testing/selftests/bpf/progs/test_btf_nokv.c
@@ -1,7 +1,7 @@
 /* SPDX-License-Identifier: GPL-2.0 */
 /* Copyright (c) 2018 Facebook */
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 int _version SEC("version") = 1;
 
diff --git a/tools/testing/selftests/bpf/progs/test_core_extern.c b/tools/testing/selftests/bpf/progs/test_core_extern.c
index 9bfc91d9d004..3ac3603ad53d 100644
--- a/tools/testing/selftests/bpf/progs/test_core_extern.c
+++ b/tools/testing/selftests/bpf/progs/test_core_extern.c
@@ -5,7 +5,7 @@
 #include <stdbool.h>
 #include <linux/ptrace.h>
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 /* non-existing BPF helper, to test dead code elimination */
 static int (*bpf_missing_helper)(const void *arg1, int arg2) = (void *) 999;
diff --git a/tools/testing/selftests/bpf/progs/test_core_reloc_arrays.c b/tools/testing/selftests/bpf/progs/test_core_reloc_arrays.c
index 053b86f6b53f..51b3f79df523 100644
--- a/tools/testing/selftests/bpf/progs/test_core_reloc_arrays.c
+++ b/tools/testing/selftests/bpf/progs/test_core_reloc_arrays.c
@@ -3,8 +3,8 @@
 
 #include <linux/bpf.h>
 #include <stdint.h>
-#include "bpf_helpers.h"
-#include "bpf_core_read.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_core_read.h>
 
 char _license[] SEC("license") = "GPL";
 
diff --git a/tools/testing/selftests/bpf/progs/test_core_reloc_bitfields_direct.c b/tools/testing/selftests/bpf/progs/test_core_reloc_bitfields_direct.c
index edc0f7c9e56d..56aec20212b5 100644
--- a/tools/testing/selftests/bpf/progs/test_core_reloc_bitfields_direct.c
+++ b/tools/testing/selftests/bpf/progs/test_core_reloc_bitfields_direct.c
@@ -3,8 +3,8 @@
 
 #include <linux/bpf.h>
 #include <stdint.h>
-#include "bpf_helpers.h"
-#include "bpf_core_read.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_core_read.h>
 
 char _license[] SEC("license") = "GPL";
 
diff --git a/tools/testing/selftests/bpf/progs/test_core_reloc_bitfields_probed.c b/tools/testing/selftests/bpf/progs/test_core_reloc_bitfields_probed.c
index 6c20e433558b..ab1e647aeb31 100644
--- a/tools/testing/selftests/bpf/progs/test_core_reloc_bitfields_probed.c
+++ b/tools/testing/selftests/bpf/progs/test_core_reloc_bitfields_probed.c
@@ -3,8 +3,8 @@
 
 #include <linux/bpf.h>
 #include <stdint.h>
-#include "bpf_helpers.h"
-#include "bpf_core_read.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_core_read.h>
 
 char _license[] SEC("license") = "GPL";
 
diff --git a/tools/testing/selftests/bpf/progs/test_core_reloc_existence.c b/tools/testing/selftests/bpf/progs/test_core_reloc_existence.c
index 1b7f0ae49cfb..7e45e2bdf6cd 100644
--- a/tools/testing/selftests/bpf/progs/test_core_reloc_existence.c
+++ b/tools/testing/selftests/bpf/progs/test_core_reloc_existence.c
@@ -3,8 +3,8 @@
 
 #include <linux/bpf.h>
 #include <stdint.h>
-#include "bpf_helpers.h"
-#include "bpf_core_read.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_core_read.h>
 
 char _license[] SEC("license") = "GPL";
 
diff --git a/tools/testing/selftests/bpf/progs/test_core_reloc_flavors.c b/tools/testing/selftests/bpf/progs/test_core_reloc_flavors.c
index b5dbeef540fd..525acc2f841b 100644
--- a/tools/testing/selftests/bpf/progs/test_core_reloc_flavors.c
+++ b/tools/testing/selftests/bpf/progs/test_core_reloc_flavors.c
@@ -3,8 +3,8 @@
 
 #include <linux/bpf.h>
 #include <stdint.h>
-#include "bpf_helpers.h"
-#include "bpf_core_read.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_core_read.h>
 
 char _license[] SEC("license") = "GPL";
 
diff --git a/tools/testing/selftests/bpf/progs/test_core_reloc_ints.c b/tools/testing/selftests/bpf/progs/test_core_reloc_ints.c
index c78ab6d28a14..6b5290739806 100644
--- a/tools/testing/selftests/bpf/progs/test_core_reloc_ints.c
+++ b/tools/testing/selftests/bpf/progs/test_core_reloc_ints.c
@@ -3,8 +3,8 @@
 
 #include <linux/bpf.h>
 #include <stdint.h>
-#include "bpf_helpers.h"
-#include "bpf_core_read.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_core_read.h>
 
 char _license[] SEC("license") = "GPL";
 
diff --git a/tools/testing/selftests/bpf/progs/test_core_reloc_kernel.c b/tools/testing/selftests/bpf/progs/test_core_reloc_kernel.c
index 270de441b60a..aba928fd60d3 100644
--- a/tools/testing/selftests/bpf/progs/test_core_reloc_kernel.c
+++ b/tools/testing/selftests/bpf/progs/test_core_reloc_kernel.c
@@ -3,8 +3,8 @@
 
 #include <linux/bpf.h>
 #include <stdint.h>
-#include "bpf_helpers.h"
-#include "bpf_core_read.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_core_read.h>
 
 char _license[] SEC("license") = "GPL";
 
diff --git a/tools/testing/selftests/bpf/progs/test_core_reloc_misc.c b/tools/testing/selftests/bpf/progs/test_core_reloc_misc.c
index 292a5c4ee76a..d5756dbdef82 100644
--- a/tools/testing/selftests/bpf/progs/test_core_reloc_misc.c
+++ b/tools/testing/selftests/bpf/progs/test_core_reloc_misc.c
@@ -3,8 +3,8 @@
 
 #include <linux/bpf.h>
 #include <stdint.h>
-#include "bpf_helpers.h"
-#include "bpf_core_read.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_core_read.h>
 
 char _license[] SEC("license") = "GPL";
 
diff --git a/tools/testing/selftests/bpf/progs/test_core_reloc_mods.c b/tools/testing/selftests/bpf/progs/test_core_reloc_mods.c
index 0b28bfacc8fd..8b533db4a7a5 100644
--- a/tools/testing/selftests/bpf/progs/test_core_reloc_mods.c
+++ b/tools/testing/selftests/bpf/progs/test_core_reloc_mods.c
@@ -3,8 +3,8 @@
 
 #include <linux/bpf.h>
 #include <stdint.h>
-#include "bpf_helpers.h"
-#include "bpf_core_read.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_core_read.h>
 
 char _license[] SEC("license") = "GPL";
 
diff --git a/tools/testing/selftests/bpf/progs/test_core_reloc_nesting.c b/tools/testing/selftests/bpf/progs/test_core_reloc_nesting.c
index 39279bf0c9db..2b4b6d49c677 100644
--- a/tools/testing/selftests/bpf/progs/test_core_reloc_nesting.c
+++ b/tools/testing/selftests/bpf/progs/test_core_reloc_nesting.c
@@ -3,8 +3,8 @@
 
 #include <linux/bpf.h>
 #include <stdint.h>
-#include "bpf_helpers.h"
-#include "bpf_core_read.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_core_read.h>
 
 char _license[] SEC("license") = "GPL";
 
diff --git a/tools/testing/selftests/bpf/progs/test_core_reloc_primitives.c b/tools/testing/selftests/bpf/progs/test_core_reloc_primitives.c
index ea57973cdd19..2a8975678aa6 100644
--- a/tools/testing/selftests/bpf/progs/test_core_reloc_primitives.c
+++ b/tools/testing/selftests/bpf/progs/test_core_reloc_primitives.c
@@ -3,8 +3,8 @@
 
 #include <linux/bpf.h>
 #include <stdint.h>
-#include "bpf_helpers.h"
-#include "bpf_core_read.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_core_read.h>
 
 char _license[] SEC("license") = "GPL";
 
diff --git a/tools/testing/selftests/bpf/progs/test_core_reloc_ptr_as_arr.c b/tools/testing/selftests/bpf/progs/test_core_reloc_ptr_as_arr.c
index d1eb59d4ea64..ca61a5183b88 100644
--- a/tools/testing/selftests/bpf/progs/test_core_reloc_ptr_as_arr.c
+++ b/tools/testing/selftests/bpf/progs/test_core_reloc_ptr_as_arr.c
@@ -3,8 +3,8 @@
 
 #include <linux/bpf.h>
 #include <stdint.h>
-#include "bpf_helpers.h"
-#include "bpf_core_read.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_core_read.h>
 
 char _license[] SEC("license") = "GPL";
 
diff --git a/tools/testing/selftests/bpf/progs/test_core_reloc_size.c b/tools/testing/selftests/bpf/progs/test_core_reloc_size.c
index 9e091124d3bd..d7fb6cfc7891 100644
--- a/tools/testing/selftests/bpf/progs/test_core_reloc_size.c
+++ b/tools/testing/selftests/bpf/progs/test_core_reloc_size.c
@@ -3,8 +3,8 @@
 
 #include <linux/bpf.h>
 #include <stdint.h>
-#include "bpf_helpers.h"
-#include "bpf_core_read.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_core_read.h>
 
 char _license[] SEC("license") = "GPL";
 
diff --git a/tools/testing/selftests/bpf/progs/test_get_stack_rawtp.c b/tools/testing/selftests/bpf/progs/test_get_stack_rawtp.c
index 6a4a8f57f174..29817a703984 100644
--- a/tools/testing/selftests/bpf/progs/test_get_stack_rawtp.c
+++ b/tools/testing/selftests/bpf/progs/test_get_stack_rawtp.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 /* Permit pretty deep stack traces */
 #define MAX_STACK_RAWTP 100
diff --git a/tools/testing/selftests/bpf/progs/test_global_data.c b/tools/testing/selftests/bpf/progs/test_global_data.c
index 32a6073acb99..dd7a4d3dbc0d 100644
--- a/tools/testing/selftests/bpf/progs/test_global_data.c
+++ b/tools/testing/selftests/bpf/progs/test_global_data.c
@@ -5,7 +5,7 @@
 #include <linux/pkt_cls.h>
 #include <string.h>
 
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 struct {
 	__uint(type, BPF_MAP_TYPE_ARRAY);
diff --git a/tools/testing/selftests/bpf/progs/test_global_func1.c b/tools/testing/selftests/bpf/progs/test_global_func1.c
index 97d57d6e244e..880260f6d536 100644
--- a/tools/testing/selftests/bpf/progs/test_global_func1.c
+++ b/tools/testing/selftests/bpf/progs/test_global_func1.c
@@ -2,7 +2,7 @@
 /* Copyright (c) 2020 Facebook */
 #include <stddef.h>
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 #ifndef MAX_STACK
 #define MAX_STACK (512 - 3 * 32 + 8)
diff --git a/tools/testing/selftests/bpf/progs/test_global_func3.c b/tools/testing/selftests/bpf/progs/test_global_func3.c
index 514ecf9f51b0..86f0ecb304fc 100644
--- a/tools/testing/selftests/bpf/progs/test_global_func3.c
+++ b/tools/testing/selftests/bpf/progs/test_global_func3.c
@@ -2,7 +2,7 @@
 /* Copyright (c) 2020 Facebook */
 #include <stddef.h>
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 __attribute__ ((noinline))
 int f1(struct __sk_buff *skb)
diff --git a/tools/testing/selftests/bpf/progs/test_global_func5.c b/tools/testing/selftests/bpf/progs/test_global_func5.c
index 86787c03cea8..260c25b827ef 100644
--- a/tools/testing/selftests/bpf/progs/test_global_func5.c
+++ b/tools/testing/selftests/bpf/progs/test_global_func5.c
@@ -2,7 +2,7 @@
 /* Copyright (c) 2020 Facebook */
 #include <stddef.h>
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 __attribute__ ((noinline))
 int f1(struct __sk_buff *skb)
diff --git a/tools/testing/selftests/bpf/progs/test_global_func6.c b/tools/testing/selftests/bpf/progs/test_global_func6.c
index e215fb3e6f02..69e19c64e10b 100644
--- a/tools/testing/selftests/bpf/progs/test_global_func6.c
+++ b/tools/testing/selftests/bpf/progs/test_global_func6.c
@@ -2,7 +2,7 @@
 /* Copyright (c) 2020 Facebook */
 #include <stddef.h>
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 __attribute__ ((noinline))
 int f1(struct __sk_buff *skb)
diff --git a/tools/testing/selftests/bpf/progs/test_global_func7.c b/tools/testing/selftests/bpf/progs/test_global_func7.c
index ff98d93916fd..309b3f6136bd 100644
--- a/tools/testing/selftests/bpf/progs/test_global_func7.c
+++ b/tools/testing/selftests/bpf/progs/test_global_func7.c
@@ -2,7 +2,7 @@
 /* Copyright (c) 2020 Facebook */
 #include <stddef.h>
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 __attribute__ ((noinline))
 void foo(struct __sk_buff *skb)
diff --git a/tools/testing/selftests/bpf/progs/test_l4lb.c b/tools/testing/selftests/bpf/progs/test_l4lb.c
index 1d652ee8e73d..33493911d87a 100644
--- a/tools/testing/selftests/bpf/progs/test_l4lb.c
+++ b/tools/testing/selftests/bpf/progs/test_l4lb.c
@@ -17,9 +17,9 @@
 #include <linux/icmpv6.h>
 #include <linux/tcp.h>
 #include <linux/udp.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #include "test_iptunnel_common.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_endian.h>
 
 int _version SEC("version") = 1;
 
diff --git a/tools/testing/selftests/bpf/progs/test_l4lb_noinline.c b/tools/testing/selftests/bpf/progs/test_l4lb_noinline.c
index 2e4efe70b1e5..28351936a438 100644
--- a/tools/testing/selftests/bpf/progs/test_l4lb_noinline.c
+++ b/tools/testing/selftests/bpf/progs/test_l4lb_noinline.c
@@ -13,9 +13,9 @@
 #include <linux/icmpv6.h>
 #include <linux/tcp.h>
 #include <linux/udp.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #include "test_iptunnel_common.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_endian.h>
 
 int _version SEC("version") = 1;
 
diff --git a/tools/testing/selftests/bpf/progs/test_lirc_mode2_kern.c b/tools/testing/selftests/bpf/progs/test_lirc_mode2_kern.c
index 4147130cc3b7..7a6620671a83 100644
--- a/tools/testing/selftests/bpf/progs/test_lirc_mode2_kern.c
+++ b/tools/testing/selftests/bpf/progs/test_lirc_mode2_kern.c
@@ -5,7 +5,7 @@
 
 #include <linux/bpf.h>
 #include <linux/lirc.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 SEC("lirc_mode2")
 int bpf_decoder(unsigned int *sample)
diff --git a/tools/testing/selftests/bpf/progs/test_lwt_ip_encap.c b/tools/testing/selftests/bpf/progs/test_lwt_ip_encap.c
index c957d6dfe6d7..d6cb986e7533 100644
--- a/tools/testing/selftests/bpf/progs/test_lwt_ip_encap.c
+++ b/tools/testing/selftests/bpf/progs/test_lwt_ip_encap.c
@@ -4,8 +4,8 @@
 #include <linux/bpf.h>
 #include <linux/ip.h>
 #include <linux/ipv6.h>
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 struct grehdr {
 	__be16 flags;
diff --git a/tools/testing/selftests/bpf/progs/test_lwt_seg6local.c b/tools/testing/selftests/bpf/progs/test_lwt_seg6local.c
index 41a3ebcd593d..48ff2b2ad5e7 100644
--- a/tools/testing/selftests/bpf/progs/test_lwt_seg6local.c
+++ b/tools/testing/selftests/bpf/progs/test_lwt_seg6local.c
@@ -3,8 +3,8 @@
 #include <errno.h>
 #include <linux/seg6_local.h>
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 /* Packet parsing state machine helpers. */
 #define cursor_advance(_cursor, _len) \
diff --git a/tools/testing/selftests/bpf/progs/test_map_in_map.c b/tools/testing/selftests/bpf/progs/test_map_in_map.c
index 113226115365..1cfeb940cf9f 100644
--- a/tools/testing/selftests/bpf/progs/test_map_in_map.c
+++ b/tools/testing/selftests/bpf/progs/test_map_in_map.c
@@ -3,7 +3,7 @@
 #include <stddef.h>
 #include <linux/bpf.h>
 #include <linux/types.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 struct {
 	__uint(type, BPF_MAP_TYPE_ARRAY_OF_MAPS);
diff --git a/tools/testing/selftests/bpf/progs/test_map_lock.c b/tools/testing/selftests/bpf/progs/test_map_lock.c
index bb7ce35f691b..b5c07ae7b68f 100644
--- a/tools/testing/selftests/bpf/progs/test_map_lock.c
+++ b/tools/testing/selftests/bpf/progs/test_map_lock.c
@@ -2,7 +2,7 @@
 // Copyright (c) 2019 Facebook
 #include <linux/bpf.h>
 #include <linux/version.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 #define VAR_NUM 16
 
diff --git a/tools/testing/selftests/bpf/progs/test_mmap.c b/tools/testing/selftests/bpf/progs/test_mmap.c
index e808791b7047..6239596cd14e 100644
--- a/tools/testing/selftests/bpf/progs/test_mmap.c
+++ b/tools/testing/selftests/bpf/progs/test_mmap.c
@@ -3,7 +3,7 @@
 
 #include <linux/bpf.h>
 #include <stdint.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 char _license[] SEC("license") = "GPL";
 
diff --git a/tools/testing/selftests/bpf/progs/test_obj_id.c b/tools/testing/selftests/bpf/progs/test_obj_id.c
index 3d30c02bdae9..98b9de2fafd0 100644
--- a/tools/testing/selftests/bpf/progs/test_obj_id.c
+++ b/tools/testing/selftests/bpf/progs/test_obj_id.c
@@ -4,7 +4,7 @@
 #include <stddef.h>
 #include <linux/bpf.h>
 #include <linux/pkt_cls.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 /* It is a dumb bpf program such that it must have no
  * issue to be loaded since testing the verifier is
diff --git a/tools/testing/selftests/bpf/progs/test_overhead.c b/tools/testing/selftests/bpf/progs/test_overhead.c
index 48748297b860..bfe9fbcb9684 100644
--- a/tools/testing/selftests/bpf/progs/test_overhead.c
+++ b/tools/testing/selftests/bpf/progs/test_overhead.c
@@ -4,8 +4,8 @@
 #include <stddef.h>
 #include <linux/bpf.h>
 #include <linux/ptrace.h>
-#include "bpf_helpers.h"
-#include "bpf_tracing.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
 #include "bpf_trace_helpers.h"
 
 struct task_struct;
diff --git a/tools/testing/selftests/bpf/progs/test_perf_buffer.c b/tools/testing/selftests/bpf/progs/test_perf_buffer.c
index 1fdc999031ac..ebfcc9f50c35 100644
--- a/tools/testing/selftests/bpf/progs/test_perf_buffer.c
+++ b/tools/testing/selftests/bpf/progs/test_perf_buffer.c
@@ -3,7 +3,7 @@
 
 #include <linux/ptrace.h>
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #include "bpf_trace_helpers.h"
 
 struct {
diff --git a/tools/testing/selftests/bpf/progs/test_pinning.c b/tools/testing/selftests/bpf/progs/test_pinning.c
index f20e7e00373f..4ef2630292b2 100644
--- a/tools/testing/selftests/bpf/progs/test_pinning.c
+++ b/tools/testing/selftests/bpf/progs/test_pinning.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 int _version SEC("version") = 1;
 
diff --git a/tools/testing/selftests/bpf/progs/test_pinning_invalid.c b/tools/testing/selftests/bpf/progs/test_pinning_invalid.c
index 51b38abe7ba1..5412e0c732c7 100644
--- a/tools/testing/selftests/bpf/progs/test_pinning_invalid.c
+++ b/tools/testing/selftests/bpf/progs/test_pinning_invalid.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 int _version SEC("version") = 1;
 
diff --git a/tools/testing/selftests/bpf/progs/test_pkt_access.c b/tools/testing/selftests/bpf/progs/test_pkt_access.c
index b77cebf71e66..7c9fcfd2b463 100644
--- a/tools/testing/selftests/bpf/progs/test_pkt_access.c
+++ b/tools/testing/selftests/bpf/progs/test_pkt_access.c
@@ -11,8 +11,8 @@
 #include <linux/in.h>
 #include <linux/tcp.h>
 #include <linux/pkt_cls.h>
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 #define barrier() __asm__ __volatile__("": : :"memory")
 int _version SEC("version") = 1;
diff --git a/tools/testing/selftests/bpf/progs/test_pkt_md_access.c b/tools/testing/selftests/bpf/progs/test_pkt_md_access.c
index 1db2623021ad..610c74ea9f64 100644
--- a/tools/testing/selftests/bpf/progs/test_pkt_md_access.c
+++ b/tools/testing/selftests/bpf/progs/test_pkt_md_access.c
@@ -5,7 +5,7 @@
 #include <string.h>
 #include <linux/bpf.h>
 #include <linux/pkt_cls.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 int _version SEC("version") = 1;
 
diff --git a/tools/testing/selftests/bpf/progs/test_probe_user.c b/tools/testing/selftests/bpf/progs/test_probe_user.c
index 5b570969e5c5..d556b1572cc6 100644
--- a/tools/testing/selftests/bpf/progs/test_probe_user.c
+++ b/tools/testing/selftests/bpf/progs/test_probe_user.c
@@ -5,8 +5,8 @@
 
 #include <netinet/in.h>
 
-#include "bpf_helpers.h"
-#include "bpf_tracing.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
 #include "bpf_trace_helpers.h"
 
 static struct sockaddr_in old;
diff --git a/tools/testing/selftests/bpf/progs/test_queue_stack_map.h b/tools/testing/selftests/bpf/progs/test_queue_stack_map.h
index 0e014d3b2b36..4dd9806ad73b 100644
--- a/tools/testing/selftests/bpf/progs/test_queue_stack_map.h
+++ b/tools/testing/selftests/bpf/progs/test_queue_stack_map.h
@@ -6,7 +6,7 @@
 #include <linux/if_ether.h>
 #include <linux/ip.h>
 #include <linux/pkt_cls.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 int _version SEC("version") = 1;
 
diff --git a/tools/testing/selftests/bpf/progs/test_rdonly_maps.c b/tools/testing/selftests/bpf/progs/test_rdonly_maps.c
index 52d94e8b214d..ecbeea2df259 100644
--- a/tools/testing/selftests/bpf/progs/test_rdonly_maps.c
+++ b/tools/testing/selftests/bpf/progs/test_rdonly_maps.c
@@ -3,7 +3,7 @@
 
 #include <linux/ptrace.h>
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 static volatile const struct {
 	unsigned a[4];
diff --git a/tools/testing/selftests/bpf/progs/test_seg6_loop.c b/tools/testing/selftests/bpf/progs/test_seg6_loop.c
index 69880c1e7700..a7278f064368 100644
--- a/tools/testing/selftests/bpf/progs/test_seg6_loop.c
+++ b/tools/testing/selftests/bpf/progs/test_seg6_loop.c
@@ -3,8 +3,8 @@
 #include <errno.h>
 #include <linux/seg6_local.h>
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 /* Packet parsing state machine helpers. */
 #define cursor_advance(_cursor, _len) \
diff --git a/tools/testing/selftests/bpf/progs/test_select_reuseport_kern.c b/tools/testing/selftests/bpf/progs/test_select_reuseport_kern.c
index b1f09f5bb1cf..d69a1f2bbbfd 100644
--- a/tools/testing/selftests/bpf/progs/test_select_reuseport_kern.c
+++ b/tools/testing/selftests/bpf/progs/test_select_reuseport_kern.c
@@ -11,8 +11,8 @@
 #include <linux/types.h>
 #include <linux/if_ether.h>
 
-#include "bpf_endian.h"
-#include "bpf_helpers.h"
+#include <bpf/bpf_endian.h>
+#include <bpf/bpf_helpers.h>
 #include "test_select_reuseport_common.h"
 
 int _version SEC("version") = 1;
diff --git a/tools/testing/selftests/bpf/progs/test_send_signal_kern.c b/tools/testing/selftests/bpf/progs/test_send_signal_kern.c
index 726733f9f20b..1acc91e87bfc 100644
--- a/tools/testing/selftests/bpf/progs/test_send_signal_kern.c
+++ b/tools/testing/selftests/bpf/progs/test_send_signal_kern.c
@@ -2,7 +2,7 @@
 // Copyright (c) 2019 Facebook
 #include <linux/bpf.h>
 #include <linux/version.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 __u32 sig = 0, pid = 0, status = 0, signal_thread = 0;
 
diff --git a/tools/testing/selftests/bpf/progs/test_sk_lookup_kern.c b/tools/testing/selftests/bpf/progs/test_sk_lookup_kern.c
index cb49ccb707d1..d2b38fa6a5b0 100644
--- a/tools/testing/selftests/bpf/progs/test_sk_lookup_kern.c
+++ b/tools/testing/selftests/bpf/progs/test_sk_lookup_kern.c
@@ -12,8 +12,8 @@
 #include <linux/pkt_cls.h>
 #include <linux/tcp.h>
 #include <sys/socket.h>
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 int _version SEC("version") = 1;
 char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/test_skb_cgroup_id_kern.c b/tools/testing/selftests/bpf/progs/test_skb_cgroup_id_kern.c
index 68cf9829f5a7..552f2090665c 100644
--- a/tools/testing/selftests/bpf/progs/test_skb_cgroup_id_kern.c
+++ b/tools/testing/selftests/bpf/progs/test_skb_cgroup_id_kern.c
@@ -6,7 +6,7 @@
 
 #include <string.h>
 
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 #define NUM_CGROUP_LEVELS	4
 
diff --git a/tools/testing/selftests/bpf/progs/test_skb_ctx.c b/tools/testing/selftests/bpf/progs/test_skb_ctx.c
index e18da87fe84f..202de3938494 100644
--- a/tools/testing/selftests/bpf/progs/test_skb_ctx.c
+++ b/tools/testing/selftests/bpf/progs/test_skb_ctx.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 int _version SEC("version") = 1;
 char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/test_skeleton.c b/tools/testing/selftests/bpf/progs/test_skeleton.c
index 4f69aac5635f..de03a90f78ca 100644
--- a/tools/testing/selftests/bpf/progs/test_skeleton.c
+++ b/tools/testing/selftests/bpf/progs/test_skeleton.c
@@ -3,7 +3,7 @@
 
 #include <stdbool.h>
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 struct s {
 	int a;
diff --git a/tools/testing/selftests/bpf/progs/test_sock_fields_kern.c b/tools/testing/selftests/bpf/progs/test_sock_fields_kern.c
index a47b003623ef..9bcaa37f476a 100644
--- a/tools/testing/selftests/bpf/progs/test_sock_fields_kern.c
+++ b/tools/testing/selftests/bpf/progs/test_sock_fields_kern.c
@@ -5,8 +5,8 @@
 #include <netinet/in.h>
 #include <stdbool.h>
 
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 enum bpf_addr_array_idx {
 	ADDR_SRV_IDX,
diff --git a/tools/testing/selftests/bpf/progs/test_spin_lock.c b/tools/testing/selftests/bpf/progs/test_spin_lock.c
index a43b999c8da2..0d31a3b3505f 100644
--- a/tools/testing/selftests/bpf/progs/test_spin_lock.c
+++ b/tools/testing/selftests/bpf/progs/test_spin_lock.c
@@ -2,7 +2,7 @@
 // Copyright (c) 2019 Facebook
 #include <linux/bpf.h>
 #include <linux/version.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 struct hmap_elem {
 	volatile int cnt;
diff --git a/tools/testing/selftests/bpf/progs/test_stacktrace_build_id.c b/tools/testing/selftests/bpf/progs/test_stacktrace_build_id.c
index f5638e26865d..0cf0134631b4 100644
--- a/tools/testing/selftests/bpf/progs/test_stacktrace_build_id.c
+++ b/tools/testing/selftests/bpf/progs/test_stacktrace_build_id.c
@@ -2,7 +2,7 @@
 // Copyright (c) 2018 Facebook
 
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 #ifndef PERF_MAX_STACK_DEPTH
 #define PERF_MAX_STACK_DEPTH         127
diff --git a/tools/testing/selftests/bpf/progs/test_stacktrace_map.c b/tools/testing/selftests/bpf/progs/test_stacktrace_map.c
index 3b7e1dca8829..00ed48672620 100644
--- a/tools/testing/selftests/bpf/progs/test_stacktrace_map.c
+++ b/tools/testing/selftests/bpf/progs/test_stacktrace_map.c
@@ -2,7 +2,7 @@
 // Copyright (c) 2018 Facebook
 
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 #ifndef PERF_MAX_STACK_DEPTH
 #define PERF_MAX_STACK_DEPTH         127
diff --git a/tools/testing/selftests/bpf/progs/test_sysctl_loop1.c b/tools/testing/selftests/bpf/progs/test_sysctl_loop1.c
index d22e438198cf..458b0d69133e 100644
--- a/tools/testing/selftests/bpf/progs/test_sysctl_loop1.c
+++ b/tools/testing/selftests/bpf/progs/test_sysctl_loop1.c
@@ -7,7 +7,7 @@
 #include <linux/stddef.h>
 #include <linux/bpf.h>
 
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 #ifndef ARRAY_SIZE
 #define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0]))
diff --git a/tools/testing/selftests/bpf/progs/test_sysctl_loop2.c b/tools/testing/selftests/bpf/progs/test_sysctl_loop2.c
index cb201cbe11e7..b2e6f9b0894d 100644
--- a/tools/testing/selftests/bpf/progs/test_sysctl_loop2.c
+++ b/tools/testing/selftests/bpf/progs/test_sysctl_loop2.c
@@ -7,7 +7,7 @@
 #include <linux/stddef.h>
 #include <linux/bpf.h>
 
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 #ifndef ARRAY_SIZE
 #define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0]))
diff --git a/tools/testing/selftests/bpf/progs/test_sysctl_prog.c b/tools/testing/selftests/bpf/progs/test_sysctl_prog.c
index 5cbbff416998..2d0b0b82a78a 100644
--- a/tools/testing/selftests/bpf/progs/test_sysctl_prog.c
+++ b/tools/testing/selftests/bpf/progs/test_sysctl_prog.c
@@ -7,7 +7,7 @@
 #include <linux/stddef.h>
 #include <linux/bpf.h>
 
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 /* Max supported length of a string with unsigned long in base 10 (pow2 - 1). */
 #define MAX_ULONG_STR_LEN 0xF
diff --git a/tools/testing/selftests/bpf/progs/test_tc_edt.c b/tools/testing/selftests/bpf/progs/test_tc_edt.c
index 0961415ba477..bf28814bfde5 100644
--- a/tools/testing/selftests/bpf/progs/test_tc_edt.c
+++ b/tools/testing/selftests/bpf/progs/test_tc_edt.c
@@ -7,8 +7,8 @@
 #include <linux/ip.h>
 #include <linux/pkt_cls.h>
 #include <linux/tcp.h>
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 /* the maximum delay we are willing to add (drop packets beyond that) */
 #define TIME_HORIZON_NS (2000 * 1000 * 1000)
diff --git a/tools/testing/selftests/bpf/progs/test_tc_tunnel.c b/tools/testing/selftests/bpf/progs/test_tc_tunnel.c
index 74370e7e286d..37bce7a7c394 100644
--- a/tools/testing/selftests/bpf/progs/test_tc_tunnel.c
+++ b/tools/testing/selftests/bpf/progs/test_tc_tunnel.c
@@ -17,8 +17,8 @@
 #include <linux/pkt_cls.h>
 #include <linux/types.h>
 
-#include "bpf_endian.h"
-#include "bpf_helpers.h"
+#include <bpf/bpf_endian.h>
+#include <bpf/bpf_helpers.h>
 
 static const int cfg_port = 8000;
 
diff --git a/tools/testing/selftests/bpf/progs/test_tcp_check_syncookie_kern.c b/tools/testing/selftests/bpf/progs/test_tcp_check_syncookie_kern.c
index d8803dfa8d32..47cbe2eeae43 100644
--- a/tools/testing/selftests/bpf/progs/test_tcp_check_syncookie_kern.c
+++ b/tools/testing/selftests/bpf/progs/test_tcp_check_syncookie_kern.c
@@ -13,8 +13,8 @@
 #include <sys/socket.h>
 #include <linux/tcp.h>
 
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 struct bpf_map_def SEC("maps") results = {
 	.type = BPF_MAP_TYPE_ARRAY,
diff --git a/tools/testing/selftests/bpf/progs/test_tcp_estats.c b/tools/testing/selftests/bpf/progs/test_tcp_estats.c
index 87b7d934ce73..adc83a54c352 100644
--- a/tools/testing/selftests/bpf/progs/test_tcp_estats.c
+++ b/tools/testing/selftests/bpf/progs/test_tcp_estats.c
@@ -36,7 +36,7 @@
 #include <linux/ipv6.h>
 #include <linux/version.h>
 #include <sys/socket.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 #define _(P) ({typeof(P) val = 0; bpf_probe_read_kernel(&val, sizeof(val), &P); val;})
 #define TCP_ESTATS_MAGIC 0xBAADBEEF
diff --git a/tools/testing/selftests/bpf/progs/test_tcpbpf_kern.c b/tools/testing/selftests/bpf/progs/test_tcpbpf_kern.c
index 7fa4595d2b66..1f1966e86e9f 100644
--- a/tools/testing/selftests/bpf/progs/test_tcpbpf_kern.c
+++ b/tools/testing/selftests/bpf/progs/test_tcpbpf_kern.c
@@ -10,8 +10,8 @@
 #include <linux/types.h>
 #include <linux/socket.h>
 #include <linux/tcp.h>
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 #include "test_tcpbpf.h"
 
 struct {
diff --git a/tools/testing/selftests/bpf/progs/test_tcpnotify_kern.c b/tools/testing/selftests/bpf/progs/test_tcpnotify_kern.c
index 08346e7765d5..ac63410bb541 100644
--- a/tools/testing/selftests/bpf/progs/test_tcpnotify_kern.c
+++ b/tools/testing/selftests/bpf/progs/test_tcpnotify_kern.c
@@ -10,8 +10,8 @@
 #include <linux/types.h>
 #include <linux/socket.h>
 #include <linux/tcp.h>
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 #include "test_tcpnotify.h"
 
 struct {
diff --git a/tools/testing/selftests/bpf/progs/test_tracepoint.c b/tools/testing/selftests/bpf/progs/test_tracepoint.c
index 04bf084517e0..4b825ee122cf 100644
--- a/tools/testing/selftests/bpf/progs/test_tracepoint.c
+++ b/tools/testing/selftests/bpf/progs/test_tracepoint.c
@@ -2,7 +2,7 @@
 // Copyright (c) 2017 Facebook
 
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 /* taken from /sys/kernel/debug/tracing/events/sched/sched_switch/format */
 struct sched_switch_args {
diff --git a/tools/testing/selftests/bpf/progs/test_tunnel_kern.c b/tools/testing/selftests/bpf/progs/test_tunnel_kern.c
index 504df69c83df..f48dbfe24ddc 100644
--- a/tools/testing/selftests/bpf/progs/test_tunnel_kern.c
+++ b/tools/testing/selftests/bpf/progs/test_tunnel_kern.c
@@ -19,8 +19,8 @@
 #include <linux/socket.h>
 #include <linux/pkt_cls.h>
 #include <linux/erspan.h>
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 #define ERROR(ret) do {\
 		char fmt[] = "ERROR line:%d ret:%d\n";\
diff --git a/tools/testing/selftests/bpf/progs/test_verif_scale1.c b/tools/testing/selftests/bpf/progs/test_verif_scale1.c
index f3236ce35f31..d38153dab3dd 100644
--- a/tools/testing/selftests/bpf/progs/test_verif_scale1.c
+++ b/tools/testing/selftests/bpf/progs/test_verif_scale1.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 // Copyright (c) 2019 Facebook
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #define ATTR __attribute__((noinline))
 #include "test_jhash.h"
 
diff --git a/tools/testing/selftests/bpf/progs/test_verif_scale2.c b/tools/testing/selftests/bpf/progs/test_verif_scale2.c
index 9897150ed516..f024154c7be7 100644
--- a/tools/testing/selftests/bpf/progs/test_verif_scale2.c
+++ b/tools/testing/selftests/bpf/progs/test_verif_scale2.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 // Copyright (c) 2019 Facebook
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #define ATTR __always_inline
 #include "test_jhash.h"
 
diff --git a/tools/testing/selftests/bpf/progs/test_verif_scale3.c b/tools/testing/selftests/bpf/progs/test_verif_scale3.c
index 1848da04ea41..9beb5bf80373 100644
--- a/tools/testing/selftests/bpf/progs/test_verif_scale3.c
+++ b/tools/testing/selftests/bpf/progs/test_verif_scale3.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 // Copyright (c) 2019 Facebook
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #define ATTR __attribute__((noinline))
 #include "test_jhash.h"
 
diff --git a/tools/testing/selftests/bpf/progs/test_xdp.c b/tools/testing/selftests/bpf/progs/test_xdp.c
index 0941c655b07b..31f9bce37491 100644
--- a/tools/testing/selftests/bpf/progs/test_xdp.c
+++ b/tools/testing/selftests/bpf/progs/test_xdp.c
@@ -16,8 +16,8 @@
 #include <linux/tcp.h>
 #include <linux/pkt_cls.h>
 #include <sys/socket.h>
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 #include "test_iptunnel_common.h"
 
 int _version SEC("version") = 1;
diff --git a/tools/testing/selftests/bpf/progs/test_xdp_bpf2bpf.c b/tools/testing/selftests/bpf/progs/test_xdp_bpf2bpf.c
index f8f105af6743..cb8a04ab7a78 100644
--- a/tools/testing/selftests/bpf/progs/test_xdp_bpf2bpf.c
+++ b/tools/testing/selftests/bpf/progs/test_xdp_bpf2bpf.c
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #include "bpf_trace_helpers.h"
 
 struct net_device {
diff --git a/tools/testing/selftests/bpf/progs/test_xdp_loop.c b/tools/testing/selftests/bpf/progs/test_xdp_loop.c
index 97175f73c3fe..fcabcda30ba3 100644
--- a/tools/testing/selftests/bpf/progs/test_xdp_loop.c
+++ b/tools/testing/selftests/bpf/progs/test_xdp_loop.c
@@ -12,8 +12,8 @@
 #include <linux/tcp.h>
 #include <linux/pkt_cls.h>
 #include <sys/socket.h>
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 #include "test_iptunnel_common.h"
 
 int _version SEC("version") = 1;
diff --git a/tools/testing/selftests/bpf/progs/test_xdp_meta.c b/tools/testing/selftests/bpf/progs/test_xdp_meta.c
index 8d0182650653..a7c4a7d49fe6 100644
--- a/tools/testing/selftests/bpf/progs/test_xdp_meta.c
+++ b/tools/testing/selftests/bpf/progs/test_xdp_meta.c
@@ -2,7 +2,7 @@
 #include <linux/if_ether.h>
 #include <linux/pkt_cls.h>
 
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 #define __round_mask(x, y) ((__typeof__(x))((y) - 1))
 #define round_up(x, y) ((((x) - 1) | __round_mask(x, y)) + 1)
diff --git a/tools/testing/selftests/bpf/progs/test_xdp_noinline.c b/tools/testing/selftests/bpf/progs/test_xdp_noinline.c
index f95bc1a17667..8beecec166d9 100644
--- a/tools/testing/selftests/bpf/progs/test_xdp_noinline.c
+++ b/tools/testing/selftests/bpf/progs/test_xdp_noinline.c
@@ -13,8 +13,8 @@
 #include <linux/icmpv6.h>
 #include <linux/tcp.h>
 #include <linux/udp.h>
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 static __u32 rol32(__u32 word, unsigned int shift)
 {
diff --git a/tools/testing/selftests/bpf/progs/test_xdp_redirect.c b/tools/testing/selftests/bpf/progs/test_xdp_redirect.c
index ef9e704be140..a5337cd9400b 100644
--- a/tools/testing/selftests/bpf/progs/test_xdp_redirect.c
+++ b/tools/testing/selftests/bpf/progs/test_xdp_redirect.c
@@ -10,7 +10,7 @@
  * General Public License for more details.
  */
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 int _version SEC("version") = 1;
 
diff --git a/tools/testing/selftests/bpf/progs/test_xdp_vlan.c b/tools/testing/selftests/bpf/progs/test_xdp_vlan.c
index 365a7d2d9f5c..134768f6b788 100644
--- a/tools/testing/selftests/bpf/progs/test_xdp_vlan.c
+++ b/tools/testing/selftests/bpf/progs/test_xdp_vlan.c
@@ -22,8 +22,8 @@
 #include <linux/in.h>
 #include <linux/pkt_cls.h>
 
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 /* linux/if_vlan.h have not exposed this as UAPI, thus mirror some here
  *
diff --git a/tools/testing/selftests/bpf/progs/xdp_dummy.c b/tools/testing/selftests/bpf/progs/xdp_dummy.c
index 43b0ef1001ed..ea25e8881992 100644
--- a/tools/testing/selftests/bpf/progs/xdp_dummy.c
+++ b/tools/testing/selftests/bpf/progs/xdp_dummy.c
@@ -2,7 +2,7 @@
 
 #define KBUILD_MODNAME "xdp_dummy"
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 SEC("xdp_dummy")
 int xdp_dummy_prog(struct xdp_md *ctx)
diff --git a/tools/testing/selftests/bpf/progs/xdp_redirect_map.c b/tools/testing/selftests/bpf/progs/xdp_redirect_map.c
index 1c5f298d7196..d037262c8937 100644
--- a/tools/testing/selftests/bpf/progs/xdp_redirect_map.c
+++ b/tools/testing/selftests/bpf/progs/xdp_redirect_map.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 struct {
 	__uint(type, BPF_MAP_TYPE_DEVMAP);
diff --git a/tools/testing/selftests/bpf/progs/xdp_tx.c b/tools/testing/selftests/bpf/progs/xdp_tx.c
index 57912e7c94b0..94e6c2b281cb 100644
--- a/tools/testing/selftests/bpf/progs/xdp_tx.c
+++ b/tools/testing/selftests/bpf/progs/xdp_tx.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 SEC("tx")
 int xdp_tx(struct xdp_md *xdp)
diff --git a/tools/testing/selftests/bpf/progs/xdping_kern.c b/tools/testing/selftests/bpf/progs/xdping_kern.c
index 112a2857f4e2..6b9ca40bd1f4 100644
--- a/tools/testing/selftests/bpf/progs/xdping_kern.c
+++ b/tools/testing/selftests/bpf/progs/xdping_kern.c
@@ -12,8 +12,8 @@
 #include <linux/if_vlan.h>
 #include <linux/ip.h>
 
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 #include "xdping.h"
 
diff --git a/tools/testing/selftests/bpf/test_cpp.cpp b/tools/testing/selftests/bpf/test_cpp.cpp
index 6fe23a10d48a..a8d2e9a87fbf 100644
--- a/tools/testing/selftests/bpf/test_cpp.cpp
+++ b/tools/testing/selftests/bpf/test_cpp.cpp
@@ -1,8 +1,8 @@
 /* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */
 #include <iostream>
-#include "libbpf.h"
-#include "bpf.h"
-#include "btf.h"
+#include <bpf/libbpf.h>
+#include <bpf/bpf.h>
+#include <bpf/btf.h>
 #include "test_core_extern.skel.h"
 
 /* do nothing, just make sure we can link successfully */
diff --git a/tools/testing/selftests/bpf/test_hashmap.c b/tools/testing/selftests/bpf/test_hashmap.c
index b64094c981e3..c490e012c23f 100644
--- a/tools/testing/selftests/bpf/test_hashmap.c
+++ b/tools/testing/selftests/bpf/test_hashmap.c
@@ -8,7 +8,7 @@
 #include <stdio.h>
 #include <errno.h>
 #include <linux/err.h>
-#include "hashmap.h"
+#include "bpf/hashmap.h"
 
 #define CHECK(condition, format...) ({					\
 	int __ret = !!(condition);					\
diff --git a/tools/testing/selftests/bpf/test_progs.h b/tools/testing/selftests/bpf/test_progs.h
index 99933a1857ca..bcfa9ef23fda 100644
--- a/tools/testing/selftests/bpf/test_progs.h
+++ b/tools/testing/selftests/bpf/test_progs.h
@@ -35,7 +35,7 @@ typedef __u16 __sum16;
 
 #include "test_iptunnel_common.h"
 #include "bpf_util.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_endian.h>
 #include "trace_helpers.h"
 #include "flow_dissector_load.h"
 
diff --git a/tools/testing/selftests/bpf/test_sock.c b/tools/testing/selftests/bpf/test_sock.c
index 0e6652733462..52bf14955797 100644
--- a/tools/testing/selftests/bpf/test_sock.c
+++ b/tools/testing/selftests/bpf/test_sock.c
@@ -13,7 +13,7 @@
 #include <bpf/bpf.h>
 
 #include "cgroup_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_endian.h>
 #include "bpf_rlimit.h"
 #include "bpf_util.h"
 
diff --git a/tools/testing/selftests/bpf/test_sockmap_kern.h b/tools/testing/selftests/bpf/test_sockmap_kern.h
index d008b41b7d8d..9b4d3a68a91a 100644
--- a/tools/testing/selftests/bpf/test_sockmap_kern.h
+++ b/tools/testing/selftests/bpf/test_sockmap_kern.h
@@ -12,8 +12,8 @@
 #include <linux/tcp.h>
 #include <linux/pkt_cls.h>
 #include <sys/socket.h>
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 /* Sockmap sample program connects a client and a backend together
  * using cgroups.
diff --git a/tools/testing/selftests/bpf/test_sysctl.c b/tools/testing/selftests/bpf/test_sysctl.c
index 40bd93a6e7ae..d196e2a4a6e0 100644
--- a/tools/testing/selftests/bpf/test_sysctl.c
+++ b/tools/testing/selftests/bpf/test_sysctl.c
@@ -13,7 +13,7 @@
 #include <bpf/bpf.h>
 #include <bpf/libbpf.h>
 
-#include "bpf_endian.h"
+#include <bpf/bpf_endian.h>
 #include "bpf_rlimit.h"
 #include "bpf_util.h"
 #include "cgroup_helpers.h"
diff --git a/tools/testing/selftests/bpf/trace_helpers.h b/tools/testing/selftests/bpf/trace_helpers.h
index aa4dcfe18050..0383c9b8adc1 100644
--- a/tools/testing/selftests/bpf/trace_helpers.h
+++ b/tools/testing/selftests/bpf/trace_helpers.h
@@ -2,7 +2,7 @@
 #ifndef __TRACE_HELPER_H
 #define __TRACE_HELPER_H
 
-#include <libbpf.h>
+#include <bpf/libbpf.h>
 
 struct ksym {
 	long addr;
-- 
cgit v1.2.3-59-g8ed1b


From 229c3b47b794e7257744224b21a95d3d9938d00a Mon Sep 17 00:00:00 2001
From: Toke Høiland-Jørgensen <toke@redhat.com>
Date: Mon, 20 Jan 2020 14:06:46 +0100
Subject: bpftool: Use consistent include paths for libbpf
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix bpftool to include libbpf header files with the bpf/ prefix, to be
consistent with external users of the library. Also ensure that all
includes of exported libbpf header files (those that are exported on 'make
install' of the library) use bracketed includes instead of quoted.

To make sure no new files are introduced that doesn't include the bpf/
prefix in its include, remove tools/lib/bpf from the include path entirely,
and use tools/lib instead.

Fixes: 6910d7d3867a ("selftests/bpf: Ensure bpf_helper_defs.h are taken from selftests dir")
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/157952560684.1683545.4765181397974997027.stgit@toke.dk
---
 tools/bpf/bpftool/Documentation/bpftool-gen.rst |  2 +-
 tools/bpf/bpftool/Makefile                      |  2 +-
 tools/bpf/bpftool/btf.c                         |  8 ++++----
 tools/bpf/bpftool/btf_dumper.c                  |  2 +-
 tools/bpf/bpftool/cgroup.c                      |  2 +-
 tools/bpf/bpftool/common.c                      |  4 ++--
 tools/bpf/bpftool/feature.c                     |  4 ++--
 tools/bpf/bpftool/gen.c                         | 10 +++++-----
 tools/bpf/bpftool/jit_disasm.c                  |  2 +-
 tools/bpf/bpftool/main.c                        |  4 ++--
 tools/bpf/bpftool/map.c                         |  4 ++--
 tools/bpf/bpftool/map_perf_ring.c               |  4 ++--
 tools/bpf/bpftool/net.c                         |  8 ++++----
 tools/bpf/bpftool/netlink_dumper.c              |  4 ++--
 tools/bpf/bpftool/perf.c                        |  2 +-
 tools/bpf/bpftool/prog.c                        |  6 +++---
 tools/bpf/bpftool/xlated_dumper.c               |  2 +-
 17 files changed, 35 insertions(+), 35 deletions(-)

diff --git a/tools/bpf/bpftool/Documentation/bpftool-gen.rst b/tools/bpf/bpftool/Documentation/bpftool-gen.rst
index 86a87da97d0b..94d91322895a 100644
--- a/tools/bpf/bpftool/Documentation/bpftool-gen.rst
+++ b/tools/bpf/bpftool/Documentation/bpftool-gen.rst
@@ -196,7 +196,7 @@ and global variables.
   #define __EXAMPLE_SKEL_H__
 
   #include <stdlib.h>
-  #include <libbpf.h>
+  #include <bpf/libbpf.h>
 
   struct example {
   	struct bpf_object_skeleton *skeleton;
diff --git a/tools/bpf/bpftool/Makefile b/tools/bpf/bpftool/Makefile
index 39bc6f0f4f0b..c4e810335810 100644
--- a/tools/bpf/bpftool/Makefile
+++ b/tools/bpf/bpftool/Makefile
@@ -45,7 +45,7 @@ CFLAGS += -DPACKAGE='"bpftool"' -D__EXPORTED_HEADERS__ \
 	-I$(srctree)/kernel/bpf/ \
 	-I$(srctree)/tools/include \
 	-I$(srctree)/tools/include/uapi \
-	-I$(srctree)/tools/lib/bpf \
+	-I$(srctree)/tools/lib \
 	-I$(srctree)/tools/perf
 CFLAGS += -DBPFTOOL_VERSION='"$(BPFTOOL_VERSION)"'
 ifneq ($(EXTRA_CFLAGS),)
diff --git a/tools/bpf/bpftool/btf.c b/tools/bpf/bpftool/btf.c
index 60c75be0666d..4ba90d81b6a1 100644
--- a/tools/bpf/bpftool/btf.c
+++ b/tools/bpf/bpftool/btf.c
@@ -8,15 +8,15 @@
 #include <stdio.h>
 #include <string.h>
 #include <unistd.h>
-#include <bpf.h>
-#include <libbpf.h>
+#include <bpf/bpf.h>
+#include <bpf/btf.h>
+#include <bpf/libbpf.h>
 #include <linux/btf.h>
 #include <linux/hashtable.h>
 #include <sys/types.h>
 #include <sys/stat.h>
 #include <unistd.h>
 
-#include "btf.h"
 #include "json_writer.h"
 #include "main.h"
 
@@ -532,7 +532,7 @@ static int do_dump(int argc, char **argv)
 		if (IS_ERR(btf)) {
 			err = PTR_ERR(btf);
 			btf = NULL;
-			p_err("failed to load BTF from %s: %s", 
+			p_err("failed to load BTF from %s: %s",
 			      *argv, strerror(err));
 			goto done;
 		}
diff --git a/tools/bpf/bpftool/btf_dumper.c b/tools/bpf/bpftool/btf_dumper.c
index d66131f69689..eb4a142016a0 100644
--- a/tools/bpf/bpftool/btf_dumper.c
+++ b/tools/bpf/bpftool/btf_dumper.c
@@ -8,8 +8,8 @@
 #include <linux/bitops.h>
 #include <linux/btf.h>
 #include <linux/err.h>
+#include <bpf/btf.h>
 
-#include "btf.h"
 #include "json_writer.h"
 #include "main.h"
 
diff --git a/tools/bpf/bpftool/cgroup.c b/tools/bpf/bpftool/cgroup.c
index 2f017caa678d..62c6a1d7cd18 100644
--- a/tools/bpf/bpftool/cgroup.c
+++ b/tools/bpf/bpftool/cgroup.c
@@ -14,7 +14,7 @@
 #include <sys/types.h>
 #include <unistd.h>
 
-#include <bpf.h>
+#include <bpf/bpf.h>
 
 #include "main.h"
 
diff --git a/tools/bpf/bpftool/common.c b/tools/bpf/bpftool/common.c
index 88264abaa738..b75b8ec5469c 100644
--- a/tools/bpf/bpftool/common.c
+++ b/tools/bpf/bpftool/common.c
@@ -20,8 +20,8 @@
 #include <sys/stat.h>
 #include <sys/vfs.h>
 
-#include <bpf.h>
-#include <libbpf.h> /* libbpf_num_possible_cpus */
+#include <bpf/bpf.h>
+#include <bpf/libbpf.h> /* libbpf_num_possible_cpus */
 
 #include "main.h"
 
diff --git a/tools/bpf/bpftool/feature.c b/tools/bpf/bpftool/feature.c
index 94735d968c34..446ba891f1e2 100644
--- a/tools/bpf/bpftool/feature.c
+++ b/tools/bpf/bpftool/feature.c
@@ -12,8 +12,8 @@
 #include <linux/filter.h>
 #include <linux/limits.h>
 
-#include <bpf.h>
-#include <libbpf.h>
+#include <bpf/bpf.h>
+#include <bpf/libbpf.h>
 #include <zlib.h>
 
 #include "main.h"
diff --git a/tools/bpf/bpftool/gen.c b/tools/bpf/bpftool/gen.c
index 7ce09a9a6999..f8113b3646f5 100644
--- a/tools/bpf/bpftool/gen.c
+++ b/tools/bpf/bpftool/gen.c
@@ -12,15 +12,15 @@
 #include <stdio.h>
 #include <string.h>
 #include <unistd.h>
-#include <bpf.h>
-#include <libbpf.h>
+#include <bpf/bpf.h>
+#include <bpf/libbpf.h>
 #include <sys/types.h>
 #include <sys/stat.h>
 #include <sys/mman.h>
 #include <unistd.h>
+#include <bpf/btf.h>
 
-#include "btf.h"
-#include "libbpf_internal.h"
+#include "bpf/libbpf_internal.h"
 #include "json_writer.h"
 #include "main.h"
 
@@ -333,7 +333,7 @@ static int do_skeleton(int argc, char **argv)
 		#define %2$s						    \n\
 									    \n\
 		#include <stdlib.h>					    \n\
-		#include <libbpf.h>					    \n\
+		#include <bpf/libbpf.h>					    \n\
 									    \n\
 		struct %1$s {						    \n\
 			struct bpf_object_skeleton *skeleton;		    \n\
diff --git a/tools/bpf/bpftool/jit_disasm.c b/tools/bpf/bpftool/jit_disasm.c
index bfed711258ce..f7f5885aa3ba 100644
--- a/tools/bpf/bpftool/jit_disasm.c
+++ b/tools/bpf/bpftool/jit_disasm.c
@@ -24,7 +24,7 @@
 #include <dis-asm.h>
 #include <sys/stat.h>
 #include <limits.h>
-#include <libbpf.h>
+#include <bpf/libbpf.h>
 
 #include "json_writer.h"
 #include "main.h"
diff --git a/tools/bpf/bpftool/main.c b/tools/bpf/bpftool/main.c
index 1fe91c558508..6d41bbfc6459 100644
--- a/tools/bpf/bpftool/main.c
+++ b/tools/bpf/bpftool/main.c
@@ -9,8 +9,8 @@
 #include <stdlib.h>
 #include <string.h>
 
-#include <bpf.h>
-#include <libbpf.h>
+#include <bpf/bpf.h>
+#include <bpf/libbpf.h>
 
 #include "main.h"
 
diff --git a/tools/bpf/bpftool/map.c b/tools/bpf/bpftool/map.c
index 86f8ab0b7e63..e6c85680b34d 100644
--- a/tools/bpf/bpftool/map.c
+++ b/tools/bpf/bpftool/map.c
@@ -15,9 +15,9 @@
 #include <sys/types.h>
 #include <sys/stat.h>
 
-#include <bpf.h>
+#include <bpf/bpf.h>
+#include <bpf/btf.h>
 
-#include "btf.h"
 #include "json_writer.h"
 #include "main.h"
 
diff --git a/tools/bpf/bpftool/map_perf_ring.c b/tools/bpf/bpftool/map_perf_ring.c
index 4c5531d1a450..d9b29c17fbb8 100644
--- a/tools/bpf/bpftool/map_perf_ring.c
+++ b/tools/bpf/bpftool/map_perf_ring.c
@@ -6,7 +6,7 @@
  */
 #include <errno.h>
 #include <fcntl.h>
-#include <libbpf.h>
+#include <bpf/libbpf.h>
 #include <poll.h>
 #include <signal.h>
 #include <stdbool.h>
@@ -21,7 +21,7 @@
 #include <sys/mman.h>
 #include <sys/syscall.h>
 
-#include <bpf.h>
+#include <bpf/bpf.h>
 #include <perf-sys.h>
 
 #include "main.h"
diff --git a/tools/bpf/bpftool/net.c b/tools/bpf/bpftool/net.c
index d93bee298e54..c5e3895b7c8b 100644
--- a/tools/bpf/bpftool/net.c
+++ b/tools/bpf/bpftool/net.c
@@ -7,7 +7,8 @@
 #include <stdlib.h>
 #include <string.h>
 #include <unistd.h>
-#include <libbpf.h>
+#include <bpf/bpf.h>
+#include <bpf/libbpf.h>
 #include <net/if.h>
 #include <linux/if.h>
 #include <linux/rtnetlink.h>
@@ -16,9 +17,8 @@
 #include <sys/stat.h>
 #include <sys/types.h>
 
-#include <bpf.h>
-#include <nlattr.h>
-#include "libbpf_internal.h"
+#include "bpf/nlattr.h"
+#include "bpf/libbpf_internal.h"
 #include "main.h"
 #include "netlink_dumper.h"
 
diff --git a/tools/bpf/bpftool/netlink_dumper.c b/tools/bpf/bpftool/netlink_dumper.c
index 550a0f537eed..5f65140b003b 100644
--- a/tools/bpf/bpftool/netlink_dumper.c
+++ b/tools/bpf/bpftool/netlink_dumper.c
@@ -3,11 +3,11 @@
 
 #include <stdlib.h>
 #include <string.h>
-#include <libbpf.h>
+#include <bpf/libbpf.h>
 #include <linux/rtnetlink.h>
 #include <linux/tc_act/tc_bpf.h>
 
-#include <nlattr.h>
+#include "bpf/nlattr.h"
 #include "main.h"
 #include "netlink_dumper.h"
 
diff --git a/tools/bpf/bpftool/perf.c b/tools/bpf/bpftool/perf.c
index b2046f33e23f..3341aa14acda 100644
--- a/tools/bpf/bpftool/perf.c
+++ b/tools/bpf/bpftool/perf.c
@@ -13,7 +13,7 @@
 #include <unistd.h>
 #include <ftw.h>
 
-#include <bpf.h>
+#include <bpf/bpf.h>
 
 #include "main.h"
 
diff --git a/tools/bpf/bpftool/prog.c b/tools/bpf/bpftool/prog.c
index 47a61ac42dc0..a3521deca869 100644
--- a/tools/bpf/bpftool/prog.c
+++ b/tools/bpf/bpftool/prog.c
@@ -17,9 +17,9 @@
 #include <linux/err.h>
 #include <linux/sizes.h>
 
-#include <bpf.h>
-#include <btf.h>
-#include <libbpf.h>
+#include <bpf/bpf.h>
+#include <bpf/btf.h>
+#include <bpf/libbpf.h>
 
 #include "cfg.h"
 #include "main.h"
diff --git a/tools/bpf/bpftool/xlated_dumper.c b/tools/bpf/bpftool/xlated_dumper.c
index 5b91ee65a080..8608cd68cdd0 100644
--- a/tools/bpf/bpftool/xlated_dumper.c
+++ b/tools/bpf/bpftool/xlated_dumper.c
@@ -7,7 +7,7 @@
 #include <stdlib.h>
 #include <string.h>
 #include <sys/types.h>
-#include <libbpf.h>
+#include <bpf/libbpf.h>
 
 #include "disasm.h"
 #include "json_writer.h"
-- 
cgit v1.2.3-59-g8ed1b


From 521fe8bb5874963d5f6fd58d5c5ad80fbc9c6b1c Mon Sep 17 00:00:00 2001
From: Toke Høiland-Jørgensen <toke@redhat.com>
Date: Mon, 20 Jan 2020 14:06:48 +0100
Subject: perf: Use consistent include paths for libbpf
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix perf to include libbpf header files with the bpf/ prefix, to
be consistent with external users of the library.

Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/157952560797.1683545.7685921032671386301.stgit@toke.dk
---
 tools/perf/examples/bpf/5sec.c             | 2 +-
 tools/perf/examples/bpf/empty.c            | 2 +-
 tools/perf/examples/bpf/sys_enter_openat.c | 2 +-
 tools/perf/include/bpf/pid_filter.h        | 2 +-
 tools/perf/include/bpf/stdio.h             | 2 +-
 tools/perf/include/bpf/unistd.h            | 2 +-
 6 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/tools/perf/examples/bpf/5sec.c b/tools/perf/examples/bpf/5sec.c
index b9c203219691..49f4f84da485 100644
--- a/tools/perf/examples/bpf/5sec.c
+++ b/tools/perf/examples/bpf/5sec.c
@@ -39,7 +39,7 @@
    Copyright (C) 2018 Red Hat, Inc., Arnaldo Carvalho de Melo <acme@redhat.com>
 */
 
-#include <bpf.h>
+#include <bpf/bpf.h>
 
 int probe(hrtimer_nanosleep, rqtp->tv_sec)(void *ctx, int err, long sec)
 {
diff --git a/tools/perf/examples/bpf/empty.c b/tools/perf/examples/bpf/empty.c
index 3776d26db9e7..7d7fb0c9fe76 100644
--- a/tools/perf/examples/bpf/empty.c
+++ b/tools/perf/examples/bpf/empty.c
@@ -1,3 +1,3 @@
-#include <bpf.h>
+#include <bpf/bpf.h>
 
 license(GPL);
diff --git a/tools/perf/examples/bpf/sys_enter_openat.c b/tools/perf/examples/bpf/sys_enter_openat.c
index 9cd124b09392..c4481c390d23 100644
--- a/tools/perf/examples/bpf/sys_enter_openat.c
+++ b/tools/perf/examples/bpf/sys_enter_openat.c
@@ -14,7 +14,7 @@
  * the return value.
  */
 
-#include <bpf.h>
+#include <bpf/bpf.h>
 
 struct syscall_enter_openat_args {
 	unsigned long long unused;
diff --git a/tools/perf/include/bpf/pid_filter.h b/tools/perf/include/bpf/pid_filter.h
index 6e61c4bdf548..607189a315b2 100644
--- a/tools/perf/include/bpf/pid_filter.h
+++ b/tools/perf/include/bpf/pid_filter.h
@@ -3,7 +3,7 @@
 #ifndef _PERF_BPF_PID_FILTER_
 #define _PERF_BPF_PID_FILTER_
 
-#include <bpf.h>
+#include <bpf/bpf.h>
 
 #define pid_filter(name) pid_map(name, bool)
 
diff --git a/tools/perf/include/bpf/stdio.h b/tools/perf/include/bpf/stdio.h
index 316af5b2ff35..7ca6fa5463ee 100644
--- a/tools/perf/include/bpf/stdio.h
+++ b/tools/perf/include/bpf/stdio.h
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0
 
-#include <bpf.h>
+#include <bpf/bpf.h>
 
 struct bpf_map SEC("maps") __bpf_stdout__ = {
        .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY,
diff --git a/tools/perf/include/bpf/unistd.h b/tools/perf/include/bpf/unistd.h
index ca7877f9a976..d1a35b6c649d 100644
--- a/tools/perf/include/bpf/unistd.h
+++ b/tools/perf/include/bpf/unistd.h
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: LGPL-2.1
 
-#include <bpf.h>
+#include <bpf/bpf.h>
 
 static int (*bpf_get_current_pid_tgid)(void) = (void *)BPF_FUNC_get_current_pid_tgid;
 
-- 
cgit v1.2.3-59-g8ed1b


From 7cf245a37ef013b2c1c5ca7ae25061de2ba7ad01 Mon Sep 17 00:00:00 2001
From: Toke Høiland-Jørgensen <toke@redhat.com>
Date: Mon, 20 Jan 2020 14:06:49 +0100
Subject: samples/bpf: Use consistent include paths for libbpf
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix all files in samples/bpf to include libbpf header files with the bpf/
prefix, to be consistent with external users of the library. Also ensure
that all includes of exported libbpf header files (those that are exported
on 'make install' of the library) use bracketed includes instead of quoted.

To make sure no new files are introduced that doesn't include the bpf/
prefix in its include, remove tools/lib/bpf from the include path entirely,
and use tools/lib instead.

Fixes: 6910d7d3867a ("selftests/bpf: Ensure bpf_helper_defs.h are taken from selftests dir")
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/157952560911.1683545.8795966751309534150.stgit@toke.dk
---
 samples/bpf/Makefile                              | 3 +--
 samples/bpf/cpustat_kern.c                        | 2 +-
 samples/bpf/fds_example.c                         | 2 +-
 samples/bpf/hbm.c                                 | 4 ++--
 samples/bpf/hbm_kern.h                            | 4 ++--
 samples/bpf/ibumad_kern.c                         | 2 +-
 samples/bpf/ibumad_user.c                         | 2 +-
 samples/bpf/lathist_kern.c                        | 2 +-
 samples/bpf/lwt_len_hist_kern.c                   | 2 +-
 samples/bpf/map_perf_test_kern.c                  | 4 ++--
 samples/bpf/offwaketime_kern.c                    | 4 ++--
 samples/bpf/offwaketime_user.c                    | 2 +-
 samples/bpf/parse_ldabs.c                         | 2 +-
 samples/bpf/parse_simple.c                        | 2 +-
 samples/bpf/parse_varlen.c                        | 2 +-
 samples/bpf/sampleip_kern.c                       | 4 ++--
 samples/bpf/sampleip_user.c                       | 2 +-
 samples/bpf/sock_flags_kern.c                     | 2 +-
 samples/bpf/sockex1_kern.c                        | 2 +-
 samples/bpf/sockex1_user.c                        | 2 +-
 samples/bpf/sockex2_kern.c                        | 2 +-
 samples/bpf/sockex2_user.c                        | 2 +-
 samples/bpf/sockex3_kern.c                        | 2 +-
 samples/bpf/spintest_kern.c                       | 4 ++--
 samples/bpf/spintest_user.c                       | 2 +-
 samples/bpf/syscall_tp_kern.c                     | 2 +-
 samples/bpf/task_fd_query_kern.c                  | 2 +-
 samples/bpf/task_fd_query_user.c                  | 2 +-
 samples/bpf/tc_l2_redirect_kern.c                 | 2 +-
 samples/bpf/tcbpf1_kern.c                         | 2 +-
 samples/bpf/tcp_basertt_kern.c                    | 4 ++--
 samples/bpf/tcp_bufs_kern.c                       | 4 ++--
 samples/bpf/tcp_clamp_kern.c                      | 4 ++--
 samples/bpf/tcp_cong_kern.c                       | 4 ++--
 samples/bpf/tcp_dumpstats_kern.c                  | 4 ++--
 samples/bpf/tcp_iw_kern.c                         | 4 ++--
 samples/bpf/tcp_rwnd_kern.c                       | 4 ++--
 samples/bpf/tcp_synrto_kern.c                     | 4 ++--
 samples/bpf/tcp_tos_reflect_kern.c                | 4 ++--
 samples/bpf/test_cgrp2_tc_kern.c                  | 2 +-
 samples/bpf/test_current_task_under_cgroup_kern.c | 2 +-
 samples/bpf/test_lwt_bpf.c                        | 2 +-
 samples/bpf/test_map_in_map_kern.c                | 4 ++--
 samples/bpf/test_overhead_kprobe_kern.c           | 4 ++--
 samples/bpf/test_overhead_raw_tp_kern.c           | 2 +-
 samples/bpf/test_overhead_tp_kern.c               | 2 +-
 samples/bpf/test_probe_write_user_kern.c          | 4 ++--
 samples/bpf/trace_event_kern.c                    | 4 ++--
 samples/bpf/trace_event_user.c                    | 2 +-
 samples/bpf/trace_output_kern.c                   | 2 +-
 samples/bpf/trace_output_user.c                   | 2 +-
 samples/bpf/tracex1_kern.c                        | 4 ++--
 samples/bpf/tracex2_kern.c                        | 4 ++--
 samples/bpf/tracex3_kern.c                        | 4 ++--
 samples/bpf/tracex4_kern.c                        | 4 ++--
 samples/bpf/tracex5_kern.c                        | 4 ++--
 samples/bpf/tracex6_kern.c                        | 2 +-
 samples/bpf/tracex7_kern.c                        | 2 +-
 samples/bpf/xdp1_kern.c                           | 2 +-
 samples/bpf/xdp1_user.c                           | 4 ++--
 samples/bpf/xdp2_kern.c                           | 2 +-
 samples/bpf/xdp2skb_meta_kern.c                   | 2 +-
 samples/bpf/xdp_adjust_tail_kern.c                | 2 +-
 samples/bpf/xdp_adjust_tail_user.c                | 4 ++--
 samples/bpf/xdp_fwd_kern.c                        | 2 +-
 samples/bpf/xdp_fwd_user.c                        | 2 +-
 samples/bpf/xdp_monitor_kern.c                    | 2 +-
 samples/bpf/xdp_redirect_cpu_kern.c               | 2 +-
 samples/bpf/xdp_redirect_cpu_user.c               | 2 +-
 samples/bpf/xdp_redirect_kern.c                   | 2 +-
 samples/bpf/xdp_redirect_map_kern.c               | 2 +-
 samples/bpf/xdp_redirect_map_user.c               | 2 +-
 samples/bpf/xdp_redirect_user.c                   | 2 +-
 samples/bpf/xdp_router_ipv4_kern.c                | 2 +-
 samples/bpf/xdp_router_ipv4_user.c                | 2 +-
 samples/bpf/xdp_rxq_info_kern.c                   | 2 +-
 samples/bpf/xdp_rxq_info_user.c                   | 4 ++--
 samples/bpf/xdp_sample_pkts_kern.c                | 2 +-
 samples/bpf/xdp_sample_pkts_user.c                | 2 +-
 samples/bpf/xdp_tx_iptunnel_kern.c                | 2 +-
 samples/bpf/xdp_tx_iptunnel_user.c                | 2 +-
 samples/bpf/xdpsock_kern.c                        | 2 +-
 samples/bpf/xdpsock_user.c                        | 6 +++---
 83 files changed, 112 insertions(+), 113 deletions(-)

diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile
index f86d713a17a5..b0e8adf7eb01 100644
--- a/samples/bpf/Makefile
+++ b/samples/bpf/Makefile
@@ -184,7 +184,6 @@ TPROGS_CFLAGS += -Wmissing-prototypes
 TPROGS_CFLAGS += -Wstrict-prototypes
 
 TPROGS_CFLAGS += -I$(objtree)/usr/include
-TPROGS_CFLAGS += -I$(srctree)/tools/lib/bpf/
 TPROGS_CFLAGS += -I$(srctree)/tools/testing/selftests/bpf/
 TPROGS_CFLAGS += -I$(srctree)/tools/lib/
 TPROGS_CFLAGS += -I$(srctree)/tools/include
@@ -305,7 +304,7 @@ $(obj)/%.o: $(src)/%.c
 	@echo "  CLANG-bpf " $@
 	$(Q)$(CLANG) $(NOSTDINC_FLAGS) $(LINUXINCLUDE) $(BPF_EXTRA_CFLAGS) \
 		-I$(obj) -I$(srctree)/tools/testing/selftests/bpf/ \
-		-I$(srctree)/tools/lib/bpf/ \
+		-I$(srctree)/tools/lib/ \
 		-D__KERNEL__ -D__BPF_TRACING__ -Wno-unused-value -Wno-pointer-sign \
 		-D__TARGET_ARCH_$(SRCARCH) -Wno-compare-distinct-pointer-types \
 		-Wno-gnu-variable-sized-type-not-at-end \
diff --git a/samples/bpf/cpustat_kern.c b/samples/bpf/cpustat_kern.c
index 68c84da065b1..a86a19d5f033 100644
--- a/samples/bpf/cpustat_kern.c
+++ b/samples/bpf/cpustat_kern.c
@@ -3,7 +3,7 @@
 #include <linux/version.h>
 #include <linux/ptrace.h>
 #include <uapi/linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 /*
  * The CPU number, cstate number and pstate number are based
diff --git a/samples/bpf/fds_example.c b/samples/bpf/fds_example.c
index 2d4b717726b6..d5992f787232 100644
--- a/samples/bpf/fds_example.c
+++ b/samples/bpf/fds_example.c
@@ -14,7 +14,7 @@
 
 #include <bpf/bpf.h>
 
-#include "libbpf.h"
+#include <bpf/libbpf.h>
 #include "bpf_insn.h"
 #include "sock_example.h"
 
diff --git a/samples/bpf/hbm.c b/samples/bpf/hbm.c
index 829b68d87687..7d7153777678 100644
--- a/samples/bpf/hbm.c
+++ b/samples/bpf/hbm.c
@@ -50,8 +50,8 @@
 #include "cgroup_helpers.h"
 #include "hbm.h"
 #include "bpf_util.h"
-#include "bpf.h"
-#include "libbpf.h"
+#include <bpf/bpf.h>
+#include <bpf/libbpf.h>
 
 bool outFlag = true;
 int minRate = 1000;		/* cgroup rate limit in Mbps */
diff --git a/samples/bpf/hbm_kern.h b/samples/bpf/hbm_kern.h
index 4edaf47876ca..e00f26f6afba 100644
--- a/samples/bpf/hbm_kern.h
+++ b/samples/bpf/hbm_kern.h
@@ -22,8 +22,8 @@
 #include <uapi/linux/pkt_cls.h>
 #include <net/ipv6.h>
 #include <net/inet_ecn.h>
-#include "bpf_endian.h"
-#include "bpf_helpers.h"
+#include <bpf/bpf_endian.h>
+#include <bpf/bpf_helpers.h>
 #include "hbm.h"
 
 #define DROP_PKT	0
diff --git a/samples/bpf/ibumad_kern.c b/samples/bpf/ibumad_kern.c
index f281df7e0089..3a91b4c1989a 100644
--- a/samples/bpf/ibumad_kern.c
+++ b/samples/bpf/ibumad_kern.c
@@ -13,7 +13,7 @@
 #define KBUILD_MODNAME "ibumad_count_pkts_by_class"
 #include <uapi/linux/bpf.h>
 
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 
 struct bpf_map_def SEC("maps") read_count = {
diff --git a/samples/bpf/ibumad_user.c b/samples/bpf/ibumad_user.c
index cb5a8f994849..fa06eef31a84 100644
--- a/samples/bpf/ibumad_user.c
+++ b/samples/bpf/ibumad_user.c
@@ -25,7 +25,7 @@
 
 #include "bpf_load.h"
 #include "bpf_util.h"
-#include "libbpf.h"
+#include <bpf/libbpf.h>
 
 static void dump_counts(int fd)
 {
diff --git a/samples/bpf/lathist_kern.c b/samples/bpf/lathist_kern.c
index 18fa088473cd..ca9c2e4e69aa 100644
--- a/samples/bpf/lathist_kern.c
+++ b/samples/bpf/lathist_kern.c
@@ -8,7 +8,7 @@
 #include <linux/version.h>
 #include <linux/ptrace.h>
 #include <uapi/linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 #define MAX_ENTRIES	20
 #define MAX_CPU		4
diff --git a/samples/bpf/lwt_len_hist_kern.c b/samples/bpf/lwt_len_hist_kern.c
index df75383280f9..9ed63e10e170 100644
--- a/samples/bpf/lwt_len_hist_kern.c
+++ b/samples/bpf/lwt_len_hist_kern.c
@@ -14,7 +14,7 @@
 #include <uapi/linux/if_ether.h>
 #include <uapi/linux/ip.h>
 #include <uapi/linux/in.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 # define printk(fmt, ...)						\
 		({							\
diff --git a/samples/bpf/map_perf_test_kern.c b/samples/bpf/map_perf_test_kern.c
index 281bcdaee58e..12e91ae64d4d 100644
--- a/samples/bpf/map_perf_test_kern.c
+++ b/samples/bpf/map_perf_test_kern.c
@@ -8,9 +8,9 @@
 #include <linux/netdevice.h>
 #include <linux/version.h>
 #include <uapi/linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #include "bpf_legacy.h"
-#include "bpf_tracing.h"
+#include <bpf/bpf_tracing.h>
 
 #define MAX_ENTRIES 1000
 #define MAX_NR_CPUS 1024
diff --git a/samples/bpf/offwaketime_kern.c b/samples/bpf/offwaketime_kern.c
index 9cb5207a692f..c4ec10dbfc3b 100644
--- a/samples/bpf/offwaketime_kern.c
+++ b/samples/bpf/offwaketime_kern.c
@@ -5,8 +5,8 @@
  * License as published by the Free Software Foundation.
  */
 #include <uapi/linux/bpf.h>
-#include "bpf_helpers.h"
-#include "bpf_tracing.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
 #include <uapi/linux/ptrace.h>
 #include <uapi/linux/perf_event.h>
 #include <linux/version.h>
diff --git a/samples/bpf/offwaketime_user.c b/samples/bpf/offwaketime_user.c
index fc8767d001f6..51c7da5341cc 100644
--- a/samples/bpf/offwaketime_user.c
+++ b/samples/bpf/offwaketime_user.c
@@ -12,7 +12,7 @@
 #include <assert.h>
 #include <stdbool.h>
 #include <sys/resource.h>
-#include "libbpf.h"
+#include <bpf/libbpf.h>
 #include "bpf_load.h"
 #include "trace_helpers.h"
 
diff --git a/samples/bpf/parse_ldabs.c b/samples/bpf/parse_ldabs.c
index ef5892377beb..c6f65f90a097 100644
--- a/samples/bpf/parse_ldabs.c
+++ b/samples/bpf/parse_ldabs.c
@@ -11,7 +11,7 @@
 #include <linux/tcp.h>
 #include <linux/udp.h>
 #include <uapi/linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #include "bpf_legacy.h"
 
 #define DEFAULT_PKTGEN_UDP_PORT	9
diff --git a/samples/bpf/parse_simple.c b/samples/bpf/parse_simple.c
index 10af53d33cc2..4a486cb1e0df 100644
--- a/samples/bpf/parse_simple.c
+++ b/samples/bpf/parse_simple.c
@@ -12,7 +12,7 @@
 #include <linux/udp.h>
 #include <uapi/linux/bpf.h>
 #include <net/ip.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 #define DEFAULT_PKTGEN_UDP_PORT 9
 
diff --git a/samples/bpf/parse_varlen.c b/samples/bpf/parse_varlen.c
index 0b6f22feb2c9..d8623846e810 100644
--- a/samples/bpf/parse_varlen.c
+++ b/samples/bpf/parse_varlen.c
@@ -14,7 +14,7 @@
 #include <linux/udp.h>
 #include <uapi/linux/bpf.h>
 #include <net/ip.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 #define DEFAULT_PKTGEN_UDP_PORT 9
 #define DEBUG 0
diff --git a/samples/bpf/sampleip_kern.c b/samples/bpf/sampleip_kern.c
index 4a190893894f..e504dc308371 100644
--- a/samples/bpf/sampleip_kern.c
+++ b/samples/bpf/sampleip_kern.c
@@ -8,8 +8,8 @@
 #include <linux/ptrace.h>
 #include <uapi/linux/bpf.h>
 #include <uapi/linux/bpf_perf_event.h>
-#include "bpf_helpers.h"
-#include "bpf_tracing.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
 
 #define MAX_IPS		8192
 
diff --git a/samples/bpf/sampleip_user.c b/samples/bpf/sampleip_user.c
index 6b5dc26d9701..b0f115f938bc 100644
--- a/samples/bpf/sampleip_user.c
+++ b/samples/bpf/sampleip_user.c
@@ -15,7 +15,7 @@
 #include <linux/ptrace.h>
 #include <linux/bpf.h>
 #include <sys/ioctl.h>
-#include "libbpf.h"
+#include <bpf/libbpf.h>
 #include "bpf_load.h"
 #include "perf-sys.h"
 #include "trace_helpers.h"
diff --git a/samples/bpf/sock_flags_kern.c b/samples/bpf/sock_flags_kern.c
index 05dcdf8a4baa..6d0ac7569d6f 100644
--- a/samples/bpf/sock_flags_kern.c
+++ b/samples/bpf/sock_flags_kern.c
@@ -3,7 +3,7 @@
 #include <linux/net.h>
 #include <uapi/linux/in.h>
 #include <uapi/linux/in6.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 SEC("cgroup/sock1")
 int bpf_prog1(struct bpf_sock *sk)
diff --git a/samples/bpf/sockex1_kern.c b/samples/bpf/sockex1_kern.c
index 2408dbfb7a21..431c956460ad 100644
--- a/samples/bpf/sockex1_kern.c
+++ b/samples/bpf/sockex1_kern.c
@@ -2,7 +2,7 @@
 #include <uapi/linux/if_ether.h>
 #include <uapi/linux/if_packet.h>
 #include <uapi/linux/ip.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #include "bpf_legacy.h"
 
 struct {
diff --git a/samples/bpf/sockex1_user.c b/samples/bpf/sockex1_user.c
index a219442afbee..3c83722877dc 100644
--- a/samples/bpf/sockex1_user.c
+++ b/samples/bpf/sockex1_user.c
@@ -3,7 +3,7 @@
 #include <assert.h>
 #include <linux/bpf.h>
 #include <bpf/bpf.h>
-#include "libbpf.h"
+#include <bpf/libbpf.h>
 #include "sock_example.h"
 #include <unistd.h>
 #include <arpa/inet.h>
diff --git a/samples/bpf/sockex2_kern.c b/samples/bpf/sockex2_kern.c
index a7bcd03bf529..a41dd520bc53 100644
--- a/samples/bpf/sockex2_kern.c
+++ b/samples/bpf/sockex2_kern.c
@@ -1,5 +1,5 @@
 #include <uapi/linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #include "bpf_legacy.h"
 #include <uapi/linux/in.h>
 #include <uapi/linux/if.h>
diff --git a/samples/bpf/sockex2_user.c b/samples/bpf/sockex2_user.c
index 6de383ddd08b..af925a5afd1d 100644
--- a/samples/bpf/sockex2_user.c
+++ b/samples/bpf/sockex2_user.c
@@ -3,7 +3,7 @@
 #include <assert.h>
 #include <linux/bpf.h>
 #include <bpf/bpf.h>
-#include "libbpf.h"
+#include <bpf/libbpf.h>
 #include "sock_example.h"
 #include <unistd.h>
 #include <arpa/inet.h>
diff --git a/samples/bpf/sockex3_kern.c b/samples/bpf/sockex3_kern.c
index 151dd842ecc0..36d4dac23549 100644
--- a/samples/bpf/sockex3_kern.c
+++ b/samples/bpf/sockex3_kern.c
@@ -5,7 +5,7 @@
  * License as published by the Free Software Foundation.
  */
 #include <uapi/linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #include "bpf_legacy.h"
 #include <uapi/linux/in.h>
 #include <uapi/linux/if.h>
diff --git a/samples/bpf/spintest_kern.c b/samples/bpf/spintest_kern.c
index 6e9478aa2938..f508af357251 100644
--- a/samples/bpf/spintest_kern.c
+++ b/samples/bpf/spintest_kern.c
@@ -9,8 +9,8 @@
 #include <linux/version.h>
 #include <uapi/linux/bpf.h>
 #include <uapi/linux/perf_event.h>
-#include "bpf_helpers.h"
-#include "bpf_tracing.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
 
 struct bpf_map_def SEC("maps") my_map = {
 	.type = BPF_MAP_TYPE_HASH,
diff --git a/samples/bpf/spintest_user.c b/samples/bpf/spintest_user.c
index 2556af2d9b3e..fb430ea2ef51 100644
--- a/samples/bpf/spintest_user.c
+++ b/samples/bpf/spintest_user.c
@@ -5,7 +5,7 @@
 #include <string.h>
 #include <assert.h>
 #include <sys/resource.h>
-#include "libbpf.h"
+#include <bpf/libbpf.h>
 #include "bpf_load.h"
 #include "trace_helpers.h"
 
diff --git a/samples/bpf/syscall_tp_kern.c b/samples/bpf/syscall_tp_kern.c
index 630ce8c4d5a2..5a62b03b1f88 100644
--- a/samples/bpf/syscall_tp_kern.c
+++ b/samples/bpf/syscall_tp_kern.c
@@ -2,7 +2,7 @@
 /* Copyright (c) 2017 Facebook
  */
 #include <uapi/linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 struct syscalls_enter_open_args {
 	unsigned long long unused;
diff --git a/samples/bpf/task_fd_query_kern.c b/samples/bpf/task_fd_query_kern.c
index fb56fc2a3e5d..278ade5427c8 100644
--- a/samples/bpf/task_fd_query_kern.c
+++ b/samples/bpf/task_fd_query_kern.c
@@ -2,7 +2,7 @@
 #include <linux/version.h>
 #include <linux/ptrace.h>
 #include <uapi/linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 SEC("kprobe/blk_mq_start_request")
 int bpf_prog1(struct pt_regs *ctx)
diff --git a/samples/bpf/task_fd_query_user.c b/samples/bpf/task_fd_query_user.c
index 4c31b305e6ef..ff2e9c1c7266 100644
--- a/samples/bpf/task_fd_query_user.c
+++ b/samples/bpf/task_fd_query_user.c
@@ -15,7 +15,7 @@
 #include <sys/stat.h>
 #include <linux/perf_event.h>
 
-#include "libbpf.h"
+#include <bpf/libbpf.h>
 #include "bpf_load.h"
 #include "bpf_util.h"
 #include "perf-sys.h"
diff --git a/samples/bpf/tc_l2_redirect_kern.c b/samples/bpf/tc_l2_redirect_kern.c
index 7ef2a12b25b2..fd2fa0004330 100644
--- a/samples/bpf/tc_l2_redirect_kern.c
+++ b/samples/bpf/tc_l2_redirect_kern.c
@@ -15,7 +15,7 @@
 #include <uapi/linux/filter.h>
 #include <uapi/linux/pkt_cls.h>
 #include <net/ipv6.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 #define _htonl __builtin_bswap32
 
diff --git a/samples/bpf/tcbpf1_kern.c b/samples/bpf/tcbpf1_kern.c
index ff43341bdfce..e9356130f84e 100644
--- a/samples/bpf/tcbpf1_kern.c
+++ b/samples/bpf/tcbpf1_kern.c
@@ -7,7 +7,7 @@
 #include <uapi/linux/tcp.h>
 #include <uapi/linux/filter.h>
 #include <uapi/linux/pkt_cls.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #include "bpf_legacy.h"
 
 /* compiler workaround */
diff --git a/samples/bpf/tcp_basertt_kern.c b/samples/bpf/tcp_basertt_kern.c
index 9dba48c2b920..8dfe09a92fec 100644
--- a/samples/bpf/tcp_basertt_kern.c
+++ b/samples/bpf/tcp_basertt_kern.c
@@ -16,8 +16,8 @@
 #include <uapi/linux/if_packet.h>
 #include <uapi/linux/ip.h>
 #include <linux/socket.h>
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 #define DEBUG 1
 
diff --git a/samples/bpf/tcp_bufs_kern.c b/samples/bpf/tcp_bufs_kern.c
index af8486f33771..6a80d08952ad 100644
--- a/samples/bpf/tcp_bufs_kern.c
+++ b/samples/bpf/tcp_bufs_kern.c
@@ -17,8 +17,8 @@
 #include <uapi/linux/if_packet.h>
 #include <uapi/linux/ip.h>
 #include <linux/socket.h>
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 #define DEBUG 1
 
diff --git a/samples/bpf/tcp_clamp_kern.c b/samples/bpf/tcp_clamp_kern.c
index 26c0fd091f3c..e88bd9ab0695 100644
--- a/samples/bpf/tcp_clamp_kern.c
+++ b/samples/bpf/tcp_clamp_kern.c
@@ -17,8 +17,8 @@
 #include <uapi/linux/if_packet.h>
 #include <uapi/linux/ip.h>
 #include <linux/socket.h>
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 #define DEBUG 1
 
diff --git a/samples/bpf/tcp_cong_kern.c b/samples/bpf/tcp_cong_kern.c
index 6d4dc4c7dd1e..2311fc9dde85 100644
--- a/samples/bpf/tcp_cong_kern.c
+++ b/samples/bpf/tcp_cong_kern.c
@@ -16,8 +16,8 @@
 #include <uapi/linux/if_packet.h>
 #include <uapi/linux/ip.h>
 #include <linux/socket.h>
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 #define DEBUG 1
 
diff --git a/samples/bpf/tcp_dumpstats_kern.c b/samples/bpf/tcp_dumpstats_kern.c
index 8557913106a0..e80d3afd24bd 100644
--- a/samples/bpf/tcp_dumpstats_kern.c
+++ b/samples/bpf/tcp_dumpstats_kern.c
@@ -4,8 +4,8 @@
  */
 #include <linux/bpf.h>
 
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 #define INTERVAL			1000000000ULL
 
diff --git a/samples/bpf/tcp_iw_kern.c b/samples/bpf/tcp_iw_kern.c
index da61d53378b3..d1444557358e 100644
--- a/samples/bpf/tcp_iw_kern.c
+++ b/samples/bpf/tcp_iw_kern.c
@@ -17,8 +17,8 @@
 #include <uapi/linux/if_packet.h>
 #include <uapi/linux/ip.h>
 #include <linux/socket.h>
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 #define DEBUG 1
 
diff --git a/samples/bpf/tcp_rwnd_kern.c b/samples/bpf/tcp_rwnd_kern.c
index d011e38b80d2..223d9c23b10c 100644
--- a/samples/bpf/tcp_rwnd_kern.c
+++ b/samples/bpf/tcp_rwnd_kern.c
@@ -16,8 +16,8 @@
 #include <uapi/linux/if_packet.h>
 #include <uapi/linux/ip.h>
 #include <linux/socket.h>
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 #define DEBUG 1
 
diff --git a/samples/bpf/tcp_synrto_kern.c b/samples/bpf/tcp_synrto_kern.c
index 720d1950322d..d58004eef124 100644
--- a/samples/bpf/tcp_synrto_kern.c
+++ b/samples/bpf/tcp_synrto_kern.c
@@ -16,8 +16,8 @@
 #include <uapi/linux/if_packet.h>
 #include <uapi/linux/ip.h>
 #include <linux/socket.h>
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 #define DEBUG 1
 
diff --git a/samples/bpf/tcp_tos_reflect_kern.c b/samples/bpf/tcp_tos_reflect_kern.c
index 369faca70a15..953fedc79ce1 100644
--- a/samples/bpf/tcp_tos_reflect_kern.c
+++ b/samples/bpf/tcp_tos_reflect_kern.c
@@ -15,8 +15,8 @@
 #include <uapi/linux/ipv6.h>
 #include <uapi/linux/in.h>
 #include <linux/socket.h>
-#include "bpf_helpers.h"
-#include "bpf_endian.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 
 #define DEBUG 1
 
diff --git a/samples/bpf/test_cgrp2_tc_kern.c b/samples/bpf/test_cgrp2_tc_kern.c
index 1547b36a7b7b..4dd532a312b9 100644
--- a/samples/bpf/test_cgrp2_tc_kern.c
+++ b/samples/bpf/test_cgrp2_tc_kern.c
@@ -10,7 +10,7 @@
 #include <uapi/linux/ipv6.h>
 #include <uapi/linux/pkt_cls.h>
 #include <uapi/linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 /* copy of 'struct ethhdr' without __packed */
 struct eth_hdr {
diff --git a/samples/bpf/test_current_task_under_cgroup_kern.c b/samples/bpf/test_current_task_under_cgroup_kern.c
index 86b28d7d6c99..6dc4f41bb6cb 100644
--- a/samples/bpf/test_current_task_under_cgroup_kern.c
+++ b/samples/bpf/test_current_task_under_cgroup_kern.c
@@ -8,7 +8,7 @@
 #include <linux/ptrace.h>
 #include <uapi/linux/bpf.h>
 #include <linux/version.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #include <uapi/linux/utsname.h>
 
 struct bpf_map_def SEC("maps") cgroup_map = {
diff --git a/samples/bpf/test_lwt_bpf.c b/samples/bpf/test_lwt_bpf.c
index bacc8013436b..1b568575ad11 100644
--- a/samples/bpf/test_lwt_bpf.c
+++ b/samples/bpf/test_lwt_bpf.c
@@ -20,7 +20,7 @@
 #include <linux/udp.h>
 #include <linux/icmpv6.h>
 #include <linux/if_ether.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #include <string.h>
 
 # define printk(fmt, ...)						\
diff --git a/samples/bpf/test_map_in_map_kern.c b/samples/bpf/test_map_in_map_kern.c
index 32ee752f19df..6cee61e8ce9b 100644
--- a/samples/bpf/test_map_in_map_kern.c
+++ b/samples/bpf/test_map_in_map_kern.c
@@ -10,9 +10,9 @@
 #include <linux/version.h>
 #include <uapi/linux/bpf.h>
 #include <uapi/linux/in6.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #include "bpf_legacy.h"
-#include "bpf_tracing.h"
+#include <bpf/bpf_tracing.h>
 
 #define MAX_NR_PORTS 65536
 
diff --git a/samples/bpf/test_overhead_kprobe_kern.c b/samples/bpf/test_overhead_kprobe_kern.c
index 8d2518e68db9..8b811c29dc79 100644
--- a/samples/bpf/test_overhead_kprobe_kern.c
+++ b/samples/bpf/test_overhead_kprobe_kern.c
@@ -7,8 +7,8 @@
 #include <linux/version.h>
 #include <linux/ptrace.h>
 #include <uapi/linux/bpf.h>
-#include "bpf_helpers.h"
-#include "bpf_tracing.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
 
 #define _(P) ({typeof(P) val = 0; bpf_probe_read(&val, sizeof(val), &P); val;})
 
diff --git a/samples/bpf/test_overhead_raw_tp_kern.c b/samples/bpf/test_overhead_raw_tp_kern.c
index d2af8bc1c805..8763181a32f3 100644
--- a/samples/bpf/test_overhead_raw_tp_kern.c
+++ b/samples/bpf/test_overhead_raw_tp_kern.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 /* Copyright (c) 2018 Facebook */
 #include <uapi/linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 SEC("raw_tracepoint/task_rename")
 int prog(struct bpf_raw_tracepoint_args *ctx)
diff --git a/samples/bpf/test_overhead_tp_kern.c b/samples/bpf/test_overhead_tp_kern.c
index 38f5c0b9da9f..eaa32693f8fc 100644
--- a/samples/bpf/test_overhead_tp_kern.c
+++ b/samples/bpf/test_overhead_tp_kern.c
@@ -5,7 +5,7 @@
  * License as published by the Free Software Foundation.
  */
 #include <uapi/linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 /* from /sys/kernel/debug/tracing/events/task/task_rename/format */
 struct task_rename {
diff --git a/samples/bpf/test_probe_write_user_kern.c b/samples/bpf/test_probe_write_user_kern.c
index b7c48f37132c..f033f36a13a3 100644
--- a/samples/bpf/test_probe_write_user_kern.c
+++ b/samples/bpf/test_probe_write_user_kern.c
@@ -8,8 +8,8 @@
 #include <linux/netdevice.h>
 #include <uapi/linux/bpf.h>
 #include <linux/version.h>
-#include "bpf_helpers.h"
-#include "bpf_tracing.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
 
 struct bpf_map_def SEC("maps") dnat_map = {
 	.type = BPF_MAP_TYPE_HASH,
diff --git a/samples/bpf/trace_event_kern.c b/samples/bpf/trace_event_kern.c
index 8dc18d233a27..da1d69e20645 100644
--- a/samples/bpf/trace_event_kern.c
+++ b/samples/bpf/trace_event_kern.c
@@ -9,8 +9,8 @@
 #include <uapi/linux/bpf.h>
 #include <uapi/linux/bpf_perf_event.h>
 #include <uapi/linux/perf_event.h>
-#include "bpf_helpers.h"
-#include "bpf_tracing.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
 
 struct key_t {
 	char comm[TASK_COMM_LEN];
diff --git a/samples/bpf/trace_event_user.c b/samples/bpf/trace_event_user.c
index 749a50f2f9f3..356171bc392b 100644
--- a/samples/bpf/trace_event_user.c
+++ b/samples/bpf/trace_event_user.c
@@ -15,7 +15,7 @@
 #include <assert.h>
 #include <errno.h>
 #include <sys/resource.h>
-#include "libbpf.h"
+#include <bpf/libbpf.h>
 #include "bpf_load.h"
 #include "perf-sys.h"
 #include "trace_helpers.h"
diff --git a/samples/bpf/trace_output_kern.c b/samples/bpf/trace_output_kern.c
index 9b96f4fb8cea..1d7d422cae6f 100644
--- a/samples/bpf/trace_output_kern.c
+++ b/samples/bpf/trace_output_kern.c
@@ -1,7 +1,7 @@
 #include <linux/ptrace.h>
 #include <linux/version.h>
 #include <uapi/linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 struct bpf_map_def SEC("maps") my_map = {
 	.type = BPF_MAP_TYPE_PERF_EVENT_ARRAY,
diff --git a/samples/bpf/trace_output_user.c b/samples/bpf/trace_output_user.c
index 8ee47699a870..60a17dd05345 100644
--- a/samples/bpf/trace_output_user.c
+++ b/samples/bpf/trace_output_user.c
@@ -15,7 +15,7 @@
 #include <sys/mman.h>
 #include <time.h>
 #include <signal.h>
-#include <libbpf.h>
+#include <bpf/libbpf.h>
 #include "bpf_load.h"
 #include "perf-sys.h"
 
diff --git a/samples/bpf/tracex1_kern.c b/samples/bpf/tracex1_kern.c
index 1a15f6605129..8e2610e14475 100644
--- a/samples/bpf/tracex1_kern.c
+++ b/samples/bpf/tracex1_kern.c
@@ -8,8 +8,8 @@
 #include <linux/netdevice.h>
 #include <uapi/linux/bpf.h>
 #include <linux/version.h>
-#include "bpf_helpers.h"
-#include "bpf_tracing.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
 
 #define _(P) ({typeof(P) val = 0; bpf_probe_read(&val, sizeof(val), &P); val;})
 
diff --git a/samples/bpf/tracex2_kern.c b/samples/bpf/tracex2_kern.c
index d70b3ea79ea7..d865bb309bcb 100644
--- a/samples/bpf/tracex2_kern.c
+++ b/samples/bpf/tracex2_kern.c
@@ -8,8 +8,8 @@
 #include <linux/netdevice.h>
 #include <linux/version.h>
 #include <uapi/linux/bpf.h>
-#include "bpf_helpers.h"
-#include "bpf_tracing.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
 
 struct bpf_map_def SEC("maps") my_map = {
 	.type = BPF_MAP_TYPE_HASH,
diff --git a/samples/bpf/tracex3_kern.c b/samples/bpf/tracex3_kern.c
index 9af546bebfa9..fe21c14feb8d 100644
--- a/samples/bpf/tracex3_kern.c
+++ b/samples/bpf/tracex3_kern.c
@@ -8,8 +8,8 @@
 #include <linux/netdevice.h>
 #include <linux/version.h>
 #include <uapi/linux/bpf.h>
-#include "bpf_helpers.h"
-#include "bpf_tracing.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
 
 struct bpf_map_def SEC("maps") my_map = {
 	.type = BPF_MAP_TYPE_HASH,
diff --git a/samples/bpf/tracex4_kern.c b/samples/bpf/tracex4_kern.c
index 2a02cbe9d9a1..b1bb9df88f8e 100644
--- a/samples/bpf/tracex4_kern.c
+++ b/samples/bpf/tracex4_kern.c
@@ -7,8 +7,8 @@
 #include <linux/ptrace.h>
 #include <linux/version.h>
 #include <uapi/linux/bpf.h>
-#include "bpf_helpers.h"
-#include "bpf_tracing.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
 
 struct pair {
 	u64 val;
diff --git a/samples/bpf/tracex5_kern.c b/samples/bpf/tracex5_kern.c
index b3557b21a8fe..481790fde864 100644
--- a/samples/bpf/tracex5_kern.c
+++ b/samples/bpf/tracex5_kern.c
@@ -10,8 +10,8 @@
 #include <uapi/linux/seccomp.h>
 #include <uapi/linux/unistd.h>
 #include "syscall_nrs.h"
-#include "bpf_helpers.h"
-#include "bpf_tracing.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
 
 #define PROG(F) SEC("kprobe/"__stringify(F)) int bpf_func_##F
 
diff --git a/samples/bpf/tracex6_kern.c b/samples/bpf/tracex6_kern.c
index 46c557afac73..96c234efa852 100644
--- a/samples/bpf/tracex6_kern.c
+++ b/samples/bpf/tracex6_kern.c
@@ -1,7 +1,7 @@
 #include <linux/ptrace.h>
 #include <linux/version.h>
 #include <uapi/linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 struct bpf_map_def SEC("maps") counters = {
 	.type = BPF_MAP_TYPE_PERF_EVENT_ARRAY,
diff --git a/samples/bpf/tracex7_kern.c b/samples/bpf/tracex7_kern.c
index 1ab308a43e0f..c5a92df8ac31 100644
--- a/samples/bpf/tracex7_kern.c
+++ b/samples/bpf/tracex7_kern.c
@@ -1,7 +1,7 @@
 #include <uapi/linux/ptrace.h>
 #include <uapi/linux/bpf.h>
 #include <linux/version.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 SEC("kprobe/open_ctree")
 int bpf_prog1(struct pt_regs *ctx)
diff --git a/samples/bpf/xdp1_kern.c b/samples/bpf/xdp1_kern.c
index db6870aee42c..34b64394ed9c 100644
--- a/samples/bpf/xdp1_kern.c
+++ b/samples/bpf/xdp1_kern.c
@@ -12,7 +12,7 @@
 #include <linux/if_vlan.h>
 #include <linux/ip.h>
 #include <linux/ipv6.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 struct {
 	__uint(type, BPF_MAP_TYPE_PERCPU_ARRAY);
diff --git a/samples/bpf/xdp1_user.c b/samples/bpf/xdp1_user.c
index 38a8852cb57f..c447ad9e3a1d 100644
--- a/samples/bpf/xdp1_user.c
+++ b/samples/bpf/xdp1_user.c
@@ -15,8 +15,8 @@
 #include <net/if.h>
 
 #include "bpf_util.h"
-#include "bpf.h"
-#include "libbpf.h"
+#include <bpf/bpf.h>
+#include <bpf/libbpf.h>
 
 static int ifindex;
 static __u32 xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
diff --git a/samples/bpf/xdp2_kern.c b/samples/bpf/xdp2_kern.c
index c74b52c6d945..c787f4b49646 100644
--- a/samples/bpf/xdp2_kern.c
+++ b/samples/bpf/xdp2_kern.c
@@ -12,7 +12,7 @@
 #include <linux/if_vlan.h>
 #include <linux/ip.h>
 #include <linux/ipv6.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 struct {
 	__uint(type, BPF_MAP_TYPE_PERCPU_ARRAY);
diff --git a/samples/bpf/xdp2skb_meta_kern.c b/samples/bpf/xdp2skb_meta_kern.c
index 0c12048ac79f..9b783316e860 100644
--- a/samples/bpf/xdp2skb_meta_kern.c
+++ b/samples/bpf/xdp2skb_meta_kern.c
@@ -12,7 +12,7 @@
 #include <uapi/linux/bpf.h>
 #include <uapi/linux/pkt_cls.h>
 
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 /*
  * This struct is stored in the XDP 'data_meta' area, which is located
diff --git a/samples/bpf/xdp_adjust_tail_kern.c b/samples/bpf/xdp_adjust_tail_kern.c
index 0f707e0fb375..ffdd548627f0 100644
--- a/samples/bpf/xdp_adjust_tail_kern.c
+++ b/samples/bpf/xdp_adjust_tail_kern.c
@@ -18,7 +18,7 @@
 #include <linux/if_vlan.h>
 #include <linux/ip.h>
 #include <linux/icmp.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 #define DEFAULT_TTL 64
 #define MAX_PCKT_SIZE 600
diff --git a/samples/bpf/xdp_adjust_tail_user.c b/samples/bpf/xdp_adjust_tail_user.c
index 008789eb6ada..ba482dc3da33 100644
--- a/samples/bpf/xdp_adjust_tail_user.c
+++ b/samples/bpf/xdp_adjust_tail_user.c
@@ -19,8 +19,8 @@
 #include <netinet/ether.h>
 #include <unistd.h>
 #include <time.h>
-#include "bpf.h"
-#include "libbpf.h"
+#include <bpf/bpf.h>
+#include <bpf/libbpf.h>
 
 #define STATS_INTERVAL_S 2U
 #define MAX_PCKT_SIZE 600
diff --git a/samples/bpf/xdp_fwd_kern.c b/samples/bpf/xdp_fwd_kern.c
index d013029aeaa2..54c099cbd639 100644
--- a/samples/bpf/xdp_fwd_kern.c
+++ b/samples/bpf/xdp_fwd_kern.c
@@ -19,7 +19,7 @@
 #include <linux/ip.h>
 #include <linux/ipv6.h>
 
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 #define IPV6_FLOWINFO_MASK              cpu_to_be32(0x0FFFFFFF)
 
diff --git a/samples/bpf/xdp_fwd_user.c b/samples/bpf/xdp_fwd_user.c
index c30f9acfdb84..74a4583d0d86 100644
--- a/samples/bpf/xdp_fwd_user.c
+++ b/samples/bpf/xdp_fwd_user.c
@@ -24,7 +24,7 @@
 #include <fcntl.h>
 #include <libgen.h>
 
-#include "libbpf.h"
+#include <bpf/libbpf.h>
 #include <bpf/bpf.h>
 
 static __u32 xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
diff --git a/samples/bpf/xdp_monitor_kern.c b/samples/bpf/xdp_monitor_kern.c
index 39458a44472e..3d33cca2d48a 100644
--- a/samples/bpf/xdp_monitor_kern.c
+++ b/samples/bpf/xdp_monitor_kern.c
@@ -4,7 +4,7 @@
  * XDP monitor tool, based on tracepoints
  */
 #include <uapi/linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 struct bpf_map_def SEC("maps") redirect_err_cnt = {
 	.type = BPF_MAP_TYPE_PERCPU_ARRAY,
diff --git a/samples/bpf/xdp_redirect_cpu_kern.c b/samples/bpf/xdp_redirect_cpu_kern.c
index cfcc31e51197..313a8fe6d125 100644
--- a/samples/bpf/xdp_redirect_cpu_kern.c
+++ b/samples/bpf/xdp_redirect_cpu_kern.c
@@ -12,7 +12,7 @@
 #include <uapi/linux/udp.h>
 
 #include <uapi/linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #include "hash_func01.h"
 
 #define MAX_CPUS 64 /* WARNING - sync with _user.c */
diff --git a/samples/bpf/xdp_redirect_cpu_user.c b/samples/bpf/xdp_redirect_cpu_user.c
index 79a2fb7d16cb..15bdf047a222 100644
--- a/samples/bpf/xdp_redirect_cpu_user.c
+++ b/samples/bpf/xdp_redirect_cpu_user.c
@@ -30,7 +30,7 @@ static const char *__doc__ =
 #define MAX_PROG 6
 
 #include <bpf/bpf.h>
-#include "libbpf.h"
+#include <bpf/libbpf.h>
 
 #include "bpf_util.h"
 
diff --git a/samples/bpf/xdp_redirect_kern.c b/samples/bpf/xdp_redirect_kern.c
index 1f0b7d05abb2..d26ec3aa215e 100644
--- a/samples/bpf/xdp_redirect_kern.c
+++ b/samples/bpf/xdp_redirect_kern.c
@@ -17,7 +17,7 @@
 #include <linux/if_vlan.h>
 #include <linux/ip.h>
 #include <linux/ipv6.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 struct {
 	__uint(type, BPF_MAP_TYPE_ARRAY);
diff --git a/samples/bpf/xdp_redirect_map_kern.c b/samples/bpf/xdp_redirect_map_kern.c
index 4631b484c432..6489352ab7a4 100644
--- a/samples/bpf/xdp_redirect_map_kern.c
+++ b/samples/bpf/xdp_redirect_map_kern.c
@@ -17,7 +17,7 @@
 #include <linux/if_vlan.h>
 #include <linux/ip.h>
 #include <linux/ipv6.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 struct {
 	__uint(type, BPF_MAP_TYPE_DEVMAP);
diff --git a/samples/bpf/xdp_redirect_map_user.c b/samples/bpf/xdp_redirect_map_user.c
index cc840661faab..35e16dee613e 100644
--- a/samples/bpf/xdp_redirect_map_user.c
+++ b/samples/bpf/xdp_redirect_map_user.c
@@ -17,7 +17,7 @@
 
 #include "bpf_util.h"
 #include <bpf/bpf.h>
-#include "libbpf.h"
+#include <bpf/libbpf.h>
 
 static int ifindex_in;
 static int ifindex_out;
diff --git a/samples/bpf/xdp_redirect_user.c b/samples/bpf/xdp_redirect_user.c
index 71dff8e3382a..9ca2bf457cda 100644
--- a/samples/bpf/xdp_redirect_user.c
+++ b/samples/bpf/xdp_redirect_user.c
@@ -17,7 +17,7 @@
 
 #include "bpf_util.h"
 #include <bpf/bpf.h>
-#include "libbpf.h"
+#include <bpf/libbpf.h>
 
 static int ifindex_in;
 static int ifindex_out;
diff --git a/samples/bpf/xdp_router_ipv4_kern.c b/samples/bpf/xdp_router_ipv4_kern.c
index bf11efc8e949..b37ca2b13063 100644
--- a/samples/bpf/xdp_router_ipv4_kern.c
+++ b/samples/bpf/xdp_router_ipv4_kern.c
@@ -12,7 +12,7 @@
 #include <linux/if_vlan.h>
 #include <linux/ip.h>
 #include <linux/ipv6.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #include <linux/slab.h>
 #include <net/ip_fib.h>
 
diff --git a/samples/bpf/xdp_router_ipv4_user.c b/samples/bpf/xdp_router_ipv4_user.c
index fef286c5add2..c2da1b51ff95 100644
--- a/samples/bpf/xdp_router_ipv4_user.c
+++ b/samples/bpf/xdp_router_ipv4_user.c
@@ -21,7 +21,7 @@
 #include <sys/ioctl.h>
 #include <sys/syscall.h>
 #include "bpf_util.h"
-#include "libbpf.h"
+#include <bpf/libbpf.h>
 #include <sys/resource.h>
 #include <libgen.h>
 
diff --git a/samples/bpf/xdp_rxq_info_kern.c b/samples/bpf/xdp_rxq_info_kern.c
index 272d0f82a6b5..5e7459f9bf3e 100644
--- a/samples/bpf/xdp_rxq_info_kern.c
+++ b/samples/bpf/xdp_rxq_info_kern.c
@@ -6,7 +6,7 @@
 #include <uapi/linux/bpf.h>
 #include <uapi/linux/if_ether.h>
 #include <uapi/linux/in.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 /* Config setup from with userspace
  *
diff --git a/samples/bpf/xdp_rxq_info_user.c b/samples/bpf/xdp_rxq_info_user.c
index fc4983fd6959..4fe47502ebed 100644
--- a/samples/bpf/xdp_rxq_info_user.c
+++ b/samples/bpf/xdp_rxq_info_user.c
@@ -22,8 +22,8 @@ static const char *__doc__ = " XDP RX-queue info extract example\n\n"
 #include <arpa/inet.h>
 #include <linux/if_link.h>
 
-#include "bpf.h"
-#include "libbpf.h"
+#include <bpf/bpf.h>
+#include <bpf/libbpf.h>
 #include "bpf_util.h"
 
 static int ifindex = -1;
diff --git a/samples/bpf/xdp_sample_pkts_kern.c b/samples/bpf/xdp_sample_pkts_kern.c
index 6c7c7e0aaeda..33377289e2a8 100644
--- a/samples/bpf/xdp_sample_pkts_kern.c
+++ b/samples/bpf/xdp_sample_pkts_kern.c
@@ -2,7 +2,7 @@
 #include <linux/ptrace.h>
 #include <linux/version.h>
 #include <uapi/linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 
 #define SAMPLE_SIZE 64ul
 #define MAX_CPUS 128
diff --git a/samples/bpf/xdp_sample_pkts_user.c b/samples/bpf/xdp_sample_pkts_user.c
index 8c1af1b7372d..991ef6f0880b 100644
--- a/samples/bpf/xdp_sample_pkts_user.c
+++ b/samples/bpf/xdp_sample_pkts_user.c
@@ -10,7 +10,7 @@
 #include <sys/sysinfo.h>
 #include <sys/ioctl.h>
 #include <signal.h>
-#include <libbpf.h>
+#include <bpf/libbpf.h>
 #include <bpf/bpf.h>
 #include <sys/resource.h>
 #include <libgen.h>
diff --git a/samples/bpf/xdp_tx_iptunnel_kern.c b/samples/bpf/xdp_tx_iptunnel_kern.c
index 6db450a5c1ca..575d57e4b8d6 100644
--- a/samples/bpf/xdp_tx_iptunnel_kern.c
+++ b/samples/bpf/xdp_tx_iptunnel_kern.c
@@ -16,7 +16,7 @@
 #include <linux/if_vlan.h>
 #include <linux/ip.h>
 #include <linux/ipv6.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #include "xdp_tx_iptunnel_common.h"
 
 struct {
diff --git a/samples/bpf/xdp_tx_iptunnel_user.c b/samples/bpf/xdp_tx_iptunnel_user.c
index 5f33b5530032..a419bee151a8 100644
--- a/samples/bpf/xdp_tx_iptunnel_user.c
+++ b/samples/bpf/xdp_tx_iptunnel_user.c
@@ -15,7 +15,7 @@
 #include <netinet/ether.h>
 #include <unistd.h>
 #include <time.h>
-#include "libbpf.h"
+#include <bpf/libbpf.h>
 #include <bpf/bpf.h>
 #include "bpf_util.h"
 #include "xdp_tx_iptunnel_common.h"
diff --git a/samples/bpf/xdpsock_kern.c b/samples/bpf/xdpsock_kern.c
index a06177c262cd..05430484375c 100644
--- a/samples/bpf/xdpsock_kern.c
+++ b/samples/bpf/xdpsock_kern.c
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <linux/bpf.h>
-#include "bpf_helpers.h"
+#include <bpf/bpf_helpers.h>
 #include "xdpsock.h"
 
 /* This XDP program is only needed for the XDP_SHARED_UMEM mode.
diff --git a/samples/bpf/xdpsock_user.c b/samples/bpf/xdpsock_user.c
index d74c4c83fc93..0b5acd722306 100644
--- a/samples/bpf/xdpsock_user.c
+++ b/samples/bpf/xdpsock_user.c
@@ -30,10 +30,10 @@
 #include <time.h>
 #include <unistd.h>
 
-#include "libbpf.h"
-#include "xsk.h"
-#include "xdpsock.h"
+#include <bpf/libbpf.h>
+#include <bpf/xsk.h>
 #include <bpf/bpf.h>
+#include "xdpsock.h"
 
 #ifndef SOL_XDP
 #define SOL_XDP 283
-- 
cgit v1.2.3-59-g8ed1b


From a9ed34c0b7a537e38564d92eb4ff9eee91c84544 Mon Sep 17 00:00:00 2001
From: Toke Høiland-Jørgensen <toke@redhat.com>
Date: Mon, 20 Jan 2020 14:06:50 +0100
Subject: tools/runqslower: Remove tools/lib/bpf from include path
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Since we are now consistently using the bpf/ prefix on #include directives,
we don't need to include tools/lib/bpf in the include path. Remove it to
make sure we don't inadvertently introduce new includes without the prefix.

Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/157952561027.1683545.1976265477926794138.stgit@toke.dk
---
 tools/bpf/runqslower/Makefile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/bpf/runqslower/Makefile b/tools/bpf/runqslower/Makefile
index b7b2230f807b..b90044caf270 100644
--- a/tools/bpf/runqslower/Makefile
+++ b/tools/bpf/runqslower/Makefile
@@ -6,7 +6,7 @@ LLVM_STRIP := llvm-strip
 DEFAULT_BPFTOOL := $(OUTPUT)/sbin/bpftool
 BPFTOOL ?= $(DEFAULT_BPFTOOL)
 LIBBPF_SRC := $(abspath ../../lib/bpf)
-INCLUDES := -I$(OUTPUT) -I$(abspath ../../lib) -I$(abspath ../../lib/bpf)
+INCLUDES := -I$(OUTPUT) -I$(abspath ../../lib)
 CFLAGS := -g -Wall
 
 # Try to detect best kernel BTF source
-- 
cgit v1.2.3-59-g8ed1b


From b6580cd8999d4f6846db9f6daa0b759c75aff33e Mon Sep 17 00:00:00 2001
From: Toke Høiland-Jørgensen <toke@redhat.com>
Date: Mon, 20 Jan 2020 14:06:51 +0100
Subject: runsqslower: Support user-specified libbpf include and object paths
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This adds support for specifying the libbpf include and object paths as
arguments to the runqslower Makefile, to support reusing the libbpf version
built as part of the selftests.

Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/157952561135.1683545.5660339645093141381.stgit@toke.dk
---
 tools/bpf/runqslower/Makefile | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/tools/bpf/runqslower/Makefile b/tools/bpf/runqslower/Makefile
index b90044caf270..faf5418609ea 100644
--- a/tools/bpf/runqslower/Makefile
+++ b/tools/bpf/runqslower/Makefile
@@ -6,7 +6,9 @@ LLVM_STRIP := llvm-strip
 DEFAULT_BPFTOOL := $(OUTPUT)/sbin/bpftool
 BPFTOOL ?= $(DEFAULT_BPFTOOL)
 LIBBPF_SRC := $(abspath ../../lib/bpf)
-INCLUDES := -I$(OUTPUT) -I$(abspath ../../lib)
+BPFOBJ := $(OUTPUT)/libbpf.a
+BPF_INCLUDE := $(OUTPUT)
+INCLUDES := -I$(BPF_INCLUDE) -I$(OUTPUT) -I$(abspath ../../lib)
 CFLAGS := -g -Wall
 
 # Try to detect best kernel BTF source
@@ -37,7 +39,7 @@ clean:
 	$(call msg,CLEAN)
 	$(Q)rm -rf $(OUTPUT) runqslower
 
-$(OUTPUT)/runqslower: $(OUTPUT)/runqslower.o $(OUTPUT)/libbpf.a
+$(OUTPUT)/runqslower: $(OUTPUT)/runqslower.o $(BPFOBJ)
 	$(call msg,BINARY,$@)
 	$(Q)$(CC) $(CFLAGS) -lelf -lz $^ -o $@
 
@@ -50,7 +52,7 @@ $(OUTPUT)/%.skel.h: $(OUTPUT)/%.bpf.o | $(BPFTOOL)
 	$(call msg,GEN-SKEL,$@)
 	$(Q)$(BPFTOOL) gen skeleton $< > $@
 
-$(OUTPUT)/%.bpf.o: %.bpf.c $(OUTPUT)/libbpf.a | $(OUTPUT)
+$(OUTPUT)/%.bpf.o: %.bpf.c $(BPFOBJ) | $(OUTPUT)
 	$(call msg,BPF,$@)
 	$(Q)$(CLANG) -g -O2 -target bpf $(INCLUDES)			      \
 		 -c $(filter %.c,$^) -o $@ &&				      \
@@ -73,9 +75,9 @@ $(OUTPUT)/vmlinux.h: $(VMLINUX_BTF_PATH) | $(OUTPUT) $(BPFTOOL)
 	fi
 	$(Q)$(BPFTOOL) btf dump file $(VMLINUX_BTF_PATH) format c > $@
 
-$(OUTPUT)/libbpf.a: | $(OUTPUT)
+$(BPFOBJ): | $(OUTPUT)
 	$(Q)$(MAKE) $(submake_extras) -C $(LIBBPF_SRC)			       \
-		    OUTPUT=$(abs_out)/ $(abs_out)/libbpf.a
+		    OUTPUT=$(abspath $(dir $@))/ $(abspath $@)
 
 $(DEFAULT_BPFTOOL):
 	$(Q)$(MAKE) $(submake_extras) -C ../bpftool			      \
-- 
cgit v1.2.3-59-g8ed1b


From a2c9652f751e3ab39738e2b30bc45293d53604fd Mon Sep 17 00:00:00 2001
From: Toke Høiland-Jørgensen <toke@redhat.com>
Date: Mon, 20 Jan 2020 14:06:52 +0100
Subject: selftests: Refactor build to remove tools/lib/bpf from include path
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

To make sure no new files are introduced that doesn't include the bpf/
prefix in its #include, remove tools/lib/bpf from the include path
entirely.

Instead, we introduce a new header files directory under the scratch tools/
dir, and add a rule to run the 'install_headers' rule from libbpf to have a
full set of consistent libbpf headers in $(OUTPUT)/tools/include/bpf, and
then use $(OUTPUT)/tools/include as the include path for selftests.

For consistency we also make sure we put all the scratch build files from
other bpftool and libbpf into tools/build/, so everything stays within
selftests/.

Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/157952561246.1683545.2762245552022369203.stgit@toke.dk
---
 tools/testing/selftests/bpf/.gitignore |  4 +--
 tools/testing/selftests/bpf/Makefile   | 59 +++++++++++++++++-----------------
 2 files changed, 30 insertions(+), 33 deletions(-)

diff --git a/tools/testing/selftests/bpf/.gitignore b/tools/testing/selftests/bpf/.gitignore
index 1d14e3ab70be..ec464859c6b6 100644
--- a/tools/testing/selftests/bpf/.gitignore
+++ b/tools/testing/selftests/bpf/.gitignore
@@ -29,8 +29,6 @@ test_tcpnotify_user
 test_libbpf
 test_tcp_check_syncookie_user
 test_sysctl
-libbpf.pc
-libbpf.so.*
 test_hashmap
 test_btf_dump
 xdping
@@ -39,4 +37,4 @@ test_cpp
 /no_alu32
 /bpf_gcc
 /tools
-bpf_helper_defs.h
+
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index 2cd91e0524cf..be8fe404a086 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -20,8 +20,8 @@ CLANG		?= clang
 LLC		?= llc
 LLVM_OBJCOPY	?= llvm-objcopy
 BPF_GCC		?= $(shell command -v bpf-gcc;)
-CFLAGS += -g -Wall -O2 $(GENFLAGS) -I$(CURDIR) -I$(APIDIR) -I$(LIBDIR)  \
-	  -I$(BPFDIR) -I$(GENDIR) -I$(TOOLSINCDIR)			\
+CFLAGS += -g -Wall -O2 $(GENFLAGS) -I$(CURDIR) -I$(APIDIR)		\
+	  -I$(INCLUDE_DIR) -I$(GENDIR) -I$(LIBDIR) -I$(TOOLSINCDIR)	\
 	  -Dbpf_prog_load=bpf_prog_test_load				\
 	  -Dbpf_load_program=bpf_test_load_program
 LDLIBS += -lcap -lelf -lz -lrt -lpthread
@@ -97,11 +97,15 @@ OVERRIDE_TARGETS := 1
 override define CLEAN
 	$(call msg,CLEAN)
 	$(RM) -r $(TEST_GEN_PROGS) $(TEST_GEN_PROGS_EXTENDED) $(TEST_GEN_FILES) $(EXTRA_CLEAN)
-	$(MAKE) -C $(BPFDIR) OUTPUT=$(OUTPUT)/ clean
 endef
 
 include ../lib.mk
 
+SCRATCH_DIR := $(OUTPUT)/tools
+BUILD_DIR := $(SCRATCH_DIR)/build
+INCLUDE_DIR := $(SCRATCH_DIR)/include
+BPFOBJ := $(BUILD_DIR)/libbpf/libbpf.a
+
 # Define simple and short `make test_progs`, `make test_sysctl`, etc targets
 # to build individual tests.
 # NOTE: Semicolon at the end is critical to override lib.mk's default static
@@ -120,7 +124,7 @@ $(OUTPUT)/urandom_read: urandom_read.c
 	$(call msg,BINARY,,$@)
 	$(CC) -o $@ $< -Wl,--build-id
 
-$(OUTPUT)/test_stub.o: test_stub.c
+$(OUTPUT)/test_stub.o: test_stub.c $(BPFOBJ)
 	$(call msg,CC,,$@)
 	$(CC) -c $(CFLAGS) -o $@ $<
 
@@ -128,12 +132,10 @@ VMLINUX_BTF_PATHS := $(abspath ../../../../vmlinux)			\
 			       /sys/kernel/btf/vmlinux			\
 			       /boot/vmlinux-$(shell uname -r)
 VMLINUX_BTF:= $(firstword $(wildcard $(VMLINUX_BTF_PATHS)))
-.PHONY: $(OUTPUT)/runqslower
-$(OUTPUT)/runqslower: force
+$(OUTPUT)/runqslower: $(BPFOBJ)
 	$(Q)$(MAKE) $(submake_extras) -C $(TOOLSDIR)/bpf/runqslower	\
-		    OUTPUT=$(OUTPUT)/tools/ VMLINUX_BTF=$(VMLINUX_BTF)
-
-BPFOBJ := $(OUTPUT)/libbpf.a
+		    OUTPUT=$(SCRATCH_DIR)/ VMLINUX_BTF=$(VMLINUX_BTF)   \
+		    BPFOBJ=$(BPFOBJ) BPF_INCLUDE=$(INCLUDE_DIR)
 
 $(TEST_GEN_PROGS) $(TEST_GEN_PROGS_EXTENDED): $(OUTPUT)/test_stub.o $(BPFOBJ)
 
@@ -151,25 +153,22 @@ $(OUTPUT)/test_netcnt: cgroup_helpers.c
 $(OUTPUT)/test_sock_fields: cgroup_helpers.c
 $(OUTPUT)/test_sysctl: cgroup_helpers.c
 
-.PHONY: force
-
-# force a rebuild of BPFOBJ when its dependencies are updated
-force:
-
-DEFAULT_BPFTOOL := $(OUTPUT)/tools/sbin/bpftool
+DEFAULT_BPFTOOL := $(SCRATCH_DIR)/sbin/bpftool
 BPFTOOL ?= $(DEFAULT_BPFTOOL)
+$(DEFAULT_BPFTOOL): $(BPFOBJ) | $(BUILD_DIR)/bpftool
+	$(Q)$(MAKE) $(submake_extras)  -C $(BPFTOOLDIR)			\
+		    OUTPUT=$(BUILD_DIR)/bpftool/			\
+		    prefix= DESTDIR=$(SCRATCH_DIR)/ install
 
-$(DEFAULT_BPFTOOL): force
-	$(Q)$(MAKE) $(submake_extras)  -C $(BPFTOOLDIR)			      \
-		    prefix= DESTDIR=$(OUTPUT)/tools/ install
-
-$(BPFOBJ): force
-	$(Q)$(MAKE) $(submake_extras) -C $(BPFDIR) OUTPUT=$(OUTPUT)/
+$(BPFOBJ): $(wildcard $(BPFDIR)/*.c $(BPFDIR)/*.h $(BPFDIR)/Makefile)          \
+	   ../../../include/uapi/linux/bpf.h                                   \
+	   | $(INCLUDE_DIR) $(BUILD_DIR)/libbpf
+	$(Q)$(MAKE) $(submake_extras) -C $(BPFDIR) OUTPUT=$(BUILD_DIR)/libbpf/ \
+		DESTDIR=$(SCRATCH_DIR) prefix= all install_headers
 
-BPF_HELPERS := $(OUTPUT)/bpf_helper_defs.h $(wildcard $(BPFDIR)/bpf_*.h)
-$(OUTPUT)/bpf_helper_defs.h: $(BPFOBJ)
-	$(Q)$(MAKE) $(submake_extras) -C $(BPFDIR)			      \
-		    OUTPUT=$(OUTPUT)/ $(OUTPUT)/bpf_helper_defs.h
+$(BUILD_DIR)/libbpf $(BUILD_DIR)/bpftool $(INCLUDE_DIR):
+	$(call msg,MKDIR,,$@)
+	mkdir -p $@
 
 # Get Clang's default includes on this system, as opposed to those seen by
 # '-target bpf'. This fixes "missing" files on some architectures/distros,
@@ -189,8 +188,8 @@ MENDIAN=$(if $(IS_LITTLE_ENDIAN),-mlittle-endian,-mbig-endian)
 
 CLANG_SYS_INCLUDES = $(call get_sys_includes,$(CLANG))
 BPF_CFLAGS = -g -D__TARGET_ARCH_$(SRCARCH) $(MENDIAN) 			\
-	     -I$(OUTPUT) -I$(CURDIR) -I$(CURDIR)/include/uapi		\
-	     -I$(APIDIR) -I$(LIBDIR) -I$(BPFDIR) -I$(abspath $(OUTPUT)/../usr/include)
+	     -I$(INCLUDE_DIR) -I$(CURDIR) -I$(CURDIR)/include/uapi	\
+	     -I$(APIDIR) -I$(abspath $(OUTPUT)/../usr/include)
 
 CLANG_CFLAGS = $(CLANG_SYS_INCLUDES) \
 	       -Wno-compare-distinct-pointer-types
@@ -279,7 +278,7 @@ $(TRUNNER_BPF_PROGS_DIR)$(if $2,-)$2-bpfobjs := y
 $(TRUNNER_BPF_OBJS): $(TRUNNER_OUTPUT)/%.o:				\
 		     $(TRUNNER_BPF_PROGS_DIR)/%.c			\
 		     $(TRUNNER_BPF_PROGS_DIR)/*.h			\
-		     $$(BPF_HELPERS) | $(TRUNNER_OUTPUT)
+		     $$(BPFOBJ) | $(TRUNNER_OUTPUT)
 	$$(call $(TRUNNER_BPF_BUILD_RULE),$$<,$$@,			\
 					  $(TRUNNER_BPF_CFLAGS),	\
 					  $(TRUNNER_BPF_LDFLAGS))
@@ -392,7 +391,7 @@ $(OUTPUT)/test_cpp: test_cpp.cpp $(OUTPUT)/test_core_extern.skel.h $(BPFOBJ)
 	$(call msg,CXX,,$@)
 	$(CXX) $(CFLAGS) $^ $(LDLIBS) -o $@
 
-EXTRA_CLEAN := $(TEST_CUSTOM_PROGS)					\
+EXTRA_CLEAN := $(TEST_CUSTOM_PROGS) $(SCRATCH_DIR)			\
 	prog_tests/tests.h map_tests/tests.h verifier/tests.h		\
 	feature								\
-	$(addprefix $(OUTPUT)/,*.o *.skel.h no_alu32 bpf_gcc tools)
+	$(addprefix $(OUTPUT)/,*.o *.skel.h no_alu32 bpf_gcc)
-- 
cgit v1.2.3-59-g8ed1b


From 49e211c0e357d8c5ccdc3c6c5fb8f94ab85d037f Mon Sep 17 00:00:00 2001
From: Chen Zhou <chenzhou10@huawei.com>
Date: Mon, 20 Jan 2020 20:49:43 +0800
Subject: net: hns3: replace snprintf with scnprintf in hns3_dbg_cmd_read

The return value of snprintf may be greater than the size of
HNS3_DBG_READ_LEN, use scnprintf instead in hns3_dbg_cmd_read.

Signed-off-by: Chen Zhou <chenzhou10@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c b/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c
index 6b328a259efc..92ee1f5cf31f 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c
@@ -297,8 +297,8 @@ static ssize_t hns3_dbg_cmd_read(struct file *filp, char __user *buffer,
 	if (!buf)
 		return -ENOMEM;
 
-	len = snprintf(buf, HNS3_DBG_READ_LEN, "%s\n",
-		       "Please echo help to cmd to get help information");
+	len = scnprintf(buf, HNS3_DBG_READ_LEN, "%s\n",
+			"Please echo help to cmd to get help information");
 	uncopy_bytes = copy_to_user(buffer, buf, len);
 
 	kfree(buf);
-- 
cgit v1.2.3-59-g8ed1b


From bea5416561b1b997adbe3a04a5665e647c589a13 Mon Sep 17 00:00:00 2001
From: Chen Zhou <chenzhou10@huawei.com>
Date: Mon, 20 Jan 2020 20:50:33 +0800
Subject: net: hns3: replace snprintf with scnprintf in hns3_update_strings

snprintf returns the number of bytes that would be written, which may be
greater than the the actual length to be written. Here use extra code to
handle this.

scnprintf returns the number of bytes that was actually written, just use
scnprintf to simplify the code.

Signed-off-by: Chen Zhou <chenzhou10@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c b/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c
index 6e0212b79438..c03856e63320 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c
@@ -423,9 +423,8 @@ static void *hns3_update_strings(u8 *data, const struct hns3_stats *stats,
 			data[ETH_GSTRING_LEN - 1] = '\0';
 
 			/* first, prepend the prefix string */
-			n1 = snprintf(data, MAX_PREFIX_SIZE, "%s%d_",
-				      prefix, i);
-			n1 = min_t(uint, n1, MAX_PREFIX_SIZE - 1);
+			n1 = scnprintf(data, MAX_PREFIX_SIZE, "%s%d_",
+				       prefix, i);
 			size_left = (ETH_GSTRING_LEN - 1) - n1;
 
 			/* now, concatenate the stats string to it */
-- 
cgit v1.2.3-59-g8ed1b


From 3231e5d2228a2078ce5982d63ea9a617e4972c00 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Mon, 20 Jan 2020 22:16:07 +0100
Subject: net: phy: rename phy_do_ioctl to phy_do_ioctl_running

We just added phy_do_ioctl, but it turned out that we need another
version of this function that doesn't check whether net_device is
running. So rename phy_do_ioctl to phy_do_ioctl_running.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 2 +-
 drivers/net/phy/phy.c                     | 6 +++---
 include/linux/phy.h                       | 2 +-
 3 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 175f951b6547..7a5fe11378aa 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -5158,7 +5158,7 @@ static const struct net_device_ops rtl_netdev_ops = {
 	.ndo_fix_features	= rtl8169_fix_features,
 	.ndo_set_features	= rtl8169_set_features,
 	.ndo_set_mac_address	= rtl_set_mac_address,
-	.ndo_do_ioctl		= phy_do_ioctl,
+	.ndo_do_ioctl		= phy_do_ioctl_running,
 	.ndo_set_rx_mode	= rtl_set_rx_mode,
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller	= rtl8169_netpoll,
diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c
index da05b3480abb..cf25fa3be123 100644
--- a/drivers/net/phy/phy.c
+++ b/drivers/net/phy/phy.c
@@ -433,19 +433,19 @@ int phy_mii_ioctl(struct phy_device *phydev, struct ifreq *ifr, int cmd)
 EXPORT_SYMBOL(phy_mii_ioctl);
 
 /**
- * phy_do_ioctl - generic ndo_do_ioctl implementation
+ * phy_do_ioctl_running - generic ndo_do_ioctl implementation
  * @dev: the net_device struct
  * @ifr: &struct ifreq for socket ioctl's
  * @cmd: ioctl cmd to execute
  */
-int phy_do_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
+int phy_do_ioctl_running(struct net_device *dev, struct ifreq *ifr, int cmd)
 {
 	if (!netif_running(dev) || !dev->phydev)
 		return -ENODEV;
 
 	return phy_mii_ioctl(dev->phydev, ifr, cmd);
 }
-EXPORT_SYMBOL(phy_do_ioctl);
+EXPORT_SYMBOL(phy_do_ioctl_running);
 
 void phy_queue_state_machine(struct phy_device *phydev, unsigned long jiffies)
 {
diff --git a/include/linux/phy.h b/include/linux/phy.h
index be6b3a1b03da..f6e714da37d8 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -1242,7 +1242,7 @@ void phy_ethtool_ksettings_get(struct phy_device *phydev,
 int phy_ethtool_ksettings_set(struct phy_device *phydev,
 			      const struct ethtool_link_ksettings *cmd);
 int phy_mii_ioctl(struct phy_device *phydev, struct ifreq *ifr, int cmd);
-int phy_do_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd);
+int phy_do_ioctl_running(struct net_device *dev, struct ifreq *ifr, int cmd);
 void phy_request_interrupt(struct phy_device *phydev);
 void phy_free_interrupt(struct phy_device *phydev);
 void phy_print_status(struct phy_device *phydev);
-- 
cgit v1.2.3-59-g8ed1b


From bbbf8430afe6906abbf879352fe10d24d380e588 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Mon, 20 Jan 2020 22:17:11 +0100
Subject: net: phy: add new version of phy_do_ioctl

Add a new version of phy_do_ioctl that doesn't check whether net_device
is running. It will typically be used if suitable drivers attach the
PHY in probe already.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/phy.c | 16 +++++++++++++---
 include/linux/phy.h   |  1 +
 2 files changed, 14 insertions(+), 3 deletions(-)

diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c
index cf25fa3be123..d76e038cf2cb 100644
--- a/drivers/net/phy/phy.c
+++ b/drivers/net/phy/phy.c
@@ -433,18 +433,28 @@ int phy_mii_ioctl(struct phy_device *phydev, struct ifreq *ifr, int cmd)
 EXPORT_SYMBOL(phy_mii_ioctl);
 
 /**
- * phy_do_ioctl_running - generic ndo_do_ioctl implementation
+ * phy_do_ioctl - generic ndo_do_ioctl implementation
  * @dev: the net_device struct
  * @ifr: &struct ifreq for socket ioctl's
  * @cmd: ioctl cmd to execute
  */
-int phy_do_ioctl_running(struct net_device *dev, struct ifreq *ifr, int cmd)
+int phy_do_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 {
-	if (!netif_running(dev) || !dev->phydev)
+	if (!dev->phydev)
 		return -ENODEV;
 
 	return phy_mii_ioctl(dev->phydev, ifr, cmd);
 }
+EXPORT_SYMBOL(phy_do_ioctl);
+
+/* same as phy_do_ioctl, but ensures that net_device is running */
+int phy_do_ioctl_running(struct net_device *dev, struct ifreq *ifr, int cmd)
+{
+	if (!netif_running(dev))
+		return -ENODEV;
+
+	return phy_do_ioctl(dev, ifr, cmd);
+}
 EXPORT_SYMBOL(phy_do_ioctl_running);
 
 void phy_queue_state_machine(struct phy_device *phydev, unsigned long jiffies)
diff --git a/include/linux/phy.h b/include/linux/phy.h
index f6e714da37d8..c570e162e05e 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -1242,6 +1242,7 @@ void phy_ethtool_ksettings_get(struct phy_device *phydev,
 int phy_ethtool_ksettings_set(struct phy_device *phydev,
 			      const struct ethtool_link_ksettings *cmd);
 int phy_mii_ioctl(struct phy_device *phydev, struct ifreq *ifr, int cmd);
+int phy_do_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd);
 int phy_do_ioctl_running(struct net_device *dev, struct ifreq *ifr, int cmd);
 void phy_request_interrupt(struct phy_device *phydev);
 void phy_free_interrupt(struct phy_device *phydev);
-- 
cgit v1.2.3-59-g8ed1b


From f1294617d2f38bd2b9f6cce516b0326858b61182 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Mon, 20 Jan 2020 22:18:37 +0100
Subject: net: convert suitable network drivers to use phy_do_ioctl

Convert suitable network drivers to use phy_do_ioctl.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/agere/et131x.c          | 11 +----------
 drivers/net/ethernet/atheros/ag71xx.c        | 10 +---------
 drivers/net/ethernet/faraday/ftgmac100.c     | 11 +----------
 drivers/net/ethernet/freescale/fec_mpc52xx.c | 12 +-----------
 drivers/net/ethernet/rdc/r6040.c             | 10 +---------
 5 files changed, 5 insertions(+), 49 deletions(-)

diff --git a/drivers/net/ethernet/agere/et131x.c b/drivers/net/ethernet/agere/et131x.c
index 3c51d8c502ed..cb6a761d5c11 100644
--- a/drivers/net/ethernet/agere/et131x.c
+++ b/drivers/net/ethernet/agere/et131x.c
@@ -3651,15 +3651,6 @@ static int et131x_close(struct net_device *netdev)
 	return del_timer_sync(&adapter->error_timer);
 }
 
-static int et131x_ioctl(struct net_device *netdev, struct ifreq *reqbuf,
-			int cmd)
-{
-	if (!netdev->phydev)
-		return -EINVAL;
-
-	return phy_mii_ioctl(netdev->phydev, reqbuf, cmd);
-}
-
 /* et131x_set_packet_filter - Configures the Rx Packet filtering */
 static int et131x_set_packet_filter(struct et131x_adapter *adapter)
 {
@@ -3899,7 +3890,7 @@ static const struct net_device_ops et131x_netdev_ops = {
 	.ndo_set_mac_address	= eth_mac_addr,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_get_stats		= et131x_stats,
-	.ndo_do_ioctl		= et131x_ioctl,
+	.ndo_do_ioctl		= phy_do_ioctl,
 };
 
 static int et131x_pci_setup(struct pci_dev *pdev,
diff --git a/drivers/net/ethernet/atheros/ag71xx.c b/drivers/net/ethernet/atheros/ag71xx.c
index 88e4e1500a8e..5ce2df482d8c 100644
--- a/drivers/net/ethernet/atheros/ag71xx.c
+++ b/drivers/net/ethernet/atheros/ag71xx.c
@@ -1394,14 +1394,6 @@ err_drop:
 	return NETDEV_TX_OK;
 }
 
-static int ag71xx_do_ioctl(struct net_device *ndev, struct ifreq *ifr, int cmd)
-{
-	if (!ndev->phydev)
-		return -EINVAL;
-
-	return phy_mii_ioctl(ndev->phydev, ifr, cmd);
-}
-
 static void ag71xx_oom_timer_handler(struct timer_list *t)
 {
 	struct ag71xx *ag = from_timer(ag, t, oom_timer);
@@ -1618,7 +1610,7 @@ static const struct net_device_ops ag71xx_netdev_ops = {
 	.ndo_open		= ag71xx_open,
 	.ndo_stop		= ag71xx_stop,
 	.ndo_start_xmit		= ag71xx_hard_start_xmit,
-	.ndo_do_ioctl		= ag71xx_do_ioctl,
+	.ndo_do_ioctl		= phy_do_ioctl,
 	.ndo_tx_timeout		= ag71xx_tx_timeout,
 	.ndo_change_mtu		= ag71xx_change_mtu,
 	.ndo_set_mac_address	= eth_mac_addr,
diff --git a/drivers/net/ethernet/faraday/ftgmac100.c b/drivers/net/ethernet/faraday/ftgmac100.c
index 48b3b72fe02e..4572797f00d7 100644
--- a/drivers/net/ethernet/faraday/ftgmac100.c
+++ b/drivers/net/ethernet/faraday/ftgmac100.c
@@ -1536,15 +1536,6 @@ static int ftgmac100_stop(struct net_device *netdev)
 	return 0;
 }
 
-/* optional */
-static int ftgmac100_do_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
-{
-	if (!netdev->phydev)
-		return -ENXIO;
-
-	return phy_mii_ioctl(netdev->phydev, ifr, cmd);
-}
-
 static void ftgmac100_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 {
 	struct ftgmac100 *priv = netdev_priv(netdev);
@@ -1597,7 +1588,7 @@ static const struct net_device_ops ftgmac100_netdev_ops = {
 	.ndo_start_xmit		= ftgmac100_hard_start_xmit,
 	.ndo_set_mac_address	= ftgmac100_set_mac_addr,
 	.ndo_validate_addr	= eth_validate_addr,
-	.ndo_do_ioctl		= ftgmac100_do_ioctl,
+	.ndo_do_ioctl		= phy_do_ioctl,
 	.ndo_tx_timeout		= ftgmac100_tx_timeout,
 	.ndo_set_rx_mode	= ftgmac100_set_rx_mode,
 	.ndo_set_features	= ftgmac100_set_features,
diff --git a/drivers/net/ethernet/freescale/fec_mpc52xx.c b/drivers/net/ethernet/freescale/fec_mpc52xx.c
index de5278485062..7a3f066e611d 100644
--- a/drivers/net/ethernet/freescale/fec_mpc52xx.c
+++ b/drivers/net/ethernet/freescale/fec_mpc52xx.c
@@ -785,16 +785,6 @@ static const struct ethtool_ops mpc52xx_fec_ethtool_ops = {
 };
 
 
-static int mpc52xx_fec_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
-{
-	struct phy_device *phydev = dev->phydev;
-
-	if (!phydev)
-		return -ENOTSUPP;
-
-	return phy_mii_ioctl(phydev, rq, cmd);
-}
-
 static const struct net_device_ops mpc52xx_fec_netdev_ops = {
 	.ndo_open = mpc52xx_fec_open,
 	.ndo_stop = mpc52xx_fec_close,
@@ -802,7 +792,7 @@ static const struct net_device_ops mpc52xx_fec_netdev_ops = {
 	.ndo_set_rx_mode = mpc52xx_fec_set_multicast_list,
 	.ndo_set_mac_address = mpc52xx_fec_set_mac_address,
 	.ndo_validate_addr = eth_validate_addr,
-	.ndo_do_ioctl = mpc52xx_fec_ioctl,
+	.ndo_do_ioctl = phy_do_ioctl,
 	.ndo_tx_timeout = mpc52xx_fec_tx_timeout,
 	.ndo_get_stats = mpc52xx_fec_get_stats,
 #ifdef CONFIG_NET_POLL_CONTROLLER
diff --git a/drivers/net/ethernet/rdc/r6040.c b/drivers/net/ethernet/rdc/r6040.c
index c23cb61bbd30..f5ecc410ff85 100644
--- a/drivers/net/ethernet/rdc/r6040.c
+++ b/drivers/net/ethernet/rdc/r6040.c
@@ -498,14 +498,6 @@ static int r6040_close(struct net_device *dev)
 	return 0;
 }
 
-static int r6040_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
-{
-	if (!dev->phydev)
-		return -EINVAL;
-
-	return phy_mii_ioctl(dev->phydev, rq, cmd);
-}
-
 static int r6040_rx(struct net_device *dev, int limit)
 {
 	struct r6040_private *priv = netdev_priv(dev);
@@ -957,7 +949,7 @@ static const struct net_device_ops r6040_netdev_ops = {
 	.ndo_set_rx_mode	= r6040_multicast_list,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_mac_address	= eth_mac_addr,
-	.ndo_do_ioctl		= r6040_ioctl,
+	.ndo_do_ioctl		= phy_do_ioctl,
 	.ndo_tx_timeout		= r6040_tx_timeout,
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller	= r6040_poll_controller,
-- 
cgit v1.2.3-59-g8ed1b


From 29237d22bc45d340792be217f40fb728ca48b6e6 Mon Sep 17 00:00:00 2001
From: Hans Wippel <ndev@hwipl.net>
Date: Tue, 21 Jan 2020 01:04:46 +0100
Subject: net/smc: allow unprivileged users to read pnet table

The current flags of the SMC_PNET_GET command only allow privileged
users to retrieve entries from the pnet table via netlink. The content
of the pnet table may be useful for all users though, e.g., for
debugging smc connection problems.

This patch removes the GENL_ADMIN_PERM flag so that unprivileged users
can read the pnet table.

Signed-off-by: Hans Wippel <ndev@hwipl.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/smc/smc_pnet.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/smc/smc_pnet.c b/net/smc/smc_pnet.c
index 82dedf052d86..2a5ed47c3e08 100644
--- a/net/smc/smc_pnet.c
+++ b/net/smc/smc_pnet.c
@@ -611,7 +611,7 @@ static const struct genl_ops smc_pnet_ops[] = {
 	{
 		.cmd = SMC_PNETID_GET,
 		.validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP,
-		.flags = GENL_ADMIN_PERM,
+		/* can be retrieved by unprivileged users */
 		.doit = smc_pnet_get,
 		.dumpit = smc_pnet_dump,
 		.start = smc_pnet_dump_start
-- 
cgit v1.2.3-59-g8ed1b


From f362e5fe0f1f9774b28c5b4c9b07b0168575cf6e Mon Sep 17 00:00:00 2001
From: Martin Schiller <ms@dev.tdt.de>
Date: Tue, 21 Jan 2020 07:00:33 +0100
Subject: wan/hdlc_x25: make lapb params configurable

This enables you to configure mode (DTE/DCE), Modulo, Window, T1, T2, N2 via
sethdlc (which needs to be patched as well).

Signed-off-by: Martin Schiller <ms@dev.tdt.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/wan/hdlc_x25.c      | 80 +++++++++++++++++++++++++++++++++++++++--
 include/uapi/linux/hdlc/ioctl.h |  9 +++++
 include/uapi/linux/if.h         |  1 +
 3 files changed, 87 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wan/hdlc_x25.c b/drivers/net/wan/hdlc_x25.c
index 5643675ff724..688233c2e1ea 100644
--- a/drivers/net/wan/hdlc_x25.c
+++ b/drivers/net/wan/hdlc_x25.c
@@ -21,8 +21,17 @@
 #include <linux/skbuff.h>
 #include <net/x25device.h>
 
+struct x25_state {
+	x25_hdlc_proto settings;
+};
+
 static int x25_ioctl(struct net_device *dev, struct ifreq *ifr);
 
+static struct x25_state *state(hdlc_device *hdlc)
+{
+	return hdlc->state;
+}
+
 /* These functions are callbacks called by LAPB layer */
 
 static void x25_connect_disconnect(struct net_device *dev, int reason, int code)
@@ -131,7 +140,6 @@ static netdev_tx_t x25_xmit(struct sk_buff *skb, struct net_device *dev)
 
 static int x25_open(struct net_device *dev)
 {
-	int result;
 	static const struct lapb_register_struct cb = {
 		.connect_confirmation = x25_connected,
 		.connect_indication = x25_connected,
@@ -140,10 +148,33 @@ static int x25_open(struct net_device *dev)
 		.data_indication = x25_data_indication,
 		.data_transmit = x25_data_transmit,
 	};
+	hdlc_device *hdlc = dev_to_hdlc(dev);
+	struct lapb_parms_struct params;
+	int result;
 
 	result = lapb_register(dev, &cb);
 	if (result != LAPB_OK)
 		return result;
+
+	result = lapb_getparms(dev, &params);
+	if (result != LAPB_OK)
+		return result;
+
+	if (state(hdlc)->settings.dce)
+		params.mode = params.mode | LAPB_DCE;
+
+	if (state(hdlc)->settings.modulo == 128)
+		params.mode = params.mode | LAPB_EXTENDED;
+
+	params.window = state(hdlc)->settings.window;
+	params.t1 = state(hdlc)->settings.t1;
+	params.t2 = state(hdlc)->settings.t2;
+	params.n2 = state(hdlc)->settings.n2;
+
+	result = lapb_setparms(dev, &params);
+	if (result != LAPB_OK)
+		return result;
+
 	return 0;
 }
 
@@ -186,7 +217,10 @@ static struct hdlc_proto proto = {
 
 static int x25_ioctl(struct net_device *dev, struct ifreq *ifr)
 {
+	x25_hdlc_proto __user *x25_s = ifr->ifr_settings.ifs_ifsu.x25;
+	const size_t size = sizeof(x25_hdlc_proto);
 	hdlc_device *hdlc = dev_to_hdlc(dev);
+	x25_hdlc_proto new_settings;
 	int result;
 
 	switch (ifr->ifr_settings.type) {
@@ -194,7 +228,13 @@ static int x25_ioctl(struct net_device *dev, struct ifreq *ifr)
 		if (dev_to_hdlc(dev)->proto != &proto)
 			return -EINVAL;
 		ifr->ifr_settings.type = IF_PROTO_X25;
-		return 0; /* return protocol only, no settable parameters */
+		if (ifr->ifr_settings.size < size) {
+			ifr->ifr_settings.size = size; /* data size wanted */
+			return -ENOBUFS;
+		}
+		if (copy_to_user(x25_s, &state(hdlc)->settings, size))
+			return -EFAULT;
+		return 0;
 
 	case IF_PROTO_X25:
 		if (!capable(CAP_NET_ADMIN))
@@ -203,12 +243,46 @@ static int x25_ioctl(struct net_device *dev, struct ifreq *ifr)
 		if (dev->flags & IFF_UP)
 			return -EBUSY;
 
+		/* backward compatibility */
+		if (ifr->ifr_settings.size = 0) {
+			new_settings.dce = 0;
+			new_settings.modulo = 8;
+			new_settings.window = 7;
+			new_settings.t1 = 3;
+			new_settings.t2 = 1;
+			new_settings.n2 = 10;
+		}
+		else {
+			if (copy_from_user(&new_settings, x25_s, size))
+				return -EFAULT;
+
+			if ((new_settings.dce != 0 &&
+			new_settings.dce != 1) ||
+			(new_settings.modulo != 8 &&
+			new_settings.modulo != 128) ||
+			new_settings.window < 1 ||
+			(new_settings.modulo == 8 &&
+			new_settings.window > 7) ||
+			(new_settings.modulo == 128 &&
+			new_settings.window > 127) ||
+			new_settings.t1 < 1 ||
+			new_settings.t1 > 255 ||
+			new_settings.t2 < 1 ||
+			new_settings.t2 > 255 ||
+			new_settings.n2 < 1 ||
+			new_settings.n2 > 255)
+				return -EINVAL;
+		}
+
 		result=hdlc->attach(dev, ENCODING_NRZ,PARITY_CRC16_PR1_CCITT);
 		if (result)
 			return result;
 
-		if ((result = attach_hdlc_protocol(dev, &proto, 0)))
+		if ((result = attach_hdlc_protocol(dev, &proto,
+						   sizeof(struct x25_state))))
 			return result;
+
+		memcpy(&state(hdlc)->settings, &new_settings, size);
 		dev->type = ARPHRD_X25;
 		call_netdevice_notifiers(NETDEV_POST_TYPE_CHANGE, dev);
 		netif_dormant_off(dev);
diff --git a/include/uapi/linux/hdlc/ioctl.h b/include/uapi/linux/hdlc/ioctl.h
index 0fe4238e8246..b06341acab5e 100644
--- a/include/uapi/linux/hdlc/ioctl.h
+++ b/include/uapi/linux/hdlc/ioctl.h
@@ -79,6 +79,15 @@ typedef struct {
     unsigned int timeout;
 } cisco_proto;
 
+typedef struct {
+	unsigned short dce; /* 1 for DCE (network side) operation */
+	unsigned int modulo; /* modulo (8 = basic / 128 = extended) */
+	unsigned int window; /* frame window size */
+	unsigned int t1; /* timeout t1 */
+	unsigned int t2; /* timeout t2 */
+	unsigned int n2; /* frame retry counter */
+} x25_hdlc_proto;
+
 /* PPP doesn't need any info now - supply length = 0 to ioctl */
 
 #endif /* __ASSEMBLY__ */
diff --git a/include/uapi/linux/if.h b/include/uapi/linux/if.h
index 4bf33344aab1..be714cd8c826 100644
--- a/include/uapi/linux/if.h
+++ b/include/uapi/linux/if.h
@@ -213,6 +213,7 @@ struct if_settings {
 		fr_proto		__user *fr;
 		fr_proto_pvc		__user *fr_pvc;
 		fr_proto_pvc_info	__user *fr_pvc_info;
+		x25_hdlc_proto		__user *x25;
 
 		/* interface settings */
 		sync_serial_settings	__user *sync;
-- 
cgit v1.2.3-59-g8ed1b


From 953c4a08dfc9ffe763a8340ac10f459d6c6cc4eb Mon Sep 17 00:00:00 2001
From: Martin Schiller <ms@dev.tdt.de>
Date: Tue, 21 Jan 2020 07:00:34 +0100
Subject: wan/hdlc_x25: fix skb handling

o call skb_reset_network_header() before hdlc->xmit()
 o change skb proto to HDLC (0x0019) before hdlc->xmit()
 o call dev_queue_xmit_nit() before hdlc->xmit()

This changes make it possible to trace (tcpdump) outgoing layer2
(ETH_P_HDLC) packets

Additionally call skb_reset_network_header() after each skb_push() /
skb_pull().

Signed-off-by: Martin Schiller <ms@dev.tdt.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/wan/hdlc_x25.c | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wan/hdlc_x25.c b/drivers/net/wan/hdlc_x25.c
index 688233c2e1ea..63c9aeed9a34 100644
--- a/drivers/net/wan/hdlc_x25.c
+++ b/drivers/net/wan/hdlc_x25.c
@@ -71,11 +71,12 @@ static int x25_data_indication(struct net_device *dev, struct sk_buff *skb)
 {
 	unsigned char *ptr;
 
-	skb_push(skb, 1);
-
 	if (skb_cow(skb, 1))
 		return NET_RX_DROP;
 
+	skb_push(skb, 1);
+	skb_reset_network_header(skb);
+
 	ptr  = skb->data;
 	*ptr = X25_IFACE_DATA;
 
@@ -88,6 +89,13 @@ static int x25_data_indication(struct net_device *dev, struct sk_buff *skb)
 static void x25_data_transmit(struct net_device *dev, struct sk_buff *skb)
 {
 	hdlc_device *hdlc = dev_to_hdlc(dev);
+
+	skb_reset_network_header(skb);
+	skb->protocol = hdlc_type_trans(skb, dev);
+
+	if (dev_nit_active(dev))
+		dev_queue_xmit_nit(skb, dev);
+
 	hdlc->xmit(skb, dev); /* Ignore return value :-( */
 }
 
@@ -102,6 +110,7 @@ static netdev_tx_t x25_xmit(struct sk_buff *skb, struct net_device *dev)
 	switch (skb->data[0]) {
 	case X25_IFACE_DATA:	/* Data to be transmitted */
 		skb_pull(skb, 1);
+		skb_reset_network_header(skb);
 		if ((result = lapb_data_request(dev, skb)) != LAPB_OK)
 			dev_kfree_skb(skb);
 		return NETDEV_TX_OK;
-- 
cgit v1.2.3-59-g8ed1b


From 09783d448bccf1ec1c91d0b9e5443b0bcb034dbb Mon Sep 17 00:00:00 2001
From: Yunsheng Lin <linyunsheng@huawei.com>
Date: Tue, 21 Jan 2020 16:42:05 +0800
Subject: net: hns3: limit the error logging in the hns3_clean_tx_ring()

The error log printed by netdev_err() in the hns3_clean_tx_ring()
may spam the kernel log.

This patch uses hns3_rl_err() to ratelimit the error log in the
hns3_clean_tx_ring().

Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hns3_enet.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
index 914a7630ac48..f5790282bd19 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
@@ -2540,8 +2540,8 @@ void hns3_clean_tx_ring(struct hns3_enet_ring *ring)
 	rmb(); /* Make sure head is ready before touch any data */
 
 	if (unlikely(!is_valid_clean_head(ring, head))) {
-		netdev_err(netdev, "wrong head (%d, %d-%d)\n", head,
-			   ring->next_to_use, ring->next_to_clean);
+		hns3_rl_err(netdev, "wrong head (%d, %d-%d)\n", head,
+			    ring->next_to_use, ring->next_to_clean);
 
 		u64_stats_update_begin(&ring->syncp);
 		ring->stats.io_err_cnt++;
-- 
cgit v1.2.3-59-g8ed1b


From 08bb3857c6c2e2a62fe0cdc593af57215e9b95bc Mon Sep 17 00:00:00 2001
From: Yunsheng Lin <linyunsheng@huawei.com>
Date: Tue, 21 Jan 2020 16:42:06 +0800
Subject: net: hns3: do not reuse pfmemalloc pages

HNS3 driver allocates pages for DMA with dev_alloc_pages(), which
calls alloc_pages_node() with the __GFP_MEMALLOC flag. So, in case
of OOM condition, HNS3 can get pages with pfmemalloc flag set.

So do not reuse the pages with pfmemalloc flag set because those
pages are reserved for special cases, such as low memory case.

Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hns3_enet.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
index f5790282bd19..06c58ed3c45e 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
@@ -2627,6 +2627,12 @@ static void hns3_nic_alloc_rx_buffers(struct hns3_enet_ring *ring,
 	writel_relaxed(i, ring->tqp->io_base + HNS3_RING_RX_RING_HEAD_REG);
 }
 
+static bool hns3_page_is_reusable(struct page *page)
+{
+	return page_to_nid(page) == numa_mem_id() &&
+		!page_is_pfmemalloc(page);
+}
+
 static void hns3_nic_reuse_page(struct sk_buff *skb, int i,
 				struct hns3_enet_ring *ring, int pull_len,
 				struct hns3_desc_cb *desc_cb)
@@ -2641,7 +2647,7 @@ static void hns3_nic_reuse_page(struct sk_buff *skb, int i,
 	/* Avoid re-using remote pages, or the stack is still using the page
 	 * when page_offset rollback to zero, flag default unreuse
 	 */
-	if (unlikely(page_to_nid(desc_cb->priv) != numa_mem_id()) ||
+	if (unlikely(!hns3_page_is_reusable(desc_cb->priv)) ||
 	    (!desc_cb->page_offset && page_count(desc_cb->priv) > 1))
 		return;
 
@@ -2860,7 +2866,7 @@ static int hns3_alloc_skb(struct hns3_enet_ring *ring, unsigned int length,
 		memcpy(__skb_put(skb, length), va, ALIGN(length, sizeof(long)));
 
 		/* We can reuse buffer as-is, just make sure it is local */
-		if (likely(page_to_nid(desc_cb->priv) == numa_mem_id()))
+		if (likely(hns3_page_is_reusable(desc_cb->priv)))
 			desc_cb->reuse_flag = 1;
 		else /* This page cannot be reused so discard it */
 			put_page(desc_cb->priv);
-- 
cgit v1.2.3-59-g8ed1b


From afb6afdb8dc67988d17e4926df8c8fe248ea555a Mon Sep 17 00:00:00 2001
From: Huazhong Tan <tanhuazhong@huawei.com>
Date: Tue, 21 Jan 2020 16:42:07 +0800
Subject: net: hns3: set VF's default reset_type to HNAE3_NONE_RESET

reset_type means what kind of reset the driver is handling now,
so after initializing or reset, the reset_type of VF should be
set to HNAE3_NONE_RESET, otherwise, this unknown default value
may be a little misleading when the device is running.

Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
index b26b8ad170d0..2a472b06a337 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
@@ -1862,6 +1862,7 @@ static void hclgevf_reset_service_task(struct hclgevf_dev *hdev)
 		hclgevf_reset_task_schedule(hdev);
 	}
 
+	hdev->reset_type = HNAE3_NONE_RESET;
 	clear_bit(HCLGEVF_STATE_RST_HANDLING, &hdev->state);
 	up(&hdev->reset_sem);
 }
@@ -2742,6 +2743,7 @@ static int hclgevf_init_hdev(struct hclgevf_dev *hdev)
 
 	hclgevf_state_init(hdev);
 	hdev->reset_level = HNAE3_VF_FUNC_RESET;
+	hdev->reset_type = HNAE3_NONE_RESET;
 
 	ret = hclgevf_misc_irq_init(hdev);
 	if (ret) {
-- 
cgit v1.2.3-59-g8ed1b


From 9027d043fc31145918b40b7a68956eaff4ddb5de Mon Sep 17 00:00:00 2001
From: Guojia Liao <liaoguojia@huawei.com>
Date: Tue, 21 Jan 2020 16:42:08 +0800
Subject: net: hns3: move duplicated macro definition into header

Macro HCLGE_GET_DFX_REG_TYPE_CNT in hclge_dbg_get_dfx_bd_num()
and macro HCLGE_DFX_REG_BD_NUM in hclge_get_dfx_reg_bd_num()
have the same meaning, so just defines HCLGE_GET_DFX_REG_TYPE_CNT
in hclge_main.h.

Signed-off-by: Guojia Liao <liaoguojia@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c | 2 --
 drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c    | 4 +---
 drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h    | 2 ++
 3 files changed, 3 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c
index f3d4cbd28913..67fad80035d3 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c
@@ -73,8 +73,6 @@ static struct hclge_dbg_reg_type_info hclge_dbg_reg_info[] = {
 
 static int hclge_dbg_get_dfx_bd_num(struct hclge_dev *hdev, int offset)
 {
-#define HCLGE_GET_DFX_REG_TYPE_CNT	4
-
 	struct hclge_desc desc[HCLGE_GET_DFX_REG_TYPE_CNT];
 	int entries_per_desc;
 	int index;
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
index 76e8aa40e1bc..8e007b740fdd 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
@@ -10204,10 +10204,8 @@ static int hclge_get_dfx_reg_bd_num(struct hclge_dev *hdev,
 				    int *bd_num_list,
 				    u32 type_num)
 {
-#define HCLGE_DFX_REG_BD_NUM	4
-
 	u32 entries_per_desc, desc_index, index, offset, i;
-	struct hclge_desc desc[HCLGE_DFX_REG_BD_NUM];
+	struct hclge_desc desc[HCLGE_GET_DFX_REG_TYPE_CNT];
 	int ret;
 
 	ret = hclge_query_bd_num_cmd_send(hdev, desc);
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h
index 1c1d6b358897..f78cbb4cc85e 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h
@@ -139,6 +139,8 @@
 #define HCLGE_PHY_MDIX_STATUS_B		6
 #define HCLGE_PHY_SPEED_DUP_RESOLVE_B	11
 
+#define HCLGE_GET_DFX_REG_TYPE_CNT	4
+
 /* Factor used to calculate offset and bitmap of VF num */
 #define HCLGE_VF_NUM_PER_CMD           64
 
-- 
cgit v1.2.3-59-g8ed1b


From cdc37385e3abaf2704adf8bb15a2c14f04762d3d Mon Sep 17 00:00:00 2001
From: Guojia Liao <liaoguojia@huawei.com>
Date: Tue, 21 Jan 2020 16:42:09 +0800
Subject: net: hns3: refine the input parameter 'size' for snprintf()

The function snprintf() writes at most size bytes (including the
terminating null byte ('\0') to str. Now, We can guarantee that the
parameter of size is lager than the length of str to be formatting
including its terminating null byte. So it's unnecessary to minus 1
for the input parameter 'size'.

Signed-off-by: Guojia Liao <liaoguojia@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hns3_enet.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
index 06c58ed3c45e..6271b693ddad 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
@@ -4716,7 +4716,7 @@ static int __init hns3_init_module(void)
 	pr_info("%s: %s\n", hns3_driver_name, hns3_copyright);
 
 	client.type = HNAE3_CLIENT_KNIC;
-	snprintf(client.name, HNAE3_CLIENT_NAME_LENGTH - 1, "%s",
+	snprintf(client.name, HNAE3_CLIENT_NAME_LENGTH, "%s",
 		 hns3_driver_name);
 
 	client.ops = &client_ops;
-- 
cgit v1.2.3-59-g8ed1b


From 6f8e330d27464a7ff436f2944b84d640e42e5e6e Mon Sep 17 00:00:00 2001
From: Yonglong Liu <liuyonglong@huawei.com>
Date: Tue, 21 Jan 2020 16:42:10 +0800
Subject: net: hns3: rewrite a log in hclge_put_vector()

When gets vector fails, hclge_put_vector() should print out
the vector instead of vector_id in the log and return the wrong
vector_id to its caller.

Signed-off-by: Yonglong Liu <liuyonglong@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
index 8e007b740fdd..07fded791d2c 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
@@ -4101,7 +4101,7 @@ static int hclge_put_vector(struct hnae3_handle *handle, int vector)
 	vector_id = hclge_get_vector_index(hdev, vector);
 	if (vector_id < 0) {
 		dev_err(&hdev->pdev->dev,
-			"Get vector index fail. vector_id =%d\n", vector_id);
+			"Get vector index fail. vector = %d\n", vector);
 		return vector_id;
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From e31053298408dc1fbb33ad2fdc9a114e70a9dfd6 Mon Sep 17 00:00:00 2001
From: Guangbin Huang <huangguangbin2@huawei.com>
Date: Tue, 21 Jan 2020 16:42:11 +0800
Subject: net: hns3: delete unnecessary blank line and space for cleanup

This patch deletes some unnecessary blank lines and spaces to clean up
code, and in hclgevf_set_vlan_filter() moves the comment to the front
of hclgevf_send_mbx_msg().

Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c        | 6 +++---
 drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c | 7 +++----
 2 files changed, 6 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c b/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c
index 92ee1f5cf31f..1d4ffc5f408a 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c
@@ -176,7 +176,7 @@ static int hns3_dbg_bd_info(struct hnae3_handle *h, const char *cmd_buf)
 		return -EINVAL;
 	}
 
-	ring  = &priv->ring[q_num];
+	ring = &priv->ring[q_num];
 	value = readl_relaxed(ring->tqp->io_base + HNS3_RING_TX_RING_TAIL_REG);
 	tx_index = (cnt == 1) ? value : tx_index;
 
@@ -209,10 +209,10 @@ static int hns3_dbg_bd_info(struct hnae3_handle *h, const char *cmd_buf)
 		 le16_to_cpu(tx_desc->tx.bdtp_fe_sc_vld_ra_ri));
 	dev_info(dev, "(TX)mss: %u\n", le16_to_cpu(tx_desc->tx.mss));
 
-	ring  = &priv->ring[q_num + h->kinfo.num_tqps];
+	ring = &priv->ring[q_num + h->kinfo.num_tqps];
 	value = readl_relaxed(ring->tqp->io_base + HNS3_RING_RX_RING_TAIL_REG);
 	rx_index = (cnt == 1) ? value : tx_index;
-	rx_desc	 = &ring->desc[rx_index];
+	rx_desc = &ring->desc[rx_index];
 
 	addr = le64_to_cpu(rx_desc->addr);
 	dev_info(dev, "RX Queue Num: %u, BD Index: %u\n", q_num, rx_index);
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
index 2a472b06a337..4b8751387be9 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
@@ -1316,14 +1316,13 @@ static int hclgevf_set_vlan_filter(struct hnae3_handle *handle,
 	msg_data[0] = is_kill;
 	memcpy(&msg_data[1], &vlan_id, sizeof(vlan_id));
 	memcpy(&msg_data[3], &proto, sizeof(proto));
-	ret = hclgevf_send_mbx_msg(hdev, HCLGE_MBX_SET_VLAN,
-				   HCLGE_MBX_VLAN_FILTER, msg_data,
-				   HCLGEVF_VLAN_MBX_MSG_LEN, true, NULL, 0);
-
 	/* when remove hw vlan filter failed, record the vlan id,
 	 * and try to remove it from hw later, to be consistence
 	 * with stack.
 	 */
+	ret = hclgevf_send_mbx_msg(hdev, HCLGE_MBX_SET_VLAN,
+				   HCLGE_MBX_VLAN_FILTER, msg_data,
+				   HCLGEVF_VLAN_MBX_MSG_LEN, true, NULL, 0);
 	if (is_kill && ret)
 		set_bit(vlan_id, hdev->vlan_del_fail_bmap);
 
-- 
cgit v1.2.3-59-g8ed1b


From 322cb97c0734555d7a10299954624363de370c9c Mon Sep 17 00:00:00 2001
From: Huazhong Tan <tanhuazhong@huawei.com>
Date: Tue, 21 Jan 2020 16:42:12 +0800
Subject: net: hns3: remove redundant print on ENOMEM

All kmalloc-based functions print enough information on failures.
So this patch removes the log in hclge_get_dfx_reg() when returns
ENOMEM.

Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
index 07fded791d2c..fafae6764660 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
@@ -10318,10 +10318,8 @@ static int hclge_get_dfx_reg(struct hclge_dev *hdev, void *data)
 
 	buf_len = sizeof(*desc_src) * bd_num_max;
 	desc_src = kzalloc(buf_len, GFP_KERNEL);
-	if (!desc_src) {
-		dev_err(&hdev->pdev->dev, "%s kzalloc failed\n", __func__);
+	if (!desc_src)
 		return -ENOMEM;
-	}
 
 	for (i = 0; i < dfx_reg_type_num; i++) {
 		bd_num = bd_num_list[i];
-- 
cgit v1.2.3-59-g8ed1b


From 60df7e91cc3efa141802dffad3a82b95c2ec479e Mon Sep 17 00:00:00 2001
From: Huazhong Tan <tanhuazhong@huawei.com>
Date: Tue, 21 Jan 2020 16:42:13 +0800
Subject: net: hns3: cleanup some coding style issue

This patch removes some unnecessary return value assignments,
some duplicated printing in the caller, refines the judgment
of 0 and uses le16_to_cpu to replace __le16_to_cpu.

Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hns3_enet.c    |  6 +-
 .../net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c |  6 +-
 .../ethernet/hisilicon/hns3/hns3pf/hclge_main.c    | 81 ++++++----------------
 .../ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c  | 28 +++-----
 4 files changed, 35 insertions(+), 86 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
index 6271b693ddad..acb796cc10d0 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
@@ -2094,10 +2094,8 @@ static int hns3_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	int ret;
 
 	ae_dev = devm_kzalloc(&pdev->dev, sizeof(*ae_dev), GFP_KERNEL);
-	if (!ae_dev) {
-		ret = -ENOMEM;
-		return ret;
-	}
+	if (!ae_dev)
+		return -ENOMEM;
 
 	ae_dev->pdev = pdev;
 	ae_dev->flag = ent->driver_data;
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c
index f8127d72d645..c85b72dc44d2 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c
@@ -1898,10 +1898,8 @@ static int hclge_handle_all_hw_msix_error(struct hclge_dev *hdev,
 
 	bd_num = max_t(u32, mpf_bd_num, pf_bd_num);
 	desc = kcalloc(bd_num, sizeof(struct hclge_desc), GFP_KERNEL);
-	if (!desc) {
-		ret = -ENOMEM;
-		goto out;
-	}
+	if (!desc)
+		return -ENOMEM;
 
 	ret = hclge_handle_mpf_msix_error(hdev, desc, mpf_bd_num,
 					  reset_requests);
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
index fafae6764660..ec5f6eeb639b 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
@@ -862,9 +862,7 @@ static int hclge_query_function_status(struct hclge_dev *hdev)
 		usleep_range(1000, 2000);
 	} while (timeout++ < HCLGE_QUERY_MAX_CNT);
 
-	ret = hclge_parse_func_status(hdev, req);
-
-	return ret;
+	return hclge_parse_func_status(hdev, req);
 }
 
 static int hclge_query_pf_resource(struct hclge_dev *hdev)
@@ -882,12 +880,12 @@ static int hclge_query_pf_resource(struct hclge_dev *hdev)
 	}
 
 	req = (struct hclge_pf_res_cmd *)desc.data;
-	hdev->num_tqps = __le16_to_cpu(req->tqp_num);
-	hdev->pkt_buf_size = __le16_to_cpu(req->buf_size) << HCLGE_BUF_UNIT_S;
+	hdev->num_tqps = le16_to_cpu(req->tqp_num);
+	hdev->pkt_buf_size = le16_to_cpu(req->buf_size) << HCLGE_BUF_UNIT_S;
 
 	if (req->tx_buf_size)
 		hdev->tx_buf_size =
-			__le16_to_cpu(req->tx_buf_size) << HCLGE_BUF_UNIT_S;
+			le16_to_cpu(req->tx_buf_size) << HCLGE_BUF_UNIT_S;
 	else
 		hdev->tx_buf_size = HCLGE_DEFAULT_TX_BUF;
 
@@ -895,7 +893,7 @@ static int hclge_query_pf_resource(struct hclge_dev *hdev)
 
 	if (req->dv_buf_size)
 		hdev->dv_buf_size =
-			__le16_to_cpu(req->dv_buf_size) << HCLGE_BUF_UNIT_S;
+			le16_to_cpu(req->dv_buf_size) << HCLGE_BUF_UNIT_S;
 	else
 		hdev->dv_buf_size = HCLGE_DEFAULT_DV;
 
@@ -903,10 +901,10 @@ static int hclge_query_pf_resource(struct hclge_dev *hdev)
 
 	if (hnae3_dev_roce_supported(hdev)) {
 		hdev->roce_base_msix_offset =
-		hnae3_get_field(__le16_to_cpu(req->msixcap_localid_ba_rocee),
+		hnae3_get_field(le16_to_cpu(req->msixcap_localid_ba_rocee),
 				HCLGE_MSIX_OFT_ROCEE_M, HCLGE_MSIX_OFT_ROCEE_S);
 		hdev->num_roce_msi =
-		hnae3_get_field(__le16_to_cpu(req->pf_intr_vector_number),
+		hnae3_get_field(le16_to_cpu(req->pf_intr_vector_number),
 				HCLGE_PF_VEC_NUM_M, HCLGE_PF_VEC_NUM_S);
 
 		/* nic's msix numbers is always equals to the roce's. */
@@ -919,7 +917,7 @@ static int hclge_query_pf_resource(struct hclge_dev *hdev)
 				hdev->roce_base_msix_offset;
 	} else {
 		hdev->num_msi =
-		hnae3_get_field(__le16_to_cpu(req->pf_intr_vector_number),
+		hnae3_get_field(le16_to_cpu(req->pf_intr_vector_number),
 				HCLGE_PF_VEC_NUM_M, HCLGE_PF_VEC_NUM_S);
 
 		hdev->num_nic_msi = hdev->num_msi;
@@ -1333,11 +1331,7 @@ static int hclge_get_cap(struct hclge_dev *hdev)
 	}
 
 	/* get pf resource */
-	ret = hclge_query_pf_resource(hdev);
-	if (ret)
-		dev_err(&hdev->pdev->dev, "query pf resource error %d.\n", ret);
-
-	return ret;
+	return hclge_query_pf_resource(hdev);
 }
 
 static void hclge_init_kdump_kernel_config(struct hclge_dev *hdev)
@@ -2621,30 +2615,21 @@ static int hclge_mac_init(struct hclge_dev *hdev)
 	hdev->hw.mac.duplex = HCLGE_MAC_FULL;
 	ret = hclge_cfg_mac_speed_dup_hw(hdev, hdev->hw.mac.speed,
 					 hdev->hw.mac.duplex);
-	if (ret) {
-		dev_err(&hdev->pdev->dev,
-			"Config mac speed dup fail ret=%d\n", ret);
+	if (ret)
 		return ret;
-	}
 
 	if (hdev->hw.mac.support_autoneg) {
 		ret = hclge_set_autoneg_en(hdev, hdev->hw.mac.autoneg);
-		if (ret) {
-			dev_err(&hdev->pdev->dev,
-				"Config mac autoneg fail ret=%d\n", ret);
+		if (ret)
 			return ret;
-		}
 	}
 
 	mac->link = 0;
 
 	if (mac->user_fec_mode & BIT(HNAE3_FEC_USER_DEF)) {
 		ret = hclge_set_fec_hw(hdev, mac->user_fec_mode);
-		if (ret) {
-			dev_err(&hdev->pdev->dev,
-				"Fec mode init fail, ret = %d\n", ret);
+		if (ret)
 			return ret;
-		}
 	}
 
 	ret = hclge_set_mac_mtu(hdev, hdev->mps);
@@ -2916,7 +2901,7 @@ static int hclge_get_status(struct hnae3_handle *handle)
 
 static struct hclge_vport *hclge_get_vf_vport(struct hclge_dev *hdev, int vf)
 {
-	if (pci_num_vf(hdev->pdev) == 0) {
+	if (!pci_num_vf(hdev->pdev)) {
 		dev_err(&hdev->pdev->dev,
 			"SRIOV is disabled, can not get vport(%d) info.\n", vf);
 		return NULL;
@@ -6584,7 +6569,7 @@ static int hclge_set_serdes_loopback(struct hclge_dev *hdev, bool en,
 
 	hclge_cfg_mac_mode(hdev, en);
 
-	ret = hclge_mac_phy_link_status_wait(hdev, en, FALSE);
+	ret = hclge_mac_phy_link_status_wait(hdev, en, false);
 	if (ret)
 		dev_err(&hdev->pdev->dev,
 			"serdes loopback config mac mode timeout\n");
@@ -6642,7 +6627,7 @@ static int hclge_set_phy_loopback(struct hclge_dev *hdev, bool en)
 
 	hclge_cfg_mac_mode(hdev, en);
 
-	ret = hclge_mac_phy_link_status_wait(hdev, en, TRUE);
+	ret = hclge_mac_phy_link_status_wait(hdev, en, true);
 	if (ret)
 		dev_err(&hdev->pdev->dev,
 			"phy loopback config mac mode timeout\n");
@@ -9394,17 +9379,13 @@ static int hclge_init_ae_dev(struct hnae3_ae_dev *ae_dev)
 	sema_init(&hdev->reset_sem, 1);
 
 	ret = hclge_pci_init(hdev);
-	if (ret) {
-		dev_err(&pdev->dev, "PCI init failed\n");
+	if (ret)
 		goto out;
-	}
 
 	/* Firmware command queue initialize */
 	ret = hclge_cmd_queue_init(hdev);
-	if (ret) {
-		dev_err(&pdev->dev, "Cmd queue init failed, ret = %d.\n", ret);
+	if (ret)
 		goto err_pci_uninit;
-	}
 
 	/* Firmware command initialize */
 	ret = hclge_cmd_init(hdev);
@@ -9412,11 +9393,8 @@ static int hclge_init_ae_dev(struct hnae3_ae_dev *ae_dev)
 		goto err_cmd_uninit;
 
 	ret = hclge_get_cap(hdev);
-	if (ret) {
-		dev_err(&pdev->dev, "get hw capability error, ret = %d.\n",
-			ret);
+	if (ret)
 		goto err_cmd_uninit;
-	}
 
 	ret = hclge_configure(hdev);
 	if (ret) {
@@ -9431,12 +9409,8 @@ static int hclge_init_ae_dev(struct hnae3_ae_dev *ae_dev)
 	}
 
 	ret = hclge_misc_irq_init(hdev);
-	if (ret) {
-		dev_err(&pdev->dev,
-			"Misc IRQ(vector0) init error, ret = %d.\n",
-			ret);
+	if (ret)
 		goto err_msi_uninit;
-	}
 
 	ret = hclge_alloc_tqps(hdev);
 	if (ret) {
@@ -9445,31 +9419,22 @@ static int hclge_init_ae_dev(struct hnae3_ae_dev *ae_dev)
 	}
 
 	ret = hclge_alloc_vport(hdev);
-	if (ret) {
-		dev_err(&pdev->dev, "Allocate vport error, ret = %d.\n", ret);
+	if (ret)
 		goto err_msi_irq_uninit;
-	}
 
 	ret = hclge_map_tqp(hdev);
-	if (ret) {
-		dev_err(&pdev->dev, "Map tqp error, ret = %d.\n", ret);
+	if (ret)
 		goto err_msi_irq_uninit;
-	}
 
 	if (hdev->hw.mac.media_type == HNAE3_MEDIA_TYPE_COPPER) {
 		ret = hclge_mac_mdio_config(hdev);
-		if (ret) {
-			dev_err(&hdev->pdev->dev,
-				"mdio config fail ret=%d\n", ret);
+		if (ret)
 			goto err_msi_irq_uninit;
-		}
 	}
 
 	ret = hclge_init_umv_space(hdev);
-	if (ret) {
-		dev_err(&pdev->dev, "umv space init error, ret=%d.\n", ret);
+	if (ret)
 		goto err_mdiobus_unreg;
-	}
 
 	ret = hclge_mac_init(hdev);
 	if (ret) {
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
index 4b8751387be9..d6597206e692 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
@@ -2597,11 +2597,11 @@ static int hclgevf_query_vf_resource(struct hclgevf_dev *hdev)
 
 	if (hnae3_dev_roce_supported(hdev)) {
 		hdev->roce_base_msix_offset =
-		hnae3_get_field(__le16_to_cpu(req->msixcap_localid_ba_rocee),
+		hnae3_get_field(le16_to_cpu(req->msixcap_localid_ba_rocee),
 				HCLGEVF_MSIX_OFT_ROCEE_M,
 				HCLGEVF_MSIX_OFT_ROCEE_S);
 		hdev->num_roce_msix =
-		hnae3_get_field(__le16_to_cpu(req->vf_intr_vector_number),
+		hnae3_get_field(le16_to_cpu(req->vf_intr_vector_number),
 				HCLGEVF_VEC_NUM_M, HCLGEVF_VEC_NUM_S);
 
 		/* nic's msix numbers is always equals to the roce's. */
@@ -2614,7 +2614,7 @@ static int hclgevf_query_vf_resource(struct hclgevf_dev *hdev)
 				hdev->roce_base_msix_offset;
 	} else {
 		hdev->num_msi =
-		hnae3_get_field(__le16_to_cpu(req->vf_intr_vector_number),
+		hnae3_get_field(le16_to_cpu(req->vf_intr_vector_number),
 				HCLGEVF_VEC_NUM_M, HCLGEVF_VEC_NUM_S);
 
 		hdev->num_nic_msix = hdev->num_msi;
@@ -2711,16 +2711,12 @@ static int hclgevf_init_hdev(struct hclgevf_dev *hdev)
 	int ret;
 
 	ret = hclgevf_pci_init(hdev);
-	if (ret) {
-		dev_err(&pdev->dev, "PCI initialization failed\n");
+	if (ret)
 		return ret;
-	}
 
 	ret = hclgevf_cmd_queue_init(hdev);
-	if (ret) {
-		dev_err(&pdev->dev, "Cmd queue init failed: %d\n", ret);
+	if (ret)
 		goto err_cmd_queue_init;
-	}
 
 	ret = hclgevf_cmd_init(hdev);
 	if (ret)
@@ -2728,11 +2724,8 @@ static int hclgevf_init_hdev(struct hclgevf_dev *hdev)
 
 	/* Get vf resource */
 	ret = hclgevf_query_vf_resource(hdev);
-	if (ret) {
-		dev_err(&hdev->pdev->dev,
-			"Query vf status error, ret = %d.\n", ret);
+	if (ret)
 		goto err_cmd_init;
-	}
 
 	ret = hclgevf_init_msi(hdev);
 	if (ret) {
@@ -2745,11 +2738,8 @@ static int hclgevf_init_hdev(struct hclgevf_dev *hdev)
 	hdev->reset_type = HNAE3_NONE_RESET;
 
 	ret = hclgevf_misc_irq_init(hdev);
-	if (ret) {
-		dev_err(&pdev->dev, "failed(%d) to init Misc IRQ(vector0)\n",
-			ret);
+	if (ret)
 		goto err_misc_irq_init;
-	}
 
 	set_bit(HCLGEVF_STATE_IRQ_INITED, &hdev->state);
 
@@ -2766,10 +2756,8 @@ static int hclgevf_init_hdev(struct hclgevf_dev *hdev)
 	}
 
 	ret = hclgevf_set_handle_info(hdev);
-	if (ret) {
-		dev_err(&pdev->dev, "failed(%d) to set handle info\n", ret);
+	if (ret)
 		goto err_config;
-	}
 
 	ret = hclgevf_config_gro(hdev, true);
 	if (ret)
-- 
cgit v1.2.3-59-g8ed1b


From e09538eac996e11d36027cb546a9ac50d07edeb1 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 21 Jan 2020 12:02:25 +0100
Subject: hdlx_x25: Fix backwards compat test.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

drivers/net/wan/hdlc_x25.c: In function ‘x25_ioctl’:
drivers/net/wan/hdlc_x25.c:256:7: warning: suggest parentheses around assignment used as truth value [-Wparentheses]
  256 |   if (ifr->ifr_settings.size = 0) {
      |       ^~~

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/wan/hdlc_x25.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wan/hdlc_x25.c b/drivers/net/wan/hdlc_x25.c
index 63c9aeed9a34..c84536b03aa8 100644
--- a/drivers/net/wan/hdlc_x25.c
+++ b/drivers/net/wan/hdlc_x25.c
@@ -253,7 +253,7 @@ static int x25_ioctl(struct net_device *dev, struct ifreq *ifr)
 			return -EBUSY;
 
 		/* backward compatibility */
-		if (ifr->ifr_settings.size = 0) {
+		if (ifr->ifr_settings.size == 0) {
 			new_settings.dce = 0;
 			new_settings.modulo = 8;
 			new_settings.window = 7;
-- 
cgit v1.2.3-59-g8ed1b


From 339821f7d2d890a61cebeb9d085643668d4df6e4 Mon Sep 17 00:00:00 2001
From: Alex Shi <alex.shi@linux.alibaba.com>
Date: Tue, 21 Jan 2020 16:49:53 +0800
Subject: net/hsr: remove seq_nr_after_or_eq

It's never used after introduced. So maybe better to remove.

Signed-off-by: Alex Shi <alex.shi@linux.alibaba.com>
Cc: Arvid Brodin <arvid.brodin@alten.se>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: netdev@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/hsr/hsr_framereg.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/net/hsr/hsr_framereg.c b/net/hsr/hsr_framereg.c
index 27dc65d7de67..364ea2cc028e 100644
--- a/net/hsr/hsr_framereg.c
+++ b/net/hsr/hsr_framereg.c
@@ -35,7 +35,6 @@ static bool seq_nr_after(u16 a, u16 b)
 }
 
 #define seq_nr_before(a, b)		seq_nr_after((b), (a))
-#define seq_nr_after_or_eq(a, b)	(!seq_nr_before((a), (b)))
 #define seq_nr_before_or_eq(a, b)	(!seq_nr_after((a), (b)))
 
 bool hsr_addr_is_self(struct hsr_priv *hsr, unsigned char *addr)
-- 
cgit v1.2.3-59-g8ed1b


From aeaec7bcebd9c85372191fb42a4478d21972d99e Mon Sep 17 00:00:00 2001
From: Alex Shi <alex.shi@linux.alibaba.com>
Date: Tue, 21 Jan 2020 16:50:07 +0800
Subject: tcp/ipv4: remove AF_INET_FAMILY

After commit 079096f103fa ("tcp/dccp: install syn_recv requests into ehash table")
the macro isn't used anymore. remove it.

Signed-off-by: Alex Shi <alex.shi@linux.alibaba.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Cc: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
Cc: netdev@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/inet_connection_sock.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 6a691fd04398..a4db79b1b643 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -610,12 +610,6 @@ no_route:
 }
 EXPORT_SYMBOL_GPL(inet_csk_route_child_sock);
 
-#if IS_ENABLED(CONFIG_IPV6)
-#define AF_INET_FAMILY(fam) ((fam) == AF_INET)
-#else
-#define AF_INET_FAMILY(fam) true
-#endif
-
 /* Decide when to expire the request and when to resend SYN-ACK */
 static inline void syn_ack_recalc(struct request_sock *req, const int thresh,
 				  const int max_retries,
-- 
cgit v1.2.3-59-g8ed1b


From d84b99ff69c14a03d76e7fdf9c04895848e6aece Mon Sep 17 00:00:00 2001
From: Chen Zhou <chenzhou10@huawei.com>
Date: Tue, 21 Jan 2020 17:24:55 +0800
Subject: drivers: net: declance: fix comparing pointer to 0

Fixes coccicheck warning:

./drivers/net/ethernet/amd/declance.c:611:14-15:
	WARNING comparing pointer to 0

Replace "skb == 0" with "!skb".

Signed-off-by: Chen Zhou <chenzhou10@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/amd/declance.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/amd/declance.c b/drivers/net/ethernet/amd/declance.c
index 6592a2db9efb..7282ce55ffb8 100644
--- a/drivers/net/ethernet/amd/declance.c
+++ b/drivers/net/ethernet/amd/declance.c
@@ -608,7 +608,7 @@ static int lance_rx(struct net_device *dev)
 			len = (*rds_ptr(rd, mblength, lp->type) & 0xfff) - 4;
 			skb = netdev_alloc_skb(dev, len + 2);
 
-			if (skb == 0) {
+			if (!skb) {
 				dev->stats.rx_dropped++;
 				*rds_ptr(rd, mblength, lp->type) = 0;
 				*rds_ptr(rd, rmd1, lp->type) =
-- 
cgit v1.2.3-59-g8ed1b


From b87121dd3fa0a4d2636c0ad02a3b3fc5161fa26b Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 20 Jan 2020 23:28:58 +0000
Subject: bpf: don't bother with getname/kern_path - use user_path_at

kernel/bpf/inode.c misuses kern_path...() - it's much simpler (and
more efficient, on top of that) to use user_path...() counterparts
rather than bothering with doing getname() manually.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20200120232858.GF8904@ZenIV.linux.org.uk
---
 kernel/bpf/inode.c | 43 +++++++++++++------------------------------
 1 file changed, 13 insertions(+), 30 deletions(-)

diff --git a/kernel/bpf/inode.c b/kernel/bpf/inode.c
index ecf42bec38c0..e11059b99f18 100644
--- a/kernel/bpf/inode.c
+++ b/kernel/bpf/inode.c
@@ -380,7 +380,7 @@ static const struct inode_operations bpf_dir_iops = {
 	.unlink		= simple_unlink,
 };
 
-static int bpf_obj_do_pin(const struct filename *pathname, void *raw,
+static int bpf_obj_do_pin(const char __user *pathname, void *raw,
 			  enum bpf_type type)
 {
 	struct dentry *dentry;
@@ -389,7 +389,7 @@ static int bpf_obj_do_pin(const struct filename *pathname, void *raw,
 	umode_t mode;
 	int ret;
 
-	dentry = kern_path_create(AT_FDCWD, pathname->name, &path, 0);
+	dentry = user_path_create(AT_FDCWD, pathname, &path, 0);
 	if (IS_ERR(dentry))
 		return PTR_ERR(dentry);
 
@@ -422,30 +422,22 @@ out:
 
 int bpf_obj_pin_user(u32 ufd, const char __user *pathname)
 {
-	struct filename *pname;
 	enum bpf_type type;
 	void *raw;
 	int ret;
 
-	pname = getname(pathname);
-	if (IS_ERR(pname))
-		return PTR_ERR(pname);
-
 	raw = bpf_fd_probe_obj(ufd, &type);
-	if (IS_ERR(raw)) {
-		ret = PTR_ERR(raw);
-		goto out;
-	}
+	if (IS_ERR(raw))
+		return PTR_ERR(raw);
 
-	ret = bpf_obj_do_pin(pname, raw, type);
+	ret = bpf_obj_do_pin(pathname, raw, type);
 	if (ret != 0)
 		bpf_any_put(raw, type);
-out:
-	putname(pname);
+
 	return ret;
 }
 
-static void *bpf_obj_do_get(const struct filename *pathname,
+static void *bpf_obj_do_get(const char __user *pathname,
 			    enum bpf_type *type, int flags)
 {
 	struct inode *inode;
@@ -453,7 +445,7 @@ static void *bpf_obj_do_get(const struct filename *pathname,
 	void *raw;
 	int ret;
 
-	ret = kern_path(pathname->name, LOOKUP_FOLLOW, &path);
+	ret = user_path_at(AT_FDCWD, pathname, LOOKUP_FOLLOW, &path);
 	if (ret)
 		return ERR_PTR(ret);
 
@@ -480,36 +472,27 @@ out:
 int bpf_obj_get_user(const char __user *pathname, int flags)
 {
 	enum bpf_type type = BPF_TYPE_UNSPEC;
-	struct filename *pname;
-	int ret = -ENOENT;
 	int f_flags;
 	void *raw;
+	int ret;
 
 	f_flags = bpf_get_file_flag(flags);
 	if (f_flags < 0)
 		return f_flags;
 
-	pname = getname(pathname);
-	if (IS_ERR(pname))
-		return PTR_ERR(pname);
-
-	raw = bpf_obj_do_get(pname, &type, f_flags);
-	if (IS_ERR(raw)) {
-		ret = PTR_ERR(raw);
-		goto out;
-	}
+	raw = bpf_obj_do_get(pathname, &type, f_flags);
+	if (IS_ERR(raw))
+		return PTR_ERR(raw);
 
 	if (type == BPF_TYPE_PROG)
 		ret = bpf_prog_new_fd(raw);
 	else if (type == BPF_TYPE_MAP)
 		ret = bpf_map_new_fd(raw, f_flags);
 	else
-		goto out;
+		return -ENOENT;
 
 	if (ret < 0)
 		bpf_any_put(raw, type);
-out:
-	putname(pname);
 	return ret;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 43a825afc91e2b06af1e8e7422198e759c2c5e20 Mon Sep 17 00:00:00 2001
From: Björn Töpel <bjorn.topel@intel.com>
Date: Mon, 20 Jan 2020 10:29:17 +0100
Subject: xsk, net: Make sock_def_readable() have external linkage
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

XDP sockets use the default implementation of struct sock's
sk_data_ready callback, which is sock_def_readable(). This function
is called in the XDP socket fast-path, and involves a retpoline. By
letting sock_def_readable() have external linkage, and being called
directly, the retpoline can be avoided.

Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20200120092917.13949-1-bjorn.topel@gmail.com
---
 include/net/sock.h | 2 ++
 net/core/sock.c    | 2 +-
 net/xdp/xsk.c      | 2 +-
 3 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/include/net/sock.h b/include/net/sock.h
index 8dff68b4c316..0891c55f1e82 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -2612,4 +2612,6 @@ static inline bool sk_dev_equal_l3scope(struct sock *sk, int dif)
 	return false;
 }
 
+void sock_def_readable(struct sock *sk);
+
 #endif	/* _SOCK_H */
diff --git a/net/core/sock.c b/net/core/sock.c
index 8459ad579f73..a4c8fac781ff 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -2786,7 +2786,7 @@ static void sock_def_error_report(struct sock *sk)
 	rcu_read_unlock();
 }
 
-static void sock_def_readable(struct sock *sk)
+void sock_def_readable(struct sock *sk)
 {
 	struct socket_wq *wq;
 
diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c
index 02ada7ab8c6e..df600487a68d 100644
--- a/net/xdp/xsk.c
+++ b/net/xdp/xsk.c
@@ -217,7 +217,7 @@ static int xsk_rcv(struct xdp_sock *xs, struct xdp_buff *xdp)
 static void xsk_flush(struct xdp_sock *xs)
 {
 	xskq_prod_submit(xs->rx);
-	xs->sk.sk_data_ready(&xs->sk);
+	sock_def_readable(&xs->sk);
 }
 
 int xsk_generic_rcv(struct xdp_sock *xs, struct xdp_buff *xdp)
-- 
cgit v1.2.3-59-g8ed1b


From 05d57f1793fb250c85028c9952c3720010baa853 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Mon, 20 Jan 2020 19:22:31 -0800
Subject: bpf: Fix trampoline usage in preempt

Though the second half of trampoline page is unused a task could be
preempted in the middle of the first half of trampoline and two
updates to trampoline would change the code from underneath the
preempted task. Hence wait for tasks to voluntarily schedule or go
to userspace. Add similar wait before freeing the trampoline.

Fixes: fec56f5890d9 ("bpf: Introduce BPF trampoline")
Reported-by: Jann Horn <jannh@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Paul E. McKenney <paulmck@kernel.org>
Link: https://lore.kernel.org/bpf/20200121032231.3292185-1-ast@kernel.org
---
 kernel/bpf/trampoline.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/kernel/bpf/trampoline.c b/kernel/bpf/trampoline.c
index 79a04417050d..7657ede7aee2 100644
--- a/kernel/bpf/trampoline.c
+++ b/kernel/bpf/trampoline.c
@@ -160,6 +160,14 @@ static int bpf_trampoline_update(struct bpf_trampoline *tr)
 	if (fexit_cnt)
 		flags = BPF_TRAMP_F_CALL_ORIG | BPF_TRAMP_F_SKIP_FRAME;
 
+	/* Though the second half of trampoline page is unused a task could be
+	 * preempted in the middle of the first half of trampoline and two
+	 * updates to trampoline would change the code from underneath the
+	 * preempted task. Hence wait for tasks to voluntarily schedule or go
+	 * to userspace.
+	 */
+	synchronize_rcu_tasks();
+
 	err = arch_prepare_bpf_trampoline(new_image, new_image + PAGE_SIZE / 2,
 					  &tr->func.model, flags,
 					  fentry, fentry_cnt,
@@ -251,6 +259,8 @@ void bpf_trampoline_put(struct bpf_trampoline *tr)
 		goto out;
 	if (WARN_ON_ONCE(!hlist_empty(&tr->progs_hlist[BPF_TRAMP_FEXIT])))
 		goto out;
+	/* wait for tasks to get out of trampoline before freeing it */
+	synchronize_rcu_tasks();
 	bpf_jit_free_exec(tr->image);
 	hlist_del(&tr->hlist);
 	kfree(tr);
-- 
cgit v1.2.3-59-g8ed1b


From f59bbfc2f6099e8655f9e8f585e10ffde17176d0 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Tue, 21 Jan 2020 18:41:38 -0800
Subject: bpf: Fix error path under memory pressure

Restore the 'if (env->cur_state)' check that was incorrectly removed during
code move. Under memory pressure env->cur_state can be freed and zeroed inside
do_check(). Hence the check is necessary.

Fixes: 51c39bb1d5d1 ("bpf: Introduce function-by-function verification")
Reported-by: syzbot+b296579ba5015704d9fa@syzkaller.appspotmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Song Liu <songliubraving@fb.com>
Link: https://lore.kernel.org/bpf/20200122024138.3385590-1-ast@kernel.org
---
 kernel/bpf/verifier.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index ca17dccc17ba..6defbec9eb62 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -9594,8 +9594,13 @@ static int do_check_common(struct bpf_verifier_env *env, int subprog)
 
 	ret = do_check(env);
 out:
-	free_verifier_state(env->cur_state, true);
-	env->cur_state = NULL;
+	/* check for NULL is necessary, since cur_state can be freed inside
+	 * do_check() under memory pressure.
+	 */
+	if (env->cur_state) {
+		free_verifier_state(env->cur_state, true);
+		env->cur_state = NULL;
+	}
 	while (!pop_stack(env, NULL, NULL));
 	free_states(env);
 	if (ret)
-- 
cgit v1.2.3-59-g8ed1b


From bdf2aca703e83eeecac2b492494687d5009a694e Mon Sep 17 00:00:00 2001
From: Alain Michaud <alainm@chromium.org>
Date: Wed, 22 Jan 2020 16:09:16 +0000
Subject: Bluetooth: adding missing const decoration to mgmt_status_table

This change simply adds a missing const decoration to the
mtmt_status_table definition.

Signed-off-by: Alain Michaud <alainm@chromium.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 net/bluetooth/mgmt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c
index 0dc610faab70..3c68a366977f 100644
--- a/net/bluetooth/mgmt.c
+++ b/net/bluetooth/mgmt.c
@@ -176,7 +176,7 @@ static const u16 mgmt_untrusted_events[] = {
 		 "\x00\x00\x00\x00\x00\x00\x00\x00"
 
 /* HCI to MGMT error code conversion table */
-static u8 mgmt_status_table[] = {
+static const u8 mgmt_status_table[] = {
 	MGMT_STATUS_SUCCESS,
 	MGMT_STATUS_UNKNOWN_COMMAND,	/* Unknown Command */
 	MGMT_STATUS_NOT_CONNECTED,	/* No Connection */
-- 
cgit v1.2.3-59-g8ed1b


From 1222653cc0489fa9304a36f7d9593ecaeb51efe1 Mon Sep 17 00:00:00 2001
From: Daniel Díaz <daniel.diaz@linaro.org>
Date: Wed, 22 Jan 2020 17:44:24 +0100
Subject: selftests/bpf: Build urandom_read with LDFLAGS and LDLIBS
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

During cross-compilation, it was discovered that LDFLAGS and
LDLIBS were not being used while building binaries, leading
to defaults which were not necessarily correct.

OpenEmbedded reported this kind of problem:

  ERROR: QA Issue: No GNU_HASH in the ELF binary [...], didn't pass LDFLAGS?

Signed-off-by: Daniel Díaz <daniel.diaz@linaro.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
---
 tools/testing/selftests/bpf/Makefile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index be8fe404a086..5f41f5bd8033 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -122,7 +122,7 @@ $(OUTPUT)/%:%.c
 
 $(OUTPUT)/urandom_read: urandom_read.c
 	$(call msg,BINARY,,$@)
-	$(CC) -o $@ $< -Wl,--build-id
+	$(CC) $(LDFLAGS) -o $@ $< $(LDLIBS) -Wl,--build-id
 
 $(OUTPUT)/test_stub.o: test_stub.c $(BPFOBJ)
 	$(call msg,CC,,$@)
-- 
cgit v1.2.3-59-g8ed1b


From 2a67a6ccb01f21b854715d86ff6432a18b97adb3 Mon Sep 17 00:00:00 2001
From: Chris Down <chris@chrisdown.name>
Date: Wed, 22 Jan 2020 00:01:10 +0000
Subject: bpf, btf: Always output invariant hit in pahole DWARF to BTF
 transform

When trying to compile with CONFIG_DEBUG_INFO_BTF enabled, I got this
error:

    % make -s
    Failed to generate BTF for vmlinux
    Try to disable CONFIG_DEBUG_INFO_BTF
    make[3]: *** [vmlinux] Error 1

Compiling again without -s shows the true error (that pahole is
missing), but since this is fatal, we should show the error
unconditionally on stderr as well, not silence it using the `info`
function. With this patch:

    % make -s
    BTF: .tmp_vmlinux.btf: pahole (pahole) is not available
    Failed to generate BTF for vmlinux
    Try to disable CONFIG_DEBUG_INFO_BTF
    make[3]: *** [vmlinux] Error 1

Signed-off-by: Chris Down <chris@chrisdown.name>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/20200122000110.GA310073@chrisdown.name
---
 scripts/link-vmlinux.sh | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/scripts/link-vmlinux.sh b/scripts/link-vmlinux.sh
index 436379940356..408b5c0b99b1 100755
--- a/scripts/link-vmlinux.sh
+++ b/scripts/link-vmlinux.sh
@@ -108,13 +108,13 @@ gen_btf()
 	local bin_arch
 
 	if ! [ -x "$(command -v ${PAHOLE})" ]; then
-		info "BTF" "${1}: pahole (${PAHOLE}) is not available"
+		echo >&2 "BTF: ${1}: pahole (${PAHOLE}) is not available"
 		return 1
 	fi
 
 	pahole_ver=$(${PAHOLE} --version | sed -E 's/v([0-9]+)\.([0-9]+)/\1\2/')
 	if [ "${pahole_ver}" -lt "113" ]; then
-		info "BTF" "${1}: pahole version $(${PAHOLE} --version) is too old, need at least v1.13"
+		echo >&2 "BTF: ${1}: pahole version $(${PAHOLE} --version) is too old, need at least v1.13"
 		return 1
 	fi
 
-- 
cgit v1.2.3-59-g8ed1b


From c5d19a6ecfce72d0352191d75f03eea4748a8c45 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Tue, 21 Jan 2020 22:05:14 +0100
Subject: net: convert additional drivers to use phy_do_ioctl

The first batch of driver conversions missed a few cases where we can
use phy_do_ioctl too.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/aurora/nb8800.c             |  7 +------
 drivers/net/ethernet/cavium/octeon/octeon_mgmt.c |  4 +---
 drivers/net/ethernet/lantiq_etop.c               |  9 +--------
 drivers/net/ethernet/marvell/pxa168_eth.c        | 11 +----------
 drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c  |  4 +---
 drivers/net/ethernet/socionext/netsec.c          |  8 +-------
 6 files changed, 6 insertions(+), 37 deletions(-)

diff --git a/drivers/net/ethernet/aurora/nb8800.c b/drivers/net/ethernet/aurora/nb8800.c
index 30b455013bf3..bc273e0db7ff 100644
--- a/drivers/net/ethernet/aurora/nb8800.c
+++ b/drivers/net/ethernet/aurora/nb8800.c
@@ -1005,18 +1005,13 @@ static int nb8800_stop(struct net_device *dev)
 	return 0;
 }
 
-static int nb8800_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
-{
-	return phy_mii_ioctl(dev->phydev, rq, cmd);
-}
-
 static const struct net_device_ops nb8800_netdev_ops = {
 	.ndo_open		= nb8800_open,
 	.ndo_stop		= nb8800_stop,
 	.ndo_start_xmit		= nb8800_xmit,
 	.ndo_set_mac_address	= nb8800_set_mac_address,
 	.ndo_set_rx_mode	= nb8800_set_rx_mode,
-	.ndo_do_ioctl		= nb8800_ioctl,
+	.ndo_do_ioctl		= phy_do_ioctl,
 	.ndo_validate_addr	= eth_validate_addr,
 };
 
diff --git a/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c b/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c
index cdd7e5da4a74..e9575887a4f8 100644
--- a/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c
+++ b/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c
@@ -790,9 +790,7 @@ static int octeon_mgmt_ioctl(struct net_device *netdev,
 	case SIOCSHWTSTAMP:
 		return octeon_mgmt_ioctl_hwtstamp(netdev, rq, cmd);
 	default:
-		if (netdev->phydev)
-			return phy_mii_ioctl(netdev->phydev, rq, cmd);
-		return -EINVAL;
+		return phy_do_ioctl(netdev, rq, cmd);
 	}
 }
 
diff --git a/drivers/net/ethernet/lantiq_etop.c b/drivers/net/ethernet/lantiq_etop.c
index 028e3e6222e9..578c31697cc0 100644
--- a/drivers/net/ethernet/lantiq_etop.c
+++ b/drivers/net/ethernet/lantiq_etop.c
@@ -509,13 +509,6 @@ ltq_etop_change_mtu(struct net_device *dev, int new_mtu)
 	return 0;
 }
 
-static int
-ltq_etop_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
-{
-	/* TODO: mii-toll reports "No MII transceiver present!." ?!*/
-	return phy_mii_ioctl(dev->phydev, rq, cmd);
-}
-
 static int
 ltq_etop_set_mac_address(struct net_device *dev, void *p)
 {
@@ -616,7 +609,7 @@ static const struct net_device_ops ltq_eth_netdev_ops = {
 	.ndo_stop = ltq_etop_stop,
 	.ndo_start_xmit = ltq_etop_tx,
 	.ndo_change_mtu = ltq_etop_change_mtu,
-	.ndo_do_ioctl = ltq_etop_ioctl,
+	.ndo_do_ioctl = phy_do_ioctl,
 	.ndo_set_mac_address = ltq_etop_set_mac_address,
 	.ndo_validate_addr = eth_validate_addr,
 	.ndo_set_rx_mode = ltq_etop_set_multicast_list,
diff --git a/drivers/net/ethernet/marvell/pxa168_eth.c b/drivers/net/ethernet/marvell/pxa168_eth.c
index 1a6877902dd6..7a0d785b826c 100644
--- a/drivers/net/ethernet/marvell/pxa168_eth.c
+++ b/drivers/net/ethernet/marvell/pxa168_eth.c
@@ -1344,15 +1344,6 @@ static int pxa168_smi_write(struct mii_bus *bus, int phy_addr, int regnum,
 	return 0;
 }
 
-static int pxa168_eth_do_ioctl(struct net_device *dev, struct ifreq *ifr,
-			       int cmd)
-{
-	if (dev->phydev)
-		return phy_mii_ioctl(dev->phydev, ifr, cmd);
-
-	return -EOPNOTSUPP;
-}
-
 #ifdef CONFIG_NET_POLL_CONTROLLER
 static void pxa168_eth_netpoll(struct net_device *dev)
 {
@@ -1387,7 +1378,7 @@ static const struct net_device_ops pxa168_eth_netdev_ops = {
 	.ndo_set_rx_mode	= pxa168_eth_set_rx_mode,
 	.ndo_set_mac_address	= pxa168_eth_set_mac_address,
 	.ndo_validate_addr	= eth_validate_addr,
-	.ndo_do_ioctl		= pxa168_eth_do_ioctl,
+	.ndo_do_ioctl		= phy_do_ioctl,
 	.ndo_change_mtu		= pxa168_eth_change_mtu,
 	.ndo_tx_timeout		= pxa168_eth_tx_timeout,
 #ifdef CONFIG_NET_POLL_CONTROLLER
diff --git a/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c b/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c
index 7d3a1c0df09c..c705743d69f7 100644
--- a/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c
+++ b/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c
@@ -1939,9 +1939,7 @@ static int sxgbe_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
 	case SIOCGMIIPHY:
 	case SIOCGMIIREG:
 	case SIOCSMIIREG:
-		if (!dev->phydev)
-			return -EINVAL;
-		ret = phy_mii_ioctl(dev->phydev, rq, cmd);
+		ret = phy_do_ioctl(dev, rq, cmd);
 		break;
 	default:
 		break;
diff --git a/drivers/net/ethernet/socionext/netsec.c b/drivers/net/ethernet/socionext/netsec.c
index 6870a6ce76a6..495f6cfdbd91 100644
--- a/drivers/net/ethernet/socionext/netsec.c
+++ b/drivers/net/ethernet/socionext/netsec.c
@@ -1740,12 +1740,6 @@ static int netsec_netdev_set_features(struct net_device *ndev,
 	return 0;
 }
 
-static int netsec_netdev_ioctl(struct net_device *ndev, struct ifreq *ifr,
-			       int cmd)
-{
-	return phy_mii_ioctl(ndev->phydev, ifr, cmd);
-}
-
 static int netsec_xdp_xmit(struct net_device *ndev, int n,
 			   struct xdp_frame **frames, u32 flags)
 {
@@ -1830,7 +1824,7 @@ static const struct net_device_ops netsec_netdev_ops = {
 	.ndo_set_features	= netsec_netdev_set_features,
 	.ndo_set_mac_address    = eth_mac_addr,
 	.ndo_validate_addr	= eth_validate_addr,
-	.ndo_do_ioctl		= netsec_netdev_ioctl,
+	.ndo_do_ioctl		= phy_do_ioctl,
 	.ndo_xdp_xmit		= netsec_xdp_xmit,
 	.ndo_bpf		= netsec_xdp,
 };
-- 
cgit v1.2.3-59-g8ed1b


From 6613babaf662cfbd283fcd0abdd758e338dd181d Mon Sep 17 00:00:00 2001
From: Alain Michaud <alainm@chromium.org>
Date: Wed, 22 Jan 2020 19:47:44 +0000
Subject: Bluetooth: fix appearance typo in mgmt.c

This change addresses a typo in the set_appearance handler.

Signed-off-by: Alain Michaud <alainm@chromium.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 net/bluetooth/mgmt.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c
index 3c68a366977f..3074363c68df 100644
--- a/net/bluetooth/mgmt.c
+++ b/net/bluetooth/mgmt.c
@@ -3291,7 +3291,7 @@ static int set_appearance(struct sock *sk, struct hci_dev *hdev, void *data,
 			  u16 len)
 {
 	struct mgmt_cp_set_appearance *cp = data;
-	u16 apperance;
+	u16 appearance;
 	int err;
 
 	BT_DBG("");
@@ -3300,12 +3300,12 @@ static int set_appearance(struct sock *sk, struct hci_dev *hdev, void *data,
 		return mgmt_cmd_status(sk, hdev->id, MGMT_OP_SET_APPEARANCE,
 				       MGMT_STATUS_NOT_SUPPORTED);
 
-	apperance = le16_to_cpu(cp->appearance);
+	appearance = le16_to_cpu(cp->appearance);
 
 	hci_dev_lock(hdev);
 
-	if (hdev->appearance != apperance) {
-		hdev->appearance = apperance;
+	if (hdev->appearance != appearance) {
+		hdev->appearance = appearance;
 
 		if (hci_dev_test_flag(hdev, HCI_LE_ADV))
 			adv_expire(hdev, MGMT_ADV_FLAG_APPEARANCE);
-- 
cgit v1.2.3-59-g8ed1b


From be8704ff07d2374bcc5c675526f95e70c6459683 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Mon, 20 Jan 2020 16:53:46 -0800
Subject: bpf: Introduce dynamic program extensions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Introduce dynamic program extensions. The users can load additional BPF
functions and replace global functions in previously loaded BPF programs while
these programs are executing.

Global functions are verified individually by the verifier based on their types only.
Hence the global function in the new program which types match older function can
safely replace that corresponding function.

This new function/program is called 'an extension' of old program. At load time
the verifier uses (attach_prog_fd, attach_btf_id) pair to identify the function
to be replaced. The BPF program type is derived from the target program into
extension program. Technically bpf_verifier_ops is copied from target program.
The BPF_PROG_TYPE_EXT program type is a placeholder. It has empty verifier_ops.
The extension program can call the same bpf helper functions as target program.
Single BPF_PROG_TYPE_EXT type is used to extend XDP, SKB and all other program
types. The verifier allows only one level of replacement. Meaning that the
extension program cannot recursively extend an extension. That also means that
the maximum stack size is increasing from 512 to 1024 bytes and maximum
function nesting level from 8 to 16. The programs don't always consume that
much. The stack usage is determined by the number of on-stack variables used by
the program. The verifier could have enforced 512 limit for combined original
plus extension program, but it makes for difficult user experience. The main
use case for extensions is to provide generic mechanism to plug external
programs into policy program or function call chaining.

BPF trampoline is used to track both fentry/fexit and program extensions
because both are using the same nop slot at the beginning of every BPF
function. Attaching fentry/fexit to a function that was replaced is not
allowed. The opposite is true as well. Replacing a function that currently
being analyzed with fentry/fexit is not allowed. The executable page allocated
by BPF trampoline is not used by program extensions. This inefficiency will be
optimized in future patches.

Function by function verification of global function supports scalars and
pointer to context only. Hence program extensions are supported for such class
of global functions only. In the future the verifier will be extended with
support to pointers to structures, arrays with sizes, etc.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/bpf/20200121005348.2769920-2-ast@kernel.org
---
 include/linux/bpf.h       |  10 ++-
 include/linux/bpf_types.h |   2 +
 include/linux/btf.h       |   5 ++
 include/uapi/linux/bpf.h  |   1 +
 kernel/bpf/btf.c          | 152 +++++++++++++++++++++++++++++++++++++++++++++-
 kernel/bpf/syscall.c      |  15 ++++-
 kernel/bpf/trampoline.c   |  41 ++++++++++++-
 kernel/bpf/verifier.c     |  85 ++++++++++++++++++++------
 8 files changed, 283 insertions(+), 28 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 8e3b8f4ad183..05d16615054c 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -465,7 +465,8 @@ void notrace __bpf_prog_exit(struct bpf_prog *prog, u64 start);
 enum bpf_tramp_prog_type {
 	BPF_TRAMP_FENTRY,
 	BPF_TRAMP_FEXIT,
-	BPF_TRAMP_MAX
+	BPF_TRAMP_MAX,
+	BPF_TRAMP_REPLACE, /* more than MAX */
 };
 
 struct bpf_trampoline {
@@ -480,6 +481,11 @@ struct bpf_trampoline {
 		void *addr;
 		bool ftrace_managed;
 	} func;
+	/* if !NULL this is BPF_PROG_TYPE_EXT program that extends another BPF
+	 * program by replacing one of its functions. func.addr is the address
+	 * of the function it replaced.
+	 */
+	struct bpf_prog *extension_prog;
 	/* list of BPF programs using this trampoline */
 	struct hlist_head progs_hlist[BPF_TRAMP_MAX];
 	/* Number of attached programs. A counter per kind. */
@@ -1107,6 +1113,8 @@ int btf_check_func_arg_match(struct bpf_verifier_env *env, int subprog,
 			     struct bpf_reg_state *regs);
 int btf_prepare_func_args(struct bpf_verifier_env *env, int subprog,
 			  struct bpf_reg_state *reg);
+int btf_check_type_match(struct bpf_verifier_env *env, struct bpf_prog *prog,
+			 struct btf *btf, const struct btf_type *t);
 
 struct bpf_prog *bpf_prog_by_id(u32 id);
 
diff --git a/include/linux/bpf_types.h b/include/linux/bpf_types.h
index 9f326e6ef885..c81d4ece79a4 100644
--- a/include/linux/bpf_types.h
+++ b/include/linux/bpf_types.h
@@ -68,6 +68,8 @@ BPF_PROG_TYPE(BPF_PROG_TYPE_SK_REUSEPORT, sk_reuseport,
 #if defined(CONFIG_BPF_JIT)
 BPF_PROG_TYPE(BPF_PROG_TYPE_STRUCT_OPS, bpf_struct_ops,
 	      void *, void *)
+BPF_PROG_TYPE(BPF_PROG_TYPE_EXT, bpf_extension,
+	      void *, void *)
 #endif
 
 BPF_MAP_TYPE(BPF_MAP_TYPE_ARRAY, array_map_ops)
diff --git a/include/linux/btf.h b/include/linux/btf.h
index 881e9b76ef49..5c1ea99b480f 100644
--- a/include/linux/btf.h
+++ b/include/linux/btf.h
@@ -107,6 +107,11 @@ static inline u16 btf_type_vlen(const struct btf_type *t)
 	return BTF_INFO_VLEN(t->info);
 }
 
+static inline u16 btf_func_linkage(const struct btf_type *t)
+{
+	return BTF_INFO_VLEN(t->info);
+}
+
 static inline bool btf_type_kflag(const struct btf_type *t)
 {
 	return BTF_INFO_KFLAG(t->info);
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 033d90a2282d..e81628eb059c 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -180,6 +180,7 @@ enum bpf_prog_type {
 	BPF_PROG_TYPE_CGROUP_SOCKOPT,
 	BPF_PROG_TYPE_TRACING,
 	BPF_PROG_TYPE_STRUCT_OPS,
+	BPF_PROG_TYPE_EXT,
 };
 
 enum bpf_attach_type {
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 832b5d7fd892..32963b6d5a9c 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -276,6 +276,11 @@ static const char * const btf_kind_str[NR_BTF_KINDS] = {
 	[BTF_KIND_DATASEC]	= "DATASEC",
 };
 
+static const char *btf_type_str(const struct btf_type *t)
+{
+	return btf_kind_str[BTF_INFO_KIND(t->info)];
+}
+
 struct btf_kind_operations {
 	s32 (*check_meta)(struct btf_verifier_env *env,
 			  const struct btf_type *t,
@@ -4115,6 +4120,148 @@ int btf_distill_func_proto(struct bpf_verifier_log *log,
 	return 0;
 }
 
+/* Compare BTFs of two functions assuming only scalars and pointers to context.
+ * t1 points to BTF_KIND_FUNC in btf1
+ * t2 points to BTF_KIND_FUNC in btf2
+ * Returns:
+ * EINVAL - function prototype mismatch
+ * EFAULT - verifier bug
+ * 0 - 99% match. The last 1% is validated by the verifier.
+ */
+int btf_check_func_type_match(struct bpf_verifier_log *log,
+			      struct btf *btf1, const struct btf_type *t1,
+			      struct btf *btf2, const struct btf_type *t2)
+{
+	const struct btf_param *args1, *args2;
+	const char *fn1, *fn2, *s1, *s2;
+	u32 nargs1, nargs2, i;
+
+	fn1 = btf_name_by_offset(btf1, t1->name_off);
+	fn2 = btf_name_by_offset(btf2, t2->name_off);
+
+	if (btf_func_linkage(t1) != BTF_FUNC_GLOBAL) {
+		bpf_log(log, "%s() is not a global function\n", fn1);
+		return -EINVAL;
+	}
+	if (btf_func_linkage(t2) != BTF_FUNC_GLOBAL) {
+		bpf_log(log, "%s() is not a global function\n", fn2);
+		return -EINVAL;
+	}
+
+	t1 = btf_type_by_id(btf1, t1->type);
+	if (!t1 || !btf_type_is_func_proto(t1))
+		return -EFAULT;
+	t2 = btf_type_by_id(btf2, t2->type);
+	if (!t2 || !btf_type_is_func_proto(t2))
+		return -EFAULT;
+
+	args1 = (const struct btf_param *)(t1 + 1);
+	nargs1 = btf_type_vlen(t1);
+	args2 = (const struct btf_param *)(t2 + 1);
+	nargs2 = btf_type_vlen(t2);
+
+	if (nargs1 != nargs2) {
+		bpf_log(log, "%s() has %d args while %s() has %d args\n",
+			fn1, nargs1, fn2, nargs2);
+		return -EINVAL;
+	}
+
+	t1 = btf_type_skip_modifiers(btf1, t1->type, NULL);
+	t2 = btf_type_skip_modifiers(btf2, t2->type, NULL);
+	if (t1->info != t2->info) {
+		bpf_log(log,
+			"Return type %s of %s() doesn't match type %s of %s()\n",
+			btf_type_str(t1), fn1,
+			btf_type_str(t2), fn2);
+		return -EINVAL;
+	}
+
+	for (i = 0; i < nargs1; i++) {
+		t1 = btf_type_skip_modifiers(btf1, args1[i].type, NULL);
+		t2 = btf_type_skip_modifiers(btf2, args2[i].type, NULL);
+
+		if (t1->info != t2->info) {
+			bpf_log(log, "arg%d in %s() is %s while %s() has %s\n",
+				i, fn1, btf_type_str(t1),
+				fn2, btf_type_str(t2));
+			return -EINVAL;
+		}
+		if (btf_type_has_size(t1) && t1->size != t2->size) {
+			bpf_log(log,
+				"arg%d in %s() has size %d while %s() has %d\n",
+				i, fn1, t1->size,
+				fn2, t2->size);
+			return -EINVAL;
+		}
+
+		/* global functions are validated with scalars and pointers
+		 * to context only. And only global functions can be replaced.
+		 * Hence type check only those types.
+		 */
+		if (btf_type_is_int(t1) || btf_type_is_enum(t1))
+			continue;
+		if (!btf_type_is_ptr(t1)) {
+			bpf_log(log,
+				"arg%d in %s() has unrecognized type\n",
+				i, fn1);
+			return -EINVAL;
+		}
+		t1 = btf_type_skip_modifiers(btf1, t1->type, NULL);
+		t2 = btf_type_skip_modifiers(btf2, t2->type, NULL);
+		if (!btf_type_is_struct(t1)) {
+			bpf_log(log,
+				"arg%d in %s() is not a pointer to context\n",
+				i, fn1);
+			return -EINVAL;
+		}
+		if (!btf_type_is_struct(t2)) {
+			bpf_log(log,
+				"arg%d in %s() is not a pointer to context\n",
+				i, fn2);
+			return -EINVAL;
+		}
+		/* This is an optional check to make program writing easier.
+		 * Compare names of structs and report an error to the user.
+		 * btf_prepare_func_args() already checked that t2 struct
+		 * is a context type. btf_prepare_func_args() will check
+		 * later that t1 struct is a context type as well.
+		 */
+		s1 = btf_name_by_offset(btf1, t1->name_off);
+		s2 = btf_name_by_offset(btf2, t2->name_off);
+		if (strcmp(s1, s2)) {
+			bpf_log(log,
+				"arg%d %s(struct %s *) doesn't match %s(struct %s *)\n",
+				i, fn1, s1, fn2, s2);
+			return -EINVAL;
+		}
+	}
+	return 0;
+}
+
+/* Compare BTFs of given program with BTF of target program */
+int btf_check_type_match(struct bpf_verifier_env *env, struct bpf_prog *prog,
+			 struct btf *btf2, const struct btf_type *t2)
+{
+	struct btf *btf1 = prog->aux->btf;
+	const struct btf_type *t1;
+	u32 btf_id = 0;
+
+	if (!prog->aux->func_info) {
+		bpf_log(&env->log, "Program extension requires BTF\n");
+		return -EINVAL;
+	}
+
+	btf_id = prog->aux->func_info[0].type_id;
+	if (!btf_id)
+		return -EFAULT;
+
+	t1 = btf_type_by_id(btf1, btf_id);
+	if (!t1 || !btf_type_is_func(t1))
+		return -EFAULT;
+
+	return btf_check_func_type_match(&env->log, btf1, t1, btf2, t2);
+}
+
 /* Compare BTF of a function with given bpf_reg_state.
  * Returns:
  * EFAULT - there is a verifier bug. Abort verification.
@@ -4224,6 +4371,7 @@ int btf_prepare_func_args(struct bpf_verifier_env *env, int subprog,
 {
 	struct bpf_verifier_log *log = &env->log;
 	struct bpf_prog *prog = env->prog;
+	enum bpf_prog_type prog_type = prog->type;
 	struct btf *btf = prog->aux->btf;
 	const struct btf_param *args;
 	const struct btf_type *t;
@@ -4261,6 +4409,8 @@ int btf_prepare_func_args(struct bpf_verifier_env *env, int subprog,
 		bpf_log(log, "Verifier bug in function %s()\n", tname);
 		return -EFAULT;
 	}
+	if (prog_type == BPF_PROG_TYPE_EXT)
+		prog_type = prog->aux->linked_prog->type;
 
 	t = btf_type_by_id(btf, t->type);
 	if (!t || !btf_type_is_func_proto(t)) {
@@ -4296,7 +4446,7 @@ int btf_prepare_func_args(struct bpf_verifier_env *env, int subprog,
 			continue;
 		}
 		if (btf_type_is_ptr(t) &&
-		    btf_get_prog_ctx_type(log, btf, t, prog->type, i)) {
+		    btf_get_prog_ctx_type(log, btf, t, prog_type, i)) {
 			reg[i + 1].type = PTR_TO_CTX;
 			continue;
 		}
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 9a840c57f6df..a91ad518c050 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -1932,13 +1932,15 @@ bpf_prog_load_check_attach(enum bpf_prog_type prog_type,
 		switch (prog_type) {
 		case BPF_PROG_TYPE_TRACING:
 		case BPF_PROG_TYPE_STRUCT_OPS:
+		case BPF_PROG_TYPE_EXT:
 			break;
 		default:
 			return -EINVAL;
 		}
 	}
 
-	if (prog_fd && prog_type != BPF_PROG_TYPE_TRACING)
+	if (prog_fd && prog_type != BPF_PROG_TYPE_TRACING &&
+	    prog_type != BPF_PROG_TYPE_EXT)
 		return -EINVAL;
 
 	switch (prog_type) {
@@ -1981,6 +1983,10 @@ bpf_prog_load_check_attach(enum bpf_prog_type prog_type,
 		default:
 			return -EINVAL;
 		}
+	case BPF_PROG_TYPE_EXT:
+		if (expected_attach_type)
+			return -EINVAL;
+		/* fallthrough */
 	default:
 		return 0;
 	}
@@ -2183,7 +2189,8 @@ static int bpf_tracing_prog_attach(struct bpf_prog *prog)
 	int tr_fd, err;
 
 	if (prog->expected_attach_type != BPF_TRACE_FENTRY &&
-	    prog->expected_attach_type != BPF_TRACE_FEXIT) {
+	    prog->expected_attach_type != BPF_TRACE_FEXIT &&
+	    prog->type != BPF_PROG_TYPE_EXT) {
 		err = -EINVAL;
 		goto out_put_prog;
 	}
@@ -2250,12 +2257,14 @@ static int bpf_raw_tracepoint_open(const union bpf_attr *attr)
 
 	if (prog->type != BPF_PROG_TYPE_RAW_TRACEPOINT &&
 	    prog->type != BPF_PROG_TYPE_TRACING &&
+	    prog->type != BPF_PROG_TYPE_EXT &&
 	    prog->type != BPF_PROG_TYPE_RAW_TRACEPOINT_WRITABLE) {
 		err = -EINVAL;
 		goto out_put_prog;
 	}
 
-	if (prog->type == BPF_PROG_TYPE_TRACING) {
+	if (prog->type == BPF_PROG_TYPE_TRACING ||
+	    prog->type == BPF_PROG_TYPE_EXT) {
 		if (attr->raw_tracepoint.name) {
 			/* The attach point for this category of programs
 			 * should be specified via btf_id during program load.
diff --git a/kernel/bpf/trampoline.c b/kernel/bpf/trampoline.c
index 7657ede7aee2..eb64c245052b 100644
--- a/kernel/bpf/trampoline.c
+++ b/kernel/bpf/trampoline.c
@@ -5,6 +5,12 @@
 #include <linux/filter.h>
 #include <linux/ftrace.h>
 
+/* dummy _ops. The verifier will operate on target program's ops. */
+const struct bpf_verifier_ops bpf_extension_verifier_ops = {
+};
+const struct bpf_prog_ops bpf_extension_prog_ops = {
+};
+
 /* btf_vmlinux has ~22k attachable functions. 1k htab is enough. */
 #define TRAMPOLINE_HASH_BITS 10
 #define TRAMPOLINE_TABLE_SIZE (1 << TRAMPOLINE_HASH_BITS)
@@ -194,8 +200,10 @@ static enum bpf_tramp_prog_type bpf_attach_type_to_tramp(enum bpf_attach_type t)
 	switch (t) {
 	case BPF_TRACE_FENTRY:
 		return BPF_TRAMP_FENTRY;
-	default:
+	case BPF_TRACE_FEXIT:
 		return BPF_TRAMP_FEXIT;
+	default:
+		return BPF_TRAMP_REPLACE;
 	}
 }
 
@@ -204,12 +212,31 @@ int bpf_trampoline_link_prog(struct bpf_prog *prog)
 	enum bpf_tramp_prog_type kind;
 	struct bpf_trampoline *tr;
 	int err = 0;
+	int cnt;
 
 	tr = prog->aux->trampoline;
 	kind = bpf_attach_type_to_tramp(prog->expected_attach_type);
 	mutex_lock(&tr->mutex);
-	if (tr->progs_cnt[BPF_TRAMP_FENTRY] + tr->progs_cnt[BPF_TRAMP_FEXIT]
-	    >= BPF_MAX_TRAMP_PROGS) {
+	if (tr->extension_prog) {
+		/* cannot attach fentry/fexit if extension prog is attached.
+		 * cannot overwrite extension prog either.
+		 */
+		err = -EBUSY;
+		goto out;
+	}
+	cnt = tr->progs_cnt[BPF_TRAMP_FENTRY] + tr->progs_cnt[BPF_TRAMP_FEXIT];
+	if (kind == BPF_TRAMP_REPLACE) {
+		/* Cannot attach extension if fentry/fexit are in use. */
+		if (cnt) {
+			err = -EBUSY;
+			goto out;
+		}
+		tr->extension_prog = prog;
+		err = bpf_arch_text_poke(tr->func.addr, BPF_MOD_JUMP, NULL,
+					 prog->bpf_func);
+		goto out;
+	}
+	if (cnt >= BPF_MAX_TRAMP_PROGS) {
 		err = -E2BIG;
 		goto out;
 	}
@@ -240,9 +267,17 @@ int bpf_trampoline_unlink_prog(struct bpf_prog *prog)
 	tr = prog->aux->trampoline;
 	kind = bpf_attach_type_to_tramp(prog->expected_attach_type);
 	mutex_lock(&tr->mutex);
+	if (kind == BPF_TRAMP_REPLACE) {
+		WARN_ON_ONCE(!tr->extension_prog);
+		err = bpf_arch_text_poke(tr->func.addr, BPF_MOD_JUMP,
+					 tr->extension_prog->bpf_func, NULL);
+		tr->extension_prog = NULL;
+		goto out;
+	}
 	hlist_del(&prog->aux->tramp_hlist);
 	tr->progs_cnt[kind]--;
 	err = bpf_trampoline_update(prog->aux->trampoline);
+out:
 	mutex_unlock(&tr->mutex);
 	return err;
 }
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 6defbec9eb62..9fe94f64f0ec 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -9564,7 +9564,7 @@ static int do_check_common(struct bpf_verifier_env *env, int subprog)
 			subprog);
 
 	regs = state->frame[state->curframe]->regs;
-	if (subprog) {
+	if (subprog || env->prog->type == BPF_PROG_TYPE_EXT) {
 		ret = btf_prepare_func_args(env, subprog, regs);
 		if (ret)
 			goto out;
@@ -9742,6 +9742,7 @@ static int check_struct_ops_btf_id(struct bpf_verifier_env *env)
 static int check_attach_btf_id(struct bpf_verifier_env *env)
 {
 	struct bpf_prog *prog = env->prog;
+	bool prog_extension = prog->type == BPF_PROG_TYPE_EXT;
 	struct bpf_prog *tgt_prog = prog->aux->linked_prog;
 	u32 btf_id = prog->aux->attach_btf_id;
 	const char prefix[] = "btf_trace_";
@@ -9757,7 +9758,7 @@ static int check_attach_btf_id(struct bpf_verifier_env *env)
 	if (prog->type == BPF_PROG_TYPE_STRUCT_OPS)
 		return check_struct_ops_btf_id(env);
 
-	if (prog->type != BPF_PROG_TYPE_TRACING)
+	if (prog->type != BPF_PROG_TYPE_TRACING && !prog_extension)
 		return 0;
 
 	if (!btf_id) {
@@ -9793,8 +9794,59 @@ static int check_attach_btf_id(struct bpf_verifier_env *env)
 			return -EINVAL;
 		}
 		conservative = aux->func_info_aux[subprog].unreliable;
+		if (prog_extension) {
+			if (conservative) {
+				verbose(env,
+					"Cannot replace static functions\n");
+				return -EINVAL;
+			}
+			if (!prog->jit_requested) {
+				verbose(env,
+					"Extension programs should be JITed\n");
+				return -EINVAL;
+			}
+			env->ops = bpf_verifier_ops[tgt_prog->type];
+		}
+		if (!tgt_prog->jited) {
+			verbose(env, "Can attach to only JITed progs\n");
+			return -EINVAL;
+		}
+		if (tgt_prog->type == prog->type) {
+			/* Cannot fentry/fexit another fentry/fexit program.
+			 * Cannot attach program extension to another extension.
+			 * It's ok to attach fentry/fexit to extension program.
+			 */
+			verbose(env, "Cannot recursively attach\n");
+			return -EINVAL;
+		}
+		if (tgt_prog->type == BPF_PROG_TYPE_TRACING &&
+		    prog_extension &&
+		    (tgt_prog->expected_attach_type == BPF_TRACE_FENTRY ||
+		     tgt_prog->expected_attach_type == BPF_TRACE_FEXIT)) {
+			/* Program extensions can extend all program types
+			 * except fentry/fexit. The reason is the following.
+			 * The fentry/fexit programs are used for performance
+			 * analysis, stats and can be attached to any program
+			 * type except themselves. When extension program is
+			 * replacing XDP function it is necessary to allow
+			 * performance analysis of all functions. Both original
+			 * XDP program and its program extension. Hence
+			 * attaching fentry/fexit to BPF_PROG_TYPE_EXT is
+			 * allowed. If extending of fentry/fexit was allowed it
+			 * would be possible to create long call chain
+			 * fentry->extension->fentry->extension beyond
+			 * reasonable stack size. Hence extending fentry is not
+			 * allowed.
+			 */
+			verbose(env, "Cannot extend fentry/fexit\n");
+			return -EINVAL;
+		}
 		key = ((u64)aux->id) << 32 | btf_id;
 	} else {
+		if (prog_extension) {
+			verbose(env, "Cannot replace kernel functions\n");
+			return -EINVAL;
+		}
 		key = btf_id;
 	}
 
@@ -9832,6 +9884,10 @@ static int check_attach_btf_id(struct bpf_verifier_env *env)
 		prog->aux->attach_func_proto = t;
 		prog->aux->attach_btf_trace = true;
 		return 0;
+	default:
+		if (!prog_extension)
+			return -EINVAL;
+		/* fallthrough */
 	case BPF_TRACE_FENTRY:
 	case BPF_TRACE_FEXIT:
 		if (!btf_type_is_func(t)) {
@@ -9839,6 +9895,9 @@ static int check_attach_btf_id(struct bpf_verifier_env *env)
 				btf_id);
 			return -EINVAL;
 		}
+		if (prog_extension &&
+		    btf_check_type_match(env, prog, btf, t))
+			return -EINVAL;
 		t = btf_type_by_id(btf, t->type);
 		if (!btf_type_is_func_proto(t))
 			return -EINVAL;
@@ -9862,18 +9921,6 @@ static int check_attach_btf_id(struct bpf_verifier_env *env)
 		if (ret < 0)
 			goto out;
 		if (tgt_prog) {
-			if (!tgt_prog->jited) {
-				/* for now */
-				verbose(env, "Can trace only JITed BPF progs\n");
-				ret = -EINVAL;
-				goto out;
-			}
-			if (tgt_prog->type == BPF_PROG_TYPE_TRACING) {
-				/* prevent cycles */
-				verbose(env, "Cannot recursively attach\n");
-				ret = -EINVAL;
-				goto out;
-			}
 			if (subprog == 0)
 				addr = (long) tgt_prog->bpf_func;
 			else
@@ -9895,8 +9942,6 @@ out:
 		if (ret)
 			bpf_trampoline_put(tr);
 		return ret;
-	default:
-		return -EINVAL;
 	}
 }
 
@@ -9966,10 +10011,6 @@ int bpf_check(struct bpf_prog **prog, union bpf_attr *attr,
 		goto skip_full_check;
 	}
 
-	ret = check_attach_btf_id(env);
-	if (ret)
-		goto skip_full_check;
-
 	env->strict_alignment = !!(attr->prog_flags & BPF_F_STRICT_ALIGNMENT);
 	if (!IS_ENABLED(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS))
 		env->strict_alignment = true;
@@ -10006,6 +10047,10 @@ int bpf_check(struct bpf_prog **prog, union bpf_attr *attr,
 	if (ret < 0)
 		goto skip_full_check;
 
+	ret = check_attach_btf_id(env);
+	if (ret)
+		goto skip_full_check;
+
 	ret = check_cfg(env);
 	if (ret < 0)
 		goto skip_full_check;
-- 
cgit v1.2.3-59-g8ed1b


From 2db6eab18b9778d55f48c804f8efebd7097e7958 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Mon, 20 Jan 2020 16:53:47 -0800
Subject: libbpf: Add support for program extensions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add minimal support for program extensions. bpf_object_open_opts() needs to be
called with attach_prog_fd = target_prog_fd and BPF program extension needs to
have in .c file section definition like SEC("freplace/func_to_be_replaced").
libbpf will search for "func_to_be_replaced" in the target_prog_fd's BTF and
will pass it in attach_btf_id to the kernel. This approach works for tests, but
more compex use case may need to request function name (and attach_btf_id that
kernel sees) to be more dynamic. Such API will be added in future patches.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/bpf/20200121005348.2769920-3-ast@kernel.org
---
 tools/include/uapi/linux/bpf.h |  1 +
 tools/lib/bpf/bpf.c            |  3 ++-
 tools/lib/bpf/libbpf.c         | 13 ++++++++++---
 tools/lib/bpf/libbpf.h         |  2 ++
 tools/lib/bpf/libbpf.map       |  2 ++
 tools/lib/bpf/libbpf_probes.c  |  1 +
 6 files changed, 18 insertions(+), 4 deletions(-)

diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index 033d90a2282d..e81628eb059c 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -180,6 +180,7 @@ enum bpf_prog_type {
 	BPF_PROG_TYPE_CGROUP_SOCKOPT,
 	BPF_PROG_TYPE_TRACING,
 	BPF_PROG_TYPE_STRUCT_OPS,
+	BPF_PROG_TYPE_EXT,
 };
 
 enum bpf_attach_type {
diff --git a/tools/lib/bpf/bpf.c b/tools/lib/bpf/bpf.c
index ed42b006533c..c6dafe563176 100644
--- a/tools/lib/bpf/bpf.c
+++ b/tools/lib/bpf/bpf.c
@@ -237,7 +237,8 @@ int bpf_load_program_xattr(const struct bpf_load_program_attr *load_attr,
 	attr.expected_attach_type = load_attr->expected_attach_type;
 	if (attr.prog_type == BPF_PROG_TYPE_STRUCT_OPS) {
 		attr.attach_btf_id = load_attr->attach_btf_id;
-	} else if (attr.prog_type == BPF_PROG_TYPE_TRACING) {
+	} else if (attr.prog_type == BPF_PROG_TYPE_TRACING ||
+		   attr.prog_type == BPF_PROG_TYPE_EXT) {
 		attr.attach_btf_id = load_attr->attach_btf_id;
 		attr.attach_prog_fd = load_attr->attach_prog_fd;
 	} else {
diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index faab96a42141..ae34b681ae82 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -4837,7 +4837,8 @@ load_program(struct bpf_program *prog, struct bpf_insn *insns, int insns_cnt,
 	load_attr.license = license;
 	if (prog->type == BPF_PROG_TYPE_STRUCT_OPS) {
 		load_attr.attach_btf_id = prog->attach_btf_id;
-	} else if (prog->type == BPF_PROG_TYPE_TRACING) {
+	} else if (prog->type == BPF_PROG_TYPE_TRACING ||
+		   prog->type == BPF_PROG_TYPE_EXT) {
 		load_attr.attach_prog_fd = prog->attach_prog_fd;
 		load_attr.attach_btf_id = prog->attach_btf_id;
 	} else {
@@ -4918,7 +4919,8 @@ int bpf_program__load(struct bpf_program *prog, char *license, __u32 kern_ver)
 {
 	int err = 0, fd, i, btf_id;
 
-	if (prog->type == BPF_PROG_TYPE_TRACING) {
+	if (prog->type == BPF_PROG_TYPE_TRACING ||
+	    prog->type == BPF_PROG_TYPE_EXT) {
 		btf_id = libbpf_find_attach_btf_id(prog);
 		if (btf_id <= 0)
 			return btf_id;
@@ -5092,7 +5094,8 @@ __bpf_object__open(const char *path, const void *obj_buf, size_t obj_buf_sz,
 
 		bpf_program__set_type(prog, prog_type);
 		bpf_program__set_expected_attach_type(prog, attach_type);
-		if (prog_type == BPF_PROG_TYPE_TRACING)
+		if (prog_type == BPF_PROG_TYPE_TRACING ||
+		    prog_type == BPF_PROG_TYPE_EXT)
 			prog->attach_prog_fd = OPTS_GET(opts, attach_prog_fd, 0);
 	}
 
@@ -6166,6 +6169,7 @@ BPF_PROG_TYPE_FNS(xdp, BPF_PROG_TYPE_XDP);
 BPF_PROG_TYPE_FNS(perf_event, BPF_PROG_TYPE_PERF_EVENT);
 BPF_PROG_TYPE_FNS(tracing, BPF_PROG_TYPE_TRACING);
 BPF_PROG_TYPE_FNS(struct_ops, BPF_PROG_TYPE_STRUCT_OPS);
+BPF_PROG_TYPE_FNS(extension, BPF_PROG_TYPE_EXT);
 
 enum bpf_attach_type
 bpf_program__get_expected_attach_type(struct bpf_program *prog)
@@ -6265,6 +6269,9 @@ static const struct bpf_sec_def section_defs[] = {
 		.expected_attach_type = BPF_TRACE_FEXIT,
 		.is_attach_btf = true,
 		.attach_fn = attach_trace),
+	SEC_DEF("freplace/", EXT,
+		.is_attach_btf = true,
+		.attach_fn = attach_trace),
 	BPF_PROG_SEC("xdp",			BPF_PROG_TYPE_XDP),
 	BPF_PROG_SEC("perf_event",		BPF_PROG_TYPE_PERF_EVENT),
 	BPF_PROG_SEC("lwt_in",			BPF_PROG_TYPE_LWT_IN),
diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index 01639f9a1062..2a5e3b087002 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -318,6 +318,7 @@ LIBBPF_API int bpf_program__set_xdp(struct bpf_program *prog);
 LIBBPF_API int bpf_program__set_perf_event(struct bpf_program *prog);
 LIBBPF_API int bpf_program__set_tracing(struct bpf_program *prog);
 LIBBPF_API int bpf_program__set_struct_ops(struct bpf_program *prog);
+LIBBPF_API int bpf_program__set_extension(struct bpf_program *prog);
 
 LIBBPF_API enum bpf_prog_type bpf_program__get_type(struct bpf_program *prog);
 LIBBPF_API void bpf_program__set_type(struct bpf_program *prog,
@@ -339,6 +340,7 @@ LIBBPF_API bool bpf_program__is_xdp(const struct bpf_program *prog);
 LIBBPF_API bool bpf_program__is_perf_event(const struct bpf_program *prog);
 LIBBPF_API bool bpf_program__is_tracing(const struct bpf_program *prog);
 LIBBPF_API bool bpf_program__is_struct_ops(const struct bpf_program *prog);
+LIBBPF_API bool bpf_program__is_extension(const struct bpf_program *prog);
 
 /*
  * No need for __attribute__((packed)), all members of 'bpf_map_def'
diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
index 64ec71ba41f1..b035122142bb 100644
--- a/tools/lib/bpf/libbpf.map
+++ b/tools/lib/bpf/libbpf.map
@@ -228,7 +228,9 @@ LIBBPF_0.0.7 {
 		bpf_prog_attach_xattr;
 		bpf_program__attach;
 		bpf_program__name;
+		bpf_program__is_extension;
 		bpf_program__is_struct_ops;
+		bpf_program__set_extension;
 		bpf_program__set_struct_ops;
 		btf__align_of;
 		libbpf_find_kernel_btf;
diff --git a/tools/lib/bpf/libbpf_probes.c b/tools/lib/bpf/libbpf_probes.c
index 8cc992bc532a..b782ebef6ac9 100644
--- a/tools/lib/bpf/libbpf_probes.c
+++ b/tools/lib/bpf/libbpf_probes.c
@@ -107,6 +107,7 @@ probe_load(enum bpf_prog_type prog_type, const struct bpf_insn *insns,
 	case BPF_PROG_TYPE_CGROUP_SOCKOPT:
 	case BPF_PROG_TYPE_TRACING:
 	case BPF_PROG_TYPE_STRUCT_OPS:
+	case BPF_PROG_TYPE_EXT:
 	default:
 		break;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 7805fe843964f81f98592bf580b9cd736ac4ad4c Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Mon, 20 Jan 2020 16:53:48 -0800
Subject: selftests/bpf: Add tests for program extensions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add program extension tests that build on top of fexit_bpf2bpf tests.
Replace three global functions in previously loaded test_pkt_access.c program
with three new implementations:
int get_skb_len(struct __sk_buff *skb);
int get_constant(long val);
int get_skb_ifindex(int val, struct __sk_buff *skb, int var);
New function return the same results as original only if arguments match.

new_get_skb_ifindex() demonstrates that 'skb' argument doesn't have to be first
and only argument of BPF program. All normal skb based accesses are available.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/bpf/20200121005348.2769920-4-ast@kernel.org
---
 .../selftests/bpf/prog_tests/fexit_bpf2bpf.c       | 20 +++++++-
 tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c  | 57 ++++++++++++++++++++++
 .../testing/selftests/bpf/progs/test_pkt_access.c  |  8 ++-
 3 files changed, 83 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/fexit_bpf2bpf.c b/tools/testing/selftests/bpf/prog_tests/fexit_bpf2bpf.c
index 7d3740d38965..db5c74d2ce6d 100644
--- a/tools/testing/selftests/bpf/prog_tests/fexit_bpf2bpf.c
+++ b/tools/testing/selftests/bpf/prog_tests/fexit_bpf2bpf.c
@@ -26,7 +26,7 @@ static void test_fexit_bpf2bpf_common(const char *obj_file,
 
 	link = calloc(sizeof(struct bpf_link *), prog_cnt);
 	prog = calloc(sizeof(struct bpf_program *), prog_cnt);
-	result = malloc(prog_cnt * sizeof(u64));
+	result = malloc((prog_cnt + 32 /* spare */) * sizeof(u64));
 	if (CHECK(!link || !prog || !result, "alloc_memory",
 		  "failed to alloc memory"))
 		goto close_prog;
@@ -106,8 +106,26 @@ static void test_target_yes_callees(void)
 				  prog_name);
 }
 
+static void test_func_replace(void)
+{
+	const char *prog_name[] = {
+		"fexit/test_pkt_access",
+		"fexit/test_pkt_access_subprog1",
+		"fexit/test_pkt_access_subprog2",
+		"fexit/test_pkt_access_subprog3",
+		"freplace/get_skb_len",
+		"freplace/get_skb_ifindex",
+		"freplace/get_constant",
+	};
+	test_fexit_bpf2bpf_common("./fexit_bpf2bpf.o",
+				  "./test_pkt_access.o",
+				  ARRAY_SIZE(prog_name),
+				  prog_name);
+}
+
 void test_fexit_bpf2bpf(void)
 {
 	test_target_no_callees();
 	test_target_yes_callees();
+	test_func_replace();
 }
diff --git a/tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c b/tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c
index ec5710767d13..c329fccf9842 100644
--- a/tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c
+++ b/tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c
@@ -1,7 +1,10 @@
 // SPDX-License-Identifier: GPL-2.0
 /* Copyright (c) 2019 Facebook */
+#include <linux/stddef.h>
+#include <linux/ipv6.h>
 #include <linux/bpf.h>
 #include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
 #include "bpf_trace_helpers.h"
 
 struct sk_buff {
@@ -94,4 +97,58 @@ int BPF_PROG(test_subprog3, int val, struct sk_buff *skb, int ret)
 	test_result_subprog3 = 1;
 	return 0;
 }
+
+__u64 test_get_skb_len = 0;
+SEC("freplace/get_skb_len")
+int new_get_skb_len(struct __sk_buff *skb)
+{
+	int len = skb->len;
+
+	if (len != 74)
+		return 0;
+	test_get_skb_len = 1;
+	return 74; /* original get_skb_len() returns skb->len */
+}
+
+__u64 test_get_skb_ifindex = 0;
+SEC("freplace/get_skb_ifindex")
+int new_get_skb_ifindex(int val, struct __sk_buff *skb, int var)
+{
+	void *data_end = (void *)(long)skb->data_end;
+	void *data = (void *)(long)skb->data;
+	struct ipv6hdr ip6, *ip6p;
+	int ifindex = skb->ifindex;
+	__u32 eth_proto;
+	__u32 nh_off;
+
+	/* check that BPF extension can read packet via direct packet access */
+	if (data + 14 + sizeof(ip6) > data_end)
+		return 0;
+	ip6p = data + 14;
+
+	if (ip6p->nexthdr != 6 || ip6p->payload_len != __bpf_constant_htons(123))
+		return 0;
+
+	/* check that legacy packet access helper works too */
+	if (bpf_skb_load_bytes(skb, 14, &ip6, sizeof(ip6)) < 0)
+		return 0;
+	ip6p = &ip6;
+	if (ip6p->nexthdr != 6 || ip6p->payload_len != __bpf_constant_htons(123))
+		return 0;
+
+	if (ifindex != 1 || val != 3 || var != 1)
+		return 0;
+	test_get_skb_ifindex = 1;
+	return 3; /* original get_skb_ifindex() returns val * ifindex * var */
+}
+
+volatile __u64 test_get_constant = 0;
+SEC("freplace/get_constant")
+int new_get_constant(long val)
+{
+	if (val != 123)
+		return 0;
+	test_get_constant = 1;
+	return test_get_constant; /* original get_constant() returns val - 122 */
+}
 char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/test_pkt_access.c b/tools/testing/selftests/bpf/progs/test_pkt_access.c
index 7c9fcfd2b463..e72eba4a93d2 100644
--- a/tools/testing/selftests/bpf/progs/test_pkt_access.c
+++ b/tools/testing/selftests/bpf/progs/test_pkt_access.c
@@ -57,12 +57,18 @@ int get_skb_len(struct __sk_buff *skb)
 	return skb->len;
 }
 
+__attribute__ ((noinline))
+int get_constant(long val)
+{
+	return val - 122;
+}
+
 int get_skb_ifindex(int, struct __sk_buff *skb, int);
 
 __attribute__ ((noinline))
 int test_pkt_access_subprog3(int val, struct __sk_buff *skb)
 {
-	return get_skb_len(skb) * get_skb_ifindex(val, skb, 1);
+	return get_skb_len(skb) * get_skb_ifindex(val, skb, get_constant(123));
 }
 
 __attribute__ ((noinline))
-- 
cgit v1.2.3-59-g8ed1b


From 5576b991e9c1a11d2cc21c4b94fc75ec27603896 Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <kafai@fb.com>
Date: Wed, 22 Jan 2020 15:36:46 -0800
Subject: bpf: Add BPF_FUNC_jiffies64

This patch adds a helper to read the 64bit jiffies.  It will be used
in a later patch to implement the bpf_cubic.c.

The helper is inlined for jit_requested and 64 BITS_PER_LONG
as the map_gen_lookup().  Other cases could be considered together
with map_gen_lookup() if needed.

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200122233646.903260-1-kafai@fb.com
---
 include/linux/bpf.h      |  1 +
 include/uapi/linux/bpf.h |  9 ++++++++-
 kernel/bpf/core.c        |  1 +
 kernel/bpf/helpers.c     | 12 ++++++++++++
 kernel/bpf/verifier.c    | 24 ++++++++++++++++++++++++
 net/core/filter.c        |  2 ++
 6 files changed, 48 insertions(+), 1 deletion(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 05d16615054c..a9687861fd7e 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -1414,6 +1414,7 @@ extern const struct bpf_func_proto bpf_get_local_storage_proto;
 extern const struct bpf_func_proto bpf_strtol_proto;
 extern const struct bpf_func_proto bpf_strtoul_proto;
 extern const struct bpf_func_proto bpf_tcp_sock_proto;
+extern const struct bpf_func_proto bpf_jiffies64_proto;
 
 /* Shared helpers among cBPF and eBPF. */
 void bpf_user_rnd_init_once(void);
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index e81628eb059c..f1d74a2bd234 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -2886,6 +2886,12 @@ union bpf_attr {
  *		**-EPERM** if no permission to send the *sig*.
  *
  *		**-EAGAIN** if bpf program can try again.
+ *
+ * u64 bpf_jiffies64(void)
+ *	Description
+ *		Obtain the 64bit jiffies
+ *	Return
+ *		The 64 bit jiffies
  */
 #define __BPF_FUNC_MAPPER(FN)		\
 	FN(unspec),			\
@@ -3005,7 +3011,8 @@ union bpf_attr {
 	FN(probe_read_user_str),	\
 	FN(probe_read_kernel_str),	\
 	FN(tcp_send_ack),		\
-	FN(send_signal_thread),
+	FN(send_signal_thread),		\
+	FN(jiffies64),
 
 /* integer value in 'imm' field of BPF_CALL instruction selects which helper
  * function eBPF program intends to call
diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c
index 29d47aae0dd1..973a20d49749 100644
--- a/kernel/bpf/core.c
+++ b/kernel/bpf/core.c
@@ -2137,6 +2137,7 @@ const struct bpf_func_proto bpf_map_pop_elem_proto __weak;
 const struct bpf_func_proto bpf_map_peek_elem_proto __weak;
 const struct bpf_func_proto bpf_spin_lock_proto __weak;
 const struct bpf_func_proto bpf_spin_unlock_proto __weak;
+const struct bpf_func_proto bpf_jiffies64_proto __weak;
 
 const struct bpf_func_proto bpf_get_prandom_u32_proto __weak;
 const struct bpf_func_proto bpf_get_smp_processor_id_proto __weak;
diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index cada974c9f4e..d8b7b110a1c5 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -11,6 +11,7 @@
 #include <linux/uidgid.h>
 #include <linux/filter.h>
 #include <linux/ctype.h>
+#include <linux/jiffies.h>
 
 #include "../../lib/kstrtox.h"
 
@@ -312,6 +313,17 @@ void copy_map_value_locked(struct bpf_map *map, void *dst, void *src,
 	preempt_enable();
 }
 
+BPF_CALL_0(bpf_jiffies64)
+{
+	return get_jiffies_64();
+}
+
+const struct bpf_func_proto bpf_jiffies64_proto = {
+	.func		= bpf_jiffies64,
+	.gpl_only	= false,
+	.ret_type	= RET_INTEGER,
+};
+
 #ifdef CONFIG_CGROUPS
 BPF_CALL_0(bpf_get_current_cgroup_id)
 {
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 9fe94f64f0ec..734abaa02123 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -9445,6 +9445,30 @@ static int fixup_bpf_calls(struct bpf_verifier_env *env)
 			goto patch_call_imm;
 		}
 
+		if (prog->jit_requested && BITS_PER_LONG == 64 &&
+		    insn->imm == BPF_FUNC_jiffies64) {
+			struct bpf_insn ld_jiffies_addr[2] = {
+				BPF_LD_IMM64(BPF_REG_0,
+					     (unsigned long)&jiffies),
+			};
+
+			insn_buf[0] = ld_jiffies_addr[0];
+			insn_buf[1] = ld_jiffies_addr[1];
+			insn_buf[2] = BPF_LDX_MEM(BPF_DW, BPF_REG_0,
+						  BPF_REG_0, 0);
+			cnt = 3;
+
+			new_prog = bpf_patch_insn_data(env, i + delta, insn_buf,
+						       cnt);
+			if (!new_prog)
+				return -ENOMEM;
+
+			delta    += cnt - 1;
+			env->prog = prog = new_prog;
+			insn      = new_prog->insnsi + i + delta;
+			continue;
+		}
+
 patch_call_imm:
 		fn = env->ops->get_func_proto(insn->imm, env->prog);
 		/* all functions that have prototype and verifier allowed
diff --git a/net/core/filter.c b/net/core/filter.c
index 17de6747d9e3..4bf3e4aa8a7a 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -5923,6 +5923,8 @@ bpf_base_func_proto(enum bpf_func_id func_id)
 		return &bpf_spin_unlock_proto;
 	case BPF_FUNC_trace_printk:
 		return bpf_get_trace_printk_proto();
+	case BPF_FUNC_jiffies64:
+		return &bpf_jiffies64_proto;
 	default:
 		return NULL;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 0a49c1a8e26266ad505704f54f9fdaf2ae14cbdb Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <kafai@fb.com>
Date: Wed, 22 Jan 2020 15:36:52 -0800
Subject: bpf: Sync uapi bpf.h to tools/

This patch sync uapi bpf.h to tools/.

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200122233652.903348-1-kafai@fb.com
---
 tools/include/uapi/linux/bpf.h | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index e81628eb059c..f1d74a2bd234 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -2886,6 +2886,12 @@ union bpf_attr {
  *		**-EPERM** if no permission to send the *sig*.
  *
  *		**-EAGAIN** if bpf program can try again.
+ *
+ * u64 bpf_jiffies64(void)
+ *	Description
+ *		Obtain the 64bit jiffies
+ *	Return
+ *		The 64 bit jiffies
  */
 #define __BPF_FUNC_MAPPER(FN)		\
 	FN(unspec),			\
@@ -3005,7 +3011,8 @@ union bpf_attr {
 	FN(probe_read_user_str),	\
 	FN(probe_read_kernel_str),	\
 	FN(tcp_send_ack),		\
-	FN(send_signal_thread),
+	FN(send_signal_thread),		\
+	FN(jiffies64),
 
 /* integer value in 'imm' field of BPF_CALL instruction selects which helper
  * function eBPF program intends to call
-- 
cgit v1.2.3-59-g8ed1b


From 6de4a9c430b57c6ebbccd2a1725f42e9be75f592 Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <kafai@fb.com>
Date: Wed, 22 Jan 2020 15:36:58 -0800
Subject: bpf: tcp: Add bpf_cubic example

This patch adds a bpf_cubic example.  Some highlights:
1. CONFIG_HZ .kconfig map is used.
2. In bictcp_update(), calculation is changed to use usec
   resolution (i.e. USEC_PER_JIFFY) instead of using jiffies.
   Thus, usecs_to_jiffies() is not used in the bpf_cubic.c.
3. In bitctcp_update() [under tcp_friendliness], the original
   "while (ca->ack_cnt > delta)" loop is changed to the equivalent
   "ca->ack_cnt / delta" operation.

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200122233658.903774-1-kafai@fb.com
---
 tools/testing/selftests/bpf/bpf_tcp_helpers.h      |  16 +
 .../testing/selftests/bpf/prog_tests/bpf_tcp_ca.c  |  25 +
 tools/testing/selftests/bpf/progs/bpf_cubic.c      | 544 +++++++++++++++++++++
 3 files changed, 585 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/progs/bpf_cubic.c

diff --git a/tools/testing/selftests/bpf/bpf_tcp_helpers.h b/tools/testing/selftests/bpf/bpf_tcp_helpers.h
index 6fee732f0297..8f21965ffc6c 100644
--- a/tools/testing/selftests/bpf/bpf_tcp_helpers.h
+++ b/tools/testing/selftests/bpf/bpf_tcp_helpers.h
@@ -6,13 +6,28 @@
 #include <linux/types.h>
 #include <bpf/bpf_helpers.h>
 #include <bpf/bpf_core_read.h>
+#include "bpf_trace_helpers.h"
+
+#define BPF_STRUCT_OPS(name, args...) \
+SEC("struct_ops/"#name) \
+BPF_PROG(name, args)
+
+#define tcp_jiffies32 ((__u32)bpf_jiffies64())
 
 struct sock_common {
 	unsigned char	skc_state;
 } __attribute__((preserve_access_index));
 
+enum sk_pacing {
+	SK_PACING_NONE		= 0,
+	SK_PACING_NEEDED	= 1,
+	SK_PACING_FQ		= 2,
+};
+
 struct sock {
 	struct sock_common	__sk_common;
+	unsigned long		sk_pacing_rate;
+	__u32			sk_pacing_status; /* see enum sk_pacing */
 } __attribute__((preserve_access_index));
 
 struct inet_sock {
@@ -54,6 +69,7 @@ struct tcp_sock {
 	__u32	max_packets_out;
 	__u32	lsndtime;
 	__u32	prior_cwnd;
+	__u64	tcp_mstamp;	/* most recent packet received/sent */
 } __attribute__((preserve_access_index));
 
 static __always_inline struct inet_connection_sock *inet_csk(const struct sock *sk)
diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
index 517318f05b1d..8482bbc67eec 100644
--- a/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
+++ b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
@@ -4,6 +4,7 @@
 #include <linux/err.h>
 #include <test_progs.h>
 #include "bpf_dctcp.skel.h"
+#include "bpf_cubic.skel.h"
 
 #define min(a, b) ((a) < (b) ? (a) : (b))
 
@@ -158,6 +159,28 @@ done:
 	close(fd);
 }
 
+static void test_cubic(void)
+{
+	struct bpf_cubic *cubic_skel;
+	struct bpf_link *link;
+
+	cubic_skel = bpf_cubic__open_and_load();
+	if (CHECK(!cubic_skel, "bpf_cubic__open_and_load", "failed\n"))
+		return;
+
+	link = bpf_map__attach_struct_ops(cubic_skel->maps.cubic);
+	if (CHECK(IS_ERR(link), "bpf_map__attach_struct_ops", "err:%ld\n",
+		  PTR_ERR(link))) {
+		bpf_cubic__destroy(cubic_skel);
+		return;
+	}
+
+	do_test("bpf_cubic");
+
+	bpf_link__destroy(link);
+	bpf_cubic__destroy(cubic_skel);
+}
+
 static void test_dctcp(void)
 {
 	struct bpf_dctcp *dctcp_skel;
@@ -184,4 +207,6 @@ void test_bpf_tcp_ca(void)
 {
 	if (test__start_subtest("dctcp"))
 		test_dctcp();
+	if (test__start_subtest("cubic"))
+		test_cubic();
 }
diff --git a/tools/testing/selftests/bpf/progs/bpf_cubic.c b/tools/testing/selftests/bpf/progs/bpf_cubic.c
new file mode 100644
index 000000000000..7897c8f4d363
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/bpf_cubic.c
@@ -0,0 +1,544 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+/* WARNING: This implemenation is not necessarily the same
+ * as the tcp_cubic.c.  The purpose is mainly for testing
+ * the kernel BPF logic.
+ *
+ * Highlights:
+ * 1. CONFIG_HZ .kconfig map is used.
+ * 2. In bictcp_update(), calculation is changed to use usec
+ *    resolution (i.e. USEC_PER_JIFFY) instead of using jiffies.
+ *    Thus, usecs_to_jiffies() is not used in the bpf_cubic.c.
+ * 3. In bitctcp_update() [under tcp_friendliness], the original
+ *    "while (ca->ack_cnt > delta)" loop is changed to the equivalent
+ *    "ca->ack_cnt / delta" operation.
+ */
+
+#include <linux/bpf.h>
+#include "bpf_tcp_helpers.h"
+
+char _license[] SEC("license") = "GPL";
+
+#define clamp(val, lo, hi) min((typeof(val))max(val, lo), hi)
+
+#define BICTCP_BETA_SCALE    1024	/* Scale factor beta calculation
+					 * max_cwnd = snd_cwnd * beta
+					 */
+#define	BICTCP_HZ		10	/* BIC HZ 2^10 = 1024 */
+
+/* Two methods of hybrid slow start */
+#define HYSTART_ACK_TRAIN	0x1
+#define HYSTART_DELAY		0x2
+
+/* Number of delay samples for detecting the increase of delay */
+#define HYSTART_MIN_SAMPLES	8
+#define HYSTART_DELAY_MIN	(4000U)	/* 4ms */
+#define HYSTART_DELAY_MAX	(16000U)	/* 16 ms */
+#define HYSTART_DELAY_THRESH(x)	clamp(x, HYSTART_DELAY_MIN, HYSTART_DELAY_MAX)
+
+static int fast_convergence = 1;
+static const int beta = 717;	/* = 717/1024 (BICTCP_BETA_SCALE) */
+static int initial_ssthresh;
+static const int bic_scale = 41;
+static int tcp_friendliness = 1;
+
+static int hystart = 1;
+static int hystart_detect = HYSTART_ACK_TRAIN | HYSTART_DELAY;
+static int hystart_low_window = 16;
+static int hystart_ack_delta_us = 2000;
+
+static const __u32 cube_rtt_scale = (bic_scale * 10);	/* 1024*c/rtt */
+static const __u32 beta_scale = 8*(BICTCP_BETA_SCALE+beta) / 3
+				/ (BICTCP_BETA_SCALE - beta);
+/* calculate the "K" for (wmax-cwnd) = c/rtt * K^3
+ *  so K = cubic_root( (wmax-cwnd)*rtt/c )
+ * the unit of K is bictcp_HZ=2^10, not HZ
+ *
+ *  c = bic_scale >> 10
+ *  rtt = 100ms
+ *
+ * the following code has been designed and tested for
+ * cwnd < 1 million packets
+ * RTT < 100 seconds
+ * HZ < 1,000,00  (corresponding to 10 nano-second)
+ */
+
+/* 1/c * 2^2*bictcp_HZ * srtt, 2^40 */
+static const __u64 cube_factor = (__u64)(1ull << (10+3*BICTCP_HZ))
+				/ (bic_scale * 10);
+
+/* BIC TCP Parameters */
+struct bictcp {
+	__u32	cnt;		/* increase cwnd by 1 after ACKs */
+	__u32	last_max_cwnd;	/* last maximum snd_cwnd */
+	__u32	last_cwnd;	/* the last snd_cwnd */
+	__u32	last_time;	/* time when updated last_cwnd */
+	__u32	bic_origin_point;/* origin point of bic function */
+	__u32	bic_K;		/* time to origin point
+				   from the beginning of the current epoch */
+	__u32	delay_min;	/* min delay (usec) */
+	__u32	epoch_start;	/* beginning of an epoch */
+	__u32	ack_cnt;	/* number of acks */
+	__u32	tcp_cwnd;	/* estimated tcp cwnd */
+	__u16	unused;
+	__u8	sample_cnt;	/* number of samples to decide curr_rtt */
+	__u8	found;		/* the exit point is found? */
+	__u32	round_start;	/* beginning of each round */
+	__u32	end_seq;	/* end_seq of the round */
+	__u32	last_ack;	/* last time when the ACK spacing is close */
+	__u32	curr_rtt;	/* the minimum rtt of current round */
+};
+
+static inline void bictcp_reset(struct bictcp *ca)
+{
+	ca->cnt = 0;
+	ca->last_max_cwnd = 0;
+	ca->last_cwnd = 0;
+	ca->last_time = 0;
+	ca->bic_origin_point = 0;
+	ca->bic_K = 0;
+	ca->delay_min = 0;
+	ca->epoch_start = 0;
+	ca->ack_cnt = 0;
+	ca->tcp_cwnd = 0;
+	ca->found = 0;
+}
+
+extern unsigned long CONFIG_HZ __kconfig;
+#define HZ CONFIG_HZ
+#define USEC_PER_MSEC	1000UL
+#define USEC_PER_SEC	1000000UL
+#define USEC_PER_JIFFY	(USEC_PER_SEC / HZ)
+
+static __always_inline __u64 div64_u64(__u64 dividend, __u64 divisor)
+{
+	return dividend / divisor;
+}
+
+#define div64_ul div64_u64
+
+#define BITS_PER_U64 (sizeof(__u64) * 8)
+static __always_inline int fls64(__u64 x)
+{
+	int num = BITS_PER_U64 - 1;
+
+	if (x == 0)
+		return 0;
+
+	if (!(x & (~0ull << (BITS_PER_U64-32)))) {
+		num -= 32;
+		x <<= 32;
+	}
+	if (!(x & (~0ull << (BITS_PER_U64-16)))) {
+		num -= 16;
+		x <<= 16;
+	}
+	if (!(x & (~0ull << (BITS_PER_U64-8)))) {
+		num -= 8;
+		x <<= 8;
+	}
+	if (!(x & (~0ull << (BITS_PER_U64-4)))) {
+		num -= 4;
+		x <<= 4;
+	}
+	if (!(x & (~0ull << (BITS_PER_U64-2)))) {
+		num -= 2;
+		x <<= 2;
+	}
+	if (!(x & (~0ull << (BITS_PER_U64-1))))
+		num -= 1;
+
+	return num + 1;
+}
+
+static __always_inline __u32 bictcp_clock_us(const struct sock *sk)
+{
+	return tcp_sk(sk)->tcp_mstamp;
+}
+
+static __always_inline void bictcp_hystart_reset(struct sock *sk)
+{
+	struct tcp_sock *tp = tcp_sk(sk);
+	struct bictcp *ca = inet_csk_ca(sk);
+
+	ca->round_start = ca->last_ack = bictcp_clock_us(sk);
+	ca->end_seq = tp->snd_nxt;
+	ca->curr_rtt = ~0U;
+	ca->sample_cnt = 0;
+}
+
+/* "struct_ops/" prefix is not a requirement
+ * It will be recognized as BPF_PROG_TYPE_STRUCT_OPS
+ * as long as it is used in one of the func ptr
+ * under SEC(".struct_ops").
+ */
+SEC("struct_ops/bictcp_init")
+void BPF_PROG(bictcp_init, struct sock *sk)
+{
+	struct bictcp *ca = inet_csk_ca(sk);
+
+	bictcp_reset(ca);
+
+	if (hystart)
+		bictcp_hystart_reset(sk);
+
+	if (!hystart && initial_ssthresh)
+		tcp_sk(sk)->snd_ssthresh = initial_ssthresh;
+}
+
+/* No prefix in SEC will also work.
+ * The remaining tcp-cubic functions have an easier way.
+ */
+SEC("no-sec-prefix-bictcp_cwnd_event")
+void BPF_PROG(bictcp_cwnd_event, struct sock *sk, enum tcp_ca_event event)
+{
+	if (event == CA_EVENT_TX_START) {
+		struct bictcp *ca = inet_csk_ca(sk);
+		__u32 now = tcp_jiffies32;
+		__s32 delta;
+
+		delta = now - tcp_sk(sk)->lsndtime;
+
+		/* We were application limited (idle) for a while.
+		 * Shift epoch_start to keep cwnd growth to cubic curve.
+		 */
+		if (ca->epoch_start && delta > 0) {
+			ca->epoch_start += delta;
+			if (after(ca->epoch_start, now))
+				ca->epoch_start = now;
+		}
+		return;
+	}
+}
+
+/*
+ * cbrt(x) MSB values for x MSB values in [0..63].
+ * Precomputed then refined by hand - Willy Tarreau
+ *
+ * For x in [0..63],
+ *   v = cbrt(x << 18) - 1
+ *   cbrt(x) = (v[x] + 10) >> 6
+ */
+static const __u8 v[] = {
+	/* 0x00 */    0,   54,   54,   54,  118,  118,  118,  118,
+	/* 0x08 */  123,  129,  134,  138,  143,  147,  151,  156,
+	/* 0x10 */  157,  161,  164,  168,  170,  173,  176,  179,
+	/* 0x18 */  181,  185,  187,  190,  192,  194,  197,  199,
+	/* 0x20 */  200,  202,  204,  206,  209,  211,  213,  215,
+	/* 0x28 */  217,  219,  221,  222,  224,  225,  227,  229,
+	/* 0x30 */  231,  232,  234,  236,  237,  239,  240,  242,
+	/* 0x38 */  244,  245,  246,  248,  250,  251,  252,  254,
+};
+
+/* calculate the cubic root of x using a table lookup followed by one
+ * Newton-Raphson iteration.
+ * Avg err ~= 0.195%
+ */
+static __always_inline __u32 cubic_root(__u64 a)
+{
+	__u32 x, b, shift;
+
+	if (a < 64) {
+		/* a in [0..63] */
+		return ((__u32)v[(__u32)a] + 35) >> 6;
+	}
+
+	b = fls64(a);
+	b = ((b * 84) >> 8) - 1;
+	shift = (a >> (b * 3));
+
+	/* it is needed for verifier's bound check on v */
+	if (shift >= 64)
+		return 0;
+
+	x = ((__u32)(((__u32)v[shift] + 10) << b)) >> 6;
+
+	/*
+	 * Newton-Raphson iteration
+	 *                         2
+	 * x    = ( 2 * x  +  a / x  ) / 3
+	 *  k+1          k         k
+	 */
+	x = (2 * x + (__u32)div64_u64(a, (__u64)x * (__u64)(x - 1)));
+	x = ((x * 341) >> 10);
+	return x;
+}
+
+/*
+ * Compute congestion window to use.
+ */
+static __always_inline void bictcp_update(struct bictcp *ca, __u32 cwnd,
+					  __u32 acked)
+{
+	__u32 delta, bic_target, max_cnt;
+	__u64 offs, t;
+
+	ca->ack_cnt += acked;	/* count the number of ACKed packets */
+
+	if (ca->last_cwnd == cwnd &&
+	    (__s32)(tcp_jiffies32 - ca->last_time) <= HZ / 32)
+		return;
+
+	/* The CUBIC function can update ca->cnt at most once per jiffy.
+	 * On all cwnd reduction events, ca->epoch_start is set to 0,
+	 * which will force a recalculation of ca->cnt.
+	 */
+	if (ca->epoch_start && tcp_jiffies32 == ca->last_time)
+		goto tcp_friendliness;
+
+	ca->last_cwnd = cwnd;
+	ca->last_time = tcp_jiffies32;
+
+	if (ca->epoch_start == 0) {
+		ca->epoch_start = tcp_jiffies32;	/* record beginning */
+		ca->ack_cnt = acked;			/* start counting */
+		ca->tcp_cwnd = cwnd;			/* syn with cubic */
+
+		if (ca->last_max_cwnd <= cwnd) {
+			ca->bic_K = 0;
+			ca->bic_origin_point = cwnd;
+		} else {
+			/* Compute new K based on
+			 * (wmax-cwnd) * (srtt>>3 / HZ) / c * 2^(3*bictcp_HZ)
+			 */
+			ca->bic_K = cubic_root(cube_factor
+					       * (ca->last_max_cwnd - cwnd));
+			ca->bic_origin_point = ca->last_max_cwnd;
+		}
+	}
+
+	/* cubic function - calc*/
+	/* calculate c * time^3 / rtt,
+	 *  while considering overflow in calculation of time^3
+	 * (so time^3 is done by using 64 bit)
+	 * and without the support of division of 64bit numbers
+	 * (so all divisions are done by using 32 bit)
+	 *  also NOTE the unit of those veriables
+	 *	  time  = (t - K) / 2^bictcp_HZ
+	 *	  c = bic_scale >> 10
+	 * rtt  = (srtt >> 3) / HZ
+	 * !!! The following code does not have overflow problems,
+	 * if the cwnd < 1 million packets !!!
+	 */
+
+	t = (__s32)(tcp_jiffies32 - ca->epoch_start) * USEC_PER_JIFFY;
+	t += ca->delay_min;
+	/* change the unit from usec to bictcp_HZ */
+	t <<= BICTCP_HZ;
+	t /= USEC_PER_SEC;
+
+	if (t < ca->bic_K)		/* t - K */
+		offs = ca->bic_K - t;
+	else
+		offs = t - ca->bic_K;
+
+	/* c/rtt * (t-K)^3 */
+	delta = (cube_rtt_scale * offs * offs * offs) >> (10+3*BICTCP_HZ);
+	if (t < ca->bic_K)                            /* below origin*/
+		bic_target = ca->bic_origin_point - delta;
+	else                                          /* above origin*/
+		bic_target = ca->bic_origin_point + delta;
+
+	/* cubic function - calc bictcp_cnt*/
+	if (bic_target > cwnd) {
+		ca->cnt = cwnd / (bic_target - cwnd);
+	} else {
+		ca->cnt = 100 * cwnd;              /* very small increment*/
+	}
+
+	/*
+	 * The initial growth of cubic function may be too conservative
+	 * when the available bandwidth is still unknown.
+	 */
+	if (ca->last_max_cwnd == 0 && ca->cnt > 20)
+		ca->cnt = 20;	/* increase cwnd 5% per RTT */
+
+tcp_friendliness:
+	/* TCP Friendly */
+	if (tcp_friendliness) {
+		__u32 scale = beta_scale;
+		__u32 n;
+
+		/* update tcp cwnd */
+		delta = (cwnd * scale) >> 3;
+		if (ca->ack_cnt > delta && delta) {
+			n = ca->ack_cnt / delta;
+			ca->ack_cnt -= n * delta;
+			ca->tcp_cwnd += n;
+		}
+
+		if (ca->tcp_cwnd > cwnd) {	/* if bic is slower than tcp */
+			delta = ca->tcp_cwnd - cwnd;
+			max_cnt = cwnd / delta;
+			if (ca->cnt > max_cnt)
+				ca->cnt = max_cnt;
+		}
+	}
+
+	/* The maximum rate of cwnd increase CUBIC allows is 1 packet per
+	 * 2 packets ACKed, meaning cwnd grows at 1.5x per RTT.
+	 */
+	ca->cnt = max(ca->cnt, 2U);
+}
+
+/* Or simply use the BPF_STRUCT_OPS to avoid the SEC boiler plate. */
+void BPF_STRUCT_OPS(bictcp_cong_avoid, struct sock *sk, __u32 ack, __u32 acked)
+{
+	struct tcp_sock *tp = tcp_sk(sk);
+	struct bictcp *ca = inet_csk_ca(sk);
+
+	if (!tcp_is_cwnd_limited(sk))
+		return;
+
+	if (tcp_in_slow_start(tp)) {
+		if (hystart && after(ack, ca->end_seq))
+			bictcp_hystart_reset(sk);
+		acked = tcp_slow_start(tp, acked);
+		if (!acked)
+			return;
+	}
+	bictcp_update(ca, tp->snd_cwnd, acked);
+	tcp_cong_avoid_ai(tp, ca->cnt, acked);
+}
+
+__u32 BPF_STRUCT_OPS(bictcp_recalc_ssthresh, struct sock *sk)
+{
+	const struct tcp_sock *tp = tcp_sk(sk);
+	struct bictcp *ca = inet_csk_ca(sk);
+
+	ca->epoch_start = 0;	/* end of epoch */
+
+	/* Wmax and fast convergence */
+	if (tp->snd_cwnd < ca->last_max_cwnd && fast_convergence)
+		ca->last_max_cwnd = (tp->snd_cwnd * (BICTCP_BETA_SCALE + beta))
+			/ (2 * BICTCP_BETA_SCALE);
+	else
+		ca->last_max_cwnd = tp->snd_cwnd;
+
+	return max((tp->snd_cwnd * beta) / BICTCP_BETA_SCALE, 2U);
+}
+
+void BPF_STRUCT_OPS(bictcp_state, struct sock *sk, __u8 new_state)
+{
+	if (new_state == TCP_CA_Loss) {
+		bictcp_reset(inet_csk_ca(sk));
+		bictcp_hystart_reset(sk);
+	}
+}
+
+#define GSO_MAX_SIZE		65536
+
+/* Account for TSO/GRO delays.
+ * Otherwise short RTT flows could get too small ssthresh, since during
+ * slow start we begin with small TSO packets and ca->delay_min would
+ * not account for long aggregation delay when TSO packets get bigger.
+ * Ideally even with a very small RTT we would like to have at least one
+ * TSO packet being sent and received by GRO, and another one in qdisc layer.
+ * We apply another 100% factor because @rate is doubled at this point.
+ * We cap the cushion to 1ms.
+ */
+static __always_inline __u32 hystart_ack_delay(struct sock *sk)
+{
+	unsigned long rate;
+
+	rate = sk->sk_pacing_rate;
+	if (!rate)
+		return 0;
+	return min((__u64)USEC_PER_MSEC,
+		   div64_ul((__u64)GSO_MAX_SIZE * 4 * USEC_PER_SEC, rate));
+}
+
+static __always_inline void hystart_update(struct sock *sk, __u32 delay)
+{
+	struct tcp_sock *tp = tcp_sk(sk);
+	struct bictcp *ca = inet_csk_ca(sk);
+	__u32 threshold;
+
+	if (hystart_detect & HYSTART_ACK_TRAIN) {
+		__u32 now = bictcp_clock_us(sk);
+
+		/* first detection parameter - ack-train detection */
+		if ((__s32)(now - ca->last_ack) <= hystart_ack_delta_us) {
+			ca->last_ack = now;
+
+			threshold = ca->delay_min + hystart_ack_delay(sk);
+
+			/* Hystart ack train triggers if we get ack past
+			 * ca->delay_min/2.
+			 * Pacing might have delayed packets up to RTT/2
+			 * during slow start.
+			 */
+			if (sk->sk_pacing_status == SK_PACING_NONE)
+				threshold >>= 1;
+
+			if ((__s32)(now - ca->round_start) > threshold) {
+				ca->found = 1;
+				tp->snd_ssthresh = tp->snd_cwnd;
+			}
+		}
+	}
+
+	if (hystart_detect & HYSTART_DELAY) {
+		/* obtain the minimum delay of more than sampling packets */
+		if (ca->sample_cnt < HYSTART_MIN_SAMPLES) {
+			if (ca->curr_rtt > delay)
+				ca->curr_rtt = delay;
+
+			ca->sample_cnt++;
+		} else {
+			if (ca->curr_rtt > ca->delay_min +
+			    HYSTART_DELAY_THRESH(ca->delay_min >> 3)) {
+				ca->found = 1;
+				tp->snd_ssthresh = tp->snd_cwnd;
+			}
+		}
+	}
+}
+
+void BPF_STRUCT_OPS(bictcp_acked, struct sock *sk,
+		    const struct ack_sample *sample)
+{
+	const struct tcp_sock *tp = tcp_sk(sk);
+	struct bictcp *ca = inet_csk_ca(sk);
+	__u32 delay;
+
+	/* Some calls are for duplicates without timetamps */
+	if (sample->rtt_us < 0)
+		return;
+
+	/* Discard delay samples right after fast recovery */
+	if (ca->epoch_start && (__s32)(tcp_jiffies32 - ca->epoch_start) < HZ)
+		return;
+
+	delay = sample->rtt_us;
+	if (delay == 0)
+		delay = 1;
+
+	/* first time call or link delay decreases */
+	if (ca->delay_min == 0 || ca->delay_min > delay)
+		ca->delay_min = delay;
+
+	/* hystart triggers when cwnd is larger than some threshold */
+	if (!ca->found && tcp_in_slow_start(tp) && hystart &&
+	    tp->snd_cwnd >= hystart_low_window)
+		hystart_update(sk, delay);
+}
+
+__u32 BPF_STRUCT_OPS(tcp_reno_undo_cwnd, struct sock *sk)
+{
+	const struct tcp_sock *tp = tcp_sk(sk);
+
+	return max(tp->snd_cwnd, tp->prior_cwnd);
+}
+
+SEC(".struct_ops")
+struct tcp_congestion_ops cubic = {
+	.init		= (void *)bictcp_init,
+	.ssthresh	= (void *)bictcp_recalc_ssthresh,
+	.cong_avoid	= (void *)bictcp_cong_avoid,
+	.set_state	= (void *)bictcp_state,
+	.undo_cwnd	= (void *)tcp_reno_undo_cwnd,
+	.cwnd_event	= (void *)bictcp_cwnd_event,
+	.pkts_acked     = (void *)bictcp_acked,
+	.name		= "bpf_cubic",
+};
-- 
cgit v1.2.3-59-g8ed1b


From ca9c74ae9be5e78541c2058db9a754947a7d4a9b Mon Sep 17 00:00:00 2001
From: Olof Johansson <olof@lixom.net>
Date: Thu, 19 Dec 2019 16:15:17 -0800
Subject: net/mlx5e: Fix printk format warning

Use "%zu" for size_t. Seen on ARM allmodconfig:

drivers/net/ethernet/mellanox/mlx5/core/wq.c: In function 'mlx5_wq_cyc_wqe_dump':
include/linux/kern_levels.h:5:18: warning: format '%ld' expects argument of type 'long int', but argument 5 has type 'size_t' {aka 'unsigned int'} [-Wformat=]

Fixes: 130c7b46c93d ("net/mlx5e: TX, Dump WQs wqe descriptors on CQE with error events")
Signed-off-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/wq.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/wq.c b/drivers/net/ethernet/mellanox/mlx5/core/wq.c
index f2a0e72285ba..02f7e4a39578 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/wq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/wq.c
@@ -89,7 +89,7 @@ void mlx5_wq_cyc_wqe_dump(struct mlx5_wq_cyc *wq, u16 ix, u8 nstrides)
 	len = nstrides << wq->fbc.log_stride;
 	wqe = mlx5_wq_cyc_get_wqe(wq, ix);
 
-	pr_info("WQE DUMP: WQ size %d WQ cur size %d, WQE index 0x%x, len: %ld\n",
+	pr_info("WQE DUMP: WQ size %d WQ cur size %d, WQE index 0x%x, len: %zu\n",
 		mlx5_wq_cyc_get_size(wq), wq->cur_sz, ix, len);
 	print_hex_dump(KERN_WARNING, "", DUMP_PREFIX_OFFSET, 16, 1, wqe, len, false);
 }
-- 
cgit v1.2.3-59-g8ed1b


From a1718505d7f67ee0ab051322f1cbc7ac42b5da82 Mon Sep 17 00:00:00 2001
From: Davide Caratti <dcaratti@redhat.com>
Date: Thu, 9 Jan 2020 12:07:59 +0100
Subject: net/mlx5e: allow TSO on VXLAN over VLAN topologies

since mlx5 hardware can segment correctly TSO packets on VXLAN over VLAN
topologies, CPU usage can improve significantly if we enable tunnel
offloads in dev->vlan_features, like it was done in the past with other
NIC drivers (e.g. mlx4, be2net and ixgbe).

Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 78737fd42616..87267c18ff8c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -4878,6 +4878,8 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev)
 		netdev->hw_enc_features |= NETIF_F_GSO_UDP_TUNNEL |
 					   NETIF_F_GSO_UDP_TUNNEL_CSUM;
 		netdev->gso_partial_features = NETIF_F_GSO_UDP_TUNNEL_CSUM;
+		netdev->vlan_features |= NETIF_F_GSO_UDP_TUNNEL |
+					 NETIF_F_GSO_UDP_TUNNEL_CSUM;
 	}
 
 	if (mlx5e_tunnel_proto_supported(mdev, IPPROTO_GRE)) {
-- 
cgit v1.2.3-59-g8ed1b


From e15cf98ee8a76472144a19a24ca73d26fefa5237 Mon Sep 17 00:00:00 2001
From: Chen Wandun <chenwandun@huawei.com>
Date: Mon, 20 Jan 2020 20:41:53 +0800
Subject: net/mlx5: make the symbol 'ESW_POOLS' static

Fix the following sparse warning:
drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.c:35:20: warning: symbol 'ESW_POOLS' was not declared. Should it be static?

Fixes: 39ac237ce009 ("net/mlx5: E-Switch, Refactor chains and priorities")
Signed-off-by: Chen Wandun <chenwandun@huawei.com>
Acked-by: Paul Blakey <paulb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.c
index 3a60eb5360bd..c5a446e295aa 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.c
@@ -32,10 +32,10 @@
  * pools.
  */
 #define ESW_SIZE (16 * 1024 * 1024)
-const unsigned int ESW_POOLS[] = { 4 * 1024 * 1024,
-				   1 * 1024 * 1024,
-				   64 * 1024,
-				   4 * 1024, };
+static const unsigned int ESW_POOLS[] = { 4 * 1024 * 1024,
+					  1 * 1024 * 1024,
+					  64 * 1024,
+					  4 * 1024, };
 
 struct mlx5_esw_chains_priv {
 	struct rhashtable chains_ht;
-- 
cgit v1.2.3-59-g8ed1b


From 6d65bc64e232896251daba7c43933f0f35859bc3 Mon Sep 17 00:00:00 2001
From: wenxu <wenxu@ucloud.cn>
Date: Tue, 7 Jan 2020 17:16:06 +0800
Subject: net/mlx5e: Add mlx5e_flower_parse_meta support

In the flowtables offload all the devices in the flowtables
share the same flow_block. An offload rule will be installed on
all the devices. This scenario is not correct.

It is no problem if there are only two devices in the flowtable,
The rule with ingress and egress on the same device can be reject
by driver.

But more than two devices in the flowtable will install the wrong
rules on hardware.

For example:
Three devices in a offload flowtables: dev_a, dev_b, dev_c

A rule ingress from dev_a and egress to dev_b:
The rule will install on device dev_a.
The rule will try to install on dev_b but failed for ingress
and egress on the same device.
The rule will install on dev_c. This is not correct.

The flowtables offload avoid this case through restricting the ingress dev
with FLOW_DISSECTOR_KEY_META.

So the mlx5e driver also should support the FLOW_DISSECTOR_KEY_META parse.

Signed-off-by: wenxu <wenxu@ucloud.cn>
Acked-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 39 +++++++++++++++++++++++++
 1 file changed, 39 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 26f559b453dc..4f184c770a45 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -1810,6 +1810,40 @@ static void *get_match_headers_value(u32 flags,
 			     outer_headers);
 }
 
+static int mlx5e_flower_parse_meta(struct net_device *filter_dev,
+				   struct flow_cls_offload *f)
+{
+	struct flow_rule *rule = flow_cls_offload_flow_rule(f);
+	struct netlink_ext_ack *extack = f->common.extack;
+	struct net_device *ingress_dev;
+	struct flow_match_meta match;
+
+	if (!flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_META))
+		return 0;
+
+	flow_rule_match_meta(rule, &match);
+	if (match.mask->ingress_ifindex != 0xFFFFFFFF) {
+		NL_SET_ERR_MSG_MOD(extack, "Unsupported ingress ifindex mask");
+		return -EINVAL;
+	}
+
+	ingress_dev = __dev_get_by_index(dev_net(filter_dev),
+					 match.key->ingress_ifindex);
+	if (!ingress_dev) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Can't find the ingress port to match on");
+		return -EINVAL;
+	}
+
+	if (ingress_dev != filter_dev) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Can't match on the ingress filter port");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
 static int __parse_cls_flower(struct mlx5e_priv *priv,
 			      struct mlx5_flow_spec *spec,
 			      struct flow_cls_offload *f,
@@ -1830,6 +1864,7 @@ static int __parse_cls_flower(struct mlx5e_priv *priv,
 	u16 addr_type = 0;
 	u8 ip_proto = 0;
 	u8 *match_level;
+	int err;
 
 	match_level = outer_match_level;
 
@@ -1873,6 +1908,10 @@ static int __parse_cls_flower(struct mlx5e_priv *priv,
 						    spec);
 	}
 
+	err = mlx5e_flower_parse_meta(filter_dev, f);
+	if (err)
+		return err;
+
 	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_BASIC)) {
 		struct flow_match_basic match;
 
-- 
cgit v1.2.3-59-g8ed1b


From a51dcc10a04ac476240befb875b0a21388a429cd Mon Sep 17 00:00:00 2001
From: Hamdan Igbaria <hamdani@mellanox.com>
Date: Tue, 24 Dec 2019 18:07:41 +0200
Subject: net/mlx5: DR, Modify set action limitation extension

Modify set actions are not supported on both tx
and rx, added a check for that.
Also refactored the code in a way that every modify
action has his own functions, this needed so in the
future we could add copy action more smoothly.

Signed-off-by: Hamdan Igbaria <hamdani@mellanox.com>
Signed-off-by: Alex Vesker <valex@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../mellanox/mlx5/core/steering/dr_action.c        | 226 +++++++++++++++------
 1 file changed, 165 insertions(+), 61 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
index 9359eed10889..ad32b88a83dc 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
@@ -1314,58 +1314,85 @@ not_found:
 }
 
 static int
-dr_action_modify_sw_to_hw(struct mlx5dr_domain *dmn,
-			  __be64 *sw_action,
-			  __be64 *hw_action,
-			  const struct dr_action_modify_field_conv **ret_hw_info)
+dr_action_modify_sw_to_hw_add(struct mlx5dr_domain *dmn,
+			      __be64 *sw_action,
+			      __be64 *hw_action,
+			      const struct dr_action_modify_field_conv **ret_hw_info)
 {
 	const struct dr_action_modify_field_conv *hw_action_info;
-	u8 offset, length, max_length, action;
+	u8 max_length;
 	u16 sw_field;
-	u8 hw_opcode;
 	u32 data;
 
 	/* Get SW modify action data */
-	action = MLX5_GET(set_action_in, sw_action, action_type);
-	length = MLX5_GET(set_action_in, sw_action, length);
-	offset = MLX5_GET(set_action_in, sw_action, offset);
 	sw_field = MLX5_GET(set_action_in, sw_action, field);
 	data = MLX5_GET(set_action_in, sw_action, data);
 
 	/* Convert SW data to HW modify action format */
 	hw_action_info = dr_action_modify_get_hw_info(sw_field);
 	if (!hw_action_info) {
-		mlx5dr_dbg(dmn, "Modify action invalid field given\n");
+		mlx5dr_dbg(dmn, "Modify add action invalid field given\n");
 		return -EINVAL;
 	}
 
 	max_length = hw_action_info->end - hw_action_info->start + 1;
 
-	switch (action) {
-	case MLX5_ACTION_TYPE_SET:
-		hw_opcode = MLX5DR_ACTION_MDFY_HW_OP_SET;
-		/* PRM defines that length zero specific length of 32bits */
-		if (!length)
-			length = 32;
+	MLX5_SET(dr_action_hw_set, hw_action,
+		 opcode, MLX5DR_ACTION_MDFY_HW_OP_ADD);
 
-		if (length + offset > max_length) {
-			mlx5dr_dbg(dmn, "Modify action length + offset exceeds limit\n");
-			return -EINVAL;
-		}
-		break;
+	MLX5_SET(dr_action_hw_set, hw_action, destination_field_code,
+		 hw_action_info->hw_field);
 
-	case MLX5_ACTION_TYPE_ADD:
-		hw_opcode = MLX5DR_ACTION_MDFY_HW_OP_ADD;
-		offset = 0;
-		length = max_length;
-		break;
+	MLX5_SET(dr_action_hw_set, hw_action, destination_left_shifter,
+		 hw_action_info->start);
 
-	default:
-		mlx5dr_info(dmn, "Unsupported action_type for modify action\n");
-		return -EOPNOTSUPP;
+	/* PRM defines that length zero specific length of 32bits */
+	MLX5_SET(dr_action_hw_set, hw_action, destination_length,
+		 max_length == 32 ? 0 : max_length);
+
+	MLX5_SET(dr_action_hw_set, hw_action, inline_data, data);
+
+	*ret_hw_info = hw_action_info;
+
+	return 0;
+}
+
+static int
+dr_action_modify_sw_to_hw_set(struct mlx5dr_domain *dmn,
+			      __be64 *sw_action,
+			      __be64 *hw_action,
+			      const struct dr_action_modify_field_conv **ret_hw_info)
+{
+	const struct dr_action_modify_field_conv *hw_action_info;
+	u8 offset, length, max_length;
+	u16 sw_field;
+	u32 data;
+
+	/* Get SW modify action data */
+	length = MLX5_GET(set_action_in, sw_action, length);
+	offset = MLX5_GET(set_action_in, sw_action, offset);
+	sw_field = MLX5_GET(set_action_in, sw_action, field);
+	data = MLX5_GET(set_action_in, sw_action, data);
+
+	/* Convert SW data to HW modify action format */
+	hw_action_info = dr_action_modify_get_hw_info(sw_field);
+	if (!hw_action_info) {
+		mlx5dr_dbg(dmn, "Modify set action invalid field given\n");
+		return -EINVAL;
 	}
 
-	MLX5_SET(dr_action_hw_set, hw_action, opcode, hw_opcode);
+	/* PRM defines that length zero specific length of 32bits */
+	length = length ? length : 32;
+
+	max_length = hw_action_info->end - hw_action_info->start + 1;
+
+	if (length + offset > max_length) {
+		mlx5dr_dbg(dmn, "Modify action length + offset exceeds limit\n");
+		return -EINVAL;
+	}
+
+	MLX5_SET(dr_action_hw_set, hw_action,
+		 opcode, MLX5DR_ACTION_MDFY_HW_OP_SET);
 
 	MLX5_SET(dr_action_hw_set, hw_action, destination_field_code,
 		 hw_action_info->hw_field);
@@ -1384,48 +1411,120 @@ dr_action_modify_sw_to_hw(struct mlx5dr_domain *dmn,
 }
 
 static int
-dr_action_modify_check_field_limitation(struct mlx5dr_domain *dmn,
-					const __be64 *sw_action)
+dr_action_modify_sw_to_hw(struct mlx5dr_domain *dmn,
+			  __be64 *sw_action,
+			  __be64 *hw_action,
+			  const struct dr_action_modify_field_conv **ret_hw_info)
 {
-	u16 sw_field;
 	u8 action;
+	int ret;
 
-	sw_field = MLX5_GET(set_action_in, sw_action, field);
+	*hw_action = 0;
+
+	/* Get SW modify action type */
 	action = MLX5_GET(set_action_in, sw_action, action_type);
 
-	/* Check if SW field is supported in current domain (RX/TX) */
-	if (action == MLX5_ACTION_TYPE_SET) {
-		if (sw_field == MLX5_ACTION_IN_FIELD_METADATA_REG_A) {
-			if (dmn->type != MLX5DR_DOMAIN_TYPE_NIC_TX) {
-				mlx5dr_dbg(dmn, "Unsupported field %d for RX/FDB set action\n",
-					   sw_field);
-				return -EINVAL;
-			}
-		}
+	switch (action) {
+	case MLX5_ACTION_TYPE_SET:
+		ret = dr_action_modify_sw_to_hw_set(dmn, sw_action,
+						    hw_action,
+						    ret_hw_info);
+		break;
 
-		if (sw_field == MLX5_ACTION_IN_FIELD_METADATA_REG_B) {
-			if (dmn->type != MLX5DR_DOMAIN_TYPE_NIC_RX) {
-				mlx5dr_dbg(dmn, "Unsupported field %d for TX/FDB set action\n",
-					   sw_field);
-				return -EINVAL;
-			}
+	case MLX5_ACTION_TYPE_ADD:
+		ret = dr_action_modify_sw_to_hw_add(dmn, sw_action,
+						    hw_action,
+						    ret_hw_info);
+		break;
+
+	default:
+		mlx5dr_info(dmn, "Unsupported action_type for modify action\n");
+		ret = -EOPNOTSUPP;
+	}
+
+	return ret;
+}
+
+static int
+dr_action_modify_check_set_field_limitation(struct mlx5dr_action *action,
+					    const __be64 *sw_action)
+{
+	u16 sw_field = MLX5_GET(set_action_in, sw_action, field);
+	struct mlx5dr_domain *dmn = action->rewrite.dmn;
+
+	if (sw_field == MLX5_ACTION_IN_FIELD_METADATA_REG_A) {
+		action->rewrite.allow_rx = 0;
+		if (dmn->type != MLX5DR_DOMAIN_TYPE_NIC_TX) {
+			mlx5dr_dbg(dmn, "Unsupported field %d for RX/FDB set action\n",
+				   sw_field);
+			return -EINVAL;
 		}
-	} else if (action == MLX5_ACTION_TYPE_ADD) {
-		if (sw_field != MLX5_ACTION_IN_FIELD_OUT_IP_TTL &&
-		    sw_field != MLX5_ACTION_IN_FIELD_OUT_IPV6_HOPLIMIT &&
-		    sw_field != MLX5_ACTION_IN_FIELD_OUT_TCP_SEQ_NUM &&
-		    sw_field != MLX5_ACTION_IN_FIELD_OUT_TCP_ACK_NUM) {
-			mlx5dr_dbg(dmn, "Unsupported field %d for add action\n", sw_field);
+	} else if (sw_field == MLX5_ACTION_IN_FIELD_METADATA_REG_B) {
+		action->rewrite.allow_tx = 0;
+		if (dmn->type != MLX5DR_DOMAIN_TYPE_NIC_RX) {
+			mlx5dr_dbg(dmn, "Unsupported field %d for TX/FDB set action\n",
+				   sw_field);
 			return -EINVAL;
 		}
-	} else {
-		mlx5dr_info(dmn, "Unsupported action %d modify action\n", action);
-		return -EOPNOTSUPP;
+	}
+
+	if (!action->rewrite.allow_rx && !action->rewrite.allow_tx) {
+		mlx5dr_dbg(dmn, "Modify SET actions not supported on both RX and TX\n");
+		return -EINVAL;
 	}
 
 	return 0;
 }
 
+static int
+dr_action_modify_check_add_field_limitation(struct mlx5dr_action *action,
+					    const __be64 *sw_action)
+{
+	u16 sw_field = MLX5_GET(set_action_in, sw_action, field);
+	struct mlx5dr_domain *dmn = action->rewrite.dmn;
+
+	if (sw_field != MLX5_ACTION_IN_FIELD_OUT_IP_TTL &&
+	    sw_field != MLX5_ACTION_IN_FIELD_OUT_IPV6_HOPLIMIT &&
+	    sw_field != MLX5_ACTION_IN_FIELD_OUT_TCP_SEQ_NUM &&
+	    sw_field != MLX5_ACTION_IN_FIELD_OUT_TCP_ACK_NUM) {
+		mlx5dr_dbg(dmn, "Unsupported field %d for add action\n",
+			   sw_field);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int
+dr_action_modify_check_field_limitation(struct mlx5dr_action *action,
+					const __be64 *sw_action)
+{
+	struct mlx5dr_domain *dmn = action->rewrite.dmn;
+	u8 action_type;
+	int ret;
+
+	action_type = MLX5_GET(set_action_in, sw_action, action_type);
+
+	switch (action_type) {
+	case MLX5_ACTION_TYPE_SET:
+		ret = dr_action_modify_check_set_field_limitation(action,
+								  sw_action);
+		break;
+
+	case MLX5_ACTION_TYPE_ADD:
+		ret = dr_action_modify_check_add_field_limitation(action,
+								  sw_action);
+		break;
+
+	default:
+		mlx5dr_info(dmn, "Unsupported action %d modify action\n",
+			    action_type);
+		ret = -EOPNOTSUPP;
+	}
+
+	return ret;
+}
+
 static bool
 dr_action_modify_check_is_ttl_modify(const u64 *sw_action)
 {
@@ -1434,7 +1533,7 @@ dr_action_modify_check_is_ttl_modify(const u64 *sw_action)
 	return sw_field == MLX5_ACTION_IN_FIELD_OUT_IP_TTL;
 }
 
-static int dr_actions_convert_modify_header(struct mlx5dr_domain *dmn,
+static int dr_actions_convert_modify_header(struct mlx5dr_action *action,
 					    u32 max_hw_actions,
 					    u32 num_sw_actions,
 					    __be64 sw_actions[],
@@ -1446,16 +1545,21 @@ static int dr_actions_convert_modify_header(struct mlx5dr_domain *dmn,
 	u16 hw_field = MLX5DR_ACTION_MDFY_HW_FLD_RESERVED;
 	u32 l3_type = MLX5DR_ACTION_MDFY_HW_HDR_L3_NONE;
 	u32 l4_type = MLX5DR_ACTION_MDFY_HW_HDR_L4_NONE;
+	struct mlx5dr_domain *dmn = action->rewrite.dmn;
 	int ret, i, hw_idx = 0;
 	__be64 *sw_action;
 	__be64 hw_action;
 
 	*modify_ttl = false;
 
+	action->rewrite.allow_rx = 1;
+	action->rewrite.allow_tx = 1;
+
 	for (i = 0; i < num_sw_actions; i++) {
 		sw_action = &sw_actions[i];
 
-		ret = dr_action_modify_check_field_limitation(dmn, sw_action);
+		ret = dr_action_modify_check_field_limitation(action,
+							      sw_action);
 		if (ret)
 			return ret;
 
@@ -1544,7 +1648,7 @@ static int dr_action_create_modify_action(struct mlx5dr_domain *dmn,
 		goto free_chunk;
 	}
 
-	ret = dr_actions_convert_modify_header(dmn,
+	ret = dr_actions_convert_modify_header(action,
 					       max_hw_actions,
 					       num_sw_actions,
 					       actions,
-- 
cgit v1.2.3-59-g8ed1b


From c21a49b360e16657f42f1a9269c3aa714738fec7 Mon Sep 17 00:00:00 2001
From: Hamdan Igbaria <hamdani@mellanox.com>
Date: Thu, 9 Jan 2020 13:27:16 +0200
Subject: net/mlx5: DR, Modify header copy support

Modify header supports ADD/SET and from this patch
also COPY. Copy allows to copy header fields and
metadata.

Signed-off-by: Hamdan Igbaria <hamdani@mellanox.com>
Signed-off-by: Alex Vesker <valex@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../mellanox/mlx5/core/steering/dr_action.c        | 150 ++++++++++++++++++---
 .../mellanox/mlx5/core/steering/mlx5_ifc_dr.h      |  16 +++
 2 files changed, 151 insertions(+), 15 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
index ad32b88a83dc..286fcec5eff2 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
@@ -1410,16 +1410,83 @@ dr_action_modify_sw_to_hw_set(struct mlx5dr_domain *dmn,
 	return 0;
 }
 
+static int
+dr_action_modify_sw_to_hw_copy(struct mlx5dr_domain *dmn,
+			       __be64 *sw_action,
+			       __be64 *hw_action,
+			       const struct dr_action_modify_field_conv **ret_dst_hw_info,
+			       const struct dr_action_modify_field_conv **ret_src_hw_info)
+{
+	u8 src_offset, dst_offset, src_max_length, dst_max_length, length;
+	const struct dr_action_modify_field_conv *hw_dst_action_info;
+	const struct dr_action_modify_field_conv *hw_src_action_info;
+	u16 src_field, dst_field;
+
+	/* Get SW modify action data */
+	src_field = MLX5_GET(copy_action_in, sw_action, src_field);
+	dst_field = MLX5_GET(copy_action_in, sw_action, dst_field);
+	src_offset = MLX5_GET(copy_action_in, sw_action, src_offset);
+	dst_offset = MLX5_GET(copy_action_in, sw_action, dst_offset);
+	length = MLX5_GET(copy_action_in, sw_action, length);
+
+	/* Convert SW data to HW modify action format */
+	hw_src_action_info = dr_action_modify_get_hw_info(src_field);
+	hw_dst_action_info = dr_action_modify_get_hw_info(dst_field);
+	if (!hw_src_action_info || !hw_dst_action_info) {
+		mlx5dr_dbg(dmn, "Modify copy action invalid field given\n");
+		return -EINVAL;
+	}
+
+	/* PRM defines that length zero specific length of 32bits */
+	length = length ? length : 32;
+
+	src_max_length = hw_src_action_info->end -
+			 hw_src_action_info->start + 1;
+	dst_max_length = hw_dst_action_info->end -
+			 hw_dst_action_info->start + 1;
+
+	if (length + src_offset > src_max_length ||
+	    length + dst_offset > dst_max_length) {
+		mlx5dr_dbg(dmn, "Modify action length + offset exceeds limit\n");
+		return -EINVAL;
+	}
+
+	MLX5_SET(dr_action_hw_copy, hw_action,
+		 opcode, MLX5DR_ACTION_MDFY_HW_OP_COPY);
+
+	MLX5_SET(dr_action_hw_copy, hw_action, destination_field_code,
+		 hw_dst_action_info->hw_field);
+
+	MLX5_SET(dr_action_hw_copy, hw_action, destination_left_shifter,
+		 hw_dst_action_info->start + dst_offset);
+
+	MLX5_SET(dr_action_hw_copy, hw_action, destination_length,
+		 length == 32 ? 0 : length);
+
+	MLX5_SET(dr_action_hw_copy, hw_action, source_field_code,
+		 hw_src_action_info->hw_field);
+
+	MLX5_SET(dr_action_hw_copy, hw_action, source_left_shifter,
+		 hw_src_action_info->start + dst_offset);
+
+	*ret_dst_hw_info = hw_dst_action_info;
+	*ret_src_hw_info = hw_src_action_info;
+
+	return 0;
+}
+
 static int
 dr_action_modify_sw_to_hw(struct mlx5dr_domain *dmn,
 			  __be64 *sw_action,
 			  __be64 *hw_action,
-			  const struct dr_action_modify_field_conv **ret_hw_info)
+			  const struct dr_action_modify_field_conv **ret_dst_hw_info,
+			  const struct dr_action_modify_field_conv **ret_src_hw_info)
 {
 	u8 action;
 	int ret;
 
 	*hw_action = 0;
+	*ret_src_hw_info = NULL;
 
 	/* Get SW modify action type */
 	action = MLX5_GET(set_action_in, sw_action, action_type);
@@ -1428,13 +1495,20 @@ dr_action_modify_sw_to_hw(struct mlx5dr_domain *dmn,
 	case MLX5_ACTION_TYPE_SET:
 		ret = dr_action_modify_sw_to_hw_set(dmn, sw_action,
 						    hw_action,
-						    ret_hw_info);
+						    ret_dst_hw_info);
 		break;
 
 	case MLX5_ACTION_TYPE_ADD:
 		ret = dr_action_modify_sw_to_hw_add(dmn, sw_action,
 						    hw_action,
-						    ret_hw_info);
+						    ret_dst_hw_info);
+		break;
+
+	case MLX5_ACTION_TYPE_COPY:
+		ret = dr_action_modify_sw_to_hw_copy(dmn, sw_action,
+						     hw_action,
+						     ret_dst_hw_info,
+						     ret_src_hw_info);
 		break;
 
 	default:
@@ -1495,6 +1569,43 @@ dr_action_modify_check_add_field_limitation(struct mlx5dr_action *action,
 	return 0;
 }
 
+static int
+dr_action_modify_check_copy_field_limitation(struct mlx5dr_action *action,
+					     const __be64 *sw_action)
+{
+	struct mlx5dr_domain *dmn = action->rewrite.dmn;
+	u16 sw_fields[2];
+	int i;
+
+	sw_fields[0] = MLX5_GET(copy_action_in, sw_action, src_field);
+	sw_fields[1] = MLX5_GET(copy_action_in, sw_action, dst_field);
+
+	for (i = 0; i < 2; i++) {
+		if (sw_fields[i] == MLX5_ACTION_IN_FIELD_METADATA_REG_A) {
+			action->rewrite.allow_rx = 0;
+			if (dmn->type != MLX5DR_DOMAIN_TYPE_NIC_TX) {
+				mlx5dr_dbg(dmn, "Unsupported field %d for RX/FDB set action\n",
+					   sw_fields[i]);
+				return -EINVAL;
+			}
+		} else if (sw_fields[i] == MLX5_ACTION_IN_FIELD_METADATA_REG_B) {
+			action->rewrite.allow_tx = 0;
+			if (dmn->type != MLX5DR_DOMAIN_TYPE_NIC_RX) {
+				mlx5dr_dbg(dmn, "Unsupported field %d for TX/FDB set action\n",
+					   sw_fields[i]);
+				return -EINVAL;
+			}
+		}
+	}
+
+	if (!action->rewrite.allow_rx && !action->rewrite.allow_tx) {
+		mlx5dr_dbg(dmn, "Modify copy actions not supported on both RX and TX\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
 static int
 dr_action_modify_check_field_limitation(struct mlx5dr_action *action,
 					const __be64 *sw_action)
@@ -1516,6 +1627,11 @@ dr_action_modify_check_field_limitation(struct mlx5dr_action *action,
 								  sw_action);
 		break;
 
+	case MLX5_ACTION_TYPE_COPY:
+		ret = dr_action_modify_check_copy_field_limitation(action,
+								   sw_action);
+		break;
+
 	default:
 		mlx5dr_info(dmn, "Unsupported action %d modify action\n",
 			    action_type);
@@ -1541,7 +1657,8 @@ static int dr_actions_convert_modify_header(struct mlx5dr_action *action,
 					    u32 *num_hw_actions,
 					    bool *modify_ttl)
 {
-	const struct dr_action_modify_field_conv *hw_action_info;
+	const struct dr_action_modify_field_conv *hw_dst_action_info;
+	const struct dr_action_modify_field_conv *hw_src_action_info;
 	u16 hw_field = MLX5DR_ACTION_MDFY_HW_FLD_RESERVED;
 	u32 l3_type = MLX5DR_ACTION_MDFY_HW_HDR_L3_NONE;
 	u32 l4_type = MLX5DR_ACTION_MDFY_HW_HDR_L4_NONE;
@@ -1570,32 +1687,35 @@ static int dr_actions_convert_modify_header(struct mlx5dr_action *action,
 		ret = dr_action_modify_sw_to_hw(dmn,
 						sw_action,
 						&hw_action,
-						&hw_action_info);
+						&hw_dst_action_info,
+						&hw_src_action_info);
 		if (ret)
 			return ret;
 
 		/* Due to a HW limitation we cannot modify 2 different L3 types */
-		if (l3_type && hw_action_info->l3_type &&
-		    hw_action_info->l3_type != l3_type) {
+		if (l3_type && hw_dst_action_info->l3_type &&
+		    hw_dst_action_info->l3_type != l3_type) {
 			mlx5dr_dbg(dmn, "Action list can't support two different L3 types\n");
 			return -EINVAL;
 		}
-		if (hw_action_info->l3_type)
-			l3_type = hw_action_info->l3_type;
+		if (hw_dst_action_info->l3_type)
+			l3_type = hw_dst_action_info->l3_type;
 
 		/* Due to a HW limitation we cannot modify two different L4 types */
-		if (l4_type && hw_action_info->l4_type &&
-		    hw_action_info->l4_type != l4_type) {
+		if (l4_type && hw_dst_action_info->l4_type &&
+		    hw_dst_action_info->l4_type != l4_type) {
 			mlx5dr_dbg(dmn, "Action list can't support two different L4 types\n");
 			return -EINVAL;
 		}
-		if (hw_action_info->l4_type)
-			l4_type = hw_action_info->l4_type;
+		if (hw_dst_action_info->l4_type)
+			l4_type = hw_dst_action_info->l4_type;
 
 		/* HW reads and executes two actions at once this means we
 		 * need to create a gap if two actions access the same field
 		 */
-		if ((hw_idx % 2) && hw_field == hw_action_info->hw_field) {
+		if ((hw_idx % 2) && (hw_field == hw_dst_action_info->hw_field ||
+				     (hw_src_action_info &&
+				      hw_field == hw_src_action_info->hw_field))) {
 			/* Check if after gap insertion the total number of HW
 			 * modify actions doesn't exceeds the limit
 			 */
@@ -1605,7 +1725,7 @@ static int dr_actions_convert_modify_header(struct mlx5dr_action *action,
 				return -EINVAL;
 			}
 		}
-		hw_field = hw_action_info->hw_field;
+		hw_field = hw_dst_action_info->hw_field;
 
 		hw_actions[hw_idx] = hw_action;
 		hw_idx++;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5_ifc_dr.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5_ifc_dr.h
index 1722f4668269..e01c3766c7de 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5_ifc_dr.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5_ifc_dr.h
@@ -32,6 +32,7 @@ enum {
 };
 
 enum {
+	MLX5DR_ACTION_MDFY_HW_OP_COPY		= 0x1,
 	MLX5DR_ACTION_MDFY_HW_OP_SET		= 0x2,
 	MLX5DR_ACTION_MDFY_HW_OP_ADD		= 0x3,
 };
@@ -625,4 +626,19 @@ struct mlx5_ifc_dr_action_hw_set_bits {
 	u8         inline_data[0x20];
 };
 
+struct mlx5_ifc_dr_action_hw_copy_bits {
+	u8         opcode[0x8];
+	u8         destination_field_code[0x8];
+	u8         reserved_at_10[0x2];
+	u8         destination_left_shifter[0x6];
+	u8         reserved_at_18[0x2];
+	u8         destination_length[0x6];
+
+	u8         reserved_at_20[0x8];
+	u8         source_field_code[0x8];
+	u8         reserved_at_30[0x2];
+	u8         source_left_shifter[0x6];
+	u8         reserved_at_38[0x8];
+};
+
 #endif /* MLX5_IFC_DR_H */
-- 
cgit v1.2.3-59-g8ed1b


From 4c7cea2f31f1dde9c578e4729e0d34f340052037 Mon Sep 17 00:00:00 2001
From: Yevgeny Kliteynik <kliteyn@mellanox.com>
Date: Mon, 20 Jan 2020 11:51:36 +0200
Subject: net/mlx5: DR, Allow connecting flow table to a lower/same level table

Allow connecting SW steering source table to a lower/same level
destination table.
Lifting this limitation is required to support Connection Tracking.

Signed-off-by: Yevgeny Kliteynik <kliteyn@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
index 286fcec5eff2..6dec2a550a10 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
@@ -677,9 +677,12 @@ int mlx5dr_actions_build_ste_arr(struct mlx5dr_matcher *matcher,
 					goto out_invalid_arg;
 				}
 				if (action->dest_tbl.tbl->level <= matcher->tbl->level) {
+					mlx5_core_warn_once(dmn->mdev,
+							    "Connecting table to a lower/same level destination table\n");
 					mlx5dr_dbg(dmn,
-						   "Destination table level should be higher than source table\n");
-					goto out_invalid_arg;
+						   "Connecting table at level %d to a destination table at level %d\n",
+						   matcher->tbl->level,
+						   action->dest_tbl.tbl->level);
 				}
 				attr.final_icm_addr = rx_rule ?
 					action->dest_tbl.tbl->rx.s_anchor->chunk->icm_addr :
-- 
cgit v1.2.3-59-g8ed1b


From 29b598ddb8d2d401257013c2af5e9f003bf84ec4 Mon Sep 17 00:00:00 2001
From: Roi Dayan <roid@mellanox.com>
Date: Wed, 23 Oct 2019 09:59:36 +0300
Subject: net/mlx5e: Move uplink rep init/cleanup code into own functions

Clean up the code and allows to call uplink rep init/cleanup
from different location later.
To be used later for a new uplink representor mode.

Signed-off-by: Roi Dayan <roid@mellanox.com>
Reviewed-by: Eli Britstein <elibr@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_rep.c | 81 +++++++++++++++---------
 1 file changed, 51 insertions(+), 30 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index 446eb4d6c983..60c79123824b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -1674,10 +1674,46 @@ static void mlx5e_cleanup_rep_rx(struct mlx5e_priv *priv)
 	mlx5e_close_drop_rq(&priv->drop_rq);
 }
 
+static int mlx5e_init_uplink_rep_tx(struct mlx5e_rep_priv *rpriv)
+{
+	struct mlx5_rep_uplink_priv *uplink_priv;
+	struct net_device *netdev;
+	struct mlx5e_priv *priv;
+	int err;
+
+	netdev = rpriv->netdev;
+	priv = netdev_priv(netdev);
+	uplink_priv = &rpriv->uplink_priv;
+
+	mutex_init(&uplink_priv->unready_flows_lock);
+	INIT_LIST_HEAD(&uplink_priv->unready_flows);
+
+	/* init shared tc flow table */
+	err = mlx5e_tc_esw_init(&uplink_priv->tc_ht);
+	if (err)
+		return err;
+
+	mlx5_init_port_tun_entropy(&uplink_priv->tun_entropy, priv->mdev);
+
+	/* init indirect block notifications */
+	INIT_LIST_HEAD(&uplink_priv->tc_indr_block_priv_list);
+	uplink_priv->netdevice_nb.notifier_call = mlx5e_nic_rep_netdevice_event;
+	err = register_netdevice_notifier(&uplink_priv->netdevice_nb);
+	if (err) {
+		mlx5_core_err(priv->mdev, "Failed to register netdev notifier\n");
+		goto tc_esw_cleanup;
+	}
+
+	return 0;
+
+tc_esw_cleanup:
+	mlx5e_tc_esw_cleanup(&uplink_priv->tc_ht);
+	return err;
+}
+
 static int mlx5e_init_rep_tx(struct mlx5e_priv *priv)
 {
 	struct mlx5e_rep_priv *rpriv = priv->ppriv;
-	struct mlx5_rep_uplink_priv *uplink_priv;
 	int err;
 
 	err = mlx5e_create_tises(priv);
@@ -1687,52 +1723,37 @@ static int mlx5e_init_rep_tx(struct mlx5e_priv *priv)
 	}
 
 	if (rpriv->rep->vport == MLX5_VPORT_UPLINK) {
-		uplink_priv = &rpriv->uplink_priv;
-
-		mutex_init(&uplink_priv->unready_flows_lock);
-		INIT_LIST_HEAD(&uplink_priv->unready_flows);
-
-		/* init shared tc flow table */
-		err = mlx5e_tc_esw_init(&uplink_priv->tc_ht);
+		err = mlx5e_init_uplink_rep_tx(rpriv);
 		if (err)
 			goto destroy_tises;
-
-		mlx5_init_port_tun_entropy(&uplink_priv->tun_entropy, priv->mdev);
-
-		/* init indirect block notifications */
-		INIT_LIST_HEAD(&uplink_priv->tc_indr_block_priv_list);
-		uplink_priv->netdevice_nb.notifier_call = mlx5e_nic_rep_netdevice_event;
-		err = register_netdevice_notifier(&uplink_priv->netdevice_nb);
-		if (err) {
-			mlx5_core_err(priv->mdev, "Failed to register netdev notifier\n");
-			goto tc_esw_cleanup;
-		}
 	}
 
 	return 0;
 
-tc_esw_cleanup:
-	mlx5e_tc_esw_cleanup(&uplink_priv->tc_ht);
 destroy_tises:
 	mlx5e_destroy_tises(priv);
 	return err;
 }
 
+static void mlx5e_cleanup_uplink_rep_tx(struct mlx5e_rep_priv *rpriv)
+{
+	/* clean indirect TC block notifications */
+	unregister_netdevice_notifier(&rpriv->uplink_priv.netdevice_nb);
+	mlx5e_rep_indr_clean_block_privs(rpriv);
+
+	/* delete shared tc flow table */
+	mlx5e_tc_esw_cleanup(&rpriv->uplink_priv.tc_ht);
+	mutex_destroy(&rpriv->uplink_priv.unready_flows_lock);
+}
+
 static void mlx5e_cleanup_rep_tx(struct mlx5e_priv *priv)
 {
 	struct mlx5e_rep_priv *rpriv = priv->ppriv;
 
 	mlx5e_destroy_tises(priv);
 
-	if (rpriv->rep->vport == MLX5_VPORT_UPLINK) {
-		/* clean indirect TC block notifications */
-		unregister_netdevice_notifier(&rpriv->uplink_priv.netdevice_nb);
-		mlx5e_rep_indr_clean_block_privs(rpriv);
-
-		/* delete shared tc flow table */
-		mlx5e_tc_esw_cleanup(&rpriv->uplink_priv.tc_ht);
-		mutex_destroy(&rpriv->uplink_priv.unready_flows_lock);
-	}
+	if (rpriv->rep->vport == MLX5_VPORT_UPLINK)
+		mlx5e_cleanup_uplink_rep_tx(rpriv);
 }
 
 static void mlx5e_rep_enable(struct mlx5e_priv *priv)
-- 
cgit v1.2.3-59-g8ed1b


From 3460c184dfb0b41118b5000ad78dd997c034b2b5 Mon Sep 17 00:00:00 2001
From: Saeed Mahameed <saeedm@mellanox.com>
Date: Mon, 20 Jan 2020 22:32:12 -0800
Subject: net/mlx5e: Profile specific stats groups

Attach stats groups array to the profiles and make the stats utility
functions (get_num, update, fill, fill_strings) generic and use the
profile->stats_grps rather the hardcoded NIC stats groups.

This will allow future extension to have per profile stats groups.

In this patch mlx5e NIC and IPoIB will still share the same stats
groups.

Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h       |  3 +-
 .../net/ethernet/mellanox/mlx5/core/en_ethtool.c   | 23 ++-------
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c  | 17 ++-----
 drivers/net/ethernet/mellanox/mlx5/core/en_stats.c | 59 +++++++++++++++++++++-
 drivers/net/ethernet/mellanox/mlx5/core/en_stats.h | 37 ++++++++------
 .../net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c  |  2 +
 6 files changed, 93 insertions(+), 48 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index fc80b59db9a8..f85d99d601a6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -892,6 +892,8 @@ struct mlx5e_profile {
 	int	(*update_rx)(struct mlx5e_priv *priv);
 	void	(*update_stats)(struct mlx5e_priv *priv);
 	void	(*update_carrier)(struct mlx5e_priv *priv);
+	unsigned int (*stats_grps_num)(struct mlx5e_priv *priv);
+	const struct mlx5e_stats_grp *stats_grps;
 	struct {
 		mlx5e_fp_handle_rx_cqe handle_rx_cqe;
 		mlx5e_fp_handle_rx_cqe handle_rx_cqe_mpwqe;
@@ -964,7 +966,6 @@ struct sk_buff *
 mlx5e_skb_from_cqe_nonlinear(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe,
 			     struct mlx5e_wqe_frag_info *wi, u32 cqe_bcnt);
 
-void mlx5e_update_stats(struct mlx5e_priv *priv);
 void mlx5e_get_stats(struct net_device *dev, struct rtnl_link_stats64 *stats);
 void mlx5e_fold_sw_stats64(struct mlx5e_priv *priv, struct rtnl_link_stats64 *s);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
index c6776f308d5e..d674cb679895 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
@@ -218,13 +218,9 @@ static const struct pflag_desc mlx5e_priv_flags[MLX5E_NUM_PFLAGS];
 
 int mlx5e_ethtool_get_sset_count(struct mlx5e_priv *priv, int sset)
 {
-	int i, num_stats = 0;
-
 	switch (sset) {
 	case ETH_SS_STATS:
-		for (i = 0; i < mlx5e_num_stats_grps; i++)
-			num_stats += mlx5e_stats_grps[i].get_num_stats(priv);
-		return num_stats;
+		return mlx5e_stats_total_num(priv);
 	case ETH_SS_PRIV_FLAGS:
 		return MLX5E_NUM_PFLAGS;
 	case ETH_SS_TEST:
@@ -242,14 +238,6 @@ static int mlx5e_get_sset_count(struct net_device *dev, int sset)
 	return mlx5e_ethtool_get_sset_count(priv, sset);
 }
 
-static void mlx5e_fill_stats_strings(struct mlx5e_priv *priv, u8 *data)
-{
-	int i, idx = 0;
-
-	for (i = 0; i < mlx5e_num_stats_grps; i++)
-		idx = mlx5e_stats_grps[i].fill_strings(priv, data, idx);
-}
-
 void mlx5e_ethtool_get_strings(struct mlx5e_priv *priv, u32 stringset, u8 *data)
 {
 	int i;
@@ -268,7 +256,7 @@ void mlx5e_ethtool_get_strings(struct mlx5e_priv *priv, u32 stringset, u8 *data)
 		break;
 
 	case ETH_SS_STATS:
-		mlx5e_fill_stats_strings(priv, data);
+		mlx5e_stats_fill_strings(priv, data);
 		break;
 	}
 }
@@ -283,14 +271,13 @@ static void mlx5e_get_strings(struct net_device *dev, u32 stringset, u8 *data)
 void mlx5e_ethtool_get_ethtool_stats(struct mlx5e_priv *priv,
 				     struct ethtool_stats *stats, u64 *data)
 {
-	int i, idx = 0;
+	int idx = 0;
 
 	mutex_lock(&priv->state_lock);
-	mlx5e_update_stats(priv);
+	mlx5e_stats_update(priv);
 	mutex_unlock(&priv->state_lock);
 
-	for (i = 0; i < mlx5e_num_stats_grps; i++)
-		idx = mlx5e_stats_grps[i].fill_stats(priv, data, idx);
+	mlx5e_stats_fill(priv, data, idx);
 }
 
 static void mlx5e_get_ethtool_stats(struct net_device *dev,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 87267c18ff8c..a1bb9eb8e3b5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -159,23 +159,14 @@ static void mlx5e_update_carrier_work(struct work_struct *work)
 	mutex_unlock(&priv->state_lock);
 }
 
-void mlx5e_update_stats(struct mlx5e_priv *priv)
-{
-	int i;
-
-	for (i = mlx5e_num_stats_grps - 1; i >= 0; i--)
-		if (mlx5e_stats_grps[i].update_stats)
-			mlx5e_stats_grps[i].update_stats(priv);
-}
-
 void mlx5e_update_ndo_stats(struct mlx5e_priv *priv)
 {
 	int i;
 
-	for (i = mlx5e_num_stats_grps - 1; i >= 0; i--)
-		if (mlx5e_stats_grps[i].update_stats_mask &
+	for (i = mlx5e_nic_stats_grps_num(priv) - 1; i >= 0; i--)
+		if (mlx5e_nic_stats_grps[i].update_stats_mask &
 		    MLX5E_NDO_UPDATE_STATS)
-			mlx5e_stats_grps[i].update_stats(priv);
+			mlx5e_nic_stats_grps[i].update_stats(priv);
 }
 
 static void mlx5e_update_stats_work(struct work_struct *work)
@@ -5197,6 +5188,8 @@ static const struct mlx5e_profile mlx5e_nic_profile = {
 	.rx_handlers.handle_rx_cqe_mpwqe = mlx5e_handle_rx_cqe_mpwrq,
 	.max_tc		   = MLX5E_MAX_NUM_TC,
 	.rq_groups	   = MLX5E_NUM_RQ_GROUPS(XSK),
+	.stats_grps	   = mlx5e_nic_stats_grps,
+	.stats_grps_num	   = mlx5e_nic_stats_grps_num,
 };
 
 /* mlx5e generic netdev management API (move to en_common.c) */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
index 4291db78efc9..a8baa6298d95 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
@@ -35,6 +35,58 @@
 #include "en_accel/ipsec.h"
 #include "en_accel/tls.h"
 
+static unsigned int stats_grps_num(struct mlx5e_priv *priv)
+{
+	return !priv->profile->stats_grps_num ? 0 :
+		priv->profile->stats_grps_num(priv);
+}
+
+unsigned int mlx5e_stats_total_num(struct mlx5e_priv *priv)
+{
+	const struct mlx5e_stats_grp *stats_grps = priv->profile->stats_grps;
+	const unsigned int num_stats_grps = stats_grps_num(priv);
+	unsigned int total = 0;
+	int i;
+
+	for (i = 0; i < num_stats_grps; i++)
+		total += stats_grps[i].get_num_stats(priv);
+
+	return total;
+}
+
+void mlx5e_stats_update(struct mlx5e_priv *priv)
+{
+	const struct mlx5e_stats_grp *stats_grps = priv->profile->stats_grps;
+	const unsigned int num_stats_grps = stats_grps_num(priv);
+	int i;
+
+	for (i = num_stats_grps - 1; i >= 0; i--)
+		if (stats_grps[i].update_stats)
+			stats_grps[i].update_stats(priv);
+}
+
+void mlx5e_stats_fill(struct mlx5e_priv *priv, u64 *data, int idx)
+{
+	const struct mlx5e_stats_grp *stats_grps = priv->profile->stats_grps;
+	const unsigned int num_stats_grps = stats_grps_num(priv);
+	int i;
+
+	for (i = 0; i < num_stats_grps; i++)
+		idx = stats_grps[i].fill_stats(priv, data, idx);
+}
+
+void mlx5e_stats_fill_strings(struct mlx5e_priv *priv, u8 *data)
+{
+	const struct mlx5e_stats_grp *stats_grps = priv->profile->stats_grps;
+	const unsigned int num_stats_grps = stats_grps_num(priv);
+	int i, idx = 0;
+
+	for (i = 0; i < num_stats_grps; i++)
+		idx = stats_grps[i].fill_strings(priv, data, idx);
+}
+
+/* Concrete NIC Stats */
+
 static const struct counter_desc sw_stats_desc[] = {
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_packets) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_bytes) },
@@ -1669,7 +1721,7 @@ static int mlx5e_grp_channels_fill_stats(struct mlx5e_priv *priv, u64 *data,
 }
 
 /* The stats groups order is opposite to the update_stats() order calls */
-const struct mlx5e_stats_grp mlx5e_stats_grps[] = {
+const struct mlx5e_stats_grp mlx5e_nic_stats_grps[] = {
 	{
 		.get_num_stats = mlx5e_grp_sw_get_num_stats,
 		.fill_strings = mlx5e_grp_sw_fill_strings,
@@ -1768,4 +1820,7 @@ const struct mlx5e_stats_grp mlx5e_stats_grps[] = {
 	},
 };
 
-const int mlx5e_num_stats_grps = ARRAY_SIZE(mlx5e_stats_grps);
+unsigned int mlx5e_nic_stats_grps_num(struct mlx5e_priv *priv)
+{
+	return ARRAY_SIZE(mlx5e_nic_stats_grps);
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
index 869f3502f631..06eeedaacb88 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
@@ -55,6 +55,26 @@ struct counter_desc {
 	size_t		offset; /* Byte offset */
 };
 
+enum {
+	MLX5E_NDO_UPDATE_STATS = BIT(0x1),
+};
+
+struct mlx5e_priv;
+struct mlx5e_stats_grp {
+	u16 update_stats_mask;
+	int (*get_num_stats)(struct mlx5e_priv *priv);
+	int (*fill_strings)(struct mlx5e_priv *priv, u8 *data, int idx);
+	int (*fill_stats)(struct mlx5e_priv *priv, u64 *data, int idx);
+	void (*update_stats)(struct mlx5e_priv *priv);
+};
+
+unsigned int mlx5e_stats_total_num(struct mlx5e_priv *priv);
+void mlx5e_stats_update(struct mlx5e_priv *priv);
+void mlx5e_stats_fill(struct mlx5e_priv *priv, u64 *data, int idx);
+void mlx5e_stats_fill_strings(struct mlx5e_priv *priv, u8 *data);
+
+/* Concrete NIC Stats */
+
 struct mlx5e_sw_stats {
 	u64 rx_packets;
 	u64 rx_bytes;
@@ -322,21 +342,8 @@ struct mlx5e_stats {
 	struct mlx5e_pcie_stats pcie;
 };
 
-enum {
-	MLX5E_NDO_UPDATE_STATS = BIT(0x1),
-};
-
-struct mlx5e_priv;
-struct mlx5e_stats_grp {
-	u16 update_stats_mask;
-	int (*get_num_stats)(struct mlx5e_priv *priv);
-	int (*fill_strings)(struct mlx5e_priv *priv, u8 *data, int idx);
-	int (*fill_stats)(struct mlx5e_priv *priv, u64 *data, int idx);
-	void (*update_stats)(struct mlx5e_priv *priv);
-};
-
-extern const struct mlx5e_stats_grp mlx5e_stats_grps[];
-extern const int mlx5e_num_stats_grps;
+extern const struct mlx5e_stats_grp mlx5e_nic_stats_grps[];
+unsigned int mlx5e_nic_stats_grps_num(struct mlx5e_priv *priv);
 
 void mlx5e_grp_802_3_update_stats(struct mlx5e_priv *priv);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c
index 7c87f523e370..bd870ffeb1be 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c
@@ -435,6 +435,8 @@ static const struct mlx5e_profile mlx5i_nic_profile = {
 	.rx_handlers.handle_rx_cqe_mpwqe = NULL, /* Not supported */
 	.max_tc		   = MLX5I_MAX_NUM_TC,
 	.rq_groups	   = MLX5E_NUM_RQ_GROUPS(REGULAR),
+	.stats_grps        = mlx5e_nic_stats_grps,
+	.stats_grps_num    = mlx5e_nic_stats_grps_num,
 };
 
 /* mlx5i netdev NDos */
-- 
cgit v1.2.3-59-g8ed1b


From 96b12796550d0f0a7613f03efa0140f39120946b Mon Sep 17 00:00:00 2001
From: Saeed Mahameed <saeedm@mellanox.com>
Date: Tue, 21 Jan 2020 00:24:53 -0800
Subject: net/mlx5e: Declare stats groups via macro

Introduce new macros to declare stats callbacks and groups, for better
code reuse and for individual groups selection per profile which will be
introduced in next patches.

Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_rep.c   |   2 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_stats.c | 266 +++++++--------------
 drivers/net/ethernet/mellanox/mlx5/core/en_stats.h |  25 +-
 3 files changed, 108 insertions(+), 185 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index 60c79123824b..f00e17f78ec9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -162,7 +162,7 @@ static void mlx5e_uplink_rep_update_hw_counters(struct mlx5e_priv *priv)
 	struct mlx5e_pport_stats *pstats = &priv->stats.pport;
 	struct rtnl_link_stats64 *vport_stats;
 
-	mlx5e_grp_802_3_update_stats(priv);
+	MLX5E_STATS_GRP_OP(802_3, update_stats)(priv);
 
 	vport_stats = &priv->stats.vf_vport;
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
index a8baa6298d95..e903a15e7289 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
@@ -198,12 +198,12 @@ static const struct counter_desc sw_stats_desc[] = {
 
 #define NUM_SW_COUNTERS			ARRAY_SIZE(sw_stats_desc)
 
-static int mlx5e_grp_sw_get_num_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_NUM_STATS(sw)
 {
 	return NUM_SW_COUNTERS;
 }
 
-static int mlx5e_grp_sw_fill_strings(struct mlx5e_priv *priv, u8 *data, int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STRS(sw)
 {
 	int i;
 
@@ -212,7 +212,7 @@ static int mlx5e_grp_sw_fill_strings(struct mlx5e_priv *priv, u8 *data, int idx)
 	return idx;
 }
 
-static int mlx5e_grp_sw_fill_stats(struct mlx5e_priv *priv, u64 *data, int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(sw)
 {
 	int i;
 
@@ -221,7 +221,7 @@ static int mlx5e_grp_sw_fill_stats(struct mlx5e_priv *priv, u64 *data, int idx)
 	return idx;
 }
 
-static void mlx5e_grp_sw_update_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(sw)
 {
 	struct mlx5e_sw_stats *s = &priv->stats.sw;
 	int i;
@@ -367,7 +367,7 @@ static const struct counter_desc drop_rq_stats_desc[] = {
 #define NUM_Q_COUNTERS			ARRAY_SIZE(q_stats_desc)
 #define NUM_DROP_RQ_COUNTERS		ARRAY_SIZE(drop_rq_stats_desc)
 
-static int mlx5e_grp_q_get_num_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_NUM_STATS(qcnt)
 {
 	int num_stats = 0;
 
@@ -380,7 +380,7 @@ static int mlx5e_grp_q_get_num_stats(struct mlx5e_priv *priv)
 	return num_stats;
 }
 
-static int mlx5e_grp_q_fill_strings(struct mlx5e_priv *priv, u8 *data, int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STRS(qcnt)
 {
 	int i;
 
@@ -395,7 +395,7 @@ static int mlx5e_grp_q_fill_strings(struct mlx5e_priv *priv, u8 *data, int idx)
 	return idx;
 }
 
-static int mlx5e_grp_q_fill_stats(struct mlx5e_priv *priv, u64 *data, int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(qcnt)
 {
 	int i;
 
@@ -408,7 +408,7 @@ static int mlx5e_grp_q_fill_stats(struct mlx5e_priv *priv, u64 *data, int idx)
 	return idx;
 }
 
-static void mlx5e_grp_q_update_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(qcnt)
 {
 	struct mlx5e_qcounter_stats *qcnt = &priv->stats.qcnt;
 	u32 out[MLX5_ST_SZ_DW(query_q_counter_out)];
@@ -443,14 +443,13 @@ static const struct counter_desc vnic_env_stats_dev_oob_desc[] = {
 	(MLX5_CAP_GEN(dev, vnic_env_int_rq_oob) ? \
 	 ARRAY_SIZE(vnic_env_stats_dev_oob_desc) : 0)
 
-static int mlx5e_grp_vnic_env_get_num_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_NUM_STATS(vnic_env)
 {
 	return NUM_VNIC_ENV_STEER_COUNTERS(priv->mdev) +
 		NUM_VNIC_ENV_DEV_OOB_COUNTERS(priv->mdev);
 }
 
-static int mlx5e_grp_vnic_env_fill_strings(struct mlx5e_priv *priv, u8 *data,
-					   int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STRS(vnic_env)
 {
 	int i;
 
@@ -464,8 +463,7 @@ static int mlx5e_grp_vnic_env_fill_strings(struct mlx5e_priv *priv, u8 *data,
 	return idx;
 }
 
-static int mlx5e_grp_vnic_env_fill_stats(struct mlx5e_priv *priv, u64 *data,
-					 int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(vnic_env)
 {
 	int i;
 
@@ -479,7 +477,7 @@ static int mlx5e_grp_vnic_env_fill_stats(struct mlx5e_priv *priv, u64 *data,
 	return idx;
 }
 
-static void mlx5e_grp_vnic_env_update_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(vnic_env)
 {
 	u32 *out = (u32 *)priv->stats.vnic.query_vnic_env_out;
 	int outlen = MLX5_ST_SZ_BYTES(query_vnic_env_out);
@@ -542,13 +540,12 @@ static const struct counter_desc vport_stats_desc[] = {
 
 #define NUM_VPORT_COUNTERS		ARRAY_SIZE(vport_stats_desc)
 
-static int mlx5e_grp_vport_get_num_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_NUM_STATS(vport)
 {
 	return NUM_VPORT_COUNTERS;
 }
 
-static int mlx5e_grp_vport_fill_strings(struct mlx5e_priv *priv, u8 *data,
-					int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STRS(vport)
 {
 	int i;
 
@@ -557,8 +554,7 @@ static int mlx5e_grp_vport_fill_strings(struct mlx5e_priv *priv, u8 *data,
 	return idx;
 }
 
-static int mlx5e_grp_vport_fill_stats(struct mlx5e_priv *priv, u64 *data,
-				      int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(vport)
 {
 	int i;
 
@@ -568,7 +564,7 @@ static int mlx5e_grp_vport_fill_stats(struct mlx5e_priv *priv, u64 *data,
 	return idx;
 }
 
-static void mlx5e_grp_vport_update_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(vport)
 {
 	int outlen = MLX5_ST_SZ_BYTES(query_vport_counter_out);
 	u32 *out = (u32 *)priv->stats.vport.query_vport_out;
@@ -607,13 +603,12 @@ static const struct counter_desc pport_802_3_stats_desc[] = {
 
 #define NUM_PPORT_802_3_COUNTERS	ARRAY_SIZE(pport_802_3_stats_desc)
 
-static int mlx5e_grp_802_3_get_num_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_NUM_STATS(802_3)
 {
 	return NUM_PPORT_802_3_COUNTERS;
 }
 
-static int mlx5e_grp_802_3_fill_strings(struct mlx5e_priv *priv, u8 *data,
-					int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STRS(802_3)
 {
 	int i;
 
@@ -622,8 +617,7 @@ static int mlx5e_grp_802_3_fill_strings(struct mlx5e_priv *priv, u8 *data,
 	return idx;
 }
 
-static int mlx5e_grp_802_3_fill_stats(struct mlx5e_priv *priv, u64 *data,
-				      int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(802_3)
 {
 	int i;
 
@@ -636,7 +630,7 @@ static int mlx5e_grp_802_3_fill_stats(struct mlx5e_priv *priv, u64 *data,
 #define MLX5_BASIC_PPCNT_SUPPORTED(mdev) \
 	(MLX5_CAP_GEN(mdev, pcam_reg) ? MLX5_CAP_PCAM_REG(mdev, ppcnt) : 1)
 
-void mlx5e_grp_802_3_update_stats(struct mlx5e_priv *priv)
+MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(802_3)
 {
 	struct mlx5e_pport_stats *pstats = &priv->stats.pport;
 	struct mlx5_core_dev *mdev = priv->mdev;
@@ -664,13 +658,12 @@ static const struct counter_desc pport_2863_stats_desc[] = {
 
 #define NUM_PPORT_2863_COUNTERS		ARRAY_SIZE(pport_2863_stats_desc)
 
-static int mlx5e_grp_2863_get_num_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_NUM_STATS(2863)
 {
 	return NUM_PPORT_2863_COUNTERS;
 }
 
-static int mlx5e_grp_2863_fill_strings(struct mlx5e_priv *priv, u8 *data,
-				       int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STRS(2863)
 {
 	int i;
 
@@ -679,8 +672,7 @@ static int mlx5e_grp_2863_fill_strings(struct mlx5e_priv *priv, u8 *data,
 	return idx;
 }
 
-static int mlx5e_grp_2863_fill_stats(struct mlx5e_priv *priv, u64 *data,
-				     int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(2863)
 {
 	int i;
 
@@ -690,7 +682,7 @@ static int mlx5e_grp_2863_fill_stats(struct mlx5e_priv *priv, u64 *data,
 	return idx;
 }
 
-static void mlx5e_grp_2863_update_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(2863)
 {
 	struct mlx5e_pport_stats *pstats = &priv->stats.pport;
 	struct mlx5_core_dev *mdev = priv->mdev;
@@ -725,13 +717,12 @@ static const struct counter_desc pport_2819_stats_desc[] = {
 
 #define NUM_PPORT_2819_COUNTERS		ARRAY_SIZE(pport_2819_stats_desc)
 
-static int mlx5e_grp_2819_get_num_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_NUM_STATS(2819)
 {
 	return NUM_PPORT_2819_COUNTERS;
 }
 
-static int mlx5e_grp_2819_fill_strings(struct mlx5e_priv *priv, u8 *data,
-				       int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STRS(2819)
 {
 	int i;
 
@@ -740,8 +731,7 @@ static int mlx5e_grp_2819_fill_strings(struct mlx5e_priv *priv, u8 *data,
 	return idx;
 }
 
-static int mlx5e_grp_2819_fill_stats(struct mlx5e_priv *priv, u64 *data,
-				     int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(2819)
 {
 	int i;
 
@@ -751,7 +741,7 @@ static int mlx5e_grp_2819_fill_stats(struct mlx5e_priv *priv, u64 *data,
 	return idx;
 }
 
-static void mlx5e_grp_2819_update_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(2819)
 {
 	struct mlx5e_pport_stats *pstats = &priv->stats.pport;
 	struct mlx5_core_dev *mdev = priv->mdev;
@@ -789,7 +779,7 @@ pport_phy_statistical_err_lanes_stats_desc[] = {
 #define NUM_PPORT_PHY_STATISTICAL_PER_LANE_COUNTERS \
 	ARRAY_SIZE(pport_phy_statistical_err_lanes_stats_desc)
 
-static int mlx5e_grp_phy_get_num_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_NUM_STATS(phy)
 {
 	struct mlx5_core_dev *mdev = priv->mdev;
 	int num_stats;
@@ -806,8 +796,7 @@ static int mlx5e_grp_phy_get_num_stats(struct mlx5e_priv *priv)
 	return num_stats;
 }
 
-static int mlx5e_grp_phy_fill_strings(struct mlx5e_priv *priv, u8 *data,
-				      int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STRS(phy)
 {
 	struct mlx5_core_dev *mdev = priv->mdev;
 	int i;
@@ -829,7 +818,7 @@ static int mlx5e_grp_phy_fill_strings(struct mlx5e_priv *priv, u8 *data,
 	return idx;
 }
 
-static int mlx5e_grp_phy_fill_stats(struct mlx5e_priv *priv, u64 *data, int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(phy)
 {
 	struct mlx5_core_dev *mdev = priv->mdev;
 	int i;
@@ -855,7 +844,7 @@ static int mlx5e_grp_phy_fill_stats(struct mlx5e_priv *priv, u64 *data, int idx)
 	return idx;
 }
 
-static void mlx5e_grp_phy_update_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(phy)
 {
 	struct mlx5e_pport_stats *pstats = &priv->stats.pport;
 	struct mlx5_core_dev *mdev = priv->mdev;
@@ -885,7 +874,7 @@ static const struct counter_desc pport_eth_ext_stats_desc[] = {
 
 #define NUM_PPORT_ETH_EXT_COUNTERS	ARRAY_SIZE(pport_eth_ext_stats_desc)
 
-static int mlx5e_grp_eth_ext_get_num_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_NUM_STATS(eth_ext)
 {
 	if (MLX5_CAP_PCAM_FEATURE((priv)->mdev, rx_buffer_fullness_counters))
 		return NUM_PPORT_ETH_EXT_COUNTERS;
@@ -893,8 +882,7 @@ static int mlx5e_grp_eth_ext_get_num_stats(struct mlx5e_priv *priv)
 	return 0;
 }
 
-static int mlx5e_grp_eth_ext_fill_strings(struct mlx5e_priv *priv, u8 *data,
-					  int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STRS(eth_ext)
 {
 	int i;
 
@@ -905,8 +893,7 @@ static int mlx5e_grp_eth_ext_fill_strings(struct mlx5e_priv *priv, u8 *data,
 	return idx;
 }
 
-static int mlx5e_grp_eth_ext_fill_stats(struct mlx5e_priv *priv, u64 *data,
-					int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(eth_ext)
 {
 	int i;
 
@@ -918,7 +905,7 @@ static int mlx5e_grp_eth_ext_fill_stats(struct mlx5e_priv *priv, u64 *data,
 	return idx;
 }
 
-static void mlx5e_grp_eth_ext_update_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(eth_ext)
 {
 	struct mlx5e_pport_stats *pstats = &priv->stats.pport;
 	struct mlx5_core_dev *mdev = priv->mdev;
@@ -959,7 +946,7 @@ static const struct counter_desc pcie_perf_stall_stats_desc[] = {
 #define NUM_PCIE_PERF_COUNTERS64	ARRAY_SIZE(pcie_perf_stats_desc64)
 #define NUM_PCIE_PERF_STALL_COUNTERS	ARRAY_SIZE(pcie_perf_stall_stats_desc)
 
-static int mlx5e_grp_pcie_get_num_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_NUM_STATS(pcie)
 {
 	int num_stats = 0;
 
@@ -975,8 +962,7 @@ static int mlx5e_grp_pcie_get_num_stats(struct mlx5e_priv *priv)
 	return num_stats;
 }
 
-static int mlx5e_grp_pcie_fill_strings(struct mlx5e_priv *priv, u8 *data,
-				       int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STRS(pcie)
 {
 	int i;
 
@@ -997,8 +983,7 @@ static int mlx5e_grp_pcie_fill_strings(struct mlx5e_priv *priv, u8 *data,
 	return idx;
 }
 
-static int mlx5e_grp_pcie_fill_stats(struct mlx5e_priv *priv, u64 *data,
-				     int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(pcie)
 {
 	int i;
 
@@ -1022,7 +1007,7 @@ static int mlx5e_grp_pcie_fill_stats(struct mlx5e_priv *priv, u64 *data,
 	return idx;
 }
 
-static void mlx5e_grp_pcie_update_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(pcie)
 {
 	struct mlx5e_pcie_stats *pcie_stats = &priv->stats.pcie;
 	struct mlx5_core_dev *mdev = priv->mdev;
@@ -1070,8 +1055,7 @@ static int mlx5e_grp_per_tc_prio_get_num_stats(struct mlx5e_priv *priv)
 	return NUM_PPORT_PER_TC_PRIO_COUNTERS * NUM_PPORT_PRIO;
 }
 
-static int mlx5e_grp_per_port_buffer_congest_fill_strings(struct mlx5e_priv *priv,
-							  u8 *data, int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STRS(per_port_buff_congest)
 {
 	struct mlx5_core_dev *mdev = priv->mdev;
 	int i, prio;
@@ -1091,8 +1075,7 @@ static int mlx5e_grp_per_port_buffer_congest_fill_strings(struct mlx5e_priv *pri
 	return idx;
 }
 
-static int mlx5e_grp_per_port_buffer_congest_fill_stats(struct mlx5e_priv *priv,
-							u64 *data, int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(per_port_buff_congest)
 {
 	struct mlx5e_pport_stats *pport = &priv->stats.pport;
 	struct mlx5_core_dev *mdev = priv->mdev;
@@ -1167,13 +1150,13 @@ static void mlx5e_grp_per_tc_congest_prio_update_stats(struct mlx5e_priv *priv)
 	}
 }
 
-static int mlx5e_grp_per_port_buffer_congest_get_num_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_NUM_STATS(per_port_buff_congest)
 {
 	return mlx5e_grp_per_tc_prio_get_num_stats(priv) +
 		mlx5e_grp_per_tc_congest_prio_get_num_stats(priv);
 }
 
-static void mlx5e_grp_per_port_buffer_congest_update_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(per_port_buff_congest)
 {
 	mlx5e_grp_per_tc_prio_update_stats(priv);
 	mlx5e_grp_per_tc_congest_prio_update_stats(priv);
@@ -1348,29 +1331,27 @@ static int mlx5e_grp_per_prio_pfc_fill_stats(struct mlx5e_priv *priv,
 	return idx;
 }
 
-static int mlx5e_grp_per_prio_get_num_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_NUM_STATS(per_prio)
 {
 	return mlx5e_grp_per_prio_traffic_get_num_stats() +
 		mlx5e_grp_per_prio_pfc_get_num_stats(priv);
 }
 
-static int mlx5e_grp_per_prio_fill_strings(struct mlx5e_priv *priv, u8 *data,
-					   int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STRS(per_prio)
 {
 	idx = mlx5e_grp_per_prio_traffic_fill_strings(priv, data, idx);
 	idx = mlx5e_grp_per_prio_pfc_fill_strings(priv, data, idx);
 	return idx;
 }
 
-static int mlx5e_grp_per_prio_fill_stats(struct mlx5e_priv *priv, u64 *data,
-					 int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(per_prio)
 {
 	idx = mlx5e_grp_per_prio_traffic_fill_stats(priv, data, idx);
 	idx = mlx5e_grp_per_prio_pfc_fill_stats(priv, data, idx);
 	return idx;
 }
 
-static void mlx5e_grp_per_prio_update_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(per_prio)
 {
 	struct mlx5e_pport_stats *pstats = &priv->stats.pport;
 	struct mlx5_core_dev *mdev = priv->mdev;
@@ -1405,13 +1386,12 @@ static const struct counter_desc mlx5e_pme_error_desc[] = {
 #define NUM_PME_STATUS_STATS		ARRAY_SIZE(mlx5e_pme_status_desc)
 #define NUM_PME_ERR_STATS		ARRAY_SIZE(mlx5e_pme_error_desc)
 
-static int mlx5e_grp_pme_get_num_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_NUM_STATS(pme)
 {
 	return NUM_PME_STATUS_STATS + NUM_PME_ERR_STATS;
 }
 
-static int mlx5e_grp_pme_fill_strings(struct mlx5e_priv *priv, u8 *data,
-				      int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STRS(pme)
 {
 	int i;
 
@@ -1424,8 +1404,7 @@ static int mlx5e_grp_pme_fill_strings(struct mlx5e_priv *priv, u8 *data,
 	return idx;
 }
 
-static int mlx5e_grp_pme_fill_stats(struct mlx5e_priv *priv, u64 *data,
-				    int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(pme)
 {
 	struct mlx5_pme_stats pme_stats;
 	int i;
@@ -1443,45 +1422,46 @@ static int mlx5e_grp_pme_fill_stats(struct mlx5e_priv *priv, u64 *data,
 	return idx;
 }
 
-static int mlx5e_grp_ipsec_get_num_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(pme) { return; }
+
+static MLX5E_DECLARE_STATS_GRP_OP_NUM_STATS(ipsec)
 {
 	return mlx5e_ipsec_get_count(priv);
 }
 
-static int mlx5e_grp_ipsec_fill_strings(struct mlx5e_priv *priv, u8 *data,
-					int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STRS(ipsec)
 {
 	return idx + mlx5e_ipsec_get_strings(priv,
 					     data + idx * ETH_GSTRING_LEN);
 }
 
-static int mlx5e_grp_ipsec_fill_stats(struct mlx5e_priv *priv, u64 *data,
-				      int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(ipsec)
 {
 	return idx + mlx5e_ipsec_get_stats(priv, data + idx);
 }
 
-static void mlx5e_grp_ipsec_update_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(ipsec)
 {
 	mlx5e_ipsec_update_stats(priv);
 }
 
-static int mlx5e_grp_tls_get_num_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_NUM_STATS(tls)
 {
 	return mlx5e_tls_get_count(priv);
 }
 
-static int mlx5e_grp_tls_fill_strings(struct mlx5e_priv *priv, u8 *data,
-				      int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STRS(tls)
 {
 	return idx + mlx5e_tls_get_strings(priv, data + idx * ETH_GSTRING_LEN);
 }
 
-static int mlx5e_grp_tls_fill_stats(struct mlx5e_priv *priv, u64 *data, int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(tls)
 {
 	return idx + mlx5e_tls_get_stats(priv, data + idx);
 }
 
+static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(tls) { return; }
+
 static const struct counter_desc rq_stats_desc[] = {
 	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, packets) },
 	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, bytes) },
@@ -1615,7 +1595,7 @@ static const struct counter_desc ch_stats_desc[] = {
 #define NUM_XSKSQ_STATS			ARRAY_SIZE(xsksq_stats_desc)
 #define NUM_CH_STATS			ARRAY_SIZE(ch_stats_desc)
 
-static int mlx5e_grp_channels_get_num_stats(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_NUM_STATS(channels)
 {
 	int max_nch = priv->max_nch;
 
@@ -1628,8 +1608,7 @@ static int mlx5e_grp_channels_get_num_stats(struct mlx5e_priv *priv)
 	       (NUM_XSKSQ_STATS * max_nch * priv->xsk.ever_used);
 }
 
-static int mlx5e_grp_channels_fill_strings(struct mlx5e_priv *priv, u8 *data,
-					   int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STRS(channels)
 {
 	bool is_xsk = priv->xsk.ever_used;
 	int max_nch = priv->max_nch;
@@ -1671,8 +1650,7 @@ static int mlx5e_grp_channels_fill_strings(struct mlx5e_priv *priv, u8 *data,
 	return idx;
 }
 
-static int mlx5e_grp_channels_fill_stats(struct mlx5e_priv *priv, u64 *data,
-					 int idx)
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(channels)
 {
 	bool is_xsk = priv->xsk.ever_used;
 	int max_nch = priv->max_nch;
@@ -1720,104 +1698,26 @@ static int mlx5e_grp_channels_fill_stats(struct mlx5e_priv *priv, u64 *data,
 	return idx;
 }
 
+static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(channels) { return; }
+
 /* The stats groups order is opposite to the update_stats() order calls */
 const struct mlx5e_stats_grp mlx5e_nic_stats_grps[] = {
-	{
-		.get_num_stats = mlx5e_grp_sw_get_num_stats,
-		.fill_strings = mlx5e_grp_sw_fill_strings,
-		.fill_stats = mlx5e_grp_sw_fill_stats,
-		.update_stats = mlx5e_grp_sw_update_stats,
-	},
-	{
-		.get_num_stats = mlx5e_grp_q_get_num_stats,
-		.fill_strings = mlx5e_grp_q_fill_strings,
-		.fill_stats = mlx5e_grp_q_fill_stats,
-		.update_stats_mask = MLX5E_NDO_UPDATE_STATS,
-		.update_stats = mlx5e_grp_q_update_stats,
-	},
-	{
-		.get_num_stats = mlx5e_grp_vnic_env_get_num_stats,
-		.fill_strings = mlx5e_grp_vnic_env_fill_strings,
-		.fill_stats = mlx5e_grp_vnic_env_fill_stats,
-		.update_stats = mlx5e_grp_vnic_env_update_stats,
-	},
-	{
-		.get_num_stats = mlx5e_grp_vport_get_num_stats,
-		.fill_strings = mlx5e_grp_vport_fill_strings,
-		.fill_stats = mlx5e_grp_vport_fill_stats,
-		.update_stats_mask = MLX5E_NDO_UPDATE_STATS,
-		.update_stats = mlx5e_grp_vport_update_stats,
-	},
-	{
-		.get_num_stats = mlx5e_grp_802_3_get_num_stats,
-		.fill_strings = mlx5e_grp_802_3_fill_strings,
-		.fill_stats = mlx5e_grp_802_3_fill_stats,
-		.update_stats_mask = MLX5E_NDO_UPDATE_STATS,
-		.update_stats = mlx5e_grp_802_3_update_stats,
-	},
-	{
-		.get_num_stats = mlx5e_grp_2863_get_num_stats,
-		.fill_strings = mlx5e_grp_2863_fill_strings,
-		.fill_stats = mlx5e_grp_2863_fill_stats,
-		.update_stats = mlx5e_grp_2863_update_stats,
-	},
-	{
-		.get_num_stats = mlx5e_grp_2819_get_num_stats,
-		.fill_strings = mlx5e_grp_2819_fill_strings,
-		.fill_stats = mlx5e_grp_2819_fill_stats,
-		.update_stats = mlx5e_grp_2819_update_stats,
-	},
-	{
-		.get_num_stats = mlx5e_grp_phy_get_num_stats,
-		.fill_strings = mlx5e_grp_phy_fill_strings,
-		.fill_stats = mlx5e_grp_phy_fill_stats,
-		.update_stats = mlx5e_grp_phy_update_stats,
-	},
-	{
-		.get_num_stats = mlx5e_grp_eth_ext_get_num_stats,
-		.fill_strings = mlx5e_grp_eth_ext_fill_strings,
-		.fill_stats = mlx5e_grp_eth_ext_fill_stats,
-		.update_stats = mlx5e_grp_eth_ext_update_stats,
-	},
-	{
-		.get_num_stats = mlx5e_grp_pcie_get_num_stats,
-		.fill_strings = mlx5e_grp_pcie_fill_strings,
-		.fill_stats = mlx5e_grp_pcie_fill_stats,
-		.update_stats = mlx5e_grp_pcie_update_stats,
-	},
-	{
-		.get_num_stats = mlx5e_grp_per_prio_get_num_stats,
-		.fill_strings = mlx5e_grp_per_prio_fill_strings,
-		.fill_stats = mlx5e_grp_per_prio_fill_stats,
-		.update_stats = mlx5e_grp_per_prio_update_stats,
-	},
-	{
-		.get_num_stats = mlx5e_grp_pme_get_num_stats,
-		.fill_strings = mlx5e_grp_pme_fill_strings,
-		.fill_stats = mlx5e_grp_pme_fill_stats,
-	},
-	{
-		.get_num_stats = mlx5e_grp_ipsec_get_num_stats,
-		.fill_strings = mlx5e_grp_ipsec_fill_strings,
-		.fill_stats = mlx5e_grp_ipsec_fill_stats,
-		.update_stats = mlx5e_grp_ipsec_update_stats,
-	},
-	{
-		.get_num_stats = mlx5e_grp_tls_get_num_stats,
-		.fill_strings = mlx5e_grp_tls_fill_strings,
-		.fill_stats = mlx5e_grp_tls_fill_stats,
-	},
-	{
-		.get_num_stats = mlx5e_grp_channels_get_num_stats,
-		.fill_strings = mlx5e_grp_channels_fill_strings,
-		.fill_stats = mlx5e_grp_channels_fill_stats,
-	},
-	{
-		.get_num_stats = mlx5e_grp_per_port_buffer_congest_get_num_stats,
-		.fill_strings = mlx5e_grp_per_port_buffer_congest_fill_strings,
-		.fill_stats = mlx5e_grp_per_port_buffer_congest_fill_stats,
-		.update_stats = mlx5e_grp_per_port_buffer_congest_update_stats,
-	},
+	MLX5E_DEFINE_STATS_GRP(sw, 0),
+	MLX5E_DEFINE_STATS_GRP(qcnt, MLX5E_NDO_UPDATE_STATS),
+	MLX5E_DEFINE_STATS_GRP(vnic_env, 0),
+	MLX5E_DEFINE_STATS_GRP(vport, MLX5E_NDO_UPDATE_STATS),
+	MLX5E_DEFINE_STATS_GRP(802_3, MLX5E_NDO_UPDATE_STATS),
+	MLX5E_DEFINE_STATS_GRP(2863, 0),
+	MLX5E_DEFINE_STATS_GRP(2819, 0),
+	MLX5E_DEFINE_STATS_GRP(phy, 0),
+	MLX5E_DEFINE_STATS_GRP(eth_ext, 0),
+	MLX5E_DEFINE_STATS_GRP(pcie, 0),
+	MLX5E_DEFINE_STATS_GRP(per_prio, 0),
+	MLX5E_DEFINE_STATS_GRP(pme, 0),
+	MLX5E_DEFINE_STATS_GRP(ipsec, 0),
+	MLX5E_DEFINE_STATS_GRP(tls, 0),
+	MLX5E_DEFINE_STATS_GRP(channels, 0),
+	MLX5E_DEFINE_STATS_GRP(per_port_buff_congest, 0),
 };
 
 unsigned int mlx5e_nic_stats_grps_num(struct mlx5e_priv *priv)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
index 06eeedaacb88..bc97964cd721 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
@@ -29,6 +29,7 @@
  * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
+
 #ifndef __MLX5_EN_STATS_H__
 #define __MLX5_EN_STATS_H__
 
@@ -68,6 +69,28 @@ struct mlx5e_stats_grp {
 	void (*update_stats)(struct mlx5e_priv *priv);
 };
 
+#define MLX5E_STATS_GRP_OP(grp, name) mlx5e_stats_grp_ ## grp ## _ ## name
+
+#define MLX5E_DECLARE_STATS_GRP_OP_NUM_STATS(grp) \
+	int MLX5E_STATS_GRP_OP(grp, num_stats)(struct mlx5e_priv *priv)
+
+#define MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(grp) \
+	void MLX5E_STATS_GRP_OP(grp, update_stats)(struct mlx5e_priv *priv)
+
+#define MLX5E_DECLARE_STATS_GRP_OP_FILL_STRS(grp) \
+	int MLX5E_STATS_GRP_OP(grp, fill_strings)(struct mlx5e_priv *priv, u8 *data, int idx)
+
+#define MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(grp) \
+	int MLX5E_STATS_GRP_OP(grp, fill_stats)(struct mlx5e_priv *priv, u64 *data, int idx)
+
+#define MLX5E_DEFINE_STATS_GRP(grp, mask) { \
+	.get_num_stats = MLX5E_STATS_GRP_OP(grp, num_stats), \
+	.fill_stats    = MLX5E_STATS_GRP_OP(grp, fill_stats), \
+	.fill_strings  = MLX5E_STATS_GRP_OP(grp, fill_strings), \
+	.update_stats  = MLX5E_STATS_GRP_OP(grp, update_stats), \
+	.update_stats_mask = mask, \
+}
+
 unsigned int mlx5e_stats_total_num(struct mlx5e_priv *priv);
 void mlx5e_stats_update(struct mlx5e_priv *priv);
 void mlx5e_stats_fill(struct mlx5e_priv *priv, u64 *data, int idx);
@@ -345,6 +368,6 @@ struct mlx5e_stats {
 extern const struct mlx5e_stats_grp mlx5e_nic_stats_grps[];
 unsigned int mlx5e_nic_stats_grps_num(struct mlx5e_priv *priv);
 
-void mlx5e_grp_802_3_update_stats(struct mlx5e_priv *priv);
+MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(802_3);
 
 #endif /* __MLX5_EN_STATS_H__ */
-- 
cgit v1.2.3-59-g8ed1b


From f0ff8e8c7206886a5360eee7d3f664480ed473c9 Mon Sep 17 00:00:00 2001
From: Saeed Mahameed <saeedm@mellanox.com>
Date: Tue, 21 Jan 2020 00:54:00 -0800
Subject: net/mlx5e: Convert stats groups array to array of group pointers

Convert stats groups array to array of "stats group" pointers to allow
sharing and individual selection of groups per profile as illustrated in
the next patches.

Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h       |  2 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c  |  4 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_stats.c | 69 ++++++++++++++--------
 drivers/net/ethernet/mellanox/mlx5/core/en_stats.h | 12 +++-
 4 files changed, 56 insertions(+), 31 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index f85d99d601a6..220ef9f06f84 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -893,7 +893,7 @@ struct mlx5e_profile {
 	void	(*update_stats)(struct mlx5e_priv *priv);
 	void	(*update_carrier)(struct mlx5e_priv *priv);
 	unsigned int (*stats_grps_num)(struct mlx5e_priv *priv);
-	const struct mlx5e_stats_grp *stats_grps;
+	mlx5e_stats_grp_t *stats_grps;
 	struct {
 		mlx5e_fp_handle_rx_cqe handle_rx_cqe;
 		mlx5e_fp_handle_rx_cqe handle_rx_cqe_mpwqe;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index a1bb9eb8e3b5..f3600ae4b0a1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -164,9 +164,9 @@ void mlx5e_update_ndo_stats(struct mlx5e_priv *priv)
 	int i;
 
 	for (i = mlx5e_nic_stats_grps_num(priv) - 1; i >= 0; i--)
-		if (mlx5e_nic_stats_grps[i].update_stats_mask &
+		if (mlx5e_nic_stats_grps[i]->update_stats_mask &
 		    MLX5E_NDO_UPDATE_STATS)
-			mlx5e_nic_stats_grps[i].update_stats(priv);
+			mlx5e_nic_stats_grps[i]->update_stats(priv);
 }
 
 static void mlx5e_update_stats_work(struct work_struct *work)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
index e903a15e7289..85730a8899c1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
@@ -43,46 +43,46 @@ static unsigned int stats_grps_num(struct mlx5e_priv *priv)
 
 unsigned int mlx5e_stats_total_num(struct mlx5e_priv *priv)
 {
-	const struct mlx5e_stats_grp *stats_grps = priv->profile->stats_grps;
+	mlx5e_stats_grp_t *stats_grps = priv->profile->stats_grps;
 	const unsigned int num_stats_grps = stats_grps_num(priv);
 	unsigned int total = 0;
 	int i;
 
 	for (i = 0; i < num_stats_grps; i++)
-		total += stats_grps[i].get_num_stats(priv);
+		total += stats_grps[i]->get_num_stats(priv);
 
 	return total;
 }
 
 void mlx5e_stats_update(struct mlx5e_priv *priv)
 {
-	const struct mlx5e_stats_grp *stats_grps = priv->profile->stats_grps;
+	mlx5e_stats_grp_t *stats_grps = priv->profile->stats_grps;
 	const unsigned int num_stats_grps = stats_grps_num(priv);
 	int i;
 
 	for (i = num_stats_grps - 1; i >= 0; i--)
-		if (stats_grps[i].update_stats)
-			stats_grps[i].update_stats(priv);
+		if (stats_grps[i]->update_stats)
+			stats_grps[i]->update_stats(priv);
 }
 
 void mlx5e_stats_fill(struct mlx5e_priv *priv, u64 *data, int idx)
 {
-	const struct mlx5e_stats_grp *stats_grps = priv->profile->stats_grps;
+	mlx5e_stats_grp_t *stats_grps = priv->profile->stats_grps;
 	const unsigned int num_stats_grps = stats_grps_num(priv);
 	int i;
 
 	for (i = 0; i < num_stats_grps; i++)
-		idx = stats_grps[i].fill_stats(priv, data, idx);
+		idx = stats_grps[i]->fill_stats(priv, data, idx);
 }
 
 void mlx5e_stats_fill_strings(struct mlx5e_priv *priv, u8 *data)
 {
-	const struct mlx5e_stats_grp *stats_grps = priv->profile->stats_grps;
+	mlx5e_stats_grp_t *stats_grps = priv->profile->stats_grps;
 	const unsigned int num_stats_grps = stats_grps_num(priv);
 	int i, idx = 0;
 
 	for (i = 0; i < num_stats_grps; i++)
-		idx = stats_grps[i].fill_strings(priv, data, idx);
+		idx = stats_grps[i]->fill_strings(priv, data, idx);
 }
 
 /* Concrete NIC Stats */
@@ -1700,24 +1700,41 @@ static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(channels)
 
 static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(channels) { return; }
 
+static MLX5E_DEFINE_STATS_GRP(sw, 0);
+static MLX5E_DEFINE_STATS_GRP(qcnt, MLX5E_NDO_UPDATE_STATS);
+static MLX5E_DEFINE_STATS_GRP(vnic_env, 0);
+static MLX5E_DEFINE_STATS_GRP(vport, MLX5E_NDO_UPDATE_STATS);
+static MLX5E_DEFINE_STATS_GRP(802_3, MLX5E_NDO_UPDATE_STATS);
+static MLX5E_DEFINE_STATS_GRP(2863, 0);
+static MLX5E_DEFINE_STATS_GRP(2819, 0);
+static MLX5E_DEFINE_STATS_GRP(phy, 0);
+static MLX5E_DEFINE_STATS_GRP(pcie, 0);
+static MLX5E_DEFINE_STATS_GRP(per_prio, 0);
+static MLX5E_DEFINE_STATS_GRP(pme, 0);
+static MLX5E_DEFINE_STATS_GRP(channels, 0);
+static MLX5E_DEFINE_STATS_GRP(per_port_buff_congest, 0);
+static MLX5E_DEFINE_STATS_GRP(eth_ext, 0);
+static MLX5E_DEFINE_STATS_GRP(ipsec, 0);
+static MLX5E_DEFINE_STATS_GRP(tls, 0);
+
 /* The stats groups order is opposite to the update_stats() order calls */
-const struct mlx5e_stats_grp mlx5e_nic_stats_grps[] = {
-	MLX5E_DEFINE_STATS_GRP(sw, 0),
-	MLX5E_DEFINE_STATS_GRP(qcnt, MLX5E_NDO_UPDATE_STATS),
-	MLX5E_DEFINE_STATS_GRP(vnic_env, 0),
-	MLX5E_DEFINE_STATS_GRP(vport, MLX5E_NDO_UPDATE_STATS),
-	MLX5E_DEFINE_STATS_GRP(802_3, MLX5E_NDO_UPDATE_STATS),
-	MLX5E_DEFINE_STATS_GRP(2863, 0),
-	MLX5E_DEFINE_STATS_GRP(2819, 0),
-	MLX5E_DEFINE_STATS_GRP(phy, 0),
-	MLX5E_DEFINE_STATS_GRP(eth_ext, 0),
-	MLX5E_DEFINE_STATS_GRP(pcie, 0),
-	MLX5E_DEFINE_STATS_GRP(per_prio, 0),
-	MLX5E_DEFINE_STATS_GRP(pme, 0),
-	MLX5E_DEFINE_STATS_GRP(ipsec, 0),
-	MLX5E_DEFINE_STATS_GRP(tls, 0),
-	MLX5E_DEFINE_STATS_GRP(channels, 0),
-	MLX5E_DEFINE_STATS_GRP(per_port_buff_congest, 0),
+mlx5e_stats_grp_t mlx5e_nic_stats_grps[] = {
+	&MLX5E_STATS_GRP(sw),
+	&MLX5E_STATS_GRP(qcnt),
+	&MLX5E_STATS_GRP(vnic_env),
+	&MLX5E_STATS_GRP(vport),
+	&MLX5E_STATS_GRP(802_3),
+	&MLX5E_STATS_GRP(2863),
+	&MLX5E_STATS_GRP(2819),
+	&MLX5E_STATS_GRP(phy),
+	&MLX5E_STATS_GRP(eth_ext),
+	&MLX5E_STATS_GRP(pcie),
+	&MLX5E_STATS_GRP(per_prio),
+	&MLX5E_STATS_GRP(pme),
+	&MLX5E_STATS_GRP(ipsec),
+	&MLX5E_STATS_GRP(tls),
+	&MLX5E_STATS_GRP(channels),
+	&MLX5E_STATS_GRP(per_port_buff_congest),
 };
 
 unsigned int mlx5e_nic_stats_grps_num(struct mlx5e_priv *priv)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
index bc97964cd721..29ad89f66bf7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
@@ -69,6 +69,8 @@ struct mlx5e_stats_grp {
 	void (*update_stats)(struct mlx5e_priv *priv);
 };
 
+typedef const struct mlx5e_stats_grp *const mlx5e_stats_grp_t;
+
 #define MLX5E_STATS_GRP_OP(grp, name) mlx5e_stats_grp_ ## grp ## _ ## name
 
 #define MLX5E_DECLARE_STATS_GRP_OP_NUM_STATS(grp) \
@@ -83,7 +85,13 @@ struct mlx5e_stats_grp {
 #define MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(grp) \
 	int MLX5E_STATS_GRP_OP(grp, fill_stats)(struct mlx5e_priv *priv, u64 *data, int idx)
 
-#define MLX5E_DEFINE_STATS_GRP(grp, mask) { \
+#define MLX5E_STATS_GRP(grp) mlx5e_stats_grp_ ## grp
+
+#define MLX5E_DECLARE_STATS_GRP(grp) \
+	const struct mlx5e_stats_grp MLX5E_STATS_GRP(grp)
+
+#define MLX5E_DEFINE_STATS_GRP(grp, mask) \
+MLX5E_DECLARE_STATS_GRP(grp) = { \
 	.get_num_stats = MLX5E_STATS_GRP_OP(grp, num_stats), \
 	.fill_stats    = MLX5E_STATS_GRP_OP(grp, fill_stats), \
 	.fill_strings  = MLX5E_STATS_GRP_OP(grp, fill_strings), \
@@ -365,7 +373,7 @@ struct mlx5e_stats {
 	struct mlx5e_pcie_stats pcie;
 };
 
-extern const struct mlx5e_stats_grp mlx5e_nic_stats_grps[];
+extern mlx5e_stats_grp_t mlx5e_nic_stats_grps[];
 unsigned int mlx5e_nic_stats_grps_num(struct mlx5e_priv *priv);
 
 MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(802_3);
-- 
cgit v1.2.3-59-g8ed1b


From 2a303f13f9cda5a8f5729ccb84b211c1ef6bed92 Mon Sep 17 00:00:00 2001
From: Saeed Mahameed <saeedm@mellanox.com>
Date: Tue, 21 Jan 2020 01:15:28 -0800
Subject: net/mlx5e: IPoIB, use separate stats groups

Don't copy all of the stats groups used for mlx5e ethernet NIC profile,
have a separate stats groups for IPoIB with the set of the needed stats
only.

Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_stats.c | 26 +++++++++++-----------
 drivers/net/ethernet/mellanox/mlx5/core/en_stats.h | 14 ++++++++++++
 .../net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c  | 26 ++++++++++++++++++++--
 3 files changed, 51 insertions(+), 15 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
index 85730a8899c1..ee5041747575 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
@@ -1700,19 +1700,19 @@ static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(channels)
 
 static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(channels) { return; }
 
-static MLX5E_DEFINE_STATS_GRP(sw, 0);
-static MLX5E_DEFINE_STATS_GRP(qcnt, MLX5E_NDO_UPDATE_STATS);
-static MLX5E_DEFINE_STATS_GRP(vnic_env, 0);
-static MLX5E_DEFINE_STATS_GRP(vport, MLX5E_NDO_UPDATE_STATS);
-static MLX5E_DEFINE_STATS_GRP(802_3, MLX5E_NDO_UPDATE_STATS);
-static MLX5E_DEFINE_STATS_GRP(2863, 0);
-static MLX5E_DEFINE_STATS_GRP(2819, 0);
-static MLX5E_DEFINE_STATS_GRP(phy, 0);
-static MLX5E_DEFINE_STATS_GRP(pcie, 0);
-static MLX5E_DEFINE_STATS_GRP(per_prio, 0);
-static MLX5E_DEFINE_STATS_GRP(pme, 0);
-static MLX5E_DEFINE_STATS_GRP(channels, 0);
-static MLX5E_DEFINE_STATS_GRP(per_port_buff_congest, 0);
+MLX5E_DEFINE_STATS_GRP(sw, 0);
+MLX5E_DEFINE_STATS_GRP(qcnt, MLX5E_NDO_UPDATE_STATS);
+MLX5E_DEFINE_STATS_GRP(vnic_env, 0);
+MLX5E_DEFINE_STATS_GRP(vport, MLX5E_NDO_UPDATE_STATS);
+MLX5E_DEFINE_STATS_GRP(802_3, MLX5E_NDO_UPDATE_STATS);
+MLX5E_DEFINE_STATS_GRP(2863, 0);
+MLX5E_DEFINE_STATS_GRP(2819, 0);
+MLX5E_DEFINE_STATS_GRP(phy, 0);
+MLX5E_DEFINE_STATS_GRP(pcie, 0);
+MLX5E_DEFINE_STATS_GRP(per_prio, 0);
+MLX5E_DEFINE_STATS_GRP(pme, 0);
+MLX5E_DEFINE_STATS_GRP(channels, 0);
+MLX5E_DEFINE_STATS_GRP(per_port_buff_congest, 0);
 static MLX5E_DEFINE_STATS_GRP(eth_ext, 0);
 static MLX5E_DEFINE_STATS_GRP(ipsec, 0);
 static MLX5E_DEFINE_STATS_GRP(tls, 0);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
index 29ad89f66bf7..4dc0b6e083f8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
@@ -378,4 +378,18 @@ unsigned int mlx5e_nic_stats_grps_num(struct mlx5e_priv *priv);
 
 MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(802_3);
 
+extern MLX5E_DECLARE_STATS_GRP(sw);
+extern MLX5E_DECLARE_STATS_GRP(qcnt);
+extern MLX5E_DECLARE_STATS_GRP(vnic_env);
+extern MLX5E_DECLARE_STATS_GRP(vport);
+extern MLX5E_DECLARE_STATS_GRP(802_3);
+extern MLX5E_DECLARE_STATS_GRP(2863);
+extern MLX5E_DECLARE_STATS_GRP(2819);
+extern MLX5E_DECLARE_STATS_GRP(phy);
+extern MLX5E_DECLARE_STATS_GRP(pcie);
+extern MLX5E_DECLARE_STATS_GRP(per_prio);
+extern MLX5E_DECLARE_STATS_GRP(pme);
+extern MLX5E_DECLARE_STATS_GRP(channels);
+extern MLX5E_DECLARE_STATS_GRP(per_port_buff_congest);
+
 #endif /* __MLX5_EN_STATS_H__ */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c
index bd870ffeb1be..56078b23f1a0 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c
@@ -419,6 +419,28 @@ static void mlx5i_cleanup_rx(struct mlx5e_priv *priv)
 	mlx5e_destroy_q_counters(priv);
 }
 
+/* The stats groups order is opposite to the update_stats() order calls */
+static mlx5e_stats_grp_t mlx5i_stats_grps[] = {
+	&MLX5E_STATS_GRP(sw),
+	&MLX5E_STATS_GRP(qcnt),
+	&MLX5E_STATS_GRP(vnic_env),
+	&MLX5E_STATS_GRP(vport),
+	&MLX5E_STATS_GRP(802_3),
+	&MLX5E_STATS_GRP(2863),
+	&MLX5E_STATS_GRP(2819),
+	&MLX5E_STATS_GRP(phy),
+	&MLX5E_STATS_GRP(pcie),
+	&MLX5E_STATS_GRP(per_prio),
+	&MLX5E_STATS_GRP(pme),
+	&MLX5E_STATS_GRP(channels),
+	&MLX5E_STATS_GRP(per_port_buff_congest),
+};
+
+static unsigned int mlx5i_stats_grps_num(struct mlx5e_priv *priv)
+{
+	return ARRAY_SIZE(mlx5i_stats_grps);
+}
+
 static const struct mlx5e_profile mlx5i_nic_profile = {
 	.init		   = mlx5i_init,
 	.cleanup	   = mlx5i_cleanup,
@@ -435,8 +457,8 @@ static const struct mlx5e_profile mlx5i_nic_profile = {
 	.rx_handlers.handle_rx_cqe_mpwqe = NULL, /* Not supported */
 	.max_tc		   = MLX5I_MAX_NUM_TC,
 	.rq_groups	   = MLX5E_NUM_RQ_GROUPS(REGULAR),
-	.stats_grps        = mlx5e_nic_stats_grps,
-	.stats_grps_num    = mlx5e_nic_stats_grps_num,
+	.stats_grps        = mlx5i_stats_grps,
+	.stats_grps_num    = mlx5i_stats_grps_num,
 };
 
 /* mlx5i netdev NDos */
-- 
cgit v1.2.3-59-g8ed1b


From 8a236b15144b29ce47f80c37bc55740a5388ecb1 Mon Sep 17 00:00:00 2001
From: Vlad Buslov <vladbu@mellanox.com>
Date: Tue, 21 Jan 2020 19:38:21 +0200
Subject: net/mlx5e: Convert rep stats to mlx5e_stats_grp-based infra

In order to support all of the supported stats that are available in legacy
mode for switchdev uplink representors, convert rep stats infrastructure to
reuse struct mlx5e_stats_grp that is already used when device is in legacy
mode. Refactor rep code to use array of mlx5e_stats_grp
structures (constructed using macros provided by stats infra) to
fill/update stats, instead of fixed hardcoded set of values. This approach
allows to easily extend representors with new stats types.

Signed-off-by: Vlad Buslov <vladbu@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_rep.c | 157 ++++++++++++++++-------
 1 file changed, 114 insertions(+), 43 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index f00e17f78ec9..01745941a11f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -117,24 +117,71 @@ static const struct counter_desc vport_rep_stats_desc[] = {
 #define NUM_VPORT_REP_SW_COUNTERS ARRAY_SIZE(sw_rep_stats_desc)
 #define NUM_VPORT_REP_HW_COUNTERS ARRAY_SIZE(vport_rep_stats_desc)
 
-static void mlx5e_rep_get_strings(struct net_device *dev,
-				  u32 stringset, uint8_t *data)
+static MLX5E_DECLARE_STATS_GRP_OP_NUM_STATS(sw_rep)
 {
-	int i, j;
+	return NUM_VPORT_REP_SW_COUNTERS;
+}
 
-	switch (stringset) {
-	case ETH_SS_STATS:
-		for (i = 0; i < NUM_VPORT_REP_SW_COUNTERS; i++)
-			strcpy(data + (i * ETH_GSTRING_LEN),
-			       sw_rep_stats_desc[i].format);
-		for (j = 0; j < NUM_VPORT_REP_HW_COUNTERS; j++, i++)
-			strcpy(data + (i * ETH_GSTRING_LEN),
-			       vport_rep_stats_desc[j].format);
-		break;
-	}
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STRS(sw_rep)
+{
+	int i;
+
+	for (i = 0; i < NUM_VPORT_REP_SW_COUNTERS; i++)
+		strcpy(data + (idx++) * ETH_GSTRING_LEN,
+		       sw_rep_stats_desc[i].format);
+	return idx;
+}
+
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(sw_rep)
+{
+	int i;
+
+	for (i = 0; i < NUM_VPORT_REP_SW_COUNTERS; i++)
+		data[idx++] = MLX5E_READ_CTR64_CPU(&priv->stats.sw,
+						   sw_rep_stats_desc, i);
+	return idx;
+}
+
+static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(sw_rep)
+{
+	struct mlx5e_sw_stats *s = &priv->stats.sw;
+	struct rtnl_link_stats64 stats64 = {};
+
+	memset(s, 0, sizeof(*s));
+	mlx5e_fold_sw_stats64(priv, &stats64);
+
+	s->rx_packets = stats64.rx_packets;
+	s->rx_bytes   = stats64.rx_bytes;
+	s->tx_packets = stats64.tx_packets;
+	s->tx_bytes   = stats64.tx_bytes;
+	s->tx_queue_dropped = stats64.tx_dropped;
+}
+
+static MLX5E_DECLARE_STATS_GRP_OP_NUM_STATS(vport_rep)
+{
+	return NUM_VPORT_REP_HW_COUNTERS;
+}
+
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STRS(vport_rep)
+{
+	int i;
+
+	for (i = 0; i < NUM_VPORT_REP_HW_COUNTERS; i++)
+		strcpy(data + (idx++) * ETH_GSTRING_LEN, vport_rep_stats_desc[i].format);
+	return idx;
+}
+
+static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(vport_rep)
+{
+	int i;
+
+	for (i = 0; i < NUM_VPORT_REP_HW_COUNTERS; i++)
+		data[idx++] = MLX5E_READ_CTR64_CPU(&priv->stats.vf_vport,
+						   vport_rep_stats_desc, i);
+	return idx;
 }
 
-static void mlx5e_rep_update_hw_counters(struct mlx5e_priv *priv)
+static void mlx5e_vf_rep_update_hw_counters(struct mlx5e_priv *priv)
 {
 	struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
 	struct mlx5e_rep_priv *rpriv = priv->ppriv;
@@ -172,49 +219,44 @@ static void mlx5e_uplink_rep_update_hw_counters(struct mlx5e_priv *priv)
 	vport_stats->tx_bytes   = PPORT_802_3_GET(pstats, a_octets_transmitted_ok);
 }
 
-static void mlx5e_rep_update_sw_counters(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(vport_rep)
 {
-	struct mlx5e_sw_stats *s = &priv->stats.sw;
-	struct rtnl_link_stats64 stats64 = {};
+	struct mlx5e_rep_priv *rpriv = priv->ppriv;
+	struct mlx5_eswitch_rep *rep = rpriv->rep;
 
-	memset(s, 0, sizeof(*s));
-	mlx5e_fold_sw_stats64(priv, &stats64);
+	if (rep->vport == MLX5_VPORT_UPLINK)
+		mlx5e_uplink_rep_update_hw_counters(priv);
+	else
+		mlx5e_vf_rep_update_hw_counters(priv);
+}
 
-	s->rx_packets = stats64.rx_packets;
-	s->rx_bytes   = stats64.rx_bytes;
-	s->tx_packets = stats64.tx_packets;
-	s->tx_bytes   = stats64.tx_bytes;
-	s->tx_queue_dropped = stats64.tx_dropped;
+static void mlx5e_rep_get_strings(struct net_device *dev,
+				  u32 stringset, uint8_t *data)
+{
+	struct mlx5e_priv *priv = netdev_priv(dev);
+
+	switch (stringset) {
+	case ETH_SS_STATS:
+		mlx5e_stats_fill_strings(priv, data);
+		break;
+	}
 }
 
 static void mlx5e_rep_get_ethtool_stats(struct net_device *dev,
 					struct ethtool_stats *stats, u64 *data)
 {
 	struct mlx5e_priv *priv = netdev_priv(dev);
-	int i, j;
-
-	if (!data)
-		return;
-
-	mutex_lock(&priv->state_lock);
-	mlx5e_rep_update_sw_counters(priv);
-	priv->profile->update_stats(priv);
-	mutex_unlock(&priv->state_lock);
 
-	for (i = 0; i < NUM_VPORT_REP_SW_COUNTERS; i++)
-		data[i] = MLX5E_READ_CTR64_CPU(&priv->stats.sw,
-					       sw_rep_stats_desc, i);
-
-	for (j = 0; j < NUM_VPORT_REP_HW_COUNTERS; j++, i++)
-		data[i] = MLX5E_READ_CTR64_CPU(&priv->stats.vf_vport,
-					       vport_rep_stats_desc, j);
+	mlx5e_ethtool_get_ethtool_stats(priv, stats, data);
 }
 
 static int mlx5e_rep_get_sset_count(struct net_device *dev, int sset)
 {
+	struct mlx5e_priv *priv = netdev_priv(dev);
+
 	switch (sset) {
 	case ETH_SS_STATS:
-		return NUM_VPORT_REP_SW_COUNTERS + NUM_VPORT_REP_HW_COUNTERS;
+		return mlx5e_stats_total_num(priv);
 	default:
 		return -EOPNOTSUPP;
 	}
@@ -1833,6 +1875,31 @@ static void mlx5e_uplink_rep_disable(struct mlx5e_priv *priv)
 	mlx5_lag_remove(mdev);
 }
 
+static MLX5E_DEFINE_STATS_GRP(sw_rep, 0);
+static MLX5E_DEFINE_STATS_GRP(vport_rep, MLX5E_NDO_UPDATE_STATS);
+
+/* The stats groups order is opposite to the update_stats() order calls */
+static mlx5e_stats_grp_t mlx5e_rep_stats_grps[] = {
+	&MLX5E_STATS_GRP(sw_rep),
+	&MLX5E_STATS_GRP(vport_rep),
+};
+
+static unsigned int mlx5e_rep_stats_grps_num(struct mlx5e_priv *priv)
+{
+	return ARRAY_SIZE(mlx5e_rep_stats_grps);
+}
+
+/* The stats groups order is opposite to the update_stats() order calls */
+static mlx5e_stats_grp_t mlx5e_ul_rep_stats_grps[] = {
+	&MLX5E_STATS_GRP(sw_rep),
+	&MLX5E_STATS_GRP(vport_rep),
+};
+
+static unsigned int mlx5e_ul_rep_stats_grps_num(struct mlx5e_priv *priv)
+{
+	return ARRAY_SIZE(mlx5e_ul_rep_stats_grps);
+}
+
 static const struct mlx5e_profile mlx5e_rep_profile = {
 	.init			= mlx5e_init_rep,
 	.cleanup		= mlx5e_cleanup_rep,
@@ -1842,11 +1909,13 @@ static const struct mlx5e_profile mlx5e_rep_profile = {
 	.cleanup_tx		= mlx5e_cleanup_rep_tx,
 	.enable		        = mlx5e_rep_enable,
 	.update_rx		= mlx5e_update_rep_rx,
-	.update_stats           = mlx5e_rep_update_hw_counters,
+	.update_stats           = mlx5e_update_ndo_stats,
 	.rx_handlers.handle_rx_cqe       = mlx5e_handle_rx_cqe_rep,
 	.rx_handlers.handle_rx_cqe_mpwqe = mlx5e_handle_rx_cqe_mpwrq,
 	.max_tc			= 1,
 	.rq_groups		= MLX5E_NUM_RQ_GROUPS(REGULAR),
+	.stats_grps		= mlx5e_rep_stats_grps,
+	.stats_grps_num		= mlx5e_rep_stats_grps_num,
 };
 
 static const struct mlx5e_profile mlx5e_uplink_rep_profile = {
@@ -1859,12 +1928,14 @@ static const struct mlx5e_profile mlx5e_uplink_rep_profile = {
 	.enable		        = mlx5e_uplink_rep_enable,
 	.disable	        = mlx5e_uplink_rep_disable,
 	.update_rx		= mlx5e_update_rep_rx,
-	.update_stats           = mlx5e_uplink_rep_update_hw_counters,
+	.update_stats           = mlx5e_update_ndo_stats,
 	.update_carrier	        = mlx5e_update_carrier,
 	.rx_handlers.handle_rx_cqe       = mlx5e_handle_rx_cqe_rep,
 	.rx_handlers.handle_rx_cqe_mpwqe = mlx5e_handle_rx_cqe_mpwrq,
 	.max_tc			= MLX5E_MAX_NUM_TC,
 	.rq_groups		= MLX5E_NUM_RQ_GROUPS(REGULAR),
+	.stats_grps		= mlx5e_ul_rep_stats_grps,
+	.stats_grps_num		= mlx5e_ul_rep_stats_grps_num,
 };
 
 static bool
-- 
cgit v1.2.3-59-g8ed1b


From 8520fa57a4e934eff631883cf80744d3a84c734b Mon Sep 17 00:00:00 2001
From: Vlad Buslov <vladbu@mellanox.com>
Date: Tue, 21 Jan 2020 21:36:40 +0200
Subject: net/mlx5e: Create q counters on uplink representors

Q counters were disabled for all types of representors to prevent an issue
where there is not enough resources to init q counters for 127 representor
instances. Enable q counters only for uplink representors to support
"rx_out_of_buffer", "rx_if_down_packets" counters in ethtool.

Signed-off-by: Vlad Buslov <vladbu@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_rep.c | 21 +++++++++++++++++++--
 1 file changed, 19 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index 01745941a11f..f11c86d1b9b7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -1716,6 +1716,23 @@ static void mlx5e_cleanup_rep_rx(struct mlx5e_priv *priv)
 	mlx5e_close_drop_rq(&priv->drop_rq);
 }
 
+static int mlx5e_init_ul_rep_rx(struct mlx5e_priv *priv)
+{
+	int err = mlx5e_init_rep_rx(priv);
+
+	if (err)
+		return err;
+
+	mlx5e_create_q_counters(priv);
+	return 0;
+}
+
+static void mlx5e_cleanup_ul_rep_rx(struct mlx5e_priv *priv)
+{
+	mlx5e_destroy_q_counters(priv);
+	mlx5e_cleanup_rep_rx(priv);
+}
+
 static int mlx5e_init_uplink_rep_tx(struct mlx5e_rep_priv *rpriv)
 {
 	struct mlx5_rep_uplink_priv *uplink_priv;
@@ -1921,8 +1938,8 @@ static const struct mlx5e_profile mlx5e_rep_profile = {
 static const struct mlx5e_profile mlx5e_uplink_rep_profile = {
 	.init			= mlx5e_init_rep,
 	.cleanup		= mlx5e_cleanup_rep,
-	.init_rx		= mlx5e_init_rep_rx,
-	.cleanup_rx		= mlx5e_cleanup_rep_rx,
+	.init_rx		= mlx5e_init_ul_rep_rx,
+	.cleanup_rx		= mlx5e_cleanup_ul_rep_rx,
 	.init_tx		= mlx5e_init_rep_tx,
 	.cleanup_tx		= mlx5e_cleanup_rep_tx,
 	.enable		        = mlx5e_uplink_rep_enable,
-- 
cgit v1.2.3-59-g8ed1b


From 7c453526dc50460c63ff28df7673570dd057c5d0 Mon Sep 17 00:00:00 2001
From: Vlad Buslov <vladbu@mellanox.com>
Date: Tue, 21 Jan 2020 20:08:42 +0200
Subject: net/mlx5e: Enable all available stats for uplink reps

Extend stats group array of uplink representor with all stats that are
available for PF in legacy mode, besides ipsec and TLS which are not
supported.

Don't output vport stats for uplink representor because they are already
handled by 802_3 group (with different names: {tx|rx}_{bytes|packets}_phy).

Signed-off-by: Vlad Buslov <vladbu@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_rep.c   | 44 ++++++++--------------
 drivers/net/ethernet/mellanox/mlx5/core/en_stats.c |  4 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_stats.h |  3 +-
 3 files changed, 18 insertions(+), 33 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index f11c86d1b9b7..09061b4c43af 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -181,7 +181,7 @@ static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(vport_rep)
 	return idx;
 }
 
-static void mlx5e_vf_rep_update_hw_counters(struct mlx5e_priv *priv)
+static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(vport_rep)
 {
 	struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
 	struct mlx5e_rep_priv *rpriv = priv->ppriv;
@@ -204,32 +204,6 @@ static void mlx5e_vf_rep_update_hw_counters(struct mlx5e_priv *priv)
 	vport_stats->tx_bytes   = vf_stats.rx_bytes;
 }
 
-static void mlx5e_uplink_rep_update_hw_counters(struct mlx5e_priv *priv)
-{
-	struct mlx5e_pport_stats *pstats = &priv->stats.pport;
-	struct rtnl_link_stats64 *vport_stats;
-
-	MLX5E_STATS_GRP_OP(802_3, update_stats)(priv);
-
-	vport_stats = &priv->stats.vf_vport;
-
-	vport_stats->rx_packets = PPORT_802_3_GET(pstats, a_frames_received_ok);
-	vport_stats->rx_bytes   = PPORT_802_3_GET(pstats, a_octets_received_ok);
-	vport_stats->tx_packets = PPORT_802_3_GET(pstats, a_frames_transmitted_ok);
-	vport_stats->tx_bytes   = PPORT_802_3_GET(pstats, a_octets_transmitted_ok);
-}
-
-static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(vport_rep)
-{
-	struct mlx5e_rep_priv *rpriv = priv->ppriv;
-	struct mlx5_eswitch_rep *rep = rpriv->rep;
-
-	if (rep->vport == MLX5_VPORT_UPLINK)
-		mlx5e_uplink_rep_update_hw_counters(priv);
-	else
-		mlx5e_vf_rep_update_hw_counters(priv);
-}
-
 static void mlx5e_rep_get_strings(struct net_device *dev,
 				  u32 stringset, uint8_t *data)
 {
@@ -1908,8 +1882,20 @@ static unsigned int mlx5e_rep_stats_grps_num(struct mlx5e_priv *priv)
 
 /* The stats groups order is opposite to the update_stats() order calls */
 static mlx5e_stats_grp_t mlx5e_ul_rep_stats_grps[] = {
-	&MLX5E_STATS_GRP(sw_rep),
-	&MLX5E_STATS_GRP(vport_rep),
+	&MLX5E_STATS_GRP(sw),
+	&MLX5E_STATS_GRP(qcnt),
+	&MLX5E_STATS_GRP(vnic_env),
+	&MLX5E_STATS_GRP(vport),
+	&MLX5E_STATS_GRP(802_3),
+	&MLX5E_STATS_GRP(2863),
+	&MLX5E_STATS_GRP(2819),
+	&MLX5E_STATS_GRP(phy),
+	&MLX5E_STATS_GRP(eth_ext),
+	&MLX5E_STATS_GRP(pcie),
+	&MLX5E_STATS_GRP(per_prio),
+	&MLX5E_STATS_GRP(pme),
+	&MLX5E_STATS_GRP(channels),
+	&MLX5E_STATS_GRP(per_port_buff_congest),
 };
 
 static unsigned int mlx5e_ul_rep_stats_grps_num(struct mlx5e_priv *priv)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
index ee5041747575..30b216d9284c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
@@ -630,7 +630,7 @@ static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(802_3)
 #define MLX5_BASIC_PPCNT_SUPPORTED(mdev) \
 	(MLX5_CAP_GEN(mdev, pcam_reg) ? MLX5_CAP_PCAM_REG(mdev, ppcnt) : 1)
 
-MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(802_3)
+static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(802_3)
 {
 	struct mlx5e_pport_stats *pstats = &priv->stats.pport;
 	struct mlx5_core_dev *mdev = priv->mdev;
@@ -1713,7 +1713,7 @@ MLX5E_DEFINE_STATS_GRP(per_prio, 0);
 MLX5E_DEFINE_STATS_GRP(pme, 0);
 MLX5E_DEFINE_STATS_GRP(channels, 0);
 MLX5E_DEFINE_STATS_GRP(per_port_buff_congest, 0);
-static MLX5E_DEFINE_STATS_GRP(eth_ext, 0);
+MLX5E_DEFINE_STATS_GRP(eth_ext, 0);
 static MLX5E_DEFINE_STATS_GRP(ipsec, 0);
 static MLX5E_DEFINE_STATS_GRP(tls, 0);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
index 4dc0b6e083f8..092b39ffa32a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
@@ -376,8 +376,6 @@ struct mlx5e_stats {
 extern mlx5e_stats_grp_t mlx5e_nic_stats_grps[];
 unsigned int mlx5e_nic_stats_grps_num(struct mlx5e_priv *priv);
 
-MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(802_3);
-
 extern MLX5E_DECLARE_STATS_GRP(sw);
 extern MLX5E_DECLARE_STATS_GRP(qcnt);
 extern MLX5E_DECLARE_STATS_GRP(vnic_env);
@@ -386,6 +384,7 @@ extern MLX5E_DECLARE_STATS_GRP(802_3);
 extern MLX5E_DECLARE_STATS_GRP(2863);
 extern MLX5E_DECLARE_STATS_GRP(2819);
 extern MLX5E_DECLARE_STATS_GRP(phy);
+extern MLX5E_DECLARE_STATS_GRP(eth_ext);
 extern MLX5E_DECLARE_STATS_GRP(pcie);
 extern MLX5E_DECLARE_STATS_GRP(per_prio);
 extern MLX5E_DECLARE_STATS_GRP(pme);
-- 
cgit v1.2.3-59-g8ed1b


From fd786fb1d2cad70b9aaba8c73872cbf63262bd58 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Tue, 21 Jan 2020 22:09:33 +0100
Subject: net: convert suitable drivers to use phy_do_ioctl_running

Convert suitable drivers to use new helper phy_do_ioctl_running.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Acked-by: Timur Tabi <timur@kernel.org>
Acked-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/allwinner/sun4i-emac.c           | 15 +--------------
 drivers/net/ethernet/amd/au1000_eth.c                 | 13 +------------
 drivers/net/ethernet/arc/emac_main.c                  | 14 +-------------
 drivers/net/ethernet/broadcom/bgmac.c                 | 10 +---------
 drivers/net/ethernet/broadcom/genet/bcmgenet.c        | 14 +-------------
 drivers/net/ethernet/dnet.c                           | 15 +--------------
 drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c | 10 +---------
 drivers/net/ethernet/hisilicon/hisi_femac.c           | 14 +-------------
 drivers/net/ethernet/hisilicon/hns/hns_enet.c         | 16 +---------------
 drivers/net/ethernet/nxp/lpc_eth.c                    | 15 +--------------
 drivers/net/ethernet/qualcomm/emac/emac.c             | 14 +-------------
 drivers/net/ethernet/renesas/sh_eth.c                 | 18 ++----------------
 drivers/net/ethernet/smsc/smsc911x.c                  | 11 +----------
 drivers/net/ethernet/smsc/smsc9420.c                  | 11 +----------
 drivers/net/ethernet/ti/cpmac.c                       | 12 +-----------
 drivers/net/ethernet/toshiba/tc35815.c                | 12 +-----------
 drivers/net/ethernet/xilinx/ll_temac_main.c           | 13 +------------
 drivers/net/usb/ax88172a.c                            | 13 +------------
 drivers/net/usb/lan78xx.c                             | 10 +---------
 19 files changed, 20 insertions(+), 230 deletions(-)

diff --git a/drivers/net/ethernet/allwinner/sun4i-emac.c b/drivers/net/ethernet/allwinner/sun4i-emac.c
index 5ea806423e4c..22cadfbeedfb 100644
--- a/drivers/net/ethernet/allwinner/sun4i-emac.c
+++ b/drivers/net/ethernet/allwinner/sun4i-emac.c
@@ -207,19 +207,6 @@ static void emac_inblk_32bit(void __iomem *reg, void *data, int count)
 	readsl(reg, data, round_up(count, 4) / 4);
 }
 
-static int emac_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
-{
-	struct phy_device *phydev = dev->phydev;
-
-	if (!netif_running(dev))
-		return -EINVAL;
-
-	if (!phydev)
-		return -ENODEV;
-
-	return phy_mii_ioctl(phydev, rq, cmd);
-}
-
 /* ethtool ops */
 static void emac_get_drvinfo(struct net_device *dev,
 			      struct ethtool_drvinfo *info)
@@ -791,7 +778,7 @@ static const struct net_device_ops emac_netdev_ops = {
 	.ndo_start_xmit		= emac_start_xmit,
 	.ndo_tx_timeout		= emac_timeout,
 	.ndo_set_rx_mode	= emac_set_rx_mode,
-	.ndo_do_ioctl		= emac_ioctl,
+	.ndo_do_ioctl		= phy_do_ioctl_running,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_mac_address	= emac_set_mac_address,
 #ifdef CONFIG_NET_POLL_CONTROLLER
diff --git a/drivers/net/ethernet/amd/au1000_eth.c b/drivers/net/ethernet/amd/au1000_eth.c
index d832c9f4d306..255847e7fa5b 100644
--- a/drivers/net/ethernet/amd/au1000_eth.c
+++ b/drivers/net/ethernet/amd/au1000_eth.c
@@ -1053,23 +1053,12 @@ static void au1000_multicast_list(struct net_device *dev)
 	writel(reg, &aup->mac->control);
 }
 
-static int au1000_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
-{
-	if (!netif_running(dev))
-		return -EINVAL;
-
-	if (!dev->phydev)
-		return -EINVAL; /* PHY not controllable */
-
-	return phy_mii_ioctl(dev->phydev, rq, cmd);
-}
-
 static const struct net_device_ops au1000_netdev_ops = {
 	.ndo_open		= au1000_open,
 	.ndo_stop		= au1000_close,
 	.ndo_start_xmit		= au1000_tx,
 	.ndo_set_rx_mode	= au1000_multicast_list,
-	.ndo_do_ioctl		= au1000_ioctl,
+	.ndo_do_ioctl		= phy_do_ioctl_running,
 	.ndo_tx_timeout		= au1000_tx_timeout,
 	.ndo_set_mac_address	= eth_mac_addr,
 	.ndo_validate_addr	= eth_validate_addr,
diff --git a/drivers/net/ethernet/arc/emac_main.c b/drivers/net/ethernet/arc/emac_main.c
index 6f2c867785fe..17bda4e8cc45 100644
--- a/drivers/net/ethernet/arc/emac_main.c
+++ b/drivers/net/ethernet/arc/emac_main.c
@@ -781,18 +781,6 @@ static int arc_emac_set_address(struct net_device *ndev, void *p)
 	return 0;
 }
 
-static int arc_emac_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
-{
-	if (!netif_running(dev))
-		return -EINVAL;
-
-	if (!dev->phydev)
-		return -ENODEV;
-
-	return phy_mii_ioctl(dev->phydev, rq, cmd);
-}
-
-
 /**
  * arc_emac_restart - Restart EMAC
  * @ndev:	Pointer to net_device structure.
@@ -857,7 +845,7 @@ static const struct net_device_ops arc_emac_netdev_ops = {
 	.ndo_set_mac_address	= arc_emac_set_address,
 	.ndo_get_stats		= arc_emac_stats,
 	.ndo_set_rx_mode	= arc_emac_set_rx_mode,
-	.ndo_do_ioctl		= arc_emac_ioctl,
+	.ndo_do_ioctl		= phy_do_ioctl_running,
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller	= arc_emac_poll_controller,
 #endif
diff --git a/drivers/net/ethernet/broadcom/bgmac.c b/drivers/net/ethernet/broadcom/bgmac.c
index 148734b166f0..1bb07a5d82c9 100644
--- a/drivers/net/ethernet/broadcom/bgmac.c
+++ b/drivers/net/ethernet/broadcom/bgmac.c
@@ -1248,14 +1248,6 @@ static int bgmac_set_mac_address(struct net_device *net_dev, void *addr)
 	return 0;
 }
 
-static int bgmac_ioctl(struct net_device *net_dev, struct ifreq *ifr, int cmd)
-{
-	if (!netif_running(net_dev))
-		return -EINVAL;
-
-	return phy_mii_ioctl(net_dev->phydev, ifr, cmd);
-}
-
 static const struct net_device_ops bgmac_netdev_ops = {
 	.ndo_open		= bgmac_open,
 	.ndo_stop		= bgmac_stop,
@@ -1263,7 +1255,7 @@ static const struct net_device_ops bgmac_netdev_ops = {
 	.ndo_set_rx_mode	= bgmac_set_rx_mode,
 	.ndo_set_mac_address	= bgmac_set_mac_address,
 	.ndo_validate_addr	= eth_validate_addr,
-	.ndo_do_ioctl           = bgmac_ioctl,
+	.ndo_do_ioctl           = phy_do_ioctl_running,
 };
 
 /**************************************************
diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
index 3ee7917e3fc0..013959285a61 100644
--- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c
+++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
@@ -1225,18 +1225,6 @@ static void bcmgenet_power_up(struct bcmgenet_priv *priv,
 	}
 }
 
-/* ioctl handle special commands that are not present in ethtool. */
-static int bcmgenet_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
-{
-	if (!netif_running(dev))
-		return -EINVAL;
-
-	if (!dev->phydev)
-		return -ENODEV;
-
-	return phy_mii_ioctl(dev->phydev, rq, cmd);
-}
-
 static struct enet_cb *bcmgenet_get_txcb(struct bcmgenet_priv *priv,
 					 struct bcmgenet_tx_ring *ring)
 {
@@ -3222,7 +3210,7 @@ static const struct net_device_ops bcmgenet_netdev_ops = {
 	.ndo_tx_timeout		= bcmgenet_timeout,
 	.ndo_set_rx_mode	= bcmgenet_set_rx_mode,
 	.ndo_set_mac_address	= bcmgenet_set_mac_addr,
-	.ndo_do_ioctl		= bcmgenet_ioctl,
+	.ndo_do_ioctl		= phy_do_ioctl_running,
 	.ndo_set_features	= bcmgenet_set_features,
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller	= bcmgenet_poll_controller,
diff --git a/drivers/net/ethernet/dnet.c b/drivers/net/ethernet/dnet.c
index e24979010969..5f8fa1145db6 100644
--- a/drivers/net/ethernet/dnet.c
+++ b/drivers/net/ethernet/dnet.c
@@ -725,19 +725,6 @@ static struct net_device_stats *dnet_get_stats(struct net_device *dev)
 	return nstat;
 }
 
-static int dnet_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
-{
-	struct phy_device *phydev = dev->phydev;
-
-	if (!netif_running(dev))
-		return -EINVAL;
-
-	if (!phydev)
-		return -ENODEV;
-
-	return phy_mii_ioctl(phydev, rq, cmd);
-}
-
 static void dnet_get_drvinfo(struct net_device *dev,
 			     struct ethtool_drvinfo *info)
 {
@@ -759,7 +746,7 @@ static const struct net_device_ops dnet_netdev_ops = {
 	.ndo_stop		= dnet_close,
 	.ndo_get_stats		= dnet_get_stats,
 	.ndo_start_xmit		= dnet_start_xmit,
-	.ndo_do_ioctl		= dnet_ioctl,
+	.ndo_do_ioctl		= phy_do_ioctl_running,
 	.ndo_set_mac_address	= eth_mac_addr,
 	.ndo_validate_addr	= eth_validate_addr,
 };
diff --git a/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c b/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c
index 80903cd58468..add61fed33ee 100644
--- a/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c
+++ b/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c
@@ -882,14 +882,6 @@ static const struct ethtool_ops fs_ethtool_ops = {
 	.set_tunable = fs_set_tunable,
 };
 
-static int fs_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
-{
-	if (!netif_running(dev))
-		return -EINVAL;
-
-	return phy_mii_ioctl(dev->phydev, rq, cmd);
-}
-
 extern int fs_mii_connect(struct net_device *dev);
 extern void fs_mii_disconnect(struct net_device *dev);
 
@@ -907,7 +899,7 @@ static const struct net_device_ops fs_enet_netdev_ops = {
 	.ndo_start_xmit		= fs_enet_start_xmit,
 	.ndo_tx_timeout		= fs_timeout,
 	.ndo_set_rx_mode	= fs_set_multicast_list,
-	.ndo_do_ioctl		= fs_ioctl,
+	.ndo_do_ioctl		= phy_do_ioctl_running,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_mac_address	= eth_mac_addr,
 #ifdef CONFIG_NET_POLL_CONTROLLER
diff --git a/drivers/net/ethernet/hisilicon/hisi_femac.c b/drivers/net/ethernet/hisilicon/hisi_femac.c
index 90ab7ade44c4..57c3bc4f7089 100644
--- a/drivers/net/ethernet/hisilicon/hisi_femac.c
+++ b/drivers/net/ethernet/hisilicon/hisi_femac.c
@@ -675,18 +675,6 @@ static void hisi_femac_net_set_rx_mode(struct net_device *dev)
 	}
 }
 
-static int hisi_femac_net_ioctl(struct net_device *dev,
-				struct ifreq *ifreq, int cmd)
-{
-	if (!netif_running(dev))
-		return -EINVAL;
-
-	if (!dev->phydev)
-		return -EINVAL;
-
-	return phy_mii_ioctl(dev->phydev, ifreq, cmd);
-}
-
 static const struct ethtool_ops hisi_femac_ethtools_ops = {
 	.get_link		= ethtool_op_get_link,
 	.get_link_ksettings	= phy_ethtool_get_link_ksettings,
@@ -697,7 +685,7 @@ static const struct net_device_ops hisi_femac_netdev_ops = {
 	.ndo_open		= hisi_femac_net_open,
 	.ndo_stop		= hisi_femac_net_close,
 	.ndo_start_xmit		= hisi_femac_net_xmit,
-	.ndo_do_ioctl		= hisi_femac_net_ioctl,
+	.ndo_do_ioctl		= phy_do_ioctl_running,
 	.ndo_set_mac_address	= hisi_femac_set_mac_address,
 	.ndo_set_rx_mode	= hisi_femac_net_set_rx_mode,
 };
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_enet.c b/drivers/net/ethernet/hisilicon/hns/hns_enet.c
index 2fbb476584bc..c117074c16e3 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_enet.c
+++ b/drivers/net/ethernet/hisilicon/hns/hns_enet.c
@@ -1497,20 +1497,6 @@ static void hns_nic_net_timeout(struct net_device *ndev, unsigned int txqueue)
 	}
 }
 
-static int hns_nic_do_ioctl(struct net_device *netdev, struct ifreq *ifr,
-			    int cmd)
-{
-	struct phy_device *phy_dev = netdev->phydev;
-
-	if (!netif_running(netdev))
-		return -EINVAL;
-
-	if (!phy_dev)
-		return -ENOTSUPP;
-
-	return phy_mii_ioctl(phy_dev, ifr, cmd);
-}
-
 static netdev_tx_t hns_nic_net_xmit(struct sk_buff *skb,
 				    struct net_device *ndev)
 {
@@ -1958,7 +1944,7 @@ static const struct net_device_ops hns_nic_netdev_ops = {
 	.ndo_tx_timeout = hns_nic_net_timeout,
 	.ndo_set_mac_address = hns_nic_net_set_mac_address,
 	.ndo_change_mtu = hns_nic_change_mtu,
-	.ndo_do_ioctl = hns_nic_do_ioctl,
+	.ndo_do_ioctl = phy_do_ioctl_running,
 	.ndo_set_features = hns_nic_set_features,
 	.ndo_fix_features = hns_nic_fix_features,
 	.ndo_get_stats64 = hns_nic_get_stats64,
diff --git a/drivers/net/ethernet/nxp/lpc_eth.c b/drivers/net/ethernet/nxp/lpc_eth.c
index 656169214cdb..d20cf03a3ea0 100644
--- a/drivers/net/ethernet/nxp/lpc_eth.c
+++ b/drivers/net/ethernet/nxp/lpc_eth.c
@@ -1149,19 +1149,6 @@ static void lpc_eth_set_multicast_list(struct net_device *ndev)
 	spin_unlock_irqrestore(&pldat->lock, flags);
 }
 
-static int lpc_eth_ioctl(struct net_device *ndev, struct ifreq *req, int cmd)
-{
-	struct phy_device *phydev = ndev->phydev;
-
-	if (!netif_running(ndev))
-		return -EINVAL;
-
-	if (!phydev)
-		return -ENODEV;
-
-	return phy_mii_ioctl(phydev, req, cmd);
-}
-
 static int lpc_eth_open(struct net_device *ndev)
 {
 	struct netdata_local *pldat = netdev_priv(ndev);
@@ -1229,7 +1216,7 @@ static const struct net_device_ops lpc_netdev_ops = {
 	.ndo_stop		= lpc_eth_close,
 	.ndo_start_xmit		= lpc_eth_hard_start_xmit,
 	.ndo_set_rx_mode	= lpc_eth_set_multicast_list,
-	.ndo_do_ioctl		= lpc_eth_ioctl,
+	.ndo_do_ioctl		= phy_do_ioctl_running,
 	.ndo_set_mac_address	= lpc_set_mac_address,
 	.ndo_validate_addr	= eth_validate_addr,
 };
diff --git a/drivers/net/ethernet/qualcomm/emac/emac.c b/drivers/net/ethernet/qualcomm/emac/emac.c
index 522fad4cb2cd..18b0c7a2d6dc 100644
--- a/drivers/net/ethernet/qualcomm/emac/emac.c
+++ b/drivers/net/ethernet/qualcomm/emac/emac.c
@@ -289,18 +289,6 @@ static void emac_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 	schedule_work(&adpt->work_thread);
 }
 
-/* IOCTL support for the interface */
-static int emac_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
-{
-	if (!netif_running(netdev))
-		return -EINVAL;
-
-	if (!netdev->phydev)
-		return -ENODEV;
-
-	return phy_mii_ioctl(netdev->phydev, ifr, cmd);
-}
-
 /**
  * emac_update_hw_stats - read the EMAC stat registers
  *
@@ -387,7 +375,7 @@ static const struct net_device_ops emac_netdev_ops = {
 	.ndo_start_xmit		= emac_start_xmit,
 	.ndo_set_mac_address	= eth_mac_addr,
 	.ndo_change_mtu		= emac_change_mtu,
-	.ndo_do_ioctl		= emac_ioctl,
+	.ndo_do_ioctl		= phy_do_ioctl_running,
 	.ndo_tx_timeout		= emac_tx_timeout,
 	.ndo_get_stats64	= emac_get_stats64,
 	.ndo_set_features       = emac_set_features,
diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c
index c922d7a553f2..58ca126518a2 100644
--- a/drivers/net/ethernet/renesas/sh_eth.c
+++ b/drivers/net/ethernet/renesas/sh_eth.c
@@ -2647,20 +2647,6 @@ static int sh_eth_close(struct net_device *ndev)
 	return 0;
 }
 
-/* ioctl to device function */
-static int sh_eth_do_ioctl(struct net_device *ndev, struct ifreq *rq, int cmd)
-{
-	struct phy_device *phydev = ndev->phydev;
-
-	if (!netif_running(ndev))
-		return -EINVAL;
-
-	if (!phydev)
-		return -ENODEV;
-
-	return phy_mii_ioctl(phydev, rq, cmd);
-}
-
 static int sh_eth_change_mtu(struct net_device *ndev, int new_mtu)
 {
 	if (netif_running(ndev))
@@ -3159,7 +3145,7 @@ static const struct net_device_ops sh_eth_netdev_ops = {
 	.ndo_get_stats		= sh_eth_get_stats,
 	.ndo_set_rx_mode	= sh_eth_set_rx_mode,
 	.ndo_tx_timeout		= sh_eth_tx_timeout,
-	.ndo_do_ioctl		= sh_eth_do_ioctl,
+	.ndo_do_ioctl		= phy_do_ioctl_running,
 	.ndo_change_mtu		= sh_eth_change_mtu,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_mac_address	= eth_mac_addr,
@@ -3175,7 +3161,7 @@ static const struct net_device_ops sh_eth_netdev_ops_tsu = {
 	.ndo_vlan_rx_add_vid	= sh_eth_vlan_rx_add_vid,
 	.ndo_vlan_rx_kill_vid	= sh_eth_vlan_rx_kill_vid,
 	.ndo_tx_timeout		= sh_eth_tx_timeout,
-	.ndo_do_ioctl		= sh_eth_do_ioctl,
+	.ndo_do_ioctl		= phy_do_ioctl_running,
 	.ndo_change_mtu		= sh_eth_change_mtu,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_mac_address	= eth_mac_addr,
diff --git a/drivers/net/ethernet/smsc/smsc911x.c b/drivers/net/ethernet/smsc/smsc911x.c
index 38068fc34141..7dc2154ccdad 100644
--- a/drivers/net/ethernet/smsc/smsc911x.c
+++ b/drivers/net/ethernet/smsc/smsc911x.c
@@ -1943,15 +1943,6 @@ static int smsc911x_set_mac_address(struct net_device *dev, void *p)
 	return 0;
 }
 
-/* Standard ioctls for mii-tool */
-static int smsc911x_do_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
-{
-	if (!netif_running(dev) || !dev->phydev)
-		return -EINVAL;
-
-	return phy_mii_ioctl(dev->phydev, ifr, cmd);
-}
-
 static void smsc911x_ethtool_getdrvinfo(struct net_device *dev,
 					struct ethtool_drvinfo *info)
 {
@@ -2151,7 +2142,7 @@ static const struct net_device_ops smsc911x_netdev_ops = {
 	.ndo_start_xmit		= smsc911x_hard_start_xmit,
 	.ndo_get_stats		= smsc911x_get_stats,
 	.ndo_set_rx_mode	= smsc911x_set_multicast_list,
-	.ndo_do_ioctl		= smsc911x_do_ioctl,
+	.ndo_do_ioctl		= phy_do_ioctl_running,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_mac_address 	= smsc911x_set_mac_address,
 #ifdef CONFIG_NET_POLL_CONTROLLER
diff --git a/drivers/net/ethernet/smsc/smsc9420.c b/drivers/net/ethernet/smsc/smsc9420.c
index a6962a41c3d2..7312e522c022 100644
--- a/drivers/net/ethernet/smsc/smsc9420.c
+++ b/drivers/net/ethernet/smsc/smsc9420.c
@@ -210,15 +210,6 @@ static int smsc9420_eeprom_reload(struct smsc9420_pdata *pd)
 	return -EIO;
 }
 
-/* Standard ioctls for mii-tool */
-static int smsc9420_do_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
-{
-	if (!netif_running(dev) || !dev->phydev)
-		return -EINVAL;
-
-	return phy_mii_ioctl(dev->phydev, ifr, cmd);
-}
-
 static void smsc9420_ethtool_get_drvinfo(struct net_device *netdev,
 					 struct ethtool_drvinfo *drvinfo)
 {
@@ -1504,7 +1495,7 @@ static const struct net_device_ops smsc9420_netdev_ops = {
 	.ndo_start_xmit		= smsc9420_hard_start_xmit,
 	.ndo_get_stats		= smsc9420_get_stats,
 	.ndo_set_rx_mode	= smsc9420_set_multicast_list,
-	.ndo_do_ioctl		= smsc9420_do_ioctl,
+	.ndo_do_ioctl		= phy_do_ioctl_running,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_mac_address 	= eth_mac_addr,
 #ifdef CONFIG_NET_POLL_CONTROLLER
diff --git a/drivers/net/ethernet/ti/cpmac.c b/drivers/net/ethernet/ti/cpmac.c
index 5e1b8292cd3f..a530afe3ce12 100644
--- a/drivers/net/ethernet/ti/cpmac.c
+++ b/drivers/net/ethernet/ti/cpmac.c
@@ -816,16 +816,6 @@ static void cpmac_tx_timeout(struct net_device *dev, unsigned int txqueue)
 	netif_tx_wake_all_queues(priv->dev);
 }
 
-static int cpmac_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
-{
-	if (!(netif_running(dev)))
-		return -EINVAL;
-	if (!dev->phydev)
-		return -EINVAL;
-
-	return phy_mii_ioctl(dev->phydev, ifr, cmd);
-}
-
 static void cpmac_get_ringparam(struct net_device *dev,
 						struct ethtool_ringparam *ring)
 {
@@ -1054,7 +1044,7 @@ static const struct net_device_ops cpmac_netdev_ops = {
 	.ndo_start_xmit		= cpmac_start_xmit,
 	.ndo_tx_timeout		= cpmac_tx_timeout,
 	.ndo_set_rx_mode	= cpmac_set_multicast_list,
-	.ndo_do_ioctl		= cpmac_ioctl,
+	.ndo_do_ioctl		= phy_do_ioctl_running,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_mac_address	= eth_mac_addr,
 };
diff --git a/drivers/net/ethernet/toshiba/tc35815.c b/drivers/net/ethernet/toshiba/tc35815.c
index 708de826200e..3fd43d30b20d 100644
--- a/drivers/net/ethernet/toshiba/tc35815.c
+++ b/drivers/net/ethernet/toshiba/tc35815.c
@@ -484,7 +484,6 @@ static int	tc35815_close(struct net_device *dev);
 static struct	net_device_stats *tc35815_get_stats(struct net_device *dev);
 static void	tc35815_set_multicast_list(struct net_device *dev);
 static void	tc35815_tx_timeout(struct net_device *dev, unsigned int txqueue);
-static int	tc35815_ioctl(struct net_device *dev, struct ifreq *rq, int cmd);
 #ifdef CONFIG_NET_POLL_CONTROLLER
 static void	tc35815_poll_controller(struct net_device *dev);
 #endif
@@ -751,7 +750,7 @@ static const struct net_device_ops tc35815_netdev_ops = {
 	.ndo_get_stats		= tc35815_get_stats,
 	.ndo_set_rx_mode	= tc35815_set_multicast_list,
 	.ndo_tx_timeout		= tc35815_tx_timeout,
-	.ndo_do_ioctl		= tc35815_ioctl,
+	.ndo_do_ioctl		= phy_do_ioctl_running,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_mac_address	= eth_mac_addr,
 #ifdef CONFIG_NET_POLL_CONTROLLER
@@ -2009,15 +2008,6 @@ static const struct ethtool_ops tc35815_ethtool_ops = {
 	.set_link_ksettings = phy_ethtool_set_link_ksettings,
 };
 
-static int tc35815_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
-{
-	if (!netif_running(dev))
-		return -EINVAL;
-	if (!dev->phydev)
-		return -ENODEV;
-	return phy_mii_ioctl(dev->phydev, rq, cmd);
-}
-
 static void tc35815_chip_reset(struct net_device *dev)
 {
 	struct tc35815_regs __iomem *tr =
diff --git a/drivers/net/ethernet/xilinx/ll_temac_main.c b/drivers/net/ethernet/xilinx/ll_temac_main.c
index 21c1b4322ea7..f5bbb19d62d0 100644
--- a/drivers/net/ethernet/xilinx/ll_temac_main.c
+++ b/drivers/net/ethernet/xilinx/ll_temac_main.c
@@ -1080,17 +1080,6 @@ temac_poll_controller(struct net_device *ndev)
 }
 #endif
 
-static int temac_ioctl(struct net_device *ndev, struct ifreq *rq, int cmd)
-{
-	if (!netif_running(ndev))
-		return -EINVAL;
-
-	if (!ndev->phydev)
-		return -EINVAL;
-
-	return phy_mii_ioctl(ndev->phydev, rq, cmd);
-}
-
 static const struct net_device_ops temac_netdev_ops = {
 	.ndo_open = temac_open,
 	.ndo_stop = temac_stop,
@@ -1098,7 +1087,7 @@ static const struct net_device_ops temac_netdev_ops = {
 	.ndo_set_rx_mode = temac_set_multicast_list,
 	.ndo_set_mac_address = temac_set_mac_address,
 	.ndo_validate_addr = eth_validate_addr,
-	.ndo_do_ioctl = temac_ioctl,
+	.ndo_do_ioctl = phy_do_ioctl_running,
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller = temac_poll_controller,
 #endif
diff --git a/drivers/net/usb/ax88172a.c b/drivers/net/usb/ax88172a.c
index af3994e0853b..4e514f5d7c6c 100644
--- a/drivers/net/usb/ax88172a.c
+++ b/drivers/net/usb/ax88172a.c
@@ -39,17 +39,6 @@ static int asix_mdio_bus_write(struct mii_bus *bus, int phy_id, int regnum,
 	return 0;
 }
 
-static int ax88172a_ioctl(struct net_device *net, struct ifreq *rq, int cmd)
-{
-	if (!netif_running(net))
-		return -EINVAL;
-
-	if (!net->phydev)
-		return -ENODEV;
-
-	return phy_mii_ioctl(net->phydev, rq, cmd);
-}
-
 /* set MAC link settings according to information from phylib */
 static void ax88172a_adjust_link(struct net_device *netdev)
 {
@@ -134,7 +123,7 @@ static const struct net_device_ops ax88172a_netdev_ops = {
 	.ndo_get_stats64	= usbnet_get_stats64,
 	.ndo_set_mac_address	= asix_set_mac_address,
 	.ndo_validate_addr	= eth_validate_addr,
-	.ndo_do_ioctl		= ax88172a_ioctl,
+	.ndo_do_ioctl		= phy_do_ioctl_running,
 	.ndo_set_rx_mode        = asix_set_multicast,
 };
 
diff --git a/drivers/net/usb/lan78xx.c b/drivers/net/usb/lan78xx.c
index c391f2521ba1..d2d61f082710 100644
--- a/drivers/net/usb/lan78xx.c
+++ b/drivers/net/usb/lan78xx.c
@@ -1663,14 +1663,6 @@ static const struct ethtool_ops lan78xx_ethtool_ops = {
 	.get_regs	= lan78xx_get_regs,
 };
 
-static int lan78xx_ioctl(struct net_device *netdev, struct ifreq *rq, int cmd)
-{
-	if (!netif_running(netdev))
-		return -EINVAL;
-
-	return phy_mii_ioctl(netdev->phydev, rq, cmd);
-}
-
 static void lan78xx_init_mac_address(struct lan78xx_net *dev)
 {
 	u32 addr_lo, addr_hi;
@@ -3676,7 +3668,7 @@ static const struct net_device_ops lan78xx_netdev_ops = {
 	.ndo_change_mtu		= lan78xx_change_mtu,
 	.ndo_set_mac_address	= lan78xx_set_mac_addr,
 	.ndo_validate_addr	= eth_validate_addr,
-	.ndo_do_ioctl		= lan78xx_ioctl,
+	.ndo_do_ioctl		= phy_do_ioctl_running,
 	.ndo_set_rx_mode	= lan78xx_set_multicast,
 	.ndo_set_features	= lan78xx_set_features,
 	.ndo_vlan_rx_add_vid	= lan78xx_vlan_rx_add_vid,
-- 
cgit v1.2.3-59-g8ed1b


From 84bf557fb02f5924c109a21a160ffc353d878487 Mon Sep 17 00:00:00 2001
From: "Mohit P. Tahiliani" <tahiliani@nitk.edu.in>
Date: Wed, 22 Jan 2020 23:52:24 +0530
Subject: net: sched: pie: move common code to pie.h

This patch moves macros, structures and small functions common
to PIE and FQ-PIE (to be added in a future commit) from the file
net/sched/sch_pie.c to the header file include/net/pie.h.
All the moved functions are made inline.

Signed-off-by: Mohit P. Tahiliani <tahiliani@nitk.edu.in>
Signed-off-by: Leslie Monis <lesliemonis@gmail.com>
Signed-off-by: Gautam Ramakrishnan <gautamramk@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/pie.h   | 96 +++++++++++++++++++++++++++++++++++++++++++++++++++++
 net/sched/sch_pie.c | 86 +----------------------------------------------
 2 files changed, 97 insertions(+), 85 deletions(-)
 create mode 100644 include/net/pie.h

diff --git a/include/net/pie.h b/include/net/pie.h
new file mode 100644
index 000000000000..440213ec83eb
--- /dev/null
+++ b/include/net/pie.h
@@ -0,0 +1,96 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+#ifndef __NET_SCHED_PIE_H
+#define __NET_SCHED_PIE_H
+
+#include <linux/ktime.h>
+#include <linux/skbuff.h>
+#include <linux/types.h>
+#include <net/inet_ecn.h>
+#include <net/pkt_sched.h>
+
+#define QUEUE_THRESHOLD 16384
+#define DQCOUNT_INVALID -1
+#define DTIME_INVALID 0xffffffffffffffff
+#define MAX_PROB 0xffffffffffffffff
+#define PIE_SCALE 8
+
+/* parameters used */
+struct pie_params {
+	psched_time_t target;	/* user specified target delay in pschedtime */
+	u32 tupdate;		/* timer frequency (in jiffies) */
+	u32 limit;		/* number of packets that can be enqueued */
+	u32 alpha;		/* alpha and beta are between 0 and 32 */
+	u32 beta;		/* and are used for shift relative to 1 */
+	bool ecn;		/* true if ecn is enabled */
+	bool bytemode;		/* to scale drop early prob based on pkt size */
+	u8 dq_rate_estimator;	/* to calculate delay using Little's law */
+};
+
+/* variables used */
+struct pie_vars {
+	u64 prob;		/* probability but scaled by u64 limit. */
+	psched_time_t burst_time;
+	psched_time_t qdelay;
+	psched_time_t qdelay_old;
+	u64 dq_count;		/* measured in bytes */
+	psched_time_t dq_tstamp;	/* drain rate */
+	u64 accu_prob;		/* accumulated drop probability */
+	u32 avg_dq_rate;	/* bytes per pschedtime tick,scaled */
+	u32 qlen_old;		/* in bytes */
+	u8 accu_prob_overflows;	/* overflows of accu_prob */
+};
+
+/* statistics gathering */
+struct pie_stats {
+	u32 packets_in;		/* total number of packets enqueued */
+	u32 dropped;		/* packets dropped due to pie_action */
+	u32 overlimit;		/* dropped due to lack of space in queue */
+	u32 maxq;		/* maximum queue size */
+	u32 ecn_mark;		/* packets marked with ECN */
+};
+
+/* private skb vars */
+struct pie_skb_cb {
+	psched_time_t enqueue_time;
+};
+
+static inline void pie_params_init(struct pie_params *params)
+{
+	params->alpha = 2;
+	params->beta = 20;
+	params->tupdate = usecs_to_jiffies(15 * USEC_PER_MSEC);	/* 15 ms */
+	params->limit = 1000;	/* default of 1000 packets */
+	params->target = PSCHED_NS2TICKS(15 * NSEC_PER_MSEC);	/* 15 ms */
+	params->ecn = false;
+	params->bytemode = false;
+	params->dq_rate_estimator = false;
+}
+
+static inline void pie_vars_init(struct pie_vars *vars)
+{
+	vars->dq_count = DQCOUNT_INVALID;
+	vars->dq_tstamp = DTIME_INVALID;
+	vars->accu_prob = 0;
+	vars->avg_dq_rate = 0;
+	/* default of 150 ms in pschedtime */
+	vars->burst_time = PSCHED_NS2TICKS(150 * NSEC_PER_MSEC);
+	vars->accu_prob_overflows = 0;
+}
+
+static inline struct pie_skb_cb *get_pie_cb(const struct sk_buff *skb)
+{
+	qdisc_cb_private_validate(skb, sizeof(struct pie_skb_cb));
+	return (struct pie_skb_cb *)qdisc_skb_cb(skb)->data;
+}
+
+static inline psched_time_t pie_get_enqueue_time(const struct sk_buff *skb)
+{
+	return get_pie_cb(skb)->enqueue_time;
+}
+
+static inline void pie_set_enqueue_time(struct sk_buff *skb)
+{
+	get_pie_cb(skb)->enqueue_time = psched_get_time();
+}
+
+#endif
diff --git a/net/sched/sch_pie.c b/net/sched/sch_pie.c
index b0b0dc46af61..7197bcaa14ba 100644
--- a/net/sched/sch_pie.c
+++ b/net/sched/sch_pie.c
@@ -19,47 +19,7 @@
 #include <linux/skbuff.h>
 #include <net/pkt_sched.h>
 #include <net/inet_ecn.h>
-
-#define QUEUE_THRESHOLD 16384
-#define DQCOUNT_INVALID -1
-#define DTIME_INVALID 0xffffffffffffffff
-#define MAX_PROB 0xffffffffffffffff
-#define PIE_SCALE 8
-
-/* parameters used */
-struct pie_params {
-	psched_time_t target;	/* user specified target delay in pschedtime */
-	u32 tupdate;		/* timer frequency (in jiffies) */
-	u32 limit;		/* number of packets that can be enqueued */
-	u32 alpha;		/* alpha and beta are between 0 and 32 */
-	u32 beta;		/* and are used for shift relative to 1 */
-	bool ecn;		/* true if ecn is enabled */
-	bool bytemode;		/* to scale drop early prob based on pkt size */
-	u8 dq_rate_estimator;	/* to calculate delay using Little's law */
-};
-
-/* variables used */
-struct pie_vars {
-	u64 prob;		/* probability but scaled by u64 limit. */
-	psched_time_t burst_time;
-	psched_time_t qdelay;
-	psched_time_t qdelay_old;
-	u64 dq_count;		/* measured in bytes */
-	psched_time_t dq_tstamp;	/* drain rate */
-	u64 accu_prob;		/* accumulated drop probability */
-	u32 avg_dq_rate;	/* bytes per pschedtime tick,scaled */
-	u32 qlen_old;		/* in bytes */
-	u8 accu_prob_overflows;	/* overflows of accu_prob */
-};
-
-/* statistics gathering */
-struct pie_stats {
-	u32 packets_in;		/* total number of packets enqueued */
-	u32 dropped;		/* packets dropped due to pie_action */
-	u32 overlimit;		/* dropped due to lack of space in queue */
-	u32 maxq;		/* maximum queue size */
-	u32 ecn_mark;		/* packets marked with ECN */
-};
+#include <net/pie.h>
 
 /* private data for the Qdisc */
 struct pie_sched_data {
@@ -70,50 +30,6 @@ struct pie_sched_data {
 	struct Qdisc *sch;
 };
 
-static void pie_params_init(struct pie_params *params)
-{
-	params->alpha = 2;
-	params->beta = 20;
-	params->tupdate = usecs_to_jiffies(15 * USEC_PER_MSEC);	/* 15 ms */
-	params->limit = 1000;	/* default of 1000 packets */
-	params->target = PSCHED_NS2TICKS(15 * NSEC_PER_MSEC);	/* 15 ms */
-	params->ecn = false;
-	params->bytemode = false;
-	params->dq_rate_estimator = false;
-}
-
-/* private skb vars */
-struct pie_skb_cb {
-	psched_time_t enqueue_time;
-};
-
-static struct pie_skb_cb *get_pie_cb(const struct sk_buff *skb)
-{
-	qdisc_cb_private_validate(skb, sizeof(struct pie_skb_cb));
-	return (struct pie_skb_cb *)qdisc_skb_cb(skb)->data;
-}
-
-static psched_time_t pie_get_enqueue_time(const struct sk_buff *skb)
-{
-	return get_pie_cb(skb)->enqueue_time;
-}
-
-static void pie_set_enqueue_time(struct sk_buff *skb)
-{
-	get_pie_cb(skb)->enqueue_time = psched_get_time();
-}
-
-static void pie_vars_init(struct pie_vars *vars)
-{
-	vars->dq_count = DQCOUNT_INVALID;
-	vars->dq_tstamp = DTIME_INVALID;
-	vars->accu_prob = 0;
-	vars->avg_dq_rate = 0;
-	/* default of 150 ms in pschedtime */
-	vars->burst_time = PSCHED_NS2TICKS(150 * NSEC_PER_MSEC);
-	vars->accu_prob_overflows = 0;
-}
-
 static bool drop_early(struct Qdisc *sch, u32 packet_size)
 {
 	struct pie_sched_data *q = qdisc_priv(sch);
-- 
cgit v1.2.3-59-g8ed1b


From 805a5a23a4c4ee126e781bd14a1deb242395e817 Mon Sep 17 00:00:00 2001
From: "Mohit P. Tahiliani" <tahiliani@nitk.edu.in>
Date: Wed, 22 Jan 2020 23:52:25 +0530
Subject: pie: use U64_MAX to denote (2^64 - 1)

Use the U64_MAX macro to denote the constant (2^64 - 1).

Signed-off-by: Mohit P. Tahiliani <tahiliani@nitk.edu.in>
Signed-off-by: Leslie Monis <lesliemonis@gmail.com>
Signed-off-by: Gautam Ramakrishnan <gautamramk@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/pie.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/net/pie.h b/include/net/pie.h
index 440213ec83eb..7ef375db5bab 100644
--- a/include/net/pie.h
+++ b/include/net/pie.h
@@ -10,8 +10,8 @@
 
 #define QUEUE_THRESHOLD 16384
 #define DQCOUNT_INVALID -1
-#define DTIME_INVALID 0xffffffffffffffff
-#define MAX_PROB 0xffffffffffffffff
+#define DTIME_INVALID U64_MAX
+#define MAX_PROB U64_MAX
 #define PIE_SCALE 8
 
 /* parameters used */
-- 
cgit v1.2.3-59-g8ed1b


From cf4eeee5ff56180b525bfb6a204071216ca4000a Mon Sep 17 00:00:00 2001
From: "Mohit P. Tahiliani" <tahiliani@nitk.edu.in>
Date: Wed, 22 Jan 2020 23:52:26 +0530
Subject: pie: rearrange macros in order of length

Rearrange macros in order of length and align the values to
improve readability.

Signed-off-by: Mohit P. Tahiliani <tahiliani@nitk.edu.in>
Signed-off-by: Leslie Monis <lesliemonis@gmail.com>
Signed-off-by: Gautam Ramakrishnan <gautamramk@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/pie.h | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/include/net/pie.h b/include/net/pie.h
index 7ef375db5bab..397c7abf0879 100644
--- a/include/net/pie.h
+++ b/include/net/pie.h
@@ -8,11 +8,11 @@
 #include <net/inet_ecn.h>
 #include <net/pkt_sched.h>
 
-#define QUEUE_THRESHOLD 16384
-#define DQCOUNT_INVALID -1
-#define DTIME_INVALID U64_MAX
-#define MAX_PROB U64_MAX
-#define PIE_SCALE 8
+#define MAX_PROB	U64_MAX
+#define DTIME_INVALID	U64_MAX
+#define QUEUE_THRESHOLD	16384
+#define DQCOUNT_INVALID	-1
+#define PIE_SCALE	8
 
 /* parameters used */
 struct pie_params {
-- 
cgit v1.2.3-59-g8ed1b


From 1dbfc5e071db3f5acc3c7c87a564bf57b838cf49 Mon Sep 17 00:00:00 2001
From: "Mohit P. Tahiliani" <tahiliani@nitk.edu.in>
Date: Wed, 22 Jan 2020 23:52:27 +0530
Subject: pie: use u8 instead of bool in pie_vars

Linux best practice recommends using u8 for true/false values in
structures.

Signed-off-by: Mohit P. Tahiliani <tahiliani@nitk.edu.in>
Signed-off-by: Leslie Monis <lesliemonis@gmail.com>
Signed-off-by: Gautam Ramakrishnan <gautamramk@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/pie.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/net/pie.h b/include/net/pie.h
index 397c7abf0879..f9c6a44bdb0c 100644
--- a/include/net/pie.h
+++ b/include/net/pie.h
@@ -21,8 +21,8 @@ struct pie_params {
 	u32 limit;		/* number of packets that can be enqueued */
 	u32 alpha;		/* alpha and beta are between 0 and 32 */
 	u32 beta;		/* and are used for shift relative to 1 */
-	bool ecn;		/* true if ecn is enabled */
-	bool bytemode;		/* to scale drop early prob based on pkt size */
+	u8 ecn;			/* true if ecn is enabled */
+	u8 bytemode;		/* to scale drop early prob based on pkt size */
 	u8 dq_rate_estimator;	/* to calculate delay using Little's law */
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From 2dfb1952a9a1fde0b515f58605c11902e69415bf Mon Sep 17 00:00:00 2001
From: "Mohit P. Tahiliani" <tahiliani@nitk.edu.in>
Date: Wed, 22 Jan 2020 23:52:28 +0530
Subject: pie: rearrange structure members and their initializations

Rearrange the members of the structure such that closely
referenced members appear together and/or fit in the same
cacheline. Also, change the order of their initializations to
match the order in which they appear in the structure.

Signed-off-by: Mohit P. Tahiliani <tahiliani@nitk.edu.in>
Signed-off-by: Leslie Monis <lesliemonis@gmail.com>
Signed-off-by: Gautam Ramakrishnan <gautamramk@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/pie.h   | 20 ++++++++++----------
 net/sched/sch_pie.c |  2 +-
 2 files changed, 11 insertions(+), 11 deletions(-)

diff --git a/include/net/pie.h b/include/net/pie.h
index f9c6a44bdb0c..ec0fbe98ec2f 100644
--- a/include/net/pie.h
+++ b/include/net/pie.h
@@ -28,13 +28,13 @@ struct pie_params {
 
 /* variables used */
 struct pie_vars {
-	u64 prob;		/* probability but scaled by u64 limit. */
-	psched_time_t burst_time;
 	psched_time_t qdelay;
 	psched_time_t qdelay_old;
-	u64 dq_count;		/* measured in bytes */
+	psched_time_t burst_time;
 	psched_time_t dq_tstamp;	/* drain rate */
+	u64 prob;		/* probability but scaled by u64 limit. */
 	u64 accu_prob;		/* accumulated drop probability */
+	u64 dq_count;		/* measured in bytes */
 	u32 avg_dq_rate;	/* bytes per pschedtime tick,scaled */
 	u32 qlen_old;		/* in bytes */
 	u8 accu_prob_overflows;	/* overflows of accu_prob */
@@ -45,8 +45,8 @@ struct pie_stats {
 	u32 packets_in;		/* total number of packets enqueued */
 	u32 dropped;		/* packets dropped due to pie_action */
 	u32 overlimit;		/* dropped due to lack of space in queue */
-	u32 maxq;		/* maximum queue size */
 	u32 ecn_mark;		/* packets marked with ECN */
+	u32 maxq;		/* maximum queue size */
 };
 
 /* private skb vars */
@@ -56,11 +56,11 @@ struct pie_skb_cb {
 
 static inline void pie_params_init(struct pie_params *params)
 {
-	params->alpha = 2;
-	params->beta = 20;
+	params->target = PSCHED_NS2TICKS(15 * NSEC_PER_MSEC);	/* 15 ms */
 	params->tupdate = usecs_to_jiffies(15 * USEC_PER_MSEC);	/* 15 ms */
 	params->limit = 1000;	/* default of 1000 packets */
-	params->target = PSCHED_NS2TICKS(15 * NSEC_PER_MSEC);	/* 15 ms */
+	params->alpha = 2;
+	params->beta = 20;
 	params->ecn = false;
 	params->bytemode = false;
 	params->dq_rate_estimator = false;
@@ -68,12 +68,12 @@ static inline void pie_params_init(struct pie_params *params)
 
 static inline void pie_vars_init(struct pie_vars *vars)
 {
-	vars->dq_count = DQCOUNT_INVALID;
+	/* default of 150 ms in pschedtime */
+	vars->burst_time = PSCHED_NS2TICKS(150 * NSEC_PER_MSEC);
 	vars->dq_tstamp = DTIME_INVALID;
 	vars->accu_prob = 0;
+	vars->dq_count = DQCOUNT_INVALID;
 	vars->avg_dq_rate = 0;
-	/* default of 150 ms in pschedtime */
-	vars->burst_time = PSCHED_NS2TICKS(150 * NSEC_PER_MSEC);
 	vars->accu_prob_overflows = 0;
 }
 
diff --git a/net/sched/sch_pie.c b/net/sched/sch_pie.c
index 7197bcaa14ba..0c583cc148f3 100644
--- a/net/sched/sch_pie.c
+++ b/net/sched/sch_pie.c
@@ -23,8 +23,8 @@
 
 /* private data for the Qdisc */
 struct pie_sched_data {
-	struct pie_params params;
 	struct pie_vars vars;
+	struct pie_params params;
 	struct pie_stats stats;
 	struct timer_list adapt_timer;
 	struct Qdisc *sch;
-- 
cgit v1.2.3-59-g8ed1b


From b42a3d7c7cfff3555d7057c20dbbe57fe75d77c0 Mon Sep 17 00:00:00 2001
From: "Mohit P. Tahiliani" <tahiliani@nitk.edu.in>
Date: Wed, 22 Jan 2020 23:52:29 +0530
Subject: pie: improve comments and commenting style

Improve the comments along with the commenting style used to
describe the members of the structures and their initial values
in the init functions.

Signed-off-by: Mohit P. Tahiliani <tahiliani@nitk.edu.in>
Signed-off-by: Leslie Monis <lesliemonis@gmail.com>
Signed-off-by: Gautam Ramakrishnan <gautamramk@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/pie.h | 85 +++++++++++++++++++++++++++++++++++++------------------
 1 file changed, 58 insertions(+), 27 deletions(-)

diff --git a/include/net/pie.h b/include/net/pie.h
index ec0fbe98ec2f..51a1984c2dce 100644
--- a/include/net/pie.h
+++ b/include/net/pie.h
@@ -14,42 +14,74 @@
 #define DQCOUNT_INVALID	-1
 #define PIE_SCALE	8
 
-/* parameters used */
+/**
+ * struct pie_params - contains pie parameters
+ * @target:		target delay in pschedtime
+ * @tudpate:		interval at which drop probability is calculated
+ * @limit:		total number of packets that can be in the queue
+ * @alpha:		parameter to control drop probability
+ * @beta:		parameter to control drop probability
+ * @ecn:		is ECN marking of packets enabled
+ * @bytemode:		is drop probability scaled based on pkt size
+ * @dq_rate_estimator:	is Little's law used for qdelay calculation
+ */
 struct pie_params {
-	psched_time_t target;	/* user specified target delay in pschedtime */
-	u32 tupdate;		/* timer frequency (in jiffies) */
-	u32 limit;		/* number of packets that can be enqueued */
-	u32 alpha;		/* alpha and beta are between 0 and 32 */
-	u32 beta;		/* and are used for shift relative to 1 */
-	u8 ecn;			/* true if ecn is enabled */
-	u8 bytemode;		/* to scale drop early prob based on pkt size */
-	u8 dq_rate_estimator;	/* to calculate delay using Little's law */
+	psched_time_t target;
+	u32 tupdate;
+	u32 limit;
+	u32 alpha;
+	u32 beta;
+	u8 ecn;
+	u8 bytemode;
+	u8 dq_rate_estimator;
 };
 
-/* variables used */
+/**
+ * struct pie_vars - contains pie variables
+ * @qdelay:			current queue delay
+ * @qdelay_old:			queue delay in previous qdelay calculation
+ * @burst_time:			burst time allowance
+ * @dq_tstamp:			timestamp at which dq rate was last calculated
+ * @prob:			drop probability
+ * @accu_prob:			accumulated drop probability
+ * @dq_count:			number of bytes dequeued in a measurement cycle
+ * @avg_dq_rate:		calculated average dq rate
+ * @qlen_old:			queue length during previous qdelay calculation
+ * @accu_prob_overflows:	number of times accu_prob overflows
+ */
 struct pie_vars {
 	psched_time_t qdelay;
 	psched_time_t qdelay_old;
 	psched_time_t burst_time;
-	psched_time_t dq_tstamp;	/* drain rate */
-	u64 prob;		/* probability but scaled by u64 limit. */
-	u64 accu_prob;		/* accumulated drop probability */
-	u64 dq_count;		/* measured in bytes */
-	u32 avg_dq_rate;	/* bytes per pschedtime tick,scaled */
-	u32 qlen_old;		/* in bytes */
-	u8 accu_prob_overflows;	/* overflows of accu_prob */
+	psched_time_t dq_tstamp;
+	u64 prob;
+	u64 accu_prob;
+	u64 dq_count;
+	u32 avg_dq_rate;
+	u32 qlen_old;
+	u8 accu_prob_overflows;
 };
 
-/* statistics gathering */
+/**
+ * struct pie_stats - contains pie stats
+ * @packets_in:	total number of packets enqueued
+ * @dropped:	packets dropped due to pie action
+ * @overlimit:	packets dropped due to lack of space in queue
+ * @ecn_mark:	packets marked with ECN
+ * @maxq:	maximum queue size
+ */
 struct pie_stats {
-	u32 packets_in;		/* total number of packets enqueued */
-	u32 dropped;		/* packets dropped due to pie_action */
-	u32 overlimit;		/* dropped due to lack of space in queue */
-	u32 ecn_mark;		/* packets marked with ECN */
-	u32 maxq;		/* maximum queue size */
+	u32 packets_in;
+	u32 dropped;
+	u32 overlimit;
+	u32 ecn_mark;
+	u32 maxq;
 };
 
-/* private skb vars */
+/**
+ * struct pie_skb_cb - contains private skb vars
+ * @enqueue_time:	timestamp when the packet is enqueued
+ */
 struct pie_skb_cb {
 	psched_time_t enqueue_time;
 };
@@ -58,7 +90,7 @@ static inline void pie_params_init(struct pie_params *params)
 {
 	params->target = PSCHED_NS2TICKS(15 * NSEC_PER_MSEC);	/* 15 ms */
 	params->tupdate = usecs_to_jiffies(15 * USEC_PER_MSEC);	/* 15 ms */
-	params->limit = 1000;	/* default of 1000 packets */
+	params->limit = 1000;
 	params->alpha = 2;
 	params->beta = 20;
 	params->ecn = false;
@@ -68,8 +100,7 @@ static inline void pie_params_init(struct pie_params *params)
 
 static inline void pie_vars_init(struct pie_vars *vars)
 {
-	/* default of 150 ms in pschedtime */
-	vars->burst_time = PSCHED_NS2TICKS(150 * NSEC_PER_MSEC);
+	vars->burst_time = PSCHED_NS2TICKS(150 * NSEC_PER_MSEC); /* 150 ms */
 	vars->dq_tstamp = DTIME_INVALID;
 	vars->accu_prob = 0;
 	vars->dq_count = DQCOUNT_INVALID;
-- 
cgit v1.2.3-59-g8ed1b


From 55f780c4a6c3046461352f4081427b077f8d06ed Mon Sep 17 00:00:00 2001
From: "Mohit P. Tahiliani" <tahiliani@nitk.edu.in>
Date: Wed, 22 Jan 2020 23:52:30 +0530
Subject: net: sched: pie: fix commenting

Fix punctuation and logical mistakes in the comments. The
logical mistake was that "dequeue_rate" is no longer the default
way to calculate queuing delay and is not needed. The default
way to calculate queue delay was changed in commit cec2975f2b70
("net: sched: pie: enable timestamp based delay calculation").

Signed-off-by: Mohit P. Tahiliani <tahiliani@nitk.edu.in>
Signed-off-by: Leslie Monis <lesliemonis@gmail.com>
Signed-off-by: Gautam Ramakrishnan <gautamramk@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sched/sch_pie.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/net/sched/sch_pie.c b/net/sched/sch_pie.c
index 0c583cc148f3..024f55569a38 100644
--- a/net/sched/sch_pie.c
+++ b/net/sched/sch_pie.c
@@ -248,10 +248,10 @@ static void pie_process_dequeue(struct Qdisc *sch, struct sk_buff *skb)
 		q->vars.dq_count = 0;
 	}
 
-	/* Calculate the average drain rate from this value.  If queue length
-	 * has receded to a small value viz., <= QUEUE_THRESHOLD bytes,reset
+	/* Calculate the average drain rate from this value. If queue length
+	 * has receded to a small value viz., <= QUEUE_THRESHOLD bytes, reset
 	 * the dq_count to -1 as we don't have enough packets to calculate the
-	 * drain rate anymore The following if block is entered only when we
+	 * drain rate anymore. The following if block is entered only when we
 	 * have a substantial queue built up (QUEUE_THRESHOLD bytes or more)
 	 * and we calculate the drain rate for the threshold here.  dq_count is
 	 * in bytes, time difference in psched_time, hence rate is in
@@ -329,8 +329,8 @@ static void calculate_probability(struct Qdisc *sch)
 		qdelay_old = q->vars.qdelay_old;
 	}
 
-	/* If qdelay is zero and qlen is not, it means qlen is very small, less
-	 * than dequeue_rate, so we do not update probabilty in this round
+	/* If qdelay is zero and qlen is not, it means qlen is very small,
+	 * so we do not update probabilty in this round.
 	 */
 	if (qdelay == 0 && qlen != 0)
 		update_prob = false;
-- 
cgit v1.2.3-59-g8ed1b


From 00ea2fb7274f568cd982a5958c66cab578aada25 Mon Sep 17 00:00:00 2001
From: "Mohit P. Tahiliani" <tahiliani@nitk.edu.in>
Date: Wed, 22 Jan 2020 23:52:31 +0530
Subject: net: sched: pie: fix alignment in struct instances

Make the alignment in the initialization of the struct instances
consistent in the file.

Signed-off-by: Mohit P. Tahiliani <tahiliani@nitk.edu.in>
Signed-off-by: Leslie Monis <lesliemonis@gmail.com>
Signed-off-by: Gautam Ramakrishnan <gautamramk@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sched/sch_pie.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/net/sched/sch_pie.c b/net/sched/sch_pie.c
index 024f55569a38..c65164659bca 100644
--- a/net/sched/sch_pie.c
+++ b/net/sched/sch_pie.c
@@ -132,14 +132,14 @@ out:
 }
 
 static const struct nla_policy pie_policy[TCA_PIE_MAX + 1] = {
-	[TCA_PIE_TARGET] = {.type = NLA_U32},
-	[TCA_PIE_LIMIT] = {.type = NLA_U32},
-	[TCA_PIE_TUPDATE] = {.type = NLA_U32},
-	[TCA_PIE_ALPHA] = {.type = NLA_U32},
-	[TCA_PIE_BETA] = {.type = NLA_U32},
-	[TCA_PIE_ECN] = {.type = NLA_U32},
-	[TCA_PIE_BYTEMODE] = {.type = NLA_U32},
-	[TCA_PIE_DQ_RATE_ESTIMATOR] = {.type = NLA_U32},
+	[TCA_PIE_TARGET]		= {.type = NLA_U32},
+	[TCA_PIE_LIMIT]			= {.type = NLA_U32},
+	[TCA_PIE_TUPDATE]		= {.type = NLA_U32},
+	[TCA_PIE_ALPHA]			= {.type = NLA_U32},
+	[TCA_PIE_BETA]			= {.type = NLA_U32},
+	[TCA_PIE_ECN]			= {.type = NLA_U32},
+	[TCA_PIE_BYTEMODE]		= {.type = NLA_U32},
+	[TCA_PIE_DQ_RATE_ESTIMATOR]	= {.type = NLA_U32},
 };
 
 static int pie_change(struct Qdisc *sch, struct nlattr *opt,
@@ -549,7 +549,7 @@ static void pie_destroy(struct Qdisc *sch)
 }
 
 static struct Qdisc_ops pie_qdisc_ops __read_mostly = {
-	.id = "pie",
+	.id		= "pie",
 	.priv_size	= sizeof(struct pie_sched_data),
 	.enqueue	= pie_qdisc_enqueue,
 	.dequeue	= pie_qdisc_dequeue,
-- 
cgit v1.2.3-59-g8ed1b


From 5205ea00cda1ac23cebfb97dfccca84722d58dfe Mon Sep 17 00:00:00 2001
From: "Mohit P. Tahiliani" <tahiliani@nitk.edu.in>
Date: Wed, 22 Jan 2020 23:52:32 +0530
Subject: net: sched: pie: export symbols to be reused by FQ-PIE

This patch makes the drop_early(), calculate_probability() and
pie_process_dequeue() functions generic enough to be used by
both PIE and FQ-PIE (to be added in a future commit). The major
change here is in the way the functions take in arguments. This
patch exports these functions and makes FQ-PIE dependent on
sch_pie.

Signed-off-by: Mohit P. Tahiliani <tahiliani@nitk.edu.in>
Signed-off-by: Leslie Monis <lesliemonis@gmail.com>
Signed-off-by: Gautam Ramakrishnan <gautamramk@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/pie.h   |   9 +++
 net/sched/sch_pie.c | 173 ++++++++++++++++++++++++++--------------------------
 2 files changed, 97 insertions(+), 85 deletions(-)

diff --git a/include/net/pie.h b/include/net/pie.h
index 51a1984c2dce..90f5db3d29e7 100644
--- a/include/net/pie.h
+++ b/include/net/pie.h
@@ -124,4 +124,13 @@ static inline void pie_set_enqueue_time(struct sk_buff *skb)
 	get_pie_cb(skb)->enqueue_time = psched_get_time();
 }
 
+bool pie_drop_early(struct Qdisc *sch, struct pie_params *params,
+		    struct pie_vars *vars, u32 qlen, u32 packet_size);
+
+void pie_process_dequeue(struct sk_buff *skb, struct pie_params *params,
+			 struct pie_vars *vars, u32 qlen);
+
+void pie_calculate_probability(struct pie_params *params, struct pie_vars *vars,
+			       u32 qlen);
+
 #endif
diff --git a/net/sched/sch_pie.c b/net/sched/sch_pie.c
index c65164659bca..915bcdb59a9f 100644
--- a/net/sched/sch_pie.c
+++ b/net/sched/sch_pie.c
@@ -30,64 +30,65 @@ struct pie_sched_data {
 	struct Qdisc *sch;
 };
 
-static bool drop_early(struct Qdisc *sch, u32 packet_size)
+bool pie_drop_early(struct Qdisc *sch, struct pie_params *params,
+		    struct pie_vars *vars, u32 qlen, u32 packet_size)
 {
-	struct pie_sched_data *q = qdisc_priv(sch);
 	u64 rnd;
-	u64 local_prob = q->vars.prob;
+	u64 local_prob = vars->prob;
 	u32 mtu = psched_mtu(qdisc_dev(sch));
 
 	/* If there is still burst allowance left skip random early drop */
-	if (q->vars.burst_time > 0)
+	if (vars->burst_time > 0)
 		return false;
 
 	/* If current delay is less than half of target, and
 	 * if drop prob is low already, disable early_drop
 	 */
-	if ((q->vars.qdelay < q->params.target / 2) &&
-	    (q->vars.prob < MAX_PROB / 5))
+	if ((vars->qdelay < params->target / 2) &&
+	    (vars->prob < MAX_PROB / 5))
 		return false;
 
-	/* If we have fewer than 2 mtu-sized packets, disable drop_early,
+	/* If we have fewer than 2 mtu-sized packets, disable pie_drop_early,
 	 * similar to min_th in RED
 	 */
-	if (sch->qstats.backlog < 2 * mtu)
+	if (qlen < 2 * mtu)
 		return false;
 
 	/* If bytemode is turned on, use packet size to compute new
 	 * probablity. Smaller packets will have lower drop prob in this case
 	 */
-	if (q->params.bytemode && packet_size <= mtu)
+	if (params->bytemode && packet_size <= mtu)
 		local_prob = (u64)packet_size * div_u64(local_prob, mtu);
 	else
-		local_prob = q->vars.prob;
+		local_prob = vars->prob;
 
 	if (local_prob == 0) {
-		q->vars.accu_prob = 0;
-		q->vars.accu_prob_overflows = 0;
+		vars->accu_prob = 0;
+		vars->accu_prob_overflows = 0;
 	}
 
-	if (local_prob > MAX_PROB - q->vars.accu_prob)
-		q->vars.accu_prob_overflows++;
+	if (local_prob > MAX_PROB - vars->accu_prob)
+		vars->accu_prob_overflows++;
 
-	q->vars.accu_prob += local_prob;
+	vars->accu_prob += local_prob;
 
-	if (q->vars.accu_prob_overflows == 0 &&
-	    q->vars.accu_prob < (MAX_PROB / 100) * 85)
+	if (vars->accu_prob_overflows == 0 &&
+	    vars->accu_prob < (MAX_PROB / 100) * 85)
 		return false;
-	if (q->vars.accu_prob_overflows == 8 &&
-	    q->vars.accu_prob >= MAX_PROB / 2)
+	if (vars->accu_prob_overflows == 8 &&
+	    vars->accu_prob >= MAX_PROB / 2)
 		return true;
 
 	prandom_bytes(&rnd, 8);
 	if (rnd < local_prob) {
-		q->vars.accu_prob = 0;
-		q->vars.accu_prob_overflows = 0;
+		vars->accu_prob = 0;
+		vars->accu_prob_overflows = 0;
 		return true;
 	}
 
 	return false;
 }
+EXPORT_SYMBOL_GPL(pie_drop_early);
 
 static int pie_qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch,
 			     struct sk_buff **to_free)
@@ -100,7 +101,8 @@ static int pie_qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch,
 		goto out;
 	}
 
-	if (!drop_early(sch, skb->len)) {
+	if (!pie_drop_early(sch, &q->params, &q->vars, sch->qstats.backlog,
+			    skb->len)) {
 		enqueue = true;
 	} else if (q->params.ecn && (q->vars.prob <= MAX_PROB / 10) &&
 		   INET_ECN_set_ce(skb)) {
@@ -212,26 +214,25 @@ static int pie_change(struct Qdisc *sch, struct nlattr *opt,
 	return 0;
 }
 
-static void pie_process_dequeue(struct Qdisc *sch, struct sk_buff *skb)
+void pie_process_dequeue(struct sk_buff *skb, struct pie_params *params,
+			 struct pie_vars *vars, u32 qlen)
 {
-	struct pie_sched_data *q = qdisc_priv(sch);
-	int qlen = sch->qstats.backlog;	/* current queue size in bytes */
 	psched_time_t now = psched_get_time();
 	u32 dtime = 0;
 
 	/* If dq_rate_estimator is disabled, calculate qdelay using the
 	 * packet timestamp.
 	 */
-	if (!q->params.dq_rate_estimator) {
-		q->vars.qdelay = now - pie_get_enqueue_time(skb);
+	if (!params->dq_rate_estimator) {
+		vars->qdelay = now - pie_get_enqueue_time(skb);
 
-		if (q->vars.dq_tstamp != DTIME_INVALID)
-			dtime = now - q->vars.dq_tstamp;
+		if (vars->dq_tstamp != DTIME_INVALID)
+			dtime = now - vars->dq_tstamp;
 
-		q->vars.dq_tstamp = now;
+		vars->dq_tstamp = now;
 
 		if (qlen == 0)
-			q->vars.qdelay = 0;
+			vars->qdelay = 0;
 
 		if (dtime == 0)
 			return;
@@ -243,9 +244,9 @@ static void pie_process_dequeue(struct Qdisc *sch, struct sk_buff *skb)
 	 * we have enough packets to calculate the drain rate. Save
 	 * current time as dq_tstamp and start measurement cycle.
 	 */
-	if (qlen >= QUEUE_THRESHOLD && q->vars.dq_count == DQCOUNT_INVALID) {
-		q->vars.dq_tstamp = psched_get_time();
-		q->vars.dq_count = 0;
+	if (qlen >= QUEUE_THRESHOLD && vars->dq_count == DQCOUNT_INVALID) {
+		vars->dq_tstamp = psched_get_time();
+		vars->dq_count = 0;
 	}
 
 	/* Calculate the average drain rate from this value. If queue length
@@ -257,25 +258,25 @@ static void pie_process_dequeue(struct Qdisc *sch, struct sk_buff *skb)
 	 * in bytes, time difference in psched_time, hence rate is in
 	 * bytes/psched_time.
 	 */
-	if (q->vars.dq_count != DQCOUNT_INVALID) {
-		q->vars.dq_count += skb->len;
+	if (vars->dq_count != DQCOUNT_INVALID) {
+		vars->dq_count += skb->len;
 
-		if (q->vars.dq_count >= QUEUE_THRESHOLD) {
-			u32 count = q->vars.dq_count << PIE_SCALE;
+		if (vars->dq_count >= QUEUE_THRESHOLD) {
+			u32 count = vars->dq_count << PIE_SCALE;
 
-			dtime = now - q->vars.dq_tstamp;
+			dtime = now - vars->dq_tstamp;
 
 			if (dtime == 0)
 				return;
 
 			count = count / dtime;
 
-			if (q->vars.avg_dq_rate == 0)
-				q->vars.avg_dq_rate = count;
+			if (vars->avg_dq_rate == 0)
+				vars->avg_dq_rate = count;
 			else
-				q->vars.avg_dq_rate =
-				    (q->vars.avg_dq_rate -
-				     (q->vars.avg_dq_rate >> 3)) + (count >> 3);
+				vars->avg_dq_rate =
+				    (vars->avg_dq_rate -
+				     (vars->avg_dq_rate >> 3)) + (count >> 3);
 
 			/* If the queue has receded below the threshold, we hold
 			 * on to the last drain rate calculated, else we reset
@@ -283,10 +284,10 @@ static void pie_process_dequeue(struct Qdisc *sch, struct sk_buff *skb)
 			 * packet is dequeued
 			 */
 			if (qlen < QUEUE_THRESHOLD) {
-				q->vars.dq_count = DQCOUNT_INVALID;
+				vars->dq_count = DQCOUNT_INVALID;
 			} else {
-				q->vars.dq_count = 0;
-				q->vars.dq_tstamp = psched_get_time();
+				vars->dq_count = 0;
+				vars->dq_tstamp = psched_get_time();
 			}
 
 			goto burst_allowance_reduction;
@@ -296,18 +297,18 @@ static void pie_process_dequeue(struct Qdisc *sch, struct sk_buff *skb)
 	return;
 
 burst_allowance_reduction:
-	if (q->vars.burst_time > 0) {
-		if (q->vars.burst_time > dtime)
-			q->vars.burst_time -= dtime;
+	if (vars->burst_time > 0) {
+		if (vars->burst_time > dtime)
+			vars->burst_time -= dtime;
 		else
-			q->vars.burst_time = 0;
+			vars->burst_time = 0;
 	}
 }
+EXPORT_SYMBOL_GPL(pie_process_dequeue);
 
-static void calculate_probability(struct Qdisc *sch)
+void pie_calculate_probability(struct pie_params *params, struct pie_vars *vars,
+			       u32 qlen)
 {
-	struct pie_sched_data *q = qdisc_priv(sch);
-	u32 qlen = sch->qstats.backlog;	/* queue size in bytes */
 	psched_time_t qdelay = 0;	/* in pschedtime */
 	psched_time_t qdelay_old = 0;	/* in pschedtime */
 	s64 delta = 0;		/* determines the change in probability */
@@ -316,17 +317,17 @@ static void calculate_probability(struct Qdisc *sch)
 	u32 power;
 	bool update_prob = true;
 
-	if (q->params.dq_rate_estimator) {
-		qdelay_old = q->vars.qdelay;
-		q->vars.qdelay_old = q->vars.qdelay;
+	if (params->dq_rate_estimator) {
+		qdelay_old = vars->qdelay;
+		vars->qdelay_old = vars->qdelay;
 
-		if (q->vars.avg_dq_rate > 0)
-			qdelay = (qlen << PIE_SCALE) / q->vars.avg_dq_rate;
+		if (vars->avg_dq_rate > 0)
+			qdelay = (qlen << PIE_SCALE) / vars->avg_dq_rate;
 		else
 			qdelay = 0;
 	} else {
-		qdelay = q->vars.qdelay;
-		qdelay_old = q->vars.qdelay_old;
+		qdelay = vars->qdelay;
+		qdelay_old = vars->qdelay_old;
 	}
 
 	/* If qdelay is zero and qlen is not, it means qlen is very small,
@@ -342,18 +343,18 @@ static void calculate_probability(struct Qdisc *sch)
 	 * probability. alpha/beta are updated locally below by scaling down
 	 * by 16 to come to 0-2 range.
 	 */
-	alpha = ((u64)q->params.alpha * (MAX_PROB / PSCHED_TICKS_PER_SEC)) >> 4;
-	beta = ((u64)q->params.beta * (MAX_PROB / PSCHED_TICKS_PER_SEC)) >> 4;
+	alpha = ((u64)params->alpha * (MAX_PROB / PSCHED_TICKS_PER_SEC)) >> 4;
+	beta = ((u64)params->beta * (MAX_PROB / PSCHED_TICKS_PER_SEC)) >> 4;
 
 	/* We scale alpha and beta differently depending on how heavy the
 	 * congestion is. Please see RFC 8033 for details.
 	 */
-	if (q->vars.prob < MAX_PROB / 10) {
+	if (vars->prob < MAX_PROB / 10) {
 		alpha >>= 1;
 		beta >>= 1;
 
 		power = 100;
-		while (q->vars.prob < div_u64(MAX_PROB, power) &&
+		while (vars->prob < div_u64(MAX_PROB, power) &&
 		       power <= 1000000) {
 			alpha >>= 2;
 			beta >>= 2;
@@ -362,14 +363,14 @@ static void calculate_probability(struct Qdisc *sch)
 	}
 
 	/* alpha and beta should be between 0 and 32, in multiples of 1/16 */
-	delta += alpha * (u64)(qdelay - q->params.target);
+	delta += alpha * (u64)(qdelay - params->target);
 	delta += beta * (u64)(qdelay - qdelay_old);
 
-	oldprob = q->vars.prob;
+	oldprob = vars->prob;
 
 	/* to ensure we increase probability in steps of no more than 2% */
 	if (delta > (s64)(MAX_PROB / (100 / 2)) &&
-	    q->vars.prob >= MAX_PROB / 10)
+	    vars->prob >= MAX_PROB / 10)
 		delta = (MAX_PROB / 100) * 2;
 
 	/* Non-linear drop:
@@ -380,12 +381,12 @@ static void calculate_probability(struct Qdisc *sch)
 	if (qdelay > (PSCHED_NS2TICKS(250 * NSEC_PER_MSEC)))
 		delta += MAX_PROB / (100 / 2);
 
-	q->vars.prob += delta;
+	vars->prob += delta;
 
 	if (delta > 0) {
 		/* prevent overflow */
-		if (q->vars.prob < oldprob) {
-			q->vars.prob = MAX_PROB;
+		if (vars->prob < oldprob) {
+			vars->prob = MAX_PROB;
 			/* Prevent normalization error. If probability is at
 			 * maximum value already, we normalize it here, and
 			 * skip the check to do a non-linear drop in the next
@@ -395,8 +396,8 @@ static void calculate_probability(struct Qdisc *sch)
 		}
 	} else {
 		/* prevent underflow */
-		if (q->vars.prob > oldprob)
-			q->vars.prob = 0;
+		if (vars->prob > oldprob)
+			vars->prob = 0;
 	}
 
 	/* Non-linear drop in probability: Reduce drop probability quickly if
@@ -405,10 +406,10 @@ static void calculate_probability(struct Qdisc *sch)
 
 	if (qdelay == 0 && qdelay_old == 0 && update_prob)
 		/* Reduce drop probability to 98.4% */
-		q->vars.prob -= q->vars.prob / 64u;
+		vars->prob -= vars->prob / 64;
 
-	q->vars.qdelay = qdelay;
-	q->vars.qlen_old = qlen;
+	vars->qdelay = qdelay;
+	vars->qlen_old = qlen;
 
 	/* We restart the measurement cycle if the following conditions are met
 	 * 1. If the delay has been low for 2 consecutive Tupdate periods
@@ -416,16 +417,17 @@ static void calculate_probability(struct Qdisc *sch)
 	 * 3. If average dq_rate_estimator is enabled, we have atleast one
 	 *    estimate for the avg_dq_rate ie., is a non-zero value
 	 */
-	if ((q->vars.qdelay < q->params.target / 2) &&
-	    (q->vars.qdelay_old < q->params.target / 2) &&
-	    q->vars.prob == 0 &&
-	    (!q->params.dq_rate_estimator || q->vars.avg_dq_rate > 0)) {
-		pie_vars_init(&q->vars);
+	if ((vars->qdelay < params->target / 2) &&
+	    (vars->qdelay_old < params->target / 2) &&
+	    vars->prob == 0 &&
+	    (!params->dq_rate_estimator || vars->avg_dq_rate > 0)) {
+		pie_vars_init(vars);
 	}
 
-	if (!q->params.dq_rate_estimator)
-		q->vars.qdelay_old = qdelay;
+	if (!params->dq_rate_estimator)
+		vars->qdelay_old = qdelay;
 }
+EXPORT_SYMBOL_GPL(pie_calculate_probability);
 
 static void pie_timer(struct timer_list *t)
 {
@@ -434,7 +436,7 @@ static void pie_timer(struct timer_list *t)
 	spinlock_t *root_lock = qdisc_lock(qdisc_root_sleeping(sch));
 
 	spin_lock(root_lock);
-	calculate_probability(sch);
+	pie_calculate_probability(&q->params, &q->vars, sch->qstats.backlog);
 
 	/* reset the timer to fire after 'tupdate'. tupdate is in jiffies. */
 	if (q->params.tupdate)
@@ -523,12 +525,13 @@ static int pie_dump_stats(struct Qdisc *sch, struct gnet_dump *d)
 
 static struct sk_buff *pie_qdisc_dequeue(struct Qdisc *sch)
 {
+	struct pie_sched_data *q = qdisc_priv(sch);
 	struct sk_buff *skb = qdisc_dequeue_head(sch);
 
 	if (!skb)
 		return NULL;
 
-	pie_process_dequeue(sch, skb);
+	pie_process_dequeue(skb, &q->params, &q->vars, sch->qstats.backlog);
 	return skb;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From ec97ecf1ebe485a17cd8395a5f35e6b80b57665a Mon Sep 17 00:00:00 2001
From: "Mohit P. Tahiliani" <tahiliani@nitk.edu.in>
Date: Wed, 22 Jan 2020 23:52:33 +0530
Subject: net: sched: add Flow Queue PIE packet scheduler

Principles:
  - Packets are classified on flows.
  - This is a Stochastic model (as we use a hash, several flows might
                                be hashed to the same slot)
  - Each flow has a PIE managed queue.
  - Flows are linked onto two (Round Robin) lists,
    so that new flows have priority on old ones.
  - For a given flow, packets are not reordered.
  - Drops during enqueue only.
  - ECN capability is off by default.
  - ECN threshold (if ECN is enabled) is at 10% by default.
  - Uses timestamps to calculate queue delay by default.

Usage:
tc qdisc ... fq_pie [ limit PACKETS ] [ flows NUMBER ]
                    [ target TIME ] [ tupdate TIME ]
                    [ alpha NUMBER ] [ beta NUMBER ]
                    [ quantum BYTES ] [ memory_limit BYTES ]
                    [ ecnprob PERCENTAGE ] [ [no]ecn ]
                    [ [no]bytemode ] [ [no_]dq_rate_estimator ]

defaults:
  limit: 10240 packets, flows: 1024
  target: 15 ms, tupdate: 15 ms (in jiffies)
  alpha: 1/8, beta : 5/4
  quantum: device MTU, memory_limit: 32 Mb
  ecnprob: 10%, ecn: off
  bytemode: off, dq_rate_estimator: off

Signed-off-by: Mohit P. Tahiliani <tahiliani@nitk.edu.in>
Signed-off-by: Sachin D. Patil <sdp.sachin@gmail.com>
Signed-off-by: V. Saicharan <vsaicharan1998@gmail.com>
Signed-off-by: Mohit Bhasi <mohitbhasi1998@gmail.com>
Signed-off-by: Leslie Monis <lesliemonis@gmail.com>
Signed-off-by: Gautam Ramakrishnan <gautamramk@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/pie.h              |   2 +
 include/uapi/linux/pkt_sched.h |  31 +++
 net/sched/Kconfig              |  13 +
 net/sched/Makefile             |   1 +
 net/sched/sch_fq_pie.c         | 562 +++++++++++++++++++++++++++++++++++++++++
 5 files changed, 609 insertions(+)
 create mode 100644 net/sched/sch_fq_pie.c

diff --git a/include/net/pie.h b/include/net/pie.h
index 90f5db3d29e7..fd5a37cb7993 100644
--- a/include/net/pie.h
+++ b/include/net/pie.h
@@ -81,9 +81,11 @@ struct pie_stats {
 /**
  * struct pie_skb_cb - contains private skb vars
  * @enqueue_time:	timestamp when the packet is enqueued
+ * @mem_usage:		size of the skb during enqueue
  */
 struct pie_skb_cb {
 	psched_time_t enqueue_time;
+	u32 mem_usage;
 };
 
 static inline void pie_params_init(struct pie_params *params)
diff --git a/include/uapi/linux/pkt_sched.h b/include/uapi/linux/pkt_sched.h
index bf5a5b1dfb0b..bbe791b24168 100644
--- a/include/uapi/linux/pkt_sched.h
+++ b/include/uapi/linux/pkt_sched.h
@@ -971,6 +971,37 @@ struct tc_pie_xstats {
 	__u32 ecn_mark;			/* packets marked with ecn*/
 };
 
+/* FQ PIE */
+enum {
+	TCA_FQ_PIE_UNSPEC,
+	TCA_FQ_PIE_LIMIT,
+	TCA_FQ_PIE_FLOWS,
+	TCA_FQ_PIE_TARGET,
+	TCA_FQ_PIE_TUPDATE,
+	TCA_FQ_PIE_ALPHA,
+	TCA_FQ_PIE_BETA,
+	TCA_FQ_PIE_QUANTUM,
+	TCA_FQ_PIE_MEMORY_LIMIT,
+	TCA_FQ_PIE_ECN_PROB,
+	TCA_FQ_PIE_ECN,
+	TCA_FQ_PIE_BYTEMODE,
+	TCA_FQ_PIE_DQ_RATE_ESTIMATOR,
+	__TCA_FQ_PIE_MAX
+};
+#define TCA_FQ_PIE_MAX   (__TCA_FQ_PIE_MAX - 1)
+
+struct tc_fq_pie_xstats {
+	__u32 packets_in;	/* total number of packets enqueued */
+	__u32 dropped;		/* packets dropped due to fq_pie_action */
+	__u32 overlimit;	/* dropped due to lack of space in queue */
+	__u32 overmemory;	/* dropped due to lack of memory in queue */
+	__u32 ecn_mark;		/* packets marked with ecn */
+	__u32 new_flow_count;	/* count of new flows created by packets */
+	__u32 new_flows_len;	/* count of flows in new list */
+	__u32 old_flows_len;	/* count of flows in old list */
+	__u32 memory_usage;	/* total memory across all queues */
+};
+
 /* CBS */
 struct tc_cbs_qopt {
 	__u8 offload;
diff --git a/net/sched/Kconfig b/net/sched/Kconfig
index b1e7ec726958..edde0e519438 100644
--- a/net/sched/Kconfig
+++ b/net/sched/Kconfig
@@ -366,6 +366,19 @@ config NET_SCH_PIE
 
 	  If unsure, say N.
 
+config NET_SCH_FQ_PIE
+	depends on NET_SCH_PIE
+	tristate "Flow Queue Proportional Integral controller Enhanced (FQ-PIE)"
+	help
+	  Say Y here if you want to use the Flow Queue Proportional Integral
+	  controller Enhanced (FQ-PIE) packet scheduling algorithm.
+	  For more information, please see https://tools.ietf.org/html/rfc8033
+
+	  To compile this driver as a module, choose M here: the module
+	  will be called sch_fq_pie.
+
+	  If unsure, say N.
+
 config NET_SCH_INGRESS
 	tristate "Ingress/classifier-action Qdisc"
 	depends on NET_CLS_ACT
diff --git a/net/sched/Makefile b/net/sched/Makefile
index bc8856b865ff..31c367a6cd09 100644
--- a/net/sched/Makefile
+++ b/net/sched/Makefile
@@ -59,6 +59,7 @@ obj-$(CONFIG_NET_SCH_CAKE)	+= sch_cake.o
 obj-$(CONFIG_NET_SCH_FQ)	+= sch_fq.o
 obj-$(CONFIG_NET_SCH_HHF)	+= sch_hhf.o
 obj-$(CONFIG_NET_SCH_PIE)	+= sch_pie.o
+obj-$(CONFIG_NET_SCH_FQ_PIE)	+= sch_fq_pie.o
 obj-$(CONFIG_NET_SCH_CBS)	+= sch_cbs.o
 obj-$(CONFIG_NET_SCH_ETF)	+= sch_etf.o
 obj-$(CONFIG_NET_SCH_TAPRIO)	+= sch_taprio.o
diff --git a/net/sched/sch_fq_pie.c b/net/sched/sch_fq_pie.c
new file mode 100644
index 000000000000..bbd0dea6b6b9
--- /dev/null
+++ b/net/sched/sch_fq_pie.c
@@ -0,0 +1,562 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Flow Queue PIE discipline
+ *
+ * Copyright (C) 2019 Mohit P. Tahiliani <tahiliani@nitk.edu.in>
+ * Copyright (C) 2019 Sachin D. Patil <sdp.sachin@gmail.com>
+ * Copyright (C) 2019 V. Saicharan <vsaicharan1998@gmail.com>
+ * Copyright (C) 2019 Mohit Bhasi <mohitbhasi1998@gmail.com>
+ * Copyright (C) 2019 Leslie Monis <lesliemonis@gmail.com>
+ * Copyright (C) 2019 Gautam Ramakrishnan <gautamramk@gmail.com>
+ */
+
+#include <linux/jhash.h>
+#include <linux/sizes.h>
+#include <linux/vmalloc.h>
+#include <net/pkt_cls.h>
+#include <net/pie.h>
+
+/* Flow Queue PIE
+ *
+ * Principles:
+ *   - Packets are classified on flows.
+ *   - This is a Stochastic model (as we use a hash, several flows might
+ *                                 be hashed to the same slot)
+ *   - Each flow has a PIE managed queue.
+ *   - Flows are linked onto two (Round Robin) lists,
+ *     so that new flows have priority on old ones.
+ *   - For a given flow, packets are not reordered.
+ *   - Drops during enqueue only.
+ *   - ECN capability is off by default.
+ *   - ECN threshold (if ECN is enabled) is at 10% by default.
+ *   - Uses timestamps to calculate queue delay by default.
+ */
+
+/**
+ * struct fq_pie_flow - contains data for each flow
+ * @vars:	pie vars associated with the flow
+ * @deficit:	number of remaining byte credits
+ * @backlog:	size of data in the flow
+ * @qlen:	number of packets in the flow
+ * @flowchain:	flowchain for the flow
+ * @head:	first packet in the flow
+ * @tail:	last packet in the flow
+ */
+struct fq_pie_flow {
+	struct pie_vars vars;
+	s32 deficit;
+	u32 backlog;
+	u32 qlen;
+	struct list_head flowchain;
+	struct sk_buff *head;
+	struct sk_buff *tail;
+};
+
+struct fq_pie_sched_data {
+	struct tcf_proto __rcu *filter_list; /* optional external classifier */
+	struct tcf_block *block;
+	struct fq_pie_flow *flows;
+	struct Qdisc *sch;
+	struct list_head old_flows;
+	struct list_head new_flows;
+	struct pie_params p_params;
+	u32 ecn_prob;
+	u32 flows_cnt;
+	u32 quantum;
+	u32 memory_limit;
+	u32 new_flow_count;
+	u32 memory_usage;
+	u32 overmemory;
+	struct pie_stats stats;
+	struct timer_list adapt_timer;
+};
+
+static unsigned int fq_pie_hash(const struct fq_pie_sched_data *q,
+				struct sk_buff *skb)
+{
+	return reciprocal_scale(skb_get_hash(skb), q->flows_cnt);
+}
+
+static unsigned int fq_pie_classify(struct sk_buff *skb, struct Qdisc *sch,
+				    int *qerr)
+{
+	struct fq_pie_sched_data *q = qdisc_priv(sch);
+	struct tcf_proto *filter;
+	struct tcf_result res;
+	int result;
+
+	if (TC_H_MAJ(skb->priority) == sch->handle &&
+	    TC_H_MIN(skb->priority) > 0 &&
+	    TC_H_MIN(skb->priority) <= q->flows_cnt)
+		return TC_H_MIN(skb->priority);
+
+	filter = rcu_dereference_bh(q->filter_list);
+	if (!filter)
+		return fq_pie_hash(q, skb) + 1;
+
+	*qerr = NET_XMIT_SUCCESS | __NET_XMIT_BYPASS;
+	result = tcf_classify(skb, filter, &res, false);
+	if (result >= 0) {
+#ifdef CONFIG_NET_CLS_ACT
+		switch (result) {
+		case TC_ACT_STOLEN:
+		case TC_ACT_QUEUED:
+		case TC_ACT_TRAP:
+			*qerr = NET_XMIT_SUCCESS | __NET_XMIT_STOLEN;
+			/* fall through */
+		case TC_ACT_SHOT:
+			return 0;
+		}
+#endif
+		if (TC_H_MIN(res.classid) <= q->flows_cnt)
+			return TC_H_MIN(res.classid);
+	}
+	return 0;
+}
+
+/* add skb to flow queue (tail add) */
+static inline void flow_queue_add(struct fq_pie_flow *flow,
+				  struct sk_buff *skb)
+{
+	if (!flow->head)
+		flow->head = skb;
+	else
+		flow->tail->next = skb;
+	flow->tail = skb;
+	skb->next = NULL;
+}
+
+static int fq_pie_qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch,
+				struct sk_buff **to_free)
+{
+	struct fq_pie_sched_data *q = qdisc_priv(sch);
+	struct fq_pie_flow *sel_flow;
+	int uninitialized_var(ret);
+	u8 memory_limited = false;
+	u8 enqueue = false;
+	u32 pkt_len;
+	u32 idx;
+
+	/* Classifies packet into corresponding flow */
+	idx = fq_pie_classify(skb, sch, &ret);
+	sel_flow = &q->flows[idx];
+
+	/* Checks whether adding a new packet would exceed memory limit */
+	get_pie_cb(skb)->mem_usage = skb->truesize;
+	memory_limited = q->memory_usage > q->memory_limit + skb->truesize;
+
+	/* Checks if the qdisc is full */
+	if (unlikely(qdisc_qlen(sch) >= sch->limit)) {
+		q->stats.overlimit++;
+		goto out;
+	} else if (unlikely(memory_limited)) {
+		q->overmemory++;
+	}
+
+	if (!pie_drop_early(sch, &q->p_params, &sel_flow->vars,
+			    sel_flow->backlog, skb->len)) {
+		enqueue = true;
+	} else if (q->p_params.ecn &&
+		   sel_flow->vars.prob <= (MAX_PROB / 100) * q->ecn_prob &&
+		   INET_ECN_set_ce(skb)) {
+		/* If packet is ecn capable, mark it if drop probability
+		 * is lower than the parameter ecn_prob, else drop it.
+		 */
+		q->stats.ecn_mark++;
+		enqueue = true;
+	}
+	if (enqueue) {
+		/* Set enqueue time only when dq_rate_estimator is disabled. */
+		if (!q->p_params.dq_rate_estimator)
+			pie_set_enqueue_time(skb);
+
+		pkt_len = qdisc_pkt_len(skb);
+		q->stats.packets_in++;
+		q->memory_usage += skb->truesize;
+		sch->qstats.backlog += pkt_len;
+		sch->q.qlen++;
+		flow_queue_add(sel_flow, skb);
+		if (list_empty(&sel_flow->flowchain)) {
+			list_add_tail(&sel_flow->flowchain, &q->new_flows);
+			q->new_flow_count++;
+			sel_flow->deficit = q->quantum;
+			sel_flow->qlen = 0;
+			sel_flow->backlog = 0;
+		}
+		sel_flow->qlen++;
+		sel_flow->backlog += pkt_len;
+		return NET_XMIT_SUCCESS;
+	}
+out:
+	q->stats.dropped++;
+	sel_flow->vars.accu_prob = 0;
+	sel_flow->vars.accu_prob_overflows = 0;
+	__qdisc_drop(skb, to_free);
+	qdisc_qstats_drop(sch);
+	return NET_XMIT_CN;
+}
+
+static const struct nla_policy fq_pie_policy[TCA_FQ_PIE_MAX + 1] = {
+	[TCA_FQ_PIE_LIMIT]		= {.type = NLA_U32},
+	[TCA_FQ_PIE_FLOWS]		= {.type = NLA_U32},
+	[TCA_FQ_PIE_TARGET]		= {.type = NLA_U32},
+	[TCA_FQ_PIE_TUPDATE]		= {.type = NLA_U32},
+	[TCA_FQ_PIE_ALPHA]		= {.type = NLA_U32},
+	[TCA_FQ_PIE_BETA]		= {.type = NLA_U32},
+	[TCA_FQ_PIE_QUANTUM]		= {.type = NLA_U32},
+	[TCA_FQ_PIE_MEMORY_LIMIT]	= {.type = NLA_U32},
+	[TCA_FQ_PIE_ECN_PROB]		= {.type = NLA_U32},
+	[TCA_FQ_PIE_ECN]		= {.type = NLA_U32},
+	[TCA_FQ_PIE_BYTEMODE]		= {.type = NLA_U32},
+	[TCA_FQ_PIE_DQ_RATE_ESTIMATOR]	= {.type = NLA_U32},
+};
+
+static inline struct sk_buff *dequeue_head(struct fq_pie_flow *flow)
+{
+	struct sk_buff *skb = flow->head;
+
+	flow->head = skb->next;
+	skb->next = NULL;
+	return skb;
+}
+
+static struct sk_buff *fq_pie_qdisc_dequeue(struct Qdisc *sch)
+{
+	struct fq_pie_sched_data *q = qdisc_priv(sch);
+	struct sk_buff *skb = NULL;
+	struct fq_pie_flow *flow;
+	struct list_head *head;
+	u32 pkt_len;
+
+begin:
+	head = &q->new_flows;
+	if (list_empty(head)) {
+		head = &q->old_flows;
+		if (list_empty(head))
+			return NULL;
+	}
+
+	flow = list_first_entry(head, struct fq_pie_flow, flowchain);
+	/* Flow has exhausted all its credits */
+	if (flow->deficit <= 0) {
+		flow->deficit += q->quantum;
+		list_move_tail(&flow->flowchain, &q->old_flows);
+		goto begin;
+	}
+
+	if (flow->head) {
+		skb = dequeue_head(flow);
+		pkt_len = qdisc_pkt_len(skb);
+		sch->qstats.backlog -= pkt_len;
+		sch->q.qlen--;
+		qdisc_bstats_update(sch, skb);
+	}
+
+	if (!skb) {
+		/* force a pass through old_flows to prevent starvation */
+		if (head == &q->new_flows && !list_empty(&q->old_flows))
+			list_move_tail(&flow->flowchain, &q->old_flows);
+		else
+			list_del_init(&flow->flowchain);
+		goto begin;
+	}
+
+	flow->qlen--;
+	flow->deficit -= pkt_len;
+	flow->backlog -= pkt_len;
+	q->memory_usage -= get_pie_cb(skb)->mem_usage;
+	pie_process_dequeue(skb, &q->p_params, &flow->vars, flow->backlog);
+	return skb;
+}
+
+static int fq_pie_change(struct Qdisc *sch, struct nlattr *opt,
+			 struct netlink_ext_ack *extack)
+{
+	struct fq_pie_sched_data *q = qdisc_priv(sch);
+	struct nlattr *tb[TCA_FQ_PIE_MAX + 1];
+	unsigned int len_dropped = 0;
+	unsigned int num_dropped = 0;
+	int err;
+
+	if (!opt)
+		return -EINVAL;
+
+	err = nla_parse_nested(tb, TCA_FQ_PIE_MAX, opt, fq_pie_policy, extack);
+	if (err < 0)
+		return err;
+
+	sch_tree_lock(sch);
+	if (tb[TCA_FQ_PIE_LIMIT]) {
+		u32 limit = nla_get_u32(tb[TCA_FQ_PIE_LIMIT]);
+
+		q->p_params.limit = limit;
+		sch->limit = limit;
+	}
+	if (tb[TCA_FQ_PIE_FLOWS]) {
+		if (q->flows) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Number of flows cannot be changed");
+			goto flow_error;
+		}
+		q->flows_cnt = nla_get_u32(tb[TCA_FQ_PIE_FLOWS]);
+		if (!q->flows_cnt || q->flows_cnt > 65536) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Number of flows must be < 65536");
+			goto flow_error;
+		}
+	}
+
+	/* convert from microseconds to pschedtime */
+	if (tb[TCA_FQ_PIE_TARGET]) {
+		/* target is in us */
+		u32 target = nla_get_u32(tb[TCA_FQ_PIE_TARGET]);
+
+		/* convert to pschedtime */
+		q->p_params.target =
+			PSCHED_NS2TICKS((u64)target * NSEC_PER_USEC);
+	}
+
+	/* tupdate is in jiffies */
+	if (tb[TCA_FQ_PIE_TUPDATE])
+		q->p_params.tupdate =
+			usecs_to_jiffies(nla_get_u32(tb[TCA_FQ_PIE_TUPDATE]));
+
+	if (tb[TCA_FQ_PIE_ALPHA])
+		q->p_params.alpha = nla_get_u32(tb[TCA_FQ_PIE_ALPHA]);
+
+	if (tb[TCA_FQ_PIE_BETA])
+		q->p_params.beta = nla_get_u32(tb[TCA_FQ_PIE_BETA]);
+
+	if (tb[TCA_FQ_PIE_QUANTUM])
+		q->quantum = nla_get_u32(tb[TCA_FQ_PIE_QUANTUM]);
+
+	if (tb[TCA_FQ_PIE_MEMORY_LIMIT])
+		q->memory_limit = nla_get_u32(tb[TCA_FQ_PIE_MEMORY_LIMIT]);
+
+	if (tb[TCA_FQ_PIE_ECN_PROB])
+		q->ecn_prob = nla_get_u32(tb[TCA_FQ_PIE_ECN_PROB]);
+
+	if (tb[TCA_FQ_PIE_ECN])
+		q->p_params.ecn = nla_get_u32(tb[TCA_FQ_PIE_ECN]);
+
+	if (tb[TCA_FQ_PIE_BYTEMODE])
+		q->p_params.bytemode = nla_get_u32(tb[TCA_FQ_PIE_BYTEMODE]);
+
+	if (tb[TCA_FQ_PIE_DQ_RATE_ESTIMATOR])
+		q->p_params.dq_rate_estimator =
+			nla_get_u32(tb[TCA_FQ_PIE_DQ_RATE_ESTIMATOR]);
+
+	/* Drop excess packets if new limit is lower */
+	while (sch->q.qlen > sch->limit) {
+		struct sk_buff *skb = fq_pie_qdisc_dequeue(sch);
+
+		kfree_skb(skb);
+		len_dropped += qdisc_pkt_len(skb);
+		num_dropped += 1;
+	}
+	qdisc_tree_reduce_backlog(sch, num_dropped, len_dropped);
+
+	sch_tree_unlock(sch);
+	return 0;
+
+flow_error:
+	sch_tree_unlock(sch);
+	return -EINVAL;
+}
+
+static void fq_pie_timer(struct timer_list *t)
+{
+	struct fq_pie_sched_data *q = from_timer(q, t, adapt_timer);
+	struct Qdisc *sch = q->sch;
+	spinlock_t *root_lock; /* to lock qdisc for probability calculations */
+	u16 idx;
+
+	root_lock = qdisc_lock(qdisc_root_sleeping(sch));
+	spin_lock(root_lock);
+
+	for (idx = 0; idx < q->flows_cnt; idx++)
+		pie_calculate_probability(&q->p_params, &q->flows[idx].vars,
+					  q->flows[idx].backlog);
+
+	/* reset the timer to fire after 'tupdate' jiffies. */
+	if (q->p_params.tupdate)
+		mod_timer(&q->adapt_timer, jiffies + q->p_params.tupdate);
+
+	spin_unlock(root_lock);
+}
+
+static int fq_pie_init(struct Qdisc *sch, struct nlattr *opt,
+		       struct netlink_ext_ack *extack)
+{
+	struct fq_pie_sched_data *q = qdisc_priv(sch);
+	int err;
+	u16 idx;
+
+	pie_params_init(&q->p_params);
+	sch->limit = 10 * 1024;
+	q->p_params.limit = sch->limit;
+	q->quantum = psched_mtu(qdisc_dev(sch));
+	q->sch = sch;
+	q->ecn_prob = 10;
+	q->flows_cnt = 1024;
+	q->memory_limit = SZ_32M;
+
+	INIT_LIST_HEAD(&q->new_flows);
+	INIT_LIST_HEAD(&q->old_flows);
+
+	if (opt) {
+		err = fq_pie_change(sch, opt, extack);
+
+		if (err)
+			return err;
+	}
+
+	err = tcf_block_get(&q->block, &q->filter_list, sch, extack);
+	if (err)
+		goto init_failure;
+
+	q->flows = kvcalloc(q->flows_cnt, sizeof(struct fq_pie_flow),
+			    GFP_KERNEL);
+	if (!q->flows) {
+		err = -ENOMEM;
+		goto init_failure;
+	}
+	for (idx = 0; idx < q->flows_cnt; idx++) {
+		struct fq_pie_flow *flow = q->flows + idx;
+
+		INIT_LIST_HEAD(&flow->flowchain);
+		pie_vars_init(&flow->vars);
+	}
+
+	timer_setup(&q->adapt_timer, fq_pie_timer, 0);
+	mod_timer(&q->adapt_timer, jiffies + HZ / 2);
+
+	return 0;
+
+init_failure:
+	q->flows_cnt = 0;
+
+	return err;
+}
+
+static int fq_pie_dump(struct Qdisc *sch, struct sk_buff *skb)
+{
+	struct fq_pie_sched_data *q = qdisc_priv(sch);
+	struct nlattr *opts;
+
+	opts = nla_nest_start(skb, TCA_OPTIONS);
+	if (!opts)
+		return -EMSGSIZE;
+
+	/* convert target from pschedtime to us */
+	if (nla_put_u32(skb, TCA_FQ_PIE_LIMIT, sch->limit) ||
+	    nla_put_u32(skb, TCA_FQ_PIE_FLOWS, q->flows_cnt) ||
+	    nla_put_u32(skb, TCA_FQ_PIE_TARGET,
+			((u32)PSCHED_TICKS2NS(q->p_params.target)) /
+			NSEC_PER_USEC) ||
+	    nla_put_u32(skb, TCA_FQ_PIE_TUPDATE,
+			jiffies_to_usecs(q->p_params.tupdate)) ||
+	    nla_put_u32(skb, TCA_FQ_PIE_ALPHA, q->p_params.alpha) ||
+	    nla_put_u32(skb, TCA_FQ_PIE_BETA, q->p_params.beta) ||
+	    nla_put_u32(skb, TCA_FQ_PIE_QUANTUM, q->quantum) ||
+	    nla_put_u32(skb, TCA_FQ_PIE_MEMORY_LIMIT, q->memory_limit) ||
+	    nla_put_u32(skb, TCA_FQ_PIE_ECN_PROB, q->ecn_prob) ||
+	    nla_put_u32(skb, TCA_FQ_PIE_ECN, q->p_params.ecn) ||
+	    nla_put_u32(skb, TCA_FQ_PIE_BYTEMODE, q->p_params.bytemode) ||
+	    nla_put_u32(skb, TCA_FQ_PIE_DQ_RATE_ESTIMATOR,
+			q->p_params.dq_rate_estimator))
+		goto nla_put_failure;
+
+	return nla_nest_end(skb, opts);
+
+nla_put_failure:
+	nla_nest_cancel(skb, opts);
+	return -EMSGSIZE;
+}
+
+static int fq_pie_dump_stats(struct Qdisc *sch, struct gnet_dump *d)
+{
+	struct fq_pie_sched_data *q = qdisc_priv(sch);
+	struct tc_fq_pie_xstats st = {
+		.packets_in	= q->stats.packets_in,
+		.overlimit	= q->stats.overlimit,
+		.overmemory	= q->overmemory,
+		.dropped	= q->stats.dropped,
+		.ecn_mark	= q->stats.ecn_mark,
+		.new_flow_count = q->new_flow_count,
+		.memory_usage   = q->memory_usage,
+	};
+	struct list_head *pos;
+
+	sch_tree_lock(sch);
+	list_for_each(pos, &q->new_flows)
+		st.new_flows_len++;
+
+	list_for_each(pos, &q->old_flows)
+		st.old_flows_len++;
+	sch_tree_unlock(sch);
+
+	return gnet_stats_copy_app(d, &st, sizeof(st));
+}
+
+static void fq_pie_reset(struct Qdisc *sch)
+{
+	struct fq_pie_sched_data *q = qdisc_priv(sch);
+	u16 idx;
+
+	INIT_LIST_HEAD(&q->new_flows);
+	INIT_LIST_HEAD(&q->old_flows);
+	for (idx = 0; idx < q->flows_cnt; idx++) {
+		struct fq_pie_flow *flow = q->flows + idx;
+
+		/* Removes all packets from flow */
+		rtnl_kfree_skbs(flow->head, flow->tail);
+		flow->head = NULL;
+
+		INIT_LIST_HEAD(&flow->flowchain);
+		pie_vars_init(&flow->vars);
+	}
+
+	sch->q.qlen = 0;
+	sch->qstats.backlog = 0;
+}
+
+static void fq_pie_destroy(struct Qdisc *sch)
+{
+	struct fq_pie_sched_data *q = qdisc_priv(sch);
+
+	tcf_block_put(q->block);
+	del_timer_sync(&q->adapt_timer);
+	kvfree(q->flows);
+}
+
+static struct Qdisc_ops fq_pie_qdisc_ops __read_mostly = {
+	.id		= "fq_pie",
+	.priv_size	= sizeof(struct fq_pie_sched_data),
+	.enqueue	= fq_pie_qdisc_enqueue,
+	.dequeue	= fq_pie_qdisc_dequeue,
+	.peek		= qdisc_peek_dequeued,
+	.init		= fq_pie_init,
+	.destroy	= fq_pie_destroy,
+	.reset		= fq_pie_reset,
+	.change		= fq_pie_change,
+	.dump		= fq_pie_dump,
+	.dump_stats	= fq_pie_dump_stats,
+	.owner		= THIS_MODULE,
+};
+
+static int __init fq_pie_module_init(void)
+{
+	return register_qdisc(&fq_pie_qdisc_ops);
+}
+
+static void __exit fq_pie_module_exit(void)
+{
+	unregister_qdisc(&fq_pie_qdisc_ops);
+}
+
+module_init(fq_pie_module_init);
+module_exit(fq_pie_module_exit);
+
+MODULE_DESCRIPTION("Flow Queue Proportional Integral controller Enhanced (FQ-PIE)");
+MODULE_AUTHOR("Mohit P. Tahiliani");
+MODULE_LICENSE("GPL");
-- 
cgit v1.2.3-59-g8ed1b


From 783da36bb90cbf45be7fa82a0d0d3910f1f6fcc7 Mon Sep 17 00:00:00 2001
From: Dan Murphy <dmurphy@ti.com>
Date: Wed, 22 Jan 2020 09:34:54 -0600
Subject: phy: dp83826: Add phy IDs for DP83826N and 826NC

Add phy IDs to the DP83822 phy driver for the DP83826N
and the DP83826NC devices.  The register map and features
are the same for basic enablement.

Signed-off-by: Dan Murphy <dmurphy@ti.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/Kconfig   | 4 ++--
 drivers/net/phy/dp83822.c | 6 ++++++
 2 files changed, 8 insertions(+), 2 deletions(-)

diff --git a/drivers/net/phy/Kconfig b/drivers/net/phy/Kconfig
index 6b5ee26795a2..cc677ddd2719 100644
--- a/drivers/net/phy/Kconfig
+++ b/drivers/net/phy/Kconfig
@@ -346,9 +346,9 @@ config DAVICOM_PHY
 	  Currently supports dm9161e and dm9131
 
 config DP83822_PHY
-	tristate "Texas Instruments DP83822/825 PHYs"
+	tristate "Texas Instruments DP83822/825/826 PHYs"
 	---help---
-	  Supports the DP83822 and DP83825I PHYs.
+	  Supports the DP83822, DP83825I, DP83826C and DP83826NC PHYs.
 
 config DP83TC811_PHY
 	tristate "Texas Instruments DP83TC811 PHY"
diff --git a/drivers/net/phy/dp83822.c b/drivers/net/phy/dp83822.c
index 8a4b1d167ce2..5159b28baa0f 100644
--- a/drivers/net/phy/dp83822.c
+++ b/drivers/net/phy/dp83822.c
@@ -16,6 +16,8 @@
 
 #define DP83822_PHY_ID	        0x2000a240
 #define DP83825I_PHY_ID		0x2000a150
+#define DP83826C_PHY_ID		0x2000a130
+#define DP83826NC_PHY_ID	0x2000a110
 
 #define DP83822_DEVADDR		0x1f
 
@@ -319,12 +321,16 @@ static int dp83822_resume(struct phy_device *phydev)
 static struct phy_driver dp83822_driver[] = {
 	DP83822_PHY_DRIVER(DP83822_PHY_ID, "TI DP83822"),
 	DP83822_PHY_DRIVER(DP83825I_PHY_ID, "TI DP83825I"),
+	DP83822_PHY_DRIVER(DP83826C_PHY_ID, "TI DP83826C"),
+	DP83822_PHY_DRIVER(DP83826NC_PHY_ID, "TI DP83826NC"),
 };
 module_phy_driver(dp83822_driver);
 
 static struct mdio_device_id __maybe_unused dp83822_tbl[] = {
 	{ DP83822_PHY_ID, 0xfffffff0 },
 	{ DP83825I_PHY_ID, 0xfffffff0 },
+	{ DP83826C_PHY_ID, 0xfffffff0 },
+	{ DP83826NC_PHY_ID, 0xfffffff0 },
 	{ },
 };
 MODULE_DEVICE_TABLE(mdio, dp83822_tbl);
-- 
cgit v1.2.3-59-g8ed1b


From 2ace13e10dbd1c8c5c9f118e5d4d0e0a77b152c7 Mon Sep 17 00:00:00 2001
From: Dan Murphy <dmurphy@ti.com>
Date: Wed, 22 Jan 2020 09:34:55 -0600
Subject: net: phy: DP83822: Add support for additional DP83825 devices

Add PHY IDs for the DP83825CS, DP83825CM and the DP83825S devices to the
DP83822 driver.

Signed-off-by: Dan Murphy <dmurphy@ti.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/Kconfig   |  3 ++-
 drivers/net/phy/dp83822.c | 12 ++++++++++--
 2 files changed, 12 insertions(+), 3 deletions(-)

diff --git a/drivers/net/phy/Kconfig b/drivers/net/phy/Kconfig
index cc677ddd2719..71fc778ce398 100644
--- a/drivers/net/phy/Kconfig
+++ b/drivers/net/phy/Kconfig
@@ -348,7 +348,8 @@ config DAVICOM_PHY
 config DP83822_PHY
 	tristate "Texas Instruments DP83822/825/826 PHYs"
 	---help---
-	  Supports the DP83822, DP83825I, DP83826C and DP83826NC PHYs.
+	  Supports the DP83822, DP83825I, DP83825CM, DP83825CS, DP83825S,
+	  DP83826C and DP83826NC PHYs.
 
 config DP83TC811_PHY
 	tristate "Texas Instruments DP83TC811 PHY"
diff --git a/drivers/net/phy/dp83822.c b/drivers/net/phy/dp83822.c
index 5159b28baa0f..fe9aa3ad52a7 100644
--- a/drivers/net/phy/dp83822.c
+++ b/drivers/net/phy/dp83822.c
@@ -1,6 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-/*
- * Driver for the Texas Instruments DP83822 PHY
+/* Driver for the Texas Instruments DP83822, DP83825 and DP83826 PHYs.
  *
  * Copyright (C) 2017 Texas Instruments Inc.
  */
@@ -15,7 +14,10 @@
 #include <linux/netdevice.h>
 
 #define DP83822_PHY_ID	        0x2000a240
+#define DP83825S_PHY_ID		0x2000a140
 #define DP83825I_PHY_ID		0x2000a150
+#define DP83825CM_PHY_ID	0x2000a160
+#define DP83825CS_PHY_ID	0x2000a170
 #define DP83826C_PHY_ID		0x2000a130
 #define DP83826NC_PHY_ID	0x2000a110
 
@@ -323,6 +325,9 @@ static struct phy_driver dp83822_driver[] = {
 	DP83822_PHY_DRIVER(DP83825I_PHY_ID, "TI DP83825I"),
 	DP83822_PHY_DRIVER(DP83826C_PHY_ID, "TI DP83826C"),
 	DP83822_PHY_DRIVER(DP83826NC_PHY_ID, "TI DP83826NC"),
+	DP83822_PHY_DRIVER(DP83825S_PHY_ID, "TI DP83825S"),
+	DP83822_PHY_DRIVER(DP83825CM_PHY_ID, "TI DP83825M"),
+	DP83822_PHY_DRIVER(DP83825CS_PHY_ID, "TI DP83825CS"),
 };
 module_phy_driver(dp83822_driver);
 
@@ -331,6 +336,9 @@ static struct mdio_device_id __maybe_unused dp83822_tbl[] = {
 	{ DP83825I_PHY_ID, 0xfffffff0 },
 	{ DP83826C_PHY_ID, 0xfffffff0 },
 	{ DP83826NC_PHY_ID, 0xfffffff0 },
+	{ DP83825S_PHY_ID, 0xfffffff0 },
+	{ DP83825CM_PHY_ID, 0xfffffff0 },
+	{ DP83825CS_PHY_ID, 0xfffffff0 },
 	{ },
 };
 MODULE_DEVICE_TABLE(mdio, dp83822_tbl);
-- 
cgit v1.2.3-59-g8ed1b


From 485ec2ea9cf556e9c120e07961b7b459d776a115 Mon Sep 17 00:00:00 2001
From: Amol Grover <frextrite@gmail.com>
Date: Thu, 23 Jan 2020 17:34:38 +0530
Subject: bpf, devmap: Pass lockdep expression to RCU lists
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

head is traversed using hlist_for_each_entry_rcu outside an RCU
read-side critical section but under the protection of dtab->index_lock.

Hence, add corresponding lockdep expression to silence false-positive
lockdep warnings, and harden RCU lists.

Fixes: 6f9d451ab1a3 ("xdp: Add devmap_hash map type for looking up devices by hashed index")
Signed-off-by: Amol Grover <frextrite@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/bpf/20200123120437.26506-1-frextrite@gmail.com
---
 kernel/bpf/devmap.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index de630f980282..f3a44f6ca86e 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -263,7 +263,8 @@ struct bpf_dtab_netdev *__dev_map_hash_lookup_elem(struct bpf_map *map, u32 key)
 	struct hlist_head *head = dev_map_index_hash(dtab, key);
 	struct bpf_dtab_netdev *dev;
 
-	hlist_for_each_entry_rcu(dev, head, index_hlist)
+	hlist_for_each_entry_rcu(dev, head, index_hlist,
+				 lockdep_is_held(&dtab->index_lock))
 		if (dev->idx == key)
 			return dev;
 
-- 
cgit v1.2.3-59-g8ed1b


From 1a6fa10640d97e16d4184fa1c24aa8c3337d4653 Mon Sep 17 00:00:00 2001
From: John Sperbeck <jsperbeck@google.com>
Date: Thu, 23 Jan 2020 15:51:44 -0800
Subject: selftests/bpf: Initialize duration variable before using

The 'duration' variable is referenced in the CHECK() macro, and there are
some uses of the macro before 'duration' is set.  The clang compiler
(validly) complains about this.

Sample error:

.../selftests/bpf/prog_tests/fexit_test.c:23:6: warning: variable 'duration' is uninitialized when used here [-Wuninitialized]
        if (CHECK(err, "prog_load sched cls", "err %d errno %d\n", err, errno))
            ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.../selftests/bpf/test_progs.h:134:25: note: expanded from macro 'CHECK'
        if (CHECK(err, "prog_load sched cls", "err %d errno %d\n", err, errno))
            ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
        _CHECK(condition, tag, duration, format)
                               ^~~~~~~~

Signed-off-by: John Sperbeck <jsperbeck@google.com>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20200123235144.93610-1-sdf@google.com
---
 tools/testing/selftests/bpf/prog_tests/fentry_test.c   | 2 +-
 tools/testing/selftests/bpf/prog_tests/fexit_bpf2bpf.c | 2 +-
 tools/testing/selftests/bpf/prog_tests/fexit_test.c    | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/fentry_test.c b/tools/testing/selftests/bpf/prog_tests/fentry_test.c
index e1a379f5f7d2..5cc06021f27d 100644
--- a/tools/testing/selftests/bpf/prog_tests/fentry_test.c
+++ b/tools/testing/selftests/bpf/prog_tests/fentry_test.c
@@ -9,7 +9,7 @@ void test_fentry_test(void)
 	struct test_pkt_access *pkt_skel = NULL;
 	struct fentry_test *fentry_skel = NULL;
 	int err, pkt_fd, i;
-	__u32 duration, retval;
+	__u32 duration = 0, retval;
 	__u64 *result;
 
 	pkt_skel = test_pkt_access__open_and_load();
diff --git a/tools/testing/selftests/bpf/prog_tests/fexit_bpf2bpf.c b/tools/testing/selftests/bpf/prog_tests/fexit_bpf2bpf.c
index db5c74d2ce6d..cde463af7071 100644
--- a/tools/testing/selftests/bpf/prog_tests/fexit_bpf2bpf.c
+++ b/tools/testing/selftests/bpf/prog_tests/fexit_bpf2bpf.c
@@ -11,7 +11,7 @@ static void test_fexit_bpf2bpf_common(const char *obj_file,
 	int err, pkt_fd, i;
 	struct bpf_link **link = NULL;
 	struct bpf_program **prog = NULL;
-	__u32 duration, retval;
+	__u32 duration = 0, retval;
 	struct bpf_map *data_map;
 	const int zero = 0;
 	u64 *result = NULL;
diff --git a/tools/testing/selftests/bpf/prog_tests/fexit_test.c b/tools/testing/selftests/bpf/prog_tests/fexit_test.c
index f99013222c74..d2c3655dd7a3 100644
--- a/tools/testing/selftests/bpf/prog_tests/fexit_test.c
+++ b/tools/testing/selftests/bpf/prog_tests/fexit_test.c
@@ -13,7 +13,7 @@ void test_fexit_test(void)
 	int err, pkt_fd, kfree_skb_fd, i;
 	struct bpf_link *link[6] = {};
 	struct bpf_program *prog[6];
-	__u32 duration, retval;
+	__u32 duration = 0, retval;
 	struct bpf_map *data_map;
 	const int zero = 0;
 	u64 result[6];
-- 
cgit v1.2.3-59-g8ed1b


From 03506297d205e5fa25b5ead0f6338b5a4a996a93 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Thu, 23 Jan 2020 21:41:48 -0800
Subject: selftests/bpf: Improve bpftool changes detection
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Detect when bpftool source code changes and trigger rebuild within
selftests/bpf Makefile. Also fix few small formatting problems.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/bpf/20200124054148.2455060-1-andriin@fb.com
---
 tools/testing/selftests/bpf/Makefile | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index 5f41f5bd8033..257a1aaaa37d 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -155,16 +155,17 @@ $(OUTPUT)/test_sysctl: cgroup_helpers.c
 
 DEFAULT_BPFTOOL := $(SCRATCH_DIR)/sbin/bpftool
 BPFTOOL ?= $(DEFAULT_BPFTOOL)
-$(DEFAULT_BPFTOOL): $(BPFOBJ) | $(BUILD_DIR)/bpftool
-	$(Q)$(MAKE) $(submake_extras)  -C $(BPFTOOLDIR)			\
-		    OUTPUT=$(BUILD_DIR)/bpftool/			\
+$(DEFAULT_BPFTOOL): $(wildcard $(BPFTOOLDIR)/*.[ch] $(BPFTOOLDIR)/Makefile)    \
+		    $(BPFOBJ) | $(BUILD_DIR)/bpftool
+	$(Q)$(MAKE) $(submake_extras)  -C $(BPFTOOLDIR)			       \
+		    OUTPUT=$(BUILD_DIR)/bpftool/			       \
 		    prefix= DESTDIR=$(SCRATCH_DIR)/ install
 
-$(BPFOBJ): $(wildcard $(BPFDIR)/*.c $(BPFDIR)/*.h $(BPFDIR)/Makefile)          \
+$(BPFOBJ): $(wildcard $(BPFDIR)/*.[ch] $(BPFDIR)/Makefile)		       \
 	   ../../../include/uapi/linux/bpf.h                                   \
 	   | $(INCLUDE_DIR) $(BUILD_DIR)/libbpf
 	$(Q)$(MAKE) $(submake_extras) -C $(BPFDIR) OUTPUT=$(BUILD_DIR)/libbpf/ \
-		DESTDIR=$(SCRATCH_DIR) prefix= all install_headers
+		    DESTDIR=$(SCRATCH_DIR) prefix= all install_headers
 
 $(BUILD_DIR)/libbpf $(BUILD_DIR)/bpftool $(INCLUDE_DIR):
 	$(call msg,MKDIR,,$@)
-- 
cgit v1.2.3-59-g8ed1b


From 41258289a8e9e3e110bab316e0aeded25fa8beb6 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Thu, 23 Jan 2020 21:43:17 -0800
Subject: bpftool: Print function linkage in BTF dump
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add printing out BTF_KIND_FUNC's linkage.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/bpf/20200124054317.2459436-1-andriin@fb.com
---
 tools/bpf/bpftool/btf.c | 27 +++++++++++++++++++++++----
 1 file changed, 23 insertions(+), 4 deletions(-)

diff --git a/tools/bpf/bpftool/btf.c b/tools/bpf/bpftool/btf.c
index 4ba90d81b6a1..b3745ed711ba 100644
--- a/tools/bpf/bpftool/btf.c
+++ b/tools/bpf/bpftool/btf.c
@@ -77,6 +77,20 @@ static const char *btf_var_linkage_str(__u32 linkage)
 	}
 }
 
+static const char *btf_func_linkage_str(const struct btf_type *t)
+{
+	switch (btf_vlen(t)) {
+	case BTF_FUNC_STATIC:
+		return "static";
+	case BTF_FUNC_GLOBAL:
+		return "global";
+	case BTF_FUNC_EXTERN:
+		return "extern";
+	default:
+		return "(unknown)";
+	}
+}
+
 static const char *btf_str(const struct btf *btf, __u32 off)
 {
 	if (!off)
@@ -231,12 +245,17 @@ static int dump_btf_type(const struct btf *btf, __u32 id,
 			printf(" fwd_kind=%s", fwd_kind);
 		break;
 	}
-	case BTF_KIND_FUNC:
-		if (json_output)
+	case BTF_KIND_FUNC: {
+		const char *linkage = btf_func_linkage_str(t);
+
+		if (json_output) {
 			jsonw_uint_field(w, "type_id", t->type);
-		else
-			printf(" type_id=%u", t->type);
+			jsonw_string_field(w, "linkage", linkage);
+		} else {
+			printf(" type_id=%u linkage=%s", t->type, linkage);
+		}
 		break;
+	}
 	case BTF_KIND_FUNC_PROTO: {
 		const struct btf_param *p = (const void *)(t + 1);
 		__u16 vlen = BTF_INFO_VLEN(t->info);
-- 
cgit v1.2.3-59-g8ed1b


From ac0e932d0e2988c125be251715d83e95839cdae3 Mon Sep 17 00:00:00 2001
From: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Date: Fri, 24 Jan 2020 13:40:19 +0200
Subject: net: bridge: check port state before br_allowed_egress

If we make sure that br_allowed_egress is called only when we have
BR_STATE_FORWARDING state then we can avoid a test later when we add
per-vlan state.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/bridge/br_forward.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c
index 86637000f275..7629b63f6f30 100644
--- a/net/bridge/br_forward.c
+++ b/net/bridge/br_forward.c
@@ -25,7 +25,7 @@ static inline int should_deliver(const struct net_bridge_port *p,
 
 	vg = nbp_vlan_group_rcu(p);
 	return ((p->flags & BR_HAIRPIN_MODE) || skb->dev != p->dev) &&
-		br_allowed_egress(vg, skb) && p->state == BR_STATE_FORWARDING &&
+		p->state == BR_STATE_FORWARDING && br_allowed_egress(vg, skb) &&
 		nbp_switchdev_allowed_egress(p, skb) &&
 		!br_skb_isolated(p, skb);
 }
-- 
cgit v1.2.3-59-g8ed1b


From 7a53e718c551752924416b33235bf338729b5115 Mon Sep 17 00:00:00 2001
From: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Date: Fri, 24 Jan 2020 13:40:20 +0200
Subject: net: bridge: vlan: add basic option dumping support

We'll be dumping the options for the whole range if they're equal. The
first range vlan will be used to extract the options. The commit doesn't
change anything yet it just adds the skeleton for the support. The dump
will happen when the first option is added.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/bridge/Makefile          |  2 +-
 net/bridge/br_private.h      |  8 ++++++++
 net/bridge/br_vlan.c         | 20 ++++++++++++++------
 net/bridge/br_vlan_options.c | 25 +++++++++++++++++++++++++
 4 files changed, 48 insertions(+), 7 deletions(-)
 create mode 100644 net/bridge/br_vlan_options.c

diff --git a/net/bridge/Makefile b/net/bridge/Makefile
index ac9ef337f0fa..49da7ae6f077 100644
--- a/net/bridge/Makefile
+++ b/net/bridge/Makefile
@@ -20,7 +20,7 @@ obj-$(CONFIG_BRIDGE_NETFILTER) += br_netfilter.o
 
 bridge-$(CONFIG_BRIDGE_IGMP_SNOOPING) += br_multicast.o br_mdb.o
 
-bridge-$(CONFIG_BRIDGE_VLAN_FILTERING) += br_vlan.o br_vlan_tunnel.o
+bridge-$(CONFIG_BRIDGE_VLAN_FILTERING) += br_vlan.o br_vlan_tunnel.o br_vlan_options.o
 
 bridge-$(CONFIG_NET_SWITCHDEV) += br_switchdev.o
 
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index a6226ff2f0cc..403df71d2cfa 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -1191,6 +1191,14 @@ static inline void br_vlan_notify(const struct net_bridge *br,
 }
 #endif
 
+/* br_vlan_options.c */
+#ifdef CONFIG_BRIDGE_VLAN_FILTERING
+bool br_vlan_opts_eq(const struct net_bridge_vlan *v1,
+		     const struct net_bridge_vlan *v2);
+bool br_vlan_opts_fill(struct sk_buff *skb, const struct net_bridge_vlan *v);
+size_t br_vlan_opts_nl_size(void);
+#endif
+
 struct nf_br_ops {
 	int (*br_dev_xmit_hook)(struct sk_buff *skb);
 };
diff --git a/net/bridge/br_vlan.c b/net/bridge/br_vlan.c
index e4f7dd10c3f8..75ec3da92b0b 100644
--- a/net/bridge/br_vlan.c
+++ b/net/bridge/br_vlan.c
@@ -1547,7 +1547,9 @@ void br_vlan_port_event(struct net_bridge_port *p, unsigned long event)
 	}
 }
 
+/* v_opts is used to dump the options which must be equal in the whole range */
 static bool br_vlan_fill_vids(struct sk_buff *skb, u16 vid, u16 vid_range,
+			      const struct net_bridge_vlan *v_opts,
 			      u16 flags)
 {
 	struct bridge_vlan_info info;
@@ -1572,6 +1574,9 @@ static bool br_vlan_fill_vids(struct sk_buff *skb, u16 vid, u16 vid_range,
 	    nla_put_u16(skb, BRIDGE_VLANDB_ENTRY_RANGE, vid_range))
 		goto out_err;
 
+	if (v_opts && !br_vlan_opts_fill(skb, v_opts))
+		goto out_err;
+
 	nla_nest_end(skb, nest);
 
 	return true;
@@ -1586,7 +1591,8 @@ static size_t rtnl_vlan_nlmsg_size(void)
 	return NLMSG_ALIGN(sizeof(struct br_vlan_msg))
 		+ nla_total_size(0) /* BRIDGE_VLANDB_ENTRY */
 		+ nla_total_size(sizeof(u16)) /* BRIDGE_VLANDB_ENTRY_RANGE */
-		+ nla_total_size(sizeof(struct bridge_vlan_info)); /* BRIDGE_VLANDB_ENTRY_INFO */
+		+ nla_total_size(sizeof(struct bridge_vlan_info)) /* BRIDGE_VLANDB_ENTRY_INFO */
+		+ br_vlan_opts_nl_size(); /* bridge vlan options */
 }
 
 void br_vlan_notify(const struct net_bridge *br,
@@ -1595,7 +1601,7 @@ void br_vlan_notify(const struct net_bridge *br,
 		    int cmd)
 {
 	struct net_bridge_vlan_group *vg;
-	struct net_bridge_vlan *v;
+	struct net_bridge_vlan *v = NULL;
 	struct br_vlan_msg *bvm;
 	struct nlmsghdr *nlh;
 	struct sk_buff *skb;
@@ -1647,7 +1653,7 @@ void br_vlan_notify(const struct net_bridge *br,
 		goto out_kfree;
 	}
 
-	if (!br_vlan_fill_vids(skb, vid, vid_range, flags))
+	if (!br_vlan_fill_vids(skb, vid, vid_range, v, flags))
 		goto out_err;
 
 	nlmsg_end(skb, nlh);
@@ -1665,7 +1671,8 @@ static bool br_vlan_can_enter_range(const struct net_bridge_vlan *v_curr,
 				    const struct net_bridge_vlan *range_end)
 {
 	return v_curr->vid - range_end->vid == 1 &&
-	       range_end->flags == v_curr->flags;
+	       range_end->flags == v_curr->flags &&
+	       br_vlan_opts_eq(v_curr, range_end);
 }
 
 static int br_vlan_dump_dev(const struct net_device *dev,
@@ -1729,7 +1736,8 @@ static int br_vlan_dump_dev(const struct net_device *dev,
 			u16 flags = br_vlan_flags(range_start, pvid);
 
 			if (!br_vlan_fill_vids(skb, range_start->vid,
-					       range_end->vid, flags)) {
+					       range_end->vid, range_start,
+					       flags)) {
 				err = -EMSGSIZE;
 				break;
 			}
@@ -1748,7 +1756,7 @@ static int br_vlan_dump_dev(const struct net_device *dev,
 	 */
 	if (!err && range_start &&
 	    !br_vlan_fill_vids(skb, range_start->vid, range_end->vid,
-			       br_vlan_flags(range_start, pvid)))
+			       range_start, br_vlan_flags(range_start, pvid)))
 		err = -EMSGSIZE;
 
 	cb->args[1] = err ? idx : 0;
diff --git a/net/bridge/br_vlan_options.c b/net/bridge/br_vlan_options.c
new file mode 100644
index 000000000000..55fcdc9c380c
--- /dev/null
+++ b/net/bridge/br_vlan_options.c
@@ -0,0 +1,25 @@
+// SPDX-License-Identifier: GPL-2.0-only
+// Copyright (c) 2020, Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
+#include <linux/kernel.h>
+#include <linux/netdevice.h>
+#include <linux/rtnetlink.h>
+#include <linux/slab.h>
+
+#include "br_private.h"
+
+/* check if the options between two vlans are equal */
+bool br_vlan_opts_eq(const struct net_bridge_vlan *v1,
+		     const struct net_bridge_vlan *v2)
+{
+	return true;
+}
+
+bool br_vlan_opts_fill(struct sk_buff *skb, const struct net_bridge_vlan *v)
+{
+	return true;
+}
+
+size_t br_vlan_opts_nl_size(void)
+{
+	return 0;
+}
-- 
cgit v1.2.3-59-g8ed1b


From a5d29ae226812ae620f58a0726525d8af9b1d751 Mon Sep 17 00:00:00 2001
From: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Date: Fri, 24 Jan 2020 13:40:21 +0200
Subject: net: bridge: vlan: add basic option setting support

This patch adds support for option modification of single vlans and
ranges. It allows to only modify options, i.e. skip create/delete by
using the BRIDGE_VLAN_INFO_ONLY_OPTS flag. When working with a range
option changes we try to pack the notifications as much as possible.

v2: do full port (all vlans) notification only when creating/deleting
    vlans for compatibility, rework the range detection when changing
    options, add more verbose extack errors and check if a vlan should
    be used (br_vlan_should_use checks)

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/uapi/linux/if_bridge.h |  1 +
 net/bridge/br_private.h        |  8 ++++
 net/bridge/br_vlan.c           | 41 ++++++++++++++++----
 net/bridge/br_vlan_options.c   | 87 ++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 130 insertions(+), 7 deletions(-)

diff --git a/include/uapi/linux/if_bridge.h b/include/uapi/linux/if_bridge.h
index ac38f0b674b8..c8b87ad52c5b 100644
--- a/include/uapi/linux/if_bridge.h
+++ b/include/uapi/linux/if_bridge.h
@@ -130,6 +130,7 @@ enum {
 #define BRIDGE_VLAN_INFO_RANGE_BEGIN	(1<<3) /* VLAN is start of vlan range */
 #define BRIDGE_VLAN_INFO_RANGE_END	(1<<4) /* VLAN is end of vlan range */
 #define BRIDGE_VLAN_INFO_BRENTRY	(1<<5) /* Global bridge VLAN entry */
+#define BRIDGE_VLAN_INFO_ONLY_OPTS	(1<<6) /* Skip create/delete/flags */
 
 struct bridge_vlan_info {
 	__u16 flags;
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index 403df71d2cfa..084904ee22a8 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -976,6 +976,8 @@ void br_vlan_notify(const struct net_bridge *br,
 		    const struct net_bridge_port *p,
 		    u16 vid, u16 vid_range,
 		    int cmd);
+bool br_vlan_can_enter_range(const struct net_bridge_vlan *v_curr,
+			     const struct net_bridge_vlan *range_end);
 
 static inline struct net_bridge_vlan_group *br_vlan_group(
 					const struct net_bridge *br)
@@ -1197,6 +1199,12 @@ bool br_vlan_opts_eq(const struct net_bridge_vlan *v1,
 		     const struct net_bridge_vlan *v2);
 bool br_vlan_opts_fill(struct sk_buff *skb, const struct net_bridge_vlan *v);
 size_t br_vlan_opts_nl_size(void);
+int br_vlan_process_options(const struct net_bridge *br,
+			    const struct net_bridge_port *p,
+			    struct net_bridge_vlan *range_start,
+			    struct net_bridge_vlan *range_end,
+			    struct nlattr **tb,
+			    struct netlink_ext_ack *extack);
 #endif
 
 struct nf_br_ops {
diff --git a/net/bridge/br_vlan.c b/net/bridge/br_vlan.c
index 75ec3da92b0b..d747756eac63 100644
--- a/net/bridge/br_vlan.c
+++ b/net/bridge/br_vlan.c
@@ -1667,8 +1667,8 @@ out_kfree:
 }
 
 /* check if v_curr can enter a range ending in range_end */
-static bool br_vlan_can_enter_range(const struct net_bridge_vlan *v_curr,
-				    const struct net_bridge_vlan *range_end)
+bool br_vlan_can_enter_range(const struct net_bridge_vlan *v_curr,
+			     const struct net_bridge_vlan *range_end)
 {
 	return v_curr->vid - range_end->vid == 1 &&
 	       range_end->flags == v_curr->flags &&
@@ -1824,11 +1824,11 @@ static int br_vlan_rtm_process_one(struct net_device *dev,
 {
 	struct bridge_vlan_info *vinfo, vrange_end, *vinfo_last = NULL;
 	struct nlattr *tb[BRIDGE_VLANDB_ENTRY_MAX + 1];
+	bool changed = false, skip_processing = false;
 	struct net_bridge_vlan_group *vg;
 	struct net_bridge_port *p = NULL;
 	int err = 0, cmdmap = 0;
 	struct net_bridge *br;
-	bool changed = false;
 
 	if (netif_is_bridge_master(dev)) {
 		br = netdev_priv(dev);
@@ -1882,16 +1882,43 @@ static int br_vlan_rtm_process_one(struct net_device *dev,
 	switch (cmd) {
 	case RTM_NEWVLAN:
 		cmdmap = RTM_SETLINK;
+		skip_processing = !!(vinfo->flags & BRIDGE_VLAN_INFO_ONLY_OPTS);
 		break;
 	case RTM_DELVLAN:
 		cmdmap = RTM_DELLINK;
 		break;
 	}
 
-	err = br_process_vlan_info(br, p, cmdmap, vinfo, &vinfo_last, &changed,
-				   extack);
-	if (changed)
-		br_ifinfo_notify(cmdmap, br, p);
+	if (!skip_processing) {
+		struct bridge_vlan_info *tmp_last = vinfo_last;
+
+		/* br_process_vlan_info may overwrite vinfo_last */
+		err = br_process_vlan_info(br, p, cmdmap, vinfo, &tmp_last,
+					   &changed, extack);
+
+		/* notify first if anything changed */
+		if (changed)
+			br_ifinfo_notify(cmdmap, br, p);
+
+		if (err)
+			return err;
+	}
+
+	/* deal with options */
+	if (cmd == RTM_NEWVLAN) {
+		struct net_bridge_vlan *range_start, *range_end;
+
+		if (vinfo_last) {
+			range_start = br_vlan_find(vg, vinfo_last->vid);
+			range_end = br_vlan_find(vg, vinfo->vid);
+		} else {
+			range_start = br_vlan_find(vg, vinfo->vid);
+			range_end = range_start;
+		}
+
+		err = br_vlan_process_options(br, p, range_start, range_end,
+					      tb, extack);
+	}
 
 	return err;
 }
diff --git a/net/bridge/br_vlan_options.c b/net/bridge/br_vlan_options.c
index 55fcdc9c380c..27275ac3e42e 100644
--- a/net/bridge/br_vlan_options.c
+++ b/net/bridge/br_vlan_options.c
@@ -23,3 +23,90 @@ size_t br_vlan_opts_nl_size(void)
 {
 	return 0;
 }
+
+static int br_vlan_process_one_opts(const struct net_bridge *br,
+				    const struct net_bridge_port *p,
+				    struct net_bridge_vlan_group *vg,
+				    struct net_bridge_vlan *v,
+				    struct nlattr **tb,
+				    bool *changed,
+				    struct netlink_ext_ack *extack)
+{
+	*changed = false;
+	return 0;
+}
+
+int br_vlan_process_options(const struct net_bridge *br,
+			    const struct net_bridge_port *p,
+			    struct net_bridge_vlan *range_start,
+			    struct net_bridge_vlan *range_end,
+			    struct nlattr **tb,
+			    struct netlink_ext_ack *extack)
+{
+	struct net_bridge_vlan *v, *curr_start = NULL, *curr_end = NULL;
+	struct net_bridge_vlan_group *vg;
+	int vid, err = 0;
+	u16 pvid;
+
+	if (p)
+		vg = nbp_vlan_group(p);
+	else
+		vg = br_vlan_group(br);
+
+	if (!range_start || !br_vlan_should_use(range_start)) {
+		NL_SET_ERR_MSG_MOD(extack, "Vlan range start doesn't exist, can't process options");
+		return -ENOENT;
+	}
+	if (!range_end || !br_vlan_should_use(range_end)) {
+		NL_SET_ERR_MSG_MOD(extack, "Vlan range end doesn't exist, can't process options");
+		return -ENOENT;
+	}
+
+	pvid = br_get_pvid(vg);
+	for (vid = range_start->vid; vid <= range_end->vid; vid++) {
+		bool changed = false;
+
+		v = br_vlan_find(vg, vid);
+		if (!v || !br_vlan_should_use(v)) {
+			NL_SET_ERR_MSG_MOD(extack, "Vlan in range doesn't exist, can't process options");
+			err = -ENOENT;
+			break;
+		}
+
+		err = br_vlan_process_one_opts(br, p, vg, v, tb, &changed,
+					       extack);
+		if (err)
+			break;
+
+		if (changed) {
+			/* vlan options changed, check for range */
+			if (!curr_start) {
+				curr_start = v;
+				curr_end = v;
+				continue;
+			}
+
+			if (v->vid == pvid ||
+			    !br_vlan_can_enter_range(v, curr_end)) {
+				br_vlan_notify(br, p, curr_start->vid,
+					       curr_end->vid, RTM_NEWVLAN);
+				curr_start = v;
+			}
+			curr_end = v;
+		} else {
+			/* nothing changed and nothing to notify yet */
+			if (!curr_start)
+				continue;
+
+			br_vlan_notify(br, p, curr_start->vid, curr_end->vid,
+				       RTM_NEWVLAN);
+			curr_start = NULL;
+			curr_end = NULL;
+		}
+	}
+	if (curr_start)
+		br_vlan_notify(br, p, curr_start->vid, curr_end->vid,
+			       RTM_NEWVLAN);
+
+	return err;
+}
-- 
cgit v1.2.3-59-g8ed1b


From a580c76d534c7360ba68042b19cb255e8420e987 Mon Sep 17 00:00:00 2001
From: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Date: Fri, 24 Jan 2020 13:40:22 +0200
Subject: net: bridge: vlan: add per-vlan state

The first per-vlan option added is state, it is needed for EVPN and for
per-vlan STP. The state allows to control the forwarding on per-vlan
basis. The vlan state is considered only if the port state is forwarding
in order to avoid conflicts and be consistent. br_allowed_egress is
called only when the state is forwarding, but the ingress case is a bit
more complicated due to the fact that we may have the transition between
port:BR_STATE_FORWARDING -> vlan:BR_STATE_LEARNING which should still
allow the bridge to learn from the packet after vlan filtering and it will
be dropped after that. Also to optimize the pvid state check we keep a
copy in the vlan group to avoid one lookup. The state members are
modified with *_ONCE() to annotate the lockless access.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/uapi/linux/if_bridge.h |  1 +
 net/bridge/br_device.c         |  3 ++-
 net/bridge/br_input.c          |  7 ++++--
 net/bridge/br_private.h        | 43 ++++++++++++++++++++++++++++++++--
 net/bridge/br_vlan.c           | 47 ++++++++++++++++++++++++++++----------
 net/bridge/br_vlan_options.c   | 52 ++++++++++++++++++++++++++++++++++++++++--
 6 files changed, 134 insertions(+), 19 deletions(-)

diff --git a/include/uapi/linux/if_bridge.h b/include/uapi/linux/if_bridge.h
index c8b87ad52c5b..42f7ca38ad80 100644
--- a/include/uapi/linux/if_bridge.h
+++ b/include/uapi/linux/if_bridge.h
@@ -191,6 +191,7 @@ enum {
 	BRIDGE_VLANDB_ENTRY_UNSPEC,
 	BRIDGE_VLANDB_ENTRY_INFO,
 	BRIDGE_VLANDB_ENTRY_RANGE,
+	BRIDGE_VLANDB_ENTRY_STATE,
 	__BRIDGE_VLANDB_ENTRY_MAX,
 };
 #define BRIDGE_VLANDB_ENTRY_MAX (__BRIDGE_VLANDB_ENTRY_MAX - 1)
diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c
index fb38add21b37..dc3d2c1dd9d5 100644
--- a/net/bridge/br_device.c
+++ b/net/bridge/br_device.c
@@ -32,6 +32,7 @@ netdev_tx_t br_dev_xmit(struct sk_buff *skb, struct net_device *dev)
 	struct net_bridge_mdb_entry *mdst;
 	struct pcpu_sw_netstats *brstats = this_cpu_ptr(br->stats);
 	const struct nf_br_ops *nf_ops;
+	u8 state = BR_STATE_FORWARDING;
 	const unsigned char *dest;
 	struct ethhdr *eth;
 	u16 vid = 0;
@@ -56,7 +57,7 @@ netdev_tx_t br_dev_xmit(struct sk_buff *skb, struct net_device *dev)
 	eth = eth_hdr(skb);
 	skb_pull(skb, ETH_HLEN);
 
-	if (!br_allowed_ingress(br, br_vlan_group_rcu(br), skb, &vid))
+	if (!br_allowed_ingress(br, br_vlan_group_rcu(br), skb, &vid, &state))
 		goto out;
 
 	if (IS_ENABLED(CONFIG_INET) &&
diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c
index 8944ceb47fe9..fcc260840028 100644
--- a/net/bridge/br_input.c
+++ b/net/bridge/br_input.c
@@ -76,11 +76,14 @@ int br_handle_frame_finish(struct net *net, struct sock *sk, struct sk_buff *skb
 	bool local_rcv, mcast_hit = false;
 	struct net_bridge *br;
 	u16 vid = 0;
+	u8 state;
 
 	if (!p || p->state == BR_STATE_DISABLED)
 		goto drop;
 
-	if (!br_allowed_ingress(p->br, nbp_vlan_group_rcu(p), skb, &vid))
+	state = p->state;
+	if (!br_allowed_ingress(p->br, nbp_vlan_group_rcu(p), skb, &vid,
+				&state))
 		goto out;
 
 	nbp_switchdev_frame_mark(p, skb);
@@ -103,7 +106,7 @@ int br_handle_frame_finish(struct net *net, struct sock *sk, struct sk_buff *skb
 		}
 	}
 
-	if (p->state == BR_STATE_LEARNING)
+	if (state == BR_STATE_LEARNING)
 		goto drop;
 
 	BR_INPUT_SKB_CB(skb)->brdev = br->dev;
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index 084904ee22a8..5153ffe79a01 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -113,6 +113,7 @@ enum {
  * @vid: VLAN id
  * @flags: bridge vlan flags
  * @priv_flags: private (in-kernel) bridge vlan flags
+ * @state: STP state (e.g. blocking, learning, forwarding)
  * @stats: per-cpu VLAN statistics
  * @br: if MASTER flag set, this points to a bridge struct
  * @port: if MASTER flag unset, this points to a port struct
@@ -133,6 +134,7 @@ struct net_bridge_vlan {
 	u16				vid;
 	u16				flags;
 	u16				priv_flags;
+	u8				state;
 	struct br_vlan_stats __percpu	*stats;
 	union {
 		struct net_bridge	*br;
@@ -157,6 +159,7 @@ struct net_bridge_vlan {
  * @vlan_list: sorted VLAN entry list
  * @num_vlans: number of total VLAN entries
  * @pvid: PVID VLAN id
+ * @pvid_state: PVID's STP state (e.g. forwarding, learning, blocking)
  *
  * IMPORTANT: Be careful when checking if there're VLAN entries using list
  *            primitives because the bridge can have entries in its list which
@@ -170,6 +173,7 @@ struct net_bridge_vlan_group {
 	struct list_head		vlan_list;
 	u16				num_vlans;
 	u16				pvid;
+	u8				pvid_state;
 };
 
 /* bridge fdb flags */
@@ -935,7 +939,7 @@ static inline int br_multicast_igmp_type(const struct sk_buff *skb)
 #ifdef CONFIG_BRIDGE_VLAN_FILTERING
 bool br_allowed_ingress(const struct net_bridge *br,
 			struct net_bridge_vlan_group *vg, struct sk_buff *skb,
-			u16 *vid);
+			u16 *vid, u8 *state);
 bool br_allowed_egress(struct net_bridge_vlan_group *vg,
 		       const struct sk_buff *skb);
 bool br_should_learn(struct net_bridge_port *p, struct sk_buff *skb, u16 *vid);
@@ -1037,7 +1041,7 @@ static inline u16 br_vlan_flags(const struct net_bridge_vlan *v, u16 pvid)
 static inline bool br_allowed_ingress(const struct net_bridge *br,
 				      struct net_bridge_vlan_group *vg,
 				      struct sk_buff *skb,
-				      u16 *vid)
+				      u16 *vid, u8 *state)
 {
 	return true;
 }
@@ -1205,6 +1209,41 @@ int br_vlan_process_options(const struct net_bridge *br,
 			    struct net_bridge_vlan *range_end,
 			    struct nlattr **tb,
 			    struct netlink_ext_ack *extack);
+
+/* vlan state manipulation helpers using *_ONCE to annotate lock-free access */
+static inline u8 br_vlan_get_state(const struct net_bridge_vlan *v)
+{
+	return READ_ONCE(v->state);
+}
+
+static inline void br_vlan_set_state(struct net_bridge_vlan *v, u8 state)
+{
+	WRITE_ONCE(v->state, state);
+}
+
+static inline u8 br_vlan_get_pvid_state(const struct net_bridge_vlan_group *vg)
+{
+	return READ_ONCE(vg->pvid_state);
+}
+
+static inline void br_vlan_set_pvid_state(struct net_bridge_vlan_group *vg,
+					  u8 state)
+{
+	WRITE_ONCE(vg->pvid_state, state);
+}
+
+/* learn_allow is true at ingress and false at egress */
+static inline bool br_vlan_state_allowed(u8 state, bool learn_allow)
+{
+	switch (state) {
+	case BR_STATE_LEARNING:
+		return learn_allow;
+	case BR_STATE_FORWARDING:
+		return true;
+	default:
+		return false;
+	}
+}
 #endif
 
 struct nf_br_ops {
diff --git a/net/bridge/br_vlan.c b/net/bridge/br_vlan.c
index d747756eac63..6b5deca08b89 100644
--- a/net/bridge/br_vlan.c
+++ b/net/bridge/br_vlan.c
@@ -34,13 +34,15 @@ static struct net_bridge_vlan *br_vlan_lookup(struct rhashtable *tbl, u16 vid)
 	return rhashtable_lookup_fast(tbl, &vid, br_vlan_rht_params);
 }
 
-static bool __vlan_add_pvid(struct net_bridge_vlan_group *vg, u16 vid)
+static bool __vlan_add_pvid(struct net_bridge_vlan_group *vg,
+			    const struct net_bridge_vlan *v)
 {
-	if (vg->pvid == vid)
+	if (vg->pvid == v->vid)
 		return false;
 
 	smp_wmb();
-	vg->pvid = vid;
+	br_vlan_set_pvid_state(vg, v->state);
+	vg->pvid = v->vid;
 
 	return true;
 }
@@ -69,7 +71,7 @@ static bool __vlan_add_flags(struct net_bridge_vlan *v, u16 flags)
 		vg = nbp_vlan_group(v->port);
 
 	if (flags & BRIDGE_VLAN_INFO_PVID)
-		ret = __vlan_add_pvid(vg, v->vid);
+		ret = __vlan_add_pvid(vg, v);
 	else
 		ret = __vlan_delete_pvid(vg, v->vid);
 
@@ -293,6 +295,9 @@ static int __vlan_add(struct net_bridge_vlan *v, u16 flags,
 		vg->num_vlans++;
 	}
 
+	/* set the state before publishing */
+	v->state = BR_STATE_FORWARDING;
+
 	err = rhashtable_lookup_insert_fast(&vg->vlan_hash, &v->vnode,
 					    br_vlan_rht_params);
 	if (err)
@@ -466,7 +471,8 @@ out:
 /* Called under RCU */
 static bool __allowed_ingress(const struct net_bridge *br,
 			      struct net_bridge_vlan_group *vg,
-			      struct sk_buff *skb, u16 *vid)
+			      struct sk_buff *skb, u16 *vid,
+			      u8 *state)
 {
 	struct br_vlan_stats *stats;
 	struct net_bridge_vlan *v;
@@ -532,13 +538,25 @@ static bool __allowed_ingress(const struct net_bridge *br,
 			skb->vlan_tci |= pvid;
 
 		/* if stats are disabled we can avoid the lookup */
-		if (!br_opt_get(br, BROPT_VLAN_STATS_ENABLED))
-			return true;
+		if (!br_opt_get(br, BROPT_VLAN_STATS_ENABLED)) {
+			if (*state == BR_STATE_FORWARDING) {
+				*state = br_vlan_get_pvid_state(vg);
+				return br_vlan_state_allowed(*state, true);
+			} else {
+				return true;
+			}
+		}
 	}
 	v = br_vlan_find(vg, *vid);
 	if (!v || !br_vlan_should_use(v))
 		goto drop;
 
+	if (*state == BR_STATE_FORWARDING) {
+		*state = br_vlan_get_state(v);
+		if (!br_vlan_state_allowed(*state, true))
+			goto drop;
+	}
+
 	if (br_opt_get(br, BROPT_VLAN_STATS_ENABLED)) {
 		stats = this_cpu_ptr(v->stats);
 		u64_stats_update_begin(&stats->syncp);
@@ -556,7 +574,7 @@ drop:
 
 bool br_allowed_ingress(const struct net_bridge *br,
 			struct net_bridge_vlan_group *vg, struct sk_buff *skb,
-			u16 *vid)
+			u16 *vid, u8 *state)
 {
 	/* If VLAN filtering is disabled on the bridge, all packets are
 	 * permitted.
@@ -566,7 +584,7 @@ bool br_allowed_ingress(const struct net_bridge *br,
 		return true;
 	}
 
-	return __allowed_ingress(br, vg, skb, vid);
+	return __allowed_ingress(br, vg, skb, vid, state);
 }
 
 /* Called under RCU. */
@@ -582,7 +600,8 @@ bool br_allowed_egress(struct net_bridge_vlan_group *vg,
 
 	br_vlan_get_tag(skb, &vid);
 	v = br_vlan_find(vg, vid);
-	if (v && br_vlan_should_use(v))
+	if (v && br_vlan_should_use(v) &&
+	    br_vlan_state_allowed(br_vlan_get_state(v), false))
 		return true;
 
 	return false;
@@ -593,6 +612,7 @@ bool br_should_learn(struct net_bridge_port *p, struct sk_buff *skb, u16 *vid)
 {
 	struct net_bridge_vlan_group *vg;
 	struct net_bridge *br = p->br;
+	struct net_bridge_vlan *v;
 
 	/* If filtering was disabled at input, let it pass. */
 	if (!br_opt_get(br, BROPT_VLAN_ENABLED))
@@ -607,13 +627,15 @@ bool br_should_learn(struct net_bridge_port *p, struct sk_buff *skb, u16 *vid)
 
 	if (!*vid) {
 		*vid = br_get_pvid(vg);
-		if (!*vid)
+		if (!*vid ||
+		    !br_vlan_state_allowed(br_vlan_get_pvid_state(vg), true))
 			return false;
 
 		return true;
 	}
 
-	if (br_vlan_find(vg, *vid))
+	v = br_vlan_find(vg, *vid);
+	if (v && br_vlan_state_allowed(br_vlan_get_state(v), true))
 		return true;
 
 	return false;
@@ -1816,6 +1838,7 @@ static const struct nla_policy br_vlan_db_policy[BRIDGE_VLANDB_ENTRY_MAX + 1] =
 	[BRIDGE_VLANDB_ENTRY_INFO]	= { .type = NLA_EXACT_LEN,
 					    .len = sizeof(struct bridge_vlan_info) },
 	[BRIDGE_VLANDB_ENTRY_RANGE]	= { .type = NLA_U16 },
+	[BRIDGE_VLANDB_ENTRY_STATE]	= { .type = NLA_U8 },
 };
 
 static int br_vlan_rtm_process_one(struct net_device *dev,
diff --git a/net/bridge/br_vlan_options.c b/net/bridge/br_vlan_options.c
index 27275ac3e42e..cd2eb194eb98 100644
--- a/net/bridge/br_vlan_options.c
+++ b/net/bridge/br_vlan_options.c
@@ -11,16 +11,54 @@
 bool br_vlan_opts_eq(const struct net_bridge_vlan *v1,
 		     const struct net_bridge_vlan *v2)
 {
-	return true;
+	return v1->state == v2->state;
 }
 
 bool br_vlan_opts_fill(struct sk_buff *skb, const struct net_bridge_vlan *v)
 {
-	return true;
+	return !nla_put_u8(skb, BRIDGE_VLANDB_ENTRY_STATE,
+			   br_vlan_get_state(v));
 }
 
 size_t br_vlan_opts_nl_size(void)
 {
+	return nla_total_size(sizeof(u8)); /* BRIDGE_VLANDB_ENTRY_STATE */
+}
+
+static int br_vlan_modify_state(struct net_bridge_vlan_group *vg,
+				struct net_bridge_vlan *v,
+				u8 state,
+				bool *changed,
+				struct netlink_ext_ack *extack)
+{
+	struct net_bridge *br;
+
+	ASSERT_RTNL();
+
+	if (state > BR_STATE_BLOCKING) {
+		NL_SET_ERR_MSG_MOD(extack, "Invalid vlan state");
+		return -EINVAL;
+	}
+
+	if (br_vlan_is_brentry(v))
+		br = v->br;
+	else
+		br = v->port->br;
+
+	if (br->stp_enabled == BR_KERNEL_STP) {
+		NL_SET_ERR_MSG_MOD(extack, "Can't modify vlan state when using kernel STP");
+		return -EBUSY;
+	}
+
+	if (v->state == state)
+		return 0;
+
+	if (v->vid == br_get_pvid(vg))
+		br_vlan_set_pvid_state(vg, state);
+
+	br_vlan_set_state(v, state);
+	*changed = true;
+
 	return 0;
 }
 
@@ -32,7 +70,17 @@ static int br_vlan_process_one_opts(const struct net_bridge *br,
 				    bool *changed,
 				    struct netlink_ext_ack *extack)
 {
+	int err;
+
 	*changed = false;
+	if (tb[BRIDGE_VLANDB_ENTRY_STATE]) {
+		u8 state = nla_get_u8(tb[BRIDGE_VLANDB_ENTRY_STATE]);
+
+		err = br_vlan_modify_state(vg, v, state, changed, extack);
+		if (err)
+			return err;
+	}
+
 	return 0;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From f870fa0b5768842cb4690c1c11f19f28b731ae6d Mon Sep 17 00:00:00 2001
From: Mat Martineau <mathew.j.martineau@linux.intel.com>
Date: Tue, 21 Jan 2020 16:56:15 -0800
Subject: mptcp: Add MPTCP socket stubs

Implements the infrastructure for MPTCP sockets.

MPTCP sockets open one in-kernel TCP socket per subflow. These subflow
sockets are only managed by the MPTCP socket that owns them and are not
visible from userspace. This commit allows a userspace program to open
an MPTCP socket with:

  sock = socket(AF_INET, SOCK_STREAM, IPPROTO_MPTCP);

The resulting socket is simply a wrapper around a single regular TCP
socket, without any of the MPTCP protocol implemented over the wire.

Co-developed-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Florian Westphal <fw@strlen.de>
Co-developed-by: Peter Krystad <peter.krystad@linux.intel.com>
Signed-off-by: Peter Krystad <peter.krystad@linux.intel.com>
Co-developed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Co-developed-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 MAINTAINERS          |   1 +
 include/net/mptcp.h  |  16 ++++++
 net/Kconfig          |   1 +
 net/Makefile         |   1 +
 net/ipv4/tcp.c       |   2 +
 net/ipv6/tcp_ipv6.c  |   7 +++
 net/mptcp/Kconfig    |  16 ++++++
 net/mptcp/Makefile   |   4 ++
 net/mptcp/protocol.c | 142 +++++++++++++++++++++++++++++++++++++++++++++++++++
 net/mptcp/protocol.h |  22 ++++++++
 10 files changed, 212 insertions(+)
 create mode 100644 net/mptcp/Kconfig
 create mode 100644 net/mptcp/Makefile
 create mode 100644 net/mptcp/protocol.c
 create mode 100644 net/mptcp/protocol.h

diff --git a/MAINTAINERS b/MAINTAINERS
index 702382b89c37..4be9c1b50183 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -11583,6 +11583,7 @@ W:	https://github.com/multipath-tcp/mptcp_net-next/wiki
 B:	https://github.com/multipath-tcp/mptcp_net-next/issues
 S:	Maintained
 F:	include/net/mptcp.h
+F:	net/mptcp/
 
 NETWORKING [TCP]
 M:	Eric Dumazet <edumazet@google.com>
diff --git a/include/net/mptcp.h b/include/net/mptcp.h
index 0573ae75c3db..98ba22379117 100644
--- a/include/net/mptcp.h
+++ b/include/net/mptcp.h
@@ -28,6 +28,8 @@ struct mptcp_ext {
 
 #ifdef CONFIG_MPTCP
 
+void mptcp_init(void);
+
 /* move the skb extension owership, with the assumption that 'to' is
  * newly allocated
  */
@@ -70,6 +72,10 @@ static inline bool mptcp_skb_can_collapse(const struct sk_buff *to,
 
 #else
 
+static inline void mptcp_init(void)
+{
+}
+
 static inline void mptcp_skb_ext_move(struct sk_buff *to,
 				      const struct sk_buff *from)
 {
@@ -82,4 +88,14 @@ static inline bool mptcp_skb_can_collapse(const struct sk_buff *to,
 }
 
 #endif /* CONFIG_MPTCP */
+
+#if IS_ENABLED(CONFIG_MPTCP_IPV6)
+int mptcpv6_init(void);
+#elif IS_ENABLED(CONFIG_IPV6)
+static inline int mptcpv6_init(void)
+{
+	return 0;
+}
+#endif
+
 #endif /* __NET_MPTCP_H */
diff --git a/net/Kconfig b/net/Kconfig
index 54916b7adb9b..b0937a700f01 100644
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -91,6 +91,7 @@ if INET
 source "net/ipv4/Kconfig"
 source "net/ipv6/Kconfig"
 source "net/netlabel/Kconfig"
+source "net/mptcp/Kconfig"
 
 endif # if INET
 
diff --git a/net/Makefile b/net/Makefile
index 848303d98d3d..07ea48160874 100644
--- a/net/Makefile
+++ b/net/Makefile
@@ -87,3 +87,4 @@ endif
 obj-$(CONFIG_QRTR)		+= qrtr/
 obj-$(CONFIG_NET_NCSI)		+= ncsi/
 obj-$(CONFIG_XDP_SOCKETS)	+= xdp/
+obj-$(CONFIG_MPTCP)		+= mptcp/
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 6711a97de3ce..7dfb78caedaf 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -271,6 +271,7 @@
 #include <net/icmp.h>
 #include <net/inet_common.h>
 #include <net/tcp.h>
+#include <net/mptcp.h>
 #include <net/xfrm.h>
 #include <net/ip.h>
 #include <net/sock.h>
@@ -4021,4 +4022,5 @@ void __init tcp_init(void)
 	tcp_metrics_init();
 	BUG_ON(tcp_register_congestion_control(&tcp_reno) != 0);
 	tcp_tasklet_init();
+	mptcp_init();
 }
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 5b5260103b65..60068ffde1d9 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -2163,9 +2163,16 @@ int __init tcpv6_init(void)
 	ret = register_pernet_subsys(&tcpv6_net_ops);
 	if (ret)
 		goto out_tcpv6_protosw;
+
+	ret = mptcpv6_init();
+	if (ret)
+		goto out_tcpv6_pernet_subsys;
+
 out:
 	return ret;
 
+out_tcpv6_pernet_subsys:
+	unregister_pernet_subsys(&tcpv6_net_ops);
 out_tcpv6_protosw:
 	inet6_unregister_protosw(&tcpv6_protosw);
 out_tcpv6_protocol:
diff --git a/net/mptcp/Kconfig b/net/mptcp/Kconfig
new file mode 100644
index 000000000000..c1a99f07a4cd
--- /dev/null
+++ b/net/mptcp/Kconfig
@@ -0,0 +1,16 @@
+
+config MPTCP
+	bool "MPTCP: Multipath TCP"
+	depends on INET
+	select SKB_EXTENSIONS
+	help
+	  Multipath TCP (MPTCP) connections send and receive data over multiple
+	  subflows in order to utilize multiple network paths. Each subflow
+	  uses the TCP protocol, and TCP options carry header information for
+	  MPTCP.
+
+config MPTCP_IPV6
+	bool "MPTCP: IPv6 support for Multipath TCP"
+	depends on MPTCP
+	select IPV6
+	default y
diff --git a/net/mptcp/Makefile b/net/mptcp/Makefile
new file mode 100644
index 000000000000..659129d1fcbf
--- /dev/null
+++ b/net/mptcp/Makefile
@@ -0,0 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0
+obj-$(CONFIG_MPTCP) += mptcp.o
+
+mptcp-y := protocol.o
diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
new file mode 100644
index 000000000000..5e24e7cf7d70
--- /dev/null
+++ b/net/mptcp/protocol.c
@@ -0,0 +1,142 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Multipath TCP
+ *
+ * Copyright (c) 2017 - 2019, Intel Corporation.
+ */
+
+#define pr_fmt(fmt) "MPTCP: " fmt
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/netdevice.h>
+#include <net/sock.h>
+#include <net/inet_common.h>
+#include <net/inet_hashtables.h>
+#include <net/protocol.h>
+#include <net/tcp.h>
+#include <net/mptcp.h>
+#include "protocol.h"
+
+static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
+{
+	struct mptcp_sock *msk = mptcp_sk(sk);
+	struct socket *subflow = msk->subflow;
+
+	if (msg->msg_flags & ~(MSG_MORE | MSG_DONTWAIT | MSG_NOSIGNAL))
+		return -EOPNOTSUPP;
+
+	return sock_sendmsg(subflow, msg);
+}
+
+static int mptcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
+			 int nonblock, int flags, int *addr_len)
+{
+	struct mptcp_sock *msk = mptcp_sk(sk);
+	struct socket *subflow = msk->subflow;
+
+	if (msg->msg_flags & ~(MSG_WAITALL | MSG_DONTWAIT))
+		return -EOPNOTSUPP;
+
+	return sock_recvmsg(subflow, msg, flags);
+}
+
+static int mptcp_init_sock(struct sock *sk)
+{
+	return 0;
+}
+
+static void mptcp_close(struct sock *sk, long timeout)
+{
+	struct mptcp_sock *msk = mptcp_sk(sk);
+
+	inet_sk_state_store(sk, TCP_CLOSE);
+
+	if (msk->subflow) {
+		pr_debug("subflow=%p", msk->subflow->sk);
+		sock_release(msk->subflow);
+	}
+
+	sock_orphan(sk);
+	sock_put(sk);
+}
+
+static int mptcp_connect(struct sock *sk, struct sockaddr *saddr, int len)
+{
+	struct mptcp_sock *msk = mptcp_sk(sk);
+	int err;
+
+	saddr->sa_family = AF_INET;
+
+	pr_debug("msk=%p, subflow=%p", msk, msk->subflow->sk);
+
+	err = kernel_connect(msk->subflow, saddr, len, 0);
+
+	sk->sk_state = TCP_ESTABLISHED;
+
+	return err;
+}
+
+static struct proto mptcp_prot = {
+	.name		= "MPTCP",
+	.owner		= THIS_MODULE,
+	.init		= mptcp_init_sock,
+	.close		= mptcp_close,
+	.accept		= inet_csk_accept,
+	.connect	= mptcp_connect,
+	.shutdown	= tcp_shutdown,
+	.sendmsg	= mptcp_sendmsg,
+	.recvmsg	= mptcp_recvmsg,
+	.hash		= inet_hash,
+	.unhash		= inet_unhash,
+	.get_port	= inet_csk_get_port,
+	.obj_size	= sizeof(struct mptcp_sock),
+	.no_autobind	= true,
+};
+
+static struct inet_protosw mptcp_protosw = {
+	.type		= SOCK_STREAM,
+	.protocol	= IPPROTO_MPTCP,
+	.prot		= &mptcp_prot,
+	.ops		= &inet_stream_ops,
+};
+
+void __init mptcp_init(void)
+{
+	if (proto_register(&mptcp_prot, 1) != 0)
+		panic("Failed to register MPTCP proto.\n");
+
+	inet_register_protosw(&mptcp_protosw);
+}
+
+#if IS_ENABLED(CONFIG_MPTCP_IPV6)
+static struct proto mptcp_v6_prot;
+
+static struct inet_protosw mptcp_v6_protosw = {
+	.type		= SOCK_STREAM,
+	.protocol	= IPPROTO_MPTCP,
+	.prot		= &mptcp_v6_prot,
+	.ops		= &inet6_stream_ops,
+	.flags		= INET_PROTOSW_ICSK,
+};
+
+int mptcpv6_init(void)
+{
+	int err;
+
+	mptcp_v6_prot = mptcp_prot;
+	strcpy(mptcp_v6_prot.name, "MPTCPv6");
+	mptcp_v6_prot.slab = NULL;
+	mptcp_v6_prot.obj_size = sizeof(struct mptcp_sock) +
+				 sizeof(struct ipv6_pinfo);
+
+	err = proto_register(&mptcp_v6_prot, 1);
+	if (err)
+		return err;
+
+	err = inet6_register_protosw(&mptcp_v6_protosw);
+	if (err)
+		proto_unregister(&mptcp_v6_prot);
+
+	return err;
+}
+#endif
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
new file mode 100644
index 000000000000..ee04a01bffd3
--- /dev/null
+++ b/net/mptcp/protocol.h
@@ -0,0 +1,22 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Multipath TCP
+ *
+ * Copyright (c) 2017 - 2019, Intel Corporation.
+ */
+
+#ifndef __MPTCP_PROTOCOL_H
+#define __MPTCP_PROTOCOL_H
+
+/* MPTCP connection sock */
+struct mptcp_sock {
+	/* inet_connection_sock must be the first member */
+	struct inet_connection_sock sk;
+	struct socket	*subflow; /* outgoing connect/listener/!mp_capable */
+};
+
+static inline struct mptcp_sock *mptcp_sk(const struct sock *sk)
+{
+	return (struct mptcp_sock *)sk;
+}
+
+#endif /* __MPTCP_PROTOCOL_H */
-- 
cgit v1.2.3-59-g8ed1b


From eda7acddf8080bb2d022a8d4b8b2345eb80c63ec Mon Sep 17 00:00:00 2001
From: Peter Krystad <peter.krystad@linux.intel.com>
Date: Tue, 21 Jan 2020 16:56:16 -0800
Subject: mptcp: Handle MPTCP TCP options

Add hooks to parse and format the MP_CAPABLE option.

This option is handled according to MPTCP version 0 (RFC6824).
MPTCP version 1 MP_CAPABLE (RFC6824bis/RFC8684) will be added later in
coordination with related code changes.

Co-developed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Co-developed-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Florian Westphal <fw@strlen.de>
Co-developed-by: Davide Caratti <dcaratti@redhat.com>
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Signed-off-by: Peter Krystad <peter.krystad@linux.intel.com>
Signed-off-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/tcp.h   | 18 ++++++++++
 include/net/mptcp.h   | 18 ++++++++++
 net/ipv4/tcp_input.c  |  5 +++
 net/ipv4/tcp_output.c | 13 +++++++
 net/mptcp/Makefile    |  2 +-
 net/mptcp/options.c   | 97 +++++++++++++++++++++++++++++++++++++++++++++++++++
 net/mptcp/protocol.h  | 29 +++++++++++++++
 7 files changed, 181 insertions(+), 1 deletion(-)
 create mode 100644 net/mptcp/options.c

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index ca6f01531e64..52798ab00394 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -78,6 +78,16 @@ struct tcp_sack_block {
 #define TCP_SACK_SEEN     (1 << 0)   /*1 = peer is SACK capable, */
 #define TCP_DSACK_SEEN    (1 << 2)   /*1 = DSACK was received from peer*/
 
+#if IS_ENABLED(CONFIG_MPTCP)
+struct mptcp_options_received {
+	u64	sndr_key;
+	u64	rcvr_key;
+	u8	mp_capable : 1,
+		mp_join : 1,
+		dss : 1;
+};
+#endif
+
 struct tcp_options_received {
 /*	PAWS/RTTM data	*/
 	int	ts_recent_stamp;/* Time we stored ts_recent (for aging) */
@@ -95,6 +105,9 @@ struct tcp_options_received {
 	u8	num_sacks;	/* Number of SACK blocks		*/
 	u16	user_mss;	/* mss requested by user in ioctl	*/
 	u16	mss_clamp;	/* Maximal mss, negotiated at connection setup */
+#if IS_ENABLED(CONFIG_MPTCP)
+	struct mptcp_options_received	mptcp;
+#endif
 };
 
 static inline void tcp_clear_options(struct tcp_options_received *rx_opt)
@@ -104,6 +117,11 @@ static inline void tcp_clear_options(struct tcp_options_received *rx_opt)
 #if IS_ENABLED(CONFIG_SMC)
 	rx_opt->smc_ok = 0;
 #endif
+#if IS_ENABLED(CONFIG_MPTCP)
+	rx_opt->mptcp.mp_capable = 0;
+	rx_opt->mptcp.mp_join = 0;
+	rx_opt->mptcp.dss = 0;
+#endif
 }
 
 /* This is the max number of SACKS that we'll generate and process. It's safe
diff --git a/include/net/mptcp.h b/include/net/mptcp.h
index 98ba22379117..3daec2ceb3ff 100644
--- a/include/net/mptcp.h
+++ b/include/net/mptcp.h
@@ -9,6 +9,7 @@
 #define __NET_MPTCP_H
 
 #include <linux/skbuff.h>
+#include <linux/tcp.h>
 #include <linux/types.h>
 
 /* MPTCP sk_buff extension data */
@@ -26,10 +27,22 @@ struct mptcp_ext {
 	/* one byte hole */
 };
 
+struct mptcp_out_options {
+#if IS_ENABLED(CONFIG_MPTCP)
+	u16 suboptions;
+	u64 sndr_key;
+	u64 rcvr_key;
+#endif
+};
+
 #ifdef CONFIG_MPTCP
 
 void mptcp_init(void);
 
+void mptcp_parse_option(const unsigned char *ptr, int opsize,
+			struct tcp_options_received *opt_rx);
+void mptcp_write_options(__be32 *ptr, struct mptcp_out_options *opts);
+
 /* move the skb extension owership, with the assumption that 'to' is
  * newly allocated
  */
@@ -76,6 +89,11 @@ static inline void mptcp_init(void)
 {
 }
 
+static inline void mptcp_parse_option(const unsigned char *ptr, int opsize,
+				      struct tcp_options_received *opt_rx)
+{
+}
+
 static inline void mptcp_skb_ext_move(struct sk_buff *to,
 				      const struct sk_buff *from)
 {
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 358365598216..3458ee13e6f0 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -79,6 +79,7 @@
 #include <trace/events/tcp.h>
 #include <linux/jump_label_ratelimit.h>
 #include <net/busy_poll.h>
+#include <net/mptcp.h>
 
 int sysctl_tcp_max_orphans __read_mostly = NR_FILE;
 
@@ -3924,6 +3925,10 @@ void tcp_parse_options(const struct net *net,
 				 */
 				break;
 #endif
+			case TCPOPT_MPTCP:
+				mptcp_parse_option(ptr, opsize, opt_rx);
+				break;
+
 			case TCPOPT_FASTOPEN:
 				tcp_parse_fastopen_option(
 					opsize - TCPOLEN_FASTOPEN_BASE,
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 786978cb2db7..0f0984f39f67 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -38,6 +38,7 @@
 #define pr_fmt(fmt) "TCP: " fmt
 
 #include <net/tcp.h>
+#include <net/mptcp.h>
 
 #include <linux/compiler.h>
 #include <linux/gfp.h>
@@ -414,6 +415,7 @@ static inline bool tcp_urg_mode(const struct tcp_sock *tp)
 #define OPTION_WSCALE		(1 << 3)
 #define OPTION_FAST_OPEN_COOKIE	(1 << 8)
 #define OPTION_SMC		(1 << 9)
+#define OPTION_MPTCP		(1 << 10)
 
 static void smc_options_write(__be32 *ptr, u16 *options)
 {
@@ -439,8 +441,17 @@ struct tcp_out_options {
 	__u8 *hash_location;	/* temporary pointer, overloaded */
 	__u32 tsval, tsecr;	/* need to include OPTION_TS */
 	struct tcp_fastopen_cookie *fastopen_cookie;	/* Fast open cookie */
+	struct mptcp_out_options mptcp;
 };
 
+static void mptcp_options_write(__be32 *ptr, struct tcp_out_options *opts)
+{
+#if IS_ENABLED(CONFIG_MPTCP)
+	if (unlikely(OPTION_MPTCP & opts->options))
+		mptcp_write_options(ptr, &opts->mptcp);
+#endif
+}
+
 /* Write previously computed TCP options to the packet.
  *
  * Beware: Something in the Internet is very sensitive to the ordering of
@@ -549,6 +560,8 @@ static void tcp_options_write(__be32 *ptr, struct tcp_sock *tp,
 	}
 
 	smc_options_write(ptr, &options);
+
+	mptcp_options_write(ptr, opts);
 }
 
 static void smc_set_option(const struct tcp_sock *tp,
diff --git a/net/mptcp/Makefile b/net/mptcp/Makefile
index 659129d1fcbf..27a846263f08 100644
--- a/net/mptcp/Makefile
+++ b/net/mptcp/Makefile
@@ -1,4 +1,4 @@
 # SPDX-License-Identifier: GPL-2.0
 obj-$(CONFIG_MPTCP) += mptcp.o
 
-mptcp-y := protocol.o
+mptcp-y := protocol.o options.o
diff --git a/net/mptcp/options.c b/net/mptcp/options.c
new file mode 100644
index 000000000000..b7a31c0e5283
--- /dev/null
+++ b/net/mptcp/options.c
@@ -0,0 +1,97 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Multipath TCP
+ *
+ * Copyright (c) 2017 - 2019, Intel Corporation.
+ */
+
+#include <linux/kernel.h>
+#include <net/tcp.h>
+#include <net/mptcp.h>
+#include "protocol.h"
+
+void mptcp_parse_option(const unsigned char *ptr, int opsize,
+			struct tcp_options_received *opt_rx)
+{
+	struct mptcp_options_received *mp_opt = &opt_rx->mptcp;
+	u8 subtype = *ptr >> 4;
+	u8 version;
+	u8 flags;
+
+	switch (subtype) {
+	case MPTCPOPT_MP_CAPABLE:
+		if (opsize != TCPOLEN_MPTCP_MPC_SYN &&
+		    opsize != TCPOLEN_MPTCP_MPC_ACK)
+			break;
+
+		version = *ptr++ & MPTCP_VERSION_MASK;
+		if (version != MPTCP_SUPPORTED_VERSION)
+			break;
+
+		flags = *ptr++;
+		if (!((flags & MPTCP_CAP_FLAG_MASK) == MPTCP_CAP_HMAC_SHA1) ||
+		    (flags & MPTCP_CAP_EXTENSIBILITY))
+			break;
+
+		/* RFC 6824, Section 3.1:
+		 * "For the Checksum Required bit (labeled "A"), if either
+		 * host requires the use of checksums, checksums MUST be used.
+		 * In other words, the only way for checksums not to be used
+		 * is if both hosts in their SYNs set A=0."
+		 *
+		 * Section 3.3.0:
+		 * "If a checksum is not present when its use has been
+		 * negotiated, the receiver MUST close the subflow with a RST as
+		 * it is considered broken."
+		 *
+		 * We don't implement DSS checksum - fall back to TCP.
+		 */
+		if (flags & MPTCP_CAP_CHECKSUM_REQD)
+			break;
+
+		mp_opt->mp_capable = 1;
+		mp_opt->sndr_key = get_unaligned_be64(ptr);
+		ptr += 8;
+
+		if (opsize == TCPOLEN_MPTCP_MPC_ACK) {
+			mp_opt->rcvr_key = get_unaligned_be64(ptr);
+			ptr += 8;
+			pr_debug("MP_CAPABLE sndr=%llu, rcvr=%llu",
+				 mp_opt->sndr_key, mp_opt->rcvr_key);
+		} else {
+			pr_debug("MP_CAPABLE sndr=%llu", mp_opt->sndr_key);
+		}
+		break;
+
+	case MPTCPOPT_DSS:
+		pr_debug("DSS");
+		mp_opt->dss = 1;
+		break;
+
+	default:
+		break;
+	}
+}
+
+void mptcp_write_options(__be32 *ptr, struct mptcp_out_options *opts)
+{
+	if ((OPTION_MPTCP_MPC_SYN |
+	     OPTION_MPTCP_MPC_ACK) & opts->suboptions) {
+		u8 len;
+
+		if (OPTION_MPTCP_MPC_SYN & opts->suboptions)
+			len = TCPOLEN_MPTCP_MPC_SYN;
+		else
+			len = TCPOLEN_MPTCP_MPC_ACK;
+
+		*ptr++ = htonl((TCPOPT_MPTCP << 24) | (len << 16) |
+			       (MPTCPOPT_MP_CAPABLE << 12) |
+			       (MPTCP_SUPPORTED_VERSION << 8) |
+			       MPTCP_CAP_HMAC_SHA1);
+		put_unaligned_be64(opts->sndr_key, ptr);
+		ptr += 2;
+		if (OPTION_MPTCP_MPC_ACK & opts->suboptions) {
+			put_unaligned_be64(opts->rcvr_key, ptr);
+			ptr += 2;
+		}
+	}
+}
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index ee04a01bffd3..c59cf8b220b0 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -7,6 +7,35 @@
 #ifndef __MPTCP_PROTOCOL_H
 #define __MPTCP_PROTOCOL_H
 
+#define MPTCP_SUPPORTED_VERSION	0
+
+/* MPTCP option bits */
+#define OPTION_MPTCP_MPC_SYN	BIT(0)
+#define OPTION_MPTCP_MPC_SYNACK	BIT(1)
+#define OPTION_MPTCP_MPC_ACK	BIT(2)
+
+/* MPTCP option subtypes */
+#define MPTCPOPT_MP_CAPABLE	0
+#define MPTCPOPT_MP_JOIN	1
+#define MPTCPOPT_DSS		2
+#define MPTCPOPT_ADD_ADDR	3
+#define MPTCPOPT_RM_ADDR	4
+#define MPTCPOPT_MP_PRIO	5
+#define MPTCPOPT_MP_FAIL	6
+#define MPTCPOPT_MP_FASTCLOSE	7
+
+/* MPTCP suboption lengths */
+#define TCPOLEN_MPTCP_MPC_SYN		12
+#define TCPOLEN_MPTCP_MPC_SYNACK	12
+#define TCPOLEN_MPTCP_MPC_ACK		20
+
+/* MPTCP MP_CAPABLE flags */
+#define MPTCP_VERSION_MASK	(0x0F)
+#define MPTCP_CAP_CHECKSUM_REQD	BIT(7)
+#define MPTCP_CAP_EXTENSIBILITY	BIT(6)
+#define MPTCP_CAP_HMAC_SHA1	BIT(0)
+#define MPTCP_CAP_FLAG_MASK	(0x3F)
+
 /* MPTCP connection sock */
 struct mptcp_sock {
 	/* inet_connection_sock must be the first member */
-- 
cgit v1.2.3-59-g8ed1b


From 2303f994b3e187091fd08148066688b08f837efc Mon Sep 17 00:00:00 2001
From: Peter Krystad <peter.krystad@linux.intel.com>
Date: Tue, 21 Jan 2020 16:56:17 -0800
Subject: mptcp: Associate MPTCP context with TCP socket

Use ULP to associate a subflow_context structure with each TCP subflow
socket. Creating these sockets requires new bind and connect functions
to make sure ULP is set up immediately when the subflow sockets are
created.

Co-developed-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Florian Westphal <fw@strlen.de>
Co-developed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Co-developed-by: Davide Caratti <dcaratti@redhat.com>
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Co-developed-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Peter Krystad <peter.krystad@linux.intel.com>
Signed-off-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/tcp.h  |   3 ++
 net/mptcp/Makefile   |   2 +-
 net/mptcp/protocol.c | 132 ++++++++++++++++++++++++++++++++++++++++++++++++---
 net/mptcp/protocol.h |  26 ++++++++++
 net/mptcp/subflow.c  | 119 ++++++++++++++++++++++++++++++++++++++++++++++
 5 files changed, 275 insertions(+), 7 deletions(-)
 create mode 100644 net/mptcp/subflow.c

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 52798ab00394..877947475814 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -397,6 +397,9 @@ struct tcp_sock {
 	u32	mtu_info; /* We received an ICMP_FRAG_NEEDED / ICMPV6_PKT_TOOBIG
 			   * while socket was owned by user.
 			   */
+#if IS_ENABLED(CONFIG_MPTCP)
+	bool	is_mptcp;
+#endif
 
 #ifdef CONFIG_TCP_MD5SIG
 /* TCP AF-Specific parts; only used by MD5 Signature support so far */
diff --git a/net/mptcp/Makefile b/net/mptcp/Makefile
index 27a846263f08..e1ee5aade8b0 100644
--- a/net/mptcp/Makefile
+++ b/net/mptcp/Makefile
@@ -1,4 +1,4 @@
 # SPDX-License-Identifier: GPL-2.0
 obj-$(CONFIG_MPTCP) += mptcp.o
 
-mptcp-y := protocol.o options.o
+mptcp-y := protocol.o subflow.o options.o
diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 5e24e7cf7d70..294b03a0393a 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -17,6 +17,53 @@
 #include <net/mptcp.h>
 #include "protocol.h"
 
+#define MPTCP_SAME_STATE TCP_MAX_STATES
+
+/* If msk has an initial subflow socket, and the MP_CAPABLE handshake has not
+ * completed yet or has failed, return the subflow socket.
+ * Otherwise return NULL.
+ */
+static struct socket *__mptcp_nmpc_socket(const struct mptcp_sock *msk)
+{
+	if (!msk->subflow)
+		return NULL;
+
+	return msk->subflow;
+}
+
+static bool __mptcp_can_create_subflow(const struct mptcp_sock *msk)
+{
+	return ((struct sock *)msk)->sk_state == TCP_CLOSE;
+}
+
+static struct socket *__mptcp_socket_create(struct mptcp_sock *msk, int state)
+{
+	struct mptcp_subflow_context *subflow;
+	struct sock *sk = (struct sock *)msk;
+	struct socket *ssock;
+	int err;
+
+	ssock = __mptcp_nmpc_socket(msk);
+	if (ssock)
+		goto set_state;
+
+	if (!__mptcp_can_create_subflow(msk))
+		return ERR_PTR(-EINVAL);
+
+	err = mptcp_subflow_create_socket(sk, &ssock);
+	if (err)
+		return ERR_PTR(err);
+
+	msk->subflow = ssock;
+	subflow = mptcp_subflow_ctx(ssock->sk);
+	subflow->request_mptcp = 1;
+
+set_state:
+	if (state != MPTCP_SAME_STATE)
+		inet_sk_state_store(sk, state);
+	return ssock;
+}
+
 static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 {
 	struct mptcp_sock *msk = mptcp_sk(sk);
@@ -48,12 +95,14 @@ static int mptcp_init_sock(struct sock *sk)
 static void mptcp_close(struct sock *sk, long timeout)
 {
 	struct mptcp_sock *msk = mptcp_sk(sk);
+	struct socket *ssock;
 
 	inet_sk_state_store(sk, TCP_CLOSE);
 
-	if (msk->subflow) {
-		pr_debug("subflow=%p", msk->subflow->sk);
-		sock_release(msk->subflow);
+	ssock = __mptcp_nmpc_socket(msk);
+	if (ssock) {
+		pr_debug("subflow=%p", mptcp_subflow_ctx(ssock->sk));
+		sock_release(ssock);
 	}
 
 	sock_orphan(sk);
@@ -67,7 +116,8 @@ static int mptcp_connect(struct sock *sk, struct sockaddr *saddr, int len)
 
 	saddr->sa_family = AF_INET;
 
-	pr_debug("msk=%p, subflow=%p", msk, msk->subflow->sk);
+	pr_debug("msk=%p, subflow=%p", msk,
+		 mptcp_subflow_ctx(msk->subflow->sk));
 
 	err = kernel_connect(msk->subflow, saddr, len, 0);
 
@@ -93,15 +143,79 @@ static struct proto mptcp_prot = {
 	.no_autobind	= true,
 };
 
+static int mptcp_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
+{
+	struct mptcp_sock *msk = mptcp_sk(sock->sk);
+	struct socket *ssock;
+	int err = -ENOTSUPP;
+
+	if (uaddr->sa_family != AF_INET) // @@ allow only IPv4 for now
+		return err;
+
+	lock_sock(sock->sk);
+	ssock = __mptcp_socket_create(msk, MPTCP_SAME_STATE);
+	if (IS_ERR(ssock)) {
+		err = PTR_ERR(ssock);
+		goto unlock;
+	}
+
+	err = ssock->ops->bind(ssock, uaddr, addr_len);
+
+unlock:
+	release_sock(sock->sk);
+	return err;
+}
+
+static int mptcp_stream_connect(struct socket *sock, struct sockaddr *uaddr,
+				int addr_len, int flags)
+{
+	struct mptcp_sock *msk = mptcp_sk(sock->sk);
+	struct socket *ssock;
+	int err;
+
+	lock_sock(sock->sk);
+	ssock = __mptcp_socket_create(msk, TCP_SYN_SENT);
+	if (IS_ERR(ssock)) {
+		err = PTR_ERR(ssock);
+		goto unlock;
+	}
+
+	err = ssock->ops->connect(ssock, uaddr, addr_len, flags);
+	inet_sk_state_store(sock->sk, inet_sk_state_load(ssock->sk));
+
+unlock:
+	release_sock(sock->sk);
+	return err;
+}
+
+static __poll_t mptcp_poll(struct file *file, struct socket *sock,
+			   struct poll_table_struct *wait)
+{
+	__poll_t mask = 0;
+
+	return mask;
+}
+
+static struct proto_ops mptcp_stream_ops;
+
 static struct inet_protosw mptcp_protosw = {
 	.type		= SOCK_STREAM,
 	.protocol	= IPPROTO_MPTCP,
 	.prot		= &mptcp_prot,
-	.ops		= &inet_stream_ops,
+	.ops		= &mptcp_stream_ops,
+	.flags		= INET_PROTOSW_ICSK,
 };
 
 void __init mptcp_init(void)
 {
+	mptcp_prot.h.hashinfo = tcp_prot.h.hashinfo;
+	mptcp_stream_ops = inet_stream_ops;
+	mptcp_stream_ops.bind = mptcp_bind;
+	mptcp_stream_ops.connect = mptcp_stream_connect;
+	mptcp_stream_ops.poll = mptcp_poll;
+
+	mptcp_subflow_init();
+
 	if (proto_register(&mptcp_prot, 1) != 0)
 		panic("Failed to register MPTCP proto.\n");
 
@@ -109,13 +223,14 @@ void __init mptcp_init(void)
 }
 
 #if IS_ENABLED(CONFIG_MPTCP_IPV6)
+static struct proto_ops mptcp_v6_stream_ops;
 static struct proto mptcp_v6_prot;
 
 static struct inet_protosw mptcp_v6_protosw = {
 	.type		= SOCK_STREAM,
 	.protocol	= IPPROTO_MPTCP,
 	.prot		= &mptcp_v6_prot,
-	.ops		= &inet6_stream_ops,
+	.ops		= &mptcp_v6_stream_ops,
 	.flags		= INET_PROTOSW_ICSK,
 };
 
@@ -133,6 +248,11 @@ int mptcpv6_init(void)
 	if (err)
 		return err;
 
+	mptcp_v6_stream_ops = inet6_stream_ops;
+	mptcp_v6_stream_ops.bind = mptcp_bind;
+	mptcp_v6_stream_ops.connect = mptcp_stream_connect;
+	mptcp_v6_stream_ops.poll = mptcp_poll;
+
 	err = inet6_register_protosw(&mptcp_v6_protosw);
 	if (err)
 		proto_unregister(&mptcp_v6_prot);
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index c59cf8b220b0..543d4d5d8985 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -48,4 +48,30 @@ static inline struct mptcp_sock *mptcp_sk(const struct sock *sk)
 	return (struct mptcp_sock *)sk;
 }
 
+/* MPTCP subflow context */
+struct mptcp_subflow_context {
+	u32	request_mptcp : 1;  /* send MP_CAPABLE */
+	struct	sock *tcp_sock;	    /* tcp sk backpointer */
+	struct	sock *conn;	    /* parent mptcp_sock */
+	struct	rcu_head rcu;
+};
+
+static inline struct mptcp_subflow_context *
+mptcp_subflow_ctx(const struct sock *sk)
+{
+	struct inet_connection_sock *icsk = inet_csk(sk);
+
+	/* Use RCU on icsk_ulp_data only for sock diag code */
+	return (__force struct mptcp_subflow_context *)icsk->icsk_ulp_data;
+}
+
+static inline struct sock *
+mptcp_subflow_tcp_sock(const struct mptcp_subflow_context *subflow)
+{
+	return subflow->tcp_sock;
+}
+
+void mptcp_subflow_init(void);
+int mptcp_subflow_create_socket(struct sock *sk, struct socket **new_sock);
+
 #endif /* __MPTCP_PROTOCOL_H */
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
new file mode 100644
index 000000000000..bf8139353653
--- /dev/null
+++ b/net/mptcp/subflow.c
@@ -0,0 +1,119 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Multipath TCP
+ *
+ * Copyright (c) 2017 - 2019, Intel Corporation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/netdevice.h>
+#include <net/sock.h>
+#include <net/inet_common.h>
+#include <net/inet_hashtables.h>
+#include <net/protocol.h>
+#include <net/tcp.h>
+#include <net/mptcp.h>
+#include "protocol.h"
+
+int mptcp_subflow_create_socket(struct sock *sk, struct socket **new_sock)
+{
+	struct mptcp_subflow_context *subflow;
+	struct net *net = sock_net(sk);
+	struct socket *sf;
+	int err;
+
+	err = sock_create_kern(net, PF_INET, SOCK_STREAM, IPPROTO_TCP, &sf);
+	if (err)
+		return err;
+
+	lock_sock(sf->sk);
+
+	/* kernel sockets do not by default acquire net ref, but TCP timer
+	 * needs it.
+	 */
+	sf->sk->sk_net_refcnt = 1;
+	get_net(net);
+	this_cpu_add(*net->core.sock_inuse, 1);
+	err = tcp_set_ulp(sf->sk, "mptcp");
+	release_sock(sf->sk);
+
+	if (err)
+		return err;
+
+	subflow = mptcp_subflow_ctx(sf->sk);
+	pr_debug("subflow=%p", subflow);
+
+	*new_sock = sf;
+	subflow->conn = sk;
+
+	return 0;
+}
+
+static struct mptcp_subflow_context *subflow_create_ctx(struct sock *sk,
+							gfp_t priority)
+{
+	struct inet_connection_sock *icsk = inet_csk(sk);
+	struct mptcp_subflow_context *ctx;
+
+	ctx = kzalloc(sizeof(*ctx), priority);
+	if (!ctx)
+		return NULL;
+
+	rcu_assign_pointer(icsk->icsk_ulp_data, ctx);
+
+	pr_debug("subflow=%p", ctx);
+
+	ctx->tcp_sock = sk;
+
+	return ctx;
+}
+
+static int subflow_ulp_init(struct sock *sk)
+{
+	struct mptcp_subflow_context *ctx;
+	struct tcp_sock *tp = tcp_sk(sk);
+	int err = 0;
+
+	/* disallow attaching ULP to a socket unless it has been
+	 * created with sock_create_kern()
+	 */
+	if (!sk->sk_kern_sock) {
+		err = -EOPNOTSUPP;
+		goto out;
+	}
+
+	ctx = subflow_create_ctx(sk, GFP_KERNEL);
+	if (!ctx) {
+		err = -ENOMEM;
+		goto out;
+	}
+
+	pr_debug("subflow=%p, family=%d", ctx, sk->sk_family);
+
+	tp->is_mptcp = 1;
+out:
+	return err;
+}
+
+static void subflow_ulp_release(struct sock *sk)
+{
+	struct mptcp_subflow_context *ctx = mptcp_subflow_ctx(sk);
+
+	if (!ctx)
+		return;
+
+	kfree_rcu(ctx, rcu);
+}
+
+static struct tcp_ulp_ops subflow_ulp_ops __read_mostly = {
+	.name		= "mptcp",
+	.owner		= THIS_MODULE,
+	.init		= subflow_ulp_init,
+	.release	= subflow_ulp_release,
+};
+
+void mptcp_subflow_init(void)
+{
+	if (tcp_register_ulp(&subflow_ulp_ops) != 0)
+		panic("MPTCP: failed to register subflows to ULP\n");
+}
-- 
cgit v1.2.3-59-g8ed1b


From cec37a6e41aae7bf3df9a3da783380a4d9325fd8 Mon Sep 17 00:00:00 2001
From: Peter Krystad <peter.krystad@linux.intel.com>
Date: Tue, 21 Jan 2020 16:56:18 -0800
Subject: mptcp: Handle MP_CAPABLE options for outgoing connections

Add hooks to tcp_output.c to add MP_CAPABLE to an outgoing SYN request,
to capture the MP_CAPABLE in the received SYN-ACK, to add MP_CAPABLE to
the final ACK of the three-way handshake.

Use the .sk_rx_dst_set() handler in the subflow proto to capture when the
responding SYN-ACK is received and notify the MPTCP connection layer.

Co-developed-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Co-developed-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Peter Krystad <peter.krystad@linux.intel.com>
Signed-off-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/tcp.h   |   3 +
 include/net/mptcp.h   |  57 +++++++++++
 net/ipv4/tcp_input.c  |   6 ++
 net/ipv4/tcp_output.c |  44 +++++++++
 net/ipv6/tcp_ipv6.c   |   6 ++
 net/mptcp/options.c   | 100 +++++++++++++++++++
 net/mptcp/protocol.c  | 163 +++++++++++++++++++++++++-----
 net/mptcp/protocol.h  |  40 +++++++-
 net/mptcp/subflow.c   | 268 +++++++++++++++++++++++++++++++++++++++++++++++++-
 9 files changed, 663 insertions(+), 24 deletions(-)

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 877947475814..e9ee06d887fa 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -137,6 +137,9 @@ struct tcp_request_sock {
 	const struct tcp_request_sock_ops *af_specific;
 	u64				snt_synack; /* first SYNACK sent time */
 	bool				tfo_listener;
+#if IS_ENABLED(CONFIG_MPTCP)
+	bool				is_mptcp;
+#endif
 	u32				txhash;
 	u32				rcv_isn;
 	u32				snt_isn;
diff --git a/include/net/mptcp.h b/include/net/mptcp.h
index 3daec2ceb3ff..eabc57c3fde4 100644
--- a/include/net/mptcp.h
+++ b/include/net/mptcp.h
@@ -39,8 +39,27 @@ struct mptcp_out_options {
 
 void mptcp_init(void);
 
+static inline bool sk_is_mptcp(const struct sock *sk)
+{
+	return tcp_sk(sk)->is_mptcp;
+}
+
+static inline bool rsk_is_mptcp(const struct request_sock *req)
+{
+	return tcp_rsk(req)->is_mptcp;
+}
+
 void mptcp_parse_option(const unsigned char *ptr, int opsize,
 			struct tcp_options_received *opt_rx);
+bool mptcp_syn_options(struct sock *sk, unsigned int *size,
+		       struct mptcp_out_options *opts);
+void mptcp_rcv_synsent(struct sock *sk);
+bool mptcp_synack_options(const struct request_sock *req, unsigned int *size,
+			  struct mptcp_out_options *opts);
+bool mptcp_established_options(struct sock *sk, struct sk_buff *skb,
+			       unsigned int *size, unsigned int remaining,
+			       struct mptcp_out_options *opts);
+
 void mptcp_write_options(__be32 *ptr, struct mptcp_out_options *opts);
 
 /* move the skb extension owership, with the assumption that 'to' is
@@ -89,11 +108,47 @@ static inline void mptcp_init(void)
 {
 }
 
+static inline bool sk_is_mptcp(const struct sock *sk)
+{
+	return false;
+}
+
+static inline bool rsk_is_mptcp(const struct request_sock *req)
+{
+	return false;
+}
+
 static inline void mptcp_parse_option(const unsigned char *ptr, int opsize,
 				      struct tcp_options_received *opt_rx)
 {
 }
 
+static inline bool mptcp_syn_options(struct sock *sk, unsigned int *size,
+				     struct mptcp_out_options *opts)
+{
+	return false;
+}
+
+static inline void mptcp_rcv_synsent(struct sock *sk)
+{
+}
+
+static inline bool mptcp_synack_options(const struct request_sock *req,
+					unsigned int *size,
+					struct mptcp_out_options *opts)
+{
+	return false;
+}
+
+static inline bool mptcp_established_options(struct sock *sk,
+					     struct sk_buff *skb,
+					     unsigned int *size,
+					     unsigned int remaining,
+					     struct mptcp_out_options *opts)
+{
+	return false;
+}
+
 static inline void mptcp_skb_ext_move(struct sk_buff *to,
 				      const struct sk_buff *from)
 {
@@ -107,6 +162,8 @@ static inline bool mptcp_skb_can_collapse(const struct sk_buff *to,
 
 #endif /* CONFIG_MPTCP */
 
+void mptcp_handle_ipv6_mapped(struct sock *sk, bool mapped);
+
 #if IS_ENABLED(CONFIG_MPTCP_IPV6)
 int mptcpv6_init(void);
 #elif IS_ENABLED(CONFIG_IPV6)
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 3458ee13e6f0..5165c8de47ee 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -5978,6 +5978,9 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb,
 		tcp_sync_mss(sk, icsk->icsk_pmtu_cookie);
 		tcp_initialize_rcv_mss(sk);
 
+		if (sk_is_mptcp(sk))
+			mptcp_rcv_synsent(sk);
+
 		/* Remember, tcp_poll() does not lock socket!
 		 * Change state from SYN-SENT only after copied_seq
 		 * is initialized. */
@@ -6600,6 +6603,9 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops,
 
 	tcp_rsk(req)->af_specific = af_ops;
 	tcp_rsk(req)->ts_off = 0;
+#if IS_ENABLED(CONFIG_MPTCP)
+	tcp_rsk(req)->is_mptcp = 0;
+#endif
 
 	tcp_clear_options(&tmp_opt);
 	tmp_opt.mss_clamp = af_ops->mss_clamp;
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 0f0984f39f67..5456076166da 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -597,6 +597,22 @@ static void smc_set_option_cond(const struct tcp_sock *tp,
 #endif
 }
 
+static void mptcp_set_option_cond(const struct request_sock *req,
+				  struct tcp_out_options *opts,
+				  unsigned int *remaining)
+{
+	if (rsk_is_mptcp(req)) {
+		unsigned int size;
+
+		if (mptcp_synack_options(req, &size, &opts->mptcp)) {
+			if (*remaining >= size) {
+				opts->options |= OPTION_MPTCP;
+				*remaining -= size;
+			}
+		}
+	}
+}
+
 /* Compute TCP options for SYN packets. This is not the final
  * network wire format yet.
  */
@@ -666,6 +682,15 @@ static unsigned int tcp_syn_options(struct sock *sk, struct sk_buff *skb,
 
 	smc_set_option(tp, opts, &remaining);
 
+	if (sk_is_mptcp(sk)) {
+		unsigned int size;
+
+		if (mptcp_syn_options(sk, &size, &opts->mptcp)) {
+			opts->options |= OPTION_MPTCP;
+			remaining -= size;
+		}
+	}
+
 	return MAX_TCP_OPTION_SPACE - remaining;
 }
 
@@ -727,6 +752,8 @@ static unsigned int tcp_synack_options(const struct sock *sk,
 		}
 	}
 
+	mptcp_set_option_cond(req, opts, &remaining);
+
 	smc_set_option_cond(tcp_sk(sk), ireq, opts, &remaining);
 
 	return MAX_TCP_OPTION_SPACE - remaining;
@@ -764,6 +791,23 @@ static unsigned int tcp_established_options(struct sock *sk, struct sk_buff *skb
 		size += TCPOLEN_TSTAMP_ALIGNED;
 	}
 
+	/* MPTCP options have precedence over SACK for the limited TCP
+	 * option space because a MPTCP connection would be forced to
+	 * fall back to regular TCP if a required multipath option is
+	 * missing. SACK still gets a chance to use whatever space is
+	 * left.
+	 */
+	if (sk_is_mptcp(sk)) {
+		unsigned int remaining = MAX_TCP_OPTION_SPACE - size;
+		unsigned int opt_size = 0;
+
+		if (mptcp_established_options(sk, skb, &opt_size, remaining,
+					      &opts->mptcp)) {
+			opts->options |= OPTION_MPTCP;
+			size += opt_size;
+		}
+	}
+
 	eff_sacks = tp->rx_opt.num_sacks + tp->rx_opt.dsack;
 	if (unlikely(eff_sacks)) {
 		const unsigned int remaining = MAX_TCP_OPTION_SPACE - size;
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 60068ffde1d9..33a578a3eb3a 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -238,6 +238,8 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
 		sin.sin_addr.s_addr = usin->sin6_addr.s6_addr32[3];
 
 		icsk->icsk_af_ops = &ipv6_mapped;
+		if (sk_is_mptcp(sk))
+			mptcp_handle_ipv6_mapped(sk, true);
 		sk->sk_backlog_rcv = tcp_v4_do_rcv;
 #ifdef CONFIG_TCP_MD5SIG
 		tp->af_specific = &tcp_sock_ipv6_mapped_specific;
@@ -248,6 +250,8 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
 		if (err) {
 			icsk->icsk_ext_hdr_len = exthdrlen;
 			icsk->icsk_af_ops = &ipv6_specific;
+			if (sk_is_mptcp(sk))
+				mptcp_handle_ipv6_mapped(sk, false);
 			sk->sk_backlog_rcv = tcp_v6_do_rcv;
 #ifdef CONFIG_TCP_MD5SIG
 			tp->af_specific = &tcp_sock_ipv6_specific;
@@ -1203,6 +1207,8 @@ static struct sock *tcp_v6_syn_recv_sock(const struct sock *sk, struct sk_buff *
 		newnp->saddr = newsk->sk_v6_rcv_saddr;
 
 		inet_csk(newsk)->icsk_af_ops = &ipv6_mapped;
+		if (sk_is_mptcp(newsk))
+			mptcp_handle_ipv6_mapped(newsk, true);
 		newsk->sk_backlog_rcv = tcp_v4_do_rcv;
 #ifdef CONFIG_TCP_MD5SIG
 		newtp->af_specific = &tcp_sock_ipv6_mapped_specific;
diff --git a/net/mptcp/options.c b/net/mptcp/options.c
index b7a31c0e5283..52ff2301b68b 100644
--- a/net/mptcp/options.c
+++ b/net/mptcp/options.c
@@ -72,14 +72,114 @@ void mptcp_parse_option(const unsigned char *ptr, int opsize,
 	}
 }
 
+void mptcp_get_options(const struct sk_buff *skb,
+		       struct tcp_options_received *opt_rx)
+{
+	const unsigned char *ptr;
+	const struct tcphdr *th = tcp_hdr(skb);
+	int length = (th->doff * 4) - sizeof(struct tcphdr);
+
+	ptr = (const unsigned char *)(th + 1);
+
+	while (length > 0) {
+		int opcode = *ptr++;
+		int opsize;
+
+		switch (opcode) {
+		case TCPOPT_EOL:
+			return;
+		case TCPOPT_NOP:	/* Ref: RFC 793 section 3.1 */
+			length--;
+			continue;
+		default:
+			opsize = *ptr++;
+			if (opsize < 2) /* "silly options" */
+				return;
+			if (opsize > length)
+				return;	/* don't parse partial options */
+			if (opcode == TCPOPT_MPTCP)
+				mptcp_parse_option(ptr, opsize, opt_rx);
+			ptr += opsize - 2;
+			length -= opsize;
+		}
+	}
+}
+
+bool mptcp_syn_options(struct sock *sk, unsigned int *size,
+		       struct mptcp_out_options *opts)
+{
+	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
+
+	if (subflow->request_mptcp) {
+		pr_debug("local_key=%llu", subflow->local_key);
+		opts->suboptions = OPTION_MPTCP_MPC_SYN;
+		opts->sndr_key = subflow->local_key;
+		*size = TCPOLEN_MPTCP_MPC_SYN;
+		return true;
+	}
+	return false;
+}
+
+void mptcp_rcv_synsent(struct sock *sk)
+{
+	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
+	struct tcp_sock *tp = tcp_sk(sk);
+
+	pr_debug("subflow=%p", subflow);
+	if (subflow->request_mptcp && tp->rx_opt.mptcp.mp_capable) {
+		subflow->mp_capable = 1;
+		subflow->remote_key = tp->rx_opt.mptcp.sndr_key;
+	} else {
+		tcp_sk(sk)->is_mptcp = 0;
+	}
+}
+
+bool mptcp_established_options(struct sock *sk, struct sk_buff *skb,
+			       unsigned int *size, unsigned int remaining,
+			       struct mptcp_out_options *opts)
+{
+	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
+
+	if (subflow->mp_capable && !subflow->fourth_ack) {
+		opts->suboptions = OPTION_MPTCP_MPC_ACK;
+		opts->sndr_key = subflow->local_key;
+		opts->rcvr_key = subflow->remote_key;
+		*size = TCPOLEN_MPTCP_MPC_ACK;
+		subflow->fourth_ack = 1;
+		pr_debug("subflow=%p, local_key=%llu, remote_key=%llu",
+			 subflow, subflow->local_key, subflow->remote_key);
+		return true;
+	}
+	return false;
+}
+
+bool mptcp_synack_options(const struct request_sock *req, unsigned int *size,
+			  struct mptcp_out_options *opts)
+{
+	struct mptcp_subflow_request_sock *subflow_req = mptcp_subflow_rsk(req);
+
+	if (subflow_req->mp_capable) {
+		opts->suboptions = OPTION_MPTCP_MPC_SYNACK;
+		opts->sndr_key = subflow_req->local_key;
+		*size = TCPOLEN_MPTCP_MPC_SYNACK;
+		pr_debug("subflow_req=%p, local_key=%llu",
+			 subflow_req, subflow_req->local_key);
+		return true;
+	}
+	return false;
+}
+
 void mptcp_write_options(__be32 *ptr, struct mptcp_out_options *opts)
 {
 	if ((OPTION_MPTCP_MPC_SYN |
+	     OPTION_MPTCP_MPC_SYNACK |
 	     OPTION_MPTCP_MPC_ACK) & opts->suboptions) {
 		u8 len;
 
 		if (OPTION_MPTCP_MPC_SYN & opts->suboptions)
 			len = TCPOLEN_MPTCP_MPC_SYN;
+		else if (OPTION_MPTCP_MPC_SYNACK & opts->suboptions)
+			len = TCPOLEN_MPTCP_MPC_SYNACK;
 		else
 			len = TCPOLEN_MPTCP_MPC_ACK;
 
diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 294b03a0393a..bdd58da1e4f6 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -25,12 +25,28 @@
  */
 static struct socket *__mptcp_nmpc_socket(const struct mptcp_sock *msk)
 {
-	if (!msk->subflow)
+	if (!msk->subflow || mptcp_subflow_ctx(msk->subflow->sk)->fourth_ack)
 		return NULL;
 
 	return msk->subflow;
 }
 
+/* if msk has a single subflow, and the mp_capable handshake is failed,
+ * return it.
+ * Otherwise returns NULL
+ */
+static struct socket *__mptcp_tcp_fallback(const struct mptcp_sock *msk)
+{
+	struct socket *ssock = __mptcp_nmpc_socket(msk);
+
+	sock_owned_by_me((const struct sock *)msk);
+
+	if (!ssock || sk_is_mptcp(ssock->sk))
+		return NULL;
+
+	return ssock;
+}
+
 static bool __mptcp_can_create_subflow(const struct mptcp_sock *msk)
 {
 	return ((struct sock *)msk)->sk_state == TCP_CLOSE;
@@ -56,6 +72,7 @@ static struct socket *__mptcp_socket_create(struct mptcp_sock *msk, int state)
 
 	msk->subflow = ssock;
 	subflow = mptcp_subflow_ctx(ssock->sk);
+	list_add(&subflow->node, &msk->conn_list);
 	subflow->request_mptcp = 1;
 
 set_state:
@@ -64,66 +81,169 @@ set_state:
 	return ssock;
 }
 
+static struct sock *mptcp_subflow_get(const struct mptcp_sock *msk)
+{
+	struct mptcp_subflow_context *subflow;
+
+	sock_owned_by_me((const struct sock *)msk);
+
+	mptcp_for_each_subflow(msk, subflow) {
+		return mptcp_subflow_tcp_sock(subflow);
+	}
+
+	return NULL;
+}
+
 static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 {
 	struct mptcp_sock *msk = mptcp_sk(sk);
-	struct socket *subflow = msk->subflow;
+	struct socket *ssock;
+	struct sock *ssk;
+	int ret;
 
 	if (msg->msg_flags & ~(MSG_MORE | MSG_DONTWAIT | MSG_NOSIGNAL))
 		return -EOPNOTSUPP;
 
-	return sock_sendmsg(subflow, msg);
+	lock_sock(sk);
+	ssock = __mptcp_tcp_fallback(msk);
+	if (ssock) {
+		pr_debug("fallback passthrough");
+		ret = sock_sendmsg(ssock, msg);
+		release_sock(sk);
+		return ret;
+	}
+
+	ssk = mptcp_subflow_get(msk);
+	if (!ssk) {
+		release_sock(sk);
+		return -ENOTCONN;
+	}
+
+	ret = sock_sendmsg(ssk->sk_socket, msg);
+
+	release_sock(sk);
+	return ret;
 }
 
 static int mptcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
 			 int nonblock, int flags, int *addr_len)
 {
 	struct mptcp_sock *msk = mptcp_sk(sk);
-	struct socket *subflow = msk->subflow;
+	struct socket *ssock;
+	struct sock *ssk;
+	int copied = 0;
 
 	if (msg->msg_flags & ~(MSG_WAITALL | MSG_DONTWAIT))
 		return -EOPNOTSUPP;
 
-	return sock_recvmsg(subflow, msg, flags);
+	lock_sock(sk);
+	ssock = __mptcp_tcp_fallback(msk);
+	if (ssock) {
+		pr_debug("fallback-read subflow=%p",
+			 mptcp_subflow_ctx(ssock->sk));
+		copied = sock_recvmsg(ssock, msg, flags);
+		release_sock(sk);
+		return copied;
+	}
+
+	ssk = mptcp_subflow_get(msk);
+	if (!ssk) {
+		release_sock(sk);
+		return -ENOTCONN;
+	}
+
+	copied = sock_recvmsg(ssk->sk_socket, msg, flags);
+
+	release_sock(sk);
+
+	return copied;
+}
+
+/* subflow sockets can be either outgoing (connect) or incoming
+ * (accept).
+ *
+ * Outgoing subflows use in-kernel sockets.
+ * Incoming subflows do not have their own 'struct socket' allocated,
+ * so we need to use tcp_close() after detaching them from the mptcp
+ * parent socket.
+ */
+static void __mptcp_close_ssk(struct sock *sk, struct sock *ssk,
+			      struct mptcp_subflow_context *subflow,
+			      long timeout)
+{
+	struct socket *sock = READ_ONCE(ssk->sk_socket);
+
+	list_del(&subflow->node);
+
+	if (sock && sock != sk->sk_socket) {
+		/* outgoing subflow */
+		sock_release(sock);
+	} else {
+		/* incoming subflow */
+		tcp_close(ssk, timeout);
+	}
 }
 
 static int mptcp_init_sock(struct sock *sk)
 {
+	struct mptcp_sock *msk = mptcp_sk(sk);
+
+	INIT_LIST_HEAD(&msk->conn_list);
+
 	return 0;
 }
 
 static void mptcp_close(struct sock *sk, long timeout)
 {
+	struct mptcp_subflow_context *subflow, *tmp;
 	struct mptcp_sock *msk = mptcp_sk(sk);
-	struct socket *ssock;
 
 	inet_sk_state_store(sk, TCP_CLOSE);
 
-	ssock = __mptcp_nmpc_socket(msk);
-	if (ssock) {
-		pr_debug("subflow=%p", mptcp_subflow_ctx(ssock->sk));
-		sock_release(ssock);
+	lock_sock(sk);
+
+	list_for_each_entry_safe(subflow, tmp, &msk->conn_list, node) {
+		struct sock *ssk = mptcp_subflow_tcp_sock(subflow);
+
+		__mptcp_close_ssk(sk, ssk, subflow, timeout);
 	}
 
-	sock_orphan(sk);
-	sock_put(sk);
+	release_sock(sk);
+	sk_common_release(sk);
 }
 
-static int mptcp_connect(struct sock *sk, struct sockaddr *saddr, int len)
+static int mptcp_get_port(struct sock *sk, unsigned short snum)
 {
 	struct mptcp_sock *msk = mptcp_sk(sk);
-	int err;
+	struct socket *ssock;
 
-	saddr->sa_family = AF_INET;
+	ssock = __mptcp_nmpc_socket(msk);
+	pr_debug("msk=%p, subflow=%p", msk, ssock);
+	if (WARN_ON_ONCE(!ssock))
+		return -EINVAL;
 
-	pr_debug("msk=%p, subflow=%p", msk,
-		 mptcp_subflow_ctx(msk->subflow->sk));
+	return inet_csk_get_port(ssock->sk, snum);
+}
 
-	err = kernel_connect(msk->subflow, saddr, len, 0);
+void mptcp_finish_connect(struct sock *ssk)
+{
+	struct mptcp_subflow_context *subflow;
+	struct mptcp_sock *msk;
+	struct sock *sk;
 
-	sk->sk_state = TCP_ESTABLISHED;
+	subflow = mptcp_subflow_ctx(ssk);
 
-	return err;
+	if (!subflow->mp_capable)
+		return;
+
+	sk = subflow->conn;
+	msk = mptcp_sk(sk);
+
+	/* the socket is not connected yet, no msk/subflow ops can access/race
+	 * accessing the field below
+	 */
+	WRITE_ONCE(msk->remote_key, subflow->remote_key);
+	WRITE_ONCE(msk->local_key, subflow->local_key);
 }
 
 static struct proto mptcp_prot = {
@@ -132,13 +252,12 @@ static struct proto mptcp_prot = {
 	.init		= mptcp_init_sock,
 	.close		= mptcp_close,
 	.accept		= inet_csk_accept,
-	.connect	= mptcp_connect,
 	.shutdown	= tcp_shutdown,
 	.sendmsg	= mptcp_sendmsg,
 	.recvmsg	= mptcp_recvmsg,
 	.hash		= inet_hash,
 	.unhash		= inet_unhash,
-	.get_port	= inet_csk_get_port,
+	.get_port	= mptcp_get_port,
 	.obj_size	= sizeof(struct mptcp_sock),
 	.no_autobind	= true,
 };
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index 543d4d5d8985..bd66e7415515 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -40,19 +40,47 @@
 struct mptcp_sock {
 	/* inet_connection_sock must be the first member */
 	struct inet_connection_sock sk;
+	u64		local_key;
+	u64		remote_key;
+	struct list_head conn_list;
 	struct socket	*subflow; /* outgoing connect/listener/!mp_capable */
 };
 
+#define mptcp_for_each_subflow(__msk, __subflow)			\
+	list_for_each_entry(__subflow, &((__msk)->conn_list), node)
+
 static inline struct mptcp_sock *mptcp_sk(const struct sock *sk)
 {
 	return (struct mptcp_sock *)sk;
 }
 
+struct mptcp_subflow_request_sock {
+	struct	tcp_request_sock sk;
+	u8	mp_capable : 1,
+		mp_join : 1,
+		backup : 1;
+	u64	local_key;
+	u64	remote_key;
+};
+
+static inline struct mptcp_subflow_request_sock *
+mptcp_subflow_rsk(const struct request_sock *rsk)
+{
+	return (struct mptcp_subflow_request_sock *)rsk;
+}
+
 /* MPTCP subflow context */
 struct mptcp_subflow_context {
-	u32	request_mptcp : 1;  /* send MP_CAPABLE */
+	struct	list_head node;/* conn_list of subflows */
+	u64	local_key;
+	u64	remote_key;
+	u32	request_mptcp : 1,  /* send MP_CAPABLE */
+		mp_capable : 1,	    /* remote is MPTCP capable */
+		fourth_ack : 1,	    /* send initial DSS */
+		conn_finished : 1;
 	struct	sock *tcp_sock;	    /* tcp sk backpointer */
 	struct	sock *conn;	    /* parent mptcp_sock */
+	const	struct inet_connection_sock_af_ops *icsk_af_ops;
 	struct	rcu_head rcu;
 };
 
@@ -74,4 +102,14 @@ mptcp_subflow_tcp_sock(const struct mptcp_subflow_context *subflow)
 void mptcp_subflow_init(void);
 int mptcp_subflow_create_socket(struct sock *sk, struct socket **new_sock);
 
+extern const struct inet_connection_sock_af_ops ipv4_specific;
+#if IS_ENABLED(CONFIG_MPTCP_IPV6)
+extern const struct inet_connection_sock_af_ops ipv6_specific;
+#endif
+
+void mptcp_get_options(const struct sk_buff *skb,
+		       struct tcp_options_received *opt_rx);
+
+void mptcp_finish_connect(struct sock *sk);
+
 #endif /* __MPTCP_PROTOCOL_H */
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index bf8139353653..df3192305967 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -12,9 +12,188 @@
 #include <net/inet_hashtables.h>
 #include <net/protocol.h>
 #include <net/tcp.h>
+#if IS_ENABLED(CONFIG_MPTCP_IPV6)
+#include <net/ip6_route.h>
+#endif
 #include <net/mptcp.h>
 #include "protocol.h"
 
+static void subflow_init_req(struct request_sock *req,
+			     const struct sock *sk_listener,
+			     struct sk_buff *skb)
+{
+	struct mptcp_subflow_context *listener = mptcp_subflow_ctx(sk_listener);
+	struct mptcp_subflow_request_sock *subflow_req = mptcp_subflow_rsk(req);
+	struct tcp_options_received rx_opt;
+
+	pr_debug("subflow_req=%p, listener=%p", subflow_req, listener);
+
+	memset(&rx_opt.mptcp, 0, sizeof(rx_opt.mptcp));
+	mptcp_get_options(skb, &rx_opt);
+
+	subflow_req->mp_capable = 0;
+
+#ifdef CONFIG_TCP_MD5SIG
+	/* no MPTCP if MD5SIG is enabled on this socket or we may run out of
+	 * TCP option space.
+	 */
+	if (rcu_access_pointer(tcp_sk(sk_listener)->md5sig_info))
+		return;
+#endif
+
+	if (rx_opt.mptcp.mp_capable && listener->request_mptcp) {
+		subflow_req->mp_capable = 1;
+		subflow_req->remote_key = rx_opt.mptcp.sndr_key;
+	}
+}
+
+static void subflow_v4_init_req(struct request_sock *req,
+				const struct sock *sk_listener,
+				struct sk_buff *skb)
+{
+	tcp_rsk(req)->is_mptcp = 1;
+
+	tcp_request_sock_ipv4_ops.init_req(req, sk_listener, skb);
+
+	subflow_init_req(req, sk_listener, skb);
+}
+
+#if IS_ENABLED(CONFIG_MPTCP_IPV6)
+static void subflow_v6_init_req(struct request_sock *req,
+				const struct sock *sk_listener,
+				struct sk_buff *skb)
+{
+	tcp_rsk(req)->is_mptcp = 1;
+
+	tcp_request_sock_ipv6_ops.init_req(req, sk_listener, skb);
+
+	subflow_init_req(req, sk_listener, skb);
+}
+#endif
+
+static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb)
+{
+	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
+
+	subflow->icsk_af_ops->sk_rx_dst_set(sk, skb);
+
+	if (subflow->conn && !subflow->conn_finished) {
+		pr_debug("subflow=%p, remote_key=%llu", mptcp_subflow_ctx(sk),
+			 subflow->remote_key);
+		mptcp_finish_connect(sk);
+		subflow->conn_finished = 1;
+	}
+}
+
+static struct request_sock_ops subflow_request_sock_ops;
+static struct tcp_request_sock_ops subflow_request_sock_ipv4_ops;
+
+static int subflow_v4_conn_request(struct sock *sk, struct sk_buff *skb)
+{
+	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
+
+	pr_debug("subflow=%p", subflow);
+
+	/* Never answer to SYNs sent to broadcast or multicast */
+	if (skb_rtable(skb)->rt_flags & (RTCF_BROADCAST | RTCF_MULTICAST))
+		goto drop;
+
+	return tcp_conn_request(&subflow_request_sock_ops,
+				&subflow_request_sock_ipv4_ops,
+				sk, skb);
+drop:
+	tcp_listendrop(sk);
+	return 0;
+}
+
+#if IS_ENABLED(CONFIG_MPTCP_IPV6)
+static struct tcp_request_sock_ops subflow_request_sock_ipv6_ops;
+static struct inet_connection_sock_af_ops subflow_v6_specific;
+static struct inet_connection_sock_af_ops subflow_v6m_specific;
+
+static int subflow_v6_conn_request(struct sock *sk, struct sk_buff *skb)
+{
+	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
+
+	pr_debug("subflow=%p", subflow);
+
+	if (skb->protocol == htons(ETH_P_IP))
+		return subflow_v4_conn_request(sk, skb);
+
+	if (!ipv6_unicast_destination(skb))
+		goto drop;
+
+	return tcp_conn_request(&subflow_request_sock_ops,
+				&subflow_request_sock_ipv6_ops, sk, skb);
+
+drop:
+	tcp_listendrop(sk);
+	return 0; /* don't send reset */
+}
+#endif
+
+static struct sock *subflow_syn_recv_sock(const struct sock *sk,
+					  struct sk_buff *skb,
+					  struct request_sock *req,
+					  struct dst_entry *dst,
+					  struct request_sock *req_unhash,
+					  bool *own_req)
+{
+	struct mptcp_subflow_context *listener = mptcp_subflow_ctx(sk);
+	struct sock *child;
+
+	pr_debug("listener=%p, req=%p, conn=%p", listener, req, listener->conn);
+
+	/* if the sk is MP_CAPABLE, we already received the client key */
+
+	child = listener->icsk_af_ops->syn_recv_sock(sk, skb, req, dst,
+						     req_unhash, own_req);
+
+	if (child && *own_req) {
+		if (!mptcp_subflow_ctx(child)) {
+			pr_debug("Closing child socket");
+			inet_sk_set_state(child, TCP_CLOSE);
+			sock_set_flag(child, SOCK_DEAD);
+			inet_csk_destroy_sock(child);
+			child = NULL;
+		}
+	}
+
+	return child;
+}
+
+static struct inet_connection_sock_af_ops subflow_specific;
+
+static struct inet_connection_sock_af_ops *
+subflow_default_af_ops(struct sock *sk)
+{
+#if IS_ENABLED(CONFIG_MPTCP_IPV6)
+	if (sk->sk_family == AF_INET6)
+		return &subflow_v6_specific;
+#endif
+	return &subflow_specific;
+}
+
+void mptcp_handle_ipv6_mapped(struct sock *sk, bool mapped)
+{
+#if IS_ENABLED(CONFIG_MPTCP_IPV6)
+	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
+	struct inet_connection_sock *icsk = inet_csk(sk);
+	struct inet_connection_sock_af_ops *target;
+
+	target = mapped ? &subflow_v6m_specific : subflow_default_af_ops(sk);
+
+	pr_debug("subflow=%p family=%d ops=%p target=%p mapped=%d",
+	         subflow, sk->sk_family, icsk->icsk_af_ops, target, mapped);
+
+	if (likely(icsk->icsk_af_ops == target))
+		return;
+
+	subflow->icsk_af_ops = icsk->icsk_af_ops;
+	icsk->icsk_af_ops = target;
+#endif
+}
+
 int mptcp_subflow_create_socket(struct sock *sk, struct socket **new_sock)
 {
 	struct mptcp_subflow_context *subflow;
@@ -22,7 +201,8 @@ int mptcp_subflow_create_socket(struct sock *sk, struct socket **new_sock)
 	struct socket *sf;
 	int err;
 
-	err = sock_create_kern(net, PF_INET, SOCK_STREAM, IPPROTO_TCP, &sf);
+	err = sock_create_kern(net, sk->sk_family, SOCK_STREAM, IPPROTO_TCP,
+			       &sf);
 	if (err)
 		return err;
 
@@ -60,6 +240,7 @@ static struct mptcp_subflow_context *subflow_create_ctx(struct sock *sk,
 		return NULL;
 
 	rcu_assign_pointer(icsk->icsk_ulp_data, ctx);
+	INIT_LIST_HEAD(&ctx->node);
 
 	pr_debug("subflow=%p", ctx);
 
@@ -70,6 +251,7 @@ static struct mptcp_subflow_context *subflow_create_ctx(struct sock *sk,
 
 static int subflow_ulp_init(struct sock *sk)
 {
+	struct inet_connection_sock *icsk = inet_csk(sk);
 	struct mptcp_subflow_context *ctx;
 	struct tcp_sock *tp = tcp_sk(sk);
 	int err = 0;
@@ -91,6 +273,8 @@ static int subflow_ulp_init(struct sock *sk)
 	pr_debug("subflow=%p, family=%d", ctx, sk->sk_family);
 
 	tp->is_mptcp = 1;
+	ctx->icsk_af_ops = icsk->icsk_af_ops;
+	icsk->icsk_af_ops = subflow_default_af_ops(sk);
 out:
 	return err;
 }
@@ -105,15 +289,97 @@ static void subflow_ulp_release(struct sock *sk)
 	kfree_rcu(ctx, rcu);
 }
 
+static void subflow_ulp_fallback(struct sock *sk)
+{
+	struct inet_connection_sock *icsk = inet_csk(sk);
+
+	icsk->icsk_ulp_ops = NULL;
+	rcu_assign_pointer(icsk->icsk_ulp_data, NULL);
+	tcp_sk(sk)->is_mptcp = 0;
+}
+
+static void subflow_ulp_clone(const struct request_sock *req,
+			      struct sock *newsk,
+			      const gfp_t priority)
+{
+	struct mptcp_subflow_request_sock *subflow_req = mptcp_subflow_rsk(req);
+	struct mptcp_subflow_context *old_ctx = mptcp_subflow_ctx(newsk);
+	struct mptcp_subflow_context *new_ctx;
+
+	if (!subflow_req->mp_capable) {
+		subflow_ulp_fallback(newsk);
+		return;
+	}
+
+	new_ctx = subflow_create_ctx(newsk, priority);
+	if (new_ctx == NULL) {
+		subflow_ulp_fallback(newsk);
+		return;
+	}
+
+	new_ctx->conn_finished = 1;
+	new_ctx->icsk_af_ops = old_ctx->icsk_af_ops;
+	new_ctx->mp_capable = 1;
+	new_ctx->fourth_ack = 1;
+	new_ctx->remote_key = subflow_req->remote_key;
+	new_ctx->local_key = subflow_req->local_key;
+}
+
 static struct tcp_ulp_ops subflow_ulp_ops __read_mostly = {
 	.name		= "mptcp",
 	.owner		= THIS_MODULE,
 	.init		= subflow_ulp_init,
 	.release	= subflow_ulp_release,
+	.clone		= subflow_ulp_clone,
 };
 
+static int subflow_ops_init(struct request_sock_ops *subflow_ops)
+{
+	subflow_ops->obj_size = sizeof(struct mptcp_subflow_request_sock);
+	subflow_ops->slab_name = "request_sock_subflow";
+
+	subflow_ops->slab = kmem_cache_create(subflow_ops->slab_name,
+					      subflow_ops->obj_size, 0,
+					      SLAB_ACCOUNT |
+					      SLAB_TYPESAFE_BY_RCU,
+					      NULL);
+	if (!subflow_ops->slab)
+		return -ENOMEM;
+
+	return 0;
+}
+
 void mptcp_subflow_init(void)
 {
+	subflow_request_sock_ops = tcp_request_sock_ops;
+	if (subflow_ops_init(&subflow_request_sock_ops) != 0)
+		panic("MPTCP: failed to init subflow request sock ops\n");
+
+	subflow_request_sock_ipv4_ops = tcp_request_sock_ipv4_ops;
+	subflow_request_sock_ipv4_ops.init_req = subflow_v4_init_req;
+
+	subflow_specific = ipv4_specific;
+	subflow_specific.conn_request = subflow_v4_conn_request;
+	subflow_specific.syn_recv_sock = subflow_syn_recv_sock;
+	subflow_specific.sk_rx_dst_set = subflow_finish_connect;
+
+#if IS_ENABLED(CONFIG_MPTCP_IPV6)
+	subflow_request_sock_ipv6_ops = tcp_request_sock_ipv6_ops;
+	subflow_request_sock_ipv6_ops.init_req = subflow_v6_init_req;
+
+	subflow_v6_specific = ipv6_specific;
+	subflow_v6_specific.conn_request = subflow_v6_conn_request;
+	subflow_v6_specific.syn_recv_sock = subflow_syn_recv_sock;
+	subflow_v6_specific.sk_rx_dst_set = subflow_finish_connect;
+
+	subflow_v6m_specific = subflow_v6_specific;
+	subflow_v6m_specific.queue_xmit = ipv4_specific.queue_xmit;
+	subflow_v6m_specific.send_check = ipv4_specific.send_check;
+	subflow_v6m_specific.net_header_len = ipv4_specific.net_header_len;
+	subflow_v6m_specific.mtu_reduced = ipv4_specific.mtu_reduced;
+	subflow_v6m_specific.net_frag_header_len = 0;
+#endif
+
 	if (tcp_register_ulp(&subflow_ulp_ops) != 0)
 		panic("MPTCP: failed to register subflows to ULP\n");
 }
-- 
cgit v1.2.3-59-g8ed1b


From cf7da0d66cc1a2a19fc5930bb746ffbb2d4cd1be Mon Sep 17 00:00:00 2001
From: Peter Krystad <peter.krystad@linux.intel.com>
Date: Tue, 21 Jan 2020 16:56:19 -0800
Subject: mptcp: Create SUBFLOW socket for incoming connections

Add subflow_request_sock type that extends tcp_request_sock
and add an is_mptcp flag to tcp_request_sock distinguish them.

Override the listen() and accept() methods of the MPTCP
socket proto_ops so they may act on the subflow socket.

Override the conn_request() and syn_recv_sock() handlers
in the inet_connection_sock to handle incoming MPTCP
SYNs and the ACK to the response SYN.

Add handling in tcp_output.c to add MP_CAPABLE to an outgoing
SYN-ACK response for a subflow_request_sock.

Co-developed-by: Davide Caratti <dcaratti@redhat.com>
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Co-developed-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Florian Westphal <fw@strlen.de>
Co-developed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Co-developed-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Peter Krystad <peter.krystad@linux.intel.com>
Signed-off-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/mptcp/protocol.c | 236 +++++++++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 231 insertions(+), 5 deletions(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index bdd58da1e4f6..e08a25eabcd5 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -14,6 +14,9 @@
 #include <net/inet_hashtables.h>
 #include <net/protocol.h>
 #include <net/tcp.h>
+#if IS_ENABLED(CONFIG_MPTCP_IPV6)
+#include <net/transp_v6.h>
+#endif
 #include <net/mptcp.h>
 #include "protocol.h"
 
@@ -212,6 +215,90 @@ static void mptcp_close(struct sock *sk, long timeout)
 	sk_common_release(sk);
 }
 
+static void mptcp_copy_inaddrs(struct sock *msk, const struct sock *ssk)
+{
+#if IS_ENABLED(CONFIG_MPTCP_IPV6)
+	const struct ipv6_pinfo *ssk6 = inet6_sk(ssk);
+	struct ipv6_pinfo *msk6 = inet6_sk(msk);
+
+	msk->sk_v6_daddr = ssk->sk_v6_daddr;
+	msk->sk_v6_rcv_saddr = ssk->sk_v6_rcv_saddr;
+
+	if (msk6 && ssk6) {
+		msk6->saddr = ssk6->saddr;
+		msk6->flow_label = ssk6->flow_label;
+	}
+#endif
+
+	inet_sk(msk)->inet_num = inet_sk(ssk)->inet_num;
+	inet_sk(msk)->inet_dport = inet_sk(ssk)->inet_dport;
+	inet_sk(msk)->inet_sport = inet_sk(ssk)->inet_sport;
+	inet_sk(msk)->inet_daddr = inet_sk(ssk)->inet_daddr;
+	inet_sk(msk)->inet_saddr = inet_sk(ssk)->inet_saddr;
+	inet_sk(msk)->inet_rcv_saddr = inet_sk(ssk)->inet_rcv_saddr;
+}
+
+static struct sock *mptcp_accept(struct sock *sk, int flags, int *err,
+				 bool kern)
+{
+	struct mptcp_sock *msk = mptcp_sk(sk);
+	struct socket *listener;
+	struct sock *newsk;
+
+	listener = __mptcp_nmpc_socket(msk);
+	if (WARN_ON_ONCE(!listener)) {
+		*err = -EINVAL;
+		return NULL;
+	}
+
+	pr_debug("msk=%p, listener=%p", msk, mptcp_subflow_ctx(listener->sk));
+	newsk = inet_csk_accept(listener->sk, flags, err, kern);
+	if (!newsk)
+		return NULL;
+
+	pr_debug("msk=%p, subflow is mptcp=%d", msk, sk_is_mptcp(newsk));
+
+	if (sk_is_mptcp(newsk)) {
+		struct mptcp_subflow_context *subflow;
+		struct sock *new_mptcp_sock;
+		struct sock *ssk = newsk;
+
+		subflow = mptcp_subflow_ctx(newsk);
+		lock_sock(sk);
+
+		local_bh_disable();
+		new_mptcp_sock = sk_clone_lock(sk, GFP_ATOMIC);
+		if (!new_mptcp_sock) {
+			*err = -ENOBUFS;
+			local_bh_enable();
+			release_sock(sk);
+			tcp_close(newsk, 0);
+			return NULL;
+		}
+
+		mptcp_init_sock(new_mptcp_sock);
+
+		msk = mptcp_sk(new_mptcp_sock);
+		msk->remote_key = subflow->remote_key;
+		msk->local_key = subflow->local_key;
+		msk->subflow = NULL;
+
+		newsk = new_mptcp_sock;
+		mptcp_copy_inaddrs(newsk, ssk);
+		list_add(&subflow->node, &msk->conn_list);
+
+		/* will be fully established at mptcp_stream_accept()
+		 * completion.
+		 */
+		inet_sk_state_store(new_mptcp_sock, TCP_SYN_RECV);
+		bh_unlock_sock(new_mptcp_sock);
+		local_bh_enable();
+		release_sock(sk);
+	}
+
+	return newsk;
+}
+
 static int mptcp_get_port(struct sock *sk, unsigned short snum)
 {
 	struct mptcp_sock *msk = mptcp_sk(sk);
@@ -246,12 +333,21 @@ void mptcp_finish_connect(struct sock *ssk)
 	WRITE_ONCE(msk->local_key, subflow->local_key);
 }
 
+static void mptcp_sock_graft(struct sock *sk, struct socket *parent)
+{
+	write_lock_bh(&sk->sk_callback_lock);
+	rcu_assign_pointer(sk->sk_wq, &parent->wq);
+	sk_set_socket(sk, parent);
+	sk->sk_uid = SOCK_INODE(parent)->i_uid;
+	write_unlock_bh(&sk->sk_callback_lock);
+}
+
 static struct proto mptcp_prot = {
 	.name		= "MPTCP",
 	.owner		= THIS_MODULE,
 	.init		= mptcp_init_sock,
 	.close		= mptcp_close,
-	.accept		= inet_csk_accept,
+	.accept		= mptcp_accept,
 	.shutdown	= tcp_shutdown,
 	.sendmsg	= mptcp_sendmsg,
 	.recvmsg	= mptcp_recvmsg,
@@ -266,10 +362,7 @@ static int mptcp_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
 {
 	struct mptcp_sock *msk = mptcp_sk(sock->sk);
 	struct socket *ssock;
-	int err = -ENOTSUPP;
-
-	if (uaddr->sa_family != AF_INET) // @@ allow only IPv4 for now
-		return err;
+	int err;
 
 	lock_sock(sock->sk);
 	ssock = __mptcp_socket_create(msk, MPTCP_SAME_STATE);
@@ -279,6 +372,8 @@ static int mptcp_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
 	}
 
 	err = ssock->ops->bind(ssock, uaddr, addr_len);
+	if (!err)
+		mptcp_copy_inaddrs(sock->sk, ssock->sk);
 
 unlock:
 	release_sock(sock->sk);
@@ -299,14 +394,139 @@ static int mptcp_stream_connect(struct socket *sock, struct sockaddr *uaddr,
 		goto unlock;
 	}
 
+#ifdef CONFIG_TCP_MD5SIG
+	/* no MPTCP if MD5SIG is enabled on this socket or we may run out of
+	 * TCP option space.
+	 */
+	if (rcu_access_pointer(tcp_sk(ssock->sk)->md5sig_info))
+		mptcp_subflow_ctx(ssock->sk)->request_mptcp = 0;
+#endif
+
 	err = ssock->ops->connect(ssock, uaddr, addr_len, flags);
 	inet_sk_state_store(sock->sk, inet_sk_state_load(ssock->sk));
+	mptcp_copy_inaddrs(sock->sk, ssock->sk);
 
 unlock:
 	release_sock(sock->sk);
 	return err;
 }
 
+static int mptcp_v4_getname(struct socket *sock, struct sockaddr *uaddr,
+			    int peer)
+{
+	if (sock->sk->sk_prot == &tcp_prot) {
+		/* we are being invoked from __sys_accept4, after
+		 * mptcp_accept() has just accepted a non-mp-capable
+		 * flow: sk is a tcp_sk, not an mptcp one.
+		 *
+		 * Hand the socket over to tcp so all further socket ops
+		 * bypass mptcp.
+		 */
+		sock->ops = &inet_stream_ops;
+	}
+
+	return inet_getname(sock, uaddr, peer);
+}
+
+#if IS_ENABLED(CONFIG_MPTCP_IPV6)
+static int mptcp_v6_getname(struct socket *sock, struct sockaddr *uaddr,
+			    int peer)
+{
+	if (sock->sk->sk_prot == &tcpv6_prot) {
+		/* we are being invoked from __sys_accept4 after
+		 * mptcp_accept() has accepted a non-mp-capable
+		 * subflow: sk is a tcp_sk, not mptcp.
+		 *
+		 * Hand the socket over to tcp so all further
+		 * socket ops bypass mptcp.
+		 */
+		sock->ops = &inet6_stream_ops;
+	}
+
+	return inet6_getname(sock, uaddr, peer);
+}
+#endif
+
+static int mptcp_listen(struct socket *sock, int backlog)
+{
+	struct mptcp_sock *msk = mptcp_sk(sock->sk);
+	struct socket *ssock;
+	int err;
+
+	pr_debug("msk=%p", msk);
+
+	lock_sock(sock->sk);
+	ssock = __mptcp_socket_create(msk, TCP_LISTEN);
+	if (IS_ERR(ssock)) {
+		err = PTR_ERR(ssock);
+		goto unlock;
+	}
+
+	err = ssock->ops->listen(ssock, backlog);
+	inet_sk_state_store(sock->sk, inet_sk_state_load(ssock->sk));
+	if (!err)
+		mptcp_copy_inaddrs(sock->sk, ssock->sk);
+
+unlock:
+	release_sock(sock->sk);
+	return err;
+}
+
+static bool is_tcp_proto(const struct proto *p)
+{
+#if IS_ENABLED(CONFIG_MPTCP_IPV6)
+	return p == &tcp_prot || p == &tcpv6_prot;
+#else
+	return p == &tcp_prot;
+#endif
+}
+
+static int mptcp_stream_accept(struct socket *sock, struct socket *newsock,
+			       int flags, bool kern)
+{
+	struct mptcp_sock *msk = mptcp_sk(sock->sk);
+	struct socket *ssock;
+	int err;
+
+	pr_debug("msk=%p", msk);
+
+	lock_sock(sock->sk);
+	if (sock->sk->sk_state != TCP_LISTEN)
+		goto unlock_fail;
+
+	ssock = __mptcp_nmpc_socket(msk);
+	if (!ssock)
+		goto unlock_fail;
+
+	sock_hold(ssock->sk);
+	release_sock(sock->sk);
+
+	err = ssock->ops->accept(sock, newsock, flags, kern);
+	if (err == 0 && !is_tcp_proto(newsock->sk->sk_prot)) {
+		struct mptcp_sock *msk = mptcp_sk(newsock->sk);
+		struct mptcp_subflow_context *subflow;
+
+		/* set ssk->sk_socket of accept()ed flows to mptcp socket.
+		 * This is needed so NOSPACE flag can be set from tcp stack.
+		 */
+		list_for_each_entry(subflow, &msk->conn_list, node) {
+			struct sock *ssk = mptcp_subflow_tcp_sock(subflow);
+
+			if (!ssk->sk_socket)
+				mptcp_sock_graft(ssk, newsock);
+		}
+
+		inet_sk_state_store(newsock->sk, TCP_ESTABLISHED);
+	}
+
+	sock_put(ssock->sk);
+	return err;
+
+unlock_fail:
+	release_sock(sock->sk);
+	return -EINVAL;
+}
+
 static __poll_t mptcp_poll(struct file *file, struct socket *sock,
 			   struct poll_table_struct *wait)
 {
@@ -332,6 +552,9 @@ void __init mptcp_init(void)
 	mptcp_stream_ops.bind = mptcp_bind;
 	mptcp_stream_ops.connect = mptcp_stream_connect;
 	mptcp_stream_ops.poll = mptcp_poll;
+	mptcp_stream_ops.accept = mptcp_stream_accept;
+	mptcp_stream_ops.getname = mptcp_v4_getname;
+	mptcp_stream_ops.listen = mptcp_listen;
 
 	mptcp_subflow_init();
 
@@ -371,6 +594,9 @@ int mptcpv6_init(void)
 	mptcp_v6_stream_ops.bind = mptcp_bind;
 	mptcp_v6_stream_ops.connect = mptcp_stream_connect;
 	mptcp_v6_stream_ops.poll = mptcp_poll;
+	mptcp_v6_stream_ops.accept = mptcp_stream_accept;
+	mptcp_v6_stream_ops.getname = mptcp_v6_getname;
+	mptcp_v6_stream_ops.listen = mptcp_listen;
 
 	err = inet6_register_protosw(&mptcp_v6_protosw);
 	if (err)
-- 
cgit v1.2.3-59-g8ed1b


From 79c0949e9a09f6a14a6dd18dc8396029423f9b68 Mon Sep 17 00:00:00 2001
From: Peter Krystad <peter.krystad@linux.intel.com>
Date: Tue, 21 Jan 2020 16:56:20 -0800
Subject: mptcp: Add key generation and token tree

Generate the local keys, IDSN, and token when creating a new socket.
Introduce the token tree to track all tokens in use using a radix tree
with the MPTCP token itself as the index.

Override the rebuild_header callback in inet_connection_sock_af_ops for
creating the local key on a new outgoing connection.

Override the init_req callback of tcp_request_sock_ops for creating the
local key on a new incoming connection.

Will be used to obtain the MPTCP parent socket to handle incoming joins.

Co-developed-by: Davide Caratti <dcaratti@redhat.com>
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Co-developed-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Florian Westphal <fw@strlen.de>
Co-developed-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Peter Krystad <peter.krystad@linux.intel.com>
Signed-off-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/mptcp/Makefile   |   2 +-
 net/mptcp/crypto.c   | 122 ++++++++++++++++++++++++++++++++
 net/mptcp/protocol.c |  16 +++++
 net/mptcp/protocol.h |  32 +++++++++
 net/mptcp/subflow.c  |  69 ++++++++++++++++--
 net/mptcp/token.c    | 195 +++++++++++++++++++++++++++++++++++++++++++++++++++
 6 files changed, 428 insertions(+), 8 deletions(-)
 create mode 100644 net/mptcp/crypto.c
 create mode 100644 net/mptcp/token.c

diff --git a/net/mptcp/Makefile b/net/mptcp/Makefile
index e1ee5aade8b0..178ae81d8b66 100644
--- a/net/mptcp/Makefile
+++ b/net/mptcp/Makefile
@@ -1,4 +1,4 @@
 # SPDX-License-Identifier: GPL-2.0
 obj-$(CONFIG_MPTCP) += mptcp.o
 
-mptcp-y := protocol.o subflow.o options.o
+mptcp-y := protocol.o subflow.o options.o token.o crypto.o
diff --git a/net/mptcp/crypto.c b/net/mptcp/crypto.c
new file mode 100644
index 000000000000..bbd6d01af211
--- /dev/null
+++ b/net/mptcp/crypto.c
@@ -0,0 +1,122 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Multipath TCP cryptographic functions
+ * Copyright (c) 2017 - 2019, Intel Corporation.
+ *
+ * Note: This code is based on mptcp_ctrl.c, mptcp_ipv4.c, and
+ *       mptcp_ipv6 from multipath-tcp.org, authored by:
+ *
+ *       Sébastien Barré <sebastien.barre@uclouvain.be>
+ *       Christoph Paasch <christoph.paasch@uclouvain.be>
+ *       Jaakko Korkeaniemi <jaakko.korkeaniemi@aalto.fi>
+ *       Gregory Detal <gregory.detal@uclouvain.be>
+ *       Fabien Duchêne <fabien.duchene@uclouvain.be>
+ *       Andreas Seelinger <Andreas.Seelinger@rwth-aachen.de>
+ *       Lavkesh Lahngir <lavkesh51@gmail.com>
+ *       Andreas Ripke <ripke@neclab.eu>
+ *       Vlad Dogaru <vlad.dogaru@intel.com>
+ *       Octavian Purdila <octavian.purdila@intel.com>
+ *       John Ronan <jronan@tssg.org>
+ *       Catalin Nicutar <catalin.nicutar@gmail.com>
+ *       Brandon Heller <brandonh@stanford.edu>
+ */
+
+#include <linux/kernel.h>
+#include <linux/cryptohash.h>
+#include <asm/unaligned.h>
+
+#include "protocol.h"
+
+struct sha1_state {
+	u32 workspace[SHA_WORKSPACE_WORDS];
+	u32 digest[SHA_DIGEST_WORDS];
+	unsigned int count;
+};
+
+static void sha1_init(struct sha1_state *state)
+{
+	sha_init(state->digest);
+	state->count = 0;
+}
+
+static void sha1_update(struct sha1_state *state, u8 *input)
+{
+	sha_transform(state->digest, input, state->workspace);
+	state->count += SHA_MESSAGE_BYTES;
+}
+
+static void sha1_pad_final(struct sha1_state *state, u8 *input,
+			   unsigned int length, __be32 *mptcp_hashed_key)
+{
+	int i;
+
+	input[length] = 0x80;
+	memset(&input[length + 1], 0, SHA_MESSAGE_BYTES - length - 9);
+	put_unaligned_be64((length + state->count) << 3,
+			   &input[SHA_MESSAGE_BYTES - 8]);
+
+	sha_transform(state->digest, input, state->workspace);
+	for (i = 0; i < SHA_DIGEST_WORDS; ++i)
+		put_unaligned_be32(state->digest[i], &mptcp_hashed_key[i]);
+
+	memzero_explicit(state->workspace, SHA_WORKSPACE_WORDS << 2);
+}
+
+void mptcp_crypto_key_sha(u64 key, u32 *token, u64 *idsn)
+{
+	__be32 mptcp_hashed_key[SHA_DIGEST_WORDS];
+	u8 input[SHA_MESSAGE_BYTES];
+	struct sha1_state state;
+
+	sha1_init(&state);
+	put_unaligned_be64(key, input);
+	sha1_pad_final(&state, input, 8, mptcp_hashed_key);
+
+	if (token)
+		*token = be32_to_cpu(mptcp_hashed_key[0]);
+	if (idsn)
+		*idsn = be64_to_cpu(*((__be64 *)&mptcp_hashed_key[3]));
+}
+
+void mptcp_crypto_hmac_sha(u64 key1, u64 key2, u32 nonce1, u32 nonce2,
+			   u32 *hash_out)
+{
+	u8 input[SHA_MESSAGE_BYTES * 2];
+	struct sha1_state state;
+	u8 key1be[8];
+	u8 key2be[8];
+	int i;
+
+	put_unaligned_be64(key1, key1be);
+	put_unaligned_be64(key2, key2be);
+
+	/* Generate key xored with ipad */
+	memset(input, 0x36, SHA_MESSAGE_BYTES);
+	for (i = 0; i < 8; i++)
+		input[i] ^= key1be[i];
+	for (i = 0; i < 8; i++)
+		input[i + 8] ^= key2be[i];
+
+	put_unaligned_be32(nonce1, &input[SHA_MESSAGE_BYTES]);
+	put_unaligned_be32(nonce2, &input[SHA_MESSAGE_BYTES + 4]);
+
+	sha1_init(&state);
+	sha1_update(&state, input);
+
+	/* emit sha256(K1 || msg) on the second input block, so we can
+	 * reuse 'input' for the last hashing
+	 */
+	sha1_pad_final(&state, &input[SHA_MESSAGE_BYTES], 8,
+		       (__force __be32 *)&input[SHA_MESSAGE_BYTES]);
+
+	/* Prepare second part of hmac */
+	memset(input, 0x5C, SHA_MESSAGE_BYTES);
+	for (i = 0; i < 8; i++)
+		input[i] ^= key1be[i];
+	for (i = 0; i < 8; i++)
+		input[i + 8] ^= key2be[i];
+
+	sha1_init(&state);
+	sha1_update(&state, input);
+	sha1_pad_final(&state, &input[SHA_MESSAGE_BYTES], SHA_DIGEST_WORDS << 2,
+		       (__be32 *)hash_out);
+}
diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index e08a25eabcd5..3f66b6a3bb28 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -201,6 +201,7 @@ static void mptcp_close(struct sock *sk, long timeout)
 	struct mptcp_subflow_context *subflow, *tmp;
 	struct mptcp_sock *msk = mptcp_sk(sk);
 
+	mptcp_token_destroy(msk->token);
 	inet_sk_state_store(sk, TCP_CLOSE);
 
 	lock_sock(sk);
@@ -281,8 +282,10 @@ static struct sock *mptcp_accept(struct sock *sk, int flags, int *err,
 		msk = mptcp_sk(new_mptcp_sock);
 		msk->remote_key = subflow->remote_key;
 		msk->local_key = subflow->local_key;
+		msk->token = subflow->token;
 		msk->subflow = NULL;
 
+		mptcp_token_update_accept(newsk, new_mptcp_sock);
 		newsk = new_mptcp_sock;
 		mptcp_copy_inaddrs(newsk, ssk);
 		list_add(&subflow->node, &msk->conn_list);
@@ -299,6 +302,10 @@ static struct sock *mptcp_accept(struct sock *sk, int flags, int *err,
 	return newsk;
 }
 
+static void mptcp_destroy(struct sock *sk)
+{
+}
+
 static int mptcp_get_port(struct sock *sk, unsigned short snum)
 {
 	struct mptcp_sock *msk = mptcp_sk(sk);
@@ -331,6 +338,7 @@ void mptcp_finish_connect(struct sock *ssk)
 	 */
 	WRITE_ONCE(msk->remote_key, subflow->remote_key);
 	WRITE_ONCE(msk->local_key, subflow->local_key);
+	WRITE_ONCE(msk->token, subflow->token);
 }
 
 static void mptcp_sock_graft(struct sock *sk, struct socket *parent)
@@ -349,6 +357,7 @@ static struct proto mptcp_prot = {
 	.close		= mptcp_close,
 	.accept		= mptcp_accept,
 	.shutdown	= tcp_shutdown,
+	.destroy	= mptcp_destroy,
 	.sendmsg	= mptcp_sendmsg,
 	.recvmsg	= mptcp_recvmsg,
 	.hash		= inet_hash,
@@ -568,6 +577,12 @@ void __init mptcp_init(void)
 static struct proto_ops mptcp_v6_stream_ops;
 static struct proto mptcp_v6_prot;
 
+static void mptcp_v6_destroy(struct sock *sk)
+{
+	mptcp_destroy(sk);
+	inet6_destroy_sock(sk);
+}
+
 static struct inet_protosw mptcp_v6_protosw = {
 	.type		= SOCK_STREAM,
 	.protocol	= IPPROTO_MPTCP,
@@ -583,6 +598,7 @@ int mptcpv6_init(void)
 	mptcp_v6_prot = mptcp_prot;
 	strcpy(mptcp_v6_prot.name, "MPTCPv6");
 	mptcp_v6_prot.slab = NULL;
+	mptcp_v6_prot.destroy = mptcp_v6_destroy;
 	mptcp_v6_prot.obj_size = sizeof(struct mptcp_sock) +
 				 sizeof(struct ipv6_pinfo);
 
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index bd66e7415515..5f43fa0275c0 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -7,6 +7,10 @@
 #ifndef __MPTCP_PROTOCOL_H
 #define __MPTCP_PROTOCOL_H
 
+#include <linux/random.h>
+#include <net/tcp.h>
+#include <net/inet_connection_sock.h>
+
 #define MPTCP_SUPPORTED_VERSION	0
 
 /* MPTCP option bits */
@@ -42,6 +46,7 @@ struct mptcp_sock {
 	struct inet_connection_sock sk;
 	u64		local_key;
 	u64		remote_key;
+	u32		token;
 	struct list_head conn_list;
 	struct socket	*subflow; /* outgoing connect/listener/!mp_capable */
 };
@@ -61,6 +66,8 @@ struct mptcp_subflow_request_sock {
 		backup : 1;
 	u64	local_key;
 	u64	remote_key;
+	u64	idsn;
+	u32	token;
 };
 
 static inline struct mptcp_subflow_request_sock *
@@ -74,6 +81,8 @@ struct mptcp_subflow_context {
 	struct	list_head node;/* conn_list of subflows */
 	u64	local_key;
 	u64	remote_key;
+	u64	idsn;
+	u32	token;
 	u32	request_mptcp : 1,  /* send MP_CAPABLE */
 		mp_capable : 1,	    /* remote is MPTCP capable */
 		fourth_ack : 1,	    /* send initial DSS */
@@ -112,4 +121,27 @@ void mptcp_get_options(const struct sk_buff *skb,
 
 void mptcp_finish_connect(struct sock *sk);
 
+int mptcp_token_new_request(struct request_sock *req);
+void mptcp_token_destroy_request(u32 token);
+int mptcp_token_new_connect(struct sock *sk);
+int mptcp_token_new_accept(u32 token);
+void mptcp_token_update_accept(struct sock *sk, struct sock *conn);
+void mptcp_token_destroy(u32 token);
+
+void mptcp_crypto_key_sha(u64 key, u32 *token, u64 *idsn);
+static inline void mptcp_crypto_key_gen_sha(u64 *key, u32 *token, u64 *idsn)
+{
+	/* we might consider a faster version that computes the key as a
+	 * hash of some information available in the MPTCP socket. Use
+	 * random data at the moment, as it's probably the safest option
+	 * in case multiple sockets are opened in different namespaces at
+	 * the same time.
+	 */
+	get_random_bytes(key, sizeof(u64));
+	mptcp_crypto_key_sha(*key, token, idsn);
+}
+
+void mptcp_crypto_hmac_sha(u64 key1, u64 key2, u32 nonce1, u32 nonce2,
+			   u32 *hash_out);
+
 #endif /* __MPTCP_PROTOCOL_H */
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index df3192305967..89b91bc7a831 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -4,6 +4,8 @@
  * Copyright (c) 2017 - 2019, Intel Corporation.
  */
 
+#define pr_fmt(fmt) "MPTCP: " fmt
+
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/netdevice.h>
@@ -18,6 +20,33 @@
 #include <net/mptcp.h>
 #include "protocol.h"
 
+static int subflow_rebuild_header(struct sock *sk)
+{
+	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
+	int err = 0;
+
+	if (subflow->request_mptcp && !subflow->token) {
+		pr_debug("subflow=%p", sk);
+		err = mptcp_token_new_connect(sk);
+	}
+
+	if (err)
+		return err;
+
+	return subflow->icsk_af_ops->rebuild_header(sk);
+}
+
+static void subflow_req_destructor(struct request_sock *req)
+{
+	struct mptcp_subflow_request_sock *subflow_req = mptcp_subflow_rsk(req);
+
+	pr_debug("subflow_req=%p", subflow_req);
+
+	if (subflow_req->mp_capable)
+		mptcp_token_destroy_request(subflow_req->token);
+	tcp_request_sock_ops.destructor(req);
+}
+
 static void subflow_init_req(struct request_sock *req,
 			     const struct sock *sk_listener,
 			     struct sk_buff *skb)
@@ -42,7 +71,12 @@ static void subflow_init_req(struct request_sock *req,
 #endif
 
 	if (rx_opt.mptcp.mp_capable && listener->request_mptcp) {
-		subflow_req->mp_capable = 1;
+		int err;
+
+		err = mptcp_token_new_request(req);
+		if (err == 0)
+			subflow_req->mp_capable = 1;
+
 		subflow_req->remote_key = rx_opt.mptcp.sndr_key;
 	}
 }
@@ -150,16 +184,28 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,
 						     req_unhash, own_req);
 
 	if (child && *own_req) {
-		if (!mptcp_subflow_ctx(child)) {
-			pr_debug("Closing child socket");
-			inet_sk_set_state(child, TCP_CLOSE);
-			sock_set_flag(child, SOCK_DEAD);
-			inet_csk_destroy_sock(child);
-			child = NULL;
+		struct mptcp_subflow_context *ctx = mptcp_subflow_ctx(child);
+
+		/* we have null ctx on TCP fallback, not fatal on MPC
+		 * handshake
+		 */
+		if (!ctx)
+			return child;
+
+		if (ctx->mp_capable) {
+			if (mptcp_token_new_accept(ctx->token))
+				goto close_child;
 		}
 	}
 
 	return child;
+
+close_child:
+	pr_debug("closing child socket");
+	tcp_send_active_reset(child, GFP_ATOMIC);
+	inet_csk_prepare_forced_close(child);
+	tcp_done(child);
+	return NULL;
 }
 
 static struct inet_connection_sock_af_ops subflow_specific;
@@ -224,6 +270,7 @@ int mptcp_subflow_create_socket(struct sock *sk, struct socket **new_sock)
 	pr_debug("subflow=%p", subflow);
 
 	*new_sock = sf;
+	sock_hold(sk);
 	subflow->conn = sk;
 
 	return 0;
@@ -286,6 +333,9 @@ static void subflow_ulp_release(struct sock *sk)
 	if (!ctx)
 		return;
 
+	if (ctx->conn)
+		sock_put(ctx->conn);
+
 	kfree_rcu(ctx, rcu);
 }
 
@@ -323,6 +373,7 @@ static void subflow_ulp_clone(const struct request_sock *req,
 	new_ctx->fourth_ack = 1;
 	new_ctx->remote_key = subflow_req->remote_key;
 	new_ctx->local_key = subflow_req->local_key;
+	new_ctx->token = subflow_req->token;
 }
 
 static struct tcp_ulp_ops subflow_ulp_ops __read_mostly = {
@@ -346,6 +397,8 @@ static int subflow_ops_init(struct request_sock_ops *subflow_ops)
 	if (!subflow_ops->slab)
 		return -ENOMEM;
 
+	subflow_ops->destructor = subflow_req_destructor;
+
 	return 0;
 }
 
@@ -362,6 +415,7 @@ void mptcp_subflow_init(void)
 	subflow_specific.conn_request = subflow_v4_conn_request;
 	subflow_specific.syn_recv_sock = subflow_syn_recv_sock;
 	subflow_specific.sk_rx_dst_set = subflow_finish_connect;
+	subflow_specific.rebuild_header = subflow_rebuild_header;
 
 #if IS_ENABLED(CONFIG_MPTCP_IPV6)
 	subflow_request_sock_ipv6_ops = tcp_request_sock_ipv6_ops;
@@ -371,6 +425,7 @@ void mptcp_subflow_init(void)
 	subflow_v6_specific.conn_request = subflow_v6_conn_request;
 	subflow_v6_specific.syn_recv_sock = subflow_syn_recv_sock;
 	subflow_v6_specific.sk_rx_dst_set = subflow_finish_connect;
+	subflow_v6_specific.rebuild_header = subflow_rebuild_header;
 
 	subflow_v6m_specific = subflow_v6_specific;
 	subflow_v6m_specific.queue_xmit = ipv4_specific.queue_xmit;
diff --git a/net/mptcp/token.c b/net/mptcp/token.c
new file mode 100644
index 000000000000..84d887806090
--- /dev/null
+++ b/net/mptcp/token.c
@@ -0,0 +1,195 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Multipath TCP token management
+ * Copyright (c) 2017 - 2019, Intel Corporation.
+ *
+ * Note: This code is based on mptcp_ctrl.c from multipath-tcp.org,
+ *       authored by:
+ *
+ *       Sébastien Barré <sebastien.barre@uclouvain.be>
+ *       Christoph Paasch <christoph.paasch@uclouvain.be>
+ *       Jaakko Korkeaniemi <jaakko.korkeaniemi@aalto.fi>
+ *       Gregory Detal <gregory.detal@uclouvain.be>
+ *       Fabien Duchêne <fabien.duchene@uclouvain.be>
+ *       Andreas Seelinger <Andreas.Seelinger@rwth-aachen.de>
+ *       Lavkesh Lahngir <lavkesh51@gmail.com>
+ *       Andreas Ripke <ripke@neclab.eu>
+ *       Vlad Dogaru <vlad.dogaru@intel.com>
+ *       Octavian Purdila <octavian.purdila@intel.com>
+ *       John Ronan <jronan@tssg.org>
+ *       Catalin Nicutar <catalin.nicutar@gmail.com>
+ *       Brandon Heller <brandonh@stanford.edu>
+ */
+
+#define pr_fmt(fmt) "MPTCP: " fmt
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/radix-tree.h>
+#include <linux/ip.h>
+#include <linux/tcp.h>
+#include <net/sock.h>
+#include <net/inet_common.h>
+#include <net/protocol.h>
+#include <net/mptcp.h>
+#include "protocol.h"
+
+static RADIX_TREE(token_tree, GFP_ATOMIC);
+static RADIX_TREE(token_req_tree, GFP_ATOMIC);
+static DEFINE_SPINLOCK(token_tree_lock);
+static int token_used __read_mostly;
+
+/**
+ * mptcp_token_new_request - create new key/idsn/token for subflow_request
+ * @req - the request socket
+ *
+ * This function is called when a new mptcp connection is coming in.
+ *
+ * It creates a unique token to identify the new mptcp connection,
+ * a secret local key and the initial data sequence number (idsn).
+ *
+ * Returns 0 on success.
+ */
+int mptcp_token_new_request(struct request_sock *req)
+{
+	struct mptcp_subflow_request_sock *subflow_req = mptcp_subflow_rsk(req);
+	int err;
+
+	while (1) {
+		u32 token;
+
+		mptcp_crypto_key_gen_sha(&subflow_req->local_key,
+					 &subflow_req->token,
+					 &subflow_req->idsn);
+		pr_debug("req=%p local_key=%llu, token=%u, idsn=%llu\n",
+			 req, subflow_req->local_key, subflow_req->token,
+			 subflow_req->idsn);
+
+		token = subflow_req->token;
+		spin_lock_bh(&token_tree_lock);
+		if (!radix_tree_lookup(&token_req_tree, token) &&
+		    !radix_tree_lookup(&token_tree, token))
+			break;
+		spin_unlock_bh(&token_tree_lock);
+	}
+
+	err = radix_tree_insert(&token_req_tree,
+				subflow_req->token, &token_used);
+	spin_unlock_bh(&token_tree_lock);
+	return err;
+}
+
+/**
+ * mptcp_token_new_connect - create new key/idsn/token for subflow
+ * @sk - the socket that will initiate a connection
+ *
+ * This function is called when a new outgoing mptcp connection is
+ * initiated.
+ *
+ * It creates a unique token to identify the new mptcp connection,
+ * a secret local key and the initial data sequence number (idsn).
+ *
+ * On success, the mptcp connection can be found again using
+ * the computed token at a later time, this is needed to process
+ * join requests.
+ *
+ * returns 0 on success.
+ */
+int mptcp_token_new_connect(struct sock *sk)
+{
+	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
+	struct sock *mptcp_sock = subflow->conn;
+	int err;
+
+	while (1) {
+		u32 token;
+
+		mptcp_crypto_key_gen_sha(&subflow->local_key, &subflow->token,
+					 &subflow->idsn);
+
+		pr_debug("ssk=%p, local_key=%llu, token=%u, idsn=%llu\n",
+			 sk, subflow->local_key, subflow->token, subflow->idsn);
+
+		token = subflow->token;
+		spin_lock_bh(&token_tree_lock);
+		if (!radix_tree_lookup(&token_req_tree, token) &&
+		    !radix_tree_lookup(&token_tree, token))
+			break;
+		spin_unlock_bh(&token_tree_lock);
+	}
+	err = radix_tree_insert(&token_tree, subflow->token, mptcp_sock);
+	spin_unlock_bh(&token_tree_lock);
+
+	return err;
+}
+
+/**
+ * mptcp_token_new_accept - insert token for later processing
+ * @token: the token to insert to the tree
+ *
+ * Called when a SYN packet creates a new logical connection, i.e.
+ * is not a join request.
+ *
+ * We don't have an mptcp socket yet at that point.
+ * This is paired with mptcp_token_update_accept, called on accept().
+ */
+int mptcp_token_new_accept(u32 token)
+{
+	int err;
+
+	spin_lock_bh(&token_tree_lock);
+	err = radix_tree_insert(&token_tree, token, &token_used);
+	spin_unlock_bh(&token_tree_lock);
+
+	return err;
+}
+
+/**
+ * mptcp_token_update_accept - update token to map to mptcp socket
+ * @conn: the new struct mptcp_sock
+ * @sk: the initial subflow for this mptcp socket
+ *
+ * Called when the first mptcp socket is created on accept to
+ * refresh the dummy mapping (done to reserve the token) with
+ * the mptcp_socket structure that wasn't allocated before.
+ */
+void mptcp_token_update_accept(struct sock *sk, struct sock *conn)
+{
+	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
+	void __rcu **slot;
+
+	spin_lock_bh(&token_tree_lock);
+	slot = radix_tree_lookup_slot(&token_tree, subflow->token);
+	WARN_ON_ONCE(!slot);
+	if (slot) {
+		WARN_ON_ONCE(rcu_access_pointer(*slot) != &token_used);
+		radix_tree_replace_slot(&token_tree, slot, conn);
+	}
+	spin_unlock_bh(&token_tree_lock);
+}
+
+/**
+ * mptcp_token_destroy_request - remove mptcp connection/token
+ * @token - token of mptcp connection to remove
+ *
+ * Remove not-yet-fully-established incoming connection identified
+ * by @token.
+ */
+void mptcp_token_destroy_request(u32 token)
+{
+	spin_lock_bh(&token_tree_lock);
+	radix_tree_delete(&token_req_tree, token);
+	spin_unlock_bh(&token_tree_lock);
+}
+
+/**
+ * mptcp_token_destroy - remove mptcp connection/token
+ * @token - token of mptcp connection to remove
+ *
+ * Remove the connection identified by @token.
+ */
+void mptcp_token_destroy(u32 token)
+{
+	spin_lock_bh(&token_tree_lock);
+	radix_tree_delete(&token_tree, token);
+	spin_unlock_bh(&token_tree_lock);
+}
-- 
cgit v1.2.3-59-g8ed1b


From 214984901aaf50cc52eebeabf14017edeb3a845f Mon Sep 17 00:00:00 2001
From: Peter Krystad <peter.krystad@linux.intel.com>
Date: Tue, 21 Jan 2020 16:56:21 -0800
Subject: mptcp: Add shutdown() socket operation

Call shutdown on all subflows in use on the given socket, or on the
fallback socket.

Co-developed-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Peter Krystad <peter.krystad@linux.intel.com>
Signed-off-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/mptcp/protocol.c | 66 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 66 insertions(+)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 3f66b6a3bb28..249b2506a66a 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -196,6 +196,29 @@ static int mptcp_init_sock(struct sock *sk)
 	return 0;
 }
 
+static void mptcp_subflow_shutdown(struct sock *ssk, int how)
+{
+	lock_sock(ssk);
+
+	switch (ssk->sk_state) {
+	case TCP_LISTEN:
+		if (!(how & RCV_SHUTDOWN))
+			break;
+		/* fall through */
+	case TCP_SYN_SENT:
+		tcp_disconnect(ssk, O_NONBLOCK);
+		break;
+	default:
+		ssk->sk_shutdown |= how;
+		tcp_shutdown(ssk, how);
+		break;
+	}
+
+	/* Wake up anyone sleeping in poll. */
+	ssk->sk_state_change(ssk);
+	release_sock(ssk);
+}
+
 static void mptcp_close(struct sock *sk, long timeout)
 {
 	struct mptcp_subflow_context *subflow, *tmp;
@@ -273,6 +296,7 @@ static struct sock *mptcp_accept(struct sock *sk, int flags, int *err,
 			*err = -ENOBUFS;
 			local_bh_enable();
 			release_sock(sk);
+			mptcp_subflow_shutdown(newsk, SHUT_RDWR + 1);
 			tcp_close(newsk, 0);
 			return NULL;
 		}
@@ -544,6 +568,46 @@ static __poll_t mptcp_poll(struct file *file, struct socket *sock,
 	return mask;
 }
 
+static int mptcp_shutdown(struct socket *sock, int how)
+{
+	struct mptcp_sock *msk = mptcp_sk(sock->sk);
+	struct mptcp_subflow_context *subflow;
+	int ret = 0;
+
+	pr_debug("sk=%p, how=%d", msk, how);
+
+	lock_sock(sock->sk);
+
+	if (how == SHUT_WR || how == SHUT_RDWR)
+		inet_sk_state_store(sock->sk, TCP_FIN_WAIT1);
+
+	how++;
+
+	if ((how & ~SHUTDOWN_MASK) || !how) {
+		ret = -EINVAL;
+		goto out_unlock;
+	}
+
+	if (sock->state == SS_CONNECTING) {
+		if ((1 << sock->sk->sk_state) &
+		    (TCPF_SYN_SENT | TCPF_SYN_RECV | TCPF_CLOSE))
+			sock->state = SS_DISCONNECTING;
+		else
+			sock->state = SS_CONNECTED;
+	}
+
+	mptcp_for_each_subflow(msk, subflow) {
+		struct sock *tcp_sk = mptcp_subflow_tcp_sock(subflow);
+
+		mptcp_subflow_shutdown(tcp_sk, how);
+	}
+
+out_unlock:
+	release_sock(sock->sk);
+
+	return ret;
+}
+
 static struct proto_ops mptcp_stream_ops;
 
 static struct inet_protosw mptcp_protosw = {
@@ -564,6 +628,7 @@ void __init mptcp_init(void)
 	mptcp_stream_ops.accept = mptcp_stream_accept;
 	mptcp_stream_ops.getname = mptcp_v4_getname;
 	mptcp_stream_ops.listen = mptcp_listen;
+	mptcp_stream_ops.shutdown = mptcp_shutdown;
 
 	mptcp_subflow_init();
 
@@ -613,6 +678,7 @@ int mptcpv6_init(void)
 	mptcp_v6_stream_ops.accept = mptcp_stream_accept;
 	mptcp_v6_stream_ops.getname = mptcp_v6_getname;
 	mptcp_v6_stream_ops.listen = mptcp_listen;
+	mptcp_v6_stream_ops.shutdown = mptcp_shutdown;
 
 	err = inet6_register_protosw(&mptcp_v6_protosw);
 	if (err)
-- 
cgit v1.2.3-59-g8ed1b


From 717e79c867ca5f27815578815feafa3f3944f06b Mon Sep 17 00:00:00 2001
From: Peter Krystad <peter.krystad@linux.intel.com>
Date: Tue, 21 Jan 2020 16:56:22 -0800
Subject: mptcp: Add setsockopt()/getsockopt() socket operations

set/getsockopt behaviour with multiple subflows is undefined.
Therefore, for now, we return -EOPNOTSUPP unless we're in fallback mode.

Co-developed-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Peter Krystad <peter.krystad@linux.intel.com>
Signed-off-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/mptcp/protocol.c | 58 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 58 insertions(+)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 249b2506a66a..875ca48de4b2 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -330,6 +330,62 @@ static void mptcp_destroy(struct sock *sk)
 {
 }
 
+static int mptcp_setsockopt(struct sock *sk, int level, int optname,
+			    char __user *uoptval, unsigned int optlen)
+{
+	struct mptcp_sock *msk = mptcp_sk(sk);
+	char __kernel *optval;
+	int ret = -EOPNOTSUPP;
+	struct socket *ssock;
+
+	/* will be treated as __user in tcp_setsockopt */
+	optval = (char __kernel __force *)uoptval;
+
+	pr_debug("msk=%p", msk);
+
+	/* @@ the meaning of setsockopt() when the socket is connected and
+	 * there are multiple subflows is not defined.
+	 */
+	lock_sock(sk);
+	ssock = __mptcp_socket_create(msk, MPTCP_SAME_STATE);
+	if (!IS_ERR(ssock)) {
+		pr_debug("subflow=%p", ssock->sk);
+		ret = kernel_setsockopt(ssock, level, optname, optval, optlen);
+	}
+	release_sock(sk);
+
+	return ret;
+}
+
+static int mptcp_getsockopt(struct sock *sk, int level, int optname,
+			    char __user *uoptval, int __user *uoption)
+{
+	struct mptcp_sock *msk = mptcp_sk(sk);
+	char __kernel *optval;
+	int ret = -EOPNOTSUPP;
+	int __kernel *option;
+	struct socket *ssock;
+
+	/* will be treated as __user in tcp_getsockopt */
+	optval = (char __kernel __force *)uoptval;
+	option = (int __kernel __force *)uoption;
+
+	pr_debug("msk=%p", msk);
+
+	/* @@ the meaning of getsockopt() when the socket is connected and
+	 * there are multiple subflows is not defined.
+	 */
+	lock_sock(sk);
+	ssock = __mptcp_socket_create(msk, MPTCP_SAME_STATE);
+	if (!IS_ERR(ssock)) {
+		pr_debug("subflow=%p", ssock->sk);
+		ret = kernel_getsockopt(ssock, level, optname, optval, option);
+	}
+	release_sock(sk);
+
+	return ret;
+}
+
 static int mptcp_get_port(struct sock *sk, unsigned short snum)
 {
 	struct mptcp_sock *msk = mptcp_sk(sk);
@@ -380,6 +436,8 @@ static struct proto mptcp_prot = {
 	.init		= mptcp_init_sock,
 	.close		= mptcp_close,
 	.accept		= mptcp_accept,
+	.setsockopt	= mptcp_setsockopt,
+	.getsockopt	= mptcp_getsockopt,
 	.shutdown	= tcp_shutdown,
 	.destroy	= mptcp_destroy,
 	.sendmsg	= mptcp_sendmsg,
-- 
cgit v1.2.3-59-g8ed1b


From 6d0060f600adfddaa43fefb96b6b12643331961e Mon Sep 17 00:00:00 2001
From: Mat Martineau <mathew.j.martineau@linux.intel.com>
Date: Tue, 21 Jan 2020 16:56:23 -0800
Subject: mptcp: Write MPTCP DSS headers to outgoing data packets

Per-packet metadata required to write the MPTCP DSS option is written to
the skb_ext area. One write to the socket may contain more than one
packet of data, which is copied to page fragments and mapped in to MPTCP
DSS segments with size determined by the available page fragments and
the maximum mapping length allowed by the MPTCP specification. If
do_tcp_sendpages() splits a DSS segment in to multiple skbs, that's ok -
the later skbs can either have duplicated DSS mapping information or
none at all, and the receiver can handle that.

The current implementation uses the subflow frag cache and tcp
sendpages to avoid excessive code duplication. More work is required to
ensure that it works correctly under memory pressure and to support
MPTCP-level retransmissions.

The MPTCP DSS checksum is not yet implemented.

Co-developed-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Co-developed-by: Peter Krystad <peter.krystad@linux.intel.com>
Signed-off-by: Peter Krystad <peter.krystad@linux.intel.com>
Co-developed-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/mptcp.h  |   1 +
 net/mptcp/options.c  | 155 +++++++++++++++++++++++++++++++++++++++++++++++++--
 net/mptcp/protocol.c | 116 +++++++++++++++++++++++++++++++++++++-
 net/mptcp/protocol.h |  20 +++++++
 4 files changed, 286 insertions(+), 6 deletions(-)

diff --git a/include/net/mptcp.h b/include/net/mptcp.h
index eabc57c3fde4..06dcc665135e 100644
--- a/include/net/mptcp.h
+++ b/include/net/mptcp.h
@@ -32,6 +32,7 @@ struct mptcp_out_options {
 	u16 suboptions;
 	u64 sndr_key;
 	u64 rcvr_key;
+	struct mptcp_ext ext_copy;
 #endif
 };
 
diff --git a/net/mptcp/options.c b/net/mptcp/options.c
index 52ff2301b68b..5ea127bc7f01 100644
--- a/net/mptcp/options.c
+++ b/net/mptcp/options.c
@@ -134,13 +134,13 @@ void mptcp_rcv_synsent(struct sock *sk)
 	}
 }
 
-bool mptcp_established_options(struct sock *sk, struct sk_buff *skb,
-			       unsigned int *size, unsigned int remaining,
-			       struct mptcp_out_options *opts)
+static bool mptcp_established_options_mp(struct sock *sk, unsigned int *size,
+					 unsigned int remaining,
+					 struct mptcp_out_options *opts)
 {
 	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
 
-	if (subflow->mp_capable && !subflow->fourth_ack) {
+	if (!subflow->fourth_ack) {
 		opts->suboptions = OPTION_MPTCP_MPC_ACK;
 		opts->sndr_key = subflow->local_key;
 		opts->rcvr_key = subflow->remote_key;
@@ -153,6 +153,112 @@ bool mptcp_established_options(struct sock *sk, struct sk_buff *skb,
 	return false;
 }
 
+static void mptcp_write_data_fin(struct mptcp_subflow_context *subflow,
+				 struct mptcp_ext *ext)
+{
+	ext->data_fin = 1;
+
+	if (!ext->use_map) {
+		/* RFC6824 requires a DSS mapping with specific values
+		 * if DATA_FIN is set but no data payload is mapped
+		 */
+		ext->use_map = 1;
+		ext->dsn64 = 1;
+		ext->data_seq = mptcp_sk(subflow->conn)->write_seq;
+		ext->subflow_seq = 0;
+		ext->data_len = 1;
+	} else {
+		/* If there's an existing DSS mapping, DATA_FIN consumes
+		 * 1 additional byte of mapping space.
+		 */
+		ext->data_len++;
+	}
+}
+
+static bool mptcp_established_options_dss(struct sock *sk, struct sk_buff *skb,
+					  unsigned int *size,
+					  unsigned int remaining,
+					  struct mptcp_out_options *opts)
+{
+	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
+	unsigned int dss_size = 0;
+	struct mptcp_ext *mpext;
+	struct mptcp_sock *msk;
+	unsigned int ack_size;
+	u8 tcp_fin;
+
+	if (skb) {
+		mpext = mptcp_get_ext(skb);
+		tcp_fin = TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN;
+	} else {
+		mpext = NULL;
+		tcp_fin = 0;
+	}
+
+	if (!skb || (mpext && mpext->use_map) || tcp_fin) {
+		unsigned int map_size;
+
+		map_size = TCPOLEN_MPTCP_DSS_BASE + TCPOLEN_MPTCP_DSS_MAP64;
+
+		remaining -= map_size;
+		dss_size = map_size;
+		if (mpext)
+			opts->ext_copy = *mpext;
+
+		if (skb && tcp_fin &&
+		    subflow->conn->sk_state != TCP_ESTABLISHED)
+			mptcp_write_data_fin(subflow, &opts->ext_copy);
+	}
+
+	ack_size = TCPOLEN_MPTCP_DSS_ACK64;
+
+	/* Add kind/length/subtype/flag overhead if mapping is not populated */
+	if (dss_size == 0)
+		ack_size += TCPOLEN_MPTCP_DSS_BASE;
+
+	dss_size += ack_size;
+
+	msk = mptcp_sk(mptcp_subflow_ctx(sk)->conn);
+	if (msk) {
+		opts->ext_copy.data_ack = msk->ack_seq;
+	} else {
+		mptcp_crypto_key_sha(mptcp_subflow_ctx(sk)->remote_key,
+				     NULL, &opts->ext_copy.data_ack);
+		opts->ext_copy.data_ack++;
+	}
+
+	opts->ext_copy.ack64 = 1;
+	opts->ext_copy.use_ack = 1;
+
+	*size = ALIGN(dss_size, 4);
+	return true;
+}
+
+bool mptcp_established_options(struct sock *sk, struct sk_buff *skb,
+			       unsigned int *size, unsigned int remaining,
+			       struct mptcp_out_options *opts)
+{
+	unsigned int opt_size = 0;
+	bool ret = false;
+
+	if (mptcp_established_options_mp(sk, &opt_size, remaining, opts))
+		ret = true;
+	else if (mptcp_established_options_dss(sk, skb, &opt_size, remaining,
+					       opts))
+		ret = true;
+
+	/* we reserved enough space for the above options, and exceeding the
+	 * TCP option space would be fatal
+	 */
+	if (WARN_ON_ONCE(opt_size > remaining))
+		return false;
+
+	*size += opt_size;
+	remaining -= opt_size;
+
+	return ret;
+}
+
 bool mptcp_synack_options(const struct request_sock *req, unsigned int *size,
 			  struct mptcp_out_options *opts)
 {
@@ -194,4 +300,45 @@ void mptcp_write_options(__be32 *ptr, struct mptcp_out_options *opts)
 			ptr += 2;
 		}
 	}
+
+	if (opts->ext_copy.use_ack || opts->ext_copy.use_map) {
+		struct mptcp_ext *mpext = &opts->ext_copy;
+		u8 len = TCPOLEN_MPTCP_DSS_BASE;
+		u8 flags = 0;
+
+		if (mpext->use_ack) {
+			len += TCPOLEN_MPTCP_DSS_ACK64;
+			flags = MPTCP_DSS_HAS_ACK | MPTCP_DSS_ACK64;
+		}
+
+		if (mpext->use_map) {
+			len += TCPOLEN_MPTCP_DSS_MAP64;
+
+			/* Use only 64-bit mapping flags for now, add
+			 * support for optional 32-bit mappings later.
+			 */
+			flags |= MPTCP_DSS_HAS_MAP | MPTCP_DSS_DSN64;
+			if (mpext->data_fin)
+				flags |= MPTCP_DSS_DATA_FIN;
+		}
+
+		*ptr++ = htonl((TCPOPT_MPTCP << 24) |
+			       (len  << 16) |
+			       (MPTCPOPT_DSS << 12) |
+			       (flags));
+
+		if (mpext->use_ack) {
+			put_unaligned_be64(mpext->data_ack, ptr);
+			ptr += 2;
+		}
+
+		if (mpext->use_map) {
+			put_unaligned_be64(mpext->data_seq, ptr);
+			ptr += 2;
+			put_unaligned_be32(mpext->subflow_seq, ptr);
+			ptr += 1;
+			put_unaligned_be32(mpext->data_len << 16 |
+					   TCPOPT_NOP << 8 | TCPOPT_NOP, ptr);
+		}
+	}
 }
diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 875ca48de4b2..8cf49193b1c0 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -97,12 +97,93 @@ static struct sock *mptcp_subflow_get(const struct mptcp_sock *msk)
 	return NULL;
 }
 
+static bool mptcp_ext_cache_refill(struct mptcp_sock *msk)
+{
+	if (!msk->cached_ext)
+		msk->cached_ext = __skb_ext_alloc();
+
+	return !!msk->cached_ext;
+}
+
+static int mptcp_sendmsg_frag(struct sock *sk, struct sock *ssk,
+			      struct msghdr *msg, long *timeo)
+{
+	int mss_now = 0, size_goal = 0, ret = 0;
+	struct mptcp_sock *msk = mptcp_sk(sk);
+	struct mptcp_ext *mpext = NULL;
+	struct page_frag *pfrag;
+	struct sk_buff *skb;
+	size_t psize;
+
+	/* use the mptcp page cache so that we can easily move the data
+	 * from one substream to another, but do per subflow memory accounting
+	 */
+	pfrag = sk_page_frag(sk);
+	while (!sk_page_frag_refill(ssk, pfrag) ||
+	       !mptcp_ext_cache_refill(msk)) {
+		ret = sk_stream_wait_memory(ssk, timeo);
+		if (ret)
+			return ret;
+	}
+
+	/* compute copy limit */
+	mss_now = tcp_send_mss(ssk, &size_goal, msg->msg_flags);
+	psize = min_t(int, pfrag->size - pfrag->offset, size_goal);
+
+	pr_debug("left=%zu", msg_data_left(msg));
+	psize = copy_page_from_iter(pfrag->page, pfrag->offset,
+				    min_t(size_t, msg_data_left(msg), psize),
+				    &msg->msg_iter);
+	pr_debug("left=%zu", msg_data_left(msg));
+	if (!psize)
+		return -EINVAL;
+
+	/* Mark the end of the previous write so the beginning of the
+	 * next write (with its own mptcp skb extension data) is not
+	 * collapsed.
+	 */
+	skb = tcp_write_queue_tail(ssk);
+	if (skb)
+		TCP_SKB_CB(skb)->eor = 1;
+
+	ret = do_tcp_sendpages(ssk, pfrag->page, pfrag->offset, psize,
+			       msg->msg_flags | MSG_SENDPAGE_NOTLAST);
+	if (ret <= 0)
+		return ret;
+	if (unlikely(ret < psize))
+		iov_iter_revert(&msg->msg_iter, psize - ret);
+
+	skb = tcp_write_queue_tail(ssk);
+	mpext = __skb_ext_set(skb, SKB_EXT_MPTCP, msk->cached_ext);
+	msk->cached_ext = NULL;
+
+	memset(mpext, 0, sizeof(*mpext));
+	mpext->data_seq = msk->write_seq;
+	mpext->subflow_seq = mptcp_subflow_ctx(ssk)->rel_write_seq;
+	mpext->data_len = ret;
+	mpext->use_map = 1;
+	mpext->dsn64 = 1;
+
+	pr_debug("data_seq=%llu subflow_seq=%u data_len=%u dsn64=%d",
+		 mpext->data_seq, mpext->subflow_seq, mpext->data_len,
+		 mpext->dsn64);
+
+	pfrag->offset += ret;
+	msk->write_seq += ret;
+	mptcp_subflow_ctx(ssk)->rel_write_seq += ret;
+
+	tcp_push(ssk, msg->msg_flags, mss_now, tcp_sk(ssk)->nonagle, size_goal);
+	return ret;
+}
+
 static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 {
 	struct mptcp_sock *msk = mptcp_sk(sk);
 	struct socket *ssock;
+	size_t copied = 0;
 	struct sock *ssk;
-	int ret;
+	int ret = 0;
+	long timeo;
 
 	if (msg->msg_flags & ~(MSG_MORE | MSG_DONTWAIT | MSG_NOSIGNAL))
 		return -EOPNOTSUPP;
@@ -116,14 +197,29 @@ static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 		return ret;
 	}
 
+	timeo = sock_sndtimeo(sk, msg->msg_flags & MSG_DONTWAIT);
+
 	ssk = mptcp_subflow_get(msk);
 	if (!ssk) {
 		release_sock(sk);
 		return -ENOTCONN;
 	}
 
-	ret = sock_sendmsg(ssk->sk_socket, msg);
+	pr_debug("conn_list->subflow=%p", ssk);
 
+	lock_sock(ssk);
+	while (msg_data_left(msg)) {
+		ret = mptcp_sendmsg_frag(sk, ssk, msg, &timeo);
+		if (ret < 0)
+			break;
+
+		copied += ret;
+	}
+
+	if (copied > 0)
+		ret = copied;
+
+	release_sock(ssk);
 	release_sock(sk);
 	return ret;
 }
@@ -235,6 +331,8 @@ static void mptcp_close(struct sock *sk, long timeout)
 		__mptcp_close_ssk(sk, ssk, subflow, timeout);
 	}
 
+	if (msk->cached_ext)
+		__skb_ext_put(msk->cached_ext);
 	release_sock(sk);
 	sk_common_release(sk);
 }
@@ -286,6 +384,7 @@ static struct sock *mptcp_accept(struct sock *sk, int flags, int *err,
 		struct mptcp_subflow_context *subflow;
 		struct sock *new_mptcp_sock;
 		struct sock *ssk = newsk;
+		u64 ack_seq;
 
 		subflow = mptcp_subflow_ctx(newsk);
 		lock_sock(sk);
@@ -310,6 +409,12 @@ static struct sock *mptcp_accept(struct sock *sk, int flags, int *err,
 		msk->subflow = NULL;
 
 		mptcp_token_update_accept(newsk, new_mptcp_sock);
+
+		mptcp_crypto_key_sha(msk->remote_key, NULL, &ack_seq);
+		msk->write_seq = subflow->idsn + 1;
+		ack_seq++;
+		msk->ack_seq = ack_seq;
+		subflow->rel_write_seq = 1;
 		newsk = new_mptcp_sock;
 		mptcp_copy_inaddrs(newsk, ssk);
 		list_add(&subflow->node, &msk->conn_list);
@@ -404,6 +509,7 @@ void mptcp_finish_connect(struct sock *ssk)
 	struct mptcp_subflow_context *subflow;
 	struct mptcp_sock *msk;
 	struct sock *sk;
+	u64 ack_seq;
 
 	subflow = mptcp_subflow_ctx(ssk);
 
@@ -413,12 +519,18 @@ void mptcp_finish_connect(struct sock *ssk)
 	sk = subflow->conn;
 	msk = mptcp_sk(sk);
 
+	mptcp_crypto_key_sha(subflow->remote_key, NULL, &ack_seq);
+	ack_seq++;
+	subflow->rel_write_seq = 1;
+
 	/* the socket is not connected yet, no msk/subflow ops can access/race
 	 * accessing the field below
 	 */
 	WRITE_ONCE(msk->remote_key, subflow->remote_key);
 	WRITE_ONCE(msk->local_key, subflow->local_key);
 	WRITE_ONCE(msk->token, subflow->token);
+	WRITE_ONCE(msk->write_seq, subflow->idsn + 1);
+	WRITE_ONCE(msk->ack_seq, ack_seq);
 }
 
 static void mptcp_sock_graft(struct sock *sk, struct socket *parent)
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index 5f43fa0275c0..384ec4804198 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -32,6 +32,10 @@
 #define TCPOLEN_MPTCP_MPC_SYN		12
 #define TCPOLEN_MPTCP_MPC_SYNACK	12
 #define TCPOLEN_MPTCP_MPC_ACK		20
+#define TCPOLEN_MPTCP_DSS_BASE		4
+#define TCPOLEN_MPTCP_DSS_ACK64		8
+#define TCPOLEN_MPTCP_DSS_MAP64		14
+#define TCPOLEN_MPTCP_DSS_CHECKSUM	2
 
 /* MPTCP MP_CAPABLE flags */
 #define MPTCP_VERSION_MASK	(0x0F)
@@ -40,14 +44,24 @@
 #define MPTCP_CAP_HMAC_SHA1	BIT(0)
 #define MPTCP_CAP_FLAG_MASK	(0x3F)
 
+/* MPTCP DSS flags */
+#define MPTCP_DSS_DATA_FIN	BIT(4)
+#define MPTCP_DSS_DSN64		BIT(3)
+#define MPTCP_DSS_HAS_MAP	BIT(2)
+#define MPTCP_DSS_ACK64		BIT(1)
+#define MPTCP_DSS_HAS_ACK	BIT(0)
+
 /* MPTCP connection sock */
 struct mptcp_sock {
 	/* inet_connection_sock must be the first member */
 	struct inet_connection_sock sk;
 	u64		local_key;
 	u64		remote_key;
+	u64		write_seq;
+	u64		ack_seq;
 	u32		token;
 	struct list_head conn_list;
+	struct skb_ext	*cached_ext;	/* for the next sendmsg */
 	struct socket	*subflow; /* outgoing connect/listener/!mp_capable */
 };
 
@@ -83,6 +97,7 @@ struct mptcp_subflow_context {
 	u64	remote_key;
 	u64	idsn;
 	u32	token;
+	u32	rel_write_seq;
 	u32	request_mptcp : 1,  /* send MP_CAPABLE */
 		mp_capable : 1,	    /* remote is MPTCP capable */
 		fourth_ack : 1,	    /* send initial DSS */
@@ -144,4 +159,9 @@ static inline void mptcp_crypto_key_gen_sha(u64 *key, u32 *token, u64 *idsn)
 void mptcp_crypto_hmac_sha(u64 key1, u64 key2, u32 nonce1, u32 nonce2,
 			   u32 *hash_out);
 
+static inline struct mptcp_ext *mptcp_get_ext(struct sk_buff *skb)
+{
+	return (struct mptcp_ext *)skb_ext_find(skb, SKB_EXT_MPTCP);
+}
+
 #endif /* __MPTCP_PROTOCOL_H */
-- 
cgit v1.2.3-59-g8ed1b


From 648ef4b88673dadb8463bf0d4b10fbf33d55def8 Mon Sep 17 00:00:00 2001
From: Mat Martineau <mathew.j.martineau@linux.intel.com>
Date: Tue, 21 Jan 2020 16:56:24 -0800
Subject: mptcp: Implement MPTCP receive path

Parses incoming DSS options and populates outgoing MPTCP ACK
fields. MPTCP fields are parsed from the TCP option header and placed in
an skb extension, allowing the upper MPTCP layer to access MPTCP
options after the skb has gone through the TCP stack.

The subflow implements its own data_ready() ops, which ensures that the
pending data is in sequence - according to MPTCP seq number - dropping
out-of-seq skbs. The DATA_READY bit flag is set if this is the case.
This allows the MPTCP socket layer to determine if more data is
available without having to consult the individual subflows.

It additionally validates the current mapping and propagates EoF events
to the connection socket.

Co-developed-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Co-developed-by: Peter Krystad <peter.krystad@linux.intel.com>
Signed-off-by: Peter Krystad <peter.krystad@linux.intel.com>
Co-developed-by: Davide Caratti <dcaratti@redhat.com>
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Co-developed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Co-developed-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/tcp.h  |  10 ++
 include/net/mptcp.h  |   8 ++
 net/ipv4/tcp_input.c |   8 +-
 net/mptcp/options.c  | 107 ++++++++++++++
 net/mptcp/protocol.c |  34 +++++
 net/mptcp/protocol.h |  64 ++++++++-
 net/mptcp/subflow.c  | 383 ++++++++++++++++++++++++++++++++++++++++++++++++++-
 7 files changed, 609 insertions(+), 5 deletions(-)

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index e9ee06d887fa..0d00dad4b85d 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -82,9 +82,19 @@ struct tcp_sack_block {
 struct mptcp_options_received {
 	u64	sndr_key;
 	u64	rcvr_key;
+	u64	data_ack;
+	u64	data_seq;
+	u32	subflow_seq;
+	u16	data_len;
 	u8	mp_capable : 1,
 		mp_join : 1,
 		dss : 1;
+	u8	use_map:1,
+		dsn64:1,
+		data_fin:1,
+		use_ack:1,
+		ack64:1,
+		__unused:3;
 };
 #endif
 
diff --git a/include/net/mptcp.h b/include/net/mptcp.h
index 06dcc665135e..8619c1fca741 100644
--- a/include/net/mptcp.h
+++ b/include/net/mptcp.h
@@ -60,6 +60,8 @@ bool mptcp_synack_options(const struct request_sock *req, unsigned int *size,
 bool mptcp_established_options(struct sock *sk, struct sk_buff *skb,
 			       unsigned int *size, unsigned int remaining,
 			       struct mptcp_out_options *opts);
+void mptcp_incoming_options(struct sock *sk, struct sk_buff *skb,
+			    struct tcp_options_received *opt_rx);
 
 void mptcp_write_options(__be32 *ptr, struct mptcp_out_options *opts);
 
@@ -150,6 +152,12 @@ static inline bool mptcp_established_options(struct sock *sk,
 	return false;
 }
 
+static inline void mptcp_incoming_options(struct sock *sk,
+					  struct sk_buff *skb,
+					  struct tcp_options_received *opt_rx)
+{
+}
+
 static inline void mptcp_skb_ext_move(struct sk_buff *to,
 				      const struct sk_buff *from)
 {
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 5165c8de47ee..28d31f2c1422 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -4770,6 +4770,9 @@ static void tcp_data_queue(struct sock *sk, struct sk_buff *skb)
 	bool fragstolen;
 	int eaten;
 
+	if (sk_is_mptcp(sk))
+		mptcp_incoming_options(sk, skb, &tp->rx_opt);
+
 	if (TCP_SKB_CB(skb)->seq == TCP_SKB_CB(skb)->end_seq) {
 		__kfree_skb(skb);
 		return;
@@ -6346,8 +6349,11 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb)
 	case TCP_CLOSE_WAIT:
 	case TCP_CLOSING:
 	case TCP_LAST_ACK:
-		if (!before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt))
+		if (!before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt)) {
+			if (sk_is_mptcp(sk))
+				mptcp_incoming_options(sk, skb, &tp->rx_opt);
 			break;
+		}
 		/* fall through */
 	case TCP_FIN_WAIT1:
 	case TCP_FIN_WAIT2:
diff --git a/net/mptcp/options.c b/net/mptcp/options.c
index 5ea127bc7f01..1fa8496f3551 100644
--- a/net/mptcp/options.c
+++ b/net/mptcp/options.c
@@ -14,6 +14,7 @@ void mptcp_parse_option(const unsigned char *ptr, int opsize,
 {
 	struct mptcp_options_received *mp_opt = &opt_rx->mptcp;
 	u8 subtype = *ptr >> 4;
+	int expected_opsize;
 	u8 version;
 	u8 flags;
 
@@ -64,7 +65,79 @@ void mptcp_parse_option(const unsigned char *ptr, int opsize,
 
 	case MPTCPOPT_DSS:
 		pr_debug("DSS");
+		ptr++;
+
+		flags = (*ptr++) & MPTCP_DSS_FLAG_MASK;
+		mp_opt->data_fin = (flags & MPTCP_DSS_DATA_FIN) != 0;
+		mp_opt->dsn64 = (flags & MPTCP_DSS_DSN64) != 0;
+		mp_opt->use_map = (flags & MPTCP_DSS_HAS_MAP) != 0;
+		mp_opt->ack64 = (flags & MPTCP_DSS_ACK64) != 0;
+		mp_opt->use_ack = (flags & MPTCP_DSS_HAS_ACK);
+
+		pr_debug("data_fin=%d dsn64=%d use_map=%d ack64=%d use_ack=%d",
+			 mp_opt->data_fin, mp_opt->dsn64,
+			 mp_opt->use_map, mp_opt->ack64,
+			 mp_opt->use_ack);
+
+		expected_opsize = TCPOLEN_MPTCP_DSS_BASE;
+
+		if (mp_opt->use_ack) {
+			if (mp_opt->ack64)
+				expected_opsize += TCPOLEN_MPTCP_DSS_ACK64;
+			else
+				expected_opsize += TCPOLEN_MPTCP_DSS_ACK32;
+		}
+
+		if (mp_opt->use_map) {
+			if (mp_opt->dsn64)
+				expected_opsize += TCPOLEN_MPTCP_DSS_MAP64;
+			else
+				expected_opsize += TCPOLEN_MPTCP_DSS_MAP32;
+		}
+
+		/* RFC 6824, Section 3.3:
+		 * If a checksum is present, but its use had
+		 * not been negotiated in the MP_CAPABLE handshake,
+		 * the checksum field MUST be ignored.
+		 */
+		if (opsize != expected_opsize &&
+		    opsize != expected_opsize + TCPOLEN_MPTCP_DSS_CHECKSUM)
+			break;
+
 		mp_opt->dss = 1;
+
+		if (mp_opt->use_ack) {
+			if (mp_opt->ack64) {
+				mp_opt->data_ack = get_unaligned_be64(ptr);
+				ptr += 8;
+			} else {
+				mp_opt->data_ack = get_unaligned_be32(ptr);
+				ptr += 4;
+			}
+
+			pr_debug("data_ack=%llu", mp_opt->data_ack);
+		}
+
+		if (mp_opt->use_map) {
+			if (mp_opt->dsn64) {
+				mp_opt->data_seq = get_unaligned_be64(ptr);
+				ptr += 8;
+			} else {
+				mp_opt->data_seq = get_unaligned_be32(ptr);
+				ptr += 4;
+			}
+
+			mp_opt->subflow_seq = get_unaligned_be32(ptr);
+			ptr += 4;
+
+			mp_opt->data_len = get_unaligned_be16(ptr);
+			ptr += 2;
+
+			pr_debug("data_seq=%llu subflow_seq=%u data_len=%u",
+				 mp_opt->data_seq, mp_opt->subflow_seq,
+				 mp_opt->data_len);
+		}
+
 		break;
 
 	default:
@@ -275,6 +348,40 @@ bool mptcp_synack_options(const struct request_sock *req, unsigned int *size,
 	return false;
 }
 
+void mptcp_incoming_options(struct sock *sk, struct sk_buff *skb,
+			    struct tcp_options_received *opt_rx)
+{
+	struct mptcp_options_received *mp_opt;
+	struct mptcp_ext *mpext;
+
+	mp_opt = &opt_rx->mptcp;
+
+	if (!mp_opt->dss)
+		return;
+
+	mpext = skb_ext_add(skb, SKB_EXT_MPTCP);
+	if (!mpext)
+		return;
+
+	memset(mpext, 0, sizeof(*mpext));
+
+	if (mp_opt->use_map) {
+		mpext->data_seq = mp_opt->data_seq;
+		mpext->subflow_seq = mp_opt->subflow_seq;
+		mpext->data_len = mp_opt->data_len;
+		mpext->use_map = 1;
+		mpext->dsn64 = mp_opt->dsn64;
+	}
+
+	if (mp_opt->use_ack) {
+		mpext->data_ack = mp_opt->data_ack;
+		mpext->use_ack = 1;
+		mpext->ack64 = mp_opt->ack64;
+	}
+
+	mpext->data_fin = mp_opt->data_fin;
+}
+
 void mptcp_write_options(__be32 *ptr, struct mptcp_out_options *opts)
 {
 	if ((OPTION_MPTCP_MPC_SYN |
diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 8cf49193b1c0..71250149180b 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -224,6 +224,33 @@ static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 	return ret;
 }
 
+int mptcp_read_actor(read_descriptor_t *desc, struct sk_buff *skb,
+		     unsigned int offset, size_t len)
+{
+	struct mptcp_read_arg *arg = desc->arg.data;
+	size_t copy_len;
+
+	copy_len = min(desc->count, len);
+
+	if (likely(arg->msg)) {
+		int err;
+
+		err = skb_copy_datagram_msg(skb, offset, arg->msg, copy_len);
+		if (err) {
+			pr_debug("error path");
+			desc->error = err;
+			return err;
+		}
+	} else {
+		pr_debug("Flushing skb payload");
+	}
+
+	desc->count -= copy_len;
+
+	pr_debug("consumed %zu bytes, %zu left", copy_len, desc->count);
+	return copy_len;
+}
+
 static int mptcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
 			 int nonblock, int flags, int *addr_len)
 {
@@ -414,7 +441,10 @@ static struct sock *mptcp_accept(struct sock *sk, int flags, int *err,
 		msk->write_seq = subflow->idsn + 1;
 		ack_seq++;
 		msk->ack_seq = ack_seq;
+		subflow->map_seq = ack_seq;
+		subflow->map_subflow_seq = 1;
 		subflow->rel_write_seq = 1;
+		subflow->tcp_sock = ssk;
 		newsk = new_mptcp_sock;
 		mptcp_copy_inaddrs(newsk, ssk);
 		list_add(&subflow->node, &msk->conn_list);
@@ -519,8 +549,12 @@ void mptcp_finish_connect(struct sock *ssk)
 	sk = subflow->conn;
 	msk = mptcp_sk(sk);
 
+	pr_debug("msk=%p, token=%u", sk, subflow->token);
+
 	mptcp_crypto_key_sha(subflow->remote_key, NULL, &ack_seq);
 	ack_seq++;
+	subflow->map_seq = ack_seq;
+	subflow->map_subflow_seq = 1;
 	subflow->rel_write_seq = 1;
 
 	/* the socket is not connected yet, no msk/subflow ops can access/race
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index 384ec4804198..c6d8217e24d4 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -33,7 +33,9 @@
 #define TCPOLEN_MPTCP_MPC_SYNACK	12
 #define TCPOLEN_MPTCP_MPC_ACK		20
 #define TCPOLEN_MPTCP_DSS_BASE		4
+#define TCPOLEN_MPTCP_DSS_ACK32		4
 #define TCPOLEN_MPTCP_DSS_ACK64		8
+#define TCPOLEN_MPTCP_DSS_MAP32		10
 #define TCPOLEN_MPTCP_DSS_MAP64		14
 #define TCPOLEN_MPTCP_DSS_CHECKSUM	2
 
@@ -50,6 +52,10 @@
 #define MPTCP_DSS_HAS_MAP	BIT(2)
 #define MPTCP_DSS_ACK64		BIT(1)
 #define MPTCP_DSS_HAS_ACK	BIT(0)
+#define MPTCP_DSS_FLAG_MASK	(0x1F)
+
+/* MPTCP socket flags */
+#define MPTCP_DATA_READY	BIT(0)
 
 /* MPTCP connection sock */
 struct mptcp_sock {
@@ -60,6 +66,7 @@ struct mptcp_sock {
 	u64		write_seq;
 	u64		ack_seq;
 	u32		token;
+	unsigned long	flags;
 	struct list_head conn_list;
 	struct skb_ext	*cached_ext;	/* for the next sendmsg */
 	struct socket	*subflow; /* outgoing connect/listener/!mp_capable */
@@ -82,6 +89,7 @@ struct mptcp_subflow_request_sock {
 	u64	remote_key;
 	u64	idsn;
 	u32	token;
+	u32	ssn_offset;
 };
 
 static inline struct mptcp_subflow_request_sock *
@@ -96,15 +104,27 @@ struct mptcp_subflow_context {
 	u64	local_key;
 	u64	remote_key;
 	u64	idsn;
+	u64	map_seq;
 	u32	token;
 	u32	rel_write_seq;
+	u32	map_subflow_seq;
+	u32	ssn_offset;
+	u32	map_data_len;
 	u32	request_mptcp : 1,  /* send MP_CAPABLE */
 		mp_capable : 1,	    /* remote is MPTCP capable */
 		fourth_ack : 1,	    /* send initial DSS */
-		conn_finished : 1;
+		conn_finished : 1,
+		map_valid : 1,
+		data_avail : 1,
+		rx_eof : 1;
+
 	struct	sock *tcp_sock;	    /* tcp sk backpointer */
 	struct	sock *conn;	    /* parent mptcp_sock */
 	const	struct inet_connection_sock_af_ops *icsk_af_ops;
+	void	(*tcp_data_ready)(struct sock *sk);
+	void	(*tcp_state_change)(struct sock *sk);
+	void	(*tcp_write_space)(struct sock *sk);
+
 	struct	rcu_head rcu;
 };
 
@@ -123,14 +143,49 @@ mptcp_subflow_tcp_sock(const struct mptcp_subflow_context *subflow)
 	return subflow->tcp_sock;
 }
 
+static inline u64
+mptcp_subflow_get_map_offset(const struct mptcp_subflow_context *subflow)
+{
+	return tcp_sk(mptcp_subflow_tcp_sock(subflow))->copied_seq -
+		      subflow->ssn_offset -
+		      subflow->map_subflow_seq;
+}
+
+static inline u64
+mptcp_subflow_get_mapped_dsn(const struct mptcp_subflow_context *subflow)
+{
+	return subflow->map_seq + mptcp_subflow_get_map_offset(subflow);
+}
+
+int mptcp_is_enabled(struct net *net);
+bool mptcp_subflow_data_available(struct sock *sk);
 void mptcp_subflow_init(void);
 int mptcp_subflow_create_socket(struct sock *sk, struct socket **new_sock);
 
+static inline void mptcp_subflow_tcp_fallback(struct sock *sk,
+					      struct mptcp_subflow_context *ctx)
+{
+	sk->sk_data_ready = ctx->tcp_data_ready;
+	sk->sk_state_change = ctx->tcp_state_change;
+	sk->sk_write_space = ctx->tcp_write_space;
+
+	inet_csk(sk)->icsk_af_ops = ctx->icsk_af_ops;
+}
+
 extern const struct inet_connection_sock_af_ops ipv4_specific;
 #if IS_ENABLED(CONFIG_MPTCP_IPV6)
 extern const struct inet_connection_sock_af_ops ipv6_specific;
 #endif
 
+void mptcp_proto_init(void);
+
+struct mptcp_read_arg {
+	struct msghdr *msg;
+};
+
+int mptcp_read_actor(read_descriptor_t *desc, struct sk_buff *skb,
+		     unsigned int offset, size_t len);
+
 void mptcp_get_options(const struct sk_buff *skb,
 		       struct tcp_options_received *opt_rx);
 
@@ -164,4 +219,11 @@ static inline struct mptcp_ext *mptcp_get_ext(struct sk_buff *skb)
 	return (struct mptcp_ext *)skb_ext_find(skb, SKB_EXT_MPTCP);
 }
 
+static inline bool before64(__u64 seq1, __u64 seq2)
+{
+	return (__s64)(seq1 - seq2) < 0;
+}
+
+#define after64(seq2, seq1)	before64(seq1, seq2)
+
 #endif /* __MPTCP_PROTOCOL_H */
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index 89b91bc7a831..528351e26371 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -78,6 +78,7 @@ static void subflow_init_req(struct request_sock *req,
 			subflow_req->mp_capable = 1;
 
 		subflow_req->remote_key = rx_opt.mptcp.sndr_key;
+		subflow_req->ssn_offset = TCP_SKB_CB(skb)->seq;
 	}
 }
 
@@ -116,6 +117,11 @@ static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb)
 			 subflow->remote_key);
 		mptcp_finish_connect(sk);
 		subflow->conn_finished = 1;
+
+		if (skb) {
+			pr_debug("synack seq=%u", TCP_SKB_CB(skb)->seq);
+			subflow->ssn_offset = TCP_SKB_CB(skb)->seq;
+		}
 	}
 }
 
@@ -210,6 +216,323 @@ close_child:
 
 static struct inet_connection_sock_af_ops subflow_specific;
 
+enum mapping_status {
+	MAPPING_OK,
+	MAPPING_INVALID,
+	MAPPING_EMPTY,
+	MAPPING_DATA_FIN
+};
+
+static u64 expand_seq(u64 old_seq, u16 old_data_len, u64 seq)
+{
+	if ((u32)seq == (u32)old_seq)
+		return old_seq;
+
+	/* Assume map covers data not mapped yet. */
+	return seq | ((old_seq + old_data_len + 1) & GENMASK_ULL(63, 32));
+}
+
+static void warn_bad_map(struct mptcp_subflow_context *subflow, u32 ssn)
+{
+	WARN_ONCE(1, "Bad mapping: ssn=%d map_seq=%d map_data_len=%d",
+		  ssn, subflow->map_subflow_seq, subflow->map_data_len);
+}
+
+static bool skb_is_fully_mapped(struct sock *ssk, struct sk_buff *skb)
+{
+	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(ssk);
+	unsigned int skb_consumed;
+
+	skb_consumed = tcp_sk(ssk)->copied_seq - TCP_SKB_CB(skb)->seq;
+	if (WARN_ON_ONCE(skb_consumed >= skb->len))
+		return true;
+
+	return skb->len - skb_consumed <= subflow->map_data_len -
+					  mptcp_subflow_get_map_offset(subflow);
+}
+
+static bool validate_mapping(struct sock *ssk, struct sk_buff *skb)
+{
+	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(ssk);
+	u32 ssn = tcp_sk(ssk)->copied_seq - subflow->ssn_offset;
+
+	if (unlikely(before(ssn, subflow->map_subflow_seq))) {
+		/* Mapping covers data later in the subflow stream,
+		 * currently unsupported.
+		 */
+		warn_bad_map(subflow, ssn);
+		return false;
+	}
+	if (unlikely(!before(ssn, subflow->map_subflow_seq +
+				  subflow->map_data_len))) {
+		/* Mapping does covers past subflow data, invalid */
+		warn_bad_map(subflow, ssn + skb->len);
+		return false;
+	}
+	return true;
+}
+
+static enum mapping_status get_mapping_status(struct sock *ssk)
+{
+	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(ssk);
+	struct mptcp_ext *mpext;
+	struct sk_buff *skb;
+	u16 data_len;
+	u64 map_seq;
+
+	skb = skb_peek(&ssk->sk_receive_queue);
+	if (!skb)
+		return MAPPING_EMPTY;
+
+	mpext = mptcp_get_ext(skb);
+	if (!mpext || !mpext->use_map) {
+		if (!subflow->map_valid && !skb->len) {
+			/* the TCP stack deliver 0 len FIN pkt to the receive
+			 * queue, that is the only 0len pkts ever expected here,
+			 * and we can admit no mapping only for 0 len pkts
+			 */
+			if (!(TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN))
+				WARN_ONCE(1, "0len seq %d:%d flags %x",
+					  TCP_SKB_CB(skb)->seq,
+					  TCP_SKB_CB(skb)->end_seq,
+					  TCP_SKB_CB(skb)->tcp_flags);
+			sk_eat_skb(ssk, skb);
+			return MAPPING_EMPTY;
+		}
+
+		if (!subflow->map_valid)
+			return MAPPING_INVALID;
+
+		goto validate_seq;
+	}
+
+	pr_debug("seq=%llu is64=%d ssn=%u data_len=%u data_fin=%d",
+		 mpext->data_seq, mpext->dsn64, mpext->subflow_seq,
+		 mpext->data_len, mpext->data_fin);
+
+	data_len = mpext->data_len;
+	if (data_len == 0) {
+		pr_err("Infinite mapping not handled");
+		return MAPPING_INVALID;
+	}
+
+	if (mpext->data_fin == 1) {
+		if (data_len == 1) {
+			pr_debug("DATA_FIN with no payload");
+			if (subflow->map_valid) {
+				/* A DATA_FIN might arrive in a DSS
+				 * option before the previous mapping
+				 * has been fully consumed. Continue
+				 * handling the existing mapping.
+				 */
+				skb_ext_del(skb, SKB_EXT_MPTCP);
+				return MAPPING_OK;
+			} else {
+				return MAPPING_DATA_FIN;
+			}
+		}
+
+		/* Adjust for DATA_FIN using 1 byte of sequence space */
+		data_len--;
+	}
+
+	if (!mpext->dsn64) {
+		map_seq = expand_seq(subflow->map_seq, subflow->map_data_len,
+				     mpext->data_seq);
+		pr_debug("expanded seq=%llu", subflow->map_seq);
+	} else {
+		map_seq = mpext->data_seq;
+	}
+
+	if (subflow->map_valid) {
+		/* Allow replacing only with an identical map */
+		if (subflow->map_seq == map_seq &&
+		    subflow->map_subflow_seq == mpext->subflow_seq &&
+		    subflow->map_data_len == data_len) {
+			skb_ext_del(skb, SKB_EXT_MPTCP);
+			return MAPPING_OK;
+		}
+
+		/* If this skb data are fully covered by the current mapping,
+		 * the new map would need caching, which is not supported
+		 */
+		if (skb_is_fully_mapped(ssk, skb))
+			return MAPPING_INVALID;
+
+		/* will validate the next map after consuming the current one */
+		return MAPPING_OK;
+	}
+
+	subflow->map_seq = map_seq;
+	subflow->map_subflow_seq = mpext->subflow_seq;
+	subflow->map_data_len = data_len;
+	subflow->map_valid = 1;
+	pr_debug("new map seq=%llu subflow_seq=%u data_len=%u",
+		 subflow->map_seq, subflow->map_subflow_seq,
+		 subflow->map_data_len);
+
+validate_seq:
+	/* we revalidate valid mapping on new skb, because we must ensure
+	 * the current skb is completely covered by the available mapping
+	 */
+	if (!validate_mapping(ssk, skb))
+		return MAPPING_INVALID;
+
+	skb_ext_del(skb, SKB_EXT_MPTCP);
+	return MAPPING_OK;
+}
+
+static bool subflow_check_data_avail(struct sock *ssk)
+{
+	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(ssk);
+	enum mapping_status status;
+	struct mptcp_sock *msk;
+	struct sk_buff *skb;
+
+	pr_debug("msk=%p ssk=%p data_avail=%d skb=%p", subflow->conn, ssk,
+		 subflow->data_avail, skb_peek(&ssk->sk_receive_queue));
+	if (subflow->data_avail)
+		return true;
+
+	if (!subflow->conn)
+		return false;
+
+	msk = mptcp_sk(subflow->conn);
+	for (;;) {
+		u32 map_remaining;
+		size_t delta;
+		u64 ack_seq;
+		u64 old_ack;
+
+		status = get_mapping_status(ssk);
+		pr_debug("msk=%p ssk=%p status=%d", msk, ssk, status);
+		if (status == MAPPING_INVALID) {
+			ssk->sk_err = EBADMSG;
+			goto fatal;
+		}
+
+		if (status != MAPPING_OK)
+			return false;
+
+		skb = skb_peek(&ssk->sk_receive_queue);
+		if (WARN_ON_ONCE(!skb))
+			return false;
+
+		old_ack = READ_ONCE(msk->ack_seq);
+		ack_seq = mptcp_subflow_get_mapped_dsn(subflow);
+		pr_debug("msk ack_seq=%llx subflow ack_seq=%llx", old_ack,
+			 ack_seq);
+		if (ack_seq == old_ack)
+			break;
+
+		/* only accept in-sequence mapping. Old values are spurious
+		 * retransmission; we can hit "future" values on active backup
+		 * subflow switch, we relay on retransmissions to get
+		 * in-sequence data.
+		 * Cuncurrent subflows support will require subflow data
+		 * reordering
+		 */
+		map_remaining = subflow->map_data_len -
+				mptcp_subflow_get_map_offset(subflow);
+		if (before64(ack_seq, old_ack))
+			delta = min_t(size_t, old_ack - ack_seq, map_remaining);
+		else
+			delta = min_t(size_t, ack_seq - old_ack, map_remaining);
+
+		/* discard mapped data */
+		pr_debug("discarding %zu bytes, current map len=%d", delta,
+			 map_remaining);
+		if (delta) {
+			struct mptcp_read_arg arg = {
+				.msg = NULL,
+			};
+			read_descriptor_t desc = {
+				.count = delta,
+				.arg.data = &arg,
+			};
+			int ret;
+
+			ret = tcp_read_sock(ssk, &desc, mptcp_read_actor);
+			if (ret < 0) {
+				ssk->sk_err = -ret;
+				goto fatal;
+			}
+			if (ret < delta)
+				return false;
+			if (delta == map_remaining)
+				subflow->map_valid = 0;
+		}
+	}
+	return true;
+
+fatal:
+	/* fatal protocol error, close the socket */
+	/* This barrier is coupled with smp_rmb() in tcp_poll() */
+	smp_wmb();
+	ssk->sk_error_report(ssk);
+	tcp_set_state(ssk, TCP_CLOSE);
+	tcp_send_active_reset(ssk, GFP_ATOMIC);
+	return false;
+}
+
+bool mptcp_subflow_data_available(struct sock *sk)
+{
+	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
+	struct sk_buff *skb;
+
+	/* check if current mapping is still valid */
+	if (subflow->map_valid &&
+	    mptcp_subflow_get_map_offset(subflow) >= subflow->map_data_len) {
+		subflow->map_valid = 0;
+		subflow->data_avail = 0;
+
+		pr_debug("Done with mapping: seq=%u data_len=%u",
+			 subflow->map_subflow_seq,
+			 subflow->map_data_len);
+	}
+
+	if (!subflow_check_data_avail(sk)) {
+		subflow->data_avail = 0;
+		return false;
+	}
+
+	skb = skb_peek(&sk->sk_receive_queue);
+	subflow->data_avail = skb &&
+		       before(tcp_sk(sk)->copied_seq, TCP_SKB_CB(skb)->end_seq);
+	return subflow->data_avail;
+}
+
+static void subflow_data_ready(struct sock *sk)
+{
+	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
+	struct sock *parent = subflow->conn;
+
+	if (!parent || !subflow->mp_capable) {
+		subflow->tcp_data_ready(sk);
+
+		if (parent)
+			parent->sk_data_ready(parent);
+		return;
+	}
+
+	if (mptcp_subflow_data_available(sk)) {
+		set_bit(MPTCP_DATA_READY, &mptcp_sk(parent)->flags);
+
+		parent->sk_data_ready(parent);
+	}
+}
+
+static void subflow_write_space(struct sock *sk)
+{
+	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
+	struct sock *parent = subflow->conn;
+
+	sk_stream_write_space(sk);
+	if (parent && sk_stream_is_writeable(sk)) {
+		sk_stream_write_space(parent);
+	}
+}
+
 static struct inet_connection_sock_af_ops *
 subflow_default_af_ops(struct sock *sk)
 {
@@ -296,6 +619,47 @@ static struct mptcp_subflow_context *subflow_create_ctx(struct sock *sk,
 	return ctx;
 }
 
+static void __subflow_state_change(struct sock *sk)
+{
+	struct socket_wq *wq;
+
+	rcu_read_lock();
+	wq = rcu_dereference(sk->sk_wq);
+	if (skwq_has_sleeper(wq))
+		wake_up_interruptible_all(&wq->wait);
+	rcu_read_unlock();
+}
+
+static bool subflow_is_done(const struct sock *sk)
+{
+	return sk->sk_shutdown & RCV_SHUTDOWN || sk->sk_state == TCP_CLOSE;
+}
+
+static void subflow_state_change(struct sock *sk)
+{
+	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
+	struct sock *parent = READ_ONCE(subflow->conn);
+
+	__subflow_state_change(sk);
+
+	/* as recvmsg() does not acquire the subflow socket for ssk selection
+	 * a fin packet carrying a DSS can be unnoticed if we don't trigger
+	 * the data available machinery here.
+	 */
+	if (parent && subflow->mp_capable && mptcp_subflow_data_available(sk)) {
+		set_bit(MPTCP_DATA_READY, &mptcp_sk(parent)->flags);
+
+		parent->sk_data_ready(parent);
+	}
+
+	if (parent && !(parent->sk_shutdown & RCV_SHUTDOWN) &&
+	    !subflow->rx_eof && subflow_is_done(sk)) {
+		subflow->rx_eof = 1;
+		parent->sk_shutdown |= RCV_SHUTDOWN;
+		__subflow_state_change(parent);
+	}
+}
+
 static int subflow_ulp_init(struct sock *sk)
 {
 	struct inet_connection_sock *icsk = inet_csk(sk);
@@ -322,6 +686,12 @@ static int subflow_ulp_init(struct sock *sk)
 	tp->is_mptcp = 1;
 	ctx->icsk_af_ops = icsk->icsk_af_ops;
 	icsk->icsk_af_ops = subflow_default_af_ops(sk);
+	ctx->tcp_data_ready = sk->sk_data_ready;
+	ctx->tcp_state_change = sk->sk_state_change;
+	ctx->tcp_write_space = sk->sk_write_space;
+	sk->sk_data_ready = subflow_data_ready;
+	sk->sk_write_space = subflow_write_space;
+	sk->sk_state_change = subflow_state_change;
 out:
 	return err;
 }
@@ -339,10 +709,12 @@ static void subflow_ulp_release(struct sock *sk)
 	kfree_rcu(ctx, rcu);
 }
 
-static void subflow_ulp_fallback(struct sock *sk)
+static void subflow_ulp_fallback(struct sock *sk,
+				 struct mptcp_subflow_context *old_ctx)
 {
 	struct inet_connection_sock *icsk = inet_csk(sk);
 
+	mptcp_subflow_tcp_fallback(sk, old_ctx);
 	icsk->icsk_ulp_ops = NULL;
 	rcu_assign_pointer(icsk->icsk_ulp_data, NULL);
 	tcp_sk(sk)->is_mptcp = 0;
@@ -357,23 +729,28 @@ static void subflow_ulp_clone(const struct request_sock *req,
 	struct mptcp_subflow_context *new_ctx;
 
 	if (!subflow_req->mp_capable) {
-		subflow_ulp_fallback(newsk);
+		subflow_ulp_fallback(newsk, old_ctx);
 		return;
 	}
 
 	new_ctx = subflow_create_ctx(newsk, priority);
 	if (new_ctx == NULL) {
-		subflow_ulp_fallback(newsk);
+		subflow_ulp_fallback(newsk, old_ctx);
 		return;
 	}
 
 	new_ctx->conn_finished = 1;
 	new_ctx->icsk_af_ops = old_ctx->icsk_af_ops;
+	new_ctx->tcp_data_ready = old_ctx->tcp_data_ready;
+	new_ctx->tcp_state_change = old_ctx->tcp_state_change;
+	new_ctx->tcp_write_space = old_ctx->tcp_write_space;
 	new_ctx->mp_capable = 1;
 	new_ctx->fourth_ack = 1;
 	new_ctx->remote_key = subflow_req->remote_key;
 	new_ctx->local_key = subflow_req->local_key;
 	new_ctx->token = subflow_req->token;
+	new_ctx->ssn_offset = subflow_req->ssn_offset;
+	new_ctx->idsn = subflow_req->idsn;
 }
 
 static struct tcp_ulp_ops subflow_ulp_ops __read_mostly = {
-- 
cgit v1.2.3-59-g8ed1b


From 1891c4a07672b300b3160ceec59755063b20222b Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Tue, 21 Jan 2020 16:56:25 -0800
Subject: mptcp: add subflow write space signalling and mptcp_poll

Add new SEND_SPACE flag to indicate that a subflow has enough space to
accept more data for transmission.

It gets cleared at the end of mptcp_sendmsg() in case ssk has run
below the free watermark.

It is (re-set) from the wspace callback.

This allows us to use msk->flags to determine the poll mask.

Co-developed-by: Peter Krystad <peter.krystad@linux.intel.com>
Signed-off-by: Peter Krystad <peter.krystad@linux.intel.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/mptcp/protocol.c | 53 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 net/mptcp/protocol.h |  1 +
 net/mptcp/subflow.c  |  3 +++
 3 files changed, 57 insertions(+)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 71250149180b..408efbe34753 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -176,6 +176,23 @@ static int mptcp_sendmsg_frag(struct sock *sk, struct sock *ssk,
 	return ret;
 }
 
+static void ssk_check_wmem(struct mptcp_sock *msk, struct sock *ssk)
+{
+	struct socket *sock;
+
+	if (likely(sk_stream_is_writeable(ssk)))
+		return;
+
+	sock = READ_ONCE(ssk->sk_socket);
+
+	if (sock) {
+		clear_bit(MPTCP_SEND_SPACE, &msk->flags);
+		smp_mb__after_atomic();
+		/* set NOSPACE only after clearing SEND_SPACE flag */
+		set_bit(SOCK_NOSPACE, &sock->flags);
+	}
+}
+
 static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 {
 	struct mptcp_sock *msk = mptcp_sk(sk);
@@ -219,6 +236,7 @@ static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 	if (copied > 0)
 		ret = copied;
 
+	ssk_check_wmem(msk, ssk);
 	release_sock(ssk);
 	release_sock(sk);
 	return ret;
@@ -315,6 +333,7 @@ static int mptcp_init_sock(struct sock *sk)
 	struct mptcp_sock *msk = mptcp_sk(sk);
 
 	INIT_LIST_HEAD(&msk->conn_list);
+	__set_bit(MPTCP_SEND_SPACE, &msk->flags);
 
 	return 0;
 }
@@ -576,6 +595,13 @@ static void mptcp_sock_graft(struct sock *sk, struct socket *parent)
 	write_unlock_bh(&sk->sk_callback_lock);
 }
 
+static bool mptcp_memory_free(const struct sock *sk, int wake)
+{
+	struct mptcp_sock *msk = mptcp_sk(sk);
+
+	return wake ? test_bit(MPTCP_SEND_SPACE, &msk->flags) : true;
+}
+
 static struct proto mptcp_prot = {
 	.name		= "MPTCP",
 	.owner		= THIS_MODULE,
@@ -591,6 +617,7 @@ static struct proto mptcp_prot = {
 	.hash		= inet_hash,
 	.unhash		= inet_unhash,
 	.get_port	= mptcp_get_port,
+	.stream_memory_free	= mptcp_memory_free,
 	.obj_size	= sizeof(struct mptcp_sock),
 	.no_autobind	= true,
 };
@@ -767,8 +794,34 @@ unlock_fail:
 static __poll_t mptcp_poll(struct file *file, struct socket *sock,
 			   struct poll_table_struct *wait)
 {
+	const struct mptcp_sock *msk;
+	struct sock *sk = sock->sk;
+	struct socket *ssock;
 	__poll_t mask = 0;
 
+	msk = mptcp_sk(sk);
+	lock_sock(sk);
+	ssock = __mptcp_nmpc_socket(msk);
+	if (ssock) {
+		mask = ssock->ops->poll(file, ssock, wait);
+		release_sock(sk);
+		return mask;
+	}
+
+	release_sock(sk);
+	sock_poll_wait(file, sock, wait);
+	lock_sock(sk);
+
+	if (test_bit(MPTCP_DATA_READY, &msk->flags))
+		mask = EPOLLIN | EPOLLRDNORM;
+	if (sk_stream_is_writeable(sk) &&
+	    test_bit(MPTCP_SEND_SPACE, &msk->flags))
+		mask |= EPOLLOUT | EPOLLWRNORM;
+	if (sk->sk_shutdown & RCV_SHUTDOWN)
+		mask |= EPOLLIN | EPOLLRDNORM | EPOLLRDHUP;
+
+	release_sock(sk);
+
 	return mask;
 }
 
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index c6d8217e24d4..59a83eb64d37 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -56,6 +56,7 @@
 
 /* MPTCP socket flags */
 #define MPTCP_DATA_READY	BIT(0)
+#define MPTCP_SEND_SPACE	BIT(1)
 
 /* MPTCP connection sock */
 struct mptcp_sock {
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index 528351e26371..9fb3eb87a20f 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -529,6 +529,9 @@ static void subflow_write_space(struct sock *sk)
 
 	sk_stream_write_space(sk);
 	if (parent && sk_stream_is_writeable(sk)) {
+		set_bit(MPTCP_SEND_SPACE, &mptcp_sk(parent)->flags);
+		smp_mb__after_atomic();
+		/* set SEND_SPACE before sk_stream_write_space clears NOSPACE */
 		sk_stream_write_space(parent);
 	}
 }
-- 
cgit v1.2.3-59-g8ed1b


From 7a6a6cbc3e592e339ad23e4e8ede9a3f6160bda8 Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Tue, 21 Jan 2020 16:56:26 -0800
Subject: mptcp: recvmsg() can drain data from multiple subflows

With the previous patch in place, the msk can detect which subflow
has the current map with a simple walk, let's update the main
loop to always select the 'current' subflow. The exit conditions now
closely mirror tcp_recvmsg() to get expected timeout and signal
behavior.

Co-developed-by: Peter Krystad <peter.krystad@linux.intel.com>
Signed-off-by: Peter Krystad <peter.krystad@linux.intel.com>
Co-developed-by: Davide Caratti <dcaratti@redhat.com>
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Co-developed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Co-developed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Co-developed-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/mptcp/protocol.c | 178 ++++++++++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 168 insertions(+), 10 deletions(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 408efbe34753..ad9c73cc20e1 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -9,6 +9,8 @@
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/netdevice.h>
+#include <linux/sched/signal.h>
+#include <linux/atomic.h>
 #include <net/sock.h>
 #include <net/inet_common.h>
 #include <net/inet_hashtables.h>
@@ -105,6 +107,21 @@ static bool mptcp_ext_cache_refill(struct mptcp_sock *msk)
 	return !!msk->cached_ext;
 }
 
+static struct sock *mptcp_subflow_recv_lookup(const struct mptcp_sock *msk)
+{
+	struct mptcp_subflow_context *subflow;
+	struct sock *sk = (struct sock *)msk;
+
+	sock_owned_by_me(sk);
+
+	mptcp_for_each_subflow(msk, subflow) {
+		if (subflow->data_avail)
+			return mptcp_subflow_tcp_sock(subflow);
+	}
+
+	return NULL;
+}
+
 static int mptcp_sendmsg_frag(struct sock *sk, struct sock *ssk,
 			      struct msghdr *msg, long *timeo)
 {
@@ -269,13 +286,37 @@ int mptcp_read_actor(read_descriptor_t *desc, struct sk_buff *skb,
 	return copy_len;
 }
 
+static void mptcp_wait_data(struct sock *sk, long *timeo)
+{
+	DEFINE_WAIT_FUNC(wait, woken_wake_function);
+	struct mptcp_sock *msk = mptcp_sk(sk);
+
+	add_wait_queue(sk_sleep(sk), &wait);
+	sk_set_bit(SOCKWQ_ASYNC_WAITDATA, sk);
+
+	sk_wait_event(sk, timeo,
+		      test_and_clear_bit(MPTCP_DATA_READY, &msk->flags), &wait);
+
+	sk_clear_bit(SOCKWQ_ASYNC_WAITDATA, sk);
+	remove_wait_queue(sk_sleep(sk), &wait);
+}
+
 static int mptcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
 			 int nonblock, int flags, int *addr_len)
 {
 	struct mptcp_sock *msk = mptcp_sk(sk);
+	struct mptcp_subflow_context *subflow;
+	bool more_data_avail = false;
+	struct mptcp_read_arg arg;
+	read_descriptor_t desc;
+	bool wait_data = false;
 	struct socket *ssock;
+	struct tcp_sock *tp;
+	bool done = false;
 	struct sock *ssk;
 	int copied = 0;
+	int target;
+	long timeo;
 
 	if (msg->msg_flags & ~(MSG_WAITALL | MSG_DONTWAIT))
 		return -EOPNOTSUPP;
@@ -290,16 +331,124 @@ static int mptcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
 		return copied;
 	}
 
-	ssk = mptcp_subflow_get(msk);
-	if (!ssk) {
-		release_sock(sk);
-		return -ENOTCONN;
+	arg.msg = msg;
+	desc.arg.data = &arg;
+	desc.error = 0;
+
+	timeo = sock_rcvtimeo(sk, nonblock);
+
+	len = min_t(size_t, len, INT_MAX);
+	target = sock_rcvlowat(sk, flags & MSG_WAITALL, len);
+
+	while (!done) {
+		u32 map_remaining;
+		int bytes_read;
+
+		ssk = mptcp_subflow_recv_lookup(msk);
+		pr_debug("msk=%p ssk=%p", msk, ssk);
+		if (!ssk)
+			goto wait_for_data;
+
+		subflow = mptcp_subflow_ctx(ssk);
+		tp = tcp_sk(ssk);
+
+		lock_sock(ssk);
+		do {
+			/* try to read as much data as available */
+			map_remaining = subflow->map_data_len -
+					mptcp_subflow_get_map_offset(subflow);
+			desc.count = min_t(size_t, len - copied, map_remaining);
+			pr_debug("reading %zu bytes, copied %d", desc.count,
+				 copied);
+			bytes_read = tcp_read_sock(ssk, &desc,
+						   mptcp_read_actor);
+			if (bytes_read < 0) {
+				if (!copied)
+					copied = bytes_read;
+				done = true;
+				goto next;
+			}
+
+			pr_debug("msk ack_seq=%llx -> %llx", msk->ack_seq,
+				 msk->ack_seq + bytes_read);
+			msk->ack_seq += bytes_read;
+			copied += bytes_read;
+			if (copied >= len) {
+				done = true;
+				goto next;
+			}
+			if (tp->urg_data && tp->urg_seq == tp->copied_seq) {
+				pr_err("Urgent data present, cannot proceed");
+				done = true;
+				goto next;
+			}
+next:
+			more_data_avail = mptcp_subflow_data_available(ssk);
+		} while (more_data_avail && !done);
+		release_sock(ssk);
+		continue;
+
+wait_for_data:
+		more_data_avail = false;
+
+		/* only the master socket status is relevant here. The exit
+		 * conditions mirror closely tcp_recvmsg()
+		 */
+		if (copied >= target)
+			break;
+
+		if (copied) {
+			if (sk->sk_err ||
+			    sk->sk_state == TCP_CLOSE ||
+			    (sk->sk_shutdown & RCV_SHUTDOWN) ||
+			    !timeo ||
+			    signal_pending(current))
+				break;
+		} else {
+			if (sk->sk_err) {
+				copied = sock_error(sk);
+				break;
+			}
+
+			if (sk->sk_shutdown & RCV_SHUTDOWN)
+				break;
+
+			if (sk->sk_state == TCP_CLOSE) {
+				copied = -ENOTCONN;
+				break;
+			}
+
+			if (!timeo) {
+				copied = -EAGAIN;
+				break;
+			}
+
+			if (signal_pending(current)) {
+				copied = sock_intr_errno(timeo);
+				break;
+			}
+		}
+
+		pr_debug("block timeout %ld", timeo);
+		wait_data = true;
+		mptcp_wait_data(sk, &timeo);
 	}
 
-	copied = sock_recvmsg(ssk->sk_socket, msg, flags);
+	if (more_data_avail) {
+		if (!test_bit(MPTCP_DATA_READY, &msk->flags))
+			set_bit(MPTCP_DATA_READY, &msk->flags);
+	} else if (!wait_data) {
+		clear_bit(MPTCP_DATA_READY, &msk->flags);
 
-	release_sock(sk);
+		/* .. race-breaker: ssk might get new data after last
+		 * data_available() returns false.
+		 */
+		ssk = mptcp_subflow_recv_lookup(msk);
+		if (unlikely(ssk))
+			set_bit(MPTCP_DATA_READY, &msk->flags);
+	}
 
+	release_sock(sk);
 	return copied;
 }
 
@@ -460,10 +609,6 @@ static struct sock *mptcp_accept(struct sock *sk, int flags, int *err,
 		msk->write_seq = subflow->idsn + 1;
 		ack_seq++;
 		msk->ack_seq = ack_seq;
-		subflow->map_seq = ack_seq;
-		subflow->map_subflow_seq = 1;
-		subflow->rel_write_seq = 1;
-		subflow->tcp_sock = ssk;
 		newsk = new_mptcp_sock;
 		mptcp_copy_inaddrs(newsk, ssk);
 		list_add(&subflow->node, &msk->conn_list);
@@ -475,6 +620,19 @@ static struct sock *mptcp_accept(struct sock *sk, int flags, int *err,
 		bh_unlock_sock(new_mptcp_sock);
 		local_bh_enable();
 		release_sock(sk);
+
+		/* the subflow can already receive packet, avoid racing with
+		 * the receive path and process the pending ones
+		 */
+		lock_sock(ssk);
+		subflow->map_seq = ack_seq;
+		subflow->map_subflow_seq = 1;
+		subflow->rel_write_seq = 1;
+		subflow->tcp_sock = ssk;
+		subflow->conn = new_mptcp_sock;
+		if (unlikely(!skb_queue_empty(&ssk->sk_receive_queue)))
+			mptcp_subflow_data_available(ssk);
+		release_sock(ssk);
 	}
 
 	return newsk;
-- 
cgit v1.2.3-59-g8ed1b


From 57040755a3e43a1ee2d4ce9c83e87de27b570104 Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Tue, 21 Jan 2020 16:56:27 -0800
Subject: mptcp: allow collapsing consecutive sendpages on the same substream

If the current sendmsg() lands on the same subflow we used last, we
can try to collapse the data.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/mptcp/protocol.c | 75 +++++++++++++++++++++++++++++++++++++++++-----------
 1 file changed, 60 insertions(+), 15 deletions(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index ad9c73cc20e1..0b21ae25bd0f 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -122,14 +122,27 @@ static struct sock *mptcp_subflow_recv_lookup(const struct mptcp_sock *msk)
 	return NULL;
 }
 
+static inline bool mptcp_skb_can_collapse_to(const struct mptcp_sock *msk,
+					     const struct sk_buff *skb,
+					     const struct mptcp_ext *mpext)
+{
+	if (!tcp_skb_can_collapse_to(skb))
+		return false;
+
+	/* can collapse only if MPTCP level sequence is in order */
+	return mpext && mpext->data_seq + mpext->data_len == msk->write_seq;
+}
+
 static int mptcp_sendmsg_frag(struct sock *sk, struct sock *ssk,
-			      struct msghdr *msg, long *timeo)
+			      struct msghdr *msg, long *timeo, int *pmss_now,
+			      int *ps_goal)
 {
-	int mss_now = 0, size_goal = 0, ret = 0;
+	int mss_now, avail_size, size_goal, ret;
 	struct mptcp_sock *msk = mptcp_sk(sk);
 	struct mptcp_ext *mpext = NULL;
+	struct sk_buff *skb, *tail;
+	bool can_collapse = false;
 	struct page_frag *pfrag;
-	struct sk_buff *skb;
 	size_t psize;
 
 	/* use the mptcp page cache so that we can easily move the data
@@ -145,8 +158,29 @@ static int mptcp_sendmsg_frag(struct sock *sk, struct sock *ssk,
 
 	/* compute copy limit */
 	mss_now = tcp_send_mss(ssk, &size_goal, msg->msg_flags);
-	psize = min_t(int, pfrag->size - pfrag->offset, size_goal);
+	*pmss_now = mss_now;
+	*ps_goal = size_goal;
+	avail_size = size_goal;
+	skb = tcp_write_queue_tail(ssk);
+	if (skb) {
+		mpext = skb_ext_find(skb, SKB_EXT_MPTCP);
+
+		/* Limit the write to the size available in the
+		 * current skb, if any, so that we create at most a new skb.
+		 * Explicitly tells TCP internals to avoid collapsing on later
+		 * queue management operation, to avoid breaking the ext <->
+		 * SSN association set here
+		 */
+		can_collapse = (size_goal - skb->len > 0) &&
+			      mptcp_skb_can_collapse_to(msk, skb, mpext);
+		if (!can_collapse)
+			TCP_SKB_CB(skb)->eor = 1;
+		else
+			avail_size = size_goal - skb->len;
+	}
+	psize = min_t(size_t, pfrag->size - pfrag->offset, avail_size);
 
+	/* Copy to page */
 	pr_debug("left=%zu", msg_data_left(msg));
 	psize = copy_page_from_iter(pfrag->page, pfrag->offset,
 				    min_t(size_t, msg_data_left(msg), psize),
@@ -155,14 +189,9 @@ static int mptcp_sendmsg_frag(struct sock *sk, struct sock *ssk,
 	if (!psize)
 		return -EINVAL;
 
-	/* Mark the end of the previous write so the beginning of the
-	 * next write (with its own mptcp skb extension data) is not
-	 * collapsed.
+	/* tell the TCP stack to delay the push so that we can safely
+	 * access the skb after the sendpages call
 	 */
-	skb = tcp_write_queue_tail(ssk);
-	if (skb)
-		TCP_SKB_CB(skb)->eor = 1;
-
 	ret = do_tcp_sendpages(ssk, pfrag->page, pfrag->offset, psize,
 			       msg->msg_flags | MSG_SENDPAGE_NOTLAST);
 	if (ret <= 0)
@@ -170,6 +199,18 @@ static int mptcp_sendmsg_frag(struct sock *sk, struct sock *ssk,
 	if (unlikely(ret < psize))
 		iov_iter_revert(&msg->msg_iter, psize - ret);
 
+	/* if the tail skb extension is still the cached one, collapsing
+	 * really happened. Note: we can't check for 'same skb' as the sk_buff
+	 * hdr on tail can be transmitted, freed and re-allocated by the
+	 * do_tcp_sendpages() call
+	 */
+	tail = tcp_write_queue_tail(ssk);
+	if (mpext && tail && mpext == skb_ext_find(tail, SKB_EXT_MPTCP)) {
+		WARN_ON_ONCE(!can_collapse);
+		mpext->data_len += ret;
+		goto out;
+	}
+
 	skb = tcp_write_queue_tail(ssk);
 	mpext = __skb_ext_set(skb, SKB_EXT_MPTCP, msk->cached_ext);
 	msk->cached_ext = NULL;
@@ -185,11 +226,11 @@ static int mptcp_sendmsg_frag(struct sock *sk, struct sock *ssk,
 		 mpext->data_seq, mpext->subflow_seq, mpext->data_len,
 		 mpext->dsn64);
 
+out:
 	pfrag->offset += ret;
 	msk->write_seq += ret;
 	mptcp_subflow_ctx(ssk)->rel_write_seq += ret;
 
-	tcp_push(ssk, msg->msg_flags, mss_now, tcp_sk(ssk)->nonagle, size_goal);
 	return ret;
 }
 
@@ -212,11 +253,11 @@ static void ssk_check_wmem(struct mptcp_sock *msk, struct sock *ssk)
 
 static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 {
+	int mss_now = 0, size_goal = 0, ret = 0;
 	struct mptcp_sock *msk = mptcp_sk(sk);
 	struct socket *ssock;
 	size_t copied = 0;
 	struct sock *ssk;
-	int ret = 0;
 	long timeo;
 
 	if (msg->msg_flags & ~(MSG_MORE | MSG_DONTWAIT | MSG_NOSIGNAL))
@@ -243,15 +284,19 @@ static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 
 	lock_sock(ssk);
 	while (msg_data_left(msg)) {
-		ret = mptcp_sendmsg_frag(sk, ssk, msg, &timeo);
+		ret = mptcp_sendmsg_frag(sk, ssk, msg, &timeo, &mss_now,
+					 &size_goal);
 		if (ret < 0)
 			break;
 
 		copied += ret;
 	}
 
-	if (copied > 0)
+	if (copied) {
 		ret = copied;
+		tcp_push(ssk, msg->msg_flags, mss_now, tcp_sk(ssk)->nonagle,
+			 size_goal);
+	}
 
 	ssk_check_wmem(msk, ssk);
 	release_sock(ssk);
-- 
cgit v1.2.3-59-g8ed1b


From 784325e9f037e5f7a7f9a46ecbb27384128f8b6e Mon Sep 17 00:00:00 2001
From: Matthieu Baerts <matthieu.baerts@tessares.net>
Date: Tue, 21 Jan 2020 16:56:28 -0800
Subject: mptcp: new sysctl to control the activation per NS

New MPTCP sockets will return -ENOPROTOOPT if MPTCP support is disabled
for the current net namespace.

We are providing here a way to control access to the feature for those
that need to turn it on or off.

The value of this new sysctl can be different per namespace. We can then
restrict the usage of MPTCP to the selected NS. In case of serious
issues with MPTCP, administrators can now easily turn MPTCP off.

Co-developed-by: Peter Krystad <peter.krystad@linux.intel.com>
Signed-off-by: Peter Krystad <peter.krystad@linux.intel.com>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/mptcp/Makefile   |   2 +-
 net/mptcp/ctrl.c     | 130 +++++++++++++++++++++++++++++++++++++++++++++++++++
 net/mptcp/protocol.c |  16 +++++--
 net/mptcp/protocol.h |   3 ++
 4 files changed, 146 insertions(+), 5 deletions(-)
 create mode 100644 net/mptcp/ctrl.c

diff --git a/net/mptcp/Makefile b/net/mptcp/Makefile
index 178ae81d8b66..4e98d9edfd0a 100644
--- a/net/mptcp/Makefile
+++ b/net/mptcp/Makefile
@@ -1,4 +1,4 @@
 # SPDX-License-Identifier: GPL-2.0
 obj-$(CONFIG_MPTCP) += mptcp.o
 
-mptcp-y := protocol.o subflow.o options.o token.o crypto.o
+mptcp-y := protocol.o subflow.o options.o token.o crypto.o ctrl.o
diff --git a/net/mptcp/ctrl.c b/net/mptcp/ctrl.c
new file mode 100644
index 000000000000..8e39585d37f3
--- /dev/null
+++ b/net/mptcp/ctrl.c
@@ -0,0 +1,130 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Multipath TCP
+ *
+ * Copyright (c) 2019, Tessares SA.
+ */
+
+#include <linux/sysctl.h>
+
+#include <net/net_namespace.h>
+#include <net/netns/generic.h>
+
+#include "protocol.h"
+
+#define MPTCP_SYSCTL_PATH "net/mptcp"
+
+static int mptcp_pernet_id;
+struct mptcp_pernet {
+	struct ctl_table_header *ctl_table_hdr;
+
+	int mptcp_enabled;
+};
+
+static struct mptcp_pernet *mptcp_get_pernet(struct net *net)
+{
+	return net_generic(net, mptcp_pernet_id);
+}
+
+int mptcp_is_enabled(struct net *net)
+{
+	return mptcp_get_pernet(net)->mptcp_enabled;
+}
+
+static struct ctl_table mptcp_sysctl_table[] = {
+	{
+		.procname = "enabled",
+		.maxlen = sizeof(int),
+		.mode = 0644,
+		/* users with CAP_NET_ADMIN or root (not and) can change this
+		 * value, same as other sysctl or the 'net' tree.
+		 */
+		.proc_handler = proc_dointvec,
+	},
+	{}
+};
+
+static void mptcp_pernet_set_defaults(struct mptcp_pernet *pernet)
+{
+	pernet->mptcp_enabled = 1;
+}
+
+static int mptcp_pernet_new_table(struct net *net, struct mptcp_pernet *pernet)
+{
+	struct ctl_table_header *hdr;
+	struct ctl_table *table;
+
+	table = mptcp_sysctl_table;
+	if (!net_eq(net, &init_net)) {
+		table = kmemdup(table, sizeof(mptcp_sysctl_table), GFP_KERNEL);
+		if (!table)
+			goto err_alloc;
+	}
+
+	table[0].data = &pernet->mptcp_enabled;
+
+	hdr = register_net_sysctl(net, MPTCP_SYSCTL_PATH, table);
+	if (!hdr)
+		goto err_reg;
+
+	pernet->ctl_table_hdr = hdr;
+
+	return 0;
+
+err_reg:
+	if (!net_eq(net, &init_net))
+		kfree(table);
+err_alloc:
+	return -ENOMEM;
+}
+
+static void mptcp_pernet_del_table(struct mptcp_pernet *pernet)
+{
+	struct ctl_table *table = pernet->ctl_table_hdr->ctl_table_arg;
+
+	unregister_net_sysctl_table(pernet->ctl_table_hdr);
+
+	kfree(table);
+}
+
+static int __net_init mptcp_net_init(struct net *net)
+{
+	struct mptcp_pernet *pernet = mptcp_get_pernet(net);
+
+	mptcp_pernet_set_defaults(pernet);
+
+	return mptcp_pernet_new_table(net, pernet);
+}
+
+/* Note: the callback will only be called per extra netns */
+static void __net_exit mptcp_net_exit(struct net *net)
+{
+	struct mptcp_pernet *pernet = mptcp_get_pernet(net);
+
+	mptcp_pernet_del_table(pernet);
+}
+
+static struct pernet_operations mptcp_pernet_ops = {
+	.init = mptcp_net_init,
+	.exit = mptcp_net_exit,
+	.id = &mptcp_pernet_id,
+	.size = sizeof(struct mptcp_pernet),
+};
+
+void __init mptcp_init(void)
+{
+	mptcp_proto_init();
+
+	if (register_pernet_subsys(&mptcp_pernet_ops) < 0)
+		panic("Failed to register MPTCP pernet subsystem.\n");
+}
+
+#if IS_ENABLED(CONFIG_MPTCP_IPV6)
+int __init mptcpv6_init(void)
+{
+	int err;
+
+	err = mptcp_proto_v6_init();
+
+	return err;
+}
+#endif
diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 0b21ae25bd0f..45e482864a19 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -522,7 +522,7 @@ static void __mptcp_close_ssk(struct sock *sk, struct sock *ssk,
 	}
 }
 
-static int mptcp_init_sock(struct sock *sk)
+static int __mptcp_init_sock(struct sock *sk)
 {
 	struct mptcp_sock *msk = mptcp_sk(sk);
 
@@ -532,6 +532,14 @@ static int mptcp_init_sock(struct sock *sk)
 	return 0;
 }
 
+static int mptcp_init_sock(struct sock *sk)
+{
+	if (!mptcp_is_enabled(sock_net(sk)))
+		return -ENOPROTOOPT;
+
+	return __mptcp_init_sock(sk);
+}
+
 static void mptcp_subflow_shutdown(struct sock *ssk, int how)
 {
 	lock_sock(ssk);
@@ -640,7 +648,7 @@ static struct sock *mptcp_accept(struct sock *sk, int flags, int *err,
 			return NULL;
 		}
 
-		mptcp_init_sock(new_mptcp_sock);
+		__mptcp_init_sock(new_mptcp_sock);
 
 		msk = mptcp_sk(new_mptcp_sock);
 		msk->remote_key = subflow->remote_key;
@@ -1078,7 +1086,7 @@ static struct inet_protosw mptcp_protosw = {
 	.flags		= INET_PROTOSW_ICSK,
 };
 
-void __init mptcp_init(void)
+void mptcp_proto_init(void)
 {
 	mptcp_prot.h.hashinfo = tcp_prot.h.hashinfo;
 	mptcp_stream_ops = inet_stream_ops;
@@ -1116,7 +1124,7 @@ static struct inet_protosw mptcp_v6_protosw = {
 	.flags		= INET_PROTOSW_ICSK,
 };
 
-int mptcpv6_init(void)
+int mptcp_proto_v6_init(void)
 {
 	int err;
 
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index 59a83eb64d37..a8fad7d78565 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -179,6 +179,9 @@ extern const struct inet_connection_sock_af_ops ipv6_specific;
 #endif
 
 void mptcp_proto_init(void);
+#if IS_ENABLED(CONFIG_MPTCP_IPV6)
+int mptcp_proto_v6_init(void);
+#endif
 
 struct mptcp_read_arg {
 	struct msghdr *msg;
-- 
cgit v1.2.3-59-g8ed1b


From 048d19d444be1e42abca19a6b969343954ae4e17 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Tue, 21 Jan 2020 16:56:29 -0800
Subject: mptcp: add basic kselftest for mptcp

Add mptcp_connect tool:
xmit two files back and forth between two processes, several net
namespaces including some adding delays, losses and reordering.
Wrapper script tests that data was transmitted without corruption.

The "-c" command line option for mptcp_connect.sh is there for debugging:

The script will use tcpdump to create one .pcap file per test case, named
according to the namespaces, protocols, and connect address in use.
For example, the first test case writes the capture to
ns1-ns1-MPTCP-MPTCP-10.0.1.1.pcap.

The stderr output from tcpdump is printed after the test completes to
show tcpdump's "packets dropped by kernel" information.

Also check that userspace can't create MPTCP sockets when mptcp.enabled
sysctl is off.

The "-b" option allows to tune/lower send buffer size.
"-m mmap" can be used to test blocking io.  Default is non-blocking
io using read/write/poll.

Will run automatically on "make kselftest".

Note that the default timeout of 45 seconds is used even if there is a
"settings" changing it to 450. 45 seconds should be enough in most cases
but this depends on the machine running the tests.

A fix to correctly read the "settings" file has been proposed upstream
but not applied yet. It is not blocking the execution of these new tests
but it would be nice to have it:

  https://patchwork.kernel.org/patch/11204935/

Co-developed-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Co-developed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Co-developed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Co-developed-by: Davide Caratti <dcaratti@redhat.com>
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 MAINTAINERS                                        |   1 +
 tools/testing/selftests/Makefile                   |   1 +
 tools/testing/selftests/net/mptcp/.gitignore       |   2 +
 tools/testing/selftests/net/mptcp/Makefile         |  13 +
 tools/testing/selftests/net/mptcp/config           |   4 +
 tools/testing/selftests/net/mptcp/mptcp_connect.c  | 832 +++++++++++++++++++++
 tools/testing/selftests/net/mptcp/mptcp_connect.sh | 595 +++++++++++++++
 tools/testing/selftests/net/mptcp/settings         |   1 +
 8 files changed, 1449 insertions(+)
 create mode 100644 tools/testing/selftests/net/mptcp/.gitignore
 create mode 100644 tools/testing/selftests/net/mptcp/Makefile
 create mode 100644 tools/testing/selftests/net/mptcp/config
 create mode 100644 tools/testing/selftests/net/mptcp/mptcp_connect.c
 create mode 100755 tools/testing/selftests/net/mptcp/mptcp_connect.sh
 create mode 100644 tools/testing/selftests/net/mptcp/settings

diff --git a/MAINTAINERS b/MAINTAINERS
index 4be9c1b50183..e7463b46565b 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -11584,6 +11584,7 @@ B:	https://github.com/multipath-tcp/mptcp_net-next/issues
 S:	Maintained
 F:	include/net/mptcp.h
 F:	net/mptcp/
+F:	tools/testing/selftests/net/mptcp/
 
 NETWORKING [TCP]
 M:	Eric Dumazet <edumazet@google.com>
diff --git a/tools/testing/selftests/Makefile b/tools/testing/selftests/Makefile
index b001c602414b..b45a8f77ee24 100644
--- a/tools/testing/selftests/Makefile
+++ b/tools/testing/selftests/Makefile
@@ -32,6 +32,7 @@ TARGETS += memory-hotplug
 TARGETS += mount
 TARGETS += mqueue
 TARGETS += net
+TARGETS += net/mptcp
 TARGETS += netfilter
 TARGETS += networking/timestamping
 TARGETS += nsfs
diff --git a/tools/testing/selftests/net/mptcp/.gitignore b/tools/testing/selftests/net/mptcp/.gitignore
new file mode 100644
index 000000000000..d72f07642738
--- /dev/null
+++ b/tools/testing/selftests/net/mptcp/.gitignore
@@ -0,0 +1,2 @@
+mptcp_connect
+*.pcap
diff --git a/tools/testing/selftests/net/mptcp/Makefile b/tools/testing/selftests/net/mptcp/Makefile
new file mode 100644
index 000000000000..93de52016dde
--- /dev/null
+++ b/tools/testing/selftests/net/mptcp/Makefile
@@ -0,0 +1,13 @@
+# SPDX-License-Identifier: GPL-2.0
+
+top_srcdir = ../../../../..
+
+CFLAGS =  -Wall -Wl,--no-as-needed -O2 -g
+
+TEST_PROGS := mptcp_connect.sh
+
+TEST_GEN_FILES = mptcp_connect
+
+EXTRA_CLEAN := *.pcap
+
+include ../../lib.mk
diff --git a/tools/testing/selftests/net/mptcp/config b/tools/testing/selftests/net/mptcp/config
new file mode 100644
index 000000000000..2499824d9e1c
--- /dev/null
+++ b/tools/testing/selftests/net/mptcp/config
@@ -0,0 +1,4 @@
+CONFIG_MPTCP=y
+CONFIG_MPTCP_IPV6=y
+CONFIG_VETH=y
+CONFIG_NET_SCH_NETEM=m
diff --git a/tools/testing/selftests/net/mptcp/mptcp_connect.c b/tools/testing/selftests/net/mptcp/mptcp_connect.c
new file mode 100644
index 000000000000..a3dccd816ae4
--- /dev/null
+++ b/tools/testing/selftests/net/mptcp/mptcp_connect.c
@@ -0,0 +1,832 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#define _GNU_SOURCE
+
+#include <errno.h>
+#include <limits.h>
+#include <fcntl.h>
+#include <string.h>
+#include <stdbool.h>
+#include <stdint.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <strings.h>
+#include <unistd.h>
+
+#include <sys/poll.h>
+#include <sys/sendfile.h>
+#include <sys/stat.h>
+#include <sys/socket.h>
+#include <sys/types.h>
+#include <sys/mman.h>
+
+#include <netdb.h>
+#include <netinet/in.h>
+
+#include <linux/tcp.h>
+
+extern int optind;
+
+#ifndef IPPROTO_MPTCP
+#define IPPROTO_MPTCP 262
+#endif
+#ifndef TCP_ULP
+#define TCP_ULP 31
+#endif
+
+static bool listen_mode;
+static int  poll_timeout;
+
+enum cfg_mode {
+	CFG_MODE_POLL,
+	CFG_MODE_MMAP,
+	CFG_MODE_SENDFILE,
+};
+
+static enum cfg_mode cfg_mode = CFG_MODE_POLL;
+static const char *cfg_host;
+static const char *cfg_port	= "12000";
+static int cfg_sock_proto	= IPPROTO_MPTCP;
+static bool tcpulp_audit;
+static int pf = AF_INET;
+static int cfg_sndbuf;
+
+static void die_usage(void)
+{
+	fprintf(stderr, "Usage: mptcp_connect [-6] [-u] [-s MPTCP|TCP] [-p port] -m mode]"
+		"[ -l ] [ -t timeout ] connect_address\n");
+	exit(1);
+}
+
+static const char *getxinfo_strerr(int err)
+{
+	if (err == EAI_SYSTEM)
+		return strerror(errno);
+
+	return gai_strerror(err);
+}
+
+static void xgetnameinfo(const struct sockaddr *addr, socklen_t addrlen,
+			 char *host, socklen_t hostlen,
+			 char *serv, socklen_t servlen)
+{
+	int flags = NI_NUMERICHOST | NI_NUMERICSERV;
+	int err = getnameinfo(addr, addrlen, host, hostlen, serv, servlen,
+			      flags);
+
+	if (err) {
+		const char *errstr = getxinfo_strerr(err);
+
+		fprintf(stderr, "Fatal: getnameinfo: %s\n", errstr);
+		exit(1);
+	}
+}
+
+static void xgetaddrinfo(const char *node, const char *service,
+			 const struct addrinfo *hints,
+			 struct addrinfo **res)
+{
+	int err = getaddrinfo(node, service, hints, res);
+
+	if (err) {
+		const char *errstr = getxinfo_strerr(err);
+
+		fprintf(stderr, "Fatal: getaddrinfo(%s:%s): %s\n",
+			node ? node : "", service ? service : "", errstr);
+		exit(1);
+	}
+}
+
+static void set_sndbuf(int fd, unsigned int size)
+{
+	int err;
+
+	err = setsockopt(fd, SOL_SOCKET, SO_SNDBUF, &size, sizeof(size));
+	if (err) {
+		perror("set SO_SNDBUF");
+		exit(1);
+	}
+}
+
+static int sock_listen_mptcp(const char * const listenaddr,
+			     const char * const port)
+{
+	int sock;
+	struct addrinfo hints = {
+		.ai_protocol = IPPROTO_TCP,
+		.ai_socktype = SOCK_STREAM,
+		.ai_flags = AI_PASSIVE | AI_NUMERICHOST
+	};
+
+	hints.ai_family = pf;
+
+	struct addrinfo *a, *addr;
+	int one = 1;
+
+	xgetaddrinfo(listenaddr, port, &hints, &addr);
+	hints.ai_family = pf;
+
+	for (a = addr; a; a = a->ai_next) {
+		sock = socket(a->ai_family, a->ai_socktype, cfg_sock_proto);
+		if (sock < 0)
+			continue;
+
+		if (-1 == setsockopt(sock, SOL_SOCKET, SO_REUSEADDR, &one,
+				     sizeof(one)))
+			perror("setsockopt");
+
+		if (bind(sock, a->ai_addr, a->ai_addrlen) == 0)
+			break; /* success */
+
+		perror("bind");
+		close(sock);
+		sock = -1;
+	}
+
+	freeaddrinfo(addr);
+
+	if (sock < 0) {
+		fprintf(stderr, "Could not create listen socket\n");
+		return sock;
+	}
+
+	if (listen(sock, 20)) {
+		perror("listen");
+		close(sock);
+		return -1;
+	}
+
+	return sock;
+}
+
+static bool sock_test_tcpulp(const char * const remoteaddr,
+			     const char * const port)
+{
+	struct addrinfo hints = {
+		.ai_protocol = IPPROTO_TCP,
+		.ai_socktype = SOCK_STREAM,
+	};
+	struct addrinfo *a, *addr;
+	int sock = -1, ret = 0;
+	bool test_pass = false;
+
+	hints.ai_family = AF_INET;
+
+	xgetaddrinfo(remoteaddr, port, &hints, &addr);
+	for (a = addr; a; a = a->ai_next) {
+		sock = socket(a->ai_family, a->ai_socktype, IPPROTO_TCP);
+		if (sock < 0) {
+			perror("socket");
+			continue;
+		}
+		ret = setsockopt(sock, IPPROTO_TCP, TCP_ULP, "mptcp",
+				 sizeof("mptcp"));
+		if (ret == -1 && errno == EOPNOTSUPP)
+			test_pass = true;
+		close(sock);
+
+		if (test_pass)
+			break;
+		if (!ret)
+			fprintf(stderr,
+				"setsockopt(TCP_ULP) returned 0\n");
+		else
+			perror("setsockopt(TCP_ULP)");
+	}
+	return test_pass;
+}
+
+static int sock_connect_mptcp(const char * const remoteaddr,
+			      const char * const port, int proto)
+{
+	struct addrinfo hints = {
+		.ai_protocol = IPPROTO_TCP,
+		.ai_socktype = SOCK_STREAM,
+	};
+	struct addrinfo *a, *addr;
+	int sock = -1;
+
+	hints.ai_family = pf;
+
+	xgetaddrinfo(remoteaddr, port, &hints, &addr);
+	for (a = addr; a; a = a->ai_next) {
+		sock = socket(a->ai_family, a->ai_socktype, proto);
+		if (sock < 0) {
+			perror("socket");
+			continue;
+		}
+
+		if (connect(sock, a->ai_addr, a->ai_addrlen) == 0)
+			break; /* success */
+
+		perror("connect()");
+		close(sock);
+		sock = -1;
+	}
+
+	freeaddrinfo(addr);
+	return sock;
+}
+
+static size_t do_rnd_write(const int fd, char *buf, const size_t len)
+{
+	unsigned int do_w;
+	ssize_t bw;
+
+	do_w = rand() & 0xffff;
+	if (do_w == 0 || do_w > len)
+		do_w = len;
+
+	bw = write(fd, buf, do_w);
+	if (bw < 0)
+		perror("write");
+
+	return bw;
+}
+
+static size_t do_write(const int fd, char *buf, const size_t len)
+{
+	size_t offset = 0;
+
+	while (offset < len) {
+		size_t written;
+		ssize_t bw;
+
+		bw = write(fd, buf + offset, len - offset);
+		if (bw < 0) {
+			perror("write");
+			return 0;
+		}
+
+		written = (size_t)bw;
+		offset += written;
+	}
+
+	return offset;
+}
+
+static ssize_t do_rnd_read(const int fd, char *buf, const size_t len)
+{
+	size_t cap = rand();
+
+	cap &= 0xffff;
+
+	if (cap == 0)
+		cap = 1;
+	else if (cap > len)
+		cap = len;
+
+	return read(fd, buf, cap);
+}
+
+static void set_nonblock(int fd)
+{
+	int flags = fcntl(fd, F_GETFL);
+
+	if (flags == -1)
+		return;
+
+	fcntl(fd, F_SETFL, flags | O_NONBLOCK);
+}
+
+static int copyfd_io_poll(int infd, int peerfd, int outfd)
+{
+	struct pollfd fds = {
+		.fd = peerfd,
+		.events = POLLIN | POLLOUT,
+	};
+	unsigned int woff = 0, wlen = 0;
+	char wbuf[8192];
+
+	set_nonblock(peerfd);
+
+	for (;;) {
+		char rbuf[8192];
+		ssize_t len;
+
+		if (fds.events == 0)
+			break;
+
+		switch (poll(&fds, 1, poll_timeout)) {
+		case -1:
+			if (errno == EINTR)
+				continue;
+			perror("poll");
+			return 1;
+		case 0:
+			fprintf(stderr, "%s: poll timed out (events: "
+				"POLLIN %u, POLLOUT %u)\n", __func__,
+				fds.events & POLLIN, fds.events & POLLOUT);
+			return 2;
+		}
+
+		if (fds.revents & POLLIN) {
+			len = do_rnd_read(peerfd, rbuf, sizeof(rbuf));
+			if (len == 0) {
+				/* no more data to receive:
+				 * peer has closed its write side
+				 */
+				fds.events &= ~POLLIN;
+
+				if ((fds.events & POLLOUT) == 0)
+					/* and nothing more to send */
+					break;
+
+			/* Else, still have data to transmit */
+			} else if (len < 0) {
+				perror("read");
+				return 3;
+			}
+
+			do_write(outfd, rbuf, len);
+		}
+
+		if (fds.revents & POLLOUT) {
+			if (wlen == 0) {
+				woff = 0;
+				wlen = read(infd, wbuf, sizeof(wbuf));
+			}
+
+			if (wlen > 0) {
+				ssize_t bw;
+
+				bw = do_rnd_write(peerfd, wbuf + woff, wlen);
+				if (bw < 0)
+					return 111;
+
+				woff += bw;
+				wlen -= bw;
+			} else if (wlen == 0) {
+				/* We have no more data to send. */
+				fds.events &= ~POLLOUT;
+
+				if ((fds.events & POLLIN) == 0)
+					/* ... and peer also closed already */
+					break;
+
+				/* ... but we still receive.
+				 * Close our write side.
+				 */
+				shutdown(peerfd, SHUT_WR);
+			} else {
+				if (errno == EINTR)
+					continue;
+				perror("read");
+				return 4;
+			}
+		}
+
+		if (fds.revents & (POLLERR | POLLNVAL)) {
+			fprintf(stderr, "Unexpected revents: "
+				"POLLERR/POLLNVAL(%x)\n", fds.revents);
+			return 5;
+		}
+	}
+
+	close(peerfd);
+	return 0;
+}
+
+static int do_recvfile(int infd, int outfd)
+{
+	ssize_t r;
+
+	do {
+		char buf[16384];
+
+		r = do_rnd_read(infd, buf, sizeof(buf));
+		if (r > 0) {
+			if (write(outfd, buf, r) != r)
+				break;
+		} else if (r < 0) {
+			perror("read");
+		}
+	} while (r > 0);
+
+	return (int)r;
+}
+
+static int do_mmap(int infd, int outfd, unsigned int size)
+{
+	char *inbuf = mmap(NULL, size, PROT_READ, MAP_SHARED, infd, 0);
+	ssize_t ret = 0, off = 0;
+	size_t rem;
+
+	if (inbuf == MAP_FAILED) {
+		perror("mmap");
+		return 1;
+	}
+
+	rem = size;
+
+	while (rem > 0) {
+		ret = write(outfd, inbuf + off, rem);
+
+		if (ret < 0) {
+			perror("write");
+			break;
+		}
+
+		off += ret;
+		rem -= ret;
+	}
+
+	munmap(inbuf, size);
+	return rem;
+}
+
+static int get_infd_size(int fd)
+{
+	struct stat sb;
+	ssize_t count;
+	int err;
+
+	err = fstat(fd, &sb);
+	if (err < 0) {
+		perror("fstat");
+		return -1;
+	}
+
+	if ((sb.st_mode & S_IFMT) != S_IFREG) {
+		fprintf(stderr, "%s: stdin is not a regular file\n", __func__);
+		return -2;
+	}
+
+	count = sb.st_size;
+	if (count > INT_MAX) {
+		fprintf(stderr, "File too large: %zu\n", count);
+		return -3;
+	}
+
+	return (int)count;
+}
+
+static int do_sendfile(int infd, int outfd, unsigned int count)
+{
+	while (count > 0) {
+		ssize_t r;
+
+		r = sendfile(outfd, infd, NULL, count);
+		if (r < 0) {
+			perror("sendfile");
+			return 3;
+		}
+
+		count -= r;
+	}
+
+	return 0;
+}
+
+static int copyfd_io_mmap(int infd, int peerfd, int outfd,
+			  unsigned int size)
+{
+	int err;
+
+	if (listen_mode) {
+		err = do_recvfile(peerfd, outfd);
+		if (err)
+			return err;
+
+		err = do_mmap(infd, peerfd, size);
+	} else {
+		err = do_mmap(infd, peerfd, size);
+		if (err)
+			return err;
+
+		shutdown(peerfd, SHUT_WR);
+
+		err = do_recvfile(peerfd, outfd);
+	}
+
+	return err;
+}
+
+static int copyfd_io_sendfile(int infd, int peerfd, int outfd,
+			      unsigned int size)
+{
+	int err;
+
+	if (listen_mode) {
+		err = do_recvfile(peerfd, outfd);
+		if (err)
+			return err;
+
+		err = do_sendfile(infd, peerfd, size);
+	} else {
+		err = do_sendfile(infd, peerfd, size);
+		if (err)
+			return err;
+		err = do_recvfile(peerfd, outfd);
+	}
+
+	return err;
+}
+
+static int copyfd_io(int infd, int peerfd, int outfd)
+{
+	int file_size;
+
+	switch (cfg_mode) {
+	case CFG_MODE_POLL:
+		return copyfd_io_poll(infd, peerfd, outfd);
+	case CFG_MODE_MMAP:
+		file_size = get_infd_size(infd);
+		if (file_size < 0)
+			return file_size;
+		return copyfd_io_mmap(infd, peerfd, outfd, file_size);
+	case CFG_MODE_SENDFILE:
+		file_size = get_infd_size(infd);
+		if (file_size < 0)
+			return file_size;
+		return copyfd_io_sendfile(infd, peerfd, outfd, file_size);
+	}
+
+	fprintf(stderr, "Invalid mode %d\n", cfg_mode);
+
+	die_usage();
+	return 1;
+}
+
+static void check_sockaddr(int pf, struct sockaddr_storage *ss,
+			   socklen_t salen)
+{
+	struct sockaddr_in6 *sin6;
+	struct sockaddr_in *sin;
+	socklen_t wanted_size = 0;
+
+	switch (pf) {
+	case AF_INET:
+		wanted_size = sizeof(*sin);
+		sin = (void *)ss;
+		if (!sin->sin_port)
+			fprintf(stderr, "accept: something wrong: ip connection from port 0");
+		break;
+	case AF_INET6:
+		wanted_size = sizeof(*sin6);
+		sin6 = (void *)ss;
+		if (!sin6->sin6_port)
+			fprintf(stderr, "accept: something wrong: ipv6 connection from port 0");
+		break;
+	default:
+		fprintf(stderr, "accept: Unknown pf %d, salen %u\n", pf, salen);
+		return;
+	}
+
+	if (salen != wanted_size)
+		fprintf(stderr, "accept: size mismatch, got %d expected %d\n",
+			(int)salen, wanted_size);
+
+	if (ss->ss_family != pf)
+		fprintf(stderr, "accept: pf mismatch, expect %d, ss_family is %d\n",
+			(int)ss->ss_family, pf);
+}
+
+static void check_getpeername(int fd, struct sockaddr_storage *ss, socklen_t salen)
+{
+	struct sockaddr_storage peerss;
+	socklen_t peersalen = sizeof(peerss);
+
+	if (getpeername(fd, (struct sockaddr *)&peerss, &peersalen) < 0) {
+		perror("getpeername");
+		return;
+	}
+
+	if (peersalen != salen) {
+		fprintf(stderr, "%s: %d vs %d\n", __func__, peersalen, salen);
+		return;
+	}
+
+	if (memcmp(ss, &peerss, peersalen)) {
+		char a[INET6_ADDRSTRLEN];
+		char b[INET6_ADDRSTRLEN];
+		char c[INET6_ADDRSTRLEN];
+		char d[INET6_ADDRSTRLEN];
+
+		xgetnameinfo((struct sockaddr *)ss, salen,
+			     a, sizeof(a), b, sizeof(b));
+
+		xgetnameinfo((struct sockaddr *)&peerss, peersalen,
+			     c, sizeof(c), d, sizeof(d));
+
+		fprintf(stderr, "%s: memcmp failure: accept %s vs peername %s, %s vs %s salen %d vs %d\n",
+			__func__, a, c, b, d, peersalen, salen);
+	}
+}
+
+static void check_getpeername_connect(int fd)
+{
+	struct sockaddr_storage ss;
+	socklen_t salen = sizeof(ss);
+	char a[INET6_ADDRSTRLEN];
+	char b[INET6_ADDRSTRLEN];
+
+	if (getpeername(fd, (struct sockaddr *)&ss, &salen) < 0) {
+		perror("getpeername");
+		return;
+	}
+
+	xgetnameinfo((struct sockaddr *)&ss, salen,
+		     a, sizeof(a), b, sizeof(b));
+
+	if (strcmp(cfg_host, a) || strcmp(cfg_port, b))
+		fprintf(stderr, "%s: %s vs %s, %s vs %s\n", __func__,
+			cfg_host, a, cfg_port, b);
+}
+
+int main_loop_s(int listensock)
+{
+	struct sockaddr_storage ss;
+	struct pollfd polls;
+	socklen_t salen;
+	int remotesock;
+
+	polls.fd = listensock;
+	polls.events = POLLIN;
+
+	switch (poll(&polls, 1, poll_timeout)) {
+	case -1:
+		perror("poll");
+		return 1;
+	case 0:
+		fprintf(stderr, "%s: timed out\n", __func__);
+		close(listensock);
+		return 2;
+	}
+
+	salen = sizeof(ss);
+	remotesock = accept(listensock, (struct sockaddr *)&ss, &salen);
+	if (remotesock >= 0) {
+		check_sockaddr(pf, &ss, salen);
+		check_getpeername(remotesock, &ss, salen);
+
+		return copyfd_io(0, remotesock, 1);
+	}
+
+	perror("accept");
+
+	return 1;
+}
+
+static void init_rng(void)
+{
+	int fd = open("/dev/urandom", O_RDONLY);
+	unsigned int foo;
+
+	if (fd > 0) {
+		int ret = read(fd, &foo, sizeof(foo));
+
+		if (ret < 0)
+			srand(fd + foo);
+		close(fd);
+	}
+
+	srand(foo);
+}
+
+int main_loop(void)
+{
+	int fd;
+
+	/* listener is ready. */
+	fd = sock_connect_mptcp(cfg_host, cfg_port, cfg_sock_proto);
+	if (fd < 0)
+		return 2;
+
+	check_getpeername_connect(fd);
+
+	if (cfg_sndbuf)
+		set_sndbuf(fd, cfg_sndbuf);
+
+	return copyfd_io(0, fd, 1);
+}
+
+int parse_proto(const char *proto)
+{
+	if (!strcasecmp(proto, "MPTCP"))
+		return IPPROTO_MPTCP;
+	if (!strcasecmp(proto, "TCP"))
+		return IPPROTO_TCP;
+
+	fprintf(stderr, "Unknown protocol: %s\n.", proto);
+	die_usage();
+
+	/* silence compiler warning */
+	return 0;
+}
+
+int parse_mode(const char *mode)
+{
+	if (!strcasecmp(mode, "poll"))
+		return CFG_MODE_POLL;
+	if (!strcasecmp(mode, "mmap"))
+		return CFG_MODE_MMAP;
+	if (!strcasecmp(mode, "sendfile"))
+		return CFG_MODE_SENDFILE;
+
+	fprintf(stderr, "Unknown test mode: %s\n", mode);
+	fprintf(stderr, "Supported modes are:\n");
+	fprintf(stderr, "\t\t\"poll\" - interleaved read/write using poll()\n");
+	fprintf(stderr, "\t\t\"mmap\" - send entire input file (mmap+write), then read response (-l will read input first)\n");
+	fprintf(stderr, "\t\t\"sendfile\" - send entire input file (sendfile), then read response (-l will read input first)\n");
+
+	die_usage();
+
+	/* silence compiler warning */
+	return 0;
+}
+
+int parse_sndbuf(const char *size)
+{
+	unsigned long s;
+
+	errno = 0;
+
+	s = strtoul(size, NULL, 0);
+
+	if (errno) {
+		fprintf(stderr, "Invalid sndbuf size %s (%s)\n",
+			size, strerror(errno));
+		die_usage();
+	}
+
+	if (s > INT_MAX) {
+		fprintf(stderr, "Invalid sndbuf size %s (%s)\n",
+			size, strerror(ERANGE));
+		die_usage();
+	}
+
+	cfg_sndbuf = s;
+
+	return 0;
+}
+
+static void parse_opts(int argc, char **argv)
+{
+	int c;
+
+	while ((c = getopt(argc, argv, "6lp:s:hut:m:b:")) != -1) {
+		switch (c) {
+		case 'l':
+			listen_mode = true;
+			break;
+		case 'p':
+			cfg_port = optarg;
+			break;
+		case 's':
+			cfg_sock_proto = parse_proto(optarg);
+			break;
+		case 'h':
+			die_usage();
+			break;
+		case 'u':
+			tcpulp_audit = true;
+			break;
+		case '6':
+			pf = AF_INET6;
+			break;
+		case 't':
+			poll_timeout = atoi(optarg) * 1000;
+			if (poll_timeout <= 0)
+				poll_timeout = -1;
+			break;
+		case 'm':
+			cfg_mode = parse_mode(optarg);
+			break;
+		case 'b':
+			cfg_sndbuf = parse_sndbuf(optarg);
+			break;
+		}
+	}
+
+	if (optind + 1 != argc)
+		die_usage();
+	cfg_host = argv[optind];
+
+	if (strchr(cfg_host, ':'))
+		pf = AF_INET6;
+}
+
+int main(int argc, char *argv[])
+{
+	init_rng();
+
+	parse_opts(argc, argv);
+
+	if (tcpulp_audit)
+		return sock_test_tcpulp(cfg_host, cfg_port) ? 0 : 1;
+
+	if (listen_mode) {
+		int fd = sock_listen_mptcp(cfg_host, cfg_port);
+
+		if (fd < 0)
+			return 1;
+
+		if (cfg_sndbuf)
+			set_sndbuf(fd, cfg_sndbuf);
+
+		return main_loop_s(fd);
+	}
+
+	return main_loop();
+}
diff --git a/tools/testing/selftests/net/mptcp/mptcp_connect.sh b/tools/testing/selftests/net/mptcp/mptcp_connect.sh
new file mode 100755
index 000000000000..d573a0feb98d
--- /dev/null
+++ b/tools/testing/selftests/net/mptcp/mptcp_connect.sh
@@ -0,0 +1,595 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+time_start=$(date +%s)
+
+optstring="b:d:e:l:r:h4cm:"
+ret=0
+sin=""
+sout=""
+cin=""
+cout=""
+ksft_skip=4
+capture=false
+timeout=30
+ipv6=true
+ethtool_random_on=true
+tc_delay="$((RANDOM%400))"
+tc_loss=$((RANDOM%101))
+tc_reorder=""
+testmode=""
+sndbuf=0
+options_log=true
+
+if [ $tc_loss -eq 100 ];then
+	tc_loss=1%
+elif [ $tc_loss -ge 10 ]; then
+	tc_loss=0.$tc_loss%
+elif [ $tc_loss -ge 1 ]; then
+	tc_loss=0.0$tc_loss%
+else
+	tc_loss=""
+fi
+
+usage() {
+	echo "Usage: $0 [ -a ]"
+	echo -e "\t-d: tc/netem delay in milliseconds, e.g. \"-d 10\" (default random)"
+	echo -e "\t-l: tc/netem loss percentage, e.g. \"-l 0.02\" (default random)"
+	echo -e "\t-r: tc/netem reorder mode, e.g. \"-r 25% 50% gap 5\", use "-r 0" to disable reordering (default random)"
+	echo -e "\t-e: ethtool features to disable, e.g.: \"-e tso -e gso\" (default: randomly disable any of tso/gso/gro)"
+	echo -e "\t-4: IPv4 only: disable IPv6 tests (default: test both IPv4 and IPv6)"
+	echo -e "\t-c: capture packets for each test using tcpdump (default: no capture)"
+	echo -e "\t-b: set sndbuf value (default: use kernel default)"
+	echo -e "\t-m: test mode (poll, sendfile; default: poll)"
+}
+
+while getopts "$optstring" option;do
+	case "$option" in
+	"h")
+		usage $0
+		exit 0
+		;;
+	"d")
+		if [ $OPTARG -ge 0 ];then
+			tc_delay="$OPTARG"
+		else
+			echo "-d requires numeric argument, got \"$OPTARG\"" 1>&2
+			exit 1
+		fi
+		;;
+	"e")
+		ethtool_args="$ethtool_args $OPTARG off"
+		ethtool_random_on=false
+		;;
+	"l")
+		tc_loss="$OPTARG"
+		;;
+	"r")
+		tc_reorder="$OPTARG"
+		;;
+	"4")
+		ipv6=false
+		;;
+	"c")
+		capture=true
+		;;
+	"b")
+		if [ $OPTARG -ge 0 ];then
+			sndbuf="$OPTARG"
+		else
+			echo "-s requires numeric argument, got \"$OPTARG\"" 1>&2
+			exit 1
+		fi
+		;;
+	"m")
+		testmode="$OPTARG"
+		;;
+	"?")
+		usage $0
+		exit 1
+		;;
+	esac
+done
+
+sec=$(date +%s)
+rndh=$(printf %x $sec)-$(mktemp -u XXXXXX)
+ns1="ns1-$rndh"
+ns2="ns2-$rndh"
+ns3="ns3-$rndh"
+ns4="ns4-$rndh"
+
+TEST_COUNT=0
+
+cleanup()
+{
+	rm -f "$cin" "$cout"
+	rm -f "$sin" "$sout"
+	rm -f "$capout"
+
+	local netns
+	for netns in "$ns1" "$ns2" "$ns3" "$ns4";do
+		ip netns del $netns
+	done
+}
+
+ip -Version > /dev/null 2>&1
+if [ $? -ne 0 ];then
+	echo "SKIP: Could not run test without ip tool"
+	exit $ksft_skip
+fi
+
+sin=$(mktemp)
+sout=$(mktemp)
+cin=$(mktemp)
+cout=$(mktemp)
+capout=$(mktemp)
+trap cleanup EXIT
+
+for i in "$ns1" "$ns2" "$ns3" "$ns4";do
+	ip netns add $i || exit $ksft_skip
+	ip -net $i link set lo up
+done
+
+#  "$ns1"              ns2                    ns3                     ns4
+# ns1eth2    ns2eth1   ns2eth3      ns3eth2   ns3eth4       ns4eth3
+#                           - drop 1% ->            reorder 25%
+#                           <- TSO off -
+
+ip link add ns1eth2 netns "$ns1" type veth peer name ns2eth1 netns "$ns2"
+ip link add ns2eth3 netns "$ns2" type veth peer name ns3eth2 netns "$ns3"
+ip link add ns3eth4 netns "$ns3" type veth peer name ns4eth3 netns "$ns4"
+
+ip -net "$ns1" addr add 10.0.1.1/24 dev ns1eth2
+ip -net "$ns1" addr add dead:beef:1::1/64 dev ns1eth2 nodad
+
+ip -net "$ns1" link set ns1eth2 up
+ip -net "$ns1" route add default via 10.0.1.2
+ip -net "$ns1" route add default via dead:beef:1::2
+
+ip -net "$ns2" addr add 10.0.1.2/24 dev ns2eth1
+ip -net "$ns2" addr add dead:beef:1::2/64 dev ns2eth1 nodad
+ip -net "$ns2" link set ns2eth1 up
+
+ip -net "$ns2" addr add 10.0.2.1/24 dev ns2eth3
+ip -net "$ns2" addr add dead:beef:2::1/64 dev ns2eth3 nodad
+ip -net "$ns2" link set ns2eth3 up
+ip -net "$ns2" route add default via 10.0.2.2
+ip -net "$ns2" route add default via dead:beef:2::2
+ip netns exec "$ns2" sysctl -q net.ipv4.ip_forward=1
+ip netns exec "$ns2" sysctl -q net.ipv6.conf.all.forwarding=1
+
+ip -net "$ns3" addr add 10.0.2.2/24 dev ns3eth2
+ip -net "$ns3" addr add dead:beef:2::2/64 dev ns3eth2 nodad
+ip -net "$ns3" link set ns3eth2 up
+
+ip -net "$ns3" addr add 10.0.3.2/24 dev ns3eth4
+ip -net "$ns3" addr add dead:beef:3::2/64 dev ns3eth4 nodad
+ip -net "$ns3" link set ns3eth4 up
+ip -net "$ns3" route add default via 10.0.2.1
+ip -net "$ns3" route add default via dead:beef:2::1
+ip netns exec "$ns3" sysctl -q net.ipv4.ip_forward=1
+ip netns exec "$ns3" sysctl -q net.ipv6.conf.all.forwarding=1
+
+ip -net "$ns4" addr add 10.0.3.1/24 dev ns4eth3
+ip -net "$ns4" addr add dead:beef:3::1/64 dev ns4eth3 nodad
+ip -net "$ns4" link set ns4eth3 up
+ip -net "$ns4" route add default via 10.0.3.2
+ip -net "$ns4" route add default via dead:beef:3::2
+
+set_ethtool_flags() {
+	local ns="$1"
+	local dev="$2"
+	local flags="$3"
+
+	ip netns exec $ns ethtool -K $dev $flags 2>/dev/null
+	[ $? -eq 0 ] && echo "INFO: set $ns dev $dev: ethtool -K $flags"
+}
+
+set_random_ethtool_flags() {
+	local flags=""
+	local r=$RANDOM
+
+	local pick1=$((r & 1))
+	local pick2=$((r & 2))
+	local pick3=$((r & 4))
+
+	[ $pick1 -ne 0 ] && flags="tso off"
+	[ $pick2 -ne 0 ] && flags="$flags gso off"
+	[ $pick3 -ne 0 ] && flags="$flags gro off"
+
+	[ -z "$flags" ] && return
+
+	set_ethtool_flags "$1" "$2" "$flags"
+}
+
+if $ethtool_random_on;then
+	set_random_ethtool_flags "$ns3" ns3eth2
+	set_random_ethtool_flags "$ns4" ns4eth3
+else
+	set_ethtool_flags "$ns3" ns3eth2 "$ethtool_args"
+	set_ethtool_flags "$ns4" ns4eth3 "$ethtool_args"
+fi
+
+print_file_err()
+{
+	ls -l "$1" 1>&2
+	echo "Trailing bytes are: "
+	tail -c 27 "$1"
+}
+
+check_transfer()
+{
+	local in=$1
+	local out=$2
+	local what=$3
+
+	cmp "$in" "$out" > /dev/null 2>&1
+	if [ $? -ne 0 ] ;then
+		echo "[ FAIL ] $what does not match (in, out):"
+		print_file_err "$in"
+		print_file_err "$out"
+
+		return 1
+	fi
+
+	return 0
+}
+
+check_mptcp_disabled()
+{
+	local disabled_ns
+	disabled_ns="ns_disabled-$sech-$(mktemp -u XXXXXX)"
+	ip netns add ${disabled_ns} || exit $ksft_skip
+
+	# net.mptcp.enabled should be enabled by default
+	if [ "$(ip netns exec ${disabled_ns} sysctl net.mptcp.enabled | awk '{ print $3 }')" -ne 1 ]; then
+		echo -e "net.mptcp.enabled sysctl is not 1 by default\t\t[ FAIL ]"
+		ret=1
+		return 1
+	fi
+	ip netns exec ${disabled_ns} sysctl -q net.mptcp.enabled=0
+
+	local err=0
+	LANG=C ip netns exec ${disabled_ns} ./mptcp_connect -t $timeout -p 10000 -s MPTCP 127.0.0.1 < "$cin" 2>&1 | \
+		grep -q "^socket: Protocol not available$" && err=1
+	ip netns delete ${disabled_ns}
+
+	if [ ${err} -eq 0 ]; then
+		echo -e "New MPTCP socket cannot be blocked via sysctl\t\t[ FAIL ]"
+		ret=1
+		return 1
+	fi
+
+	echo -e "New MPTCP socket can be blocked via sysctl\t\t[ OK ]"
+	return 0
+}
+
+check_mptcp_ulp_setsockopt()
+{
+	local t retval
+	t="ns_ulp-$sech-$(mktemp -u XXXXXX)"
+
+	ip netns add ${t} || exit $ksft_skip
+	if ! ip netns exec ${t} ./mptcp_connect -u -p 10000 -s TCP 127.0.0.1 2>&1; then
+		printf "setsockopt(..., TCP_ULP, \"mptcp\", ...) allowed\t[ FAIL ]\n"
+		retval=1
+		ret=$retval
+	else
+		printf "setsockopt(..., TCP_ULP, \"mptcp\", ...) blocked\t[ OK ]\n"
+		retval=0
+	fi
+	ip netns del ${t}
+	return $retval
+}
+
+# $1: IP address
+is_v6()
+{
+	[ -z "${1##*:*}" ]
+}
+
+do_ping()
+{
+	local listener_ns="$1"
+	local connector_ns="$2"
+	local connect_addr="$3"
+	local ping_args="-q -c 1"
+
+	if is_v6 "${connect_addr}"; then
+		$ipv6 || return 0
+		ping_args="${ping_args} -6"
+	fi
+
+	ip netns exec ${connector_ns} ping ${ping_args} $connect_addr >/dev/null
+	if [ $? -ne 0 ] ; then
+		echo "$listener_ns -> $connect_addr connectivity [ FAIL ]" 1>&2
+		ret=1
+
+		return 1
+	fi
+
+	return 0
+}
+
+# $1: ns, $2: port
+wait_local_port_listen()
+{
+	local listener_ns="${1}"
+	local port="${2}"
+
+	local port_hex i
+
+	port_hex="$(printf "%04X" "${port}")"
+	for i in $(seq 10); do
+		ip netns exec "${listener_ns}" cat /proc/net/tcp* | \
+			awk "BEGIN {rc=1} {if (\$2 ~ /:${port_hex}\$/ && \$4 ~ /0A/) {rc=0; exit}} END {exit rc}" &&
+			break
+		sleep 0.1
+	done
+}
+
+do_transfer()
+{
+	local listener_ns="$1"
+	local connector_ns="$2"
+	local cl_proto="$3"
+	local srv_proto="$4"
+	local connect_addr="$5"
+	local local_addr="$6"
+	local extra_args=""
+
+	local port
+	port=$((10000+$TEST_COUNT))
+	TEST_COUNT=$((TEST_COUNT+1))
+
+	if [ "$sndbuf" -gt 0 ]; then
+		extra_args="$extra_args -b $sndbuf"
+	fi
+
+	if [ -n "$testmode" ]; then
+		extra_args="$extra_args -m $testmode"
+	fi
+
+	if [ -n "$extra_args" ] && $options_log; then
+		options_log=false
+		echo "INFO: extra options: $extra_args"
+	fi
+
+	:> "$cout"
+	:> "$sout"
+	:> "$capout"
+
+	local addr_port
+	addr_port=$(printf "%s:%d" ${connect_addr} ${port})
+	printf "%.3s %-5s -> %.3s (%-20s) %-5s\t" ${connector_ns} ${cl_proto} ${listener_ns} ${addr_port} ${srv_proto}
+
+	if $capture; then
+		local capuser
+		if [ -z $SUDO_USER ] ; then
+			capuser=""
+		else
+			capuser="-Z $SUDO_USER"
+		fi
+
+		local capfile="${listener_ns}-${connector_ns}-${cl_proto}-${srv_proto}-${connect_addr}.pcap"
+
+		ip netns exec ${listener_ns} tcpdump -i any -s 65535 -B 32768 $capuser -w $capfile > "$capout" 2>&1 &
+		local cappid=$!
+
+		sleep 1
+	fi
+
+	ip netns exec ${listener_ns} ./mptcp_connect -t $timeout -l -p $port -s ${srv_proto} $extra_args $local_addr < "$sin" > "$sout" &
+	local spid=$!
+
+	wait_local_port_listen "${listener_ns}" "${port}"
+
+	local start
+	start=$(date +%s%3N)
+	ip netns exec ${connector_ns} ./mptcp_connect -t $timeout -p $port -s ${cl_proto} $extra_args $connect_addr < "$cin" > "$cout" &
+	local cpid=$!
+
+	wait $cpid
+	local retc=$?
+	wait $spid
+	local rets=$?
+
+	local stop
+	stop=$(date +%s%3N)
+
+	if $capture; then
+		sleep 1
+		kill $cappid
+	fi
+
+	local duration
+	duration=$((stop-start))
+	duration=$(printf "(duration %05sms)" $duration)
+	if [ ${rets} -ne 0 ] || [ ${retc} -ne 0 ]; then
+		echo "$duration [ FAIL ] client exit code $retc, server $rets" 1>&2
+		echo "\nnetns ${listener_ns} socket stat for $port:" 1>&2
+		ip netns exec ${listener_ns} ss -nita 1>&2 -o "sport = :$port"
+		echo "\nnetns ${connector_ns} socket stat for $port:" 1>&2
+		ip netns exec ${connector_ns} ss -nita 1>&2 -o "dport = :$port"
+
+		cat "$capout"
+		return 1
+	fi
+
+	check_transfer $sin $cout "file received by client"
+	retc=$?
+	check_transfer $cin $sout "file received by server"
+	rets=$?
+
+	if [ $retc -eq 0 ] && [ $rets -eq 0 ];then
+		echo "$duration [ OK ]"
+		cat "$capout"
+		return 0
+	fi
+
+	cat "$capout"
+	return 1
+}
+
+make_file()
+{
+	local name=$1
+	local who=$2
+
+	local SIZE TSIZE
+	SIZE=$((RANDOM % (1024 * 8)))
+	TSIZE=$((SIZE * 1024))
+
+	dd if=/dev/urandom of="$name" bs=1024 count=$SIZE 2> /dev/null
+
+	SIZE=$((RANDOM % 1024))
+	SIZE=$((SIZE + 128))
+	TSIZE=$((TSIZE + SIZE))
+	dd if=/dev/urandom conv=notrunc of="$name" bs=1 count=$SIZE 2> /dev/null
+	echo -e "\nMPTCP_TEST_FILE_END_MARKER" >> "$name"
+
+	echo "Created $name (size $TSIZE) containing data sent by $who"
+}
+
+run_tests_lo()
+{
+	local listener_ns="$1"
+	local connector_ns="$2"
+	local connect_addr="$3"
+	local loopback="$4"
+	local lret=0
+
+	# skip if test programs are running inside same netns for subsequent runs.
+	if [ $loopback -eq 0 ] && [ ${listener_ns} = ${connector_ns} ]; then
+		return 0
+	fi
+
+	# skip if we don't want v6
+	if ! $ipv6 && is_v6 "${connect_addr}"; then
+		return 0
+	fi
+
+	local local_addr
+	if is_v6 "${connect_addr}"; then
+		local_addr="::"
+	else
+		local_addr="0.0.0.0"
+	fi
+
+	do_transfer ${listener_ns} ${connector_ns} MPTCP MPTCP ${connect_addr} ${local_addr}
+	lret=$?
+	if [ $lret -ne 0 ]; then
+		ret=$lret
+		return 1
+	fi
+
+	# don't bother testing fallback tcp except for loopback case.
+	if [ ${listener_ns} != ${connector_ns} ]; then
+		return 0
+	fi
+
+	do_transfer ${listener_ns} ${connector_ns} MPTCP TCP ${connect_addr} ${local_addr}
+	lret=$?
+	if [ $lret -ne 0 ]; then
+		ret=$lret
+		return 1
+	fi
+
+	do_transfer ${listener_ns} ${connector_ns} TCP MPTCP ${connect_addr} ${local_addr}
+	lret=$?
+	if [ $lret -ne 0 ]; then
+		ret=$lret
+		return 1
+	fi
+
+	return 0
+}
+
+run_tests()
+{
+	run_tests_lo $1 $2 $3 0
+}
+
+make_file "$cin" "client"
+make_file "$sin" "server"
+
+check_mptcp_disabled
+
+check_mptcp_ulp_setsockopt
+
+echo "INFO: validating network environment with pings"
+for sender in "$ns1" "$ns2" "$ns3" "$ns4";do
+	do_ping "$ns1" $sender 10.0.1.1
+	do_ping "$ns1" $sender dead:beef:1::1
+
+	do_ping "$ns2" $sender 10.0.1.2
+	do_ping "$ns2" $sender dead:beef:1::2
+	do_ping "$ns2" $sender 10.0.2.1
+	do_ping "$ns2" $sender dead:beef:2::1
+
+	do_ping "$ns3" $sender 10.0.2.2
+	do_ping "$ns3" $sender dead:beef:2::2
+	do_ping "$ns3" $sender 10.0.3.2
+	do_ping "$ns3" $sender dead:beef:3::2
+
+	do_ping "$ns4" $sender 10.0.3.1
+	do_ping "$ns4" $sender dead:beef:3::1
+done
+
+[ -n "$tc_loss" ] && tc -net "$ns2" qdisc add dev ns2eth3 root netem loss random $tc_loss
+echo -n "INFO: Using loss of $tc_loss "
+test "$tc_delay" -gt 0 && echo -n "delay $tc_delay ms "
+
+if [ -z "${tc_reorder}" ]; then
+	reorder1=$((RANDOM%10))
+	reorder1=$((100 - reorder1))
+	reorder2=$((RANDOM%100))
+
+	if [ $tc_delay -gt 0 ] && [ $reorder1 -lt 100 ] && [ $reorder2 -gt 0 ]; then
+		tc_reorder="reorder ${reorder1}% ${reorder2}%"
+		echo -n "$tc_reorder "
+	fi
+elif [ "$tc_reorder" = "0" ];then
+	tc_reorder=""
+elif [ "$tc_delay" -gt 0 ];then
+	# reordering requires some delay
+	tc_reorder="reorder $tc_reorder"
+	echo -n "$tc_reorder "
+fi
+
+echo "on ns3eth4"
+
+tc -net "$ns3" qdisc add dev ns3eth4 root netem delay ${tc_delay}ms $tc_reorder
+
+for sender in $ns1 $ns2 $ns3 $ns4;do
+	run_tests_lo "$ns1" "$sender" 10.0.1.1 1
+	if [ $ret -ne 0 ] ;then
+		echo "FAIL: Could not even run loopback test" 1>&2
+		exit $ret
+	fi
+	run_tests_lo "$ns1" $sender dead:beef:1::1 1
+	if [ $ret -ne 0 ] ;then
+		echo "FAIL: Could not even run loopback v6 test" 2>&1
+		exit $ret
+	fi
+
+	run_tests "$ns2" $sender 10.0.1.2
+	run_tests "$ns2" $sender dead:beef:1::2
+	run_tests "$ns2" $sender 10.0.2.1
+	run_tests "$ns2" $sender dead:beef:2::1
+
+	run_tests "$ns3" $sender 10.0.2.2
+	run_tests "$ns3" $sender dead:beef:2::2
+	run_tests "$ns3" $sender 10.0.3.2
+	run_tests "$ns3" $sender dead:beef:3::2
+
+	run_tests "$ns4" $sender 10.0.3.1
+	run_tests "$ns4" $sender dead:beef:3::1
+done
+
+time_end=$(date +%s)
+time_run=$((time_end-time_start))
+
+echo "Time: ${time_run} seconds"
+
+exit $ret
diff --git a/tools/testing/selftests/net/mptcp/settings b/tools/testing/selftests/net/mptcp/settings
new file mode 100644
index 000000000000..026384c189c9
--- /dev/null
+++ b/tools/testing/selftests/net/mptcp/settings
@@ -0,0 +1 @@
+timeout=450
-- 
cgit v1.2.3-59-g8ed1b


From 65492c5a6ab5df5091a77562dbcca2d2dc3877c0 Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Tue, 21 Jan 2020 16:56:30 -0800
Subject: mptcp: move from sha1 (v0) to sha256 (v1)

For simplicity's sake use directly sha256 primitives (and pull them
as a required build dep).
Add optional, boot-time self-tests for the hmac function.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/mptcp/Kconfig    |  10 ++++
 net/mptcp/crypto.c   | 140 +++++++++++++++++++++++++++++++--------------------
 net/mptcp/options.c  |   9 +++-
 net/mptcp/protocol.h |   4 +-
 4 files changed, 104 insertions(+), 59 deletions(-)

diff --git a/net/mptcp/Kconfig b/net/mptcp/Kconfig
index c1a99f07a4cd..5db56d2218c5 100644
--- a/net/mptcp/Kconfig
+++ b/net/mptcp/Kconfig
@@ -3,6 +3,7 @@ config MPTCP
 	bool "MPTCP: Multipath TCP"
 	depends on INET
 	select SKB_EXTENSIONS
+	select CRYPTO_LIB_SHA256
 	help
 	  Multipath TCP (MPTCP) connections send and receive data over multiple
 	  subflows in order to utilize multiple network paths. Each subflow
@@ -14,3 +15,12 @@ config MPTCP_IPV6
 	depends on MPTCP
 	select IPV6
 	default y
+
+config MPTCP_HMAC_TEST
+	bool "Tests for MPTCP HMAC implementation"
+	default n
+	help
+	  This option enable boot time self-test for the HMAC implementation
+	  used by the MPTCP code
+
+	  Say N if you are unsure.
diff --git a/net/mptcp/crypto.c b/net/mptcp/crypto.c
index bbd6d01af211..40d1bb18fd60 100644
--- a/net/mptcp/crypto.c
+++ b/net/mptcp/crypto.c
@@ -21,67 +21,36 @@
  */
 
 #include <linux/kernel.h>
-#include <linux/cryptohash.h>
+#include <crypto/sha.h>
 #include <asm/unaligned.h>
 
 #include "protocol.h"
 
-struct sha1_state {
-	u32 workspace[SHA_WORKSPACE_WORDS];
-	u32 digest[SHA_DIGEST_WORDS];
-	unsigned int count;
-};
-
-static void sha1_init(struct sha1_state *state)
-{
-	sha_init(state->digest);
-	state->count = 0;
-}
-
-static void sha1_update(struct sha1_state *state, u8 *input)
-{
-	sha_transform(state->digest, input, state->workspace);
-	state->count += SHA_MESSAGE_BYTES;
-}
-
-static void sha1_pad_final(struct sha1_state *state, u8 *input,
-			   unsigned int length, __be32 *mptcp_hashed_key)
-{
-	int i;
-
-	input[length] = 0x80;
-	memset(&input[length + 1], 0, SHA_MESSAGE_BYTES - length - 9);
-	put_unaligned_be64((length + state->count) << 3,
-			   &input[SHA_MESSAGE_BYTES - 8]);
-
-	sha_transform(state->digest, input, state->workspace);
-	for (i = 0; i < SHA_DIGEST_WORDS; ++i)
-		put_unaligned_be32(state->digest[i], &mptcp_hashed_key[i]);
-
-	memzero_explicit(state->workspace, SHA_WORKSPACE_WORDS << 2);
-}
+#define SHA256_DIGEST_WORDS (SHA256_DIGEST_SIZE / 4)
 
 void mptcp_crypto_key_sha(u64 key, u32 *token, u64 *idsn)
 {
-	__be32 mptcp_hashed_key[SHA_DIGEST_WORDS];
-	u8 input[SHA_MESSAGE_BYTES];
-	struct sha1_state state;
+	__be32 mptcp_hashed_key[SHA256_DIGEST_WORDS];
+	__be64 input = cpu_to_be64(key);
+	struct sha256_state state;
 
-	sha1_init(&state);
-	put_unaligned_be64(key, input);
-	sha1_pad_final(&state, input, 8, mptcp_hashed_key);
+	sha256_init(&state);
+	sha256_update(&state, (__force u8 *)&input, sizeof(input));
+	sha256_final(&state, (u8 *)mptcp_hashed_key);
 
 	if (token)
 		*token = be32_to_cpu(mptcp_hashed_key[0]);
 	if (idsn)
-		*idsn = be64_to_cpu(*((__be64 *)&mptcp_hashed_key[3]));
+		*idsn = be64_to_cpu(*((__be64 *)&mptcp_hashed_key[6]));
 }
 
 void mptcp_crypto_hmac_sha(u64 key1, u64 key2, u32 nonce1, u32 nonce2,
-			   u32 *hash_out)
+			   void *hmac)
 {
-	u8 input[SHA_MESSAGE_BYTES * 2];
-	struct sha1_state state;
+	u8 input[SHA256_BLOCK_SIZE + SHA256_DIGEST_SIZE];
+	__be32 mptcp_hashed_key[SHA256_DIGEST_WORDS];
+	__be32 *hash_out = (__force __be32 *)hmac;
+	struct sha256_state state;
 	u8 key1be[8];
 	u8 key2be[8];
 	int i;
@@ -96,17 +65,16 @@ void mptcp_crypto_hmac_sha(u64 key1, u64 key2, u32 nonce1, u32 nonce2,
 	for (i = 0; i < 8; i++)
 		input[i + 8] ^= key2be[i];
 
-	put_unaligned_be32(nonce1, &input[SHA_MESSAGE_BYTES]);
-	put_unaligned_be32(nonce2, &input[SHA_MESSAGE_BYTES + 4]);
+	put_unaligned_be32(nonce1, &input[SHA256_BLOCK_SIZE]);
+	put_unaligned_be32(nonce2, &input[SHA256_BLOCK_SIZE + 4]);
 
-	sha1_init(&state);
-	sha1_update(&state, input);
+	sha256_init(&state);
+	sha256_update(&state, input, SHA256_BLOCK_SIZE + 8);
 
 	/* emit sha256(K1 || msg) on the second input block, so we can
 	 * reuse 'input' for the last hashing
 	 */
-	sha1_pad_final(&state, &input[SHA_MESSAGE_BYTES], 8,
-		       (__force __be32 *)&input[SHA_MESSAGE_BYTES]);
+	sha256_final(&state, &input[SHA256_BLOCK_SIZE]);
 
 	/* Prepare second part of hmac */
 	memset(input, 0x5C, SHA_MESSAGE_BYTES);
@@ -115,8 +83,70 @@ void mptcp_crypto_hmac_sha(u64 key1, u64 key2, u32 nonce1, u32 nonce2,
 	for (i = 0; i < 8; i++)
 		input[i + 8] ^= key2be[i];
 
-	sha1_init(&state);
-	sha1_update(&state, input);
-	sha1_pad_final(&state, &input[SHA_MESSAGE_BYTES], SHA_DIGEST_WORDS << 2,
-		       (__be32 *)hash_out);
+	sha256_init(&state);
+	sha256_update(&state, input, SHA256_BLOCK_SIZE + SHA256_DIGEST_SIZE);
+	sha256_final(&state, (u8 *)mptcp_hashed_key);
+
+	/* takes only first 160 bits */
+	for (i = 0; i < 5; i++)
+		hash_out[i] = mptcp_hashed_key[i];
+}
+
+#ifdef CONFIG_MPTCP_HMAC_TEST
+struct test_cast {
+	char *key;
+	char *msg;
+	char *result;
+};
+
+/* we can't reuse RFC 4231 test vectors, as we have constraint on the
+ * input and key size, and we truncate the output.
+ */
+static struct test_cast tests[] = {
+	{
+		.key = "0b0b0b0b0b0b0b0b",
+		.msg = "48692054",
+		.result = "8385e24fb4235ac37556b6b886db106284a1da67",
+	},
+	{
+		.key = "aaaaaaaaaaaaaaaa",
+		.msg = "dddddddd",
+		.result = "2c5e219164ff1dca1c4a92318d847bb6b9d44492",
+	},
+	{
+		.key = "0102030405060708",
+		.msg = "cdcdcdcd",
+		.result = "e73b9ba9969969cefb04aa0d6df18ec2fcc075b6",
+	},
+};
+
+static int __init test_mptcp_crypto(void)
+{
+	char hmac[20], hmac_hex[41];
+	u32 nonce1, nonce2;
+	u64 key1, key2;
+	int i, j;
+
+	for (i = 0; i < ARRAY_SIZE(tests); ++i) {
+		/* mptcp hmap will convert to be before computing the hmac */
+		key1 = be64_to_cpu(*((__be64 *)&tests[i].key[0]));
+		key2 = be64_to_cpu(*((__be64 *)&tests[i].key[8]));
+		nonce1 = be32_to_cpu(*((__be32 *)&tests[i].msg[0]));
+		nonce2 = be32_to_cpu(*((__be32 *)&tests[i].msg[4]));
+
+		mptcp_crypto_hmac_sha(key1, key2, nonce1, nonce2, hmac);
+		for (j = 0; j < 20; ++j)
+			sprintf(&hmac_hex[j << 1], "%02x", hmac[j] & 0xff);
+		hmac_hex[40] = 0;
+
+		if (memcmp(hmac_hex, tests[i].result, 40))
+			pr_err("test %d failed, got %s expected %s", i,
+			       hmac_hex, tests[i].result);
+		else
+			pr_info("test %d [ ok ]", i);
+	}
+	return 0;
 }
+
+late_initcall(test_mptcp_crypto);
+#endif
diff --git a/net/mptcp/options.c b/net/mptcp/options.c
index 1fa8496f3551..1aec742ca8e1 100644
--- a/net/mptcp/options.c
+++ b/net/mptcp/options.c
@@ -9,6 +9,11 @@
 #include <net/mptcp.h>
 #include "protocol.h"
 
+static bool mptcp_cap_flag_sha256(u8 flags)
+{
+	return (flags & MPTCP_CAP_FLAG_MASK) == MPTCP_CAP_HMAC_SHA256;
+}
+
 void mptcp_parse_option(const unsigned char *ptr, int opsize,
 			struct tcp_options_received *opt_rx)
 {
@@ -29,7 +34,7 @@ void mptcp_parse_option(const unsigned char *ptr, int opsize,
 			break;
 
 		flags = *ptr++;
-		if (!((flags & MPTCP_CAP_FLAG_MASK) == MPTCP_CAP_HMAC_SHA1) ||
+		if (!mptcp_cap_flag_sha256(flags) ||
 		    (flags & MPTCP_CAP_EXTENSIBILITY))
 			break;
 
@@ -399,7 +404,7 @@ void mptcp_write_options(__be32 *ptr, struct mptcp_out_options *opts)
 		*ptr++ = htonl((TCPOPT_MPTCP << 24) | (len << 16) |
 			       (MPTCPOPT_MP_CAPABLE << 12) |
 			       (MPTCP_SUPPORTED_VERSION << 8) |
-			       MPTCP_CAP_HMAC_SHA1);
+			       MPTCP_CAP_HMAC_SHA256);
 		put_unaligned_be64(opts->sndr_key, ptr);
 		ptr += 2;
 		if (OPTION_MPTCP_MPC_ACK & opts->suboptions) {
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index a8fad7d78565..a355bb1cf31b 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -43,7 +43,7 @@
 #define MPTCP_VERSION_MASK	(0x0F)
 #define MPTCP_CAP_CHECKSUM_REQD	BIT(7)
 #define MPTCP_CAP_EXTENSIBILITY	BIT(6)
-#define MPTCP_CAP_HMAC_SHA1	BIT(0)
+#define MPTCP_CAP_HMAC_SHA256	BIT(0)
 #define MPTCP_CAP_FLAG_MASK	(0x3F)
 
 /* MPTCP DSS flags */
@@ -216,7 +216,7 @@ static inline void mptcp_crypto_key_gen_sha(u64 *key, u32 *token, u64 *idsn)
 }
 
 void mptcp_crypto_hmac_sha(u64 key1, u64 key2, u32 nonce1, u32 nonce2,
-			   u32 *hash_out);
+			   void *hash_out);
 
 static inline struct mptcp_ext *mptcp_get_ext(struct sk_buff *skb)
 {
-- 
cgit v1.2.3-59-g8ed1b


From cc7972ea1932335e0a0ee00ac8a24b3e8304630d Mon Sep 17 00:00:00 2001
From: Christoph Paasch <cpaasch@apple.com>
Date: Tue, 21 Jan 2020 16:56:31 -0800
Subject: mptcp: parse and emit MP_CAPABLE option according to v1 spec

This implements MP_CAPABLE options parsing and writing according
to RFC 6824 bis / RFC 8684: MPTCP v1.

Local key is sent on syn/ack, and both keys are sent on 3rd ack.
MP_CAPABLE messages len are updated accordingly. We need the skbuff to
correctly emit the above, so we push the skbuff struct as an argument
all the way from tcp code to the relevant mptcp callbacks.

When processing incoming MP_CAPABLE + data, build a full blown DSS-like
map info, to simplify later processing.  On child socket creation, we
need to record the remote key, if available.

Signed-off-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/tcp.h   |   3 +-
 include/net/mptcp.h   |  17 +++---
 net/ipv4/tcp_input.c  |   2 +-
 net/ipv4/tcp_output.c |   2 +-
 net/mptcp/options.c   | 162 ++++++++++++++++++++++++++++++++++++++++----------
 net/mptcp/protocol.h  |   6 +-
 net/mptcp/subflow.c   |  14 ++++-
 7 files changed, 160 insertions(+), 46 deletions(-)

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 0d00dad4b85d..4e2124607d32 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -94,7 +94,8 @@ struct mptcp_options_received {
 		data_fin:1,
 		use_ack:1,
 		ack64:1,
-		__unused:3;
+		mpc_map:1,
+		__unused:2;
 };
 #endif
 
diff --git a/include/net/mptcp.h b/include/net/mptcp.h
index 8619c1fca741..27627e2d1bc2 100644
--- a/include/net/mptcp.h
+++ b/include/net/mptcp.h
@@ -23,7 +23,8 @@ struct mptcp_ext {
 			data_fin:1,
 			use_ack:1,
 			ack64:1,
-			__unused:3;
+			mpc_map:1,
+			__unused:2;
 	/* one byte hole */
 };
 
@@ -50,10 +51,10 @@ static inline bool rsk_is_mptcp(const struct request_sock *req)
 	return tcp_rsk(req)->is_mptcp;
 }
 
-void mptcp_parse_option(const unsigned char *ptr, int opsize,
-			struct tcp_options_received *opt_rx);
-bool mptcp_syn_options(struct sock *sk, unsigned int *size,
-		       struct mptcp_out_options *opts);
+void mptcp_parse_option(const struct sk_buff *skb, const unsigned char *ptr,
+			int opsize, struct tcp_options_received *opt_rx);
+bool mptcp_syn_options(struct sock *sk, const struct sk_buff *skb,
+		       unsigned int *size, struct mptcp_out_options *opts);
 void mptcp_rcv_synsent(struct sock *sk);
 bool mptcp_synack_options(const struct request_sock *req, unsigned int *size,
 			  struct mptcp_out_options *opts);
@@ -121,12 +122,14 @@ static inline bool rsk_is_mptcp(const struct request_sock *req)
 	return false;
 }
 
-static inline void mptcp_parse_option(const unsigned char *ptr, int opsize,
+static inline void mptcp_parse_option(const struct sk_buff *skb,
+				      const unsigned char *ptr, int opsize,
 				      struct tcp_options_received *opt_rx)
 {
 }
 
-static inline bool mptcp_syn_options(struct sock *sk, unsigned int *size,
+static inline bool mptcp_syn_options(struct sock *sk, const struct sk_buff *skb,
+				     unsigned int *size,
 				     struct mptcp_out_options *opts)
 {
 	return false;
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 28d31f2c1422..2f475b897c11 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -3926,7 +3926,7 @@ void tcp_parse_options(const struct net *net,
 				break;
 #endif
 			case TCPOPT_MPTCP:
-				mptcp_parse_option(ptr, opsize, opt_rx);
+				mptcp_parse_option(skb, ptr, opsize, opt_rx);
 				break;
 
 			case TCPOPT_FASTOPEN:
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 5456076166da..fec4b3a4b22d 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -685,7 +685,7 @@ static unsigned int tcp_syn_options(struct sock *sk, struct sk_buff *skb,
 	if (sk_is_mptcp(sk)) {
 		unsigned int size;
 
-		if (mptcp_syn_options(sk, &size, &opts->mptcp)) {
+		if (mptcp_syn_options(sk, skb, &size, &opts->mptcp)) {
 			opts->options |= OPTION_MPTCP;
 			remaining -= size;
 		}
diff --git a/net/mptcp/options.c b/net/mptcp/options.c
index 1aec742ca8e1..8f82ff9a5a8e 100644
--- a/net/mptcp/options.c
+++ b/net/mptcp/options.c
@@ -14,8 +14,8 @@ static bool mptcp_cap_flag_sha256(u8 flags)
 	return (flags & MPTCP_CAP_FLAG_MASK) == MPTCP_CAP_HMAC_SHA256;
 }
 
-void mptcp_parse_option(const unsigned char *ptr, int opsize,
-			struct tcp_options_received *opt_rx)
+void mptcp_parse_option(const struct sk_buff *skb, const unsigned char *ptr,
+			int opsize, struct tcp_options_received *opt_rx)
 {
 	struct mptcp_options_received *mp_opt = &opt_rx->mptcp;
 	u8 subtype = *ptr >> 4;
@@ -25,13 +25,29 @@ void mptcp_parse_option(const unsigned char *ptr, int opsize,
 
 	switch (subtype) {
 	case MPTCPOPT_MP_CAPABLE:
-		if (opsize != TCPOLEN_MPTCP_MPC_SYN &&
-		    opsize != TCPOLEN_MPTCP_MPC_ACK)
+		/* strict size checking */
+		if (!(TCP_SKB_CB(skb)->tcp_flags & TCPHDR_SYN)) {
+			if (skb->len > tcp_hdr(skb)->doff << 2)
+				expected_opsize = TCPOLEN_MPTCP_MPC_ACK_DATA;
+			else
+				expected_opsize = TCPOLEN_MPTCP_MPC_ACK;
+		} else {
+			if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_ACK)
+				expected_opsize = TCPOLEN_MPTCP_MPC_SYNACK;
+			else
+				expected_opsize = TCPOLEN_MPTCP_MPC_SYN;
+		}
+		if (opsize != expected_opsize)
 			break;
 
+		/* try to be gentle vs future versions on the initial syn */
 		version = *ptr++ & MPTCP_VERSION_MASK;
-		if (version != MPTCP_SUPPORTED_VERSION)
+		if (opsize != TCPOLEN_MPTCP_MPC_SYN) {
+			if (version != MPTCP_SUPPORTED_VERSION)
+				break;
+		} else if (version < MPTCP_SUPPORTED_VERSION) {
 			break;
+		}
 
 		flags = *ptr++;
 		if (!mptcp_cap_flag_sha256(flags) ||
@@ -55,23 +71,40 @@ void mptcp_parse_option(const unsigned char *ptr, int opsize,
 			break;
 
 		mp_opt->mp_capable = 1;
-		mp_opt->sndr_key = get_unaligned_be64(ptr);
-		ptr += 8;
-
-		if (opsize == TCPOLEN_MPTCP_MPC_ACK) {
+		if (opsize >= TCPOLEN_MPTCP_MPC_SYNACK) {
+			mp_opt->sndr_key = get_unaligned_be64(ptr);
+			ptr += 8;
+		}
+		if (opsize >= TCPOLEN_MPTCP_MPC_ACK) {
 			mp_opt->rcvr_key = get_unaligned_be64(ptr);
 			ptr += 8;
-			pr_debug("MP_CAPABLE sndr=%llu, rcvr=%llu",
-				 mp_opt->sndr_key, mp_opt->rcvr_key);
-		} else {
-			pr_debug("MP_CAPABLE sndr=%llu", mp_opt->sndr_key);
 		}
+		if (opsize == TCPOLEN_MPTCP_MPC_ACK_DATA) {
+			/* Section 3.1.:
+			 * "the data parameters in a MP_CAPABLE are semantically
+			 * equivalent to those in a DSS option and can be used
+			 * interchangeably."
+			 */
+			mp_opt->dss = 1;
+			mp_opt->use_map = 1;
+			mp_opt->mpc_map = 1;
+			mp_opt->data_len = get_unaligned_be16(ptr);
+			ptr += 2;
+		}
+		pr_debug("MP_CAPABLE version=%x, flags=%x, optlen=%d sndr=%llu, rcvr=%llu len=%d",
+			 version, flags, opsize, mp_opt->sndr_key,
+			 mp_opt->rcvr_key, mp_opt->data_len);
 		break;
 
 	case MPTCPOPT_DSS:
 		pr_debug("DSS");
 		ptr++;
 
+		/* we must clear 'mpc_map' be able to detect MP_CAPABLE
+		 * map vs DSS map in mptcp_incoming_options(), and reconstruct
+		 * map info accordingly
+		 */
+		mp_opt->mpc_map = 0;
 		flags = (*ptr++) & MPTCP_DSS_FLAG_MASK;
 		mp_opt->data_fin = (flags & MPTCP_DSS_DATA_FIN) != 0;
 		mp_opt->dsn64 = (flags & MPTCP_DSS_DSN64) != 0;
@@ -176,18 +209,22 @@ void mptcp_get_options(const struct sk_buff *skb,
 			if (opsize > length)
 				return;	/* don't parse partial options */
 			if (opcode == TCPOPT_MPTCP)
-				mptcp_parse_option(ptr, opsize, opt_rx);
+				mptcp_parse_option(skb, ptr, opsize, opt_rx);
 			ptr += opsize - 2;
 			length -= opsize;
 		}
 	}
 }
 
-bool mptcp_syn_options(struct sock *sk, unsigned int *size,
-		       struct mptcp_out_options *opts)
+bool mptcp_syn_options(struct sock *sk, const struct sk_buff *skb,
+		       unsigned int *size, struct mptcp_out_options *opts)
 {
 	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
 
+	/* we will use snd_isn to detect first pkt [re]transmission
+	 * in mptcp_established_options_mp()
+	 */
+	subflow->snd_isn = TCP_SKB_CB(skb)->end_seq;
 	if (subflow->request_mptcp) {
 		pr_debug("local_key=%llu", subflow->local_key);
 		opts->suboptions = OPTION_MPTCP_MPC_SYN;
@@ -212,20 +249,52 @@ void mptcp_rcv_synsent(struct sock *sk)
 	}
 }
 
-static bool mptcp_established_options_mp(struct sock *sk, unsigned int *size,
+static bool mptcp_established_options_mp(struct sock *sk, struct sk_buff *skb,
+					 unsigned int *size,
 					 unsigned int remaining,
 					 struct mptcp_out_options *opts)
 {
 	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
+	struct mptcp_ext *mpext;
+	unsigned int data_len;
+
+	pr_debug("subflow=%p fourth_ack=%d seq=%x:%x remaining=%d", subflow,
+		 subflow->fourth_ack, subflow->snd_isn,
+		 skb ? TCP_SKB_CB(skb)->seq : 0, remaining);
+
+	if (subflow->mp_capable && !subflow->fourth_ack && skb &&
+	    subflow->snd_isn == TCP_SKB_CB(skb)->seq) {
+		/* When skb is not available, we better over-estimate the
+		 * emitted options len. A full DSS option is longer than
+		 * TCPOLEN_MPTCP_MPC_ACK_DATA, so let's the caller try to fit
+		 * that.
+		 */
+		mpext = mptcp_get_ext(skb);
+		data_len = mpext ? mpext->data_len : 0;
 
-	if (!subflow->fourth_ack) {
+		/* we will check ext_copy.data_len in mptcp_write_options() to
+		 * discriminate between TCPOLEN_MPTCP_MPC_ACK_DATA and
+		 * TCPOLEN_MPTCP_MPC_ACK
+		 */
+		opts->ext_copy.data_len = data_len;
 		opts->suboptions = OPTION_MPTCP_MPC_ACK;
 		opts->sndr_key = subflow->local_key;
 		opts->rcvr_key = subflow->remote_key;
-		*size = TCPOLEN_MPTCP_MPC_ACK;
-		subflow->fourth_ack = 1;
-		pr_debug("subflow=%p, local_key=%llu, remote_key=%llu",
-			 subflow, subflow->local_key, subflow->remote_key);
+
+		/* Section 3.1.
+		 * The MP_CAPABLE option is carried on the SYN, SYN/ACK, and ACK
+		 * packets that start the first subflow of an MPTCP connection,
+		 * as well as the first packet that carries data
+		 */
+		if (data_len > 0)
+			*size = ALIGN(TCPOLEN_MPTCP_MPC_ACK_DATA, 4);
+		else
+			*size = TCPOLEN_MPTCP_MPC_ACK;
+
+		pr_debug("subflow=%p, local_key=%llu, remote_key=%llu map_len=%d",
+			 subflow, subflow->local_key, subflow->remote_key,
+			 data_len);
+
 		return true;
 	}
 	return false;
@@ -319,7 +388,7 @@ bool mptcp_established_options(struct sock *sk, struct sk_buff *skb,
 	unsigned int opt_size = 0;
 	bool ret = false;
 
-	if (mptcp_established_options_mp(sk, &opt_size, remaining, opts))
+	if (mptcp_established_options_mp(sk, skb, &opt_size, remaining, opts))
 		ret = true;
 	else if (mptcp_established_options_dss(sk, skb, &opt_size, remaining,
 					       opts))
@@ -371,11 +440,26 @@ void mptcp_incoming_options(struct sock *sk, struct sk_buff *skb,
 	memset(mpext, 0, sizeof(*mpext));
 
 	if (mp_opt->use_map) {
-		mpext->data_seq = mp_opt->data_seq;
-		mpext->subflow_seq = mp_opt->subflow_seq;
+		if (mp_opt->mpc_map) {
+			struct mptcp_subflow_context *subflow =
+				mptcp_subflow_ctx(sk);
+
+			/* this is an MP_CAPABLE carrying MPTCP data
+			 * we know this map the first chunk of data
+			 */
+			mptcp_crypto_key_sha(subflow->remote_key, NULL,
+					     &mpext->data_seq);
+			mpext->data_seq++;
+			mpext->subflow_seq = 1;
+			mpext->dsn64 = 1;
+			mpext->mpc_map = 1;
+		} else {
+			mpext->data_seq = mp_opt->data_seq;
+			mpext->subflow_seq = mp_opt->subflow_seq;
+			mpext->dsn64 = mp_opt->dsn64;
+		}
 		mpext->data_len = mp_opt->data_len;
 		mpext->use_map = 1;
-		mpext->dsn64 = mp_opt->dsn64;
 	}
 
 	if (mp_opt->use_ack) {
@@ -389,8 +473,7 @@ void mptcp_incoming_options(struct sock *sk, struct sk_buff *skb,
 
 void mptcp_write_options(__be32 *ptr, struct mptcp_out_options *opts)
 {
-	if ((OPTION_MPTCP_MPC_SYN |
-	     OPTION_MPTCP_MPC_SYNACK |
+	if ((OPTION_MPTCP_MPC_SYN | OPTION_MPTCP_MPC_SYNACK |
 	     OPTION_MPTCP_MPC_ACK) & opts->suboptions) {
 		u8 len;
 
@@ -398,6 +481,8 @@ void mptcp_write_options(__be32 *ptr, struct mptcp_out_options *opts)
 			len = TCPOLEN_MPTCP_MPC_SYN;
 		else if (OPTION_MPTCP_MPC_SYNACK & opts->suboptions)
 			len = TCPOLEN_MPTCP_MPC_SYNACK;
+		else if (opts->ext_copy.data_len)
+			len = TCPOLEN_MPTCP_MPC_ACK_DATA;
 		else
 			len = TCPOLEN_MPTCP_MPC_ACK;
 
@@ -405,14 +490,27 @@ void mptcp_write_options(__be32 *ptr, struct mptcp_out_options *opts)
 			       (MPTCPOPT_MP_CAPABLE << 12) |
 			       (MPTCP_SUPPORTED_VERSION << 8) |
 			       MPTCP_CAP_HMAC_SHA256);
+
+		if (!((OPTION_MPTCP_MPC_SYNACK | OPTION_MPTCP_MPC_ACK) &
+		    opts->suboptions))
+			goto mp_capable_done;
+
 		put_unaligned_be64(opts->sndr_key, ptr);
 		ptr += 2;
-		if (OPTION_MPTCP_MPC_ACK & opts->suboptions) {
-			put_unaligned_be64(opts->rcvr_key, ptr);
-			ptr += 2;
-		}
+		if (!((OPTION_MPTCP_MPC_ACK) & opts->suboptions))
+			goto mp_capable_done;
+
+		put_unaligned_be64(opts->rcvr_key, ptr);
+		ptr += 2;
+		if (!opts->ext_copy.data_len)
+			goto mp_capable_done;
+
+		put_unaligned_be32(opts->ext_copy.data_len << 16 |
+				   TCPOPT_NOP << 8 | TCPOPT_NOP, ptr);
+		ptr += 1;
 	}
 
+mp_capable_done:
 	if (opts->ext_copy.use_ack || opts->ext_copy.use_map) {
 		struct mptcp_ext *mpext = &opts->ext_copy;
 		u8 len = TCPOLEN_MPTCP_DSS_BASE;
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index a355bb1cf31b..36b90024d34d 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -11,7 +11,7 @@
 #include <net/tcp.h>
 #include <net/inet_connection_sock.h>
 
-#define MPTCP_SUPPORTED_VERSION	0
+#define MPTCP_SUPPORTED_VERSION	1
 
 /* MPTCP option bits */
 #define OPTION_MPTCP_MPC_SYN	BIT(0)
@@ -29,9 +29,10 @@
 #define MPTCPOPT_MP_FASTCLOSE	7
 
 /* MPTCP suboption lengths */
-#define TCPOLEN_MPTCP_MPC_SYN		12
+#define TCPOLEN_MPTCP_MPC_SYN		4
 #define TCPOLEN_MPTCP_MPC_SYNACK	12
 #define TCPOLEN_MPTCP_MPC_ACK		20
+#define TCPOLEN_MPTCP_MPC_ACK_DATA	22
 #define TCPOLEN_MPTCP_DSS_BASE		4
 #define TCPOLEN_MPTCP_DSS_ACK32		4
 #define TCPOLEN_MPTCP_DSS_ACK64		8
@@ -106,6 +107,7 @@ struct mptcp_subflow_context {
 	u64	remote_key;
 	u64	idsn;
 	u64	map_seq;
+	u32	snd_isn;
 	u32	token;
 	u32	rel_write_seq;
 	u32	map_subflow_seq;
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index 9fb3eb87a20f..8892855f4f52 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -77,7 +77,6 @@ static void subflow_init_req(struct request_sock *req,
 		if (err == 0)
 			subflow_req->mp_capable = 1;
 
-		subflow_req->remote_key = rx_opt.mptcp.sndr_key;
 		subflow_req->ssn_offset = TCP_SKB_CB(skb)->seq;
 	}
 }
@@ -180,11 +179,22 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,
 					  bool *own_req)
 {
 	struct mptcp_subflow_context *listener = mptcp_subflow_ctx(sk);
+	struct mptcp_subflow_request_sock *subflow_req;
+	struct tcp_options_received opt_rx;
 	struct sock *child;
 
 	pr_debug("listener=%p, req=%p, conn=%p", listener, req, listener->conn);
 
-	/* if the sk is MP_CAPABLE, we already received the client key */
+	/* if the sk is MP_CAPABLE, we need to fetch the client key */
+	subflow_req = mptcp_subflow_rsk(req);
+	if (subflow_req->mp_capable) {
+		opt_rx.mptcp.mp_capable = 0;
+		mptcp_get_options(skb, &opt_rx);
+		if (!opt_rx.mptcp.mp_capable)
+			subflow_req->mp_capable = 0;
+		else
+			subflow_req->remote_key = opt_rx.mptcp.sndr_key;
+	}
 
 	child = listener->icsk_af_ops->syn_recv_sock(sk, skb, req, dst,
 						     req_unhash, own_req);
-- 
cgit v1.2.3-59-g8ed1b


From d22f4988ffecbe284e4d00e897525adbd0edd801 Mon Sep 17 00:00:00 2001
From: Christoph Paasch <cpaasch@apple.com>
Date: Tue, 21 Jan 2020 16:56:32 -0800
Subject: mptcp: process MP_CAPABLE data option

This patch implements the handling of MP_CAPABLE + data option, as per
RFC 6824 bis / RFC 8684: MPTCP v1.

On the server side we can receive the remote key after that the connection
is established. We need to explicitly track the 'missing remote key'
status and avoid emitting a mptcp ack until we get such info.

When a late/retransmitted/OoO pkt carrying MP_CAPABLE[+data] option
is received, we have to propagate the mptcp seq number info to
the msk socket. To avoid ABBA locking issue, explicitly check for
that in recvmsg(), where we own msk and subflow sock locks.

The above also means that an established mp_capable subflow - still
waiting for the remote key - can be 'downgraded' to plain TCP.

Such change could potentially block a reader waiting for new data
forever - as they hook to msk, while later wake-up after the downgrade
will be on subflow only.

The above issue is not handled here, we likely have to get rid of
msk->fallback to handle that cleanly.

Signed-off-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/mptcp/options.c  | 56 +++++++++++++++++++++++++++++++++++++++++-----------
 net/mptcp/protocol.c | 16 ++++++++-------
 net/mptcp/protocol.h | 10 +++++++---
 net/mptcp/subflow.c  | 40 ++++++++++++++++++++++++++++++++-----
 4 files changed, 95 insertions(+), 27 deletions(-)

diff --git a/net/mptcp/options.c b/net/mptcp/options.c
index 8f82ff9a5a8e..45acd877bef3 100644
--- a/net/mptcp/options.c
+++ b/net/mptcp/options.c
@@ -243,6 +243,7 @@ void mptcp_rcv_synsent(struct sock *sk)
 	pr_debug("subflow=%p", subflow);
 	if (subflow->request_mptcp && tp->rx_opt.mptcp.mp_capable) {
 		subflow->mp_capable = 1;
+		subflow->can_ack = 1;
 		subflow->remote_key = tp->rx_opt.mptcp.sndr_key;
 	} else {
 		tcp_sk(sk)->is_mptcp = 0;
@@ -332,6 +333,7 @@ static bool mptcp_established_options_dss(struct sock *sk, struct sk_buff *skb,
 	struct mptcp_ext *mpext;
 	struct mptcp_sock *msk;
 	unsigned int ack_size;
+	bool ret = false;
 	u8 tcp_fin;
 
 	if (skb) {
@@ -355,6 +357,14 @@ static bool mptcp_established_options_dss(struct sock *sk, struct sk_buff *skb,
 		if (skb && tcp_fin &&
 		    subflow->conn->sk_state != TCP_ESTABLISHED)
 			mptcp_write_data_fin(subflow, &opts->ext_copy);
+		ret = true;
+	}
+
+	opts->ext_copy.use_ack = 0;
+	msk = mptcp_sk(subflow->conn);
+	if (!msk || !READ_ONCE(msk->can_ack)) {
+		*size = ALIGN(dss_size, 4);
+		return ret;
 	}
 
 	ack_size = TCPOLEN_MPTCP_DSS_ACK64;
@@ -365,15 +375,7 @@ static bool mptcp_established_options_dss(struct sock *sk, struct sk_buff *skb,
 
 	dss_size += ack_size;
 
-	msk = mptcp_sk(mptcp_subflow_ctx(sk)->conn);
-	if (msk) {
-		opts->ext_copy.data_ack = msk->ack_seq;
-	} else {
-		mptcp_crypto_key_sha(mptcp_subflow_ctx(sk)->remote_key,
-				     NULL, &opts->ext_copy.data_ack);
-		opts->ext_copy.data_ack++;
-	}
-
+	opts->ext_copy.data_ack = msk->ack_seq;
 	opts->ext_copy.ack64 = 1;
 	opts->ext_copy.use_ack = 1;
 
@@ -422,13 +424,46 @@ bool mptcp_synack_options(const struct request_sock *req, unsigned int *size,
 	return false;
 }
 
+static bool check_fourth_ack(struct mptcp_subflow_context *subflow,
+			     struct sk_buff *skb,
+			     struct mptcp_options_received *mp_opt)
+{
+	/* here we can process OoO, in-window pkts, only in-sequence 4th ack
+	 * are relevant
+	 */
+	if (likely(subflow->fourth_ack ||
+		   TCP_SKB_CB(skb)->seq != subflow->ssn_offset + 1))
+		return true;
+
+	if (mp_opt->use_ack)
+		subflow->fourth_ack = 1;
+
+	if (subflow->can_ack)
+		return true;
+
+	/* If the first established packet does not contain MP_CAPABLE + data
+	 * then fallback to TCP
+	 */
+	if (!mp_opt->mp_capable) {
+		subflow->mp_capable = 0;
+		tcp_sk(mptcp_subflow_tcp_sock(subflow))->is_mptcp = 0;
+		return false;
+	}
+	subflow->remote_key = mp_opt->sndr_key;
+	subflow->can_ack = 1;
+	return true;
+}
+
 void mptcp_incoming_options(struct sock *sk, struct sk_buff *skb,
 			    struct tcp_options_received *opt_rx)
 {
+	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
 	struct mptcp_options_received *mp_opt;
 	struct mptcp_ext *mpext;
 
 	mp_opt = &opt_rx->mptcp;
+	if (!check_fourth_ack(subflow, skb, mp_opt))
+		return;
 
 	if (!mp_opt->dss)
 		return;
@@ -441,9 +476,6 @@ void mptcp_incoming_options(struct sock *sk, struct sk_buff *skb,
 
 	if (mp_opt->use_map) {
 		if (mp_opt->mpc_map) {
-			struct mptcp_subflow_context *subflow =
-				mptcp_subflow_ctx(sk);
-
 			/* this is an MP_CAPABLE carrying MPTCP data
 			 * we know this map the first chunk of data
 			 */
diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 45e482864a19..1b64dfaa5f63 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -30,7 +30,7 @@
  */
 static struct socket *__mptcp_nmpc_socket(const struct mptcp_sock *msk)
 {
-	if (!msk->subflow || mptcp_subflow_ctx(msk->subflow->sk)->fourth_ack)
+	if (!msk->subflow || READ_ONCE(msk->can_ack))
 		return NULL;
 
 	return msk->subflow;
@@ -651,17 +651,20 @@ static struct sock *mptcp_accept(struct sock *sk, int flags, int *err,
 		__mptcp_init_sock(new_mptcp_sock);
 
 		msk = mptcp_sk(new_mptcp_sock);
-		msk->remote_key = subflow->remote_key;
 		msk->local_key = subflow->local_key;
 		msk->token = subflow->token;
 		msk->subflow = NULL;
 
 		mptcp_token_update_accept(newsk, new_mptcp_sock);
 
-		mptcp_crypto_key_sha(msk->remote_key, NULL, &ack_seq);
 		msk->write_seq = subflow->idsn + 1;
-		ack_seq++;
-		msk->ack_seq = ack_seq;
+		if (subflow->can_ack) {
+			msk->can_ack = true;
+			msk->remote_key = subflow->remote_key;
+			mptcp_crypto_key_sha(msk->remote_key, NULL, &ack_seq);
+			ack_seq++;
+			msk->ack_seq = ack_seq;
+		}
 		newsk = new_mptcp_sock;
 		mptcp_copy_inaddrs(newsk, ssk);
 		list_add(&subflow->node, &msk->conn_list);
@@ -678,8 +681,6 @@ static struct sock *mptcp_accept(struct sock *sk, int flags, int *err,
 		 * the receive path and process the pending ones
 		 */
 		lock_sock(ssk);
-		subflow->map_seq = ack_seq;
-		subflow->map_subflow_seq = 1;
 		subflow->rel_write_seq = 1;
 		subflow->tcp_sock = ssk;
 		subflow->conn = new_mptcp_sock;
@@ -795,6 +796,7 @@ void mptcp_finish_connect(struct sock *ssk)
 	WRITE_ONCE(msk->token, subflow->token);
 	WRITE_ONCE(msk->write_seq, subflow->idsn + 1);
 	WRITE_ONCE(msk->ack_seq, ack_seq);
+	WRITE_ONCE(msk->can_ack, 1);
 }
 
 static void mptcp_sock_graft(struct sock *sk, struct socket *parent)
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index 36b90024d34d..10eaa7c7381b 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -69,6 +69,7 @@ struct mptcp_sock {
 	u64		ack_seq;
 	u32		token;
 	unsigned long	flags;
+	bool		can_ack;
 	struct list_head conn_list;
 	struct skb_ext	*cached_ext;	/* for the next sendmsg */
 	struct socket	*subflow; /* outgoing connect/listener/!mp_capable */
@@ -84,9 +85,10 @@ static inline struct mptcp_sock *mptcp_sk(const struct sock *sk)
 
 struct mptcp_subflow_request_sock {
 	struct	tcp_request_sock sk;
-	u8	mp_capable : 1,
+	u16	mp_capable : 1,
 		mp_join : 1,
-		backup : 1;
+		backup : 1,
+		remote_key_valid : 1;
 	u64	local_key;
 	u64	remote_key;
 	u64	idsn;
@@ -118,8 +120,10 @@ struct mptcp_subflow_context {
 		fourth_ack : 1,	    /* send initial DSS */
 		conn_finished : 1,
 		map_valid : 1,
+		mpc_map : 1,
 		data_avail : 1,
-		rx_eof : 1;
+		rx_eof : 1,
+		can_ack : 1;	    /* only after processing the remote a key */
 
 	struct	sock *tcp_sock;	    /* tcp sk backpointer */
 	struct	sock *conn;	    /* parent mptcp_sock */
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index 8892855f4f52..8cfa1d29d59c 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -61,6 +61,7 @@ static void subflow_init_req(struct request_sock *req,
 	mptcp_get_options(skb, &rx_opt);
 
 	subflow_req->mp_capable = 0;
+	subflow_req->remote_key_valid = 0;
 
 #ifdef CONFIG_TCP_MD5SIG
 	/* no MPTCP if MD5SIG is enabled on this socket or we may run out of
@@ -185,17 +186,28 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,
 
 	pr_debug("listener=%p, req=%p, conn=%p", listener, req, listener->conn);
 
-	/* if the sk is MP_CAPABLE, we need to fetch the client key */
+	/* if the sk is MP_CAPABLE, we try to fetch the client key */
 	subflow_req = mptcp_subflow_rsk(req);
 	if (subflow_req->mp_capable) {
+		if (TCP_SKB_CB(skb)->seq != subflow_req->ssn_offset + 1) {
+			/* here we can receive and accept an in-window,
+			 * out-of-order pkt, which will not carry the MP_CAPABLE
+			 * opt even on mptcp enabled paths
+			 */
+			goto create_child;
+		}
+
 		opt_rx.mptcp.mp_capable = 0;
 		mptcp_get_options(skb, &opt_rx);
-		if (!opt_rx.mptcp.mp_capable)
-			subflow_req->mp_capable = 0;
-		else
+		if (opt_rx.mptcp.mp_capable) {
 			subflow_req->remote_key = opt_rx.mptcp.sndr_key;
+			subflow_req->remote_key_valid = 1;
+		} else {
+			subflow_req->mp_capable = 0;
+		}
 	}
 
+create_child:
 	child = listener->icsk_af_ops->syn_recv_sock(sk, skb, req, dst,
 						     req_unhash, own_req);
 
@@ -377,6 +389,7 @@ static enum mapping_status get_mapping_status(struct sock *ssk)
 	subflow->map_subflow_seq = mpext->subflow_seq;
 	subflow->map_data_len = data_len;
 	subflow->map_valid = 1;
+	subflow->mpc_map = mpext->mpc_map;
 	pr_debug("new map seq=%llu subflow_seq=%u data_len=%u",
 		 subflow->map_seq, subflow->map_subflow_seq,
 		 subflow->map_data_len);
@@ -428,6 +441,19 @@ static bool subflow_check_data_avail(struct sock *ssk)
 		if (WARN_ON_ONCE(!skb))
 			return false;
 
+		/* if msk lacks the remote key, this subflow must provide an
+		 * MP_CAPABLE-based mapping
+		 */
+		if (unlikely(!READ_ONCE(msk->can_ack))) {
+			if (!subflow->mpc_map) {
+				ssk->sk_err = EBADMSG;
+				goto fatal;
+			}
+			WRITE_ONCE(msk->remote_key, subflow->remote_key);
+			WRITE_ONCE(msk->ack_seq, subflow->map_seq);
+			WRITE_ONCE(msk->can_ack, true);
+		}
+
 		old_ack = READ_ONCE(msk->ack_seq);
 		ack_seq = mptcp_subflow_get_mapped_dsn(subflow);
 		pr_debug("msk ack_seq=%llx subflow ack_seq=%llx", old_ack,
@@ -752,13 +778,17 @@ static void subflow_ulp_clone(const struct request_sock *req,
 		return;
 	}
 
+	/* see comments in subflow_syn_recv_sock(), MPTCP connection is fully
+	 * established only after we receive the remote key
+	 */
 	new_ctx->conn_finished = 1;
 	new_ctx->icsk_af_ops = old_ctx->icsk_af_ops;
 	new_ctx->tcp_data_ready = old_ctx->tcp_data_ready;
 	new_ctx->tcp_state_change = old_ctx->tcp_state_change;
 	new_ctx->tcp_write_space = old_ctx->tcp_write_space;
 	new_ctx->mp_capable = 1;
-	new_ctx->fourth_ack = 1;
+	new_ctx->fourth_ack = subflow_req->remote_key_valid;
+	new_ctx->can_ack = subflow_req->remote_key_valid;
 	new_ctx->remote_key = subflow_req->remote_key;
 	new_ctx->local_key = subflow_req->local_key;
 	new_ctx->token = subflow_req->token;
-- 
cgit v1.2.3-59-g8ed1b


From 8ab183deb26a3b79f8021afa9e83cc1bbd812031 Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Tue, 21 Jan 2020 16:56:33 -0800
Subject: mptcp: cope with later TCP fallback

With MPTCP v1, passive connections can fallback to TCP after the
subflow becomes established:

syn + MP_CAPABLE ->
               <- syn, ack + MP_CAPABLE

ack, seq = 3    ->
        // OoO packet is accepted because in-sequence
        // passive socket is created, is in ESTABLISHED
	// status and tentatively as MP_CAPABLE

ack, seq = 2     ->
        // no MP_CAPABLE opt, subflow should fallback to TCP

We can't use the 'subflow' socket fallback, as we don't have
it available for passive connection.

Instead, when the fallback is detected, replace the mptcp
socket with the underlying TCP subflow. Beyond covering
the above scenario, it makes a TCP fallback socket as efficient
as plain TCP ones.

Co-developed-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/mptcp/protocol.c | 119 +++++++++++++++++++++++++++++++++++++++++++--------
 net/mptcp/protocol.h |   1 +
 2 files changed, 103 insertions(+), 17 deletions(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 1b64dfaa5f63..ee916b3686fe 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -24,6 +24,58 @@
 
 #define MPTCP_SAME_STATE TCP_MAX_STATES
 
+static void __mptcp_close(struct sock *sk, long timeout);
+
+static const struct proto_ops * tcp_proto_ops(struct sock *sk)
+{
+#if IS_ENABLED(CONFIG_IPV6)
+	if (sk->sk_family == AF_INET6)
+		return &inet6_stream_ops;
+#endif
+	return &inet_stream_ops;
+}
+
+/* MP_CAPABLE handshake failed, convert msk to plain tcp, replacing
+ * socket->sk and stream ops and destroying msk
+ * return the msk socket, as we can't access msk anymore after this function
+ * completes
+ * Called with msk lock held, releases such lock before returning
+ */
+static struct socket *__mptcp_fallback_to_tcp(struct mptcp_sock *msk,
+					      struct sock *ssk)
+{
+	struct mptcp_subflow_context *subflow;
+	struct socket *sock;
+	struct sock *sk;
+
+	sk = (struct sock *)msk;
+	sock = sk->sk_socket;
+	subflow = mptcp_subflow_ctx(ssk);
+
+	/* detach the msk socket */
+	list_del_init(&subflow->node);
+	sock_orphan(sk);
+	sock->sk = NULL;
+
+	/* socket is now TCP */
+	lock_sock(ssk);
+	sock_graft(ssk, sock);
+	if (subflow->conn) {
+		/* We can't release the ULP data on a live socket,
+		 * restore the tcp callback
+		 */
+		mptcp_subflow_tcp_fallback(ssk, subflow);
+		sock_put(subflow->conn);
+		subflow->conn = NULL;
+	}
+	release_sock(ssk);
+	sock->ops = tcp_proto_ops(ssk);
+
+	/* destroy the left-over msk sock */
+	__mptcp_close(sk, 0);
+	return sock;
+}
+
 /* If msk has an initial subflow socket, and the MP_CAPABLE handshake has not
  * completed yet or has failed, return the subflow socket.
  * Otherwise return NULL.
@@ -36,25 +88,37 @@ static struct socket *__mptcp_nmpc_socket(const struct mptcp_sock *msk)
 	return msk->subflow;
 }
 
-/* if msk has a single subflow, and the mp_capable handshake is failed,
- * return it.
+static bool __mptcp_needs_tcp_fallback(const struct mptcp_sock *msk)
+{
+	return msk->first && !sk_is_mptcp(msk->first);
+}
+
+/* if the mp_capable handshake has failed, it fallbacks msk to plain TCP,
+ * releases the socket lock and returns a reference to the now TCP socket.
  * Otherwise returns NULL
  */
-static struct socket *__mptcp_tcp_fallback(const struct mptcp_sock *msk)
+static struct socket *__mptcp_tcp_fallback(struct mptcp_sock *msk)
 {
-	struct socket *ssock = __mptcp_nmpc_socket(msk);
-
 	sock_owned_by_me((const struct sock *)msk);
 
-	if (!ssock || sk_is_mptcp(ssock->sk))
+	if (likely(!__mptcp_needs_tcp_fallback(msk)))
 		return NULL;
 
-	return ssock;
+	if (msk->subflow) {
+		/* the first subflow is an active connection, discart the
+		 * paired socket
+		 */
+		msk->subflow->sk = NULL;
+		sock_release(msk->subflow);
+		msk->subflow = NULL;
+	}
+
+	return __mptcp_fallback_to_tcp(msk, msk->first);
 }
 
 static bool __mptcp_can_create_subflow(const struct mptcp_sock *msk)
 {
-	return ((struct sock *)msk)->sk_state == TCP_CLOSE;
+	return !msk->first;
 }
 
 static struct socket *__mptcp_socket_create(struct mptcp_sock *msk, int state)
@@ -75,6 +139,7 @@ static struct socket *__mptcp_socket_create(struct mptcp_sock *msk, int state)
 	if (err)
 		return ERR_PTR(err);
 
+	msk->first = ssock->sk;
 	msk->subflow = ssock;
 	subflow = mptcp_subflow_ctx(ssock->sk);
 	list_add(&subflow->node, &msk->conn_list);
@@ -154,6 +219,8 @@ static int mptcp_sendmsg_frag(struct sock *sk, struct sock *ssk,
 		ret = sk_stream_wait_memory(ssk, timeo);
 		if (ret)
 			return ret;
+		if (unlikely(__mptcp_needs_tcp_fallback(msk)))
+			return 0;
 	}
 
 	/* compute copy limit */
@@ -265,11 +332,11 @@ static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 
 	lock_sock(sk);
 	ssock = __mptcp_tcp_fallback(msk);
-	if (ssock) {
+	if (unlikely(ssock)) {
+fallback:
 		pr_debug("fallback passthrough");
 		ret = sock_sendmsg(ssock, msg);
-		release_sock(sk);
-		return ret;
+		return ret >= 0 ? ret + copied : (copied ? copied : ret);
 	}
 
 	timeo = sock_sndtimeo(sk, msg->msg_flags & MSG_DONTWAIT);
@@ -288,6 +355,11 @@ static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 					 &size_goal);
 		if (ret < 0)
 			break;
+		if (ret == 0 && unlikely(__mptcp_needs_tcp_fallback(msk))) {
+			release_sock(ssk);
+			ssock = __mptcp_tcp_fallback(msk);
+			goto fallback;
+		}
 
 		copied += ret;
 	}
@@ -368,11 +440,11 @@ static int mptcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
 
 	lock_sock(sk);
 	ssock = __mptcp_tcp_fallback(msk);
-	if (ssock) {
+	if (unlikely(ssock)) {
+fallback:
 		pr_debug("fallback-read subflow=%p",
 			 mptcp_subflow_ctx(ssock->sk));
 		copied = sock_recvmsg(ssock, msg, flags);
-		release_sock(sk);
 		return copied;
 	}
 
@@ -477,6 +549,8 @@ wait_for_data:
 		pr_debug("block timeout %ld", timeo);
 		wait_data = true;
 		mptcp_wait_data(sk, &timeo);
+		if (unlikely(__mptcp_tcp_fallback(msk)))
+			goto fallback;
 	}
 
 	if (more_data_avail) {
@@ -529,6 +603,8 @@ static int __mptcp_init_sock(struct sock *sk)
 	INIT_LIST_HEAD(&msk->conn_list);
 	__set_bit(MPTCP_SEND_SPACE, &msk->flags);
 
+	msk->first = NULL;
+
 	return 0;
 }
 
@@ -563,7 +639,8 @@ static void mptcp_subflow_shutdown(struct sock *ssk, int how)
 	release_sock(ssk);
 }
 
-static void mptcp_close(struct sock *sk, long timeout)
+/* Called with msk lock held, releases such lock before returning */
+static void __mptcp_close(struct sock *sk, long timeout)
 {
 	struct mptcp_subflow_context *subflow, *tmp;
 	struct mptcp_sock *msk = mptcp_sk(sk);
@@ -571,8 +648,6 @@ static void mptcp_close(struct sock *sk, long timeout)
 	mptcp_token_destroy(msk->token);
 	inet_sk_state_store(sk, TCP_CLOSE);
 
-	lock_sock(sk);
-
 	list_for_each_entry_safe(subflow, tmp, &msk->conn_list, node) {
 		struct sock *ssk = mptcp_subflow_tcp_sock(subflow);
 
@@ -585,6 +660,12 @@ static void mptcp_close(struct sock *sk, long timeout)
 	sk_common_release(sk);
 }
 
+static void mptcp_close(struct sock *sk, long timeout)
+{
+	lock_sock(sk);
+	__mptcp_close(sk, timeout);
+}
+
 static void mptcp_copy_inaddrs(struct sock *msk, const struct sock *ssk)
 {
 #if IS_ENABLED(CONFIG_MPTCP_IPV6)
@@ -654,6 +735,7 @@ static struct sock *mptcp_accept(struct sock *sk, int flags, int *err,
 		msk->local_key = subflow->local_key;
 		msk->token = subflow->token;
 		msk->subflow = NULL;
+		msk->first = newsk;
 
 		mptcp_token_update_accept(newsk, new_mptcp_sock);
 
@@ -1007,8 +1089,8 @@ unlock_fail:
 static __poll_t mptcp_poll(struct file *file, struct socket *sock,
 			   struct poll_table_struct *wait)
 {
-	const struct mptcp_sock *msk;
 	struct sock *sk = sock->sk;
+	struct mptcp_sock *msk;
 	struct socket *ssock;
 	__poll_t mask = 0;
 
@@ -1024,6 +1106,9 @@ static __poll_t mptcp_poll(struct file *file, struct socket *sock,
 	release_sock(sk);
 	sock_poll_wait(file, sock, wait);
 	lock_sock(sk);
+	ssock = __mptcp_tcp_fallback(msk);
+	if (unlikely(ssock))
+		return ssock->ops->poll(file, ssock, NULL);
 
 	if (test_bit(MPTCP_DATA_READY, &msk->flags))
 		mask = EPOLLIN | EPOLLRDNORM;
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index 10eaa7c7381b..8a99a2930284 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -73,6 +73,7 @@ struct mptcp_sock {
 	struct list_head conn_list;
 	struct skb_ext	*cached_ext;	/* for the next sendmsg */
 	struct socket	*subflow; /* outgoing connect/listener/!mp_capable */
+	struct sock	*first;
 };
 
 #define mptcp_for_each_subflow(__msk, __subflow)			\
-- 
cgit v1.2.3-59-g8ed1b


From 268d3636dfb22254324774de1f8875174b3be064 Mon Sep 17 00:00:00 2001
From: Maxim Mikityanskiy <maxtram95@gmail.com>
Date: Fri, 24 Jan 2020 19:15:35 +0200
Subject: Bluetooth: btrtl: Use kvmalloc for FW allocations

Currently, kmemdup is applied to the firmware data, and it invokes
kmalloc under the hood. The firmware size and patch_length are big (more
than PAGE_SIZE), and on some low-end systems (like ASUS E202SA) kmalloc
may fail to allocate a contiguous chunk under high memory usage and
fragmentation:

Bluetooth: hci0: RTL: examining hci_ver=06 hci_rev=000a lmp_ver=06 lmp_subver=8821
Bluetooth: hci0: RTL: rom_version status=0 version=1
Bluetooth: hci0: RTL: loading rtl_bt/rtl8821a_fw.bin
kworker/u9:2: page allocation failure: order:4, mode:0x40cc0(GFP_KERNEL|__GFP_COMP), nodemask=(null),cpuset=/,mems_allowed=0
<stack trace follows>

As firmware load happens on each resume, Bluetooth will stop working
after several iterations, when the kernel fails to allocate an order-4
page.

This patch replaces kmemdup with kvmalloc+memcpy. It's not required to
have a contiguous chunk here, because it's not mapped to the device
directly.

Signed-off-by: Maxim Mikityanskiy <maxtram95@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 drivers/bluetooth/btrtl.c | 20 +++++++++++---------
 1 file changed, 11 insertions(+), 9 deletions(-)

diff --git a/drivers/bluetooth/btrtl.c b/drivers/bluetooth/btrtl.c
index f838537f9f89..577cfa3329db 100644
--- a/drivers/bluetooth/btrtl.c
+++ b/drivers/bluetooth/btrtl.c
@@ -370,11 +370,11 @@ static int rtlbt_parse_firmware(struct hci_dev *hdev,
 	 * the end.
 	 */
 	len = patch_length;
-	buf = kmemdup(btrtl_dev->fw_data + patch_offset, patch_length,
-		      GFP_KERNEL);
+	buf = kvmalloc(patch_length, GFP_KERNEL);
 	if (!buf)
 		return -ENOMEM;
 
+	memcpy(buf, btrtl_dev->fw_data + patch_offset, patch_length - 4);
 	memcpy(buf + patch_length - 4, &epatch_info->fw_version, 4);
 
 	*_buf = buf;
@@ -460,8 +460,10 @@ static int rtl_load_file(struct hci_dev *hdev, const char *name, u8 **buff)
 	if (ret < 0)
 		return ret;
 	ret = fw->size;
-	*buff = kmemdup(fw->data, ret, GFP_KERNEL);
-	if (!*buff)
+	*buff = kvmalloc(fw->size, GFP_KERNEL);
+	if (*buff)
+		memcpy(*buff, fw->data, ret);
+	else
 		ret = -ENOMEM;
 
 	release_firmware(fw);
@@ -499,14 +501,14 @@ static int btrtl_setup_rtl8723b(struct hci_dev *hdev,
 		goto out;
 
 	if (btrtl_dev->cfg_len > 0) {
-		tbuff = kzalloc(ret + btrtl_dev->cfg_len, GFP_KERNEL);
+		tbuff = kvzalloc(ret + btrtl_dev->cfg_len, GFP_KERNEL);
 		if (!tbuff) {
 			ret = -ENOMEM;
 			goto out;
 		}
 
 		memcpy(tbuff, fw_data, ret);
-		kfree(fw_data);
+		kvfree(fw_data);
 
 		memcpy(tbuff + ret, btrtl_dev->cfg_data, btrtl_dev->cfg_len);
 		ret += btrtl_dev->cfg_len;
@@ -519,14 +521,14 @@ static int btrtl_setup_rtl8723b(struct hci_dev *hdev,
 	ret = rtl_download_firmware(hdev, fw_data, ret);
 
 out:
-	kfree(fw_data);
+	kvfree(fw_data);
 	return ret;
 }
 
 void btrtl_free(struct btrtl_device_info *btrtl_dev)
 {
-	kfree(btrtl_dev->fw_data);
-	kfree(btrtl_dev->cfg_data);
+	kvfree(btrtl_dev->fw_data);
+	kvfree(btrtl_dev->cfg_data);
 	kfree(btrtl_dev);
 }
 EXPORT_SYMBOL_GPL(btrtl_free);
-- 
cgit v1.2.3-59-g8ed1b


From c31dbb1e41d1857b403f9bf58c87f5898519a0bc Mon Sep 17 00:00:00 2001
From: Lorenz Bauer <lmb@cloudflare.com>
Date: Fri, 24 Jan 2020 11:27:51 +0000
Subject: selftests: bpf: Use a temporary file in test_sockmap

Use a proper temporary file for sendpage tests. This means that running
the tests doesn't clutter the working directory, and allows running the
test on read-only filesystems.

Fixes: 16962b2404ac ("bpf: sockmap, add selftests")
Signed-off-by: Lorenz Bauer <lmb@cloudflare.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Jakub Sitnicki <jakub@cloudflare.com>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20200124112754.19664-2-lmb@cloudflare.com
---
 tools/testing/selftests/bpf/test_sockmap.c | 15 +++++----------
 1 file changed, 5 insertions(+), 10 deletions(-)

diff --git a/tools/testing/selftests/bpf/test_sockmap.c b/tools/testing/selftests/bpf/test_sockmap.c
index 4a851513c842..779e11da979c 100644
--- a/tools/testing/selftests/bpf/test_sockmap.c
+++ b/tools/testing/selftests/bpf/test_sockmap.c
@@ -331,7 +331,7 @@ static int msg_loop_sendpage(int fd, int iov_length, int cnt,
 	FILE *file;
 	int i, fp;
 
-	file = fopen(".sendpage_tst.tmp", "w+");
+	file = tmpfile();
 	if (!file) {
 		perror("create file for sendpage");
 		return 1;
@@ -340,13 +340,8 @@ static int msg_loop_sendpage(int fd, int iov_length, int cnt,
 		fwrite(&k, sizeof(char), 1, file);
 	fflush(file);
 	fseek(file, 0, SEEK_SET);
-	fclose(file);
 
-	fp = open(".sendpage_tst.tmp", O_RDONLY);
-	if (fp < 0) {
-		perror("reopen file for sendpage");
-		return 1;
-	}
+	fp = fileno(file);
 
 	clock_gettime(CLOCK_MONOTONIC, &s->start);
 	for (i = 0; i < cnt; i++) {
@@ -354,11 +349,11 @@ static int msg_loop_sendpage(int fd, int iov_length, int cnt,
 
 		if (!drop && sent < 0) {
 			perror("send loop error");
-			close(fp);
+			fclose(file);
 			return sent;
 		} else if (drop && sent >= 0) {
 			printf("sendpage loop error expected: %i\n", sent);
-			close(fp);
+			fclose(file);
 			return -EIO;
 		}
 
@@ -366,7 +361,7 @@ static int msg_loop_sendpage(int fd, int iov_length, int cnt,
 			s->bytes_sent += sent;
 	}
 	clock_gettime(CLOCK_MONOTONIC, &s->end);
-	close(fp);
+	fclose(file);
 	return 0;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 8bec4f665e0baecb5f1b683379fc10b3745eb612 Mon Sep 17 00:00:00 2001
From: Lorenz Bauer <lmb@cloudflare.com>
Date: Fri, 24 Jan 2020 11:27:52 +0000
Subject: selftests: bpf: Ignore FIN packets for reuseport tests

The reuseport tests currently suffer from a race condition: FIN
packets count towards DROP_ERR_SKB_DATA, since they don't contain
a valid struct cmd. Tests will spuriously fail depending on whether
check_results is called before or after the FIN is processed.

Exit the BPF program early if FIN is set.

Fixes: 91134d849a0e ("bpf: Test BPF_PROG_TYPE_SK_REUSEPORT")
Signed-off-by: Lorenz Bauer <lmb@cloudflare.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Jakub Sitnicki <jakub@cloudflare.com>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20200124112754.19664-3-lmb@cloudflare.com
---
 tools/testing/selftests/bpf/progs/test_select_reuseport_kern.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/tools/testing/selftests/bpf/progs/test_select_reuseport_kern.c b/tools/testing/selftests/bpf/progs/test_select_reuseport_kern.c
index d69a1f2bbbfd..26e77dcc7e91 100644
--- a/tools/testing/selftests/bpf/progs/test_select_reuseport_kern.c
+++ b/tools/testing/selftests/bpf/progs/test_select_reuseport_kern.c
@@ -113,6 +113,12 @@ int _select_by_skb_data(struct sk_reuseport_md *reuse_md)
 		data_check.skb_ports[0] = th->source;
 		data_check.skb_ports[1] = th->dest;
 
+		if (th->fin)
+			/* The connection is being torn down at the end of a
+			 * test. It can't contain a cmd, so return early.
+			 */
+			return SK_PASS;
+
 		if ((th->doff << 2) + sizeof(*cmd) > data_check.len)
 			GOTO_DONE(DROP_ERR_SKB_DATA);
 		if (bpf_skb_load_bytes(reuse_md, th->doff << 2, &cmd_copy,
-- 
cgit v1.2.3-59-g8ed1b


From 603fba9dfd0b2e2a844ece9ed98ce874c38aa98e Mon Sep 17 00:00:00 2001
From: Lorenz Bauer <lmb@cloudflare.com>
Date: Fri, 24 Jan 2020 11:27:53 +0000
Subject: selftests: bpf: Make reuseport test output more legible

Include the name of the mismatching result in human readable format
when reporting an error. The new output looks like the following:

  unexpected result
   result: [1, 0, 0, 0, 0, 0]
  expected: [0, 0, 0, 0, 0, 0]
  mismatch on DROP_ERR_INNER_MAP (bpf_prog_linum:153)
  check_results:FAIL:382

Signed-off-by: Lorenz Bauer <lmb@cloudflare.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Jakub Sitnicki <jakub@cloudflare.com>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20200124112754.19664-4-lmb@cloudflare.com
---
 .../selftests/bpf/prog_tests/select_reuseport.c    | 28 ++++++++++++++++++----
 1 file changed, 24 insertions(+), 4 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/select_reuseport.c b/tools/testing/selftests/bpf/prog_tests/select_reuseport.c
index 2c37ae7dc214..e7e56929751c 100644
--- a/tools/testing/selftests/bpf/prog_tests/select_reuseport.c
+++ b/tools/testing/selftests/bpf/prog_tests/select_reuseport.c
@@ -316,6 +316,26 @@ static void check_data(int type, sa_family_t family, const struct cmd *cmd,
 		       expected.len, result.len, get_linum());
 }
 
+static const char *result_to_str(enum result res)
+{
+	switch (res) {
+	case DROP_ERR_INNER_MAP:
+		return "DROP_ERR_INNER_MAP";
+	case DROP_ERR_SKB_DATA:
+		return "DROP_ERR_SKB_DATA";
+	case DROP_ERR_SK_SELECT_REUSEPORT:
+		return "DROP_ERR_SK_SELECT_REUSEPORT";
+	case DROP_MISC:
+		return "DROP_MISC";
+	case PASS:
+		return "PASS";
+	case PASS_ERR_SK_SELECT_REUSEPORT:
+		return "PASS_ERR_SK_SELECT_REUSEPORT";
+	default:
+		return "UNKNOWN";
+	}
+}
+
 static void check_results(void)
 {
 	__u32 results[NR_RESULTS];
@@ -351,10 +371,10 @@ static void check_results(void)
 		printf(", %u", expected_results[i]);
 	printf("]\n");
 
-	RET_IF(expected_results[broken] != results[broken],
-	       "unexpected result",
-	       "expected_results[%u] != results[%u] bpf_prog_linum:%ld\n",
-	       broken, broken, get_linum());
+	printf("mismatch on %s (bpf_prog_linum:%ld)\n", result_to_str(broken),
+	       get_linum());
+
+	CHECK_FAIL(true);
 }
 
 static int send_data(int type, sa_family_t family, void *data, size_t len,
-- 
cgit v1.2.3-59-g8ed1b


From 51bad0f05616c43d6d34b0a19bcc9bdab8e8fb39 Mon Sep 17 00:00:00 2001
From: Lorenz Bauer <lmb@cloudflare.com>
Date: Fri, 24 Jan 2020 11:27:54 +0000
Subject: selftests: bpf: Reset global state between reuseport test runs

Currently, there is a lot of false positives if a single reuseport test
fails. This is because expected_results and the result map are not cleared.

Zero both after individual test runs, which fixes the mentioned false
positives.

Fixes: 91134d849a0e ("bpf: Test BPF_PROG_TYPE_SK_REUSEPORT")
Signed-off-by: Lorenz Bauer <lmb@cloudflare.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Jakub Sitnicki <jakub@cloudflare.com>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20200124112754.19664-5-lmb@cloudflare.com
---
 .../testing/selftests/bpf/prog_tests/select_reuseport.c  | 16 ++++++++++++++--
 1 file changed, 14 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/select_reuseport.c b/tools/testing/selftests/bpf/prog_tests/select_reuseport.c
index e7e56929751c..098bcae5f827 100644
--- a/tools/testing/selftests/bpf/prog_tests/select_reuseport.c
+++ b/tools/testing/selftests/bpf/prog_tests/select_reuseport.c
@@ -33,7 +33,7 @@
 #define REUSEPORT_ARRAY_SIZE 32
 
 static int result_map, tmp_index_ovr_map, linum_map, data_check_map;
-static enum result expected_results[NR_RESULTS];
+static __u32 expected_results[NR_RESULTS];
 static int sk_fds[REUSEPORT_ARRAY_SIZE];
 static int reuseport_array = -1, outer_map = -1;
 static int select_by_skb_data_prog;
@@ -697,7 +697,19 @@ static void setup_per_test(int type, sa_family_t family, bool inany,
 
 static void cleanup_per_test(bool no_inner_map)
 {
-	int i, err;
+	int i, err, zero = 0;
+
+	memset(expected_results, 0, sizeof(expected_results));
+
+	for (i = 0; i < NR_RESULTS; i++) {
+		err = bpf_map_update_elem(result_map, &i, &zero, BPF_ANY);
+		RET_IF(err, "reset elem in result_map",
+		       "i:%u err:%d errno:%d\n", i, err, errno);
+	}
+
+	err = bpf_map_update_elem(linum_map, &zero, &zero, BPF_ANY);
+	RET_IF(err, "reset line number in linum_map", "err:%d errno:%d\n",
+	       err, errno);
 
 	for (i = 0; i < REUSEPORT_ARRAY_SIZE; i++)
 		close(sk_fds[i]);
-- 
cgit v1.2.3-59-g8ed1b


From d7a252708dbc950ca2064310217e8b9f85846e2f Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Thu, 23 Jan 2020 21:38:37 -0800
Subject: libbpf: Improve handling of failed CO-RE relocations

Previously, if libbpf failed to resolve CO-RE relocation for some
instructions, it would either return error immediately, or, if
.relaxed_core_relocs option was set, would replace relocatable offset/imm part
of an instruction with a bogus value (-1). Neither approach is good, because
there are many possible scenarios where relocation is expected to fail (e.g.,
when some field knowingly can be missing on specific kernel versions). On the
other hand, replacing offset with invalid one can hide programmer errors, if
this relocation failue wasn't anticipated.

This patch deprecates .relaxed_core_relocs option and changes the approach to
always replacing instruction, for which relocation failed, with invalid BPF
helper call instruction. For cases where this is expected, BPF program should
already ensure that that instruction is unreachable, in which case this
invalid instruction is going to be silently ignored. But if instruction wasn't
guarded, BPF program will be rejected at verification step with verifier log
pointing precisely to the place in assembly where the problem is.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20200124053837.2434679-1-andriin@fb.com
---
 tools/lib/bpf/libbpf.c | 95 +++++++++++++++++++++++++++++---------------------
 tools/lib/bpf/libbpf.h |  6 +++-
 2 files changed, 61 insertions(+), 40 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index ae34b681ae82..39f1b7633a7c 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -345,7 +345,6 @@ struct bpf_object {
 
 	bool loaded;
 	bool has_pseudo_calls;
-	bool relaxed_core_relocs;
 
 	/*
 	 * Information when doing elf related work. Only valid if fd
@@ -4238,25 +4237,38 @@ static int bpf_core_calc_field_relo(const struct bpf_program *prog,
  */
 static int bpf_core_reloc_insn(struct bpf_program *prog,
 			       const struct bpf_field_reloc *relo,
+			       int relo_idx,
 			       const struct bpf_core_spec *local_spec,
 			       const struct bpf_core_spec *targ_spec)
 {
-	bool failed = false, validate = true;
 	__u32 orig_val, new_val;
 	struct bpf_insn *insn;
+	bool validate = true;
 	int insn_idx, err;
 	__u8 class;
 
 	if (relo->insn_off % sizeof(struct bpf_insn))
 		return -EINVAL;
 	insn_idx = relo->insn_off / sizeof(struct bpf_insn);
+	insn = &prog->insns[insn_idx];
+	class = BPF_CLASS(insn->code);
 
 	if (relo->kind == BPF_FIELD_EXISTS) {
 		orig_val = 1; /* can't generate EXISTS relo w/o local field */
 		new_val = targ_spec ? 1 : 0;
 	} else if (!targ_spec) {
-		failed = true;
-		new_val = (__u32)-1;
+		pr_debug("prog '%s': relo #%d: substituting insn #%d w/ invalid insn\n",
+			 bpf_program__title(prog, false), relo_idx, insn_idx);
+		insn->code = BPF_JMP | BPF_CALL;
+		insn->dst_reg = 0;
+		insn->src_reg = 0;
+		insn->off = 0;
+		/* if this instruction is reachable (not a dead code),
+		 * verifier will complain with the following message:
+		 * invalid func unknown#195896080
+		 */
+		insn->imm = 195896080; /* => 0xbad2310 => "bad relo" */
+		return 0;
 	} else {
 		err = bpf_core_calc_field_relo(prog, relo, local_spec,
 					       &orig_val, &validate);
@@ -4268,50 +4280,47 @@ static int bpf_core_reloc_insn(struct bpf_program *prog,
 			return err;
 	}
 
-	insn = &prog->insns[insn_idx];
-	class = BPF_CLASS(insn->code);
-
 	switch (class) {
 	case BPF_ALU:
 	case BPF_ALU64:
 		if (BPF_SRC(insn->code) != BPF_K)
 			return -EINVAL;
-		if (!failed && validate && insn->imm != orig_val) {
-			pr_warn("prog '%s': unexpected insn #%d (ALU/ALU64) value: got %u, exp %u -> %u\n",
-				bpf_program__title(prog, false), insn_idx,
-				insn->imm, orig_val, new_val);
+		if (validate && insn->imm != orig_val) {
+			pr_warn("prog '%s': relo #%d: unexpected insn #%d (ALU/ALU64) value: got %u, exp %u -> %u\n",
+				bpf_program__title(prog, false), relo_idx,
+				insn_idx, insn->imm, orig_val, new_val);
 			return -EINVAL;
 		}
 		orig_val = insn->imm;
 		insn->imm = new_val;
-		pr_debug("prog '%s': patched insn #%d (ALU/ALU64)%s imm %u -> %u\n",
-			 bpf_program__title(prog, false), insn_idx,
-			 failed ? " w/ failed reloc" : "", orig_val, new_val);
+		pr_debug("prog '%s': relo #%d: patched insn #%d (ALU/ALU64) imm %u -> %u\n",
+			 bpf_program__title(prog, false), relo_idx, insn_idx,
+			 orig_val, new_val);
 		break;
 	case BPF_LDX:
 	case BPF_ST:
 	case BPF_STX:
-		if (!failed && validate && insn->off != orig_val) {
-			pr_warn("prog '%s': unexpected insn #%d (LD/LDX/ST/STX) value: got %u, exp %u -> %u\n",
-				bpf_program__title(prog, false), insn_idx,
-				insn->off, orig_val, new_val);
+		if (validate && insn->off != orig_val) {
+			pr_warn("prog '%s': relo #%d: unexpected insn #%d (LD/LDX/ST/STX) value: got %u, exp %u -> %u\n",
+				bpf_program__title(prog, false), relo_idx,
+				insn_idx, insn->off, orig_val, new_val);
 			return -EINVAL;
 		}
 		if (new_val > SHRT_MAX) {
-			pr_warn("prog '%s': insn #%d (LD/LDX/ST/STX) value too big: %u\n",
-				bpf_program__title(prog, false), insn_idx,
-				new_val);
+			pr_warn("prog '%s': relo #%d: insn #%d (LDX/ST/STX) value too big: %u\n",
+				bpf_program__title(prog, false), relo_idx,
+				insn_idx, new_val);
 			return -ERANGE;
 		}
 		orig_val = insn->off;
 		insn->off = new_val;
-		pr_debug("prog '%s': patched insn #%d (LD/LDX/ST/STX)%s off %u -> %u\n",
-			 bpf_program__title(prog, false), insn_idx,
-			 failed ? " w/ failed reloc" : "", orig_val, new_val);
+		pr_debug("prog '%s': relo #%d: patched insn #%d (LDX/ST/STX) off %u -> %u\n",
+			 bpf_program__title(prog, false), relo_idx, insn_idx,
+			 orig_val, new_val);
 		break;
 	default:
-		pr_warn("prog '%s': trying to relocate unrecognized insn #%d, code:%x, src:%x, dst:%x, off:%x, imm:%x\n",
-			bpf_program__title(prog, false),
+		pr_warn("prog '%s': relo #%d: trying to relocate unrecognized insn #%d, code:%x, src:%x, dst:%x, off:%x, imm:%x\n",
+			bpf_program__title(prog, false), relo_idx,
 			insn_idx, insn->code, insn->src_reg, insn->dst_reg,
 			insn->off, insn->imm);
 		return -EINVAL;
@@ -4510,24 +4519,33 @@ static int bpf_core_reloc_field(struct bpf_program *prog,
 	}
 
 	/*
-	 * For BPF_FIELD_EXISTS relo or when relaxed CO-RE reloc mode is
-	 * requested, it's expected that we might not find any candidates.
-	 * In this case, if field wasn't found in any candidate, the list of
-	 * candidates shouldn't change at all, we'll just handle relocating
-	 * appropriately, depending on relo's kind.
+	 * For BPF_FIELD_EXISTS relo or when used BPF program has field
+	 * existence checks or kernel version/config checks, it's expected
+	 * that we might not find any candidates. In this case, if field
+	 * wasn't found in any candidate, the list of candidates shouldn't
+	 * change at all, we'll just handle relocating appropriately,
+	 * depending on relo's kind.
 	 */
 	if (j > 0)
 		cand_ids->len = j;
 
-	if (j == 0 && !prog->obj->relaxed_core_relocs &&
-	    relo->kind != BPF_FIELD_EXISTS) {
-		pr_warn("prog '%s': relo #%d: no matching targets found for [%d] %s + %s\n",
-			prog_name, relo_idx, local_id, local_name, spec_str);
-		return -ESRCH;
-	}
+	/*
+	 * If no candidates were found, it might be both a programmer error,
+	 * as well as expected case, depending whether instruction w/
+	 * relocation is guarded in some way that makes it unreachable (dead
+	 * code) if relocation can't be resolved. This is handled in
+	 * bpf_core_reloc_insn() uniformly by replacing that instruction with
+	 * BPF helper call insn (using invalid helper ID). If that instruction
+	 * is indeed unreachable, then it will be ignored and eliminated by
+	 * verifier. If it was an error, then verifier will complain and point
+	 * to a specific instruction number in its log.
+	 */
+	if (j == 0)
+		pr_debug("prog '%s': relo #%d: no matching targets found for [%d] %s + %s\n",
+			 prog_name, relo_idx, local_id, local_name, spec_str);
 
 	/* bpf_core_reloc_insn should know how to handle missing targ_spec */
-	err = bpf_core_reloc_insn(prog, relo, &local_spec,
+	err = bpf_core_reloc_insn(prog, relo, relo_idx, &local_spec,
 				  j ? &targ_spec : NULL);
 	if (err) {
 		pr_warn("prog '%s': relo #%d: failed to patch insn at offset %d: %d\n",
@@ -5057,7 +5075,6 @@ __bpf_object__open(const char *path, const void *obj_buf, size_t obj_buf_sz,
 	if (IS_ERR(obj))
 		return obj;
 
-	obj->relaxed_core_relocs = OPTS_GET(opts, relaxed_core_relocs, false);
 	kconfig = OPTS_GET(opts, kconfig, NULL);
 	if (kconfig) {
 		obj->kconfig = strdup(kconfig);
diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index 2a5e3b087002..3fe12c9d1f92 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -77,7 +77,11 @@ struct bpf_object_open_opts {
 	const char *object_name;
 	/* parse map definitions non-strictly, allowing extra attributes/data */
 	bool relaxed_maps;
-	/* process CO-RE relocations non-strictly, allowing them to fail */
+	/* DEPRECATED: handle CO-RE relocations non-strictly, allowing failures.
+	 * Value is ignored. Relocations always are processed non-strictly.
+	 * Non-relocatable instructions are replaced with invalid ones to
+	 * prevent accidental errors.
+	 * */
 	bool relaxed_core_relocs;
 	/* maps that set the 'pinning' attribute in their definition will have
 	 * their pin_path attribute set to a file in this directory, and be
-- 
cgit v1.2.3-59-g8ed1b


From c90ed40cefe187a20fc565650b119aa696abc2ed Mon Sep 17 00:00:00 2001
From: Tony Nguyen <anthony.l.nguyen@intel.com>
Date: Fri, 17 Jan 2020 07:39:12 -0800
Subject: ice: Enable writing hardware filtering tables

Enable the driver to write the filtering hardware tables to allow for
changing of RSS rules. Upon loading of DDP package, a minimal configuration
should be written to hardware.

Introduce and initialize structures for storing configuration and make
the top level calls to configure the RSS tables to initial values. A packet
segment will be created but nothing is written to hardware yet.

Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Signed-off-by: Henry Tieman <henry.w.tieman@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ice/Makefile        |   3 +-
 drivers/net/ethernet/intel/ice/ice_common.c    |   6 +-
 drivers/net/ethernet/intel/ice/ice_flex_pipe.c |  31 ++-
 drivers/net/ethernet/intel/ice/ice_flow.c      | 250 +++++++++++++++++++++++++
 drivers/net/ethernet/intel/ice/ice_flow.h      | 114 +++++++++++
 drivers/net/ethernet/intel/ice/ice_lib.c       |  87 ++++++++-
 drivers/net/ethernet/intel/ice/ice_type.h      |   4 +
 7 files changed, 491 insertions(+), 4 deletions(-)
 create mode 100644 drivers/net/ethernet/intel/ice/ice_flow.c
 create mode 100644 drivers/net/ethernet/intel/ice/ice_flow.h

diff --git a/drivers/net/ethernet/intel/ice/Makefile b/drivers/net/ethernet/intel/ice/Makefile
index 7cb829132d28..59544b0fc086 100644
--- a/drivers/net/ethernet/intel/ice/Makefile
+++ b/drivers/net/ethernet/intel/ice/Makefile
@@ -17,7 +17,8 @@ ice-y := ice_main.o	\
 	 ice_lib.o	\
 	 ice_txrx_lib.o	\
 	 ice_txrx.o	\
-	 ice_flex_pipe.o	\
+	 ice_flex_pipe.o \
+	 ice_flow.o	\
 	 ice_ethtool.o
 ice-$(CONFIG_PCI_IOV) += ice_virtchnl_pf.o ice_sriov.o
 ice-$(CONFIG_DCB) += ice_dcb.o ice_dcb_nl.o ice_dcb_lib.o
diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index a03b4fdc01e6..dd9af9c63755 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -4,6 +4,7 @@
 #include "ice_common.h"
 #include "ice_sched.h"
 #include "ice_adminq_cmd.h"
+#include "ice_flow.h"
 
 #define ICE_PF_RESET_WAIT_COUNT	200
 
@@ -3406,7 +3407,10 @@ enum ice_status ice_replay_vsi(struct ice_hw *hw, u16 vsi_handle)
 		if (status)
 			return status;
 	}
-
+	/* Replay per VSI all RSS configurations */
+	status = ice_replay_rss_cfg(hw, vsi_handle);
+	if (status)
+		return status;
 	/* Replay per VSI all filters */
 	status = ice_replay_vsi_all_fltr(hw, vsi_handle);
 	return status;
diff --git a/drivers/net/ethernet/intel/ice/ice_flex_pipe.c b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c
index cbd53b586c36..f37eb536e7b7 100644
--- a/drivers/net/ethernet/intel/ice/ice_flex_pipe.c
+++ b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c
@@ -3,6 +3,7 @@
 
 #include "ice_common.h"
 #include "ice_flex_pipe.h"
+#include "ice_flow.h"
 
 /**
  * ice_pkg_val_buf
@@ -1379,11 +1380,18 @@ void ice_fill_blk_tbls(struct ice_hw *hw)
  */
 void ice_free_hw_tbls(struct ice_hw *hw)
 {
+	struct ice_rss_cfg *r, *rt;
 	u8 i;
 
 	for (i = 0; i < ICE_BLK_COUNT; i++) {
-		hw->blk[i].is_list_init = false;
+		if (hw->blk[i].is_list_init) {
+			struct ice_es *es = &hw->blk[i].es;
 
+			mutex_destroy(&es->prof_map_lock);
+			mutex_destroy(&hw->fl_profs_locks[i]);
+
+			hw->blk[i].is_list_init = false;
+		}
 		devm_kfree(ice_hw_to_dev(hw), hw->blk[i].xlt1.ptypes);
 		devm_kfree(ice_hw_to_dev(hw), hw->blk[i].xlt1.ptg_tbl);
 		devm_kfree(ice_hw_to_dev(hw), hw->blk[i].xlt1.t);
@@ -1397,9 +1405,25 @@ void ice_free_hw_tbls(struct ice_hw *hw)
 		devm_kfree(ice_hw_to_dev(hw), hw->blk[i].es.written);
 	}
 
+	list_for_each_entry_safe(r, rt, &hw->rss_list_head, l_entry) {
+		list_del(&r->l_entry);
+		devm_kfree(ice_hw_to_dev(hw), r);
+	}
+	mutex_destroy(&hw->rss_locks);
 	memset(hw->blk, 0, sizeof(hw->blk));
 }
 
+/**
+ * ice_init_flow_profs - init flow profile locks and list heads
+ * @hw: pointer to the hardware structure
+ * @blk_idx: HW block index
+ */
+static void ice_init_flow_profs(struct ice_hw *hw, u8 blk_idx)
+{
+	mutex_init(&hw->fl_profs_locks[blk_idx]);
+	INIT_LIST_HEAD(&hw->fl_profs[blk_idx]);
+}
+
 /**
  * ice_clear_hw_tbls - clear HW tables and flow profiles
  * @hw: pointer to the hardware structure
@@ -1443,6 +1467,8 @@ enum ice_status ice_init_hw_tbls(struct ice_hw *hw)
 {
 	u8 i;
 
+	mutex_init(&hw->rss_locks);
+	INIT_LIST_HEAD(&hw->rss_list_head);
 	for (i = 0; i < ICE_BLK_COUNT; i++) {
 		struct ice_prof_redir *prof_redir = &hw->blk[i].prof_redir;
 		struct ice_prof_tcam *prof = &hw->blk[i].prof;
@@ -1454,6 +1480,9 @@ enum ice_status ice_init_hw_tbls(struct ice_hw *hw)
 		if (hw->blk[i].is_list_init)
 			continue;
 
+		ice_init_flow_profs(hw, i);
+		mutex_init(&es->prof_map_lock);
+		INIT_LIST_HEAD(&es->prof_map);
 		hw->blk[i].is_list_init = true;
 
 		hw->blk[i].overwrite = blk_sizes[i].overwrite;
diff --git a/drivers/net/ethernet/intel/ice/ice_flow.c b/drivers/net/ethernet/intel/ice/ice_flow.c
new file mode 100644
index 000000000000..4828c64abe98
--- /dev/null
+++ b/drivers/net/ethernet/intel/ice/ice_flow.c
@@ -0,0 +1,250 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2019, Intel Corporation. */
+
+#include "ice_common.h"
+#include "ice_flow.h"
+
+/* Describe properties of a protocol header field */
+struct ice_flow_field_info {
+	enum ice_flow_seg_hdr hdr;
+	s16 off;	/* Offset from start of a protocol header, in bits */
+	u16 size;	/* Size of fields in bits */
+};
+
+#define ICE_FLOW_FLD_INFO(_hdr, _offset_bytes, _size_bytes) { \
+	.hdr = _hdr, \
+	.off = (_offset_bytes) * BITS_PER_BYTE, \
+	.size = (_size_bytes) * BITS_PER_BYTE, \
+}
+
+/* Table containing properties of supported protocol header fields */
+static const
+struct ice_flow_field_info ice_flds_info[ICE_FLOW_FIELD_IDX_MAX] = {
+	/* IPv4 / IPv6 */
+	/* ICE_FLOW_FIELD_IDX_IPV4_SA */
+	ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_IPV4, 12, sizeof(struct in_addr)),
+	/* ICE_FLOW_FIELD_IDX_IPV4_DA */
+	ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_IPV4, 16, sizeof(struct in_addr)),
+	/* ICE_FLOW_FIELD_IDX_IPV6_SA */
+	ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_IPV6, 8, sizeof(struct in6_addr)),
+	/* ICE_FLOW_FIELD_IDX_IPV6_DA */
+	ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_IPV6, 24, sizeof(struct in6_addr)),
+	/* Transport */
+	/* ICE_FLOW_FIELD_IDX_TCP_SRC_PORT */
+	ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_TCP, 0, sizeof(__be16)),
+	/* ICE_FLOW_FIELD_IDX_TCP_DST_PORT */
+	ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_TCP, 2, sizeof(__be16)),
+	/* ICE_FLOW_FIELD_IDX_UDP_SRC_PORT */
+	ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_UDP, 0, sizeof(__be16)),
+	/* ICE_FLOW_FIELD_IDX_UDP_DST_PORT */
+	ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_UDP, 2, sizeof(__be16)),
+};
+
+/**
+ * ice_flow_set_fld_ext - specifies locations of field from entry's input buffer
+ * @seg: packet segment the field being set belongs to
+ * @fld: field to be set
+ * @type: type of the field
+ * @val_loc: if not ICE_FLOW_FLD_OFF_INVAL, location of the value to match from
+ *           entry's input buffer
+ * @mask_loc: if not ICE_FLOW_FLD_OFF_INVAL, location of mask value from entry's
+ *            input buffer
+ * @last_loc: if not ICE_FLOW_FLD_OFF_INVAL, location of last/upper value from
+ *            entry's input buffer
+ *
+ * This helper function stores information of a field being matched, including
+ * the type of the field and the locations of the value to match, the mask, and
+ * and the upper-bound value in the start of the input buffer for a flow entry.
+ * This function should only be used for fixed-size data structures.
+ *
+ * This function also opportunistically determines the protocol headers to be
+ * present based on the fields being set. Some fields cannot be used alone to
+ * determine the protocol headers present. Sometimes, fields for particular
+ * protocol headers are not matched. In those cases, the protocol headers
+ * must be explicitly set.
+ */
+static void
+ice_flow_set_fld_ext(struct ice_flow_seg_info *seg, enum ice_flow_field fld,
+		     enum ice_flow_fld_match_type type, u16 val_loc,
+		     u16 mask_loc, u16 last_loc)
+{
+	u64 bit = BIT_ULL(fld);
+
+	seg->match |= bit;
+	if (type == ICE_FLOW_FLD_TYPE_RANGE)
+		seg->range |= bit;
+
+	seg->fields[fld].type = type;
+	seg->fields[fld].src.val = val_loc;
+	seg->fields[fld].src.mask = mask_loc;
+	seg->fields[fld].src.last = last_loc;
+
+	ICE_FLOW_SET_HDRS(seg, ice_flds_info[fld].hdr);
+}
+
+/**
+ * ice_flow_set_fld - specifies locations of field from entry's input buffer
+ * @seg: packet segment the field being set belongs to
+ * @fld: field to be set
+ * @val_loc: if not ICE_FLOW_FLD_OFF_INVAL, location of the value to match from
+ *           entry's input buffer
+ * @mask_loc: if not ICE_FLOW_FLD_OFF_INVAL, location of mask value from entry's
+ *            input buffer
+ * @last_loc: if not ICE_FLOW_FLD_OFF_INVAL, location of last/upper value from
+ *            entry's input buffer
+ * @range: indicate if field being matched is to be in a range
+ *
+ * This function specifies the locations, in the form of byte offsets from the
+ * start of the input buffer for a flow entry, from where the value to match,
+ * the mask value, and upper value can be extracted. These locations are then
+ * stored in the flow profile. When adding a flow entry associated with the
+ * flow profile, these locations will be used to quickly extract the values and
+ * create the content of a match entry. This function should only be used for
+ * fixed-size data structures.
+ */
+static void
+ice_flow_set_fld(struct ice_flow_seg_info *seg, enum ice_flow_field fld,
+		 u16 val_loc, u16 mask_loc, u16 last_loc, bool range)
+{
+	enum ice_flow_fld_match_type t = range ?
+		ICE_FLOW_FLD_TYPE_RANGE : ICE_FLOW_FLD_TYPE_REG;
+
+	ice_flow_set_fld_ext(seg, fld, t, val_loc, mask_loc, last_loc);
+}
+
+#define ICE_FLOW_RSS_SEG_HDR_L3_MASKS \
+	(ICE_FLOW_SEG_HDR_IPV4 | ICE_FLOW_SEG_HDR_IPV6)
+
+#define ICE_FLOW_RSS_SEG_HDR_L4_MASKS \
+	(ICE_FLOW_SEG_HDR_TCP | ICE_FLOW_SEG_HDR_UDP | ICE_FLOW_SEG_HDR_SCTP)
+
+#define ICE_FLOW_RSS_SEG_HDR_VAL_MASKS \
+	(ICE_FLOW_RSS_SEG_HDR_L3_MASKS | \
+	 ICE_FLOW_RSS_SEG_HDR_L4_MASKS)
+
+/**
+ * ice_flow_set_rss_seg_info - setup packet segments for RSS
+ * @segs: pointer to the flow field segment(s)
+ * @hash_fields: fields to be hashed on for the segment(s)
+ * @flow_hdr: protocol header fields within a packet segment
+ *
+ * Helper function to extract fields from hash bitmap and use flow
+ * header value to set flow field segment for further use in flow
+ * profile entry or removal.
+ */
+static enum ice_status
+ice_flow_set_rss_seg_info(struct ice_flow_seg_info *segs, u64 hash_fields,
+			  u32 flow_hdr)
+{
+	u64 val;
+	u8 i;
+
+	for_each_set_bit(i, (unsigned long *)&hash_fields,
+			 ICE_FLOW_FIELD_IDX_MAX)
+		ice_flow_set_fld(segs, (enum ice_flow_field)i,
+				 ICE_FLOW_FLD_OFF_INVAL, ICE_FLOW_FLD_OFF_INVAL,
+				 ICE_FLOW_FLD_OFF_INVAL, false);
+
+	ICE_FLOW_SET_HDRS(segs, flow_hdr);
+
+	if (segs->hdrs & ~ICE_FLOW_RSS_SEG_HDR_VAL_MASKS)
+		return ICE_ERR_PARAM;
+
+	val = (u64)(segs->hdrs & ICE_FLOW_RSS_SEG_HDR_L3_MASKS);
+	if (val && !is_power_of_2(val))
+		return ICE_ERR_CFG;
+
+	val = (u64)(segs->hdrs & ICE_FLOW_RSS_SEG_HDR_L4_MASKS);
+	if (val && !is_power_of_2(val))
+		return ICE_ERR_CFG;
+
+	return 0;
+}
+
+#define ICE_RSS_OUTER_HEADERS	1
+
+/**
+ * ice_add_rss_cfg_sync - add an RSS configuration
+ * @hashed_flds: hash bit fields (ICE_FLOW_HASH_*) to configure
+ * @addl_hdrs: protocol header fields
+ * @segs_cnt: packet segment count
+ *
+ * Assumption: lock has already been acquired for RSS list
+ */
+static enum ice_status
+ice_add_rss_cfg_sync(u64 hashed_flds, u32 addl_hdrs, u8 segs_cnt)
+{
+	struct ice_flow_seg_info *segs;
+	enum ice_status status;
+
+	if (!segs_cnt || segs_cnt > ICE_FLOW_SEG_MAX)
+		return ICE_ERR_PARAM;
+
+	segs = kcalloc(segs_cnt, sizeof(*segs), GFP_KERNEL);
+	if (!segs)
+		return ICE_ERR_NO_MEMORY;
+
+	/* Construct the packet segment info from the hashed fields */
+	status = ice_flow_set_rss_seg_info(&segs[segs_cnt - 1], hashed_flds,
+					   addl_hdrs);
+
+	kfree(segs);
+	return status;
+}
+
+/**
+ * ice_add_rss_cfg - add an RSS configuration with specified hashed fields
+ * @hw: pointer to the hardware structure
+ * @vsi_handle: software VSI handle
+ * @hashed_flds: hash bit fields (ICE_FLOW_HASH_*) to configure
+ * @addl_hdrs: protocol header fields
+ *
+ * This function will generate a flow profile based on fields associated with
+ * the input fields to hash on, the flow type and use the VSI number to add
+ * a flow entry to the profile.
+ */
+enum ice_status
+ice_add_rss_cfg(struct ice_hw *hw, u16 vsi_handle, u64 hashed_flds,
+		u32 addl_hdrs)
+{
+	enum ice_status status;
+
+	if (hashed_flds == ICE_HASH_INVALID ||
+	    !ice_is_vsi_valid(hw, vsi_handle))
+		return ICE_ERR_PARAM;
+
+	mutex_lock(&hw->rss_locks);
+	status = ice_add_rss_cfg_sync(hashed_flds, addl_hdrs,
+				      ICE_RSS_OUTER_HEADERS);
+	mutex_unlock(&hw->rss_locks);
+
+	return status;
+}
+
+/**
+ * ice_replay_rss_cfg - replay RSS configurations associated with VSI
+ * @hw: pointer to the hardware structure
+ * @vsi_handle: software VSI handle
+ */
+enum ice_status ice_replay_rss_cfg(struct ice_hw *hw, u16 vsi_handle)
+{
+	enum ice_status status = 0;
+	struct ice_rss_cfg *r;
+
+	if (!ice_is_vsi_valid(hw, vsi_handle))
+		return ICE_ERR_PARAM;
+
+	mutex_lock(&hw->rss_locks);
+	list_for_each_entry(r, &hw->rss_list_head, l_entry) {
+		if (test_bit(vsi_handle, r->vsis)) {
+			status = ice_add_rss_cfg_sync(r->hashed_flds,
+						      r->packet_hdr,
+						      ICE_RSS_OUTER_HEADERS);
+			if (status)
+				break;
+		}
+	}
+	mutex_unlock(&hw->rss_locks);
+
+	return status;
+}
diff --git a/drivers/net/ethernet/intel/ice/ice_flow.h b/drivers/net/ethernet/intel/ice/ice_flow.h
new file mode 100644
index 000000000000..48c0fc09d5ff
--- /dev/null
+++ b/drivers/net/ethernet/intel/ice/ice_flow.h
@@ -0,0 +1,114 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) 2019, Intel Corporation. */
+
+#ifndef _ICE_FLOW_H_
+#define _ICE_FLOW_H_
+
+#define ICE_FLOW_FLD_OFF_INVAL		0xffff
+
+/* Generate flow hash field from flow field type(s) */
+#define ICE_FLOW_HASH_IPV4	\
+	(BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_SA) | \
+	 BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_DA))
+#define ICE_FLOW_HASH_IPV6	\
+	(BIT_ULL(ICE_FLOW_FIELD_IDX_IPV6_SA) | \
+	 BIT_ULL(ICE_FLOW_FIELD_IDX_IPV6_DA))
+#define ICE_FLOW_HASH_TCP_PORT	\
+	(BIT_ULL(ICE_FLOW_FIELD_IDX_TCP_SRC_PORT) | \
+	 BIT_ULL(ICE_FLOW_FIELD_IDX_TCP_DST_PORT))
+#define ICE_FLOW_HASH_UDP_PORT	\
+	(BIT_ULL(ICE_FLOW_FIELD_IDX_UDP_SRC_PORT) | \
+	 BIT_ULL(ICE_FLOW_FIELD_IDX_UDP_DST_PORT))
+
+#define ICE_HASH_INVALID	0
+#define ICE_HASH_TCP_IPV4	(ICE_FLOW_HASH_IPV4 | ICE_FLOW_HASH_TCP_PORT)
+#define ICE_HASH_TCP_IPV6	(ICE_FLOW_HASH_IPV6 | ICE_FLOW_HASH_TCP_PORT)
+#define ICE_HASH_UDP_IPV4	(ICE_FLOW_HASH_IPV4 | ICE_FLOW_HASH_UDP_PORT)
+#define ICE_HASH_UDP_IPV6	(ICE_FLOW_HASH_IPV6 | ICE_FLOW_HASH_UDP_PORT)
+
+/* Protocol header fields within a packet segment. A segment consists of one or
+ * more protocol headers that make up a logical group of protocol headers. Each
+ * logical group of protocol headers encapsulates or is encapsulated using/by
+ * tunneling or encapsulation protocols for network virtualization such as GRE,
+ * VxLAN, etc.
+ */
+enum ice_flow_seg_hdr {
+	ICE_FLOW_SEG_HDR_NONE		= 0x00000000,
+	ICE_FLOW_SEG_HDR_IPV4		= 0x00000004,
+	ICE_FLOW_SEG_HDR_IPV6		= 0x00000008,
+	ICE_FLOW_SEG_HDR_TCP		= 0x00000040,
+	ICE_FLOW_SEG_HDR_UDP		= 0x00000080,
+	ICE_FLOW_SEG_HDR_SCTP		= 0x00000100,
+};
+
+enum ice_flow_field {
+	/* L3 */
+	ICE_FLOW_FIELD_IDX_IPV4_SA,
+	ICE_FLOW_FIELD_IDX_IPV4_DA,
+	ICE_FLOW_FIELD_IDX_IPV6_SA,
+	ICE_FLOW_FIELD_IDX_IPV6_DA,
+	/* L4 */
+	ICE_FLOW_FIELD_IDX_TCP_SRC_PORT,
+	ICE_FLOW_FIELD_IDX_TCP_DST_PORT,
+	ICE_FLOW_FIELD_IDX_UDP_SRC_PORT,
+	ICE_FLOW_FIELD_IDX_UDP_DST_PORT,
+	/* The total number of enums must not exceed 64 */
+	ICE_FLOW_FIELD_IDX_MAX
+};
+
+#define ICE_FLOW_SEG_MAX		2
+#define ICE_FLOW_SET_HDRS(seg, val)	((seg)->hdrs |= (u32)(val))
+
+struct ice_flow_seg_xtrct {
+	u8 prot_id;	/* Protocol ID of extracted header field */
+	u16 off;	/* Starting offset of the field in header in bytes */
+	u8 idx;		/* Index of FV entry used */
+	u8 disp;	/* Displacement of field in bits fr. FV entry's start */
+};
+
+enum ice_flow_fld_match_type {
+	ICE_FLOW_FLD_TYPE_REG,		/* Value, mask */
+	ICE_FLOW_FLD_TYPE_RANGE,	/* Value, mask, last (upper bound) */
+	ICE_FLOW_FLD_TYPE_PREFIX,	/* IP address, prefix, size of prefix */
+	ICE_FLOW_FLD_TYPE_SIZE,		/* Value, mask, size of match */
+};
+
+struct ice_flow_fld_loc {
+	/* Describe offsets of field information relative to the beginning of
+	 * input buffer provided when adding flow entries.
+	 */
+	u16 val;	/* Offset where the value is located */
+	u16 mask;	/* Offset where the mask/prefix value is located */
+	u16 last;	/* Length or offset where the upper value is located */
+};
+
+struct ice_flow_fld_info {
+	enum ice_flow_fld_match_type type;
+	/* Location where to retrieve data from an input buffer */
+	struct ice_flow_fld_loc src;
+	/* Location where to put the data into the final entry buffer */
+	struct ice_flow_fld_loc entry;
+	struct ice_flow_seg_xtrct xtrct;
+};
+
+struct ice_flow_seg_info {
+	u32 hdrs;	/* Bitmask indicating protocol headers present */
+	u64 match;	/* Bitmask indicating header fields to be matched */
+	u64 range;	/* Bitmask indicating header fields matched as ranges */
+
+	struct ice_flow_fld_info fields[ICE_FLOW_FIELD_IDX_MAX];
+};
+
+struct ice_rss_cfg {
+	struct list_head l_entry;
+	/* bitmap of VSIs added to the RSS entry */
+	DECLARE_BITMAP(vsis, ICE_MAX_VSI);
+	u64 hashed_flds;
+	u32 packet_hdr;
+};
+
+enum ice_status ice_replay_rss_cfg(struct ice_hw *hw, u16 vsi_handle);
+enum ice_status
+ice_add_rss_cfg(struct ice_hw *hw, u16 vsi_handle, u64 hashed_flds,
+		u32 addl_hdrs);
+#endif /* _ICE_FLOW_H_ */
diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c
index 4cfad81ba496..dd230c6664ee 100644
--- a/drivers/net/ethernet/intel/ice/ice_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_lib.c
@@ -3,6 +3,7 @@
 
 #include "ice.h"
 #include "ice_base.h"
+#include "ice_flow.h"
 #include "ice_lib.h"
 #include "ice_dcb_lib.h"
 
@@ -1086,6 +1087,88 @@ ice_vsi_cfg_rss_exit:
 	return err;
 }
 
+/**
+ * ice_vsi_set_rss_flow_fld - Sets RSS input set for different flows
+ * @vsi: VSI to be configured
+ *
+ * This function will only be called after successful download package call
+ * during initialization of PF. Since the downloaded package will erase the
+ * RSS section, this function will configure RSS input sets for different
+ * flow types. The last profile added has the highest priority, therefore 2
+ * tuple profiles (i.e. IPv4 src/dst) are added before 4 tuple profiles
+ * (i.e. IPv4 src/dst TCP src/dst port).
+ */
+static void ice_vsi_set_rss_flow_fld(struct ice_vsi *vsi)
+{
+	u16 vsi_handle = vsi->idx, vsi_num = vsi->vsi_num;
+	struct ice_pf *pf = vsi->back;
+	struct ice_hw *hw = &pf->hw;
+	enum ice_status status;
+	struct device *dev;
+
+	dev = ice_pf_to_dev(pf);
+	if (ice_is_safe_mode(pf)) {
+		dev_dbg(dev, "Advanced RSS disabled. Package download failed, vsi num = %d\n",
+			vsi_num);
+		return;
+	}
+	/* configure RSS for IPv4 with input set IP src/dst */
+	status = ice_add_rss_cfg(hw, vsi_handle, ICE_FLOW_HASH_IPV4,
+				 ICE_FLOW_SEG_HDR_IPV4);
+	if (status)
+		dev_dbg(dev, "ice_add_rss_cfg failed for ipv4 flow, vsi = %d, error = %d\n",
+			vsi_num, status);
+
+	/* configure RSS for IPv6 with input set IPv6 src/dst */
+	status = ice_add_rss_cfg(hw, vsi_handle, ICE_FLOW_HASH_IPV6,
+				 ICE_FLOW_SEG_HDR_IPV6);
+	if (status)
+		dev_dbg(dev, "ice_add_rss_cfg failed for ipv6 flow, vsi = %d, error = %d\n",
+			vsi_num, status);
+
+	/* configure RSS for tcp4 with input set IP src/dst, TCP src/dst */
+	status = ice_add_rss_cfg(hw, vsi_handle, ICE_HASH_TCP_IPV4,
+				 ICE_FLOW_SEG_HDR_TCP | ICE_FLOW_SEG_HDR_IPV4);
+	if (status)
+		dev_dbg(dev, "ice_add_rss_cfg failed for tcp4 flow, vsi = %d, error = %d\n",
+			vsi_num, status);
+
+	/* configure RSS for udp4 with input set IP src/dst, UDP src/dst */
+	status = ice_add_rss_cfg(hw, vsi_handle, ICE_HASH_UDP_IPV4,
+				 ICE_FLOW_SEG_HDR_UDP | ICE_FLOW_SEG_HDR_IPV4);
+	if (status)
+		dev_dbg(dev, "ice_add_rss_cfg failed for udp4 flow, vsi = %d, error = %d\n",
+			vsi_num, status);
+
+	/* configure RSS for sctp4 with input set IP src/dst */
+	status = ice_add_rss_cfg(hw, vsi_handle, ICE_FLOW_HASH_IPV4,
+				 ICE_FLOW_SEG_HDR_SCTP | ICE_FLOW_SEG_HDR_IPV4);
+	if (status)
+		dev_dbg(dev, "ice_add_rss_cfg failed for sctp4 flow, vsi = %d, error = %d\n",
+			vsi_num, status);
+
+	/* configure RSS for tcp6 with input set IPv6 src/dst, TCP src/dst */
+	status = ice_add_rss_cfg(hw, vsi_handle, ICE_HASH_TCP_IPV6,
+				 ICE_FLOW_SEG_HDR_TCP | ICE_FLOW_SEG_HDR_IPV6);
+	if (status)
+		dev_dbg(dev, "ice_add_rss_cfg failed for tcp6 flow, vsi = %d, error = %d\n",
+			vsi_num, status);
+
+	/* configure RSS for udp6 with input set IPv6 src/dst, UDP src/dst */
+	status = ice_add_rss_cfg(hw, vsi_handle, ICE_HASH_UDP_IPV6,
+				 ICE_FLOW_SEG_HDR_UDP | ICE_FLOW_SEG_HDR_IPV6);
+	if (status)
+		dev_dbg(dev, "ice_add_rss_cfg failed for udp6 flow, vsi = %d, error = %d\n",
+			vsi_num, status);
+
+	/* configure RSS for sctp6 with input set IPv6 src/dst */
+	status = ice_add_rss_cfg(hw, vsi_handle, ICE_FLOW_HASH_IPV6,
+				 ICE_FLOW_SEG_HDR_SCTP | ICE_FLOW_SEG_HDR_IPV6);
+	if (status)
+		dev_dbg(dev, "ice_add_rss_cfg failed for sctp6 flow, vsi = %d, error = %d\n",
+			vsi_num, status);
+}
+
 /**
  * ice_add_mac_to_list - Add a MAC address filter entry to the list
  * @vsi: the VSI to be forwarded to
@@ -1901,8 +1984,10 @@ ice_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi,
 		 * receive traffic on first queue. Hence no need to capture
 		 * return value
 		 */
-		if (test_bit(ICE_FLAG_RSS_ENA, pf->flags))
+		if (test_bit(ICE_FLAG_RSS_ENA, pf->flags)) {
 			ice_vsi_cfg_rss_lut_key(vsi);
+			ice_vsi_set_rss_flow_fld(vsi);
+		}
 		break;
 	case ICE_VSI_VF:
 		/* VF driver will take care of creating netdev for this type and
diff --git a/drivers/net/ethernet/intel/ice/ice_type.h b/drivers/net/ethernet/intel/ice/ice_type.h
index c4854a987130..ac38af105f90 100644
--- a/drivers/net/ethernet/intel/ice/ice_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_type.h
@@ -559,6 +559,10 @@ struct ice_hw {
 
 	/* HW block tables */
 	struct ice_blk_info blk[ICE_BLK_COUNT];
+	struct mutex fl_profs_locks[ICE_BLK_COUNT];	/* lock fltr profiles */
+	struct list_head fl_profs[ICE_BLK_COUNT];
+	struct mutex rss_locks;	/* protect RSS configuration */
+	struct list_head rss_list_head;
 };
 
 /* Statistics collected by each port, VSI, VEB, and S-channel */
-- 
cgit v1.2.3-59-g8ed1b


From 35b9211c0a2427e8f39e534f442f43804fc8d5ca Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andriin@fb.com>
Date: Fri, 24 Jan 2020 12:18:46 -0800
Subject: libbpf: Fix realloc usage in bpf_core_find_cands

Fix bug requesting invalid size of reallocated array when constructing CO-RE
relocation candidate list. This can cause problems if there are many potential
candidates and a very fine-grained memory allocator bucket sizes are used.

Fixes: ddc7c3042614 ("libbpf: implement BPF CO-RE offset relocation algorithm")
Reported-by: William Smith <williampsmith@fb.com>
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20200124201847.212528-1-andriin@fb.com
---
 tools/lib/bpf/libbpf.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 39f1b7633a7c..514b1a524abb 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -3869,7 +3869,9 @@ static struct ids_vec *bpf_core_find_cands(const struct btf *local_btf,
 		if (strncmp(local_name, targ_name, local_essent_len) == 0) {
 			pr_debug("[%d] %s: found candidate [%d] %s\n",
 				 local_type_id, local_name, i, targ_name);
-			new_ids = realloc(cand_ids->data, cand_ids->len + 1);
+			new_ids = reallocarray(cand_ids->data,
+					       cand_ids->len + 1,
+					       sizeof(*cand_ids->data));
 			if (!new_ids) {
 				err = -ENOMEM;
 				goto err_out;
-- 
cgit v1.2.3-59-g8ed1b


From 31ad4e4ee1e4e2e12985e46aa6263c8bd5ad4271 Mon Sep 17 00:00:00 2001
From: Tony Nguyen <anthony.l.nguyen@intel.com>
Date: Fri, 17 Jan 2020 07:39:13 -0800
Subject: ice: Allocate flow profile

Create an extraction sequence based on the packet header protocols to be
programmed and allocate a flow profile for the extraction sequence.

Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Signed-off-by: Henry Tieman <henry.w.tieman@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_adminq_cmd.h    |   6 +
 drivers/net/ethernet/intel/ice/ice_common.c        |  75 ++++
 drivers/net/ethernet/intel/ice/ice_common.h        |   6 +
 drivers/net/ethernet/intel/ice/ice_flex_pipe.c     | 216 +++++++++
 drivers/net/ethernet/intel/ice/ice_flex_pipe.h     |   3 +
 drivers/net/ethernet/intel/ice/ice_flex_type.h     |  15 +
 drivers/net/ethernet/intel/ice/ice_flow.c          | 490 ++++++++++++++++++++-
 drivers/net/ethernet/intel/ice/ice_flow.h          |  23 +
 drivers/net/ethernet/intel/ice/ice_lan_tx_rx.h     |   8 +
 drivers/net/ethernet/intel/ice/ice_protocol_type.h |  24 +
 drivers/net/ethernet/intel/ice/ice_switch.c        |  36 --
 drivers/net/ethernet/intel/ice/ice_type.h          |   2 +
 12 files changed, 864 insertions(+), 40 deletions(-)
 create mode 100644 drivers/net/ethernet/intel/ice/ice_protocol_type.h

diff --git a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
index 5421fc413f94..ad15a772b609 100644
--- a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
@@ -232,6 +232,12 @@ struct ice_aqc_get_sw_cfg_resp {
  */
 #define ICE_AQC_RES_TYPE_VSI_LIST_REP			0x03
 #define ICE_AQC_RES_TYPE_VSI_LIST_PRUNE			0x04
+#define ICE_AQC_RES_TYPE_HASH_PROF_BLDR_PROFID		0x60
+
+#define ICE_AQC_RES_TYPE_FLAG_SCAN_BOTTOM		BIT(12)
+#define ICE_AQC_RES_TYPE_FLAG_IGNORE_INDEX		BIT(13)
+
+#define ICE_AQC_RES_TYPE_FLAG_DEDICATED			0x00
 
 /* Allocate Resources command (indirect 0x0208)
  * Free Resources command (indirect 0x0209)
diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index dd9af9c63755..8f86962fd052 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -1497,6 +1497,81 @@ void ice_release_res(struct ice_hw *hw, enum ice_aq_res_ids res)
 	}
 }
 
+/**
+ * ice_aq_alloc_free_res - command to allocate/free resources
+ * @hw: pointer to the HW struct
+ * @num_entries: number of resource entries in buffer
+ * @buf: Indirect buffer to hold data parameters and response
+ * @buf_size: size of buffer for indirect commands
+ * @opc: pass in the command opcode
+ * @cd: pointer to command details structure or NULL
+ *
+ * Helper function to allocate/free resources using the admin queue commands
+ */
+enum ice_status
+ice_aq_alloc_free_res(struct ice_hw *hw, u16 num_entries,
+		      struct ice_aqc_alloc_free_res_elem *buf, u16 buf_size,
+		      enum ice_adminq_opc opc, struct ice_sq_cd *cd)
+{
+	struct ice_aqc_alloc_free_res_cmd *cmd;
+	struct ice_aq_desc desc;
+
+	cmd = &desc.params.sw_res_ctrl;
+
+	if (!buf)
+		return ICE_ERR_PARAM;
+
+	if (buf_size < (num_entries * sizeof(buf->elem[0])))
+		return ICE_ERR_PARAM;
+
+	ice_fill_dflt_direct_cmd_desc(&desc, opc);
+
+	desc.flags |= cpu_to_le16(ICE_AQ_FLAG_RD);
+
+	cmd->num_entries = cpu_to_le16(num_entries);
+
+	return ice_aq_send_cmd(hw, &desc, buf, buf_size, cd);
+}
+
+/**
+ * ice_alloc_hw_res - allocate resource
+ * @hw: pointer to the HW struct
+ * @type: type of resource
+ * @num: number of resources to allocate
+ * @btm: allocate from bottom
+ * @res: pointer to array that will receive the resources
+ */
+enum ice_status
+ice_alloc_hw_res(struct ice_hw *hw, u16 type, u16 num, bool btm, u16 *res)
+{
+	struct ice_aqc_alloc_free_res_elem *buf;
+	enum ice_status status;
+	u16 buf_len;
+
+	buf_len = struct_size(buf, elem, num - 1);
+	buf = kzalloc(buf_len, GFP_KERNEL);
+	if (!buf)
+		return ICE_ERR_NO_MEMORY;
+
+	/* Prepare buffer to allocate resource. */
+	buf->num_elems = cpu_to_le16(num);
+	buf->res_type = cpu_to_le16(type | ICE_AQC_RES_TYPE_FLAG_DEDICATED |
+				    ICE_AQC_RES_TYPE_FLAG_IGNORE_INDEX);
+	if (btm)
+		buf->res_type |= cpu_to_le16(ICE_AQC_RES_TYPE_FLAG_SCAN_BOTTOM);
+
+	status = ice_aq_alloc_free_res(hw, 1, buf, buf_len,
+				       ice_aqc_opc_alloc_res, NULL);
+	if (status)
+		goto ice_alloc_res_exit;
+
+	memcpy(res, buf->elem, sizeof(buf->elem) * num);
+
+ice_alloc_res_exit:
+	kfree(buf);
+	return status;
+}
+
 /**
  * ice_get_num_per_func - determine number of resources per PF
  * @hw: pointer to the HW structure
diff --git a/drivers/net/ethernet/intel/ice/ice_common.h b/drivers/net/ethernet/intel/ice/ice_common.h
index b22aa561e253..81e8ac925b9c 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.h
+++ b/drivers/net/ethernet/intel/ice/ice_common.h
@@ -34,10 +34,16 @@ enum ice_status
 ice_acquire_res(struct ice_hw *hw, enum ice_aq_res_ids res,
 		enum ice_aq_res_access_type access, u32 timeout);
 void ice_release_res(struct ice_hw *hw, enum ice_aq_res_ids res);
+enum ice_status
+ice_alloc_hw_res(struct ice_hw *hw, u16 type, u16 num, bool btm, u16 *res);
 enum ice_status ice_init_nvm(struct ice_hw *hw);
 enum ice_status
 ice_read_sr_buf(struct ice_hw *hw, u16 offset, u16 *words, u16 *data);
 enum ice_status
+ice_aq_alloc_free_res(struct ice_hw *hw, u16 num_entries,
+		      struct ice_aqc_alloc_free_res_elem *buf, u16 buf_size,
+		      enum ice_adminq_opc opc, struct ice_sq_cd *cd);
+enum ice_status
 ice_sq_send_cmd(struct ice_hw *hw, struct ice_ctl_q_info *cq,
 		struct ice_aq_desc *desc, void *buf, u16 buf_size,
 		struct ice_sq_cd *cd);
diff --git a/drivers/net/ethernet/intel/ice/ice_flex_pipe.c b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c
index f37eb536e7b7..5c09d3105b8b 100644
--- a/drivers/net/ethernet/intel/ice/ice_flex_pipe.c
+++ b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c
@@ -1118,6 +1118,117 @@ ice_vsig_add_mv_vsi(struct ice_hw *hw, enum ice_block blk, u16 vsi, u16 vsig)
 	return 0;
 }
 
+/**
+ * ice_find_prof_id - find profile ID for a given field vector
+ * @hw: pointer to the hardware structure
+ * @blk: HW block
+ * @fv: field vector to search for
+ * @prof_id: receives the profile ID
+ */
+static enum ice_status
+ice_find_prof_id(struct ice_hw *hw, enum ice_block blk,
+		 struct ice_fv_word *fv, u8 *prof_id)
+{
+	struct ice_es *es = &hw->blk[blk].es;
+	u16 off, i;
+
+	for (i = 0; i < es->count; i++) {
+		off = i * es->fvw;
+
+		if (memcmp(&es->t[off], fv, es->fvw * sizeof(*fv)))
+			continue;
+
+		*prof_id = i;
+		return 0;
+	}
+
+	return ICE_ERR_DOES_NOT_EXIST;
+}
+
+/**
+ * ice_prof_id_rsrc_type - get profile ID resource type for a block type
+ * @blk: the block type
+ * @rsrc_type: pointer to variable to receive the resource type
+ */
+static bool ice_prof_id_rsrc_type(enum ice_block blk, u16 *rsrc_type)
+{
+	switch (blk) {
+	case ICE_BLK_RSS:
+		*rsrc_type = ICE_AQC_RES_TYPE_HASH_PROF_BLDR_PROFID;
+		break;
+	default:
+		return false;
+	}
+	return true;
+}
+
+/**
+ * ice_alloc_prof_id - allocate profile ID
+ * @hw: pointer to the HW struct
+ * @blk: the block to allocate the profile ID for
+ * @prof_id: pointer to variable to receive the profile ID
+ *
+ * This function allocates a new profile ID, which also corresponds to a Field
+ * Vector (Extraction Sequence) entry.
+ */
+static enum ice_status
+ice_alloc_prof_id(struct ice_hw *hw, enum ice_block blk, u8 *prof_id)
+{
+	enum ice_status status;
+	u16 res_type;
+	u16 get_prof;
+
+	if (!ice_prof_id_rsrc_type(blk, &res_type))
+		return ICE_ERR_PARAM;
+
+	status = ice_alloc_hw_res(hw, res_type, 1, false, &get_prof);
+	if (!status)
+		*prof_id = (u8)get_prof;
+
+	return status;
+}
+
+/**
+ * ice_prof_inc_ref - increment reference count for profile
+ * @hw: pointer to the HW struct
+ * @blk: the block from which to free the profile ID
+ * @prof_id: the profile ID for which to increment the reference count
+ */
+static enum ice_status
+ice_prof_inc_ref(struct ice_hw *hw, enum ice_block blk, u8 prof_id)
+{
+	if (prof_id > hw->blk[blk].es.count)
+		return ICE_ERR_PARAM;
+
+	hw->blk[blk].es.ref_count[prof_id]++;
+
+	return 0;
+}
+
+/**
+ * ice_write_es - write an extraction sequence to hardware
+ * @hw: pointer to the HW struct
+ * @blk: the block in which to write the extraction sequence
+ * @prof_id: the profile ID to write
+ * @fv: pointer to the extraction sequence to write - NULL to clear extraction
+ */
+static void
+ice_write_es(struct ice_hw *hw, enum ice_block blk, u8 prof_id,
+	     struct ice_fv_word *fv)
+{
+	u16 off;
+
+	off = prof_id * hw->blk[blk].es.fvw;
+	if (!fv) {
+		memset(&hw->blk[blk].es.t[off], 0,
+		       hw->blk[blk].es.fvw * sizeof(*fv));
+		hw->blk[blk].es.written[prof_id] = false;
+	} else {
+		memcpy(&hw->blk[blk].es.t[off], fv,
+		       hw->blk[blk].es.fvw * sizeof(*fv));
+	}
+}
+
 /* Block / table section IDs */
 static const u32 ice_blk_sids[ICE_BLK_COUNT][ICE_SID_OFF_COUNT] = {
 	/* SWITCH */
@@ -1576,3 +1687,108 @@ err:
 	ice_free_hw_tbls(hw);
 	return ICE_ERR_NO_MEMORY;
 }
+
+/**
+ * ice_add_prof - add profile
+ * @hw: pointer to the HW struct
+ * @blk: hardware block
+ * @id: profile tracking ID
+ * @ptypes: array of bitmaps indicating ptypes (ICE_FLOW_PTYPE_MAX bits)
+ * @es: extraction sequence (length of array is determined by the block)
+ *
+ * This function registers a profile, which matches a set of PTGs with a
+ * particular extraction sequence. While the hardware profile is allocated
+ * it will not be written until the first call to ice_add_flow that specifies
+ * the ID value used here.
+ */
+enum ice_status
+ice_add_prof(struct ice_hw *hw, enum ice_block blk, u64 id, u8 ptypes[],
+	     struct ice_fv_word *es)
+{
+	u32 bytes = DIV_ROUND_UP(ICE_FLOW_PTYPE_MAX, BITS_PER_BYTE);
+	DECLARE_BITMAP(ptgs_used, ICE_XLT1_CNT);
+	struct ice_prof_map *prof;
+	enum ice_status status;
+	u32 byte = 0;
+	u8 prof_id;
+
+	bitmap_zero(ptgs_used, ICE_XLT1_CNT);
+
+	mutex_lock(&hw->blk[blk].es.prof_map_lock);
+
+	/* search for existing profile */
+	status = ice_find_prof_id(hw, blk, es, &prof_id);
+	if (status) {
+		/* allocate profile ID */
+		status = ice_alloc_prof_id(hw, blk, &prof_id);
+		if (status)
+			goto err_ice_add_prof;
+
+		/* and write new es */
+		ice_write_es(hw, blk, prof_id, es);
+	}
+
+	ice_prof_inc_ref(hw, blk, prof_id);
+
+	/* add profile info */
+	prof = devm_kzalloc(ice_hw_to_dev(hw), sizeof(*prof), GFP_KERNEL);
+	if (!prof)
+		goto err_ice_add_prof;
+
+	prof->profile_cookie = id;
+	prof->prof_id = prof_id;
+	prof->ptg_cnt = 0;
+	prof->context = 0;
+
+	/* build list of ptgs */
+	while (bytes && prof->ptg_cnt < ICE_MAX_PTG_PER_PROFILE) {
+		u32 bit;
+
+		if (!ptypes[byte]) {
+			bytes--;
+			byte++;
+			continue;
+		}
+
+		/* Examine 8 bits per byte */
+		for_each_set_bit(bit, (unsigned long *)&ptypes[byte],
+				 BITS_PER_BYTE) {
+			u16 ptype;
+			u8 ptg;
+			u8 m;
+
+			ptype = byte * BITS_PER_BYTE + bit;
+
+			/* The package should place all ptypes in a non-zero
+			 * PTG, so the following call should never fail.
+			 */
+			if (ice_ptg_find_ptype(hw, blk, ptype, &ptg))
+				continue;
+
+			/* If PTG is already added, skip and continue */
+			if (test_bit(ptg, ptgs_used))
+				continue;
+
+			set_bit(ptg, ptgs_used);
+			prof->ptg[prof->ptg_cnt] = ptg;
+
+			if (++prof->ptg_cnt >= ICE_MAX_PTG_PER_PROFILE)
+				break;
+
+			/* nothing left in byte, then exit */
+			m = ~((1 << (bit + 1)) - 1);
+			if (!(ptypes[byte] & m))
+				break;
+		}
+
+		bytes--;
+		byte++;
+	}
+
+	list_add(&prof->list, &hw->blk[blk].es.prof_map);
+	status = 0;
+
+err_ice_add_prof:
+	mutex_unlock(&hw->blk[blk].es.prof_map_lock);
+	return status;
+}
diff --git a/drivers/net/ethernet/intel/ice/ice_flex_pipe.h b/drivers/net/ethernet/intel/ice/ice_flex_pipe.h
index 37eb282742d1..8cb7d7f09e0b 100644
--- a/drivers/net/ethernet/intel/ice/ice_flex_pipe.h
+++ b/drivers/net/ethernet/intel/ice/ice_flex_pipe.h
@@ -18,6 +18,9 @@
 
 #define ICE_PKG_CNT 4
 
+enum ice_status
+ice_add_prof(struct ice_hw *hw, enum ice_block blk, u64 id, u8 ptypes[],
+	     struct ice_fv_word *es);
 enum ice_status ice_init_pkg(struct ice_hw *hw, u8 *buff, u32 len);
 enum ice_status
 ice_copy_and_init_pkg(struct ice_hw *hw, const u8 *buf, u32 len);
diff --git a/drivers/net/ethernet/intel/ice/ice_flex_type.h b/drivers/net/ethernet/intel/ice/ice_flex_type.h
index 5d5a7eaffa30..3005f111fb3b 100644
--- a/drivers/net/ethernet/intel/ice/ice_flex_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_flex_type.h
@@ -3,6 +3,9 @@
 
 #ifndef _ICE_FLEX_TYPE_H_
 #define _ICE_FLEX_TYPE_H_
+
+#define ICE_FV_OFFSET_INVAL	0x1FF
+
 /* Extraction Sequence (Field Vector) Table */
 struct ice_fv_word {
 	u8 prot_id;
@@ -280,6 +283,17 @@ struct ice_ptg_ptype {
 	u8 ptg;
 };
 
+#define ICE_MAX_PTG_PER_PROFILE		32
+
+struct ice_prof_map {
+	struct list_head list;
+	u64 profile_cookie;
+	u64 context;
+	u8 prof_id;
+	u8 ptg_cnt;
+	u8 ptg[ICE_MAX_PTG_PER_PROFILE];
+};
+
 struct ice_vsig_entry {
 	struct list_head prop_lst;
 	struct ice_vsig_vsi *first_vsi;
@@ -371,4 +385,5 @@ struct ice_blk_info {
 	u8 is_list_init;
 };
 
+#define ICE_FLOW_PTYPE_MAX		ICE_XLT1_CNT
 #endif /* _ICE_FLEX_TYPE_H_ */
diff --git a/drivers/net/ethernet/intel/ice/ice_flow.c b/drivers/net/ethernet/intel/ice/ice_flow.c
index 4828c64abe98..24fe04f8baa2 100644
--- a/drivers/net/ethernet/intel/ice/ice_flow.c
+++ b/drivers/net/ethernet/intel/ice/ice_flow.c
@@ -40,6 +40,457 @@ struct ice_flow_field_info ice_flds_info[ICE_FLOW_FIELD_IDX_MAX] = {
 	ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_UDP, 2, sizeof(__be16)),
 };
 
+/* Bitmaps indicating relevant packet types for a particular protocol header
+ *
+ * Packet types for packets with an Outer/First/Single IPv4 header
+ */
+static const u32 ice_ptypes_ipv4_ofos[] = {
+	0x1DC00000, 0x04000800, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+};
+
+/* Packet types for packets with an Innermost/Last IPv4 header */
+static const u32 ice_ptypes_ipv4_il[] = {
+	0xE0000000, 0xB807700E, 0x80000003, 0xE01DC03B,
+	0x0000000E, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+};
+
+/* Packet types for packets with an Outer/First/Single IPv6 header */
+static const u32 ice_ptypes_ipv6_ofos[] = {
+	0x00000000, 0x00000000, 0x77000000, 0x10002000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+};
+
+/* Packet types for packets with an Innermost/Last IPv6 header */
+static const u32 ice_ptypes_ipv6_il[] = {
+	0x00000000, 0x03B80770, 0x000001DC, 0x0EE00000,
+	0x00000770, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+};
+
+/* UDP Packet types for non-tunneled packets or tunneled
+ * packets with inner UDP.
+ */
+static const u32 ice_ptypes_udp_il[] = {
+	0x81000000, 0x20204040, 0x04000010, 0x80810102,
+	0x00000040, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+};
+
+/* Packet types for packets with an Innermost/Last TCP header */
+static const u32 ice_ptypes_tcp_il[] = {
+	0x04000000, 0x80810102, 0x10000040, 0x02040408,
+	0x00000102, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+};
+
+/* Packet types for packets with an Innermost/Last SCTP header */
+static const u32 ice_ptypes_sctp_il[] = {
+	0x08000000, 0x01020204, 0x20000081, 0x04080810,
+	0x00000204, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+};
+
+/* Manage parameters and info. used during the creation of a flow profile */
+struct ice_flow_prof_params {
+	enum ice_block blk;
+	u16 entry_length; /* # of bytes formatted entry will require */
+	u8 es_cnt;
+	struct ice_flow_prof *prof;
+
+	/* For ACL, the es[0] will have the data of ICE_RX_MDID_PKT_FLAGS_15_0
+	 * This will give us the direction flags.
+	 */
+	struct ice_fv_word es[ICE_MAX_FV_WORDS];
+	DECLARE_BITMAP(ptypes, ICE_FLOW_PTYPE_MAX);
+};
+
+#define ICE_FLOW_SEG_HDRS_L3_MASK	\
+	(ICE_FLOW_SEG_HDR_IPV4 | ICE_FLOW_SEG_HDR_IPV6)
+#define ICE_FLOW_SEG_HDRS_L4_MASK	\
+	(ICE_FLOW_SEG_HDR_TCP | ICE_FLOW_SEG_HDR_UDP | ICE_FLOW_SEG_HDR_SCTP)
+
+/**
+ * ice_flow_val_hdrs - validates packet segments for valid protocol headers
+ * @segs: array of one or more packet segments that describe the flow
+ * @segs_cnt: number of packet segments provided
+ */
+static enum ice_status
+ice_flow_val_hdrs(struct ice_flow_seg_info *segs, u8 segs_cnt)
+{
+	u8 i;
+
+	for (i = 0; i < segs_cnt; i++) {
+		/* Multiple L3 headers */
+		if (segs[i].hdrs & ICE_FLOW_SEG_HDRS_L3_MASK &&
+		    !is_power_of_2(segs[i].hdrs & ICE_FLOW_SEG_HDRS_L3_MASK))
+			return ICE_ERR_PARAM;
+
+		/* Multiple L4 headers */
+		if (segs[i].hdrs & ICE_FLOW_SEG_HDRS_L4_MASK &&
+		    !is_power_of_2(segs[i].hdrs & ICE_FLOW_SEG_HDRS_L4_MASK))
+			return ICE_ERR_PARAM;
+	}
+
+	return 0;
+}
+
+/**
+ * ice_flow_proc_seg_hdrs - process protocol headers present in pkt segments
+ * @params: information about the flow to be processed
+ *
+ * This function identifies the packet types associated with the protocol
+ * headers being present in packet segments of the specified flow profile.
+ */
+static enum ice_status
+ice_flow_proc_seg_hdrs(struct ice_flow_prof_params *params)
+{
+	struct ice_flow_prof *prof;
+	u8 i;
+
+	memset(params->ptypes, 0xff, sizeof(params->ptypes));
+
+	prof = params->prof;
+
+	for (i = 0; i < params->prof->segs_cnt; i++) {
+		const unsigned long *src;
+		u32 hdrs;
+
+		hdrs = prof->segs[i].hdrs;
+
+		if (hdrs & ICE_FLOW_SEG_HDR_IPV4) {
+			src = !i ? (const unsigned long *)ice_ptypes_ipv4_ofos :
+				(const unsigned long *)ice_ptypes_ipv4_il;
+			bitmap_and(params->ptypes, params->ptypes, src,
+				   ICE_FLOW_PTYPE_MAX);
+		} else if (hdrs & ICE_FLOW_SEG_HDR_IPV6) {
+			src = !i ? (const unsigned long *)ice_ptypes_ipv6_ofos :
+				(const unsigned long *)ice_ptypes_ipv6_il;
+			bitmap_and(params->ptypes, params->ptypes, src,
+				   ICE_FLOW_PTYPE_MAX);
+		}
+
+		if (hdrs & ICE_FLOW_SEG_HDR_UDP) {
+			src = (const unsigned long *)ice_ptypes_udp_il;
+			bitmap_and(params->ptypes, params->ptypes, src,
+				   ICE_FLOW_PTYPE_MAX);
+		} else if (hdrs & ICE_FLOW_SEG_HDR_TCP) {
+			bitmap_and(params->ptypes, params->ptypes,
+				   (const unsigned long *)ice_ptypes_tcp_il,
+				   ICE_FLOW_PTYPE_MAX);
+		} else if (hdrs & ICE_FLOW_SEG_HDR_SCTP) {
+			src = (const unsigned long *)ice_ptypes_sctp_il;
+			bitmap_and(params->ptypes, params->ptypes, src,
+				   ICE_FLOW_PTYPE_MAX);
+		}
+	}
+
+	return 0;
+}
+
+/**
+ * ice_flow_xtract_fld - Create an extraction sequence entry for the given field
+ * @hw: pointer to the HW struct
+ * @params: information about the flow to be processed
+ * @seg: packet segment index of the field to be extracted
+ * @fld: ID of field to be extracted
+ *
+ * This function determines the protocol ID, offset, and size of the given
+ * field. It then allocates one or more extraction sequence entries for the
+ * given field, and fill the entries with protocol ID and offset information.
+ */
+static enum ice_status
+ice_flow_xtract_fld(struct ice_hw *hw, struct ice_flow_prof_params *params,
+		    u8 seg, enum ice_flow_field fld)
+{
+	enum ice_prot_id prot_id = ICE_PROT_ID_INVAL;
+	u8 fv_words = hw->blk[params->blk].es.fvw;
+	struct ice_flow_fld_info *flds;
+	u16 cnt, ese_bits, i;
+	u16 off;
+
+	flds = params->prof->segs[seg].fields;
+
+	switch (fld) {
+	case ICE_FLOW_FIELD_IDX_IPV4_SA:
+	case ICE_FLOW_FIELD_IDX_IPV4_DA:
+		prot_id = seg == 0 ? ICE_PROT_IPV4_OF_OR_S : ICE_PROT_IPV4_IL;
+		break;
+	case ICE_FLOW_FIELD_IDX_IPV6_SA:
+	case ICE_FLOW_FIELD_IDX_IPV6_DA:
+		prot_id = seg == 0 ? ICE_PROT_IPV6_OF_OR_S : ICE_PROT_IPV6_IL;
+		break;
+	case ICE_FLOW_FIELD_IDX_TCP_SRC_PORT:
+	case ICE_FLOW_FIELD_IDX_TCP_DST_PORT:
+		prot_id = ICE_PROT_TCP_IL;
+		break;
+	case ICE_FLOW_FIELD_IDX_UDP_SRC_PORT:
+	case ICE_FLOW_FIELD_IDX_UDP_DST_PORT:
+		prot_id = ICE_PROT_UDP_IL_OR_S;
+		break;
+	default:
+		return ICE_ERR_NOT_IMPL;
+	}
+
+	/* Each extraction sequence entry is a word in size, and extracts a
+	 * word-aligned offset from a protocol header.
+	 */
+	ese_bits = ICE_FLOW_FV_EXTRACT_SZ * BITS_PER_BYTE;
+
+	flds[fld].xtrct.prot_id = prot_id;
+	flds[fld].xtrct.off = (ice_flds_info[fld].off / ese_bits) *
+		ICE_FLOW_FV_EXTRACT_SZ;
+	flds[fld].xtrct.disp = (u8)(ice_flds_info[fld].off % ese_bits);
+	flds[fld].xtrct.idx = params->es_cnt;
+
+	/* Adjust the next field-entry index after accommodating the number of
+	 * entries this field consumes
+	 */
+	cnt = DIV_ROUND_UP(flds[fld].xtrct.disp + ice_flds_info[fld].size,
+			   ese_bits);
+
+	/* Fill in the extraction sequence entries needed for this field */
+	off = flds[fld].xtrct.off;
+	for (i = 0; i < cnt; i++) {
+		u8 idx;
+
+		/* Make sure the number of extraction sequence required
+		 * does not exceed the block's capability
+		 */
+		if (params->es_cnt >= fv_words)
+			return ICE_ERR_MAX_LIMIT;
+
+		/* some blocks require a reversed field vector layout */
+		if (hw->blk[params->blk].es.reverse)
+			idx = fv_words - params->es_cnt - 1;
+		else
+			idx = params->es_cnt;
+
+		params->es[idx].prot_id = prot_id;
+		params->es[idx].off = off;
+		params->es_cnt++;
+
+		off += ICE_FLOW_FV_EXTRACT_SZ;
+	}
+
+	return 0;
+}
+
+/**
+ * ice_flow_create_xtrct_seq - Create an extraction sequence for given segments
+ * @hw: pointer to the HW struct
+ * @params: information about the flow to be processed
+ *
+ * This function iterates through all matched fields in the given segments, and
+ * creates an extraction sequence for the fields.
+ */
+static enum ice_status
+ice_flow_create_xtrct_seq(struct ice_hw *hw,
+			  struct ice_flow_prof_params *params)
+{
+	struct ice_flow_prof *prof = params->prof;
+	enum ice_status status = 0;
+	u8 i;
+
+	for (i = 0; i < prof->segs_cnt; i++) {
+		u8 j;
+
+		for_each_set_bit(j, (unsigned long *)&prof->segs[i].match,
+				 ICE_FLOW_FIELD_IDX_MAX) {
+			status = ice_flow_xtract_fld(hw, params, i,
+						     (enum ice_flow_field)j);
+			if (status)
+				return status;
+		}
+	}
+
+	return status;
+}
+
+/**
+ * ice_flow_proc_segs - process all packet segments associated with a profile
+ * @hw: pointer to the HW struct
+ * @params: information about the flow to be processed
+ */
+static enum ice_status
+ice_flow_proc_segs(struct ice_hw *hw, struct ice_flow_prof_params *params)
+{
+	enum ice_status status;
+
+	status = ice_flow_proc_seg_hdrs(params);
+	if (status)
+		return status;
+
+	status = ice_flow_create_xtrct_seq(hw, params);
+	if (status)
+		return status;
+
+	switch (params->blk) {
+	case ICE_BLK_RSS:
+		/* Only header information is provided for RSS configuration.
+		 * No further processing is needed.
+		 */
+		status = 0;
+		break;
+	default:
+		return ICE_ERR_NOT_IMPL;
+	}
+
+	return status;
+}
+
+/**
+ * ice_flow_add_prof_sync - Add a flow profile for packet segments and fields
+ * @hw: pointer to the HW struct
+ * @blk: classification stage
+ * @dir: flow direction
+ * @prof_id: unique ID to identify this flow profile
+ * @segs: array of one or more packet segments that describe the flow
+ * @segs_cnt: number of packet segments provided
+ * @prof: stores the returned flow profile added
+ *
+ * Assumption: the caller has acquired the lock to the profile list
+ */
+static enum ice_status
+ice_flow_add_prof_sync(struct ice_hw *hw, enum ice_block blk,
+		       enum ice_flow_dir dir, u64 prof_id,
+		       struct ice_flow_seg_info *segs, u8 segs_cnt,
+		       struct ice_flow_prof **prof)
+{
+	struct ice_flow_prof_params params;
+	enum ice_status status;
+	u8 i;
+
+	if (!prof)
+		return ICE_ERR_BAD_PTR;
+
+	memset(&params, 0, sizeof(params));
+	params.prof = devm_kzalloc(ice_hw_to_dev(hw), sizeof(*params.prof),
+				   GFP_KERNEL);
+	if (!params.prof)
+		return ICE_ERR_NO_MEMORY;
+
+	/* initialize extraction sequence to all invalid (0xff) */
+	for (i = 0; i < ICE_MAX_FV_WORDS; i++) {
+		params.es[i].prot_id = ICE_PROT_INVALID;
+		params.es[i].off = ICE_FV_OFFSET_INVAL;
+	}
+
+	params.blk = blk;
+	params.prof->id = prof_id;
+	params.prof->dir = dir;
+	params.prof->segs_cnt = segs_cnt;
+
+	/* Make a copy of the segments that need to be persistent in the flow
+	 * profile instance
+	 */
+	for (i = 0; i < segs_cnt; i++)
+		memcpy(&params.prof->segs[i], &segs[i], sizeof(*segs));
+
+	status = ice_flow_proc_segs(hw, &params);
+	if (status) {
+		ice_debug(hw, ICE_DBG_FLOW,
+			  "Error processing a flow's packet segments\n");
+		goto out;
+	}
+
+	/* Add a HW profile for this flow profile */
+	status = ice_add_prof(hw, blk, prof_id, (u8 *)params.ptypes, params.es);
+	if (status) {
+		ice_debug(hw, ICE_DBG_FLOW, "Error adding a HW flow profile\n");
+		goto out;
+	}
+
+	INIT_LIST_HEAD(&params.prof->entries);
+	mutex_init(&params.prof->entries_lock);
+	*prof = params.prof;
+
+out:
+	if (status)
+		devm_kfree(ice_hw_to_dev(hw), params.prof);
+
+	return status;
+}
+
+/**
+ * ice_flow_add_prof - Add a flow profile for packet segments and matched fields
+ * @hw: pointer to the HW struct
+ * @blk: classification stage
+ * @dir: flow direction
+ * @prof_id: unique ID to identify this flow profile
+ * @segs: array of one or more packet segments that describe the flow
+ * @segs_cnt: number of packet segments provided
+ */
+static enum ice_status
+ice_flow_add_prof(struct ice_hw *hw, enum ice_block blk, enum ice_flow_dir dir,
+		  u64 prof_id, struct ice_flow_seg_info *segs, u8 segs_cnt)
+{
+	struct ice_flow_prof *prof = NULL;
+	enum ice_status status;
+
+	if (segs_cnt > ICE_FLOW_SEG_MAX)
+		return ICE_ERR_MAX_LIMIT;
+
+	if (!segs_cnt)
+		return ICE_ERR_PARAM;
+
+	if (!segs)
+		return ICE_ERR_BAD_PTR;
+
+	status = ice_flow_val_hdrs(segs, segs_cnt);
+	if (status)
+		return status;
+
+	mutex_lock(&hw->fl_profs_locks[blk]);
+
+	status = ice_flow_add_prof_sync(hw, blk, dir, prof_id, segs, segs_cnt,
+					&prof);
+	if (!status)
+		list_add(&prof->l_entry, &hw->fl_profs[blk]);
+
+	mutex_unlock(&hw->fl_profs_locks[blk]);
+
+	return status;
+}
+
 /**
  * ice_flow_set_fld_ext - specifies locations of field from entry's input buffer
  * @seg: packet segment the field being set belongs to
@@ -161,10 +612,28 @@ ice_flow_set_rss_seg_info(struct ice_flow_seg_info *segs, u64 hash_fields,
 	return 0;
 }
 
+#define ICE_FLOW_PROF_HASH_S	0
+#define ICE_FLOW_PROF_HASH_M	(0xFFFFFFFFULL << ICE_FLOW_PROF_HASH_S)
+#define ICE_FLOW_PROF_HDR_S	32
+#define ICE_FLOW_PROF_HDR_M	(0x3FFFFFFFULL << ICE_FLOW_PROF_HDR_S)
+#define ICE_FLOW_PROF_ENCAP_S	63
+#define ICE_FLOW_PROF_ENCAP_M	(BIT_ULL(ICE_FLOW_PROF_ENCAP_S))
+
 #define ICE_RSS_OUTER_HEADERS	1
 
+/* Flow profile ID format:
+ * [0:31] - Packet match fields
+ * [32:62] - Protocol header
+ * [63] - Encapsulation flag, 0 if non-tunneled, 1 if tunneled
+ */
+#define ICE_FLOW_GEN_PROFID(hash, hdr, segs_cnt) \
+	(u64)(((u64)(hash) & ICE_FLOW_PROF_HASH_M) | \
+	      (((u64)(hdr) << ICE_FLOW_PROF_HDR_S) & ICE_FLOW_PROF_HDR_M) | \
+	      ((u8)((segs_cnt) - 1) ? ICE_FLOW_PROF_ENCAP_M : 0))
+
 /**
  * ice_add_rss_cfg_sync - add an RSS configuration
+ * @hw: pointer to the hardware structure
  * @hashed_flds: hash bit fields (ICE_FLOW_HASH_*) to configure
  * @addl_hdrs: protocol header fields
  * @segs_cnt: packet segment count
@@ -172,7 +641,8 @@ ice_flow_set_rss_seg_info(struct ice_flow_seg_info *segs, u64 hash_fields,
  * Assumption: lock has already been acquired for RSS list
  */
 static enum ice_status
-ice_add_rss_cfg_sync(u64 hashed_flds, u32 addl_hdrs, u8 segs_cnt)
+ice_add_rss_cfg_sync(struct ice_hw *hw, u64 hashed_flds, u32 addl_hdrs,
+		     u8 segs_cnt)
 {
 	struct ice_flow_seg_info *segs;
 	enum ice_status status;
@@ -187,7 +657,19 @@ ice_add_rss_cfg_sync(u64 hashed_flds, u32 addl_hdrs, u8 segs_cnt)
 	/* Construct the packet segment info from the hashed fields */
 	status = ice_flow_set_rss_seg_info(&segs[segs_cnt - 1], hashed_flds,
 					   addl_hdrs);
-
+	if (status)
+		goto exit;
+
+	/* Create a new flow profile with generated profile and packet
+	 * segment information.
+	 */
+	status = ice_flow_add_prof(hw, ICE_BLK_RSS, ICE_FLOW_RX,
+				   ICE_FLOW_GEN_PROFID(hashed_flds,
+						       segs[segs_cnt - 1].hdrs,
+						       segs_cnt),
+				   segs, segs_cnt);
+
+exit:
 	kfree(segs);
 	return status;
 }
@@ -214,7 +696,7 @@ ice_add_rss_cfg(struct ice_hw *hw, u16 vsi_handle, u64 hashed_flds,
 		return ICE_ERR_PARAM;
 
 	mutex_lock(&hw->rss_locks);
-	status = ice_add_rss_cfg_sync(hashed_flds, addl_hdrs,
+	status = ice_add_rss_cfg_sync(hw, hashed_flds, addl_hdrs,
 				      ICE_RSS_OUTER_HEADERS);
 	mutex_unlock(&hw->rss_locks);
 
@@ -237,7 +719,7 @@ enum ice_status ice_replay_rss_cfg(struct ice_hw *hw, u16 vsi_handle)
 	mutex_lock(&hw->rss_locks);
 	list_for_each_entry(r, &hw->rss_list_head, l_entry) {
 		if (test_bit(vsi_handle, r->vsis)) {
-			status = ice_add_rss_cfg_sync(r->hashed_flds,
+			status = ice_add_rss_cfg_sync(hw, r->hashed_flds,
 						      r->packet_hdr,
 						      ICE_RSS_OUTER_HEADERS);
 			if (status)
diff --git a/drivers/net/ethernet/intel/ice/ice_flow.h b/drivers/net/ethernet/intel/ice/ice_flow.h
index 48c0fc09d5ff..05b0dab4793c 100644
--- a/drivers/net/ethernet/intel/ice/ice_flow.h
+++ b/drivers/net/ethernet/intel/ice/ice_flow.h
@@ -56,7 +56,13 @@ enum ice_flow_field {
 	ICE_FLOW_FIELD_IDX_MAX
 };
 
+enum ice_flow_dir {
+	ICE_FLOW_RX		= 0x02,
+};
+
 #define ICE_FLOW_SEG_MAX		2
+#define ICE_FLOW_FV_EXTRACT_SZ		2
+
 #define ICE_FLOW_SET_HDRS(seg, val)	((seg)->hdrs |= (u32)(val))
 
 struct ice_flow_seg_xtrct {
@@ -99,6 +105,23 @@ struct ice_flow_seg_info {
 	struct ice_flow_fld_info fields[ICE_FLOW_FIELD_IDX_MAX];
 };
 
+struct ice_flow_prof {
+	struct list_head l_entry;
+
+	u64 id;
+	enum ice_flow_dir dir;
+	u8 segs_cnt;
+
+	/* Keep track of flow entries associated with this flow profile */
+	struct mutex entries_lock;
+	struct list_head entries;
+
+	struct ice_flow_seg_info segs[ICE_FLOW_SEG_MAX];
+
+	/* software VSI handles referenced by this flow profile */
+	DECLARE_BITMAP(vsis, ICE_MAX_VSI);
+};
+
 struct ice_rss_cfg {
 	struct list_head l_entry;
 	/* bitmap of VSIs added to the RSS entry */
diff --git a/drivers/net/ethernet/intel/ice/ice_lan_tx_rx.h b/drivers/net/ethernet/intel/ice/ice_lan_tx_rx.h
index 0997d352709b..878e125d8b42 100644
--- a/drivers/net/ethernet/intel/ice/ice_lan_tx_rx.h
+++ b/drivers/net/ethernet/intel/ice/ice_lan_tx_rx.h
@@ -199,6 +199,14 @@ enum ice_rxdid {
 /* Receive Flex Descriptor Rx opcode values */
 #define ICE_RX_OPC_MDID		0x01
 
+/* Receive Descriptor MDID values that access packet flags */
+enum ice_flex_mdid_pkt_flags {
+	ICE_RX_MDID_PKT_FLAGS_15_0	= 20,
+	ICE_RX_MDID_PKT_FLAGS_31_16,
+	ICE_RX_MDID_PKT_FLAGS_47_32,
+	ICE_RX_MDID_PKT_FLAGS_63_48,
+};
+
 /* Receive Descriptor MDID values */
 enum ice_flex_rx_mdid {
 	ICE_RX_MDID_FLOW_ID_LOWER	= 5,
diff --git a/drivers/net/ethernet/intel/ice/ice_protocol_type.h b/drivers/net/ethernet/intel/ice/ice_protocol_type.h
new file mode 100644
index 000000000000..97db46ce7a21
--- /dev/null
+++ b/drivers/net/ethernet/intel/ice/ice_protocol_type.h
@@ -0,0 +1,24 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) 2019, Intel Corporation. */
+
+#ifndef _ICE_PROTOCOL_TYPE_H_
+#define _ICE_PROTOCOL_TYPE_H_
+/* Decoders for ice_prot_id:
+ * - F: First
+ * - I: Inner
+ * - L: Last
+ * - O: Outer
+ * - S: Single
+ */
+enum ice_prot_id {
+	ICE_PROT_ID_INVAL	= 0,
+	ICE_PROT_IPV4_OF_OR_S	= 32,
+	ICE_PROT_IPV4_IL	= 33,
+	ICE_PROT_IPV6_OF_OR_S	= 40,
+	ICE_PROT_IPV6_IL	= 41,
+	ICE_PROT_TCP_IL		= 49,
+	ICE_PROT_UDP_IL_OR_S	= 53,
+	ICE_PROT_META_ID	= 255, /* when offset == metadata */
+	ICE_PROT_INVALID	= 255  /* when offset == ICE_FV_OFFSET_INVAL */
+};
+#endif /* _ICE_PROTOCOL_TYPE_H_ */
diff --git a/drivers/net/ethernet/intel/ice/ice_switch.c b/drivers/net/ethernet/intel/ice/ice_switch.c
index b5a53f862a83..431266081a80 100644
--- a/drivers/net/ethernet/intel/ice/ice_switch.c
+++ b/drivers/net/ethernet/intel/ice/ice_switch.c
@@ -49,42 +49,6 @@ static const u8 dummy_eth_header[DUMMY_ETH_HDR_LEN] = { 0x2, 0, 0, 0, 0, 0,
 	 sizeof(((struct ice_sw_rule_vsi_list *)0)->vsi) + \
 	 ((n) * sizeof(((struct ice_sw_rule_vsi_list *)0)->vsi)))
 
-/**
- * ice_aq_alloc_free_res - command to allocate/free resources
- * @hw: pointer to the HW struct
- * @num_entries: number of resource entries in buffer
- * @buf: Indirect buffer to hold data parameters and response
- * @buf_size: size of buffer for indirect commands
- * @opc: pass in the command opcode
- * @cd: pointer to command details structure or NULL
- *
- * Helper function to allocate/free resources using the admin queue commands
- */
-static enum ice_status
-ice_aq_alloc_free_res(struct ice_hw *hw, u16 num_entries,
-		      struct ice_aqc_alloc_free_res_elem *buf, u16 buf_size,
-		      enum ice_adminq_opc opc, struct ice_sq_cd *cd)
-{
-	struct ice_aqc_alloc_free_res_cmd *cmd;
-	struct ice_aq_desc desc;
-
-	cmd = &desc.params.sw_res_ctrl;
-
-	if (!buf)
-		return ICE_ERR_PARAM;
-
-	if (buf_size < (num_entries * sizeof(buf->elem[0])))
-		return ICE_ERR_PARAM;
-
-	ice_fill_dflt_direct_cmd_desc(&desc, opc);
-
-	desc.flags |= cpu_to_le16(ICE_AQ_FLAG_RD);
-
-	cmd->num_entries = cpu_to_le16(num_entries);
-
-	return ice_aq_send_cmd(hw, &desc, buf, buf_size, cd);
-}
-
 /**
  * ice_init_def_sw_recp - initialize the recipe book keeping tables
  * @hw: pointer to the HW struct
diff --git a/drivers/net/ethernet/intel/ice/ice_type.h b/drivers/net/ethernet/intel/ice/ice_type.h
index ac38af105f90..b361ffabb0ca 100644
--- a/drivers/net/ethernet/intel/ice/ice_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_type.h
@@ -13,6 +13,7 @@
 #include "ice_controlq.h"
 #include "ice_lan_tx_rx.h"
 #include "ice_flex_type.h"
+#include "ice_protocol_type.h"
 
 static inline bool ice_is_tc_ena(unsigned long bitmap, u8 tc)
 {
@@ -41,6 +42,7 @@ static inline u32 ice_round_to_num(u32 N, u32 R)
 #define ICE_DBG_QCTX		BIT_ULL(6)
 #define ICE_DBG_NVM		BIT_ULL(7)
 #define ICE_DBG_LAN		BIT_ULL(8)
+#define ICE_DBG_FLOW		BIT_ULL(9)
 #define ICE_DBG_SW		BIT_ULL(13)
 #define ICE_DBG_SCHED		BIT_ULL(14)
 #define ICE_DBG_PKG		BIT_ULL(16)
-- 
cgit v1.2.3-59-g8ed1b


From e42f1ac626e7f799717d006e0f8393b6d6f9fc8c Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Fri, 24 Jan 2020 16:04:02 -0800
Subject: mptcp: do not inherit inet proto ops

We need to initialise the struct ourselves, else we expose tcp-specific
callbacks such as tcp_splice_read which will then trigger splat because
the socket is an mptcp one:

BUG: KASAN: slab-out-of-bounds in tcp_mstamp_refresh+0x80/0xa0 net/ipv4/tcp_output.c:57
Write of size 8 at addr ffff888116aa21d0 by task syz-executor.0/5478

CPU: 1 PID: 5478 Comm: syz-executor.0 Not tainted 5.5.0-rc6 #3
Call Trace:
 tcp_mstamp_refresh+0x80/0xa0 net/ipv4/tcp_output.c:57
 tcp_rcv_space_adjust+0x72/0x7f0 net/ipv4/tcp_input.c:612
 tcp_read_sock+0x622/0x990 net/ipv4/tcp.c:1674
 tcp_splice_read+0x20b/0xb40 net/ipv4/tcp.c:791
 do_splice+0x1259/0x1560 fs/splice.c:1205

To prevent build error with ipv6, add the recv/sendmsg function
declaration to ipv6.h.  The functions are already accessible "thanks"
to retpoline related work, but they are currently only made visible
by socket.c specific INDIRECT_CALLABLE macros.

Reported-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ipv6.h   |  3 +++
 net/mptcp/protocol.c | 70 ++++++++++++++++++++++++++++++++++++++--------------
 2 files changed, 54 insertions(+), 19 deletions(-)

diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 4e95f6df508c..cec1a54401f2 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -1113,6 +1113,9 @@ int inet6_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg);
 
 int inet6_hash_connect(struct inet_timewait_death_row *death_row,
 			      struct sock *sk);
+int inet6_sendmsg(struct socket *sock, struct msghdr *msg, size_t size);
+int inet6_recvmsg(struct socket *sock, struct msghdr *msg, size_t size,
+		  int flags);
 
 /*
  * reassembly.c
diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index ee916b3686fe..41d49126d29a 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -1163,7 +1163,31 @@ out_unlock:
 	return ret;
 }
 
-static struct proto_ops mptcp_stream_ops;
+static const struct proto_ops mptcp_stream_ops = {
+	.family		   = PF_INET,
+	.owner		   = THIS_MODULE,
+	.release	   = inet_release,
+	.bind		   = mptcp_bind,
+	.connect	   = mptcp_stream_connect,
+	.socketpair	   = sock_no_socketpair,
+	.accept		   = mptcp_stream_accept,
+	.getname	   = mptcp_v4_getname,
+	.poll		   = mptcp_poll,
+	.ioctl		   = inet_ioctl,
+	.gettstamp	   = sock_gettstamp,
+	.listen		   = mptcp_listen,
+	.shutdown	   = mptcp_shutdown,
+	.setsockopt	   = sock_common_setsockopt,
+	.getsockopt	   = sock_common_getsockopt,
+	.sendmsg	   = inet_sendmsg,
+	.recvmsg	   = inet_recvmsg,
+	.mmap		   = sock_no_mmap,
+	.sendpage	   = inet_sendpage,
+#ifdef CONFIG_COMPAT
+	.compat_setsockopt = compat_sock_common_setsockopt,
+	.compat_getsockopt = compat_sock_common_getsockopt,
+#endif
+};
 
 static struct inet_protosw mptcp_protosw = {
 	.type		= SOCK_STREAM,
@@ -1176,14 +1200,6 @@ static struct inet_protosw mptcp_protosw = {
 void mptcp_proto_init(void)
 {
 	mptcp_prot.h.hashinfo = tcp_prot.h.hashinfo;
-	mptcp_stream_ops = inet_stream_ops;
-	mptcp_stream_ops.bind = mptcp_bind;
-	mptcp_stream_ops.connect = mptcp_stream_connect;
-	mptcp_stream_ops.poll = mptcp_poll;
-	mptcp_stream_ops.accept = mptcp_stream_accept;
-	mptcp_stream_ops.getname = mptcp_v4_getname;
-	mptcp_stream_ops.listen = mptcp_listen;
-	mptcp_stream_ops.shutdown = mptcp_shutdown;
 
 	mptcp_subflow_init();
 
@@ -1194,7 +1210,32 @@ void mptcp_proto_init(void)
 }
 
 #if IS_ENABLED(CONFIG_MPTCP_IPV6)
-static struct proto_ops mptcp_v6_stream_ops;
+static const struct proto_ops mptcp_v6_stream_ops = {
+	.family		   = PF_INET6,
+	.owner		   = THIS_MODULE,
+	.release	   = inet6_release,
+	.bind		   = mptcp_bind,
+	.connect	   = mptcp_stream_connect,
+	.socketpair	   = sock_no_socketpair,
+	.accept		   = mptcp_stream_accept,
+	.getname	   = mptcp_v6_getname,
+	.poll		   = mptcp_poll,
+	.ioctl		   = inet6_ioctl,
+	.gettstamp	   = sock_gettstamp,
+	.listen		   = mptcp_listen,
+	.shutdown	   = mptcp_shutdown,
+	.setsockopt	   = sock_common_setsockopt,
+	.getsockopt	   = sock_common_getsockopt,
+	.sendmsg	   = inet6_sendmsg,
+	.recvmsg	   = inet6_recvmsg,
+	.mmap		   = sock_no_mmap,
+	.sendpage	   = inet_sendpage,
+#ifdef CONFIG_COMPAT
+	.compat_setsockopt = compat_sock_common_setsockopt,
+	.compat_getsockopt = compat_sock_common_getsockopt,
+#endif
+};
+
 static struct proto mptcp_v6_prot;
 
 static void mptcp_v6_destroy(struct sock *sk)
@@ -1226,15 +1267,6 @@ int mptcp_proto_v6_init(void)
 	if (err)
 		return err;
 
-	mptcp_v6_stream_ops = inet6_stream_ops;
-	mptcp_v6_stream_ops.bind = mptcp_bind;
-	mptcp_v6_stream_ops.connect = mptcp_stream_connect;
-	mptcp_v6_stream_ops.poll = mptcp_poll;
-	mptcp_v6_stream_ops.accept = mptcp_stream_accept;
-	mptcp_v6_stream_ops.getname = mptcp_v6_getname;
-	mptcp_v6_stream_ops.listen = mptcp_listen;
-	mptcp_v6_stream_ops.shutdown = mptcp_shutdown;
-
 	err = inet6_register_protosw(&mptcp_v6_protosw);
 	if (err)
 		proto_unregister(&mptcp_v6_prot);
-- 
cgit v1.2.3-59-g8ed1b


From edc7e4898d5fdb164de1ab9d3a0c30ef9888e4f1 Mon Sep 17 00:00:00 2001
From: Mat Martineau <mathew.j.martineau@linux.intel.com>
Date: Fri, 24 Jan 2020 16:04:03 -0800
Subject: mptcp: Fix code formatting

checkpatch.pl had a few complaints in the last set of MPTCP patches:

ERROR: code indent should use tabs where possible
+^I         subflow, sk->sk_family, icsk->icsk_af_ops, target, mapped);$

CHECK: Comparison to NULL could be written "!new_ctx"
+	if (new_ctx == NULL) {

ERROR: "foo * bar" should be "foo *bar"
+static const struct proto_ops * tcp_proto_ops(struct sock *sk)

Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/mptcp/protocol.c | 2 +-
 net/mptcp/subflow.c  | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 41d49126d29a..39fdca79ce90 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -26,7 +26,7 @@
 
 static void __mptcp_close(struct sock *sk, long timeout);
 
-static const struct proto_ops * tcp_proto_ops(struct sock *sk)
+static const struct proto_ops *tcp_proto_ops(struct sock *sk)
 {
 #if IS_ENABLED(CONFIG_IPV6)
 	if (sk->sk_family == AF_INET6)
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index 8cfa1d29d59c..1662e1178949 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -592,7 +592,7 @@ void mptcp_handle_ipv6_mapped(struct sock *sk, bool mapped)
 	target = mapped ? &subflow_v6m_specific : subflow_default_af_ops(sk);
 
 	pr_debug("subflow=%p family=%d ops=%p target=%p mapped=%d",
-	         subflow, sk->sk_family, icsk->icsk_af_ops, target, mapped);
+		 subflow, sk->sk_family, icsk->icsk_af_ops, target, mapped);
 
 	if (likely(icsk->icsk_af_ops == target))
 		return;
@@ -773,7 +773,7 @@ static void subflow_ulp_clone(const struct request_sock *req,
 	}
 
 	new_ctx = subflow_create_ctx(newsk, priority);
-	if (new_ctx == NULL) {
+	if (!new_ctx) {
 		subflow_ulp_fallback(newsk, old_ctx);
 		return;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 79ac522402fca97deef3e07be9600bfab20d47a8 Mon Sep 17 00:00:00 2001
From: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Date: Thu, 23 Jan 2020 16:20:02 +0200
Subject: net: atm: use %*ph to print small buffer

Use %*ph format to print small buffer as hex string.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/atm/atm_sysfs.c | 22 ++++++----------
 net/atm/lec.c       | 76 ++++++++++++++---------------------------------------
 2 files changed, 28 insertions(+), 70 deletions(-)

diff --git a/net/atm/atm_sysfs.c b/net/atm/atm_sysfs.c
index 39b94ca5f65d..aa1b57161f3b 100644
--- a/net/atm/atm_sysfs.c
+++ b/net/atm/atm_sysfs.c
@@ -33,23 +33,17 @@ static ssize_t show_atmaddress(struct device *cdev,
 	unsigned long flags;
 	struct atm_dev *adev = to_atm_dev(cdev);
 	struct atm_dev_addr *aaddr;
-	int bin[] = { 1, 2, 10, 6, 1 }, *fmt = bin;
-	int i, j, count = 0;
+	int count = 0;
 
 	spin_lock_irqsave(&adev->lock, flags);
 	list_for_each_entry(aaddr, &adev->local, entry) {
-		for (i = 0, j = 0; i < ATM_ESA_LEN; ++i, ++j) {
-			if (j == *fmt) {
-				count += scnprintf(buf + count,
-						   PAGE_SIZE - count, ".");
-				++fmt;
-				j = 0;
-			}
-			count += scnprintf(buf + count,
-					   PAGE_SIZE - count, "%02x",
-					   aaddr->addr.sas_addr.prv[i]);
-		}
-		count += scnprintf(buf + count, PAGE_SIZE - count, "\n");
+		count += scnprintf(buf + count, PAGE_SIZE - count,
+				   "%1phN.%2phN.%10phN.%6phN.%1phN\n",
+				   &aaddr->addr.sas_addr.prv[0],
+				   &aaddr->addr.sas_addr.prv[1],
+				   &aaddr->addr.sas_addr.prv[3],
+				   &aaddr->addr.sas_addr.prv[13],
+				   &aaddr->addr.sas_addr.prv[19]);
 	}
 	spin_unlock_irqrestore(&adev->lock, flags);
 
diff --git a/net/atm/lec.c b/net/atm/lec.c
index b57368e70aab..25fa3a7b72bd 100644
--- a/net/atm/lec.c
+++ b/net/atm/lec.c
@@ -799,14 +799,9 @@ static const char *lec_arp_get_status_string(unsigned char status)
 
 static void lec_info(struct seq_file *seq, struct lec_arp_table *entry)
 {
-	int i;
-
-	for (i = 0; i < ETH_ALEN; i++)
-		seq_printf(seq, "%2.2x", entry->mac_addr[i] & 0xff);
-	seq_printf(seq, " ");
-	for (i = 0; i < ATM_ESA_LEN; i++)
-		seq_printf(seq, "%2.2x", entry->atm_addr[i] & 0xff);
-	seq_printf(seq, " %s %4.4x", lec_arp_get_status_string(entry->status),
+	seq_printf(seq, "%pM ", entry->mac_addr);
+	seq_printf(seq, "%*phN ", ATM_ESA_LEN, entry->atm_addr);
+	seq_printf(seq, "%s %4.4x", lec_arp_get_status_string(entry->status),
 		   entry->flags & 0xffff);
 	if (entry->vcc)
 		seq_printf(seq, "%3d %3d ", entry->vcc->vpi, entry->vcc->vci);
@@ -1354,7 +1349,7 @@ static void dump_arp_table(struct lec_priv *priv)
 {
 	struct lec_arp_table *rulla;
 	char buf[256];
-	int i, j, offset;
+	int i, offset;
 
 	pr_info("Dump %p:\n", priv);
 	for (i = 0; i < LEC_ARP_TABLE_SIZE; i++) {
@@ -1362,14 +1357,10 @@ static void dump_arp_table(struct lec_priv *priv)
 				     &priv->lec_arp_tables[i], next) {
 			offset = 0;
 			offset += sprintf(buf, "%d: %p\n", i, rulla);
-			offset += sprintf(buf + offset, "Mac: %pM",
+			offset += sprintf(buf + offset, "Mac: %pM ",
 					  rulla->mac_addr);
-			offset += sprintf(buf + offset, " Atm:");
-			for (j = 0; j < ATM_ESA_LEN; j++) {
-				offset += sprintf(buf + offset,
-						  "%2.2x ",
-						  rulla->atm_addr[j] & 0xff);
-			}
+			offset += sprintf(buf + offset, "Atm: %*ph ", ATM_ESA_LEN,
+					  rulla->atm_addr);
 			offset += sprintf(buf + offset,
 					  "Vcc vpi:%d vci:%d, Recv_vcc vpi:%d vci:%d Last_used:%lx, Timestamp:%lx, No_tries:%d ",
 					  rulla->vcc ? rulla->vcc->vpi : 0,
@@ -1392,12 +1383,9 @@ static void dump_arp_table(struct lec_priv *priv)
 		pr_info("No forward\n");
 	hlist_for_each_entry(rulla, &priv->lec_no_forward, next) {
 		offset = 0;
-		offset += sprintf(buf + offset, "Mac: %pM", rulla->mac_addr);
-		offset += sprintf(buf + offset, " Atm:");
-		for (j = 0; j < ATM_ESA_LEN; j++) {
-			offset += sprintf(buf + offset, "%2.2x ",
-					  rulla->atm_addr[j] & 0xff);
-		}
+		offset += sprintf(buf + offset, "Mac: %pM ", rulla->mac_addr);
+		offset += sprintf(buf + offset, "Atm: %*ph ", ATM_ESA_LEN,
+				  rulla->atm_addr);
 		offset += sprintf(buf + offset,
 				  "Vcc vpi:%d vci:%d, Recv_vcc vpi:%d vci:%d Last_used:%lx, Timestamp:%lx, No_tries:%d ",
 				  rulla->vcc ? rulla->vcc->vpi : 0,
@@ -1417,12 +1405,9 @@ static void dump_arp_table(struct lec_priv *priv)
 		pr_info("Empty ones\n");
 	hlist_for_each_entry(rulla, &priv->lec_arp_empty_ones, next) {
 		offset = 0;
-		offset += sprintf(buf + offset, "Mac: %pM", rulla->mac_addr);
-		offset += sprintf(buf + offset, " Atm:");
-		for (j = 0; j < ATM_ESA_LEN; j++) {
-			offset += sprintf(buf + offset, "%2.2x ",
-					  rulla->atm_addr[j] & 0xff);
-		}
+		offset += sprintf(buf + offset, "Mac: %pM ", rulla->mac_addr);
+		offset += sprintf(buf + offset, "Atm: %*ph ", ATM_ESA_LEN,
+				  rulla->atm_addr);
 		offset += sprintf(buf + offset,
 				  "Vcc vpi:%d vci:%d, Recv_vcc vpi:%d vci:%d Last_used:%lx, Timestamp:%lx, No_tries:%d ",
 				  rulla->vcc ? rulla->vcc->vpi : 0,
@@ -1442,12 +1427,9 @@ static void dump_arp_table(struct lec_priv *priv)
 		pr_info("Multicast Forward VCCs\n");
 	hlist_for_each_entry(rulla, &priv->mcast_fwds, next) {
 		offset = 0;
-		offset += sprintf(buf + offset, "Mac: %pM", rulla->mac_addr);
-		offset += sprintf(buf + offset, " Atm:");
-		for (j = 0; j < ATM_ESA_LEN; j++) {
-			offset += sprintf(buf + offset, "%2.2x ",
-					  rulla->atm_addr[j] & 0xff);
-		}
+		offset += sprintf(buf + offset, "Mac: %pM ", rulla->mac_addr);
+		offset += sprintf(buf + offset, "Atm: %*ph ", ATM_ESA_LEN,
+				  rulla->atm_addr);
 		offset += sprintf(buf + offset,
 				  "Vcc vpi:%d vci:%d, Recv_vcc vpi:%d vci:%d Last_used:%lx, Timestamp:%lx, No_tries:%d ",
 				  rulla->vcc ? rulla->vcc->vpi : 0,
@@ -1973,17 +1955,8 @@ lec_vcc_added(struct lec_priv *priv, const struct atmlec_ioc *ioc_data,
 		 * Vcc which we don't want to make default vcc,
 		 * attach it anyway.
 		 */
-		pr_debug("LEC_ARP:Attaching data direct, not default: %2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x\n",
-			 ioc_data->atm_addr[0], ioc_data->atm_addr[1],
-			 ioc_data->atm_addr[2], ioc_data->atm_addr[3],
-			 ioc_data->atm_addr[4], ioc_data->atm_addr[5],
-			 ioc_data->atm_addr[6], ioc_data->atm_addr[7],
-			 ioc_data->atm_addr[8], ioc_data->atm_addr[9],
-			 ioc_data->atm_addr[10], ioc_data->atm_addr[11],
-			 ioc_data->atm_addr[12], ioc_data->atm_addr[13],
-			 ioc_data->atm_addr[14], ioc_data->atm_addr[15],
-			 ioc_data->atm_addr[16], ioc_data->atm_addr[17],
-			 ioc_data->atm_addr[18], ioc_data->atm_addr[19]);
+		pr_debug("LEC_ARP:Attaching data direct, not default: %*phN\n",
+			 ATM_ESA_LEN, ioc_data->atm_addr);
 		entry = make_entry(priv, bus_mac);
 		if (entry == NULL)
 			goto out;
@@ -1999,17 +1972,8 @@ lec_vcc_added(struct lec_priv *priv, const struct atmlec_ioc *ioc_data,
 		dump_arp_table(priv);
 		goto out;
 	}
-	pr_debug("LEC_ARP:Attaching data direct, default: %2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x%2.2x\n",
-		 ioc_data->atm_addr[0], ioc_data->atm_addr[1],
-		 ioc_data->atm_addr[2], ioc_data->atm_addr[3],
-		 ioc_data->atm_addr[4], ioc_data->atm_addr[5],
-		 ioc_data->atm_addr[6], ioc_data->atm_addr[7],
-		 ioc_data->atm_addr[8], ioc_data->atm_addr[9],
-		 ioc_data->atm_addr[10], ioc_data->atm_addr[11],
-		 ioc_data->atm_addr[12], ioc_data->atm_addr[13],
-		 ioc_data->atm_addr[14], ioc_data->atm_addr[15],
-		 ioc_data->atm_addr[16], ioc_data->atm_addr[17],
-		 ioc_data->atm_addr[18], ioc_data->atm_addr[19]);
+	pr_debug("LEC_ARP:Attaching data direct, default: %*phN\n",
+		 ATM_ESA_LEN, ioc_data->atm_addr);
 	for (i = 0; i < LEC_ARP_TABLE_SIZE; i++) {
 		hlist_for_each_entry(entry,
 				     &priv->lec_arp_tables[i], next) {
-- 
cgit v1.2.3-59-g8ed1b


From 53c677009581ac5dbe46c4cf667e408ebfe3049a Mon Sep 17 00:00:00 2001
From: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Date: Thu, 23 Jan 2020 16:27:29 +0200
Subject: net: fddi: skfp: Use print_hex_dump() helper

Use the print_hex_dump() helper, instead of open-coding the same operations.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/fddi/skfp/skfddi.c | 16 +---------------
 1 file changed, 1 insertion(+), 15 deletions(-)

diff --git a/drivers/net/fddi/skfp/skfddi.c b/drivers/net/fddi/skfp/skfddi.c
index 15754451cb87..69c29a2ef95d 100644
--- a/drivers/net/fddi/skfp/skfddi.c
+++ b/drivers/net/fddi/skfp/skfddi.c
@@ -1520,22 +1520,8 @@ void mac_drv_tx_complete(struct s_smc *smc, volatile struct s_smt_fp_txd *txd)
 #ifdef DUMPPACKETS
 void dump_data(unsigned char *Data, int length)
 {
-	int i, j;
-	unsigned char s[255], sh[10];
-	if (length > 64) {
-		length = 64;
-	}
 	printk(KERN_INFO "---Packet start---\n");
-	for (i = 0, j = 0; i < length / 8; i++, j += 8)
-		printk(KERN_INFO "%02x %02x %02x %02x %02x %02x %02x %02x\n",
-		       Data[j + 0], Data[j + 1], Data[j + 2], Data[j + 3],
-		       Data[j + 4], Data[j + 5], Data[j + 6], Data[j + 7]);
-	strcpy(s, "");
-	for (i = 0; i < length % 8; i++) {
-		sprintf(sh, "%02x ", Data[j + i]);
-		strcat(s, sh);
-	}
-	printk(KERN_INFO "%s\n", s);
+	print_hex_dump(KERN_INFO, "", DUMP_PREFIX_NONE, 16, 1, Data, min_t(size_t, length, 64), false);
 	printk(KERN_INFO "------------------\n");
 }				// dump_data
 #else
-- 
cgit v1.2.3-59-g8ed1b


From 6ec8b6cd79a4360e375da99d848d63f8d4fb08b3 Mon Sep 17 00:00:00 2001
From: Moshe Shemesh <moshe@mellanox.com>
Date: Thu, 23 Jan 2020 19:57:13 +0200
Subject: devlink: Add health recover notifications on devlink flows

Devlink health recover notifications were added only on driver direct
updates of health_state through devlink_health_reporter_state_update().
Add notifications on updates of health_state by devlink flows of report
and recover.

Moved functions devlink_nl_health_reporter_fill() and
devlink_recover_notify() to avoid forward declaration.

Fixes: 97ff3bd37fac ("devlink: add devink notification when reporter update health state")
Signed-off-by: Moshe Shemesh <moshe@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/devlink.c | 176 +++++++++++++++++++++++++++--------------------------
 1 file changed, 89 insertions(+), 87 deletions(-)

diff --git a/net/core/devlink.c b/net/core/devlink.c
index 64367eeb21e6..ca1df0ec3c97 100644
--- a/net/core/devlink.c
+++ b/net/core/devlink.c
@@ -4843,6 +4843,93 @@ devlink_health_reporter_destroy(struct devlink_health_reporter *reporter)
 }
 EXPORT_SYMBOL_GPL(devlink_health_reporter_destroy);
 
+static int
+devlink_nl_health_reporter_fill(struct sk_buff *msg,
+				struct devlink *devlink,
+				struct devlink_health_reporter *reporter,
+				enum devlink_command cmd, u32 portid,
+				u32 seq, int flags)
+{
+	struct nlattr *reporter_attr;
+	void *hdr;
+
+	hdr = genlmsg_put(msg, portid, seq, &devlink_nl_family, flags, cmd);
+	if (!hdr)
+		return -EMSGSIZE;
+
+	if (devlink_nl_put_handle(msg, devlink))
+		goto genlmsg_cancel;
+
+	reporter_attr = nla_nest_start_noflag(msg,
+					      DEVLINK_ATTR_HEALTH_REPORTER);
+	if (!reporter_attr)
+		goto genlmsg_cancel;
+	if (nla_put_string(msg, DEVLINK_ATTR_HEALTH_REPORTER_NAME,
+			   reporter->ops->name))
+		goto reporter_nest_cancel;
+	if (nla_put_u8(msg, DEVLINK_ATTR_HEALTH_REPORTER_STATE,
+		       reporter->health_state))
+		goto reporter_nest_cancel;
+	if (nla_put_u64_64bit(msg, DEVLINK_ATTR_HEALTH_REPORTER_ERR_COUNT,
+			      reporter->error_count, DEVLINK_ATTR_PAD))
+		goto reporter_nest_cancel;
+	if (nla_put_u64_64bit(msg, DEVLINK_ATTR_HEALTH_REPORTER_RECOVER_COUNT,
+			      reporter->recovery_count, DEVLINK_ATTR_PAD))
+		goto reporter_nest_cancel;
+	if (reporter->ops->recover &&
+	    nla_put_u64_64bit(msg, DEVLINK_ATTR_HEALTH_REPORTER_GRACEFUL_PERIOD,
+			      reporter->graceful_period,
+			      DEVLINK_ATTR_PAD))
+		goto reporter_nest_cancel;
+	if (reporter->ops->recover &&
+	    nla_put_u8(msg, DEVLINK_ATTR_HEALTH_REPORTER_AUTO_RECOVER,
+		       reporter->auto_recover))
+		goto reporter_nest_cancel;
+	if (reporter->dump_fmsg &&
+	    nla_put_u64_64bit(msg, DEVLINK_ATTR_HEALTH_REPORTER_DUMP_TS,
+			      jiffies_to_msecs(reporter->dump_ts),
+			      DEVLINK_ATTR_PAD))
+		goto reporter_nest_cancel;
+	if (reporter->dump_fmsg &&
+	    nla_put_u64_64bit(msg, DEVLINK_ATTR_HEALTH_REPORTER_DUMP_TS_NS,
+			      reporter->dump_real_ts, DEVLINK_ATTR_PAD))
+		goto reporter_nest_cancel;
+
+	nla_nest_end(msg, reporter_attr);
+	genlmsg_end(msg, hdr);
+	return 0;
+
+reporter_nest_cancel:
+	nla_nest_end(msg, reporter_attr);
+genlmsg_cancel:
+	genlmsg_cancel(msg, hdr);
+	return -EMSGSIZE;
+}
+
+static void devlink_recover_notify(struct devlink_health_reporter *reporter,
+				   enum devlink_command cmd)
+{
+	struct sk_buff *msg;
+	int err;
+
+	WARN_ON(cmd != DEVLINK_CMD_HEALTH_REPORTER_RECOVER);
+
+	msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
+	if (!msg)
+		return;
+
+	err = devlink_nl_health_reporter_fill(msg, reporter->devlink,
+					      reporter, cmd, 0, 0, 0);
+	if (err) {
+		nlmsg_free(msg);
+		return;
+	}
+
+	genlmsg_multicast_netns(&devlink_nl_family,
+				devlink_net(reporter->devlink),
+				msg, 0, DEVLINK_MCGRP_CONFIG, GFP_KERNEL);
+}
+
 void
 devlink_health_reporter_recovery_done(struct devlink_health_reporter *reporter)
 {
@@ -4869,6 +4956,7 @@ devlink_health_reporter_recover(struct devlink_health_reporter *reporter,
 
 	devlink_health_reporter_recovery_done(reporter);
 	reporter->health_state = DEVLINK_HEALTH_REPORTER_STATE_HEALTHY;
+	devlink_recover_notify(reporter, DEVLINK_CMD_HEALTH_REPORTER_RECOVER);
 
 	return 0;
 }
@@ -4935,6 +5023,7 @@ int devlink_health_report(struct devlink_health_reporter *reporter,
 	reporter->error_count++;
 	prev_health_state = reporter->health_state;
 	reporter->health_state = DEVLINK_HEALTH_REPORTER_STATE_ERROR;
+	devlink_recover_notify(reporter, DEVLINK_CMD_HEALTH_REPORTER_RECOVER);
 
 	/* abort if the previous error wasn't recovered */
 	if (reporter->auto_recover &&
@@ -5017,93 +5106,6 @@ devlink_health_reporter_put(struct devlink_health_reporter *reporter)
 	refcount_dec(&reporter->refcount);
 }
 
-static int
-devlink_nl_health_reporter_fill(struct sk_buff *msg,
-				struct devlink *devlink,
-				struct devlink_health_reporter *reporter,
-				enum devlink_command cmd, u32 portid,
-				u32 seq, int flags)
-{
-	struct nlattr *reporter_attr;
-	void *hdr;
-
-	hdr = genlmsg_put(msg, portid, seq, &devlink_nl_family, flags, cmd);
-	if (!hdr)
-		return -EMSGSIZE;
-
-	if (devlink_nl_put_handle(msg, devlink))
-		goto genlmsg_cancel;
-
-	reporter_attr = nla_nest_start_noflag(msg,
-					      DEVLINK_ATTR_HEALTH_REPORTER);
-	if (!reporter_attr)
-		goto genlmsg_cancel;
-	if (nla_put_string(msg, DEVLINK_ATTR_HEALTH_REPORTER_NAME,
-			   reporter->ops->name))
-		goto reporter_nest_cancel;
-	if (nla_put_u8(msg, DEVLINK_ATTR_HEALTH_REPORTER_STATE,
-		       reporter->health_state))
-		goto reporter_nest_cancel;
-	if (nla_put_u64_64bit(msg, DEVLINK_ATTR_HEALTH_REPORTER_ERR_COUNT,
-			      reporter->error_count, DEVLINK_ATTR_PAD))
-		goto reporter_nest_cancel;
-	if (nla_put_u64_64bit(msg, DEVLINK_ATTR_HEALTH_REPORTER_RECOVER_COUNT,
-			      reporter->recovery_count, DEVLINK_ATTR_PAD))
-		goto reporter_nest_cancel;
-	if (reporter->ops->recover &&
-	    nla_put_u64_64bit(msg, DEVLINK_ATTR_HEALTH_REPORTER_GRACEFUL_PERIOD,
-			      reporter->graceful_period,
-			      DEVLINK_ATTR_PAD))
-		goto reporter_nest_cancel;
-	if (reporter->ops->recover &&
-	    nla_put_u8(msg, DEVLINK_ATTR_HEALTH_REPORTER_AUTO_RECOVER,
-		       reporter->auto_recover))
-		goto reporter_nest_cancel;
-	if (reporter->dump_fmsg &&
-	    nla_put_u64_64bit(msg, DEVLINK_ATTR_HEALTH_REPORTER_DUMP_TS,
-			      jiffies_to_msecs(reporter->dump_ts),
-			      DEVLINK_ATTR_PAD))
-		goto reporter_nest_cancel;
-	if (reporter->dump_fmsg &&
-	    nla_put_u64_64bit(msg, DEVLINK_ATTR_HEALTH_REPORTER_DUMP_TS_NS,
-			      reporter->dump_real_ts, DEVLINK_ATTR_PAD))
-		goto reporter_nest_cancel;
-
-	nla_nest_end(msg, reporter_attr);
-	genlmsg_end(msg, hdr);
-	return 0;
-
-reporter_nest_cancel:
-	nla_nest_end(msg, reporter_attr);
-genlmsg_cancel:
-	genlmsg_cancel(msg, hdr);
-	return -EMSGSIZE;
-}
-
-static void devlink_recover_notify(struct devlink_health_reporter *reporter,
-				   enum devlink_command cmd)
-{
-	struct sk_buff *msg;
-	int err;
-
-	WARN_ON(cmd != DEVLINK_CMD_HEALTH_REPORTER_RECOVER);
-
-	msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
-	if (!msg)
-		return;
-
-	err = devlink_nl_health_reporter_fill(msg, reporter->devlink,
-					      reporter, cmd, 0, 0, 0);
-	if (err) {
-		nlmsg_free(msg);
-		return;
-	}
-
-	genlmsg_multicast_netns(&devlink_nl_family,
-				devlink_net(reporter->devlink),
-				msg, 0, DEVLINK_MCGRP_CONFIG, GFP_KERNEL);
-}
-
 void
 devlink_health_reporter_state_update(struct devlink_health_reporter *reporter,
 				     enum devlink_health_reporter_state state)
-- 
cgit v1.2.3-59-g8ed1b


From 351e1581395fcc7fb952bbd7dda01238f69968fd Mon Sep 17 00:00:00 2001
From: Haiyang Zhang <haiyangz@microsoft.com>
Date: Thu, 23 Jan 2020 13:52:34 -0800
Subject: hv_netvsc: Add XDP support

This patch adds support of XDP in native mode for hv_netvsc driver, and
transparently sets the XDP program on the associated VF NIC as well.

Setting / unsetting XDP program on synthetic NIC (netvsc) propagates to
VF NIC automatically. Setting / unsetting XDP program on VF NIC directly
is not recommended, also not propagated to synthetic NIC, and may be
overwritten by setting of synthetic NIC.

The Azure/Hyper-V synthetic NIC receive buffer doesn't provide headroom
for XDP. We thought about re-use the RNDIS header space, but it's too
small. So we decided to copy the packets to a page buffer for XDP. And,
most of our VMs on Azure have Accelerated  Network (SRIOV) enabled, so
most of the packets run on VF NIC. The synthetic NIC is considered as a
fallback data-path. So the data copy on netvsc won't impact performance
significantly.

XDP program cannot run with LRO (RSC) enabled, so you need to disable LRO
before running XDP:
        ethtool -K eth0 lro off

XDP actions not yet supported:
        XDP_REDIRECT

Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/hyperv/Makefile       |   2 +-
 drivers/net/hyperv/hyperv_net.h   |  21 +++-
 drivers/net/hyperv/netvsc.c       |  31 +++++-
 drivers/net/hyperv/netvsc_bpf.c   | 209 ++++++++++++++++++++++++++++++++++++++
 drivers/net/hyperv/netvsc_drv.c   | 183 +++++++++++++++++++++++++++------
 drivers/net/hyperv/rndis_filter.c |   2 +-
 6 files changed, 409 insertions(+), 39 deletions(-)
 create mode 100644 drivers/net/hyperv/netvsc_bpf.c

diff --git a/drivers/net/hyperv/Makefile b/drivers/net/hyperv/Makefile
index 3a2aa0708166..0db7ccaec4a4 100644
--- a/drivers/net/hyperv/Makefile
+++ b/drivers/net/hyperv/Makefile
@@ -1,4 +1,4 @@
 # SPDX-License-Identifier: GPL-2.0-only
 obj-$(CONFIG_HYPERV_NET) += hv_netvsc.o
 
-hv_netvsc-y := netvsc_drv.o netvsc.o rndis_filter.o netvsc_trace.o
+hv_netvsc-y := netvsc_drv.o netvsc.o rndis_filter.o netvsc_trace.o netvsc_bpf.o
diff --git a/drivers/net/hyperv/hyperv_net.h b/drivers/net/hyperv/hyperv_net.h
index dc44819946e6..abda736e7c7d 100644
--- a/drivers/net/hyperv/hyperv_net.h
+++ b/drivers/net/hyperv/hyperv_net.h
@@ -142,6 +142,8 @@ struct netvsc_device_info {
 	u32  send_section_size;
 	u32  recv_section_size;
 
+	struct bpf_prog *bprog;
+
 	u8 rss_key[NETVSC_HASH_KEYLEN];
 };
 
@@ -189,7 +191,8 @@ int netvsc_send(struct net_device *net,
 		struct hv_netvsc_packet *packet,
 		struct rndis_message *rndis_msg,
 		struct hv_page_buffer *page_buffer,
-		struct sk_buff *skb);
+		struct sk_buff *skb,
+		bool xdp_tx);
 void netvsc_linkstatus_callback(struct net_device *net,
 				struct rndis_message *resp);
 int netvsc_recv_callback(struct net_device *net,
@@ -198,6 +201,16 @@ int netvsc_recv_callback(struct net_device *net,
 void netvsc_channel_cb(void *context);
 int netvsc_poll(struct napi_struct *napi, int budget);
 
+u32 netvsc_run_xdp(struct net_device *ndev, struct netvsc_channel *nvchan,
+		   struct xdp_buff *xdp);
+unsigned int netvsc_xdp_fraglen(unsigned int len);
+struct bpf_prog *netvsc_xdp_get(struct netvsc_device *nvdev);
+int netvsc_xdp_set(struct net_device *dev, struct bpf_prog *prog,
+		   struct netlink_ext_ack *extack,
+		   struct netvsc_device *nvdev);
+int netvsc_vf_setxdp(struct net_device *vf_netdev, struct bpf_prog *prog);
+int netvsc_bpf(struct net_device *dev, struct netdev_bpf *bpf);
+
 int rndis_set_subchannel(struct net_device *ndev,
 			 struct netvsc_device *nvdev,
 			 struct netvsc_device_info *dev_info);
@@ -832,6 +845,8 @@ struct nvsp_message {
 #define RNDIS_MAX_PKT_DEFAULT 8
 #define RNDIS_PKT_ALIGN_DEFAULT 8
 
+#define NETVSC_XDP_HDRM 256
+
 struct multi_send_data {
 	struct sk_buff *skb; /* skb containing the pkt */
 	struct hv_netvsc_packet *pkt; /* netvsc pkt pending */
@@ -867,6 +882,7 @@ struct netvsc_stats {
 	u64 bytes;
 	u64 broadcast;
 	u64 multicast;
+	u64 xdp_drop;
 	struct u64_stats_sync syncp;
 };
 
@@ -972,6 +988,9 @@ struct netvsc_channel {
 	atomic_t queue_sends;
 	struct nvsc_rsc rsc;
 
+	struct bpf_prog __rcu *bpf_prog;
+	struct xdp_rxq_info xdp_rxq;
+
 	struct netvsc_stats tx_stats;
 	struct netvsc_stats rx_stats;
 };
diff --git a/drivers/net/hyperv/netvsc.c b/drivers/net/hyperv/netvsc.c
index eab83e71567a..ae3f3084c2ed 100644
--- a/drivers/net/hyperv/netvsc.c
+++ b/drivers/net/hyperv/netvsc.c
@@ -122,8 +122,10 @@ static void free_netvsc_device(struct rcu_head *head)
 	vfree(nvdev->send_buf);
 	kfree(nvdev->send_section_map);
 
-	for (i = 0; i < VRSS_CHANNEL_MAX; i++)
+	for (i = 0; i < VRSS_CHANNEL_MAX; i++) {
+		xdp_rxq_info_unreg(&nvdev->chan_table[i].xdp_rxq);
 		vfree(nvdev->chan_table[i].mrc.slots);
+	}
 
 	kfree(nvdev);
 }
@@ -900,7 +902,8 @@ int netvsc_send(struct net_device *ndev,
 		struct hv_netvsc_packet *packet,
 		struct rndis_message *rndis_msg,
 		struct hv_page_buffer *pb,
-		struct sk_buff *skb)
+		struct sk_buff *skb,
+		bool xdp_tx)
 {
 	struct net_device_context *ndev_ctx = netdev_priv(ndev);
 	struct netvsc_device *net_device
@@ -923,10 +926,11 @@ int netvsc_send(struct net_device *ndev,
 	packet->send_buf_index = NETVSC_INVALID_INDEX;
 	packet->cp_partial = false;
 
-	/* Send control message directly without accessing msd (Multi-Send
-	 * Data) field which may be changed during data packet processing.
+	/* Send a control message or XDP packet directly without accessing
+	 * msd (Multi-Send Data) field which may be changed during data packet
+	 * processing.
 	 */
-	if (!skb)
+	if (!skb || xdp_tx)
 		return netvsc_send_pkt(device, packet, net_device, pb, skb);
 
 	/* batch packets in send buffer if possible */
@@ -1392,6 +1396,21 @@ struct netvsc_device *netvsc_device_add(struct hv_device *device,
 		nvchan->net_device = net_device;
 		u64_stats_init(&nvchan->tx_stats.syncp);
 		u64_stats_init(&nvchan->rx_stats.syncp);
+
+		ret = xdp_rxq_info_reg(&nvchan->xdp_rxq, ndev, i);
+
+		if (ret) {
+			netdev_err(ndev, "xdp_rxq_info_reg fail: %d\n", ret);
+			goto cleanup2;
+		}
+
+		ret = xdp_rxq_info_reg_mem_model(&nvchan->xdp_rxq,
+						 MEM_TYPE_PAGE_SHARED, NULL);
+
+		if (ret) {
+			netdev_err(ndev, "xdp reg_mem_model fail: %d\n", ret);
+			goto cleanup2;
+		}
 	}
 
 	/* Enable NAPI handler before init callbacks */
@@ -1437,6 +1456,8 @@ close:
 
 cleanup:
 	netif_napi_del(&net_device->chan_table[0].napi);
+
+cleanup2:
 	free_netvsc_device(&net_device->rcu);
 
 	return ERR_PTR(ret);
diff --git a/drivers/net/hyperv/netvsc_bpf.c b/drivers/net/hyperv/netvsc_bpf.c
new file mode 100644
index 000000000000..20adfe544294
--- /dev/null
+++ b/drivers/net/hyperv/netvsc_bpf.c
@@ -0,0 +1,209 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright (c) 2019, Microsoft Corporation.
+ *
+ * Author:
+ *   Haiyang Zhang <haiyangz@microsoft.com>
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/ethtool.h>
+#include <linux/bpf.h>
+#include <linux/bpf_trace.h>
+#include <linux/kernel.h>
+#include <net/xdp.h>
+
+#include <linux/mutex.h>
+#include <linux/rtnetlink.h>
+
+#include "hyperv_net.h"
+
+u32 netvsc_run_xdp(struct net_device *ndev, struct netvsc_channel *nvchan,
+		   struct xdp_buff *xdp)
+{
+	void *data = nvchan->rsc.data[0];
+	u32 len = nvchan->rsc.len[0];
+	struct page *page = NULL;
+	struct bpf_prog *prog;
+	u32 act = XDP_PASS;
+
+	xdp->data_hard_start = NULL;
+
+	rcu_read_lock();
+	prog = rcu_dereference(nvchan->bpf_prog);
+
+	if (!prog)
+		goto out;
+
+	/* allocate page buffer for data */
+	page = alloc_page(GFP_ATOMIC);
+	if (!page) {
+		act = XDP_DROP;
+		goto out;
+	}
+
+	xdp->data_hard_start = page_address(page);
+	xdp->data = xdp->data_hard_start + NETVSC_XDP_HDRM;
+	xdp_set_data_meta_invalid(xdp);
+	xdp->data_end = xdp->data + len;
+	xdp->rxq = &nvchan->xdp_rxq;
+	xdp->handle = 0;
+
+	memcpy(xdp->data, data, len);
+
+	act = bpf_prog_run_xdp(prog, xdp);
+
+	switch (act) {
+	case XDP_PASS:
+	case XDP_TX:
+	case XDP_DROP:
+		break;
+
+	case XDP_ABORTED:
+		trace_xdp_exception(ndev, prog, act);
+		break;
+
+	default:
+		bpf_warn_invalid_xdp_action(act);
+	}
+
+out:
+	rcu_read_unlock();
+
+	if (page && act != XDP_PASS && act != XDP_TX) {
+		__free_page(page);
+		xdp->data_hard_start = NULL;
+	}
+
+	return act;
+}
+
+unsigned int netvsc_xdp_fraglen(unsigned int len)
+{
+	return SKB_DATA_ALIGN(len) +
+	       SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
+}
+
+struct bpf_prog *netvsc_xdp_get(struct netvsc_device *nvdev)
+{
+	return rtnl_dereference(nvdev->chan_table[0].bpf_prog);
+}
+
+int netvsc_xdp_set(struct net_device *dev, struct bpf_prog *prog,
+		   struct netlink_ext_ack *extack,
+		   struct netvsc_device *nvdev)
+{
+	struct bpf_prog *old_prog;
+	int buf_max, i;
+
+	old_prog = netvsc_xdp_get(nvdev);
+
+	if (!old_prog && !prog)
+		return 0;
+
+	buf_max = NETVSC_XDP_HDRM + netvsc_xdp_fraglen(dev->mtu + ETH_HLEN);
+	if (prog && buf_max > PAGE_SIZE) {
+		netdev_err(dev, "XDP: mtu:%u too large, buf_max:%u\n",
+			   dev->mtu, buf_max);
+		NL_SET_ERR_MSG_MOD(extack, "XDP: mtu too large");
+
+		return -EOPNOTSUPP;
+	}
+
+	if (prog && (dev->features & NETIF_F_LRO)) {
+		netdev_err(dev, "XDP: not support LRO\n");
+		NL_SET_ERR_MSG_MOD(extack, "XDP: not support LRO");
+
+		return -EOPNOTSUPP;
+	}
+
+	if (prog)
+		bpf_prog_add(prog, nvdev->num_chn);
+
+	for (i = 0; i < nvdev->num_chn; i++)
+		rcu_assign_pointer(nvdev->chan_table[i].bpf_prog, prog);
+
+	if (old_prog)
+		for (i = 0; i < nvdev->num_chn; i++)
+			bpf_prog_put(old_prog);
+
+	return 0;
+}
+
+int netvsc_vf_setxdp(struct net_device *vf_netdev, struct bpf_prog *prog)
+{
+	struct netdev_bpf xdp;
+	bpf_op_t ndo_bpf;
+
+	ASSERT_RTNL();
+
+	if (!vf_netdev)
+		return 0;
+
+	ndo_bpf = vf_netdev->netdev_ops->ndo_bpf;
+	if (!ndo_bpf)
+		return 0;
+
+	memset(&xdp, 0, sizeof(xdp));
+
+	xdp.command = XDP_SETUP_PROG;
+	xdp.prog = prog;
+
+	return ndo_bpf(vf_netdev, &xdp);
+}
+
+static u32 netvsc_xdp_query(struct netvsc_device *nvdev)
+{
+	struct bpf_prog *prog = netvsc_xdp_get(nvdev);
+
+	if (prog)
+		return prog->aux->id;
+
+	return 0;
+}
+
+int netvsc_bpf(struct net_device *dev, struct netdev_bpf *bpf)
+{
+	struct net_device_context *ndevctx = netdev_priv(dev);
+	struct netvsc_device *nvdev = rtnl_dereference(ndevctx->nvdev);
+	struct net_device *vf_netdev = rtnl_dereference(ndevctx->vf_netdev);
+	struct netlink_ext_ack *extack = bpf->extack;
+	int ret;
+
+	if (!nvdev || nvdev->destroy) {
+		if (bpf->command == XDP_QUERY_PROG) {
+			bpf->prog_id = 0;
+			return 0; /* Query must always succeed */
+		} else {
+			return -ENODEV;
+		}
+	}
+
+	switch (bpf->command) {
+	case XDP_SETUP_PROG:
+		ret = netvsc_xdp_set(dev, bpf->prog, extack, nvdev);
+
+		if (ret)
+			return ret;
+
+		ret = netvsc_vf_setxdp(vf_netdev, bpf->prog);
+
+		if (ret) {
+			netdev_err(dev, "vf_setxdp failed:%d\n", ret);
+			NL_SET_ERR_MSG_MOD(extack, "vf_setxdp failed");
+
+			netvsc_xdp_set(dev, NULL, extack, nvdev);
+		}
+
+		return ret;
+
+	case XDP_QUERY_PROG:
+		bpf->prog_id = netvsc_xdp_query(nvdev);
+		return 0;
+
+	default:
+		return -EINVAL;
+	}
+}
diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c
index f3f9eb8a402a..8fc71bd49894 100644
--- a/drivers/net/hyperv/netvsc_drv.c
+++ b/drivers/net/hyperv/netvsc_drv.c
@@ -25,6 +25,7 @@
 #include <linux/slab.h>
 #include <linux/rtnetlink.h>
 #include <linux/netpoll.h>
+#include <linux/bpf.h>
 
 #include <net/arp.h>
 #include <net/route.h>
@@ -519,7 +520,7 @@ static int netvsc_vf_xmit(struct net_device *net, struct net_device *vf_netdev,
 	return rc;
 }
 
-static int netvsc_start_xmit(struct sk_buff *skb, struct net_device *net)
+static int netvsc_xmit(struct sk_buff *skb, struct net_device *net, bool xdp_tx)
 {
 	struct net_device_context *net_device_ctx = netdev_priv(net);
 	struct hv_netvsc_packet *packet = NULL;
@@ -686,7 +687,7 @@ static int netvsc_start_xmit(struct sk_buff *skb, struct net_device *net)
 	/* timestamp packet in software */
 	skb_tx_timestamp(skb);
 
-	ret = netvsc_send(net, packet, rndis_msg, pb, skb);
+	ret = netvsc_send(net, packet, rndis_msg, pb, skb, xdp_tx);
 	if (likely(ret == 0))
 		return NETDEV_TX_OK;
 
@@ -709,6 +710,11 @@ no_memory:
 	goto drop;
 }
 
+static int netvsc_start_xmit(struct sk_buff *skb, struct net_device *ndev)
+{
+	return netvsc_xmit(skb, ndev, false);
+}
+
 /*
  * netvsc_linkstatus_callback - Link up/down notification
  */
@@ -751,6 +757,22 @@ void netvsc_linkstatus_callback(struct net_device *net,
 	schedule_delayed_work(&ndev_ctx->dwork, 0);
 }
 
+static void netvsc_xdp_xmit(struct sk_buff *skb, struct net_device *ndev)
+{
+	int rc;
+
+	skb->queue_mapping = skb_get_rx_queue(skb);
+	__skb_push(skb, ETH_HLEN);
+
+	rc = netvsc_xmit(skb, ndev, true);
+
+	if (dev_xmit_complete(rc))
+		return;
+
+	dev_kfree_skb_any(skb);
+	ndev->stats.tx_dropped++;
+}
+
 static void netvsc_comp_ipcsum(struct sk_buff *skb)
 {
 	struct iphdr *iph = (struct iphdr *)skb->data;
@@ -760,7 +782,8 @@ static void netvsc_comp_ipcsum(struct sk_buff *skb)
 }
 
 static struct sk_buff *netvsc_alloc_recv_skb(struct net_device *net,
-					     struct netvsc_channel *nvchan)
+					     struct netvsc_channel *nvchan,
+					     struct xdp_buff *xdp)
 {
 	struct napi_struct *napi = &nvchan->napi;
 	const struct ndis_pkt_8021q_info *vlan = nvchan->rsc.vlan;
@@ -768,18 +791,37 @@ static struct sk_buff *netvsc_alloc_recv_skb(struct net_device *net,
 						nvchan->rsc.csum_info;
 	const u32 *hash_info = nvchan->rsc.hash_info;
 	struct sk_buff *skb;
+	void *xbuf = xdp->data_hard_start;
 	int i;
 
-	skb = napi_alloc_skb(napi, nvchan->rsc.pktlen);
-	if (!skb)
-		return skb;
+	if (xbuf) {
+		unsigned int hdroom = xdp->data - xdp->data_hard_start;
+		unsigned int xlen = xdp->data_end - xdp->data;
+		unsigned int frag_size = netvsc_xdp_fraglen(hdroom + xlen);
 
-	/*
-	 * Copy to skb. This copy is needed here since the memory pointed by
-	 * hv_netvsc_packet cannot be deallocated
-	 */
-	for (i = 0; i < nvchan->rsc.cnt; i++)
-		skb_put_data(skb, nvchan->rsc.data[i], nvchan->rsc.len[i]);
+		skb = build_skb(xbuf, frag_size);
+
+		if (!skb) {
+			__free_page(virt_to_page(xbuf));
+			return NULL;
+		}
+
+		skb_reserve(skb, hdroom);
+		skb_put(skb, xlen);
+		skb->dev = napi->dev;
+	} else {
+		skb = napi_alloc_skb(napi, nvchan->rsc.pktlen);
+
+		if (!skb)
+			return NULL;
+
+		/* Copy to skb. This copy is needed here since the memory
+		 * pointed by hv_netvsc_packet cannot be deallocated.
+		 */
+		for (i = 0; i < nvchan->rsc.cnt; i++)
+			skb_put_data(skb, nvchan->rsc.data[i],
+				     nvchan->rsc.len[i]);
+	}
 
 	skb->protocol = eth_type_trans(skb, net);
 
@@ -829,13 +871,25 @@ int netvsc_recv_callback(struct net_device *net,
 	struct vmbus_channel *channel = nvchan->channel;
 	u16 q_idx = channel->offermsg.offer.sub_channel_index;
 	struct sk_buff *skb;
-	struct netvsc_stats *rx_stats;
+	struct netvsc_stats *rx_stats = &nvchan->rx_stats;
+	struct xdp_buff xdp;
+	u32 act;
 
 	if (net->reg_state != NETREG_REGISTERED)
 		return NVSP_STAT_FAIL;
 
+	act = netvsc_run_xdp(net, nvchan, &xdp);
+
+	if (act != XDP_PASS && act != XDP_TX) {
+		u64_stats_update_begin(&rx_stats->syncp);
+		rx_stats->xdp_drop++;
+		u64_stats_update_end(&rx_stats->syncp);
+
+		return NVSP_STAT_SUCCESS; /* consumed by XDP */
+	}
+
 	/* Allocate a skb - TODO direct I/O to pages? */
-	skb = netvsc_alloc_recv_skb(net, nvchan);
+	skb = netvsc_alloc_recv_skb(net, nvchan, &xdp);
 
 	if (unlikely(!skb)) {
 		++net_device_ctx->eth_stats.rx_no_memory;
@@ -849,7 +903,6 @@ int netvsc_recv_callback(struct net_device *net,
 	 * on the synthetic device because modifying the VF device
 	 * statistics will not work correctly.
 	 */
-	rx_stats = &nvchan->rx_stats;
 	u64_stats_update_begin(&rx_stats->syncp);
 	rx_stats->packets++;
 	rx_stats->bytes += nvchan->rsc.pktlen;
@@ -860,6 +913,11 @@ int netvsc_recv_callback(struct net_device *net,
 		++rx_stats->multicast;
 	u64_stats_update_end(&rx_stats->syncp);
 
+	if (act == XDP_TX) {
+		netvsc_xdp_xmit(skb, net);
+		return NVSP_STAT_SUCCESS;
+	}
+
 	napi_gro_receive(&nvchan->napi, skb);
 	return NVSP_STAT_SUCCESS;
 }
@@ -886,10 +944,11 @@ static void netvsc_get_channels(struct net_device *net,
 /* Alloc struct netvsc_device_info, and initialize it from either existing
  * struct netvsc_device, or from default values.
  */
-static struct netvsc_device_info *netvsc_devinfo_get
-			(struct netvsc_device *nvdev)
+static
+struct netvsc_device_info *netvsc_devinfo_get(struct netvsc_device *nvdev)
 {
 	struct netvsc_device_info *dev_info;
+	struct bpf_prog *prog;
 
 	dev_info = kzalloc(sizeof(*dev_info), GFP_ATOMIC);
 
@@ -897,6 +956,8 @@ static struct netvsc_device_info *netvsc_devinfo_get
 		return NULL;
 
 	if (nvdev) {
+		ASSERT_RTNL();
+
 		dev_info->num_chn = nvdev->num_chn;
 		dev_info->send_sections = nvdev->send_section_cnt;
 		dev_info->send_section_size = nvdev->send_section_size;
@@ -905,6 +966,12 @@ static struct netvsc_device_info *netvsc_devinfo_get
 
 		memcpy(dev_info->rss_key, nvdev->extension->rss_key,
 		       NETVSC_HASH_KEYLEN);
+
+		prog = netvsc_xdp_get(nvdev);
+		if (prog) {
+			bpf_prog_inc(prog);
+			dev_info->bprog = prog;
+		}
 	} else {
 		dev_info->num_chn = VRSS_CHANNEL_DEFAULT;
 		dev_info->send_sections = NETVSC_DEFAULT_TX;
@@ -916,6 +983,17 @@ static struct netvsc_device_info *netvsc_devinfo_get
 	return dev_info;
 }
 
+/* Free struct netvsc_device_info */
+static void netvsc_devinfo_put(struct netvsc_device_info *dev_info)
+{
+	if (dev_info->bprog) {
+		ASSERT_RTNL();
+		bpf_prog_put(dev_info->bprog);
+	}
+
+	kfree(dev_info);
+}
+
 static int netvsc_detach(struct net_device *ndev,
 			 struct netvsc_device *nvdev)
 {
@@ -927,6 +1005,8 @@ static int netvsc_detach(struct net_device *ndev,
 	if (cancel_work_sync(&nvdev->subchan_work))
 		nvdev->num_chn = 1;
 
+	netvsc_xdp_set(ndev, NULL, NULL, nvdev);
+
 	/* If device was up (receiving) then shutdown */
 	if (netif_running(ndev)) {
 		netvsc_tx_disable(nvdev, ndev);
@@ -960,7 +1040,8 @@ static int netvsc_attach(struct net_device *ndev,
 	struct hv_device *hdev = ndev_ctx->device_ctx;
 	struct netvsc_device *nvdev;
 	struct rndis_device *rdev;
-	int ret;
+	struct bpf_prog *prog;
+	int ret = 0;
 
 	nvdev = rndis_filter_device_add(hdev, dev_info);
 	if (IS_ERR(nvdev))
@@ -976,6 +1057,13 @@ static int netvsc_attach(struct net_device *ndev,
 		}
 	}
 
+	prog = dev_info->bprog;
+	if (prog) {
+		ret = netvsc_xdp_set(ndev, prog, NULL, nvdev);
+		if (ret)
+			goto err1;
+	}
+
 	/* In any case device is now ready */
 	netif_device_attach(ndev);
 
@@ -985,7 +1073,7 @@ static int netvsc_attach(struct net_device *ndev,
 	if (netif_running(ndev)) {
 		ret = rndis_filter_open(nvdev);
 		if (ret)
-			goto err;
+			goto err2;
 
 		rdev = nvdev->extension;
 		if (!rdev->link_state)
@@ -994,9 +1082,10 @@ static int netvsc_attach(struct net_device *ndev,
 
 	return 0;
 
-err:
+err2:
 	netif_device_detach(ndev);
 
+err1:
 	rndis_filter_device_remove(hdev, nvdev);
 
 	return ret;
@@ -1046,7 +1135,7 @@ static int netvsc_set_channels(struct net_device *net,
 	}
 
 out:
-	kfree(device_info);
+	netvsc_devinfo_put(device_info);
 	return ret;
 }
 
@@ -1153,7 +1242,7 @@ rollback_vf:
 		dev_set_mtu(vf_netdev, orig_mtu);
 
 out:
-	kfree(device_info);
+	netvsc_devinfo_put(device_info);
 	return ret;
 }
 
@@ -1378,8 +1467,8 @@ static const struct {
 /* statistics per queue (rx/tx packets/bytes) */
 #define NETVSC_PCPU_STATS_LEN (num_present_cpus() * ARRAY_SIZE(pcpu_stats))
 
-/* 4 statistics per queue (rx/tx packets/bytes) */
-#define NETVSC_QUEUE_STATS_LEN(dev) ((dev)->num_chn * 4)
+/* 5 statistics per queue (rx/tx packets/bytes, rx xdp_drop) */
+#define NETVSC_QUEUE_STATS_LEN(dev) ((dev)->num_chn * 5)
 
 static int netvsc_get_sset_count(struct net_device *dev, int string_set)
 {
@@ -1411,6 +1500,7 @@ static void netvsc_get_ethtool_stats(struct net_device *dev,
 	struct netvsc_ethtool_pcpu_stats *pcpu_sum;
 	unsigned int start;
 	u64 packets, bytes;
+	u64 xdp_drop;
 	int i, j, cpu;
 
 	if (!nvdev)
@@ -1439,9 +1529,11 @@ static void netvsc_get_ethtool_stats(struct net_device *dev,
 			start = u64_stats_fetch_begin_irq(&qstats->syncp);
 			packets = qstats->packets;
 			bytes = qstats->bytes;
+			xdp_drop = qstats->xdp_drop;
 		} while (u64_stats_fetch_retry_irq(&qstats->syncp, start));
 		data[i++] = packets;
 		data[i++] = bytes;
+		data[i++] = xdp_drop;
 	}
 
 	pcpu_sum = kvmalloc_array(num_possible_cpus(),
@@ -1489,6 +1581,8 @@ static void netvsc_get_strings(struct net_device *dev, u32 stringset, u8 *data)
 			p += ETH_GSTRING_LEN;
 			sprintf(p, "rx_queue_%u_bytes", i);
 			p += ETH_GSTRING_LEN;
+			sprintf(p, "rx_queue_%u_xdp_drop", i);
+			p += ETH_GSTRING_LEN;
 		}
 
 		for_each_present_cpu(cpu) {
@@ -1785,10 +1879,27 @@ static int netvsc_set_ringparam(struct net_device *ndev,
 	}
 
 out:
-	kfree(device_info);
+	netvsc_devinfo_put(device_info);
 	return ret;
 }
 
+static netdev_features_t netvsc_fix_features(struct net_device *ndev,
+					     netdev_features_t features)
+{
+	struct net_device_context *ndevctx = netdev_priv(ndev);
+	struct netvsc_device *nvdev = rtnl_dereference(ndevctx->nvdev);
+
+	if (!nvdev || nvdev->destroy)
+		return features;
+
+	if ((features & NETIF_F_LRO) && netvsc_xdp_get(nvdev)) {
+		features ^= NETIF_F_LRO;
+		netdev_info(ndev, "Skip LRO - unsupported with XDP\n");
+	}
+
+	return features;
+}
+
 static int netvsc_set_features(struct net_device *ndev,
 			       netdev_features_t features)
 {
@@ -1875,12 +1986,14 @@ static const struct net_device_ops device_ops = {
 	.ndo_start_xmit =		netvsc_start_xmit,
 	.ndo_change_rx_flags =		netvsc_change_rx_flags,
 	.ndo_set_rx_mode =		netvsc_set_rx_mode,
+	.ndo_fix_features =		netvsc_fix_features,
 	.ndo_set_features =		netvsc_set_features,
 	.ndo_change_mtu =		netvsc_change_mtu,
 	.ndo_validate_addr =		eth_validate_addr,
 	.ndo_set_mac_address =		netvsc_set_mac_addr,
 	.ndo_select_queue =		netvsc_select_queue,
 	.ndo_get_stats64 =		netvsc_get_stats64,
+	.ndo_bpf =			netvsc_bpf,
 };
 
 /*
@@ -2167,6 +2280,7 @@ static int netvsc_register_vf(struct net_device *vf_netdev)
 {
 	struct net_device_context *net_device_ctx;
 	struct netvsc_device *netvsc_dev;
+	struct bpf_prog *prog;
 	struct net_device *ndev;
 	int ret;
 
@@ -2211,6 +2325,9 @@ static int netvsc_register_vf(struct net_device *vf_netdev)
 	vf_netdev->wanted_features = ndev->features;
 	netdev_update_features(vf_netdev);
 
+	prog = netvsc_xdp_get(netvsc_dev);
+	netvsc_vf_setxdp(vf_netdev, prog);
+
 	return NOTIFY_OK;
 }
 
@@ -2252,6 +2369,8 @@ static int netvsc_unregister_vf(struct net_device *vf_netdev)
 
 	netdev_info(ndev, "VF unregistering: %s\n", vf_netdev->name);
 
+	netvsc_vf_setxdp(vf_netdev, NULL);
+
 	netdev_rx_handler_unregister(vf_netdev);
 	netdev_upper_dev_unlink(vf_netdev, ndev);
 	RCU_INIT_POINTER(net_device_ctx->vf_netdev, NULL);
@@ -2363,14 +2482,14 @@ static int netvsc_probe(struct hv_device *dev,
 	list_add(&net_device_ctx->list, &netvsc_dev_list);
 	rtnl_unlock();
 
-	kfree(device_info);
+	netvsc_devinfo_put(device_info);
 	return 0;
 
 register_failed:
 	rtnl_unlock();
 	rndis_filter_device_remove(dev, nvdev);
 rndis_failed:
-	kfree(device_info);
+	netvsc_devinfo_put(device_info);
 devinfo_failed:
 	free_percpu(net_device_ctx->vf_stats);
 no_stats:
@@ -2398,8 +2517,10 @@ static int netvsc_remove(struct hv_device *dev)
 
 	rtnl_lock();
 	nvdev = rtnl_dereference(ndev_ctx->nvdev);
-	if (nvdev)
+	if (nvdev) {
 		cancel_work_sync(&nvdev->subchan_work);
+		netvsc_xdp_set(net, NULL, NULL, nvdev);
+	}
 
 	/*
 	 * Call to the vsc driver to let it know that the device is being
@@ -2472,11 +2593,11 @@ static int netvsc_resume(struct hv_device *dev)
 
 	ret = netvsc_attach(net, device_info);
 
-	rtnl_unlock();
-
-	kfree(device_info);
+	netvsc_devinfo_put(device_info);
 	net_device_ctx->saved_netvsc_dev_info = NULL;
 
+	rtnl_unlock();
+
 	return ret;
 }
 static const struct hv_vmbus_device_id id_table[] = {
diff --git a/drivers/net/hyperv/rndis_filter.c b/drivers/net/hyperv/rndis_filter.c
index e66d77dc28c8..b81ceba38218 100644
--- a/drivers/net/hyperv/rndis_filter.c
+++ b/drivers/net/hyperv/rndis_filter.c
@@ -235,7 +235,7 @@ static int rndis_filter_send_request(struct rndis_device *dev,
 	trace_rndis_send(dev->ndev, 0, &req->request_msg);
 
 	rcu_read_lock_bh();
-	ret = netvsc_send(dev->ndev, packet, NULL, pb, NULL);
+	ret = netvsc_send(dev->ndev, packet, NULL, pb, NULL, false);
 	rcu_read_unlock_bh();
 
 	return ret;
-- 
cgit v1.2.3-59-g8ed1b


From 12fa74383ed4d1ffa283f77c1e7fe038e8182405 Mon Sep 17 00:00:00 2001
From: Haiyang Zhang <haiyangz@microsoft.com>
Date: Thu, 23 Jan 2020 13:52:35 -0800
Subject: hv_netvsc: Update document for XDP support

Added the new section in the document regarding XDP support
by hv_netvsc driver.

Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../networking/device_drivers/microsoft/netvsc.txt  | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/Documentation/networking/device_drivers/microsoft/netvsc.txt b/Documentation/networking/device_drivers/microsoft/netvsc.txt
index 3bfa635bbbd5..cd63556b27a0 100644
--- a/Documentation/networking/device_drivers/microsoft/netvsc.txt
+++ b/Documentation/networking/device_drivers/microsoft/netvsc.txt
@@ -82,3 +82,24 @@ Features
   contain one or more packets. The send buffer is an optimization, the driver
   will use slower method to handle very large packets or if the send buffer
   area is exhausted.
+
+  XDP support
+  -----------
+  XDP (eXpress Data Path) is a feature that runs eBPF bytecode at the early
+  stage when packets arrive at a NIC card. The goal is to increase performance
+  for packet processing, reducing the overhead of SKB allocation and other
+  upper network layers.
+
+  hv_netvsc supports XDP in native mode, and transparently sets the XDP
+  program on the associated VF NIC as well.
+
+  Setting / unsetting XDP program on synthetic NIC (netvsc) propagates to
+  VF NIC automatically. Setting / unsetting XDP program on VF NIC directly
+  is not recommended, also not propagated to synthetic NIC, and may be
+  overwritten by setting of synthetic NIC.
+
+  XDP program cannot run with LRO (RSC) enabled, so you need to disable LRO
+  before running XDP:
+	ethtool -K eth0 lro off
+
+  XDP_REDIRECT action is not yet supported.
-- 
cgit v1.2.3-59-g8ed1b


From c2070152747ebc294f018e4b28bf2d9f719a6b36 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@mellanox.com>
Date: Fri, 24 Jan 2020 15:23:05 +0200
Subject: net: sched: sch_tbf: Don't overwrite backlog before dumping

In 2011, in commit b0460e4484f9 ("sch_tbf: report backlog information"),
TBF started copying backlog depth from the child Qdisc before dumping, with
the motivation that the backlog was otherwise not visible in "tc -s qdisc
show".

Later, in 2016, in commit 8d5958f424b6 ("sch_tbf: update backlog as well"),
TBF got a full-blown backlog tracking. However it kept copying the child's
backlog over before dumping.

That line is now unnecessary, so remove it.

As shown in the following example, backlog is still reported correctly:

    # tc -s qdisc show dev veth0 invisible
    qdisc tbf 1: root refcnt 2 rate 1Mbit burst 128Kb lat 82.8s
     Sent 505475370 bytes 406985 pkt (dropped 0, overlimits 812544 requeues 0)
     backlog 81972b 66p requeues 0
    qdisc bfifo 0: parent 1:1 limit 10Mb
     Sent 505475370 bytes 406985 pkt (dropped 0, overlimits 0 requeues 0)
     backlog 81972b 66p requeues 0

Signed-off-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sched/sch_tbf.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c
index 2cd94973795c..7ae317958090 100644
--- a/net/sched/sch_tbf.c
+++ b/net/sched/sch_tbf.c
@@ -441,7 +441,6 @@ static int tbf_dump(struct Qdisc *sch, struct sk_buff *skb)
 	struct nlattr *nest;
 	struct tc_tbf_qopt opt;
 
-	sch->qstats.backlog = q->qdisc->qstats.backlog;
 	nest = nla_nest_start_noflag(skb, TCA_OPTIONS);
 	if (nest == NULL)
 		goto nla_put_failure;
-- 
cgit v1.2.3-59-g8ed1b


From ef6aadcc76c97e25f62adc4e9d19684d3e5d0b87 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@mellanox.com>
Date: Fri, 24 Jan 2020 15:23:06 +0200
Subject: net: sched: Make TBF Qdisc offloadable

Invoke ndo_setup_tc as appropriate to signal init / replacement, destroying
and dumping of TBF Qdisc.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  1 +
 include/net/pkt_cls.h     | 22 +++++++++++++++++++
 net/sched/sch_tbf.c       | 55 +++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 78 insertions(+)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 5ec3537fbdb1..11bdf6cb30bd 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -850,6 +850,7 @@ enum tc_setup_type {
 	TC_SETUP_QDISC_TAPRIO,
 	TC_SETUP_FT,
 	TC_SETUP_QDISC_ETS,
+	TC_SETUP_QDISC_TBF,
 };
 
 /* These structures hold the attributes of bpf state that are being passed
diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h
index 47b115e2012a..ce036492986a 100644
--- a/include/net/pkt_cls.h
+++ b/include/net/pkt_cls.h
@@ -854,4 +854,26 @@ struct tc_ets_qopt_offload {
 	};
 };
 
+enum tc_tbf_command {
+	TC_TBF_REPLACE,
+	TC_TBF_DESTROY,
+	TC_TBF_STATS,
+};
+
+struct tc_tbf_qopt_offload_replace_params {
+	struct psched_ratecfg rate;
+	u32 max_size;
+	struct gnet_stats_queue *qstats;
+};
+
+struct tc_tbf_qopt_offload {
+	enum tc_tbf_command command;
+	u32 handle;
+	u32 parent;
+	union {
+		struct tc_tbf_qopt_offload_replace_params replace_params;
+		struct tc_qopt_offload_stats stats;
+	};
+};
+
 #endif
diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c
index 7ae317958090..78e79029dc63 100644
--- a/net/sched/sch_tbf.c
+++ b/net/sched/sch_tbf.c
@@ -15,6 +15,7 @@
 #include <linux/skbuff.h>
 #include <net/netlink.h>
 #include <net/sch_generic.h>
+#include <net/pkt_cls.h>
 #include <net/pkt_sched.h>
 
 
@@ -137,6 +138,52 @@ static u64 psched_ns_t2l(const struct psched_ratecfg *r,
 	return len;
 }
 
+static void tbf_offload_change(struct Qdisc *sch)
+{
+	struct tbf_sched_data *q = qdisc_priv(sch);
+	struct net_device *dev = qdisc_dev(sch);
+	struct tc_tbf_qopt_offload qopt;
+
+	if (!tc_can_offload(dev) || !dev->netdev_ops->ndo_setup_tc)
+		return;
+
+	qopt.command = TC_TBF_REPLACE;
+	qopt.handle = sch->handle;
+	qopt.parent = sch->parent;
+	qopt.replace_params.rate = q->rate;
+	qopt.replace_params.max_size = q->max_size;
+	qopt.replace_params.qstats = &sch->qstats;
+
+	dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_QDISC_TBF, &qopt);
+}
+
+static void tbf_offload_destroy(struct Qdisc *sch)
+{
+	struct net_device *dev = qdisc_dev(sch);
+	struct tc_tbf_qopt_offload qopt;
+
+	if (!tc_can_offload(dev) || !dev->netdev_ops->ndo_setup_tc)
+		return;
+
+	qopt.command = TC_TBF_DESTROY;
+	qopt.handle = sch->handle;
+	qopt.parent = sch->parent;
+	dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_QDISC_TBF, &qopt);
+}
+
+static int tbf_offload_dump(struct Qdisc *sch)
+{
+	struct tc_tbf_qopt_offload qopt;
+
+	qopt.command = TC_TBF_STATS;
+	qopt.handle = sch->handle;
+	qopt.parent = sch->parent;
+	qopt.stats.bstats = &sch->bstats;
+	qopt.stats.qstats = &sch->qstats;
+
+	return qdisc_offload_dump_helper(sch, TC_SETUP_QDISC_TBF, &qopt);
+}
+
 /* GSO packet is too big, segment it so that tbf can transmit
  * each segment in time
  */
@@ -407,6 +454,8 @@ static int tbf_change(struct Qdisc *sch, struct nlattr *opt,
 
 	sch_tree_unlock(sch);
 	err = 0;
+
+	tbf_offload_change(sch);
 done:
 	return err;
 }
@@ -432,6 +481,7 @@ static void tbf_destroy(struct Qdisc *sch)
 	struct tbf_sched_data *q = qdisc_priv(sch);
 
 	qdisc_watchdog_cancel(&q->watchdog);
+	tbf_offload_destroy(sch);
 	qdisc_put(q->qdisc);
 }
 
@@ -440,6 +490,11 @@ static int tbf_dump(struct Qdisc *sch, struct sk_buff *skb)
 	struct tbf_sched_data *q = qdisc_priv(sch);
 	struct nlattr *nest;
 	struct tc_tbf_qopt opt;
+	int err;
+
+	err = tbf_offload_dump(sch);
+	if (err)
+		return err;
 
 	nest = nla_nest_start_noflag(skb, TCA_OPTIONS);
 	if (nest == NULL)
-- 
cgit v1.2.3-59-g8ed1b


From cf9af379cd672d6873ae97ec416148708c6575e9 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@mellanox.com>
Date: Fri, 24 Jan 2020 15:23:07 +0200
Subject: mlxsw: spectrum_qdisc: Extract a per-TC stat function

Extract from mlxsw_sp_qdisc_get_prio_stats() two new functions:
mlxsw_sp_qdisc_collect_tc_stats() to accumulate stats for that one TC only,
and mlxsw_sp_qdisc_update_stats() that makes the stats relative to base
values stored earlier. Use them from mlxsw_sp_qdisc_get_red_stats().

Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../net/ethernet/mellanox/mlxsw/spectrum_qdisc.c   | 119 ++++++++++++---------
 1 file changed, 70 insertions(+), 49 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c
index d57c9b15f45e..ef63b8cbbc94 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c
@@ -227,6 +227,52 @@ mlxsw_sp_qdisc_bstats_per_priority_get(struct mlxsw_sp_port_xstats *xstats,
 	}
 }
 
+static void
+mlxsw_sp_qdisc_collect_tc_stats(struct mlxsw_sp_port *mlxsw_sp_port,
+				struct mlxsw_sp_qdisc *mlxsw_sp_qdisc,
+				u64 *p_tx_bytes, u64 *p_tx_packets,
+				u64 *p_drops, u64 *p_backlog)
+{
+	u8 tclass_num = mlxsw_sp_qdisc->tclass_num;
+	struct mlxsw_sp_port_xstats *xstats;
+	u64 tx_bytes, tx_packets;
+
+	xstats = &mlxsw_sp_port->periodic_hw_stats.xstats;
+	mlxsw_sp_qdisc_bstats_per_priority_get(xstats,
+					       mlxsw_sp_qdisc->prio_bitmap,
+					       &tx_packets, &tx_bytes);
+
+	*p_tx_packets += tx_packets;
+	*p_tx_bytes += tx_bytes;
+	*p_drops += xstats->wred_drop[tclass_num] +
+		    mlxsw_sp_xstats_tail_drop(xstats, tclass_num);
+	*p_backlog += mlxsw_sp_xstats_backlog(xstats, tclass_num);
+}
+
+static void
+mlxsw_sp_qdisc_update_stats(struct mlxsw_sp *mlxsw_sp,
+			    struct mlxsw_sp_qdisc *mlxsw_sp_qdisc,
+			    u64 tx_bytes, u64 tx_packets,
+			    u64 drops, u64 backlog,
+			    struct tc_qopt_offload_stats *stats_ptr)
+{
+	struct mlxsw_sp_qdisc_stats *stats_base = &mlxsw_sp_qdisc->stats_base;
+
+	tx_bytes -= stats_base->tx_bytes;
+	tx_packets -= stats_base->tx_packets;
+	drops -= stats_base->drops;
+	backlog -= stats_base->backlog;
+
+	_bstats_update(stats_ptr->bstats, tx_bytes, tx_packets);
+	stats_ptr->qstats->drops += drops;
+	stats_ptr->qstats->backlog += mlxsw_sp_cells_bytes(mlxsw_sp, backlog);
+
+	stats_base->backlog += backlog;
+	stats_base->drops += drops;
+	stats_base->tx_bytes += tx_bytes;
+	stats_base->tx_packets += tx_packets;
+}
+
 static int
 mlxsw_sp_tclass_congestion_enable(struct mlxsw_sp_port *mlxsw_sp_port,
 				  int tclass_num, u32 min, u32 max,
@@ -403,41 +449,30 @@ mlxsw_sp_qdisc_get_red_stats(struct mlxsw_sp_port *mlxsw_sp_port,
 			     struct mlxsw_sp_qdisc *mlxsw_sp_qdisc,
 			     struct tc_qopt_offload_stats *stats_ptr)
 {
-	u64 tx_bytes, tx_packets, overlimits, drops, backlog;
 	u8 tclass_num = mlxsw_sp_qdisc->tclass_num;
 	struct mlxsw_sp_qdisc_stats *stats_base;
 	struct mlxsw_sp_port_xstats *xstats;
+	u64 tx_packets = 0;
+	u64 tx_bytes = 0;
+	u64 backlog = 0;
+	u64 overlimits;
+	u64 drops = 0;
 
 	xstats = &mlxsw_sp_port->periodic_hw_stats.xstats;
 	stats_base = &mlxsw_sp_qdisc->stats_base;
 
-	mlxsw_sp_qdisc_bstats_per_priority_get(xstats,
-					       mlxsw_sp_qdisc->prio_bitmap,
-					       &tx_packets, &tx_bytes);
-	tx_bytes = tx_bytes - stats_base->tx_bytes;
-	tx_packets = tx_packets - stats_base->tx_packets;
-
+	mlxsw_sp_qdisc_collect_tc_stats(mlxsw_sp_port, mlxsw_sp_qdisc,
+					&tx_bytes, &tx_packets,
+					&drops, &backlog);
 	overlimits = xstats->wred_drop[tclass_num] + xstats->ecn -
 		     stats_base->overlimits;
-	drops = xstats->wred_drop[tclass_num] +
-		mlxsw_sp_xstats_tail_drop(xstats, tclass_num) -
-		stats_base->drops;
-	backlog = mlxsw_sp_xstats_backlog(xstats, tclass_num);
 
-	_bstats_update(stats_ptr->bstats, tx_bytes, tx_packets);
+	mlxsw_sp_qdisc_update_stats(mlxsw_sp_port->mlxsw_sp, mlxsw_sp_qdisc,
+				    tx_bytes, tx_packets, drops, backlog,
+				    stats_ptr);
 	stats_ptr->qstats->overlimits += overlimits;
-	stats_ptr->qstats->drops += drops;
-	stats_ptr->qstats->backlog +=
-				mlxsw_sp_cells_bytes(mlxsw_sp_port->mlxsw_sp,
-						     backlog) -
-				mlxsw_sp_cells_bytes(mlxsw_sp_port->mlxsw_sp,
-						     stats_base->backlog);
-
-	stats_base->backlog = backlog;
-	stats_base->drops +=  drops;
 	stats_base->overlimits += overlimits;
-	stats_base->tx_bytes += tx_bytes;
-	stats_base->tx_packets += tx_packets;
+
 	return 0;
 }
 
@@ -628,37 +663,23 @@ mlxsw_sp_qdisc_get_prio_stats(struct mlxsw_sp_port *mlxsw_sp_port,
 			      struct mlxsw_sp_qdisc *mlxsw_sp_qdisc,
 			      struct tc_qopt_offload_stats *stats_ptr)
 {
-	u64 tx_bytes, tx_packets, drops = 0, backlog = 0;
-	struct mlxsw_sp_qdisc_stats *stats_base;
-	struct mlxsw_sp_port_xstats *xstats;
-	struct rtnl_link_stats64 *stats;
+	struct mlxsw_sp_qdisc *tc_qdisc;
+	u64 tx_packets = 0;
+	u64 tx_bytes = 0;
+	u64 backlog = 0;
+	u64 drops = 0;
 	int i;
 
-	xstats = &mlxsw_sp_port->periodic_hw_stats.xstats;
-	stats = &mlxsw_sp_port->periodic_hw_stats.stats;
-	stats_base = &mlxsw_sp_qdisc->stats_base;
-
-	tx_bytes = stats->tx_bytes - stats_base->tx_bytes;
-	tx_packets = stats->tx_packets - stats_base->tx_packets;
-
 	for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) {
-		drops += mlxsw_sp_xstats_tail_drop(xstats, i);
-		drops += xstats->wred_drop[i];
-		backlog += mlxsw_sp_xstats_backlog(xstats, i);
+		tc_qdisc = &mlxsw_sp_port->tclass_qdiscs[i];
+		mlxsw_sp_qdisc_collect_tc_stats(mlxsw_sp_port, tc_qdisc,
+						&tx_bytes, &tx_packets,
+						&drops, &backlog);
 	}
-	drops = drops - stats_base->drops;
 
-	_bstats_update(stats_ptr->bstats, tx_bytes, tx_packets);
-	stats_ptr->qstats->drops += drops;
-	stats_ptr->qstats->backlog +=
-				mlxsw_sp_cells_bytes(mlxsw_sp_port->mlxsw_sp,
-						     backlog) -
-				mlxsw_sp_cells_bytes(mlxsw_sp_port->mlxsw_sp,
-						     stats_base->backlog);
-	stats_base->backlog = backlog;
-	stats_base->drops += drops;
-	stats_base->tx_bytes += tx_bytes;
-	stats_base->tx_packets += tx_packets;
+	mlxsw_sp_qdisc_update_stats(mlxsw_sp_port->mlxsw_sp, mlxsw_sp_qdisc,
+				    tx_bytes, tx_packets, drops, backlog,
+				    stats_ptr);
 	return 0;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 3d0d592193dce954bf5ccac5128080830dd8e982 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@mellanox.com>
Date: Fri, 24 Jan 2020 15:23:08 +0200
Subject: mlxsw: spectrum_qdisc: Add mlxsw_sp_qdisc_get_class_stats()

Add a wrapper around mlxsw_sp_qdisc_collect_tc_stats() and
mlxsw_sp_qdisc_update_stats() for the simple case of doing both in one go:
mlxsw_sp_qdisc_get_class_stats(). Dispatch to that function from
mlxsw_sp_qdisc_get_red_stats(). This new function will be useful for other
leaf Qdiscs as well.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../net/ethernet/mellanox/mlxsw/spectrum_qdisc.c   | 29 ++++++++++++++--------
 1 file changed, 19 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c
index ef63b8cbbc94..a0f07e951607 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c
@@ -273,6 +273,24 @@ mlxsw_sp_qdisc_update_stats(struct mlxsw_sp *mlxsw_sp,
 	stats_base->tx_packets += tx_packets;
 }
 
+static void
+mlxsw_sp_qdisc_get_tc_stats(struct mlxsw_sp_port *mlxsw_sp_port,
+			    struct mlxsw_sp_qdisc *mlxsw_sp_qdisc,
+			    struct tc_qopt_offload_stats *stats_ptr)
+{
+	u64 tx_packets = 0;
+	u64 tx_bytes = 0;
+	u64 backlog = 0;
+	u64 drops = 0;
+
+	mlxsw_sp_qdisc_collect_tc_stats(mlxsw_sp_port, mlxsw_sp_qdisc,
+					&tx_bytes, &tx_packets,
+					&drops, &backlog);
+	mlxsw_sp_qdisc_update_stats(mlxsw_sp_port->mlxsw_sp, mlxsw_sp_qdisc,
+				    tx_bytes, tx_packets, drops, backlog,
+				    stats_ptr);
+}
+
 static int
 mlxsw_sp_tclass_congestion_enable(struct mlxsw_sp_port *mlxsw_sp_port,
 				  int tclass_num, u32 min, u32 max,
@@ -452,24 +470,15 @@ mlxsw_sp_qdisc_get_red_stats(struct mlxsw_sp_port *mlxsw_sp_port,
 	u8 tclass_num = mlxsw_sp_qdisc->tclass_num;
 	struct mlxsw_sp_qdisc_stats *stats_base;
 	struct mlxsw_sp_port_xstats *xstats;
-	u64 tx_packets = 0;
-	u64 tx_bytes = 0;
-	u64 backlog = 0;
 	u64 overlimits;
-	u64 drops = 0;
 
 	xstats = &mlxsw_sp_port->periodic_hw_stats.xstats;
 	stats_base = &mlxsw_sp_qdisc->stats_base;
 
-	mlxsw_sp_qdisc_collect_tc_stats(mlxsw_sp_port, mlxsw_sp_qdisc,
-					&tx_bytes, &tx_packets,
-					&drops, &backlog);
+	mlxsw_sp_qdisc_get_tc_stats(mlxsw_sp_port, mlxsw_sp_qdisc, stats_ptr);
 	overlimits = xstats->wred_drop[tclass_num] + xstats->ecn -
 		     stats_base->overlimits;
 
-	mlxsw_sp_qdisc_update_stats(mlxsw_sp_port->mlxsw_sp, mlxsw_sp_qdisc,
-				    tx_bytes, tx_packets, drops, backlog,
-				    stats_ptr);
 	stats_ptr->qstats->overlimits += overlimits;
 	stats_base->overlimits += overlimits;
 
-- 
cgit v1.2.3-59-g8ed1b


From be1d5a8a7728cc734fc31b837796e7addc4ba832 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@mellanox.com>
Date: Fri, 24 Jan 2020 15:23:09 +0200
Subject: mlxsw: spectrum_qdisc: Extract a common leaf unoffload function

When the RED Qdisc is unoffloaded, it needs to reduce the reported backlog
by the amount that is in the HW, so that only the SW backlog is contained
in the counter. The same thing will need to be done by TBF, and likely any
other leaf Qdisc as well.

Extract a helper mlxsw_sp_qdisc_leaf_unoffload() and call it from
mlxsw_sp_qdisc_red_unoffload().

Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c | 19 ++++++++++++++-----
 1 file changed, 14 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c
index a0f07e951607..57b014a95bc8 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c
@@ -421,19 +421,28 @@ mlxsw_sp_qdisc_red_replace(struct mlxsw_sp_port *mlxsw_sp_port,
 }
 
 static void
-mlxsw_sp_qdisc_red_unoffload(struct mlxsw_sp_port *mlxsw_sp_port,
-			     struct mlxsw_sp_qdisc *mlxsw_sp_qdisc,
-			     void *params)
+mlxsw_sp_qdisc_leaf_unoffload(struct mlxsw_sp_port *mlxsw_sp_port,
+			      struct mlxsw_sp_qdisc *mlxsw_sp_qdisc,
+			      struct gnet_stats_queue *qstats)
 {
-	struct tc_red_qopt_offload_params *p = params;
 	u64 backlog;
 
 	backlog = mlxsw_sp_cells_bytes(mlxsw_sp_port->mlxsw_sp,
 				       mlxsw_sp_qdisc->stats_base.backlog);
-	p->qstats->backlog -= backlog;
+	qstats->backlog -= backlog;
 	mlxsw_sp_qdisc->stats_base.backlog = 0;
 }
 
+static void
+mlxsw_sp_qdisc_red_unoffload(struct mlxsw_sp_port *mlxsw_sp_port,
+			     struct mlxsw_sp_qdisc *mlxsw_sp_qdisc,
+			     void *params)
+{
+	struct tc_red_qopt_offload_params *p = params;
+
+	mlxsw_sp_qdisc_leaf_unoffload(mlxsw_sp_port, mlxsw_sp_qdisc, p->qstats);
+}
+
 static int
 mlxsw_sp_qdisc_get_red_xstats(struct mlxsw_sp_port *mlxsw_sp_port,
 			      struct mlxsw_sp_qdisc *mlxsw_sp_qdisc,
-- 
cgit v1.2.3-59-g8ed1b


From 23effa2479bae2fbea69493afaab9914ee3862ca Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@mellanox.com>
Date: Fri, 24 Jan 2020 15:23:10 +0200
Subject: mlxsw: reg: Add max_shaper_bs to QoS ETS Element Configuration

The QEEC register configures scheduling elements. One of the bits of
configuration is the burst size to use for the shaper installed on the
element. Add the necessary fields to support this configuration.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/reg.h | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h
index 0b80e75e87c3..afd712d8fd46 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/reg.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h
@@ -3602,6 +3602,21 @@ MLXSW_ITEM32(reg, qeec, dwrr, 0x18, 15, 1);
  */
 MLXSW_ITEM32(reg, qeec, dwrr_weight, 0x18, 0, 8);
 
+/* reg_qeec_max_shaper_bs
+ * Max shaper burst size
+ * Burst size is 2^max_shaper_bs * 512 bits
+ * For Spectrum-1: Range is: 5..25
+ * For Spectrum-2: Range is: 11..25
+ * Reserved when ptps = 1
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, qeec, max_shaper_bs, 0x1C, 0, 6);
+
+#define MLXSW_REG_QEEC_HIGHEST_SHAPER_BS	25
+#define MLXSW_REG_QEEC_LOWEST_SHAPER_BS_SP1	5
+#define MLXSW_REG_QEEC_LOWEST_SHAPER_BS_SP2	11
+#define MLXSW_REG_QEEC_LOWEST_SHAPER_BS_SP3	5
+
 static inline void mlxsw_reg_qeec_pack(char *payload, u8 local_port,
 				       enum mlxsw_reg_qeec_hr hr, u8 index,
 				       u8 next_index)
-- 
cgit v1.2.3-59-g8ed1b


From 92afbfedb77d59a4077d292b88a316f3de706344 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@mellanox.com>
Date: Fri, 24 Jan 2020 15:23:11 +0200
Subject: mlxsw: reg: Increase MLXSW_REG_QEEC_MAS_DIS

As the port speeds grow, the current value of "unlimited shaper",
200000000Kbps, might become lower than the actually supported speeds. Bump
it to the maximum value that fits in the corresponding QEEC field, which is
about 2.1Tbps.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/reg.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h
index afd712d8fd46..dd6685156396 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/reg.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h
@@ -3563,8 +3563,8 @@ MLXSW_ITEM32(reg, qeec, min_shaper_rate, 0x0C, 0, 28);
  */
 MLXSW_ITEM32(reg, qeec, mase, 0x10, 31, 1);
 
-/* A large max rate will disable the max shaper. */
-#define MLXSW_REG_QEEC_MAS_DIS	200000000	/* Kbps */
+/* The largest max shaper value possible to disable the shaper. */
+#define MLXSW_REG_QEEC_MAS_DIS	((1u << 31) - 1)	/* Kbps */
 
 /* reg_qeec_max_shaper_rate
  * Max shaper information rate.
-- 
cgit v1.2.3-59-g8ed1b


From 47259544e0fc0659f74f6ca6680745658969a803 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@mellanox.com>
Date: Fri, 24 Jan 2020 15:23:12 +0200
Subject: mlxsw: spectrum: Add lowest_shaper_bs to struct mlxsw_sp

Lower limit of burst size configuration is dependent on system type. Add a
datum to track the value. Initialize as appropriate in mlxsw_spX_init().

Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum.c | 3 +++
 drivers/net/ethernet/mellanox/mlxsw/spectrum.h | 1 +
 2 files changed, 4 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index 9eb3ac7669f7..72d24595df7a 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -5173,6 +5173,7 @@ static int mlxsw_sp1_init(struct mlxsw_core *mlxsw_core,
 	mlxsw_sp->span_ops = &mlxsw_sp1_span_ops;
 	mlxsw_sp->listeners = mlxsw_sp1_listener;
 	mlxsw_sp->listeners_count = ARRAY_SIZE(mlxsw_sp1_listener);
+	mlxsw_sp->lowest_shaper_bs = MLXSW_REG_QEEC_LOWEST_SHAPER_BS_SP1;
 
 	return mlxsw_sp_init(mlxsw_core, mlxsw_bus_info, extack);
 }
@@ -5197,6 +5198,7 @@ static int mlxsw_sp2_init(struct mlxsw_core *mlxsw_core,
 	mlxsw_sp->port_type_speed_ops = &mlxsw_sp2_port_type_speed_ops;
 	mlxsw_sp->ptp_ops = &mlxsw_sp2_ptp_ops;
 	mlxsw_sp->span_ops = &mlxsw_sp2_span_ops;
+	mlxsw_sp->lowest_shaper_bs = MLXSW_REG_QEEC_LOWEST_SHAPER_BS_SP2;
 
 	return mlxsw_sp_init(mlxsw_core, mlxsw_bus_info, extack);
 }
@@ -5219,6 +5221,7 @@ static int mlxsw_sp3_init(struct mlxsw_core *mlxsw_core,
 	mlxsw_sp->port_type_speed_ops = &mlxsw_sp2_port_type_speed_ops;
 	mlxsw_sp->ptp_ops = &mlxsw_sp2_ptp_ops;
 	mlxsw_sp->span_ops = &mlxsw_sp2_span_ops;
+	mlxsw_sp->lowest_shaper_bs = MLXSW_REG_QEEC_LOWEST_SHAPER_BS_SP3;
 
 	return mlxsw_sp_init(mlxsw_core, mlxsw_bus_info, extack);
 }
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
index 5f3b74360dc8..bbc23939964e 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
@@ -189,6 +189,7 @@ struct mlxsw_sp {
 	const struct mlxsw_sp_span_ops *span_ops;
 	const struct mlxsw_listener *listeners;
 	size_t listeners_count;
+	u32 lowest_shaper_bs;
 };
 
 static inline struct mlxsw_sp_upper *
-- 
cgit v1.2.3-59-g8ed1b


From dbacf8ba5860e79d5191b0868bffa9e351d205e4 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@mellanox.com>
Date: Fri, 24 Jan 2020 15:23:13 +0200
Subject: mlxsw: spectrum: Configure shaper rate and burst size together

In order to allow configuration of burst size together with shaper rate,
extend mlxsw_sp_port_ets_maxrate_set() with a burst_size argument. Convert
call sites to pass 0 (for default).

Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum.c     | 11 ++++++-----
 drivers/net/ethernet/mellanox/mlxsw/spectrum.h     |  2 +-
 drivers/net/ethernet/mellanox/mlxsw/spectrum_dcb.c |  5 +++--
 3 files changed, 10 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index 72d24595df7a..021664c62bba 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -3577,7 +3577,7 @@ int mlxsw_sp_port_ets_set(struct mlxsw_sp_port *mlxsw_sp_port,
 
 int mlxsw_sp_port_ets_maxrate_set(struct mlxsw_sp_port *mlxsw_sp_port,
 				  enum mlxsw_reg_qeec_hr hr, u8 index,
-				  u8 next_index, u32 maxrate)
+				  u8 next_index, u32 maxrate, u8 burst_size)
 {
 	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
 	char qeec_pl[MLXSW_REG_QEEC_LEN];
@@ -3586,6 +3586,7 @@ int mlxsw_sp_port_ets_maxrate_set(struct mlxsw_sp_port *mlxsw_sp_port,
 			    next_index);
 	mlxsw_reg_qeec_mase_set(qeec_pl, true);
 	mlxsw_reg_qeec_max_shaper_rate_set(qeec_pl, maxrate);
+	mlxsw_reg_qeec_max_shaper_bs_set(qeec_pl, burst_size);
 	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(qeec), qeec_pl);
 }
 
@@ -3654,14 +3655,14 @@ static int mlxsw_sp_port_ets_init(struct mlxsw_sp_port *mlxsw_sp_port)
 	 */
 	err = mlxsw_sp_port_ets_maxrate_set(mlxsw_sp_port,
 					    MLXSW_REG_QEEC_HR_PORT, 0, 0,
-					    MLXSW_REG_QEEC_MAS_DIS);
+					    MLXSW_REG_QEEC_MAS_DIS, 0);
 	if (err)
 		return err;
 	for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) {
 		err = mlxsw_sp_port_ets_maxrate_set(mlxsw_sp_port,
 						    MLXSW_REG_QEEC_HR_SUBGROUP,
 						    i, 0,
-						    MLXSW_REG_QEEC_MAS_DIS);
+						    MLXSW_REG_QEEC_MAS_DIS, 0);
 		if (err)
 			return err;
 	}
@@ -3669,14 +3670,14 @@ static int mlxsw_sp_port_ets_init(struct mlxsw_sp_port *mlxsw_sp_port)
 		err = mlxsw_sp_port_ets_maxrate_set(mlxsw_sp_port,
 						    MLXSW_REG_QEEC_HR_TC,
 						    i, i,
-						    MLXSW_REG_QEEC_MAS_DIS);
+						    MLXSW_REG_QEEC_MAS_DIS, 0);
 		if (err)
 			return err;
 
 		err = mlxsw_sp_port_ets_maxrate_set(mlxsw_sp_port,
 						    MLXSW_REG_QEEC_HR_TC,
 						    i + 8, i,
-						    MLXSW_REG_QEEC_MAS_DIS);
+						    MLXSW_REG_QEEC_MAS_DIS, 0);
 		if (err)
 			return err;
 	}
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
index bbc23939964e..a2393c02a5e5 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
@@ -488,7 +488,7 @@ int __mlxsw_sp_port_headroom_set(struct mlxsw_sp_port *mlxsw_sp_port, int mtu,
 				 struct ieee_pfc *my_pfc);
 int mlxsw_sp_port_ets_maxrate_set(struct mlxsw_sp_port *mlxsw_sp_port,
 				  enum mlxsw_reg_qeec_hr hr, u8 index,
-				  u8 next_index, u32 maxrate);
+				  u8 next_index, u32 maxrate, u8 burst_size);
 enum mlxsw_reg_spms_state mlxsw_sp_stp_spms_state(u8 stp_state);
 int mlxsw_sp_port_vid_stp_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid,
 			      u8 state);
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_dcb.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_dcb.c
index db66f2b56a6d..49a72a8f1f57 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_dcb.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_dcb.c
@@ -526,7 +526,7 @@ static int mlxsw_sp_dcbnl_ieee_setmaxrate(struct net_device *dev,
 		err = mlxsw_sp_port_ets_maxrate_set(mlxsw_sp_port,
 						    MLXSW_REG_QEEC_HR_SUBGROUP,
 						    i, 0,
-						    maxrate->tc_maxrate[i]);
+						    maxrate->tc_maxrate[i], 0);
 		if (err) {
 			netdev_err(dev, "Failed to set maxrate for TC %d\n", i);
 			goto err_port_ets_maxrate_set;
@@ -541,7 +541,8 @@ err_port_ets_maxrate_set:
 	for (i--; i >= 0; i--)
 		mlxsw_sp_port_ets_maxrate_set(mlxsw_sp_port,
 					      MLXSW_REG_QEEC_HR_SUBGROUP,
-					      i, 0, my_maxrate->tc_maxrate[i]);
+					      i, 0,
+					      my_maxrate->tc_maxrate[i], 0);
 	return err;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From a44f58c41bfbb921c1c4586bed9ad26aa2c6208a Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@mellanox.com>
Date: Fri, 24 Jan 2020 15:23:14 +0200
Subject: mlxsw: spectrum_qdisc: Support offloading of TBF Qdisc

React to the TC messages that were introduced in a preceding patch and
configure egress maximum shaper as appropriate. TBF can be used as a root
qdisc or under one of PRIO or strict ETS bands.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum.c     |   2 +
 drivers/net/ethernet/mellanox/mlxsw/spectrum.h     |   2 +
 .../net/ethernet/mellanox/mlxsw/spectrum_qdisc.c   | 199 +++++++++++++++++++++
 3 files changed, 203 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index 021664c62bba..7358b5bc7eb6 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -1796,6 +1796,8 @@ static int mlxsw_sp_setup_tc(struct net_device *dev, enum tc_setup_type type,
 		return mlxsw_sp_setup_tc_prio(mlxsw_sp_port, type_data);
 	case TC_SETUP_QDISC_ETS:
 		return mlxsw_sp_setup_tc_ets(mlxsw_sp_port, type_data);
+	case TC_SETUP_QDISC_TBF:
+		return mlxsw_sp_setup_tc_tbf(mlxsw_sp_port, type_data);
 	default:
 		return -EOPNOTSUPP;
 	}
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
index a2393c02a5e5..a0f1f9dceec5 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
@@ -862,6 +862,8 @@ int mlxsw_sp_setup_tc_prio(struct mlxsw_sp_port *mlxsw_sp_port,
 			   struct tc_prio_qopt_offload *p);
 int mlxsw_sp_setup_tc_ets(struct mlxsw_sp_port *mlxsw_sp_port,
 			  struct tc_ets_qopt_offload *p);
+int mlxsw_sp_setup_tc_tbf(struct mlxsw_sp_port *mlxsw_sp_port,
+			  struct tc_tbf_qopt_offload *p);
 
 /* spectrum_fid.c */
 bool mlxsw_sp_fid_is_dummy(struct mlxsw_sp *mlxsw_sp, u16 fid_index);
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c
index 57b014a95bc8..79a2801d59f6 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c
@@ -19,6 +19,7 @@ enum mlxsw_sp_qdisc_type {
 	MLXSW_SP_QDISC_RED,
 	MLXSW_SP_QDISC_PRIO,
 	MLXSW_SP_QDISC_ETS,
+	MLXSW_SP_QDISC_TBF,
 };
 
 struct mlxsw_sp_qdisc_ops {
@@ -540,6 +541,204 @@ int mlxsw_sp_setup_tc_red(struct mlxsw_sp_port *mlxsw_sp_port,
 	}
 }
 
+static void
+mlxsw_sp_setup_tc_qdisc_leaf_clean_stats(struct mlxsw_sp_port *mlxsw_sp_port,
+					 struct mlxsw_sp_qdisc *mlxsw_sp_qdisc)
+{
+	u64 backlog_cells = 0;
+	u64 tx_packets = 0;
+	u64 tx_bytes = 0;
+	u64 drops = 0;
+
+	mlxsw_sp_qdisc_collect_tc_stats(mlxsw_sp_port, mlxsw_sp_qdisc,
+					&tx_bytes, &tx_packets,
+					&drops, &backlog_cells);
+
+	mlxsw_sp_qdisc->stats_base.tx_packets = tx_packets;
+	mlxsw_sp_qdisc->stats_base.tx_bytes = tx_bytes;
+	mlxsw_sp_qdisc->stats_base.drops = drops;
+	mlxsw_sp_qdisc->stats_base.backlog = 0;
+}
+
+static int
+mlxsw_sp_qdisc_tbf_destroy(struct mlxsw_sp_port *mlxsw_sp_port,
+			   struct mlxsw_sp_qdisc *mlxsw_sp_qdisc)
+{
+	struct mlxsw_sp_qdisc *root_qdisc = mlxsw_sp_port->root_qdisc;
+
+	if (root_qdisc != mlxsw_sp_qdisc)
+		root_qdisc->stats_base.backlog -=
+					mlxsw_sp_qdisc->stats_base.backlog;
+
+	return mlxsw_sp_port_ets_maxrate_set(mlxsw_sp_port,
+					     MLXSW_REG_QEEC_HR_SUBGROUP,
+					     mlxsw_sp_qdisc->tclass_num, 0,
+					     MLXSW_REG_QEEC_MAS_DIS, 0);
+}
+
+static int
+mlxsw_sp_qdisc_tbf_bs(struct mlxsw_sp_port *mlxsw_sp_port,
+		      u32 max_size, u8 *p_burst_size)
+{
+	/* TBF burst size is configured in bytes. The ASIC burst size value is
+	 * ((2 ^ bs) * 512 bits. Convert the TBF bytes to 512-bit units.
+	 */
+	u32 bs512 = max_size / 64;
+	u8 bs = fls(bs512);
+
+	if (!bs)
+		return -EINVAL;
+	--bs;
+
+	/* Demand a power of two. */
+	if ((1 << bs) != bs512)
+		return -EINVAL;
+
+	if (bs < mlxsw_sp_port->mlxsw_sp->lowest_shaper_bs ||
+	    bs > MLXSW_REG_QEEC_HIGHEST_SHAPER_BS)
+		return -EINVAL;
+
+	*p_burst_size = bs;
+	return 0;
+}
+
+static u32
+mlxsw_sp_qdisc_tbf_max_size(u8 bs)
+{
+	return (1U << bs) * 64;
+}
+
+static u64
+mlxsw_sp_qdisc_tbf_rate_kbps(struct tc_tbf_qopt_offload_replace_params *p)
+{
+	/* TBF interface is in bytes/s, whereas Spectrum ASIC is configured in
+	 * Kbits/s.
+	 */
+	return p->rate.rate_bytes_ps / 1000 * 8;
+}
+
+static int
+mlxsw_sp_qdisc_tbf_check_params(struct mlxsw_sp_port *mlxsw_sp_port,
+				struct mlxsw_sp_qdisc *mlxsw_sp_qdisc,
+				void *params)
+{
+	struct tc_tbf_qopt_offload_replace_params *p = params;
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	u64 rate_kbps = mlxsw_sp_qdisc_tbf_rate_kbps(p);
+	u8 burst_size;
+	int err;
+
+	if (rate_kbps >= MLXSW_REG_QEEC_MAS_DIS) {
+		dev_err(mlxsw_sp_port->mlxsw_sp->bus_info->dev,
+			"spectrum: TBF: rate of %lluKbps must be below %u\n",
+			rate_kbps, MLXSW_REG_QEEC_MAS_DIS);
+		return -EINVAL;
+	}
+
+	err = mlxsw_sp_qdisc_tbf_bs(mlxsw_sp_port, p->max_size, &burst_size);
+	if (err) {
+		u8 highest_shaper_bs = MLXSW_REG_QEEC_HIGHEST_SHAPER_BS;
+
+		dev_err(mlxsw_sp->bus_info->dev,
+			"spectrum: TBF: invalid burst size of %u, must be a power of two between %u and %u",
+			p->max_size,
+			mlxsw_sp_qdisc_tbf_max_size(mlxsw_sp->lowest_shaper_bs),
+			mlxsw_sp_qdisc_tbf_max_size(highest_shaper_bs));
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int
+mlxsw_sp_qdisc_tbf_replace(struct mlxsw_sp_port *mlxsw_sp_port,
+			   struct mlxsw_sp_qdisc *mlxsw_sp_qdisc,
+			   void *params)
+{
+	struct tc_tbf_qopt_offload_replace_params *p = params;
+	u64 rate_kbps = mlxsw_sp_qdisc_tbf_rate_kbps(p);
+	u8 burst_size;
+	int err;
+
+	err = mlxsw_sp_qdisc_tbf_bs(mlxsw_sp_port, p->max_size, &burst_size);
+	if (WARN_ON_ONCE(err))
+		/* check_params above was supposed to reject this value. */
+		return -EINVAL;
+
+	/* Configure subgroup shaper, so that both UC and MC traffic is subject
+	 * to shaping. That is unlike RED, however UC queue lengths are going to
+	 * be different than MC ones due to different pool and quota
+	 * configurations, so the configuration is not applicable. For shaper on
+	 * the other hand, subjecting the overall stream to the configured
+	 * shaper makes sense. Also note that that is what we do for
+	 * ieee_setmaxrate().
+	 */
+	return mlxsw_sp_port_ets_maxrate_set(mlxsw_sp_port,
+					     MLXSW_REG_QEEC_HR_SUBGROUP,
+					     mlxsw_sp_qdisc->tclass_num, 0,
+					     rate_kbps, burst_size);
+}
+
+static void
+mlxsw_sp_qdisc_tbf_unoffload(struct mlxsw_sp_port *mlxsw_sp_port,
+			     struct mlxsw_sp_qdisc *mlxsw_sp_qdisc,
+			     void *params)
+{
+	struct tc_tbf_qopt_offload_replace_params *p = params;
+
+	mlxsw_sp_qdisc_leaf_unoffload(mlxsw_sp_port, mlxsw_sp_qdisc, p->qstats);
+}
+
+static int
+mlxsw_sp_qdisc_get_tbf_stats(struct mlxsw_sp_port *mlxsw_sp_port,
+			     struct mlxsw_sp_qdisc *mlxsw_sp_qdisc,
+			     struct tc_qopt_offload_stats *stats_ptr)
+{
+	mlxsw_sp_qdisc_get_tc_stats(mlxsw_sp_port, mlxsw_sp_qdisc,
+				    stats_ptr);
+	return 0;
+}
+
+static struct mlxsw_sp_qdisc_ops mlxsw_sp_qdisc_ops_tbf = {
+	.type = MLXSW_SP_QDISC_TBF,
+	.check_params = mlxsw_sp_qdisc_tbf_check_params,
+	.replace = mlxsw_sp_qdisc_tbf_replace,
+	.unoffload = mlxsw_sp_qdisc_tbf_unoffload,
+	.destroy = mlxsw_sp_qdisc_tbf_destroy,
+	.get_stats = mlxsw_sp_qdisc_get_tbf_stats,
+	.clean_stats = mlxsw_sp_setup_tc_qdisc_leaf_clean_stats,
+};
+
+int mlxsw_sp_setup_tc_tbf(struct mlxsw_sp_port *mlxsw_sp_port,
+			  struct tc_tbf_qopt_offload *p)
+{
+	struct mlxsw_sp_qdisc *mlxsw_sp_qdisc;
+
+	mlxsw_sp_qdisc = mlxsw_sp_qdisc_find(mlxsw_sp_port, p->parent, false);
+	if (!mlxsw_sp_qdisc)
+		return -EOPNOTSUPP;
+
+	if (p->command == TC_TBF_REPLACE)
+		return mlxsw_sp_qdisc_replace(mlxsw_sp_port, p->handle,
+					      mlxsw_sp_qdisc,
+					      &mlxsw_sp_qdisc_ops_tbf,
+					      &p->replace_params);
+
+	if (!mlxsw_sp_qdisc_compare(mlxsw_sp_qdisc, p->handle,
+				    MLXSW_SP_QDISC_TBF))
+		return -EOPNOTSUPP;
+
+	switch (p->command) {
+	case TC_TBF_DESTROY:
+		return mlxsw_sp_qdisc_destroy(mlxsw_sp_port, mlxsw_sp_qdisc);
+	case TC_TBF_STATS:
+		return mlxsw_sp_qdisc_get_stats(mlxsw_sp_port, mlxsw_sp_qdisc,
+						&p->stats);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
 static int
 __mlxsw_sp_qdisc_ets_destroy(struct mlxsw_sp_port *mlxsw_sp_port)
 {
-- 
cgit v1.2.3-59-g8ed1b


From adc6c7ec11293117ec2763ccbe17d58349e011cf Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@mellanox.com>
Date: Fri, 24 Jan 2020 15:23:15 +0200
Subject: selftests: Move two functions from mlxsw's qos_lib to lib

The function humanize() is used for converting value in bits/s to a
human-friendly approximate value in Kbps, Mbps or Gbps. There is nothing
hardware-specific in that, so move the function to lib.sh.

Similarly for the rate() function, which just does a bit of math to
calculate a rate, given two counter values and a time interval.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../testing/selftests/drivers/net/mlxsw/qos_lib.sh | 24 ----------------------
 tools/testing/selftests/net/forwarding/lib.sh      | 24 ++++++++++++++++++++++
 2 files changed, 24 insertions(+), 24 deletions(-)

diff --git a/tools/testing/selftests/drivers/net/mlxsw/qos_lib.sh b/tools/testing/selftests/drivers/net/mlxsw/qos_lib.sh
index a5937069ac16..faa51012cdac 100644
--- a/tools/testing/selftests/drivers/net/mlxsw/qos_lib.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/qos_lib.sh
@@ -1,29 +1,5 @@
 # SPDX-License-Identifier: GPL-2.0
 
-humanize()
-{
-	local speed=$1; shift
-
-	for unit in bps Kbps Mbps Gbps; do
-		if (($(echo "$speed < 1024" | bc))); then
-			break
-		fi
-
-		speed=$(echo "scale=1; $speed / 1024" | bc)
-	done
-
-	echo "$speed${unit}"
-}
-
-rate()
-{
-	local t0=$1; shift
-	local t1=$1; shift
-	local interval=$1; shift
-
-	echo $((8 * (t1 - t0) / interval))
-}
-
 check_rate()
 {
 	local rate=$1; shift
diff --git a/tools/testing/selftests/net/forwarding/lib.sh b/tools/testing/selftests/net/forwarding/lib.sh
index 8dc5fac98cbc..10cdfc0adca8 100644
--- a/tools/testing/selftests/net/forwarding/lib.sh
+++ b/tools/testing/selftests/net/forwarding/lib.sh
@@ -588,6 +588,30 @@ ethtool_stats_get()
 	ethtool -S $dev | grep "^ *$stat:" | head -n 1 | cut -d: -f2
 }
 
+humanize()
+{
+	local speed=$1; shift
+
+	for unit in bps Kbps Mbps Gbps; do
+		if (($(echo "$speed < 1024" | bc))); then
+			break
+		fi
+
+		speed=$(echo "scale=1; $speed / 1024" | bc)
+	done
+
+	echo "$speed${unit}"
+}
+
+rate()
+{
+	local t0=$1; shift
+	local t1=$1; shift
+	local interval=$1; shift
+
+	echo $((8 * (t1 - t0) / interval))
+}
+
 mac_get()
 {
 	local if_name=$1
-- 
cgit v1.2.3-59-g8ed1b


From 4121d9479b24c20ff7d3e7b62a7cf330fae7b2af Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@mellanox.com>
Date: Fri, 24 Jan 2020 15:23:16 +0200
Subject: selftests: forwarding: lib: Add helpers for busywaiting

The function busywait() is handy as a safety-latched variant of a while
loop. Many selftests deal specifically with counter values, and busywaiting
on them is likely to be rather common (it is not quite common now, but
busywait() has not been around for very long). To facilitate expressing
simply what is tested, introduce two helpers:

- until_counter_is(), which can be used as a predicate passed to
  busywait(), which holds when expression, which is itself passed as an
  argument to until_counter_is(), reaches a desired value.

- busywait_for_counter(), which is useful for waiting until a given counter
  changes "by" (as opposed to "to") a certain amount.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/net/forwarding/lib.sh | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/tools/testing/selftests/net/forwarding/lib.sh b/tools/testing/selftests/net/forwarding/lib.sh
index 10cdfc0adca8..096340f064db 100644
--- a/tools/testing/selftests/net/forwarding/lib.sh
+++ b/tools/testing/selftests/net/forwarding/lib.sh
@@ -248,6 +248,24 @@ busywait()
 	done
 }
 
+until_counter_is()
+{
+	local value=$1; shift
+	local current=$("$@")
+
+	echo $((current))
+	((current >= value))
+}
+
+busywait_for_counter()
+{
+	local timeout=$1; shift
+	local delta=$1; shift
+
+	local base=$("$@")
+	busywait "$timeout" until_counter_is $((base + delta)) "$@"
+}
+
 setup_wait_dev()
 {
 	local dev=$1; shift
-- 
cgit v1.2.3-59-g8ed1b


From c143139b7fd36cb66afb4423ac14a1efd8a45609 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@mellanox.com>
Date: Fri, 24 Jan 2020 15:23:17 +0200
Subject: selftests: forwarding: lib: Allow reading TC rule byte counters

The function tc_rule_stats_get() fetches a packet counter of a given TC
rule. Extend it to support byte counters as well by adding an optional
argument with selector.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/net/forwarding/lib.sh | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/net/forwarding/lib.sh b/tools/testing/selftests/net/forwarding/lib.sh
index 096340f064db..2f5da414aaa7 100644
--- a/tools/testing/selftests/net/forwarding/lib.sh
+++ b/tools/testing/selftests/net/forwarding/lib.sh
@@ -593,9 +593,10 @@ tc_rule_stats_get()
 	local dev=$1; shift
 	local pref=$1; shift
 	local dir=$1; shift
+	local selector=${1:-.packets}; shift
 
 	tc -j -s filter show dev $dev ${dir:-ingress} pref $pref \
-	    | jq '.[1].options.actions[].stats.packets'
+	    | jq ".[1].options.actions[].stats$selector"
 }
 
 ethtool_stats_get()
-- 
cgit v1.2.3-59-g8ed1b


From e814c58d454d952778fbb70bd0e71564f8c49f28 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@mellanox.com>
Date: Fri, 24 Jan 2020 15:23:18 +0200
Subject: selftests: mlxsw: Add a TBF selftest

Add a test that runs traffic across a port throttled with TBF. The test
checks that the observed throughput is within +-5% from the installed
shaper.

To allow checking both the software datapath and the offloaded one, make
the test suitable for inclusion from driver-specific wrapper. Introduce
such wrappers for mlxsw.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../selftests/drivers/net/mlxsw/sch_tbf_ets.sh     |   9 +
 .../selftests/drivers/net/mlxsw/sch_tbf_prio.sh    |   9 +
 .../selftests/drivers/net/mlxsw/sch_tbf_root.sh    |   9 +
 .../selftests/net/forwarding/sch_tbf_core.sh       | 233 +++++++++++++++++++++
 .../selftests/net/forwarding/sch_tbf_ets.sh        |   6 +
 .../selftests/net/forwarding/sch_tbf_etsprio.sh    |  39 ++++
 .../selftests/net/forwarding/sch_tbf_prio.sh       |   6 +
 .../selftests/net/forwarding/sch_tbf_root.sh       |  33 +++
 8 files changed, 344 insertions(+)
 create mode 100755 tools/testing/selftests/drivers/net/mlxsw/sch_tbf_ets.sh
 create mode 100755 tools/testing/selftests/drivers/net/mlxsw/sch_tbf_prio.sh
 create mode 100755 tools/testing/selftests/drivers/net/mlxsw/sch_tbf_root.sh
 create mode 100644 tools/testing/selftests/net/forwarding/sch_tbf_core.sh
 create mode 100755 tools/testing/selftests/net/forwarding/sch_tbf_ets.sh
 create mode 100644 tools/testing/selftests/net/forwarding/sch_tbf_etsprio.sh
 create mode 100755 tools/testing/selftests/net/forwarding/sch_tbf_prio.sh
 create mode 100755 tools/testing/selftests/net/forwarding/sch_tbf_root.sh

diff --git a/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_ets.sh b/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_ets.sh
new file mode 100755
index 000000000000..c6ce0b448bf3
--- /dev/null
+++ b/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_ets.sh
@@ -0,0 +1,9 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+source qos_lib.sh
+bail_on_lldpad
+
+lib_dir=$(dirname $0)/../../../net/forwarding
+TCFLAGS=skip_sw
+source $lib_dir/sch_tbf_ets.sh
diff --git a/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_prio.sh b/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_prio.sh
new file mode 100755
index 000000000000..8d245f331619
--- /dev/null
+++ b/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_prio.sh
@@ -0,0 +1,9 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+source qos_lib.sh
+bail_on_lldpad
+
+lib_dir=$(dirname $0)/../../../net/forwarding
+TCFLAGS=skip_sw
+source $lib_dir/sch_tbf_prio.sh
diff --git a/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_root.sh b/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_root.sh
new file mode 100755
index 000000000000..013886061f15
--- /dev/null
+++ b/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_root.sh
@@ -0,0 +1,9 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+source qos_lib.sh
+bail_on_lldpad
+
+lib_dir=$(dirname $0)/../../../net/forwarding
+TCFLAGS=skip_sw
+source $lib_dir/sch_tbf_root.sh
diff --git a/tools/testing/selftests/net/forwarding/sch_tbf_core.sh b/tools/testing/selftests/net/forwarding/sch_tbf_core.sh
new file mode 100644
index 000000000000..d1f26cb7cd73
--- /dev/null
+++ b/tools/testing/selftests/net/forwarding/sch_tbf_core.sh
@@ -0,0 +1,233 @@
+# SPDX-License-Identifier: GPL-2.0
+
+# This test sends a stream of traffic from H1 through a switch, to H2. On the
+# egress port from the switch ($swp2), a shaper is installed. The test verifies
+# that the rates on the port match the configured shaper.
+#
+# In order to test per-class shaping, $swp2 actually contains TBF under PRIO or
+# ETS, with two different configurations. Traffic is prioritized using 802.1p.
+#
+# +-------------------------------------------+
+# | H1                                        |
+# |     + $h1.10                  $h1.11 +    |
+# |     | 192.0.2.1/28     192.0.2.17/28 |    |
+# |     |                                |    |
+# |     \______________    _____________/     |
+# |                    \ /                    |
+# |                     + $h1                 |
+# +---------------------|---------------------+
+#                       |
+# +---------------------|---------------------+
+# | SW                  + $swp1               |
+# |     _______________/ \_______________     |
+# |    /                                 \    |
+# |  +-|--------------+   +--------------|-+  |
+# |  | + $swp1.10     |   |     $swp1.11 + |  |
+# |  |                |   |                |  |
+# |  |     BR10       |   |       BR11     |  |
+# |  |                |   |                |  |
+# |  | + $swp2.10     |   |     $swp2.11 + |  |
+# |  +-|--------------+   +--------------|-+  |
+# |    \_______________   ______________/     |
+# |                    \ /                    |
+# |                     + $swp2               |
+# +---------------------|---------------------+
+#                       |
+# +---------------------|---------------------+
+# | H2                  + $h2                 |
+# |      ______________/ \______________      |
+# |     /                               \     |
+# |     |                               |     |
+# |     + $h2.10                 $h2.11 +     |
+# |       192.0.2.2/28    192.0.2.18/28       |
+# +-------------------------------------------+
+
+NUM_NETIFS=4
+CHECK_TC="yes"
+source $lib_dir/lib.sh
+
+ipaddr()
+{
+	local host=$1; shift
+	local vlan=$1; shift
+
+	echo 192.0.2.$((16 * (vlan - 10) + host))
+}
+
+host_create()
+{
+	local dev=$1; shift
+	local host=$1; shift
+
+	simple_if_init $dev
+	mtu_set $dev 10000
+
+	vlan_create $dev 10 v$dev $(ipaddr $host 10)/28
+	ip link set dev $dev.10 type vlan egress 0:0
+
+	vlan_create $dev 11 v$dev $(ipaddr $host 11)/28
+	ip link set dev $dev.11 type vlan egress 0:1
+}
+
+host_destroy()
+{
+	local dev=$1; shift
+
+	vlan_destroy $dev 11
+	vlan_destroy $dev 10
+	mtu_restore $dev
+	simple_if_fini $dev
+}
+
+h1_create()
+{
+	host_create $h1 1
+}
+
+h1_destroy()
+{
+	host_destroy $h1
+}
+
+h2_create()
+{
+	host_create $h2 2
+
+	tc qdisc add dev $h2 clsact
+	tc filter add dev $h2 ingress pref 1010 prot 802.1q \
+	   flower $TCFLAGS vlan_id 10 action pass
+	tc filter add dev $h2 ingress pref 1011 prot 802.1q \
+	   flower $TCFLAGS vlan_id 11 action pass
+}
+
+h2_destroy()
+{
+	tc qdisc del dev $h2 clsact
+	host_destroy $h2
+}
+
+switch_create()
+{
+	local intf
+	local vlan
+
+	ip link add dev br10 type bridge
+	ip link add dev br11 type bridge
+
+	for intf in $swp1 $swp2; do
+		ip link set dev $intf up
+		mtu_set $intf 10000
+
+		for vlan in 10 11; do
+			vlan_create $intf $vlan
+			ip link set dev $intf.$vlan master br$vlan
+			ip link set dev $intf.$vlan up
+		done
+	done
+
+	for vlan in 10 11; do
+		ip link set dev $swp1.$vlan type vlan ingress 0:0 1:1
+	done
+
+	ip link set dev br10 up
+	ip link set dev br11 up
+}
+
+switch_destroy()
+{
+	local intf
+	local vlan
+
+	# A test may have been interrupted mid-run, with Qdisc installed. Delete
+	# it here.
+	tc qdisc del dev $swp2 root 2>/dev/null
+
+	ip link set dev br11 down
+	ip link set dev br10 down
+
+	for intf in $swp2 $swp1; do
+		for vlan in 11 10; do
+			ip link set dev $intf.$vlan down
+			ip link set dev $intf.$vlan nomaster
+			vlan_destroy $intf $vlan
+		done
+
+		mtu_restore $intf
+		ip link set dev $intf down
+	done
+
+	ip link del dev br11
+	ip link del dev br10
+}
+
+setup_prepare()
+{
+	h1=${NETIFS[p1]}
+	swp1=${NETIFS[p2]}
+
+	swp2=${NETIFS[p3]}
+	h2=${NETIFS[p4]}
+
+	swp3=${NETIFS[p5]}
+	h3=${NETIFS[p6]}
+
+	swp4=${NETIFS[p7]}
+	swp5=${NETIFS[p8]}
+
+	h2_mac=$(mac_get $h2)
+
+	vrf_prepare
+
+	h1_create
+	h2_create
+	switch_create
+}
+
+cleanup()
+{
+	pre_cleanup
+
+	switch_destroy
+	h2_destroy
+	h1_destroy
+
+	vrf_cleanup
+}
+
+ping_ipv4()
+{
+	ping_test $h1.10 $(ipaddr 2 10) " vlan 10"
+	ping_test $h1.11 $(ipaddr 2 11) " vlan 11"
+}
+
+tbf_get_counter()
+{
+	local vlan=$1; shift
+
+	tc_rule_stats_get $h2 10$vlan ingress .bytes
+}
+
+do_tbf_test()
+{
+	local vlan=$1; shift
+	local mbit=$1; shift
+
+	start_traffic $h1.$vlan $(ipaddr 1 $vlan) $(ipaddr 2 $vlan) $h2_mac
+	sleep 5 # Wait for the burst to dwindle
+
+	local t2=$(busywait_for_counter 1000 +1 tbf_get_counter $vlan)
+	sleep 10
+	local t3=$(tbf_get_counter $vlan)
+	stop_traffic
+
+	RET=0
+
+	# Note: TBF uses 10^6 Mbits, not 2^20 ones.
+	local er=$((mbit * 1000 * 1000))
+	local nr=$(rate $t2 $t3 10)
+	local nr_pct=$((100 * (nr - er) / er))
+	((-5 <= nr_pct && nr_pct <= 5))
+	check_err $? "Expected rate $(humanize $er), got $(humanize $nr), which is $nr_pct% off. Required accuracy is +-5%."
+
+	log_test "TC $((vlan - 10)): TBF rate ${mbit}Mbit"
+}
diff --git a/tools/testing/selftests/net/forwarding/sch_tbf_ets.sh b/tools/testing/selftests/net/forwarding/sch_tbf_ets.sh
new file mode 100755
index 000000000000..84fb6cab88e4
--- /dev/null
+++ b/tools/testing/selftests/net/forwarding/sch_tbf_ets.sh
@@ -0,0 +1,6 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+QDISC="ets strict"
+: ${lib_dir:=.}
+source $lib_dir/sch_tbf_etsprio.sh
diff --git a/tools/testing/selftests/net/forwarding/sch_tbf_etsprio.sh b/tools/testing/selftests/net/forwarding/sch_tbf_etsprio.sh
new file mode 100644
index 000000000000..8bd85da1905a
--- /dev/null
+++ b/tools/testing/selftests/net/forwarding/sch_tbf_etsprio.sh
@@ -0,0 +1,39 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+ALL_TESTS="
+	ping_ipv4
+	tbf_test
+"
+source $lib_dir/sch_tbf_core.sh
+
+tbf_test_one()
+{
+	local bs=$1; shift
+
+	tc qdisc replace dev $swp2 parent 10:3 handle 103: tbf \
+	   rate 400Mbit burst $bs limit 1M
+	tc qdisc replace dev $swp2 parent 10:2 handle 102: tbf \
+	   rate 800Mbit burst $bs limit 1M
+
+	do_tbf_test 10 400 $bs
+	do_tbf_test 11 800 $bs
+}
+
+tbf_test()
+{
+	# This test is used for both ETS and PRIO. Even though we only need two
+	# bands, PRIO demands a minimum of three.
+	tc qdisc add dev $swp2 root handle 10: $QDISC 3 priomap 2 1 0
+	tbf_test_one 128K
+	tc qdisc del dev $swp2 root
+}
+
+trap cleanup EXIT
+
+setup_prepare
+setup_wait
+
+tests_run
+
+exit $EXIT_STATUS
diff --git a/tools/testing/selftests/net/forwarding/sch_tbf_prio.sh b/tools/testing/selftests/net/forwarding/sch_tbf_prio.sh
new file mode 100755
index 000000000000..9c8cb1cb9ba4
--- /dev/null
+++ b/tools/testing/selftests/net/forwarding/sch_tbf_prio.sh
@@ -0,0 +1,6 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+QDISC="prio bands"
+: ${lib_dir:=.}
+source $lib_dir/sch_tbf_etsprio.sh
diff --git a/tools/testing/selftests/net/forwarding/sch_tbf_root.sh b/tools/testing/selftests/net/forwarding/sch_tbf_root.sh
new file mode 100755
index 000000000000..72aa21ba88c7
--- /dev/null
+++ b/tools/testing/selftests/net/forwarding/sch_tbf_root.sh
@@ -0,0 +1,33 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+ALL_TESTS="
+	ping_ipv4
+	tbf_test
+"
+: ${lib_dir:=.}
+source $lib_dir/sch_tbf_core.sh
+
+tbf_test_one()
+{
+	local bs=$1; shift
+
+	tc qdisc replace dev $swp2 root handle 108: tbf \
+	   rate 400Mbit burst $bs limit 1M
+	do_tbf_test 10 400 $bs
+}
+
+tbf_test()
+{
+	tbf_test_one 128K
+	tc qdisc del dev $swp2 root
+}
+
+trap cleanup EXIT
+
+setup_prepare
+setup_wait
+
+tests_run
+
+exit $EXIT_STATUS
-- 
cgit v1.2.3-59-g8ed1b


From cc974003615afa044fa62c7520ae690091fa684a Mon Sep 17 00:00:00 2001
From: Marcel Holtmann <marcel@holtmann.org>
Date: Sat, 25 Jan 2020 09:23:47 +0100
Subject: Bluetooth: Add missing checks for HCI_ISODATA_PKT packet type

The checks for HCI_ISODATA_PKT packet type are required in a few
additional locations to allow sending/receiving of this new packet type.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
---
 net/bluetooth/hci_core.c |  4 +++-
 net/bluetooth/hci_sock.c | 12 ++++++++----
 2 files changed, 11 insertions(+), 5 deletions(-)

diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c
index 1ca7508b6ca7..cbbc34a006d1 100644
--- a/net/bluetooth/hci_core.c
+++ b/net/bluetooth/hci_core.c
@@ -3565,7 +3565,8 @@ int hci_recv_frame(struct hci_dev *hdev, struct sk_buff *skb)
 
 	if (hci_skb_pkt_type(skb) != HCI_EVENT_PKT &&
 	    hci_skb_pkt_type(skb) != HCI_ACLDATA_PKT &&
-	    hci_skb_pkt_type(skb) != HCI_SCODATA_PKT) {
+	    hci_skb_pkt_type(skb) != HCI_SCODATA_PKT &&
+	    hci_skb_pkt_type(skb) != HCI_ISODATA_PKT) {
 		kfree_skb(skb);
 		return -EINVAL;
 	}
@@ -4543,6 +4544,7 @@ static void hci_rx_work(struct work_struct *work)
 			switch (hci_skb_pkt_type(skb)) {
 			case HCI_ACLDATA_PKT:
 			case HCI_SCODATA_PKT:
+			case HCI_ISODATA_PKT:
 				kfree_skb(skb);
 				continue;
 			}
diff --git a/net/bluetooth/hci_sock.c b/net/bluetooth/hci_sock.c
index 3ae508674ef7..5756b124c2cb 100644
--- a/net/bluetooth/hci_sock.c
+++ b/net/bluetooth/hci_sock.c
@@ -211,7 +211,8 @@ void hci_send_to_sock(struct hci_dev *hdev, struct sk_buff *skb)
 			if (hci_skb_pkt_type(skb) != HCI_COMMAND_PKT &&
 			    hci_skb_pkt_type(skb) != HCI_EVENT_PKT &&
 			    hci_skb_pkt_type(skb) != HCI_ACLDATA_PKT &&
-			    hci_skb_pkt_type(skb) != HCI_SCODATA_PKT)
+			    hci_skb_pkt_type(skb) != HCI_SCODATA_PKT &&
+			    hci_skb_pkt_type(skb) != HCI_ISODATA_PKT)
 				continue;
 			if (is_filtered_packet(sk, skb))
 				continue;
@@ -220,7 +221,8 @@ void hci_send_to_sock(struct hci_dev *hdev, struct sk_buff *skb)
 				continue;
 			if (hci_skb_pkt_type(skb) != HCI_EVENT_PKT &&
 			    hci_skb_pkt_type(skb) != HCI_ACLDATA_PKT &&
-			    hci_skb_pkt_type(skb) != HCI_SCODATA_PKT)
+			    hci_skb_pkt_type(skb) != HCI_SCODATA_PKT &&
+			    hci_skb_pkt_type(skb) != HCI_ISODATA_PKT)
 				continue;
 		} else {
 			/* Don't send frame to other channel types */
@@ -1768,7 +1770,8 @@ static int hci_sock_sendmsg(struct socket *sock, struct msghdr *msg,
 		 */
 		if (hci_skb_pkt_type(skb) != HCI_COMMAND_PKT &&
 		    hci_skb_pkt_type(skb) != HCI_ACLDATA_PKT &&
-		    hci_skb_pkt_type(skb) != HCI_SCODATA_PKT) {
+		    hci_skb_pkt_type(skb) != HCI_SCODATA_PKT &&
+		    hci_skb_pkt_type(skb) != HCI_ISODATA_PKT) {
 			err = -EINVAL;
 			goto drop;
 		}
@@ -1812,7 +1815,8 @@ static int hci_sock_sendmsg(struct socket *sock, struct msghdr *msg,
 		}
 
 		if (hci_skb_pkt_type(skb) != HCI_ACLDATA_PKT &&
-		    hci_skb_pkt_type(skb) != HCI_SCODATA_PKT) {
+		    hci_skb_pkt_type(skb) != HCI_SCODATA_PKT &&
+		    hci_skb_pkt_type(skb) != HCI_ISODATA_PKT) {
 			err = -EINVAL;
 			goto drop;
 		}
-- 
cgit v1.2.3-59-g8ed1b


From 18f81241b74fb49d576c83fbbab9a0b6e3bb20d4 Mon Sep 17 00:00:00 2001
From: Marcel Holtmann <marcel@holtmann.org>
Date: Sat, 25 Jan 2020 09:19:51 +0100
Subject: Bluetooth: Move {min,max}_key_size debugfs into hci_debugfs_create_le

The debugfs entries for {min,max}_key_size are created during SMP
registration and thus it might lead to multiple attempts to create the
same entries. Avoid this by moving them to the LE controller init
section.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
---
 net/bluetooth/hci_debugfs.c | 61 +++++++++++++++++++++++++++++
 net/bluetooth/smp.c         | 93 ---------------------------------------------
 2 files changed, 61 insertions(+), 93 deletions(-)

diff --git a/net/bluetooth/hci_debugfs.c b/net/bluetooth/hci_debugfs.c
index 1c8100bc4e04..6b1314c738b8 100644
--- a/net/bluetooth/hci_debugfs.c
+++ b/net/bluetooth/hci_debugfs.c
@@ -26,6 +26,7 @@
 #include <net/bluetooth/bluetooth.h>
 #include <net/bluetooth/hci_core.h>
 
+#include "smp.h"
 #include "hci_debugfs.h"
 
 #define DEFINE_QUIRK_ATTRIBUTE(__name, __quirk)				      \
@@ -989,6 +990,62 @@ static int adv_max_interval_get(void *data, u64 *val)
 DEFINE_SIMPLE_ATTRIBUTE(adv_max_interval_fops, adv_max_interval_get,
 			adv_max_interval_set, "%llu\n");
 
+static int min_key_size_set(void *data, u64 val)
+{
+	struct hci_dev *hdev = data;
+
+	if (val > hdev->le_max_key_size || val < SMP_MIN_ENC_KEY_SIZE)
+		return -EINVAL;
+
+	hci_dev_lock(hdev);
+	hdev->le_min_key_size = val;
+	hci_dev_unlock(hdev);
+
+	return 0;
+}
+
+static int min_key_size_get(void *data, u64 *val)
+{
+	struct hci_dev *hdev = data;
+
+	hci_dev_lock(hdev);
+	*val = hdev->le_min_key_size;
+	hci_dev_unlock(hdev);
+
+	return 0;
+}
+
+DEFINE_SIMPLE_ATTRIBUTE(min_key_size_fops, min_key_size_get,
+			min_key_size_set, "%llu\n");
+
+static int max_key_size_set(void *data, u64 val)
+{
+	struct hci_dev *hdev = data;
+
+	if (val > SMP_MAX_ENC_KEY_SIZE || val < hdev->le_min_key_size)
+		return -EINVAL;
+
+	hci_dev_lock(hdev);
+	hdev->le_max_key_size = val;
+	hci_dev_unlock(hdev);
+
+	return 0;
+}
+
+static int max_key_size_get(void *data, u64 *val)
+{
+	struct hci_dev *hdev = data;
+
+	hci_dev_lock(hdev);
+	*val = hdev->le_max_key_size;
+	hci_dev_unlock(hdev);
+
+	return 0;
+}
+
+DEFINE_SIMPLE_ATTRIBUTE(max_key_size_fops, max_key_size_get,
+			max_key_size_set, "%llu\n");
+
 static int auth_payload_timeout_set(void *data, u64 val)
 {
 	struct hci_dev *hdev = data;
@@ -1071,6 +1128,10 @@ void hci_debugfs_create_le(struct hci_dev *hdev)
 			    &adv_max_interval_fops);
 	debugfs_create_u16("discov_interleaved_timeout", 0644, hdev->debugfs,
 			   &hdev->discov_interleaved_timeout);
+	debugfs_create_file("min_key_size", 0644, hdev->debugfs, hdev,
+			    &min_key_size_fops);
+	debugfs_create_file("max_key_size", 0644, hdev->debugfs, hdev,
+			    &max_key_size_fops);
 	debugfs_create_file("auth_payload_timeout", 0644, hdev->debugfs, hdev,
 			    &auth_payload_timeout_fops);
 
diff --git a/net/bluetooth/smp.c b/net/bluetooth/smp.c
index 4ece170c518e..204f14f8b507 100644
--- a/net/bluetooth/smp.c
+++ b/net/bluetooth/smp.c
@@ -3373,94 +3373,6 @@ static const struct file_operations force_bredr_smp_fops = {
 	.llseek		= default_llseek,
 };
 
-static ssize_t le_min_key_size_read(struct file *file,
-				     char __user *user_buf,
-				     size_t count, loff_t *ppos)
-{
-	struct hci_dev *hdev = file->private_data;
-	char buf[4];
-
-	snprintf(buf, sizeof(buf), "%2u\n", hdev->le_min_key_size);
-
-	return simple_read_from_buffer(user_buf, count, ppos, buf, strlen(buf));
-}
-
-static ssize_t le_min_key_size_write(struct file *file,
-				      const char __user *user_buf,
-				      size_t count, loff_t *ppos)
-{
-	struct hci_dev *hdev = file->private_data;
-	char buf[32];
-	size_t buf_size = min(count, (sizeof(buf) - 1));
-	u8 key_size;
-
-	if (copy_from_user(buf, user_buf, buf_size))
-		return -EFAULT;
-
-	buf[buf_size] = '\0';
-
-	sscanf(buf, "%hhu", &key_size);
-
-	if (key_size > hdev->le_max_key_size ||
-	    key_size < SMP_MIN_ENC_KEY_SIZE)
-		return -EINVAL;
-
-	hdev->le_min_key_size = key_size;
-
-	return count;
-}
-
-static const struct file_operations le_min_key_size_fops = {
-	.open		= simple_open,
-	.read		= le_min_key_size_read,
-	.write		= le_min_key_size_write,
-	.llseek		= default_llseek,
-};
-
-static ssize_t le_max_key_size_read(struct file *file,
-				     char __user *user_buf,
-				     size_t count, loff_t *ppos)
-{
-	struct hci_dev *hdev = file->private_data;
-	char buf[4];
-
-	snprintf(buf, sizeof(buf), "%2u\n", hdev->le_max_key_size);
-
-	return simple_read_from_buffer(user_buf, count, ppos, buf, strlen(buf));
-}
-
-static ssize_t le_max_key_size_write(struct file *file,
-				      const char __user *user_buf,
-				      size_t count, loff_t *ppos)
-{
-	struct hci_dev *hdev = file->private_data;
-	char buf[32];
-	size_t buf_size = min(count, (sizeof(buf) - 1));
-	u8 key_size;
-
-	if (copy_from_user(buf, user_buf, buf_size))
-		return -EFAULT;
-
-	buf[buf_size] = '\0';
-
-	sscanf(buf, "%hhu", &key_size);
-
-	if (key_size > SMP_MAX_ENC_KEY_SIZE ||
-	    key_size < hdev->le_min_key_size)
-		return -EINVAL;
-
-	hdev->le_max_key_size = key_size;
-
-	return count;
-}
-
-static const struct file_operations le_max_key_size_fops = {
-	.open		= simple_open,
-	.read		= le_max_key_size_read,
-	.write		= le_max_key_size_write,
-	.llseek		= default_llseek,
-};
-
 int smp_register(struct hci_dev *hdev)
 {
 	struct l2cap_chan *chan;
@@ -3485,11 +3397,6 @@ int smp_register(struct hci_dev *hdev)
 
 	hdev->smp_data = chan;
 
-	debugfs_create_file("le_min_key_size", 0644, hdev->debugfs, hdev,
-			    &le_min_key_size_fops);
-	debugfs_create_file("le_max_key_size", 0644, hdev->debugfs, hdev,
-			    &le_max_key_size_fops);
-
 	/* If the controller does not support BR/EDR Secure Connections
 	 * feature, then the BR/EDR SMP channel shall not be present.
 	 *
-- 
cgit v1.2.3-59-g8ed1b


From 84ad7a7ab69f112c0c4b878c9be91b950a1fb1f8 Mon Sep 17 00:00:00 2001
From: Jiri Olsa <jolsa@kernel.org>
Date: Thu, 23 Jan 2020 17:15:06 +0100
Subject: bpf: Allow BTF ctx access for string pointers

When accessing the context we allow access to arguments with
scalar type and pointer to struct. But we deny access for
pointer to scalar type, which is the case for many functions.

Alexei suggested to take conservative approach and allow
currently only string pointer access, which is the case
for most functions now:

Adding check if the pointer is to string type and allow access to it.

Suggested-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200123161508.915203-2-jolsa@kernel.org
---
 kernel/bpf/btf.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 32963b6d5a9c..b7c1660fb594 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -3669,6 +3669,19 @@ struct btf *bpf_prog_get_target_btf(const struct bpf_prog *prog)
 	}
 }
 
+static bool is_string_ptr(struct btf *btf, const struct btf_type *t)
+{
+	/* t comes in already as a pointer */
+	t = btf_type_by_id(btf, t->type);
+
+	/* allow const */
+	if (BTF_INFO_KIND(t->info) == BTF_KIND_CONST)
+		t = btf_type_by_id(btf, t->type);
+
+	/* char, signed char, unsigned char */
+	return btf_type_is_int(t) && t->size == 1;
+}
+
 bool btf_ctx_access(int off, int size, enum bpf_access_type type,
 		    const struct bpf_prog *prog,
 		    struct bpf_insn_access_aux *info)
@@ -3735,6 +3748,9 @@ bool btf_ctx_access(int off, int size, enum bpf_access_type type,
 		 */
 		return true;
 
+	if (is_string_ptr(btf, t))
+		return true;
+
 	/* this is a pointer to another type */
 	info->reg_type = PTR_TO_BTF_ID;
 
-- 
cgit v1.2.3-59-g8ed1b


From e9b4e606c2289d6610113253922bb8c9ac7f68b0 Mon Sep 17 00:00:00 2001
From: Jiri Olsa <jolsa@kernel.org>
Date: Thu, 23 Jan 2020 17:15:07 +0100
Subject: bpf: Allow to resolve bpf trampoline and dispatcher in unwind

When unwinding the stack we need to identify each address
to successfully continue. Adding latch tree to keep trampolines
for quick lookup during the unwind.

The patch uses first 48 bytes for latch tree node, leaving 4048
bytes from the rest of the page for trampoline or dispatcher
generated code.

It's still enough not to affect trampoline and dispatcher progs
maximum counts.

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200123161508.915203-3-jolsa@kernel.org
---
 include/linux/bpf.h     | 12 +++++++-
 kernel/bpf/dispatcher.c |  4 +--
 kernel/bpf/trampoline.c | 80 ++++++++++++++++++++++++++++++++++++++++++++-----
 kernel/extable.c        |  7 +++--
 4 files changed, 90 insertions(+), 13 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index a9687861fd7e..8e9ad3943cd9 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -525,7 +525,6 @@ struct bpf_trampoline *bpf_trampoline_lookup(u64 key);
 int bpf_trampoline_link_prog(struct bpf_prog *prog);
 int bpf_trampoline_unlink_prog(struct bpf_prog *prog);
 void bpf_trampoline_put(struct bpf_trampoline *tr);
-void *bpf_jit_alloc_exec_page(void);
 #define BPF_DISPATCHER_INIT(name) {			\
 	.mutex = __MUTEX_INITIALIZER(name.mutex),	\
 	.func = &name##func,				\
@@ -557,6 +556,13 @@ void *bpf_jit_alloc_exec_page(void);
 #define BPF_DISPATCHER_PTR(name) (&name)
 void bpf_dispatcher_change_prog(struct bpf_dispatcher *d, struct bpf_prog *from,
 				struct bpf_prog *to);
+struct bpf_image {
+	struct latch_tree_node tnode;
+	unsigned char data[];
+};
+#define BPF_IMAGE_SIZE (PAGE_SIZE - sizeof(struct bpf_image))
+bool is_bpf_image_address(unsigned long address);
+void *bpf_image_alloc(void);
 #else
 static inline struct bpf_trampoline *bpf_trampoline_lookup(u64 key)
 {
@@ -578,6 +584,10 @@ static inline void bpf_trampoline_put(struct bpf_trampoline *tr) {}
 static inline void bpf_dispatcher_change_prog(struct bpf_dispatcher *d,
 					      struct bpf_prog *from,
 					      struct bpf_prog *to) {}
+static inline bool is_bpf_image_address(unsigned long address)
+{
+	return false;
+}
 #endif
 
 struct bpf_func_info_aux {
diff --git a/kernel/bpf/dispatcher.c b/kernel/bpf/dispatcher.c
index 204ee61a3904..b3e5b214fed8 100644
--- a/kernel/bpf/dispatcher.c
+++ b/kernel/bpf/dispatcher.c
@@ -113,7 +113,7 @@ static void bpf_dispatcher_update(struct bpf_dispatcher *d, int prev_num_progs)
 		noff = 0;
 	} else {
 		old = d->image + d->image_off;
-		noff = d->image_off ^ (PAGE_SIZE / 2);
+		noff = d->image_off ^ (BPF_IMAGE_SIZE / 2);
 	}
 
 	new = d->num_progs ? d->image + noff : NULL;
@@ -140,7 +140,7 @@ void bpf_dispatcher_change_prog(struct bpf_dispatcher *d, struct bpf_prog *from,
 
 	mutex_lock(&d->mutex);
 	if (!d->image) {
-		d->image = bpf_jit_alloc_exec_page();
+		d->image = bpf_image_alloc();
 		if (!d->image)
 			goto out;
 	}
diff --git a/kernel/bpf/trampoline.c b/kernel/bpf/trampoline.c
index eb64c245052b..6b264a92064b 100644
--- a/kernel/bpf/trampoline.c
+++ b/kernel/bpf/trampoline.c
@@ -4,6 +4,7 @@
 #include <linux/bpf.h>
 #include <linux/filter.h>
 #include <linux/ftrace.h>
+#include <linux/rbtree_latch.h>
 
 /* dummy _ops. The verifier will operate on target program's ops. */
 const struct bpf_verifier_ops bpf_extension_verifier_ops = {
@@ -16,11 +17,12 @@ const struct bpf_prog_ops bpf_extension_prog_ops = {
 #define TRAMPOLINE_TABLE_SIZE (1 << TRAMPOLINE_HASH_BITS)
 
 static struct hlist_head trampoline_table[TRAMPOLINE_TABLE_SIZE];
+static struct latch_tree_root image_tree __cacheline_aligned;
 
-/* serializes access to trampoline_table */
+/* serializes access to trampoline_table and image_tree */
 static DEFINE_MUTEX(trampoline_mutex);
 
-void *bpf_jit_alloc_exec_page(void)
+static void *bpf_jit_alloc_exec_page(void)
 {
 	void *image;
 
@@ -36,6 +38,64 @@ void *bpf_jit_alloc_exec_page(void)
 	return image;
 }
 
+static __always_inline bool image_tree_less(struct latch_tree_node *a,
+				      struct latch_tree_node *b)
+{
+	struct bpf_image *ia = container_of(a, struct bpf_image, tnode);
+	struct bpf_image *ib = container_of(b, struct bpf_image, tnode);
+
+	return ia < ib;
+}
+
+static __always_inline int image_tree_comp(void *addr, struct latch_tree_node *n)
+{
+	void *image = container_of(n, struct bpf_image, tnode);
+
+	if (addr < image)
+		return -1;
+	if (addr >= image + PAGE_SIZE)
+		return 1;
+
+	return 0;
+}
+
+static const struct latch_tree_ops image_tree_ops = {
+	.less	= image_tree_less,
+	.comp	= image_tree_comp,
+};
+
+static void *__bpf_image_alloc(bool lock)
+{
+	struct bpf_image *image;
+
+	image = bpf_jit_alloc_exec_page();
+	if (!image)
+		return NULL;
+
+	if (lock)
+		mutex_lock(&trampoline_mutex);
+	latch_tree_insert(&image->tnode, &image_tree, &image_tree_ops);
+	if (lock)
+		mutex_unlock(&trampoline_mutex);
+	return image->data;
+}
+
+void *bpf_image_alloc(void)
+{
+	return __bpf_image_alloc(true);
+}
+
+bool is_bpf_image_address(unsigned long addr)
+{
+	bool ret;
+
+	rcu_read_lock();
+	ret = latch_tree_find((void *) addr, &image_tree, &image_tree_ops) != NULL;
+	rcu_read_unlock();
+
+	return ret;
+}
+
 struct bpf_trampoline *bpf_trampoline_lookup(u64 key)
 {
 	struct bpf_trampoline *tr;
@@ -56,7 +116,7 @@ struct bpf_trampoline *bpf_trampoline_lookup(u64 key)
 		goto out;
 
 	/* is_root was checked earlier. No need for bpf_jit_charge_modmem() */
-	image = bpf_jit_alloc_exec_page();
+	image = __bpf_image_alloc(false);
 	if (!image) {
 		kfree(tr);
 		tr = NULL;
@@ -131,14 +191,14 @@ static int register_fentry(struct bpf_trampoline *tr, void *new_addr)
 }
 
 /* Each call __bpf_prog_enter + call bpf_func + call __bpf_prog_exit is ~50
- * bytes on x86.  Pick a number to fit into PAGE_SIZE / 2
+ * bytes on x86.  Pick a number to fit into BPF_IMAGE_SIZE / 2
  */
 #define BPF_MAX_TRAMP_PROGS 40
 
 static int bpf_trampoline_update(struct bpf_trampoline *tr)
 {
-	void *old_image = tr->image + ((tr->selector + 1) & 1) * PAGE_SIZE/2;
-	void *new_image = tr->image + (tr->selector & 1) * PAGE_SIZE/2;
+	void *old_image = tr->image + ((tr->selector + 1) & 1) * BPF_IMAGE_SIZE/2;
+	void *new_image = tr->image + (tr->selector & 1) * BPF_IMAGE_SIZE/2;
 	struct bpf_prog *progs_to_run[BPF_MAX_TRAMP_PROGS];
 	int fentry_cnt = tr->progs_cnt[BPF_TRAMP_FENTRY];
 	int fexit_cnt = tr->progs_cnt[BPF_TRAMP_FEXIT];
@@ -174,7 +234,7 @@ static int bpf_trampoline_update(struct bpf_trampoline *tr)
 	 */
 	synchronize_rcu_tasks();
 
-	err = arch_prepare_bpf_trampoline(new_image, new_image + PAGE_SIZE / 2,
+	err = arch_prepare_bpf_trampoline(new_image, new_image + BPF_IMAGE_SIZE / 2,
 					  &tr->func.model, flags,
 					  fentry, fentry_cnt,
 					  fexit, fexit_cnt,
@@ -284,6 +344,8 @@ out:
 
 void bpf_trampoline_put(struct bpf_trampoline *tr)
 {
+	struct bpf_image *image;
+
 	if (!tr)
 		return;
 	mutex_lock(&trampoline_mutex);
@@ -294,9 +356,11 @@ void bpf_trampoline_put(struct bpf_trampoline *tr)
 		goto out;
 	if (WARN_ON_ONCE(!hlist_empty(&tr->progs_hlist[BPF_TRAMP_FEXIT])))
 		goto out;
+	image = container_of(tr->image, struct bpf_image, data);
+	latch_tree_erase(&image->tnode, &image_tree, &image_tree_ops);
 	/* wait for tasks to get out of trampoline before freeing it */
 	synchronize_rcu_tasks();
-	bpf_jit_free_exec(tr->image);
+	bpf_jit_free_exec(image);
 	hlist_del(&tr->hlist);
 	kfree(tr);
 out:
diff --git a/kernel/extable.c b/kernel/extable.c
index f6920a11e28a..a0024f27d3a1 100644
--- a/kernel/extable.c
+++ b/kernel/extable.c
@@ -131,8 +131,9 @@ int kernel_text_address(unsigned long addr)
 	 * triggers a stack trace, or a WARN() that happens during
 	 * coming back from idle, or cpu on or offlining.
 	 *
-	 * is_module_text_address() as well as the kprobe slots
-	 * and is_bpf_text_address() require RCU to be watching.
+	 * is_module_text_address() as well as the kprobe slots,
+	 * is_bpf_text_address() and is_bpf_image_address require
+	 * RCU to be watching.
 	 */
 	no_rcu = !rcu_is_watching();
 
@@ -148,6 +149,8 @@ int kernel_text_address(unsigned long addr)
 		goto out;
 	if (is_bpf_text_address(addr))
 		goto out;
+	if (is_bpf_image_address(addr))
+		goto out;
 	ret = 0;
 out:
 	if (no_rcu)
-- 
cgit v1.2.3-59-g8ed1b


From d633d57902a510debd4ec5b7a374a009c8c2d620 Mon Sep 17 00:00:00 2001
From: Jiri Olsa <jolsa@kernel.org>
Date: Thu, 23 Jan 2020 17:15:08 +0100
Subject: selftest/bpf: Add test for allowed trampolines count

There's limit of 40 programs tht can be attached
to trampoline for one function. Adding test that
tries to attach that many plus one extra that needs
to fail.

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20200123161508.915203-4-jolsa@kernel.org
---
 .../selftests/bpf/prog_tests/trampoline_count.c    | 112 +++++++++++++++++++++
 .../selftests/bpf/progs/test_trampoline_count.c    |  21 ++++
 2 files changed, 133 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/trampoline_count.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_trampoline_count.c

diff --git a/tools/testing/selftests/bpf/prog_tests/trampoline_count.c b/tools/testing/selftests/bpf/prog_tests/trampoline_count.c
new file mode 100644
index 000000000000..1235f3d1cc50
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/trampoline_count.c
@@ -0,0 +1,112 @@
+// SPDX-License-Identifier: GPL-2.0-only
+#define _GNU_SOURCE
+#include <sched.h>
+#include <sys/prctl.h>
+#include <test_progs.h>
+
+#define MAX_TRAMP_PROGS 40
+
+struct inst {
+	struct bpf_object *obj;
+	struct bpf_link   *link_fentry;
+	struct bpf_link   *link_fexit;
+};
+
+static int test_task_rename(void)
+{
+	int fd, duration = 0, err;
+	char buf[] = "test_overhead";
+
+	fd = open("/proc/self/comm", O_WRONLY|O_TRUNC);
+	if (CHECK(fd < 0, "open /proc", "err %d", errno))
+		return -1;
+	err = write(fd, buf, sizeof(buf));
+	if (err < 0) {
+		CHECK(err < 0, "task rename", "err %d", errno);
+		close(fd);
+		return -1;
+	}
+	close(fd);
+	return 0;
+}
+
+static struct bpf_link *load(struct bpf_object *obj, const char *name)
+{
+	struct bpf_program *prog;
+	int duration = 0;
+
+	prog = bpf_object__find_program_by_title(obj, name);
+	if (CHECK(!prog, "find_probe", "prog '%s' not found\n", name))
+		return ERR_PTR(-EINVAL);
+	return bpf_program__attach_trace(prog);
+}
+
+void test_trampoline_count(void)
+{
+	const char *fentry_name = "fentry/__set_task_comm";
+	const char *fexit_name = "fexit/__set_task_comm";
+	const char *object = "test_trampoline_count.o";
+	struct inst inst[MAX_TRAMP_PROGS] = { 0 };
+	int err, i = 0, duration = 0;
+	struct bpf_object *obj;
+	struct bpf_link *link;
+	char comm[16] = {};
+
+	/* attach 'allowed' 40 trampoline programs */
+	for (i = 0; i < MAX_TRAMP_PROGS; i++) {
+		obj = bpf_object__open_file(object, NULL);
+		if (CHECK(IS_ERR(obj), "obj_open_file", "err %ld\n", PTR_ERR(obj)))
+			goto cleanup;
+
+		err = bpf_object__load(obj);
+		if (CHECK(err, "obj_load", "err %d\n", err))
+			goto cleanup;
+		inst[i].obj = obj;
+
+		if (rand() % 2) {
+			link = load(obj, fentry_name);
+			if (CHECK(IS_ERR(link), "attach prog", "err %ld\n", PTR_ERR(link)))
+				goto cleanup;
+			inst[i].link_fentry = link;
+		} else {
+			link = load(obj, fexit_name);
+			if (CHECK(IS_ERR(link), "attach prog", "err %ld\n", PTR_ERR(link)))
+				goto cleanup;
+			inst[i].link_fexit = link;
+		}
+	}
+
+	/* and try 1 extra.. */
+	obj = bpf_object__open_file(object, NULL);
+	if (CHECK(IS_ERR(obj), "obj_open_file", "err %ld\n", PTR_ERR(obj)))
+		goto cleanup;
+
+	err = bpf_object__load(obj);
+	if (CHECK(err, "obj_load", "err %d\n", err))
+		goto cleanup_extra;
+
+	/* ..that needs to fail */
+	link = load(obj, fentry_name);
+	if (CHECK(!IS_ERR(link), "cannot attach over the limit", "err %ld\n", PTR_ERR(link))) {
+		bpf_link__destroy(link);
+		goto cleanup_extra;
+	}
+
+	/* with E2BIG error */
+	CHECK(PTR_ERR(link) != -E2BIG, "proper error check", "err %ld\n", PTR_ERR(link));
+
+	/* and finaly execute the probe */
+	if (CHECK_FAIL(prctl(PR_GET_NAME, comm, 0L, 0L, 0L)))
+		goto cleanup_extra;
+	CHECK_FAIL(test_task_rename());
+	CHECK_FAIL(prctl(PR_SET_NAME, comm, 0L, 0L, 0L));
+
+cleanup_extra:
+	bpf_object__close(obj);
+cleanup:
+	while (--i) {
+		bpf_link__destroy(inst[i].link_fentry);
+		bpf_link__destroy(inst[i].link_fexit);
+		bpf_object__close(inst[i].obj);
+	}
+}
diff --git a/tools/testing/selftests/bpf/progs/test_trampoline_count.c b/tools/testing/selftests/bpf/progs/test_trampoline_count.c
new file mode 100644
index 000000000000..e51e6e3a81c2
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_trampoline_count.c
@@ -0,0 +1,21 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <stdbool.h>
+#include <stddef.h>
+#include <linux/bpf.h>
+#include "bpf_trace_helpers.h"
+
+struct task_struct;
+
+SEC("fentry/__set_task_comm")
+int BPF_PROG(prog1, struct task_struct *tsk, const char *buf, bool exec)
+{
+	return 0;
+}
+
+SEC("fexit/__set_task_comm")
+int BPF_PROG(prog2, struct task_struct *tsk, const char *buf, bool exec)
+{
+	return 0;
+}
+
+char _license[] SEC("license") = "GPL";
-- 
cgit v1.2.3-59-g8ed1b


From 451f2c4406e035295169a3abe127e377af694398 Mon Sep 17 00:00:00 2001
From: Tony Nguyen <anthony.l.nguyen@intel.com>
Date: Fri, 17 Jan 2020 07:39:14 -0800
Subject: ice: Populate TCAM filter software structures

Store the TCAM entry with the profile data and the VSI group in the
respective SW structures. This will be subsequently used to write out
the tables to hardware.

Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Signed-off-by: Henry Tieman <henry.w.tieman@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_adminq_cmd.h |    1 +
 drivers/net/ethernet/intel/ice/ice_common.c     |   33 +
 drivers/net/ethernet/intel/ice/ice_common.h     |    2 +
 drivers/net/ethernet/intel/ice/ice_flex_pipe.c  | 1318 +++++++++++++++++++++++
 drivers/net/ethernet/intel/ice/ice_flex_pipe.h  |    2 +
 drivers/net/ethernet/intel/ice/ice_flex_type.h  |   51 +
 drivers/net/ethernet/intel/ice/ice_flow.c       |   61 +-
 drivers/net/ethernet/intel/ice/ice_status.h     |    1 +
 8 files changed, 1459 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
index ad15a772b609..9967973a7c42 100644
--- a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
@@ -233,6 +233,7 @@ struct ice_aqc_get_sw_cfg_resp {
 #define ICE_AQC_RES_TYPE_VSI_LIST_REP			0x03
 #define ICE_AQC_RES_TYPE_VSI_LIST_PRUNE			0x04
 #define ICE_AQC_RES_TYPE_HASH_PROF_BLDR_PROFID		0x60
+#define ICE_AQC_RES_TYPE_HASH_PROF_BLDR_TCAM		0x61
 
 #define ICE_AQC_RES_TYPE_FLAG_SCAN_BOTTOM		BIT(12)
 #define ICE_AQC_RES_TYPE_FLAG_IGNORE_INDEX		BIT(13)
diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index 8f86962fd052..0207e28c2682 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -1572,6 +1572,39 @@ ice_alloc_res_exit:
 	return status;
 }
 
+/**
+ * ice_free_hw_res - free allocated HW resource
+ * @hw: pointer to the HW struct
+ * @type: type of resource to free
+ * @num: number of resources
+ * @res: pointer to array that contains the resources to free
+ */
+enum ice_status
+ice_free_hw_res(struct ice_hw *hw, u16 type, u16 num, u16 *res)
+{
+	struct ice_aqc_alloc_free_res_elem *buf;
+	enum ice_status status;
+	u16 buf_len;
+
+	buf_len = struct_size(buf, elem, num - 1);
+	buf = kzalloc(buf_len, GFP_KERNEL);
+	if (!buf)
+		return ICE_ERR_NO_MEMORY;
+
+	/* Prepare buffer to free resource. */
+	buf->num_elems = cpu_to_le16(num);
+	buf->res_type = cpu_to_le16(type);
+	memcpy(buf->elem, res, sizeof(buf->elem) * num);
+
+	status = ice_aq_alloc_free_res(hw, num, buf, buf_len,
+				       ice_aqc_opc_free_res, NULL);
+	if (status)
+		ice_debug(hw, ICE_DBG_SW, "CQ CMD Buffer:\n");
+
+	kfree(buf);
+	return status;
+}
+
 /**
  * ice_get_num_per_func - determine number of resources per PF
  * @hw: pointer to the HW structure
diff --git a/drivers/net/ethernet/intel/ice/ice_common.h b/drivers/net/ethernet/intel/ice/ice_common.h
index 81e8ac925b9c..b5c013fdaaf9 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.h
+++ b/drivers/net/ethernet/intel/ice/ice_common.h
@@ -36,6 +36,8 @@ ice_acquire_res(struct ice_hw *hw, enum ice_aq_res_ids res,
 void ice_release_res(struct ice_hw *hw, enum ice_aq_res_ids res);
 enum ice_status
 ice_alloc_hw_res(struct ice_hw *hw, u16 type, u16 num, bool btm, u16 *res);
+enum ice_status
+ice_free_hw_res(struct ice_hw *hw, u16 type, u16 num, u16 *res);
 enum ice_status ice_init_nvm(struct ice_hw *hw);
 enum ice_status
 ice_read_sr_buf(struct ice_hw *hw, u16 offset, u16 *words, u16 *data);
diff --git a/drivers/net/ethernet/intel/ice/ice_flex_pipe.c b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c
index 5c09d3105b8b..8c93c303a4a5 100644
--- a/drivers/net/ethernet/intel/ice/ice_flex_pipe.c
+++ b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c
@@ -159,6 +159,176 @@ ice_pkg_enum_section(struct ice_seg *ice_seg, struct ice_pkg_enum *state,
 	return state->sect;
 }
 
+/* Key creation */
+
+#define ICE_DC_KEY	0x1	/* don't care */
+#define ICE_DC_KEYINV	0x1
+#define ICE_NM_KEY	0x0	/* never match */
+#define ICE_NM_KEYINV	0x0
+#define ICE_0_KEY	0x1	/* match 0 */
+#define ICE_0_KEYINV	0x0
+#define ICE_1_KEY	0x0	/* match 1 */
+#define ICE_1_KEYINV	0x1
+
+/**
+ * ice_gen_key_word - generate 16-bits of a key/mask word
+ * @val: the value
+ * @valid: valid bits mask (change only the valid bits)
+ * @dont_care: don't care mask
+ * @nvr_mtch: never match mask
+ * @key: pointer to an array of where the resulting key portion
+ * @key_inv: pointer to an array of where the resulting key invert portion
+ *
+ * This function generates 16-bits from a 8-bit value, an 8-bit don't care mask
+ * and an 8-bit never match mask. The 16-bits of output are divided into 8 bits
+ * of key and 8 bits of key invert.
+ *
+ *     '0' =    b01, always match a 0 bit
+ *     '1' =    b10, always match a 1 bit
+ *     '?' =    b11, don't care bit (always matches)
+ *     '~' =    b00, never match bit
+ *
+ * Input:
+ *          val:         b0  1  0  1  0  1
+ *          dont_care:   b0  0  1  1  0  0
+ *          never_mtch:  b0  0  0  0  1  1
+ *          ------------------------------
+ * Result:  key:        b01 10 11 11 00 00
+ */
+static enum ice_status
+ice_gen_key_word(u8 val, u8 valid, u8 dont_care, u8 nvr_mtch, u8 *key,
+		 u8 *key_inv)
+{
+	u8 in_key = *key, in_key_inv = *key_inv;
+	u8 i;
+
+	/* 'dont_care' and 'nvr_mtch' masks cannot overlap */
+	if ((dont_care ^ nvr_mtch) != (dont_care | nvr_mtch))
+		return ICE_ERR_CFG;
+
+	*key = 0;
+	*key_inv = 0;
+
+	/* encode the 8 bits into 8-bit key and 8-bit key invert */
+	for (i = 0; i < 8; i++) {
+		*key >>= 1;
+		*key_inv >>= 1;
+
+		if (!(valid & 0x1)) { /* change only valid bits */
+			*key |= (in_key & 0x1) << 7;
+			*key_inv |= (in_key_inv & 0x1) << 7;
+		} else if (dont_care & 0x1) { /* don't care bit */
+			*key |= ICE_DC_KEY << 7;
+			*key_inv |= ICE_DC_KEYINV << 7;
+		} else if (nvr_mtch & 0x1) { /* never match bit */
+			*key |= ICE_NM_KEY << 7;
+			*key_inv |= ICE_NM_KEYINV << 7;
+		} else if (val & 0x01) { /* exact 1 match */
+			*key |= ICE_1_KEY << 7;
+			*key_inv |= ICE_1_KEYINV << 7;
+		} else { /* exact 0 match */
+			*key |= ICE_0_KEY << 7;
+			*key_inv |= ICE_0_KEYINV << 7;
+		}
+
+		dont_care >>= 1;
+		nvr_mtch >>= 1;
+		valid >>= 1;
+		val >>= 1;
+		in_key >>= 1;
+		in_key_inv >>= 1;
+	}
+
+	return 0;
+}
+
+/**
+ * ice_bits_max_set - determine if the number of bits set is within a maximum
+ * @mask: pointer to the byte array which is the mask
+ * @size: the number of bytes in the mask
+ * @max: the max number of set bits
+ *
+ * This function determines if there are at most 'max' number of bits set in an
+ * array. Returns true if the number for bits set is <= max or will return false
+ * otherwise.
+ */
+static bool ice_bits_max_set(const u8 *mask, u16 size, u16 max)
+{
+	u16 count = 0;
+	u16 i;
+
+	/* check each byte */
+	for (i = 0; i < size; i++) {
+		/* if 0, go to next byte */
+		if (!mask[i])
+			continue;
+
+		/* We know there is at least one set bit in this byte because of
+		 * the above check; if we already have found 'max' number of
+		 * bits set, then we can return failure now.
+		 */
+		if (count == max)
+			return false;
+
+		/* count the bits in this byte, checking threshold */
+		count += hweight8(mask[i]);
+		if (count > max)
+			return false;
+	}
+
+	return true;
+}
+
+/**
+ * ice_set_key - generate a variable sized key with multiples of 16-bits
+ * @key: pointer to where the key will be stored
+ * @size: the size of the complete key in bytes (must be even)
+ * @val: array of 8-bit values that makes up the value portion of the key
+ * @upd: array of 8-bit masks that determine what key portion to update
+ * @dc: array of 8-bit masks that make up the don't care mask
+ * @nm: array of 8-bit masks that make up the never match mask
+ * @off: the offset of the first byte in the key to update
+ * @len: the number of bytes in the key update
+ *
+ * This function generates a key from a value, a don't care mask and a never
+ * match mask.
+ * upd, dc, and nm are optional parameters, and can be NULL:
+ *	upd == NULL --> udp mask is all 1's (update all bits)
+ *	dc == NULL --> dc mask is all 0's (no don't care bits)
+ *	nm == NULL --> nm mask is all 0's (no never match bits)
+ */
+static enum ice_status
+ice_set_key(u8 *key, u16 size, u8 *val, u8 *upd, u8 *dc, u8 *nm, u16 off,
+	    u16 len)
+{
+	u16 half_size;
+	u16 i;
+
+	/* size must be a multiple of 2 bytes. */
+	if (size % 2)
+		return ICE_ERR_CFG;
+
+	half_size = size / 2;
+	if (off + len > half_size)
+		return ICE_ERR_CFG;
+
+	/* Make sure at most one bit is set in the never match mask. Having more
+	 * than one never match mask bit set will cause HW to consume excessive
+	 * power otherwise; this is a power management efficiency check.
+	 */
+#define ICE_NVR_MTCH_BITS_MAX	1
+	if (nm && !ice_bits_max_set(nm, len, ICE_NVR_MTCH_BITS_MAX))
+		return ICE_ERR_CFG;
+
+	for (i = 0; i < len; i++)
+		if (ice_gen_key_word(val[i], upd ? upd[i] : 0xff,
+				     dc ? dc[i] : 0, nm ? nm[i] : 0,
+				     key + off + i, key + half_size + off + i))
+			return ICE_ERR_CFG;
+
+	return 0;
+}
+
 /**
  * ice_acquire_global_cfg_lock
  * @hw: pointer to the HW structure
@@ -952,6 +1122,48 @@ enum ice_sid_all {
 	ICE_SID_OFF_COUNT,
 };
 
+/* Characteristic handling */
+
+/**
+ * ice_match_prop_lst - determine if properties of two lists match
+ * @list1: first properties list
+ * @list2: second properties list
+ *
+ * Count, cookies and the order must match in order to be considered equivalent.
+ */
+static bool
+ice_match_prop_lst(struct list_head *list1, struct list_head *list2)
+{
+	struct ice_vsig_prof *tmp1;
+	struct ice_vsig_prof *tmp2;
+	u16 chk_count = 0;
+	u16 count = 0;
+
+	/* compare counts */
+	list_for_each_entry(tmp1, list1, list)
+		count++;
+	list_for_each_entry(tmp2, list2, list)
+		chk_count++;
+	if (!count || count != chk_count)
+		return false;
+
+	tmp1 = list_first_entry(list1, struct ice_vsig_prof, list);
+	tmp2 = list_first_entry(list2, struct ice_vsig_prof, list);
+
+	/* profile cookies must compare, and in the exact same order to take
+	 * into account priority
+	 */
+	while (count--) {
+		if (tmp2->profile_cookie != tmp1->profile_cookie)
+			return false;
+
+		tmp1 = list_next_entry(tmp1, list);
+		tmp2 = list_next_entry(tmp2, list);
+	}
+
+	return true;
+}
+
 /* VSIG Management */
 
 /**
@@ -999,6 +1211,117 @@ static u16 ice_vsig_alloc_val(struct ice_hw *hw, enum ice_block blk, u16 vsig)
 	return ICE_VSIG_VALUE(idx, hw->pf_id);
 }
 
+/**
+ * ice_vsig_alloc - Finds a free entry and allocates a new VSIG
+ * @hw: pointer to the hardware structure
+ * @blk: HW block
+ *
+ * This function will iterate through the VSIG list and mark the first
+ * unused entry for the new VSIG entry as used and return that value.
+ */
+static u16 ice_vsig_alloc(struct ice_hw *hw, enum ice_block blk)
+{
+	u16 i;
+
+	for (i = 1; i < ICE_MAX_VSIGS; i++)
+		if (!hw->blk[blk].xlt2.vsig_tbl[i].in_use)
+			return ice_vsig_alloc_val(hw, blk, i);
+
+	return ICE_DEFAULT_VSIG;
+}
+
+/**
+ * ice_find_dup_props_vsig - find VSI group with a specified set of properties
+ * @hw: pointer to the hardware structure
+ * @blk: HW block
+ * @chs: characteristic list
+ * @vsig: returns the VSIG with the matching profiles, if found
+ *
+ * Each VSIG is associated with a characteristic set; i.e. all VSIs under
+ * a group have the same characteristic set. To check if there exists a VSIG
+ * which has the same characteristics as the input characteristics; this
+ * function will iterate through the XLT2 list and return the VSIG that has a
+ * matching configuration. In order to make sure that priorities are accounted
+ * for, the list must match exactly, including the order in which the
+ * characteristics are listed.
+ */
+static enum ice_status
+ice_find_dup_props_vsig(struct ice_hw *hw, enum ice_block blk,
+			struct list_head *chs, u16 *vsig)
+{
+	struct ice_xlt2 *xlt2 = &hw->blk[blk].xlt2;
+	u16 i;
+
+	for (i = 0; i < xlt2->count; i++)
+		if (xlt2->vsig_tbl[i].in_use &&
+		    ice_match_prop_lst(chs, &xlt2->vsig_tbl[i].prop_lst)) {
+			*vsig = ICE_VSIG_VALUE(i, hw->pf_id);
+			return 0;
+		}
+
+	return ICE_ERR_DOES_NOT_EXIST;
+}
+
+/**
+ * ice_vsig_free - free VSI group
+ * @hw: pointer to the hardware structure
+ * @blk: HW block
+ * @vsig: VSIG to remove
+ *
+ * The function will remove all VSIs associated with the input VSIG and move
+ * them to the DEFAULT_VSIG and mark the VSIG available.
+ */
+static enum ice_status
+ice_vsig_free(struct ice_hw *hw, enum ice_block blk, u16 vsig)
+{
+	struct ice_vsig_prof *dtmp, *del;
+	struct ice_vsig_vsi *vsi_cur;
+	u16 idx;
+
+	idx = vsig & ICE_VSIG_IDX_M;
+	if (idx >= ICE_MAX_VSIGS)
+		return ICE_ERR_PARAM;
+
+	if (!hw->blk[blk].xlt2.vsig_tbl[idx].in_use)
+		return ICE_ERR_DOES_NOT_EXIST;
+
+	hw->blk[blk].xlt2.vsig_tbl[idx].in_use = false;
+
+	vsi_cur = hw->blk[blk].xlt2.vsig_tbl[idx].first_vsi;
+	/* If the VSIG has at least 1 VSI then iterate through the
+	 * list and remove the VSIs before deleting the group.
+	 */
+	if (vsi_cur) {
+		/* remove all vsis associated with this VSIG XLT2 entry */
+		do {
+			struct ice_vsig_vsi *tmp = vsi_cur->next_vsi;
+
+			vsi_cur->vsig = ICE_DEFAULT_VSIG;
+			vsi_cur->changed = 1;
+			vsi_cur->next_vsi = NULL;
+			vsi_cur = tmp;
+		} while (vsi_cur);
+
+		/* NULL terminate head of VSI list */
+		hw->blk[blk].xlt2.vsig_tbl[idx].first_vsi = NULL;
+	}
+
+	/* free characteristic list */
+	list_for_each_entry_safe(del, dtmp,
+				 &hw->blk[blk].xlt2.vsig_tbl[idx].prop_lst,
+				 list) {
+		list_del(&del->list);
+		devm_kfree(ice_hw_to_dev(hw), del);
+	}
+
+	/* if VSIG characteristic list was cleared for reset
+	 * re-initialize the list head
+	 */
+	INIT_LIST_HEAD(&hw->blk[blk].xlt2.vsig_tbl[idx].prop_lst);
+
+	return 0;
+}
+
 /**
  * ice_vsig_remove_vsi - remove VSI from VSIG
  * @hw: pointer to the hardware structure
@@ -1162,6 +1485,62 @@ static bool ice_prof_id_rsrc_type(enum ice_block blk, u16 *rsrc_type)
 	return true;
 }
 
+/**
+ * ice_tcam_ent_rsrc_type - get TCAM entry resource type for a block type
+ * @blk: the block type
+ * @rsrc_type: pointer to variable to receive the resource type
+ */
+static bool ice_tcam_ent_rsrc_type(enum ice_block blk, u16 *rsrc_type)
+{
+	switch (blk) {
+	case ICE_BLK_RSS:
+		*rsrc_type = ICE_AQC_RES_TYPE_HASH_PROF_BLDR_TCAM;
+		break;
+	default:
+		return false;
+	}
+	return true;
+}
+
+/**
+ * ice_alloc_tcam_ent - allocate hardware TCAM entry
+ * @hw: pointer to the HW struct
+ * @blk: the block to allocate the TCAM for
+ * @tcam_idx: pointer to variable to receive the TCAM entry
+ *
+ * This function allocates a new entry in a Profile ID TCAM for a specific
+ * block.
+ */
+static enum ice_status
+ice_alloc_tcam_ent(struct ice_hw *hw, enum ice_block blk, u16 *tcam_idx)
+{
+	u16 res_type;
+
+	if (!ice_tcam_ent_rsrc_type(blk, &res_type))
+		return ICE_ERR_PARAM;
+
+	return ice_alloc_hw_res(hw, res_type, 1, true, tcam_idx);
+}
+
+/**
+ * ice_free_tcam_ent - free hardware TCAM entry
+ * @hw: pointer to the HW struct
+ * @blk: the block from which to free the TCAM entry
+ * @tcam_idx: the TCAM entry to free
+ *
+ * This function frees an entry in a Profile ID TCAM for a specific block.
+ */
+static enum ice_status
+ice_free_tcam_ent(struct ice_hw *hw, enum ice_block blk, u16 tcam_idx)
+{
+	u16 res_type;
+
+	if (!ice_tcam_ent_rsrc_type(blk, &res_type))
+		return ICE_ERR_PARAM;
+
+	return ice_free_hw_res(hw, res_type, 1, &tcam_idx);
+}
+
 /**
  * ice_alloc_prof_id - allocate profile ID
  * @hw: pointer to the HW struct
@@ -1688,6 +2067,146 @@ err:
 	return ICE_ERR_NO_MEMORY;
 }
 
+/**
+ * ice_prof_gen_key - generate profile ID key
+ * @hw: pointer to the HW struct
+ * @blk: the block in which to write profile ID to
+ * @ptg: packet type group (PTG) portion of key
+ * @vsig: VSIG portion of key
+ * @cdid: CDID portion of key
+ * @flags: flag portion of key
+ * @vl_msk: valid mask
+ * @dc_msk: don't care mask
+ * @nm_msk: never match mask
+ * @key: output of profile ID key
+ */
+static enum ice_status
+ice_prof_gen_key(struct ice_hw *hw, enum ice_block blk, u8 ptg, u16 vsig,
+		 u8 cdid, u16 flags, u8 vl_msk[ICE_TCAM_KEY_VAL_SZ],
+		 u8 dc_msk[ICE_TCAM_KEY_VAL_SZ], u8 nm_msk[ICE_TCAM_KEY_VAL_SZ],
+		 u8 key[ICE_TCAM_KEY_SZ])
+{
+	struct ice_prof_id_key inkey;
+
+	inkey.xlt1 = ptg;
+	inkey.xlt2_cdid = cpu_to_le16(vsig);
+	inkey.flags = cpu_to_le16(flags);
+
+	switch (hw->blk[blk].prof.cdid_bits) {
+	case 0:
+		break;
+	case 2:
+#define ICE_CD_2_M 0xC000U
+#define ICE_CD_2_S 14
+		inkey.xlt2_cdid &= ~cpu_to_le16(ICE_CD_2_M);
+		inkey.xlt2_cdid |= cpu_to_le16(BIT(cdid) << ICE_CD_2_S);
+		break;
+	case 4:
+#define ICE_CD_4_M 0xF000U
+#define ICE_CD_4_S 12
+		inkey.xlt2_cdid &= ~cpu_to_le16(ICE_CD_4_M);
+		inkey.xlt2_cdid |= cpu_to_le16(BIT(cdid) << ICE_CD_4_S);
+		break;
+	case 8:
+#define ICE_CD_8_M 0xFF00U
+#define ICE_CD_8_S 16
+		inkey.xlt2_cdid &= ~cpu_to_le16(ICE_CD_8_M);
+		inkey.xlt2_cdid |= cpu_to_le16(BIT(cdid) << ICE_CD_8_S);
+		break;
+	default:
+		ice_debug(hw, ICE_DBG_PKG, "Error in profile config\n");
+		break;
+	}
+
+	return ice_set_key(key, ICE_TCAM_KEY_SZ, (u8 *)&inkey, vl_msk, dc_msk,
+			   nm_msk, 0, ICE_TCAM_KEY_SZ / 2);
+}
+
+/**
+ * ice_tcam_write_entry - write TCAM entry
+ * @hw: pointer to the HW struct
+ * @blk: the block in which to write profile ID to
+ * @idx: the entry index to write to
+ * @prof_id: profile ID
+ * @ptg: packet type group (PTG) portion of key
+ * @vsig: VSIG portion of key
+ * @cdid: CDID portion of key
+ * @flags: flag portion of key
+ * @vl_msk: valid mask
+ * @dc_msk: don't care mask
+ * @nm_msk: never match mask
+ */
+static enum ice_status
+ice_tcam_write_entry(struct ice_hw *hw, enum ice_block blk, u16 idx,
+		     u8 prof_id, u8 ptg, u16 vsig, u8 cdid, u16 flags,
+		     u8 vl_msk[ICE_TCAM_KEY_VAL_SZ],
+		     u8 dc_msk[ICE_TCAM_KEY_VAL_SZ],
+		     u8 nm_msk[ICE_TCAM_KEY_VAL_SZ])
+{
+	struct ice_prof_tcam_entry;
+	enum ice_status status;
+
+	status = ice_prof_gen_key(hw, blk, ptg, vsig, cdid, flags, vl_msk,
+				  dc_msk, nm_msk, hw->blk[blk].prof.t[idx].key);
+	if (!status) {
+		hw->blk[blk].prof.t[idx].addr = cpu_to_le16(idx);
+		hw->blk[blk].prof.t[idx].prof_id = prof_id;
+	}
+
+	return status;
+}
+
+/**
+ * ice_vsig_get_ref - returns number of VSIs belong to a VSIG
+ * @hw: pointer to the hardware structure
+ * @blk: HW block
+ * @vsig: VSIG to query
+ * @refs: pointer to variable to receive the reference count
+ */
+static enum ice_status
+ice_vsig_get_ref(struct ice_hw *hw, enum ice_block blk, u16 vsig, u16 *refs)
+{
+	u16 idx = vsig & ICE_VSIG_IDX_M;
+	struct ice_vsig_vsi *ptr;
+
+	*refs = 0;
+
+	if (!hw->blk[blk].xlt2.vsig_tbl[idx].in_use)
+		return ICE_ERR_DOES_NOT_EXIST;
+
+	ptr = hw->blk[blk].xlt2.vsig_tbl[idx].first_vsi;
+	while (ptr) {
+		(*refs)++;
+		ptr = ptr->next_vsi;
+	}
+
+	return 0;
+}
+
+/**
+ * ice_has_prof_vsig - check to see if VSIG has a specific profile
+ * @hw: pointer to the hardware structure
+ * @blk: HW block
+ * @vsig: VSIG to check against
+ * @hdl: profile handle
+ */
+static bool
+ice_has_prof_vsig(struct ice_hw *hw, enum ice_block blk, u16 vsig, u64 hdl)
+{
+	u16 idx = vsig & ICE_VSIG_IDX_M;
+	struct ice_vsig_prof *ent;
+
+	list_for_each_entry(ent, &hw->blk[blk].xlt2.vsig_tbl[idx].prop_lst,
+			    list)
+		if (ent->profile_cookie == hdl)
+			return true;
+
+	ice_debug(hw, ICE_DBG_INIT,
+		  "Characteristic list for VSI group %d not found.\n",
+		  vsig);
+	return false;
+}
+
 /**
  * ice_add_prof - add profile
  * @hw: pointer to the HW struct
@@ -1792,3 +2311,802 @@ err_ice_add_prof:
 	mutex_unlock(&hw->blk[blk].es.prof_map_lock);
 	return status;
 }
+
+/**
+ * ice_search_prof_id_low - Search for a profile tracking ID low level
+ * @hw: pointer to the HW struct
+ * @blk: hardware block
+ * @id: profile tracking ID
+ *
+ * This will search for a profile tracking ID which was previously added. This
+ * version assumes that the caller has already acquired the prof map lock.
+ */
+static struct ice_prof_map *
+ice_search_prof_id_low(struct ice_hw *hw, enum ice_block blk, u64 id)
+{
+	struct ice_prof_map *entry = NULL;
+	struct ice_prof_map *map;
+
+	list_for_each_entry(map, &hw->blk[blk].es.prof_map, list)
+		if (map->profile_cookie == id) {
+			entry = map;
+			break;
+		}
+
+	return entry;
+}
+
+/**
+ * ice_search_prof_id - Search for a profile tracking ID
+ * @hw: pointer to the HW struct
+ * @blk: hardware block
+ * @id: profile tracking ID
+ *
+ * This will search for a profile tracking ID which was previously added.
+ */
+static struct ice_prof_map *
+ice_search_prof_id(struct ice_hw *hw, enum ice_block blk, u64 id)
+{
+	struct ice_prof_map *entry;
+
+	mutex_lock(&hw->blk[blk].es.prof_map_lock);
+	entry = ice_search_prof_id_low(hw, blk, id);
+	mutex_unlock(&hw->blk[blk].es.prof_map_lock);
+
+	return entry;
+}
+
+/**
+ * ice_rel_tcam_idx - release a TCAM index
+ * @hw: pointer to the HW struct
+ * @blk: hardware block
+ * @idx: the index to release
+ */
+static enum ice_status
+ice_rel_tcam_idx(struct ice_hw *hw, enum ice_block blk, u16 idx)
+{
+	/* Masks to invoke a never match entry */
+	u8 vl_msk[ICE_TCAM_KEY_VAL_SZ] = { 0xFF, 0xFF, 0xFF, 0xFF, 0xFF };
+	u8 dc_msk[ICE_TCAM_KEY_VAL_SZ] = { 0xFE, 0xFF, 0xFF, 0xFF, 0xFF };
+	u8 nm_msk[ICE_TCAM_KEY_VAL_SZ] = { 0x01, 0x00, 0x00, 0x00, 0x00 };
+	enum ice_status status;
+
+	/* write the TCAM entry */
+	status = ice_tcam_write_entry(hw, blk, idx, 0, 0, 0, 0, 0, vl_msk,
+				      dc_msk, nm_msk);
+	if (status)
+		return status;
+
+	/* release the TCAM entry */
+	status = ice_free_tcam_ent(hw, blk, idx);
+
+	return status;
+}
+
+/**
+ * ice_rem_prof_id - remove one profile from a VSIG
+ * @hw: pointer to the HW struct
+ * @blk: hardware block
+ * @prof: pointer to profile structure to remove
+ */
+static enum ice_status
+ice_rem_prof_id(struct ice_hw *hw, enum ice_block blk,
+		struct ice_vsig_prof *prof)
+{
+	enum ice_status status;
+	u16 i;
+
+	for (i = 0; i < prof->tcam_count; i++)
+		if (prof->tcam[i].in_use) {
+			prof->tcam[i].in_use = false;
+			status = ice_rel_tcam_idx(hw, blk,
+						  prof->tcam[i].tcam_idx);
+			if (status)
+				return ICE_ERR_HW_TABLE;
+		}
+
+	return 0;
+}
+
+/**
+ * ice_rem_vsig - remove VSIG
+ * @hw: pointer to the HW struct
+ * @blk: hardware block
+ * @vsig: the VSIG to remove
+ * @chg: the change list
+ */
+static enum ice_status
+ice_rem_vsig(struct ice_hw *hw, enum ice_block blk, u16 vsig,
+	     struct list_head *chg)
+{
+	u16 idx = vsig & ICE_VSIG_IDX_M;
+	struct ice_vsig_vsi *vsi_cur;
+	struct ice_vsig_prof *d, *t;
+	enum ice_status status;
+
+	/* remove TCAM entries */
+	list_for_each_entry_safe(d, t,
+				 &hw->blk[blk].xlt2.vsig_tbl[idx].prop_lst,
+				 list) {
+		status = ice_rem_prof_id(hw, blk, d);
+		if (status)
+			return status;
+
+		list_del(&d->list);
+		devm_kfree(ice_hw_to_dev(hw), d);
+	}
+
+	/* Move all VSIS associated with this VSIG to the default VSIG */
+	vsi_cur = hw->blk[blk].xlt2.vsig_tbl[idx].first_vsi;
+	/* If the VSIG has at least 1 VSI then iterate through the list
+	 * and remove the VSIs before deleting the group.
+	 */
+	if (vsi_cur)
+		do {
+			struct ice_vsig_vsi *tmp = vsi_cur->next_vsi;
+			struct ice_chs_chg *p;
+
+			p = devm_kzalloc(ice_hw_to_dev(hw), sizeof(*p),
+					 GFP_KERNEL);
+			if (!p)
+				return ICE_ERR_NO_MEMORY;
+
+			p->type = ICE_VSIG_REM;
+			p->orig_vsig = vsig;
+			p->vsig = ICE_DEFAULT_VSIG;
+			p->vsi = vsi_cur - hw->blk[blk].xlt2.vsis;
+
+			list_add(&p->list_entry, chg);
+
+			vsi_cur = tmp;
+		} while (vsi_cur);
+
+	return ice_vsig_free(hw, blk, vsig);
+}
+
+/**
+ * ice_get_prof - get profile
+ * @hw: pointer to the HW struct
+ * @blk: hardware block
+ * @hdl: profile handle
+ * @chg: change list
+ */
+static enum ice_status
+ice_get_prof(struct ice_hw *hw, enum ice_block blk, u64 hdl,
+	     struct list_head *chg)
+{
+	struct ice_prof_map *map;
+	struct ice_chs_chg *p;
+	u16 i;
+
+	/* Get the details on the profile specified by the handle ID */
+	map = ice_search_prof_id(hw, blk, hdl);
+	if (!map)
+		return ICE_ERR_DOES_NOT_EXIST;
+
+	for (i = 0; i < map->ptg_cnt; i++)
+		if (!hw->blk[blk].es.written[map->prof_id]) {
+			/* add ES to change list */
+			p = devm_kzalloc(ice_hw_to_dev(hw), sizeof(*p),
+					 GFP_KERNEL);
+			if (!p)
+				goto err_ice_get_prof;
+
+			p->type = ICE_PTG_ES_ADD;
+			p->ptype = 0;
+			p->ptg = map->ptg[i];
+			p->add_ptg = 0;
+
+			p->add_prof = 1;
+			p->prof_id = map->prof_id;
+
+			hw->blk[blk].es.written[map->prof_id] = true;
+
+			list_add(&p->list_entry, chg);
+		}
+
+	return 0;
+
+err_ice_get_prof:
+	/* let caller clean up the change list */
+	return ICE_ERR_NO_MEMORY;
+}
+
+/**
+ * ice_get_profs_vsig - get a copy of the list of profiles from a VSIG
+ * @hw: pointer to the HW struct
+ * @blk: hardware block
+ * @vsig: VSIG from which to copy the list
+ * @lst: output list
+ *
+ * This routine makes a copy of the list of profiles in the specified VSIG.
+ */
+static enum ice_status
+ice_get_profs_vsig(struct ice_hw *hw, enum ice_block blk, u16 vsig,
+		   struct list_head *lst)
+{
+	struct ice_vsig_prof *ent1, *ent2;
+	u16 idx = vsig & ICE_VSIG_IDX_M;
+
+	list_for_each_entry(ent1, &hw->blk[blk].xlt2.vsig_tbl[idx].prop_lst,
+			    list) {
+		struct ice_vsig_prof *p;
+
+		/* copy to the input list */
+		p = devm_kmemdup(ice_hw_to_dev(hw), ent1, sizeof(*p),
+				 GFP_KERNEL);
+		if (!p)
+			goto err_ice_get_profs_vsig;
+
+		list_add_tail(&p->list, lst);
+	}
+
+	return 0;
+
+err_ice_get_profs_vsig:
+	list_for_each_entry_safe(ent1, ent2, lst, list) {
+		list_del(&ent1->list);
+		devm_kfree(ice_hw_to_dev(hw), ent1);
+	}
+
+	return ICE_ERR_NO_MEMORY;
+}
+
+/**
+ * ice_add_prof_to_lst - add profile entry to a list
+ * @hw: pointer to the HW struct
+ * @blk: hardware block
+ * @lst: the list to be added to
+ * @hdl: profile handle of entry to add
+ */
+static enum ice_status
+ice_add_prof_to_lst(struct ice_hw *hw, enum ice_block blk,
+		    struct list_head *lst, u64 hdl)
+{
+	struct ice_prof_map *map;
+	struct ice_vsig_prof *p;
+	u16 i;
+
+	map = ice_search_prof_id(hw, blk, hdl);
+	if (!map)
+		return ICE_ERR_DOES_NOT_EXIST;
+
+	p = devm_kzalloc(ice_hw_to_dev(hw), sizeof(*p), GFP_KERNEL);
+	if (!p)
+		return ICE_ERR_NO_MEMORY;
+
+	p->profile_cookie = map->profile_cookie;
+	p->prof_id = map->prof_id;
+	p->tcam_count = map->ptg_cnt;
+
+	for (i = 0; i < map->ptg_cnt; i++) {
+		p->tcam[i].prof_id = map->prof_id;
+		p->tcam[i].tcam_idx = ICE_INVALID_TCAM;
+		p->tcam[i].ptg = map->ptg[i];
+	}
+
+	list_add(&p->list, lst);
+
+	return 0;
+}
+
+/**
+ * ice_move_vsi - move VSI to another VSIG
+ * @hw: pointer to the HW struct
+ * @blk: hardware block
+ * @vsi: the VSI to move
+ * @vsig: the VSIG to move the VSI to
+ * @chg: the change list
+ */
+static enum ice_status
+ice_move_vsi(struct ice_hw *hw, enum ice_block blk, u16 vsi, u16 vsig,
+	     struct list_head *chg)
+{
+	enum ice_status status;
+	struct ice_chs_chg *p;
+	u16 orig_vsig;
+
+	p = devm_kzalloc(ice_hw_to_dev(hw), sizeof(*p), GFP_KERNEL);
+	if (!p)
+		return ICE_ERR_NO_MEMORY;
+
+	status = ice_vsig_find_vsi(hw, blk, vsi, &orig_vsig);
+	if (!status)
+		status = ice_vsig_add_mv_vsi(hw, blk, vsi, vsig);
+
+	if (status) {
+		devm_kfree(ice_hw_to_dev(hw), p);
+		return status;
+	}
+
+	p->type = ICE_VSI_MOVE;
+	p->vsi = vsi;
+	p->orig_vsig = orig_vsig;
+	p->vsig = vsig;
+
+	list_add(&p->list_entry, chg);
+
+	return 0;
+}
+
+/**
+ * ice_prof_tcam_ena_dis - add enable or disable TCAM change
+ * @hw: pointer to the HW struct
+ * @blk: hardware block
+ * @enable: true to enable, false to disable
+ * @vsig: the VSIG of the TCAM entry
+ * @tcam: pointer the TCAM info structure of the TCAM to disable
+ * @chg: the change list
+ *
+ * This function appends an enable or disable TCAM entry in the change log
+ */
+static enum ice_status
+ice_prof_tcam_ena_dis(struct ice_hw *hw, enum ice_block blk, bool enable,
+		      u16 vsig, struct ice_tcam_inf *tcam,
+		      struct list_head *chg)
+{
+	enum ice_status status;
+	struct ice_chs_chg *p;
+
+	/* Default: enable means change the low flag bit to don't care */
+	u8 dc_msk[ICE_TCAM_KEY_VAL_SZ] = { 0x01, 0x00, 0x00, 0x00, 0x00 };
+	u8 nm_msk[ICE_TCAM_KEY_VAL_SZ] = { 0x00, 0x00, 0x00, 0x00, 0x00 };
+	u8 vl_msk[ICE_TCAM_KEY_VAL_SZ] = { 0x01, 0x00, 0x00, 0x00, 0x00 };
+
+	/* if disabling, free the TCAM */
+	if (!enable) {
+		status = ice_free_tcam_ent(hw, blk, tcam->tcam_idx);
+		tcam->tcam_idx = 0;
+		tcam->in_use = 0;
+		return status;
+	}
+
+	/* for re-enabling, reallocate a TCAM */
+	status = ice_alloc_tcam_ent(hw, blk, &tcam->tcam_idx);
+	if (status)
+		return status;
+
+	/* add TCAM to change list */
+	p = devm_kzalloc(ice_hw_to_dev(hw), sizeof(*p), GFP_KERNEL);
+	if (!p)
+		return ICE_ERR_NO_MEMORY;
+
+	status = ice_tcam_write_entry(hw, blk, tcam->tcam_idx, tcam->prof_id,
+				      tcam->ptg, vsig, 0, 0, vl_msk, dc_msk,
+				      nm_msk);
+	if (status)
+		goto err_ice_prof_tcam_ena_dis;
+
+	tcam->in_use = 1;
+
+	p->type = ICE_TCAM_ADD;
+	p->add_tcam_idx = true;
+	p->prof_id = tcam->prof_id;
+	p->ptg = tcam->ptg;
+	p->vsig = 0;
+	p->tcam_idx = tcam->tcam_idx;
+
+	/* log change */
+	list_add(&p->list_entry, chg);
+
+	return 0;
+
+err_ice_prof_tcam_ena_dis:
+	devm_kfree(ice_hw_to_dev(hw), p);
+	return status;
+}
+
+/**
+ * ice_adj_prof_priorities - adjust profile based on priorities
+ * @hw: pointer to the HW struct
+ * @blk: hardware block
+ * @vsig: the VSIG for which to adjust profile priorities
+ * @chg: the change list
+ */
+static enum ice_status
+ice_adj_prof_priorities(struct ice_hw *hw, enum ice_block blk, u16 vsig,
+			struct list_head *chg)
+{
+	DECLARE_BITMAP(ptgs_used, ICE_XLT1_CNT);
+	struct ice_vsig_prof *t;
+	enum ice_status status;
+	u16 idx;
+
+	bitmap_zero(ptgs_used, ICE_XLT1_CNT);
+	idx = vsig & ICE_VSIG_IDX_M;
+
+	/* Priority is based on the order in which the profiles are added. The
+	 * newest added profile has highest priority and the oldest added
+	 * profile has the lowest priority. Since the profile property list for
+	 * a VSIG is sorted from newest to oldest, this code traverses the list
+	 * in order and enables the first of each PTG that it finds (that is not
+	 * already enabled); it also disables any duplicate PTGs that it finds
+	 * in the older profiles (that are currently enabled).
+	 */
+
+	list_for_each_entry(t, &hw->blk[blk].xlt2.vsig_tbl[idx].prop_lst,
+			    list) {
+		u16 i;
+
+		for (i = 0; i < t->tcam_count; i++) {
+			/* Scan the priorities from newest to oldest.
+			 * Make sure that the newest profiles take priority.
+			 */
+			if (test_bit(t->tcam[i].ptg, ptgs_used) &&
+			    t->tcam[i].in_use) {
+				/* need to mark this PTG as never match, as it
+				 * was already in use and therefore duplicate
+				 * (and lower priority)
+				 */
+				status = ice_prof_tcam_ena_dis(hw, blk, false,
+							       vsig,
+							       &t->tcam[i],
+							       chg);
+				if (status)
+					return status;
+			} else if (!test_bit(t->tcam[i].ptg, ptgs_used) &&
+				   !t->tcam[i].in_use) {
+				/* need to enable this PTG, as it in not in use
+				 * and not enabled (highest priority)
+				 */
+				status = ice_prof_tcam_ena_dis(hw, blk, true,
+							       vsig,
+							       &t->tcam[i],
+							       chg);
+				if (status)
+					return status;
+			}
+
+			/* keep track of used ptgs */
+			set_bit(t->tcam[i].ptg, ptgs_used);
+		}
+	}
+
+	return 0;
+}
+
+/**
+ * ice_add_prof_id_vsig - add profile to VSIG
+ * @hw: pointer to the HW struct
+ * @blk: hardware block
+ * @vsig: the VSIG to which this profile is to be added
+ * @hdl: the profile handle indicating the profile to add
+ * @chg: the change list
+ */
+static enum ice_status
+ice_add_prof_id_vsig(struct ice_hw *hw, enum ice_block blk, u16 vsig, u64 hdl,
+		     struct list_head *chg)
+{
+	/* Masks that ignore flags */
+	u8 vl_msk[ICE_TCAM_KEY_VAL_SZ] = { 0xFF, 0xFF, 0xFF, 0xFF, 0xFF };
+	u8 dc_msk[ICE_TCAM_KEY_VAL_SZ] = { 0xFF, 0xFF, 0x00, 0x00, 0x00 };
+	u8 nm_msk[ICE_TCAM_KEY_VAL_SZ] = { 0x00, 0x00, 0x00, 0x00, 0x00 };
+	struct ice_prof_map *map;
+	struct ice_vsig_prof *t;
+	struct ice_chs_chg *p;
+	u16 i;
+
+	/* Get the details on the profile specified by the handle ID */
+	map = ice_search_prof_id(hw, blk, hdl);
+	if (!map)
+		return ICE_ERR_DOES_NOT_EXIST;
+
+	/* Error, if this VSIG already has this profile */
+	if (ice_has_prof_vsig(hw, blk, vsig, hdl))
+		return ICE_ERR_ALREADY_EXISTS;
+
+	/* new VSIG profile structure */
+	t = devm_kzalloc(ice_hw_to_dev(hw), sizeof(*t), GFP_KERNEL);
+	if (!t)
+		return ICE_ERR_NO_MEMORY;
+
+	t->profile_cookie = map->profile_cookie;
+	t->prof_id = map->prof_id;
+	t->tcam_count = map->ptg_cnt;
+
+	/* create TCAM entries */
+	for (i = 0; i < map->ptg_cnt; i++) {
+		enum ice_status status;
+		u16 tcam_idx;
+
+		/* add TCAM to change list */
+		p = devm_kzalloc(ice_hw_to_dev(hw), sizeof(*p), GFP_KERNEL);
+		if (!p)
+			goto err_ice_add_prof_id_vsig;
+
+		/* allocate the TCAM entry index */
+		status = ice_alloc_tcam_ent(hw, blk, &tcam_idx);
+		if (status) {
+			devm_kfree(ice_hw_to_dev(hw), p);
+			goto err_ice_add_prof_id_vsig;
+		}
+
+		t->tcam[i].ptg = map->ptg[i];
+		t->tcam[i].prof_id = map->prof_id;
+		t->tcam[i].tcam_idx = tcam_idx;
+		t->tcam[i].in_use = true;
+
+		p->type = ICE_TCAM_ADD;
+		p->add_tcam_idx = true;
+		p->prof_id = t->tcam[i].prof_id;
+		p->ptg = t->tcam[i].ptg;
+		p->vsig = vsig;
+		p->tcam_idx = t->tcam[i].tcam_idx;
+
+		/* write the TCAM entry */
+		status = ice_tcam_write_entry(hw, blk, t->tcam[i].tcam_idx,
+					      t->tcam[i].prof_id,
+					      t->tcam[i].ptg, vsig, 0, 0,
+					      vl_msk, dc_msk, nm_msk);
+		if (status)
+			goto err_ice_add_prof_id_vsig;
+
+		/* log change */
+		list_add(&p->list_entry, chg);
+	}
+
+	/* add profile to VSIG */
+	list_add(&t->list,
+		 &hw->blk[blk].xlt2.vsig_tbl[(vsig & ICE_VSIG_IDX_M)].prop_lst);
+
+	return 0;
+
+err_ice_add_prof_id_vsig:
+	/* let caller clean up the change list */
+	devm_kfree(ice_hw_to_dev(hw), t);
+	return ICE_ERR_NO_MEMORY;
+}
+
+/**
+ * ice_create_prof_id_vsig - add a new VSIG with a single profile
+ * @hw: pointer to the HW struct
+ * @blk: hardware block
+ * @vsi: the initial VSI that will be in VSIG
+ * @hdl: the profile handle of the profile that will be added to the VSIG
+ * @chg: the change list
+ */
+static enum ice_status
+ice_create_prof_id_vsig(struct ice_hw *hw, enum ice_block blk, u16 vsi, u64 hdl,
+			struct list_head *chg)
+{
+	enum ice_status status;
+	struct ice_chs_chg *p;
+	u16 new_vsig;
+
+	p = devm_kzalloc(ice_hw_to_dev(hw), sizeof(*p), GFP_KERNEL);
+	if (!p)
+		return ICE_ERR_NO_MEMORY;
+
+	new_vsig = ice_vsig_alloc(hw, blk);
+	if (!new_vsig) {
+		status = ICE_ERR_HW_TABLE;
+		goto err_ice_create_prof_id_vsig;
+	}
+
+	status = ice_move_vsi(hw, blk, vsi, new_vsig, chg);
+	if (status)
+		goto err_ice_create_prof_id_vsig;
+
+	status = ice_add_prof_id_vsig(hw, blk, new_vsig, hdl, chg);
+	if (status)
+		goto err_ice_create_prof_id_vsig;
+
+	p->type = ICE_VSIG_ADD;
+	p->vsi = vsi;
+	p->orig_vsig = ICE_DEFAULT_VSIG;
+	p->vsig = new_vsig;
+
+	list_add(&p->list_entry, chg);
+
+	return 0;
+
+err_ice_create_prof_id_vsig:
+	/* let caller clean up the change list */
+	devm_kfree(ice_hw_to_dev(hw), p);
+	return status;
+}
+
+/**
+ * ice_create_vsig_from_lst - create a new VSIG with a list of profiles
+ * @hw: pointer to the HW struct
+ * @blk: hardware block
+ * @vsi: the initial VSI that will be in VSIG
+ * @lst: the list of profile that will be added to the VSIG
+ * @chg: the change list
+ */
+static enum ice_status
+ice_create_vsig_from_lst(struct ice_hw *hw, enum ice_block blk, u16 vsi,
+			 struct list_head *lst, struct list_head *chg)
+{
+	struct ice_vsig_prof *t;
+	enum ice_status status;
+	u16 vsig;
+
+	vsig = ice_vsig_alloc(hw, blk);
+	if (!vsig)
+		return ICE_ERR_HW_TABLE;
+
+	status = ice_move_vsi(hw, blk, vsi, vsig, chg);
+	if (status)
+		return status;
+
+	list_for_each_entry(t, lst, list) {
+		status = ice_add_prof_id_vsig(hw, blk, vsig, t->profile_cookie,
+					      chg);
+		if (status)
+			return status;
+	}
+
+	return 0;
+}
+
+/**
+ * ice_find_prof_vsig - find a VSIG with a specific profile handle
+ * @hw: pointer to the HW struct
+ * @blk: hardware block
+ * @hdl: the profile handle of the profile to search for
+ * @vsig: returns the VSIG with the matching profile
+ */
+static bool
+ice_find_prof_vsig(struct ice_hw *hw, enum ice_block blk, u64 hdl, u16 *vsig)
+{
+	struct ice_vsig_prof *t;
+	enum ice_status status;
+	struct list_head lst;
+
+	INIT_LIST_HEAD(&lst);
+
+	t = kzalloc(sizeof(*t), GFP_KERNEL);
+	if (!t)
+		return false;
+
+	t->profile_cookie = hdl;
+	list_add(&t->list, &lst);
+
+	status = ice_find_dup_props_vsig(hw, blk, &lst, vsig);
+
+	list_del(&t->list);
+	kfree(t);
+
+	return !status;
+}
+
+/**
+ * ice_add_prof_id_flow - add profile flow
+ * @hw: pointer to the HW struct
+ * @blk: hardware block
+ * @vsi: the VSI to enable with the profile specified by ID
+ * @hdl: profile handle
+ *
+ * Calling this function will update the hardware tables to enable the
+ * profile indicated by the ID parameter for the VSIs specified in the VSI
+ * array. Once successfully called, the flow will be enabled.
+ */
+enum ice_status
+ice_add_prof_id_flow(struct ice_hw *hw, enum ice_block blk, u16 vsi, u64 hdl)
+{
+	struct ice_vsig_prof *tmp1, *del1;
+	struct ice_chs_chg *tmp, *del;
+	struct list_head union_lst;
+	enum ice_status status;
+	struct list_head chg;
+	u16 vsig;
+
+	INIT_LIST_HEAD(&union_lst);
+	INIT_LIST_HEAD(&chg);
+
+	/* Get profile */
+	status = ice_get_prof(hw, blk, hdl, &chg);
+	if (status)
+		return status;
+
+	/* determine if VSI is already part of a VSIG */
+	status = ice_vsig_find_vsi(hw, blk, vsi, &vsig);
+	if (!status && vsig) {
+		bool only_vsi;
+		u16 or_vsig;
+		u16 ref;
+
+		/* found in VSIG */
+		or_vsig = vsig;
+
+		/* make sure that there is no overlap/conflict between the new
+		 * characteristics and the existing ones; we don't support that
+		 * scenario
+		 */
+		if (ice_has_prof_vsig(hw, blk, vsig, hdl)) {
+			status = ICE_ERR_ALREADY_EXISTS;
+			goto err_ice_add_prof_id_flow;
+		}
+
+		/* last VSI in the VSIG? */
+		status = ice_vsig_get_ref(hw, blk, vsig, &ref);
+		if (status)
+			goto err_ice_add_prof_id_flow;
+		only_vsi = (ref == 1);
+
+		/* create a union of the current profiles and the one being
+		 * added
+		 */
+		status = ice_get_profs_vsig(hw, blk, vsig, &union_lst);
+		if (status)
+			goto err_ice_add_prof_id_flow;
+
+		status = ice_add_prof_to_lst(hw, blk, &union_lst, hdl);
+		if (status)
+			goto err_ice_add_prof_id_flow;
+
+		/* search for an existing VSIG with an exact charc match */
+		status = ice_find_dup_props_vsig(hw, blk, &union_lst, &vsig);
+		if (!status) {
+			/* move VSI to the VSIG that matches */
+			status = ice_move_vsi(hw, blk, vsi, vsig, &chg);
+			if (status)
+				goto err_ice_add_prof_id_flow;
+
+			/* VSI has been moved out of or_vsig. If the or_vsig had
+			 * only that VSI it is now empty and can be removed.
+			 */
+			if (only_vsi) {
+				status = ice_rem_vsig(hw, blk, or_vsig, &chg);
+				if (status)
+					goto err_ice_add_prof_id_flow;
+			}
+		} else if (only_vsi) {
+			/* If the original VSIG only contains one VSI, then it
+			 * will be the requesting VSI. In this case the VSI is
+			 * not sharing entries and we can simply add the new
+			 * profile to the VSIG.
+			 */
+			status = ice_add_prof_id_vsig(hw, blk, vsig, hdl, &chg);
+			if (status)
+				goto err_ice_add_prof_id_flow;
+
+			/* Adjust priorities */
+			status = ice_adj_prof_priorities(hw, blk, vsig, &chg);
+			if (status)
+				goto err_ice_add_prof_id_flow;
+		} else {
+			/* No match, so we need a new VSIG */
+			status = ice_create_vsig_from_lst(hw, blk, vsi,
+							  &union_lst, &chg);
+			if (status)
+				goto err_ice_add_prof_id_flow;
+
+			/* Adjust priorities */
+			status = ice_adj_prof_priorities(hw, blk, vsig, &chg);
+			if (status)
+				goto err_ice_add_prof_id_flow;
+		}
+	} else {
+		/* need to find or add a VSIG */
+		/* search for an existing VSIG with an exact charc match */
+		if (ice_find_prof_vsig(hw, blk, hdl, &vsig)) {
+			/* found an exact match */
+			/* add or move VSI to the VSIG that matches */
+			status = ice_move_vsi(hw, blk, vsi, vsig, &chg);
+			if (status)
+				goto err_ice_add_prof_id_flow;
+		} else {
+			/* we did not find an exact match */
+			/* we need to add a VSIG */
+			status = ice_create_prof_id_vsig(hw, blk, vsi, hdl,
+							 &chg);
+			if (status)
+				goto err_ice_add_prof_id_flow;
+		}
+	}
+
+err_ice_add_prof_id_flow:
+	list_for_each_entry_safe(del, tmp, &chg, list_entry) {
+		list_del(&del->list_entry);
+		devm_kfree(ice_hw_to_dev(hw), del);
+	}
+
+	list_for_each_entry_safe(del1, tmp1, &union_lst, list) {
+		list_del(&del1->list);
+		devm_kfree(ice_hw_to_dev(hw), del1);
+	}
+
+	return status;
+}
diff --git a/drivers/net/ethernet/intel/ice/ice_flex_pipe.h b/drivers/net/ethernet/intel/ice/ice_flex_pipe.h
index 8cb7d7f09e0b..33e4510da24c 100644
--- a/drivers/net/ethernet/intel/ice/ice_flex_pipe.h
+++ b/drivers/net/ethernet/intel/ice/ice_flex_pipe.h
@@ -21,6 +21,8 @@
 enum ice_status
 ice_add_prof(struct ice_hw *hw, enum ice_block blk, u64 id, u8 ptypes[],
 	     struct ice_fv_word *es);
+enum ice_status
+ice_add_prof_id_flow(struct ice_hw *hw, enum ice_block blk, u16 vsi, u64 hdl);
 enum ice_status ice_init_pkg(struct ice_hw *hw, u8 *buff, u32 len);
 enum ice_status
 ice_copy_and_init_pkg(struct ice_hw *hw, const u8 *buf, u32 len);
diff --git a/drivers/net/ethernet/intel/ice/ice_flex_type.h b/drivers/net/ethernet/intel/ice/ice_flex_type.h
index 3005f111fb3b..9d95d51bc760 100644
--- a/drivers/net/ethernet/intel/ice/ice_flex_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_flex_type.h
@@ -283,6 +283,7 @@ struct ice_ptg_ptype {
 	u8 ptg;
 };
 
+#define ICE_MAX_TCAM_PER_PROFILE	32
 #define ICE_MAX_PTG_PER_PROFILE		32
 
 struct ice_prof_map {
@@ -294,6 +295,23 @@ struct ice_prof_map {
 	u8 ptg[ICE_MAX_PTG_PER_PROFILE];
 };
 
+#define ICE_INVALID_TCAM	0xFFFF
+
+struct ice_tcam_inf {
+	u16 tcam_idx;
+	u8 ptg;
+	u8 prof_id;
+	u8 in_use;
+};
+
+struct ice_vsig_prof {
+	struct list_head list;
+	u64 profile_cookie;
+	u8 prof_id;
+	u8 tcam_count;
+	struct ice_tcam_inf tcam[ICE_MAX_TCAM_PER_PROFILE];
+};
+
 struct ice_vsig_entry {
 	struct list_head prop_lst;
 	struct ice_vsig_vsi *first_vsi;
@@ -343,6 +361,13 @@ struct ice_xlt2 {
 	u16 count;
 };
 
+/* Profile ID Management */
+struct ice_prof_id_key {
+	__le16 flags;
+	u8 xlt1;
+	__le16 xlt2_cdid;
+} __packed;
+
 /* Keys are made up of two values, each one-half the size of the key.
  * For TCAM, the entire key is 80 bits wide (or 2, 40-bit wide values)
  */
@@ -385,5 +410,31 @@ struct ice_blk_info {
 	u8 is_list_init;
 };
 
+enum ice_chg_type {
+	ICE_TCAM_NONE = 0,
+	ICE_PTG_ES_ADD,
+	ICE_TCAM_ADD,
+	ICE_VSIG_ADD,
+	ICE_VSIG_REM,
+	ICE_VSI_MOVE,
+};
+
+struct ice_chs_chg {
+	struct list_head list_entry;
+	enum ice_chg_type type;
+
+	u8 add_ptg;
+	u8 add_vsig;
+	u8 add_tcam_idx;
+	u8 add_prof;
+	u16 ptype;
+	u8 ptg;
+	u8 prof_id;
+	u16 vsi;
+	u16 vsig;
+	u16 orig_vsig;
+	u16 tcam_idx;
+};
+
 #define ICE_FLOW_PTYPE_MAX		ICE_XLT1_CNT
 #endif /* _ICE_FLEX_TYPE_H_ */
diff --git a/drivers/net/ethernet/intel/ice/ice_flow.c b/drivers/net/ethernet/intel/ice/ice_flow.c
index 24fe04f8baa2..1bbf4b6ed7d2 100644
--- a/drivers/net/ethernet/intel/ice/ice_flow.c
+++ b/drivers/net/ethernet/intel/ice/ice_flow.c
@@ -450,6 +450,38 @@ out:
 	return status;
 }
 
+/**
+ * ice_flow_assoc_prof - associate a VSI with a flow profile
+ * @hw: pointer to the hardware structure
+ * @blk: classification stage
+ * @prof: pointer to flow profile
+ * @vsi_handle: software VSI handle
+ *
+ * Assumption: the caller has acquired the lock to the profile list
+ * and the software VSI handle has been validated
+ */
+static enum ice_status
+ice_flow_assoc_prof(struct ice_hw *hw, enum ice_block blk,
+		    struct ice_flow_prof *prof, u16 vsi_handle)
+{
+	enum ice_status status = 0;
+
+	if (!test_bit(vsi_handle, prof->vsis)) {
+		status = ice_add_prof_id_flow(hw, blk,
+					      ice_get_hw_vsi_num(hw,
+								 vsi_handle),
+					      prof->id);
+		if (!status)
+			set_bit(vsi_handle, prof->vsis);
+		else
+			ice_debug(hw, ICE_DBG_FLOW,
+				  "HW profile add failed, %d\n",
+				  status);
+	}
+
+	return status;
+}
+
 /**
  * ice_flow_add_prof - Add a flow profile for packet segments and matched fields
  * @hw: pointer to the HW struct
@@ -458,12 +490,13 @@ out:
  * @prof_id: unique ID to identify this flow profile
  * @segs: array of one or more packet segments that describe the flow
  * @segs_cnt: number of packet segments provided
+ * @prof: stores the returned flow profile added
  */
 static enum ice_status
 ice_flow_add_prof(struct ice_hw *hw, enum ice_block blk, enum ice_flow_dir dir,
-		  u64 prof_id, struct ice_flow_seg_info *segs, u8 segs_cnt)
+		  u64 prof_id, struct ice_flow_seg_info *segs, u8 segs_cnt,
+		  struct ice_flow_prof **prof)
 {
-	struct ice_flow_prof *prof = NULL;
 	enum ice_status status;
 
 	if (segs_cnt > ICE_FLOW_SEG_MAX)
@@ -482,9 +515,9 @@ ice_flow_add_prof(struct ice_hw *hw, enum ice_block blk, enum ice_flow_dir dir,
 	mutex_lock(&hw->fl_profs_locks[blk]);
 
 	status = ice_flow_add_prof_sync(hw, blk, dir, prof_id, segs, segs_cnt,
-					&prof);
+					prof);
 	if (!status)
-		list_add(&prof->l_entry, &hw->fl_profs[blk]);
+		list_add(&(*prof)->l_entry, &hw->fl_profs[blk]);
 
 	mutex_unlock(&hw->fl_profs_locks[blk]);
 
@@ -634,6 +667,7 @@ ice_flow_set_rss_seg_info(struct ice_flow_seg_info *segs, u64 hash_fields,
 /**
  * ice_add_rss_cfg_sync - add an RSS configuration
  * @hw: pointer to the hardware structure
+ * @vsi_handle: software VSI handle
  * @hashed_flds: hash bit fields (ICE_FLOW_HASH_*) to configure
  * @addl_hdrs: protocol header fields
  * @segs_cnt: packet segment count
@@ -641,9 +675,11 @@ ice_flow_set_rss_seg_info(struct ice_flow_seg_info *segs, u64 hash_fields,
  * Assumption: lock has already been acquired for RSS list
  */
 static enum ice_status
-ice_add_rss_cfg_sync(struct ice_hw *hw, u64 hashed_flds, u32 addl_hdrs,
-		     u8 segs_cnt)
+ice_add_rss_cfg_sync(struct ice_hw *hw, u16 vsi_handle, u64 hashed_flds,
+		     u32 addl_hdrs, u8 segs_cnt)
 {
+	const enum ice_block blk = ICE_BLK_RSS;
+	struct ice_flow_prof *prof = NULL;
 	struct ice_flow_seg_info *segs;
 	enum ice_status status;
 
@@ -663,11 +699,15 @@ ice_add_rss_cfg_sync(struct ice_hw *hw, u64 hashed_flds, u32 addl_hdrs,
 	/* Create a new flow profile with generated profile and packet
 	 * segment information.
 	 */
-	status = ice_flow_add_prof(hw, ICE_BLK_RSS, ICE_FLOW_RX,
+	status = ice_flow_add_prof(hw, blk, ICE_FLOW_RX,
 				   ICE_FLOW_GEN_PROFID(hashed_flds,
 						       segs[segs_cnt - 1].hdrs,
 						       segs_cnt),
-				   segs, segs_cnt);
+				   segs, segs_cnt, &prof);
+	if (status)
+		goto exit;
+
+	status = ice_flow_assoc_prof(hw, blk, prof, vsi_handle);
 
 exit:
 	kfree(segs);
@@ -696,7 +736,7 @@ ice_add_rss_cfg(struct ice_hw *hw, u16 vsi_handle, u64 hashed_flds,
 		return ICE_ERR_PARAM;
 
 	mutex_lock(&hw->rss_locks);
-	status = ice_add_rss_cfg_sync(hw, hashed_flds, addl_hdrs,
+	status = ice_add_rss_cfg_sync(hw, vsi_handle, hashed_flds, addl_hdrs,
 				      ICE_RSS_OUTER_HEADERS);
 	mutex_unlock(&hw->rss_locks);
 
@@ -719,7 +759,8 @@ enum ice_status ice_replay_rss_cfg(struct ice_hw *hw, u16 vsi_handle)
 	mutex_lock(&hw->rss_locks);
 	list_for_each_entry(r, &hw->rss_list_head, l_entry) {
 		if (test_bit(vsi_handle, r->vsis)) {
-			status = ice_add_rss_cfg_sync(hw, r->hashed_flds,
+			status = ice_add_rss_cfg_sync(hw, vsi_handle,
+						      r->hashed_flds,
 						      r->packet_hdr,
 						      ICE_RSS_OUTER_HEADERS);
 			if (status)
diff --git a/drivers/net/ethernet/intel/ice/ice_status.h b/drivers/net/ethernet/intel/ice/ice_status.h
index c01597885629..a9a8bc3aca42 100644
--- a/drivers/net/ethernet/intel/ice/ice_status.h
+++ b/drivers/net/ethernet/intel/ice/ice_status.h
@@ -26,6 +26,7 @@ enum ice_status {
 	ICE_ERR_IN_USE				= -16,
 	ICE_ERR_MAX_LIMIT			= -17,
 	ICE_ERR_RESET_ONGOING			= -18,
+	ICE_ERR_HW_TABLE			= -19,
 	ICE_ERR_NVM_CHECKSUM			= -51,
 	ICE_ERR_BUF_TOO_SHORT			= -52,
 	ICE_ERR_NVM_BLANK_MODE			= -53,
-- 
cgit v1.2.3-59-g8ed1b


From 43dbfc7bb8f0bd79a98f3fd628266a853d4bae61 Mon Sep 17 00:00:00 2001
From: Tony Nguyen <anthony.l.nguyen@intel.com>
Date: Fri, 17 Jan 2020 07:39:15 -0800
Subject: ice: Enable writing filtering tables

Write the hardware tables based on the populated software structures.

Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Signed-off-by: Henry Tieman <henry.w.tieman@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_adminq_cmd.h |   1 +
 drivers/net/ethernet/intel/ice/ice_flex_pipe.c  | 604 ++++++++++++++++++++++++
 drivers/net/ethernet/intel/ice/ice_flex_type.h  |  46 ++
 3 files changed, 651 insertions(+)

diff --git a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
index 9967973a7c42..4459bc564b11 100644
--- a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
@@ -1856,6 +1856,7 @@ enum ice_adminq_opc {
 
 	/* package commands */
 	ice_aqc_opc_download_pkg			= 0x0C40,
+	ice_aqc_opc_update_pkg				= 0x0C42,
 	ice_aqc_opc_get_pkg_info_list			= 0x0C43,
 
 	/* debug commands */
diff --git a/drivers/net/ethernet/intel/ice/ice_flex_pipe.c b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c
index 8c93c303a4a5..6dca611b408a 100644
--- a/drivers/net/ethernet/intel/ice/ice_flex_pipe.c
+++ b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c
@@ -5,6 +5,86 @@
 #include "ice_flex_pipe.h"
 #include "ice_flow.h"
 
+static const u32 ice_sect_lkup[ICE_BLK_COUNT][ICE_SECT_COUNT] = {
+	/* SWITCH */
+	{
+		ICE_SID_XLT0_SW,
+		ICE_SID_XLT_KEY_BUILDER_SW,
+		ICE_SID_XLT1_SW,
+		ICE_SID_XLT2_SW,
+		ICE_SID_PROFID_TCAM_SW,
+		ICE_SID_PROFID_REDIR_SW,
+		ICE_SID_FLD_VEC_SW,
+		ICE_SID_CDID_KEY_BUILDER_SW,
+		ICE_SID_CDID_REDIR_SW
+	},
+
+	/* ACL */
+	{
+		ICE_SID_XLT0_ACL,
+		ICE_SID_XLT_KEY_BUILDER_ACL,
+		ICE_SID_XLT1_ACL,
+		ICE_SID_XLT2_ACL,
+		ICE_SID_PROFID_TCAM_ACL,
+		ICE_SID_PROFID_REDIR_ACL,
+		ICE_SID_FLD_VEC_ACL,
+		ICE_SID_CDID_KEY_BUILDER_ACL,
+		ICE_SID_CDID_REDIR_ACL
+	},
+
+	/* FD */
+	{
+		ICE_SID_XLT0_FD,
+		ICE_SID_XLT_KEY_BUILDER_FD,
+		ICE_SID_XLT1_FD,
+		ICE_SID_XLT2_FD,
+		ICE_SID_PROFID_TCAM_FD,
+		ICE_SID_PROFID_REDIR_FD,
+		ICE_SID_FLD_VEC_FD,
+		ICE_SID_CDID_KEY_BUILDER_FD,
+		ICE_SID_CDID_REDIR_FD
+	},
+
+	/* RSS */
+	{
+		ICE_SID_XLT0_RSS,
+		ICE_SID_XLT_KEY_BUILDER_RSS,
+		ICE_SID_XLT1_RSS,
+		ICE_SID_XLT2_RSS,
+		ICE_SID_PROFID_TCAM_RSS,
+		ICE_SID_PROFID_REDIR_RSS,
+		ICE_SID_FLD_VEC_RSS,
+		ICE_SID_CDID_KEY_BUILDER_RSS,
+		ICE_SID_CDID_REDIR_RSS
+	},
+
+	/* PE */
+	{
+		ICE_SID_XLT0_PE,
+		ICE_SID_XLT_KEY_BUILDER_PE,
+		ICE_SID_XLT1_PE,
+		ICE_SID_XLT2_PE,
+		ICE_SID_PROFID_TCAM_PE,
+		ICE_SID_PROFID_REDIR_PE,
+		ICE_SID_FLD_VEC_PE,
+		ICE_SID_CDID_KEY_BUILDER_PE,
+		ICE_SID_CDID_REDIR_PE
+	}
+};
+
+/**
+ * ice_sect_id - returns section ID
+ * @blk: block type
+ * @sect: section type
+ *
+ * This helper function returns the proper section ID given a block type and a
+ * section type.
+ */
+static u32 ice_sect_id(enum ice_block blk, enum ice_sect sect)
+{
+	return ice_sect_lkup[blk][sect];
+}
+
 /**
  * ice_pkg_val_buf
  * @buf: pointer to the ice buffer
@@ -375,6 +455,31 @@ static void ice_release_global_cfg_lock(struct ice_hw *hw)
 	ice_release_res(hw, ICE_GLOBAL_CFG_LOCK_RES_ID);
 }
 
+/**
+ * ice_acquire_change_lock
+ * @hw: pointer to the HW structure
+ * @access: access type (read or write)
+ *
+ * This function will request ownership of the change lock.
+ */
+static enum ice_status
+ice_acquire_change_lock(struct ice_hw *hw, enum ice_aq_res_access_type access)
+{
+	return ice_acquire_res(hw, ICE_CHANGE_LOCK_RES_ID, access,
+			       ICE_CHANGE_LOCK_TIMEOUT);
+}
+
+/**
+ * ice_release_change_lock
+ * @hw: pointer to the HW structure
+ *
+ * This function will release the change lock using the proper Admin Command.
+ */
+static void ice_release_change_lock(struct ice_hw *hw)
+{
+	ice_release_res(hw, ICE_CHANGE_LOCK_RES_ID);
+}
+
 /**
  * ice_aq_download_pkg
  * @hw: pointer to the hardware structure
@@ -423,6 +528,54 @@ ice_aq_download_pkg(struct ice_hw *hw, struct ice_buf_hdr *pkg_buf,
 	return status;
 }
 
+/**
+ * ice_aq_update_pkg
+ * @hw: pointer to the hardware structure
+ * @pkg_buf: the package cmd buffer
+ * @buf_size: the size of the package cmd buffer
+ * @last_buf: last buffer indicator
+ * @error_offset: returns error offset
+ * @error_info: returns error information
+ * @cd: pointer to command details structure or NULL
+ *
+ * Update Package (0x0C42)
+ */
+static enum ice_status
+ice_aq_update_pkg(struct ice_hw *hw, struct ice_buf_hdr *pkg_buf, u16 buf_size,
+		  bool last_buf, u32 *error_offset, u32 *error_info,
+		  struct ice_sq_cd *cd)
+{
+	struct ice_aqc_download_pkg *cmd;
+	struct ice_aq_desc desc;
+	enum ice_status status;
+
+	if (error_offset)
+		*error_offset = 0;
+	if (error_info)
+		*error_info = 0;
+
+	cmd = &desc.params.download_pkg;
+	ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_update_pkg);
+	desc.flags |= cpu_to_le16(ICE_AQ_FLAG_RD);
+
+	if (last_buf)
+		cmd->flags |= ICE_AQC_DOWNLOAD_PKG_LAST_BUF;
+
+	status = ice_aq_send_cmd(hw, &desc, pkg_buf, buf_size, cd);
+	if (status == ICE_ERR_AQ_ERROR) {
+		/* Read error from buffer only when the FW returned an error */
+		struct ice_aqc_download_pkg_resp *resp;
+
+		resp = (struct ice_aqc_download_pkg_resp *)pkg_buf;
+		if (error_offset)
+			*error_offset = le32_to_cpu(resp->error_offset);
+		if (error_info)
+			*error_info = le32_to_cpu(resp->error_info);
+	}
+
+	return status;
+}
+
 /**
  * ice_find_seg_in_pkg
  * @hw: pointer to the hardware structure
@@ -457,6 +610,44 @@ ice_find_seg_in_pkg(struct ice_hw *hw, u32 seg_type,
 	return NULL;
 }
 
+/**
+ * ice_update_pkg
+ * @hw: pointer to the hardware structure
+ * @bufs: pointer to an array of buffers
+ * @count: the number of buffers in the array
+ *
+ * Obtains change lock and updates package.
+ */
+static enum ice_status
+ice_update_pkg(struct ice_hw *hw, struct ice_buf *bufs, u32 count)
+{
+	enum ice_status status;
+	u32 offset, info, i;
+
+	status = ice_acquire_change_lock(hw, ICE_RES_WRITE);
+	if (status)
+		return status;
+
+	for (i = 0; i < count; i++) {
+		struct ice_buf_hdr *bh = (struct ice_buf_hdr *)(bufs + i);
+		bool last = ((i + 1) == count);
+
+		status = ice_aq_update_pkg(hw, bh, le16_to_cpu(bh->data_end),
+					   last, &offset, &info, NULL);
+
+		if (status) {
+			ice_debug(hw, ICE_DBG_PKG,
+				  "Update pkg failed: err %d off %d inf %d\n",
+				  status, offset, info);
+			break;
+		}
+	}
+
+	ice_release_change_lock(hw);
+
+	return status;
+}
+
 /**
  * ice_dwnld_cfg_bufs
  * @hw: pointer to the hardware structure
@@ -938,6 +1129,169 @@ enum ice_status ice_copy_and_init_pkg(struct ice_hw *hw, const u8 *buf, u32 len)
 	return status;
 }
 
+/**
+ * ice_pkg_buf_alloc
+ * @hw: pointer to the HW structure
+ *
+ * Allocates a package buffer and returns a pointer to the buffer header.
+ * Note: all package contents must be in Little Endian form.
+ */
+static struct ice_buf_build *ice_pkg_buf_alloc(struct ice_hw *hw)
+{
+	struct ice_buf_build *bld;
+	struct ice_buf_hdr *buf;
+
+	bld = devm_kzalloc(ice_hw_to_dev(hw), sizeof(*bld), GFP_KERNEL);
+	if (!bld)
+		return NULL;
+
+	buf = (struct ice_buf_hdr *)bld;
+	buf->data_end = cpu_to_le16(offsetof(struct ice_buf_hdr,
+					     section_entry));
+	return bld;
+}
+
+/**
+ * ice_pkg_buf_free
+ * @hw: pointer to the HW structure
+ * @bld: pointer to pkg build (allocated by ice_pkg_buf_alloc())
+ *
+ * Frees a package buffer
+ */
+static void ice_pkg_buf_free(struct ice_hw *hw, struct ice_buf_build *bld)
+{
+	devm_kfree(ice_hw_to_dev(hw), bld);
+}
+
+/**
+ * ice_pkg_buf_reserve_section
+ * @bld: pointer to pkg build (allocated by ice_pkg_buf_alloc())
+ * @count: the number of sections to reserve
+ *
+ * Reserves one or more section table entries in a package buffer. This routine
+ * can be called multiple times as long as they are made before calling
+ * ice_pkg_buf_alloc_section(). Once ice_pkg_buf_alloc_section()
+ * is called once, the number of sections that can be allocated will not be able
+ * to be increased; not using all reserved sections is fine, but this will
+ * result in some wasted space in the buffer.
+ * Note: all package contents must be in Little Endian form.
+ */
+static enum ice_status
+ice_pkg_buf_reserve_section(struct ice_buf_build *bld, u16 count)
+{
+	struct ice_buf_hdr *buf;
+	u16 section_count;
+	u16 data_end;
+
+	if (!bld)
+		return ICE_ERR_PARAM;
+
+	buf = (struct ice_buf_hdr *)&bld->buf;
+
+	/* already an active section, can't increase table size */
+	section_count = le16_to_cpu(buf->section_count);
+	if (section_count > 0)
+		return ICE_ERR_CFG;
+
+	if (bld->reserved_section_table_entries + count > ICE_MAX_S_COUNT)
+		return ICE_ERR_CFG;
+	bld->reserved_section_table_entries += count;
+
+	data_end = le16_to_cpu(buf->data_end) +
+		   (count * sizeof(buf->section_entry[0]));
+	buf->data_end = cpu_to_le16(data_end);
+
+	return 0;
+}
+
+/**
+ * ice_pkg_buf_alloc_section
+ * @bld: pointer to pkg build (allocated by ice_pkg_buf_alloc())
+ * @type: the section type value
+ * @size: the size of the section to reserve (in bytes)
+ *
+ * Reserves memory in the buffer for a section's content and updates the
+ * buffers' status accordingly. This routine returns a pointer to the first
+ * byte of the section start within the buffer, which is used to fill in the
+ * section contents.
+ * Note: all package contents must be in Little Endian form.
+ */
+static void *
+ice_pkg_buf_alloc_section(struct ice_buf_build *bld, u32 type, u16 size)
+{
+	struct ice_buf_hdr *buf;
+	u16 sect_count;
+	u16 data_end;
+
+	if (!bld || !type || !size)
+		return NULL;
+
+	buf = (struct ice_buf_hdr *)&bld->buf;
+
+	/* check for enough space left in buffer */
+	data_end = le16_to_cpu(buf->data_end);
+
+	/* section start must align on 4 byte boundary */
+	data_end = ALIGN(data_end, 4);
+
+	if ((data_end + size) > ICE_MAX_S_DATA_END)
+		return NULL;
+
+	/* check for more available section table entries */
+	sect_count = le16_to_cpu(buf->section_count);
+	if (sect_count < bld->reserved_section_table_entries) {
+		void *section_ptr = ((u8 *)buf) + data_end;
+
+		buf->section_entry[sect_count].offset = cpu_to_le16(data_end);
+		buf->section_entry[sect_count].size = cpu_to_le16(size);
+		buf->section_entry[sect_count].type = cpu_to_le32(type);
+
+		data_end += size;
+		buf->data_end = cpu_to_le16(data_end);
+
+		buf->section_count = cpu_to_le16(sect_count + 1);
+		return section_ptr;
+	}
+
+	/* no free section table entries */
+	return NULL;
+}
+
+/**
+ * ice_pkg_buf_get_active_sections
+ * @bld: pointer to pkg build (allocated by ice_pkg_buf_alloc())
+ *
+ * Returns the number of active sections. Before using the package buffer
+ * in an update package command, the caller should make sure that there is at
+ * least one active section - otherwise, the buffer is not legal and should
+ * not be used.
+ * Note: all package contents must be in Little Endian form.
+ */
+static u16 ice_pkg_buf_get_active_sections(struct ice_buf_build *bld)
+{
+	struct ice_buf_hdr *buf;
+
+	if (!bld)
+		return 0;
+
+	buf = (struct ice_buf_hdr *)&bld->buf;
+	return le16_to_cpu(buf->section_count);
+}
+
+/**
+ * ice_pkg_buf
+ * @bld: pointer to pkg build (allocated by ice_pkg_buf_alloc())
+ *
+ * Return a pointer to the buffer's header
+ */
+static struct ice_buf *ice_pkg_buf(struct ice_buf_build *bld)
+{
+	if (!bld)
+		return NULL;
+
+	return &bld->buf;
+}
+
 /* PTG Management */
 
 /**
@@ -2207,6 +2561,252 @@ ice_has_prof_vsig(struct ice_hw *hw, enum ice_block blk, u16 vsig, u64 hdl)
 	return false;
 }
 
+/**
+ * ice_prof_bld_es - build profile ID extraction sequence changes
+ * @hw: pointer to the HW struct
+ * @blk: hardware block
+ * @bld: the update package buffer build to add to
+ * @chgs: the list of changes to make in hardware
+ */
+static enum ice_status
+ice_prof_bld_es(struct ice_hw *hw, enum ice_block blk,
+		struct ice_buf_build *bld, struct list_head *chgs)
+{
+	u16 vec_size = hw->blk[blk].es.fvw * sizeof(struct ice_fv_word);
+	struct ice_chs_chg *tmp;
+
+	list_for_each_entry(tmp, chgs, list_entry)
+		if (tmp->type == ICE_PTG_ES_ADD && tmp->add_prof) {
+			u16 off = tmp->prof_id * hw->blk[blk].es.fvw;
+			struct ice_pkg_es *p;
+			u32 id;
+
+			id = ice_sect_id(blk, ICE_VEC_TBL);
+			p = (struct ice_pkg_es *)
+				ice_pkg_buf_alloc_section(bld, id, sizeof(*p) +
+							  vec_size -
+							  sizeof(p->es[0]));
+
+			if (!p)
+				return ICE_ERR_MAX_LIMIT;
+
+			p->count = cpu_to_le16(1);
+			p->offset = cpu_to_le16(tmp->prof_id);
+
+			memcpy(p->es, &hw->blk[blk].es.t[off], vec_size);
+		}
+
+	return 0;
+}
+
+/**
+ * ice_prof_bld_tcam - build profile ID TCAM changes
+ * @hw: pointer to the HW struct
+ * @blk: hardware block
+ * @bld: the update package buffer build to add to
+ * @chgs: the list of changes to make in hardware
+ */
+static enum ice_status
+ice_prof_bld_tcam(struct ice_hw *hw, enum ice_block blk,
+		  struct ice_buf_build *bld, struct list_head *chgs)
+{
+	struct ice_chs_chg *tmp;
+
+	list_for_each_entry(tmp, chgs, list_entry)
+		if (tmp->type == ICE_TCAM_ADD && tmp->add_tcam_idx) {
+			struct ice_prof_id_section *p;
+			u32 id;
+
+			id = ice_sect_id(blk, ICE_PROF_TCAM);
+			p = (struct ice_prof_id_section *)
+				ice_pkg_buf_alloc_section(bld, id, sizeof(*p));
+
+			if (!p)
+				return ICE_ERR_MAX_LIMIT;
+
+			p->count = cpu_to_le16(1);
+			p->entry[0].addr = cpu_to_le16(tmp->tcam_idx);
+			p->entry[0].prof_id = tmp->prof_id;
+
+			memcpy(p->entry[0].key,
+			       &hw->blk[blk].prof.t[tmp->tcam_idx].key,
+			       sizeof(hw->blk[blk].prof.t->key));
+		}
+
+	return 0;
+}
+
+/**
+ * ice_prof_bld_xlt1 - build XLT1 changes
+ * @blk: hardware block
+ * @bld: the update package buffer build to add to
+ * @chgs: the list of changes to make in hardware
+ */
+static enum ice_status
+ice_prof_bld_xlt1(enum ice_block blk, struct ice_buf_build *bld,
+		  struct list_head *chgs)
+{
+	struct ice_chs_chg *tmp;
+
+	list_for_each_entry(tmp, chgs, list_entry)
+		if (tmp->type == ICE_PTG_ES_ADD && tmp->add_ptg) {
+			struct ice_xlt1_section *p;
+			u32 id;
+
+			id = ice_sect_id(blk, ICE_XLT1);
+			p = (struct ice_xlt1_section *)
+				ice_pkg_buf_alloc_section(bld, id, sizeof(*p));
+
+			if (!p)
+				return ICE_ERR_MAX_LIMIT;
+
+			p->count = cpu_to_le16(1);
+			p->offset = cpu_to_le16(tmp->ptype);
+			p->value[0] = tmp->ptg;
+		}
+
+	return 0;
+}
+
+/**
+ * ice_prof_bld_xlt2 - build XLT2 changes
+ * @blk: hardware block
+ * @bld: the update package buffer build to add to
+ * @chgs: the list of changes to make in hardware
+ */
+static enum ice_status
+ice_prof_bld_xlt2(enum ice_block blk, struct ice_buf_build *bld,
+		  struct list_head *chgs)
+{
+	struct ice_chs_chg *tmp;
+
+	list_for_each_entry(tmp, chgs, list_entry) {
+		struct ice_xlt2_section *p;
+		u32 id;
+
+		switch (tmp->type) {
+		case ICE_VSIG_ADD:
+		case ICE_VSI_MOVE:
+		case ICE_VSIG_REM:
+			id = ice_sect_id(blk, ICE_XLT2);
+			p = (struct ice_xlt2_section *)
+				ice_pkg_buf_alloc_section(bld, id, sizeof(*p));
+
+			if (!p)
+				return ICE_ERR_MAX_LIMIT;
+
+			p->count = cpu_to_le16(1);
+			p->offset = cpu_to_le16(tmp->vsi);
+			p->value[0] = cpu_to_le16(tmp->vsig);
+			break;
+		default:
+			break;
+		}
+	}
+
+	return 0;
+}
+
+/**
+ * ice_upd_prof_hw - update hardware using the change list
+ * @hw: pointer to the HW struct
+ * @blk: hardware block
+ * @chgs: the list of changes to make in hardware
+ */
+static enum ice_status
+ice_upd_prof_hw(struct ice_hw *hw, enum ice_block blk,
+		struct list_head *chgs)
+{
+	struct ice_buf_build *b;
+	struct ice_chs_chg *tmp;
+	enum ice_status status;
+	u16 pkg_sects;
+	u16 xlt1 = 0;
+	u16 xlt2 = 0;
+	u16 tcam = 0;
+	u16 es = 0;
+	u16 sects;
+
+	/* count number of sections we need */
+	list_for_each_entry(tmp, chgs, list_entry) {
+		switch (tmp->type) {
+		case ICE_PTG_ES_ADD:
+			if (tmp->add_ptg)
+				xlt1++;
+			if (tmp->add_prof)
+				es++;
+			break;
+		case ICE_TCAM_ADD:
+			tcam++;
+			break;
+		case ICE_VSIG_ADD:
+		case ICE_VSI_MOVE:
+		case ICE_VSIG_REM:
+			xlt2++;
+			break;
+		default:
+			break;
+		}
+	}
+	sects = xlt1 + xlt2 + tcam + es;
+
+	if (!sects)
+		return 0;
+
+	/* Build update package buffer */
+	b = ice_pkg_buf_alloc(hw);
+	if (!b)
+		return ICE_ERR_NO_MEMORY;
+
+	status = ice_pkg_buf_reserve_section(b, sects);
+	if (status)
+		goto error_tmp;
+
+	/* Preserve order of table update: ES, TCAM, PTG, VSIG */
+	if (es) {
+		status = ice_prof_bld_es(hw, blk, b, chgs);
+		if (status)
+			goto error_tmp;
+	}
+
+	if (tcam) {
+		status = ice_prof_bld_tcam(hw, blk, b, chgs);
+		if (status)
+			goto error_tmp;
+	}
+
+	if (xlt1) {
+		status = ice_prof_bld_xlt1(blk, b, chgs);
+		if (status)
+			goto error_tmp;
+	}
+
+	if (xlt2) {
+		status = ice_prof_bld_xlt2(blk, b, chgs);
+		if (status)
+			goto error_tmp;
+	}
+
+	/* After package buffer build check if the section count in buffer is
+	 * non-zero and matches the number of sections detected for package
+	 * update.
+	 */
+	pkg_sects = ice_pkg_buf_get_active_sections(b);
+	if (!pkg_sects || pkg_sects != sects) {
+		status = ICE_ERR_INVAL_SIZE;
+		goto error_tmp;
+	}
+
+	/* update package */
+	status = ice_update_pkg(hw, ice_pkg_buf(b), 1);
+	if (status == ICE_ERR_AQ_ERROR)
+		ice_debug(hw, ICE_DBG_INIT, "Unable to update HW profile\n");
+
+error_tmp:
+	ice_pkg_buf_free(hw, b);
+	return status;
+}
+
 /**
  * ice_add_prof - add profile
  * @hw: pointer to the HW struct
@@ -3097,6 +3697,10 @@ ice_add_prof_id_flow(struct ice_hw *hw, enum ice_block blk, u16 vsi, u64 hdl)
 		}
 	}
 
+	/* update hardware */
+	if (!status)
+		status = ice_upd_prof_hw(hw, blk, &chg);
+
 err_ice_add_prof_id_flow:
 	list_for_each_entry_safe(del, tmp, &chg, list_entry) {
 		list_del(&del->list_entry);
diff --git a/drivers/net/ethernet/intel/ice/ice_flex_type.h b/drivers/net/ethernet/intel/ice/ice_flex_type.h
index 9d95d51bc760..0fb3fe3ff3ea 100644
--- a/drivers/net/ethernet/intel/ice/ice_flex_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_flex_type.h
@@ -108,37 +108,57 @@ struct ice_buf_hdr {
 	sizeof(struct ice_buf_hdr) - (hd_sz)) / (ent_sz))
 
 /* ice package section IDs */
+#define ICE_SID_XLT0_SW			10
+#define ICE_SID_XLT_KEY_BUILDER_SW	11
 #define ICE_SID_XLT1_SW			12
 #define ICE_SID_XLT2_SW			13
 #define ICE_SID_PROFID_TCAM_SW		14
 #define ICE_SID_PROFID_REDIR_SW		15
 #define ICE_SID_FLD_VEC_SW		16
+#define ICE_SID_CDID_KEY_BUILDER_SW	17
+#define ICE_SID_CDID_REDIR_SW		18
 
+#define ICE_SID_XLT0_ACL		20
+#define ICE_SID_XLT_KEY_BUILDER_ACL	21
 #define ICE_SID_XLT1_ACL		22
 #define ICE_SID_XLT2_ACL		23
 #define ICE_SID_PROFID_TCAM_ACL		24
 #define ICE_SID_PROFID_REDIR_ACL	25
 #define ICE_SID_FLD_VEC_ACL		26
+#define ICE_SID_CDID_KEY_BUILDER_ACL	27
+#define ICE_SID_CDID_REDIR_ACL		28
 
+#define ICE_SID_XLT0_FD			30
+#define ICE_SID_XLT_KEY_BUILDER_FD	31
 #define ICE_SID_XLT1_FD			32
 #define ICE_SID_XLT2_FD			33
 #define ICE_SID_PROFID_TCAM_FD		34
 #define ICE_SID_PROFID_REDIR_FD		35
 #define ICE_SID_FLD_VEC_FD		36
+#define ICE_SID_CDID_KEY_BUILDER_FD	37
+#define ICE_SID_CDID_REDIR_FD		38
 
+#define ICE_SID_XLT0_RSS		40
+#define ICE_SID_XLT_KEY_BUILDER_RSS	41
 #define ICE_SID_XLT1_RSS		42
 #define ICE_SID_XLT2_RSS		43
 #define ICE_SID_PROFID_TCAM_RSS		44
 #define ICE_SID_PROFID_REDIR_RSS	45
 #define ICE_SID_FLD_VEC_RSS		46
+#define ICE_SID_CDID_KEY_BUILDER_RSS	47
+#define ICE_SID_CDID_REDIR_RSS		48
 
 #define ICE_SID_RXPARSER_BOOST_TCAM	56
 
+#define ICE_SID_XLT0_PE			80
+#define ICE_SID_XLT_KEY_BUILDER_PE	81
 #define ICE_SID_XLT1_PE			82
 #define ICE_SID_XLT2_PE			83
 #define ICE_SID_PROFID_TCAM_PE		84
 #define ICE_SID_PROFID_REDIR_PE		85
 #define ICE_SID_FLD_VEC_PE		86
+#define ICE_SID_CDID_KEY_BUILDER_PE	87
+#define ICE_SID_CDID_REDIR_PE		88
 
 /* Label Metadata section IDs */
 #define ICE_SID_LBL_FIRST		0x80000010
@@ -155,6 +175,19 @@ enum ice_block {
 	ICE_BLK_COUNT
 };
 
+enum ice_sect {
+	ICE_XLT0 = 0,
+	ICE_XLT_KB,
+	ICE_XLT1,
+	ICE_XLT2,
+	ICE_PROF_TCAM,
+	ICE_PROF_REDIR,
+	ICE_VEC_TBL,
+	ICE_CDID_KB,
+	ICE_CDID_REDIR,
+	ICE_SECT_COUNT
+};
+
 /* package labels */
 struct ice_label {
 	__le16 value;
@@ -237,6 +270,13 @@ struct ice_prof_redir_section {
 	u8 redir_value[1];
 };
 
+/* package buffer building */
+
+struct ice_buf_build {
+	struct ice_buf buf;
+	u16 reserved_section_table_entries;
+};
+
 struct ice_pkg_enum {
 	struct ice_buf_table *buf_table;
 	u32 buf_idx;
@@ -251,6 +291,12 @@ struct ice_pkg_enum {
 	void *(*handler)(u32 sect_type, void *section, u32 index, u32 *offset);
 };
 
+struct ice_pkg_es {
+	__le16 count;
+	__le16 offset;
+	struct ice_fv_word es[1];
+};
+
 struct ice_es {
 	u32 sid;
 	u16 count;
-- 
cgit v1.2.3-59-g8ed1b


From 2c61054c5fdaf3df267c4fa95c8cfc33f7665615 Mon Sep 17 00:00:00 2001
From: Tony Nguyen <anthony.l.nguyen@intel.com>
Date: Fri, 17 Jan 2020 07:39:16 -0800
Subject: ice: Optimize table usage

Attempt to optimize TCAM entries and reduce table resource usage by
searching for profiles that can be reused. Provide resource cleanup
of both hardware and software structures.

Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Signed-off-by: Henry Tieman <henry.w.tieman@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_flex_pipe.c | 394 +++++++++++++++++++++++++
 drivers/net/ethernet/intel/ice/ice_flex_pipe.h |   4 +
 drivers/net/ethernet/intel/ice/ice_flow.c      | 338 +++++++++++++++++++++
 drivers/net/ethernet/intel/ice/ice_flow.h      |  10 +
 drivers/net/ethernet/intel/ice/ice_lib.c       |  28 +-
 5 files changed, 773 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_flex_pipe.c b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c
index 6dca611b408a..99208946224c 100644
--- a/drivers/net/ethernet/intel/ice/ice_flex_pipe.c
+++ b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c
@@ -1921,6 +1921,26 @@ ice_alloc_prof_id(struct ice_hw *hw, enum ice_block blk, u8 *prof_id)
 	return status;
 }
 
+/**
+ * ice_free_prof_id - free profile ID
+ * @hw: pointer to the HW struct
+ * @blk: the block from which to free the profile ID
+ * @prof_id: the profile ID to free
+ *
+ * This function frees a profile ID, which also corresponds to a Field Vector.
+ */
+static enum ice_status
+ice_free_prof_id(struct ice_hw *hw, enum ice_block blk, u8 prof_id)
+{
+	u16 tmp_prof_id = (u16)prof_id;
+	u16 res_type;
+
+	if (!ice_prof_id_rsrc_type(blk, &res_type))
+		return ICE_ERR_PARAM;
+
+	return ice_free_hw_res(hw, res_type, 1, &tmp_prof_id);
+}
+
 /**
  * ice_prof_inc_ref - increment reference count for profile
  * @hw: pointer to the HW struct
@@ -1962,6 +1982,28 @@ ice_write_es(struct ice_hw *hw, enum ice_block blk, u8 prof_id,
 	}
 }
 
+/**
+ * ice_prof_dec_ref - decrement reference count for profile
+ * @hw: pointer to the HW struct
+ * @blk: the block from which to free the profile ID
+ * @prof_id: the profile ID for which to decrement the reference count
+ */
+static enum ice_status
+ice_prof_dec_ref(struct ice_hw *hw, enum ice_block blk, u8 prof_id)
+{
+	if (prof_id > hw->blk[blk].es.count)
+		return ICE_ERR_PARAM;
+
+	if (hw->blk[blk].es.ref_count[prof_id] > 0) {
+		if (!--hw->blk[blk].es.ref_count[prof_id]) {
+			ice_write_es(hw, blk, prof_id, NULL);
+			return ice_free_prof_id(hw, blk, prof_id);
+		}
+	}
+
+	return 0;
+}
+
 /* Block / table section IDs */
 static const u32 ice_blk_sids[ICE_BLK_COUNT][ICE_SID_OFF_COUNT] = {
 	/* SWITCH */
@@ -2218,6 +2260,64 @@ void ice_fill_blk_tbls(struct ice_hw *hw)
 	ice_init_sw_db(hw);
 }
 
+/**
+ * ice_free_prof_map - free profile map
+ * @hw: pointer to the hardware structure
+ * @blk_idx: HW block index
+ */
+static void ice_free_prof_map(struct ice_hw *hw, u8 blk_idx)
+{
+	struct ice_es *es = &hw->blk[blk_idx].es;
+	struct ice_prof_map *del, *tmp;
+
+	mutex_lock(&es->prof_map_lock);
+	list_for_each_entry_safe(del, tmp, &es->prof_map, list) {
+		list_del(&del->list);
+		devm_kfree(ice_hw_to_dev(hw), del);
+	}
+	INIT_LIST_HEAD(&es->prof_map);
+	mutex_unlock(&es->prof_map_lock);
+}
+
+/**
+ * ice_free_flow_profs - free flow profile entries
+ * @hw: pointer to the hardware structure
+ * @blk_idx: HW block index
+ */
+static void ice_free_flow_profs(struct ice_hw *hw, u8 blk_idx)
+{
+	struct ice_flow_prof *p, *tmp;
+
+	mutex_lock(&hw->fl_profs_locks[blk_idx]);
+	list_for_each_entry_safe(p, tmp, &hw->fl_profs[blk_idx], l_entry) {
+		list_del(&p->l_entry);
+		devm_kfree(ice_hw_to_dev(hw), p);
+	}
+	mutex_unlock(&hw->fl_profs_locks[blk_idx]);
+
+	/* if driver is in reset and tables are being cleared
+	 * re-initialize the flow profile list heads
+	 */
+	INIT_LIST_HEAD(&hw->fl_profs[blk_idx]);
+}
+
+/**
+ * ice_free_vsig_tbl - free complete VSIG table entries
+ * @hw: pointer to the hardware structure
+ * @blk: the HW block on which to free the VSIG table entries
+ */
+static void ice_free_vsig_tbl(struct ice_hw *hw, enum ice_block blk)
+{
+	u16 i;
+
+	if (!hw->blk[blk].xlt2.vsig_tbl)
+		return;
+
+	for (i = 1; i < ICE_MAX_VSIGS; i++)
+		if (hw->blk[blk].xlt2.vsig_tbl[i].in_use)
+			ice_vsig_free(hw, blk, i);
+}
+
 /**
  * ice_free_hw_tbls - free hardware table memory
  * @hw: pointer to the hardware structure
@@ -2231,11 +2331,15 @@ void ice_free_hw_tbls(struct ice_hw *hw)
 		if (hw->blk[i].is_list_init) {
 			struct ice_es *es = &hw->blk[i].es;
 
+			ice_free_prof_map(hw, i);
 			mutex_destroy(&es->prof_map_lock);
+
+			ice_free_flow_profs(hw, i);
 			mutex_destroy(&hw->fl_profs_locks[i]);
 
 			hw->blk[i].is_list_init = false;
 		}
+		ice_free_vsig_tbl(hw, (enum ice_block)i);
 		devm_kfree(ice_hw_to_dev(hw), hw->blk[i].xlt1.ptypes);
 		devm_kfree(ice_hw_to_dev(hw), hw->blk[i].xlt1.ptg_tbl);
 		devm_kfree(ice_hw_to_dev(hw), hw->blk[i].xlt1.t);
@@ -2283,6 +2387,13 @@ void ice_clear_hw_tbls(struct ice_hw *hw)
 		struct ice_xlt2 *xlt2 = &hw->blk[i].xlt2;
 		struct ice_es *es = &hw->blk[i].es;
 
+		if (hw->blk[i].is_list_init) {
+			ice_free_prof_map(hw, i);
+			ice_free_flow_profs(hw, i);
+		}
+
+		ice_free_vsig_tbl(hw, (enum ice_block)i);
+
 		memset(xlt1->ptypes, 0, xlt1->count * sizeof(*xlt1->ptypes));
 		memset(xlt1->ptg_tbl, 0,
 		       ICE_MAX_PTGS * sizeof(*xlt1->ptg_tbl));
@@ -2956,6 +3067,25 @@ ice_search_prof_id(struct ice_hw *hw, enum ice_block blk, u64 id)
 	return entry;
 }
 
+/**
+ * ice_vsig_prof_id_count - count profiles in a VSIG
+ * @hw: pointer to the HW struct
+ * @blk: hardware block
+ * @vsig: VSIG to remove the profile from
+ */
+static u16
+ice_vsig_prof_id_count(struct ice_hw *hw, enum ice_block blk, u16 vsig)
+{
+	u16 idx = vsig & ICE_VSIG_IDX_M, count = 0;
+	struct ice_vsig_prof *p;
+
+	list_for_each_entry(p, &hw->blk[blk].xlt2.vsig_tbl[idx].prop_lst,
+			    list)
+		count++;
+
+	return count;
+}
+
 /**
  * ice_rel_tcam_idx - release a TCAM index
  * @hw: pointer to the HW struct
@@ -3064,6 +3194,117 @@ ice_rem_vsig(struct ice_hw *hw, enum ice_block blk, u16 vsig,
 	return ice_vsig_free(hw, blk, vsig);
 }
 
+/**
+ * ice_rem_prof_id_vsig - remove a specific profile from a VSIG
+ * @hw: pointer to the HW struct
+ * @blk: hardware block
+ * @vsig: VSIG to remove the profile from
+ * @hdl: profile handle indicating which profile to remove
+ * @chg: list to receive a record of changes
+ */
+static enum ice_status
+ice_rem_prof_id_vsig(struct ice_hw *hw, enum ice_block blk, u16 vsig, u64 hdl,
+		     struct list_head *chg)
+{
+	u16 idx = vsig & ICE_VSIG_IDX_M;
+	struct ice_vsig_prof *p, *t;
+	enum ice_status status;
+
+	list_for_each_entry_safe(p, t,
+				 &hw->blk[blk].xlt2.vsig_tbl[idx].prop_lst,
+				 list)
+		if (p->profile_cookie == hdl) {
+			if (ice_vsig_prof_id_count(hw, blk, vsig) == 1)
+				/* this is the last profile, remove the VSIG */
+				return ice_rem_vsig(hw, blk, vsig, chg);
+
+			status = ice_rem_prof_id(hw, blk, p);
+			if (!status) {
+				list_del(&p->list);
+				devm_kfree(ice_hw_to_dev(hw), p);
+			}
+			return status;
+		}
+
+	return ICE_ERR_DOES_NOT_EXIST;
+}
+
+/**
+ * ice_rem_flow_all - remove all flows with a particular profile
+ * @hw: pointer to the HW struct
+ * @blk: hardware block
+ * @id: profile tracking ID
+ */
+static enum ice_status
+ice_rem_flow_all(struct ice_hw *hw, enum ice_block blk, u64 id)
+{
+	struct ice_chs_chg *del, *tmp;
+	enum ice_status status;
+	struct list_head chg;
+	u16 i;
+
+	INIT_LIST_HEAD(&chg);
+
+	for (i = 1; i < ICE_MAX_VSIGS; i++)
+		if (hw->blk[blk].xlt2.vsig_tbl[i].in_use) {
+			if (ice_has_prof_vsig(hw, blk, i, id)) {
+				status = ice_rem_prof_id_vsig(hw, blk, i, id,
+							      &chg);
+				if (status)
+					goto err_ice_rem_flow_all;
+			}
+		}
+
+	status = ice_upd_prof_hw(hw, blk, &chg);
+
+err_ice_rem_flow_all:
+	list_for_each_entry_safe(del, tmp, &chg, list_entry) {
+		list_del(&del->list_entry);
+		devm_kfree(ice_hw_to_dev(hw), del);
+	}
+
+	return status;
+}
+
+/**
+ * ice_rem_prof - remove profile
+ * @hw: pointer to the HW struct
+ * @blk: hardware block
+ * @id: profile tracking ID
+ *
+ * This will remove the profile specified by the ID parameter, which was
+ * previously created through ice_add_prof. If any existing entries
+ * are associated with this profile, they will be removed as well.
+ */
+enum ice_status ice_rem_prof(struct ice_hw *hw, enum ice_block blk, u64 id)
+{
+	struct ice_prof_map *pmap;
+	enum ice_status status;
+
+	mutex_lock(&hw->blk[blk].es.prof_map_lock);
+
+	pmap = ice_search_prof_id_low(hw, blk, id);
+	if (!pmap) {
+		status = ICE_ERR_DOES_NOT_EXIST;
+		goto err_ice_rem_prof;
+	}
+
+	/* remove all flows with this profile */
+	status = ice_rem_flow_all(hw, blk, pmap->profile_cookie);
+	if (status)
+		goto err_ice_rem_prof;
+
+	/* dereference profile, and possibly remove */
+	ice_prof_dec_ref(hw, blk, pmap->prof_id);
+
+	list_del(&pmap->list);
+	devm_kfree(ice_hw_to_dev(hw), pmap);
+
+err_ice_rem_prof:
+	mutex_unlock(&hw->blk[blk].es.prof_map_lock);
+	return status;
+}
+
 /**
  * ice_get_prof - get profile
  * @hw: pointer to the HW struct
@@ -3714,3 +3955,156 @@ err_ice_add_prof_id_flow:
 
 	return status;
 }
+
+/**
+ * ice_rem_prof_from_list - remove a profile from list
+ * @hw: pointer to the HW struct
+ * @lst: list to remove the profile from
+ * @hdl: the profile handle indicating the profile to remove
+ */
+static enum ice_status
+ice_rem_prof_from_list(struct ice_hw *hw, struct list_head *lst, u64 hdl)
+{
+	struct ice_vsig_prof *ent, *tmp;
+
+	list_for_each_entry_safe(ent, tmp, lst, list)
+		if (ent->profile_cookie == hdl) {
+			list_del(&ent->list);
+			devm_kfree(ice_hw_to_dev(hw), ent);
+			return 0;
+		}
+
+	return ICE_ERR_DOES_NOT_EXIST;
+}
+
+/**
+ * ice_rem_prof_id_flow - remove flow
+ * @hw: pointer to the HW struct
+ * @blk: hardware block
+ * @vsi: the VSI from which to remove the profile specified by ID
+ * @hdl: profile tracking handle
+ *
+ * Calling this function will update the hardware tables to remove the
+ * profile indicated by the ID parameter for the VSIs specified in the VSI
+ * array. Once successfully called, the flow will be disabled.
+ */
+enum ice_status
+ice_rem_prof_id_flow(struct ice_hw *hw, enum ice_block blk, u16 vsi, u64 hdl)
+{
+	struct ice_vsig_prof *tmp1, *del1;
+	struct ice_chs_chg *tmp, *del;
+	struct list_head chg, copy;
+	enum ice_status status;
+	u16 vsig;
+
+	INIT_LIST_HEAD(&copy);
+	INIT_LIST_HEAD(&chg);
+
+	/* determine if VSI is already part of a VSIG */
+	status = ice_vsig_find_vsi(hw, blk, vsi, &vsig);
+	if (!status && vsig) {
+		bool last_profile;
+		bool only_vsi;
+		u16 ref;
+
+		/* found in VSIG */
+		last_profile = ice_vsig_prof_id_count(hw, blk, vsig) == 1;
+		status = ice_vsig_get_ref(hw, blk, vsig, &ref);
+		if (status)
+			goto err_ice_rem_prof_id_flow;
+		only_vsi = (ref == 1);
+
+		if (only_vsi) {
+			/* If the original VSIG only contains one reference,
+			 * which will be the requesting VSI, then the VSI is not
+			 * sharing entries and we can simply remove the specific
+			 * characteristics from the VSIG.
+			 */
+
+			if (last_profile) {
+				/* If there are no profiles left for this VSIG,
+				 * then simply remove the the VSIG.
+				 */
+				status = ice_rem_vsig(hw, blk, vsig, &chg);
+				if (status)
+					goto err_ice_rem_prof_id_flow;
+			} else {
+				status = ice_rem_prof_id_vsig(hw, blk, vsig,
+							      hdl, &chg);
+				if (status)
+					goto err_ice_rem_prof_id_flow;
+
+				/* Adjust priorities */
+				status = ice_adj_prof_priorities(hw, blk, vsig,
+								 &chg);
+				if (status)
+					goto err_ice_rem_prof_id_flow;
+			}
+
+		} else {
+			/* Make a copy of the VSIG's list of Profiles */
+			status = ice_get_profs_vsig(hw, blk, vsig, &copy);
+			if (status)
+				goto err_ice_rem_prof_id_flow;
+
+			/* Remove specified profile entry from the list */
+			status = ice_rem_prof_from_list(hw, &copy, hdl);
+			if (status)
+				goto err_ice_rem_prof_id_flow;
+
+			if (list_empty(&copy)) {
+				status = ice_move_vsi(hw, blk, vsi,
+						      ICE_DEFAULT_VSIG, &chg);
+				if (status)
+					goto err_ice_rem_prof_id_flow;
+
+			} else if (!ice_find_dup_props_vsig(hw, blk, &copy,
+							    &vsig)) {
+				/* found an exact match */
+				/* add or move VSI to the VSIG that matches */
+				/* Search for a VSIG with a matching profile
+				 * list
+				 */
+
+				/* Found match, move VSI to the matching VSIG */
+				status = ice_move_vsi(hw, blk, vsi, vsig, &chg);
+				if (status)
+					goto err_ice_rem_prof_id_flow;
+			} else {
+				/* since no existing VSIG supports this
+				 * characteristic pattern, we need to create a
+				 * new VSIG and TCAM entries
+				 */
+				status = ice_create_vsig_from_lst(hw, blk, vsi,
+								  &copy, &chg);
+				if (status)
+					goto err_ice_rem_prof_id_flow;
+
+				/* Adjust priorities */
+				status = ice_adj_prof_priorities(hw, blk, vsig,
+								 &chg);
+				if (status)
+					goto err_ice_rem_prof_id_flow;
+			}
+		}
+	} else {
+		status = ICE_ERR_DOES_NOT_EXIST;
+	}
+
+	/* update hardware tables */
+	if (!status)
+		status = ice_upd_prof_hw(hw, blk, &chg);
+
+err_ice_rem_prof_id_flow:
+	list_for_each_entry_safe(del, tmp, &chg, list_entry) {
+		list_del(&del->list_entry);
+		devm_kfree(ice_hw_to_dev(hw), del);
+	}
+
+	list_for_each_entry_safe(del1, tmp1, &copy, list) {
+		list_del(&del1->list);
+		devm_kfree(ice_hw_to_dev(hw), del1);
+	}
+
+	return status;
+}
diff --git a/drivers/net/ethernet/intel/ice/ice_flex_pipe.h b/drivers/net/ethernet/intel/ice/ice_flex_pipe.h
index 33e4510da24c..c7b5e1a6ea2b 100644
--- a/drivers/net/ethernet/intel/ice/ice_flex_pipe.h
+++ b/drivers/net/ethernet/intel/ice/ice_flex_pipe.h
@@ -23,6 +23,8 @@ ice_add_prof(struct ice_hw *hw, enum ice_block blk, u64 id, u8 ptypes[],
 	     struct ice_fv_word *es);
 enum ice_status
 ice_add_prof_id_flow(struct ice_hw *hw, enum ice_block blk, u16 vsi, u64 hdl);
+enum ice_status
+ice_rem_prof_id_flow(struct ice_hw *hw, enum ice_block blk, u16 vsi, u64 hdl);
 enum ice_status ice_init_pkg(struct ice_hw *hw, u8 *buff, u32 len);
 enum ice_status
 ice_copy_and_init_pkg(struct ice_hw *hw, const u8 *buf, u32 len);
@@ -31,4 +33,6 @@ void ice_free_seg(struct ice_hw *hw);
 void ice_fill_blk_tbls(struct ice_hw *hw);
 void ice_clear_hw_tbls(struct ice_hw *hw);
 void ice_free_hw_tbls(struct ice_hw *hw);
+enum ice_status
+ice_rem_prof(struct ice_hw *hw, enum ice_block blk, u64 id);
 #endif /* _ICE_FLEX_PIPE_H_ */
diff --git a/drivers/net/ethernet/intel/ice/ice_flow.c b/drivers/net/ethernet/intel/ice/ice_flow.c
index 1bbf4b6ed7d2..ffce2284d8fa 100644
--- a/drivers/net/ethernet/intel/ice/ice_flow.c
+++ b/drivers/net/ethernet/intel/ice/ice_flow.c
@@ -377,6 +377,77 @@ ice_flow_proc_segs(struct ice_hw *hw, struct ice_flow_prof_params *params)
 	return status;
 }
 
+#define ICE_FLOW_FIND_PROF_CHK_FLDS	0x00000001
+#define ICE_FLOW_FIND_PROF_CHK_VSI	0x00000002
+#define ICE_FLOW_FIND_PROF_NOT_CHK_DIR	0x00000004
+
+/**
+ * ice_flow_find_prof_conds - Find a profile matching headers and conditions
+ * @hw: pointer to the HW struct
+ * @blk: classification stage
+ * @dir: flow direction
+ * @segs: array of one or more packet segments that describe the flow
+ * @segs_cnt: number of packet segments provided
+ * @vsi_handle: software VSI handle to check VSI (ICE_FLOW_FIND_PROF_CHK_VSI)
+ * @conds: additional conditions to be checked (ICE_FLOW_FIND_PROF_CHK_*)
+ */
+static struct ice_flow_prof *
+ice_flow_find_prof_conds(struct ice_hw *hw, enum ice_block blk,
+			 enum ice_flow_dir dir, struct ice_flow_seg_info *segs,
+			 u8 segs_cnt, u16 vsi_handle, u32 conds)
+{
+	struct ice_flow_prof *p, *prof = NULL;
+
+	mutex_lock(&hw->fl_profs_locks[blk]);
+	list_for_each_entry(p, &hw->fl_profs[blk], l_entry)
+		if ((p->dir == dir || conds & ICE_FLOW_FIND_PROF_NOT_CHK_DIR) &&
+		    segs_cnt && segs_cnt == p->segs_cnt) {
+			u8 i;
+
+			/* Check for profile-VSI association if specified */
+			if ((conds & ICE_FLOW_FIND_PROF_CHK_VSI) &&
+			    ice_is_vsi_valid(hw, vsi_handle) &&
+			    !test_bit(vsi_handle, p->vsis))
+				continue;
+
+			/* Protocol headers must be checked. Matched fields are
+			 * checked if specified.
+			 */
+			for (i = 0; i < segs_cnt; i++)
+				if (segs[i].hdrs != p->segs[i].hdrs ||
+				    ((conds & ICE_FLOW_FIND_PROF_CHK_FLDS) &&
+				     segs[i].match != p->segs[i].match))
+					break;
+
+			/* A match is found if all segments are matched */
+			if (i == segs_cnt) {
+				prof = p;
+				break;
+			}
+		}
+	mutex_unlock(&hw->fl_profs_locks[blk]);
+
+	return prof;
+}
+
+/**
+ * ice_flow_find_prof_id - Look up a profile with given profile ID
+ * @hw: pointer to the HW struct
+ * @blk: classification stage
+ * @prof_id: unique ID to identify this flow profile
+ */
+static struct ice_flow_prof *
+ice_flow_find_prof_id(struct ice_hw *hw, enum ice_block blk, u64 prof_id)
+{
+	struct ice_flow_prof *p;
+
+	list_for_each_entry(p, &hw->fl_profs[blk], l_entry)
+		if (p->id == prof_id)
+			return p;
+
+	return NULL;
+}
+
 /**
  * ice_flow_add_prof_sync - Add a flow profile for packet segments and fields
  * @hw: pointer to the HW struct
@@ -450,6 +521,31 @@ out:
 	return status;
 }
 
+/**
+ * ice_flow_rem_prof_sync - remove a flow profile
+ * @hw: pointer to the hardware structure
+ * @blk: classification stage
+ * @prof: pointer to flow profile to remove
+ *
+ * Assumption: the caller has acquired the lock to the profile list
+ */
+static enum ice_status
+ice_flow_rem_prof_sync(struct ice_hw *hw, enum ice_block blk,
+		       struct ice_flow_prof *prof)
+{
+	enum ice_status status;
+
+	/* Remove all hardware profiles associated with this flow profile */
+	status = ice_rem_prof(hw, blk, prof->id);
+	if (!status) {
+		list_del(&prof->l_entry);
+		mutex_destroy(&prof->entries_lock);
+		devm_kfree(ice_hw_to_dev(hw), prof);
+	}
+
+	return status;
+}
+
 /**
  * ice_flow_assoc_prof - associate a VSI with a flow profile
  * @hw: pointer to the hardware structure
@@ -482,6 +578,38 @@ ice_flow_assoc_prof(struct ice_hw *hw, enum ice_block blk,
 	return status;
 }
 
+/**
+ * ice_flow_disassoc_prof - disassociate a VSI from a flow profile
+ * @hw: pointer to the hardware structure
+ * @blk: classification stage
+ * @prof: pointer to flow profile
+ * @vsi_handle: software VSI handle
+ *
+ * Assumption: the caller has acquired the lock to the profile list
+ * and the software VSI handle has been validated
+ */
+static enum ice_status
+ice_flow_disassoc_prof(struct ice_hw *hw, enum ice_block blk,
+		       struct ice_flow_prof *prof, u16 vsi_handle)
+{
+	enum ice_status status = 0;
+
+	if (test_bit(vsi_handle, prof->vsis)) {
+		status = ice_rem_prof_id_flow(hw, blk,
+					      ice_get_hw_vsi_num(hw,
+								 vsi_handle),
+					      prof->id);
+		if (!status)
+			clear_bit(vsi_handle, prof->vsis);
+		else
+			ice_debug(hw, ICE_DBG_FLOW,
+				  "HW profile remove failed, %d\n",
+				  status);
+	}
+
+	return status;
+}
+
 /**
  * ice_flow_add_prof - Add a flow profile for packet segments and matched fields
  * @hw: pointer to the HW struct
@@ -524,6 +652,35 @@ ice_flow_add_prof(struct ice_hw *hw, enum ice_block blk, enum ice_flow_dir dir,
 	return status;
 }
 
+/**
+ * ice_flow_rem_prof - Remove a flow profile and all entries associated with it
+ * @hw: pointer to the HW struct
+ * @blk: the block for which the flow profile is to be removed
+ * @prof_id: unique ID of the flow profile to be removed
+ */
+static enum ice_status
+ice_flow_rem_prof(struct ice_hw *hw, enum ice_block blk, u64 prof_id)
+{
+	struct ice_flow_prof *prof;
+	enum ice_status status;
+
+	mutex_lock(&hw->fl_profs_locks[blk]);
+
+	prof = ice_flow_find_prof_id(hw, blk, prof_id);
+	if (!prof) {
+		status = ICE_ERR_DOES_NOT_EXIST;
+		goto out;
+	}
+
+	/* prof becomes invalid after the call */
+	status = ice_flow_rem_prof_sync(hw, blk, prof);
+
+out:
+	mutex_unlock(&hw->fl_profs_locks[blk]);
+
+	return status;
+}
+
 /**
  * ice_flow_set_fld_ext - specifies locations of field from entry's input buffer
  * @seg: packet segment the field being set belongs to
@@ -645,6 +802,132 @@ ice_flow_set_rss_seg_info(struct ice_flow_seg_info *segs, u64 hash_fields,
 	return 0;
 }
 
+/**
+ * ice_rem_vsi_rss_list - remove VSI from RSS list
+ * @hw: pointer to the hardware structure
+ * @vsi_handle: software VSI handle
+ *
+ * Remove the VSI from all RSS configurations in the list.
+ */
+void ice_rem_vsi_rss_list(struct ice_hw *hw, u16 vsi_handle)
+{
+	struct ice_rss_cfg *r, *tmp;
+
+	if (list_empty(&hw->rss_list_head))
+		return;
+
+	mutex_lock(&hw->rss_locks);
+	list_for_each_entry_safe(r, tmp, &hw->rss_list_head, l_entry)
+		if (test_and_clear_bit(vsi_handle, r->vsis))
+			if (bitmap_empty(r->vsis, ICE_MAX_VSI)) {
+				list_del(&r->l_entry);
+				devm_kfree(ice_hw_to_dev(hw), r);
+			}
+	mutex_unlock(&hw->rss_locks);
+}
+
+/**
+ * ice_rem_vsi_rss_cfg - remove RSS configurations associated with VSI
+ * @hw: pointer to the hardware structure
+ * @vsi_handle: software VSI handle
+ *
+ * This function will iterate through all flow profiles and disassociate
+ * the VSI from that profile. If the flow profile has no VSIs it will
+ * be removed.
+ */
+enum ice_status ice_rem_vsi_rss_cfg(struct ice_hw *hw, u16 vsi_handle)
+{
+	const enum ice_block blk = ICE_BLK_RSS;
+	struct ice_flow_prof *p, *t;
+	enum ice_status status = 0;
+
+	if (!ice_is_vsi_valid(hw, vsi_handle))
+		return ICE_ERR_PARAM;
+
+	if (list_empty(&hw->fl_profs[blk]))
+		return 0;
+
+	mutex_lock(&hw->fl_profs_locks[blk]);
+	list_for_each_entry_safe(p, t, &hw->fl_profs[blk], l_entry)
+		if (test_bit(vsi_handle, p->vsis)) {
+			status = ice_flow_disassoc_prof(hw, blk, p, vsi_handle);
+			if (status)
+				break;
+
+			if (bitmap_empty(p->vsis, ICE_MAX_VSI)) {
+				status = ice_flow_rem_prof_sync(hw, blk, p);
+				if (status)
+					break;
+			}
+		}
+	mutex_unlock(&hw->fl_profs_locks[blk]);
+
+	return status;
+}
+
+/**
+ * ice_rem_rss_list - remove RSS configuration from list
+ * @hw: pointer to the hardware structure
+ * @vsi_handle: software VSI handle
+ * @prof: pointer to flow profile
+ *
+ * Assumption: lock has already been acquired for RSS list
+ */
+static void
+ice_rem_rss_list(struct ice_hw *hw, u16 vsi_handle, struct ice_flow_prof *prof)
+{
+	struct ice_rss_cfg *r, *tmp;
+
+	/* Search for RSS hash fields associated to the VSI that match the
+	 * hash configurations associated to the flow profile. If found
+	 * remove from the RSS entry list of the VSI context and delete entry.
+	 */
+	list_for_each_entry_safe(r, tmp, &hw->rss_list_head, l_entry)
+		if (r->hashed_flds == prof->segs[prof->segs_cnt - 1].match &&
+		    r->packet_hdr == prof->segs[prof->segs_cnt - 1].hdrs) {
+			clear_bit(vsi_handle, r->vsis);
+			if (bitmap_empty(r->vsis, ICE_MAX_VSI)) {
+				list_del(&r->l_entry);
+				devm_kfree(ice_hw_to_dev(hw), r);
+			}
+			return;
+		}
+}
+
+/**
+ * ice_add_rss_list - add RSS configuration to list
+ * @hw: pointer to the hardware structure
+ * @vsi_handle: software VSI handle
+ * @prof: pointer to flow profile
+ *
+ * Assumption: lock has already been acquired for RSS list
+ */
+static enum ice_status
+ice_add_rss_list(struct ice_hw *hw, u16 vsi_handle, struct ice_flow_prof *prof)
+{
+	struct ice_rss_cfg *r, *rss_cfg;
+
+	list_for_each_entry(r, &hw->rss_list_head, l_entry)
+		if (r->hashed_flds == prof->segs[prof->segs_cnt - 1].match &&
+		    r->packet_hdr == prof->segs[prof->segs_cnt - 1].hdrs) {
+			set_bit(vsi_handle, r->vsis);
+			return 0;
+		}
+
+	rss_cfg = devm_kzalloc(ice_hw_to_dev(hw), sizeof(*rss_cfg),
+			       GFP_KERNEL);
+	if (!rss_cfg)
+		return ICE_ERR_NO_MEMORY;
+
+	rss_cfg->hashed_flds = prof->segs[prof->segs_cnt - 1].match;
+	rss_cfg->packet_hdr = prof->segs[prof->segs_cnt - 1].hdrs;
+	set_bit(vsi_handle, rss_cfg->vsis);
+
+	list_add_tail(&rss_cfg->l_entry, &hw->rss_list_head);
+
+	return 0;
+}
+
 #define ICE_FLOW_PROF_HASH_S	0
 #define ICE_FLOW_PROF_HASH_M	(0xFFFFFFFFULL << ICE_FLOW_PROF_HASH_S)
 #define ICE_FLOW_PROF_HDR_S	32
@@ -696,6 +979,52 @@ ice_add_rss_cfg_sync(struct ice_hw *hw, u16 vsi_handle, u64 hashed_flds,
 	if (status)
 		goto exit;
 
+	/* Search for a flow profile that has matching headers, hash fields
+	 * and has the input VSI associated to it. If found, no further
+	 * operations required and exit.
+	 */
+	prof = ice_flow_find_prof_conds(hw, blk, ICE_FLOW_RX, segs, segs_cnt,
+					vsi_handle,
+					ICE_FLOW_FIND_PROF_CHK_FLDS |
+					ICE_FLOW_FIND_PROF_CHK_VSI);
+	if (prof)
+		goto exit;
+
+	/* Check if a flow profile exists with the same protocol headers and
+	 * associated with the input VSI. If so disassociate the VSI from
+	 * this profile. The VSI will be added to a new profile created with
+	 * the protocol header and new hash field configuration.
+	 */
+	prof = ice_flow_find_prof_conds(hw, blk, ICE_FLOW_RX, segs, segs_cnt,
+					vsi_handle, ICE_FLOW_FIND_PROF_CHK_VSI);
+	if (prof) {
+		status = ice_flow_disassoc_prof(hw, blk, prof, vsi_handle);
+		if (!status)
+			ice_rem_rss_list(hw, vsi_handle, prof);
+		else
+			goto exit;
+
+		/* Remove profile if it has no VSIs associated */
+		if (bitmap_empty(prof->vsis, ICE_MAX_VSI)) {
+			status = ice_flow_rem_prof(hw, blk, prof->id);
+			if (status)
+				goto exit;
+		}
+	}
+
+	/* Search for a profile that has same match fields only. If this
+	 * exists then associate the VSI to this profile.
+	 */
+	prof = ice_flow_find_prof_conds(hw, blk, ICE_FLOW_RX, segs, segs_cnt,
+					vsi_handle,
+					ICE_FLOW_FIND_PROF_CHK_FLDS);
+	if (prof) {
+		status = ice_flow_assoc_prof(hw, blk, prof, vsi_handle);
+		if (!status)
+			status = ice_add_rss_list(hw, vsi_handle, prof);
+		goto exit;
+	}
+
 	/* Create a new flow profile with generated profile and packet
 	 * segment information.
 	 */
@@ -708,6 +1037,15 @@ ice_add_rss_cfg_sync(struct ice_hw *hw, u16 vsi_handle, u64 hashed_flds,
 		goto exit;
 
 	status = ice_flow_assoc_prof(hw, blk, prof, vsi_handle);
+	/* If association to a new flow profile failed then this profile can
+	 * be removed.
+	 */
+	if (status) {
+		ice_flow_rem_prof(hw, blk, prof->id);
+		goto exit;
+	}
+
+	status = ice_add_rss_list(hw, vsi_handle, prof);
 
 exit:
 	kfree(segs);
diff --git a/drivers/net/ethernet/intel/ice/ice_flow.h b/drivers/net/ethernet/intel/ice/ice_flow.h
index 05b0dab4793c..38669b077209 100644
--- a/drivers/net/ethernet/intel/ice/ice_flow.h
+++ b/drivers/net/ethernet/intel/ice/ice_flow.h
@@ -4,6 +4,7 @@
 #ifndef _ICE_FLOW_H_
 #define _ICE_FLOW_H_
 
+#define ICE_FLOW_ENTRY_HANDLE_INVAL	0
 #define ICE_FLOW_FLD_OFF_INVAL		0xffff
 
 /* Generate flow hash field from flow field type(s) */
@@ -60,6 +61,12 @@ enum ice_flow_dir {
 	ICE_FLOW_RX		= 0x02,
 };
 
+enum ice_flow_priority {
+	ICE_FLOW_PRIO_LOW,
+	ICE_FLOW_PRIO_NORMAL,
+	ICE_FLOW_PRIO_HIGH
+};
+
 #define ICE_FLOW_SEG_MAX		2
 #define ICE_FLOW_FV_EXTRACT_SZ		2
 
@@ -130,7 +137,10 @@ struct ice_rss_cfg {
 	u32 packet_hdr;
 };
 
+enum ice_status ice_flow_rem_entry(struct ice_hw *hw, u64 entry_h);
+void ice_rem_vsi_rss_list(struct ice_hw *hw, u16 vsi_handle);
 enum ice_status ice_replay_rss_cfg(struct ice_hw *hw, u16 vsi_handle);
+enum ice_status ice_rem_vsi_rss_cfg(struct ice_hw *hw, u16 vsi_handle);
 enum ice_status
 ice_add_rss_cfg(struct ice_hw *hw, u16 vsi_handle, u64 hashed_flds,
 		u32 addl_hdrs);
diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c
index dd230c6664ee..caa3edd5d3d6 100644
--- a/drivers/net/ethernet/intel/ice/ice_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_lib.c
@@ -494,7 +494,28 @@ bool ice_is_safe_mode(struct ice_pf *pf)
 }
 
 /**
- * ice_rss_clean - Delete RSS related VSI structures that hold user inputs
+ * ice_vsi_clean_rss_flow_fld - Delete RSS configuration
+ * @vsi: the VSI being cleaned up
+ *
+ * This function deletes RSS input set for all flows that were configured
+ * for this VSI
+ */
+static void ice_vsi_clean_rss_flow_fld(struct ice_vsi *vsi)
+{
+	struct ice_pf *pf = vsi->back;
+	enum ice_status status;
+
+	if (ice_is_safe_mode(pf))
+		return;
+
+	status = ice_rem_vsi_rss_cfg(&pf->hw, vsi->idx);
+	if (status)
+		dev_dbg(ice_pf_to_dev(pf), "ice_rem_vsi_rss_cfg failed for vsi = %d, error = %d\n",
+			vsi->vsi_num, status);
+}
+
+/**
+ * ice_rss_clean - Delete RSS related VSI structures and configuration
  * @vsi: the VSI being removed
  */
 static void ice_rss_clean(struct ice_vsi *vsi)
@@ -508,6 +529,11 @@ static void ice_rss_clean(struct ice_vsi *vsi)
 		devm_kfree(dev, vsi->rss_hkey_user);
 	if (vsi->rss_lut_user)
 		devm_kfree(dev, vsi->rss_lut_user);
+
+	ice_vsi_clean_rss_flow_fld(vsi);
+	/* remove RSS replay list */
+	if (!ice_is_safe_mode(pf))
+		ice_rem_vsi_rss_list(&pf->hw, vsi->idx);
 }
 
 /**
-- 
cgit v1.2.3-59-g8ed1b


From 1c01c8c6c969a9da1b0783c2b4138debc4d868c2 Mon Sep 17 00:00:00 2001
From: Md Fahad Iqbal Polash <md.fahad.iqbal.polash@intel.com>
Date: Fri, 17 Jan 2020 07:39:17 -0800
Subject: ice: Initilialize VF RSS tables

Set configuration for hardware RSS tables for VFs.

Signed-off-by: Md Fahad Iqbal Polash <md.fahad.iqbal.polash@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_flow.c          | 135 +++++++++++++++++++++
 drivers/net/ethernet/intel/ice/ice_flow.h          |  59 +++++++++
 drivers/net/ethernet/intel/ice/ice_lib.c           |  31 ++++-
 drivers/net/ethernet/intel/ice/ice_protocol_type.h |   1 +
 4 files changed, 225 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_flow.c b/drivers/net/ethernet/intel/ice/ice_flow.c
index ffce2284d8fa..4ecac0ae565a 100644
--- a/drivers/net/ethernet/intel/ice/ice_flow.c
+++ b/drivers/net/ethernet/intel/ice/ice_flow.c
@@ -38,6 +38,11 @@ struct ice_flow_field_info ice_flds_info[ICE_FLOW_FIELD_IDX_MAX] = {
 	ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_UDP, 0, sizeof(__be16)),
 	/* ICE_FLOW_FIELD_IDX_UDP_DST_PORT */
 	ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_UDP, 2, sizeof(__be16)),
+	/* ICE_FLOW_FIELD_IDX_SCTP_SRC_PORT */
+	ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_SCTP, 0, sizeof(__be16)),
+	/* ICE_FLOW_FIELD_IDX_SCTP_DST_PORT */
+	ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_SCTP, 2, sizeof(__be16)),
+
 };
 
 /* Bitmaps indicating relevant packet types for a particular protocol header
@@ -266,6 +271,10 @@ ice_flow_xtract_fld(struct ice_hw *hw, struct ice_flow_prof_params *params,
 	case ICE_FLOW_FIELD_IDX_UDP_DST_PORT:
 		prot_id = ICE_PROT_UDP_IL_OR_S;
 		break;
+	case ICE_FLOW_FIELD_IDX_SCTP_SRC_PORT:
+	case ICE_FLOW_FIELD_IDX_SCTP_DST_PORT:
+		prot_id = ICE_PROT_SCTP_IL;
+		break;
 	default:
 		return ICE_ERR_NOT_IMPL;
 	}
@@ -1081,6 +1090,132 @@ ice_add_rss_cfg(struct ice_hw *hw, u16 vsi_handle, u64 hashed_flds,
 	return status;
 }
 
+/* Mapping of AVF hash bit fields to an L3-L4 hash combination.
+ * As the ice_flow_avf_hdr_field represent individual bit shifts in a hash,
+ * convert its values to their appropriate flow L3, L4 values.
+ */
+#define ICE_FLOW_AVF_RSS_IPV4_MASKS \
+	(BIT_ULL(ICE_AVF_FLOW_FIELD_IPV4_OTHER) | \
+	 BIT_ULL(ICE_AVF_FLOW_FIELD_FRAG_IPV4))
+#define ICE_FLOW_AVF_RSS_TCP_IPV4_MASKS \
+	(BIT_ULL(ICE_AVF_FLOW_FIELD_IPV4_TCP_SYN_NO_ACK) | \
+	 BIT_ULL(ICE_AVF_FLOW_FIELD_IPV4_TCP))
+#define ICE_FLOW_AVF_RSS_UDP_IPV4_MASKS \
+	(BIT_ULL(ICE_AVF_FLOW_FIELD_UNICAST_IPV4_UDP) | \
+	 BIT_ULL(ICE_AVF_FLOW_FIELD_MULTICAST_IPV4_UDP) | \
+	 BIT_ULL(ICE_AVF_FLOW_FIELD_IPV4_UDP))
+#define ICE_FLOW_AVF_RSS_ALL_IPV4_MASKS \
+	(ICE_FLOW_AVF_RSS_TCP_IPV4_MASKS | ICE_FLOW_AVF_RSS_UDP_IPV4_MASKS | \
+	 ICE_FLOW_AVF_RSS_IPV4_MASKS | BIT_ULL(ICE_AVF_FLOW_FIELD_IPV4_SCTP))
+
+#define ICE_FLOW_AVF_RSS_IPV6_MASKS \
+	(BIT_ULL(ICE_AVF_FLOW_FIELD_IPV6_OTHER) | \
+	 BIT_ULL(ICE_AVF_FLOW_FIELD_FRAG_IPV6))
+#define ICE_FLOW_AVF_RSS_UDP_IPV6_MASKS \
+	(BIT_ULL(ICE_AVF_FLOW_FIELD_UNICAST_IPV6_UDP) | \
+	 BIT_ULL(ICE_AVF_FLOW_FIELD_MULTICAST_IPV6_UDP) | \
+	 BIT_ULL(ICE_AVF_FLOW_FIELD_IPV6_UDP))
+#define ICE_FLOW_AVF_RSS_TCP_IPV6_MASKS \
+	(BIT_ULL(ICE_AVF_FLOW_FIELD_IPV6_TCP_SYN_NO_ACK) | \
+	 BIT_ULL(ICE_AVF_FLOW_FIELD_IPV6_TCP))
+#define ICE_FLOW_AVF_RSS_ALL_IPV6_MASKS \
+	(ICE_FLOW_AVF_RSS_TCP_IPV6_MASKS | ICE_FLOW_AVF_RSS_UDP_IPV6_MASKS | \
+	 ICE_FLOW_AVF_RSS_IPV6_MASKS | BIT_ULL(ICE_AVF_FLOW_FIELD_IPV6_SCTP))
+
+/**
+ * ice_add_avf_rss_cfg - add an RSS configuration for AVF driver
+ * @hw: pointer to the hardware structure
+ * @vsi_handle: software VSI handle
+ * @avf_hash: hash bit fields (ICE_AVF_FLOW_FIELD_*) to configure
+ *
+ * This function will take the hash bitmap provided by the AVF driver via a
+ * message, convert it to ICE-compatible values, and configure RSS flow
+ * profiles.
+ */
+enum ice_status
+ice_add_avf_rss_cfg(struct ice_hw *hw, u16 vsi_handle, u64 avf_hash)
+{
+	enum ice_status status = 0;
+	u64 hash_flds;
+
+	if (avf_hash == ICE_AVF_FLOW_FIELD_INVALID ||
+	    !ice_is_vsi_valid(hw, vsi_handle))
+		return ICE_ERR_PARAM;
+
+	/* Make sure no unsupported bits are specified */
+	if (avf_hash & ~(ICE_FLOW_AVF_RSS_ALL_IPV4_MASKS |
+			 ICE_FLOW_AVF_RSS_ALL_IPV6_MASKS))
+		return ICE_ERR_CFG;
+
+	hash_flds = avf_hash;
+
+	/* Always create an L3 RSS configuration for any L4 RSS configuration */
+	if (hash_flds & ICE_FLOW_AVF_RSS_ALL_IPV4_MASKS)
+		hash_flds |= ICE_FLOW_AVF_RSS_IPV4_MASKS;
+
+	if (hash_flds & ICE_FLOW_AVF_RSS_ALL_IPV6_MASKS)
+		hash_flds |= ICE_FLOW_AVF_RSS_IPV6_MASKS;
+
+	/* Create the corresponding RSS configuration for each valid hash bit */
+	while (hash_flds) {
+		u64 rss_hash = ICE_HASH_INVALID;
+
+		if (hash_flds & ICE_FLOW_AVF_RSS_ALL_IPV4_MASKS) {
+			if (hash_flds & ICE_FLOW_AVF_RSS_IPV4_MASKS) {
+				rss_hash = ICE_FLOW_HASH_IPV4;
+				hash_flds &= ~ICE_FLOW_AVF_RSS_IPV4_MASKS;
+			} else if (hash_flds &
+				   ICE_FLOW_AVF_RSS_TCP_IPV4_MASKS) {
+				rss_hash = ICE_FLOW_HASH_IPV4 |
+					ICE_FLOW_HASH_TCP_PORT;
+				hash_flds &= ~ICE_FLOW_AVF_RSS_TCP_IPV4_MASKS;
+			} else if (hash_flds &
+				   ICE_FLOW_AVF_RSS_UDP_IPV4_MASKS) {
+				rss_hash = ICE_FLOW_HASH_IPV4 |
+					ICE_FLOW_HASH_UDP_PORT;
+				hash_flds &= ~ICE_FLOW_AVF_RSS_UDP_IPV4_MASKS;
+			} else if (hash_flds &
+				   BIT_ULL(ICE_AVF_FLOW_FIELD_IPV4_SCTP)) {
+				rss_hash = ICE_FLOW_HASH_IPV4 |
+					ICE_FLOW_HASH_SCTP_PORT;
+				hash_flds &=
+					~BIT_ULL(ICE_AVF_FLOW_FIELD_IPV4_SCTP);
+			}
+		} else if (hash_flds & ICE_FLOW_AVF_RSS_ALL_IPV6_MASKS) {
+			if (hash_flds & ICE_FLOW_AVF_RSS_IPV6_MASKS) {
+				rss_hash = ICE_FLOW_HASH_IPV6;
+				hash_flds &= ~ICE_FLOW_AVF_RSS_IPV6_MASKS;
+			} else if (hash_flds &
+				   ICE_FLOW_AVF_RSS_TCP_IPV6_MASKS) {
+				rss_hash = ICE_FLOW_HASH_IPV6 |
+					ICE_FLOW_HASH_TCP_PORT;
+				hash_flds &= ~ICE_FLOW_AVF_RSS_TCP_IPV6_MASKS;
+			} else if (hash_flds &
+				   ICE_FLOW_AVF_RSS_UDP_IPV6_MASKS) {
+				rss_hash = ICE_FLOW_HASH_IPV6 |
+					ICE_FLOW_HASH_UDP_PORT;
+				hash_flds &= ~ICE_FLOW_AVF_RSS_UDP_IPV6_MASKS;
+			} else if (hash_flds &
+				   BIT_ULL(ICE_AVF_FLOW_FIELD_IPV6_SCTP)) {
+				rss_hash = ICE_FLOW_HASH_IPV6 |
+					ICE_FLOW_HASH_SCTP_PORT;
+				hash_flds &=
+					~BIT_ULL(ICE_AVF_FLOW_FIELD_IPV6_SCTP);
+			}
+		}
+
+		if (rss_hash == ICE_HASH_INVALID)
+			return ICE_ERR_OUT_OF_RANGE;
+
+		status = ice_add_rss_cfg(hw, vsi_handle, rss_hash,
+					 ICE_FLOW_SEG_HDR_NONE);
+		if (status)
+			break;
+	}
+
+	return status;
+}
+
 /**
  * ice_replay_rss_cfg - replay RSS configurations associated with VSI
  * @hw: pointer to the hardware structure
diff --git a/drivers/net/ethernet/intel/ice/ice_flow.h b/drivers/net/ethernet/intel/ice/ice_flow.h
index 38669b077209..475a025d3a2d 100644
--- a/drivers/net/ethernet/intel/ice/ice_flow.h
+++ b/drivers/net/ethernet/intel/ice/ice_flow.h
@@ -20,6 +20,9 @@
 #define ICE_FLOW_HASH_UDP_PORT	\
 	(BIT_ULL(ICE_FLOW_FIELD_IDX_UDP_SRC_PORT) | \
 	 BIT_ULL(ICE_FLOW_FIELD_IDX_UDP_DST_PORT))
+#define ICE_FLOW_HASH_SCTP_PORT	\
+	(BIT_ULL(ICE_FLOW_FIELD_IDX_SCTP_SRC_PORT) | \
+	 BIT_ULL(ICE_FLOW_FIELD_IDX_SCTP_DST_PORT))
 
 #define ICE_HASH_INVALID	0
 #define ICE_HASH_TCP_IPV4	(ICE_FLOW_HASH_IPV4 | ICE_FLOW_HASH_TCP_PORT)
@@ -53,10 +56,64 @@ enum ice_flow_field {
 	ICE_FLOW_FIELD_IDX_TCP_DST_PORT,
 	ICE_FLOW_FIELD_IDX_UDP_SRC_PORT,
 	ICE_FLOW_FIELD_IDX_UDP_DST_PORT,
+	ICE_FLOW_FIELD_IDX_SCTP_SRC_PORT,
+	ICE_FLOW_FIELD_IDX_SCTP_DST_PORT,
 	/* The total number of enums must not exceed 64 */
 	ICE_FLOW_FIELD_IDX_MAX
 };
 
+/* Flow headers and fields for AVF support */
+enum ice_flow_avf_hdr_field {
+	/* Values 0 - 28 are reserved for future use */
+	ICE_AVF_FLOW_FIELD_INVALID		= 0,
+	ICE_AVF_FLOW_FIELD_UNICAST_IPV4_UDP	= 29,
+	ICE_AVF_FLOW_FIELD_MULTICAST_IPV4_UDP,
+	ICE_AVF_FLOW_FIELD_IPV4_UDP,
+	ICE_AVF_FLOW_FIELD_IPV4_TCP_SYN_NO_ACK,
+	ICE_AVF_FLOW_FIELD_IPV4_TCP,
+	ICE_AVF_FLOW_FIELD_IPV4_SCTP,
+	ICE_AVF_FLOW_FIELD_IPV4_OTHER,
+	ICE_AVF_FLOW_FIELD_FRAG_IPV4,
+	/* Values 37-38 are reserved */
+	ICE_AVF_FLOW_FIELD_UNICAST_IPV6_UDP	= 39,
+	ICE_AVF_FLOW_FIELD_MULTICAST_IPV6_UDP,
+	ICE_AVF_FLOW_FIELD_IPV6_UDP,
+	ICE_AVF_FLOW_FIELD_IPV6_TCP_SYN_NO_ACK,
+	ICE_AVF_FLOW_FIELD_IPV6_TCP,
+	ICE_AVF_FLOW_FIELD_IPV6_SCTP,
+	ICE_AVF_FLOW_FIELD_IPV6_OTHER,
+	ICE_AVF_FLOW_FIELD_FRAG_IPV6,
+	ICE_AVF_FLOW_FIELD_RSVD47,
+	ICE_AVF_FLOW_FIELD_FCOE_OX,
+	ICE_AVF_FLOW_FIELD_FCOE_RX,
+	ICE_AVF_FLOW_FIELD_FCOE_OTHER,
+	/* Values 51-62 are reserved */
+	ICE_AVF_FLOW_FIELD_L2_PAYLOAD		= 63,
+	ICE_AVF_FLOW_FIELD_MAX
+};
+
+/* Supported RSS offloads  This macro is defined to support
+ * VIRTCHNL_OP_GET_RSS_HENA_CAPS ops. PF driver sends the RSS hardware
+ * capabilities to the caller of this ops.
+ */
+#define ICE_DEFAULT_RSS_HENA ( \
+	BIT_ULL(ICE_AVF_FLOW_FIELD_IPV4_UDP) | \
+	BIT_ULL(ICE_AVF_FLOW_FIELD_IPV4_SCTP) | \
+	BIT_ULL(ICE_AVF_FLOW_FIELD_IPV4_TCP) | \
+	BIT_ULL(ICE_AVF_FLOW_FIELD_IPV4_OTHER) | \
+	BIT_ULL(ICE_AVF_FLOW_FIELD_FRAG_IPV4) | \
+	BIT_ULL(ICE_AVF_FLOW_FIELD_IPV6_UDP) | \
+	BIT_ULL(ICE_AVF_FLOW_FIELD_IPV6_TCP) | \
+	BIT_ULL(ICE_AVF_FLOW_FIELD_IPV6_SCTP) | \
+	BIT_ULL(ICE_AVF_FLOW_FIELD_IPV6_OTHER) | \
+	BIT_ULL(ICE_AVF_FLOW_FIELD_FRAG_IPV6) | \
+	BIT_ULL(ICE_AVF_FLOW_FIELD_IPV4_TCP_SYN_NO_ACK) | \
+	BIT_ULL(ICE_AVF_FLOW_FIELD_UNICAST_IPV4_UDP) | \
+	BIT_ULL(ICE_AVF_FLOW_FIELD_MULTICAST_IPV4_UDP) | \
+	BIT_ULL(ICE_AVF_FLOW_FIELD_IPV6_TCP_SYN_NO_ACK) | \
+	BIT_ULL(ICE_AVF_FLOW_FIELD_UNICAST_IPV6_UDP) | \
+	BIT_ULL(ICE_AVF_FLOW_FIELD_MULTICAST_IPV6_UDP))
+
 enum ice_flow_dir {
 	ICE_FLOW_RX		= 0x02,
 };
@@ -140,6 +197,8 @@ struct ice_rss_cfg {
 enum ice_status ice_flow_rem_entry(struct ice_hw *hw, u64 entry_h);
 void ice_rem_vsi_rss_list(struct ice_hw *hw, u16 vsi_handle);
 enum ice_status ice_replay_rss_cfg(struct ice_hw *hw, u16 vsi_handle);
+enum ice_status
+ice_add_avf_rss_cfg(struct ice_hw *hw, u16 vsi_handle, u64 hashed_flds);
 enum ice_status ice_rem_vsi_rss_cfg(struct ice_hw *hw, u16 vsi_handle);
 enum ice_status
 ice_add_rss_cfg(struct ice_hw *hw, u16 vsi_handle, u64 hashed_flds,
diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c
index caa3edd5d3d6..1874c9f51a32 100644
--- a/drivers/net/ethernet/intel/ice/ice_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_lib.c
@@ -1113,6 +1113,33 @@ ice_vsi_cfg_rss_exit:
 	return err;
 }
 
+/**
+ * ice_vsi_set_vf_rss_flow_fld - Sets VF VSI RSS input set for different flows
+ * @vsi: VSI to be configured
+ *
+ * This function will only be called during the VF VSI setup. Upon successful
+ * completion of package download, this function will configure default RSS
+ * input sets for VF VSI.
+ */
+static void ice_vsi_set_vf_rss_flow_fld(struct ice_vsi *vsi)
+{
+	struct ice_pf *pf = vsi->back;
+	enum ice_status status;
+	struct device *dev;
+
+	dev = ice_pf_to_dev(pf);
+	if (ice_is_safe_mode(pf)) {
+		dev_dbg(dev, "Advanced RSS disabled. Package download failed, vsi num = %d\n",
+			vsi->vsi_num);
+		return;
+	}
+
+	status = ice_add_avf_rss_cfg(&pf->hw, vsi->idx, ICE_DEFAULT_RSS_HENA);
+	if (status)
+		dev_dbg(dev, "ice_add_avf_rss_cfg failed for vsi = %d, error = %d\n",
+			vsi->vsi_num, status);
+}
+
 /**
  * ice_vsi_set_rss_flow_fld - Sets RSS input set for different flows
  * @vsi: VSI to be configured
@@ -2037,8 +2064,10 @@ ice_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi,
 		 * receive traffic on first queue. Hence no need to capture
 		 * return value
 		 */
-		if (test_bit(ICE_FLAG_RSS_ENA, pf->flags))
+		if (test_bit(ICE_FLAG_RSS_ENA, pf->flags)) {
 			ice_vsi_cfg_rss_lut_key(vsi);
+			ice_vsi_set_vf_rss_flow_fld(vsi);
+		}
 		break;
 	case ICE_VSI_LB:
 		ret = ice_vsi_alloc_rings(vsi);
diff --git a/drivers/net/ethernet/intel/ice/ice_protocol_type.h b/drivers/net/ethernet/intel/ice/ice_protocol_type.h
index 97db46ce7a21..71647566964e 100644
--- a/drivers/net/ethernet/intel/ice/ice_protocol_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_protocol_type.h
@@ -18,6 +18,7 @@ enum ice_prot_id {
 	ICE_PROT_IPV6_IL	= 41,
 	ICE_PROT_TCP_IL		= 49,
 	ICE_PROT_UDP_IL_OR_S	= 53,
+	ICE_PROT_SCTP_IL	= 96,
 	ICE_PROT_META_ID	= 255, /* when offset == metadata */
 	ICE_PROT_INVALID	= 255  /* when offset == ICE_FV_OFFSET_INVAL */
 };
-- 
cgit v1.2.3-59-g8ed1b


From 6876fb6404a50d764cc1409870207b62f6e21dff Mon Sep 17 00:00:00 2001
From: Md Fahad Iqbal Polash <md.fahad.iqbal.polash@intel.com>
Date: Fri, 17 Jan 2020 07:39:18 -0800
Subject: ice: Implement ethtool get/set rx-flow-hash

Provide support to change or retrieve RSS hash options for a flow type.
The supported flow-types are: tcp4, tcp6, udp4, udp6, sctp4, sctp6.

Signed-off-by: Md Fahad Iqbal Polash <md.fahad.iqbal.polash@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_ethtool.c | 243 +++++++++++++++++++++++++++
 drivers/net/ethernet/intel/ice/ice_flow.c    |  29 ++++
 drivers/net/ethernet/intel/ice/ice_flow.h    |   1 +
 3 files changed, 273 insertions(+)

diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c
index f395457b728f..90c6a3ca20c9 100644
--- a/drivers/net/ethernet/intel/ice/ice_ethtool.c
+++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c
@@ -4,6 +4,7 @@
 /* ethtool support for ice */
 
 #include "ice.h"
+#include "ice_flow.h"
 #include "ice_lib.h"
 #include "ice_dcb_lib.h"
 
@@ -2533,6 +2534,243 @@ done:
 	return err;
 }
 
+/**
+ * ice_parse_hdrs - parses headers from RSS hash input
+ * @nfc: ethtool rxnfc command
+ *
+ * This function parses the rxnfc command and returns intended
+ * header types for RSS configuration
+ */
+static u32 ice_parse_hdrs(struct ethtool_rxnfc *nfc)
+{
+	u32 hdrs = ICE_FLOW_SEG_HDR_NONE;
+
+	switch (nfc->flow_type) {
+	case TCP_V4_FLOW:
+		hdrs |= ICE_FLOW_SEG_HDR_TCP | ICE_FLOW_SEG_HDR_IPV4;
+		break;
+	case UDP_V4_FLOW:
+		hdrs |= ICE_FLOW_SEG_HDR_UDP | ICE_FLOW_SEG_HDR_IPV4;
+		break;
+	case SCTP_V4_FLOW:
+		hdrs |= ICE_FLOW_SEG_HDR_SCTP | ICE_FLOW_SEG_HDR_IPV4;
+		break;
+	case TCP_V6_FLOW:
+		hdrs |= ICE_FLOW_SEG_HDR_TCP | ICE_FLOW_SEG_HDR_IPV6;
+		break;
+	case UDP_V6_FLOW:
+		hdrs |= ICE_FLOW_SEG_HDR_UDP | ICE_FLOW_SEG_HDR_IPV6;
+		break;
+	case SCTP_V6_FLOW:
+		hdrs |= ICE_FLOW_SEG_HDR_SCTP | ICE_FLOW_SEG_HDR_IPV6;
+		break;
+	default:
+		break;
+	}
+	return hdrs;
+}
+
+#define ICE_FLOW_HASH_FLD_IPV4_SA	BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_SA)
+#define ICE_FLOW_HASH_FLD_IPV6_SA	BIT_ULL(ICE_FLOW_FIELD_IDX_IPV6_SA)
+#define ICE_FLOW_HASH_FLD_IPV4_DA	BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_DA)
+#define ICE_FLOW_HASH_FLD_IPV6_DA	BIT_ULL(ICE_FLOW_FIELD_IDX_IPV6_DA)
+#define ICE_FLOW_HASH_FLD_TCP_SRC_PORT	BIT_ULL(ICE_FLOW_FIELD_IDX_TCP_SRC_PORT)
+#define ICE_FLOW_HASH_FLD_TCP_DST_PORT	BIT_ULL(ICE_FLOW_FIELD_IDX_TCP_DST_PORT)
+#define ICE_FLOW_HASH_FLD_UDP_SRC_PORT	BIT_ULL(ICE_FLOW_FIELD_IDX_UDP_SRC_PORT)
+#define ICE_FLOW_HASH_FLD_UDP_DST_PORT	BIT_ULL(ICE_FLOW_FIELD_IDX_UDP_DST_PORT)
+#define ICE_FLOW_HASH_FLD_SCTP_SRC_PORT	\
+	BIT_ULL(ICE_FLOW_FIELD_IDX_SCTP_SRC_PORT)
+#define ICE_FLOW_HASH_FLD_SCTP_DST_PORT	\
+	BIT_ULL(ICE_FLOW_FIELD_IDX_SCTP_DST_PORT)
+
+/**
+ * ice_parse_hash_flds - parses hash fields from RSS hash input
+ * @nfc: ethtool rxnfc command
+ *
+ * This function parses the rxnfc command and returns intended
+ * hash fields for RSS configuration
+ */
+static u64 ice_parse_hash_flds(struct ethtool_rxnfc *nfc)
+{
+	u64 hfld = ICE_HASH_INVALID;
+
+	if (nfc->data & RXH_IP_SRC || nfc->data & RXH_IP_DST) {
+		switch (nfc->flow_type) {
+		case TCP_V4_FLOW:
+		case UDP_V4_FLOW:
+		case SCTP_V4_FLOW:
+			if (nfc->data & RXH_IP_SRC)
+				hfld |= ICE_FLOW_HASH_FLD_IPV4_SA;
+			if (nfc->data & RXH_IP_DST)
+				hfld |= ICE_FLOW_HASH_FLD_IPV4_DA;
+			break;
+		case TCP_V6_FLOW:
+		case UDP_V6_FLOW:
+		case SCTP_V6_FLOW:
+			if (nfc->data & RXH_IP_SRC)
+				hfld |= ICE_FLOW_HASH_FLD_IPV6_SA;
+			if (nfc->data & RXH_IP_DST)
+				hfld |= ICE_FLOW_HASH_FLD_IPV6_DA;
+			break;
+		default:
+			break;
+		}
+	}
+
+	if (nfc->data & RXH_L4_B_0_1 || nfc->data & RXH_L4_B_2_3) {
+		switch (nfc->flow_type) {
+		case TCP_V4_FLOW:
+		case TCP_V6_FLOW:
+			if (nfc->data & RXH_L4_B_0_1)
+				hfld |= ICE_FLOW_HASH_FLD_TCP_SRC_PORT;
+			if (nfc->data & RXH_L4_B_2_3)
+				hfld |= ICE_FLOW_HASH_FLD_TCP_DST_PORT;
+			break;
+		case UDP_V4_FLOW:
+		case UDP_V6_FLOW:
+			if (nfc->data & RXH_L4_B_0_1)
+				hfld |= ICE_FLOW_HASH_FLD_UDP_SRC_PORT;
+			if (nfc->data & RXH_L4_B_2_3)
+				hfld |= ICE_FLOW_HASH_FLD_UDP_DST_PORT;
+			break;
+		case SCTP_V4_FLOW:
+		case SCTP_V6_FLOW:
+			if (nfc->data & RXH_L4_B_0_1)
+				hfld |= ICE_FLOW_HASH_FLD_SCTP_SRC_PORT;
+			if (nfc->data & RXH_L4_B_2_3)
+				hfld |= ICE_FLOW_HASH_FLD_SCTP_DST_PORT;
+			break;
+		default:
+			break;
+		}
+	}
+
+	return hfld;
+}
+
+/**
+ * ice_set_rss_hash_opt - Enable/Disable flow types for RSS hash
+ * @vsi: the VSI being configured
+ * @nfc: ethtool rxnfc command
+ *
+ * Returns Success if the flow input set is supported.
+ */
+static int
+ice_set_rss_hash_opt(struct ice_vsi *vsi, struct ethtool_rxnfc *nfc)
+{
+	struct ice_pf *pf = vsi->back;
+	enum ice_status status;
+	struct device *dev;
+	u64 hashed_flds;
+	u32 hdrs;
+
+	dev = ice_pf_to_dev(pf);
+	if (ice_is_safe_mode(pf)) {
+		dev_dbg(dev, "Advanced RSS disabled. Package download failed, vsi num = %d\n",
+			vsi->vsi_num);
+		return -EINVAL;
+	}
+
+	hashed_flds = ice_parse_hash_flds(nfc);
+	if (hashed_flds == ICE_HASH_INVALID) {
+		dev_dbg(dev, "Invalid hash fields, vsi num = %d\n",
+			vsi->vsi_num);
+		return -EINVAL;
+	}
+
+	hdrs = ice_parse_hdrs(nfc);
+	if (hdrs == ICE_FLOW_SEG_HDR_NONE) {
+		dev_dbg(dev, "Header type is not valid, vsi num = %d\n",
+			vsi->vsi_num);
+		return -EINVAL;
+	}
+
+	status = ice_add_rss_cfg(&pf->hw, vsi->idx, hashed_flds, hdrs);
+	if (status) {
+		dev_dbg(dev, "ice_add_rss_cfg failed, vsi num = %d, error = %d\n",
+			vsi->vsi_num, status);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+/**
+ * ice_get_rss_hash_opt - Retrieve hash fields for a given flow-type
+ * @vsi: the VSI being configured
+ * @nfc: ethtool rxnfc command
+ */
+static void
+ice_get_rss_hash_opt(struct ice_vsi *vsi, struct ethtool_rxnfc *nfc)
+{
+	struct ice_pf *pf = vsi->back;
+	struct device *dev;
+	u64 hash_flds;
+	u32 hdrs;
+
+	dev = ice_pf_to_dev(pf);
+
+	nfc->data = 0;
+	if (ice_is_safe_mode(pf)) {
+		dev_dbg(dev, "Advanced RSS disabled. Package download failed, vsi num = %d\n",
+			vsi->vsi_num);
+		return;
+	}
+
+	hdrs = ice_parse_hdrs(nfc);
+	if (hdrs == ICE_FLOW_SEG_HDR_NONE) {
+		dev_dbg(dev, "Header type is not valid, vsi num = %d\n",
+			vsi->vsi_num);
+		return;
+	}
+
+	hash_flds = ice_get_rss_cfg(&pf->hw, vsi->idx, hdrs);
+	if (hash_flds == ICE_HASH_INVALID) {
+		dev_dbg(dev, "No hash fields found for the given header type, vsi num = %d\n",
+			vsi->vsi_num);
+		return;
+	}
+
+	if (hash_flds & ICE_FLOW_HASH_FLD_IPV4_SA ||
+	    hash_flds & ICE_FLOW_HASH_FLD_IPV6_SA)
+		nfc->data |= (u64)RXH_IP_SRC;
+
+	if (hash_flds & ICE_FLOW_HASH_FLD_IPV4_DA ||
+	    hash_flds & ICE_FLOW_HASH_FLD_IPV6_DA)
+		nfc->data |= (u64)RXH_IP_DST;
+
+	if (hash_flds & ICE_FLOW_HASH_FLD_TCP_SRC_PORT ||
+	    hash_flds & ICE_FLOW_HASH_FLD_UDP_SRC_PORT ||
+	    hash_flds & ICE_FLOW_HASH_FLD_SCTP_SRC_PORT)
+		nfc->data |= (u64)RXH_L4_B_0_1;
+
+	if (hash_flds & ICE_FLOW_HASH_FLD_TCP_DST_PORT ||
+	    hash_flds & ICE_FLOW_HASH_FLD_UDP_DST_PORT ||
+	    hash_flds & ICE_FLOW_HASH_FLD_SCTP_DST_PORT)
+		nfc->data |= (u64)RXH_L4_B_2_3;
+}
+
+/**
+ * ice_set_rxnfc - command to set Rx flow rules.
+ * @netdev: network interface device structure
+ * @cmd: ethtool rxnfc command
+ *
+ * Returns 0 for success and negative values for errors
+ */
+static int ice_set_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd)
+{
+	struct ice_netdev_priv *np = netdev_priv(netdev);
+	struct ice_vsi *vsi = np->vsi;
+
+	switch (cmd->cmd) {
+	case ETHTOOL_SRXFH:
+		return ice_set_rss_hash_opt(vsi, cmd);
+	default:
+		break;
+	}
+	return -EOPNOTSUPP;
+}
+
 /**
  * ice_get_rxnfc - command to get Rx flow classification rules
  * @netdev: network interface device structure
@@ -2554,6 +2792,10 @@ ice_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd,
 		cmd->data = vsi->rss_size;
 		ret = 0;
 		break;
+	case ETHTOOL_GRXFH:
+		ice_get_rss_hash_opt(vsi, cmd);
+		ret = 0;
+		break;
 	default:
 		break;
 	}
@@ -3857,6 +4099,7 @@ static const struct ethtool_ops ice_ethtool_ops = {
 	.set_priv_flags		= ice_set_priv_flags,
 	.get_sset_count		= ice_get_sset_count,
 	.get_rxnfc		= ice_get_rxnfc,
+	.set_rxnfc		= ice_set_rxnfc,
 	.get_ringparam		= ice_get_ringparam,
 	.set_ringparam		= ice_set_ringparam,
 	.nway_reset		= ice_nway_reset,
diff --git a/drivers/net/ethernet/intel/ice/ice_flow.c b/drivers/net/ethernet/intel/ice/ice_flow.c
index 4ecac0ae565a..a05ceb59863b 100644
--- a/drivers/net/ethernet/intel/ice/ice_flow.c
+++ b/drivers/net/ethernet/intel/ice/ice_flow.c
@@ -1244,3 +1244,32 @@ enum ice_status ice_replay_rss_cfg(struct ice_hw *hw, u16 vsi_handle)
 
 	return status;
 }
+
+/**
+ * ice_get_rss_cfg - returns hashed fields for the given header types
+ * @hw: pointer to the hardware structure
+ * @vsi_handle: software VSI handle
+ * @hdrs: protocol header type
+ *
+ * This function will return the match fields of the first instance of flow
+ * profile having the given header types and containing input VSI
+ */
+u64 ice_get_rss_cfg(struct ice_hw *hw, u16 vsi_handle, u32 hdrs)
+{
+	struct ice_rss_cfg *r, *rss_cfg = NULL;
+
+	/* verify if the protocol header is non zero and VSI is valid */
+	if (hdrs == ICE_FLOW_SEG_HDR_NONE || !ice_is_vsi_valid(hw, vsi_handle))
+		return ICE_HASH_INVALID;
+
+	mutex_lock(&hw->rss_locks);
+	list_for_each_entry(r, &hw->rss_list_head, l_entry)
+		if (test_bit(vsi_handle, r->vsis) &&
+		    r->packet_hdr == hdrs) {
+			rss_cfg = r;
+			break;
+		}
+	mutex_unlock(&hw->rss_locks);
+
+	return rss_cfg ? rss_cfg->hashed_flds : ICE_HASH_INVALID;
+}
diff --git a/drivers/net/ethernet/intel/ice/ice_flow.h b/drivers/net/ethernet/intel/ice/ice_flow.h
index 475a025d3a2d..5558627bd5eb 100644
--- a/drivers/net/ethernet/intel/ice/ice_flow.h
+++ b/drivers/net/ethernet/intel/ice/ice_flow.h
@@ -203,4 +203,5 @@ enum ice_status ice_rem_vsi_rss_cfg(struct ice_hw *hw, u16 vsi_handle);
 enum ice_status
 ice_add_rss_cfg(struct ice_hw *hw, u16 vsi_handle, u64 hashed_flds,
 		u32 addl_hdrs);
+u64 ice_get_rss_cfg(struct ice_hw *hw, u16 vsi_handle, u32 hdrs);
 #endif /* _ICE_FLOW_H_ */
-- 
cgit v1.2.3-59-g8ed1b


From 18a8d3586310743d451c94db46319797e72727da Mon Sep 17 00:00:00 2001
From: Tony Nguyen <anthony.l.nguyen@intel.com>
Date: Fri, 17 Jan 2020 07:39:19 -0800
Subject: ice: Bump version

Bump version to 0.8.2-k

Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index eb9d00608e9a..5ae671609f98 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -13,7 +13,7 @@
 
 #define DRV_VERSION_MAJOR 0
 #define DRV_VERSION_MINOR 8
-#define DRV_VERSION_BUILD 1
+#define DRV_VERSION_BUILD 2
 
 #define DRV_VERSION	__stringify(DRV_VERSION_MAJOR) "." \
 			__stringify(DRV_VERSION_MINOR) "." \
-- 
cgit v1.2.3-59-g8ed1b


From 11eb85ec42dc8c7a7ec519b90ccf2eeae9409de8 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@oracle.com>
Date: Wed, 15 Jan 2020 20:49:04 +0300
Subject: Bluetooth: Fix race condition in hci_release_sock()

Syzbot managed to trigger a use after free "KASAN: use-after-free Write
in hci_sock_bind".  I have reviewed the code manually and one possibly
cause I have found is that we are not holding lock_sock(sk) when we do
the hci_dev_put(hdev) in hci_sock_release().  My theory is that the bind
and the release are racing against each other which results in this use
after free.

Reported-by: syzbot+eba992608adf3d796bcc@syzkaller.appspotmail.com
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
---
 net/bluetooth/hci_sock.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/net/bluetooth/hci_sock.c b/net/bluetooth/hci_sock.c
index 5756b124c2cb..9c4a093f8960 100644
--- a/net/bluetooth/hci_sock.c
+++ b/net/bluetooth/hci_sock.c
@@ -839,6 +839,8 @@ static int hci_sock_release(struct socket *sock)
 	if (!sk)
 		return 0;
 
+	lock_sock(sk);
+
 	switch (hci_pi(sk)->channel) {
 	case HCI_CHANNEL_MONITOR:
 		atomic_dec(&monitor_promisc);
@@ -886,6 +888,7 @@ static int hci_sock_release(struct socket *sock)
 	skb_queue_purge(&sk->sk_receive_queue);
 	skb_queue_purge(&sk->sk_write_queue);
 
+	release_sock(sk);
 	sock_put(sk);
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From d27c768f3b617a5a94844e58b18f773d2e40f41c Mon Sep 17 00:00:00 2001
From: zhengbin <zhengbin13@huawei.com>
Date: Tue, 24 Dec 2019 22:16:03 +0800
Subject: ath9k: use true,false for bool variable

Fixes coccicheck warning:

drivers/net/wireless/ath/ath9k/ar9003_aic.c:409:2-12: WARNING: Assignment of 0/1 to bool variable

Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: zhengbin <zhengbin13@huawei.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath9k/ar9003_aic.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath9k/ar9003_aic.c b/drivers/net/wireless/ath/ath9k/ar9003_aic.c
index 547cd46da260..d0f1e8bcc846 100644
--- a/drivers/net/wireless/ath/ath9k/ar9003_aic.c
+++ b/drivers/net/wireless/ath/ath9k/ar9003_aic.c
@@ -406,7 +406,7 @@ static bool ar9003_aic_cal_post_process(struct ath_hw *ah)
 		sram.com_att_6db =
 			ar9003_aic_find_index(1, fixed_com_att_db);
 
-		sram.valid = 1;
+		sram.valid = true;
 
 		sram.rot_dir_att_db =
 			min(max(rot_dir_path_att_db,
-- 
cgit v1.2.3-59-g8ed1b


From f89ee9927a49d805ae355c6d88a22e11a7ec083a Mon Sep 17 00:00:00 2001
From: zhengbin <zhengbin13@huawei.com>
Date: Tue, 24 Dec 2019 22:16:05 +0800
Subject: wil6210: use true,false for bool variable

Fixes coccicheck warning:

drivers/net/wireless/ath/wil6210/main.c:765:1-14: WARNING: Assignment of 0/1 to bool variable
drivers/net/wireless/ath/wil6210/txrx.c:1143:1-19: WARNING: Assignment of 0/1 to bool variable
drivers/net/wireless/ath/wil6210/wmi.c:1516:4-23: WARNING: Assignment of 0/1 to bool variable
drivers/net/wireless/ath/wil6210/wmi.c:1523:4-23: WARNING: Assignment of 0/1 to bool variable
drivers/net/wireless/ath/wil6210/wmi.c:1538:4-30: WARNING: Assignment of 0/1 to bool variable
drivers/net/wireless/ath/wil6210/wmi.c:1545:4-30: WARNING: Assignment of 0/1 to bool variable

Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: zhengbin <zhengbin13@huawei.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/wil6210/main.c | 2 +-
 drivers/net/wireless/ath/wil6210/txrx.c | 2 +-
 drivers/net/wireless/ath/wil6210/wmi.c  | 8 ++++----
 3 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/ath/wil6210/main.c b/drivers/net/wireless/ath/wil6210/main.c
index 6d39547e20fd..3ba5b2550a8c 100644
--- a/drivers/net/wireless/ath/wil6210/main.c
+++ b/drivers/net/wireless/ath/wil6210/main.c
@@ -762,7 +762,7 @@ int wil_priv_init(struct wil6210_priv *wil)
 	 */
 	wil->rx_buff_id_count = WIL_RX_BUFF_ARR_SIZE_DEFAULT;
 
-	wil->amsdu_en = 1;
+	wil->amsdu_en = true;
 
 	return 0;
 
diff --git a/drivers/net/wireless/ath/wil6210/txrx.c b/drivers/net/wireless/ath/wil6210/txrx.c
index 17118d643d7e..bc8c15fb609d 100644
--- a/drivers/net/wireless/ath/wil6210/txrx.c
+++ b/drivers/net/wireless/ath/wil6210/txrx.c
@@ -1140,7 +1140,7 @@ static int wil_tx_desc_map(union wil_tx_desc *desc, dma_addr_t pa,
 void wil_tx_data_init(struct wil_ring_tx_data *txdata)
 {
 	spin_lock_bh(&txdata->lock);
-	txdata->dot1x_open = 0;
+	txdata->dot1x_open = false;
 	txdata->enabled = 0;
 	txdata->idle = 0;
 	txdata->last_idle = 0;
diff --git a/drivers/net/wireless/ath/wil6210/wmi.c b/drivers/net/wireless/ath/wil6210/wmi.c
index dcba0a4c47b4..23e1ed6a9d6d 100644
--- a/drivers/net/wireless/ath/wil6210/wmi.c
+++ b/drivers/net/wireless/ath/wil6210/wmi.c
@@ -1513,14 +1513,14 @@ static void wmi_link_stats_parse(struct wil6210_vif *vif, u64 tsf,
 			if (vif->fw_stats_ready) {
 				/* clean old statistics */
 				vif->fw_stats_tsf = 0;
-				vif->fw_stats_ready = 0;
+				vif->fw_stats_ready = false;
 			}
 
 			wil_link_stats_store_basic(vif, payload + hdr_size);
 
 			if (!has_next) {
 				vif->fw_stats_tsf = tsf;
-				vif->fw_stats_ready = 1;
+				vif->fw_stats_ready = true;
 			}
 
 			break;
@@ -1535,14 +1535,14 @@ static void wmi_link_stats_parse(struct wil6210_vif *vif, u64 tsf,
 			if (wil->fw_stats_global.ready) {
 				/* clean old statistics */
 				wil->fw_stats_global.tsf = 0;
-				wil->fw_stats_global.ready = 0;
+				wil->fw_stats_global.ready = false;
 			}
 
 			wil_link_stats_store_global(vif, payload + hdr_size);
 
 			if (!has_next) {
 				wil->fw_stats_global.tsf = tsf;
-				wil->fw_stats_global.ready = 1;
+				wil->fw_stats_global.ready = true;
 			}
 
 			break;
-- 
cgit v1.2.3-59-g8ed1b


From d239380196c4e27a26fa4bea73d2bf994c14ec2d Mon Sep 17 00:00:00 2001
From: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Date: Thu, 19 Dec 2019 13:15:38 +0000
Subject: ath10k: pci: Only dump ATH10K_MEM_REGION_TYPE_IOREG when safe

ath10k_pci_dump_memory_reg() will try to access memory of type
ATH10K_MEM_REGION_TYPE_IOREG however, if a hardware restart is in progress
this can crash a system.

Individual ioread32() time has been observed to jump from 15-20 ticks to >
80k ticks followed by a secure-watchdog bite and a system reset.

Work around this corner case by only issuing the read transaction when the
driver state is ATH10K_STATE_ON.

Tested-on: QCA9988 PCI 10.4-3.9.0.2-00044

Fixes: 219cc084c6706 ("ath10k: add memory dump support QCA9984")
Signed-off-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/pci.c | 19 +++++++++++++++++--
 1 file changed, 17 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/pci.c b/drivers/net/wireless/ath/ath10k/pci.c
index bb44f5a0941b..4822a65f6f3c 100644
--- a/drivers/net/wireless/ath/ath10k/pci.c
+++ b/drivers/net/wireless/ath/ath10k/pci.c
@@ -1604,11 +1604,22 @@ static int ath10k_pci_dump_memory_reg(struct ath10k *ar,
 {
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
 	u32 i;
+	int ret;
+
+	mutex_lock(&ar->conf_mutex);
+	if (ar->state != ATH10K_STATE_ON) {
+		ath10k_warn(ar, "Skipping pci_dump_memory_reg invalid state\n");
+		ret = -EIO;
+		goto done;
+	}
 
 	for (i = 0; i < region->len; i += 4)
 		*(u32 *)(buf + i) = ioread32(ar_pci->mem + region->start + i);
 
-	return region->len;
+	ret = region->len;
+done:
+	mutex_unlock(&ar->conf_mutex);
+	return ret;
 }
 
 /* if an error happened returns < 0, otherwise the length */
@@ -1704,7 +1715,11 @@ static void ath10k_pci_dump_memory(struct ath10k *ar,
 			count = ath10k_pci_dump_memory_sram(ar, current_region, buf);
 			break;
 		case ATH10K_MEM_REGION_TYPE_IOREG:
-			count = ath10k_pci_dump_memory_reg(ar, current_region, buf);
+			ret = ath10k_pci_dump_memory_reg(ar, current_region, buf);
+			if (ret < 0)
+				break;
+
+			count = ret;
 			break;
 		default:
 			ret = ath10k_pci_dump_memory_generic(ar, current_region, buf);
-- 
cgit v1.2.3-59-g8ed1b


From 63ec5cbc31f7c37870b7fbd33d9f1a50ba320221 Mon Sep 17 00:00:00 2001
From: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Date: Thu, 19 Dec 2019 13:15:39 +0000
Subject: ath10k: pci: Fix comment on ath10k_pci_dump_memory_sram

The description of ath10k_pci_dump_memory_sram() is inaccurate, an error
can never be returned, it is always the length. Update the comment to
reflect.

Fixes: 219cc084c6706 ("ath10k: add memory dump support QCA9984")
Signed-off-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/pci.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath10k/pci.c b/drivers/net/wireless/ath/ath10k/pci.c
index 4822a65f6f3c..ded7a220a4aa 100644
--- a/drivers/net/wireless/ath/ath10k/pci.c
+++ b/drivers/net/wireless/ath/ath10k/pci.c
@@ -1578,7 +1578,7 @@ static int ath10k_pci_set_ram_config(struct ath10k *ar, u32 config)
 	return 0;
 }
 
-/* if an error happened returns < 0, otherwise the length */
+/* Always returns the length */
 static int ath10k_pci_dump_memory_sram(struct ath10k *ar,
 				       const struct ath10k_mem_region *region,
 				       u8 *buf)
-- 
cgit v1.2.3-59-g8ed1b


From 57a3b60ddbfa87e8707c453c69915a3c5d2c9e50 Mon Sep 17 00:00:00 2001
From: Bjorn Andersson <bjorn.andersson@linaro.org>
Date: Sun, 22 Dec 2019 21:48:54 -0800
Subject: ath10k: Add optional qdss clk

The WiFi firmware found on sm8150 requires that the QDSS clock is
ticking in order to operate, so add an optional clock to the binding to
allow this to be specified in the sm8150 dts and add the clock to the
list of clocks in the driver.

Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 Documentation/devicetree/bindings/net/wireless/qcom,ath10k.txt | 2 +-
 drivers/net/wireless/ath/ath10k/snoc.c                         | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/Documentation/devicetree/bindings/net/wireless/qcom,ath10k.txt b/Documentation/devicetree/bindings/net/wireless/qcom,ath10k.txt
index 017128394a3e..3fc2cce4626b 100644
--- a/Documentation/devicetree/bindings/net/wireless/qcom,ath10k.txt
+++ b/Documentation/devicetree/bindings/net/wireless/qcom,ath10k.txt
@@ -50,7 +50,7 @@ Optional properties:
           entry in clock-names.
 - clock-names: Should contain the clock names "wifi_wcss_cmd", "wifi_wcss_ref",
 	       "wifi_wcss_rtc" for "qcom,ipq4019-wifi" compatible target and
-	       "cxo_ref_clk_pin" for "qcom,wcn3990-wifi"
+	       "cxo_ref_clk_pin" and optionally "qdss" for "qcom,wcn3990-wifi"
 	       compatible target.
 - qcom,msi_addr: MSI interrupt address.
 - qcom,msi_base: Base value to add before writing MSI data into
diff --git a/drivers/net/wireless/ath/ath10k/snoc.c b/drivers/net/wireless/ath/ath10k/snoc.c
index 7e85c4916e7f..aeb4cca92c35 100644
--- a/drivers/net/wireless/ath/ath10k/snoc.c
+++ b/drivers/net/wireless/ath/ath10k/snoc.c
@@ -46,7 +46,7 @@ static const char * const ath10k_regulators[] = {
 };
 
 static const char * const ath10k_clocks[] = {
-	"cxo_ref_clk_pin",
+	"cxo_ref_clk_pin", "qdss",
 };
 
 static void ath10k_snoc_htc_tx_cb(struct ath10k_ce_pipe *ce_state);
-- 
cgit v1.2.3-59-g8ed1b


From 0f7ab288ade970b7754d89b594e872ea40af8f66 Mon Sep 17 00:00:00 2001
From: zhengbin <zhengbin13@huawei.com>
Date: Tue, 24 Dec 2019 22:16:04 +0800
Subject: ath10k: use true,false for bool variable

Fixes coccicheck warning:

drivers/net/wireless/ath/ath10k/htt_rx.c:2143:2-31: WARNING: Assignment of 0/1 to bool variable

Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: zhengbin <zhengbin13@huawei.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/htt_rx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c
index 7fb15180c685..38a5814cf345 100644
--- a/drivers/net/wireless/ath/ath10k/htt_rx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_rx.c
@@ -2140,7 +2140,7 @@ static bool ath10k_htt_rx_pn_check_replay_hl(struct ath10k *ar,
 	if (last_pn_valid)
 		pn_invalid = ath10k_htt_rx_pn_cmp48(&new_pn, last_pn);
 	else
-		peer->tids_last_pn_valid[tid] = 1;
+		peer->tids_last_pn_valid[tid] = true;
 
 	if (!pn_invalid)
 		last_pn->pn48 = new_pn.pn48;
-- 
cgit v1.2.3-59-g8ed1b


From 218f646d235a84d716c603100a6aa5447bc3c9a2 Mon Sep 17 00:00:00 2001
From: Wen Gong <wgong@codeaurora.org>
Date: Tue, 31 Dec 2019 17:27:06 +0800
Subject: ath10k: drop RX skb with invalid length for sdio

When simulate random transfer fail for sdio write and read, it crash
sometimes.

Test steps:
1. Add config and update kernel:
CONFIG_FAIL_MMC_REQUEST=y
CONFIG_FAULT_INJECTION=y
CONFIG_FAULT_INJECTION_DEBUG_FS=y

2. run simulate fail:
cd /sys/kernel/debug/mmc1/fail_mmc_request
echo 10 > probability
echo 10 > times # repeat until hitting issues

3. it crash, the act len of ath10k_htc_hdr is higher than allocate len, it cause panic:
[   99.723482] skbuff: skb_over_panic: text:00000000caa0f780 len:57013 put:57013 head:000000004116f24a data:0000000019ecb4dc tail:0xdef5 end:0x640 dev:<NULL>
[   99.737697] ------------[ cut here ]------------
[   99.742327] kernel BUG at /mnt/host/source/src/third_party/kernel/v4.19/net/core/skbuff.c:104!
[   99.750937] Internal error: Oops - BUG: 0 [#1] PREEMPT SMP
[   99.831154] Process kworker/0:2 (pid: 151, stack limit = 0x00000000728010bf)
[   99.838200] CPU: 0 PID: 151 Comm: kworker/0:2 Tainted: G W 4.19.85 #48
[   99.846022] Hardware name: MediaTek krane sku0 board (DT)
[   99.851429] Workqueue: events sdio_irq_work
[   99.855614] pstate: 60000005 (nZCv daif -PAN -UAO)
[   99.860402] pc : skb_panic+0x64/0x68
[   99.863974] lr : skb_panic+0x64/0x68
[   99.867542] sp : ffffff8008833a90
[   99.870850] x29: ffffff8008833ac0 x28: ffffffe52e337370
[   99.876159] x27: ffffffe52e328a90 x26: 000000000000e0d0
[   99.881469] x25: ffffffe52e336b60 x24: 000000000000deb5
[   99.886779] x23: ffffffe52e340680 x22: ffffffe4efd47e00
[   99.892088] x21: 000000000000deb5 x20: ffffffa516d85b4c
[   99.897397] x19: ffffffa526928037 x18: 0000000000000000
[   99.902706] x17: 000000000000003c x16: ffffffa5265b6c80
[   99.908015] x15: 0000000000000006 x14: 3a76656420303436
[   99.913325] x13: 0000000000029bf0 x12: 0000000000000000
[   99.918634] x11: 0000000000000000 x10: 0000000000000000
[   99.923943] x9 : a3b907e4b2783000 x8 : a3b907e4b2783000
[   99.929253] x7 : 0000000000000000 x6 : ffffffa526f66d76
[   99.934563] x5 : 0000000000000000 x4 : 0000000000000000
[   99.939872] x3 : 000000000002a5ab x2 : ffffffe53feed918
[   99.945182] x1 : ffffffe53fee4a08 x0 : 000000000000008e
[   99.950491] Call trace:
[   99.952937]  skb_panic+0x64/0x68
[   99.956165]  skb_put+0x7c/0x84
[   99.959224]  ath10k_sdio_irq_handler+0x740/0xbb8 [ath10k_sdio]
[   99.965055]  process_sdio_pending_irqs+0x58/0x1a4
[   99.969758]  sdio_run_irqs+0x34/0x60
[   99.973329]  sdio_irq_work+0x1c/0x28
[   99.974930] cros-ec-spi spi2.0: SPI transfer timed out
[   99.976904]  process_one_work+0x210/0x410
[   99.976911]  worker_thread+0x234/0x3dc
[   99.976923]  kthread+0x120/0x130
[   99.982090] cros-ec-spi spi2.0: spi transfer failed: -110
[   99.986054]  ret_from_fork+0x10/0x18
[   99.986063] Code: aa1403e2 2a1503e4 a90023e9 97e37d1a (d4210000)
[   99.986068] ---[ end trace cb6d948c5a0fd6c7 ]---
[  100.017250] Kernel panic - not syncing: Fatal exception
[  100.018879] cros-ec-spi spi2.0: Command xfer error (err:-110)
[  100.023659] SMP: stopping secondary CPUs
[  100.023703] Kernel Offset: 0x251dc00000 from 0xffffff8008000000
[  100.023707] CPU features: 0x0,2188200c
[  100.023709] Memory Limit: none

The simulate fail of sdio is not a real sdio transter fail, it only
set an error status in mmc_should_fail_request after the transfer end,
actually the transfer is success, then sdio_io_rw_ext_helper will
return error status and stop transfer the left data. For example,
the really RX len is 286 bytes, then it will split to 2 blocks in
sdio_io_rw_ext_helper, one is 256 bytes, left is 30 bytes, if the
first 256 bytes get an error status by mmc_should_fail_request,then
the left 30 bytes will not read in this RX operation. Then when the
next RX arrive, the left 30 bytes will be considered as the header
of the read, the top 8 bytes will be considered as ath10k_htc_hdr,
but actually the 8 bytes is not the ath10k_htc_hdr, so the act_len
from this ath10k_htc_hdr is not correct, if it is a big value, such
as 57013, it will trigger skb_panic.

Drop the skb with invalid length will be reasonable.

This patch only effect sdio chips.

Tested with QCA6174 SDIO with firmware WLAN.RMH.4.4.1-00029.

Signed-off-by: Wen Gong <wgong@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/sdio.c | 24 ++++++++++++++++++------
 1 file changed, 18 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/sdio.c b/drivers/net/wireless/ath/ath10k/sdio.c
index 6fdf71b8b676..e5316b911e1d 100644
--- a/drivers/net/wireless/ath/ath10k/sdio.c
+++ b/drivers/net/wireless/ath/ath10k/sdio.c
@@ -642,16 +642,23 @@ static int ath10k_sdio_mbox_rx_fetch(struct ath10k *ar)
 
 	ret = ath10k_sdio_readsb(ar, ar_sdio->mbox_info.htc_addr,
 				 skb->data, pkt->alloc_len);
-
-	if (ret) {
-		ar_sdio->n_rx_pkts = 0;
-		ath10k_sdio_mbox_free_rx_pkt(pkt);
-		return ret;
-	}
+	if (ret)
+		goto err;
 
 	htc_hdr = (struct ath10k_htc_hdr *)skb->data;
 	pkt->act_len = le16_to_cpu(htc_hdr->len) + sizeof(*htc_hdr);
+
+	if (pkt->act_len > pkt->alloc_len) {
+		ret = -EINVAL;
+		goto err;
+	}
+
 	skb_put(skb, pkt->act_len);
+	return 0;
+
+err:
+	ar_sdio->n_rx_pkts = 0;
+	ath10k_sdio_mbox_free_rx_pkt(pkt);
 
 	return ret;
 }
@@ -687,6 +694,11 @@ static int ath10k_sdio_mbox_rx_fetch_bundle(struct ath10k *ar)
 		htc_hdr = (struct ath10k_htc_hdr *)(ar_sdio->vsg_buffer + pkt_offset);
 		pkt->act_len = le16_to_cpu(htc_hdr->len) + sizeof(*htc_hdr);
 
+		if (pkt->act_len > pkt->alloc_len ) {
+			ret = -EINVAL;
+			goto err;
+		}
+
 		skb_put_data(pkt->skb, htc_hdr, pkt->act_len);
 		pkt_offset += pkt->alloc_len;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From a1769bb68a850508a492e3674ab1e5e479b11254 Mon Sep 17 00:00:00 2001
From: Zhi Chen <zhichen@codeaurora.org>
Date: Tue, 14 Jan 2020 12:35:21 +0800
Subject: Revert "ath10k: fix DMA related firmware crashes on multiple devices"

This reverts commit 76d164f582150fd0259ec0fcbc485470bcd8033e.
PCIe hung issue was observed on multiple platforms. The issue was reproduced
when DUT was configured as AP and associated with 50+ STAs.

For QCA9984/QCA9888, the DMA_BURST_SIZE register controls the AXI burst size
of the RD/WR access to the HOST MEM.
0 - No split , RAW read/write transfer size from MAC is put out on bus
    as burst length
1 - Split at 256 byte boundary
2,3 - Reserved

With PCIe protocol analyzer, we can see DMA Read crossing 4KB boundary when
issue happened. It broke PCIe spec and caused PCIe stuck. So revert
the default value from 0 to 1.

Tested:  IPQ8064 + QCA9984 with firmware 10.4-3.10-00047
         QCS404 + QCA9984 with firmware 10.4-3.9.0.2--00044
         Synaptics AS370 + QCA9888  with firmware 10.4-3.9.0.2--00040

Signed-off-by: Zhi Chen <zhichen@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/hw.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath10k/hw.h b/drivers/net/wireless/ath/ath10k/hw.h
index 21b7a2a873b0..775fd62fb92d 100644
--- a/drivers/net/wireless/ath/ath10k/hw.h
+++ b/drivers/net/wireless/ath/ath10k/hw.h
@@ -816,7 +816,7 @@ ath10k_is_rssi_enable(struct ath10k_hw_params *hw,
 
 #define TARGET_10_4_TX_DBG_LOG_SIZE		1024
 #define TARGET_10_4_NUM_WDS_ENTRIES		32
-#define TARGET_10_4_DMA_BURST_SIZE		0
+#define TARGET_10_4_DMA_BURST_SIZE		1
 #define TARGET_10_4_MAC_AGGR_DELIM		0
 #define TARGET_10_4_RX_SKIP_DEFRAG_TIMEOUT_DUP_DETECTION_CHECK 1
 #define TARGET_10_4_VOW_CONFIG			0
-- 
cgit v1.2.3-59-g8ed1b


From 6ca71a10aee8d5613446a180f85aa5efbb370b10 Mon Sep 17 00:00:00 2001
From: Govind Singh <govinds@codeaurora.org>
Date: Fri, 3 Jan 2020 19:14:13 +0530
Subject: dt: bindings: add dt entry flag to skip SCM call for msa region

Add boolean context flag to disable SCM call for statically
mapped msa region.

Signed-off-by: Govind Singh <govinds@codeaurora.org>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 Documentation/devicetree/bindings/net/wireless/qcom,ath10k.txt | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/wireless/qcom,ath10k.txt b/Documentation/devicetree/bindings/net/wireless/qcom,ath10k.txt
index 3fc2cce4626b..616c87746d6f 100644
--- a/Documentation/devicetree/bindings/net/wireless/qcom,ath10k.txt
+++ b/Documentation/devicetree/bindings/net/wireless/qcom,ath10k.txt
@@ -88,6 +88,9 @@ Optional properties:
 		    of the host capability QMI request
 - qcom,xo-cal-data: xo cal offset to be configured in xo trim register.
 
+- qcom,msa-fixed-perm: Boolean context flag to disable SCM call for statically
+		       mapped msa region.
+
 Example (to supply PCI based wifi block details):
 
 In this example, the node is defined as child node of the PCI controller.
@@ -185,4 +188,5 @@ wifi@18000000 {
 		vdd-3.3-ch0-supply = <&vreg_l25a_3p3>;
 		memory-region = <&wifi_msa_mem>;
 		iommus = <&apps_smmu 0x0040 0x1>;
+		qcom,msa-fixed-perm;
 };
-- 
cgit v1.2.3-59-g8ed1b


From ab000ea6a3939972b6eaca78046c4cd7d3aeee22 Mon Sep 17 00:00:00 2001
From: Govind Singh <govinds@codeaurora.org>
Date: Fri, 3 Jan 2020 19:14:14 +0530
Subject: ath10k: Don't call SCM interface for statically mapped msa region

For some targets ex: QCS404, SCM permissions for MSA region is
statically configured in TrustZone fw. Add SCM call disable option
for such targets to avoid duplicate permissions.

Testing: Tested on WCN3990 HW
Tested FW: WLAN.HL.3.1-01040-QCAHLSWMTPLZ-1

Signed-off-by: Govind Singh <govinds@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/qmi.c | 9 +++++++++
 drivers/net/wireless/ath/ath10k/qmi.h | 1 +
 2 files changed, 10 insertions(+)

diff --git a/drivers/net/wireless/ath/ath10k/qmi.c b/drivers/net/wireless/ath/ath10k/qmi.c
index 7b524b614f22..85dce43c5439 100644
--- a/drivers/net/wireless/ath/ath10k/qmi.c
+++ b/drivers/net/wireless/ath/ath10k/qmi.c
@@ -84,6 +84,9 @@ static int ath10k_qmi_setup_msa_permissions(struct ath10k_qmi *qmi)
 	int ret;
 	int i;
 
+	if (qmi->msa_fixed_perm)
+		return 0;
+
 	for (i = 0; i < qmi->nr_mem_region; i++) {
 		ret = ath10k_qmi_map_msa_permission(qmi, &qmi->mem_region[i]);
 		if (ret)
@@ -102,6 +105,9 @@ static void ath10k_qmi_remove_msa_permission(struct ath10k_qmi *qmi)
 {
 	int i;
 
+	if (qmi->msa_fixed_perm)
+		return;
+
 	for (i = 0; i < qmi->nr_mem_region; i++)
 		ath10k_qmi_unmap_msa_permission(qmi, &qmi->mem_region[i]);
 }
@@ -1035,6 +1041,9 @@ static int ath10k_qmi_setup_msa_resources(struct ath10k_qmi *qmi, u32 msa_size)
 		qmi->msa_mem_size = msa_size;
 	}
 
+	if (of_property_read_bool(dev->of_node, "qcom,msa-fixed-perm"))
+		qmi->msa_fixed_perm = true;
+
 	ath10k_dbg(ar, ATH10K_DBG_QMI, "msa pa: %pad , msa va: 0x%p\n",
 		   &qmi->msa_pa,
 		   qmi->msa_va);
diff --git a/drivers/net/wireless/ath/ath10k/qmi.h b/drivers/net/wireless/ath/ath10k/qmi.h
index 40aafb875ed0..dc257375f161 100644
--- a/drivers/net/wireless/ath/ath10k/qmi.h
+++ b/drivers/net/wireless/ath/ath10k/qmi.h
@@ -104,6 +104,7 @@ struct ath10k_qmi {
 	bool fw_ready;
 	char fw_build_timestamp[MAX_TIMESTAMP_LEN + 1];
 	struct ath10k_qmi_cal_data cal_data[MAX_NUM_CAL_V01];
+	bool msa_fixed_perm;
 };
 
 int ath10k_qmi_wlan_enable(struct ath10k *ar,
-- 
cgit v1.2.3-59-g8ed1b


From a45ceea5015d5be4bab4f6ce59cffcbeceba7e6c Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Tue, 7 Jan 2020 22:50:04 +0100
Subject: ath11k: fix debugfs build failure

When CONFIG_ATH11K_DEBUGFS is disabled, but CONFIG_MAC80211_DEBUGFS
is turned on, the driver fails to build:

drivers/net/wireless/ath/ath11k/debugfs_sta.c: In function 'ath11k_dbg_sta_open_htt_peer_stats':
drivers/net/wireless/ath/ath11k/debugfs_sta.c:416:4: error: 'struct ath11k' has no member named 'debug'
  ar->debug.htt_stats.stats_req = stats_req;
    ^~

It appears that just using the former symbol is sufficient here,
adding a Kconfig dependency takes care of the corner cases.

Fixes: d5c65159f289 ("ath11k: driver for Qualcomm IEEE 802.11ax devices")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/Kconfig  |  2 +-
 drivers/net/wireless/ath/ath11k/Makefile |  3 +--
 drivers/net/wireless/ath/ath11k/debug.h  | 22 ++++++++++------------
 drivers/net/wireless/ath/ath11k/mac.c    |  2 +-
 4 files changed, 13 insertions(+), 16 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/Kconfig b/drivers/net/wireless/ath/ath11k/Kconfig
index cfab4fb86aef..c88e16d4022b 100644
--- a/drivers/net/wireless/ath/ath11k/Kconfig
+++ b/drivers/net/wireless/ath/ath11k/Kconfig
@@ -22,7 +22,7 @@ config ATH11K_DEBUG
 
 config ATH11K_DEBUGFS
 	bool "QCA ath11k debugfs support"
-	depends on ATH11K && DEBUG_FS
+	depends on ATH11K && DEBUG_FS && MAC80211_DEBUGFS
 	---help---
 	  Enable ath11k debugfs support
 
diff --git a/drivers/net/wireless/ath/ath11k/Makefile b/drivers/net/wireless/ath/ath11k/Makefile
index a91d75c1cfeb..2761d07d938e 100644
--- a/drivers/net/wireless/ath/ath11k/Makefile
+++ b/drivers/net/wireless/ath/ath11k/Makefile
@@ -17,8 +17,7 @@ ath11k-y += core.o \
 	    ce.o \
 	    peer.o
 
-ath11k-$(CONFIG_ATH11K_DEBUGFS) += debug_htt_stats.o
-ath11k-$(CONFIG_MAC80211_DEBUGFS) += debugfs_sta.o
+ath11k-$(CONFIG_ATH11K_DEBUGFS) += debug_htt_stats.o debugfs_sta.o
 ath11k-$(CONFIG_NL80211_TESTMODE) += testmode.o
 ath11k-$(CONFIG_ATH11K_TRACING) += trace.o
 
diff --git a/drivers/net/wireless/ath/ath11k/debug.h b/drivers/net/wireless/ath/ath11k/debug.h
index a317a7bdb9a2..8e8d5588b541 100644
--- a/drivers/net/wireless/ath/ath11k/debug.h
+++ b/drivers/net/wireless/ath/ath11k/debug.h
@@ -172,6 +172,16 @@ static inline int ath11k_debug_is_extd_rx_stats_enabled(struct ath11k *ar)
 {
 	return ar->debug.extd_rx_stats;
 }
+
+void ath11k_sta_add_debugfs(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+			    struct ieee80211_sta *sta, struct dentry *dir);
+void
+ath11k_accumulate_per_peer_tx_stats(struct ath11k_sta *arsta,
+				    struct ath11k_per_peer_tx_stats *peer_stats,
+				    u8 legacy_rate_idx);
+void ath11k_update_per_peer_stats_from_txcompl(struct ath11k *ar,
+					       struct sk_buff *msdu,
+					       struct hal_tx_status *ts);
 #else
 static inline int ath11k_debug_soc_create(struct ath11k_base *ab)
 {
@@ -243,19 +253,7 @@ static inline bool ath11k_debug_is_pktlog_peer_valid(struct ath11k *ar, u8 *addr
 {
 	return false;
 }
-#endif /* CONFIG_ATH11K_DEBUGFS */
 
-#ifdef CONFIG_MAC80211_DEBUGFS
-void ath11k_sta_add_debugfs(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
-			    struct ieee80211_sta *sta, struct dentry *dir);
-void
-ath11k_accumulate_per_peer_tx_stats(struct ath11k_sta *arsta,
-				    struct ath11k_per_peer_tx_stats *peer_stats,
-				    u8 legacy_rate_idx);
-void ath11k_update_per_peer_stats_from_txcompl(struct ath11k *ar,
-					       struct sk_buff *msdu,
-					       struct hal_tx_status *ts);
-#else /* !CONFIG_MAC80211_DEBUGFS */
 static inline void
 ath11k_accumulate_per_peer_tx_stats(struct ath11k_sta *arsta,
 				    struct ath11k_per_peer_tx_stats *peer_stats,
diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index 556eef9881a7..0ed3e4d19f7a 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -5468,7 +5468,7 @@ static const struct ieee80211_ops ath11k_ops = {
 	.flush				= ath11k_mac_op_flush,
 	.sta_statistics			= ath11k_mac_op_sta_statistics,
 	CFG80211_TESTMODE_CMD(ath11k_tm_cmd)
-#ifdef CONFIG_MAC80211_DEBUGFS
+#ifdef CONFIG_ATH11K_DEBUGFS
 	.sta_add_debugfs		= ath11k_sta_add_debugfs,
 #endif
 };
-- 
cgit v1.2.3-59-g8ed1b


From 1e93a78113b4eaf009b3c994595db44356544b05 Mon Sep 17 00:00:00 2001
From: Anilkumar Kolli <akolli@codeaurora.org>
Date: Tue, 17 Dec 2019 17:21:40 +0100
Subject: ath11k: enable HE tlvs in ppdu stats for pktlog lite

This patch enables HE tlvs in ppdu stats for pktlog lite mode.

Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Anilkumar Kolli <akolli@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/debug.c | 15 +++++++++++++++
 drivers/net/wireless/ath/ath11k/dp.h    |  8 ++++++++
 2 files changed, 23 insertions(+)

diff --git a/drivers/net/wireless/ath/ath11k/debug.c b/drivers/net/wireless/ath/ath11k/debug.c
index f48daf17f2d2..8d485171b0b3 100644
--- a/drivers/net/wireless/ath/ath11k/debug.c
+++ b/drivers/net/wireless/ath/ath11k/debug.c
@@ -931,7 +931,22 @@ static ssize_t ath11k_write_pktlog_filter(struct file *file,
 			    HTT_RX_FILTER_TLV_FLAGS_PACKET_HEADER |
 			    HTT_RX_FILTER_TLV_FLAGS_ATTENTION;
 	} else if (mode == ATH11K_PKTLOG_MODE_LITE) {
+		ret = ath11k_dp_tx_htt_h2t_ppdu_stats_req(ar,
+							  HTT_PPDU_STATS_TAG_PKTLOG);
+		if (ret) {
+			ath11k_err(ar->ab, "failed to enable pktlog lite: %d\n", ret);
+			goto out;
+		}
+
 		rx_filter = HTT_RX_FILTER_TLV_LITE_MODE;
+	} else {
+		ret = ath11k_dp_tx_htt_h2t_ppdu_stats_req(ar,
+							  HTT_PPDU_STATS_TAG_DEFAULT);
+		if (ret) {
+			ath11k_err(ar->ab, "failed to send htt ppdu stats req: %d\n",
+				   ret);
+			goto out;
+		}
 	}
 
 	tlv_filter.rx_filter = rx_filter;
diff --git a/drivers/net/wireless/ath/ath11k/dp.h b/drivers/net/wireless/ath/ath11k/dp.h
index 2f0980f2c762..6ef5be4201b2 100644
--- a/drivers/net/wireless/ath/ath11k/dp.h
+++ b/drivers/net/wireless/ath/ath11k/dp.h
@@ -507,6 +507,14 @@ enum htt_ppdu_stats_tag_type {
 				   | BIT(HTT_PPDU_STATS_TAG_USR_COMPLTN_FLUSH) \
 				   | BIT(HTT_PPDU_STATS_TAG_USR_COMMON_ARRAY))
 
+#define HTT_PPDU_STATS_TAG_PKTLOG  (BIT(HTT_PPDU_STATS_TAG_USR_MPDU_ENQ_BITMAP_64) | \
+				    BIT(HTT_PPDU_STATS_TAG_USR_MPDU_ENQ_BITMAP_256) | \
+				    BIT(HTT_PPDU_STATS_TAG_USR_COMPLTN_BA_BITMAP_64) | \
+				    BIT(HTT_PPDU_STATS_TAG_USR_COMPLTN_BA_BITMAP_256) | \
+				    BIT(HTT_PPDU_STATS_TAG_INFO) | \
+				    BIT(HTT_PPDU_STATS_TAG_TX_MGMTCTRL_PAYLOAD) | \
+				    HTT_PPDU_STATS_TAG_DEFAULT)
+
 /* HTT_H2T_MSG_TYPE_RX_RING_SELECTION_CFG Message
  *
  * details:
-- 
cgit v1.2.3-59-g8ed1b


From 5815719dd2714a369448159a562f6fdf75ffe87b Mon Sep 17 00:00:00 2001
From: Bhagavathi Perumal S <bperumal@codeaurora.org>
Date: Tue, 17 Dec 2019 17:49:17 +0100
Subject: ath11k: set TxBf parameters after vdev start

The channel info parameters are required by the firmware to process TxBf
parameters. Currently TxBf is passed prior to the channel info. This patch
moves TxBf setup after the channel setup.

Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Bhagavathi Perumal S <bperumal@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/mac.c | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index 0ed3e4d19f7a..f9ab02e84d13 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -4212,12 +4212,6 @@ static int ath11k_mac_op_add_interface(struct ieee80211_hw *hw,
 			    arvif->vdev_id, ret);
 	}
 
-	ret = ath11k_mac_set_txbf_conf(arvif);
-	if (ret) {
-		ath11k_warn(ar->ab, "failed to set txbf conf for vdev %d: %d\n",
-			    arvif->vdev_id, ret);
-	}
-
 	ath11k_dp_vdev_tx_attach(ar, arvif);
 
 	mutex_unlock(&ar->conf_mutex);
@@ -4567,6 +4561,11 @@ ath11k_mac_vdev_start_restart(struct ath11k_vif *arvif,
 			   arg.channel.freq, arg.vdev_id);
 	}
 
+	ret = ath11k_mac_set_txbf_conf(arvif);
+	if (ret)
+		ath11k_warn(ab, "failed to set txbf conf for vdev %d: %d\n",
+			    arvif->vdev_id, ret);
+
 	return 0;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From eefca584140b42551e642277515ca26be0ca5ef3 Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.king@canonical.com>
Date: Sat, 21 Dec 2019 00:40:46 +0000
Subject: ath11k: ensure ts.flags is initialized before bit-wise or'ing in
 values

Currently the structure ts is not inititalized and ts.flags contains
garbage values from the stack.  This is being passed into function
ath11k_dp_tx_status_parse that bit-wise or'ing in settings into the
ts.flags field.  To avoid flags (and other fields) from containing
garbage, initialize the structure to zero before use.

Addresses-Coverity: ("Uninitialized scalar variable)"
Fixes: d5c65159f289 ("ath11k: driver for Qualcomm IEEE 802.11ax devices")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/dp_tx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp_tx.c b/drivers/net/wireless/ath/ath11k/dp_tx.c
index 918305dda106..04ad1a20e459 100644
--- a/drivers/net/wireless/ath/ath11k/dp_tx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_tx.c
@@ -461,7 +461,7 @@ void ath11k_dp_tx_completion_handler(struct ath11k_base *ab, int ring_id)
 	int hal_ring_id = dp->tx_ring[ring_id].tcl_comp_ring.ring_id;
 	struct hal_srng *status_ring = &ab->hal.srng_list[hal_ring_id];
 	struct sk_buff *msdu;
-	struct hal_tx_status ts;
+	struct hal_tx_status ts = { 0 };
 	struct dp_tx_ring *tx_ring = &dp->tx_ring[ring_id];
 	u32 *desc;
 	u32 msdu_id;
-- 
cgit v1.2.3-59-g8ed1b


From 875603b3de9fb5dc11acf80b6df45709adec68b2 Mon Sep 17 00:00:00 2001
From: Bhagavathi Perumal S <bperumal@codeaurora.org>
Date: Thu, 9 Jan 2020 17:18:50 +0530
Subject: ath11k: Add missing pdev rx rate stats

This adds missing rx rate info stats like pilot evm,
per chain rssi, per user ul ppdu and mpdu counts and
ul ofdma rate info etc.

And add null checks for memory alloc failures.

Signed-off-by: Bhagavathi Perumal S <bperumal@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/debug_htt_stats.c | 205 ++++++++++++++++++----
 drivers/net/wireless/ath/ath11k/debug_htt_stats.h |  42 +++++
 2 files changed, 214 insertions(+), 33 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/debug_htt_stats.c b/drivers/net/wireless/ath/ath11k/debug_htt_stats.c
index 090fffa5e53c..9939e909628f 100644
--- a/drivers/net/wireless/ath/ath11k/debug_htt_stats.c
+++ b/drivers/net/wireless/ath/ath11k/debug_htt_stats.c
@@ -776,11 +776,14 @@ static inline void htt_print_tx_peer_rate_stats_tlv(const void *tag_buf,
 	u32 len = stats_req->buf_len;
 	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
 	char str_buf[HTT_MAX_STRING_LEN] = {0};
-	char *tx_gi[HTT_TX_PEER_STATS_NUM_GI_COUNTERS];
+	char *tx_gi[HTT_TX_PEER_STATS_NUM_GI_COUNTERS] = {NULL};
 	u8 j;
 
-	for (j = 0; j < HTT_TX_PEER_STATS_NUM_GI_COUNTERS; j++)
+	for (j = 0; j < HTT_TX_PEER_STATS_NUM_GI_COUNTERS; j++) {
 		tx_gi[j] = kmalloc(HTT_MAX_STRING_LEN, GFP_ATOMIC);
+		if (!tx_gi[j])
+			goto fail;
+	}
 
 	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_PEER_RATE_STATS_TLV:");
 	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_ldpc = %u",
@@ -841,15 +844,16 @@ static inline void htt_print_tx_peer_rate_stats_tlv(const void *tag_buf,
 			HTT_TX_PDEV_STATS_NUM_DCM_COUNTERS);
 	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_dcm = %s\n", str_buf);
 
-	for (j = 0; j < HTT_TX_PEER_STATS_NUM_GI_COUNTERS; j++)
-		kfree(tx_gi[j]);
-
 	if (len >= buf_len)
 		buf[buf_len - 1] = 0;
 	else
 		buf[len] = 0;
 
 	stats_req->buf_len = len;
+
+fail:
+	for (j = 0; j < HTT_TX_PEER_STATS_NUM_GI_COUNTERS; j++)
+		kfree(tx_gi[j]);
 }
 
 static inline void htt_print_rx_peer_rate_stats_tlv(const void *tag_buf,
@@ -860,15 +864,21 @@ static inline void htt_print_rx_peer_rate_stats_tlv(const void *tag_buf,
 	u32 len = stats_req->buf_len;
 	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
 	u8 j;
-	char *rssi_chain[HTT_RX_PEER_STATS_NUM_SPATIAL_STREAMS];
-	char *rx_gi[HTT_RX_PEER_STATS_NUM_GI_COUNTERS];
+	char *rssi_chain[HTT_RX_PEER_STATS_NUM_SPATIAL_STREAMS] = {NULL};
+	char *rx_gi[HTT_RX_PEER_STATS_NUM_GI_COUNTERS] = {NULL};
 	char str_buf[HTT_MAX_STRING_LEN] = {0};
 
-	for (j = 0; j < HTT_RX_PEER_STATS_NUM_SPATIAL_STREAMS; j++)
+	for (j = 0; j < HTT_RX_PEER_STATS_NUM_SPATIAL_STREAMS; j++) {
 		rssi_chain[j] = kmalloc(HTT_MAX_STRING_LEN, GFP_ATOMIC);
+		if (!rssi_chain[j])
+			goto fail;
+	}
 
-	for (j = 0; j < HTT_RX_PEER_STATS_NUM_GI_COUNTERS; j++)
+	for (j = 0; j < HTT_RX_PEER_STATS_NUM_GI_COUNTERS; j++) {
 		rx_gi[j] = kmalloc(HTT_MAX_STRING_LEN, GFP_ATOMIC);
+		if (!rx_gi[j])
+			goto fail;
+	}
 
 	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_RX_PEER_RATE_STATS_TLV:");
 	len += HTT_DBG_OUT(buf + len, buf_len - len, "nsts = %u",
@@ -928,18 +938,19 @@ static inline void htt_print_rx_peer_rate_stats_tlv(const void *tag_buf,
 			HTT_RX_PDEV_STATS_NUM_PREAMBLE_TYPES);
 	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_pream = %s\n", str_buf);
 
-	for (j = 0; j < HTT_RX_PEER_STATS_NUM_SPATIAL_STREAMS; j++)
-		kfree(rssi_chain[j]);
-
-	for (j = 0; j < HTT_RX_PEER_STATS_NUM_GI_COUNTERS; j++)
-		kfree(rx_gi[j]);
-
 	if (len >= buf_len)
 		buf[buf_len - 1] = 0;
 	else
 		buf[len] = 0;
 
 	stats_req->buf_len = len;
+
+fail:
+	for (j = 0; j < HTT_RX_PEER_STATS_NUM_SPATIAL_STREAMS; j++)
+		kfree(rssi_chain[j]);
+
+	for (j = 0; j < HTT_RX_PEER_STATS_NUM_GI_COUNTERS; j++)
+		kfree(rx_gi[j]);
 }
 
 static inline void
@@ -2832,10 +2843,13 @@ static inline void htt_print_tx_pdev_rate_stats_tlv(const void *tag_buf,
 	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
 	u8 j;
 	char str_buf[HTT_MAX_STRING_LEN] = {0};
-	char *tx_gi[HTT_TX_PEER_STATS_NUM_GI_COUNTERS];
+	char *tx_gi[HTT_TX_PEER_STATS_NUM_GI_COUNTERS] = {NULL};
 
-	for (j = 0; j < HTT_TX_PEER_STATS_NUM_GI_COUNTERS; j++)
+	for (j = 0; j < HTT_TX_PEER_STATS_NUM_GI_COUNTERS; j++) {
 		tx_gi[j] = kmalloc(HTT_MAX_STRING_LEN, GFP_ATOMIC);
+		if (!tx_gi[j])
+			goto fail;
+	}
 
 	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_TX_PDEV_RATE_STATS_TLV:");
 	len += HTT_DBG_OUT(buf + len, buf_len - len, "mac_id = %u",
@@ -2988,15 +3002,15 @@ static inline void htt_print_tx_pdev_rate_stats_tlv(const void *tag_buf,
 			HTT_TX_PDEV_STATS_NUM_DCM_COUNTERS);
 	len += HTT_DBG_OUT(buf + len, buf_len - len, "tx_dcm = %s\n", str_buf);
 
-	for (j = 0; j < HTT_TX_PEER_STATS_NUM_GI_COUNTERS; j++)
-		kfree(tx_gi[j]);
-
 	if (len >= buf_len)
 		buf[buf_len - 1] = 0;
 	else
 		buf[len] = 0;
 
 	stats_req->buf_len = len;
+fail:
+	for (j = 0; j < HTT_TX_PEER_STATS_NUM_GI_COUNTERS; j++)
+		kfree(tx_gi[j]);
 }
 
 static inline void htt_print_rx_pdev_rate_stats_tlv(const void *tag_buf,
@@ -3006,16 +3020,30 @@ static inline void htt_print_rx_pdev_rate_stats_tlv(const void *tag_buf,
 	u8 *buf = stats_req->buf;
 	u32 len = stats_req->buf_len;
 	u32 buf_len = ATH11K_HTT_STATS_BUF_SIZE;
-	u8 j;
-	char *rssi_chain[HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS];
-	char *rx_gi[HTT_RX_PDEV_STATS_NUM_GI_COUNTERS];
+	u8 i, j;
+	u16 index = 0;
+	char *rssi_chain[HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS] = {NULL};
+	char *rx_gi[HTT_RX_PDEV_STATS_NUM_GI_COUNTERS] = {NULL};
 	char str_buf[HTT_MAX_STRING_LEN] = {0};
+	char *rx_pilot_evm_db[HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS] = {NULL};
 
-	for (j = 0; j < HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS; j++)
+	for (j = 0; j < HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS; j++) {
 		rssi_chain[j] = kmalloc(HTT_MAX_STRING_LEN, GFP_ATOMIC);
+		if (!rssi_chain[j])
+			goto fail;
+	}
 
-	for (j = 0; j < HTT_RX_PDEV_STATS_NUM_GI_COUNTERS; j++)
+	for (j = 0; j < HTT_RX_PDEV_STATS_NUM_GI_COUNTERS; j++) {
 		rx_gi[j] = kmalloc(HTT_MAX_STRING_LEN, GFP_ATOMIC);
+		if (!rx_gi[j])
+			goto fail;
+	}
+
+	for (j = 0; j < HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS; j++) {
+		rx_pilot_evm_db[j] = kmalloc(HTT_MAX_STRING_LEN, GFP_ATOMIC);
+		if (!rx_pilot_evm_db[j])
+			goto fail;
+	}
 
 	len += HTT_DBG_OUT(buf + len, buf_len - len, "HTT_RX_PDEV_RATE_STATS_TLV:");
 	len += HTT_DBG_OUT(buf + len, buf_len - len, "mac_id = %u",
@@ -3059,6 +3087,32 @@ static inline void htt_print_rx_pdev_rate_stats_tlv(const void *tag_buf,
 	ARRAY_TO_STRING(str_buf, htt_stats_buf->rx_bw,
 			HTT_RX_PDEV_STATS_NUM_BW_COUNTERS);
 	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_bw = %s ", str_buf);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_evm_nss_count = %u",
+			htt_stats_buf->nss_count);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_evm_pilot_count = %u",
+			htt_stats_buf->pilot_count);
+
+	for (j = 0; j < HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS; j++) {
+		index = 0;
+
+		for (i = 0; i < HTT_RX_PDEV_STATS_RXEVM_MAX_PILOTS_PER_NSS; i++)
+			index += snprintf(&rx_pilot_evm_db[j][index],
+					  HTT_MAX_STRING_LEN - index,
+					  " %u:%d,",
+					  i,
+					  htt_stats_buf->rx_pilot_evm_db[j][i]);
+		len += HTT_DBG_OUT(buf + len, buf_len - len, "pilot_evm_dB[%u] = %s ",
+				   j, rx_pilot_evm_db[j]);
+	}
+
+	index = 0;
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	for (i = 0; i < HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS; i++)
+		index += snprintf(&str_buf[index],
+				  HTT_MAX_STRING_LEN - index,
+				  " %u:%d,", i, htt_stats_buf->rx_pilot_evm_db_mean[i]);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "pilot_evm_dB_mean = %s ", str_buf);
 
 	for (j = 0; j < HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS; j++) {
 		ARRAY_TO_STRING(rssi_chain[j], htt_stats_buf->rssi_chain[j],
@@ -3079,12 +3133,6 @@ static inline void htt_print_rx_pdev_rate_stats_tlv(const void *tag_buf,
 			HTT_RX_PDEV_STATS_NUM_PREAMBLE_TYPES);
 	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_pream = %s", str_buf);
 
-	for (j = 0; j < HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS; j++)
-		kfree(rssi_chain[j]);
-
-	for (j = 0; j < HTT_RX_PDEV_STATS_NUM_GI_COUNTERS; j++)
-		kfree(rx_gi[j]);
-
 	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_11ax_su_ext = %u",
 			   htt_stats_buf->rx_11ax_su_ext);
 	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_11ac_mumimo = %u",
@@ -3110,8 +3158,89 @@ static inline void htt_print_rx_pdev_rate_stats_tlv(const void *tag_buf,
 
 	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_active_dur_us_low = %u",
 			   htt_stats_buf->rx_active_dur_us_low);
-	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_active_dur_us_high = %u\n",
-			   htt_stats_buf->rx_active_dur_us_high);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_active_dur_us_high = %u",
+			htt_stats_buf->rx_active_dur_us_high);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_11ax_ul_ofdma = %u",
+			htt_stats_buf->rx_11ax_ul_ofdma);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->ul_ofdma_rx_mcs,
+			HTT_RX_PDEV_STATS_NUM_MCS_COUNTERS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ul_ofdma_rx_mcs = %s ", str_buf);
+
+	for (j = 0; j < HTT_RX_PDEV_STATS_NUM_GI_COUNTERS; j++) {
+		ARRAY_TO_STRING(rx_gi[j], htt_stats_buf->ul_ofdma_rx_gi[j],
+				HTT_RX_PDEV_STATS_NUM_MCS_COUNTERS);
+		len += HTT_DBG_OUT(buf + len, buf_len - len, "ul_ofdma_rx_gi[%u] = %s ",
+				   j, rx_gi[j]);
+	}
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->ul_ofdma_rx_nss,
+			HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ul_ofdma_rx_nss = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->ul_ofdma_rx_bw,
+			HTT_RX_PDEV_STATS_NUM_BW_COUNTERS);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ul_ofdma_rx_bw = %s ", str_buf);
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ul_ofdma_rx_stbc = %u",
+			htt_stats_buf->ul_ofdma_rx_stbc);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "ul_ofdma_rx_ldpc = %u",
+			htt_stats_buf->ul_ofdma_rx_ldpc);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->rx_ulofdma_non_data_ppdu,
+			HTT_RX_PDEV_MAX_OFDMA_NUM_USER);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_ulofdma_non_data_ppdu = %s ",
+			   str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->rx_ulofdma_data_ppdu,
+			HTT_RX_PDEV_MAX_OFDMA_NUM_USER);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_ulofdma_data_ppdu = %s ",
+			   str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->rx_ulofdma_mpdu_ok,
+			HTT_RX_PDEV_MAX_OFDMA_NUM_USER);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_ulofdma_mpdu_ok = %s ", str_buf);
+
+	memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+	ARRAY_TO_STRING(str_buf, htt_stats_buf->rx_ulofdma_mpdu_fail,
+			HTT_RX_PDEV_MAX_OFDMA_NUM_USER);
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "rx_ulofdma_mpdu_fail = %s",
+			   str_buf);
+
+	for (j = 0; j < HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS; j++) {
+		index = 0;
+		memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+		for (i = 0; i < HTT_RX_PDEV_MAX_OFDMA_NUM_USER; i++)
+			index += snprintf(&str_buf[index],
+					  HTT_MAX_STRING_LEN - index,
+					  " %u:%d,",
+					  i, htt_stats_buf->rx_ul_fd_rssi[j][i]);
+		len += HTT_DBG_OUT(buf + len, buf_len - len,
+				   "rx_ul_fd_rssi: nss[%u] = %s", j, str_buf);
+	}
+
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "per_chain_rssi_pkt_type = %#x",
+			   htt_stats_buf->per_chain_rssi_pkt_type);
+
+	for (j = 0; j < HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS; j++) {
+		index = 0;
+		memset(str_buf, 0x0, HTT_MAX_STRING_LEN);
+		for (i = 0; i < HTT_RX_PDEV_STATS_NUM_BW_COUNTERS; i++)
+			index += snprintf(&str_buf[index],
+					  HTT_MAX_STRING_LEN - index,
+					  " %u:%d,",
+					  i,
+					  htt_stats_buf->rx_per_chain_rssi_in_dbm[j][i]);
+		len += HTT_DBG_OUT(buf + len, buf_len - len,
+				   "rx_per_chain_rssi_in_dbm[%u] = %s ", j, str_buf);
+	}
+	len += HTT_DBG_OUT(buf + len, buf_len - len, "\n");
 
 	if (len >= buf_len)
 		buf[buf_len - 1] = 0;
@@ -3119,6 +3248,16 @@ static inline void htt_print_rx_pdev_rate_stats_tlv(const void *tag_buf,
 		buf[len] = 0;
 
 	stats_req->buf_len = len;
+
+fail:
+	for (j = 0; j < HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS; j++)
+		kfree(rssi_chain[j]);
+
+	for (j = 0; j < HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS; j++)
+		kfree(rx_pilot_evm_db[j]);
+
+	for (i = 0; i < HTT_RX_PDEV_STATS_NUM_GI_COUNTERS; i++)
+		kfree(rx_gi[i]);
 }
 
 static inline void htt_print_rx_soc_fw_stats_tlv(const void *tag_buf,
diff --git a/drivers/net/wireless/ath/ath11k/debug_htt_stats.h b/drivers/net/wireless/ath/ath11k/debug_htt_stats.h
index 618f1946bf49..4bdb62dd7b8d 100644
--- a/drivers/net/wireless/ath/ath11k/debug_htt_stats.h
+++ b/drivers/net/wireless/ath/ath11k/debug_htt_stats.h
@@ -1231,6 +1231,8 @@ struct htt_tx_pdev_rate_stats_tlv {
 #define HTT_RX_PDEV_STATS_NUM_BW_COUNTERS          4
 #define HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS      8
 #define HTT_RX_PDEV_STATS_NUM_PREAMBLE_TYPES       HTT_STATS_PREAM_COUNT
+#define HTT_RX_PDEV_MAX_OFDMA_NUM_USER             8
+#define HTT_RX_PDEV_STATS_RXEVM_MAX_PILOTS_PER_NSS 16
 
 struct htt_rx_pdev_rate_stats_tlv {
 	u32 mac_id__word;
@@ -1269,6 +1271,46 @@ struct htt_rx_pdev_rate_stats_tlv {
 	u32 rx_legacy_ofdm_rate[HTT_RX_PDEV_STATS_NUM_LEGACY_OFDM_STATS];
 	u32 rx_active_dur_us_low;
 	u32 rx_active_dur_us_high;
+
+	u32 rx_11ax_ul_ofdma;
+
+	u32 ul_ofdma_rx_mcs[HTT_RX_PDEV_STATS_NUM_MCS_COUNTERS];
+	u32 ul_ofdma_rx_gi[HTT_TX_PDEV_STATS_NUM_GI_COUNTERS]
+			  [HTT_RX_PDEV_STATS_NUM_MCS_COUNTERS];
+	u32 ul_ofdma_rx_nss[HTT_TX_PDEV_STATS_NUM_SPATIAL_STREAMS];
+	u32 ul_ofdma_rx_bw[HTT_TX_PDEV_STATS_NUM_BW_COUNTERS];
+	u32 ul_ofdma_rx_stbc;
+	u32 ul_ofdma_rx_ldpc;
+
+	/* record the stats for each user index */
+	u32 rx_ulofdma_non_data_ppdu[HTT_RX_PDEV_MAX_OFDMA_NUM_USER]; /* ppdu level */
+	u32 rx_ulofdma_data_ppdu[HTT_RX_PDEV_MAX_OFDMA_NUM_USER];     /* ppdu level */
+	u32 rx_ulofdma_mpdu_ok[HTT_RX_PDEV_MAX_OFDMA_NUM_USER];       /* mpdu level */
+	u32 rx_ulofdma_mpdu_fail[HTT_RX_PDEV_MAX_OFDMA_NUM_USER];     /* mpdu level */
+
+	u32 nss_count;
+	u32 pilot_count;
+	/* RxEVM stats in dB */
+	s32 rx_pilot_evm_db[HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS]
+			   [HTT_RX_PDEV_STATS_RXEVM_MAX_PILOTS_PER_NSS];
+	/* rx_pilot_evm_db_mean:
+	 * EVM mean across pilots, computed as
+	 *     mean(10*log10(rx_pilot_evm_linear)) = mean(rx_pilot_evm_db)
+	 */
+	s32 rx_pilot_evm_db_mean[HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS];
+	s8 rx_ul_fd_rssi[HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS]
+			[HTT_RX_PDEV_MAX_OFDMA_NUM_USER]; /* dBm units */
+	/* per_chain_rssi_pkt_type:
+	 * This field shows what type of rx frame the per-chain RSSI was computed
+	 * on, by recording the frame type and sub-type as bit-fields within this
+	 * field:
+	 * BIT [3 : 0]    :- IEEE80211_FC0_TYPE
+	 * BIT [7 : 4]    :- IEEE80211_FC0_SUBTYPE
+	 * BIT [31 : 8]   :- Reserved
+	 */
+	u32 per_chain_rssi_pkt_type;
+	s8 rx_per_chain_rssi_in_dbm[HTT_RX_PDEV_STATS_NUM_SPATIAL_STREAMS]
+				   [HTT_RX_PDEV_STATS_NUM_BW_COUNTERS];
 };
 
 /* == RX PDEV/SOC STATS == */
-- 
cgit v1.2.3-59-g8ed1b


From 3b4516838eaabacc56770f12407d5f791c88838a Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.king@canonical.com>
Date: Sat, 11 Jan 2020 09:08:24 +0000
Subject: ath11k: avoid null pointer dereference when pointer band is null

In the unlikely event that cap->supported_bands has neither
WMI_HOST_WLAN_2G_CAP set or WMI_HOST_WLAN_5G_CAP set then pointer
band is null and a null dereference occurs when assigning
band->n_iftype_data.  Move the assignment to the if blocks to
avoid this.  Cleans up static analysis warnings.

Addresses-Coverity: ("Explicit null dereference")
Fixes: 9f056ed8ee01 ("ath11k: add HE support")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/mac.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index f9ab02e84d13..6640662f5ede 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -3520,8 +3520,8 @@ static int ath11k_mac_copy_he_cap(struct ath11k *ar,
 static void ath11k_mac_setup_he_cap(struct ath11k *ar,
 				    struct ath11k_pdev_cap *cap)
 {
-	struct ieee80211_supported_band *band = NULL;
-	int count = 0;
+	struct ieee80211_supported_band *band;
+	int count;
 
 	if (cap->supported_bands & WMI_HOST_WLAN_2G_CAP) {
 		count = ath11k_mac_copy_he_cap(ar, cap,
@@ -3529,6 +3529,7 @@ static void ath11k_mac_setup_he_cap(struct ath11k *ar,
 					       NL80211_BAND_2GHZ);
 		band = &ar->mac.sbands[NL80211_BAND_2GHZ];
 		band->iftype_data = ar->mac.iftype[NL80211_BAND_2GHZ];
+		band->n_iftype_data = count;
 	}
 
 	if (cap->supported_bands & WMI_HOST_WLAN_5G_CAP) {
@@ -3537,9 +3538,8 @@ static void ath11k_mac_setup_he_cap(struct ath11k *ar,
 					       NL80211_BAND_5GHZ);
 		band = &ar->mac.sbands[NL80211_BAND_5GHZ];
 		band->iftype_data = ar->mac.iftype[NL80211_BAND_5GHZ];
+		band->n_iftype_data = count;
 	}
-
-	band->n_iftype_data = count;
 }
 
 static int __ath11k_set_antenna(struct ath11k *ar, u32 tx_ant, u32 rx_ant)
-- 
cgit v1.2.3-59-g8ed1b


From 93634c61056e34522e824d56ff77ee2332ed7f8a Mon Sep 17 00:00:00 2001
From: John Crispin <john@phrozen.org>
Date: Tue, 17 Dec 2019 21:20:57 +0100
Subject: ath11k: make sure to also report the RX bandwidth inside radiotap

Add IEEE80211_RADIOTAP_HE_DATA1_BW_RU_ALLOC_KNOWN to the list of known
fields. Not doing so will result in wireshark not calculating the
bitrate correctly.

Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/dp_rx.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c b/drivers/net/wireless/ath/ath11k/dp_rx.c
index 3a3dc7680622..b08da839b7d9 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
@@ -2067,7 +2067,8 @@ static void ath11k_dp_rx_deliver_msdu(struct ath11k *ar, struct napi_struct *nap
 				      struct sk_buff *msdu)
 {
 	static const struct ieee80211_radiotap_he known = {
-		.data1 = cpu_to_le16(IEEE80211_RADIOTAP_HE_DATA1_DATA_MCS_KNOWN),
+		.data1 = cpu_to_le16(IEEE80211_RADIOTAP_HE_DATA1_DATA_MCS_KNOWN |
+				     IEEE80211_RADIOTAP_HE_DATA1_BW_RU_ALLOC_KNOWN),
 		.data2 = cpu_to_le16(IEEE80211_RADIOTAP_HE_DATA2_GI_KNOWN),
 	};
 	struct ieee80211_rx_status *status;
-- 
cgit v1.2.3-59-g8ed1b


From 7195c8747e8757daaf48d862e1314f8bcfdb3207 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@oracle.com>
Date: Wed, 15 Jan 2020 20:46:52 +0300
Subject: ath11k: fix up some error paths

There are two error paths where "ret" wasn't set.  Also one error path
we set the error code to -EINVAL but we should just preserve the error
code from ath11k_hal_srng_get_entrysize().  That function only returns
-EINVAL so this doesn't change anything.

I removed the "ret = 0;" initializers so that hopefully GCC will be able
to detect these sorts of bugs in the future.

Fixes: d5c65159f289 ("ath11k: driver for Qualcomm IEEE 802.11ax devices")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath11k/dp_tx.c | 22 +++++++++++-----------
 1 file changed, 11 insertions(+), 11 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp_tx.c b/drivers/net/wireless/ath/ath11k/dp_tx.c
index 04ad1a20e459..6d7d33761caf 100644
--- a/drivers/net/wireless/ath/ath11k/dp_tx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_tx.c
@@ -630,7 +630,7 @@ int ath11k_dp_tx_htt_srng_setup(struct ath11k_base *ab, u32 ring_id,
 	dma_addr_t hp_addr, tp_addr;
 	enum htt_srng_ring_type htt_ring_type;
 	enum htt_srng_ring_id htt_ring_id;
-	int ret = 0;
+	int ret;
 
 	skb = ath11k_htc_alloc_skb(ab, len);
 	if (!skb)
@@ -642,9 +642,10 @@ int ath11k_dp_tx_htt_srng_setup(struct ath11k_base *ab, u32 ring_id,
 	hp_addr = ath11k_hal_srng_get_hp_addr(ab, srng);
 	tp_addr = ath11k_hal_srng_get_tp_addr(ab, srng);
 
-	if (ath11k_dp_tx_get_ring_id_type(ab, mac_id, ring_id,
-					  ring_type, &htt_ring_type,
-					  &htt_ring_id))
+	ret = ath11k_dp_tx_get_ring_id_type(ab, mac_id, ring_id,
+					    ring_type, &htt_ring_type,
+					    &htt_ring_id);
+	if (ret)
 		goto err_free;
 
 	skb_put(skb, len);
@@ -669,10 +670,8 @@ int ath11k_dp_tx_htt_srng_setup(struct ath11k_base *ab, u32 ring_id,
 				 HAL_ADDR_MSB_REG_SHIFT;
 
 	ret = ath11k_hal_srng_get_entrysize(ring_type);
-	if (ret < 0) {
-		ret = -EINVAL;
+	if (ret < 0)
 		goto err_free;
-	}
 
 	ring_entry_sz = ret;
 
@@ -817,7 +816,7 @@ int ath11k_dp_tx_htt_rx_filter_setup(struct ath11k_base *ab, u32 ring_id,
 	int len = sizeof(*cmd);
 	enum htt_srng_ring_type htt_ring_type;
 	enum htt_srng_ring_id htt_ring_id;
-	int ret = 0;
+	int ret;
 
 	skb = ath11k_htc_alloc_skb(ab, len);
 	if (!skb)
@@ -826,9 +825,10 @@ int ath11k_dp_tx_htt_rx_filter_setup(struct ath11k_base *ab, u32 ring_id,
 	memset(&params, 0, sizeof(params));
 	ath11k_hal_srng_get_params(ab, srng, &params);
 
-	if (ath11k_dp_tx_get_ring_id_type(ab, mac_id, ring_id,
-					  ring_type, &htt_ring_type,
-					  &htt_ring_id))
+	ret = ath11k_dp_tx_get_ring_id_type(ab, mac_id, ring_id,
+					    ring_type, &htt_ring_type,
+					    &htt_ring_id);
+	if (ret)
 		goto err_free;
 
 	skb_put(skb, len);
-- 
cgit v1.2.3-59-g8ed1b


From 70e5b8f445fd27fde0c5583460e82539a7242424 Mon Sep 17 00:00:00 2001
From: Brian Norris <briannorris@chromium.org>
Date: Fri, 6 Dec 2019 11:45:35 -0800
Subject: mwifiex: drop most magic numbers from
 mwifiex_process_tdls_action_frame()

Before commit 1e58252e334d ("mwifiex: Fix heap overflow in
mmwifiex_process_tdls_action_frame()"),
mwifiex_process_tdls_action_frame() already had too many magic numbers.
But this commit just added a ton more, in the name of checking for
buffer overflows. That seems like a really bad idea.

Let's make these magic numbers a little less magic, by
(a) factoring out 'pos[1]' as 'ie_len'
(b) using 'sizeof' on the appropriate source or destination fields where
    possible, instead of bare numbers
(c) dropping redundant checks, per below.

Regarding redundant checks: the beginning of the loop has this:

                if (pos + 2 + pos[1] > end)
                        break;

but then individual 'case's include stuff like this:

 			if (pos > end - 3)
 				return;
 			if (pos[1] != 1)
				return;

Note that the second 'return' (validating the length, pos[1]) combined
with the above condition (ensuring 'pos + 2 + length' doesn't exceed
'end'), makes the first 'return' (whose 'if' can be reworded as 'pos >
end - pos[1] - 2') redundant. Rather than unwind the magic numbers
there, just drop those conditions.

Fixes: 1e58252e334d ("mwifiex: Fix heap overflow in mmwifiex_process_tdls_action_frame()")
Signed-off-by: Brian Norris <briannorris@chromium.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/marvell/mwifiex/tdls.c | 75 +++++++++++------------------
 1 file changed, 28 insertions(+), 47 deletions(-)

diff --git a/drivers/net/wireless/marvell/mwifiex/tdls.c b/drivers/net/wireless/marvell/mwifiex/tdls.c
index 7caf1d26124a..f8f282ce39bd 100644
--- a/drivers/net/wireless/marvell/mwifiex/tdls.c
+++ b/drivers/net/wireless/marvell/mwifiex/tdls.c
@@ -894,7 +894,7 @@ void mwifiex_process_tdls_action_frame(struct mwifiex_private *priv,
 	u8 *peer, *pos, *end;
 	u8 i, action, basic;
 	u16 cap = 0;
-	int ie_len = 0;
+	int ies_len = 0;
 
 	if (len < (sizeof(struct ethhdr) + 3))
 		return;
@@ -916,7 +916,7 @@ void mwifiex_process_tdls_action_frame(struct mwifiex_private *priv,
 		pos = buf + sizeof(struct ethhdr) + 4;
 		/* payload 1+ category 1 + action 1 + dialog 1 */
 		cap = get_unaligned_le16(pos);
-		ie_len = len - sizeof(struct ethhdr) - TDLS_REQ_FIX_LEN;
+		ies_len = len - sizeof(struct ethhdr) - TDLS_REQ_FIX_LEN;
 		pos += 2;
 		break;
 
@@ -926,7 +926,7 @@ void mwifiex_process_tdls_action_frame(struct mwifiex_private *priv,
 		/* payload 1+ category 1 + action 1 + dialog 1 + status code 2*/
 		pos = buf + sizeof(struct ethhdr) + 6;
 		cap = get_unaligned_le16(pos);
-		ie_len = len - sizeof(struct ethhdr) - TDLS_RESP_FIX_LEN;
+		ies_len = len - sizeof(struct ethhdr) - TDLS_RESP_FIX_LEN;
 		pos += 2;
 		break;
 
@@ -934,7 +934,7 @@ void mwifiex_process_tdls_action_frame(struct mwifiex_private *priv,
 		if (len < (sizeof(struct ethhdr) + TDLS_CONFIRM_FIX_LEN))
 			return;
 		pos = buf + sizeof(struct ethhdr) + TDLS_CONFIRM_FIX_LEN;
-		ie_len = len - sizeof(struct ethhdr) - TDLS_CONFIRM_FIX_LEN;
+		ies_len = len - sizeof(struct ethhdr) - TDLS_CONFIRM_FIX_LEN;
 		break;
 	default:
 		mwifiex_dbg(priv->adapter, ERROR, "Unknown TDLS frame type.\n");
@@ -947,33 +947,33 @@ void mwifiex_process_tdls_action_frame(struct mwifiex_private *priv,
 
 	sta_ptr->tdls_cap.capab = cpu_to_le16(cap);
 
-	for (end = pos + ie_len; pos + 1 < end; pos += 2 + pos[1]) {
-		if (pos + 2 + pos[1] > end)
+	for (end = pos + ies_len; pos + 1 < end; pos += 2 + pos[1]) {
+		u8 ie_len = pos[1];
+
+		if (pos + 2 + ie_len > end)
 			break;
 
 		switch (*pos) {
 		case WLAN_EID_SUPP_RATES:
-			if (pos[1] > 32)
+			if (ie_len > sizeof(sta_ptr->tdls_cap.rates))
 				return;
-			sta_ptr->tdls_cap.rates_len = pos[1];
-			for (i = 0; i < pos[1]; i++)
+			sta_ptr->tdls_cap.rates_len = ie_len;
+			for (i = 0; i < ie_len; i++)
 				sta_ptr->tdls_cap.rates[i] = pos[i + 2];
 			break;
 
 		case WLAN_EID_EXT_SUPP_RATES:
-			if (pos[1] > 32)
+			if (ie_len > sizeof(sta_ptr->tdls_cap.rates))
 				return;
 			basic = sta_ptr->tdls_cap.rates_len;
-			if (pos[1] > 32 - basic)
+			if (ie_len > sizeof(sta_ptr->tdls_cap.rates) - basic)
 				return;
-			for (i = 0; i < pos[1]; i++)
+			for (i = 0; i < ie_len; i++)
 				sta_ptr->tdls_cap.rates[basic + i] = pos[i + 2];
-			sta_ptr->tdls_cap.rates_len += pos[1];
+			sta_ptr->tdls_cap.rates_len += ie_len;
 			break;
 		case WLAN_EID_HT_CAPABILITY:
-			if (pos > end - sizeof(struct ieee80211_ht_cap) - 2)
-				return;
-			if (pos[1] != sizeof(struct ieee80211_ht_cap))
+			if (ie_len != sizeof(struct ieee80211_ht_cap))
 				return;
 			/* copy the ie's value into ht_capb*/
 			memcpy((u8 *)&sta_ptr->tdls_cap.ht_capb, pos + 2,
@@ -981,59 +981,45 @@ void mwifiex_process_tdls_action_frame(struct mwifiex_private *priv,
 			sta_ptr->is_11n_enabled = 1;
 			break;
 		case WLAN_EID_HT_OPERATION:
-			if (pos > end -
-			    sizeof(struct ieee80211_ht_operation) - 2)
-				return;
-			if (pos[1] != sizeof(struct ieee80211_ht_operation))
+			if (ie_len != sizeof(struct ieee80211_ht_operation))
 				return;
 			/* copy the ie's value into ht_oper*/
 			memcpy(&sta_ptr->tdls_cap.ht_oper, pos + 2,
 			       sizeof(struct ieee80211_ht_operation));
 			break;
 		case WLAN_EID_BSS_COEX_2040:
-			if (pos > end - 3)
-				return;
-			if (pos[1] != 1)
+			if (ie_len != sizeof(pos[2]))
 				return;
 			sta_ptr->tdls_cap.coex_2040 = pos[2];
 			break;
 		case WLAN_EID_EXT_CAPABILITY:
-			if (pos > end - sizeof(struct ieee_types_header))
-				return;
-			if (pos[1] < sizeof(struct ieee_types_header))
+			if (ie_len < sizeof(struct ieee_types_header))
 				return;
-			if (pos[1] > 8)
+			if (ie_len > 8)
 				return;
 			memcpy((u8 *)&sta_ptr->tdls_cap.extcap, pos,
 			       sizeof(struct ieee_types_header) +
-			       min_t(u8, pos[1], 8));
+			       min_t(u8, ie_len, 8));
 			break;
 		case WLAN_EID_RSN:
-			if (pos > end - sizeof(struct ieee_types_header))
+			if (ie_len < sizeof(struct ieee_types_header))
 				return;
-			if (pos[1] < sizeof(struct ieee_types_header))
-				return;
-			if (pos[1] > IEEE_MAX_IE_SIZE -
+			if (ie_len > IEEE_MAX_IE_SIZE -
 			    sizeof(struct ieee_types_header))
 				return;
 			memcpy((u8 *)&sta_ptr->tdls_cap.rsn_ie, pos,
 			       sizeof(struct ieee_types_header) +
-			       min_t(u8, pos[1], IEEE_MAX_IE_SIZE -
+			       min_t(u8, ie_len, IEEE_MAX_IE_SIZE -
 				     sizeof(struct ieee_types_header)));
 			break;
 		case WLAN_EID_QOS_CAPA:
-			if (pos > end - 3)
-				return;
-			if (pos[1] != 1)
+			if (ie_len != sizeof(pos[2]))
 				return;
 			sta_ptr->tdls_cap.qos_info = pos[2];
 			break;
 		case WLAN_EID_VHT_OPERATION:
 			if (priv->adapter->is_hw_11ac_capable) {
-				if (pos > end -
-				    sizeof(struct ieee80211_vht_operation) - 2)
-					return;
-				if (pos[1] !=
+				if (ie_len !=
 				    sizeof(struct ieee80211_vht_operation))
 					return;
 				/* copy the ie's value into vhtoper*/
@@ -1043,10 +1029,7 @@ void mwifiex_process_tdls_action_frame(struct mwifiex_private *priv,
 			break;
 		case WLAN_EID_VHT_CAPABILITY:
 			if (priv->adapter->is_hw_11ac_capable) {
-				if (pos > end -
-				    sizeof(struct ieee80211_vht_cap) - 2)
-					return;
-				if (pos[1] != sizeof(struct ieee80211_vht_cap))
+				if (ie_len != sizeof(struct ieee80211_vht_cap))
 					return;
 				/* copy the ie's value into vhtcap*/
 				memcpy((u8 *)&sta_ptr->tdls_cap.vhtcap, pos + 2,
@@ -1056,9 +1039,7 @@ void mwifiex_process_tdls_action_frame(struct mwifiex_private *priv,
 			break;
 		case WLAN_EID_AID:
 			if (priv->adapter->is_hw_11ac_capable) {
-				if (pos > end - 4)
-					return;
-				if (pos[1] != 2)
+				if (ie_len != sizeof(u16))
 					return;
 				sta_ptr->tdls_cap.aid =
 					get_unaligned_le16((pos + 2));
-- 
cgit v1.2.3-59-g8ed1b


From d04bf42891a7ea43d15384262ae6ddd8e0461642 Mon Sep 17 00:00:00 2001
From: Ganapathi Bhat <ganapathi.bhat@nxp.com>
Date: Mon, 13 Jan 2020 10:42:46 +0000
Subject: MAINTAINERS: update for mwifiex driver maintainers

Remove Nishant Sarmukadam from Maintainer list, as he is no
longer working in NXP.

Signed-off-by: Ganapathi Bhat <ganapathi.bhat@nxp.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 MAINTAINERS | 1 -
 1 file changed, 1 deletion(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index 2549f10eb0b1..c2979c6893c5 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -9960,7 +9960,6 @@ F:	drivers/net/ethernet/marvell/mvneta.*
 
 MARVELL MWIFIEX WIRELESS DRIVER
 M:	Amitkumar Karwar <amitkarwar@gmail.com>
-M:	Nishant Sarmukadam <nishants@marvell.com>
 M:	Ganapathi Bhat <ganapathi.bhat@nxp.com>
 M:	Xinming Hu <huxinming820@gmail.com>
 L:	linux-wireless@vger.kernel.org
-- 
cgit v1.2.3-59-g8ed1b


From 8ca769e2b88f9e564aecdfab54fcc5a9d20190c1 Mon Sep 17 00:00:00 2001
From: Julian Wiedmann <jwi@linux.ibm.com>
Date: Sat, 25 Jan 2020 16:52:59 +0100
Subject: s390/qeth: shift some bridgeport code around

qeth_l2_setup_bridgeport_attrs() is entirely unrelated to sysfs
functionality, move it where it belongs.
While at it merge all the bridgeport-specific code in the set-online
path together.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Reviewed-by: Alexandra Winter <wintera@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/s390/net/qeth_l2.h      |  1 -
 drivers/s390/net/qeth_l2_main.c | 26 ++++++++++++++++++++++----
 drivers/s390/net/qeth_l2_sys.c  | 34 ----------------------------------
 3 files changed, 22 insertions(+), 39 deletions(-)

diff --git a/drivers/s390/net/qeth_l2.h b/drivers/s390/net/qeth_l2.h
index ddc615b431a8..adf25c9fd2b3 100644
--- a/drivers/s390/net/qeth_l2.h
+++ b/drivers/s390/net/qeth_l2.h
@@ -13,7 +13,6 @@ extern const struct attribute_group *qeth_l2_attr_groups[];
 
 int qeth_l2_create_device_attributes(struct device *);
 void qeth_l2_remove_device_attributes(struct device *);
-void qeth_l2_setup_bridgeport_attrs(struct qeth_card *card);
 int qeth_bridgeport_query_ports(struct qeth_card *card,
 				enum qeth_sbp_roles *role,
 				enum qeth_sbp_states *state);
diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c
index 7175b5d8a23c..7da306e267c9 100644
--- a/drivers/s390/net/qeth_l2_main.c
+++ b/drivers/s390/net/qeth_l2_main.c
@@ -728,6 +728,24 @@ static void qeth_l2_trace_features(struct qeth_card *card)
 		      sizeof(card->options.vnicc.sup_chars));
 }
 
+static void qeth_l2_setup_bridgeport_attrs(struct qeth_card *card)
+{
+	if (!card->options.sbp.reflect_promisc &&
+	    card->options.sbp.role != QETH_SBP_ROLE_NONE) {
+		/* Conditional to avoid spurious error messages */
+		qeth_bridgeport_setrole(card, card->options.sbp.role);
+		/* Let the callback function refresh the stored role value. */
+		qeth_bridgeport_query_ports(card, &card->options.sbp.role,
+					    NULL);
+	}
+	if (card->options.sbp.hostnotification) {
+		if (qeth_bridgeport_an_set(card, 1))
+			card->options.sbp.hostnotification = 0;
+	} else {
+		qeth_bridgeport_an_set(card, 0);
+	}
+}
+
 static int qeth_l2_set_online(struct ccwgroup_device *gdev)
 {
 	struct qeth_card *card = dev_get_drvdata(&gdev->dev);
@@ -748,9 +766,11 @@ static int qeth_l2_set_online(struct ccwgroup_device *gdev)
 
 	mutex_lock(&card->sbp_lock);
 	qeth_bridgeport_query_support(card);
-	if (card->options.sbp.supported_funcs)
+	if (card->options.sbp.supported_funcs) {
+		qeth_l2_setup_bridgeport_attrs(card);
 		dev_info(&card->gdev->dev,
-		"The device represents a Bridge Capable Port\n");
+			 "The device represents a Bridge Capable Port\n");
+	}
 	mutex_unlock(&card->sbp_lock);
 
 	qeth_l2_register_dev_addr(card);
@@ -761,8 +781,6 @@ static int qeth_l2_set_online(struct ccwgroup_device *gdev)
 	qeth_trace_features(card);
 	qeth_l2_trace_features(card);
 
-	qeth_l2_setup_bridgeport_attrs(card);
-
 	card->state = CARD_STATE_HARDSETUP;
 	qeth_print_status_message(card);
 
diff --git a/drivers/s390/net/qeth_l2_sys.c b/drivers/s390/net/qeth_l2_sys.c
index 7fa325cf6f8d..86bcae992f72 100644
--- a/drivers/s390/net/qeth_l2_sys.c
+++ b/drivers/s390/net/qeth_l2_sys.c
@@ -246,40 +246,6 @@ static struct attribute_group qeth_l2_bridgeport_attr_group = {
 	.attrs = qeth_l2_bridgeport_attrs,
 };
 
-/**
- * qeth_l2_setup_bridgeport_attrs() - set/restore attrs when turning online.
- * @card:			      qeth_card structure pointer
- *
- * Note: this function is called with conf_mutex held by the caller
- */
-void qeth_l2_setup_bridgeport_attrs(struct qeth_card *card)
-{
-	int rc;
-
-	if (!card)
-		return;
-	if (!card->options.sbp.supported_funcs)
-		return;
-
-	mutex_lock(&card->sbp_lock);
-	if (!card->options.sbp.reflect_promisc &&
-	    card->options.sbp.role != QETH_SBP_ROLE_NONE) {
-		/* Conditional to avoid spurious error messages */
-		qeth_bridgeport_setrole(card, card->options.sbp.role);
-		/* Let the callback function refresh the stored role value. */
-		qeth_bridgeport_query_ports(card,
-			&card->options.sbp.role, NULL);
-	}
-	if (card->options.sbp.hostnotification) {
-		rc = qeth_bridgeport_an_set(card, 1);
-		if (rc)
-			card->options.sbp.hostnotification = 0;
-	} else {
-		qeth_bridgeport_an_set(card, 0);
-	}
-	mutex_unlock(&card->sbp_lock);
-}
-
 /* VNIC CHARS support */
 
 /* convert sysfs attr name to VNIC characteristic */
-- 
cgit v1.2.3-59-g8ed1b


From aa3ad3920301f2cf6d406209b14acde0c50f2323 Mon Sep 17 00:00:00 2001
From: Julian Wiedmann <jwi@linux.ibm.com>
Date: Sat, 25 Jan 2020 16:53:00 +0100
Subject: s390/qeth: consolidate QDIO queue setup

Move some duplicated logic into a shared code path.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/s390/net/qeth_core.h      | 1 -
 drivers/s390/net/qeth_core_main.c | 9 +++++++--
 drivers/s390/net/qeth_l2_main.c   | 8 +-------
 drivers/s390/net/qeth_l3_main.c   | 8 +-------
 4 files changed, 9 insertions(+), 17 deletions(-)

diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h
index 7c37e94fb28b..09d8211fd542 100644
--- a/drivers/s390/net/qeth_core.h
+++ b/drivers/s390/net/qeth_core.h
@@ -994,7 +994,6 @@ int qeth_core_hardsetup_card(struct qeth_card *card, bool *carrier_ok);
 int qeth_stop_channel(struct qeth_channel *channel);
 
 void qeth_print_status_message(struct qeth_card *);
-int qeth_init_qdio_queues(struct qeth_card *);
 int qeth_send_ipa_cmd(struct qeth_card *, struct qeth_cmd_buffer *,
 		  int (*reply_cb)
 		  (struct qeth_card *, struct qeth_reply *, unsigned long),
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index c19cc3978e1f..52e5f086444b 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -2665,7 +2665,7 @@ static unsigned int qeth_tx_select_bulk_max(struct qeth_card *card,
 	return card->ssqd.mmwc ? card->ssqd.mmwc : 1;
 }
 
-int qeth_init_qdio_queues(struct qeth_card *card)
+static int qeth_init_qdio_queues(struct qeth_card *card)
 {
 	unsigned int i;
 	int rc;
@@ -2713,7 +2713,6 @@ int qeth_init_qdio_queues(struct qeth_card *card)
 	}
 	return 0;
 }
-EXPORT_SYMBOL_GPL(qeth_init_qdio_queues);
 
 static void qeth_ipa_finalize_cmd(struct qeth_card *card,
 				  struct qeth_cmd_buffer *iob)
@@ -5026,6 +5025,12 @@ retriable:
 	if (rc)
 		goto out;
 
+	rc = qeth_init_qdio_queues(card);
+	if (rc) {
+		QETH_CARD_TEXT_(card, 2, "9err%d", rc);
+		goto out;
+	}
+
 	return 0;
 out:
 	dev_warn(&card->gdev->dev, "The qeth device driver failed to recover "
diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c
index 7da306e267c9..5d9bb0597465 100644
--- a/drivers/s390/net/qeth_l2_main.c
+++ b/drivers/s390/net/qeth_l2_main.c
@@ -288,11 +288,11 @@ static void qeth_l2_stop_card(struct qeth_card *card)
 	}
 	if (card->state == CARD_STATE_HARDSETUP) {
 		qeth_drain_output_queues(card);
-		qeth_clear_working_pool_list(card);
 		card->state = CARD_STATE_DOWN;
 	}
 
 	qeth_qdio_clear_card(card, 0);
+	qeth_clear_working_pool_list(card);
 	flush_workqueue(card->event_wq);
 	card->info.mac_bits &= ~QETH_LAYER2_MAC_REGISTERED;
 	card->info.promisc_mode = 0;
@@ -787,12 +787,6 @@ static int qeth_l2_set_online(struct ccwgroup_device *gdev)
 	/* softsetup */
 	QETH_CARD_TEXT(card, 2, "softsetp");
 
-	rc = qeth_init_qdio_queues(card);
-	if (rc) {
-		QETH_CARD_TEXT_(card, 2, "6err%d", rc);
-		rc = -ENODEV;
-		goto out_remove;
-	}
 	card->state = CARD_STATE_SOFTSETUP;
 
 	qeth_set_allowed_threads(card, 0xffffffff, 0);
diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index 6c0bffd11138..ee1bdaaa26f6 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -1184,11 +1184,11 @@ static void qeth_l3_stop_card(struct qeth_card *card)
 	}
 	if (card->state == CARD_STATE_HARDSETUP) {
 		qeth_drain_output_queues(card);
-		qeth_clear_working_pool_list(card);
 		card->state = CARD_STATE_DOWN;
 	}
 
 	qeth_qdio_clear_card(card, 0);
+	qeth_clear_working_pool_list(card);
 	flush_workqueue(card->event_wq);
 	card->info.promisc_mode = 0;
 }
@@ -2097,12 +2097,6 @@ static int qeth_l3_set_online(struct ccwgroup_device *gdev)
 			QETH_CARD_TEXT_(card, 2, "5err%04x", rc);
 	}
 
-	rc = qeth_init_qdio_queues(card);
-	if (rc) {
-		QETH_CARD_TEXT_(card, 2, "6err%d", rc);
-		rc = -ENODEV;
-		goto out_remove;
-	}
 	card->state = CARD_STATE_SOFTSETUP;
 
 	qeth_set_allowed_threads(card, 0xffffffff, 0);
-- 
cgit v1.2.3-59-g8ed1b


From 91003f354e6bbe8d225b4d54127d80c694d201d9 Mon Sep 17 00:00:00 2001
From: Julian Wiedmann <jwi@linux.ibm.com>
Date: Sat, 25 Jan 2020 16:53:01 +0100
Subject: s390/qeth: consolidate online/offline code

Large parts of the online/offline code are identical now, and cleaning
up the remaining stuff is easier with a shared core.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/s390/net/qeth_core.h      |   9 ++--
 drivers/s390/net/qeth_core_main.c | 110 +++++++++++++++++++++++++++++++++-----
 drivers/s390/net/qeth_l2_main.c   |  85 ++---------------------------
 drivers/s390/net/qeth_l3_main.c   |  87 ++----------------------------
 4 files changed, 107 insertions(+), 184 deletions(-)

diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h
index 09d8211fd542..d052a265da1c 100644
--- a/drivers/s390/net/qeth_core.h
+++ b/drivers/s390/net/qeth_core.h
@@ -727,11 +727,10 @@ struct qeth_osn_info {
 
 struct qeth_discipline {
 	const struct device_type *devtype;
-	int (*recover)(void *ptr);
 	int (*setup) (struct ccwgroup_device *);
 	void (*remove) (struct ccwgroup_device *);
-	int (*set_online) (struct ccwgroup_device *);
-	int (*set_offline) (struct ccwgroup_device *);
+	int (*set_online)(struct qeth_card *card);
+	void (*set_offline)(struct qeth_card *card);
 	int (*do_ioctl)(struct net_device *dev, struct ifreq *rq, int cmd);
 	int (*control_event_handler)(struct qeth_card *card,
 					struct qeth_ipa_cmd *cmd);
@@ -987,11 +986,9 @@ struct net_device *qeth_clone_netdev(struct net_device *orig);
 struct qeth_card *qeth_get_card_by_busid(char *bus_id);
 void qeth_set_allowed_threads(struct qeth_card *, unsigned long , int);
 int qeth_threads_running(struct qeth_card *, unsigned long);
-int qeth_do_run_thread(struct qeth_card *, unsigned long);
-void qeth_clear_thread_start_bit(struct qeth_card *, unsigned long);
-void qeth_clear_thread_running_bit(struct qeth_card *, unsigned long);
 int qeth_core_hardsetup_card(struct qeth_card *card, bool *carrier_ok);
 int qeth_stop_channel(struct qeth_channel *channel);
+int qeth_set_offline(struct qeth_card *card, bool resetting);
 
 void qeth_print_status_message(struct qeth_card *);
 int qeth_send_ipa_cmd(struct qeth_card *, struct qeth_cmd_buffer *,
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index 52e5f086444b..d66a7433908c 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -823,7 +823,8 @@ static int qeth_set_thread_start_bit(struct qeth_card *card,
 	return 0;
 }
 
-void qeth_clear_thread_start_bit(struct qeth_card *card, unsigned long thread)
+static void qeth_clear_thread_start_bit(struct qeth_card *card,
+					unsigned long thread)
 {
 	unsigned long flags;
 
@@ -832,9 +833,9 @@ void qeth_clear_thread_start_bit(struct qeth_card *card, unsigned long thread)
 	spin_unlock_irqrestore(&card->thread_mask_lock, flags);
 	wake_up(&card->wait_q);
 }
-EXPORT_SYMBOL_GPL(qeth_clear_thread_start_bit);
 
-void qeth_clear_thread_running_bit(struct qeth_card *card, unsigned long thread)
+static void qeth_clear_thread_running_bit(struct qeth_card *card,
+					  unsigned long thread)
 {
 	unsigned long flags;
 
@@ -843,7 +844,6 @@ void qeth_clear_thread_running_bit(struct qeth_card *card, unsigned long thread)
 	spin_unlock_irqrestore(&card->thread_mask_lock, flags);
 	wake_up_all(&card->wait_q);
 }
-EXPORT_SYMBOL_GPL(qeth_clear_thread_running_bit);
 
 static int __qeth_do_run_thread(struct qeth_card *card, unsigned long thread)
 {
@@ -864,7 +864,7 @@ static int __qeth_do_run_thread(struct qeth_card *card, unsigned long thread)
 	return rc;
 }
 
-int qeth_do_run_thread(struct qeth_card *card, unsigned long thread)
+static int qeth_do_run_thread(struct qeth_card *card, unsigned long thread)
 {
 	int rc = 0;
 
@@ -872,7 +872,6 @@ int qeth_do_run_thread(struct qeth_card *card, unsigned long thread)
 		   (rc = __qeth_do_run_thread(card, thread)) >= 0);
 	return rc;
 }
-EXPORT_SYMBOL_GPL(qeth_do_run_thread);
 
 void qeth_schedule_recovery(struct qeth_card *card)
 {
@@ -880,7 +879,6 @@ void qeth_schedule_recovery(struct qeth_card *card)
 	if (qeth_set_thread_start_bit(card, QETH_RECOVER_THREAD) == 0)
 		schedule_work(&card->kernel_thread_starter);
 }
-EXPORT_SYMBOL_GPL(qeth_schedule_recovery);
 
 static int qeth_get_problem(struct qeth_card *card, struct ccw_device *cdev,
 			    struct irb *irb)
@@ -1287,6 +1285,7 @@ static int qeth_do_start_thread(struct qeth_card *card, unsigned long thread)
 	return rc;
 }
 
+static int qeth_do_reset(void *data);
 static void qeth_start_kernel_thread(struct work_struct *work)
 {
 	struct task_struct *ts;
@@ -1298,8 +1297,7 @@ static void qeth_start_kernel_thread(struct work_struct *work)
 	    card->write.state != CH_STATE_UP)
 		return;
 	if (qeth_do_start_thread(card, QETH_RECOVER_THREAD)) {
-		ts = kthread_run(card->discipline->recover, (void *)card,
-				"qeth_recover");
+		ts = kthread_run(qeth_do_reset, card, "qeth_recover");
 		if (IS_ERR(ts)) {
 			qeth_clear_thread_start_bit(card, QETH_RECOVER_THREAD);
 			qeth_clear_thread_running_bit(card,
@@ -3095,7 +3093,6 @@ int qeth_hw_trap(struct qeth_card *card, enum qeth_diags_trap_action action)
 	}
 	return qeth_send_ipa_cmd(card, iob, qeth_hw_trap_cb, NULL);
 }
-EXPORT_SYMBOL_GPL(qeth_hw_trap);
 
 static int qeth_check_qdio_errors(struct qeth_card *card,
 				  struct qdio_buffer *buf,
@@ -5041,6 +5038,89 @@ out:
 }
 EXPORT_SYMBOL_GPL(qeth_core_hardsetup_card);
 
+static int qeth_set_online(struct qeth_card *card)
+{
+	int rc;
+
+	mutex_lock(&card->discipline_mutex);
+	mutex_lock(&card->conf_mutex);
+	QETH_CARD_TEXT(card, 2, "setonlin");
+
+	rc = card->discipline->set_online(card);
+
+	mutex_unlock(&card->conf_mutex);
+	mutex_unlock(&card->discipline_mutex);
+
+	return rc;
+}
+
+int qeth_set_offline(struct qeth_card *card, bool resetting)
+{
+	int rc, rc2, rc3;
+
+	mutex_lock(&card->discipline_mutex);
+	mutex_lock(&card->conf_mutex);
+	QETH_CARD_TEXT(card, 3, "setoffl");
+
+	if ((!resetting && card->info.hwtrap) || card->info.hwtrap == 2) {
+		qeth_hw_trap(card, QETH_DIAGS_TRAP_DISARM);
+		card->info.hwtrap = 1;
+	}
+
+	rtnl_lock();
+	card->info.open_when_online = card->dev->flags & IFF_UP;
+	dev_close(card->dev);
+	netif_device_detach(card->dev);
+	netif_carrier_off(card->dev);
+	rtnl_unlock();
+
+	card->discipline->set_offline(card);
+
+	rc  = qeth_stop_channel(&card->data);
+	rc2 = qeth_stop_channel(&card->write);
+	rc3 = qeth_stop_channel(&card->read);
+	if (!rc)
+		rc = (rc2) ? rc2 : rc3;
+	if (rc)
+		QETH_CARD_TEXT_(card, 2, "1err%d", rc);
+	qdio_free(CARD_DDEV(card));
+
+	/* let user_space know that device is offline */
+	kobject_uevent(&card->gdev->dev.kobj, KOBJ_CHANGE);
+
+	mutex_unlock(&card->conf_mutex);
+	mutex_unlock(&card->discipline_mutex);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(qeth_set_offline);
+
+static int qeth_do_reset(void *data)
+{
+	struct qeth_card *card = data;
+	int rc;
+
+	QETH_CARD_TEXT(card, 2, "recover1");
+	if (!qeth_do_run_thread(card, QETH_RECOVER_THREAD))
+		return 0;
+	QETH_CARD_TEXT(card, 2, "recover2");
+	dev_warn(&card->gdev->dev,
+		 "A recovery process has been started for the device\n");
+
+	qeth_set_offline(card, true);
+	rc = qeth_set_online(card);
+	if (!rc) {
+		dev_info(&card->gdev->dev,
+			 "Device successfully recovered!\n");
+	} else {
+		ccwgroup_set_offline(card->gdev);
+		dev_warn(&card->gdev->dev,
+			 "The qeth device driver failed to recover an error on the device\n");
+	}
+	qeth_clear_thread_start_bit(card, QETH_RECOVER_THREAD);
+	qeth_clear_thread_running_bit(card, QETH_RECOVER_THREAD);
+	return 0;
+}
+
 #if IS_ENABLED(CONFIG_QETH_L3)
 static void qeth_l3_rebuild_skb(struct qeth_card *card, struct sk_buff *skb,
 				struct qeth_hdr *hdr)
@@ -5977,7 +6057,8 @@ static int qeth_core_set_online(struct ccwgroup_device *gdev)
 			goto err;
 		}
 	}
-	rc = card->discipline->set_online(gdev);
+
+	rc = qeth_set_online(card);
 err:
 	return rc;
 }
@@ -5985,7 +6066,8 @@ err:
 static int qeth_core_set_offline(struct ccwgroup_device *gdev)
 {
 	struct qeth_card *card = dev_get_drvdata(&gdev->dev);
-	return card->discipline->set_offline(gdev);
+
+	return qeth_set_offline(card, false);
 }
 
 static void qeth_core_shutdown(struct ccwgroup_device *gdev)
@@ -6008,7 +6090,7 @@ static int qeth_suspend(struct ccwgroup_device *gdev)
 	if (gdev->state == CCWGROUP_OFFLINE)
 		return 0;
 
-	card->discipline->set_offline(gdev);
+	qeth_set_offline(card, false);
 	return 0;
 }
 
@@ -6017,7 +6099,7 @@ static int qeth_resume(struct ccwgroup_device *gdev)
 	struct qeth_card *card = dev_get_drvdata(&gdev->dev);
 	int rc;
 
-	rc = card->discipline->set_online(gdev);
+	rc = qeth_set_online(card);
 
 	qeth_set_allowed_threads(card, 0xffffffff, 0);
 	if (rc)
diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c
index 5d9bb0597465..c36b6c1fc33f 100644
--- a/drivers/s390/net/qeth_l2_main.c
+++ b/drivers/s390/net/qeth_l2_main.c
@@ -24,7 +24,6 @@
 #include "qeth_core.h"
 #include "qeth_l2.h"
 
-static int qeth_l2_set_offline(struct ccwgroup_device *);
 static void qeth_bridgeport_query_support(struct qeth_card *card);
 static void qeth_bridge_state_change(struct qeth_card *card,
 					struct qeth_ipa_cmd *cmd);
@@ -610,7 +609,7 @@ static void qeth_l2_remove_device(struct ccwgroup_device *cgdev)
 	wait_event(card->wait_q, qeth_threads_running(card, 0xffffffff) == 0);
 
 	if (cgdev->state == CCWGROUP_ONLINE)
-		qeth_l2_set_offline(cgdev);
+		qeth_set_offline(card, false);
 
 	cancel_work_sync(&card->close_dev_work);
 	if (qeth_netdev_is_registered(card->dev))
@@ -746,17 +745,13 @@ static void qeth_l2_setup_bridgeport_attrs(struct qeth_card *card)
 	}
 }
 
-static int qeth_l2_set_online(struct ccwgroup_device *gdev)
+static int qeth_l2_set_online(struct qeth_card *card)
 {
-	struct qeth_card *card = dev_get_drvdata(&gdev->dev);
+	struct ccwgroup_device *gdev = card->gdev;
 	struct net_device *dev = card->dev;
 	int rc = 0;
 	bool carrier_ok;
 
-	mutex_lock(&card->discipline_mutex);
-	mutex_lock(&card->conf_mutex);
-	QETH_CARD_TEXT(card, 2, "setonlin");
-
 	rc = qeth_core_hardsetup_card(card, &carrier_ok);
 	if (rc) {
 		QETH_CARD_TEXT_(card, 2, "2err%04x", rc);
@@ -813,8 +808,6 @@ static int qeth_l2_set_online(struct ccwgroup_device *gdev)
 	}
 	/* let user_space know that device is online */
 	kobject_uevent(&gdev->dev.kobj, KOBJ_CHANGE);
-	mutex_unlock(&card->conf_mutex);
-	mutex_unlock(&card->discipline_mutex);
 	return 0;
 
 out_remove:
@@ -823,81 +816,12 @@ out_remove:
 	qeth_stop_channel(&card->write);
 	qeth_stop_channel(&card->read);
 	qdio_free(CARD_DDEV(card));
-
-	mutex_unlock(&card->conf_mutex);
-	mutex_unlock(&card->discipline_mutex);
 	return rc;
 }
 
-static int __qeth_l2_set_offline(struct ccwgroup_device *cgdev,
-					int recovery_mode)
+static void qeth_l2_set_offline(struct qeth_card *card)
 {
-	struct qeth_card *card = dev_get_drvdata(&cgdev->dev);
-	int rc = 0, rc2 = 0, rc3 = 0;
-
-	mutex_lock(&card->discipline_mutex);
-	mutex_lock(&card->conf_mutex);
-	QETH_CARD_TEXT(card, 3, "setoffl");
-
-	if ((!recovery_mode && card->info.hwtrap) || card->info.hwtrap == 2) {
-		qeth_hw_trap(card, QETH_DIAGS_TRAP_DISARM);
-		card->info.hwtrap = 1;
-	}
-
-	rtnl_lock();
-	card->info.open_when_online = card->dev->flags & IFF_UP;
-	dev_close(card->dev);
-	netif_device_detach(card->dev);
-	netif_carrier_off(card->dev);
-	rtnl_unlock();
-
 	qeth_l2_stop_card(card);
-	rc  = qeth_stop_channel(&card->data);
-	rc2 = qeth_stop_channel(&card->write);
-	rc3 = qeth_stop_channel(&card->read);
-	if (!rc)
-		rc = (rc2) ? rc2 : rc3;
-	if (rc)
-		QETH_CARD_TEXT_(card, 2, "1err%d", rc);
-	qdio_free(CARD_DDEV(card));
-
-	/* let user_space know that device is offline */
-	kobject_uevent(&cgdev->dev.kobj, KOBJ_CHANGE);
-	mutex_unlock(&card->conf_mutex);
-	mutex_unlock(&card->discipline_mutex);
-	return 0;
-}
-
-static int qeth_l2_set_offline(struct ccwgroup_device *cgdev)
-{
-	return __qeth_l2_set_offline(cgdev, 0);
-}
-
-static int qeth_l2_recover(void *ptr)
-{
-	struct qeth_card *card;
-	int rc = 0;
-
-	card = (struct qeth_card *) ptr;
-	QETH_CARD_TEXT(card, 2, "recover1");
-	if (!qeth_do_run_thread(card, QETH_RECOVER_THREAD))
-		return 0;
-	QETH_CARD_TEXT(card, 2, "recover2");
-	dev_warn(&card->gdev->dev,
-		"A recovery process has been started for the device\n");
-	__qeth_l2_set_offline(card->gdev, 1);
-	rc = qeth_l2_set_online(card->gdev);
-	if (!rc)
-		dev_info(&card->gdev->dev,
-			"Device successfully recovered!\n");
-	else {
-		ccwgroup_set_offline(card->gdev);
-		dev_warn(&card->gdev->dev, "The qeth device driver "
-				"failed to recover an error on the device\n");
-	}
-	qeth_clear_thread_start_bit(card, QETH_RECOVER_THREAD);
-	qeth_clear_thread_running_bit(card, QETH_RECOVER_THREAD);
-	return 0;
 }
 
 static int __init qeth_l2_init(void)
@@ -934,7 +858,6 @@ static int qeth_l2_control_event(struct qeth_card *card,
 
 struct qeth_discipline qeth_l2_discipline = {
 	.devtype = &qeth_l2_devtype,
-	.recover = qeth_l2_recover,
 	.setup = qeth_l2_probe_device,
 	.remove = qeth_l2_remove_device,
 	.set_online = qeth_l2_set_online,
diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index ee1bdaaa26f6..eb2d9c427b10 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -37,8 +37,6 @@
 
 #include "qeth_l3.h"
 
-
-static int qeth_l3_set_offline(struct ccwgroup_device *);
 static int qeth_l3_register_addr_entry(struct qeth_card *,
 		struct qeth_ipaddr *);
 static int qeth_l3_deregister_addr_entry(struct qeth_card *,
@@ -2044,7 +2042,7 @@ static void qeth_l3_remove_device(struct ccwgroup_device *cgdev)
 	wait_event(card->wait_q, qeth_threads_running(card, 0xffffffff) == 0);
 
 	if (cgdev->state == CCWGROUP_ONLINE)
-		qeth_l3_set_offline(cgdev);
+		qeth_set_offline(card, false);
 
 	cancel_work_sync(&card->close_dev_work);
 	if (qeth_netdev_is_registered(card->dev))
@@ -2056,17 +2054,13 @@ static void qeth_l3_remove_device(struct ccwgroup_device *cgdev)
 	qeth_l3_clear_ipato_list(card);
 }
 
-static int qeth_l3_set_online(struct ccwgroup_device *gdev)
+static int qeth_l3_set_online(struct qeth_card *card)
 {
-	struct qeth_card *card = dev_get_drvdata(&gdev->dev);
+	struct ccwgroup_device *gdev = card->gdev;
 	struct net_device *dev = card->dev;
 	int rc = 0;
 	bool carrier_ok;
 
-	mutex_lock(&card->discipline_mutex);
-	mutex_lock(&card->conf_mutex);
-	QETH_CARD_TEXT(card, 2, "setonlin");
-
 	rc = qeth_core_hardsetup_card(card, &carrier_ok);
 	if (rc) {
 		QETH_CARD_TEXT_(card, 2, "2err%04x", rc);
@@ -2125,8 +2119,6 @@ static int qeth_l3_set_online(struct ccwgroup_device *gdev)
 	qeth_trace_features(card);
 	/* let user_space know that device is online */
 	kobject_uevent(&gdev->dev.kobj, KOBJ_CHANGE);
-	mutex_unlock(&card->conf_mutex);
-	mutex_unlock(&card->discipline_mutex);
 	return 0;
 out_remove:
 	qeth_l3_stop_card(card);
@@ -2134,82 +2126,12 @@ out_remove:
 	qeth_stop_channel(&card->write);
 	qeth_stop_channel(&card->read);
 	qdio_free(CARD_DDEV(card));
-
-	mutex_unlock(&card->conf_mutex);
-	mutex_unlock(&card->discipline_mutex);
 	return rc;
 }
 
-static int __qeth_l3_set_offline(struct ccwgroup_device *cgdev,
-			int recovery_mode)
+static void qeth_l3_set_offline(struct qeth_card *card)
 {
-	struct qeth_card *card = dev_get_drvdata(&cgdev->dev);
-	int rc = 0, rc2 = 0, rc3 = 0;
-
-	mutex_lock(&card->discipline_mutex);
-	mutex_lock(&card->conf_mutex);
-	QETH_CARD_TEXT(card, 3, "setoffl");
-
-	if ((!recovery_mode && card->info.hwtrap) || card->info.hwtrap == 2) {
-		qeth_hw_trap(card, QETH_DIAGS_TRAP_DISARM);
-		card->info.hwtrap = 1;
-	}
-
-	rtnl_lock();
-	card->info.open_when_online = card->dev->flags & IFF_UP;
-	dev_close(card->dev);
-	netif_device_detach(card->dev);
-	netif_carrier_off(card->dev);
-	rtnl_unlock();
-
 	qeth_l3_stop_card(card);
-	rc  = qeth_stop_channel(&card->data);
-	rc2 = qeth_stop_channel(&card->write);
-	rc3 = qeth_stop_channel(&card->read);
-	if (!rc)
-		rc = (rc2) ? rc2 : rc3;
-	if (rc)
-		QETH_CARD_TEXT_(card, 2, "1err%d", rc);
-	qdio_free(CARD_DDEV(card));
-
-	/* let user_space know that device is offline */
-	kobject_uevent(&cgdev->dev.kobj, KOBJ_CHANGE);
-	mutex_unlock(&card->conf_mutex);
-	mutex_unlock(&card->discipline_mutex);
-	return 0;
-}
-
-static int qeth_l3_set_offline(struct ccwgroup_device *cgdev)
-{
-	return __qeth_l3_set_offline(cgdev, 0);
-}
-
-static int qeth_l3_recover(void *ptr)
-{
-	struct qeth_card *card;
-	int rc = 0;
-
-	card = (struct qeth_card *) ptr;
-	QETH_CARD_TEXT(card, 2, "recover1");
-	QETH_CARD_HEX(card, 2, &card, sizeof(void *));
-	if (!qeth_do_run_thread(card, QETH_RECOVER_THREAD))
-		return 0;
-	QETH_CARD_TEXT(card, 2, "recover2");
-	dev_warn(&card->gdev->dev,
-		"A recovery process has been started for the device\n");
-	__qeth_l3_set_offline(card->gdev, 1);
-	rc = qeth_l3_set_online(card->gdev);
-	if (!rc)
-		dev_info(&card->gdev->dev,
-			"Device successfully recovered!\n");
-	else {
-		ccwgroup_set_offline(card->gdev);
-		dev_warn(&card->gdev->dev, "The qeth device driver "
-				"failed to recover an error on the device\n");
-	}
-	qeth_clear_thread_start_bit(card, QETH_RECOVER_THREAD);
-	qeth_clear_thread_running_bit(card, QETH_RECOVER_THREAD);
-	return 0;
 }
 
 /* Returns zero if the command is successfully "consumed" */
@@ -2221,7 +2143,6 @@ static int qeth_l3_control_event(struct qeth_card *card,
 
 struct qeth_discipline qeth_l3_discipline = {
 	.devtype = &qeth_l3_devtype,
-	.recover = qeth_l3_recover,
 	.setup = qeth_l3_probe_device,
 	.remove = qeth_l3_remove_device,
 	.set_online = qeth_l3_set_online,
-- 
cgit v1.2.3-59-g8ed1b


From 7f92316c5393b583c5217bb2cfc789d6820ec393 Mon Sep 17 00:00:00 2001
From: Julian Wiedmann <jwi@linux.ibm.com>
Date: Sat, 25 Jan 2020 16:53:02 +0100
Subject: s390/qeth: make cmd/reply matching more flexible

When data is received on the READ channel, the matching logic for cmds
that are waiting for a reply is currently hard-coded into the channel's
main IO callback.
Move this into a per-cmd callback, so that we can apply custom matching
logic for each individual cmd.

This also allows us to remove the coarse-grained check for unexpected
non-IPA replies, since they will no longer match against _all_ pending
cmds.

Note that IDX cmds use _no_ matcher, since their reply is synchronously
received as part of the cmd's IO.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/s390/net/qeth_core.h      | 14 +++++++++++++-
 drivers/s390/net/qeth_core_main.c | 34 ++++++++++++++++++++++++----------
 drivers/s390/net/qeth_l2_main.c   |  3 ++-
 3 files changed, 39 insertions(+), 12 deletions(-)

diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h
index d052a265da1c..950d01551373 100644
--- a/drivers/s390/net/qeth_core.h
+++ b/drivers/s390/net/qeth_core.h
@@ -608,6 +608,8 @@ struct qeth_cmd_buffer {
 	long timeout;
 	unsigned char *data;
 	void (*finalize)(struct qeth_card *card, struct qeth_cmd_buffer *iob);
+	bool (*match)(struct qeth_cmd_buffer *iob,
+		      struct qeth_cmd_buffer *reply);
 	void (*callback)(struct qeth_card *card, struct qeth_cmd_buffer *iob,
 			 unsigned int data_length);
 	int rc;
@@ -618,6 +620,14 @@ static inline void qeth_get_cmd(struct qeth_cmd_buffer *iob)
 	refcount_inc(&iob->ref_count);
 }
 
+static inline struct qeth_ipa_cmd *__ipa_reply(struct qeth_cmd_buffer *iob)
+{
+	if (!IS_IPA(iob->data))
+		return NULL;
+
+	return (struct qeth_ipa_cmd *) PDU_ENCAPSULATION(iob->data);
+}
+
 static inline struct qeth_ipa_cmd *__ipa_cmd(struct qeth_cmd_buffer *iob)
 {
 	return (struct qeth_ipa_cmd *)(iob->data + IPA_PDU_HEADER_SIZE);
@@ -1023,7 +1033,9 @@ void qeth_setadp_promisc_mode(struct qeth_card *card, bool enable);
 int qeth_setadpparms_change_macaddr(struct qeth_card *);
 void qeth_tx_timeout(struct net_device *, unsigned int txqueue);
 void qeth_prepare_ipa_cmd(struct qeth_card *card, struct qeth_cmd_buffer *iob,
-			  u16 cmd_length);
+			  u16 cmd_length,
+			  bool (*match)(struct qeth_cmd_buffer *iob,
+					struct qeth_cmd_buffer *reply));
 int qeth_query_switch_attributes(struct qeth_card *card,
 				  struct qeth_switch_info *sw_info);
 int qeth_query_card_info(struct qeth_card *card,
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index d66a7433908c..0f5f36e63823 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -748,8 +748,8 @@ static void qeth_issue_next_read_cb(struct qeth_card *card,
 		goto out;
 	}
 
-	if (IS_IPA(iob->data)) {
-		cmd = (struct qeth_ipa_cmd *) PDU_ENCAPSULATION(iob->data);
+	cmd = __ipa_reply(iob);
+	if (cmd) {
 		cmd = qeth_check_ipa_data(card, cmd);
 		if (!cmd)
 			goto out;
@@ -758,17 +758,12 @@ static void qeth_issue_next_read_cb(struct qeth_card *card,
 			card->osn_info.assist_cb(card->dev, cmd);
 			goto out;
 		}
-	} else {
-		/* non-IPA commands should only flow during initialization */
-		if (card->state != CARD_STATE_DOWN)
-			goto out;
 	}
 
 	/* match against pending cmd requests */
 	spin_lock_irqsave(&card->lock, flags);
 	list_for_each_entry(tmp, &card->cmd_waiter_list, list) {
-		if (!IS_IPA(tmp->data) ||
-		    __ipa_cmd(tmp)->hdr.seqno == cmd->hdr.seqno) {
+		if (tmp->match && tmp->match(tmp, iob)) {
 			request = tmp;
 			/* take the object outside the lock */
 			qeth_get_cmd(request);
@@ -1688,6 +1683,13 @@ static void qeth_mpc_finalize_cmd(struct qeth_card *card,
 	iob->callback = qeth_release_buffer_cb;
 }
 
+static bool qeth_mpc_match_reply(struct qeth_cmd_buffer *iob,
+				 struct qeth_cmd_buffer *reply)
+{
+	/* MPC cmds are issued strictly in sequence. */
+	return !IS_IPA(reply->data);
+}
+
 static struct qeth_cmd_buffer *qeth_mpc_alloc_cmd(struct qeth_card *card,
 						  void *data,
 						  unsigned int data_length)
@@ -1702,6 +1704,7 @@ static struct qeth_cmd_buffer *qeth_mpc_alloc_cmd(struct qeth_card *card,
 	qeth_setup_ccw(__ccw_from_cmd(iob), CCW_CMD_WRITE, 0, data_length,
 		       iob->data);
 	iob->finalize = qeth_mpc_finalize_cmd;
+	iob->match = qeth_mpc_match_reply;
 	return iob;
 }
 
@@ -2722,7 +2725,9 @@ static void qeth_ipa_finalize_cmd(struct qeth_card *card,
 }
 
 void qeth_prepare_ipa_cmd(struct qeth_card *card, struct qeth_cmd_buffer *iob,
-			  u16 cmd_length)
+			  u16 cmd_length,
+			  bool (*match)(struct qeth_cmd_buffer *iob,
+					struct qeth_cmd_buffer *reply))
 {
 	u8 prot_type = qeth_mpc_select_prot_type(card);
 	u16 total_length = iob->length;
@@ -2730,6 +2735,7 @@ void qeth_prepare_ipa_cmd(struct qeth_card *card, struct qeth_cmd_buffer *iob,
 	qeth_setup_ccw(__ccw_from_cmd(iob), CCW_CMD_WRITE, 0, total_length,
 		       iob->data);
 	iob->finalize = qeth_ipa_finalize_cmd;
+	iob->match = match;
 
 	memcpy(iob->data, IPA_PDU_HEADER, IPA_PDU_HEADER_SIZE);
 	memcpy(QETH_IPA_PDU_LEN_TOTAL(iob->data), &total_length, 2);
@@ -2742,6 +2748,14 @@ void qeth_prepare_ipa_cmd(struct qeth_card *card, struct qeth_cmd_buffer *iob,
 }
 EXPORT_SYMBOL_GPL(qeth_prepare_ipa_cmd);
 
+static bool qeth_ipa_match_reply(struct qeth_cmd_buffer *iob,
+				 struct qeth_cmd_buffer *reply)
+{
+	struct qeth_ipa_cmd *ipa_reply = __ipa_reply(reply);
+
+	return ipa_reply && (__ipa_cmd(iob)->hdr.seqno == ipa_reply->hdr.seqno);
+}
+
 struct qeth_cmd_buffer *qeth_ipa_alloc_cmd(struct qeth_card *card,
 					   enum qeth_ipa_cmds cmd_code,
 					   enum qeth_prot_versions prot,
@@ -2757,7 +2771,7 @@ struct qeth_cmd_buffer *qeth_ipa_alloc_cmd(struct qeth_card *card,
 	if (!iob)
 		return NULL;
 
-	qeth_prepare_ipa_cmd(card, iob, data_length);
+	qeth_prepare_ipa_cmd(card, iob, data_length, qeth_ipa_match_reply);
 
 	hdr = &__ipa_cmd(iob)->hdr;
 	hdr->command = cmd_code;
diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c
index c36b6c1fc33f..d5ffd40937d3 100644
--- a/drivers/s390/net/qeth_l2_main.c
+++ b/drivers/s390/net/qeth_l2_main.c
@@ -896,7 +896,8 @@ int qeth_osn_assist(struct net_device *dev, void *data, int data_len)
 	if (!iob)
 		return -ENOMEM;
 
-	qeth_prepare_ipa_cmd(card, iob, (u16) data_len);
+	qeth_prepare_ipa_cmd(card, iob, (u16) data_len, NULL);
+
 	memcpy(__ipa_cmd(iob), data, data_len);
 	iob->callback = qeth_osn_assist_cb;
 	return qeth_send_ipa_cmd(card, iob, NULL, NULL);
-- 
cgit v1.2.3-59-g8ed1b


From 17d9723e69d44491d04289213006fea1647bd0a6 Mon Sep 17 00:00:00 2001
From: Julian Wiedmann <jwi@linux.ibm.com>
Date: Sat, 25 Jan 2020 16:53:03 +0100
Subject: s390/qeth: remove HARDSETUP state

qeth_l?_stop_card() is _never_ called while in HARDSETUP state, and
there's no other usage of the card state that relies on the
DOWN -> HARDSETUP -> SOFTSETUP transition.

As related cleanup, remove the check in qeth_realloc_buffer_pool() as it
is already done by the callers.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/s390/net/qeth_core.h      |  1 -
 drivers/s390/net/qeth_core_main.c |  3 ---
 drivers/s390/net/qeth_core_sys.c  |  2 --
 drivers/s390/net/qeth_l2_main.c   |  4 ----
 drivers/s390/net/qeth_l3_main.c   | 14 +++-----------
 5 files changed, 3 insertions(+), 21 deletions(-)

diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h
index 950d01551373..9575a627a1e1 100644
--- a/drivers/s390/net/qeth_core.h
+++ b/drivers/s390/net/qeth_core.h
@@ -558,7 +558,6 @@ enum qeth_channel_states {
  */
 enum qeth_card_states {
 	CARD_STATE_DOWN,
-	CARD_STATE_HARDSETUP,
 	CARD_STATE_SOFTSETUP,
 };
 
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index 0f5f36e63823..9639938581f5 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -247,9 +247,6 @@ int qeth_realloc_buffer_pool(struct qeth_card *card, int bufcnt)
 {
 	QETH_CARD_TEXT(card, 2, "realcbp");
 
-	if (card->state != CARD_STATE_DOWN)
-		return -EPERM;
-
 	/* TODO: steel/add buffers from/to a running card's buffer pool (?) */
 	qeth_clear_working_pool_list(card);
 	qeth_free_buffer_pool(card);
diff --git a/drivers/s390/net/qeth_core_sys.c b/drivers/s390/net/qeth_core_sys.c
index 7bd86027f559..2bd9993aa60b 100644
--- a/drivers/s390/net/qeth_core_sys.c
+++ b/drivers/s390/net/qeth_core_sys.c
@@ -24,8 +24,6 @@ static ssize_t qeth_dev_state_show(struct device *dev,
 	switch (card->state) {
 	case CARD_STATE_DOWN:
 		return sprintf(buf, "DOWN\n");
-	case CARD_STATE_HARDSETUP:
-		return sprintf(buf, "HARDSETUP\n");
 	case CARD_STATE_SOFTSETUP:
 		if (card->dev->flags & IFF_UP)
 			return sprintf(buf, "UP (LAN %s)\n",
diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c
index d5ffd40937d3..692bd2623401 100644
--- a/drivers/s390/net/qeth_l2_main.c
+++ b/drivers/s390/net/qeth_l2_main.c
@@ -283,9 +283,6 @@ static void qeth_l2_stop_card(struct qeth_card *card)
 
 	if (card->state == CARD_STATE_SOFTSETUP) {
 		qeth_clear_ipacmd_list(card);
-		card->state = CARD_STATE_HARDSETUP;
-	}
-	if (card->state == CARD_STATE_HARDSETUP) {
 		qeth_drain_output_queues(card);
 		card->state = CARD_STATE_DOWN;
 	}
@@ -776,7 +773,6 @@ static int qeth_l2_set_online(struct qeth_card *card)
 	qeth_trace_features(card);
 	qeth_l2_trace_features(card);
 
-	card->state = CARD_STATE_HARDSETUP;
 	qeth_print_status_message(card);
 
 	/* softsetup */
diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index eb2d9c427b10..317d56647a4a 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -901,7 +901,7 @@ out:
 	return rc;
 }
 
-static int qeth_l3_start_ipassists(struct qeth_card *card)
+static void qeth_l3_start_ipassists(struct qeth_card *card)
 {
 	QETH_CARD_TEXT(card, 3, "strtipas");
 
@@ -911,7 +911,6 @@ static int qeth_l3_start_ipassists(struct qeth_card *card)
 	qeth_l3_start_ipa_multicast(card);		/* go on*/
 	qeth_l3_start_ipa_ipv6(card);		/* go on*/
 	qeth_l3_start_ipa_broadcast(card);		/* go on*/
-	return 0;
 }
 
 static int qeth_l3_iqd_read_initial_mac_cb(struct qeth_card *card,
@@ -1178,9 +1177,6 @@ static void qeth_l3_stop_card(struct qeth_card *card)
 	if (card->state == CARD_STATE_SOFTSETUP) {
 		qeth_l3_clear_ip_htable(card, 1);
 		qeth_clear_ipacmd_list(card);
-		card->state = CARD_STATE_HARDSETUP;
-	}
-	if (card->state == CARD_STATE_HARDSETUP) {
 		qeth_drain_output_queues(card);
 		card->state = CARD_STATE_DOWN;
 	}
@@ -2068,7 +2064,6 @@ static int qeth_l3_set_online(struct qeth_card *card)
 		goto out_remove;
 	}
 
-	card->state = CARD_STATE_HARDSETUP;
 	qeth_print_status_message(card);
 
 	/* softsetup */
@@ -2078,11 +2073,8 @@ static int qeth_l3_set_online(struct qeth_card *card)
 	if (rc)
 		QETH_CARD_TEXT_(card, 2, "2err%04x", rc);
 	if (!card->options.sniffer) {
-		rc = qeth_l3_start_ipassists(card);
-		if (rc) {
-			QETH_CARD_TEXT_(card, 2, "3err%d", rc);
-			goto out_remove;
-		}
+		qeth_l3_start_ipassists(card);
+
 		rc = qeth_l3_setrouting_v4(card);
 		if (rc)
 			QETH_CARD_TEXT_(card, 2, "4err%04x", rc);
-- 
cgit v1.2.3-59-g8ed1b


From 6ba8b3b6bd772f575f7736c8fd893c6981fcce16 Mon Sep 17 00:00:00 2001
From: Rakesh Pillai <pillair@codeaurora.org>
Date: Tue, 21 Jan 2020 12:42:28 +0530
Subject: ath10k: Correct the DMA direction for management tx buffers

The management packets, send to firmware via WMI, are
mapped using the direction DMA_TO_DEVICE. Currently in
case of wmi cleanup, these buffers are being unmapped
using an incorrect DMA direction. This can cause unwanted
behavior when the host driver is handling a restart
of the wlan firmware.

We might see a trace like below

[<ffffff8008098b18>] __dma_inv_area+0x28/0x58
[<ffffff8001176734>] ath10k_wmi_mgmt_tx_clean_up_pending+0x60/0xb0 [ath10k_core]
[<ffffff80088c7c50>] idr_for_each+0x78/0xe4
[<ffffff80011766a4>] ath10k_wmi_detach+0x4c/0x7c [ath10k_core]
[<ffffff8001163d7c>] ath10k_core_stop+0x58/0x68 [ath10k_core]
[<ffffff800114fb74>] ath10k_halt+0xec/0x13c [ath10k_core]
[<ffffff8001165110>] ath10k_core_restart+0x11c/0x1a8 [ath10k_core]
[<ffffff80080c36bc>] process_one_work+0x16c/0x31c

Fix the incorrect DMA direction during the wmi
management tx buffer cleanup.

Tested HW: WCN3990
Tested FW: WLAN.HL.3.1-00784-QCAHLSWMTPLZ-1

Fixes: dc405152bb6 ("ath10k: handle mgmt tx completion event")
Signed-off-by: Rakesh Pillai <pillair@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/wmi.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath10k/wmi.c b/drivers/net/wireless/ath/ath10k/wmi.c
index 13f7531d945d..61885d4d662c 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.c
+++ b/drivers/net/wireless/ath/ath10k/wmi.c
@@ -9490,7 +9490,7 @@ static int ath10k_wmi_mgmt_tx_clean_up_pending(int msdu_id, void *ptr,
 
 	msdu = pkt_addr->vaddr;
 	dma_unmap_single(ar->dev, pkt_addr->paddr,
-			 msdu->len, DMA_FROM_DEVICE);
+			 msdu->len, DMA_TO_DEVICE);
 	ieee80211_free_txskb(ar->hw, msdu);
 
 	return 0;
-- 
cgit v1.2.3-59-g8ed1b


From 32efcc06d2a15fa87585614d12d6c2308cc2d3f3 Mon Sep 17 00:00:00 2001
From: Abdul Kabbani <akabbani@google.com>
Date: Fri, 24 Jan 2020 16:34:02 -0500
Subject: tcp: export count for rehash attempts

Using IPv6 flow-label to swiftly route around avoid congested or
disconnected network path can greatly improve TCP reliability.

This patch adds SNMP counters and a OPT_STATS counter to track both
host-level and connection-level statistics. Network administrators
can use these counters to evaluate the impact of this new ability better.

Export count for rehash attempts to
1) two SNMP counters: TcpTimeoutRehash (rehash due to timeouts),
   and TcpDuplicateDataRehash (rehash due to receiving duplicate
   packets)
2) Timestamping API SOF_TIMESTAMPING_OPT_STATS.

Signed-off-by: Abdul Kabbani <akabbani@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Kevin(Yudong) Yang <yyd@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/tcp.h       | 2 ++
 include/uapi/linux/snmp.h | 2 ++
 include/uapi/linux/tcp.h  | 1 +
 net/ipv4/proc.c           | 2 ++
 net/ipv4/tcp.c            | 2 ++
 net/ipv4/tcp_input.c      | 4 +++-
 net/ipv4/tcp_timer.c      | 6 ++++++
 7 files changed, 18 insertions(+), 1 deletion(-)

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 4e2124607d32..1cf73e6f85ca 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -386,6 +386,8 @@ struct tcp_sock {
 #define BPF_SOCK_OPS_TEST_FLAG(TP, ARG) 0
 #endif
 
+	u16 timeout_rehash;	/* Timeout-triggered rehash attempts */
+
 	u32 rcv_ooopack; /* Received out-of-order packets, for tcpinfo */
 
 /* Receiver side RTT estimation */
diff --git a/include/uapi/linux/snmp.h b/include/uapi/linux/snmp.h
index 7eee233e78d2..7d91f4debc48 100644
--- a/include/uapi/linux/snmp.h
+++ b/include/uapi/linux/snmp.h
@@ -285,6 +285,8 @@ enum
 	LINUX_MIB_TCPRCVQDROP,			/* TCPRcvQDrop */
 	LINUX_MIB_TCPWQUEUETOOBIG,		/* TCPWqueueTooBig */
 	LINUX_MIB_TCPFASTOPENPASSIVEALTKEY,	/* TCPFastOpenPassiveAltKey */
+	LINUX_MIB_TCPTIMEOUTREHASH,		/* TCPTimeoutRehash */
+	LINUX_MIB_TCPDUPLICATEDATAREHASH,	/* TCPDuplicateDataRehash */
 	__LINUX_MIB_MAX
 };
 
diff --git a/include/uapi/linux/tcp.h b/include/uapi/linux/tcp.h
index d87184e673ca..fd9eb8f6bcae 100644
--- a/include/uapi/linux/tcp.h
+++ b/include/uapi/linux/tcp.h
@@ -311,6 +311,7 @@ enum {
 	TCP_NLA_DSACK_DUPS,	/* DSACK blocks received */
 	TCP_NLA_REORD_SEEN,	/* reordering events seen */
 	TCP_NLA_SRTT,		/* smoothed RTT in usecs */
+	TCP_NLA_TIMEOUT_REHASH, /* Timeout-triggered rehash attempts */
 };
 
 /* for TCP_MD5SIG socket option */
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index cc90243ccf76..2580303249e2 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -289,6 +289,8 @@ static const struct snmp_mib snmp4_net_list[] = {
 	SNMP_MIB_ITEM("TCPRcvQDrop", LINUX_MIB_TCPRCVQDROP),
 	SNMP_MIB_ITEM("TCPWqueueTooBig", LINUX_MIB_TCPWQUEUETOOBIG),
 	SNMP_MIB_ITEM("TCPFastOpenPassiveAltKey", LINUX_MIB_TCPFASTOPENPASSIVEALTKEY),
+	SNMP_MIB_ITEM("TcpTimeoutRehash", LINUX_MIB_TCPTIMEOUTREHASH),
+	SNMP_MIB_ITEM("TcpDuplicateDataRehash", LINUX_MIB_TCPDUPLICATEDATAREHASH),
 	SNMP_MIB_SENTINEL
 };
 
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 2857c853e2fd..484485ae74c2 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -3337,6 +3337,7 @@ static size_t tcp_opt_stats_get_size(void)
 		nla_total_size(sizeof(u32)) + /* TCP_NLA_DSACK_DUPS */
 		nla_total_size(sizeof(u32)) + /* TCP_NLA_REORD_SEEN */
 		nla_total_size(sizeof(u32)) + /* TCP_NLA_SRTT */
+		nla_total_size(sizeof(u16)) + /* TCP_NLA_TIMEOUT_REHASH */
 		0;
 }
 
@@ -3391,6 +3392,7 @@ struct sk_buff *tcp_get_timestamping_opt_stats(const struct sock *sk)
 	nla_put_u32(stats, TCP_NLA_DSACK_DUPS, tp->dsack_dups);
 	nla_put_u32(stats, TCP_NLA_REORD_SEEN, tp->reord_seen);
 	nla_put_u32(stats, TCP_NLA_SRTT, tp->srtt_us >> 3);
+	nla_put_u16(stats, TCP_NLA_TIMEOUT_REHASH, tp->timeout_rehash);
 
 	return stats;
 }
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 4915de65d278..e8b840a4767e 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -4271,8 +4271,10 @@ static void tcp_rcv_spurious_retrans(struct sock *sk, const struct sk_buff *skb)
 	 * The receiver remembers and reflects via DSACKs. Leverage the
 	 * DSACK state and change the txhash to re-route speculatively.
 	 */
-	if (TCP_SKB_CB(skb)->seq == tcp_sk(sk)->duplicate_sack[0].start_seq)
+	if (TCP_SKB_CB(skb)->seq == tcp_sk(sk)->duplicate_sack[0].start_seq) {
 		sk_rethink_txhash(sk);
+		NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPDUPLICATEDATAREHASH);
+	}
 }
 
 static void tcp_send_dupack(struct sock *sk, const struct sk_buff *skb)
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index 1097b438befe..c3f26dcd6704 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -223,6 +223,9 @@ static int tcp_write_timeout(struct sock *sk)
 			dst_negative_advice(sk);
 		} else {
 			sk_rethink_txhash(sk);
+			tp->timeout_rehash++;
+			__NET_INC_STATS(sock_net(sk),
+					LINUX_MIB_TCPTIMEOUTREHASH);
 		}
 		retry_until = icsk->icsk_syn_retries ? : net->ipv4.sysctl_tcp_syn_retries;
 		expired = icsk->icsk_retransmits >= retry_until;
@@ -234,6 +237,9 @@ static int tcp_write_timeout(struct sock *sk)
 			dst_negative_advice(sk);
 		} else {
 			sk_rethink_txhash(sk);
+			tp->timeout_rehash++;
+			__NET_INC_STATS(sock_net(sk),
+					LINUX_MIB_TCPTIMEOUTREHASH);
 		}
 
 		retry_until = net->ipv4.sysctl_tcp_retries2;
-- 
cgit v1.2.3-59-g8ed1b


From 79a4b788451dd5afd4ef62d5ff9bcde219ce5286 Mon Sep 17 00:00:00 2001
From: Stephen Boyd <swboyd@chromium.org>
Date: Thu, 23 Jan 2020 15:24:56 -0800
Subject: ath10k: Add newlines to printk messages

Some printks in here don't have newlines at the end, meaning the log
will be sort of hard to read. Add newlines.

Signed-off-by: Stephen Boyd <swboyd@chromium.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/snoc.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/snoc.c b/drivers/net/wireless/ath/ath10k/snoc.c
index aeb4cca92c35..2189a0d01f5a 100644
--- a/drivers/net/wireless/ath/ath10k/snoc.c
+++ b/drivers/net/wireless/ath/ath10k/snoc.c
@@ -582,7 +582,7 @@ static void ath10k_snoc_process_rx_cb(struct ath10k_ce_pipe *ce_state,
 				 max_nbytes, DMA_FROM_DEVICE);
 
 		if (unlikely(max_nbytes < nbytes)) {
-			ath10k_warn(ar, "rxed more than expected (nbytes %d, max %d)",
+			ath10k_warn(ar, "rxed more than expected (nbytes %d, max %d)\n",
 				    nbytes, max_nbytes);
 			dev_kfree_skb_any(skb);
 			continue;
@@ -1201,7 +1201,7 @@ static int ath10k_snoc_request_irq(struct ath10k *ar)
 				  irqflags, ce_name[id], ar);
 		if (ret) {
 			ath10k_err(ar,
-				   "failed to register IRQ handler for CE %d: %d",
+				   "failed to register IRQ handler for CE %d: %d\n",
 				   id, ret);
 			goto err_irq;
 		}
@@ -1485,7 +1485,7 @@ static int ath10k_snoc_probe(struct platform_device *pdev)
 
 	ret = dma_set_mask_and_coherent(dev, drv_data->dma_mask);
 	if (ret) {
-		dev_err(dev, "failed to set dma mask: %d", ret);
+		dev_err(dev, "failed to set dma mask: %d\n", ret);
 		return ret;
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From fa43e99dd4b76201eb323ee3918c68132e8abba9 Mon Sep 17 00:00:00 2001
From: Stephen Boyd <swboyd@chromium.org>
Date: Thu, 23 Jan 2020 15:29:44 -0800
Subject: ath10k: Use device_get_match_data() to simplify code

Use device_get_match_data() here to simplify the code a bit.

Signed-off-by: Stephen Boyd <swboyd@chromium.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ath10k/snoc.c | 11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/snoc.c b/drivers/net/wireless/ath/ath10k/snoc.c
index 2189a0d01f5a..21081b4a27d7 100644
--- a/drivers/net/wireless/ath/ath10k/snoc.c
+++ b/drivers/net/wireless/ath/ath10k/snoc.c
@@ -1466,7 +1466,6 @@ MODULE_DEVICE_TABLE(of, ath10k_snoc_dt_match);
 static int ath10k_snoc_probe(struct platform_device *pdev)
 {
 	const struct ath10k_snoc_drv_priv *drv_data;
-	const struct of_device_id *of_id;
 	struct ath10k_snoc *ar_snoc;
 	struct device *dev;
 	struct ath10k *ar;
@@ -1474,15 +1473,13 @@ static int ath10k_snoc_probe(struct platform_device *pdev)
 	int ret;
 	u32 i;
 
-	of_id = of_match_device(ath10k_snoc_dt_match, &pdev->dev);
-	if (!of_id) {
-		dev_err(&pdev->dev, "failed to find matching device tree id\n");
+	dev = &pdev->dev;
+	drv_data = device_get_match_data(dev);
+	if (!drv_data) {
+		dev_err(dev, "failed to find matching device tree id\n");
 		return -EINVAL;
 	}
 
-	drv_data = of_id->data;
-	dev = &pdev->dev;
-
 	ret = dma_set_mask_and_coherent(dev, drv_data->dma_mask);
 	if (ret) {
 		dev_err(dev, "failed to set dma mask: %d\n", ret);
-- 
cgit v1.2.3-59-g8ed1b


From 5b362498a79631f283578b64bf6f4d15ed4cc19a Mon Sep 17 00:00:00 2001
From: Mert Dirik <mertdirik@gmail.com>
Date: Thu, 16 Jan 2020 14:11:25 +0300
Subject: ar5523: Add USB ID of SMCWUSBT-G2 wireless adapter

Add the required USB ID for running SMCWUSBT-G2 wireless adapter (SMC
"EZ Connect g").

This device uses ar5523 chipset and requires firmware to be loaded. Even
though pid of the device is 4507, this patch adds it as 4506 so that
AR5523_DEVICE_UG macro can set the AR5523_FLAG_PRE_FIRMWARE flag for pid
4507.

Signed-off-by: Mert Dirik <mertdirik@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ar5523/ar5523.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/wireless/ath/ar5523/ar5523.c b/drivers/net/wireless/ath/ar5523/ar5523.c
index da2d179430ca..4c57e79e5779 100644
--- a/drivers/net/wireless/ath/ar5523/ar5523.c
+++ b/drivers/net/wireless/ath/ar5523/ar5523.c
@@ -1770,6 +1770,8 @@ static const struct usb_device_id ar5523_id_table[] = {
 	AR5523_DEVICE_UX(0x0846, 0x4300),	/* Netgear / WG111U */
 	AR5523_DEVICE_UG(0x0846, 0x4250),	/* Netgear / WG111T */
 	AR5523_DEVICE_UG(0x0846, 0x5f00),	/* Netgear / WPN111 */
+	AR5523_DEVICE_UG(0x083a, 0x4506),	/* SMC / EZ Connect
+						   SMCWUSBT-G2 */
 	AR5523_DEVICE_UG(0x157e, 0x3006),	/* Umedia / AR5523_1 */
 	AR5523_DEVICE_UX(0x157e, 0x3205),	/* Umedia / AR5523_2 */
 	AR5523_DEVICE_UG(0x157e, 0x3006),	/* Umedia / TEW444UBEU */
-- 
cgit v1.2.3-59-g8ed1b


From d92e8fa8ce882107e3f5be1244ff77ea1a74e63e Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.king@canonical.com>
Date: Thu, 23 Jan 2020 00:18:01 +0000
Subject: ar5523: fix spelling mistake "to" -> "too"

There is a spelling mistake in a ar5523_err message. Fix it.

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/ar5523/ar5523.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ar5523/ar5523.c b/drivers/net/wireless/ath/ar5523/ar5523.c
index 4c57e79e5779..49cc4b7ed516 100644
--- a/drivers/net/wireless/ath/ar5523/ar5523.c
+++ b/drivers/net/wireless/ath/ar5523/ar5523.c
@@ -74,7 +74,7 @@ static void ar5523_read_reply(struct ar5523 *ar, struct ar5523_cmd_hdr *hdr,
 
 	if (cmd->odata) {
 		if (cmd->olen < olen) {
-			ar5523_err(ar, "olen to small %d < %d\n",
+			ar5523_err(ar, "olen too small %d < %d\n",
 				   cmd->olen, olen);
 			cmd->olen = 0;
 			cmd->res = -EOVERFLOW;
-- 
cgit v1.2.3-59-g8ed1b


From d7809bd9eae67b4252cbc4672431610227cbb729 Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.king@canonical.com>
Date: Thu, 23 Jan 2020 00:51:17 +0000
Subject: wcn36xx: fix spelling mistake "to" -> "too"

There is a spelling mistake in a wcn36xx_err message. Fix it.

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ath/wcn36xx/smd.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/wcn36xx/smd.c b/drivers/net/wireless/ath/wcn36xx/smd.c
index 523550f94a3f..77269ac7f352 100644
--- a/drivers/net/wireless/ath/wcn36xx/smd.c
+++ b/drivers/net/wireless/ath/wcn36xx/smd.c
@@ -1620,7 +1620,7 @@ int wcn36xx_smd_send_beacon(struct wcn36xx *wcn, struct ieee80211_vif *vif,
 	msg_body.beacon_length6 = msg_body.beacon_length + 6;
 
 	if (msg_body.beacon_length > BEACON_TEMPLATE_SIZE) {
-		wcn36xx_err("Beacon is to big: beacon size=%d\n",
+		wcn36xx_err("Beacon is too big: beacon size=%d\n",
 			      msg_body.beacon_length);
 		ret = -ENOMEM;
 		goto out;
-- 
cgit v1.2.3-59-g8ed1b


From cf33a7728bf27e6f8c5e9190410db85c11103699 Mon Sep 17 00:00:00 2001
From: Maital Hahn <maitalm@ti.com>
Date: Thu, 16 Jan 2020 08:33:53 -0600
Subject: wlcore: mesh: Add support for RX Broadcast Key

In order to support authentication of equals peers,
need to save RX Broadcast key per peer (on top of 1 TX broadcast key
and unicast key per peer).

Signed-off-by: Maital Hahn <maitalm@ti.com>
Acked-by: Guy Mishol <guym@ti.com>
Signed-off-by: Hari Nagalla <hnagalla@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/ti/wlcore/cmd.c      |  6 ++++--
 drivers/net/wireless/ti/wlcore/cmd.h      |  2 +-
 drivers/net/wireless/ti/wlcore/main.c     | 23 +++++++++++++++--------
 drivers/net/wireless/ti/wlcore/wlcore_i.h |  1 +
 4 files changed, 21 insertions(+), 11 deletions(-)

diff --git a/drivers/net/wireless/ti/wlcore/cmd.c b/drivers/net/wireless/ti/wlcore/cmd.c
index 2a48fc6ad56c..6ef8fc9ae627 100644
--- a/drivers/net/wireless/ti/wlcore/cmd.c
+++ b/drivers/net/wireless/ti/wlcore/cmd.c
@@ -1429,7 +1429,7 @@ out:
 int wl1271_cmd_set_ap_key(struct wl1271 *wl, struct wl12xx_vif *wlvif,
 			  u16 action, u8 id, u8 key_type,
 			  u8 key_size, const u8 *key, u8 hlid, u32 tx_seq_32,
-			  u16 tx_seq_16)
+			  u16 tx_seq_16, bool is_pairwise)
 {
 	struct wl1271_cmd_set_keys *cmd;
 	int ret = 0;
@@ -1444,8 +1444,10 @@ int wl1271_cmd_set_ap_key(struct wl1271 *wl, struct wl12xx_vif *wlvif,
 			lid_type = WEP_DEFAULT_LID_TYPE;
 		else
 			lid_type = BROADCAST_LID_TYPE;
-	} else {
+	} else if (is_pairwise) {
 		lid_type = UNICAST_LID_TYPE;
+	} else {
+		lid_type = BROADCAST_LID_TYPE;
 	}
 
 	wl1271_debug(DEBUG_CRYPT, "ap key action: %d id: %d lid: %d type: %d"
diff --git a/drivers/net/wireless/ti/wlcore/cmd.h b/drivers/net/wireless/ti/wlcore/cmd.h
index 084375ba4abf..bfad7b5a1ac6 100644
--- a/drivers/net/wireless/ti/wlcore/cmd.h
+++ b/drivers/net/wireless/ti/wlcore/cmd.h
@@ -65,7 +65,7 @@ int wl1271_cmd_set_sta_key(struct wl1271 *wl, struct wl12xx_vif *wlvif,
 int wl1271_cmd_set_ap_key(struct wl1271 *wl, struct wl12xx_vif *wlvif,
 			  u16 action, u8 id, u8 key_type,
 			  u8 key_size, const u8 *key, u8 hlid, u32 tx_seq_32,
-			  u16 tx_seq_16);
+			  u16 tx_seq_16, bool is_pairwise);
 int wl12xx_cmd_set_peer_state(struct wl1271 *wl, struct wl12xx_vif *wlvif,
 			      u8 hlid);
 int wl12xx_roc(struct wl1271 *wl, struct wl12xx_vif *wlvif, u8 role_id,
diff --git a/drivers/net/wireless/ti/wlcore/main.c b/drivers/net/wireless/ti/wlcore/main.c
index e994995d79ab..ed049c9f7e29 100644
--- a/drivers/net/wireless/ti/wlcore/main.c
+++ b/drivers/net/wireless/ti/wlcore/main.c
@@ -3273,7 +3273,7 @@ out:
 static int wl1271_record_ap_key(struct wl1271 *wl, struct wl12xx_vif *wlvif,
 				u8 id, u8 key_type, u8 key_size,
 				const u8 *key, u8 hlid, u32 tx_seq_32,
-				u16 tx_seq_16)
+				u16 tx_seq_16, bool is_pairwise)
 {
 	struct wl1271_ap_key *ap_key;
 	int i;
@@ -3311,6 +3311,7 @@ static int wl1271_record_ap_key(struct wl1271 *wl, struct wl12xx_vif *wlvif,
 	ap_key->hlid = hlid;
 	ap_key->tx_seq_32 = tx_seq_32;
 	ap_key->tx_seq_16 = tx_seq_16;
+	ap_key->is_pairwise = is_pairwise;
 
 	wlvif->ap.recorded_keys[i] = ap_key;
 	return 0;
@@ -3346,7 +3347,7 @@ static int wl1271_ap_init_hwenc(struct wl1271 *wl, struct wl12xx_vif *wlvif)
 					    key->id, key->key_type,
 					    key->key_size, key->key,
 					    hlid, key->tx_seq_32,
-					    key->tx_seq_16);
+					    key->tx_seq_16, key->is_pairwise);
 		if (ret < 0)
 			goto out;
 
@@ -3369,7 +3370,8 @@ out:
 static int wl1271_set_key(struct wl1271 *wl, struct wl12xx_vif *wlvif,
 		       u16 action, u8 id, u8 key_type,
 		       u8 key_size, const u8 *key, u32 tx_seq_32,
-		       u16 tx_seq_16, struct ieee80211_sta *sta)
+		       u16 tx_seq_16, struct ieee80211_sta *sta,
+		       bool is_pairwise)
 {
 	int ret;
 	bool is_ap = (wlvif->bss_type == BSS_TYPE_AP_BSS);
@@ -3396,12 +3398,12 @@ static int wl1271_set_key(struct wl1271 *wl, struct wl12xx_vif *wlvif,
 			ret = wl1271_record_ap_key(wl, wlvif, id,
 					     key_type, key_size,
 					     key, hlid, tx_seq_32,
-					     tx_seq_16);
+					     tx_seq_16, is_pairwise);
 		} else {
 			ret = wl1271_cmd_set_ap_key(wl, wlvif, action,
 					     id, key_type, key_size,
 					     key, hlid, tx_seq_32,
-					     tx_seq_16);
+					     tx_seq_16, is_pairwise);
 		}
 
 		if (ret < 0)
@@ -3501,6 +3503,7 @@ int wlcore_set_key(struct wl1271 *wl, enum set_key_cmd cmd,
 	u16 tx_seq_16 = 0;
 	u8 key_type;
 	u8 hlid;
+	bool is_pairwise;
 
 	wl1271_debug(DEBUG_MAC80211, "mac80211 set key");
 
@@ -3550,12 +3553,14 @@ int wlcore_set_key(struct wl1271 *wl, enum set_key_cmd cmd,
 		return -EOPNOTSUPP;
 	}
 
+	is_pairwise = key_conf->flags & IEEE80211_KEY_FLAG_PAIRWISE;
+
 	switch (cmd) {
 	case SET_KEY:
 		ret = wl1271_set_key(wl, wlvif, KEY_ADD_OR_REPLACE,
 				 key_conf->keyidx, key_type,
 				 key_conf->keylen, key_conf->key,
-				 tx_seq_32, tx_seq_16, sta);
+				 tx_seq_32, tx_seq_16, sta, is_pairwise);
 		if (ret < 0) {
 			wl1271_error("Could not add or replace key");
 			return ret;
@@ -3581,7 +3586,7 @@ int wlcore_set_key(struct wl1271 *wl, enum set_key_cmd cmd,
 		ret = wl1271_set_key(wl, wlvif, KEY_REMOVE,
 				     key_conf->keyidx, key_type,
 				     key_conf->keylen, key_conf->key,
-				     0, 0, sta);
+				     0, 0, sta, is_pairwise);
 		if (ret < 0) {
 			wl1271_error("Could not remove key");
 			return ret;
@@ -6223,6 +6228,7 @@ static int wl1271_init_ieee80211(struct wl1271 *wl)
 
 	ieee80211_hw_set(wl->hw, SUPPORT_FAST_XMIT);
 	ieee80211_hw_set(wl->hw, CHANCTX_STA_CSA);
+	ieee80211_hw_set(wl->hw, SUPPORTS_PER_STA_GTK);
 	ieee80211_hw_set(wl->hw, QUEUE_CONTROL);
 	ieee80211_hw_set(wl->hw, TX_AMPDU_SETUP_IN_HW);
 	ieee80211_hw_set(wl->hw, AMPDU_AGGREGATION);
@@ -6267,7 +6273,8 @@ static int wl1271_init_ieee80211(struct wl1271 *wl)
 
 	wl->hw->wiphy->flags |= WIPHY_FLAG_AP_UAPSD |
 				WIPHY_FLAG_HAS_REMAIN_ON_CHANNEL |
-				WIPHY_FLAG_HAS_CHANNEL_SWITCH;
+				WIPHY_FLAG_HAS_CHANNEL_SWITCH |
++				WIPHY_FLAG_IBSS_RSN;
 
 	wl->hw->wiphy->features |= NL80211_FEATURE_AP_SCAN;
 
diff --git a/drivers/net/wireless/ti/wlcore/wlcore_i.h b/drivers/net/wireless/ti/wlcore/wlcore_i.h
index 6fab60b0e367..eefae3f867b9 100644
--- a/drivers/net/wireless/ti/wlcore/wlcore_i.h
+++ b/drivers/net/wireless/ti/wlcore/wlcore_i.h
@@ -212,6 +212,7 @@ struct wl1271_ap_key {
 	u8 hlid;
 	u32 tx_seq_32;
 	u16 tx_seq_16;
+	bool is_pairwise;
 };
 
 enum wl12xx_flags {
-- 
cgit v1.2.3-59-g8ed1b


From b61156fba74f659d0bc2de8f2dbf5bad9f4b8faf Mon Sep 17 00:00:00 2001
From: Nathan Chancellor <natechancellor@gmail.com>
Date: Tue, 17 Dec 2019 18:15:46 -0700
Subject: hostap: Adjust indentation in prism2_hostapd_add_sta

Clang warns:

../drivers/net/wireless/intersil/hostap/hostap_ap.c:2511:3: warning:
misleading indentation; statement is not part of the previous 'if'
[-Wmisleading-indentation]
        if (sta->tx_supp_rates & WLAN_RATE_5M5)
        ^
../drivers/net/wireless/intersil/hostap/hostap_ap.c:2509:2: note:
previous statement is here
        if (sta->tx_supp_rates & WLAN_RATE_2M)
        ^
1 warning generated.

This warning occurs because there is a space before the tab on this
line. Remove it so that the indentation is consistent with the Linux
kernel coding style and clang no longer warns.

Fixes: ff1d2767d5a4 ("Add HostAP wireless driver.")
Link: https://github.com/ClangBuiltLinux/linux/issues/813
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/intersil/hostap/hostap_ap.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intersil/hostap/hostap_ap.c b/drivers/net/wireless/intersil/hostap/hostap_ap.c
index 0094b1d2b577..3ec46f48cfde 100644
--- a/drivers/net/wireless/intersil/hostap/hostap_ap.c
+++ b/drivers/net/wireless/intersil/hostap/hostap_ap.c
@@ -2508,7 +2508,7 @@ static int prism2_hostapd_add_sta(struct ap_data *ap,
 		sta->supported_rates[0] = 2;
 	if (sta->tx_supp_rates & WLAN_RATE_2M)
 		sta->supported_rates[1] = 4;
- 	if (sta->tx_supp_rates & WLAN_RATE_5M5)
+	if (sta->tx_supp_rates & WLAN_RATE_5M5)
 		sta->supported_rates[2] = 11;
 	if (sta->tx_supp_rates & WLAN_RATE_11M)
 		sta->supported_rates[3] = 22;
-- 
cgit v1.2.3-59-g8ed1b


From 25cab7e71da32d46651f5fc9373eef9b79938629 Mon Sep 17 00:00:00 2001
From: Chin-Yen Lee <timlee@realtek.com>
Date: Thu, 19 Dec 2019 16:58:10 +0800
Subject: rtw88: pci: reset ring index when release skbs in tx ring

When skbs queued for each TX ring are relased in PCI stop flow,
the ring index should be reset at the same time.

Signed-off-by: Chin-Yen Lee <timlee@realtek.com>
Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtw88/pci.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wireless/realtek/rtw88/pci.c b/drivers/net/wireless/realtek/rtw88/pci.c
index a58e8276a41a..068f1bec88e6 100644
--- a/drivers/net/wireless/realtek/rtw88/pci.c
+++ b/drivers/net/wireless/realtek/rtw88/pci.c
@@ -506,6 +506,7 @@ static void rtw_pci_dma_release(struct rtw_dev *rtwdev, struct rtw_pci *rtwpci)
 	struct rtw_pci_tx_ring *tx_ring;
 	u8 queue;
 
+	rtw_pci_reset_trx_ring(rtwdev);
 	for (queue = 0; queue < RTK_MAX_TX_QUEUE_NUM; queue++) {
 		tx_ring = &rtwpci->tx_rings[queue];
 		rtw_pci_free_tx_ring_skbs(rtwdev, tx_ring);
-- 
cgit v1.2.3-59-g8ed1b


From fd30e891e0bc68735b52d480d078959e5fb30e70 Mon Sep 17 00:00:00 2001
From: Chin-Yen Lee <timlee@realtek.com>
Date: Thu, 19 Dec 2019 16:58:11 +0800
Subject: rtw88: pci: reset dma when reset pci trx ring

When PCI trx rings are reset, the DMA engine should also be reset.
Otherswise, the rx_tag of rx flow is not synchronous to hw.

Remove DMA reset when rtw_pci_start() as we added it in
rtw_pci_setup().

Signed-off-by: Chin-Yen Lee <timlee@realtek.com>
Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
Reviewed-by: Chris Chiu <chiu@endlessm.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtw88/pci.c | 19 ++++++++++---------
 1 file changed, 10 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw88/pci.c b/drivers/net/wireless/realtek/rtw88/pci.c
index 068f1bec88e6..c394df494c95 100644
--- a/drivers/net/wireless/realtek/rtw88/pci.c
+++ b/drivers/net/wireless/realtek/rtw88/pci.c
@@ -486,13 +486,6 @@ static void rtw_pci_disable_interrupt(struct rtw_dev *rtwdev,
 	rtwpci->irq_enabled = false;
 }
 
-static int rtw_pci_setup(struct rtw_dev *rtwdev)
-{
-	rtw_pci_reset_trx_ring(rtwdev);
-
-	return 0;
-}
-
 static void rtw_pci_dma_reset(struct rtw_dev *rtwdev, struct rtw_pci *rtwpci)
 {
 	/* reset dma and rx tag */
@@ -501,6 +494,16 @@ static void rtw_pci_dma_reset(struct rtw_dev *rtwdev, struct rtw_pci *rtwpci)
 	rtwpci->rx_tag = 0;
 }
 
+static int rtw_pci_setup(struct rtw_dev *rtwdev)
+{
+	struct rtw_pci *rtwpci = (struct rtw_pci *)rtwdev->priv;
+
+	rtw_pci_reset_trx_ring(rtwdev);
+	rtw_pci_dma_reset(rtwdev, rtwpci);
+
+	return 0;
+}
+
 static void rtw_pci_dma_release(struct rtw_dev *rtwdev, struct rtw_pci *rtwpci)
 {
 	struct rtw_pci_tx_ring *tx_ring;
@@ -518,8 +521,6 @@ static int rtw_pci_start(struct rtw_dev *rtwdev)
 	struct rtw_pci *rtwpci = (struct rtw_pci *)rtwdev->priv;
 	unsigned long flags;
 
-	rtw_pci_dma_reset(rtwdev, rtwpci);
-
 	spin_lock_irqsave(&rtwpci->irq_lock, flags);
 	rtw_pci_enable_interrupt(rtwdev, rtwpci);
 	spin_unlock_irqrestore(&rtwpci->irq_lock, flags);
-- 
cgit v1.2.3-59-g8ed1b


From 786221048e8f74bafb87b4ba8b613cd5bc0ea159 Mon Sep 17 00:00:00 2001
From: Yan-Hsuan Chuang <yhchuang@realtek.com>
Date: Thu, 19 Dec 2019 16:58:12 +0800
Subject: rtw88: add interface config for 8822c

Some devices need to configure interface/HCI related reigsters
in power on flow. Add interface_cfg for HCI for the settings.

The driver only supports RTL8822BE/RTL8822CE now, and since
RTL8822BE does not need to configure PCIE, the configuration
is only added for RTL8822CE. Without it, some of the RTL8822CE
device can crash and disconnected to host in suspend/wowlan mode.

Signed-off-by: Tzu-En Huang <tehuang@realtek.com>
Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtw88/hci.h |  6 ++++++
 drivers/net/wireless/realtek/rtw88/mac.c |  2 ++
 drivers/net/wireless/realtek/rtw88/pci.c | 17 +++++++++++++++++
 drivers/net/wireless/realtek/rtw88/reg.h |  2 ++
 4 files changed, 27 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw88/hci.h b/drivers/net/wireless/realtek/rtw88/hci.h
index 3d91aea942c3..85a81a578fd5 100644
--- a/drivers/net/wireless/realtek/rtw88/hci.h
+++ b/drivers/net/wireless/realtek/rtw88/hci.h
@@ -15,6 +15,7 @@ struct rtw_hci_ops {
 	void (*stop)(struct rtw_dev *rtwdev);
 	void (*deep_ps)(struct rtw_dev *rtwdev, bool enter);
 	void (*link_ps)(struct rtw_dev *rtwdev, bool enter);
+	void (*interface_cfg)(struct rtw_dev *rtwdev);
 
 	int (*write_data_rsvd_page)(struct rtw_dev *rtwdev, u8 *buf, u32 size);
 	int (*write_data_h2c)(struct rtw_dev *rtwdev, u8 *buf, u32 size);
@@ -59,6 +60,11 @@ static inline void rtw_hci_link_ps(struct rtw_dev *rtwdev, bool enter)
 	rtwdev->hci.ops->link_ps(rtwdev, enter);
 }
 
+static inline void rtw_hci_interface_cfg(struct rtw_dev *rtwdev)
+{
+	rtwdev->hci.ops->interface_cfg(rtwdev);
+}
+
 static inline int
 rtw_hci_write_data_rsvd_page(struct rtw_dev *rtwdev, u8 *buf, u32 size)
 {
diff --git a/drivers/net/wireless/realtek/rtw88/mac.c b/drivers/net/wireless/realtek/rtw88/mac.c
index 507970387b2a..03a82de053ea 100644
--- a/drivers/net/wireless/realtek/rtw88/mac.c
+++ b/drivers/net/wireless/realtek/rtw88/mac.c
@@ -1034,5 +1034,7 @@ int rtw_mac_init(struct rtw_dev *rtwdev)
 	if (ret)
 		return ret;
 
+	rtw_hci_interface_cfg(rtwdev);
+
 	return 0;
 }
diff --git a/drivers/net/wireless/realtek/rtw88/pci.c b/drivers/net/wireless/realtek/rtw88/pci.c
index c394df494c95..a60d9b107006 100644
--- a/drivers/net/wireless/realtek/rtw88/pci.c
+++ b/drivers/net/wireless/realtek/rtw88/pci.c
@@ -6,6 +6,7 @@
 #include <linux/pci.h>
 #include "main.h"
 #include "pci.h"
+#include "reg.h"
 #include "tx.h"
 #include "rx.h"
 #include "fw.h"
@@ -1224,6 +1225,21 @@ static void rtw_pci_link_cfg(struct rtw_dev *rtwdev)
 	rtwpci->link_ctrl = link_ctrl;
 }
 
+static void rtw_pci_interface_cfg(struct rtw_dev *rtwdev)
+{
+	struct rtw_chip_info *chip = rtwdev->chip;
+
+	switch (chip->id) {
+	case RTW_CHIP_TYPE_8822C:
+		if (rtwdev->hal.cut_version >= RTW_CHIP_VER_CUT_D)
+			rtw_write32_mask(rtwdev, REG_HCI_MIX_CFG,
+					 BIT_PCIE_EMAC_PDN_AUX_TO_FAST_CLK, 1);
+		break;
+	default:
+		break;
+	}
+}
+
 static void rtw_pci_phy_cfg(struct rtw_dev *rtwdev)
 {
 	struct rtw_chip_info *chip = rtwdev->chip;
@@ -1332,6 +1348,7 @@ static struct rtw_hci_ops rtw_pci_ops = {
 	.stop = rtw_pci_stop,
 	.deep_ps = rtw_pci_deep_ps,
 	.link_ps = rtw_pci_link_ps,
+	.interface_cfg = rtw_pci_interface_cfg,
 
 	.read8 = rtw_pci_read8,
 	.read16 = rtw_pci_read16,
diff --git a/drivers/net/wireless/realtek/rtw88/reg.h b/drivers/net/wireless/realtek/rtw88/reg.h
index 7e817bc997eb..cd6b14c379ba 100644
--- a/drivers/net/wireless/realtek/rtw88/reg.h
+++ b/drivers/net/wireless/realtek/rtw88/reg.h
@@ -194,6 +194,8 @@
 #define REG_H2C_INFO		0x0254
 
 #define REG_INT_MIG		0x0304
+#define REG_HCI_MIX_CFG		0x03FC
+#define BIT_PCIE_EMAC_PDN_AUX_TO_FAST_CLK BIT(26)
 
 #define REG_FWHW_TXQ_CTRL	0x0420
 #define BIT_EN_BCNQ_DL		BIT(22)
-- 
cgit v1.2.3-59-g8ed1b


From c8e5695eae9959fc5774c0f490f2450be8bad3de Mon Sep 17 00:00:00 2001
From: Chin-Yen Lee <timlee@realtek.com>
Date: Thu, 19 Dec 2019 16:58:13 +0800
Subject: rtw88: load wowlan firmware if wowlan is supported

Driver used to download normal firmware only,
but some devices support wowlan and require to
download wowlan firmware when system suspends.
So modify rtw_load_firmware() and its callback to
allow driver to download both normal and wowlan firmware.

Signed-off-by: Chin-Yen Lee <timlee@realtek.com>
Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtw88/main.c | 72 +++++++++++++++++++++++++------
 drivers/net/wireless/realtek/rtw88/main.h | 10 +++++
 2 files changed, 70 insertions(+), 12 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw88/main.c b/drivers/net/wireless/realtek/rtw88/main.c
index ae61415e1665..54e8a0cef14b 100644
--- a/drivers/net/wireless/realtek/rtw88/main.c
+++ b/drivers/net/wireless/realtek/rtw88/main.c
@@ -793,6 +793,26 @@ void rtw_update_sta_info(struct rtw_dev *rtwdev, struct rtw_sta_info *si)
 	rtw_fw_send_ra_info(rtwdev, si);
 }
 
+static int rtw_wait_firmware_completion(struct rtw_dev *rtwdev)
+{
+	struct rtw_chip_info *chip = rtwdev->chip;
+	struct rtw_fw_state *fw;
+
+	fw = &rtwdev->fw;
+	wait_for_completion(&fw->completion);
+	if (!fw->firmware)
+		return -EINVAL;
+
+	if (chip->wow_fw_name) {
+		fw = &rtwdev->wow_fw;
+		wait_for_completion(&fw->completion);
+		if (!fw->firmware)
+			return -EINVAL;
+	}
+
+	return 0;
+}
+
 static int rtw_power_on(struct rtw_dev *rtwdev)
 {
 	struct rtw_chip_info *chip = rtwdev->chip;
@@ -813,11 +833,10 @@ static int rtw_power_on(struct rtw_dev *rtwdev)
 		goto err;
 	}
 
-	wait_for_completion(&fw->completion);
-	if (!fw->firmware) {
-		ret = -EINVAL;
-		rtw_err(rtwdev, "failed to load firmware\n");
-		goto err;
+	ret = rtw_wait_firmware_completion(rtwdev);
+	if (ret) {
+		rtw_err(rtwdev, "failed to wait firmware completion\n");
+		goto err_off;
 	}
 
 	ret = rtw_download_firmware(rtwdev, fw);
@@ -1020,8 +1039,8 @@ static void rtw_unset_supported_band(struct ieee80211_hw *hw,
 
 static void rtw_load_firmware_cb(const struct firmware *firmware, void *context)
 {
-	struct rtw_dev *rtwdev = context;
-	struct rtw_fw_state *fw = &rtwdev->fw;
+	struct rtw_fw_state *fw = context;
+	struct rtw_dev *rtwdev = fw->rtwdev;
 	const struct rtw_fw_hdr *fw_hdr;
 
 	if (!firmware || !firmware->data) {
@@ -1043,17 +1062,35 @@ static void rtw_load_firmware_cb(const struct firmware *firmware, void *context)
 		 fw->version, fw->sub_version, fw->sub_index, fw->h2c_version);
 }
 
-static int rtw_load_firmware(struct rtw_dev *rtwdev, const char *fw_name)
+static int rtw_load_firmware(struct rtw_dev *rtwdev, enum rtw_fw_type type)
 {
-	struct rtw_fw_state *fw = &rtwdev->fw;
+	const char *fw_name;
+	struct rtw_fw_state *fw;
 	int ret;
 
+	switch (type) {
+	case RTW_WOWLAN_FW:
+		fw = &rtwdev->wow_fw;
+		fw_name = rtwdev->chip->wow_fw_name;
+		break;
+
+	case RTW_NORMAL_FW:
+		fw = &rtwdev->fw;
+		fw_name = rtwdev->chip->fw_name;
+		break;
+
+	default:
+		rtw_warn(rtwdev, "unsupported firmware type\n");
+		return -ENOENT;
+	}
+
+	fw->rtwdev = rtwdev;
 	init_completion(&fw->completion);
 
 	ret = request_firmware_nowait(THIS_MODULE, true, fw_name, rtwdev->dev,
-				      GFP_KERNEL, rtwdev, rtw_load_firmware_cb);
+				      GFP_KERNEL, fw, rtw_load_firmware_cb);
 	if (ret) {
-		rtw_err(rtwdev, "async firmware request failed\n");
+		rtw_err(rtwdev, "failed to async firmware request\n");
 		return ret;
 	}
 
@@ -1372,12 +1409,19 @@ int rtw_core_init(struct rtw_dev *rtwdev)
 			  BIT_HTC_LOC_CTRL | BIT_APP_PHYSTS |
 			  BIT_AB | BIT_AM | BIT_APM;
 
-	ret = rtw_load_firmware(rtwdev, rtwdev->chip->fw_name);
+	ret = rtw_load_firmware(rtwdev, RTW_NORMAL_FW);
 	if (ret) {
 		rtw_warn(rtwdev, "no firmware loaded\n");
 		return ret;
 	}
 
+	if (chip->wow_fw_name) {
+		ret = rtw_load_firmware(rtwdev, RTW_WOWLAN_FW);
+		if (ret) {
+			rtw_warn(rtwdev, "no wow firmware loaded\n");
+			return ret;
+		}
+	}
 	return 0;
 }
 EXPORT_SYMBOL(rtw_core_init);
@@ -1385,12 +1429,16 @@ EXPORT_SYMBOL(rtw_core_init);
 void rtw_core_deinit(struct rtw_dev *rtwdev)
 {
 	struct rtw_fw_state *fw = &rtwdev->fw;
+	struct rtw_fw_state *wow_fw = &rtwdev->wow_fw;
 	struct rtw_rsvd_page *rsvd_pkt, *tmp;
 	unsigned long flags;
 
 	if (fw->firmware)
 		release_firmware(fw->firmware);
 
+	if (wow_fw->firmware)
+		release_firmware(wow_fw->firmware);
+
 	tasklet_kill(&rtwdev->tx_tasklet);
 	spin_lock_irqsave(&rtwdev->tx_report.q_lock, flags);
 	skb_queue_purge(&rtwdev->tx_report.queue);
diff --git a/drivers/net/wireless/realtek/rtw88/main.h b/drivers/net/wireless/realtek/rtw88/main.h
index d012eefcd0da..b7690666fd57 100644
--- a/drivers/net/wireless/realtek/rtw88/main.h
+++ b/drivers/net/wireless/realtek/rtw88/main.h
@@ -193,6 +193,11 @@ enum rtw_rx_queue_type {
 	RTK_MAX_RX_QUEUE_NUM
 };
 
+enum rtw_fw_type {
+	RTW_NORMAL_FW = 0x0,
+	RTW_WOWLAN_FW = 0x1,
+};
+
 enum rtw_rate_index {
 	RTW_RATEID_BGN_40M_2SS	= 0,
 	RTW_RATEID_BGN_40M_1SS	= 1,
@@ -1030,6 +1035,8 @@ struct rtw_chip_info {
 	u8 bfer_su_max_num;
 	u8 bfer_mu_max_num;
 
+	const char *wow_fw_name;
+
 	/* coex paras */
 	u32 coex_para_ver;
 	u8 bt_desired_ver;
@@ -1456,6 +1463,7 @@ struct rtw_fifo_conf {
 
 struct rtw_fw_state {
 	const struct firmware *firmware;
+	struct rtw_dev *rtwdev;
 	struct completion completion;
 	u16 version;
 	u8 sub_version;
@@ -1580,6 +1588,8 @@ struct rtw_dev {
 
 	u8 mp_mode;
 
+	struct rtw_fw_state wow_fw;
+
 	/* hci related data, must be last */
 	u8 priv[0] __aligned(sizeof(void *));
 };
-- 
cgit v1.2.3-59-g8ed1b


From 44bc17f7f5b3b2cc4084eba6307ba750078a8a73 Mon Sep 17 00:00:00 2001
From: Chin-Yen Lee <timlee@realtek.com>
Date: Thu, 19 Dec 2019 16:58:14 +0800
Subject: rtw88: support wowlan feature for 8822c

Wake on WLAN(wowlan) is a feature which allows devices
to be woken up from suspend state through wlan events.

When user enables wowlan feature and then let the device
enter suspend state, wowlan firmware will be loaded by
the driver and periodically monitors wifi packets.
Power consumption of wifi chip will be reduced in this
state.

If wowlan firmware detects that specific wlan event
happens, it will issue wakeup signal to trigger resume
process. Driver will load normal firmware and let wifi
chip return to the original state.

Currently supported wlan events include receiving magic packet,
rekey packet and deauth packet, and disconnecting from AP.

Signed-off-by: Chin-Yen Lee <timlee@realtek.com>
Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtw88/Makefile   |   1 +
 drivers/net/wireless/realtek/rtw88/debug.h    |   1 +
 drivers/net/wireless/realtek/rtw88/fw.c       |  86 +++++
 drivers/net/wireless/realtek/rtw88/fw.h       |  69 ++++
 drivers/net/wireless/realtek/rtw88/mac80211.c |  44 +++
 drivers/net/wireless/realtek/rtw88/main.c     |   3 +
 drivers/net/wireless/realtek/rtw88/main.h     |  18 +
 drivers/net/wireless/realtek/rtw88/pci.c      |  18 +
 drivers/net/wireless/realtek/rtw88/reg.h      |  13 +
 drivers/net/wireless/realtek/rtw88/rtw8822c.c |  12 +
 drivers/net/wireless/realtek/rtw88/util.h     |   2 +
 drivers/net/wireless/realtek/rtw88/wow.c      | 531 ++++++++++++++++++++++++++
 drivers/net/wireless/realtek/rtw88/wow.h      |  38 ++
 13 files changed, 836 insertions(+)
 create mode 100644 drivers/net/wireless/realtek/rtw88/wow.c
 create mode 100644 drivers/net/wireless/realtek/rtw88/wow.h

diff --git a/drivers/net/wireless/realtek/rtw88/Makefile b/drivers/net/wireless/realtek/rtw88/Makefile
index 15e12155a04c..cac148d13cf1 100644
--- a/drivers/net/wireless/realtek/rtw88/Makefile
+++ b/drivers/net/wireless/realtek/rtw88/Makefile
@@ -15,6 +15,7 @@ rtw88-y += main.o \
 	   ps.o \
 	   sec.o \
 	   bf.o \
+	   wow.o \
 	   regd.o
 
 rtw88-$(CONFIG_RTW88_8822BE)	+= rtw8822b.o rtw8822b_table.o
diff --git a/drivers/net/wireless/realtek/rtw88/debug.h b/drivers/net/wireless/realtek/rtw88/debug.h
index cd28f675e9cb..a0f36f29b4a6 100644
--- a/drivers/net/wireless/realtek/rtw88/debug.h
+++ b/drivers/net/wireless/realtek/rtw88/debug.h
@@ -18,6 +18,7 @@ enum rtw_debug_mask {
 	RTW_DBG_DEBUGFS		= 0x00000200,
 	RTW_DBG_PS		= 0x00000400,
 	RTW_DBG_BF		= 0x00000800,
+	RTW_DBG_WOW		= 0x00001000,
 
 	RTW_DBG_ALL		= 0xffffffff
 };
diff --git a/drivers/net/wireless/realtek/rtw88/fw.c b/drivers/net/wireless/realtek/rtw88/fw.c
index b8c581161f61..16ff59576154 100644
--- a/drivers/net/wireless/realtek/rtw88/fw.c
+++ b/drivers/net/wireless/realtek/rtw88/fw.c
@@ -10,6 +10,7 @@
 #include "sec.h"
 #include "debug.h"
 #include "util.h"
+#include "wow.h"
 
 static void rtw_fw_c2h_cmd_handle_ext(struct rtw_dev *rtwdev,
 				      struct sk_buff *skb)
@@ -482,6 +483,91 @@ void rtw_fw_set_pwr_mode(struct rtw_dev *rtwdev)
 	rtw_fw_send_h2c_command(rtwdev, h2c_pkt);
 }
 
+void rtw_fw_set_keep_alive_cmd(struct rtw_dev *rtwdev, bool enable)
+{
+	u8 h2c_pkt[H2C_PKT_SIZE] = {0};
+	struct rtw_fw_wow_keep_alive_para mode = {
+		.adopt = true,
+		.pkt_type = KEEP_ALIVE_NULL_PKT,
+		.period = 5,
+	};
+
+	SET_H2C_CMD_ID_CLASS(h2c_pkt, H2C_CMD_KEEP_ALIVE);
+	SET_KEEP_ALIVE_ENABLE(h2c_pkt, enable);
+	SET_KEEP_ALIVE_ADOPT(h2c_pkt, mode.adopt);
+	SET_KEEP_ALIVE_PKT_TYPE(h2c_pkt, mode.pkt_type);
+	SET_KEEP_ALIVE_CHECK_PERIOD(h2c_pkt, mode.period);
+
+	rtw_fw_send_h2c_command(rtwdev, h2c_pkt);
+}
+
+void rtw_fw_set_disconnect_decision_cmd(struct rtw_dev *rtwdev, bool enable)
+{
+	struct rtw_wow_param *rtw_wow = &rtwdev->wow;
+	u8 h2c_pkt[H2C_PKT_SIZE] = {0};
+	struct rtw_fw_wow_disconnect_para mode = {
+		.adopt = true,
+		.period = 30,
+		.retry_count = 5,
+	};
+
+	SET_H2C_CMD_ID_CLASS(h2c_pkt, H2C_CMD_DISCONNECT_DECISION);
+
+	if (test_bit(RTW_WOW_FLAG_EN_DISCONNECT, rtw_wow->flags)) {
+		SET_DISCONNECT_DECISION_ENABLE(h2c_pkt, enable);
+		SET_DISCONNECT_DECISION_ADOPT(h2c_pkt, mode.adopt);
+		SET_DISCONNECT_DECISION_CHECK_PERIOD(h2c_pkt, mode.period);
+		SET_DISCONNECT_DECISION_TRY_PKT_NUM(h2c_pkt, mode.retry_count);
+	}
+
+	rtw_fw_send_h2c_command(rtwdev, h2c_pkt);
+}
+
+void rtw_fw_set_wowlan_ctrl_cmd(struct rtw_dev *rtwdev, bool enable)
+{
+	struct rtw_wow_param *rtw_wow = &rtwdev->wow;
+	u8 h2c_pkt[H2C_PKT_SIZE] = {0};
+
+	SET_H2C_CMD_ID_CLASS(h2c_pkt, H2C_CMD_WOWLAN);
+
+	SET_WOWLAN_FUNC_ENABLE(h2c_pkt, enable);
+	if (rtw_wow_mgd_linked(rtwdev)) {
+		if (test_bit(RTW_WOW_FLAG_EN_MAGIC_PKT, rtw_wow->flags))
+			SET_WOWLAN_MAGIC_PKT_ENABLE(h2c_pkt, enable);
+		if (test_bit(RTW_WOW_FLAG_EN_DISCONNECT, rtw_wow->flags))
+			SET_WOWLAN_DEAUTH_WAKEUP_ENABLE(h2c_pkt, enable);
+		if (test_bit(RTW_WOW_FLAG_EN_REKEY_PKT, rtw_wow->flags))
+			SET_WOWLAN_REKEY_WAKEUP_ENABLE(h2c_pkt, enable);
+	}
+
+	rtw_fw_send_h2c_command(rtwdev, h2c_pkt);
+}
+
+void rtw_fw_set_aoac_global_info_cmd(struct rtw_dev *rtwdev,
+				     u8 pairwise_key_enc,
+				     u8 group_key_enc)
+{
+	u8 h2c_pkt[H2C_PKT_SIZE] = {0};
+
+	SET_H2C_CMD_ID_CLASS(h2c_pkt, H2C_CMD_AOAC_GLOBAL_INFO);
+
+	SET_AOAC_GLOBAL_INFO_PAIRWISE_ENC_ALG(h2c_pkt, pairwise_key_enc);
+	SET_AOAC_GLOBAL_INFO_GROUP_ENC_ALG(h2c_pkt, group_key_enc);
+
+	rtw_fw_send_h2c_command(rtwdev, h2c_pkt);
+}
+
+void rtw_fw_set_remote_wake_ctrl_cmd(struct rtw_dev *rtwdev, bool enable)
+{
+	u8 h2c_pkt[H2C_PKT_SIZE] = {0};
+
+	SET_H2C_CMD_ID_CLASS(h2c_pkt, H2C_CMD_REMOTE_WAKE_CTRL);
+
+	SET_REMOTE_WAKECTRL_ENABLE(h2c_pkt, enable);
+
+	rtw_fw_send_h2c_command(rtwdev, h2c_pkt);
+}
+
 static u8 rtw_get_rsvd_page_location(struct rtw_dev *rtwdev,
 				     enum rtw_rsvd_packet_type type)
 {
diff --git a/drivers/net/wireless/realtek/rtw88/fw.h b/drivers/net/wireless/realtek/rtw88/fw.h
index 73d1b9ca8efc..9ffc1b1ea895 100644
--- a/drivers/net/wireless/realtek/rtw88/fw.h
+++ b/drivers/net/wireless/realtek/rtw88/fw.h
@@ -100,6 +100,23 @@ struct rtw_rsvd_page {
 	bool add_txdesc;
 };
 
+enum rtw_keep_alive_pkt_type {
+	KEEP_ALIVE_NULL_PKT = 0,
+	KEEP_ALIVE_ARP_RSP = 1,
+};
+
+struct rtw_fw_wow_keep_alive_para {
+	bool adopt;
+	u8 pkt_type;
+	u8 period;		/* unit: sec */
+};
+
+struct rtw_fw_wow_disconnect_para {
+	bool adopt;
+	u8 period;		/* unit: sec */
+	u8 retry_count;
+};
+
 struct rtw_fw_hdr {
 	__le16 signature;
 	u8 category;
@@ -198,6 +215,11 @@ static inline void rtw_h2c_pkt_set_header(u8 *h2c_pkt, u8 sub_id)
 #define H2C_CMD_QUERY_BT_MP_INFO	0x67
 #define H2C_CMD_BT_WIFI_CONTROL		0x69
 
+#define H2C_CMD_KEEP_ALIVE		0x03
+#define H2C_CMD_DISCONNECT_DECISION	0x04
+#define H2C_CMD_WOWLAN			0x80
+#define H2C_CMD_REMOTE_WAKE_CTRL	0x81
+#define H2C_CMD_AOAC_GLOBAL_INFO	0x82
 #define SET_H2C_CMD_ID_CLASS(h2c_pkt, value)				       \
 	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, GENMASK(7, 0))
 
@@ -301,6 +323,45 @@ static inline void rtw_h2c_pkt_set_header(u8 *h2c_pkt, u8 sub_id)
 #define SET_BT_WIFI_CONTROL_DATA5(h2c_pkt, value)                              \
 	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x01, value, GENMASK(23, 16))
 
+#define SET_KEEP_ALIVE_ENABLE(h2c_pkt, value)				       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, BIT(8))
+#define SET_KEEP_ALIVE_ADOPT(h2c_pkt, value)				       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, BIT(9))
+#define SET_KEEP_ALIVE_PKT_TYPE(h2c_pkt, value)				       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, BIT(10))
+#define SET_KEEP_ALIVE_CHECK_PERIOD(h2c_pkt, value)			       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, GENMASK(23, 16))
+
+#define SET_DISCONNECT_DECISION_ENABLE(h2c_pkt, value)			       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, BIT(8))
+#define SET_DISCONNECT_DECISION_ADOPT(h2c_pkt, value)			       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, BIT(9))
+#define SET_DISCONNECT_DECISION_CHECK_PERIOD(h2c_pkt, value)		       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, GENMASK(23, 16))
+#define SET_DISCONNECT_DECISION_TRY_PKT_NUM(h2c_pkt, value)		       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, GENMASK(31, 24))
+
+#define SET_WOWLAN_FUNC_ENABLE(h2c_pkt, value)				       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, BIT(8))
+#define SET_WOWLAN_MAGIC_PKT_ENABLE(h2c_pkt, value)			       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, BIT(10))
+#define SET_WOWLAN_UNICAST_PKT_ENABLE(h2c_pkt, value)			       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, BIT(11))
+#define SET_WOWLAN_REKEY_WAKEUP_ENABLE(h2c_pkt, value)			       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, BIT(14))
+#define SET_WOWLAN_DEAUTH_WAKEUP_ENABLE(h2c_pkt, value)			       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, BIT(15))
+
+#define SET_REMOTE_WAKECTRL_ENABLE(h2c_pkt, value)			       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, BIT(8))
+#define SET_REMOTE_WAKE_CTRL_NLO_OFFLOAD_EN(h2c_pkt, value)		       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, BIT(12))
+
+#define SET_AOAC_GLOBAL_INFO_PAIRWISE_ENC_ALG(h2c_pkt, value)		       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, GENMASK(15, 8))
+#define SET_AOAC_GLOBAL_INFO_GROUP_ENC_ALG(h2c_pkt, value)		       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, GENMASK(23, 16))
+
 static inline struct rtw_c2h_cmd *get_c2h_from_skb(struct sk_buff *skb)
 {
 	u32 pkt_offset;
@@ -340,4 +401,12 @@ int rtw_fw_download_rsvd_page(struct rtw_dev *rtwdev,
 void rtw_send_rsvd_page_h2c(struct rtw_dev *rtwdev);
 int rtw_dump_drv_rsvd_page(struct rtw_dev *rtwdev,
 			   u32 offset, u32 size, u32 *buf);
+void rtw_fw_set_remote_wake_ctrl_cmd(struct rtw_dev *rtwdev, bool enable);
+void rtw_fw_set_wowlan_ctrl_cmd(struct rtw_dev *rtwdev, bool enable);
+void rtw_fw_set_keep_alive_cmd(struct rtw_dev *rtwdev, bool enable);
+void rtw_fw_set_disconnect_decision_cmd(struct rtw_dev *rtwdev, bool enable);
+void rtw_fw_set_aoac_global_info_cmd(struct rtw_dev *rtwdev,
+				     u8 pairwise_key_enc,
+				     u8 group_key_enc);
+
 #endif
diff --git a/drivers/net/wireless/realtek/rtw88/mac80211.c b/drivers/net/wireless/realtek/rtw88/mac80211.c
index 34a1c3b53cd4..1150e42539a9 100644
--- a/drivers/net/wireless/realtek/rtw88/mac80211.c
+++ b/drivers/net/wireless/realtek/rtw88/mac80211.c
@@ -12,6 +12,7 @@
 #include "reg.h"
 #include "bf.h"
 #include "debug.h"
+#include "wow.h"
 
 static void rtw_ops_tx(struct ieee80211_hw *hw,
 		       struct ieee80211_tx_control *control,
@@ -735,6 +736,44 @@ static int rtw_ops_set_bitrate_mask(struct ieee80211_hw *hw,
 	return 0;
 }
 
+#ifdef CONFIG_PM
+static int rtw_ops_suspend(struct ieee80211_hw *hw,
+			   struct cfg80211_wowlan *wowlan)
+{
+	struct rtw_dev *rtwdev = hw->priv;
+	int ret;
+
+	mutex_lock(&rtwdev->mutex);
+	ret = rtw_wow_suspend(rtwdev, wowlan);
+	if (ret)
+		rtw_err(rtwdev, "failed to suspend for wow %d\n", ret);
+	mutex_unlock(&rtwdev->mutex);
+
+	return ret ? 1 : 0;
+}
+
+static int rtw_ops_resume(struct ieee80211_hw *hw)
+{
+	struct rtw_dev *rtwdev = hw->priv;
+	int ret;
+
+	mutex_lock(&rtwdev->mutex);
+	ret = rtw_wow_resume(rtwdev);
+	if (ret)
+		rtw_err(rtwdev, "failed to resume for wow %d\n", ret);
+	mutex_unlock(&rtwdev->mutex);
+
+	return ret ? 1 : 0;
+}
+
+static void rtw_ops_set_wakeup(struct ieee80211_hw *hw, bool enabled)
+{
+	struct rtw_dev *rtwdev = hw->priv;
+
+	device_set_wakeup_enable(rtwdev->dev, enabled);
+}
+#endif
+
 const struct ieee80211_ops rtw_ops = {
 	.tx			= rtw_ops_tx,
 	.wake_tx_queue		= rtw_ops_wake_tx_queue,
@@ -757,5 +796,10 @@ const struct ieee80211_ops rtw_ops = {
 	.sta_statistics		= rtw_ops_sta_statistics,
 	.flush			= rtw_ops_flush,
 	.set_bitrate_mask	= rtw_ops_set_bitrate_mask,
+#ifdef CONFIG_PM
+	.suspend		= rtw_ops_suspend,
+	.resume			= rtw_ops_resume,
+	.set_wakeup		= rtw_ops_set_wakeup,
+#endif
 };
 EXPORT_SYMBOL(rtw_ops);
diff --git a/drivers/net/wireless/realtek/rtw88/main.c b/drivers/net/wireless/realtek/rtw88/main.c
index 54e8a0cef14b..1ac48ccdb3c8 100644
--- a/drivers/net/wireless/realtek/rtw88/main.c
+++ b/drivers/net/wireless/realtek/rtw88/main.c
@@ -1493,6 +1493,9 @@ int rtw_register_hw(struct rtw_dev *rtwdev, struct ieee80211_hw *hw)
 
 	wiphy_ext_feature_set(hw->wiphy, NL80211_EXT_FEATURE_CAN_REPLACE_PTK0);
 
+#ifdef CONFIG_PM
+	hw->wiphy->wowlan = rtwdev->chip->wowlan_stub;
+#endif
 	rtw_set_supported_band(hw, rtwdev->chip);
 	SET_IEEE80211_PERM_ADDR(hw, rtwdev->efuse.addr);
 
diff --git a/drivers/net/wireless/realtek/rtw88/main.h b/drivers/net/wireless/realtek/rtw88/main.h
index b7690666fd57..e403d24adbcf 100644
--- a/drivers/net/wireless/realtek/rtw88/main.h
+++ b/drivers/net/wireless/realtek/rtw88/main.h
@@ -342,6 +342,7 @@ enum rtw_flags {
 	RTW_FLAG_LEISURE_PS_DEEP,
 	RTW_FLAG_DIG_DISABLE,
 	RTW_FLAG_BUSY_TRAFFIC,
+	RTW_FLAG_WOWLAN,
 
 	NUM_OF_RTW_FLAGS,
 };
@@ -372,6 +373,15 @@ enum rtw_snr {
 	RTW_SNR_NUM
 };
 
+enum rtw_wow_flags {
+	RTW_WOW_FLAG_EN_MAGIC_PKT,
+	RTW_WOW_FLAG_EN_REKEY_PKT,
+	RTW_WOW_FLAG_EN_DISCONNECT,
+
+	/* keep it last */
+	RTW_WOW_FLAG_MAX,
+};
+
 /* the power index is represented by differences, which cck-1s & ht40-1s are
  * the base values, so for 1s's differences, there are only ht20 & ofdm
  */
@@ -907,6 +917,12 @@ struct rtw_intf_phy_para {
 	u16 platform;
 };
 
+struct rtw_wow_param {
+	struct ieee80211_vif *wow_vif;
+	DECLARE_BITMAP(flags, RTW_WOW_FLAG_MAX);
+	u8 txpause;
+};
+
 struct rtw_intf_phy_para_table {
 	struct rtw_intf_phy_para *usb2_para;
 	struct rtw_intf_phy_para *usb3_para;
@@ -1036,6 +1052,7 @@ struct rtw_chip_info {
 	u8 bfer_mu_max_num;
 
 	const char *wow_fw_name;
+	const struct wiphy_wowlan_support *wowlan_stub;
 
 	/* coex paras */
 	u32 coex_para_ver;
@@ -1589,6 +1606,7 @@ struct rtw_dev {
 	u8 mp_mode;
 
 	struct rtw_fw_state wow_fw;
+	struct rtw_wow_param wow;
 
 	/* hci related data, must be last */
 	u8 priv[0] __aligned(sizeof(void *));
diff --git a/drivers/net/wireless/realtek/rtw88/pci.c b/drivers/net/wireless/realtek/rtw88/pci.c
index a60d9b107006..f7770304ba42 100644
--- a/drivers/net/wireless/realtek/rtw88/pci.c
+++ b/drivers/net/wireless/realtek/rtw88/pci.c
@@ -1282,6 +1282,23 @@ static void rtw_pci_phy_cfg(struct rtw_dev *rtwdev)
 	rtw_pci_link_cfg(rtwdev);
 }
 
+#ifdef CONFIG_PM
+static int rtw_pci_suspend(struct device *dev)
+{
+	return 0;
+}
+
+static int rtw_pci_resume(struct device *dev)
+{
+	return 0;
+}
+
+static SIMPLE_DEV_PM_OPS(rtw_pm_ops, rtw_pci_suspend, rtw_pci_resume);
+#define RTW_PM_OPS (&rtw_pm_ops)
+#else
+#define RTW_PM_OPS NULL
+#endif
+
 static int rtw_pci_claim(struct rtw_dev *rtwdev, struct pci_dev *pdev)
 {
 	int ret;
@@ -1508,6 +1525,7 @@ static struct pci_driver rtw_pci_driver = {
 	.id_table = rtw_pci_id_table,
 	.probe = rtw_pci_probe,
 	.remove = rtw_pci_remove,
+	.driver.pm = RTW_PM_OPS,
 };
 module_pci_driver(rtw_pci_driver);
 
diff --git a/drivers/net/wireless/realtek/rtw88/reg.h b/drivers/net/wireless/realtek/rtw88/reg.h
index cd6b14c379ba..77f31420599f 100644
--- a/drivers/net/wireless/realtek/rtw88/reg.h
+++ b/drivers/net/wireless/realtek/rtw88/reg.h
@@ -9,6 +9,8 @@
 #define BIT_FEN_CPUEN		BIT(2)
 #define BIT_FEN_BB_GLB_RST	BIT(1)
 #define BIT_FEN_BB_RSTB		BIT(0)
+#define BIT_R_DIS_PRST		BIT(6)
+#define BIT_WLOCK_1C_B6		BIT(5)
 #define REG_SYS_PW_CTRL		0x0004
 #define REG_SYS_CLK_CTRL	0x0008
 #define BIT_CPU_CLK_EN		BIT(14)
@@ -160,8 +162,12 @@
 #define REG_CR			0x0100
 #define REG_TRXFF_BNDY		0x0114
 #define REG_RXFF_BNDY		0x011C
+#define REG_FE1IMR		0x0120
+#define BIT_FS_RXDONE		BIT(16)
 #define REG_PKTBUF_DBG_CTRL	0x0140
 #define REG_C2HEVT		0x01A0
+#define REG_MCUTST_II		0x01C4
+#define REG_WOWLAN_WAKE_REASON	0x01C7
 #define REG_HMETFR		0x01CC
 #define REG_HMEBOX0		0x01D0
 #define REG_HMEBOX1		0x01D4
@@ -192,11 +198,18 @@
 #define REG_H2C_TAIL		0x0248
 #define REG_H2C_READ_ADDR	0x024C
 #define REG_H2C_INFO		0x0254
+#define REG_RXPKT_NUM		0x0284
+#define BIT_RXDMA_REQ		BIT(19)
+#define BIT_RW_RELEASE		BIT(18)
+#define BIT_RXDMA_IDLE		BIT(17)
+#define REG_RXPKTNUM		0x02B0
 
 #define REG_INT_MIG		0x0304
 #define REG_HCI_MIX_CFG		0x03FC
 #define BIT_PCIE_EMAC_PDN_AUX_TO_FAST_CLK BIT(26)
 
+#define REG_BCNQ_INFO		0x0418
+#define BIT_MGQ_CPU_EMPTY	BIT(24)
 #define REG_FWHW_TXQ_CTRL	0x0420
 #define BIT_EN_BCNQ_DL		BIT(22)
 #define BIT_EN_WR_FREE_TAIL	BIT(20)
diff --git a/drivers/net/wireless/realtek/rtw88/rtw8822c.c b/drivers/net/wireless/realtek/rtw88/rtw8822c.c
index 174029836833..e10fe672f79c 100644
--- a/drivers/net/wireless/realtek/rtw88/rtw8822c.c
+++ b/drivers/net/wireless/realtek/rtw88/rtw8822c.c
@@ -4060,6 +4060,13 @@ static const struct rtw_pwr_track_tbl rtw8822c_rtw_pwr_track_tbl = {
 	.pwrtrk_2g_ccka_p = rtw8822c_pwrtrk_2g_cck_a_p,
 };
 
+#ifdef CONFIG_PM
+static const struct wiphy_wowlan_support rtw_wowlan_stub_8822c = {
+	.flags = WIPHY_WOWLAN_MAGIC_PKT | WIPHY_WOWLAN_GTK_REKEY_FAILURE |
+		 WIPHY_WOWLAN_DISCONNECT | WIPHY_WOWLAN_SUPPORTS_GTK_REKEY,
+};
+#endif
+
 struct rtw_chip_info rtw8822c_hw_spec = {
 	.ops = &rtw8822c_ops,
 	.id = RTW_CHIP_TYPE_8822C,
@@ -4106,6 +4113,10 @@ struct rtw_chip_info rtw8822c_hw_spec = {
 	.bfer_su_max_num = 2,
 	.bfer_mu_max_num = 1,
 
+#ifdef CONFIG_PM
+	.wow_fw_name = "rtw88/rtw8822c_wow_fw.bin",
+	.wowlan_stub = &rtw_wowlan_stub_8822c,
+#endif
 	.coex_para_ver = 0x19062706,
 	.bt_desired_ver = 0x6,
 	.scbd_support = true,
@@ -4135,3 +4146,4 @@ struct rtw_chip_info rtw8822c_hw_spec = {
 EXPORT_SYMBOL(rtw8822c_hw_spec);
 
 MODULE_FIRMWARE("rtw88/rtw8822c_fw.bin");
+MODULE_FIRMWARE("rtw88/rtw8822c_wow_fw.bin");
diff --git a/drivers/net/wireless/realtek/rtw88/util.h b/drivers/net/wireless/realtek/rtw88/util.h
index 7bd2843b0bce..41c10e7144df 100644
--- a/drivers/net/wireless/realtek/rtw88/util.h
+++ b/drivers/net/wireless/realtek/rtw88/util.h
@@ -15,6 +15,8 @@ struct rtw_dev;
 			IEEE80211_IFACE_ITER_NORMAL, iterator, data)
 #define rtw_iterate_stas_atomic(rtwdev, iterator, data)                        \
 	ieee80211_iterate_stations_atomic(rtwdev->hw, iterator, data)
+#define rtw_iterate_keys(rtwdev, vif, iterator, data)			       \
+	ieee80211_iter_keys(rtwdev->hw, vif, iterator, data)
 
 static inline u8 *get_hdr_bssid(struct ieee80211_hdr *hdr)
 {
diff --git a/drivers/net/wireless/realtek/rtw88/wow.c b/drivers/net/wireless/realtek/rtw88/wow.c
new file mode 100644
index 000000000000..70289bccd5e4
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtw88/wow.c
@@ -0,0 +1,531 @@
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+/* Copyright(c) 2018-2019  Realtek Corporation
+ */
+
+#include "main.h"
+#include "fw.h"
+#include "wow.h"
+#include "reg.h"
+#include "debug.h"
+#include "mac.h"
+#include "ps.h"
+
+static void rtw_wow_show_wakeup_reason(struct rtw_dev *rtwdev)
+{
+	u8 reason;
+
+	reason = rtw_read8(rtwdev, REG_WOWLAN_WAKE_REASON);
+
+	if (reason == RTW_WOW_RSN_RX_DEAUTH)
+		rtw_dbg(rtwdev, RTW_DBG_WOW, "WOW: Rx deauth\n");
+	else if (reason == RTW_WOW_RSN_DISCONNECT)
+		rtw_dbg(rtwdev, RTW_DBG_WOW, "WOW: AP is off\n");
+	else if (reason == RTW_WOW_RSN_RX_MAGIC_PKT)
+		rtw_dbg(rtwdev, RTW_DBG_WOW, "WOW: Rx magic packet\n");
+	else if (reason == RTW_WOW_RSN_RX_GTK_REKEY)
+		rtw_dbg(rtwdev, RTW_DBG_WOW, "WOW: Rx gtk rekey\n");
+	else if (reason == RTW_WOW_RSN_RX_PTK_REKEY)
+		rtw_dbg(rtwdev, RTW_DBG_WOW, "WOW: Rx ptk rekey\n");
+	else
+		rtw_warn(rtwdev, "Unknown wakeup reason %x\n", reason);
+}
+
+static void rtw_wow_bb_stop(struct rtw_dev *rtwdev)
+{
+	struct rtw_wow_param *rtw_wow = &rtwdev->wow;
+
+	/* wait 100ms for firmware to finish TX */
+	msleep(100);
+
+	if (!rtw_read32_mask(rtwdev, REG_BCNQ_INFO, BIT_MGQ_CPU_EMPTY))
+		rtw_warn(rtwdev, "Wrong status of MGQ_CPU empty!\n");
+
+	rtw_wow->txpause = rtw_read8(rtwdev, REG_TXPAUSE);
+	rtw_write8(rtwdev, REG_TXPAUSE, 0xff);
+	rtw_write8_clr(rtwdev, REG_SYS_FUNC_EN, BIT_FEN_BB_RSTB);
+}
+
+static void rtw_wow_bb_start(struct rtw_dev *rtwdev)
+{
+	struct rtw_wow_param *rtw_wow = &rtwdev->wow;
+
+	rtw_write8_set(rtwdev, REG_SYS_FUNC_EN, BIT_FEN_BB_RSTB);
+	rtw_write8(rtwdev, REG_TXPAUSE, rtw_wow->txpause);
+}
+
+static void rtw_wow_rx_dma_stop(struct rtw_dev *rtwdev)
+{
+	/* wait 100ms for HW to finish rx dma */
+	msleep(100);
+
+	rtw_write32_set(rtwdev, REG_RXPKT_NUM, BIT_RW_RELEASE);
+
+	if (!check_hw_ready(rtwdev, REG_RXPKT_NUM, BIT_RXDMA_IDLE, 1))
+		rtw_err(rtwdev, "failed to stop rx dma\n");
+}
+
+static void rtw_wow_rx_dma_start(struct rtw_dev *rtwdev)
+{
+	rtw_write32_clr(rtwdev, REG_RXPKT_NUM, BIT_RW_RELEASE);
+}
+
+static bool rtw_wow_check_fw_status(struct rtw_dev *rtwdev, bool wow_enable)
+{
+	bool ret;
+
+	/* wait 100ms for wow firmware to finish work */
+	msleep(100);
+
+	if (wow_enable) {
+		if (!rtw_read8(rtwdev, REG_WOWLAN_WAKE_REASON))
+			ret = 0;
+	} else {
+		if (rtw_read32_mask(rtwdev, REG_FE1IMR, BIT_FS_RXDONE) == 0 &&
+		    rtw_read32_mask(rtwdev, REG_RXPKT_NUM, BIT_RW_RELEASE) == 0)
+			ret = 0;
+	}
+
+	if (ret)
+		rtw_err(rtwdev, "failed to check wow status %s\n",
+			wow_enable ? "enabled" : "disabled");
+
+	return ret;
+}
+
+static void rtw_wow_fw_security_type_iter(struct ieee80211_hw *hw,
+					  struct ieee80211_vif *vif,
+					  struct ieee80211_sta *sta,
+					  struct ieee80211_key_conf *key,
+					  void *data)
+{
+	struct rtw_fw_key_type_iter_data *iter_data = data;
+	struct rtw_dev *rtwdev = hw->priv;
+	u8 hw_key_type;
+
+	if (vif != rtwdev->wow.wow_vif)
+		return;
+
+	switch (key->cipher) {
+	case WLAN_CIPHER_SUITE_WEP40:
+		hw_key_type = RTW_CAM_WEP40;
+		break;
+	case WLAN_CIPHER_SUITE_WEP104:
+		hw_key_type = RTW_CAM_WEP104;
+		break;
+	case WLAN_CIPHER_SUITE_TKIP:
+		hw_key_type = RTW_CAM_TKIP;
+		key->flags |= IEEE80211_KEY_FLAG_GENERATE_MMIC;
+		break;
+	case WLAN_CIPHER_SUITE_CCMP:
+		hw_key_type = RTW_CAM_AES;
+		key->flags |= IEEE80211_KEY_FLAG_SW_MGMT_TX;
+		break;
+	default:
+		rtw_err(rtwdev, "Unsupported key type for wowlan mode\n");
+		hw_key_type = 0;
+		break;
+	}
+
+	if (sta)
+		iter_data->pairwise_key_type = hw_key_type;
+	else
+		iter_data->group_key_type = hw_key_type;
+}
+
+static void rtw_wow_fw_security_type(struct rtw_dev *rtwdev)
+{
+	struct rtw_fw_key_type_iter_data data = {};
+	struct ieee80211_vif *wow_vif = rtwdev->wow.wow_vif;
+
+	data.rtwdev = rtwdev;
+	rtw_iterate_keys(rtwdev, wow_vif,
+			 rtw_wow_fw_security_type_iter, &data);
+	rtw_fw_set_aoac_global_info_cmd(rtwdev, data.pairwise_key_type,
+					data.group_key_type);
+}
+
+static int rtw_wow_fw_start(struct rtw_dev *rtwdev)
+{
+	if (rtw_wow_mgd_linked(rtwdev)) {
+		rtw_send_rsvd_page_h2c(rtwdev);
+		rtw_wow_fw_security_type(rtwdev);
+		rtw_fw_set_disconnect_decision_cmd(rtwdev, true);
+		rtw_fw_set_keep_alive_cmd(rtwdev, true);
+	}
+
+	rtw_fw_set_wowlan_ctrl_cmd(rtwdev, true);
+	rtw_fw_set_remote_wake_ctrl_cmd(rtwdev, true);
+
+	return rtw_wow_check_fw_status(rtwdev, true);
+}
+
+static int rtw_wow_fw_stop(struct rtw_dev *rtwdev)
+{
+	if (rtw_wow_mgd_linked(rtwdev)) {
+		rtw_fw_set_disconnect_decision_cmd(rtwdev, false);
+		rtw_fw_set_keep_alive_cmd(rtwdev, false);
+	}
+
+	rtw_fw_set_wowlan_ctrl_cmd(rtwdev, false);
+	rtw_fw_set_remote_wake_ctrl_cmd(rtwdev, false);
+
+	return rtw_wow_check_fw_status(rtwdev, false);
+}
+
+static void rtw_wow_avoid_reset_mac(struct rtw_dev *rtwdev)
+{
+	/* When resuming from wowlan mode, some hosts issue signal
+	 * (PCIE: PREST, USB: SE0RST) to device, and lead to reset
+	 * mac core. If it happens, the connection to AP will be lost.
+	 * Setting REG_RSV_CTRL Register can avoid this process.
+	 */
+	switch (rtw_hci_type(rtwdev)) {
+	case RTW_HCI_TYPE_PCIE:
+	case RTW_HCI_TYPE_USB:
+		rtw_write8(rtwdev, REG_RSV_CTRL, BIT_WLOCK_1C_B6);
+		rtw_write8(rtwdev, REG_RSV_CTRL,
+			   BIT_WLOCK_1C_B6 | BIT_R_DIS_PRST);
+		break;
+	default:
+		rtw_warn(rtwdev, "Unsupported hci type to disable reset MAC\n");
+		break;
+	}
+}
+
+static void rtw_wow_fw_media_status_iter(void *data, struct ieee80211_sta *sta)
+{
+	struct rtw_sta_info *si = (struct rtw_sta_info *)sta->drv_priv;
+	struct rtw_fw_media_status_iter_data *iter_data = data;
+	struct rtw_dev *rtwdev = iter_data->rtwdev;
+
+	rtw_fw_media_status_report(rtwdev, si->mac_id, iter_data->connect);
+}
+
+static void rtw_wow_fw_media_status(struct rtw_dev *rtwdev, bool connect)
+{
+	struct rtw_fw_media_status_iter_data data;
+
+	data.rtwdev = rtwdev;
+	data.connect = connect;
+
+	rtw_iterate_stas_atomic(rtwdev, rtw_wow_fw_media_status_iter, &data);
+}
+
+static void rtw_wow_config_linked_rsvd_page(struct rtw_dev *rtwdev)
+{
+	rtw_add_rsvd_page(rtwdev, RSVD_PS_POLL, true);
+	rtw_add_rsvd_page(rtwdev, RSVD_QOS_NULL, true);
+	rtw_add_rsvd_page(rtwdev, RSVD_NULL, true);
+	rtw_add_rsvd_page(rtwdev, RSVD_LPS_PG_DPK, true);
+	rtw_add_rsvd_page(rtwdev, RSVD_LPS_PG_INFO, true);
+}
+
+static void rtw_wow_config_rsvd_page(struct rtw_dev *rtwdev)
+{
+	rtw_reset_rsvd_page(rtwdev);
+
+	if (rtw_wow_mgd_linked(rtwdev))
+		rtw_wow_config_linked_rsvd_page(rtwdev);
+}
+
+static int rtw_wow_dl_fw_rsvd_page(struct rtw_dev *rtwdev)
+{
+	struct ieee80211_vif *wow_vif = rtwdev->wow.wow_vif;
+
+	rtw_wow_config_rsvd_page(rtwdev);
+
+	return rtw_fw_download_rsvd_page(rtwdev, wow_vif);
+}
+
+static int rtw_wow_swap_fw(struct rtw_dev *rtwdev, enum rtw_fw_type type)
+{
+	struct rtw_fw_state *fw;
+	int ret;
+
+	switch (type) {
+	case RTW_WOWLAN_FW:
+		fw = &rtwdev->wow_fw;
+		break;
+
+	case RTW_NORMAL_FW:
+		fw = &rtwdev->fw;
+		break;
+
+	default:
+		rtw_warn(rtwdev, "unsupported firmware type to swap\n");
+		return -ENOENT;
+	}
+
+	ret = rtw_download_firmware(rtwdev, fw);
+	if (ret)
+		goto out;
+
+	rtw_fw_send_general_info(rtwdev);
+	rtw_fw_send_phydm_info(rtwdev);
+	rtw_wow_fw_media_status(rtwdev, true);
+
+out:
+	return ret;
+}
+
+static int rtw_wow_leave_linked_ps(struct rtw_dev *rtwdev)
+{
+	if (!test_bit(RTW_FLAG_WOWLAN, rtwdev->flags))
+		cancel_delayed_work_sync(&rtwdev->watch_dog_work);
+
+	return 0;
+}
+
+static int rtw_wow_leave_ps(struct rtw_dev *rtwdev)
+{
+	int ret = 0;
+
+	if (rtw_wow_mgd_linked(rtwdev))
+		ret = rtw_wow_leave_linked_ps(rtwdev);
+
+	return ret;
+}
+
+static int rtw_wow_enter_linked_ps(struct rtw_dev *rtwdev)
+{
+	struct rtw_wow_param *rtw_wow = &rtwdev->wow;
+	struct ieee80211_vif *wow_vif = rtw_wow->wow_vif;
+	struct rtw_vif *rtwvif = (struct rtw_vif *)wow_vif->drv_priv;
+
+	rtw_enter_lps(rtwdev, rtwvif->port);
+
+	return 0;
+}
+
+static int rtw_wow_enter_ps(struct rtw_dev *rtwdev)
+{
+	int ret = 0;
+
+	if (rtw_wow_mgd_linked(rtwdev))
+		ret = rtw_wow_enter_linked_ps(rtwdev);
+
+	return ret;
+}
+
+static void rtw_wow_stop_trx(struct rtw_dev *rtwdev)
+{
+	rtw_wow_bb_stop(rtwdev);
+	rtw_wow_rx_dma_stop(rtwdev);
+}
+
+static int rtw_wow_start(struct rtw_dev *rtwdev)
+{
+	int ret;
+
+	ret = rtw_wow_fw_start(rtwdev);
+	if (ret)
+		goto out;
+
+	rtw_hci_stop(rtwdev);
+	rtw_wow_bb_start(rtwdev);
+	rtw_wow_avoid_reset_mac(rtwdev);
+
+out:
+	return ret;
+}
+
+static int rtw_wow_enable(struct rtw_dev *rtwdev)
+{
+	int ret = 0;
+
+	rtw_wow_stop_trx(rtwdev);
+
+	ret = rtw_wow_swap_fw(rtwdev, RTW_WOWLAN_FW);
+	if (ret) {
+		rtw_err(rtwdev, "failed to swap wow fw\n");
+		goto error;
+	}
+
+	set_bit(RTW_FLAG_WOWLAN, rtwdev->flags);
+
+	ret = rtw_wow_dl_fw_rsvd_page(rtwdev);
+	if (ret) {
+		rtw_err(rtwdev, "failed to download wowlan rsvd page\n");
+		goto error;
+	}
+
+	ret = rtw_wow_start(rtwdev);
+	if (ret) {
+		rtw_err(rtwdev, "failed to start wow\n");
+		goto error;
+	}
+
+	return ret;
+
+error:
+	clear_bit(RTW_FLAG_WOWLAN, rtwdev->flags);
+	return ret;
+}
+
+static int rtw_wow_stop(struct rtw_dev *rtwdev)
+{
+	int ret;
+
+	/* some HCI related registers will be reset after resume,
+	 * need to set them again.
+	 */
+	ret = rtw_hci_setup(rtwdev);
+	if (ret) {
+		rtw_err(rtwdev, "failed to setup hci\n");
+		return ret;
+	}
+
+	ret = rtw_hci_start(rtwdev);
+	if (ret) {
+		rtw_err(rtwdev, "failed to start hci\n");
+		return ret;
+	}
+
+	ret = rtw_wow_fw_stop(rtwdev);
+	if (ret)
+		rtw_err(rtwdev, "failed to stop wowlan fw\n");
+
+	rtw_wow_bb_stop(rtwdev);
+
+	return ret;
+}
+
+static void rtw_wow_resume_trx(struct rtw_dev *rtwdev)
+{
+	rtw_wow_rx_dma_start(rtwdev);
+	rtw_wow_bb_start(rtwdev);
+	ieee80211_queue_delayed_work(rtwdev->hw, &rtwdev->watch_dog_work,
+				     RTW_WATCH_DOG_DELAY_TIME);
+}
+
+static int rtw_wow_disable(struct rtw_dev *rtwdev)
+{
+	int ret;
+
+	clear_bit(RTW_FLAG_WOWLAN, rtwdev->flags);
+
+	ret = rtw_wow_stop(rtwdev);
+	if (ret) {
+		rtw_err(rtwdev, "failed to stop wow\n");
+		goto out;
+	}
+
+	ret = rtw_wow_swap_fw(rtwdev, RTW_NORMAL_FW);
+	if (ret) {
+		rtw_err(rtwdev, "failed to swap normal fw\n");
+		goto out;
+	}
+
+	ret = rtw_wow_dl_fw_rsvd_page(rtwdev);
+	if (ret)
+		rtw_err(rtwdev, "failed to download normal rsvd page\n");
+
+out:
+	rtw_wow_resume_trx(rtwdev);
+	return ret;
+}
+
+static void rtw_wow_vif_iter(void *data, u8 *mac, struct ieee80211_vif *vif)
+{
+	struct rtw_dev *rtwdev = data;
+	struct rtw_vif *rtwvif = (struct rtw_vif *)vif->drv_priv;
+	struct rtw_wow_param *rtw_wow = &rtwdev->wow;
+
+	/* Current wowlan function support setting of only one STATION vif.
+	 * So when one suitable vif is found, stop the iteration.
+	 */
+	if (rtw_wow->wow_vif || vif->type != NL80211_IFTYPE_STATION)
+		return;
+
+	switch (rtwvif->net_type) {
+	case RTW_NET_MGD_LINKED:
+		rtw_wow->wow_vif = vif;
+		break;
+	default:
+		break;
+	}
+}
+
+static int rtw_wow_set_wakeups(struct rtw_dev *rtwdev,
+			       struct cfg80211_wowlan *wowlan)
+{
+	struct rtw_wow_param *rtw_wow = &rtwdev->wow;
+
+	if (wowlan->disconnect)
+		set_bit(RTW_WOW_FLAG_EN_DISCONNECT, rtw_wow->flags);
+	if (wowlan->magic_pkt)
+		set_bit(RTW_WOW_FLAG_EN_MAGIC_PKT, rtw_wow->flags);
+	if (wowlan->gtk_rekey_failure)
+		set_bit(RTW_WOW_FLAG_EN_REKEY_PKT, rtw_wow->flags);
+
+	rtw_iterate_vifs_atomic(rtwdev, rtw_wow_vif_iter, rtwdev);
+	if (!rtw_wow->wow_vif)
+		return -EPERM;
+
+	return 0;
+}
+
+static void rtw_wow_clear_wakeups(struct rtw_dev *rtwdev)
+{
+	struct rtw_wow_param *rtw_wow = &rtwdev->wow;
+
+	memset(rtw_wow, 0, sizeof(rtwdev->wow));
+}
+
+int rtw_wow_suspend(struct rtw_dev *rtwdev, struct cfg80211_wowlan *wowlan)
+{
+	int ret = 0;
+
+	ret = rtw_wow_set_wakeups(rtwdev, wowlan);
+	if (ret) {
+		rtw_err(rtwdev, "failed to set wakeup event\n");
+		goto out;
+	}
+
+	ret = rtw_wow_leave_ps(rtwdev);
+	if (ret) {
+		rtw_err(rtwdev, "failed to leave ps from normal mode\n");
+		goto out;
+	}
+
+	ret = rtw_wow_enable(rtwdev);
+	if (ret) {
+		rtw_err(rtwdev, "failed to enable wow\n");
+		goto out;
+	}
+
+	ret = rtw_wow_enter_ps(rtwdev);
+	if (ret)
+		rtw_err(rtwdev, "failed to enter ps for wow\n");
+
+out:
+	return ret;
+}
+
+int rtw_wow_resume(struct rtw_dev *rtwdev)
+{
+	int ret;
+
+	/* If wowlan mode is not enabled, do nothing */
+	if (!test_bit(RTW_FLAG_WOWLAN, rtwdev->flags)) {
+		rtw_err(rtwdev, "wow is not enabled\n");
+		ret = -EPERM;
+		goto out;
+	}
+
+	ret = rtw_wow_leave_ps(rtwdev);
+	if (ret) {
+		rtw_err(rtwdev, "failed to leave ps from wowlan mode\n");
+		goto out;
+	}
+
+	rtw_wow_show_wakeup_reason(rtwdev);
+
+	ret = rtw_wow_disable(rtwdev);
+	if (ret)
+		rtw_err(rtwdev, "failed to disable wow\n");
+
+out:
+	rtw_wow_clear_wakeups(rtwdev);
+	return ret;
+}
diff --git a/drivers/net/wireless/realtek/rtw88/wow.h b/drivers/net/wireless/realtek/rtw88/wow.h
new file mode 100644
index 000000000000..617806a2f28d
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtw88/wow.h
@@ -0,0 +1,38 @@
+/* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
+/* Copyright(c) 2018-2019  Realtek Corporation
+ */
+
+#ifndef __RTW_WOW_H__
+#define __RTW_WOW_H__
+
+enum rtw_wake_reason {
+	RTW_WOW_RSN_RX_PTK_REKEY = 0x1,
+	RTW_WOW_RSN_RX_GTK_REKEY = 0x2,
+	RTW_WOW_RSN_RX_DEAUTH = 0x8,
+	RTW_WOW_RSN_DISCONNECT = 0x10,
+	RTW_WOW_RSN_RX_MAGIC_PKT = 0x21,
+};
+
+struct rtw_fw_media_status_iter_data {
+	struct rtw_dev *rtwdev;
+	u8 connect;
+};
+
+struct rtw_fw_key_type_iter_data {
+	struct rtw_dev *rtwdev;
+	u8 group_key_type;
+	u8 pairwise_key_type;
+};
+
+static inline bool rtw_wow_mgd_linked(struct rtw_dev *rtwdev)
+{
+	struct ieee80211_vif *wow_vif = rtwdev->wow.wow_vif;
+	struct rtw_vif *rtwvif = (struct rtw_vif *)wow_vif->drv_priv;
+
+	return (rtwvif->net_type == RTW_NET_MGD_LINKED);
+}
+
+int rtw_wow_suspend(struct rtw_dev *rtwdev, struct cfg80211_wowlan *wowlan);
+int rtw_wow_resume(struct rtw_dev *rtwdev);
+
+#endif
-- 
cgit v1.2.3-59-g8ed1b


From e3e400df30078830871b927ca24826667d97d1f6 Mon Sep 17 00:00:00 2001
From: Chin-Yen Lee <timlee@realtek.com>
Date: Thu, 19 Dec 2019 16:58:15 +0800
Subject: rtw88: Add wowlan pattern match support

Pattern match is an option of wowlan to allow the device
to be woken up from suspend mode when receiving packets
matched user-designed patterns.

The patterns are written into hardware cam in suspend flow
if users have set up them. If packets matched designed
pattern are received, wowlan firmware will get an interrupt
and then wake up the device.

Signed-off-by: Chin-Yen Lee <timlee@realtek.com>
Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtw88/fw.c       |   6 +
 drivers/net/wireless/realtek/rtw88/fw.h       |   4 +
 drivers/net/wireless/realtek/rtw88/main.h     |  14 ++
 drivers/net/wireless/realtek/rtw88/reg.h      |  14 ++
 drivers/net/wireless/realtek/rtw88/rtw8822c.c |   3 +
 drivers/net/wireless/realtek/rtw88/wow.c      | 224 ++++++++++++++++++++++++++
 drivers/net/wireless/realtek/rtw88/wow.h      |   9 ++
 7 files changed, 274 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw88/fw.c b/drivers/net/wireless/realtek/rtw88/fw.c
index 16ff59576154..0523851cf98d 100644
--- a/drivers/net/wireless/realtek/rtw88/fw.c
+++ b/drivers/net/wireless/realtek/rtw88/fw.c
@@ -538,6 +538,8 @@ void rtw_fw_set_wowlan_ctrl_cmd(struct rtw_dev *rtwdev, bool enable)
 			SET_WOWLAN_DEAUTH_WAKEUP_ENABLE(h2c_pkt, enable);
 		if (test_bit(RTW_WOW_FLAG_EN_REKEY_PKT, rtw_wow->flags))
 			SET_WOWLAN_REKEY_WAKEUP_ENABLE(h2c_pkt, enable);
+		if (rtw_wow->pattern_cnt)
+			SET_WOWLAN_PATTERN_MATCH_ENABLE(h2c_pkt, enable);
 	}
 
 	rtw_fw_send_h2c_command(rtwdev, h2c_pkt);
@@ -596,6 +598,7 @@ void rtw_fw_set_pg_info(struct rtw_dev *rtwdev)
 	LPS_PG_INFO_LOC(h2c_pkt, loc_pg);
 	LPS_PG_DPK_LOC(h2c_pkt, loc_dpk);
 	LPS_PG_SEC_CAM_EN(h2c_pkt, conf->sec_cam_backup);
+	LPS_PG_PATTERN_CAM_EN(h2c_pkt, conf->pattern_cam_backup);
 
 	rtw_fw_send_h2c_command(rtwdev, h2c_pkt);
 }
@@ -677,6 +680,7 @@ static struct sk_buff *rtw_lps_pg_info_get(struct ieee80211_hw *hw,
 	struct rtw_chip_info *chip = rtwdev->chip;
 	struct rtw_lps_conf *conf = &rtwdev->lps_conf;
 	struct rtw_lps_pg_info_hdr *pg_info_hdr;
+	struct rtw_wow_param *rtw_wow = &rtwdev->wow;
 	struct sk_buff *skb;
 	u32 size;
 
@@ -691,8 +695,10 @@ static struct sk_buff *rtw_lps_pg_info_get(struct ieee80211_hw *hw,
 	pg_info_hdr->macid = find_first_bit(rtwdev->mac_id_map, RTW_MAX_MAC_ID_NUM);
 	pg_info_hdr->sec_cam_count =
 		rtw_sec_cam_pg_backup(rtwdev, pg_info_hdr->sec_cam);
+	pg_info_hdr->pattern_count = rtw_wow->pattern_cnt;
 
 	conf->sec_cam_backup = pg_info_hdr->sec_cam_count != 0;
+	conf->pattern_cam_backup = rtw_wow->pattern_cnt != 0;
 
 	return skb;
 }
diff --git a/drivers/net/wireless/realtek/rtw88/fw.h b/drivers/net/wireless/realtek/rtw88/fw.h
index 9ffc1b1ea895..9ea27cca11ac 100644
--- a/drivers/net/wireless/realtek/rtw88/fw.h
+++ b/drivers/net/wireless/realtek/rtw88/fw.h
@@ -246,6 +246,8 @@ static inline void rtw_h2c_pkt_set_header(u8 *h2c_pkt, u8 sub_id)
 	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, GENMASK(31, 24))
 #define LPS_PG_SEC_CAM_EN(h2c_pkt, value)                                      \
 	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, BIT(8))
+#define LPS_PG_PATTERN_CAM_EN(h2c_pkt, value)				       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, BIT(10))
 #define SET_RSSI_INFO_MACID(h2c_pkt, value)                                    \
 	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, GENMASK(15, 8))
 #define SET_RSSI_INFO_RSSI(h2c_pkt, value)                                     \
@@ -343,6 +345,8 @@ static inline void rtw_h2c_pkt_set_header(u8 *h2c_pkt, u8 sub_id)
 
 #define SET_WOWLAN_FUNC_ENABLE(h2c_pkt, value)				       \
 	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, BIT(8))
+#define SET_WOWLAN_PATTERN_MATCH_ENABLE(h2c_pkt, value)			       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, BIT(9))
 #define SET_WOWLAN_MAGIC_PKT_ENABLE(h2c_pkt, value)			       \
 	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, BIT(10))
 #define SET_WOWLAN_UNICAST_PKT_ENABLE(h2c_pkt, value)			       \
diff --git a/drivers/net/wireless/realtek/rtw88/main.h b/drivers/net/wireless/realtek/rtw88/main.h
index e403d24adbcf..74368ef954fa 100644
--- a/drivers/net/wireless/realtek/rtw88/main.h
+++ b/drivers/net/wireless/realtek/rtw88/main.h
@@ -19,6 +19,10 @@
 #define RTW_MAX_SEC_CAM_NUM		32
 #define MAX_PG_CAM_BACKUP_NUM		8
 
+#define RTW_MAX_PATTERN_NUM		12
+#define RTW_MAX_PATTERN_MASK_SIZE	16
+#define RTW_MAX_PATTERN_SIZE		128
+
 #define RTW_WATCH_DOG_DELAY_TIME	round_jiffies_relative(HZ * 2)
 
 #define RFREG_MASK			0xfffff
@@ -623,6 +627,7 @@ struct rtw_lps_conf {
 	u8 smart_ps;
 	u8 port_id;
 	bool sec_cam_backup;
+	bool pattern_cam_backup;
 };
 
 enum rtw_hw_key_type {
@@ -917,10 +922,19 @@ struct rtw_intf_phy_para {
 	u16 platform;
 };
 
+struct rtw_wow_pattern {
+	u16 crc;
+	u8 type;
+	u8 valid;
+	u8 mask[RTW_MAX_PATTERN_MASK_SIZE];
+};
+
 struct rtw_wow_param {
 	struct ieee80211_vif *wow_vif;
 	DECLARE_BITMAP(flags, RTW_WOW_FLAG_MAX);
 	u8 txpause;
+	u8 pattern_cnt;
+	struct rtw_wow_pattern patterns[RTW_MAX_PATTERN_NUM];
 };
 
 struct rtw_intf_phy_para_table {
diff --git a/drivers/net/wireless/realtek/rtw88/reg.h b/drivers/net/wireless/realtek/rtw88/reg.h
index 77f31420599f..9d94534c9674 100644
--- a/drivers/net/wireless/realtek/rtw88/reg.h
+++ b/drivers/net/wireless/realtek/rtw88/reg.h
@@ -338,6 +338,20 @@
 #define BIT_RFMOD_80M		BIT(8)
 #define BIT_RFMOD_40M		BIT(7)
 #define REG_WMAC_TRXPTCL_CTL_H	0x066C
+#define REG_WKFMCAM_CMD		0x0698
+#define BIT_WKFCAM_POLLING_V1	BIT(31)
+#define BIT_WKFCAM_CLR_V1	BIT(30)
+#define BIT_WKFCAM_WE		BIT(16)
+#define BIT_SHIFT_WKFCAM_ADDR_V2	8
+#define BIT_MASK_WKFCAM_ADDR_V2		0xff
+#define BIT_WKFCAM_ADDR_V2(x)						       \
+	(((x) & BIT_MASK_WKFCAM_ADDR_V2) << BIT_SHIFT_WKFCAM_ADDR_V2)
+#define REG_WKFMCAM_RWD         0x069C
+#define BIT_WKFMCAM_VALID	BIT(31)
+#define BIT_WKFMCAM_BC		BIT(26)
+#define BIT_WKFMCAM_MC		BIT(25)
+#define BIT_WKFMCAM_UC		BIT(24)
+
 #define REG_RXFLTMAP0		0x06A0
 #define REG_RXFLTMAP1		0x06A2
 #define REG_RXFLTMAP2		0x06A4
diff --git a/drivers/net/wireless/realtek/rtw88/rtw8822c.c b/drivers/net/wireless/realtek/rtw88/rtw8822c.c
index e10fe672f79c..4836c0044b05 100644
--- a/drivers/net/wireless/realtek/rtw88/rtw8822c.c
+++ b/drivers/net/wireless/realtek/rtw88/rtw8822c.c
@@ -4064,6 +4064,9 @@ static const struct rtw_pwr_track_tbl rtw8822c_rtw_pwr_track_tbl = {
 static const struct wiphy_wowlan_support rtw_wowlan_stub_8822c = {
 	.flags = WIPHY_WOWLAN_MAGIC_PKT | WIPHY_WOWLAN_GTK_REKEY_FAILURE |
 		 WIPHY_WOWLAN_DISCONNECT | WIPHY_WOWLAN_SUPPORTS_GTK_REKEY,
+	.n_patterns = RTW_MAX_PATTERN_NUM,
+	.pattern_max_len = RTW_MAX_PATTERN_SIZE,
+	.pattern_min_len = 1,
 };
 #endif
 
diff --git a/drivers/net/wireless/realtek/rtw88/wow.c b/drivers/net/wireless/realtek/rtw88/wow.c
index 70289bccd5e4..f4645f3c98e0 100644
--- a/drivers/net/wireless/realtek/rtw88/wow.c
+++ b/drivers/net/wireless/realtek/rtw88/wow.c
@@ -26,10 +26,220 @@ static void rtw_wow_show_wakeup_reason(struct rtw_dev *rtwdev)
 		rtw_dbg(rtwdev, RTW_DBG_WOW, "WOW: Rx gtk rekey\n");
 	else if (reason == RTW_WOW_RSN_RX_PTK_REKEY)
 		rtw_dbg(rtwdev, RTW_DBG_WOW, "WOW: Rx ptk rekey\n");
+	else if (reason == RTW_WOW_RSN_RX_PATTERN_MATCH)
+		rtw_dbg(rtwdev, RTW_DBG_WOW, "WOW: Rx pattern match packet\n");
 	else
 		rtw_warn(rtwdev, "Unknown wakeup reason %x\n", reason);
 }
 
+static void rtw_wow_pattern_write_cam(struct rtw_dev *rtwdev, u8 addr,
+				      u32 wdata)
+{
+	rtw_write32(rtwdev, REG_WKFMCAM_RWD, wdata);
+	rtw_write32(rtwdev, REG_WKFMCAM_CMD, BIT_WKFCAM_POLLING_V1 |
+		    BIT_WKFCAM_WE | BIT_WKFCAM_ADDR_V2(addr));
+
+	if (!check_hw_ready(rtwdev, REG_WKFMCAM_CMD, BIT_WKFCAM_POLLING_V1, 0))
+		rtw_err(rtwdev, "failed to write pattern cam\n");
+}
+
+static void rtw_wow_pattern_write_cam_ent(struct rtw_dev *rtwdev, u8 id,
+					  struct rtw_wow_pattern *rtw_pattern)
+{
+	int i;
+	u8 addr;
+	u32 wdata;
+
+	for (i = 0; i < RTW_MAX_PATTERN_MASK_SIZE / 4; i++) {
+		addr = (id << 3) + i;
+		wdata = rtw_pattern->mask[i * 4];
+		wdata |= rtw_pattern->mask[i * 4 + 1] << 8;
+		wdata |= rtw_pattern->mask[i * 4 + 2] << 16;
+		wdata |= rtw_pattern->mask[i * 4 + 3] << 24;
+		rtw_wow_pattern_write_cam(rtwdev, addr, wdata);
+	}
+
+	wdata = rtw_pattern->crc;
+	addr = (id << 3) + RTW_MAX_PATTERN_MASK_SIZE / 4;
+
+	switch (rtw_pattern->type) {
+	case RTW_PATTERN_BROADCAST:
+		wdata |= BIT_WKFMCAM_BC | BIT_WKFMCAM_VALID;
+		break;
+	case RTW_PATTERN_MULTICAST:
+		wdata |= BIT_WKFMCAM_MC | BIT_WKFMCAM_VALID;
+		break;
+	case RTW_PATTERN_UNICAST:
+		wdata |= BIT_WKFMCAM_UC | BIT_WKFMCAM_VALID;
+		break;
+	default:
+		break;
+	}
+	rtw_wow_pattern_write_cam(rtwdev, addr, wdata);
+}
+
+/* RTK internal CRC16 for Pattern Cam */
+static u16 __rtw_cal_crc16(u8 data, u16 crc)
+{
+	u8 shift_in, data_bit;
+	u8 crc_bit4, crc_bit11, crc_bit15;
+	u16 crc_result;
+	int index;
+
+	for (index = 0; index < 8; index++) {
+		crc_bit15 = ((crc & BIT(15)) ? 1 : 0);
+		data_bit = (data & (BIT(0) << index) ? 1 : 0);
+		shift_in = crc_bit15 ^ data_bit;
+
+		crc_result = crc << 1;
+
+		if (shift_in == 0)
+			crc_result &= (~BIT(0));
+		else
+			crc_result |= BIT(0);
+
+		crc_bit11 = ((crc & BIT(11)) ? 1 : 0) ^ shift_in;
+
+		if (crc_bit11 == 0)
+			crc_result &= (~BIT(12));
+		else
+			crc_result |= BIT(12);
+
+		crc_bit4 = ((crc & BIT(4)) ? 1 : 0) ^ shift_in;
+
+		if (crc_bit4 == 0)
+			crc_result &= (~BIT(5));
+		else
+			crc_result |= BIT(5);
+
+		crc = crc_result;
+	}
+	return crc;
+}
+
+static u16 rtw_calc_crc(u8 *pdata, int length)
+{
+	u16 crc = 0xffff;
+	int i;
+
+	for (i = 0; i < length; i++)
+		crc = __rtw_cal_crc16(pdata[i], crc);
+
+	/* get 1' complement */
+	return ~crc;
+}
+
+static void rtw_wow_pattern_generate(struct rtw_dev *rtwdev,
+				     struct rtw_vif *rtwvif,
+				     const struct cfg80211_pkt_pattern *pkt_pattern,
+				     struct rtw_wow_pattern *rtw_pattern)
+{
+	const u8 *mask;
+	const u8 *pattern;
+	u8 mask_hw[RTW_MAX_PATTERN_MASK_SIZE] = {0};
+	u8 content[RTW_MAX_PATTERN_SIZE] = {0};
+	u8 mac_addr[ETH_ALEN] = {0};
+	u8 mask_len;
+	u16 count;
+	int len;
+	int i;
+
+	pattern = pkt_pattern->pattern;
+	len = pkt_pattern->pattern_len;
+	mask = pkt_pattern->mask;
+
+	ether_addr_copy(mac_addr, rtwvif->mac_addr);
+	memset(rtw_pattern, 0, sizeof(*rtw_pattern));
+
+	mask_len = DIV_ROUND_UP(len, 8);
+
+	if (is_broadcast_ether_addr(pattern))
+		rtw_pattern->type = RTW_PATTERN_BROADCAST;
+	else if (is_multicast_ether_addr(pattern))
+		rtw_pattern->type = RTW_PATTERN_MULTICAST;
+	else if (ether_addr_equal(pattern, mac_addr))
+		rtw_pattern->type = RTW_PATTERN_UNICAST;
+	else
+		rtw_pattern->type = RTW_PATTERN_INVALID;
+
+	/* translate mask from os to mask for hw
+	 * pattern from OS uses 'ethenet frame', like this:
+	 * |    6   |    6   |   2  |     20    |  Variable  |  4  |
+	 * |--------+--------+------+-----------+------------+-----|
+	 * |    802.3 Mac Header    | IP Header | TCP Packet | FCS |
+	 * |   DA   |   SA   | Type |
+	 *
+	 * BUT, packet catched by our HW is in '802.11 frame', begin from LLC
+	 * |     24 or 30      |    6   |   2  |     20    |  Variable  |  4  |
+	 * |-------------------+--------+------+-----------+------------+-----|
+	 * | 802.11 MAC Header |       LLC     | IP Header | TCP Packet | FCS |
+	 *		       | Others | Tpye |
+	 *
+	 * Therefore, we need translate mask_from_OS to mask_to_hw.
+	 * We should left-shift mask by 6 bits, then set the new bit[0~5] = 0,
+	 * because new mask[0~5] means 'SA', but our HW packet begins from LLC,
+	 * bit[0~5] corresponds to first 6 Bytes in LLC, they just don't match.
+	 */
+
+	/* Shift 6 bits */
+	for (i = 0; i < mask_len - 1; i++) {
+		mask_hw[i] = u8_get_bits(mask[i], GENMASK(7, 6));
+		mask_hw[i] |= u8_get_bits(mask[i + 1], GENMASK(5, 0)) << 2;
+	}
+	mask_hw[i] = u8_get_bits(mask[i], GENMASK(7, 6));
+
+	/* Set bit 0-5 to zero */
+	mask_hw[0] &= (~GENMASK(5, 0));
+
+	memcpy(rtw_pattern->mask, mask_hw, RTW_MAX_PATTERN_MASK_SIZE);
+
+	/* To get the wake up pattern from the mask.
+	 * We do not count first 12 bits which means
+	 * DA[6] and SA[6] in the pattern to match HW design.
+	 */
+	count = 0;
+	for (i = 12; i < len; i++) {
+		if ((mask[i / 8] >> (i % 8)) & 0x01) {
+			content[count] = pattern[i];
+			count++;
+		}
+	}
+
+	rtw_pattern->crc = rtw_calc_crc(content, count);
+}
+
+static void rtw_wow_pattern_clear_cam(struct rtw_dev *rtwdev)
+{
+	bool ret;
+
+	rtw_write32(rtwdev, REG_WKFMCAM_CMD, BIT_WKFCAM_POLLING_V1 |
+		    BIT_WKFCAM_CLR_V1);
+
+	ret = check_hw_ready(rtwdev, REG_WKFMCAM_CMD, BIT_WKFCAM_POLLING_V1, 0);
+	if (!ret)
+		rtw_err(rtwdev, "failed to clean pattern cam\n");
+}
+
+static void rtw_wow_pattern_write(struct rtw_dev *rtwdev)
+{
+	struct rtw_wow_param *rtw_wow = &rtwdev->wow;
+	struct rtw_wow_pattern *rtw_pattern = rtw_wow->patterns;
+	int i = 0;
+
+	for (i = 0; i < rtw_wow->pattern_cnt; i++)
+		rtw_wow_pattern_write_cam_ent(rtwdev, i, rtw_pattern + i);
+}
+
+static void rtw_wow_pattern_clear(struct rtw_dev *rtwdev)
+{
+	struct rtw_wow_param *rtw_wow = &rtwdev->wow;
+
+	rtw_wow_pattern_clear_cam(rtwdev);
+
+	rtw_wow->pattern_cnt = 0;
+	memset(rtw_wow->patterns, 0, sizeof(rtw_wow->patterns));
+}
+
 static void rtw_wow_bb_stop(struct rtw_dev *rtwdev)
 {
 	struct rtw_wow_param *rtw_wow = &rtwdev->wow;
@@ -148,6 +358,7 @@ static int rtw_wow_fw_start(struct rtw_dev *rtwdev)
 {
 	if (rtw_wow_mgd_linked(rtwdev)) {
 		rtw_send_rsvd_page_h2c(rtwdev);
+		rtw_wow_pattern_write(rtwdev);
 		rtw_wow_fw_security_type(rtwdev);
 		rtw_fw_set_disconnect_decision_cmd(rtwdev, true);
 		rtw_fw_set_keep_alive_cmd(rtwdev, true);
@@ -164,6 +375,7 @@ static int rtw_wow_fw_stop(struct rtw_dev *rtwdev)
 	if (rtw_wow_mgd_linked(rtwdev)) {
 		rtw_fw_set_disconnect_decision_cmd(rtwdev, false);
 		rtw_fw_set_keep_alive_cmd(rtwdev, false);
+		rtw_wow_pattern_clear(rtwdev);
 	}
 
 	rtw_fw_set_wowlan_ctrl_cmd(rtwdev, false);
@@ -450,6 +662,9 @@ static int rtw_wow_set_wakeups(struct rtw_dev *rtwdev,
 			       struct cfg80211_wowlan *wowlan)
 {
 	struct rtw_wow_param *rtw_wow = &rtwdev->wow;
+	struct rtw_wow_pattern *rtw_patterns = rtw_wow->patterns;
+	struct rtw_vif *rtwvif;
+	int i;
 
 	if (wowlan->disconnect)
 		set_bit(RTW_WOW_FLAG_EN_DISCONNECT, rtw_wow->flags);
@@ -462,6 +677,15 @@ static int rtw_wow_set_wakeups(struct rtw_dev *rtwdev,
 	if (!rtw_wow->wow_vif)
 		return -EPERM;
 
+	rtwvif = (struct rtw_vif *)rtw_wow->wow_vif->drv_priv;
+	if (wowlan->n_patterns && wowlan->patterns) {
+		rtw_wow->pattern_cnt = wowlan->n_patterns;
+		for (i = 0; i < wowlan->n_patterns; i++)
+			rtw_wow_pattern_generate(rtwdev, rtwvif,
+						 wowlan->patterns + i,
+						 rtw_patterns + i);
+	}
+
 	return 0;
 }
 
diff --git a/drivers/net/wireless/realtek/rtw88/wow.h b/drivers/net/wireless/realtek/rtw88/wow.h
index 617806a2f28d..54f742f4c955 100644
--- a/drivers/net/wireless/realtek/rtw88/wow.h
+++ b/drivers/net/wireless/realtek/rtw88/wow.h
@@ -5,12 +5,21 @@
 #ifndef __RTW_WOW_H__
 #define __RTW_WOW_H__
 
+enum rtw_wow_pattern_type {
+	RTW_PATTERN_BROADCAST = 0,
+	RTW_PATTERN_MULTICAST,
+	RTW_PATTERN_UNICAST,
+	RTW_PATTERN_VALID,
+	RTW_PATTERN_INVALID,
+};
+
 enum rtw_wake_reason {
 	RTW_WOW_RSN_RX_PTK_REKEY = 0x1,
 	RTW_WOW_RSN_RX_GTK_REKEY = 0x2,
 	RTW_WOW_RSN_RX_DEAUTH = 0x8,
 	RTW_WOW_RSN_DISCONNECT = 0x10,
 	RTW_WOW_RSN_RX_MAGIC_PKT = 0x21,
+	RTW_WOW_RSN_RX_PATTERN_MATCH = 0x23,
 };
 
 struct rtw_fw_media_status_iter_data {
-- 
cgit v1.2.3-59-g8ed1b


From b6c12908a33e4e413523e73464cd5957690f1a76 Mon Sep 17 00:00:00 2001
From: Chin-Yen Lee <timlee@realtek.com>
Date: Thu, 19 Dec 2019 16:58:16 +0800
Subject: rtw88: Add wowlan net-detect support

Net-detect is an option of wowlan to allow the device to
be woken up from suspend mode when configured network is detected.

When user enables net-detect and lets the device enter suspend
state, wowlan firmware will periodically scan until beacon or
probe response of configured networks are received.
Between two scans, wowlan firmware keeps wifi chip in idle mode
to reduce power consumption. If configured networks are detected,
wowlan firmware will trigger resume process.

Signed-off-by: Chin-Yen Lee <timlee@realtek.com>
Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtw88/fw.c       | 292 ++++++++++++++++++++++++--
 drivers/net/wireless/realtek/rtw88/fw.h       | 113 ++++++++++
 drivers/net/wireless/realtek/rtw88/main.c     |   1 +
 drivers/net/wireless/realtek/rtw88/main.h     |  25 +++
 drivers/net/wireless/realtek/rtw88/rtw8822c.c |   5 +-
 drivers/net/wireless/realtek/rtw88/wow.c      | 139 +++++++++++-
 drivers/net/wireless/realtek/rtw88/wow.h      |  11 +
 7 files changed, 571 insertions(+), 15 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw88/fw.c b/drivers/net/wireless/realtek/rtw88/fw.c
index 0523851cf98d..e3e91ae0d3d2 100644
--- a/drivers/net/wireless/realtek/rtw88/fw.c
+++ b/drivers/net/wireless/realtek/rtw88/fw.c
@@ -567,6 +567,9 @@ void rtw_fw_set_remote_wake_ctrl_cmd(struct rtw_dev *rtwdev, bool enable)
 
 	SET_REMOTE_WAKECTRL_ENABLE(h2c_pkt, enable);
 
+	if (rtw_wow_no_link(rtwdev))
+		SET_REMOTE_WAKE_CTRL_NLO_OFFLOAD_EN(h2c_pkt, enable);
+
 	rtw_fw_send_h2c_command(rtwdev, h2c_pkt);
 }
 
@@ -584,6 +587,26 @@ static u8 rtw_get_rsvd_page_location(struct rtw_dev *rtwdev,
 	return location;
 }
 
+void rtw_fw_set_nlo_info(struct rtw_dev *rtwdev, bool enable)
+{
+	u8 h2c_pkt[H2C_PKT_SIZE] = {0};
+	u8 loc_nlo;
+
+	loc_nlo = rtw_get_rsvd_page_location(rtwdev, RSVD_NLO_INFO);
+
+	SET_H2C_CMD_ID_CLASS(h2c_pkt, H2C_CMD_NLO_INFO);
+
+	SET_NLO_FUN_EN(h2c_pkt, enable);
+	if (enable) {
+		if (rtw_fw_lps_deep_mode)
+			SET_NLO_PS_32K(h2c_pkt, enable);
+		SET_NLO_IGNORE_SECURITY(h2c_pkt, enable);
+		SET_NLO_LOC_NLO_INFO(h2c_pkt, loc_nlo);
+	}
+
+	rtw_fw_send_h2c_command(rtwdev, h2c_pkt);
+}
+
 void rtw_fw_set_pg_info(struct rtw_dev *rtwdev)
 {
 	struct rtw_lps_conf *conf = &rtwdev->lps_conf;
@@ -603,6 +626,40 @@ void rtw_fw_set_pg_info(struct rtw_dev *rtwdev)
 	rtw_fw_send_h2c_command(rtwdev, h2c_pkt);
 }
 
+u8 rtw_get_rsvd_page_probe_req_location(struct rtw_dev *rtwdev,
+					struct cfg80211_ssid *ssid)
+{
+	struct rtw_rsvd_page *rsvd_pkt;
+	u8 location = 0;
+
+	list_for_each_entry(rsvd_pkt, &rtwdev->rsvd_page_list, list) {
+		if (rsvd_pkt->type != RSVD_PROBE_REQ)
+			continue;
+		if ((!ssid && !rsvd_pkt->ssid) ||
+		    rtw_ssid_equal(rsvd_pkt->ssid, ssid))
+			location = rsvd_pkt->page;
+	}
+
+	return location;
+}
+
+u16 rtw_get_rsvd_page_probe_req_size(struct rtw_dev *rtwdev,
+				     struct cfg80211_ssid *ssid)
+{
+	struct rtw_rsvd_page *rsvd_pkt;
+	u16 size = 0;
+
+	list_for_each_entry(rsvd_pkt, &rtwdev->rsvd_page_list, list) {
+		if (rsvd_pkt->type != RSVD_PROBE_REQ)
+			continue;
+		if ((!ssid && !rsvd_pkt->ssid) ||
+		    rtw_ssid_equal(rsvd_pkt->ssid, ssid))
+			size = rsvd_pkt->skb->len;
+	}
+
+	return size;
+}
+
 void rtw_send_rsvd_page_h2c(struct rtw_dev *rtwdev)
 {
 	u8 h2c_pkt[H2C_PKT_SIZE] = {0};
@@ -648,6 +705,95 @@ rtw_beacon_get(struct ieee80211_hw *hw, struct ieee80211_vif *vif)
 	return skb_new;
 }
 
+static struct sk_buff *rtw_nlo_info_get(struct ieee80211_hw *hw)
+{
+	struct rtw_dev *rtwdev = hw->priv;
+	struct rtw_chip_info *chip = rtwdev->chip;
+	struct rtw_pno_request *pno_req = &rtwdev->wow.pno_req;
+	struct rtw_nlo_info_hdr *nlo_hdr;
+	struct cfg80211_ssid *ssid;
+	struct sk_buff *skb;
+	u8 *pos, loc;
+	u32 size;
+	int i;
+
+	if (!pno_req->inited || !pno_req->match_set_cnt)
+		return NULL;
+
+	size = sizeof(struct rtw_nlo_info_hdr) + pno_req->match_set_cnt *
+		      IEEE80211_MAX_SSID_LEN + chip->tx_pkt_desc_sz;
+
+	skb = alloc_skb(size, GFP_KERNEL);
+	if (!skb)
+		return NULL;
+
+	skb_reserve(skb, chip->tx_pkt_desc_sz);
+
+	nlo_hdr = skb_put_zero(skb, sizeof(struct rtw_nlo_info_hdr));
+
+	nlo_hdr->nlo_count = pno_req->match_set_cnt;
+	nlo_hdr->hidden_ap_count = pno_req->match_set_cnt;
+
+	/* pattern check for firmware */
+	memset(nlo_hdr->pattern_check, 0xA5, FW_NLO_INFO_CHECK_SIZE);
+
+	for (i = 0; i < pno_req->match_set_cnt; i++)
+		nlo_hdr->ssid_len[i] = pno_req->match_sets[i].ssid.ssid_len;
+
+	for (i = 0; i < pno_req->match_set_cnt; i++) {
+		ssid = &pno_req->match_sets[i].ssid;
+		loc  = rtw_get_rsvd_page_probe_req_location(rtwdev, ssid);
+		if (!loc) {
+			rtw_err(rtwdev, "failed to get probe req rsvd loc\n");
+			kfree(skb);
+			return NULL;
+		}
+		nlo_hdr->location[i] = loc;
+	}
+
+	for (i = 0; i < pno_req->match_set_cnt; i++) {
+		pos = skb_put_zero(skb, IEEE80211_MAX_SSID_LEN);
+		memcpy(pos, pno_req->match_sets[i].ssid.ssid,
+		       pno_req->match_sets[i].ssid.ssid_len);
+	}
+
+	return skb;
+}
+
+static struct sk_buff *rtw_cs_channel_info_get(struct ieee80211_hw *hw)
+{
+	struct rtw_dev *rtwdev = hw->priv;
+	struct rtw_chip_info *chip = rtwdev->chip;
+	struct rtw_pno_request *pno_req = &rtwdev->wow.pno_req;
+	struct ieee80211_channel *channels = pno_req->channels;
+	struct sk_buff *skb;
+	int count =  pno_req->channel_cnt;
+	u8 *pos;
+	int i = 0;
+
+	skb = alloc_skb(4 * count + chip->tx_pkt_desc_sz, GFP_KERNEL);
+	if (!skb)
+		return NULL;
+
+	skb_reserve(skb, chip->tx_pkt_desc_sz);
+
+	for (i = 0; i < count; i++) {
+		pos = skb_put_zero(skb, 4);
+
+		CHSW_INFO_SET_CH(pos, channels[i].hw_value);
+
+		if (channels[i].flags & IEEE80211_CHAN_RADAR)
+			CHSW_INFO_SET_ACTION_ID(pos, 0);
+		else
+			CHSW_INFO_SET_ACTION_ID(pos, 1);
+		CHSW_INFO_SET_TIMEOUT(pos, 1);
+		CHSW_INFO_SET_PRI_CH_IDX(pos, 1);
+		CHSW_INFO_SET_BW(pos, 0);
+	}
+
+	return skb;
+}
+
 static struct sk_buff *rtw_lps_pg_dpk_get(struct ieee80211_hw *hw)
 {
 	struct rtw_dev *rtwdev = hw->priv;
@@ -705,11 +851,12 @@ static struct sk_buff *rtw_lps_pg_info_get(struct ieee80211_hw *hw,
 
 static struct sk_buff *rtw_get_rsvd_page_skb(struct ieee80211_hw *hw,
 					     struct ieee80211_vif *vif,
-					     enum rtw_rsvd_packet_type type)
+					     struct rtw_rsvd_page *rsvd_pkt)
 {
 	struct sk_buff *skb_new;
+	struct cfg80211_ssid *ssid;
 
-	switch (type) {
+	switch (rsvd_pkt->type) {
 	case RSVD_BEACON:
 		skb_new = rtw_beacon_get(hw, vif);
 		break;
@@ -731,6 +878,21 @@ static struct sk_buff *rtw_get_rsvd_page_skb(struct ieee80211_hw *hw,
 	case RSVD_LPS_PG_INFO:
 		skb_new = rtw_lps_pg_info_get(hw, vif);
 		break;
+	case RSVD_PROBE_REQ:
+		ssid = (struct cfg80211_ssid *)rsvd_pkt->ssid;
+		if (ssid)
+			skb_new = ieee80211_probereq_get(hw, vif->addr,
+							 ssid->ssid,
+							 ssid->ssid_len, 0);
+		else
+			skb_new = ieee80211_probereq_get(hw, vif->addr, NULL, 0, 0);
+		break;
+	case RSVD_NLO_INFO:
+		skb_new = rtw_nlo_info_get(hw);
+		break;
+	case RSVD_CH_INFO:
+		skb_new = rtw_cs_channel_info_get(hw);
+		break;
 	default:
 		return NULL;
 	}
@@ -772,25 +934,53 @@ static void rtw_rsvd_page_list_to_buf(struct rtw_dev *rtwdev, u8 page_size,
 		memcpy(buf, skb->data, skb->len);
 }
 
+static struct rtw_rsvd_page *rtw_alloc_rsvd_page(struct rtw_dev *rtwdev,
+						 enum rtw_rsvd_packet_type type,
+						 bool txdesc)
+{
+	struct rtw_rsvd_page *rsvd_pkt = NULL;
+
+	rsvd_pkt = kzalloc(sizeof(*rsvd_pkt), GFP_KERNEL);
+
+	if (!rsvd_pkt)
+		return NULL;
+
+	rsvd_pkt->type = type;
+	rsvd_pkt->add_txdesc = txdesc;
+
+	return rsvd_pkt;
+}
+
+static void rtw_insert_rsvd_page(struct rtw_dev *rtwdev,
+				 struct rtw_rsvd_page *rsvd_pkt)
+{
+	lockdep_assert_held(&rtwdev->mutex);
+	list_add_tail(&rsvd_pkt->list, &rtwdev->rsvd_page_list);
+}
+
 void rtw_add_rsvd_page(struct rtw_dev *rtwdev, enum rtw_rsvd_packet_type type,
 		       bool txdesc)
 {
 	struct rtw_rsvd_page *rsvd_pkt;
 
-	lockdep_assert_held(&rtwdev->mutex);
+	rsvd_pkt = rtw_alloc_rsvd_page(rtwdev, type, txdesc);
+	if (!rsvd_pkt)
+		return;
 
-	list_for_each_entry(rsvd_pkt, &rtwdev->rsvd_page_list, list) {
-		if (rsvd_pkt->type == type)
-			return;
-	}
+	rtw_insert_rsvd_page(rtwdev, rsvd_pkt);
+}
 
-	rsvd_pkt = kmalloc(sizeof(*rsvd_pkt), GFP_KERNEL);
+void rtw_add_rsvd_page_probe_req(struct rtw_dev *rtwdev,
+				 struct cfg80211_ssid *ssid)
+{
+	struct rtw_rsvd_page *rsvd_pkt;
+
+	rsvd_pkt = rtw_alloc_rsvd_page(rtwdev, RSVD_PROBE_REQ, true);
 	if (!rsvd_pkt)
 		return;
 
-	rsvd_pkt->type = type;
-	rsvd_pkt->add_txdesc = txdesc;
-	list_add_tail(&rsvd_pkt->list, &rtwdev->rsvd_page_list);
+	rsvd_pkt->ssid = ssid;
+	rtw_insert_rsvd_page(rtwdev, rsvd_pkt);
 }
 
 void rtw_reset_rsvd_page(struct rtw_dev *rtwdev)
@@ -887,7 +1077,7 @@ static u8 *rtw_build_rsvd_page(struct rtw_dev *rtwdev,
 	page_margin = page_size - tx_desc_sz;
 
 	list_for_each_entry(rsvd_pkt, &rtwdev->rsvd_page_list, list) {
-		iter = rtw_get_rsvd_page_skb(hw, vif, rsvd_pkt->type);
+		iter = rtw_get_rsvd_page_skb(hw, vif, rsvd_pkt);
 		if (!iter) {
 			rtw_err(rtwdev, "failed to build rsvd packet\n");
 			goto release_skb;
@@ -1065,3 +1255,81 @@ out:
 	rtw_write8(rtwdev, REG_RCR + 2, rcr);
 	return 0;
 }
+
+static void __rtw_fw_update_pkt(struct rtw_dev *rtwdev, u8 pkt_id, u16 size,
+				u8 location)
+{
+	u8 h2c_pkt[H2C_PKT_SIZE] = {0};
+	u16 total_size = H2C_PKT_HDR_SIZE + H2C_PKT_UPDATE_PKT_LEN;
+
+	rtw_h2c_pkt_set_header(h2c_pkt, H2C_PKT_UPDATE_PKT);
+
+	SET_PKT_H2C_TOTAL_LEN(h2c_pkt, total_size);
+	UPDATE_PKT_SET_PKT_ID(h2c_pkt, pkt_id);
+	UPDATE_PKT_SET_LOCATION(h2c_pkt, location);
+
+	/* include txdesc size */
+	UPDATE_PKT_SET_SIZE(h2c_pkt, size);
+
+	rtw_fw_send_h2c_packet(rtwdev, h2c_pkt);
+}
+
+void rtw_fw_update_pkt_probe_req(struct rtw_dev *rtwdev,
+				 struct cfg80211_ssid *ssid)
+{
+	u8 loc;
+	u32 size;
+
+	loc = rtw_get_rsvd_page_probe_req_location(rtwdev, ssid);
+	if (!loc) {
+		rtw_err(rtwdev, "failed to get probe_req rsvd loc\n");
+		return;
+	}
+
+	size = rtw_get_rsvd_page_probe_req_size(rtwdev, ssid);
+	if (!size) {
+		rtw_err(rtwdev, "failed to get probe_req rsvd size\n");
+		return;
+	}
+
+	__rtw_fw_update_pkt(rtwdev, RTW_PACKET_PROBE_REQ, size, loc);
+}
+
+void rtw_fw_channel_switch(struct rtw_dev *rtwdev, bool enable)
+{
+	struct rtw_pno_request *rtw_pno_req = &rtwdev->wow.pno_req;
+	u8 h2c_pkt[H2C_PKT_SIZE] = {0};
+	u16 total_size = H2C_PKT_HDR_SIZE + H2C_PKT_CH_SWITCH_LEN;
+	u8 loc_ch_info;
+	const struct rtw_ch_switch_option cs_option = {
+		.dest_ch_en = 1,
+		.dest_ch = 1,
+		.periodic_option = 2,
+		.normal_period = 5,
+		.normal_period_sel = 0,
+		.normal_cycle = 10,
+		.slow_period = 1,
+		.slow_period_sel = 1,
+	};
+
+	rtw_h2c_pkt_set_header(h2c_pkt, H2C_PKT_CH_SWITCH);
+	SET_PKT_H2C_TOTAL_LEN(h2c_pkt, total_size);
+
+	CH_SWITCH_SET_START(h2c_pkt, enable);
+	CH_SWITCH_SET_DEST_CH_EN(h2c_pkt, cs_option.dest_ch_en);
+	CH_SWITCH_SET_DEST_CH(h2c_pkt, cs_option.dest_ch);
+	CH_SWITCH_SET_NORMAL_PERIOD(h2c_pkt, cs_option.normal_period);
+	CH_SWITCH_SET_NORMAL_PERIOD_SEL(h2c_pkt, cs_option.normal_period_sel);
+	CH_SWITCH_SET_SLOW_PERIOD(h2c_pkt, cs_option.slow_period);
+	CH_SWITCH_SET_SLOW_PERIOD_SEL(h2c_pkt, cs_option.slow_period_sel);
+	CH_SWITCH_SET_NORMAL_CYCLE(h2c_pkt, cs_option.normal_cycle);
+	CH_SWITCH_SET_PERIODIC_OPT(h2c_pkt, cs_option.periodic_option);
+
+	CH_SWITCH_SET_CH_NUM(h2c_pkt, rtw_pno_req->channel_cnt);
+	CH_SWITCH_SET_INFO_SIZE(h2c_pkt, rtw_pno_req->channel_cnt * 4);
+
+	loc_ch_info = rtw_get_rsvd_page_location(rtwdev, RSVD_CH_INFO);
+	CH_SWITCH_SET_INFO_LOC(h2c_pkt, loc_ch_info);
+
+	rtw_fw_send_h2c_packet(rtwdev, h2c_pkt);
+}
diff --git a/drivers/net/wireless/realtek/rtw88/fw.h b/drivers/net/wireless/realtek/rtw88/fw.h
index 9ea27cca11ac..ccd27bd45775 100644
--- a/drivers/net/wireless/realtek/rtw88/fw.h
+++ b/drivers/net/wireless/realtek/rtw88/fw.h
@@ -12,6 +12,8 @@
 #define FW_HDR_SIZE			64
 #define FW_HDR_CHKSUM_SIZE		8
 
+#define FW_NLO_INFO_CHECK_SIZE		4
+
 #define FIFO_PAGE_SIZE_SHIFT		12
 #define FIFO_PAGE_SIZE			4096
 #define RSVD_PAGE_START_ADDR		0x780
@@ -45,6 +47,9 @@ enum rtw_rsvd_packet_type {
 	RSVD_QOS_NULL,
 	RSVD_LPS_PG_DPK,
 	RSVD_LPS_PG_INFO,
+	RSVD_PROBE_REQ,
+	RSVD_NLO_INFO,
+	RSVD_CH_INFO,
 };
 
 enum rtw_fw_rf_type {
@@ -98,6 +103,7 @@ struct rtw_rsvd_page {
 	enum rtw_rsvd_packet_type type;
 	u8 page;
 	bool add_txdesc;
+	struct cfg80211_ssid *ssid;
 };
 
 enum rtw_keep_alive_pkt_type {
@@ -105,6 +111,24 @@ enum rtw_keep_alive_pkt_type {
 	KEEP_ALIVE_ARP_RSP = 1,
 };
 
+struct rtw_nlo_info_hdr {
+	u8 nlo_count;
+	u8 hidden_ap_count;
+	u8 rsvd1[2];
+	u8 pattern_check[FW_NLO_INFO_CHECK_SIZE];
+	u8 rsvd2[8];
+	u8 ssid_len[16];
+	u8 chiper[16];
+	u8 rsvd3[16];
+	u8 location[8];
+} __packed;
+
+enum rtw_packet_type {
+	RTW_PACKET_PROBE_REQ = 0x00,
+
+	RTW_PACKET_UNDEFINE = 0x7FFFFFFF,
+};
+
 struct rtw_fw_wow_keep_alive_para {
 	bool adopt;
 	u8 pkt_type;
@@ -117,6 +141,21 @@ struct rtw_fw_wow_disconnect_para {
 	u8 retry_count;
 };
 
+struct rtw_ch_switch_option {
+	u8 periodic_option;
+	u32 tsf_high;
+	u32 tsf_low;
+	u8 dest_ch_en;
+	u8 absolute_time_en;
+	u8 dest_ch;
+	u8 normal_period;
+	u8 normal_period_sel;
+	u8 normal_cycle;
+	u8 slow_period;
+	u8 slow_period_sel;
+	u8 nlo_en;
+};
+
 struct rtw_fw_hdr {
 	__le16 signature;
 	u8 category;
@@ -163,6 +202,12 @@ struct rtw_fw_hdr {
 #define H2C_PKT_PHYDM_INFO 0x11
 #define H2C_PKT_IQK 0x0E
 
+#define H2C_PKT_CH_SWITCH 0x02
+#define H2C_PKT_UPDATE_PKT 0x0C
+
+#define H2C_PKT_CH_SWITCH_LEN 0x20
+#define H2C_PKT_UPDATE_PKT_LEN 0x4
+
 #define SET_PKT_H2C_CATEGORY(h2c_pkt, value)                                   \
 	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, GENMASK(6, 0))
 #define SET_PKT_H2C_CMD_ID(h2c_pkt, value)                                     \
@@ -199,6 +244,57 @@ static inline void rtw_h2c_pkt_set_header(u8 *h2c_pkt, u8 sub_id)
 #define IQK_SET_SEGMENT_IQK(h2c_pkt, value)                                    \
 	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x02, value, BIT(1))
 
+#define CHSW_INFO_SET_CH(pkt, value)					       \
+	le32p_replace_bits((__le32 *)(pkt) + 0x00, value, GENMASK(7, 0))
+#define CHSW_INFO_SET_PRI_CH_IDX(pkt, value)				       \
+	le32p_replace_bits((__le32 *)(pkt) + 0x00, value, GENMASK(11, 8))
+#define CHSW_INFO_SET_BW(pkt, value)					       \
+	le32p_replace_bits((__le32 *)(pkt) + 0x00, value, GENMASK(15, 12))
+#define CHSW_INFO_SET_TIMEOUT(pkt, value)				       \
+	le32p_replace_bits((__le32 *)(pkt) + 0x00, value, GENMASK(23, 16))
+#define CHSW_INFO_SET_ACTION_ID(pkt, value)				       \
+	le32p_replace_bits((__le32 *)(pkt) + 0x00, value, GENMASK(30, 24))
+
+#define UPDATE_PKT_SET_SIZE(h2c_pkt, value)				       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x02, value, GENMASK(15, 0))
+#define UPDATE_PKT_SET_PKT_ID(h2c_pkt, value)				       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x02, value, GENMASK(23, 16))
+#define UPDATE_PKT_SET_LOCATION(h2c_pkt, value)				       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x02, value, GENMASK(31, 24))
+
+#define CH_SWITCH_SET_START(h2c_pkt, value)				       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x02, value, BIT(0))
+#define CH_SWITCH_SET_DEST_CH_EN(h2c_pkt, value)			       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x02, value, BIT(1))
+#define CH_SWITCH_SET_ABSOLUTE_TIME(h2c_pkt, value)			       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x02, value, BIT(2))
+#define CH_SWITCH_SET_PERIODIC_OPT(h2c_pkt, value)			       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x02, value, GENMASK(4, 3))
+#define CH_SWITCH_SET_INFO_LOC(h2c_pkt, value)				       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x02, value, GENMASK(15, 8))
+#define CH_SWITCH_SET_CH_NUM(h2c_pkt, value)				       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x02, value, GENMASK(23, 16))
+#define CH_SWITCH_SET_PRI_CH_IDX(h2c_pkt, value)			       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x02, value, GENMASK(27, 24))
+#define CH_SWITCH_SET_DEST_CH(h2c_pkt, value)				       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x03, value, GENMASK(7, 0))
+#define CH_SWITCH_SET_NORMAL_PERIOD(h2c_pkt, value)			       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x03, value, GENMASK(13, 8))
+#define CH_SWITCH_SET_NORMAL_PERIOD_SEL(h2c_pkt, value)			       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x03, value, GENMASK(15, 14))
+#define CH_SWITCH_SET_SLOW_PERIOD(h2c_pkt, value)			       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x03, value, GENMASK(21, 16))
+#define CH_SWITCH_SET_SLOW_PERIOD_SEL(h2c_pkt, value)			       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x03, value, GENMASK(23, 22))
+#define CH_SWITCH_SET_NORMAL_CYCLE(h2c_pkt, value)			       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x03, value, GENMASK(31, 24))
+#define CH_SWITCH_SET_TSF_HIGH(h2c_pkt, value)				       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x04, value, GENMASK(31, 0))
+#define CH_SWITCH_SET_TSF_LOW(h2c_pkt, value)				       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x05, value, GENMASK(31, 0))
+#define CH_SWITCH_SET_INFO_SIZE(h2c_pkt, value)				       \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x06, value, GENMASK(15, 0))
+
 /* Command H2C */
 #define H2C_CMD_RSVD_PAGE		0x0
 #define H2C_CMD_MEDIA_STATUS_RPT	0x01
@@ -220,6 +316,8 @@ static inline void rtw_h2c_pkt_set_header(u8 *h2c_pkt, u8 sub_id)
 #define H2C_CMD_WOWLAN			0x80
 #define H2C_CMD_REMOTE_WAKE_CTRL	0x81
 #define H2C_CMD_AOAC_GLOBAL_INFO	0x82
+#define H2C_CMD_NLO_INFO		0x8C
+
 #define SET_H2C_CMD_ID_CLASS(h2c_pkt, value)				       \
 	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, GENMASK(7, 0))
 
@@ -366,6 +464,15 @@ static inline void rtw_h2c_pkt_set_header(u8 *h2c_pkt, u8 sub_id)
 #define SET_AOAC_GLOBAL_INFO_GROUP_ENC_ALG(h2c_pkt, value)		       \
 	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, GENMASK(23, 16))
 
+#define SET_NLO_FUN_EN(h2c_pkt, value)                                         \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, BIT(8))
+#define SET_NLO_PS_32K(h2c_pkt, value)                                         \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, BIT(9))
+#define SET_NLO_IGNORE_SECURITY(h2c_pkt, value)                                \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, BIT(10))
+#define SET_NLO_LOC_NLO_INFO(h2c_pkt, value)                                   \
+	le32p_replace_bits((__le32 *)(h2c_pkt) + 0x00, value, GENMASK(23, 16))
+
 static inline struct rtw_c2h_cmd *get_c2h_from_skb(struct sk_buff *skb)
 {
 	u32 pkt_offset;
@@ -397,6 +504,8 @@ void rtw_fw_send_ra_info(struct rtw_dev *rtwdev, struct rtw_sta_info *si);
 void rtw_fw_media_status_report(struct rtw_dev *rtwdev, u8 mac_id, bool conn);
 void rtw_add_rsvd_page(struct rtw_dev *rtwdev, enum rtw_rsvd_packet_type type,
 		       bool txdesc);
+void rtw_add_rsvd_page_probe_req(struct rtw_dev *rtwdev,
+				 struct cfg80211_ssid *ssid);
 int rtw_fw_write_data_rsvd_page(struct rtw_dev *rtwdev, u16 pg_addr,
 				u8 *buf, u32 size);
 void rtw_reset_rsvd_page(struct rtw_dev *rtwdev);
@@ -413,4 +522,8 @@ void rtw_fw_set_aoac_global_info_cmd(struct rtw_dev *rtwdev,
 				     u8 pairwise_key_enc,
 				     u8 group_key_enc);
 
+void rtw_fw_set_nlo_info(struct rtw_dev *rtwdev, bool enable);
+void rtw_fw_update_pkt_probe_req(struct rtw_dev *rtwdev,
+				 struct cfg80211_ssid *ssid);
+void rtw_fw_channel_switch(struct rtw_dev *rtwdev, bool enable);
 #endif
diff --git a/drivers/net/wireless/realtek/rtw88/main.c b/drivers/net/wireless/realtek/rtw88/main.c
index 1ac48ccdb3c8..3d26c4a6e8da 100644
--- a/drivers/net/wireless/realtek/rtw88/main.c
+++ b/drivers/net/wireless/realtek/rtw88/main.c
@@ -1495,6 +1495,7 @@ int rtw_register_hw(struct rtw_dev *rtwdev, struct ieee80211_hw *hw)
 
 #ifdef CONFIG_PM
 	hw->wiphy->wowlan = rtwdev->chip->wowlan_stub;
+	hw->wiphy->max_sched_scan_ssids = rtwdev->chip->max_sched_scan_ssids;
 #endif
 	rtw_set_supported_band(hw, rtwdev->chip);
 	SET_IEEE80211_PERM_ADDR(hw, rtwdev->efuse.addr);
diff --git a/drivers/net/wireless/realtek/rtw88/main.h b/drivers/net/wireless/realtek/rtw88/main.h
index 74368ef954fa..4fa17aab3838 100644
--- a/drivers/net/wireless/realtek/rtw88/main.h
+++ b/drivers/net/wireless/realtek/rtw88/main.h
@@ -929,12 +929,24 @@ struct rtw_wow_pattern {
 	u8 mask[RTW_MAX_PATTERN_MASK_SIZE];
 };
 
+struct rtw_pno_request {
+	bool inited;
+	u32 match_set_cnt;
+	struct cfg80211_match_set *match_sets;
+	u8 channel_cnt;
+	struct ieee80211_channel *channels;
+	struct cfg80211_sched_scan_plan scan_plan;
+};
+
 struct rtw_wow_param {
 	struct ieee80211_vif *wow_vif;
 	DECLARE_BITMAP(flags, RTW_WOW_FLAG_MAX);
 	u8 txpause;
 	u8 pattern_cnt;
 	struct rtw_wow_pattern patterns[RTW_MAX_PATTERN_NUM];
+
+	bool ips_enabled;
+	struct rtw_pno_request pno_req;
 };
 
 struct rtw_intf_phy_para_table {
@@ -1067,6 +1079,7 @@ struct rtw_chip_info {
 
 	const char *wow_fw_name;
 	const struct wiphy_wowlan_support *wowlan_stub;
+	const u8 max_sched_scan_ssids;
 
 	/* coex paras */
 	u32 coex_para_ver;
@@ -1647,6 +1660,18 @@ static inline struct ieee80211_vif *rtwvif_to_vif(struct rtw_vif *rtwvif)
 	return container_of(p, struct ieee80211_vif, drv_priv);
 }
 
+static inline bool rtw_ssid_equal(struct cfg80211_ssid *a,
+				  struct cfg80211_ssid *b)
+{
+	if (!a || !b || a->ssid_len != b->ssid_len)
+		return false;
+
+	if (memcmp(a->ssid, b->ssid, a->ssid_len))
+		return false;
+
+	return true;
+}
+
 void rtw_get_channel_params(struct cfg80211_chan_def *chandef,
 			    struct rtw_channel_params *ch_param);
 bool check_hw_ready(struct rtw_dev *rtwdev, u32 addr, u32 mask, u32 target);
diff --git a/drivers/net/wireless/realtek/rtw88/rtw8822c.c b/drivers/net/wireless/realtek/rtw88/rtw8822c.c
index 4836c0044b05..655ca9265f68 100644
--- a/drivers/net/wireless/realtek/rtw88/rtw8822c.c
+++ b/drivers/net/wireless/realtek/rtw88/rtw8822c.c
@@ -4063,10 +4063,12 @@ static const struct rtw_pwr_track_tbl rtw8822c_rtw_pwr_track_tbl = {
 #ifdef CONFIG_PM
 static const struct wiphy_wowlan_support rtw_wowlan_stub_8822c = {
 	.flags = WIPHY_WOWLAN_MAGIC_PKT | WIPHY_WOWLAN_GTK_REKEY_FAILURE |
-		 WIPHY_WOWLAN_DISCONNECT | WIPHY_WOWLAN_SUPPORTS_GTK_REKEY,
+		 WIPHY_WOWLAN_DISCONNECT | WIPHY_WOWLAN_SUPPORTS_GTK_REKEY |
+		 WIPHY_WOWLAN_NET_DETECT,
 	.n_patterns = RTW_MAX_PATTERN_NUM,
 	.pattern_max_len = RTW_MAX_PATTERN_SIZE,
 	.pattern_min_len = 1,
+	.max_nd_match_sets = 4,
 };
 #endif
 
@@ -4119,6 +4121,7 @@ struct rtw_chip_info rtw8822c_hw_spec = {
 #ifdef CONFIG_PM
 	.wow_fw_name = "rtw88/rtw8822c_wow_fw.bin",
 	.wowlan_stub = &rtw_wowlan_stub_8822c,
+	.max_sched_scan_ssids = 4,
 #endif
 	.coex_para_ver = 0x19062706,
 	.bt_desired_ver = 0x6,
diff --git a/drivers/net/wireless/realtek/rtw88/wow.c b/drivers/net/wireless/realtek/rtw88/wow.c
index f4645f3c98e0..af5c27e1bb07 100644
--- a/drivers/net/wireless/realtek/rtw88/wow.c
+++ b/drivers/net/wireless/realtek/rtw88/wow.c
@@ -28,6 +28,8 @@ static void rtw_wow_show_wakeup_reason(struct rtw_dev *rtwdev)
 		rtw_dbg(rtwdev, RTW_DBG_WOW, "WOW: Rx ptk rekey\n");
 	else if (reason == RTW_WOW_RSN_RX_PATTERN_MATCH)
 		rtw_dbg(rtwdev, RTW_DBG_WOW, "WOW: Rx pattern match packet\n");
+	else if (reason == RTW_WOW_RSN_RX_NLO)
+		rtw_dbg(rtwdev, RTW_DBG_WOW, "Rx NLO\n");
 	else
 		rtw_warn(rtwdev, "Unknown wakeup reason %x\n", reason);
 }
@@ -362,6 +364,10 @@ static int rtw_wow_fw_start(struct rtw_dev *rtwdev)
 		rtw_wow_fw_security_type(rtwdev);
 		rtw_fw_set_disconnect_decision_cmd(rtwdev, true);
 		rtw_fw_set_keep_alive_cmd(rtwdev, true);
+	} else if (rtw_wow_no_link(rtwdev)) {
+		rtw_fw_set_nlo_info(rtwdev, true);
+		rtw_fw_update_pkt_probe_req(rtwdev, NULL);
+		rtw_fw_channel_switch(rtwdev, true);
 	}
 
 	rtw_fw_set_wowlan_ctrl_cmd(rtwdev, true);
@@ -376,6 +382,9 @@ static int rtw_wow_fw_stop(struct rtw_dev *rtwdev)
 		rtw_fw_set_disconnect_decision_cmd(rtwdev, false);
 		rtw_fw_set_keep_alive_cmd(rtwdev, false);
 		rtw_wow_pattern_clear(rtwdev);
+	} else if (rtw_wow_no_link(rtwdev)) {
+		rtw_fw_channel_switch(rtwdev, false);
+		rtw_fw_set_nlo_info(rtwdev, false);
 	}
 
 	rtw_fw_set_wowlan_ctrl_cmd(rtwdev, false);
@@ -423,6 +432,22 @@ static void rtw_wow_fw_media_status(struct rtw_dev *rtwdev, bool connect)
 	rtw_iterate_stas_atomic(rtwdev, rtw_wow_fw_media_status_iter, &data);
 }
 
+static void rtw_wow_config_pno_rsvd_page(struct rtw_dev *rtwdev)
+{
+	struct rtw_wow_param *rtw_wow = &rtwdev->wow;
+	struct rtw_pno_request *rtw_pno_req = &rtw_wow->pno_req;
+	struct cfg80211_ssid *ssid;
+	int i;
+
+	for (i = 0 ; i < rtw_pno_req->match_set_cnt; i++) {
+		ssid = &rtw_pno_req->match_sets[i].ssid;
+		rtw_add_rsvd_page_probe_req(rtwdev, ssid);
+	}
+	rtw_add_rsvd_page_probe_req(rtwdev, NULL);
+	rtw_add_rsvd_page(rtwdev, RSVD_NLO_INFO, false);
+	rtw_add_rsvd_page(rtwdev, RSVD_CH_INFO, true);
+}
+
 static void rtw_wow_config_linked_rsvd_page(struct rtw_dev *rtwdev)
 {
 	rtw_add_rsvd_page(rtwdev, RSVD_PS_POLL, true);
@@ -436,8 +461,12 @@ static void rtw_wow_config_rsvd_page(struct rtw_dev *rtwdev)
 {
 	rtw_reset_rsvd_page(rtwdev);
 
-	if (rtw_wow_mgd_linked(rtwdev))
+	if (rtw_wow_mgd_linked(rtwdev)) {
 		rtw_wow_config_linked_rsvd_page(rtwdev);
+	} else if (test_bit(RTW_FLAG_WOWLAN, rtwdev->flags) &&
+		   rtw_wow_no_link(rtwdev)) {
+		rtw_wow_config_pno_rsvd_page(rtwdev);
+	}
 }
 
 static int rtw_wow_dl_fw_rsvd_page(struct rtw_dev *rtwdev)
@@ -480,11 +509,75 @@ out:
 	return ret;
 }
 
+static void rtw_wow_check_pno(struct rtw_dev *rtwdev,
+			      struct cfg80211_sched_scan_request *nd_config)
+{
+	struct rtw_wow_param *rtw_wow = &rtwdev->wow;
+	struct rtw_pno_request *pno_req = &rtw_wow->pno_req;
+	struct ieee80211_channel *channel;
+	int i, size;
+
+	if (!nd_config->n_match_sets || !nd_config->n_channels)
+		goto err;
+
+	pno_req->match_set_cnt = nd_config->n_match_sets;
+	size = sizeof(*pno_req->match_sets) * pno_req->match_set_cnt;
+	pno_req->match_sets = kmemdup(nd_config->match_sets, size, GFP_KERNEL);
+	if (!pno_req->match_sets)
+		goto err;
+
+	pno_req->channel_cnt = nd_config->n_channels;
+	size = sizeof(*nd_config->channels[0]) * nd_config->n_channels;
+	pno_req->channels = kmalloc(size, GFP_KERNEL);
+	if (!pno_req->channels)
+		goto channel_err;
+
+	for (i = 0 ; i < pno_req->channel_cnt; i++) {
+		channel = pno_req->channels + i;
+		memcpy(channel, nd_config->channels[i], sizeof(*channel));
+	}
+
+	pno_req->scan_plan = *nd_config->scan_plans;
+	pno_req->inited = true;
+
+	rtw_dbg(rtwdev, RTW_DBG_WOW, "WOW: net-detect is enabled\n");
+
+	return;
+
+channel_err:
+	kfree(pno_req->match_sets);
+
+err:
+	rtw_dbg(rtwdev, RTW_DBG_WOW, "WOW: net-detect is disabled\n");
+}
+
 static int rtw_wow_leave_linked_ps(struct rtw_dev *rtwdev)
 {
 	if (!test_bit(RTW_FLAG_WOWLAN, rtwdev->flags))
 		cancel_delayed_work_sync(&rtwdev->watch_dog_work);
 
+	rtw_leave_lps(rtwdev);
+
+	return 0;
+}
+
+static int rtw_wow_leave_no_link_ps(struct rtw_dev *rtwdev)
+{
+	struct rtw_wow_param *rtw_wow = &rtwdev->wow;
+	int ret = 0;
+
+	if (test_bit(RTW_FLAG_WOWLAN, rtwdev->flags)) {
+		if (rtw_fw_lps_deep_mode)
+			rtw_leave_lps_deep(rtwdev);
+	} else {
+		if (test_bit(RTW_FLAG_INACTIVE_PS, rtwdev->flags)) {
+			rtw_wow->ips_enabled = true;
+			ret = rtw_leave_ips(rtwdev);
+			if (ret)
+				return ret;
+		}
+	}
+
 	return 0;
 }
 
@@ -494,6 +587,18 @@ static int rtw_wow_leave_ps(struct rtw_dev *rtwdev)
 
 	if (rtw_wow_mgd_linked(rtwdev))
 		ret = rtw_wow_leave_linked_ps(rtwdev);
+	else if (rtw_wow_no_link(rtwdev))
+		ret = rtw_wow_leave_no_link_ps(rtwdev);
+
+	return ret;
+}
+
+static int rtw_wow_restore_ps(struct rtw_dev *rtwdev)
+{
+	int ret = 0;
+
+	if (rtw_wow_no_link(rtwdev) && rtwdev->wow.ips_enabled)
+		ret = rtw_enter_ips(rtwdev);
 
 	return ret;
 }
@@ -509,12 +614,22 @@ static int rtw_wow_enter_linked_ps(struct rtw_dev *rtwdev)
 	return 0;
 }
 
+static int rtw_wow_enter_no_link_ps(struct rtw_dev *rtwdev)
+{
+	/* firmware enters deep ps by itself if supported */
+	set_bit(RTW_FLAG_LEISURE_PS_DEEP, rtwdev->flags);
+
+	return 0;
+}
+
 static int rtw_wow_enter_ps(struct rtw_dev *rtwdev)
 {
 	int ret = 0;
 
 	if (rtw_wow_mgd_linked(rtwdev))
 		ret = rtw_wow_enter_linked_ps(rtwdev);
+	else if (rtw_wow_no_link(rtwdev) && rtw_fw_lps_deep_mode)
+		ret = rtw_wow_enter_no_link_ps(rtwdev);
 
 	return ret;
 }
@@ -653,6 +768,10 @@ static void rtw_wow_vif_iter(void *data, u8 *mac, struct ieee80211_vif *vif)
 	case RTW_NET_MGD_LINKED:
 		rtw_wow->wow_vif = vif;
 		break;
+	case RTW_NET_NO_LINK:
+		if (rtw_wow->pno_req.inited)
+			rtwdev->wow.wow_vif = vif;
+		break;
 	default:
 		break;
 	}
@@ -673,6 +792,9 @@ static int rtw_wow_set_wakeups(struct rtw_dev *rtwdev,
 	if (wowlan->gtk_rekey_failure)
 		set_bit(RTW_WOW_FLAG_EN_REKEY_PKT, rtw_wow->flags);
 
+	if (wowlan->nd_config)
+		rtw_wow_check_pno(rtwdev, wowlan->nd_config);
+
 	rtw_iterate_vifs_atomic(rtwdev, rtw_wow_vif_iter, rtwdev);
 	if (!rtw_wow->wow_vif)
 		return -EPERM;
@@ -692,6 +814,12 @@ static int rtw_wow_set_wakeups(struct rtw_dev *rtwdev,
 static void rtw_wow_clear_wakeups(struct rtw_dev *rtwdev)
 {
 	struct rtw_wow_param *rtw_wow = &rtwdev->wow;
+	struct rtw_pno_request *pno_req = &rtw_wow->pno_req;
+
+	if (pno_req->inited) {
+		kfree(pno_req->channels);
+		kfree(pno_req->match_sets);
+	}
 
 	memset(rtw_wow, 0, sizeof(rtwdev->wow));
 }
@@ -715,6 +843,7 @@ int rtw_wow_suspend(struct rtw_dev *rtwdev, struct cfg80211_wowlan *wowlan)
 	ret = rtw_wow_enable(rtwdev);
 	if (ret) {
 		rtw_err(rtwdev, "failed to enable wow\n");
+		rtw_wow_restore_ps(rtwdev);
 		goto out;
 	}
 
@@ -746,8 +875,14 @@ int rtw_wow_resume(struct rtw_dev *rtwdev)
 	rtw_wow_show_wakeup_reason(rtwdev);
 
 	ret = rtw_wow_disable(rtwdev);
-	if (ret)
+	if (ret) {
 		rtw_err(rtwdev, "failed to disable wow\n");
+		goto out;
+	}
+
+	ret = rtw_wow_restore_ps(rtwdev);
+	if (ret)
+		rtw_err(rtwdev, "failed to restore ps to normal mode\n");
 
 out:
 	rtw_wow_clear_wakeups(rtwdev);
diff --git a/drivers/net/wireless/realtek/rtw88/wow.h b/drivers/net/wireless/realtek/rtw88/wow.h
index 54f742f4c955..289368a2cba4 100644
--- a/drivers/net/wireless/realtek/rtw88/wow.h
+++ b/drivers/net/wireless/realtek/rtw88/wow.h
@@ -5,6 +5,8 @@
 #ifndef __RTW_WOW_H__
 #define __RTW_WOW_H__
 
+#define PNO_CHECK_BYTE 4
+
 enum rtw_wow_pattern_type {
 	RTW_PATTERN_BROADCAST = 0,
 	RTW_PATTERN_MULTICAST,
@@ -20,6 +22,7 @@ enum rtw_wake_reason {
 	RTW_WOW_RSN_DISCONNECT = 0x10,
 	RTW_WOW_RSN_RX_MAGIC_PKT = 0x21,
 	RTW_WOW_RSN_RX_PATTERN_MATCH = 0x23,
+	RTW_WOW_RSN_RX_NLO = 0x55,
 };
 
 struct rtw_fw_media_status_iter_data {
@@ -41,6 +44,14 @@ static inline bool rtw_wow_mgd_linked(struct rtw_dev *rtwdev)
 	return (rtwvif->net_type == RTW_NET_MGD_LINKED);
 }
 
+static inline bool rtw_wow_no_link(struct rtw_dev *rtwdev)
+{
+	struct ieee80211_vif *wow_vif = rtwdev->wow.wow_vif;
+	struct rtw_vif *rtwvif = (struct rtw_vif *)wow_vif->drv_priv;
+
+	return (rtwvif->net_type == RTW_NET_NO_LINK);
+}
+
 int rtw_wow_suspend(struct rtw_dev *rtwdev, struct cfg80211_wowlan *wowlan);
 int rtw_wow_resume(struct rtw_dev *rtwdev);
 
-- 
cgit v1.2.3-59-g8ed1b


From 35a68fa5f96a80797e11b6952a47c5a84939a7bf Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Fri, 20 Dec 2019 17:21:46 +0800
Subject: rtw88: fix rate mask for 1SS chip

The rate mask is used to tell firmware the supported rate depends on
negotiation. We loop 2 times for all VHT/HT 2SS rate mask first, and then
only keep the part according to chip's NSS.

This commit fixes the logic error of '&' operations for VHT/HT rate, and
we should run this logic before adding legacy rate.

To access HT MCS map, index 0/1 represent MCS 0-7/8-15 respectively. Use
NL80211_BAND_xxx is incorrect, so fix it as well.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
Reviewed-by: Chris Chiu <chiu@endlessm.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtw88/main.c | 12 +++++-------
 1 file changed, 5 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw88/main.c b/drivers/net/wireless/realtek/rtw88/main.c
index 3d26c4a6e8da..9144b2696618 100644
--- a/drivers/net/wireless/realtek/rtw88/main.c
+++ b/drivers/net/wireless/realtek/rtw88/main.c
@@ -706,8 +706,8 @@ void rtw_update_sta_info(struct rtw_dev *rtwdev, struct rtw_sta_info *si)
 		if (sta->vht_cap.cap & IEEE80211_VHT_CAP_SHORT_GI_80)
 			is_support_sgi = true;
 	} else if (sta->ht_cap.ht_supported) {
-		ra_mask |= (sta->ht_cap.mcs.rx_mask[NL80211_BAND_5GHZ] << 20) |
-			   (sta->ht_cap.mcs.rx_mask[NL80211_BAND_2GHZ] << 12);
+		ra_mask |= (sta->ht_cap.mcs.rx_mask[1] << 20) |
+			   (sta->ht_cap.mcs.rx_mask[0] << 12);
 		if (sta->ht_cap.cap & IEEE80211_HT_CAP_RX_STBC)
 			stbc_en = HT_STBC_EN;
 		if (sta->ht_cap.cap & IEEE80211_HT_CAP_LDPC_CODING)
@@ -717,6 +717,9 @@ void rtw_update_sta_info(struct rtw_dev *rtwdev, struct rtw_sta_info *si)
 			is_support_sgi = true;
 	}
 
+	if (efuse->hw_cap.nss == 1)
+		ra_mask &= RA_MASK_VHT_RATES_1SS | RA_MASK_HT_RATES_1SS;
+
 	if (hal->current_band_type == RTW_BAND_5G) {
 		ra_mask |= (u64)sta->supp_rates[NL80211_BAND_5GHZ] << 4;
 		if (sta->vht_cap.vht_supported) {
@@ -750,11 +753,6 @@ void rtw_update_sta_info(struct rtw_dev *rtwdev, struct rtw_sta_info *si)
 		wireless_set = 0;
 	}
 
-	if (efuse->hw_cap.nss == 1) {
-		ra_mask &= RA_MASK_VHT_RATES_1SS;
-		ra_mask &= RA_MASK_HT_RATES_1SS;
-	}
-
 	switch (sta->bandwidth) {
 	case IEEE80211_STA_RX_BW_80:
 		bw_mode = RTW_CHANNEL_WIDTH_80;
-- 
cgit v1.2.3-59-g8ed1b


From 73a2d0b83424f7150a1f607470dac67c95b8b2d4 Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Fri, 20 Dec 2019 17:21:47 +0800
Subject: rtw88: fix TX secondary channel offset of 40M if current bw is 20M or
 40M

TX secondary channel offset is valid only if current bandwidth is 80M,
otherwise leave this value as zero. The wrong value of txsc40 causes
MAC unpredictable behavior.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtw88/mac.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw88/mac.c b/drivers/net/wireless/realtek/rtw88/mac.c
index 03a82de053ea..cadf0abbe16b 100644
--- a/drivers/net/wireless/realtek/rtw88/mac.c
+++ b/drivers/net/wireless/realtek/rtw88/mac.c
@@ -16,10 +16,12 @@ void rtw_set_channel_mac(struct rtw_dev *rtwdev, u8 channel, u8 bw,
 	u8 value8;
 
 	txsc20 = primary_ch_idx;
-	if (txsc20 == 1 || txsc20 == 3)
-		txsc40 = 9;
-	else
-		txsc40 = 10;
+	if (bw == RTW_CHANNEL_WIDTH_80) {
+		if (txsc20 == 1 || txsc20 == 3)
+			txsc40 = 9;
+		else
+			txsc40 = 10;
+	}
 	rtw_write8(rtwdev, REG_DATA_SC,
 		   BIT_TXSC_20M(txsc20) | BIT_TXSC_40M(txsc40));
 
-- 
cgit v1.2.3-59-g8ed1b


From e339b649352437c529616daa0c15db04c76f0e85 Mon Sep 17 00:00:00 2001
From: Tzu-En Huang <tehuang@realtek.com>
Date: Fri, 20 Dec 2019 17:21:49 +0800
Subject: rtw88: 8822c: update power sequence to v15

Update card enable power sequence flow, to fix CMD11 fail after
reboot and wrong PLL clock.

Signed-off-by: Tzu-En Huang <tehuang@realtek.com>
Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtw88/rtw8822c.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw88/rtw8822c.c b/drivers/net/wireless/realtek/rtw88/rtw8822c.c
index 655ca9265f68..3865097696d4 100644
--- a/drivers/net/wireless/realtek/rtw88/rtw8822c.c
+++ b/drivers/net/wireless/realtek/rtw88/rtw8822c.c
@@ -3487,12 +3487,12 @@ static struct rtw_pwr_seq_cmd trans_cardemu_to_act_8822c[] = {
 	 RTW_PWR_CUT_ALL_MSK,
 	 RTW_PWR_INTF_ALL_MSK,
 	 RTW_PWR_ADDR_MAC,
-	 RTW_PWR_CMD_WRITE, BIT(7), 0},
-	{0x0005,
+	 RTW_PWR_CMD_WRITE, (BIT(4) | BIT(3)), 0},
+	{0x1018,
 	 RTW_PWR_CUT_ALL_MSK,
 	 RTW_PWR_INTF_ALL_MSK,
 	 RTW_PWR_ADDR_MAC,
-	 RTW_PWR_CMD_WRITE, (BIT(4) | BIT(3)), 0},
+	 RTW_PWR_CMD_WRITE, BIT(2), BIT(2)},
 	{0x0005,
 	 RTW_PWR_CUT_ALL_MSK,
 	 RTW_PWR_INTF_ALL_MSK,
-- 
cgit v1.2.3-59-g8ed1b


From 3f43f10bd619b1acebb570bebf989265920aeb27 Mon Sep 17 00:00:00 2001
From: Tzu-En Huang <tehuang@realtek.com>
Date: Fri, 20 Dec 2019 17:21:51 +0800
Subject: rtw88: remove unused spinlock

dm_lock is never used. Thus, remove this redundant spinlock.

Signed-off-by: Tzu-En Huang <tehuang@realtek.com>
Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtw88/main.c | 1 -
 drivers/net/wireless/realtek/rtw88/main.h | 3 ---
 2 files changed, 4 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw88/main.c b/drivers/net/wireless/realtek/rtw88/main.c
index 9144b2696618..2a60f8333df0 100644
--- a/drivers/net/wireless/realtek/rtw88/main.c
+++ b/drivers/net/wireless/realtek/rtw88/main.c
@@ -1376,7 +1376,6 @@ int rtw_core_init(struct rtw_dev *rtwdev)
 	skb_queue_head_init(&rtwdev->coex.queue);
 	skb_queue_head_init(&rtwdev->tx_report.queue);
 
-	spin_lock_init(&rtwdev->dm_lock);
 	spin_lock_init(&rtwdev->rf_lock);
 	spin_lock_init(&rtwdev->h2c.lock);
 	spin_lock_init(&rtwdev->txq_lock);
diff --git a/drivers/net/wireless/realtek/rtw88/main.h b/drivers/net/wireless/realtek/rtw88/main.h
index 4fa17aab3838..0c9bac08ebf7 100644
--- a/drivers/net/wireless/realtek/rtw88/main.h
+++ b/drivers/net/wireless/realtek/rtw88/main.h
@@ -1586,9 +1586,6 @@ struct rtw_dev {
 	/* ensures exclusive access from mac80211 callbacks */
 	struct mutex mutex;
 
-	/* lock for dm to use */
-	spinlock_t dm_lock;
-
 	/* read/write rf register */
 	spinlock_t rf_lock;
 
-- 
cgit v1.2.3-59-g8ed1b


From 962562cde15463f712e9cc007086f6140930199d Mon Sep 17 00:00:00 2001
From: Yan-Hsuan Chuang <yhchuang@realtek.com>
Date: Fri, 20 Dec 2019 17:21:52 +0800
Subject: rtw88: remove unused variable 'in_lps'

Unused, will not be used neither, because the hardware/firmware
can only support one vif for LPS currnetly. If there's more than
one vif, than driver will never enter LPS. So remove it.

Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtw88/mac80211.c | 1 -
 drivers/net/wireless/realtek/rtw88/main.h     | 1 -
 2 files changed, 2 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw88/mac80211.c b/drivers/net/wireless/realtek/rtw88/mac80211.c
index 1150e42539a9..c8be4b4a9371 100644
--- a/drivers/net/wireless/realtek/rtw88/mac80211.c
+++ b/drivers/net/wireless/realtek/rtw88/mac80211.c
@@ -158,7 +158,6 @@ static int rtw_ops_add_interface(struct ieee80211_hw *hw,
 	rtwvif->stats.rx_unicast = 0;
 	rtwvif->stats.tx_cnt = 0;
 	rtwvif->stats.rx_cnt = 0;
-	rtwvif->in_lps = false;
 	memset(&rtwvif->bfee, 0, sizeof(struct rtw_bfee));
 	rtwvif->conf = &rtw_vif_port[port];
 	rtw_txq_init(rtwdev, vif->txq);
diff --git a/drivers/net/wireless/realtek/rtw88/main.h b/drivers/net/wireless/realtek/rtw88/main.h
index 0c9bac08ebf7..107b17c86f9f 100644
--- a/drivers/net/wireless/realtek/rtw88/main.h
+++ b/drivers/net/wireless/realtek/rtw88/main.h
@@ -747,7 +747,6 @@ struct rtw_vif {
 	const struct rtw_vif_port *conf;
 
 	struct rtw_traffic_stats stats;
-	bool in_lps;
 
 	struct rtw_bfee bfee;
 };
-- 
cgit v1.2.3-59-g8ed1b


From 65ae64d375751cad79c6e74222d348f72e310e4c Mon Sep 17 00:00:00 2001
From: Yan-Hsuan Chuang <yhchuang@realtek.com>
Date: Fri, 20 Dec 2019 17:21:53 +0800
Subject: rtw88: remove unused vif pointer in struct rtw_vif

As driver can easily get vif with container_of(), we can
just remove it.

Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtw88/mac80211.c | 1 -
 drivers/net/wireless/realtek/rtw88/main.h     | 1 -
 2 files changed, 2 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw88/mac80211.c b/drivers/net/wireless/realtek/rtw88/mac80211.c
index c8be4b4a9371..6fc33e11d08c 100644
--- a/drivers/net/wireless/realtek/rtw88/mac80211.c
+++ b/drivers/net/wireless/realtek/rtw88/mac80211.c
@@ -153,7 +153,6 @@ static int rtw_ops_add_interface(struct ieee80211_hw *hw,
 	u8 bcn_ctrl = 0;
 
 	rtwvif->port = port;
-	rtwvif->vif = vif;
 	rtwvif->stats.tx_unicast = 0;
 	rtwvif->stats.rx_unicast = 0;
 	rtwvif->stats.tx_cnt = 0;
diff --git a/drivers/net/wireless/realtek/rtw88/main.h b/drivers/net/wireless/realtek/rtw88/main.h
index 107b17c86f9f..f334d201bfb5 100644
--- a/drivers/net/wireless/realtek/rtw88/main.h
+++ b/drivers/net/wireless/realtek/rtw88/main.h
@@ -736,7 +736,6 @@ struct rtw_bf_info {
 };
 
 struct rtw_vif {
-	struct ieee80211_vif *vif;
 	enum rtw_net_type net_type;
 	u16 aid;
 	u8 mac_addr[ETH_ALEN];
-- 
cgit v1.2.3-59-g8ed1b


From fc83c616d4d95c386e2bac7994b635b24662bca4 Mon Sep 17 00:00:00 2001
From: Chin-Yen Lee <timlee@realtek.com>
Date: Fri, 20 Dec 2019 17:21:54 +0800
Subject: rtw88: use rtw_hci_stop() instead of rtwdev->hci.ops->stop()

Fix typo, should use rtw_hci_stop()

Signed-off-by: Chin-Yen Lee <timlee@realtek.com>
Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtw88/main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw88/main.c b/drivers/net/wireless/realtek/rtw88/main.c
index 2a60f8333df0..2845d2838f7b 100644
--- a/drivers/net/wireless/realtek/rtw88/main.c
+++ b/drivers/net/wireless/realtek/rtw88/main.c
@@ -898,7 +898,7 @@ int rtw_core_start(struct rtw_dev *rtwdev)
 
 static void rtw_power_off(struct rtw_dev *rtwdev)
 {
-	rtwdev->hci.ops->stop(rtwdev);
+	rtw_hci_stop(rtwdev);
 	rtw_mac_power_off(rtwdev);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From f48abf064ade7a08015bc3a50a873437375843b2 Mon Sep 17 00:00:00 2001
From: Yan-Hsuan Chuang <yhchuang@realtek.com>
Date: Fri, 20 Dec 2019 17:21:55 +0800
Subject: rtw88: assign NULL to skb after being kfree()'ed

Should assign NULL to skb after kfree(), in case of driver
trying to free the same skb again.

This could happen if driver failed to allocate an skb when
building reserved page.

Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtw88/fw.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw88/fw.c b/drivers/net/wireless/realtek/rtw88/fw.c
index e3e91ae0d3d2..243441453ead 100644
--- a/drivers/net/wireless/realtek/rtw88/fw.c
+++ b/drivers/net/wireless/realtek/rtw88/fw.c
@@ -1139,13 +1139,16 @@ static u8 *rtw_build_rsvd_page(struct rtw_dev *rtwdev,
 			page += rtw_len_to_page(rsvd_pkt->skb->len, page_size);
 
 		kfree_skb(rsvd_pkt->skb);
+		rsvd_pkt->skb = NULL;
 	}
 
 	return buf;
 
 release_skb:
-	list_for_each_entry(rsvd_pkt, &rtwdev->rsvd_page_list, list)
+	list_for_each_entry(rsvd_pkt, &rtwdev->rsvd_page_list, list) {
 		kfree_skb(rsvd_pkt->skb);
+		rsvd_pkt->skb = NULL;
+	}
 
 	return NULL;
 }
-- 
cgit v1.2.3-59-g8ed1b


From bbdd1d854e0a97b771ab642fc07992b28e853036 Mon Sep 17 00:00:00 2001
From: Zong-Zhe Yang <kevin_yang@realtek.com>
Date: Fri, 20 Dec 2019 17:21:56 +0800
Subject: rtw88: change max_num_of_tx_queue() definition to inline in pci.h

It's more reasonable to define max_num_of_tx_queue() as an inline function.

Signed-off-by: Zong-Zhe Yang <kevin_yang@realtek.com>
Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtw88/pci.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw88/pci.h b/drivers/net/wireless/realtek/rtw88/pci.h
index 49bf29a92152..1580cfc57361 100644
--- a/drivers/net/wireless/realtek/rtw88/pci.h
+++ b/drivers/net/wireless/realtek/rtw88/pci.h
@@ -210,7 +210,7 @@ struct rtw_pci {
 	void __iomem *mmap;
 };
 
-static u32 max_num_of_tx_queue(u8 queue)
+static inline u32 max_num_of_tx_queue(u8 queue)
 {
 	u32 max_num;
 
-- 
cgit v1.2.3-59-g8ed1b


From fd156bdf62bb06271084f45eff9384b24f734fd8 Mon Sep 17 00:00:00 2001
From: Amadeusz Sławiński <amade@asmblr.net>
Date: Mon, 23 Dec 2019 13:37:07 +0100
Subject: rtlwifi: rtl8192cu: Fix typo
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Replace USB_VENDER_ID_REALTEK with USB_VENDOR_ID_REALTEK.

Signed-off-by: Amadeusz Sławiński <amade@asmblr.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 .../net/wireless/realtek/rtlwifi/rtl8192cu/sw.c    | 34 +++++++++++-----------
 1 file changed, 17 insertions(+), 17 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/sw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/sw.c
index ab3e4aebad39..6b954059e830 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/sw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/sw.c
@@ -252,45 +252,45 @@ static struct rtl_hal_cfg rtl92cu_hal_cfg = {
 	.maps[RTL_RC_HT_RATEMCS15] = DESC_RATEMCS15,
 };
 
-#define USB_VENDER_ID_REALTEK		0x0bda
+#define USB_VENDOR_ID_REALTEK		0x0bda
 
 /* 2010-10-19 DID_USB_V3.4 */
 static const struct usb_device_id rtl8192c_usb_ids[] = {
 
 	/*=== Realtek demoboard ===*/
 	/* Default ID */
-	{RTL_USB_DEVICE(USB_VENDER_ID_REALTEK, 0x8191, rtl92cu_hal_cfg)},
+	{RTL_USB_DEVICE(USB_VENDOR_ID_REALTEK, 0x8191, rtl92cu_hal_cfg)},
 
 	/****** 8188CU ********/
 	/* RTL8188CTV */
-	{RTL_USB_DEVICE(USB_VENDER_ID_REALTEK, 0x018a, rtl92cu_hal_cfg)},
+	{RTL_USB_DEVICE(USB_VENDOR_ID_REALTEK, 0x018a, rtl92cu_hal_cfg)},
 	/* 8188CE-VAU USB minCard */
-	{RTL_USB_DEVICE(USB_VENDER_ID_REALTEK, 0x8170, rtl92cu_hal_cfg)},
+	{RTL_USB_DEVICE(USB_VENDOR_ID_REALTEK, 0x8170, rtl92cu_hal_cfg)},
 	/* 8188cu 1*1 dongle */
-	{RTL_USB_DEVICE(USB_VENDER_ID_REALTEK, 0x8176, rtl92cu_hal_cfg)},
+	{RTL_USB_DEVICE(USB_VENDOR_ID_REALTEK, 0x8176, rtl92cu_hal_cfg)},
 	/* 8188cu 1*1 dongle, (b/g mode only) */
-	{RTL_USB_DEVICE(USB_VENDER_ID_REALTEK, 0x8177, rtl92cu_hal_cfg)},
+	{RTL_USB_DEVICE(USB_VENDOR_ID_REALTEK, 0x8177, rtl92cu_hal_cfg)},
 	/* 8188cu Slim Solo */
-	{RTL_USB_DEVICE(USB_VENDER_ID_REALTEK, 0x817a, rtl92cu_hal_cfg)},
+	{RTL_USB_DEVICE(USB_VENDOR_ID_REALTEK, 0x817a, rtl92cu_hal_cfg)},
 	/* 8188cu Slim Combo */
-	{RTL_USB_DEVICE(USB_VENDER_ID_REALTEK, 0x817b, rtl92cu_hal_cfg)},
+	{RTL_USB_DEVICE(USB_VENDOR_ID_REALTEK, 0x817b, rtl92cu_hal_cfg)},
 	/* 8188RU High-power USB Dongle */
-	{RTL_USB_DEVICE(USB_VENDER_ID_REALTEK, 0x817d, rtl92cu_hal_cfg)},
+	{RTL_USB_DEVICE(USB_VENDOR_ID_REALTEK, 0x817d, rtl92cu_hal_cfg)},
 	/* 8188CE-VAU USB minCard (b/g mode only) */
-	{RTL_USB_DEVICE(USB_VENDER_ID_REALTEK, 0x817e, rtl92cu_hal_cfg)},
+	{RTL_USB_DEVICE(USB_VENDOR_ID_REALTEK, 0x817e, rtl92cu_hal_cfg)},
 	/* 8188RU in Alfa AWUS036NHR */
-	{RTL_USB_DEVICE(USB_VENDER_ID_REALTEK, 0x817f, rtl92cu_hal_cfg)},
+	{RTL_USB_DEVICE(USB_VENDOR_ID_REALTEK, 0x817f, rtl92cu_hal_cfg)},
 	/* RTL8188CUS-VL */
-	{RTL_USB_DEVICE(USB_VENDER_ID_REALTEK, 0x818a, rtl92cu_hal_cfg)},
-	{RTL_USB_DEVICE(USB_VENDER_ID_REALTEK, 0x819a, rtl92cu_hal_cfg)},
+	{RTL_USB_DEVICE(USB_VENDOR_ID_REALTEK, 0x818a, rtl92cu_hal_cfg)},
+	{RTL_USB_DEVICE(USB_VENDOR_ID_REALTEK, 0x819a, rtl92cu_hal_cfg)},
 	/* 8188 Combo for BC4 */
-	{RTL_USB_DEVICE(USB_VENDER_ID_REALTEK, 0x8754, rtl92cu_hal_cfg)},
+	{RTL_USB_DEVICE(USB_VENDOR_ID_REALTEK, 0x8754, rtl92cu_hal_cfg)},
 
 	/****** 8192CU ********/
 	/* 8192cu 2*2 */
-	{RTL_USB_DEVICE(USB_VENDER_ID_REALTEK, 0x8178, rtl92cu_hal_cfg)},
+	{RTL_USB_DEVICE(USB_VENDOR_ID_REALTEK, 0x8178, rtl92cu_hal_cfg)},
 	/* 8192CE-VAU USB minCard */
-	{RTL_USB_DEVICE(USB_VENDER_ID_REALTEK, 0x817c, rtl92cu_hal_cfg)},
+	{RTL_USB_DEVICE(USB_VENDOR_ID_REALTEK, 0x817c, rtl92cu_hal_cfg)},
 
 	/*=== Customer ID ===*/
 	/****** 8188CU ********/
@@ -329,7 +329,7 @@ static const struct usb_device_id rtl8192c_usb_ids[] = {
 
 	/****** 8188 RU ********/
 	/* Netcore */
-	{RTL_USB_DEVICE(USB_VENDER_ID_REALTEK, 0x317f, rtl92cu_hal_cfg)},
+	{RTL_USB_DEVICE(USB_VENDOR_ID_REALTEK, 0x317f, rtl92cu_hal_cfg)},
 
 	/****** 8188CUS Slim Solo********/
 	{RTL_USB_DEVICE(0x04f2, 0xaff7, rtl92cu_hal_cfg)}, /*Xavi*/
-- 
cgit v1.2.3-59-g8ed1b


From 5d3f9145f54ee81b33752f68285eb15ef644e08a Mon Sep 17 00:00:00 2001
From: Amadeusz Sławiński <amade@asmblr.net>
Date: Mon, 23 Dec 2019 13:37:08 +0100
Subject: rtlwifi: rtl8188ee: Make functions static & rm sw.h
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Some of functions which were exposed in sw.h, are only used in sw.c, so
just make them static. This makes sw.h unnecessary, so remove it.

Signed-off-by: Amadeusz Sławiński <amade@asmblr.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtlwifi/rtl8188ee/sw.c |  7 +++----
 drivers/net/wireless/realtek/rtlwifi/rtl8188ee/sw.h | 12 ------------
 2 files changed, 3 insertions(+), 16 deletions(-)
 delete mode 100644 drivers/net/wireless/realtek/rtlwifi/rtl8188ee/sw.h

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/sw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/sw.c
index a0eda51e833c..4865639ac9ea 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/sw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/sw.c
@@ -9,7 +9,6 @@
 #include "phy.h"
 #include "dm.h"
 #include "hw.h"
-#include "sw.h"
 #include "trx.h"
 #include "led.h"
 #include "table.h"
@@ -59,7 +58,7 @@ static void rtl88e_init_aspm_vars(struct ieee80211_hw *hw)
 	rtlpci->const_support_pciaspm = rtlpriv->cfg->mod_params->aspm_support;
 }
 
-int rtl88e_init_sw_vars(struct ieee80211_hw *hw)
+static int rtl88e_init_sw_vars(struct ieee80211_hw *hw)
 {
 	int err = 0;
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
@@ -173,7 +172,7 @@ int rtl88e_init_sw_vars(struct ieee80211_hw *hw)
 	return err;
 }
 
-void rtl88e_deinit_sw_vars(struct ieee80211_hw *hw)
+static void rtl88e_deinit_sw_vars(struct ieee80211_hw *hw)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 
@@ -189,7 +188,7 @@ void rtl88e_deinit_sw_vars(struct ieee80211_hw *hw)
 }
 
 /* get bt coexist status */
-bool rtl88e_get_btc_status(void)
+static bool rtl88e_get_btc_status(void)
 {
 	return false;
 }
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/sw.h b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/sw.h
deleted file mode 100644
index 1407151503f8..000000000000
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/sw.h
+++ /dev/null
@@ -1,12 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright(c) 2009-2013  Realtek Corporation.*/
-
-#ifndef __RTL92CE_SW_H__
-#define __RTL92CE_SW_H__
-
-int rtl88e_init_sw_vars(struct ieee80211_hw *hw);
-void rtl88e_deinit_sw_vars(struct ieee80211_hw *hw);
-bool rtl88e_get_btc_status(void);
-
-
-#endif
-- 
cgit v1.2.3-59-g8ed1b


From be913e3f49ac943c230dca5de40974f3a5f7ae90 Mon Sep 17 00:00:00 2001
From: Amadeusz Sławiński <amade@asmblr.net>
Date: Mon, 23 Dec 2019 13:37:09 +0100
Subject: rtlwifi: rtl8192ce: Make functions static & rm sw.h
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Some of functions which were exposed in sw.h, are only used in sw.c, so
just make them static. The rtl92c_init_var_map function is not defined
anywhere, while declared in sw.h. Two other functions are also declared
in phy.h (which is included in sw.c) and their definitions are in phy.c
Overall sw.h is unnecessary and can be removed.

Signed-off-by: Amadeusz Sławiński <amade@asmblr.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtlwifi/rtl8192ce/sw.c |  5 ++---
 drivers/net/wireless/realtek/rtlwifi/rtl8192ce/sw.h | 15 ---------------
 2 files changed, 2 insertions(+), 18 deletions(-)
 delete mode 100644 drivers/net/wireless/realtek/rtlwifi/rtl8192ce/sw.h

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/sw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/sw.c
index 900788e4018c..ed68c850f9a2 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/sw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/sw.c
@@ -14,7 +14,6 @@
 #include "../rtl8192c/phy_common.h"
 #include "hw.h"
 #include "rf.h"
-#include "sw.h"
 #include "trx.h"
 #include "led.h"
 
@@ -65,7 +64,7 @@ static void rtl92c_init_aspm_vars(struct ieee80211_hw *hw)
 	rtlpci->const_support_pciaspm = rtlpriv->cfg->mod_params->aspm_support;
 }
 
-int rtl92c_init_sw_vars(struct ieee80211_hw *hw)
+static int rtl92c_init_sw_vars(struct ieee80211_hw *hw)
 {
 	int err;
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
@@ -161,7 +160,7 @@ int rtl92c_init_sw_vars(struct ieee80211_hw *hw)
 	return 0;
 }
 
-void rtl92c_deinit_sw_vars(struct ieee80211_hw *hw)
+static void rtl92c_deinit_sw_vars(struct ieee80211_hw *hw)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/sw.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/sw.h
deleted file mode 100644
index f2d121a60159..000000000000
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/sw.h
+++ /dev/null
@@ -1,15 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright(c) 2009-2012  Realtek Corporation.*/
-
-#ifndef __RTL92CE_SW_H__
-#define __RTL92CE_SW_H__
-
-int rtl92c_init_sw_vars(struct ieee80211_hw *hw);
-void rtl92c_deinit_sw_vars(struct ieee80211_hw *hw);
-void rtl92c_init_var_map(struct ieee80211_hw *hw);
-bool _rtl92ce_phy_config_bb_with_headerfile(struct ieee80211_hw *hw,
-					    u8 configtype);
-bool _rtl92ce_phy_config_bb_with_pgheaderfile(struct ieee80211_hw *hw,
-					      u8 configtype);
-
-#endif
-- 
cgit v1.2.3-59-g8ed1b


From a3cda3c363ca747758a3bcc6fc23529003a1c983 Mon Sep 17 00:00:00 2001
From: Amadeusz Sławiński <amade@asmblr.net>
Date: Mon, 23 Dec 2019 13:37:10 +0100
Subject: rtlwifi: rtl8192cu: Remove sw.h header
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

It has one define, which is already defined in include from reg.h.
All functions are declared in their own headers and included in *.c
files belonging to them.
This makes sw.h unnecessary, so we can remove it.

Signed-off-by: Amadeusz Sławiński <amade@asmblr.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 .../net/wireless/realtek/rtlwifi/rtl8192cu/sw.c    |  1 -
 .../net/wireless/realtek/rtlwifi/rtl8192cu/sw.h    | 27 ----------------------
 2 files changed, 28 deletions(-)
 delete mode 100644 drivers/net/wireless/realtek/rtlwifi/rtl8192cu/sw.h

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/sw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/sw.c
index 6b954059e830..b53daf1b29f7 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/sw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/sw.c
@@ -12,7 +12,6 @@
 #include "mac.h"
 #include "dm.h"
 #include "rf.h"
-#include "sw.h"
 #include "trx.h"
 #include "led.h"
 #include "hw.h"
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/sw.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/sw.h
deleted file mode 100644
index 662440c4cdc9..000000000000
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/sw.h
+++ /dev/null
@@ -1,27 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright(c) 2009-2012  Realtek Corporation.*/
-
-#ifndef __RTL92CU_SW_H__
-#define __RTL92CU_SW_H__
-
-#define EFUSE_MAX_SECTION	16
-
-void rtl92cu_phy_rf6052_set_cck_txpower(struct ieee80211_hw *hw,
-					u8 *powerlevel);
-void rtl92cu_phy_rf6052_set_ofdm_txpower(struct ieee80211_hw *hw,
-					u8 *ppowerlevel, u8 channel);
-bool _rtl92cu_phy_config_bb_with_headerfile(struct ieee80211_hw *hw,
-					    u8 configtype);
-bool _rtl92cu_phy_config_bb_with_pgheaderfile(struct ieee80211_hw *hw,
-						    u8 configtype);
-void _rtl92cu_phy_lc_calibrate(struct ieee80211_hw *hw, bool is2t);
-void rtl92cu_phy_set_rf_reg(struct ieee80211_hw *hw,
-			   enum radio_path rfpath,
-			   u32 regaddr, u32 bitmask, u32 data);
-bool rtl92cu_phy_set_rf_power_state(struct ieee80211_hw *hw,
-				   enum rf_pwrstate rfpwr_state);
-u32 rtl92cu_phy_query_rf_reg(struct ieee80211_hw *hw,
-			    enum radio_path rfpath, u32 regaddr, u32 bitmask);
-void rtl92cu_phy_set_bw_mode_callback(struct ieee80211_hw *hw);
-
-#endif
-- 
cgit v1.2.3-59-g8ed1b


From fef8a2d969af25638f48411930f491088d6ccea0 Mon Sep 17 00:00:00 2001
From: Amadeusz Sławiński <amade@asmblr.net>
Date: Mon, 23 Dec 2019 13:37:11 +0100
Subject: rtlwifi: rtl8192ee: Make functions static & rm sw.h
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Some of functions which were exposed in sw.h, are only used in sw.c, so
just make them static. This makes sw.h unnecessary, so remove it.

Signed-off-by: Amadeusz Sławiński <amade@asmblr.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtlwifi/rtl8192ee/sw.c |  7 +++----
 drivers/net/wireless/realtek/rtlwifi/rtl8192ee/sw.h | 11 -----------
 2 files changed, 3 insertions(+), 15 deletions(-)
 delete mode 100644 drivers/net/wireless/realtek/rtlwifi/rtl8192ee/sw.h

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/sw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/sw.c
index b6ee7dae5943..b337d599b6f4 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/sw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/sw.c
@@ -9,7 +9,6 @@
 #include "phy.h"
 #include "dm.h"
 #include "hw.h"
-#include "sw.h"
 #include "fw.h"
 #include "trx.h"
 #include "led.h"
@@ -65,7 +64,7 @@ static void rtl92ee_init_aspm_vars(struct ieee80211_hw *hw)
 	rtlpci->const_support_pciaspm = rtlpriv->cfg->mod_params->aspm_support;
 }
 
-int rtl92ee_init_sw_vars(struct ieee80211_hw *hw)
+static int rtl92ee_init_sw_vars(struct ieee80211_hw *hw)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 	struct rtl_pci *rtlpci = rtl_pcidev(rtl_pcipriv(hw));
@@ -164,7 +163,7 @@ int rtl92ee_init_sw_vars(struct ieee80211_hw *hw)
 	return 0;
 }
 
-void rtl92ee_deinit_sw_vars(struct ieee80211_hw *hw)
+static void rtl92ee_deinit_sw_vars(struct ieee80211_hw *hw)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 
@@ -175,7 +174,7 @@ void rtl92ee_deinit_sw_vars(struct ieee80211_hw *hw)
 }
 
 /* get bt coexist status */
-bool rtl92ee_get_btc_status(void)
+static bool rtl92ee_get_btc_status(void)
 {
 	return true;
 }
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/sw.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/sw.h
deleted file mode 100644
index 36e29a2da0fd..000000000000
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/sw.h
+++ /dev/null
@@ -1,11 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright(c) 2009-2014  Realtek Corporation.*/
-
-#ifndef __RTL92E_SW_H__
-#define __RTL92E_SW_H__
-
-int rtl92ee_init_sw_vars(struct ieee80211_hw *hw);
-void rtl92ee_deinit_sw_vars(struct ieee80211_hw *hw);
-bool rtl92ee_get_btc_status(void);
-
-#endif
-- 
cgit v1.2.3-59-g8ed1b


From 5b2640835a3443e629ee2af4b8367f6abdeb142a Mon Sep 17 00:00:00 2001
From: Amadeusz Sławiński <amade@asmblr.net>
Date: Mon, 23 Dec 2019 13:37:12 +0100
Subject: rtlwifi: rtl8192se: Remove sw.h header
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

It has one define, which is already defined in include from reg.h.
All the declared functions are not implemented anywhere, sw.c has
ones with similar names which are already static.

Signed-off-by: Amadeusz Sławiński <amade@asmblr.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtlwifi/rtl8192se/sw.c |  1 -
 drivers/net/wireless/realtek/rtlwifi/rtl8192se/sw.h | 13 -------------
 2 files changed, 14 deletions(-)
 delete mode 100644 drivers/net/wireless/realtek/rtlwifi/rtl8192se/sw.h

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192se/sw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192se/sw.c
index 1c7ee569f4bf..7a54497b7df2 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192se/sw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192se/sw.c
@@ -11,7 +11,6 @@
 #include "dm.h"
 #include "fw.h"
 #include "hw.h"
-#include "sw.h"
 #include "trx.h"
 #include "led.h"
 
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192se/sw.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192se/sw.h
deleted file mode 100644
index a31efba0e6b5..000000000000
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192se/sw.h
+++ /dev/null
@@ -1,13 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright(c) 2009-2012  Realtek Corporation.*/
-
-#ifndef __REALTEK_PCI92SE_SW_H__
-#define __REALTEK_PCI92SE_SW_H__
-
-#define EFUSE_MAX_SECTION	16
-
-int rtl92se_init_sw(struct ieee80211_hw *hw);
-void rtl92se_deinit_sw(struct ieee80211_hw *hw);
-void rtl92se_init_var_map(struct ieee80211_hw *hw);
-
-#endif
-- 
cgit v1.2.3-59-g8ed1b


From c218acfaa637d02830549cf27798921f820ec818 Mon Sep 17 00:00:00 2001
From: Amadeusz Sławiński <amade@asmblr.net>
Date: Mon, 23 Dec 2019 13:37:13 +0100
Subject: rtlwifi: rtl8723ae: Make functions static & rm sw.h
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Some of functions which were exposed in sw.h, are only used in sw.c, so
just make them static. This makes sw.h unnecessary, so remove it.

Signed-off-by: Amadeusz Sławiński <amade@asmblr.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtlwifi/rtl8723ae/sw.c |  7 +++----
 drivers/net/wireless/realtek/rtlwifi/rtl8723ae/sw.h | 13 -------------
 2 files changed, 3 insertions(+), 17 deletions(-)
 delete mode 100644 drivers/net/wireless/realtek/rtlwifi/rtl8723ae/sw.h

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/sw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/sw.c
index 5702ac6deebf..ea86d5bf33d2 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/sw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/sw.c
@@ -11,7 +11,6 @@
 #include "fw.h"
 #include "../rtl8723com/fw_common.h"
 #include "hw.h"
-#include "sw.h"
 #include "trx.h"
 #include "led.h"
 #include "table.h"
@@ -67,7 +66,7 @@ static void rtl8723e_init_aspm_vars(struct ieee80211_hw *hw)
 	rtlpci->const_support_pciaspm = rtlpriv->cfg->mod_params->aspm_support;
 }
 
-int rtl8723e_init_sw_vars(struct ieee80211_hw *hw)
+static int rtl8723e_init_sw_vars(struct ieee80211_hw *hw)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 	struct rtl_pci *rtlpci = rtl_pcidev(rtl_pcipriv(hw));
@@ -166,7 +165,7 @@ int rtl8723e_init_sw_vars(struct ieee80211_hw *hw)
 	return 0;
 }
 
-void rtl8723e_deinit_sw_vars(struct ieee80211_hw *hw)
+static void rtl8723e_deinit_sw_vars(struct ieee80211_hw *hw)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 
@@ -177,7 +176,7 @@ void rtl8723e_deinit_sw_vars(struct ieee80211_hw *hw)
 }
 
 /* get bt coexist status */
-bool rtl8723e_get_btc_status(void)
+static bool rtl8723e_get_btc_status(void)
 {
 	return true;
 }
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/sw.h b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/sw.h
deleted file mode 100644
index 200ce39a0dd7..000000000000
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/sw.h
+++ /dev/null
@@ -1,13 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright(c) 2009-2012  Realtek Corporation.*/
-
-#ifndef __RTL8723E_SW_H__
-#define __RTL8723E_SW_H__
-
-int rtl8723e_init_sw_vars(struct ieee80211_hw *hw);
-void rtl8723e_deinit_sw_vars(struct ieee80211_hw *hw);
-void rtl8723e_init_var_map(struct ieee80211_hw *hw);
-bool rtl8723e_get_btc_status(void);
-
-
-#endif
-- 
cgit v1.2.3-59-g8ed1b


From 02a214e29ea4415437d9e38c6bf5604e564397da Mon Sep 17 00:00:00 2001
From: Amadeusz Sławiński <amade@asmblr.net>
Date: Mon, 23 Dec 2019 13:37:14 +0100
Subject: rtlwifi: rtl8723be: Make functions static & rm sw.h
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Some of functions which were exposed in sw.h, are only used in sw.c, so
just make them static. This makes sw.h unnecessary, so remove it.

Signed-off-by: Amadeusz Sławiński <amade@asmblr.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtlwifi/rtl8723be/sw.c |  7 +++----
 drivers/net/wireless/realtek/rtlwifi/rtl8723be/sw.h | 13 -------------
 2 files changed, 3 insertions(+), 17 deletions(-)
 delete mode 100644 drivers/net/wireless/realtek/rtlwifi/rtl8723be/sw.h

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/sw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/sw.c
index 3c8528f0ecb3..36209ac5b208 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/sw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/sw.c
@@ -13,7 +13,6 @@
 #include "hw.h"
 #include "fw.h"
 #include "../rtl8723com/fw_common.h"
-#include "sw.h"
 #include "trx.h"
 #include "led.h"
 #include "table.h"
@@ -64,7 +63,7 @@ static void rtl8723be_init_aspm_vars(struct ieee80211_hw *hw)
 	rtlpci->const_support_pciaspm = rtlpriv->cfg->mod_params->aspm_support;
 }
 
-int rtl8723be_init_sw_vars(struct ieee80211_hw *hw)
+static int rtl8723be_init_sw_vars(struct ieee80211_hw *hw)
 {
 	int err = 0;
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
@@ -170,7 +169,7 @@ int rtl8723be_init_sw_vars(struct ieee80211_hw *hw)
 	return 0;
 }
 
-void rtl8723be_deinit_sw_vars(struct ieee80211_hw *hw)
+static void rtl8723be_deinit_sw_vars(struct ieee80211_hw *hw)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 
@@ -181,7 +180,7 @@ void rtl8723be_deinit_sw_vars(struct ieee80211_hw *hw)
 }
 
 /* get bt coexist status */
-bool rtl8723be_get_btc_status(void)
+static bool rtl8723be_get_btc_status(void)
 {
 	return true;
 }
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/sw.h b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/sw.h
deleted file mode 100644
index 6ecacf9fbfd7..000000000000
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/sw.h
+++ /dev/null
@@ -1,13 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright(c) 2009-2014  Realtek Corporation.*/
-
-#ifndef __RTL8723BE_SW_H__
-#define __RTL8723BE_SW_H__
-
-int rtl8723be_init_sw_vars(struct ieee80211_hw *hw);
-void rtl8723be_deinit_sw_vars(struct ieee80211_hw *hw);
-void rtl8723be_init_var_map(struct ieee80211_hw *hw);
-bool rtl8723be_get_btc_status(void);
-
-
-#endif
-- 
cgit v1.2.3-59-g8ed1b


From 8ddd4a2a6b5930b8318b994e416db4491795a8aa Mon Sep 17 00:00:00 2001
From: Amadeusz Sławiński <amade@asmblr.net>
Date: Mon, 23 Dec 2019 13:37:15 +0100
Subject: rtlwifi: rtl8821ae: Make functions static & rm sw.h
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Some of functions which were exposed in sw.h, are only used in sw.c, so
just make them static. This makes sw.h unnecessary, so remove it.

Signed-off-by: Amadeusz Sławiński <amade@asmblr.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtlwifi/rtl8821ae/sw.c |  7 +++----
 drivers/net/wireless/realtek/rtlwifi/rtl8821ae/sw.h | 12 ------------
 2 files changed, 3 insertions(+), 16 deletions(-)
 delete mode 100644 drivers/net/wireless/realtek/rtlwifi/rtl8821ae/sw.h

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/sw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/sw.c
index 3def6a2b3450..d8df816753cb 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/sw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/sw.c
@@ -10,7 +10,6 @@
 #include "dm.h"
 #include "hw.h"
 #include "fw.h"
-#include "sw.h"
 #include "trx.h"
 #include "led.h"
 #include "table.h"
@@ -65,7 +64,7 @@ static void rtl8821ae_init_aspm_vars(struct ieee80211_hw *hw)
 }
 
 /*InitializeVariables8812E*/
-int rtl8821ae_init_sw_vars(struct ieee80211_hw *hw)
+static int rtl8821ae_init_sw_vars(struct ieee80211_hw *hw)
 {
 	int err = 0;
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
@@ -211,7 +210,7 @@ int rtl8821ae_init_sw_vars(struct ieee80211_hw *hw)
 	return 0;
 }
 
-void rtl8821ae_deinit_sw_vars(struct ieee80211_hw *hw)
+static void rtl8821ae_deinit_sw_vars(struct ieee80211_hw *hw)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 
@@ -228,7 +227,7 @@ void rtl8821ae_deinit_sw_vars(struct ieee80211_hw *hw)
 }
 
 /* get bt coexist status */
-bool rtl8821ae_get_btc_status(void)
+static bool rtl8821ae_get_btc_status(void)
 {
 	return true;
 }
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/sw.h b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/sw.h
deleted file mode 100644
index 9d7610f84b20..000000000000
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/sw.h
+++ /dev/null
@@ -1,12 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright(c) 2009-2010  Realtek Corporation.*/
-
-#ifndef __RTL8821AE_SW_H__
-#define __RTL8821AE_SW_H__
-
-int rtl8821ae_init_sw_vars(struct ieee80211_hw *hw);
-void rtl8821ae_deinit_sw_vars(struct ieee80211_hw *hw);
-void rtl8821ae_init_var_map(struct ieee80211_hw *hw);
-bool rtl8821ae_get_btc_status(void);
-
-#endif
-- 
cgit v1.2.3-59-g8ed1b


From 891984bccf64138cb721f87abae838af5636281a Mon Sep 17 00:00:00 2001
From: zhengbin <zhengbin13@huawei.com>
Date: Tue, 24 Dec 2019 22:16:01 +0800
Subject: rtw88: use true,false for bool variable

Fixes coccicheck warning:

drivers/net/wireless/realtek/rtw88/phy.c:1437:1-24: WARNING: Assignment of 0/1 to bool variable

Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: zhengbin <zhengbin13@huawei.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtw88/phy.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw88/phy.c b/drivers/net/wireless/realtek/rtw88/phy.c
index a3e1e9578b65..eea9d888fbf1 100644
--- a/drivers/net/wireless/realtek/rtw88/phy.c
+++ b/drivers/net/wireless/realtek/rtw88/phy.c
@@ -1434,7 +1434,7 @@ static void rtw_load_rfk_table(struct rtw_dev *rtwdev)
 
 	rtw_load_table(rtwdev, chip->rfk_init_tbl);
 
-	dpk_info->is_dpk_pwr_on = 1;
+	dpk_info->is_dpk_pwr_on = true;
 }
 
 void rtw_phy_load_tables(struct rtw_dev *rtwdev)
-- 
cgit v1.2.3-59-g8ed1b


From ab36bb72d00e24ae637a87ce8cde568435322a56 Mon Sep 17 00:00:00 2001
From: zhengbin <zhengbin13@huawei.com>
Date: Tue, 24 Dec 2019 22:16:02 +0800
Subject: cw1200: use true,false for bool variable

Fixes coccicheck warning:

drivers/net/wireless/st/cw1200/txrx.c:718:6-16: WARNING: Assignment of 0/1 to bool variable

Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: zhengbin <zhengbin13@huawei.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/st/cw1200/txrx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/st/cw1200/txrx.c b/drivers/net/wireless/st/cw1200/txrx.c
index 2dfcdb145944..400dd585916b 100644
--- a/drivers/net/wireless/st/cw1200/txrx.c
+++ b/drivers/net/wireless/st/cw1200/txrx.c
@@ -715,7 +715,7 @@ void cw1200_tx(struct ieee80211_hw *dev,
 	};
 	struct ieee80211_sta *sta;
 	struct wsm_tx *wsm;
-	bool tid_update = 0;
+	bool tid_update = false;
 	u8 flags = 0;
 	int ret;
 
-- 
cgit v1.2.3-59-g8ed1b


From b92c017deda819e45a0f054f6df6b53e645d7fe4 Mon Sep 17 00:00:00 2001
From: zhengbin <zhengbin13@huawei.com>
Date: Tue, 24 Dec 2019 22:16:06 +0800
Subject: brcmfmac: use true,false for bool variable

Fixes coccicheck warning:

drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwsignal.c:911:2-24: WARNING: Assignment of 0/1 to bool variable

Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: zhengbin <zhengbin13@huawei.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwsignal.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwsignal.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwsignal.c
index 2bd892df83cc..5e1a11c07551 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwsignal.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwsignal.c
@@ -908,7 +908,7 @@ static u8 brcmf_fws_hdrpush(struct brcmf_fws_info *fws, struct sk_buff *skb)
 	wlh += wlh[1] + 2;
 
 	if (entry->send_tim_signal) {
-		entry->send_tim_signal = 0;
+		entry->send_tim_signal = false;
 		wlh[0] = BRCMF_FWS_TYPE_PENDING_TRAFFIC_BMP;
 		wlh[1] = BRCMF_FWS_TYPE_PENDING_TRAFFIC_BMP_LEN;
 		wlh[2] = entry->mac_handle;
-- 
cgit v1.2.3-59-g8ed1b


From 8c8e60fb86a90a30721bbd797f58f96b3980dcc1 Mon Sep 17 00:00:00 2001
From: Jean-Philippe Brucker <jean-philippe@linaro.org>
Date: Thu, 26 Dec 2019 10:20:33 +0100
Subject: brcmfmac: sdio: Fix OOB interrupt initialization on brcm43362

Commit 262f2b53f679 ("brcmfmac: call brcmf_attach() just before calling
brcmf_bus_started()") changed the initialization order of the brcmfmac
SDIO driver. Unfortunately since brcmf_sdiod_intr_register() is now
called before the sdiodev->bus_if initialization, it reads the wrong
chip ID and fails to initialize the GPIO on brcm43362. Thus the chip
cannot send interrupts and fails to probe:

[   12.517023] brcmfmac: brcmf_sdio_bus_rxctl: resumed on timeout
[   12.531214] ieee80211 phy0: brcmf_bus_started: failed: -110
[   12.536976] ieee80211 phy0: brcmf_attach: dongle is not responding: err=-110
[   12.566467] brcmfmac: brcmf_sdio_firmware_callback: brcmf_attach failed

Initialize the bus interface earlier to ensure that
brcmf_sdiod_intr_register() properly sets up the OOB interrupt.

BugLink: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=908438
Fixes: 262f2b53f679 ("brcmfmac: call brcmf_attach() just before calling brcmf_bus_started()")
Signed-off-by: Jean-Philippe Brucker <jean-philippe@linaro.org>
Reviewed-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
index f9df95bc7fa1..f9047db6a11d 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
@@ -4243,6 +4243,12 @@ static void brcmf_sdio_firmware_callback(struct device *dev, int err,
 	}
 
 	if (err == 0) {
+		/* Assign bus interface call back */
+		sdiod->bus_if->dev = sdiod->dev;
+		sdiod->bus_if->ops = &brcmf_sdio_bus_ops;
+		sdiod->bus_if->chip = bus->ci->chip;
+		sdiod->bus_if->chiprev = bus->ci->chiprev;
+
 		/* Allow full data communication using DPC from now on. */
 		brcmf_sdiod_change_state(bus->sdiodev, BRCMF_SDIOD_DATA);
 
@@ -4259,12 +4265,6 @@ static void brcmf_sdio_firmware_callback(struct device *dev, int err,
 
 	sdio_release_host(sdiod->func1);
 
-	/* Assign bus interface call back */
-	sdiod->bus_if->dev = sdiod->dev;
-	sdiod->bus_if->ops = &brcmf_sdio_bus_ops;
-	sdiod->bus_if->chip = bus->ci->chip;
-	sdiod->bus_if->chiprev = bus->ci->chiprev;
-
 	err = brcmf_alloc(sdiod->dev, sdiod->settings);
 	if (err) {
 		brcmf_err("brcmf_alloc failed\n");
-- 
cgit v1.2.3-59-g8ed1b


From 24332f8068ff6df7f16aefee45d514de1de4de80 Mon Sep 17 00:00:00 2001
From: Rafał Miłecki <rafal@milecki.pl>
Date: Thu, 26 Dec 2019 14:30:49 +0100
Subject: brcmfmac: simplify building interface combinations
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Move similar/duplicated code out of combination specific code blocks.
This simplifies code a bit and allows adding more combinations later.
A list of combinations remains unchanged.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 .../broadcom/brcm80211/brcmfmac/cfg80211.c         | 43 +++++++---------------
 1 file changed, 14 insertions(+), 29 deletions(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
index 6eb3064c3721..3ecb91aa765e 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
@@ -6547,12 +6547,13 @@ static int brcmf_setup_ifmodes(struct wiphy *wiphy, struct brcmf_if *ifp)
 	struct ieee80211_iface_limit *c0_limits = NULL;
 	struct ieee80211_iface_limit *p2p_limits = NULL;
 	struct ieee80211_iface_limit *mbss_limits = NULL;
-	bool mbss, p2p, rsdb;
+	bool mbss, p2p, rsdb, mchan;
 	int i, c, n_combos;
 
 	mbss = brcmf_feat_is_enabled(ifp, BRCMF_FEAT_MBSS);
 	p2p = brcmf_feat_is_enabled(ifp, BRCMF_FEAT_P2P);
 	rsdb = brcmf_feat_is_enabled(ifp, BRCMF_FEAT_RSDB);
+	mchan = brcmf_feat_is_enabled(ifp, BRCMF_FEAT_MCHAN);
 
 	n_combos = 1 + !!(p2p && !rsdb) + !!mbss;
 	combo = kcalloc(n_combos, sizeof(*combo), GFP_KERNEL);
@@ -6562,6 +6563,10 @@ static int brcmf_setup_ifmodes(struct wiphy *wiphy, struct brcmf_if *ifp)
 	wiphy->interface_modes = BIT(NL80211_IFTYPE_STATION) |
 				 BIT(NL80211_IFTYPE_ADHOC) |
 				 BIT(NL80211_IFTYPE_AP);
+	if (p2p)
+		wiphy->interface_modes |= BIT(NL80211_IFTYPE_P2P_CLIENT) |
+					  BIT(NL80211_IFTYPE_P2P_GO) |
+					  BIT(NL80211_IFTYPE_P2P_DEVICE);
 
 	c = 0;
 	i = 0;
@@ -6573,48 +6578,28 @@ static int brcmf_setup_ifmodes(struct wiphy *wiphy, struct brcmf_if *ifp)
 		c0_limits = kcalloc(2, sizeof(*c0_limits), GFP_KERNEL);
 	if (!c0_limits)
 		goto err;
-	if (p2p && rsdb) {
-		combo[c].num_different_channels = 2;
-		wiphy->interface_modes |= BIT(NL80211_IFTYPE_P2P_CLIENT) |
-					  BIT(NL80211_IFTYPE_P2P_GO) |
-					  BIT(NL80211_IFTYPE_P2P_DEVICE);
-		c0_limits[i].max = 2;
-		c0_limits[i++].types = BIT(NL80211_IFTYPE_STATION);
+
+	combo[c].num_different_channels = 1 + (rsdb || (p2p && mchan));
+	c0_limits[i].max = 1 + rsdb;
+	c0_limits[i++].types = BIT(NL80211_IFTYPE_STATION);
+	if (p2p) {
 		c0_limits[i].max = 1;
 		c0_limits[i++].types = BIT(NL80211_IFTYPE_P2P_DEVICE);
-		c0_limits[i].max = 2;
+		c0_limits[i].max = 1 + rsdb;
 		c0_limits[i++].types = BIT(NL80211_IFTYPE_P2P_CLIENT) |
 				       BIT(NL80211_IFTYPE_P2P_GO);
+	}
+	if (p2p && rsdb) {
 		c0_limits[i].max = 2;
 		c0_limits[i++].types = BIT(NL80211_IFTYPE_AP);
 		combo[c].max_interfaces = 5;
 	} else if (p2p) {
-		if (brcmf_feat_is_enabled(ifp, BRCMF_FEAT_MCHAN))
-			combo[c].num_different_channels = 2;
-		else
-			combo[c].num_different_channels = 1;
-		c0_limits[i].max = 1;
-		c0_limits[i++].types = BIT(NL80211_IFTYPE_STATION);
-		wiphy->interface_modes |= BIT(NL80211_IFTYPE_P2P_CLIENT) |
-					  BIT(NL80211_IFTYPE_P2P_GO) |
-					  BIT(NL80211_IFTYPE_P2P_DEVICE);
-		c0_limits[i].max = 1;
-		c0_limits[i++].types = BIT(NL80211_IFTYPE_P2P_DEVICE);
-		c0_limits[i].max = 1;
-		c0_limits[i++].types = BIT(NL80211_IFTYPE_P2P_CLIENT) |
-				       BIT(NL80211_IFTYPE_P2P_GO);
 		combo[c].max_interfaces = i;
 	} else if (rsdb) {
-		combo[c].num_different_channels = 2;
-		c0_limits[i].max = 2;
-		c0_limits[i++].types = BIT(NL80211_IFTYPE_STATION);
 		c0_limits[i].max = 2;
 		c0_limits[i++].types = BIT(NL80211_IFTYPE_AP);
 		combo[c].max_interfaces = 3;
 	} else {
-		combo[c].num_different_channels = 1;
-		c0_limits[i].max = 1;
-		c0_limits[i++].types = BIT(NL80211_IFTYPE_STATION);
 		c0_limits[i].max = 1;
 		c0_limits[i++].types = BIT(NL80211_IFTYPE_AP);
 		combo[c].max_interfaces = i;
-- 
cgit v1.2.3-59-g8ed1b


From 20f2c5fa3af060401c72e444999470a4cab641cf Mon Sep 17 00:00:00 2001
From: Rafał Miłecki <rafal@milecki.pl>
Date: Thu, 26 Dec 2019 14:30:50 +0100
Subject: brcmfmac: add initial support for monitor mode
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Report monitor interface availability using cfg80211 and support it in
the add_virtual_intf() and del_virtual_intf() callbacks. This new
feature is conditional and depends on firmware flagging monitor packets.
Receiving monitor frames is already handled by the brcmf_netif_mon_rx().

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 .../broadcom/brcm80211/brcmfmac/cfg80211.c         | 112 ++++++++++++++++++---
 .../wireless/broadcom/brcm80211/brcmfmac/core.c    |  68 ++++++++++++-
 .../wireless/broadcom/brcm80211/brcmfmac/core.h    |   2 +
 .../wireless/broadcom/brcm80211/brcmfmac/feature.c |   1 +
 .../wireless/broadcom/brcm80211/brcmfmac/feature.h |   2 +
 .../wireless/broadcom/brcm80211/brcmfmac/fwil.h    |   2 +
 6 files changed, 174 insertions(+), 13 deletions(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
index 3ecb91aa765e..a2328d3eee03 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
@@ -11,6 +11,7 @@
 #include <linux/vmalloc.h>
 #include <net/cfg80211.h>
 #include <net/netlink.h>
+#include <uapi/linux/if_arp.h>
 
 #include <brcmu_utils.h>
 #include <defs.h>
@@ -619,6 +620,82 @@ static bool brcmf_is_ibssmode(struct brcmf_cfg80211_vif *vif)
 	return vif->wdev.iftype == NL80211_IFTYPE_ADHOC;
 }
 
+/**
+ * brcmf_mon_add_vif() - create monitor mode virtual interface
+ *
+ * @wiphy: wiphy device of new interface.
+ * @name: name of the new interface.
+ */
+static struct wireless_dev *brcmf_mon_add_vif(struct wiphy *wiphy,
+					      const char *name)
+{
+	struct brcmf_cfg80211_info *cfg = wiphy_to_cfg(wiphy);
+	struct brcmf_cfg80211_vif *vif;
+	struct net_device *ndev;
+	struct brcmf_if *ifp;
+	int err;
+
+	if (cfg->pub->mon_if) {
+		err = -EEXIST;
+		goto err_out;
+	}
+
+	vif = brcmf_alloc_vif(cfg, NL80211_IFTYPE_MONITOR);
+	if (IS_ERR(vif)) {
+		err = PTR_ERR(vif);
+		goto err_out;
+	}
+
+	ndev = alloc_netdev(sizeof(*ifp), name, NET_NAME_UNKNOWN, ether_setup);
+	if (!ndev) {
+		err = -ENOMEM;
+		goto err_free_vif;
+	}
+	ndev->type = ARPHRD_IEEE80211_RADIOTAP;
+	ndev->ieee80211_ptr = &vif->wdev;
+	ndev->needs_free_netdev = true;
+	ndev->priv_destructor = brcmf_cfg80211_free_netdev;
+	SET_NETDEV_DEV(ndev, wiphy_dev(cfg->wiphy));
+
+	ifp = netdev_priv(ndev);
+	ifp->vif = vif;
+	ifp->ndev = ndev;
+	ifp->drvr = cfg->pub;
+
+	vif->ifp = ifp;
+	vif->wdev.netdev = ndev;
+
+	err = brcmf_net_mon_attach(ifp);
+	if (err) {
+		brcmf_err("Failed to attach %s device\n", ndev->name);
+		free_netdev(ndev);
+		goto err_free_vif;
+	}
+
+	cfg->pub->mon_if = ifp;
+
+	return &vif->wdev;
+
+err_free_vif:
+	brcmf_free_vif(vif);
+err_out:
+	return ERR_PTR(err);
+}
+
+static int brcmf_mon_del_vif(struct wiphy *wiphy, struct wireless_dev *wdev)
+{
+	struct brcmf_cfg80211_info *cfg = wiphy_to_cfg(wiphy);
+	struct net_device *ndev = wdev->netdev;
+
+	ndev->netdev_ops->ndo_stop(ndev);
+
+	brcmf_net_detach(ndev, true);
+
+	cfg->pub->mon_if = NULL;
+
+	return 0;
+}
+
 static struct wireless_dev *brcmf_cfg80211_add_iface(struct wiphy *wiphy,
 						     const char *name,
 						     unsigned char name_assign_type,
@@ -641,9 +718,10 @@ static struct wireless_dev *brcmf_cfg80211_add_iface(struct wiphy *wiphy,
 	case NL80211_IFTYPE_STATION:
 	case NL80211_IFTYPE_AP_VLAN:
 	case NL80211_IFTYPE_WDS:
-	case NL80211_IFTYPE_MONITOR:
 	case NL80211_IFTYPE_MESH_POINT:
 		return ERR_PTR(-EOPNOTSUPP);
+	case NL80211_IFTYPE_MONITOR:
+		return brcmf_mon_add_vif(wiphy, name);
 	case NL80211_IFTYPE_AP:
 		wdev = brcmf_ap_add_vif(wiphy, name, params);
 		break;
@@ -826,9 +904,10 @@ int brcmf_cfg80211_del_iface(struct wiphy *wiphy, struct wireless_dev *wdev)
 	case NL80211_IFTYPE_STATION:
 	case NL80211_IFTYPE_AP_VLAN:
 	case NL80211_IFTYPE_WDS:
-	case NL80211_IFTYPE_MONITOR:
 	case NL80211_IFTYPE_MESH_POINT:
 		return -EOPNOTSUPP;
+	case NL80211_IFTYPE_MONITOR:
+		return brcmf_mon_del_vif(wiphy, wdev);
 	case NL80211_IFTYPE_AP:
 		return brcmf_cfg80211_del_ap_iface(wiphy, wdev);
 	case NL80211_IFTYPE_P2P_CLIENT:
@@ -6547,9 +6626,10 @@ static int brcmf_setup_ifmodes(struct wiphy *wiphy, struct brcmf_if *ifp)
 	struct ieee80211_iface_limit *c0_limits = NULL;
 	struct ieee80211_iface_limit *p2p_limits = NULL;
 	struct ieee80211_iface_limit *mbss_limits = NULL;
-	bool mbss, p2p, rsdb, mchan;
-	int i, c, n_combos;
+	bool mon_flag, mbss, p2p, rsdb, mchan;
+	int i, c, n_combos, n_limits;
 
+	mon_flag = brcmf_feat_is_enabled(ifp, BRCMF_FEAT_MONITOR_FLAG);
 	mbss = brcmf_feat_is_enabled(ifp, BRCMF_FEAT_MBSS);
 	p2p = brcmf_feat_is_enabled(ifp, BRCMF_FEAT_P2P);
 	rsdb = brcmf_feat_is_enabled(ifp, BRCMF_FEAT_RSDB);
@@ -6563,6 +6643,8 @@ static int brcmf_setup_ifmodes(struct wiphy *wiphy, struct brcmf_if *ifp)
 	wiphy->interface_modes = BIT(NL80211_IFTYPE_STATION) |
 				 BIT(NL80211_IFTYPE_ADHOC) |
 				 BIT(NL80211_IFTYPE_AP);
+	if (mon_flag)
+		wiphy->interface_modes |= BIT(NL80211_IFTYPE_MONITOR);
 	if (p2p)
 		wiphy->interface_modes |= BIT(NL80211_IFTYPE_P2P_CLIENT) |
 					  BIT(NL80211_IFTYPE_P2P_GO) |
@@ -6570,18 +6652,18 @@ static int brcmf_setup_ifmodes(struct wiphy *wiphy, struct brcmf_if *ifp)
 
 	c = 0;
 	i = 0;
-	if (p2p && rsdb)
-		c0_limits = kcalloc(4, sizeof(*c0_limits), GFP_KERNEL);
-	else if (p2p)
-		c0_limits = kcalloc(3, sizeof(*c0_limits), GFP_KERNEL);
-	else
-		c0_limits = kcalloc(2, sizeof(*c0_limits), GFP_KERNEL);
+	n_limits = 1 + mon_flag + (p2p ? 2 : 0) + (rsdb || !p2p);
+	c0_limits = kcalloc(n_limits, sizeof(*c0_limits), GFP_KERNEL);
 	if (!c0_limits)
 		goto err;
 
 	combo[c].num_different_channels = 1 + (rsdb || (p2p && mchan));
 	c0_limits[i].max = 1 + rsdb;
 	c0_limits[i++].types = BIT(NL80211_IFTYPE_STATION);
+	if (mon_flag) {
+		c0_limits[i].max = 1;
+		c0_limits[i++].types = BIT(NL80211_IFTYPE_MONITOR);
+	}
 	if (p2p) {
 		c0_limits[i].max = 1;
 		c0_limits[i++].types = BIT(NL80211_IFTYPE_P2P_DEVICE);
@@ -6630,14 +6712,20 @@ static int brcmf_setup_ifmodes(struct wiphy *wiphy, struct brcmf_if *ifp)
 	if (mbss) {
 		c++;
 		i = 0;
-		mbss_limits = kcalloc(1, sizeof(*mbss_limits), GFP_KERNEL);
+		n_limits = 1 + mon_flag;
+		mbss_limits = kcalloc(n_limits, sizeof(*mbss_limits),
+				      GFP_KERNEL);
 		if (!mbss_limits)
 			goto err;
 		mbss_limits[i].max = 4;
 		mbss_limits[i++].types = BIT(NL80211_IFTYPE_AP);
+		if (mon_flag) {
+			mbss_limits[i].max = 1;
+			mbss_limits[i++].types = BIT(NL80211_IFTYPE_MONITOR);
+		}
 		combo[c].beacon_int_infra_match = true;
 		combo[c].num_different_channels = 1;
-		combo[c].max_interfaces = 4;
+		combo[c].max_interfaces = 4 + mon_flag;
 		combo[c].n_limits = i;
 		combo[c].limits = mbss_limits;
 	}
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c
index d3ddd97fe768..23627c953a5e 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c
@@ -673,7 +673,7 @@ fail:
 	return -EBADE;
 }
 
-static void brcmf_net_detach(struct net_device *ndev, bool rtnl_locked)
+void brcmf_net_detach(struct net_device *ndev, bool rtnl_locked)
 {
 	if (ndev->reg_state == NETREG_REGISTERED) {
 		if (rtnl_locked)
@@ -686,6 +686,72 @@ static void brcmf_net_detach(struct net_device *ndev, bool rtnl_locked)
 	}
 }
 
+static int brcmf_net_mon_open(struct net_device *ndev)
+{
+	struct brcmf_if *ifp = netdev_priv(ndev);
+	struct brcmf_pub *drvr = ifp->drvr;
+	u32 monitor;
+	int err;
+
+	brcmf_dbg(TRACE, "Enter\n");
+
+	err = brcmf_fil_cmd_int_get(ifp, BRCMF_C_GET_MONITOR, &monitor);
+	if (err) {
+		bphy_err(drvr, "BRCMF_C_GET_MONITOR error (%d)\n", err);
+		return err;
+	} else if (monitor) {
+		bphy_err(drvr, "Monitor mode is already enabled\n");
+		return -EEXIST;
+	}
+
+	monitor = 3;
+	err = brcmf_fil_cmd_int_set(ifp, BRCMF_C_SET_MONITOR, monitor);
+	if (err)
+		bphy_err(drvr, "BRCMF_C_SET_MONITOR error (%d)\n", err);
+
+	return err;
+}
+
+static int brcmf_net_mon_stop(struct net_device *ndev)
+{
+	struct brcmf_if *ifp = netdev_priv(ndev);
+	struct brcmf_pub *drvr = ifp->drvr;
+	u32 monitor;
+	int err;
+
+	brcmf_dbg(TRACE, "Enter\n");
+
+	monitor = 0;
+	err = brcmf_fil_cmd_int_set(ifp, BRCMF_C_SET_MONITOR, monitor);
+	if (err)
+		bphy_err(drvr, "BRCMF_C_SET_MONITOR error (%d)\n", err);
+
+	return err;
+}
+
+static const struct net_device_ops brcmf_netdev_ops_mon = {
+	.ndo_open = brcmf_net_mon_open,
+	.ndo_stop = brcmf_net_mon_stop,
+};
+
+int brcmf_net_mon_attach(struct brcmf_if *ifp)
+{
+	struct brcmf_pub *drvr = ifp->drvr;
+	struct net_device *ndev;
+	int err;
+
+	brcmf_dbg(TRACE, "Enter\n");
+
+	ndev = ifp->ndev;
+	ndev->netdev_ops = &brcmf_netdev_ops_mon;
+
+	err = register_netdevice(ndev);
+	if (err)
+		bphy_err(drvr, "Failed to register %s device\n", ndev->name);
+
+	return err;
+}
+
 void brcmf_net_setcarrier(struct brcmf_if *ifp, bool on)
 {
 	struct net_device *ndev;
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.h
index 6699637d3bf8..33b2ab3b54b0 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.h
@@ -210,6 +210,8 @@ void brcmf_txflowblock_if(struct brcmf_if *ifp,
 void brcmf_txfinalize(struct brcmf_if *ifp, struct sk_buff *txp, bool success);
 void brcmf_netif_rx(struct brcmf_if *ifp, struct sk_buff *skb);
 void brcmf_netif_mon_rx(struct brcmf_if *ifp, struct sk_buff *skb);
+void brcmf_net_detach(struct net_device *ndev, bool rtnl_locked);
+int brcmf_net_mon_attach(struct brcmf_if *ifp);
 void brcmf_net_setcarrier(struct brcmf_if *ifp, bool on);
 int __init brcmf_core_init(void);
 void __exit brcmf_core_exit(void);
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.c
index 1c9c74cc958e..5da0dda0d899 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.c
@@ -38,6 +38,7 @@ static const struct brcmf_feat_fwcap brcmf_fwcap_map[] = {
 	{ BRCMF_FEAT_MCHAN, "mchan" },
 	{ BRCMF_FEAT_P2P, "p2p" },
 	{ BRCMF_FEAT_MONITOR, "monitor" },
+	{ BRCMF_FEAT_MONITOR_FLAG, "rtap" },
 	{ BRCMF_FEAT_MONITOR_FMT_RADIOTAP, "rtap" },
 	{ BRCMF_FEAT_DOT11H, "802.11h" },
 	{ BRCMF_FEAT_SAE, "sae" },
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.h
index 280a1f6412d4..cda3fc1bab7f 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.h
@@ -23,6 +23,7 @@
  * GSCAN: enhanced scan offload feature.
  * FWSUP: Firmware supplicant.
  * MONITOR: firmware can pass monitor packets to host.
+ * MONITOR_FLAG: firmware flags monitor packets.
  * MONITOR_FMT_RADIOTAP: firmware provides monitor packets with radiotap header
  * MONITOR_FMT_HW_RX_HDR: firmware provides monitor packets with hw/ucode header
  * DOT11H: firmware supports 802.11h
@@ -44,6 +45,7 @@
 	BRCMF_FEAT_DEF(GSCAN) \
 	BRCMF_FEAT_DEF(FWSUP) \
 	BRCMF_FEAT_DEF(MONITOR) \
+	BRCMF_FEAT_DEF(MONITOR_FLAG) \
 	BRCMF_FEAT_DEF(MONITOR_FMT_RADIOTAP) \
 	BRCMF_FEAT_DEF(MONITOR_FMT_HW_RX_HDR) \
 	BRCMF_FEAT_DEF(DOT11H) \
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil.h
index 0ff6f5212a94..ae4cf4372908 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil.h
@@ -49,6 +49,8 @@
 #define BRCMF_C_GET_PM				85
 #define BRCMF_C_SET_PM				86
 #define BRCMF_C_GET_REVINFO			98
+#define BRCMF_C_GET_MONITOR			107
+#define BRCMF_C_SET_MONITOR			108
 #define BRCMF_C_GET_CURR_RATESET		114
 #define BRCMF_C_GET_AP				117
 #define BRCMF_C_SET_AP				118
-- 
cgit v1.2.3-59-g8ed1b


From f4f84ff8377d4cedf18317747bc407b2cf657d0f Mon Sep 17 00:00:00 2001
From: Yan-Hsuan Chuang <yhchuang@realtek.com>
Date: Tue, 7 Jan 2020 16:08:07 +0800
Subject: rtw88: fix potential NULL skb access in TX ISR

Sometimes the TX queue may be empty and we could possible
dequeue a NULL pointer, crash the kernel. If the skb is NULL
then there is nothing to do, just leave the ISR.

And the TX queue should not be empty here, so print an error
to see if there is anything wrong for DMA ring.

Fixes: e3037485c68e ("rtw88: new Realtek 802.11ac driver")
Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtw88/pci.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw88/pci.c b/drivers/net/wireless/realtek/rtw88/pci.c
index f7770304ba42..1fbc14c149ec 100644
--- a/drivers/net/wireless/realtek/rtw88/pci.c
+++ b/drivers/net/wireless/realtek/rtw88/pci.c
@@ -835,6 +835,11 @@ static void rtw_pci_tx_isr(struct rtw_dev *rtwdev, struct rtw_pci *rtwpci,
 
 	while (count--) {
 		skb = skb_dequeue(&ring->queue);
+		if (!skb) {
+			rtw_err(rtwdev, "failed to dequeue %d skb TX queue %d, BD=0x%08x, rp %d -> %d\n",
+				count, hw_queue, bd_idx, ring->r.rp, cur_rp);
+			break;
+		}
 		tx_data = rtw_pci_get_tx_data(skb);
 		pci_unmap_single(rtwpci->pdev, tx_data->dma, skb->len,
 				 PCI_DMA_TODEVICE);
-- 
cgit v1.2.3-59-g8ed1b


From 6b6c150b8464e91af26be805f88d5e58c6d626ae Mon Sep 17 00:00:00 2001
From: Yan-Hsuan Chuang <yhchuang@realtek.com>
Date: Tue, 7 Jan 2020 22:27:29 +0800
Subject: rtw88: use shorter delay time to poll PS state

When TX packet arrives, driver should leave deep PS state to make
sure the DMA is working. After requested to leave deep PS state,
driver needs to poll the PS state to check if the mode has been
changed successfully. The driver used to check the state of the
hardware every 20 msecs, which means upon the first failure of
state check, the CPU is delayed 20 msecs for next check. This is
harmful for some time-sensitive applications such as media players.

So, use shorter delay time each check from 20 msecs to 100 usecs.
The state should be changed in several tries. But we still need
to reserve ~15 msecs in total in case of the state just took too
long to be changed successfully. If the states of driver and the
hardware is not synchronized, the power state could be locked
forever, which mean we could never enter/leave the PS state.

Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com>
Reviewed-by: Chris Chiu <chiu@endlessm.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtw88/ps.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw88/ps.c b/drivers/net/wireless/realtek/rtw88/ps.c
index 913e6f47130f..7a189a9926fe 100644
--- a/drivers/net/wireless/realtek/rtw88/ps.c
+++ b/drivers/net/wireless/realtek/rtw88/ps.c
@@ -91,11 +91,11 @@ void rtw_power_mode_change(struct rtw_dev *rtwdev, bool enter)
 			return;
 
 		/* check confirm power mode has left power save state */
-		for (polling_cnt = 0; polling_cnt < 3; polling_cnt++) {
+		for (polling_cnt = 0; polling_cnt < 50; polling_cnt++) {
 			polling = rtw_read8(rtwdev, rtwdev->hci.cpwm_addr);
 			if ((polling ^ confirm) & BIT_RPWM_TOGGLE)
 				return;
-			mdelay(20);
+			udelay(100);
 		}
 
 		/* in case of fw/hw missed the request, retry */
-- 
cgit v1.2.3-59-g8ed1b


From 627b0d094240c38393b2f2d40626c33a8fff6103 Mon Sep 17 00:00:00 2001
From: yuehaibing <yuehaibing@huawei.com>
Date: Wed, 8 Jan 2020 21:57:48 +0800
Subject: brcmfmac: Remove always false 'idx < 0' statement

idx is declared as u32, it will never less than 0.

Signed-off-by: yuehaibing <yuehaibing@huawei.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/msgbuf.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/msgbuf.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/msgbuf.c
index e3dd8623be4e..8bb4f1fa790e 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/msgbuf.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/msgbuf.c
@@ -365,7 +365,7 @@ brcmf_msgbuf_get_pktid(struct device *dev, struct brcmf_msgbuf_pktids *pktids,
 	struct brcmf_msgbuf_pktid *pktid;
 	struct sk_buff *skb;
 
-	if (idx < 0 || idx >= pktids->array_size) {
+	if (idx >= pktids->array_size) {
 		brcmf_err("Invalid packet id %d (max %d)\n", idx,
 			  pktids->array_size);
 		return NULL;
-- 
cgit v1.2.3-59-g8ed1b


From e92b07a20b5f039ccd8776dc7c6323a3e1d726d9 Mon Sep 17 00:00:00 2001
From: Sergey Matyukevich <sergey.matyukevich.os@quantenna.com>
Date: Thu, 9 Jan 2020 16:17:51 +0300
Subject: qtnfmac: cleanup slave_radar access function

Currently this parameter is global, it is not specific to mac.
So this function does not need any input parameters.

Signed-off-by: Sergey Matyukevich <sergey.matyukevich.os@quantenna.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/quantenna/qtnfmac/cfg80211.c |  2 +-
 drivers/net/wireless/quantenna/qtnfmac/core.c     | 10 +++++-----
 drivers/net/wireless/quantenna/qtnfmac/core.h     |  2 +-
 3 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/quantenna/qtnfmac/cfg80211.c b/drivers/net/wireless/quantenna/qtnfmac/cfg80211.c
index 59d089e092f9..25a9b512b8ce 100644
--- a/drivers/net/wireless/quantenna/qtnfmac/cfg80211.c
+++ b/drivers/net/wireless/quantenna/qtnfmac/cfg80211.c
@@ -1056,7 +1056,7 @@ static void qtnf_cfg80211_reg_notifier(struct wiphy *wiphy,
 	pr_debug("MAC%u: initiator=%d alpha=%c%c\n", mac->macid, req->initiator,
 		 req->alpha2[0], req->alpha2[1]);
 
-	ret = qtnf_cmd_reg_notify(mac, req, qtnf_mac_slave_radar_get(wiphy));
+	ret = qtnf_cmd_reg_notify(mac, req, qtnf_slave_radar_get());
 	if (ret) {
 		pr_err("MAC%u: failed to update region to %c%c: %d\n",
 		       mac->macid, req->alpha2[0], req->alpha2[1], ret);
diff --git a/drivers/net/wireless/quantenna/qtnfmac/core.c b/drivers/net/wireless/quantenna/qtnfmac/core.c
index 648dfc38bd70..9bd952a62373 100644
--- a/drivers/net/wireless/quantenna/qtnfmac/core.c
+++ b/drivers/net/wireless/quantenna/qtnfmac/core.c
@@ -23,6 +23,11 @@ MODULE_PARM_DESC(slave_radar, "set 0 to disable radar detection in slave mode");
 
 static struct dentry *qtnf_debugfs_dir;
 
+bool qtnf_slave_radar_get(void)
+{
+	return slave_radar;
+}
+
 struct qtnf_wmac *qtnf_core_get_mac(const struct qtnf_bus *bus, u8 macid)
 {
 	struct qtnf_wmac *mac = NULL;
@@ -456,11 +461,6 @@ static struct qtnf_wmac *qtnf_core_mac_alloc(struct qtnf_bus *bus,
 	return mac;
 }
 
-bool qtnf_mac_slave_radar_get(struct wiphy *wiphy)
-{
-	return slave_radar;
-}
-
 static const struct ethtool_ops qtnf_ethtool_ops = {
 	.get_drvinfo = cfg80211_get_drvinfo,
 };
diff --git a/drivers/net/wireless/quantenna/qtnfmac/core.h b/drivers/net/wireless/quantenna/qtnfmac/core.h
index 116ec16aa15b..949219c6cb8c 100644
--- a/drivers/net/wireless/quantenna/qtnfmac/core.h
+++ b/drivers/net/wireless/quantenna/qtnfmac/core.h
@@ -133,7 +133,7 @@ struct qtnf_vif *qtnf_mac_get_free_vif(struct qtnf_wmac *mac);
 struct qtnf_vif *qtnf_mac_get_base_vif(struct qtnf_wmac *mac);
 void qtnf_mac_iface_comb_free(struct qtnf_wmac *mac);
 void qtnf_mac_ext_caps_free(struct qtnf_wmac *mac);
-bool qtnf_mac_slave_radar_get(struct wiphy *wiphy);
+bool qtnf_slave_radar_get(void);
 struct wiphy *qtnf_wiphy_allocate(struct qtnf_bus *bus);
 int qtnf_core_net_attach(struct qtnf_wmac *mac, struct qtnf_vif *priv,
 			 const char *name, unsigned char name_assign_type);
-- 
cgit v1.2.3-59-g8ed1b


From 155b424c86f8169b30d30beb46ad216c5e61749a Mon Sep 17 00:00:00 2001
From: Sergey Matyukevich <sergey.matyukevich.os@quantenna.com>
Date: Thu, 9 Jan 2020 16:17:52 +0300
Subject: qtnfmac: add module param to configure DFS offload

Firmware may support DFS offload. However the final decision on whether
to use it or not should be up to the user. So even if firmware supports
DFS offload, it should be enabled only if user explicitly requests it.
For this purpose introduce kernel param dfs_offload which is disabled
by default.

Signed-off-by: Sergey Matyukevich <sergey.matyukevich.os@quantenna.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/quantenna/qtnfmac/cfg80211.c | 9 ++++++---
 drivers/net/wireless/quantenna/qtnfmac/commands.c | 3 ++-
 drivers/net/wireless/quantenna/qtnfmac/commands.h | 2 +-
 drivers/net/wireless/quantenna/qtnfmac/core.c     | 9 +++++++++
 drivers/net/wireless/quantenna/qtnfmac/core.h     | 1 +
 drivers/net/wireless/quantenna/qtnfmac/qlink.h    | 5 +++--
 6 files changed, 22 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/quantenna/qtnfmac/cfg80211.c b/drivers/net/wireless/quantenna/qtnfmac/cfg80211.c
index 25a9b512b8ce..8849faa5bc10 100644
--- a/drivers/net/wireless/quantenna/qtnfmac/cfg80211.c
+++ b/drivers/net/wireless/quantenna/qtnfmac/cfg80211.c
@@ -1056,7 +1056,8 @@ static void qtnf_cfg80211_reg_notifier(struct wiphy *wiphy,
 	pr_debug("MAC%u: initiator=%d alpha=%c%c\n", mac->macid, req->initiator,
 		 req->alpha2[0], req->alpha2[1]);
 
-	ret = qtnf_cmd_reg_notify(mac, req, qtnf_slave_radar_get());
+	ret = qtnf_cmd_reg_notify(mac, req, qtnf_slave_radar_get(),
+				  qtnf_dfs_offload_get());
 	if (ret) {
 		pr_err("MAC%u: failed to update region to %c%c: %d\n",
 		       mac->macid, req->alpha2[0], req->alpha2[1], ret);
@@ -1078,7 +1079,8 @@ struct wiphy *qtnf_wiphy_allocate(struct qtnf_bus *bus)
 {
 	struct wiphy *wiphy;
 
-	if (bus->hw_info.hw_capab & QLINK_HW_CAPAB_DFS_OFFLOAD)
+	if (qtnf_dfs_offload_get() &&
+	    (bus->hw_info.hw_capab & QLINK_HW_CAPAB_DFS_OFFLOAD))
 		qtn_cfg80211_ops.start_radar_detection = NULL;
 
 	if (!(bus->hw_info.hw_capab & QLINK_HW_CAPAB_PWR_MGMT))
@@ -1163,7 +1165,8 @@ int qtnf_wiphy_register(struct qtnf_hw_info *hw_info, struct qtnf_wmac *mac)
 			WIPHY_FLAG_NETNS_OK;
 	wiphy->flags &= ~WIPHY_FLAG_PS_ON_BY_DEFAULT;
 
-	if (hw_info->hw_capab & QLINK_HW_CAPAB_DFS_OFFLOAD)
+	if (qtnf_dfs_offload_get() &&
+	    (hw_info->hw_capab & QLINK_HW_CAPAB_DFS_OFFLOAD))
 		wiphy_ext_feature_set(wiphy, NL80211_EXT_FEATURE_DFS_OFFLOAD);
 
 	if (hw_info->hw_capab & QLINK_HW_CAPAB_SCAN_DWELL)
diff --git a/drivers/net/wireless/quantenna/qtnfmac/commands.c b/drivers/net/wireless/quantenna/qtnfmac/commands.c
index 548f6ff6d0f2..fbeff01f902a 100644
--- a/drivers/net/wireless/quantenna/qtnfmac/commands.c
+++ b/drivers/net/wireless/quantenna/qtnfmac/commands.c
@@ -2454,7 +2454,7 @@ out:
 }
 
 int qtnf_cmd_reg_notify(struct qtnf_wmac *mac, struct regulatory_request *req,
-			bool slave_radar)
+			bool slave_radar, bool dfs_offload)
 {
 	struct wiphy *wiphy = priv_to_wiphy(mac);
 	struct qtnf_bus *bus = mac->bus;
@@ -2517,6 +2517,7 @@ int qtnf_cmd_reg_notify(struct qtnf_wmac *mac, struct regulatory_request *req,
 	}
 
 	cmd->slave_radar = slave_radar;
+	cmd->dfs_offload = dfs_offload;
 	cmd->num_channels = 0;
 
 	for (band = 0; band < NUM_NL80211_BANDS; band++) {
diff --git a/drivers/net/wireless/quantenna/qtnfmac/commands.h b/drivers/net/wireless/quantenna/qtnfmac/commands.h
index 761755bf9ede..ab273257b078 100644
--- a/drivers/net/wireless/quantenna/qtnfmac/commands.h
+++ b/drivers/net/wireless/quantenna/qtnfmac/commands.h
@@ -58,7 +58,7 @@ int qtnf_cmd_send_disconnect(struct qtnf_vif *vif,
 int qtnf_cmd_send_updown_intf(struct qtnf_vif *vif,
 			      bool up);
 int qtnf_cmd_reg_notify(struct qtnf_wmac *mac, struct regulatory_request *req,
-			bool slave_radar);
+			bool slave_radar, bool dfs_offload);
 int qtnf_cmd_get_chan_stats(struct qtnf_wmac *mac, u16 channel,
 			    struct qtnf_chan_stats *stats);
 int qtnf_cmd_send_chan_switch(struct qtnf_vif *vif,
diff --git a/drivers/net/wireless/quantenna/qtnfmac/core.c b/drivers/net/wireless/quantenna/qtnfmac/core.c
index 9bd952a62373..b6e53415bea1 100644
--- a/drivers/net/wireless/quantenna/qtnfmac/core.c
+++ b/drivers/net/wireless/quantenna/qtnfmac/core.c
@@ -21,6 +21,10 @@ static bool slave_radar = true;
 module_param(slave_radar, bool, 0644);
 MODULE_PARM_DESC(slave_radar, "set 0 to disable radar detection in slave mode");
 
+static bool dfs_offload;
+module_param(dfs_offload, bool, 0644);
+MODULE_PARM_DESC(dfs_offload, "set 1 to enable DFS offload to firmware");
+
 static struct dentry *qtnf_debugfs_dir;
 
 bool qtnf_slave_radar_get(void)
@@ -28,6 +32,11 @@ bool qtnf_slave_radar_get(void)
 	return slave_radar;
 }
 
+bool qtnf_dfs_offload_get(void)
+{
+	return dfs_offload;
+}
+
 struct qtnf_wmac *qtnf_core_get_mac(const struct qtnf_bus *bus, u8 macid)
 {
 	struct qtnf_wmac *mac = NULL;
diff --git a/drivers/net/wireless/quantenna/qtnfmac/core.h b/drivers/net/wireless/quantenna/qtnfmac/core.h
index 949219c6cb8c..d715e1cd0006 100644
--- a/drivers/net/wireless/quantenna/qtnfmac/core.h
+++ b/drivers/net/wireless/quantenna/qtnfmac/core.h
@@ -134,6 +134,7 @@ struct qtnf_vif *qtnf_mac_get_base_vif(struct qtnf_wmac *mac);
 void qtnf_mac_iface_comb_free(struct qtnf_wmac *mac);
 void qtnf_mac_ext_caps_free(struct qtnf_wmac *mac);
 bool qtnf_slave_radar_get(void);
+bool qtnf_dfs_offload_get(void);
 struct wiphy *qtnf_wiphy_allocate(struct qtnf_bus *bus);
 int qtnf_core_net_attach(struct qtnf_wmac *mac, struct qtnf_vif *priv,
 			 const char *name, unsigned char name_assign_type);
diff --git a/drivers/net/wireless/quantenna/qtnfmac/qlink.h b/drivers/net/wireless/quantenna/qtnfmac/qlink.h
index 75527f1bb306..799c622a160d 100644
--- a/drivers/net/wireless/quantenna/qtnfmac/qlink.h
+++ b/drivers/net/wireless/quantenna/qtnfmac/qlink.h
@@ -596,8 +596,9 @@ enum qlink_user_reg_hint_type {
  *	of &enum qlink_user_reg_hint_type.
  * @num_channels: number of &struct qlink_tlv_channel in a variable portion of a
  *	payload.
- * @slave_radar: whether slave device should enable radar detection.
  * @dfs_region: one of &enum qlink_dfs_regions.
+ * @slave_radar: whether slave device should enable radar detection.
+ * @dfs_offload: enable or disable DFS offload to firmware.
  * @info: variable portion of regulatory notifier callback.
  */
 struct qlink_cmd_reg_notify {
@@ -608,7 +609,7 @@ struct qlink_cmd_reg_notify {
 	u8 num_channels;
 	u8 dfs_region;
 	u8 slave_radar;
-	u8 rsvd[1];
+	u8 dfs_offload;
 	u8 info[0];
 } __packed;
 
-- 
cgit v1.2.3-59-g8ed1b


From e14291752e568bf936907be7f11ae3cc39d89fd2 Mon Sep 17 00:00:00 2001
From: Sergey Matyukevich <sergey.matyukevich.os@quantenna.com>
Date: Thu, 9 Jan 2020 16:17:53 +0300
Subject: qtnfmac: control qtnfmac wireless interfaces bridging

Bridging qtnfmac interfaces is possible only if the following two
conditions are fulfilled:
- firmware advertises proper support with QLINK_HW_CAPAB_HW_BRIDGE
- kernel is built with CONFIG_NET_SWITCHDEV support

Otherwise adding qtnfmac wireless interfaces into the same bridge
should not be allowed since packets flooded by kernel may break
internal forwarding rules between interfaces.

This patch disables adding qtnfmac wireless interfaces into the
same bridge if no support is provided either by card or by kernel.

Signed-off-by: Sergey Matyukevich <sergey.matyukevich.os@quantenna.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/quantenna/qtnfmac/core.c | 52 +++++++++++++++++----------
 1 file changed, 34 insertions(+), 18 deletions(-)

diff --git a/drivers/net/wireless/quantenna/qtnfmac/core.c b/drivers/net/wireless/quantenna/qtnfmac/core.c
index b6e53415bea1..4320180f8c07 100644
--- a/drivers/net/wireless/quantenna/qtnfmac/core.c
+++ b/drivers/net/wireless/quantenna/qtnfmac/core.c
@@ -225,9 +225,6 @@ static int qtnf_netdev_port_parent_id(struct net_device *ndev,
 	const struct qtnf_vif *vif = qtnf_netdev_get_priv(ndev);
 	const struct qtnf_bus *bus = vif->mac->bus;
 
-	if (!(bus->hw_info.hw_capab & QLINK_HW_CAPAB_HW_BRIDGE))
-		return -EOPNOTSUPP;
-
 	ppid->id_len = sizeof(bus->hw_id);
 	memcpy(&ppid->id, bus->hw_id, ppid->id_len);
 
@@ -665,12 +662,24 @@ bool qtnf_netdev_is_qtn(const struct net_device *ndev)
 	return ndev->netdev_ops == &qtnf_netdev_ops;
 }
 
+static int qtnf_check_br_ports(struct net_device *dev, void *data)
+{
+	struct net_device *ndev = data;
+
+	if (dev != ndev && netdev_port_same_parent_id(dev, ndev))
+		return -ENOTSUPP;
+
+	return 0;
+}
+
 static int qtnf_core_netdevice_event(struct notifier_block *nb,
 				     unsigned long event, void *ptr)
 {
 	struct net_device *ndev = netdev_notifier_info_to_dev(ptr);
 	const struct netdev_notifier_changeupper_info *info;
+	struct net_device *brdev;
 	struct qtnf_vif *vif;
+	struct qtnf_bus *bus;
 	int br_domain;
 	int ret = 0;
 
@@ -681,25 +690,34 @@ static int qtnf_core_netdevice_event(struct notifier_block *nb,
 		return NOTIFY_OK;
 
 	vif = qtnf_netdev_get_priv(ndev);
+	bus = vif->mac->bus;
 
 	switch (event) {
 	case NETDEV_CHANGEUPPER:
 		info = ptr;
+		brdev = info->upper_dev;
 
-		if (!netif_is_bridge_master(info->upper_dev))
+		if (!netif_is_bridge_master(brdev))
 			break;
 
 		pr_debug("[VIF%u.%u] change bridge: %s %s\n",
-			 vif->mac->macid, vif->vifid,
-			 netdev_name(info->upper_dev),
+			 vif->mac->macid, vif->vifid, netdev_name(brdev),
 			 info->linking ? "add" : "del");
 
-		if (info->linking)
-			br_domain = info->upper_dev->ifindex;
-		else
-			br_domain = ndev->ifindex;
+		if (IS_ENABLED(CONFIG_NET_SWITCHDEV) &&
+		    (bus->hw_info.hw_capab & QLINK_HW_CAPAB_HW_BRIDGE)) {
+			if (info->linking)
+				br_domain = brdev->ifindex;
+			else
+				br_domain = ndev->ifindex;
+
+			ret = qtnf_cmd_netdev_changeupper(vif, br_domain);
+		} else {
+			ret = netdev_walk_all_lower_dev(brdev,
+							qtnf_check_br_ports,
+							ndev);
+		}
 
-		ret = qtnf_cmd_netdev_changeupper(vif, br_domain);
 		break;
 	default:
 		break;
@@ -772,13 +790,11 @@ int qtnf_core_attach(struct qtnf_bus *bus)
 		}
 	}
 
-	if (bus->hw_info.hw_capab & QLINK_HW_CAPAB_HW_BRIDGE) {
-		bus->netdev_nb.notifier_call = qtnf_core_netdevice_event;
-		ret = register_netdevice_notifier(&bus->netdev_nb);
-		if (ret) {
-			pr_err("failed to register netdev notifier: %d\n", ret);
-			goto error;
-		}
+	bus->netdev_nb.notifier_call = qtnf_core_netdevice_event;
+	ret = register_netdevice_notifier(&bus->netdev_nb);
+	if (ret) {
+		pr_err("failed to register netdev notifier: %d\n", ret);
+		goto error;
 	}
 
 	bus->fw_state = QTNF_FW_STATE_RUNNING;
-- 
cgit v1.2.3-59-g8ed1b


From b73f0aac731ce87a656d9911cb9e2e79eb721d1a Mon Sep 17 00:00:00 2001
From: Sergey Matyukevich <sergey.matyukevich.os@quantenna.com>
Date: Thu, 9 Jan 2020 16:17:54 +0300
Subject: qtnfmac: add support for STA HE rates

Add HE rates into STA info. Report HE Rx/Tx MCS if STA supports them.

Signed-off-by: Sergey Matyukevich <sergey.matyukevich.os@quantenna.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/quantenna/qtnfmac/commands.c | 2 ++
 drivers/net/wireless/quantenna/qtnfmac/qlink.h    | 1 +
 2 files changed, 3 insertions(+)

diff --git a/drivers/net/wireless/quantenna/qtnfmac/commands.c b/drivers/net/wireless/quantenna/qtnfmac/commands.c
index fbeff01f902a..fa932f771856 100644
--- a/drivers/net/wireless/quantenna/qtnfmac/commands.c
+++ b/drivers/net/wireless/quantenna/qtnfmac/commands.c
@@ -510,6 +510,8 @@ qtnf_sta_info_parse_rate(struct rate_info *rate_dst,
 		rate_dst->flags |= RATE_INFO_FLAGS_MCS;
 	else if (rate_src->flags & QLINK_STA_INFO_RATE_FLAG_VHT_MCS)
 		rate_dst->flags |= RATE_INFO_FLAGS_VHT_MCS;
+	else if (rate_src->flags & QLINK_STA_INFO_RATE_FLAG_HE_MCS)
+		rate_dst->flags |= RATE_INFO_FLAGS_HE_MCS;
 
 	if (rate_src->flags & QLINK_STA_INFO_RATE_FLAG_SHORT_GI)
 		rate_dst->flags |= RATE_INFO_FLAGS_SHORT_GI;
diff --git a/drivers/net/wireless/quantenna/qtnfmac/qlink.h b/drivers/net/wireless/quantenna/qtnfmac/qlink.h
index 799c622a160d..19e432b28de5 100644
--- a/drivers/net/wireless/quantenna/qtnfmac/qlink.h
+++ b/drivers/net/wireless/quantenna/qtnfmac/qlink.h
@@ -949,6 +949,7 @@ enum qlink_sta_info_rate_flags {
 	QLINK_STA_INFO_RATE_FLAG_VHT_MCS	= BIT(1),
 	QLINK_STA_INFO_RATE_FLAG_SHORT_GI	= BIT(2),
 	QLINK_STA_INFO_RATE_FLAG_60G		= BIT(3),
+	QLINK_STA_INFO_RATE_FLAG_HE_MCS		= BIT(4),
 };
 
 /**
-- 
cgit v1.2.3-59-g8ed1b


From ed7791d9d0abc7e24223ab1663931c8e5351fea8 Mon Sep 17 00:00:00 2001
From: Mikhail Karpenko <mkarpenko@quantenna.com>
Date: Thu, 9 Jan 2020 16:17:55 +0300
Subject: qtnfmac: add support for TWT responder and spatial reuse

Add support for 11ax features: TWT responder and spatial reuse.
Add separate structure for spatial reuse parameters and pass this
structure to firmware along with other parameters in start_ap
command. Pass TWT responder value to firmware. Bump qlink
protocol version.

Signed-off-by: Mikhail Karpenko <mkarpenko@quantenna.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/quantenna/qtnfmac/commands.c |  8 ++++
 drivers/net/wireless/quantenna/qtnfmac/qlink.h    | 46 ++++++++++++++++++++++-
 2 files changed, 53 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/quantenna/qtnfmac/commands.c b/drivers/net/wireless/quantenna/qtnfmac/commands.c
index fa932f771856..d0d7ec8794c4 100644
--- a/drivers/net/wireless/quantenna/qtnfmac/commands.c
+++ b/drivers/net/wireless/quantenna/qtnfmac/commands.c
@@ -257,6 +257,14 @@ int qtnf_cmd_send_start_ap(struct qtnf_vif *vif,
 	cmd->pbss = s->pbss;
 	cmd->ht_required = s->ht_required;
 	cmd->vht_required = s->vht_required;
+	cmd->twt_responder = s->twt_responder;
+	if (s->he_obss_pd.enable) {
+		cmd->sr_params.sr_control |= QLINK_SR_SRG_INFORMATION_PRESENT;
+		cmd->sr_params.srg_obss_pd_min_offset =
+			s->he_obss_pd.min_offset;
+		cmd->sr_params.srg_obss_pd_max_offset =
+			s->he_obss_pd.max_offset;
+	}
 
 	aen = &cmd->aen;
 	aen->auth_type = s->auth_type;
diff --git a/drivers/net/wireless/quantenna/qtnfmac/qlink.h b/drivers/net/wireless/quantenna/qtnfmac/qlink.h
index 19e432b28de5..b2edb03819d1 100644
--- a/drivers/net/wireless/quantenna/qtnfmac/qlink.h
+++ b/drivers/net/wireless/quantenna/qtnfmac/qlink.h
@@ -6,7 +6,7 @@
 
 #include <linux/ieee80211.h>
 
-#define QLINK_PROTO_VER		15
+#define QLINK_PROTO_VER		16
 
 #define QLINK_MACID_RSVD		0xFF
 #define QLINK_VIFID_RSVD		0xFF
@@ -196,6 +196,45 @@ struct qlink_sta_info_state {
 	__le32 value;
 } __packed;
 
+/**
+ * enum qlink_sr_ctrl_flags - control flags for spatial reuse parameter set
+ *
+ * @QLINK_SR_PSR_DISALLOWED: indicates whether or not PSR-based spatial reuse
+ * transmissions are allowed for STAs associated with the AP
+ * @QLINK_SR_NON_SRG_OBSS_PD_SR_DISALLOWED: indicates whether or not
+ * Non-SRG OBSS PD spatial reuse transmissions are allowed for STAs associated
+ * with the AP
+ * @NON_SRG_OFFSET_PRESENT: indicates whether or not Non-SRG OBSS PD Max offset
+ * field is valid in the element
+ * @QLINK_SR_SRG_INFORMATION_PRESENT: indicates whether or not SRG OBSS PD
+ * Min/Max offset fields ore valid in the element
+ */
+enum qlink_sr_ctrl_flags {
+	QLINK_SR_PSR_DISALLOWED                = BIT(0),
+	QLINK_SR_NON_SRG_OBSS_PD_SR_DISALLOWED = BIT(1),
+	QLINK_SR_NON_SRG_OFFSET_PRESENT        = BIT(2),
+	QLINK_SR_SRG_INFORMATION_PRESENT       = BIT(3),
+};
+
+/**
+ * struct qlink_sr_params - spatial reuse parameters
+ *
+ * @sr_control: spatial reuse control field; flags contained in this field are
+ * defined in @qlink_sr_ctrl_flags
+ * @non_srg_obss_pd_max: added to -82 dBm to generate the value of the
+ * Non-SRG OBSS PD Max parameter
+ * @srg_obss_pd_min_offset: added to -82 dBm to generate the value of the
+ * SRG OBSS PD Min parameter
+ * @srg_obss_pd_max_offset: added to -82 dBm to generate the value of the
+ * SRG PBSS PD Max parameter
+ */
+struct qlink_sr_params {
+	u8 sr_control;
+	u8 non_srg_obss_pd_max;
+	u8 srg_obss_pd_min_offset;
+	u8 srg_obss_pd_max_offset;
+} __packed;
+
 /* QLINK Command messages related definitions
  */
 
@@ -651,6 +690,8 @@ enum qlink_hidden_ssid {
  * @ht_required: stations must support HT
  * @vht_required: stations must support VHT
  * @aen: encryption info
+ * @sr_params: spatial reuse parameters
+ * @twt_responder: enable Target Wake Time
  * @info: variable configurations
  */
 struct qlink_cmd_start_ap {
@@ -666,6 +707,9 @@ struct qlink_cmd_start_ap {
 	u8 ht_required;
 	u8 vht_required;
 	struct qlink_auth_encr aen;
+	struct qlink_sr_params sr_params;
+	u8 twt_responder;
+	u8 rsvd[3];
 	u8 info[0];
 } __packed;
 
-- 
cgit v1.2.3-59-g8ed1b


From c3a913357c3e3d2225a83508f06ece01f5bada20 Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.king@canonical.com>
Date: Tue, 14 Jan 2020 16:56:01 +0000
Subject: rtlwifi: rtl8188ee: remove redundant assignment to variable cond

Variable cond is being assigned with a value that is never
read, it is assigned a new value later on. The assignment is
redundant and can be removed.

Addresses-Coverity: ("Unused value")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Acked-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtlwifi/rtl8188ee/phy.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/phy.c b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/phy.c
index 5ca900f97d66..d13983ec09ad 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/phy.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/phy.c
@@ -264,7 +264,7 @@ static bool _rtl88e_check_condition(struct ieee80211_hw *hw,
 	u32 _board = rtlefuse->board_type; /*need efuse define*/
 	u32 _interface = rtlhal->interface;
 	u32 _platform = 0x08;/*SupportPlatform */
-	u32 cond = condition;
+	u32 cond;
 
 	if (condition == 0xCDCDCDCD)
 		return true;
-- 
cgit v1.2.3-59-g8ed1b


From cc071a6f26aae3321cf193dc2e8c35090709b8ab Mon Sep 17 00:00:00 2001
From: YueHaibing <yuehaibing@huawei.com>
Date: Tue, 21 Jan 2020 10:09:58 +0800
Subject: rtlwifi: rtl8821ae: remove unused variables

drivers/net/wireless/realtek/rtlwifi/rtl8821ae/dm.c:142:17:
 warning: cckswing_table_ch1ch13 defined but not used [-Wunused-const-variable=]
drivers/net/wireless/realtek/rtlwifi/rtl8821ae/dm.c:178:17:
 warning: cckswing_table_ch14 defined but not used [-Wunused-const-variable=]
drivers/net/wireless/realtek/rtlwifi/rtl8821ae/dm.c:96:18:
 warning: ofdmswing_table defined but not used [-Wunused-const-variable=]

These variable is never used, so remove them.

Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 .../net/wireless/realtek/rtlwifi/rtl8821ae/dm.c    | 118 ---------------------
 1 file changed, 118 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/dm.c b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/dm.c
index b54230433a6b..f57e8794f0ec 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/dm.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/dm.c
@@ -93,124 +93,6 @@ static const u32 rtl8821ae_txscaling_table[TXSCALE_TABLE_SIZE] = {
 	0x3FE  /* 36, +6.0dB */
 };
 
-static const u32 ofdmswing_table[] = {
-	0x0b40002d, /* 0, -15.0dB */
-	0x0c000030, /* 1, -14.5dB */
-	0x0cc00033, /* 2, -14.0dB */
-	0x0d800036, /* 3, -13.5dB */
-	0x0e400039, /* 4, -13.0dB */
-	0x0f00003c, /* 5, -12.5dB */
-	0x10000040, /* 6, -12.0dB */
-	0x11000044, /* 7, -11.5dB */
-	0x12000048, /* 8, -11.0dB */
-	0x1300004c, /* 9, -10.5dB */
-	0x14400051, /* 10, -10.0dB */
-	0x15800056, /* 11, -9.5dB */
-	0x16c0005b, /* 12, -9.0dB */
-	0x18000060, /* 13, -8.5dB */
-	0x19800066, /* 14, -8.0dB */
-	0x1b00006c, /* 15, -7.5dB */
-	0x1c800072, /* 16, -7.0dB */
-	0x1e400079, /* 17, -6.5dB */
-	0x20000080, /* 18, -6.0dB */
-	0x22000088, /* 19, -5.5dB */
-	0x24000090, /* 20, -5.0dB */
-	0x26000098, /* 21, -4.5dB */
-	0x288000a2, /* 22, -4.0dB */
-	0x2ac000ab, /* 23, -3.5dB */
-	0x2d4000b5, /* 24, -3.0dB */
-	0x300000c0, /* 25, -2.5dB */
-	0x32c000cb, /* 26, -2.0dB */
-	0x35c000d7, /* 27, -1.5dB */
-	0x390000e4, /* 28, -1.0dB */
-	0x3c8000f2, /* 29, -0.5dB */
-	0x40000100, /* 30, +0dB */
-	0x43c0010f, /* 31, +0.5dB */
-	0x47c0011f, /* 32, +1.0dB */
-	0x4c000130, /* 33, +1.5dB */
-	0x50800142, /* 34, +2.0dB */
-	0x55400155, /* 35, +2.5dB */
-	0x5a400169, /* 36, +3.0dB */
-	0x5fc0017f, /* 37, +3.5dB */
-	0x65400195, /* 38, +4.0dB */
-	0x6b8001ae, /* 39, +4.5dB */
-	0x71c001c7, /* 40, +5.0dB */
-	0x788001e2, /* 41, +5.5dB */
-	0x7f8001fe  /* 42, +6.0dB */
-};
-
-static const u8 cckswing_table_ch1ch13[CCK_TABLE_SIZE][8] = {
-	{0x09, 0x08, 0x07, 0x06, 0x04, 0x03, 0x01, 0x01}, /* 0, -16.0dB */
-	{0x09, 0x09, 0x08, 0x06, 0x05, 0x03, 0x01, 0x01}, /* 1, -15.5dB */
-	{0x0a, 0x09, 0x08, 0x07, 0x05, 0x03, 0x02, 0x01}, /* 2, -15.0dB */
-	{0x0a, 0x0a, 0x09, 0x07, 0x05, 0x03, 0x02, 0x01}, /* 3, -14.5dB */
-	{0x0b, 0x0a, 0x09, 0x08, 0x06, 0x04, 0x02, 0x01}, /* 4, -14.0dB */
-	{0x0b, 0x0b, 0x0a, 0x08, 0x06, 0x04, 0x02, 0x01}, /* 5, -13.5dB */
-	{0x0c, 0x0c, 0x0a, 0x09, 0x06, 0x04, 0x02, 0x01}, /* 6, -13.0dB */
-	{0x0d, 0x0c, 0x0b, 0x09, 0x07, 0x04, 0x02, 0x01}, /* 7, -12.5dB */
-	{0x0d, 0x0d, 0x0c, 0x0a, 0x07, 0x05, 0x02, 0x01}, /* 8, -12.0dB */
-	{0x0e, 0x0e, 0x0c, 0x0a, 0x08, 0x05, 0x02, 0x01}, /* 9, -11.5dB */
-	{0x0f, 0x0f, 0x0d, 0x0b, 0x08, 0x05, 0x03, 0x01}, /* 10, -11.0dB */
-	{0x10, 0x10, 0x0e, 0x0b, 0x08, 0x05, 0x03, 0x01}, /* 11, -10.5dB */
-	{0x11, 0x11, 0x0f, 0x0c, 0x09, 0x06, 0x03, 0x01}, /* 12, -10.0dB */
-	{0x12, 0x12, 0x0f, 0x0c, 0x09, 0x06, 0x03, 0x01}, /* 13, -9.5dB */
-	{0x13, 0x13, 0x10, 0x0d, 0x0a, 0x06, 0x03, 0x01}, /* 14, -9.0dB */
-	{0x14, 0x14, 0x11, 0x0e, 0x0b, 0x07, 0x03, 0x02}, /* 15, -8.5dB */
-	{0x16, 0x15, 0x12, 0x0f, 0x0b, 0x07, 0x04, 0x01}, /* 16, -8.0dB */
-	{0x17, 0x16, 0x13, 0x10, 0x0c, 0x08, 0x04, 0x02}, /* 17, -7.5dB */
-	{0x18, 0x17, 0x15, 0x11, 0x0c, 0x08, 0x04, 0x02}, /* 18, -7.0dB */
-	{0x1a, 0x19, 0x16, 0x12, 0x0d, 0x09, 0x04, 0x02}, /* 19, -6.5dB */
-	{0x1b, 0x1a, 0x17, 0x13, 0x0e, 0x09, 0x04, 0x02}, /* 20, -6.0dB */
-	{0x1d, 0x1c, 0x18, 0x14, 0x0f, 0x0a, 0x05, 0x02}, /* 21, -5.5dB */
-	{0x1f, 0x1e, 0x1a, 0x15, 0x10, 0x0a, 0x05, 0x02}, /* 22, -5.0dB */
-	{0x20, 0x20, 0x1b, 0x16, 0x11, 0x08, 0x05, 0x02}, /* 23, -4.5dB */
-	{0x22, 0x21, 0x1d, 0x18, 0x11, 0x0b, 0x06, 0x02}, /* 24, -4.0dB */
-	{0x24, 0x23, 0x1f, 0x19, 0x13, 0x0c, 0x06, 0x03}, /* 25, -3.5dB */
-	{0x26, 0x25, 0x21, 0x1b, 0x14, 0x0d, 0x06, 0x03}, /* 26, -3.0dB */
-	{0x28, 0x28, 0x22, 0x1c, 0x15, 0x0d, 0x07, 0x03}, /* 27, -2.5dB */
-	{0x2b, 0x2a, 0x25, 0x1e, 0x16, 0x0e, 0x07, 0x03}, /* 28, -2.0dB */
-	{0x2d, 0x2d, 0x27, 0x1f, 0x18, 0x0f, 0x08, 0x03}, /* 29, -1.5dB */
-	{0x30, 0x2f, 0x29, 0x21, 0x19, 0x10, 0x08, 0x03}, /* 30, -1.0dB */
-	{0x33, 0x32, 0x2b, 0x23, 0x1a, 0x11, 0x08, 0x04}, /* 31, -0.5dB */
-	{0x36, 0x35, 0x2e, 0x25, 0x1c, 0x12, 0x09, 0x04} /* 32, +0dB */
-};
-
-static const u8 cckswing_table_ch14[CCK_TABLE_SIZE][8] = {
-	{0x09, 0x08, 0x07, 0x04, 0x00, 0x00, 0x00, 0x00}, /* 0, -16.0dB */
-	{0x09, 0x09, 0x08, 0x05, 0x00, 0x00, 0x00, 0x00}, /* 1, -15.5dB */
-	{0x0a, 0x09, 0x08, 0x05, 0x00, 0x00, 0x00, 0x00}, /* 2, -15.0dB */
-	{0x0a, 0x0a, 0x09, 0x05, 0x00, 0x00, 0x00, 0x00}, /* 3, -14.5dB */
-	{0x0b, 0x0a, 0x09, 0x05, 0x00, 0x00, 0x00, 0x00}, /* 4, -14.0dB */
-	{0x0b, 0x0b, 0x0a, 0x06, 0x00, 0x00, 0x00, 0x00}, /* 5, -13.5dB */
-	{0x0c, 0x0c, 0x0a, 0x06, 0x00, 0x00, 0x00, 0x00}, /* 6, -13.0dB */
-	{0x0d, 0x0c, 0x0b, 0x06, 0x00, 0x00, 0x00, 0x00}, /* 7, -12.5dB */
-	{0x0d, 0x0d, 0x0c, 0x07, 0x00, 0x00, 0x00, 0x00}, /* 8, -12.0dB */
-	{0x0e, 0x0e, 0x0c, 0x07, 0x00, 0x00, 0x00, 0x00}, /* 9, -11.5dB */
-	{0x0f, 0x0f, 0x0d, 0x08, 0x00, 0x00, 0x00, 0x00}, /* 10, -11.0dB */
-	{0x10, 0x10, 0x0e, 0x08, 0x00, 0x00, 0x00, 0x00}, /* 11, -10.5dB */
-	{0x11, 0x11, 0x0f, 0x09, 0x00, 0x00, 0x00, 0x00}, /* 12, -10.0dB */
-	{0x12, 0x12, 0x0f, 0x09, 0x00, 0x00, 0x00, 0x00}, /* 13, -9.5dB */
-	{0x13, 0x13, 0x10, 0x0a, 0x00, 0x00, 0x00, 0x00}, /* 14, -9.0dB */
-	{0x14, 0x14, 0x11, 0x0a, 0x00, 0x00, 0x00, 0x00}, /* 15, -8.5dB */
-	{0x16, 0x15, 0x12, 0x0b, 0x00, 0x00, 0x00, 0x00}, /* 16, -8.0dB */
-	{0x17, 0x16, 0x13, 0x0b, 0x00, 0x00, 0x00, 0x00}, /* 17, -7.5dB */
-	{0x18, 0x17, 0x15, 0x0c, 0x00, 0x00, 0x00, 0x00}, /* 18, -7.0dB */
-	{0x1a, 0x19, 0x16, 0x0d, 0x00, 0x00, 0x00, 0x00}, /* 19, -6.5dB */
-	{0x1b, 0x1a, 0x17, 0x0e, 0x00, 0x00, 0x00, 0x00}, /* 20, -6.0dB */
-	{0x1d, 0x1c, 0x18, 0x0e, 0x00, 0x00, 0x00, 0x00}, /* 21, -5.5dB */
-	{0x1f, 0x1e, 0x1a, 0x0f, 0x00, 0x00, 0x00, 0x00}, /* 22, -5.0dB */
-	{0x20, 0x20, 0x1b, 0x10, 0x00, 0x00, 0x00, 0x00}, /* 23, -4.5dB */
-	{0x22, 0x21, 0x1d, 0x11, 0x00, 0x00, 0x00, 0x00}, /* 24, -4.0dB */
-	{0x24, 0x23, 0x1f, 0x12, 0x00, 0x00, 0x00, 0x00}, /* 25, -3.5dB */
-	{0x26, 0x25, 0x21, 0x13, 0x00, 0x00, 0x00, 0x00}, /* 26, -3.0dB */
-	{0x28, 0x28, 0x24, 0x14, 0x00, 0x00, 0x00, 0x00}, /* 27, -2.5dB */
-	{0x2b, 0x2a, 0x25, 0x15, 0x00, 0x00, 0x00, 0x00}, /* 28, -2.0dB */
-	{0x2d, 0x2d, 0x17, 0x17, 0x00, 0x00, 0x00, 0x00}, /* 29, -1.5dB */
-	{0x30, 0x2f, 0x29, 0x18, 0x00, 0x00, 0x00, 0x00}, /* 30, -1.0dB */
-	{0x33, 0x32, 0x2b, 0x19, 0x00, 0x00, 0x00, 0x00}, /* 31, -0.5dB */
-	{0x36, 0x35, 0x2e, 0x1b, 0x00, 0x00, 0x00, 0x00} /* 32, +0dB */
-};
-
 static const u32 edca_setting_dl[PEER_MAX] = {
 	0xa44f,		/* 0 UNKNOWN */
 	0x5ea44f,	/* 1 REALTEK_90 */
-- 
cgit v1.2.3-59-g8ed1b


From 253e5aba937973fd29bd5c559d21e35d0642242e Mon Sep 17 00:00:00 2001
From: YueHaibing <yuehaibing@huawei.com>
Date: Tue, 21 Jan 2020 10:19:24 +0800
Subject: rtlwifi: rtl8192ee: remove unused variables

drivers/net/wireless/realtek/rtlwifi/rtl8192ee/dm.c:15:18:
 warning: ofdmswing_table defined but not used [-Wunused-const-variable=]
drivers/net/wireless/realtek/rtlwifi/rtl8192ee/dm.c:61:17:
 warning: cckswing_table_ch1ch13 defined but not used [-Wunused-const-variable=]
drivers/net/wireless/realtek/rtlwifi/rtl8192ee/dm.c:97:17:
 warning: cckswing_table_ch14 defined but not used [-Wunused-const-variable=]

These variable is never used, so remove them.

Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 .../net/wireless/realtek/rtlwifi/rtl8192ee/dm.c    | 118 ---------------------
 1 file changed, 118 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/dm.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/dm.c
index 648f9108ed4b..551aa86825ed 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/dm.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/dm.c
@@ -12,124 +12,6 @@
 #include "fw.h"
 #include "trx.h"
 
-static const u32 ofdmswing_table[OFDM_TABLE_SIZE] = {
-	0x7f8001fe,		/* 0, +6.0dB */
-	0x788001e2,		/* 1, +5.5dB */
-	0x71c001c7,		/* 2, +5.0dB */
-	0x6b8001ae,		/* 3, +4.5dB */
-	0x65400195,		/* 4, +4.0dB */
-	0x5fc0017f,		/* 5, +3.5dB */
-	0x5a400169,		/* 6, +3.0dB */
-	0x55400155,		/* 7, +2.5dB */
-	0x50800142,		/* 8, +2.0dB */
-	0x4c000130,		/* 9, +1.5dB */
-	0x47c0011f,		/* 10, +1.0dB */
-	0x43c0010f,		/* 11, +0.5dB */
-	0x40000100,		/* 12, +0dB */
-	0x3c8000f2,		/* 13, -0.5dB */
-	0x390000e4,		/* 14, -1.0dB */
-	0x35c000d7,		/* 15, -1.5dB */
-	0x32c000cb,		/* 16, -2.0dB */
-	0x300000c0,		/* 17, -2.5dB */
-	0x2d4000b5,		/* 18, -3.0dB */
-	0x2ac000ab,		/* 19, -3.5dB */
-	0x288000a2,		/* 20, -4.0dB */
-	0x26000098,		/* 21, -4.5dB */
-	0x24000090,		/* 22, -5.0dB */
-	0x22000088,		/* 23, -5.5dB */
-	0x20000080,		/* 24, -6.0dB */
-	0x1e400079,		/* 25, -6.5dB */
-	0x1c800072,		/* 26, -7.0dB */
-	0x1b00006c,		/* 27. -7.5dB */
-	0x19800066,		/* 28, -8.0dB */
-	0x18000060,		/* 29, -8.5dB */
-	0x16c0005b,		/* 30, -9.0dB */
-	0x15800056,		/* 31, -9.5dB */
-	0x14400051,		/* 32, -10.0dB */
-	0x1300004c,		/* 33, -10.5dB */
-	0x12000048,		/* 34, -11.0dB */
-	0x11000044,		/* 35, -11.5dB */
-	0x10000040,		/* 36, -12.0dB */
-	0x0f00003c,		/* 37, -12.5dB */
-	0x0e400039,		/* 38, -13.0dB */
-	0x0d800036,		/* 39, -13.5dB */
-	0x0cc00033,		/* 40, -14.0dB */
-	0x0c000030,		/* 41, -14.5dB */
-	0x0b40002d,		/* 42, -15.0dB */
-};
-
-static const u8 cckswing_table_ch1ch13[CCK_TABLE_SIZE][8] = {
-	{0x36, 0x35, 0x2e, 0x25, 0x1c, 0x12, 0x09, 0x04}, /* 0, +0dB */
-	{0x33, 0x32, 0x2b, 0x23, 0x1a, 0x11, 0x08, 0x04}, /* 1, -0.5dB */
-	{0x30, 0x2f, 0x29, 0x21, 0x19, 0x10, 0x08, 0x03}, /* 2, -1.0dB */
-	{0x2d, 0x2d, 0x27, 0x1f, 0x18, 0x0f, 0x08, 0x03}, /* 3, -1.5dB */
-	{0x2b, 0x2a, 0x25, 0x1e, 0x16, 0x0e, 0x07, 0x03}, /* 4, -2.0dB */
-	{0x28, 0x28, 0x22, 0x1c, 0x15, 0x0d, 0x07, 0x03}, /* 5, -2.5dB */
-	{0x26, 0x25, 0x21, 0x1b, 0x14, 0x0d, 0x06, 0x03}, /* 6, -3.0dB */
-	{0x24, 0x23, 0x1f, 0x19, 0x13, 0x0c, 0x06, 0x03}, /* 7, -3.5dB */
-	{0x22, 0x21, 0x1d, 0x18, 0x11, 0x0b, 0x06, 0x02}, /* 8, -4.0dB */
-	{0x20, 0x20, 0x1b, 0x16, 0x11, 0x08, 0x05, 0x02}, /* 9, -4.5dB */
-	{0x1f, 0x1e, 0x1a, 0x15, 0x10, 0x0a, 0x05, 0x02}, /* 10, -5.0dB */
-	{0x1d, 0x1c, 0x18, 0x14, 0x0f, 0x0a, 0x05, 0x02}, /* 11, -5.5dB */
-	{0x1b, 0x1a, 0x17, 0x13, 0x0e, 0x09, 0x04, 0x02}, /* 12, -6.0dB */
-	{0x1a, 0x19, 0x16, 0x12, 0x0d, 0x09, 0x04, 0x02}, /* 13, -6.5dB */
-	{0x18, 0x17, 0x15, 0x11, 0x0c, 0x08, 0x04, 0x02}, /* 14, -7.0dB */
-	{0x17, 0x16, 0x13, 0x10, 0x0c, 0x08, 0x04, 0x02}, /* 15, -7.5dB */
-	{0x16, 0x15, 0x12, 0x0f, 0x0b, 0x07, 0x04, 0x01}, /* 16, -8.0dB */
-	{0x14, 0x14, 0x11, 0x0e, 0x0b, 0x07, 0x03, 0x02}, /* 17, -8.5dB */
-	{0x13, 0x13, 0x10, 0x0d, 0x0a, 0x06, 0x03, 0x01}, /* 18, -9.0dB */
-	{0x12, 0x12, 0x0f, 0x0c, 0x09, 0x06, 0x03, 0x01}, /* 19, -9.5dB */
-	{0x11, 0x11, 0x0f, 0x0c, 0x09, 0x06, 0x03, 0x01}, /* 20, -10.0dB */
-	{0x10, 0x10, 0x0e, 0x0b, 0x08, 0x05, 0x03, 0x01}, /* 21, -10.5dB */
-	{0x0f, 0x0f, 0x0d, 0x0b, 0x08, 0x05, 0x03, 0x01}, /* 22, -11.0dB */
-	{0x0e, 0x0e, 0x0c, 0x0a, 0x08, 0x05, 0x02, 0x01}, /* 23, -11.5dB */
-	{0x0d, 0x0d, 0x0c, 0x0a, 0x07, 0x05, 0x02, 0x01}, /* 24, -12.0dB */
-	{0x0d, 0x0c, 0x0b, 0x09, 0x07, 0x04, 0x02, 0x01}, /* 25, -12.5dB */
-	{0x0c, 0x0c, 0x0a, 0x09, 0x06, 0x04, 0x02, 0x01}, /* 26, -13.0dB */
-	{0x0b, 0x0b, 0x0a, 0x08, 0x06, 0x04, 0x02, 0x01}, /* 27, -13.5dB */
-	{0x0b, 0x0a, 0x09, 0x08, 0x06, 0x04, 0x02, 0x01}, /* 28, -14.0dB */
-	{0x0a, 0x0a, 0x09, 0x07, 0x05, 0x03, 0x02, 0x01}, /* 29, -14.5dB */
-	{0x0a, 0x09, 0x08, 0x07, 0x05, 0x03, 0x02, 0x01}, /* 30, -15.0dB */
-	{0x09, 0x09, 0x08, 0x06, 0x05, 0x03, 0x01, 0x01}, /* 31, -15.5dB */
-	{0x09, 0x08, 0x07, 0x06, 0x04, 0x03, 0x01, 0x01}  /* 32, -16.0dB */
-};
-
-static const u8 cckswing_table_ch14[CCK_TABLE_SIZE][8] = {
-	{0x36, 0x35, 0x2e, 0x1b, 0x00, 0x00, 0x00, 0x00}, /* 0, +0dB */
-	{0x33, 0x32, 0x2b, 0x19, 0x00, 0x00, 0x00, 0x00}, /* 1, -0.5dB */
-	{0x30, 0x2f, 0x29, 0x18, 0x00, 0x00, 0x00, 0x00}, /* 2, -1.0dB */
-	{0x2d, 0x2d, 0x17, 0x17, 0x00, 0x00, 0x00, 0x00}, /* 3, -1.5dB */
-	{0x2b, 0x2a, 0x25, 0x15, 0x00, 0x00, 0x00, 0x00}, /* 4, -2.0dB */
-	{0x28, 0x28, 0x24, 0x14, 0x00, 0x00, 0x00, 0x00}, /* 5, -2.5dB */
-	{0x26, 0x25, 0x21, 0x13, 0x00, 0x00, 0x00, 0x00}, /* 6, -3.0dB */
-	{0x24, 0x23, 0x1f, 0x12, 0x00, 0x00, 0x00, 0x00}, /* 7, -3.5dB */
-	{0x22, 0x21, 0x1d, 0x11, 0x00, 0x00, 0x00, 0x00}, /* 8, -4.0dB */
-	{0x20, 0x20, 0x1b, 0x10, 0x00, 0x00, 0x00, 0x00}, /* 9, -4.5dB */
-	{0x1f, 0x1e, 0x1a, 0x0f, 0x00, 0x00, 0x00, 0x00}, /* 10, -5.0dB */
-	{0x1d, 0x1c, 0x18, 0x0e, 0x00, 0x00, 0x00, 0x00}, /* 11, -5.5dB */
-	{0x1b, 0x1a, 0x17, 0x0e, 0x00, 0x00, 0x00, 0x00}, /* 12, -6.0dB */
-	{0x1a, 0x19, 0x16, 0x0d, 0x00, 0x00, 0x00, 0x00}, /* 13, -6.5dB */
-	{0x18, 0x17, 0x15, 0x0c, 0x00, 0x00, 0x00, 0x00}, /* 14, -7.0dB */
-	{0x17, 0x16, 0x13, 0x0b, 0x00, 0x00, 0x00, 0x00}, /* 15, -7.5dB */
-	{0x16, 0x15, 0x12, 0x0b, 0x00, 0x00, 0x00, 0x00}, /* 16, -8.0dB */
-	{0x14, 0x14, 0x11, 0x0a, 0x00, 0x00, 0x00, 0x00}, /* 17, -8.5dB */
-	{0x13, 0x13, 0x10, 0x0a, 0x00, 0x00, 0x00, 0x00}, /* 18, -9.0dB */
-	{0x12, 0x12, 0x0f, 0x09, 0x00, 0x00, 0x00, 0x00}, /* 19, -9.5dB */
-	{0x11, 0x11, 0x0f, 0x09, 0x00, 0x00, 0x00, 0x00}, /* 20, -10.0dB */
-	{0x10, 0x10, 0x0e, 0x08, 0x00, 0x00, 0x00, 0x00}, /* 21, -10.5dB */
-	{0x0f, 0x0f, 0x0d, 0x08, 0x00, 0x00, 0x00, 0x00}, /* 22, -11.0dB */
-	{0x0e, 0x0e, 0x0c, 0x07, 0x00, 0x00, 0x00, 0x00}, /* 23, -11.5dB */
-	{0x0d, 0x0d, 0x0c, 0x07, 0x00, 0x00, 0x00, 0x00}, /* 24, -12.0dB */
-	{0x0d, 0x0c, 0x0b, 0x06, 0x00, 0x00, 0x00, 0x00}, /* 25, -12.5dB */
-	{0x0c, 0x0c, 0x0a, 0x06, 0x00, 0x00, 0x00, 0x00}, /* 26, -13.0dB */
-	{0x0b, 0x0b, 0x0a, 0x06, 0x00, 0x00, 0x00, 0x00}, /* 27, -13.5dB */
-	{0x0b, 0x0a, 0x09, 0x05, 0x00, 0x00, 0x00, 0x00}, /* 28, -14.0dB */
-	{0x0a, 0x0a, 0x09, 0x05, 0x00, 0x00, 0x00, 0x00}, /* 29, -14.5dB */
-	{0x0a, 0x09, 0x08, 0x05, 0x00, 0x00, 0x00, 0x00}, /* 30, -15.0dB */
-	{0x09, 0x09, 0x08, 0x05, 0x00, 0x00, 0x00, 0x00}, /* 31, -15.5dB */
-	{0x09, 0x08, 0x07, 0x04, 0x00, 0x00, 0x00, 0x00}  /* 32, -16.0dB */
-};
-
 static void rtl92ee_dm_false_alarm_counter_statistics(struct ieee80211_hw *hw)
 {
 	u32 ret_value;
-- 
cgit v1.2.3-59-g8ed1b


From c5f9852411098474ab21f5d7b1b84e5cdd59ca5a Mon Sep 17 00:00:00 2001
From: YueHaibing <yuehaibing@huawei.com>
Date: Tue, 21 Jan 2020 11:04:28 +0800
Subject: rtlwifi: rtl8723ae: remove unused variables

drivers/net/wireless/realtek/rtlwifi/rtl8723ae/dm.c:16:18:
 warning: ofdmswing_table defined but not used [-Wunused-const-variable=]
drivers/net/wireless/realtek/rtlwifi/rtl8723ae/dm.c:56:17:
 warning: cckswing_table_ch1ch13 defined but not used [-Wunused-const-variable=]
drivers/net/wireless/realtek/rtlwifi/rtl8723ae/dm.c:92:17:
 warning: cckswing_table_ch14 defined but not used [-Wunused-const-variable=]

These variable is never used, so remove them.

Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 .../net/wireless/realtek/rtlwifi/rtl8723ae/dm.c    | 112 ---------------------
 1 file changed, 112 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/dm.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/dm.c
index d8260c7afe09..c61a92df9d73 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/dm.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/dm.c
@@ -13,118 +13,6 @@
 #include "fw.h"
 #include "hal_btc.h"
 
-static const u32 ofdmswing_table[OFDM_TABLE_SIZE] = {
-	0x7f8001fe,
-	0x788001e2,
-	0x71c001c7,
-	0x6b8001ae,
-	0x65400195,
-	0x5fc0017f,
-	0x5a400169,
-	0x55400155,
-	0x50800142,
-	0x4c000130,
-	0x47c0011f,
-	0x43c0010f,
-	0x40000100,
-	0x3c8000f2,
-	0x390000e4,
-	0x35c000d7,
-	0x32c000cb,
-	0x300000c0,
-	0x2d4000b5,
-	0x2ac000ab,
-	0x288000a2,
-	0x26000098,
-	0x24000090,
-	0x22000088,
-	0x20000080,
-	0x1e400079,
-	0x1c800072,
-	0x1b00006c,
-	0x19800066,
-	0x18000060,
-	0x16c0005b,
-	0x15800056,
-	0x14400051,
-	0x1300004c,
-	0x12000048,
-	0x11000044,
-	0x10000040,
-};
-
-static const u8 cckswing_table_ch1ch13[CCK_TABLE_SIZE][8] = {
-	{0x36, 0x35, 0x2e, 0x25, 0x1c, 0x12, 0x09, 0x04},
-	{0x33, 0x32, 0x2b, 0x23, 0x1a, 0x11, 0x08, 0x04},
-	{0x30, 0x2f, 0x29, 0x21, 0x19, 0x10, 0x08, 0x03},
-	{0x2d, 0x2d, 0x27, 0x1f, 0x18, 0x0f, 0x08, 0x03},
-	{0x2b, 0x2a, 0x25, 0x1e, 0x16, 0x0e, 0x07, 0x03},
-	{0x28, 0x28, 0x22, 0x1c, 0x15, 0x0d, 0x07, 0x03},
-	{0x26, 0x25, 0x21, 0x1b, 0x14, 0x0d, 0x06, 0x03},
-	{0x24, 0x23, 0x1f, 0x19, 0x13, 0x0c, 0x06, 0x03},
-	{0x22, 0x21, 0x1d, 0x18, 0x11, 0x0b, 0x06, 0x02},
-	{0x20, 0x20, 0x1b, 0x16, 0x11, 0x08, 0x05, 0x02},
-	{0x1f, 0x1e, 0x1a, 0x15, 0x10, 0x0a, 0x05, 0x02},
-	{0x1d, 0x1c, 0x18, 0x14, 0x0f, 0x0a, 0x05, 0x02},
-	{0x1b, 0x1a, 0x17, 0x13, 0x0e, 0x09, 0x04, 0x02},
-	{0x1a, 0x19, 0x16, 0x12, 0x0d, 0x09, 0x04, 0x02},
-	{0x18, 0x17, 0x15, 0x11, 0x0c, 0x08, 0x04, 0x02},
-	{0x17, 0x16, 0x13, 0x10, 0x0c, 0x08, 0x04, 0x02},
-	{0x16, 0x15, 0x12, 0x0f, 0x0b, 0x07, 0x04, 0x01},
-	{0x14, 0x14, 0x11, 0x0e, 0x0b, 0x07, 0x03, 0x02},
-	{0x13, 0x13, 0x10, 0x0d, 0x0a, 0x06, 0x03, 0x01},
-	{0x12, 0x12, 0x0f, 0x0c, 0x09, 0x06, 0x03, 0x01},
-	{0x11, 0x11, 0x0f, 0x0c, 0x09, 0x06, 0x03, 0x01},
-	{0x10, 0x10, 0x0e, 0x0b, 0x08, 0x05, 0x03, 0x01},
-	{0x0f, 0x0f, 0x0d, 0x0b, 0x08, 0x05, 0x03, 0x01},
-	{0x0e, 0x0e, 0x0c, 0x0a, 0x08, 0x05, 0x02, 0x01},
-	{0x0d, 0x0d, 0x0c, 0x0a, 0x07, 0x05, 0x02, 0x01},
-	{0x0d, 0x0c, 0x0b, 0x09, 0x07, 0x04, 0x02, 0x01},
-	{0x0c, 0x0c, 0x0a, 0x09, 0x06, 0x04, 0x02, 0x01},
-	{0x0b, 0x0b, 0x0a, 0x08, 0x06, 0x04, 0x02, 0x01},
-	{0x0b, 0x0a, 0x09, 0x08, 0x06, 0x04, 0x02, 0x01},
-	{0x0a, 0x0a, 0x09, 0x07, 0x05, 0x03, 0x02, 0x01},
-	{0x0a, 0x09, 0x08, 0x07, 0x05, 0x03, 0x02, 0x01},
-	{0x09, 0x09, 0x08, 0x06, 0x05, 0x03, 0x01, 0x01},
-	{0x09, 0x08, 0x07, 0x06, 0x04, 0x03, 0x01, 0x01}
-};
-
-static const u8 cckswing_table_ch14[CCK_TABLE_SIZE][8] = {
-	{0x36, 0x35, 0x2e, 0x1b, 0x00, 0x00, 0x00, 0x00},
-	{0x33, 0x32, 0x2b, 0x19, 0x00, 0x00, 0x00, 0x00},
-	{0x30, 0x2f, 0x29, 0x18, 0x00, 0x00, 0x00, 0x00},
-	{0x2d, 0x2d, 0x17, 0x17, 0x00, 0x00, 0x00, 0x00},
-	{0x2b, 0x2a, 0x25, 0x15, 0x00, 0x00, 0x00, 0x00},
-	{0x28, 0x28, 0x24, 0x14, 0x00, 0x00, 0x00, 0x00},
-	{0x26, 0x25, 0x21, 0x13, 0x00, 0x00, 0x00, 0x00},
-	{0x24, 0x23, 0x1f, 0x12, 0x00, 0x00, 0x00, 0x00},
-	{0x22, 0x21, 0x1d, 0x11, 0x00, 0x00, 0x00, 0x00},
-	{0x20, 0x20, 0x1b, 0x10, 0x00, 0x00, 0x00, 0x00},
-	{0x1f, 0x1e, 0x1a, 0x0f, 0x00, 0x00, 0x00, 0x00},
-	{0x1d, 0x1c, 0x18, 0x0e, 0x00, 0x00, 0x00, 0x00},
-	{0x1b, 0x1a, 0x17, 0x0e, 0x00, 0x00, 0x00, 0x00},
-	{0x1a, 0x19, 0x16, 0x0d, 0x00, 0x00, 0x00, 0x00},
-	{0x18, 0x17, 0x15, 0x0c, 0x00, 0x00, 0x00, 0x00},
-	{0x17, 0x16, 0x13, 0x0b, 0x00, 0x00, 0x00, 0x00},
-	{0x16, 0x15, 0x12, 0x0b, 0x00, 0x00, 0x00, 0x00},
-	{0x14, 0x14, 0x11, 0x0a, 0x00, 0x00, 0x00, 0x00},
-	{0x13, 0x13, 0x10, 0x0a, 0x00, 0x00, 0x00, 0x00},
-	{0x12, 0x12, 0x0f, 0x09, 0x00, 0x00, 0x00, 0x00},
-	{0x11, 0x11, 0x0f, 0x09, 0x00, 0x00, 0x00, 0x00},
-	{0x10, 0x10, 0x0e, 0x08, 0x00, 0x00, 0x00, 0x00},
-	{0x0f, 0x0f, 0x0d, 0x08, 0x00, 0x00, 0x00, 0x00},
-	{0x0e, 0x0e, 0x0c, 0x07, 0x00, 0x00, 0x00, 0x00},
-	{0x0d, 0x0d, 0x0c, 0x07, 0x00, 0x00, 0x00, 0x00},
-	{0x0d, 0x0c, 0x0b, 0x06, 0x00, 0x00, 0x00, 0x00},
-	{0x0c, 0x0c, 0x0a, 0x06, 0x00, 0x00, 0x00, 0x00},
-	{0x0b, 0x0b, 0x0a, 0x06, 0x00, 0x00, 0x00, 0x00},
-	{0x0b, 0x0a, 0x09, 0x05, 0x00, 0x00, 0x00, 0x00},
-	{0x0a, 0x0a, 0x09, 0x05, 0x00, 0x00, 0x00, 0x00},
-	{0x0a, 0x09, 0x08, 0x05, 0x00, 0x00, 0x00, 0x00},
-	{0x09, 0x09, 0x08, 0x05, 0x00, 0x00, 0x00, 0x00},
-	{0x09, 0x08, 0x07, 0x04, 0x00, 0x00, 0x00, 0x00}
-};
-
 static u8 rtl8723e_dm_initial_gain_min_pwdb(struct ieee80211_hw *hw)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
-- 
cgit v1.2.3-59-g8ed1b


From f76c34082b24534c813a9e3e175ef778c11e9950 Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.king@canonical.com>
Date: Wed, 22 Jan 2020 09:33:40 +0000
Subject: rtlwifi: btcoex: fix spelling mistake "initilized" -> "initialized"

There is a spelling mistake in one of the fields in the btc_coexist struct,
fix it.

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtc8192e2ant.c | 2 +-
 drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtcoutsrc.c    | 2 +-
 drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtcoutsrc.h    | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtc8192e2ant.c b/drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtc8192e2ant.c
index 3c96c320236c..658ff425c256 100644
--- a/drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtc8192e2ant.c
+++ b/drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtc8192e2ant.c
@@ -862,7 +862,7 @@ static void btc8192e2ant_set_sw_rf_rx_lpf_corner(struct btc_coexist *btcoexist,
 		/* Resume RF Rx LPF corner
 		 * After initialized, we can use coex_dm->btRf0x1eBackup
 		 */
-		if (btcoexist->initilized) {
+		if (btcoexist->initialized) {
 			RT_TRACE(rtlpriv, COMP_BT_COEXIST, DBG_LOUD,
 				 "[BTCoex], Resume RF Rx LPF corner!!\n");
 			btcoexist->btc_set_rf_reg(btcoexist, BTC_RF_A, 0x1e,
diff --git a/drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtcoutsrc.c b/drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtcoutsrc.c
index 191dafd03189..a4940a3842de 100644
--- a/drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtcoutsrc.c
+++ b/drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtcoutsrc.c
@@ -1461,7 +1461,7 @@ void exhalbtc_init_coex_dm(struct btc_coexist *btcoexist)
 			ex_btc8192e2ant_init_coex_dm(btcoexist);
 	}
 
-	btcoexist->initilized = true;
+	btcoexist->initialized = true;
 }
 
 void exhalbtc_ips_notify(struct btc_coexist *btcoexist, u8 type)
diff --git a/drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtcoutsrc.h b/drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtcoutsrc.h
index 8c0a7fdbf200..a96a995dd850 100644
--- a/drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtcoutsrc.h
+++ b/drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtcoutsrc.h
@@ -679,7 +679,7 @@ struct btc_coexist {
 	bool auto_report_2ant;
 	bool dbg_mode_1ant;
 	bool dbg_mode_2ant;
-	bool initilized;
+	bool initialized;
 	bool stop_coex_dm;
 	bool manual_control;
 	struct btc_statistics statistics;
-- 
cgit v1.2.3-59-g8ed1b


From c2f9a4e4a5abfc84c01b738496b3fd2d471e0b18 Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.king@canonical.com>
Date: Sun, 26 Jan 2020 00:09:54 +0000
Subject: iwlegacy: ensure loop counter addr does not wrap and cause an
 infinite loop

The loop counter addr is a u16 where as the upper limit of the loop
is an int. In the unlikely event that the il->cfg->eeprom_size is
greater than 64K then we end up with an infinite loop since addr will
wrap around an never reach upper loop limit. Fix this by making addr
an int.

Addresses-Coverity: ("Infinite loop")
Fixes: be663ab67077 ("iwlwifi: split the drivers for agn and legacy devices 3945/4965")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Acked-by: Stanislaw Gruszka <stf_xl@wp.pl>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
---
 drivers/net/wireless/intel/iwlegacy/common.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlegacy/common.c b/drivers/net/wireless/intel/iwlegacy/common.c
index d966b29b45ee..348c17ce72f5 100644
--- a/drivers/net/wireless/intel/iwlegacy/common.c
+++ b/drivers/net/wireless/intel/iwlegacy/common.c
@@ -699,7 +699,7 @@ il_eeprom_init(struct il_priv *il)
 	u32 gp = _il_rd(il, CSR_EEPROM_GP);
 	int sz;
 	int ret;
-	u16 addr;
+	int addr;
 
 	/* allocate eeprom */
 	sz = il->cfg->eeprom_size;
-- 
cgit v1.2.3-59-g8ed1b


From 20a1452c35425b2cef76f21f8395ef069dfddfa9 Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Wed, 22 Jan 2020 00:17:51 +0100
Subject: netfilter: nf_tables: add nft_setelem_parse_key()

Add helper function to parse the set element key netlink attribute.

v4: No changes
v3: New patch

[sbrivio: refactor error paths and labels; use NFT_DATA_VALUE_MAXLEN
  instead of sizeof(*key) in helper, value can be longer than that;
  rebase]
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nf_tables_api.c | 91 +++++++++++++++++++++----------------------
 1 file changed, 45 insertions(+), 46 deletions(-)

diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index 7e63b481cc86..58e3b285a3d1 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -4524,11 +4524,28 @@ static int nft_setelem_parse_flags(const struct nft_set *set,
 	return 0;
 }
 
+static int nft_setelem_parse_key(struct nft_ctx *ctx, struct nft_set *set,
+				 struct nft_data *key, struct nlattr *attr)
+{
+	struct nft_data_desc desc;
+	int err;
+
+	err = nft_data_init(ctx, key, NFT_DATA_VALUE_MAXLEN, &desc, attr);
+	if (err < 0)
+		return err;
+
+	if (desc.type != NFT_DATA_VALUE || desc.len != set->klen) {
+		nft_data_release(key, desc.type);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
 static int nft_get_set_elem(struct nft_ctx *ctx, struct nft_set *set,
 			    const struct nlattr *attr)
 {
 	struct nlattr *nla[NFTA_SET_ELEM_MAX + 1];
-	struct nft_data_desc desc;
 	struct nft_set_elem elem;
 	struct sk_buff *skb;
 	uint32_t flags = 0;
@@ -4547,17 +4564,11 @@ static int nft_get_set_elem(struct nft_ctx *ctx, struct nft_set *set,
 	if (err < 0)
 		return err;
 
-	err = nft_data_init(ctx, &elem.key.val, sizeof(elem.key), &desc,
-			    nla[NFTA_SET_ELEM_KEY]);
+	err = nft_setelem_parse_key(ctx, set, &elem.key.val,
+				    nla[NFTA_SET_ELEM_KEY]);
 	if (err < 0)
 		return err;
 
-	err = -EINVAL;
-	if (desc.type != NFT_DATA_VALUE || desc.len != set->klen) {
-		nft_data_release(&elem.key.val, desc.type);
-		return err;
-	}
-
 	priv = set->ops->get(ctx->net, set, &elem, flags);
 	if (IS_ERR(priv))
 		return PTR_ERR(priv);
@@ -4756,13 +4767,13 @@ static int nft_add_set_elem(struct nft_ctx *ctx, struct nft_set *set,
 {
 	struct nlattr *nla[NFTA_SET_ELEM_MAX + 1];
 	u8 genmask = nft_genmask_next(ctx->net);
-	struct nft_data_desc d1, d2;
 	struct nft_set_ext_tmpl tmpl;
 	struct nft_set_ext *ext, *ext2;
 	struct nft_set_elem elem;
 	struct nft_set_binding *binding;
 	struct nft_object *obj = NULL;
 	struct nft_userdata *udata;
+	struct nft_data_desc desc;
 	struct nft_data data;
 	enum nft_registers dreg;
 	struct nft_trans *trans;
@@ -4828,15 +4839,12 @@ static int nft_add_set_elem(struct nft_ctx *ctx, struct nft_set *set,
 			return err;
 	}
 
-	err = nft_data_init(ctx, &elem.key.val, sizeof(elem.key), &d1,
-			    nla[NFTA_SET_ELEM_KEY]);
+	err = nft_setelem_parse_key(ctx, set, &elem.key.val,
+				    nla[NFTA_SET_ELEM_KEY]);
 	if (err < 0)
 		goto err1;
-	err = -EINVAL;
-	if (d1.type != NFT_DATA_VALUE || d1.len != set->klen)
-		goto err2;
 
-	nft_set_ext_add_length(&tmpl, NFT_SET_EXT_KEY, d1.len);
+	nft_set_ext_add_length(&tmpl, NFT_SET_EXT_KEY, set->klen);
 	if (timeout > 0) {
 		nft_set_ext_add(&tmpl, NFT_SET_EXT_EXPIRATION);
 		if (timeout != set->timeout)
@@ -4859,13 +4867,13 @@ static int nft_add_set_elem(struct nft_ctx *ctx, struct nft_set *set,
 	}
 
 	if (nla[NFTA_SET_ELEM_DATA] != NULL) {
-		err = nft_data_init(ctx, &data, sizeof(data), &d2,
+		err = nft_data_init(ctx, &data, sizeof(data), &desc,
 				    nla[NFTA_SET_ELEM_DATA]);
 		if (err < 0)
 			goto err2;
 
 		err = -EINVAL;
-		if (set->dtype != NFT_DATA_VERDICT && d2.len != set->dlen)
+		if (set->dtype != NFT_DATA_VERDICT && desc.len != set->dlen)
 			goto err3;
 
 		dreg = nft_type_to_reg(set->dtype);
@@ -4882,18 +4890,18 @@ static int nft_add_set_elem(struct nft_ctx *ctx, struct nft_set *set,
 
 			err = nft_validate_register_store(&bind_ctx, dreg,
 							  &data,
-							  d2.type, d2.len);
+							  desc.type, desc.len);
 			if (err < 0)
 				goto err3;
 
-			if (d2.type == NFT_DATA_VERDICT &&
+			if (desc.type == NFT_DATA_VERDICT &&
 			    (data.verdict.code == NFT_GOTO ||
 			     data.verdict.code == NFT_JUMP))
 				nft_validate_state_update(ctx->net,
 							  NFT_VALIDATE_NEED);
 		}
 
-		nft_set_ext_add_length(&tmpl, NFT_SET_EXT_DATA, d2.len);
+		nft_set_ext_add_length(&tmpl, NFT_SET_EXT_DATA, desc.len);
 	}
 
 	/* The full maximum length of userdata can exceed the maximum
@@ -4976,9 +4984,9 @@ err4:
 	kfree(elem.priv);
 err3:
 	if (nla[NFTA_SET_ELEM_DATA] != NULL)
-		nft_data_release(&data, d2.type);
+		nft_data_release(&data, desc.type);
 err2:
-	nft_data_release(&elem.key.val, d1.type);
+	nft_data_release(&elem.key.val, NFT_DATA_VALUE);
 err1:
 	return err;
 }
@@ -5074,7 +5082,6 @@ static int nft_del_setelem(struct nft_ctx *ctx, struct nft_set *set,
 {
 	struct nlattr *nla[NFTA_SET_ELEM_MAX + 1];
 	struct nft_set_ext_tmpl tmpl;
-	struct nft_data_desc desc;
 	struct nft_set_elem elem;
 	struct nft_set_ext *ext;
 	struct nft_trans *trans;
@@ -5085,11 +5092,10 @@ static int nft_del_setelem(struct nft_ctx *ctx, struct nft_set *set,
 	err = nla_parse_nested_deprecated(nla, NFTA_SET_ELEM_MAX, attr,
 					  nft_set_elem_policy, NULL);
 	if (err < 0)
-		goto err1;
+		return err;
 
-	err = -EINVAL;
 	if (nla[NFTA_SET_ELEM_KEY] == NULL)
-		goto err1;
+		return -EINVAL;
 
 	nft_set_ext_prepare(&tmpl);
 
@@ -5099,37 +5105,31 @@ static int nft_del_setelem(struct nft_ctx *ctx, struct nft_set *set,
 	if (flags != 0)
 		nft_set_ext_add(&tmpl, NFT_SET_EXT_FLAGS);
 
-	err = nft_data_init(ctx, &elem.key.val, sizeof(elem.key), &desc,
-			    nla[NFTA_SET_ELEM_KEY]);
+	err = nft_setelem_parse_key(ctx, set, &elem.key.val,
+				    nla[NFTA_SET_ELEM_KEY]);
 	if (err < 0)
-		goto err1;
-
-	err = -EINVAL;
-	if (desc.type != NFT_DATA_VALUE || desc.len != set->klen)
-		goto err2;
+		return err;
 
-	nft_set_ext_add_length(&tmpl, NFT_SET_EXT_KEY, desc.len);
+	nft_set_ext_add_length(&tmpl, NFT_SET_EXT_KEY, set->klen);
 
 	err = -ENOMEM;
 	elem.priv = nft_set_elem_init(set, &tmpl, elem.key.val.data, NULL, 0,
 				      0, GFP_KERNEL);
 	if (elem.priv == NULL)
-		goto err2;
+		goto fail_elem;
 
 	ext = nft_set_elem_ext(set, elem.priv);
 	if (flags)
 		*nft_set_ext_flags(ext) = flags;
 
 	trans = nft_trans_elem_alloc(ctx, NFT_MSG_DELSETELEM, set);
-	if (trans == NULL) {
-		err = -ENOMEM;
-		goto err3;
-	}
+	if (trans == NULL)
+		goto fail_trans;
 
 	priv = set->ops->deactivate(ctx->net, set, &elem);
 	if (priv == NULL) {
 		err = -ENOENT;
-		goto err4;
+		goto fail_ops;
 	}
 	kfree(elem.priv);
 	elem.priv = priv;
@@ -5140,13 +5140,12 @@ static int nft_del_setelem(struct nft_ctx *ctx, struct nft_set *set,
 	list_add_tail(&trans->list, &ctx->net->nft.commit_list);
 	return 0;
 
-err4:
+fail_ops:
 	kfree(trans);
-err3:
+fail_trans:
 	kfree(elem.priv);
-err2:
-	nft_data_release(&elem.key.val, desc.type);
-err1:
+fail_elem:
+	nft_data_release(&elem.key.val, NFT_DATA_VALUE);
 	return err;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 7b225d0b5c6dda5fefab578175f210c6fc7e389a Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Wed, 22 Jan 2020 00:17:52 +0100
Subject: netfilter: nf_tables: add NFTA_SET_ELEM_KEY_END attribute

Add NFTA_SET_ELEM_KEY_END attribute to convey the closing element of the
interval between kernel and userspace.

This patch also adds the NFT_SET_EXT_KEY_END extension to store the
closing element value in this interval.

v4: No changes
v3: New patch

[sbrivio: refactor error paths and labels; add corresponding
  nft_set_ext_type for new key; rebase]
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_tables.h        | 14 +++++-
 include/uapi/linux/netfilter/nf_tables.h |  2 +
 net/netfilter/nf_tables_api.c            | 85 +++++++++++++++++++++++---------
 net/netfilter/nft_dynset.c               |  2 +-
 4 files changed, 79 insertions(+), 24 deletions(-)

diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h
index fe7c50acc681..504c0aa93805 100644
--- a/include/net/netfilter/nf_tables.h
+++ b/include/net/netfilter/nf_tables.h
@@ -231,6 +231,7 @@ struct nft_userdata {
  *	struct nft_set_elem - generic representation of set elements
  *
  *	@key: element key
+ *	@key_end: closing element key
  *	@priv: element private data and extensions
  */
 struct nft_set_elem {
@@ -238,6 +239,10 @@ struct nft_set_elem {
 		u32		buf[NFT_DATA_VALUE_MAXLEN / sizeof(u32)];
 		struct nft_data	val;
 	} key;
+	union {
+		u32		buf[NFT_DATA_VALUE_MAXLEN / sizeof(u32)];
+		struct nft_data	val;
+	} key_end;
 	void			*priv;
 };
 
@@ -502,6 +507,7 @@ void nf_tables_destroy_set(const struct nft_ctx *ctx, struct nft_set *set);
  *	enum nft_set_extensions - set extension type IDs
  *
  *	@NFT_SET_EXT_KEY: element key
+ *	@NFT_SET_EXT_KEY_END: upper bound element key, for ranges
  *	@NFT_SET_EXT_DATA: mapping data
  *	@NFT_SET_EXT_FLAGS: element flags
  *	@NFT_SET_EXT_TIMEOUT: element timeout
@@ -513,6 +519,7 @@ void nf_tables_destroy_set(const struct nft_ctx *ctx, struct nft_set *set);
  */
 enum nft_set_extensions {
 	NFT_SET_EXT_KEY,
+	NFT_SET_EXT_KEY_END,
 	NFT_SET_EXT_DATA,
 	NFT_SET_EXT_FLAGS,
 	NFT_SET_EXT_TIMEOUT,
@@ -606,6 +613,11 @@ static inline struct nft_data *nft_set_ext_key(const struct nft_set_ext *ext)
 	return nft_set_ext(ext, NFT_SET_EXT_KEY);
 }
 
+static inline struct nft_data *nft_set_ext_key_end(const struct nft_set_ext *ext)
+{
+	return nft_set_ext(ext, NFT_SET_EXT_KEY_END);
+}
+
 static inline struct nft_data *nft_set_ext_data(const struct nft_set_ext *ext)
 {
 	return nft_set_ext(ext, NFT_SET_EXT_DATA);
@@ -655,7 +667,7 @@ static inline struct nft_object **nft_set_ext_obj(const struct nft_set_ext *ext)
 
 void *nft_set_elem_init(const struct nft_set *set,
 			const struct nft_set_ext_tmpl *tmpl,
-			const u32 *key, const u32 *data,
+			const u32 *key, const u32 *key_end, const u32 *data,
 			u64 timeout, u64 expiration, gfp_t gfp);
 void nft_set_elem_destroy(const struct nft_set *set, void *elem,
 			  bool destroy_expr);
diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h
index 261864736b26..c13106496bd2 100644
--- a/include/uapi/linux/netfilter/nf_tables.h
+++ b/include/uapi/linux/netfilter/nf_tables.h
@@ -370,6 +370,7 @@ enum nft_set_elem_flags {
  * @NFTA_SET_ELEM_USERDATA: user data (NLA_BINARY)
  * @NFTA_SET_ELEM_EXPR: expression (NLA_NESTED: nft_expr_attributes)
  * @NFTA_SET_ELEM_OBJREF: stateful object reference (NLA_STRING)
+ * @NFTA_SET_ELEM_KEY_END: closing key value (NLA_NESTED: nft_data)
  */
 enum nft_set_elem_attributes {
 	NFTA_SET_ELEM_UNSPEC,
@@ -382,6 +383,7 @@ enum nft_set_elem_attributes {
 	NFTA_SET_ELEM_EXPR,
 	NFTA_SET_ELEM_PAD,
 	NFTA_SET_ELEM_OBJREF,
+	NFTA_SET_ELEM_KEY_END,
 	__NFTA_SET_ELEM_MAX
 };
 #define NFTA_SET_ELEM_MAX	(__NFTA_SET_ELEM_MAX - 1)
diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index 58e3b285a3d1..5f645a85538a 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -4215,6 +4215,9 @@ const struct nft_set_ext_type nft_set_ext_types[] = {
 		.len	= sizeof(struct nft_userdata),
 		.align	= __alignof__(struct nft_userdata),
 	},
+	[NFT_SET_EXT_KEY_END]		= {
+		.align	= __alignof__(u32),
+	},
 };
 EXPORT_SYMBOL_GPL(nft_set_ext_types);
 
@@ -4233,6 +4236,7 @@ static const struct nla_policy nft_set_elem_policy[NFTA_SET_ELEM_MAX + 1] = {
 	[NFTA_SET_ELEM_EXPR]		= { .type = NLA_NESTED },
 	[NFTA_SET_ELEM_OBJREF]		= { .type = NLA_STRING,
 					    .len = NFT_OBJ_MAXNAMELEN - 1 },
+	[NFTA_SET_ELEM_KEY_END]		= { .type = NLA_NESTED },
 };
 
 static const struct nla_policy nft_set_elem_list_policy[NFTA_SET_ELEM_LIST_MAX + 1] = {
@@ -4282,6 +4286,11 @@ static int nf_tables_fill_setelem(struct sk_buff *skb,
 			  NFT_DATA_VALUE, set->klen) < 0)
 		goto nla_put_failure;
 
+	if (nft_set_ext_exists(ext, NFT_SET_EXT_KEY_END) &&
+	    nft_data_dump(skb, NFTA_SET_ELEM_KEY_END, nft_set_ext_key_end(ext),
+			  NFT_DATA_VALUE, set->klen) < 0)
+		goto nla_put_failure;
+
 	if (nft_set_ext_exists(ext, NFT_SET_EXT_DATA) &&
 	    nft_data_dump(skb, NFTA_SET_ELEM_DATA, nft_set_ext_data(ext),
 			  set->dtype == NFT_DATA_VERDICT ? NFT_DATA_VERDICT : NFT_DATA_VALUE,
@@ -4569,6 +4578,13 @@ static int nft_get_set_elem(struct nft_ctx *ctx, struct nft_set *set,
 	if (err < 0)
 		return err;
 
+	if (nla[NFTA_SET_ELEM_KEY_END]) {
+		err = nft_setelem_parse_key(ctx, set, &elem.key_end.val,
+					    nla[NFTA_SET_ELEM_KEY_END]);
+		if (err < 0)
+			return err;
+	}
+
 	priv = set->ops->get(ctx->net, set, &elem, flags);
 	if (IS_ERR(priv))
 		return PTR_ERR(priv);
@@ -4694,8 +4710,8 @@ static struct nft_trans *nft_trans_elem_alloc(struct nft_ctx *ctx,
 
 void *nft_set_elem_init(const struct nft_set *set,
 			const struct nft_set_ext_tmpl *tmpl,
-			const u32 *key, const u32 *data,
-			u64 timeout, u64 expiration, gfp_t gfp)
+			const u32 *key, const u32 *key_end,
+			const u32 *data, u64 timeout, u64 expiration, gfp_t gfp)
 {
 	struct nft_set_ext *ext;
 	void *elem;
@@ -4708,6 +4724,8 @@ void *nft_set_elem_init(const struct nft_set *set,
 	nft_set_ext_init(ext, tmpl);
 
 	memcpy(nft_set_ext_key(ext), key, set->klen);
+	if (nft_set_ext_exists(ext, NFT_SET_EXT_KEY_END))
+		memcpy(nft_set_ext_key_end(ext), key_end, set->klen);
 	if (nft_set_ext_exists(ext, NFT_SET_EXT_DATA))
 		memcpy(nft_set_ext_data(ext), data, set->dlen);
 	if (nft_set_ext_exists(ext, NFT_SET_EXT_EXPIRATION)) {
@@ -4842,9 +4860,19 @@ static int nft_add_set_elem(struct nft_ctx *ctx, struct nft_set *set,
 	err = nft_setelem_parse_key(ctx, set, &elem.key.val,
 				    nla[NFTA_SET_ELEM_KEY]);
 	if (err < 0)
-		goto err1;
+		return err;
 
 	nft_set_ext_add_length(&tmpl, NFT_SET_EXT_KEY, set->klen);
+
+	if (nla[NFTA_SET_ELEM_KEY_END]) {
+		err = nft_setelem_parse_key(ctx, set, &elem.key_end.val,
+					    nla[NFTA_SET_ELEM_KEY_END]);
+		if (err < 0)
+			goto err_parse_key;
+
+		nft_set_ext_add_length(&tmpl, NFT_SET_EXT_KEY_END, set->klen);
+	}
+
 	if (timeout > 0) {
 		nft_set_ext_add(&tmpl, NFT_SET_EXT_EXPIRATION);
 		if (timeout != set->timeout)
@@ -4854,14 +4882,14 @@ static int nft_add_set_elem(struct nft_ctx *ctx, struct nft_set *set,
 	if (nla[NFTA_SET_ELEM_OBJREF] != NULL) {
 		if (!(set->flags & NFT_SET_OBJECT)) {
 			err = -EINVAL;
-			goto err2;
+			goto err_parse_key_end;
 		}
 		obj = nft_obj_lookup(ctx->net, ctx->table,
 				     nla[NFTA_SET_ELEM_OBJREF],
 				     set->objtype, genmask);
 		if (IS_ERR(obj)) {
 			err = PTR_ERR(obj);
-			goto err2;
+			goto err_parse_key_end;
 		}
 		nft_set_ext_add(&tmpl, NFT_SET_EXT_OBJREF);
 	}
@@ -4870,11 +4898,11 @@ static int nft_add_set_elem(struct nft_ctx *ctx, struct nft_set *set,
 		err = nft_data_init(ctx, &data, sizeof(data), &desc,
 				    nla[NFTA_SET_ELEM_DATA]);
 		if (err < 0)
-			goto err2;
+			goto err_parse_key_end;
 
 		err = -EINVAL;
 		if (set->dtype != NFT_DATA_VERDICT && desc.len != set->dlen)
-			goto err3;
+			goto err_parse_data;
 
 		dreg = nft_type_to_reg(set->dtype);
 		list_for_each_entry(binding, &set->bindings, list) {
@@ -4892,7 +4920,7 @@ static int nft_add_set_elem(struct nft_ctx *ctx, struct nft_set *set,
 							  &data,
 							  desc.type, desc.len);
 			if (err < 0)
-				goto err3;
+				goto err_parse_data;
 
 			if (desc.type == NFT_DATA_VERDICT &&
 			    (data.verdict.code == NFT_GOTO ||
@@ -4917,10 +4945,11 @@ static int nft_add_set_elem(struct nft_ctx *ctx, struct nft_set *set,
 	}
 
 	err = -ENOMEM;
-	elem.priv = nft_set_elem_init(set, &tmpl, elem.key.val.data, data.data,
+	elem.priv = nft_set_elem_init(set, &tmpl, elem.key.val.data,
+				      elem.key_end.val.data, data.data,
 				      timeout, expiration, GFP_KERNEL);
 	if (elem.priv == NULL)
-		goto err3;
+		goto err_parse_data;
 
 	ext = nft_set_elem_ext(set, elem.priv);
 	if (flags)
@@ -4937,7 +4966,7 @@ static int nft_add_set_elem(struct nft_ctx *ctx, struct nft_set *set,
 
 	trans = nft_trans_elem_alloc(ctx, NFT_MSG_NEWSETELEM, set);
 	if (trans == NULL)
-		goto err4;
+		goto err_trans;
 
 	ext->genmask = nft_genmask_cur(ctx->net) | NFT_SET_ELEM_BUSY_MASK;
 	err = set->ops->insert(ctx->net, set, &elem, &ext2);
@@ -4948,7 +4977,7 @@ static int nft_add_set_elem(struct nft_ctx *ctx, struct nft_set *set,
 			    nft_set_ext_exists(ext, NFT_SET_EXT_OBJREF) ^
 			    nft_set_ext_exists(ext2, NFT_SET_EXT_OBJREF)) {
 				err = -EBUSY;
-				goto err5;
+				goto err_element_clash;
 			}
 			if ((nft_set_ext_exists(ext, NFT_SET_EXT_DATA) &&
 			     nft_set_ext_exists(ext2, NFT_SET_EXT_DATA) &&
@@ -4961,33 +4990,35 @@ static int nft_add_set_elem(struct nft_ctx *ctx, struct nft_set *set,
 			else if (!(nlmsg_flags & NLM_F_EXCL))
 				err = 0;
 		}
-		goto err5;
+		goto err_element_clash;
 	}
 
 	if (set->size &&
 	    !atomic_add_unless(&set->nelems, 1, set->size + set->ndeact)) {
 		err = -ENFILE;
-		goto err6;
+		goto err_set_full;
 	}
 
 	nft_trans_elem(trans) = elem;
 	list_add_tail(&trans->list, &ctx->net->nft.commit_list);
 	return 0;
 
-err6:
+err_set_full:
 	set->ops->remove(ctx->net, set, &elem);
-err5:
+err_element_clash:
 	kfree(trans);
-err4:
+err_trans:
 	if (obj)
 		obj->use--;
 	kfree(elem.priv);
-err3:
+err_parse_data:
 	if (nla[NFTA_SET_ELEM_DATA] != NULL)
 		nft_data_release(&data, desc.type);
-err2:
+err_parse_key_end:
+	nft_data_release(&elem.key_end.val, NFT_DATA_VALUE);
+err_parse_key:
 	nft_data_release(&elem.key.val, NFT_DATA_VALUE);
-err1:
+
 	return err;
 }
 
@@ -5112,9 +5143,19 @@ static int nft_del_setelem(struct nft_ctx *ctx, struct nft_set *set,
 
 	nft_set_ext_add_length(&tmpl, NFT_SET_EXT_KEY, set->klen);
 
+	if (nla[NFTA_SET_ELEM_KEY_END]) {
+		err = nft_setelem_parse_key(ctx, set, &elem.key_end.val,
+					    nla[NFTA_SET_ELEM_KEY_END]);
+		if (err < 0)
+			return err;
+
+		nft_set_ext_add_length(&tmpl, NFT_SET_EXT_KEY_END, set->klen);
+	}
+
 	err = -ENOMEM;
-	elem.priv = nft_set_elem_init(set, &tmpl, elem.key.val.data, NULL, 0,
-				      0, GFP_KERNEL);
+	elem.priv = nft_set_elem_init(set, &tmpl, elem.key.val.data,
+				      elem.key_end.val.data, NULL, 0, 0,
+				      GFP_KERNEL);
 	if (elem.priv == NULL)
 		goto fail_elem;
 
diff --git a/net/netfilter/nft_dynset.c b/net/netfilter/nft_dynset.c
index 8887295414dc..683785225a3e 100644
--- a/net/netfilter/nft_dynset.c
+++ b/net/netfilter/nft_dynset.c
@@ -54,7 +54,7 @@ static void *nft_dynset_new(struct nft_set *set, const struct nft_expr *expr,
 
 	timeout = priv->timeout ? : set->timeout;
 	elem = nft_set_elem_init(set, &priv->tmpl,
-				 &regs->data[priv->sreg_key],
+				 &regs->data[priv->sreg_key], NULL,
 				 &regs->data[priv->sreg_data],
 				 timeout, 0, GFP_ATOMIC);
 	if (elem == NULL)
-- 
cgit v1.2.3-59-g8ed1b


From f3a2181e16f1dcbf5446ed43f6b5d9f56c459f85 Mon Sep 17 00:00:00 2001
From: Stefano Brivio <sbrivio@redhat.com>
Date: Wed, 22 Jan 2020 00:17:53 +0100
Subject: netfilter: nf_tables: Support for sets with multiple ranged fields

Introduce a new nested netlink attribute, NFTA_SET_DESC_CONCAT, used
to specify the length of each field in a set concatenation.

This allows set implementations to support concatenation of multiple
ranged items, as they can divide the input key into matching data for
every single field. Such set implementations would be selected as
they specify support for NFT_SET_INTERVAL and allow desc->field_count
to be greater than one. Explicitly disallow this for nft_set_rbtree.

In order to specify the interval for a set entry, userspace would
include in NFTA_SET_DESC_CONCAT attributes field lengths, and pass
range endpoints as two separate keys, represented by attributes
NFTA_SET_ELEM_KEY and NFTA_SET_ELEM_KEY_END.

While at it, export the number of 32-bit registers available for
packet matching, as nftables will need this to know the maximum
number of field lengths that can be specified.

For example, "packets with an IPv4 address between 192.0.2.0 and
192.0.2.42, with destination port between 22 and 25", can be
expressed as two concatenated elements:

  NFTA_SET_ELEM_KEY:            192.0.2.0 . 22
  NFTA_SET_ELEM_KEY_END:        192.0.2.42 . 25

and NFTA_SET_DESC_CONCAT attribute would contain:

  NFTA_LIST_ELEM
    NFTA_SET_FIELD_LEN:		4
  NFTA_LIST_ELEM
    NFTA_SET_FIELD_LEN:		2

v4: No changes
v3: Complete rework, NFTA_SET_DESC_CONCAT instead of NFTA_SET_SUBKEY
v2: No changes

Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_tables.h        |  8 +++
 include/uapi/linux/netfilter/nf_tables.h | 15 ++++++
 net/netfilter/nf_tables_api.c            | 90 +++++++++++++++++++++++++++++++-
 net/netfilter/nft_set_rbtree.c           |  3 ++
 4 files changed, 115 insertions(+), 1 deletion(-)

diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h
index 504c0aa93805..4170c033d461 100644
--- a/include/net/netfilter/nf_tables.h
+++ b/include/net/netfilter/nf_tables.h
@@ -264,11 +264,15 @@ struct nft_set_iter {
  *	@klen: key length
  *	@dlen: data length
  *	@size: number of set elements
+ *	@field_len: length of each field in concatenation, bytes
+ *	@field_count: number of concatenated fields in element
  */
 struct nft_set_desc {
 	unsigned int		klen;
 	unsigned int		dlen;
 	unsigned int		size;
+	u8			field_len[NFT_REG32_COUNT];
+	u8			field_count;
 };
 
 /**
@@ -409,6 +413,8 @@ void nft_unregister_set(struct nft_set_type *type);
  * 	@dtype: data type (verdict or numeric type defined by userspace)
  * 	@objtype: object type (see NFT_OBJECT_* definitions)
  * 	@size: maximum set size
+ *	@field_len: length of each field in concatenation, bytes
+ *	@field_count: number of concatenated fields in element
  *	@use: number of rules references to this set
  * 	@nelems: number of elements
  * 	@ndeact: number of deactivated elements queued for removal
@@ -435,6 +441,8 @@ struct nft_set {
 	u32				dtype;
 	u32				objtype;
 	u32				size;
+	u8				field_len[NFT_REG32_COUNT];
+	u8				field_count;
 	u32				use;
 	atomic_t			nelems;
 	u32				ndeact;
diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h
index c13106496bd2..065218a20bb7 100644
--- a/include/uapi/linux/netfilter/nf_tables.h
+++ b/include/uapi/linux/netfilter/nf_tables.h
@@ -48,6 +48,7 @@ enum nft_registers {
 
 #define NFT_REG_SIZE	16
 #define NFT_REG32_SIZE	4
+#define NFT_REG32_COUNT	(NFT_REG32_15 - NFT_REG32_00 + 1)
 
 /**
  * enum nft_verdicts - nf_tables internal verdicts
@@ -301,14 +302,28 @@ enum nft_set_policies {
  * enum nft_set_desc_attributes - set element description
  *
  * @NFTA_SET_DESC_SIZE: number of elements in set (NLA_U32)
+ * @NFTA_SET_DESC_CONCAT: description of field concatenation (NLA_NESTED)
  */
 enum nft_set_desc_attributes {
 	NFTA_SET_DESC_UNSPEC,
 	NFTA_SET_DESC_SIZE,
+	NFTA_SET_DESC_CONCAT,
 	__NFTA_SET_DESC_MAX
 };
 #define NFTA_SET_DESC_MAX	(__NFTA_SET_DESC_MAX - 1)
 
+/**
+ * enum nft_set_field_attributes - attributes of concatenated fields
+ *
+ * @NFTA_SET_FIELD_LEN: length of single field, in bits (NLA_U32)
+ */
+enum nft_set_field_attributes {
+	NFTA_SET_FIELD_UNSPEC,
+	NFTA_SET_FIELD_LEN,
+	__NFTA_SET_FIELD_MAX
+};
+#define NFTA_SET_FIELD_MAX	(__NFTA_SET_FIELD_MAX - 1)
+
 /**
  * enum nft_set_attributes - nf_tables set netlink attributes
  *
diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index 5f645a85538a..d1318bdf49ca 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -3391,6 +3391,7 @@ static const struct nla_policy nft_set_policy[NFTA_SET_MAX + 1] = {
 
 static const struct nla_policy nft_set_desc_policy[NFTA_SET_DESC_MAX + 1] = {
 	[NFTA_SET_DESC_SIZE]		= { .type = NLA_U32 },
+	[NFTA_SET_DESC_CONCAT]		= { .type = NLA_NESTED },
 };
 
 static int nft_ctx_init_from_setattr(struct nft_ctx *ctx, struct net *net,
@@ -3557,6 +3558,33 @@ static __be64 nf_jiffies64_to_msecs(u64 input)
 	return cpu_to_be64(jiffies64_to_msecs(input));
 }
 
+static int nf_tables_fill_set_concat(struct sk_buff *skb,
+				     const struct nft_set *set)
+{
+	struct nlattr *concat, *field;
+	int i;
+
+	concat = nla_nest_start_noflag(skb, NFTA_SET_DESC_CONCAT);
+	if (!concat)
+		return -ENOMEM;
+
+	for (i = 0; i < set->field_count; i++) {
+		field = nla_nest_start_noflag(skb, NFTA_LIST_ELEM);
+		if (!field)
+			return -ENOMEM;
+
+		if (nla_put_be32(skb, NFTA_SET_FIELD_LEN,
+				 htonl(set->field_len[i])))
+			return -ENOMEM;
+
+		nla_nest_end(skb, field);
+	}
+
+	nla_nest_end(skb, concat);
+
+	return 0;
+}
+
 static int nf_tables_fill_set(struct sk_buff *skb, const struct nft_ctx *ctx,
 			      const struct nft_set *set, u16 event, u16 flags)
 {
@@ -3620,11 +3648,17 @@ static int nf_tables_fill_set(struct sk_buff *skb, const struct nft_ctx *ctx,
 		goto nla_put_failure;
 
 	desc = nla_nest_start_noflag(skb, NFTA_SET_DESC);
+
 	if (desc == NULL)
 		goto nla_put_failure;
 	if (set->size &&
 	    nla_put_be32(skb, NFTA_SET_DESC_SIZE, htonl(set->size)))
 		goto nla_put_failure;
+
+	if (set->field_count > 1 &&
+	    nf_tables_fill_set_concat(skb, set))
+		goto nla_put_failure;
+
 	nla_nest_end(skb, desc);
 
 	nlmsg_end(skb, nlh);
@@ -3797,6 +3831,53 @@ err:
 	return err;
 }
 
+static const struct nla_policy nft_concat_policy[NFTA_SET_FIELD_MAX + 1] = {
+	[NFTA_SET_FIELD_LEN]	= { .type = NLA_U32 },
+};
+
+static int nft_set_desc_concat_parse(const struct nlattr *attr,
+				     struct nft_set_desc *desc)
+{
+	struct nlattr *tb[NFTA_SET_FIELD_MAX + 1];
+	u32 len;
+	int err;
+
+	err = nla_parse_nested_deprecated(tb, NFTA_SET_FIELD_MAX, attr,
+					  nft_concat_policy, NULL);
+	if (err < 0)
+		return err;
+
+	if (!tb[NFTA_SET_FIELD_LEN])
+		return -EINVAL;
+
+	len = ntohl(nla_get_be32(tb[NFTA_SET_FIELD_LEN]));
+
+	if (len * BITS_PER_BYTE / 32 > NFT_REG32_COUNT)
+		return -E2BIG;
+
+	desc->field_len[desc->field_count++] = len;
+
+	return 0;
+}
+
+static int nft_set_desc_concat(struct nft_set_desc *desc,
+			       const struct nlattr *nla)
+{
+	struct nlattr *attr;
+	int rem, err;
+
+	nla_for_each_nested(attr, nla, rem) {
+		if (nla_type(attr) != NFTA_LIST_ELEM)
+			return -EINVAL;
+
+		err = nft_set_desc_concat_parse(attr, desc);
+		if (err < 0)
+			return err;
+	}
+
+	return 0;
+}
+
 static int nf_tables_set_desc_parse(struct nft_set_desc *desc,
 				    const struct nlattr *nla)
 {
@@ -3810,8 +3891,10 @@ static int nf_tables_set_desc_parse(struct nft_set_desc *desc,
 
 	if (da[NFTA_SET_DESC_SIZE] != NULL)
 		desc->size = ntohl(nla_get_be32(da[NFTA_SET_DESC_SIZE]));
+	if (da[NFTA_SET_DESC_CONCAT])
+		err = nft_set_desc_concat(desc, da[NFTA_SET_DESC_CONCAT]);
 
-	return 0;
+	return err;
 }
 
 static int nf_tables_newset(struct net *net, struct sock *nlsk,
@@ -3834,6 +3917,7 @@ static int nf_tables_newset(struct net *net, struct sock *nlsk,
 	unsigned char *udata;
 	u16 udlen;
 	int err;
+	int i;
 
 	if (nla[NFTA_SET_TABLE] == NULL ||
 	    nla[NFTA_SET_NAME] == NULL ||
@@ -4012,6 +4096,10 @@ static int nf_tables_newset(struct net *net, struct sock *nlsk,
 	set->gc_int = gc_int;
 	set->handle = nf_tables_alloc_handle(table);
 
+	set->field_count = desc.field_count;
+	for (i = 0; i < desc.field_count; i++)
+		set->field_len[i] = desc.field_len[i];
+
 	err = ops->init(set, &desc, nla);
 	if (err < 0)
 		goto err3;
diff --git a/net/netfilter/nft_set_rbtree.c b/net/netfilter/nft_set_rbtree.c
index a9f804f7a04a..5000b938ab1e 100644
--- a/net/netfilter/nft_set_rbtree.c
+++ b/net/netfilter/nft_set_rbtree.c
@@ -466,6 +466,9 @@ static void nft_rbtree_destroy(const struct nft_set *set)
 static bool nft_rbtree_estimate(const struct nft_set_desc *desc, u32 features,
 				struct nft_set_estimate *est)
 {
+	if (desc->field_count > 1)
+		return false;
+
 	if (desc->size)
 		est->size = sizeof(struct nft_rbtree) +
 			    desc->size * sizeof(struct nft_rbtree_elem);
-- 
cgit v1.2.3-59-g8ed1b


From 2092767168f0681aa03727448b801600a364c013 Mon Sep 17 00:00:00 2001
From: Stefano Brivio <sbrivio@redhat.com>
Date: Wed, 22 Jan 2020 00:17:54 +0100
Subject: bitmap: Introduce bitmap_cut(): cut bits and shift remaining

The new bitmap function bitmap_cut() copies bits from source to
destination by removing the region specified by parameters first
and cut, and remapping the bits above the cut region by right
shifting them.

Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/linux/bitmap.h |  4 +++
 lib/bitmap.c           | 66 ++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 70 insertions(+)

diff --git a/include/linux/bitmap.h b/include/linux/bitmap.h
index ff335b22f23c..f0f3a9fffa6a 100644
--- a/include/linux/bitmap.h
+++ b/include/linux/bitmap.h
@@ -53,6 +53,7 @@
  *  bitmap_find_next_zero_area_off(buf, len, pos, n, mask)  as above
  *  bitmap_shift_right(dst, src, n, nbits)      *dst = *src >> n
  *  bitmap_shift_left(dst, src, n, nbits)       *dst = *src << n
+ *  bitmap_cut(dst, src, first, n, nbits)       Cut n bits from first, copy rest
  *  bitmap_replace(dst, old, new, mask, nbits)  *dst = (*old & ~(*mask)) | (*new & *mask)
  *  bitmap_remap(dst, src, old, new, nbits)     *dst = map(old, new)(src)
  *  bitmap_bitremap(oldbit, old, new, nbits)    newbit = map(old, new)(oldbit)
@@ -133,6 +134,9 @@ extern void __bitmap_shift_right(unsigned long *dst, const unsigned long *src,
 				unsigned int shift, unsigned int nbits);
 extern void __bitmap_shift_left(unsigned long *dst, const unsigned long *src,
 				unsigned int shift, unsigned int nbits);
+extern void bitmap_cut(unsigned long *dst, const unsigned long *src,
+		       unsigned int first, unsigned int cut,
+		       unsigned int nbits);
 extern int __bitmap_and(unsigned long *dst, const unsigned long *bitmap1,
 			const unsigned long *bitmap2, unsigned int nbits);
 extern void __bitmap_or(unsigned long *dst, const unsigned long *bitmap1,
diff --git a/lib/bitmap.c b/lib/bitmap.c
index 4250519d7d1c..6e175fbd69a9 100644
--- a/lib/bitmap.c
+++ b/lib/bitmap.c
@@ -168,6 +168,72 @@ void __bitmap_shift_left(unsigned long *dst, const unsigned long *src,
 }
 EXPORT_SYMBOL(__bitmap_shift_left);
 
+/**
+ * bitmap_cut() - remove bit region from bitmap and right shift remaining bits
+ * @dst: destination bitmap, might overlap with src
+ * @src: source bitmap
+ * @first: start bit of region to be removed
+ * @cut: number of bits to remove
+ * @nbits: bitmap size, in bits
+ *
+ * Set the n-th bit of @dst iff the n-th bit of @src is set and
+ * n is less than @first, or the m-th bit of @src is set for any
+ * m such that @first <= n < nbits, and m = n + @cut.
+ *
+ * In pictures, example for a big-endian 32-bit architecture:
+ *
+ * @src:
+ * 31                                   63
+ * |                                    |
+ * 10000000 11000001 11110010 00010101  10000000 11000001 01110010 00010101
+ *                 |  |              |                                    |
+ *                16  14             0                                   32
+ *
+ * if @cut is 3, and @first is 14, bits 14-16 in @src are cut and @dst is:
+ *
+ * 31                                   63
+ * |                                    |
+ * 10110000 00011000 00110010 00010101  00010000 00011000 00101110 01000010
+ *                    |              |                                    |
+ *                    14 (bit 17     0                                   32
+ *                        from @src)
+ *
+ * Note that @dst and @src might overlap partially or entirely.
+ *
+ * This is implemented in the obvious way, with a shift and carry
+ * step for each moved bit. Optimisation is left as an exercise
+ * for the compiler.
+ */
+void bitmap_cut(unsigned long *dst, const unsigned long *src,
+		unsigned int first, unsigned int cut, unsigned int nbits)
+{
+	unsigned int len = BITS_TO_LONGS(nbits);
+	unsigned long keep = 0, carry;
+	int i;
+
+	memmove(dst, src, len * sizeof(*dst));
+
+	if (first % BITS_PER_LONG) {
+		keep = src[first / BITS_PER_LONG] &
+		       (~0UL >> (BITS_PER_LONG - first % BITS_PER_LONG));
+	}
+
+	while (cut--) {
+		for (i = first / BITS_PER_LONG; i < len; i++) {
+			if (i < len - 1)
+				carry = dst[i + 1] & 1UL;
+			else
+				carry = 0;
+
+			dst[i] = (dst[i] >> 1) | (carry << (BITS_PER_LONG - 1));
+		}
+	}
+
+	dst[first / BITS_PER_LONG] &= ~0UL << (first % BITS_PER_LONG);
+	dst[first / BITS_PER_LONG] |= keep;
+}
+EXPORT_SYMBOL(bitmap_cut);
+
 int __bitmap_and(unsigned long *dst, const unsigned long *bitmap1,
 				const unsigned long *bitmap2, unsigned int bits)
 {
-- 
cgit v1.2.3-59-g8ed1b


From 3c4287f62044a90e73a561aa05fc46e62da173da Mon Sep 17 00:00:00 2001
From: Stefano Brivio <sbrivio@redhat.com>
Date: Wed, 22 Jan 2020 00:17:55 +0100
Subject: nf_tables: Add set type for arbitrary concatenation of ranges

This new set type allows for intervals in concatenated fields,
which are expressed in the usual way, that is, simple byte
concatenation with padding to 32 bits for single fields, and
given as ranges by specifying start and end elements containing,
each, the full concatenation of start and end values for the
single fields.

Ranges are expanded to composing netmasks, for each field: these
are inserted as rules in per-field lookup tables. Bits to be
classified are divided in 4-bit groups, and for each group, the
lookup table contains 4^2 buckets, representing all the possible
values of a bit group. This approach was inspired by the Grouper
algorithm:
	http://www.cse.usf.edu/~ligatti/projects/grouper/

Matching is performed by a sequence of AND operations between
bucket values, with buckets selected according to the value of
packet bits, for each group. The result of this sequence tells
us which rules matched for a given field.

In order to concatenate several ranged fields, per-field rules
are mapped using mapping arrays, one per field, that specify
which rules should be considered while matching the next field.
The mapping array for the last field contains a reference to
the element originally inserted.

The notes in nft_set_pipapo.c cover the algorithm in deeper
detail.

A pure hash-based approach is of no use here, as ranges need
to be classified. An implementation based on "proxying" the
existing red-black tree set type, creating a tree for each
field, was considered, but deemed impractical due to the fact
that elements would need to be shared between trees, at least
as long as we want to keep UAPI changes to a minimum.

A stand-alone implementation of this algorithm is available at:
	https://pipapo.lameexcu.se
together with notes about possible future optimisations
(in pipapo.c).

This algorithm was designed with data locality in mind, and can
be highly optimised for SIMD instruction sets, as the bulk of
the matching work is done with repetitive, simple bitwise
operations.

At this point, without further optimisations, nft_concat_range.sh
reports, for one AMD Epyc 7351 thread (2.9GHz, 512 KiB L1D$, 8 MiB
L2$):

TEST: performance
  net,port                                                      [ OK ]
    baseline (drop from netdev hook):              10190076pps
    baseline hash (non-ranged entries):             6179564pps
    baseline rbtree (match on first field only):    2950341pps
    set with  1000 full, ranged entries:            2304165pps
  port,net                                                      [ OK ]
    baseline (drop from netdev hook):              10143615pps
    baseline hash (non-ranged entries):             6135776pps
    baseline rbtree (match on first field only):    4311934pps
    set with   100 full, ranged entries:            4131471pps
  net6,port                                                     [ OK ]
    baseline (drop from netdev hook):               9730404pps
    baseline hash (non-ranged entries):             4809557pps
    baseline rbtree (match on first field only):    1501699pps
    set with  1000 full, ranged entries:            1092557pps
  port,proto                                                    [ OK ]
    baseline (drop from netdev hook):              10812426pps
    baseline hash (non-ranged entries):             6929353pps
    baseline rbtree (match on first field only):    3027105pps
    set with 30000 full, ranged entries:             284147pps
  net6,port,mac                                                 [ OK ]
    baseline (drop from netdev hook):               9660114pps
    baseline hash (non-ranged entries):             3778877pps
    baseline rbtree (match on first field only):    3179379pps
    set with    10 full, ranged entries:            2082880pps
  net6,port,mac,proto                                           [ OK ]
    baseline (drop from netdev hook):               9718324pps
    baseline hash (non-ranged entries):             3799021pps
    baseline rbtree (match on first field only):    1506689pps
    set with  1000 full, ranged entries:             783810pps
  net,mac                                                       [ OK ]
    baseline (drop from netdev hook):              10190029pps
    baseline hash (non-ranged entries):             5172218pps
    baseline rbtree (match on first field only):    2946863pps
    set with  1000 full, ranged entries:            1279122pps

v4:
 - fix build for 32-bit architectures: 64-bit division needs
   div_u64() (kbuild test robot <lkp@intel.com>)
v3:
 - rework interface for field length specification,
   NFT_SET_SUBKEY disappears and information is stored in
   description
 - remove scratch area to store closing element of ranges,
   as elements now come with an actual attribute to specify
   the upper range limit (Pablo Neira Ayuso)
 - also remove pointer to 'start' element from mapping table,
   closing key is now accessible via extension data
 - use bytes right away instead of bits for field lengths,
   this way we can also double the inner loop of the lookup
   function to take care of upper and lower bits in a single
   iteration (minor performance improvement)
 - make it clearer that set operations are actually atomic
   API-wise, but we can't e.g. implement flush() as one-shot
   action
 - fix type for 'dup' in nft_pipapo_insert(), check for
   duplicates only in the next generation, and in general take
   care of differentiating generation mask cases depending on
   the operation (Pablo Neira Ayuso)
 - report C implementation matching rate in commit message, so
   that AVX2 implementation can be compared (Pablo Neira Ayuso)
v2:
 - protect access to scratch maps in nft_pipapo_lookup() with
   local_bh_disable/enable() (Florian Westphal)
 - drop rcu_read_lock/unlock() from nft_pipapo_lookup(), it's
   already implied (Florian Westphal)
 - explain why partial allocation failures don't need handling
   in pipapo_realloc_scratch(), rename 'm' to clone and update
   related kerneldoc to make it clear we're not operating on
   the live copy (Florian Westphal)
 - add expicit check for priv->start_elem in
   nft_pipapo_insert() to avoid ending up in nft_pipapo_walk()
   with a NULL start element, and also zero it out in every
   operation that might make it invalid, so that insertion
   doesn't proceed with an invalid element (Florian Westphal)

Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_tables_core.h |    1 +
 net/netfilter/Makefile                 |    3 +-
 net/netfilter/nf_tables_set_core.c     |    2 +
 net/netfilter/nft_set_pipapo.c         | 2102 ++++++++++++++++++++++++++++++++
 4 files changed, 2107 insertions(+), 1 deletion(-)
 create mode 100644 net/netfilter/nft_set_pipapo.c

diff --git a/include/net/netfilter/nf_tables_core.h b/include/net/netfilter/nf_tables_core.h
index 2656155b4069..29e7e1021267 100644
--- a/include/net/netfilter/nf_tables_core.h
+++ b/include/net/netfilter/nf_tables_core.h
@@ -74,6 +74,7 @@ extern struct nft_set_type nft_set_hash_type;
 extern struct nft_set_type nft_set_hash_fast_type;
 extern struct nft_set_type nft_set_rbtree_type;
 extern struct nft_set_type nft_set_bitmap_type;
+extern struct nft_set_type nft_set_pipapo_type;
 
 struct nft_expr;
 struct nft_regs;
diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile
index 5e9b2eb24349..3f572e5a975e 100644
--- a/net/netfilter/Makefile
+++ b/net/netfilter/Makefile
@@ -81,7 +81,8 @@ nf_tables-objs := nf_tables_core.o nf_tables_api.o nft_chain_filter.o \
 		  nft_chain_route.o nf_tables_offload.o
 
 nf_tables_set-objs := nf_tables_set_core.o \
-		      nft_set_hash.o nft_set_bitmap.o nft_set_rbtree.o
+		      nft_set_hash.o nft_set_bitmap.o nft_set_rbtree.o \
+		      nft_set_pipapo.o
 
 obj-$(CONFIG_NF_TABLES)		+= nf_tables.o
 obj-$(CONFIG_NF_TABLES_SET)	+= nf_tables_set.o
diff --git a/net/netfilter/nf_tables_set_core.c b/net/netfilter/nf_tables_set_core.c
index a9fce8d10051..586b621007eb 100644
--- a/net/netfilter/nf_tables_set_core.c
+++ b/net/netfilter/nf_tables_set_core.c
@@ -9,12 +9,14 @@ static int __init nf_tables_set_module_init(void)
 	nft_register_set(&nft_set_rhash_type);
 	nft_register_set(&nft_set_bitmap_type);
 	nft_register_set(&nft_set_rbtree_type);
+	nft_register_set(&nft_set_pipapo_type);
 
 	return 0;
 }
 
 static void __exit nf_tables_set_module_exit(void)
 {
+	nft_unregister_set(&nft_set_pipapo_type);
 	nft_unregister_set(&nft_set_rbtree_type);
 	nft_unregister_set(&nft_set_bitmap_type);
 	nft_unregister_set(&nft_set_rhash_type);
diff --git a/net/netfilter/nft_set_pipapo.c b/net/netfilter/nft_set_pipapo.c
new file mode 100644
index 000000000000..f0cb1e13af50
--- /dev/null
+++ b/net/netfilter/nft_set_pipapo.c
@@ -0,0 +1,2102 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+/* PIPAPO: PIle PAcket POlicies: set for arbitrary concatenations of ranges
+ *
+ * Copyright (c) 2019-2020 Red Hat GmbH
+ *
+ * Author: Stefano Brivio <sbrivio@redhat.com>
+ */
+
+/**
+ * DOC: Theory of Operation
+ *
+ *
+ * Problem
+ * -------
+ *
+ * Match packet bytes against entries composed of ranged or non-ranged packet
+ * field specifiers, mapping them to arbitrary references. For example:
+ *
+ * ::
+ *
+ *               --- fields --->
+ *      |    [net],[port],[net]... => [reference]
+ *   entries [net],[port],[net]... => [reference]
+ *      |    [net],[port],[net]... => [reference]
+ *      V    ...
+ *
+ * where [net] fields can be IP ranges or netmasks, and [port] fields are port
+ * ranges. Arbitrary packet fields can be matched.
+ *
+ *
+ * Algorithm Overview
+ * ------------------
+ *
+ * This algorithm is loosely inspired by [Ligatti 2010], and fundamentally
+ * relies on the consideration that every contiguous range in a space of b bits
+ * can be converted into b * 2 netmasks, from Theorem 3 in [Rottenstreich 2010],
+ * as also illustrated in Section 9 of [Kogan 2014].
+ *
+ * Classification against a number of entries, that require matching given bits
+ * of a packet field, is performed by grouping those bits in sets of arbitrary
+ * size, and classifying packet bits one group at a time.
+ *
+ * Example:
+ *   to match the source port (16 bits) of a packet, we can divide those 16 bits
+ *   in 4 groups of 4 bits each. Given the entry:
+ *      0000 0001 0101 1001
+ *   and a packet with source port:
+ *      0000 0001 1010 1001
+ *   first and second groups match, but the third doesn't. We conclude that the
+ *   packet doesn't match the given entry.
+ *
+ * Translate the set to a sequence of lookup tables, one per field. Each table
+ * has two dimensions: bit groups to be matched for a single packet field, and
+ * all the possible values of said groups (buckets). Input entries are
+ * represented as one or more rules, depending on the number of composing
+ * netmasks for the given field specifier, and a group match is indicated as a
+ * set bit, with number corresponding to the rule index, in all the buckets
+ * whose value matches the entry for a given group.
+ *
+ * Rules are mapped between fields through an array of x, n pairs, with each
+ * item mapping a matched rule to one or more rules. The position of the pair in
+ * the array indicates the matched rule to be mapped to the next field, x
+ * indicates the first rule index in the next field, and n the amount of
+ * next-field rules the current rule maps to.
+ *
+ * The mapping array for the last field maps to the desired references.
+ *
+ * To match, we perform table lookups using the values of grouped packet bits,
+ * and use a sequence of bitwise operations to progressively evaluate rule
+ * matching.
+ *
+ * A stand-alone, reference implementation, also including notes about possible
+ * future optimisations, is available at:
+ *    https://pipapo.lameexcu.se/
+ *
+ * Insertion
+ * ---------
+ *
+ * - For each packet field:
+ *
+ *   - divide the b packet bits we want to classify into groups of size t,
+ *     obtaining ceil(b / t) groups
+ *
+ *      Example: match on destination IP address, with t = 4: 32 bits, 8 groups
+ *      of 4 bits each
+ *
+ *   - allocate a lookup table with one column ("bucket") for each possible
+ *     value of a group, and with one row for each group
+ *
+ *      Example: 8 groups, 2^4 buckets:
+ *
+ * ::
+ *
+ *                     bucket
+ *      group  0   1   2   3   4   5   6   7   8   9  10  11  12  13  14  15
+ *        0
+ *        1
+ *        2
+ *        3
+ *        4
+ *        5
+ *        6
+ *        7
+ *
+ *   - map the bits we want to classify for the current field, for a given
+ *     entry, to a single rule for non-ranged and netmask set items, and to one
+ *     or multiple rules for ranges. Ranges are expanded to composing netmasks
+ *     by pipapo_expand().
+ *
+ *      Example: 2 entries, 10.0.0.5:1024 and 192.168.1.0-192.168.2.1:2048
+ *      - rule #0: 10.0.0.5
+ *      - rule #1: 192.168.1.0/24
+ *      - rule #2: 192.168.2.0/31
+ *
+ *   - insert references to the rules in the lookup table, selecting buckets
+ *     according to bit values of a rule in the given group. This is done by
+ *     pipapo_insert().
+ *
+ *      Example: given:
+ *      - rule #0: 10.0.0.5 mapping to buckets
+ *        < 0 10  0 0   0 0  0 5 >
+ *      - rule #1: 192.168.1.0/24 mapping to buckets
+ *        < 12 0  10 8  0 1  < 0..15 > < 0..15 > >
+ *      - rule #2: 192.168.2.0/31 mapping to buckets
+ *        < 12 0  10 8  0 2  0 < 0..1 > >
+ *
+ *      these bits are set in the lookup table:
+ *
+ * ::
+ *
+ *                     bucket
+ *      group  0   1   2   3   4   5   6   7   8   9  10  11  12  13  14  15
+ *        0    0                                              1,2
+ *        1   1,2                                      0
+ *        2    0                                      1,2
+ *        3    0                              1,2
+ *        4  0,1,2
+ *        5    0   1   2
+ *        6  0,1,2 1   1   1   1   1   1   1   1   1   1   1   1   1   1   1
+ *        7   1,2 1,2  1   1   1  0,1  1   1   1   1   1   1   1   1   1   1
+ *
+ *   - if this is not the last field in the set, fill a mapping array that maps
+ *     rules from the lookup table to rules belonging to the same entry in
+ *     the next lookup table, done by pipapo_map().
+ *
+ *     Note that as rules map to contiguous ranges of rules, given how netmask
+ *     expansion and insertion is performed, &union nft_pipapo_map_bucket stores
+ *     this information as pairs of first rule index, rule count.
+ *
+ *      Example: 2 entries, 10.0.0.5:1024 and 192.168.1.0-192.168.2.1:2048,
+ *      given lookup table #0 for field 0 (see example above):
+ *
+ * ::
+ *
+ *                     bucket
+ *      group  0   1   2   3   4   5   6   7   8   9  10  11  12  13  14  15
+ *        0    0                                              1,2
+ *        1   1,2                                      0
+ *        2    0                                      1,2
+ *        3    0                              1,2
+ *        4  0,1,2
+ *        5    0   1   2
+ *        6  0,1,2 1   1   1   1   1   1   1   1   1   1   1   1   1   1   1
+ *        7   1,2 1,2  1   1   1  0,1  1   1   1   1   1   1   1   1   1   1
+ *
+ *      and lookup table #1 for field 1 with:
+ *      - rule #0: 1024 mapping to buckets
+ *        < 0  0  4  0 >
+ *      - rule #1: 2048 mapping to buckets
+ *        < 0  0  5  0 >
+ *
+ * ::
+ *
+ *                     bucket
+ *      group  0   1   2   3   4   5   6   7   8   9  10  11  12  13  14  15
+ *        0   0,1
+ *        1   0,1
+ *        2                    0   1
+ *        3   0,1
+ *
+ *      we need to map rules for 10.0.0.5 in lookup table #0 (rule #0) to 1024
+ *      in lookup table #1 (rule #0) and rules for 192.168.1.0-192.168.2.1
+ *      (rules #1, #2) to 2048 in lookup table #2 (rule #1):
+ *
+ * ::
+ *
+ *       rule indices in current field: 0    1    2
+ *       map to rules in next field:    0    1    1
+ *
+ *   - if this is the last field in the set, fill a mapping array that maps
+ *     rules from the last lookup table to element pointers, also done by
+ *     pipapo_map().
+ *
+ *     Note that, in this implementation, we have two elements (start, end) for
+ *     each entry. The pointer to the end element is stored in this array, and
+ *     the pointer to the start element is linked from it.
+ *
+ *      Example: entry 10.0.0.5:1024 has a corresponding &struct nft_pipapo_elem
+ *      pointer, 0x66, and element for 192.168.1.0-192.168.2.1:2048 is at 0x42.
+ *      From the rules of lookup table #1 as mapped above:
+ *
+ * ::
+ *
+ *       rule indices in last field:    0    1
+ *       map to elements:             0x42  0x66
+ *
+ *
+ * Matching
+ * --------
+ *
+ * We use a result bitmap, with the size of a single lookup table bucket, to
+ * represent the matching state that applies at every algorithm step. This is
+ * done by pipapo_lookup().
+ *
+ * - For each packet field:
+ *
+ *   - start with an all-ones result bitmap (res_map in pipapo_lookup())
+ *
+ *   - perform a lookup into the table corresponding to the current field,
+ *     for each group, and at every group, AND the current result bitmap with
+ *     the value from the lookup table bucket
+ *
+ * ::
+ *
+ *      Example: 192.168.1.5 < 12 0  10 8  0 1  0 5 >, with lookup table from
+ *      insertion examples.
+ *      Lookup table buckets are at least 3 bits wide, we'll assume 8 bits for
+ *      convenience in this example. Initial result bitmap is 0xff, the steps
+ *      below show the value of the result bitmap after each group is processed:
+ *
+ *                     bucket
+ *      group  0   1   2   3   4   5   6   7   8   9  10  11  12  13  14  15
+ *        0    0                                              1,2
+ *        result bitmap is now: 0xff & 0x6 [bucket 12] = 0x6
+ *
+ *        1   1,2                                      0
+ *        result bitmap is now: 0x6 & 0x6 [bucket 0] = 0x6
+ *
+ *        2    0                                      1,2
+ *        result bitmap is now: 0x6 & 0x6 [bucket 10] = 0x6
+ *
+ *        3    0                              1,2
+ *        result bitmap is now: 0x6 & 0x6 [bucket 8] = 0x6
+ *
+ *        4  0,1,2
+ *        result bitmap is now: 0x6 & 0x7 [bucket 0] = 0x6
+ *
+ *        5    0   1   2
+ *        result bitmap is now: 0x6 & 0x2 [bucket 1] = 0x2
+ *
+ *        6  0,1,2 1   1   1   1   1   1   1   1   1   1   1   1   1   1   1
+ *        result bitmap is now: 0x2 & 0x7 [bucket 0] = 0x2
+ *
+ *        7   1,2 1,2  1   1   1  0,1  1   1   1   1   1   1   1   1   1   1
+ *        final result bitmap for this field is: 0x2 & 0x3 [bucket 5] = 0x2
+ *
+ *   - at the next field, start with a new, all-zeroes result bitmap. For each
+ *     bit set in the previous result bitmap, fill the new result bitmap
+ *     (fill_map in pipapo_lookup()) with the rule indices from the
+ *     corresponding buckets of the mapping field for this field, done by
+ *     pipapo_refill()
+ *
+ *      Example: with mapping table from insertion examples, with the current
+ *      result bitmap from the previous example, 0x02:
+ *
+ * ::
+ *
+ *       rule indices in current field: 0    1    2
+ *       map to rules in next field:    0    1    1
+ *
+ *      the new result bitmap will be 0x02: rule 1 was set, and rule 1 will be
+ *      set.
+ *
+ *      We can now extend this example to cover the second iteration of the step
+ *      above (lookup and AND bitmap): assuming the port field is
+ *      2048 < 0  0  5  0 >, with starting result bitmap 0x2, and lookup table
+ *      for "port" field from pre-computation example:
+ *
+ * ::
+ *
+ *                     bucket
+ *      group  0   1   2   3   4   5   6   7   8   9  10  11  12  13  14  15
+ *        0   0,1
+ *        1   0,1
+ *        2                    0   1
+ *        3   0,1
+ *
+ *       operations are: 0x2 & 0x3 [bucket 0] & 0x3 [bucket 0] & 0x2 [bucket 5]
+ *       & 0x3 [bucket 0], resulting bitmap is 0x2.
+ *
+ *   - if this is the last field in the set, look up the value from the mapping
+ *     array corresponding to the final result bitmap
+ *
+ *      Example: 0x2 resulting bitmap from 192.168.1.5:2048, mapping array for
+ *      last field from insertion example:
+ *
+ * ::
+ *
+ *       rule indices in last field:    0    1
+ *       map to elements:             0x42  0x66
+ *
+ *      the matching element is at 0x42.
+ *
+ *
+ * References
+ * ----------
+ *
+ * [Ligatti 2010]
+ *      A Packet-classification Algorithm for Arbitrary Bitmask Rules, with
+ *      Automatic Time-space Tradeoffs
+ *      Jay Ligatti, Josh Kuhn, and Chris Gage.
+ *      Proceedings of the IEEE International Conference on Computer
+ *      Communication Networks (ICCCN), August 2010.
+ *      http://www.cse.usf.edu/~ligatti/papers/grouper-conf.pdf
+ *
+ * [Rottenstreich 2010]
+ *      Worst-Case TCAM Rule Expansion
+ *      Ori Rottenstreich and Isaac Keslassy.
+ *      2010 Proceedings IEEE INFOCOM, San Diego, CA, 2010.
+ *      http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.212.4592&rep=rep1&type=pdf
+ *
+ * [Kogan 2014]
+ *      SAX-PAC (Scalable And eXpressive PAcket Classification)
+ *      Kirill Kogan, Sergey Nikolenko, Ori Rottenstreich, William Culhane,
+ *      and Patrick Eugster.
+ *      Proceedings of the 2014 ACM conference on SIGCOMM, August 2014.
+ *      http://www.sigcomm.org/sites/default/files/ccr/papers/2014/August/2619239-2626294.pdf
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/log2.h>
+#include <linux/module.h>
+#include <linux/netlink.h>
+#include <linux/netfilter.h>
+#include <linux/netfilter/nf_tables.h>
+#include <net/netfilter/nf_tables_core.h>
+#include <uapi/linux/netfilter/nf_tables.h>
+#include <net/ipv6.h>			/* For the maximum length of a field */
+#include <linux/bitmap.h>
+#include <linux/bitops.h>
+
+/* Count of concatenated fields depends on count of 32-bit nftables registers */
+#define NFT_PIPAPO_MAX_FIELDS		NFT_REG32_COUNT
+
+/* Largest supported field size */
+#define NFT_PIPAPO_MAX_BYTES		(sizeof(struct in6_addr))
+#define NFT_PIPAPO_MAX_BITS		(NFT_PIPAPO_MAX_BYTES * BITS_PER_BYTE)
+
+/* Number of bits to be grouped together in lookup table buckets, arbitrary */
+#define NFT_PIPAPO_GROUP_BITS		4
+#define NFT_PIPAPO_GROUPS_PER_BYTE	(BITS_PER_BYTE / NFT_PIPAPO_GROUP_BITS)
+
+/* Fields are padded to 32 bits in input registers */
+#define NFT_PIPAPO_GROUPS_PADDED_SIZE(x)				\
+	(round_up((x) / NFT_PIPAPO_GROUPS_PER_BYTE, sizeof(u32)))
+#define NFT_PIPAPO_GROUPS_PADDING(x)					\
+	(NFT_PIPAPO_GROUPS_PADDED_SIZE((x)) - (x) / NFT_PIPAPO_GROUPS_PER_BYTE)
+
+/* Number of buckets, given by 2 ^ n, with n grouped bits */
+#define NFT_PIPAPO_BUCKETS		(1 << NFT_PIPAPO_GROUP_BITS)
+
+/* Each n-bit range maps to up to n * 2 rules */
+#define NFT_PIPAPO_MAP_NBITS		(const_ilog2(NFT_PIPAPO_MAX_BITS * 2))
+
+/* Use the rest of mapping table buckets for rule indices, but it makes no sense
+ * to exceed 32 bits
+ */
+#if BITS_PER_LONG == 64
+#define NFT_PIPAPO_MAP_TOBITS		32
+#else
+#define NFT_PIPAPO_MAP_TOBITS		(BITS_PER_LONG - NFT_PIPAPO_MAP_NBITS)
+#endif
+
+/* ...which gives us the highest allowed index for a rule */
+#define NFT_PIPAPO_RULE0_MAX		((1UL << (NFT_PIPAPO_MAP_TOBITS - 1)) \
+					- (1UL << NFT_PIPAPO_MAP_NBITS))
+
+#define nft_pipapo_for_each_field(field, index, match)		\
+	for ((field) = (match)->f, (index) = 0;			\
+	     (index) < (match)->field_count;			\
+	     (index)++, (field)++)
+
+/**
+ * union nft_pipapo_map_bucket - Bucket of mapping table
+ * @to:		First rule number (in next field) this rule maps to
+ * @n:		Number of rules (in next field) this rule maps to
+ * @e:		If there's no next field, pointer to element this rule maps to
+ */
+union nft_pipapo_map_bucket {
+	struct {
+#if BITS_PER_LONG == 64
+		static_assert(NFT_PIPAPO_MAP_TOBITS <= 32);
+		u32 to;
+
+		static_assert(NFT_PIPAPO_MAP_NBITS <= 32);
+		u32 n;
+#else
+		unsigned long to:NFT_PIPAPO_MAP_TOBITS;
+		unsigned long  n:NFT_PIPAPO_MAP_NBITS;
+#endif
+	};
+	struct nft_pipapo_elem *e;
+};
+
+/**
+ * struct nft_pipapo_field - Lookup, mapping tables and related data for a field
+ * @groups:	Amount of 4-bit groups
+ * @rules:	Number of inserted rules
+ * @bsize:	Size of each bucket in lookup table, in longs
+ * @lt:		Lookup table: 'groups' rows of NFT_PIPAPO_BUCKETS buckets
+ * @mt:		Mapping table: one bucket per rule
+ */
+struct nft_pipapo_field {
+	int groups;
+	unsigned long rules;
+	size_t bsize;
+	unsigned long *lt;
+	union nft_pipapo_map_bucket *mt;
+};
+
+/**
+ * struct nft_pipapo_match - Data used for lookup and matching
+ * @field_count		Amount of fields in set
+ * @scratch:		Preallocated per-CPU maps for partial matching results
+ * @bsize_max:		Maximum lookup table bucket size of all fields, in longs
+ * @rcu			Matching data is swapped on commits
+ * @f:			Fields, with lookup and mapping tables
+ */
+struct nft_pipapo_match {
+	int field_count;
+	unsigned long * __percpu *scratch;
+	size_t bsize_max;
+	struct rcu_head rcu;
+	struct nft_pipapo_field f[0];
+};
+
+/* Current working bitmap index, toggled between field matches */
+static DEFINE_PER_CPU(bool, nft_pipapo_scratch_index);
+
+/**
+ * struct nft_pipapo - Representation of a set
+ * @match:	Currently in-use matching data
+ * @clone:	Copy where pending insertions and deletions are kept
+ * @groups:	Total amount of 4-bit groups for fields in this set
+ * @width:	Total bytes to be matched for one packet, including padding
+ * @dirty:	Working copy has pending insertions or deletions
+ * @last_gc:	Timestamp of last garbage collection run, jiffies
+ */
+struct nft_pipapo {
+	struct nft_pipapo_match __rcu *match;
+	struct nft_pipapo_match *clone;
+	int groups;
+	int width;
+	bool dirty;
+	unsigned long last_gc;
+};
+
+struct nft_pipapo_elem;
+
+/**
+ * struct nft_pipapo_elem - API-facing representation of single set element
+ * @ext:	nftables API extensions
+ */
+struct nft_pipapo_elem {
+	struct nft_set_ext ext;
+};
+
+/**
+ * pipapo_refill() - For each set bit, set bits from selected mapping table item
+ * @map:	Bitmap to be scanned for set bits
+ * @len:	Length of bitmap in longs
+ * @rules:	Number of rules in field
+ * @dst:	Destination bitmap
+ * @mt:		Mapping table containing bit set specifiers
+ * @match_only:	Find a single bit and return, don't fill
+ *
+ * Iteration over set bits with __builtin_ctzl(): Daniel Lemire, public domain.
+ *
+ * For each bit set in map, select the bucket from mapping table with index
+ * corresponding to the position of the bit set. Use start bit and amount of
+ * bits specified in bucket to fill region in dst.
+ *
+ * Return: -1 on no match, bit position on 'match_only', 0 otherwise.
+ */
+static int pipapo_refill(unsigned long *map, int len, int rules,
+			 unsigned long *dst, union nft_pipapo_map_bucket *mt,
+			 bool match_only)
+{
+	unsigned long bitset;
+	int k, ret = -1;
+
+	for (k = 0; k < len; k++) {
+		bitset = map[k];
+		while (bitset) {
+			unsigned long t = bitset & -bitset;
+			int r = __builtin_ctzl(bitset);
+			int i = k * BITS_PER_LONG + r;
+
+			if (unlikely(i >= rules)) {
+				map[k] = 0;
+				return -1;
+			}
+
+			if (unlikely(match_only)) {
+				bitmap_clear(map, i, 1);
+				return i;
+			}
+
+			ret = 0;
+
+			bitmap_set(dst, mt[i].to, mt[i].n);
+
+			bitset ^= t;
+		}
+		map[k] = 0;
+	}
+
+	return ret;
+}
+
+/**
+ * nft_pipapo_lookup() - Lookup function
+ * @net:	Network namespace
+ * @set:	nftables API set representation
+ * @elem:	nftables API element representation containing key data
+ * @ext:	nftables API extension pointer, filled with matching reference
+ *
+ * For more details, see DOC: Theory of Operation.
+ *
+ * Return: true on match, false otherwise.
+ */
+static bool nft_pipapo_lookup(const struct net *net, const struct nft_set *set,
+			      const u32 *key, const struct nft_set_ext **ext)
+{
+	struct nft_pipapo *priv = nft_set_priv(set);
+	unsigned long *res_map, *fill_map;
+	u8 genmask = nft_genmask_cur(net);
+	const u8 *rp = (const u8 *)key;
+	struct nft_pipapo_match *m;
+	struct nft_pipapo_field *f;
+	bool map_index;
+	int i;
+
+	local_bh_disable();
+
+	map_index = raw_cpu_read(nft_pipapo_scratch_index);
+
+	m = rcu_dereference(priv->match);
+
+	if (unlikely(!m || !*raw_cpu_ptr(m->scratch)))
+		goto out;
+
+	res_map  = *raw_cpu_ptr(m->scratch) + (map_index ? m->bsize_max : 0);
+	fill_map = *raw_cpu_ptr(m->scratch) + (map_index ? 0 : m->bsize_max);
+
+	memset(res_map, 0xff, m->bsize_max * sizeof(*res_map));
+
+	nft_pipapo_for_each_field(f, i, m) {
+		bool last = i == m->field_count - 1;
+		unsigned long *lt = f->lt;
+		int b, group;
+
+		/* For each 4-bit group: select lookup table bucket depending on
+		 * packet bytes value, then AND bucket value
+		 */
+		for (group = 0; group < f->groups; group += 2) {
+			u8 v;
+
+			v = *rp >> 4;
+			__bitmap_and(res_map, res_map, lt + v * f->bsize,
+				     f->bsize * BITS_PER_LONG);
+			lt += f->bsize * NFT_PIPAPO_BUCKETS;
+
+			v = *rp & 0x0f;
+			rp++;
+			__bitmap_and(res_map, res_map, lt + v * f->bsize,
+				     f->bsize * BITS_PER_LONG);
+			lt += f->bsize * NFT_PIPAPO_BUCKETS;
+		}
+
+		/* Now populate the bitmap for the next field, unless this is
+		 * the last field, in which case return the matched 'ext'
+		 * pointer if any.
+		 *
+		 * Now res_map contains the matching bitmap, and fill_map is the
+		 * bitmap for the next field.
+		 */
+next_match:
+		b = pipapo_refill(res_map, f->bsize, f->rules, fill_map, f->mt,
+				  last);
+		if (b < 0) {
+			raw_cpu_write(nft_pipapo_scratch_index, map_index);
+			local_bh_enable();
+
+			return false;
+		}
+
+		if (last) {
+			*ext = &f->mt[b].e->ext;
+			if (unlikely(nft_set_elem_expired(*ext) ||
+				     !nft_set_elem_active(*ext, genmask)))
+				goto next_match;
+
+			/* Last field: we're just returning the key without
+			 * filling the initial bitmap for the next field, so the
+			 * current inactive bitmap is clean and can be reused as
+			 * *next* bitmap (not initial) for the next packet.
+			 */
+			raw_cpu_write(nft_pipapo_scratch_index, map_index);
+			local_bh_enable();
+
+			return true;
+		}
+
+		/* Swap bitmap indices: res_map is the initial bitmap for the
+		 * next field, and fill_map is guaranteed to be all-zeroes at
+		 * this point.
+		 */
+		map_index = !map_index;
+		swap(res_map, fill_map);
+
+		rp += NFT_PIPAPO_GROUPS_PADDING(f->groups);
+	}
+
+out:
+	local_bh_enable();
+	return false;
+}
+
+/**
+ * pipapo_get() - Get matching element reference given key data
+ * @net:	Network namespace
+ * @set:	nftables API set representation
+ * @data:	Key data to be matched against existing elements
+ * @genmask:	If set, check that element is active in given genmask
+ *
+ * This is essentially the same as the lookup function, except that it matches
+ * key data against the uncommitted copy and doesn't use preallocated maps for
+ * bitmap results.
+ *
+ * Return: pointer to &struct nft_pipapo_elem on match, error pointer otherwise.
+ */
+static struct nft_pipapo_elem *pipapo_get(const struct net *net,
+					  const struct nft_set *set,
+					  const u8 *data, u8 genmask)
+{
+	struct nft_pipapo_elem *ret = ERR_PTR(-ENOENT);
+	struct nft_pipapo *priv = nft_set_priv(set);
+	struct nft_pipapo_match *m = priv->clone;
+	unsigned long *res_map, *fill_map = NULL;
+	struct nft_pipapo_field *f;
+	int i;
+
+	res_map = kmalloc_array(m->bsize_max, sizeof(*res_map), GFP_ATOMIC);
+	if (!res_map) {
+		ret = ERR_PTR(-ENOMEM);
+		goto out;
+	}
+
+	fill_map = kcalloc(m->bsize_max, sizeof(*res_map), GFP_ATOMIC);
+	if (!fill_map) {
+		ret = ERR_PTR(-ENOMEM);
+		goto out;
+	}
+
+	memset(res_map, 0xff, m->bsize_max * sizeof(*res_map));
+
+	nft_pipapo_for_each_field(f, i, m) {
+		bool last = i == m->field_count - 1;
+		unsigned long *lt = f->lt;
+		int b, group;
+
+		/* For each 4-bit group: select lookup table bucket depending on
+		 * packet bytes value, then AND bucket value
+		 */
+		for (group = 0; group < f->groups; group++) {
+			u8 v;
+
+			if (group % 2) {
+				v = *data & 0x0f;
+				data++;
+			} else {
+				v = *data >> 4;
+			}
+			__bitmap_and(res_map, res_map, lt + v * f->bsize,
+				     f->bsize * BITS_PER_LONG);
+
+			lt += f->bsize * NFT_PIPAPO_BUCKETS;
+		}
+
+		/* Now populate the bitmap for the next field, unless this is
+		 * the last field, in which case return the matched 'ext'
+		 * pointer if any.
+		 *
+		 * Now res_map contains the matching bitmap, and fill_map is the
+		 * bitmap for the next field.
+		 */
+next_match:
+		b = pipapo_refill(res_map, f->bsize, f->rules, fill_map, f->mt,
+				  last);
+		if (b < 0)
+			goto out;
+
+		if (last) {
+			if (nft_set_elem_expired(&f->mt[b].e->ext) ||
+			    (genmask &&
+			     !nft_set_elem_active(&f->mt[b].e->ext, genmask)))
+				goto next_match;
+
+			ret = f->mt[b].e;
+			goto out;
+		}
+
+		data += NFT_PIPAPO_GROUPS_PADDING(f->groups);
+
+		/* Swap bitmap indices: fill_map will be the initial bitmap for
+		 * the next field (i.e. the new res_map), and res_map is
+		 * guaranteed to be all-zeroes at this point, ready to be filled
+		 * according to the next mapping table.
+		 */
+		swap(res_map, fill_map);
+	}
+
+out:
+	kfree(fill_map);
+	kfree(res_map);
+	return ret;
+}
+
+/**
+ * nft_pipapo_get() - Get matching element reference given key data
+ * @net:	Network namespace
+ * @set:	nftables API set representation
+ * @elem:	nftables API element representation containing key data
+ * @flags:	Unused
+ */
+void *nft_pipapo_get(const struct net *net, const struct nft_set *set,
+		     const struct nft_set_elem *elem, unsigned int flags)
+{
+	return pipapo_get(net, set, (const u8 *)elem->key.val.data,
+			  nft_genmask_cur(net));
+}
+
+/**
+ * pipapo_resize() - Resize lookup or mapping table, or both
+ * @f:		Field containing lookup and mapping tables
+ * @old_rules:	Previous amount of rules in field
+ * @rules:	New amount of rules
+ *
+ * Increase, decrease or maintain tables size depending on new amount of rules,
+ * and copy data over. In case the new size is smaller, throw away data for
+ * highest-numbered rules.
+ *
+ * Return: 0 on success, -ENOMEM on allocation failure.
+ */
+static int pipapo_resize(struct nft_pipapo_field *f, int old_rules, int rules)
+{
+	long *new_lt = NULL, *new_p, *old_lt = f->lt, *old_p;
+	union nft_pipapo_map_bucket *new_mt, *old_mt = f->mt;
+	size_t new_bucket_size, copy;
+	int group, bucket;
+
+	new_bucket_size = DIV_ROUND_UP(rules, BITS_PER_LONG);
+
+	if (new_bucket_size == f->bsize)
+		goto mt;
+
+	if (new_bucket_size > f->bsize)
+		copy = f->bsize;
+	else
+		copy = new_bucket_size;
+
+	new_lt = kvzalloc(f->groups * NFT_PIPAPO_BUCKETS * new_bucket_size *
+			  sizeof(*new_lt), GFP_KERNEL);
+	if (!new_lt)
+		return -ENOMEM;
+
+	new_p = new_lt;
+	old_p = old_lt;
+	for (group = 0; group < f->groups; group++) {
+		for (bucket = 0; bucket < NFT_PIPAPO_BUCKETS; bucket++) {
+			memcpy(new_p, old_p, copy * sizeof(*new_p));
+			new_p += copy;
+			old_p += copy;
+
+			if (new_bucket_size > f->bsize)
+				new_p += new_bucket_size - f->bsize;
+			else
+				old_p += f->bsize - new_bucket_size;
+		}
+	}
+
+mt:
+	new_mt = kvmalloc(rules * sizeof(*new_mt), GFP_KERNEL);
+	if (!new_mt) {
+		kvfree(new_lt);
+		return -ENOMEM;
+	}
+
+	memcpy(new_mt, f->mt, min(old_rules, rules) * sizeof(*new_mt));
+	if (rules > old_rules) {
+		memset(new_mt + old_rules, 0,
+		       (rules - old_rules) * sizeof(*new_mt));
+	}
+
+	if (new_lt) {
+		f->bsize = new_bucket_size;
+		f->lt = new_lt;
+		kvfree(old_lt);
+	}
+
+	f->mt = new_mt;
+	kvfree(old_mt);
+
+	return 0;
+}
+
+/**
+ * pipapo_bucket_set() - Set rule bit in bucket given group and group value
+ * @f:		Field containing lookup table
+ * @rule:	Rule index
+ * @group:	Group index
+ * @v:		Value of bit group
+ */
+static void pipapo_bucket_set(struct nft_pipapo_field *f, int rule, int group,
+			      int v)
+{
+	unsigned long *pos;
+
+	pos = f->lt + f->bsize * NFT_PIPAPO_BUCKETS * group;
+	pos += f->bsize * v;
+
+	__set_bit(rule, pos);
+}
+
+/**
+ * pipapo_insert() - Insert new rule in field given input key and mask length
+ * @f:		Field containing lookup table
+ * @k:		Input key for classification, without nftables padding
+ * @mask_bits:	Length of mask; matches field length for non-ranged entry
+ *
+ * Insert a new rule reference in lookup buckets corresponding to k and
+ * mask_bits.
+ *
+ * Return: 1 on success (one rule inserted), negative error code on failure.
+ */
+static int pipapo_insert(struct nft_pipapo_field *f, const uint8_t *k,
+			 int mask_bits)
+{
+	int rule = f->rules++, group, ret;
+
+	ret = pipapo_resize(f, f->rules - 1, f->rules);
+	if (ret)
+		return ret;
+
+	for (group = 0; group < f->groups; group++) {
+		int i, v;
+		u8 mask;
+
+		if (group % 2)
+			v = k[group / 2] & 0x0f;
+		else
+			v = k[group / 2] >> 4;
+
+		if (mask_bits >= (group + 1) * 4) {
+			/* Not masked */
+			pipapo_bucket_set(f, rule, group, v);
+		} else if (mask_bits <= group * 4) {
+			/* Completely masked */
+			for (i = 0; i < NFT_PIPAPO_BUCKETS; i++)
+				pipapo_bucket_set(f, rule, group, i);
+		} else {
+			/* The mask limit falls on this group */
+			mask = 0x0f >> (mask_bits - group * 4);
+			for (i = 0; i < NFT_PIPAPO_BUCKETS; i++) {
+				if ((i & ~mask) == (v & ~mask))
+					pipapo_bucket_set(f, rule, group, i);
+			}
+		}
+	}
+
+	return 1;
+}
+
+/**
+ * pipapo_step_diff() - Check if setting @step bit in netmask would change it
+ * @base:	Mask we are expanding
+ * @step:	Step bit for given expansion step
+ * @len:	Total length of mask space (set and unset bits), bytes
+ *
+ * Convenience function for mask expansion.
+ *
+ * Return: true if step bit changes mask (i.e. isn't set), false otherwise.
+ */
+static bool pipapo_step_diff(u8 *base, int step, int len)
+{
+	/* Network order, byte-addressed */
+#ifdef __BIG_ENDIAN__
+	return !(BIT(step % BITS_PER_BYTE) & base[step / BITS_PER_BYTE]);
+#else
+	return !(BIT(step % BITS_PER_BYTE) &
+		 base[len - 1 - step / BITS_PER_BYTE]);
+#endif
+}
+
+/**
+ * pipapo_step_after_end() - Check if mask exceeds range end with given step
+ * @base:	Mask we are expanding
+ * @end:	End of range
+ * @step:	Step bit for given expansion step, highest bit to be set
+ * @len:	Total length of mask space (set and unset bits), bytes
+ *
+ * Convenience function for mask expansion.
+ *
+ * Return: true if mask exceeds range setting step bits, false otherwise.
+ */
+static bool pipapo_step_after_end(const u8 *base, const u8 *end, int step,
+				  int len)
+{
+	u8 tmp[NFT_PIPAPO_MAX_BYTES];
+	int i;
+
+	memcpy(tmp, base, len);
+
+	/* Network order, byte-addressed */
+	for (i = 0; i <= step; i++)
+#ifdef __BIG_ENDIAN__
+		tmp[i / BITS_PER_BYTE] |= BIT(i % BITS_PER_BYTE);
+#else
+		tmp[len - 1 - i / BITS_PER_BYTE] |= BIT(i % BITS_PER_BYTE);
+#endif
+
+	return memcmp(tmp, end, len) > 0;
+}
+
+/**
+ * pipapo_base_sum() - Sum step bit to given len-sized netmask base with carry
+ * @base:	Netmask base
+ * @step:	Step bit to sum
+ * @len:	Netmask length, bytes
+ */
+static void pipapo_base_sum(u8 *base, int step, int len)
+{
+	bool carry = false;
+	int i;
+
+	/* Network order, byte-addressed */
+#ifdef __BIG_ENDIAN__
+	for (i = step / BITS_PER_BYTE; i < len; i++) {
+#else
+	for (i = len - 1 - step / BITS_PER_BYTE; i >= 0; i--) {
+#endif
+		if (carry)
+			base[i]++;
+		else
+			base[i] += 1 << (step % BITS_PER_BYTE);
+
+		if (base[i])
+			break;
+
+		carry = true;
+	}
+}
+
+/**
+ * pipapo_expand() - Expand to composing netmasks, insert into lookup table
+ * @f:		Field containing lookup table
+ * @start:	Start of range
+ * @end:	End of range
+ * @len:	Length of value in bits
+ *
+ * Expand range to composing netmasks and insert corresponding rule references
+ * in lookup buckets.
+ *
+ * Return: number of inserted rules on success, negative error code on failure.
+ */
+static int pipapo_expand(struct nft_pipapo_field *f,
+			 const u8 *start, const u8 *end, int len)
+{
+	int step, masks = 0, bytes = DIV_ROUND_UP(len, BITS_PER_BYTE);
+	u8 base[NFT_PIPAPO_MAX_BYTES];
+
+	memcpy(base, start, bytes);
+	while (memcmp(base, end, bytes) <= 0) {
+		int err;
+
+		step = 0;
+		while (pipapo_step_diff(base, step, bytes)) {
+			if (pipapo_step_after_end(base, end, step, bytes))
+				break;
+
+			step++;
+			if (step >= len) {
+				if (!masks) {
+					pipapo_insert(f, base, 0);
+					masks = 1;
+				}
+				goto out;
+			}
+		}
+
+		err = pipapo_insert(f, base, len - step);
+
+		if (err < 0)
+			return err;
+
+		masks++;
+		pipapo_base_sum(base, step, bytes);
+	}
+out:
+	return masks;
+}
+
+/**
+ * pipapo_map() - Insert rules in mapping tables, mapping them between fields
+ * @m:		Matching data, including mapping table
+ * @map:	Table of rule maps: array of first rule and amount of rules
+ *		in next field a given rule maps to, for each field
+ * @ext:	For last field, nft_set_ext pointer matching rules map to
+ */
+static void pipapo_map(struct nft_pipapo_match *m,
+		       union nft_pipapo_map_bucket map[NFT_PIPAPO_MAX_FIELDS],
+		       struct nft_pipapo_elem *e)
+{
+	struct nft_pipapo_field *f;
+	int i, j;
+
+	for (i = 0, f = m->f; i < m->field_count - 1; i++, f++) {
+		for (j = 0; j < map[i].n; j++) {
+			f->mt[map[i].to + j].to = map[i + 1].to;
+			f->mt[map[i].to + j].n = map[i + 1].n;
+		}
+	}
+
+	/* Last field: map to ext instead of mapping to next field */
+	for (j = 0; j < map[i].n; j++)
+		f->mt[map[i].to + j].e = e;
+}
+
+/**
+ * pipapo_realloc_scratch() - Reallocate scratch maps for partial match results
+ * @clone:	Copy of matching data with pending insertions and deletions
+ * @bsize_max	Maximum bucket size, scratch maps cover two buckets
+ *
+ * Return: 0 on success, -ENOMEM on failure.
+ */
+static int pipapo_realloc_scratch(struct nft_pipapo_match *clone,
+				  unsigned long bsize_max)
+{
+	int i;
+
+	for_each_possible_cpu(i) {
+		unsigned long *scratch;
+
+		scratch = kzalloc_node(bsize_max * sizeof(*scratch) * 2,
+				       GFP_KERNEL, cpu_to_node(i));
+		if (!scratch) {
+			/* On failure, there's no need to undo previous
+			 * allocations: this means that some scratch maps have
+			 * a bigger allocated size now (this is only called on
+			 * insertion), but the extra space won't be used by any
+			 * CPU as new elements are not inserted and m->bsize_max
+			 * is not updated.
+			 */
+			return -ENOMEM;
+		}
+
+		kfree(*per_cpu_ptr(clone->scratch, i));
+
+		*per_cpu_ptr(clone->scratch, i) = scratch;
+	}
+
+	return 0;
+}
+
+/**
+ * nft_pipapo_insert() - Validate and insert ranged elements
+ * @net:	Network namespace
+ * @set:	nftables API set representation
+ * @elem:	nftables API element representation containing key data
+ * @ext2:	Filled with pointer to &struct nft_set_ext in inserted element
+ *
+ * Return: 0 on success, error pointer on failure.
+ */
+static int nft_pipapo_insert(const struct net *net, const struct nft_set *set,
+			     const struct nft_set_elem *elem,
+			     struct nft_set_ext **ext2)
+{
+	const struct nft_set_ext *ext = nft_set_elem_ext(set, elem->priv);
+	union nft_pipapo_map_bucket rulemap[NFT_PIPAPO_MAX_FIELDS];
+	const u8 *start = (const u8 *)elem->key.val.data, *end;
+	struct nft_pipapo_elem *e = elem->priv, *dup;
+	struct nft_pipapo *priv = nft_set_priv(set);
+	struct nft_pipapo_match *m = priv->clone;
+	u8 genmask = nft_genmask_next(net);
+	struct nft_pipapo_field *f;
+	int i, bsize_max, err = 0;
+
+	dup = pipapo_get(net, set, start, genmask);
+	if (PTR_ERR(dup) == -ENOENT) {
+		if (nft_set_ext_exists(ext, NFT_SET_EXT_KEY_END)) {
+			end = (const u8 *)nft_set_ext_key_end(ext)->data;
+			dup = pipapo_get(net, set, end, nft_genmask_next(net));
+		} else {
+			end = start;
+		}
+	}
+
+	if (PTR_ERR(dup) != -ENOENT) {
+		if (IS_ERR(dup))
+			return PTR_ERR(dup);
+		*ext2 = &dup->ext;
+		return -EEXIST;
+	}
+
+	/* Validate */
+	nft_pipapo_for_each_field(f, i, m) {
+		const u8 *start_p = start, *end_p = end;
+
+		if (f->rules >= (unsigned long)NFT_PIPAPO_RULE0_MAX)
+			return -ENOSPC;
+
+		if (memcmp(start_p, end_p,
+			   f->groups / NFT_PIPAPO_GROUPS_PER_BYTE) > 0)
+			return -EINVAL;
+
+		start_p += NFT_PIPAPO_GROUPS_PADDED_SIZE(f->groups);
+		end_p += NFT_PIPAPO_GROUPS_PADDED_SIZE(f->groups);
+	}
+
+	/* Insert */
+	priv->dirty = true;
+
+	bsize_max = m->bsize_max;
+
+	nft_pipapo_for_each_field(f, i, m) {
+		int ret;
+
+		rulemap[i].to = f->rules;
+
+		ret = memcmp(start, end,
+			     f->groups / NFT_PIPAPO_GROUPS_PER_BYTE);
+		if (!ret) {
+			ret = pipapo_insert(f, start,
+					    f->groups * NFT_PIPAPO_GROUP_BITS);
+		} else {
+			ret = pipapo_expand(f, start, end,
+					    f->groups * NFT_PIPAPO_GROUP_BITS);
+		}
+
+		if (f->bsize > bsize_max)
+			bsize_max = f->bsize;
+
+		rulemap[i].n = ret;
+
+		start += NFT_PIPAPO_GROUPS_PADDED_SIZE(f->groups);
+		end += NFT_PIPAPO_GROUPS_PADDED_SIZE(f->groups);
+	}
+
+	if (!*this_cpu_ptr(m->scratch) || bsize_max > m->bsize_max) {
+		err = pipapo_realloc_scratch(m, bsize_max);
+		if (err)
+			return err;
+
+		this_cpu_write(nft_pipapo_scratch_index, false);
+
+		m->bsize_max = bsize_max;
+	}
+
+	*ext2 = &e->ext;
+
+	pipapo_map(m, rulemap, e);
+
+	return 0;
+}
+
+/**
+ * pipapo_clone() - Clone matching data to create new working copy
+ * @old:	Existing matching data
+ *
+ * Return: copy of matching data passed as 'old', error pointer on failure
+ */
+static struct nft_pipapo_match *pipapo_clone(struct nft_pipapo_match *old)
+{
+	struct nft_pipapo_field *dst, *src;
+	struct nft_pipapo_match *new;
+	int i;
+
+	new = kmalloc(sizeof(*new) + sizeof(*dst) * old->field_count,
+		      GFP_KERNEL);
+	if (!new)
+		return ERR_PTR(-ENOMEM);
+
+	new->field_count = old->field_count;
+	new->bsize_max = old->bsize_max;
+
+	new->scratch = alloc_percpu(*new->scratch);
+	if (!new->scratch)
+		goto out_scratch;
+
+	rcu_head_init(&new->rcu);
+
+	src = old->f;
+	dst = new->f;
+
+	for (i = 0; i < old->field_count; i++) {
+		memcpy(dst, src, offsetof(struct nft_pipapo_field, lt));
+
+		dst->lt = kvzalloc(src->groups * NFT_PIPAPO_BUCKETS *
+				   src->bsize * sizeof(*dst->lt),
+				   GFP_KERNEL);
+		if (!dst->lt)
+			goto out_lt;
+
+		memcpy(dst->lt, src->lt,
+		       src->bsize * sizeof(*dst->lt) *
+		       src->groups * NFT_PIPAPO_BUCKETS);
+
+		dst->mt = kvmalloc(src->rules * sizeof(*src->mt), GFP_KERNEL);
+		if (!dst->mt)
+			goto out_mt;
+
+		memcpy(dst->mt, src->mt, src->rules * sizeof(*src->mt));
+		src++;
+		dst++;
+	}
+
+	return new;
+
+out_mt:
+	kvfree(dst->lt);
+out_lt:
+	for (dst--; i > 0; i--) {
+		kvfree(dst->mt);
+		kvfree(dst->lt);
+		dst--;
+	}
+	free_percpu(new->scratch);
+out_scratch:
+	kfree(new);
+
+	return ERR_PTR(-ENOMEM);
+}
+
+/**
+ * pipapo_rules_same_key() - Get number of rules originated from the same entry
+ * @f:		Field containing mapping table
+ * @first:	Index of first rule in set of rules mapping to same entry
+ *
+ * Using the fact that all rules in a field that originated from the same entry
+ * will map to the same set of rules in the next field, or to the same element
+ * reference, return the cardinality of the set of rules that originated from
+ * the same entry as the rule with index @first, @first rule included.
+ *
+ * In pictures:
+ *				rules
+ *	field #0		0    1    2    3    4
+ *		map to:		0    1   2-4  2-4  5-9
+ *				.    .    .......   . ...
+ *				|    |    |    | \   \
+ *				|    |    |    |  \   \
+ *				|    |    |    |   \   \
+ *				'    '    '    '    '   \
+ *	in field #1		0    1    2    3    4    5 ...
+ *
+ * if this is called for rule 2 on field #0, it will return 3, as also rules 2
+ * and 3 in field 0 map to the same set of rules (2, 3, 4) in the next field.
+ *
+ * For the last field in a set, we can rely on associated entries to map to the
+ * same element references.
+ *
+ * Return: Number of rules that originated from the same entry as @first.
+ */
+static int pipapo_rules_same_key(struct nft_pipapo_field *f, int first)
+{
+	struct nft_pipapo_elem *e = NULL; /* Keep gcc happy */
+	int r;
+
+	for (r = first; r < f->rules; r++) {
+		if (r != first && e != f->mt[r].e)
+			return r - first;
+
+		e = f->mt[r].e;
+	}
+
+	if (r != first)
+		return r - first;
+
+	return 0;
+}
+
+/**
+ * pipapo_unmap() - Remove rules from mapping tables, renumber remaining ones
+ * @mt:		Mapping array
+ * @rules:	Original amount of rules in mapping table
+ * @start:	First rule index to be removed
+ * @n:		Amount of rules to be removed
+ * @to_offset:	First rule index, in next field, this group of rules maps to
+ * @is_last:	If this is the last field, delete reference from mapping array
+ *
+ * This is used to unmap rules from the mapping table for a single field,
+ * maintaining consistency and compactness for the existing ones.
+ *
+ * In pictures: let's assume that we want to delete rules 2 and 3 from the
+ * following mapping array:
+ *
+ *                 rules
+ *               0      1      2      3      4
+ *      map to:  4-10   4-10   11-15  11-15  16-18
+ *
+ * the result will be:
+ *
+ *                 rules
+ *               0      1      2
+ *      map to:  4-10   4-10   11-13
+ *
+ * for fields before the last one. In case this is the mapping table for the
+ * last field in a set, and rules map to pointers to &struct nft_pipapo_elem:
+ *
+ *                      rules
+ *                        0      1      2      3      4
+ *  element pointers:  0x42   0x42   0x33   0x33   0x44
+ *
+ * the result will be:
+ *
+ *                      rules
+ *                        0      1      2
+ *  element pointers:  0x42   0x42   0x44
+ */
+static void pipapo_unmap(union nft_pipapo_map_bucket *mt, int rules,
+			 int start, int n, int to_offset, bool is_last)
+{
+	int i;
+
+	memmove(mt + start, mt + start + n, (rules - start - n) * sizeof(*mt));
+	memset(mt + rules - n, 0, n * sizeof(*mt));
+
+	if (is_last)
+		return;
+
+	for (i = start; i < rules - n; i++)
+		mt[i].to -= to_offset;
+}
+
+/**
+ * pipapo_drop() - Delete entry from lookup and mapping tables, given rule map
+ * @m:		Matching data
+ * @rulemap	Table of rule maps, arrays of first rule and amount of rules
+ *		in next field a given entry maps to, for each field
+ *
+ * For each rule in lookup table buckets mapping to this set of rules, drop
+ * all bits set in lookup table mapping. In pictures, assuming we want to drop
+ * rules 0 and 1 from this lookup table:
+ *
+ *                     bucket
+ *      group  0   1   2   3   4   5   6   7   8   9  10  11  12  13  14  15
+ *        0    0                                              1,2
+ *        1   1,2                                      0
+ *        2    0                                      1,2
+ *        3    0                              1,2
+ *        4  0,1,2
+ *        5    0   1   2
+ *        6  0,1,2 1   1   1   1   1   1   1   1   1   1   1   1   1   1   1
+ *        7   1,2 1,2  1   1   1  0,1  1   1   1   1   1   1   1   1   1   1
+ *
+ * rule 2 becomes rule 0, and the result will be:
+ *
+ *                     bucket
+ *      group  0   1   2   3   4   5   6   7   8   9  10  11  12  13  14  15
+ *        0                                                    0
+ *        1    0
+ *        2                                            0
+ *        3                                    0
+ *        4    0
+ *        5            0
+ *        6    0
+ *        7    0   0
+ *
+ * once this is done, call unmap() to drop all the corresponding rule references
+ * from mapping tables.
+ */
+static void pipapo_drop(struct nft_pipapo_match *m,
+			union nft_pipapo_map_bucket rulemap[])
+{
+	struct nft_pipapo_field *f;
+	int i;
+
+	nft_pipapo_for_each_field(f, i, m) {
+		int g;
+
+		for (g = 0; g < f->groups; g++) {
+			unsigned long *pos;
+			int b;
+
+			pos = f->lt + g * NFT_PIPAPO_BUCKETS * f->bsize;
+
+			for (b = 0; b < NFT_PIPAPO_BUCKETS; b++) {
+				bitmap_cut(pos, pos, rulemap[i].to,
+					   rulemap[i].n,
+					   f->bsize * BITS_PER_LONG);
+
+				pos += f->bsize;
+			}
+		}
+
+		pipapo_unmap(f->mt, f->rules, rulemap[i].to, rulemap[i].n,
+			     rulemap[i + 1].n, i == m->field_count - 1);
+		if (pipapo_resize(f, f->rules, f->rules - rulemap[i].n)) {
+			/* We can ignore this, a failure to shrink tables down
+			 * doesn't make tables invalid.
+			 */
+			;
+		}
+		f->rules -= rulemap[i].n;
+	}
+}
+
+/**
+ * pipapo_gc() - Drop expired entries from set, destroy start and end elements
+ * @set:	nftables API set representation
+ * @m:		Matching data
+ */
+static void pipapo_gc(const struct nft_set *set, struct nft_pipapo_match *m)
+{
+	struct nft_pipapo *priv = nft_set_priv(set);
+	int rules_f0, first_rule = 0;
+
+	while ((rules_f0 = pipapo_rules_same_key(m->f, first_rule))) {
+		union nft_pipapo_map_bucket rulemap[NFT_PIPAPO_MAX_FIELDS];
+		struct nft_pipapo_field *f;
+		struct nft_pipapo_elem *e;
+		int i, start, rules_fx;
+
+		start = first_rule;
+		rules_fx = rules_f0;
+
+		nft_pipapo_for_each_field(f, i, m) {
+			rulemap[i].to = start;
+			rulemap[i].n = rules_fx;
+
+			if (i < m->field_count - 1) {
+				rules_fx = f->mt[start].n;
+				start = f->mt[start].to;
+			}
+		}
+
+		/* Pick the last field, and its last index */
+		f--;
+		i--;
+		e = f->mt[rulemap[i].to].e;
+		if (nft_set_elem_expired(&e->ext) &&
+		    !nft_set_elem_mark_busy(&e->ext)) {
+			priv->dirty = true;
+			pipapo_drop(m, rulemap);
+
+			rcu_barrier();
+			nft_set_elem_destroy(set, e, true);
+
+			/* And check again current first rule, which is now the
+			 * first we haven't checked.
+			 */
+		} else {
+			first_rule += rules_f0;
+		}
+	}
+
+	priv->last_gc = jiffies;
+}
+
+/**
+ * pipapo_free_fields() - Free per-field tables contained in matching data
+ * @m:		Matching data
+ */
+static void pipapo_free_fields(struct nft_pipapo_match *m)
+{
+	struct nft_pipapo_field *f;
+	int i;
+
+	nft_pipapo_for_each_field(f, i, m) {
+		kvfree(f->lt);
+		kvfree(f->mt);
+	}
+}
+
+/**
+ * pipapo_reclaim_match - RCU callback to free fields from old matching data
+ * @rcu:	RCU head
+ */
+static void pipapo_reclaim_match(struct rcu_head *rcu)
+{
+	struct nft_pipapo_match *m;
+	int i;
+
+	m = container_of(rcu, struct nft_pipapo_match, rcu);
+
+	for_each_possible_cpu(i)
+		kfree(*per_cpu_ptr(m->scratch, i));
+
+	free_percpu(m->scratch);
+
+	pipapo_free_fields(m);
+
+	kfree(m);
+}
+
+/**
+ * pipapo_commit() - Replace lookup data with current working copy
+ * @set:	nftables API set representation
+ *
+ * While at it, check if we should perform garbage collection on the working
+ * copy before committing it for lookup, and don't replace the table if the
+ * working copy doesn't have pending changes.
+ *
+ * We also need to create a new working copy for subsequent insertions and
+ * deletions.
+ */
+static void pipapo_commit(const struct nft_set *set)
+{
+	struct nft_pipapo *priv = nft_set_priv(set);
+	struct nft_pipapo_match *new_clone, *old;
+
+	if (time_after_eq(jiffies, priv->last_gc + nft_set_gc_interval(set)))
+		pipapo_gc(set, priv->clone);
+
+	if (!priv->dirty)
+		return;
+
+	new_clone = pipapo_clone(priv->clone);
+	if (IS_ERR(new_clone))
+		return;
+
+	priv->dirty = false;
+
+	old = rcu_access_pointer(priv->match);
+	rcu_assign_pointer(priv->match, priv->clone);
+	if (old)
+		call_rcu(&old->rcu, pipapo_reclaim_match);
+
+	priv->clone = new_clone;
+}
+
+/**
+ * nft_pipapo_activate() - Mark element reference as active given key, commit
+ * @net:	Network namespace
+ * @set:	nftables API set representation
+ * @elem:	nftables API element representation containing key data
+ *
+ * On insertion, elements are added to a copy of the matching data currently
+ * in use for lookups, and not directly inserted into current lookup data, so
+ * we'll take care of that by calling pipapo_commit() here. Both
+ * nft_pipapo_insert() and nft_pipapo_activate() are called once for each
+ * element, hence we can't purpose either one as a real commit operation.
+ */
+static void nft_pipapo_activate(const struct net *net,
+				const struct nft_set *set,
+				const struct nft_set_elem *elem)
+{
+	struct nft_pipapo_elem *e;
+
+	e = pipapo_get(net, set, (const u8 *)elem->key.val.data, 0);
+	if (IS_ERR(e))
+		return;
+
+	nft_set_elem_change_active(net, set, &e->ext);
+	nft_set_elem_clear_busy(&e->ext);
+
+	pipapo_commit(set);
+}
+
+/**
+ * pipapo_deactivate() - Check that element is in set, mark as inactive
+ * @net:	Network namespace
+ * @set:	nftables API set representation
+ * @data:	Input key data
+ * @ext:	nftables API extension pointer, used to check for end element
+ *
+ * This is a convenience function that can be called from both
+ * nft_pipapo_deactivate() and nft_pipapo_flush(), as they are in fact the same
+ * operation.
+ *
+ * Return: deactivated element if found, NULL otherwise.
+ */
+static void *pipapo_deactivate(const struct net *net, const struct nft_set *set,
+			       const u8 *data, const struct nft_set_ext *ext)
+{
+	struct nft_pipapo_elem *e;
+
+	e = pipapo_get(net, set, data, nft_genmask_next(net));
+	if (IS_ERR(e))
+		return NULL;
+
+	nft_set_elem_change_active(net, set, &e->ext);
+
+	return e;
+}
+
+/**
+ * nft_pipapo_deactivate() - Call pipapo_deactivate() to make element inactive
+ * @net:	Network namespace
+ * @set:	nftables API set representation
+ * @elem:	nftables API element representation containing key data
+ *
+ * Return: deactivated element if found, NULL otherwise.
+ */
+static void *nft_pipapo_deactivate(const struct net *net,
+				   const struct nft_set *set,
+				   const struct nft_set_elem *elem)
+{
+	const struct nft_set_ext *ext = nft_set_elem_ext(set, elem->priv);
+
+	return pipapo_deactivate(net, set, (const u8 *)elem->key.val.data, ext);
+}
+
+/**
+ * nft_pipapo_flush() - Call pipapo_deactivate() to make element inactive
+ * @net:	Network namespace
+ * @set:	nftables API set representation
+ * @elem:	nftables API element representation containing key data
+ *
+ * This is functionally the same as nft_pipapo_deactivate(), with a slightly
+ * different interface, and it's also called once for each element in a set
+ * being flushed, so we can't implement, strictly speaking, a flush operation,
+ * which would otherwise be as simple as allocating an empty copy of the
+ * matching data.
+ *
+ * Note that we could in theory do that, mark the set as flushed, and ignore
+ * subsequent calls, but we would leak all the elements after the first one,
+ * because they wouldn't then be freed as result of API calls.
+ *
+ * Return: true if element was found and deactivated.
+ */
+static bool nft_pipapo_flush(const struct net *net, const struct nft_set *set,
+			     void *elem)
+{
+	struct nft_pipapo_elem *e = elem;
+
+	return pipapo_deactivate(net, set, (const u8 *)nft_set_ext_key(&e->ext),
+				 &e->ext);
+}
+
+/**
+ * pipapo_get_boundaries() - Get byte interval for associated rules
+ * @f:		Field including lookup table
+ * @first_rule:	First rule (lowest index)
+ * @rule_count:	Number of associated rules
+ * @left:	Byte expression for left boundary (start of range)
+ * @right:	Byte expression for right boundary (end of range)
+ *
+ * Given the first rule and amount of rules that originated from the same entry,
+ * build the original range associated with the entry, and calculate the length
+ * of the originating netmask.
+ *
+ * In pictures:
+ *
+ *                     bucket
+ *      group  0   1   2   3   4   5   6   7   8   9  10  11  12  13  14  15
+ *        0                                                   1,2
+ *        1   1,2
+ *        2                                           1,2
+ *        3                                   1,2
+ *        4   1,2
+ *        5        1   2
+ *        6   1,2  1   1   1   1   1   1   1   1   1   1   1   1   1   1   1
+ *        7   1,2 1,2  1   1   1   1   1   1   1   1   1   1   1   1   1   1
+ *
+ * this is the lookup table corresponding to the IPv4 range
+ * 192.168.1.0-192.168.2.1, which was expanded to the two composing netmasks,
+ * rule #1: 192.168.1.0/24, and rule #2: 192.168.2.0/31.
+ *
+ * This function fills @left and @right with the byte values of the leftmost
+ * and rightmost bucket indices for the lowest and highest rule indices,
+ * respectively. If @first_rule is 1 and @rule_count is 2, we obtain, in
+ * nibbles:
+ *   left:  < 12, 0, 10, 8, 0, 1, 0, 0 >
+ *   right: < 12, 0, 10, 8, 0, 2, 2, 1 >
+ * corresponding to bytes:
+ *   left:  < 192, 168, 1, 0 >
+ *   right: < 192, 168, 2, 1 >
+ * with mask length irrelevant here, unused on return, as the range is already
+ * defined by its start and end points. The mask length is relevant for a single
+ * ranged entry instead: if @first_rule is 1 and @rule_count is 1, we ignore
+ * rule 2 above: @left becomes < 192, 168, 1, 0 >, @right becomes
+ * < 192, 168, 1, 255 >, and the mask length, calculated from the distances
+ * between leftmost and rightmost bucket indices for each group, would be 24.
+ *
+ * Return: mask length, in bits.
+ */
+static int pipapo_get_boundaries(struct nft_pipapo_field *f, int first_rule,
+				 int rule_count, u8 *left, u8 *right)
+{
+	u8 *l = left, *r = right;
+	int g, mask_len = 0;
+
+	for (g = 0; g < f->groups; g++) {
+		int b, x0, x1;
+
+		x0 = -1;
+		x1 = -1;
+		for (b = 0; b < NFT_PIPAPO_BUCKETS; b++) {
+			unsigned long *pos;
+
+			pos = f->lt + (g * NFT_PIPAPO_BUCKETS + b) * f->bsize;
+			if (test_bit(first_rule, pos) && x0 == -1)
+				x0 = b;
+			if (test_bit(first_rule + rule_count - 1, pos))
+				x1 = b;
+		}
+
+		if (g % 2) {
+			*(l++) |= x0 & 0x0f;
+			*(r++) |= x1 & 0x0f;
+		} else {
+			*l |= x0 << 4;
+			*r |= x1 << 4;
+		}
+
+		if (x1 - x0 == 0)
+			mask_len += 4;
+		else if (x1 - x0 == 1)
+			mask_len += 3;
+		else if (x1 - x0 == 3)
+			mask_len += 2;
+		else if (x1 - x0 == 7)
+			mask_len += 1;
+	}
+
+	return mask_len;
+}
+
+/**
+ * pipapo_match_field() - Match rules against byte ranges
+ * @f:		Field including the lookup table
+ * @first_rule:	First of associated rules originating from same entry
+ * @rule_count:	Amount of associated rules
+ * @start:	Start of range to be matched
+ * @end:	End of range to be matched
+ *
+ * Return: true on match, false otherwise.
+ */
+static bool pipapo_match_field(struct nft_pipapo_field *f,
+			       int first_rule, int rule_count,
+			       const u8 *start, const u8 *end)
+{
+	u8 right[NFT_PIPAPO_MAX_BYTES] = { 0 };
+	u8 left[NFT_PIPAPO_MAX_BYTES] = { 0 };
+
+	pipapo_get_boundaries(f, first_rule, rule_count, left, right);
+
+	return !memcmp(start, left, f->groups / NFT_PIPAPO_GROUPS_PER_BYTE) &&
+	       !memcmp(end, right, f->groups / NFT_PIPAPO_GROUPS_PER_BYTE);
+}
+
+/**
+ * nft_pipapo_remove() - Remove element given key, commit
+ * @net:	Network namespace
+ * @set:	nftables API set representation
+ * @elem:	nftables API element representation containing key data
+ *
+ * Similarly to nft_pipapo_activate(), this is used as commit operation by the
+ * API, but it's called once per element in the pending transaction, so we can't
+ * implement this as a single commit operation. Closest we can get is to remove
+ * the matched element here, if any, and commit the updated matching data.
+ */
+static void nft_pipapo_remove(const struct net *net, const struct nft_set *set,
+			      const struct nft_set_elem *elem)
+{
+	const u8 *data = (const u8 *)elem->key.val.data;
+	struct nft_pipapo *priv = nft_set_priv(set);
+	struct nft_pipapo_match *m = priv->clone;
+	int rules_f0, first_rule = 0;
+	struct nft_pipapo_elem *e;
+
+	e = pipapo_get(net, set, data, 0);
+	if (IS_ERR(e))
+		return;
+
+	while ((rules_f0 = pipapo_rules_same_key(m->f, first_rule))) {
+		union nft_pipapo_map_bucket rulemap[NFT_PIPAPO_MAX_FIELDS];
+		const u8 *match_start, *match_end;
+		struct nft_pipapo_field *f;
+		int i, start, rules_fx;
+
+		match_start = data;
+		match_end = (const u8 *)nft_set_ext_key_end(&e->ext)->data;
+
+		start = first_rule;
+		rules_fx = rules_f0;
+
+		nft_pipapo_for_each_field(f, i, m) {
+			if (!pipapo_match_field(f, start, rules_fx,
+						match_start, match_end))
+				break;
+
+			rulemap[i].to = start;
+			rulemap[i].n = rules_fx;
+
+			rules_fx = f->mt[start].n;
+			start = f->mt[start].to;
+
+			match_start += NFT_PIPAPO_GROUPS_PADDED_SIZE(f->groups);
+			match_end += NFT_PIPAPO_GROUPS_PADDED_SIZE(f->groups);
+		}
+
+		if (i == m->field_count) {
+			priv->dirty = true;
+			pipapo_drop(m, rulemap);
+			pipapo_commit(set);
+			return;
+		}
+
+		first_rule += rules_f0;
+	}
+}
+
+/**
+ * nft_pipapo_walk() - Walk over elements
+ * @ctx:	nftables API context
+ * @set:	nftables API set representation
+ * @iter:	Iterator
+ *
+ * As elements are referenced in the mapping array for the last field, directly
+ * scan that array: there's no need to follow rule mappings from the first
+ * field.
+ */
+static void nft_pipapo_walk(const struct nft_ctx *ctx, struct nft_set *set,
+			    struct nft_set_iter *iter)
+{
+	struct nft_pipapo *priv = nft_set_priv(set);
+	struct nft_pipapo_match *m;
+	struct nft_pipapo_field *f;
+	int i, r;
+
+	rcu_read_lock();
+	m = rcu_dereference(priv->match);
+
+	if (unlikely(!m))
+		goto out;
+
+	for (i = 0, f = m->f; i < m->field_count - 1; i++, f++)
+		;
+
+	for (r = 0; r < f->rules; r++) {
+		struct nft_pipapo_elem *e;
+		struct nft_set_elem elem;
+
+		if (r < f->rules - 1 && f->mt[r + 1].e == f->mt[r].e)
+			continue;
+
+		if (iter->count < iter->skip)
+			goto cont;
+
+		e = f->mt[r].e;
+		if (nft_set_elem_expired(&e->ext))
+			goto cont;
+
+		elem.priv = e;
+
+		iter->err = iter->fn(ctx, set, iter, &elem);
+		if (iter->err < 0)
+			goto out;
+
+cont:
+		iter->count++;
+	}
+
+out:
+	rcu_read_unlock();
+}
+
+/**
+ * nft_pipapo_privsize() - Return the size of private data for the set
+ * @nla:	netlink attributes, ignored as size doesn't depend on them
+ * @desc:	Set description, ignored as size doesn't depend on it
+ *
+ * Return: size of private data for this set implementation, in bytes
+ */
+static u64 nft_pipapo_privsize(const struct nlattr * const nla[],
+			       const struct nft_set_desc *desc)
+{
+	return sizeof(struct nft_pipapo);
+}
+
+/**
+ * nft_pipapo_estimate() - Estimate set size, space and lookup complexity
+ * @desc:	Set description, element count and field description used here
+ * @features:	Flags: NFT_SET_INTERVAL needs to be there
+ * @est:	Storage for estimation data
+ *
+ * The size for this set type can vary dramatically, as it depends on the number
+ * of rules (composing netmasks) the entries expand to. We compute the worst
+ * case here.
+ *
+ * In general, for a non-ranged entry or a single composing netmask, we need
+ * one bit in each of the sixteen NFT_PIPAPO_BUCKETS, for each 4-bit group (that
+ * is, each input bit needs four bits of matching data), plus a bucket in the
+ * mapping table for each field.
+ *
+ * Return: true only for compatible range concatenations
+ */
+static bool nft_pipapo_estimate(const struct nft_set_desc *desc, u32 features,
+				struct nft_set_estimate *est)
+{
+	unsigned long entry_size;
+	int i;
+
+	if (!(features & NFT_SET_INTERVAL) || desc->field_count <= 1)
+		return false;
+
+	for (i = 0, entry_size = 0; i < desc->field_count; i++) {
+		unsigned long rules;
+
+		if (desc->field_len[i] > NFT_PIPAPO_MAX_BYTES)
+			return false;
+
+		/* Worst-case ranges for each concatenated field: each n-bit
+		 * field can expand to up to n * 2 rules in each bucket, and
+		 * each rule also needs a mapping bucket.
+		 */
+		rules = ilog2(desc->field_len[i] * BITS_PER_BYTE) * 2;
+		entry_size += rules * NFT_PIPAPO_BUCKETS / BITS_PER_BYTE;
+		entry_size += rules * sizeof(union nft_pipapo_map_bucket);
+	}
+
+	/* Rules in lookup and mapping tables are needed for each entry */
+	est->size = desc->size * entry_size;
+	if (est->size && div_u64(est->size, desc->size) != entry_size)
+		return false;
+
+	est->size += sizeof(struct nft_pipapo) +
+		     sizeof(struct nft_pipapo_match) * 2;
+
+	est->size += sizeof(struct nft_pipapo_field) * desc->field_count;
+
+	est->lookup = NFT_SET_CLASS_O_LOG_N;
+
+	est->space = NFT_SET_CLASS_O_N;
+
+	return true;
+}
+
+/**
+ * nft_pipapo_init() - Initialise data for a set instance
+ * @set:	nftables API set representation
+ * @desc:	Set description
+ * @nla:	netlink attributes
+ *
+ * Validate number and size of fields passed as NFTA_SET_DESC_CONCAT netlink
+ * attributes, initialise internal set parameters, current instance of matching
+ * data and a copy for subsequent insertions.
+ *
+ * Return: 0 on success, negative error code on failure.
+ */
+static int nft_pipapo_init(const struct nft_set *set,
+			   const struct nft_set_desc *desc,
+			   const struct nlattr * const nla[])
+{
+	struct nft_pipapo *priv = nft_set_priv(set);
+	struct nft_pipapo_match *m;
+	struct nft_pipapo_field *f;
+	int err, i;
+
+	if (desc->field_count > NFT_PIPAPO_MAX_FIELDS)
+		return -EINVAL;
+
+	m = kmalloc(sizeof(*priv->match) + sizeof(*f) * desc->field_count,
+		    GFP_KERNEL);
+	if (!m)
+		return -ENOMEM;
+
+	m->field_count = desc->field_count;
+	m->bsize_max = 0;
+
+	m->scratch = alloc_percpu(unsigned long *);
+	if (!m->scratch) {
+		err = -ENOMEM;
+		goto out_free;
+	}
+	for_each_possible_cpu(i)
+		*per_cpu_ptr(m->scratch, i) = NULL;
+
+	rcu_head_init(&m->rcu);
+
+	nft_pipapo_for_each_field(f, i, m) {
+		f->groups = desc->field_len[i] * NFT_PIPAPO_GROUPS_PER_BYTE;
+		priv->groups += f->groups;
+
+		priv->width += round_up(desc->field_len[i], sizeof(u32));
+
+		f->bsize = 0;
+		f->rules = 0;
+		f->lt = NULL;
+		f->mt = NULL;
+	}
+
+	/* Create an initial clone of matching data for next insertion */
+	priv->clone = pipapo_clone(m);
+	if (IS_ERR(priv->clone)) {
+		err = PTR_ERR(priv->clone);
+		goto out_free;
+	}
+
+	priv->dirty = false;
+
+	rcu_assign_pointer(priv->match, m);
+
+	return 0;
+
+out_free:
+	free_percpu(m->scratch);
+	kfree(m);
+
+	return err;
+}
+
+/**
+ * nft_pipapo_destroy() - Free private data for set and all committed elements
+ * @set:	nftables API set representation
+ */
+static void nft_pipapo_destroy(const struct nft_set *set)
+{
+	struct nft_pipapo *priv = nft_set_priv(set);
+	struct nft_pipapo_match *m;
+	struct nft_pipapo_field *f;
+	int i, r, cpu;
+
+	m = rcu_dereference_protected(priv->match, true);
+	if (m) {
+		rcu_barrier();
+
+		for (i = 0, f = m->f; i < m->field_count - 1; i++, f++)
+			;
+
+		for (r = 0; r < f->rules; r++) {
+			struct nft_pipapo_elem *e;
+
+			if (r < f->rules - 1 && f->mt[r + 1].e == f->mt[r].e)
+				continue;
+
+			e = f->mt[r].e;
+
+			nft_set_elem_destroy(set, e, true);
+		}
+
+		for_each_possible_cpu(cpu)
+			kfree(*per_cpu_ptr(m->scratch, cpu));
+		free_percpu(m->scratch);
+
+		pipapo_free_fields(m);
+		kfree(m);
+		priv->match = NULL;
+	}
+
+	if (priv->clone) {
+		for_each_possible_cpu(cpu)
+			kfree(*per_cpu_ptr(priv->clone->scratch, cpu));
+		free_percpu(priv->clone->scratch);
+
+		pipapo_free_fields(priv->clone);
+		kfree(priv->clone);
+		priv->clone = NULL;
+	}
+}
+
+/**
+ * nft_pipapo_gc_init() - Initialise garbage collection
+ * @set:	nftables API set representation
+ *
+ * Instead of actually setting up a periodic work for garbage collection, as
+ * this operation requires a swap of matching data with the working copy, we'll
+ * do that opportunistically with other commit operations if the interval is
+ * elapsed, so we just need to set the current jiffies timestamp here.
+ */
+static void nft_pipapo_gc_init(const struct nft_set *set)
+{
+	struct nft_pipapo *priv = nft_set_priv(set);
+
+	priv->last_gc = jiffies;
+}
+
+struct nft_set_type nft_set_pipapo_type __read_mostly = {
+	.owner		= THIS_MODULE,
+	.features	= NFT_SET_INTERVAL | NFT_SET_MAP | NFT_SET_OBJECT |
+			  NFT_SET_TIMEOUT,
+	.ops		= {
+		.lookup		= nft_pipapo_lookup,
+		.insert		= nft_pipapo_insert,
+		.activate	= nft_pipapo_activate,
+		.deactivate	= nft_pipapo_deactivate,
+		.flush		= nft_pipapo_flush,
+		.remove		= nft_pipapo_remove,
+		.walk		= nft_pipapo_walk,
+		.get		= nft_pipapo_get,
+		.privsize	= nft_pipapo_privsize,
+		.estimate	= nft_pipapo_estimate,
+		.init		= nft_pipapo_init,
+		.destroy	= nft_pipapo_destroy,
+		.gc_init	= nft_pipapo_gc_init,
+		.elemsize	= offsetof(struct nft_pipapo_elem, ext),
+	},
+};
-- 
cgit v1.2.3-59-g8ed1b


From 611973c1e06faef31d034deeb3ae7b7960b1f043 Mon Sep 17 00:00:00 2001
From: Stefano Brivio <sbrivio@redhat.com>
Date: Wed, 22 Jan 2020 00:17:56 +0100
Subject: selftests: netfilter: Introduce tests for sets with range
 concatenation

This test covers functionality and stability of the newly added
nftables set implementation supporting concatenation of ranged
fields.

For some selected set expression types, test:
- correctness, by checking that packets match or don't
- concurrency, by attempting races between insertion, deletion, lookup
- timeout feature, checking that packets don't match expired entries

and (roughly) estimate matching rates, comparing to baselines for
simple drop on netdev ingress hook and for hash and rbtrees sets.

In order to send packets, this needs one of sendip, netcat or bash.
To flood with traffic, iperf3, iperf and netperf are supported. For
performance measurements, this relies on the sample pktgen script
pktgen_bench_xmit_mode_netif_receive.sh.

If none of the tools suitable for a given test are available, specific
tests will be skipped.

Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 tools/testing/selftests/netfilter/Makefile         |    3 +-
 .../selftests/netfilter/nft_concat_range.sh        | 1481 ++++++++++++++++++++
 2 files changed, 1483 insertions(+), 1 deletion(-)
 create mode 100755 tools/testing/selftests/netfilter/nft_concat_range.sh

diff --git a/tools/testing/selftests/netfilter/Makefile b/tools/testing/selftests/netfilter/Makefile
index de1032b5ddea..08194aa44006 100644
--- a/tools/testing/selftests/netfilter/Makefile
+++ b/tools/testing/selftests/netfilter/Makefile
@@ -2,6 +2,7 @@
 # Makefile for netfilter selftests
 
 TEST_PROGS := nft_trans_stress.sh nft_nat.sh bridge_brouter.sh \
-	conntrack_icmp_related.sh nft_flowtable.sh ipvs.sh
+	conntrack_icmp_related.sh nft_flowtable.sh ipvs.sh \
+	nft_concat_range.sh
 
 include ../lib.mk
diff --git a/tools/testing/selftests/netfilter/nft_concat_range.sh b/tools/testing/selftests/netfilter/nft_concat_range.sh
new file mode 100755
index 000000000000..aca21dde102a
--- /dev/null
+++ b/tools/testing/selftests/netfilter/nft_concat_range.sh
@@ -0,0 +1,1481 @@
+#!/bin/sh
+# SPDX-License-Identifier: GPL-2.0
+#
+# nft_concat_range.sh - Tests for sets with concatenation of ranged fields
+#
+# Copyright (c) 2019 Red Hat GmbH
+#
+# Author: Stefano Brivio <sbrivio@redhat.com>
+#
+# shellcheck disable=SC2154,SC2034,SC2016,SC2030,SC2031
+# ^ Configuration and templates sourced with eval, counters reused in subshells
+
+KSELFTEST_SKIP=4
+
+# Available test groups:
+# - correctness: check that packets match given entries, and only those
+# - concurrency: attempt races between insertion, deletion and lookup
+# - timeout: check that packets match entries until they expire
+# - performance: estimate matching rate, compare with rbtree and hash baselines
+TESTS="correctness concurrency timeout"
+[ "${quicktest}" != "1" ] && TESTS="${TESTS} performance"
+
+# Set types, defined by TYPE_ variables below
+TYPES="net_port port_net net6_port port_proto net6_port_mac net6_port_mac_proto
+       net_port_net net_mac net_mac_icmp net6_mac_icmp net6_port_net6_port
+       net_port_mac_proto_net"
+
+# List of possible paths to pktgen script from kernel tree for performance tests
+PKTGEN_SCRIPT_PATHS="
+	../../../samples/pktgen/pktgen_bench_xmit_mode_netif_receive.sh
+	pktgen/pktgen_bench_xmit_mode_netif_receive.sh"
+
+# Definition of set types:
+# display	display text for test report
+# type_spec	nftables set type specifier
+# chain_spec	nftables type specifier for rules mapping to set
+# dst		call sequence of format_*() functions for destination fields
+# src		call sequence of format_*() functions for source fields
+# start		initial integer used to generate addresses and ports
+# count		count of entries to generate and match
+# src_delta	number summed to destination generator for source fields
+# tools		list of tools for correctness and timeout tests, any can be used
+# proto		L4 protocol of test packets
+#
+# race_repeat	race attempts per thread, 0 disables concurrency test for type
+# flood_tools	list of tools for concurrency tests, any can be used
+# flood_proto	L4 protocol of test packets for concurrency tests
+# flood_spec	nftables type specifier for concurrency tests
+#
+# perf_duration	duration of single pktgen injection test
+# perf_spec	nftables type specifier for performance tests
+# perf_dst	format_*() functions for destination fields in performance test
+# perf_src	format_*() functions for source fields in performance test
+# perf_entries	number of set entries for performance test
+# perf_proto	L3 protocol of test packets
+TYPE_net_port="
+display		net,port
+type_spec	ipv4_addr . inet_service
+chain_spec	ip daddr . udp dport
+dst		addr4 port
+src		 
+start		1
+count		5
+src_delta	2000
+tools		sendip nc bash
+proto		udp
+
+race_repeat	3
+flood_tools	iperf3 iperf netperf
+flood_proto	udp
+flood_spec	ip daddr . udp dport
+
+perf_duration	5
+perf_spec	ip daddr . udp dport
+perf_dst	addr4 port
+perf_src	 
+perf_entries	1000
+perf_proto	ipv4
+"
+
+TYPE_port_net="
+display		port,net
+type_spec	inet_service . ipv4_addr
+chain_spec	udp dport . ip daddr
+dst		port addr4
+src		 
+start		1
+count		5
+src_delta	2000
+tools		sendip nc bash
+proto		udp
+
+race_repeat	3
+flood_tools	iperf3 iperf netperf
+flood_proto	udp
+flood_spec	udp dport . ip daddr
+
+perf_duration	5
+perf_spec	udp dport . ip daddr
+perf_dst	port addr4
+perf_src	 
+perf_entries	100
+perf_proto	ipv4
+"
+
+TYPE_net6_port="
+display		net6,port
+type_spec	ipv6_addr . inet_service
+chain_spec	ip6 daddr . udp dport
+dst		addr6 port
+src		 
+start		10
+count		5
+src_delta	2000
+tools		sendip nc bash
+proto		udp6
+
+race_repeat	3
+flood_tools	iperf3 iperf netperf
+flood_proto	tcp6
+flood_spec	ip6 daddr . udp dport
+
+perf_duration	5
+perf_spec	ip6 daddr . udp dport
+perf_dst	addr6 port
+perf_src	 
+perf_entries	1000
+perf_proto	ipv6
+"
+
+TYPE_port_proto="
+display		port,proto
+type_spec	inet_service . inet_proto
+chain_spec	udp dport . meta l4proto
+dst		port proto
+src		 
+start		1
+count		5
+src_delta	2000
+tools		sendip nc bash
+proto		udp
+
+race_repeat	0
+
+perf_duration	5
+perf_spec	udp dport . meta l4proto
+perf_dst	port proto
+perf_src	 
+perf_entries	30000
+perf_proto	ipv4
+"
+
+TYPE_net6_port_mac="
+display		net6,port,mac
+type_spec	ipv6_addr . inet_service . ether_addr
+chain_spec	ip6 daddr . udp dport . ether saddr
+dst		addr6 port
+src		mac
+start		10
+count		5
+src_delta	2000
+tools		sendip nc bash
+proto		udp6
+
+race_repeat	0
+
+perf_duration	5
+perf_spec	ip6 daddr . udp dport . ether daddr
+perf_dst	addr6 port mac
+perf_src	 
+perf_entries	10
+perf_proto	ipv6
+"
+
+TYPE_net6_port_mac_proto="
+display		net6,port,mac,proto
+type_spec	ipv6_addr . inet_service . ether_addr . inet_proto
+chain_spec	ip6 daddr . udp dport . ether saddr . meta l4proto
+dst		addr6 port
+src		mac proto
+start		10
+count		5
+src_delta	2000
+tools		sendip nc bash
+proto		udp6
+
+race_repeat	0
+
+perf_duration	5
+perf_spec	ip6 daddr . udp dport . ether daddr . meta l4proto
+perf_dst	addr6 port mac proto
+perf_src	 
+perf_entries	1000
+perf_proto	ipv6
+"
+
+TYPE_net_port_net="
+display		net,port,net
+type_spec	ipv4_addr . inet_service . ipv4_addr
+chain_spec	ip daddr . udp dport . ip saddr
+dst		addr4 port
+src		addr4
+start		1
+count		5
+src_delta	2000
+tools		sendip nc bash
+proto		udp
+
+race_repeat	3
+flood_tools	iperf3 iperf netperf
+flood_proto	tcp
+flood_spec	ip daddr . udp dport . ip saddr
+
+perf_duration	0
+"
+
+TYPE_net6_port_net6_port="
+display		net6,port,net6,port
+type_spec	ipv6_addr . inet_service . ipv6_addr . inet_service
+chain_spec	ip6 daddr . udp dport . ip6 saddr . udp sport
+dst		addr6 port
+src		addr6 port
+start		10
+count		5
+src_delta	2000
+tools		sendip nc
+proto		udp6
+
+race_repeat	3
+flood_tools	iperf3 iperf netperf
+flood_proto	tcp6
+flood_spec	ip6 daddr . tcp dport . ip6 saddr . tcp sport
+
+perf_duration	0
+"
+
+TYPE_net_port_mac_proto_net="
+display		net,port,mac,proto,net
+type_spec	ipv4_addr . inet_service . ether_addr . inet_proto . ipv4_addr
+chain_spec	ip daddr . udp dport . ether saddr . meta l4proto . ip saddr
+dst		addr4 port
+src		mac proto addr4
+start		1
+count		5
+src_delta	2000
+tools		sendip nc bash
+proto		udp
+
+race_repeat	0
+
+perf_duration	0
+"
+
+TYPE_net_mac="
+display		net,mac
+type_spec	ipv4_addr . ether_addr
+chain_spec	ip daddr . ether saddr
+dst		addr4
+src		mac
+start		1
+count		5
+src_delta	2000
+tools		sendip nc bash
+proto		udp
+
+race_repeat	0
+
+perf_duration	5
+perf_spec	ip daddr . ether daddr
+perf_dst	addr4 mac
+perf_src	 
+perf_entries	1000
+perf_proto	ipv4
+"
+
+TYPE_net_mac_icmp="
+display		net,mac - ICMP
+type_spec	ipv4_addr . ether_addr
+chain_spec	ip daddr . ether saddr
+dst		addr4
+src		mac
+start		1
+count		5
+src_delta	2000
+tools		ping
+proto		icmp
+
+race_repeat	0
+
+perf_duration	0
+"
+
+TYPE_net6_mac_icmp="
+display		net6,mac - ICMPv6
+type_spec	ipv6_addr . ether_addr
+chain_spec	ip6 daddr . ether saddr
+dst		addr6
+src		mac
+start		10
+count		50
+src_delta	2000
+tools		ping
+proto		icmp6
+
+race_repeat	0
+
+perf_duration	0
+"
+
+TYPE_net_port_proto_net="
+display		net,port,proto,net
+type_spec	ipv4_addr . inet_service . inet_proto . ipv4_addr
+chain_spec	ip daddr . udp dport . meta l4proto . ip saddr
+dst		addr4 port proto
+src		addr4
+start		1
+count		5
+src_delta	2000
+tools		sendip nc
+proto		udp
+
+race_repeat	3
+flood_tools	iperf3 iperf netperf
+flood_proto	tcp
+flood_spec	ip daddr . tcp dport . meta l4proto . ip saddr
+
+perf_duration	0
+"
+
+# Set template for all tests, types and rules are filled in depending on test
+set_template='
+flush ruleset
+
+table inet filter {
+	counter test {
+		packets 0 bytes 0
+	}
+
+	set test {
+		type ${type_spec}
+		flags interval,timeout
+	}
+
+	chain input {
+		type filter hook prerouting priority 0; policy accept;
+		${chain_spec} @test counter name \"test\"
+	}
+}
+
+table netdev perf {
+	counter test {
+		packets 0 bytes 0
+	}
+
+	counter match {
+		packets 0 bytes 0
+	}
+
+	set test {
+		type ${type_spec}
+		flags interval
+	}
+
+	set norange {
+		type ${type_spec}
+	}
+
+	set noconcat {
+		type ${type_spec%% *}
+		flags interval
+	}
+
+	chain test {
+		type filter hook ingress device veth_a priority 0;
+	}
+}
+'
+
+err_buf=
+info_buf=
+
+# Append string to error buffer
+err() {
+	err_buf="${err_buf}${1}
+"
+}
+
+# Append string to information buffer
+info() {
+	info_buf="${info_buf}${1}
+"
+}
+
+# Flush error buffer to stdout
+err_flush() {
+	printf "%s" "${err_buf}"
+	err_buf=
+}
+
+# Flush information buffer to stdout
+info_flush() {
+	printf "%s" "${info_buf}"
+	info_buf=
+}
+
+# Setup veth pair: this namespace receives traffic, B generates it
+setup_veth() {
+	ip netns add B
+	ip link add veth_a type veth peer name veth_b || return 1
+
+	ip link set veth_a up
+	ip link set veth_b netns B
+
+	ip -n B link set veth_b up
+
+	ip addr add dev veth_a 10.0.0.1
+	ip route add default dev veth_a
+
+	ip -6 addr add fe80::1/64 dev veth_a nodad
+	ip -6 addr add 2001:db8::1/64 dev veth_a nodad
+	ip -6 route add default dev veth_a
+
+	ip -n B route add default dev veth_b
+
+	ip -6 -n B addr add fe80::2/64 dev veth_b nodad
+	ip -6 -n B addr add 2001:db8::2/64 dev veth_b nodad
+	ip -6 -n B route add default dev veth_b
+
+	B() {
+		ip netns exec B "$@" >/dev/null 2>&1
+	}
+
+	sleep 2
+}
+
+# Fill in set template and initialise set
+setup_set() {
+	eval "echo \"${set_template}\"" | nft -f -
+}
+
+# Check that at least one of the needed tools is available
+check_tools() {
+	__tools=
+	for tool in ${tools}; do
+		if [ "${tool}" = "nc" ] && [ "${proto}" = "udp6" ] && \
+		   ! nc -u -w0 1.1.1.1 1 2>/dev/null; then
+			# Some GNU netcat builds might not support IPv6
+			__tools="${__tools} netcat-openbsd"
+			continue
+		fi
+		__tools="${__tools} ${tool}"
+
+		command -v "${tool}" >/dev/null && return 0
+	done
+	err "need one of:${__tools}, skipping" && return 1
+}
+
+# Set up function to send ICMP packets
+setup_send_icmp() {
+	send_icmp() {
+		B ping -c1 -W1 "${dst_addr4}" >/dev/null 2>&1
+	}
+}
+
+# Set up function to send ICMPv6 packets
+setup_send_icmp6() {
+	if command -v ping6 >/dev/null; then
+		send_icmp6() {
+			ip -6 addr add "${dst_addr6}" dev veth_a nodad \
+				2>/dev/null
+			B ping6 -q -c1 -W1 "${dst_addr6}"
+		}
+	else
+		send_icmp6() {
+			ip -6 addr add "${dst_addr6}" dev veth_a nodad \
+				2>/dev/null
+			B ping -q -6 -c1 -W1 "${dst_addr6}"
+		}
+	fi
+}
+
+# Set up function to send single UDP packets on IPv4
+setup_send_udp() {
+	if command -v sendip >/dev/null; then
+		send_udp() {
+			[ -n "${src_port}" ] && src_port="-us ${src_port}"
+			[ -n "${dst_port}" ] && dst_port="-ud ${dst_port}"
+			[ -n "${src_addr4}" ] && src_addr4="-is ${src_addr4}"
+
+			# shellcheck disable=SC2086 # sendip needs split options
+			B sendip -p ipv4 -p udp ${src_addr4} ${src_port} \
+						${dst_port} "${dst_addr4}"
+
+			src_port=
+			dst_port=
+			src_addr4=
+		}
+	elif command -v nc >/dev/null; then
+		if nc -u -w0 1.1.1.1 1 2>/dev/null; then
+			# OpenBSD netcat
+			nc_opt="-w0"
+		else
+			# GNU netcat
+			nc_opt="-q0"
+		fi
+
+		send_udp() {
+			if [ -n "${src_addr4}" ]; then
+				B ip addr add "${src_addr4}" dev veth_b
+				__src_addr4="-s ${src_addr4}"
+			fi
+			ip addr add "${dst_addr4}" dev veth_a 2>/dev/null
+			[ -n "${src_port}" ] && src_port="-p ${src_port}"
+
+			echo "" | B nc -u "${nc_opt}" "${__src_addr4}" \
+				  "${src_port}" "${dst_addr4}" "${dst_port}"
+
+			src_addr4=
+			src_port=
+		}
+	elif [ -z "$(bash -c 'type -p')" ]; then
+		send_udp() {
+			ip addr add "${dst_addr4}" dev veth_a 2>/dev/null
+			if [ -n "${src_addr4}" ]; then
+				B ip addr add "${src_addr4}/16" dev veth_b
+				B ip route add default dev veth_b
+			fi
+
+			B bash -c "echo > /dev/udp/${dst_addr4}/${dst_port}"
+
+			if [ -n "${src_addr4}" ]; then
+				B ip addr del "${src_addr4}/16" dev veth_b
+			fi
+			src_addr4=
+		}
+	else
+		return 1
+	fi
+}
+
+# Set up function to send single UDP packets on IPv6
+setup_send_udp6() {
+	if command -v sendip >/dev/null; then
+		send_udp6() {
+			[ -n "${src_port}" ] && src_port="-us ${src_port}"
+			[ -n "${dst_port}" ] && dst_port="-ud ${dst_port}"
+			if [ -n "${src_addr6}" ]; then
+				src_addr6="-6s ${src_addr6}"
+			else
+				src_addr6="-6s 2001:db8::2"
+			fi
+			ip -6 addr add "${dst_addr6}" dev veth_a nodad \
+				2>/dev/null
+
+			# shellcheck disable=SC2086 # this needs split options
+			B sendip -p ipv6 -p udp ${src_addr6} ${src_port} \
+						${dst_port} "${dst_addr6}"
+
+			src_port=
+			dst_port=
+			src_addr6=
+		}
+	elif command -v nc >/dev/null && nc -u -w0 1.1.1.1 1 2>/dev/null; then
+		# GNU netcat might not work with IPv6, try next tool
+		send_udp6() {
+			ip -6 addr add "${dst_addr6}" dev veth_a nodad \
+				2>/dev/null
+			if [ -n "${src_addr6}" ]; then
+				B ip addr add "${src_addr6}" dev veth_b nodad
+			else
+				src_addr6="2001:db8::2"
+			fi
+			[ -n "${src_port}" ] && src_port="-p ${src_port}"
+
+			# shellcheck disable=SC2086 # this needs split options
+			echo "" | B nc -u w0 "-s${src_addr6}" ${src_port} \
+					       ${dst_addr6} ${dst_port}
+
+			src_addr6=
+			src_port=
+		}
+	elif [ -z "$(bash -c 'type -p')" ]; then
+		send_udp6() {
+			ip -6 addr add "${dst_addr6}" dev veth_a nodad \
+				2>/dev/null
+			B ip addr add "${src_addr6}" dev veth_b nodad
+			B bash -c "echo > /dev/udp/${dst_addr6}/${dst_port}"
+			ip -6 addr del "${dst_addr6}" dev veth_a 2>/dev/null
+		}
+	else
+		return 1
+	fi
+}
+
+# Set up function to send TCP traffic on IPv4
+setup_flood_tcp() {
+	if command -v iperf3 >/dev/null; then
+		flood_tcp() {
+			[ -n "${dst_port}" ] && dst_port="-p ${dst_port}"
+			if [ -n "${src_addr4}" ]; then
+				B ip addr add "${src_addr4}/16" dev veth_b
+				src_addr4="-B ${src_addr4}"
+			else
+				B ip addr add dev veth_b 10.0.0.2
+				src_addr4="-B 10.0.0.2"
+			fi
+			if [ -n "${src_port}" ]; then
+				src_port="--cport ${src_port}"
+			fi
+			B ip route add default dev veth_b 2>/dev/null
+			ip addr add "${dst_addr4}" dev veth_a 2>/dev/null
+
+			# shellcheck disable=SC2086 # this needs split options
+			iperf3 -s -DB "${dst_addr4}" ${dst_port} >/dev/null 2>&1
+			sleep 2
+
+			# shellcheck disable=SC2086 # this needs split options
+			B iperf3 -c "${dst_addr4}" ${dst_port} ${src_port} \
+				${src_addr4} -l16 -t 1000
+
+			src_addr4=
+			src_port=
+			dst_port=
+		}
+	elif command -v iperf >/dev/null; then
+		flood_tcp() {
+			[ -n "${dst_port}" ] && dst_port="-p ${dst_port}"
+			if [ -n "${src_addr4}" ]; then
+				B ip addr add "${src_addr4}/16" dev veth_b
+				src_addr4="-B ${src_addr4}"
+			else
+				B ip addr add dev veth_b 10.0.0.2 2>/dev/null
+				src_addr4="-B 10.0.0.2"
+			fi
+			if [ -n "${src_port}" ]; then
+				src_addr4="${src_addr4}:${src_port}"
+			fi
+			B ip route add default dev veth_b
+			ip addr add "${dst_addr4}" dev veth_a 2>/dev/null
+
+			# shellcheck disable=SC2086 # this needs split options
+			iperf -s -DB "${dst_addr4}" ${dst_port} >/dev/null 2>&1
+			sleep 2
+
+			# shellcheck disable=SC2086 # this needs split options
+			B iperf -c "${dst_addr4}" ${dst_port} ${src_addr4} \
+				-l20 -t 1000
+
+			src_addr4=
+			src_port=
+			dst_port=
+		}
+	elif command -v netperf >/dev/null; then
+		flood_tcp() {
+			[ -n "${dst_port}" ] && dst_port="-p ${dst_port}"
+			if [ -n "${src_addr4}" ]; then
+				B ip addr add "${src_addr4}/16" dev veth_b
+			else
+				B ip addr add dev veth_b 10.0.0.2
+				src_addr4="10.0.0.2"
+			fi
+			if [ -n "${src_port}" ]; then
+				dst_port="${dst_port},${src_port}"
+			fi
+			B ip route add default dev veth_b
+			ip addr add "${dst_addr4}" dev veth_a 2>/dev/null
+
+			# shellcheck disable=SC2086 # this needs split options
+			netserver -4 ${dst_port} -L "${dst_addr4}" \
+				>/dev/null 2>&1
+			sleep 2
+
+			# shellcheck disable=SC2086 # this needs split options
+			B netperf -4 -H "${dst_addr4}" ${dst_port} \
+				-L "${src_addr4}" -l 1000 -t TCP_STREAM
+
+			src_addr4=
+			src_port=
+			dst_port=
+		}
+	else
+		return 1
+	fi
+}
+
+# Set up function to send TCP traffic on IPv6
+setup_flood_tcp6() {
+	if command -v iperf3 >/dev/null; then
+		flood_tcp6() {
+			[ -n "${dst_port}" ] && dst_port="-p ${dst_port}"
+			if [ -n "${src_addr6}" ]; then
+				B ip addr add "${src_addr6}" dev veth_b nodad
+				src_addr6="-B ${src_addr6}"
+			else
+				src_addr6="-B 2001:db8::2"
+			fi
+			if [ -n "${src_port}" ]; then
+				src_port="--cport ${src_port}"
+			fi
+			B ip route add default dev veth_b
+			ip -6 addr add "${dst_addr6}" dev veth_a nodad \
+				2>/dev/null
+
+			# shellcheck disable=SC2086 # this needs split options
+			iperf3 -s -DB "${dst_addr6}" ${dst_port} >/dev/null 2>&1
+			sleep 2
+
+			# shellcheck disable=SC2086 # this needs split options
+			B iperf3 -c "${dst_addr6}" ${dst_port} \
+				${src_port} ${src_addr6} -l16 -t 1000
+
+			src_addr6=
+			src_port=
+			dst_port=
+		}
+	elif command -v iperf >/dev/null; then
+		flood_tcp6() {
+			[ -n "${dst_port}" ] && dst_port="-p ${dst_port}"
+			if [ -n "${src_addr6}" ]; then
+				B ip addr add "${src_addr6}" dev veth_b nodad
+				src_addr6="-B ${src_addr6}"
+			else
+				src_addr6="-B 2001:db8::2"
+			fi
+			if [ -n "${src_port}" ]; then
+				src_addr6="${src_addr6}:${src_port}"
+			fi
+			B ip route add default dev veth_b
+			ip -6 addr add "${dst_addr6}" dev veth_a nodad \
+				2>/dev/null
+
+			# shellcheck disable=SC2086 # this needs split options
+			iperf -s -VDB "${dst_addr6}" ${dst_port} >/dev/null 2>&1
+			sleep 2
+
+			# shellcheck disable=SC2086 # this needs split options
+			B iperf -c "${dst_addr6}" -V ${dst_port} \
+				${src_addr6} -l1 -t 1000
+
+			src_addr6=
+			src_port=
+			dst_port=
+		}
+	elif command -v netperf >/dev/null; then
+		flood_tcp6() {
+			[ -n "${dst_port}" ] && dst_port="-p ${dst_port}"
+			if [ -n "${src_addr6}" ]; then
+				B ip addr add "${src_addr6}" dev veth_b nodad
+			else
+				src_addr6="2001:db8::2"
+			fi
+			if [ -n "${src_port}" ]; then
+				dst_port="${dst_port},${src_port}"
+			fi
+			B ip route add default dev veth_b
+			ip -6 addr add "${dst_addr6}" dev veth_a nodad \
+				2>/dev/null
+
+			# shellcheck disable=SC2086 # this needs split options
+			netserver -6 ${dst_port} -L "${dst_addr6}" \
+				>/dev/null 2>&1
+			sleep 2
+
+			# shellcheck disable=SC2086 # this needs split options
+			B netperf -6 -H "${dst_addr6}" ${dst_port} \
+				-L "${src_addr6}" -l 1000 -t TCP_STREAM
+
+			src_addr6=
+			src_port=
+			dst_port=
+		}
+	else
+		return 1
+	fi
+}
+
+# Set up function to send UDP traffic on IPv4
+setup_flood_udp() {
+	if command -v iperf3 >/dev/null; then
+		flood_udp() {
+			[ -n "${dst_port}" ] && dst_port="-p ${dst_port}"
+			if [ -n "${src_addr4}" ]; then
+				B ip addr add "${src_addr4}/16" dev veth_b
+				src_addr4="-B ${src_addr4}"
+			else
+				B ip addr add dev veth_b 10.0.0.2 2>/dev/null
+				src_addr4="-B 10.0.0.2"
+			fi
+			if [ -n "${src_port}" ]; then
+				src_port="--cport ${src_port}"
+			fi
+			B ip route add default dev veth_b
+			ip addr add "${dst_addr4}" dev veth_a 2>/dev/null
+
+			# shellcheck disable=SC2086 # this needs split options
+			iperf3 -s -DB "${dst_addr4}" ${dst_port}
+			sleep 2
+
+			# shellcheck disable=SC2086 # this needs split options
+			B iperf3 -u -c "${dst_addr4}" -Z -b 100M -l16 -t1000 \
+				${dst_port} ${src_port} ${src_addr4}
+
+			src_addr4=
+			src_port=
+			dst_port=
+		}
+	elif command -v iperf >/dev/null; then
+		flood_udp() {
+			[ -n "${dst_port}" ] && dst_port="-p ${dst_port}"
+			if [ -n "${src_addr4}" ]; then
+				B ip addr add "${src_addr4}/16" dev veth_b
+				src_addr4="-B ${src_addr4}"
+			else
+				B ip addr add dev veth_b 10.0.0.2
+				src_addr4="-B 10.0.0.2"
+			fi
+			if [ -n "${src_port}" ]; then
+				src_addr4="${src_addr4}:${src_port}"
+			fi
+			B ip route add default dev veth_b
+			ip addr add "${dst_addr4}" dev veth_a 2>/dev/null
+
+			# shellcheck disable=SC2086 # this needs split options
+			iperf -u -sDB "${dst_addr4}" ${dst_port} >/dev/null 2>&1
+			sleep 2
+
+			# shellcheck disable=SC2086 # this needs split options
+			B iperf -u -c "${dst_addr4}" -b 100M -l1 -t1000 \
+				${dst_port} ${src_addr4}
+
+			src_addr4=
+			src_port=
+			dst_port=
+		}
+	elif command -v netperf >/dev/null; then
+		flood_udp() {
+			[ -n "${dst_port}" ] && dst_port="-p ${dst_port}"
+			if [ -n "${src_addr4}" ]; then
+				B ip addr add "${src_addr4}/16" dev veth_b
+			else
+				B ip addr add dev veth_b 10.0.0.2
+				src_addr4="10.0.0.2"
+			fi
+			if [ -n "${src_port}" ]; then
+				dst_port="${dst_port},${src_port}"
+			fi
+			B ip route add default dev veth_b
+			ip addr add "${dst_addr4}" dev veth_a 2>/dev/null
+
+			# shellcheck disable=SC2086 # this needs split options
+			netserver -4 ${dst_port} -L "${dst_addr4}" \
+				>/dev/null 2>&1
+			sleep 2
+
+			# shellcheck disable=SC2086 # this needs split options
+			B netperf -4 -H "${dst_addr4}" ${dst_port} \
+				-L "${src_addr4}" -l 1000 -t UDP_STREAM
+
+			src_addr4=
+			src_port=
+			dst_port=
+		}
+	else
+		return 1
+	fi
+}
+
+# Find pktgen script and set up function to start pktgen injection
+setup_perf() {
+	for pktgen_script_path in ${PKTGEN_SCRIPT_PATHS} __notfound; do
+		command -v "${pktgen_script_path}" >/dev/null && break
+	done
+	[ "${pktgen_script_path}" = "__notfound" ] && return 1
+
+	perf_ipv4() {
+		${pktgen_script_path} -s80 \
+			-i veth_a -d "${dst_addr4}" -p "${dst_port}" \
+			-m "${dst_mac}" \
+			-t $(($(nproc) / 5 + 1)) -b10000 -n0 2>/dev/null &
+		perf_pid=$!
+	}
+	perf_ipv6() {
+		IP6=6 ${pktgen_script_path} -s100 \
+			-i veth_a -d "${dst_addr6}" -p "${dst_port}" \
+			-m "${dst_mac}" \
+			-t $(($(nproc) / 5 + 1)) -b10000 -n0 2>/dev/null &
+		perf_pid=$!
+	}
+}
+
+# Clean up before each test
+cleanup() {
+	nft reset counter inet filter test	>/dev/null 2>&1
+	nft flush ruleset			>/dev/null 2>&1
+	ip link del dummy0			2>/dev/null
+	ip route del default			2>/dev/null
+	ip -6 route del default			2>/dev/null
+	ip netns del B				2>/dev/null
+	ip link del veth_a			2>/dev/null
+	timeout=
+	killall iperf3				2>/dev/null
+	killall iperf				2>/dev/null
+	killall netperf				2>/dev/null
+	killall netserver			2>/dev/null
+	rm -f ${tmp}
+	sleep 2
+}
+
+# Entry point for setup functions
+setup() {
+	if [ "$(id -u)" -ne 0 ]; then
+		echo "  need to run as root"
+		exit ${KSELFTEST_SKIP}
+	fi
+
+	cleanup
+	check_tools || return 1
+	for arg do
+		if ! eval setup_"${arg}"; then
+			err "  ${arg} not supported"
+			return 1
+		fi
+	done
+}
+
+# Format integer into IPv4 address, summing 10.0.0.5 (arbitrary) to it
+format_addr4() {
+	a=$((${1} + 16777216 * 10 + 5))
+	printf "%i.%i.%i.%i"						\
+	       "$((a / 16777216))" "$((a % 16777216 / 65536))"	\
+	       "$((a % 65536 / 256))" "$((a % 256))"
+}
+
+# Format integer into IPv6 address, summing 2001:db8:: to it
+format_addr6() {
+	printf "2001:db8::%04x:%04x" "$((${1} / 65536))" "$((${1} % 65536))"
+}
+
+# Format integer into EUI-48 address, summing 00:01:00:00:00:00 to it
+format_mac() {
+	printf "00:01:%02x:%02x:%02x:%02x" \
+	       "$((${1} / 16777216))" "$((${1} % 16777216 / 65536))"	\
+	       "$((${1} % 65536 / 256))" "$((${1} % 256))"
+}
+
+# Format integer into port, avoid 0 port
+format_port() {
+	printf "%i" "$((${1} % 65534 + 1))"
+}
+
+# Drop suffixed '6' from L4 protocol, if any
+format_proto() {
+	printf "%s" "${proto}" | tr -d 6
+}
+
+# Format destination and source fields into nft concatenated type
+format() {
+	__start=
+	__end=
+	__expr="{ "
+
+	for f in ${dst}; do
+		[ "${__expr}" != "{ " ] && __expr="${__expr} . "
+
+		__start="$(eval format_"${f}" "${start}")"
+		__end="$(eval format_"${f}" "${end}")"
+
+		if [ "${f}" = "proto" ]; then
+			__expr="${__expr}${__start}"
+		else
+			__expr="${__expr}${__start}-${__end}"
+		fi
+	done
+	for f in ${src}; do
+		__expr="${__expr} . "
+		__start="$(eval format_"${f}" "${srcstart}")"
+		__end="$(eval format_"${f}" "${srcend}")"
+
+		if [ "${f}" = "proto" ]; then
+			__expr="${__expr}${__start}"
+		else
+			__expr="${__expr}${__start}-${__end}"
+		fi
+	done
+
+	if [ -n "${timeout}" ]; then
+		echo "${__expr} timeout ${timeout}s }"
+	else
+		echo "${__expr} }"
+	fi
+}
+
+# Format destination and source fields into nft type, start element only
+format_norange() {
+	__expr="{ "
+
+	for f in ${dst}; do
+		[ "${__expr}" != "{ " ] && __expr="${__expr} . "
+
+		__expr="${__expr}$(eval format_"${f}" "${start}")"
+	done
+	for f in ${src}; do
+		__expr="${__expr} . $(eval format_"${f}" "${start}")"
+	done
+
+	echo "${__expr} }"
+}
+
+# Format first destination field into nft type
+format_noconcat() {
+	for f in ${dst}; do
+		__start="$(eval format_"${f}" "${start}")"
+		__end="$(eval format_"${f}" "${end}")"
+
+		if [ "${f}" = "proto" ]; then
+			echo "{ ${__start} }"
+		else
+			echo "{ ${__start}-${__end} }"
+		fi
+		return
+	done
+}
+
+# Add single entry to 'test' set in 'inet filter' table
+add() {
+	if ! nft add element inet filter test "${1}"; then
+		err "Failed to add ${1} given ruleset:"
+		err "$(nft list ruleset -a)"
+		return 1
+	fi
+}
+
+# Format and output entries for sets in 'netdev perf' table
+add_perf() {
+	if [ "${1}" = "test" ]; then
+		echo "add element netdev perf test $(format)"
+	elif [ "${1}" = "norange" ]; then
+		echo "add element netdev perf norange $(format_norange)"
+	elif [ "${1}" = "noconcat" ]; then
+		echo "add element netdev perf noconcat $(format_noconcat)"
+	fi
+}
+
+# Add single entry to 'norange' set in 'netdev perf' table
+add_perf_norange() {
+	if ! nft add element netdev perf norange "${1}"; then
+		err "Failed to add ${1} given ruleset:"
+		err "$(nft list ruleset -a)"
+		return 1
+	fi
+}
+
+# Add single entry to 'noconcat' set in 'netdev perf' table
+add_perf_noconcat() {
+	if ! nft add element netdev perf noconcat "${1}"; then
+		err "Failed to add ${1} given ruleset:"
+		err "$(nft list ruleset -a)"
+		return 1
+	fi
+}
+
+# Delete single entry from set
+del() {
+	if ! nft delete element inet filter test "${1}"; then
+		err "Failed to delete ${1} given ruleset:"
+		err "$(nft list ruleset -a)"
+		return 1
+	fi
+}
+
+# Return packet count from 'test' counter in 'inet filter' table
+count_packets() {
+	found=0
+	for token in $(nft list counter inet filter test); do
+		[ ${found} -eq 1 ] && echo "${token}" && return
+		[ "${token}" = "packets" ] && found=1
+	done
+}
+
+# Return packet count from 'test' counter in 'netdev perf' table
+count_perf_packets() {
+	found=0
+	for token in $(nft list counter netdev perf test); do
+		[ ${found} -eq 1 ] && echo "${token}" && return
+		[ "${token}" = "packets" ] && found=1
+	done
+}
+
+# Set MAC addresses, send traffic according to specifier
+flood() {
+	ip link set veth_a address "$(format_mac "${1}")"
+	ip -n B link set veth_b address "$(format_mac "${2}")"
+
+	for f in ${dst}; do
+		eval dst_"$f"=\$\(format_\$f "${1}"\)
+	done
+	for f in ${src}; do
+		eval src_"$f"=\$\(format_\$f "${2}"\)
+	done
+	eval flood_\$proto
+}
+
+# Set MAC addresses, start pktgen injection
+perf() {
+	dst_mac="$(format_mac "${1}")"
+	ip link set veth_a address "${dst_mac}"
+
+	for f in ${dst}; do
+		eval dst_"$f"=\$\(format_\$f "${1}"\)
+	done
+	for f in ${src}; do
+		eval src_"$f"=\$\(format_\$f "${2}"\)
+	done
+	eval perf_\$perf_proto
+}
+
+# Set MAC addresses, send single packet, check that it matches, reset counter
+send_match() {
+	ip link set veth_a address "$(format_mac "${1}")"
+	ip -n B link set veth_b address "$(format_mac "${2}")"
+
+	for f in ${dst}; do
+		eval dst_"$f"=\$\(format_\$f "${1}"\)
+	done
+	for f in ${src}; do
+		eval src_"$f"=\$\(format_\$f "${2}"\)
+	done
+	eval send_\$proto
+	if [ "$(count_packets)" != "1" ]; then
+		err "${proto} packet to:"
+		err "  $(for f in ${dst}; do
+			 eval format_\$f "${1}"; printf ' '; done)"
+		err "from:"
+		err "  $(for f in ${src}; do
+			 eval format_\$f "${2}"; printf ' '; done)"
+		err "should have matched ruleset:"
+		err "$(nft list ruleset -a)"
+		return 1
+	fi
+	nft reset counter inet filter test >/dev/null
+}
+
+# Set MAC addresses, send single packet, check that it doesn't match
+send_nomatch() {
+	ip link set veth_a address "$(format_mac "${1}")"
+	ip -n B link set veth_b address "$(format_mac "${2}")"
+
+	for f in ${dst}; do
+		eval dst_"$f"=\$\(format_\$f "${1}"\)
+	done
+	for f in ${src}; do
+		eval src_"$f"=\$\(format_\$f "${2}"\)
+	done
+	eval send_\$proto
+	if [ "$(count_packets)" != "0" ]; then
+		err "${proto} packet to:"
+		err "  $(for f in ${dst}; do
+			 eval format_\$f "${1}"; printf ' '; done)"
+		err "from:"
+		err "  $(for f in ${src}; do
+			 eval format_\$f "${2}"; printf ' '; done)"
+		err "should not have matched ruleset:"
+		err "$(nft list ruleset -a)"
+		return 1
+	fi
+}
+
+# Correctness test template:
+# - add ranged element, check that packets match it
+# - check that packets outside range don't match it
+# - remove some elements, check that packets don't match anymore
+test_correctness() {
+	setup veth send_"${proto}" set || return ${KSELFTEST_SKIP}
+
+	range_size=1
+	for i in $(seq "${start}" $((start + count))); do
+		end=$((start + range_size))
+
+		# Avoid negative or zero-sized port ranges
+		if [ $((end / 65534)) -gt $((start / 65534)) ]; then
+			start=${end}
+			end=$((end + 1))
+		fi
+		srcstart=$((start + src_delta))
+		srcend=$((end + src_delta))
+
+		add "$(format)" || return 1
+		for j in $(seq ${start} $((range_size / 2 + 1)) ${end}); do
+			send_match "${j}" $((j + src_delta)) || return 1
+		done
+		send_nomatch $((end + 1)) $((end + 1 + src_delta)) || return 1
+
+		# Delete elements now and then
+		if [ $((i % 3)) -eq 0 ]; then
+			del "$(format)" || return 1
+			for j in $(seq ${start} \
+				   $((range_size / 2 + 1)) ${end}); do
+				send_nomatch "${j}" $((j + src_delta)) \
+					|| return 1
+			done
+		fi
+
+		range_size=$((range_size + 1))
+		start=$((end + range_size))
+	done
+}
+
+# Concurrency test template:
+# - add all the elements
+# - start a thread for each physical thread that:
+#   - adds all the elements
+#   - flushes the set
+#   - adds all the elements
+#   - flushes the entire ruleset
+#   - adds the set back
+#   - adds all the elements
+#   - delete all the elements
+test_concurrency() {
+	proto=${flood_proto}
+	tools=${flood_tools}
+	chain_spec=${flood_spec}
+	setup veth flood_"${proto}" set || return ${KSELFTEST_SKIP}
+
+	range_size=1
+	cstart=${start}
+	flood_pids=
+	for i in $(seq ${start} $((start + count))); do
+		end=$((start + range_size))
+		srcstart=$((start + src_delta))
+		srcend=$((end + src_delta))
+
+		add "$(format)" || return 1
+
+		flood "${i}" $((i + src_delta)) & flood_pids="${flood_pids} $!"
+
+		range_size=$((range_size + 1))
+		start=$((end + range_size))
+	done
+
+	sleep 10
+
+	pids=
+	for c in $(seq 1 "$(nproc)"); do (
+		for r in $(seq 1 "${race_repeat}"); do
+			range_size=1
+
+			# $start needs to be local to this subshell
+			# shellcheck disable=SC2030
+			start=${cstart}
+			for i in $(seq ${start} $((start + count))); do
+				end=$((start + range_size))
+				srcstart=$((start + src_delta))
+				srcend=$((end + src_delta))
+
+				add "$(format)" 2>/dev/null
+
+				range_size=$((range_size + 1))
+				start=$((end + range_size))
+			done
+
+			nft flush inet filter test 2>/dev/null
+
+			range_size=1
+			start=${cstart}
+			for i in $(seq ${start} $((start + count))); do
+				end=$((start + range_size))
+				srcstart=$((start + src_delta))
+				srcend=$((end + src_delta))
+
+				add "$(format)" 2>/dev/null
+
+				range_size=$((range_size + 1))
+				start=$((end + range_size))
+			done
+
+			nft flush ruleset
+			setup set 2>/dev/null
+
+			range_size=1
+			start=${cstart}
+			for i in $(seq ${start} $((start + count))); do
+				end=$((start + range_size))
+				srcstart=$((start + src_delta))
+				srcend=$((end + src_delta))
+
+				add "$(format)" 2>/dev/null
+
+				range_size=$((range_size + 1))
+				start=$((end + range_size))
+			done
+
+			range_size=1
+			start=${cstart}
+			for i in $(seq ${start} $((start + count))); do
+				end=$((start + range_size))
+				srcstart=$((start + src_delta))
+				srcend=$((end + src_delta))
+
+				del "$(format)" 2>/dev/null
+
+				range_size=$((range_size + 1))
+				start=$((end + range_size))
+			done
+		done
+	) & pids="${pids} $!"
+	done
+
+	# shellcheck disable=SC2046,SC2086 # word splitting wanted here
+	wait $(for pid in ${pids}; do echo ${pid}; done)
+	# shellcheck disable=SC2046,SC2086
+	kill $(for pid in ${flood_pids}; do echo ${pid}; done) 2>/dev/null
+	# shellcheck disable=SC2046,SC2086
+	wait $(for pid in ${flood_pids}; do echo ${pid}; done) 2>/dev/null
+
+	return 0
+}
+
+# Timeout test template:
+# - add all the elements with 3s timeout while checking that packets match
+# - wait 3s after the last insertion, check that packets don't match any entry
+test_timeout() {
+	setup veth send_"${proto}" set || return ${KSELFTEST_SKIP}
+
+	timeout=3
+	range_size=1
+	for i in $(seq "${start}" $((start + count))); do
+		end=$((start + range_size))
+		srcstart=$((start + src_delta))
+		srcend=$((end + src_delta))
+
+		add "$(format)" || return 1
+
+		for j in $(seq ${start} $((range_size / 2 + 1)) ${end}); do
+			send_match "${j}" $((j + src_delta)) || return 1
+		done
+
+		range_size=$((range_size + 1))
+		start=$((end + range_size))
+	done
+	sleep 3
+	for i in $(seq ${start} $((start + count))); do
+		end=$((start + range_size))
+		srcstart=$((start + src_delta))
+		srcend=$((end + src_delta))
+
+		for j in $(seq ${start} $((range_size / 2 + 1)) ${end}); do
+			send_nomatch "${j}" $((j + src_delta)) || return 1
+		done
+
+		range_size=$((range_size + 1))
+		start=$((end + range_size))
+	done
+}
+
+# Performance test template:
+# - add concatenated ranged entries
+# - add non-ranged concatenated entries (for hash set matching rate baseline)
+# - add ranged entries with first field only (for rbhash baseline)
+# - start pktgen injection directly on device rx path of this namespace
+# - measure drop only rate, hash and rbtree baselines, then matching rate
+test_performance() {
+	chain_spec=${perf_spec}
+	dst="${perf_dst}"
+	src="${perf_src}"
+	setup veth perf set || return ${KSELFTEST_SKIP}
+
+	first=${start}
+	range_size=1
+	for set in test norange noconcat; do
+		start=${first}
+		for i in $(seq ${start} $((start + perf_entries))); do
+			end=$((start + range_size))
+			srcstart=$((start + src_delta))
+			srcend=$((end + src_delta))
+
+			if [ $((end / 65534)) -gt $((start / 65534)) ]; then
+				start=${end}
+				end=$((end + 1))
+			elif [ ${start} -eq ${end} ]; then
+				end=$((start + 1))
+			fi
+
+			add_perf ${set}
+
+			start=$((end + range_size))
+		done > "${tmp}"
+		nft -f "${tmp}"
+	done
+
+	perf $((end - 1)) ${srcstart}
+
+	sleep 2
+
+	nft add rule netdev perf test counter name \"test\" drop
+	nft reset counter netdev perf test >/dev/null 2>&1
+	sleep "${perf_duration}"
+	pps="$(printf %10s $(($(count_perf_packets) / perf_duration)))"
+	info "    baseline (drop from netdev hook):            ${pps}pps"
+	handle="$(nft -a list chain netdev perf test | grep counter)"
+	handle="${handle##* }"
+	nft delete rule netdev perf test handle "${handle}"
+
+	nft add rule "netdev perf test ${chain_spec} @norange \
+		counter name \"test\" drop"
+	nft reset counter netdev perf test >/dev/null 2>&1
+	sleep "${perf_duration}"
+	pps="$(printf %10s $(($(count_perf_packets) / perf_duration)))"
+	info "    baseline hash (non-ranged entries):          ${pps}pps"
+	handle="$(nft -a list chain netdev perf test | grep counter)"
+	handle="${handle##* }"
+	nft delete rule netdev perf test handle "${handle}"
+
+	nft add rule "netdev perf test ${chain_spec%%. *} @noconcat \
+		counter name \"test\" drop"
+	nft reset counter netdev perf test >/dev/null 2>&1
+	sleep "${perf_duration}"
+	pps="$(printf %10s $(($(count_perf_packets) / perf_duration)))"
+	info "    baseline rbtree (match on first field only): ${pps}pps"
+	handle="$(nft -a list chain netdev perf test | grep counter)"
+	handle="${handle##* }"
+	nft delete rule netdev perf test handle "${handle}"
+
+	nft add rule "netdev perf test ${chain_spec} @test \
+		counter name \"test\" drop"
+	nft reset counter netdev perf test >/dev/null 2>&1
+	sleep "${perf_duration}"
+	pps="$(printf %10s $(($(count_perf_packets) / perf_duration)))"
+	p5="$(printf %5s "${perf_entries}")"
+	info "    set with ${p5} full, ranged entries:         ${pps}pps"
+	kill "${perf_pid}"
+}
+
+# Run everything in a separate network namespace
+[ "${1}" != "run" ] && { unshare -n "${0}" run; exit $?; }
+tmp="$(mktemp)"
+trap cleanup EXIT
+
+# Entry point for test runs
+passed=0
+for name in ${TESTS}; do
+	printf "TEST: %s\n" "${name}"
+	for type in ${TYPES}; do
+		eval desc=\$TYPE_"${type}"
+		IFS='
+'
+		for __line in ${desc}; do
+			# shellcheck disable=SC2086
+			eval ${__line%%	*}=\"${__line##*	}\";
+		done
+		IFS=' 	
+'
+
+		if [ "${name}" = "concurrency" ] && \
+		   [ "${race_repeat}" = "0" ]; then
+			continue
+		fi
+		if [ "${name}" = "performance" ] && \
+		   [ "${perf_duration}" = "0" ]; then
+			continue
+		fi
+
+		printf "  %-60s  " "${display}"
+		eval test_"${name}"
+		ret=$?
+
+		if [ $ret -eq 0 ]; then
+			printf "[ OK ]\n"
+			info_flush
+			passed=$((passed + 1))
+		elif [ $ret -eq 1 ]; then
+			printf "[FAIL]\n"
+			err_flush
+			exit 1
+		elif [ $ret -eq ${KSELFTEST_SKIP} ]; then
+			printf "[SKIP]\n"
+			err_flush
+		fi
+	done
+done
+
+[ ${passed} -eq 0 ] && exit ${KSELFTEST_SKIP}
-- 
cgit v1.2.3-59-g8ed1b


From 2e24cd755552350b94a7617617c6877b8cbcb701 Mon Sep 17 00:00:00 2001
From: Cong Wang <xiyou.wangcong@gmail.com>
Date: Thu, 23 Jan 2020 16:26:18 -0800
Subject: net_sched: fix ops->bind_class() implementations

The current implementations of ops->bind_class() are merely
searching for classid and updating class in the struct tcf_result,
without invoking either of cl_ops->bind_tcf() or
cl_ops->unbind_tcf(). This breaks the design of them as qdisc's
like cbq use them to count filters too. This is why syzbot triggered
the warning in cbq_destroy_class().

In order to fix this, we have to call cl_ops->bind_tcf() and
cl_ops->unbind_tcf() like the filter binding path. This patch does
so by refactoring out two helper functions __tcf_bind_filter()
and __tcf_unbind_filter(), which are lockless and accept a Qdisc
pointer, then teaching each implementation to call them correctly.

Note, we merely pass the Qdisc pointer as an opaque pointer to
each filter, they only need to pass it down to the helper
functions without understanding it at all.

Fixes: 07d79fc7d94e ("net_sched: add reverse binding for tc class")
Reported-and-tested-by: syzbot+0a0596220218fcb603a8@syzkaller.appspotmail.com
Reported-and-tested-by: syzbot+63bdb6006961d8c917c6@syzkaller.appspotmail.com
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Cc: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/pkt_cls.h     | 33 +++++++++++++++++++--------------
 include/net/sch_generic.h |  3 ++-
 net/sched/cls_basic.c     | 11 ++++++++---
 net/sched/cls_bpf.c       | 11 ++++++++---
 net/sched/cls_flower.c    | 11 ++++++++---
 net/sched/cls_fw.c        | 11 ++++++++---
 net/sched/cls_matchall.c  | 11 ++++++++---
 net/sched/cls_route.c     | 11 ++++++++---
 net/sched/cls_rsvp.h      | 11 ++++++++---
 net/sched/cls_tcindex.c   | 11 ++++++++---
 net/sched/cls_u32.c       | 11 ++++++++---
 net/sched/sch_api.c       |  6 ++++--
 12 files changed, 97 insertions(+), 44 deletions(-)

diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h
index ce036492986a..a972244ab193 100644
--- a/include/net/pkt_cls.h
+++ b/include/net/pkt_cls.h
@@ -141,31 +141,38 @@ __cls_set_class(unsigned long *clp, unsigned long cl)
 	return xchg(clp, cl);
 }
 
-static inline unsigned long
-cls_set_class(struct Qdisc *q, unsigned long *clp, unsigned long cl)
+static inline void
+__tcf_bind_filter(struct Qdisc *q, struct tcf_result *r, unsigned long base)
 {
-	unsigned long old_cl;
+	unsigned long cl;
 
-	sch_tree_lock(q);
-	old_cl = __cls_set_class(clp, cl);
-	sch_tree_unlock(q);
-	return old_cl;
+	cl = q->ops->cl_ops->bind_tcf(q, base, r->classid);
+	cl = __cls_set_class(&r->class, cl);
+	if (cl)
+		q->ops->cl_ops->unbind_tcf(q, cl);
 }
 
 static inline void
 tcf_bind_filter(struct tcf_proto *tp, struct tcf_result *r, unsigned long base)
 {
 	struct Qdisc *q = tp->chain->block->q;
-	unsigned long cl;
 
 	/* Check q as it is not set for shared blocks. In that case,
 	 * setting class is not supported.
 	 */
 	if (!q)
 		return;
-	cl = q->ops->cl_ops->bind_tcf(q, base, r->classid);
-	cl = cls_set_class(q, &r->class, cl);
-	if (cl)
+	sch_tree_lock(q);
+	__tcf_bind_filter(q, r, base);
+	sch_tree_unlock(q);
+}
+
+static inline void
+__tcf_unbind_filter(struct Qdisc *q, struct tcf_result *r)
+{
+	unsigned long cl;
+
+	if ((cl = __cls_set_class(&r->class, 0)) != 0)
 		q->ops->cl_ops->unbind_tcf(q, cl);
 }
 
@@ -173,12 +180,10 @@ static inline void
 tcf_unbind_filter(struct tcf_proto *tp, struct tcf_result *r)
 {
 	struct Qdisc *q = tp->chain->block->q;
-	unsigned long cl;
 
 	if (!q)
 		return;
-	if ((cl = __cls_set_class(&r->class, 0)) != 0)
-		q->ops->cl_ops->unbind_tcf(q, cl);
+	__tcf_unbind_filter(q, r);
 }
 
 struct tcf_exts {
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index fceddf89592a..151208704ed2 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -318,7 +318,8 @@ struct tcf_proto_ops {
 					  void *type_data);
 	void			(*hw_del)(struct tcf_proto *tp,
 					  void *type_data);
-	void			(*bind_class)(void *, u32, unsigned long);
+	void			(*bind_class)(void *, u32, unsigned long,
+					      void *, unsigned long);
 	void *			(*tmplt_create)(struct net *net,
 						struct tcf_chain *chain,
 						struct nlattr **tca,
diff --git a/net/sched/cls_basic.c b/net/sched/cls_basic.c
index 4aafbe3d435c..f256a7c69093 100644
--- a/net/sched/cls_basic.c
+++ b/net/sched/cls_basic.c
@@ -263,12 +263,17 @@ skip:
 	}
 }
 
-static void basic_bind_class(void *fh, u32 classid, unsigned long cl)
+static void basic_bind_class(void *fh, u32 classid, unsigned long cl, void *q,
+			     unsigned long base)
 {
 	struct basic_filter *f = fh;
 
-	if (f && f->res.classid == classid)
-		f->res.class = cl;
+	if (f && f->res.classid == classid) {
+		if (cl)
+			__tcf_bind_filter(q, &f->res, base);
+		else
+			__tcf_unbind_filter(q, &f->res);
+	}
 }
 
 static int basic_dump(struct net *net, struct tcf_proto *tp, void *fh,
diff --git a/net/sched/cls_bpf.c b/net/sched/cls_bpf.c
index 8229ed4a67be..6e3e63db0e01 100644
--- a/net/sched/cls_bpf.c
+++ b/net/sched/cls_bpf.c
@@ -631,12 +631,17 @@ nla_put_failure:
 	return -1;
 }
 
-static void cls_bpf_bind_class(void *fh, u32 classid, unsigned long cl)
+static void cls_bpf_bind_class(void *fh, u32 classid, unsigned long cl,
+			       void *q, unsigned long base)
 {
 	struct cls_bpf_prog *prog = fh;
 
-	if (prog && prog->res.classid == classid)
-		prog->res.class = cl;
+	if (prog && prog->res.classid == classid) {
+		if (cl)
+			__tcf_bind_filter(q, &prog->res, base);
+		else
+			__tcf_unbind_filter(q, &prog->res);
+	}
 }
 
 static void cls_bpf_walk(struct tcf_proto *tp, struct tcf_walker *arg,
diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c
index b0f42e62dd76..f9c0d1e8d380 100644
--- a/net/sched/cls_flower.c
+++ b/net/sched/cls_flower.c
@@ -2765,12 +2765,17 @@ nla_put_failure:
 	return -EMSGSIZE;
 }
 
-static void fl_bind_class(void *fh, u32 classid, unsigned long cl)
+static void fl_bind_class(void *fh, u32 classid, unsigned long cl, void *q,
+			  unsigned long base)
 {
 	struct cls_fl_filter *f = fh;
 
-	if (f && f->res.classid == classid)
-		f->res.class = cl;
+	if (f && f->res.classid == classid) {
+		if (cl)
+			__tcf_bind_filter(q, &f->res, base);
+		else
+			__tcf_unbind_filter(q, &f->res);
+	}
 }
 
 static bool fl_delete_empty(struct tcf_proto *tp)
diff --git a/net/sched/cls_fw.c b/net/sched/cls_fw.c
index c9496c920d6f..ec945294626a 100644
--- a/net/sched/cls_fw.c
+++ b/net/sched/cls_fw.c
@@ -419,12 +419,17 @@ nla_put_failure:
 	return -1;
 }
 
-static void fw_bind_class(void *fh, u32 classid, unsigned long cl)
+static void fw_bind_class(void *fh, u32 classid, unsigned long cl, void *q,
+			  unsigned long base)
 {
 	struct fw_filter *f = fh;
 
-	if (f && f->res.classid == classid)
-		f->res.class = cl;
+	if (f && f->res.classid == classid) {
+		if (cl)
+			__tcf_bind_filter(q, &f->res, base);
+		else
+			__tcf_unbind_filter(q, &f->res);
+	}
 }
 
 static struct tcf_proto_ops cls_fw_ops __read_mostly = {
diff --git a/net/sched/cls_matchall.c b/net/sched/cls_matchall.c
index 7fc2eb62aa98..039cc86974f4 100644
--- a/net/sched/cls_matchall.c
+++ b/net/sched/cls_matchall.c
@@ -393,12 +393,17 @@ nla_put_failure:
 	return -1;
 }
 
-static void mall_bind_class(void *fh, u32 classid, unsigned long cl)
+static void mall_bind_class(void *fh, u32 classid, unsigned long cl, void *q,
+			    unsigned long base)
 {
 	struct cls_mall_head *head = fh;
 
-	if (head && head->res.classid == classid)
-		head->res.class = cl;
+	if (head && head->res.classid == classid) {
+		if (cl)
+			__tcf_bind_filter(q, &head->res, base);
+		else
+			__tcf_unbind_filter(q, &head->res);
+	}
 }
 
 static struct tcf_proto_ops cls_mall_ops __read_mostly = {
diff --git a/net/sched/cls_route.c b/net/sched/cls_route.c
index 2d9e0b4484ea..6f8786b06bde 100644
--- a/net/sched/cls_route.c
+++ b/net/sched/cls_route.c
@@ -641,12 +641,17 @@ nla_put_failure:
 	return -1;
 }
 
-static void route4_bind_class(void *fh, u32 classid, unsigned long cl)
+static void route4_bind_class(void *fh, u32 classid, unsigned long cl, void *q,
+			      unsigned long base)
 {
 	struct route4_filter *f = fh;
 
-	if (f && f->res.classid == classid)
-		f->res.class = cl;
+	if (f && f->res.classid == classid) {
+		if (cl)
+			__tcf_bind_filter(q, &f->res, base);
+		else
+			__tcf_unbind_filter(q, &f->res);
+	}
 }
 
 static struct tcf_proto_ops cls_route4_ops __read_mostly = {
diff --git a/net/sched/cls_rsvp.h b/net/sched/cls_rsvp.h
index 2f3c03b25d5d..c22624131949 100644
--- a/net/sched/cls_rsvp.h
+++ b/net/sched/cls_rsvp.h
@@ -738,12 +738,17 @@ nla_put_failure:
 	return -1;
 }
 
-static void rsvp_bind_class(void *fh, u32 classid, unsigned long cl)
+static void rsvp_bind_class(void *fh, u32 classid, unsigned long cl, void *q,
+			    unsigned long base)
 {
 	struct rsvp_filter *f = fh;
 
-	if (f && f->res.classid == classid)
-		f->res.class = cl;
+	if (f && f->res.classid == classid) {
+		if (cl)
+			__tcf_bind_filter(q, &f->res, base);
+		else
+			__tcf_unbind_filter(q, &f->res);
+	}
 }
 
 static struct tcf_proto_ops RSVP_OPS __read_mostly = {
diff --git a/net/sched/cls_tcindex.c b/net/sched/cls_tcindex.c
index e573e5a5c794..3d4a1280352f 100644
--- a/net/sched/cls_tcindex.c
+++ b/net/sched/cls_tcindex.c
@@ -654,12 +654,17 @@ nla_put_failure:
 	return -1;
 }
 
-static void tcindex_bind_class(void *fh, u32 classid, unsigned long cl)
+static void tcindex_bind_class(void *fh, u32 classid, unsigned long cl,
+			       void *q, unsigned long base)
 {
 	struct tcindex_filter_result *r = fh;
 
-	if (r && r->res.classid == classid)
-		r->res.class = cl;
+	if (r && r->res.classid == classid) {
+		if (cl)
+			__tcf_bind_filter(q, &r->res, base);
+		else
+			__tcf_unbind_filter(q, &r->res);
+	}
 }
 
 static struct tcf_proto_ops cls_tcindex_ops __read_mostly = {
diff --git a/net/sched/cls_u32.c b/net/sched/cls_u32.c
index a0e6fac613de..e15ff335953d 100644
--- a/net/sched/cls_u32.c
+++ b/net/sched/cls_u32.c
@@ -1255,12 +1255,17 @@ static int u32_reoffload(struct tcf_proto *tp, bool add, flow_setup_cb_t *cb,
 	return 0;
 }
 
-static void u32_bind_class(void *fh, u32 classid, unsigned long cl)
+static void u32_bind_class(void *fh, u32 classid, unsigned long cl, void *q,
+			   unsigned long base)
 {
 	struct tc_u_knode *n = fh;
 
-	if (n && n->res.classid == classid)
-		n->res.class = cl;
+	if (n && n->res.classid == classid) {
+		if (cl)
+			__tcf_bind_filter(q, &n->res, base);
+		else
+			__tcf_unbind_filter(q, &n->res);
+	}
 }
 
 static int u32_dump(struct net *net, struct tcf_proto *tp, void *fh,
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 1047825d9f48..943ad3425380 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -1891,8 +1891,9 @@ static int tclass_del_notify(struct net *net,
 
 struct tcf_bind_args {
 	struct tcf_walker w;
-	u32 classid;
+	unsigned long base;
 	unsigned long cl;
+	u32 classid;
 };
 
 static int tcf_node_bind(struct tcf_proto *tp, void *n, struct tcf_walker *arg)
@@ -1903,7 +1904,7 @@ static int tcf_node_bind(struct tcf_proto *tp, void *n, struct tcf_walker *arg)
 		struct Qdisc *q = tcf_block_q(tp->chain->block);
 
 		sch_tree_lock(q);
-		tp->ops->bind_class(n, a->classid, a->cl);
+		tp->ops->bind_class(n, a->classid, a->cl, q, a->base);
 		sch_tree_unlock(q);
 	}
 	return 0;
@@ -1936,6 +1937,7 @@ static void tc_bind_tclass(struct Qdisc *q, u32 portid, u32 clid,
 
 			arg.w.fn = tcf_node_bind;
 			arg.classid = clid;
+			arg.base = cl;
 			arg.cl = new_cl;
 			tp->ops->walk(tp, &arg.w, true);
 		}
-- 
cgit v1.2.3-59-g8ed1b


From 760d228e322e99cdf6d81b4b60a268b8f13cf67a Mon Sep 17 00:00:00 2001
From: Cong Wang <xiyou.wangcong@gmail.com>
Date: Thu, 23 Jan 2020 17:27:08 -0800
Subject: net_sched: walk through all child classes in tc_bind_tclass()

In a complex TC class hierarchy like this:

tc qdisc add dev eth0 root handle 1:0 cbq bandwidth 100Mbit         \
  avpkt 1000 cell 8
tc class add dev eth0 parent 1:0 classid 1:1 cbq bandwidth 100Mbit  \
  rate 6Mbit weight 0.6Mbit prio 8 allot 1514 cell 8 maxburst 20      \
  avpkt 1000 bounded

tc filter add dev eth0 parent 1:0 protocol ip prio 1 u32 match ip \
  sport 80 0xffff flowid 1:3
tc filter add dev eth0 parent 1:0 protocol ip prio 1 u32 match ip \
  sport 25 0xffff flowid 1:4

tc class add dev eth0 parent 1:1 classid 1:3 cbq bandwidth 100Mbit  \
  rate 5Mbit weight 0.5Mbit prio 5 allot 1514 cell 8 maxburst 20      \
  avpkt 1000
tc class add dev eth0 parent 1:1 classid 1:4 cbq bandwidth 100Mbit  \
  rate 3Mbit weight 0.3Mbit prio 5 allot 1514 cell 8 maxburst 20      \
  avpkt 1000

where filters are installed on qdisc 1:0, so we can't merely
search from class 1:1 when creating class 1:3 and class 1:4. We have
to walk through all the child classes of the direct parent qdisc.
Otherwise we would miss filters those need reverse binding.

Fixes: 07d79fc7d94e ("net_sched: add reverse binding for tc class")
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Cc: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sched/sch_api.c | 41 ++++++++++++++++++++++++++++++-----------
 1 file changed, 30 insertions(+), 11 deletions(-)

diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 943ad3425380..50794125bf02 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -1910,22 +1910,24 @@ static int tcf_node_bind(struct tcf_proto *tp, void *n, struct tcf_walker *arg)
 	return 0;
 }
 
-static void tc_bind_tclass(struct Qdisc *q, u32 portid, u32 clid,
-			   unsigned long new_cl)
+struct tc_bind_class_args {
+	struct qdisc_walker w;
+	unsigned long new_cl;
+	u32 portid;
+	u32 clid;
+};
+
+static int tc_bind_class_walker(struct Qdisc *q, unsigned long cl,
+				struct qdisc_walker *w)
 {
+	struct tc_bind_class_args *a = (struct tc_bind_class_args *)w;
 	const struct Qdisc_class_ops *cops = q->ops->cl_ops;
 	struct tcf_block *block;
 	struct tcf_chain *chain;
-	unsigned long cl;
 
-	cl = cops->find(q, portid);
-	if (!cl)
-		return;
-	if (!cops->tcf_block)
-		return;
 	block = cops->tcf_block(q, cl, NULL);
 	if (!block)
-		return;
+		return 0;
 	for (chain = tcf_get_next_chain(block, NULL);
 	     chain;
 	     chain = tcf_get_next_chain(block, chain)) {
@@ -1936,12 +1938,29 @@ static void tc_bind_tclass(struct Qdisc *q, u32 portid, u32 clid,
 			struct tcf_bind_args arg = {};
 
 			arg.w.fn = tcf_node_bind;
-			arg.classid = clid;
+			arg.classid = a->clid;
 			arg.base = cl;
-			arg.cl = new_cl;
+			arg.cl = a->new_cl;
 			tp->ops->walk(tp, &arg.w, true);
 		}
 	}
+
+	return 0;
+}
+
+static void tc_bind_tclass(struct Qdisc *q, u32 portid, u32 clid,
+			   unsigned long new_cl)
+{
+	const struct Qdisc_class_ops *cops = q->ops->cl_ops;
+	struct tc_bind_class_args args = {};
+
+	if (!cops->tcf_block)
+		return;
+	args.portid = portid;
+	args.clid = clid;
+	args.new_cl = new_cl;
+	args.w.fn = tc_bind_class_walker;
+	q->ops->cl_ops->walk(q, &args.w);
 }
 
 #else
-- 
cgit v1.2.3-59-g8ed1b


From 07fdbee134b3530ef30c709aa4251ca04ea9e3f8 Mon Sep 17 00:00:00 2001
From: Andrey Ignatov <rdna@fb.com>
Date: Fri, 24 Jan 2020 14:41:42 -0800
Subject: tools/bpf: Allow overriding llvm tools for runqslower

tools/testing/selftests/bpf/Makefile supports overriding clang, llc and
other tools so that custom ones can be used instead of those from PATH.
It's convinient and heavily used by some users.

Apply same rules to runqslower/Makefile.

Signed-off-by: Andrey Ignatov <rdna@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/20200124224142.1833678-1-rdna@fb.com
---
 tools/bpf/runqslower/Makefile | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/tools/bpf/runqslower/Makefile b/tools/bpf/runqslower/Makefile
index faf5418609ea..0c021352beed 100644
--- a/tools/bpf/runqslower/Makefile
+++ b/tools/bpf/runqslower/Makefile
@@ -1,8 +1,8 @@
 # SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause)
 OUTPUT := .output
-CLANG := clang
-LLC := llc
-LLVM_STRIP := llvm-strip
+CLANG ?= clang
+LLC ?= llc
+LLVM_STRIP ?= llvm-strip
 DEFAULT_BPFTOOL := $(OUTPUT)/sbin/bpftool
 BPFTOOL ?= $(DEFAULT_BPFTOOL)
 LIBBPF_SRC := $(abspath ../../lib/bpf)
-- 
cgit v1.2.3-59-g8ed1b


From f9e95555757915fc194288862d2978e370fe316b Mon Sep 17 00:00:00 2001
From: Stephen Worley <sworley@cumulusnetworks.com>
Date: Fri, 24 Jan 2020 16:53:27 -0500
Subject: net: include struct nhmsg size in nh nlmsg size

Include the size of struct nhmsg size when calculating
how much of a payload to allocate in a new netlink nexthop
notification message.

Without this, we will fail to fill the skbuff at certain nexthop
group sizes.

You can reproduce the failure with the following iproute2 commands:

ip link add dummy1 type dummy
ip link add dummy2 type dummy
ip link add dummy3 type dummy
ip link add dummy4 type dummy
ip link add dummy5 type dummy
ip link add dummy6 type dummy
ip link add dummy7 type dummy
ip link add dummy8 type dummy
ip link add dummy9 type dummy
ip link add dummy10 type dummy
ip link add dummy11 type dummy
ip link add dummy12 type dummy
ip link add dummy13 type dummy
ip link add dummy14 type dummy
ip link add dummy15 type dummy
ip link add dummy16 type dummy
ip link add dummy17 type dummy
ip link add dummy18 type dummy
ip link add dummy19 type dummy

ip ro add 1.1.1.1/32 dev dummy1
ip ro add 1.1.1.2/32 dev dummy2
ip ro add 1.1.1.3/32 dev dummy3
ip ro add 1.1.1.4/32 dev dummy4
ip ro add 1.1.1.5/32 dev dummy5
ip ro add 1.1.1.6/32 dev dummy6
ip ro add 1.1.1.7/32 dev dummy7
ip ro add 1.1.1.8/32 dev dummy8
ip ro add 1.1.1.9/32 dev dummy9
ip ro add 1.1.1.10/32 dev dummy10
ip ro add 1.1.1.11/32 dev dummy11
ip ro add 1.1.1.12/32 dev dummy12
ip ro add 1.1.1.13/32 dev dummy13
ip ro add 1.1.1.14/32 dev dummy14
ip ro add 1.1.1.15/32 dev dummy15
ip ro add 1.1.1.16/32 dev dummy16
ip ro add 1.1.1.17/32 dev dummy17
ip ro add 1.1.1.18/32 dev dummy18
ip ro add 1.1.1.19/32 dev dummy19

ip next add id 1 via 1.1.1.1 dev dummy1
ip next add id 2 via 1.1.1.2 dev dummy2
ip next add id 3 via 1.1.1.3 dev dummy3
ip next add id 4 via 1.1.1.4 dev dummy4
ip next add id 5 via 1.1.1.5 dev dummy5
ip next add id 6 via 1.1.1.6 dev dummy6
ip next add id 7 via 1.1.1.7 dev dummy7
ip next add id 8 via 1.1.1.8 dev dummy8
ip next add id 9 via 1.1.1.9 dev dummy9
ip next add id 10 via 1.1.1.10 dev dummy10
ip next add id 11 via 1.1.1.11 dev dummy11
ip next add id 12 via 1.1.1.12 dev dummy12
ip next add id 13 via 1.1.1.13 dev dummy13
ip next add id 14 via 1.1.1.14 dev dummy14
ip next add id 15 via 1.1.1.15 dev dummy15
ip next add id 16 via 1.1.1.16 dev dummy16
ip next add id 17 via 1.1.1.17 dev dummy17
ip next add id 18 via 1.1.1.18 dev dummy18
ip next add id 19 via 1.1.1.19 dev dummy19

ip next add id 1111 group 1/2/3/4/5/6/7/8/9/10/11/12/13/14/15/16/17/18/19
ip next del id 1111

Fixes: 430a049190de ("nexthop: Add support for nexthop groups")
Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/nexthop.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/net/ipv4/nexthop.c b/net/ipv4/nexthop.c
index 511eaa94e2d1..d072c326dd64 100644
--- a/net/ipv4/nexthop.c
+++ b/net/ipv4/nexthop.c
@@ -321,7 +321,9 @@ static size_t nh_nlmsg_size_single(struct nexthop *nh)
 
 static size_t nh_nlmsg_size(struct nexthop *nh)
 {
-	size_t sz = nla_total_size(4);    /* NHA_ID */
+	size_t sz = NLMSG_ALIGN(sizeof(struct nhmsg));
+
+	sz += nla_total_size(4); /* NHA_ID */
 
 	if (nh->is_group)
 		sz += nh_nlmsg_size_grp(nh);
-- 
cgit v1.2.3-59-g8ed1b


From 90435a7891a2259b0f74c5a1bc5600d0d64cba8f Mon Sep 17 00:00:00 2001
From: Vasily Averin <vvs@virtuozzo.com>
Date: Sat, 25 Jan 2020 12:10:02 +0300
Subject: bpf: map_seq_next should always increase position index

If seq_file .next fuction does not change position index,
read after some lseek can generate an unexpected output.

See also: https://bugzilla.kernel.org/show_bug.cgi?id=206283

v1 -> v2: removed missed increment in end of function

Signed-off-by: Vasily Averin <vvs@virtuozzo.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/eca84fdd-c374-a154-d874-6c7b55fc3bc4@virtuozzo.com
---
 kernel/bpf/inode.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/kernel/bpf/inode.c b/kernel/bpf/inode.c
index e11059b99f18..bd2fd8eab470 100644
--- a/kernel/bpf/inode.c
+++ b/kernel/bpf/inode.c
@@ -196,6 +196,7 @@ static void *map_seq_next(struct seq_file *m, void *v, loff_t *pos)
 	void *key = map_iter(m)->key;
 	void *prev_key;
 
+	(*pos)++;
 	if (map_iter(m)->done)
 		return NULL;
 
@@ -208,8 +209,6 @@ static void *map_seq_next(struct seq_file *m, void *v, loff_t *pos)
 		map_iter(m)->done = true;
 		return NULL;
 	}
-
-	++(*pos);
 	return key;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 55cd9f67f1e45de8517cdaab985fb8e56c0bc1d8 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Fri, 24 Jan 2020 14:57:20 -0800
Subject: net_sched: ematch: reject invalid TCF_EM_SIMPLE

It is possible for malicious userspace to set TCF_EM_SIMPLE bit
even for matches that should not have this bit set.

This can fool two places using tcf_em_is_simple()

1) tcf_em_tree_destroy() -> memory leak of em->data
   if ops->destroy() is NULL

2) tcf_em_tree_dump() wrongly report/leak 4 low-order bytes
   of a kernel pointer.

BUG: memory leak
unreferenced object 0xffff888121850a40 (size 32):
  comm "syz-executor927", pid 7193, jiffies 4294941655 (age 19.840s)
  hex dump (first 32 bytes):
    00 00 00 00 01 00 00 00 00 00 00 00 00 00 00 00  ................
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
  backtrace:
    [<00000000f67036ea>] kmemleak_alloc_recursive include/linux/kmemleak.h:43 [inline]
    [<00000000f67036ea>] slab_post_alloc_hook mm/slab.h:586 [inline]
    [<00000000f67036ea>] slab_alloc mm/slab.c:3320 [inline]
    [<00000000f67036ea>] __do_kmalloc mm/slab.c:3654 [inline]
    [<00000000f67036ea>] __kmalloc_track_caller+0x165/0x300 mm/slab.c:3671
    [<00000000fab0cc8e>] kmemdup+0x27/0x60 mm/util.c:127
    [<00000000d9992e0a>] kmemdup include/linux/string.h:453 [inline]
    [<00000000d9992e0a>] em_nbyte_change+0x5b/0x90 net/sched/em_nbyte.c:32
    [<000000007e04f711>] tcf_em_validate net/sched/ematch.c:241 [inline]
    [<000000007e04f711>] tcf_em_tree_validate net/sched/ematch.c:359 [inline]
    [<000000007e04f711>] tcf_em_tree_validate+0x332/0x46f net/sched/ematch.c:300
    [<000000007a769204>] basic_set_parms net/sched/cls_basic.c:157 [inline]
    [<000000007a769204>] basic_change+0x1d7/0x5f0 net/sched/cls_basic.c:219
    [<00000000e57a5997>] tc_new_tfilter+0x566/0xf70 net/sched/cls_api.c:2104
    [<0000000074b68559>] rtnetlink_rcv_msg+0x3b2/0x4b0 net/core/rtnetlink.c:5415
    [<00000000b7fe53fb>] netlink_rcv_skb+0x61/0x170 net/netlink/af_netlink.c:2477
    [<00000000e83a40d0>] rtnetlink_rcv+0x1d/0x30 net/core/rtnetlink.c:5442
    [<00000000d62ba933>] netlink_unicast_kernel net/netlink/af_netlink.c:1302 [inline]
    [<00000000d62ba933>] netlink_unicast+0x223/0x310 net/netlink/af_netlink.c:1328
    [<0000000088070f72>] netlink_sendmsg+0x2c0/0x570 net/netlink/af_netlink.c:1917
    [<00000000f70b15ea>] sock_sendmsg_nosec net/socket.c:639 [inline]
    [<00000000f70b15ea>] sock_sendmsg+0x54/0x70 net/socket.c:659
    [<00000000ef95a9be>] ____sys_sendmsg+0x2d0/0x300 net/socket.c:2330
    [<00000000b650f1ab>] ___sys_sendmsg+0x8a/0xd0 net/socket.c:2384
    [<0000000055bfa74a>] __sys_sendmsg+0x80/0xf0 net/socket.c:2417
    [<000000002abac183>] __do_sys_sendmsg net/socket.c:2426 [inline]
    [<000000002abac183>] __se_sys_sendmsg net/socket.c:2424 [inline]
    [<000000002abac183>] __x64_sys_sendmsg+0x23/0x30 net/socket.c:2424

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: syzbot+03c4738ed29d5d366ddf@syzkaller.appspotmail.com
Cc: Cong Wang <xiyou.wangcong@gmail.com>
Acked-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sched/ematch.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/net/sched/ematch.c b/net/sched/ematch.c
index d0140a92694a..dd3b8c11a2e0 100644
--- a/net/sched/ematch.c
+++ b/net/sched/ematch.c
@@ -238,6 +238,9 @@ static int tcf_em_validate(struct tcf_proto *tp,
 			goto errout;
 
 		if (em->ops->change) {
+			err = -EINVAL;
+			if (em_hdr->flags & TCF_EM_SIMPLE)
+				goto errout;
 			err = em->ops->change(net, data, data_len, em);
 			if (err < 0)
 				goto errout;
-- 
cgit v1.2.3-59-g8ed1b


From 122d74fac84204b9a98263636f6f9a3b2e665639 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Fri, 24 Jan 2020 23:08:04 +0000
Subject: rxrpc: Fix use-after-free in rxrpc_receive_data()

The subpacket scanning loop in rxrpc_receive_data() references the
subpacket count in the private data part of the sk_buff in the loop
termination condition.  However, when the final subpacket is pasted into
the ring buffer, the function is no longer has a ref on the sk_buff and
should not be looking at sp->* any more.  This point is actually marked in
the code when skb is cleared (but sp is not - which is an error).

Fix this by caching sp->nr_subpackets in a local variable and using that
instead.

Also clear 'sp' to catch accesses after that point.

This can show up as an oops in rxrpc_get_skb() if sp->nr_subpackets gets
trashed by the sk_buff getting freed and reused in the meantime.

Fixes: e2de6c404898 ("rxrpc: Use info in skbuff instead of reparsing a jumbo packet")
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/rxrpc/input.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c
index 86bd133b4fa0..96d54e5bf7bc 100644
--- a/net/rxrpc/input.c
+++ b/net/rxrpc/input.c
@@ -413,7 +413,7 @@ static void rxrpc_input_data(struct rxrpc_call *call, struct sk_buff *skb)
 {
 	struct rxrpc_skb_priv *sp = rxrpc_skb(skb);
 	enum rxrpc_call_state state;
-	unsigned int j;
+	unsigned int j, nr_subpackets;
 	rxrpc_serial_t serial = sp->hdr.serial, ack_serial = 0;
 	rxrpc_seq_t seq0 = sp->hdr.seq, hard_ack;
 	bool immediate_ack = false, jumbo_bad = false;
@@ -457,7 +457,8 @@ static void rxrpc_input_data(struct rxrpc_call *call, struct sk_buff *skb)
 	call->ackr_prev_seq = seq0;
 	hard_ack = READ_ONCE(call->rx_hard_ack);
 
-	if (sp->nr_subpackets > 1) {
+	nr_subpackets = sp->nr_subpackets;
+	if (nr_subpackets > 1) {
 		if (call->nr_jumbo_bad > 3) {
 			ack = RXRPC_ACK_NOSPACE;
 			ack_serial = serial;
@@ -465,11 +466,11 @@ static void rxrpc_input_data(struct rxrpc_call *call, struct sk_buff *skb)
 		}
 	}
 
-	for (j = 0; j < sp->nr_subpackets; j++) {
+	for (j = 0; j < nr_subpackets; j++) {
 		rxrpc_serial_t serial = sp->hdr.serial + j;
 		rxrpc_seq_t seq = seq0 + j;
 		unsigned int ix = seq & RXRPC_RXTX_BUFF_MASK;
-		bool terminal = (j == sp->nr_subpackets - 1);
+		bool terminal = (j == nr_subpackets - 1);
 		bool last = terminal && (sp->rx_flags & RXRPC_SKB_INCL_LAST);
 		u8 flags, annotation = j;
 
@@ -506,7 +507,7 @@ static void rxrpc_input_data(struct rxrpc_call *call, struct sk_buff *skb)
 		}
 
 		if (call->rxtx_buffer[ix]) {
-			rxrpc_input_dup_data(call, seq, sp->nr_subpackets > 1,
+			rxrpc_input_dup_data(call, seq, nr_subpackets > 1,
 					     &jumbo_bad);
 			if (ack != RXRPC_ACK_DUPLICATE) {
 				ack = RXRPC_ACK_DUPLICATE;
@@ -564,6 +565,7 @@ static void rxrpc_input_data(struct rxrpc_call *call, struct sk_buff *skb)
 			 * ring.
 			 */
 			skb = NULL;
+			sp = NULL;
 		}
 
 		if (last) {
-- 
cgit v1.2.3-59-g8ed1b


From 79572c98c554dcdb080bca547c871a51716dcdf8 Mon Sep 17 00:00:00 2001
From: Sven Auhagen <sven.auhagen@voleatech.de>
Date: Sat, 25 Jan 2020 08:07:03 +0000
Subject: mvneta driver disallow XDP program on hardware buffer management

Recently XDP Support was added to the mvneta driver
for software buffer management only.
It is still possible to attach an XDP program if
hardware buffer management is used.
It is not doing anything at that point.

The patch disallows attaching XDP programs to mvneta
if hardware buffer management is used.

I am sorry about that. It is my first submission and I am having
some troubles with the format of my emails.

v4 -> v5:
- Remove extra tabs

v3 -> v4:
- Please ignore v3 I accidentally submitted
  my other patch with git-send-mail and v4 is correct

v2 -> v3:
- My mailserver corrupted the patch
  resubmission with git-send-email

v1 -> v2:
- Fixing the patches indentation

Signed-off-by: Sven Auhagen <sven.auhagen@voleatech.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/marvell/mvneta.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index 0449d4b28ade..2dfbfdff45a8 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -4226,6 +4226,12 @@ static int mvneta_xdp_setup(struct net_device *dev, struct bpf_prog *prog,
 		return -EOPNOTSUPP;
 	}
 
+	if (pp->bm_priv) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Hardware Buffer Management not supported on XDP");
+		return -EOPNOTSUPP;
+	}
+
 	need_update = !!pp->xdp_prog != !!prog;
 	if (running && need_update)
 		mvneta_stop(dev);
-- 
cgit v1.2.3-59-g8ed1b


From 3b33583265ed3b0ae76eddbabf9d038b4076d1a9 Mon Sep 17 00:00:00 2001
From: Steffen Klassert <steffen.klassert@secunet.com>
Date: Sat, 25 Jan 2020 11:26:42 +0100
Subject: net: Add fraglist GRO/GSO feature flags

This adds new Fraglist GRO/GSO feature flags. They will be used
to configure fraglist GRO/GSO what will be implemented with some
followup paches.

Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdev_features.h | 6 +++++-
 include/linux/netdevice.h       | 1 +
 include/linux/skbuff.h          | 2 ++
 net/ethtool/common.c            | 1 +
 4 files changed, 9 insertions(+), 1 deletion(-)

diff --git a/include/linux/netdev_features.h b/include/linux/netdev_features.h
index 4b19c544c59a..b239507da2a0 100644
--- a/include/linux/netdev_features.h
+++ b/include/linux/netdev_features.h
@@ -53,8 +53,9 @@ enum {
 	NETIF_F_GSO_ESP_BIT,		/* ... ESP with TSO */
 	NETIF_F_GSO_UDP_BIT,		/* ... UFO, deprecated except tuntap */
 	NETIF_F_GSO_UDP_L4_BIT,		/* ... UDP payload GSO (not UFO) */
+	NETIF_F_GSO_FRAGLIST_BIT,		/* ... Fraglist GSO */
 	/**/NETIF_F_GSO_LAST =		/* last bit, see GSO_MASK */
-		NETIF_F_GSO_UDP_L4_BIT,
+		NETIF_F_GSO_FRAGLIST_BIT,
 
 	NETIF_F_FCOE_CRC_BIT,		/* FCoE CRC32 */
 	NETIF_F_SCTP_CRC_BIT,		/* SCTP checksum offload */
@@ -80,6 +81,7 @@ enum {
 
 	NETIF_F_GRO_HW_BIT,		/* Hardware Generic receive offload */
 	NETIF_F_HW_TLS_RECORD_BIT,	/* Offload TLS record */
+	NETIF_F_GRO_FRAGLIST_BIT,	/* Fraglist GRO */
 
 	/*
 	 * Add your fresh new feature above and remember to update
@@ -150,6 +152,8 @@ enum {
 #define NETIF_F_GSO_UDP_L4	__NETIF_F(GSO_UDP_L4)
 #define NETIF_F_HW_TLS_TX	__NETIF_F(HW_TLS_TX)
 #define NETIF_F_HW_TLS_RX	__NETIF_F(HW_TLS_RX)
+#define NETIF_F_GRO_FRAGLIST	__NETIF_F(GRO_FRAGLIST)
+#define NETIF_F_GSO_FRAGLIST	__NETIF_F(GSO_FRAGLIST)
 
 /* Finds the next feature with the highest number of the range of start till 0.
  */
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 78e9c6c1b131..fcc76b890f50 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -4570,6 +4570,7 @@ static inline bool net_gso_ok(netdev_features_t features, int gso_type)
 	BUILD_BUG_ON(SKB_GSO_ESP != (NETIF_F_GSO_ESP >> NETIF_F_GSO_SHIFT));
 	BUILD_BUG_ON(SKB_GSO_UDP != (NETIF_F_GSO_UDP >> NETIF_F_GSO_SHIFT));
 	BUILD_BUG_ON(SKB_GSO_UDP_L4 != (NETIF_F_GSO_UDP_L4 >> NETIF_F_GSO_SHIFT));
+	BUILD_BUG_ON(SKB_GSO_FRAGLIST != (NETIF_F_GSO_FRAGLIST >> NETIF_F_GSO_SHIFT));
 
 	return (features & feature) == feature;
 }
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 26beae7db264..23aaaf08e1e9 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -592,6 +592,8 @@ enum {
 	SKB_GSO_UDP = 1 << 16,
 
 	SKB_GSO_UDP_L4 = 1 << 17,
+
+	SKB_GSO_FRAGLIST = 1 << 18,
 };
 
 #if BITS_PER_LONG > 32
diff --git a/net/ethtool/common.c b/net/ethtool/common.c
index e621b1694d2f..c7b8956c3827 100644
--- a/net/ethtool/common.c
+++ b/net/ethtool/common.c
@@ -59,6 +59,7 @@ const char netdev_features_strings[NETDEV_FEATURE_COUNT][ETH_GSTRING_LEN] = {
 	[NETIF_F_HW_TLS_RECORD_BIT] =	"tls-hw-record",
 	[NETIF_F_HW_TLS_TX_BIT] =	 "tls-hw-tx-offload",
 	[NETIF_F_HW_TLS_RX_BIT] =	 "tls-hw-rx-offload",
+	[NETIF_F_GRO_FRAGLIST_BIT] =	 "rx-gro-list",
 };
 
 const char
-- 
cgit v1.2.3-59-g8ed1b


From 1a3c998f3a27ab6ecf56bdbb17e27e55fd6d47cd Mon Sep 17 00:00:00 2001
From: Steffen Klassert <steffen.klassert@secunet.com>
Date: Sat, 25 Jan 2020 11:26:43 +0100
Subject: net: Add a netdev software feature set that defaults to off.

The previous patch added the NETIF_F_GRO_FRAGLIST feature.
This is a software feature that should default to off.
Current software features default to on, so add a new
feature set that defaults to off.

Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdev_features.h | 3 +++
 net/core/dev.c                  | 2 +-
 2 files changed, 4 insertions(+), 1 deletion(-)

diff --git a/include/linux/netdev_features.h b/include/linux/netdev_features.h
index b239507da2a0..34d050bb1ae6 100644
--- a/include/linux/netdev_features.h
+++ b/include/linux/netdev_features.h
@@ -230,6 +230,9 @@ static inline int find_next_netdev_feature(u64 feature, unsigned long start)
 /* changeable features with no special hardware requirements */
 #define NETIF_F_SOFT_FEATURES	(NETIF_F_GSO | NETIF_F_GRO)
 
+/* Changeable features with no special hardware requirements that defaults to off. */
+#define NETIF_F_SOFT_FEATURES_OFF	NETIF_F_GRO_FRAGLIST
+
 #define NETIF_F_VLAN_FEATURES	(NETIF_F_HW_VLAN_CTAG_FILTER | \
 				 NETIF_F_HW_VLAN_CTAG_RX | \
 				 NETIF_F_HW_VLAN_CTAG_TX | \
diff --git a/net/core/dev.c b/net/core/dev.c
index c806b078097b..a3b154a4b4f9 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -9283,7 +9283,7 @@ int register_netdevice(struct net_device *dev)
 	/* Transfer changeable features to wanted_features and enable
 	 * software offloads (GSO and GRO).
 	 */
-	dev->hw_features |= NETIF_F_SOFT_FEATURES;
+	dev->hw_features |= (NETIF_F_SOFT_FEATURES | NETIF_F_SOFT_FEATURES_OFF);
 	dev->features |= NETIF_F_SOFT_FEATURES;
 
 	if (dev->netdev_ops->ndo_udp_tunnel_add) {
-- 
cgit v1.2.3-59-g8ed1b


From 3a1296a38d0cf62bffb9a03c585cbd5dbf15d596 Mon Sep 17 00:00:00 2001
From: Steffen Klassert <steffen.klassert@secunet.com>
Date: Sat, 25 Jan 2020 11:26:44 +0100
Subject: net: Support GRO/GSO fraglist chaining.

This patch adds the core functions to chain/unchain
GSO skbs at the frag_list pointer. This also adds
a new GSO type SKB_GSO_FRAGLIST and a is_flist
flag to napi_gro_cb which indicates that this
flow will be GROed by fraglist chaining.

Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  4 ++-
 include/linux/skbuff.h    |  2 ++
 net/core/dev.c            |  2 +-
 net/core/skbuff.c         | 91 +++++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 97 insertions(+), 2 deletions(-)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index fcc76b890f50..20445f94eb1c 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -2326,7 +2326,8 @@ struct napi_gro_cb {
 	/* Number of gro_receive callbacks this packet already went through */
 	u8 recursion_counter:4;
 
-	/* 1 bit hole */
+	/* GRO is done by frag_list pointer chaining. */
+	u8	is_flist:1;
 
 	/* used to support CHECKSUM_COMPLETE for tunneling protocols */
 	__wsum	csum;
@@ -2694,6 +2695,7 @@ struct net_device *dev_get_by_napi_id(unsigned int napi_id);
 int netdev_get_name(struct net *net, char *name, int ifindex);
 int dev_restart(struct net_device *dev);
 int skb_gro_receive(struct sk_buff *p, struct sk_buff *skb);
+int skb_gro_receive_list(struct sk_buff *p, struct sk_buff *skb);
 
 static inline unsigned int skb_gro_offset(const struct sk_buff *skb)
 {
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 23aaaf08e1e9..3d13a4b717e9 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -3535,6 +3535,8 @@ void skb_scrub_packet(struct sk_buff *skb, bool xnet);
 bool skb_gso_validate_network_len(const struct sk_buff *skb, unsigned int mtu);
 bool skb_gso_validate_mac_len(const struct sk_buff *skb, unsigned int len);
 struct sk_buff *skb_segment(struct sk_buff *skb, netdev_features_t features);
+struct sk_buff *skb_segment_list(struct sk_buff *skb, netdev_features_t features,
+				 unsigned int offset);
 struct sk_buff *skb_vlan_untag(struct sk_buff *skb);
 int skb_ensure_writable(struct sk_buff *skb, int write_len);
 int __skb_vlan_pop(struct sk_buff *skb, u16 *vlan_tci);
diff --git a/net/core/dev.c b/net/core/dev.c
index a3b154a4b4f9..ce8900dbd9ea 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3249,7 +3249,7 @@ struct sk_buff *__skb_gso_segment(struct sk_buff *skb,
 
 	segs = skb_mac_gso_segment(skb, features);
 
-	if (unlikely(skb_needs_check(skb, tx_path) && !IS_ERR(segs)))
+	if (segs != skb && unlikely(skb_needs_check(skb, tx_path) && !IS_ERR(segs)))
 		skb_warn_bad_offload(skb);
 
 	return segs;
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 48a7029529c9..864cb9e9622f 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -3639,6 +3639,97 @@ static inline skb_frag_t skb_head_frag_to_page_desc(struct sk_buff *frag_skb)
 	return head_frag;
 }
 
+struct sk_buff *skb_segment_list(struct sk_buff *skb,
+				 netdev_features_t features,
+				 unsigned int offset)
+{
+	struct sk_buff *list_skb = skb_shinfo(skb)->frag_list;
+	unsigned int tnl_hlen = skb_tnl_header_len(skb);
+	unsigned int delta_truesize = 0;
+	unsigned int delta_len = 0;
+	struct sk_buff *tail = NULL;
+	struct sk_buff *nskb;
+
+	skb_push(skb, -skb_network_offset(skb) + offset);
+
+	skb_shinfo(skb)->frag_list = NULL;
+
+	do {
+		nskb = list_skb;
+		list_skb = list_skb->next;
+
+		if (!tail)
+			skb->next = nskb;
+		else
+			tail->next = nskb;
+
+		tail = nskb;
+
+		delta_len += nskb->len;
+		delta_truesize += nskb->truesize;
+
+		skb_push(nskb, -skb_network_offset(nskb) + offset);
+
+		 __copy_skb_header(nskb, skb);
+
+		skb_headers_offset_update(nskb, skb_headroom(nskb) - skb_headroom(skb));
+		skb_copy_from_linear_data_offset(skb, -tnl_hlen,
+						 nskb->data - tnl_hlen,
+						 offset + tnl_hlen);
+
+		if (skb_needs_linearize(nskb, features) &&
+		    __skb_linearize(nskb))
+			goto err_linearize;
+
+	} while (list_skb);
+
+	skb->truesize = skb->truesize - delta_truesize;
+	skb->data_len = skb->data_len - delta_len;
+	skb->len = skb->len - delta_len;
+
+	skb_gso_reset(skb);
+
+	skb->prev = tail;
+
+	if (skb_needs_linearize(skb, features) &&
+	    __skb_linearize(skb))
+		goto err_linearize;
+
+	skb_get(skb);
+
+	return skb;
+
+err_linearize:
+	kfree_skb_list(skb->next);
+	skb->next = NULL;
+	return ERR_PTR(-ENOMEM);
+}
+EXPORT_SYMBOL_GPL(skb_segment_list);
+
+int skb_gro_receive_list(struct sk_buff *p, struct sk_buff *skb)
+{
+	if (unlikely(p->len + skb->len >= 65536))
+		return -E2BIG;
+
+	if (NAPI_GRO_CB(p)->last == p)
+		skb_shinfo(p)->frag_list = skb;
+	else
+		NAPI_GRO_CB(p)->last->next = skb;
+
+	skb_pull(skb, skb_gro_offset(skb));
+
+	NAPI_GRO_CB(p)->last = skb;
+	NAPI_GRO_CB(p)->count++;
+	p->data_len += skb->len;
+	p->truesize += skb->truesize;
+	p->len += skb->len;
+
+	NAPI_GRO_CB(skb)->same_flow = 1;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(skb_gro_receive_list);
+
 /**
  *	skb_segment - Perform protocol segmentation on skb.
  *	@head_skb: buffer to segment
-- 
cgit v1.2.3-59-g8ed1b


From 9fd1ff5d2ac7181844735806b0a703c942365291 Mon Sep 17 00:00:00 2001
From: Steffen Klassert <steffen.klassert@secunet.com>
Date: Sat, 25 Jan 2020 11:26:45 +0100
Subject: udp: Support UDP fraglist GRO/GSO.

This patch extends UDP GRO to support fraglist GRO/GSO
by using the previously introduced infrastructure.
If the feature is enabled, all UDP packets are going to
fraglist GRO (local input and forward).

After validating the csum,  we mark ip_summed as
CHECKSUM_UNNECESSARY for fraglist GRO packets to
make sure that the csum is not touched.

Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/udp.h      |   2 +-
 net/ipv4/udp_offload.c | 104 ++++++++++++++++++++++++++++++++++++++-----------
 net/ipv6/udp_offload.c |  27 ++++++++++++-
 3 files changed, 107 insertions(+), 26 deletions(-)

diff --git a/include/net/udp.h b/include/net/udp.h
index bad74f780831..44e0e52b585c 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -167,7 +167,7 @@ typedef struct sock *(*udp_lookup_t)(struct sk_buff *skb, __be16 sport,
 				     __be16 dport);
 
 struct sk_buff *udp_gro_receive(struct list_head *head, struct sk_buff *skb,
-				struct udphdr *uh, udp_lookup_t lookup);
+				struct udphdr *uh, struct sock *sk);
 int udp_gro_complete(struct sk_buff *skb, int nhoff, udp_lookup_t lookup);
 
 struct sk_buff *__udp_gso_segment(struct sk_buff *gso_skb,
diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c
index b25e42100ceb..1a98583a79f4 100644
--- a/net/ipv4/udp_offload.c
+++ b/net/ipv4/udp_offload.c
@@ -184,6 +184,20 @@ out_unlock:
 }
 EXPORT_SYMBOL(skb_udp_tunnel_segment);
 
+static struct sk_buff *__udp_gso_segment_list(struct sk_buff *skb,
+					      netdev_features_t features)
+{
+	unsigned int mss = skb_shinfo(skb)->gso_size;
+
+	skb = skb_segment_list(skb, features, skb_mac_header_len(skb));
+	if (IS_ERR(skb))
+		return skb;
+
+	udp_hdr(skb)->len = htons(sizeof(struct udphdr) + mss);
+
+	return skb;
+}
+
 struct sk_buff *__udp_gso_segment(struct sk_buff *gso_skb,
 				  netdev_features_t features)
 {
@@ -196,6 +210,9 @@ struct sk_buff *__udp_gso_segment(struct sk_buff *gso_skb,
 	__sum16 check;
 	__be16 newlen;
 
+	if (skb_shinfo(gso_skb)->gso_type & SKB_GSO_FRAGLIST)
+		return __udp_gso_segment_list(gso_skb, features);
+
 	mss = skb_shinfo(gso_skb)->gso_size;
 	if (gso_skb->len <= sizeof(*uh) + mss)
 		return ERR_PTR(-EINVAL);
@@ -354,6 +371,7 @@ static struct sk_buff *udp_gro_receive_segment(struct list_head *head,
 	struct udphdr *uh2;
 	struct sk_buff *p;
 	unsigned int ulen;
+	int ret = 0;
 
 	/* requires non zero csum, for symmetry with GSO */
 	if (!uh->check) {
@@ -369,7 +387,6 @@ static struct sk_buff *udp_gro_receive_segment(struct list_head *head,
 	}
 	/* pull encapsulating udp header */
 	skb_gro_pull(skb, sizeof(struct udphdr));
-	skb_gro_postpull_rcsum(skb, uh, sizeof(struct udphdr));
 
 	list_for_each_entry(p, head, list) {
 		if (!NAPI_GRO_CB(p)->same_flow)
@@ -383,14 +400,40 @@ static struct sk_buff *udp_gro_receive_segment(struct list_head *head,
 			continue;
 		}
 
+		if (NAPI_GRO_CB(skb)->is_flist != NAPI_GRO_CB(p)->is_flist) {
+			NAPI_GRO_CB(skb)->flush = 1;
+			return p;
+		}
+
 		/* Terminate the flow on len mismatch or if it grow "too much".
 		 * Under small packet flood GRO count could elsewhere grow a lot
 		 * leading to excessive truesize values.
 		 * On len mismatch merge the first packet shorter than gso_size,
 		 * otherwise complete the GRO packet.
 		 */
-		if (ulen > ntohs(uh2->len) || skb_gro_receive(p, skb) ||
-		    ulen != ntohs(uh2->len) ||
+		if (ulen > ntohs(uh2->len)) {
+			pp = p;
+		} else {
+			if (NAPI_GRO_CB(skb)->is_flist) {
+				if (!pskb_may_pull(skb, skb_gro_offset(skb))) {
+					NAPI_GRO_CB(skb)->flush = 1;
+					return NULL;
+				}
+				if ((skb->ip_summed != p->ip_summed) ||
+				    (skb->csum_level != p->csum_level)) {
+					NAPI_GRO_CB(skb)->flush = 1;
+					return NULL;
+				}
+				ret = skb_gro_receive_list(p, skb);
+			} else {
+				skb_gro_postpull_rcsum(skb, uh,
+						       sizeof(struct udphdr));
+
+				ret = skb_gro_receive(p, skb);
+			}
+		}
+
+		if (ret || ulen != ntohs(uh2->len) ||
 		    NAPI_GRO_CB(p)->count >= UDP_GRO_CNT_MAX)
 			pp = p;
 
@@ -401,36 +444,29 @@ static struct sk_buff *udp_gro_receive_segment(struct list_head *head,
 	return NULL;
 }
 
-INDIRECT_CALLABLE_DECLARE(struct sock *udp6_lib_lookup_skb(struct sk_buff *skb,
-						   __be16 sport, __be16 dport));
 struct sk_buff *udp_gro_receive(struct list_head *head, struct sk_buff *skb,
-				struct udphdr *uh, udp_lookup_t lookup)
+				struct udphdr *uh, struct sock *sk)
 {
 	struct sk_buff *pp = NULL;
 	struct sk_buff *p;
 	struct udphdr *uh2;
 	unsigned int off = skb_gro_offset(skb);
 	int flush = 1;
-	struct sock *sk;
 
-	rcu_read_lock();
-	sk = INDIRECT_CALL_INET(lookup, udp6_lib_lookup_skb,
-				udp4_lib_lookup_skb, skb, uh->source, uh->dest);
-	if (!sk)
-		goto out_unlock;
+	if (skb->dev->features & NETIF_F_GRO_FRAGLIST)
+		NAPI_GRO_CB(skb)->is_flist = sk ? !udp_sk(sk)->gro_enabled: 1;
 
-	if (udp_sk(sk)->gro_enabled) {
+	if ((sk && udp_sk(sk)->gro_enabled) || NAPI_GRO_CB(skb)->is_flist) {
 		pp = call_gro_receive(udp_gro_receive_segment, head, skb);
-		rcu_read_unlock();
 		return pp;
 	}
 
-	if (NAPI_GRO_CB(skb)->encap_mark ||
+	if (!sk || NAPI_GRO_CB(skb)->encap_mark ||
 	    (skb->ip_summed != CHECKSUM_PARTIAL &&
 	     NAPI_GRO_CB(skb)->csum_cnt == 0 &&
 	     !NAPI_GRO_CB(skb)->csum_valid) ||
 	    !udp_sk(sk)->gro_receive)
-		goto out_unlock;
+		goto out;
 
 	/* mark that this skb passed once through the tunnel gro layer */
 	NAPI_GRO_CB(skb)->encap_mark = 1;
@@ -457,8 +493,7 @@ struct sk_buff *udp_gro_receive(struct list_head *head, struct sk_buff *skb,
 	skb_gro_postpull_rcsum(skb, uh, sizeof(struct udphdr));
 	pp = call_gro_receive_sk(udp_sk(sk)->gro_receive, sk, head, skb);
 
-out_unlock:
-	rcu_read_unlock();
+out:
 	skb_gro_flush_final(skb, pp, flush);
 	return pp;
 }
@@ -468,8 +503,10 @@ INDIRECT_CALLABLE_SCOPE
 struct sk_buff *udp4_gro_receive(struct list_head *head, struct sk_buff *skb)
 {
 	struct udphdr *uh = udp_gro_udphdr(skb);
+	struct sk_buff *pp;
+	struct sock *sk;
 
-	if (unlikely(!uh) || !static_branch_unlikely(&udp_encap_needed_key))
+	if (unlikely(!uh))
 		goto flush;
 
 	/* Don't bother verifying checksum if we're going to flush anyway. */
@@ -484,7 +521,11 @@ struct sk_buff *udp4_gro_receive(struct list_head *head, struct sk_buff *skb)
 					     inet_gro_compute_pseudo);
 skip:
 	NAPI_GRO_CB(skb)->is_ipv6 = 0;
-	return udp_gro_receive(head, skb, uh, udp4_lib_lookup_skb);
+	rcu_read_lock();
+	sk = static_branch_unlikely(&udp_encap_needed_key) ? udp4_lib_lookup_skb(skb, uh->source, uh->dest) : NULL;
+	pp = udp_gro_receive(head, skb, uh, sk);
+	rcu_read_unlock();
+	return pp;
 
 flush:
 	NAPI_GRO_CB(skb)->flush = 1;
@@ -517,9 +558,7 @@ int udp_gro_complete(struct sk_buff *skb, int nhoff,
 	rcu_read_lock();
 	sk = INDIRECT_CALL_INET(lookup, udp6_lib_lookup_skb,
 				udp4_lib_lookup_skb, skb, uh->source, uh->dest);
-	if (sk && udp_sk(sk)->gro_enabled) {
-		err = udp_gro_complete_segment(skb);
-	} else if (sk && udp_sk(sk)->gro_complete) {
+	if (sk && udp_sk(sk)->gro_complete) {
 		skb_shinfo(skb)->gso_type = uh->check ? SKB_GSO_UDP_TUNNEL_CSUM
 					: SKB_GSO_UDP_TUNNEL;
 
@@ -529,6 +568,8 @@ int udp_gro_complete(struct sk_buff *skb, int nhoff,
 		skb->encapsulation = 1;
 		err = udp_sk(sk)->gro_complete(sk, skb,
 				nhoff + sizeof(struct udphdr));
+	} else {
+		err = udp_gro_complete_segment(skb);
 	}
 	rcu_read_unlock();
 
@@ -544,6 +585,23 @@ INDIRECT_CALLABLE_SCOPE int udp4_gro_complete(struct sk_buff *skb, int nhoff)
 	const struct iphdr *iph = ip_hdr(skb);
 	struct udphdr *uh = (struct udphdr *)(skb->data + nhoff);
 
+	if (NAPI_GRO_CB(skb)->is_flist) {
+		uh->len = htons(skb->len - nhoff);
+
+		skb_shinfo(skb)->gso_type |= (SKB_GSO_FRAGLIST|SKB_GSO_UDP_L4);
+		skb_shinfo(skb)->gso_segs = NAPI_GRO_CB(skb)->count;
+
+		if (skb->ip_summed == CHECKSUM_UNNECESSARY) {
+			if (skb->csum_level < SKB_MAX_CSUM_LEVEL)
+				skb->csum_level++;
+		} else {
+			skb->ip_summed = CHECKSUM_UNNECESSARY;
+			skb->csum_level = 0;
+		}
+
+		return 0;
+	}
+
 	if (uh->check)
 		uh->check = ~udp_v4_check(skb->len - nhoff, iph->saddr,
 					  iph->daddr, 0);
diff --git a/net/ipv6/udp_offload.c b/net/ipv6/udp_offload.c
index f0d5fc27d0b5..584157a07759 100644
--- a/net/ipv6/udp_offload.c
+++ b/net/ipv6/udp_offload.c
@@ -115,8 +115,10 @@ INDIRECT_CALLABLE_SCOPE
 struct sk_buff *udp6_gro_receive(struct list_head *head, struct sk_buff *skb)
 {
 	struct udphdr *uh = udp_gro_udphdr(skb);
+	struct sk_buff *pp;
+	struct sock *sk;
 
-	if (unlikely(!uh) || !static_branch_unlikely(&udpv6_encap_needed_key))
+	if (unlikely(!uh))
 		goto flush;
 
 	/* Don't bother verifying checksum if we're going to flush anyway. */
@@ -132,7 +134,11 @@ struct sk_buff *udp6_gro_receive(struct list_head *head, struct sk_buff *skb)
 
 skip:
 	NAPI_GRO_CB(skb)->is_ipv6 = 1;
-	return udp_gro_receive(head, skb, uh, udp6_lib_lookup_skb);
+	rcu_read_lock();
+	sk = static_branch_unlikely(&udpv6_encap_needed_key) ? udp6_lib_lookup_skb(skb, uh->source, uh->dest) : NULL;
+	pp = udp_gro_receive(head, skb, uh, sk);
+	rcu_read_unlock();
+	return pp;
 
 flush:
 	NAPI_GRO_CB(skb)->flush = 1;
@@ -144,6 +150,23 @@ INDIRECT_CALLABLE_SCOPE int udp6_gro_complete(struct sk_buff *skb, int nhoff)
 	const struct ipv6hdr *ipv6h = ipv6_hdr(skb);
 	struct udphdr *uh = (struct udphdr *)(skb->data + nhoff);
 
+	if (NAPI_GRO_CB(skb)->is_flist) {
+		uh->len = htons(skb->len - nhoff);
+
+		skb_shinfo(skb)->gso_type |= (SKB_GSO_FRAGLIST|SKB_GSO_UDP_L4);
+		skb_shinfo(skb)->gso_segs = NAPI_GRO_CB(skb)->count;
+
+		if (skb->ip_summed == CHECKSUM_UNNECESSARY) {
+			if (skb->csum_level < SKB_MAX_CSUM_LEVEL)
+				skb->csum_level++;
+		} else {
+			skb->ip_summed = CHECKSUM_UNNECESSARY;
+			skb->csum_level = 0;
+		}
+
+		return 0;
+	}
+
 	if (uh->check)
 		uh->check = ~udp_v6_check(skb->len - nhoff, &ipv6h->saddr,
 					  &ipv6h->daddr, 0);
-- 
cgit v1.2.3-59-g8ed1b


From cd94ef06392ffd49e0a0e1c28bc5cd44f37f1f6b Mon Sep 17 00:00:00 2001
From: Kuniyuki Iwashima <kuniyu@amazon.co.jp>
Date: Sat, 25 Jan 2020 10:41:02 +0000
Subject: soreuseport: Cleanup duplicate initialization of
 more_reuse->max_socks.

reuseport_grow() does not need to initialize the more_reuse->max_socks
again. It is already initialized in __reuseport_alloc().

Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.co.jp>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/sock_reuseport.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/net/core/sock_reuseport.c b/net/core/sock_reuseport.c
index f19f179538b9..91e9f2223c39 100644
--- a/net/core/sock_reuseport.c
+++ b/net/core/sock_reuseport.c
@@ -107,7 +107,6 @@ static struct sock_reuseport *reuseport_grow(struct sock_reuseport *reuse)
 	if (!more_reuse)
 		return NULL;
 
-	more_reuse->max_socks = more_socks_size;
 	more_reuse->num_socks = reuse->num_socks;
 	more_reuse->prog = reuse->prog;
 	more_reuse->reuseport_id = reuse->reuseport_id;
-- 
cgit v1.2.3-59-g8ed1b


From 48b3a1379fc6603c1ff26893ea05322c1c41e31c Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@mellanox.com>
Date: Sat, 25 Jan 2020 12:17:06 +0100
Subject: net: call call_netdevice_unregister_net_notifiers from unregister

The function does the same thing as the existing code, so rather call
call_netdevice_unregister_net_notifiers() instead of code duplication.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/dev.c | 14 +++-----------
 1 file changed, 3 insertions(+), 11 deletions(-)

diff --git a/net/core/dev.c b/net/core/dev.c
index ce8900dbd9ea..ee4b1e64d663 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1764,7 +1764,6 @@ EXPORT_SYMBOL(register_netdevice_notifier);
 
 int unregister_netdevice_notifier(struct notifier_block *nb)
 {
-	struct net_device *dev;
 	struct net *net;
 	int err;
 
@@ -1775,16 +1774,9 @@ int unregister_netdevice_notifier(struct notifier_block *nb)
 	if (err)
 		goto unlock;
 
-	for_each_net(net) {
-		for_each_netdev(net, dev) {
-			if (dev->flags & IFF_UP) {
-				call_netdevice_notifier(nb, NETDEV_GOING_DOWN,
-							dev);
-				call_netdevice_notifier(nb, NETDEV_DOWN, dev);
-			}
-			call_netdevice_notifier(nb, NETDEV_UNREGISTER, dev);
-		}
-	}
+	for_each_net(net)
+		call_netdevice_unregister_net_notifiers(nb, net);
+
 unlock:
 	rtnl_unlock();
 	up_write(&pernet_ops_rwsem);
-- 
cgit v1.2.3-59-g8ed1b


From 1f637703d8b63f1ba411b4c798e998e3f828b6cb Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@mellanox.com>
Date: Sat, 25 Jan 2020 12:17:07 +0100
Subject: net: push code from net notifier reg/unreg into helpers

Push the code which is done under rtnl lock in net notifier register and
unregister function into separate helpers.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/dev.c | 60 +++++++++++++++++++++++++++++++++++++---------------------
 1 file changed, 38 insertions(+), 22 deletions(-)

diff --git a/net/core/dev.c b/net/core/dev.c
index ee4b1e64d663..b521b509a653 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1784,6 +1784,42 @@ unlock:
 }
 EXPORT_SYMBOL(unregister_netdevice_notifier);
 
+static int __register_netdevice_notifier_net(struct net *net,
+					     struct notifier_block *nb,
+					     bool ignore_call_fail)
+{
+	int err;
+
+	err = raw_notifier_chain_register(&net->netdev_chain, nb);
+	if (err)
+		return err;
+	if (dev_boot_phase)
+		return 0;
+
+	err = call_netdevice_register_net_notifiers(nb, net);
+	if (err && !ignore_call_fail)
+		goto chain_unregister;
+
+	return 0;
+
+chain_unregister:
+	raw_notifier_chain_unregister(&net->netdev_chain, nb);
+	return err;
+}
+
+static int __unregister_netdevice_notifier_net(struct net *net,
+					       struct notifier_block *nb)
+{
+	int err;
+
+	err = raw_notifier_chain_unregister(&net->netdev_chain, nb);
+	if (err)
+		return err;
+
+	call_netdevice_unregister_net_notifiers(nb, net);
+	return 0;
+}
+
 /**
  * register_netdevice_notifier_net - register a per-netns network notifier block
  * @net: network namespace
@@ -1804,23 +1840,9 @@ int register_netdevice_notifier_net(struct net *net, struct notifier_block *nb)
 	int err;
 
 	rtnl_lock();
-	err = raw_notifier_chain_register(&net->netdev_chain, nb);
-	if (err)
-		goto unlock;
-	if (dev_boot_phase)
-		goto unlock;
-
-	err = call_netdevice_register_net_notifiers(nb, net);
-	if (err)
-		goto chain_unregister;
-
-unlock:
+	err = __register_netdevice_notifier_net(net, nb, false);
 	rtnl_unlock();
 	return err;
-
-chain_unregister:
-	raw_notifier_chain_unregister(&netdev_chain, nb);
-	goto unlock;
 }
 EXPORT_SYMBOL(register_netdevice_notifier_net);
 
@@ -1846,13 +1868,7 @@ int unregister_netdevice_notifier_net(struct net *net,
 	int err;
 
 	rtnl_lock();
-	err = raw_notifier_chain_unregister(&net->netdev_chain, nb);
-	if (err)
-		goto unlock;
-
-	call_netdevice_unregister_net_notifiers(nb, net);
-
-unlock:
+	err = __unregister_netdevice_notifier_net(net, nb);
 	rtnl_unlock();
 	return err;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 93642e14bd50e59b11cf6389ce3fc243e932777a Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@mellanox.com>
Date: Sat, 25 Jan 2020 12:17:08 +0100
Subject: net: introduce dev_net notifier register/unregister variants

Introduce dev_net variants of netdev notifier register/unregister functions
and allow per-net notifier to follow the netdevice into the namespace it is
moved to.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 17 +++++++++++++++++
 net/core/dev.c            | 46 ++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 63 insertions(+)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 20445f94eb1c..4626188a754b 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -939,6 +939,11 @@ struct netdev_name_node {
 int netdev_name_node_alt_create(struct net_device *dev, const char *name);
 int netdev_name_node_alt_destroy(struct net_device *dev, const char *name);
 
+struct netdev_net_notifier {
+	struct list_head list;
+	struct notifier_block *nb;
+};
+
 /*
  * This structure defines the management hooks for network devices.
  * The following hooks can be defined; unless noted otherwise, they are
@@ -1793,6 +1798,10 @@ enum netdev_priv_flags {
  *
  *	@wol_enabled:	Wake-on-LAN is enabled
  *
+ *	@net_notifier_list:	List of per-net netdev notifier block
+ *				that follow this device when it is moved
+ *				to another network namespace.
+ *
  *	FIXME: cleanup struct net_device such that network protocol info
  *	moves out.
  */
@@ -2085,6 +2094,8 @@ struct net_device {
 	struct lock_class_key	addr_list_lock_key;
 	bool			proto_down;
 	unsigned		wol_enabled:1;
+
+	struct list_head	net_notifier_list;
 };
 #define to_net_dev(d) container_of(d, struct net_device, dev)
 
@@ -2529,6 +2540,12 @@ int unregister_netdevice_notifier(struct notifier_block *nb);
 int register_netdevice_notifier_net(struct net *net, struct notifier_block *nb);
 int unregister_netdevice_notifier_net(struct net *net,
 				      struct notifier_block *nb);
+int register_netdevice_notifier_dev_net(struct net_device *dev,
+					struct notifier_block *nb,
+					struct netdev_net_notifier *nn);
+int unregister_netdevice_notifier_dev_net(struct net_device *dev,
+					  struct notifier_block *nb,
+					  struct netdev_net_notifier *nn);
 
 struct netdev_notifier_info {
 	struct net_device	*dev;
diff --git a/net/core/dev.c b/net/core/dev.c
index b521b509a653..38bc35da39f7 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1874,6 +1874,48 @@ int unregister_netdevice_notifier_net(struct net *net,
 }
 EXPORT_SYMBOL(unregister_netdevice_notifier_net);
 
+int register_netdevice_notifier_dev_net(struct net_device *dev,
+					struct notifier_block *nb,
+					struct netdev_net_notifier *nn)
+{
+	int err;
+
+	rtnl_lock();
+	err = __register_netdevice_notifier_net(dev_net(dev), nb, false);
+	if (!err) {
+		nn->nb = nb;
+		list_add(&nn->list, &dev->net_notifier_list);
+	}
+	rtnl_unlock();
+	return err;
+}
+EXPORT_SYMBOL(register_netdevice_notifier_dev_net);
+
+int unregister_netdevice_notifier_dev_net(struct net_device *dev,
+					  struct notifier_block *nb,
+					  struct netdev_net_notifier *nn)
+{
+	int err;
+
+	rtnl_lock();
+	list_del(&nn->list);
+	err = __unregister_netdevice_notifier_net(dev_net(dev), nb);
+	rtnl_unlock();
+	return err;
+}
+EXPORT_SYMBOL(unregister_netdevice_notifier_dev_net);
+
+static void move_netdevice_notifiers_dev_net(struct net_device *dev,
+					     struct net *net)
+{
+	struct netdev_net_notifier *nn;
+
+	list_for_each_entry(nn, &dev->net_notifier_list, list) {
+		__unregister_netdevice_notifier_net(dev_net(dev), nn->nb);
+		__register_netdevice_notifier_net(net, nn->nb, true);
+	}
+}
+
 /**
  *	call_netdevice_notifiers_info - call all network notifier blocks
  *	@val: value passed unmodified to notifier function
@@ -9786,6 +9828,7 @@ struct net_device *alloc_netdev_mqs(int sizeof_priv, const char *name,
 	INIT_LIST_HEAD(&dev->adj_list.lower);
 	INIT_LIST_HEAD(&dev->ptype_all);
 	INIT_LIST_HEAD(&dev->ptype_specific);
+	INIT_LIST_HEAD(&dev->net_notifier_list);
 #ifdef CONFIG_NET_SCHED
 	hash_init(dev->qdisc_hash);
 #endif
@@ -10049,6 +10092,9 @@ int dev_change_net_namespace(struct net_device *dev, struct net *net, const char
 	kobject_uevent(&dev->dev.kobj, KOBJ_REMOVE);
 	netdev_adjacent_del_links(dev);
 
+	/* Move per-net netdevice notifiers that are following the netdevice */
+	move_netdevice_notifiers_dev_net(dev, net);
+
 	/* Actually switch the network namespace */
 	dev_net_set(dev, net);
 	dev->ifindex = new_ifindex;
-- 
cgit v1.2.3-59-g8ed1b


From d48834f9d4b4e7611cc84ac2c5018f1b420d59c2 Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@mellanox.com>
Date: Sat, 25 Jan 2020 12:17:09 +0100
Subject: mlx5: Use dev_net netdevice notifier registrations

Register the dev_net notifier and allow the per-net notifier to follow
the device into different namespace.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/fs.h     |  1 +
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c   |  3 ++-
 drivers/net/ethernet/mellanox/mlx5/core/en_rep.c    | 13 ++++++++++---
 drivers/net/ethernet/mellanox/mlx5/core/en_rep.h    |  1 +
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c     |  9 +++++++--
 drivers/net/ethernet/mellanox/mlx5/core/lag.c       |  8 +++++---
 drivers/net/ethernet/mellanox/mlx5/core/lag.h       |  1 +
 drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h |  2 +-
 8 files changed, 28 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/fs.h b/drivers/net/ethernet/mellanox/mlx5/core/en/fs.h
index d48292ccda29..0416f7712109 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/fs.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/fs.h
@@ -21,6 +21,7 @@ struct mlx5e_tc_table {
 	DECLARE_HASHTABLE(hairpin_tbl, 8);
 
 	struct notifier_block     netdevice_nb;
+	struct netdev_net_notifier	netdevice_nn;
 };
 
 struct mlx5e_flow_table {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index f3600ae4b0a1..454d3459bd8b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -5144,6 +5144,7 @@ static void mlx5e_nic_enable(struct mlx5e_priv *priv)
 
 static void mlx5e_nic_disable(struct mlx5e_priv *priv)
 {
+	struct net_device *netdev = priv->netdev;
 	struct mlx5_core_dev *mdev = priv->mdev;
 
 #ifdef CONFIG_MLX5_CORE_EN_DCB
@@ -5164,7 +5165,7 @@ static void mlx5e_nic_disable(struct mlx5e_priv *priv)
 		mlx5e_monitor_counter_cleanup(priv);
 
 	mlx5e_disable_async_events(priv);
-	mlx5_lag_remove(mdev);
+	mlx5_lag_remove(mdev, netdev);
 }
 
 int mlx5e_update_nic_rx(struct mlx5e_priv *priv)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index 09061b4c43af..7b48ccacebe2 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -1731,7 +1731,9 @@ static int mlx5e_init_uplink_rep_tx(struct mlx5e_rep_priv *rpriv)
 	/* init indirect block notifications */
 	INIT_LIST_HEAD(&uplink_priv->tc_indr_block_priv_list);
 	uplink_priv->netdevice_nb.notifier_call = mlx5e_nic_rep_netdevice_event;
-	err = register_netdevice_notifier(&uplink_priv->netdevice_nb);
+	err = register_netdevice_notifier_dev_net(rpriv->netdev,
+						  &uplink_priv->netdevice_nb,
+						  &uplink_priv->netdevice_nn);
 	if (err) {
 		mlx5_core_err(priv->mdev, "Failed to register netdev notifier\n");
 		goto tc_esw_cleanup;
@@ -1770,8 +1772,12 @@ destroy_tises:
 
 static void mlx5e_cleanup_uplink_rep_tx(struct mlx5e_rep_priv *rpriv)
 {
+	struct mlx5_rep_uplink_priv *uplink_priv = &rpriv->uplink_priv;
+
 	/* clean indirect TC block notifications */
-	unregister_netdevice_notifier(&rpriv->uplink_priv.netdevice_nb);
+	unregister_netdevice_notifier_dev_net(rpriv->netdev,
+					      &uplink_priv->netdevice_nb,
+					      &uplink_priv->netdevice_nn);
 	mlx5e_rep_indr_clean_block_privs(rpriv);
 
 	/* delete shared tc flow table */
@@ -1855,6 +1861,7 @@ static void mlx5e_uplink_rep_enable(struct mlx5e_priv *priv)
 
 static void mlx5e_uplink_rep_disable(struct mlx5e_priv *priv)
 {
+	struct net_device *netdev = priv->netdev;
 	struct mlx5_core_dev *mdev = priv->mdev;
 	struct mlx5e_rep_priv *rpriv = priv->ppriv;
 
@@ -1863,7 +1870,7 @@ static void mlx5e_uplink_rep_disable(struct mlx5e_priv *priv)
 #endif
 	mlx5_notifier_unregister(mdev, &priv->events_nb);
 	cancel_work_sync(&rpriv->uplink_priv.reoffload_flows_work);
-	mlx5_lag_remove(mdev);
+	mlx5_lag_remove(mdev, netdev);
 }
 
 static MLX5E_DEFINE_STATS_GRP(sw_rep, 0);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h
index 31f83c8adcc9..3f756d51435f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h
@@ -73,6 +73,7 @@ struct mlx5_rep_uplink_priv {
 	 */
 	struct list_head	    tc_indr_block_priv_list;
 	struct notifier_block	    netdevice_nb;
+	struct netdev_net_notifier  netdevice_nn;
 
 	struct mlx5_tun_entropy tun_entropy;
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 915afb2a9b18..74091f72c9a8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -4251,7 +4251,10 @@ int mlx5e_tc_nic_init(struct mlx5e_priv *priv)
 		return err;
 
 	tc->netdevice_nb.notifier_call = mlx5e_tc_netdev_event;
-	if (register_netdevice_notifier(&tc->netdevice_nb)) {
+	err = register_netdevice_notifier_dev_net(priv->netdev,
+						  &tc->netdevice_nb,
+						  &tc->netdevice_nn);
+	if (err) {
 		tc->netdevice_nb.notifier_call = NULL;
 		mlx5_core_warn(priv->mdev, "Failed to register netdev notifier\n");
 	}
@@ -4273,7 +4276,9 @@ void mlx5e_tc_nic_cleanup(struct mlx5e_priv *priv)
 	struct mlx5e_tc_table *tc = &priv->fs.tc;
 
 	if (tc->netdevice_nb.notifier_call)
-		unregister_netdevice_notifier(&tc->netdevice_nb);
+		unregister_netdevice_notifier_dev_net(priv->netdev,
+						      &tc->netdevice_nb,
+						      &tc->netdevice_nn);
 
 	mutex_destroy(&tc->mod_hdr.lock);
 	mutex_destroy(&tc->hairpin_tbl_lock);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag.c b/drivers/net/ethernet/mellanox/mlx5/core/lag.c
index fc0d9583475d..b91eabc09fbc 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lag.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lag.c
@@ -586,7 +586,8 @@ void mlx5_lag_add(struct mlx5_core_dev *dev, struct net_device *netdev)
 
 	if (!ldev->nb.notifier_call) {
 		ldev->nb.notifier_call = mlx5_lag_netdev_event;
-		if (register_netdevice_notifier(&ldev->nb)) {
+		if (register_netdevice_notifier_dev_net(netdev, &ldev->nb,
+							&ldev->nn)) {
 			ldev->nb.notifier_call = NULL;
 			mlx5_core_err(dev, "Failed to register LAG netdev notifier\n");
 		}
@@ -599,7 +600,7 @@ void mlx5_lag_add(struct mlx5_core_dev *dev, struct net_device *netdev)
 }
 
 /* Must be called with intf_mutex held */
-void mlx5_lag_remove(struct mlx5_core_dev *dev)
+void mlx5_lag_remove(struct mlx5_core_dev *dev, struct net_device *netdev)
 {
 	struct mlx5_lag *ldev;
 	int i;
@@ -619,7 +620,8 @@ void mlx5_lag_remove(struct mlx5_core_dev *dev)
 
 	if (i == MLX5_MAX_PORTS) {
 		if (ldev->nb.notifier_call)
-			unregister_netdevice_notifier(&ldev->nb);
+			unregister_netdevice_notifier_dev_net(netdev, &ldev->nb,
+							      &ldev->nn);
 		mlx5_lag_mp_cleanup(ldev);
 		cancel_delayed_work_sync(&ldev->bond_work);
 		mlx5_lag_dev_free(ldev);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag.h b/drivers/net/ethernet/mellanox/mlx5/core/lag.h
index f1068aac6406..316ab09e2664 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lag.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lag.h
@@ -44,6 +44,7 @@ struct mlx5_lag {
 	struct workqueue_struct   *wq;
 	struct delayed_work       bond_work;
 	struct notifier_block     nb;
+	struct netdev_net_notifier	nn;
 	struct lag_mp             lag_mp;
 };
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
index da67b28d6e23..fcce9e0fc82c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
@@ -157,7 +157,7 @@ int mlx5_query_qcam_reg(struct mlx5_core_dev *mdev, u32 *qcam,
 			u8 feature_group, u8 access_reg_group);
 
 void mlx5_lag_add(struct mlx5_core_dev *dev, struct net_device *netdev);
-void mlx5_lag_remove(struct mlx5_core_dev *dev);
+void mlx5_lag_remove(struct mlx5_core_dev *dev, struct net_device *netdev);
 
 int mlx5_irq_table_init(struct mlx5_core_dev *dev);
 void mlx5_irq_table_cleanup(struct mlx5_core_dev *dev);
-- 
cgit v1.2.3-59-g8ed1b


From b5e82e3c89c78ee0407ea8e8087af5519b6c7bae Mon Sep 17 00:00:00 2001
From: Lorenzo Bianconi <lorenzo@kernel.org>
Date: Sat, 25 Jan 2020 12:48:50 +0100
Subject: net: socionext: fix possible user-after-free in netsec_process_rx

Fix possible use-after-free in in netsec_process_rx that can occurs if
the first packet is sent to the normal networking stack and the
following one is dropped by the bpf program attached to the xdp hook.
Fix the issue defining the skb pointer in the 'budget' loop

Fixes: ba2b232108d3c ("net: netsec: add XDP support")
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/socionext/netsec.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/socionext/netsec.c b/drivers/net/ethernet/socionext/netsec.c
index 495f6cfdbd91..1f685221f96c 100644
--- a/drivers/net/ethernet/socionext/netsec.c
+++ b/drivers/net/ethernet/socionext/netsec.c
@@ -935,7 +935,6 @@ static int netsec_process_rx(struct netsec_priv *priv, int budget)
 	struct netsec_rx_pkt_info rx_info;
 	enum dma_data_direction dma_dir;
 	struct bpf_prog *xdp_prog;
-	struct sk_buff *skb = NULL;
 	u16 xdp_xmit = 0;
 	u32 xdp_act = 0;
 	int done = 0;
@@ -949,6 +948,7 @@ static int netsec_process_rx(struct netsec_priv *priv, int budget)
 		struct netsec_de *de = dring->vaddr + (DESC_SZ * idx);
 		struct netsec_desc *desc = &dring->desc[idx];
 		struct page *page = virt_to_page(desc->addr);
+		struct sk_buff *skb = NULL;
 		u32 xdp_result = XDP_PASS;
 		u16 pkt_len, desc_len;
 		dma_addr_t dma_handle;
-- 
cgit v1.2.3-59-g8ed1b


From 02758cb6dac31a2b4bd9e535cffbe718acd46404 Mon Sep 17 00:00:00 2001
From: Lorenzo Bianconi <lorenzo@kernel.org>
Date: Sat, 25 Jan 2020 12:48:51 +0100
Subject: net: socionext: fix xdp_result initialization in netsec_process_rx

Fix xdp_result initialization in netsec_process_rx in order to not
increase rx counters if there is no bpf program attached to the xdp hook
and napi_gro_receive returns GRO_DROP

Fixes: ba2b232108d3c ("net: netsec: add XDP support")
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/socionext/netsec.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/socionext/netsec.c b/drivers/net/ethernet/socionext/netsec.c
index 1f685221f96c..e8224b543dfc 100644
--- a/drivers/net/ethernet/socionext/netsec.c
+++ b/drivers/net/ethernet/socionext/netsec.c
@@ -948,8 +948,8 @@ static int netsec_process_rx(struct netsec_priv *priv, int budget)
 		struct netsec_de *de = dring->vaddr + (DESC_SZ * idx);
 		struct netsec_desc *desc = &dring->desc[idx];
 		struct page *page = virt_to_page(desc->addr);
+		u32 xdp_result = NETSEC_XDP_PASS;
 		struct sk_buff *skb = NULL;
-		u32 xdp_result = XDP_PASS;
 		u16 pkt_len, desc_len;
 		dma_addr_t dma_handle;
 		struct xdp_buff xdp;
-- 
cgit v1.2.3-59-g8ed1b


From a85dd3a5170c8812cd835ea968ccadf0ebf1648e Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Sat, 25 Jan 2020 13:42:14 +0100
Subject: net: remove eth_change_mtu

All usage of this function was removed three years ago, and the
function was marked as deprecated:
a52ad514fdf3 ("net: deprecate eth_change_mtu, remove usage")
So I think we can remove it now.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/etherdevice.h |  1 -
 net/ethernet/eth.c          | 16 ----------------
 2 files changed, 17 deletions(-)

diff --git a/include/linux/etherdevice.h b/include/linux/etherdevice.h
index f6564b572d77..8801f1f986e5 100644
--- a/include/linux/etherdevice.h
+++ b/include/linux/etherdevice.h
@@ -43,7 +43,6 @@ __be16 eth_header_parse_protocol(const struct sk_buff *skb);
 int eth_prepare_mac_addr_change(struct net_device *dev, void *p);
 void eth_commit_mac_addr_change(struct net_device *dev, void *p);
 int eth_mac_addr(struct net_device *dev, void *p);
-int eth_change_mtu(struct net_device *dev, int new_mtu);
 int eth_validate_addr(struct net_device *dev);
 
 struct net_device *alloc_etherdev_mqs(int sizeof_priv, unsigned int txqs,
diff --git a/net/ethernet/eth.c b/net/ethernet/eth.c
index 9040fe55e0f5..c8b903302ff2 100644
--- a/net/ethernet/eth.c
+++ b/net/ethernet/eth.c
@@ -335,22 +335,6 @@ int eth_mac_addr(struct net_device *dev, void *p)
 }
 EXPORT_SYMBOL(eth_mac_addr);
 
-/**
- * eth_change_mtu - set new MTU size
- * @dev: network device
- * @new_mtu: new Maximum Transfer Unit
- *
- * Allow changing MTU size. Needs to be overridden for devices
- * supporting jumbo frames.
- */
-int eth_change_mtu(struct net_device *dev, int new_mtu)
-{
-	netdev_warn(dev, "%s is deprecated\n", __func__);
-	dev->mtu = new_mtu;
-	return 0;
-}
-EXPORT_SYMBOL(eth_change_mtu);
-
 int eth_validate_addr(struct net_device *dev)
 {
 	if (!is_valid_ether_addr(dev->dev_addr))
-- 
cgit v1.2.3-59-g8ed1b


From 6dc43cd3aae0ffeeaebe427e1bf5e5faf9de7d42 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Sat, 25 Jan 2020 23:01:11 +0200
Subject: net: dsa: Fix use-after-free in probing of DSA switch tree

DSA sets up a switch tree little by little. Every switch of the N
members of the tree calls dsa_register_switch, and (N - 1) will just
touch the dst->ports list with their ports and quickly exit. Only the
last switch that calls dsa_register_switch will find all DSA links
complete in dsa_tree_setup_routing_table, and not return zero as a
result but instead go ahead and set up the entire DSA switch tree
(practically on behalf of the other switches too).

The trouble is that the (N - 1) switches don't clean up after themselves
after they get an error such as EPROBE_DEFER. Their footprint left in
dst->ports by dsa_switch_touch_ports is still there. And switch N, the
one responsible with actually setting up the tree, is going to work with
those stale dp, dp->ds and dp->ds->dev pointers. In particular ds and
ds->dev might get freed by the device driver.

Be there a 2-switch tree and the following calling order:
- Switch 1 calls dsa_register_switch
  - Calls dsa_switch_touch_ports, populates dst->ports
  - Calls dsa_port_parse_cpu, gets -EPROBE_DEFER, exits.
- Switch 2 calls dsa_register_switch
  - Calls dsa_switch_touch_ports, populates dst->ports
  - Probe doesn't get deferred, so it goes ahead.
  - Calls dsa_tree_setup_routing_table, which returns "complete == true"
    due to Switch 1 having called dsa_switch_touch_ports before.
  - Because the DSA links are complete, it calls dsa_tree_setup_switches
    now.
  - dsa_tree_setup_switches iterates through dst->ports, initializing
    the Switch 1 ds structure (invalid) and the Switch 2 ds structure
    (valid).
  - Undefined behavior (use after free, sometimes NULL pointers, etc).

Real example below (debugging prints added by me, as well as guards
against NULL pointers):

[    5.477947] dsa_tree_setup_switches: Setting up port 0 of switch ffffff803df0b980 (dev ffffff803f775c00)
[    6.313002] dsa_tree_setup_switches: Setting up port 1 of switch ffffff803df0b980 (dev ffffff803f775c00)
[    6.319932] dsa_tree_setup_switches: Setting up port 2 of switch ffffff803df0b980 (dev ffffff803f775c00)
[    6.329693] dsa_tree_setup_switches: Setting up port 3 of switch ffffff803df0b980 (dev ffffff803f775c00)
[    6.339458] dsa_tree_setup_switches: Setting up port 4 of switch ffffff803df0b980 (dev ffffff803f775c00)
[    6.349226] dsa_tree_setup_switches: Setting up port 5 of switch ffffff803df0b980 (dev ffffff803f775c00)
[    6.358991] dsa_tree_setup_switches: Setting up port 6 of switch ffffff803df0b980 (dev ffffff803f775c00)
[    6.368758] dsa_tree_setup_switches: Setting up port 7 of switch ffffff803df0b980 (dev ffffff803f775c00)
[    6.378524] dsa_tree_setup_switches: Setting up port 8 of switch ffffff803df0b980 (dev ffffff803f775c00)
[    6.388291] dsa_tree_setup_switches: Setting up port 9 of switch ffffff803df0b980 (dev ffffff803f775c00)
[    6.398057] dsa_tree_setup_switches: Setting up port 10 of switch ffffff803df0b980 (dev ffffff803f775c00)
[    6.407912] dsa_tree_setup_switches: Setting up port 0 of switch ffffff803da02f80 (dev 0000000000000000)
[    6.417682] dsa_tree_setup_switches: Setting up port 1 of switch ffffff803da02f80 (dev 0000000000000000)
[    6.427446] dsa_tree_setup_switches: Setting up port 2 of switch ffffff803da02f80 (dev 0000000000000000)
[    6.437212] dsa_tree_setup_switches: Setting up port 3 of switch ffffff803da02f80 (dev 0000000000000000)
[    6.446979] dsa_tree_setup_switches: Setting up port 4 of switch ffffff803da02f80 (dev 0000000000000000)
[    6.456744] dsa_tree_setup_switches: Setting up port 5 of switch ffffff803da02f80 (dev 0000000000000000)
[    6.466512] dsa_tree_setup_switches: Setting up port 6 of switch ffffff803da02f80 (dev 0000000000000000)
[    6.476277] dsa_tree_setup_switches: Setting up port 7 of switch ffffff803da02f80 (dev 0000000000000000)
[    6.486043] dsa_tree_setup_switches: Setting up port 8 of switch ffffff803da02f80 (dev 0000000000000000)
[    6.495810] dsa_tree_setup_switches: Setting up port 9 of switch ffffff803da02f80 (dev 0000000000000000)
[    6.505577] dsa_tree_setup_switches: Setting up port 10 of switch ffffff803da02f80 (dev 0000000000000000)
[    6.515433] dsa_tree_setup_switches: Setting up port 0 of switch ffffff803db15b80 (dev ffffff803d8e4800)
[    7.354120] dsa_tree_setup_switches: Setting up port 1 of switch ffffff803db15b80 (dev ffffff803d8e4800)
[    7.361045] dsa_tree_setup_switches: Setting up port 2 of switch ffffff803db15b80 (dev ffffff803d8e4800)
[    7.370805] dsa_tree_setup_switches: Setting up port 3 of switch ffffff803db15b80 (dev ffffff803d8e4800)
[    7.380571] dsa_tree_setup_switches: Setting up port 4 of switch ffffff803db15b80 (dev ffffff803d8e4800)
[    7.390337] dsa_tree_setup_switches: Setting up port 5 of switch ffffff803db15b80 (dev ffffff803d8e4800)
[    7.400104] dsa_tree_setup_switches: Setting up port 6 of switch ffffff803db15b80 (dev ffffff803d8e4800)
[    7.409872] dsa_tree_setup_switches: Setting up port 7 of switch ffffff803db15b80 (dev ffffff803d8e4800)
[    7.419637] dsa_tree_setup_switches: Setting up port 8 of switch ffffff803db15b80 (dev ffffff803d8e4800)
[    7.429403] dsa_tree_setup_switches: Setting up port 9 of switch ffffff803db15b80 (dev ffffff803d8e4800)
[    7.439169] dsa_tree_setup_switches: Setting up port 10 of switch ffffff803db15b80 (dev ffffff803d8e4800)

The solution is to recognize that the functions that call
dsa_switch_touch_ports (dsa_switch_parse_of, dsa_switch_parse) have side
effects, and therefore one should clean up their side effects on error
path. The cleanup of dst->ports was taken from dsa_switch_remove and
moved into a dedicated dsa_switch_release_ports function, which should
really be per-switch (free only the members of dst->ports that are also
members of ds, instead of all switch ports).

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/dsa/dsa2.c | 36 +++++++++++++++++++++++++-----------
 1 file changed, 25 insertions(+), 11 deletions(-)

diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c
index c6d81f2baf4e..e7c30b472034 100644
--- a/net/dsa/dsa2.c
+++ b/net/dsa/dsa2.c
@@ -849,6 +849,19 @@ static int dsa_switch_parse(struct dsa_switch *ds, struct dsa_chip_data *cd)
 	return dsa_switch_parse_ports(ds, cd);
 }
 
+static void dsa_switch_release_ports(struct dsa_switch *ds)
+{
+	struct dsa_switch_tree *dst = ds->dst;
+	struct dsa_port *dp, *next;
+
+	list_for_each_entry_safe(dp, next, &dst->ports, list) {
+		if (dp->ds != ds)
+			continue;
+		list_del(&dp->list);
+		kfree(dp);
+	}
+}
+
 static int dsa_switch_probe(struct dsa_switch *ds)
 {
 	struct dsa_switch_tree *dst;
@@ -865,12 +878,17 @@ static int dsa_switch_probe(struct dsa_switch *ds)
 	if (!ds->num_ports)
 		return -EINVAL;
 
-	if (np)
+	if (np) {
 		err = dsa_switch_parse_of(ds, np);
-	else if (pdata)
+		if (err)
+			dsa_switch_release_ports(ds);
+	} else if (pdata) {
 		err = dsa_switch_parse(ds, pdata);
-	else
+		if (err)
+			dsa_switch_release_ports(ds);
+	} else {
 		err = -ENODEV;
+	}
 
 	if (err)
 		return err;
@@ -878,8 +896,10 @@ static int dsa_switch_probe(struct dsa_switch *ds)
 	dst = ds->dst;
 	dsa_tree_get(dst);
 	err = dsa_tree_setup(dst);
-	if (err)
+	if (err) {
+		dsa_switch_release_ports(ds);
 		dsa_tree_put(dst);
+	}
 
 	return err;
 }
@@ -900,15 +920,9 @@ EXPORT_SYMBOL_GPL(dsa_register_switch);
 static void dsa_switch_remove(struct dsa_switch *ds)
 {
 	struct dsa_switch_tree *dst = ds->dst;
-	struct dsa_port *dp, *next;
 
 	dsa_tree_teardown(dst);
-
-	list_for_each_entry_safe(dp, next, &dst->ports, list) {
-		list_del(&dp->list);
-		kfree(dp);
-	}
-
+	dsa_switch_release_ports(ds);
 	dsa_tree_put(dst);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 6dd4b4f3936e17fedea1308bc70e9716f68bf232 Mon Sep 17 00:00:00 2001
From: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Date: Sat, 25 Jan 2020 22:18:47 +0100
Subject: mlxsw: minimal: Fix an error handling path in 'mlxsw_m_port_create()'

An 'alloc_etherdev()' called is not ballanced by a corresponding
'free_netdev()' call in one error handling path.

Slighly reorder the error handling code to catch the missed case.

Fixes: c100e47caa8e ("mlxsw: minimal: Add ethtool support")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/minimal.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/minimal.c b/drivers/net/ethernet/mellanox/mlxsw/minimal.c
index 2b543911ae00..c4caeeadcba9 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/minimal.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/minimal.c
@@ -213,8 +213,8 @@ mlxsw_m_port_create(struct mlxsw_m *mlxsw_m, u8 local_port, u8 module)
 
 err_register_netdev:
 	mlxsw_m->ports[local_port] = NULL;
-	free_netdev(dev);
 err_dev_addr_get:
+	free_netdev(dev);
 err_alloc_etherdev:
 	mlxsw_core_port_fini(mlxsw_m->core, local_port);
 	return err;
-- 
cgit v1.2.3-59-g8ed1b


From a8ec173a3f29fd777681c72835d4d8642ebea899 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Sun, 26 Jan 2020 10:40:44 +0100
Subject: r8169: don't set min_mtu/max_mtu if not needed

Defaults for min_mtu and max_mtu are set by ether_setup(), which is
called from devm_alloc_etherdev(). Let rtl_jumbo_max() only return
a positive value if actually jumbo packets are supported. This also
allows to remove constant Jumbo_1K which is a little misleading anyway.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 12 +++++-------
 1 file changed, 5 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 7a5fe11378aa..aaa316be6183 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -85,7 +85,6 @@
 #define RTL_R16(tp, reg)		readw(tp->mmio_addr + (reg))
 #define RTL_R32(tp, reg)		readl(tp->mmio_addr + (reg))
 
-#define JUMBO_1K	ETH_DATA_LEN
 #define JUMBO_4K	(4*1024 - ETH_HLEN - 2)
 #define JUMBO_6K	(6*1024 - ETH_HLEN - 2)
 #define JUMBO_7K	(7*1024 - ETH_HLEN - 2)
@@ -1494,7 +1493,7 @@ static netdev_features_t rtl8169_fix_features(struct net_device *dev,
 	if (dev->mtu > TD_MSS_MAX)
 		features &= ~NETIF_F_ALL_TSO;
 
-	if (dev->mtu > JUMBO_1K &&
+	if (dev->mtu > ETH_DATA_LEN &&
 	    tp->mac_version > RTL_GIGA_MAC_VER_06)
 		features &= ~(NETIF_F_CSUM_MASK | NETIF_F_ALL_TSO);
 
@@ -5360,7 +5359,7 @@ static int rtl_jumbo_max(struct rtl8169_private *tp)
 {
 	/* Non-GBit versions don't support jumbo frames */
 	if (!tp->supports_gmii)
-		return JUMBO_1K;
+		return 0;
 
 	switch (tp->mac_version) {
 	/* RTL8169 */
@@ -5591,10 +5590,9 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	dev->hw_features |= NETIF_F_RXALL;
 	dev->hw_features |= NETIF_F_RXFCS;
 
-	/* MTU range: 60 - hw-specific max */
-	dev->min_mtu = ETH_ZLEN;
 	jumbo_max = rtl_jumbo_max(tp);
-	dev->max_mtu = jumbo_max;
+	if (jumbo_max)
+		dev->max_mtu = jumbo_max;
 
 	rtl_set_irq_mask(tp);
 
@@ -5624,7 +5622,7 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		   (RTL_R32(tp, TxConfig) >> 20) & 0xfcf,
 		   pci_irq_vector(pdev, 0));
 
-	if (jumbo_max > JUMBO_1K)
+	if (jumbo_max)
 		netif_info(tp, probe, dev,
 			   "jumbo features [frames: %d bytes, tx checksumming: %s]\n",
 			   jumbo_max, tp->mac_version <= RTL_GIGA_MAC_VER_06 ?
-- 
cgit v1.2.3-59-g8ed1b


From 42a84a8cd0ff0cbff5a4595e1304c4567a30267d Mon Sep 17 00:00:00 2001
From: John Fastabend <john.fastabend@gmail.com>
Date: Sun, 26 Jan 2020 16:14:00 -0800
Subject: bpf, xdp: Update devmap comments to reflect napi/rcu usage
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Now that we rely on synchronize_rcu and call_rcu waiting to
exit perempt-disable regions (NAPI) lets update the comments
to reflect this.

Fixes: 0536b85239b84 ("xdp: Simplify devmap cleanup")
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Björn Töpel <bjorn.topel@intel.com>
Acked-by: Song Liu <songliubraving@fb.com>
Link: https://lore.kernel.org/bpf/1580084042-11598-2-git-send-email-john.fastabend@gmail.com
---
 kernel/bpf/devmap.c | 21 +++++++++++----------
 1 file changed, 11 insertions(+), 10 deletions(-)

diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index f3a44f6ca86e..373c6a30d8a5 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -190,10 +190,12 @@ static void dev_map_free(struct bpf_map *map)
 
 	/* At this point bpf_prog->aux->refcnt == 0 and this map->refcnt == 0,
 	 * so the programs (can be more than one that used this map) were
-	 * disconnected from events. Wait for outstanding critical sections in
-	 * these programs to complete. The rcu critical section only guarantees
-	 * no further reads against netdev_map. It does __not__ ensure pending
-	 * flush operations (if any) are complete.
+	 * disconnected from events. The following synchronize_rcu() guarantees
+	 * both rcu read critical sections complete and waits for
+	 * preempt-disable regions (NAPI being the relevant context here) so we
+	 * are certain there will be no further reads against the netdev_map and
+	 * all flush operations are complete. Flush operations can only be done
+	 * from NAPI context for this reason.
 	 */
 
 	spin_lock(&dev_map_lock);
@@ -503,12 +505,11 @@ static int dev_map_delete_elem(struct bpf_map *map, void *key)
 		return -EINVAL;
 
 	/* Use call_rcu() here to ensure any rcu critical sections have
-	 * completed, but this does not guarantee a flush has happened
-	 * yet. Because driver side rcu_read_lock/unlock only protects the
-	 * running XDP program. However, for pending flush operations the
-	 * dev and ctx are stored in another per cpu map. And additionally,
-	 * the driver tear down ensures all soft irqs are complete before
-	 * removing the net device in the case of dev_put equals zero.
+	 * completed as well as any flush operations because call_rcu
+	 * will wait for preempt-disable region to complete, NAPI in this
+	 * context.  And additionally, the driver tear down ensures all
+	 * soft irqs are complete before removing the net device in the
+	 * case of dev_put equals zero.
 	 */
 	old_dev = xchg(&dtab->netdev_map[k], NULL);
 	if (old_dev)
-- 
cgit v1.2.3-59-g8ed1b


From 9719c6b98db42462bfb9aa172f017bdfd6d66c6f Mon Sep 17 00:00:00 2001
From: John Fastabend <john.fastabend@gmail.com>
Date: Sun, 26 Jan 2020 16:14:01 -0800
Subject: bpf, xdp: virtio_net use access ptr macro for xdp enable check

virtio_net currently relies on rcu critical section to access the xdp
program in its xdp_xmit handler. However, the pointer to the xdp program
is only used to do a NULL pointer comparison to determine if xdp is
enabled or not.

Use rcu_access_pointer() instead of rcu_dereference() to reflect this.
Then later when we drop rcu_read critical section virtio_net will not
need in special handling.

Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Link: https://lore.kernel.org/bpf/1580084042-11598-3-git-send-email-john.fastabend@gmail.com
---
 drivers/net/virtio_net.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index c458cd313281..2fe7a3188282 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -501,7 +501,7 @@ static int virtnet_xdp_xmit(struct net_device *dev,
 	/* Only allow ndo_xdp_xmit if XDP is loaded on dev, as this
 	 * indicate XDP resources have been successfully allocated.
 	 */
-	xdp_prog = rcu_dereference(rq->xdp_prog);
+	xdp_prog = rcu_access_pointer(rq->xdp_prog);
 	if (!xdp_prog)
 		return -ENXIO;
 
-- 
cgit v1.2.3-59-g8ed1b


From b23bfa5633b19bf1db87b36a76b2225c734f794c Mon Sep 17 00:00:00 2001
From: John Fastabend <john.fastabend@gmail.com>
Date: Sun, 26 Jan 2020 16:14:02 -0800
Subject: bpf, xdp: Remove no longer required rcu_read_{un}lock()

Now that we depend on rcu_call() and synchronize_rcu() to also wait
for preempt_disabled region to complete the rcu read critical section
in __dev_map_flush() is no longer required. Except in a few special
cases in drivers that need it for other reasons.

These originally ensured the map reference was safe while a map was
also being free'd. And additionally that bpf program updates via
ndo_bpf did not happen while flush updates were in flight. But flush
by new rules can only be called from preempt-disabled NAPI context.
The synchronize_rcu from the map free path and the rcu_call from the
delete path will ensure the reference there is safe. So lets remove
the rcu_read_lock and rcu_read_unlock pair to avoid any confusion
around how this is being protected.

If the rcu_read_lock was required it would mean errors in the above
logic and the original patch would also be wrong.

Now that we have done above we put the rcu_read_lock in the driver
code where it is needed in a driver dependent way. I think this
helps readability of the code so we know where and why we are
taking read locks. Most drivers will not need rcu_read_locks here
and further XDP drivers already have rcu_read_locks in their code
paths for reading xdp programs on RX side so this makes it symmetric
where we don't have half of rcu critical sections define in driver
and the other half in devmap.

Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Link: https://lore.kernel.org/bpf/1580084042-11598-4-git-send-email-john.fastabend@gmail.com
---
 drivers/net/veth.c  | 6 +++++-
 kernel/bpf/devmap.c | 5 +++--
 2 files changed, 8 insertions(+), 3 deletions(-)

diff --git a/drivers/net/veth.c b/drivers/net/veth.c
index 1c89017beebb..8cdc4415fa70 100644
--- a/drivers/net/veth.c
+++ b/drivers/net/veth.c
@@ -377,6 +377,7 @@ static int veth_xdp_xmit(struct net_device *dev, int n,
 	unsigned int max_len;
 	struct veth_rq *rq;
 
+	rcu_read_lock();
 	if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK)) {
 		ret = -EINVAL;
 		goto drop;
@@ -418,11 +419,14 @@ static int veth_xdp_xmit(struct net_device *dev, int n,
 	if (flags & XDP_XMIT_FLUSH)
 		__veth_xdp_flush(rq);
 
-	if (likely(!drops))
+	if (likely(!drops)) {
+		rcu_read_unlock();
 		return n;
+	}
 
 	ret = n - drops;
 drop:
+	rcu_read_unlock();
 	atomic64_add(drops, &priv->dropped);
 
 	return ret;
diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index 373c6a30d8a5..58bdca5d978a 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -366,16 +366,17 @@ error:
  * from NET_RX_SOFTIRQ. Either way the poll routine must complete before the
  * net device can be torn down. On devmap tear down we ensure the flush list
  * is empty before completing to ensure all flush operations have completed.
+ * When drivers update the bpf program they may need to ensure any flush ops
+ * are also complete. Using synchronize_rcu or call_rcu will suffice for this
+ * because both wait for napi context to exit.
  */
 void __dev_flush(void)
 {
 	struct list_head *flush_list = this_cpu_ptr(&dev_flush_list);
 	struct xdp_dev_bulk_queue *bq, *tmp;
 
-	rcu_read_lock();
 	list_for_each_entry_safe(bq, tmp, flush_list, flush_node)
 		bq_xmit_all(bq, XDP_XMIT_FLUSH);
-	rcu_read_unlock();
 }
 
 /* rcu_read_lock (from syscall and BPF contexts) ensures that if a delete and/or
-- 
cgit v1.2.3-59-g8ed1b


From 5a44c71ccda60a50073c5d7fe3f694cdfa3ab0c2 Mon Sep 17 00:00:00 2001
From: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Date: Sun, 26 Jan 2020 11:44:29 +0100
Subject: drivers: net: xgene: Fix the order of the arguments of
 'alloc_etherdev_mqs()'

'alloc_etherdev_mqs()' expects first 'tx', then 'rx'. The semantic here
looks reversed.

Reorder the arguments passed to 'alloc_etherdev_mqs()' in order to keep
the correct semantic.

In fact, this is a no-op because both XGENE_NUM_[RT]X_RING are 8.

Fixes: 107dec2749fe ("drivers: net: xgene: Add support for multiple queues")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/apm/xgene/xgene_enet_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c
index e284b6753725..6aee2f0fc0db 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c
@@ -2020,7 +2020,7 @@ static int xgene_enet_probe(struct platform_device *pdev)
 	int ret;
 
 	ndev = alloc_etherdev_mqs(sizeof(struct xgene_enet_pdata),
-				  XGENE_NUM_RX_RING, XGENE_NUM_TX_RING);
+				  XGENE_NUM_TX_RING, XGENE_NUM_RX_RING);
 	if (!ndev)
 		return -ENOMEM;
 
-- 
cgit v1.2.3-59-g8ed1b


From 59fb9b62fb6c929a756563152a89f39b07cf8893 Mon Sep 17 00:00:00 2001
From: Yoshiki Komachi <komachi.yoshiki@gmail.com>
Date: Fri, 17 Jan 2020 16:05:32 +0900
Subject: flow_dissector: Fix to use new variables for port ranges in bpf hook

This patch applies new flag (FLOW_DISSECTOR_KEY_PORTS_RANGE) and
field (tp_range) to BPF flow dissector to generate appropriate flow
keys when classified by specified port ranges.

Fixes: 8ffb055beae5 ("cls_flower: Fix the behavior using port ranges with hw-offload")
Signed-off-by: Yoshiki Komachi <komachi.yoshiki@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Petar Penkov <ppenkov@google.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20200117070533.402240-2-komachi.yoshiki@gmail.com
---
 net/core/flow_dissector.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c
index f560b4902060..a1670dff0629 100644
--- a/net/core/flow_dissector.c
+++ b/net/core/flow_dissector.c
@@ -834,10 +834,10 @@ static void __skb_flow_bpf_to_target(const struct bpf_flow_keys *flow_keys,
 				     struct flow_dissector *flow_dissector,
 				     void *target_container)
 {
+	struct flow_dissector_key_ports *key_ports = NULL;
 	struct flow_dissector_key_control *key_control;
 	struct flow_dissector_key_basic *key_basic;
 	struct flow_dissector_key_addrs *key_addrs;
-	struct flow_dissector_key_ports *key_ports;
 	struct flow_dissector_key_tags *key_tags;
 
 	key_control = skb_flow_dissector_target(flow_dissector,
@@ -876,10 +876,17 @@ static void __skb_flow_bpf_to_target(const struct bpf_flow_keys *flow_keys,
 		key_control->addr_type = FLOW_DISSECTOR_KEY_IPV6_ADDRS;
 	}
 
-	if (dissector_uses_key(flow_dissector, FLOW_DISSECTOR_KEY_PORTS)) {
+	if (dissector_uses_key(flow_dissector, FLOW_DISSECTOR_KEY_PORTS))
 		key_ports = skb_flow_dissector_target(flow_dissector,
 						      FLOW_DISSECTOR_KEY_PORTS,
 						      target_container);
+	else if (dissector_uses_key(flow_dissector,
+				    FLOW_DISSECTOR_KEY_PORTS_RANGE))
+		key_ports = skb_flow_dissector_target(flow_dissector,
+						      FLOW_DISSECTOR_KEY_PORTS_RANGE,
+						      target_container);
+
+	if (key_ports) {
 		key_ports->src = flow_keys->sport;
 		key_ports->dst = flow_keys->dport;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 265bb359061db6ef825dec3912f341a604966371 Mon Sep 17 00:00:00 2001
From: Yoshiki Komachi <komachi.yoshiki@gmail.com>
Date: Fri, 17 Jan 2020 16:05:33 +0900
Subject: selftests/bpf: Add test based on port range for BPF flow dissector

Add a simple test to make sure that a filter based on specified port
range classifies packets correctly.

Signed-off-by: Yoshiki Komachi <komachi.yoshiki@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Petar Penkov <ppenkov@google.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20200117070533.402240-3-komachi.yoshiki@gmail.com
---
 tools/testing/selftests/bpf/test_flow_dissector.sh | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/tools/testing/selftests/bpf/test_flow_dissector.sh b/tools/testing/selftests/bpf/test_flow_dissector.sh
index a8485ae103d1..174b72a64a4c 100755
--- a/tools/testing/selftests/bpf/test_flow_dissector.sh
+++ b/tools/testing/selftests/bpf/test_flow_dissector.sh
@@ -139,6 +139,20 @@ echo "Testing IPv4 + GRE..."
 
 tc filter del dev lo ingress pref 1337
 
+echo "Testing port range..."
+# Drops all IP/UDP packets coming from port 8-10
+tc filter add dev lo parent ffff: protocol ip pref 1337 flower ip_proto \
+	udp src_port 8-10 action drop
+
+# Send 10 IPv4/UDP packets from port 7. Filter should not drop any.
+./test_flow_dissector -i 4 -f 7
+# Send 10 IPv4/UDP packets from port 9. Filter should drop all.
+./test_flow_dissector -i 4 -f 9 -F
+# Send 10 IPv4/UDP packets from port 11. Filter should not drop any.
+./test_flow_dissector -i 4 -f 11
+
+tc filter del dev lo ingress pref 1337
+
 echo "Testing IPv6..."
 # Drops all IPv6/UDP packets coming from port 9
 tc filter add dev lo parent ffff: protocol ipv6 pref 1337 flower ip_proto \
-- 
cgit v1.2.3-59-g8ed1b


From d2c4b444fd13f4699ab8c4c49062cd2c7516c241 Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Sun, 26 Jan 2020 23:11:01 +0100
Subject: ethtool: fix kernel-doc descriptions

Fix missing or incorrect function argument and struct member descriptions.

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ethtool/netlink.c | 26 ++++++++++++++------------
 net/ethtool/strset.c  |  1 +
 2 files changed, 15 insertions(+), 12 deletions(-)

diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index 86b79f9bc08d..0af43bbdb9b2 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -134,11 +134,12 @@ nla_put_failure:
 
 /**
  * ethnl_reply_init() - Create skb for a reply and fill device identification
- * @payload: payload length (without netlink and genetlink header)
- * @dev:     device the reply is about (may be null)
- * @cmd:     ETHTOOL_MSG_* message type for reply
- * @info:    genetlink info of the received packet we respond to
- * @ehdrp:   place to store payload pointer returned by genlmsg_new()
+ * @payload:      payload length (without netlink and genetlink header)
+ * @dev:          device the reply is about (may be null)
+ * @cmd:          ETHTOOL_MSG_* message type for reply
+ * @hdr_attrtype: attribute type for common header
+ * @info:         genetlink info of the received packet we respond to
+ * @ehdrp:        place to store payload pointer returned by genlmsg_new()
  *
  * Return: pointer to allocated skb on success, NULL on error
  */
@@ -188,10 +189,11 @@ static int ethnl_multicast(struct sk_buff *skb, struct net_device *dev)
 
 /**
  * struct ethnl_dump_ctx - context structure for generic dumpit() callback
- * @ops:      request ops of currently processed message type
- * @req_info: parsed request header of processed request
- * @pos_hash: saved iteration position - hashbucket
- * @pos_idx:  saved iteration position - index
+ * @ops:        request ops of currently processed message type
+ * @req_info:   parsed request header of processed request
+ * @reply_data: data needed to compose the reply
+ * @pos_hash:   saved iteration position - hashbucket
+ * @pos_idx:    saved iteration position - index
  *
  * These parameters are kept in struct netlink_callback as context preserved
  * between iterations. They are initialized by ethnl_default_start() and used
@@ -268,9 +270,9 @@ out:
 
 /**
  * ethnl_init_reply_data() - Initialize reply data for GET request
- * @req_info: pointer to embedded struct ethnl_req_info
- * @ops:      instance of struct ethnl_request_ops describing the layout
- * @dev:      network device to initialize the reply for
+ * @reply_data: pointer to embedded struct ethnl_reply_data
+ * @ops:        instance of struct ethnl_request_ops describing the layout
+ * @dev:        network device to initialize the reply for
  *
  * Fills the reply data part with zeros and sets the dev member. Must be called
  * before calling the ->fill_reply() callback (for each iteration when handling
diff --git a/net/ethtool/strset.c b/net/ethtool/strset.c
index 82a059c13c1c..948d967f1eca 100644
--- a/net/ethtool/strset.c
+++ b/net/ethtool/strset.c
@@ -85,6 +85,7 @@ get_stringset_policy[ETHTOOL_A_STRINGSET_MAX + 1] = {
 
 /**
  * strset_include() - test if a string set should be included in reply
+ * @info: parsed client request
  * @data: pointer to request data structure
  * @id:   id of string set to check (ETH_SS_* constants)
  */
-- 
cgit v1.2.3-59-g8ed1b


From 6a94b8ccf6b77f005ab1b36a878e1d81df0c033e Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Sun, 26 Jan 2020 23:11:04 +0100
Subject: ethtool: provide message mask with DEBUG_GET request

Implement DEBUG_GET request to get debugging settings for a device. At the
moment, only message mask corresponding to message level as reported by
ETHTOOL_GMSGLVL ioctl request is provided. (It is called message level in
ioctl interface but almost all drivers interpret it as a bit mask.)

As part of the implementation, provide symbolic names for message mask bits
as ETH_SS_MSG_CLASSES string set.

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/ethtool-netlink.rst | 34 +++++++++++-
 include/linux/netdevice.h                    | 56 +++++++++++++------
 include/uapi/linux/ethtool.h                 |  2 +
 include/uapi/linux/ethtool_netlink.h         | 14 +++++
 net/ethtool/Makefile                         |  2 +-
 net/ethtool/common.c                         | 19 +++++++
 net/ethtool/common.h                         |  1 +
 net/ethtool/debug.c                          | 80 ++++++++++++++++++++++++++++
 net/ethtool/netlink.c                        |  8 +++
 net/ethtool/netlink.h                        |  1 +
 net/ethtool/strset.c                         |  5 ++
 11 files changed, 205 insertions(+), 17 deletions(-)
 create mode 100644 net/ethtool/debug.c

diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
index c60afba69e3c..76a411d3102d 100644
--- a/Documentation/networking/ethtool-netlink.rst
+++ b/Documentation/networking/ethtool-netlink.rst
@@ -185,6 +185,7 @@ Userspace to kernel:
   ``ETHTOOL_MSG_LINKMODES_GET``         get link modes info
   ``ETHTOOL_MSG_LINKMODES_SET``         set link modes info
   ``ETHTOOL_MSG_LINKSTATE_GET``         get link state
+  ``ETHTOOL_MSG_DEBUG_GET``             get debugging settings
   ===================================== ================================
 
 Kernel to userspace:
@@ -196,6 +197,7 @@ Kernel to userspace:
   ``ETHTOOL_MSG_LINKMODES_GET_REPLY``   link modes info
   ``ETHTOOL_MSG_LINKMODES_NTF``         link modes notification
   ``ETHTOOL_MSG_LINKSTATE_GET_REPLY``   link state info
+  ``ETHTOOL_MSG_DEBUG_GET_REPLY``       debugging settings
   ===================================== ================================
 
 ``GET`` requests are sent by userspace applications to retrieve device
@@ -423,6 +425,36 @@ define their own handler.
 devices supporting the request).
 
 
+DEBUG_GET
+=========
+
+Requests debugging settings of a device. At the moment, only message mask is
+provided.
+
+Request contents:
+
+  ====================================  ======  ==========================
+  ``ETHTOOL_A_DEBUG_HEADER``            nested  request header
+  ====================================  ======  ==========================
+
+Kernel response contents:
+
+  ====================================  ======  ==========================
+  ``ETHTOOL_A_DEBUG_HEADER``            nested  reply header
+  ``ETHTOOL_A_DEBUG_MSGMASK``           bitset  message mask
+  ====================================  ======  ==========================
+
+The message mask (``ETHTOOL_A_DEBUG_MSGMASK``) is equal to message level as
+provided by ``ETHTOOL_GMSGLVL`` and set by ``ETHTOOL_SMSGLVL`` in ioctl
+interface. While it is called message level there for historical reasons, most
+drivers and almost all newer drivers use it as a mask of enabled message
+classes (represented by ``NETIF_MSG_*`` constants); therefore netlink
+interface follows its actual use in practice.
+
+``DEBUG_GET`` allows dump requests (kernel returns reply messages for all
+devices supporting the request).
+
+
 Request translation
 ===================
 
@@ -441,7 +473,7 @@ have their netlink replacement yet.
   ``ETHTOOL_GREGS``                   n/a
   ``ETHTOOL_GWOL``                    n/a
   ``ETHTOOL_SWOL``                    n/a
-  ``ETHTOOL_GMSGLVL``                 n/a
+  ``ETHTOOL_GMSGLVL``                 ``ETHTOOL_MSG_DEBUG_GET``
   ``ETHTOOL_SMSGLVL``                 n/a
   ``ETHTOOL_NWAY_RST``                n/a
   ``ETHTOOL_GLINK``                   ``ETHTOOL_MSG_LINKSTATE_GET``
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 4626188a754b..a9c6b5c61d27 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -3913,22 +3913,48 @@ void netif_device_attach(struct net_device *dev);
  */
 
 enum {
-	NETIF_MSG_DRV		= 0x0001,
-	NETIF_MSG_PROBE		= 0x0002,
-	NETIF_MSG_LINK		= 0x0004,
-	NETIF_MSG_TIMER		= 0x0008,
-	NETIF_MSG_IFDOWN	= 0x0010,
-	NETIF_MSG_IFUP		= 0x0020,
-	NETIF_MSG_RX_ERR	= 0x0040,
-	NETIF_MSG_TX_ERR	= 0x0080,
-	NETIF_MSG_TX_QUEUED	= 0x0100,
-	NETIF_MSG_INTR		= 0x0200,
-	NETIF_MSG_TX_DONE	= 0x0400,
-	NETIF_MSG_RX_STATUS	= 0x0800,
-	NETIF_MSG_PKTDATA	= 0x1000,
-	NETIF_MSG_HW		= 0x2000,
-	NETIF_MSG_WOL		= 0x4000,
+	NETIF_MSG_DRV_BIT,
+	NETIF_MSG_PROBE_BIT,
+	NETIF_MSG_LINK_BIT,
+	NETIF_MSG_TIMER_BIT,
+	NETIF_MSG_IFDOWN_BIT,
+	NETIF_MSG_IFUP_BIT,
+	NETIF_MSG_RX_ERR_BIT,
+	NETIF_MSG_TX_ERR_BIT,
+	NETIF_MSG_TX_QUEUED_BIT,
+	NETIF_MSG_INTR_BIT,
+	NETIF_MSG_TX_DONE_BIT,
+	NETIF_MSG_RX_STATUS_BIT,
+	NETIF_MSG_PKTDATA_BIT,
+	NETIF_MSG_HW_BIT,
+	NETIF_MSG_WOL_BIT,
+
+	/* When you add a new bit above, update netif_msg_class_names array
+	 * in net/ethtool/common.c
+	 */
+	NETIF_MSG_CLASS_COUNT,
 };
+/* Both ethtool_ops interface and internal driver implementation use u32 */
+static_assert(NETIF_MSG_CLASS_COUNT <= 32);
+
+#define __NETIF_MSG_BIT(bit)	((u32)1 << (bit))
+#define __NETIF_MSG(name)	__NETIF_MSG_BIT(NETIF_MSG_ ## name ## _BIT)
+
+#define NETIF_MSG_DRV		__NETIF_MSG(DRV)
+#define NETIF_MSG_PROBE		__NETIF_MSG(PROBE)
+#define NETIF_MSG_LINK		__NETIF_MSG(LINK)
+#define NETIF_MSG_TIMER		__NETIF_MSG(TIMER)
+#define NETIF_MSG_IFDOWN	__NETIF_MSG(IFDOWN)
+#define NETIF_MSG_IFUP		__NETIF_MSG(IFUP)
+#define NETIF_MSG_RX_ERR	__NETIF_MSG(RX_ERR)
+#define NETIF_MSG_TX_ERR	__NETIF_MSG(TX_ERR)
+#define NETIF_MSG_TX_QUEUED	__NETIF_MSG(TX_QUEUED)
+#define NETIF_MSG_INTR		__NETIF_MSG(INTR)
+#define NETIF_MSG_TX_DONE	__NETIF_MSG(TX_DONE)
+#define NETIF_MSG_RX_STATUS	__NETIF_MSG(RX_STATUS)
+#define NETIF_MSG_PKTDATA	__NETIF_MSG(PKTDATA)
+#define NETIF_MSG_HW		__NETIF_MSG(HW)
+#define NETIF_MSG_WOL		__NETIF_MSG(WOL)
 
 #define netif_msg_drv(p)	((p)->msg_enable & NETIF_MSG_DRV)
 #define netif_msg_probe(p)	((p)->msg_enable & NETIF_MSG_PROBE)
diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h
index 116bcbf09c74..456fb2aa0fad 100644
--- a/include/uapi/linux/ethtool.h
+++ b/include/uapi/linux/ethtool.h
@@ -594,6 +594,7 @@ struct ethtool_pauseparam {
  * @ETH_SS_PHY_STATS: Statistic names, for use with %ETHTOOL_GPHYSTATS
  * @ETH_SS_PHY_TUNABLES: PHY tunable names
  * @ETH_SS_LINK_MODES: link mode names
+ * @ETH_SS_MSG_CLASSES: debug message class names
  */
 enum ethtool_stringset {
 	ETH_SS_TEST		= 0,
@@ -606,6 +607,7 @@ enum ethtool_stringset {
 	ETH_SS_PHY_STATS,
 	ETH_SS_PHY_TUNABLES,
 	ETH_SS_LINK_MODES,
+	ETH_SS_MSG_CLASSES,
 
 	/* add new constants above here */
 	ETH_SS_COUNT
diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
index 02f82f42a889..0d39e04567cb 100644
--- a/include/uapi/linux/ethtool_netlink.h
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -20,6 +20,7 @@ enum {
 	ETHTOOL_MSG_LINKMODES_GET,
 	ETHTOOL_MSG_LINKMODES_SET,
 	ETHTOOL_MSG_LINKSTATE_GET,
+	ETHTOOL_MSG_DEBUG_GET,
 
 	/* add new constants above here */
 	__ETHTOOL_MSG_USER_CNT,
@@ -35,6 +36,7 @@ enum {
 	ETHTOOL_MSG_LINKMODES_GET_REPLY,
 	ETHTOOL_MSG_LINKMODES_NTF,
 	ETHTOOL_MSG_LINKSTATE_GET_REPLY,
+	ETHTOOL_MSG_DEBUG_GET_REPLY,
 
 	/* add new constants above here */
 	__ETHTOOL_MSG_KERNEL_CNT,
@@ -195,6 +197,18 @@ enum {
 	ETHTOOL_A_LINKSTATE_MAX = __ETHTOOL_A_LINKSTATE_CNT - 1
 };
 
+/* DEBUG */
+
+enum {
+	ETHTOOL_A_DEBUG_UNSPEC,
+	ETHTOOL_A_DEBUG_HEADER,			/* nest - _A_HEADER_* */
+	ETHTOOL_A_DEBUG_MSGMASK,		/* bitset */
+
+	/* add new constants above here */
+	__ETHTOOL_A_DEBUG_CNT,
+	ETHTOOL_A_DEBUG_MAX = __ETHTOOL_A_DEBUG_CNT - 1
+};
+
 /* generic netlink info */
 #define ETHTOOL_GENL_NAME "ethtool"
 #define ETHTOOL_GENL_VERSION 1
diff --git a/net/ethtool/Makefile b/net/ethtool/Makefile
index 9a1332fb0cc6..c120c820a4f5 100644
--- a/net/ethtool/Makefile
+++ b/net/ethtool/Makefile
@@ -5,4 +5,4 @@ obj-y				+= ioctl.o common.o
 obj-$(CONFIG_ETHTOOL_NETLINK)	+= ethtool_nl.o
 
 ethtool_nl-y	:= netlink.o bitset.o strset.o linkinfo.o linkmodes.o \
-		   linkstate.o
+		   linkstate.o debug.o
diff --git a/net/ethtool/common.c b/net/ethtool/common.c
index c7b8956c3827..aa1183a65a76 100644
--- a/net/ethtool/common.c
+++ b/net/ethtool/common.c
@@ -171,6 +171,25 @@ const char link_mode_names[][ETH_GSTRING_LEN] = {
 };
 static_assert(ARRAY_SIZE(link_mode_names) == __ETHTOOL_LINK_MODE_MASK_NBITS);
 
+const char netif_msg_class_names[][ETH_GSTRING_LEN] = {
+	[NETIF_MSG_DRV_BIT]		= "drv",
+	[NETIF_MSG_PROBE_BIT]		= "probe",
+	[NETIF_MSG_LINK_BIT]		= "link",
+	[NETIF_MSG_TIMER_BIT]		= "timer",
+	[NETIF_MSG_IFDOWN_BIT]		= "ifdown",
+	[NETIF_MSG_IFUP_BIT]		= "ifup",
+	[NETIF_MSG_RX_ERR_BIT]		= "rx_err",
+	[NETIF_MSG_TX_ERR_BIT]		= "tx_err",
+	[NETIF_MSG_TX_QUEUED_BIT]	= "tx_queued",
+	[NETIF_MSG_INTR_BIT]		= "intr",
+	[NETIF_MSG_TX_DONE_BIT]		= "tx_done",
+	[NETIF_MSG_RX_STATUS_BIT]	= "rx_status",
+	[NETIF_MSG_PKTDATA_BIT]		= "pktdata",
+	[NETIF_MSG_HW_BIT]		= "hw",
+	[NETIF_MSG_WOL_BIT]		= "wol",
+};
+static_assert(ARRAY_SIZE(netif_msg_class_names) == NETIF_MSG_CLASS_COUNT);
+
 /* return false if legacy contained non-0 deprecated fields
  * maxtxpkt/maxrxpkt. rest of ksettings always updated
  */
diff --git a/net/ethtool/common.h b/net/ethtool/common.h
index 5c5f7dc90cd4..064c5c3aa990 100644
--- a/net/ethtool/common.h
+++ b/net/ethtool/common.h
@@ -19,6 +19,7 @@ tunable_strings[__ETHTOOL_TUNABLE_COUNT][ETH_GSTRING_LEN];
 extern const char
 phy_tunable_strings[__ETHTOOL_PHY_TUNABLE_COUNT][ETH_GSTRING_LEN];
 extern const char link_mode_names[][ETH_GSTRING_LEN];
+extern const char netif_msg_class_names[][ETH_GSTRING_LEN];
 
 int __ethtool_get_link(struct net_device *dev);
 
diff --git a/net/ethtool/debug.c b/net/ethtool/debug.c
new file mode 100644
index 000000000000..cc121a23be5f
--- /dev/null
+++ b/net/ethtool/debug.c
@@ -0,0 +1,80 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include "netlink.h"
+#include "common.h"
+#include "bitset.h"
+
+struct debug_req_info {
+	struct ethnl_req_info		base;
+};
+
+struct debug_reply_data {
+	struct ethnl_reply_data		base;
+	u32				msg_mask;
+};
+
+#define DEBUG_REPDATA(__reply_base) \
+	container_of(__reply_base, struct debug_reply_data, base)
+
+static const struct nla_policy
+debug_get_policy[ETHTOOL_A_DEBUG_MAX + 1] = {
+	[ETHTOOL_A_DEBUG_UNSPEC]	= { .type = NLA_REJECT },
+	[ETHTOOL_A_DEBUG_HEADER]	= { .type = NLA_NESTED },
+	[ETHTOOL_A_DEBUG_MSGMASK]	= { .type = NLA_REJECT },
+};
+
+static int debug_prepare_data(const struct ethnl_req_info *req_base,
+			      struct ethnl_reply_data *reply_base,
+			      struct genl_info *info)
+{
+	struct debug_reply_data *data = DEBUG_REPDATA(reply_base);
+	struct net_device *dev = reply_base->dev;
+	int ret;
+
+	if (!dev->ethtool_ops->get_msglevel)
+		return -EOPNOTSUPP;
+
+	ret = ethnl_ops_begin(dev);
+	if (ret < 0)
+		return ret;
+	data->msg_mask = dev->ethtool_ops->get_msglevel(dev);
+	ethnl_ops_complete(dev);
+
+	return 0;
+}
+
+static int debug_reply_size(const struct ethnl_req_info *req_base,
+			    const struct ethnl_reply_data *reply_base)
+{
+	const struct debug_reply_data *data = DEBUG_REPDATA(reply_base);
+	bool compact = req_base->flags & ETHTOOL_FLAG_COMPACT_BITSETS;
+
+	return ethnl_bitset32_size(&data->msg_mask, NULL, NETIF_MSG_CLASS_COUNT,
+				   netif_msg_class_names, compact);
+}
+
+static int debug_fill_reply(struct sk_buff *skb,
+			    const struct ethnl_req_info *req_base,
+			    const struct ethnl_reply_data *reply_base)
+{
+	const struct debug_reply_data *data = DEBUG_REPDATA(reply_base);
+	bool compact = req_base->flags & ETHTOOL_FLAG_COMPACT_BITSETS;
+
+	return ethnl_put_bitset32(skb, ETHTOOL_A_DEBUG_MSGMASK, &data->msg_mask,
+				  NULL, NETIF_MSG_CLASS_COUNT,
+				  netif_msg_class_names, compact);
+}
+
+const struct ethnl_request_ops ethnl_debug_request_ops = {
+	.request_cmd		= ETHTOOL_MSG_DEBUG_GET,
+	.reply_cmd		= ETHTOOL_MSG_DEBUG_GET_REPLY,
+	.hdr_attr		= ETHTOOL_A_DEBUG_HEADER,
+	.max_attr		= ETHTOOL_A_DEBUG_MAX,
+	.req_info_size		= sizeof(struct debug_req_info),
+	.reply_data_size	= sizeof(struct debug_reply_data),
+	.request_policy		= debug_get_policy,
+
+	.prepare_data		= debug_prepare_data,
+	.reply_size		= debug_reply_size,
+	.fill_reply		= debug_fill_reply,
+};
diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index 0af43bbdb9b2..bdaf583e392b 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -213,6 +213,7 @@ ethnl_default_requests[__ETHTOOL_MSG_USER_CNT] = {
 	[ETHTOOL_MSG_LINKINFO_GET]	= &ethnl_linkinfo_request_ops,
 	[ETHTOOL_MSG_LINKMODES_GET]	= &ethnl_linkmodes_request_ops,
 	[ETHTOOL_MSG_LINKSTATE_GET]	= &ethnl_linkstate_request_ops,
+	[ETHTOOL_MSG_DEBUG_GET]		= &ethnl_debug_request_ops,
 };
 
 static struct ethnl_dump_ctx *ethnl_dump_context(struct netlink_callback *cb)
@@ -664,6 +665,13 @@ static const struct genl_ops ethtool_genl_ops[] = {
 		.dumpit	= ethnl_default_dumpit,
 		.done	= ethnl_default_done,
 	},
+	{
+		.cmd	= ETHTOOL_MSG_DEBUG_GET,
+		.doit	= ethnl_default_doit,
+		.start	= ethnl_default_start,
+		.dumpit	= ethnl_default_dumpit,
+		.done	= ethnl_default_done,
+	},
 };
 
 static const struct genl_multicast_group ethtool_nl_mcgrps[] = {
diff --git a/net/ethtool/netlink.h b/net/ethtool/netlink.h
index da9d6521a4eb..9bd8ef671501 100644
--- a/net/ethtool/netlink.h
+++ b/net/ethtool/netlink.h
@@ -334,6 +334,7 @@ extern const struct ethnl_request_ops ethnl_strset_request_ops;
 extern const struct ethnl_request_ops ethnl_linkinfo_request_ops;
 extern const struct ethnl_request_ops ethnl_linkmodes_request_ops;
 extern const struct ethnl_request_ops ethnl_linkstate_request_ops;
+extern const struct ethnl_request_ops ethnl_debug_request_ops;
 
 int ethnl_set_linkinfo(struct sk_buff *skb, struct genl_info *info);
 int ethnl_set_linkmodes(struct sk_buff *skb, struct genl_info *info);
diff --git a/net/ethtool/strset.c b/net/ethtool/strset.c
index 948d967f1eca..7a45c25355b8 100644
--- a/net/ethtool/strset.c
+++ b/net/ethtool/strset.c
@@ -50,6 +50,11 @@ static const struct strset_info info_template[] = {
 		.count		= __ETHTOOL_LINK_MODE_MASK_NBITS,
 		.strings	= link_mode_names,
 	},
+	[ETH_SS_MSG_CLASSES] = {
+		.per_dev	= false,
+		.count		= NETIF_MSG_CLASS_COUNT,
+		.strings	= netif_msg_class_names,
+	},
 };
 
 struct strset_req_info {
-- 
cgit v1.2.3-59-g8ed1b


From e54d04e3afead22d8e7d6edaaac487a1205bac39 Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Sun, 26 Jan 2020 23:11:07 +0100
Subject: ethtool: set message mask with DEBUG_SET request

Implement DEBUG_SET netlink request to set debugging settings for a device.
At the moment, only message mask corresponding to message level as set by
ETHTOOL_SMSGLVL ioctl request can be set. (It is called message level in
ioctl interface but almost all drivers interpret it as a bit mask.)

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/ethtool-netlink.rst | 20 ++++++++++-
 include/uapi/linux/ethtool_netlink.h         |  1 +
 net/ethtool/debug.c                          | 53 ++++++++++++++++++++++++++++
 net/ethtool/netlink.c                        |  5 +++
 net/ethtool/netlink.h                        |  1 +
 5 files changed, 79 insertions(+), 1 deletion(-)

diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
index 76a411d3102d..58473c05319f 100644
--- a/Documentation/networking/ethtool-netlink.rst
+++ b/Documentation/networking/ethtool-netlink.rst
@@ -186,6 +186,7 @@ Userspace to kernel:
   ``ETHTOOL_MSG_LINKMODES_SET``         set link modes info
   ``ETHTOOL_MSG_LINKSTATE_GET``         get link state
   ``ETHTOOL_MSG_DEBUG_GET``             get debugging settings
+  ``ETHTOOL_MSG_DEBUG_SET``             set debugging settings
   ===================================== ================================
 
 Kernel to userspace:
@@ -455,6 +456,23 @@ interface follows its actual use in practice.
 devices supporting the request).
 
 
+DEBUG_SET
+=========
+
+Set or update debugging settings of a device. At the moment, only message mask
+is supported.
+
+Request contents:
+
+  ====================================  ======  ==========================
+  ``ETHTOOL_A_DEBUG_HEADER``            nested  request header
+  ``ETHTOOL_A_DEBUG_MSGMASK``           bitset  message mask
+  ====================================  ======  ==========================
+
+``ETHTOOL_A_DEBUG_MSGMASK`` bit set allows setting or modifying mask of
+enabled debugging message types for the device.
+
+
 Request translation
 ===================
 
@@ -474,7 +492,7 @@ have their netlink replacement yet.
   ``ETHTOOL_GWOL``                    n/a
   ``ETHTOOL_SWOL``                    n/a
   ``ETHTOOL_GMSGLVL``                 ``ETHTOOL_MSG_DEBUG_GET``
-  ``ETHTOOL_SMSGLVL``                 n/a
+  ``ETHTOOL_SMSGLVL``                 ``ETHTOOL_MSG_DEBUG_SET``
   ``ETHTOOL_NWAY_RST``                n/a
   ``ETHTOOL_GLINK``                   ``ETHTOOL_MSG_LINKSTATE_GET``
   ``ETHTOOL_GEEPROM``                 n/a
diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
index 0d39e04567cb..8f0b6fd277d5 100644
--- a/include/uapi/linux/ethtool_netlink.h
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -21,6 +21,7 @@ enum {
 	ETHTOOL_MSG_LINKMODES_SET,
 	ETHTOOL_MSG_LINKSTATE_GET,
 	ETHTOOL_MSG_DEBUG_GET,
+	ETHTOOL_MSG_DEBUG_SET,
 
 	/* add new constants above here */
 	__ETHTOOL_MSG_USER_CNT,
diff --git a/net/ethtool/debug.c b/net/ethtool/debug.c
index cc121a23be5f..6fc3ef8113c0 100644
--- a/net/ethtool/debug.c
+++ b/net/ethtool/debug.c
@@ -78,3 +78,56 @@ const struct ethnl_request_ops ethnl_debug_request_ops = {
 	.reply_size		= debug_reply_size,
 	.fill_reply		= debug_fill_reply,
 };
+
+/* DEBUG_SET */
+
+static const struct nla_policy
+debug_set_policy[ETHTOOL_A_DEBUG_MAX + 1] = {
+	[ETHTOOL_A_DEBUG_UNSPEC]	= { .type = NLA_REJECT },
+	[ETHTOOL_A_DEBUG_HEADER]	= { .type = NLA_NESTED },
+	[ETHTOOL_A_DEBUG_MSGMASK]	= { .type = NLA_NESTED },
+};
+
+int ethnl_set_debug(struct sk_buff *skb, struct genl_info *info)
+{
+	struct nlattr *tb[ETHTOOL_A_DEBUG_MAX + 1];
+	struct ethnl_req_info req_info = {};
+	struct net_device *dev;
+	bool mod = false;
+	u32 msg_mask;
+	int ret;
+
+	ret = nlmsg_parse(info->nlhdr, GENL_HDRLEN, tb,
+			  ETHTOOL_A_DEBUG_MAX, debug_set_policy,
+			  info->extack);
+	if (ret < 0)
+		return ret;
+	ret = ethnl_parse_header(&req_info, tb[ETHTOOL_A_DEBUG_HEADER],
+				 genl_info_net(info), info->extack, true);
+	if (ret < 0)
+		return ret;
+	dev = req_info.dev;
+	if (!dev->ethtool_ops->get_msglevel || !dev->ethtool_ops->set_msglevel)
+		return -EOPNOTSUPP;
+
+	rtnl_lock();
+	ret = ethnl_ops_begin(dev);
+	if (ret < 0)
+		goto out_rtnl;
+
+	msg_mask = dev->ethtool_ops->get_msglevel(dev);
+	ret = ethnl_update_bitset32(&msg_mask, NETIF_MSG_CLASS_COUNT,
+				    tb[ETHTOOL_A_DEBUG_MSGMASK],
+				    netif_msg_class_names, info->extack, &mod);
+	if (ret < 0 || !mod)
+		goto out_ops;
+
+	dev->ethtool_ops->set_msglevel(dev, msg_mask);
+
+out_ops:
+	ethnl_ops_complete(dev);
+out_rtnl:
+	rtnl_unlock();
+	dev_put(dev);
+	return ret;
+}
diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index bdaf583e392b..bcdc42e0bd14 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -672,6 +672,11 @@ static const struct genl_ops ethtool_genl_ops[] = {
 		.dumpit	= ethnl_default_dumpit,
 		.done	= ethnl_default_done,
 	},
+	{
+		.cmd	= ETHTOOL_MSG_DEBUG_SET,
+		.flags	= GENL_UNS_ADMIN_PERM,
+		.doit	= ethnl_set_debug,
+	},
 };
 
 static const struct genl_multicast_group ethtool_nl_mcgrps[] = {
diff --git a/net/ethtool/netlink.h b/net/ethtool/netlink.h
index 9bd8ef671501..772723e536e8 100644
--- a/net/ethtool/netlink.h
+++ b/net/ethtool/netlink.h
@@ -338,5 +338,6 @@ extern const struct ethnl_request_ops ethnl_debug_request_ops;
 
 int ethnl_set_linkinfo(struct sk_buff *skb, struct genl_info *info);
 int ethnl_set_linkmodes(struct sk_buff *skb, struct genl_info *info);
+int ethnl_set_debug(struct sk_buff *skb, struct genl_info *info);
 
 #endif /* _NET_ETHTOOL_NETLINK_H */
-- 
cgit v1.2.3-59-g8ed1b


From 0bda7af39d2ba6ef83bd18e26bcb027f1630004e Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Sun, 26 Jan 2020 23:11:10 +0100
Subject: ethtool: add DEBUG_NTF notification

Send ETHTOOL_MSG_DEBUG_NTF notification message whenever debugging message
mask for a device are modified using ETHTOOL_MSG_DEBUG_SET netlink message
or ETHTOOL_SMSGLVL ioctl request.

The notification message has the same format as reply to DEBUG_GET request.
As with other ethtool notifications, netlink requests only trigger the
notification if the mask is actually changed while ioctl request trigger it
whenever the request results in calling the ethtool_ops handler.

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/ethtool-netlink.rst | 1 +
 include/uapi/linux/ethtool_netlink.h         | 1 +
 net/ethtool/debug.c                          | 1 +
 net/ethtool/ioctl.c                          | 2 ++
 net/ethtool/netlink.c                        | 2 ++
 5 files changed, 7 insertions(+)

diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
index 58473c05319f..2263f885d18f 100644
--- a/Documentation/networking/ethtool-netlink.rst
+++ b/Documentation/networking/ethtool-netlink.rst
@@ -199,6 +199,7 @@ Kernel to userspace:
   ``ETHTOOL_MSG_LINKMODES_NTF``         link modes notification
   ``ETHTOOL_MSG_LINKSTATE_GET_REPLY``   link state info
   ``ETHTOOL_MSG_DEBUG_GET_REPLY``       debugging settings
+  ``ETHTOOL_MSG_DEBUG_NTF``             debugging settings notification
   ===================================== ================================
 
 ``GET`` requests are sent by userspace applications to retrieve device
diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
index 8f0b6fd277d5..67a06b94bf28 100644
--- a/include/uapi/linux/ethtool_netlink.h
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -38,6 +38,7 @@ enum {
 	ETHTOOL_MSG_LINKMODES_NTF,
 	ETHTOOL_MSG_LINKSTATE_GET_REPLY,
 	ETHTOOL_MSG_DEBUG_GET_REPLY,
+	ETHTOOL_MSG_DEBUG_NTF,
 
 	/* add new constants above here */
 	__ETHTOOL_MSG_KERNEL_CNT,
diff --git a/net/ethtool/debug.c b/net/ethtool/debug.c
index 6fc3ef8113c0..aaef4843e6ba 100644
--- a/net/ethtool/debug.c
+++ b/net/ethtool/debug.c
@@ -123,6 +123,7 @@ int ethnl_set_debug(struct sk_buff *skb, struct genl_info *info)
 		goto out_ops;
 
 	dev->ethtool_ops->set_msglevel(dev, msg_mask);
+	ethtool_notify(dev, ETHTOOL_MSG_DEBUG_NTF, NULL);
 
 out_ops:
 	ethnl_ops_complete(dev);
diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
index 182bffbffa78..46e0b31782fc 100644
--- a/net/ethtool/ioctl.c
+++ b/net/ethtool/ioctl.c
@@ -2545,6 +2545,8 @@ int dev_ethtool(struct net *net, struct ifreq *ifr)
 	case ETHTOOL_SMSGLVL:
 		rc = ethtool_set_value_void(dev, useraddr,
 				       dev->ethtool_ops->set_msglevel);
+		if (!rc)
+			ethtool_notify(dev, ETHTOOL_MSG_DEBUG_NTF, NULL);
 		break;
 	case ETHTOOL_GEEE:
 		rc = ethtool_get_eee(dev, useraddr);
diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index bcdc42e0bd14..9a0a6c2f8dbb 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -524,6 +524,7 @@ static const struct ethnl_request_ops *
 ethnl_default_notify_ops[ETHTOOL_MSG_KERNEL_MAX + 1] = {
 	[ETHTOOL_MSG_LINKINFO_NTF]	= &ethnl_linkinfo_request_ops,
 	[ETHTOOL_MSG_LINKMODES_NTF]	= &ethnl_linkmodes_request_ops,
+	[ETHTOOL_MSG_DEBUG_NTF]		= &ethnl_debug_request_ops,
 };
 
 /* default notification handler */
@@ -607,6 +608,7 @@ typedef void (*ethnl_notify_handler_t)(struct net_device *dev, unsigned int cmd,
 static const ethnl_notify_handler_t ethnl_notify_handlers[] = {
 	[ETHTOOL_MSG_LINKINFO_NTF]	= ethnl_default_notify,
 	[ETHTOOL_MSG_LINKMODES_NTF]	= ethnl_default_notify,
+	[ETHTOOL_MSG_DEBUG_NTF]		= ethnl_default_notify,
 };
 
 void ethtool_notify(struct net_device *dev, unsigned int cmd, const void *data)
-- 
cgit v1.2.3-59-g8ed1b


From 51ea22b04ea0c210f9ce87b8a600965dbe476bc2 Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Sun, 26 Jan 2020 23:11:13 +0100
Subject: ethtool: provide WoL settings with WOL_GET request

Implement WOL_GET request to get wake-on-lan settings for a device,
traditionally available via ETHTOOL_GWOL ioctl request.

As part of the implementation, provide symbolic names for wake-on-line
modes as ETH_SS_WOL_MODES string set.

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/ethtool-netlink.rst | 30 ++++++++-
 include/uapi/linux/ethtool.h                 |  4 ++
 include/uapi/linux/ethtool_netlink.h         | 15 +++++
 net/ethtool/Makefile                         |  2 +-
 net/ethtool/common.c                         | 12 ++++
 net/ethtool/common.h                         |  1 +
 net/ethtool/netlink.c                        |  9 +++
 net/ethtool/netlink.h                        |  1 +
 net/ethtool/strset.c                         |  5 ++
 net/ethtool/wol.c                            | 99 ++++++++++++++++++++++++++++
 10 files changed, 176 insertions(+), 2 deletions(-)
 create mode 100644 net/ethtool/wol.c

diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
index 2263f885d18f..5fd85e3ea96e 100644
--- a/Documentation/networking/ethtool-netlink.rst
+++ b/Documentation/networking/ethtool-netlink.rst
@@ -187,6 +187,7 @@ Userspace to kernel:
   ``ETHTOOL_MSG_LINKSTATE_GET``         get link state
   ``ETHTOOL_MSG_DEBUG_GET``             get debugging settings
   ``ETHTOOL_MSG_DEBUG_SET``             set debugging settings
+  ``ETHTOOL_MSG_WOL_GET``               get wake-on-lan settings
   ===================================== ================================
 
 Kernel to userspace:
@@ -200,6 +201,7 @@ Kernel to userspace:
   ``ETHTOOL_MSG_LINKSTATE_GET_REPLY``   link state info
   ``ETHTOOL_MSG_DEBUG_GET_REPLY``       debugging settings
   ``ETHTOOL_MSG_DEBUG_NTF``             debugging settings notification
+  ``ETHTOOL_MSG_WOL_GET_REPLY``         wake-on-lan settings
   ===================================== ================================
 
 ``GET`` requests are sent by userspace applications to retrieve device
@@ -474,6 +476,32 @@ Request contents:
 enabled debugging message types for the device.
 
 
+WOL_GET
+=======
+
+Query device wake-on-lan settings. Unlike most "GET" type requests,
+``ETHTOOL_MSG_WOL_GET`` requires (netns) ``CAP_NET_ADMIN`` privileges as it
+(potentially) provides SecureOn(tm) password which is confidential.
+
+Request contents:
+
+  ====================================  ======  ==========================
+  ``ETHTOOL_A_WOL_HEADER``              nested  request header
+  ====================================  ======  ==========================
+
+Kernel response contents:
+
+  ====================================  ======  ==========================
+  ``ETHTOOL_A_WOL_HEADER``              nested  reply header
+  ``ETHTOOL_A_WOL_MODES``               bitset  mask of enabled WoL modes
+  ``ETHTOOL_A_WOL_SOPASS``              binary  SecureOn(tm) password
+  ====================================  ======  ==========================
+
+In reply, ``ETHTOOL_A_WOL_MODES`` mask consists of modes supported by the
+device, value of modes which are enabled. ``ETHTOOL_A_WOL_SOPASS`` is only
+included in reply if ``WAKE_MAGICSECURE`` mode is supported.
+
+
 Request translation
 ===================
 
@@ -490,7 +518,7 @@ have their netlink replacement yet.
                                       ``ETHTOOL_MSG_LINKMODES_SET``
   ``ETHTOOL_GDRVINFO``                n/a
   ``ETHTOOL_GREGS``                   n/a
-  ``ETHTOOL_GWOL``                    n/a
+  ``ETHTOOL_GWOL``                    ``ETHTOOL_MSG_WOL_GET``
   ``ETHTOOL_SWOL``                    n/a
   ``ETHTOOL_GMSGLVL``                 ``ETHTOOL_MSG_DEBUG_GET``
   ``ETHTOOL_SMSGLVL``                 ``ETHTOOL_MSG_DEBUG_SET``
diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h
index 456fb2aa0fad..4295ebfa2f91 100644
--- a/include/uapi/linux/ethtool.h
+++ b/include/uapi/linux/ethtool.h
@@ -595,6 +595,7 @@ struct ethtool_pauseparam {
  * @ETH_SS_PHY_TUNABLES: PHY tunable names
  * @ETH_SS_LINK_MODES: link mode names
  * @ETH_SS_MSG_CLASSES: debug message class names
+ * @ETH_SS_WOL_MODES: wake-on-lan modes
  */
 enum ethtool_stringset {
 	ETH_SS_TEST		= 0,
@@ -608,6 +609,7 @@ enum ethtool_stringset {
 	ETH_SS_PHY_TUNABLES,
 	ETH_SS_LINK_MODES,
 	ETH_SS_MSG_CLASSES,
+	ETH_SS_WOL_MODES,
 
 	/* add new constants above here */
 	ETH_SS_COUNT
@@ -1695,6 +1697,8 @@ static inline int ethtool_validate_duplex(__u8 duplex)
 #define WAKE_MAGICSECURE	(1 << 6) /* only meaningful if WAKE_MAGIC */
 #define WAKE_FILTER		(1 << 7)
 
+#define WOL_MODE_COUNT		8
+
 /* L2-L4 network traffic flow types */
 #define	TCP_V4_FLOW	0x01	/* hash or spec (tcp_ip4_spec) */
 #define	UDP_V4_FLOW	0x02	/* hash or spec (udp_ip4_spec) */
diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
index 67a06b94bf28..dcc5c32dc018 100644
--- a/include/uapi/linux/ethtool_netlink.h
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -22,6 +22,7 @@ enum {
 	ETHTOOL_MSG_LINKSTATE_GET,
 	ETHTOOL_MSG_DEBUG_GET,
 	ETHTOOL_MSG_DEBUG_SET,
+	ETHTOOL_MSG_WOL_GET,
 
 	/* add new constants above here */
 	__ETHTOOL_MSG_USER_CNT,
@@ -39,6 +40,7 @@ enum {
 	ETHTOOL_MSG_LINKSTATE_GET_REPLY,
 	ETHTOOL_MSG_DEBUG_GET_REPLY,
 	ETHTOOL_MSG_DEBUG_NTF,
+	ETHTOOL_MSG_WOL_GET_REPLY,
 
 	/* add new constants above here */
 	__ETHTOOL_MSG_KERNEL_CNT,
@@ -211,6 +213,19 @@ enum {
 	ETHTOOL_A_DEBUG_MAX = __ETHTOOL_A_DEBUG_CNT - 1
 };
 
+/* WOL */
+
+enum {
+	ETHTOOL_A_WOL_UNSPEC,
+	ETHTOOL_A_WOL_HEADER,			/* nest - _A_HEADER_* */
+	ETHTOOL_A_WOL_MODES,			/* bitset */
+	ETHTOOL_A_WOL_SOPASS,			/* binary */
+
+	/* add new constants above here */
+	__ETHTOOL_A_WOL_CNT,
+	ETHTOOL_A_WOL_MAX = __ETHTOOL_A_WOL_CNT - 1
+};
+
 /* generic netlink info */
 #define ETHTOOL_GENL_NAME "ethtool"
 #define ETHTOOL_GENL_VERSION 1
diff --git a/net/ethtool/Makefile b/net/ethtool/Makefile
index c120c820a4f5..424545a4aaec 100644
--- a/net/ethtool/Makefile
+++ b/net/ethtool/Makefile
@@ -5,4 +5,4 @@ obj-y				+= ioctl.o common.o
 obj-$(CONFIG_ETHTOOL_NETLINK)	+= ethtool_nl.o
 
 ethtool_nl-y	:= netlink.o bitset.o strset.o linkinfo.o linkmodes.o \
-		   linkstate.o debug.o
+		   linkstate.o debug.o wol.o
diff --git a/net/ethtool/common.c b/net/ethtool/common.c
index aa1183a65a76..636ec6d5110e 100644
--- a/net/ethtool/common.c
+++ b/net/ethtool/common.c
@@ -190,6 +190,18 @@ const char netif_msg_class_names[][ETH_GSTRING_LEN] = {
 };
 static_assert(ARRAY_SIZE(netif_msg_class_names) == NETIF_MSG_CLASS_COUNT);
 
+const char wol_mode_names[][ETH_GSTRING_LEN] = {
+	[const_ilog2(WAKE_PHY)]		= "phy",
+	[const_ilog2(WAKE_UCAST)]	= "ucast",
+	[const_ilog2(WAKE_MCAST)]	= "mcast",
+	[const_ilog2(WAKE_BCAST)]	= "bcast",
+	[const_ilog2(WAKE_ARP)]		= "arp",
+	[const_ilog2(WAKE_MAGIC)]	= "magic",
+	[const_ilog2(WAKE_MAGICSECURE)]	= "magicsecure",
+	[const_ilog2(WAKE_FILTER)]	= "filter",
+};
+static_assert(ARRAY_SIZE(wol_mode_names) == WOL_MODE_COUNT);
+
 /* return false if legacy contained non-0 deprecated fields
  * maxtxpkt/maxrxpkt. rest of ksettings always updated
  */
diff --git a/net/ethtool/common.h b/net/ethtool/common.h
index 064c5c3aa990..40ba74e0b9bb 100644
--- a/net/ethtool/common.h
+++ b/net/ethtool/common.h
@@ -20,6 +20,7 @@ extern const char
 phy_tunable_strings[__ETHTOOL_PHY_TUNABLE_COUNT][ETH_GSTRING_LEN];
 extern const char link_mode_names[][ETH_GSTRING_LEN];
 extern const char netif_msg_class_names[][ETH_GSTRING_LEN];
+extern const char wol_mode_names[][ETH_GSTRING_LEN];
 
 int __ethtool_get_link(struct net_device *dev);
 
diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index 9a0a6c2f8dbb..eeb6d8594e1b 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -214,6 +214,7 @@ ethnl_default_requests[__ETHTOOL_MSG_USER_CNT] = {
 	[ETHTOOL_MSG_LINKMODES_GET]	= &ethnl_linkmodes_request_ops,
 	[ETHTOOL_MSG_LINKSTATE_GET]	= &ethnl_linkstate_request_ops,
 	[ETHTOOL_MSG_DEBUG_GET]		= &ethnl_debug_request_ops,
+	[ETHTOOL_MSG_WOL_GET]		= &ethnl_wol_request_ops,
 };
 
 static struct ethnl_dump_ctx *ethnl_dump_context(struct netlink_callback *cb)
@@ -679,6 +680,14 @@ static const struct genl_ops ethtool_genl_ops[] = {
 		.flags	= GENL_UNS_ADMIN_PERM,
 		.doit	= ethnl_set_debug,
 	},
+	{
+		.cmd	= ETHTOOL_MSG_WOL_GET,
+		.flags	= GENL_UNS_ADMIN_PERM,
+		.doit	= ethnl_default_doit,
+		.start	= ethnl_default_start,
+		.dumpit	= ethnl_default_dumpit,
+		.done	= ethnl_default_done,
+	},
 };
 
 static const struct genl_multicast_group ethtool_nl_mcgrps[] = {
diff --git a/net/ethtool/netlink.h b/net/ethtool/netlink.h
index 772723e536e8..9fcd6f87b396 100644
--- a/net/ethtool/netlink.h
+++ b/net/ethtool/netlink.h
@@ -335,6 +335,7 @@ extern const struct ethnl_request_ops ethnl_linkinfo_request_ops;
 extern const struct ethnl_request_ops ethnl_linkmodes_request_ops;
 extern const struct ethnl_request_ops ethnl_linkstate_request_ops;
 extern const struct ethnl_request_ops ethnl_debug_request_ops;
+extern const struct ethnl_request_ops ethnl_wol_request_ops;
 
 int ethnl_set_linkinfo(struct sk_buff *skb, struct genl_info *info);
 int ethnl_set_linkmodes(struct sk_buff *skb, struct genl_info *info);
diff --git a/net/ethtool/strset.c b/net/ethtool/strset.c
index 7a45c25355b8..8e5911887b4c 100644
--- a/net/ethtool/strset.c
+++ b/net/ethtool/strset.c
@@ -55,6 +55,11 @@ static const struct strset_info info_template[] = {
 		.count		= NETIF_MSG_CLASS_COUNT,
 		.strings	= netif_msg_class_names,
 	},
+	[ETH_SS_WOL_MODES] = {
+		.per_dev	= false,
+		.count		= WOL_MODE_COUNT,
+		.strings	= wol_mode_names,
+	},
 };
 
 struct strset_req_info {
diff --git a/net/ethtool/wol.c b/net/ethtool/wol.c
new file mode 100644
index 000000000000..7c9a1ef622ce
--- /dev/null
+++ b/net/ethtool/wol.c
@@ -0,0 +1,99 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include "netlink.h"
+#include "common.h"
+#include "bitset.h"
+
+struct wol_req_info {
+	struct ethnl_req_info		base;
+};
+
+struct wol_reply_data {
+	struct ethnl_reply_data		base;
+	struct ethtool_wolinfo		wol;
+	bool				show_sopass;
+};
+
+#define WOL_REPDATA(__reply_base) \
+	container_of(__reply_base, struct wol_reply_data, base)
+
+static const struct nla_policy
+wol_get_policy[ETHTOOL_A_WOL_MAX + 1] = {
+	[ETHTOOL_A_WOL_UNSPEC]		= { .type = NLA_REJECT },
+	[ETHTOOL_A_WOL_HEADER]		= { .type = NLA_NESTED },
+	[ETHTOOL_A_WOL_MODES]		= { .type = NLA_REJECT },
+	[ETHTOOL_A_WOL_SOPASS]		= { .type = NLA_REJECT },
+};
+
+static int wol_prepare_data(const struct ethnl_req_info *req_base,
+			    struct ethnl_reply_data *reply_base,
+			    struct genl_info *info)
+{
+	struct wol_reply_data *data = WOL_REPDATA(reply_base);
+	struct net_device *dev = reply_base->dev;
+	int ret;
+
+	if (!dev->ethtool_ops->get_wol)
+		return -EOPNOTSUPP;
+
+	ret = ethnl_ops_begin(dev);
+	if (ret < 0)
+		return ret;
+	dev->ethtool_ops->get_wol(dev, &data->wol);
+	ethnl_ops_complete(dev);
+	data->show_sopass = data->wol.supported & WAKE_MAGICSECURE;
+
+	return 0;
+}
+
+static int wol_reply_size(const struct ethnl_req_info *req_base,
+			  const struct ethnl_reply_data *reply_base)
+{
+	bool compact = req_base->flags & ETHTOOL_FLAG_COMPACT_BITSETS;
+	const struct wol_reply_data *data = WOL_REPDATA(reply_base);
+	int len;
+
+	len = ethnl_bitset32_size(&data->wol.wolopts, &data->wol.supported,
+				  WOL_MODE_COUNT, wol_mode_names, compact);
+	if (len < 0)
+		return len;
+	if (data->show_sopass)
+		len += nla_total_size(sizeof(data->wol.sopass));
+
+	return len;
+}
+
+static int wol_fill_reply(struct sk_buff *skb,
+			  const struct ethnl_req_info *req_base,
+			  const struct ethnl_reply_data *reply_base)
+{
+	bool compact = req_base->flags & ETHTOOL_FLAG_COMPACT_BITSETS;
+	const struct wol_reply_data *data = WOL_REPDATA(reply_base);
+	int ret;
+
+	ret = ethnl_put_bitset32(skb, ETHTOOL_A_WOL_MODES, &data->wol.wolopts,
+				 &data->wol.supported, WOL_MODE_COUNT,
+				 wol_mode_names, compact);
+	if (ret < 0)
+		return ret;
+	if (data->show_sopass &&
+	    nla_put(skb, ETHTOOL_A_WOL_SOPASS, sizeof(data->wol.sopass),
+		    data->wol.sopass))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+const struct ethnl_request_ops ethnl_wol_request_ops = {
+	.request_cmd		= ETHTOOL_MSG_WOL_GET,
+	.reply_cmd		= ETHTOOL_MSG_WOL_GET_REPLY,
+	.hdr_attr		= ETHTOOL_A_WOL_HEADER,
+	.max_attr		= ETHTOOL_A_WOL_MAX,
+	.req_info_size		= sizeof(struct wol_req_info),
+	.reply_data_size	= sizeof(struct wol_reply_data),
+	.request_policy		= wol_get_policy,
+
+	.prepare_data		= wol_prepare_data,
+	.reply_size		= wol_reply_size,
+	.fill_reply		= wol_fill_reply,
+};
-- 
cgit v1.2.3-59-g8ed1b


From 8d425b19b305a77cb1ba67b84e7c1ca547de032f Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Sun, 26 Jan 2020 23:11:16 +0100
Subject: ethtool: set wake-on-lan settings with WOL_SET request

Implement WOL_SET netlink request to set wake-on-lan settings. This is
equivalent to ETHTOOL_SWOL ioctl request.

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/ethtool-netlink.rst | 20 +++++++-
 include/uapi/linux/ethtool_netlink.h         |  1 +
 net/ethtool/netlink.c                        |  5 ++
 net/ethtool/netlink.h                        |  1 +
 net/ethtool/wol.c                            | 76 ++++++++++++++++++++++++++++
 5 files changed, 102 insertions(+), 1 deletion(-)

diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
index 5fd85e3ea96e..f16f74bbb546 100644
--- a/Documentation/networking/ethtool-netlink.rst
+++ b/Documentation/networking/ethtool-netlink.rst
@@ -188,6 +188,7 @@ Userspace to kernel:
   ``ETHTOOL_MSG_DEBUG_GET``             get debugging settings
   ``ETHTOOL_MSG_DEBUG_SET``             set debugging settings
   ``ETHTOOL_MSG_WOL_GET``               get wake-on-lan settings
+  ``ETHTOOL_MSG_WOL_SET``               set wake-on-lan settings
   ===================================== ================================
 
 Kernel to userspace:
@@ -502,6 +503,23 @@ device, value of modes which are enabled. ``ETHTOOL_A_WOL_SOPASS`` is only
 included in reply if ``WAKE_MAGICSECURE`` mode is supported.
 
 
+WOL_SET
+=======
+
+Set or update wake-on-lan settings.
+
+Request contents:
+
+  ====================================  ======  ==========================
+  ``ETHTOOL_A_WOL_HEADER``              nested  request header
+  ``ETHTOOL_A_WOL_MODES``               bitset  enabled WoL modes
+  ``ETHTOOL_A_WOL_SOPASS``              binary  SecureOn(tm) password
+  ====================================  ======  ==========================
+
+``ETHTOOL_A_WOL_SOPASS`` is only allowed for devices supporting
+``WAKE_MAGICSECURE`` mode.
+
+
 Request translation
 ===================
 
@@ -519,7 +537,7 @@ have their netlink replacement yet.
   ``ETHTOOL_GDRVINFO``                n/a
   ``ETHTOOL_GREGS``                   n/a
   ``ETHTOOL_GWOL``                    ``ETHTOOL_MSG_WOL_GET``
-  ``ETHTOOL_SWOL``                    n/a
+  ``ETHTOOL_SWOL``                    ``ETHTOOL_MSG_WOL_SET``
   ``ETHTOOL_GMSGLVL``                 ``ETHTOOL_MSG_DEBUG_GET``
   ``ETHTOOL_SMSGLVL``                 ``ETHTOOL_MSG_DEBUG_SET``
   ``ETHTOOL_NWAY_RST``                n/a
diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
index dcc5c32dc018..59de35695521 100644
--- a/include/uapi/linux/ethtool_netlink.h
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -23,6 +23,7 @@ enum {
 	ETHTOOL_MSG_DEBUG_GET,
 	ETHTOOL_MSG_DEBUG_SET,
 	ETHTOOL_MSG_WOL_GET,
+	ETHTOOL_MSG_WOL_SET,
 
 	/* add new constants above here */
 	__ETHTOOL_MSG_USER_CNT,
diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index eeb6d8594e1b..2c375f9095fe 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -688,6 +688,11 @@ static const struct genl_ops ethtool_genl_ops[] = {
 		.dumpit	= ethnl_default_dumpit,
 		.done	= ethnl_default_done,
 	},
+	{
+		.cmd	= ETHTOOL_MSG_WOL_SET,
+		.flags	= GENL_UNS_ADMIN_PERM,
+		.doit	= ethnl_set_wol,
+	},
 };
 
 static const struct genl_multicast_group ethtool_nl_mcgrps[] = {
diff --git a/net/ethtool/netlink.h b/net/ethtool/netlink.h
index 9fcd6f87b396..60efd87686ad 100644
--- a/net/ethtool/netlink.h
+++ b/net/ethtool/netlink.h
@@ -340,5 +340,6 @@ extern const struct ethnl_request_ops ethnl_wol_request_ops;
 int ethnl_set_linkinfo(struct sk_buff *skb, struct genl_info *info);
 int ethnl_set_linkmodes(struct sk_buff *skb, struct genl_info *info);
 int ethnl_set_debug(struct sk_buff *skb, struct genl_info *info);
+int ethnl_set_wol(struct sk_buff *skb, struct genl_info *info);
 
 #endif /* _NET_ETHTOOL_NETLINK_H */
diff --git a/net/ethtool/wol.c b/net/ethtool/wol.c
index 7c9a1ef622ce..a2724378fac4 100644
--- a/net/ethtool/wol.c
+++ b/net/ethtool/wol.c
@@ -97,3 +97,79 @@ const struct ethnl_request_ops ethnl_wol_request_ops = {
 	.reply_size		= wol_reply_size,
 	.fill_reply		= wol_fill_reply,
 };
+
+/* WOL_SET */
+
+static const struct nla_policy
+wol_set_policy[ETHTOOL_A_WOL_MAX + 1] = {
+	[ETHTOOL_A_WOL_UNSPEC]		= { .type = NLA_REJECT },
+	[ETHTOOL_A_WOL_HEADER]		= { .type = NLA_NESTED },
+	[ETHTOOL_A_WOL_MODES]		= { .type = NLA_NESTED },
+	[ETHTOOL_A_WOL_SOPASS]		= { .type = NLA_BINARY,
+					    .len = SOPASS_MAX },
+};
+
+int ethnl_set_wol(struct sk_buff *skb, struct genl_info *info)
+{
+	struct ethtool_wolinfo wol = { .cmd = ETHTOOL_GWOL };
+	struct nlattr *tb[ETHTOOL_A_WOL_MAX + 1];
+	struct ethnl_req_info req_info = {};
+	struct net_device *dev;
+	bool mod = false;
+	int ret;
+
+	ret = nlmsg_parse(info->nlhdr, GENL_HDRLEN, tb, ETHTOOL_A_WOL_MAX,
+			  wol_set_policy, info->extack);
+	if (ret < 0)
+		return ret;
+	ret = ethnl_parse_header(&req_info, tb[ETHTOOL_A_WOL_HEADER],
+				 genl_info_net(info), info->extack, true);
+	if (ret < 0)
+		return ret;
+	dev = req_info.dev;
+	if (!dev->ethtool_ops->get_wol || !dev->ethtool_ops->set_wol)
+		return -EOPNOTSUPP;
+
+	rtnl_lock();
+	ret = ethnl_ops_begin(dev);
+	if (ret < 0)
+		goto out_rtnl;
+
+	dev->ethtool_ops->get_wol(dev, &wol);
+	ret = ethnl_update_bitset32(&wol.wolopts, WOL_MODE_COUNT,
+				    tb[ETHTOOL_A_WOL_MODES], wol_mode_names,
+				    info->extack, &mod);
+	if (ret < 0)
+		goto out_ops;
+	if (wol.wolopts & ~wol.supported) {
+		NL_SET_ERR_MSG_ATTR(info->extack, tb[ETHTOOL_A_WOL_MODES],
+				    "cannot enable unsupported WoL mode");
+		ret = -EINVAL;
+		goto out_ops;
+	}
+	if (tb[ETHTOOL_A_WOL_SOPASS]) {
+		if (!(wol.supported & WAKE_MAGICSECURE)) {
+			NL_SET_ERR_MSG_ATTR(info->extack,
+					    tb[ETHTOOL_A_WOL_SOPASS],
+					    "magicsecure not supported, cannot set password");
+			ret = -EINVAL;
+			goto out_ops;
+		}
+		ethnl_update_binary(wol.sopass, sizeof(wol.sopass),
+				    tb[ETHTOOL_A_WOL_SOPASS], &mod);
+	}
+
+	if (!mod)
+		goto out_ops;
+	ret = dev->ethtool_ops->set_wol(dev, &wol);
+	if (ret)
+		goto out_ops;
+	dev->wol_enabled = !!wol.wolopts;
+
+out_ops:
+	ethnl_ops_complete(dev);
+out_rtnl:
+	rtnl_unlock();
+	dev_put(dev);
+	return ret;
+}
-- 
cgit v1.2.3-59-g8ed1b


From 67bffa79231f15ba372007972563cc8aa5e48bfa Mon Sep 17 00:00:00 2001
From: Michal Kubecek <mkubecek@suse.cz>
Date: Sun, 26 Jan 2020 23:11:19 +0100
Subject: ethtool: add WOL_NTF notification

Send ETHTOOL_MSG_WOL_NTF notification whenever wake-on-lan settings of
a device are modified using ETHTOOL_MSG_WOL_SET netlink message or
ETHTOOL_SWOL ioctl request.

As notifications can be received by anyone, do not include SecureOn(tm)
password in notification messages.

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/ethtool-netlink.rst | 5 +++--
 include/uapi/linux/ethtool_netlink.h         | 1 +
 net/ethtool/ioctl.c                          | 1 +
 net/ethtool/netlink.c                        | 2 ++
 net/ethtool/wol.c                            | 4 +++-
 5 files changed, 10 insertions(+), 3 deletions(-)

diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
index f16f74bbb546..f1f868479ceb 100644
--- a/Documentation/networking/ethtool-netlink.rst
+++ b/Documentation/networking/ethtool-netlink.rst
@@ -193,7 +193,7 @@ Userspace to kernel:
 
 Kernel to userspace:
 
-  ===================================== ================================
+  ===================================== =================================
   ``ETHTOOL_MSG_STRSET_GET_REPLY``      string set contents
   ``ETHTOOL_MSG_LINKINFO_GET_REPLY``    link settings
   ``ETHTOOL_MSG_LINKINFO_NTF``          link settings notification
@@ -203,7 +203,8 @@ Kernel to userspace:
   ``ETHTOOL_MSG_DEBUG_GET_REPLY``       debugging settings
   ``ETHTOOL_MSG_DEBUG_NTF``             debugging settings notification
   ``ETHTOOL_MSG_WOL_GET_REPLY``         wake-on-lan settings
-  ===================================== ================================
+  ``ETHTOOL_MSG_WOL_NTF``               wake-on-lan settings notification
+  ===================================== =================================
 
 ``GET`` requests are sent by userspace applications to retrieve device
 information. They usually do not contain any message specific attributes.
diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
index 59de35695521..7e0b460f872c 100644
--- a/include/uapi/linux/ethtool_netlink.h
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -42,6 +42,7 @@ enum {
 	ETHTOOL_MSG_DEBUG_GET_REPLY,
 	ETHTOOL_MSG_DEBUG_NTF,
 	ETHTOOL_MSG_WOL_GET_REPLY,
+	ETHTOOL_MSG_WOL_NTF,
 
 	/* add new constants above here */
 	__ETHTOOL_MSG_KERNEL_CNT,
diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
index 46e0b31782fc..b88dd14e41c6 100644
--- a/net/ethtool/ioctl.c
+++ b/net/ethtool/ioctl.c
@@ -1316,6 +1316,7 @@ static int ethtool_set_wol(struct net_device *dev, char __user *useraddr)
 		return ret;
 
 	dev->wol_enabled = !!wol.wolopts;
+	ethtool_notify(dev, ETHTOOL_MSG_WOL_NTF, NULL);
 
 	return 0;
 }
diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index 2c375f9095fe..180c194fab07 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -526,6 +526,7 @@ ethnl_default_notify_ops[ETHTOOL_MSG_KERNEL_MAX + 1] = {
 	[ETHTOOL_MSG_LINKINFO_NTF]	= &ethnl_linkinfo_request_ops,
 	[ETHTOOL_MSG_LINKMODES_NTF]	= &ethnl_linkmodes_request_ops,
 	[ETHTOOL_MSG_DEBUG_NTF]		= &ethnl_debug_request_ops,
+	[ETHTOOL_MSG_WOL_NTF]		= &ethnl_wol_request_ops,
 };
 
 /* default notification handler */
@@ -610,6 +611,7 @@ static const ethnl_notify_handler_t ethnl_notify_handlers[] = {
 	[ETHTOOL_MSG_LINKINFO_NTF]	= ethnl_default_notify,
 	[ETHTOOL_MSG_LINKMODES_NTF]	= ethnl_default_notify,
 	[ETHTOOL_MSG_DEBUG_NTF]		= ethnl_default_notify,
+	[ETHTOOL_MSG_WOL_NTF]		= ethnl_default_notify,
 };
 
 void ethtool_notify(struct net_device *dev, unsigned int cmd, const void *data)
diff --git a/net/ethtool/wol.c b/net/ethtool/wol.c
index a2724378fac4..e1b8a65b64c4 100644
--- a/net/ethtool/wol.c
+++ b/net/ethtool/wol.c
@@ -41,7 +41,8 @@ static int wol_prepare_data(const struct ethnl_req_info *req_base,
 		return ret;
 	dev->ethtool_ops->get_wol(dev, &data->wol);
 	ethnl_ops_complete(dev);
-	data->show_sopass = data->wol.supported & WAKE_MAGICSECURE;
+	/* do not include password in notifications */
+	data->show_sopass = info && (data->wol.supported & WAKE_MAGICSECURE);
 
 	return 0;
 }
@@ -165,6 +166,7 @@ int ethnl_set_wol(struct sk_buff *skb, struct genl_info *info)
 	if (ret)
 		goto out_ops;
 	dev->wol_enabled = !!wol.wolopts;
+	ethtool_notify(dev, ETHTOOL_MSG_WOL_NTF, NULL);
 
 out_ops:
 	ethnl_ops_complete(dev);
-- 
cgit v1.2.3-59-g8ed1b


From 83d8f5e92d034a4c4a04d780107e73af31a38504 Mon Sep 17 00:00:00 2001
From: Michael Chan <michael.chan@broadcom.com>
Date: Mon, 27 Jan 2020 04:56:13 -0500
Subject: bnxt_en: Improve link up detection.

In bnxt_update_phy_setting(), ethtool_get_link_ksettings() and
bnxt_disable_an_for_lpbk(), we inconsistently use netif_carrier_ok()
to determine link.  Instead, we should use bp->link_info.link_up
which has the true link state.  The netif_carrier state may be off
during self-test and while the device is being reset and may not always
reflect the true link state.

By always using bp->link_info.link_up, the code is now more
consistent and more correct.  Some unnecessary link toggles are
now prevented with this patch.

Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c         |  2 +-
 drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 16 ++++++++--------
 2 files changed, 9 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 198c69dceeef..4b6f746c4357 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -9064,7 +9064,7 @@ static int bnxt_update_phy_setting(struct bnxt *bp)
 	/* The last close may have shutdown the link, so need to call
 	 * PHY_CFG to bring it back up.
 	 */
-	if (!netif_carrier_ok(bp->dev))
+	if (!bp->link_info.link_up)
 		update_link = true;
 
 	if (!bnxt_eee_config_ok(bp))
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index 08d56ec7b68a..6171fa8b3677 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -1462,15 +1462,15 @@ static int bnxt_get_link_ksettings(struct net_device *dev,
 		ethtool_link_ksettings_add_link_mode(lk_ksettings,
 						     advertising, Autoneg);
 		base->autoneg = AUTONEG_ENABLE;
-		if (link_info->phy_link_status == BNXT_LINK_LINK)
+		base->duplex = DUPLEX_UNKNOWN;
+		if (link_info->phy_link_status == BNXT_LINK_LINK) {
 			bnxt_fw_to_ethtool_lp_adv(link_info, lk_ksettings);
+			if (link_info->duplex & BNXT_LINK_DUPLEX_FULL)
+				base->duplex = DUPLEX_FULL;
+			else
+				base->duplex = DUPLEX_HALF;
+		}
 		ethtool_speed = bnxt_fw_to_ethtool_speed(link_info->link_speed);
-		if (!netif_carrier_ok(dev))
-			base->duplex = DUPLEX_UNKNOWN;
-		else if (link_info->duplex & BNXT_LINK_DUPLEX_FULL)
-			base->duplex = DUPLEX_FULL;
-		else
-			base->duplex = DUPLEX_HALF;
 	} else {
 		base->autoneg = AUTONEG_DISABLE;
 		ethtool_speed =
@@ -2707,7 +2707,7 @@ static int bnxt_disable_an_for_lpbk(struct bnxt *bp,
 		return rc;
 
 	fw_speed = PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_1GB;
-	if (netif_carrier_ok(bp->dev))
+	if (bp->link_info.link_up)
 		fw_speed = bp->link_info.link_speed;
 	else if (fw_advertising & BNXT_LINK_SPEED_MSK_10GB)
 		fw_speed = PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_10GB;
-- 
cgit v1.2.3-59-g8ed1b


From 43a5107dc1acbf88a8a194beb9ff4e8563a2c7a3 Mon Sep 17 00:00:00 2001
From: Michael Chan <michael.chan@broadcom.com>
Date: Mon, 27 Jan 2020 04:56:14 -0500
Subject: bnxt_en: Improve bnxt_probe_phy().

If the 2nd parameter fw_dflt is not set, we are calling bnxt_probe_phy()
after the firmware has reset.  There is no need to query the current
PHY settings from firmware as these settings may be different from
the ethtool settings that the driver will re-establish later.  So
return earlier in bnxt_probe_phy() to save one firmware call.

Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 4b6f746c4357..4d790bca124e 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -11471,6 +11471,9 @@ static int bnxt_probe_phy(struct bnxt *bp, bool fw_dflt)
 			   rc);
 		return rc;
 	}
+	if (!fw_dflt)
+		return 0;
+
 	rc = bnxt_update_link(bp, false);
 	if (rc) {
 		netdev_err(bp->dev, "Probe phy can't update link (rc: %x)\n",
@@ -11484,9 +11487,6 @@ static int bnxt_probe_phy(struct bnxt *bp, bool fw_dflt)
 	if (link_info->auto_link_speeds && !link_info->support_auto_speeds)
 		link_info->support_auto_speeds = link_info->support_speeds;
 
-	if (!fw_dflt)
-		return 0;
-
 	bnxt_init_ethtool_link_settings(bp);
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 1d86859fdf31a0d50cc82b5d0d6bfb5fe98f6c00 Mon Sep 17 00:00:00 2001
From: Michael Chan <michael.chan@broadcom.com>
Date: Mon, 27 Jan 2020 04:56:15 -0500
Subject: bnxt_en: Remove the setting of dev_port.

The dev_port is meant to distinguish the network ports belonging to
the same PCI function.  Our devices only have one network port
associated with each PCI function and so we should not set it for
correctness.

Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 4d790bca124e..6c0c926d8ca5 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -6998,7 +6998,6 @@ static int __bnxt_hwrm_func_qcaps(struct bnxt *bp)
 
 		pf->fw_fid = le16_to_cpu(resp->fid);
 		pf->port_id = le16_to_cpu(resp->port_id);
-		bp->dev->dev_port = pf->port_id;
 		memcpy(pf->mac_addr, resp->mac_address, ETH_ALEN);
 		pf->first_vf_id = le16_to_cpu(resp->first_vf_id);
 		pf->max_vfs = le16_to_cpu(resp->max_vfs);
-- 
cgit v1.2.3-59-g8ed1b


From c66c06c5e28a55920c87b386876461c139aa1878 Mon Sep 17 00:00:00 2001
From: Michael Chan <michael.chan@broadcom.com>
Date: Mon, 27 Jan 2020 04:56:16 -0500
Subject: bnxt_en: Support UDP RSS hashing on 575XX chips.

575XX (P5) chips have the same UDP RSS hashing capability as P4 chips,
so we can enable it on P5 chips.

Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 6c0c926d8ca5..fb4a65fe575d 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -10568,7 +10568,7 @@ static void bnxt_set_dflt_rss_hash_type(struct bnxt *bp)
 			   VNIC_RSS_CFG_REQ_HASH_TYPE_TCP_IPV4 |
 			   VNIC_RSS_CFG_REQ_HASH_TYPE_IPV6 |
 			   VNIC_RSS_CFG_REQ_HASH_TYPE_TCP_IPV6;
-	if (BNXT_CHIP_P4(bp) && bp->hwrm_spec_code >= 0x10501) {
+	if (BNXT_CHIP_P4_PLUS(bp) && bp->hwrm_spec_code >= 0x10501) {
 		bp->flags |= BNXT_FLAG_UDP_RSS_CAP;
 		bp->rss_hash_cfg |= VNIC_RSS_CFG_REQ_HASH_TYPE_UDP_IPV4 |
 				    VNIC_RSS_CFG_REQ_HASH_TYPE_UDP_IPV6;
-- 
cgit v1.2.3-59-g8ed1b


From f47d0e19ae99329177423db80f86a601f8cd8e3e Mon Sep 17 00:00:00 2001
From: Michael Chan <michael.chan@broadcom.com>
Date: Mon, 27 Jan 2020 04:56:17 -0500
Subject: bnxt_en: Do not accept fragments for aRFS flow steering.

In bnxt_rx_flow_steer(), if the dissected packet is a fragment, do not
proceed to create the ntuple filter and return error instead.  Otherwise
we would create a filter with 0 source and destination ports because
the dissected ports would not be available for fragments.

Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index fb4a65fe575d..fb67e6656206 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -11100,6 +11100,7 @@ static int bnxt_rx_flow_steer(struct net_device *dev, const struct sk_buff *skb,
 	struct ethhdr *eth = (struct ethhdr *)skb_mac_header(skb);
 	int rc = 0, idx, bit_id, l2_idx = 0;
 	struct hlist_head *head;
+	u32 flags;
 
 	if (!ether_addr_equal(dev->dev_addr, eth->h_dest)) {
 		struct bnxt_vnic_info *vnic = &bp->vnic_info[0];
@@ -11139,8 +11140,9 @@ static int bnxt_rx_flow_steer(struct net_device *dev, const struct sk_buff *skb,
 		rc = -EPROTONOSUPPORT;
 		goto err_free;
 	}
-	if ((fkeys->control.flags & FLOW_DIS_ENCAPSULATION) &&
-	    bp->hwrm_spec_code < 0x10601) {
+	flags = fkeys->control.flags;
+	if (((flags & FLOW_DIS_ENCAPSULATION) &&
+	     bp->hwrm_spec_code < 0x10601) || (flags & FLOW_DIS_IS_FRAGMENT)) {
 		rc = -EPROTONOSUPPORT;
 		goto err_free;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 87d67f59d6a33f741b5c6a42fa01b99ea2b14b7d Mon Sep 17 00:00:00 2001
From: Pavan Chebbi <pavan.chebbi@broadcom.com>
Date: Mon, 27 Jan 2020 04:56:18 -0500
Subject: bnxt_en: Periodically check and remove aged-out ntuple filters

Currently the only time we check and remove expired filters is
when we are inserting new filters.
Improving the aRFS expiry handling by adding code to do the above
work periodically.

Signed-off-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index fb67e6656206..7d5367277a9d 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -10040,6 +10040,13 @@ static void bnxt_timer(struct timer_list *t)
 		bnxt_queue_sp_work(bp);
 	}
 
+#ifdef CONFIG_RFS_ACCEL
+	if ((bp->flags & BNXT_FLAG_RFS) && bp->ntp_fltr_count) {
+		set_bit(BNXT_RX_NTP_FLTR_SP_EVENT, &bp->sp_event);
+		bnxt_queue_sp_work(bp);
+	}
+#endif /*CONFIG_RFS_ACCEL*/
+
 	if (bp->link_info.phy_retry) {
 		if (time_after(jiffies, bp->link_info.phy_retry_expires)) {
 			bp->link_info.phy_retry = false;
-- 
cgit v1.2.3-59-g8ed1b


From 5313845f491f22761c675f7009dd77b5ae64172d Mon Sep 17 00:00:00 2001
From: Michael Chan <michael.chan@broadcom.com>
Date: Mon, 27 Jan 2020 04:56:19 -0500
Subject: bnxt_en: Disable workaround for lost interrupts on 575XX B0 and newer
 chips.

The hardware bug has been fixed on B0 and newer chips, so disable the
workaround on these chips.

Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 4 +++-
 drivers/net/ethernet/broadcom/bnxt/bnxt.h | 2 ++
 2 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 7d5367277a9d..676f4da23c95 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -7288,6 +7288,7 @@ static int bnxt_hwrm_ver_get(struct bnxt *bp)
 		bp->hwrm_max_ext_req_len = HWRM_MAX_REQ_LEN;
 
 	bp->chip_num = le16_to_cpu(resp->chip_num);
+	bp->chip_rev = resp->chip_rev;
 	if (bp->chip_num == CHIP_NUM_58700 && !resp->chip_rev &&
 	    !resp->chip_metal)
 		bp->flags |= BNXT_FLAG_CHIP_NITRO_A0;
@@ -10057,7 +10058,8 @@ static void bnxt_timer(struct timer_list *t)
 		}
 	}
 
-	if ((bp->flags & BNXT_FLAG_CHIP_P5) && netif_carrier_ok(dev)) {
+	if ((bp->flags & BNXT_FLAG_CHIP_P5) && !bp->chip_rev &&
+	    netif_carrier_ok(dev)) {
 		set_bit(BNXT_RING_COAL_NOW_SP_EVENT, &bp->sp_event);
 		bnxt_queue_sp_work(bp);
 	}
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index f14335433a64..cda434b5fa66 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -1457,6 +1457,8 @@ struct bnxt {
 #define CHIP_NUM_58804		0xd804
 #define CHIP_NUM_58808		0xd808
 
+	u8			chip_rev;
+
 #define BNXT_CHIP_NUM_5730X(chip_num)		\
 	((chip_num) >= CHIP_NUM_57301 &&	\
 	 (chip_num) <= CHIP_NUM_57304)
-- 
cgit v1.2.3-59-g8ed1b


From d6292ade7f6f109bffe9e1065de37f688fa0107f Mon Sep 17 00:00:00 2001
From: Vasundhara Volam <vasundhara-v.volam@broadcom.com>
Date: Mon, 27 Jan 2020 04:56:20 -0500
Subject: bnxt_en: Refactor bnxt_dl_register()

Define bnxt_dl_params_register() and bnxt_dl_params_unregister()
functions and move params register/unregister code to these newly
defined functions. This patch is in preparation to register
devlink irrespective of firmware spec. version in the next patch.

Signed-off-by: Vasundhara Volam <vasundhara-v.volam@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c | 60 ++++++++++++++---------
 1 file changed, 36 insertions(+), 24 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c
index 0c3d224637b9..9253eed60f55 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c
@@ -485,6 +485,38 @@ static const struct devlink_param bnxt_dl_params[] = {
 static const struct devlink_param bnxt_dl_port_params[] = {
 };
 
+static int bnxt_dl_params_register(struct bnxt *bp)
+{
+	int rc;
+
+	rc = devlink_params_register(bp->dl, bnxt_dl_params,
+				     ARRAY_SIZE(bnxt_dl_params));
+	if (rc) {
+		netdev_warn(bp->dev, "devlink_params_register failed. rc=%d",
+			    rc);
+		return rc;
+	}
+	rc = devlink_port_params_register(&bp->dl_port, bnxt_dl_port_params,
+					  ARRAY_SIZE(bnxt_dl_port_params));
+	if (rc) {
+		netdev_err(bp->dev, "devlink_port_params_register failed");
+		devlink_params_unregister(bp->dl, bnxt_dl_params,
+					  ARRAY_SIZE(bnxt_dl_params));
+		return rc;
+	}
+	devlink_params_publish(bp->dl);
+
+	return 0;
+}
+
+static void bnxt_dl_params_unregister(struct bnxt *bp)
+{
+	devlink_params_unregister(bp->dl, bnxt_dl_params,
+				  ARRAY_SIZE(bnxt_dl_params));
+	devlink_port_params_unregister(&bp->dl_port, bnxt_dl_port_params,
+				       ARRAY_SIZE(bnxt_dl_port_params));
+}
+
 int bnxt_dl_register(struct bnxt *bp)
 {
 	struct devlink *dl;
@@ -520,40 +552,24 @@ int bnxt_dl_register(struct bnxt *bp)
 	if (!BNXT_PF(bp))
 		return 0;
 
-	rc = devlink_params_register(dl, bnxt_dl_params,
-				     ARRAY_SIZE(bnxt_dl_params));
-	if (rc) {
-		netdev_warn(bp->dev, "devlink_params_register failed. rc=%d",
-			    rc);
-		goto err_dl_unreg;
-	}
-
 	devlink_port_attrs_set(&bp->dl_port, DEVLINK_PORT_FLAVOUR_PHYSICAL,
 			       bp->pf.port_id, false, 0,
 			       bp->switch_id, sizeof(bp->switch_id));
 	rc = devlink_port_register(dl, &bp->dl_port, bp->pf.port_id);
 	if (rc) {
 		netdev_err(bp->dev, "devlink_port_register failed");
-		goto err_dl_param_unreg;
+		goto err_dl_unreg;
 	}
 	devlink_port_type_eth_set(&bp->dl_port, bp->dev);
 
-	rc = devlink_port_params_register(&bp->dl_port, bnxt_dl_port_params,
-					  ARRAY_SIZE(bnxt_dl_port_params));
-	if (rc) {
-		netdev_err(bp->dev, "devlink_port_params_register failed");
+	rc = bnxt_dl_params_register(bp);
+	if (rc)
 		goto err_dl_port_unreg;
-	}
-
-	devlink_params_publish(dl);
 
 	return 0;
 
 err_dl_port_unreg:
 	devlink_port_unregister(&bp->dl_port);
-err_dl_param_unreg:
-	devlink_params_unregister(dl, bnxt_dl_params,
-				  ARRAY_SIZE(bnxt_dl_params));
 err_dl_unreg:
 	devlink_unregister(dl);
 err_dl_free:
@@ -570,12 +586,8 @@ void bnxt_dl_unregister(struct bnxt *bp)
 		return;
 
 	if (BNXT_PF(bp)) {
-		devlink_port_params_unregister(&bp->dl_port,
-					       bnxt_dl_port_params,
-					       ARRAY_SIZE(bnxt_dl_port_params));
+		bnxt_dl_params_unregister(bp);
 		devlink_port_unregister(&bp->dl_port);
-		devlink_params_unregister(dl, bnxt_dl_params,
-					  ARRAY_SIZE(bnxt_dl_params));
 	}
 	devlink_unregister(dl);
 	devlink_free(dl);
-- 
cgit v1.2.3-59-g8ed1b


From 002870ebdabb75809d753e3c171c68c4e70a0e91 Mon Sep 17 00:00:00 2001
From: Vasundhara Volam <vasundhara-v.volam@broadcom.com>
Date: Mon, 27 Jan 2020 04:56:21 -0500
Subject: bnxt_en: Register devlink irrespective of firmware spec version

This will allow to register for devlink port and use port features.
Also register params only if firmware spec version is at least 0x10600
which will support reading/setting numbered variables in NVRAM.

Signed-off-by: Vasundhara Volam <vasundhara-v.volam@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c
index 9253eed60f55..a8cdfbdd5598 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c
@@ -489,6 +489,9 @@ static int bnxt_dl_params_register(struct bnxt *bp)
 {
 	int rc;
 
+	if (bp->hwrm_spec_code < 0x10600)
+		return 0;
+
 	rc = devlink_params_register(bp->dl, bnxt_dl_params,
 				     ARRAY_SIZE(bnxt_dl_params));
 	if (rc) {
@@ -511,6 +514,9 @@ static int bnxt_dl_params_register(struct bnxt *bp)
 
 static void bnxt_dl_params_unregister(struct bnxt *bp)
 {
+	if (bp->hwrm_spec_code < 0x10600)
+		return;
+
 	devlink_params_unregister(bp->dl, bnxt_dl_params,
 				  ARRAY_SIZE(bnxt_dl_params));
 	devlink_port_params_unregister(&bp->dl_port, bnxt_dl_port_params,
@@ -522,11 +528,6 @@ int bnxt_dl_register(struct bnxt *bp)
 	struct devlink *dl;
 	int rc;
 
-	if (bp->hwrm_spec_code < 0x10600) {
-		netdev_warn(bp->dev, "Firmware does not support NVM params");
-		return -ENOTSUPP;
-	}
-
 	if (BNXT_PF(bp))
 		dl = devlink_alloc(&bnxt_dl_ops, sizeof(struct bnxt_dl));
 	else
-- 
cgit v1.2.3-59-g8ed1b


From cda2cab0771183932d6ba73c5ac63bb63decdadf Mon Sep 17 00:00:00 2001
From: Vasundhara Volam <vasundhara-v.volam@broadcom.com>
Date: Mon, 27 Jan 2020 04:56:22 -0500
Subject: bnxt_en: Move devlink_register before registering netdev

Latest kernels get the phys_port_name via devlink, if
ndo_get_phys_port_name is not defined. To provide the phys_port_name
correctly, register devlink before registering netdev.

Also call devlink_port_type_eth_set() after registering netdev as
devlink port updates the netdev structure and notifies user.

Cc: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Vasundhara Volam <vasundhara-v.volam@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c         | 12 ++++++++----
 drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c |  1 -
 2 files changed, 8 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 676f4da23c95..857941502eaf 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -11445,9 +11445,9 @@ static void bnxt_remove_one(struct pci_dev *pdev)
 		bnxt_sriov_disable(bp);
 
 	bnxt_dl_fw_reporters_destroy(bp, true);
-	bnxt_dl_unregister(bp);
 	pci_disable_pcie_error_reporting(pdev);
 	unregister_netdev(dev);
+	bnxt_dl_unregister(bp);
 	bnxt_shutdown_tc(bp);
 	bnxt_cancel_sp_work(bp);
 	bp->sp_event = 0;
@@ -11917,11 +11917,14 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		bnxt_init_tc(bp);
 	}
 
+	bnxt_dl_register(bp);
+
 	rc = register_netdev(dev);
 	if (rc)
-		goto init_err_cleanup_tc;
+		goto init_err_cleanup;
 
-	bnxt_dl_register(bp);
+	if (BNXT_PF(bp))
+		devlink_port_type_eth_set(&bp->dl_port, bp->dev);
 	bnxt_dl_fw_reporters_create(bp);
 
 	netdev_info(dev, "%s found at mem %lx, node addr %pM\n",
@@ -11931,7 +11934,8 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	return 0;
 
-init_err_cleanup_tc:
+init_err_cleanup:
+	bnxt_dl_unregister(bp);
 	bnxt_shutdown_tc(bp);
 	bnxt_clear_int_mode(bp);
 
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c
index a8cdfbdd5598..f2d9cd6892d1 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c
@@ -561,7 +561,6 @@ int bnxt_dl_register(struct bnxt *bp)
 		netdev_err(bp->dev, "devlink_port_register failed");
 		goto err_dl_unreg;
 	}
-	devlink_port_type_eth_set(&bp->dl_port, bp->dev);
 
 	rc = bnxt_dl_params_register(bp);
 	if (rc)
-- 
cgit v1.2.3-59-g8ed1b


From 8159cbe3e0b24489f97ca0fb7df7a1710f03a26b Mon Sep 17 00:00:00 2001
From: Vasundhara Volam <vasundhara-v.volam@broadcom.com>
Date: Mon, 27 Jan 2020 04:56:23 -0500
Subject: bnxt_en: Add support to update progress of flash update

This patch adds status notification to devlink flash update
while flashing is in progress.

$ devlink dev flash pci/0000:05:00.0 file 103.pkg
Preparing to flash
Flashing done

Cc: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Vasundhara Volam <vasundhara-v.volam@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c
index f2d9cd6892d1..265a68c7fe8e 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c
@@ -21,6 +21,7 @@ bnxt_dl_flash_update(struct devlink *dl, const char *filename,
 		     const char *region, struct netlink_ext_ack *extack)
 {
 	struct bnxt *bp = bnxt_get_bp_from_dl(dl);
+	int rc;
 
 	if (region)
 		return -EOPNOTSUPP;
@@ -31,7 +32,18 @@ bnxt_dl_flash_update(struct devlink *dl, const char *filename,
 		return -EPERM;
 	}
 
-	return bnxt_flash_package_from_file(bp->dev, filename, 0);
+	devlink_flash_update_begin_notify(dl);
+	devlink_flash_update_status_notify(dl, "Preparing to flash", region, 0,
+					   0);
+	rc = bnxt_flash_package_from_file(bp->dev, filename, 0);
+	if (!rc)
+		devlink_flash_update_status_notify(dl, "Flashing done", region,
+						   0, 0);
+	else
+		devlink_flash_update_status_notify(dl, "Flashing failed",
+						   region, 0, 0);
+	devlink_flash_update_end_notify(dl);
+	return rc;
 }
 
 static int bnxt_fw_reporter_diagnose(struct devlink_health_reporter *reporter,
-- 
cgit v1.2.3-59-g8ed1b


From b014232f7f56f6db10b8540b0b97ae8c7eeef28e Mon Sep 17 00:00:00 2001
From: Vasundhara Volam <vasundhara-v.volam@broadcom.com>
Date: Mon, 27 Jan 2020 04:56:24 -0500
Subject: bnxt_en: Rename switch_id to dsn

Instead of switch_id, renaming it to dsn will be more meaningful
so that it can be used to display device serial number in follow up
patch via devlink_info command.

Signed-off-by: Vasundhara Volam <vasundhara-v.volam@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c         | 6 +++---
 drivers/net/ethernet/broadcom/bnxt/bnxt.h         | 2 +-
 drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c | 4 ++--
 3 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 857941502eaf..483935b001c8 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -11388,8 +11388,8 @@ int bnxt_get_port_parent_id(struct net_device *dev,
 	if (!BNXT_PF(bp) || !(bp->flags & BNXT_FLAG_DSN_VALID))
 		return -EOPNOTSUPP;
 
-	ppid->id_len = sizeof(bp->switch_id);
-	memcpy(ppid->id, bp->switch_id, ppid->id_len);
+	ppid->id_len = sizeof(bp->dsn);
+	memcpy(ppid->id, bp->dsn, ppid->id_len);
 
 	return 0;
 }
@@ -11870,7 +11870,7 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	if (BNXT_PF(bp)) {
 		/* Read the adapter's DSN to use as the eswitch switch_id */
-		bnxt_pcie_dsn_get(bp, bp->switch_id);
+		rc = bnxt_pcie_dsn_get(bp, bp->dsn);
 	}
 
 	/* MTU range: 60 - FW defined max */
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index cda434b5fa66..cabef0b4f5fb 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -1848,7 +1848,7 @@ struct bnxt {
 	enum devlink_eswitch_mode eswitch_mode;
 	struct bnxt_vf_rep	**vf_reps; /* array of vf-rep ptrs */
 	u16			*cfa_code_map; /* cfa_code -> vf_idx map */
-	u8			switch_id[8];
+	u8			dsn[8];
 	struct bnxt_tc_info	*tc_info;
 	struct list_head	tc_indr_block_list;
 	struct notifier_block	tc_netdev_nb;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c
index 265a68c7fe8e..35e2a22534c1 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c
@@ -566,8 +566,8 @@ int bnxt_dl_register(struct bnxt *bp)
 		return 0;
 
 	devlink_port_attrs_set(&bp->dl_port, DEVLINK_PORT_FLAVOUR_PHYSICAL,
-			       bp->pf.port_id, false, 0,
-			       bp->switch_id, sizeof(bp->switch_id));
+			       bp->pf.port_id, false, 0, bp->dsn,
+			       sizeof(bp->dsn));
 	rc = devlink_port_register(dl, &bp->dl_port, bp->pf.port_id);
 	if (rc) {
 		netdev_err(bp->dev, "devlink_port_register failed");
-- 
cgit v1.2.3-59-g8ed1b


From 41c0d917d11edf9a34461f56e14f067aadb36101 Mon Sep 17 00:00:00 2001
From: Vasundhara Volam <vasundhara-v.volam@broadcom.com>
Date: Mon, 27 Jan 2020 04:56:25 -0500
Subject: devlink: add macro for "fw.roce"

Add definition and documentation for the new generic info "fw.roce".

v2: Remove board.nvm_cfg since fw.psid is similar.

Cc: Jiri Pirko <jiri@mellanox.com>
Cc: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Vasundhara Volam <vasundhara-v.volam@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/devlink/devlink-info.rst | 6 ++++++
 include/net/devlink.h                             | 2 ++
 2 files changed, 8 insertions(+)

diff --git a/Documentation/networking/devlink/devlink-info.rst b/Documentation/networking/devlink/devlink-info.rst
index 0385f15028b1..70981dd1b981 100644
--- a/Documentation/networking/devlink/devlink-info.rst
+++ b/Documentation/networking/devlink/devlink-info.rst
@@ -92,3 +92,9 @@ fw.psid
 -------
 
 Unique identifier of the firmware parameter set.
+
+fw.roce
+-------
+
+RoCE firmware version which is responsible for handling roce
+management.
diff --git a/include/net/devlink.h b/include/net/devlink.h
index 5e46c24bb6e6..ce5cea428fdc 100644
--- a/include/net/devlink.h
+++ b/include/net/devlink.h
@@ -487,6 +487,8 @@ enum devlink_param_generic_id {
 #define DEVLINK_INFO_VERSION_GENERIC_FW_NCSI	"fw.ncsi"
 /* FW parameter set id */
 #define DEVLINK_INFO_VERSION_GENERIC_FW_PSID	"fw.psid"
+/* RoCE FW version */
+#define DEVLINK_INFO_VERSION_GENERIC_FW_ROCE	"fw.roce"
 
 struct devlink_region;
 struct devlink_info_req;
-- 
cgit v1.2.3-59-g8ed1b


From 9599e036b161243d7c62399a1b6c250573e08a43 Mon Sep 17 00:00:00 2001
From: Vasundhara Volam <vasundhara-v.volam@broadcom.com>
Date: Mon, 27 Jan 2020 04:56:26 -0500
Subject: bnxt_en: Add support for devlink info command

Display the following information via devlink info command:
  - Driver name
  - Board id
  - Broad revision
  - Board Serial number
  - Board FW version
  - FW parameter set version
  - FW App version
  - FW management version
  - FW RoCE version

  Standard output example:
  $ devlink dev info pci/0000:3b:00.0
  pci/0000:3b:00.0:
  driver bnxt_en
  serial_number 00-10-18-FF-FE-AD-05-00
  versions:
      fixed:
        asic.id D802
        asic.rev 1
      running:
        fw 216.1.124.0
        fw.psid 0.0.0
        fw.app 216.1.122.0
        fw.mgmt 864.0.32.0
        fw.roce 216.1.15.0

[ This version has incorporated changes suggested by Jakub Kicinski to
  use generic devlink version tags. ]

v2: Use fw.psid

Cc: Jiri Pirko <jiri@mellanox.com>
Cc: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Vasundhara Volam <vasundhara-v.volam@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c | 134 ++++++++++++++++++++++
 drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.h |   4 +
 2 files changed, 138 insertions(+)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c
index 35e2a22534c1..eec0168330b7 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c
@@ -284,11 +284,15 @@ void bnxt_dl_health_recovery_done(struct bnxt *bp)
 		devlink_health_reporter_recovery_done(hlth->fw_reset_reporter);
 }
 
+static int bnxt_dl_info_get(struct devlink *dl, struct devlink_info_req *req,
+			    struct netlink_ext_ack *extack);
+
 static const struct devlink_ops bnxt_dl_ops = {
 #ifdef CONFIG_BNXT_SRIOV
 	.eswitch_mode_set = bnxt_dl_eswitch_mode_set,
 	.eswitch_mode_get = bnxt_dl_eswitch_mode_get,
 #endif /* CONFIG_BNXT_SRIOV */
+	.info_get	  = bnxt_dl_info_get,
 	.flash_update	  = bnxt_dl_flash_update,
 };
 
@@ -355,6 +359,136 @@ static void bnxt_copy_from_nvm_data(union devlink_param_value *dst,
 		dst->vu8 = (u8)val32;
 }
 
+static int bnxt_hwrm_get_nvm_cfg_ver(struct bnxt *bp,
+				     union devlink_param_value *nvm_cfg_ver)
+{
+	struct hwrm_nvm_get_variable_input req = {0};
+	union bnxt_nvm_data *data;
+	dma_addr_t data_dma_addr;
+	int rc;
+
+	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_NVM_GET_VARIABLE, -1, -1);
+	data = dma_alloc_coherent(&bp->pdev->dev, sizeof(*data),
+				  &data_dma_addr, GFP_KERNEL);
+	if (!data)
+		return -ENOMEM;
+
+	req.dest_data_addr = cpu_to_le64(data_dma_addr);
+	req.data_len = cpu_to_le16(BNXT_NVM_CFG_VER_BITS);
+	req.option_num = cpu_to_le16(NVM_OFF_NVM_CFG_VER);
+
+	rc = hwrm_send_message_silent(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	if (!rc)
+		bnxt_copy_from_nvm_data(nvm_cfg_ver, data,
+					BNXT_NVM_CFG_VER_BITS,
+					BNXT_NVM_CFG_VER_BYTES);
+
+	dma_free_coherent(&bp->pdev->dev, sizeof(*data), data, data_dma_addr);
+	return rc;
+}
+
+static int bnxt_dl_info_get(struct devlink *dl, struct devlink_info_req *req,
+			    struct netlink_ext_ack *extack)
+{
+	struct bnxt *bp = bnxt_get_bp_from_dl(dl);
+	union devlink_param_value nvm_cfg_ver;
+	struct hwrm_ver_get_output *ver_resp;
+	char mgmt_ver[FW_VER_STR_LEN];
+	char roce_ver[FW_VER_STR_LEN];
+	char fw_ver[FW_VER_STR_LEN];
+	char buf[32];
+	int rc;
+
+	rc = devlink_info_driver_name_put(req, DRV_MODULE_NAME);
+	if (rc)
+		return rc;
+
+	sprintf(buf, "%X", bp->chip_num);
+	rc = devlink_info_version_fixed_put(req,
+			DEVLINK_INFO_VERSION_GENERIC_ASIC_ID, buf);
+	if (rc)
+		return rc;
+
+	ver_resp = &bp->ver_resp;
+	sprintf(buf, "%X", ver_resp->chip_rev);
+	rc = devlink_info_version_fixed_put(req,
+			DEVLINK_INFO_VERSION_GENERIC_ASIC_REV, buf);
+	if (rc)
+		return rc;
+
+	if (BNXT_PF(bp)) {
+		sprintf(buf, "%02X-%02X-%02X-%02X-%02X-%02X-%02X-%02X",
+			bp->dsn[7], bp->dsn[6], bp->dsn[5], bp->dsn[4],
+			bp->dsn[3], bp->dsn[2], bp->dsn[1], bp->dsn[0]);
+		rc = devlink_info_serial_number_put(req, buf);
+		if (rc)
+			return rc;
+	}
+
+	if (strlen(ver_resp->active_pkg_name)) {
+		rc =
+		    devlink_info_version_running_put(req,
+					DEVLINK_INFO_VERSION_GENERIC_FW,
+					ver_resp->active_pkg_name);
+		if (rc)
+			return rc;
+	}
+
+	if (BNXT_PF(bp) && !bnxt_hwrm_get_nvm_cfg_ver(bp, &nvm_cfg_ver)) {
+		u32 ver = nvm_cfg_ver.vu32;
+
+		sprintf(buf, "%X.%X.%X", (ver >> 16) & 0xF, (ver >> 8) & 0xF,
+			ver & 0xF);
+		rc = devlink_info_version_running_put(req,
+				DEVLINK_INFO_VERSION_GENERIC_FW_PSID, buf);
+		if (rc)
+			return rc;
+	}
+
+	if (ver_resp->flags & VER_GET_RESP_FLAGS_EXT_VER_AVAIL) {
+		snprintf(fw_ver, FW_VER_STR_LEN, "%d.%d.%d.%d",
+			 ver_resp->hwrm_fw_major, ver_resp->hwrm_fw_minor,
+			 ver_resp->hwrm_fw_build, ver_resp->hwrm_fw_patch);
+
+		snprintf(mgmt_ver, FW_VER_STR_LEN, "%d.%d.%d.%d",
+			 ver_resp->mgmt_fw_major, ver_resp->mgmt_fw_minor,
+			 ver_resp->mgmt_fw_build, ver_resp->mgmt_fw_patch);
+
+		snprintf(roce_ver, FW_VER_STR_LEN, "%d.%d.%d.%d",
+			 ver_resp->roce_fw_major, ver_resp->roce_fw_minor,
+			 ver_resp->roce_fw_build, ver_resp->roce_fw_patch);
+	} else {
+		snprintf(fw_ver, FW_VER_STR_LEN, "%d.%d.%d.%d",
+			 ver_resp->hwrm_fw_maj_8b, ver_resp->hwrm_fw_min_8b,
+			 ver_resp->hwrm_fw_bld_8b, ver_resp->hwrm_fw_rsvd_8b);
+
+		snprintf(mgmt_ver, FW_VER_STR_LEN, "%d.%d.%d.%d",
+			 ver_resp->mgmt_fw_maj_8b, ver_resp->mgmt_fw_min_8b,
+			 ver_resp->mgmt_fw_bld_8b, ver_resp->mgmt_fw_rsvd_8b);
+
+		snprintf(roce_ver, FW_VER_STR_LEN, "%d.%d.%d.%d",
+			 ver_resp->roce_fw_maj_8b, ver_resp->roce_fw_min_8b,
+			 ver_resp->roce_fw_bld_8b, ver_resp->roce_fw_rsvd_8b);
+	}
+	rc = devlink_info_version_running_put(req,
+			DEVLINK_INFO_VERSION_GENERIC_FW_APP, fw_ver);
+	if (rc)
+		return rc;
+
+	if (!(bp->flags & BNXT_FLAG_CHIP_P5)) {
+		rc = devlink_info_version_running_put(req,
+			DEVLINK_INFO_VERSION_GENERIC_FW_MGMT, mgmt_ver);
+		if (rc)
+			return rc;
+
+		rc = devlink_info_version_running_put(req,
+			DEVLINK_INFO_VERSION_GENERIC_FW_ROCE, roce_ver);
+		if (rc)
+			return rc;
+	}
+	return 0;
+}
+
 static int bnxt_hwrm_nvm_req(struct bnxt *bp, u32 param_id, void *msg,
 			     int msg_len, union devlink_param_value *val)
 {
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.h
index 08aaa4441c78..95f893f2a74d 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.h
@@ -38,6 +38,10 @@ static inline void bnxt_link_bp_to_dl(struct bnxt *bp, struct devlink *dl)
 #define NVM_OFF_IGNORE_ARI		164
 #define NVM_OFF_DIS_GRE_VER_CHECK	171
 #define NVM_OFF_ENABLE_SRIOV		401
+#define NVM_OFF_NVM_CFG_VER		602
+
+#define BNXT_NVM_CFG_VER_BITS		24
+#define BNXT_NVM_CFG_VER_BYTES		4
 
 #define BNXT_MSIX_VEC_MAX	1280
 #define BNXT_MSIX_VEC_MIN_MAX	128
-- 
cgit v1.2.3-59-g8ed1b


From 92439915e0e37495389bdaec5bda877293dd68b4 Mon Sep 17 00:00:00 2001
From: Vasundhara Volam <vasundhara-v.volam@broadcom.com>
Date: Mon, 27 Jan 2020 04:56:27 -0500
Subject: devlink: document devlink info versions reported by bnxt_en driver

Add the set of info versions reported by bnxt_en driver, including
a description of what the version represents, and what modes (fixed,
running, stored) it reports.

v2: Use fw.psid.

Cc: Jiri Pirko <jiri@mellanox.com>
Cc: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Vasundhara Volam <vasundhara-v.volam@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/devlink/bnxt.rst | 33 +++++++++++++++++++++++++++++++
 1 file changed, 33 insertions(+)

diff --git a/Documentation/networking/devlink/bnxt.rst b/Documentation/networking/devlink/bnxt.rst
index 79e746d22a53..82ef9ec46707 100644
--- a/Documentation/networking/devlink/bnxt.rst
+++ b/Documentation/networking/devlink/bnxt.rst
@@ -39,3 +39,36 @@ parameters.
      - Generic Routing Encapsulation (GRE) version check will be enabled in
        the device. If disabled, the device will skip the version check for
        incoming packets.
+
+Info versions
+=============
+
+The ``bnxt_en`` driver reports the following versions
+
+.. list-table:: devlink info versions implemented
+      :widths: 5 5 90
+
+   * - Name
+     - Type
+     - Description
+   * - ``asic.id``
+     - fixed
+     - ASIC design identifier
+   * - ``asic.rev``
+     - fixed
+     - ASIC design revision
+   * - ``fw.psid``
+     - stored, running
+     - Firmware parameter set version of the board
+   * - ``fw``
+     - stored, running
+     - Overall board firmware version
+   * - ``fw.app``
+     - stored, running
+     - Data path firmware version
+   * - ``fw.mgmt``
+     - stored, running
+     - Management firmware version
+   * - ``fw.roce``
+     - stored, running
+     - RoCE management firmware version
-- 
cgit v1.2.3-59-g8ed1b


From 90c914d2a39367b18b25b67b0a69335df8b92d56 Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Mon, 27 Jan 2020 11:13:27 +0000
Subject: sfc: rename mcdi filtering functions/structs

Minor style fixes included due to name lengths changing.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/ef10.c | 558 ++++++++++++++++++++--------------------
 1 file changed, 282 insertions(+), 276 deletions(-)

diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c
index 4997f61de3d6..648a86ea25e0 100644
--- a/drivers/net/ethernet/sfc/ef10.c
+++ b/drivers/net/ethernet/sfc/ef10.c
@@ -44,7 +44,7 @@ enum {
  * Huntington has a single 8K filter table shared between all filter
  * types and both ports.
  */
-#define HUNT_FILTER_TBL_ROWS 8192
+#define EFX_MCDI_FILTER_TBL_ROWS 8192
 
 #define EFX_EF10_FILTER_ID_INVALID 0xffff
 
@@ -57,7 +57,7 @@ struct efx_ef10_vlan {
 	u16 vid;
 };
 
-enum efx_ef10_default_filters {
+enum efx_mcdi_filter_default_filters {
 	EFX_EF10_BCAST,
 	EFX_EF10_UCDEF,
 	EFX_EF10_MCDEF,
@@ -78,7 +78,7 @@ enum efx_ef10_default_filters {
 };
 
 /* Per-VLAN filters information */
-struct efx_ef10_filter_vlan {
+struct efx_mcdi_filter_vlan {
 	struct list_head list;
 	u16 vid;
 	u16 uc[EFX_EF10_FILTER_DEV_UC_MAX];
@@ -86,11 +86,11 @@ struct efx_ef10_filter_vlan {
 	u16 default_filters[EFX_EF10_NUM_DEFAULT_FILTERS];
 };
 
-struct efx_ef10_dev_addr {
+struct efx_mcdi_dev_addr {
 	u8 addr[ETH_ALEN];
 };
 
-struct efx_ef10_filter_table {
+struct efx_mcdi_filter_table {
 /* The MCDI match masks supported by this fw & hw, in order of priority */
 	u32 rx_match_mcdi_flags[
 		MC_CMD_GET_PARSER_DISP_INFO_OUT_SUPPORTED_MATCHES_MAXNUM * 2];
@@ -106,8 +106,8 @@ struct efx_ef10_filter_table {
 		u64 handle;		/* firmware handle */
 	} *entry;
 /* Shadow of net_device address lists, guarded by mac_lock */
-	struct efx_ef10_dev_addr dev_uc_list[EFX_EF10_FILTER_DEV_UC_MAX];
-	struct efx_ef10_dev_addr dev_mc_list[EFX_EF10_FILTER_DEV_MC_MAX];
+	struct efx_mcdi_dev_addr dev_uc_list[EFX_EF10_FILTER_DEV_UC_MAX];
+	struct efx_mcdi_dev_addr dev_mc_list[EFX_EF10_FILTER_DEV_MC_MAX];
 	int dev_uc_count;
 	int dev_mc_count;
 	bool uc_promisc;
@@ -122,28 +122,28 @@ struct efx_ef10_filter_table {
 /* An arbitrary search limit for the software hash table */
 #define EFX_EF10_FILTER_SEARCH_LIMIT 200
 
-static void efx_ef10_rx_free_indir_table(struct efx_nic *efx);
-static void efx_ef10_filter_table_remove(struct efx_nic *efx);
-static int efx_ef10_filter_add_vlan(struct efx_nic *efx, u16 vid);
-static void efx_ef10_filter_del_vlan_internal(struct efx_nic *efx,
-					      struct efx_ef10_filter_vlan *vlan);
-static void efx_ef10_filter_del_vlan(struct efx_nic *efx, u16 vid);
+static void efx_mcdi_rx_free_indir_table(struct efx_nic *efx);
+static void efx_mcdi_filter_table_remove(struct efx_nic *efx);
+static int efx_mcdi_filter_add_vlan(struct efx_nic *efx, u16 vid);
+static void efx_mcdi_filter_del_vlan_internal(struct efx_nic *efx,
+					      struct efx_mcdi_filter_vlan *vlan);
+static void efx_mcdi_filter_del_vlan(struct efx_nic *efx, u16 vid);
 static int efx_ef10_set_udp_tnl_ports(struct efx_nic *efx, bool unloading);
 
 static u32 efx_ef10_filter_get_unsafe_id(u32 filter_id)
 {
 	WARN_ON_ONCE(filter_id == EFX_EF10_FILTER_ID_INVALID);
-	return filter_id & (HUNT_FILTER_TBL_ROWS - 1);
+	return filter_id & (EFX_MCDI_FILTER_TBL_ROWS - 1);
 }
 
 static unsigned int efx_ef10_filter_get_unsafe_pri(u32 filter_id)
 {
-	return filter_id / (HUNT_FILTER_TBL_ROWS * 2);
+	return filter_id / (EFX_MCDI_FILTER_TBL_ROWS * 2);
 }
 
 static u32 efx_ef10_make_filter_id(unsigned int pri, u16 idx)
 {
-	return pri * HUNT_FILTER_TBL_ROWS * 2 + idx;
+	return pri * EFX_MCDI_FILTER_TBL_ROWS * 2 + idx;
 }
 
 static int efx_ef10_get_warm_boot_count(struct efx_nic *efx)
@@ -546,7 +546,7 @@ static int efx_ef10_add_vlan(struct efx_nic *efx, u16 vid)
 	if (efx->filter_state) {
 		mutex_lock(&efx->mac_lock);
 		down_write(&efx->filter_sem);
-		rc = efx_ef10_filter_add_vlan(efx, vlan->vid);
+		rc = efx_mcdi_filter_add_vlan(efx, vlan->vid);
 		up_write(&efx->filter_sem);
 		mutex_unlock(&efx->mac_lock);
 		if (rc)
@@ -575,7 +575,7 @@ static void efx_ef10_del_vlan_internal(struct efx_nic *efx,
 
 	if (efx->filter_state) {
 		down_write(&efx->filter_sem);
-		efx_ef10_filter_del_vlan(efx, vlan->vid);
+		efx_mcdi_filter_del_vlan(efx, vlan->vid);
 		up_write(&efx->filter_sem);
 	}
 
@@ -1038,7 +1038,7 @@ static void efx_ef10_remove(struct efx_nic *efx)
 
 	efx_mcdi_mon_remove(efx);
 
-	efx_ef10_rx_free_indir_table(efx);
+	efx_mcdi_rx_free_indir_table(efx);
 
 	if (nic_data->wc_membase)
 		iounmap(nic_data->wc_membase);
@@ -1426,7 +1426,7 @@ static int efx_ef10_init_nic(struct efx_nic *efx)
 	return 0;
 }
 
-static void efx_ef10_reset_mc_allocations(struct efx_nic *efx)
+static void efx_ef10_table_reset_mc_allocations(struct efx_nic *efx)
 {
 	struct efx_ef10_nic_data *nic_data = efx->nic_data;
 #ifdef CONFIG_SFC_SRIOV
@@ -1507,7 +1507,7 @@ static int efx_ef10_reset(struct efx_nic *efx, enum reset_type reset_type)
 	 */
 	if ((reset_type == RESET_TYPE_ALL ||
 	     reset_type == RESET_TYPE_MCDI_TIMEOUT) && !rc)
-		efx_ef10_reset_mc_allocations(efx);
+		efx_ef10_table_reset_mc_allocations(efx);
 	return rc;
 }
 
@@ -2123,7 +2123,7 @@ static void efx_ef10_mcdi_reboot_detected(struct efx_nic *efx)
 	struct efx_ef10_nic_data *nic_data = efx->nic_data;
 
 	/* All our allocations have been reset */
-	efx_ef10_reset_mc_allocations(efx);
+	efx_ef10_table_reset_mc_allocations(efx);
 
 	/* The datapath firmware might have been changed */
 	nic_data->must_check_datapath_caps = true;
@@ -2512,7 +2512,8 @@ static void efx_ef10_tx_write(struct efx_tx_queue *tx_queue)
 					 RSS_MODE_HASH_ADDRS << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_UDP_IPV6_RSS_MODE_LBN |\
 					 RSS_MODE_HASH_ADDRS << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_OTHER_IPV6_RSS_MODE_LBN)
 
-static int efx_ef10_get_rss_flags(struct efx_nic *efx, u32 context, u32 *flags)
+static int efx_mcdi_get_rss_context_flags(struct efx_nic *efx, u32 context,
+					  u32 *flags)
 {
 	/* Firmware had a bug (sfc bug 61952) where it would not actually
 	 * fill in the flags field in the response to MC_CMD_RSS_CONTEXT_GET_FLAGS.
@@ -2558,15 +2559,15 @@ static int efx_ef10_get_rss_flags(struct efx_nic *efx, u32 context, u32 *flags)
  * Defaults are 4-tuple for TCP and 2-tuple for UDP and other-IP, so we
  * just need to set the UDP ports flags (for both IP versions).
  */
-static void efx_ef10_set_rss_flags(struct efx_nic *efx,
-				   struct efx_rss_context *ctx)
+static void efx_mcdi_set_rss_context_flags(struct efx_nic *efx,
+					   struct efx_rss_context *ctx)
 {
 	MCDI_DECLARE_BUF(inbuf, MC_CMD_RSS_CONTEXT_SET_FLAGS_IN_LEN);
 	u32 flags;
 
 	BUILD_BUG_ON(MC_CMD_RSS_CONTEXT_SET_FLAGS_OUT_LEN != 0);
 
-	if (efx_ef10_get_rss_flags(efx, ctx->context_id, &flags) != 0)
+	if (efx_mcdi_get_rss_context_flags(efx, ctx->context_id, &flags) != 0)
 		return;
 	MCDI_SET_DWORD(inbuf, RSS_CONTEXT_SET_FLAGS_IN_RSS_CONTEXT_ID,
 		       ctx->context_id);
@@ -2579,9 +2580,9 @@ static void efx_ef10_set_rss_flags(struct efx_nic *efx,
 		ctx->rx_hash_udp_4tuple = true;
 }
 
-static int efx_ef10_alloc_rss_context(struct efx_nic *efx, bool exclusive,
-				      struct efx_rss_context *ctx,
-				      unsigned *context_size)
+static int efx_mcdi_filter_alloc_rss_context(struct efx_nic *efx, bool exclusive,
+					     struct efx_rss_context *ctx,
+					     unsigned *context_size)
 {
 	MCDI_DECLARE_BUF(inbuf, MC_CMD_RSS_CONTEXT_ALLOC_IN_LEN);
 	MCDI_DECLARE_BUF(outbuf, MC_CMD_RSS_CONTEXT_ALLOC_OUT_LEN);
@@ -2627,12 +2628,12 @@ static int efx_ef10_alloc_rss_context(struct efx_nic *efx, bool exclusive,
 
 	if (nic_data->datapath_caps &
 	    1 << MC_CMD_GET_CAPABILITIES_OUT_ADDITIONAL_RSS_MODES_LBN)
-		efx_ef10_set_rss_flags(efx, ctx);
+		efx_mcdi_set_rss_context_flags(efx, ctx);
 
 	return 0;
 }
 
-static int efx_ef10_free_rss_context(struct efx_nic *efx, u32 context)
+static int efx_mcdi_filter_free_rss_context(struct efx_nic *efx, u32 context)
 {
 	MCDI_DECLARE_BUF(inbuf, MC_CMD_RSS_CONTEXT_FREE_IN_LEN);
 
@@ -2642,8 +2643,8 @@ static int efx_ef10_free_rss_context(struct efx_nic *efx, u32 context)
 			    NULL, 0, NULL);
 }
 
-static int efx_ef10_populate_rss_table(struct efx_nic *efx, u32 context,
-				       const u32 *rx_indir_table, const u8 *key)
+static int efx_mcdi_filter_populate_rss_table(struct efx_nic *efx, u32 context,
+					      const u32 *rx_indir_table, const u8 *key)
 {
 	MCDI_DECLARE_BUF(tablebuf, MC_CMD_RSS_CONTEXT_SET_TABLE_IN_LEN);
 	MCDI_DECLARE_BUF(keybuf, MC_CMD_RSS_CONTEXT_SET_KEY_IN_LEN);
@@ -2680,23 +2681,25 @@ static int efx_ef10_populate_rss_table(struct efx_nic *efx, u32 context,
 			    sizeof(keybuf), NULL, 0, NULL);
 }
 
-static void efx_ef10_rx_free_indir_table(struct efx_nic *efx)
+static void efx_mcdi_rx_free_indir_table(struct efx_nic *efx)
 {
 	int rc;
 
 	if (efx->rss_context.context_id != EFX_MCDI_RSS_CONTEXT_INVALID) {
-		rc = efx_ef10_free_rss_context(efx, efx->rss_context.context_id);
+		rc = efx_mcdi_filter_free_rss_context(efx,
+						   efx->rss_context.context_id);
 		WARN_ON(rc != 0);
 	}
 	efx->rss_context.context_id = EFX_MCDI_RSS_CONTEXT_INVALID;
 }
 
-static int efx_ef10_rx_push_shared_rss_config(struct efx_nic *efx,
-					      unsigned *context_size)
+static int efx_mcdi_filter_rx_push_shared_rss_config(struct efx_nic *efx,
+						     unsigned *context_size)
 {
 	struct efx_ef10_nic_data *nic_data = efx->nic_data;
-	int rc = efx_ef10_alloc_rss_context(efx, false, &efx->rss_context,
-					    context_size);
+	int rc = efx_mcdi_filter_alloc_rss_context(efx, false,
+						   &efx->rss_context,
+						   context_size);
 
 	if (rc != 0)
 		return rc;
@@ -2706,9 +2709,9 @@ static int efx_ef10_rx_push_shared_rss_config(struct efx_nic *efx,
 	return 0;
 }
 
-static int efx_ef10_rx_push_exclusive_rss_config(struct efx_nic *efx,
-						 const u32 *rx_indir_table,
-						 const u8 *key)
+static int efx_mcdi_filter_rx_push_exclusive_rss_config(struct efx_nic *efx,
+						        const u32 *rx_indir_table,
+						        const u8 *key)
 {
 	u32 old_rx_rss_context = efx->rss_context.context_id;
 	struct efx_ef10_nic_data *nic_data = efx->nic_data;
@@ -2716,22 +2719,24 @@ static int efx_ef10_rx_push_exclusive_rss_config(struct efx_nic *efx,
 
 	if (efx->rss_context.context_id == EFX_MCDI_RSS_CONTEXT_INVALID ||
 	    !nic_data->rx_rss_context_exclusive) {
-		rc = efx_ef10_alloc_rss_context(efx, true, &efx->rss_context,
-						NULL);
+		rc = efx_mcdi_filter_alloc_rss_context(efx, true,
+						       &efx->rss_context,
+						       NULL);
 		if (rc == -EOPNOTSUPP)
 			return rc;
 		else if (rc != 0)
 			goto fail1;
 	}
 
-	rc = efx_ef10_populate_rss_table(efx, efx->rss_context.context_id,
-					 rx_indir_table, key);
+	rc = efx_mcdi_filter_populate_rss_table(efx,
+						efx->rss_context.context_id,
+						rx_indir_table, key);
 	if (rc != 0)
 		goto fail2;
 
 	if (efx->rss_context.context_id != old_rx_rss_context &&
 	    old_rx_rss_context != EFX_MCDI_RSS_CONTEXT_INVALID)
-		WARN_ON(efx_ef10_free_rss_context(efx, old_rx_rss_context) != 0);
+		WARN_ON(efx_mcdi_filter_free_rss_context(efx, old_rx_rss_context) != 0);
 	nic_data->rx_rss_context_exclusive = true;
 	if (rx_indir_table != efx->rss_context.rx_indir_table)
 		memcpy(efx->rss_context.rx_indir_table, rx_indir_table,
@@ -2744,7 +2749,7 @@ static int efx_ef10_rx_push_exclusive_rss_config(struct efx_nic *efx,
 
 fail2:
 	if (old_rx_rss_context != efx->rss_context.context_id) {
-		WARN_ON(efx_ef10_free_rss_context(efx, efx->rss_context.context_id) != 0);
+		WARN_ON(efx_mcdi_filter_free_rss_context(efx, efx->rss_context.context_id) != 0);
 		efx->rss_context.context_id = old_rx_rss_context;
 	}
 fail1:
@@ -2752,7 +2757,7 @@ fail1:
 	return rc;
 }
 
-static int efx_ef10_rx_push_rss_context_config(struct efx_nic *efx,
+static int efx_mcdi_rx_push_rss_context_config(struct efx_nic *efx,
 					       struct efx_rss_context *ctx,
 					       const u32 *rx_indir_table,
 					       const u8 *key)
@@ -2762,16 +2767,16 @@ static int efx_ef10_rx_push_rss_context_config(struct efx_nic *efx,
 	WARN_ON(!mutex_is_locked(&efx->rss_lock));
 
 	if (ctx->context_id == EFX_MCDI_RSS_CONTEXT_INVALID) {
-		rc = efx_ef10_alloc_rss_context(efx, true, ctx, NULL);
+		rc = efx_mcdi_filter_alloc_rss_context(efx, true, ctx, NULL);
 		if (rc)
 			return rc;
 	}
 
 	if (!rx_indir_table) /* Delete this context */
-		return efx_ef10_free_rss_context(efx, ctx->context_id);
+		return efx_mcdi_filter_free_rss_context(efx, ctx->context_id);
 
-	rc = efx_ef10_populate_rss_table(efx, ctx->context_id,
-					 rx_indir_table, key);
+	rc = efx_mcdi_filter_populate_rss_table(efx, ctx->context_id,
+						rx_indir_table, key);
 	if (rc)
 		return rc;
 
@@ -2782,7 +2787,7 @@ static int efx_ef10_rx_push_rss_context_config(struct efx_nic *efx,
 	return 0;
 }
 
-static int efx_ef10_rx_pull_rss_context_config(struct efx_nic *efx,
+static int efx_mcdi_rx_pull_rss_context_config(struct efx_nic *efx,
 					       struct efx_rss_context *ctx)
 {
 	MCDI_DECLARE_BUF(inbuf, MC_CMD_RSS_CONTEXT_GET_TABLE_IN_LEN);
@@ -2834,17 +2839,17 @@ static int efx_ef10_rx_pull_rss_context_config(struct efx_nic *efx,
 	return 0;
 }
 
-static int efx_ef10_rx_pull_rss_config(struct efx_nic *efx)
+static int efx_mcdi_rx_pull_rss_config(struct efx_nic *efx)
 {
 	int rc;
 
 	mutex_lock(&efx->rss_lock);
-	rc = efx_ef10_rx_pull_rss_context_config(efx, &efx->rss_context);
+	rc = efx_mcdi_rx_pull_rss_context_config(efx, &efx->rss_context);
 	mutex_unlock(&efx->rss_lock);
 	return rc;
 }
 
-static void efx_ef10_rx_restore_rss_contexts(struct efx_nic *efx)
+static void efx_mcdi_rx_restore_rss_contexts(struct efx_nic *efx)
 {
 	struct efx_ef10_nic_data *nic_data = efx->nic_data;
 	struct efx_rss_context *ctx;
@@ -2859,7 +2864,7 @@ static void efx_ef10_rx_restore_rss_contexts(struct efx_nic *efx)
 		/* previous NIC RSS context is gone */
 		ctx->context_id = EFX_MCDI_RSS_CONTEXT_INVALID;
 		/* so try to allocate a new one */
-		rc = efx_ef10_rx_push_rss_context_config(efx, ctx,
+		rc = efx_mcdi_rx_push_rss_context_config(efx, ctx,
 							 ctx->rx_indir_table,
 							 ctx->rx_hash_key);
 		if (rc)
@@ -2871,7 +2876,7 @@ static void efx_ef10_rx_restore_rss_contexts(struct efx_nic *efx)
 	nic_data->must_restore_rss_contexts = false;
 }
 
-static int efx_ef10_pf_rx_push_rss_config(struct efx_nic *efx, bool user,
+static int efx_mcdi_pf_rx_push_rss_config(struct efx_nic *efx, bool user,
 					  const u32 *rx_indir_table,
 					  const u8 *key)
 {
@@ -2883,7 +2888,7 @@ static int efx_ef10_pf_rx_push_rss_config(struct efx_nic *efx, bool user,
 	if (!key)
 		key = efx->rss_context.rx_hash_key;
 
-	rc = efx_ef10_rx_push_exclusive_rss_config(efx, rx_indir_table, key);
+	rc = efx_mcdi_filter_rx_push_exclusive_rss_config(efx, rx_indir_table, key);
 
 	if (rc == -ENOBUFS && !user) {
 		unsigned context_size;
@@ -2896,7 +2901,8 @@ static int efx_ef10_pf_rx_push_rss_config(struct efx_nic *efx, bool user,
 			mismatch = rx_indir_table[i] !=
 				ethtool_rxfh_indir_default(i, efx->rss_spread);
 
-		rc = efx_ef10_rx_push_shared_rss_config(efx, &context_size);
+		rc = efx_mcdi_filter_rx_push_shared_rss_config(efx,
+							       &context_size);
 		if (rc == 0) {
 			if (context_size != efx->rss_spread)
 				netif_warn(efx, probe, efx->net_dev,
@@ -2920,7 +2926,7 @@ static int efx_ef10_pf_rx_push_rss_config(struct efx_nic *efx, bool user,
 	return rc;
 }
 
-static int efx_ef10_vf_rx_push_rss_config(struct efx_nic *efx, bool user,
+static int efx_mcdi_vf_rx_push_rss_config(struct efx_nic *efx, bool user,
 					  const u32 *rx_indir_table
 					  __attribute__ ((unused)),
 					  const u8 *key
@@ -2930,7 +2936,7 @@ static int efx_ef10_vf_rx_push_rss_config(struct efx_nic *efx, bool user,
 		return -EOPNOTSUPP;
 	if (efx->rss_context.context_id != EFX_MCDI_RSS_CONTEXT_INVALID)
 		return 0;
-	return efx_ef10_rx_push_shared_rss_config(efx, NULL);
+	return efx_mcdi_filter_rx_push_shared_rss_config(efx, NULL);
 }
 
 /* This creates an entry in the RX descriptor queue */
@@ -3710,7 +3716,7 @@ static bool efx_ef10_filter_is_exclusive(const struct efx_filter_spec *spec)
 }
 
 static struct efx_filter_spec *
-efx_ef10_filter_entry_spec(const struct efx_ef10_filter_table *table,
+efx_ef10_filter_entry_spec(const struct efx_mcdi_filter_table *table,
 			   unsigned int filter_idx)
 {
 	return (struct efx_filter_spec *)(table->entry[filter_idx].spec &
@@ -3718,14 +3724,14 @@ efx_ef10_filter_entry_spec(const struct efx_ef10_filter_table *table,
 }
 
 static unsigned int
-efx_ef10_filter_entry_flags(const struct efx_ef10_filter_table *table,
+efx_ef10_filter_entry_flags(const struct efx_mcdi_filter_table *table,
 			   unsigned int filter_idx)
 {
 	return table->entry[filter_idx].spec & EFX_EF10_FILTER_FLAGS;
 }
 
 static void
-efx_ef10_filter_set_entry(struct efx_ef10_filter_table *table,
+efx_ef10_filter_set_entry(struct efx_mcdi_filter_table *table,
 			  unsigned int filter_idx,
 			  const struct efx_filter_spec *spec,
 			  unsigned int flags)
@@ -3831,7 +3837,7 @@ efx_ef10_filter_push_prep_set_match_fields(struct efx_nic *efx,
 		       match_fields);
 }
 
-static void efx_ef10_filter_push_prep(struct efx_nic *efx,
+static void efx_mcdi_filter_push_prep(struct efx_nic *efx,
 				      const struct efx_filter_spec *spec,
 				      efx_dword_t *inbuf, u64 handle,
 				      struct efx_rss_context *ctx,
@@ -3880,7 +3886,7 @@ static void efx_ef10_filter_push_prep(struct efx_nic *efx,
 		MCDI_SET_DWORD(inbuf, FILTER_OP_IN_RX_CONTEXT, ctx->context_id);
 }
 
-static int efx_ef10_filter_push(struct efx_nic *efx,
+static int efx_mcdi_filter_push(struct efx_nic *efx,
 				const struct efx_filter_spec *spec, u64 *handle,
 				struct efx_rss_context *ctx, bool replacing)
 {
@@ -3889,7 +3895,7 @@ static int efx_ef10_filter_push(struct efx_nic *efx,
 	size_t outlen;
 	int rc;
 
-	efx_ef10_filter_push_prep(efx, spec, inbuf, *handle, ctx, replacing);
+	efx_mcdi_filter_push_prep(efx, spec, inbuf, *handle, ctx, replacing);
 	rc = efx_mcdi_rpc_quiet(efx, MC_CMD_FILTER_OP, inbuf, sizeof(inbuf),
 				outbuf, sizeof(outbuf), &outlen);
 	if (rc && spec->priority != EFX_FILTER_PRI_HINT)
@@ -3902,7 +3908,7 @@ static int efx_ef10_filter_push(struct efx_nic *efx,
 	return rc;
 }
 
-static u32 efx_ef10_filter_mcdi_flags_from_spec(const struct efx_filter_spec *spec)
+static u32 efx_mcdi_filter_mcdi_flags_from_spec(const struct efx_filter_spec *spec)
 {
 	enum efx_encap_type encap_type = efx_filter_get_encap_type(spec);
 	unsigned int match_flags = spec->match_flags;
@@ -3962,10 +3968,10 @@ static u32 efx_ef10_filter_mcdi_flags_from_spec(const struct efx_filter_spec *sp
 	return mcdi_flags;
 }
 
-static int efx_ef10_filter_pri(struct efx_ef10_filter_table *table,
+static int efx_mcdi_filter_pri(struct efx_mcdi_filter_table *table,
 			       const struct efx_filter_spec *spec)
 {
-	u32 mcdi_flags = efx_ef10_filter_mcdi_flags_from_spec(spec);
+	u32 mcdi_flags = efx_mcdi_filter_mcdi_flags_from_spec(spec);
 	unsigned int match_pri;
 
 	for (match_pri = 0;
@@ -3977,13 +3983,13 @@ static int efx_ef10_filter_pri(struct efx_ef10_filter_table *table,
 	return -EPROTONOSUPPORT;
 }
 
-static s32 efx_ef10_filter_insert_locked(struct efx_nic *efx,
+static s32 efx_mcdi_filter_insert_locked(struct efx_nic *efx,
 					 struct efx_filter_spec *spec,
 					 bool replace_equal)
 {
 	DECLARE_BITMAP(mc_rem_map, EFX_EF10_FILTER_SEARCH_LIMIT);
 	struct efx_ef10_nic_data *nic_data = efx->nic_data;
-	struct efx_ef10_filter_table *table;
+	struct efx_mcdi_filter_table *table;
 	struct efx_filter_spec *saved_spec;
 	struct efx_rss_context *ctx = NULL;
 	unsigned int match_pri, hash;
@@ -4007,7 +4013,7 @@ static s32 efx_ef10_filter_insert_locked(struct efx_nic *efx,
 		goto out_unlock;
 	}
 
-	rc = efx_ef10_filter_pri(table, spec);
+	rc = efx_mcdi_filter_pri(table, spec);
 	if (rc < 0)
 		goto out_unlock;
 	match_pri = rc;
@@ -4038,7 +4044,7 @@ static s32 efx_ef10_filter_insert_locked(struct efx_nic *efx,
 	 * else a free slot to insert at.
 	 */
 	for (depth = 1; depth < EFX_EF10_FILTER_SEARCH_LIMIT; depth++) {
-		i = (hash + depth) & (HUNT_FILTER_TBL_ROWS - 1);
+		i = (hash + depth) & (EFX_MCDI_FILTER_TBL_ROWS - 1);
 		saved_spec = efx_ef10_filter_entry_spec(table, i);
 
 		if (!saved_spec) {
@@ -4108,7 +4114,7 @@ static s32 efx_ef10_filter_insert_locked(struct efx_nic *efx,
 	efx_ef10_filter_set_entry(table, ins_index, saved_spec, priv_flags);
 
 	/* Actually insert the filter on the HW */
-	rc = efx_ef10_filter_push(efx, spec, &table->entry[ins_index].handle,
+	rc = efx_mcdi_filter_push(efx, spec, &table->entry[ins_index].handle,
 				  ctx, replacing);
 
 	if (rc == -EINVAL && nic_data->must_realloc_vis)
@@ -4155,7 +4161,7 @@ static s32 efx_ef10_filter_insert_locked(struct efx_nic *efx,
 			if (!test_bit(depth, mc_rem_map))
 				continue;
 
-			i = (hash + depth) & (HUNT_FILTER_TBL_ROWS - 1);
+			i = (hash + depth) & (EFX_MCDI_FILTER_TBL_ROWS - 1);
 			saved_spec = efx_ef10_filter_entry_spec(table, i);
 			priv_flags = efx_ef10_filter_entry_flags(table, i);
 
@@ -4190,20 +4196,20 @@ out_unlock:
 	return rc;
 }
 
-static s32 efx_ef10_filter_insert(struct efx_nic *efx,
+static s32 efx_mcdi_filter_insert(struct efx_nic *efx,
 				  struct efx_filter_spec *spec,
 				  bool replace_equal)
 {
 	s32 ret;
 
 	down_read(&efx->filter_sem);
-	ret = efx_ef10_filter_insert_locked(efx, spec, replace_equal);
+	ret = efx_mcdi_filter_insert_locked(efx, spec, replace_equal);
 	up_read(&efx->filter_sem);
 
 	return ret;
 }
 
-static void efx_ef10_filter_update_rx_scatter(struct efx_nic *efx)
+static void efx_mcdi_update_rx_scatter(struct efx_nic *efx)
 {
 	/* no need to do anything here on EF10 */
 }
@@ -4215,12 +4221,12 @@ static void efx_ef10_filter_update_rx_scatter(struct efx_nic *efx)
  * Caller must hold efx->filter_sem for read, and efx->filter_state->lock
  * for write.
  */
-static int efx_ef10_filter_remove_internal(struct efx_nic *efx,
+static int efx_mcdi_filter_remove_internal(struct efx_nic *efx,
 					   unsigned int priority_mask,
 					   u32 filter_id, bool by_index)
 {
 	unsigned int filter_idx = efx_ef10_filter_get_unsafe_id(filter_id);
-	struct efx_ef10_filter_table *table = efx->filter_state;
+	struct efx_mcdi_filter_table *table = efx->filter_state;
 	MCDI_DECLARE_BUF(inbuf,
 			 MC_CMD_FILTER_OP_IN_HANDLE_OFST +
 			 MC_CMD_FILTER_OP_IN_HANDLE_LEN);
@@ -4231,7 +4237,7 @@ static int efx_ef10_filter_remove_internal(struct efx_nic *efx,
 	spec = efx_ef10_filter_entry_spec(table, filter_idx);
 	if (!spec ||
 	    (!by_index &&
-	     efx_ef10_filter_pri(table, spec) !=
+	     efx_mcdi_filter_pri(table, spec) !=
 	     efx_ef10_filter_get_unsafe_pri(filter_id)))
 		return -ENOENT;
 
@@ -4257,7 +4263,7 @@ static int efx_ef10_filter_remove_internal(struct efx_nic *efx,
 				   EFX_FILTER_FLAG_RX_RSS : 0));
 		new_spec.dmaq_id = 0;
 		new_spec.rss_context = 0;
-		rc = efx_ef10_filter_push(efx, &new_spec,
+		rc = efx_mcdi_filter_push(efx, &new_spec,
 					  &table->entry[filter_idx].handle,
 					  &efx->rss_context,
 					  true);
@@ -4290,17 +4296,17 @@ static int efx_ef10_filter_remove_internal(struct efx_nic *efx,
 	return rc;
 }
 
-static int efx_ef10_filter_remove_safe(struct efx_nic *efx,
+static int efx_mcdi_filter_remove_safe(struct efx_nic *efx,
 				       enum efx_filter_priority priority,
 				       u32 filter_id)
 {
-	struct efx_ef10_filter_table *table;
+	struct efx_mcdi_filter_table *table;
 	int rc;
 
 	down_read(&efx->filter_sem);
 	table = efx->filter_state;
 	down_write(&table->lock);
-	rc = efx_ef10_filter_remove_internal(efx, 1U << priority, filter_id,
+	rc = efx_mcdi_filter_remove_internal(efx, 1U << priority, filter_id,
 					     false);
 	up_write(&table->lock);
 	up_read(&efx->filter_sem);
@@ -4308,28 +4314,28 @@ static int efx_ef10_filter_remove_safe(struct efx_nic *efx,
 }
 
 /* Caller must hold efx->filter_sem for read */
-static void efx_ef10_filter_remove_unsafe(struct efx_nic *efx,
+static void efx_mcdi_filter_remove_unsafe(struct efx_nic *efx,
 					  enum efx_filter_priority priority,
 					  u32 filter_id)
 {
-	struct efx_ef10_filter_table *table = efx->filter_state;
+	struct efx_mcdi_filter_table *table = efx->filter_state;
 
 	if (filter_id == EFX_EF10_FILTER_ID_INVALID)
 		return;
 
 	down_write(&table->lock);
-	efx_ef10_filter_remove_internal(efx, 1U << priority, filter_id,
+	efx_mcdi_filter_remove_internal(efx, 1U << priority, filter_id,
 					true);
 	up_write(&table->lock);
 }
 
-static int efx_ef10_filter_get_safe(struct efx_nic *efx,
+static int efx_mcdi_filter_get_safe(struct efx_nic *efx,
 				    enum efx_filter_priority priority,
 				    u32 filter_id, struct efx_filter_spec *spec)
 {
 	unsigned int filter_idx = efx_ef10_filter_get_unsafe_id(filter_id);
 	const struct efx_filter_spec *saved_spec;
-	struct efx_ef10_filter_table *table;
+	struct efx_mcdi_filter_table *table;
 	int rc;
 
 	down_read(&efx->filter_sem);
@@ -4337,7 +4343,7 @@ static int efx_ef10_filter_get_safe(struct efx_nic *efx,
 	down_read(&table->lock);
 	saved_spec = efx_ef10_filter_entry_spec(table, filter_idx);
 	if (saved_spec && saved_spec->priority == priority &&
-	    efx_ef10_filter_pri(table, saved_spec) ==
+	    efx_mcdi_filter_pri(table, saved_spec) ==
 	    efx_ef10_filter_get_unsafe_pri(filter_id)) {
 		*spec = *saved_spec;
 		rc = 0;
@@ -4349,10 +4355,10 @@ static int efx_ef10_filter_get_safe(struct efx_nic *efx,
 	return rc;
 }
 
-static int efx_ef10_filter_clear_rx(struct efx_nic *efx,
+static int efx_mcdi_filter_clear_rx(struct efx_nic *efx,
 				    enum efx_filter_priority priority)
 {
-	struct efx_ef10_filter_table *table;
+	struct efx_mcdi_filter_table *table;
 	unsigned int priority_mask;
 	unsigned int i;
 	int rc;
@@ -4363,8 +4369,8 @@ static int efx_ef10_filter_clear_rx(struct efx_nic *efx,
 	down_read(&efx->filter_sem);
 	table = efx->filter_state;
 	down_write(&table->lock);
-	for (i = 0; i < HUNT_FILTER_TBL_ROWS; i++) {
-		rc = efx_ef10_filter_remove_internal(efx, priority_mask,
+	for (i = 0; i < EFX_MCDI_FILTER_TBL_ROWS; i++) {
+		rc = efx_mcdi_filter_remove_internal(efx, priority_mask,
 						     i, true);
 		if (rc && rc != -ENOENT)
 			break;
@@ -4376,17 +4382,17 @@ static int efx_ef10_filter_clear_rx(struct efx_nic *efx,
 	return rc;
 }
 
-static u32 efx_ef10_filter_count_rx_used(struct efx_nic *efx,
+static u32 efx_mcdi_filter_count_rx_used(struct efx_nic *efx,
 					 enum efx_filter_priority priority)
 {
-	struct efx_ef10_filter_table *table;
+	struct efx_mcdi_filter_table *table;
 	unsigned int filter_idx;
 	s32 count = 0;
 
 	down_read(&efx->filter_sem);
 	table = efx->filter_state;
 	down_read(&table->lock);
-	for (filter_idx = 0; filter_idx < HUNT_FILTER_TBL_ROWS; filter_idx++) {
+	for (filter_idx = 0; filter_idx < EFX_MCDI_FILTER_TBL_ROWS; filter_idx++) {
 		if (table->entry[filter_idx].spec &&
 		    efx_ef10_filter_entry_spec(table, filter_idx)->priority ==
 		    priority)
@@ -4397,18 +4403,18 @@ static u32 efx_ef10_filter_count_rx_used(struct efx_nic *efx,
 	return count;
 }
 
-static u32 efx_ef10_filter_get_rx_id_limit(struct efx_nic *efx)
+static u32 efx_mcdi_filter_get_rx_id_limit(struct efx_nic *efx)
 {
-	struct efx_ef10_filter_table *table = efx->filter_state;
+	struct efx_mcdi_filter_table *table = efx->filter_state;
 
-	return table->rx_match_count * HUNT_FILTER_TBL_ROWS * 2;
+	return table->rx_match_count * EFX_MCDI_FILTER_TBL_ROWS * 2;
 }
 
-static s32 efx_ef10_filter_get_rx_ids(struct efx_nic *efx,
+static s32 efx_mcdi_filter_get_rx_ids(struct efx_nic *efx,
 				      enum efx_filter_priority priority,
 				      u32 *buf, u32 size)
 {
-	struct efx_ef10_filter_table *table;
+	struct efx_mcdi_filter_table *table;
 	struct efx_filter_spec *spec;
 	unsigned int filter_idx;
 	s32 count = 0;
@@ -4417,7 +4423,7 @@ static s32 efx_ef10_filter_get_rx_ids(struct efx_nic *efx,
 	table = efx->filter_state;
 	down_read(&table->lock);
 
-	for (filter_idx = 0; filter_idx < HUNT_FILTER_TBL_ROWS; filter_idx++) {
+	for (filter_idx = 0; filter_idx < EFX_MCDI_FILTER_TBL_ROWS; filter_idx++) {
 		spec = efx_ef10_filter_entry_spec(table, filter_idx);
 		if (spec && spec->priority == priority) {
 			if (count == size) {
@@ -4426,7 +4432,7 @@ static s32 efx_ef10_filter_get_rx_ids(struct efx_nic *efx,
 			}
 			buf[count++] =
 				efx_ef10_make_filter_id(
-					efx_ef10_filter_pri(table, spec),
+					efx_mcdi_filter_pri(table, spec),
 					filter_idx);
 		}
 	}
@@ -4437,11 +4443,11 @@ static s32 efx_ef10_filter_get_rx_ids(struct efx_nic *efx,
 
 #ifdef CONFIG_RFS_ACCEL
 
-static bool efx_ef10_filter_rfs_expire_one(struct efx_nic *efx, u32 flow_id,
+static bool efx_mcdi_filter_rfs_expire_one(struct efx_nic *efx, u32 flow_id,
 					   unsigned int filter_idx)
 {
 	struct efx_filter_spec *spec, saved_spec;
-	struct efx_ef10_filter_table *table;
+	struct efx_mcdi_filter_table *table;
 	struct efx_arfs_rule *rule = NULL;
 	bool ret = true, force = false;
 	u16 arfs_id;
@@ -4486,7 +4492,7 @@ expire:
 	 * the rule is not removed by efx_rps_hash_del() below.
 	 */
 	if (ret)
-		ret = efx_ef10_filter_remove_internal(efx, 1U << spec->priority,
+		ret = efx_mcdi_filter_remove_internal(efx, 1U << spec->priority,
 						      filter_idx, true) == 0;
 	/* While we can't safely dereference rule (we dropped the lock), we can
 	 * still test it for NULL.
@@ -4505,7 +4511,7 @@ out_unlock:
 
 #endif /* CONFIG_RFS_ACCEL */
 
-static int efx_ef10_filter_match_flags_from_mcdi(bool encap, u32 mcdi_flags)
+static int efx_mcdi_filter_match_flags_from_mcdi(bool encap, u32 mcdi_flags)
 {
 	int match_flags = 0;
 
@@ -4562,12 +4568,12 @@ static int efx_ef10_filter_match_flags_from_mcdi(bool encap, u32 mcdi_flags)
 	return match_flags;
 }
 
-static void efx_ef10_filter_cleanup_vlans(struct efx_nic *efx)
+static void efx_mcdi_filter_cleanup_vlans(struct efx_nic *efx)
 {
-	struct efx_ef10_filter_table *table = efx->filter_state;
-	struct efx_ef10_filter_vlan *vlan, *next_vlan;
+	struct efx_mcdi_filter_table *table = efx->filter_state;
+	struct efx_mcdi_filter_vlan *vlan, *next_vlan;
 
-	/* See comment in efx_ef10_filter_table_remove() */
+	/* See comment in efx_mcdi_filter_table_remove() */
 	if (!efx_rwsem_assert_write_locked(&efx->filter_sem))
 		return;
 
@@ -4575,10 +4581,10 @@ static void efx_ef10_filter_cleanup_vlans(struct efx_nic *efx)
 		return;
 
 	list_for_each_entry_safe(vlan, next_vlan, &table->vlan_list, list)
-		efx_ef10_filter_del_vlan_internal(efx, vlan);
+		efx_mcdi_filter_del_vlan_internal(efx, vlan);
 }
 
-static bool efx_ef10_filter_match_supported(struct efx_ef10_filter_table *table,
+static bool efx_mcdi_filter_match_supported(struct efx_mcdi_filter_table *table,
 					    bool encap,
 					    enum efx_filter_match_flags match_flags)
 {
@@ -4588,7 +4594,7 @@ static bool efx_ef10_filter_match_supported(struct efx_ef10_filter_table *table,
 	for (match_pri = 0;
 	     match_pri < table->rx_match_count;
 	     match_pri++) {
-		mf = efx_ef10_filter_match_flags_from_mcdi(encap,
+		mf = efx_mcdi_filter_match_flags_from_mcdi(encap,
 				table->rx_match_mcdi_flags[match_pri]);
 		if (mf == match_flags)
 			return true;
@@ -4598,8 +4604,8 @@ static bool efx_ef10_filter_match_supported(struct efx_ef10_filter_table *table,
 }
 
 static int
-efx_ef10_filter_table_probe_matches(struct efx_nic *efx,
-				    struct efx_ef10_filter_table *table,
+efx_mcdi_filter_table_probe_matches(struct efx_nic *efx,
+				    struct efx_mcdi_filter_table *table,
 				    bool encap)
 {
 	MCDI_DECLARE_BUF(inbuf, MC_CMD_GET_PARSER_DISP_INFO_IN_LEN);
@@ -4628,7 +4634,7 @@ efx_ef10_filter_table_probe_matches(struct efx_nic *efx,
 				outbuf,
 				GET_PARSER_DISP_INFO_OUT_SUPPORTED_MATCHES,
 				pd_match_pri);
-		rc = efx_ef10_filter_match_flags_from_mcdi(encap, mcdi_flags);
+		rc = efx_mcdi_filter_match_flags_from_mcdi(encap, mcdi_flags);
 		if (rc < 0) {
 			netif_dbg(efx, probe, efx->net_dev,
 				  "%s: fw flags %#x pri %u not supported in driver\n",
@@ -4646,11 +4652,11 @@ efx_ef10_filter_table_probe_matches(struct efx_nic *efx,
 	return 0;
 }
 
-static int efx_ef10_filter_table_probe(struct efx_nic *efx)
+static int efx_mcdi_filter_table_probe(struct efx_nic *efx)
 {
 	struct efx_ef10_nic_data *nic_data = efx->nic_data;
 	struct net_device *net_dev = efx->net_dev;
-	struct efx_ef10_filter_table *table;
+	struct efx_mcdi_filter_table *table;
 	struct efx_ef10_vlan *vlan;
 	int rc;
 
@@ -4665,18 +4671,18 @@ static int efx_ef10_filter_table_probe(struct efx_nic *efx)
 		return -ENOMEM;
 
 	table->rx_match_count = 0;
-	rc = efx_ef10_filter_table_probe_matches(efx, table, false);
+	rc = efx_mcdi_filter_table_probe_matches(efx, table, false);
 	if (rc)
 		goto fail;
 	if (nic_data->datapath_caps &
 		   (1 << MC_CMD_GET_CAPABILITIES_OUT_VXLAN_NVGRE_LBN))
-		rc = efx_ef10_filter_table_probe_matches(efx, table, true);
+		rc = efx_mcdi_filter_table_probe_matches(efx, table, true);
 	if (rc)
 		goto fail;
 	if ((efx_supported_features(efx) & NETIF_F_HW_VLAN_CTAG_FILTER) &&
-	    !(efx_ef10_filter_match_supported(table, false,
+	    !(efx_mcdi_filter_match_supported(table, false,
 		(EFX_FILTER_MATCH_OUTER_VID | EFX_FILTER_MATCH_LOC_MAC)) &&
-	      efx_ef10_filter_match_supported(table, false,
+	      efx_mcdi_filter_match_supported(table, false,
 		(EFX_FILTER_MATCH_OUTER_VID | EFX_FILTER_MATCH_LOC_MAC_IG)))) {
 		netif_info(efx, probe, net_dev,
 			   "VLAN filters are not supported in this firmware variant\n");
@@ -4685,7 +4691,7 @@ static int efx_ef10_filter_table_probe(struct efx_nic *efx)
 		net_dev->hw_features &= ~NETIF_F_HW_VLAN_CTAG_FILTER;
 	}
 
-	table->entry = vzalloc(array_size(HUNT_FILTER_TBL_ROWS,
+	table->entry = vzalloc(array_size(EFX_MCDI_FILTER_TBL_ROWS,
 					  sizeof(*table->entry)));
 	if (!table->entry) {
 		rc = -ENOMEM;
@@ -4701,7 +4707,7 @@ static int efx_ef10_filter_table_probe(struct efx_nic *efx)
 	efx->filter_state = table;
 
 	list_for_each_entry(vlan, &nic_data->vlan_list, list) {
-		rc = efx_ef10_filter_add_vlan(efx, vlan->vid);
+		rc = efx_mcdi_filter_add_vlan(efx, vlan->vid);
 		if (rc)
 			goto fail_add_vlan;
 	}
@@ -4709,7 +4715,7 @@ static int efx_ef10_filter_table_probe(struct efx_nic *efx)
 	return 0;
 
 fail_add_vlan:
-	efx_ef10_filter_cleanup_vlans(efx);
+	efx_mcdi_filter_cleanup_vlans(efx);
 	efx->filter_state = NULL;
 fail:
 	kfree(table);
@@ -4717,14 +4723,14 @@ fail:
 }
 
 /* Caller must hold efx->filter_sem for read if race against
- * efx_ef10_filter_table_remove() is possible
+ * efx_mcdi_filter_table_remove() is possible
  */
-static void efx_ef10_filter_table_restore(struct efx_nic *efx)
+static void efx_mcdi_filter_table_restore(struct efx_nic *efx)
 {
-	struct efx_ef10_filter_table *table = efx->filter_state;
+	struct efx_mcdi_filter_table *table = efx->filter_state;
 	struct efx_ef10_nic_data *nic_data = efx->nic_data;
 	unsigned int invalid_filters = 0, failed = 0;
-	struct efx_ef10_filter_vlan *vlan;
+	struct efx_mcdi_filter_vlan *vlan;
 	struct efx_filter_spec *spec;
 	struct efx_rss_context *ctx;
 	unsigned int filter_idx;
@@ -4743,12 +4749,12 @@ static void efx_ef10_filter_table_restore(struct efx_nic *efx)
 	down_write(&table->lock);
 	mutex_lock(&efx->rss_lock);
 
-	for (filter_idx = 0; filter_idx < HUNT_FILTER_TBL_ROWS; filter_idx++) {
+	for (filter_idx = 0; filter_idx < EFX_MCDI_FILTER_TBL_ROWS; filter_idx++) {
 		spec = efx_ef10_filter_entry_spec(table, filter_idx);
 		if (!spec)
 			continue;
 
-		mcdi_flags = efx_ef10_filter_mcdi_flags_from_spec(spec);
+		mcdi_flags = efx_mcdi_filter_mcdi_flags_from_spec(spec);
 		match_pri = 0;
 		while (match_pri < table->rx_match_count &&
 		       table->rx_match_mcdi_flags[match_pri] != mcdi_flags)
@@ -4778,7 +4784,7 @@ static void efx_ef10_filter_table_restore(struct efx_nic *efx)
 			}
 		}
 
-		rc = efx_ef10_filter_push(efx, spec,
+		rc = efx_mcdi_filter_push(efx, spec,
 					  &table->entry[filter_idx].handle,
 					  ctx, false);
 		if (rc)
@@ -4815,15 +4821,15 @@ not_restored:
 		nic_data->must_restore_filters = false;
 }
 
-static void efx_ef10_filter_table_remove(struct efx_nic *efx)
+static void efx_mcdi_filter_table_remove(struct efx_nic *efx)
 {
-	struct efx_ef10_filter_table *table = efx->filter_state;
+	struct efx_mcdi_filter_table *table = efx->filter_state;
 	MCDI_DECLARE_BUF(inbuf, MC_CMD_FILTER_OP_EXT_IN_LEN);
 	struct efx_filter_spec *spec;
 	unsigned int filter_idx;
 	int rc;
 
-	efx_ef10_filter_cleanup_vlans(efx);
+	efx_mcdi_filter_cleanup_vlans(efx);
 	efx->filter_state = NULL;
 	/* If we were called without locking, then it's not safe to free
 	 * the table as others might be using it.  So we just WARN, leak
@@ -4837,7 +4843,7 @@ static void efx_ef10_filter_table_remove(struct efx_nic *efx)
 	if (!table)
 		return;
 
-	for (filter_idx = 0; filter_idx < HUNT_FILTER_TBL_ROWS; filter_idx++) {
+	for (filter_idx = 0; filter_idx < EFX_MCDI_FILTER_TBL_ROWS; filter_idx++) {
 		spec = efx_ef10_filter_entry_spec(table, filter_idx);
 		if (!spec)
 			continue;
@@ -4861,9 +4867,9 @@ static void efx_ef10_filter_table_remove(struct efx_nic *efx)
 	kfree(table);
 }
 
-static void efx_ef10_filter_mark_one_old(struct efx_nic *efx, uint16_t *id)
+static void efx_mcdi_filter_mark_one_old(struct efx_nic *efx, uint16_t *id)
 {
-	struct efx_ef10_filter_table *table = efx->filter_state;
+	struct efx_mcdi_filter_table *table = efx->filter_state;
 	unsigned int filter_idx;
 
 	efx_rwsem_assert_write_locked(&table->lock);
@@ -4880,38 +4886,38 @@ static void efx_ef10_filter_mark_one_old(struct efx_nic *efx, uint16_t *id)
 }
 
 /* Mark old per-VLAN filters that may need to be removed */
-static void _efx_ef10_filter_vlan_mark_old(struct efx_nic *efx,
-					   struct efx_ef10_filter_vlan *vlan)
+static void _efx_mcdi_filter_vlan_mark_old(struct efx_nic *efx,
+					   struct efx_mcdi_filter_vlan *vlan)
 {
-	struct efx_ef10_filter_table *table = efx->filter_state;
+	struct efx_mcdi_filter_table *table = efx->filter_state;
 	unsigned int i;
 
 	for (i = 0; i < table->dev_uc_count; i++)
-		efx_ef10_filter_mark_one_old(efx, &vlan->uc[i]);
+		efx_mcdi_filter_mark_one_old(efx, &vlan->uc[i]);
 	for (i = 0; i < table->dev_mc_count; i++)
-		efx_ef10_filter_mark_one_old(efx, &vlan->mc[i]);
+		efx_mcdi_filter_mark_one_old(efx, &vlan->mc[i]);
 	for (i = 0; i < EFX_EF10_NUM_DEFAULT_FILTERS; i++)
-		efx_ef10_filter_mark_one_old(efx, &vlan->default_filters[i]);
+		efx_mcdi_filter_mark_one_old(efx, &vlan->default_filters[i]);
 }
 
 /* Mark old filters that may need to be removed.
  * Caller must hold efx->filter_sem for read if race against
- * efx_ef10_filter_table_remove() is possible
+ * efx_mcdi_filter_table_remove() is possible
  */
-static void efx_ef10_filter_mark_old(struct efx_nic *efx)
+static void efx_mcdi_filter_mark_old(struct efx_nic *efx)
 {
-	struct efx_ef10_filter_table *table = efx->filter_state;
-	struct efx_ef10_filter_vlan *vlan;
+	struct efx_mcdi_filter_table *table = efx->filter_state;
+	struct efx_mcdi_filter_vlan *vlan;
 
 	down_write(&table->lock);
 	list_for_each_entry(vlan, &table->vlan_list, list)
-		_efx_ef10_filter_vlan_mark_old(efx, vlan);
+		_efx_mcdi_filter_vlan_mark_old(efx, vlan);
 	up_write(&table->lock);
 }
 
-static void efx_ef10_filter_uc_addr_list(struct efx_nic *efx)
+static void efx_mcdi_filter_uc_addr_list(struct efx_nic *efx)
 {
-	struct efx_ef10_filter_table *table = efx->filter_state;
+	struct efx_mcdi_filter_table *table = efx->filter_state;
 	struct net_device *net_dev = efx->net_dev;
 	struct netdev_hw_addr *uc;
 	unsigned int i;
@@ -4931,9 +4937,9 @@ static void efx_ef10_filter_uc_addr_list(struct efx_nic *efx)
 	table->dev_uc_count = i;
 }
 
-static void efx_ef10_filter_mc_addr_list(struct efx_nic *efx)
+static void efx_mcdi_filter_mc_addr_list(struct efx_nic *efx)
 {
-	struct efx_ef10_filter_table *table = efx->filter_state;
+	struct efx_mcdi_filter_table *table = efx->filter_state;
 	struct net_device *net_dev = efx->net_dev;
 	struct netdev_hw_addr *mc;
 	unsigned int i;
@@ -4955,12 +4961,12 @@ static void efx_ef10_filter_mc_addr_list(struct efx_nic *efx)
 	table->dev_mc_count = i;
 }
 
-static int efx_ef10_filter_insert_addr_list(struct efx_nic *efx,
-					    struct efx_ef10_filter_vlan *vlan,
+static int efx_mcdi_filter_insert_addr_list(struct efx_nic *efx,
+					    struct efx_mcdi_filter_vlan *vlan,
 					    bool multicast, bool rollback)
 {
-	struct efx_ef10_filter_table *table = efx->filter_state;
-	struct efx_ef10_dev_addr *addr_list;
+	struct efx_mcdi_filter_table *table = efx->filter_state;
+	struct efx_mcdi_dev_addr *addr_list;
 	enum efx_filter_flags filter_flags;
 	struct efx_filter_spec spec;
 	u8 baddr[ETH_ALEN];
@@ -4986,7 +4992,7 @@ static int efx_ef10_filter_insert_addr_list(struct efx_nic *efx,
 		EFX_WARN_ON_PARANOID(ids[i] != EFX_EF10_FILTER_ID_INVALID);
 		efx_filter_init_rx(&spec, EFX_FILTER_PRI_AUTO, filter_flags, 0);
 		efx_filter_set_eth_local(&spec, vlan->vid, addr_list[i].addr);
-		rc = efx_ef10_filter_insert_locked(efx, &spec, true);
+		rc = efx_mcdi_filter_insert_locked(efx, &spec, true);
 		if (rc < 0) {
 			if (rollback) {
 				netif_info(efx, drv, efx->net_dev,
@@ -4994,7 +5000,7 @@ static int efx_ef10_filter_insert_addr_list(struct efx_nic *efx,
 					   rc);
 				/* Fall back to promiscuous */
 				for (j = 0; j < i; j++) {
-					efx_ef10_filter_remove_unsafe(
+					efx_mcdi_filter_remove_unsafe(
 						efx, EFX_FILTER_PRI_AUTO,
 						ids[j]);
 					ids[j] = EFX_EF10_FILTER_ID_INVALID;
@@ -5015,13 +5021,13 @@ static int efx_ef10_filter_insert_addr_list(struct efx_nic *efx,
 		efx_filter_init_rx(&spec, EFX_FILTER_PRI_AUTO, filter_flags, 0);
 		eth_broadcast_addr(baddr);
 		efx_filter_set_eth_local(&spec, vlan->vid, baddr);
-		rc = efx_ef10_filter_insert_locked(efx, &spec, true);
+		rc = efx_mcdi_filter_insert_locked(efx, &spec, true);
 		if (rc < 0) {
 			netif_warn(efx, drv, efx->net_dev,
 				   "Broadcast filter insert failed rc=%d\n", rc);
 			/* Fall back to promiscuous */
 			for (j = 0; j < i; j++) {
-				efx_ef10_filter_remove_unsafe(
+				efx_mcdi_filter_remove_unsafe(
 					efx, EFX_FILTER_PRI_AUTO,
 					ids[j]);
 				ids[j] = EFX_EF10_FILTER_ID_INVALID;
@@ -5036,8 +5042,8 @@ static int efx_ef10_filter_insert_addr_list(struct efx_nic *efx,
 	return 0;
 }
 
-static int efx_ef10_filter_insert_def(struct efx_nic *efx,
-				      struct efx_ef10_filter_vlan *vlan,
+static int efx_mcdi_filter_insert_def(struct efx_nic *efx,
+				      struct efx_mcdi_filter_vlan *vlan,
 				      enum efx_encap_type encap_type,
 				      bool multicast, bool rollback)
 {
@@ -5071,7 +5077,7 @@ static int efx_ef10_filter_insert_def(struct efx_nic *efx,
 	if (vlan->vid != EFX_FILTER_VID_UNSPEC)
 		efx_filter_set_eth_local(&spec, vlan->vid, NULL);
 
-	rc = efx_ef10_filter_insert_locked(efx, &spec, true);
+	rc = efx_mcdi_filter_insert_locked(efx, &spec, true);
 	if (rc < 0) {
 		const char *um = multicast ? "Multicast" : "Unicast";
 		const char *encap_name = "";
@@ -5101,7 +5107,7 @@ static int efx_ef10_filter_insert_def(struct efx_nic *efx,
 			       encap_name, encap_ipv, um, rc);
 	} else if (multicast) {
 		/* mapping from encap types to default filter IDs (multicast) */
-		static enum efx_ef10_default_filters map[] = {
+		static enum efx_mcdi_filter_default_filters map[] = {
 			[EFX_ENCAP_TYPE_NONE] = EFX_EF10_MCDEF,
 			[EFX_ENCAP_TYPE_VXLAN] = EFX_EF10_VXLAN4_MCDEF,
 			[EFX_ENCAP_TYPE_NVGRE] = EFX_EF10_NVGRE4_MCDEF,
@@ -5131,14 +5137,14 @@ static int efx_ef10_filter_insert_def(struct efx_nic *efx,
 					   filter_flags, 0);
 			eth_broadcast_addr(baddr);
 			efx_filter_set_eth_local(&spec, vlan->vid, baddr);
-			rc = efx_ef10_filter_insert_locked(efx, &spec, true);
+			rc = efx_mcdi_filter_insert_locked(efx, &spec, true);
 			if (rc < 0) {
 				netif_warn(efx, drv, efx->net_dev,
 					   "Broadcast filter insert failed rc=%d\n",
 					   rc);
 				if (rollback) {
 					/* Roll back the mc_def filter */
-					efx_ef10_filter_remove_unsafe(
+					efx_mcdi_filter_remove_unsafe(
 							efx, EFX_FILTER_PRI_AUTO,
 							*id);
 					*id = EFX_EF10_FILTER_ID_INVALID;
@@ -5155,7 +5161,7 @@ static int efx_ef10_filter_insert_def(struct efx_nic *efx,
 		rc = 0;
 	} else {
 		/* mapping from encap types to default filter IDs (unicast) */
-		static enum efx_ef10_default_filters map[] = {
+		static enum efx_mcdi_filter_default_filters map[] = {
 			[EFX_ENCAP_TYPE_NONE] = EFX_EF10_UCDEF,
 			[EFX_ENCAP_TYPE_VXLAN] = EFX_EF10_VXLAN4_UCDEF,
 			[EFX_ENCAP_TYPE_NVGRE] = EFX_EF10_NVGRE4_UCDEF,
@@ -5184,19 +5190,19 @@ static int efx_ef10_filter_insert_def(struct efx_nic *efx,
 }
 
 /* Remove filters that weren't renewed. */
-static void efx_ef10_filter_remove_old(struct efx_nic *efx)
+static void efx_mcdi_filter_remove_old(struct efx_nic *efx)
 {
-	struct efx_ef10_filter_table *table = efx->filter_state;
+	struct efx_mcdi_filter_table *table = efx->filter_state;
 	int remove_failed = 0;
 	int remove_noent = 0;
 	int rc;
 	int i;
 
 	down_write(&table->lock);
-	for (i = 0; i < HUNT_FILTER_TBL_ROWS; i++) {
+	for (i = 0; i < EFX_MCDI_FILTER_TBL_ROWS; i++) {
 		if (READ_ONCE(table->entry[i].spec) &
 		    EFX_EF10_FILTER_FLAG_AUTO_OLD) {
-			rc = efx_ef10_filter_remove_internal(efx,
+			rc = efx_mcdi_filter_remove_internal(efx,
 					1U << EFX_FILTER_PRI_AUTO, i, true);
 			if (rc == -ENOENT)
 				remove_noent++;
@@ -5229,7 +5235,7 @@ static int efx_ef10_vport_set_mac_address(struct efx_nic *efx)
 	efx_device_detach_sync(efx);
 	efx_net_stop(efx->net_dev);
 	down_write(&efx->filter_sem);
-	efx_ef10_filter_table_remove(efx);
+	efx_mcdi_filter_table_remove(efx);
 	up_write(&efx->filter_sem);
 
 	rc = efx_ef10_vadaptor_free(efx, nic_data->vport_id);
@@ -5261,7 +5267,7 @@ restore_vadaptor:
 		goto reset_nic;
 restore_filters:
 	down_write(&efx->filter_sem);
-	rc2 = efx_ef10_filter_table_probe(efx);
+	rc2 = efx_mcdi_filter_table_probe(efx);
 	up_write(&efx->filter_sem);
 	if (rc2)
 		goto reset_nic;
@@ -5283,12 +5289,12 @@ reset_nic:
 }
 
 /* Caller must hold efx->filter_sem for read if race against
- * efx_ef10_filter_table_remove() is possible
+ * efx_mcdi_filter_table_remove() is possible
  */
-static void efx_ef10_filter_vlan_sync_rx_mode(struct efx_nic *efx,
-					      struct efx_ef10_filter_vlan *vlan)
+static void efx_mcdi_filter_vlan_sync_rx_mode(struct efx_nic *efx,
+					      struct efx_mcdi_filter_vlan *vlan)
 {
-	struct efx_ef10_filter_table *table = efx->filter_state;
+	struct efx_mcdi_filter_table *table = efx->filter_state;
 	struct efx_ef10_nic_data *nic_data = efx->nic_data;
 
 	/* Do not install unspecified VID if VLAN filtering is enabled.
@@ -5299,32 +5305,32 @@ static void efx_ef10_filter_vlan_sync_rx_mode(struct efx_nic *efx,
 
 	/* Insert/renew unicast filters */
 	if (table->uc_promisc) {
-		efx_ef10_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_NONE,
+		efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_NONE,
 					   false, false);
-		efx_ef10_filter_insert_addr_list(efx, vlan, false, false);
+		efx_mcdi_filter_insert_addr_list(efx, vlan, false, false);
 	} else {
 		/* If any of the filters failed to insert, fall back to
 		 * promiscuous mode - add in the uc_def filter.  But keep
 		 * our individual unicast filters.
 		 */
-		if (efx_ef10_filter_insert_addr_list(efx, vlan, false, false))
-			efx_ef10_filter_insert_def(efx, vlan,
+		if (efx_mcdi_filter_insert_addr_list(efx, vlan, false, false))
+			efx_mcdi_filter_insert_def(efx, vlan,
 						   EFX_ENCAP_TYPE_NONE,
 						   false, false);
 	}
-	efx_ef10_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_VXLAN,
+	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_VXLAN,
 				   false, false);
-	efx_ef10_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_VXLAN |
+	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_VXLAN |
 					      EFX_ENCAP_FLAG_IPV6,
 				   false, false);
-	efx_ef10_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_NVGRE,
+	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_NVGRE,
 				   false, false);
-	efx_ef10_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_NVGRE |
+	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_NVGRE |
 					      EFX_ENCAP_FLAG_IPV6,
 				   false, false);
-	efx_ef10_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_GENEVE,
+	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_GENEVE,
 				   false, false);
-	efx_ef10_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_GENEVE |
+	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_GENEVE |
 					      EFX_ENCAP_FLAG_IPV6,
 				   false, false);
 
@@ -5334,18 +5340,18 @@ static void efx_ef10_filter_vlan_sync_rx_mode(struct efx_nic *efx,
 	 */
 	if (nic_data->workaround_26807 &&
 	    table->mc_promisc_last != table->mc_promisc)
-		efx_ef10_filter_remove_old(efx);
+		efx_mcdi_filter_remove_old(efx);
 	if (table->mc_promisc) {
 		if (nic_data->workaround_26807) {
 			/* If we failed to insert promiscuous filters, rollback
 			 * and fall back to individual multicast filters
 			 */
-			if (efx_ef10_filter_insert_def(efx, vlan,
+			if (efx_mcdi_filter_insert_def(efx, vlan,
 						       EFX_ENCAP_TYPE_NONE,
 						       true, true)) {
 				/* Changing promisc state, so remove old filters */
-				efx_ef10_filter_remove_old(efx);
-				efx_ef10_filter_insert_addr_list(efx, vlan,
+				efx_mcdi_filter_remove_old(efx);
+				efx_mcdi_filter_insert_addr_list(efx, vlan,
 								 true, false);
 			}
 		} else {
@@ -5353,11 +5359,11 @@ static void efx_ef10_filter_vlan_sync_rx_mode(struct efx_nic *efx,
 			 * rollback.  Regardless, also insert the mc_list,
 			 * unless it's incomplete due to overflow
 			 */
-			efx_ef10_filter_insert_def(efx, vlan,
+			efx_mcdi_filter_insert_def(efx, vlan,
 						   EFX_ENCAP_TYPE_NONE,
 						   true, false);
 			if (!table->mc_overflow)
-				efx_ef10_filter_insert_addr_list(efx, vlan,
+				efx_mcdi_filter_insert_addr_list(efx, vlan,
 								 true, false);
 		}
 	} else {
@@ -5366,42 +5372,42 @@ static void efx_ef10_filter_vlan_sync_rx_mode(struct efx_nic *efx,
 		 * that fails, roll back again and insert as many of our
 		 * individual multicast filters as we can.
 		 */
-		if (efx_ef10_filter_insert_addr_list(efx, vlan, true, true)) {
+		if (efx_mcdi_filter_insert_addr_list(efx, vlan, true, true)) {
 			/* Changing promisc state, so remove old filters */
 			if (nic_data->workaround_26807)
-				efx_ef10_filter_remove_old(efx);
-			if (efx_ef10_filter_insert_def(efx, vlan,
+				efx_mcdi_filter_remove_old(efx);
+			if (efx_mcdi_filter_insert_def(efx, vlan,
 						       EFX_ENCAP_TYPE_NONE,
 						       true, true))
-				efx_ef10_filter_insert_addr_list(efx, vlan,
+				efx_mcdi_filter_insert_addr_list(efx, vlan,
 								 true, false);
 		}
 	}
-	efx_ef10_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_VXLAN,
+	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_VXLAN,
 				   true, false);
-	efx_ef10_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_VXLAN |
+	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_VXLAN |
 					      EFX_ENCAP_FLAG_IPV6,
 				   true, false);
-	efx_ef10_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_NVGRE,
+	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_NVGRE,
 				   true, false);
-	efx_ef10_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_NVGRE |
+	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_NVGRE |
 					      EFX_ENCAP_FLAG_IPV6,
 				   true, false);
-	efx_ef10_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_GENEVE,
+	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_GENEVE,
 				   true, false);
-	efx_ef10_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_GENEVE |
+	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_GENEVE |
 					      EFX_ENCAP_FLAG_IPV6,
 				   true, false);
 }
 
 /* Caller must hold efx->filter_sem for read if race against
- * efx_ef10_filter_table_remove() is possible
+ * efx_mcdi_filter_table_remove() is possible
  */
-static void efx_ef10_filter_sync_rx_mode(struct efx_nic *efx)
+static void efx_mcdi_filter_sync_rx_mode(struct efx_nic *efx)
 {
-	struct efx_ef10_filter_table *table = efx->filter_state;
+	struct efx_mcdi_filter_table *table = efx->filter_state;
 	struct net_device *net_dev = efx->net_dev;
-	struct efx_ef10_filter_vlan *vlan;
+	struct efx_mcdi_filter_vlan *vlan;
 	bool vlan_filter;
 
 	if (!efx_dev_registered(efx))
@@ -5410,14 +5416,14 @@ static void efx_ef10_filter_sync_rx_mode(struct efx_nic *efx)
 	if (!table)
 		return;
 
-	efx_ef10_filter_mark_old(efx);
+	efx_mcdi_filter_mark_old(efx);
 
 	/* Copy/convert the address lists; add the primary station
 	 * address and broadcast address
 	 */
 	netif_addr_lock_bh(net_dev);
-	efx_ef10_filter_uc_addr_list(efx);
-	efx_ef10_filter_mc_addr_list(efx);
+	efx_mcdi_filter_uc_addr_list(efx);
+	efx_mcdi_filter_mc_addr_list(efx);
 	netif_addr_unlock_bh(net_dev);
 
 	/* If VLAN filtering changes, all old filters are finally removed.
@@ -5427,20 +5433,20 @@ static void efx_ef10_filter_sync_rx_mode(struct efx_nic *efx)
 	vlan_filter = !!(net_dev->features & NETIF_F_HW_VLAN_CTAG_FILTER);
 	if (table->vlan_filter != vlan_filter) {
 		table->vlan_filter = vlan_filter;
-		efx_ef10_filter_remove_old(efx);
+		efx_mcdi_filter_remove_old(efx);
 	}
 
 	list_for_each_entry(vlan, &table->vlan_list, list)
-		efx_ef10_filter_vlan_sync_rx_mode(efx, vlan);
+		efx_mcdi_filter_vlan_sync_rx_mode(efx, vlan);
 
-	efx_ef10_filter_remove_old(efx);
+	efx_mcdi_filter_remove_old(efx);
 	table->mc_promisc_last = table->mc_promisc;
 }
 
-static struct efx_ef10_filter_vlan *efx_ef10_filter_find_vlan(struct efx_nic *efx, u16 vid)
+static struct efx_mcdi_filter_vlan *efx_mcdi_filter_find_vlan(struct efx_nic *efx, u16 vid)
 {
-	struct efx_ef10_filter_table *table = efx->filter_state;
-	struct efx_ef10_filter_vlan *vlan;
+	struct efx_mcdi_filter_table *table = efx->filter_state;
+	struct efx_mcdi_filter_vlan *vlan;
 
 	WARN_ON(!rwsem_is_locked(&efx->filter_sem));
 
@@ -5452,16 +5458,16 @@ static struct efx_ef10_filter_vlan *efx_ef10_filter_find_vlan(struct efx_nic *ef
 	return NULL;
 }
 
-static int efx_ef10_filter_add_vlan(struct efx_nic *efx, u16 vid)
+static int efx_mcdi_filter_add_vlan(struct efx_nic *efx, u16 vid)
 {
-	struct efx_ef10_filter_table *table = efx->filter_state;
-	struct efx_ef10_filter_vlan *vlan;
+	struct efx_mcdi_filter_table *table = efx->filter_state;
+	struct efx_mcdi_filter_vlan *vlan;
 	unsigned int i;
 
 	if (!efx_rwsem_assert_write_locked(&efx->filter_sem))
 		return -EINVAL;
 
-	vlan = efx_ef10_filter_find_vlan(efx, vid);
+	vlan = efx_mcdi_filter_find_vlan(efx, vid);
 	if (WARN_ON(vlan)) {
 		netif_err(efx, drv, efx->net_dev,
 			  "VLAN %u already added\n", vid);
@@ -5484,52 +5490,52 @@ static int efx_ef10_filter_add_vlan(struct efx_nic *efx, u16 vid)
 	list_add_tail(&vlan->list, &table->vlan_list);
 
 	if (efx_dev_registered(efx))
-		efx_ef10_filter_vlan_sync_rx_mode(efx, vlan);
+		efx_mcdi_filter_vlan_sync_rx_mode(efx, vlan);
 
 	return 0;
 }
 
-static void efx_ef10_filter_del_vlan_internal(struct efx_nic *efx,
-					      struct efx_ef10_filter_vlan *vlan)
+static void efx_mcdi_filter_del_vlan_internal(struct efx_nic *efx,
+					      struct efx_mcdi_filter_vlan *vlan)
 {
 	unsigned int i;
 
-	/* See comment in efx_ef10_filter_table_remove() */
+	/* See comment in efx_mcdi_filter_table_remove() */
 	if (!efx_rwsem_assert_write_locked(&efx->filter_sem))
 		return;
 
 	list_del(&vlan->list);
 
 	for (i = 0; i < ARRAY_SIZE(vlan->uc); i++)
-		efx_ef10_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO,
+		efx_mcdi_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO,
 					      vlan->uc[i]);
 	for (i = 0; i < ARRAY_SIZE(vlan->mc); i++)
-		efx_ef10_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO,
+		efx_mcdi_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO,
 					      vlan->mc[i]);
 	for (i = 0; i < EFX_EF10_NUM_DEFAULT_FILTERS; i++)
 		if (vlan->default_filters[i] != EFX_EF10_FILTER_ID_INVALID)
-			efx_ef10_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO,
+			efx_mcdi_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO,
 						      vlan->default_filters[i]);
 
 	kfree(vlan);
 }
 
-static void efx_ef10_filter_del_vlan(struct efx_nic *efx, u16 vid)
+static void efx_mcdi_filter_del_vlan(struct efx_nic *efx, u16 vid)
 {
-	struct efx_ef10_filter_vlan *vlan;
+	struct efx_mcdi_filter_vlan *vlan;
 
-	/* See comment in efx_ef10_filter_table_remove() */
+	/* See comment in efx_mcdi_filter_table_remove() */
 	if (!efx_rwsem_assert_write_locked(&efx->filter_sem))
 		return;
 
-	vlan = efx_ef10_filter_find_vlan(efx, vid);
+	vlan = efx_mcdi_filter_find_vlan(efx, vid);
 	if (!vlan) {
 		netif_err(efx, drv, efx->net_dev,
 			  "VLAN %u not found in filter state\n", vid);
 		return;
 	}
 
-	efx_ef10_filter_del_vlan_internal(efx, vlan);
+	efx_mcdi_filter_del_vlan_internal(efx, vlan);
 }
 
 static int efx_ef10_set_mac_address(struct efx_nic *efx)
@@ -5544,7 +5550,7 @@ static int efx_ef10_set_mac_address(struct efx_nic *efx)
 
 	mutex_lock(&efx->mac_lock);
 	down_write(&efx->filter_sem);
-	efx_ef10_filter_table_remove(efx);
+	efx_mcdi_filter_table_remove(efx);
 
 	ether_addr_copy(MCDI_PTR(inbuf, VADAPTOR_SET_MAC_IN_MACADDR),
 			efx->net_dev->dev_addr);
@@ -5553,7 +5559,7 @@ static int efx_ef10_set_mac_address(struct efx_nic *efx)
 	rc = efx_mcdi_rpc_quiet(efx, MC_CMD_VADAPTOR_SET_MAC, inbuf,
 				sizeof(inbuf), NULL, 0, NULL);
 
-	efx_ef10_filter_table_probe(efx);
+	efx_mcdi_filter_table_probe(efx);
 	up_write(&efx->filter_sem);
 	mutex_unlock(&efx->mac_lock);
 
@@ -5615,14 +5621,14 @@ static int efx_ef10_set_mac_address(struct efx_nic *efx)
 
 static int efx_ef10_mac_reconfigure(struct efx_nic *efx)
 {
-	efx_ef10_filter_sync_rx_mode(efx);
+	efx_mcdi_filter_sync_rx_mode(efx);
 
 	return efx_mcdi_set_mac(efx);
 }
 
 static int efx_ef10_mac_reconfigure_vf(struct efx_nic *efx)
 {
-	efx_ef10_filter_sync_rx_mode(efx);
+	efx_mcdi_filter_sync_rx_mode(efx);
 
 	return 0;
 }
@@ -6337,8 +6343,8 @@ const struct efx_nic_type efx_hunt_a0_vf_nic_type = {
 	.tx_remove = efx_mcdi_tx_remove,
 	.tx_write = efx_ef10_tx_write,
 	.tx_limit_len = efx_ef10_tx_limit_len,
-	.rx_push_rss_config = efx_ef10_vf_rx_push_rss_config,
-	.rx_pull_rss_config = efx_ef10_rx_pull_rss_config,
+	.rx_push_rss_config = efx_mcdi_vf_rx_push_rss_config,
+	.rx_pull_rss_config = efx_mcdi_rx_pull_rss_config,
 	.rx_probe = efx_mcdi_rx_probe,
 	.rx_init = efx_mcdi_rx_init,
 	.rx_remove = efx_mcdi_rx_remove,
@@ -6351,19 +6357,19 @@ const struct efx_nic_type efx_hunt_a0_vf_nic_type = {
 	.ev_process = efx_ef10_ev_process,
 	.ev_read_ack = efx_ef10_ev_read_ack,
 	.ev_test_generate = efx_ef10_ev_test_generate,
-	.filter_table_probe = efx_ef10_filter_table_probe,
-	.filter_table_restore = efx_ef10_filter_table_restore,
-	.filter_table_remove = efx_ef10_filter_table_remove,
-	.filter_update_rx_scatter = efx_ef10_filter_update_rx_scatter,
-	.filter_insert = efx_ef10_filter_insert,
-	.filter_remove_safe = efx_ef10_filter_remove_safe,
-	.filter_get_safe = efx_ef10_filter_get_safe,
-	.filter_clear_rx = efx_ef10_filter_clear_rx,
-	.filter_count_rx_used = efx_ef10_filter_count_rx_used,
-	.filter_get_rx_id_limit = efx_ef10_filter_get_rx_id_limit,
-	.filter_get_rx_ids = efx_ef10_filter_get_rx_ids,
+	.filter_table_probe = efx_mcdi_filter_table_probe,
+	.filter_table_restore = efx_mcdi_filter_table_restore,
+	.filter_table_remove = efx_mcdi_filter_table_remove,
+	.filter_update_rx_scatter = efx_mcdi_update_rx_scatter,
+	.filter_insert = efx_mcdi_filter_insert,
+	.filter_remove_safe = efx_mcdi_filter_remove_safe,
+	.filter_get_safe = efx_mcdi_filter_get_safe,
+	.filter_clear_rx = efx_mcdi_filter_clear_rx,
+	.filter_count_rx_used = efx_mcdi_filter_count_rx_used,
+	.filter_get_rx_id_limit = efx_mcdi_filter_get_rx_id_limit,
+	.filter_get_rx_ids = efx_mcdi_filter_get_rx_ids,
 #ifdef CONFIG_RFS_ACCEL
-	.filter_rfs_expire_one = efx_ef10_filter_rfs_expire_one,
+	.filter_rfs_expire_one = efx_mcdi_filter_rfs_expire_one,
 #endif
 #ifdef CONFIG_SFC_MTD
 	.mtd_probe = efx_port_dummy_op_int,
@@ -6393,7 +6399,7 @@ const struct efx_nic_type efx_hunt_a0_vf_nic_type = {
 	.timer_period_max = 1 << ERF_DD_EVQ_IND_TIMER_VAL_WIDTH,
 	.offload_features = EF10_OFFLOAD_FEATURES,
 	.mcdi_max_ver = 2,
-	.max_rx_ip_filters = HUNT_FILTER_TBL_ROWS,
+	.max_rx_ip_filters = EFX_MCDI_FILTER_TBL_ROWS,
 	.hwtstamp_filters = 1 << HWTSTAMP_FILTER_NONE |
 			    1 << HWTSTAMP_FILTER_ALL,
 	.rx_hash_key_size = 40,
@@ -6446,11 +6452,11 @@ const struct efx_nic_type efx_hunt_a0_nic_type = {
 	.tx_remove = efx_mcdi_tx_remove,
 	.tx_write = efx_ef10_tx_write,
 	.tx_limit_len = efx_ef10_tx_limit_len,
-	.rx_push_rss_config = efx_ef10_pf_rx_push_rss_config,
-	.rx_pull_rss_config = efx_ef10_rx_pull_rss_config,
-	.rx_push_rss_context_config = efx_ef10_rx_push_rss_context_config,
-	.rx_pull_rss_context_config = efx_ef10_rx_pull_rss_context_config,
-	.rx_restore_rss_contexts = efx_ef10_rx_restore_rss_contexts,
+	.rx_push_rss_config = efx_mcdi_pf_rx_push_rss_config,
+	.rx_pull_rss_config = efx_mcdi_rx_pull_rss_config,
+	.rx_push_rss_context_config = efx_mcdi_rx_push_rss_context_config,
+	.rx_pull_rss_context_config = efx_mcdi_rx_pull_rss_context_config,
+	.rx_restore_rss_contexts = efx_mcdi_rx_restore_rss_contexts,
 	.rx_probe = efx_mcdi_rx_probe,
 	.rx_init = efx_mcdi_rx_init,
 	.rx_remove = efx_mcdi_rx_remove,
@@ -6463,19 +6469,19 @@ const struct efx_nic_type efx_hunt_a0_nic_type = {
 	.ev_process = efx_ef10_ev_process,
 	.ev_read_ack = efx_ef10_ev_read_ack,
 	.ev_test_generate = efx_ef10_ev_test_generate,
-	.filter_table_probe = efx_ef10_filter_table_probe,
-	.filter_table_restore = efx_ef10_filter_table_restore,
-	.filter_table_remove = efx_ef10_filter_table_remove,
-	.filter_update_rx_scatter = efx_ef10_filter_update_rx_scatter,
-	.filter_insert = efx_ef10_filter_insert,
-	.filter_remove_safe = efx_ef10_filter_remove_safe,
-	.filter_get_safe = efx_ef10_filter_get_safe,
-	.filter_clear_rx = efx_ef10_filter_clear_rx,
-	.filter_count_rx_used = efx_ef10_filter_count_rx_used,
-	.filter_get_rx_id_limit = efx_ef10_filter_get_rx_id_limit,
-	.filter_get_rx_ids = efx_ef10_filter_get_rx_ids,
+	.filter_table_probe = efx_mcdi_filter_table_probe,
+	.filter_table_restore = efx_mcdi_filter_table_restore,
+	.filter_table_remove = efx_mcdi_filter_table_remove,
+	.filter_update_rx_scatter = efx_mcdi_update_rx_scatter,
+	.filter_insert = efx_mcdi_filter_insert,
+	.filter_remove_safe = efx_mcdi_filter_remove_safe,
+	.filter_get_safe = efx_mcdi_filter_get_safe,
+	.filter_clear_rx = efx_mcdi_filter_clear_rx,
+	.filter_count_rx_used = efx_mcdi_filter_count_rx_used,
+	.filter_get_rx_id_limit = efx_mcdi_filter_get_rx_id_limit,
+	.filter_get_rx_ids = efx_mcdi_filter_get_rx_ids,
 #ifdef CONFIG_RFS_ACCEL
-	.filter_rfs_expire_one = efx_ef10_filter_rfs_expire_one,
+	.filter_rfs_expire_one = efx_mcdi_filter_rfs_expire_one,
 #endif
 #ifdef CONFIG_SFC_MTD
 	.mtd_probe = efx_ef10_mtd_probe,
@@ -6528,7 +6534,7 @@ const struct efx_nic_type efx_hunt_a0_nic_type = {
 	.timer_period_max = 1 << ERF_DD_EVQ_IND_TIMER_VAL_WIDTH,
 	.offload_features = EF10_OFFLOAD_FEATURES,
 	.mcdi_max_ver = 2,
-	.max_rx_ip_filters = HUNT_FILTER_TBL_ROWS,
+	.max_rx_ip_filters = EFX_MCDI_FILTER_TBL_ROWS,
 	.hwtstamp_filters = 1 << HWTSTAMP_FILTER_NONE |
 			    1 << HWTSTAMP_FILTER_ALL,
 	.rx_hash_key_size = 40,
-- 
cgit v1.2.3-59-g8ed1b


From 00aaf7e56f890b0b5fd4568ad0d82c6c1d89a2d4 Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Mon, 27 Jan 2020 11:13:40 +0000
Subject: sfc: create header for mcdi filtering code

Moved structs, enums, and added function prototypes.

The affected functions are no longer static.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/ef10.c         | 191 +++++++++-----------------------
 drivers/net/ethernet/sfc/mcdi_filters.h | 159 ++++++++++++++++++++++++++
 2 files changed, 211 insertions(+), 139 deletions(-)
 create mode 100644 drivers/net/ethernet/sfc/mcdi_filters.h

diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c
index 648a86ea25e0..4c203f135999 100644
--- a/drivers/net/ethernet/sfc/ef10.c
+++ b/drivers/net/ethernet/sfc/ef10.c
@@ -13,6 +13,7 @@
 #include "mcdi_port_common.h"
 #include "mcdi_functions.h"
 #include "nic.h"
+#include "mcdi_filters.h"
 #include "workarounds.h"
 #include "selftest.h"
 #include "ef10_sriov.h"
@@ -32,102 +33,19 @@ enum {
 /* TODO: this should really be from the mcdi protocol export */
 #define EFX_EF10_MAX_SHARED_RSS_CONTEXT_SIZE 64UL
 
-/* The filter table(s) are managed by firmware and we have write-only
- * access.  When removing filters we must identify them to the
- * firmware by a 64-bit handle, but this is too wide for Linux kernel
- * interfaces (32-bit for RX NFC, 16-bit for RFS).  Also, we need to
- * be able to tell in advance whether a requested insertion will
- * replace an existing filter.  Therefore we maintain a software hash
- * table, which should be at least as large as the hardware hash
- * table.
- *
- * Huntington has a single 8K filter table shared between all filter
- * types and both ports.
- */
-#define EFX_MCDI_FILTER_TBL_ROWS 8192
-
 #define EFX_EF10_FILTER_ID_INVALID 0xffff
 
-#define EFX_EF10_FILTER_DEV_UC_MAX	32
-#define EFX_EF10_FILTER_DEV_MC_MAX	256
-
 /* VLAN list entry */
 struct efx_ef10_vlan {
 	struct list_head list;
 	u16 vid;
 };
 
-enum efx_mcdi_filter_default_filters {
-	EFX_EF10_BCAST,
-	EFX_EF10_UCDEF,
-	EFX_EF10_MCDEF,
-	EFX_EF10_VXLAN4_UCDEF,
-	EFX_EF10_VXLAN4_MCDEF,
-	EFX_EF10_VXLAN6_UCDEF,
-	EFX_EF10_VXLAN6_MCDEF,
-	EFX_EF10_NVGRE4_UCDEF,
-	EFX_EF10_NVGRE4_MCDEF,
-	EFX_EF10_NVGRE6_UCDEF,
-	EFX_EF10_NVGRE6_MCDEF,
-	EFX_EF10_GENEVE4_UCDEF,
-	EFX_EF10_GENEVE4_MCDEF,
-	EFX_EF10_GENEVE6_UCDEF,
-	EFX_EF10_GENEVE6_MCDEF,
-
-	EFX_EF10_NUM_DEFAULT_FILTERS
-};
-
-/* Per-VLAN filters information */
-struct efx_mcdi_filter_vlan {
-	struct list_head list;
-	u16 vid;
-	u16 uc[EFX_EF10_FILTER_DEV_UC_MAX];
-	u16 mc[EFX_EF10_FILTER_DEV_MC_MAX];
-	u16 default_filters[EFX_EF10_NUM_DEFAULT_FILTERS];
-};
-
-struct efx_mcdi_dev_addr {
-	u8 addr[ETH_ALEN];
-};
-
-struct efx_mcdi_filter_table {
-/* The MCDI match masks supported by this fw & hw, in order of priority */
-	u32 rx_match_mcdi_flags[
-		MC_CMD_GET_PARSER_DISP_INFO_OUT_SUPPORTED_MATCHES_MAXNUM * 2];
-	unsigned int rx_match_count;
-
-	struct rw_semaphore lock; /* Protects entries */
-	struct {
-		unsigned long spec;	/* pointer to spec plus flag bits */
-/* AUTO_OLD is used to mark and sweep MAC filters for the device address lists. */
-/* unused flag	1UL */
-#define EFX_EF10_FILTER_FLAG_AUTO_OLD	2UL
-#define EFX_EF10_FILTER_FLAGS		3UL
-		u64 handle;		/* firmware handle */
-	} *entry;
-/* Shadow of net_device address lists, guarded by mac_lock */
-	struct efx_mcdi_dev_addr dev_uc_list[EFX_EF10_FILTER_DEV_UC_MAX];
-	struct efx_mcdi_dev_addr dev_mc_list[EFX_EF10_FILTER_DEV_MC_MAX];
-	int dev_uc_count;
-	int dev_mc_count;
-	bool uc_promisc;
-	bool mc_promisc;
-/* Whether in multicast promiscuous mode when last changed */
-	bool mc_promisc_last;
-	bool mc_overflow; /* Too many MC addrs; should always imply mc_promisc */
-	bool vlan_filter;
-	struct list_head vlan_list;
-};
-
 /* An arbitrary search limit for the software hash table */
 #define EFX_EF10_FILTER_SEARCH_LIMIT 200
 
-static void efx_mcdi_rx_free_indir_table(struct efx_nic *efx);
-static void efx_mcdi_filter_table_remove(struct efx_nic *efx);
-static int efx_mcdi_filter_add_vlan(struct efx_nic *efx, u16 vid);
 static void efx_mcdi_filter_del_vlan_internal(struct efx_nic *efx,
 					      struct efx_mcdi_filter_vlan *vlan);
-static void efx_mcdi_filter_del_vlan(struct efx_nic *efx, u16 vid);
 static int efx_ef10_set_udp_tnl_ports(struct efx_nic *efx, bool unloading);
 
 static u32 efx_ef10_filter_get_unsafe_id(u32 filter_id)
@@ -2512,8 +2430,7 @@ static void efx_ef10_tx_write(struct efx_tx_queue *tx_queue)
 					 RSS_MODE_HASH_ADDRS << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_UDP_IPV6_RSS_MODE_LBN |\
 					 RSS_MODE_HASH_ADDRS << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_OTHER_IPV6_RSS_MODE_LBN)
 
-static int efx_mcdi_get_rss_context_flags(struct efx_nic *efx, u32 context,
-					  u32 *flags)
+int efx_mcdi_get_rss_context_flags(struct efx_nic *efx, u32 context, u32 *flags)
 {
 	/* Firmware had a bug (sfc bug 61952) where it would not actually
 	 * fill in the flags field in the response to MC_CMD_RSS_CONTEXT_GET_FLAGS.
@@ -2559,8 +2476,8 @@ static int efx_mcdi_get_rss_context_flags(struct efx_nic *efx, u32 context,
  * Defaults are 4-tuple for TCP and 2-tuple for UDP and other-IP, so we
  * just need to set the UDP ports flags (for both IP versions).
  */
-static void efx_mcdi_set_rss_context_flags(struct efx_nic *efx,
-					   struct efx_rss_context *ctx)
+void efx_mcdi_set_rss_context_flags(struct efx_nic *efx,
+				    struct efx_rss_context *ctx)
 {
 	MCDI_DECLARE_BUF(inbuf, MC_CMD_RSS_CONTEXT_SET_FLAGS_IN_LEN);
 	u32 flags;
@@ -2681,7 +2598,7 @@ static int efx_mcdi_filter_populate_rss_table(struct efx_nic *efx, u32 context,
 			    sizeof(keybuf), NULL, 0, NULL);
 }
 
-static void efx_mcdi_rx_free_indir_table(struct efx_nic *efx)
+void efx_mcdi_rx_free_indir_table(struct efx_nic *efx)
 {
 	int rc;
 
@@ -2757,10 +2674,10 @@ fail1:
 	return rc;
 }
 
-static int efx_mcdi_rx_push_rss_context_config(struct efx_nic *efx,
-					       struct efx_rss_context *ctx,
-					       const u32 *rx_indir_table,
-					       const u8 *key)
+int efx_mcdi_rx_push_rss_context_config(struct efx_nic *efx,
+					struct efx_rss_context *ctx,
+					const u32 *rx_indir_table,
+					const u8 *key)
 {
 	int rc;
 
@@ -2787,8 +2704,8 @@ static int efx_mcdi_rx_push_rss_context_config(struct efx_nic *efx,
 	return 0;
 }
 
-static int efx_mcdi_rx_pull_rss_context_config(struct efx_nic *efx,
-					       struct efx_rss_context *ctx)
+int efx_mcdi_rx_pull_rss_context_config(struct efx_nic *efx,
+					struct efx_rss_context *ctx)
 {
 	MCDI_DECLARE_BUF(inbuf, MC_CMD_RSS_CONTEXT_GET_TABLE_IN_LEN);
 	MCDI_DECLARE_BUF(tablebuf, MC_CMD_RSS_CONTEXT_GET_TABLE_OUT_LEN);
@@ -2839,7 +2756,7 @@ static int efx_mcdi_rx_pull_rss_context_config(struct efx_nic *efx,
 	return 0;
 }
 
-static int efx_mcdi_rx_pull_rss_config(struct efx_nic *efx)
+int efx_mcdi_rx_pull_rss_config(struct efx_nic *efx)
 {
 	int rc;
 
@@ -2849,7 +2766,7 @@ static int efx_mcdi_rx_pull_rss_config(struct efx_nic *efx)
 	return rc;
 }
 
-static void efx_mcdi_rx_restore_rss_contexts(struct efx_nic *efx)
+void efx_mcdi_rx_restore_rss_contexts(struct efx_nic *efx)
 {
 	struct efx_ef10_nic_data *nic_data = efx->nic_data;
 	struct efx_rss_context *ctx;
@@ -2876,9 +2793,9 @@ static void efx_mcdi_rx_restore_rss_contexts(struct efx_nic *efx)
 	nic_data->must_restore_rss_contexts = false;
 }
 
-static int efx_mcdi_pf_rx_push_rss_config(struct efx_nic *efx, bool user,
-					  const u32 *rx_indir_table,
-					  const u8 *key)
+int efx_mcdi_pf_rx_push_rss_config(struct efx_nic *efx, bool user,
+				   const u32 *rx_indir_table,
+				   const u8 *key)
 {
 	int rc;
 
@@ -2926,11 +2843,11 @@ static int efx_mcdi_pf_rx_push_rss_config(struct efx_nic *efx, bool user,
 	return rc;
 }
 
-static int efx_mcdi_vf_rx_push_rss_config(struct efx_nic *efx, bool user,
-					  const u32 *rx_indir_table
-					  __attribute__ ((unused)),
-					  const u8 *key
-					  __attribute__ ((unused)))
+int efx_mcdi_vf_rx_push_rss_config(struct efx_nic *efx, bool user,
+				   const u32 *rx_indir_table
+				   __attribute__ ((unused)),
+				   const u8 *key
+				   __attribute__ ((unused)))
 {
 	if (user)
 		return -EOPNOTSUPP;
@@ -4196,9 +4113,9 @@ out_unlock:
 	return rc;
 }
 
-static s32 efx_mcdi_filter_insert(struct efx_nic *efx,
-				  struct efx_filter_spec *spec,
-				  bool replace_equal)
+s32 efx_mcdi_filter_insert(struct efx_nic *efx,
+			   struct efx_filter_spec *spec,
+			   bool replace_equal)
 {
 	s32 ret;
 
@@ -4209,11 +4126,6 @@ static s32 efx_mcdi_filter_insert(struct efx_nic *efx,
 	return ret;
 }
 
-static void efx_mcdi_update_rx_scatter(struct efx_nic *efx)
-{
-	/* no need to do anything here on EF10 */
-}
-
 /* Remove a filter.
  * If !by_index, remove by ID
  * If by_index, remove by index
@@ -4296,9 +4208,9 @@ static int efx_mcdi_filter_remove_internal(struct efx_nic *efx,
 	return rc;
 }
 
-static int efx_mcdi_filter_remove_safe(struct efx_nic *efx,
-				       enum efx_filter_priority priority,
-				       u32 filter_id)
+int efx_mcdi_filter_remove_safe(struct efx_nic *efx,
+				enum efx_filter_priority priority,
+				u32 filter_id)
 {
 	struct efx_mcdi_filter_table *table;
 	int rc;
@@ -4329,9 +4241,9 @@ static void efx_mcdi_filter_remove_unsafe(struct efx_nic *efx,
 	up_write(&table->lock);
 }
 
-static int efx_mcdi_filter_get_safe(struct efx_nic *efx,
-				    enum efx_filter_priority priority,
-				    u32 filter_id, struct efx_filter_spec *spec)
+int efx_mcdi_filter_get_safe(struct efx_nic *efx,
+			     enum efx_filter_priority priority,
+			     u32 filter_id, struct efx_filter_spec *spec)
 {
 	unsigned int filter_idx = efx_ef10_filter_get_unsafe_id(filter_id);
 	const struct efx_filter_spec *saved_spec;
@@ -4355,8 +4267,8 @@ static int efx_mcdi_filter_get_safe(struct efx_nic *efx,
 	return rc;
 }
 
-static int efx_mcdi_filter_clear_rx(struct efx_nic *efx,
-				    enum efx_filter_priority priority)
+int efx_mcdi_filter_clear_rx(struct efx_nic *efx,
+			     enum efx_filter_priority priority)
 {
 	struct efx_mcdi_filter_table *table;
 	unsigned int priority_mask;
@@ -4382,8 +4294,8 @@ static int efx_mcdi_filter_clear_rx(struct efx_nic *efx,
 	return rc;
 }
 
-static u32 efx_mcdi_filter_count_rx_used(struct efx_nic *efx,
-					 enum efx_filter_priority priority)
+u32 efx_mcdi_filter_count_rx_used(struct efx_nic *efx,
+				  enum efx_filter_priority priority)
 {
 	struct efx_mcdi_filter_table *table;
 	unsigned int filter_idx;
@@ -4403,16 +4315,16 @@ static u32 efx_mcdi_filter_count_rx_used(struct efx_nic *efx,
 	return count;
 }
 
-static u32 efx_mcdi_filter_get_rx_id_limit(struct efx_nic *efx)
+u32 efx_mcdi_filter_get_rx_id_limit(struct efx_nic *efx)
 {
 	struct efx_mcdi_filter_table *table = efx->filter_state;
 
 	return table->rx_match_count * EFX_MCDI_FILTER_TBL_ROWS * 2;
 }
 
-static s32 efx_mcdi_filter_get_rx_ids(struct efx_nic *efx,
-				      enum efx_filter_priority priority,
-				      u32 *buf, u32 size)
+s32 efx_mcdi_filter_get_rx_ids(struct efx_nic *efx,
+			       enum efx_filter_priority priority,
+			       u32 *buf, u32 size)
 {
 	struct efx_mcdi_filter_table *table;
 	struct efx_filter_spec *spec;
@@ -4443,8 +4355,8 @@ static s32 efx_mcdi_filter_get_rx_ids(struct efx_nic *efx,
 
 #ifdef CONFIG_RFS_ACCEL
 
-static bool efx_mcdi_filter_rfs_expire_one(struct efx_nic *efx, u32 flow_id,
-					   unsigned int filter_idx)
+bool efx_mcdi_filter_rfs_expire_one(struct efx_nic *efx, u32 flow_id,
+				    unsigned int filter_idx)
 {
 	struct efx_filter_spec *spec, saved_spec;
 	struct efx_mcdi_filter_table *table;
@@ -4568,7 +4480,7 @@ static int efx_mcdi_filter_match_flags_from_mcdi(bool encap, u32 mcdi_flags)
 	return match_flags;
 }
 
-static void efx_mcdi_filter_cleanup_vlans(struct efx_nic *efx)
+void efx_mcdi_filter_cleanup_vlans(struct efx_nic *efx)
 {
 	struct efx_mcdi_filter_table *table = efx->filter_state;
 	struct efx_mcdi_filter_vlan *vlan, *next_vlan;
@@ -4584,9 +4496,9 @@ static void efx_mcdi_filter_cleanup_vlans(struct efx_nic *efx)
 		efx_mcdi_filter_del_vlan_internal(efx, vlan);
 }
 
-static bool efx_mcdi_filter_match_supported(struct efx_mcdi_filter_table *table,
-					    bool encap,
-					    enum efx_filter_match_flags match_flags)
+bool efx_mcdi_filter_match_supported(struct efx_mcdi_filter_table *table,
+				     bool encap,
+				     enum efx_filter_match_flags match_flags)
 {
 	unsigned int match_pri;
 	int mf;
@@ -4652,7 +4564,7 @@ efx_mcdi_filter_table_probe_matches(struct efx_nic *efx,
 	return 0;
 }
 
-static int efx_mcdi_filter_table_probe(struct efx_nic *efx)
+int efx_mcdi_filter_table_probe(struct efx_nic *efx)
 {
 	struct efx_ef10_nic_data *nic_data = efx->nic_data;
 	struct net_device *net_dev = efx->net_dev;
@@ -4725,7 +4637,7 @@ fail:
 /* Caller must hold efx->filter_sem for read if race against
  * efx_mcdi_filter_table_remove() is possible
  */
-static void efx_mcdi_filter_table_restore(struct efx_nic *efx)
+void efx_mcdi_filter_table_restore(struct efx_nic *efx)
 {
 	struct efx_mcdi_filter_table *table = efx->filter_state;
 	struct efx_ef10_nic_data *nic_data = efx->nic_data;
@@ -4821,7 +4733,7 @@ not_restored:
 		nic_data->must_restore_filters = false;
 }
 
-static void efx_mcdi_filter_table_remove(struct efx_nic *efx)
+void efx_mcdi_filter_table_remove(struct efx_nic *efx)
 {
 	struct efx_mcdi_filter_table *table = efx->filter_state;
 	MCDI_DECLARE_BUF(inbuf, MC_CMD_FILTER_OP_EXT_IN_LEN);
@@ -5403,7 +5315,7 @@ static void efx_mcdi_filter_vlan_sync_rx_mode(struct efx_nic *efx,
 /* Caller must hold efx->filter_sem for read if race against
  * efx_mcdi_filter_table_remove() is possible
  */
-static void efx_mcdi_filter_sync_rx_mode(struct efx_nic *efx)
+void efx_mcdi_filter_sync_rx_mode(struct efx_nic *efx)
 {
 	struct efx_mcdi_filter_table *table = efx->filter_state;
 	struct net_device *net_dev = efx->net_dev;
@@ -5443,7 +5355,8 @@ static void efx_mcdi_filter_sync_rx_mode(struct efx_nic *efx)
 	table->mc_promisc_last = table->mc_promisc;
 }
 
-static struct efx_mcdi_filter_vlan *efx_mcdi_filter_find_vlan(struct efx_nic *efx, u16 vid)
+struct efx_mcdi_filter_vlan *efx_mcdi_filter_find_vlan(struct efx_nic *efx,
+						       u16 vid)
 {
 	struct efx_mcdi_filter_table *table = efx->filter_state;
 	struct efx_mcdi_filter_vlan *vlan;
@@ -5458,7 +5371,7 @@ static struct efx_mcdi_filter_vlan *efx_mcdi_filter_find_vlan(struct efx_nic *ef
 	return NULL;
 }
 
-static int efx_mcdi_filter_add_vlan(struct efx_nic *efx, u16 vid)
+int efx_mcdi_filter_add_vlan(struct efx_nic *efx, u16 vid)
 {
 	struct efx_mcdi_filter_table *table = efx->filter_state;
 	struct efx_mcdi_filter_vlan *vlan;
@@ -5520,7 +5433,7 @@ static void efx_mcdi_filter_del_vlan_internal(struct efx_nic *efx,
 	kfree(vlan);
 }
 
-static void efx_mcdi_filter_del_vlan(struct efx_nic *efx, u16 vid)
+void efx_mcdi_filter_del_vlan(struct efx_nic *efx, u16 vid)
 {
 	struct efx_mcdi_filter_vlan *vlan;
 
diff --git a/drivers/net/ethernet/sfc/mcdi_filters.h b/drivers/net/ethernet/sfc/mcdi_filters.h
new file mode 100644
index 000000000000..1837f4f5d661
--- /dev/null
+++ b/drivers/net/ethernet/sfc/mcdi_filters.h
@@ -0,0 +1,159 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/****************************************************************************
+ * Driver for Solarflare network controllers and boards
+ * Copyright 2019 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+#ifndef EFX_MCDI_FILTERS_H
+#define EFX_MCDI_FILTERS_H
+
+#include "net_driver.h"
+#include "filter.h"
+#include "mcdi_pcol.h"
+
+#define EFX_EF10_FILTER_DEV_UC_MAX	32
+#define EFX_EF10_FILTER_DEV_MC_MAX	256
+
+enum efx_mcdi_filter_default_filters {
+	EFX_EF10_BCAST,
+	EFX_EF10_UCDEF,
+	EFX_EF10_MCDEF,
+	EFX_EF10_VXLAN4_UCDEF,
+	EFX_EF10_VXLAN4_MCDEF,
+	EFX_EF10_VXLAN6_UCDEF,
+	EFX_EF10_VXLAN6_MCDEF,
+	EFX_EF10_NVGRE4_UCDEF,
+	EFX_EF10_NVGRE4_MCDEF,
+	EFX_EF10_NVGRE6_UCDEF,
+	EFX_EF10_NVGRE6_MCDEF,
+	EFX_EF10_GENEVE4_UCDEF,
+	EFX_EF10_GENEVE4_MCDEF,
+	EFX_EF10_GENEVE6_UCDEF,
+	EFX_EF10_GENEVE6_MCDEF,
+
+	EFX_EF10_NUM_DEFAULT_FILTERS
+};
+
+/* Per-VLAN filters information */
+struct efx_mcdi_filter_vlan {
+	struct list_head list;
+	u16 vid;
+	u16 uc[EFX_EF10_FILTER_DEV_UC_MAX];
+	u16 mc[EFX_EF10_FILTER_DEV_MC_MAX];
+	u16 default_filters[EFX_EF10_NUM_DEFAULT_FILTERS];
+};
+
+struct efx_mcdi_dev_addr {
+	u8 addr[ETH_ALEN];
+};
+
+struct efx_mcdi_filter_table {
+/* The MCDI match masks supported by this fw & hw, in order of priority */
+	u32 rx_match_mcdi_flags[
+		MC_CMD_GET_PARSER_DISP_INFO_OUT_SUPPORTED_MATCHES_MAXNUM * 2];
+	unsigned int rx_match_count;
+
+	struct rw_semaphore lock; /* Protects entries */
+	struct {
+		unsigned long spec;	/* pointer to spec plus flag bits */
+/* AUTO_OLD is used to mark and sweep MAC filters for the device address lists. */
+/* unused flag	1UL */
+#define EFX_EF10_FILTER_FLAG_AUTO_OLD	2UL
+#define EFX_EF10_FILTER_FLAGS		3UL
+		u64 handle;		/* firmware handle */
+	} *entry;
+/* Shadow of net_device address lists, guarded by mac_lock */
+	struct efx_mcdi_dev_addr dev_uc_list[EFX_EF10_FILTER_DEV_UC_MAX];
+	struct efx_mcdi_dev_addr dev_mc_list[EFX_EF10_FILTER_DEV_MC_MAX];
+	int dev_uc_count;
+	int dev_mc_count;
+	bool uc_promisc;
+	bool mc_promisc;
+/* Whether in multicast promiscuous mode when last changed */
+	bool mc_promisc_last;
+	bool mc_overflow; /* Too many MC addrs; should always imply mc_promisc */
+	bool vlan_filter;
+	struct list_head vlan_list;
+};
+
+int efx_mcdi_filter_table_probe(struct efx_nic *efx);
+void efx_mcdi_filter_table_remove(struct efx_nic *efx);
+void efx_mcdi_filter_table_restore(struct efx_nic *efx);
+
+/*
+ * The filter table(s) are managed by firmware and we have write-only
+ * access.  When removing filters we must identify them to the
+ * firmware by a 64-bit handle, but this is too wide for Linux kernel
+ * interfaces (32-bit for RX NFC, 16-bit for RFS).  Also, we need to
+ * be able to tell in advance whether a requested insertion will
+ * replace an existing filter.  Therefore we maintain a software hash
+ * table, which should be at least as large as the hardware hash
+ * table.
+ *
+ * Huntington has a single 8K filter table shared between all filter
+ * types and both ports.
+ */
+#define EFX_MCDI_FILTER_TBL_ROWS 8192
+
+bool efx_mcdi_filter_match_supported(struct efx_mcdi_filter_table *table,
+				     bool encap,
+				     enum efx_filter_match_flags match_flags);
+
+void efx_mcdi_filter_sync_rx_mode(struct efx_nic *efx);
+s32 efx_mcdi_filter_insert(struct efx_nic *efx, struct efx_filter_spec *spec,
+			   bool replace_equal);
+int efx_mcdi_filter_remove_safe(struct efx_nic *efx,
+				enum efx_filter_priority priority,
+				u32 filter_id);
+int efx_mcdi_filter_get_safe(struct efx_nic *efx,
+			     enum efx_filter_priority priority,
+			     u32 filter_id, struct efx_filter_spec *spec);
+
+u32 efx_mcdi_filter_count_rx_used(struct efx_nic *efx,
+				  enum efx_filter_priority priority);
+int efx_mcdi_filter_clear_rx(struct efx_nic *efx,
+			     enum efx_filter_priority priority);
+u32 efx_mcdi_filter_get_rx_id_limit(struct efx_nic *efx);
+s32 efx_mcdi_filter_get_rx_ids(struct efx_nic *efx,
+			       enum efx_filter_priority priority,
+			       u32 *buf, u32 size);
+
+void efx_mcdi_filter_cleanup_vlans(struct efx_nic *efx);
+int efx_mcdi_filter_add_vlan(struct efx_nic *efx, u16 vid);
+struct efx_mcdi_filter_vlan *efx_mcdi_filter_find_vlan(struct efx_nic *efx, u16 vid);
+void efx_mcdi_filter_del_vlan(struct efx_nic *efx, u16 vid);
+
+void efx_mcdi_rx_free_indir_table(struct efx_nic *efx);
+int efx_mcdi_rx_push_rss_context_config(struct efx_nic *efx,
+					struct efx_rss_context *ctx,
+					const u32 *rx_indir_table,
+					const u8 *key);
+int efx_mcdi_pf_rx_push_rss_config(struct efx_nic *efx, bool user,
+				   const u32 *rx_indir_table,
+				   const u8 *key);
+int efx_mcdi_vf_rx_push_rss_config(struct efx_nic *efx, bool user,
+				   const u32 *rx_indir_table
+				   __attribute__ ((unused)),
+				   const u8 *key
+				   __attribute__ ((unused)));
+int efx_mcdi_rx_pull_rss_config(struct efx_nic *efx);
+int efx_mcdi_rx_pull_rss_context_config(struct efx_nic *efx,
+					struct efx_rss_context *ctx);
+int efx_mcdi_get_rss_context_flags(struct efx_nic *efx, u32 context,
+				   u32 *flags);
+void efx_mcdi_set_rss_context_flags(struct efx_nic *efx,
+				    struct efx_rss_context *ctx);
+void efx_mcdi_rx_restore_rss_contexts(struct efx_nic *efx);
+
+static inline void efx_mcdi_update_rx_scatter(struct efx_nic *efx)
+{
+	/* no need to do anything here */
+}
+
+bool efx_mcdi_filter_rfs_expire_one(struct efx_nic *efx, u32 flow_id,
+				    unsigned int filter_idx);
+
+#endif
-- 
cgit v1.2.3-59-g8ed1b


From 6c77065bbdff817cea4c391720b120d4dd81e1c8 Mon Sep 17 00:00:00 2001
From: "Alex Maftei (amaftei)" <amaftei@solarflare.com>
Date: Mon, 27 Jan 2020 11:13:55 +0000
Subject: sfc: move mcdi filtering code

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/Makefile       |    2 +-
 drivers/net/ethernet/sfc/ef10.c         | 2333 +------------------------------
 drivers/net/ethernet/sfc/mcdi_filters.c | 2270 ++++++++++++++++++++++++++++++
 3 files changed, 2315 insertions(+), 2290 deletions(-)
 create mode 100644 drivers/net/ethernet/sfc/mcdi_filters.c

diff --git a/drivers/net/ethernet/sfc/Makefile b/drivers/net/ethernet/sfc/Makefile
index 890fd65caa2d..87d093da22ca 100644
--- a/drivers/net/ethernet/sfc/Makefile
+++ b/drivers/net/ethernet/sfc/Makefile
@@ -4,7 +4,7 @@ sfc-y			+= efx.o efx_common.o efx_channels.o nic.o \
 			   tx.o tx_common.o tx_tso.o rx.o rx_common.o \
 			   selftest.o ethtool.o ethtool_common.o ptp.o \
 			   mcdi.o mcdi_port.o mcdi_port_common.o \
-			   mcdi_functions.o mcdi_mon.o
+			   mcdi_functions.o mcdi_filters.o mcdi_mon.o
 sfc-$(CONFIG_SFC_MTD)	+= mtd.o
 sfc-$(CONFIG_SFC_SRIOV)	+= sriov.o siena_sriov.o ef10_sriov.o
 
diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c
index 4c203f135999..397f3bc7da3b 100644
--- a/drivers/net/ethernet/sfc/ef10.c
+++ b/drivers/net/ethernet/sfc/ef10.c
@@ -29,11 +29,6 @@ enum {
 	EFX_EF10_TEST = 1,
 	EFX_EF10_REFILL,
 };
-/* The maximum size of a shared RSS context */
-/* TODO: this should really be from the mcdi protocol export */
-#define EFX_EF10_MAX_SHARED_RSS_CONTEXT_SIZE 64UL
-
-#define EFX_EF10_FILTER_ID_INVALID 0xffff
 
 /* VLAN list entry */
 struct efx_ef10_vlan {
@@ -41,29 +36,8 @@ struct efx_ef10_vlan {
 	u16 vid;
 };
 
-/* An arbitrary search limit for the software hash table */
-#define EFX_EF10_FILTER_SEARCH_LIMIT 200
-
-static void efx_mcdi_filter_del_vlan_internal(struct efx_nic *efx,
-					      struct efx_mcdi_filter_vlan *vlan);
 static int efx_ef10_set_udp_tnl_ports(struct efx_nic *efx, bool unloading);
 
-static u32 efx_ef10_filter_get_unsafe_id(u32 filter_id)
-{
-	WARN_ON_ONCE(filter_id == EFX_EF10_FILTER_ID_INVALID);
-	return filter_id & (EFX_MCDI_FILTER_TBL_ROWS - 1);
-}
-
-static unsigned int efx_ef10_filter_get_unsafe_pri(u32 filter_id)
-{
-	return filter_id / (EFX_MCDI_FILTER_TBL_ROWS * 2);
-}
-
-static u32 efx_ef10_make_filter_id(unsigned int pri, u16 idx)
-{
-	return pri * EFX_MCDI_FILTER_TBL_ROWS * 2 + idx;
-}
-
 static int efx_ef10_get_warm_boot_count(struct efx_nic *efx)
 {
 	efx_dword_t reg;
@@ -2415,447 +2389,6 @@ static void efx_ef10_tx_write(struct efx_tx_queue *tx_queue)
 	}
 }
 
-#define RSS_MODE_HASH_ADDRS	(1 << RSS_MODE_HASH_SRC_ADDR_LBN |\
-				 1 << RSS_MODE_HASH_DST_ADDR_LBN)
-#define RSS_MODE_HASH_PORTS	(1 << RSS_MODE_HASH_SRC_PORT_LBN |\
-				 1 << RSS_MODE_HASH_DST_PORT_LBN)
-#define RSS_CONTEXT_FLAGS_DEFAULT	(1 << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_TOEPLITZ_IPV4_EN_LBN |\
-					 1 << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_TOEPLITZ_TCPV4_EN_LBN |\
-					 1 << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_TOEPLITZ_IPV6_EN_LBN |\
-					 1 << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_TOEPLITZ_TCPV6_EN_LBN |\
-					 (RSS_MODE_HASH_ADDRS | RSS_MODE_HASH_PORTS) << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_TCP_IPV4_RSS_MODE_LBN |\
-					 RSS_MODE_HASH_ADDRS << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_UDP_IPV4_RSS_MODE_LBN |\
-					 RSS_MODE_HASH_ADDRS << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_OTHER_IPV4_RSS_MODE_LBN |\
-					 (RSS_MODE_HASH_ADDRS | RSS_MODE_HASH_PORTS) << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_TCP_IPV6_RSS_MODE_LBN |\
-					 RSS_MODE_HASH_ADDRS << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_UDP_IPV6_RSS_MODE_LBN |\
-					 RSS_MODE_HASH_ADDRS << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_OTHER_IPV6_RSS_MODE_LBN)
-
-int efx_mcdi_get_rss_context_flags(struct efx_nic *efx, u32 context, u32 *flags)
-{
-	/* Firmware had a bug (sfc bug 61952) where it would not actually
-	 * fill in the flags field in the response to MC_CMD_RSS_CONTEXT_GET_FLAGS.
-	 * This meant that it would always contain whatever was previously
-	 * in the MCDI buffer.  Fortunately, all firmware versions with
-	 * this bug have the same default flags value for a newly-allocated
-	 * RSS context, and the only time we want to get the flags is just
-	 * after allocating.  Moreover, the response has a 32-bit hole
-	 * where the context ID would be in the request, so we can use an
-	 * overlength buffer in the request and pre-fill the flags field
-	 * with what we believe the default to be.  Thus if the firmware
-	 * has the bug, it will leave our pre-filled value in the flags
-	 * field of the response, and we will get the right answer.
-	 *
-	 * However, this does mean that this function should NOT be used if
-	 * the RSS context flags might not be their defaults - it is ONLY
-	 * reliably correct for a newly-allocated RSS context.
-	 */
-	MCDI_DECLARE_BUF(inbuf, MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_LEN);
-	MCDI_DECLARE_BUF(outbuf, MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_LEN);
-	size_t outlen;
-	int rc;
-
-	/* Check we have a hole for the context ID */
-	BUILD_BUG_ON(MC_CMD_RSS_CONTEXT_GET_FLAGS_IN_LEN != MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_FLAGS_OFST);
-	MCDI_SET_DWORD(inbuf, RSS_CONTEXT_GET_FLAGS_IN_RSS_CONTEXT_ID, context);
-	MCDI_SET_DWORD(inbuf, RSS_CONTEXT_GET_FLAGS_OUT_FLAGS,
-		       RSS_CONTEXT_FLAGS_DEFAULT);
-	rc = efx_mcdi_rpc(efx, MC_CMD_RSS_CONTEXT_GET_FLAGS, inbuf,
-			  sizeof(inbuf), outbuf, sizeof(outbuf), &outlen);
-	if (rc == 0) {
-		if (outlen < MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_LEN)
-			rc = -EIO;
-		else
-			*flags = MCDI_DWORD(outbuf, RSS_CONTEXT_GET_FLAGS_OUT_FLAGS);
-	}
-	return rc;
-}
-
-/* Attempt to enable 4-tuple UDP hashing on the specified RSS context.
- * If we fail, we just leave the RSS context at its default hash settings,
- * which is safe but may slightly reduce performance.
- * Defaults are 4-tuple for TCP and 2-tuple for UDP and other-IP, so we
- * just need to set the UDP ports flags (for both IP versions).
- */
-void efx_mcdi_set_rss_context_flags(struct efx_nic *efx,
-				    struct efx_rss_context *ctx)
-{
-	MCDI_DECLARE_BUF(inbuf, MC_CMD_RSS_CONTEXT_SET_FLAGS_IN_LEN);
-	u32 flags;
-
-	BUILD_BUG_ON(MC_CMD_RSS_CONTEXT_SET_FLAGS_OUT_LEN != 0);
-
-	if (efx_mcdi_get_rss_context_flags(efx, ctx->context_id, &flags) != 0)
-		return;
-	MCDI_SET_DWORD(inbuf, RSS_CONTEXT_SET_FLAGS_IN_RSS_CONTEXT_ID,
-		       ctx->context_id);
-	flags |= RSS_MODE_HASH_PORTS << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_UDP_IPV4_RSS_MODE_LBN;
-	flags |= RSS_MODE_HASH_PORTS << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_UDP_IPV6_RSS_MODE_LBN;
-	MCDI_SET_DWORD(inbuf, RSS_CONTEXT_SET_FLAGS_IN_FLAGS, flags);
-	if (!efx_mcdi_rpc(efx, MC_CMD_RSS_CONTEXT_SET_FLAGS, inbuf, sizeof(inbuf),
-			  NULL, 0, NULL))
-		/* Succeeded, so UDP 4-tuple is now enabled */
-		ctx->rx_hash_udp_4tuple = true;
-}
-
-static int efx_mcdi_filter_alloc_rss_context(struct efx_nic *efx, bool exclusive,
-					     struct efx_rss_context *ctx,
-					     unsigned *context_size)
-{
-	MCDI_DECLARE_BUF(inbuf, MC_CMD_RSS_CONTEXT_ALLOC_IN_LEN);
-	MCDI_DECLARE_BUF(outbuf, MC_CMD_RSS_CONTEXT_ALLOC_OUT_LEN);
-	struct efx_ef10_nic_data *nic_data = efx->nic_data;
-	size_t outlen;
-	int rc;
-	u32 alloc_type = exclusive ?
-				MC_CMD_RSS_CONTEXT_ALLOC_IN_TYPE_EXCLUSIVE :
-				MC_CMD_RSS_CONTEXT_ALLOC_IN_TYPE_SHARED;
-	unsigned rss_spread = exclusive ?
-				efx->rss_spread :
-				min(rounddown_pow_of_two(efx->rss_spread),
-				    EFX_EF10_MAX_SHARED_RSS_CONTEXT_SIZE);
-
-	if (!exclusive && rss_spread == 1) {
-		ctx->context_id = EFX_MCDI_RSS_CONTEXT_INVALID;
-		if (context_size)
-			*context_size = 1;
-		return 0;
-	}
-
-	if (nic_data->datapath_caps &
-	    1 << MC_CMD_GET_CAPABILITIES_OUT_RX_RSS_LIMITED_LBN)
-		return -EOPNOTSUPP;
-
-	MCDI_SET_DWORD(inbuf, RSS_CONTEXT_ALLOC_IN_UPSTREAM_PORT_ID,
-		       nic_data->vport_id);
-	MCDI_SET_DWORD(inbuf, RSS_CONTEXT_ALLOC_IN_TYPE, alloc_type);
-	MCDI_SET_DWORD(inbuf, RSS_CONTEXT_ALLOC_IN_NUM_QUEUES, rss_spread);
-
-	rc = efx_mcdi_rpc(efx, MC_CMD_RSS_CONTEXT_ALLOC, inbuf, sizeof(inbuf),
-		outbuf, sizeof(outbuf), &outlen);
-	if (rc != 0)
-		return rc;
-
-	if (outlen < MC_CMD_RSS_CONTEXT_ALLOC_OUT_LEN)
-		return -EIO;
-
-	ctx->context_id = MCDI_DWORD(outbuf, RSS_CONTEXT_ALLOC_OUT_RSS_CONTEXT_ID);
-
-	if (context_size)
-		*context_size = rss_spread;
-
-	if (nic_data->datapath_caps &
-	    1 << MC_CMD_GET_CAPABILITIES_OUT_ADDITIONAL_RSS_MODES_LBN)
-		efx_mcdi_set_rss_context_flags(efx, ctx);
-
-	return 0;
-}
-
-static int efx_mcdi_filter_free_rss_context(struct efx_nic *efx, u32 context)
-{
-	MCDI_DECLARE_BUF(inbuf, MC_CMD_RSS_CONTEXT_FREE_IN_LEN);
-
-	MCDI_SET_DWORD(inbuf, RSS_CONTEXT_FREE_IN_RSS_CONTEXT_ID,
-		       context);
-	return efx_mcdi_rpc(efx, MC_CMD_RSS_CONTEXT_FREE, inbuf, sizeof(inbuf),
-			    NULL, 0, NULL);
-}
-
-static int efx_mcdi_filter_populate_rss_table(struct efx_nic *efx, u32 context,
-					      const u32 *rx_indir_table, const u8 *key)
-{
-	MCDI_DECLARE_BUF(tablebuf, MC_CMD_RSS_CONTEXT_SET_TABLE_IN_LEN);
-	MCDI_DECLARE_BUF(keybuf, MC_CMD_RSS_CONTEXT_SET_KEY_IN_LEN);
-	int i, rc;
-
-	MCDI_SET_DWORD(tablebuf, RSS_CONTEXT_SET_TABLE_IN_RSS_CONTEXT_ID,
-		       context);
-	BUILD_BUG_ON(ARRAY_SIZE(efx->rss_context.rx_indir_table) !=
-		     MC_CMD_RSS_CONTEXT_SET_TABLE_IN_INDIRECTION_TABLE_LEN);
-
-	/* This iterates over the length of efx->rss_context.rx_indir_table, but
-	 * copies bytes from rx_indir_table.  That's because the latter is a
-	 * pointer rather than an array, but should have the same length.
-	 * The efx->rss_context.rx_hash_key loop below is similar.
-	 */
-	for (i = 0; i < ARRAY_SIZE(efx->rss_context.rx_indir_table); ++i)
-		MCDI_PTR(tablebuf,
-			 RSS_CONTEXT_SET_TABLE_IN_INDIRECTION_TABLE)[i] =
-				(u8) rx_indir_table[i];
-
-	rc = efx_mcdi_rpc(efx, MC_CMD_RSS_CONTEXT_SET_TABLE, tablebuf,
-			  sizeof(tablebuf), NULL, 0, NULL);
-	if (rc != 0)
-		return rc;
-
-	MCDI_SET_DWORD(keybuf, RSS_CONTEXT_SET_KEY_IN_RSS_CONTEXT_ID,
-		       context);
-	BUILD_BUG_ON(ARRAY_SIZE(efx->rss_context.rx_hash_key) !=
-		     MC_CMD_RSS_CONTEXT_SET_KEY_IN_TOEPLITZ_KEY_LEN);
-	for (i = 0; i < ARRAY_SIZE(efx->rss_context.rx_hash_key); ++i)
-		MCDI_PTR(keybuf, RSS_CONTEXT_SET_KEY_IN_TOEPLITZ_KEY)[i] = key[i];
-
-	return efx_mcdi_rpc(efx, MC_CMD_RSS_CONTEXT_SET_KEY, keybuf,
-			    sizeof(keybuf), NULL, 0, NULL);
-}
-
-void efx_mcdi_rx_free_indir_table(struct efx_nic *efx)
-{
-	int rc;
-
-	if (efx->rss_context.context_id != EFX_MCDI_RSS_CONTEXT_INVALID) {
-		rc = efx_mcdi_filter_free_rss_context(efx,
-						   efx->rss_context.context_id);
-		WARN_ON(rc != 0);
-	}
-	efx->rss_context.context_id = EFX_MCDI_RSS_CONTEXT_INVALID;
-}
-
-static int efx_mcdi_filter_rx_push_shared_rss_config(struct efx_nic *efx,
-						     unsigned *context_size)
-{
-	struct efx_ef10_nic_data *nic_data = efx->nic_data;
-	int rc = efx_mcdi_filter_alloc_rss_context(efx, false,
-						   &efx->rss_context,
-						   context_size);
-
-	if (rc != 0)
-		return rc;
-
-	nic_data->rx_rss_context_exclusive = false;
-	efx_set_default_rx_indir_table(efx, &efx->rss_context);
-	return 0;
-}
-
-static int efx_mcdi_filter_rx_push_exclusive_rss_config(struct efx_nic *efx,
-						        const u32 *rx_indir_table,
-						        const u8 *key)
-{
-	u32 old_rx_rss_context = efx->rss_context.context_id;
-	struct efx_ef10_nic_data *nic_data = efx->nic_data;
-	int rc;
-
-	if (efx->rss_context.context_id == EFX_MCDI_RSS_CONTEXT_INVALID ||
-	    !nic_data->rx_rss_context_exclusive) {
-		rc = efx_mcdi_filter_alloc_rss_context(efx, true,
-						       &efx->rss_context,
-						       NULL);
-		if (rc == -EOPNOTSUPP)
-			return rc;
-		else if (rc != 0)
-			goto fail1;
-	}
-
-	rc = efx_mcdi_filter_populate_rss_table(efx,
-						efx->rss_context.context_id,
-						rx_indir_table, key);
-	if (rc != 0)
-		goto fail2;
-
-	if (efx->rss_context.context_id != old_rx_rss_context &&
-	    old_rx_rss_context != EFX_MCDI_RSS_CONTEXT_INVALID)
-		WARN_ON(efx_mcdi_filter_free_rss_context(efx, old_rx_rss_context) != 0);
-	nic_data->rx_rss_context_exclusive = true;
-	if (rx_indir_table != efx->rss_context.rx_indir_table)
-		memcpy(efx->rss_context.rx_indir_table, rx_indir_table,
-		       sizeof(efx->rss_context.rx_indir_table));
-	if (key != efx->rss_context.rx_hash_key)
-		memcpy(efx->rss_context.rx_hash_key, key,
-		       efx->type->rx_hash_key_size);
-
-	return 0;
-
-fail2:
-	if (old_rx_rss_context != efx->rss_context.context_id) {
-		WARN_ON(efx_mcdi_filter_free_rss_context(efx, efx->rss_context.context_id) != 0);
-		efx->rss_context.context_id = old_rx_rss_context;
-	}
-fail1:
-	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
-	return rc;
-}
-
-int efx_mcdi_rx_push_rss_context_config(struct efx_nic *efx,
-					struct efx_rss_context *ctx,
-					const u32 *rx_indir_table,
-					const u8 *key)
-{
-	int rc;
-
-	WARN_ON(!mutex_is_locked(&efx->rss_lock));
-
-	if (ctx->context_id == EFX_MCDI_RSS_CONTEXT_INVALID) {
-		rc = efx_mcdi_filter_alloc_rss_context(efx, true, ctx, NULL);
-		if (rc)
-			return rc;
-	}
-
-	if (!rx_indir_table) /* Delete this context */
-		return efx_mcdi_filter_free_rss_context(efx, ctx->context_id);
-
-	rc = efx_mcdi_filter_populate_rss_table(efx, ctx->context_id,
-						rx_indir_table, key);
-	if (rc)
-		return rc;
-
-	memcpy(ctx->rx_indir_table, rx_indir_table,
-	       sizeof(efx->rss_context.rx_indir_table));
-	memcpy(ctx->rx_hash_key, key, efx->type->rx_hash_key_size);
-
-	return 0;
-}
-
-int efx_mcdi_rx_pull_rss_context_config(struct efx_nic *efx,
-					struct efx_rss_context *ctx)
-{
-	MCDI_DECLARE_BUF(inbuf, MC_CMD_RSS_CONTEXT_GET_TABLE_IN_LEN);
-	MCDI_DECLARE_BUF(tablebuf, MC_CMD_RSS_CONTEXT_GET_TABLE_OUT_LEN);
-	MCDI_DECLARE_BUF(keybuf, MC_CMD_RSS_CONTEXT_GET_KEY_OUT_LEN);
-	size_t outlen;
-	int rc, i;
-
-	WARN_ON(!mutex_is_locked(&efx->rss_lock));
-
-	BUILD_BUG_ON(MC_CMD_RSS_CONTEXT_GET_TABLE_IN_LEN !=
-		     MC_CMD_RSS_CONTEXT_GET_KEY_IN_LEN);
-
-	if (ctx->context_id == EFX_MCDI_RSS_CONTEXT_INVALID)
-		return -ENOENT;
-
-	MCDI_SET_DWORD(inbuf, RSS_CONTEXT_GET_TABLE_IN_RSS_CONTEXT_ID,
-		       ctx->context_id);
-	BUILD_BUG_ON(ARRAY_SIZE(ctx->rx_indir_table) !=
-		     MC_CMD_RSS_CONTEXT_GET_TABLE_OUT_INDIRECTION_TABLE_LEN);
-	rc = efx_mcdi_rpc(efx, MC_CMD_RSS_CONTEXT_GET_TABLE, inbuf, sizeof(inbuf),
-			  tablebuf, sizeof(tablebuf), &outlen);
-	if (rc != 0)
-		return rc;
-
-	if (WARN_ON(outlen != MC_CMD_RSS_CONTEXT_GET_TABLE_OUT_LEN))
-		return -EIO;
-
-	for (i = 0; i < ARRAY_SIZE(ctx->rx_indir_table); i++)
-		ctx->rx_indir_table[i] = MCDI_PTR(tablebuf,
-				RSS_CONTEXT_GET_TABLE_OUT_INDIRECTION_TABLE)[i];
-
-	MCDI_SET_DWORD(inbuf, RSS_CONTEXT_GET_KEY_IN_RSS_CONTEXT_ID,
-		       ctx->context_id);
-	BUILD_BUG_ON(ARRAY_SIZE(ctx->rx_hash_key) !=
-		     MC_CMD_RSS_CONTEXT_SET_KEY_IN_TOEPLITZ_KEY_LEN);
-	rc = efx_mcdi_rpc(efx, MC_CMD_RSS_CONTEXT_GET_KEY, inbuf, sizeof(inbuf),
-			  keybuf, sizeof(keybuf), &outlen);
-	if (rc != 0)
-		return rc;
-
-	if (WARN_ON(outlen != MC_CMD_RSS_CONTEXT_GET_KEY_OUT_LEN))
-		return -EIO;
-
-	for (i = 0; i < ARRAY_SIZE(ctx->rx_hash_key); ++i)
-		ctx->rx_hash_key[i] = MCDI_PTR(
-				keybuf, RSS_CONTEXT_GET_KEY_OUT_TOEPLITZ_KEY)[i];
-
-	return 0;
-}
-
-int efx_mcdi_rx_pull_rss_config(struct efx_nic *efx)
-{
-	int rc;
-
-	mutex_lock(&efx->rss_lock);
-	rc = efx_mcdi_rx_pull_rss_context_config(efx, &efx->rss_context);
-	mutex_unlock(&efx->rss_lock);
-	return rc;
-}
-
-void efx_mcdi_rx_restore_rss_contexts(struct efx_nic *efx)
-{
-	struct efx_ef10_nic_data *nic_data = efx->nic_data;
-	struct efx_rss_context *ctx;
-	int rc;
-
-	WARN_ON(!mutex_is_locked(&efx->rss_lock));
-
-	if (!nic_data->must_restore_rss_contexts)
-		return;
-
-	list_for_each_entry(ctx, &efx->rss_context.list, list) {
-		/* previous NIC RSS context is gone */
-		ctx->context_id = EFX_MCDI_RSS_CONTEXT_INVALID;
-		/* so try to allocate a new one */
-		rc = efx_mcdi_rx_push_rss_context_config(efx, ctx,
-							 ctx->rx_indir_table,
-							 ctx->rx_hash_key);
-		if (rc)
-			netif_warn(efx, probe, efx->net_dev,
-				   "failed to restore RSS context %u, rc=%d"
-				   "; RSS filters may fail to be applied\n",
-				   ctx->user_id, rc);
-	}
-	nic_data->must_restore_rss_contexts = false;
-}
-
-int efx_mcdi_pf_rx_push_rss_config(struct efx_nic *efx, bool user,
-				   const u32 *rx_indir_table,
-				   const u8 *key)
-{
-	int rc;
-
-	if (efx->rss_spread == 1)
-		return 0;
-
-	if (!key)
-		key = efx->rss_context.rx_hash_key;
-
-	rc = efx_mcdi_filter_rx_push_exclusive_rss_config(efx, rx_indir_table, key);
-
-	if (rc == -ENOBUFS && !user) {
-		unsigned context_size;
-		bool mismatch = false;
-		size_t i;
-
-		for (i = 0;
-		     i < ARRAY_SIZE(efx->rss_context.rx_indir_table) && !mismatch;
-		     i++)
-			mismatch = rx_indir_table[i] !=
-				ethtool_rxfh_indir_default(i, efx->rss_spread);
-
-		rc = efx_mcdi_filter_rx_push_shared_rss_config(efx,
-							       &context_size);
-		if (rc == 0) {
-			if (context_size != efx->rss_spread)
-				netif_warn(efx, probe, efx->net_dev,
-					   "Could not allocate an exclusive RSS"
-					   " context; allocated a shared one of"
-					   " different size."
-					   " Wanted %u, got %u.\n",
-					   efx->rss_spread, context_size);
-			else if (mismatch)
-				netif_warn(efx, probe, efx->net_dev,
-					   "Could not allocate an exclusive RSS"
-					   " context; allocated a shared one but"
-					   " could not apply custom"
-					   " indirection.\n");
-			else
-				netif_info(efx, probe, efx->net_dev,
-					   "Could not allocate an exclusive RSS"
-					   " context; allocated a shared one.\n");
-		}
-	}
-	return rc;
-}
-
-int efx_mcdi_vf_rx_push_rss_config(struct efx_nic *efx, bool user,
-				   const u32 *rx_indir_table
-				   __attribute__ ((unused)),
-				   const u8 *key
-				   __attribute__ ((unused)))
-{
-	if (user)
-		return -EOPNOTSUPP;
-	if (efx->rss_context.context_id != EFX_MCDI_RSS_CONTEXT_INVALID)
-		return 0;
-	return efx_mcdi_filter_rx_push_shared_rss_config(efx, NULL);
-}
-
 /* This creates an entry in the RX descriptor queue */
 static inline void
 efx_ef10_build_rx_desc(struct efx_rx_queue *rx_queue, unsigned int index)
@@ -3607,1588 +3140,61 @@ static void efx_ef10_prepare_flr(struct efx_nic *efx)
 	atomic_set(&efx->active_queues, 0);
 }
 
-/* Decide whether a filter should be exclusive or else should allow
- * delivery to additional recipients.  Currently we decide that
- * filters for specific local unicast MAC and IP addresses are
- * exclusive.
- */
-static bool efx_ef10_filter_is_exclusive(const struct efx_filter_spec *spec)
-{
-	if (spec->match_flags & EFX_FILTER_MATCH_LOC_MAC &&
-	    !is_multicast_ether_addr(spec->loc_mac))
-		return true;
-
-	if ((spec->match_flags &
-	     (EFX_FILTER_MATCH_ETHER_TYPE | EFX_FILTER_MATCH_LOC_HOST)) ==
-	    (EFX_FILTER_MATCH_ETHER_TYPE | EFX_FILTER_MATCH_LOC_HOST)) {
-		if (spec->ether_type == htons(ETH_P_IP) &&
-		    !ipv4_is_multicast(spec->loc_host[0]))
-			return true;
-		if (spec->ether_type == htons(ETH_P_IPV6) &&
-		    ((const u8 *)spec->loc_host)[0] != 0xff)
-			return true;
-	}
-
-	return false;
-}
-
-static struct efx_filter_spec *
-efx_ef10_filter_entry_spec(const struct efx_mcdi_filter_table *table,
-			   unsigned int filter_idx)
-{
-	return (struct efx_filter_spec *)(table->entry[filter_idx].spec &
-					  ~EFX_EF10_FILTER_FLAGS);
-}
-
-static unsigned int
-efx_ef10_filter_entry_flags(const struct efx_mcdi_filter_table *table,
-			   unsigned int filter_idx)
+static int efx_ef10_vport_set_mac_address(struct efx_nic *efx)
 {
-	return table->entry[filter_idx].spec & EFX_EF10_FILTER_FLAGS;
-}
+	struct efx_ef10_nic_data *nic_data = efx->nic_data;
+	u8 mac_old[ETH_ALEN];
+	int rc, rc2;
 
-static void
-efx_ef10_filter_set_entry(struct efx_mcdi_filter_table *table,
-			  unsigned int filter_idx,
-			  const struct efx_filter_spec *spec,
-			  unsigned int flags)
-{
-	table->entry[filter_idx].spec =	(unsigned long)spec | flags;
-}
+	/* Only reconfigure a PF-created vport */
+	if (is_zero_ether_addr(nic_data->vport_mac))
+		return 0;
 
-static void
-efx_ef10_filter_push_prep_set_match_fields(struct efx_nic *efx,
-					   const struct efx_filter_spec *spec,
-					   efx_dword_t *inbuf)
-{
-	enum efx_encap_type encap_type = efx_filter_get_encap_type(spec);
-	u32 match_fields = 0, uc_match, mc_match;
+	efx_device_detach_sync(efx);
+	efx_net_stop(efx->net_dev);
+	down_write(&efx->filter_sem);
+	efx_mcdi_filter_table_remove(efx);
+	up_write(&efx->filter_sem);
 
-	MCDI_SET_DWORD(inbuf, FILTER_OP_IN_OP,
-		       efx_ef10_filter_is_exclusive(spec) ?
-		       MC_CMD_FILTER_OP_IN_OP_INSERT :
-		       MC_CMD_FILTER_OP_IN_OP_SUBSCRIBE);
+	rc = efx_ef10_vadaptor_free(efx, nic_data->vport_id);
+	if (rc)
+		goto restore_filters;
 
-	/* Convert match flags and values.  Unlike almost
-	 * everything else in MCDI, these fields are in
-	 * network byte order.
-	 */
-#define COPY_VALUE(value, mcdi_field)					     \
-	do {							     \
-		match_fields |=					     \
-			1 << MC_CMD_FILTER_OP_IN_MATCH_ ##	     \
-			mcdi_field ## _LBN;			     \
-		BUILD_BUG_ON(					     \
-			MC_CMD_FILTER_OP_IN_ ## mcdi_field ## _LEN < \
-			sizeof(value));				     \
-		memcpy(MCDI_PTR(inbuf, FILTER_OP_IN_ ##	mcdi_field), \
-		       &value, sizeof(value));			     \
-	} while (0)
-#define COPY_FIELD(gen_flag, gen_field, mcdi_field)			     \
-	if (spec->match_flags & EFX_FILTER_MATCH_ ## gen_flag) {     \
-		COPY_VALUE(spec->gen_field, mcdi_field);	     \
-	}
-	/* Handle encap filters first.  They will always be mismatch
-	 * (unknown UC or MC) filters
-	 */
-	if (encap_type) {
-		/* ether_type and outer_ip_proto need to be variables
-		 * because COPY_VALUE wants to memcpy them
-		 */
-		__be16 ether_type =
-			htons(encap_type & EFX_ENCAP_FLAG_IPV6 ?
-			      ETH_P_IPV6 : ETH_P_IP);
-		u8 vni_type = MC_CMD_FILTER_OP_EXT_IN_VNI_TYPE_GENEVE;
-		u8 outer_ip_proto;
-
-		switch (encap_type & EFX_ENCAP_TYPES_MASK) {
-		case EFX_ENCAP_TYPE_VXLAN:
-			vni_type = MC_CMD_FILTER_OP_EXT_IN_VNI_TYPE_VXLAN;
-			/* fallthrough */
-		case EFX_ENCAP_TYPE_GENEVE:
-			COPY_VALUE(ether_type, ETHER_TYPE);
-			outer_ip_proto = IPPROTO_UDP;
-			COPY_VALUE(outer_ip_proto, IP_PROTO);
-			/* We always need to set the type field, even
-			 * though we're not matching on the TNI.
-			 */
-			MCDI_POPULATE_DWORD_1(inbuf,
-				FILTER_OP_EXT_IN_VNI_OR_VSID,
-				FILTER_OP_EXT_IN_VNI_TYPE,
-				vni_type);
-			break;
-		case EFX_ENCAP_TYPE_NVGRE:
-			COPY_VALUE(ether_type, ETHER_TYPE);
-			outer_ip_proto = IPPROTO_GRE;
-			COPY_VALUE(outer_ip_proto, IP_PROTO);
-			break;
-		default:
-			WARN_ON(1);
-		}
+	ether_addr_copy(mac_old, nic_data->vport_mac);
+	rc = efx_ef10_vport_del_mac(efx, nic_data->vport_id,
+				    nic_data->vport_mac);
+	if (rc)
+		goto restore_vadaptor;
 
-		uc_match = MC_CMD_FILTER_OP_EXT_IN_MATCH_IFRM_UNKNOWN_UCAST_DST_LBN;
-		mc_match = MC_CMD_FILTER_OP_EXT_IN_MATCH_IFRM_UNKNOWN_MCAST_DST_LBN;
+	rc = efx_ef10_vport_add_mac(efx, nic_data->vport_id,
+				    efx->net_dev->dev_addr);
+	if (!rc) {
+		ether_addr_copy(nic_data->vport_mac, efx->net_dev->dev_addr);
 	} else {
-		uc_match = MC_CMD_FILTER_OP_EXT_IN_MATCH_UNKNOWN_UCAST_DST_LBN;
-		mc_match = MC_CMD_FILTER_OP_EXT_IN_MATCH_UNKNOWN_MCAST_DST_LBN;
+		rc2 = efx_ef10_vport_add_mac(efx, nic_data->vport_id, mac_old);
+		if (rc2) {
+			/* Failed to add original MAC, so clear vport_mac */
+			eth_zero_addr(nic_data->vport_mac);
+			goto reset_nic;
+		}
 	}
 
-	if (spec->match_flags & EFX_FILTER_MATCH_LOC_MAC_IG)
-		match_fields |=
-			is_multicast_ether_addr(spec->loc_mac) ?
-			1 << mc_match :
-			1 << uc_match;
-	COPY_FIELD(REM_HOST, rem_host, SRC_IP);
-	COPY_FIELD(LOC_HOST, loc_host, DST_IP);
-	COPY_FIELD(REM_MAC, rem_mac, SRC_MAC);
-	COPY_FIELD(REM_PORT, rem_port, SRC_PORT);
-	COPY_FIELD(LOC_MAC, loc_mac, DST_MAC);
-	COPY_FIELD(LOC_PORT, loc_port, DST_PORT);
-	COPY_FIELD(ETHER_TYPE, ether_type, ETHER_TYPE);
-	COPY_FIELD(INNER_VID, inner_vid, INNER_VLAN);
-	COPY_FIELD(OUTER_VID, outer_vid, OUTER_VLAN);
-	COPY_FIELD(IP_PROTO, ip_proto, IP_PROTO);
-#undef COPY_FIELD
-#undef COPY_VALUE
-	MCDI_SET_DWORD(inbuf, FILTER_OP_IN_MATCH_FIELDS,
-		       match_fields);
-}
+restore_vadaptor:
+	rc2 = efx_ef10_vadaptor_alloc(efx, nic_data->vport_id);
+	if (rc2)
+		goto reset_nic;
+restore_filters:
+	down_write(&efx->filter_sem);
+	rc2 = efx_mcdi_filter_table_probe(efx);
+	up_write(&efx->filter_sem);
+	if (rc2)
+		goto reset_nic;
 
-static void efx_mcdi_filter_push_prep(struct efx_nic *efx,
-				      const struct efx_filter_spec *spec,
-				      efx_dword_t *inbuf, u64 handle,
-				      struct efx_rss_context *ctx,
-				      bool replacing)
-{
-	struct efx_ef10_nic_data *nic_data = efx->nic_data;
-	u32 flags = spec->flags;
+	rc2 = efx_net_open(efx->net_dev);
+	if (rc2)
+		goto reset_nic;
 
-	memset(inbuf, 0, MC_CMD_FILTER_OP_EXT_IN_LEN);
-
-	/* If RSS filter, caller better have given us an RSS context */
-	if (flags & EFX_FILTER_FLAG_RX_RSS) {
-		/* We don't have the ability to return an error, so we'll just
-		 * log a warning and disable RSS for the filter.
-		 */
-		if (WARN_ON_ONCE(!ctx))
-			flags &= ~EFX_FILTER_FLAG_RX_RSS;
-		else if (WARN_ON_ONCE(ctx->context_id == EFX_MCDI_RSS_CONTEXT_INVALID))
-			flags &= ~EFX_FILTER_FLAG_RX_RSS;
-	}
-
-	if (replacing) {
-		MCDI_SET_DWORD(inbuf, FILTER_OP_IN_OP,
-			       MC_CMD_FILTER_OP_IN_OP_REPLACE);
-		MCDI_SET_QWORD(inbuf, FILTER_OP_IN_HANDLE, handle);
-	} else {
-		efx_ef10_filter_push_prep_set_match_fields(efx, spec, inbuf);
-	}
-
-	MCDI_SET_DWORD(inbuf, FILTER_OP_IN_PORT_ID, nic_data->vport_id);
-	MCDI_SET_DWORD(inbuf, FILTER_OP_IN_RX_DEST,
-		       spec->dmaq_id == EFX_FILTER_RX_DMAQ_ID_DROP ?
-		       MC_CMD_FILTER_OP_IN_RX_DEST_DROP :
-		       MC_CMD_FILTER_OP_IN_RX_DEST_HOST);
-	MCDI_SET_DWORD(inbuf, FILTER_OP_IN_TX_DOMAIN, 0);
-	MCDI_SET_DWORD(inbuf, FILTER_OP_IN_TX_DEST,
-		       MC_CMD_FILTER_OP_IN_TX_DEST_DEFAULT);
-	MCDI_SET_DWORD(inbuf, FILTER_OP_IN_RX_QUEUE,
-		       spec->dmaq_id == EFX_FILTER_RX_DMAQ_ID_DROP ?
-		       0 : spec->dmaq_id);
-	MCDI_SET_DWORD(inbuf, FILTER_OP_IN_RX_MODE,
-		       (flags & EFX_FILTER_FLAG_RX_RSS) ?
-		       MC_CMD_FILTER_OP_IN_RX_MODE_RSS :
-		       MC_CMD_FILTER_OP_IN_RX_MODE_SIMPLE);
-	if (flags & EFX_FILTER_FLAG_RX_RSS)
-		MCDI_SET_DWORD(inbuf, FILTER_OP_IN_RX_CONTEXT, ctx->context_id);
-}
-
-static int efx_mcdi_filter_push(struct efx_nic *efx,
-				const struct efx_filter_spec *spec, u64 *handle,
-				struct efx_rss_context *ctx, bool replacing)
-{
-	MCDI_DECLARE_BUF(inbuf, MC_CMD_FILTER_OP_EXT_IN_LEN);
-	MCDI_DECLARE_BUF(outbuf, MC_CMD_FILTER_OP_EXT_OUT_LEN);
-	size_t outlen;
-	int rc;
-
-	efx_mcdi_filter_push_prep(efx, spec, inbuf, *handle, ctx, replacing);
-	rc = efx_mcdi_rpc_quiet(efx, MC_CMD_FILTER_OP, inbuf, sizeof(inbuf),
-				outbuf, sizeof(outbuf), &outlen);
-	if (rc && spec->priority != EFX_FILTER_PRI_HINT)
-		efx_mcdi_display_error(efx, MC_CMD_FILTER_OP, sizeof(inbuf),
-				       outbuf, outlen, rc);
-	if (rc == 0)
-		*handle = MCDI_QWORD(outbuf, FILTER_OP_OUT_HANDLE);
-	if (rc == -ENOSPC)
-		rc = -EBUSY; /* to match efx_farch_filter_insert() */
-	return rc;
-}
-
-static u32 efx_mcdi_filter_mcdi_flags_from_spec(const struct efx_filter_spec *spec)
-{
-	enum efx_encap_type encap_type = efx_filter_get_encap_type(spec);
-	unsigned int match_flags = spec->match_flags;
-	unsigned int uc_match, mc_match;
-	u32 mcdi_flags = 0;
-
-#define MAP_FILTER_TO_MCDI_FLAG(gen_flag, mcdi_field, encap) {		\
-		unsigned int  old_match_flags = match_flags;		\
-		match_flags &= ~EFX_FILTER_MATCH_ ## gen_flag;		\
-		if (match_flags != old_match_flags)			\
-			mcdi_flags |=					\
-				(1 << ((encap) ?			\
-				       MC_CMD_FILTER_OP_EXT_IN_MATCH_IFRM_ ## \
-				       mcdi_field ## _LBN :		\
-				       MC_CMD_FILTER_OP_EXT_IN_MATCH_ ##\
-				       mcdi_field ## _LBN));		\
-	}
-	/* inner or outer based on encap type */
-	MAP_FILTER_TO_MCDI_FLAG(REM_HOST, SRC_IP, encap_type);
-	MAP_FILTER_TO_MCDI_FLAG(LOC_HOST, DST_IP, encap_type);
-	MAP_FILTER_TO_MCDI_FLAG(REM_MAC, SRC_MAC, encap_type);
-	MAP_FILTER_TO_MCDI_FLAG(REM_PORT, SRC_PORT, encap_type);
-	MAP_FILTER_TO_MCDI_FLAG(LOC_MAC, DST_MAC, encap_type);
-	MAP_FILTER_TO_MCDI_FLAG(LOC_PORT, DST_PORT, encap_type);
-	MAP_FILTER_TO_MCDI_FLAG(ETHER_TYPE, ETHER_TYPE, encap_type);
-	MAP_FILTER_TO_MCDI_FLAG(IP_PROTO, IP_PROTO, encap_type);
-	/* always outer */
-	MAP_FILTER_TO_MCDI_FLAG(INNER_VID, INNER_VLAN, false);
-	MAP_FILTER_TO_MCDI_FLAG(OUTER_VID, OUTER_VLAN, false);
-#undef MAP_FILTER_TO_MCDI_FLAG
-
-	/* special handling for encap type, and mismatch */
-	if (encap_type) {
-		match_flags &= ~EFX_FILTER_MATCH_ENCAP_TYPE;
-		mcdi_flags |=
-			(1 << MC_CMD_FILTER_OP_EXT_IN_MATCH_ETHER_TYPE_LBN);
-		mcdi_flags |= (1 << MC_CMD_FILTER_OP_EXT_IN_MATCH_IP_PROTO_LBN);
-
-		uc_match = MC_CMD_FILTER_OP_EXT_IN_MATCH_IFRM_UNKNOWN_UCAST_DST_LBN;
-		mc_match = MC_CMD_FILTER_OP_EXT_IN_MATCH_IFRM_UNKNOWN_MCAST_DST_LBN;
-	} else {
-		uc_match = MC_CMD_FILTER_OP_EXT_IN_MATCH_UNKNOWN_UCAST_DST_LBN;
-		mc_match = MC_CMD_FILTER_OP_EXT_IN_MATCH_UNKNOWN_MCAST_DST_LBN;
-	}
-
-	if (match_flags & EFX_FILTER_MATCH_LOC_MAC_IG) {
-		match_flags &= ~EFX_FILTER_MATCH_LOC_MAC_IG;
-		mcdi_flags |=
-			is_multicast_ether_addr(spec->loc_mac) ?
-			1 << mc_match :
-			1 << uc_match;
-	}
-
-	/* Did we map them all? */
-	WARN_ON_ONCE(match_flags);
-
-	return mcdi_flags;
-}
-
-static int efx_mcdi_filter_pri(struct efx_mcdi_filter_table *table,
-			       const struct efx_filter_spec *spec)
-{
-	u32 mcdi_flags = efx_mcdi_filter_mcdi_flags_from_spec(spec);
-	unsigned int match_pri;
-
-	for (match_pri = 0;
-	     match_pri < table->rx_match_count;
-	     match_pri++)
-		if (table->rx_match_mcdi_flags[match_pri] == mcdi_flags)
-			return match_pri;
-
-	return -EPROTONOSUPPORT;
-}
-
-static s32 efx_mcdi_filter_insert_locked(struct efx_nic *efx,
-					 struct efx_filter_spec *spec,
-					 bool replace_equal)
-{
-	DECLARE_BITMAP(mc_rem_map, EFX_EF10_FILTER_SEARCH_LIMIT);
-	struct efx_ef10_nic_data *nic_data = efx->nic_data;
-	struct efx_mcdi_filter_table *table;
-	struct efx_filter_spec *saved_spec;
-	struct efx_rss_context *ctx = NULL;
-	unsigned int match_pri, hash;
-	unsigned int priv_flags;
-	bool rss_locked = false;
-	bool replacing = false;
-	unsigned int depth, i;
-	int ins_index = -1;
-	DEFINE_WAIT(wait);
-	bool is_mc_recip;
-	s32 rc;
-
-	WARN_ON(!rwsem_is_locked(&efx->filter_sem));
-	table = efx->filter_state;
-	down_write(&table->lock);
-
-	/* For now, only support RX filters */
-	if ((spec->flags & (EFX_FILTER_FLAG_RX | EFX_FILTER_FLAG_TX)) !=
-	    EFX_FILTER_FLAG_RX) {
-		rc = -EINVAL;
-		goto out_unlock;
-	}
-
-	rc = efx_mcdi_filter_pri(table, spec);
-	if (rc < 0)
-		goto out_unlock;
-	match_pri = rc;
-
-	hash = efx_filter_spec_hash(spec);
-	is_mc_recip = efx_filter_is_mc_recipient(spec);
-	if (is_mc_recip)
-		bitmap_zero(mc_rem_map, EFX_EF10_FILTER_SEARCH_LIMIT);
-
-	if (spec->flags & EFX_FILTER_FLAG_RX_RSS) {
-		mutex_lock(&efx->rss_lock);
-		rss_locked = true;
-		if (spec->rss_context)
-			ctx = efx_find_rss_context_entry(efx, spec->rss_context);
-		else
-			ctx = &efx->rss_context;
-		if (!ctx) {
-			rc = -ENOENT;
-			goto out_unlock;
-		}
-		if (ctx->context_id == EFX_MCDI_RSS_CONTEXT_INVALID) {
-			rc = -EOPNOTSUPP;
-			goto out_unlock;
-		}
-	}
-
-	/* Find any existing filters with the same match tuple or
-	 * else a free slot to insert at.
-	 */
-	for (depth = 1; depth < EFX_EF10_FILTER_SEARCH_LIMIT; depth++) {
-		i = (hash + depth) & (EFX_MCDI_FILTER_TBL_ROWS - 1);
-		saved_spec = efx_ef10_filter_entry_spec(table, i);
-
-		if (!saved_spec) {
-			if (ins_index < 0)
-				ins_index = i;
-		} else if (efx_filter_spec_equal(spec, saved_spec)) {
-			if (spec->priority < saved_spec->priority &&
-			    spec->priority != EFX_FILTER_PRI_AUTO) {
-				rc = -EPERM;
-				goto out_unlock;
-			}
-			if (!is_mc_recip) {
-				/* This is the only one */
-				if (spec->priority ==
-				    saved_spec->priority &&
-				    !replace_equal) {
-					rc = -EEXIST;
-					goto out_unlock;
-				}
-				ins_index = i;
-				break;
-			} else if (spec->priority >
-				   saved_spec->priority ||
-				   (spec->priority ==
-				    saved_spec->priority &&
-				    replace_equal)) {
-				if (ins_index < 0)
-					ins_index = i;
-				else
-					__set_bit(depth, mc_rem_map);
-			}
-		}
-	}
-
-	/* Once we reach the maximum search depth, use the first suitable
-	 * slot, or return -EBUSY if there was none
-	 */
-	if (ins_index < 0) {
-		rc = -EBUSY;
-		goto out_unlock;
-	}
-
-	/* Create a software table entry if necessary. */
-	saved_spec = efx_ef10_filter_entry_spec(table, ins_index);
-	if (saved_spec) {
-		if (spec->priority == EFX_FILTER_PRI_AUTO &&
-		    saved_spec->priority >= EFX_FILTER_PRI_AUTO) {
-			/* Just make sure it won't be removed */
-			if (saved_spec->priority > EFX_FILTER_PRI_AUTO)
-				saved_spec->flags |= EFX_FILTER_FLAG_RX_OVER_AUTO;
-			table->entry[ins_index].spec &=
-				~EFX_EF10_FILTER_FLAG_AUTO_OLD;
-			rc = ins_index;
-			goto out_unlock;
-		}
-		replacing = true;
-		priv_flags = efx_ef10_filter_entry_flags(table, ins_index);
-	} else {
-		saved_spec = kmalloc(sizeof(*spec), GFP_ATOMIC);
-		if (!saved_spec) {
-			rc = -ENOMEM;
-			goto out_unlock;
-		}
-		*saved_spec = *spec;
-		priv_flags = 0;
-	}
-	efx_ef10_filter_set_entry(table, ins_index, saved_spec, priv_flags);
-
-	/* Actually insert the filter on the HW */
-	rc = efx_mcdi_filter_push(efx, spec, &table->entry[ins_index].handle,
-				  ctx, replacing);
-
-	if (rc == -EINVAL && nic_data->must_realloc_vis)
-		/* The MC rebooted under us, causing it to reject our filter
-		 * insertion as pointing to an invalid VI (spec->dmaq_id).
-		 */
-		rc = -EAGAIN;
-
-	/* Finalise the software table entry */
-	if (rc == 0) {
-		if (replacing) {
-			/* Update the fields that may differ */
-			if (saved_spec->priority == EFX_FILTER_PRI_AUTO)
-				saved_spec->flags |=
-					EFX_FILTER_FLAG_RX_OVER_AUTO;
-			saved_spec->priority = spec->priority;
-			saved_spec->flags &= EFX_FILTER_FLAG_RX_OVER_AUTO;
-			saved_spec->flags |= spec->flags;
-			saved_spec->rss_context = spec->rss_context;
-			saved_spec->dmaq_id = spec->dmaq_id;
-		}
-	} else if (!replacing) {
-		kfree(saved_spec);
-		saved_spec = NULL;
-	} else {
-		/* We failed to replace, so the old filter is still present.
-		 * Roll back the software table to reflect this.  In fact the
-		 * efx_ef10_filter_set_entry() call below will do the right
-		 * thing, so nothing extra is needed here.
-		 */
-	}
-	efx_ef10_filter_set_entry(table, ins_index, saved_spec, priv_flags);
-
-	/* Remove and finalise entries for lower-priority multicast
-	 * recipients
-	 */
-	if (is_mc_recip) {
-		MCDI_DECLARE_BUF(inbuf, MC_CMD_FILTER_OP_EXT_IN_LEN);
-		unsigned int depth, i;
-
-		memset(inbuf, 0, sizeof(inbuf));
-
-		for (depth = 0; depth < EFX_EF10_FILTER_SEARCH_LIMIT; depth++) {
-			if (!test_bit(depth, mc_rem_map))
-				continue;
-
-			i = (hash + depth) & (EFX_MCDI_FILTER_TBL_ROWS - 1);
-			saved_spec = efx_ef10_filter_entry_spec(table, i);
-			priv_flags = efx_ef10_filter_entry_flags(table, i);
-
-			if (rc == 0) {
-				MCDI_SET_DWORD(inbuf, FILTER_OP_IN_OP,
-					       MC_CMD_FILTER_OP_IN_OP_UNSUBSCRIBE);
-				MCDI_SET_QWORD(inbuf, FILTER_OP_IN_HANDLE,
-					       table->entry[i].handle);
-				rc = efx_mcdi_rpc(efx, MC_CMD_FILTER_OP,
-						  inbuf, sizeof(inbuf),
-						  NULL, 0, NULL);
-			}
-
-			if (rc == 0) {
-				kfree(saved_spec);
-				saved_spec = NULL;
-				priv_flags = 0;
-			}
-			efx_ef10_filter_set_entry(table, i, saved_spec,
-						  priv_flags);
-		}
-	}
-
-	/* If successful, return the inserted filter ID */
-	if (rc == 0)
-		rc = efx_ef10_make_filter_id(match_pri, ins_index);
-
-out_unlock:
-	if (rss_locked)
-		mutex_unlock(&efx->rss_lock);
-	up_write(&table->lock);
-	return rc;
-}
-
-s32 efx_mcdi_filter_insert(struct efx_nic *efx,
-			   struct efx_filter_spec *spec,
-			   bool replace_equal)
-{
-	s32 ret;
-
-	down_read(&efx->filter_sem);
-	ret = efx_mcdi_filter_insert_locked(efx, spec, replace_equal);
-	up_read(&efx->filter_sem);
-
-	return ret;
-}
-
-/* Remove a filter.
- * If !by_index, remove by ID
- * If by_index, remove by index
- * Filter ID may come from userland and must be range-checked.
- * Caller must hold efx->filter_sem for read, and efx->filter_state->lock
- * for write.
- */
-static int efx_mcdi_filter_remove_internal(struct efx_nic *efx,
-					   unsigned int priority_mask,
-					   u32 filter_id, bool by_index)
-{
-	unsigned int filter_idx = efx_ef10_filter_get_unsafe_id(filter_id);
-	struct efx_mcdi_filter_table *table = efx->filter_state;
-	MCDI_DECLARE_BUF(inbuf,
-			 MC_CMD_FILTER_OP_IN_HANDLE_OFST +
-			 MC_CMD_FILTER_OP_IN_HANDLE_LEN);
-	struct efx_filter_spec *spec;
-	DEFINE_WAIT(wait);
-	int rc;
-
-	spec = efx_ef10_filter_entry_spec(table, filter_idx);
-	if (!spec ||
-	    (!by_index &&
-	     efx_mcdi_filter_pri(table, spec) !=
-	     efx_ef10_filter_get_unsafe_pri(filter_id)))
-		return -ENOENT;
-
-	if (spec->flags & EFX_FILTER_FLAG_RX_OVER_AUTO &&
-	    priority_mask == (1U << EFX_FILTER_PRI_AUTO)) {
-		/* Just remove flags */
-		spec->flags &= ~EFX_FILTER_FLAG_RX_OVER_AUTO;
-		table->entry[filter_idx].spec &= ~EFX_EF10_FILTER_FLAG_AUTO_OLD;
-		return 0;
-	}
-
-	if (!(priority_mask & (1U << spec->priority)))
-		return -ENOENT;
-
-	if (spec->flags & EFX_FILTER_FLAG_RX_OVER_AUTO) {
-		/* Reset to an automatic filter */
-
-		struct efx_filter_spec new_spec = *spec;
-
-		new_spec.priority = EFX_FILTER_PRI_AUTO;
-		new_spec.flags = (EFX_FILTER_FLAG_RX |
-				  (efx_rss_active(&efx->rss_context) ?
-				   EFX_FILTER_FLAG_RX_RSS : 0));
-		new_spec.dmaq_id = 0;
-		new_spec.rss_context = 0;
-		rc = efx_mcdi_filter_push(efx, &new_spec,
-					  &table->entry[filter_idx].handle,
-					  &efx->rss_context,
-					  true);
-
-		if (rc == 0)
-			*spec = new_spec;
-	} else {
-		/* Really remove the filter */
-
-		MCDI_SET_DWORD(inbuf, FILTER_OP_IN_OP,
-			       efx_ef10_filter_is_exclusive(spec) ?
-			       MC_CMD_FILTER_OP_IN_OP_REMOVE :
-			       MC_CMD_FILTER_OP_IN_OP_UNSUBSCRIBE);
-		MCDI_SET_QWORD(inbuf, FILTER_OP_IN_HANDLE,
-			       table->entry[filter_idx].handle);
-		rc = efx_mcdi_rpc_quiet(efx, MC_CMD_FILTER_OP,
-					inbuf, sizeof(inbuf), NULL, 0, NULL);
-
-		if ((rc == 0) || (rc == -ENOENT)) {
-			/* Filter removed OK or didn't actually exist */
-			kfree(spec);
-			efx_ef10_filter_set_entry(table, filter_idx, NULL, 0);
-		} else {
-			efx_mcdi_display_error(efx, MC_CMD_FILTER_OP,
-					       MC_CMD_FILTER_OP_EXT_IN_LEN,
-					       NULL, 0, rc);
-		}
-	}
-
-	return rc;
-}
-
-int efx_mcdi_filter_remove_safe(struct efx_nic *efx,
-				enum efx_filter_priority priority,
-				u32 filter_id)
-{
-	struct efx_mcdi_filter_table *table;
-	int rc;
-
-	down_read(&efx->filter_sem);
-	table = efx->filter_state;
-	down_write(&table->lock);
-	rc = efx_mcdi_filter_remove_internal(efx, 1U << priority, filter_id,
-					     false);
-	up_write(&table->lock);
-	up_read(&efx->filter_sem);
-	return rc;
-}
-
-/* Caller must hold efx->filter_sem for read */
-static void efx_mcdi_filter_remove_unsafe(struct efx_nic *efx,
-					  enum efx_filter_priority priority,
-					  u32 filter_id)
-{
-	struct efx_mcdi_filter_table *table = efx->filter_state;
-
-	if (filter_id == EFX_EF10_FILTER_ID_INVALID)
-		return;
-
-	down_write(&table->lock);
-	efx_mcdi_filter_remove_internal(efx, 1U << priority, filter_id,
-					true);
-	up_write(&table->lock);
-}
-
-int efx_mcdi_filter_get_safe(struct efx_nic *efx,
-			     enum efx_filter_priority priority,
-			     u32 filter_id, struct efx_filter_spec *spec)
-{
-	unsigned int filter_idx = efx_ef10_filter_get_unsafe_id(filter_id);
-	const struct efx_filter_spec *saved_spec;
-	struct efx_mcdi_filter_table *table;
-	int rc;
-
-	down_read(&efx->filter_sem);
-	table = efx->filter_state;
-	down_read(&table->lock);
-	saved_spec = efx_ef10_filter_entry_spec(table, filter_idx);
-	if (saved_spec && saved_spec->priority == priority &&
-	    efx_mcdi_filter_pri(table, saved_spec) ==
-	    efx_ef10_filter_get_unsafe_pri(filter_id)) {
-		*spec = *saved_spec;
-		rc = 0;
-	} else {
-		rc = -ENOENT;
-	}
-	up_read(&table->lock);
-	up_read(&efx->filter_sem);
-	return rc;
-}
-
-int efx_mcdi_filter_clear_rx(struct efx_nic *efx,
-			     enum efx_filter_priority priority)
-{
-	struct efx_mcdi_filter_table *table;
-	unsigned int priority_mask;
-	unsigned int i;
-	int rc;
-
-	priority_mask = (((1U << (priority + 1)) - 1) &
-			 ~(1U << EFX_FILTER_PRI_AUTO));
-
-	down_read(&efx->filter_sem);
-	table = efx->filter_state;
-	down_write(&table->lock);
-	for (i = 0; i < EFX_MCDI_FILTER_TBL_ROWS; i++) {
-		rc = efx_mcdi_filter_remove_internal(efx, priority_mask,
-						     i, true);
-		if (rc && rc != -ENOENT)
-			break;
-		rc = 0;
-	}
-
-	up_write(&table->lock);
-	up_read(&efx->filter_sem);
-	return rc;
-}
-
-u32 efx_mcdi_filter_count_rx_used(struct efx_nic *efx,
-				  enum efx_filter_priority priority)
-{
-	struct efx_mcdi_filter_table *table;
-	unsigned int filter_idx;
-	s32 count = 0;
-
-	down_read(&efx->filter_sem);
-	table = efx->filter_state;
-	down_read(&table->lock);
-	for (filter_idx = 0; filter_idx < EFX_MCDI_FILTER_TBL_ROWS; filter_idx++) {
-		if (table->entry[filter_idx].spec &&
-		    efx_ef10_filter_entry_spec(table, filter_idx)->priority ==
-		    priority)
-			++count;
-	}
-	up_read(&table->lock);
-	up_read(&efx->filter_sem);
-	return count;
-}
-
-u32 efx_mcdi_filter_get_rx_id_limit(struct efx_nic *efx)
-{
-	struct efx_mcdi_filter_table *table = efx->filter_state;
-
-	return table->rx_match_count * EFX_MCDI_FILTER_TBL_ROWS * 2;
-}
-
-s32 efx_mcdi_filter_get_rx_ids(struct efx_nic *efx,
-			       enum efx_filter_priority priority,
-			       u32 *buf, u32 size)
-{
-	struct efx_mcdi_filter_table *table;
-	struct efx_filter_spec *spec;
-	unsigned int filter_idx;
-	s32 count = 0;
-
-	down_read(&efx->filter_sem);
-	table = efx->filter_state;
-	down_read(&table->lock);
-
-	for (filter_idx = 0; filter_idx < EFX_MCDI_FILTER_TBL_ROWS; filter_idx++) {
-		spec = efx_ef10_filter_entry_spec(table, filter_idx);
-		if (spec && spec->priority == priority) {
-			if (count == size) {
-				count = -EMSGSIZE;
-				break;
-			}
-			buf[count++] =
-				efx_ef10_make_filter_id(
-					efx_mcdi_filter_pri(table, spec),
-					filter_idx);
-		}
-	}
-	up_read(&table->lock);
-	up_read(&efx->filter_sem);
-	return count;
-}
-
-#ifdef CONFIG_RFS_ACCEL
-
-bool efx_mcdi_filter_rfs_expire_one(struct efx_nic *efx, u32 flow_id,
-				    unsigned int filter_idx)
-{
-	struct efx_filter_spec *spec, saved_spec;
-	struct efx_mcdi_filter_table *table;
-	struct efx_arfs_rule *rule = NULL;
-	bool ret = true, force = false;
-	u16 arfs_id;
-
-	down_read(&efx->filter_sem);
-	table = efx->filter_state;
-	down_write(&table->lock);
-	spec = efx_ef10_filter_entry_spec(table, filter_idx);
-
-	if (!spec || spec->priority != EFX_FILTER_PRI_HINT)
-		goto out_unlock;
-
-	spin_lock_bh(&efx->rps_hash_lock);
-	if (!efx->rps_hash_table) {
-		/* In the absence of the table, we always return 0 to ARFS. */
-		arfs_id = 0;
-	} else {
-		rule = efx_rps_hash_find(efx, spec);
-		if (!rule)
-			/* ARFS table doesn't know of this filter, so remove it */
-			goto expire;
-		arfs_id = rule->arfs_id;
-		ret = efx_rps_check_rule(rule, filter_idx, &force);
-		if (force)
-			goto expire;
-		if (!ret) {
-			spin_unlock_bh(&efx->rps_hash_lock);
-			goto out_unlock;
-		}
-	}
-	if (!rps_may_expire_flow(efx->net_dev, spec->dmaq_id, flow_id, arfs_id))
-		ret = false;
-	else if (rule)
-		rule->filter_id = EFX_ARFS_FILTER_ID_REMOVING;
-expire:
-	saved_spec = *spec; /* remove operation will kfree spec */
-	spin_unlock_bh(&efx->rps_hash_lock);
-	/* At this point (since we dropped the lock), another thread might queue
-	 * up a fresh insertion request (but the actual insertion will be held
-	 * up by our possession of the filter table lock).  In that case, it
-	 * will set rule->filter_id to EFX_ARFS_FILTER_ID_PENDING, meaning that
-	 * the rule is not removed by efx_rps_hash_del() below.
-	 */
-	if (ret)
-		ret = efx_mcdi_filter_remove_internal(efx, 1U << spec->priority,
-						      filter_idx, true) == 0;
-	/* While we can't safely dereference rule (we dropped the lock), we can
-	 * still test it for NULL.
-	 */
-	if (ret && rule) {
-		/* Expiring, so remove entry from ARFS table */
-		spin_lock_bh(&efx->rps_hash_lock);
-		efx_rps_hash_del(efx, &saved_spec);
-		spin_unlock_bh(&efx->rps_hash_lock);
-	}
-out_unlock:
-	up_write(&table->lock);
-	up_read(&efx->filter_sem);
-	return ret;
-}
-
-#endif /* CONFIG_RFS_ACCEL */
-
-static int efx_mcdi_filter_match_flags_from_mcdi(bool encap, u32 mcdi_flags)
-{
-	int match_flags = 0;
-
-#define MAP_FLAG(gen_flag, mcdi_field) do {				\
-		u32 old_mcdi_flags = mcdi_flags;			\
-		mcdi_flags &= ~(1 << MC_CMD_FILTER_OP_EXT_IN_MATCH_ ##	\
-				     mcdi_field ## _LBN);		\
-		if (mcdi_flags != old_mcdi_flags)			\
-			match_flags |= EFX_FILTER_MATCH_ ## gen_flag;	\
-	} while (0)
-
-	if (encap) {
-		/* encap filters must specify encap type */
-		match_flags |= EFX_FILTER_MATCH_ENCAP_TYPE;
-		/* and imply ethertype and ip proto */
-		mcdi_flags &=
-			~(1 << MC_CMD_FILTER_OP_EXT_IN_MATCH_IP_PROTO_LBN);
-		mcdi_flags &=
-			~(1 << MC_CMD_FILTER_OP_EXT_IN_MATCH_ETHER_TYPE_LBN);
-		/* VLAN tags refer to the outer packet */
-		MAP_FLAG(INNER_VID, INNER_VLAN);
-		MAP_FLAG(OUTER_VID, OUTER_VLAN);
-		/* everything else refers to the inner packet */
-		MAP_FLAG(LOC_MAC_IG, IFRM_UNKNOWN_UCAST_DST);
-		MAP_FLAG(LOC_MAC_IG, IFRM_UNKNOWN_MCAST_DST);
-		MAP_FLAG(REM_HOST, IFRM_SRC_IP);
-		MAP_FLAG(LOC_HOST, IFRM_DST_IP);
-		MAP_FLAG(REM_MAC, IFRM_SRC_MAC);
-		MAP_FLAG(REM_PORT, IFRM_SRC_PORT);
-		MAP_FLAG(LOC_MAC, IFRM_DST_MAC);
-		MAP_FLAG(LOC_PORT, IFRM_DST_PORT);
-		MAP_FLAG(ETHER_TYPE, IFRM_ETHER_TYPE);
-		MAP_FLAG(IP_PROTO, IFRM_IP_PROTO);
-	} else {
-		MAP_FLAG(LOC_MAC_IG, UNKNOWN_UCAST_DST);
-		MAP_FLAG(LOC_MAC_IG, UNKNOWN_MCAST_DST);
-		MAP_FLAG(REM_HOST, SRC_IP);
-		MAP_FLAG(LOC_HOST, DST_IP);
-		MAP_FLAG(REM_MAC, SRC_MAC);
-		MAP_FLAG(REM_PORT, SRC_PORT);
-		MAP_FLAG(LOC_MAC, DST_MAC);
-		MAP_FLAG(LOC_PORT, DST_PORT);
-		MAP_FLAG(ETHER_TYPE, ETHER_TYPE);
-		MAP_FLAG(INNER_VID, INNER_VLAN);
-		MAP_FLAG(OUTER_VID, OUTER_VLAN);
-		MAP_FLAG(IP_PROTO, IP_PROTO);
-	}
-#undef MAP_FLAG
-
-	/* Did we map them all? */
-	if (mcdi_flags)
-		return -EINVAL;
-
-	return match_flags;
-}
-
-void efx_mcdi_filter_cleanup_vlans(struct efx_nic *efx)
-{
-	struct efx_mcdi_filter_table *table = efx->filter_state;
-	struct efx_mcdi_filter_vlan *vlan, *next_vlan;
-
-	/* See comment in efx_mcdi_filter_table_remove() */
-	if (!efx_rwsem_assert_write_locked(&efx->filter_sem))
-		return;
-
-	if (!table)
-		return;
-
-	list_for_each_entry_safe(vlan, next_vlan, &table->vlan_list, list)
-		efx_mcdi_filter_del_vlan_internal(efx, vlan);
-}
-
-bool efx_mcdi_filter_match_supported(struct efx_mcdi_filter_table *table,
-				     bool encap,
-				     enum efx_filter_match_flags match_flags)
-{
-	unsigned int match_pri;
-	int mf;
-
-	for (match_pri = 0;
-	     match_pri < table->rx_match_count;
-	     match_pri++) {
-		mf = efx_mcdi_filter_match_flags_from_mcdi(encap,
-				table->rx_match_mcdi_flags[match_pri]);
-		if (mf == match_flags)
-			return true;
-	}
-
-	return false;
-}
-
-static int
-efx_mcdi_filter_table_probe_matches(struct efx_nic *efx,
-				    struct efx_mcdi_filter_table *table,
-				    bool encap)
-{
-	MCDI_DECLARE_BUF(inbuf, MC_CMD_GET_PARSER_DISP_INFO_IN_LEN);
-	MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_PARSER_DISP_INFO_OUT_LENMAX);
-	unsigned int pd_match_pri, pd_match_count;
-	size_t outlen;
-	int rc;
-
-	/* Find out which RX filter types are supported, and their priorities */
-	MCDI_SET_DWORD(inbuf, GET_PARSER_DISP_INFO_IN_OP,
-		       encap ?
-		       MC_CMD_GET_PARSER_DISP_INFO_IN_OP_GET_SUPPORTED_ENCAP_RX_MATCHES :
-		       MC_CMD_GET_PARSER_DISP_INFO_IN_OP_GET_SUPPORTED_RX_MATCHES);
-	rc = efx_mcdi_rpc(efx, MC_CMD_GET_PARSER_DISP_INFO,
-			  inbuf, sizeof(inbuf), outbuf, sizeof(outbuf),
-			  &outlen);
-	if (rc)
-		return rc;
-
-	pd_match_count = MCDI_VAR_ARRAY_LEN(
-		outlen, GET_PARSER_DISP_INFO_OUT_SUPPORTED_MATCHES);
-
-	for (pd_match_pri = 0; pd_match_pri < pd_match_count; pd_match_pri++) {
-		u32 mcdi_flags =
-			MCDI_ARRAY_DWORD(
-				outbuf,
-				GET_PARSER_DISP_INFO_OUT_SUPPORTED_MATCHES,
-				pd_match_pri);
-		rc = efx_mcdi_filter_match_flags_from_mcdi(encap, mcdi_flags);
-		if (rc < 0) {
-			netif_dbg(efx, probe, efx->net_dev,
-				  "%s: fw flags %#x pri %u not supported in driver\n",
-				  __func__, mcdi_flags, pd_match_pri);
-		} else {
-			netif_dbg(efx, probe, efx->net_dev,
-				  "%s: fw flags %#x pri %u supported as driver flags %#x pri %u\n",
-				  __func__, mcdi_flags, pd_match_pri,
-				  rc, table->rx_match_count);
-			table->rx_match_mcdi_flags[table->rx_match_count] = mcdi_flags;
-			table->rx_match_count++;
-		}
-	}
-
-	return 0;
-}
-
-int efx_mcdi_filter_table_probe(struct efx_nic *efx)
-{
-	struct efx_ef10_nic_data *nic_data = efx->nic_data;
-	struct net_device *net_dev = efx->net_dev;
-	struct efx_mcdi_filter_table *table;
-	struct efx_ef10_vlan *vlan;
-	int rc;
-
-	if (!efx_rwsem_assert_write_locked(&efx->filter_sem))
-		return -EINVAL;
-
-	if (efx->filter_state) /* already probed */
-		return 0;
-
-	table = kzalloc(sizeof(*table), GFP_KERNEL);
-	if (!table)
-		return -ENOMEM;
-
-	table->rx_match_count = 0;
-	rc = efx_mcdi_filter_table_probe_matches(efx, table, false);
-	if (rc)
-		goto fail;
-	if (nic_data->datapath_caps &
-		   (1 << MC_CMD_GET_CAPABILITIES_OUT_VXLAN_NVGRE_LBN))
-		rc = efx_mcdi_filter_table_probe_matches(efx, table, true);
-	if (rc)
-		goto fail;
-	if ((efx_supported_features(efx) & NETIF_F_HW_VLAN_CTAG_FILTER) &&
-	    !(efx_mcdi_filter_match_supported(table, false,
-		(EFX_FILTER_MATCH_OUTER_VID | EFX_FILTER_MATCH_LOC_MAC)) &&
-	      efx_mcdi_filter_match_supported(table, false,
-		(EFX_FILTER_MATCH_OUTER_VID | EFX_FILTER_MATCH_LOC_MAC_IG)))) {
-		netif_info(efx, probe, net_dev,
-			   "VLAN filters are not supported in this firmware variant\n");
-		net_dev->features &= ~NETIF_F_HW_VLAN_CTAG_FILTER;
-		efx->fixed_features &= ~NETIF_F_HW_VLAN_CTAG_FILTER;
-		net_dev->hw_features &= ~NETIF_F_HW_VLAN_CTAG_FILTER;
-	}
-
-	table->entry = vzalloc(array_size(EFX_MCDI_FILTER_TBL_ROWS,
-					  sizeof(*table->entry)));
-	if (!table->entry) {
-		rc = -ENOMEM;
-		goto fail;
-	}
-
-	table->mc_promisc_last = false;
-	table->vlan_filter =
-		!!(efx->net_dev->features & NETIF_F_HW_VLAN_CTAG_FILTER);
-	INIT_LIST_HEAD(&table->vlan_list);
-	init_rwsem(&table->lock);
-
-	efx->filter_state = table;
-
-	list_for_each_entry(vlan, &nic_data->vlan_list, list) {
-		rc = efx_mcdi_filter_add_vlan(efx, vlan->vid);
-		if (rc)
-			goto fail_add_vlan;
-	}
-
-	return 0;
-
-fail_add_vlan:
-	efx_mcdi_filter_cleanup_vlans(efx);
-	efx->filter_state = NULL;
-fail:
-	kfree(table);
-	return rc;
-}
-
-/* Caller must hold efx->filter_sem for read if race against
- * efx_mcdi_filter_table_remove() is possible
- */
-void efx_mcdi_filter_table_restore(struct efx_nic *efx)
-{
-	struct efx_mcdi_filter_table *table = efx->filter_state;
-	struct efx_ef10_nic_data *nic_data = efx->nic_data;
-	unsigned int invalid_filters = 0, failed = 0;
-	struct efx_mcdi_filter_vlan *vlan;
-	struct efx_filter_spec *spec;
-	struct efx_rss_context *ctx;
-	unsigned int filter_idx;
-	u32 mcdi_flags;
-	int match_pri;
-	int rc, i;
-
-	WARN_ON(!rwsem_is_locked(&efx->filter_sem));
-
-	if (!nic_data->must_restore_filters)
-		return;
-
-	if (!table)
-		return;
-
-	down_write(&table->lock);
-	mutex_lock(&efx->rss_lock);
-
-	for (filter_idx = 0; filter_idx < EFX_MCDI_FILTER_TBL_ROWS; filter_idx++) {
-		spec = efx_ef10_filter_entry_spec(table, filter_idx);
-		if (!spec)
-			continue;
-
-		mcdi_flags = efx_mcdi_filter_mcdi_flags_from_spec(spec);
-		match_pri = 0;
-		while (match_pri < table->rx_match_count &&
-		       table->rx_match_mcdi_flags[match_pri] != mcdi_flags)
-			++match_pri;
-		if (match_pri >= table->rx_match_count) {
-			invalid_filters++;
-			goto not_restored;
-		}
-		if (spec->rss_context)
-			ctx = efx_find_rss_context_entry(efx, spec->rss_context);
-		else
-			ctx = &efx->rss_context;
-		if (spec->flags & EFX_FILTER_FLAG_RX_RSS) {
-			if (!ctx) {
-				netif_warn(efx, drv, efx->net_dev,
-					   "Warning: unable to restore a filter with nonexistent RSS context %u.\n",
-					   spec->rss_context);
-				invalid_filters++;
-				goto not_restored;
-			}
-			if (ctx->context_id == EFX_MCDI_RSS_CONTEXT_INVALID) {
-				netif_warn(efx, drv, efx->net_dev,
-					   "Warning: unable to restore a filter with RSS context %u as it was not created.\n",
-					   spec->rss_context);
-				invalid_filters++;
-				goto not_restored;
-			}
-		}
-
-		rc = efx_mcdi_filter_push(efx, spec,
-					  &table->entry[filter_idx].handle,
-					  ctx, false);
-		if (rc)
-			failed++;
-
-		if (rc) {
-not_restored:
-			list_for_each_entry(vlan, &table->vlan_list, list)
-				for (i = 0; i < EFX_EF10_NUM_DEFAULT_FILTERS; ++i)
-					if (vlan->default_filters[i] == filter_idx)
-						vlan->default_filters[i] =
-							EFX_EF10_FILTER_ID_INVALID;
-
-			kfree(spec);
-			efx_ef10_filter_set_entry(table, filter_idx, NULL, 0);
-		}
-	}
-
-	mutex_unlock(&efx->rss_lock);
-	up_write(&table->lock);
-
-	/* This can happen validly if the MC's capabilities have changed, so
-	 * is not an error.
-	 */
-	if (invalid_filters)
-		netif_dbg(efx, drv, efx->net_dev,
-			  "Did not restore %u filters that are now unsupported.\n",
-			  invalid_filters);
-
-	if (failed)
-		netif_err(efx, hw, efx->net_dev,
-			  "unable to restore %u filters\n", failed);
-	else
-		nic_data->must_restore_filters = false;
-}
-
-void efx_mcdi_filter_table_remove(struct efx_nic *efx)
-{
-	struct efx_mcdi_filter_table *table = efx->filter_state;
-	MCDI_DECLARE_BUF(inbuf, MC_CMD_FILTER_OP_EXT_IN_LEN);
-	struct efx_filter_spec *spec;
-	unsigned int filter_idx;
-	int rc;
-
-	efx_mcdi_filter_cleanup_vlans(efx);
-	efx->filter_state = NULL;
-	/* If we were called without locking, then it's not safe to free
-	 * the table as others might be using it.  So we just WARN, leak
-	 * the memory, and potentially get an inconsistent filter table
-	 * state.
-	 * This should never actually happen.
-	 */
-	if (!efx_rwsem_assert_write_locked(&efx->filter_sem))
-		return;
-
-	if (!table)
-		return;
-
-	for (filter_idx = 0; filter_idx < EFX_MCDI_FILTER_TBL_ROWS; filter_idx++) {
-		spec = efx_ef10_filter_entry_spec(table, filter_idx);
-		if (!spec)
-			continue;
-
-		MCDI_SET_DWORD(inbuf, FILTER_OP_IN_OP,
-			       efx_ef10_filter_is_exclusive(spec) ?
-			       MC_CMD_FILTER_OP_IN_OP_REMOVE :
-			       MC_CMD_FILTER_OP_IN_OP_UNSUBSCRIBE);
-		MCDI_SET_QWORD(inbuf, FILTER_OP_IN_HANDLE,
-			       table->entry[filter_idx].handle);
-		rc = efx_mcdi_rpc_quiet(efx, MC_CMD_FILTER_OP, inbuf,
-					sizeof(inbuf), NULL, 0, NULL);
-		if (rc)
-			netif_info(efx, drv, efx->net_dev,
-				   "%s: filter %04x remove failed\n",
-				   __func__, filter_idx);
-		kfree(spec);
-	}
-
-	vfree(table->entry);
-	kfree(table);
-}
-
-static void efx_mcdi_filter_mark_one_old(struct efx_nic *efx, uint16_t *id)
-{
-	struct efx_mcdi_filter_table *table = efx->filter_state;
-	unsigned int filter_idx;
-
-	efx_rwsem_assert_write_locked(&table->lock);
-
-	if (*id != EFX_EF10_FILTER_ID_INVALID) {
-		filter_idx = efx_ef10_filter_get_unsafe_id(*id);
-		if (!table->entry[filter_idx].spec)
-			netif_dbg(efx, drv, efx->net_dev,
-				  "marked null spec old %04x:%04x\n", *id,
-				  filter_idx);
-		table->entry[filter_idx].spec |= EFX_EF10_FILTER_FLAG_AUTO_OLD;
-		*id = EFX_EF10_FILTER_ID_INVALID;
-	}
-}
-
-/* Mark old per-VLAN filters that may need to be removed */
-static void _efx_mcdi_filter_vlan_mark_old(struct efx_nic *efx,
-					   struct efx_mcdi_filter_vlan *vlan)
-{
-	struct efx_mcdi_filter_table *table = efx->filter_state;
-	unsigned int i;
-
-	for (i = 0; i < table->dev_uc_count; i++)
-		efx_mcdi_filter_mark_one_old(efx, &vlan->uc[i]);
-	for (i = 0; i < table->dev_mc_count; i++)
-		efx_mcdi_filter_mark_one_old(efx, &vlan->mc[i]);
-	for (i = 0; i < EFX_EF10_NUM_DEFAULT_FILTERS; i++)
-		efx_mcdi_filter_mark_one_old(efx, &vlan->default_filters[i]);
-}
-
-/* Mark old filters that may need to be removed.
- * Caller must hold efx->filter_sem for read if race against
- * efx_mcdi_filter_table_remove() is possible
- */
-static void efx_mcdi_filter_mark_old(struct efx_nic *efx)
-{
-	struct efx_mcdi_filter_table *table = efx->filter_state;
-	struct efx_mcdi_filter_vlan *vlan;
-
-	down_write(&table->lock);
-	list_for_each_entry(vlan, &table->vlan_list, list)
-		_efx_mcdi_filter_vlan_mark_old(efx, vlan);
-	up_write(&table->lock);
-}
-
-static void efx_mcdi_filter_uc_addr_list(struct efx_nic *efx)
-{
-	struct efx_mcdi_filter_table *table = efx->filter_state;
-	struct net_device *net_dev = efx->net_dev;
-	struct netdev_hw_addr *uc;
-	unsigned int i;
-
-	table->uc_promisc = !!(net_dev->flags & IFF_PROMISC);
-	ether_addr_copy(table->dev_uc_list[0].addr, net_dev->dev_addr);
-	i = 1;
-	netdev_for_each_uc_addr(uc, net_dev) {
-		if (i >= EFX_EF10_FILTER_DEV_UC_MAX) {
-			table->uc_promisc = true;
-			break;
-		}
-		ether_addr_copy(table->dev_uc_list[i].addr, uc->addr);
-		i++;
-	}
-
-	table->dev_uc_count = i;
-}
-
-static void efx_mcdi_filter_mc_addr_list(struct efx_nic *efx)
-{
-	struct efx_mcdi_filter_table *table = efx->filter_state;
-	struct net_device *net_dev = efx->net_dev;
-	struct netdev_hw_addr *mc;
-	unsigned int i;
-
-	table->mc_overflow = false;
-	table->mc_promisc = !!(net_dev->flags & (IFF_PROMISC | IFF_ALLMULTI));
-
-	i = 0;
-	netdev_for_each_mc_addr(mc, net_dev) {
-		if (i >= EFX_EF10_FILTER_DEV_MC_MAX) {
-			table->mc_promisc = true;
-			table->mc_overflow = true;
-			break;
-		}
-		ether_addr_copy(table->dev_mc_list[i].addr, mc->addr);
-		i++;
-	}
-
-	table->dev_mc_count = i;
-}
-
-static int efx_mcdi_filter_insert_addr_list(struct efx_nic *efx,
-					    struct efx_mcdi_filter_vlan *vlan,
-					    bool multicast, bool rollback)
-{
-	struct efx_mcdi_filter_table *table = efx->filter_state;
-	struct efx_mcdi_dev_addr *addr_list;
-	enum efx_filter_flags filter_flags;
-	struct efx_filter_spec spec;
-	u8 baddr[ETH_ALEN];
-	unsigned int i, j;
-	int addr_count;
-	u16 *ids;
-	int rc;
-
-	if (multicast) {
-		addr_list = table->dev_mc_list;
-		addr_count = table->dev_mc_count;
-		ids = vlan->mc;
-	} else {
-		addr_list = table->dev_uc_list;
-		addr_count = table->dev_uc_count;
-		ids = vlan->uc;
-	}
-
-	filter_flags = efx_rss_active(&efx->rss_context) ? EFX_FILTER_FLAG_RX_RSS : 0;
-
-	/* Insert/renew filters */
-	for (i = 0; i < addr_count; i++) {
-		EFX_WARN_ON_PARANOID(ids[i] != EFX_EF10_FILTER_ID_INVALID);
-		efx_filter_init_rx(&spec, EFX_FILTER_PRI_AUTO, filter_flags, 0);
-		efx_filter_set_eth_local(&spec, vlan->vid, addr_list[i].addr);
-		rc = efx_mcdi_filter_insert_locked(efx, &spec, true);
-		if (rc < 0) {
-			if (rollback) {
-				netif_info(efx, drv, efx->net_dev,
-					   "efx_ef10_filter_insert failed rc=%d\n",
-					   rc);
-				/* Fall back to promiscuous */
-				for (j = 0; j < i; j++) {
-					efx_mcdi_filter_remove_unsafe(
-						efx, EFX_FILTER_PRI_AUTO,
-						ids[j]);
-					ids[j] = EFX_EF10_FILTER_ID_INVALID;
-				}
-				return rc;
-			} else {
-				/* keep invalid ID, and carry on */
-			}
-		} else {
-			ids[i] = efx_ef10_filter_get_unsafe_id(rc);
-		}
-	}
-
-	if (multicast && rollback) {
-		/* Also need an Ethernet broadcast filter */
-		EFX_WARN_ON_PARANOID(vlan->default_filters[EFX_EF10_BCAST] !=
-				     EFX_EF10_FILTER_ID_INVALID);
-		efx_filter_init_rx(&spec, EFX_FILTER_PRI_AUTO, filter_flags, 0);
-		eth_broadcast_addr(baddr);
-		efx_filter_set_eth_local(&spec, vlan->vid, baddr);
-		rc = efx_mcdi_filter_insert_locked(efx, &spec, true);
-		if (rc < 0) {
-			netif_warn(efx, drv, efx->net_dev,
-				   "Broadcast filter insert failed rc=%d\n", rc);
-			/* Fall back to promiscuous */
-			for (j = 0; j < i; j++) {
-				efx_mcdi_filter_remove_unsafe(
-					efx, EFX_FILTER_PRI_AUTO,
-					ids[j]);
-				ids[j] = EFX_EF10_FILTER_ID_INVALID;
-			}
-			return rc;
-		} else {
-			vlan->default_filters[EFX_EF10_BCAST] =
-				efx_ef10_filter_get_unsafe_id(rc);
-		}
-	}
-
-	return 0;
-}
-
-static int efx_mcdi_filter_insert_def(struct efx_nic *efx,
-				      struct efx_mcdi_filter_vlan *vlan,
-				      enum efx_encap_type encap_type,
-				      bool multicast, bool rollback)
-{
-	struct efx_ef10_nic_data *nic_data = efx->nic_data;
-	enum efx_filter_flags filter_flags;
-	struct efx_filter_spec spec;
-	u8 baddr[ETH_ALEN];
-	int rc;
-	u16 *id;
-
-	filter_flags = efx_rss_active(&efx->rss_context) ? EFX_FILTER_FLAG_RX_RSS : 0;
-
-	efx_filter_init_rx(&spec, EFX_FILTER_PRI_AUTO, filter_flags, 0);
-
-	if (multicast)
-		efx_filter_set_mc_def(&spec);
-	else
-		efx_filter_set_uc_def(&spec);
-
-	if (encap_type) {
-		if (nic_data->datapath_caps &
-		    (1 << MC_CMD_GET_CAPABILITIES_OUT_VXLAN_NVGRE_LBN))
-			efx_filter_set_encap_type(&spec, encap_type);
-		else
-			/* don't insert encap filters on non-supporting
-			 * platforms. ID will be left as INVALID.
-			 */
-			return 0;
-	}
-
-	if (vlan->vid != EFX_FILTER_VID_UNSPEC)
-		efx_filter_set_eth_local(&spec, vlan->vid, NULL);
-
-	rc = efx_mcdi_filter_insert_locked(efx, &spec, true);
-	if (rc < 0) {
-		const char *um = multicast ? "Multicast" : "Unicast";
-		const char *encap_name = "";
-		const char *encap_ipv = "";
-
-		if ((encap_type & EFX_ENCAP_TYPES_MASK) ==
-		    EFX_ENCAP_TYPE_VXLAN)
-			encap_name = "VXLAN ";
-		else if ((encap_type & EFX_ENCAP_TYPES_MASK) ==
-			 EFX_ENCAP_TYPE_NVGRE)
-			encap_name = "NVGRE ";
-		else if ((encap_type & EFX_ENCAP_TYPES_MASK) ==
-			 EFX_ENCAP_TYPE_GENEVE)
-			encap_name = "GENEVE ";
-		if (encap_type & EFX_ENCAP_FLAG_IPV6)
-			encap_ipv = "IPv6 ";
-		else if (encap_type)
-			encap_ipv = "IPv4 ";
-
-		/* unprivileged functions can't insert mismatch filters
-		 * for encapsulated or unicast traffic, so downgrade
-		 * those warnings to debug.
-		 */
-		netif_cond_dbg(efx, drv, efx->net_dev,
-			       rc == -EPERM && (encap_type || !multicast), warn,
-			       "%s%s%s mismatch filter insert failed rc=%d\n",
-			       encap_name, encap_ipv, um, rc);
-	} else if (multicast) {
-		/* mapping from encap types to default filter IDs (multicast) */
-		static enum efx_mcdi_filter_default_filters map[] = {
-			[EFX_ENCAP_TYPE_NONE] = EFX_EF10_MCDEF,
-			[EFX_ENCAP_TYPE_VXLAN] = EFX_EF10_VXLAN4_MCDEF,
-			[EFX_ENCAP_TYPE_NVGRE] = EFX_EF10_NVGRE4_MCDEF,
-			[EFX_ENCAP_TYPE_GENEVE] = EFX_EF10_GENEVE4_MCDEF,
-			[EFX_ENCAP_TYPE_VXLAN | EFX_ENCAP_FLAG_IPV6] =
-				EFX_EF10_VXLAN6_MCDEF,
-			[EFX_ENCAP_TYPE_NVGRE | EFX_ENCAP_FLAG_IPV6] =
-				EFX_EF10_NVGRE6_MCDEF,
-			[EFX_ENCAP_TYPE_GENEVE | EFX_ENCAP_FLAG_IPV6] =
-				EFX_EF10_GENEVE6_MCDEF,
-		};
-
-		/* quick bounds check (BCAST result impossible) */
-		BUILD_BUG_ON(EFX_EF10_BCAST != 0);
-		if (encap_type >= ARRAY_SIZE(map) || map[encap_type] == 0) {
-			WARN_ON(1);
-			return -EINVAL;
-		}
-		/* then follow map */
-		id = &vlan->default_filters[map[encap_type]];
-
-		EFX_WARN_ON_PARANOID(*id != EFX_EF10_FILTER_ID_INVALID);
-		*id = efx_ef10_filter_get_unsafe_id(rc);
-		if (!nic_data->workaround_26807 && !encap_type) {
-			/* Also need an Ethernet broadcast filter */
-			efx_filter_init_rx(&spec, EFX_FILTER_PRI_AUTO,
-					   filter_flags, 0);
-			eth_broadcast_addr(baddr);
-			efx_filter_set_eth_local(&spec, vlan->vid, baddr);
-			rc = efx_mcdi_filter_insert_locked(efx, &spec, true);
-			if (rc < 0) {
-				netif_warn(efx, drv, efx->net_dev,
-					   "Broadcast filter insert failed rc=%d\n",
-					   rc);
-				if (rollback) {
-					/* Roll back the mc_def filter */
-					efx_mcdi_filter_remove_unsafe(
-							efx, EFX_FILTER_PRI_AUTO,
-							*id);
-					*id = EFX_EF10_FILTER_ID_INVALID;
-					return rc;
-				}
-			} else {
-				EFX_WARN_ON_PARANOID(
-					vlan->default_filters[EFX_EF10_BCAST] !=
-					EFX_EF10_FILTER_ID_INVALID);
-				vlan->default_filters[EFX_EF10_BCAST] =
-					efx_ef10_filter_get_unsafe_id(rc);
-			}
-		}
-		rc = 0;
-	} else {
-		/* mapping from encap types to default filter IDs (unicast) */
-		static enum efx_mcdi_filter_default_filters map[] = {
-			[EFX_ENCAP_TYPE_NONE] = EFX_EF10_UCDEF,
-			[EFX_ENCAP_TYPE_VXLAN] = EFX_EF10_VXLAN4_UCDEF,
-			[EFX_ENCAP_TYPE_NVGRE] = EFX_EF10_NVGRE4_UCDEF,
-			[EFX_ENCAP_TYPE_GENEVE] = EFX_EF10_GENEVE4_UCDEF,
-			[EFX_ENCAP_TYPE_VXLAN | EFX_ENCAP_FLAG_IPV6] =
-				EFX_EF10_VXLAN6_UCDEF,
-			[EFX_ENCAP_TYPE_NVGRE | EFX_ENCAP_FLAG_IPV6] =
-				EFX_EF10_NVGRE6_UCDEF,
-			[EFX_ENCAP_TYPE_GENEVE | EFX_ENCAP_FLAG_IPV6] =
-				EFX_EF10_GENEVE6_UCDEF,
-		};
-
-		/* quick bounds check (BCAST result impossible) */
-		BUILD_BUG_ON(EFX_EF10_BCAST != 0);
-		if (encap_type >= ARRAY_SIZE(map) || map[encap_type] == 0) {
-			WARN_ON(1);
-			return -EINVAL;
-		}
-		/* then follow map */
-		id = &vlan->default_filters[map[encap_type]];
-		EFX_WARN_ON_PARANOID(*id != EFX_EF10_FILTER_ID_INVALID);
-		*id = rc;
-		rc = 0;
-	}
-	return rc;
-}
-
-/* Remove filters that weren't renewed. */
-static void efx_mcdi_filter_remove_old(struct efx_nic *efx)
-{
-	struct efx_mcdi_filter_table *table = efx->filter_state;
-	int remove_failed = 0;
-	int remove_noent = 0;
-	int rc;
-	int i;
-
-	down_write(&table->lock);
-	for (i = 0; i < EFX_MCDI_FILTER_TBL_ROWS; i++) {
-		if (READ_ONCE(table->entry[i].spec) &
-		    EFX_EF10_FILTER_FLAG_AUTO_OLD) {
-			rc = efx_mcdi_filter_remove_internal(efx,
-					1U << EFX_FILTER_PRI_AUTO, i, true);
-			if (rc == -ENOENT)
-				remove_noent++;
-			else if (rc)
-				remove_failed++;
-		}
-	}
-	up_write(&table->lock);
-
-	if (remove_failed)
-		netif_info(efx, drv, efx->net_dev,
-			   "%s: failed to remove %d filters\n",
-			   __func__, remove_failed);
-	if (remove_noent)
-		netif_info(efx, drv, efx->net_dev,
-			   "%s: failed to remove %d non-existent filters\n",
-			   __func__, remove_noent);
-}
-
-static int efx_ef10_vport_set_mac_address(struct efx_nic *efx)
-{
-	struct efx_ef10_nic_data *nic_data = efx->nic_data;
-	u8 mac_old[ETH_ALEN];
-	int rc, rc2;
-
-	/* Only reconfigure a PF-created vport */
-	if (is_zero_ether_addr(nic_data->vport_mac))
-		return 0;
-
-	efx_device_detach_sync(efx);
-	efx_net_stop(efx->net_dev);
-	down_write(&efx->filter_sem);
-	efx_mcdi_filter_table_remove(efx);
-	up_write(&efx->filter_sem);
-
-	rc = efx_ef10_vadaptor_free(efx, nic_data->vport_id);
-	if (rc)
-		goto restore_filters;
-
-	ether_addr_copy(mac_old, nic_data->vport_mac);
-	rc = efx_ef10_vport_del_mac(efx, nic_data->vport_id,
-				    nic_data->vport_mac);
-	if (rc)
-		goto restore_vadaptor;
-
-	rc = efx_ef10_vport_add_mac(efx, nic_data->vport_id,
-				    efx->net_dev->dev_addr);
-	if (!rc) {
-		ether_addr_copy(nic_data->vport_mac, efx->net_dev->dev_addr);
-	} else {
-		rc2 = efx_ef10_vport_add_mac(efx, nic_data->vport_id, mac_old);
-		if (rc2) {
-			/* Failed to add original MAC, so clear vport_mac */
-			eth_zero_addr(nic_data->vport_mac);
-			goto reset_nic;
-		}
-	}
-
-restore_vadaptor:
-	rc2 = efx_ef10_vadaptor_alloc(efx, nic_data->vport_id);
-	if (rc2)
-		goto reset_nic;
-restore_filters:
-	down_write(&efx->filter_sem);
-	rc2 = efx_mcdi_filter_table_probe(efx);
-	up_write(&efx->filter_sem);
-	if (rc2)
-		goto reset_nic;
-
-	rc2 = efx_net_open(efx->net_dev);
-	if (rc2)
-		goto reset_nic;
-
-	efx_device_attach_if_not_resetting(efx);
+	efx_device_attach_if_not_resetting(efx);
 
 	return rc;
 
@@ -5200,257 +3206,6 @@ reset_nic:
 	return rc ? rc : rc2;
 }
 
-/* Caller must hold efx->filter_sem for read if race against
- * efx_mcdi_filter_table_remove() is possible
- */
-static void efx_mcdi_filter_vlan_sync_rx_mode(struct efx_nic *efx,
-					      struct efx_mcdi_filter_vlan *vlan)
-{
-	struct efx_mcdi_filter_table *table = efx->filter_state;
-	struct efx_ef10_nic_data *nic_data = efx->nic_data;
-
-	/* Do not install unspecified VID if VLAN filtering is enabled.
-	 * Do not install all specified VIDs if VLAN filtering is disabled.
-	 */
-	if ((vlan->vid == EFX_FILTER_VID_UNSPEC) == table->vlan_filter)
-		return;
-
-	/* Insert/renew unicast filters */
-	if (table->uc_promisc) {
-		efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_NONE,
-					   false, false);
-		efx_mcdi_filter_insert_addr_list(efx, vlan, false, false);
-	} else {
-		/* If any of the filters failed to insert, fall back to
-		 * promiscuous mode - add in the uc_def filter.  But keep
-		 * our individual unicast filters.
-		 */
-		if (efx_mcdi_filter_insert_addr_list(efx, vlan, false, false))
-			efx_mcdi_filter_insert_def(efx, vlan,
-						   EFX_ENCAP_TYPE_NONE,
-						   false, false);
-	}
-	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_VXLAN,
-				   false, false);
-	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_VXLAN |
-					      EFX_ENCAP_FLAG_IPV6,
-				   false, false);
-	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_NVGRE,
-				   false, false);
-	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_NVGRE |
-					      EFX_ENCAP_FLAG_IPV6,
-				   false, false);
-	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_GENEVE,
-				   false, false);
-	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_GENEVE |
-					      EFX_ENCAP_FLAG_IPV6,
-				   false, false);
-
-	/* Insert/renew multicast filters */
-	/* If changing promiscuous state with cascaded multicast filters, remove
-	 * old filters first, so that packets are dropped rather than duplicated
-	 */
-	if (nic_data->workaround_26807 &&
-	    table->mc_promisc_last != table->mc_promisc)
-		efx_mcdi_filter_remove_old(efx);
-	if (table->mc_promisc) {
-		if (nic_data->workaround_26807) {
-			/* If we failed to insert promiscuous filters, rollback
-			 * and fall back to individual multicast filters
-			 */
-			if (efx_mcdi_filter_insert_def(efx, vlan,
-						       EFX_ENCAP_TYPE_NONE,
-						       true, true)) {
-				/* Changing promisc state, so remove old filters */
-				efx_mcdi_filter_remove_old(efx);
-				efx_mcdi_filter_insert_addr_list(efx, vlan,
-								 true, false);
-			}
-		} else {
-			/* If we failed to insert promiscuous filters, don't
-			 * rollback.  Regardless, also insert the mc_list,
-			 * unless it's incomplete due to overflow
-			 */
-			efx_mcdi_filter_insert_def(efx, vlan,
-						   EFX_ENCAP_TYPE_NONE,
-						   true, false);
-			if (!table->mc_overflow)
-				efx_mcdi_filter_insert_addr_list(efx, vlan,
-								 true, false);
-		}
-	} else {
-		/* If any filters failed to insert, rollback and fall back to
-		 * promiscuous mode - mc_def filter and maybe broadcast.  If
-		 * that fails, roll back again and insert as many of our
-		 * individual multicast filters as we can.
-		 */
-		if (efx_mcdi_filter_insert_addr_list(efx, vlan, true, true)) {
-			/* Changing promisc state, so remove old filters */
-			if (nic_data->workaround_26807)
-				efx_mcdi_filter_remove_old(efx);
-			if (efx_mcdi_filter_insert_def(efx, vlan,
-						       EFX_ENCAP_TYPE_NONE,
-						       true, true))
-				efx_mcdi_filter_insert_addr_list(efx, vlan,
-								 true, false);
-		}
-	}
-	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_VXLAN,
-				   true, false);
-	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_VXLAN |
-					      EFX_ENCAP_FLAG_IPV6,
-				   true, false);
-	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_NVGRE,
-				   true, false);
-	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_NVGRE |
-					      EFX_ENCAP_FLAG_IPV6,
-				   true, false);
-	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_GENEVE,
-				   true, false);
-	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_GENEVE |
-					      EFX_ENCAP_FLAG_IPV6,
-				   true, false);
-}
-
-/* Caller must hold efx->filter_sem for read if race against
- * efx_mcdi_filter_table_remove() is possible
- */
-void efx_mcdi_filter_sync_rx_mode(struct efx_nic *efx)
-{
-	struct efx_mcdi_filter_table *table = efx->filter_state;
-	struct net_device *net_dev = efx->net_dev;
-	struct efx_mcdi_filter_vlan *vlan;
-	bool vlan_filter;
-
-	if (!efx_dev_registered(efx))
-		return;
-
-	if (!table)
-		return;
-
-	efx_mcdi_filter_mark_old(efx);
-
-	/* Copy/convert the address lists; add the primary station
-	 * address and broadcast address
-	 */
-	netif_addr_lock_bh(net_dev);
-	efx_mcdi_filter_uc_addr_list(efx);
-	efx_mcdi_filter_mc_addr_list(efx);
-	netif_addr_unlock_bh(net_dev);
-
-	/* If VLAN filtering changes, all old filters are finally removed.
-	 * Do it in advance to avoid conflicts for unicast untagged and
-	 * VLAN 0 tagged filters.
-	 */
-	vlan_filter = !!(net_dev->features & NETIF_F_HW_VLAN_CTAG_FILTER);
-	if (table->vlan_filter != vlan_filter) {
-		table->vlan_filter = vlan_filter;
-		efx_mcdi_filter_remove_old(efx);
-	}
-
-	list_for_each_entry(vlan, &table->vlan_list, list)
-		efx_mcdi_filter_vlan_sync_rx_mode(efx, vlan);
-
-	efx_mcdi_filter_remove_old(efx);
-	table->mc_promisc_last = table->mc_promisc;
-}
-
-struct efx_mcdi_filter_vlan *efx_mcdi_filter_find_vlan(struct efx_nic *efx,
-						       u16 vid)
-{
-	struct efx_mcdi_filter_table *table = efx->filter_state;
-	struct efx_mcdi_filter_vlan *vlan;
-
-	WARN_ON(!rwsem_is_locked(&efx->filter_sem));
-
-	list_for_each_entry(vlan, &table->vlan_list, list) {
-		if (vlan->vid == vid)
-			return vlan;
-	}
-
-	return NULL;
-}
-
-int efx_mcdi_filter_add_vlan(struct efx_nic *efx, u16 vid)
-{
-	struct efx_mcdi_filter_table *table = efx->filter_state;
-	struct efx_mcdi_filter_vlan *vlan;
-	unsigned int i;
-
-	if (!efx_rwsem_assert_write_locked(&efx->filter_sem))
-		return -EINVAL;
-
-	vlan = efx_mcdi_filter_find_vlan(efx, vid);
-	if (WARN_ON(vlan)) {
-		netif_err(efx, drv, efx->net_dev,
-			  "VLAN %u already added\n", vid);
-		return -EALREADY;
-	}
-
-	vlan = kzalloc(sizeof(*vlan), GFP_KERNEL);
-	if (!vlan)
-		return -ENOMEM;
-
-	vlan->vid = vid;
-
-	for (i = 0; i < ARRAY_SIZE(vlan->uc); i++)
-		vlan->uc[i] = EFX_EF10_FILTER_ID_INVALID;
-	for (i = 0; i < ARRAY_SIZE(vlan->mc); i++)
-		vlan->mc[i] = EFX_EF10_FILTER_ID_INVALID;
-	for (i = 0; i < EFX_EF10_NUM_DEFAULT_FILTERS; i++)
-		vlan->default_filters[i] = EFX_EF10_FILTER_ID_INVALID;
-
-	list_add_tail(&vlan->list, &table->vlan_list);
-
-	if (efx_dev_registered(efx))
-		efx_mcdi_filter_vlan_sync_rx_mode(efx, vlan);
-
-	return 0;
-}
-
-static void efx_mcdi_filter_del_vlan_internal(struct efx_nic *efx,
-					      struct efx_mcdi_filter_vlan *vlan)
-{
-	unsigned int i;
-
-	/* See comment in efx_mcdi_filter_table_remove() */
-	if (!efx_rwsem_assert_write_locked(&efx->filter_sem))
-		return;
-
-	list_del(&vlan->list);
-
-	for (i = 0; i < ARRAY_SIZE(vlan->uc); i++)
-		efx_mcdi_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO,
-					      vlan->uc[i]);
-	for (i = 0; i < ARRAY_SIZE(vlan->mc); i++)
-		efx_mcdi_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO,
-					      vlan->mc[i]);
-	for (i = 0; i < EFX_EF10_NUM_DEFAULT_FILTERS; i++)
-		if (vlan->default_filters[i] != EFX_EF10_FILTER_ID_INVALID)
-			efx_mcdi_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO,
-						      vlan->default_filters[i]);
-
-	kfree(vlan);
-}
-
-void efx_mcdi_filter_del_vlan(struct efx_nic *efx, u16 vid)
-{
-	struct efx_mcdi_filter_vlan *vlan;
-
-	/* See comment in efx_mcdi_filter_table_remove() */
-	if (!efx_rwsem_assert_write_locked(&efx->filter_sem))
-		return;
-
-	vlan = efx_mcdi_filter_find_vlan(efx, vid);
-	if (!vlan) {
-		netif_err(efx, drv, efx->net_dev,
-			  "VLAN %u not found in filter state\n", vid);
-		return;
-	}
-
-	efx_mcdi_filter_del_vlan_internal(efx, vlan);
-}
-
 static int efx_ef10_set_mac_address(struct efx_nic *efx)
 {
 	MCDI_DECLARE_BUF(inbuf, MC_CMD_VADAPTOR_SET_MAC_IN_LEN);
diff --git a/drivers/net/ethernet/sfc/mcdi_filters.c b/drivers/net/ethernet/sfc/mcdi_filters.c
new file mode 100644
index 000000000000..4310ae5bd898
--- /dev/null
+++ b/drivers/net/ethernet/sfc/mcdi_filters.c
@@ -0,0 +1,2270 @@
+#include "mcdi_filters.h"
+#include "mcdi.h"
+#include "nic.h"
+#include "rx_common.h"
+
+/* The maximum size of a shared RSS context */
+/* TODO: this should really be from the mcdi protocol export */
+#define EFX_EF10_MAX_SHARED_RSS_CONTEXT_SIZE 64UL
+
+#define EFX_EF10_FILTER_ID_INVALID 0xffff
+
+/* An arbitrary search limit for the software hash table */
+#define EFX_EF10_FILTER_SEARCH_LIMIT 200
+
+static struct efx_filter_spec *
+efx_mcdi_filter_entry_spec(const struct efx_mcdi_filter_table *table,
+			   unsigned int filter_idx)
+{
+	return (struct efx_filter_spec *)(table->entry[filter_idx].spec &
+					  ~EFX_EF10_FILTER_FLAGS);
+}
+
+static unsigned int
+efx_mcdi_filter_entry_flags(const struct efx_mcdi_filter_table *table,
+			   unsigned int filter_idx)
+{
+	return table->entry[filter_idx].spec & EFX_EF10_FILTER_FLAGS;
+}
+
+static u32 efx_mcdi_filter_get_unsafe_id(u32 filter_id)
+{
+	WARN_ON_ONCE(filter_id == EFX_EF10_FILTER_ID_INVALID);
+	return filter_id & (EFX_MCDI_FILTER_TBL_ROWS - 1);
+}
+
+static unsigned int efx_mcdi_filter_get_unsafe_pri(u32 filter_id)
+{
+	return filter_id / (EFX_MCDI_FILTER_TBL_ROWS * 2);
+}
+
+static u32 efx_mcdi_filter_make_filter_id(unsigned int pri, u16 idx)
+{
+	return pri * EFX_MCDI_FILTER_TBL_ROWS * 2 + idx;
+}
+
+/*
+ * Decide whether a filter should be exclusive or else should allow
+ * delivery to additional recipients.  Currently we decide that
+ * filters for specific local unicast MAC and IP addresses are
+ * exclusive.
+ */
+static bool efx_mcdi_filter_is_exclusive(const struct efx_filter_spec *spec)
+{
+	if (spec->match_flags & EFX_FILTER_MATCH_LOC_MAC &&
+	    !is_multicast_ether_addr(spec->loc_mac))
+		return true;
+
+	if ((spec->match_flags &
+	     (EFX_FILTER_MATCH_ETHER_TYPE | EFX_FILTER_MATCH_LOC_HOST)) ==
+	    (EFX_FILTER_MATCH_ETHER_TYPE | EFX_FILTER_MATCH_LOC_HOST)) {
+		if (spec->ether_type == htons(ETH_P_IP) &&
+		    !ipv4_is_multicast(spec->loc_host[0]))
+			return true;
+		if (spec->ether_type == htons(ETH_P_IPV6) &&
+		    ((const u8 *)spec->loc_host)[0] != 0xff)
+			return true;
+	}
+
+	return false;
+}
+
+static void
+efx_mcdi_filter_set_entry(struct efx_mcdi_filter_table *table,
+			  unsigned int filter_idx,
+			  const struct efx_filter_spec *spec,
+			  unsigned int flags)
+{
+	table->entry[filter_idx].spec =	(unsigned long)spec | flags;
+}
+
+static void
+efx_mcdi_filter_push_prep_set_match_fields(struct efx_nic *efx,
+					   const struct efx_filter_spec *spec,
+					   efx_dword_t *inbuf)
+{
+	enum efx_encap_type encap_type = efx_filter_get_encap_type(spec);
+	u32 match_fields = 0, uc_match, mc_match;
+
+	MCDI_SET_DWORD(inbuf, FILTER_OP_IN_OP,
+		       efx_mcdi_filter_is_exclusive(spec) ?
+		       MC_CMD_FILTER_OP_IN_OP_INSERT :
+		       MC_CMD_FILTER_OP_IN_OP_SUBSCRIBE);
+
+	/*
+	 * Convert match flags and values.  Unlike almost
+	 * everything else in MCDI, these fields are in
+	 * network byte order.
+	 */
+#define COPY_VALUE(value, mcdi_field)					     \
+	do {							     \
+		match_fields |=					     \
+			1 << MC_CMD_FILTER_OP_IN_MATCH_ ##	     \
+			mcdi_field ## _LBN;			     \
+		BUILD_BUG_ON(					     \
+			MC_CMD_FILTER_OP_IN_ ## mcdi_field ## _LEN < \
+			sizeof(value));				     \
+		memcpy(MCDI_PTR(inbuf, FILTER_OP_IN_ ##	mcdi_field), \
+		       &value, sizeof(value));			     \
+	} while (0)
+#define COPY_FIELD(gen_flag, gen_field, mcdi_field)			     \
+	if (spec->match_flags & EFX_FILTER_MATCH_ ## gen_flag) {     \
+		COPY_VALUE(spec->gen_field, mcdi_field);	     \
+	}
+	/*
+	 * Handle encap filters first.  They will always be mismatch
+	 * (unknown UC or MC) filters
+	 */
+	if (encap_type) {
+		/*
+		 * ether_type and outer_ip_proto need to be variables
+		 * because COPY_VALUE wants to memcpy them
+		 */
+		__be16 ether_type =
+			htons(encap_type & EFX_ENCAP_FLAG_IPV6 ?
+			      ETH_P_IPV6 : ETH_P_IP);
+		u8 vni_type = MC_CMD_FILTER_OP_EXT_IN_VNI_TYPE_GENEVE;
+		u8 outer_ip_proto;
+
+		switch (encap_type & EFX_ENCAP_TYPES_MASK) {
+		case EFX_ENCAP_TYPE_VXLAN:
+			vni_type = MC_CMD_FILTER_OP_EXT_IN_VNI_TYPE_VXLAN;
+			/* fallthrough */
+		case EFX_ENCAP_TYPE_GENEVE:
+			COPY_VALUE(ether_type, ETHER_TYPE);
+			outer_ip_proto = IPPROTO_UDP;
+			COPY_VALUE(outer_ip_proto, IP_PROTO);
+			/*
+			 * We always need to set the type field, even
+			 * though we're not matching on the TNI.
+			 */
+			MCDI_POPULATE_DWORD_1(inbuf,
+				FILTER_OP_EXT_IN_VNI_OR_VSID,
+				FILTER_OP_EXT_IN_VNI_TYPE,
+				vni_type);
+			break;
+		case EFX_ENCAP_TYPE_NVGRE:
+			COPY_VALUE(ether_type, ETHER_TYPE);
+			outer_ip_proto = IPPROTO_GRE;
+			COPY_VALUE(outer_ip_proto, IP_PROTO);
+			break;
+		default:
+			WARN_ON(1);
+		}
+
+		uc_match = MC_CMD_FILTER_OP_EXT_IN_MATCH_IFRM_UNKNOWN_UCAST_DST_LBN;
+		mc_match = MC_CMD_FILTER_OP_EXT_IN_MATCH_IFRM_UNKNOWN_MCAST_DST_LBN;
+	} else {
+		uc_match = MC_CMD_FILTER_OP_EXT_IN_MATCH_UNKNOWN_UCAST_DST_LBN;
+		mc_match = MC_CMD_FILTER_OP_EXT_IN_MATCH_UNKNOWN_MCAST_DST_LBN;
+	}
+
+	if (spec->match_flags & EFX_FILTER_MATCH_LOC_MAC_IG)
+		match_fields |=
+			is_multicast_ether_addr(spec->loc_mac) ?
+			1 << mc_match :
+			1 << uc_match;
+	COPY_FIELD(REM_HOST, rem_host, SRC_IP);
+	COPY_FIELD(LOC_HOST, loc_host, DST_IP);
+	COPY_FIELD(REM_MAC, rem_mac, SRC_MAC);
+	COPY_FIELD(REM_PORT, rem_port, SRC_PORT);
+	COPY_FIELD(LOC_MAC, loc_mac, DST_MAC);
+	COPY_FIELD(LOC_PORT, loc_port, DST_PORT);
+	COPY_FIELD(ETHER_TYPE, ether_type, ETHER_TYPE);
+	COPY_FIELD(INNER_VID, inner_vid, INNER_VLAN);
+	COPY_FIELD(OUTER_VID, outer_vid, OUTER_VLAN);
+	COPY_FIELD(IP_PROTO, ip_proto, IP_PROTO);
+#undef COPY_FIELD
+#undef COPY_VALUE
+	MCDI_SET_DWORD(inbuf, FILTER_OP_IN_MATCH_FIELDS,
+		       match_fields);
+}
+
+static void efx_mcdi_filter_push_prep(struct efx_nic *efx,
+				      const struct efx_filter_spec *spec,
+				      efx_dword_t *inbuf, u64 handle,
+				      struct efx_rss_context *ctx,
+				      bool replacing)
+{
+	struct efx_ef10_nic_data *nic_data = efx->nic_data;
+	u32 flags = spec->flags;
+
+	memset(inbuf, 0, MC_CMD_FILTER_OP_EXT_IN_LEN);
+
+	/* If RSS filter, caller better have given us an RSS context */
+	if (flags & EFX_FILTER_FLAG_RX_RSS) {
+		/*
+		 * We don't have the ability to return an error, so we'll just
+		 * log a warning and disable RSS for the filter.
+		 */
+		if (WARN_ON_ONCE(!ctx))
+			flags &= ~EFX_FILTER_FLAG_RX_RSS;
+		else if (WARN_ON_ONCE(ctx->context_id == EFX_MCDI_RSS_CONTEXT_INVALID))
+			flags &= ~EFX_FILTER_FLAG_RX_RSS;
+	}
+
+	if (replacing) {
+		MCDI_SET_DWORD(inbuf, FILTER_OP_IN_OP,
+			       MC_CMD_FILTER_OP_IN_OP_REPLACE);
+		MCDI_SET_QWORD(inbuf, FILTER_OP_IN_HANDLE, handle);
+	} else {
+		efx_mcdi_filter_push_prep_set_match_fields(efx, spec, inbuf);
+	}
+
+	MCDI_SET_DWORD(inbuf, FILTER_OP_IN_PORT_ID, nic_data->vport_id);
+	MCDI_SET_DWORD(inbuf, FILTER_OP_IN_RX_DEST,
+		       spec->dmaq_id == EFX_FILTER_RX_DMAQ_ID_DROP ?
+		       MC_CMD_FILTER_OP_IN_RX_DEST_DROP :
+		       MC_CMD_FILTER_OP_IN_RX_DEST_HOST);
+	MCDI_SET_DWORD(inbuf, FILTER_OP_IN_TX_DOMAIN, 0);
+	MCDI_SET_DWORD(inbuf, FILTER_OP_IN_TX_DEST,
+		       MC_CMD_FILTER_OP_IN_TX_DEST_DEFAULT);
+	MCDI_SET_DWORD(inbuf, FILTER_OP_IN_RX_QUEUE,
+		       spec->dmaq_id == EFX_FILTER_RX_DMAQ_ID_DROP ?
+		       0 : spec->dmaq_id);
+	MCDI_SET_DWORD(inbuf, FILTER_OP_IN_RX_MODE,
+		       (flags & EFX_FILTER_FLAG_RX_RSS) ?
+		       MC_CMD_FILTER_OP_IN_RX_MODE_RSS :
+		       MC_CMD_FILTER_OP_IN_RX_MODE_SIMPLE);
+	if (flags & EFX_FILTER_FLAG_RX_RSS)
+		MCDI_SET_DWORD(inbuf, FILTER_OP_IN_RX_CONTEXT, ctx->context_id);
+}
+
+static int efx_mcdi_filter_push(struct efx_nic *efx,
+				const struct efx_filter_spec *spec, u64 *handle,
+				struct efx_rss_context *ctx, bool replacing)
+{
+	MCDI_DECLARE_BUF(inbuf, MC_CMD_FILTER_OP_EXT_IN_LEN);
+	MCDI_DECLARE_BUF(outbuf, MC_CMD_FILTER_OP_EXT_OUT_LEN);
+	size_t outlen;
+	int rc;
+
+	efx_mcdi_filter_push_prep(efx, spec, inbuf, *handle, ctx, replacing);
+	rc = efx_mcdi_rpc_quiet(efx, MC_CMD_FILTER_OP, inbuf, sizeof(inbuf),
+				outbuf, sizeof(outbuf), &outlen);
+	if (rc && spec->priority != EFX_FILTER_PRI_HINT)
+		efx_mcdi_display_error(efx, MC_CMD_FILTER_OP, sizeof(inbuf),
+				       outbuf, outlen, rc);
+	if (rc == 0)
+		*handle = MCDI_QWORD(outbuf, FILTER_OP_OUT_HANDLE);
+	if (rc == -ENOSPC)
+		rc = -EBUSY; /* to match efx_farch_filter_insert() */
+	return rc;
+}
+
+static u32 efx_mcdi_filter_mcdi_flags_from_spec(const struct efx_filter_spec *spec)
+{
+	enum efx_encap_type encap_type = efx_filter_get_encap_type(spec);
+	unsigned int match_flags = spec->match_flags;
+	unsigned int uc_match, mc_match;
+	u32 mcdi_flags = 0;
+
+#define MAP_FILTER_TO_MCDI_FLAG(gen_flag, mcdi_field, encap) {		\
+		unsigned int  old_match_flags = match_flags;		\
+		match_flags &= ~EFX_FILTER_MATCH_ ## gen_flag;		\
+		if (match_flags != old_match_flags)			\
+			mcdi_flags |=					\
+				(1 << ((encap) ?			\
+				       MC_CMD_FILTER_OP_EXT_IN_MATCH_IFRM_ ## \
+				       mcdi_field ## _LBN :		\
+				       MC_CMD_FILTER_OP_EXT_IN_MATCH_ ##\
+				       mcdi_field ## _LBN));		\
+	}
+	/* inner or outer based on encap type */
+	MAP_FILTER_TO_MCDI_FLAG(REM_HOST, SRC_IP, encap_type);
+	MAP_FILTER_TO_MCDI_FLAG(LOC_HOST, DST_IP, encap_type);
+	MAP_FILTER_TO_MCDI_FLAG(REM_MAC, SRC_MAC, encap_type);
+	MAP_FILTER_TO_MCDI_FLAG(REM_PORT, SRC_PORT, encap_type);
+	MAP_FILTER_TO_MCDI_FLAG(LOC_MAC, DST_MAC, encap_type);
+	MAP_FILTER_TO_MCDI_FLAG(LOC_PORT, DST_PORT, encap_type);
+	MAP_FILTER_TO_MCDI_FLAG(ETHER_TYPE, ETHER_TYPE, encap_type);
+	MAP_FILTER_TO_MCDI_FLAG(IP_PROTO, IP_PROTO, encap_type);
+	/* always outer */
+	MAP_FILTER_TO_MCDI_FLAG(INNER_VID, INNER_VLAN, false);
+	MAP_FILTER_TO_MCDI_FLAG(OUTER_VID, OUTER_VLAN, false);
+#undef MAP_FILTER_TO_MCDI_FLAG
+
+	/* special handling for encap type, and mismatch */
+	if (encap_type) {
+		match_flags &= ~EFX_FILTER_MATCH_ENCAP_TYPE;
+		mcdi_flags |=
+			(1 << MC_CMD_FILTER_OP_EXT_IN_MATCH_ETHER_TYPE_LBN);
+		mcdi_flags |= (1 << MC_CMD_FILTER_OP_EXT_IN_MATCH_IP_PROTO_LBN);
+
+		uc_match = MC_CMD_FILTER_OP_EXT_IN_MATCH_IFRM_UNKNOWN_UCAST_DST_LBN;
+		mc_match = MC_CMD_FILTER_OP_EXT_IN_MATCH_IFRM_UNKNOWN_MCAST_DST_LBN;
+	} else {
+		uc_match = MC_CMD_FILTER_OP_EXT_IN_MATCH_UNKNOWN_UCAST_DST_LBN;
+		mc_match = MC_CMD_FILTER_OP_EXT_IN_MATCH_UNKNOWN_MCAST_DST_LBN;
+	}
+
+	if (match_flags & EFX_FILTER_MATCH_LOC_MAC_IG) {
+		match_flags &= ~EFX_FILTER_MATCH_LOC_MAC_IG;
+		mcdi_flags |=
+			is_multicast_ether_addr(spec->loc_mac) ?
+			1 << mc_match :
+			1 << uc_match;
+	}
+
+	/* Did we map them all? */
+	WARN_ON_ONCE(match_flags);
+
+	return mcdi_flags;
+}
+
+static int efx_mcdi_filter_pri(struct efx_mcdi_filter_table *table,
+			       const struct efx_filter_spec *spec)
+{
+	u32 mcdi_flags = efx_mcdi_filter_mcdi_flags_from_spec(spec);
+	unsigned int match_pri;
+
+	for (match_pri = 0;
+	     match_pri < table->rx_match_count;
+	     match_pri++)
+		if (table->rx_match_mcdi_flags[match_pri] == mcdi_flags)
+			return match_pri;
+
+	return -EPROTONOSUPPORT;
+}
+
+static s32 efx_mcdi_filter_insert_locked(struct efx_nic *efx,
+					 struct efx_filter_spec *spec,
+					 bool replace_equal)
+{
+	DECLARE_BITMAP(mc_rem_map, EFX_EF10_FILTER_SEARCH_LIMIT);
+	struct efx_ef10_nic_data *nic_data = efx->nic_data;
+	struct efx_mcdi_filter_table *table;
+	struct efx_filter_spec *saved_spec;
+	struct efx_rss_context *ctx = NULL;
+	unsigned int match_pri, hash;
+	unsigned int priv_flags;
+	bool rss_locked = false;
+	bool replacing = false;
+	unsigned int depth, i;
+	int ins_index = -1;
+	DEFINE_WAIT(wait);
+	bool is_mc_recip;
+	s32 rc;
+
+	WARN_ON(!rwsem_is_locked(&efx->filter_sem));
+	table = efx->filter_state;
+	down_write(&table->lock);
+
+	/* For now, only support RX filters */
+	if ((spec->flags & (EFX_FILTER_FLAG_RX | EFX_FILTER_FLAG_TX)) !=
+	    EFX_FILTER_FLAG_RX) {
+		rc = -EINVAL;
+		goto out_unlock;
+	}
+
+	rc = efx_mcdi_filter_pri(table, spec);
+	if (rc < 0)
+		goto out_unlock;
+	match_pri = rc;
+
+	hash = efx_filter_spec_hash(spec);
+	is_mc_recip = efx_filter_is_mc_recipient(spec);
+	if (is_mc_recip)
+		bitmap_zero(mc_rem_map, EFX_EF10_FILTER_SEARCH_LIMIT);
+
+	if (spec->flags & EFX_FILTER_FLAG_RX_RSS) {
+		mutex_lock(&efx->rss_lock);
+		rss_locked = true;
+		if (spec->rss_context)
+			ctx = efx_find_rss_context_entry(efx, spec->rss_context);
+		else
+			ctx = &efx->rss_context;
+		if (!ctx) {
+			rc = -ENOENT;
+			goto out_unlock;
+		}
+		if (ctx->context_id == EFX_MCDI_RSS_CONTEXT_INVALID) {
+			rc = -EOPNOTSUPP;
+			goto out_unlock;
+		}
+	}
+
+	/* Find any existing filters with the same match tuple or
+	 * else a free slot to insert at.
+	 */
+	for (depth = 1; depth < EFX_EF10_FILTER_SEARCH_LIMIT; depth++) {
+		i = (hash + depth) & (EFX_MCDI_FILTER_TBL_ROWS - 1);
+		saved_spec = efx_mcdi_filter_entry_spec(table, i);
+
+		if (!saved_spec) {
+			if (ins_index < 0)
+				ins_index = i;
+		} else if (efx_filter_spec_equal(spec, saved_spec)) {
+			if (spec->priority < saved_spec->priority &&
+			    spec->priority != EFX_FILTER_PRI_AUTO) {
+				rc = -EPERM;
+				goto out_unlock;
+			}
+			if (!is_mc_recip) {
+				/* This is the only one */
+				if (spec->priority ==
+				    saved_spec->priority &&
+				    !replace_equal) {
+					rc = -EEXIST;
+					goto out_unlock;
+				}
+				ins_index = i;
+				break;
+			} else if (spec->priority >
+				   saved_spec->priority ||
+				   (spec->priority ==
+				    saved_spec->priority &&
+				    replace_equal)) {
+				if (ins_index < 0)
+					ins_index = i;
+				else
+					__set_bit(depth, mc_rem_map);
+			}
+		}
+	}
+
+	/* Once we reach the maximum search depth, use the first suitable
+	 * slot, or return -EBUSY if there was none
+	 */
+	if (ins_index < 0) {
+		rc = -EBUSY;
+		goto out_unlock;
+	}
+
+	/* Create a software table entry if necessary. */
+	saved_spec = efx_mcdi_filter_entry_spec(table, ins_index);
+	if (saved_spec) {
+		if (spec->priority == EFX_FILTER_PRI_AUTO &&
+		    saved_spec->priority >= EFX_FILTER_PRI_AUTO) {
+			/* Just make sure it won't be removed */
+			if (saved_spec->priority > EFX_FILTER_PRI_AUTO)
+				saved_spec->flags |= EFX_FILTER_FLAG_RX_OVER_AUTO;
+			table->entry[ins_index].spec &=
+				~EFX_EF10_FILTER_FLAG_AUTO_OLD;
+			rc = ins_index;
+			goto out_unlock;
+		}
+		replacing = true;
+		priv_flags = efx_mcdi_filter_entry_flags(table, ins_index);
+	} else {
+		saved_spec = kmalloc(sizeof(*spec), GFP_ATOMIC);
+		if (!saved_spec) {
+			rc = -ENOMEM;
+			goto out_unlock;
+		}
+		*saved_spec = *spec;
+		priv_flags = 0;
+	}
+	efx_mcdi_filter_set_entry(table, ins_index, saved_spec, priv_flags);
+
+	/* Actually insert the filter on the HW */
+	rc = efx_mcdi_filter_push(efx, spec, &table->entry[ins_index].handle,
+				  ctx, replacing);
+
+	if (rc == -EINVAL && nic_data->must_realloc_vis)
+		/* The MC rebooted under us, causing it to reject our filter
+		 * insertion as pointing to an invalid VI (spec->dmaq_id).
+		 */
+		rc = -EAGAIN;
+
+	/* Finalise the software table entry */
+	if (rc == 0) {
+		if (replacing) {
+			/* Update the fields that may differ */
+			if (saved_spec->priority == EFX_FILTER_PRI_AUTO)
+				saved_spec->flags |=
+					EFX_FILTER_FLAG_RX_OVER_AUTO;
+			saved_spec->priority = spec->priority;
+			saved_spec->flags &= EFX_FILTER_FLAG_RX_OVER_AUTO;
+			saved_spec->flags |= spec->flags;
+			saved_spec->rss_context = spec->rss_context;
+			saved_spec->dmaq_id = spec->dmaq_id;
+		}
+	} else if (!replacing) {
+		kfree(saved_spec);
+		saved_spec = NULL;
+	} else {
+		/* We failed to replace, so the old filter is still present.
+		 * Roll back the software table to reflect this.  In fact the
+		 * efx_mcdi_filter_set_entry() call below will do the right
+		 * thing, so nothing extra is needed here.
+		 */
+	}
+	efx_mcdi_filter_set_entry(table, ins_index, saved_spec, priv_flags);
+
+	/* Remove and finalise entries for lower-priority multicast
+	 * recipients
+	 */
+	if (is_mc_recip) {
+		MCDI_DECLARE_BUF(inbuf, MC_CMD_FILTER_OP_EXT_IN_LEN);
+		unsigned int depth, i;
+
+		memset(inbuf, 0, sizeof(inbuf));
+
+		for (depth = 0; depth < EFX_EF10_FILTER_SEARCH_LIMIT; depth++) {
+			if (!test_bit(depth, mc_rem_map))
+				continue;
+
+			i = (hash + depth) & (EFX_MCDI_FILTER_TBL_ROWS - 1);
+			saved_spec = efx_mcdi_filter_entry_spec(table, i);
+			priv_flags = efx_mcdi_filter_entry_flags(table, i);
+
+			if (rc == 0) {
+				MCDI_SET_DWORD(inbuf, FILTER_OP_IN_OP,
+					       MC_CMD_FILTER_OP_IN_OP_UNSUBSCRIBE);
+				MCDI_SET_QWORD(inbuf, FILTER_OP_IN_HANDLE,
+					       table->entry[i].handle);
+				rc = efx_mcdi_rpc(efx, MC_CMD_FILTER_OP,
+						  inbuf, sizeof(inbuf),
+						  NULL, 0, NULL);
+			}
+
+			if (rc == 0) {
+				kfree(saved_spec);
+				saved_spec = NULL;
+				priv_flags = 0;
+			}
+			efx_mcdi_filter_set_entry(table, i, saved_spec,
+						  priv_flags);
+		}
+	}
+
+	/* If successful, return the inserted filter ID */
+	if (rc == 0)
+		rc = efx_mcdi_filter_make_filter_id(match_pri, ins_index);
+
+out_unlock:
+	if (rss_locked)
+		mutex_unlock(&efx->rss_lock);
+	up_write(&table->lock);
+	return rc;
+}
+
+s32 efx_mcdi_filter_insert(struct efx_nic *efx, struct efx_filter_spec *spec,
+			   bool replace_equal)
+{
+	s32 ret;
+
+	down_read(&efx->filter_sem);
+	ret = efx_mcdi_filter_insert_locked(efx, spec, replace_equal);
+	up_read(&efx->filter_sem);
+
+	return ret;
+}
+
+/*
+ * Remove a filter.
+ * If !by_index, remove by ID
+ * If by_index, remove by index
+ * Filter ID may come from userland and must be range-checked.
+ * Caller must hold efx->filter_sem for read, and efx->filter_state->lock
+ * for write.
+ */
+static int efx_mcdi_filter_remove_internal(struct efx_nic *efx,
+					   unsigned int priority_mask,
+					   u32 filter_id, bool by_index)
+{
+	unsigned int filter_idx = efx_mcdi_filter_get_unsafe_id(filter_id);
+	struct efx_mcdi_filter_table *table = efx->filter_state;
+	MCDI_DECLARE_BUF(inbuf,
+			 MC_CMD_FILTER_OP_IN_HANDLE_OFST +
+			 MC_CMD_FILTER_OP_IN_HANDLE_LEN);
+	struct efx_filter_spec *spec;
+	DEFINE_WAIT(wait);
+	int rc;
+
+	spec = efx_mcdi_filter_entry_spec(table, filter_idx);
+	if (!spec ||
+	    (!by_index &&
+	     efx_mcdi_filter_pri(table, spec) !=
+	     efx_mcdi_filter_get_unsafe_pri(filter_id)))
+		return -ENOENT;
+
+	if (spec->flags & EFX_FILTER_FLAG_RX_OVER_AUTO &&
+	    priority_mask == (1U << EFX_FILTER_PRI_AUTO)) {
+		/* Just remove flags */
+		spec->flags &= ~EFX_FILTER_FLAG_RX_OVER_AUTO;
+		table->entry[filter_idx].spec &= ~EFX_EF10_FILTER_FLAG_AUTO_OLD;
+		return 0;
+	}
+
+	if (!(priority_mask & (1U << spec->priority)))
+		return -ENOENT;
+
+	if (spec->flags & EFX_FILTER_FLAG_RX_OVER_AUTO) {
+		/* Reset to an automatic filter */
+
+		struct efx_filter_spec new_spec = *spec;
+
+		new_spec.priority = EFX_FILTER_PRI_AUTO;
+		new_spec.flags = (EFX_FILTER_FLAG_RX |
+				  (efx_rss_active(&efx->rss_context) ?
+				   EFX_FILTER_FLAG_RX_RSS : 0));
+		new_spec.dmaq_id = 0;
+		new_spec.rss_context = 0;
+		rc = efx_mcdi_filter_push(efx, &new_spec,
+					  &table->entry[filter_idx].handle,
+					  &efx->rss_context,
+					  true);
+
+		if (rc == 0)
+			*spec = new_spec;
+	} else {
+		/* Really remove the filter */
+
+		MCDI_SET_DWORD(inbuf, FILTER_OP_IN_OP,
+			       efx_mcdi_filter_is_exclusive(spec) ?
+			       MC_CMD_FILTER_OP_IN_OP_REMOVE :
+			       MC_CMD_FILTER_OP_IN_OP_UNSUBSCRIBE);
+		MCDI_SET_QWORD(inbuf, FILTER_OP_IN_HANDLE,
+			       table->entry[filter_idx].handle);
+		rc = efx_mcdi_rpc_quiet(efx, MC_CMD_FILTER_OP,
+					inbuf, sizeof(inbuf), NULL, 0, NULL);
+
+		if ((rc == 0) || (rc == -ENOENT)) {
+			/* Filter removed OK or didn't actually exist */
+			kfree(spec);
+			efx_mcdi_filter_set_entry(table, filter_idx, NULL, 0);
+		} else {
+			efx_mcdi_display_error(efx, MC_CMD_FILTER_OP,
+					       MC_CMD_FILTER_OP_EXT_IN_LEN,
+					       NULL, 0, rc);
+		}
+	}
+
+	return rc;
+}
+
+/* Remove filters that weren't renewed. */
+static void efx_mcdi_filter_remove_old(struct efx_nic *efx)
+{
+	struct efx_mcdi_filter_table *table = efx->filter_state;
+	int remove_failed = 0;
+	int remove_noent = 0;
+	int rc;
+	int i;
+
+	down_write(&table->lock);
+	for (i = 0; i < EFX_MCDI_FILTER_TBL_ROWS; i++) {
+		if (READ_ONCE(table->entry[i].spec) &
+		    EFX_EF10_FILTER_FLAG_AUTO_OLD) {
+			rc = efx_mcdi_filter_remove_internal(efx,
+					1U << EFX_FILTER_PRI_AUTO, i, true);
+			if (rc == -ENOENT)
+				remove_noent++;
+			else if (rc)
+				remove_failed++;
+		}
+	}
+	up_write(&table->lock);
+
+	if (remove_failed)
+		netif_info(efx, drv, efx->net_dev,
+			   "%s: failed to remove %d filters\n",
+			   __func__, remove_failed);
+	if (remove_noent)
+		netif_info(efx, drv, efx->net_dev,
+			   "%s: failed to remove %d non-existent filters\n",
+			   __func__, remove_noent);
+}
+
+int efx_mcdi_filter_remove_safe(struct efx_nic *efx,
+				enum efx_filter_priority priority,
+				u32 filter_id)
+{
+	struct efx_mcdi_filter_table *table;
+	int rc;
+
+	down_read(&efx->filter_sem);
+	table = efx->filter_state;
+	down_write(&table->lock);
+	rc = efx_mcdi_filter_remove_internal(efx, 1U << priority, filter_id,
+					     false);
+	up_write(&table->lock);
+	up_read(&efx->filter_sem);
+	return rc;
+}
+
+/* Caller must hold efx->filter_sem for read */
+static void efx_mcdi_filter_remove_unsafe(struct efx_nic *efx,
+					  enum efx_filter_priority priority,
+					  u32 filter_id)
+{
+	struct efx_mcdi_filter_table *table = efx->filter_state;
+
+	if (filter_id == EFX_EF10_FILTER_ID_INVALID)
+		return;
+
+	down_write(&table->lock);
+	efx_mcdi_filter_remove_internal(efx, 1U << priority, filter_id,
+					true);
+	up_write(&table->lock);
+}
+
+int efx_mcdi_filter_get_safe(struct efx_nic *efx,
+			     enum efx_filter_priority priority,
+			     u32 filter_id, struct efx_filter_spec *spec)
+{
+	unsigned int filter_idx = efx_mcdi_filter_get_unsafe_id(filter_id);
+	const struct efx_filter_spec *saved_spec;
+	struct efx_mcdi_filter_table *table;
+	int rc;
+
+	down_read(&efx->filter_sem);
+	table = efx->filter_state;
+	down_read(&table->lock);
+	saved_spec = efx_mcdi_filter_entry_spec(table, filter_idx);
+	if (saved_spec && saved_spec->priority == priority &&
+	    efx_mcdi_filter_pri(table, saved_spec) ==
+	    efx_mcdi_filter_get_unsafe_pri(filter_id)) {
+		*spec = *saved_spec;
+		rc = 0;
+	} else {
+		rc = -ENOENT;
+	}
+	up_read(&table->lock);
+	up_read(&efx->filter_sem);
+	return rc;
+}
+
+static int efx_mcdi_filter_insert_addr_list(struct efx_nic *efx,
+					    struct efx_mcdi_filter_vlan *vlan,
+					    bool multicast, bool rollback)
+{
+	struct efx_mcdi_filter_table *table = efx->filter_state;
+	struct efx_mcdi_dev_addr *addr_list;
+	enum efx_filter_flags filter_flags;
+	struct efx_filter_spec spec;
+	u8 baddr[ETH_ALEN];
+	unsigned int i, j;
+	int addr_count;
+	u16 *ids;
+	int rc;
+
+	if (multicast) {
+		addr_list = table->dev_mc_list;
+		addr_count = table->dev_mc_count;
+		ids = vlan->mc;
+	} else {
+		addr_list = table->dev_uc_list;
+		addr_count = table->dev_uc_count;
+		ids = vlan->uc;
+	}
+
+	filter_flags = efx_rss_active(&efx->rss_context) ? EFX_FILTER_FLAG_RX_RSS : 0;
+
+	/* Insert/renew filters */
+	for (i = 0; i < addr_count; i++) {
+		EFX_WARN_ON_PARANOID(ids[i] != EFX_EF10_FILTER_ID_INVALID);
+		efx_filter_init_rx(&spec, EFX_FILTER_PRI_AUTO, filter_flags, 0);
+		efx_filter_set_eth_local(&spec, vlan->vid, addr_list[i].addr);
+		rc = efx_mcdi_filter_insert_locked(efx, &spec, true);
+		if (rc < 0) {
+			if (rollback) {
+				netif_info(efx, drv, efx->net_dev,
+					   "efx_mcdi_filter_insert failed rc=%d\n",
+					   rc);
+				/* Fall back to promiscuous */
+				for (j = 0; j < i; j++) {
+					efx_mcdi_filter_remove_unsafe(
+						efx, EFX_FILTER_PRI_AUTO,
+						ids[j]);
+					ids[j] = EFX_EF10_FILTER_ID_INVALID;
+				}
+				return rc;
+			} else {
+				/* keep invalid ID, and carry on */
+			}
+		} else {
+			ids[i] = efx_mcdi_filter_get_unsafe_id(rc);
+		}
+	}
+
+	if (multicast && rollback) {
+		/* Also need an Ethernet broadcast filter */
+		EFX_WARN_ON_PARANOID(vlan->default_filters[EFX_EF10_BCAST] !=
+				     EFX_EF10_FILTER_ID_INVALID);
+		efx_filter_init_rx(&spec, EFX_FILTER_PRI_AUTO, filter_flags, 0);
+		eth_broadcast_addr(baddr);
+		efx_filter_set_eth_local(&spec, vlan->vid, baddr);
+		rc = efx_mcdi_filter_insert_locked(efx, &spec, true);
+		if (rc < 0) {
+			netif_warn(efx, drv, efx->net_dev,
+				   "Broadcast filter insert failed rc=%d\n", rc);
+			/* Fall back to promiscuous */
+			for (j = 0; j < i; j++) {
+				efx_mcdi_filter_remove_unsafe(
+					efx, EFX_FILTER_PRI_AUTO,
+					ids[j]);
+				ids[j] = EFX_EF10_FILTER_ID_INVALID;
+			}
+			return rc;
+		} else {
+			vlan->default_filters[EFX_EF10_BCAST] =
+				efx_mcdi_filter_get_unsafe_id(rc);
+		}
+	}
+
+	return 0;
+}
+
+static int efx_mcdi_filter_insert_def(struct efx_nic *efx,
+				      struct efx_mcdi_filter_vlan *vlan,
+				      enum efx_encap_type encap_type,
+				      bool multicast, bool rollback)
+{
+	struct efx_ef10_nic_data *nic_data = efx->nic_data;
+	enum efx_filter_flags filter_flags;
+	struct efx_filter_spec spec;
+	u8 baddr[ETH_ALEN];
+	int rc;
+	u16 *id;
+
+	filter_flags = efx_rss_active(&efx->rss_context) ? EFX_FILTER_FLAG_RX_RSS : 0;
+
+	efx_filter_init_rx(&spec, EFX_FILTER_PRI_AUTO, filter_flags, 0);
+
+	if (multicast)
+		efx_filter_set_mc_def(&spec);
+	else
+		efx_filter_set_uc_def(&spec);
+
+	if (encap_type) {
+		if (nic_data->datapath_caps &
+		    (1 << MC_CMD_GET_CAPABILITIES_OUT_VXLAN_NVGRE_LBN))
+			efx_filter_set_encap_type(&spec, encap_type);
+		else
+			/*
+			 * don't insert encap filters on non-supporting
+			 * platforms. ID will be left as INVALID.
+			 */
+			return 0;
+	}
+
+	if (vlan->vid != EFX_FILTER_VID_UNSPEC)
+		efx_filter_set_eth_local(&spec, vlan->vid, NULL);
+
+	rc = efx_mcdi_filter_insert_locked(efx, &spec, true);
+	if (rc < 0) {
+		const char *um = multicast ? "Multicast" : "Unicast";
+		const char *encap_name = "";
+		const char *encap_ipv = "";
+
+		if ((encap_type & EFX_ENCAP_TYPES_MASK) ==
+		    EFX_ENCAP_TYPE_VXLAN)
+			encap_name = "VXLAN ";
+		else if ((encap_type & EFX_ENCAP_TYPES_MASK) ==
+			 EFX_ENCAP_TYPE_NVGRE)
+			encap_name = "NVGRE ";
+		else if ((encap_type & EFX_ENCAP_TYPES_MASK) ==
+			 EFX_ENCAP_TYPE_GENEVE)
+			encap_name = "GENEVE ";
+		if (encap_type & EFX_ENCAP_FLAG_IPV6)
+			encap_ipv = "IPv6 ";
+		else if (encap_type)
+			encap_ipv = "IPv4 ";
+
+		/*
+		 * unprivileged functions can't insert mismatch filters
+		 * for encapsulated or unicast traffic, so downgrade
+		 * those warnings to debug.
+		 */
+		netif_cond_dbg(efx, drv, efx->net_dev,
+			       rc == -EPERM && (encap_type || !multicast), warn,
+			       "%s%s%s mismatch filter insert failed rc=%d\n",
+			       encap_name, encap_ipv, um, rc);
+	} else if (multicast) {
+		/* mapping from encap types to default filter IDs (multicast) */
+		static enum efx_mcdi_filter_default_filters map[] = {
+			[EFX_ENCAP_TYPE_NONE] = EFX_EF10_MCDEF,
+			[EFX_ENCAP_TYPE_VXLAN] = EFX_EF10_VXLAN4_MCDEF,
+			[EFX_ENCAP_TYPE_NVGRE] = EFX_EF10_NVGRE4_MCDEF,
+			[EFX_ENCAP_TYPE_GENEVE] = EFX_EF10_GENEVE4_MCDEF,
+			[EFX_ENCAP_TYPE_VXLAN | EFX_ENCAP_FLAG_IPV6] =
+				EFX_EF10_VXLAN6_MCDEF,
+			[EFX_ENCAP_TYPE_NVGRE | EFX_ENCAP_FLAG_IPV6] =
+				EFX_EF10_NVGRE6_MCDEF,
+			[EFX_ENCAP_TYPE_GENEVE | EFX_ENCAP_FLAG_IPV6] =
+				EFX_EF10_GENEVE6_MCDEF,
+		};
+
+		/* quick bounds check (BCAST result impossible) */
+		BUILD_BUG_ON(EFX_EF10_BCAST != 0);
+		if (encap_type >= ARRAY_SIZE(map) || map[encap_type] == 0) {
+			WARN_ON(1);
+			return -EINVAL;
+		}
+		/* then follow map */
+		id = &vlan->default_filters[map[encap_type]];
+
+		EFX_WARN_ON_PARANOID(*id != EFX_EF10_FILTER_ID_INVALID);
+		*id = efx_mcdi_filter_get_unsafe_id(rc);
+		if (!nic_data->workaround_26807 && !encap_type) {
+			/* Also need an Ethernet broadcast filter */
+			efx_filter_init_rx(&spec, EFX_FILTER_PRI_AUTO,
+					   filter_flags, 0);
+			eth_broadcast_addr(baddr);
+			efx_filter_set_eth_local(&spec, vlan->vid, baddr);
+			rc = efx_mcdi_filter_insert_locked(efx, &spec, true);
+			if (rc < 0) {
+				netif_warn(efx, drv, efx->net_dev,
+					   "Broadcast filter insert failed rc=%d\n",
+					   rc);
+				if (rollback) {
+					/* Roll back the mc_def filter */
+					efx_mcdi_filter_remove_unsafe(
+							efx, EFX_FILTER_PRI_AUTO,
+							*id);
+					*id = EFX_EF10_FILTER_ID_INVALID;
+					return rc;
+				}
+			} else {
+				EFX_WARN_ON_PARANOID(
+					vlan->default_filters[EFX_EF10_BCAST] !=
+					EFX_EF10_FILTER_ID_INVALID);
+				vlan->default_filters[EFX_EF10_BCAST] =
+					efx_mcdi_filter_get_unsafe_id(rc);
+			}
+		}
+		rc = 0;
+	} else {
+		/* mapping from encap types to default filter IDs (unicast) */
+		static enum efx_mcdi_filter_default_filters map[] = {
+			[EFX_ENCAP_TYPE_NONE] = EFX_EF10_UCDEF,
+			[EFX_ENCAP_TYPE_VXLAN] = EFX_EF10_VXLAN4_UCDEF,
+			[EFX_ENCAP_TYPE_NVGRE] = EFX_EF10_NVGRE4_UCDEF,
+			[EFX_ENCAP_TYPE_GENEVE] = EFX_EF10_GENEVE4_UCDEF,
+			[EFX_ENCAP_TYPE_VXLAN | EFX_ENCAP_FLAG_IPV6] =
+				EFX_EF10_VXLAN6_UCDEF,
+			[EFX_ENCAP_TYPE_NVGRE | EFX_ENCAP_FLAG_IPV6] =
+				EFX_EF10_NVGRE6_UCDEF,
+			[EFX_ENCAP_TYPE_GENEVE | EFX_ENCAP_FLAG_IPV6] =
+				EFX_EF10_GENEVE6_UCDEF,
+		};
+
+		/* quick bounds check (BCAST result impossible) */
+		BUILD_BUG_ON(EFX_EF10_BCAST != 0);
+		if (encap_type >= ARRAY_SIZE(map) || map[encap_type] == 0) {
+			WARN_ON(1);
+			return -EINVAL;
+		}
+		/* then follow map */
+		id = &vlan->default_filters[map[encap_type]];
+		EFX_WARN_ON_PARANOID(*id != EFX_EF10_FILTER_ID_INVALID);
+		*id = rc;
+		rc = 0;
+	}
+	return rc;
+}
+
+/*
+ * Caller must hold efx->filter_sem for read if race against
+ * efx_mcdi_filter_table_remove() is possible
+ */
+static void efx_mcdi_filter_vlan_sync_rx_mode(struct efx_nic *efx,
+					      struct efx_mcdi_filter_vlan *vlan)
+{
+	struct efx_mcdi_filter_table *table = efx->filter_state;
+	struct efx_ef10_nic_data *nic_data = efx->nic_data;
+
+	/*
+	 * Do not install unspecified VID if VLAN filtering is enabled.
+	 * Do not install all specified VIDs if VLAN filtering is disabled.
+	 */
+	if ((vlan->vid == EFX_FILTER_VID_UNSPEC) == table->vlan_filter)
+		return;
+
+	/* Insert/renew unicast filters */
+	if (table->uc_promisc) {
+		efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_NONE,
+					   false, false);
+		efx_mcdi_filter_insert_addr_list(efx, vlan, false, false);
+	} else {
+		/*
+		 * If any of the filters failed to insert, fall back to
+		 * promiscuous mode - add in the uc_def filter.  But keep
+		 * our individual unicast filters.
+		 */
+		if (efx_mcdi_filter_insert_addr_list(efx, vlan, false, false))
+			efx_mcdi_filter_insert_def(efx, vlan,
+						   EFX_ENCAP_TYPE_NONE,
+						   false, false);
+	}
+	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_VXLAN,
+				   false, false);
+	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_VXLAN |
+					      EFX_ENCAP_FLAG_IPV6,
+				   false, false);
+	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_NVGRE,
+				   false, false);
+	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_NVGRE |
+					      EFX_ENCAP_FLAG_IPV6,
+				   false, false);
+	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_GENEVE,
+				   false, false);
+	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_GENEVE |
+					      EFX_ENCAP_FLAG_IPV6,
+				   false, false);
+
+	/*
+	 * Insert/renew multicast filters
+	 *
+	 * If changing promiscuous state with cascaded multicast filters, remove
+	 * old filters first, so that packets are dropped rather than duplicated
+	 */
+	if (nic_data->workaround_26807 &&
+	    table->mc_promisc_last != table->mc_promisc)
+		efx_mcdi_filter_remove_old(efx);
+	if (table->mc_promisc) {
+		if (nic_data->workaround_26807) {
+			/*
+			 * If we failed to insert promiscuous filters, rollback
+			 * and fall back to individual multicast filters
+			 */
+			if (efx_mcdi_filter_insert_def(efx, vlan,
+						       EFX_ENCAP_TYPE_NONE,
+						       true, true)) {
+				/* Changing promisc state, so remove old filters */
+				efx_mcdi_filter_remove_old(efx);
+				efx_mcdi_filter_insert_addr_list(efx, vlan,
+								 true, false);
+			}
+		} else {
+			/*
+			 * If we failed to insert promiscuous filters, don't
+			 * rollback.  Regardless, also insert the mc_list,
+			 * unless it's incomplete due to overflow
+			 */
+			efx_mcdi_filter_insert_def(efx, vlan,
+						   EFX_ENCAP_TYPE_NONE,
+						   true, false);
+			if (!table->mc_overflow)
+				efx_mcdi_filter_insert_addr_list(efx, vlan,
+								 true, false);
+		}
+	} else {
+		/*
+		 * If any filters failed to insert, rollback and fall back to
+		 * promiscuous mode - mc_def filter and maybe broadcast.  If
+		 * that fails, roll back again and insert as many of our
+		 * individual multicast filters as we can.
+		 */
+		if (efx_mcdi_filter_insert_addr_list(efx, vlan, true, true)) {
+			/* Changing promisc state, so remove old filters */
+			if (nic_data->workaround_26807)
+				efx_mcdi_filter_remove_old(efx);
+			if (efx_mcdi_filter_insert_def(efx, vlan,
+						       EFX_ENCAP_TYPE_NONE,
+						       true, true))
+				efx_mcdi_filter_insert_addr_list(efx, vlan,
+								 true, false);
+		}
+	}
+	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_VXLAN,
+				   true, false);
+	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_VXLAN |
+					      EFX_ENCAP_FLAG_IPV6,
+				   true, false);
+	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_NVGRE,
+				   true, false);
+	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_NVGRE |
+					      EFX_ENCAP_FLAG_IPV6,
+				   true, false);
+	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_GENEVE,
+				   true, false);
+	efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_GENEVE |
+					      EFX_ENCAP_FLAG_IPV6,
+				   true, false);
+}
+
+int efx_mcdi_filter_clear_rx(struct efx_nic *efx,
+			     enum efx_filter_priority priority)
+{
+	struct efx_mcdi_filter_table *table;
+	unsigned int priority_mask;
+	unsigned int i;
+	int rc;
+
+	priority_mask = (((1U << (priority + 1)) - 1) &
+			 ~(1U << EFX_FILTER_PRI_AUTO));
+
+	down_read(&efx->filter_sem);
+	table = efx->filter_state;
+	down_write(&table->lock);
+	for (i = 0; i < EFX_MCDI_FILTER_TBL_ROWS; i++) {
+		rc = efx_mcdi_filter_remove_internal(efx, priority_mask,
+						     i, true);
+		if (rc && rc != -ENOENT)
+			break;
+		rc = 0;
+	}
+
+	up_write(&table->lock);
+	up_read(&efx->filter_sem);
+	return rc;
+}
+
+u32 efx_mcdi_filter_count_rx_used(struct efx_nic *efx,
+				 enum efx_filter_priority priority)
+{
+	struct efx_mcdi_filter_table *table;
+	unsigned int filter_idx;
+	s32 count = 0;
+
+	down_read(&efx->filter_sem);
+	table = efx->filter_state;
+	down_read(&table->lock);
+	for (filter_idx = 0; filter_idx < EFX_MCDI_FILTER_TBL_ROWS; filter_idx++) {
+		if (table->entry[filter_idx].spec &&
+		    efx_mcdi_filter_entry_spec(table, filter_idx)->priority ==
+		    priority)
+			++count;
+	}
+	up_read(&table->lock);
+	up_read(&efx->filter_sem);
+	return count;
+}
+
+u32 efx_mcdi_filter_get_rx_id_limit(struct efx_nic *efx)
+{
+	struct efx_mcdi_filter_table *table = efx->filter_state;
+
+	return table->rx_match_count * EFX_MCDI_FILTER_TBL_ROWS * 2;
+}
+
+s32 efx_mcdi_filter_get_rx_ids(struct efx_nic *efx,
+			       enum efx_filter_priority priority,
+			       u32 *buf, u32 size)
+{
+	struct efx_mcdi_filter_table *table;
+	struct efx_filter_spec *spec;
+	unsigned int filter_idx;
+	s32 count = 0;
+
+	down_read(&efx->filter_sem);
+	table = efx->filter_state;
+	down_read(&table->lock);
+
+	for (filter_idx = 0; filter_idx < EFX_MCDI_FILTER_TBL_ROWS; filter_idx++) {
+		spec = efx_mcdi_filter_entry_spec(table, filter_idx);
+		if (spec && spec->priority == priority) {
+			if (count == size) {
+				count = -EMSGSIZE;
+				break;
+			}
+			buf[count++] =
+				efx_mcdi_filter_make_filter_id(
+					efx_mcdi_filter_pri(table, spec),
+					filter_idx);
+		}
+	}
+	up_read(&table->lock);
+	up_read(&efx->filter_sem);
+	return count;
+}
+
+static int efx_mcdi_filter_match_flags_from_mcdi(bool encap, u32 mcdi_flags)
+{
+	int match_flags = 0;
+
+#define MAP_FLAG(gen_flag, mcdi_field) do {				\
+		u32 old_mcdi_flags = mcdi_flags;			\
+		mcdi_flags &= ~(1 << MC_CMD_FILTER_OP_EXT_IN_MATCH_ ##	\
+				     mcdi_field ## _LBN);		\
+		if (mcdi_flags != old_mcdi_flags)			\
+			match_flags |= EFX_FILTER_MATCH_ ## gen_flag;	\
+	} while (0)
+
+	if (encap) {
+		/* encap filters must specify encap type */
+		match_flags |= EFX_FILTER_MATCH_ENCAP_TYPE;
+		/* and imply ethertype and ip proto */
+		mcdi_flags &=
+			~(1 << MC_CMD_FILTER_OP_EXT_IN_MATCH_IP_PROTO_LBN);
+		mcdi_flags &=
+			~(1 << MC_CMD_FILTER_OP_EXT_IN_MATCH_ETHER_TYPE_LBN);
+		/* VLAN tags refer to the outer packet */
+		MAP_FLAG(INNER_VID, INNER_VLAN);
+		MAP_FLAG(OUTER_VID, OUTER_VLAN);
+		/* everything else refers to the inner packet */
+		MAP_FLAG(LOC_MAC_IG, IFRM_UNKNOWN_UCAST_DST);
+		MAP_FLAG(LOC_MAC_IG, IFRM_UNKNOWN_MCAST_DST);
+		MAP_FLAG(REM_HOST, IFRM_SRC_IP);
+		MAP_FLAG(LOC_HOST, IFRM_DST_IP);
+		MAP_FLAG(REM_MAC, IFRM_SRC_MAC);
+		MAP_FLAG(REM_PORT, IFRM_SRC_PORT);
+		MAP_FLAG(LOC_MAC, IFRM_DST_MAC);
+		MAP_FLAG(LOC_PORT, IFRM_DST_PORT);
+		MAP_FLAG(ETHER_TYPE, IFRM_ETHER_TYPE);
+		MAP_FLAG(IP_PROTO, IFRM_IP_PROTO);
+	} else {
+		MAP_FLAG(LOC_MAC_IG, UNKNOWN_UCAST_DST);
+		MAP_FLAG(LOC_MAC_IG, UNKNOWN_MCAST_DST);
+		MAP_FLAG(REM_HOST, SRC_IP);
+		MAP_FLAG(LOC_HOST, DST_IP);
+		MAP_FLAG(REM_MAC, SRC_MAC);
+		MAP_FLAG(REM_PORT, SRC_PORT);
+		MAP_FLAG(LOC_MAC, DST_MAC);
+		MAP_FLAG(LOC_PORT, DST_PORT);
+		MAP_FLAG(ETHER_TYPE, ETHER_TYPE);
+		MAP_FLAG(INNER_VID, INNER_VLAN);
+		MAP_FLAG(OUTER_VID, OUTER_VLAN);
+		MAP_FLAG(IP_PROTO, IP_PROTO);
+	}
+#undef MAP_FLAG
+
+	/* Did we map them all? */
+	if (mcdi_flags)
+		return -EINVAL;
+
+	return match_flags;
+}
+
+bool efx_mcdi_filter_match_supported(struct efx_mcdi_filter_table *table,
+				     bool encap,
+				     enum efx_filter_match_flags match_flags)
+{
+	unsigned int match_pri;
+	int mf;
+
+	for (match_pri = 0;
+	     match_pri < table->rx_match_count;
+	     match_pri++) {
+		mf = efx_mcdi_filter_match_flags_from_mcdi(encap,
+				table->rx_match_mcdi_flags[match_pri]);
+		if (mf == match_flags)
+			return true;
+	}
+
+	return false;
+}
+
+static int
+efx_mcdi_filter_table_probe_matches(struct efx_nic *efx,
+				    struct efx_mcdi_filter_table *table,
+				    bool encap)
+{
+	MCDI_DECLARE_BUF(inbuf, MC_CMD_GET_PARSER_DISP_INFO_IN_LEN);
+	MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_PARSER_DISP_INFO_OUT_LENMAX);
+	unsigned int pd_match_pri, pd_match_count;
+	size_t outlen;
+	int rc;
+
+	/* Find out which RX filter types are supported, and their priorities */
+	MCDI_SET_DWORD(inbuf, GET_PARSER_DISP_INFO_IN_OP,
+		       encap ?
+		       MC_CMD_GET_PARSER_DISP_INFO_IN_OP_GET_SUPPORTED_ENCAP_RX_MATCHES :
+		       MC_CMD_GET_PARSER_DISP_INFO_IN_OP_GET_SUPPORTED_RX_MATCHES);
+	rc = efx_mcdi_rpc(efx, MC_CMD_GET_PARSER_DISP_INFO,
+			  inbuf, sizeof(inbuf), outbuf, sizeof(outbuf),
+			  &outlen);
+	if (rc)
+		return rc;
+
+	pd_match_count = MCDI_VAR_ARRAY_LEN(
+		outlen, GET_PARSER_DISP_INFO_OUT_SUPPORTED_MATCHES);
+
+	for (pd_match_pri = 0; pd_match_pri < pd_match_count; pd_match_pri++) {
+		u32 mcdi_flags =
+			MCDI_ARRAY_DWORD(
+				outbuf,
+				GET_PARSER_DISP_INFO_OUT_SUPPORTED_MATCHES,
+				pd_match_pri);
+		rc = efx_mcdi_filter_match_flags_from_mcdi(encap, mcdi_flags);
+		if (rc < 0) {
+			netif_dbg(efx, probe, efx->net_dev,
+				  "%s: fw flags %#x pri %u not supported in driver\n",
+				  __func__, mcdi_flags, pd_match_pri);
+		} else {
+			netif_dbg(efx, probe, efx->net_dev,
+				  "%s: fw flags %#x pri %u supported as driver flags %#x pri %u\n",
+				  __func__, mcdi_flags, pd_match_pri,
+				  rc, table->rx_match_count);
+			table->rx_match_mcdi_flags[table->rx_match_count] = mcdi_flags;
+			table->rx_match_count++;
+		}
+	}
+
+	return 0;
+}
+
+int efx_mcdi_filter_table_probe(struct efx_nic *efx)
+{
+	struct efx_ef10_nic_data *nic_data = efx->nic_data;
+	struct net_device *net_dev = efx->net_dev;
+	struct efx_mcdi_filter_table *table;
+	struct efx_mcdi_filter_vlan *vlan;
+	int rc;
+
+	if (!efx_rwsem_assert_write_locked(&efx->filter_sem))
+		return -EINVAL;
+
+	if (efx->filter_state) /* already probed */
+		return 0;
+
+	table = kzalloc(sizeof(*table), GFP_KERNEL);
+	if (!table)
+		return -ENOMEM;
+
+	table->rx_match_count = 0;
+	rc = efx_mcdi_filter_table_probe_matches(efx, table, false);
+	if (rc)
+		goto fail;
+	if (nic_data->datapath_caps &
+		   (1 << MC_CMD_GET_CAPABILITIES_OUT_VXLAN_NVGRE_LBN))
+		rc = efx_mcdi_filter_table_probe_matches(efx, table, true);
+	if (rc)
+		goto fail;
+	if ((efx_supported_features(efx) & NETIF_F_HW_VLAN_CTAG_FILTER) &&
+	    !(efx_mcdi_filter_match_supported(table, false,
+		(EFX_FILTER_MATCH_OUTER_VID | EFX_FILTER_MATCH_LOC_MAC)) &&
+	      efx_mcdi_filter_match_supported(table, false,
+		(EFX_FILTER_MATCH_OUTER_VID | EFX_FILTER_MATCH_LOC_MAC_IG)))) {
+		netif_info(efx, probe, net_dev,
+			   "VLAN filters are not supported in this firmware variant\n");
+		net_dev->features &= ~NETIF_F_HW_VLAN_CTAG_FILTER;
+		efx->fixed_features &= ~NETIF_F_HW_VLAN_CTAG_FILTER;
+		net_dev->hw_features &= ~NETIF_F_HW_VLAN_CTAG_FILTER;
+	}
+
+	table->entry = vzalloc(array_size(EFX_MCDI_FILTER_TBL_ROWS,
+					  sizeof(*table->entry)));
+	if (!table->entry) {
+		rc = -ENOMEM;
+		goto fail;
+	}
+
+	table->mc_promisc_last = false;
+	table->vlan_filter =
+		!!(efx->net_dev->features & NETIF_F_HW_VLAN_CTAG_FILTER);
+	INIT_LIST_HEAD(&table->vlan_list);
+	init_rwsem(&table->lock);
+
+	efx->filter_state = table;
+
+	list_for_each_entry(vlan, &nic_data->vlan_list, list) {
+		rc = efx_mcdi_filter_add_vlan(efx, vlan->vid);
+		if (rc)
+			goto fail_add_vlan;
+	}
+
+	return 0;
+
+fail_add_vlan:
+	efx_mcdi_filter_cleanup_vlans(efx);
+	efx->filter_state = NULL;
+fail:
+	kfree(table);
+	return rc;
+}
+
+/*
+ * Caller must hold efx->filter_sem for read if race against
+ * efx_mcdi_filter_table_remove() is possible
+ */
+void efx_mcdi_filter_table_restore(struct efx_nic *efx)
+{
+	struct efx_mcdi_filter_table *table = efx->filter_state;
+	struct efx_ef10_nic_data *nic_data = efx->nic_data;
+	unsigned int invalid_filters = 0, failed = 0;
+	struct efx_mcdi_filter_vlan *vlan;
+	struct efx_filter_spec *spec;
+	struct efx_rss_context *ctx;
+	unsigned int filter_idx;
+	u32 mcdi_flags;
+	int match_pri;
+	int rc, i;
+
+	WARN_ON(!rwsem_is_locked(&efx->filter_sem));
+
+	if (!nic_data->must_restore_filters)
+		return;
+
+	if (!table)
+		return;
+
+	down_write(&table->lock);
+	mutex_lock(&efx->rss_lock);
+
+	for (filter_idx = 0; filter_idx < EFX_MCDI_FILTER_TBL_ROWS; filter_idx++) {
+		spec = efx_mcdi_filter_entry_spec(table, filter_idx);
+		if (!spec)
+			continue;
+
+		mcdi_flags = efx_mcdi_filter_mcdi_flags_from_spec(spec);
+		match_pri = 0;
+		while (match_pri < table->rx_match_count &&
+		       table->rx_match_mcdi_flags[match_pri] != mcdi_flags)
+			++match_pri;
+		if (match_pri >= table->rx_match_count) {
+			invalid_filters++;
+			goto not_restored;
+		}
+		if (spec->rss_context)
+			ctx = efx_find_rss_context_entry(efx, spec->rss_context);
+		else
+			ctx = &efx->rss_context;
+		if (spec->flags & EFX_FILTER_FLAG_RX_RSS) {
+			if (!ctx) {
+				netif_warn(efx, drv, efx->net_dev,
+					   "Warning: unable to restore a filter with nonexistent RSS context %u.\n",
+					   spec->rss_context);
+				invalid_filters++;
+				goto not_restored;
+			}
+			if (ctx->context_id == EFX_MCDI_RSS_CONTEXT_INVALID) {
+				netif_warn(efx, drv, efx->net_dev,
+					   "Warning: unable to restore a filter with RSS context %u as it was not created.\n",
+					   spec->rss_context);
+				invalid_filters++;
+				goto not_restored;
+			}
+		}
+
+		rc = efx_mcdi_filter_push(efx, spec,
+					  &table->entry[filter_idx].handle,
+					  ctx, false);
+		if (rc)
+			failed++;
+
+		if (rc) {
+not_restored:
+			list_for_each_entry(vlan, &table->vlan_list, list)
+				for (i = 0; i < EFX_EF10_NUM_DEFAULT_FILTERS; ++i)
+					if (vlan->default_filters[i] == filter_idx)
+						vlan->default_filters[i] =
+							EFX_EF10_FILTER_ID_INVALID;
+
+			kfree(spec);
+			efx_mcdi_filter_set_entry(table, filter_idx, NULL, 0);
+		}
+	}
+
+	mutex_unlock(&efx->rss_lock);
+	up_write(&table->lock);
+
+	/*
+	 * This can happen validly if the MC's capabilities have changed, so
+	 * is not an error.
+	 */
+	if (invalid_filters)
+		netif_dbg(efx, drv, efx->net_dev,
+			  "Did not restore %u filters that are now unsupported.\n",
+			  invalid_filters);
+
+	if (failed)
+		netif_err(efx, hw, efx->net_dev,
+			  "unable to restore %u filters\n", failed);
+	else
+		nic_data->must_restore_filters = false;
+}
+
+void efx_mcdi_filter_table_remove(struct efx_nic *efx)
+{
+	struct efx_mcdi_filter_table *table = efx->filter_state;
+	MCDI_DECLARE_BUF(inbuf, MC_CMD_FILTER_OP_EXT_IN_LEN);
+	struct efx_filter_spec *spec;
+	unsigned int filter_idx;
+	int rc;
+
+	efx_mcdi_filter_cleanup_vlans(efx);
+	efx->filter_state = NULL;
+	/*
+	 * If we were called without locking, then it's not safe to free
+	 * the table as others might be using it.  So we just WARN, leak
+	 * the memory, and potentially get an inconsistent filter table
+	 * state.
+	 * This should never actually happen.
+	 */
+	if (!efx_rwsem_assert_write_locked(&efx->filter_sem))
+		return;
+
+	if (!table)
+		return;
+
+	for (filter_idx = 0; filter_idx < EFX_MCDI_FILTER_TBL_ROWS; filter_idx++) {
+		spec = efx_mcdi_filter_entry_spec(table, filter_idx);
+		if (!spec)
+			continue;
+
+		MCDI_SET_DWORD(inbuf, FILTER_OP_IN_OP,
+			       efx_mcdi_filter_is_exclusive(spec) ?
+			       MC_CMD_FILTER_OP_IN_OP_REMOVE :
+			       MC_CMD_FILTER_OP_IN_OP_UNSUBSCRIBE);
+		MCDI_SET_QWORD(inbuf, FILTER_OP_IN_HANDLE,
+			       table->entry[filter_idx].handle);
+		rc = efx_mcdi_rpc_quiet(efx, MC_CMD_FILTER_OP, inbuf,
+					sizeof(inbuf), NULL, 0, NULL);
+		if (rc)
+			netif_info(efx, drv, efx->net_dev,
+				   "%s: filter %04x remove failed\n",
+				   __func__, filter_idx);
+		kfree(spec);
+	}
+
+	vfree(table->entry);
+	kfree(table);
+}
+
+static void efx_mcdi_filter_mark_one_old(struct efx_nic *efx, uint16_t *id)
+{
+	struct efx_mcdi_filter_table *table = efx->filter_state;
+	unsigned int filter_idx;
+
+	efx_rwsem_assert_write_locked(&table->lock);
+
+	if (*id != EFX_EF10_FILTER_ID_INVALID) {
+		filter_idx = efx_mcdi_filter_get_unsafe_id(*id);
+		if (!table->entry[filter_idx].spec)
+			netif_dbg(efx, drv, efx->net_dev,
+				  "marked null spec old %04x:%04x\n", *id,
+				  filter_idx);
+		table->entry[filter_idx].spec |= EFX_EF10_FILTER_FLAG_AUTO_OLD;
+		*id = EFX_EF10_FILTER_ID_INVALID;
+	}
+}
+
+/* Mark old per-VLAN filters that may need to be removed */
+static void _efx_mcdi_filter_vlan_mark_old(struct efx_nic *efx,
+					   struct efx_mcdi_filter_vlan *vlan)
+{
+	struct efx_mcdi_filter_table *table = efx->filter_state;
+	unsigned int i;
+
+	for (i = 0; i < table->dev_uc_count; i++)
+		efx_mcdi_filter_mark_one_old(efx, &vlan->uc[i]);
+	for (i = 0; i < table->dev_mc_count; i++)
+		efx_mcdi_filter_mark_one_old(efx, &vlan->mc[i]);
+	for (i = 0; i < EFX_EF10_NUM_DEFAULT_FILTERS; i++)
+		efx_mcdi_filter_mark_one_old(efx, &vlan->default_filters[i]);
+}
+
+/*
+ * Mark old filters that may need to be removed.
+ * Caller must hold efx->filter_sem for read if race against
+ * efx_mcdi_filter_table_remove() is possible
+ */
+static void efx_mcdi_filter_mark_old(struct efx_nic *efx)
+{
+	struct efx_mcdi_filter_table *table = efx->filter_state;
+	struct efx_mcdi_filter_vlan *vlan;
+
+	down_write(&table->lock);
+	list_for_each_entry(vlan, &table->vlan_list, list)
+		_efx_mcdi_filter_vlan_mark_old(efx, vlan);
+	up_write(&table->lock);
+}
+
+int efx_mcdi_filter_add_vlan(struct efx_nic *efx, u16 vid)
+{
+	struct efx_mcdi_filter_table *table = efx->filter_state;
+	struct efx_mcdi_filter_vlan *vlan;
+	unsigned int i;
+
+	if (!efx_rwsem_assert_write_locked(&efx->filter_sem))
+		return -EINVAL;
+
+	vlan = efx_mcdi_filter_find_vlan(efx, vid);
+	if (WARN_ON(vlan)) {
+		netif_err(efx, drv, efx->net_dev,
+			  "VLAN %u already added\n", vid);
+		return -EALREADY;
+	}
+
+	vlan = kzalloc(sizeof(*vlan), GFP_KERNEL);
+	if (!vlan)
+		return -ENOMEM;
+
+	vlan->vid = vid;
+
+	for (i = 0; i < ARRAY_SIZE(vlan->uc); i++)
+		vlan->uc[i] = EFX_EF10_FILTER_ID_INVALID;
+	for (i = 0; i < ARRAY_SIZE(vlan->mc); i++)
+		vlan->mc[i] = EFX_EF10_FILTER_ID_INVALID;
+	for (i = 0; i < EFX_EF10_NUM_DEFAULT_FILTERS; i++)
+		vlan->default_filters[i] = EFX_EF10_FILTER_ID_INVALID;
+
+	list_add_tail(&vlan->list, &table->vlan_list);
+
+	if (efx_dev_registered(efx))
+		efx_mcdi_filter_vlan_sync_rx_mode(efx, vlan);
+
+	return 0;
+}
+
+static void efx_mcdi_filter_del_vlan_internal(struct efx_nic *efx,
+					      struct efx_mcdi_filter_vlan *vlan)
+{
+	unsigned int i;
+
+	/* See comment in efx_mcdi_filter_table_remove() */
+	if (!efx_rwsem_assert_write_locked(&efx->filter_sem))
+		return;
+
+	list_del(&vlan->list);
+
+	for (i = 0; i < ARRAY_SIZE(vlan->uc); i++)
+		efx_mcdi_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO,
+					      vlan->uc[i]);
+	for (i = 0; i < ARRAY_SIZE(vlan->mc); i++)
+		efx_mcdi_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO,
+					      vlan->mc[i]);
+	for (i = 0; i < EFX_EF10_NUM_DEFAULT_FILTERS; i++)
+		if (vlan->default_filters[i] != EFX_EF10_FILTER_ID_INVALID)
+			efx_mcdi_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO,
+						      vlan->default_filters[i]);
+
+	kfree(vlan);
+}
+
+void efx_mcdi_filter_del_vlan(struct efx_nic *efx, u16 vid)
+{
+	struct efx_mcdi_filter_vlan *vlan;
+
+	/* See comment in efx_mcdi_filter_table_remove() */
+	if (!efx_rwsem_assert_write_locked(&efx->filter_sem))
+		return;
+
+	vlan = efx_mcdi_filter_find_vlan(efx, vid);
+	if (!vlan) {
+		netif_err(efx, drv, efx->net_dev,
+			  "VLAN %u not found in filter state\n", vid);
+		return;
+	}
+
+	efx_mcdi_filter_del_vlan_internal(efx, vlan);
+}
+
+struct efx_mcdi_filter_vlan *efx_mcdi_filter_find_vlan(struct efx_nic *efx,
+						       u16 vid)
+{
+	struct efx_mcdi_filter_table *table = efx->filter_state;
+	struct efx_mcdi_filter_vlan *vlan;
+
+	WARN_ON(!rwsem_is_locked(&efx->filter_sem));
+
+	list_for_each_entry(vlan, &table->vlan_list, list) {
+		if (vlan->vid == vid)
+			return vlan;
+	}
+
+	return NULL;
+}
+
+void efx_mcdi_filter_cleanup_vlans(struct efx_nic *efx)
+{
+	struct efx_mcdi_filter_table *table = efx->filter_state;
+	struct efx_mcdi_filter_vlan *vlan, *next_vlan;
+
+	/* See comment in efx_mcdi_filter_table_remove() */
+	if (!efx_rwsem_assert_write_locked(&efx->filter_sem))
+		return;
+
+	if (!table)
+		return;
+
+	list_for_each_entry_safe(vlan, next_vlan, &table->vlan_list, list)
+		efx_mcdi_filter_del_vlan_internal(efx, vlan);
+}
+
+static void efx_mcdi_filter_uc_addr_list(struct efx_nic *efx)
+{
+	struct efx_mcdi_filter_table *table = efx->filter_state;
+	struct net_device *net_dev = efx->net_dev;
+	struct netdev_hw_addr *uc;
+	unsigned int i;
+
+	table->uc_promisc = !!(net_dev->flags & IFF_PROMISC);
+	ether_addr_copy(table->dev_uc_list[0].addr, net_dev->dev_addr);
+	i = 1;
+	netdev_for_each_uc_addr(uc, net_dev) {
+		if (i >= EFX_EF10_FILTER_DEV_UC_MAX) {
+			table->uc_promisc = true;
+			break;
+		}
+		ether_addr_copy(table->dev_uc_list[i].addr, uc->addr);
+		i++;
+	}
+
+	table->dev_uc_count = i;
+}
+
+static void efx_mcdi_filter_mc_addr_list(struct efx_nic *efx)
+{
+	struct efx_mcdi_filter_table *table = efx->filter_state;
+	struct net_device *net_dev = efx->net_dev;
+	struct netdev_hw_addr *mc;
+	unsigned int i;
+
+	table->mc_overflow = false;
+	table->mc_promisc = !!(net_dev->flags & (IFF_PROMISC | IFF_ALLMULTI));
+
+	i = 0;
+	netdev_for_each_mc_addr(mc, net_dev) {
+		if (i >= EFX_EF10_FILTER_DEV_MC_MAX) {
+			table->mc_promisc = true;
+			table->mc_overflow = true;
+			break;
+		}
+		ether_addr_copy(table->dev_mc_list[i].addr, mc->addr);
+		i++;
+	}
+
+	table->dev_mc_count = i;
+}
+
+/*
+ * Caller must hold efx->filter_sem for read if race against
+ * efx_mcdi_filter_table_remove() is possible
+ */
+void efx_mcdi_filter_sync_rx_mode(struct efx_nic *efx)
+{
+	struct efx_mcdi_filter_table *table = efx->filter_state;
+	struct net_device *net_dev = efx->net_dev;
+	struct efx_mcdi_filter_vlan *vlan;
+	bool vlan_filter;
+
+	if (!efx_dev_registered(efx))
+		return;
+
+	if (!table)
+		return;
+
+	efx_mcdi_filter_mark_old(efx);
+
+	/*
+	 * Copy/convert the address lists; add the primary station
+	 * address and broadcast address
+	 */
+	netif_addr_lock_bh(net_dev);
+	efx_mcdi_filter_uc_addr_list(efx);
+	efx_mcdi_filter_mc_addr_list(efx);
+	netif_addr_unlock_bh(net_dev);
+
+	/*
+	 * If VLAN filtering changes, all old filters are finally removed.
+	 * Do it in advance to avoid conflicts for unicast untagged and
+	 * VLAN 0 tagged filters.
+	 */
+	vlan_filter = !!(net_dev->features & NETIF_F_HW_VLAN_CTAG_FILTER);
+	if (table->vlan_filter != vlan_filter) {
+		table->vlan_filter = vlan_filter;
+		efx_mcdi_filter_remove_old(efx);
+	}
+
+	list_for_each_entry(vlan, &table->vlan_list, list)
+		efx_mcdi_filter_vlan_sync_rx_mode(efx, vlan);
+
+	efx_mcdi_filter_remove_old(efx);
+	table->mc_promisc_last = table->mc_promisc;
+}
+
+#ifdef CONFIG_RFS_ACCEL
+
+bool efx_mcdi_filter_rfs_expire_one(struct efx_nic *efx, u32 flow_id,
+				    unsigned int filter_idx)
+{
+	struct efx_filter_spec *spec, saved_spec;
+	struct efx_mcdi_filter_table *table;
+	struct efx_arfs_rule *rule = NULL;
+	bool ret = true, force = false;
+	u16 arfs_id;
+
+	down_read(&efx->filter_sem);
+	table = efx->filter_state;
+	down_write(&table->lock);
+	spec = efx_mcdi_filter_entry_spec(table, filter_idx);
+
+	if (!spec || spec->priority != EFX_FILTER_PRI_HINT)
+		goto out_unlock;
+
+	spin_lock_bh(&efx->rps_hash_lock);
+	if (!efx->rps_hash_table) {
+		/* In the absence of the table, we always return 0 to ARFS. */
+		arfs_id = 0;
+	} else {
+		rule = efx_rps_hash_find(efx, spec);
+		if (!rule)
+			/* ARFS table doesn't know of this filter, so remove it */
+			goto expire;
+		arfs_id = rule->arfs_id;
+		ret = efx_rps_check_rule(rule, filter_idx, &force);
+		if (force)
+			goto expire;
+		if (!ret) {
+			spin_unlock_bh(&efx->rps_hash_lock);
+			goto out_unlock;
+		}
+	}
+	if (!rps_may_expire_flow(efx->net_dev, spec->dmaq_id, flow_id, arfs_id))
+		ret = false;
+	else if (rule)
+		rule->filter_id = EFX_ARFS_FILTER_ID_REMOVING;
+expire:
+	saved_spec = *spec; /* remove operation will kfree spec */
+	spin_unlock_bh(&efx->rps_hash_lock);
+	/*
+	 * At this point (since we dropped the lock), another thread might queue
+	 * up a fresh insertion request (but the actual insertion will be held
+	 * up by our possession of the filter table lock).  In that case, it
+	 * will set rule->filter_id to EFX_ARFS_FILTER_ID_PENDING, meaning that
+	 * the rule is not removed by efx_rps_hash_del() below.
+	 */
+	if (ret)
+		ret = efx_mcdi_filter_remove_internal(efx, 1U << spec->priority,
+						      filter_idx, true) == 0;
+	/*
+	 * While we can't safely dereference rule (we dropped the lock), we can
+	 * still test it for NULL.
+	 */
+	if (ret && rule) {
+		/* Expiring, so remove entry from ARFS table */
+		spin_lock_bh(&efx->rps_hash_lock);
+		efx_rps_hash_del(efx, &saved_spec);
+		spin_unlock_bh(&efx->rps_hash_lock);
+	}
+out_unlock:
+	up_write(&table->lock);
+	up_read(&efx->filter_sem);
+	return ret;
+}
+
+#endif /* CONFIG_RFS_ACCEL */
+
+#define RSS_MODE_HASH_ADDRS	(1 << RSS_MODE_HASH_SRC_ADDR_LBN |\
+				 1 << RSS_MODE_HASH_DST_ADDR_LBN)
+#define RSS_MODE_HASH_PORTS	(1 << RSS_MODE_HASH_SRC_PORT_LBN |\
+				 1 << RSS_MODE_HASH_DST_PORT_LBN)
+#define RSS_CONTEXT_FLAGS_DEFAULT	(1 << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_TOEPLITZ_IPV4_EN_LBN |\
+					 1 << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_TOEPLITZ_TCPV4_EN_LBN |\
+					 1 << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_TOEPLITZ_IPV6_EN_LBN |\
+					 1 << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_TOEPLITZ_TCPV6_EN_LBN |\
+					 (RSS_MODE_HASH_ADDRS | RSS_MODE_HASH_PORTS) << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_TCP_IPV4_RSS_MODE_LBN |\
+					 RSS_MODE_HASH_ADDRS << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_UDP_IPV4_RSS_MODE_LBN |\
+					 RSS_MODE_HASH_ADDRS << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_OTHER_IPV4_RSS_MODE_LBN |\
+					 (RSS_MODE_HASH_ADDRS | RSS_MODE_HASH_PORTS) << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_TCP_IPV6_RSS_MODE_LBN |\
+					 RSS_MODE_HASH_ADDRS << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_UDP_IPV6_RSS_MODE_LBN |\
+					 RSS_MODE_HASH_ADDRS << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_OTHER_IPV6_RSS_MODE_LBN)
+
+int efx_mcdi_get_rss_context_flags(struct efx_nic *efx, u32 context, u32 *flags)
+{
+	/*
+	 * Firmware had a bug (sfc bug 61952) where it would not actually
+	 * fill in the flags field in the response to MC_CMD_RSS_CONTEXT_GET_FLAGS.
+	 * This meant that it would always contain whatever was previously
+	 * in the MCDI buffer.  Fortunately, all firmware versions with
+	 * this bug have the same default flags value for a newly-allocated
+	 * RSS context, and the only time we want to get the flags is just
+	 * after allocating.  Moreover, the response has a 32-bit hole
+	 * where the context ID would be in the request, so we can use an
+	 * overlength buffer in the request and pre-fill the flags field
+	 * with what we believe the default to be.  Thus if the firmware
+	 * has the bug, it will leave our pre-filled value in the flags
+	 * field of the response, and we will get the right answer.
+	 *
+	 * However, this does mean that this function should NOT be used if
+	 * the RSS context flags might not be their defaults - it is ONLY
+	 * reliably correct for a newly-allocated RSS context.
+	 */
+	MCDI_DECLARE_BUF(inbuf, MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_LEN);
+	MCDI_DECLARE_BUF(outbuf, MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_LEN);
+	size_t outlen;
+	int rc;
+
+	/* Check we have a hole for the context ID */
+	BUILD_BUG_ON(MC_CMD_RSS_CONTEXT_GET_FLAGS_IN_LEN != MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_FLAGS_OFST);
+	MCDI_SET_DWORD(inbuf, RSS_CONTEXT_GET_FLAGS_IN_RSS_CONTEXT_ID, context);
+	MCDI_SET_DWORD(inbuf, RSS_CONTEXT_GET_FLAGS_OUT_FLAGS,
+		       RSS_CONTEXT_FLAGS_DEFAULT);
+	rc = efx_mcdi_rpc(efx, MC_CMD_RSS_CONTEXT_GET_FLAGS, inbuf,
+			  sizeof(inbuf), outbuf, sizeof(outbuf), &outlen);
+	if (rc == 0) {
+		if (outlen < MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_LEN)
+			rc = -EIO;
+		else
+			*flags = MCDI_DWORD(outbuf, RSS_CONTEXT_GET_FLAGS_OUT_FLAGS);
+	}
+	return rc;
+}
+
+/*
+ * Attempt to enable 4-tuple UDP hashing on the specified RSS context.
+ * If we fail, we just leave the RSS context at its default hash settings,
+ * which is safe but may slightly reduce performance.
+ * Defaults are 4-tuple for TCP and 2-tuple for UDP and other-IP, so we
+ * just need to set the UDP ports flags (for both IP versions).
+ */
+void efx_mcdi_set_rss_context_flags(struct efx_nic *efx,
+				    struct efx_rss_context *ctx)
+{
+	MCDI_DECLARE_BUF(inbuf, MC_CMD_RSS_CONTEXT_SET_FLAGS_IN_LEN);
+	u32 flags;
+
+	BUILD_BUG_ON(MC_CMD_RSS_CONTEXT_SET_FLAGS_OUT_LEN != 0);
+
+	if (efx_mcdi_get_rss_context_flags(efx, ctx->context_id, &flags) != 0)
+		return;
+	MCDI_SET_DWORD(inbuf, RSS_CONTEXT_SET_FLAGS_IN_RSS_CONTEXT_ID,
+		       ctx->context_id);
+	flags |= RSS_MODE_HASH_PORTS << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_UDP_IPV4_RSS_MODE_LBN;
+	flags |= RSS_MODE_HASH_PORTS << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_UDP_IPV6_RSS_MODE_LBN;
+	MCDI_SET_DWORD(inbuf, RSS_CONTEXT_SET_FLAGS_IN_FLAGS, flags);
+	if (!efx_mcdi_rpc(efx, MC_CMD_RSS_CONTEXT_SET_FLAGS, inbuf, sizeof(inbuf),
+			  NULL, 0, NULL))
+		/* Succeeded, so UDP 4-tuple is now enabled */
+		ctx->rx_hash_udp_4tuple = true;
+}
+
+static int efx_mcdi_filter_alloc_rss_context(struct efx_nic *efx, bool exclusive,
+					     struct efx_rss_context *ctx,
+					     unsigned *context_size)
+{
+	MCDI_DECLARE_BUF(inbuf, MC_CMD_RSS_CONTEXT_ALLOC_IN_LEN);
+	MCDI_DECLARE_BUF(outbuf, MC_CMD_RSS_CONTEXT_ALLOC_OUT_LEN);
+	struct efx_ef10_nic_data *nic_data = efx->nic_data;
+	size_t outlen;
+	int rc;
+	u32 alloc_type = exclusive ?
+				MC_CMD_RSS_CONTEXT_ALLOC_IN_TYPE_EXCLUSIVE :
+				MC_CMD_RSS_CONTEXT_ALLOC_IN_TYPE_SHARED;
+	unsigned rss_spread = exclusive ?
+				efx->rss_spread :
+				min(rounddown_pow_of_two(efx->rss_spread),
+				    EFX_EF10_MAX_SHARED_RSS_CONTEXT_SIZE);
+
+	if (!exclusive && rss_spread == 1) {
+		ctx->context_id = EFX_MCDI_RSS_CONTEXT_INVALID;
+		if (context_size)
+			*context_size = 1;
+		return 0;
+	}
+
+	if (nic_data->datapath_caps &
+	    1 << MC_CMD_GET_CAPABILITIES_OUT_RX_RSS_LIMITED_LBN)
+		return -EOPNOTSUPP;
+
+	MCDI_SET_DWORD(inbuf, RSS_CONTEXT_ALLOC_IN_UPSTREAM_PORT_ID,
+		       nic_data->vport_id);
+	MCDI_SET_DWORD(inbuf, RSS_CONTEXT_ALLOC_IN_TYPE, alloc_type);
+	MCDI_SET_DWORD(inbuf, RSS_CONTEXT_ALLOC_IN_NUM_QUEUES, rss_spread);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_RSS_CONTEXT_ALLOC, inbuf, sizeof(inbuf),
+		outbuf, sizeof(outbuf), &outlen);
+	if (rc != 0)
+		return rc;
+
+	if (outlen < MC_CMD_RSS_CONTEXT_ALLOC_OUT_LEN)
+		return -EIO;
+
+	ctx->context_id = MCDI_DWORD(outbuf, RSS_CONTEXT_ALLOC_OUT_RSS_CONTEXT_ID);
+
+	if (context_size)
+		*context_size = rss_spread;
+
+	if (nic_data->datapath_caps &
+	    1 << MC_CMD_GET_CAPABILITIES_OUT_ADDITIONAL_RSS_MODES_LBN)
+		efx_mcdi_set_rss_context_flags(efx, ctx);
+
+	return 0;
+}
+
+static int efx_mcdi_filter_free_rss_context(struct efx_nic *efx, u32 context)
+{
+	MCDI_DECLARE_BUF(inbuf, MC_CMD_RSS_CONTEXT_FREE_IN_LEN);
+
+	MCDI_SET_DWORD(inbuf, RSS_CONTEXT_FREE_IN_RSS_CONTEXT_ID,
+		       context);
+	return efx_mcdi_rpc(efx, MC_CMD_RSS_CONTEXT_FREE, inbuf, sizeof(inbuf),
+			    NULL, 0, NULL);
+}
+
+static int efx_mcdi_filter_populate_rss_table(struct efx_nic *efx, u32 context,
+				       const u32 *rx_indir_table, const u8 *key)
+{
+	MCDI_DECLARE_BUF(tablebuf, MC_CMD_RSS_CONTEXT_SET_TABLE_IN_LEN);
+	MCDI_DECLARE_BUF(keybuf, MC_CMD_RSS_CONTEXT_SET_KEY_IN_LEN);
+	int i, rc;
+
+	MCDI_SET_DWORD(tablebuf, RSS_CONTEXT_SET_TABLE_IN_RSS_CONTEXT_ID,
+		       context);
+	BUILD_BUG_ON(ARRAY_SIZE(efx->rss_context.rx_indir_table) !=
+		     MC_CMD_RSS_CONTEXT_SET_TABLE_IN_INDIRECTION_TABLE_LEN);
+
+	/* This iterates over the length of efx->rss_context.rx_indir_table, but
+	 * copies bytes from rx_indir_table.  That's because the latter is a
+	 * pointer rather than an array, but should have the same length.
+	 * The efx->rss_context.rx_hash_key loop below is similar.
+	 */
+	for (i = 0; i < ARRAY_SIZE(efx->rss_context.rx_indir_table); ++i)
+		MCDI_PTR(tablebuf,
+			 RSS_CONTEXT_SET_TABLE_IN_INDIRECTION_TABLE)[i] =
+				(u8) rx_indir_table[i];
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_RSS_CONTEXT_SET_TABLE, tablebuf,
+			  sizeof(tablebuf), NULL, 0, NULL);
+	if (rc != 0)
+		return rc;
+
+	MCDI_SET_DWORD(keybuf, RSS_CONTEXT_SET_KEY_IN_RSS_CONTEXT_ID,
+		       context);
+	BUILD_BUG_ON(ARRAY_SIZE(efx->rss_context.rx_hash_key) !=
+		     MC_CMD_RSS_CONTEXT_SET_KEY_IN_TOEPLITZ_KEY_LEN);
+	for (i = 0; i < ARRAY_SIZE(efx->rss_context.rx_hash_key); ++i)
+		MCDI_PTR(keybuf, RSS_CONTEXT_SET_KEY_IN_TOEPLITZ_KEY)[i] = key[i];
+
+	return efx_mcdi_rpc(efx, MC_CMD_RSS_CONTEXT_SET_KEY, keybuf,
+			    sizeof(keybuf), NULL, 0, NULL);
+}
+
+void efx_mcdi_rx_free_indir_table(struct efx_nic *efx)
+{
+	int rc;
+
+	if (efx->rss_context.context_id != EFX_MCDI_RSS_CONTEXT_INVALID) {
+		rc = efx_mcdi_filter_free_rss_context(efx, efx->rss_context.context_id);
+		WARN_ON(rc != 0);
+	}
+	efx->rss_context.context_id = EFX_MCDI_RSS_CONTEXT_INVALID;
+}
+
+static int efx_mcdi_filter_rx_push_shared_rss_config(struct efx_nic *efx,
+					      unsigned *context_size)
+{
+	struct efx_ef10_nic_data *nic_data = efx->nic_data;
+	int rc = efx_mcdi_filter_alloc_rss_context(efx, false, &efx->rss_context,
+					    context_size);
+
+	if (rc != 0)
+		return rc;
+
+	nic_data->rx_rss_context_exclusive = false;
+	efx_set_default_rx_indir_table(efx, &efx->rss_context);
+	return 0;
+}
+
+static int efx_mcdi_filter_rx_push_exclusive_rss_config(struct efx_nic *efx,
+						 const u32 *rx_indir_table,
+						 const u8 *key)
+{
+	u32 old_rx_rss_context = efx->rss_context.context_id;
+	struct efx_ef10_nic_data *nic_data = efx->nic_data;
+	int rc;
+
+	if (efx->rss_context.context_id == EFX_MCDI_RSS_CONTEXT_INVALID ||
+	    !nic_data->rx_rss_context_exclusive) {
+		rc = efx_mcdi_filter_alloc_rss_context(efx, true, &efx->rss_context,
+						NULL);
+		if (rc == -EOPNOTSUPP)
+			return rc;
+		else if (rc != 0)
+			goto fail1;
+	}
+
+	rc = efx_mcdi_filter_populate_rss_table(efx, efx->rss_context.context_id,
+					 rx_indir_table, key);
+	if (rc != 0)
+		goto fail2;
+
+	if (efx->rss_context.context_id != old_rx_rss_context &&
+	    old_rx_rss_context != EFX_MCDI_RSS_CONTEXT_INVALID)
+		WARN_ON(efx_mcdi_filter_free_rss_context(efx, old_rx_rss_context) != 0);
+	nic_data->rx_rss_context_exclusive = true;
+	if (rx_indir_table != efx->rss_context.rx_indir_table)
+		memcpy(efx->rss_context.rx_indir_table, rx_indir_table,
+		       sizeof(efx->rss_context.rx_indir_table));
+	if (key != efx->rss_context.rx_hash_key)
+		memcpy(efx->rss_context.rx_hash_key, key,
+		       efx->type->rx_hash_key_size);
+
+	return 0;
+
+fail2:
+	if (old_rx_rss_context != efx->rss_context.context_id) {
+		WARN_ON(efx_mcdi_filter_free_rss_context(efx, efx->rss_context.context_id) != 0);
+		efx->rss_context.context_id = old_rx_rss_context;
+	}
+fail1:
+	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+	return rc;
+}
+
+int efx_mcdi_rx_push_rss_context_config(struct efx_nic *efx,
+					struct efx_rss_context *ctx,
+					const u32 *rx_indir_table,
+					const u8 *key)
+{
+	int rc;
+
+	WARN_ON(!mutex_is_locked(&efx->rss_lock));
+
+	if (ctx->context_id == EFX_MCDI_RSS_CONTEXT_INVALID) {
+		rc = efx_mcdi_filter_alloc_rss_context(efx, true, ctx, NULL);
+		if (rc)
+			return rc;
+	}
+
+	if (!rx_indir_table) /* Delete this context */
+		return efx_mcdi_filter_free_rss_context(efx, ctx->context_id);
+
+	rc = efx_mcdi_filter_populate_rss_table(efx, ctx->context_id,
+					 rx_indir_table, key);
+	if (rc)
+		return rc;
+
+	memcpy(ctx->rx_indir_table, rx_indir_table,
+	       sizeof(efx->rss_context.rx_indir_table));
+	memcpy(ctx->rx_hash_key, key, efx->type->rx_hash_key_size);
+
+	return 0;
+}
+
+int efx_mcdi_rx_pull_rss_context_config(struct efx_nic *efx,
+					struct efx_rss_context *ctx)
+{
+	MCDI_DECLARE_BUF(inbuf, MC_CMD_RSS_CONTEXT_GET_TABLE_IN_LEN);
+	MCDI_DECLARE_BUF(tablebuf, MC_CMD_RSS_CONTEXT_GET_TABLE_OUT_LEN);
+	MCDI_DECLARE_BUF(keybuf, MC_CMD_RSS_CONTEXT_GET_KEY_OUT_LEN);
+	size_t outlen;
+	int rc, i;
+
+	WARN_ON(!mutex_is_locked(&efx->rss_lock));
+
+	BUILD_BUG_ON(MC_CMD_RSS_CONTEXT_GET_TABLE_IN_LEN !=
+		     MC_CMD_RSS_CONTEXT_GET_KEY_IN_LEN);
+
+	if (ctx->context_id == EFX_MCDI_RSS_CONTEXT_INVALID)
+		return -ENOENT;
+
+	MCDI_SET_DWORD(inbuf, RSS_CONTEXT_GET_TABLE_IN_RSS_CONTEXT_ID,
+		       ctx->context_id);
+	BUILD_BUG_ON(ARRAY_SIZE(ctx->rx_indir_table) !=
+		     MC_CMD_RSS_CONTEXT_GET_TABLE_OUT_INDIRECTION_TABLE_LEN);
+	rc = efx_mcdi_rpc(efx, MC_CMD_RSS_CONTEXT_GET_TABLE, inbuf, sizeof(inbuf),
+			  tablebuf, sizeof(tablebuf), &outlen);
+	if (rc != 0)
+		return rc;
+
+	if (WARN_ON(outlen != MC_CMD_RSS_CONTEXT_GET_TABLE_OUT_LEN))
+		return -EIO;
+
+	for (i = 0; i < ARRAY_SIZE(ctx->rx_indir_table); i++)
+		ctx->rx_indir_table[i] = MCDI_PTR(tablebuf,
+				RSS_CONTEXT_GET_TABLE_OUT_INDIRECTION_TABLE)[i];
+
+	MCDI_SET_DWORD(inbuf, RSS_CONTEXT_GET_KEY_IN_RSS_CONTEXT_ID,
+		       ctx->context_id);
+	BUILD_BUG_ON(ARRAY_SIZE(ctx->rx_hash_key) !=
+		     MC_CMD_RSS_CONTEXT_SET_KEY_IN_TOEPLITZ_KEY_LEN);
+	rc = efx_mcdi_rpc(efx, MC_CMD_RSS_CONTEXT_GET_KEY, inbuf, sizeof(inbuf),
+			  keybuf, sizeof(keybuf), &outlen);
+	if (rc != 0)
+		return rc;
+
+	if (WARN_ON(outlen != MC_CMD_RSS_CONTEXT_GET_KEY_OUT_LEN))
+		return -EIO;
+
+	for (i = 0; i < ARRAY_SIZE(ctx->rx_hash_key); ++i)
+		ctx->rx_hash_key[i] = MCDI_PTR(
+				keybuf, RSS_CONTEXT_GET_KEY_OUT_TOEPLITZ_KEY)[i];
+
+	return 0;
+}
+
+int efx_mcdi_rx_pull_rss_config(struct efx_nic *efx)
+{
+	int rc;
+
+	mutex_lock(&efx->rss_lock);
+	rc = efx_mcdi_rx_pull_rss_context_config(efx, &efx->rss_context);
+	mutex_unlock(&efx->rss_lock);
+	return rc;
+}
+
+void efx_mcdi_rx_restore_rss_contexts(struct efx_nic *efx)
+{
+	struct efx_ef10_nic_data *nic_data = efx->nic_data;
+	struct efx_rss_context *ctx;
+	int rc;
+
+	WARN_ON(!mutex_is_locked(&efx->rss_lock));
+
+	if (!nic_data->must_restore_rss_contexts)
+		return;
+
+	list_for_each_entry(ctx, &efx->rss_context.list, list) {
+		/* previous NIC RSS context is gone */
+		ctx->context_id = EFX_MCDI_RSS_CONTEXT_INVALID;
+		/* so try to allocate a new one */
+		rc = efx_mcdi_rx_push_rss_context_config(efx, ctx,
+							 ctx->rx_indir_table,
+							 ctx->rx_hash_key);
+		if (rc)
+			netif_warn(efx, probe, efx->net_dev,
+				   "failed to restore RSS context %u, rc=%d"
+				   "; RSS filters may fail to be applied\n",
+				   ctx->user_id, rc);
+	}
+	nic_data->must_restore_rss_contexts = false;
+}
+
+int efx_mcdi_pf_rx_push_rss_config(struct efx_nic *efx, bool user,
+				   const u32 *rx_indir_table,
+				   const u8 *key)
+{
+	int rc;
+
+	if (efx->rss_spread == 1)
+		return 0;
+
+	if (!key)
+		key = efx->rss_context.rx_hash_key;
+
+	rc = efx_mcdi_filter_rx_push_exclusive_rss_config(efx, rx_indir_table, key);
+
+	if (rc == -ENOBUFS && !user) {
+		unsigned context_size;
+		bool mismatch = false;
+		size_t i;
+
+		for (i = 0;
+		     i < ARRAY_SIZE(efx->rss_context.rx_indir_table) && !mismatch;
+		     i++)
+			mismatch = rx_indir_table[i] !=
+				ethtool_rxfh_indir_default(i, efx->rss_spread);
+
+		rc = efx_mcdi_filter_rx_push_shared_rss_config(efx, &context_size);
+		if (rc == 0) {
+			if (context_size != efx->rss_spread)
+				netif_warn(efx, probe, efx->net_dev,
+					   "Could not allocate an exclusive RSS"
+					   " context; allocated a shared one of"
+					   " different size."
+					   " Wanted %u, got %u.\n",
+					   efx->rss_spread, context_size);
+			else if (mismatch)
+				netif_warn(efx, probe, efx->net_dev,
+					   "Could not allocate an exclusive RSS"
+					   " context; allocated a shared one but"
+					   " could not apply custom"
+					   " indirection.\n");
+			else
+				netif_info(efx, probe, efx->net_dev,
+					   "Could not allocate an exclusive RSS"
+					   " context; allocated a shared one.\n");
+		}
+	}
+	return rc;
+}
+
+int efx_mcdi_vf_rx_push_rss_config(struct efx_nic *efx, bool user,
+				   const u32 *rx_indir_table
+				   __attribute__ ((unused)),
+				   const u8 *key
+				   __attribute__ ((unused)))
+{
+	if (user)
+		return -EOPNOTSUPP;
+	if (efx->rss_context.context_id != EFX_MCDI_RSS_CONTEXT_INVALID)
+		return 0;
+	return efx_mcdi_filter_rx_push_shared_rss_config(efx, NULL);
+}
-- 
cgit v1.2.3-59-g8ed1b


From 6a7e25c7fb482dba3e80fec953f1907bcb24d52c Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@mellanox.com>
Date: Mon, 27 Jan 2020 09:20:28 +0200
Subject: net/core: Replace driver version to be kernel version

In order to stop useless driver version bumps and unify output
presented by ethtool -i, let's set default version string.

As Linus said in [1]: "Things are supposed to be backwards and
forwards compatible, because we don't accept breakage in user
space anyway. So versioning is pointless, and only causes
problems."

They cause problems when users start to see version changes
and expect specific set of features which will be different
for stable@, vanilla and distribution kernels.

Distribution kernels are based on some kernel version with extra
patches on top, for example, in RedHat world this "extra" is a lot
and for them your driver version say nothing. Users who run vanilla
kernels won't use driver version information too, because running
such kernels requires knowledge and understanding.

Another set of problems are related to difference in versioning scheme
and such doesn't allow to write meaningful automation which will work
sanely on all ethtool capable devices.

Before this change:
[leonro@erver ~]$ ethtool -i eth0
driver: virtio_net
version: 1.0.0
After this change and once ->version assignment will be deleted
from virtio_net:
[leonro@server ~]$ ethtool -i eth0
driver: virtio_net
version: 5.5.0-rc6+

Link: https://lore.kernel.org/ksummit-discuss/CA+55aFx9A=5cc0QZ7CySC4F2K7eYaEfzkdYEc9JaNgCcV25=rg@mail.gmail.com/
Link: https://lore.kernel.org/linux-rdma/20200122152627.14903-1-michal.kalderon@marvell.com/T/#md460ff8f976c532a89d6860411c3c50bb811038b
Link: https://lore.kernel.org/linux-rdma/20200127060835.GA570@unicorn.suse.cz
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Acked-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ethtool/ioctl.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
index b88dd14e41c6..b987052d91ef 100644
--- a/net/ethtool/ioctl.c
+++ b/net/ethtool/ioctl.c
@@ -17,6 +17,7 @@
 #include <linux/phy.h>
 #include <linux/bitops.h>
 #include <linux/uaccess.h>
+#include <linux/vermagic.h>
 #include <linux/vmalloc.h>
 #include <linux/sfp.h>
 #include <linux/slab.h>
@@ -655,6 +656,7 @@ static noinline_for_stack int ethtool_get_drvinfo(struct net_device *dev,
 
 	memset(&info, 0, sizeof(info));
 	info.cmd = ETHTOOL_GDRVINFO;
+	strlcpy(info.version, UTS_RELEASE, sizeof(info.version));
 	if (ops->get_drvinfo) {
 		ops->get_drvinfo(dev, &info);
 	} else if (dev->dev.parent && dev->dev.parent->driver) {
-- 
cgit v1.2.3-59-g8ed1b


From c312840cd79061af37158cb42590931cfa364c1b Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Mon, 27 Jan 2020 13:49:33 +0100
Subject: Revert "pktgen: Allow configuration of IPv6 source address range"

This reverts commit 7786a1af2a6bceb07860ec720e74714004438834.

It causes build failures on 32-bit, for example:

   net/core/pktgen.o: In function `mod_cur_headers':
>> pktgen.c:(.text.mod_cur_headers+0xba0): undefined reference to `__umoddi3'

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/pktgen.c | 98 -------------------------------------------------------
 1 file changed, 98 deletions(-)

diff --git a/net/core/pktgen.c b/net/core/pktgen.c
index 890be1b4877e..294bfcf0ce0e 100644
--- a/net/core/pktgen.c
+++ b/net/core/pktgen.c
@@ -323,10 +323,6 @@ struct pktgen_dev {
 	struct in6_addr max_in6_daddr;
 	struct in6_addr min_in6_saddr;
 	struct in6_addr max_in6_saddr;
-	u64  max_in6_h;
-	u64  max_in6_l;
-	u64  min_in6_h;
-	u64  min_in6_l;
 
 	/* If we're doing ranges, random or incremental, then this
 	 * defines the min/max for those ranges.
@@ -1359,59 +1355,6 @@ static ssize_t pktgen_if_write(struct file *file,
 		sprintf(pg_result, "OK: dst6_max=%s", buf);
 		return count;
 	}
-	if (!strcmp(name, "src6_min")) {
-		len = strn_len(&user_buffer[i], sizeof(buf) - 1);
-		if (len < 0)
-			return len;
-
-		pkt_dev->flags |= F_IPV6;
-
-		if (copy_from_user(buf, &user_buffer[i], len))
-			return -EFAULT;
-		buf[len] = 0;
-
-		in6_pton(buf, -1, pkt_dev->min_in6_saddr.s6_addr, -1, NULL);
-		snprintf(buf, sizeof(buf), "%pI6c", &pkt_dev->min_in6_saddr);
-
-		memcpy(&pkt_dev->min_in6_h, pkt_dev->min_in6_saddr.s6_addr, 8);
-		memcpy(&pkt_dev->min_in6_l, pkt_dev->min_in6_saddr.s6_addr + 8, 8);
-		pkt_dev->min_in6_h = be64_to_cpu(pkt_dev->min_in6_h);
-		pkt_dev->min_in6_l = be64_to_cpu(pkt_dev->min_in6_l);
-
-		pkt_dev->cur_in6_saddr = pkt_dev->min_in6_saddr;
-		if (debug)
-			pr_debug("src6_min set to: %s\n", buf);
-
-		i += len;
-		sprintf(pg_result, "OK: src6_min=%s", buf);
-		return count;
-	}
-	if (!strcmp(name, "src6_max")) {
-		len = strn_len(&user_buffer[i], sizeof(buf) - 1);
-		if (len < 0)
-			return len;
-
-		pkt_dev->flags |= F_IPV6;
-
-		if (copy_from_user(buf, &user_buffer[i], len))
-			return -EFAULT;
-		buf[len] = 0;
-
-		in6_pton(buf, -1, pkt_dev->max_in6_saddr.s6_addr, -1, NULL);
-		snprintf(buf, sizeof(buf), "%pI6c", &pkt_dev->max_in6_saddr);
-
-		memcpy(&pkt_dev->max_in6_h, pkt_dev->max_in6_saddr.s6_addr, 8);
-		memcpy(&pkt_dev->max_in6_l, pkt_dev->max_in6_saddr.s6_addr + 8, 8);
-		pkt_dev->max_in6_h = be64_to_cpu(pkt_dev->max_in6_h);
-		pkt_dev->max_in6_l = be64_to_cpu(pkt_dev->max_in6_l);
-
-		if (debug)
-			pr_debug("src6_max set to: %s\n", buf);
-
-		i += len;
-		sprintf(pg_result, "OK: src6_max=%s", buf);
-		return count;
-	}
 	if (!strcmp(name, "src6")) {
 		len = strn_len(&user_buffer[i], sizeof(buf) - 1);
 		if (len < 0)
@@ -2343,45 +2286,6 @@ static void set_cur_queue_map(struct pktgen_dev *pkt_dev)
 	pkt_dev->cur_queue_map  = pkt_dev->cur_queue_map % pkt_dev->odev->real_num_tx_queues;
 }
 
-/* generate ipv6 source addr */
-static void set_src_in6_addr(struct pktgen_dev *pkt_dev)
-{
-	u64 min6, max6, rand, i;
-	struct in6_addr addr6;
-	__be64 addr_l, *t;
-
-	min6 = pkt_dev->min_in6_l;
-	max6 = pkt_dev->max_in6_l;
-
-	/* only generate source address in least significant 64 bits range
-	 * most significant 64 bits must be equal
-	 */
-	if (pkt_dev->max_in6_h != pkt_dev->min_in6_h || min6 >= max6)
-		return;
-
-	addr6 = pkt_dev->min_in6_saddr;
-	t = (__be64 *)addr6.s6_addr + 1;
-
-	if (pkt_dev->flags & F_IPSRC_RND) {
-		do {
-			prandom_bytes(&rand, sizeof(rand));
-			rand = rand % (max6 - min6) + min6;
-			addr_l = cpu_to_be64(rand);
-			memcpy(t, &addr_l, 8);
-		} while (ipv6_addr_loopback(&addr6) ||
-			 ipv6_addr_v4mapped(&addr6) ||
-			 ipv6_addr_is_multicast(&addr6));
-	} else {
-		addr6 = pkt_dev->cur_in6_saddr;
-		i = be64_to_cpu(*t);
-		if (++i > max6)
-			i = min6;
-		addr_l = cpu_to_be64(i);
-		memcpy(t, &addr_l, 8);
-	}
-	pkt_dev->cur_in6_saddr = addr6;
-}
-
 /* Increment/randomize headers according to flags and current values
  * for IP src/dest, UDP src/dst port, MAC-Addr src/dst
  */
@@ -2550,8 +2454,6 @@ static void mod_cur_headers(struct pktgen_dev *pkt_dev)
 		}
 	} else {		/* IPV6 * */
 
-		set_src_in6_addr(pkt_dev);
-
 		if (!ipv6_addr_any(&pkt_dev->min_in6_daddr)) {
 			int i;
 
-- 
cgit v1.2.3-59-g8ed1b


From 165475779ba1d4758429f48d65960473df9dc35a Mon Sep 17 00:00:00 2001
From: Sunil Goutham <sgoutham@marvell.com>
Date: Mon, 27 Jan 2020 18:35:15 +0530
Subject: octeontx2-pf: Add Marvell OcteonTX2 NIC driver

This patch adds template for the Marvell's OcteonTX2 network
controller's physical function driver. Just the probe, PCI
specific initialization and netdev registration.

Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/marvell/octeontx2/Kconfig     |   8 +
 drivers/net/ethernet/marvell/octeontx2/Makefile    |   2 +
 .../net/ethernet/marvell/octeontx2/nic/Makefile    |  10 +
 .../ethernet/marvell/octeontx2/nic/otx2_common.h   |  77 ++++++++
 .../net/ethernet/marvell/octeontx2/nic/otx2_pf.c   | 214 +++++++++++++++++++++
 .../net/ethernet/marvell/octeontx2/nic/otx2_reg.h  |  51 +++++
 6 files changed, 362 insertions(+)
 create mode 100644 drivers/net/ethernet/marvell/octeontx2/nic/Makefile
 create mode 100644 drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
 create mode 100644 drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
 create mode 100644 drivers/net/ethernet/marvell/octeontx2/nic/otx2_reg.h

diff --git a/drivers/net/ethernet/marvell/octeontx2/Kconfig b/drivers/net/ethernet/marvell/octeontx2/Kconfig
index fb34fbd62088..ced514c05c97 100644
--- a/drivers/net/ethernet/marvell/octeontx2/Kconfig
+++ b/drivers/net/ethernet/marvell/octeontx2/Kconfig
@@ -25,3 +25,11 @@ config NDC_DIS_DYNAMIC_CACHING
 	  This config option disables caching of dynamic entries such as NIX SQEs
 	  , NPA stack pages etc in NDC. Also locks down NIX SQ/CQ/RQ/RSS and
 	  NPA Aura/Pool contexts.
+
+config OCTEONTX2_PF
+	tristate "Marvell OcteonTX2 NIC Physical Function driver"
+	select OCTEONTX2_MBOX
+	depends on (64BIT && COMPILE_TEST) || ARM64
+	depends on PCI
+	help
+	  This driver supports Marvell's OcteonTX2 NIC physical function.
diff --git a/drivers/net/ethernet/marvell/octeontx2/Makefile b/drivers/net/ethernet/marvell/octeontx2/Makefile
index e579dcd54c97..0064a69e0f72 100644
--- a/drivers/net/ethernet/marvell/octeontx2/Makefile
+++ b/drivers/net/ethernet/marvell/octeontx2/Makefile
@@ -3,4 +3,6 @@
 # Makefile for Marvell OcteonTX2 device drivers.
 #
 
+obj-$(CONFIG_OCTEONTX2_MBOX) += af/
 obj-$(CONFIG_OCTEONTX2_AF) += af/
+obj-$(CONFIG_OCTEONTX2_PF) += nic/
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/Makefile b/drivers/net/ethernet/marvell/octeontx2/nic/Makefile
new file mode 100644
index 000000000000..622b803e1a5d
--- /dev/null
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/Makefile
@@ -0,0 +1,10 @@
+# SPDX-License-Identifier: GPL-2.0
+#
+# Makefile for Marvell's OcteonTX2 ethernet device drivers
+#
+
+obj-$(CONFIG_OCTEONTX2_PF) += octeontx2_nicpf.o
+
+octeontx2_nicpf-y := otx2_pf.o
+
+ccflags-y += -I$(srctree)/drivers/net/ethernet/marvell/octeontx2/af
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
new file mode 100644
index 000000000000..9d52ab364471
--- /dev/null
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
@@ -0,0 +1,77 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Marvell OcteonTx2 RVU Ethernet driver
+ *
+ * Copyright (C) 2020 Marvell International Ltd.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef OTX2_COMMON_H
+#define OTX2_COMMON_H
+
+#include <linux/pci.h>
+
+#include "otx2_reg.h"
+
+/* PCI device IDs */
+#define PCI_DEVID_OCTEONTX2_RVU_PF              0xA063
+
+/* PCI BAR nos */
+#define PCI_CFG_REG_BAR_NUM                     2
+
+struct otx2_hw {
+	struct pci_dev		*pdev;
+	u16                     rx_queues;
+	u16                     tx_queues;
+	u16			max_queues;
+};
+
+struct otx2_nic {
+	void __iomem		*reg_base;
+	struct net_device	*netdev;
+
+	struct otx2_hw		hw;
+	struct pci_dev		*pdev;
+	struct device		*dev;
+};
+
+/* Register read/write APIs */
+static inline void __iomem *otx2_get_regaddr(struct otx2_nic *nic, u64 offset)
+{
+	u64 blkaddr;
+
+	switch ((offset >> RVU_FUNC_BLKADDR_SHIFT) & RVU_FUNC_BLKADDR_MASK) {
+	case BLKTYPE_NIX:
+		blkaddr = BLKADDR_NIX0;
+		break;
+	case BLKTYPE_NPA:
+		blkaddr = BLKADDR_NPA;
+		break;
+	default:
+		blkaddr = BLKADDR_RVUM;
+		break;
+	};
+
+	offset &= ~(RVU_FUNC_BLKADDR_MASK << RVU_FUNC_BLKADDR_SHIFT);
+	offset |= (blkaddr << RVU_FUNC_BLKADDR_SHIFT);
+
+	return nic->reg_base + offset;
+}
+
+static inline void otx2_write64(struct otx2_nic *nic, u64 offset, u64 val)
+{
+	void __iomem *addr = otx2_get_regaddr(nic, offset);
+
+	writeq(val, addr);
+}
+
+static inline u64 otx2_read64(struct otx2_nic *nic, u64 offset)
+{
+	void __iomem *addr = otx2_get_regaddr(nic, offset);
+
+	return readq(addr);
+}
+
+#endif /* OTX2_COMMON_H */
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
new file mode 100644
index 000000000000..d3af2006ce3c
--- /dev/null
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
@@ -0,0 +1,214 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Marvell OcteonTx2 RVU Physcial Function ethernet driver
+ *
+ * Copyright (C) 2020 Marvell International Ltd.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/module.h>
+#include <linux/interrupt.h>
+#include <linux/pci.h>
+#include <linux/etherdevice.h>
+#include <linux/of.h>
+#include <linux/if_vlan.h>
+#include <linux/iommu.h>
+#include <net/ip.h>
+
+#include "otx2_common.h"
+
+#define DRV_NAME	"octeontx2-nicpf"
+#define DRV_STRING	"Marvell OcteonTX2 NIC Physical Function Driver"
+#define DRV_VERSION	"1.0"
+
+/* Supported devices */
+static const struct pci_device_id otx2_pf_id_table[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_CAVIUM, PCI_DEVID_OCTEONTX2_RVU_PF) },
+	{ 0, }  /* end of table */
+};
+
+MODULE_AUTHOR("Marvell International Ltd.");
+MODULE_DESCRIPTION(DRV_STRING);
+MODULE_LICENSE("GPL v2");
+MODULE_VERSION(DRV_VERSION);
+MODULE_DEVICE_TABLE(pci, otx2_pf_id_table);
+
+static int otx2_set_real_num_queues(struct net_device *netdev,
+				    int tx_queues, int rx_queues)
+{
+	int err;
+
+	err = netif_set_real_num_tx_queues(netdev, tx_queues);
+	if (err) {
+		netdev_err(netdev,
+			   "Failed to set no of Tx queues: %d\n", tx_queues);
+		return err;
+	}
+
+	err = netif_set_real_num_rx_queues(netdev, rx_queues);
+	if (err)
+		netdev_err(netdev,
+			   "Failed to set no of Rx queues: %d\n", rx_queues);
+	return err;
+}
+
+static int otx2_open(struct net_device *netdev)
+{
+	netif_carrier_off(netdev);
+
+	return 0;
+}
+
+static int otx2_stop(struct net_device *netdev)
+{
+	return 0;
+}
+
+static const struct net_device_ops otx2_netdev_ops = {
+	.ndo_open		= otx2_open,
+	.ndo_stop		= otx2_stop,
+};
+
+static int otx2_check_pf_usable(struct otx2_nic *nic)
+{
+	u64 rev;
+
+	rev = otx2_read64(nic, RVU_PF_BLOCK_ADDRX_DISC(BLKADDR_RVUM));
+	rev = (rev >> 12) & 0xFF;
+	/* Check if AF has setup revision for RVUM block,
+	 * otherwise this driver probe should be deferred
+	 * until AF driver comes up.
+	 */
+	if (!rev) {
+		dev_warn(nic->dev,
+			 "AF is not initialized, deferring probe\n");
+		return -EPROBE_DEFER;
+	}
+	return 0;
+}
+
+static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id)
+{
+	struct device *dev = &pdev->dev;
+	struct net_device *netdev;
+	struct otx2_nic *pf;
+	struct otx2_hw *hw;
+	int err, qcount;
+
+	err = pcim_enable_device(pdev);
+	if (err) {
+		dev_err(dev, "Failed to enable PCI device\n");
+		return err;
+	}
+
+	err = pci_request_regions(pdev, DRV_NAME);
+	if (err) {
+		dev_err(dev, "PCI request regions failed 0x%x\n", err);
+		return err;
+	}
+
+	err = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(48));
+	if (err) {
+		dev_err(dev, "DMA mask config failed, abort\n");
+		goto err_release_regions;
+	}
+
+	pci_set_master(pdev);
+
+	/* Set number of queues */
+	qcount = min_t(int, num_online_cpus(), num_online_cpus());
+
+	netdev = alloc_etherdev_mqs(sizeof(*pf), qcount, qcount);
+	if (!netdev) {
+		err = -ENOMEM;
+		goto err_release_regions;
+	}
+
+	pci_set_drvdata(pdev, netdev);
+	SET_NETDEV_DEV(netdev, &pdev->dev);
+	pf = netdev_priv(netdev);
+	pf->netdev = netdev;
+	pf->pdev = pdev;
+	pf->dev = dev;
+
+	hw = &pf->hw;
+	hw->pdev = pdev;
+	hw->rx_queues = qcount;
+	hw->tx_queues = qcount;
+	hw->max_queues = qcount;
+
+	/* Map CSRs */
+	pf->reg_base = pcim_iomap(pdev, PCI_CFG_REG_BAR_NUM, 0);
+	if (!pf->reg_base) {
+		dev_err(dev, "Unable to map physical function CSRs, aborting\n");
+		err = -ENOMEM;
+		goto err_free_netdev;
+	}
+
+	err = otx2_check_pf_usable(pf);
+	if (err)
+		goto err_free_netdev;
+
+	err = otx2_set_real_num_queues(netdev, hw->tx_queues, hw->rx_queues);
+	if (err)
+		goto err_free_netdev;
+
+	netdev->netdev_ops = &otx2_netdev_ops;
+
+	err = register_netdev(netdev);
+	if (err) {
+		dev_err(dev, "Failed to register netdevice\n");
+		goto err_free_netdev;
+	}
+
+	return 0;
+
+err_free_netdev:
+	pci_set_drvdata(pdev, NULL);
+	free_netdev(netdev);
+err_release_regions:
+	pci_release_regions(pdev);
+	return err;
+}
+
+static void otx2_remove(struct pci_dev *pdev)
+{
+	struct net_device *netdev = pci_get_drvdata(pdev);
+	struct otx2_nic *pf;
+
+	if (!netdev)
+		return;
+
+	pf = netdev_priv(netdev);
+
+	unregister_netdev(netdev);
+	pci_free_irq_vectors(pf->pdev);
+	pci_set_drvdata(pdev, NULL);
+	free_netdev(netdev);
+	pci_release_regions(pdev);
+}
+
+static struct pci_driver otx2_pf_driver = {
+	.name = DRV_NAME,
+	.id_table = otx2_pf_id_table,
+	.probe = otx2_probe,
+	.shutdown = otx2_remove,
+	.remove = otx2_remove,
+};
+
+static int __init otx2_rvupf_init_module(void)
+{
+	pr_info("%s: %s\n", DRV_NAME, DRV_STRING);
+
+	return pci_register_driver(&otx2_pf_driver);
+}
+
+static void __exit otx2_rvupf_cleanup_module(void)
+{
+	pci_unregister_driver(&otx2_pf_driver);
+}
+
+module_init(otx2_rvupf_init_module);
+module_exit(otx2_rvupf_cleanup_module);
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_reg.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_reg.h
new file mode 100644
index 000000000000..d0bd64ab4f1a
--- /dev/null
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_reg.h
@@ -0,0 +1,51 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Marvell OcteonTx2 RVU Ethernet driver
+ *
+ * Copyright (C) 2020 Marvell International Ltd.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef OTX2_REG_H
+#define OTX2_REG_H
+
+#include <rvu_struct.h>
+
+/* RVU PF registers */
+#define	RVU_PF_VFX_PFVF_MBOX0		    (0x00000)
+#define	RVU_PF_VFX_PFVF_MBOX1		    (0x00008)
+#define RVU_PF_VFX_PFVF_MBOXX(a, b)         (0x0 | (a) << 12 | (b) << 3)
+#define RVU_PF_VF_BAR4_ADDR                 (0x10)
+#define RVU_PF_BLOCK_ADDRX_DISC(a)          (0x200 | (a) << 3)
+#define RVU_PF_VFME_STATUSX(a)              (0x800 | (a) << 3)
+#define RVU_PF_VFTRPENDX(a)                 (0x820 | (a) << 3)
+#define RVU_PF_VFTRPEND_W1SX(a)             (0x840 | (a) << 3)
+#define RVU_PF_VFPF_MBOX_INTX(a)            (0x880 | (a) << 3)
+#define RVU_PF_VFPF_MBOX_INT_W1SX(a)        (0x8A0 | (a) << 3)
+#define RVU_PF_VFPF_MBOX_INT_ENA_W1SX(a)    (0x8C0 | (a) << 3)
+#define RVU_PF_VFPF_MBOX_INT_ENA_W1CX(a)    (0x8E0 | (a) << 3)
+#define RVU_PF_VFFLR_INTX(a)                (0x900 | (a) << 3)
+#define RVU_PF_VFFLR_INT_W1SX(a)            (0x920 | (a) << 3)
+#define RVU_PF_VFFLR_INT_ENA_W1SX(a)        (0x940 | (a) << 3)
+#define RVU_PF_VFFLR_INT_ENA_W1CX(a)        (0x960 | (a) << 3)
+#define RVU_PF_VFME_INTX(a)                 (0x980 | (a) << 3)
+#define RVU_PF_VFME_INT_W1SX(a)             (0x9A0 | (a) << 3)
+#define RVU_PF_VFME_INT_ENA_W1SX(a)         (0x9C0 | (a) << 3)
+#define RVU_PF_VFME_INT_ENA_W1CX(a)         (0x9E0 | (a) << 3)
+#define RVU_PF_PFAF_MBOX0                   (0xC00)
+#define RVU_PF_PFAF_MBOX1                   (0xC08)
+#define RVU_PF_PFAF_MBOXX(a)                (0xC00 | (a) << 3)
+#define RVU_PF_INT                          (0xc20)
+#define RVU_PF_INT_W1S                      (0xc28)
+#define RVU_PF_INT_ENA_W1S                  (0xc30)
+#define RVU_PF_INT_ENA_W1C                  (0xc38)
+#define RVU_PF_MSIX_VECX_ADDR(a)            (0x000 | (a) << 4)
+#define RVU_PF_MSIX_VECX_CTL(a)             (0x008 | (a) << 4)
+#define RVU_PF_MSIX_PBAX(a)                 (0xF0000 | (a) << 3)
+
+#define RVU_FUNC_BLKADDR_SHIFT		20
+#define RVU_FUNC_BLKADDR_MASK		0x1FULL
+
+#endif /* OTX2_REG_H */
-- 
cgit v1.2.3-59-g8ed1b


From 5a6d7c9daef392370d873f8a626bb525629cdd6a Mon Sep 17 00:00:00 2001
From: Sunil Goutham <sgoutham@marvell.com>
Date: Mon, 27 Jan 2020 18:35:16 +0530
Subject: octeontx2-pf: Mailbox communication with AF

In the resource virtualization unit (RVU) each of the PF and AF
(admin function) share a 64KB of reserved memory region for
communication. This patch initializes PF <=> AF mailbox IRQs,
registers handlers for processing these communication messages.
Also adds support to process these messages in both directions
ie responses to PF initiated DOWN (PF => AF) messages and AF
initiated UP messages (AF => PF).

Mbox communication APIs and message formats are defined in AF driver
(drivers/net/ethernet/marvell/octeontx2/af), mbox.h from AF driver is
included here to avoid duplication.

Signed-off-by: Geetha sowjanya <gakula@marvell.com>
Signed-off-by: Christina Jacob <cjacob@marvell.com>
Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com>
Signed-off-by: Aleksey Makarov <amakarov@marvell.com>
Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../net/ethernet/marvell/octeontx2/nic/Makefile    |   2 +-
 .../ethernet/marvell/octeontx2/nic/otx2_common.c   |  28 ++
 .../ethernet/marvell/octeontx2/nic/otx2_common.h   | 166 ++++++++++
 .../net/ethernet/marvell/octeontx2/nic/otx2_pf.c   | 359 ++++++++++++++++++++-
 4 files changed, 552 insertions(+), 3 deletions(-)
 create mode 100644 drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c

diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/Makefile b/drivers/net/ethernet/marvell/octeontx2/nic/Makefile
index 622b803e1a5d..339fde892a36 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/Makefile
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/Makefile
@@ -5,6 +5,6 @@
 
 obj-$(CONFIG_OCTEONTX2_PF) += octeontx2_nicpf.o
 
-octeontx2_nicpf-y := otx2_pf.o
+octeontx2_nicpf-y := otx2_pf.o otx2_common.o
 
 ccflags-y += -I$(srctree)/drivers/net/ethernet/marvell/octeontx2/af
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
new file mode 100644
index 000000000000..cbab325872bc
--- /dev/null
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
@@ -0,0 +1,28 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Marvell OcteonTx2 RVU Ethernet driver
+ *
+ * Copyright (C) 2020 Marvell International Ltd.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/interrupt.h>
+#include <linux/pci.h>
+
+#include "otx2_reg.h"
+#include "otx2_common.h"
+
+#define M(_name, _id, _fn_name, _req_type, _rsp_type)			\
+int __weak								\
+otx2_mbox_up_handler_ ## _fn_name(struct otx2_nic *pfvf,		\
+				struct _req_type *req,			\
+				struct _rsp_type *rsp)			\
+{									\
+	/* Nothing to do here */					\
+	return 0;							\
+}									\
+EXPORT_SYMBOL(otx2_mbox_up_handler_ ## _fn_name);
+MBOX_UP_CGX_MESSAGES
+#undef M
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
index 9d52ab364471..a21eaaf6e824 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
@@ -13,6 +13,7 @@
 
 #include <linux/pci.h>
 
+#include <mbox.h>
 #include "otx2_reg.h"
 
 /* PCI device IDs */
@@ -20,12 +21,31 @@
 
 /* PCI BAR nos */
 #define PCI_CFG_REG_BAR_NUM                     2
+#define PCI_MBOX_BAR_NUM                        4
+
+#define NAME_SIZE                               32
+
+struct mbox {
+	struct otx2_mbox	mbox;
+	struct work_struct	mbox_wrk;
+	struct otx2_mbox	mbox_up;
+	struct work_struct	mbox_up_wrk;
+	struct otx2_nic		*pfvf;
+	void			*bbuf_base; /* Bounce buffer for mbox memory */
+	struct mutex		lock;	/* serialize mailbox access */
+	int			num_msgs; /* mbox number of messages */
+	int			up_num_msgs; /* mbox_up number of messages */
+};
 
 struct otx2_hw {
 	struct pci_dev		*pdev;
 	u16                     rx_queues;
 	u16                     tx_queues;
 	u16			max_queues;
+
+	/* MSI-X */
+	char			*irq_name;
+	cpumask_var_t           *affinity_mask;
 };
 
 struct otx2_nic {
@@ -35,6 +55,12 @@ struct otx2_nic {
 	struct otx2_hw		hw;
 	struct pci_dev		*pdev;
 	struct device		*dev;
+
+	/* Mbox */
+	struct mbox		mbox;
+	struct workqueue_struct *mbox_wq;
+
+	u16			pcifunc; /* RVU PF_FUNC */
 };
 
 /* Register read/write APIs */
@@ -74,4 +100,144 @@ static inline u64 otx2_read64(struct otx2_nic *nic, u64 offset)
 	return readq(addr);
 }
 
+/* Mbox bounce buffer APIs */
+static inline int otx2_mbox_bbuf_init(struct mbox *mbox, struct pci_dev *pdev)
+{
+	struct otx2_mbox *otx2_mbox;
+	struct otx2_mbox_dev *mdev;
+
+	mbox->bbuf_base = devm_kmalloc(&pdev->dev, MBOX_SIZE, GFP_KERNEL);
+	if (!mbox->bbuf_base)
+		return -ENOMEM;
+
+	/* Overwrite mbox mbase to point to bounce buffer, so that PF/VF
+	 * prepare all mbox messages in bounce buffer instead of directly
+	 * in hw mbox memory.
+	 */
+	otx2_mbox = &mbox->mbox;
+	mdev = &otx2_mbox->dev[0];
+	mdev->mbase = mbox->bbuf_base;
+
+	otx2_mbox = &mbox->mbox_up;
+	mdev = &otx2_mbox->dev[0];
+	mdev->mbase = mbox->bbuf_base;
+	return 0;
+}
+
+static inline void otx2_sync_mbox_bbuf(struct otx2_mbox *mbox, int devid)
+{
+	u16 msgs_offset = ALIGN(sizeof(struct mbox_hdr), MBOX_MSG_ALIGN);
+	void *hw_mbase = mbox->hwbase + (devid * MBOX_SIZE);
+	struct otx2_mbox_dev *mdev = &mbox->dev[devid];
+	struct mbox_hdr *hdr;
+	u64 msg_size;
+
+	if (mdev->mbase == hw_mbase)
+		return;
+
+	hdr = hw_mbase + mbox->rx_start;
+	msg_size = hdr->msg_size;
+
+	if (msg_size > mbox->rx_size - msgs_offset)
+		msg_size = mbox->rx_size - msgs_offset;
+
+	/* Copy mbox messages from mbox memory to bounce buffer */
+	memcpy(mdev->mbase + mbox->rx_start,
+	       hw_mbase + mbox->rx_start, msg_size + msgs_offset);
+}
+
+static inline void otx2_mbox_lock_init(struct mbox *mbox)
+{
+	mutex_init(&mbox->lock);
+}
+
+static inline void otx2_mbox_lock(struct mbox *mbox)
+{
+	mutex_lock(&mbox->lock);
+}
+
+static inline void otx2_mbox_unlock(struct mbox *mbox)
+{
+	mutex_unlock(&mbox->lock);
+}
+
+/* Mbox APIs */
+static inline int otx2_sync_mbox_msg(struct mbox *mbox)
+{
+	int err;
+
+	if (!otx2_mbox_nonempty(&mbox->mbox, 0))
+		return 0;
+	otx2_mbox_msg_send(&mbox->mbox, 0);
+	err = otx2_mbox_wait_for_rsp(&mbox->mbox, 0);
+	if (err)
+		return err;
+
+	return otx2_mbox_check_rsp_msgs(&mbox->mbox, 0);
+}
+
+static inline int otx2_sync_mbox_up_msg(struct mbox *mbox, int devid)
+{
+	int err;
+
+	if (!otx2_mbox_nonempty(&mbox->mbox_up, devid))
+		return 0;
+	otx2_mbox_msg_send(&mbox->mbox_up, devid);
+	err = otx2_mbox_wait_for_rsp(&mbox->mbox_up, devid);
+	if (err)
+		return err;
+
+	return otx2_mbox_check_rsp_msgs(&mbox->mbox_up, devid);
+}
+
+/* Use this API to send mbox msgs in atomic context
+ * where sleeping is not allowed
+ */
+static inline int otx2_sync_mbox_msg_busy_poll(struct mbox *mbox)
+{
+	int err;
+
+	if (!otx2_mbox_nonempty(&mbox->mbox, 0))
+		return 0;
+	otx2_mbox_msg_send(&mbox->mbox, 0);
+	err = otx2_mbox_busy_poll_for_rsp(&mbox->mbox, 0);
+	if (err)
+		return err;
+
+	return otx2_mbox_check_rsp_msgs(&mbox->mbox, 0);
+}
+
+#define M(_name, _id, _fn_name, _req_type, _rsp_type)                   \
+static struct _req_type __maybe_unused					\
+*otx2_mbox_alloc_msg_ ## _fn_name(struct mbox *mbox)                    \
+{									\
+	struct _req_type *req;						\
+									\
+	req = (struct _req_type *)otx2_mbox_alloc_msg_rsp(		\
+		&mbox->mbox, 0, sizeof(struct _req_type),		\
+		sizeof(struct _rsp_type));				\
+	if (!req)							\
+		return NULL;						\
+	req->hdr.sig = OTX2_MBOX_REQ_SIG;				\
+	req->hdr.id = _id;						\
+	return req;							\
+}
+
+MBOX_MESSAGES
+#undef M
+
+#define M(_name, _id, _fn_name, _req_type, _rsp_type)			\
+int									\
+otx2_mbox_up_handler_ ## _fn_name(struct otx2_nic *pfvf,		\
+				struct _req_type *req,			\
+				struct _rsp_type *rsp);			\
+
+MBOX_UP_CGX_MESSAGES
+#undef M
+
+#define	RVU_PFVF_PF_SHIFT	10
+#define	RVU_PFVF_PF_MASK	0x3F
+#define	RVU_PFVF_FUNC_SHIFT	0
+#define	RVU_PFVF_FUNC_MASK	0x3FF
+
 #endif /* OTX2_COMMON_H */
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
index d3af2006ce3c..e21bc10be3dd 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
@@ -35,6 +35,322 @@ MODULE_LICENSE("GPL v2");
 MODULE_VERSION(DRV_VERSION);
 MODULE_DEVICE_TABLE(pci, otx2_pf_id_table);
 
+enum {
+	TYPE_PFAF,
+	TYPE_PFVF,
+};
+
+static void otx2_queue_work(struct mbox *mw, struct workqueue_struct *mbox_wq,
+			    int first, int mdevs, u64 intr, int type)
+{
+	struct otx2_mbox_dev *mdev;
+	struct otx2_mbox *mbox;
+	struct mbox_hdr *hdr;
+	int i;
+
+	for (i = first; i < mdevs; i++) {
+		/* start from 0 */
+		if (!(intr & BIT_ULL(i - first)))
+			continue;
+
+		mbox = &mw->mbox;
+		mdev = &mbox->dev[i];
+		if (type == TYPE_PFAF)
+			otx2_sync_mbox_bbuf(mbox, i);
+		hdr = mdev->mbase + mbox->rx_start;
+		/* The hdr->num_msgs is set to zero immediately in the interrupt
+		 * handler to  ensure that it holds a correct value next time
+		 * when the interrupt handler is called.
+		 * pf->mbox.num_msgs holds the data for use in pfaf_mbox_handler
+		 * pf>mbox.up_num_msgs holds the data for use in
+		 * pfaf_mbox_up_handler.
+		 */
+		if (hdr->num_msgs) {
+			mw[i].num_msgs = hdr->num_msgs;
+			hdr->num_msgs = 0;
+			if (type == TYPE_PFAF)
+				memset(mbox->hwbase + mbox->rx_start, 0,
+				       ALIGN(sizeof(struct mbox_hdr),
+					     sizeof(u64)));
+
+			queue_work(mbox_wq, &mw[i].mbox_wrk);
+		}
+
+		mbox = &mw->mbox_up;
+		mdev = &mbox->dev[i];
+		if (type == TYPE_PFAF)
+			otx2_sync_mbox_bbuf(mbox, i);
+		hdr = mdev->mbase + mbox->rx_start;
+		if (hdr->num_msgs) {
+			mw[i].up_num_msgs = hdr->num_msgs;
+			hdr->num_msgs = 0;
+			if (type == TYPE_PFAF)
+				memset(mbox->hwbase + mbox->rx_start, 0,
+				       ALIGN(sizeof(struct mbox_hdr),
+					     sizeof(u64)));
+
+			queue_work(mbox_wq, &mw[i].mbox_up_wrk);
+		}
+	}
+}
+
+static void otx2_process_pfaf_mbox_msg(struct otx2_nic *pf,
+				       struct mbox_msghdr *msg)
+{
+	if (msg->id >= MBOX_MSG_MAX) {
+		dev_err(pf->dev,
+			"Mbox msg with unknown ID 0x%x\n", msg->id);
+		return;
+	}
+
+	if (msg->sig != OTX2_MBOX_RSP_SIG) {
+		dev_err(pf->dev,
+			"Mbox msg with wrong signature %x, ID 0x%x\n",
+			 msg->sig, msg->id);
+		return;
+	}
+
+	switch (msg->id) {
+	case MBOX_MSG_READY:
+		pf->pcifunc = msg->pcifunc;
+		break;
+	default:
+		if (msg->rc)
+			dev_err(pf->dev,
+				"Mbox msg response has err %d, ID 0x%x\n",
+				msg->rc, msg->id);
+		break;
+	}
+}
+
+static void otx2_pfaf_mbox_handler(struct work_struct *work)
+{
+	struct otx2_mbox_dev *mdev;
+	struct mbox_hdr *rsp_hdr;
+	struct mbox_msghdr *msg;
+	struct otx2_mbox *mbox;
+	struct mbox *af_mbox;
+	struct otx2_nic *pf;
+	int offset, id;
+
+	af_mbox = container_of(work, struct mbox, mbox_wrk);
+	mbox = &af_mbox->mbox;
+	mdev = &mbox->dev[0];
+	rsp_hdr = (struct mbox_hdr *)(mdev->mbase + mbox->rx_start);
+
+	offset = mbox->rx_start + ALIGN(sizeof(*rsp_hdr), MBOX_MSG_ALIGN);
+	pf = af_mbox->pfvf;
+
+	for (id = 0; id < af_mbox->num_msgs; id++) {
+		msg = (struct mbox_msghdr *)(mdev->mbase + offset);
+		otx2_process_pfaf_mbox_msg(pf, msg);
+		offset = mbox->rx_start + msg->next_msgoff;
+		mdev->msgs_acked++;
+	}
+
+	otx2_mbox_reset(mbox, 0);
+}
+
+static int otx2_process_mbox_msg_up(struct otx2_nic *pf,
+				    struct mbox_msghdr *req)
+{
+	/* Check if valid, if not reply with a invalid msg */
+	if (req->sig != OTX2_MBOX_REQ_SIG) {
+		otx2_reply_invalid_msg(&pf->mbox.mbox_up, 0, 0, req->id);
+		return -ENODEV;
+	}
+
+	switch (req->id) {
+#define M(_name, _id, _fn_name, _req_type, _rsp_type)			\
+	case _id: {							\
+		struct _rsp_type *rsp;					\
+		int err;						\
+									\
+		rsp = (struct _rsp_type *)otx2_mbox_alloc_msg(		\
+			&pf->mbox.mbox_up, 0,				\
+			sizeof(struct _rsp_type));			\
+		if (!rsp)						\
+			return -ENOMEM;					\
+									\
+		rsp->hdr.id = _id;					\
+		rsp->hdr.sig = OTX2_MBOX_RSP_SIG;			\
+		rsp->hdr.pcifunc = 0;					\
+		rsp->hdr.rc = 0;					\
+									\
+		err = otx2_mbox_up_handler_ ## _fn_name(		\
+			pf, (struct _req_type *)req, rsp);		\
+		return err;						\
+	}
+MBOX_UP_CGX_MESSAGES
+#undef M
+		break;
+	default:
+		otx2_reply_invalid_msg(&pf->mbox.mbox_up, 0, 0, req->id);
+		return -ENODEV;
+	}
+	return 0;
+}
+
+static void otx2_pfaf_mbox_up_handler(struct work_struct *work)
+{
+	struct mbox *af_mbox = container_of(work, struct mbox, mbox_up_wrk);
+	struct otx2_mbox *mbox = &af_mbox->mbox_up;
+	struct otx2_mbox_dev *mdev = &mbox->dev[0];
+	struct otx2_nic *pf = af_mbox->pfvf;
+	int offset, id, devid = 0;
+	struct mbox_hdr *rsp_hdr;
+	struct mbox_msghdr *msg;
+
+	rsp_hdr = (struct mbox_hdr *)(mdev->mbase + mbox->rx_start);
+
+	offset = mbox->rx_start + ALIGN(sizeof(*rsp_hdr), MBOX_MSG_ALIGN);
+
+	for (id = 0; id < af_mbox->up_num_msgs; id++) {
+		msg = (struct mbox_msghdr *)(mdev->mbase + offset);
+
+		devid = msg->pcifunc & RVU_PFVF_FUNC_MASK;
+		/* Skip processing VF's messages */
+		if (!devid)
+			otx2_process_mbox_msg_up(pf, msg);
+		offset = mbox->rx_start + msg->next_msgoff;
+	}
+
+	otx2_mbox_msg_send(mbox, 0);
+}
+
+static irqreturn_t otx2_pfaf_mbox_intr_handler(int irq, void *pf_irq)
+{
+	struct otx2_nic *pf = (struct otx2_nic *)pf_irq;
+	struct mbox *mbox;
+
+	/* Clear the IRQ */
+	otx2_write64(pf, RVU_PF_INT, BIT_ULL(0));
+
+	mbox = &pf->mbox;
+	otx2_queue_work(mbox, pf->mbox_wq, 0, 1, 1, TYPE_PFAF);
+
+	return IRQ_HANDLED;
+}
+
+static void otx2_disable_mbox_intr(struct otx2_nic *pf)
+{
+	int vector = pci_irq_vector(pf->pdev, RVU_PF_INT_VEC_AFPF_MBOX);
+
+	/* Disable AF => PF mailbox IRQ */
+	otx2_write64(pf, RVU_PF_INT_ENA_W1C, BIT_ULL(0));
+	free_irq(vector, pf);
+}
+
+static int otx2_register_mbox_intr(struct otx2_nic *pf, bool probe_af)
+{
+	struct otx2_hw *hw = &pf->hw;
+	struct msg_req *req;
+	char *irq_name;
+	int err;
+
+	/* Register mailbox interrupt handler */
+	irq_name = &hw->irq_name[RVU_PF_INT_VEC_AFPF_MBOX * NAME_SIZE];
+	snprintf(irq_name, NAME_SIZE, "RVUPFAF Mbox");
+	err = request_irq(pci_irq_vector(pf->pdev, RVU_PF_INT_VEC_AFPF_MBOX),
+			  otx2_pfaf_mbox_intr_handler, 0, irq_name, pf);
+	if (err) {
+		dev_err(pf->dev,
+			"RVUPF: IRQ registration failed for PFAF mbox irq\n");
+		return err;
+	}
+
+	/* Enable mailbox interrupt for msgs coming from AF.
+	 * First clear to avoid spurious interrupts, if any.
+	 */
+	otx2_write64(pf, RVU_PF_INT, BIT_ULL(0));
+	otx2_write64(pf, RVU_PF_INT_ENA_W1S, BIT_ULL(0));
+
+	if (!probe_af)
+		return 0;
+
+	/* Check mailbox communication with AF */
+	req = otx2_mbox_alloc_msg_ready(&pf->mbox);
+	if (!req) {
+		otx2_disable_mbox_intr(pf);
+		return -ENOMEM;
+	}
+	err = otx2_sync_mbox_msg(&pf->mbox);
+	if (err) {
+		dev_warn(pf->dev,
+			 "AF not responding to mailbox, deferring probe\n");
+		otx2_disable_mbox_intr(pf);
+		return -EPROBE_DEFER;
+	}
+
+	return 0;
+}
+
+static void otx2_pfaf_mbox_destroy(struct otx2_nic *pf)
+{
+	struct mbox *mbox = &pf->mbox;
+
+	if (pf->mbox_wq) {
+		flush_workqueue(pf->mbox_wq);
+		destroy_workqueue(pf->mbox_wq);
+		pf->mbox_wq = NULL;
+	}
+
+	if (mbox->mbox.hwbase)
+		iounmap((void __iomem *)mbox->mbox.hwbase);
+
+	otx2_mbox_destroy(&mbox->mbox);
+	otx2_mbox_destroy(&mbox->mbox_up);
+}
+
+static int otx2_pfaf_mbox_init(struct otx2_nic *pf)
+{
+	struct mbox *mbox = &pf->mbox;
+	void __iomem *hwbase;
+	int err;
+
+	mbox->pfvf = pf;
+	pf->mbox_wq = alloc_workqueue("otx2_pfaf_mailbox",
+				      WQ_UNBOUND | WQ_HIGHPRI |
+				      WQ_MEM_RECLAIM, 1);
+	if (!pf->mbox_wq)
+		return -ENOMEM;
+
+	/* Mailbox is a reserved memory (in RAM) region shared between
+	 * admin function (i.e AF) and this PF, shouldn't be mapped as
+	 * device memory to allow unaligned accesses.
+	 */
+	hwbase = ioremap_wc(pci_resource_start(pf->pdev, PCI_MBOX_BAR_NUM),
+			    pci_resource_len(pf->pdev, PCI_MBOX_BAR_NUM));
+	if (!hwbase) {
+		dev_err(pf->dev, "Unable to map PFAF mailbox region\n");
+		err = -ENOMEM;
+		goto exit;
+	}
+
+	err = otx2_mbox_init(&mbox->mbox, hwbase, pf->pdev, pf->reg_base,
+			     MBOX_DIR_PFAF, 1);
+	if (err)
+		goto exit;
+
+	err = otx2_mbox_init(&mbox->mbox_up, hwbase, pf->pdev, pf->reg_base,
+			     MBOX_DIR_PFAF_UP, 1);
+	if (err)
+		goto exit;
+
+	err = otx2_mbox_bbuf_init(mbox, pf->pdev);
+	if (err)
+		goto exit;
+
+	INIT_WORK(&mbox->mbox_wrk, otx2_pfaf_mbox_handler);
+	INIT_WORK(&mbox->mbox_up_wrk, otx2_pfaf_mbox_up_handler);
+	otx2_mbox_lock_init(&pf->mbox);
+
+	return 0;
+exit:
+	otx2_pfaf_mbox_destroy(pf);
+	return err;
+}
+
 static int otx2_set_real_num_queues(struct net_device *netdev,
 				    int tx_queues, int rx_queues)
 {
@@ -96,6 +412,7 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	struct otx2_nic *pf;
 	struct otx2_hw *hw;
 	int err, qcount;
+	int num_vec;
 
 	err = pcim_enable_device(pdev);
 	if (err) {
@@ -139,6 +456,17 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	hw->tx_queues = qcount;
 	hw->max_queues = qcount;
 
+	num_vec = pci_msix_vec_count(pdev);
+	hw->irq_name = devm_kmalloc_array(&hw->pdev->dev, num_vec, NAME_SIZE,
+					  GFP_KERNEL);
+	if (!hw->irq_name)
+		goto err_free_netdev;
+
+	hw->affinity_mask = devm_kcalloc(&hw->pdev->dev, num_vec,
+					 sizeof(cpumask_var_t), GFP_KERNEL);
+	if (!hw->affinity_mask)
+		goto err_free_netdev;
+
 	/* Map CSRs */
 	pf->reg_base = pcim_iomap(pdev, PCI_CFG_REG_BAR_NUM, 0);
 	if (!pf->reg_base) {
@@ -151,20 +479,44 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	if (err)
 		goto err_free_netdev;
 
+	err = pci_alloc_irq_vectors(hw->pdev, RVU_PF_INT_VEC_CNT,
+				    RVU_PF_INT_VEC_CNT, PCI_IRQ_MSIX);
+	if (err < 0) {
+		dev_err(dev, "%s: Failed to alloc %d IRQ vectors\n",
+			__func__, num_vec);
+		goto err_free_netdev;
+	}
+
+	/* Init PF <=> AF mailbox stuff */
+	err = otx2_pfaf_mbox_init(pf);
+	if (err)
+		goto err_free_irq_vectors;
+
+	/* Register mailbox interrupt */
+	err = otx2_register_mbox_intr(pf, true);
+	if (err)
+		goto err_mbox_destroy;
+
 	err = otx2_set_real_num_queues(netdev, hw->tx_queues, hw->rx_queues);
 	if (err)
-		goto err_free_netdev;
+		goto err_disable_mbox_intr;
 
 	netdev->netdev_ops = &otx2_netdev_ops;
 
 	err = register_netdev(netdev);
 	if (err) {
 		dev_err(dev, "Failed to register netdevice\n");
-		goto err_free_netdev;
+		goto err_disable_mbox_intr;
 	}
 
 	return 0;
 
+err_disable_mbox_intr:
+	otx2_disable_mbox_intr(pf);
+err_mbox_destroy:
+	otx2_pfaf_mbox_destroy(pf);
+err_free_irq_vectors:
+	pci_free_irq_vectors(hw->pdev);
 err_free_netdev:
 	pci_set_drvdata(pdev, NULL);
 	free_netdev(netdev);
@@ -184,9 +536,12 @@ static void otx2_remove(struct pci_dev *pdev)
 	pf = netdev_priv(netdev);
 
 	unregister_netdev(netdev);
+	otx2_disable_mbox_intr(pf);
+	otx2_pfaf_mbox_destroy(pf);
 	pci_free_irq_vectors(pf->pdev);
 	pci_set_drvdata(pdev, NULL);
 	free_netdev(netdev);
+
 	pci_release_regions(pdev);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 05fcc9e08955b7c04afb7f53402902f4a86e553c Mon Sep 17 00:00:00 2001
From: Sunil Goutham <sgoutham@marvell.com>
Date: Mon, 27 Jan 2020 18:35:17 +0530
Subject: octeontx2-pf: Attach NIX and NPA block LFs

For a PF to function as a NIC, NPA (for Rx buffers, Tx descriptors etc)
and NIX (for rcv, send and completion queues) are the minimum resources
needed. So request admin function (AF) to attach one each of NIX and NPA
block LFs (local functions).

Only AF can configure a LF's contexts, so request AF to allocate memory
for NPA aura/pool and NIX RQ/SQ/CQ HW contexts. Upon receiving response,
save some of the HW constants like number of pointers per stack page,
size of send queue buffer (SQBs, where SQEs are queued by HW) e.t.c which
are later used to initialize queues.

A HW context here is like a state machine maintained for a descriptor
queue. eg size, head/tail pointers, irq etc etc. HW maintains this in
memory.

Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../ethernet/marvell/octeontx2/nic/otx2_common.c   | 167 +++++++++++++++++++++
 .../ethernet/marvell/octeontx2/nic/otx2_common.h   |  36 +++++
 .../net/ethernet/marvell/octeontx2/nic/otx2_pf.c   |  67 ++++++++-
 .../ethernet/marvell/octeontx2/nic/otx2_struct.h   |  20 +++
 4 files changed, 286 insertions(+), 4 deletions(-)
 create mode 100644 drivers/net/ethernet/marvell/octeontx2/nic/otx2_struct.h

diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
index cbab325872bc..1fa09e9cfeb2 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
@@ -13,6 +13,173 @@
 
 #include "otx2_reg.h"
 #include "otx2_common.h"
+#include "otx2_struct.h"
+
+int otx2_config_nix(struct otx2_nic *pfvf)
+{
+	struct nix_lf_alloc_req  *nixlf;
+	struct nix_lf_alloc_rsp *rsp;
+	int err;
+
+	pfvf->qset.xqe_size = NIX_XQESZ_W16 ? 128 : 512;
+
+	/* Get memory to put this msg */
+	nixlf = otx2_mbox_alloc_msg_nix_lf_alloc(&pfvf->mbox);
+	if (!nixlf)
+		return -ENOMEM;
+
+	/* Set RQ/SQ/CQ counts */
+	nixlf->rq_cnt = pfvf->hw.rx_queues;
+	nixlf->sq_cnt = pfvf->hw.tx_queues;
+	nixlf->cq_cnt = pfvf->qset.cq_cnt;
+	nixlf->xqe_sz = NIX_XQESZ_W16;
+	/* We don't know absolute NPA LF idx attached.
+	 * AF will replace 'RVU_DEFAULT_PF_FUNC' with
+	 * NPA LF attached to this RVU PF/VF.
+	 */
+	nixlf->npa_func = RVU_DEFAULT_PF_FUNC;
+	/* Disable alignment pad, enable L2 length check,
+	 * enable L4 TCP/UDP checksum verification.
+	 */
+	nixlf->rx_cfg = BIT_ULL(33) | BIT_ULL(35) | BIT_ULL(37);
+
+	err = otx2_sync_mbox_msg(&pfvf->mbox);
+	if (err)
+		return err;
+
+	rsp = (struct nix_lf_alloc_rsp *)otx2_mbox_get_rsp(&pfvf->mbox.mbox, 0,
+							   &nixlf->hdr);
+	if (IS_ERR(rsp))
+		return PTR_ERR(rsp);
+
+	if (rsp->qints < 1)
+		return -ENXIO;
+
+	return rsp->hdr.rc;
+}
+
+int otx2_config_npa(struct otx2_nic *pfvf)
+{
+	struct otx2_qset *qset = &pfvf->qset;
+	struct npa_lf_alloc_req  *npalf;
+	struct otx2_hw *hw = &pfvf->hw;
+	int aura_cnt;
+
+	/* Pool - Stack of free buffer pointers
+	 * Aura - Alloc/frees pointers from/to pool for NIX DMA.
+	 */
+
+	if (!hw->pool_cnt)
+		return -EINVAL;
+
+	qset->pool = devm_kzalloc(pfvf->dev, sizeof(struct otx2_pool) *
+				  hw->pool_cnt, GFP_KERNEL);
+	if (!qset->pool)
+		return -ENOMEM;
+
+	/* Get memory to put this msg */
+	npalf = otx2_mbox_alloc_msg_npa_lf_alloc(&pfvf->mbox);
+	if (!npalf)
+		return -ENOMEM;
+
+	/* Set aura and pool counts */
+	npalf->nr_pools = hw->pool_cnt;
+	aura_cnt = ilog2(roundup_pow_of_two(hw->pool_cnt));
+	npalf->aura_sz = (aura_cnt >= ilog2(128)) ? (aura_cnt - 6) : 1;
+
+	return otx2_sync_mbox_msg(&pfvf->mbox);
+}
+
+int otx2_detach_resources(struct mbox *mbox)
+{
+	struct rsrc_detach *detach;
+
+	otx2_mbox_lock(mbox);
+	detach = otx2_mbox_alloc_msg_detach_resources(mbox);
+	if (!detach) {
+		otx2_mbox_unlock(mbox);
+		return -ENOMEM;
+	}
+
+	/* detach all */
+	detach->partial = false;
+
+	/* Send detach request to AF */
+	otx2_mbox_msg_send(&mbox->mbox, 0);
+	otx2_mbox_unlock(mbox);
+	return 0;
+}
+
+int otx2_attach_npa_nix(struct otx2_nic *pfvf)
+{
+	struct rsrc_attach *attach;
+	struct msg_req *msix;
+	int err;
+
+	otx2_mbox_lock(&pfvf->mbox);
+	/* Get memory to put this msg */
+	attach = otx2_mbox_alloc_msg_attach_resources(&pfvf->mbox);
+	if (!attach) {
+		otx2_mbox_unlock(&pfvf->mbox);
+		return -ENOMEM;
+	}
+
+	attach->npalf = true;
+	attach->nixlf = true;
+
+	/* Send attach request to AF */
+	err = otx2_sync_mbox_msg(&pfvf->mbox);
+	if (err) {
+		otx2_mbox_unlock(&pfvf->mbox);
+		return err;
+	}
+
+	/* Get NPA and NIX MSIX vector offsets */
+	msix = otx2_mbox_alloc_msg_msix_offset(&pfvf->mbox);
+	if (!msix) {
+		otx2_mbox_unlock(&pfvf->mbox);
+		return -ENOMEM;
+	}
+
+	err = otx2_sync_mbox_msg(&pfvf->mbox);
+	if (err) {
+		otx2_mbox_unlock(&pfvf->mbox);
+		return err;
+	}
+	otx2_mbox_unlock(&pfvf->mbox);
+
+	if (pfvf->hw.npa_msixoff == MSIX_VECTOR_INVALID ||
+	    pfvf->hw.nix_msixoff == MSIX_VECTOR_INVALID) {
+		dev_err(pfvf->dev,
+			"RVUPF: Invalid MSIX vector offset for NPA/NIX\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+/* Mbox message handlers */
+void mbox_handler_npa_lf_alloc(struct otx2_nic *pfvf,
+			       struct npa_lf_alloc_rsp *rsp)
+{
+	pfvf->hw.stack_pg_ptrs = rsp->stack_pg_ptrs;
+	pfvf->hw.stack_pg_bytes = rsp->stack_pg_bytes;
+}
+
+void mbox_handler_nix_lf_alloc(struct otx2_nic *pfvf,
+			       struct nix_lf_alloc_rsp *rsp)
+{
+	pfvf->hw.sqb_size = rsp->sqb_size;
+	pfvf->hw.rx_chan_base = rsp->rx_chan_base;
+	pfvf->hw.tx_chan_base = rsp->tx_chan_base;
+}
+
+void mbox_handler_msix_offset(struct otx2_nic *pfvf,
+			      struct msix_offset_rsp *rsp)
+{
+	pfvf->hw.npa_msixoff = rsp->npa_msixoff;
+	pfvf->hw.nix_msixoff = rsp->nix_msixoff;
+}
 
 #define M(_name, _id, _fn_name, _req_type, _rsp_type)			\
 int __weak								\
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
index a21eaaf6e824..cdb1c567d5dc 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
@@ -25,6 +25,17 @@
 
 #define NAME_SIZE                               32
 
+struct otx2_pool {
+	struct qmem		*stack;
+};
+
+struct otx2_qset {
+#define OTX2_MAX_CQ_CNT		64
+	u16			cq_cnt;
+	u16			xqe_size; /* Size of CQE i.e 128 or 512 bytes */
+	struct otx2_pool	*pool;
+};
+
 struct mbox {
 	struct otx2_mbox	mbox;
 	struct work_struct	mbox_wrk;
@@ -42,8 +53,19 @@ struct otx2_hw {
 	u16                     rx_queues;
 	u16                     tx_queues;
 	u16			max_queues;
+	u16			pool_cnt;
+
+	/* NPA */
+	u32			stack_pg_ptrs;  /* No of ptrs per stack page */
+	u32			stack_pg_bytes; /* Size of stack page */
+	u16			sqb_size;
+
+	u16			rx_chan_base;
+	u16			tx_chan_base;
 
 	/* MSI-X */
+	u16			npa_msixoff; /* Offset of NPA vectors */
+	u16			nix_msixoff; /* Offset of NIX vectors */
 	char			*irq_name;
 	cpumask_var_t           *affinity_mask;
 };
@@ -52,6 +74,7 @@ struct otx2_nic {
 	void __iomem		*reg_base;
 	struct net_device	*netdev;
 
+	struct otx2_qset	qset;
 	struct otx2_hw		hw;
 	struct pci_dev		*pdev;
 	struct device		*dev;
@@ -240,4 +263,17 @@ MBOX_UP_CGX_MESSAGES
 #define	RVU_PFVF_FUNC_SHIFT	0
 #define	RVU_PFVF_FUNC_MASK	0x3FF
 
+/* RVU block related APIs */
+int otx2_attach_npa_nix(struct otx2_nic *pfvf);
+int otx2_detach_resources(struct mbox *mbox);
+int otx2_config_npa(struct otx2_nic *pfvf);
+int otx2_config_nix(struct otx2_nic *pfvf);
+
+/* Mbox handlers */
+void mbox_handler_msix_offset(struct otx2_nic *pfvf,
+			      struct msix_offset_rsp *rsp);
+void mbox_handler_npa_lf_alloc(struct otx2_nic *pfvf,
+			       struct npa_lf_alloc_rsp *rsp);
+void mbox_handler_nix_lf_alloc(struct otx2_nic *pfvf,
+			       struct nix_lf_alloc_rsp *rsp);
 #endif /* OTX2_COMMON_H */
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
index e21bc10be3dd..ef5dba442abf 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
@@ -114,6 +114,15 @@ static void otx2_process_pfaf_mbox_msg(struct otx2_nic *pf,
 	case MBOX_MSG_READY:
 		pf->pcifunc = msg->pcifunc;
 		break;
+	case MBOX_MSG_MSIX_OFFSET:
+		mbox_handler_msix_offset(pf, (struct msix_offset_rsp *)msg);
+		break;
+	case MBOX_MSG_NPA_LF_ALLOC:
+		mbox_handler_npa_lf_alloc(pf, (struct npa_lf_alloc_rsp *)msg);
+		break;
+	case MBOX_MSG_NIX_LF_ALLOC:
+		mbox_handler_nix_lf_alloc(pf, (struct nix_lf_alloc_rsp *)msg);
+		break;
 	default:
 		if (msg->rc)
 			dev_err(pf->dev,
@@ -372,9 +381,20 @@ static int otx2_set_real_num_queues(struct net_device *netdev,
 
 static int otx2_open(struct net_device *netdev)
 {
+	struct otx2_nic *pf = netdev_priv(netdev);
+	int err = 0;
+
 	netif_carrier_off(netdev);
 
-	return 0;
+	pf->qset.cq_cnt = pf->hw.rx_queues + pf->hw.tx_queues;
+
+	/* NPA init */
+	err = otx2_config_npa(pf);
+	if (err)
+		return err;
+
+	/* NIX init */
+	return otx2_config_nix(pf);
 }
 
 static int otx2_stop(struct net_device *netdev)
@@ -405,6 +425,31 @@ static int otx2_check_pf_usable(struct otx2_nic *nic)
 	return 0;
 }
 
+static int otx2_realloc_msix_vectors(struct otx2_nic *pf)
+{
+	struct otx2_hw *hw = &pf->hw;
+	int num_vec, err;
+
+	/* NPA interrupts are inot registered, so alloc only
+	 * upto NIX vector offset.
+	 */
+	num_vec = hw->nix_msixoff;
+#define NIX_LF_CINT_VEC_START			0x40
+	num_vec += NIX_LF_CINT_VEC_START + hw->max_queues;
+
+	otx2_disable_mbox_intr(pf);
+	pci_free_irq_vectors(hw->pdev);
+	pci_free_irq_vectors(hw->pdev);
+	err = pci_alloc_irq_vectors(hw->pdev, num_vec, num_vec, PCI_IRQ_MSIX);
+	if (err < 0) {
+		dev_err(pf->dev, "%s: Failed to realloc %d IRQ vectors\n",
+			__func__, num_vec);
+		return err;
+	}
+
+	return otx2_register_mbox_intr(pf, false);
+}
+
 static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 {
 	struct device *dev = &pdev->dev;
@@ -435,7 +480,7 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	pci_set_master(pdev);
 
 	/* Set number of queues */
-	qcount = min_t(int, num_online_cpus(), num_online_cpus());
+	qcount = min_t(int, num_online_cpus(), OTX2_MAX_CQ_CNT);
 
 	netdev = alloc_etherdev_mqs(sizeof(*pf), qcount, qcount);
 	if (!netdev) {
@@ -497,20 +542,33 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	if (err)
 		goto err_mbox_destroy;
 
-	err = otx2_set_real_num_queues(netdev, hw->tx_queues, hw->rx_queues);
+	/* Request AF to attach NPA and NIX LFs to this PF.
+	 * NIX and NPA LFs are needed for this PF to function as a NIC.
+	 */
+	err = otx2_attach_npa_nix(pf);
 	if (err)
 		goto err_disable_mbox_intr;
 
+	err = otx2_realloc_msix_vectors(pf);
+	if (err)
+		goto err_detach_rsrc;
+
+	err = otx2_set_real_num_queues(netdev, hw->tx_queues, hw->rx_queues);
+	if (err)
+		goto err_detach_rsrc;
+
 	netdev->netdev_ops = &otx2_netdev_ops;
 
 	err = register_netdev(netdev);
 	if (err) {
 		dev_err(dev, "Failed to register netdevice\n");
-		goto err_disable_mbox_intr;
+		goto err_detach_rsrc;
 	}
 
 	return 0;
 
+err_detach_rsrc:
+	otx2_detach_resources(&pf->mbox);
 err_disable_mbox_intr:
 	otx2_disable_mbox_intr(pf);
 err_mbox_destroy:
@@ -536,6 +594,7 @@ static void otx2_remove(struct pci_dev *pdev)
 	pf = netdev_priv(netdev);
 
 	unregister_netdev(netdev);
+	otx2_detach_resources(&pf->mbox);
 	otx2_disable_mbox_intr(pf);
 	otx2_pfaf_mbox_destroy(pf);
 	pci_free_irq_vectors(pf->pdev);
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_struct.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_struct.h
new file mode 100644
index 000000000000..e37f89f6c965
--- /dev/null
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_struct.h
@@ -0,0 +1,20 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Marvell OcteonTx2 RVU Ethernet driver
+ *
+ * Copyright (C) 2020 Marvell International Ltd.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef OTX2_STRUCT_H
+#define OTX2_STRUCT_H
+
+/* NIX WQE/CQE size 128 byte or 512 byte */
+enum nix_cqesz_e {
+	NIX_XQESZ_W64 = 0x0,
+	NIX_XQESZ_W16 = 0x1,
+};
+
+#endif /* OTX2_STRUCT_H */
-- 
cgit v1.2.3-59-g8ed1b


From caa2da34fd25a37e9fd43343b6966fb9d730a6d5 Mon Sep 17 00:00:00 2001
From: Sunil Goutham <sgoutham@marvell.com>
Date: Mon, 27 Jan 2020 18:35:18 +0530
Subject: octeontx2-pf: Initialize and config queues

This patch does the initialization of all queues ie the
receive buffer pools, receive and transmit queues, completion
or notification queues etc. Allocates all required resources
(eg transmit schedulers, receive buffers etc) and configures
them for proper functioning of queues. Also sets up receive
queue's RED dropping levels.

Co-developed-by: Geetha sowjanya <gakula@marvell.com>
Signed-off-by: Geetha sowjanya <gakula@marvell.com>
Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/marvell/octeontx2/af/common.h |   9 +-
 .../ethernet/marvell/octeontx2/nic/otx2_common.c   | 723 +++++++++++++++++++++
 .../ethernet/marvell/octeontx2/nic/otx2_common.h   | 131 +++-
 .../net/ethernet/marvell/octeontx2/nic/otx2_pf.c   | 235 ++++++-
 .../net/ethernet/marvell/octeontx2/nic/otx2_reg.h  |  96 +++
 .../ethernet/marvell/octeontx2/nic/otx2_struct.h   |  11 +
 .../net/ethernet/marvell/octeontx2/nic/otx2_txrx.h | 102 +++
 7 files changed, 1290 insertions(+), 17 deletions(-)
 create mode 100644 drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h

diff --git a/drivers/net/ethernet/marvell/octeontx2/af/common.h b/drivers/net/ethernet/marvell/octeontx2/af/common.h
index 784207bae5f8..cd33c2e6ca5f 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/common.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/common.h
@@ -143,8 +143,13 @@ enum nix_scheduler {
 	NIX_TXSCH_LVL_CNT = 0x5,
 };
 
-#define TXSCH_TL1_DFLT_RR_QTM      ((1 << 24) - 1)
-#define TXSCH_TL1_DFLT_RR_PRIO     (0x1ull)
+#define TXSCH_RR_QTM_MAX		((1 << 24) - 1)
+#define TXSCH_TL1_DFLT_RR_QTM		TXSCH_RR_QTM_MAX
+#define TXSCH_TL1_DFLT_RR_PRIO		(0x1ull)
+#define MAX_SCHED_WEIGHT		0xFF
+#define DFLT_RR_WEIGHT			71
+#define DFLT_RR_QTM	((DFLT_RR_WEIGHT * TXSCH_RR_QTM_MAX) \
+			 / MAX_SCHED_WEIGHT)
 
 /* Min/Max packet sizes, excluding FCS */
 #define	NIC_HW_MIN_FRS			40
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
index 1fa09e9cfeb2..3ebbf0462b8f 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
@@ -15,6 +15,388 @@
 #include "otx2_common.h"
 #include "otx2_struct.h"
 
+dma_addr_t otx2_alloc_rbuf(struct otx2_nic *pfvf, struct otx2_pool *pool,
+			   gfp_t gfp)
+{
+	dma_addr_t iova;
+
+	/* Check if request can be accommodated in previous allocated page */
+	if (pool->page && ((pool->page_offset + pool->rbsize) <=
+	    (PAGE_SIZE << pool->rbpage_order))) {
+		pool->pageref++;
+		goto ret;
+	}
+
+	otx2_get_page(pool);
+
+	/* Allocate a new page */
+	pool->page = alloc_pages(gfp | __GFP_COMP | __GFP_NOWARN,
+				 pool->rbpage_order);
+	if (unlikely(!pool->page))
+		return -ENOMEM;
+
+	pool->page_offset = 0;
+ret:
+	iova = (u64)otx2_dma_map_page(pfvf, pool->page, pool->page_offset,
+				      pool->rbsize, DMA_FROM_DEVICE);
+	if (!iova) {
+		if (!pool->page_offset)
+			__free_pages(pool->page, pool->rbpage_order);
+		pool->page = NULL;
+		return -ENOMEM;
+	}
+	pool->page_offset += pool->rbsize;
+	return iova;
+}
+
+static int otx2_get_link(struct otx2_nic *pfvf)
+{
+	int link = 0;
+	u16 map;
+
+	/* cgx lmac link */
+	if (pfvf->hw.tx_chan_base >= CGX_CHAN_BASE) {
+		map = pfvf->hw.tx_chan_base & 0x7FF;
+		link = 4 * ((map >> 8) & 0xF) + ((map >> 4) & 0xF);
+	}
+	/* LBK channel */
+	if (pfvf->hw.tx_chan_base < SDP_CHAN_BASE)
+		link = 12;
+
+	return link;
+}
+
+int otx2_txschq_config(struct otx2_nic *pfvf, int lvl)
+{
+	struct otx2_hw *hw = &pfvf->hw;
+	struct nix_txschq_config *req;
+	u64 schq, parent;
+
+	req = otx2_mbox_alloc_msg_nix_txschq_cfg(&pfvf->mbox);
+	if (!req)
+		return -ENOMEM;
+
+	req->lvl = lvl;
+	req->num_regs = 1;
+
+	schq = hw->txschq_list[lvl][0];
+	/* Set topology e.t.c configuration */
+	if (lvl == NIX_TXSCH_LVL_SMQ) {
+		req->reg[0] = NIX_AF_SMQX_CFG(schq);
+		req->regval[0] |= (0x20ULL << 51) | (0x80ULL << 39) |
+				  (0x2ULL << 36);
+		req->num_regs++;
+		/* MDQ config */
+		parent =  hw->txschq_list[NIX_TXSCH_LVL_TL4][0];
+		req->reg[1] = NIX_AF_MDQX_PARENT(schq);
+		req->regval[1] = parent << 16;
+		req->num_regs++;
+		/* Set DWRR quantum */
+		req->reg[2] = NIX_AF_MDQX_SCHEDULE(schq);
+		req->regval[2] =  DFLT_RR_QTM;
+	} else if (lvl == NIX_TXSCH_LVL_TL4) {
+		parent =  hw->txschq_list[NIX_TXSCH_LVL_TL3][0];
+		req->reg[0] = NIX_AF_TL4X_PARENT(schq);
+		req->regval[0] = parent << 16;
+		req->num_regs++;
+		req->reg[1] = NIX_AF_TL4X_SCHEDULE(schq);
+		req->regval[1] = DFLT_RR_QTM;
+	} else if (lvl == NIX_TXSCH_LVL_TL3) {
+		parent = hw->txschq_list[NIX_TXSCH_LVL_TL2][0];
+		req->reg[0] = NIX_AF_TL3X_PARENT(schq);
+		req->regval[0] = parent << 16;
+		req->num_regs++;
+		req->reg[1] = NIX_AF_TL3X_SCHEDULE(schq);
+		req->regval[1] = DFLT_RR_QTM;
+	} else if (lvl == NIX_TXSCH_LVL_TL2) {
+		parent =  hw->txschq_list[NIX_TXSCH_LVL_TL1][0];
+		req->reg[0] = NIX_AF_TL2X_PARENT(schq);
+		req->regval[0] = parent << 16;
+
+		req->num_regs++;
+		req->reg[1] = NIX_AF_TL2X_SCHEDULE(schq);
+		req->regval[1] = TXSCH_TL1_DFLT_RR_PRIO << 24 | DFLT_RR_QTM;
+
+		req->num_regs++;
+		req->reg[2] = NIX_AF_TL3_TL2X_LINKX_CFG(schq,
+							otx2_get_link(pfvf));
+		/* Enable this queue and backpressure */
+		req->regval[2] = BIT_ULL(13) | BIT_ULL(12);
+
+	} else if (lvl == NIX_TXSCH_LVL_TL1) {
+		/* Default config for TL1.
+		 * For VF this is always ignored.
+		 */
+
+		/* Set DWRR quantum */
+		req->reg[0] = NIX_AF_TL1X_SCHEDULE(schq);
+		req->regval[0] = TXSCH_TL1_DFLT_RR_QTM;
+
+		req->num_regs++;
+		req->reg[1] = NIX_AF_TL1X_TOPOLOGY(schq);
+		req->regval[1] = (TXSCH_TL1_DFLT_RR_PRIO << 1);
+
+		req->num_regs++;
+		req->reg[2] = NIX_AF_TL1X_CIR(schq);
+		req->regval[2] = 0;
+	}
+
+	return otx2_sync_mbox_msg(&pfvf->mbox);
+}
+
+int otx2_txsch_alloc(struct otx2_nic *pfvf)
+{
+	struct nix_txsch_alloc_req *req;
+	int lvl;
+
+	/* Get memory to put this msg */
+	req = otx2_mbox_alloc_msg_nix_txsch_alloc(&pfvf->mbox);
+	if (!req)
+		return -ENOMEM;
+
+	/* Request one schq per level */
+	for (lvl = 0; lvl < NIX_TXSCH_LVL_CNT; lvl++)
+		req->schq[lvl] = 1;
+
+	return otx2_sync_mbox_msg(&pfvf->mbox);
+}
+
+int otx2_txschq_stop(struct otx2_nic *pfvf)
+{
+	struct nix_txsch_free_req *free_req;
+	int lvl, schq, err;
+
+	otx2_mbox_lock(&pfvf->mbox);
+	/* Free the transmit schedulers */
+	free_req = otx2_mbox_alloc_msg_nix_txsch_free(&pfvf->mbox);
+	if (!free_req) {
+		otx2_mbox_unlock(&pfvf->mbox);
+		return -ENOMEM;
+	}
+
+	free_req->flags = TXSCHQ_FREE_ALL;
+	err = otx2_sync_mbox_msg(&pfvf->mbox);
+	otx2_mbox_unlock(&pfvf->mbox);
+
+	/* Clear the txschq list */
+	for (lvl = 0; lvl < NIX_TXSCH_LVL_CNT; lvl++) {
+		for (schq = 0; schq < MAX_TXSCHQ_PER_FUNC; schq++)
+			pfvf->hw.txschq_list[lvl][schq] = 0;
+	}
+	return err;
+}
+
+void otx2_sqb_flush(struct otx2_nic *pfvf)
+{
+	int qidx, sqe_tail, sqe_head;
+	u64 incr, *ptr, val;
+
+	ptr = (u64 *)otx2_get_regaddr(pfvf, NIX_LF_SQ_OP_STATUS);
+	for (qidx = 0; qidx < pfvf->hw.tx_queues; qidx++) {
+		incr = (u64)qidx << 32;
+		while (1) {
+			val = otx2_atomic64_add(incr, ptr);
+			sqe_head = (val >> 20) & 0x3F;
+			sqe_tail = (val >> 28) & 0x3F;
+			if (sqe_head == sqe_tail)
+				break;
+			usleep_range(1, 3);
+		}
+	}
+}
+
+/* RED and drop levels of CQ on packet reception.
+ * For CQ level is measure of emptiness ( 0x0 = full, 255 = empty).
+ */
+#define RQ_PASS_LVL_CQ(skid, qsize)	((((skid) + 16) * 256) / (qsize))
+#define RQ_DROP_LVL_CQ(skid, qsize)	(((skid) * 256) / (qsize))
+
+/* RED and drop levels of AURA for packet reception.
+ * For AURA level is measure of fullness (0x0 = empty, 255 = full).
+ * Eg: For RQ length 1K, for pass/drop level 204/230.
+ * RED accepts pkts if free pointers > 102 & <= 205.
+ * Drops pkts if free pointers < 102.
+ */
+#define RQ_PASS_LVL_AURA (255 - ((95 * 256) / 100)) /* RED when 95% is full */
+#define RQ_DROP_LVL_AURA (255 - ((99 * 256) / 100)) /* Drop when 99% is full */
+
+/* Send skid of 2000 packets required for CQ size of 4K CQEs. */
+#define SEND_CQ_SKID	2000
+
+static int otx2_rq_init(struct otx2_nic *pfvf, u16 qidx, u16 lpb_aura)
+{
+	struct otx2_qset *qset = &pfvf->qset;
+	struct nix_aq_enq_req *aq;
+
+	/* Get memory to put this msg */
+	aq = otx2_mbox_alloc_msg_nix_aq_enq(&pfvf->mbox);
+	if (!aq)
+		return -ENOMEM;
+
+	aq->rq.cq = qidx;
+	aq->rq.ena = 1;
+	aq->rq.pb_caching = 1;
+	aq->rq.lpb_aura = lpb_aura; /* Use large packet buffer aura */
+	aq->rq.lpb_sizem1 = (DMA_BUFFER_LEN(pfvf->rbsize) / 8) - 1;
+	aq->rq.xqe_imm_size = 0; /* Copying of packet to CQE not needed */
+	aq->rq.flow_tagw = 32; /* Copy full 32bit flow_tag to CQE header */
+	aq->rq.lpb_drop_ena = 1; /* Enable RED dropping for AURA */
+	aq->rq.xqe_drop_ena = 1; /* Enable RED dropping for CQ/SSO */
+	aq->rq.xqe_pass = RQ_PASS_LVL_CQ(pfvf->hw.rq_skid, qset->rqe_cnt);
+	aq->rq.xqe_drop = RQ_DROP_LVL_CQ(pfvf->hw.rq_skid, qset->rqe_cnt);
+	aq->rq.lpb_aura_pass = RQ_PASS_LVL_AURA;
+	aq->rq.lpb_aura_drop = RQ_DROP_LVL_AURA;
+
+	/* Fill AQ info */
+	aq->qidx = qidx;
+	aq->ctype = NIX_AQ_CTYPE_RQ;
+	aq->op = NIX_AQ_INSTOP_INIT;
+
+	return otx2_sync_mbox_msg(&pfvf->mbox);
+}
+
+static int otx2_sq_init(struct otx2_nic *pfvf, u16 qidx, u16 sqb_aura)
+{
+	struct otx2_qset *qset = &pfvf->qset;
+	struct otx2_snd_queue *sq;
+	struct nix_aq_enq_req *aq;
+	struct otx2_pool *pool;
+	int err;
+
+	pool = &pfvf->qset.pool[sqb_aura];
+	sq = &qset->sq[qidx];
+	sq->sqe_size = NIX_SQESZ_W16 ? 64 : 128;
+	sq->sqe_cnt = qset->sqe_cnt;
+
+	err = qmem_alloc(pfvf->dev, &sq->sqe, 1, sq->sqe_size);
+	if (err)
+		return err;
+
+	sq->sqe_base = sq->sqe->base;
+
+	sq->sqe_per_sqb = (pfvf->hw.sqb_size / sq->sqe_size) - 1;
+	sq->num_sqbs = (qset->sqe_cnt + sq->sqe_per_sqb) / sq->sqe_per_sqb;
+	sq->aura_id = sqb_aura;
+	sq->aura_fc_addr = pool->fc_addr->base;
+	sq->lmt_addr = (__force u64 *)(pfvf->reg_base + LMT_LF_LMTLINEX(qidx));
+	sq->io_addr = (__force u64)otx2_get_regaddr(pfvf, NIX_LF_OP_SENDX(0));
+
+	/* Get memory to put this msg */
+	aq = otx2_mbox_alloc_msg_nix_aq_enq(&pfvf->mbox);
+	if (!aq)
+		return -ENOMEM;
+
+	aq->sq.cq = pfvf->hw.rx_queues + qidx;
+	aq->sq.max_sqe_size = NIX_MAXSQESZ_W16; /* 128 byte */
+	aq->sq.cq_ena = 1;
+	aq->sq.ena = 1;
+	/* Only one SMQ is allocated, map all SQ's to that SMQ  */
+	aq->sq.smq = pfvf->hw.txschq_list[NIX_TXSCH_LVL_SMQ][0];
+	aq->sq.smq_rr_quantum = DFLT_RR_QTM;
+	aq->sq.default_chan = pfvf->hw.tx_chan_base;
+	aq->sq.sqe_stype = NIX_STYPE_STF; /* Cache SQB */
+	aq->sq.sqb_aura = sqb_aura;
+	/* Due pipelining impact minimum 2000 unused SQ CQE's
+	 * need to maintain to avoid CQ overflow.
+	 */
+	aq->sq.cq_limit = ((SEND_CQ_SKID * 256) / (sq->sqe_cnt));
+
+	/* Fill AQ info */
+	aq->qidx = qidx;
+	aq->ctype = NIX_AQ_CTYPE_SQ;
+	aq->op = NIX_AQ_INSTOP_INIT;
+
+	return otx2_sync_mbox_msg(&pfvf->mbox);
+}
+
+static int otx2_cq_init(struct otx2_nic *pfvf, u16 qidx)
+{
+	struct otx2_qset *qset = &pfvf->qset;
+	struct nix_aq_enq_req *aq;
+	struct otx2_cq_queue *cq;
+	int err, pool_id;
+
+	cq = &qset->cq[qidx];
+	cq->cq_idx = qidx;
+	if (qidx < pfvf->hw.rx_queues) {
+		cq->cq_type = CQ_RX;
+		cq->cqe_cnt = qset->rqe_cnt;
+	} else {
+		cq->cq_type = CQ_TX;
+		cq->cqe_cnt = qset->sqe_cnt;
+	}
+	cq->cqe_size = pfvf->qset.xqe_size;
+
+	/* Allocate memory for CQEs */
+	err = qmem_alloc(pfvf->dev, &cq->cqe, cq->cqe_cnt, cq->cqe_size);
+	if (err)
+		return err;
+
+	/* Save CQE CPU base for faster reference */
+	cq->cqe_base = cq->cqe->base;
+	/* In case where all RQs auras point to single pool,
+	 * all CQs receive buffer pool also point to same pool.
+	 */
+	pool_id = ((cq->cq_type == CQ_RX) &&
+		   (pfvf->hw.rqpool_cnt != pfvf->hw.rx_queues)) ? 0 : qidx;
+	cq->rbpool = &qset->pool[pool_id];
+
+	/* Get memory to put this msg */
+	aq = otx2_mbox_alloc_msg_nix_aq_enq(&pfvf->mbox);
+	if (!aq)
+		return -ENOMEM;
+
+	aq->cq.ena = 1;
+	aq->cq.qsize = Q_SIZE(cq->cqe_cnt, 4);
+	aq->cq.caching = 1;
+	aq->cq.base = cq->cqe->iova;
+	aq->cq.avg_level = 255;
+
+	if (qidx < pfvf->hw.rx_queues) {
+		aq->cq.drop = RQ_DROP_LVL_CQ(pfvf->hw.rq_skid, cq->cqe_cnt);
+		aq->cq.drop_ena = 1;
+	}
+
+	/* Fill AQ info */
+	aq->qidx = qidx;
+	aq->ctype = NIX_AQ_CTYPE_CQ;
+	aq->op = NIX_AQ_INSTOP_INIT;
+
+	return otx2_sync_mbox_msg(&pfvf->mbox);
+}
+
+int otx2_config_nix_queues(struct otx2_nic *pfvf)
+{
+	int qidx, err;
+
+	/* Initialize RX queues */
+	for (qidx = 0; qidx < pfvf->hw.rx_queues; qidx++) {
+		u16 lpb_aura = otx2_get_pool_idx(pfvf, AURA_NIX_RQ, qidx);
+
+		err = otx2_rq_init(pfvf, qidx, lpb_aura);
+		if (err)
+			return err;
+	}
+
+	/* Initialize TX queues */
+	for (qidx = 0; qidx < pfvf->hw.tx_queues; qidx++) {
+		u16 sqb_aura = otx2_get_pool_idx(pfvf, AURA_NIX_SQ, qidx);
+
+		err = otx2_sq_init(pfvf, qidx, sqb_aura);
+		if (err)
+			return err;
+	}
+
+	/* Initialize completion queues */
+	for (qidx = 0; qidx < pfvf->qset.cq_cnt; qidx++) {
+		err = otx2_cq_init(pfvf, qidx);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
 int otx2_config_nix(struct otx2_nic *pfvf)
 {
 	struct nix_lf_alloc_req  *nixlf;
@@ -58,6 +440,302 @@ int otx2_config_nix(struct otx2_nic *pfvf)
 	return rsp->hdr.rc;
 }
 
+void otx2_sq_free_sqbs(struct otx2_nic *pfvf)
+{
+	struct otx2_qset *qset = &pfvf->qset;
+	struct otx2_hw *hw = &pfvf->hw;
+	struct otx2_snd_queue *sq;
+	int sqb, qidx;
+	u64 iova, pa;
+
+	for (qidx = 0; qidx < hw->tx_queues; qidx++) {
+		sq = &qset->sq[qidx];
+		if (!sq->sqb_ptrs)
+			continue;
+		for (sqb = 0; sqb < sq->sqb_count; sqb++) {
+			if (!sq->sqb_ptrs[sqb])
+				continue;
+			iova = sq->sqb_ptrs[sqb];
+			pa = otx2_iova_to_phys(pfvf->iommu_domain, iova);
+			dma_unmap_page_attrs(pfvf->dev, iova, hw->sqb_size,
+					     DMA_FROM_DEVICE,
+					     DMA_ATTR_SKIP_CPU_SYNC);
+			put_page(virt_to_page(phys_to_virt(pa)));
+		}
+		sq->sqb_count = 0;
+	}
+}
+
+void otx2_free_aura_ptr(struct otx2_nic *pfvf, int type)
+{
+	int pool_id, pool_start = 0, pool_end = 0, size = 0;
+	u64 iova, pa;
+
+	if (type == AURA_NIX_SQ) {
+		pool_start = otx2_get_pool_idx(pfvf, type, 0);
+		pool_end =  pool_start + pfvf->hw.sqpool_cnt;
+		size = pfvf->hw.sqb_size;
+	}
+	if (type == AURA_NIX_RQ) {
+		pool_start = otx2_get_pool_idx(pfvf, type, 0);
+		pool_end = pfvf->hw.rqpool_cnt;
+		size = pfvf->rbsize;
+	}
+
+	/* Free SQB and RQB pointers from the aura pool */
+	for (pool_id = pool_start; pool_id < pool_end; pool_id++) {
+		iova = otx2_aura_allocptr(pfvf, pool_id);
+		while (iova) {
+			if (type == AURA_NIX_RQ)
+				iova -= OTX2_HEAD_ROOM;
+
+			pa = otx2_iova_to_phys(pfvf->iommu_domain, iova);
+			dma_unmap_page_attrs(pfvf->dev, iova, size,
+					     DMA_FROM_DEVICE,
+					     DMA_ATTR_SKIP_CPU_SYNC);
+			put_page(virt_to_page(phys_to_virt(pa)));
+			iova = otx2_aura_allocptr(pfvf, pool_id);
+		}
+	}
+}
+
+void otx2_aura_pool_free(struct otx2_nic *pfvf)
+{
+	struct otx2_pool *pool;
+	int pool_id;
+
+	if (!pfvf->qset.pool)
+		return;
+
+	for (pool_id = 0; pool_id < pfvf->hw.pool_cnt; pool_id++) {
+		pool = &pfvf->qset.pool[pool_id];
+		qmem_free(pfvf->dev, pool->stack);
+		qmem_free(pfvf->dev, pool->fc_addr);
+	}
+	devm_kfree(pfvf->dev, pfvf->qset.pool);
+}
+
+static int otx2_aura_init(struct otx2_nic *pfvf, int aura_id,
+			  int pool_id, int numptrs)
+{
+	struct npa_aq_enq_req *aq;
+	struct otx2_pool *pool;
+	int err;
+
+	pool = &pfvf->qset.pool[pool_id];
+
+	/* Allocate memory for HW to update Aura count.
+	 * Alloc one cache line, so that it fits all FC_STYPE modes.
+	 */
+	if (!pool->fc_addr) {
+		err = qmem_alloc(pfvf->dev, &pool->fc_addr, 1, OTX2_ALIGN);
+		if (err)
+			return err;
+	}
+
+	/* Initialize this aura's context via AF */
+	aq = otx2_mbox_alloc_msg_npa_aq_enq(&pfvf->mbox);
+	if (!aq) {
+		/* Shared mbox memory buffer is full, flush it and retry */
+		err = otx2_sync_mbox_msg(&pfvf->mbox);
+		if (err)
+			return err;
+		aq = otx2_mbox_alloc_msg_npa_aq_enq(&pfvf->mbox);
+		if (!aq)
+			return -ENOMEM;
+	}
+
+	aq->aura_id = aura_id;
+	/* Will be filled by AF with correct pool context address */
+	aq->aura.pool_addr = pool_id;
+	aq->aura.pool_caching = 1;
+	aq->aura.shift = ilog2(numptrs) - 8;
+	aq->aura.count = numptrs;
+	aq->aura.limit = numptrs;
+	aq->aura.avg_level = 255;
+	aq->aura.ena = 1;
+	aq->aura.fc_ena = 1;
+	aq->aura.fc_addr = pool->fc_addr->iova;
+	aq->aura.fc_hyst_bits = 0; /* Store count on all updates */
+
+	/* Fill AQ info */
+	aq->ctype = NPA_AQ_CTYPE_AURA;
+	aq->op = NPA_AQ_INSTOP_INIT;
+
+	return 0;
+}
+
+static int otx2_pool_init(struct otx2_nic *pfvf, u16 pool_id,
+			  int stack_pages, int numptrs, int buf_size)
+{
+	struct npa_aq_enq_req *aq;
+	struct otx2_pool *pool;
+	int err;
+
+	pool = &pfvf->qset.pool[pool_id];
+	/* Alloc memory for stack which is used to store buffer pointers */
+	err = qmem_alloc(pfvf->dev, &pool->stack,
+			 stack_pages, pfvf->hw.stack_pg_bytes);
+	if (err)
+		return err;
+
+	pool->rbsize = buf_size;
+	pool->rbpage_order = get_order(buf_size);
+
+	/* Initialize this pool's context via AF */
+	aq = otx2_mbox_alloc_msg_npa_aq_enq(&pfvf->mbox);
+	if (!aq) {
+		/* Shared mbox memory buffer is full, flush it and retry */
+		err = otx2_sync_mbox_msg(&pfvf->mbox);
+		if (err) {
+			qmem_free(pfvf->dev, pool->stack);
+			return err;
+		}
+		aq = otx2_mbox_alloc_msg_npa_aq_enq(&pfvf->mbox);
+		if (!aq) {
+			qmem_free(pfvf->dev, pool->stack);
+			return -ENOMEM;
+		}
+	}
+
+	aq->aura_id = pool_id;
+	aq->pool.stack_base = pool->stack->iova;
+	aq->pool.stack_caching = 1;
+	aq->pool.ena = 1;
+	aq->pool.buf_size = buf_size / 128;
+	aq->pool.stack_max_pages = stack_pages;
+	aq->pool.shift = ilog2(numptrs) - 8;
+	aq->pool.ptr_start = 0;
+	aq->pool.ptr_end = ~0ULL;
+
+	/* Fill AQ info */
+	aq->ctype = NPA_AQ_CTYPE_POOL;
+	aq->op = NPA_AQ_INSTOP_INIT;
+
+	return 0;
+}
+
+int otx2_sq_aura_pool_init(struct otx2_nic *pfvf)
+{
+	int qidx, pool_id, stack_pages, num_sqbs;
+	struct otx2_qset *qset = &pfvf->qset;
+	struct otx2_hw *hw = &pfvf->hw;
+	struct otx2_snd_queue *sq;
+	struct otx2_pool *pool;
+	int err, ptr;
+	s64 bufptr;
+
+	/* Calculate number of SQBs needed.
+	 *
+	 * For a 128byte SQE, and 4K size SQB, 31 SQEs will fit in one SQB.
+	 * Last SQE is used for pointing to next SQB.
+	 */
+	num_sqbs = (hw->sqb_size / 128) - 1;
+	num_sqbs = (qset->sqe_cnt + num_sqbs) / num_sqbs;
+
+	/* Get no of stack pages needed */
+	stack_pages =
+		(num_sqbs + hw->stack_pg_ptrs - 1) / hw->stack_pg_ptrs;
+
+	for (qidx = 0; qidx < hw->tx_queues; qidx++) {
+		pool_id = otx2_get_pool_idx(pfvf, AURA_NIX_SQ, qidx);
+		/* Initialize aura context */
+		err = otx2_aura_init(pfvf, pool_id, pool_id, num_sqbs);
+		if (err)
+			goto fail;
+
+		/* Initialize pool context */
+		err = otx2_pool_init(pfvf, pool_id, stack_pages,
+				     num_sqbs, hw->sqb_size);
+		if (err)
+			goto fail;
+	}
+
+	/* Flush accumulated messages */
+	err = otx2_sync_mbox_msg(&pfvf->mbox);
+	if (err)
+		goto fail;
+
+	/* Allocate pointers and free them to aura/pool */
+	for (qidx = 0; qidx < hw->tx_queues; qidx++) {
+		pool_id = otx2_get_pool_idx(pfvf, AURA_NIX_SQ, qidx);
+		pool = &pfvf->qset.pool[pool_id];
+
+		sq = &qset->sq[qidx];
+		sq->sqb_count = 0;
+		sq->sqb_ptrs = kcalloc(num_sqbs, sizeof(u64 *), GFP_KERNEL);
+		if (!sq->sqb_ptrs)
+			return -ENOMEM;
+
+		for (ptr = 0; ptr < num_sqbs; ptr++) {
+			bufptr = otx2_alloc_rbuf(pfvf, pool, GFP_KERNEL);
+			if (bufptr <= 0)
+				return bufptr;
+			otx2_aura_freeptr(pfvf, pool_id, bufptr);
+			sq->sqb_ptrs[sq->sqb_count++] = (u64)bufptr;
+		}
+		otx2_get_page(pool);
+	}
+
+	return 0;
+fail:
+	otx2_mbox_reset(&pfvf->mbox.mbox, 0);
+	otx2_aura_pool_free(pfvf);
+	return err;
+}
+
+int otx2_rq_aura_pool_init(struct otx2_nic *pfvf)
+{
+	struct otx2_hw *hw = &pfvf->hw;
+	int stack_pages, pool_id, rq;
+	struct otx2_pool *pool;
+	int err, ptr, num_ptrs;
+	s64 bufptr;
+
+	num_ptrs = pfvf->qset.rqe_cnt;
+
+	stack_pages =
+		(num_ptrs + hw->stack_pg_ptrs - 1) / hw->stack_pg_ptrs;
+
+	for (rq = 0; rq < hw->rx_queues; rq++) {
+		pool_id = otx2_get_pool_idx(pfvf, AURA_NIX_RQ, rq);
+		/* Initialize aura context */
+		err = otx2_aura_init(pfvf, pool_id, pool_id, num_ptrs);
+		if (err)
+			goto fail;
+	}
+	for (pool_id = 0; pool_id < hw->rqpool_cnt; pool_id++) {
+		err = otx2_pool_init(pfvf, pool_id, stack_pages,
+				     num_ptrs, pfvf->rbsize);
+		if (err)
+			goto fail;
+	}
+
+	/* Flush accumulated messages */
+	err = otx2_sync_mbox_msg(&pfvf->mbox);
+	if (err)
+		goto fail;
+
+	/* Allocate pointers and free them to aura/pool */
+	for (pool_id = 0; pool_id < hw->rqpool_cnt; pool_id++) {
+		pool = &pfvf->qset.pool[pool_id];
+		for (ptr = 0; ptr < num_ptrs; ptr++) {
+			bufptr = otx2_alloc_rbuf(pfvf, pool, GFP_KERNEL);
+			if (bufptr <= 0)
+				return bufptr;
+			otx2_aura_freeptr(pfvf, pool_id,
+					  bufptr + OTX2_HEAD_ROOM);
+		}
+		otx2_get_page(pool);
+	}
+
+	return 0;
+fail:
+	otx2_mbox_reset(&pfvf->mbox.mbox, 0);
+	otx2_aura_pool_free(pfvf);
+	return err;
+}
+
 int otx2_config_npa(struct otx2_nic *pfvf)
 {
 	struct otx2_qset *qset = &pfvf->qset;
@@ -134,6 +812,14 @@ int otx2_attach_npa_nix(struct otx2_nic *pfvf)
 		return err;
 	}
 
+	pfvf->nix_blkaddr = BLKADDR_NIX0;
+
+	/* If the platform has two NIX blocks then LF may be
+	 * allocated from NIX1.
+	 */
+	if (otx2_read64(pfvf, RVU_PF_BLOCK_ADDRX_DISC(BLKADDR_NIX1)) & 0x1FFULL)
+		pfvf->nix_blkaddr = BLKADDR_NIX1;
+
 	/* Get NPA and NIX MSIX vector offsets */
 	msix = otx2_mbox_alloc_msg_msix_offset(&pfvf->mbox);
 	if (!msix) {
@@ -158,6 +844,43 @@ int otx2_attach_npa_nix(struct otx2_nic *pfvf)
 	return 0;
 }
 
+void otx2_ctx_disable(struct mbox *mbox, int type, bool npa)
+{
+	struct hwctx_disable_req *req;
+
+	otx2_mbox_lock(mbox);
+	/* Request AQ to disable this context */
+	if (npa)
+		req = otx2_mbox_alloc_msg_npa_hwctx_disable(mbox);
+	else
+		req = otx2_mbox_alloc_msg_nix_hwctx_disable(mbox);
+
+	if (!req) {
+		otx2_mbox_unlock(mbox);
+		return;
+	}
+
+	req->ctype = type;
+
+	if (otx2_sync_mbox_msg(mbox))
+		dev_err(mbox->pfvf->dev, "%s failed to disable context\n",
+			__func__);
+
+	otx2_mbox_unlock(mbox);
+}
+
+void mbox_handler_nix_txsch_alloc(struct otx2_nic *pf,
+				  struct nix_txsch_alloc_rsp *rsp)
+{
+	int lvl, schq;
+
+	/* Setup transmit scheduler list */
+	for (lvl = 0; lvl < NIX_TXSCH_LVL_CNT; lvl++)
+		for (schq = 0; schq < rsp->schq[lvl]; schq++)
+			pf->hw.txschq_list[lvl][schq] =
+				rsp->schq_list[lvl][schq];
+}
+
 /* Mbox message handlers */
 void mbox_handler_npa_lf_alloc(struct otx2_nic *pfvf,
 			       struct npa_lf_alloc_rsp *rsp)
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
index cdb1c567d5dc..a4f1c604cb60 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
@@ -12,9 +12,11 @@
 #define OTX2_COMMON_H
 
 #include <linux/pci.h>
+#include <linux/iommu.h>
 
 #include <mbox.h>
 #include "otx2_reg.h"
+#include "otx2_txrx.h"
 
 /* PCI device IDs */
 #define PCI_DEVID_OCTEONTX2_RVU_PF              0xA063
@@ -25,15 +27,9 @@
 
 #define NAME_SIZE                               32
 
-struct otx2_pool {
-	struct qmem		*stack;
-};
-
-struct otx2_qset {
-#define OTX2_MAX_CQ_CNT		64
-	u16			cq_cnt;
-	u16			xqe_size; /* Size of CQE i.e 128 or 512 bytes */
-	struct otx2_pool	*pool;
+enum arua_mapped_qtypes {
+	AURA_NIX_RQ,
+	AURA_NIX_SQ,
 };
 
 struct mbox {
@@ -54,14 +50,21 @@ struct otx2_hw {
 	u16                     tx_queues;
 	u16			max_queues;
 	u16			pool_cnt;
+	u16			rqpool_cnt;
+	u16			sqpool_cnt;
 
 	/* NPA */
 	u32			stack_pg_ptrs;  /* No of ptrs per stack page */
 	u32			stack_pg_bytes; /* Size of stack page */
 	u16			sqb_size;
 
+	/* NIX */
+	u16		txschq_list[NIX_TXSCH_LVL_CNT][MAX_TXSCHQ_PER_FUNC];
+
+	/* HW settings, coalescing etc */
 	u16			rx_chan_base;
 	u16			tx_chan_base;
+	u16			rq_skid;
 
 	/* MSI-X */
 	u16			npa_msixoff; /* Offset of NPA vectors */
@@ -73,6 +76,8 @@ struct otx2_hw {
 struct otx2_nic {
 	void __iomem		*reg_base;
 	struct net_device	*netdev;
+	void			*iommu_domain;
+	u16			rbsize; /* Receive buffer size */
 
 	struct otx2_qset	qset;
 	struct otx2_hw		hw;
@@ -84,6 +89,9 @@ struct otx2_nic {
 	struct workqueue_struct *mbox_wq;
 
 	u16			pcifunc; /* RVU PF_FUNC */
+
+	/* Block address of NIX either BLKADDR_NIX0 or BLKADDR_NIX1 */
+	int			nix_blkaddr;
 };
 
 /* Register read/write APIs */
@@ -93,7 +101,7 @@ static inline void __iomem *otx2_get_regaddr(struct otx2_nic *nic, u64 offset)
 
 	switch ((offset >> RVU_FUNC_BLKADDR_SHIFT) & RVU_FUNC_BLKADDR_MASK) {
 	case BLKTYPE_NIX:
-		blkaddr = BLKADDR_NIX0;
+		blkaddr = nic->nix_blkaddr;
 		break;
 	case BLKTYPE_NPA:
 		blkaddr = BLKADDR_NPA;
@@ -184,6 +192,72 @@ static inline void otx2_mbox_unlock(struct mbox *mbox)
 	mutex_unlock(&mbox->lock);
 }
 
+/* With the absence of API for 128-bit IO memory access for arm64,
+ * implement required operations at place.
+ */
+#if defined(CONFIG_ARM64)
+static inline void otx2_write128(u64 lo, u64 hi, void __iomem *addr)
+{
+	__asm__ volatile("stp %x[x0], %x[x1], [%x[p1],#0]!"
+			 ::[x0]"r"(lo), [x1]"r"(hi), [p1]"r"(addr));
+}
+
+static inline u64 otx2_atomic64_add(u64 incr, u64 *ptr)
+{
+	u64 result;
+
+	__asm__ volatile(".cpu   generic+lse\n"
+			 "ldadd %x[i], %x[r], [%[b]]"
+			 : [r]"=r"(result), "+m"(*ptr)
+			 : [i]"r"(incr), [b]"r"(ptr)
+			 : "memory");
+	return result;
+}
+
+#else
+#define otx2_write128(lo, hi, addr)
+#define otx2_atomic64_add(incr, ptr)		({ *ptr += incr; })
+#endif
+
+/* Alloc pointer from pool/aura */
+static inline u64 otx2_aura_allocptr(struct otx2_nic *pfvf, int aura)
+{
+	u64 *ptr = (u64 *)otx2_get_regaddr(pfvf,
+			   NPA_LF_AURA_OP_ALLOCX(0));
+	u64 incr = (u64)aura | BIT_ULL(63);
+
+	return otx2_atomic64_add(incr, ptr);
+}
+
+/* Free pointer to a pool/aura */
+static inline void otx2_aura_freeptr(struct otx2_nic *pfvf,
+				     int aura, s64 buf)
+{
+	otx2_write128((u64)buf, (u64)aura | BIT_ULL(63),
+		      otx2_get_regaddr(pfvf, NPA_LF_AURA_OP_FREE0));
+}
+
+/* Update page ref count */
+static inline void otx2_get_page(struct otx2_pool *pool)
+{
+	if (!pool->page)
+		return;
+
+	if (pool->pageref)
+		page_ref_add(pool->page, pool->pageref);
+	pool->pageref = 0;
+	pool->page = NULL;
+}
+
+static inline int otx2_get_pool_idx(struct otx2_nic *pfvf, int type, int idx)
+{
+	if (type == AURA_NIX_SQ)
+		return pfvf->hw.rqpool_cnt + idx;
+
+	 /* AURA_NIX_RQ */
+	return idx;
+}
+
 /* Mbox APIs */
 static inline int otx2_sync_mbox_msg(struct mbox *mbox)
 {
@@ -263,11 +337,46 @@ MBOX_UP_CGX_MESSAGES
 #define	RVU_PFVF_FUNC_SHIFT	0
 #define	RVU_PFVF_FUNC_MASK	0x3FF
 
+static inline dma_addr_t otx2_dma_map_page(struct otx2_nic *pfvf,
+					   struct page *page,
+					   size_t offset, size_t size,
+					   enum dma_data_direction dir)
+{
+	dma_addr_t iova;
+
+	iova = dma_map_page_attrs(pfvf->dev, page,
+				  offset, size, dir, DMA_ATTR_SKIP_CPU_SYNC);
+	if (unlikely(dma_mapping_error(pfvf->dev, iova)))
+		return (dma_addr_t)NULL;
+	return iova;
+}
+
+static inline void otx2_dma_unmap_page(struct otx2_nic *pfvf,
+				       dma_addr_t addr, size_t size,
+				       enum dma_data_direction dir)
+{
+	dma_unmap_page_attrs(pfvf->dev, addr, size,
+			     dir, DMA_ATTR_SKIP_CPU_SYNC);
+}
+
 /* RVU block related APIs */
 int otx2_attach_npa_nix(struct otx2_nic *pfvf);
 int otx2_detach_resources(struct mbox *mbox);
 int otx2_config_npa(struct otx2_nic *pfvf);
+int otx2_sq_aura_pool_init(struct otx2_nic *pfvf);
+int otx2_rq_aura_pool_init(struct otx2_nic *pfvf);
+void otx2_aura_pool_free(struct otx2_nic *pfvf);
+void otx2_free_aura_ptr(struct otx2_nic *pfvf, int type);
+void otx2_sq_free_sqbs(struct otx2_nic *pfvf);
 int otx2_config_nix(struct otx2_nic *pfvf);
+int otx2_config_nix_queues(struct otx2_nic *pfvf);
+int otx2_txschq_config(struct otx2_nic *pfvf, int lvl);
+int otx2_txsch_alloc(struct otx2_nic *pfvf);
+int otx2_txschq_stop(struct otx2_nic *pfvf);
+void otx2_sqb_flush(struct otx2_nic *pfvf);
+dma_addr_t otx2_alloc_rbuf(struct otx2_nic *pfvf, struct otx2_pool *pool,
+			   gfp_t gfp);
+void otx2_ctx_disable(struct mbox *mbox, int type, bool npa);
 
 /* Mbox handlers */
 void mbox_handler_msix_offset(struct otx2_nic *pfvf,
@@ -276,4 +385,6 @@ void mbox_handler_npa_lf_alloc(struct otx2_nic *pfvf,
 			       struct npa_lf_alloc_rsp *rsp);
 void mbox_handler_nix_lf_alloc(struct otx2_nic *pfvf,
 			       struct nix_lf_alloc_rsp *rsp);
+void mbox_handler_nix_txsch_alloc(struct otx2_nic *pf,
+				  struct nix_txsch_alloc_rsp *rsp);
 #endif /* OTX2_COMMON_H */
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
index ef5dba442abf..735188900de4 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
@@ -17,7 +17,10 @@
 #include <linux/iommu.h>
 #include <net/ip.h>
 
+#include "otx2_reg.h"
 #include "otx2_common.h"
+#include "otx2_txrx.h"
+#include "otx2_struct.h"
 
 #define DRV_NAME	"octeontx2-nicpf"
 #define DRV_STRING	"Marvell OcteonTX2 NIC Physical Function Driver"
@@ -123,6 +126,10 @@ static void otx2_process_pfaf_mbox_msg(struct otx2_nic *pf,
 	case MBOX_MSG_NIX_LF_ALLOC:
 		mbox_handler_nix_lf_alloc(pf, (struct nix_lf_alloc_rsp *)msg);
 		break;
+	case MBOX_MSG_NIX_TXSCH_ALLOC:
+		mbox_handler_nix_txsch_alloc(pf,
+					     (struct nix_txsch_alloc_rsp *)msg);
+		break;
 	default:
 		if (msg->rc)
 			dev_err(pf->dev,
@@ -379,26 +386,231 @@ static int otx2_set_real_num_queues(struct net_device *netdev,
 	return err;
 }
 
+static void otx2_free_cq_res(struct otx2_nic *pf)
+{
+	struct otx2_qset *qset = &pf->qset;
+	struct otx2_cq_queue *cq;
+	int qidx;
+
+	/* Disable CQs */
+	otx2_ctx_disable(&pf->mbox, NIX_AQ_CTYPE_CQ, false);
+	for (qidx = 0; qidx < qset->cq_cnt; qidx++) {
+		cq = &qset->cq[qidx];
+		qmem_free(pf->dev, cq->cqe);
+	}
+}
+
+static void otx2_free_sq_res(struct otx2_nic *pf)
+{
+	struct otx2_qset *qset = &pf->qset;
+	struct otx2_snd_queue *sq;
+	int qidx;
+
+	/* Disable SQs */
+	otx2_ctx_disable(&pf->mbox, NIX_AQ_CTYPE_SQ, false);
+	/* Free SQB pointers */
+	otx2_sq_free_sqbs(pf);
+	for (qidx = 0; qidx < pf->hw.tx_queues; qidx++) {
+		sq = &qset->sq[qidx];
+		qmem_free(pf->dev, sq->sqe);
+		kfree(sq->sqb_ptrs);
+	}
+}
+
+static int otx2_init_hw_resources(struct otx2_nic *pf)
+{
+	struct mbox *mbox = &pf->mbox;
+	struct otx2_hw *hw = &pf->hw;
+	struct msg_req *req;
+	int err = 0, lvl;
+
+	/* Set required NPA LF's pool counts
+	 * Auras and Pools are used in a 1:1 mapping,
+	 * so, aura count = pool count.
+	 */
+	hw->rqpool_cnt = hw->rx_queues;
+	hw->sqpool_cnt = hw->tx_queues;
+	hw->pool_cnt = hw->rqpool_cnt + hw->sqpool_cnt;
+
+	/* Get the size of receive buffers to allocate */
+	pf->rbsize = RCV_FRAG_LEN(pf->netdev->mtu);
+
+	otx2_mbox_lock(mbox);
+	/* NPA init */
+	err = otx2_config_npa(pf);
+	if (err)
+		goto exit;
+
+	/* NIX init */
+	err = otx2_config_nix(pf);
+	if (err)
+		goto err_free_npa_lf;
+
+	/* Init Auras and pools used by NIX RQ, for free buffer ptrs */
+	err = otx2_rq_aura_pool_init(pf);
+	if (err) {
+		otx2_mbox_unlock(mbox);
+		goto err_free_nix_lf;
+	}
+	/* Init Auras and pools used by NIX SQ, for queueing SQEs */
+	err = otx2_sq_aura_pool_init(pf);
+	if (err) {
+		otx2_mbox_unlock(mbox);
+		goto err_free_rq_ptrs;
+	}
+
+	err = otx2_txsch_alloc(pf);
+	if (err) {
+		otx2_mbox_unlock(mbox);
+		goto err_free_sq_ptrs;
+	}
+
+	err = otx2_config_nix_queues(pf);
+	if (err) {
+		otx2_mbox_unlock(mbox);
+		goto err_free_txsch;
+	}
+	for (lvl = 0; lvl < NIX_TXSCH_LVL_CNT; lvl++) {
+		err = otx2_txschq_config(pf, lvl);
+		if (err) {
+			otx2_mbox_unlock(mbox);
+			goto err_free_nix_queues;
+		}
+	}
+	otx2_mbox_unlock(mbox);
+	return err;
+
+err_free_nix_queues:
+	otx2_free_sq_res(pf);
+	otx2_free_cq_res(pf);
+	otx2_ctx_disable(mbox, NIX_AQ_CTYPE_RQ, false);
+err_free_txsch:
+	if (otx2_txschq_stop(pf))
+		dev_err(pf->dev, "%s failed to stop TX schedulers\n", __func__);
+err_free_sq_ptrs:
+	otx2_sq_free_sqbs(pf);
+err_free_rq_ptrs:
+	otx2_free_aura_ptr(pf, AURA_NIX_RQ);
+	otx2_ctx_disable(mbox, NPA_AQ_CTYPE_POOL, true);
+	otx2_ctx_disable(mbox, NPA_AQ_CTYPE_AURA, true);
+	otx2_aura_pool_free(pf);
+err_free_nix_lf:
+	otx2_mbox_lock(mbox);
+	req = otx2_mbox_alloc_msg_nix_lf_free(mbox);
+	if (req) {
+		if (otx2_sync_mbox_msg(mbox))
+			dev_err(pf->dev, "%s failed to free nixlf\n", __func__);
+	}
+err_free_npa_lf:
+	/* Reset NPA LF */
+	req = otx2_mbox_alloc_msg_npa_lf_free(mbox);
+	if (req) {
+		if (otx2_sync_mbox_msg(mbox))
+			dev_err(pf->dev, "%s failed to free npalf\n", __func__);
+	}
+exit:
+	otx2_mbox_unlock(mbox);
+	return err;
+}
+
+static void otx2_free_hw_resources(struct otx2_nic *pf)
+{
+	struct mbox *mbox = &pf->mbox;
+	struct msg_req *req;
+	int err;
+
+	/* Ensure all SQE are processed */
+	otx2_sqb_flush(pf);
+
+	/* Stop transmission */
+	err = otx2_txschq_stop(pf);
+	if (err)
+		dev_err(pf->dev, "RVUPF: Failed to stop/free TX schedulers\n");
+
+	/* Disable RQs */
+	otx2_ctx_disable(mbox, NIX_AQ_CTYPE_RQ, false);
+
+	otx2_free_sq_res(pf);
+
+	/* Free RQ buffer pointers*/
+	otx2_free_aura_ptr(pf, AURA_NIX_RQ);
+
+	otx2_free_cq_res(pf);
+
+	otx2_mbox_lock(mbox);
+	/* Reset NIX LF */
+	req = otx2_mbox_alloc_msg_nix_lf_free(mbox);
+	if (req) {
+		if (otx2_sync_mbox_msg(mbox))
+			dev_err(pf->dev, "%s failed to free nixlf\n", __func__);
+	}
+	otx2_mbox_unlock(mbox);
+
+	/* Disable NPA Pool and Aura hw context */
+	otx2_ctx_disable(mbox, NPA_AQ_CTYPE_POOL, true);
+	otx2_ctx_disable(mbox, NPA_AQ_CTYPE_AURA, true);
+	otx2_aura_pool_free(pf);
+
+	otx2_mbox_lock(mbox);
+	/* Reset NPA LF */
+	req = otx2_mbox_alloc_msg_npa_lf_free(mbox);
+	if (req) {
+		if (otx2_sync_mbox_msg(mbox))
+			dev_err(pf->dev, "%s failed to free npalf\n", __func__);
+	}
+	otx2_mbox_unlock(mbox);
+}
+
 static int otx2_open(struct net_device *netdev)
 {
 	struct otx2_nic *pf = netdev_priv(netdev);
+	struct otx2_qset *qset = &pf->qset;
 	int err = 0;
 
 	netif_carrier_off(netdev);
 
 	pf->qset.cq_cnt = pf->hw.rx_queues + pf->hw.tx_queues;
 
-	/* NPA init */
-	err = otx2_config_npa(pf);
+	/* CQ size of RQ */
+	qset->rqe_cnt = qset->rqe_cnt ? qset->rqe_cnt : Q_COUNT(Q_SIZE_256);
+	/* CQ size of SQ */
+	qset->sqe_cnt = qset->sqe_cnt ? qset->sqe_cnt : Q_COUNT(Q_SIZE_4K);
+
+	err = -ENOMEM;
+	qset->cq = kcalloc(pf->qset.cq_cnt,
+			   sizeof(struct otx2_cq_queue), GFP_KERNEL);
+	if (!qset->cq)
+		goto err_free_mem;
+
+	qset->sq = kcalloc(pf->hw.tx_queues,
+			   sizeof(struct otx2_snd_queue), GFP_KERNEL);
+	if (!qset->sq)
+		goto err_free_mem;
+
+	err = otx2_init_hw_resources(pf);
 	if (err)
-		return err;
+		goto err_free_mem;
 
-	/* NIX init */
-	return otx2_config_nix(pf);
+	return 0;
+err_free_mem:
+	kfree(qset->sq);
+	kfree(qset->cq);
+	return err;
 }
 
 static int otx2_stop(struct net_device *netdev)
 {
+	struct otx2_nic *pf = netdev_priv(netdev);
+	struct otx2_qset *qset = &pf->qset;
+
+	otx2_free_hw_resources(pf);
+
+	kfree(qset->sq);
+	kfree(qset->cq);
+
+	/* Do not clear RQ/SQ ringsize settings */
+	memset((void *)qset + offsetof(struct otx2_qset, sqe_cnt), 0,
+	       sizeof(*qset) - offsetof(struct otx2_qset, sqe_cnt));
 	return 0;
 }
 
@@ -557,6 +769,19 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	if (err)
 		goto err_detach_rsrc;
 
+	/* NPA's pool is a stack to which SW frees buffer pointers via Aura.
+	 * HW allocates buffer pointer from stack and uses it for DMA'ing
+	 * ingress packet. In some scenarios HW can free back allocated buffer
+	 * pointers to pool. This makes it impossible for SW to maintain a
+	 * parallel list where physical addresses of buffer pointers (IOVAs)
+	 * given to HW can be saved for later reference.
+	 *
+	 * So the only way to convert Rx packet's buffer address is to use
+	 * IOMMU's iova_to_phys() handler which translates the address by
+	 * walking through the translation tables.
+	 */
+	pf->iommu_domain = iommu_get_domain_for_dev(dev);
+
 	netdev->netdev_ops = &otx2_netdev_ops;
 
 	err = register_netdev(netdev);
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_reg.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_reg.h
index d0bd64ab4f1a..7963d418886a 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_reg.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_reg.h
@@ -48,4 +48,100 @@
 #define RVU_FUNC_BLKADDR_SHIFT		20
 #define RVU_FUNC_BLKADDR_MASK		0x1FULL
 
+/* NPA LF registers */
+#define NPA_LFBASE			(BLKTYPE_NPA << RVU_FUNC_BLKADDR_SHIFT)
+#define NPA_LF_AURA_OP_ALLOCX(a)	(NPA_LFBASE | 0x10 | (a) << 3)
+#define NPA_LF_AURA_OP_FREE0            (NPA_LFBASE | 0x20)
+#define NPA_LF_AURA_OP_FREE1            (NPA_LFBASE | 0x28)
+#define NPA_LF_AURA_OP_CNT              (NPA_LFBASE | 0x30)
+#define NPA_LF_AURA_OP_LIMIT            (NPA_LFBASE | 0x50)
+#define NPA_LF_AURA_OP_INT              (NPA_LFBASE | 0x60)
+#define NPA_LF_AURA_OP_THRESH           (NPA_LFBASE | 0x70)
+#define NPA_LF_POOL_OP_PC               (NPA_LFBASE | 0x100)
+#define NPA_LF_POOL_OP_AVAILABLE        (NPA_LFBASE | 0x110)
+#define NPA_LF_POOL_OP_PTR_START0       (NPA_LFBASE | 0x120)
+#define NPA_LF_POOL_OP_PTR_START1       (NPA_LFBASE | 0x128)
+#define NPA_LF_POOL_OP_PTR_END0         (NPA_LFBASE | 0x130)
+#define NPA_LF_POOL_OP_PTR_END1         (NPA_LFBASE | 0x138)
+#define NPA_LF_POOL_OP_INT              (NPA_LFBASE | 0x160)
+#define NPA_LF_POOL_OP_THRESH           (NPA_LFBASE | 0x170)
+#define NPA_LF_ERR_INT                  (NPA_LFBASE | 0x200)
+#define NPA_LF_ERR_INT_W1S              (NPA_LFBASE | 0x208)
+#define NPA_LF_ERR_INT_ENA_W1C          (NPA_LFBASE | 0x210)
+#define NPA_LF_ERR_INT_ENA_W1S          (NPA_LFBASE | 0x218)
+#define NPA_LF_RAS                      (NPA_LFBASE | 0x220)
+#define NPA_LF_RAS_W1S                  (NPA_LFBASE | 0x228)
+#define NPA_LF_RAS_ENA_W1C              (NPA_LFBASE | 0x230)
+#define NPA_LF_RAS_ENA_W1S              (NPA_LFBASE | 0x238)
+#define NPA_LF_QINTX_CNT(a)             (NPA_LFBASE | 0x300 | (a) << 12)
+#define NPA_LF_QINTX_INT(a)             (NPA_LFBASE | 0x310 | (a) << 12)
+#define NPA_LF_QINTX_INT_W1S(a)         (NPA_LFBASE | 0x318 | (a) << 12)
+#define NPA_LF_QINTX_ENA_W1S(a)         (NPA_LFBASE | 0x320 | (a) << 12)
+#define NPA_LF_QINTX_ENA_W1C(a)         (NPA_LFBASE | 0x330 | (a) << 12)
+
+/* NIX LF registers */
+#define	NIX_LFBASE			(BLKTYPE_NIX << RVU_FUNC_BLKADDR_SHIFT)
+#define	NIX_LF_RX_SECRETX(a)		(NIX_LFBASE | 0x0 | (a) << 3)
+#define	NIX_LF_CFG			(NIX_LFBASE | 0x100)
+#define	NIX_LF_GINT			(NIX_LFBASE | 0x200)
+#define	NIX_LF_GINT_W1S			(NIX_LFBASE | 0x208)
+#define	NIX_LF_GINT_ENA_W1C		(NIX_LFBASE | 0x210)
+#define	NIX_LF_GINT_ENA_W1S		(NIX_LFBASE | 0x218)
+#define	NIX_LF_ERR_INT			(NIX_LFBASE | 0x220)
+#define	NIX_LF_ERR_INT_W1S		(NIX_LFBASE | 0x228)
+#define	NIX_LF_ERR_INT_ENA_W1C		(NIX_LFBASE | 0x230)
+#define	NIX_LF_ERR_INT_ENA_W1S		(NIX_LFBASE | 0x238)
+#define	NIX_LF_RAS			(NIX_LFBASE | 0x240)
+#define	NIX_LF_RAS_W1S			(NIX_LFBASE | 0x248)
+#define	NIX_LF_RAS_ENA_W1C		(NIX_LFBASE | 0x250)
+#define	NIX_LF_RAS_ENA_W1S		(NIX_LFBASE | 0x258)
+#define	NIX_LF_SQ_OP_ERR_DBG		(NIX_LFBASE | 0x260)
+#define	NIX_LF_MNQ_ERR_DBG		(NIX_LFBASE | 0x270)
+#define	NIX_LF_SEND_ERR_DBG		(NIX_LFBASE | 0x280)
+#define	NIX_LF_TX_STATX(a)		(NIX_LFBASE | 0x300 | (a) << 3)
+#define	NIX_LF_RX_STATX(a)		(NIX_LFBASE | 0x400 | (a) << 3)
+#define	NIX_LF_OP_SENDX(a)		(NIX_LFBASE | 0x800 | (a) << 3)
+#define	NIX_LF_RQ_OP_INT		(NIX_LFBASE | 0x900)
+#define	NIX_LF_RQ_OP_OCTS		(NIX_LFBASE | 0x910)
+#define	NIX_LF_RQ_OP_PKTS		(NIX_LFBASE | 0x920)
+#define	NIX_LF_OP_IPSEC_DYNO_CN		(NIX_LFBASE | 0x980)
+#define	NIX_LF_SQ_OP_INT		(NIX_LFBASE | 0xa00)
+#define	NIX_LF_SQ_OP_OCTS		(NIX_LFBASE | 0xa10)
+#define	NIX_LF_SQ_OP_PKTS		(NIX_LFBASE | 0xa20)
+#define	NIX_LF_SQ_OP_STATUS		(NIX_LFBASE | 0xa30)
+#define	NIX_LF_CQ_OP_INT		(NIX_LFBASE | 0xb00)
+#define	NIX_LF_CQ_OP_DOOR		(NIX_LFBASE | 0xb30)
+#define	NIX_LF_CQ_OP_STATUS		(NIX_LFBASE | 0xb40)
+#define	NIX_LF_QINTX_CNT(a)		(NIX_LFBASE | 0xC00 | (a) << 12)
+#define	NIX_LF_QINTX_INT(a)		(NIX_LFBASE | 0xC10 | (a) << 12)
+#define	NIX_LF_QINTX_INT_W1S(a)		(NIX_LFBASE | 0xC18 | (a) << 12)
+#define	NIX_LF_QINTX_ENA_W1S(a)		(NIX_LFBASE | 0xC20 | (a) << 12)
+#define	NIX_LF_QINTX_ENA_W1C(a)		(NIX_LFBASE | 0xC30 | (a) << 12)
+#define	NIX_LF_CINTX_CNT(a)		(NIX_LFBASE | 0xD00 | (a) << 12)
+#define	NIX_LF_CINTX_WAIT(a)		(NIX_LFBASE | 0xD10 | (a) << 12)
+#define	NIX_LF_CINTX_INT(a)		(NIX_LFBASE | 0xD20 | (a) << 12)
+#define	NIX_LF_CINTX_INT_W1S(a)		(NIX_LFBASE | 0xD30 | (a) << 12)
+#define	NIX_LF_CINTX_ENA_W1S(a)		(NIX_LFBASE | 0xD40 | (a) << 12)
+#define	NIX_LF_CINTX_ENA_W1C(a)		(NIX_LFBASE | 0xD50 | (a) << 12)
+
+/* NIX AF transmit scheduler registers */
+#define NIX_AF_SMQX_CFG(a)		(0x700 | (a) << 16)
+#define NIX_AF_TL1X_SCHEDULE(a)		(0xC00 | (a) << 16)
+#define NIX_AF_TL1X_CIR(a)		(0xC20 | (a) << 16)
+#define NIX_AF_TL1X_TOPOLOGY(a)		(0xC80 | (a) << 16)
+#define NIX_AF_TL2X_PARENT(a)		(0xE88 | (a) << 16)
+#define NIX_AF_TL2X_SCHEDULE(a)		(0xE00 | (a) << 16)
+#define NIX_AF_TL3X_PARENT(a)		(0x1088 | (a) << 16)
+#define NIX_AF_TL3X_SCHEDULE(a)		(0x1000 | (a) << 16)
+#define NIX_AF_TL4X_PARENT(a)		(0x1288 | (a) << 16)
+#define NIX_AF_TL4X_SCHEDULE(a)		(0x1200 | (a) << 16)
+#define NIX_AF_MDQX_SCHEDULE(a)		(0x1400 | (a) << 16)
+#define NIX_AF_MDQX_PARENT(a)		(0x1480 | (a) << 16)
+#define NIX_AF_TL3_TL2X_LINKX_CFG(a, b)	(0x1700 | (a) << 16 | (b) << 3)
+
+/* LMT LF registers */
+#define LMT_LFBASE			BIT_ULL(RVU_FUNC_BLKADDR_SHIFT)
+#define LMT_LF_LMTLINEX(a)		(LMT_LFBASE | 0x000 | (a) << 12)
+#define LMT_LF_LMTCANCEL		(LMT_LFBASE | 0x400)
+
 #endif /* OTX2_REG_H */
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_struct.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_struct.h
index e37f89f6c965..13086b05e872 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_struct.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_struct.h
@@ -17,4 +17,15 @@ enum nix_cqesz_e {
 	NIX_XQESZ_W16 = 0x1,
 };
 
+enum nix_sqes_e {
+	NIX_SQESZ_W16 = 0x0,
+	NIX_SQESZ_W8 = 0x1,
+};
+
+enum nix_send_ldtype {
+	NIX_SEND_LDTYPE_LDD  = 0x0,
+	NIX_SEND_LDTYPE_LDT  = 0x1,
+	NIX_SEND_LDTYPE_LDWB = 0x2,
+};
+
 #endif /* OTX2_STRUCT_H */
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h
new file mode 100644
index 000000000000..ce6efcf2e8cf
--- /dev/null
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h
@@ -0,0 +1,102 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Marvell OcteonTx2 RVU Ethernet driver
+ *
+ * Copyright (C) 2020 Marvell International Ltd.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef OTX2_TXRX_H
+#define OTX2_TXRX_H
+
+#include <linux/etherdevice.h>
+#include <linux/iommu.h>
+#include <linux/if_vlan.h>
+
+#define LBK_CHAN_BASE	0x000
+#define SDP_CHAN_BASE	0x700
+#define CGX_CHAN_BASE	0x800
+
+#define OTX2_DATA_ALIGN(X)	ALIGN(X, OTX2_ALIGN)
+#define OTX2_HEAD_ROOM		OTX2_ALIGN
+
+/* Rx buffer size should be in multiples of 128bytes */
+#define RCV_FRAG_LEN1(x)				\
+		((OTX2_HEAD_ROOM + OTX2_DATA_ALIGN(x)) + \
+		OTX2_DATA_ALIGN(sizeof(struct skb_shared_info)))
+
+/* Prefer 2048 byte buffers for better last level cache
+ * utilization or data distribution across regions.
+ */
+#define RCV_FRAG_LEN(x)	\
+		((RCV_FRAG_LEN1(x) < 2048) ? 2048 : RCV_FRAG_LEN1(x))
+
+#define DMA_BUFFER_LEN(x)		\
+		((x) - OTX2_HEAD_ROOM - \
+		OTX2_DATA_ALIGN(sizeof(struct skb_shared_info)))
+
+struct otx2_snd_queue {
+	u8			aura_id;
+	u16			sqe_size;
+	u32			sqe_cnt;
+	u16			num_sqbs;
+	u8			sqe_per_sqb;
+	u64			 io_addr;
+	u64			*aura_fc_addr;
+	u64			*lmt_addr;
+	void			*sqe_base;
+	struct qmem		*sqe;
+	u16			sqb_count;
+	u64			*sqb_ptrs;
+} ____cacheline_aligned_in_smp;
+
+struct otx2_pool {
+	struct qmem		*stack;
+	struct qmem		*fc_addr;
+	u8			rbpage_order;
+	u16			rbsize;
+	u32			page_offset;
+	u16			pageref;
+	struct page		*page;
+};
+
+enum cq_type {
+	CQ_RX,
+	CQ_TX,
+	CQS_PER_CINT = 2, /* RQ + SQ */
+};
+
+struct otx2_cq_queue {
+	u8			cq_idx;
+	u8			cq_type;
+	u16			cqe_size;
+	u16			pool_ptrs;
+	u32			cqe_cnt;
+	void			*cqe_base;
+	struct qmem		*cqe;
+	struct otx2_pool	*rbpool;
+} ____cacheline_aligned_in_smp;
+
+struct otx2_qset {
+	u32			rqe_cnt;
+	u32			sqe_cnt; /* Keep these two at top */
+#define OTX2_MAX_CQ_CNT		64
+	u16			cq_cnt;
+	u16			xqe_size;
+	struct otx2_pool	*pool;
+	struct otx2_cq_queue	*cq;
+	struct otx2_snd_queue	*sq;
+};
+
+/* Translate IOVA to physical address */
+static inline u64 otx2_iova_to_phys(void *iommu_domain, dma_addr_t dma_addr)
+{
+	/* Translation is installed only when IOMMU is present */
+	if (likely(iommu_domain))
+		return iommu_iova_to_phys(iommu_domain, dma_addr);
+	return dma_addr;
+}
+
+#endif /* OTX2_TXRX_H */
-- 
cgit v1.2.3-59-g8ed1b


From 04a21ef303df74737259190d7d7cf77a129bb56a Mon Sep 17 00:00:00 2001
From: Sunil Goutham <sgoutham@marvell.com>
Date: Mon, 27 Jan 2020 18:35:19 +0530
Subject: octeontx2-pf: Setup interrupts and NAPI handler

Completion queue (CQ) is the one with which HW notifies SW on a packet
reception or transmission. Each of the RQ and SQ are mapped to a unique
CQ and again both CQs are mapped to same interrupt ie the CINT. So that
each core has one interrupt source in whose handler both Rx and Tx
notifications are processed.

Also
- Registered a NAPI handler for the CINT.
- Setup coalescing parameters.
- IRQ affinity hints etc

Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../net/ethernet/marvell/octeontx2/nic/Makefile    |   2 +-
 .../ethernet/marvell/octeontx2/nic/otx2_common.c   |  55 +++++++++
 .../ethernet/marvell/octeontx2/nic/otx2_common.h   |  54 +++++++++
 .../net/ethernet/marvell/octeontx2/nic/otx2_pf.c   | 125 ++++++++++++++++++++-
 .../net/ethernet/marvell/octeontx2/nic/otx2_txrx.c |  68 +++++++++++
 .../net/ethernet/marvell/octeontx2/nic/otx2_txrx.h |  38 ++++++-
 6 files changed, 332 insertions(+), 10 deletions(-)
 create mode 100644 drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c

diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/Makefile b/drivers/net/ethernet/marvell/octeontx2/nic/Makefile
index 339fde892a36..0484d709fe68 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/Makefile
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/Makefile
@@ -5,6 +5,6 @@
 
 obj-$(CONFIG_OCTEONTX2_PF) += octeontx2_nicpf.o
 
-octeontx2_nicpf-y := otx2_pf.o otx2_common.o
+octeontx2_nicpf-y := otx2_pf.o otx2_common.o otx2_txrx.o
 
 ccflags-y += -I$(srctree)/drivers/net/ethernet/marvell/octeontx2/af
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
index 3ebbf0462b8f..8f7b2cf40d7e 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
@@ -15,6 +15,20 @@
 #include "otx2_common.h"
 #include "otx2_struct.h"
 
+void otx2_config_irq_coalescing(struct otx2_nic *pfvf, int qidx)
+{
+	/* Configure CQE interrupt coalescing parameters
+	 *
+	 * HW triggers an irq when ECOUNT > cq_ecount_wait, hence
+	 * set 1 less than cq_ecount_wait. And cq_time_wait is in
+	 * usecs, convert that to 100ns count.
+	 */
+	otx2_write64(pfvf, NIX_LF_CINTX_WAIT(qidx),
+		     ((u64)(pfvf->hw.cq_time_wait * 10) << 48) |
+		     ((u64)pfvf->hw.cq_qcount_wait << 32) |
+		     (pfvf->hw.cq_ecount_wait - 1));
+}
+
 dma_addr_t otx2_alloc_rbuf(struct otx2_nic *pfvf, struct otx2_pool *pool,
 			   gfp_t gfp)
 {
@@ -904,6 +918,47 @@ void mbox_handler_msix_offset(struct otx2_nic *pfvf,
 	pfvf->hw.nix_msixoff = rsp->nix_msixoff;
 }
 
+void otx2_free_cints(struct otx2_nic *pfvf, int n)
+{
+	struct otx2_qset *qset = &pfvf->qset;
+	struct otx2_hw *hw = &pfvf->hw;
+	int irq, qidx;
+
+	for (qidx = 0, irq = hw->nix_msixoff + NIX_LF_CINT_VEC_START;
+	     qidx < n;
+	     qidx++, irq++) {
+		int vector = pci_irq_vector(pfvf->pdev, irq);
+
+		irq_set_affinity_hint(vector, NULL);
+		free_cpumask_var(hw->affinity_mask[irq]);
+		free_irq(vector, &qset->napi[qidx]);
+	}
+}
+
+void otx2_set_cints_affinity(struct otx2_nic *pfvf)
+{
+	struct otx2_hw *hw = &pfvf->hw;
+	int vec, cpu, irq, cint;
+
+	vec = hw->nix_msixoff + NIX_LF_CINT_VEC_START;
+	cpu = cpumask_first(cpu_online_mask);
+
+	/* CQ interrupts */
+	for (cint = 0; cint < pfvf->hw.cint_cnt; cint++, vec++) {
+		if (!alloc_cpumask_var(&hw->affinity_mask[vec], GFP_KERNEL))
+			return;
+
+		cpumask_set_cpu(cpu, hw->affinity_mask[vec]);
+
+		irq = pci_irq_vector(pfvf->pdev, vec);
+		irq_set_affinity_hint(irq, hw->affinity_mask[vec]);
+
+		cpu = cpumask_next(cpu, cpu_online_mask);
+		if (unlikely(cpu >= nr_cpu_ids))
+			cpu = 0;
+	}
+}
+
 #define M(_name, _id, _fn_name, _req_type, _rsp_type)			\
 int __weak								\
 otx2_mbox_up_handler_ ## _fn_name(struct otx2_nic *pfvf,		\
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
index a4f1c604cb60..92e08f57b07d 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
@@ -21,6 +21,8 @@
 /* PCI device IDs */
 #define PCI_DEVID_OCTEONTX2_RVU_PF              0xA063
 
+#define PCI_SUBSYS_DEVID_96XX_RVU_PFVF		0xB200
+
 /* PCI BAR nos */
 #define PCI_CFG_REG_BAR_NUM                     2
 #define PCI_MBOX_BAR_NUM                        4
@@ -32,6 +34,13 @@ enum arua_mapped_qtypes {
 	AURA_NIX_SQ,
 };
 
+/* NIX LF interrupts range*/
+#define NIX_LF_QINT_VEC_START			0x00
+#define NIX_LF_CINT_VEC_START			0x40
+#define NIX_LF_GINT_VEC				0x80
+#define NIX_LF_ERR_VEC				0x81
+#define NIX_LF_POISON_VEC			0x82
+
 struct mbox {
 	struct otx2_mbox	mbox;
 	struct work_struct	mbox_wrk;
@@ -64,9 +73,13 @@ struct otx2_hw {
 	/* HW settings, coalescing etc */
 	u16			rx_chan_base;
 	u16			tx_chan_base;
+	u16			cq_qcount_wait;
+	u16			cq_ecount_wait;
 	u16			rq_skid;
+	u8			cq_time_wait;
 
 	/* MSI-X */
+	u8			cint_cnt; /* CQ interrupt count */
 	u16			npa_msixoff; /* Offset of NPA vectors */
 	u16			nix_msixoff; /* Offset of NIX vectors */
 	char			*irq_name;
@@ -94,6 +107,36 @@ struct otx2_nic {
 	int			nix_blkaddr;
 };
 
+static inline bool is_96xx_A0(struct pci_dev *pdev)
+{
+	return (pdev->revision == 0x00) &&
+		(pdev->subsystem_device == PCI_SUBSYS_DEVID_96XX_RVU_PFVF);
+}
+
+static inline bool is_96xx_B0(struct pci_dev *pdev)
+{
+	return (pdev->revision == 0x01) &&
+		(pdev->subsystem_device == PCI_SUBSYS_DEVID_96XX_RVU_PFVF);
+}
+
+static inline void otx2_setup_dev_hw_settings(struct otx2_nic *pfvf)
+{
+	pfvf->hw.cq_time_wait = CQ_TIMER_THRESH_DEFAULT;
+	pfvf->hw.cq_ecount_wait = CQ_CQE_THRESH_DEFAULT;
+	pfvf->hw.cq_qcount_wait = CQ_QCOUNT_DEFAULT;
+
+	if (is_96xx_A0(pfvf->pdev)) {
+		/* Time based irq coalescing is not supported */
+		pfvf->hw.cq_qcount_wait = 0x0;
+
+		/* Due to HW issue previous silicons required minimum
+		 * 600 unused CQE to avoid CQ overflow.
+		 */
+		pfvf->hw.rq_skid = 600;
+		pfvf->qset.rqe_cnt = Q_COUNT(Q_SIZE_1K);
+	}
+}
+
 /* Register read/write APIs */
 static inline void __iomem *otx2_get_regaddr(struct otx2_nic *nic, u64 offset)
 {
@@ -337,6 +380,11 @@ MBOX_UP_CGX_MESSAGES
 #define	RVU_PFVF_FUNC_SHIFT	0
 #define	RVU_PFVF_FUNC_MASK	0x3FF
 
+static inline int rvu_get_pf(u16 pcifunc)
+{
+	return (pcifunc >> RVU_PFVF_PF_SHIFT) & RVU_PFVF_PF_MASK;
+}
+
 static inline dma_addr_t otx2_dma_map_page(struct otx2_nic *pfvf,
 					   struct page *page,
 					   size_t offset, size_t size,
@@ -359,6 +407,12 @@ static inline void otx2_dma_unmap_page(struct otx2_nic *pfvf,
 			     dir, DMA_ATTR_SKIP_CPU_SYNC);
 }
 
+/* MSI-X APIs */
+void otx2_free_cints(struct otx2_nic *pfvf, int n);
+void otx2_set_cints_affinity(struct otx2_nic *pfvf);
+
+void otx2_config_irq_coalescing(struct otx2_nic *pfvf, int qidx);
+
 /* RVU block related APIs */
 int otx2_attach_npa_nix(struct otx2_nic *pfvf);
 int otx2_detach_resources(struct mbox *mbox);
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
index 735188900de4..32c8bc422024 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
@@ -386,6 +386,38 @@ static int otx2_set_real_num_queues(struct net_device *netdev,
 	return err;
 }
 
+static irqreturn_t otx2_cq_intr_handler(int irq, void *cq_irq)
+{
+	struct otx2_cq_poll *cq_poll = (struct otx2_cq_poll *)cq_irq;
+	struct otx2_nic *pf = (struct otx2_nic *)cq_poll->dev;
+	int qidx = cq_poll->cint_idx;
+
+	/* Disable interrupts.
+	 *
+	 * Completion interrupts behave in a level-triggered interrupt
+	 * fashion, and hence have to be cleared only after it is serviced.
+	 */
+	otx2_write64(pf, NIX_LF_CINTX_ENA_W1C(qidx), BIT_ULL(0));
+
+	/* Schedule NAPI */
+	napi_schedule_irqoff(&cq_poll->napi);
+
+	return IRQ_HANDLED;
+}
+
+static void otx2_disable_napi(struct otx2_nic *pf)
+{
+	struct otx2_qset *qset = &pf->qset;
+	struct otx2_cq_poll *cq_poll;
+	int qidx;
+
+	for (qidx = 0; qidx < pf->hw.cint_cnt; qidx++) {
+		cq_poll = &qset->napi[qidx];
+		napi_disable(&cq_poll->napi);
+		netif_napi_del(&cq_poll->napi);
+	}
+}
+
 static void otx2_free_cq_res(struct otx2_nic *pf)
 {
 	struct otx2_qset *qset = &pf->qset;
@@ -564,12 +596,21 @@ static void otx2_free_hw_resources(struct otx2_nic *pf)
 static int otx2_open(struct net_device *netdev)
 {
 	struct otx2_nic *pf = netdev_priv(netdev);
+	struct otx2_cq_poll *cq_poll = NULL;
 	struct otx2_qset *qset = &pf->qset;
-	int err = 0;
+	int err = 0, qidx, vec;
+	char *irq_name;
 
 	netif_carrier_off(netdev);
 
 	pf->qset.cq_cnt = pf->hw.rx_queues + pf->hw.tx_queues;
+	/* RQ and SQs are mapped to different CQs,
+	 * so find out max CQ IRQs (i.e CINTs) needed.
+	 */
+	pf->hw.cint_cnt = max(pf->hw.rx_queues, pf->hw.tx_queues);
+	qset->napi = kcalloc(pf->hw.cint_cnt, sizeof(*cq_poll), GFP_KERNEL);
+	if (!qset->napi)
+		return -ENOMEM;
 
 	/* CQ size of RQ */
 	qset->rqe_cnt = qset->rqe_cnt ? qset->rqe_cnt : Q_COUNT(Q_SIZE_256);
@@ -591,23 +632,100 @@ static int otx2_open(struct net_device *netdev)
 	if (err)
 		goto err_free_mem;
 
+	/* Register NAPI handler */
+	for (qidx = 0; qidx < pf->hw.cint_cnt; qidx++) {
+		cq_poll = &qset->napi[qidx];
+		cq_poll->cint_idx = qidx;
+		/* RQ0 & SQ0 are mapped to CINT0 and so on..
+		 * 'cq_ids[0]' points to RQ's CQ and
+		 * 'cq_ids[1]' points to SQ's CQ and
+		 */
+		cq_poll->cq_ids[CQ_RX] =
+			(qidx <  pf->hw.rx_queues) ? qidx : CINT_INVALID_CQ;
+		cq_poll->cq_ids[CQ_TX] = (qidx < pf->hw.tx_queues) ?
+				      qidx + pf->hw.rx_queues : CINT_INVALID_CQ;
+		cq_poll->dev = (void *)pf;
+		netif_napi_add(netdev, &cq_poll->napi,
+			       otx2_napi_handler, NAPI_POLL_WEIGHT);
+		napi_enable(&cq_poll->napi);
+	}
+
+	/* Register CQ IRQ handlers */
+	vec = pf->hw.nix_msixoff + NIX_LF_CINT_VEC_START;
+	for (qidx = 0; qidx < pf->hw.cint_cnt; qidx++) {
+		irq_name = &pf->hw.irq_name[vec * NAME_SIZE];
+
+		snprintf(irq_name, NAME_SIZE, "%s-rxtx-%d", pf->netdev->name,
+			 qidx);
+
+		err = request_irq(pci_irq_vector(pf->pdev, vec),
+				  otx2_cq_intr_handler, 0, irq_name,
+				  &qset->napi[qidx]);
+		if (err) {
+			dev_err(pf->dev,
+				"RVUPF%d: IRQ registration failed for CQ%d\n",
+				rvu_get_pf(pf->pcifunc), qidx);
+			goto err_free_cints;
+		}
+		vec++;
+
+		otx2_config_irq_coalescing(pf, qidx);
+
+		/* Enable CQ IRQ */
+		otx2_write64(pf, NIX_LF_CINTX_INT(qidx), BIT_ULL(0));
+		otx2_write64(pf, NIX_LF_CINTX_ENA_W1S(qidx), BIT_ULL(0));
+	}
+
+	otx2_set_cints_affinity(pf);
+
 	return 0;
+
+err_free_cints:
+	otx2_free_cints(pf, qidx);
+	otx2_disable_napi(pf);
+	otx2_free_hw_resources(pf);
 err_free_mem:
 	kfree(qset->sq);
 	kfree(qset->cq);
+	kfree(qset->napi);
 	return err;
 }
 
 static int otx2_stop(struct net_device *netdev)
 {
 	struct otx2_nic *pf = netdev_priv(netdev);
+	struct otx2_cq_poll *cq_poll = NULL;
 	struct otx2_qset *qset = &pf->qset;
+	int qidx, vec;
+
+	netif_carrier_off(netdev);
+	netif_tx_stop_all_queues(netdev);
+
+	/* Cleanup CQ NAPI and IRQ */
+	vec = pf->hw.nix_msixoff + NIX_LF_CINT_VEC_START;
+	for (qidx = 0; qidx < pf->hw.cint_cnt; qidx++) {
+		/* Disable interrupt */
+		otx2_write64(pf, NIX_LF_CINTX_ENA_W1C(qidx), BIT_ULL(0));
+
+		synchronize_irq(pci_irq_vector(pf->pdev, vec));
+
+		cq_poll = &qset->napi[qidx];
+		napi_synchronize(&cq_poll->napi);
+		vec++;
+	}
+
+	netif_tx_disable(netdev);
 
 	otx2_free_hw_resources(pf);
+	otx2_free_cints(pf, pf->hw.cint_cnt);
+	otx2_disable_napi(pf);
+
+	for (qidx = 0; qidx < netdev->num_tx_queues; qidx++)
+		netdev_tx_reset_queue(netdev_get_tx_queue(netdev, qidx));
 
 	kfree(qset->sq);
 	kfree(qset->cq);
-
+	kfree(qset->napi);
 	/* Do not clear RQ/SQ ringsize settings */
 	memset((void *)qset + offsetof(struct otx2_qset, sqe_cnt), 0,
 	       sizeof(*qset) - offsetof(struct otx2_qset, sqe_cnt));
@@ -646,7 +764,6 @@ static int otx2_realloc_msix_vectors(struct otx2_nic *pf)
 	 * upto NIX vector offset.
 	 */
 	num_vec = hw->nix_msixoff;
-#define NIX_LF_CINT_VEC_START			0x40
 	num_vec += NIX_LF_CINT_VEC_START + hw->max_queues;
 
 	otx2_disable_mbox_intr(pf);
@@ -769,6 +886,8 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	if (err)
 		goto err_detach_rsrc;
 
+	otx2_setup_dev_hw_settings(pf);
+
 	/* NPA's pool is a stack to which SW frees buffer pointers via Aura.
 	 * HW allocates buffer pointer from stack and uses it for DMA'ing
 	 * ingress packet. In some scenarios HW can free back allocated buffer
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c
new file mode 100644
index 000000000000..b07082ec7d98
--- /dev/null
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c
@@ -0,0 +1,68 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Marvell OcteonTx2 RVU Ethernet driver
+ *
+ * Copyright (C) 2020 Marvell International Ltd.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/etherdevice.h>
+#include <net/ip.h>
+
+#include "otx2_reg.h"
+#include "otx2_common.h"
+#include "otx2_struct.h"
+#include "otx2_txrx.h"
+
+static int otx2_rx_napi_handler(struct otx2_nic *pfvf,
+				struct napi_struct *napi,
+				struct otx2_cq_queue *cq, int budget)
+{
+	 /* Nothing to do, for now */
+	return 0;
+}
+
+static int otx2_tx_napi_handler(struct otx2_nic *pfvf,
+				struct otx2_cq_queue *cq, int budget)
+{
+	 /* Nothing to do, for now */
+	return 0;
+}
+
+int otx2_napi_handler(struct napi_struct *napi, int budget)
+{
+	struct otx2_cq_poll *cq_poll;
+	int workdone = 0, cq_idx, i;
+	struct otx2_cq_queue *cq;
+	struct otx2_qset *qset;
+	struct otx2_nic *pfvf;
+
+	cq_poll = container_of(napi, struct otx2_cq_poll, napi);
+	pfvf = (struct otx2_nic *)cq_poll->dev;
+	qset = &pfvf->qset;
+
+	for (i = CQS_PER_CINT - 1; i >= 0; i--) {
+		cq_idx = cq_poll->cq_ids[i];
+		if (unlikely(cq_idx == CINT_INVALID_CQ))
+			continue;
+		cq = &qset->cq[cq_idx];
+		if (cq->cq_type == CQ_RX) {
+			workdone += otx2_rx_napi_handler(pfvf, napi,
+							 cq, budget);
+		} else {
+			workdone += otx2_tx_napi_handler(pfvf, cq, budget);
+		}
+	}
+
+	/* Clear the IRQ */
+	otx2_write64(pfvf, NIX_LF_CINTX_INT(cq_poll->cint_idx), BIT_ULL(0));
+
+	if (workdone < budget && napi_complete_done(napi, workdone)) {
+		/* Re-enable interrupts */
+		otx2_write64(pfvf, NIX_LF_CINTX_ENA_W1S(cq_poll->cint_idx),
+			     BIT_ULL(0));
+	}
+	return workdone;
+}
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h
index ce6efcf2e8cf..a81bdc633884 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h
@@ -37,6 +37,22 @@
 		((x) - OTX2_HEAD_ROOM - \
 		OTX2_DATA_ALIGN(sizeof(struct skb_shared_info)))
 
+/* IRQ triggered when NIX_LF_CINTX_CNT[ECOUNT]
+ * is equal to this value.
+ */
+#define CQ_CQE_THRESH_DEFAULT	10
+
+/* IRQ triggered when NIX_LF_CINTX_CNT[ECOUNT]
+ * is nonzero and this much time elapses after that.
+ */
+#define CQ_TIMER_THRESH_DEFAULT	1  /* 1 usec */
+#define CQ_TIMER_THRESH_MAX     25 /* 25 usec */
+
+/* Min number of CQs (of the ones mapped to this CINT)
+ * with valid CQEs.
+ */
+#define CQ_QCOUNT_DEFAULT	1
+
 struct otx2_snd_queue {
 	u8			aura_id;
 	u16			sqe_size;
@@ -52,6 +68,20 @@ struct otx2_snd_queue {
 	u64			*sqb_ptrs;
 } ____cacheline_aligned_in_smp;
 
+enum cq_type {
+	CQ_RX,
+	CQ_TX,
+	CQS_PER_CINT = 2, /* RQ + SQ */
+};
+
+struct otx2_cq_poll {
+	void			*dev;
+#define CINT_INVALID_CQ		255
+	u8			cint_idx;
+	u8			cq_ids[CQS_PER_CINT];
+	struct napi_struct	napi;
+};
+
 struct otx2_pool {
 	struct qmem		*stack;
 	struct qmem		*fc_addr;
@@ -62,12 +92,6 @@ struct otx2_pool {
 	struct page		*page;
 };
 
-enum cq_type {
-	CQ_RX,
-	CQ_TX,
-	CQS_PER_CINT = 2, /* RQ + SQ */
-};
-
 struct otx2_cq_queue {
 	u8			cq_idx;
 	u8			cq_type;
@@ -86,6 +110,7 @@ struct otx2_qset {
 	u16			cq_cnt;
 	u16			xqe_size;
 	struct otx2_pool	*pool;
+	struct otx2_cq_poll	*napi;
 	struct otx2_cq_queue	*cq;
 	struct otx2_snd_queue	*sq;
 };
@@ -99,4 +124,5 @@ static inline u64 otx2_iova_to_phys(void *iommu_domain, dma_addr_t dma_addr)
 	return dma_addr;
 }
 
+int otx2_napi_handler(struct napi_struct *napi, int budget);
 #endif /* OTX2_TXRX_H */
-- 
cgit v1.2.3-59-g8ed1b


From abe0254333124598d1e7452c19e9e753c6bfaada Mon Sep 17 00:00:00 2001
From: Sunil Goutham <sgoutham@marvell.com>
Date: Mon, 27 Jan 2020 18:35:20 +0530
Subject: octeontx2-pf: Receive packet handling support

Added receive packet handling (NAPI) support, error stats, RX_ALL
capability config option to passon error pkts to stack upon user request.

In subsequent patches these error stats will be added to ethttool.

Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../ethernet/marvell/octeontx2/nic/otx2_common.c   |   2 +
 .../ethernet/marvell/octeontx2/nic/otx2_common.h   |  44 +++++
 .../net/ethernet/marvell/octeontx2/nic/otx2_pf.c   |  16 +-
 .../ethernet/marvell/octeontx2/nic/otx2_struct.h   | 108 +++++++++++++
 .../net/ethernet/marvell/octeontx2/nic/otx2_txrx.c | 180 ++++++++++++++++++++-
 .../net/ethernet/marvell/octeontx2/nic/otx2_txrx.h |   2 +
 6 files changed, 349 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
index 8f7b2cf40d7e..6b548313be3b 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
@@ -334,6 +334,7 @@ static int otx2_cq_init(struct otx2_nic *pfvf, u16 qidx)
 	cq->cq_idx = qidx;
 	if (qidx < pfvf->hw.rx_queues) {
 		cq->cq_type = CQ_RX;
+		cq->cint_idx = qidx;
 		cq->cqe_cnt = qset->rqe_cnt;
 	} else {
 		cq->cq_type = CQ_TX;
@@ -364,6 +365,7 @@ static int otx2_cq_init(struct otx2_nic *pfvf, u16 qidx)
 	aq->cq.qsize = Q_SIZE(cq->cqe_cnt, 4);
 	aq->cq.caching = 1;
 	aq->cq.base = cq->cqe->iova;
+	aq->cq.cint_idx = cq->cint_idx;
 	aq->cq.avg_level = 255;
 
 	if (qidx < pfvf->hw.rx_queues) {
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
index 92e08f57b07d..622c4f8f3ca0 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
@@ -41,6 +41,46 @@ enum arua_mapped_qtypes {
 #define NIX_LF_ERR_VEC				0x81
 #define NIX_LF_POISON_VEC			0x82
 
+/* NIX (or NPC) RX errors */
+enum otx2_errlvl {
+	NPC_ERRLVL_RE,
+	NPC_ERRLVL_LID_LA,
+	NPC_ERRLVL_LID_LB,
+	NPC_ERRLVL_LID_LC,
+	NPC_ERRLVL_LID_LD,
+	NPC_ERRLVL_LID_LE,
+	NPC_ERRLVL_LID_LF,
+	NPC_ERRLVL_LID_LG,
+	NPC_ERRLVL_LID_LH,
+	NPC_ERRLVL_NIX = 0x0F,
+};
+
+enum otx2_errcodes_re {
+	/* NPC_ERRLVL_RE errcodes */
+	ERRCODE_FCS = 0x7,
+	ERRCODE_FCS_RCV = 0x8,
+	ERRCODE_UNDERSIZE = 0x10,
+	ERRCODE_OVERSIZE = 0x11,
+	ERRCODE_OL2_LEN_MISMATCH = 0x12,
+	/* NPC_ERRLVL_NIX errcodes */
+	ERRCODE_OL3_LEN = 0x10,
+	ERRCODE_OL4_LEN = 0x11,
+	ERRCODE_OL4_CSUM = 0x12,
+	ERRCODE_IL3_LEN = 0x20,
+	ERRCODE_IL4_LEN = 0x21,
+	ERRCODE_IL4_CSUM = 0x22,
+};
+
+/* Driver counted stats */
+struct otx2_drv_stats {
+	atomic_t rx_fcs_errs;
+	atomic_t rx_oversize_errs;
+	atomic_t rx_undersize_errs;
+	atomic_t rx_csum_errs;
+	atomic_t rx_len_errs;
+	atomic_t rx_other_errs;
+};
+
 struct mbox {
 	struct otx2_mbox	mbox;
 	struct work_struct	mbox_wrk;
@@ -84,6 +124,9 @@ struct otx2_hw {
 	u16			nix_msixoff; /* Offset of NIX vectors */
 	char			*irq_name;
 	cpumask_var_t           *affinity_mask;
+
+	/* Stats */
+	struct otx2_drv_stats	drv_stats;
 };
 
 struct otx2_nic {
@@ -431,6 +474,7 @@ void otx2_sqb_flush(struct otx2_nic *pfvf);
 dma_addr_t otx2_alloc_rbuf(struct otx2_nic *pfvf, struct otx2_pool *pool,
 			   gfp_t gfp);
 void otx2_ctx_disable(struct mbox *mbox, int type, bool npa);
+void otx2_cleanup_rx_cqes(struct otx2_nic *pfvf, struct otx2_cq_queue *cq);
 
 /* Mbox handlers */
 void mbox_handler_msix_offset(struct otx2_nic *pfvf,
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
index 32c8bc422024..37362ef36683 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
@@ -547,9 +547,11 @@ exit:
 
 static void otx2_free_hw_resources(struct otx2_nic *pf)
 {
+	struct otx2_qset *qset = &pf->qset;
 	struct mbox *mbox = &pf->mbox;
+	struct otx2_cq_queue *cq;
 	struct msg_req *req;
-	int err;
+	int qidx, err;
 
 	/* Ensure all SQE are processed */
 	otx2_sqb_flush(pf);
@@ -562,6 +564,13 @@ static void otx2_free_hw_resources(struct otx2_nic *pf)
 	/* Disable RQs */
 	otx2_ctx_disable(mbox, NIX_AQ_CTYPE_RQ, false);
 
+	/*Dequeue all CQEs */
+	for (qidx = 0; qidx < qset->cq_cnt; qidx++) {
+		cq = &qset->cq[qidx];
+		if (cq->cq_type == CQ_RX)
+			otx2_cleanup_rx_cqes(pf, cq);
+	}
+
 	otx2_free_sq_res(pf);
 
 	/* Free RQ buffer pointers*/
@@ -901,6 +910,11 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	 */
 	pf->iommu_domain = iommu_get_domain_for_dev(dev);
 
+	netdev->hw_features = NETIF_F_RXCSUM;
+	netdev->features |= netdev->hw_features;
+
+	netdev->hw_features |= NETIF_F_RXALL;
+
 	netdev->netdev_ops = &otx2_netdev_ops;
 
 	err = register_netdev(netdev);
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_struct.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_struct.h
index 13086b05e872..dad73cb5e6f8 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_struct.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_struct.h
@@ -28,4 +28,112 @@ enum nix_send_ldtype {
 	NIX_SEND_LDTYPE_LDWB = 0x2,
 };
 
+/* NIX wqe/cqe types */
+enum nix_xqe_type {
+	NIX_XQE_TYPE_INVALID   = 0x0,
+	NIX_XQE_TYPE_RX        = 0x1,
+	NIX_XQE_TYPE_RX_IPSECS = 0x2,
+	NIX_XQE_TYPE_RX_IPSECH = 0x3,
+	NIX_XQE_TYPE_RX_IPSECD = 0x4,
+	NIX_XQE_TYPE_SEND      = 0x8,
+};
+
+/* NIX CQE/SQE subdescriptor types */
+enum nix_subdc {
+	NIX_SUBDC_NOP  = 0x0,
+	NIX_SUBDC_EXT  = 0x1,
+	NIX_SUBDC_CRC  = 0x2,
+	NIX_SUBDC_IMM  = 0x3,
+	NIX_SUBDC_SG   = 0x4,
+	NIX_SUBDC_MEM  = 0x5,
+	NIX_SUBDC_JUMP = 0x6,
+	NIX_SUBDC_WORK = 0x7,
+	NIX_SUBDC_SOD  = 0xf,
+};
+
+/* NIX CQE header structure */
+struct nix_cqe_hdr_s {
+	u64 flow_tag              : 32;
+	u64 q                     : 20;
+	u64 reserved_52_57        : 6;
+	u64 node                  : 2;
+	u64 cqe_type              : 4;
+};
+
+/* NIX CQE RX parse structure */
+struct nix_rx_parse_s {
+	u64 chan         : 12;
+	u64 desc_sizem1  : 5;
+	u64 rsvd_17      : 1;
+	u64 express      : 1;
+	u64 wqwd         : 1;
+	u64 errlev       : 4;
+	u64 errcode      : 8;
+	u64 latype       : 4;
+	u64 lbtype       : 4;
+	u64 lctype       : 4;
+	u64 ldtype       : 4;
+	u64 letype       : 4;
+	u64 lftype       : 4;
+	u64 lgtype       : 4;
+	u64 lhtype       : 4;
+	u64 pkt_lenm1    : 16; /* W1 */
+	u64 l2m          : 1;
+	u64 l2b          : 1;
+	u64 l3m          : 1;
+	u64 l3b          : 1;
+	u64 vtag0_valid  : 1;
+	u64 vtag0_gone   : 1;
+	u64 vtag1_valid  : 1;
+	u64 vtag1_gone   : 1;
+	u64 pkind        : 6;
+	u64 rsvd_95_94   : 2;
+	u64 vtag0_tci    : 16;
+	u64 vtag1_tci    : 16;
+	u64 laflags      : 8; /* W2 */
+	u64 lbflags      : 8;
+	u64 lcflags      : 8;
+	u64 ldflags      : 8;
+	u64 leflags      : 8;
+	u64 lfflags      : 8;
+	u64 lgflags      : 8;
+	u64 lhflags      : 8;
+	u64 eoh_ptr      : 8; /* W3 */
+	u64 wqe_aura     : 20;
+	u64 pb_aura      : 20;
+	u64 match_id     : 16;
+	u64 laptr        : 8; /* W4 */
+	u64 lbptr        : 8;
+	u64 lcptr        : 8;
+	u64 ldptr        : 8;
+	u64 leptr        : 8;
+	u64 lfptr        : 8;
+	u64 lgptr        : 8;
+	u64 lhptr        : 8;
+	u64 vtag0_ptr    : 8; /* W5 */
+	u64 vtag1_ptr    : 8;
+	u64 flow_key_alg : 5;
+	u64 rsvd_383_341 : 43;
+	u64 rsvd_447_384;     /* W6 */
+};
+
+/* NIX CQE RX scatter/gather subdescriptor structure */
+struct nix_rx_sg_s {
+	u64 seg_size   : 16; /* W0 */
+	u64 seg2_size  : 16;
+	u64 seg3_size  : 16;
+	u64 segs       : 2;
+	u64 rsvd_59_50 : 10;
+	u64 subdc      : 4;
+	u64 seg_addr;
+	u64 seg2_addr;
+	u64 seg3_addr;
+};
+
+struct nix_cqe_rx_s {
+	struct nix_cqe_hdr_s  hdr;
+	struct nix_rx_parse_s parse;
+	struct nix_rx_sg_s sg;
+};
+
 #endif /* OTX2_STRUCT_H */
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c
index b07082ec7d98..0c1519a05281 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c
@@ -16,12 +16,167 @@
 #include "otx2_struct.h"
 #include "otx2_txrx.h"
 
+#define CQE_ADDR(CQ, idx) ((CQ)->cqe_base + ((CQ)->cqe_size * (idx)))
+
+static struct nix_cqe_hdr_s *otx2_get_next_cqe(struct otx2_cq_queue *cq)
+{
+	struct nix_cqe_hdr_s *cqe_hdr;
+
+	cqe_hdr = (struct nix_cqe_hdr_s *)CQE_ADDR(cq, cq->cq_head);
+	if (cqe_hdr->cqe_type == NIX_XQE_TYPE_INVALID)
+		return NULL;
+
+	cq->cq_head++;
+	cq->cq_head &= (cq->cqe_cnt - 1);
+
+	return cqe_hdr;
+}
+
+static void otx2_skb_add_frag(struct otx2_nic *pfvf, struct sk_buff *skb,
+			      u64 iova, int len)
+{
+	struct page *page;
+	void *va;
+
+	va = phys_to_virt(otx2_iova_to_phys(pfvf->iommu_domain, iova));
+	page = virt_to_page(va);
+	skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, page,
+			va - page_address(page), len, pfvf->rbsize);
+
+	otx2_dma_unmap_page(pfvf, iova - OTX2_HEAD_ROOM,
+			    pfvf->rbsize, DMA_FROM_DEVICE);
+}
+
+static bool otx2_check_rcv_errors(struct otx2_nic *pfvf,
+				  struct nix_cqe_rx_s *cqe, int qidx)
+{
+	struct otx2_drv_stats *stats = &pfvf->hw.drv_stats;
+	struct nix_rx_parse_s *parse = &cqe->parse;
+
+	if (parse->errlev == NPC_ERRLVL_RE) {
+		switch (parse->errcode) {
+		case ERRCODE_FCS:
+		case ERRCODE_FCS_RCV:
+			atomic_inc(&stats->rx_fcs_errs);
+			break;
+		case ERRCODE_UNDERSIZE:
+			atomic_inc(&stats->rx_undersize_errs);
+			break;
+		case ERRCODE_OVERSIZE:
+			atomic_inc(&stats->rx_oversize_errs);
+			break;
+		case ERRCODE_OL2_LEN_MISMATCH:
+			atomic_inc(&stats->rx_len_errs);
+			break;
+		default:
+			atomic_inc(&stats->rx_other_errs);
+			break;
+		}
+	} else if (parse->errlev == NPC_ERRLVL_NIX) {
+		switch (parse->errcode) {
+		case ERRCODE_OL3_LEN:
+		case ERRCODE_OL4_LEN:
+		case ERRCODE_IL3_LEN:
+		case ERRCODE_IL4_LEN:
+			atomic_inc(&stats->rx_len_errs);
+			break;
+		case ERRCODE_OL4_CSUM:
+		case ERRCODE_IL4_CSUM:
+			atomic_inc(&stats->rx_csum_errs);
+			break;
+		default:
+			atomic_inc(&stats->rx_other_errs);
+			break;
+		}
+	} else {
+		atomic_inc(&stats->rx_other_errs);
+		/* For now ignore all the NPC parser errors and
+		 * pass the packets to stack.
+		 */
+		return false;
+	}
+
+	/* If RXALL is enabled pass on packets to stack. */
+	if (cqe->sg.segs && (pfvf->netdev->features & NETIF_F_RXALL))
+		return false;
+
+	/* Free buffer back to pool */
+	if (cqe->sg.segs)
+		otx2_aura_freeptr(pfvf, qidx, cqe->sg.seg_addr & ~0x07ULL);
+	return true;
+}
+
+static void otx2_rcv_pkt_handler(struct otx2_nic *pfvf,
+				 struct napi_struct *napi,
+				 struct otx2_cq_queue *cq,
+				 struct nix_cqe_rx_s *cqe)
+{
+	struct nix_rx_parse_s *parse = &cqe->parse;
+	struct sk_buff *skb = NULL;
+
+	if (unlikely(parse->errlev || parse->errcode)) {
+		if (otx2_check_rcv_errors(pfvf, cqe, cq->cq_idx))
+			return;
+	}
+
+	skb = napi_get_frags(napi);
+	if (unlikely(!skb))
+		return;
+
+	otx2_skb_add_frag(pfvf, skb, cqe->sg.seg_addr, cqe->sg.seg_size);
+	cq->pool_ptrs++;
+
+	skb_record_rx_queue(skb, cq->cq_idx);
+	if (pfvf->netdev->features & NETIF_F_RXCSUM)
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+	napi_gro_frags(napi);
+}
+
 static int otx2_rx_napi_handler(struct otx2_nic *pfvf,
 				struct napi_struct *napi,
 				struct otx2_cq_queue *cq, int budget)
 {
-	 /* Nothing to do, for now */
-	return 0;
+	struct nix_cqe_rx_s *cqe;
+	int processed_cqe = 0;
+	s64 bufptr;
+
+	while (likely(processed_cqe < budget)) {
+		cqe = (struct nix_cqe_rx_s *)CQE_ADDR(cq, cq->cq_head);
+		if (cqe->hdr.cqe_type == NIX_XQE_TYPE_INVALID ||
+		    !cqe->sg.seg_addr) {
+			if (!processed_cqe)
+				return 0;
+			break;
+		}
+		cq->cq_head++;
+		cq->cq_head &= (cq->cqe_cnt - 1);
+
+		otx2_rcv_pkt_handler(pfvf, napi, cq, cqe);
+
+		cqe->hdr.cqe_type = NIX_XQE_TYPE_INVALID;
+		cqe->sg.seg_addr = 0x00;
+		processed_cqe++;
+	}
+
+	/* Free CQEs to HW */
+	otx2_write64(pfvf, NIX_LF_CQ_OP_DOOR,
+		     ((u64)cq->cq_idx << 32) | processed_cqe);
+
+	if (unlikely(!cq->pool_ptrs))
+		return 0;
+
+	/* Refill pool with new buffers */
+	while (cq->pool_ptrs) {
+		bufptr = otx2_alloc_rbuf(pfvf, cq->rbpool, GFP_ATOMIC);
+		if (unlikely(bufptr <= 0))
+			break;
+		otx2_aura_freeptr(pfvf, cq->cq_idx, bufptr + OTX2_HEAD_ROOM);
+		cq->pool_ptrs--;
+	}
+	otx2_get_page(cq->rbpool);
+
+	return processed_cqe;
 }
 
 static int otx2_tx_napi_handler(struct otx2_nic *pfvf,
@@ -66,3 +221,24 @@ int otx2_napi_handler(struct napi_struct *napi, int budget)
 	}
 	return workdone;
 }
+
+void otx2_cleanup_rx_cqes(struct otx2_nic *pfvf, struct otx2_cq_queue *cq)
+{
+	struct nix_cqe_rx_s *cqe;
+	int processed_cqe = 0;
+	u64 iova, pa;
+
+	while ((cqe = (struct nix_cqe_rx_s *)otx2_get_next_cqe(cq))) {
+		if (!cqe->sg.subdc)
+			continue;
+		iova = cqe->sg.seg_addr - OTX2_HEAD_ROOM;
+		pa = otx2_iova_to_phys(pfvf->iommu_domain, iova);
+		otx2_dma_unmap_page(pfvf, iova, pfvf->rbsize, DMA_FROM_DEVICE);
+		put_page(virt_to_page(phys_to_virt(pa)));
+		processed_cqe++;
+	}
+
+	/* Free CQEs to HW */
+	otx2_write64(pfvf, NIX_LF_CQ_OP_DOOR,
+		     ((u64)cq->cq_idx << 32) | processed_cqe);
+}
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h
index a81bdc633884..0944c176efa3 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h
@@ -95,9 +95,11 @@ struct otx2_pool {
 struct otx2_cq_queue {
 	u8			cq_idx;
 	u8			cq_type;
+	u8			cint_idx; /* CQ interrupt id */
 	u16			cqe_size;
 	u16			pool_ptrs;
 	u32			cqe_cnt;
+	u32			cq_head;
 	void			*cqe_base;
 	struct qmem		*cqe;
 	struct otx2_pool	*rbpool;
-- 
cgit v1.2.3-59-g8ed1b


From 3ca6c4c882a7f34085b170d93cf0d0e843aa00e6 Mon Sep 17 00:00:00 2001
From: Sunil Goutham <sgoutham@marvell.com>
Date: Mon, 27 Jan 2020 18:35:21 +0530
Subject: octeontx2-pf: Add packet transmission support

This patch adds the packet transmission support.
For a given skb prepares send queue descriptors (SQEs) and pushes them
to HW. Here driver doesn't maintain it's own SQ rings, SQEs are pushed
to HW using a silicon specific operations called LMTST. From the
instuction HW derives the transmit queue number and queues the SQE to
that queue. These LMTST instructions are designed to avoid queue
maintenance in SW and lockless behavior ie when multiple cores are trying
to add SQEs to same queue then HW will takecare of serialization, no need
for SW to hold locks.

Also supports scatter/gather.

Co-developed-by: Geetha sowjanya <gakula@marvell.com>
Signed-off-by: Geetha sowjanya <gakula@marvell.com>
Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../ethernet/marvell/octeontx2/nic/otx2_common.c   |   7 +
 .../ethernet/marvell/octeontx2/nic/otx2_common.h   |  13 +
 .../net/ethernet/marvell/octeontx2/nic/otx2_pf.c   |  38 ++-
 .../ethernet/marvell/octeontx2/nic/otx2_struct.h   | 109 ++++++++
 .../net/ethernet/marvell/octeontx2/nic/otx2_txrx.c | 291 ++++++++++++++++++++-
 .../net/ethernet/marvell/octeontx2/nic/otx2_txrx.h |  14 +
 6 files changed, 470 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
index 6b548313be3b..e9bd2d6f6ebc 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
@@ -287,9 +287,15 @@ static int otx2_sq_init(struct otx2_nic *pfvf, u16 qidx, u16 sqb_aura)
 		return err;
 
 	sq->sqe_base = sq->sqe->base;
+	sq->sg = kcalloc(qset->sqe_cnt, sizeof(struct sg_list), GFP_KERNEL);
+	if (!sq->sg)
+		return -ENOMEM;
 
+	sq->head = 0;
 	sq->sqe_per_sqb = (pfvf->hw.sqb_size / sq->sqe_size) - 1;
 	sq->num_sqbs = (qset->sqe_cnt + sq->sqe_per_sqb) / sq->sqe_per_sqb;
+	/* Set SQE threshold to 10% of total SQEs */
+	sq->sqe_thresh = ((sq->num_sqbs * sq->sqe_per_sqb) * 10) / 100;
 	sq->aura_id = sqb_aura;
 	sq->aura_fc_addr = pool->fc_addr->base;
 	sq->lmt_addr = (__force u64 *)(pfvf->reg_base + LMT_LF_LMTLINEX(qidx));
@@ -338,6 +344,7 @@ static int otx2_cq_init(struct otx2_nic *pfvf, u16 qidx)
 		cq->cqe_cnt = qset->rqe_cnt;
 	} else {
 		cq->cq_type = CQ_TX;
+		cq->cint_idx = qidx - pfvf->hw.rx_queues;
 		cq->cqe_cnt = qset->sqe_cnt;
 	}
 	cq->cqe_size = pfvf->qset.xqe_size;
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
index 622c4f8f3ca0..11948540dd4a 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
@@ -300,9 +300,21 @@ static inline u64 otx2_atomic64_add(u64 incr, u64 *ptr)
 	return result;
 }
 
+static inline u64 otx2_lmt_flush(uint64_t addr)
+{
+	u64 result = 0;
+
+	__asm__ volatile(".cpu  generic+lse\n"
+			 "ldeor xzr,%x[rf],[%[rs]]"
+			 : [rf]"=r"(result)
+			 : [rs]"r"(addr));
+	return result;
+}
+
 #else
 #define otx2_write128(lo, hi, addr)
 #define otx2_atomic64_add(incr, ptr)		({ *ptr += incr; })
+#define otx2_lmt_flush(addr)			({ 0; })
 #endif
 
 /* Alloc pointer from pool/aura */
@@ -475,6 +487,7 @@ dma_addr_t otx2_alloc_rbuf(struct otx2_nic *pfvf, struct otx2_pool *pool,
 			   gfp_t gfp);
 void otx2_ctx_disable(struct mbox *mbox, int type, bool npa);
 void otx2_cleanup_rx_cqes(struct otx2_nic *pfvf, struct otx2_cq_queue *cq);
+void otx2_cleanup_tx_cqes(struct otx2_nic *pfvf, struct otx2_cq_queue *cq);
 
 /* Mbox handlers */
 void mbox_handler_msix_offset(struct otx2_nic *pfvf,
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
index 37362ef36683..a18bf5facaaf 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
@@ -445,6 +445,7 @@ static void otx2_free_sq_res(struct otx2_nic *pf)
 	for (qidx = 0; qidx < pf->hw.tx_queues; qidx++) {
 		sq = &qset->sq[qidx];
 		qmem_free(pf->dev, sq->sqe);
+		kfree(sq->sg);
 		kfree(sq->sqb_ptrs);
 	}
 }
@@ -569,6 +570,8 @@ static void otx2_free_hw_resources(struct otx2_nic *pf)
 		cq = &qset->cq[qidx];
 		if (cq->cq_type == CQ_RX)
 			otx2_cleanup_rx_cqes(pf, cq);
+		else
+			otx2_cleanup_tx_cqes(pf, cq);
 	}
 
 	otx2_free_sq_res(pf);
@@ -741,9 +744,41 @@ static int otx2_stop(struct net_device *netdev)
 	return 0;
 }
 
+static netdev_tx_t otx2_xmit(struct sk_buff *skb, struct net_device *netdev)
+{
+	struct otx2_nic *pf = netdev_priv(netdev);
+	int qidx = skb_get_queue_mapping(skb);
+	struct otx2_snd_queue *sq;
+	struct netdev_queue *txq;
+
+	/* Check for minimum packet length */
+	if (skb->len <= ETH_HLEN) {
+		dev_kfree_skb(skb);
+		return NETDEV_TX_OK;
+	}
+
+	sq = &pf->qset.sq[qidx];
+	txq = netdev_get_tx_queue(netdev, qidx);
+
+	if (!otx2_sq_append_skb(netdev, sq, skb, qidx)) {
+		netif_tx_stop_queue(txq);
+
+		/* Check again, incase SQBs got freed up */
+		smp_mb();
+		if (((sq->num_sqbs - *sq->aura_fc_addr) * sq->sqe_per_sqb)
+							> sq->sqe_thresh)
+			netif_tx_wake_queue(txq);
+
+		return NETDEV_TX_BUSY;
+	}
+
+	return NETDEV_TX_OK;
+}
+
 static const struct net_device_ops otx2_netdev_ops = {
 	.ndo_open		= otx2_open,
 	.ndo_stop		= otx2_stop,
+	.ndo_start_xmit		= otx2_xmit,
 };
 
 static int otx2_check_pf_usable(struct otx2_nic *nic)
@@ -910,7 +945,8 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	 */
 	pf->iommu_domain = iommu_get_domain_for_dev(dev);
 
-	netdev->hw_features = NETIF_F_RXCSUM;
+	netdev->hw_features = (NETIF_F_RXCSUM | NETIF_F_IP_CSUM |
+			       NETIF_F_IPV6_CSUM | NETIF_F_SG);
 	netdev->features |= netdev->hw_features;
 
 	netdev->hw_features |= NETIF_F_RXALL;
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_struct.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_struct.h
index dad73cb5e6f8..04a9f12be2e9 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_struct.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_struct.h
@@ -28,6 +28,21 @@ enum nix_send_ldtype {
 	NIX_SEND_LDTYPE_LDWB = 0x2,
 };
 
+/* CSUM offload */
+enum nix_sendl3type {
+	NIX_SENDL3TYPE_NONE = 0x0,
+	NIX_SENDL3TYPE_IP4 = 0x2,
+	NIX_SENDL3TYPE_IP4_CKSUM = 0x3,
+	NIX_SENDL3TYPE_IP6 = 0x4,
+};
+
+enum nix_sendl4type {
+	NIX_SENDL4TYPE_NONE,
+	NIX_SENDL4TYPE_TCP_CKSUM,
+	NIX_SENDL4TYPE_SCTP_CKSUM,
+	NIX_SENDL4TYPE_UDP_CKSUM,
+};
+
 /* NIX wqe/cqe types */
 enum nix_xqe_type {
 	NIX_XQE_TYPE_INVALID   = 0x0,
@@ -51,6 +66,20 @@ enum nix_subdc {
 	NIX_SUBDC_SOD  = 0xf,
 };
 
+/* Algorithm for nix_sqe_mem_s header (value of the `alg` field) */
+enum nix_sendmemalg {
+	NIX_SENDMEMALG_E_SET       = 0x0,
+	NIX_SENDMEMALG_E_SETTSTMP  = 0x1,
+	NIX_SENDMEMALG_E_SETRSLT   = 0x2,
+	NIX_SENDMEMALG_E_ADD       = 0x8,
+	NIX_SENDMEMALG_E_SUB       = 0x9,
+	NIX_SENDMEMALG_E_ADDLEN    = 0xa,
+	NIX_SENDMEMALG_E_SUBLEN    = 0xb,
+	NIX_SENDMEMALG_E_ADDMBUF   = 0xc,
+	NIX_SENDMEMALG_E_SUBMBUF   = 0xd,
+	NIX_SENDMEMALG_E_ENUM_LAST = 0xe,
+};
+
 /* NIX CQE header structure */
 struct nix_cqe_hdr_s {
 	u64 flow_tag              : 32;
@@ -130,10 +159,90 @@ struct nix_rx_sg_s {
 	u64 seg3_addr;
 };
 
+struct nix_send_comp_s {
+	u64 status	: 8;
+	u64 sqe_id	: 16;
+	u64 rsvd_24_63	: 40;
+};
+
 struct nix_cqe_rx_s {
 	struct nix_cqe_hdr_s  hdr;
 	struct nix_rx_parse_s parse;
 	struct nix_rx_sg_s sg;
 };
 
+struct nix_cqe_tx_s {
+	struct nix_cqe_hdr_s  hdr;
+	struct nix_send_comp_s comp;
+};
+
+/* NIX SQE header structure */
+struct nix_sqe_hdr_s {
+	u64 total		: 18; /* W0 */
+	u64 reserved_18		: 1;
+	u64 df			: 1;
+	u64 aura		: 20;
+	u64 sizem1		: 3;
+	u64 pnc			: 1;
+	u64 sq			: 20;
+	u64 ol3ptr		: 8; /* W1 */
+	u64 ol4ptr		: 8;
+	u64 il3ptr		: 8;
+	u64 il4ptr		: 8;
+	u64 ol3type		: 4;
+	u64 ol4type		: 4;
+	u64 il3type		: 4;
+	u64 il4type		: 4;
+	u64 sqe_id		: 16;
+
+};
+
+/* NIX send extended header subdescriptor structure */
+struct nix_sqe_ext_s {
+	u64 lso_mps       : 14; /* W0 */
+	u64 lso           : 1;
+	u64 tstmp         : 1;
+	u64 lso_sb        : 8;
+	u64 lso_format    : 5;
+	u64 rsvd_31_29    : 3;
+	u64 shp_chg       : 9;
+	u64 shp_dis       : 1;
+	u64 shp_ra        : 2;
+	u64 markptr       : 8;
+	u64 markform      : 7;
+	u64 mark_en       : 1;
+	u64 subdc         : 4;
+	u64 vlan0_ins_ptr : 8; /* W1 */
+	u64 vlan0_ins_tci : 16;
+	u64 vlan1_ins_ptr : 8;
+	u64 vlan1_ins_tci : 16;
+	u64 vlan0_ins_ena : 1;
+	u64 vlan1_ins_ena : 1;
+	u64 rsvd_127_114  : 14;
+};
+
+struct nix_sqe_sg_s {
+	u64 seg1_size	: 16;
+	u64 seg2_size	: 16;
+	u64 seg3_size	: 16;
+	u64 segs	: 2;
+	u64 rsvd_54_50	: 5;
+	u64 i1		: 1;
+	u64 i2		: 1;
+	u64 i3		: 1;
+	u64 ld_type	: 2;
+	u64 subdc	: 4;
+};
+
+/* NIX send memory subdescriptor structure */
+struct nix_sqe_mem_s {
+	u64 offset        : 16; /* W0 */
+	u64 rsvd_52_16    : 37;
+	u64 wmem          : 1;
+	u64 dsz           : 2;
+	u64 alg           : 4;
+	u64 subdc         : 4;
+	u64 addr; /* W1 */
+};
+
 #endif /* OTX2_STRUCT_H */
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c
index 0c1519a05281..f19f5d044110 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c
@@ -32,6 +32,74 @@ static struct nix_cqe_hdr_s *otx2_get_next_cqe(struct otx2_cq_queue *cq)
 	return cqe_hdr;
 }
 
+static unsigned int frag_num(unsigned int i)
+{
+#ifdef __BIG_ENDIAN
+	return (i & ~3) + 3 - (i & 3);
+#else
+	return i;
+#endif
+}
+
+static dma_addr_t otx2_dma_map_skb_frag(struct otx2_nic *pfvf,
+					struct sk_buff *skb, int seg, int *len)
+{
+	const skb_frag_t *frag;
+	struct page *page;
+	int offset;
+
+	/* First segment is always skb->data */
+	if (!seg) {
+		page = virt_to_page(skb->data);
+		offset = offset_in_page(skb->data);
+		*len = skb_headlen(skb);
+	} else {
+		frag = &skb_shinfo(skb)->frags[seg - 1];
+		page = skb_frag_page(frag);
+		offset = skb_frag_off(frag);
+		*len = skb_frag_size(frag);
+	}
+	return otx2_dma_map_page(pfvf, page, offset, *len, DMA_TO_DEVICE);
+}
+
+static void otx2_dma_unmap_skb_frags(struct otx2_nic *pfvf, struct sg_list *sg)
+{
+	int seg;
+
+	for (seg = 0; seg < sg->num_segs; seg++) {
+		otx2_dma_unmap_page(pfvf, sg->dma_addr[seg],
+				    sg->size[seg], DMA_TO_DEVICE);
+	}
+	sg->num_segs = 0;
+}
+
+static void otx2_snd_pkt_handler(struct otx2_nic *pfvf,
+				 struct otx2_cq_queue *cq,
+				 struct otx2_snd_queue *sq,
+				 struct nix_cqe_tx_s *cqe,
+				 int budget, int *tx_pkts, int *tx_bytes)
+{
+	struct nix_send_comp_s *snd_comp = &cqe->comp;
+	struct sk_buff *skb = NULL;
+	struct sg_list *sg;
+
+	if (unlikely(snd_comp->status))
+		net_err_ratelimited("%s: TX%d: Error in send CQ status:%x\n",
+				    pfvf->netdev->name, cq->cint_idx,
+				    snd_comp->status);
+
+	sg = &sq->sg[snd_comp->sqe_id];
+	skb = (struct sk_buff *)sg->skb;
+	if (unlikely(!skb))
+		return;
+
+	*tx_bytes += skb->len;
+	(*tx_pkts)++;
+	otx2_dma_unmap_skb_frags(pfvf, sg);
+	napi_consume_skb(skb, budget);
+	sg->skb = (u64)NULL;
+}
+
 static void otx2_skb_add_frag(struct otx2_nic *pfvf, struct sk_buff *skb,
 			      u64 iova, int len)
 {
@@ -182,7 +250,39 @@ static int otx2_rx_napi_handler(struct otx2_nic *pfvf,
 static int otx2_tx_napi_handler(struct otx2_nic *pfvf,
 				struct otx2_cq_queue *cq, int budget)
 {
-	 /* Nothing to do, for now */
+	int tx_pkts = 0, tx_bytes = 0;
+	struct nix_cqe_tx_s *cqe;
+	int processed_cqe = 0;
+
+	while (likely(processed_cqe < budget)) {
+		cqe = (struct nix_cqe_tx_s *)otx2_get_next_cqe(cq);
+		if (unlikely(!cqe)) {
+			if (!processed_cqe)
+				return 0;
+			break;
+		}
+		otx2_snd_pkt_handler(pfvf, cq, &pfvf->qset.sq[cq->cint_idx],
+				     cqe, budget, &tx_pkts, &tx_bytes);
+
+		cqe->hdr.cqe_type = NIX_XQE_TYPE_INVALID;
+		processed_cqe++;
+	}
+
+	/* Free CQEs to HW */
+	otx2_write64(pfvf, NIX_LF_CQ_OP_DOOR,
+		     ((u64)cq->cq_idx << 32) | processed_cqe);
+
+	if (likely(tx_pkts)) {
+		struct netdev_queue *txq;
+
+		txq = netdev_get_tx_queue(pfvf->netdev, cq->cint_idx);
+		netdev_tx_completed_queue(txq, tx_pkts, tx_bytes);
+		/* Check if queue was stopped earlier due to ring full */
+		smp_mb();
+		if (netif_tx_queue_stopped(txq) &&
+		    netif_carrier_ok(pfvf->netdev))
+			netif_tx_wake_queue(txq);
+	}
 	return 0;
 }
 
@@ -222,6 +322,169 @@ int otx2_napi_handler(struct napi_struct *napi, int budget)
 	return workdone;
 }
 
+static void otx2_sqe_flush(struct otx2_snd_queue *sq, int size)
+{
+	u64 status;
+
+	/* Packet data stores should finish before SQE is flushed to HW */
+	dma_wmb();
+
+	do {
+		memcpy(sq->lmt_addr, sq->sqe_base, size);
+		status = otx2_lmt_flush(sq->io_addr);
+	} while (status == 0);
+
+	sq->head++;
+	sq->head &= (sq->sqe_cnt - 1);
+}
+
+#define MAX_SEGS_PER_SG	3
+/* Add SQE scatter/gather subdescriptor structure */
+static bool otx2_sqe_add_sg(struct otx2_nic *pfvf, struct otx2_snd_queue *sq,
+			    struct sk_buff *skb, int num_segs, int *offset)
+{
+	struct nix_sqe_sg_s *sg = NULL;
+	u64 dma_addr, *iova = NULL;
+	u16 *sg_lens = NULL;
+	int seg, len;
+
+	sq->sg[sq->head].num_segs = 0;
+
+	for (seg = 0; seg < num_segs; seg++) {
+		if ((seg % MAX_SEGS_PER_SG) == 0) {
+			sg = (struct nix_sqe_sg_s *)(sq->sqe_base + *offset);
+			sg->ld_type = NIX_SEND_LDTYPE_LDD;
+			sg->subdc = NIX_SUBDC_SG;
+			sg->segs = 0;
+			sg_lens = (void *)sg;
+			iova = (void *)sg + sizeof(*sg);
+			/* Next subdc always starts at a 16byte boundary.
+			 * So if sg->segs is whether 2 or 3, offset += 16bytes.
+			 */
+			if ((num_segs - seg) >= (MAX_SEGS_PER_SG - 1))
+				*offset += sizeof(*sg) + (3 * sizeof(u64));
+			else
+				*offset += sizeof(*sg) + sizeof(u64);
+		}
+		dma_addr = otx2_dma_map_skb_frag(pfvf, skb, seg, &len);
+		if (dma_mapping_error(pfvf->dev, dma_addr))
+			return false;
+
+		sg_lens[frag_num(seg % MAX_SEGS_PER_SG)] = len;
+		sg->segs++;
+		*iova++ = dma_addr;
+
+		/* Save DMA mapping info for later unmapping */
+		sq->sg[sq->head].dma_addr[seg] = dma_addr;
+		sq->sg[sq->head].size[seg] = len;
+		sq->sg[sq->head].num_segs++;
+	}
+
+	sq->sg[sq->head].skb = (u64)skb;
+	return true;
+}
+
+/* Add SQE header subdescriptor structure */
+static void otx2_sqe_add_hdr(struct otx2_nic *pfvf, struct otx2_snd_queue *sq,
+			     struct nix_sqe_hdr_s *sqe_hdr,
+			     struct sk_buff *skb, u16 qidx)
+{
+	int proto = 0;
+
+	/* Check if SQE was framed before, if yes then no need to
+	 * set these constants again and again.
+	 */
+	if (!sqe_hdr->total) {
+		/* Don't free Tx buffers to Aura */
+		sqe_hdr->df = 1;
+		sqe_hdr->aura = sq->aura_id;
+		/* Post a CQE Tx after pkt transmission */
+		sqe_hdr->pnc = 1;
+		sqe_hdr->sq = qidx;
+	}
+	sqe_hdr->total = skb->len;
+	/* Set SQE identifier which will be used later for freeing SKB */
+	sqe_hdr->sqe_id = sq->head;
+
+	/* Offload TCP/UDP checksum to HW */
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
+		sqe_hdr->ol3ptr = skb_network_offset(skb);
+		sqe_hdr->ol4ptr = skb_transport_offset(skb);
+		/* get vlan protocol Ethertype */
+		if (eth_type_vlan(skb->protocol))
+			skb->protocol = vlan_get_protocol(skb);
+
+		if (skb->protocol == htons(ETH_P_IP)) {
+			proto = ip_hdr(skb)->protocol;
+			/* In case of TSO, HW needs this to be explicitly set.
+			 * So set this always, instead of adding a check.
+			 */
+			sqe_hdr->ol3type = NIX_SENDL3TYPE_IP4_CKSUM;
+		} else if (skb->protocol == htons(ETH_P_IPV6)) {
+			proto = ipv6_hdr(skb)->nexthdr;
+		}
+
+		if (proto == IPPROTO_TCP)
+			sqe_hdr->ol4type = NIX_SENDL4TYPE_TCP_CKSUM;
+		else if (proto == IPPROTO_UDP)
+			sqe_hdr->ol4type = NIX_SENDL4TYPE_UDP_CKSUM;
+	}
+}
+
+bool otx2_sq_append_skb(struct net_device *netdev, struct otx2_snd_queue *sq,
+			struct sk_buff *skb, u16 qidx)
+{
+	struct netdev_queue *txq = netdev_get_tx_queue(netdev, qidx);
+	struct otx2_nic *pfvf = netdev_priv(netdev);
+	int offset, num_segs, free_sqe;
+	struct nix_sqe_hdr_s *sqe_hdr;
+
+	/* Check if there is room for new SQE.
+	 * 'Num of SQBs freed to SQ's pool - SQ's Aura count'
+	 * will give free SQE count.
+	 */
+	free_sqe = (sq->num_sqbs - *sq->aura_fc_addr) * sq->sqe_per_sqb;
+
+	if (!free_sqe || free_sqe < sq->sqe_thresh)
+		return false;
+
+	num_segs = skb_shinfo(skb)->nr_frags + 1;
+
+	/* If SKB doesn't fit in a single SQE, linearize it.
+	 * TODO: Consider adding JUMP descriptor instead.
+	 */
+	if (unlikely(num_segs > OTX2_MAX_FRAGS_IN_SQE)) {
+		if (__skb_linearize(skb)) {
+			dev_kfree_skb_any(skb);
+			return true;
+		}
+		num_segs = skb_shinfo(skb)->nr_frags + 1;
+	}
+
+	/* Set SQE's SEND_HDR.
+	 * Do not clear the first 64bit as it contains constant info.
+	 */
+	memset(sq->sqe_base + 8, 0, sq->sqe_size - 8);
+	sqe_hdr = (struct nix_sqe_hdr_s *)(sq->sqe_base);
+	otx2_sqe_add_hdr(pfvf, sq, sqe_hdr, skb, qidx);
+	offset = sizeof(*sqe_hdr);
+
+	/* Add SG subdesc with data frags */
+	if (!otx2_sqe_add_sg(pfvf, sq, skb, num_segs, &offset)) {
+		otx2_dma_unmap_skb_frags(pfvf, &sq->sg[sq->head]);
+		return false;
+	}
+
+	sqe_hdr->sizem1 = (offset / 16) - 1;
+
+	netdev_tx_sent_queue(txq, skb->len);
+
+	/* Flush SQE to HW */
+	otx2_sqe_flush(sq, offset);
+
+	return true;
+}
+
 void otx2_cleanup_rx_cqes(struct otx2_nic *pfvf, struct otx2_cq_queue *cq)
 {
 	struct nix_cqe_rx_s *cqe;
@@ -242,3 +505,29 @@ void otx2_cleanup_rx_cqes(struct otx2_nic *pfvf, struct otx2_cq_queue *cq)
 	otx2_write64(pfvf, NIX_LF_CQ_OP_DOOR,
 		     ((u64)cq->cq_idx << 32) | processed_cqe);
 }
+
+void otx2_cleanup_tx_cqes(struct otx2_nic *pfvf, struct otx2_cq_queue *cq)
+{
+	struct sk_buff *skb = NULL;
+	struct otx2_snd_queue *sq;
+	struct nix_cqe_tx_s *cqe;
+	int processed_cqe = 0;
+	struct sg_list *sg;
+
+	sq = &pfvf->qset.sq[cq->cint_idx];
+
+	while ((cqe = (struct nix_cqe_tx_s *)otx2_get_next_cqe(cq))) {
+		sg = &sq->sg[cqe->comp.sqe_id];
+		skb = (struct sk_buff *)sg->skb;
+		if (skb) {
+			otx2_dma_unmap_skb_frags(pfvf, sg);
+			dev_kfree_skb_any(skb);
+			sg->skb = (u64)NULL;
+		}
+		processed_cqe++;
+	}
+
+	/* Free CQEs to HW */
+	otx2_write64(pfvf, NIX_LF_CQ_OP_DOOR,
+		     ((u64)cq->cq_idx << 32) | processed_cqe);
+}
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h
index 0944c176efa3..d9683c3bbc91 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h
@@ -22,6 +22,8 @@
 #define OTX2_DATA_ALIGN(X)	ALIGN(X, OTX2_ALIGN)
 #define OTX2_HEAD_ROOM		OTX2_ALIGN
 
+#define OTX2_MAX_FRAGS_IN_SQE	9
+
 /* Rx buffer size should be in multiples of 128bytes */
 #define RCV_FRAG_LEN1(x)				\
 		((OTX2_HEAD_ROOM + OTX2_DATA_ALIGN(x)) + \
@@ -53,17 +55,27 @@
  */
 #define CQ_QCOUNT_DEFAULT	1
 
+struct sg_list {
+	u16	num_segs;
+	u64	skb;
+	u64	size[OTX2_MAX_FRAGS_IN_SQE];
+	u64	dma_addr[OTX2_MAX_FRAGS_IN_SQE];
+};
+
 struct otx2_snd_queue {
 	u8			aura_id;
+	u16			head;
 	u16			sqe_size;
 	u32			sqe_cnt;
 	u16			num_sqbs;
+	u16			sqe_thresh;
 	u8			sqe_per_sqb;
 	u64			 io_addr;
 	u64			*aura_fc_addr;
 	u64			*lmt_addr;
 	void			*sqe_base;
 	struct qmem		*sqe;
+	struct sg_list		*sg;
 	u16			sqb_count;
 	u64			*sqb_ptrs;
 } ____cacheline_aligned_in_smp;
@@ -127,4 +139,6 @@ static inline u64 otx2_iova_to_phys(void *iommu_domain, dma_addr_t dma_addr)
 }
 
 int otx2_napi_handler(struct napi_struct *napi, int budget);
+bool otx2_sq_append_skb(struct net_device *netdev, struct otx2_snd_queue *sq,
+			struct sk_buff *skb, u16 qidx);
 #endif /* OTX2_TXRX_H */
-- 
cgit v1.2.3-59-g8ed1b


From 50fe6c02e5ad50727fb1069bb31099385fb59883 Mon Sep 17 00:00:00 2001
From: Linu Cherian <lcherian@marvell.com>
Date: Mon, 27 Jan 2020 18:35:22 +0530
Subject: octeontx2-pf: Register and handle link notifications

PF and AF (admin function) shares 64KB of reserved memory region for
communication. This region is shared for
 - Messages sent by PF and responses sent by AF.
 - Notifications sent by AF and ACKs sent by PF.

This patch adds infrastructure to handle notifications sent
by AF and adds handlers to process them.

One of the main usecase of notifications from AF is physical
link changes. So this patch adds registration of PF with AF
to receive link status change notifications and also adds
the handler for that notification.

Signed-off-by: Linu Cherian <lcherian@marvell.com>
Signed-off-by: Tomasz Duszynski <tduszynski@marvell.com>
Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../ethernet/marvell/octeontx2/nic/otx2_common.h   |  5 ++
 .../net/ethernet/marvell/octeontx2/nic/otx2_pf.c   | 79 ++++++++++++++++++++++
 .../net/ethernet/marvell/octeontx2/nic/otx2_txrx.c | 25 +++++++
 3 files changed, 109 insertions(+)

diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
index 11948540dd4a..16d7dd70a2af 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
@@ -135,6 +135,9 @@ struct otx2_nic {
 	void			*iommu_domain;
 	u16			rbsize; /* Receive buffer size */
 
+#define OTX2_FLAG_INTF_DOWN			BIT_ULL(2)
+	u64			flags;
+
 	struct otx2_qset	qset;
 	struct otx2_hw		hw;
 	struct pci_dev		*pdev;
@@ -145,6 +148,7 @@ struct otx2_nic {
 	struct workqueue_struct *mbox_wq;
 
 	u16			pcifunc; /* RVU PF_FUNC */
+	struct cgx_link_user_info linfo;
 
 	/* Block address of NIX either BLKADDR_NIX0 or BLKADDR_NIX1 */
 	int			nix_blkaddr;
@@ -485,6 +489,7 @@ int otx2_txschq_stop(struct otx2_nic *pfvf);
 void otx2_sqb_flush(struct otx2_nic *pfvf);
 dma_addr_t otx2_alloc_rbuf(struct otx2_nic *pfvf, struct otx2_pool *pool,
 			   gfp_t gfp);
+int otx2_rxtx_enable(struct otx2_nic *pfvf, bool enable);
 void otx2_ctx_disable(struct mbox *mbox, int type, bool npa);
 void otx2_cleanup_rx_cqes(struct otx2_nic *pfvf, struct otx2_cq_queue *cq);
 void otx2_cleanup_tx_cqes(struct otx2_nic *pfvf, struct otx2_cq_queue *cq);
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
index a18bf5facaaf..8cb1cf56d388 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
@@ -167,6 +167,38 @@ static void otx2_pfaf_mbox_handler(struct work_struct *work)
 	otx2_mbox_reset(mbox, 0);
 }
 
+static void otx2_handle_link_event(struct otx2_nic *pf)
+{
+	struct cgx_link_user_info *linfo = &pf->linfo;
+	struct net_device *netdev = pf->netdev;
+
+	pr_info("%s NIC Link is %s %d Mbps %s duplex\n", netdev->name,
+		linfo->link_up ? "UP" : "DOWN", linfo->speed,
+		linfo->full_duplex ? "Full" : "Half");
+	if (linfo->link_up) {
+		netif_carrier_on(netdev);
+		netif_tx_start_all_queues(netdev);
+	} else {
+		netif_tx_stop_all_queues(netdev);
+		netif_carrier_off(netdev);
+	}
+}
+
+int otx2_mbox_up_handler_cgx_link_event(struct otx2_nic *pf,
+					struct cgx_link_info_msg *msg,
+					struct msg_rsp *rsp)
+{
+	/* Copy the link info sent by AF */
+	pf->linfo = msg->link_info;
+
+	/* interface has not been fully configured yet */
+	if (pf->flags & OTX2_FLAG_INTF_DOWN)
+		return 0;
+
+	otx2_handle_link_event(pf);
+	return 0;
+}
+
 static int otx2_process_mbox_msg_up(struct otx2_nic *pf,
 				    struct mbox_msghdr *req)
 {
@@ -367,6 +399,27 @@ exit:
 	return err;
 }
 
+static int otx2_cgx_config_linkevents(struct otx2_nic *pf, bool enable)
+{
+	struct msg_req *msg;
+	int err;
+
+	otx2_mbox_lock(&pf->mbox);
+	if (enable)
+		msg = otx2_mbox_alloc_msg_cgx_start_linkevents(&pf->mbox);
+	else
+		msg = otx2_mbox_alloc_msg_cgx_stop_linkevents(&pf->mbox);
+
+	if (!msg) {
+		otx2_mbox_unlock(&pf->mbox);
+		return -ENOMEM;
+	}
+
+	err = otx2_sync_mbox_msg(&pf->mbox);
+	otx2_mbox_unlock(&pf->mbox);
+	return err;
+}
+
 static int otx2_set_real_num_queues(struct net_device *netdev,
 				    int tx_queues, int rx_queues)
 {
@@ -690,6 +743,18 @@ static int otx2_open(struct net_device *netdev)
 
 	otx2_set_cints_affinity(pf);
 
+	pf->flags &= ~OTX2_FLAG_INTF_DOWN;
+	/* 'intf_down' may be checked on any cpu */
+	smp_wmb();
+
+	/* we have already received link status notification */
+	if (pf->linfo.link_up && !(pf->pcifunc & RVU_PFVF_FUNC_MASK))
+		otx2_handle_link_event(pf);
+
+	err = otx2_rxtx_enable(pf, true);
+	if (err)
+		goto err_free_cints;
+
 	return 0;
 
 err_free_cints:
@@ -713,6 +778,13 @@ static int otx2_stop(struct net_device *netdev)
 	netif_carrier_off(netdev);
 	netif_tx_stop_all_queues(netdev);
 
+	pf->flags |= OTX2_FLAG_INTF_DOWN;
+	/* 'intf_down' may be checked on any cpu */
+	smp_wmb();
+
+	/* First stop packet Rx/Tx */
+	otx2_rxtx_enable(pf, false);
+
 	/* Cleanup CQ NAPI and IRQ */
 	vec = pf->hw.nix_msixoff + NIX_LF_CINT_VEC_START;
 	for (qidx = 0; qidx < pf->hw.cint_cnt; qidx++) {
@@ -867,6 +939,7 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	pf->netdev = netdev;
 	pf->pdev = pdev;
 	pf->dev = dev;
+	pf->flags |= OTX2_FLAG_INTF_DOWN;
 
 	hw = &pf->hw;
 	hw->pdev = pdev;
@@ -959,6 +1032,9 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 		goto err_detach_rsrc;
 	}
 
+	/* Enable link notifications */
+	otx2_cgx_config_linkevents(pf, true);
+
 	return 0;
 
 err_detach_rsrc:
@@ -987,6 +1063,9 @@ static void otx2_remove(struct pci_dev *pdev)
 
 	pf = netdev_priv(netdev);
 
+	/* Disable link notifications */
+	otx2_cgx_config_linkevents(pf, false);
+
 	unregister_netdev(netdev);
 	otx2_detach_resources(&pf->mbox);
 	otx2_disable_mbox_intr(pf);
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c
index f19f5d044110..87b579a121ab 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c
@@ -315,6 +315,10 @@ int otx2_napi_handler(struct napi_struct *napi, int budget)
 	otx2_write64(pfvf, NIX_LF_CINTX_INT(cq_poll->cint_idx), BIT_ULL(0));
 
 	if (workdone < budget && napi_complete_done(napi, workdone)) {
+		/* If interface is going down, don't re-enable IRQ */
+		if (pfvf->flags & OTX2_FLAG_INTF_DOWN)
+			return workdone;
+
 		/* Re-enable interrupts */
 		otx2_write64(pfvf, NIX_LF_CINTX_ENA_W1S(cq_poll->cint_idx),
 			     BIT_ULL(0));
@@ -531,3 +535,24 @@ void otx2_cleanup_tx_cqes(struct otx2_nic *pfvf, struct otx2_cq_queue *cq)
 	otx2_write64(pfvf, NIX_LF_CQ_OP_DOOR,
 		     ((u64)cq->cq_idx << 32) | processed_cqe);
 }
+
+int otx2_rxtx_enable(struct otx2_nic *pfvf, bool enable)
+{
+	struct msg_req *msg;
+	int err;
+
+	otx2_mbox_lock(&pfvf->mbox);
+	if (enable)
+		msg = otx2_mbox_alloc_msg_nix_lf_start_rx(&pfvf->mbox);
+	else
+		msg = otx2_mbox_alloc_msg_nix_lf_stop_rx(&pfvf->mbox);
+
+	if (!msg) {
+		otx2_mbox_unlock(&pfvf->mbox);
+		return -ENOMEM;
+	}
+
+	err = otx2_sync_mbox_msg(&pfvf->mbox);
+	otx2_mbox_unlock(&pfvf->mbox);
+	return err;
+}
-- 
cgit v1.2.3-59-g8ed1b


From 34bfe0ebedb76aeb964341d96cf7c5135d14e1bf Mon Sep 17 00:00:00 2001
From: Sunil Goutham <sgoutham@marvell.com>
Date: Mon, 27 Jan 2020 18:35:23 +0530
Subject: octeontx2-pf: MTU, MAC and RX mode config support

This patch addes support to change interface MTU, MAC address
retrieval and config, RX mode ie unicast, multicast and promiscuous.
Also added link loopback support

Signed-off-by: Tomasz Duszynski <tduszynski@marvell.com>
Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com>
Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/marvell/octeontx2/af/mbox.h   |   8 +-
 .../net/ethernet/marvell/octeontx2/af/rvu_nix.c    |  17 ++++
 .../ethernet/marvell/octeontx2/nic/otx2_common.c   | 108 +++++++++++++++++++++
 .../ethernet/marvell/octeontx2/nic/otx2_common.h   |   8 +-
 .../net/ethernet/marvell/octeontx2/nic/otx2_pf.c   | 103 +++++++++++++++++++-
 .../net/ethernet/marvell/octeontx2/nic/otx2_txrx.h |   4 +
 6 files changed, 242 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
index a589748f1240..8bbc1f1d81f5 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
@@ -210,7 +210,8 @@ M(NIX_SET_RX_CFG,	0x8010, nix_set_rx_cfg, nix_rx_cfg, msg_rsp)	\
 M(NIX_LSO_FORMAT_CFG,	0x8011, nix_lso_format_cfg,			\
 				 nix_lso_format_cfg,			\
 				 nix_lso_format_cfg_rsp)		\
-M(NIX_RXVLAN_ALLOC,	0x8012, nix_rxvlan_alloc, msg_req, msg_rsp)
+M(NIX_RXVLAN_ALLOC,	0x8012, nix_rxvlan_alloc, msg_req, msg_rsp)	\
+M(NIX_GET_MAC_ADDR, 0x8018, nix_get_mac_addr, msg_req, nix_get_mac_addr_rsp) \
 
 /* Messages initiated by AF (range 0xC00 - 0xDFF) */
 #define MBOX_UP_CGX_MESSAGES						\
@@ -618,6 +619,11 @@ struct nix_set_mac_addr {
 	u8 mac_addr[ETH_ALEN]; /* MAC address to be set for this pcifunc */
 };
 
+struct nix_get_mac_addr_rsp {
+	struct mbox_msghdr hdr;
+	u8 mac_addr[ETH_ALEN];
+};
+
 struct nix_mark_format_cfg {
 	struct mbox_msghdr hdr;
 	u8 offset;
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
index 8a59f7d53fbf..eb5e542424e7 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
@@ -2546,6 +2546,23 @@ int rvu_mbox_handler_nix_set_mac_addr(struct rvu *rvu,
 	return 0;
 }
 
+int rvu_mbox_handler_nix_get_mac_addr(struct rvu *rvu,
+				      struct msg_req *req,
+				      struct nix_get_mac_addr_rsp *rsp)
+{
+	u16 pcifunc = req->hdr.pcifunc;
+	struct rvu_pfvf *pfvf;
+
+	if (!is_nixlf_attached(rvu, pcifunc))
+		return NIX_AF_ERR_AF_LF_INVALID;
+
+	pfvf = rvu_get_pfvf(rvu, pcifunc);
+
+	ether_addr_copy(rsp->mac_addr, pfvf->mac_addr);
+
+	return 0;
+}
+
 int rvu_mbox_handler_nix_set_rx_mode(struct rvu *rvu, struct nix_rx_mode *req,
 				     struct msg_rsp *rsp)
 {
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
index e9bd2d6f6ebc..836b3c860e62 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
@@ -15,6 +15,97 @@
 #include "otx2_common.h"
 #include "otx2_struct.h"
 
+/* Sync MAC address with RVU AF */
+static int otx2_hw_set_mac_addr(struct otx2_nic *pfvf, u8 *mac)
+{
+	struct nix_set_mac_addr *req;
+	int err;
+
+	otx2_mbox_lock(&pfvf->mbox);
+	req = otx2_mbox_alloc_msg_nix_set_mac_addr(&pfvf->mbox);
+	if (!req) {
+		otx2_mbox_unlock(&pfvf->mbox);
+		return -ENOMEM;
+	}
+
+	ether_addr_copy(req->mac_addr, mac);
+
+	err = otx2_sync_mbox_msg(&pfvf->mbox);
+	otx2_mbox_unlock(&pfvf->mbox);
+	return err;
+}
+
+static int otx2_hw_get_mac_addr(struct otx2_nic *pfvf,
+				struct net_device *netdev)
+{
+	struct nix_get_mac_addr_rsp *rsp;
+	struct mbox_msghdr *msghdr;
+	struct msg_req *req;
+	int err;
+
+	otx2_mbox_lock(&pfvf->mbox);
+	req = otx2_mbox_alloc_msg_nix_get_mac_addr(&pfvf->mbox);
+	if (!req) {
+		otx2_mbox_unlock(&pfvf->mbox);
+		return -ENOMEM;
+	}
+
+	err = otx2_sync_mbox_msg(&pfvf->mbox);
+	if (err) {
+		otx2_mbox_unlock(&pfvf->mbox);
+		return err;
+	}
+
+	msghdr = otx2_mbox_get_rsp(&pfvf->mbox.mbox, 0, &req->hdr);
+	if (!msghdr) {
+		otx2_mbox_unlock(&pfvf->mbox);
+		return -ENOMEM;
+	}
+	rsp = (struct nix_get_mac_addr_rsp *)msghdr;
+	ether_addr_copy(netdev->dev_addr, rsp->mac_addr);
+	otx2_mbox_unlock(&pfvf->mbox);
+
+	return 0;
+}
+
+int otx2_set_mac_address(struct net_device *netdev, void *p)
+{
+	struct otx2_nic *pfvf = netdev_priv(netdev);
+	struct sockaddr *addr = p;
+
+	if (!is_valid_ether_addr(addr->sa_data))
+		return -EADDRNOTAVAIL;
+
+	if (!otx2_hw_set_mac_addr(pfvf, addr->sa_data))
+		memcpy(netdev->dev_addr, addr->sa_data, netdev->addr_len);
+	else
+		return -EPERM;
+
+	return 0;
+}
+
+int otx2_hw_set_mtu(struct otx2_nic *pfvf, int mtu)
+{
+	struct nix_frs_cfg *req;
+	int err;
+
+	otx2_mbox_lock(&pfvf->mbox);
+	req = otx2_mbox_alloc_msg_nix_set_hw_frs(&pfvf->mbox);
+	if (!req) {
+		otx2_mbox_unlock(&pfvf->mbox);
+		return -ENOMEM;
+	}
+
+	/* SMQ config limits maximum pkt size that can be transmitted */
+	req->update_smq = true;
+	pfvf->max_frs = mtu +  OTX2_ETH_HLEN;
+	req->maxlen = pfvf->max_frs;
+
+	err = otx2_sync_mbox_msg(&pfvf->mbox);
+	otx2_mbox_unlock(&pfvf->mbox);
+	return err;
+}
+
 void otx2_config_irq_coalescing(struct otx2_nic *pfvf, int qidx)
 {
 	/* Configure CQE interrupt coalescing parameters
@@ -63,6 +154,20 @@ ret:
 	return iova;
 }
 
+void otx2_get_mac_from_af(struct net_device *netdev)
+{
+	struct otx2_nic *pfvf = netdev_priv(netdev);
+	int err;
+
+	err = otx2_hw_get_mac_addr(pfvf, netdev);
+	if (err)
+		dev_warn(pfvf->dev, "Failed to read mac from hardware\n");
+
+	/* If AF doesn't provide a valid MAC, generate a random one */
+	if (!is_valid_ether_addr(netdev->dev_addr))
+		eth_hw_addr_random(netdev);
+}
+
 static int otx2_get_link(struct otx2_nic *pfvf)
 {
 	int link = 0;
@@ -97,6 +202,9 @@ int otx2_txschq_config(struct otx2_nic *pfvf, int lvl)
 	/* Set topology e.t.c configuration */
 	if (lvl == NIX_TXSCH_LVL_SMQ) {
 		req->reg[0] = NIX_AF_SMQX_CFG(schq);
+		req->regval[0] = ((pfvf->netdev->mtu  + OTX2_ETH_HLEN) << 8) |
+				   OTX2_MIN_MTU;
+
 		req->regval[0] |= (0x20ULL << 51) | (0x80ULL << 39) |
 				  (0x2ULL << 36);
 		req->num_regs++;
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
index 16d7dd70a2af..8c29b63b319e 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
@@ -133,6 +133,7 @@ struct otx2_nic {
 	void __iomem		*reg_base;
 	struct net_device	*netdev;
 	void			*iommu_domain;
+	u16			max_frs;
 	u16			rbsize; /* Receive buffer size */
 
 #define OTX2_FLAG_INTF_DOWN			BIT_ULL(2)
@@ -469,7 +470,9 @@ static inline void otx2_dma_unmap_page(struct otx2_nic *pfvf,
 /* MSI-X APIs */
 void otx2_free_cints(struct otx2_nic *pfvf, int n);
 void otx2_set_cints_affinity(struct otx2_nic *pfvf);
-
+int otx2_set_mac_address(struct net_device *netdev, void *p);
+int otx2_hw_set_mtu(struct otx2_nic *pfvf, int mtu);
+void otx2_get_mac_from_af(struct net_device *netdev);
 void otx2_config_irq_coalescing(struct otx2_nic *pfvf, int qidx);
 
 /* RVU block related APIs */
@@ -503,4 +506,7 @@ void mbox_handler_nix_lf_alloc(struct otx2_nic *pfvf,
 			       struct nix_lf_alloc_rsp *rsp);
 void mbox_handler_nix_txsch_alloc(struct otx2_nic *pf,
 				  struct nix_txsch_alloc_rsp *rsp);
+
+int otx2_open(struct net_device *netdev);
+int otx2_stop(struct net_device *netdev);
 #endif /* OTX2_COMMON_H */
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
index 8cb1cf56d388..60a67b90f2b0 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
@@ -43,6 +43,24 @@ enum {
 	TYPE_PFVF,
 };
 
+static int otx2_change_mtu(struct net_device *netdev, int new_mtu)
+{
+	bool if_up = netif_running(netdev);
+	int err = 0;
+
+	if (if_up)
+		otx2_stop(netdev);
+
+	netdev_info(netdev, "Changing MTU from %d to %d\n",
+		    netdev->mtu, new_mtu);
+	netdev->mtu = new_mtu;
+
+	if (if_up)
+		err = otx2_open(netdev);
+
+	return err;
+}
+
 static void otx2_queue_work(struct mbox *mw, struct workqueue_struct *mbox_wq,
 			    int first, int mdevs, u64 intr, int type)
 {
@@ -420,6 +438,27 @@ static int otx2_cgx_config_linkevents(struct otx2_nic *pf, bool enable)
 	return err;
 }
 
+static int otx2_cgx_config_loopback(struct otx2_nic *pf, bool enable)
+{
+	struct msg_req *msg;
+	int err;
+
+	otx2_mbox_lock(&pf->mbox);
+	if (enable)
+		msg = otx2_mbox_alloc_msg_cgx_intlbk_enable(&pf->mbox);
+	else
+		msg = otx2_mbox_alloc_msg_cgx_intlbk_disable(&pf->mbox);
+
+	if (!msg) {
+		otx2_mbox_unlock(&pf->mbox);
+		return -ENOMEM;
+	}
+
+	err = otx2_sync_mbox_msg(&pf->mbox);
+	otx2_mbox_unlock(&pf->mbox);
+	return err;
+}
+
 static int otx2_set_real_num_queues(struct net_device *netdev,
 				    int tx_queues, int rx_queues)
 {
@@ -519,7 +558,7 @@ static int otx2_init_hw_resources(struct otx2_nic *pf)
 	hw->pool_cnt = hw->rqpool_cnt + hw->sqpool_cnt;
 
 	/* Get the size of receive buffers to allocate */
-	pf->rbsize = RCV_FRAG_LEN(pf->netdev->mtu);
+	pf->rbsize = RCV_FRAG_LEN(pf->netdev->mtu + OTX2_ETH_HLEN);
 
 	otx2_mbox_lock(mbox);
 	/* NPA init */
@@ -658,7 +697,7 @@ static void otx2_free_hw_resources(struct otx2_nic *pf)
 	otx2_mbox_unlock(mbox);
 }
 
-static int otx2_open(struct net_device *netdev)
+int otx2_open(struct net_device *netdev)
 {
 	struct otx2_nic *pf = netdev_priv(netdev);
 	struct otx2_cq_poll *cq_poll = NULL;
@@ -715,6 +754,11 @@ static int otx2_open(struct net_device *netdev)
 		napi_enable(&cq_poll->napi);
 	}
 
+	/* Set maximum frame size allowed in HW */
+	err = otx2_hw_set_mtu(pf, netdev->mtu);
+	if (err)
+		goto err_disable_napi;
+
 	/* Register CQ IRQ handlers */
 	vec = pf->hw.nix_msixoff + NIX_LF_CINT_VEC_START;
 	for (qidx = 0; qidx < pf->hw.cint_cnt; qidx++) {
@@ -759,6 +803,7 @@ static int otx2_open(struct net_device *netdev)
 
 err_free_cints:
 	otx2_free_cints(pf, qidx);
+err_disable_napi:
 	otx2_disable_napi(pf);
 	otx2_free_hw_resources(pf);
 err_free_mem:
@@ -768,7 +813,7 @@ err_free_mem:
 	return err;
 }
 
-static int otx2_stop(struct net_device *netdev)
+int otx2_stop(struct net_device *netdev)
 {
 	struct otx2_nic *pf = netdev_priv(netdev);
 	struct otx2_cq_poll *cq_poll = NULL;
@@ -847,10 +892,53 @@ static netdev_tx_t otx2_xmit(struct sk_buff *skb, struct net_device *netdev)
 	return NETDEV_TX_OK;
 }
 
+static void otx2_set_rx_mode(struct net_device *netdev)
+{
+	struct otx2_nic *pf = netdev_priv(netdev);
+	struct nix_rx_mode *req;
+
+	if (!(netdev->flags & IFF_UP))
+		return;
+
+	otx2_mbox_lock(&pf->mbox);
+	req = otx2_mbox_alloc_msg_nix_set_rx_mode(&pf->mbox);
+	if (!req) {
+		otx2_mbox_unlock(&pf->mbox);
+		return;
+	}
+
+	req->mode = NIX_RX_MODE_UCAST;
+
+	/* We don't support MAC address filtering yet */
+	if (netdev->flags & IFF_PROMISC)
+		req->mode |= NIX_RX_MODE_PROMISC;
+	else if (netdev->flags & (IFF_ALLMULTI | IFF_MULTICAST))
+		req->mode |= NIX_RX_MODE_ALLMULTI;
+
+	otx2_sync_mbox_msg(&pf->mbox);
+	otx2_mbox_unlock(&pf->mbox);
+}
+
+static int otx2_set_features(struct net_device *netdev,
+			     netdev_features_t features)
+{
+	netdev_features_t changed = features ^ netdev->features;
+	struct otx2_nic *pf = netdev_priv(netdev);
+
+	if ((changed & NETIF_F_LOOPBACK) && netif_running(netdev))
+		return otx2_cgx_config_loopback(pf,
+						features & NETIF_F_LOOPBACK);
+	return 0;
+}
+
 static const struct net_device_ops otx2_netdev_ops = {
 	.ndo_open		= otx2_open,
 	.ndo_stop		= otx2_stop,
 	.ndo_start_xmit		= otx2_xmit,
+	.ndo_set_mac_address    = otx2_set_mac_address,
+	.ndo_change_mtu		= otx2_change_mtu,
+	.ndo_set_rx_mode	= otx2_set_rx_mode,
+	.ndo_set_features	= otx2_set_features,
 };
 
 static int otx2_check_pf_usable(struct otx2_nic *nic)
@@ -1005,6 +1093,9 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 
 	otx2_setup_dev_hw_settings(pf);
 
+	/* Assign default mac address */
+	otx2_get_mac_from_af(netdev);
+
 	/* NPA's pool is a stack to which SW frees buffer pointers via Aura.
 	 * HW allocates buffer pointer from stack and uses it for DMA'ing
 	 * ingress packet. In some scenarios HW can free back allocated buffer
@@ -1022,10 +1113,14 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 			       NETIF_F_IPV6_CSUM | NETIF_F_SG);
 	netdev->features |= netdev->hw_features;
 
-	netdev->hw_features |= NETIF_F_RXALL;
+	netdev->hw_features |= NETIF_F_LOOPBACK | NETIF_F_RXALL;
 
 	netdev->netdev_ops = &otx2_netdev_ops;
 
+	/* MTU range: 64 - 9190 */
+	netdev->min_mtu = OTX2_MIN_MTU;
+	netdev->max_mtu = OTX2_MAX_MTU;
+
 	err = register_netdev(netdev);
 	if (err) {
 		dev_err(dev, "Failed to register netdevice\n");
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h
index d9683c3bbc91..bad225979aec 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h
@@ -22,6 +22,10 @@
 #define OTX2_DATA_ALIGN(X)	ALIGN(X, OTX2_ALIGN)
 #define OTX2_HEAD_ROOM		OTX2_ALIGN
 
+#define	OTX2_ETH_HLEN		(VLAN_ETH_HLEN + VLAN_HLEN)
+#define	OTX2_MIN_MTU		64
+#define	OTX2_MAX_MTU		(9212 - OTX2_ETH_HLEN)
+
 #define OTX2_MAX_FRAGS_IN_SQE	9
 
 /* Rx buffer size should be in multiples of 128bytes */
-- 
cgit v1.2.3-59-g8ed1b


From 4ff7d1488a8496fddcfdf40c42489eaa64892f3e Mon Sep 17 00:00:00 2001
From: Geetha sowjanya <gakula@marvell.com>
Date: Mon, 27 Jan 2020 18:35:24 +0530
Subject: octeontx2-pf: Error handling support

HW reports many errors on the receive and transmit paths.
Such as incorrect queue configuration, pkt transmission errors,
LMTST instruction errors, transmit queue full etc. These are reported
via QINT interrupt. Most of the errors are fatal and needs
reinitialization.

Also added support to allocate receive buffers in non-atomic context
when allocation fails in NAPI context.

Signed-off-by: Geetha sowjanya <gakula@marvell.com>
Signed-off-by: Aleksey Makarov <amakarov@marvell.com>
Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../ethernet/marvell/octeontx2/nic/otx2_common.c   |  63 ++++++++++
 .../ethernet/marvell/octeontx2/nic/otx2_common.h   |  13 ++
 .../net/ethernet/marvell/octeontx2/nic/otx2_pf.c   | 133 ++++++++++++++++++++-
 .../ethernet/marvell/octeontx2/nic/otx2_struct.h   |  28 +++++
 .../net/ethernet/marvell/octeontx2/nic/otx2_txrx.c |  20 +++-
 .../net/ethernet/marvell/octeontx2/nic/otx2_txrx.h |   1 +
 6 files changed, 255 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
index 836b3c860e62..2b503066d801 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
@@ -154,6 +154,13 @@ ret:
 	return iova;
 }
 
+void otx2_tx_timeout(struct net_device *netdev, unsigned int txq)
+{
+	struct otx2_nic *pfvf = netdev_priv(netdev);
+
+	schedule_work(&pfvf->reset_task);
+}
+
 void otx2_get_mac_from_af(struct net_device *netdev)
 {
 	struct otx2_nic *pfvf = netdev_priv(netdev);
@@ -362,6 +369,7 @@ static int otx2_rq_init(struct otx2_nic *pfvf, u16 qidx, u16 lpb_aura)
 	aq->rq.lpb_sizem1 = (DMA_BUFFER_LEN(pfvf->rbsize) / 8) - 1;
 	aq->rq.xqe_imm_size = 0; /* Copying of packet to CQE not needed */
 	aq->rq.flow_tagw = 32; /* Copy full 32bit flow_tag to CQE header */
+	aq->rq.qint_idx = 0;
 	aq->rq.lpb_drop_ena = 1; /* Enable RED dropping for AURA */
 	aq->rq.xqe_drop_ena = 1; /* Enable RED dropping for CQ/SSO */
 	aq->rq.xqe_pass = RQ_PASS_LVL_CQ(pfvf->hw.rq_skid, qset->rqe_cnt);
@@ -424,6 +432,8 @@ static int otx2_sq_init(struct otx2_nic *pfvf, u16 qidx, u16 sqb_aura)
 	aq->sq.default_chan = pfvf->hw.tx_chan_base;
 	aq->sq.sqe_stype = NIX_STYPE_STF; /* Cache SQB */
 	aq->sq.sqb_aura = sqb_aura;
+	aq->sq.sq_int_ena = NIX_SQINT_BITS;
+	aq->sq.qint_idx = 0;
 	/* Due pipelining impact minimum 2000 unused SQ CQE's
 	 * need to maintain to avoid CQ overflow.
 	 */
@@ -470,6 +480,7 @@ static int otx2_cq_init(struct otx2_nic *pfvf, u16 qidx)
 	pool_id = ((cq->cq_type == CQ_RX) &&
 		   (pfvf->hw.rqpool_cnt != pfvf->hw.rx_queues)) ? 0 : qidx;
 	cq->rbpool = &qset->pool[pool_id];
+	cq->refill_task_sched = false;
 
 	/* Get memory to put this msg */
 	aq = otx2_mbox_alloc_msg_nix_aq_enq(&pfvf->mbox);
@@ -481,6 +492,8 @@ static int otx2_cq_init(struct otx2_nic *pfvf, u16 qidx)
 	aq->cq.caching = 1;
 	aq->cq.base = cq->cqe->iova;
 	aq->cq.cint_idx = cq->cint_idx;
+	aq->cq.cq_err_int_ena = NIX_CQERRINT_BITS;
+	aq->cq.qint_idx = 0;
 	aq->cq.avg_level = 255;
 
 	if (qidx < pfvf->hw.rx_queues) {
@@ -496,6 +509,45 @@ static int otx2_cq_init(struct otx2_nic *pfvf, u16 qidx)
 	return otx2_sync_mbox_msg(&pfvf->mbox);
 }
 
+static void otx2_pool_refill_task(struct work_struct *work)
+{
+	struct otx2_cq_queue *cq;
+	struct otx2_pool *rbpool;
+	struct refill_work *wrk;
+	int qidx, free_ptrs = 0;
+	struct otx2_nic *pfvf;
+	s64 bufptr;
+
+	wrk = container_of(work, struct refill_work, pool_refill_work.work);
+	pfvf = wrk->pf;
+	qidx = wrk - pfvf->refill_wrk;
+	cq = &pfvf->qset.cq[qidx];
+	rbpool = cq->rbpool;
+	free_ptrs = cq->pool_ptrs;
+
+	while (cq->pool_ptrs) {
+		bufptr = otx2_alloc_rbuf(pfvf, rbpool, GFP_KERNEL);
+		if (bufptr <= 0) {
+			/* Schedule a WQ if we fails to free atleast half of the
+			 * pointers else enable napi for this RQ.
+			 */
+			if (!((free_ptrs - cq->pool_ptrs) > free_ptrs / 2)) {
+				struct delayed_work *dwork;
+
+				dwork = &wrk->pool_refill_work;
+				schedule_delayed_work(dwork,
+						      msecs_to_jiffies(100));
+			} else {
+				cq->refill_task_sched = false;
+			}
+			return;
+		}
+		otx2_aura_freeptr(pfvf, qidx, bufptr + OTX2_HEAD_ROOM);
+		cq->pool_ptrs--;
+	}
+	cq->refill_task_sched = false;
+}
+
 int otx2_config_nix_queues(struct otx2_nic *pfvf)
 {
 	int qidx, err;
@@ -525,6 +577,17 @@ int otx2_config_nix_queues(struct otx2_nic *pfvf)
 			return err;
 	}
 
+	/* Initialize work queue for receive buffer refill */
+	pfvf->refill_wrk = devm_kcalloc(pfvf->dev, pfvf->qset.cq_cnt,
+					sizeof(struct refill_work), GFP_KERNEL);
+	if (!pfvf->refill_wrk)
+		return -ENOMEM;
+
+	for (qidx = 0; qidx < pfvf->qset.cq_cnt; qidx++) {
+		pfvf->refill_wrk[qidx].pf = pfvf;
+		INIT_DELAYED_WORK(&pfvf->refill_wrk[qidx].pool_refill_work,
+				  otx2_pool_refill_task);
+	}
 	return 0;
 }
 
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
index 8c29b63b319e..2a812110995f 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
@@ -129,6 +129,11 @@ struct otx2_hw {
 	struct otx2_drv_stats	drv_stats;
 };
 
+struct refill_work {
+	struct delayed_work pool_refill_work;
+	struct otx2_nic *pf;
+};
+
 struct otx2_nic {
 	void __iomem		*reg_base;
 	struct net_device	*netdev;
@@ -151,6 +156,10 @@ struct otx2_nic {
 	u16			pcifunc; /* RVU PF_FUNC */
 	struct cgx_link_user_info linfo;
 
+	u64			reset_count;
+	struct work_struct	reset_task;
+	struct refill_work	*refill_wrk;
+
 	/* Block address of NIX either BLKADDR_NIX0 or BLKADDR_NIX1 */
 	int			nix_blkaddr;
 };
@@ -435,6 +444,9 @@ otx2_mbox_up_handler_ ## _fn_name(struct otx2_nic *pfvf,		\
 MBOX_UP_CGX_MESSAGES
 #undef M
 
+/* Time to wait before watchdog kicks off */
+#define OTX2_TX_TIMEOUT		(100 * HZ)
+
 #define	RVU_PFVF_PF_SHIFT	10
 #define	RVU_PFVF_PF_MASK	0x3F
 #define	RVU_PFVF_FUNC_SHIFT	0
@@ -472,6 +484,7 @@ void otx2_free_cints(struct otx2_nic *pfvf, int n);
 void otx2_set_cints_affinity(struct otx2_nic *pfvf);
 int otx2_set_mac_address(struct net_device *netdev, void *p);
 int otx2_hw_set_mtu(struct otx2_nic *pfvf, int mtu);
+void otx2_tx_timeout(struct net_device *netdev, unsigned int txq);
 void otx2_get_mac_from_af(struct net_device *netdev);
 void otx2_config_irq_coalescing(struct otx2_nic *pfvf, int qidx);
 
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
index 60a67b90f2b0..171bab0ed662 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
@@ -478,6 +478,85 @@ static int otx2_set_real_num_queues(struct net_device *netdev,
 	return err;
 }
 
+static irqreturn_t otx2_q_intr_handler(int irq, void *data)
+{
+	struct otx2_nic *pf = data;
+	u64 val, *ptr;
+	u64 qidx = 0;
+
+	/* CQ */
+	for (qidx = 0; qidx < pf->qset.cq_cnt; qidx++) {
+		ptr = otx2_get_regaddr(pf, NIX_LF_CQ_OP_INT);
+		val = otx2_atomic64_add((qidx << 44), ptr);
+
+		otx2_write64(pf, NIX_LF_CQ_OP_INT, (qidx << 44) |
+			     (val & NIX_CQERRINT_BITS));
+		if (!(val & (NIX_CQERRINT_BITS | BIT_ULL(42))))
+			continue;
+
+		if (val & BIT_ULL(42)) {
+			netdev_err(pf->netdev, "CQ%lld: error reading NIX_LF_CQ_OP_INT, NIX_LF_ERR_INT 0x%llx\n",
+				   qidx, otx2_read64(pf, NIX_LF_ERR_INT));
+		} else {
+			if (val & BIT_ULL(NIX_CQERRINT_DOOR_ERR))
+				netdev_err(pf->netdev, "CQ%lld: Doorbell error",
+					   qidx);
+			if (val & BIT_ULL(NIX_CQERRINT_CQE_FAULT))
+				netdev_err(pf->netdev, "CQ%lld: Memory fault on CQE write to LLC/DRAM",
+					   qidx);
+		}
+
+		schedule_work(&pf->reset_task);
+	}
+
+	/* SQ */
+	for (qidx = 0; qidx < pf->hw.tx_queues; qidx++) {
+		ptr = otx2_get_regaddr(pf, NIX_LF_SQ_OP_INT);
+		val = otx2_atomic64_add((qidx << 44), ptr);
+		otx2_write64(pf, NIX_LF_SQ_OP_INT, (qidx << 44) |
+			     (val & NIX_SQINT_BITS));
+
+		if (!(val & (NIX_SQINT_BITS | BIT_ULL(42))))
+			continue;
+
+		if (val & BIT_ULL(42)) {
+			netdev_err(pf->netdev, "SQ%lld: error reading NIX_LF_SQ_OP_INT, NIX_LF_ERR_INT 0x%llx\n",
+				   qidx, otx2_read64(pf, NIX_LF_ERR_INT));
+		} else {
+			if (val & BIT_ULL(NIX_SQINT_LMT_ERR)) {
+				netdev_err(pf->netdev, "SQ%lld: LMT store error NIX_LF_SQ_OP_ERR_DBG:0x%llx",
+					   qidx,
+					   otx2_read64(pf,
+						       NIX_LF_SQ_OP_ERR_DBG));
+				otx2_write64(pf, NIX_LF_SQ_OP_ERR_DBG,
+					     BIT_ULL(44));
+			}
+			if (val & BIT_ULL(NIX_SQINT_MNQ_ERR)) {
+				netdev_err(pf->netdev, "SQ%lld: Meta-descriptor enqueue error NIX_LF_MNQ_ERR_DGB:0x%llx\n",
+					   qidx,
+					   otx2_read64(pf, NIX_LF_MNQ_ERR_DBG));
+				otx2_write64(pf, NIX_LF_MNQ_ERR_DBG,
+					     BIT_ULL(44));
+			}
+			if (val & BIT_ULL(NIX_SQINT_SEND_ERR)) {
+				netdev_err(pf->netdev, "SQ%lld: Send error, NIX_LF_SEND_ERR_DBG 0x%llx",
+					   qidx,
+					   otx2_read64(pf,
+						       NIX_LF_SEND_ERR_DBG));
+				otx2_write64(pf, NIX_LF_SEND_ERR_DBG,
+					     BIT_ULL(44));
+			}
+			if (val & BIT_ULL(NIX_SQINT_SQB_ALLOC_FAIL))
+				netdev_err(pf->netdev, "SQ%lld: SQB allocation failed",
+					   qidx);
+		}
+
+		schedule_work(&pf->reset_task);
+	}
+
+	return IRQ_HANDLED;
+}
+
 static irqreturn_t otx2_cq_intr_handler(int irq, void *cq_irq)
 {
 	struct otx2_cq_poll *cq_poll = (struct otx2_cq_poll *)cq_irq;
@@ -759,6 +838,24 @@ int otx2_open(struct net_device *netdev)
 	if (err)
 		goto err_disable_napi;
 
+	/* Register Queue IRQ handlers */
+	vec = pf->hw.nix_msixoff + NIX_LF_QINT_VEC_START;
+	irq_name = &pf->hw.irq_name[vec * NAME_SIZE];
+
+	snprintf(irq_name, NAME_SIZE, "%s-qerr", pf->netdev->name);
+
+	err = request_irq(pci_irq_vector(pf->pdev, vec),
+			  otx2_q_intr_handler, 0, irq_name, pf);
+	if (err) {
+		dev_err(pf->dev,
+			"RVUPF%d: IRQ registration failed for QERR\n",
+			rvu_get_pf(pf->pcifunc));
+		goto err_disable_napi;
+	}
+
+	/* Enable QINT IRQ */
+	otx2_write64(pf, NIX_LF_QINTX_ENA_W1S(0), BIT_ULL(0));
+
 	/* Register CQ IRQ handlers */
 	vec = pf->hw.nix_msixoff + NIX_LF_CINT_VEC_START;
 	for (qidx = 0; qidx < pf->hw.cint_cnt; qidx++) {
@@ -803,6 +900,11 @@ int otx2_open(struct net_device *netdev)
 
 err_free_cints:
 	otx2_free_cints(pf, qidx);
+	vec = pci_irq_vector(pf->pdev,
+			     pf->hw.nix_msixoff + NIX_LF_QINT_VEC_START);
+	otx2_write64(pf, NIX_LF_QINTX_ENA_W1C(0), BIT_ULL(0));
+	synchronize_irq(vec);
+	free_irq(vec, pf);
 err_disable_napi:
 	otx2_disable_napi(pf);
 	otx2_free_hw_resources(pf);
@@ -818,7 +920,7 @@ int otx2_stop(struct net_device *netdev)
 	struct otx2_nic *pf = netdev_priv(netdev);
 	struct otx2_cq_poll *cq_poll = NULL;
 	struct otx2_qset *qset = &pf->qset;
-	int qidx, vec;
+	int qidx, vec, wrk;
 
 	netif_carrier_off(netdev);
 	netif_tx_stop_all_queues(netdev);
@@ -830,6 +932,13 @@ int otx2_stop(struct net_device *netdev)
 	/* First stop packet Rx/Tx */
 	otx2_rxtx_enable(pf, false);
 
+	/* Cleanup Queue IRQ */
+	vec = pci_irq_vector(pf->pdev,
+			     pf->hw.nix_msixoff + NIX_LF_QINT_VEC_START);
+	otx2_write64(pf, NIX_LF_QINTX_ENA_W1C(0), BIT_ULL(0));
+	synchronize_irq(vec);
+	free_irq(vec, pf);
+
 	/* Cleanup CQ NAPI and IRQ */
 	vec = pf->hw.nix_msixoff + NIX_LF_CINT_VEC_START;
 	for (qidx = 0; qidx < pf->hw.cint_cnt; qidx++) {
@@ -852,6 +961,10 @@ int otx2_stop(struct net_device *netdev)
 	for (qidx = 0; qidx < netdev->num_tx_queues; qidx++)
 		netdev_tx_reset_queue(netdev_get_tx_queue(netdev, qidx));
 
+	for (wrk = 0; wrk < pf->qset.cq_cnt; wrk++)
+		cancel_delayed_work_sync(&pf->refill_wrk[wrk].pool_refill_work);
+	devm_kfree(pf->dev, pf->refill_wrk);
+
 	kfree(qset->sq);
 	kfree(qset->cq);
 	kfree(qset->napi);
@@ -931,6 +1044,19 @@ static int otx2_set_features(struct net_device *netdev,
 	return 0;
 }
 
+static void otx2_reset_task(struct work_struct *work)
+{
+	struct otx2_nic *pf = container_of(work, struct otx2_nic, reset_task);
+
+	if (!netif_running(pf->netdev))
+		return;
+
+	otx2_stop(pf->netdev);
+	pf->reset_count++;
+	otx2_open(pf->netdev);
+	netif_trans_update(pf->netdev);
+}
+
 static const struct net_device_ops otx2_netdev_ops = {
 	.ndo_open		= otx2_open,
 	.ndo_stop		= otx2_stop,
@@ -939,6 +1065,7 @@ static const struct net_device_ops otx2_netdev_ops = {
 	.ndo_change_mtu		= otx2_change_mtu,
 	.ndo_set_rx_mode	= otx2_set_rx_mode,
 	.ndo_set_features	= otx2_set_features,
+	.ndo_tx_timeout		= otx2_tx_timeout,
 };
 
 static int otx2_check_pf_usable(struct otx2_nic *nic)
@@ -1115,12 +1242,16 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 
 	netdev->hw_features |= NETIF_F_LOOPBACK | NETIF_F_RXALL;
 
+	netdev->watchdog_timeo = OTX2_TX_TIMEOUT;
+
 	netdev->netdev_ops = &otx2_netdev_ops;
 
 	/* MTU range: 64 - 9190 */
 	netdev->min_mtu = OTX2_MIN_MTU;
 	netdev->max_mtu = OTX2_MAX_MTU;
 
+	INIT_WORK(&pf->reset_task, otx2_reset_task);
+
 	err = register_netdev(netdev);
 	if (err) {
 		dev_err(dev, "Failed to register netdevice\n");
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_struct.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_struct.h
index 04a9f12be2e9..cba59ddf71bb 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_struct.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_struct.h
@@ -245,4 +245,32 @@ struct nix_sqe_mem_s {
 	u64 addr; /* W1 */
 };
 
+enum nix_cqerrint_e {
+	NIX_CQERRINT_DOOR_ERR = 0,
+	NIX_CQERRINT_WR_FULL = 1,
+	NIX_CQERRINT_CQE_FAULT = 2,
+};
+
+#define NIX_CQERRINT_BITS (BIT_ULL(NIX_CQERRINT_DOOR_ERR) | \
+			   BIT_ULL(NIX_CQERRINT_CQE_FAULT))
+
+enum nix_rqint_e {
+	NIX_RQINT_DROP = 0,
+	NIX_RQINT_RED = 1,
+};
+
+#define NIX_RQINT_BITS (BIT_ULL(NIX_RQINT_DROP) | BIT_ULL(NIX_RQINT_RED))
+
+enum nix_sqint_e {
+	NIX_SQINT_LMT_ERR = 0,
+	NIX_SQINT_MNQ_ERR = 1,
+	NIX_SQINT_SEND_ERR = 2,
+	NIX_SQINT_SQB_ALLOC_FAIL = 3,
+};
+
+#define NIX_SQINT_BITS (BIT_ULL(NIX_SQINT_LMT_ERR) | \
+			BIT_ULL(NIX_SQINT_MNQ_ERR) | \
+			BIT_ULL(NIX_SQINT_SEND_ERR) | \
+			BIT_ULL(NIX_SQINT_SQB_ALLOC_FAIL))
+
 #endif /* OTX2_STRUCT_H */
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c
index 87b579a121ab..94dac84de8dc 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c
@@ -237,12 +237,23 @@ static int otx2_rx_napi_handler(struct otx2_nic *pfvf,
 	/* Refill pool with new buffers */
 	while (cq->pool_ptrs) {
 		bufptr = otx2_alloc_rbuf(pfvf, cq->rbpool, GFP_ATOMIC);
-		if (unlikely(bufptr <= 0))
+		if (unlikely(bufptr <= 0)) {
+			struct refill_work *work;
+			struct delayed_work *dwork;
+
+			work = &pfvf->refill_wrk[cq->cq_idx];
+			dwork = &work->pool_refill_work;
+			/* Schedule a task if no other task is running */
+			if (!cq->refill_task_sched) {
+				cq->refill_task_sched = true;
+				schedule_delayed_work(dwork,
+						      msecs_to_jiffies(100));
+			}
 			break;
+		}
 		otx2_aura_freeptr(pfvf, cq->cq_idx, bufptr + OTX2_HEAD_ROOM);
 		cq->pool_ptrs--;
 	}
-	otx2_get_page(cq->rbpool);
 
 	return processed_cqe;
 }
@@ -304,6 +315,11 @@ int otx2_napi_handler(struct napi_struct *napi, int budget)
 			continue;
 		cq = &qset->cq[cq_idx];
 		if (cq->cq_type == CQ_RX) {
+			/* If the RQ refill WQ task is running, skip napi
+			 * scheduler for this queue.
+			 */
+			if (cq->refill_task_sched)
+				continue;
 			workdone += otx2_rx_napi_handler(pfvf, napi,
 							 cq, budget);
 		} else {
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h
index bad225979aec..a889b491ea52 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h
@@ -112,6 +112,7 @@ struct otx2_cq_queue {
 	u8			cq_idx;
 	u8			cq_type;
 	u8			cint_idx; /* CQ interrupt id */
+	u8			refill_task_sched;
 	u16			cqe_size;
 	u16			pool_ptrs;
 	u32			cqe_cnt;
-- 
cgit v1.2.3-59-g8ed1b


From 85069e95e53161f8bcc1aa266f47c60c1f3e4ec3 Mon Sep 17 00:00:00 2001
From: Sunil Goutham <sgoutham@marvell.com>
Date: Mon, 27 Jan 2020 18:35:25 +0530
Subject: octeontx2-pf: Receive side scaling support

Adds receive side scaling (RSS) support to distribute
pkts/flows across multiple queues. Sets up key, indirection
table etc. Also added extraction of HW calculated rxhash and
adding to same to SKB ie NETIF_F_RXHASH offload support.

Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../ethernet/marvell/octeontx2/nic/otx2_common.c   | 122 +++++++++++++++++++++
 .../ethernet/marvell/octeontx2/nic/otx2_common.h   |  14 +++
 .../net/ethernet/marvell/octeontx2/nic/otx2_pf.c   |   8 +-
 .../net/ethernet/marvell/octeontx2/nic/otx2_txrx.c |  24 ++++
 4 files changed, 167 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
index 2b503066d801..96751588e841 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
@@ -106,6 +106,126 @@ int otx2_hw_set_mtu(struct otx2_nic *pfvf, int mtu)
 	return err;
 }
 
+static int otx2_set_flowkey_cfg(struct otx2_nic *pfvf)
+{
+	struct otx2_rss_info *rss = &pfvf->hw.rss_info;
+	struct nix_rss_flowkey_cfg *req;
+	int err;
+
+	otx2_mbox_lock(&pfvf->mbox);
+	req = otx2_mbox_alloc_msg_nix_rss_flowkey_cfg(&pfvf->mbox);
+	if (!req) {
+		otx2_mbox_unlock(&pfvf->mbox);
+		return -ENOMEM;
+	}
+	req->mcam_index = -1; /* Default or reserved index */
+	req->flowkey_cfg = rss->flowkey_cfg;
+	req->group = DEFAULT_RSS_CONTEXT_GROUP;
+
+	err = otx2_sync_mbox_msg(&pfvf->mbox);
+	otx2_mbox_unlock(&pfvf->mbox);
+	return err;
+}
+
+static int otx2_set_rss_table(struct otx2_nic *pfvf)
+{
+	struct otx2_rss_info *rss = &pfvf->hw.rss_info;
+	struct mbox *mbox = &pfvf->mbox;
+	struct nix_aq_enq_req *aq;
+	int idx, err;
+
+	otx2_mbox_lock(mbox);
+	/* Get memory to put this msg */
+	for (idx = 0; idx < rss->rss_size; idx++) {
+		aq = otx2_mbox_alloc_msg_nix_aq_enq(mbox);
+		if (!aq) {
+			/* The shared memory buffer can be full.
+			 * Flush it and retry
+			 */
+			err = otx2_sync_mbox_msg(mbox);
+			if (err) {
+				otx2_mbox_unlock(mbox);
+				return err;
+			}
+			aq = otx2_mbox_alloc_msg_nix_aq_enq(mbox);
+			if (!aq) {
+				otx2_mbox_unlock(mbox);
+				return -ENOMEM;
+			}
+		}
+
+		aq->rss.rq = rss->ind_tbl[idx];
+
+		/* Fill AQ info */
+		aq->qidx = idx;
+		aq->ctype = NIX_AQ_CTYPE_RSS;
+		aq->op = NIX_AQ_INSTOP_INIT;
+	}
+	err = otx2_sync_mbox_msg(mbox);
+	otx2_mbox_unlock(mbox);
+	return err;
+}
+
+static void otx2_set_rss_key(struct otx2_nic *pfvf)
+{
+	struct otx2_rss_info *rss = &pfvf->hw.rss_info;
+	u64 *key = (u64 *)&rss->key[4];
+	int idx;
+
+	/* 352bit or 44byte key needs to be configured as below
+	 * NIX_LF_RX_SECRETX0 = key<351:288>
+	 * NIX_LF_RX_SECRETX1 = key<287:224>
+	 * NIX_LF_RX_SECRETX2 = key<223:160>
+	 * NIX_LF_RX_SECRETX3 = key<159:96>
+	 * NIX_LF_RX_SECRETX4 = key<95:32>
+	 * NIX_LF_RX_SECRETX5<63:32> = key<31:0>
+	 */
+	otx2_write64(pfvf, NIX_LF_RX_SECRETX(5),
+		     (u64)(*((u32 *)&rss->key)) << 32);
+	idx = sizeof(rss->key) / sizeof(u64);
+	while (idx > 0) {
+		idx--;
+		otx2_write64(pfvf, NIX_LF_RX_SECRETX(idx), *key++);
+	}
+}
+
+int otx2_rss_init(struct otx2_nic *pfvf)
+{
+	struct otx2_rss_info *rss = &pfvf->hw.rss_info;
+	int idx, ret = 0;
+
+	rss->rss_size = sizeof(rss->ind_tbl);
+
+	/* Init RSS key if it is not setup already */
+	if (!rss->enable)
+		netdev_rss_key_fill(rss->key, sizeof(rss->key));
+	otx2_set_rss_key(pfvf);
+
+	if (!netif_is_rxfh_configured(pfvf->netdev)) {
+		/* Default indirection table */
+		for (idx = 0; idx < rss->rss_size; idx++)
+			rss->ind_tbl[idx] =
+				ethtool_rxfh_indir_default(idx,
+							   pfvf->hw.rx_queues);
+	}
+	ret = otx2_set_rss_table(pfvf);
+	if (ret)
+		return ret;
+
+	/* Flowkey or hash config to be used for generating flow tag */
+	rss->flowkey_cfg = rss->enable ? rss->flowkey_cfg :
+			   NIX_FLOW_KEY_TYPE_IPV4 | NIX_FLOW_KEY_TYPE_IPV6 |
+			   NIX_FLOW_KEY_TYPE_TCP | NIX_FLOW_KEY_TYPE_UDP |
+			   NIX_FLOW_KEY_TYPE_SCTP;
+
+	ret = otx2_set_flowkey_cfg(pfvf);
+	if (ret)
+		return ret;
+
+	rss->enable = true;
+	return 0;
+}
+
 void otx2_config_irq_coalescing(struct otx2_nic *pfvf, int qidx)
 {
 	/* Configure CQE interrupt coalescing parameters
@@ -608,6 +728,8 @@ int otx2_config_nix(struct otx2_nic *pfvf)
 	nixlf->rq_cnt = pfvf->hw.rx_queues;
 	nixlf->sq_cnt = pfvf->hw.tx_queues;
 	nixlf->cq_cnt = pfvf->qset.cq_cnt;
+	nixlf->rss_sz = MAX_RSS_INDIR_TBL_SIZE;
+	nixlf->rss_grps = 1; /* Single RSS indir table supported, for now */
 	nixlf->xqe_sz = NIX_XQESZ_W16;
 	/* We don't know absolute NPA LF idx attached.
 	 * AF will replace 'RVU_DEFAULT_PF_FUNC' with
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
index 2a812110995f..f9149efc8b03 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
@@ -41,6 +41,16 @@ enum arua_mapped_qtypes {
 #define NIX_LF_ERR_VEC				0x81
 #define NIX_LF_POISON_VEC			0x82
 
+/* RSS configuration */
+struct otx2_rss_info {
+	u8 enable;
+	u32 flowkey_cfg;
+	u16 rss_size;
+	u8  ind_tbl[MAX_RSS_INDIR_TBL_SIZE];
+#define RSS_HASH_KEY_SIZE	44   /* 352 bit key */
+	u8  key[RSS_HASH_KEY_SIZE];
+};
+
 /* NIX (or NPC) RX errors */
 enum otx2_errlvl {
 	NPC_ERRLVL_RE,
@@ -95,6 +105,7 @@ struct mbox {
 
 struct otx2_hw {
 	struct pci_dev		*pdev;
+	struct otx2_rss_info	rss_info;
 	u16                     rx_queues;
 	u16                     tx_queues;
 	u16			max_queues;
@@ -510,6 +521,9 @@ void otx2_ctx_disable(struct mbox *mbox, int type, bool npa);
 void otx2_cleanup_rx_cqes(struct otx2_nic *pfvf, struct otx2_cq_queue *cq);
 void otx2_cleanup_tx_cqes(struct otx2_nic *pfvf, struct otx2_cq_queue *cq);
 
+/* RSS configuration APIs*/
+int otx2_rss_init(struct otx2_nic *pfvf);
+
 /* Mbox handlers */
 void mbox_handler_msix_offset(struct otx2_nic *pfvf,
 			      struct msix_offset_rsp *rsp);
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
index 171bab0ed662..5f78215979f5 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
@@ -838,6 +838,11 @@ int otx2_open(struct net_device *netdev)
 	if (err)
 		goto err_disable_napi;
 
+	/* Initialize RSS */
+	err = otx2_rss_init(pf);
+	if (err)
+		goto err_disable_napi;
+
 	/* Register Queue IRQ handlers */
 	vec = pf->hw.nix_msixoff + NIX_LF_QINT_VEC_START;
 	irq_name = &pf->hw.irq_name[vec * NAME_SIZE];
@@ -1237,7 +1242,8 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	pf->iommu_domain = iommu_get_domain_for_dev(dev);
 
 	netdev->hw_features = (NETIF_F_RXCSUM | NETIF_F_IP_CSUM |
-			       NETIF_F_IPV6_CSUM | NETIF_F_SG);
+			       NETIF_F_IPV6_CSUM | NETIF_F_RXHASH |
+			       NETIF_F_SG);
 	netdev->features |= netdev->hw_features;
 
 	netdev->hw_features |= NETIF_F_LOOPBACK | NETIF_F_RXALL;
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c
index 94dac84de8dc..513b126feaac 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c
@@ -115,6 +115,28 @@ static void otx2_skb_add_frag(struct otx2_nic *pfvf, struct sk_buff *skb,
 			    pfvf->rbsize, DMA_FROM_DEVICE);
 }
 
+static void otx2_set_rxhash(struct otx2_nic *pfvf,
+			    struct nix_cqe_rx_s *cqe, struct sk_buff *skb)
+{
+	enum pkt_hash_types hash_type = PKT_HASH_TYPE_NONE;
+	struct otx2_rss_info *rss;
+	u32 hash = 0;
+
+	if (!(pfvf->netdev->features & NETIF_F_RXHASH))
+		return;
+
+	rss = &pfvf->hw.rss_info;
+	if (rss->flowkey_cfg) {
+		if (rss->flowkey_cfg &
+		    ~(NIX_FLOW_KEY_TYPE_IPV4 | NIX_FLOW_KEY_TYPE_IPV6))
+			hash_type = PKT_HASH_TYPE_L4;
+		else
+			hash_type = PKT_HASH_TYPE_L3;
+		hash = cqe->hdr.flow_tag;
+	}
+	skb_set_hash(skb, hash, hash_type);
+}
+
 static bool otx2_check_rcv_errors(struct otx2_nic *pfvf,
 				  struct nix_cqe_rx_s *cqe, int qidx)
 {
@@ -194,6 +216,8 @@ static void otx2_rcv_pkt_handler(struct otx2_nic *pfvf,
 	otx2_skb_add_frag(pfvf, skb, cqe->sg.seg_addr, cqe->sg.seg_size);
 	cq->pool_ptrs++;
 
+	otx2_set_rxhash(pfvf, cqe, skb);
+
 	skb_record_rx_queue(skb, cq->cq_idx);
 	if (pfvf->netdev->features & NETIF_F_RXCSUM)
 		skb->ip_summed = CHECKSUM_UNNECESSARY;
-- 
cgit v1.2.3-59-g8ed1b


From 86d7476078b87874ce5dee39cbbd094055c73d49 Mon Sep 17 00:00:00 2001
From: Sunil Goutham <sgoutham@marvell.com>
Date: Mon, 27 Jan 2020 18:35:26 +0530
Subject: octeontx2-pf: TCP segmentation offload support

Adds TCP segmentation offload (TSO) support. First version
of the silicon didn't support TSO offload, for this driver
level TSO support is added.

Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../ethernet/marvell/octeontx2/nic/otx2_common.c   |   8 +
 .../ethernet/marvell/octeontx2/nic/otx2_common.h   |  11 +
 .../net/ethernet/marvell/octeontx2/nic/otx2_pf.c   |   9 +-
 .../net/ethernet/marvell/octeontx2/nic/otx2_txrx.c | 247 ++++++++++++++++++++-
 .../net/ethernet/marvell/octeontx2/nic/otx2_txrx.h |   2 +
 5 files changed, 273 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
index 96751588e841..c674171104c9 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
@@ -10,6 +10,7 @@
 
 #include <linux/interrupt.h>
 #include <linux/pci.h>
+#include <net/tso.h>
 
 #include "otx2_reg.h"
 #include "otx2_common.h"
@@ -522,6 +523,11 @@ static int otx2_sq_init(struct otx2_nic *pfvf, u16 qidx, u16 sqb_aura)
 	if (err)
 		return err;
 
+	err = qmem_alloc(pfvf->dev, &sq->tso_hdrs, qset->sqe_cnt,
+			 TSO_HEADER_SIZE);
+	if (err)
+		return err;
+
 	sq->sqe_base = sq->sqe->base;
 	sq->sg = kcalloc(qset->sqe_cnt, sizeof(struct sg_list), GFP_KERNEL);
 	if (!sq->sg)
@@ -1211,6 +1217,8 @@ void mbox_handler_nix_lf_alloc(struct otx2_nic *pfvf,
 	pfvf->hw.sqb_size = rsp->sqb_size;
 	pfvf->hw.rx_chan_base = rsp->rx_chan_base;
 	pfvf->hw.tx_chan_base = rsp->tx_chan_base;
+	pfvf->hw.lso_tsov4_idx = rsp->lso_tsov4_idx;
+	pfvf->hw.lso_tsov6_idx = rsp->lso_tsov6_idx;
 }
 
 void mbox_handler_msix_offset(struct otx2_nic *pfvf,
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
index f9149efc8b03..1598396870af 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
@@ -129,6 +129,11 @@ struct otx2_hw {
 	u16			rq_skid;
 	u8			cq_time_wait;
 
+	/* For TSO segmentation */
+	u8			lso_tsov4_idx;
+	u8			lso_tsov6_idx;
+	u8			hw_tso;
+
 	/* MSI-X */
 	u8			cint_cnt; /* CQ interrupt count */
 	u16			npa_msixoff; /* Offset of NPA vectors */
@@ -189,11 +194,17 @@ static inline bool is_96xx_B0(struct pci_dev *pdev)
 
 static inline void otx2_setup_dev_hw_settings(struct otx2_nic *pfvf)
 {
+	struct otx2_hw *hw = &pfvf->hw;
+
 	pfvf->hw.cq_time_wait = CQ_TIMER_THRESH_DEFAULT;
 	pfvf->hw.cq_ecount_wait = CQ_CQE_THRESH_DEFAULT;
 	pfvf->hw.cq_qcount_wait = CQ_QCOUNT_DEFAULT;
 
+	hw->hw_tso = true;
+
 	if (is_96xx_A0(pfvf->pdev)) {
+		hw->hw_tso = false;
+
 		/* Time based irq coalescing is not supported */
 		pfvf->hw.cq_qcount_wait = 0x0;
 
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
index 5f78215979f5..d0809369c367 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
@@ -616,6 +616,7 @@ static void otx2_free_sq_res(struct otx2_nic *pf)
 	for (qidx = 0; qidx < pf->hw.tx_queues; qidx++) {
 		sq = &qset->sq[qidx];
 		qmem_free(pf->dev, sq->sqe);
+		qmem_free(pf->dev, sq->tso_hdrs);
 		kfree(sq->sg);
 		kfree(sq->sqb_ptrs);
 	}
@@ -986,8 +987,9 @@ static netdev_tx_t otx2_xmit(struct sk_buff *skb, struct net_device *netdev)
 	struct otx2_snd_queue *sq;
 	struct netdev_queue *txq;
 
-	/* Check for minimum packet length */
-	if (skb->len <= ETH_HLEN) {
+	/* Check for minimum and maximum packet length */
+	if (skb->len <= ETH_HLEN ||
+	    (!skb_shinfo(skb)->gso_size && skb->len > pf->max_frs)) {
 		dev_kfree_skb(skb);
 		return NETDEV_TX_OK;
 	}
@@ -1243,11 +1245,12 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 
 	netdev->hw_features = (NETIF_F_RXCSUM | NETIF_F_IP_CSUM |
 			       NETIF_F_IPV6_CSUM | NETIF_F_RXHASH |
-			       NETIF_F_SG);
+			       NETIF_F_SG | NETIF_F_TSO | NETIF_F_TSO6);
 	netdev->features |= netdev->hw_features;
 
 	netdev->hw_features |= NETIF_F_LOOPBACK | NETIF_F_RXALL;
 
+	netdev->gso_max_segs = OTX2_MAX_GSO_SEGS;
 	netdev->watchdog_timeo = OTX2_TX_TIMEOUT;
 
 	netdev->netdev_ops = &otx2_netdev_ops;
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c
index 513b126feaac..6a7ca3b4b9be 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c
@@ -10,6 +10,7 @@
 
 #include <linux/etherdevice.h>
 #include <net/ip.h>
+#include <net/tso.h>
 
 #include "otx2_reg.h"
 #include "otx2_common.h"
@@ -428,6 +429,38 @@ static bool otx2_sqe_add_sg(struct otx2_nic *pfvf, struct otx2_snd_queue *sq,
 	return true;
 }
 
+/* Add SQE extended header subdescriptor */
+static void otx2_sqe_add_ext(struct otx2_nic *pfvf, struct otx2_snd_queue *sq,
+			     struct sk_buff *skb, int *offset)
+{
+	struct nix_sqe_ext_s *ext;
+
+	ext = (struct nix_sqe_ext_s *)(sq->sqe_base + *offset);
+	ext->subdc = NIX_SUBDC_EXT;
+	if (skb_shinfo(skb)->gso_size) {
+		ext->lso = 1;
+		ext->lso_sb = skb_transport_offset(skb) + tcp_hdrlen(skb);
+		ext->lso_mps = skb_shinfo(skb)->gso_size;
+
+		/* Only TSOv4 and TSOv6 GSO offloads are supported */
+		if (skb_shinfo(skb)->gso_type & SKB_GSO_TCPV4) {
+			ext->lso_format = pfvf->hw.lso_tsov4_idx;
+
+			/* HW adds payload size to 'ip_hdr->tot_len' while
+			 * sending TSO segment, hence set payload length
+			 * in IP header of the packet to just header length.
+			 */
+			ip_hdr(skb)->tot_len =
+				htons(ext->lso_sb - skb_network_offset(skb));
+		} else {
+			ext->lso_format = pfvf->hw.lso_tsov6_idx;
+			ipv6_hdr(skb)->payload_len =
+				htons(ext->lso_sb - skb_network_offset(skb));
+		}
+	}
+	*offset += sizeof(*ext);
+}
+
 /* Add SQE header subdescriptor structure */
 static void otx2_sqe_add_hdr(struct otx2_nic *pfvf, struct otx2_snd_queue *sq,
 			     struct nix_sqe_hdr_s *sqe_hdr,
@@ -475,6 +508,209 @@ static void otx2_sqe_add_hdr(struct otx2_nic *pfvf, struct otx2_snd_queue *sq,
 	}
 }
 
+static int otx2_dma_map_tso_skb(struct otx2_nic *pfvf,
+				struct otx2_snd_queue *sq,
+				struct sk_buff *skb, int sqe, int hdr_len)
+{
+	int num_segs = skb_shinfo(skb)->nr_frags + 1;
+	struct sg_list *sg = &sq->sg[sqe];
+	u64 dma_addr;
+	int seg, len;
+
+	sg->num_segs = 0;
+
+	/* Get payload length at skb->data */
+	len = skb_headlen(skb) - hdr_len;
+
+	for (seg = 0; seg < num_segs; seg++) {
+		/* Skip skb->data, if there is no payload */
+		if (!seg && !len)
+			continue;
+		dma_addr = otx2_dma_map_skb_frag(pfvf, skb, seg, &len);
+		if (dma_mapping_error(pfvf->dev, dma_addr))
+			goto unmap;
+
+		/* Save DMA mapping info for later unmapping */
+		sg->dma_addr[sg->num_segs] = dma_addr;
+		sg->size[sg->num_segs] = len;
+		sg->num_segs++;
+	}
+	return 0;
+unmap:
+	otx2_dma_unmap_skb_frags(pfvf, sg);
+	return -EINVAL;
+}
+
+static u64 otx2_tso_frag_dma_addr(struct otx2_snd_queue *sq,
+				  struct sk_buff *skb, int seg,
+				  u64 seg_addr, int hdr_len, int sqe)
+{
+	struct sg_list *sg = &sq->sg[sqe];
+	const skb_frag_t *frag;
+	int offset;
+
+	if (seg < 0)
+		return sg->dma_addr[0] + (seg_addr - (u64)skb->data);
+
+	frag = &skb_shinfo(skb)->frags[seg];
+	offset = seg_addr - (u64)skb_frag_address(frag);
+	if (skb_headlen(skb) - hdr_len)
+		seg++;
+	return sg->dma_addr[seg] + offset;
+}
+
+static void otx2_sqe_tso_add_sg(struct otx2_snd_queue *sq,
+				struct sg_list *list, int *offset)
+{
+	struct nix_sqe_sg_s *sg = NULL;
+	u16 *sg_lens = NULL;
+	u64 *iova = NULL;
+	int seg;
+
+	/* Add SG descriptors with buffer addresses */
+	for (seg = 0; seg < list->num_segs; seg++) {
+		if ((seg % MAX_SEGS_PER_SG) == 0) {
+			sg = (struct nix_sqe_sg_s *)(sq->sqe_base + *offset);
+			sg->ld_type = NIX_SEND_LDTYPE_LDD;
+			sg->subdc = NIX_SUBDC_SG;
+			sg->segs = 0;
+			sg_lens = (void *)sg;
+			iova = (void *)sg + sizeof(*sg);
+			/* Next subdc always starts at a 16byte boundary.
+			 * So if sg->segs is whether 2 or 3, offset += 16bytes.
+			 */
+			if ((list->num_segs - seg) >= (MAX_SEGS_PER_SG - 1))
+				*offset += sizeof(*sg) + (3 * sizeof(u64));
+			else
+				*offset += sizeof(*sg) + sizeof(u64);
+		}
+		sg_lens[frag_num(seg % MAX_SEGS_PER_SG)] = list->size[seg];
+		*iova++ = list->dma_addr[seg];
+		sg->segs++;
+	}
+}
+
+static void otx2_sq_append_tso(struct otx2_nic *pfvf, struct otx2_snd_queue *sq,
+			       struct sk_buff *skb, u16 qidx)
+{
+	struct netdev_queue *txq = netdev_get_tx_queue(pfvf->netdev, qidx);
+	int hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb);
+	int tcp_data, seg_len, pkt_len, offset;
+	struct nix_sqe_hdr_s *sqe_hdr;
+	int first_sqe = sq->head;
+	struct sg_list list;
+	struct tso_t tso;
+
+	/* Map SKB's fragments to DMA.
+	 * It's done here to avoid mapping for every TSO segment's packet.
+	 */
+	if (otx2_dma_map_tso_skb(pfvf, sq, skb, first_sqe, hdr_len)) {
+		dev_kfree_skb_any(skb);
+		return;
+	}
+
+	netdev_tx_sent_queue(txq, skb->len);
+
+	tso_start(skb, &tso);
+	tcp_data = skb->len - hdr_len;
+	while (tcp_data > 0) {
+		char *hdr;
+
+		seg_len = min_t(int, skb_shinfo(skb)->gso_size, tcp_data);
+		tcp_data -= seg_len;
+
+		/* Set SQE's SEND_HDR */
+		memset(sq->sqe_base, 0, sq->sqe_size);
+		sqe_hdr = (struct nix_sqe_hdr_s *)(sq->sqe_base);
+		otx2_sqe_add_hdr(pfvf, sq, sqe_hdr, skb, qidx);
+		offset = sizeof(*sqe_hdr);
+
+		/* Add TSO segment's pkt header */
+		hdr = sq->tso_hdrs->base + (sq->head * TSO_HEADER_SIZE);
+		tso_build_hdr(skb, hdr, &tso, seg_len, tcp_data == 0);
+		list.dma_addr[0] =
+			sq->tso_hdrs->iova + (sq->head * TSO_HEADER_SIZE);
+		list.size[0] = hdr_len;
+		list.num_segs = 1;
+
+		/* Add TSO segment's payload data fragments */
+		pkt_len = hdr_len;
+		while (seg_len > 0) {
+			int size;
+
+			size = min_t(int, tso.size, seg_len);
+
+			list.size[list.num_segs] = size;
+			list.dma_addr[list.num_segs] =
+				otx2_tso_frag_dma_addr(sq, skb,
+						       tso.next_frag_idx - 1,
+						       (u64)tso.data, hdr_len,
+						       first_sqe);
+			list.num_segs++;
+			pkt_len += size;
+			seg_len -= size;
+			tso_build_data(skb, &tso, size);
+		}
+		sqe_hdr->total = pkt_len;
+		otx2_sqe_tso_add_sg(sq, &list, &offset);
+
+		/* DMA mappings and skb needs to be freed only after last
+		 * TSO segment is transmitted out. So set 'PNC' only for
+		 * last segment. Also point last segment's sqe_id to first
+		 * segment's SQE index where skb address and DMA mappings
+		 * are saved.
+		 */
+		if (!tcp_data) {
+			sqe_hdr->pnc = 1;
+			sqe_hdr->sqe_id = first_sqe;
+			sq->sg[first_sqe].skb = (u64)skb;
+		} else {
+			sqe_hdr->pnc = 0;
+		}
+
+		sqe_hdr->sizem1 = (offset / 16) - 1;
+
+		/* Flush SQE to HW */
+		otx2_sqe_flush(sq, offset);
+	}
+}
+
+static bool is_hw_tso_supported(struct otx2_nic *pfvf,
+				struct sk_buff *skb)
+{
+	int payload_len, last_seg_size;
+
+	if (!pfvf->hw.hw_tso)
+		return false;
+
+	/* HW has an issue due to which when the payload of the last LSO
+	 * segment is shorter than 16 bytes, some header fields may not
+	 * be correctly modified, hence don't offload such TSO segments.
+	 */
+	if (!is_96xx_B0(pfvf->pdev))
+		return true;
+
+	payload_len = skb->len - (skb_transport_offset(skb) + tcp_hdrlen(skb));
+	last_seg_size = payload_len % skb_shinfo(skb)->gso_size;
+	if (last_seg_size && last_seg_size < 16)
+		return false;
+
+	return true;
+}
+
+static int otx2_get_sqe_count(struct otx2_nic *pfvf, struct sk_buff *skb)
+{
+	if (!skb_shinfo(skb)->gso_size)
+		return 1;
+
+	/* HW TSO */
+	if (is_hw_tso_supported(pfvf, skb))
+		return 1;
+
+	/* SW TSO */
+	return skb_shinfo(skb)->gso_segs;
+}
+
 bool otx2_sq_append_skb(struct net_device *netdev, struct otx2_snd_queue *sq,
 			struct sk_buff *skb, u16 qidx)
 {
@@ -489,7 +725,8 @@ bool otx2_sq_append_skb(struct net_device *netdev, struct otx2_snd_queue *sq,
 	 */
 	free_sqe = (sq->num_sqbs - *sq->aura_fc_addr) * sq->sqe_per_sqb;
 
-	if (!free_sqe || free_sqe < sq->sqe_thresh)
+	if (free_sqe < sq->sqe_thresh ||
+	    free_sqe < otx2_get_sqe_count(pfvf, skb))
 		return false;
 
 	num_segs = skb_shinfo(skb)->nr_frags + 1;
@@ -505,6 +742,11 @@ bool otx2_sq_append_skb(struct net_device *netdev, struct otx2_snd_queue *sq,
 		num_segs = skb_shinfo(skb)->nr_frags + 1;
 	}
 
+	if (skb_shinfo(skb)->gso_size && !is_hw_tso_supported(pfvf, skb)) {
+		otx2_sq_append_tso(pfvf, sq, skb, qidx);
+		return true;
+	}
+
 	/* Set SQE's SEND_HDR.
 	 * Do not clear the first 64bit as it contains constant info.
 	 */
@@ -513,6 +755,9 @@ bool otx2_sq_append_skb(struct net_device *netdev, struct otx2_snd_queue *sq,
 	otx2_sqe_add_hdr(pfvf, sq, sqe_hdr, skb, qidx);
 	offset = sizeof(*sqe_hdr);
 
+	/* Add extended header if needed */
+	otx2_sqe_add_ext(pfvf, sq, skb, &offset);
+
 	/* Add SG subdesc with data frags */
 	if (!otx2_sqe_add_sg(pfvf, sq, skb, num_segs, &offset)) {
 		otx2_dma_unmap_skb_frags(pfvf, &sq->sg[sq->head]);
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h
index a889b491ea52..107a261492ba 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h
@@ -26,6 +26,7 @@
 #define	OTX2_MIN_MTU		64
 #define	OTX2_MAX_MTU		(9212 - OTX2_ETH_HLEN)
 
+#define OTX2_MAX_GSO_SEGS	255
 #define OTX2_MAX_FRAGS_IN_SQE	9
 
 /* Rx buffer size should be in multiples of 128bytes */
@@ -79,6 +80,7 @@ struct otx2_snd_queue {
 	u64			*lmt_addr;
 	void			*sqe_base;
 	struct qmem		*sqe;
+	struct qmem		*tso_hdrs;
 	struct sg_list		*sg;
 	u16			sqb_count;
 	u64			*sqb_ptrs;
-- 
cgit v1.2.3-59-g8ed1b


From e239d0c78c42e49558ea2c0e5857c4f9ca064f48 Mon Sep 17 00:00:00 2001
From: Geetha sowjanya <gakula@marvell.com>
Date: Mon, 27 Jan 2020 18:35:27 +0530
Subject: octeontx2-pf: Add ndo_get_stats64

Added ndo_get_stats64 which returns stats maintained by HW.

Signed-off-by: Geetha sowjanya <gakula@marvell.com>
Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../ethernet/marvell/octeontx2/nic/otx2_common.c   | 47 +++++++++++++++++++++
 .../ethernet/marvell/octeontx2/nic/otx2_common.h   | 49 ++++++++++++++++++++++
 .../net/ethernet/marvell/octeontx2/nic/otx2_pf.c   |  1 +
 3 files changed, 97 insertions(+)

diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
index c674171104c9..c4be7872d9f4 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
@@ -16,6 +16,53 @@
 #include "otx2_common.h"
 #include "otx2_struct.h"
 
+void otx2_get_dev_stats(struct otx2_nic *pfvf)
+{
+	struct otx2_dev_stats *dev_stats = &pfvf->hw.dev_stats;
+
+#define OTX2_GET_RX_STATS(reg) \
+	 otx2_read64(pfvf, NIX_LF_RX_STATX(reg))
+#define OTX2_GET_TX_STATS(reg) \
+	 otx2_read64(pfvf, NIX_LF_TX_STATX(reg))
+
+	dev_stats->rx_bytes = OTX2_GET_RX_STATS(RX_OCTS);
+	dev_stats->rx_drops = OTX2_GET_RX_STATS(RX_DROP);
+	dev_stats->rx_bcast_frames = OTX2_GET_RX_STATS(RX_BCAST);
+	dev_stats->rx_mcast_frames = OTX2_GET_RX_STATS(RX_MCAST);
+	dev_stats->rx_ucast_frames = OTX2_GET_RX_STATS(RX_UCAST);
+	dev_stats->rx_frames = dev_stats->rx_bcast_frames +
+			       dev_stats->rx_mcast_frames +
+			       dev_stats->rx_ucast_frames;
+
+	dev_stats->tx_bytes = OTX2_GET_TX_STATS(TX_OCTS);
+	dev_stats->tx_drops = OTX2_GET_TX_STATS(TX_DROP);
+	dev_stats->tx_bcast_frames = OTX2_GET_TX_STATS(TX_BCAST);
+	dev_stats->tx_mcast_frames = OTX2_GET_TX_STATS(TX_MCAST);
+	dev_stats->tx_ucast_frames = OTX2_GET_TX_STATS(TX_UCAST);
+	dev_stats->tx_frames = dev_stats->tx_bcast_frames +
+			       dev_stats->tx_mcast_frames +
+			       dev_stats->tx_ucast_frames;
+}
+
+void otx2_get_stats64(struct net_device *netdev,
+		      struct rtnl_link_stats64 *stats)
+{
+	struct otx2_nic *pfvf = netdev_priv(netdev);
+	struct otx2_dev_stats *dev_stats;
+
+	otx2_get_dev_stats(pfvf);
+
+	dev_stats = &pfvf->hw.dev_stats;
+	stats->rx_bytes = dev_stats->rx_bytes;
+	stats->rx_packets = dev_stats->rx_frames;
+	stats->rx_dropped = dev_stats->rx_drops;
+	stats->multicast = dev_stats->rx_mcast_frames;
+
+	stats->tx_bytes = dev_stats->tx_bytes;
+	stats->tx_packets = dev_stats->tx_frames;
+	stats->tx_dropped = dev_stats->tx_drops;
+}
+
 /* Sync MAC address with RVU AF */
 static int otx2_hw_set_mac_addr(struct otx2_nic *pfvf, u8 *mac)
 {
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
index 1598396870af..ce7a552b478a 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
@@ -81,6 +81,49 @@ enum otx2_errcodes_re {
 	ERRCODE_IL4_CSUM = 0x22,
 };
 
+/* NIX TX stats */
+enum nix_stat_lf_tx {
+	TX_UCAST	= 0x0,
+	TX_BCAST	= 0x1,
+	TX_MCAST	= 0x2,
+	TX_DROP		= 0x3,
+	TX_OCTS		= 0x4,
+	TX_STATS_ENUM_LAST,
+};
+
+/* NIX RX stats */
+enum nix_stat_lf_rx {
+	RX_OCTS		= 0x0,
+	RX_UCAST	= 0x1,
+	RX_BCAST	= 0x2,
+	RX_MCAST	= 0x3,
+	RX_DROP		= 0x4,
+	RX_DROP_OCTS	= 0x5,
+	RX_FCS		= 0x6,
+	RX_ERR		= 0x7,
+	RX_DRP_BCAST	= 0x8,
+	RX_DRP_MCAST	= 0x9,
+	RX_DRP_L3BCAST	= 0xa,
+	RX_DRP_L3MCAST	= 0xb,
+	RX_STATS_ENUM_LAST,
+};
+
+struct otx2_dev_stats {
+	u64 rx_bytes;
+	u64 rx_frames;
+	u64 rx_ucast_frames;
+	u64 rx_bcast_frames;
+	u64 rx_mcast_frames;
+	u64 rx_drops;
+
+	u64 tx_bytes;
+	u64 tx_frames;
+	u64 tx_ucast_frames;
+	u64 tx_bcast_frames;
+	u64 tx_mcast_frames;
+	u64 tx_drops;
+};
+
 /* Driver counted stats */
 struct otx2_drv_stats {
 	atomic_t rx_fcs_errs;
@@ -142,6 +185,7 @@ struct otx2_hw {
 	cpumask_var_t           *affinity_mask;
 
 	/* Stats */
+	struct otx2_dev_stats	dev_stats;
 	struct otx2_drv_stats	drv_stats;
 };
 
@@ -545,6 +589,11 @@ void mbox_handler_nix_lf_alloc(struct otx2_nic *pfvf,
 void mbox_handler_nix_txsch_alloc(struct otx2_nic *pf,
 				  struct nix_txsch_alloc_rsp *rsp);
 
+/* Device stats APIs */
+void otx2_get_dev_stats(struct otx2_nic *pfvf);
+void otx2_get_stats64(struct net_device *netdev,
+		      struct rtnl_link_stats64 *stats);
+
 int otx2_open(struct net_device *netdev);
 int otx2_stop(struct net_device *netdev);
 #endif /* OTX2_COMMON_H */
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
index d0809369c367..3092634ef291 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
@@ -1073,6 +1073,7 @@ static const struct net_device_ops otx2_netdev_ops = {
 	.ndo_set_rx_mode	= otx2_set_rx_mode,
 	.ndo_set_features	= otx2_set_features,
 	.ndo_tx_timeout		= otx2_tx_timeout,
+	.ndo_get_stats64	= otx2_get_stats64,
 };
 
 static int otx2_check_pf_usable(struct otx2_nic *nic)
-- 
cgit v1.2.3-59-g8ed1b


From d45d8979840d9c9ac93d3fe8cfc8e794b7228445 Mon Sep 17 00:00:00 2001
From: Christina Jacob <cjacob@marvell.com>
Date: Mon, 27 Jan 2020 18:35:28 +0530
Subject: octeontx2-pf: Add basic ethtool support

This patch adds ethtool support for
 - Driver stats, Tx/Rx perqueue and CGX LMAC stats
 - Set/show Rx/Tx queue count
 - Set/show Rx/Tx ring sizes
 - Set/show IRQ coalescing parameters

Signed-off-by: Christina Jacob <cjacob@marvell.com>
Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../net/ethernet/marvell/octeontx2/nic/Makefile    |   2 +-
 .../ethernet/marvell/octeontx2/nic/otx2_common.c   |  82 ++++-
 .../ethernet/marvell/octeontx2/nic/otx2_common.h   |  10 +
 .../ethernet/marvell/octeontx2/nic/otx2_ethtool.c  | 408 +++++++++++++++++++++
 .../net/ethernet/marvell/octeontx2/nic/otx2_pf.c   |  16 +-
 .../net/ethernet/marvell/octeontx2/nic/otx2_txrx.h |  11 +
 6 files changed, 525 insertions(+), 4 deletions(-)
 create mode 100644 drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c

diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/Makefile b/drivers/net/ethernet/marvell/octeontx2/nic/Makefile
index 0484d709fe68..41bf00cf5b1d 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/Makefile
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/Makefile
@@ -5,6 +5,6 @@
 
 obj-$(CONFIG_OCTEONTX2_PF) += octeontx2_nicpf.o
 
-octeontx2_nicpf-y := otx2_pf.o otx2_common.o otx2_txrx.o
+octeontx2_nicpf-y := otx2_pf.o otx2_common.o otx2_txrx.o otx2_ethtool.o
 
 ccflags-y += -I$(srctree)/drivers/net/ethernet/marvell/octeontx2/af
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
index c4be7872d9f4..07d45596b62e 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
@@ -16,6 +16,72 @@
 #include "otx2_common.h"
 #include "otx2_struct.h"
 
+static void otx2_nix_rq_op_stats(struct queue_stats *stats,
+				 struct otx2_nic *pfvf, int qidx)
+{
+	u64 incr = (u64)qidx << 32;
+	u64 *ptr;
+
+	ptr = (u64 *)otx2_get_regaddr(pfvf, NIX_LF_RQ_OP_OCTS);
+	stats->bytes = otx2_atomic64_add(incr, ptr);
+
+	ptr = (u64 *)otx2_get_regaddr(pfvf, NIX_LF_RQ_OP_PKTS);
+	stats->pkts = otx2_atomic64_add(incr, ptr);
+}
+
+static void otx2_nix_sq_op_stats(struct queue_stats *stats,
+				 struct otx2_nic *pfvf, int qidx)
+{
+	u64 incr = (u64)qidx << 32;
+	u64 *ptr;
+
+	ptr = (u64 *)otx2_get_regaddr(pfvf, NIX_LF_SQ_OP_OCTS);
+	stats->bytes = otx2_atomic64_add(incr, ptr);
+
+	ptr = (u64 *)otx2_get_regaddr(pfvf, NIX_LF_SQ_OP_PKTS);
+	stats->pkts = otx2_atomic64_add(incr, ptr);
+}
+
+void otx2_update_lmac_stats(struct otx2_nic *pfvf)
+{
+	struct msg_req *req;
+
+	if (!netif_running(pfvf->netdev))
+		return;
+
+	otx2_mbox_lock(&pfvf->mbox);
+	req = otx2_mbox_alloc_msg_cgx_stats(&pfvf->mbox);
+	if (!req) {
+		otx2_mbox_unlock(&pfvf->mbox);
+		return;
+	}
+
+	otx2_sync_mbox_msg(&pfvf->mbox);
+	otx2_mbox_unlock(&pfvf->mbox);
+}
+
+int otx2_update_rq_stats(struct otx2_nic *pfvf, int qidx)
+{
+	struct otx2_rcv_queue *rq = &pfvf->qset.rq[qidx];
+
+	if (!pfvf->qset.rq)
+		return 0;
+
+	otx2_nix_rq_op_stats(&rq->stats, pfvf, qidx);
+	return 1;
+}
+
+int otx2_update_sq_stats(struct otx2_nic *pfvf, int qidx)
+{
+	struct otx2_snd_queue *sq = &pfvf->qset.sq[qidx];
+
+	if (!pfvf->qset.sq)
+		return 0;
+
+	otx2_nix_sq_op_stats(&sq->stats, pfvf, qidx);
+	return 1;
+}
+
 void otx2_get_dev_stats(struct otx2_nic *pfvf)
 {
 	struct otx2_dev_stats *dev_stats = &pfvf->hw.dev_stats;
@@ -590,6 +656,9 @@ static int otx2_sq_init(struct otx2_nic *pfvf, u16 qidx, u16 sqb_aura)
 	sq->lmt_addr = (__force u64 *)(pfvf->reg_base + LMT_LF_LMTLINEX(qidx));
 	sq->io_addr = (__force u64)otx2_get_regaddr(pfvf, NIX_LF_OP_SENDX(0));
 
+	sq->stats.bytes = 0;
+	sq->stats.pkts = 0;
+
 	/* Get memory to put this msg */
 	aq = otx2_mbox_alloc_msg_nix_aq_enq(&pfvf->mbox);
 	if (!aq)
@@ -1238,6 +1307,18 @@ void otx2_ctx_disable(struct mbox *mbox, int type, bool npa)
 	otx2_mbox_unlock(mbox);
 }
 
+/* Mbox message handlers */
+void mbox_handler_cgx_stats(struct otx2_nic *pfvf,
+			    struct cgx_stats_rsp *rsp)
+{
+	int id;
+
+	for (id = 0; id < CGX_RX_STATS_COUNT; id++)
+		pfvf->hw.cgx_rx_stats[id] = rsp->rx_stats[id];
+	for (id = 0; id < CGX_TX_STATS_COUNT; id++)
+		pfvf->hw.cgx_tx_stats[id] = rsp->tx_stats[id];
+}
+
 void mbox_handler_nix_txsch_alloc(struct otx2_nic *pf,
 				  struct nix_txsch_alloc_rsp *rsp)
 {
@@ -1250,7 +1331,6 @@ void mbox_handler_nix_txsch_alloc(struct otx2_nic *pf,
 				rsp->schq_list[lvl][schq];
 }
 
-/* Mbox message handlers */
 void mbox_handler_npa_lf_alloc(struct otx2_nic *pfvf,
 			       struct npa_lf_alloc_rsp *rsp)
 {
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
index ce7a552b478a..95557e8425a3 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
@@ -187,6 +187,8 @@ struct otx2_hw {
 	/* Stats */
 	struct otx2_dev_stats	dev_stats;
 	struct otx2_drv_stats	drv_stats;
+	u64			cgx_rx_stats[CGX_RX_STATS_COUNT];
+	u64			cgx_tx_stats[CGX_TX_STATS_COUNT];
 };
 
 struct refill_work {
@@ -588,12 +590,20 @@ void mbox_handler_nix_lf_alloc(struct otx2_nic *pfvf,
 			       struct nix_lf_alloc_rsp *rsp);
 void mbox_handler_nix_txsch_alloc(struct otx2_nic *pf,
 				  struct nix_txsch_alloc_rsp *rsp);
+void mbox_handler_cgx_stats(struct otx2_nic *pfvf,
+			    struct cgx_stats_rsp *rsp);
 
 /* Device stats APIs */
 void otx2_get_dev_stats(struct otx2_nic *pfvf);
 void otx2_get_stats64(struct net_device *netdev,
 		      struct rtnl_link_stats64 *stats);
+void otx2_update_lmac_stats(struct otx2_nic *pfvf);
+int otx2_update_rq_stats(struct otx2_nic *pfvf, int qidx);
+int otx2_update_sq_stats(struct otx2_nic *pfvf, int qidx);
+void otx2_set_ethtool_ops(struct net_device *netdev);
 
 int otx2_open(struct net_device *netdev);
 int otx2_stop(struct net_device *netdev);
+int otx2_set_real_num_queues(struct net_device *netdev,
+			     int tx_queues, int rx_queues);
 #endif /* OTX2_COMMON_H */
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c
new file mode 100644
index 000000000000..b1f61e069ed8
--- /dev/null
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c
@@ -0,0 +1,408 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Marvell OcteonTx2 RVU Ethernet driver
+ *
+ * Copyright (C) 2020 Marvell International Ltd.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/pci.h>
+#include <linux/ethtool.h>
+#include <linux/stddef.h>
+#include <linux/etherdevice.h>
+#include <linux/log2.h>
+
+#include "otx2_common.h"
+
+#define DRV_NAME	"octeontx2-nicpf"
+
+struct otx2_stat {
+	char name[ETH_GSTRING_LEN];
+	unsigned int index;
+};
+
+/* HW device stats */
+#define OTX2_DEV_STAT(stat) { \
+	.name = #stat, \
+	.index = offsetof(struct otx2_dev_stats, stat) / sizeof(u64), \
+}
+
+static const struct otx2_stat otx2_dev_stats[] = {
+	OTX2_DEV_STAT(rx_ucast_frames),
+	OTX2_DEV_STAT(rx_bcast_frames),
+	OTX2_DEV_STAT(rx_mcast_frames),
+
+	OTX2_DEV_STAT(tx_ucast_frames),
+	OTX2_DEV_STAT(tx_bcast_frames),
+	OTX2_DEV_STAT(tx_mcast_frames),
+};
+
+/* Driver level stats */
+#define OTX2_DRV_STAT(stat) { \
+	.name = #stat, \
+	.index = offsetof(struct otx2_drv_stats, stat) / sizeof(atomic_t), \
+}
+
+static const struct otx2_stat otx2_drv_stats[] = {
+	OTX2_DRV_STAT(rx_fcs_errs),
+	OTX2_DRV_STAT(rx_oversize_errs),
+	OTX2_DRV_STAT(rx_undersize_errs),
+	OTX2_DRV_STAT(rx_csum_errs),
+	OTX2_DRV_STAT(rx_len_errs),
+	OTX2_DRV_STAT(rx_other_errs),
+};
+
+static const struct otx2_stat otx2_queue_stats[] = {
+	{ "bytes", 0 },
+	{ "frames", 1 },
+};
+
+static const unsigned int otx2_n_dev_stats = ARRAY_SIZE(otx2_dev_stats);
+static const unsigned int otx2_n_drv_stats = ARRAY_SIZE(otx2_drv_stats);
+static const unsigned int otx2_n_queue_stats = ARRAY_SIZE(otx2_queue_stats);
+
+static void otx2_dev_open(struct net_device *netdev)
+{
+	otx2_open(netdev);
+}
+
+static void otx2_dev_stop(struct net_device *netdev)
+{
+	otx2_stop(netdev);
+}
+
+static void otx2_get_drvinfo(struct net_device *netdev,
+			     struct ethtool_drvinfo *info)
+{
+	struct otx2_nic *pfvf = netdev_priv(netdev);
+
+	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
+	strlcpy(info->bus_info, pci_name(pfvf->pdev), sizeof(info->bus_info));
+}
+
+static void otx2_get_qset_strings(struct otx2_nic *pfvf, u8 **data, int qset)
+{
+	int start_qidx = qset * pfvf->hw.rx_queues;
+	int qidx, stats;
+
+	for (qidx = 0; qidx < pfvf->hw.rx_queues; qidx++) {
+		for (stats = 0; stats < otx2_n_queue_stats; stats++) {
+			sprintf(*data, "rxq%d: %s", qidx + start_qidx,
+				otx2_queue_stats[stats].name);
+			*data += ETH_GSTRING_LEN;
+		}
+	}
+	for (qidx = 0; qidx < pfvf->hw.tx_queues; qidx++) {
+		for (stats = 0; stats < otx2_n_queue_stats; stats++) {
+			sprintf(*data, "txq%d: %s", qidx + start_qidx,
+				otx2_queue_stats[stats].name);
+			*data += ETH_GSTRING_LEN;
+		}
+	}
+}
+
+static void otx2_get_strings(struct net_device *netdev, u32 sset, u8 *data)
+{
+	struct otx2_nic *pfvf = netdev_priv(netdev);
+	int stats;
+
+	if (sset != ETH_SS_STATS)
+		return;
+
+	for (stats = 0; stats < otx2_n_dev_stats; stats++) {
+		memcpy(data, otx2_dev_stats[stats].name, ETH_GSTRING_LEN);
+		data += ETH_GSTRING_LEN;
+	}
+
+	for (stats = 0; stats < otx2_n_drv_stats; stats++) {
+		memcpy(data, otx2_drv_stats[stats].name, ETH_GSTRING_LEN);
+		data += ETH_GSTRING_LEN;
+	}
+
+	otx2_get_qset_strings(pfvf, &data, 0);
+
+	for (stats = 0; stats < CGX_RX_STATS_COUNT; stats++) {
+		sprintf(data, "cgx_rxstat%d: ", stats);
+		data += ETH_GSTRING_LEN;
+	}
+
+	for (stats = 0; stats < CGX_TX_STATS_COUNT; stats++) {
+		sprintf(data, "cgx_txstat%d: ", stats);
+		data += ETH_GSTRING_LEN;
+	}
+
+	strcpy(data, "reset_count");
+	data += ETH_GSTRING_LEN;
+}
+
+static void otx2_get_qset_stats(struct otx2_nic *pfvf,
+				struct ethtool_stats *stats, u64 **data)
+{
+	int stat, qidx;
+
+	if (!pfvf)
+		return;
+	for (qidx = 0; qidx < pfvf->hw.rx_queues; qidx++) {
+		if (!otx2_update_rq_stats(pfvf, qidx)) {
+			for (stat = 0; stat < otx2_n_queue_stats; stat++)
+				*((*data)++) = 0;
+			continue;
+		}
+		for (stat = 0; stat < otx2_n_queue_stats; stat++)
+			*((*data)++) = ((u64 *)&pfvf->qset.rq[qidx].stats)
+				[otx2_queue_stats[stat].index];
+	}
+
+	for (qidx = 0; qidx < pfvf->hw.tx_queues; qidx++) {
+		if (!otx2_update_sq_stats(pfvf, qidx)) {
+			for (stat = 0; stat < otx2_n_queue_stats; stat++)
+				*((*data)++) = 0;
+			continue;
+		}
+		for (stat = 0; stat < otx2_n_queue_stats; stat++)
+			*((*data)++) = ((u64 *)&pfvf->qset.sq[qidx].stats)
+				[otx2_queue_stats[stat].index];
+	}
+}
+
+/* Get device and per queue statistics */
+static void otx2_get_ethtool_stats(struct net_device *netdev,
+				   struct ethtool_stats *stats, u64 *data)
+{
+	struct otx2_nic *pfvf = netdev_priv(netdev);
+	int stat;
+
+	otx2_get_dev_stats(pfvf);
+	for (stat = 0; stat < otx2_n_dev_stats; stat++)
+		*(data++) = ((u64 *)&pfvf->hw.dev_stats)
+				[otx2_dev_stats[stat].index];
+
+	for (stat = 0; stat < otx2_n_drv_stats; stat++)
+		*(data++) = atomic_read(&((atomic_t *)&pfvf->hw.drv_stats)
+						[otx2_drv_stats[stat].index]);
+
+	otx2_get_qset_stats(pfvf, stats, &data);
+	otx2_update_lmac_stats(pfvf);
+	for (stat = 0; stat < CGX_RX_STATS_COUNT; stat++)
+		*(data++) = pfvf->hw.cgx_rx_stats[stat];
+	for (stat = 0; stat < CGX_TX_STATS_COUNT; stat++)
+		*(data++) = pfvf->hw.cgx_tx_stats[stat];
+	*(data++) = pfvf->reset_count;
+}
+
+static int otx2_get_sset_count(struct net_device *netdev, int sset)
+{
+	struct otx2_nic *pfvf = netdev_priv(netdev);
+	int qstats_count;
+
+	if (sset != ETH_SS_STATS)
+		return -EINVAL;
+
+	qstats_count = otx2_n_queue_stats *
+		       (pfvf->hw.rx_queues + pfvf->hw.tx_queues);
+
+	return otx2_n_dev_stats + otx2_n_drv_stats + qstats_count +
+		CGX_RX_STATS_COUNT + CGX_TX_STATS_COUNT + 1;
+}
+
+/* Get no of queues device supports and current queue count */
+static void otx2_get_channels(struct net_device *dev,
+			      struct ethtool_channels *channel)
+{
+	struct otx2_nic *pfvf = netdev_priv(dev);
+
+	channel->max_rx = pfvf->hw.max_queues;
+	channel->max_tx = pfvf->hw.max_queues;
+
+	channel->rx_count = pfvf->hw.rx_queues;
+	channel->tx_count = pfvf->hw.tx_queues;
+}
+
+/* Set no of Tx, Rx queues to be used */
+static int otx2_set_channels(struct net_device *dev,
+			     struct ethtool_channels *channel)
+{
+	struct otx2_nic *pfvf = netdev_priv(dev);
+	bool if_up = netif_running(dev);
+	int err = 0;
+
+	if (!channel->rx_count || !channel->tx_count)
+		return -EINVAL;
+
+	if (if_up)
+		otx2_dev_stop(dev);
+
+	err = otx2_set_real_num_queues(dev, channel->tx_count,
+				       channel->rx_count);
+	if (err)
+		goto fail;
+
+	pfvf->hw.rx_queues = channel->rx_count;
+	pfvf->hw.tx_queues = channel->tx_count;
+	pfvf->qset.cq_cnt = pfvf->hw.tx_queues +  pfvf->hw.rx_queues;
+
+fail:
+	if (if_up)
+		otx2_dev_open(dev);
+
+	netdev_info(dev, "Setting num Tx rings to %d, Rx rings to %d success\n",
+		    pfvf->hw.tx_queues, pfvf->hw.rx_queues);
+
+	return err;
+}
+
+static void otx2_get_ringparam(struct net_device *netdev,
+			       struct ethtool_ringparam *ring)
+{
+	struct otx2_nic *pfvf = netdev_priv(netdev);
+	struct otx2_qset *qs = &pfvf->qset;
+
+	ring->rx_max_pending = Q_COUNT(Q_SIZE_MAX);
+	ring->rx_pending = qs->rqe_cnt ? qs->rqe_cnt : Q_COUNT(Q_SIZE_256);
+	ring->tx_max_pending = Q_COUNT(Q_SIZE_MAX);
+	ring->tx_pending = qs->sqe_cnt ? qs->sqe_cnt : Q_COUNT(Q_SIZE_4K);
+}
+
+static int otx2_set_ringparam(struct net_device *netdev,
+			      struct ethtool_ringparam *ring)
+{
+	struct otx2_nic *pfvf = netdev_priv(netdev);
+	bool if_up = netif_running(netdev);
+	struct otx2_qset *qs = &pfvf->qset;
+	u32 rx_count, tx_count;
+
+	if (ring->rx_mini_pending || ring->rx_jumbo_pending)
+		return -EINVAL;
+
+	/* Permitted lengths are 16 64 256 1K 4K 16K 64K 256K 1M  */
+	rx_count = ring->rx_pending;
+	/* On some silicon variants a skid or reserved CQEs are
+	 * needed to avoid CQ overflow.
+	 */
+	if (rx_count < pfvf->hw.rq_skid)
+		rx_count =  pfvf->hw.rq_skid;
+	rx_count = Q_COUNT(Q_SIZE(rx_count, 3));
+
+	/* Due pipelining impact minimum 2000 unused SQ CQE's
+	 * need to be maintained to avoid CQ overflow, hence the
+	 * minimum 4K size.
+	 */
+	tx_count = clamp_t(u32, ring->tx_pending,
+			   Q_COUNT(Q_SIZE_4K), Q_COUNT(Q_SIZE_MAX));
+	tx_count = Q_COUNT(Q_SIZE(tx_count, 3));
+
+	if (tx_count == qs->sqe_cnt && rx_count == qs->rqe_cnt)
+		return 0;
+
+	if (if_up)
+		otx2_dev_stop(netdev);
+
+	/* Assigned to the nearest possible exponent. */
+	qs->sqe_cnt = tx_count;
+	qs->rqe_cnt = rx_count;
+
+	if (if_up)
+		otx2_dev_open(netdev);
+	return 0;
+}
+
+static int otx2_get_coalesce(struct net_device *netdev,
+			     struct ethtool_coalesce *cmd)
+{
+	struct otx2_nic *pfvf = netdev_priv(netdev);
+	struct otx2_hw *hw = &pfvf->hw;
+
+	cmd->rx_coalesce_usecs = hw->cq_time_wait;
+	cmd->rx_max_coalesced_frames = hw->cq_ecount_wait;
+	cmd->tx_coalesce_usecs = hw->cq_time_wait;
+	cmd->tx_max_coalesced_frames = hw->cq_ecount_wait;
+
+	return 0;
+}
+
+static int otx2_set_coalesce(struct net_device *netdev,
+			     struct ethtool_coalesce *ec)
+{
+	struct otx2_nic *pfvf = netdev_priv(netdev);
+	struct otx2_hw *hw = &pfvf->hw;
+	int qidx;
+
+	if (ec->use_adaptive_rx_coalesce || ec->use_adaptive_tx_coalesce ||
+	    ec->rx_coalesce_usecs_irq || ec->rx_max_coalesced_frames_irq ||
+	    ec->tx_coalesce_usecs_irq || ec->tx_max_coalesced_frames_irq ||
+	    ec->stats_block_coalesce_usecs || ec->pkt_rate_low ||
+	    ec->rx_coalesce_usecs_low || ec->rx_max_coalesced_frames_low ||
+	    ec->tx_coalesce_usecs_low || ec->tx_max_coalesced_frames_low ||
+	    ec->pkt_rate_high || ec->rx_coalesce_usecs_high ||
+	    ec->rx_max_coalesced_frames_high || ec->tx_coalesce_usecs_high ||
+	    ec->tx_max_coalesced_frames_high || ec->rate_sample_interval)
+		return -EOPNOTSUPP;
+
+	if (!ec->rx_max_coalesced_frames || !ec->tx_max_coalesced_frames)
+		return 0;
+
+	/* 'cq_time_wait' is 8bit and is in multiple of 100ns,
+	 * so clamp the user given value to the range of 1 to 25usec.
+	 */
+	ec->rx_coalesce_usecs = clamp_t(u32, ec->rx_coalesce_usecs,
+					1, CQ_TIMER_THRESH_MAX);
+	ec->tx_coalesce_usecs = clamp_t(u32, ec->tx_coalesce_usecs,
+					1, CQ_TIMER_THRESH_MAX);
+
+	/* Rx and Tx are mapped to same CQ, check which one
+	 * is changed, if both then choose the min.
+	 */
+	if (hw->cq_time_wait == ec->rx_coalesce_usecs)
+		hw->cq_time_wait = ec->tx_coalesce_usecs;
+	else if (hw->cq_time_wait == ec->tx_coalesce_usecs)
+		hw->cq_time_wait = ec->rx_coalesce_usecs;
+	else
+		hw->cq_time_wait = min_t(u8, ec->rx_coalesce_usecs,
+					 ec->tx_coalesce_usecs);
+
+	/* Max ecount_wait supported is 16bit,
+	 * so clamp the user given value to the range of 1 to 64k.
+	 */
+	ec->rx_max_coalesced_frames = clamp_t(u32, ec->rx_max_coalesced_frames,
+					      1, U16_MAX);
+	ec->tx_max_coalesced_frames = clamp_t(u32, ec->tx_max_coalesced_frames,
+					      1, U16_MAX);
+
+	/* Rx and Tx are mapped to same CQ, check which one
+	 * is changed, if both then choose the min.
+	 */
+	if (hw->cq_ecount_wait == ec->rx_max_coalesced_frames)
+		hw->cq_ecount_wait = ec->tx_max_coalesced_frames;
+	else if (hw->cq_ecount_wait == ec->tx_max_coalesced_frames)
+		hw->cq_ecount_wait = ec->rx_max_coalesced_frames;
+	else
+		hw->cq_ecount_wait = min_t(u16, ec->rx_max_coalesced_frames,
+					   ec->tx_max_coalesced_frames);
+
+	if (netif_running(netdev)) {
+		for (qidx = 0; qidx < pfvf->hw.cint_cnt; qidx++)
+			otx2_config_irq_coalescing(pfvf, qidx);
+	}
+
+	return 0;
+}
+
+static const struct ethtool_ops otx2_ethtool_ops = {
+	.get_drvinfo		= otx2_get_drvinfo,
+	.get_strings		= otx2_get_strings,
+	.get_ethtool_stats	= otx2_get_ethtool_stats,
+	.get_sset_count		= otx2_get_sset_count,
+	.set_channels		= otx2_set_channels,
+	.get_channels		= otx2_get_channels,
+	.get_ringparam		= otx2_get_ringparam,
+	.set_ringparam		= otx2_set_ringparam,
+	.get_coalesce		= otx2_get_coalesce,
+	.set_coalesce		= otx2_set_coalesce,
+};
+
+void otx2_set_ethtool_ops(struct net_device *netdev)
+{
+	netdev->ethtool_ops = &otx2_ethtool_ops;
+}
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
index 3092634ef291..85f9b9ba6bd5 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
@@ -148,6 +148,9 @@ static void otx2_process_pfaf_mbox_msg(struct otx2_nic *pf,
 		mbox_handler_nix_txsch_alloc(pf,
 					     (struct nix_txsch_alloc_rsp *)msg);
 		break;
+	case MBOX_MSG_CGX_STATS:
+		mbox_handler_cgx_stats(pf, (struct cgx_stats_rsp *)msg);
+		break;
 	default:
 		if (msg->rc)
 			dev_err(pf->dev,
@@ -459,8 +462,8 @@ static int otx2_cgx_config_loopback(struct otx2_nic *pf, bool enable)
 	return err;
 }
 
-static int otx2_set_real_num_queues(struct net_device *netdev,
-				    int tx_queues, int rx_queues)
+int otx2_set_real_num_queues(struct net_device *netdev,
+			     int tx_queues, int rx_queues)
 {
 	int err;
 
@@ -812,6 +815,11 @@ int otx2_open(struct net_device *netdev)
 	if (!qset->sq)
 		goto err_free_mem;
 
+	qset->rq = kcalloc(pf->hw.rx_queues,
+			   sizeof(struct otx2_rcv_queue), GFP_KERNEL);
+	if (!qset->rq)
+		goto err_free_mem;
+
 	err = otx2_init_hw_resources(pf);
 	if (err)
 		goto err_free_mem;
@@ -917,6 +925,7 @@ err_disable_napi:
 err_free_mem:
 	kfree(qset->sq);
 	kfree(qset->cq);
+	kfree(qset->rq);
 	kfree(qset->napi);
 	return err;
 }
@@ -973,6 +982,7 @@ int otx2_stop(struct net_device *netdev)
 
 	kfree(qset->sq);
 	kfree(qset->cq);
+	kfree(qset->rq);
 	kfree(qset->napi);
 	/* Do not clear RQ/SQ ringsize settings */
 	memset((void *)qset + offsetof(struct otx2_qset, sqe_cnt), 0,
@@ -1268,6 +1278,8 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 		goto err_detach_rsrc;
 	}
 
+	otx2_set_ethtool_ops(netdev);
+
 	/* Enable link notifications */
 	otx2_cgx_config_linkevents(pf, true);
 
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h
index 107a261492ba..4ab32d3adb78 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h
@@ -60,6 +60,15 @@
  */
 #define CQ_QCOUNT_DEFAULT	1
 
+struct queue_stats {
+	u64	bytes;
+	u64	pkts;
+};
+
+struct otx2_rcv_queue {
+	struct queue_stats	stats;
+};
+
 struct sg_list {
 	u16	num_segs;
 	u64	skb;
@@ -82,6 +91,7 @@ struct otx2_snd_queue {
 	struct qmem		*sqe;
 	struct qmem		*tso_hdrs;
 	struct sg_list		*sg;
+	struct queue_stats	stats;
 	u16			sqb_count;
 	u64			*sqb_ptrs;
 } ____cacheline_aligned_in_smp;
@@ -134,6 +144,7 @@ struct otx2_qset {
 	struct otx2_cq_poll	*napi;
 	struct otx2_cq_queue	*cq;
 	struct otx2_snd_queue	*sq;
+	struct otx2_rcv_queue	*rq;
 };
 
 /* Translate IOVA to physical address */
-- 
cgit v1.2.3-59-g8ed1b


From 6e92d71bf88208ddcbd48b3c831c0e394d0f56e2 Mon Sep 17 00:00:00 2001
From: Sunil Goutham <sgoutham@marvell.com>
Date: Mon, 27 Jan 2020 18:35:29 +0530
Subject: octeontx2-pf: ethtool RSS config support

Added support to show or configure RSS hash key, indirection table,
2,4 tuple via ethtool. Also added debug msg_level support
to dump messages when HW reports errors in packet received
or transmitted.

Signed-off-by: Prakash Brahmajyosyula <bprakash@marvell.com>
Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../ethernet/marvell/octeontx2/nic/otx2_common.c   |   6 +-
 .../ethernet/marvell/octeontx2/nic/otx2_common.h   |   6 +
 .../ethernet/marvell/octeontx2/nic/otx2_ethtool.c  | 254 +++++++++++++++++++++
 .../net/ethernet/marvell/octeontx2/nic/otx2_txrx.c |   7 +-
 4 files changed, 269 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
index 07d45596b62e..8247d21d0432 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
@@ -220,7 +220,7 @@ int otx2_hw_set_mtu(struct otx2_nic *pfvf, int mtu)
 	return err;
 }
 
-static int otx2_set_flowkey_cfg(struct otx2_nic *pfvf)
+int otx2_set_flowkey_cfg(struct otx2_nic *pfvf)
 {
 	struct otx2_rss_info *rss = &pfvf->hw.rss_info;
 	struct nix_rss_flowkey_cfg *req;
@@ -241,7 +241,7 @@ static int otx2_set_flowkey_cfg(struct otx2_nic *pfvf)
 	return err;
 }
 
-static int otx2_set_rss_table(struct otx2_nic *pfvf)
+int otx2_set_rss_table(struct otx2_nic *pfvf)
 {
 	struct otx2_rss_info *rss = &pfvf->hw.rss_info;
 	struct mbox *mbox = &pfvf->mbox;
@@ -280,7 +280,7 @@ static int otx2_set_rss_table(struct otx2_nic *pfvf)
 	return err;
 }
 
-static void otx2_set_rss_key(struct otx2_nic *pfvf)
+void otx2_set_rss_key(struct otx2_nic *pfvf)
 {
 	struct otx2_rss_info *rss = &pfvf->hw.rss_info;
 	u64 *key = (u64 *)&rss->key[4];
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
index 95557e8425a3..320f3b7bf57f 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
@@ -222,6 +222,9 @@ struct otx2_nic {
 	struct work_struct	reset_task;
 	struct refill_work	*refill_wrk;
 
+	/* Ethtool stuff */
+	u32			msg_enable;
+
 	/* Block address of NIX either BLKADDR_NIX0 or BLKADDR_NIX1 */
 	int			nix_blkaddr;
 };
@@ -580,6 +583,9 @@ void otx2_cleanup_tx_cqes(struct otx2_nic *pfvf, struct otx2_cq_queue *cq);
 
 /* RSS configuration APIs*/
 int otx2_rss_init(struct otx2_nic *pfvf);
+int otx2_set_flowkey_cfg(struct otx2_nic *pfvf);
+void otx2_set_rss_key(struct otx2_nic *pfvf);
+int otx2_set_rss_table(struct otx2_nic *pfvf);
 
 /* Mbox handlers */
 void mbox_handler_msix_offset(struct otx2_nic *pfvf,
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c
index b1f61e069ed8..60fcf82dd8cb 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c
@@ -389,7 +389,253 @@ static int otx2_set_coalesce(struct net_device *netdev,
 	return 0;
 }
 
+static int otx2_get_rss_hash_opts(struct otx2_nic *pfvf,
+				  struct ethtool_rxnfc *nfc)
+{
+	struct otx2_rss_info *rss = &pfvf->hw.rss_info;
+
+	if (!(rss->flowkey_cfg &
+	    (NIX_FLOW_KEY_TYPE_IPV4 | NIX_FLOW_KEY_TYPE_IPV6)))
+		return 0;
+
+	/* Mimimum is IPv4 and IPv6, SIP/DIP */
+	nfc->data = RXH_IP_SRC | RXH_IP_DST;
+
+	switch (nfc->flow_type) {
+	case TCP_V4_FLOW:
+	case TCP_V6_FLOW:
+		if (rss->flowkey_cfg & NIX_FLOW_KEY_TYPE_TCP)
+			nfc->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		break;
+	case UDP_V4_FLOW:
+	case UDP_V6_FLOW:
+		if (rss->flowkey_cfg & NIX_FLOW_KEY_TYPE_UDP)
+			nfc->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		break;
+	case SCTP_V4_FLOW:
+	case SCTP_V6_FLOW:
+		if (rss->flowkey_cfg & NIX_FLOW_KEY_TYPE_SCTP)
+			nfc->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+		break;
+	case AH_ESP_V4_FLOW:
+	case AH_V4_FLOW:
+	case ESP_V4_FLOW:
+	case IPV4_FLOW:
+	case AH_ESP_V6_FLOW:
+	case AH_V6_FLOW:
+	case ESP_V6_FLOW:
+	case IPV6_FLOW:
+		break;
+	default:
+		return -EINVAL;
+	}
+	return 0;
+}
+
+static int otx2_set_rss_hash_opts(struct otx2_nic *pfvf,
+				  struct ethtool_rxnfc *nfc)
+{
+	struct otx2_rss_info *rss = &pfvf->hw.rss_info;
+	u32 rxh_l4 = RXH_L4_B_0_1 | RXH_L4_B_2_3;
+	u32 rss_cfg = rss->flowkey_cfg;
+
+	if (!rss->enable) {
+		netdev_err(pfvf->netdev,
+			   "RSS is disabled, cannot change settings\n");
+		return -EIO;
+	}
+
+	/* Mimimum is IPv4 and IPv6, SIP/DIP */
+	if (!(nfc->data & RXH_IP_SRC) || !(nfc->data & RXH_IP_DST))
+		return -EINVAL;
+
+	switch (nfc->flow_type) {
+	case TCP_V4_FLOW:
+	case TCP_V6_FLOW:
+		/* Different config for v4 and v6 is not supported.
+		 * Both of them have to be either 4-tuple or 2-tuple.
+		 */
+		switch (nfc->data & rxh_l4) {
+		case 0:
+			rss_cfg &= ~NIX_FLOW_KEY_TYPE_TCP;
+			break;
+		case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
+			rss_cfg |= NIX_FLOW_KEY_TYPE_TCP;
+			break;
+		default:
+			return -EINVAL;
+		}
+		break;
+	case UDP_V4_FLOW:
+	case UDP_V6_FLOW:
+		switch (nfc->data & rxh_l4) {
+		case 0:
+			rss_cfg &= ~NIX_FLOW_KEY_TYPE_UDP;
+			break;
+		case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
+			rss_cfg |= NIX_FLOW_KEY_TYPE_UDP;
+			break;
+		default:
+			return -EINVAL;
+		}
+		break;
+	case SCTP_V4_FLOW:
+	case SCTP_V6_FLOW:
+		switch (nfc->data & rxh_l4) {
+		case 0:
+			rss_cfg &= ~NIX_FLOW_KEY_TYPE_SCTP;
+			break;
+		case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
+			rss_cfg |= NIX_FLOW_KEY_TYPE_SCTP;
+			break;
+		default:
+			return -EINVAL;
+		}
+		break;
+	case IPV4_FLOW:
+	case IPV6_FLOW:
+		rss_cfg = NIX_FLOW_KEY_TYPE_IPV4 | NIX_FLOW_KEY_TYPE_IPV6;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	rss->flowkey_cfg = rss_cfg;
+	otx2_set_flowkey_cfg(pfvf);
+	return 0;
+}
+
+static int otx2_get_rxnfc(struct net_device *dev,
+			  struct ethtool_rxnfc *nfc, u32 *rules)
+{
+	struct otx2_nic *pfvf = netdev_priv(dev);
+	int ret = -EOPNOTSUPP;
+
+	switch (nfc->cmd) {
+	case ETHTOOL_GRXRINGS:
+		nfc->data = pfvf->hw.rx_queues;
+		ret = 0;
+		break;
+	case ETHTOOL_GRXFH:
+		return otx2_get_rss_hash_opts(pfvf, nfc);
+	default:
+		break;
+	}
+	return ret;
+}
+
+static int otx2_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *nfc)
+{
+	struct otx2_nic *pfvf = netdev_priv(dev);
+	int ret = -EOPNOTSUPP;
+
+	switch (nfc->cmd) {
+	case ETHTOOL_SRXFH:
+		ret = otx2_set_rss_hash_opts(pfvf, nfc);
+		break;
+	default:
+		break;
+	}
+
+	return ret;
+}
+
+static u32 otx2_get_rxfh_key_size(struct net_device *netdev)
+{
+	struct otx2_nic *pfvf = netdev_priv(netdev);
+	struct otx2_rss_info *rss;
+
+	rss = &pfvf->hw.rss_info;
+
+	return sizeof(rss->key);
+}
+
+static u32 otx2_get_rxfh_indir_size(struct net_device *dev)
+{
+	struct otx2_nic *pfvf = netdev_priv(dev);
+
+	return pfvf->hw.rss_info.rss_size;
+}
+
+/* Get RSS configuration */
+static int otx2_get_rxfh(struct net_device *dev, u32 *indir,
+			 u8 *hkey, u8 *hfunc)
+{
+	struct otx2_nic *pfvf = netdev_priv(dev);
+	struct otx2_rss_info *rss;
+	int idx;
+
+	rss = &pfvf->hw.rss_info;
+
+	if (indir) {
+		for (idx = 0; idx < rss->rss_size; idx++)
+			indir[idx] = rss->ind_tbl[idx];
+	}
+
+	if (hkey)
+		memcpy(hkey, rss->key, sizeof(rss->key));
+
+	if (hfunc)
+		*hfunc = ETH_RSS_HASH_TOP;
+
+	return 0;
+}
+
+/* Configure RSS table and hash key */
+static int otx2_set_rxfh(struct net_device *dev, const u32 *indir,
+			 const u8 *hkey, const u8 hfunc)
+{
+	struct otx2_nic *pfvf = netdev_priv(dev);
+	struct otx2_rss_info *rss;
+	int idx;
+
+	if (hfunc != ETH_RSS_HASH_NO_CHANGE && hfunc != ETH_RSS_HASH_TOP)
+		return -EOPNOTSUPP;
+
+	rss = &pfvf->hw.rss_info;
+
+	if (!rss->enable) {
+		netdev_err(dev, "RSS is disabled, cannot change settings\n");
+		return -EIO;
+	}
+
+	if (indir) {
+		for (idx = 0; idx < rss->rss_size; idx++)
+			rss->ind_tbl[idx] = indir[idx];
+	}
+
+	if (hkey) {
+		memcpy(rss->key, hkey, sizeof(rss->key));
+		otx2_set_rss_key(pfvf);
+	}
+
+	otx2_set_rss_table(pfvf);
+	return 0;
+}
+
+static u32 otx2_get_msglevel(struct net_device *netdev)
+{
+	struct otx2_nic *pfvf = netdev_priv(netdev);
+
+	return pfvf->msg_enable;
+}
+
+static void otx2_set_msglevel(struct net_device *netdev, u32 val)
+{
+	struct otx2_nic *pfvf = netdev_priv(netdev);
+
+	pfvf->msg_enable = val;
+}
+
+static u32 otx2_get_link(struct net_device *netdev)
+{
+	struct otx2_nic *pfvf = netdev_priv(netdev);
+
+	return pfvf->linfo.link_up;
+}
+
 static const struct ethtool_ops otx2_ethtool_ops = {
+	.get_link		= otx2_get_link,
 	.get_drvinfo		= otx2_get_drvinfo,
 	.get_strings		= otx2_get_strings,
 	.get_ethtool_stats	= otx2_get_ethtool_stats,
@@ -400,6 +646,14 @@ static const struct ethtool_ops otx2_ethtool_ops = {
 	.set_ringparam		= otx2_set_ringparam,
 	.get_coalesce		= otx2_get_coalesce,
 	.set_coalesce		= otx2_set_coalesce,
+	.get_rxnfc		= otx2_get_rxnfc,
+	.set_rxnfc              = otx2_set_rxnfc,
+	.get_rxfh_key_size	= otx2_get_rxfh_key_size,
+	.get_rxfh_indir_size	= otx2_get_rxfh_indir_size,
+	.get_rxfh		= otx2_get_rxfh,
+	.set_rxfh		= otx2_set_rxfh,
+	.get_msglevel		= otx2_get_msglevel,
+	.set_msglevel		= otx2_set_msglevel,
 };
 
 void otx2_set_ethtool_ops(struct net_device *netdev)
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c
index 6a7ca3b4b9be..bef4c20fe314 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c
@@ -84,7 +84,7 @@ static void otx2_snd_pkt_handler(struct otx2_nic *pfvf,
 	struct sk_buff *skb = NULL;
 	struct sg_list *sg;
 
-	if (unlikely(snd_comp->status))
+	if (unlikely(snd_comp->status) && netif_msg_tx_err(pfvf))
 		net_err_ratelimited("%s: TX%d: Error in send CQ status:%x\n",
 				    pfvf->netdev->name, cq->cint_idx,
 				    snd_comp->status);
@@ -144,6 +144,11 @@ static bool otx2_check_rcv_errors(struct otx2_nic *pfvf,
 	struct otx2_drv_stats *stats = &pfvf->hw.drv_stats;
 	struct nix_rx_parse_s *parse = &cqe->parse;
 
+	if (netif_msg_rx_err(pfvf))
+		netdev_err(pfvf->netdev,
+			   "RQ%d: Error pkt with errlev:0x%x errcode:0x%x\n",
+			   qidx, parse->errlev, parse->errcode);
+
 	if (parse->errlev == NPC_ERRLVL_RE) {
 		switch (parse->errcode) {
 		case ERRCODE_FCS:
-- 
cgit v1.2.3-59-g8ed1b


From 493aeb26e12ad4f0f7ad7f320b5706b6902e6eed Mon Sep 17 00:00:00 2001
From: Sunil Goutham <sgoutham@marvell.com>
Date: Mon, 27 Jan 2020 18:35:30 +0530
Subject: Documentation: net: octeontx2: Add RVU HW and drivers overview

Added high level overview of OcteonTx2 RVU HW and functionality of
various drivers which will be upstreamed.

Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/device_drivers/index.rst  |   1 +
 .../device_drivers/marvell/octeontx2.rst           | 159 +++++++++++++++++++++
 MAINTAINERS                                        |   1 +
 3 files changed, 161 insertions(+)
 create mode 100644 Documentation/networking/device_drivers/marvell/octeontx2.rst

diff --git a/Documentation/networking/device_drivers/index.rst b/Documentation/networking/device_drivers/index.rst
index 4bc6ff29976a..a191faaf97de 100644
--- a/Documentation/networking/device_drivers/index.rst
+++ b/Documentation/networking/device_drivers/index.rst
@@ -22,6 +22,7 @@ Contents:
    intel/iavf
    intel/ice
    google/gve
+   marvell/octeontx2
    mellanox/mlx5
    netronome/nfp
    pensando/ionic
diff --git a/Documentation/networking/device_drivers/marvell/octeontx2.rst b/Documentation/networking/device_drivers/marvell/octeontx2.rst
new file mode 100644
index 000000000000..88f508338c5f
--- /dev/null
+++ b/Documentation/networking/device_drivers/marvell/octeontx2.rst
@@ -0,0 +1,159 @@
+.. SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+
+====================================
+Marvell OcteonTx2 RVU Kernel Drivers
+====================================
+
+Copyright (c) 2020 Marvell International Ltd.
+
+Contents
+========
+
+- `Overview`_
+- `Drivers`_
+- `Basic packet flow`_
+
+Overview
+========
+
+Resource virtualization unit (RVU) on Marvell's OcteonTX2 SOC maps HW
+resources from the network, crypto and other functional blocks into
+PCI-compatible physical and virtual functions. Each functional block
+again has multiple local functions (LFs) for provisioning to PCI devices.
+RVU supports multiple PCIe SRIOV physical functions (PFs) and virtual
+functions (VFs). PF0 is called the administrative / admin function (AF)
+and has privileges to provision RVU functional block's LFs to each of the
+PF/VF.
+
+RVU managed networking functional blocks
+ - Network pool or buffer allocator (NPA)
+ - Network interface controller (NIX)
+ - Network parser CAM (NPC)
+ - Schedule/Synchronize/Order unit (SSO)
+ - Loopback interface (LBK)
+
+RVU managed non-networking functional blocks
+ - Crypto accelerator (CPT)
+ - Scheduled timers unit (TIM)
+ - Schedule/Synchronize/Order unit (SSO)
+   Used for both networking and non networking usecases
+
+Resource provisioning examples
+ - A PF/VF with NIX-LF & NPA-LF resources works as a pure network device
+ - A PF/VF with CPT-LF resource works as a pure crypto offload device.
+
+RVU functional blocks are highly configurable as per software requirements.
+
+Firmware setups following stuff before kernel boots
+ - Enables required number of RVU PFs based on number of physical links.
+ - Number of VFs per PF are either static or configurable at compile time.
+   Based on config, firmware assigns VFs to each of the PFs.
+ - Also assigns MSIX vectors to each of PF and VFs.
+ - These are not changed after kernel boot.
+
+Drivers
+=======
+
+Linux kernel will have multiple drivers registering to different PF and VFs
+of RVU. Wrt networking there will be 3 flavours of drivers.
+
+Admin Function driver
+---------------------
+
+As mentioned above RVU PF0 is called the admin function (AF), this driver
+supports resource provisioning and configuration of functional blocks.
+Doesn't handle any I/O. It sets up few basic stuff but most of the
+funcionality is achieved via configuration requests from PFs and VFs.
+
+PF/VFs communicates with AF via a shared memory region (mailbox). Upon
+receiving requests AF does resource provisioning and other HW configuration.
+AF is always attached to host kernel, but PFs and their VFs may be used by host
+kernel itself, or attached to VMs or to userspace applications like
+DPDK etc. So AF has to handle provisioning/configuration requests sent
+by any device from any domain.
+
+AF driver also interacts with underlying firmware to
+ - Manage physical ethernet links ie CGX LMACs.
+ - Retrieve information like speed, duplex, autoneg etc
+ - Retrieve PHY EEPROM and stats.
+ - Configure FEC, PAM modes
+ - etc
+
+From pure networking side AF driver supports following functionality.
+ - Map a physical link to a RVU PF to which a netdev is registered.
+ - Attach NIX and NPA block LFs to RVU PF/VF which provide buffer pools, RQs, SQs
+   for regular networking functionality.
+ - Flow control (pause frames) enable/disable/config.
+ - HW PTP timestamping related config.
+ - NPC parser profile config, basically how to parse pkt and what info to extract.
+ - NPC extract profile config, what to extract from the pkt to match data in MCAM entries.
+ - Manage NPC MCAM entries, upon request can frame and install requested packet forwarding rules.
+ - Defines receive side scaling (RSS) algorithms.
+ - Defines segmentation offload algorithms (eg TSO)
+ - VLAN stripping, capture and insertion config.
+ - SSO and TIM blocks config which provide packet scheduling support.
+ - Debugfs support, to check current resource provising, current status of
+   NPA pools, NIX RQ, SQ and CQs, various stats etc which helps in debugging issues.
+ - And many more.
+
+Physical Function driver
+------------------------
+
+This RVU PF handles IO, is mapped to a physical ethernet link and this
+driver registers a netdev. This supports SR-IOV. As said above this driver
+communicates with AF with a mailbox. To retrieve information from physical
+links this driver talks to AF and AF gets that info from firmware and responds
+back ie cannot talk to firmware directly.
+
+Supports ethtool for configuring links, RSS, queue count, queue size,
+flow control, ntuple filters, dump PHY EEPROM, config FEC etc.
+
+Virtual Function driver
+-----------------------
+
+There are two types VFs, VFs that share the physical link with their parent
+SR-IOV PF and the VFs which work in pairs using internal HW loopback channels (LBK).
+
+Type1:
+ - These VFs and their parent PF share a physical link and used for outside communication.
+ - VFs cannot communicate with AF directly, they send mbox message to PF and PF
+   forwards that to AF. AF after processing, responds back to PF and PF forwards
+   the reply to VF.
+ - From functionality point of view there is no difference between PF and VF as same type
+   HW resources are attached to both. But user would be able to configure few stuff only
+   from PF as PF is treated as owner/admin of the link.
+
+Type2:
+ - RVU PF0 ie admin function creates these VFs and maps them to loopback block's channels.
+ - A set of two VFs (VF0 & VF1, VF2 & VF3 .. so on) works as a pair ie pkts sent out of
+   VF0 will be received by VF1 and viceversa.
+ - These VFs can be used by applications or virtual machines to communicate between them
+   without sending traffic outside. There is no switch present in HW, hence the support
+   for loopback VFs.
+ - These communicate directly with AF (PF0) via mbox.
+
+Except for the IO channels or links used for packet reception and transmission there is
+no other difference between these VF types. AF driver takes care of IO channel mapping,
+hence same VF driver works for both types of devices.
+
+Basic packet flow
+=================
+
+Ingress
+-------
+
+1. CGX LMAC receives packet.
+2. Forwards the packet to the NIX block.
+3. Then submitted to NPC block for parsing and then MCAM lookup to get the destination RVU device.
+4. NIX LF attached to the destination RVU device allocates a buffer from RQ mapped buffer pool of NPA block LF.
+5. RQ may be selected by RSS or by configuring MCAM rule with a RQ number.
+6. Packet is DMA'ed and driver is notified.
+
+Egress
+------
+
+1. Driver prepares a send descriptor and submits to SQ for transmission.
+2. The SQ is already configured (by AF) to transmit on a specific link/channel.
+3. The SQ descriptor ring is maintained in buffers allocated from SQ mapped pool of NPA block LF.
+4. NIX block transmits the pkt on the designated channel.
+5. NPC MCAM entries can be installed to divert pkt onto a different channel.
diff --git a/MAINTAINERS b/MAINTAINERS
index 31084bdb1639..681ba1db3cf2 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -10000,6 +10000,7 @@ M:	Jerin Jacob <jerinj@marvell.com>
 L:	netdev@vger.kernel.org
 S:	Supported
 F:	drivers/net/ethernet/marvell/octeontx2/af/
+F:	Documentation/networking/device_drivers/marvell/octeontx2.rst
 
 MATROX FRAMEBUFFER DRIVER
 L:	linux-fbdev@vger.kernel.org
-- 
cgit v1.2.3-59-g8ed1b


From 688b3e829d60a290c9a0a2c8b0b99f816ed37a8c Mon Sep 17 00:00:00 2001
From: Sunil Goutham <sgoutham@marvell.com>
Date: Mon, 27 Jan 2020 18:35:31 +0530
Subject: MAINTAINERS: Add entry for Marvell OcteonTX2 Physical Function driver

Added maintainers entry for Marvell OcteonTX2 SOC's physical
function NIC driver.

Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 MAINTAINERS | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index 681ba1db3cf2..da8a77384dd4 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -10002,6 +10002,15 @@ S:	Supported
 F:	drivers/net/ethernet/marvell/octeontx2/af/
 F:	Documentation/networking/device_drivers/marvell/octeontx2.rst
 
+MARVELL OCTEONTX2 PHYSICAL FUNCTION DRIVER
+M:	Sunil Goutham <sgoutham@marvell.com>
+M:	Geetha sowjanya <gakula@marvell.com>
+M:	Subbaraya Sundeep <sbhatta@marvell.com>
+M:	hariprasad <hkelam@marvell.com>
+L:	netdev@vger.kernel.org
+S:	Supported
+F:	drivers/net/ethernet/marvell/octeontx2/nic/
+
 MATROX FRAMEBUFFER DRIVER
 L:	linux-fbdev@vger.kernel.org
 S:	Orphan
-- 
cgit v1.2.3-59-g8ed1b


From 2924e0699963b839f88f8c4e855929ea49185870 Mon Sep 17 00:00:00 2001
From: Michal Kalderon <michal.kalderon@marvell.com>
Date: Mon, 27 Jan 2020 15:26:07 +0200
Subject: qed: FW 8.42.2.0 Internal ram offsets modifications

IRO stands for internal RAM offsets. Updating the FW binary produces
different iro offsets. This file contains the different values,
and a new representation of the values.
Update the FW version

Signed-off-by: Ariel Elior <ariel.elior@marvell.com>
Signed-off-by: Michal Kalderon <michal.kalderon@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/qlogic/qed/qed.h     |   4 +-
 drivers/net/ethernet/qlogic/qed/qed_hsi.h | 419 ++++++++++++++++--------------
 include/linux/qed/common_hsi.h            |   4 +-
 3 files changed, 234 insertions(+), 193 deletions(-)

diff --git a/drivers/net/ethernet/qlogic/qed/qed.h b/drivers/net/ethernet/qlogic/qed/qed.h
index 89fe091c958d..8ec46bf409c2 100644
--- a/drivers/net/ethernet/qlogic/qed/qed.h
+++ b/drivers/net/ethernet/qlogic/qed/qed.h
@@ -796,8 +796,8 @@ struct qed_dev {
 	u8				cache_shift;
 
 	/* Init */
-	const struct iro		*iro_arr;
-#define IRO (p_hwfn->cdev->iro_arr)
+	const u32 *iro_arr;
+#define IRO ((const struct iro *)p_hwfn->cdev->iro_arr)
 
 	/* HW functions */
 	u8				num_hwfns;
diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
index cf3ceb62e397..aaff7117291c 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
@@ -4282,325 +4282,366 @@ void qed_set_rdma_error_level(struct qed_hwfn *p_hwfn,
 	(IRO[7].base + ((queue_zone_id) * IRO[7].m1))
 #define USTORM_COMMON_QUEUE_CONS_SIZE			(IRO[7].size)
 
+/* Xstorm common PQ info */
+#define XSTORM_PQ_INFO_OFFSET(pq_id) \
+	(IRO[8].base + ((pq_id) * IRO[8].m1))
+#define XSTORM_PQ_INFO_SIZE				(IRO[8].size)
+
 /* Xstorm Integration Test Data */
-#define XSTORM_INTEG_TEST_DATA_OFFSET			(IRO[8].base)
-#define XSTORM_INTEG_TEST_DATA_SIZE			(IRO[8].size)
+#define XSTORM_INTEG_TEST_DATA_OFFSET			(IRO[9].base)
+#define XSTORM_INTEG_TEST_DATA_SIZE			(IRO[9].size)
 
 /* Ystorm Integration Test Data */
-#define YSTORM_INTEG_TEST_DATA_OFFSET			(IRO[9].base)
-#define YSTORM_INTEG_TEST_DATA_SIZE			(IRO[9].size)
+#define YSTORM_INTEG_TEST_DATA_OFFSET			(IRO[10].base)
+#define YSTORM_INTEG_TEST_DATA_SIZE			(IRO[10].size)
 
 /* Pstorm Integration Test Data */
-#define PSTORM_INTEG_TEST_DATA_OFFSET			(IRO[10].base)
-#define PSTORM_INTEG_TEST_DATA_SIZE			(IRO[10].size)
+#define PSTORM_INTEG_TEST_DATA_OFFSET			(IRO[11].base)
+#define PSTORM_INTEG_TEST_DATA_SIZE			(IRO[11].size)
 
 /* Tstorm Integration Test Data */
-#define TSTORM_INTEG_TEST_DATA_OFFSET			(IRO[11].base)
-#define TSTORM_INTEG_TEST_DATA_SIZE			(IRO[11].size)
+#define TSTORM_INTEG_TEST_DATA_OFFSET			(IRO[12].base)
+#define TSTORM_INTEG_TEST_DATA_SIZE			(IRO[12].size)
 
 /* Mstorm Integration Test Data */
-#define MSTORM_INTEG_TEST_DATA_OFFSET			(IRO[12].base)
-#define MSTORM_INTEG_TEST_DATA_SIZE			(IRO[12].size)
+#define MSTORM_INTEG_TEST_DATA_OFFSET			(IRO[13].base)
+#define MSTORM_INTEG_TEST_DATA_SIZE			(IRO[13].size)
 
 /* Ustorm Integration Test Data */
-#define USTORM_INTEG_TEST_DATA_OFFSET			(IRO[13].base)
-#define USTORM_INTEG_TEST_DATA_SIZE			(IRO[13].size)
+#define USTORM_INTEG_TEST_DATA_OFFSET			(IRO[14].base)
+#define USTORM_INTEG_TEST_DATA_SIZE			(IRO[14].size)
+
+/* Xstorm overlay buffer host address */
+#define XSTORM_OVERLAY_BUF_ADDR_OFFSET			(IRO[15].base)
+#define XSTORM_OVERLAY_BUF_ADDR_SIZE			(IRO[15].size)
+
+/* Ystorm overlay buffer host address */
+#define YSTORM_OVERLAY_BUF_ADDR_OFFSET			(IRO[16].base)
+#define YSTORM_OVERLAY_BUF_ADDR_SIZE			(IRO[16].size)
+
+/* Pstorm overlay buffer host address */
+#define PSTORM_OVERLAY_BUF_ADDR_OFFSET			(IRO[17].base)
+#define PSTORM_OVERLAY_BUF_ADDR_SIZE			(IRO[17].size)
+
+/* Tstorm overlay buffer host address */
+#define TSTORM_OVERLAY_BUF_ADDR_OFFSET			(IRO[18].base)
+#define TSTORM_OVERLAY_BUF_ADDR_SIZE			(IRO[18].size)
+
+/* Mstorm overlay buffer host address */
+#define MSTORM_OVERLAY_BUF_ADDR_OFFSET			(IRO[19].base)
+#define MSTORM_OVERLAY_BUF_ADDR_SIZE			(IRO[19].size)
+
+/* Ustorm overlay buffer host address */
+#define USTORM_OVERLAY_BUF_ADDR_OFFSET			(IRO[20].base)
+#define USTORM_OVERLAY_BUF_ADDR_SIZE			(IRO[20].size)
 
 /* Tstorm producers */
 #define TSTORM_LL2_RX_PRODS_OFFSET(core_rx_queue_id) \
-	(IRO[14].base + ((core_rx_queue_id) * IRO[14].m1))
-#define TSTORM_LL2_RX_PRODS_SIZE			(IRO[14].size)
+	(IRO[21].base + ((core_rx_queue_id) * IRO[21].m1))
+#define TSTORM_LL2_RX_PRODS_SIZE			(IRO[21].size)
 
 /* Tstorm LightL2 queue statistics */
 #define CORE_LL2_TSTORM_PER_QUEUE_STAT_OFFSET(core_rx_queue_id) \
-	(IRO[15].base + ((core_rx_queue_id) * IRO[15].m1))
-#define CORE_LL2_TSTORM_PER_QUEUE_STAT_SIZE		(IRO[15].size)
+	(IRO[22].base + ((core_rx_queue_id) * IRO[22].m1))
+#define CORE_LL2_TSTORM_PER_QUEUE_STAT_SIZE		(IRO[22].size)
 
 /* Ustorm LiteL2 queue statistics */
 #define CORE_LL2_USTORM_PER_QUEUE_STAT_OFFSET(core_rx_queue_id) \
-	(IRO[16].base + ((core_rx_queue_id) * IRO[16].m1))
-#define CORE_LL2_USTORM_PER_QUEUE_STAT_SIZE		(IRO[16].size)
+	(IRO[23].base + ((core_rx_queue_id) * IRO[23].m1))
+#define CORE_LL2_USTORM_PER_QUEUE_STAT_SIZE		(IRO[23].size)
 
 /* Pstorm LiteL2 queue statistics */
 #define CORE_LL2_PSTORM_PER_QUEUE_STAT_OFFSET(core_tx_stats_id) \
-	(IRO[17].base + ((core_tx_stats_id) * IRO[17].m1))
-#define CORE_LL2_PSTORM_PER_QUEUE_STAT_SIZE		(IRO[17].size)
+	(IRO[24].base + ((core_tx_stats_id) * IRO[24].m1))
+#define CORE_LL2_PSTORM_PER_QUEUE_STAT_SIZE		(IRO[24].size)
 
 /* Mstorm queue statistics */
 #define MSTORM_QUEUE_STAT_OFFSET(stat_counter_id) \
-	(IRO[18].base + ((stat_counter_id) * IRO[18].m1))
-#define MSTORM_QUEUE_STAT_SIZE				(IRO[18].size)
+	(IRO[25].base + ((stat_counter_id) * IRO[25].m1))
+#define MSTORM_QUEUE_STAT_SIZE				(IRO[25].size)
 
-/* Mstorm ETH PF queues producers */
-#define MSTORM_ETH_PF_PRODS_OFFSET(queue_id) \
-	(IRO[19].base + ((queue_id) * IRO[19].m1))
-#define MSTORM_ETH_PF_PRODS_SIZE			(IRO[19].size)
+/* TPA agregation timeout in us resolution (on ASIC) */
+#define MSTORM_TPA_TIMEOUT_US_OFFSET			(IRO[26].base)
+#define MSTORM_TPA_TIMEOUT_US_SIZE			(IRO[26].size)
 
 /* Mstorm ETH VF queues producers offset in RAM. Used in default VF zone size
- * mode.
+ * mode
  */
 #define MSTORM_ETH_VF_PRODS_OFFSET(vf_id, vf_queue_id) \
-	(IRO[20].base + ((vf_id) * IRO[20].m1) + ((vf_queue_id) * IRO[20].m2))
-#define MSTORM_ETH_VF_PRODS_SIZE			(IRO[20].size)
+	(IRO[27].base + ((vf_id) * IRO[27].m1) + ((vf_queue_id) * IRO[27].m2))
+#define MSTORM_ETH_VF_PRODS_SIZE			(IRO[27].size)
 
-/* TPA agregation timeout in us resolution (on ASIC) */
-#define MSTORM_TPA_TIMEOUT_US_OFFSET			(IRO[21].base)
-#define MSTORM_TPA_TIMEOUT_US_SIZE			(IRO[21].size)
+/* Mstorm ETH PF queues producers */
+#define MSTORM_ETH_PF_PRODS_OFFSET(queue_id) \
+	(IRO[28].base + ((queue_id) * IRO[28].m1))
+#define MSTORM_ETH_PF_PRODS_SIZE			(IRO[28].size)
 
 /* Mstorm pf statistics */
 #define MSTORM_ETH_PF_STAT_OFFSET(pf_id) \
-	(IRO[22].base + ((pf_id) * IRO[22].m1))
-#define MSTORM_ETH_PF_STAT_SIZE				(IRO[22].size)
+	(IRO[29].base + ((pf_id) * IRO[29].m1))
+#define MSTORM_ETH_PF_STAT_SIZE				(IRO[29].size)
 
 /* Ustorm queue statistics */
 #define USTORM_QUEUE_STAT_OFFSET(stat_counter_id) \
-	(IRO[23].base + ((stat_counter_id) * IRO[23].m1))
-#define USTORM_QUEUE_STAT_SIZE				(IRO[23].size)
+	(IRO[30].base + ((stat_counter_id) * IRO[30].m1))
+#define USTORM_QUEUE_STAT_SIZE				(IRO[30].size)
 
 /* Ustorm pf statistics */
-#define USTORM_ETH_PF_STAT_OFFSET(pf_id)\
-	(IRO[24].base + ((pf_id) * IRO[24].m1))
-#define USTORM_ETH_PF_STAT_SIZE				(IRO[24].size)
+#define USTORM_ETH_PF_STAT_OFFSET(pf_id) \
+	(IRO[31].base + ((pf_id) * IRO[31].m1))
+#define USTORM_ETH_PF_STAT_SIZE				(IRO[31].size)
 
 /* Pstorm queue statistics */
-#define PSTORM_QUEUE_STAT_OFFSET(stat_counter_id) \
-	(IRO[25].base + ((stat_counter_id) * IRO[25].m1))
-#define PSTORM_QUEUE_STAT_SIZE				(IRO[25].size)
+#define PSTORM_QUEUE_STAT_OFFSET(stat_counter_id)	\
+	(IRO[32].base + ((stat_counter_id) * IRO[32].m1))
+#define PSTORM_QUEUE_STAT_SIZE				(IRO[32].size)
 
 /* Pstorm pf statistics */
 #define PSTORM_ETH_PF_STAT_OFFSET(pf_id) \
-	(IRO[26].base + ((pf_id) * IRO[26].m1))
-#define PSTORM_ETH_PF_STAT_SIZE				(IRO[26].size)
+	(IRO[33].base + ((pf_id) * IRO[33].m1))
+#define PSTORM_ETH_PF_STAT_SIZE				(IRO[33].size)
 
 /* Control frame's EthType configuration for TX control frame security */
-#define PSTORM_CTL_FRAME_ETHTYPE_OFFSET(eth_type_id) \
-	(IRO[27].base + ((eth_type_id) * IRO[27].m1))
-#define PSTORM_CTL_FRAME_ETHTYPE_SIZE			(IRO[27].size)
+#define PSTORM_CTL_FRAME_ETHTYPE_OFFSET(eth_type_id)	\
+	(IRO[34].base + ((eth_type_id) * IRO[34].m1))
+#define PSTORM_CTL_FRAME_ETHTYPE_SIZE			(IRO[34].size)
 
 /* Tstorm last parser message */
-#define TSTORM_ETH_PRS_INPUT_OFFSET			(IRO[28].base)
-#define TSTORM_ETH_PRS_INPUT_SIZE			(IRO[28].size)
+#define TSTORM_ETH_PRS_INPUT_OFFSET			(IRO[35].base)
+#define TSTORM_ETH_PRS_INPUT_SIZE			(IRO[35].size)
 
 /* Tstorm Eth limit Rx rate */
-#define ETH_RX_RATE_LIMIT_OFFSET(pf_id) \
-	(IRO[29].base + ((pf_id) * IRO[29].m1))
-#define ETH_RX_RATE_LIMIT_SIZE				(IRO[29].size)
+#define ETH_RX_RATE_LIMIT_OFFSET(pf_id)	\
+	(IRO[36].base + ((pf_id) * IRO[36].m1))
+#define ETH_RX_RATE_LIMIT_SIZE				(IRO[36].size)
 
 /* RSS indirection table entry update command per PF offset in TSTORM PF BAR0.
- * Use eth_tstorm_rss_update_data for update.
+ * Use eth_tstorm_rss_update_data for update
  */
 #define TSTORM_ETH_RSS_UPDATE_OFFSET(pf_id) \
-	(IRO[30].base + ((pf_id) * IRO[30].m1))
-#define TSTORM_ETH_RSS_UPDATE_SIZE			(IRO[30].size)
+	(IRO[37].base + ((pf_id) * IRO[37].m1))
+#define TSTORM_ETH_RSS_UPDATE_SIZE			(IRO[37].size)
 
 /* Xstorm queue zone */
 #define XSTORM_ETH_QUEUE_ZONE_OFFSET(queue_id) \
-	(IRO[31].base + ((queue_id) * IRO[31].m1))
-#define XSTORM_ETH_QUEUE_ZONE_SIZE			(IRO[31].size)
+	(IRO[38].base + ((queue_id) * IRO[38].m1))
+#define XSTORM_ETH_QUEUE_ZONE_SIZE			(IRO[38].size)
 
 /* Ystorm cqe producer */
 #define YSTORM_TOE_CQ_PROD_OFFSET(rss_id) \
-	(IRO[32].base + ((rss_id) * IRO[32].m1))
-#define YSTORM_TOE_CQ_PROD_SIZE				(IRO[32].size)
+	(IRO[39].base + ((rss_id) * IRO[39].m1))
+#define YSTORM_TOE_CQ_PROD_SIZE				(IRO[39].size)
 
 /* Ustorm cqe producer */
 #define USTORM_TOE_CQ_PROD_OFFSET(rss_id) \
-	(IRO[33].base + ((rss_id) * IRO[33].m1))
-#define USTORM_TOE_CQ_PROD_SIZE				(IRO[33].size)
+	(IRO[40].base + ((rss_id) * IRO[40].m1))
+#define USTORM_TOE_CQ_PROD_SIZE				(IRO[40].size)
 
 /* Ustorm grq producer */
 #define USTORM_TOE_GRQ_PROD_OFFSET(pf_id) \
-	(IRO[34].base + ((pf_id) * IRO[34].m1))
-#define USTORM_TOE_GRQ_PROD_SIZE			(IRO[34].size)
+	(IRO[41].base + ((pf_id) * IRO[41].m1))
+#define USTORM_TOE_GRQ_PROD_SIZE			(IRO[41].size)
 
 /* Tstorm cmdq-cons of given command queue-id */
 #define TSTORM_SCSI_CMDQ_CONS_OFFSET(cmdq_queue_id) \
-	(IRO[35].base + ((cmdq_queue_id) * IRO[35].m1))
-#define TSTORM_SCSI_CMDQ_CONS_SIZE			(IRO[35].size)
+	(IRO[42].base + ((cmdq_queue_id) * IRO[42].m1))
+#define TSTORM_SCSI_CMDQ_CONS_SIZE			(IRO[42].size)
 
 /* Tstorm (reflects M-Storm) bdq-external-producer of given function ID,
- * BDqueue-id.
+ * BDqueue-id
  */
-#define TSTORM_SCSI_BDQ_EXT_PROD_OFFSET(func_id, bdq_id) \
-	(IRO[36].base + ((func_id) * IRO[36].m1) + ((bdq_id) * IRO[36].m2))
-#define TSTORM_SCSI_BDQ_EXT_PROD_SIZE			(IRO[36].size)
+#define TSTORM_SCSI_BDQ_EXT_PROD_OFFSET(storage_func_id, bdq_id) \
+	(IRO[43].base + ((storage_func_id) * IRO[43].m1) + \
+	 ((bdq_id) * IRO[43].m2))
+#define TSTORM_SCSI_BDQ_EXT_PROD_SIZE			(IRO[43].size)
 
 /* Mstorm bdq-external-producer of given BDQ resource ID, BDqueue-id */
-#define MSTORM_SCSI_BDQ_EXT_PROD_OFFSET(func_id, bdq_id) \
-	(IRO[37].base + ((func_id) * IRO[37].m1) + ((bdq_id) * IRO[37].m2))
-#define MSTORM_SCSI_BDQ_EXT_PROD_SIZE			(IRO[37].size)
+#define MSTORM_SCSI_BDQ_EXT_PROD_OFFSET(storage_func_id, bdq_id) \
+	(IRO[44].base + ((storage_func_id) * IRO[44].m1) + \
+	 ((bdq_id) * IRO[44].m2))
+#define MSTORM_SCSI_BDQ_EXT_PROD_SIZE			(IRO[44].size)
 
 /* Tstorm iSCSI RX stats */
-#define TSTORM_ISCSI_RX_STATS_OFFSET(pf_id) \
-	(IRO[38].base + ((pf_id) * IRO[38].m1))
-#define TSTORM_ISCSI_RX_STATS_SIZE			(IRO[38].size)
+#define TSTORM_ISCSI_RX_STATS_OFFSET(storage_func_id) \
+	(IRO[45].base + ((storage_func_id) * IRO[45].m1))
+#define TSTORM_ISCSI_RX_STATS_SIZE			(IRO[45].size)
 
 /* Mstorm iSCSI RX stats */
-#define MSTORM_ISCSI_RX_STATS_OFFSET(pf_id) \
-	(IRO[39].base + ((pf_id) * IRO[39].m1))
-#define MSTORM_ISCSI_RX_STATS_SIZE			(IRO[39].size)
+#define MSTORM_ISCSI_RX_STATS_OFFSET(storage_func_id) \
+	(IRO[46].base + ((storage_func_id) * IRO[46].m1))
+#define MSTORM_ISCSI_RX_STATS_SIZE			(IRO[46].size)
 
 /* Ustorm iSCSI RX stats */
-#define USTORM_ISCSI_RX_STATS_OFFSET(pf_id) \
-	(IRO[40].base + ((pf_id) * IRO[40].m1))
-#define USTORM_ISCSI_RX_STATS_SIZE			(IRO[40].size)
+#define USTORM_ISCSI_RX_STATS_OFFSET(storage_func_id) \
+	(IRO[47].base + ((storage_func_id) * IRO[47].m1))
+#define USTORM_ISCSI_RX_STATS_SIZE			(IRO[47].size)
 
 /* Xstorm iSCSI TX stats */
-#define XSTORM_ISCSI_TX_STATS_OFFSET(pf_id) \
-	(IRO[41].base + ((pf_id) * IRO[41].m1))
-#define XSTORM_ISCSI_TX_STATS_SIZE			(IRO[41].size)
+#define XSTORM_ISCSI_TX_STATS_OFFSET(storage_func_id) \
+	(IRO[48].base + ((storage_func_id) * IRO[48].m1))
+#define XSTORM_ISCSI_TX_STATS_SIZE			(IRO[48].size)
 
 /* Ystorm iSCSI TX stats */
-#define YSTORM_ISCSI_TX_STATS_OFFSET(pf_id) \
-	(IRO[42].base + ((pf_id) * IRO[42].m1))
-#define YSTORM_ISCSI_TX_STATS_SIZE			(IRO[42].size)
+#define YSTORM_ISCSI_TX_STATS_OFFSET(storage_func_id) \
+	(IRO[49].base + ((storage_func_id) * IRO[49].m1))
+#define YSTORM_ISCSI_TX_STATS_SIZE			(IRO[49].size)
 
 /* Pstorm iSCSI TX stats */
-#define PSTORM_ISCSI_TX_STATS_OFFSET(pf_id) \
-	(IRO[43].base + ((pf_id) * IRO[43].m1))
-#define PSTORM_ISCSI_TX_STATS_SIZE			(IRO[43].size)
+#define PSTORM_ISCSI_TX_STATS_OFFSET(storage_func_id) \
+	(IRO[50].base + ((storage_func_id) * IRO[50].m1))
+#define PSTORM_ISCSI_TX_STATS_SIZE			(IRO[50].size)
 
 /* Tstorm FCoE RX stats */
 #define TSTORM_FCOE_RX_STATS_OFFSET(pf_id) \
-	(IRO[44].base + ((pf_id) * IRO[44].m1))
-#define TSTORM_FCOE_RX_STATS_SIZE			(IRO[44].size)
+	(IRO[51].base + ((pf_id) * IRO[51].m1))
+#define TSTORM_FCOE_RX_STATS_SIZE			(IRO[51].size)
 
 /* Pstorm FCoE TX stats */
 #define PSTORM_FCOE_TX_STATS_OFFSET(pf_id) \
-	(IRO[45].base + ((pf_id) * IRO[45].m1))
-#define PSTORM_FCOE_TX_STATS_SIZE			(IRO[45].size)
+	(IRO[52].base + ((pf_id) * IRO[52].m1))
+#define PSTORM_FCOE_TX_STATS_SIZE			(IRO[52].size)
 
 /* Pstorm RDMA queue statistics */
 #define PSTORM_RDMA_QUEUE_STAT_OFFSET(rdma_stat_counter_id) \
-	(IRO[46].base + ((rdma_stat_counter_id) * IRO[46].m1))
-#define PSTORM_RDMA_QUEUE_STAT_SIZE			(IRO[46].size)
+	(IRO[53].base + ((rdma_stat_counter_id) * IRO[53].m1))
+#define PSTORM_RDMA_QUEUE_STAT_SIZE			(IRO[53].size)
 
 /* Tstorm RDMA queue statistics */
 #define TSTORM_RDMA_QUEUE_STAT_OFFSET(rdma_stat_counter_id) \
-	(IRO[47].base + ((rdma_stat_counter_id) * IRO[47].m1))
-#define TSTORM_RDMA_QUEUE_STAT_SIZE			(IRO[47].size)
+	(IRO[54].base + ((rdma_stat_counter_id) * IRO[54].m1))
+#define TSTORM_RDMA_QUEUE_STAT_SIZE			(IRO[54].size)
 
 /* Xstorm error level for assert */
 #define XSTORM_RDMA_ASSERT_LEVEL_OFFSET(pf_id) \
-	(IRO[48].base +	((pf_id) * IRO[48].m1))
-#define XSTORM_RDMA_ASSERT_LEVEL_SIZE			(IRO[48].size)
+	(IRO[55].base + ((pf_id) * IRO[55].m1))
+#define XSTORM_RDMA_ASSERT_LEVEL_SIZE			(IRO[55].size)
 
 /* Ystorm error level for assert */
 #define YSTORM_RDMA_ASSERT_LEVEL_OFFSET(pf_id) \
-	(IRO[49].base + ((pf_id) * IRO[49].m1))
-#define YSTORM_RDMA_ASSERT_LEVEL_SIZE			(IRO[49].size)
+	(IRO[56].base + ((pf_id) * IRO[56].m1))
+#define YSTORM_RDMA_ASSERT_LEVEL_SIZE			(IRO[56].size)
 
 /* Pstorm error level for assert */
 #define PSTORM_RDMA_ASSERT_LEVEL_OFFSET(pf_id) \
-	(IRO[50].base +	((pf_id) * IRO[50].m1))
-#define PSTORM_RDMA_ASSERT_LEVEL_SIZE			(IRO[50].size)
+	(IRO[57].base + ((pf_id) * IRO[57].m1))
+#define PSTORM_RDMA_ASSERT_LEVEL_SIZE			(IRO[57].size)
 
 /* Tstorm error level for assert */
 #define TSTORM_RDMA_ASSERT_LEVEL_OFFSET(pf_id) \
-	(IRO[51].base +	((pf_id) * IRO[51].m1))
-#define TSTORM_RDMA_ASSERT_LEVEL_SIZE			(IRO[51].size)
+	(IRO[58].base + ((pf_id) * IRO[58].m1))
+#define TSTORM_RDMA_ASSERT_LEVEL_SIZE			(IRO[58].size)
 
 /* Mstorm error level for assert */
 #define MSTORM_RDMA_ASSERT_LEVEL_OFFSET(pf_id) \
-	(IRO[52].base + ((pf_id) * IRO[52].m1))
-#define MSTORM_RDMA_ASSERT_LEVEL_SIZE			(IRO[52].size)
+	(IRO[59].base + ((pf_id) * IRO[59].m1))
+#define MSTORM_RDMA_ASSERT_LEVEL_SIZE			(IRO[59].size)
 
 /* Ustorm error level for assert */
 #define USTORM_RDMA_ASSERT_LEVEL_OFFSET(pf_id) \
-	(IRO[53].base + ((pf_id) * IRO[53].m1))
-#define USTORM_RDMA_ASSERT_LEVEL_SIZE			(IRO[53].size)
+	(IRO[60].base + ((pf_id) * IRO[60].m1))
+#define USTORM_RDMA_ASSERT_LEVEL_SIZE			(IRO[60].size)
 
 /* Xstorm iWARP rxmit stats */
 #define XSTORM_IWARP_RXMIT_STATS_OFFSET(pf_id) \
-	(IRO[54].base +	((pf_id) * IRO[54].m1))
-#define XSTORM_IWARP_RXMIT_STATS_SIZE			(IRO[54].size)
+	(IRO[61].base + ((pf_id) * IRO[61].m1))
+#define XSTORM_IWARP_RXMIT_STATS_SIZE			(IRO[61].size)
 
 /* Tstorm RoCE Event Statistics */
-#define TSTORM_ROCE_EVENTS_STAT_OFFSET(roce_pf_id) \
-	(IRO[55].base + ((roce_pf_id) * IRO[55].m1))
-#define TSTORM_ROCE_EVENTS_STAT_SIZE			(IRO[55].size)
+#define TSTORM_ROCE_EVENTS_STAT_OFFSET(roce_pf_id)	\
+	(IRO[62].base + ((roce_pf_id) * IRO[62].m1))
+#define TSTORM_ROCE_EVENTS_STAT_SIZE			(IRO[62].size)
 
 /* DCQCN Received Statistics */
-#define YSTORM_ROCE_DCQCN_RECEIVED_STATS_OFFSET(roce_pf_id) \
-	(IRO[56].base + ((roce_pf_id) * IRO[56].m1))
-#define YSTORM_ROCE_DCQCN_RECEIVED_STATS_SIZE		(IRO[56].size)
+#define YSTORM_ROCE_DCQCN_RECEIVED_STATS_OFFSET(roce_pf_id)\
+	(IRO[63].base + ((roce_pf_id) * IRO[63].m1))
+#define YSTORM_ROCE_DCQCN_RECEIVED_STATS_SIZE		(IRO[63].size)
 
 /* RoCE Error Statistics */
-#define YSTORM_ROCE_ERROR_STATS_OFFSET(roce_pf_id) \
-	(IRO[57].base + ((roce_pf_id) * IRO[57].m1))
-#define YSTORM_ROCE_ERROR_STATS_SIZE			(IRO[57].size)
+#define YSTORM_ROCE_ERROR_STATS_OFFSET(roce_pf_id)	\
+	(IRO[64].base + ((roce_pf_id) * IRO[64].m1))
+#define YSTORM_ROCE_ERROR_STATS_SIZE			(IRO[64].size)
 
 /* DCQCN Sent Statistics */
-#define PSTORM_ROCE_DCQCN_SENT_STATS_OFFSET(roce_pf_id) \
-	(IRO[58].base + ((roce_pf_id) * IRO[58].m1))
-#define PSTORM_ROCE_DCQCN_SENT_STATS_SIZE		(IRO[58].size)
+#define PSTORM_ROCE_DCQCN_SENT_STATS_OFFSET(roce_pf_id)	\
+	(IRO[65].base + ((roce_pf_id) * IRO[65].m1))
+#define PSTORM_ROCE_DCQCN_SENT_STATS_SIZE		(IRO[65].size)
 
 /* RoCE CQEs Statistics */
-#define USTORM_ROCE_CQE_STATS_OFFSET(roce_pf_id) \
-	(IRO[59].base + ((roce_pf_id) * IRO[59].m1))
-#define USTORM_ROCE_CQE_STATS_SIZE			(IRO[59].size)
-
-static const struct iro iro_arr[60] = {
-	{0x0, 0x0, 0x0, 0x0, 0x8},
-	{0x4cb8, 0x88, 0x0, 0x0, 0x88},
-	{0x6530, 0x20, 0x0, 0x0, 0x20},
-	{0xb00, 0x8, 0x0, 0x0, 0x4},
-	{0xa80, 0x8, 0x0, 0x0, 0x4},
-	{0x0, 0x8, 0x0, 0x0, 0x2},
-	{0x80, 0x8, 0x0, 0x0, 0x4},
-	{0x84, 0x8, 0x0, 0x0, 0x2},
-	{0x4c48, 0x0, 0x0, 0x0, 0x78},
-	{0x3e38, 0x0, 0x0, 0x0, 0x78},
-	{0x3ef8, 0x0, 0x0, 0x0, 0x78},
-	{0x4c40, 0x0, 0x0, 0x0, 0x78},
-	{0x4998, 0x0, 0x0, 0x0, 0x78},
-	{0x7f50, 0x0, 0x0, 0x0, 0x78},
-	{0xa28, 0x8, 0x0, 0x0, 0x8},
-	{0x6210, 0x10, 0x0, 0x0, 0x10},
-	{0xb820, 0x30, 0x0, 0x0, 0x30},
-	{0xa990, 0x30, 0x0, 0x0, 0x30},
-	{0x4b68, 0x80, 0x0, 0x0, 0x40},
-	{0x1f8, 0x4, 0x0, 0x0, 0x4},
-	{0x53a8, 0x80, 0x4, 0x0, 0x4},
-	{0xc7d0, 0x0, 0x0, 0x0, 0x4},
-	{0x4ba8, 0x80, 0x0, 0x0, 0x20},
-	{0x8158, 0x40, 0x0, 0x0, 0x30},
-	{0xe770, 0x60, 0x0, 0x0, 0x60},
-	{0x4090, 0x80, 0x0, 0x0, 0x38},
-	{0xfea8, 0x78, 0x0, 0x0, 0x78},
-	{0x1f8, 0x4, 0x0, 0x0, 0x4},
-	{0xaf20, 0x0, 0x0, 0x0, 0xf0},
-	{0xb010, 0x8, 0x0, 0x0, 0x8},
-	{0xc00, 0x8, 0x0, 0x0, 0x8},
-	{0x1f8, 0x8, 0x0, 0x0, 0x8},
-	{0xac0, 0x8, 0x0, 0x0, 0x8},
-	{0x2578, 0x8, 0x0, 0x0, 0x8},
-	{0x24f8, 0x8, 0x0, 0x0, 0x8},
-	{0x0, 0x8, 0x0, 0x0, 0x8},
-	{0x400, 0x18, 0x8, 0x0, 0x8},
-	{0xb78, 0x18, 0x8, 0x0, 0x2},
-	{0xd898, 0x50, 0x0, 0x0, 0x3c},
-	{0x12908, 0x18, 0x0, 0x0, 0x10},
-	{0x11aa8, 0x40, 0x0, 0x0, 0x18},
-	{0xa588, 0x50, 0x0, 0x0, 0x20},
-	{0x8f00, 0x40, 0x0, 0x0, 0x28},
-	{0x10e30, 0x18, 0x0, 0x0, 0x10},
-	{0xde48, 0x48, 0x0, 0x0, 0x38},
-	{0x11298, 0x20, 0x0, 0x0, 0x20},
-	{0x40c8, 0x80, 0x0, 0x0, 0x10},
-	{0x5048, 0x10, 0x0, 0x0, 0x10},
-	{0xc748, 0x8, 0x0, 0x0, 0x1},
-	{0xa928, 0x8, 0x0, 0x0, 0x1},
-	{0x11a30, 0x8, 0x0, 0x0, 0x1},
-	{0xf030, 0x8, 0x0, 0x0, 0x1},
-	{0x13028, 0x8, 0x0, 0x0, 0x1},
-	{0x12c58, 0x8, 0x0, 0x0, 0x1},
-	{0xc9b8, 0x30, 0x0, 0x0, 0x10},
-	{0xed90, 0x28, 0x0, 0x0, 0x28},
-	{0xad20, 0x18, 0x0, 0x0, 0x18},
-	{0xaea0, 0x8, 0x0, 0x0, 0x8},
-	{0x13c38, 0x8, 0x0, 0x0, 0x8},
-	{0x13c50, 0x18, 0x0, 0x0, 0x18},
+#define USTORM_ROCE_CQE_STATS_OFFSET(roce_pf_id)	\
+	(IRO[66].base + ((roce_pf_id) * IRO[66].m1))
+#define USTORM_ROCE_CQE_STATS_SIZE			(IRO[66].size)
+
+/* IRO Array */
+static const u32 iro_arr[] = {
+	0x00000000, 0x00000000, 0x00080000,
+	0x00003288, 0x00000088, 0x00880000,
+	0x000058e8, 0x00000020, 0x00200000,
+	0x00000b00, 0x00000008, 0x00040000,
+	0x00000a80, 0x00000008, 0x00040000,
+	0x00000000, 0x00000008, 0x00020000,
+	0x00000080, 0x00000008, 0x00040000,
+	0x00000084, 0x00000008, 0x00020000,
+	0x00005718, 0x00000004, 0x00040000,
+	0x00004dd0, 0x00000000, 0x00780000,
+	0x00003e40, 0x00000000, 0x00780000,
+	0x00004480, 0x00000000, 0x00780000,
+	0x00003210, 0x00000000, 0x00780000,
+	0x00003b50, 0x00000000, 0x00780000,
+	0x00007f58, 0x00000000, 0x00780000,
+	0x00005f58, 0x00000000, 0x00080000,
+	0x00007100, 0x00000000, 0x00080000,
+	0x0000aea0, 0x00000000, 0x00080000,
+	0x00004398, 0x00000000, 0x00080000,
+	0x0000a5a0, 0x00000000, 0x00080000,
+	0x0000bde8, 0x00000000, 0x00080000,
+	0x00000020, 0x00000004, 0x00040000,
+	0x000056c8, 0x00000010, 0x00100000,
+	0x0000c210, 0x00000030, 0x00300000,
+	0x0000b088, 0x00000038, 0x00380000,
+	0x00003d20, 0x00000080, 0x00400000,
+	0x0000bf60, 0x00000000, 0x00040000,
+	0x00004560, 0x00040080, 0x00040000,
+	0x000001f8, 0x00000004, 0x00040000,
+	0x00003d60, 0x00000080, 0x00200000,
+	0x00008960, 0x00000040, 0x00300000,
+	0x0000e840, 0x00000060, 0x00600000,
+	0x00004618, 0x00000080, 0x00380000,
+	0x00010738, 0x000000c0, 0x00c00000,
+	0x000001f8, 0x00000002, 0x00020000,
+	0x0000a2a0, 0x00000000, 0x01080000,
+	0x0000a3a8, 0x00000008, 0x00080000,
+	0x000001c0, 0x00000008, 0x00080000,
+	0x000001f8, 0x00000008, 0x00080000,
+	0x00000ac0, 0x00000008, 0x00080000,
+	0x00002578, 0x00000008, 0x00080000,
+	0x000024f8, 0x00000008, 0x00080000,
+	0x00000280, 0x00000008, 0x00080000,
+	0x00000680, 0x00080018, 0x00080000,
+	0x00000b78, 0x00080018, 0x00020000,
+	0x0000c640, 0x00000050, 0x003c0000,
+	0x00012038, 0x00000018, 0x00100000,
+	0x00011b00, 0x00000040, 0x00180000,
+	0x000095d0, 0x00000050, 0x00200000,
+	0x00008b10, 0x00000040, 0x00280000,
+	0x00011640, 0x00000018, 0x00100000,
+	0x0000c828, 0x00000048, 0x00380000,
+	0x00011710, 0x00000020, 0x00200000,
+	0x00004650, 0x00000080, 0x00100000,
+	0x00003618, 0x00000010, 0x00100000,
+	0x0000a968, 0x00000008, 0x00010000,
+	0x000097a0, 0x00000008, 0x00010000,
+	0x00011990, 0x00000008, 0x00010000,
+	0x0000f018, 0x00000008, 0x00010000,
+	0x00012628, 0x00000008, 0x00010000,
+	0x00011da8, 0x00000008, 0x00010000,
+	0x0000aa78, 0x00000030, 0x00100000,
+	0x0000d768, 0x00000028, 0x00280000,
+	0x00009a58, 0x00000018, 0x00180000,
+	0x00009bd8, 0x00000008, 0x00080000,
+	0x00013a18, 0x00000008, 0x00080000,
+	0x000126e8, 0x00000018, 0x00180000,
+	0x0000e608, 0x00500288, 0x00100000,
+	0x00012970, 0x00000138, 0x00280000,
 };
 
 /* Runtime array offsets */
diff --git a/include/linux/qed/common_hsi.h b/include/linux/qed/common_hsi.h
index 03f59a28fefd..3f437e826a4c 100644
--- a/include/linux/qed/common_hsi.h
+++ b/include/linux/qed/common_hsi.h
@@ -109,8 +109,8 @@
 #define MAX_NUM_LL2_TX_STATS_COUNTERS	48
 
 #define FW_MAJOR_VERSION	8
-#define FW_MINOR_VERSION        37
-#define FW_REVISION_VERSION     7
+#define FW_MINOR_VERSION	42
+#define FW_REVISION_VERSION	2
 #define FW_ENGINEERING_VERSION	0
 
 /***********************/
-- 
cgit v1.2.3-59-g8ed1b


From 6aebde8dc767088a3171b864755f26ceae2f8bc7 Mon Sep 17 00:00:00 2001
From: Michal Kalderon <michal.kalderon@marvell.com>
Date: Mon, 27 Jan 2020 15:26:08 +0200
Subject: qed: FW 8.42.2.0 Expose new registers and change windows

This patch contains register initialization related changes.
- Modifications to the runtime offsets - these are defines used
  by the driver or firmware functions to set values that are used
  by the initialization functions to set device register values.
- Global window values changes to provide different device register
  ranges.
- Additional device registers addresses were added to the register file,
  used in later stages.

Signed-off-by: Ariel Elior <ariel.elior@marvell.com>
Signed-off-by: Michal Kalderon <michal.kalderon@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/qlogic/qed/qed_hsi.h          | 961 ++++++++++-----------
 .../net/ethernet/qlogic/qed/qed_init_fw_funcs.c    |   9 +-
 drivers/net/ethernet/qlogic/qed/qed_init_ops.c     |  18 +-
 drivers/net/ethernet/qlogic/qed/qed_reg_addr.h     |  38 +
 4 files changed, 488 insertions(+), 538 deletions(-)

diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
index aaff7117291c..efabe0ca5cbe 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
@@ -4645,529 +4645,444 @@ static const u32 iro_arr[] = {
 };
 
 /* Runtime array offsets */
-#define DORQ_REG_PF_MAX_ICID_0_RT_OFFSET			0
-#define DORQ_REG_PF_MAX_ICID_1_RT_OFFSET			1
-#define DORQ_REG_PF_MAX_ICID_2_RT_OFFSET			2
-#define DORQ_REG_PF_MAX_ICID_3_RT_OFFSET			3
-#define DORQ_REG_PF_MAX_ICID_4_RT_OFFSET			4
-#define DORQ_REG_PF_MAX_ICID_5_RT_OFFSET			5
-#define DORQ_REG_PF_MAX_ICID_6_RT_OFFSET			6
-#define DORQ_REG_PF_MAX_ICID_7_RT_OFFSET			7
-#define DORQ_REG_VF_MAX_ICID_0_RT_OFFSET			8
-#define DORQ_REG_VF_MAX_ICID_1_RT_OFFSET			9
-#define DORQ_REG_VF_MAX_ICID_2_RT_OFFSET			10
-#define DORQ_REG_VF_MAX_ICID_3_RT_OFFSET			11
-#define DORQ_REG_VF_MAX_ICID_4_RT_OFFSET			12
-#define DORQ_REG_VF_MAX_ICID_5_RT_OFFSET			13
-#define DORQ_REG_VF_MAX_ICID_6_RT_OFFSET			14
-#define DORQ_REG_VF_MAX_ICID_7_RT_OFFSET			15
-#define DORQ_REG_PF_WAKE_ALL_RT_OFFSET				16
-#define DORQ_REG_TAG1_ETHERTYPE_RT_OFFSET			17
-#define DORQ_REG_GLB_MAX_ICID_0_RT_OFFSET			18
-#define DORQ_REG_GLB_MAX_ICID_1_RT_OFFSET			19
-#define DORQ_REG_GLB_RANGE2CONN_TYPE_0_RT_OFFSET		20
-#define DORQ_REG_GLB_RANGE2CONN_TYPE_1_RT_OFFSET		21
-#define DORQ_REG_PRV_PF_MAX_ICID_2_RT_OFFSET			22
-#define DORQ_REG_PRV_PF_MAX_ICID_3_RT_OFFSET			23
-#define DORQ_REG_PRV_PF_MAX_ICID_4_RT_OFFSET			24
-#define DORQ_REG_PRV_PF_MAX_ICID_5_RT_OFFSET			25
-#define DORQ_REG_PRV_VF_MAX_ICID_2_RT_OFFSET			26
-#define DORQ_REG_PRV_VF_MAX_ICID_3_RT_OFFSET			27
-#define DORQ_REG_PRV_VF_MAX_ICID_4_RT_OFFSET			28
-#define DORQ_REG_PRV_VF_MAX_ICID_5_RT_OFFSET			29
-#define DORQ_REG_PRV_PF_RANGE2CONN_TYPE_2_RT_OFFSET		30
-#define DORQ_REG_PRV_PF_RANGE2CONN_TYPE_3_RT_OFFSET		31
-#define DORQ_REG_PRV_PF_RANGE2CONN_TYPE_4_RT_OFFSET		32
-#define DORQ_REG_PRV_PF_RANGE2CONN_TYPE_5_RT_OFFSET		33
-#define DORQ_REG_PRV_VF_RANGE2CONN_TYPE_2_RT_OFFSET		34
-#define DORQ_REG_PRV_VF_RANGE2CONN_TYPE_3_RT_OFFSET		35
-#define DORQ_REG_PRV_VF_RANGE2CONN_TYPE_4_RT_OFFSET		36
-#define DORQ_REG_PRV_VF_RANGE2CONN_TYPE_5_RT_OFFSET		37
-#define IGU_REG_PF_CONFIGURATION_RT_OFFSET			38
-#define IGU_REG_VF_CONFIGURATION_RT_OFFSET			39
-#define IGU_REG_ATTN_MSG_ADDR_L_RT_OFFSET			40
-#define IGU_REG_ATTN_MSG_ADDR_H_RT_OFFSET			41
-#define IGU_REG_LEADING_EDGE_LATCH_RT_OFFSET			42
-#define IGU_REG_TRAILING_EDGE_LATCH_RT_OFFSET			43
-#define CAU_REG_CQE_AGG_UNIT_SIZE_RT_OFFSET			44
-#define CAU_REG_SB_VAR_MEMORY_RT_OFFSET				45
-#define CAU_REG_SB_VAR_MEMORY_RT_SIZE				1024
-#define CAU_REG_SB_ADDR_MEMORY_RT_OFFSET			1069
-#define CAU_REG_SB_ADDR_MEMORY_RT_SIZE				1024
-#define CAU_REG_PI_MEMORY_RT_OFFSET				2093
-#define CAU_REG_PI_MEMORY_RT_SIZE				4416
-#define PRS_REG_SEARCH_RESP_INITIATOR_TYPE_RT_OFFSET		6509
-#define PRS_REG_TASK_ID_MAX_INITIATOR_PF_RT_OFFSET		6510
-#define PRS_REG_TASK_ID_MAX_INITIATOR_VF_RT_OFFSET		6511
-#define PRS_REG_TASK_ID_MAX_TARGET_PF_RT_OFFSET			6512
-#define PRS_REG_TASK_ID_MAX_TARGET_VF_RT_OFFSET			6513
-#define PRS_REG_SEARCH_TCP_RT_OFFSET				6514
-#define PRS_REG_SEARCH_FCOE_RT_OFFSET				6515
-#define PRS_REG_SEARCH_ROCE_RT_OFFSET				6516
-#define PRS_REG_ROCE_DEST_QP_MAX_VF_RT_OFFSET			6517
-#define PRS_REG_ROCE_DEST_QP_MAX_PF_RT_OFFSET			6518
-#define PRS_REG_SEARCH_OPENFLOW_RT_OFFSET			6519
-#define PRS_REG_SEARCH_NON_IP_AS_OPENFLOW_RT_OFFSET		6520
-#define PRS_REG_OPENFLOW_SUPPORT_ONLY_KNOWN_OVER_IP_RT_OFFSET	6521
-#define PRS_REG_OPENFLOW_SEARCH_KEY_MASK_RT_OFFSET		6522
-#define PRS_REG_TAG_ETHERTYPE_0_RT_OFFSET			6523
-#define PRS_REG_LIGHT_L2_ETHERTYPE_EN_RT_OFFSET			6524
-#define SRC_REG_FIRSTFREE_RT_OFFSET				6525
-#define SRC_REG_FIRSTFREE_RT_SIZE				2
-#define SRC_REG_LASTFREE_RT_OFFSET				6527
-#define SRC_REG_LASTFREE_RT_SIZE				2
-#define SRC_REG_COUNTFREE_RT_OFFSET				6529
-#define SRC_REG_NUMBER_HASH_BITS_RT_OFFSET			6530
-#define PSWRQ2_REG_CDUT_P_SIZE_RT_OFFSET			6531
-#define PSWRQ2_REG_CDUC_P_SIZE_RT_OFFSET			6532
-#define PSWRQ2_REG_TM_P_SIZE_RT_OFFSET				6533
-#define PSWRQ2_REG_QM_P_SIZE_RT_OFFSET				6534
-#define PSWRQ2_REG_SRC_P_SIZE_RT_OFFSET				6535
-#define PSWRQ2_REG_TSDM_P_SIZE_RT_OFFSET			6536
-#define PSWRQ2_REG_TM_FIRST_ILT_RT_OFFSET			6537
-#define PSWRQ2_REG_TM_LAST_ILT_RT_OFFSET			6538
-#define PSWRQ2_REG_QM_FIRST_ILT_RT_OFFSET			6539
-#define PSWRQ2_REG_QM_LAST_ILT_RT_OFFSET			6540
-#define PSWRQ2_REG_SRC_FIRST_ILT_RT_OFFSET			6541
-#define PSWRQ2_REG_SRC_LAST_ILT_RT_OFFSET			6542
-#define PSWRQ2_REG_CDUC_FIRST_ILT_RT_OFFSET			6543
-#define PSWRQ2_REG_CDUC_LAST_ILT_RT_OFFSET			6544
-#define PSWRQ2_REG_CDUT_FIRST_ILT_RT_OFFSET			6545
-#define PSWRQ2_REG_CDUT_LAST_ILT_RT_OFFSET			6546
-#define PSWRQ2_REG_TSDM_FIRST_ILT_RT_OFFSET			6547
-#define PSWRQ2_REG_TSDM_LAST_ILT_RT_OFFSET			6548
-#define PSWRQ2_REG_TM_NUMBER_OF_PF_BLOCKS_RT_OFFSET		6549
-#define PSWRQ2_REG_CDUT_NUMBER_OF_PF_BLOCKS_RT_OFFSET		6550
-#define PSWRQ2_REG_CDUC_NUMBER_OF_PF_BLOCKS_RT_OFFSET		6551
-#define PSWRQ2_REG_TM_VF_BLOCKS_RT_OFFSET			6552
-#define PSWRQ2_REG_CDUT_VF_BLOCKS_RT_OFFSET			6553
-#define PSWRQ2_REG_CDUC_VF_BLOCKS_RT_OFFSET			6554
-#define PSWRQ2_REG_TM_BLOCKS_FACTOR_RT_OFFSET			6555
-#define PSWRQ2_REG_CDUT_BLOCKS_FACTOR_RT_OFFSET			6556
-#define PSWRQ2_REG_CDUC_BLOCKS_FACTOR_RT_OFFSET			6557
-#define PSWRQ2_REG_VF_BASE_RT_OFFSET				6558
-#define PSWRQ2_REG_VF_LAST_ILT_RT_OFFSET			6559
-#define PSWRQ2_REG_DRAM_ALIGN_WR_RT_OFFSET			6560
-#define PSWRQ2_REG_DRAM_ALIGN_RD_RT_OFFSET			6561
-#define PSWRQ2_REG_TGSRC_FIRST_ILT_RT_OFFSET			6562
-#define PSWRQ2_REG_RGSRC_FIRST_ILT_RT_OFFSET			6563
-#define PSWRQ2_REG_TGSRC_LAST_ILT_RT_OFFSET			6564
-#define PSWRQ2_REG_RGSRC_LAST_ILT_RT_OFFSET			6565
-#define PSWRQ2_REG_ILT_MEMORY_RT_OFFSET				6566
-#define PSWRQ2_REG_ILT_MEMORY_RT_SIZE				26414
-#define PGLUE_REG_B_VF_BASE_RT_OFFSET				32980
-#define PGLUE_REG_B_MSDM_OFFSET_MASK_B_RT_OFFSET		32981
-#define PGLUE_REG_B_MSDM_VF_SHIFT_B_RT_OFFSET			32982
-#define PGLUE_REG_B_CACHE_LINE_SIZE_RT_OFFSET			32983
-#define PGLUE_REG_B_PF_BAR0_SIZE_RT_OFFSET			32984
-#define PGLUE_REG_B_PF_BAR1_SIZE_RT_OFFSET			32985
-#define PGLUE_REG_B_VF_BAR1_SIZE_RT_OFFSET			32986
-#define TM_REG_VF_ENABLE_CONN_RT_OFFSET				32987
-#define TM_REG_PF_ENABLE_CONN_RT_OFFSET				32988
-#define TM_REG_PF_ENABLE_TASK_RT_OFFSET				32989
-#define TM_REG_GROUP_SIZE_RESOLUTION_CONN_RT_OFFSET		32990
-#define TM_REG_GROUP_SIZE_RESOLUTION_TASK_RT_OFFSET		32991
-#define TM_REG_CONFIG_CONN_MEM_RT_OFFSET			32992
-#define TM_REG_CONFIG_CONN_MEM_RT_SIZE				416
-#define TM_REG_CONFIG_TASK_MEM_RT_OFFSET			33408
-#define TM_REG_CONFIG_TASK_MEM_RT_SIZE				608
-#define QM_REG_MAXPQSIZE_0_RT_OFFSET				34016
-#define QM_REG_MAXPQSIZE_1_RT_OFFSET				34017
-#define QM_REG_MAXPQSIZE_2_RT_OFFSET				34018
-#define QM_REG_MAXPQSIZETXSEL_0_RT_OFFSET			34019
-#define QM_REG_MAXPQSIZETXSEL_1_RT_OFFSET			34020
-#define QM_REG_MAXPQSIZETXSEL_2_RT_OFFSET			34021
-#define QM_REG_MAXPQSIZETXSEL_3_RT_OFFSET			34022
-#define QM_REG_MAXPQSIZETXSEL_4_RT_OFFSET			34023
-#define QM_REG_MAXPQSIZETXSEL_5_RT_OFFSET			34024
-#define QM_REG_MAXPQSIZETXSEL_6_RT_OFFSET			34025
-#define QM_REG_MAXPQSIZETXSEL_7_RT_OFFSET			34026
-#define QM_REG_MAXPQSIZETXSEL_8_RT_OFFSET			34027
-#define QM_REG_MAXPQSIZETXSEL_9_RT_OFFSET			34028
-#define QM_REG_MAXPQSIZETXSEL_10_RT_OFFSET			34029
-#define QM_REG_MAXPQSIZETXSEL_11_RT_OFFSET			34030
-#define QM_REG_MAXPQSIZETXSEL_12_RT_OFFSET			34031
-#define QM_REG_MAXPQSIZETXSEL_13_RT_OFFSET			34032
-#define QM_REG_MAXPQSIZETXSEL_14_RT_OFFSET			34033
-#define QM_REG_MAXPQSIZETXSEL_15_RT_OFFSET			34034
-#define QM_REG_MAXPQSIZETXSEL_16_RT_OFFSET			34035
-#define QM_REG_MAXPQSIZETXSEL_17_RT_OFFSET			34036
-#define QM_REG_MAXPQSIZETXSEL_18_RT_OFFSET			34037
-#define QM_REG_MAXPQSIZETXSEL_19_RT_OFFSET			34038
-#define QM_REG_MAXPQSIZETXSEL_20_RT_OFFSET			34039
-#define QM_REG_MAXPQSIZETXSEL_21_RT_OFFSET			34040
-#define QM_REG_MAXPQSIZETXSEL_22_RT_OFFSET			34041
-#define QM_REG_MAXPQSIZETXSEL_23_RT_OFFSET			34042
-#define QM_REG_MAXPQSIZETXSEL_24_RT_OFFSET			34043
-#define QM_REG_MAXPQSIZETXSEL_25_RT_OFFSET			34044
-#define QM_REG_MAXPQSIZETXSEL_26_RT_OFFSET			34045
-#define QM_REG_MAXPQSIZETXSEL_27_RT_OFFSET			34046
-#define QM_REG_MAXPQSIZETXSEL_28_RT_OFFSET			34047
-#define QM_REG_MAXPQSIZETXSEL_29_RT_OFFSET			34048
-#define QM_REG_MAXPQSIZETXSEL_30_RT_OFFSET			34049
-#define QM_REG_MAXPQSIZETXSEL_31_RT_OFFSET			34050
-#define QM_REG_MAXPQSIZETXSEL_32_RT_OFFSET			34051
-#define QM_REG_MAXPQSIZETXSEL_33_RT_OFFSET			34052
-#define QM_REG_MAXPQSIZETXSEL_34_RT_OFFSET			34053
-#define QM_REG_MAXPQSIZETXSEL_35_RT_OFFSET			34054
-#define QM_REG_MAXPQSIZETXSEL_36_RT_OFFSET			34055
-#define QM_REG_MAXPQSIZETXSEL_37_RT_OFFSET			34056
-#define QM_REG_MAXPQSIZETXSEL_38_RT_OFFSET			34057
-#define QM_REG_MAXPQSIZETXSEL_39_RT_OFFSET			34058
-#define QM_REG_MAXPQSIZETXSEL_40_RT_OFFSET			34059
-#define QM_REG_MAXPQSIZETXSEL_41_RT_OFFSET			34060
-#define QM_REG_MAXPQSIZETXSEL_42_RT_OFFSET			34061
-#define QM_REG_MAXPQSIZETXSEL_43_RT_OFFSET			34062
-#define QM_REG_MAXPQSIZETXSEL_44_RT_OFFSET			34063
-#define QM_REG_MAXPQSIZETXSEL_45_RT_OFFSET			34064
-#define QM_REG_MAXPQSIZETXSEL_46_RT_OFFSET			34065
-#define QM_REG_MAXPQSIZETXSEL_47_RT_OFFSET			34066
-#define QM_REG_MAXPQSIZETXSEL_48_RT_OFFSET			34067
-#define QM_REG_MAXPQSIZETXSEL_49_RT_OFFSET			34068
-#define QM_REG_MAXPQSIZETXSEL_50_RT_OFFSET			34069
-#define QM_REG_MAXPQSIZETXSEL_51_RT_OFFSET			34070
-#define QM_REG_MAXPQSIZETXSEL_52_RT_OFFSET			34071
-#define QM_REG_MAXPQSIZETXSEL_53_RT_OFFSET			34072
-#define QM_REG_MAXPQSIZETXSEL_54_RT_OFFSET			34073
-#define QM_REG_MAXPQSIZETXSEL_55_RT_OFFSET			34074
-#define QM_REG_MAXPQSIZETXSEL_56_RT_OFFSET			34075
-#define QM_REG_MAXPQSIZETXSEL_57_RT_OFFSET			34076
-#define QM_REG_MAXPQSIZETXSEL_58_RT_OFFSET			34077
-#define QM_REG_MAXPQSIZETXSEL_59_RT_OFFSET			34078
-#define QM_REG_MAXPQSIZETXSEL_60_RT_OFFSET			34079
-#define QM_REG_MAXPQSIZETXSEL_61_RT_OFFSET			34080
-#define QM_REG_MAXPQSIZETXSEL_62_RT_OFFSET			34081
-#define QM_REG_MAXPQSIZETXSEL_63_RT_OFFSET			34082
-#define QM_REG_BASEADDROTHERPQ_RT_OFFSET			34083
-#define QM_REG_BASEADDROTHERPQ_RT_SIZE				128
-#define QM_REG_PTRTBLOTHER_RT_OFFSET				34211
-#define QM_REG_PTRTBLOTHER_RT_SIZE				256
-#define QM_REG_AFULLQMBYPTHRPFWFQ_RT_OFFSET			34467
-#define QM_REG_AFULLQMBYPTHRVPWFQ_RT_OFFSET			34468
-#define QM_REG_AFULLQMBYPTHRPFRL_RT_OFFSET			34469
-#define QM_REG_AFULLQMBYPTHRGLBLRL_RT_OFFSET			34470
-#define QM_REG_AFULLOPRTNSTCCRDMASK_RT_OFFSET			34471
-#define QM_REG_WRROTHERPQGRP_0_RT_OFFSET			34472
-#define QM_REG_WRROTHERPQGRP_1_RT_OFFSET			34473
-#define QM_REG_WRROTHERPQGRP_2_RT_OFFSET			34474
-#define QM_REG_WRROTHERPQGRP_3_RT_OFFSET			34475
-#define QM_REG_WRROTHERPQGRP_4_RT_OFFSET			34476
-#define QM_REG_WRROTHERPQGRP_5_RT_OFFSET			34477
-#define QM_REG_WRROTHERPQGRP_6_RT_OFFSET			34478
-#define QM_REG_WRROTHERPQGRP_7_RT_OFFSET			34479
-#define QM_REG_WRROTHERPQGRP_8_RT_OFFSET			34480
-#define QM_REG_WRROTHERPQGRP_9_RT_OFFSET			34481
-#define QM_REG_WRROTHERPQGRP_10_RT_OFFSET			34482
-#define QM_REG_WRROTHERPQGRP_11_RT_OFFSET			34483
-#define QM_REG_WRROTHERPQGRP_12_RT_OFFSET			34484
-#define QM_REG_WRROTHERPQGRP_13_RT_OFFSET			34485
-#define QM_REG_WRROTHERPQGRP_14_RT_OFFSET			34486
-#define QM_REG_WRROTHERPQGRP_15_RT_OFFSET			34487
-#define QM_REG_WRROTHERGRPWEIGHT_0_RT_OFFSET			34488
-#define QM_REG_WRROTHERGRPWEIGHT_1_RT_OFFSET			34489
-#define QM_REG_WRROTHERGRPWEIGHT_2_RT_OFFSET			34490
-#define QM_REG_WRROTHERGRPWEIGHT_3_RT_OFFSET			34491
-#define QM_REG_WRRTXGRPWEIGHT_0_RT_OFFSET			34492
-#define QM_REG_WRRTXGRPWEIGHT_1_RT_OFFSET			34493
-#define QM_REG_PQTX2PF_0_RT_OFFSET				34494
-#define QM_REG_PQTX2PF_1_RT_OFFSET				34495
-#define QM_REG_PQTX2PF_2_RT_OFFSET				34496
-#define QM_REG_PQTX2PF_3_RT_OFFSET				34497
-#define QM_REG_PQTX2PF_4_RT_OFFSET				34498
-#define QM_REG_PQTX2PF_5_RT_OFFSET				34499
-#define QM_REG_PQTX2PF_6_RT_OFFSET				34500
-#define QM_REG_PQTX2PF_7_RT_OFFSET				34501
-#define QM_REG_PQTX2PF_8_RT_OFFSET				34502
-#define QM_REG_PQTX2PF_9_RT_OFFSET				34503
-#define QM_REG_PQTX2PF_10_RT_OFFSET				34504
-#define QM_REG_PQTX2PF_11_RT_OFFSET				34505
-#define QM_REG_PQTX2PF_12_RT_OFFSET				34506
-#define QM_REG_PQTX2PF_13_RT_OFFSET				34507
-#define QM_REG_PQTX2PF_14_RT_OFFSET				34508
-#define QM_REG_PQTX2PF_15_RT_OFFSET				34509
-#define QM_REG_PQTX2PF_16_RT_OFFSET				34510
-#define QM_REG_PQTX2PF_17_RT_OFFSET				34511
-#define QM_REG_PQTX2PF_18_RT_OFFSET				34512
-#define QM_REG_PQTX2PF_19_RT_OFFSET				34513
-#define QM_REG_PQTX2PF_20_RT_OFFSET				34514
-#define QM_REG_PQTX2PF_21_RT_OFFSET				34515
-#define QM_REG_PQTX2PF_22_RT_OFFSET				34516
-#define QM_REG_PQTX2PF_23_RT_OFFSET				34517
-#define QM_REG_PQTX2PF_24_RT_OFFSET				34518
-#define QM_REG_PQTX2PF_25_RT_OFFSET				34519
-#define QM_REG_PQTX2PF_26_RT_OFFSET				34520
-#define QM_REG_PQTX2PF_27_RT_OFFSET				34521
-#define QM_REG_PQTX2PF_28_RT_OFFSET				34522
-#define QM_REG_PQTX2PF_29_RT_OFFSET				34523
-#define QM_REG_PQTX2PF_30_RT_OFFSET				34524
-#define QM_REG_PQTX2PF_31_RT_OFFSET				34525
-#define QM_REG_PQTX2PF_32_RT_OFFSET				34526
-#define QM_REG_PQTX2PF_33_RT_OFFSET				34527
-#define QM_REG_PQTX2PF_34_RT_OFFSET				34528
-#define QM_REG_PQTX2PF_35_RT_OFFSET				34529
-#define QM_REG_PQTX2PF_36_RT_OFFSET				34530
-#define QM_REG_PQTX2PF_37_RT_OFFSET				34531
-#define QM_REG_PQTX2PF_38_RT_OFFSET				34532
-#define QM_REG_PQTX2PF_39_RT_OFFSET				34533
-#define QM_REG_PQTX2PF_40_RT_OFFSET				34534
-#define QM_REG_PQTX2PF_41_RT_OFFSET				34535
-#define QM_REG_PQTX2PF_42_RT_OFFSET				34536
-#define QM_REG_PQTX2PF_43_RT_OFFSET				34537
-#define QM_REG_PQTX2PF_44_RT_OFFSET				34538
-#define QM_REG_PQTX2PF_45_RT_OFFSET				34539
-#define QM_REG_PQTX2PF_46_RT_OFFSET				34540
-#define QM_REG_PQTX2PF_47_RT_OFFSET				34541
-#define QM_REG_PQTX2PF_48_RT_OFFSET				34542
-#define QM_REG_PQTX2PF_49_RT_OFFSET				34543
-#define QM_REG_PQTX2PF_50_RT_OFFSET				34544
-#define QM_REG_PQTX2PF_51_RT_OFFSET				34545
-#define QM_REG_PQTX2PF_52_RT_OFFSET				34546
-#define QM_REG_PQTX2PF_53_RT_OFFSET				34547
-#define QM_REG_PQTX2PF_54_RT_OFFSET				34548
-#define QM_REG_PQTX2PF_55_RT_OFFSET				34549
-#define QM_REG_PQTX2PF_56_RT_OFFSET				34550
-#define QM_REG_PQTX2PF_57_RT_OFFSET				34551
-#define QM_REG_PQTX2PF_58_RT_OFFSET				34552
-#define QM_REG_PQTX2PF_59_RT_OFFSET				34553
-#define QM_REG_PQTX2PF_60_RT_OFFSET				34554
-#define QM_REG_PQTX2PF_61_RT_OFFSET				34555
-#define QM_REG_PQTX2PF_62_RT_OFFSET				34556
-#define QM_REG_PQTX2PF_63_RT_OFFSET				34557
-#define QM_REG_PQOTHER2PF_0_RT_OFFSET				34558
-#define QM_REG_PQOTHER2PF_1_RT_OFFSET				34559
-#define QM_REG_PQOTHER2PF_2_RT_OFFSET				34560
-#define QM_REG_PQOTHER2PF_3_RT_OFFSET				34561
-#define QM_REG_PQOTHER2PF_4_RT_OFFSET				34562
-#define QM_REG_PQOTHER2PF_5_RT_OFFSET				34563
-#define QM_REG_PQOTHER2PF_6_RT_OFFSET				34564
-#define QM_REG_PQOTHER2PF_7_RT_OFFSET				34565
-#define QM_REG_PQOTHER2PF_8_RT_OFFSET				34566
-#define QM_REG_PQOTHER2PF_9_RT_OFFSET				34567
-#define QM_REG_PQOTHER2PF_10_RT_OFFSET				34568
-#define QM_REG_PQOTHER2PF_11_RT_OFFSET				34569
-#define QM_REG_PQOTHER2PF_12_RT_OFFSET				34570
-#define QM_REG_PQOTHER2PF_13_RT_OFFSET				34571
-#define QM_REG_PQOTHER2PF_14_RT_OFFSET				34572
-#define QM_REG_PQOTHER2PF_15_RT_OFFSET				34573
-#define QM_REG_RLGLBLPERIOD_0_RT_OFFSET				34574
-#define QM_REG_RLGLBLPERIOD_1_RT_OFFSET				34575
-#define QM_REG_RLGLBLPERIODTIMER_0_RT_OFFSET			34576
-#define QM_REG_RLGLBLPERIODTIMER_1_RT_OFFSET			34577
-#define QM_REG_RLGLBLPERIODSEL_0_RT_OFFSET			34578
-#define QM_REG_RLGLBLPERIODSEL_1_RT_OFFSET			34579
-#define QM_REG_RLGLBLPERIODSEL_2_RT_OFFSET			34580
-#define QM_REG_RLGLBLPERIODSEL_3_RT_OFFSET			34581
-#define QM_REG_RLGLBLPERIODSEL_4_RT_OFFSET			34582
-#define QM_REG_RLGLBLPERIODSEL_5_RT_OFFSET			34583
-#define QM_REG_RLGLBLPERIODSEL_6_RT_OFFSET			34584
-#define QM_REG_RLGLBLPERIODSEL_7_RT_OFFSET			34585
-#define QM_REG_RLGLBLINCVAL_RT_OFFSET				34586
-#define QM_REG_RLGLBLINCVAL_RT_SIZE				256
-#define QM_REG_RLGLBLUPPERBOUND_RT_OFFSET			34842
-#define QM_REG_RLGLBLUPPERBOUND_RT_SIZE				256
-#define QM_REG_RLGLBLCRD_RT_OFFSET				35098
-#define QM_REG_RLGLBLCRD_RT_SIZE				256
-#define QM_REG_RLGLBLENABLE_RT_OFFSET				35354
-#define QM_REG_RLPFPERIOD_RT_OFFSET				35355
-#define QM_REG_RLPFPERIODTIMER_RT_OFFSET			35356
-#define QM_REG_RLPFINCVAL_RT_OFFSET				35357
-#define QM_REG_RLPFINCVAL_RT_SIZE				16
-#define QM_REG_RLPFUPPERBOUND_RT_OFFSET				35373
-#define QM_REG_RLPFUPPERBOUND_RT_SIZE				16
-#define QM_REG_RLPFCRD_RT_OFFSET				35389
-#define QM_REG_RLPFCRD_RT_SIZE					16
-#define QM_REG_RLPFENABLE_RT_OFFSET				35405
-#define QM_REG_RLPFVOQENABLE_RT_OFFSET				35406
-#define QM_REG_WFQPFWEIGHT_RT_OFFSET				35407
-#define QM_REG_WFQPFWEIGHT_RT_SIZE				16
-#define QM_REG_WFQPFUPPERBOUND_RT_OFFSET			35423
-#define QM_REG_WFQPFUPPERBOUND_RT_SIZE				16
-#define QM_REG_WFQPFCRD_RT_OFFSET				35439
-#define QM_REG_WFQPFCRD_RT_SIZE					256
-#define QM_REG_WFQPFENABLE_RT_OFFSET				35695
-#define QM_REG_WFQVPENABLE_RT_OFFSET				35696
-#define QM_REG_BASEADDRTXPQ_RT_OFFSET				35697
-#define QM_REG_BASEADDRTXPQ_RT_SIZE				512
-#define QM_REG_TXPQMAP_RT_OFFSET				36209
-#define QM_REG_TXPQMAP_RT_SIZE					512
-#define QM_REG_WFQVPWEIGHT_RT_OFFSET				36721
-#define QM_REG_WFQVPWEIGHT_RT_SIZE				512
-#define QM_REG_WFQVPCRD_RT_OFFSET				37233
-#define QM_REG_WFQVPCRD_RT_SIZE					512
-#define QM_REG_WFQVPMAP_RT_OFFSET				37745
-#define QM_REG_WFQVPMAP_RT_SIZE					512
-#define QM_REG_PTRTBLTX_RT_OFFSET				38257
-#define QM_REG_PTRTBLTX_RT_SIZE					1024
-#define QM_REG_WFQPFCRD_MSB_RT_OFFSET				39281
-#define QM_REG_WFQPFCRD_MSB_RT_SIZE				320
-#define QM_REG_VOQCRDLINE_RT_OFFSET				39601
-#define QM_REG_VOQCRDLINE_RT_SIZE				36
-#define QM_REG_VOQINITCRDLINE_RT_OFFSET				39637
-#define QM_REG_VOQINITCRDLINE_RT_SIZE				36
-#define QM_REG_RLPFVOQENABLE_MSB_RT_OFFSET			39673
-#define NIG_REG_TAG_ETHERTYPE_0_RT_OFFSET			39674
-#define NIG_REG_BRB_GATE_DNTFWD_PORT_RT_OFFSET			39675
-#define NIG_REG_OUTER_TAG_VALUE_LIST0_RT_OFFSET			39676
-#define NIG_REG_OUTER_TAG_VALUE_LIST1_RT_OFFSET			39677
-#define NIG_REG_OUTER_TAG_VALUE_LIST2_RT_OFFSET			39678
-#define NIG_REG_OUTER_TAG_VALUE_LIST3_RT_OFFSET			39679
-#define NIG_REG_LLH_FUNC_TAGMAC_CLS_TYPE_RT_OFFSET		39680
-#define NIG_REG_LLH_FUNC_TAG_EN_RT_OFFSET			39681
-#define NIG_REG_LLH_FUNC_TAG_EN_RT_SIZE				4
-#define NIG_REG_LLH_FUNC_TAG_VALUE_RT_OFFSET			39685
-#define NIG_REG_LLH_FUNC_TAG_VALUE_RT_SIZE			4
-#define NIG_REG_LLH_FUNC_FILTER_VALUE_RT_OFFSET			39689
-#define NIG_REG_LLH_FUNC_FILTER_VALUE_RT_SIZE			32
-#define NIG_REG_LLH_FUNC_FILTER_EN_RT_OFFSET			39721
-#define NIG_REG_LLH_FUNC_FILTER_EN_RT_SIZE			16
-#define NIG_REG_LLH_FUNC_FILTER_MODE_RT_OFFSET			39737
-#define NIG_REG_LLH_FUNC_FILTER_MODE_RT_SIZE			16
-#define NIG_REG_LLH_FUNC_FILTER_PROTOCOL_TYPE_RT_OFFSET		39753
-#define NIG_REG_LLH_FUNC_FILTER_PROTOCOL_TYPE_RT_SIZE		16
-#define NIG_REG_LLH_FUNC_FILTER_HDR_SEL_RT_OFFSET		39769
-#define NIG_REG_LLH_FUNC_FILTER_HDR_SEL_RT_SIZE			16
-#define NIG_REG_TX_EDPM_CTRL_RT_OFFSET				39785
-#define NIG_REG_PPF_TO_ENGINE_SEL_RT_OFFSET                             39786
-#define NIG_REG_PPF_TO_ENGINE_SEL_RT_SIZE                               8
-#define NIG_REG_LLH_PF_CLS_FUNC_FILTER_VALUE_RT_OFFSET                  39794
-#define NIG_REG_LLH_PF_CLS_FUNC_FILTER_VALUE_RT_SIZE                    1024
-#define NIG_REG_LLH_PF_CLS_FUNC_FILTER_EN_RT_OFFSET                     40818
-#define NIG_REG_LLH_PF_CLS_FUNC_FILTER_EN_RT_SIZE                       512
-#define NIG_REG_LLH_PF_CLS_FUNC_FILTER_MODE_RT_OFFSET                   41330
-#define NIG_REG_LLH_PF_CLS_FUNC_FILTER_MODE_RT_SIZE                     512
-#define NIG_REG_LLH_PF_CLS_FUNC_FILTER_PROTOCOL_TYPE_RT_OFFSET          41842
-#define NIG_REG_LLH_PF_CLS_FUNC_FILTER_PROTOCOL_TYPE_RT_SIZE            512
-#define NIG_REG_LLH_PF_CLS_FUNC_FILTER_HDR_SEL_RT_OFFSET                42354
-#define NIG_REG_LLH_PF_CLS_FUNC_FILTER_HDR_SEL_RT_SIZE                  512
-#define NIG_REG_LLH_PF_CLS_FILTERS_MAP_RT_OFFSET                        42866
-#define NIG_REG_LLH_PF_CLS_FILTERS_MAP_RT_SIZE                          32
-#define CDU_REG_CID_ADDR_PARAMS_RT_OFFSET                               42898
-#define CDU_REG_SEGMENT0_PARAMS_RT_OFFSET                               42899
-#define CDU_REG_SEGMENT1_PARAMS_RT_OFFSET                               42900
-#define CDU_REG_PF_SEG0_TYPE_OFFSET_RT_OFFSET                           42901
-#define CDU_REG_PF_SEG1_TYPE_OFFSET_RT_OFFSET                           42902
-#define CDU_REG_PF_SEG2_TYPE_OFFSET_RT_OFFSET                           42903
-#define CDU_REG_PF_SEG3_TYPE_OFFSET_RT_OFFSET                           42904
-#define CDU_REG_PF_FL_SEG0_TYPE_OFFSET_RT_OFFSET                        42905
-#define CDU_REG_PF_FL_SEG1_TYPE_OFFSET_RT_OFFSET                        42906
-#define CDU_REG_PF_FL_SEG2_TYPE_OFFSET_RT_OFFSET                        42907
-#define CDU_REG_PF_FL_SEG3_TYPE_OFFSET_RT_OFFSET                        42908
-#define CDU_REG_VF_SEG_TYPE_OFFSET_RT_OFFSET                            42909
-#define CDU_REG_VF_FL_SEG_TYPE_OFFSET_RT_OFFSET                         42910
-#define PBF_REG_TAG_ETHERTYPE_0_RT_OFFSET                               42911
-#define PBF_REG_BTB_SHARED_AREA_SIZE_RT_OFFSET                          42912
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ0_RT_OFFSET                        42913
-#define PBF_REG_BTB_GUARANTEED_VOQ0_RT_OFFSET                           42914
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ0_RT_OFFSET                    42915
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ1_RT_OFFSET                        42916
-#define PBF_REG_BTB_GUARANTEED_VOQ1_RT_OFFSET                           42917
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ1_RT_OFFSET                    42918
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ2_RT_OFFSET                        42919
-#define PBF_REG_BTB_GUARANTEED_VOQ2_RT_OFFSET                           42920
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ2_RT_OFFSET                    42921
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ3_RT_OFFSET                        42922
-#define PBF_REG_BTB_GUARANTEED_VOQ3_RT_OFFSET                           42923
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ3_RT_OFFSET                    42924
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ4_RT_OFFSET                        42925
-#define PBF_REG_BTB_GUARANTEED_VOQ4_RT_OFFSET                           42926
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ4_RT_OFFSET                    42927
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ5_RT_OFFSET                        42928
-#define PBF_REG_BTB_GUARANTEED_VOQ5_RT_OFFSET                           42929
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ5_RT_OFFSET                    42930
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ6_RT_OFFSET                        42931
-#define PBF_REG_BTB_GUARANTEED_VOQ6_RT_OFFSET                           42932
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ6_RT_OFFSET                    42933
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ7_RT_OFFSET                        42934
-#define PBF_REG_BTB_GUARANTEED_VOQ7_RT_OFFSET                           42935
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ7_RT_OFFSET                    42936
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ8_RT_OFFSET                        42937
-#define PBF_REG_BTB_GUARANTEED_VOQ8_RT_OFFSET                           42938
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ8_RT_OFFSET                    42939
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ9_RT_OFFSET                        42940
-#define PBF_REG_BTB_GUARANTEED_VOQ9_RT_OFFSET                           42941
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ9_RT_OFFSET                    42942
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ10_RT_OFFSET                       42943
-#define PBF_REG_BTB_GUARANTEED_VOQ10_RT_OFFSET                          42944
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ10_RT_OFFSET                   42945
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ11_RT_OFFSET                       42946
-#define PBF_REG_BTB_GUARANTEED_VOQ11_RT_OFFSET                          42947
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ11_RT_OFFSET                   42948
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ12_RT_OFFSET                       42949
-#define PBF_REG_BTB_GUARANTEED_VOQ12_RT_OFFSET                          42950
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ12_RT_OFFSET                   42951
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ13_RT_OFFSET                       42952
-#define PBF_REG_BTB_GUARANTEED_VOQ13_RT_OFFSET                          42953
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ13_RT_OFFSET                   42954
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ14_RT_OFFSET                       42955
-#define PBF_REG_BTB_GUARANTEED_VOQ14_RT_OFFSET                          42956
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ14_RT_OFFSET                   42957
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ15_RT_OFFSET                       42958
-#define PBF_REG_BTB_GUARANTEED_VOQ15_RT_OFFSET                          42959
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ15_RT_OFFSET                   42960
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ16_RT_OFFSET                       42961
-#define PBF_REG_BTB_GUARANTEED_VOQ16_RT_OFFSET                          42962
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ16_RT_OFFSET                   42963
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ17_RT_OFFSET                       42964
-#define PBF_REG_BTB_GUARANTEED_VOQ17_RT_OFFSET                          42965
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ17_RT_OFFSET                   42966
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ18_RT_OFFSET                       42967
-#define PBF_REG_BTB_GUARANTEED_VOQ18_RT_OFFSET                          42968
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ18_RT_OFFSET                   42969
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ19_RT_OFFSET                       42970
-#define PBF_REG_BTB_GUARANTEED_VOQ19_RT_OFFSET                          42971
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ19_RT_OFFSET                   42972
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ20_RT_OFFSET                       42973
-#define PBF_REG_BTB_GUARANTEED_VOQ20_RT_OFFSET                          42974
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ20_RT_OFFSET                   42975
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ21_RT_OFFSET                       42976
-#define PBF_REG_BTB_GUARANTEED_VOQ21_RT_OFFSET                          42977
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ21_RT_OFFSET                   42978
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ22_RT_OFFSET                       42979
-#define PBF_REG_BTB_GUARANTEED_VOQ22_RT_OFFSET                          42980
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ22_RT_OFFSET                   42981
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ23_RT_OFFSET                       42982
-#define PBF_REG_BTB_GUARANTEED_VOQ23_RT_OFFSET                          42983
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ23_RT_OFFSET                   42984
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ24_RT_OFFSET                       42985
-#define PBF_REG_BTB_GUARANTEED_VOQ24_RT_OFFSET                          42986
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ24_RT_OFFSET                   42987
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ25_RT_OFFSET                       42988
-#define PBF_REG_BTB_GUARANTEED_VOQ25_RT_OFFSET                          42989
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ25_RT_OFFSET                   42990
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ26_RT_OFFSET                       42991
-#define PBF_REG_BTB_GUARANTEED_VOQ26_RT_OFFSET                          42992
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ26_RT_OFFSET                   42993
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ27_RT_OFFSET                       42994
-#define PBF_REG_BTB_GUARANTEED_VOQ27_RT_OFFSET                          42995
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ27_RT_OFFSET                   42996
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ28_RT_OFFSET                       42997
-#define PBF_REG_BTB_GUARANTEED_VOQ28_RT_OFFSET                          42998
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ28_RT_OFFSET                   42999
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ29_RT_OFFSET                       43000
-#define PBF_REG_BTB_GUARANTEED_VOQ29_RT_OFFSET                          43001
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ29_RT_OFFSET                   43002
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ30_RT_OFFSET                       43003
-#define PBF_REG_BTB_GUARANTEED_VOQ30_RT_OFFSET                          43004
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ30_RT_OFFSET                   43005
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ31_RT_OFFSET                       43006
-#define PBF_REG_BTB_GUARANTEED_VOQ31_RT_OFFSET                          43007
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ31_RT_OFFSET                   43008
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ32_RT_OFFSET                       43009
-#define PBF_REG_BTB_GUARANTEED_VOQ32_RT_OFFSET                          43010
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ32_RT_OFFSET                   43011
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ33_RT_OFFSET                       43012
-#define PBF_REG_BTB_GUARANTEED_VOQ33_RT_OFFSET                          43013
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ33_RT_OFFSET                   43014
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ34_RT_OFFSET                       43015
-#define PBF_REG_BTB_GUARANTEED_VOQ34_RT_OFFSET                          43016
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ34_RT_OFFSET                   43017
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ35_RT_OFFSET                       43018
-#define PBF_REG_BTB_GUARANTEED_VOQ35_RT_OFFSET                          43019
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ35_RT_OFFSET                   43020
-#define XCM_REG_CON_PHY_Q3_RT_OFFSET                                    43021
-
-#define RUNTIME_ARRAY_SIZE 43022
-
+#define DORQ_REG_PF_MAX_ICID_0_RT_OFFSET				0
+#define DORQ_REG_PF_MAX_ICID_1_RT_OFFSET				1
+#define DORQ_REG_PF_MAX_ICID_2_RT_OFFSET				2
+#define DORQ_REG_PF_MAX_ICID_3_RT_OFFSET				3
+#define DORQ_REG_PF_MAX_ICID_4_RT_OFFSET				4
+#define DORQ_REG_PF_MAX_ICID_5_RT_OFFSET				5
+#define DORQ_REG_PF_MAX_ICID_6_RT_OFFSET				6
+#define DORQ_REG_PF_MAX_ICID_7_RT_OFFSET				7
+#define DORQ_REG_VF_MAX_ICID_0_RT_OFFSET				8
+#define DORQ_REG_VF_MAX_ICID_1_RT_OFFSET				9
+#define DORQ_REG_VF_MAX_ICID_2_RT_OFFSET				10
+#define DORQ_REG_VF_MAX_ICID_3_RT_OFFSET				11
+#define DORQ_REG_VF_MAX_ICID_4_RT_OFFSET				12
+#define DORQ_REG_VF_MAX_ICID_5_RT_OFFSET				13
+#define DORQ_REG_VF_MAX_ICID_6_RT_OFFSET				14
+#define DORQ_REG_VF_MAX_ICID_7_RT_OFFSET				15
+#define DORQ_REG_VF_ICID_BIT_SHIFT_NORM_RT_OFFSET			16
+#define DORQ_REG_PF_WAKE_ALL_RT_OFFSET					17
+#define DORQ_REG_TAG1_ETHERTYPE_RT_OFFSET				18
+#define IGU_REG_PF_CONFIGURATION_RT_OFFSET				19
+#define IGU_REG_VF_CONFIGURATION_RT_OFFSET				20
+#define IGU_REG_ATTN_MSG_ADDR_L_RT_OFFSET				21
+#define IGU_REG_ATTN_MSG_ADDR_H_RT_OFFSET				22
+#define IGU_REG_LEADING_EDGE_LATCH_RT_OFFSET				23
+#define IGU_REG_TRAILING_EDGE_LATCH_RT_OFFSET				24
+#define CAU_REG_CQE_AGG_UNIT_SIZE_RT_OFFSET				25
+#define CAU_REG_SB_VAR_MEMORY_RT_OFFSET					26
+#define CAU_REG_SB_VAR_MEMORY_RT_SIZE					736
+#define CAU_REG_SB_ADDR_MEMORY_RT_OFFSET				762
+#define CAU_REG_SB_ADDR_MEMORY_RT_SIZE					736
+#define CAU_REG_PI_MEMORY_RT_OFFSET					1498
+#define CAU_REG_PI_MEMORY_RT_SIZE					4416
+#define PRS_REG_SEARCH_RESP_INITIATOR_TYPE_RT_OFFSET			5914
+#define PRS_REG_TASK_ID_MAX_INITIATOR_PF_RT_OFFSET			5915
+#define PRS_REG_TASK_ID_MAX_INITIATOR_VF_RT_OFFSET			5916
+#define PRS_REG_TASK_ID_MAX_TARGET_PF_RT_OFFSET				5917
+#define PRS_REG_TASK_ID_MAX_TARGET_VF_RT_OFFSET				5918
+#define PRS_REG_SEARCH_TCP_RT_OFFSET					5919
+#define PRS_REG_SEARCH_FCOE_RT_OFFSET					5920
+#define PRS_REG_SEARCH_ROCE_RT_OFFSET					5921
+#define PRS_REG_ROCE_DEST_QP_MAX_VF_RT_OFFSET				5922
+#define PRS_REG_ROCE_DEST_QP_MAX_PF_RT_OFFSET				5923
+#define PRS_REG_SEARCH_OPENFLOW_RT_OFFSET				5924
+#define PRS_REG_SEARCH_NON_IP_AS_OPENFLOW_RT_OFFSET			5925
+#define PRS_REG_OPENFLOW_SUPPORT_ONLY_KNOWN_OVER_IP_RT_OFFSET		5926
+#define PRS_REG_OPENFLOW_SEARCH_KEY_MASK_RT_OFFSET			5927
+#define PRS_REG_TAG_ETHERTYPE_0_RT_OFFSET				5928
+#define PRS_REG_LIGHT_L2_ETHERTYPE_EN_RT_OFFSET				5929
+#define SRC_REG_FIRSTFREE_RT_OFFSET					5930
+#define SRC_REG_FIRSTFREE_RT_SIZE					2
+#define SRC_REG_LASTFREE_RT_OFFSET					5932
+#define SRC_REG_LASTFREE_RT_SIZE					2
+#define SRC_REG_COUNTFREE_RT_OFFSET					5934
+#define SRC_REG_NUMBER_HASH_BITS_RT_OFFSET				5935
+#define PSWRQ2_REG_CDUT_P_SIZE_RT_OFFSET				5936
+#define PSWRQ2_REG_CDUC_P_SIZE_RT_OFFSET				5937
+#define PSWRQ2_REG_TM_P_SIZE_RT_OFFSET					5938
+#define PSWRQ2_REG_QM_P_SIZE_RT_OFFSET					5939
+#define PSWRQ2_REG_SRC_P_SIZE_RT_OFFSET					5940
+#define PSWRQ2_REG_TSDM_P_SIZE_RT_OFFSET				5941
+#define PSWRQ2_REG_TM_FIRST_ILT_RT_OFFSET				5942
+#define PSWRQ2_REG_TM_LAST_ILT_RT_OFFSET				5943
+#define PSWRQ2_REG_QM_FIRST_ILT_RT_OFFSET				5944
+#define PSWRQ2_REG_QM_LAST_ILT_RT_OFFSET				5945
+#define PSWRQ2_REG_SRC_FIRST_ILT_RT_OFFSET				5946
+#define PSWRQ2_REG_SRC_LAST_ILT_RT_OFFSET				5947
+#define PSWRQ2_REG_CDUC_FIRST_ILT_RT_OFFSET				5948
+#define PSWRQ2_REG_CDUC_LAST_ILT_RT_OFFSET				5949
+#define PSWRQ2_REG_CDUT_FIRST_ILT_RT_OFFSET				5950
+#define PSWRQ2_REG_CDUT_LAST_ILT_RT_OFFSET				5951
+#define PSWRQ2_REG_TSDM_FIRST_ILT_RT_OFFSET				5952
+#define PSWRQ2_REG_TSDM_LAST_ILT_RT_OFFSET				5953
+#define PSWRQ2_REG_TM_NUMBER_OF_PF_BLOCKS_RT_OFFSET			5954
+#define PSWRQ2_REG_CDUT_NUMBER_OF_PF_BLOCKS_RT_OFFSET			5955
+#define PSWRQ2_REG_CDUC_NUMBER_OF_PF_BLOCKS_RT_OFFSET			5956
+#define PSWRQ2_REG_TM_VF_BLOCKS_RT_OFFSET				5957
+#define PSWRQ2_REG_CDUT_VF_BLOCKS_RT_OFFSET				5958
+#define PSWRQ2_REG_CDUC_VF_BLOCKS_RT_OFFSET				5959
+#define PSWRQ2_REG_TM_BLOCKS_FACTOR_RT_OFFSET				5960
+#define PSWRQ2_REG_CDUT_BLOCKS_FACTOR_RT_OFFSET				5961
+#define PSWRQ2_REG_CDUC_BLOCKS_FACTOR_RT_OFFSET				5962
+#define PSWRQ2_REG_VF_BASE_RT_OFFSET					5963
+#define PSWRQ2_REG_VF_LAST_ILT_RT_OFFSET				5964
+#define PSWRQ2_REG_DRAM_ALIGN_WR_RT_OFFSET				5965
+#define PSWRQ2_REG_DRAM_ALIGN_RD_RT_OFFSET				5966
+#define PSWRQ2_REG_ILT_MEMORY_RT_OFFSET					5967
+#define PSWRQ2_REG_ILT_MEMORY_RT_SIZE					22000
+#define PGLUE_REG_B_VF_BASE_RT_OFFSET					27967
+#define PGLUE_REG_B_MSDM_OFFSET_MASK_B_RT_OFFSET			27968
+#define PGLUE_REG_B_MSDM_VF_SHIFT_B_RT_OFFSET				27969
+#define PGLUE_REG_B_CACHE_LINE_SIZE_RT_OFFSET				27970
+#define PGLUE_REG_B_PF_BAR0_SIZE_RT_OFFSET				27971
+#define PGLUE_REG_B_PF_BAR1_SIZE_RT_OFFSET				27972
+#define PGLUE_REG_B_VF_BAR1_SIZE_RT_OFFSET				27973
+#define TM_REG_VF_ENABLE_CONN_RT_OFFSET					27974
+#define TM_REG_PF_ENABLE_CONN_RT_OFFSET					27975
+#define TM_REG_PF_ENABLE_TASK_RT_OFFSET					27976
+#define TM_REG_GROUP_SIZE_RESOLUTION_CONN_RT_OFFSET			27977
+#define TM_REG_GROUP_SIZE_RESOLUTION_TASK_RT_OFFSET			27978
+#define TM_REG_CONFIG_CONN_MEM_RT_OFFSET				27979
+#define TM_REG_CONFIG_CONN_MEM_RT_SIZE					416
+#define TM_REG_CONFIG_TASK_MEM_RT_OFFSET				28395
+#define TM_REG_CONFIG_TASK_MEM_RT_SIZE					512
+#define QM_REG_MAXPQSIZE_0_RT_OFFSET					28907
+#define QM_REG_MAXPQSIZE_1_RT_OFFSET					28908
+#define QM_REG_MAXPQSIZE_2_RT_OFFSET					28909
+#define QM_REG_MAXPQSIZETXSEL_0_RT_OFFSET				28910
+#define QM_REG_MAXPQSIZETXSEL_1_RT_OFFSET				28911
+#define QM_REG_MAXPQSIZETXSEL_2_RT_OFFSET				28912
+#define QM_REG_MAXPQSIZETXSEL_3_RT_OFFSET				28913
+#define QM_REG_MAXPQSIZETXSEL_4_RT_OFFSET				28914
+#define QM_REG_MAXPQSIZETXSEL_5_RT_OFFSET				28915
+#define QM_REG_MAXPQSIZETXSEL_6_RT_OFFSET				28916
+#define QM_REG_MAXPQSIZETXSEL_7_RT_OFFSET				28917
+#define QM_REG_MAXPQSIZETXSEL_8_RT_OFFSET				28918
+#define QM_REG_MAXPQSIZETXSEL_9_RT_OFFSET				28919
+#define QM_REG_MAXPQSIZETXSEL_10_RT_OFFSET				28920
+#define QM_REG_MAXPQSIZETXSEL_11_RT_OFFSET				28921
+#define QM_REG_MAXPQSIZETXSEL_12_RT_OFFSET				28922
+#define QM_REG_MAXPQSIZETXSEL_13_RT_OFFSET				28923
+#define QM_REG_MAXPQSIZETXSEL_14_RT_OFFSET				28924
+#define QM_REG_MAXPQSIZETXSEL_15_RT_OFFSET				28925
+#define QM_REG_MAXPQSIZETXSEL_16_RT_OFFSET				28926
+#define QM_REG_MAXPQSIZETXSEL_17_RT_OFFSET				28927
+#define QM_REG_MAXPQSIZETXSEL_18_RT_OFFSET				28928
+#define QM_REG_MAXPQSIZETXSEL_19_RT_OFFSET				28929
+#define QM_REG_MAXPQSIZETXSEL_20_RT_OFFSET				28930
+#define QM_REG_MAXPQSIZETXSEL_21_RT_OFFSET				28931
+#define QM_REG_MAXPQSIZETXSEL_22_RT_OFFSET				28932
+#define QM_REG_MAXPQSIZETXSEL_23_RT_OFFSET				28933
+#define QM_REG_MAXPQSIZETXSEL_24_RT_OFFSET				28934
+#define QM_REG_MAXPQSIZETXSEL_25_RT_OFFSET				28935
+#define QM_REG_MAXPQSIZETXSEL_26_RT_OFFSET				28936
+#define QM_REG_MAXPQSIZETXSEL_27_RT_OFFSET				28937
+#define QM_REG_MAXPQSIZETXSEL_28_RT_OFFSET				28938
+#define QM_REG_MAXPQSIZETXSEL_29_RT_OFFSET				28939
+#define QM_REG_MAXPQSIZETXSEL_30_RT_OFFSET				28940
+#define QM_REG_MAXPQSIZETXSEL_31_RT_OFFSET				28941
+#define QM_REG_MAXPQSIZETXSEL_32_RT_OFFSET				28942
+#define QM_REG_MAXPQSIZETXSEL_33_RT_OFFSET				28943
+#define QM_REG_MAXPQSIZETXSEL_34_RT_OFFSET				28944
+#define QM_REG_MAXPQSIZETXSEL_35_RT_OFFSET				28945
+#define QM_REG_MAXPQSIZETXSEL_36_RT_OFFSET				28946
+#define QM_REG_MAXPQSIZETXSEL_37_RT_OFFSET				28947
+#define QM_REG_MAXPQSIZETXSEL_38_RT_OFFSET				28948
+#define QM_REG_MAXPQSIZETXSEL_39_RT_OFFSET				28949
+#define QM_REG_MAXPQSIZETXSEL_40_RT_OFFSET				28950
+#define QM_REG_MAXPQSIZETXSEL_41_RT_OFFSET				28951
+#define QM_REG_MAXPQSIZETXSEL_42_RT_OFFSET				28952
+#define QM_REG_MAXPQSIZETXSEL_43_RT_OFFSET				28953
+#define QM_REG_MAXPQSIZETXSEL_44_RT_OFFSET				28954
+#define QM_REG_MAXPQSIZETXSEL_45_RT_OFFSET				28955
+#define QM_REG_MAXPQSIZETXSEL_46_RT_OFFSET				28956
+#define QM_REG_MAXPQSIZETXSEL_47_RT_OFFSET				28957
+#define QM_REG_MAXPQSIZETXSEL_48_RT_OFFSET				28958
+#define QM_REG_MAXPQSIZETXSEL_49_RT_OFFSET				28959
+#define QM_REG_MAXPQSIZETXSEL_50_RT_OFFSET				28960
+#define QM_REG_MAXPQSIZETXSEL_51_RT_OFFSET				28961
+#define QM_REG_MAXPQSIZETXSEL_52_RT_OFFSET				28962
+#define QM_REG_MAXPQSIZETXSEL_53_RT_OFFSET				28963
+#define QM_REG_MAXPQSIZETXSEL_54_RT_OFFSET				28964
+#define QM_REG_MAXPQSIZETXSEL_55_RT_OFFSET				28965
+#define QM_REG_MAXPQSIZETXSEL_56_RT_OFFSET				28966
+#define QM_REG_MAXPQSIZETXSEL_57_RT_OFFSET				28967
+#define QM_REG_MAXPQSIZETXSEL_58_RT_OFFSET				28968
+#define QM_REG_MAXPQSIZETXSEL_59_RT_OFFSET				28969
+#define QM_REG_MAXPQSIZETXSEL_60_RT_OFFSET				28970
+#define QM_REG_MAXPQSIZETXSEL_61_RT_OFFSET				28971
+#define QM_REG_MAXPQSIZETXSEL_62_RT_OFFSET				28972
+#define QM_REG_MAXPQSIZETXSEL_63_RT_OFFSET				28973
+#define QM_REG_BASEADDROTHERPQ_RT_OFFSET				28974
+#define QM_REG_BASEADDROTHERPQ_RT_SIZE					128
+#define QM_REG_PTRTBLOTHER_RT_OFFSET					29102
+#define QM_REG_PTRTBLOTHER_RT_SIZE					256
+#define QM_REG_VOQCRDLINE_RT_OFFSET					29358
+#define QM_REG_VOQCRDLINE_RT_SIZE					20
+#define QM_REG_VOQINITCRDLINE_RT_OFFSET					29378
+#define QM_REG_VOQINITCRDLINE_RT_SIZE					20
+#define QM_REG_AFULLQMBYPTHRPFWFQ_RT_OFFSET				29398
+#define QM_REG_AFULLQMBYPTHRVPWFQ_RT_OFFSET				29399
+#define QM_REG_AFULLQMBYPTHRPFRL_RT_OFFSET				29400
+#define QM_REG_AFULLQMBYPTHRGLBLRL_RT_OFFSET				29401
+#define QM_REG_AFULLOPRTNSTCCRDMASK_RT_OFFSET				29402
+#define QM_REG_WRROTHERPQGRP_0_RT_OFFSET				29403
+#define QM_REG_WRROTHERPQGRP_1_RT_OFFSET				29404
+#define QM_REG_WRROTHERPQGRP_2_RT_OFFSET				29405
+#define QM_REG_WRROTHERPQGRP_3_RT_OFFSET				29406
+#define QM_REG_WRROTHERPQGRP_4_RT_OFFSET				29407
+#define QM_REG_WRROTHERPQGRP_5_RT_OFFSET				29408
+#define QM_REG_WRROTHERPQGRP_6_RT_OFFSET				29409
+#define QM_REG_WRROTHERPQGRP_7_RT_OFFSET				29410
+#define QM_REG_WRROTHERPQGRP_8_RT_OFFSET				29411
+#define QM_REG_WRROTHERPQGRP_9_RT_OFFSET				29412
+#define QM_REG_WRROTHERPQGRP_10_RT_OFFSET				29413
+#define QM_REG_WRROTHERPQGRP_11_RT_OFFSET				29414
+#define QM_REG_WRROTHERPQGRP_12_RT_OFFSET				29415
+#define QM_REG_WRROTHERPQGRP_13_RT_OFFSET				29416
+#define QM_REG_WRROTHERPQGRP_14_RT_OFFSET				29417
+#define QM_REG_WRROTHERPQGRP_15_RT_OFFSET				29418
+#define QM_REG_WRROTHERGRPWEIGHT_0_RT_OFFSET				29419
+#define QM_REG_WRROTHERGRPWEIGHT_1_RT_OFFSET				29420
+#define QM_REG_WRROTHERGRPWEIGHT_2_RT_OFFSET				29421
+#define QM_REG_WRROTHERGRPWEIGHT_3_RT_OFFSET				29422
+#define QM_REG_WRRTXGRPWEIGHT_0_RT_OFFSET				29423
+#define QM_REG_WRRTXGRPWEIGHT_1_RT_OFFSET				29424
+#define QM_REG_PQTX2PF_0_RT_OFFSET					29425
+#define QM_REG_PQTX2PF_1_RT_OFFSET					29426
+#define QM_REG_PQTX2PF_2_RT_OFFSET					29427
+#define QM_REG_PQTX2PF_3_RT_OFFSET					29428
+#define QM_REG_PQTX2PF_4_RT_OFFSET					29429
+#define QM_REG_PQTX2PF_5_RT_OFFSET					29430
+#define QM_REG_PQTX2PF_6_RT_OFFSET					29431
+#define QM_REG_PQTX2PF_7_RT_OFFSET					29432
+#define QM_REG_PQTX2PF_8_RT_OFFSET					29433
+#define QM_REG_PQTX2PF_9_RT_OFFSET					29434
+#define QM_REG_PQTX2PF_10_RT_OFFSET					29435
+#define QM_REG_PQTX2PF_11_RT_OFFSET					29436
+#define QM_REG_PQTX2PF_12_RT_OFFSET					29437
+#define QM_REG_PQTX2PF_13_RT_OFFSET					29438
+#define QM_REG_PQTX2PF_14_RT_OFFSET					29439
+#define QM_REG_PQTX2PF_15_RT_OFFSET					29440
+#define QM_REG_PQTX2PF_16_RT_OFFSET					29441
+#define QM_REG_PQTX2PF_17_RT_OFFSET					29442
+#define QM_REG_PQTX2PF_18_RT_OFFSET					29443
+#define QM_REG_PQTX2PF_19_RT_OFFSET					29444
+#define QM_REG_PQTX2PF_20_RT_OFFSET					29445
+#define QM_REG_PQTX2PF_21_RT_OFFSET					29446
+#define QM_REG_PQTX2PF_22_RT_OFFSET					29447
+#define QM_REG_PQTX2PF_23_RT_OFFSET					29448
+#define QM_REG_PQTX2PF_24_RT_OFFSET					29449
+#define QM_REG_PQTX2PF_25_RT_OFFSET					29450
+#define QM_REG_PQTX2PF_26_RT_OFFSET					29451
+#define QM_REG_PQTX2PF_27_RT_OFFSET					29452
+#define QM_REG_PQTX2PF_28_RT_OFFSET					29453
+#define QM_REG_PQTX2PF_29_RT_OFFSET					29454
+#define QM_REG_PQTX2PF_30_RT_OFFSET					29455
+#define QM_REG_PQTX2PF_31_RT_OFFSET					29456
+#define QM_REG_PQTX2PF_32_RT_OFFSET					29457
+#define QM_REG_PQTX2PF_33_RT_OFFSET					29458
+#define QM_REG_PQTX2PF_34_RT_OFFSET					29459
+#define QM_REG_PQTX2PF_35_RT_OFFSET					29460
+#define QM_REG_PQTX2PF_36_RT_OFFSET					29461
+#define QM_REG_PQTX2PF_37_RT_OFFSET					29462
+#define QM_REG_PQTX2PF_38_RT_OFFSET					29463
+#define QM_REG_PQTX2PF_39_RT_OFFSET					29464
+#define QM_REG_PQTX2PF_40_RT_OFFSET					29465
+#define QM_REG_PQTX2PF_41_RT_OFFSET					29466
+#define QM_REG_PQTX2PF_42_RT_OFFSET					29467
+#define QM_REG_PQTX2PF_43_RT_OFFSET					29468
+#define QM_REG_PQTX2PF_44_RT_OFFSET					29469
+#define QM_REG_PQTX2PF_45_RT_OFFSET					29470
+#define QM_REG_PQTX2PF_46_RT_OFFSET					29471
+#define QM_REG_PQTX2PF_47_RT_OFFSET					29472
+#define QM_REG_PQTX2PF_48_RT_OFFSET					29473
+#define QM_REG_PQTX2PF_49_RT_OFFSET					29474
+#define QM_REG_PQTX2PF_50_RT_OFFSET					29475
+#define QM_REG_PQTX2PF_51_RT_OFFSET					29476
+#define QM_REG_PQTX2PF_52_RT_OFFSET					29477
+#define QM_REG_PQTX2PF_53_RT_OFFSET					29478
+#define QM_REG_PQTX2PF_54_RT_OFFSET					29479
+#define QM_REG_PQTX2PF_55_RT_OFFSET					29480
+#define QM_REG_PQTX2PF_56_RT_OFFSET					29481
+#define QM_REG_PQTX2PF_57_RT_OFFSET					29482
+#define QM_REG_PQTX2PF_58_RT_OFFSET					29483
+#define QM_REG_PQTX2PF_59_RT_OFFSET					29484
+#define QM_REG_PQTX2PF_60_RT_OFFSET					29485
+#define QM_REG_PQTX2PF_61_RT_OFFSET					29486
+#define QM_REG_PQTX2PF_62_RT_OFFSET					29487
+#define QM_REG_PQTX2PF_63_RT_OFFSET					29488
+#define QM_REG_PQOTHER2PF_0_RT_OFFSET					29489
+#define QM_REG_PQOTHER2PF_1_RT_OFFSET					29490
+#define QM_REG_PQOTHER2PF_2_RT_OFFSET					29491
+#define QM_REG_PQOTHER2PF_3_RT_OFFSET					29492
+#define QM_REG_PQOTHER2PF_4_RT_OFFSET					29493
+#define QM_REG_PQOTHER2PF_5_RT_OFFSET					29494
+#define QM_REG_PQOTHER2PF_6_RT_OFFSET					29495
+#define QM_REG_PQOTHER2PF_7_RT_OFFSET					29496
+#define QM_REG_PQOTHER2PF_8_RT_OFFSET					29497
+#define QM_REG_PQOTHER2PF_9_RT_OFFSET					29498
+#define QM_REG_PQOTHER2PF_10_RT_OFFSET					29499
+#define QM_REG_PQOTHER2PF_11_RT_OFFSET					29500
+#define QM_REG_PQOTHER2PF_12_RT_OFFSET					29501
+#define QM_REG_PQOTHER2PF_13_RT_OFFSET					29502
+#define QM_REG_PQOTHER2PF_14_RT_OFFSET					29503
+#define QM_REG_PQOTHER2PF_15_RT_OFFSET					29504
+#define QM_REG_RLGLBLPERIOD_0_RT_OFFSET					29505
+#define QM_REG_RLGLBLPERIOD_1_RT_OFFSET					29506
+#define QM_REG_RLGLBLPERIODTIMER_0_RT_OFFSET				29507
+#define QM_REG_RLGLBLPERIODTIMER_1_RT_OFFSET				29508
+#define QM_REG_RLGLBLPERIODSEL_0_RT_OFFSET				29509
+#define QM_REG_RLGLBLPERIODSEL_1_RT_OFFSET				29510
+#define QM_REG_RLGLBLPERIODSEL_2_RT_OFFSET				29511
+#define QM_REG_RLGLBLPERIODSEL_3_RT_OFFSET				29512
+#define QM_REG_RLGLBLPERIODSEL_4_RT_OFFSET				29513
+#define QM_REG_RLGLBLPERIODSEL_5_RT_OFFSET				29514
+#define QM_REG_RLGLBLPERIODSEL_6_RT_OFFSET				29515
+#define QM_REG_RLGLBLPERIODSEL_7_RT_OFFSET				29516
+#define QM_REG_RLGLBLINCVAL_RT_OFFSET					29517
+#define QM_REG_RLGLBLINCVAL_RT_SIZE					256
+#define QM_REG_RLGLBLUPPERBOUND_RT_OFFSET				29773
+#define QM_REG_RLGLBLUPPERBOUND_RT_SIZE					256
+#define QM_REG_RLGLBLCRD_RT_OFFSET					30029
+#define QM_REG_RLGLBLCRD_RT_SIZE					256
+#define QM_REG_RLGLBLENABLE_RT_OFFSET					30285
+#define QM_REG_RLPFPERIOD_RT_OFFSET					30286
+#define QM_REG_RLPFPERIODTIMER_RT_OFFSET				30287
+#define QM_REG_RLPFINCVAL_RT_OFFSET					30288
+#define QM_REG_RLPFINCVAL_RT_SIZE					16
+#define QM_REG_RLPFUPPERBOUND_RT_OFFSET					30304
+#define QM_REG_RLPFUPPERBOUND_RT_SIZE					16
+#define QM_REG_RLPFCRD_RT_OFFSET					30320
+#define QM_REG_RLPFCRD_RT_SIZE						16
+#define QM_REG_RLPFENABLE_RT_OFFSET					30336
+#define QM_REG_RLPFVOQENABLE_RT_OFFSET					30337
+#define QM_REG_WFQPFWEIGHT_RT_OFFSET					30338
+#define QM_REG_WFQPFWEIGHT_RT_SIZE					16
+#define QM_REG_WFQPFUPPERBOUND_RT_OFFSET				30354
+#define QM_REG_WFQPFUPPERBOUND_RT_SIZE					16
+#define QM_REG_WFQPFCRD_RT_OFFSET					30370
+#define QM_REG_WFQPFCRD_RT_SIZE						160
+#define QM_REG_WFQPFENABLE_RT_OFFSET					30530
+#define QM_REG_WFQVPENABLE_RT_OFFSET					30531
+#define QM_REG_BASEADDRTXPQ_RT_OFFSET					30532
+#define QM_REG_BASEADDRTXPQ_RT_SIZE					512
+#define QM_REG_TXPQMAP_RT_OFFSET					31044
+#define QM_REG_TXPQMAP_RT_SIZE						512
+#define QM_REG_WFQVPWEIGHT_RT_OFFSET					31556
+#define QM_REG_WFQVPWEIGHT_RT_SIZE					512
+#define QM_REG_WFQVPCRD_RT_OFFSET					32068
+#define QM_REG_WFQVPCRD_RT_SIZE						512
+#define QM_REG_WFQVPMAP_RT_OFFSET					32580
+#define QM_REG_WFQVPMAP_RT_SIZE						512
+#define QM_REG_PTRTBLTX_RT_OFFSET					33092
+#define QM_REG_PTRTBLTX_RT_SIZE						1024
+#define QM_REG_WFQPFCRD_MSB_RT_OFFSET					34116
+#define QM_REG_WFQPFCRD_MSB_RT_SIZE					160
+#define NIG_REG_TAG_ETHERTYPE_0_RT_OFFSET				34276
+#define NIG_REG_BRB_GATE_DNTFWD_PORT_RT_OFFSET				34277
+#define NIG_REG_OUTER_TAG_VALUE_LIST0_RT_OFFSET				34278
+#define NIG_REG_OUTER_TAG_VALUE_LIST1_RT_OFFSET				34279
+#define NIG_REG_OUTER_TAG_VALUE_LIST2_RT_OFFSET				34280
+#define NIG_REG_OUTER_TAG_VALUE_LIST3_RT_OFFSET				34281
+#define NIG_REG_LLH_FUNC_TAGMAC_CLS_TYPE_RT_OFFSET			34282
+#define NIG_REG_LLH_FUNC_TAG_EN_RT_OFFSET				34283
+#define NIG_REG_LLH_FUNC_TAG_EN_RT_SIZE					4
+#define NIG_REG_LLH_FUNC_TAG_VALUE_RT_OFFSET				34287
+#define NIG_REG_LLH_FUNC_TAG_VALUE_RT_SIZE				4
+#define NIG_REG_LLH_FUNC_FILTER_VALUE_RT_OFFSET				34291
+#define NIG_REG_LLH_FUNC_FILTER_VALUE_RT_SIZE				32
+#define NIG_REG_LLH_FUNC_FILTER_EN_RT_OFFSET				34323
+#define NIG_REG_LLH_FUNC_FILTER_EN_RT_SIZE				16
+#define NIG_REG_LLH_FUNC_FILTER_MODE_RT_OFFSET				34339
+#define NIG_REG_LLH_FUNC_FILTER_MODE_RT_SIZE				16
+#define NIG_REG_LLH_FUNC_FILTER_PROTOCOL_TYPE_RT_OFFSET			34355
+#define NIG_REG_LLH_FUNC_FILTER_PROTOCOL_TYPE_RT_SIZE			16
+#define NIG_REG_LLH_FUNC_FILTER_HDR_SEL_RT_OFFSET			34371
+#define NIG_REG_LLH_FUNC_FILTER_HDR_SEL_RT_SIZE				16
+#define NIG_REG_TX_EDPM_CTRL_RT_OFFSET					34387
+#define NIG_REG_PPF_TO_ENGINE_SEL_RT_OFFSET				34388
+#define NIG_REG_PPF_TO_ENGINE_SEL_RT_SIZE				8
+#define CDU_REG_CID_ADDR_PARAMS_RT_OFFSET				34396
+#define CDU_REG_SEGMENT0_PARAMS_RT_OFFSET				34397
+#define CDU_REG_SEGMENT1_PARAMS_RT_OFFSET				34398
+#define CDU_REG_PF_SEG0_TYPE_OFFSET_RT_OFFSET				34399
+#define CDU_REG_PF_SEG1_TYPE_OFFSET_RT_OFFSET				34400
+#define CDU_REG_PF_SEG2_TYPE_OFFSET_RT_OFFSET				34401
+#define CDU_REG_PF_SEG3_TYPE_OFFSET_RT_OFFSET				34402
+#define CDU_REG_PF_FL_SEG0_TYPE_OFFSET_RT_OFFSET			34403
+#define CDU_REG_PF_FL_SEG1_TYPE_OFFSET_RT_OFFSET			34404
+#define CDU_REG_PF_FL_SEG2_TYPE_OFFSET_RT_OFFSET			34405
+#define CDU_REG_PF_FL_SEG3_TYPE_OFFSET_RT_OFFSET			34406
+#define CDU_REG_VF_SEG_TYPE_OFFSET_RT_OFFSET				34407
+#define CDU_REG_VF_FL_SEG_TYPE_OFFSET_RT_OFFSET				34408
+#define PBF_REG_TAG_ETHERTYPE_0_RT_OFFSET				34409
+#define PBF_REG_BTB_SHARED_AREA_SIZE_RT_OFFSET				34410
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ0_RT_OFFSET			34411
+#define PBF_REG_BTB_GUARANTEED_VOQ0_RT_OFFSET				34412
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ0_RT_OFFSET			34413
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ1_RT_OFFSET			34414
+#define PBF_REG_BTB_GUARANTEED_VOQ1_RT_OFFSET				34415
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ1_RT_OFFSET			34416
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ2_RT_OFFSET			34417
+#define PBF_REG_BTB_GUARANTEED_VOQ2_RT_OFFSET				34418
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ2_RT_OFFSET			34419
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ3_RT_OFFSET			34420
+#define PBF_REG_BTB_GUARANTEED_VOQ3_RT_OFFSET				34421
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ3_RT_OFFSET			34422
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ4_RT_OFFSET			34423
+#define PBF_REG_BTB_GUARANTEED_VOQ4_RT_OFFSET				34424
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ4_RT_OFFSET			34425
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ5_RT_OFFSET			34426
+#define PBF_REG_BTB_GUARANTEED_VOQ5_RT_OFFSET				34427
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ5_RT_OFFSET			34428
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ6_RT_OFFSET			34429
+#define PBF_REG_BTB_GUARANTEED_VOQ6_RT_OFFSET				34430
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ6_RT_OFFSET			34431
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ7_RT_OFFSET			34432
+#define PBF_REG_BTB_GUARANTEED_VOQ7_RT_OFFSET				34433
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ7_RT_OFFSET			34434
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ8_RT_OFFSET			34435
+#define PBF_REG_BTB_GUARANTEED_VOQ8_RT_OFFSET				34436
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ8_RT_OFFSET			34437
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ9_RT_OFFSET			34438
+#define PBF_REG_BTB_GUARANTEED_VOQ9_RT_OFFSET				34439
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ9_RT_OFFSET			34440
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ10_RT_OFFSET			34441
+#define PBF_REG_BTB_GUARANTEED_VOQ10_RT_OFFSET				34442
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ10_RT_OFFSET			34443
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ11_RT_OFFSET			34444
+#define PBF_REG_BTB_GUARANTEED_VOQ11_RT_OFFSET				34445
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ11_RT_OFFSET			34446
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ12_RT_OFFSET			34447
+#define PBF_REG_BTB_GUARANTEED_VOQ12_RT_OFFSET				34448
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ12_RT_OFFSET			34449
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ13_RT_OFFSET			34450
+#define PBF_REG_BTB_GUARANTEED_VOQ13_RT_OFFSET				34451
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ13_RT_OFFSET			34452
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ14_RT_OFFSET			34453
+#define PBF_REG_BTB_GUARANTEED_VOQ14_RT_OFFSET				34454
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ14_RT_OFFSET			34455
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ15_RT_OFFSET			34456
+#define PBF_REG_BTB_GUARANTEED_VOQ15_RT_OFFSET				34457
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ15_RT_OFFSET			34458
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ16_RT_OFFSET			34459
+#define PBF_REG_BTB_GUARANTEED_VOQ16_RT_OFFSET				34460
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ16_RT_OFFSET			34461
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ17_RT_OFFSET			34462
+#define PBF_REG_BTB_GUARANTEED_VOQ17_RT_OFFSET				34463
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ17_RT_OFFSET			34464
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ18_RT_OFFSET			34465
+#define PBF_REG_BTB_GUARANTEED_VOQ18_RT_OFFSET				34466
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ18_RT_OFFSET			34467
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ19_RT_OFFSET			34468
+#define PBF_REG_BTB_GUARANTEED_VOQ19_RT_OFFSET				34469
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ19_RT_OFFSET			34470
+#define XCM_REG_CON_PHY_Q3_RT_OFFSET					34471
+
+#define RUNTIME_ARRAY_SIZE 34472
 
 /* Init Callbacks */
 #define DMAE_READY_CB	0
diff --git a/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c b/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c
index d6430dfebd83..2307f8842c9e 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c
@@ -44,9 +44,9 @@
 #define CDU_VALIDATION_DEFAULT_CFG	61
 
 static u16 con_region_offsets[3][NUM_OF_CONNECTION_TYPES_E4] = {
-	{400, 336, 352, 304, 304, 384, 416, 352},	/* region 3 offsets */
-	{528, 496, 416, 448, 448, 512, 544, 480},	/* region 4 offsets */
-	{608, 544, 496, 512, 576, 592, 624, 560}	/* region 5 offsets */
+	{400, 336, 352, 368, 304, 384, 416, 352},	/* region 3 offsets */
+	{528, 496, 416, 512, 448, 512, 544, 480},	/* region 4 offsets */
+	{608, 544, 496, 576, 576, 592, 624, 560}	/* region 5 offsets */
 };
 
 static u16 task_region_offsets[1][NUM_OF_CONNECTION_TYPES_E4] = {
@@ -228,9 +228,6 @@ static void qed_enable_pf_rl(struct qed_hwfn *p_hwfn, bool pf_rl_en)
 		STORE_RT_REG(p_hwfn,
 			     QM_REG_RLPFVOQENABLE_RT_OFFSET,
 			     (u32)voq_bit_mask);
-		if (num_ext_voqs >= 32)
-			STORE_RT_REG(p_hwfn, QM_REG_RLPFVOQENABLE_MSB_RT_OFFSET,
-				     (u32)(voq_bit_mask >> 32));
 
 		/* Write RL period */
 		STORE_RT_REG(p_hwfn,
diff --git a/drivers/net/ethernet/qlogic/qed/qed_init_ops.c b/drivers/net/ethernet/qlogic/qed/qed_init_ops.c
index a868d7f88601..ef22a3596295 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_init_ops.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_init_ops.c
@@ -54,15 +54,15 @@ static u32 pxp_global_win[] = {
 	0x1c80, /* win 3: addr=0x1c80000, size=4096 bytes */
 	0x1d00, /* win 4: addr=0x1d00000, size=4096 bytes */
 	0x1d01, /* win 5: addr=0x1d01000, size=4096 bytes */
-	0x1d80, /* win 6: addr=0x1d80000, size=4096 bytes */
-	0x1d81, /* win 7: addr=0x1d81000, size=4096 bytes */
-	0x1d82, /* win 8: addr=0x1d82000, size=4096 bytes */
-	0x1e00, /* win 9: addr=0x1e00000, size=4096 bytes */
-	0x1e80, /* win 10: addr=0x1e80000, size=4096 bytes */
-	0x1f00, /* win 11: addr=0x1f00000, size=4096 bytes */
-	0,
-	0,
-	0,
+	0x1d02, /* win 6: addr=0x1d02000, size=4096 bytes */
+	0x1d80, /* win 7: addr=0x1d80000, size=4096 bytes */
+	0x1d81, /* win 8: addr=0x1d81000, size=4096 bytes */
+	0x1d82, /* win 9: addr=0x1d82000, size=4096 bytes */
+	0x1e00, /* win 10: addr=0x1e00000, size=4096 bytes */
+	0x1e01, /* win 11: addr=0x1e01000, size=4096 bytes */
+	0x1e80, /* win 12: addr=0x1e80000, size=4096 bytes */
+	0x1f00, /* win 13: addr=0x1f00000, size=4096 bytes */
+	0x1c08, /* win 14: addr=0x1c08000, size=4096 bytes */
 	0,
 	0,
 	0,
diff --git a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h
index 60f850c3bdd6..3dcb6ff58e73 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h
@@ -178,6 +178,8 @@
 	0x008c80UL
 #define  MCP_REG_SCRATCH	\
 	0xe20000UL
+#define MCP_REG_SCRATCH_SIZE \
+	57344
 #define  CNIG_REG_NW_PORT_MODE_BB \
 	0x218200UL
 #define  MISCS_REG_CHIP_NUM \
@@ -212,6 +214,8 @@
 	0x580900UL
 #define  DBG_REG_CLIENT_ENABLE \
 	0x010004UL
+#define DBG_REG_TIMESTAMP_VALID_EN \
+	0x010b58UL
 #define  DMAE_REG_INIT \
 	0x00c000UL
 #define  DORQ_REG_IFEN \
@@ -350,6 +354,10 @@
 	0x24000cUL
 #define PSWRQ2_REG_ILT_MEMORY \
 	0x260000UL
+#define PSWRQ2_REG_ILT_MEMORY_SIZE_BB \
+	15200
+#define PSWRQ2_REG_ILT_MEMORY_SIZE_K2 \
+	22000
 #define  PSWHST_REG_DISCARD_INTERNAL_WRITES \
 	0x2a0040UL
 #define  PSWHST2_REG_DBGSYN_ALMOST_FULL_THR \
@@ -1453,6 +1461,8 @@
 	0x1401404UL
 #define XSEM_REG_DBG_FRAME_MODE_BB_K2	\
 	0x1401408UL
+#define XSEM_REG_DBG_GPRE_VECT \
+	0x1401410UL
 #define XSEM_REG_DBG_MODE1_CFG_BB_K2 \
 	0x1401420UL
 #define XSEM_REG_FAST_MEMORY \
@@ -1465,6 +1475,8 @@
 	0x1501404UL
 #define YSEM_REG_DBG_FRAME_MODE_BB_K2	\
 	0x1501408UL
+#define YSEM_REG_DBG_GPRE_VECT \
+	0x1501410UL
 #define YSEM_REG_DBG_MODE1_CFG_BB_K2 \
 	0x1501420UL
 #define YSEM_REG_FAST_MEMORY \
@@ -1479,6 +1491,8 @@
 	0x1601404UL
 #define PSEM_REG_DBG_FRAME_MODE_BB_K2	\
 	0x1601408UL
+#define PSEM_REG_DBG_GPRE_VECT \
+	0x1601410UL
 #define PSEM_REG_DBG_MODE1_CFG_BB_K2 \
 	0x1601420UL
 #define PSEM_REG_FAST_MEMORY \
@@ -1493,6 +1507,8 @@
 	0x1701404UL
 #define TSEM_REG_DBG_FRAME_MODE_BB_K2	\
 	0x1701408UL
+#define TSEM_REG_DBG_GPRE_VECT \
+	0x1701410UL
 #define TSEM_REG_DBG_MODE1_CFG_BB_K2 \
 	0x1701420UL
 #define TSEM_REG_FAST_MEMORY \
@@ -1507,12 +1523,16 @@
 	0x1801404UL
 #define MSEM_REG_DBG_FRAME_MODE_BB_K2	\
 	0x1801408UL
+#define MSEM_REG_DBG_GPRE_VECT \
+	0x1801410UL
 #define MSEM_REG_DBG_MODE1_CFG_BB_K2 \
 	0x1801420UL
 #define MSEM_REG_FAST_MEMORY \
 	0x1840000UL
 #define USEM_REG_SLOW_DBG_EMPTY_BB_K2	\
 	0x1901140UL
+#define SEM_FAST_REG_INT_RAM_SIZE \
+	20480
 #define USEM_REG_SYNC_DBG_EMPTY	\
 	0x1901160UL
 #define USEM_REG_SLOW_DBG_ACTIVE_BB_K2 \
@@ -1521,14 +1541,26 @@
 	0x1901404UL
 #define USEM_REG_DBG_FRAME_MODE_BB_K2	\
 	0x1901408UL
+#define USEM_REG_DBG_GPRE_VECT \
+	0x1901410UL
 #define USEM_REG_DBG_MODE1_CFG_BB_K2 \
 	0x1901420UL
 #define USEM_REG_FAST_MEMORY \
 	0x1940000UL
+#define SEM_FAST_REG_DBG_MODE23_SRC_DISABLE \
+	0x000748UL
+#define SEM_FAST_REG_DBG_MODE4_SRC_DISABLE \
+	0x00074cUL
+#define SEM_FAST_REG_DBG_MODE6_SRC_DISABLE \
+	0x000750UL
+#define SEM_FAST_REG_DEBUG_ACTIVE \
+	0x000740UL
 #define SEM_FAST_REG_INT_RAM \
 	0x020000UL
 #define SEM_FAST_REG_INT_RAM_SIZE_BB_K2 \
 	20480
+#define SEM_FAST_REG_RECORD_FILTER_ENABLE \
+	0x000768UL
 #define GRC_REG_TRACE_FIFO_VALID_DATA \
 	0x050064UL
 #define GRC_REG_NUMBER_VALID_OVERRIDE_WINDOW \
@@ -1583,14 +1615,20 @@
 	0x181530UL
 #define DBG_REG_DBG_BLOCK_ON \
 	0x010454UL
+#define DBG_REG_FILTER_ENABLE \
+	0x0109d0UL
 #define DBG_REG_FRAMING_MODE \
 	0x010058UL
+#define DBG_REG_TRIGGER_ENABLE \
+	0x01054cUL
 #define SEM_FAST_REG_VFC_DATA_WR \
 	0x000b40UL
 #define SEM_FAST_REG_VFC_ADDR \
 	0x000b44UL
 #define SEM_FAST_REG_VFC_DATA_RD \
 	0x000b48UL
+#define SEM_FAST_REG_VFC_STATUS	\
+	0x000b4cUL
 #define RSS_REG_RSS_RAM_DATA \
 	0x238c20UL
 #define RSS_REG_RSS_RAM_DATA_SIZE \
-- 
cgit v1.2.3-59-g8ed1b


From 92fae6fb231f66ce0e71ed348e788180c1a603f9 Mon Sep 17 00:00:00 2001
From: Michal Kalderon <michal.kalderon@marvell.com>
Date: Mon, 27 Jan 2020 15:26:09 +0200
Subject: qed: FW 8.42.2.0 Queue Manager changes

This patch contains changes in initialization and usage of the QM blocks.
Instead of setting a rate limiter per vport the rate limiters are now a
global resource and set independentaly.

The patch also contains a field name change:
vport_wfq which is part of vport_params was renamed to wfq as the vport
prefix is redundant.

Signed-off-by: Ariel Elior <ariel.elior@marvell.com>
Signed-off-by: Michal Kalderon <michal.kalderon@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/qlogic/qed/qed_cxt.c          |   1 -
 drivers/net/ethernet/qlogic/qed/qed_dev.c          |  21 +-
 drivers/net/ethernet/qlogic/qed/qed_hsi.h          |  41 ++--
 .../net/ethernet/qlogic/qed/qed_init_fw_funcs.c    | 238 ++++++++++-----------
 drivers/net/ethernet/qlogic/qed/qed_sriov.c        |   8 +-
 5 files changed, 149 insertions(+), 160 deletions(-)

diff --git a/drivers/net/ethernet/qlogic/qed/qed_cxt.c b/drivers/net/ethernet/qlogic/qed/qed_cxt.c
index 8e1bdf58b9e7..5ebafbc379db 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_cxt.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_cxt.c
@@ -1522,7 +1522,6 @@ void qed_qm_init_pf(struct qed_hwfn *p_hwfn,
 	params.num_vports = qm_info->num_vports;
 	params.pf_wfq = qm_info->pf_wfq;
 	params.pf_rl = qm_info->pf_rl;
-	params.link_speed = p_link->speed;
 	params.pq_params = qm_info->qm_pq_params;
 	params.vport_params = qm_info->qm_vport_params;
 
diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c
index a1ebc2b1ca0b..249fcc3dc138 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_dev.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c
@@ -1571,7 +1571,7 @@ static void qed_init_qm_vport_params(struct qed_hwfn *p_hwfn)
 
 	/* all vports participate in weighted fair queueing */
 	for (i = 0; i < qed_init_qm_get_num_vports(p_hwfn); i++)
-		qm_info->qm_vport_params[i].vport_wfq = 1;
+		qm_info->qm_vport_params[i].wfq = 1;
 }
 
 /* initialize qm port params */
@@ -2034,9 +2034,8 @@ static void qed_dp_init_qm_params(struct qed_hwfn *p_hwfn)
 		vport = &(qm_info->qm_vport_params[i]);
 		DP_VERBOSE(p_hwfn,
 			   NETIF_MSG_HW,
-			   "vport idx %d, vport_rl %d, wfq %d, first_tx_pq_id [ ",
-			   qm_info->start_vport + i,
-			   vport->vport_rl, vport->vport_wfq);
+			   "vport idx %d, wfq %d, first_tx_pq_id [ ",
+			   qm_info->start_vport + i, vport->wfq);
 		for (tc = 0; tc < NUM_OF_TCS; tc++)
 			DP_VERBOSE(p_hwfn,
 				   NETIF_MSG_HW,
@@ -2049,11 +2048,11 @@ static void qed_dp_init_qm_params(struct qed_hwfn *p_hwfn)
 		pq = &(qm_info->qm_pq_params[i]);
 		DP_VERBOSE(p_hwfn,
 			   NETIF_MSG_HW,
-			   "pq idx %d, port %d, vport_id %d, tc %d, wrr_grp %d, rl_valid %d\n",
+			   "pq idx %d, port %d, vport_id %d, tc %d, wrr_grp %d, rl_valid %d rl_id %d\n",
 			   qm_info->start_pq + i,
 			   pq->port_id,
 			   pq->vport_id,
-			   pq->tc_id, pq->wrr_group, pq->rl_valid);
+			   pq->tc_id, pq->wrr_group, pq->rl_valid, pq->rl_id);
 	}
 }
 
@@ -2623,7 +2622,7 @@ static int qed_hw_init_common(struct qed_hwfn *p_hwfn,
 	params.max_phys_tcs_per_port = qm_info->max_phys_tcs_per_port;
 	params.pf_rl_en = qm_info->pf_rl_en;
 	params.pf_wfq_en = qm_info->pf_wfq_en;
-	params.vport_rl_en = qm_info->vport_rl_en;
+	params.global_rl_en = qm_info->vport_rl_en;
 	params.vport_wfq_en = qm_info->vport_wfq_en;
 	params.port_params = qm_info->qm_port_params;
 
@@ -5087,11 +5086,11 @@ static void qed_configure_wfq_for_all_vports(struct qed_hwfn *p_hwfn,
 	for (i = 0; i < p_hwfn->qm_info.num_vports; i++) {
 		u32 wfq_speed = p_hwfn->qm_info.wfq_data[i].min_speed;
 
-		vport_params[i].vport_wfq = (wfq_speed * QED_WFQ_UNIT) /
+		vport_params[i].wfq = (wfq_speed * QED_WFQ_UNIT) /
 						min_pf_rate;
 		qed_init_vport_wfq(p_hwfn, p_ptt,
 				   vport_params[i].first_tx_pq_id,
-				   vport_params[i].vport_wfq);
+				   vport_params[i].wfq);
 	}
 }
 
@@ -5102,7 +5101,7 @@ static void qed_init_wfq_default_param(struct qed_hwfn *p_hwfn,
 	int i;
 
 	for (i = 0; i < p_hwfn->qm_info.num_vports; i++)
-		p_hwfn->qm_info.qm_vport_params[i].vport_wfq = 1;
+		p_hwfn->qm_info.qm_vport_params[i].wfq = 1;
 }
 
 static void qed_disable_wfq_for_all_vports(struct qed_hwfn *p_hwfn,
@@ -5118,7 +5117,7 @@ static void qed_disable_wfq_for_all_vports(struct qed_hwfn *p_hwfn,
 		qed_init_wfq_default_param(p_hwfn, min_pf_rate);
 		qed_init_vport_wfq(p_hwfn, p_ptt,
 				   vport_params[i].first_tx_pq_id,
-				   vport_params[i].vport_wfq);
+				   vport_params[i].wfq);
 	}
 }
 
diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
index efabe0ca5cbe..5ac68ec01a04 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
@@ -2630,13 +2630,18 @@ struct init_nig_pri_tc_map_req {
 	struct init_nig_pri_tc_map_entry pri[NUM_OF_VLAN_PRIORITIES];
 };
 
+/* QM per global RL init parameters */
+struct init_qm_global_rl_params {
+	u32 rate_limit;
+};
+
 /* QM per-port init parameters */
 struct init_qm_port_params {
-	u8 active;
-	u8 active_phys_tcs;
+	u16 active_phys_tcs;
 	u16 num_pbf_cmd_lines;
 	u16 num_btb_blocks;
-	u16 reserved;
+	u8 active;
+	u8 reserved;
 };
 
 /* QM per-PQ init parameters */
@@ -2645,15 +2650,14 @@ struct init_qm_pq_params {
 	u8 tc_id;
 	u8 wrr_group;
 	u8 rl_valid;
+	u16 rl_id;
 	u8 port_id;
-	u8 reserved0;
-	u16 reserved1;
+	u8 reserved;
 };
 
 /* QM per-vport init parameters */
 struct init_qm_vport_params {
-	u32 vport_rl;
-	u16 vport_wfq;
+	u16 wfq;
 	u16 first_tx_pq_id[NUM_OF_TCS];
 };
 
@@ -3982,7 +3986,7 @@ struct qed_qm_common_rt_init_params {
 	u8 max_phys_tcs_per_port;
 	bool pf_rl_en;
 	bool pf_wfq_en;
-	bool vport_rl_en;
+	bool global_rl_en;
 	bool vport_wfq_en;
 	struct init_qm_port_params *port_params;
 };
@@ -4001,11 +4005,10 @@ struct qed_qm_pf_rt_init_params {
 	u16 start_pq;
 	u16 num_pf_pqs;
 	u16 num_vf_pqs;
-	u8 start_vport;
-	u8 num_vports;
+	u16 start_vport;
+	u16 num_vports;
 	u16 pf_wfq;
 	u32 pf_rl;
-	u32 link_speed;
 	struct init_qm_pq_params *pq_params;
 	struct init_qm_vport_params *vport_params;
 };
@@ -4054,22 +4057,22 @@ int qed_init_pf_rl(struct qed_hwfn *p_hwfn,
  */
 int qed_init_vport_wfq(struct qed_hwfn *p_hwfn,
 		       struct qed_ptt *p_ptt,
-		       u16 first_tx_pq_id[NUM_OF_TCS], u16 vport_wfq);
+		       u16 first_tx_pq_id[NUM_OF_TCS], u16 wfq);
 
 /**
- * @brief qed_init_vport_rl - Initializes the rate limit of the specified VPORT
+ * @brief qed_init_global_rl - Initializes the rate limit of the specified
+ * rate limiter
  *
  * @param p_hwfn
  * @param p_ptt - ptt window used for writing the registers
- * @param vport_id - VPORT ID
- * @param vport_rl - rate limit in Mb/sec units
- * @param link_speed - link speed in Mbps.
+ * @param rl_id - RL ID
+ * @param rate_limit - rate limit in Mb/sec units
  *
  * @return 0 on success, -1 on error.
  */
-int qed_init_vport_rl(struct qed_hwfn *p_hwfn,
-		      struct qed_ptt *p_ptt,
-		      u8 vport_id, u32 vport_rl, u32 link_speed);
+int qed_init_global_rl(struct qed_hwfn *p_hwfn,
+		       struct qed_ptt *p_ptt,
+		       u16 rl_id, u32 rate_limit);
 
 /**
  * @brief qed_send_qm_stop_cmd  Sends a stop command to the QM
diff --git a/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c b/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c
index 2307f8842c9e..965529b1a9c5 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c
@@ -61,6 +61,9 @@ static u16 task_region_offsets[1][NUM_OF_CONNECTION_TYPES_E4] = {
 								0x100) - 1 : 0)
 #define QM_INVALID_PQ_ID		0xffff
 
+/* Max link speed (in Mbps) */
+#define QM_MAX_LINK_SPEED               100000
+
 /* Feature enable */
 #define QM_BYPASS_EN	1
 #define QM_BYTE_CRD_EN	1
@@ -128,8 +131,6 @@ static u16 task_region_offsets[1][NUM_OF_CONNECTION_TYPES_E4] = {
 /* Pure LB CmdQ lines (+spare) */
 #define PBF_CMDQ_PURE_LB_LINES	150
 
-#define PBF_CMDQ_LINES_E5_RSVD_RATIO	8
-
 #define PBF_CMDQ_LINES_RT_OFFSET(ext_voq) \
 	(PBF_REG_YCMD_QS_NUM_LINES_VOQ0_RT_OFFSET + \
 	 (ext_voq) * (PBF_REG_YCMD_QS_NUM_LINES_VOQ1_RT_OFFSET - \
@@ -140,6 +141,9 @@ static u16 task_region_offsets[1][NUM_OF_CONNECTION_TYPES_E4] = {
 	 (ext_voq) * (PBF_REG_BTB_GUARANTEED_VOQ1_RT_OFFSET - \
 		PBF_REG_BTB_GUARANTEED_VOQ0_RT_OFFSET))
 
+/* Returns the VOQ line credit for the specified number of PBF command lines.
+ * PBF lines are specified in 256b units.
+ */
 #define QM_VOQ_LINE_CRD(pbf_cmd_lines) \
 	((((pbf_cmd_lines) - 4) * 2) | QM_LINE_CRD_REG_SIGN_BIT)
 
@@ -178,14 +182,14 @@ static u16 task_region_offsets[1][NUM_OF_CONNECTION_TYPES_E4] = {
 		  cmd ## _ ## field, \
 		  value)
 
-#define QM_INIT_TX_PQ_MAP(p_hwfn, map, chip, pq_id, rl_valid, vp_pq_id, rl_id, \
+#define QM_INIT_TX_PQ_MAP(p_hwfn, map, chip, pq_id, vp_pq_id, rl_valid, rl_id, \
 			  ext_voq, wrr) \
 	do { \
 		typeof(map) __map; \
 		memset(&__map, 0, sizeof(__map)); \
 		SET_FIELD(__map.reg, QM_RF_PQ_MAP_ ## chip ## _PQ_VALID, 1); \
 		SET_FIELD(__map.reg, QM_RF_PQ_MAP_ ## chip ## _RL_VALID, \
-			  rl_valid); \
+			  rl_valid ? 1 : 0);\
 		SET_FIELD(__map.reg, QM_RF_PQ_MAP_ ## chip ## _VP_PQ_ID, \
 			  vp_pq_id); \
 		SET_FIELD(__map.reg, QM_RF_PQ_MAP_ ## chip ## _RL_ID, rl_id); \
@@ -200,9 +204,12 @@ static u16 task_region_offsets[1][NUM_OF_CONNECTION_TYPES_E4] = {
 #define WRITE_PQ_INFO_TO_RAM	1
 #define PQ_INFO_ELEMENT(vp, pf, tc, port, rl_valid, rl) \
 	(((vp) << 0) | ((pf) << 12) | ((tc) << 16) | ((port) << 20) | \
-	((rl_valid) << 22) | ((rl) << 24))
+	((rl_valid ? 1 : 0) << 22) | (((rl) & 255) << 24) | \
+	(((rl) >> 8) << 9))
+
 #define PQ_INFO_RAM_GRC_ADDRESS(pq_id) \
-	(XSEM_REG_FAST_MEMORY + SEM_FAST_REG_INT_RAM + 21776 + (pq_id) * 4)
+	XSEM_REG_FAST_MEMORY + SEM_FAST_REG_INT_RAM + \
+	XSTORM_PQ_INFO_OFFSET(pq_id)
 
 /******************** INTERNAL IMPLEMENTATION *********************/
 
@@ -256,12 +263,12 @@ static void qed_enable_pf_wfq(struct qed_hwfn *p_hwfn, bool pf_wfq_en)
 			     QM_WFQ_UPPER_BOUND);
 }
 
-/* Prepare VPORT RL enable/disable runtime init values */
-static void qed_enable_vport_rl(struct qed_hwfn *p_hwfn, bool vport_rl_en)
+/* Prepare global RL enable/disable runtime init values */
+static void qed_enable_global_rl(struct qed_hwfn *p_hwfn, bool global_rl_en)
 {
 	STORE_RT_REG(p_hwfn, QM_REG_RLGLBLENABLE_RT_OFFSET,
-		     vport_rl_en ? 1 : 0);
-	if (vport_rl_en) {
+		     global_rl_en ? 1 : 0);
+	if (global_rl_en) {
 		/* Write RL period (use timer 0 only) */
 		STORE_RT_REG(p_hwfn,
 			     QM_REG_RLGLBLPERIOD_0_RT_OFFSET,
@@ -328,8 +335,7 @@ static void qed_cmdq_lines_rt_init(
 			continue;
 
 		/* Find number of command queue lines to divide between the
-		 * active physical TCs. In E5, 1/8 of the lines are reserved.
-		 * the lines for pure LB TC are subtracted.
+		 * active physical TCs.
 		 */
 		phys_lines = port_params[port_id].num_pbf_cmd_lines;
 		phys_lines -= PBF_CMDQ_PURE_LB_LINES;
@@ -358,11 +364,30 @@ static void qed_cmdq_lines_rt_init(
 		ext_voq = qed_get_ext_voq(p_hwfn,
 					  port_id,
 					  PURE_LB_TC, max_phys_tcs_per_port);
-		qed_cmdq_lines_voq_rt_init(p_hwfn,
-					   ext_voq, PBF_CMDQ_PURE_LB_LINES);
+		qed_cmdq_lines_voq_rt_init(p_hwfn, ext_voq,
+					   PBF_CMDQ_PURE_LB_LINES);
 	}
 }
 
+/* Prepare runtime init values to allocate guaranteed BTB blocks for the
+ * specified port. The guaranteed BTB space is divided between the TCs as
+ * follows (shared space Is currently not used):
+ * 1. Parameters:
+ *    B - BTB blocks for this port
+ *    C - Number of physical TCs for this port
+ * 2. Calculation:
+ *    a. 38 blocks (9700B jumbo frame) are allocated for global per port
+ *	 headroom.
+ *    b. B = B - 38 (remainder after global headroom allocation).
+ *    c. MAX(38,B/(C+0.7)) blocks are allocated for the pure LB VOQ.
+ *    d. B = B - MAX(38, B/(C+0.7)) (remainder after pure LB allocation).
+ *    e. B/C blocks are allocated for each physical TC.
+ * Assumptions:
+ * - MTU is up to 9700 bytes (38 blocks)
+ * - All TCs are considered symmetrical (same rate and packet size)
+ * - No optimization for lossy TC (all are considered lossless). Shared space
+ *   is not enabled and allocated for each TC.
+ */
 static void qed_btb_blocks_rt_init(
 	struct qed_hwfn *p_hwfn,
 	u8 max_ports_per_engine,
@@ -421,6 +446,34 @@ static void qed_btb_blocks_rt_init(
 	}
 }
 
+/* Prepare runtime init values for the specified RL.
+ * Set max link speed (100Gbps) per rate limiter.
+ * Return -1 on error.
+ */
+static int qed_global_rl_rt_init(struct qed_hwfn *p_hwfn)
+{
+	u32 upper_bound = QM_VP_RL_UPPER_BOUND(QM_MAX_LINK_SPEED) |
+			  (u32)QM_RL_CRD_REG_SIGN_BIT;
+	u32 inc_val;
+	u16 rl_id;
+
+	/* Go over all global RLs */
+	for (rl_id = 0; rl_id < MAX_QM_GLOBAL_RLS; rl_id++) {
+		inc_val = QM_RL_INC_VAL(QM_MAX_LINK_SPEED);
+
+		STORE_RT_REG(p_hwfn,
+			     QM_REG_RLGLBLCRD_RT_OFFSET + rl_id,
+			     (u32)QM_RL_CRD_REG_SIGN_BIT);
+		STORE_RT_REG(p_hwfn,
+			     QM_REG_RLGLBLUPPERBOUND_RT_OFFSET + rl_id,
+			     upper_bound);
+		STORE_RT_REG(p_hwfn,
+			     QM_REG_RLGLBLINCVAL_RT_OFFSET + rl_id, inc_val);
+	}
+
+	return 0;
+}
+
 /* Prepare Tx PQ mapping runtime init values for the specified PF */
 static void qed_tx_pq_map_rt_init(struct qed_hwfn *p_hwfn,
 				  struct qed_ptt *p_ptt,
@@ -457,18 +510,17 @@ static void qed_tx_pq_map_rt_init(struct qed_hwfn *p_hwfn,
 
 	/* Go over all Tx PQs */
 	for (i = 0, pq_id = p_params->start_pq; i < num_pqs; i++, pq_id++) {
-		u8 ext_voq, vport_id_in_pf, tc_id = pq_params[i].tc_id;
-		u32 max_qm_global_rls = MAX_QM_GLOBAL_RLS;
+		u16 *p_first_tx_pq_id, vport_id_in_pf;
 		struct qm_rf_pq_map_e4 tx_pq_map;
-		bool is_vf_pq, rl_valid;
-		u16 *p_first_tx_pq_id;
+		u8 tc_id = pq_params[i].tc_id;
+		bool is_vf_pq;
+		u8 ext_voq;
 
 		ext_voq = qed_get_ext_voq(p_hwfn,
 					  pq_params[i].port_id,
 					  tc_id,
 					  p_params->max_phys_tcs_per_port);
 		is_vf_pq = (i >= p_params->num_pf_pqs);
-		rl_valid = pq_params[i].rl_valid > 0;
 
 		/* Update first Tx PQ of VPORT/TC */
 		vport_id_in_pf = pq_params[i].vport_id - p_params->start_vport;
@@ -489,21 +541,14 @@ static void qed_tx_pq_map_rt_init(struct qed_hwfn *p_hwfn,
 				     map_val);
 		}
 
-		/* Check RL ID */
-		if (rl_valid && pq_params[i].vport_id >= max_qm_global_rls) {
-			DP_NOTICE(p_hwfn,
-				  "Invalid VPORT ID for rate limiter configuration\n");
-			rl_valid = false;
-		}
-
 		/* Prepare PQ map entry */
 		QM_INIT_TX_PQ_MAP(p_hwfn,
 				  tx_pq_map,
 				  E4,
 				  pq_id,
-				  rl_valid ? 1 : 0,
 				  *p_first_tx_pq_id,
-				  rl_valid ? pq_params[i].vport_id : 0,
+				  pq_params[i].rl_valid,
+				  pq_params[i].rl_id,
 				  ext_voq, pq_params[i].wrr_group);
 
 		/* Set PQ base address */
@@ -526,9 +571,8 @@ static void qed_tx_pq_map_rt_init(struct qed_hwfn *p_hwfn,
 						  p_params->pf_id,
 						  tc_id,
 						  pq_params[i].port_id,
-						  rl_valid ? 1 : 0,
-						  rl_valid ?
-						  pq_params[i].vport_id : 0);
+						  pq_params[i].rl_valid,
+						  pq_params[i].rl_id);
 			qed_wr(p_hwfn, p_ptt, PQ_INFO_RAM_GRC_ADDRESS(pq_id),
 			       pq_info);
 		}
@@ -666,19 +710,19 @@ static int qed_pf_rl_rt_init(struct qed_hwfn *p_hwfn, u8 pf_id, u32 pf_rl)
  * Return -1 on error.
  */
 static int qed_vp_wfq_rt_init(struct qed_hwfn *p_hwfn,
-			      u8 num_vports,
+			      u16 num_vports,
 			      struct init_qm_vport_params *vport_params)
 {
-	u16 vport_pq_id;
+	u16 vport_pq_id, i;
 	u32 inc_val;
-	u8 tc, i;
+	u8 tc;
 
 	/* Go over all PF VPORTs */
 	for (i = 0; i < num_vports; i++) {
-		if (!vport_params[i].vport_wfq)
+		if (!vport_params[i].wfq)
 			continue;
 
-		inc_val = QM_WFQ_INC_VAL(vport_params[i].vport_wfq);
+		inc_val = QM_WFQ_INC_VAL(vport_params[i].wfq);
 		if (inc_val > QM_WFQ_MAX_INC_VAL) {
 			DP_NOTICE(p_hwfn,
 				  "Invalid VPORT WFQ weight configuration\n");
@@ -703,48 +747,6 @@ static int qed_vp_wfq_rt_init(struct qed_hwfn *p_hwfn,
 	return 0;
 }
 
-/* Prepare VPORT RL runtime init values for the specified VPORTs.
- * Return -1 on error.
- */
-static int qed_vport_rl_rt_init(struct qed_hwfn *p_hwfn,
-				u8 start_vport,
-				u8 num_vports,
-				u32 link_speed,
-				struct init_qm_vport_params *vport_params)
-{
-	u8 i, vport_id;
-	u32 inc_val;
-
-	if (start_vport + num_vports >= MAX_QM_GLOBAL_RLS) {
-		DP_NOTICE(p_hwfn,
-			  "Invalid VPORT ID for rate limiter configuration\n");
-		return -1;
-	}
-
-	/* Go over all PF VPORTs */
-	for (i = 0, vport_id = start_vport; i < num_vports; i++, vport_id++) {
-		inc_val = QM_RL_INC_VAL(vport_params[i].vport_rl ?
-			  vport_params[i].vport_rl :
-			  link_speed);
-		if (inc_val > QM_VP_RL_MAX_INC_VAL(link_speed)) {
-			DP_NOTICE(p_hwfn,
-				  "Invalid VPORT rate-limit configuration\n");
-			return -1;
-		}
-
-		STORE_RT_REG(p_hwfn, QM_REG_RLGLBLCRD_RT_OFFSET + vport_id,
-			     (u32)QM_RL_CRD_REG_SIGN_BIT);
-		STORE_RT_REG(p_hwfn,
-			     QM_REG_RLGLBLUPPERBOUND_RT_OFFSET + vport_id,
-			     QM_VP_RL_UPPER_BOUND(link_speed) |
-			     (u32)QM_RL_CRD_REG_SIGN_BIT);
-		STORE_RT_REG(p_hwfn, QM_REG_RLGLBLINCVAL_RT_OFFSET + vport_id,
-			     inc_val);
-	}
-
-	return 0;
-}
-
 static bool qed_poll_on_qm_cmd_ready(struct qed_hwfn *p_hwfn,
 				     struct qed_ptt *p_ptt)
 {
@@ -796,23 +798,20 @@ u32 qed_qm_pf_mem_size(u32 num_pf_cids,
 int qed_qm_common_rt_init(struct qed_hwfn *p_hwfn,
 			  struct qed_qm_common_rt_init_params *p_params)
 {
-	/* Init AFullOprtnstcCrdMask */
-	u32 mask = (QM_OPPOR_LINE_VOQ_DEF <<
-		    QM_RF_OPPORTUNISTIC_MASK_LINEVOQ_SHIFT) |
-		   (QM_BYTE_CRD_EN << QM_RF_OPPORTUNISTIC_MASK_BYTEVOQ_SHIFT) |
-		   (p_params->pf_wfq_en <<
-		    QM_RF_OPPORTUNISTIC_MASK_PFWFQ_SHIFT) |
-		   (p_params->vport_wfq_en <<
-		    QM_RF_OPPORTUNISTIC_MASK_VPWFQ_SHIFT) |
-		   (p_params->pf_rl_en <<
-		    QM_RF_OPPORTUNISTIC_MASK_PFRL_SHIFT) |
-		   (p_params->vport_rl_en <<
-		    QM_RF_OPPORTUNISTIC_MASK_VPQCNRL_SHIFT) |
-		   (QM_OPPOR_FW_STOP_DEF <<
-		    QM_RF_OPPORTUNISTIC_MASK_FWPAUSE_SHIFT) |
-		   (QM_OPPOR_PQ_EMPTY_DEF <<
-		    QM_RF_OPPORTUNISTIC_MASK_QUEUEEMPTY_SHIFT);
+	u32 mask = 0;
 
+	/* Init AFullOprtnstcCrdMask */
+	SET_FIELD(mask, QM_RF_OPPORTUNISTIC_MASK_LINEVOQ,
+		  QM_OPPOR_LINE_VOQ_DEF);
+	SET_FIELD(mask, QM_RF_OPPORTUNISTIC_MASK_BYTEVOQ, QM_BYTE_CRD_EN);
+	SET_FIELD(mask, QM_RF_OPPORTUNISTIC_MASK_PFWFQ, p_params->pf_wfq_en);
+	SET_FIELD(mask, QM_RF_OPPORTUNISTIC_MASK_VPWFQ, p_params->vport_wfq_en);
+	SET_FIELD(mask, QM_RF_OPPORTUNISTIC_MASK_PFRL, p_params->pf_rl_en);
+	SET_FIELD(mask, QM_RF_OPPORTUNISTIC_MASK_VPQCNRL,
+		  p_params->global_rl_en);
+	SET_FIELD(mask, QM_RF_OPPORTUNISTIC_MASK_FWPAUSE, QM_OPPOR_FW_STOP_DEF);
+	SET_FIELD(mask,
+		  QM_RF_OPPORTUNISTIC_MASK_QUEUEEMPTY, QM_OPPOR_PQ_EMPTY_DEF);
 	STORE_RT_REG(p_hwfn, QM_REG_AFULLOPRTNSTCCRDMASK_RT_OFFSET, mask);
 
 	/* Enable/disable PF RL */
@@ -821,8 +820,8 @@ int qed_qm_common_rt_init(struct qed_hwfn *p_hwfn,
 	/* Enable/disable PF WFQ */
 	qed_enable_pf_wfq(p_hwfn, p_params->pf_wfq_en);
 
-	/* Enable/disable VPORT RL */
-	qed_enable_vport_rl(p_hwfn, p_params->vport_rl_en);
+	/* Enable/disable global RL */
+	qed_enable_global_rl(p_hwfn, p_params->global_rl_en);
 
 	/* Enable/disable VPORT WFQ */
 	qed_enable_vport_wfq(p_hwfn, p_params->vport_wfq_en);
@@ -839,6 +838,8 @@ int qed_qm_common_rt_init(struct qed_hwfn *p_hwfn,
 			       p_params->max_phys_tcs_per_port,
 			       p_params->port_params);
 
+	qed_global_rl_rt_init(p_hwfn);
+
 	return 0;
 }
 
@@ -850,7 +851,9 @@ int qed_qm_pf_rt_init(struct qed_hwfn *p_hwfn,
 	u32 other_mem_size_4kb = QM_PQ_MEM_4KB(p_params->num_pf_cids +
 					       p_params->num_tids) *
 				 QM_OTHER_PQS_PER_PF;
-	u8 tc, i;
+	u16 i;
+	u8 tc;
+
 
 	/* Clear first Tx PQ ID array for each VPORT */
 	for (i = 0; i < p_params->num_vports; i++)
@@ -875,16 +878,10 @@ int qed_qm_pf_rt_init(struct qed_hwfn *p_hwfn,
 	if (qed_pf_rl_rt_init(p_hwfn, p_params->pf_id, p_params->pf_rl))
 		return -1;
 
-	/* Set VPORT WFQ */
+	/* Init VPORT WFQ */
 	if (qed_vp_wfq_rt_init(p_hwfn, p_params->num_vports, vport_params))
 		return -1;
 
-	/* Set VPORT RL */
-	if (qed_vport_rl_rt_init(p_hwfn, p_params->start_vport,
-				 p_params->num_vports, p_params->link_speed,
-				 vport_params))
-		return -1;
-
 	return 0;
 }
 
@@ -922,18 +919,19 @@ int qed_init_pf_rl(struct qed_hwfn *p_hwfn,
 
 int qed_init_vport_wfq(struct qed_hwfn *p_hwfn,
 		       struct qed_ptt *p_ptt,
-		       u16 first_tx_pq_id[NUM_OF_TCS], u16 vport_wfq)
+		       u16 first_tx_pq_id[NUM_OF_TCS], u16 wfq)
 {
 	u16 vport_pq_id;
 	u32 inc_val;
 	u8 tc;
 
-	inc_val = QM_WFQ_INC_VAL(vport_wfq);
+	inc_val = QM_WFQ_INC_VAL(wfq);
 	if (!inc_val || inc_val > QM_WFQ_MAX_INC_VAL) {
-		DP_NOTICE(p_hwfn, "Invalid VPORT WFQ weight configuration\n");
+		DP_NOTICE(p_hwfn, "Invalid VPORT WFQ configuration.\n");
 		return -1;
 	}
 
+	/* A VPORT can have several VPORT PQ IDs for various TCs */
 	for (tc = 0; tc < NUM_OF_TCS; tc++) {
 		vport_pq_id = first_tx_pq_id[tc];
 		if (vport_pq_id != QM_INVALID_PQ_ID)
@@ -945,28 +943,20 @@ int qed_init_vport_wfq(struct qed_hwfn *p_hwfn,
 	return 0;
 }
 
-int qed_init_vport_rl(struct qed_hwfn *p_hwfn,
-		      struct qed_ptt *p_ptt,
-		      u8 vport_id, u32 vport_rl, u32 link_speed)
+int qed_init_global_rl(struct qed_hwfn *p_hwfn,
+		       struct qed_ptt *p_ptt, u16 rl_id, u32 rate_limit)
 {
-	u32 inc_val, max_qm_global_rls = MAX_QM_GLOBAL_RLS;
-
-	if (vport_id >= max_qm_global_rls) {
-		DP_NOTICE(p_hwfn,
-			  "Invalid VPORT ID for rate limiter configuration\n");
-		return -1;
-	}
+	u32 inc_val;
 
-	inc_val = QM_RL_INC_VAL(vport_rl ? vport_rl : link_speed);
-	if (inc_val > QM_VP_RL_MAX_INC_VAL(link_speed)) {
-		DP_NOTICE(p_hwfn, "Invalid VPORT rate-limit configuration\n");
+	inc_val = QM_RL_INC_VAL(rate_limit);
+	if (inc_val > QM_VP_RL_MAX_INC_VAL(rate_limit)) {
+		DP_NOTICE(p_hwfn, "Invalid rate limit configuration.\n");
 		return -1;
 	}
 
-	qed_wr(p_hwfn,
-	       p_ptt,
-	       QM_REG_RLGLBLCRD + vport_id * 4, (u32)QM_RL_CRD_REG_SIGN_BIT);
-	qed_wr(p_hwfn, p_ptt, QM_REG_RLGLBLINCVAL + vport_id * 4, inc_val);
+	qed_wr(p_hwfn, p_ptt,
+	       QM_REG_RLGLBLCRD + rl_id * 4, (u32)QM_RL_CRD_REG_SIGN_BIT);
+	qed_wr(p_hwfn, p_ptt, QM_REG_RLGLBLINCVAL + rl_id * 4, inc_val);
 
 	return 0;
 }
diff --git a/drivers/net/ethernet/qlogic/qed/qed_sriov.c b/drivers/net/ethernet/qlogic/qed/qed_sriov.c
index dcb5c917f373..fe4b740de14c 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_sriov.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_sriov.c
@@ -4354,9 +4354,9 @@ qed_iov_bulletin_get_forced_vlan(struct qed_hwfn *p_hwfn, u16 rel_vf_id)
 static int qed_iov_configure_tx_rate(struct qed_hwfn *p_hwfn,
 				     struct qed_ptt *p_ptt, int vfid, int val)
 {
-	struct qed_mcp_link_state *p_link;
 	struct qed_vf_info *vf;
 	u8 abs_vp_id = 0;
+	u16 rl_id;
 	int rc;
 
 	vf = qed_iov_get_vf_info(p_hwfn, (u16)vfid, true);
@@ -4367,10 +4367,8 @@ static int qed_iov_configure_tx_rate(struct qed_hwfn *p_hwfn,
 	if (rc)
 		return rc;
 
-	p_link = &QED_LEADING_HWFN(p_hwfn->cdev)->mcp_info->link_output;
-
-	return qed_init_vport_rl(p_hwfn, p_ptt, abs_vp_id, (u32)val,
-				 p_link->speed);
+	rl_id = abs_vp_id;	/* The "rl_id" is set as the "vport_id" */
+	return qed_init_global_rl(p_hwfn, p_ptt, rl_id, (u32)val);
 }
 
 static int
-- 
cgit v1.2.3-59-g8ed1b


From 63ddca3052ce2397ffd71899390d552cd40ddb6f Mon Sep 17 00:00:00 2001
From: Michal Kalderon <michal.kalderon@marvell.com>
Date: Mon, 27 Jan 2020 15:26:10 +0200
Subject: qed: FW 8.42.2.0 Parser offsets modified

Convert storm ram line to regpair rather than two distinct u32
to better represent the u64 width of the ram.
Convert some defines to be hex instead of negative values
these values also changed by FW from previous value.

Signed-off-by: Ariel Elior <ariel.elior@marvell.com>
Signed-off-by: Michal Kalderon <michal.kalderon@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../net/ethernet/qlogic/qed/qed_init_fw_funcs.c    | 51 +++++++++++-----------
 1 file changed, 25 insertions(+), 26 deletions(-)

diff --git a/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c b/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c
index 965529b1a9c5..40b8a52dbfa3 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c
@@ -1000,7 +1000,6 @@ bool qed_send_qm_stop_cmd(struct qed_hwfn *p_hwfn,
 	return true;
 }
 
-
 #define SET_TUNNEL_TYPE_ENABLE_BIT(var, offset, enable) \
 	do { \
 		typeof(var) *__p_var = &(var); \
@@ -1008,8 +1007,9 @@ bool qed_send_qm_stop_cmd(struct qed_hwfn *p_hwfn,
 		*__p_var = (*__p_var & ~BIT(__offset)) | \
 			   ((enable) ? BIT(__offset) : 0); \
 	} while (0)
-#define PRS_ETH_TUNN_OUTPUT_FORMAT        -188897008
-#define PRS_ETH_OUTPUT_FORMAT             -46832
+
+#define PRS_ETH_TUNN_OUTPUT_FORMAT     0xF4DAB910
+#define PRS_ETH_OUTPUT_FORMAT          0xFFFF4910
 
 void qed_set_vxlan_dest_port(struct qed_hwfn *p_hwfn,
 			     struct qed_ptt *p_ptt, u16 dest_port)
@@ -1153,8 +1153,8 @@ void qed_set_geneve_enable(struct qed_hwfn *p_hwfn,
 	       ip_geneve_enable ? 1 : 0);
 }
 
-#define PRS_ETH_VXLAN_NO_L2_ENABLE_OFFSET   4
-#define PRS_ETH_VXLAN_NO_L2_OUTPUT_FORMAT      -927094512
+#define PRS_ETH_VXLAN_NO_L2_ENABLE_OFFSET      3
+#define PRS_ETH_VXLAN_NO_L2_OUTPUT_FORMAT   -925189872
 
 void qed_set_vxlan_no_l2_enable(struct qed_hwfn *p_hwfn,
 				struct qed_ptt *p_ptt, bool enable)
@@ -1219,7 +1219,8 @@ void qed_gft_config(struct qed_hwfn *p_hwfn,
 		    bool udp,
 		    bool ipv4, bool ipv6, enum gft_profile_type profile_type)
 {
-	u32 reg_val, cam_line, ram_line_lo, ram_line_hi, search_non_ip_as_gft;
+	u32 reg_val, cam_line, search_non_ip_as_gft;
+	struct regpair ram_line = { };
 
 	if (!ipv6 && !ipv4)
 		DP_NOTICE(p_hwfn,
@@ -1285,35 +1286,33 @@ void qed_gft_config(struct qed_hwfn *p_hwfn,
 	    qed_rd(p_hwfn, p_ptt, PRS_REG_GFT_CAM + CAM_LINE_SIZE * pf_id);
 
 	/* Write line to RAM - compare to filter 4 tuple */
-	ram_line_lo = 0;
-	ram_line_hi = 0;
 
 	/* Search no IP as GFT */
 	search_non_ip_as_gft = 0;
 
 	/* Tunnel type */
-	SET_FIELD(ram_line_lo, GFT_RAM_LINE_TUNNEL_DST_PORT, 1);
-	SET_FIELD(ram_line_lo, GFT_RAM_LINE_TUNNEL_OVER_IP_PROTOCOL, 1);
+	SET_FIELD(ram_line.lo, GFT_RAM_LINE_TUNNEL_DST_PORT, 1);
+	SET_FIELD(ram_line.lo, GFT_RAM_LINE_TUNNEL_OVER_IP_PROTOCOL, 1);
 
 	if (profile_type == GFT_PROFILE_TYPE_4_TUPLE) {
-		SET_FIELD(ram_line_hi, GFT_RAM_LINE_DST_IP, 1);
-		SET_FIELD(ram_line_hi, GFT_RAM_LINE_SRC_IP, 1);
-		SET_FIELD(ram_line_hi, GFT_RAM_LINE_OVER_IP_PROTOCOL, 1);
-		SET_FIELD(ram_line_lo, GFT_RAM_LINE_ETHERTYPE, 1);
-		SET_FIELD(ram_line_lo, GFT_RAM_LINE_SRC_PORT, 1);
-		SET_FIELD(ram_line_lo, GFT_RAM_LINE_DST_PORT, 1);
+		SET_FIELD(ram_line.hi, GFT_RAM_LINE_DST_IP, 1);
+		SET_FIELD(ram_line.hi, GFT_RAM_LINE_SRC_IP, 1);
+		SET_FIELD(ram_line.hi, GFT_RAM_LINE_OVER_IP_PROTOCOL, 1);
+		SET_FIELD(ram_line.lo, GFT_RAM_LINE_ETHERTYPE, 1);
+		SET_FIELD(ram_line.lo, GFT_RAM_LINE_SRC_PORT, 1);
+		SET_FIELD(ram_line.lo, GFT_RAM_LINE_DST_PORT, 1);
 	} else if (profile_type == GFT_PROFILE_TYPE_L4_DST_PORT) {
-		SET_FIELD(ram_line_hi, GFT_RAM_LINE_OVER_IP_PROTOCOL, 1);
-		SET_FIELD(ram_line_lo, GFT_RAM_LINE_ETHERTYPE, 1);
-		SET_FIELD(ram_line_lo, GFT_RAM_LINE_DST_PORT, 1);
+		SET_FIELD(ram_line.hi, GFT_RAM_LINE_OVER_IP_PROTOCOL, 1);
+		SET_FIELD(ram_line.lo, GFT_RAM_LINE_ETHERTYPE, 1);
+		SET_FIELD(ram_line.lo, GFT_RAM_LINE_DST_PORT, 1);
 	} else if (profile_type == GFT_PROFILE_TYPE_IP_DST_ADDR) {
-		SET_FIELD(ram_line_hi, GFT_RAM_LINE_DST_IP, 1);
-		SET_FIELD(ram_line_lo, GFT_RAM_LINE_ETHERTYPE, 1);
+		SET_FIELD(ram_line.hi, GFT_RAM_LINE_DST_IP, 1);
+		SET_FIELD(ram_line.lo, GFT_RAM_LINE_ETHERTYPE, 1);
 	} else if (profile_type == GFT_PROFILE_TYPE_IP_SRC_ADDR) {
-		SET_FIELD(ram_line_hi, GFT_RAM_LINE_SRC_IP, 1);
-		SET_FIELD(ram_line_lo, GFT_RAM_LINE_ETHERTYPE, 1);
+		SET_FIELD(ram_line.hi, GFT_RAM_LINE_SRC_IP, 1);
+		SET_FIELD(ram_line.lo, GFT_RAM_LINE_ETHERTYPE, 1);
 	} else if (profile_type == GFT_PROFILE_TYPE_TUNNEL_TYPE) {
-		SET_FIELD(ram_line_lo, GFT_RAM_LINE_TUNNEL_ETHERTYPE, 1);
+		SET_FIELD(ram_line.lo, GFT_RAM_LINE_TUNNEL_ETHERTYPE, 1);
 
 		/* Allow tunneled traffic without inner IP */
 		search_non_ip_as_gft = 1;
@@ -1324,11 +1323,11 @@ void qed_gft_config(struct qed_hwfn *p_hwfn,
 	qed_wr(p_hwfn,
 	       p_ptt,
 	       PRS_REG_GFT_PROFILE_MASK_RAM + RAM_LINE_SIZE * pf_id,
-	       ram_line_lo);
+	       ram_line.lo);
 	qed_wr(p_hwfn,
 	       p_ptt,
 	       PRS_REG_GFT_PROFILE_MASK_RAM + RAM_LINE_SIZE * pf_id + REG_SIZE,
-	       ram_line_hi);
+	       ram_line.hi);
 
 	/* Set default profile so that no filter match will happen */
 	qed_wr(p_hwfn,
-- 
cgit v1.2.3-59-g8ed1b


From 804c5702fc583afbdb6860f79e9d4058570ccc43 Mon Sep 17 00:00:00 2001
From: Michal Kalderon <michal.kalderon@marvell.com>
Date: Mon, 27 Jan 2020 15:26:11 +0200
Subject: qed: Use dmae to write to widebus registers in fw_funcs

There are several wide-bus registers written to by the fw_funcs
that require using the dmae for atomicity. Therefore using the dmae
channel functionality was added to the fw_funcs file, since the code
is very similar to the previously used code, the structures used were
moved to qed_hsi. Due to FW conventions, the names of the flags in the
struct changed. Since this required slight modification in the places
that set the flags the code was modified to use GET/SET FIELD macros.

Signed-off-by: Ariel Elior <ariel.elior@marvell.com>
Signed-off-by: Michal Kalderon <michal.kalderon@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/qlogic/qed/qed_dev.c          |  2 +-
 drivers/net/ethernet/qlogic/qed/qed_dev_api.h      | 24 ------
 drivers/net/ethernet/qlogic/qed/qed_hsi.h          | 41 ++++++++++-
 drivers/net/ethernet/qlogic/qed/qed_hw.c           | 67 ++++++++---------
 .../net/ethernet/qlogic/qed/qed_init_fw_funcs.c    | 86 ++++++++++++++++------
 drivers/net/ethernet/qlogic/qed/qed_init_ops.c     |  2 +-
 drivers/net/ethernet/qlogic/qed/qed_sriov.c        | 11 +--
 7 files changed, 142 insertions(+), 91 deletions(-)

diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c
index 249fcc3dc138..479d98e6187a 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_dev.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c
@@ -907,7 +907,7 @@ qed_llh_access_filter(struct qed_hwfn *p_hwfn,
 	/* Filter value */
 	addr = NIG_REG_LLH_FUNC_FILTER_VALUE + 2 * filter_idx * 0x4;
 
-	params.flags = QED_DMAE_FLAG_PF_DST;
+	SET_FIELD(params.flags, QED_DMAE_PARAMS_DST_PF_VALID, 0x1);
 	params.dst_pfid = pfid;
 	rc = qed_dmae_host2grc(p_hwfn,
 			       p_ptt,
diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev_api.h b/drivers/net/ethernet/qlogic/qed/qed_dev_api.h
index 47376d4d071f..eb4808b3bf67 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_dev_api.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_dev_api.h
@@ -230,30 +230,6 @@ enum qed_dmae_address_type_t {
 	QED_DMAE_ADDRESS_GRC
 };
 
-/* value of flags If QED_DMAE_FLAG_RW_REPL_SRC flag is set and the
- * source is a block of length DMAE_MAX_RW_SIZE and the
- * destination is larger, the source block will be duplicated as
- * many times as required to fill the destination block. This is
- * used mostly to write a zeroed buffer to destination address
- * using DMA
- */
-#define QED_DMAE_FLAG_RW_REPL_SRC	0x00000001
-#define QED_DMAE_FLAG_VF_SRC		0x00000002
-#define QED_DMAE_FLAG_VF_DST		0x00000004
-#define QED_DMAE_FLAG_COMPLETION_DST	0x00000008
-#define QED_DMAE_FLAG_PORT		0x00000010
-#define QED_DMAE_FLAG_PF_SRC		0x00000020
-#define QED_DMAE_FLAG_PF_DST		0x00000040
-
-struct qed_dmae_params {
-	u32 flags; /* consists of QED_DMAE_FLAG_* values */
-	u8 src_vfid;
-	u8 dst_vfid;
-	u8 port_id;
-	u8 src_pfid;
-	u8 dst_pfid;
-};
-
 /**
  * @brief qed_dmae_host2grc - copy data from source addr to
  * dmae registers using the given ptt
diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
index 5ac68ec01a04..cced2ce365de 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
@@ -1435,7 +1435,11 @@ struct dmae_cmd {
 	__le16 crc16;
 	__le16 crc16_c;
 	__le16 crc10;
-	__le16 reserved;
+	__le16 error_bit_reserved;
+#define DMAE_CMD_ERROR_BIT_MASK        0x1
+#define DMAE_CMD_ERROR_BIT_SHIFT       0
+#define DMAE_CMD_RESERVED_MASK	       0x7FFF
+#define DMAE_CMD_RESERVED_SHIFT        1
 	__le16 xsum16;
 	__le16 xsum8;
 };
@@ -1566,6 +1570,41 @@ struct e4_ystorm_core_conn_ag_ctx {
 	__le32 reg3;
 };
 
+/* DMAE parameters */
+struct qed_dmae_params {
+	u32 flags;
+/* If QED_DMAE_PARAMS_RW_REPL_SRC flag is set and the
+ * source is a block of length DMAE_MAX_RW_SIZE and the
+ * destination is larger, the source block will be duplicated as
+ * many times as required to fill the destination block. This is
+ * used mostly to write a zeroed buffer to destination address
+ * using DMA
+ */
+#define QED_DMAE_PARAMS_RW_REPL_SRC_MASK	0x1
+#define QED_DMAE_PARAMS_RW_REPL_SRC_SHIFT	0
+#define QED_DMAE_PARAMS_SRC_VF_VALID_MASK	0x1
+#define QED_DMAE_PARAMS_SRC_VF_VALID_SHIFT	1
+#define QED_DMAE_PARAMS_DST_VF_VALID_MASK	0x1
+#define QED_DMAE_PARAMS_DST_VF_VALID_SHIFT	2
+#define QED_DMAE_PARAMS_COMPLETION_DST_MASK	0x1
+#define QED_DMAE_PARAMS_COMPLETION_DST_SHIFT	3
+#define QED_DMAE_PARAMS_PORT_VALID_MASK		0x1
+#define QED_DMAE_PARAMS_PORT_VALID_SHIFT	4
+#define QED_DMAE_PARAMS_SRC_PF_VALID_MASK	0x1
+#define QED_DMAE_PARAMS_SRC_PF_VALID_SHIFT	5
+#define QED_DMAE_PARAMS_DST_PF_VALID_MASK	0x1
+#define QED_DMAE_PARAMS_DST_PF_VALID_SHIFT	6
+#define QED_DMAE_PARAMS_RESERVED_MASK		0x1FFFFFF
+#define QED_DMAE_PARAMS_RESERVED_SHIFT		7
+	u8 src_vfid;
+	u8 dst_vfid;
+	u8 port_id;
+	u8 src_pfid;
+	u8 dst_pfid;
+	u8 reserved1;
+	__le16 reserved2;
+};
+
 /* IGU cleanup command */
 struct igu_cleanup {
 	__le32 sb_id_and_flags;
diff --git a/drivers/net/ethernet/qlogic/qed/qed_hw.c b/drivers/net/ethernet/qlogic/qed/qed_hw.c
index a4de9e3ef72c..4ab8cfaf63d1 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_hw.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_hw.c
@@ -393,7 +393,7 @@ u32 qed_vfid_to_concrete(struct qed_hwfn *p_hwfn, u8 vfid)
 
 /* DMAE */
 #define QED_DMAE_FLAGS_IS_SET(params, flag) \
-	((params) != NULL && ((params)->flags & QED_DMAE_FLAG_##flag))
+	((params) != NULL && GET_FIELD((params)->flags, QED_DMAE_PARAMS_##flag))
 
 static void qed_dmae_opcode(struct qed_hwfn *p_hwfn,
 			    const u8 is_src_type_grc,
@@ -408,62 +408,55 @@ static void qed_dmae_opcode(struct qed_hwfn *p_hwfn,
 	 * 0- The source is the PCIe
 	 * 1- The source is the GRC.
 	 */
-	opcode |= (is_src_type_grc ? DMAE_CMD_SRC_MASK_GRC
-				   : DMAE_CMD_SRC_MASK_PCIE) <<
-		   DMAE_CMD_SRC_SHIFT;
-	src_pfid = QED_DMAE_FLAGS_IS_SET(p_params, PF_SRC) ?
-		   p_params->src_pfid : p_hwfn->rel_pf_id;
-	opcode |= ((src_pfid & DMAE_CMD_SRC_PF_ID_MASK) <<
-		   DMAE_CMD_SRC_PF_ID_SHIFT);
+	SET_FIELD(opcode, DMAE_CMD_SRC,
+		  (is_src_type_grc ? dmae_cmd_src_grc : dmae_cmd_src_pcie));
+	src_pfid = QED_DMAE_FLAGS_IS_SET(p_params, SRC_PF_VALID) ?
+	    p_params->src_pfid : p_hwfn->rel_pf_id;
+	SET_FIELD(opcode, DMAE_CMD_SRC_PF_ID, src_pfid);
 
 	/* The destination of the DMA can be: 0-None 1-PCIe 2-GRC 3-None */
-	opcode |= (is_dst_type_grc ? DMAE_CMD_DST_MASK_GRC
-				   : DMAE_CMD_DST_MASK_PCIE) <<
-		   DMAE_CMD_DST_SHIFT;
-	dst_pfid = QED_DMAE_FLAGS_IS_SET(p_params, PF_DST) ?
-		   p_params->dst_pfid : p_hwfn->rel_pf_id;
-	opcode |= ((dst_pfid & DMAE_CMD_DST_PF_ID_MASK) <<
-		   DMAE_CMD_DST_PF_ID_SHIFT);
+	SET_FIELD(opcode, DMAE_CMD_DST,
+		  (is_dst_type_grc ? dmae_cmd_dst_grc : dmae_cmd_dst_pcie));
+	dst_pfid = QED_DMAE_FLAGS_IS_SET(p_params, DST_PF_VALID) ?
+	    p_params->dst_pfid : p_hwfn->rel_pf_id;
+	SET_FIELD(opcode, DMAE_CMD_DST_PF_ID, dst_pfid);
+
 
 	/* Whether to write a completion word to the completion destination:
 	 * 0-Do not write a completion word
 	 * 1-Write the completion word
 	 */
-	opcode |= (DMAE_CMD_COMP_WORD_EN_MASK << DMAE_CMD_COMP_WORD_EN_SHIFT);
-	opcode |= (DMAE_CMD_SRC_ADDR_RESET_MASK <<
-		   DMAE_CMD_SRC_ADDR_RESET_SHIFT);
+	SET_FIELD(opcode, DMAE_CMD_COMP_WORD_EN, 1);
+	SET_FIELD(opcode, DMAE_CMD_SRC_ADDR_RESET, 1);
 
 	if (QED_DMAE_FLAGS_IS_SET(p_params, COMPLETION_DST))
-		opcode |= (1 << DMAE_CMD_COMP_FUNC_SHIFT);
+		SET_FIELD(opcode, DMAE_CMD_COMP_FUNC, 1);
 
-	opcode |= (DMAE_CMD_ENDIANITY << DMAE_CMD_ENDIANITY_MODE_SHIFT);
+	/* swapping mode 3 - big endian */
+	SET_FIELD(opcode, DMAE_CMD_ENDIANITY_MODE, DMAE_CMD_ENDIANITY);
 
-	port_id = (QED_DMAE_FLAGS_IS_SET(p_params, PORT)) ?
-		   p_params->port_id : p_hwfn->port_id;
-	opcode |= (port_id << DMAE_CMD_PORT_ID_SHIFT);
+	port_id = (QED_DMAE_FLAGS_IS_SET(p_params, PORT_VALID)) ?
+	    p_params->port_id : p_hwfn->port_id;
+	SET_FIELD(opcode, DMAE_CMD_PORT_ID, port_id);
 
 	/* reset source address in next go */
-	opcode |= (DMAE_CMD_SRC_ADDR_RESET_MASK <<
-		   DMAE_CMD_SRC_ADDR_RESET_SHIFT);
+	SET_FIELD(opcode, DMAE_CMD_SRC_ADDR_RESET, 1);
 
 	/* reset dest address in next go */
-	opcode |= (DMAE_CMD_DST_ADDR_RESET_MASK <<
-		   DMAE_CMD_DST_ADDR_RESET_SHIFT);
+	SET_FIELD(opcode, DMAE_CMD_DST_ADDR_RESET, 1);
 
 	/* SRC/DST VFID: all 1's - pf, otherwise VF id */
-	if (QED_DMAE_FLAGS_IS_SET(p_params, VF_SRC)) {
-		opcode |= 1 << DMAE_CMD_SRC_VF_ID_VALID_SHIFT;
-		opcode_b |= p_params->src_vfid << DMAE_CMD_SRC_VF_ID_SHIFT;
+	if (QED_DMAE_FLAGS_IS_SET(p_params, SRC_VF_VALID)) {
+		SET_FIELD(opcode, DMAE_CMD_SRC_VF_ID_VALID, 1);
+		SET_FIELD(opcode_b, DMAE_CMD_SRC_VF_ID, p_params->src_vfid);
 	} else {
-		opcode_b |= DMAE_CMD_SRC_VF_ID_MASK <<
-			    DMAE_CMD_SRC_VF_ID_SHIFT;
+		SET_FIELD(opcode_b, DMAE_CMD_SRC_VF_ID, 0xFF);
 	}
-
-	if (QED_DMAE_FLAGS_IS_SET(p_params, VF_DST)) {
-		opcode |= 1 << DMAE_CMD_DST_VF_ID_VALID_SHIFT;
-		opcode_b |= p_params->dst_vfid << DMAE_CMD_DST_VF_ID_SHIFT;
+	if (QED_DMAE_FLAGS_IS_SET(p_params, DST_VF_VALID)) {
+		SET_FIELD(opcode, DMAE_CMD_DST_VF_ID_VALID, 1);
+		SET_FIELD(opcode_b, DMAE_CMD_DST_VF_ID, p_params->dst_vfid);
 	} else {
-		opcode_b |= DMAE_CMD_DST_VF_ID_MASK << DMAE_CMD_DST_VF_ID_SHIFT;
+		SET_FIELD(opcode_b, DMAE_CMD_DST_VF_ID, 0xFF);
 	}
 
 	p_hwfn->dmae_info.p_dmae_cmd->opcode = cpu_to_le32(opcode);
diff --git a/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c b/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c
index 40b8a52dbfa3..6fdf9ab98685 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c
@@ -1011,6 +1011,56 @@ bool qed_send_qm_stop_cmd(struct qed_hwfn *p_hwfn,
 #define PRS_ETH_TUNN_OUTPUT_FORMAT     0xF4DAB910
 #define PRS_ETH_OUTPUT_FORMAT          0xFFFF4910
 
+#define ARR_REG_WR(dev, ptt, addr, arr,	arr_size) \
+	do { \
+		u32 i; \
+		\
+		for (i = 0; i < (arr_size); i++) \
+			qed_wr(dev, ptt, \
+			       ((addr) + (4 * i)), \
+			       ((u32 *)&(arr))[i]); \
+	} while (0)
+
+/**
+ * @brief qed_dmae_to_grc - is an internal function - writes from host to
+ * wide-bus registers (split registers are not supported yet)
+ *
+ * @param p_hwfn - HW device data
+ * @param p_ptt - ptt window used for writing the registers.
+ * @param p_data - pointer to source data.
+ * @param addr - Destination register address.
+ * @param len_in_dwords - data length in DWARDS (u32)
+ */
+static int qed_dmae_to_grc(struct qed_hwfn *p_hwfn,
+			   struct qed_ptt *p_ptt,
+			   u32 *p_data, u32 addr, u32 len_in_dwords)
+{
+	struct qed_dmae_params params = {};
+	int rc;
+
+	if (!p_data)
+		return -1;
+
+	/* Set DMAE params */
+	SET_FIELD(params.flags, QED_DMAE_PARAMS_COMPLETION_DST, 1);
+
+	/* Execute DMAE command */
+	rc = qed_dmae_host2grc(p_hwfn, p_ptt,
+			       (u64)(uintptr_t)(p_data),
+			       addr, len_in_dwords, &params);
+
+	/* If not read using DMAE, read using GRC */
+	if (rc) {
+		DP_VERBOSE(p_hwfn,
+			   QED_MSG_DEBUG,
+			   "Failed writing to chip using DMAE, using GRC instead\n");
+		/* write to registers using GRC */
+		ARR_REG_WR(p_hwfn, p_ptt, addr, p_data, len_in_dwords);
+	}
+
+	return len_in_dwords;
+}
+
 void qed_set_vxlan_dest_port(struct qed_hwfn *p_hwfn,
 			     struct qed_ptt *p_ptt, u16 dest_port)
 {
@@ -1195,6 +1245,8 @@ void qed_set_vxlan_no_l2_enable(struct qed_hwfn *p_hwfn,
 
 void qed_gft_disable(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, u16 pf_id)
 {
+	struct regpair ram_line = { };
+
 	/* Disable gft search for PF */
 	qed_wr(p_hwfn, p_ptt, PRS_REG_SEARCH_GFT, 0);
 
@@ -1204,12 +1256,9 @@ void qed_gft_disable(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, u16 pf_id)
 	qed_wr(p_hwfn, p_ptt, PRS_REG_GFT_CAM + CAM_LINE_SIZE * pf_id, 0);
 
 	/* Zero ramline */
-	qed_wr(p_hwfn,
-	       p_ptt, PRS_REG_GFT_PROFILE_MASK_RAM + RAM_LINE_SIZE * pf_id, 0);
-	qed_wr(p_hwfn,
-	       p_ptt,
-	       PRS_REG_GFT_PROFILE_MASK_RAM + RAM_LINE_SIZE * pf_id + REG_SIZE,
-	       0);
+	qed_dmae_to_grc(p_hwfn, p_ptt, (u32 *)&ram_line,
+			PRS_REG_GFT_PROFILE_MASK_RAM + RAM_LINE_SIZE * pf_id,
+			sizeof(ram_line) / REG_SIZE);
 }
 
 void qed_gft_config(struct qed_hwfn *p_hwfn,
@@ -1320,24 +1369,17 @@ void qed_gft_config(struct qed_hwfn *p_hwfn,
 
 	qed_wr(p_hwfn,
 	       p_ptt, PRS_REG_SEARCH_NON_IP_AS_GFT, search_non_ip_as_gft);
-	qed_wr(p_hwfn,
-	       p_ptt,
-	       PRS_REG_GFT_PROFILE_MASK_RAM + RAM_LINE_SIZE * pf_id,
-	       ram_line.lo);
-	qed_wr(p_hwfn,
-	       p_ptt,
-	       PRS_REG_GFT_PROFILE_MASK_RAM + RAM_LINE_SIZE * pf_id + REG_SIZE,
-	       ram_line.hi);
+	qed_dmae_to_grc(p_hwfn, p_ptt, (u32 *)&ram_line,
+			PRS_REG_GFT_PROFILE_MASK_RAM + RAM_LINE_SIZE * pf_id,
+			sizeof(ram_line) / REG_SIZE);
 
 	/* Set default profile so that no filter match will happen */
-	qed_wr(p_hwfn,
-	       p_ptt,
-	       PRS_REG_GFT_PROFILE_MASK_RAM + RAM_LINE_SIZE *
-	       PRS_GFT_CAM_LINES_NO_MATCH, 0xffffffff);
-	qed_wr(p_hwfn,
-	       p_ptt,
-	       PRS_REG_GFT_PROFILE_MASK_RAM + RAM_LINE_SIZE *
-	       PRS_GFT_CAM_LINES_NO_MATCH + REG_SIZE, 0x3ff);
+	ram_line.lo = 0xffffffff;
+	ram_line.hi = 0x3ff;
+	qed_dmae_to_grc(p_hwfn, p_ptt, (u32 *)&ram_line,
+			PRS_REG_GFT_PROFILE_MASK_RAM + RAM_LINE_SIZE *
+			PRS_GFT_CAM_LINES_NO_MATCH,
+			sizeof(ram_line) / REG_SIZE);
 
 	/* Enable gft search */
 	qed_wr(p_hwfn, p_ptt, PRS_REG_SEARCH_GFT, 1);
diff --git a/drivers/net/ethernet/qlogic/qed/qed_init_ops.c b/drivers/net/ethernet/qlogic/qed/qed_init_ops.c
index ef22a3596295..31222b3e3936 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_init_ops.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_init_ops.c
@@ -215,7 +215,7 @@ static int qed_init_fill_dmae(struct qed_hwfn *p_hwfn,
 	 * 3. p_hwfb->temp_data,
 	 * 4. fill_count
 	 */
-	params.flags = QED_DMAE_FLAG_RW_REPL_SRC;
+	SET_FIELD(params.flags, QED_DMAE_PARAMS_RW_REPL_SRC, 0x1);
 	return qed_dmae_host2grc(p_hwfn, p_ptt,
 				 (uintptr_t)(&zero_buffer[0]),
 				 addr, fill_count, &params);
diff --git a/drivers/net/ethernet/qlogic/qed/qed_sriov.c b/drivers/net/ethernet/qlogic/qed/qed_sriov.c
index fe4b740de14c..66876af814c4 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_sriov.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_sriov.c
@@ -352,7 +352,7 @@ static int qed_iov_post_vf_bulletin(struct qed_hwfn *p_hwfn,
 
 	/* propagate bulletin board via dmae to vm memory */
 	memset(&params, 0, sizeof(params));
-	params.flags = QED_DMAE_FLAG_VF_DST;
+	SET_FIELD(params.flags, QED_DMAE_PARAMS_DST_VF_VALID, 0x1);
 	params.dst_vfid = p_vf->abs_vf_id;
 	return qed_dmae_host2host(p_hwfn, p_ptt, p_vf->bulletin.phys,
 				  p_vf->vf_bulletin, p_vf->bulletin.size / 4,
@@ -1225,8 +1225,8 @@ static void qed_iov_send_response(struct qed_hwfn *p_hwfn,
 
 	eng_vf_id = p_vf->abs_vf_id;
 
-	memset(&params, 0, sizeof(struct qed_dmae_params));
-	params.flags = QED_DMAE_FLAG_VF_DST;
+	memset(&params, 0, sizeof(params));
+	SET_FIELD(params.flags, QED_DMAE_PARAMS_DST_VF_VALID, 0x1);
 	params.dst_vfid = eng_vf_id;
 
 	qed_dmae_host2host(p_hwfn, p_ptt, mbx->reply_phys + sizeof(u64),
@@ -4103,8 +4103,9 @@ static int qed_iov_copy_vf_msg(struct qed_hwfn *p_hwfn, struct qed_ptt *ptt,
 	if (!vf_info)
 		return -EINVAL;
 
-	memset(&params, 0, sizeof(struct qed_dmae_params));
-	params.flags = QED_DMAE_FLAG_VF_SRC | QED_DMAE_FLAG_COMPLETION_DST;
+	memset(&params, 0, sizeof(params));
+	SET_FIELD(params.flags, QED_DMAE_PARAMS_SRC_VF_VALID, 0x1);
+	SET_FIELD(params.flags, QED_DMAE_PARAMS_COMPLETION_DST, 0x1);
 	params.src_vfid = vf_info->abs_vf_id;
 
 	if (qed_dmae_host2host(p_hwfn, ptt,
-- 
cgit v1.2.3-59-g8ed1b


From 997af5df230e3288ec1f5b332955f9be643e450b Mon Sep 17 00:00:00 2001
From: Michal Kalderon <michal.kalderon@marvell.com>
Date: Mon, 27 Jan 2020 15:26:12 +0200
Subject: qed: FW 8.42.2.0 Additional ll2 type

LL2 queues were a limited resource due to FW constraints.
This FW introduced a new resource which is a context based ll2 queue
(memory on host). The additional ll2 queues are required for RDMA SRIOV.
The code refers to the previous ll2 queues as ram-based or legacy, and the
new queues as ctx-based.
This change decreased the "legacy" ram-based queues therefore the first ll2
queue used for iWARP was converted to the ctx-based ll2 queue.
This feature also exposed a bug in the DIRECT_REG_WR64 macro implementation
which didn't have an effect in other use cases.

Signed-off-by: Ariel Elior <ariel.elior@marvell.com>
Signed-off-by: Michal Kalderon <michal.kalderon@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/qlogic/qed/qed.h       |   3 +-
 drivers/net/ethernet/qlogic/qed/qed_dev.c   |  20 ++--
 drivers/net/ethernet/qlogic/qed/qed_hsi.h   |  42 +++++++-
 drivers/net/ethernet/qlogic/qed/qed_iscsi.c |   5 +-
 drivers/net/ethernet/qlogic/qed/qed_iwarp.c |   8 +-
 drivers/net/ethernet/qlogic/qed/qed_ll2.c   | 149 ++++++++++++++++++++++++----
 drivers/net/ethernet/qlogic/qed/qed_ll2.h   |  14 +++
 drivers/net/ethernet/qlogic/qed/qed_mcp.c   |   5 +-
 include/linux/qed/common_hsi.h              |  15 ++-
 include/linux/qed/qed_if.h                  |   4 +-
 include/linux/qed/qed_ll2_if.h              |   7 ++
 11 files changed, 235 insertions(+), 37 deletions(-)

diff --git a/drivers/net/ethernet/qlogic/qed/qed.h b/drivers/net/ethernet/qlogic/qed/qed.h
index 8ec46bf409c2..cfa45fb9a5a4 100644
--- a/drivers/net/ethernet/qlogic/qed/qed.h
+++ b/drivers/net/ethernet/qlogic/qed/qed.h
@@ -253,7 +253,8 @@ enum qed_resources {
 	QED_VLAN,
 	QED_RDMA_CNQ_RAM,
 	QED_ILT,
-	QED_LL2_QUEUE,
+	QED_LL2_RAM_QUEUE,
+	QED_LL2_CTX_QUEUE,
 	QED_CMDQS_CQS,
 	QED_RDMA_STATS_QUEUE,
 	QED_BDQ,
diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c
index 479d98e6187a..898c1f8d1530 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_dev.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c
@@ -3565,8 +3565,10 @@ const char *qed_hw_get_resc_name(enum qed_resources res_id)
 		return "RDMA_CNQ_RAM";
 	case QED_ILT:
 		return "ILT";
-	case QED_LL2_QUEUE:
-		return "LL2_QUEUE";
+	case QED_LL2_RAM_QUEUE:
+		return "LL2_RAM_QUEUE";
+	case QED_LL2_CTX_QUEUE:
+		return "LL2_CTX_QUEUE";
 	case QED_CMDQS_CQS:
 		return "CMDQS_CQS";
 	case QED_RDMA_STATS_QUEUE:
@@ -3615,8 +3617,11 @@ qed_hw_set_soft_resc_size(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt)
 
 	for (res_id = 0; res_id < QED_MAX_RESC; res_id++) {
 		switch (res_id) {
-		case QED_LL2_QUEUE:
-			resc_max_val = MAX_NUM_LL2_RX_QUEUES;
+		case QED_LL2_RAM_QUEUE:
+			resc_max_val = MAX_NUM_LL2_RX_RAM_QUEUES;
+			break;
+		case QED_LL2_CTX_QUEUE:
+			resc_max_val = MAX_NUM_LL2_RX_CTX_QUEUES;
 			break;
 		case QED_RDMA_CNQ_RAM:
 			/* No need for a case for QED_CMDQS_CQS since
@@ -3691,8 +3696,11 @@ int qed_hw_get_dflt_resc(struct qed_hwfn *p_hwfn,
 		*p_resc_num = (b_ah ? PXP_NUM_ILT_RECORDS_K2 :
 			       PXP_NUM_ILT_RECORDS_BB) / num_funcs;
 		break;
-	case QED_LL2_QUEUE:
-		*p_resc_num = MAX_NUM_LL2_RX_QUEUES / num_funcs;
+	case QED_LL2_RAM_QUEUE:
+		*p_resc_num = MAX_NUM_LL2_RX_RAM_QUEUES / num_funcs;
+		break;
+	case QED_LL2_CTX_QUEUE:
+		*p_resc_num = MAX_NUM_LL2_RX_CTX_QUEUES / num_funcs;
 		break;
 	case QED_RDMA_CNQ_RAM:
 	case QED_CMDQS_CQS:
diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
index cced2ce365de..f0e5195cae01 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
@@ -98,6 +98,7 @@ enum core_event_opcode {
 	CORE_EVENT_RX_QUEUE_STOP,
 	CORE_EVENT_RX_QUEUE_FLUSH,
 	CORE_EVENT_TX_QUEUE_UPDATE,
+	CORE_EVENT_QUEUE_STATS_QUERY,
 	MAX_CORE_EVENT_OPCODE
 };
 
@@ -116,7 +117,7 @@ struct core_ll2_port_stats {
 	struct regpair gsi_crcchksm_error;
 };
 
-/* Ethernet TX Per Queue Stats */
+/* LL2 TX Per Queue Stats */
 struct core_ll2_pstorm_per_queue_stat {
 	struct regpair sent_ucast_bytes;
 	struct regpair sent_mcast_bytes;
@@ -124,13 +125,13 @@ struct core_ll2_pstorm_per_queue_stat {
 	struct regpair sent_ucast_pkts;
 	struct regpair sent_mcast_pkts;
 	struct regpair sent_bcast_pkts;
+	struct regpair error_drop_pkts;
 };
 
 /* Light-L2 RX Producers in Tstorm RAM */
 struct core_ll2_rx_prod {
 	__le16 bd_prod;
 	__le16 cqe_prod;
-	__le32 reserved;
 };
 
 struct core_ll2_tstorm_per_queue_stat {
@@ -147,6 +148,18 @@ struct core_ll2_ustorm_per_queue_stat {
 	struct regpair rcv_bcast_pkts;
 };
 
+/* Structure for doorbell data, in PWM mode, for RX producers update. */
+struct core_pwm_prod_update_data {
+	__le16 icid; /* internal CID */
+	u8 reserved0;
+	u8 params;
+#define CORE_PWM_PROD_UPDATE_DATA_AGG_CMD_MASK	  0x3
+#define CORE_PWM_PROD_UPDATE_DATA_AGG_CMD_SHIFT   0
+#define CORE_PWM_PROD_UPDATE_DATA_RESERVED1_MASK  0x3F	/* Set 0 */
+#define CORE_PWM_PROD_UPDATE_DATA_RESERVED1_SHIFT 2
+	struct core_ll2_rx_prod prod; /* Producers */
+};
+
 /* Core Ramrod Command IDs (light L2) */
 enum core_ramrod_cmd_id {
 	CORE_RAMROD_UNUSED,
@@ -156,6 +169,7 @@ enum core_ramrod_cmd_id {
 	CORE_RAMROD_TX_QUEUE_STOP,
 	CORE_RAMROD_RX_QUEUE_FLUSH,
 	CORE_RAMROD_TX_QUEUE_UPDATE,
+	CORE_RAMROD_QUEUE_STATS_QUERY,
 	MAX_CORE_RAMROD_CMD_ID
 };
 
@@ -274,8 +288,11 @@ struct core_rx_start_ramrod_data {
 	u8 mf_si_mcast_accept_all;
 	struct core_rx_action_on_error action_on_error;
 	u8 gsi_offload_flag;
+	u8 vport_id_valid;
+	u8 vport_id;
+	u8 zero_prod_flg;
 	u8 wipe_inner_vlan_pri_en;
-	u8 reserved[5];
+	u8 reserved[2];
 };
 
 /* Ramrod data for rx queue stop ramrod */
@@ -352,8 +369,11 @@ struct core_tx_start_ramrod_data {
 	__le16 pbl_size;
 	__le16 qm_pq_id;
 	u8 gsi_offload_flag;
+	u8 ctx_stats_en;
+	u8 vport_id_valid;
 	u8 vport_id;
-	u8 resrved[2];
+	u8 enforce_security_flag;
+	u8 reserved[7];
 };
 
 /* Ramrod data for tx queue stop ramrod */
@@ -761,7 +781,7 @@ struct e4_tstorm_core_conn_ag_ctx {
 	__le16 word1;
 	__le16 word2;
 	__le16 word3;
-	__le32 reg9;
+	__le32 ll2_rx_prod;
 	__le32 reg10;
 };
 
@@ -844,6 +864,11 @@ struct ustorm_core_conn_st_ctx {
 	__le32 reserved[4];
 };
 
+/* The core storm context for the Tstorm */
+struct tstorm_core_conn_st_ctx {
+	__le32 reserved[4];
+};
+
 /* core connection context */
 struct e4_core_conn_context {
 	struct ystorm_core_conn_st_ctx ystorm_st_context;
@@ -857,6 +882,8 @@ struct e4_core_conn_context {
 	struct mstorm_core_conn_st_ctx mstorm_st_context;
 	struct ustorm_core_conn_st_ctx ustorm_st_context;
 	struct regpair ustorm_st_padding[2];
+	struct tstorm_core_conn_st_ctx tstorm_st_context;
+	struct regpair tstorm_st_padding[2];
 };
 
 struct eth_mstorm_per_pf_stat {
@@ -12483,6 +12510,11 @@ enum resource_id_enum {
 	RESOURCE_LL2_QUEUE_E = 15,
 	RESOURCE_RDMA_STATS_QUEUE_E = 16,
 	RESOURCE_BDQ_E = 17,
+	RESOURCE_QCN_E = 18,
+	RESOURCE_LLH_FILTER_E = 19,
+	RESOURCE_VF_MAC_ADDR = 20,
+	RESOURCE_LL2_CQS_E = 21,
+	RESOURCE_VF_CNQS = 22,
 	RESOURCE_MAX_NUM,
 	RESOURCE_NUM_INVALID = 0xFFFFFFFF
 };
diff --git a/drivers/net/ethernet/qlogic/qed/qed_iscsi.c b/drivers/net/ethernet/qlogic/qed/qed_iscsi.c
index 5585c18053ec..8ee668d66b4b 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_iscsi.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_iscsi.c
@@ -213,8 +213,9 @@ qed_sp_iscsi_func_start(struct qed_hwfn *p_hwfn,
 	p_init->num_sq_pages_in_ring = p_params->num_sq_pages_in_ring;
 	p_init->num_r2tq_pages_in_ring = p_params->num_r2tq_pages_in_ring;
 	p_init->num_uhq_pages_in_ring = p_params->num_uhq_pages_in_ring;
-	p_init->ll2_rx_queue_id = p_hwfn->hw_info.resc_start[QED_LL2_QUEUE] +
-				  p_params->ll2_ooo_queue_id;
+	p_init->ll2_rx_queue_id =
+	    p_hwfn->hw_info.resc_start[QED_LL2_RAM_QUEUE] +
+	    p_params->ll2_ooo_queue_id;
 
 	p_init->func_params.log_page_size = p_params->log_page_size;
 	val = p_params->num_tasks;
diff --git a/drivers/net/ethernet/qlogic/qed/qed_iwarp.c b/drivers/net/ethernet/qlogic/qed/qed_iwarp.c
index 65ec16a31658..d2fe61a5cf56 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_iwarp.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_iwarp.c
@@ -137,8 +137,8 @@ qed_iwarp_init_fw_ramrod(struct qed_hwfn *p_hwfn,
 			 struct iwarp_init_func_ramrod_data *p_ramrod)
 {
 	p_ramrod->iwarp.ll2_ooo_q_index =
-		RESC_START(p_hwfn, QED_LL2_QUEUE) +
-		p_hwfn->p_rdma_info->iwarp.ll2_ooo_handle;
+	    RESC_START(p_hwfn, QED_LL2_RAM_QUEUE) +
+	    p_hwfn->p_rdma_info->iwarp.ll2_ooo_handle;
 
 	p_ramrod->tcp.max_fin_rt = QED_IWARP_MAX_FIN_RT_DEFAULT;
 
@@ -2651,6 +2651,8 @@ qed_iwarp_ll2_start(struct qed_hwfn *p_hwfn,
 
 	memset(&data, 0, sizeof(data));
 	data.input.conn_type = QED_LL2_TYPE_IWARP;
+	/* SYN will use ctx based queues */
+	data.input.rx_conn_type = QED_LL2_RX_TYPE_CTX;
 	data.input.mtu = params->max_mtu;
 	data.input.rx_num_desc = QED_IWARP_LL2_SYN_RX_SIZE;
 	data.input.tx_num_desc = QED_IWARP_LL2_SYN_TX_SIZE;
@@ -2683,6 +2685,8 @@ qed_iwarp_ll2_start(struct qed_hwfn *p_hwfn,
 
 	/* Start OOO connection */
 	data.input.conn_type = QED_LL2_TYPE_OOO;
+	/* OOO/unaligned will use legacy ll2 queues (ram based) */
+	data.input.rx_conn_type = QED_LL2_RX_TYPE_LEGACY;
 	data.input.mtu = params->max_mtu;
 
 	n_ooo_bufs = (QED_IWARP_MAX_OOO * rcv_wnd_size) /
diff --git a/drivers/net/ethernet/qlogic/qed/qed_ll2.c b/drivers/net/ethernet/qlogic/qed/qed_ll2.c
index 19a1a58d60f8..037e5978787e 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_ll2.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_ll2.c
@@ -962,7 +962,7 @@ static int qed_sp_ll2_rx_queue_start(struct qed_hwfn *p_hwfn,
 		return rc;
 
 	p_ramrod = &p_ent->ramrod.core_rx_queue_start;
-
+	memset(p_ramrod, 0, sizeof(*p_ramrod));
 	p_ramrod->sb_id = cpu_to_le16(qed_int_get_sp_sb_id(p_hwfn));
 	p_ramrod->sb_index = p_rx->rx_sb_index;
 	p_ramrod->complete_event_flg = 1;
@@ -996,6 +996,8 @@ static int qed_sp_ll2_rx_queue_start(struct qed_hwfn *p_hwfn,
 
 	p_ramrod->action_on_error.error_type = action_on_error;
 	p_ramrod->gsi_offload_flag = p_ll2_conn->input.gsi_enable;
+	p_ramrod->zero_prod_flg = 1;
+
 	return qed_spq_post(p_hwfn, p_ent, NULL);
 }
 
@@ -1317,6 +1319,25 @@ qed_ll2_set_cbs(struct qed_ll2_info *p_ll2_info, const struct qed_ll2_cbs *cbs)
 	return 0;
 }
 
+static void _qed_ll2_calc_allowed_conns(struct qed_hwfn *p_hwfn,
+					struct qed_ll2_acquire_data *data,
+					u8 *start_idx, u8 *last_idx)
+{
+	/* LL2 queues handles will be split as follows:
+	 * First will be the legacy queues, and then the ctx based.
+	 */
+	if (data->input.rx_conn_type == QED_LL2_RX_TYPE_LEGACY) {
+		*start_idx = QED_LL2_LEGACY_CONN_BASE_PF;
+		*last_idx = *start_idx +
+			QED_MAX_NUM_OF_LEGACY_LL2_CONNS_PF;
+	} else {
+		/* QED_LL2_RX_TYPE_CTX */
+		*start_idx = QED_LL2_CTX_CONN_BASE_PF;
+		*last_idx = *start_idx +
+			QED_MAX_NUM_OF_CTX_LL2_CONNS_PF;
+	}
+}
+
 static enum core_error_handle
 qed_ll2_get_error_choice(enum qed_ll2_error_handle err)
 {
@@ -1337,14 +1358,16 @@ int qed_ll2_acquire_connection(void *cxt, struct qed_ll2_acquire_data *data)
 	struct qed_hwfn *p_hwfn = cxt;
 	qed_int_comp_cb_t comp_rx_cb, comp_tx_cb;
 	struct qed_ll2_info *p_ll2_info = NULL;
-	u8 i, *p_tx_max;
+	u8 i, first_idx, last_idx, *p_tx_max;
 	int rc;
 
 	if (!data->p_connection_handle || !p_hwfn->p_ll2_info)
 		return -EINVAL;
 
+	_qed_ll2_calc_allowed_conns(p_hwfn, data, &first_idx, &last_idx);
+
 	/* Find a free connection to be used */
-	for (i = 0; (i < QED_MAX_NUM_OF_LL2_CONNECTIONS); i++) {
+	for (i = first_idx; i < last_idx; i++) {
 		mutex_lock(&p_hwfn->p_ll2_info[i].mutex);
 		if (p_hwfn->p_ll2_info[i].b_active) {
 			mutex_unlock(&p_hwfn->p_ll2_info[i].mutex);
@@ -1448,6 +1471,7 @@ static int qed_ll2_establish_connection_rx(struct qed_hwfn *p_hwfn,
 	enum qed_ll2_error_handle error_input;
 	enum core_error_handle error_mode;
 	u8 action_on_error = 0;
+	int rc;
 
 	if (!QED_LL2_RX_REGISTERED(p_ll2_conn))
 		return 0;
@@ -1461,7 +1485,18 @@ static int qed_ll2_establish_connection_rx(struct qed_hwfn *p_hwfn,
 	error_mode = qed_ll2_get_error_choice(error_input);
 	SET_FIELD(action_on_error, CORE_RX_ACTION_ON_ERROR_NO_BUFF, error_mode);
 
-	return qed_sp_ll2_rx_queue_start(p_hwfn, p_ll2_conn, action_on_error);
+	rc = qed_sp_ll2_rx_queue_start(p_hwfn, p_ll2_conn, action_on_error);
+	if (rc)
+		return rc;
+
+	if (p_ll2_conn->rx_queue.ctx_based) {
+		rc = qed_db_recovery_add(p_hwfn->cdev,
+					 p_ll2_conn->rx_queue.set_prod_addr,
+					 &p_ll2_conn->rx_queue.db_data,
+					 DB_REC_WIDTH_64B, DB_REC_KERNEL);
+	}
+
+	return rc;
 }
 
 static void
@@ -1475,13 +1510,41 @@ qed_ll2_establish_connection_ooo(struct qed_hwfn *p_hwfn,
 	qed_ooo_submit_rx_buffers(p_hwfn, p_ll2_conn);
 }
 
+static inline u8 qed_ll2_handle_to_queue_id(struct qed_hwfn *p_hwfn,
+					    u8 handle,
+					    u8 ll2_queue_type)
+{
+	u8 qid;
+
+	if (ll2_queue_type == QED_LL2_RX_TYPE_LEGACY)
+		return p_hwfn->hw_info.resc_start[QED_LL2_RAM_QUEUE] + handle;
+
+	/* QED_LL2_RX_TYPE_CTX
+	 * FW distinguishes between the legacy queues (ram based) and the
+	 * ctx based queues by the queue_id.
+	 * The first MAX_NUM_LL2_RX_RAM_QUEUES queues are legacy
+	 * and the queue ids above that are ctx base.
+	 */
+	qid = p_hwfn->hw_info.resc_start[QED_LL2_CTX_QUEUE] +
+	      MAX_NUM_LL2_RX_RAM_QUEUES;
+
+	/* See comment on the acquire connection for how the ll2
+	 * queues handles are divided.
+	 */
+	qid += (handle - QED_MAX_NUM_OF_LEGACY_LL2_CONNS_PF);
+
+	return qid;
+}
+
 int qed_ll2_establish_connection(void *cxt, u8 connection_handle)
 {
-	struct qed_hwfn *p_hwfn = cxt;
-	struct qed_ll2_info *p_ll2_conn;
+	struct e4_core_conn_context *p_cxt;
 	struct qed_ll2_tx_packet *p_pkt;
+	struct qed_ll2_info *p_ll2_conn;
+	struct qed_hwfn *p_hwfn = cxt;
 	struct qed_ll2_rx_queue *p_rx;
 	struct qed_ll2_tx_queue *p_tx;
+	struct qed_cxt_info cxt_info;
 	struct qed_ptt *p_ptt;
 	int rc = -EINVAL;
 	u32 i, capacity;
@@ -1539,13 +1602,46 @@ int qed_ll2_establish_connection(void *cxt, u8 connection_handle)
 	rc = qed_cxt_acquire_cid(p_hwfn, PROTOCOLID_CORE, &p_ll2_conn->cid);
 	if (rc)
 		goto out;
+	cxt_info.iid = p_ll2_conn->cid;
+	rc = qed_cxt_get_cid_info(p_hwfn, &cxt_info);
+	if (rc) {
+		DP_NOTICE(p_hwfn, "Cannot find context info for cid=%d\n",
+			  p_ll2_conn->cid);
+		goto out;
+	}
+
+	p_cxt = cxt_info.p_cxt;
+
+	memset(p_cxt, 0, sizeof(*p_cxt));
 
-	qid = p_hwfn->hw_info.resc_start[QED_LL2_QUEUE] + connection_handle;
+	qid = qed_ll2_handle_to_queue_id(p_hwfn, connection_handle,
+					 p_ll2_conn->input.rx_conn_type);
 	p_ll2_conn->queue_id = qid;
 	p_ll2_conn->tx_stats_id = qid;
-	p_rx->set_prod_addr = (u8 __iomem *)p_hwfn->regview +
-					    GTT_BAR0_MAP_REG_TSDM_RAM +
-					    TSTORM_LL2_RX_PRODS_OFFSET(qid);
+
+	DP_VERBOSE(p_hwfn, QED_MSG_LL2,
+		   "Establishing ll2 queue. PF %d ctx_based=%d abs qid=%d\n",
+		   p_hwfn->rel_pf_id, p_ll2_conn->input.rx_conn_type, qid);
+
+	if (p_ll2_conn->input.rx_conn_type == QED_LL2_RX_TYPE_LEGACY) {
+		p_rx->set_prod_addr = p_hwfn->regview +
+		    GTT_BAR0_MAP_REG_TSDM_RAM + TSTORM_LL2_RX_PRODS_OFFSET(qid);
+	} else {
+		/* QED_LL2_RX_TYPE_CTX - using doorbell */
+		p_rx->ctx_based = 1;
+
+		p_rx->set_prod_addr = p_hwfn->doorbells +
+			p_hwfn->dpi_start_offset +
+			DB_ADDR_SHIFT(DQ_PWM_OFFSET_TCM_LL2_PROD_UPDATE);
+
+		/* prepare db data */
+		p_rx->db_data.icid = cpu_to_le16((u16)p_ll2_conn->cid);
+		SET_FIELD(p_rx->db_data.params,
+			  CORE_PWM_PROD_UPDATE_DATA_AGG_CMD, DB_AGG_CMD_SET);
+		SET_FIELD(p_rx->db_data.params,
+			  CORE_PWM_PROD_UPDATE_DATA_RESERVED1, 0);
+	}
+
 	p_tx->doorbell_addr = (u8 __iomem *)p_hwfn->doorbells +
 					    qed_db_addr(p_ll2_conn->cid,
 							DQ_DEMS_LEGACY);
@@ -1556,7 +1652,6 @@ int qed_ll2_establish_connection(void *cxt, u8 connection_handle)
 		  DQ_XCM_CORE_TX_BD_PROD_CMD);
 	p_tx->db_msg.agg_flags = DQ_XCM_CORE_DQ_CF_CMD;
 
-
 	rc = qed_ll2_establish_connection_rx(p_hwfn, p_ll2_conn);
 	if (rc)
 		goto out;
@@ -1590,7 +1685,7 @@ static void qed_ll2_post_rx_buffer_notify_fw(struct qed_hwfn *p_hwfn,
 					     struct qed_ll2_rx_packet *p_curp)
 {
 	struct qed_ll2_rx_packet *p_posting_packet = NULL;
-	struct core_ll2_rx_prod rx_prod = { 0, 0, 0 };
+	struct core_ll2_rx_prod rx_prod = { 0, 0 };
 	bool b_notify_fw = false;
 	u16 bd_prod, cq_prod;
 
@@ -1615,13 +1710,27 @@ static void qed_ll2_post_rx_buffer_notify_fw(struct qed_hwfn *p_hwfn,
 
 	bd_prod = qed_chain_get_prod_idx(&p_rx->rxq_chain);
 	cq_prod = qed_chain_get_prod_idx(&p_rx->rcq_chain);
-	rx_prod.bd_prod = cpu_to_le16(bd_prod);
-	rx_prod.cqe_prod = cpu_to_le16(cq_prod);
+	if (p_rx->ctx_based) {
+		/* update producer by giving a doorbell */
+		p_rx->db_data.prod.bd_prod = cpu_to_le16(bd_prod);
+		p_rx->db_data.prod.cqe_prod = cpu_to_le16(cq_prod);
+		/* Make sure chain element is updated before ringing the
+		 * doorbell
+		 */
+		dma_wmb();
+		DIRECT_REG_WR64(p_rx->set_prod_addr,
+				*((u64 *)&p_rx->db_data));
+	} else {
+		rx_prod.bd_prod = cpu_to_le16(bd_prod);
+		rx_prod.cqe_prod = cpu_to_le16(cq_prod);
 
-	/* Make sure chain element is updated before ringing the doorbell */
-	dma_wmb();
+		/* Make sure chain element is updated before ringing the
+		 * doorbell
+		 */
+		dma_wmb();
 
-	DIRECT_REG_WR(p_rx->set_prod_addr, *((u32 *)&rx_prod));
+		DIRECT_REG_WR(p_rx->set_prod_addr, *((u32 *)&rx_prod));
+	}
 }
 
 int qed_ll2_post_rx_buffer(void *cxt,
@@ -1965,6 +2074,12 @@ int qed_ll2_terminate_connection(void *cxt, u8 connection_handle)
 	if (QED_LL2_RX_REGISTERED(p_ll2_conn)) {
 		p_ll2_conn->rx_queue.b_cb_registered = false;
 		smp_wmb(); /* Make sure this is seen by ll2_lb_rxq_completion */
+
+		if (p_ll2_conn->rx_queue.ctx_based)
+			qed_db_recovery_del(p_hwfn->cdev,
+					    p_ll2_conn->rx_queue.set_prod_addr,
+					    &p_ll2_conn->rx_queue.db_data);
+
 		rc = qed_sp_ll2_rx_queue_stop(p_hwfn, p_ll2_conn);
 		if (rc)
 			goto out;
diff --git a/drivers/net/ethernet/qlogic/qed/qed_ll2.h b/drivers/net/ethernet/qlogic/qed/qed_ll2.h
index 5f01fbd3c073..288642d526b7 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_ll2.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_ll2.h
@@ -46,6 +46,18 @@
 #include "qed_sp.h"
 
 #define QED_MAX_NUM_OF_LL2_CONNECTIONS                    (4)
+/* LL2 queues handles will be split as follows:
+ * first will be legacy queues, and then the ctx based queues.
+ */
+#define QED_MAX_NUM_OF_LL2_CONNS_PF            (4)
+#define QED_MAX_NUM_OF_LEGACY_LL2_CONNS_PF   (3)
+
+#define QED_MAX_NUM_OF_CTX_LL2_CONNS_PF	\
+	(QED_MAX_NUM_OF_LL2_CONNS_PF - QED_MAX_NUM_OF_LEGACY_LL2_CONNS_PF)
+
+#define QED_LL2_LEGACY_CONN_BASE_PF     0
+#define QED_LL2_CTX_CONN_BASE_PF        QED_MAX_NUM_OF_LEGACY_LL2_CONNS_PF
+
 
 struct qed_ll2_rx_packet {
 	struct list_head list_entry;
@@ -79,6 +91,7 @@ struct qed_ll2_rx_queue {
 	struct qed_chain rxq_chain;
 	struct qed_chain rcq_chain;
 	u8 rx_sb_index;
+	u8 ctx_based;
 	bool b_cb_registered;
 	__le16 *p_fw_cons;
 	struct list_head active_descq;
@@ -86,6 +99,7 @@ struct qed_ll2_rx_queue {
 	struct list_head posting_descq;
 	struct qed_ll2_rx_packet *descq_array;
 	void __iomem *set_prod_addr;
+	struct core_pwm_prod_update_data db_data;
 };
 
 struct qed_ll2_tx_queue {
diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.c b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
index 36ddb89856a8..aeac359941de 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_mcp.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
@@ -3261,9 +3261,12 @@ static enum resource_id_enum qed_mcp_get_mfw_res_id(enum qed_resources res_id)
 	case QED_ILT:
 		mfw_res_id = RESOURCE_ILT_E;
 		break;
-	case QED_LL2_QUEUE:
+	case QED_LL2_RAM_QUEUE:
 		mfw_res_id = RESOURCE_LL2_QUEUE_E;
 		break;
+	case QED_LL2_CTX_QUEUE:
+		mfw_res_id = RESOURCE_LL2_CQS_E;
+		break;
 	case QED_RDMA_CNQ_RAM:
 	case QED_CMDQS_CQS:
 		/* CNQ/CMDQS are the same resource */
diff --git a/include/linux/qed/common_hsi.h b/include/linux/qed/common_hsi.h
index 3f437e826a4c..a2b7826b36f0 100644
--- a/include/linux/qed/common_hsi.h
+++ b/include/linux/qed/common_hsi.h
@@ -105,8 +105,15 @@
 
 #define CORE_SPQE_PAGE_SIZE_BYTES	4096
 
-#define MAX_NUM_LL2_RX_QUEUES		48
-#define MAX_NUM_LL2_TX_STATS_COUNTERS	48
+/* Number of LL2 RAM based queues */
+#define MAX_NUM_LL2_RX_RAM_QUEUES 32
+
+/* Number of LL2 context based queues */
+#define MAX_NUM_LL2_RX_CTX_QUEUES 208
+#define MAX_NUM_LL2_RX_QUEUES \
+	(MAX_NUM_LL2_RX_RAM_QUEUES + MAX_NUM_LL2_RX_CTX_QUEUES)
+
+#define MAX_NUM_LL2_TX_STATS_COUNTERS  48
 
 #define FW_MAJOR_VERSION	8
 #define FW_MINOR_VERSION	42
@@ -340,6 +347,10 @@
 #define DQ_PWM_OFFSET_TCM_ROCE_RQ_PROD		(DQ_PWM_OFFSET_TCM16_BASE + 1)
 #define DQ_PWM_OFFSET_TCM_IWARP_RQ_PROD		(DQ_PWM_OFFSET_TCM16_BASE + 3)
 
+/* DQ_DEMS_AGG_VAL_BASE */
+#define DQ_PWM_OFFSET_TCM_LL2_PROD_UPDATE \
+	(DQ_PWM_OFFSET_TCM32_BASE + DQ_TCM_AGG_VAL_SEL_REG9 - 4)
+
 #define	DQ_REGION_SHIFT			(12)
 
 /* DPM */
diff --git a/include/linux/qed/qed_if.h b/include/linux/qed/qed_if.h
index b5db1ee96d78..9bcb2f419004 100644
--- a/include/linux/qed/qed_if.h
+++ b/include/linux/qed/qed_if.h
@@ -463,7 +463,7 @@ enum qed_db_rec_space {
 
 #define DIRECT_REG_RD(reg_addr) readl((void __iomem *)(reg_addr))
 
-#define DIRECT_REG_WR64(reg_addr, val) writeq((u32)val,	\
+#define DIRECT_REG_WR64(reg_addr, val) writeq((u64)val,	\
 					      (void __iomem *)(reg_addr))
 
 #define QED_COALESCE_MAX 0x1FF
@@ -1177,6 +1177,8 @@ struct qed_common_ops {
 #define GET_FIELD(value, name) \
 	(((value) >> (name ## _SHIFT)) & name ## _MASK)
 
+#define DB_ADDR_SHIFT(addr) ((addr) << DB_PWM_ADDR_OFFSET_SHIFT)
+
 /* Debug print definitions */
 #define DP_ERR(cdev, fmt, ...)					\
 	do {							\
diff --git a/include/linux/qed/qed_ll2_if.h b/include/linux/qed/qed_ll2_if.h
index 5eb022953aca..1313c34d9a68 100644
--- a/include/linux/qed/qed_ll2_if.h
+++ b/include/linux/qed/qed_ll2_if.h
@@ -52,6 +52,12 @@ enum qed_ll2_conn_type {
 	QED_LL2_TYPE_ROCE,
 	QED_LL2_TYPE_IWARP,
 	QED_LL2_TYPE_RESERVED3,
+	MAX_QED_LL2_CONN_TYPE
+};
+
+enum qed_ll2_rx_conn_type {
+	QED_LL2_RX_TYPE_LEGACY,
+	QED_LL2_RX_TYPE_CTX,
 	MAX_QED_LL2_RX_CONN_TYPE
 };
 
@@ -165,6 +171,7 @@ struct qed_ll2_cbs {
 };
 
 struct qed_ll2_acquire_data_inputs {
+	enum qed_ll2_rx_conn_type rx_conn_type;
 	enum qed_ll2_conn_type conn_type;
 	u16 mtu;
 	u16 rx_num_desc;
-- 
cgit v1.2.3-59-g8ed1b


From 1392d19ff1d6ddd370cefa73b552a0262f9c35ea Mon Sep 17 00:00:00 2001
From: Michal Kalderon <michal.kalderon@marvell.com>
Date: Mon, 27 Jan 2020 15:26:13 +0200
Subject: qed: Add abstraction for different hsi values per chip

The number of BTB blocks was modified to be different between the two chip
flavors supported (BB/K2) as a result, this lead to a re-write of selecting
the default hsi value based on the chip.
This patch creates a lookup table for hsi values per chip rather than
ask again and again for every value.

Signed-off-by: Ariel Elior <ariel.elior@marvell.com>
Signed-off-by: Michal Kalderon <michal.kalderon@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/qlogic/qed/qed.h     | 56 +++++++++++++++++++++++-----
 drivers/net/ethernet/qlogic/qed/qed_dev.c | 62 +++++++++++++++++++++----------
 include/linux/qed/common_hsi.h            |  4 +-
 3 files changed, 90 insertions(+), 32 deletions(-)

diff --git a/drivers/net/ethernet/qlogic/qed/qed.h b/drivers/net/ethernet/qlogic/qed/qed.h
index cfa45fb9a5a4..cfbfe7441ecb 100644
--- a/drivers/net/ethernet/qlogic/qed/qed.h
+++ b/drivers/net/ethernet/qlogic/qed/qed.h
@@ -532,6 +532,23 @@ struct qed_nvm_image_info {
 	bool valid;
 };
 
+enum qed_hsi_def_type {
+	QED_HSI_DEF_MAX_NUM_VFS,
+	QED_HSI_DEF_MAX_NUM_L2_QUEUES,
+	QED_HSI_DEF_MAX_NUM_PORTS,
+	QED_HSI_DEF_MAX_SB_PER_PATH,
+	QED_HSI_DEF_MAX_NUM_PFS,
+	QED_HSI_DEF_MAX_NUM_VPORTS,
+	QED_HSI_DEF_NUM_ETH_RSS_ENGINE,
+	QED_HSI_DEF_MAX_QM_TX_QUEUES,
+	QED_HSI_DEF_NUM_PXP_ILT_RECORDS,
+	QED_HSI_DEF_NUM_RDMA_STATISTIC_COUNTERS,
+	QED_HSI_DEF_MAX_QM_GLOBAL_RLS,
+	QED_HSI_DEF_MAX_PBF_CMD_LINES,
+	QED_HSI_DEF_MAX_BTB_BLOCKS,
+	QED_NUM_HSI_DEFS
+};
+
 #define DRV_MODULE_VERSION		      \
 	__stringify(QED_MAJOR_VERSION) "."    \
 	__stringify(QED_MINOR_VERSION) "."    \
@@ -869,16 +886,35 @@ struct qed_dev {
 	bool				iwarp_cmt;
 };
 
-#define NUM_OF_VFS(dev)         (QED_IS_BB(dev) ? MAX_NUM_VFS_BB \
-						: MAX_NUM_VFS_K2)
-#define NUM_OF_L2_QUEUES(dev)   (QED_IS_BB(dev) ? MAX_NUM_L2_QUEUES_BB \
-						: MAX_NUM_L2_QUEUES_K2)
-#define NUM_OF_PORTS(dev)       (QED_IS_BB(dev) ? MAX_NUM_PORTS_BB \
-						: MAX_NUM_PORTS_K2)
-#define NUM_OF_SBS(dev)         (QED_IS_BB(dev) ? MAX_SB_PER_PATH_BB \
-						: MAX_SB_PER_PATH_K2)
-#define NUM_OF_ENG_PFS(dev)     (QED_IS_BB(dev) ? MAX_NUM_PFS_BB \
-						: MAX_NUM_PFS_K2)
+u32 qed_get_hsi_def_val(struct qed_dev *cdev, enum qed_hsi_def_type type);
+
+#define NUM_OF_VFS(dev)	\
+	qed_get_hsi_def_val(dev, QED_HSI_DEF_MAX_NUM_VFS)
+#define NUM_OF_L2_QUEUES(dev) \
+	qed_get_hsi_def_val(dev, QED_HSI_DEF_MAX_NUM_L2_QUEUES)
+#define NUM_OF_PORTS(dev) \
+	qed_get_hsi_def_val(dev, QED_HSI_DEF_MAX_NUM_PORTS)
+#define NUM_OF_SBS(dev)	\
+	qed_get_hsi_def_val(dev, QED_HSI_DEF_MAX_SB_PER_PATH)
+#define NUM_OF_ENG_PFS(dev) \
+	qed_get_hsi_def_val(dev, QED_HSI_DEF_MAX_NUM_PFS)
+#define NUM_OF_VPORTS(dev) \
+	qed_get_hsi_def_val(dev, QED_HSI_DEF_MAX_NUM_VPORTS)
+#define NUM_OF_RSS_ENGINES(dev)	\
+	qed_get_hsi_def_val(dev, QED_HSI_DEF_NUM_ETH_RSS_ENGINE)
+#define NUM_OF_QM_TX_QUEUES(dev) \
+	qed_get_hsi_def_val(dev, QED_HSI_DEF_MAX_QM_TX_QUEUES)
+#define NUM_OF_PXP_ILT_RECORDS(dev) \
+	qed_get_hsi_def_val(dev, QED_HSI_DEF_NUM_PXP_ILT_RECORDS)
+#define NUM_OF_RDMA_STATISTIC_COUNTERS(dev) \
+	qed_get_hsi_def_val(dev, QED_HSI_DEF_NUM_RDMA_STATISTIC_COUNTERS)
+#define NUM_OF_QM_GLOBAL_RLS(dev) \
+	qed_get_hsi_def_val(dev, QED_HSI_DEF_MAX_QM_GLOBAL_RLS)
+#define NUM_OF_PBF_CMD_LINES(dev) \
+	qed_get_hsi_def_val(dev, QED_HSI_DEF_MAX_PBF_CMD_LINES)
+#define NUM_OF_BTB_BLOCKS(dev) \
+	qed_get_hsi_def_val(dev, QED_HSI_DEF_MAX_BTB_BLOCKS)
+
 
 /**
  * @brief qed_concrete_to_sw_fid - get the sw function id from
diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c
index 898c1f8d1530..e3e0376c13d6 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_dev.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c
@@ -1579,6 +1579,7 @@ static void qed_init_qm_port_params(struct qed_hwfn *p_hwfn)
 {
 	/* Initialize qm port parameters */
 	u8 i, active_phys_tcs, num_ports = p_hwfn->cdev->num_ports_in_engine;
+	struct qed_dev *cdev = p_hwfn->cdev;
 
 	/* indicate how ooo and high pri traffic is dealt with */
 	active_phys_tcs = num_ports == MAX_NUM_PORTS_K2 ?
@@ -1588,11 +1589,13 @@ static void qed_init_qm_port_params(struct qed_hwfn *p_hwfn)
 	for (i = 0; i < num_ports; i++) {
 		struct init_qm_port_params *p_qm_port =
 		    &p_hwfn->qm_info.qm_port_params[i];
+		u16 pbf_max_cmd_lines;
 
 		p_qm_port->active = 1;
 		p_qm_port->active_phys_tcs = active_phys_tcs;
-		p_qm_port->num_pbf_cmd_lines = PBF_MAX_CMD_LINES / num_ports;
-		p_qm_port->num_btb_blocks = BTB_MAX_BLOCKS / num_ports;
+		pbf_max_cmd_lines = (u16)NUM_OF_PBF_CMD_LINES(cdev);
+		p_qm_port->num_pbf_cmd_lines = pbf_max_cmd_lines / num_ports;
+		p_qm_port->num_btb_blocks = NUM_OF_BTB_BLOCKS(cdev) / num_ports;
 	}
 }
 
@@ -3607,14 +3610,39 @@ __qed_hw_set_soft_resc_size(struct qed_hwfn *p_hwfn,
 	return 0;
 }
 
+static u32 qed_hsi_def_val[][MAX_CHIP_IDS] = {
+	{MAX_NUM_VFS_BB, MAX_NUM_VFS_K2},
+	{MAX_NUM_L2_QUEUES_BB, MAX_NUM_L2_QUEUES_K2},
+	{MAX_NUM_PORTS_BB, MAX_NUM_PORTS_K2},
+	{MAX_SB_PER_PATH_BB, MAX_SB_PER_PATH_K2,},
+	{MAX_NUM_PFS_BB, MAX_NUM_PFS_K2},
+	{MAX_NUM_VPORTS_BB, MAX_NUM_VPORTS_K2},
+	{ETH_RSS_ENGINE_NUM_BB, ETH_RSS_ENGINE_NUM_K2},
+	{MAX_QM_TX_QUEUES_BB, MAX_QM_TX_QUEUES_K2},
+	{PXP_NUM_ILT_RECORDS_BB, PXP_NUM_ILT_RECORDS_K2},
+	{RDMA_NUM_STATISTIC_COUNTERS_BB, RDMA_NUM_STATISTIC_COUNTERS_K2},
+	{MAX_QM_GLOBAL_RLS, MAX_QM_GLOBAL_RLS},
+	{PBF_MAX_CMD_LINES, PBF_MAX_CMD_LINES},
+	{BTB_MAX_BLOCKS_BB, BTB_MAX_BLOCKS_K2},
+};
+
+u32 qed_get_hsi_def_val(struct qed_dev *cdev, enum qed_hsi_def_type type)
+{
+	enum chip_ids chip_id = QED_IS_BB(cdev) ? CHIP_BB : CHIP_K2;
+
+	if (type >= QED_NUM_HSI_DEFS) {
+		DP_ERR(cdev, "Unexpected HSI definition type [%d]\n", type);
+		return 0;
+	}
+
+	return qed_hsi_def_val[type][chip_id];
+}
 static int
 qed_hw_set_soft_resc_size(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt)
 {
-	bool b_ah = QED_IS_AH(p_hwfn->cdev);
 	u32 resc_max_val, mcp_resp;
 	u8 res_id;
 	int rc;
-
 	for (res_id = 0; res_id < QED_MAX_RESC; res_id++) {
 		switch (res_id) {
 		case QED_LL2_RAM_QUEUE:
@@ -3630,8 +3658,8 @@ qed_hw_set_soft_resc_size(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt)
 			resc_max_val = NUM_OF_GLOBAL_QUEUES;
 			break;
 		case QED_RDMA_STATS_QUEUE:
-			resc_max_val = b_ah ? RDMA_NUM_STATISTIC_COUNTERS_K2
-			    : RDMA_NUM_STATISTIC_COUNTERS_BB;
+			resc_max_val =
+			    NUM_OF_RDMA_STATISTIC_COUNTERS(p_hwfn->cdev);
 			break;
 		case QED_BDQ:
 			resc_max_val = BDQ_NUM_RESOURCES;
@@ -3664,28 +3692,24 @@ int qed_hw_get_dflt_resc(struct qed_hwfn *p_hwfn,
 			 u32 *p_resc_num, u32 *p_resc_start)
 {
 	u8 num_funcs = p_hwfn->num_funcs_on_engine;
-	bool b_ah = QED_IS_AH(p_hwfn->cdev);
+	struct qed_dev *cdev = p_hwfn->cdev;
 
 	switch (res_id) {
 	case QED_L2_QUEUE:
-		*p_resc_num = (b_ah ? MAX_NUM_L2_QUEUES_K2 :
-			       MAX_NUM_L2_QUEUES_BB) / num_funcs;
+		*p_resc_num = NUM_OF_L2_QUEUES(cdev) / num_funcs;
 		break;
 	case QED_VPORT:
-		*p_resc_num = (b_ah ? MAX_NUM_VPORTS_K2 :
-			       MAX_NUM_VPORTS_BB) / num_funcs;
+		*p_resc_num = NUM_OF_VPORTS(cdev) / num_funcs;
 		break;
 	case QED_RSS_ENG:
-		*p_resc_num = (b_ah ? ETH_RSS_ENGINE_NUM_K2 :
-			       ETH_RSS_ENGINE_NUM_BB) / num_funcs;
+		*p_resc_num = NUM_OF_RSS_ENGINES(cdev) / num_funcs;
 		break;
 	case QED_PQ:
-		*p_resc_num = (b_ah ? MAX_QM_TX_QUEUES_K2 :
-			       MAX_QM_TX_QUEUES_BB) / num_funcs;
+		*p_resc_num = NUM_OF_QM_TX_QUEUES(cdev) / num_funcs;
 		*p_resc_num &= ~0x7;	/* The granularity of the PQs is 8 */
 		break;
 	case QED_RL:
-		*p_resc_num = MAX_QM_GLOBAL_RLS / num_funcs;
+		*p_resc_num = NUM_OF_QM_GLOBAL_RLS(cdev) / num_funcs;
 		break;
 	case QED_MAC:
 	case QED_VLAN:
@@ -3693,8 +3717,7 @@ int qed_hw_get_dflt_resc(struct qed_hwfn *p_hwfn,
 		*p_resc_num = ETH_NUM_MAC_FILTERS / num_funcs;
 		break;
 	case QED_ILT:
-		*p_resc_num = (b_ah ? PXP_NUM_ILT_RECORDS_K2 :
-			       PXP_NUM_ILT_RECORDS_BB) / num_funcs;
+		*p_resc_num = NUM_OF_PXP_ILT_RECORDS(cdev) / num_funcs;
 		break;
 	case QED_LL2_RAM_QUEUE:
 		*p_resc_num = MAX_NUM_LL2_RX_RAM_QUEUES / num_funcs;
@@ -3708,8 +3731,7 @@ int qed_hw_get_dflt_resc(struct qed_hwfn *p_hwfn,
 		*p_resc_num = NUM_OF_GLOBAL_QUEUES / num_funcs;
 		break;
 	case QED_RDMA_STATS_QUEUE:
-		*p_resc_num = (b_ah ? RDMA_NUM_STATISTIC_COUNTERS_K2 :
-			       RDMA_NUM_STATISTIC_COUNTERS_BB) / num_funcs;
+		*p_resc_num = NUM_OF_RDMA_STATISTIC_COUNTERS(cdev) / num_funcs;
 		break;
 	case QED_BDQ:
 		if (p_hwfn->hw_info.personality != QED_PCI_ISCSI &&
diff --git a/include/linux/qed/common_hsi.h b/include/linux/qed/common_hsi.h
index a2b7826b36f0..718ce72e5965 100644
--- a/include/linux/qed/common_hsi.h
+++ b/include/linux/qed/common_hsi.h
@@ -663,8 +663,8 @@
 #define PBF_MAX_CMD_LINES	3328
 
 /* Number of BTB blocks. Each block is 256B. */
-#define BTB_MAX_BLOCKS		1440
-
+#define BTB_MAX_BLOCKS_BB 1440
+#define BTB_MAX_BLOCKS_K2 1840
 /*****************/
 /* PRS CONSTANTS */
 /*****************/
-- 
cgit v1.2.3-59-g8ed1b


From 6459d93619b5bc21f775e7eb12bc4d051743d7aa Mon Sep 17 00:00:00 2001
From: Michal Kalderon <michal.kalderon@marvell.com>
Date: Mon, 27 Jan 2020 15:26:14 +0200
Subject: qed: FW 8.42.2.0 iscsi/fcoe changes

- Remove struct iscsi_slow_path_hdr and field fw_cid from several structs
- Remove struct iscsi_spe_func_dstry
- Remove fields pbe_page_size_log and pbl_page_size_log from struct
  iscsi_conn_offload_param

Signed-off-by: Manish Rangankar <manish.rangankar@marvell.com>
Signed-off-by: Saurav Kashyap <saurav.kashyap@marvell.com>
Signed-off-by: Ariel Elior <ariel.elior@marvell.com>
Signed-off-by: Michal Kalderon <michal.kalderon@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/qlogic/qed/qed_fcoe.c  |  2 +
 drivers/net/ethernet/qlogic/qed/qed_hsi.h   |  8 ++--
 drivers/net/ethernet/qlogic/qed/qed_iscsi.c | 31 --------------
 drivers/net/ethernet/qlogic/qed/qed_sp.h    |  2 -
 include/linux/qed/iscsi_common.h            | 64 +++++++++++------------------
 include/linux/qed/storage_common.h          |  3 +-
 6 files changed, 32 insertions(+), 78 deletions(-)

diff --git a/drivers/net/ethernet/qlogic/qed/qed_fcoe.c b/drivers/net/ethernet/qlogic/qed/qed_fcoe.c
index de31a382f58e..4c7fa391fd33 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_fcoe.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_fcoe.c
@@ -167,6 +167,8 @@ qed_sp_fcoe_func_start(struct qed_hwfn *p_hwfn,
 		goto err;
 	}
 	p_cxt = cxt_info.p_cxt;
+	memset(p_cxt, 0, sizeof(*p_cxt));
+
 	SET_FIELD(p_cxt->tstorm_ag_context.flags3,
 		  E4_TSTORM_FCOE_CONN_AG_CTX_DUMMY_TIMER_CF_EN, 1);
 
diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
index f0e5195cae01..0c3d714e6518 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
@@ -11496,8 +11496,8 @@ struct e4_tstorm_iscsi_conn_ag_ctx {
 	u8 flags3;
 #define E4_TSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q0_MASK		0x3
 #define E4_TSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q0_SHIFT		0
-#define E4_TSTORM_ISCSI_CONN_AG_CTX_CF10_MASK			0x3
-#define E4_TSTORM_ISCSI_CONN_AG_CTX_CF10_SHIFT			2
+#define E4_TSTORM_ISCSI_CONN_AG_CTX_FLUSH_OOO_ISLES_CF_MASK	0x3
+#define E4_TSTORM_ISCSI_CONN_AG_CTX_FLUSH_OOO_ISLES_CF_SHIFT	2
 #define E4_TSTORM_ISCSI_CONN_AG_CTX_CF0EN_MASK			0x1
 #define E4_TSTORM_ISCSI_CONN_AG_CTX_CF0EN_SHIFT			4
 #define E4_TSTORM_ISCSI_CONN_AG_CTX_P2T_FLUSH_CF_EN_MASK	0x1
@@ -11519,8 +11519,8 @@ struct e4_tstorm_iscsi_conn_ag_ctx {
 #define E4_TSTORM_ISCSI_CONN_AG_CTX_CF8EN_SHIFT		4
 #define E4_TSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q0_EN_MASK	0x1
 #define E4_TSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q0_EN_SHIFT	5
-#define E4_TSTORM_ISCSI_CONN_AG_CTX_CF10EN_MASK		0x1
-#define E4_TSTORM_ISCSI_CONN_AG_CTX_CF10EN_SHIFT	6
+#define E4_TSTORM_ISCSI_CONN_AG_CTX_FLUSH_OOO_ISLES_CF_EN_MASK	0x1
+#define E4_TSTORM_ISCSI_CONN_AG_CTX_FLUSH_OOO_ISLES_CF_EN_SHIFT	6
 #define E4_TSTORM_ISCSI_CONN_AG_CTX_RULE0EN_MASK	0x1
 #define E4_TSTORM_ISCSI_CONN_AG_CTX_RULE0EN_SHIFT	7
 	u8 flags5;
diff --git a/drivers/net/ethernet/qlogic/qed/qed_iscsi.c b/drivers/net/ethernet/qlogic/qed/qed_iscsi.c
index 8ee668d66b4b..7245a615517a 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_iscsi.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_iscsi.c
@@ -204,10 +204,6 @@ qed_sp_iscsi_func_start(struct qed_hwfn *p_hwfn,
 		return -EINVAL;
 	}
 
-	SET_FIELD(p_init->hdr.flags,
-		  ISCSI_SLOW_PATH_HDR_LAYER_CODE, ISCSI_SLOW_PATH_LAYER_CODE);
-	p_init->hdr.op_code = ISCSI_RAMROD_CMD_ID_INIT_FUNC;
-
 	val = p_params->half_way_close_timeout;
 	p_init->half_way_close_timeout = cpu_to_le16(val);
 	p_init->num_sq_pages_in_ring = p_params->num_sq_pages_in_ring;
@@ -332,12 +328,7 @@ static int qed_sp_iscsi_conn_offload(struct qed_hwfn *p_hwfn,
 	p_conn->physical_q1 = cpu_to_le16(physical_q);
 	p_ramrod->iscsi.physical_q1 = cpu_to_le16(physical_q);
 
-	p_ramrod->hdr.op_code = ISCSI_RAMROD_CMD_ID_OFFLOAD_CONN;
-	SET_FIELD(p_ramrod->hdr.flags, ISCSI_SLOW_PATH_HDR_LAYER_CODE,
-		  p_conn->layer_code);
-
 	p_ramrod->conn_id = cpu_to_le16(p_conn->conn_id);
-	p_ramrod->fw_cid = cpu_to_le32(p_conn->icid);
 
 	DMA_REGPAIR_LE(p_ramrod->iscsi.sq_pbl_addr, p_conn->sq_pbl_addr);
 
@@ -493,12 +484,8 @@ static int qed_sp_iscsi_conn_update(struct qed_hwfn *p_hwfn,
 		return rc;
 
 	p_ramrod = &p_ent->ramrod.iscsi_conn_update;
-	p_ramrod->hdr.op_code = ISCSI_RAMROD_CMD_ID_UPDATE_CONN;
-	SET_FIELD(p_ramrod->hdr.flags,
-		  ISCSI_SLOW_PATH_HDR_LAYER_CODE, p_conn->layer_code);
 
 	p_ramrod->conn_id = cpu_to_le16(p_conn->conn_id);
-	p_ramrod->fw_cid = cpu_to_le32(p_conn->icid);
 	p_ramrod->flags = p_conn->update_flag;
 	p_ramrod->max_seq_size = cpu_to_le32(p_conn->max_seq_size);
 	dval = p_conn->max_recv_pdu_length;
@@ -538,12 +525,8 @@ qed_sp_iscsi_mac_update(struct qed_hwfn *p_hwfn,
 		return rc;
 
 	p_ramrod = &p_ent->ramrod.iscsi_conn_mac_update;
-	p_ramrod->hdr.op_code = ISCSI_RAMROD_CMD_ID_MAC_UPDATE;
-	SET_FIELD(p_ramrod->hdr.flags,
-		  ISCSI_SLOW_PATH_HDR_LAYER_CODE, p_conn->layer_code);
 
 	p_ramrod->conn_id = cpu_to_le16(p_conn->conn_id);
-	p_ramrod->fw_cid = cpu_to_le32(p_conn->icid);
 	ucval = p_conn->remote_mac[1];
 	((u8 *)(&p_ramrod->remote_mac_addr_hi))[0] = ucval;
 	ucval = p_conn->remote_mac[0];
@@ -584,12 +567,8 @@ static int qed_sp_iscsi_conn_terminate(struct qed_hwfn *p_hwfn,
 		return rc;
 
 	p_ramrod = &p_ent->ramrod.iscsi_conn_terminate;
-	p_ramrod->hdr.op_code = ISCSI_RAMROD_CMD_ID_TERMINATION_CONN;
-	SET_FIELD(p_ramrod->hdr.flags,
-		  ISCSI_SLOW_PATH_HDR_LAYER_CODE, p_conn->layer_code);
 
 	p_ramrod->conn_id = cpu_to_le16(p_conn->conn_id);
-	p_ramrod->fw_cid = cpu_to_le32(p_conn->icid);
 	p_ramrod->abortive = p_conn->abortive_dsconnect;
 
 	DMA_REGPAIR_LE(p_ramrod->query_params_addr,
@@ -604,7 +583,6 @@ static int qed_sp_iscsi_conn_clear_sq(struct qed_hwfn *p_hwfn,
 				      enum spq_mode comp_mode,
 				      struct qed_spq_comp_cb *p_comp_addr)
 {
-	struct iscsi_slow_path_hdr *p_ramrod = NULL;
 	struct qed_spq_entry *p_ent = NULL;
 	struct qed_sp_init_data init_data;
 	int rc = -EINVAL;
@@ -622,11 +600,6 @@ static int qed_sp_iscsi_conn_clear_sq(struct qed_hwfn *p_hwfn,
 	if (rc)
 		return rc;
 
-	p_ramrod = &p_ent->ramrod.iscsi_empty;
-	p_ramrod->op_code = ISCSI_RAMROD_CMD_ID_CLEAR_SQ;
-	SET_FIELD(p_ramrod->flags,
-		  ISCSI_SLOW_PATH_HDR_LAYER_CODE, p_conn->layer_code);
-
 	return qed_spq_post(p_hwfn, p_ent, NULL);
 }
 
@@ -634,7 +607,6 @@ static int qed_sp_iscsi_func_stop(struct qed_hwfn *p_hwfn,
 				  enum spq_mode comp_mode,
 				  struct qed_spq_comp_cb *p_comp_addr)
 {
-	struct iscsi_spe_func_dstry *p_ramrod = NULL;
 	struct qed_spq_entry *p_ent = NULL;
 	struct qed_sp_init_data init_data;
 	int rc = 0;
@@ -652,9 +624,6 @@ static int qed_sp_iscsi_func_stop(struct qed_hwfn *p_hwfn,
 	if (rc)
 		return rc;
 
-	p_ramrod = &p_ent->ramrod.iscsi_destroy;
-	p_ramrod->hdr.op_code = ISCSI_RAMROD_CMD_ID_DESTROY_FUNC;
-
 	rc = qed_spq_post(p_hwfn, p_ent, NULL);
 
 	qed_spq_unregister_async_cb(p_hwfn, PROTOCOLID_ISCSI);
diff --git a/drivers/net/ethernet/qlogic/qed/qed_sp.h b/drivers/net/ethernet/qlogic/qed/qed_sp.h
index 96ab77ae6af5..b7b4fbbbccfe 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_sp.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_sp.h
@@ -120,9 +120,7 @@ union ramrod_data {
 	struct fcoe_conn_terminate_ramrod_params fcoe_conn_terminate;
 	struct fcoe_stat_ramrod_params fcoe_stat;
 
-	struct iscsi_slow_path_hdr iscsi_empty;
 	struct iscsi_init_ramrod_params iscsi_init;
-	struct iscsi_spe_func_dstry iscsi_destroy;
 	struct iscsi_spe_conn_offload iscsi_conn_offload;
 	struct iscsi_conn_update_ramrod_params iscsi_conn_update;
 	struct iscsi_spe_conn_mac_update iscsi_conn_mac_update;
diff --git a/include/linux/qed/iscsi_common.h b/include/linux/qed/iscsi_common.h
index 66aba505ec56..2f0a771a9176 100644
--- a/include/linux/qed/iscsi_common.h
+++ b/include/linux/qed/iscsi_common.h
@@ -999,7 +999,6 @@ struct iscsi_conn_offload_params {
 	struct regpair r2tq_pbl_addr;
 	struct regpair xhq_pbl_addr;
 	struct regpair uhq_pbl_addr;
-	__le32 initial_ack;
 	__le16 physical_q0;
 	__le16 physical_q1;
 	u8 flags;
@@ -1011,10 +1010,10 @@ struct iscsi_conn_offload_params {
 #define ISCSI_CONN_OFFLOAD_PARAMS_RESTRICTED_MODE_SHIFT	2
 #define ISCSI_CONN_OFFLOAD_PARAMS_RESERVED1_MASK	0x1F
 #define ISCSI_CONN_OFFLOAD_PARAMS_RESERVED1_SHIFT	3
-	u8 pbl_page_size_log;
-	u8 pbe_page_size_log;
 	u8 default_cq;
+	__le16 reserved0;
 	__le32 stat_sn;
+	__le32 initial_ack;
 };
 
 /* iSCSI connection statistics */
@@ -1029,25 +1028,14 @@ struct iscsi_conn_stats_params {
 	__le32 reserved;
 };
 
-/* spe message header */
-struct iscsi_slow_path_hdr {
-	u8 op_code;
-	u8 flags;
-#define ISCSI_SLOW_PATH_HDR_RESERVED0_MASK	0xF
-#define ISCSI_SLOW_PATH_HDR_RESERVED0_SHIFT	0
-#define ISCSI_SLOW_PATH_HDR_LAYER_CODE_MASK	0x7
-#define ISCSI_SLOW_PATH_HDR_LAYER_CODE_SHIFT	4
-#define ISCSI_SLOW_PATH_HDR_RESERVED1_MASK	0x1
-#define ISCSI_SLOW_PATH_HDR_RESERVED1_SHIFT	7
-};
 
 /* iSCSI connection update params passed by driver to FW in ISCSI update
  *ramrod.
  */
 struct iscsi_conn_update_ramrod_params {
-	struct iscsi_slow_path_hdr hdr;
+	__le16 reserved0;
 	__le16 conn_id;
-	__le32 fw_cid;
+	__le32 reserved1;
 	u8 flags;
 #define ISCSI_CONN_UPDATE_RAMROD_PARAMS_HD_EN_MASK		0x1
 #define ISCSI_CONN_UPDATE_RAMROD_PARAMS_HD_EN_SHIFT		0
@@ -1065,7 +1053,7 @@ struct iscsi_conn_update_ramrod_params {
 #define ISCSI_CONN_UPDATE_RAMROD_PARAMS_DIF_ON_IMM_EN_SHIFT	6
 #define ISCSI_CONN_UPDATE_RAMROD_PARAMS_LUN_MAPPER_EN_MASK	0x1
 #define ISCSI_CONN_UPDATE_RAMROD_PARAMS_LUN_MAPPER_EN_SHIFT	7
-	u8 reserved0[3];
+	u8 reserved3[3];
 	__le32 max_seq_size;
 	__le32 max_send_pdu_length;
 	__le32 max_recv_pdu_length;
@@ -1251,22 +1239,22 @@ enum iscsi_ramrod_cmd_id {
 
 /* iSCSI connection termination request */
 struct iscsi_spe_conn_mac_update {
-	struct iscsi_slow_path_hdr hdr;
+	__le16 reserved0;
 	__le16 conn_id;
-	__le32 fw_cid;
+	__le32 reserved1;
 	__le16 remote_mac_addr_lo;
 	__le16 remote_mac_addr_mid;
 	__le16 remote_mac_addr_hi;
-	u8 reserved0[2];
+	u8 reserved2[2];
 };
 
 /* iSCSI and TCP connection (Option 1) offload params passed by driver to FW in
  * iSCSI offload ramrod.
  */
 struct iscsi_spe_conn_offload {
-	struct iscsi_slow_path_hdr hdr;
+	__le16 reserved0;
 	__le16 conn_id;
-	__le32 fw_cid;
+	__le32 reserved1;
 	struct iscsi_conn_offload_params iscsi;
 	struct tcp_offload_params tcp;
 };
@@ -1275,44 +1263,36 @@ struct iscsi_spe_conn_offload {
  * iSCSI offload ramrod.
  */
 struct iscsi_spe_conn_offload_option2 {
-	struct iscsi_slow_path_hdr hdr;
+	__le16 reserved0;
 	__le16 conn_id;
-	__le32 fw_cid;
+	__le32 reserved1;
 	struct iscsi_conn_offload_params iscsi;
 	struct tcp_offload_params_opt2 tcp;
 };
 
 /* iSCSI collect connection statistics request */
 struct iscsi_spe_conn_statistics {
-	struct iscsi_slow_path_hdr hdr;
+	__le16 reserved0;
 	__le16 conn_id;
-	__le32 fw_cid;
+	__le32 reserved1;
 	u8 reset_stats;
-	u8 reserved0[7];
+	u8 reserved2[7];
 	struct regpair stats_cnts_addr;
 };
 
 /* iSCSI connection termination request */
 struct iscsi_spe_conn_termination {
-	struct iscsi_slow_path_hdr hdr;
+	__le16 reserved0;
 	__le16 conn_id;
-	__le32 fw_cid;
+	__le32 reserved1;
 	u8 abortive;
-	u8 reserved0[7];
+	u8 reserved2[7];
 	struct regpair queue_cnts_addr;
 	struct regpair query_params_addr;
 };
 
-/* iSCSI firmware function destroy parameters */
-struct iscsi_spe_func_dstry {
-	struct iscsi_slow_path_hdr hdr;
-	__le16 reserved0;
-	__le32 reserved1;
-};
-
 /* iSCSI firmware function init parameters */
 struct iscsi_spe_func_init {
-	struct iscsi_slow_path_hdr hdr;
 	__le16 half_way_close_timeout;
 	u8 num_sq_pages_in_ring;
 	u8 num_r2tq_pages_in_ring;
@@ -1324,8 +1304,12 @@ struct iscsi_spe_func_init {
 #define ISCSI_SPE_FUNC_INIT_RESERVED0_MASK	0x7F
 #define ISCSI_SPE_FUNC_INIT_RESERVED0_SHIFT	1
 	struct iscsi_debug_modes debug_mode;
-	__le16 reserved1;
-	__le32 reserved2;
+	u8 params;
+#define ISCSI_SPE_FUNC_INIT_MAX_SYN_RT_MASK	0xF
+#define ISCSI_SPE_FUNC_INIT_MAX_SYN_RT_SHIFT	0
+#define ISCSI_SPE_FUNC_INIT_RESERVED1_MASK	0xF
+#define ISCSI_SPE_FUNC_INIT_RESERVED1_SHIFT	4
+	u8 reserved2[7];
 	struct scsi_init_func_params func_params;
 	struct scsi_init_func_queues q_params;
 };
diff --git a/include/linux/qed/storage_common.h b/include/linux/qed/storage_common.h
index 505c0b48a761..9a973ffbbff5 100644
--- a/include/linux/qed/storage_common.h
+++ b/include/linux/qed/storage_common.h
@@ -107,8 +107,9 @@ struct scsi_drv_cmdq {
 struct scsi_init_func_params {
 	__le16 num_tasks;
 	u8 log_page_size;
+	u8 log_page_size_conn;
 	u8 debug_mode;
-	u8 reserved2[12];
+	u8 reserved2[11];
 };
 
 /* SCSI RQ/CQ/CMDQ firmware function init parameters */
-- 
cgit v1.2.3-59-g8ed1b


From 0500a70d6e071040ffdaadebb966986afa83c5e9 Mon Sep 17 00:00:00 2001
From: Michal Kalderon <michal.kalderon@marvell.com>
Date: Mon, 27 Jan 2020 15:26:15 +0200
Subject: qed: FW 8.42.2.0 HSI changes

This patch contains several HSI changes. The changes are part of
features like RDMA VF and OVS, the patch also contains a fix to
how the init code determines if the dmae is ready to be used.

Signed-off-by: Ariel Elior <ariel.elior@marvell.com>
Signed-off-by: Michal Kalderon <michal.kalderon@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/qlogic/qed/qed_hsi.h      | 317 ++++++++++++++++---------
 drivers/net/ethernet/qlogic/qed/qed_init_ops.c |   6 +-
 drivers/net/ethernet/qlogic/qed/qed_mcp.c      |   2 +
 drivers/net/ethernet/qlogic/qed/qed_roce.c     |   2 +-
 drivers/net/ethernet/qlogic/qede/qede_fp.c     |   8 +-
 include/linux/qed/common_hsi.h                 |  21 +-
 include/linux/qed/eth_common.h                 |  78 ++++--
 7 files changed, 280 insertions(+), 154 deletions(-)

diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
index 0c3d714e6518..06b94508115a 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
@@ -250,7 +250,8 @@ struct core_rx_gsi_offload_cqe {
 	__le16 src_mac_addrlo;
 	__le16 qp_id;
 	__le32 src_qp;
-	__le32 reserved[3];
+	struct core_rx_cqe_opaque_data opaque_data;
+	__le32 reserved;
 };
 
 /* Core RX CQE for Light L2 */
@@ -405,7 +406,7 @@ struct ystorm_core_conn_st_ctx {
 
 /* The core storm context for the Pstorm */
 struct pstorm_core_conn_st_ctx {
-	__le32 reserved[4];
+	__le32 reserved[20];
 };
 
 /* Core Slowpath Connection storm context of Xstorm */
@@ -856,12 +857,12 @@ struct e4_ustorm_core_conn_ag_ctx {
 
 /* The core storm context for the Mstorm */
 struct mstorm_core_conn_st_ctx {
-	__le32 reserved[24];
+	__le32 reserved[40];
 };
 
 /* The core storm context for the Ustorm */
 struct ustorm_core_conn_st_ctx {
-	__le32 reserved[4];
+	__le32 reserved[20];
 };
 
 /* The core storm context for the Tstorm */
@@ -915,12 +916,21 @@ struct eth_pstorm_per_pf_stat {
 	struct regpair sent_gre_bytes;
 	struct regpair sent_vxlan_bytes;
 	struct regpair sent_geneve_bytes;
+	struct regpair sent_mpls_bytes;
+	struct regpair sent_gre_mpls_bytes;
+	struct regpair sent_udp_mpls_bytes;
 	struct regpair sent_gre_pkts;
 	struct regpair sent_vxlan_pkts;
 	struct regpair sent_geneve_pkts;
+	struct regpair sent_mpls_pkts;
+	struct regpair sent_gre_mpls_pkts;
+	struct regpair sent_udp_mpls_pkts;
 	struct regpair gre_drop_pkts;
 	struct regpair vxlan_drop_pkts;
 	struct regpair geneve_drop_pkts;
+	struct regpair mpls_drop_pkts;
+	struct regpair gre_mpls_drop_pkts;
+	struct regpair udp_mpls_drop_pkts;
 };
 
 /* Ethernet TX Per Queue Stats */
@@ -1010,7 +1020,8 @@ union event_ring_data {
 struct event_ring_entry {
 	u8 protocol_id;
 	u8 opcode;
-	__le16 reserved0;
+	u8 reserved0;
+	u8 vf_id;
 	__le16 echo;
 	u8 fw_return_code;
 	u8 flags;
@@ -1088,7 +1099,20 @@ enum malicious_vf_error_id {
 	ETH_CONTROL_PACKET_VIOLATION,
 	ETH_ANTI_SPOOFING_ERR,
 	ETH_PACKET_SIZE_TOO_LARGE,
-	MAX_MALICIOUS_VF_ERROR_ID
+	CORE_ILLEGAL_VLAN_MODE,
+	CORE_ILLEGAL_NBDS,
+	CORE_FIRST_BD_WO_SOP,
+	CORE_INSUFFICIENT_BDS,
+	CORE_PACKET_TOO_SMALL,
+	CORE_ILLEGAL_INBAND_TAGS,
+	CORE_VLAN_INSERT_AND_INBAND_VLAN,
+	CORE_MTU_VIOLATION,
+	CORE_CONTROL_PACKET_VIOLATION,
+	CORE_ANTI_SPOOFING_ERR,
+	CORE_PACKET_SIZE_TOO_LARGE,
+	CORE_ILLEGAL_BD_FLAGS,
+	CORE_GSI_PACKET_VIOLATION,
+	MAX_MALICIOUS_VF_ERROR_ID,
 };
 
 /* Mstorm non-triggering VF zone */
@@ -1394,6 +1418,16 @@ enum vf_zone_size_mode {
 	MAX_VF_ZONE_SIZE_MODE
 };
 
+/* Xstorm non-triggering VF zone */
+struct xstorm_non_trigger_vf_zone {
+	struct regpair non_edpm_ack_pkts;
+};
+
+/* Tstorm VF zone */
+struct xstorm_vf_zone {
+	struct xstorm_non_trigger_vf_zone non_trigger;
+};
+
 /* Attentions status block */
 struct atten_status_block {
 	__le32 atten_bits;
@@ -2748,8 +2782,8 @@ enum chip_ids {
 };
 
 struct fw_asserts_ram_section {
-	u16 section_ram_line_offset;
-	u16 section_ram_line_size;
+	__le16 section_ram_line_offset;
+	__le16 section_ram_line_size;
 	u8 list_dword_offset;
 	u8 list_element_dword_size;
 	u8 list_num_elements;
@@ -2799,6 +2833,7 @@ enum init_modes {
 	MODE_PORTS_PER_ENG_4,
 	MODE_100G,
 	MODE_RESERVED6,
+	MODE_RESERVED7,
 	MAX_INIT_MODES
 };
 
@@ -2833,6 +2868,7 @@ enum bin_init_buffer_type {
 	BIN_BUF_INIT_VAL,
 	BIN_BUF_INIT_MODE_TREE,
 	BIN_BUF_INIT_IRO,
+	BIN_BUF_INIT_OVERLAYS,
 	MAX_BIN_INIT_BUFFER_TYPE
 };
 
@@ -2929,10 +2965,8 @@ struct init_if_phase_op {
 	u32 op_data;
 #define INIT_IF_PHASE_OP_OP_MASK		0xF
 #define INIT_IF_PHASE_OP_OP_SHIFT		0
-#define INIT_IF_PHASE_OP_DMAE_ENABLE_MASK	0x1
-#define INIT_IF_PHASE_OP_DMAE_ENABLE_SHIFT	4
-#define INIT_IF_PHASE_OP_RESERVED1_MASK		0x7FF
-#define INIT_IF_PHASE_OP_RESERVED1_SHIFT	5
+#define INIT_IF_PHASE_OP_RESERVED1_MASK		0xFFF
+#define INIT_IF_PHASE_OP_RESERVED1_SHIFT	4
 #define INIT_IF_PHASE_OP_CMD_OFFSET_MASK	0xFFFF
 #define INIT_IF_PHASE_OP_CMD_OFFSET_SHIFT	16
 	u32 phase_data;
@@ -4226,7 +4260,7 @@ void qed_gft_disable(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, u16 pf_id);
 /**
  * @brief qed_gft_config - Enable and configure HW for GFT
  *
- * @param p_hwfn
+ * @param p_hwfn - HW device data
  * @param p_ptt - ptt window used for writing the registers.
  * @param pf_id - pf on which to enable GFT.
  * @param tcp - set profile tcp packets.
@@ -5673,9 +5707,9 @@ struct e4_eth_conn_context {
 	struct pstorm_eth_conn_st_ctx pstorm_st_context;
 	struct xstorm_eth_conn_st_ctx xstorm_st_context;
 	struct e4_xstorm_eth_conn_ag_ctx xstorm_ag_context;
+	struct e4_tstorm_eth_conn_ag_ctx tstorm_ag_context;
 	struct ystorm_eth_conn_st_ctx ystorm_st_context;
 	struct e4_ystorm_eth_conn_ag_ctx ystorm_ag_context;
-	struct e4_tstorm_eth_conn_ag_ctx tstorm_ag_context;
 	struct e4_ustorm_eth_conn_ag_ctx ustorm_ag_context;
 	struct ustorm_eth_conn_st_ctx ustorm_st_context;
 	struct mstorm_eth_conn_st_ctx mstorm_st_context;
@@ -5705,6 +5739,16 @@ enum eth_error_code {
 	ETH_FILTERS_VNI_ADD_FAIL_FULL,
 	ETH_FILTERS_VNI_ADD_FAIL_DUP,
 	ETH_FILTERS_GFT_UPDATE_FAIL,
+	ETH_RX_QUEUE_FAIL_LOAD_VF_DATA,
+	ETH_FILTERS_GFS_ADD_FILTER_FAIL_MAX_HOPS,
+	ETH_FILTERS_GFS_ADD_FILTER_FAIL_NO_FREE_ENRTY,
+	ETH_FILTERS_GFS_ADD_FILTER_FAIL_ALREADY_EXISTS,
+	ETH_FILTERS_GFS_ADD_FILTER_FAIL_PCI_ERROR,
+	ETH_FILTERS_GFS_ADD_FINLER_FAIL_MAGIC_NUM_ERROR,
+	ETH_FILTERS_GFS_DEL_FILTER_FAIL_MAX_HOPS,
+	ETH_FILTERS_GFS_DEL_FILTER_FAIL_NO_MATCH_ENRTY,
+	ETH_FILTERS_GFS_DEL_FILTER_FAIL_PCI_ERROR,
+	ETH_FILTERS_GFS_DEL_FILTER_FAIL_MAGIC_NUM_ERROR,
 	MAX_ETH_ERROR_CODE
 };
 
@@ -5728,6 +5772,11 @@ enum eth_event_opcode {
 	ETH_EVENT_RX_CREATE_GFT_ACTION,
 	ETH_EVENT_RX_GFT_UPDATE_FILTER,
 	ETH_EVENT_TX_QUEUE_UPDATE,
+	ETH_EVENT_RGFS_ADD_FILTER,
+	ETH_EVENT_RGFS_DEL_FILTER,
+	ETH_EVENT_TGFS_ADD_FILTER,
+	ETH_EVENT_TGFS_DEL_FILTER,
+	ETH_EVENT_GFS_COUNTERS_REPORT_REQUEST,
 	MAX_ETH_EVENT_OPCODE
 };
 
@@ -5820,18 +5869,31 @@ enum eth_ramrod_cmd_id {
 	ETH_RAMROD_RX_CREATE_GFT_ACTION,
 	ETH_RAMROD_GFT_UPDATE_FILTER,
 	ETH_RAMROD_TX_QUEUE_UPDATE,
+	ETH_RAMROD_RGFS_FILTER_ADD,
+	ETH_RAMROD_RGFS_FILTER_DEL,
+	ETH_RAMROD_TGFS_FILTER_ADD,
+	ETH_RAMROD_TGFS_FILTER_DEL,
+	ETH_RAMROD_GFS_COUNTERS_REPORT_REQUEST,
 	MAX_ETH_RAMROD_CMD_ID
 };
 
 /* Return code from eth sp ramrods */
 struct eth_return_code {
 	u8 value;
-#define ETH_RETURN_CODE_ERR_CODE_MASK	0x1F
-#define ETH_RETURN_CODE_ERR_CODE_SHIFT	0
-#define ETH_RETURN_CODE_RESERVED_MASK	0x3
-#define ETH_RETURN_CODE_RESERVED_SHIFT	5
-#define ETH_RETURN_CODE_RX_TX_MASK	0x1
-#define ETH_RETURN_CODE_RX_TX_SHIFT	7
+#define ETH_RETURN_CODE_ERR_CODE_MASK  0x3F
+#define ETH_RETURN_CODE_ERR_CODE_SHIFT 0
+#define ETH_RETURN_CODE_RESERVED_MASK  0x1
+#define ETH_RETURN_CODE_RESERVED_SHIFT 6
+#define ETH_RETURN_CODE_RX_TX_MASK     0x1
+#define ETH_RETURN_CODE_RX_TX_SHIFT    7
+};
+
+/* tx destination enum */
+enum eth_tx_dst_mode_config_enum {
+	ETH_TX_DST_MODE_CONFIG_DISABLE,
+	ETH_TX_DST_MODE_CONFIG_FORWARD_DATA_IN_BD,
+	ETH_TX_DST_MODE_CONFIG_FORWARD_DATA_IN_VPORT,
+	MAX_ETH_TX_DST_MODE_CONFIG_ENUM
 };
 
 /* What to do in case an error occurs */
@@ -5858,8 +5920,10 @@ struct eth_tx_err_vals {
 #define ETH_TX_ERR_VALS_MTU_VIOLATION_SHIFT			5
 #define ETH_TX_ERR_VALS_ILLEGAL_CONTROL_FRAME_MASK		0x1
 #define ETH_TX_ERR_VALS_ILLEGAL_CONTROL_FRAME_SHIFT		6
-#define ETH_TX_ERR_VALS_RESERVED_MASK				0x1FF
-#define ETH_TX_ERR_VALS_RESERVED_SHIFT				7
+#define ETH_TX_ERR_VALS_ILLEGAL_BD_FLAGS_MASK			0x1
+#define ETH_TX_ERR_VALS_ILLEGAL_BD_FLAGS_SHIFT			7
+#define ETH_TX_ERR_VALS_RESERVED_MASK				0xFF
+#define ETH_TX_ERR_VALS_RESERVED_SHIFT				8
 };
 
 /* vport rss configuration data */
@@ -5889,7 +5953,6 @@ struct eth_vport_rss_config {
 	u8 tbl_size;
 	__le32 reserved2[2];
 	__le16 indirection_table[ETH_RSS_IND_TABLE_ENTRIES_NUM];
-
 	__le32 rss_key[ETH_RSS_KEY_SIZE_REGS];
 	__le32 reserved3[2];
 };
@@ -6091,7 +6154,7 @@ struct rx_update_gft_filter_data {
 	u8 inner_vlan_removal_en;
 };
 
-/* Ramrod data for rx queue start ramrod */
+/* Ramrod data for tx queue start ramrod */
 struct tx_queue_start_ramrod_data {
 	__le16 sb_id;
 	u8 sb_index;
@@ -6104,16 +6167,14 @@ struct tx_queue_start_ramrod_data {
 #define TX_QUEUE_START_RAMROD_DATA_DISABLE_OPPORTUNISTIC_SHIFT	0
 #define TX_QUEUE_START_RAMROD_DATA_TEST_MODE_PKT_DUP_MASK	0x1
 #define TX_QUEUE_START_RAMROD_DATA_TEST_MODE_PKT_DUP_SHIFT	1
-#define TX_QUEUE_START_RAMROD_DATA_TEST_MODE_TX_DEST_MASK	0x1
-#define TX_QUEUE_START_RAMROD_DATA_TEST_MODE_TX_DEST_SHIFT	2
 #define TX_QUEUE_START_RAMROD_DATA_PMD_MODE_MASK		0x1
-#define TX_QUEUE_START_RAMROD_DATA_PMD_MODE_SHIFT		3
+#define TX_QUEUE_START_RAMROD_DATA_PMD_MODE_SHIFT		2
 #define TX_QUEUE_START_RAMROD_DATA_NOTIFY_EN_MASK		0x1
-#define TX_QUEUE_START_RAMROD_DATA_NOTIFY_EN_SHIFT		4
+#define TX_QUEUE_START_RAMROD_DATA_NOTIFY_EN_SHIFT		3
 #define TX_QUEUE_START_RAMROD_DATA_PIN_CONTEXT_MASK		0x1
-#define TX_QUEUE_START_RAMROD_DATA_PIN_CONTEXT_SHIFT		5
-#define TX_QUEUE_START_RAMROD_DATA_RESERVED1_MASK		0x3
-#define TX_QUEUE_START_RAMROD_DATA_RESERVED1_SHIFT		6
+#define TX_QUEUE_START_RAMROD_DATA_PIN_CONTEXT_SHIFT		4
+#define TX_QUEUE_START_RAMROD_DATA_RESERVED1_MASK		0x7
+#define TX_QUEUE_START_RAMROD_DATA_RESERVED1_SHIFT		5
 	u8 pxp_st_hint;
 	u8 pxp_tph_valid_bd;
 	u8 pxp_tph_valid_pkt;
@@ -6169,18 +6230,22 @@ struct vport_start_ramrod_data {
 	__le16 default_vlan;
 	u8 tx_switching_en;
 	u8 anti_spoofing_en;
-
 	u8 default_vlan_en;
-
 	u8 handle_ptp_pkts;
 	u8 silent_vlan_removal_en;
 	u8 untagged;
 	struct eth_tx_err_vals tx_err_behav;
-
 	u8 zero_placement_offset;
 	u8 ctl_frame_mac_check_en;
 	u8 ctl_frame_ethtype_check_en;
+	u8 reserved0;
+	u8 reserved1;
+	u8 tx_dst_port_mode_config;
+	u8 dst_vport_id;
+	u8 tx_dst_port_mode;
+	u8 dst_vport_id_valid;
 	u8 wipe_inner_vlan_pri_en;
+	u8 reserved2[2];
 	struct eth_in_to_in_pri_map_cfg in_to_in_vlan_pri_map_cfg;
 };
 
@@ -6740,19 +6805,6 @@ struct e4_xstorm_eth_hw_conn_ag_ctx {
 	__le16 conn_dpi;
 };
 
-/* GFT CAM line struct */
-struct gft_cam_line {
-	__le32 camline;
-#define GFT_CAM_LINE_VALID_MASK		0x1
-#define GFT_CAM_LINE_VALID_SHIFT	0
-#define GFT_CAM_LINE_DATA_MASK		0x3FFF
-#define GFT_CAM_LINE_DATA_SHIFT		1
-#define GFT_CAM_LINE_MASK_BITS_MASK	0x3FFF
-#define GFT_CAM_LINE_MASK_BITS_SHIFT	15
-#define GFT_CAM_LINE_RESERVED1_MASK	0x7
-#define GFT_CAM_LINE_RESERVED1_SHIFT	29
-};
-
 /* GFT CAM line struct with fields breakout */
 struct gft_cam_line_mapped {
 	__le32 camline;
@@ -6782,10 +6834,6 @@ struct gft_cam_line_mapped {
 #define GFT_CAM_LINE_MAPPED_RESERVED1_SHIFT			29
 };
 
-union gft_cam_line_union {
-	struct gft_cam_line cam_line;
-	struct gft_cam_line_mapped cam_line_mapped;
-};
 
 /* Used in gft_profile_key: Indication for ip version */
 enum gft_profile_ip_version {
@@ -7064,6 +7112,11 @@ struct mstorm_rdma_task_st_ctx {
 	struct regpair temp[4];
 };
 
+/* The roce task context of Ustorm */
+struct ustorm_rdma_task_st_ctx {
+	struct regpair temp[6];
+};
+
 struct e4_ustorm_rdma_task_ag_ctx {
 	u8 reserved;
 	u8 state;
@@ -7073,8 +7126,8 @@ struct e4_ustorm_rdma_task_ag_ctx {
 #define E4_USTORM_RDMA_TASK_AG_CTX_CONNECTION_TYPE_SHIFT	0
 #define E4_USTORM_RDMA_TASK_AG_CTX_EXIST_IN_QM0_MASK		0x1
 #define E4_USTORM_RDMA_TASK_AG_CTX_EXIST_IN_QM0_SHIFT		4
-#define E4_USTORM_RDMA_TASK_AG_CTX_DIF_RUNT_VALID_MASK		0x1
-#define E4_USTORM_RDMA_TASK_AG_CTX_DIF_RUNT_VALID_SHIFT		5
+#define E4_USTORM_RDMA_TASK_AG_CTX_BIT1_MASK			0x1
+#define E4_USTORM_RDMA_TASK_AG_CTX_BIT1_SHIFT			5
 #define E4_USTORM_RDMA_TASK_AG_CTX_DIF_WRITE_RESULT_CF_MASK	0x3
 #define E4_USTORM_RDMA_TASK_AG_CTX_DIF_WRITE_RESULT_CF_SHIFT	6
 	u8 flags1;
@@ -7104,29 +7157,29 @@ struct e4_ustorm_rdma_task_ag_ctx {
 #define E4_USTORM_RDMA_TASK_AG_CTX_RULE2EN_MASK			0x1
 #define E4_USTORM_RDMA_TASK_AG_CTX_RULE2EN_SHIFT		7
 	u8 flags3;
-#define E4_USTORM_RDMA_TASK_AG_CTX_RULE3EN_MASK		0x1
-#define E4_USTORM_RDMA_TASK_AG_CTX_RULE3EN_SHIFT	0
-#define E4_USTORM_RDMA_TASK_AG_CTX_RULE4EN_MASK		0x1
-#define E4_USTORM_RDMA_TASK_AG_CTX_RULE4EN_SHIFT	1
-#define E4_USTORM_RDMA_TASK_AG_CTX_RULE5EN_MASK		0x1
-#define E4_USTORM_RDMA_TASK_AG_CTX_RULE5EN_SHIFT	2
-#define E4_USTORM_RDMA_TASK_AG_CTX_RULE6EN_MASK		0x1
-#define E4_USTORM_RDMA_TASK_AG_CTX_RULE6EN_SHIFT	3
-#define E4_USTORM_RDMA_TASK_AG_CTX_DIF_ERROR_TYPE_MASK	0xF
-#define E4_USTORM_RDMA_TASK_AG_CTX_DIF_ERROR_TYPE_SHIFT	4
+#define E4_USTORM_RDMA_TASK_AG_CTX_DIF_RXMIT_PROD_CONS_EN_MASK	0x1
+#define E4_USTORM_RDMA_TASK_AG_CTX_DIF_RXMIT_PROD_CONS_EN_SHIFT	0
+#define E4_USTORM_RDMA_TASK_AG_CTX_RULE4EN_MASK			0x1
+#define E4_USTORM_RDMA_TASK_AG_CTX_RULE4EN_SHIFT		1
+#define E4_USTORM_RDMA_TASK_AG_CTX_DIF_WRITE_PROD_CONS_EN_MASK	0x1
+#define E4_USTORM_RDMA_TASK_AG_CTX_DIF_WRITE_PROD_CONS_EN_SHIFT	2
+#define E4_USTORM_RDMA_TASK_AG_CTX_RULE6EN_MASK			0x1
+#define E4_USTORM_RDMA_TASK_AG_CTX_RULE6EN_SHIFT		3
+#define E4_USTORM_RDMA_TASK_AG_CTX_DIF_ERROR_TYPE_MASK		0xF
+#define E4_USTORM_RDMA_TASK_AG_CTX_DIF_ERROR_TYPE_SHIFT		4
 	__le32 dif_err_intervals;
 	__le32 dif_error_1st_interval;
-	__le32 sq_cons;
-	__le32 dif_runt_value;
+	__le32 dif_rxmit_cons;
+	__le32 dif_rxmit_prod;
 	__le32 sge_index;
-	__le32 reg5;
+	__le32 sq_cons;
 	u8 byte2;
 	u8 byte3;
-	__le16 word1;
-	__le16 word2;
+	__le16 dif_write_cons;
+	__le16 dif_write_prod;
 	__le16 word3;
-	__le32 reg6;
-	__le32 reg7;
+	__le32 dif_error_buffer_address_lo;
+	__le32 dif_error_buffer_address_hi;
 };
 
 /* RDMA task context */
@@ -7137,6 +7190,8 @@ struct e4_rdma_task_context {
 	struct e4_mstorm_rdma_task_ag_ctx mstorm_ag_context;
 	struct mstorm_rdma_task_st_ctx mstorm_st_context;
 	struct rdif_task_context rdif_context;
+	struct ustorm_rdma_task_st_ctx ustorm_st_context;
+	struct regpair ustorm_st_padding[2];
 	struct e4_ustorm_rdma_task_ag_ctx ustorm_ag_context;
 };
 
@@ -7172,7 +7227,12 @@ struct rdma_create_cq_ramrod_data {
 	u8 pbl_log_page_size;
 	u8 toggle_bit;
 	__le16 int_timeout;
-	__le16 reserved1;
+	u8 vf_id;
+	u8 flags;
+#define RDMA_CREATE_CQ_RAMROD_DATA_VF_ID_VALID_MASK  0x1
+#define RDMA_CREATE_CQ_RAMROD_DATA_VF_ID_VALID_SHIFT 0
+#define RDMA_CREATE_CQ_RAMROD_DATA_RESERVED1_MASK    0x7F
+#define RDMA_CREATE_CQ_RAMROD_DATA_RESERVED1_SHIFT   1
 };
 
 /* rdma deregister tid ramrod data */
@@ -7216,6 +7276,7 @@ enum rdma_fw_return_code {
 	RDMA_RETURN_DEREGISTER_MR_BAD_STATE_ERR,
 	RDMA_RETURN_RESIZE_CQ_ERR,
 	RDMA_RETURN_NIG_DRAIN_REQ,
+	RDMA_RETURN_GENERAL_ERR,
 	MAX_RDMA_FW_RETURN_CODE
 };
 
@@ -7229,7 +7290,10 @@ struct rdma_init_func_hdr {
 	u8 relaxed_ordering;
 	__le16 first_reg_srq_id;
 	__le32 reg_srq_base_addr;
-	__le32 reserved;
+	u8 searcher_mode;
+	u8 pvrdma_mode;
+	u8 max_num_ns_log;
+	u8 reserved;
 };
 
 /* rdma function init ramrod data */
@@ -7319,16 +7383,20 @@ struct rdma_resize_cq_ramrod_data {
 #define RDMA_RESIZE_CQ_RAMROD_DATA_TOGGLE_BIT_SHIFT		0
 #define RDMA_RESIZE_CQ_RAMROD_DATA_IS_TWO_LEVEL_PBL_MASK	0x1
 #define RDMA_RESIZE_CQ_RAMROD_DATA_IS_TWO_LEVEL_PBL_SHIFT	1
-#define RDMA_RESIZE_CQ_RAMROD_DATA_RESERVED_MASK		0x3F
-#define RDMA_RESIZE_CQ_RAMROD_DATA_RESERVED_SHIFT		2
+#define RDMA_RESIZE_CQ_RAMROD_DATA_VF_ID_VALID_MASK		0x1
+#define RDMA_RESIZE_CQ_RAMROD_DATA_VF_ID_VALID_SHIFT		2
+#define RDMA_RESIZE_CQ_RAMROD_DATA_RESERVED_MASK		0x1F
+#define RDMA_RESIZE_CQ_RAMROD_DATA_RESERVED_SHIFT		3
 	u8 pbl_log_page_size;
 	__le16 pbl_num_pages;
 	__le32 max_cqes;
 	struct regpair pbl_addr;
 	struct regpair output_params_addr;
+	u8 vf_id;
+	u8 reserved1[7];
 };
 
-/* The rdma storm context of Mstorm */
+/* The rdma SRQ context */
 struct rdma_srq_context {
 	struct regpair temp[8];
 };
@@ -7375,6 +7443,7 @@ enum rdma_tid_type {
 	MAX_RDMA_TID_TYPE
 };
 
+/* The rdma XRC SRQ context */
 struct rdma_xrc_srq_context {
 	struct regpair temp[9];
 };
@@ -7556,12 +7625,12 @@ struct e4_xstorm_roce_conn_ag_ctx {
 #define E4_XSTORM_ROCE_CONN_AG_CTX_BIT10_SHIFT            2
 #define E4_XSTORM_ROCE_CONN_AG_CTX_BIT11_MASK             0x1
 #define E4_XSTORM_ROCE_CONN_AG_CTX_BIT11_SHIFT            3
-#define E4_XSTORM_ROCE_CONN_AG_CTX_BIT12_MASK             0x1
-#define E4_XSTORM_ROCE_CONN_AG_CTX_BIT12_SHIFT            4
+#define E4_XSTORM_ROCE_CONN_AG_CTX_MSDM_FLUSH_MASK        0x1
+#define E4_XSTORM_ROCE_CONN_AG_CTX_MSDM_FLUSH_SHIFT       4
 #define E4_XSTORM_ROCE_CONN_AG_CTX_MSEM_FLUSH_MASK        0x1
 #define E4_XSTORM_ROCE_CONN_AG_CTX_MSEM_FLUSH_SHIFT       5
-#define E4_XSTORM_ROCE_CONN_AG_CTX_MSDM_FLUSH_MASK        0x1
-#define E4_XSTORM_ROCE_CONN_AG_CTX_MSDM_FLUSH_SHIFT       6
+#define E4_XSTORM_ROCE_CONN_AG_CTX_BIT14_MASK	       0x1
+#define E4_XSTORM_ROCE_CONN_AG_CTX_BIT14_SHIFT	       6
 #define E4_XSTORM_ROCE_CONN_AG_CTX_YSTORM_FLUSH_MASK      0x1
 #define E4_XSTORM_ROCE_CONN_AG_CTX_YSTORM_FLUSH_SHIFT     7
 	u8 flags2;
@@ -7885,9 +7954,9 @@ struct mstorm_roce_conn_st_ctx {
 	struct regpair temp[6];
 };
 
-/* The roce storm context of Ystorm */
+/* The roce storm context of Ustorm */
 struct ustorm_roce_conn_st_ctx {
-	struct regpair temp[12];
+	struct regpair temp[14];
 };
 
 /* roce connection context */
@@ -7905,6 +7974,7 @@ struct e4_roce_conn_context {
 	struct mstorm_roce_conn_st_ctx mstorm_st_context;
 	struct regpair mstorm_st_padding[2];
 	struct ustorm_roce_conn_st_ctx ustorm_st_context;
+	struct regpair ustorm_st_padding[2];
 };
 
 /* roce cqes statistics */
@@ -7959,12 +8029,17 @@ struct roce_create_qp_req_ramrod_data {
 	struct regpair qp_handle_for_cqe;
 	struct regpair qp_handle_for_async;
 	u8 stats_counter_id;
-	u8 reserved3[6];
+	u8 vf_id;
+	u8 vport_id;
 	u8 flags2;
 #define ROCE_CREATE_QP_REQ_RAMROD_DATA_EDPM_MODE_MASK			0x1
 #define ROCE_CREATE_QP_REQ_RAMROD_DATA_EDPM_MODE_SHIFT			0
-#define ROCE_CREATE_QP_REQ_RAMROD_DATA_RESERVED_MASK			0x7F
-#define ROCE_CREATE_QP_REQ_RAMROD_DATA_RESERVED_SHIFT			1
+#define ROCE_CREATE_QP_REQ_RAMROD_DATA_VF_ID_VALID_MASK			0x1
+#define ROCE_CREATE_QP_REQ_RAMROD_DATA_VF_ID_VALID_SHIFT		1
+#define ROCE_CREATE_QP_REQ_RAMROD_DATA_RESERVED_MASK			0x3F
+#define ROCE_CREATE_QP_REQ_RAMROD_DATA_RESERVED_SHIFT			2
+	u8 name_space;
+	u8 reserved3[3];
 	__le16 regular_latency_phy_queue;
 	__le16 dpi;
 };
@@ -7992,8 +8067,10 @@ struct roce_create_qp_resp_ramrod_data {
 #define ROCE_CREATE_QP_RESP_RAMROD_DATA_MIN_RNR_NAK_TIMER_SHIFT		11
 #define ROCE_CREATE_QP_RESP_RAMROD_DATA_XRC_FLAG_MASK             0x1
 #define ROCE_CREATE_QP_RESP_RAMROD_DATA_XRC_FLAG_SHIFT            16
-#define ROCE_CREATE_QP_RESP_RAMROD_DATA_RESERVED_MASK             0x7FFF
-#define ROCE_CREATE_QP_RESP_RAMROD_DATA_RESERVED_SHIFT            17
+#define ROCE_CREATE_QP_RESP_RAMROD_DATA_VF_ID_VALID_MASK	0x1
+#define ROCE_CREATE_QP_RESP_RAMROD_DATA_VF_ID_VALID_SHIFT	17
+#define ROCE_CREATE_QP_RESP_RAMROD_DATA_RESERVED_MASK		0x3FFF
+#define ROCE_CREATE_QP_RESP_RAMROD_DATA_RESERVED_SHIFT		18
 	__le16 xrc_domain;
 	u8 max_ird;
 	u8 traffic_class;
@@ -8020,10 +8097,14 @@ struct roce_create_qp_resp_ramrod_data {
 	struct regpair qp_handle_for_cqe;
 	struct regpair qp_handle_for_async;
 	__le16 low_latency_phy_queue;
-	u8 reserved2[2];
+	u8 vf_id;
+	u8 vport_id;
 	__le32 cq_cid;
 	__le16 regular_latency_phy_queue;
 	__le16 dpi;
+	__le32 src_qp_id;
+	u8 name_space;
+	u8 reserved3[3];
 };
 
 /* roce DCQCN received statistics */
@@ -8057,6 +8138,8 @@ struct roce_destroy_qp_resp_output_params {
 /* RoCE destroy qp responder ramrod data */
 struct roce_destroy_qp_resp_ramrod_data {
 	struct regpair output_params_addr;
+	__le32 src_qp_id;
+	__le32 reserved;
 };
 
 /* roce error statistics */
@@ -8140,8 +8223,8 @@ struct roce_modify_qp_req_ramrod_data {
 #define ROCE_MODIFY_QP_REQ_RAMROD_DATA_PRI_FLG_SHIFT			9
 #define ROCE_MODIFY_QP_REQ_RAMROD_DATA_PRI_MASK				0x7
 #define ROCE_MODIFY_QP_REQ_RAMROD_DATA_PRI_SHIFT			10
-#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_PHYSICAL_QUEUES_FLG_MASK		0x1
-#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_PHYSICAL_QUEUES_FLG_SHIFT	13
+#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_PHYSICAL_QUEUE_FLG_MASK		0x1
+#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_PHYSICAL_QUEUE_FLG_SHIFT		13
 #define ROCE_MODIFY_QP_REQ_RAMROD_DATA_RESERVED1_MASK			0x3
 #define ROCE_MODIFY_QP_REQ_RAMROD_DATA_RESERVED1_SHIFT			14
 	u8 fields;
@@ -8187,8 +8270,8 @@ struct roce_modify_qp_resp_ramrod_data {
 #define ROCE_MODIFY_QP_RESP_RAMROD_DATA_MIN_RNR_NAK_TIMER_FLG_SHIFT	8
 #define ROCE_MODIFY_QP_RESP_RAMROD_DATA_RDMA_OPS_EN_FLG_MASK		0x1
 #define ROCE_MODIFY_QP_RESP_RAMROD_DATA_RDMA_OPS_EN_FLG_SHIFT		9
-#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_PHYSICAL_QUEUES_FLG_MASK	0x1
-#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_PHYSICAL_QUEUES_FLG_SHIFT	10
+#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_PHYSICAL_QUEUE_FLG_MASK		0x1
+#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_PHYSICAL_QUEUE_FLG_SHIFT	10
 #define ROCE_MODIFY_QP_RESP_RAMROD_DATA_RESERVED1_MASK			0x1F
 #define ROCE_MODIFY_QP_RESP_RAMROD_DATA_RESERVED1_SHIFT			11
 	u8 fields;
@@ -8229,7 +8312,7 @@ struct roce_query_qp_req_ramrod_data {
 /* RoCE query qp responder output params */
 struct roce_query_qp_resp_output_params {
 	__le32 psn;
-	__le32 err_flag;
+	__le32 flags;
 #define ROCE_QUERY_QP_RESP_OUTPUT_PARAMS_ERROR_FLG_MASK  0x1
 #define ROCE_QUERY_QP_RESP_OUTPUT_PARAMS_ERROR_FLG_SHIFT 0
 #define ROCE_QUERY_QP_RESP_OUTPUT_PARAMS_RESERVED0_MASK  0x7FFFFFFF
@@ -8296,12 +8379,12 @@ struct e4_xstorm_roce_conn_ag_ctx_dq_ext_ld_part {
 #define E4XSTORMROCECONNAGCTXDQEXTLDPART_BIT10_SHIFT		2
 #define E4XSTORMROCECONNAGCTXDQEXTLDPART_BIT11_MASK		0x1
 #define E4XSTORMROCECONNAGCTXDQEXTLDPART_BIT11_SHIFT		3
-#define E4XSTORMROCECONNAGCTXDQEXTLDPART_BIT12_MASK		0x1
-#define E4XSTORMROCECONNAGCTXDQEXTLDPART_BIT12_SHIFT		4
-#define E4XSTORMROCECONNAGCTXDQEXTLDPART_MSEM_FLUSH_MASK        0x1
-#define E4XSTORMROCECONNAGCTXDQEXTLDPART_MSEM_FLUSH_SHIFT       5
-#define E4XSTORMROCECONNAGCTXDQEXTLDPART_MSDM_FLUSH_MASK        0x1
-#define E4XSTORMROCECONNAGCTXDQEXTLDPART_MSDM_FLUSH_SHIFT       6
+#define E4XSTORMROCECONNAGCTXDQEXTLDPART_MSDM_FLUSH_MASK	0x1
+#define E4XSTORMROCECONNAGCTXDQEXTLDPART_MSDM_FLUSH_SHIFT	4
+#define E4XSTORMROCECONNAGCTXDQEXTLDPART_MSEM_FLUSH_MASK	0x1
+#define E4XSTORMROCECONNAGCTXDQEXTLDPART_MSEM_FLUSH_SHIFT	5
+#define E4XSTORMROCECONNAGCTXDQEXTLDPART_BIT14_MASK		0x1
+#define E4XSTORMROCECONNAGCTXDQEXTLDPART_BIT14_SHIFT		6
 #define E4XSTORMROCECONNAGCTXDQEXTLDPART_YSTORM_FLUSH_MASK	0x1
 #define E4XSTORMROCECONNAGCTXDQEXTLDPART_YSTORM_FLUSH_SHIFT	7
 	u8 flags2;
@@ -8674,8 +8757,8 @@ struct e4_tstorm_roce_req_conn_ag_ctx {
 	u8 flags5;
 #define E4_TSTORM_ROCE_REQ_CONN_AG_CTX_RULE1EN_MASK		0x1
 #define E4_TSTORM_ROCE_REQ_CONN_AG_CTX_RULE1EN_SHIFT		0
-#define E4_TSTORM_ROCE_REQ_CONN_AG_CTX_RULE2EN_MASK		0x1
-#define E4_TSTORM_ROCE_REQ_CONN_AG_CTX_RULE2EN_SHIFT		1
+#define E4_TSTORM_ROCE_REQ_CONN_AG_CTX_DIF_CNT_EN_MASK		0x1
+#define E4_TSTORM_ROCE_REQ_CONN_AG_CTX_DIF_CNT_EN_SHIFT		1
 #define E4_TSTORM_ROCE_REQ_CONN_AG_CTX_RULE3EN_MASK		0x1
 #define E4_TSTORM_ROCE_REQ_CONN_AG_CTX_RULE3EN_SHIFT		2
 #define E4_TSTORM_ROCE_REQ_CONN_AG_CTX_RULE4EN_MASK		0x1
@@ -8688,13 +8771,13 @@ struct e4_tstorm_roce_req_conn_ag_ctx {
 #define E4_TSTORM_ROCE_REQ_CONN_AG_CTX_RULE7EN_SHIFT		6
 #define E4_TSTORM_ROCE_REQ_CONN_AG_CTX_RULE8EN_MASK		0x1
 #define E4_TSTORM_ROCE_REQ_CONN_AG_CTX_RULE8EN_SHIFT		7
-	__le32 reg0;
+	__le32 dif_rxmit_cnt;
 	__le32 snd_nxt_psn;
 	__le32 snd_max_psn;
 	__le32 orq_prod;
 	__le32 reg4;
-	__le32 reg5;
-	__le32 reg6;
+	__le32 dif_acked_cnt;
+	__le32 dif_cnt;
 	__le32 reg7;
 	__le32 reg8;
 	u8 tx_cqe_error_type;
@@ -8705,7 +8788,7 @@ struct e4_tstorm_roce_req_conn_ag_ctx {
 	__le16 snd_sq_cons;
 	__le16 conn_dpi;
 	__le16 force_comp_cons;
-	__le32 reg9;
+	__le32 dif_rxmit_acked_cnt;
 	__le32 reg10;
 };
 
@@ -8980,10 +9063,10 @@ struct e4_xstorm_roce_req_conn_ag_ctx {
 #define E4_XSTORM_ROCE_REQ_CONN_AG_CTX_BIT10_SHIFT		2
 #define E4_XSTORM_ROCE_REQ_CONN_AG_CTX_BIT11_MASK		0x1
 #define E4_XSTORM_ROCE_REQ_CONN_AG_CTX_BIT11_SHIFT		3
-#define E4_XSTORM_ROCE_REQ_CONN_AG_CTX_BIT12_MASK		0x1
-#define E4_XSTORM_ROCE_REQ_CONN_AG_CTX_BIT12_SHIFT		4
-#define E4_XSTORM_ROCE_REQ_CONN_AG_CTX_BIT13_MASK		0x1
-#define E4_XSTORM_ROCE_REQ_CONN_AG_CTX_BIT13_SHIFT		5
+#define E4_XSTORM_ROCE_REQ_CONN_AG_CTX_MSDM_FLUSH_MASK		0x1
+#define E4_XSTORM_ROCE_REQ_CONN_AG_CTX_MSDM_FLUSH_SHIFT		4
+#define E4_XSTORM_ROCE_REQ_CONN_AG_CTX_MSEM_FLUSH_MASK		0x1
+#define E4_XSTORM_ROCE_REQ_CONN_AG_CTX_MSEM_FLUSH_SHIFT		5
 #define E4_XSTORM_ROCE_REQ_CONN_AG_CTX_ERROR_STATE_MASK		0x1
 #define E4_XSTORM_ROCE_REQ_CONN_AG_CTX_ERROR_STATE_SHIFT	6
 #define E4_XSTORM_ROCE_REQ_CONN_AG_CTX_YSTORM_FLUSH_MASK	0x1
@@ -9209,10 +9292,10 @@ struct e4_xstorm_roce_resp_conn_ag_ctx {
 #define E4_XSTORM_ROCE_RESP_CONN_AG_CTX_BIT10_SHIFT		2
 #define E4_XSTORM_ROCE_RESP_CONN_AG_CTX_BIT11_MASK		0x1
 #define E4_XSTORM_ROCE_RESP_CONN_AG_CTX_BIT11_SHIFT		3
-#define E4_XSTORM_ROCE_RESP_CONN_AG_CTX_BIT12_MASK		0x1
-#define E4_XSTORM_ROCE_RESP_CONN_AG_CTX_BIT12_SHIFT		4
-#define E4_XSTORM_ROCE_RESP_CONN_AG_CTX_BIT13_MASK		0x1
-#define E4_XSTORM_ROCE_RESP_CONN_AG_CTX_BIT13_SHIFT		5
+#define E4_XSTORM_ROCE_RESP_CONN_AG_CTX_MSDM_FLUSH_MASK		0x1
+#define E4_XSTORM_ROCE_RESP_CONN_AG_CTX_MSDM_FLUSH_SHIFT	4
+#define E4_XSTORM_ROCE_RESP_CONN_AG_CTX_MSEM_FLUSH_MASK		0x1
+#define E4_XSTORM_ROCE_RESP_CONN_AG_CTX_MSEM_FLUSH_SHIFT	5
 #define E4_XSTORM_ROCE_RESP_CONN_AG_CTX_ERROR_STATE_MASK	0x1
 #define E4_XSTORM_ROCE_RESP_CONN_AG_CTX_ERROR_STATE_SHIFT	6
 #define E4_XSTORM_ROCE_RESP_CONN_AG_CTX_YSTORM_FLUSH_MASK	0x1
@@ -9939,7 +10022,7 @@ struct mstorm_iwarp_conn_st_ctx {
 
 /* The iwarp storm context of Ustorm */
 struct ustorm_iwarp_conn_st_ctx {
-	__le32 reserved[24];
+	struct regpair reserved[14];
 };
 
 /* iwarp connection context */
@@ -9957,6 +10040,7 @@ struct e4_iwarp_conn_context {
 	struct regpair tstorm_st_padding[2];
 	struct mstorm_iwarp_conn_st_ctx mstorm_st_context;
 	struct ustorm_iwarp_conn_st_ctx ustorm_st_context;
+	struct regpair ustorm_st_padding[2];
 };
 
 /* iWARP create QP params passed by driver to FW in CreateQP Request Ramrod */
@@ -10009,7 +10093,8 @@ enum iwarp_eqe_async_opcode {
 
 struct iwarp_eqe_data_mpa_async_completion {
 	__le16 ulp_data_len;
-	u8 reserved[6];
+	u8 rtr_type_sent;
+	u8 reserved[5];
 };
 
 struct iwarp_eqe_data_tcp_async_completion {
@@ -10034,7 +10119,7 @@ enum iwarp_eqe_sync_opcode {
 
 /* iWARP EQE completion status */
 enum iwarp_fw_return_code {
-	IWARP_CONN_ERROR_TCP_CONNECT_INVALID_PACKET = 5,
+	IWARP_CONN_ERROR_TCP_CONNECT_INVALID_PACKET = 6,
 	IWARP_CONN_ERROR_TCP_CONNECTION_RST,
 	IWARP_CONN_ERROR_TCP_CONNECT_TIMEOUT,
 	IWARP_CONN_ERROR_MPA_ERROR_REJECT,
@@ -10203,8 +10288,8 @@ struct iwarp_rxmit_stats_drv {
  * offload ramrod.
  */
 struct iwarp_tcp_offload_ramrod_data {
-	struct iwarp_offload_params iwarp;
 	struct tcp_offload_params_opt2 tcp;
+	struct iwarp_offload_params iwarp;
 };
 
 /* iWARP MPA negotiation types */
diff --git a/drivers/net/ethernet/qlogic/qed/qed_init_ops.c b/drivers/net/ethernet/qlogic/qed/qed_init_ops.c
index 31222b3e3936..5ca117a8734a 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_init_ops.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_init_ops.c
@@ -490,10 +490,10 @@ static u32 qed_init_cmd_phase(struct qed_hwfn *p_hwfn,
 int qed_init_run(struct qed_hwfn *p_hwfn,
 		 struct qed_ptt *p_ptt, int phase, int phase_id, int modes)
 {
+	bool b_dmae = (phase != PHASE_ENGINE);
 	struct qed_dev *cdev = p_hwfn->cdev;
 	u32 cmd_num, num_init_ops;
 	union init_op *init_ops;
-	bool b_dmae = false;
 	int rc = 0;
 
 	num_init_ops = cdev->fw_data->init_ops_size;
@@ -522,7 +522,6 @@ int qed_init_run(struct qed_hwfn *p_hwfn,
 		case INIT_OP_IF_PHASE:
 			cmd_num += qed_init_cmd_phase(p_hwfn, &cmd->if_phase,
 						      phase, phase_id);
-			b_dmae = GET_FIELD(data, INIT_IF_PHASE_OP_DMAE_ENABLE);
 			break;
 		case INIT_OP_DELAY:
 			/* qed_init_run is always invoked from
@@ -533,6 +532,9 @@ int qed_init_run(struct qed_hwfn *p_hwfn,
 
 		case INIT_OP_CALLBACK:
 			rc = qed_init_cmd_cb(p_hwfn, p_ptt, &cmd->callback);
+			if (phase == PHASE_ENGINE &&
+			    cmd->callback.callback_id == DMAE_READY_CB)
+				b_dmae = true;
 			break;
 		}
 
diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.c b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
index aeac359941de..af981ff5595e 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_mcp.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
@@ -48,6 +48,8 @@
 #include "qed_reg_addr.h"
 #include "qed_sriov.h"
 
+#define GRCBASE_MCP     0xe00000
+
 #define QED_MCP_RESP_ITER_US	10
 
 #define QED_DRV_MB_MAX_RETRIES	(500 * 1000)	/* Account for 5 sec */
diff --git a/drivers/net/ethernet/qlogic/qed/qed_roce.c b/drivers/net/ethernet/qlogic/qed/qed_roce.c
index e49fada85410..37e70562a964 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_roce.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_roce.c
@@ -900,7 +900,7 @@ int qed_roce_query_qp(struct qed_hwfn *p_hwfn,
 		goto err_resp;
 
 	out_params->rq_psn = le32_to_cpu(p_resp_ramrod_res->psn);
-	rq_err_state = GET_FIELD(le32_to_cpu(p_resp_ramrod_res->err_flag),
+	rq_err_state = GET_FIELD(le32_to_cpu(p_resp_ramrod_res->flags),
 				 ROCE_QUERY_QP_RESP_OUTPUT_PARAMS_ERROR_FLG);
 
 	dma_free_coherent(&p_hwfn->cdev->pdev->dev, sizeof(*p_resp_ramrod_res),
diff --git a/drivers/net/ethernet/qlogic/qede/qede_fp.c b/drivers/net/ethernet/qlogic/qede/qede_fp.c
index 004c0bfec41d..c6c20776b474 100644
--- a/drivers/net/ethernet/qlogic/qede/qede_fp.c
+++ b/drivers/net/ethernet/qlogic/qede/qede_fp.c
@@ -848,13 +848,13 @@ static void qede_tpa_start(struct qede_dev *edev,
 	qede_set_gro_params(edev, tpa_info->skb, cqe);
 
 cons_buf: /* We still need to handle bd_len_list to consume buffers */
-	if (likely(cqe->ext_bd_len_list[0]))
+	if (likely(cqe->bw_ext_bd_len_list[0]))
 		qede_fill_frag_skb(edev, rxq, cqe->tpa_agg_index,
-				   le16_to_cpu(cqe->ext_bd_len_list[0]));
+				   le16_to_cpu(cqe->bw_ext_bd_len_list[0]));
 
-	if (unlikely(cqe->ext_bd_len_list[1])) {
+	if (unlikely(cqe->bw_ext_bd_len_list[1])) {
 		DP_ERR(edev,
-		       "Unlikely - got a TPA aggregation with more than one ext_bd_len_list entry in the TPA start\n");
+		       "Unlikely - got a TPA aggregation with more than one bw_ext_bd_len_list entry in the TPA start\n");
 		tpa_info->state = QEDE_AGG_STATE_ERROR;
 	}
 }
diff --git a/include/linux/qed/common_hsi.h b/include/linux/qed/common_hsi.h
index 718ce72e5965..2c4737e6694a 100644
--- a/include/linux/qed/common_hsi.h
+++ b/include/linux/qed/common_hsi.h
@@ -76,7 +76,6 @@
 
 #define FW_ASSERT_GENERAL_ATTN_IDX		32
 
-#define MAX_PINNED_CCFC				32
 
 /* Queue Zone sizes in bytes */
 #define TSTORM_QZONE_SIZE	8
@@ -139,10 +138,10 @@
 #define MAX_NUM_VFS	(MAX_NUM_VFS_K2)
 
 #define MAX_NUM_FUNCTIONS_BB	(MAX_NUM_PFS_BB + MAX_NUM_VFS_BB)
-#define MAX_NUM_FUNCTIONS	(MAX_NUM_PFS + MAX_NUM_VFS)
 
 #define MAX_FUNCTION_NUMBER_BB	(MAX_NUM_PFS + MAX_NUM_VFS_BB)
-#define MAX_FUNCTION_NUMBER	(MAX_NUM_PFS + MAX_NUM_VFS)
+#define MAX_FUNCTION_NUMBER_K2  (MAX_NUM_PFS + MAX_NUM_VFS_K2)
+#define MAX_NUM_FUNCTIONS	(MAX_FUNCTION_NUMBER_K2)
 
 #define MAX_NUM_VPORTS_K2	(208)
 #define MAX_NUM_VPORTS_BB	(160)
@@ -229,6 +228,7 @@
 #define DQ_XCM_TOE_TX_BD_PROD_CMD		DQ_XCM_AGG_VAL_SEL_WORD4
 #define DQ_XCM_TOE_MORE_TO_SEND_SEQ_CMD		DQ_XCM_AGG_VAL_SEL_REG3
 #define DQ_XCM_TOE_LOCAL_ADV_WND_SEQ_CMD	DQ_XCM_AGG_VAL_SEL_REG4
+#define DQ_XCM_ROCE_ACK_EDPM_DORQ_SEQ_CMD	DQ_XCM_AGG_VAL_SEL_WORD5
 
 /* UCM agg val selection (HW) */
 #define	DQ_UCM_AGG_VAL_SEL_WORD0	0
@@ -406,6 +406,7 @@
 
 /* Number of Protocol Indices per Status Block */
 #define PIS_PER_SB_E4	12
+#define MAX_PIS_PER_SB	PIS_PER_SB
 
 #define CAU_HC_STOPPED_STATE	3
 #define CAU_HC_DISABLE_STATE	4
@@ -436,8 +437,6 @@
 #define IGU_MEM_PBA_MSIX_RESERVED_UPPER	0x03ff
 
 #define IGU_CMD_INT_ACK_BASE		0x0400
-#define IGU_CMD_INT_ACK_UPPER		(IGU_CMD_INT_ACK_BASE +	\
-					 MAX_TOT_SB_PER_PATH - 1)
 #define IGU_CMD_INT_ACK_RESERVED_UPPER	0x05ff
 
 #define IGU_CMD_ATTN_BIT_UPD_UPPER	0x05f0
@@ -450,8 +449,6 @@
 #define IGU_REG_SISR_MDPC_WOMASK_UPPER		0x05f6
 
 #define IGU_CMD_PROD_UPD_BASE			0x0600
-#define IGU_CMD_PROD_UPD_UPPER			(IGU_CMD_PROD_UPD_BASE +\
-						 MAX_TOT_SB_PER_PATH - 1)
 #define IGU_CMD_PROD_UPD_RESERVED_UPPER		0x07ff
 
 /*****************/
@@ -741,6 +738,8 @@ enum protocol_type {
 	PROTOCOLID_PREROCE,
 	PROTOCOLID_COMMON,
 	PROTOCOLID_RESERVED1,
+	PROTOCOLID_RDMA,
+	PROTOCOLID_SCSI,
 	MAX_PROTOCOL_TYPE
 };
 
@@ -761,6 +760,10 @@ union rdma_eqe_data {
 	struct rdma_eqe_destroy_qp rdma_destroy_qp_data;
 };
 
+struct tstorm_queue_zone {
+	__le32 reserved[2];
+};
+
 /* Ustorm Queue Zone */
 struct ustorm_eth_queue_zone {
 	struct coalescing_timeset int_coalescing_timeset;
@@ -883,8 +886,8 @@ struct db_l2_dpm_data {
 #define DB_L2_DPM_DATA_RESERVED0_SHIFT 27
 #define DB_L2_DPM_DATA_SGE_NUM_MASK	0x7
 #define DB_L2_DPM_DATA_SGE_NUM_SHIFT	28
-#define DB_L2_DPM_DATA_GFS_SRC_EN_MASK	0x1
-#define DB_L2_DPM_DATA_GFS_SRC_EN_SHIFT	31
+#define DB_L2_DPM_DATA_TGFS_SRC_EN_MASK  0x1
+#define DB_L2_DPM_DATA_TGFS_SRC_EN_SHIFT 31
 };
 
 /* Structure for SGE in a DPM doorbell of type DPM_L2_BD */
diff --git a/include/linux/qed/eth_common.h b/include/linux/qed/eth_common.h
index d9416ad5ef59..95f5fd615852 100644
--- a/include/linux/qed/eth_common.h
+++ b/include/linux/qed/eth_common.h
@@ -38,9 +38,11 @@
 /********************/
 
 #define ETH_HSI_VER_MAJOR		3
-#define ETH_HSI_VER_MINOR		10
+#define ETH_HSI_VER_MINOR		11
 
-#define ETH_HSI_VER_NO_PKT_LEN_TUNN	5
+#define ETH_HSI_VER_NO_PKT_LEN_TUNN         5
+/* Maximum number of pinned L2 connections (CIDs) */
+#define ETH_PINNED_CONN_MAX_NUM             32
 
 #define ETH_CACHE_LINE_SIZE		64
 #define ETH_RX_CQE_GAP			32
@@ -61,6 +63,7 @@
 #define ETH_TX_MIN_BDS_PER_TUNN_IPV6_WITH_EXT_PKT	3
 #define ETH_TX_MIN_BDS_PER_IPV6_WITH_EXT_PKT		2
 #define ETH_TX_MIN_BDS_PER_PKT_W_LOOPBACK_MODE		2
+#define ETH_TX_MIN_BDS_PER_PKT_W_VPORT_FORWARDING	4
 #define ETH_TX_MAX_NON_LSO_PKT_LEN		(9700 - (4 + 4 + 12 + 8))
 #define ETH_TX_MAX_LSO_HDR_BYTES			510
 #define ETH_TX_LSO_WINDOW_BDS_NUM			(18 - 1)
@@ -75,9 +78,8 @@
 #define ETH_NUM_STATISTIC_COUNTERS_QUAD_VF_ZONE \
 	(ETH_NUM_STATISTIC_COUNTERS - 3 * MAX_NUM_VFS / 4)
 
-/* Maximum number of buffers, used for RX packet placement */
 #define ETH_RX_MAX_BUFF_PER_PKT		5
-#define ETH_RX_BD_THRESHOLD		12
+#define ETH_RX_BD_THRESHOLD             16
 
 /* Num of MAC/VLAN filters */
 #define ETH_NUM_MAC_FILTERS		512
@@ -96,24 +98,24 @@
 #define ETH_RSS_ENGINE_NUM_BB		127
 
 /* TPA constants */
-#define ETH_TPA_MAX_AGGS_NUM		64
-#define ETH_TPA_CQE_START_LEN_LIST_SIZE	ETH_RX_MAX_BUFF_PER_PKT
-#define ETH_TPA_CQE_CONT_LEN_LIST_SIZE	6
-#define ETH_TPA_CQE_END_LEN_LIST_SIZE	4
+#define ETH_TPA_MAX_AGGS_NUM                64
+#define ETH_TPA_CQE_START_BW_LEN_LIST_SIZE  2
+#define ETH_TPA_CQE_CONT_LEN_LIST_SIZE      6
+#define ETH_TPA_CQE_END_LEN_LIST_SIZE       4
 
 /* Control frame check constants */
-#define ETH_CTL_FRAME_ETH_TYPE_NUM	4
+#define ETH_CTL_FRAME_ETH_TYPE_NUM        4
 
 /* GFS constants */
 #define ETH_GFT_TRASHCAN_VPORT         0x1FF	/* GFT drop flow vport number */
 
 /* Destination port mode */
-enum dest_port_mode {
-	DEST_PORT_PHY,
-	DEST_PORT_LOOPBACK,
-	DEST_PORT_PHY_LOOPBACK,
-	DEST_PORT_DROP,
-	MAX_DEST_PORT_MODE
+enum dst_port_mode {
+	DST_PORT_PHY,
+	DST_PORT_LOOPBACK,
+	DST_PORT_PHY_LOOPBACK,
+	DST_PORT_DROP,
+	MAX_DST_PORT_MODE
 };
 
 /* Ethernet address type */
@@ -167,8 +169,8 @@ struct eth_tx_data_2nd_bd {
 #define ETH_TX_DATA_2ND_BD_TUNN_INNER_L2_HDR_SIZE_W_SHIFT	0
 #define ETH_TX_DATA_2ND_BD_TUNN_INNER_ETH_TYPE_MASK		0x3
 #define ETH_TX_DATA_2ND_BD_TUNN_INNER_ETH_TYPE_SHIFT		4
-#define ETH_TX_DATA_2ND_BD_DEST_PORT_MODE_MASK			0x3
-#define ETH_TX_DATA_2ND_BD_DEST_PORT_MODE_SHIFT			6
+#define ETH_TX_DATA_2ND_BD_DST_PORT_MODE_MASK			0x3
+#define ETH_TX_DATA_2ND_BD_DST_PORT_MODE_SHIFT			6
 #define ETH_TX_DATA_2ND_BD_START_BD_MASK			0x1
 #define ETH_TX_DATA_2ND_BD_START_BD_SHIFT			8
 #define ETH_TX_DATA_2ND_BD_TUNN_TYPE_MASK			0x3
@@ -244,8 +246,9 @@ struct eth_fast_path_rx_reg_cqe {
 	struct eth_tunnel_parsing_flags tunnel_pars_flags;
 	u8 bd_num;
 	u8 reserved;
-	__le16 flow_id;
-	u8 reserved1[11];
+	__le16 reserved2;
+	__le32 flow_id_or_resource_id;
+	u8 reserved1[7];
 	struct eth_pmd_flow_flags pmd_flags;
 };
 
@@ -296,9 +299,10 @@ struct eth_fast_path_rx_tpa_start_cqe {
 	struct eth_tunnel_parsing_flags tunnel_pars_flags;
 	u8 tpa_agg_index;
 	u8 header_len;
-	__le16 ext_bd_len_list[ETH_TPA_CQE_START_LEN_LIST_SIZE];
-	__le16 flow_id;
-	u8 reserved;
+	__le16 bw_ext_bd_len_list[ETH_TPA_CQE_START_BW_LEN_LIST_SIZE];
+	__le16 reserved2;
+	__le32 flow_id_or_resource_id;
+	u8 reserved[3];
 	struct eth_pmd_flow_flags pmd_flags;
 };
 
@@ -407,6 +411,29 @@ struct eth_tx_3rd_bd {
 	struct eth_tx_data_3rd_bd data;
 };
 
+/* The parsing information data for the forth tx bd of a given packet. */
+struct eth_tx_data_4th_bd {
+	u8 dst_vport_id;
+	u8 reserved4;
+	__le16 bitfields;
+#define ETH_TX_DATA_4TH_BD_DST_VPORT_ID_VALID_MASK  0x1
+#define ETH_TX_DATA_4TH_BD_DST_VPORT_ID_VALID_SHIFT 0
+#define ETH_TX_DATA_4TH_BD_RESERVED1_MASK           0x7F
+#define ETH_TX_DATA_4TH_BD_RESERVED1_SHIFT          1
+#define ETH_TX_DATA_4TH_BD_START_BD_MASK            0x1
+#define ETH_TX_DATA_4TH_BD_START_BD_SHIFT           8
+#define ETH_TX_DATA_4TH_BD_RESERVED2_MASK           0x7F
+#define ETH_TX_DATA_4TH_BD_RESERVED2_SHIFT          9
+	__le16 reserved3;
+};
+
+/* The forth tx bd of a given packet */
+struct eth_tx_4th_bd {
+	struct regpair addr; /* Single continuous buffer */
+	__le16 nbytes; /* Number of bytes in this BD */
+	struct eth_tx_data_4th_bd data; /* Parsing information data */
+};
+
 /* Complementary information for the regular tx bd of a given packet */
 struct eth_tx_data_bd {
 	__le16 reserved0;
@@ -431,6 +458,7 @@ union eth_tx_bd_types {
 	struct eth_tx_1st_bd first_bd;
 	struct eth_tx_2nd_bd second_bd;
 	struct eth_tx_3rd_bd third_bd;
+	struct eth_tx_4th_bd fourth_bd;
 	struct eth_tx_bd reg_bd;
 };
 
@@ -443,6 +471,12 @@ enum eth_tx_tunn_type {
 	MAX_ETH_TX_TUNN_TYPE
 };
 
+/* Mstorm Queue Zone */
+struct mstorm_eth_queue_zone {
+	struct eth_rx_prod_data rx_producers;
+	__le32 reserved[3];
+};
+
 /* Ystorm Queue Zone */
 struct xstorm_eth_queue_zone {
 	struct coalescing_timeset int_coalescing_timeset;
-- 
cgit v1.2.3-59-g8ed1b


From 30d5f85895fad1b3fec34b86cb779d4f56f85e32 Mon Sep 17 00:00:00 2001
From: Michal Kalderon <michal.kalderon@marvell.com>
Date: Mon, 27 Jan 2020 15:26:16 +0200
Subject: qed: FW 8.42.2.0 Add fw overlay feature

This feature enables the FW to page out FW code when required

Signed-off-by: Ariel Elior <ariel.elior@marvell.com>
Signed-off-by: Michal Kalderon <michal.kalderon@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/qlogic/qed/qed.h              |   3 +
 drivers/net/ethernet/qlogic/qed/qed_dev.c          |  18 ++-
 drivers/net/ethernet/qlogic/qed/qed_hsi.h          |  58 +++++++++
 .../net/ethernet/qlogic/qed/qed_init_fw_funcs.c    | 141 +++++++++++++++++++++
 drivers/net/ethernet/qlogic/qed/qed_init_ops.c     |   5 +
 5 files changed, 224 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/qlogic/qed/qed.h b/drivers/net/ethernet/qlogic/qed/qed.h
index cfbfe7441ecb..fedb1ab0b223 100644
--- a/drivers/net/ethernet/qlogic/qed/qed.h
+++ b/drivers/net/ethernet/qlogic/qed/qed.h
@@ -462,6 +462,8 @@ struct qed_fw_data {
 	const u8		*modes_tree_buf;
 	union init_op		*init_ops;
 	const u32		*arr_data;
+	const u32		*fw_overlays;
+	u32			fw_overlays_len;
 	u32			init_ops_size;
 };
 
@@ -686,6 +688,7 @@ struct qed_hwfn {
 	/* Nvm images number and attributes */
 	struct qed_nvm_image_info nvm_info;
 
+	struct phys_mem_desc *fw_overlay_mem;
 	struct qed_ptt *p_arfs_ptt;
 
 	struct qed_simd_fp_handler	simd_proto_handler[64];
diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c
index e3e0376c13d6..df97810b09b9 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_dev.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c
@@ -1412,6 +1412,7 @@ void qed_resc_free(struct qed_dev *cdev)
 		qed_dmae_info_free(p_hwfn);
 		qed_dcbx_info_free(p_hwfn);
 		qed_dbg_user_data_free(p_hwfn);
+		qed_fw_overlay_mem_free(p_hwfn, p_hwfn->fw_overlay_mem);
 
 		/* Destroy doorbell recovery mechanism */
 		qed_db_recovery_teardown(p_hwfn);
@@ -2893,6 +2894,8 @@ static int qed_hw_init_pf(struct qed_hwfn *p_hwfn,
 	if (rc)
 		return rc;
 
+	qed_fw_overlay_init_ram(p_hwfn, p_ptt, p_hwfn->fw_overlay_mem);
+
 	/* Pure runtime initializations - directly to the HW  */
 	qed_int_igu_init_pure_rt(p_hwfn, p_ptt, true, true);
 
@@ -3002,8 +3005,10 @@ int qed_hw_init(struct qed_dev *cdev, struct qed_hw_init_params *p_params)
 	u32 load_code, resp, param, drv_mb_param;
 	bool b_default_mtu = true;
 	struct qed_hwfn *p_hwfn;
-	int rc = 0, i;
+	const u32 *fw_overlays;
+	u32 fw_overlays_len;
 	u16 ether_type;
+	int rc = 0, i;
 
 	if ((p_params->int_mode == QED_INT_MODE_MSI) && (cdev->num_hwfns > 1)) {
 		DP_NOTICE(cdev, "MSI mode is not supported for CMT devices\n");
@@ -3104,6 +3109,17 @@ int qed_hw_init(struct qed_dev *cdev, struct qed_hw_init_params *p_params)
 		 */
 		qed_pglueb_clear_err(p_hwfn, p_hwfn->p_main_ptt);
 
+		fw_overlays = cdev->fw_data->fw_overlays;
+		fw_overlays_len = cdev->fw_data->fw_overlays_len;
+		p_hwfn->fw_overlay_mem =
+		    qed_fw_overlay_mem_alloc(p_hwfn, fw_overlays,
+					     fw_overlays_len);
+		if (!p_hwfn->fw_overlay_mem) {
+			DP_NOTICE(p_hwfn,
+				  "Failed to allocate fw overlay memory\n");
+			goto load_err;
+		}
+
 		switch (load_code) {
 		case FW_MSG_CODE_DRV_LOAD_ENGINE:
 			rc = qed_hw_init_common(p_hwfn, p_hwfn->p_main_ptt,
diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
index 06b94508115a..67c048415d3b 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
@@ -1843,6 +1843,19 @@ struct sdm_op_gen {
 #define SDM_OP_GEN_RESERVED_SHIFT	20
 };
 
+/* Physical memory descriptor */
+struct phys_mem_desc {
+	dma_addr_t phys_addr;
+	void *virt_addr;
+	u32 size;		/* In bytes */
+};
+
+/* Virtual memory descriptor */
+struct virt_mem_desc {
+	void *ptr;
+	u32 size;		/* In bytes */
+};
+
 /****************************************/
 /* Debug Tools HSI constants and macros */
 /****************************************/
@@ -2872,6 +2885,15 @@ enum bin_init_buffer_type {
 	MAX_BIN_INIT_BUFFER_TYPE
 };
 
+/* FW overlay buffer header */
+struct fw_overlay_buf_hdr {
+	u32 data;
+#define FW_OVERLAY_BUF_HDR_STORM_ID_MASK  0xFF
+#define FW_OVERLAY_BUF_HDR_STORM_ID_SHIFT 0
+#define FW_OVERLAY_BUF_HDR_BUF_SIZE_MASK  0xFFFFFF
+#define FW_OVERLAY_BUF_HDR_BUF_SIZE_SHIFT 8
+};
+
 /* init array header: raw */
 struct init_array_raw_hdr {
 	u32 data;
@@ -4345,6 +4367,42 @@ void qed_memset_task_ctx(void *p_ctx_mem, u32 ctx_size, u8 ctx_type);
 void qed_set_rdma_error_level(struct qed_hwfn *p_hwfn,
 			      struct qed_ptt *p_ptt,
 			      u8 assert_level[NUM_STORMS]);
+/**
+ * @brief qed_fw_overlay_mem_alloc - Allocates and fills the FW overlay memory.
+ *
+ * @param p_hwfn - HW device data
+ * @param fw_overlay_in_buf - the input FW overlay buffer.
+ * @param buf_size - the size of the input FW overlay buffer in bytes.
+ *		     must be aligned to dwords.
+ * @param fw_overlay_out_mem - OUT: a pointer to the allocated overlays memory.
+ *
+ * @return a pointer to the allocated overlays memory,
+ * or NULL in case of failures.
+ */
+struct phys_mem_desc *
+qed_fw_overlay_mem_alloc(struct qed_hwfn *p_hwfn,
+			 const u32 * const fw_overlay_in_buf,
+			 u32 buf_size_in_bytes);
+
+/**
+ * @brief qed_fw_overlay_init_ram - Initializes the FW overlay RAM.
+ *
+ * @param p_hwfn - HW device data.
+ * @param p_ptt - ptt window used for writing the registers.
+ * @param fw_overlay_mem - the allocated FW overlay memory.
+ */
+void qed_fw_overlay_init_ram(struct qed_hwfn *p_hwfn,
+			     struct qed_ptt *p_ptt,
+			     struct phys_mem_desc *fw_overlay_mem);
+
+/**
+ * @brief qed_fw_overlay_mem_free - Frees the FW overlay memory.
+ *
+ * @param p_hwfn - HW device data.
+ * @param fw_overlay_mem - the allocated FW overlay memory to free.
+ */
+void qed_fw_overlay_mem_free(struct qed_hwfn *p_hwfn,
+			     struct phys_mem_desc *fw_overlay_mem);
 
 /* Ystorm flow control mode. Use enum fw_flow_ctrl_mode */
 #define YSTORM_FLOW_CONTROL_MODE_OFFSET			(IRO[0].base)
diff --git a/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c b/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c
index 6fdf9ab98685..2f1049b0b93a 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c
@@ -1572,3 +1572,144 @@ void qed_set_rdma_error_level(struct qed_hwfn *p_hwfn,
 		qed_wr(p_hwfn, p_ptt, ram_addr, assert_level[storm_id]);
 	}
 }
+
+#define PHYS_ADDR_DWORDS        DIV_ROUND_UP(sizeof(dma_addr_t), 4)
+#define OVERLAY_HDR_SIZE_DWORDS (sizeof(struct fw_overlay_buf_hdr) / 4)
+
+static u32 qed_get_overlay_addr_ram_addr(struct qed_hwfn *p_hwfn, u8 storm_id)
+{
+	switch (storm_id) {
+	case 0:
+		return TSEM_REG_FAST_MEMORY + SEM_FAST_REG_INT_RAM +
+		    TSTORM_OVERLAY_BUF_ADDR_OFFSET;
+	case 1:
+		return MSEM_REG_FAST_MEMORY + SEM_FAST_REG_INT_RAM +
+		    MSTORM_OVERLAY_BUF_ADDR_OFFSET;
+	case 2:
+		return USEM_REG_FAST_MEMORY + SEM_FAST_REG_INT_RAM +
+		    USTORM_OVERLAY_BUF_ADDR_OFFSET;
+	case 3:
+		return XSEM_REG_FAST_MEMORY + SEM_FAST_REG_INT_RAM +
+		    XSTORM_OVERLAY_BUF_ADDR_OFFSET;
+	case 4:
+		return YSEM_REG_FAST_MEMORY + SEM_FAST_REG_INT_RAM +
+		    YSTORM_OVERLAY_BUF_ADDR_OFFSET;
+	case 5:
+		return PSEM_REG_FAST_MEMORY + SEM_FAST_REG_INT_RAM +
+		    PSTORM_OVERLAY_BUF_ADDR_OFFSET;
+
+	default:
+		return 0;
+	}
+}
+
+struct phys_mem_desc *qed_fw_overlay_mem_alloc(struct qed_hwfn *p_hwfn,
+					       const u32 * const
+					       fw_overlay_in_buf,
+					       u32 buf_size_in_bytes)
+{
+	u32 buf_size = buf_size_in_bytes / sizeof(u32), buf_offset = 0;
+	struct phys_mem_desc *allocated_mem;
+
+	if (!buf_size)
+		return NULL;
+
+	allocated_mem = kcalloc(NUM_STORMS, sizeof(struct phys_mem_desc),
+				GFP_KERNEL);
+	if (!allocated_mem)
+		return NULL;
+
+	memset(allocated_mem, 0, NUM_STORMS * sizeof(struct phys_mem_desc));
+
+	/* For each Storm, set physical address in RAM */
+	while (buf_offset < buf_size) {
+		struct phys_mem_desc *storm_mem_desc;
+		struct fw_overlay_buf_hdr *hdr;
+		u32 storm_buf_size;
+		u8 storm_id;
+
+		hdr =
+		    (struct fw_overlay_buf_hdr *)&fw_overlay_in_buf[buf_offset];
+		storm_buf_size = GET_FIELD(hdr->data,
+					   FW_OVERLAY_BUF_HDR_BUF_SIZE);
+		storm_id = GET_FIELD(hdr->data, FW_OVERLAY_BUF_HDR_STORM_ID);
+		storm_mem_desc = allocated_mem + storm_id;
+		storm_mem_desc->size = storm_buf_size * sizeof(u32);
+
+		/* Allocate physical memory for Storm's overlays buffer */
+		storm_mem_desc->virt_addr =
+		    dma_alloc_coherent(&p_hwfn->cdev->pdev->dev,
+				       storm_mem_desc->size,
+				       &storm_mem_desc->phys_addr, GFP_KERNEL);
+		if (!storm_mem_desc->virt_addr)
+			break;
+
+		/* Skip overlays buffer header */
+		buf_offset += OVERLAY_HDR_SIZE_DWORDS;
+
+		/* Copy Storm's overlays buffer to allocated memory */
+		memcpy(storm_mem_desc->virt_addr,
+		       &fw_overlay_in_buf[buf_offset], storm_mem_desc->size);
+
+		/* Advance to next Storm */
+		buf_offset += storm_buf_size;
+	}
+
+	/* If memory allocation has failed, free all allocated memory */
+	if (buf_offset < buf_size) {
+		qed_fw_overlay_mem_free(p_hwfn, allocated_mem);
+		return NULL;
+	}
+
+	return allocated_mem;
+}
+
+void qed_fw_overlay_init_ram(struct qed_hwfn *p_hwfn,
+			     struct qed_ptt *p_ptt,
+			     struct phys_mem_desc *fw_overlay_mem)
+{
+	u8 storm_id;
+
+	for (storm_id = 0; storm_id < NUM_STORMS; storm_id++) {
+		struct phys_mem_desc *storm_mem_desc =
+		    (struct phys_mem_desc *)fw_overlay_mem + storm_id;
+		u32 ram_addr, i;
+
+		/* Skip Storms with no FW overlays */
+		if (!storm_mem_desc->virt_addr)
+			continue;
+
+		/* Calculate overlay RAM GRC address of current PF */
+		ram_addr = qed_get_overlay_addr_ram_addr(p_hwfn, storm_id) +
+			   sizeof(dma_addr_t) * p_hwfn->rel_pf_id;
+
+		/* Write Storm's overlay physical address to RAM */
+		for (i = 0; i < PHYS_ADDR_DWORDS; i++, ram_addr += sizeof(u32))
+			qed_wr(p_hwfn, p_ptt, ram_addr,
+			       ((u32 *)&storm_mem_desc->phys_addr)[i]);
+	}
+}
+
+void qed_fw_overlay_mem_free(struct qed_hwfn *p_hwfn,
+			     struct phys_mem_desc *fw_overlay_mem)
+{
+	u8 storm_id;
+
+	if (!fw_overlay_mem)
+		return;
+
+	for (storm_id = 0; storm_id < NUM_STORMS; storm_id++) {
+		struct phys_mem_desc *storm_mem_desc =
+		    (struct phys_mem_desc *)fw_overlay_mem + storm_id;
+
+		/* Free Storm's physical memory */
+		if (storm_mem_desc->virt_addr)
+			dma_free_coherent(&p_hwfn->cdev->pdev->dev,
+					  storm_mem_desc->size,
+					  storm_mem_desc->virt_addr,
+					  storm_mem_desc->phys_addr);
+	}
+
+	/* Free allocated virtual memory */
+	kfree(fw_overlay_mem);
+}
diff --git a/drivers/net/ethernet/qlogic/qed/qed_init_ops.c b/drivers/net/ethernet/qlogic/qed/qed_init_ops.c
index 5ca117a8734a..36f998c89e74 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_init_ops.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_init_ops.c
@@ -589,5 +589,10 @@ int qed_init_fw_data(struct qed_dev *cdev, const u8 *data)
 	len = buf_hdr[BIN_BUF_INIT_CMD].length;
 	fw->init_ops_size = len / sizeof(struct init_raw_op);
 
+	offset = buf_hdr[BIN_BUF_INIT_OVERLAYS].offset;
+	fw->fw_overlays = (u32 *)(data + offset);
+	len = buf_hdr[BIN_BUF_INIT_OVERLAYS].length;
+	fw->fw_overlays_len = len;
+
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 8a52bbab39c9791480cbae86c69ad0d47f62972e Mon Sep 17 00:00:00 2001
From: Michal Kalderon <michal.kalderon@marvell.com>
Date: Mon, 27 Jan 2020 15:26:17 +0200
Subject: qed: Debug feature: ilt and mdump

Part of the FW drop includes new debug capabilities implemented in the
qed_debug file. This patch dumps additional information during ethtool -d
for better debugging. The data dumped is the ilt (internal logical table)
and information gathered by the management firmware incase there was a
crash and driver was not able to extract the information (mdump).

Signed-off-by: Ariel Elior <ariel.elior@marvell.com>
Signed-off-by: Michal Kalderon <michal.kalderon@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/qlogic/qed/qed.h       |   1 +
 drivers/net/ethernet/qlogic/qed/qed_cxt.c   | 357 +++++++++----------
 drivers/net/ethernet/qlogic/qed/qed_cxt.h   | 130 +++++++
 drivers/net/ethernet/qlogic/qed/qed_debug.c | 521 +++++++++++++++++++++++++++-
 drivers/net/ethernet/qlogic/qed/qed_debug.h |   4 +
 drivers/net/ethernet/qlogic/qed/qed_hsi.h   |  18 +
 drivers/net/ethernet/qlogic/qed/qed_mcp.c   |   3 +
 include/linux/qed/qed_if.h                  |   1 +
 8 files changed, 821 insertions(+), 214 deletions(-)

diff --git a/drivers/net/ethernet/qlogic/qed/qed.h b/drivers/net/ethernet/qlogic/qed/qed.h
index fedb1ab0b223..a6dc40681a32 100644
--- a/drivers/net/ethernet/qlogic/qed/qed.h
+++ b/drivers/net/ethernet/qlogic/qed/qed.h
@@ -877,6 +877,7 @@ struct qed_dev {
 	struct qed_cb_ll2_info		*ll2;
 	u8				ll2_mac_address[ETH_ALEN];
 #endif
+	bool disable_ilt_dump;
 	DECLARE_HASHTABLE(connections, 10);
 	const struct firmware		*firmware;
 
diff --git a/drivers/net/ethernet/qlogic/qed/qed_cxt.c b/drivers/net/ethernet/qlogic/qed/qed_cxt.c
index 5ebafbc379db..fbfff2b1dc93 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_cxt.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_cxt.c
@@ -50,12 +50,6 @@
 #include "qed_reg_addr.h"
 #include "qed_sriov.h"
 
-/* Max number of connection types in HW (DQ/CDU etc.) */
-#define MAX_CONN_TYPES		PROTOCOLID_COMMON
-#define NUM_TASK_TYPES		2
-#define NUM_TASK_PF_SEGMENTS	4
-#define NUM_TASK_VF_SEGMENTS	1
-
 /* QM constants */
 #define QM_PQ_ELEMENT_SIZE	4 /* in bytes */
 
@@ -123,126 +117,6 @@ struct src_ent {
 /* Alignment is inherent to the type1_task_context structure */
 #define TYPE1_TASK_CXT_SIZE(p_hwfn) sizeof(union type1_task_context)
 
-/* PF per protocl configuration object */
-#define TASK_SEGMENTS   (NUM_TASK_PF_SEGMENTS + NUM_TASK_VF_SEGMENTS)
-#define TASK_SEGMENT_VF (NUM_TASK_PF_SEGMENTS)
-
-struct qed_tid_seg {
-	u32 count;
-	u8 type;
-	bool has_fl_mem;
-};
-
-struct qed_conn_type_cfg {
-	u32 cid_count;
-	u32 cids_per_vf;
-	struct qed_tid_seg tid_seg[TASK_SEGMENTS];
-};
-
-/* ILT Client configuration, Per connection type (protocol) resources. */
-#define ILT_CLI_PF_BLOCKS	(1 + NUM_TASK_PF_SEGMENTS * 2)
-#define ILT_CLI_VF_BLOCKS       (1 + NUM_TASK_VF_SEGMENTS * 2)
-#define CDUC_BLK		(0)
-#define SRQ_BLK                 (0)
-#define CDUT_SEG_BLK(n)         (1 + (u8)(n))
-#define CDUT_FL_SEG_BLK(n, X)   (1 + (n) + NUM_TASK_ ## X ## _SEGMENTS)
-
-enum ilt_clients {
-	ILT_CLI_CDUC,
-	ILT_CLI_CDUT,
-	ILT_CLI_QM,
-	ILT_CLI_TM,
-	ILT_CLI_SRC,
-	ILT_CLI_TSDM,
-	ILT_CLI_MAX
-};
-
-struct ilt_cfg_pair {
-	u32 reg;
-	u32 val;
-};
-
-struct qed_ilt_cli_blk {
-	u32 total_size; /* 0 means not active */
-	u32 real_size_in_page;
-	u32 start_line;
-	u32 dynamic_line_cnt;
-};
-
-struct qed_ilt_client_cfg {
-	bool active;
-
-	/* ILT boundaries */
-	struct ilt_cfg_pair first;
-	struct ilt_cfg_pair last;
-	struct ilt_cfg_pair p_size;
-
-	/* ILT client blocks for PF */
-	struct qed_ilt_cli_blk pf_blks[ILT_CLI_PF_BLOCKS];
-	u32 pf_total_lines;
-
-	/* ILT client blocks for VFs */
-	struct qed_ilt_cli_blk vf_blks[ILT_CLI_VF_BLOCKS];
-	u32 vf_total_lines;
-};
-
-/* Per Path -
- *      ILT shadow table
- *      Protocol acquired CID lists
- *      PF start line in ILT
- */
-struct qed_dma_mem {
-	dma_addr_t p_phys;
-	void *p_virt;
-	size_t size;
-};
-
-struct qed_cid_acquired_map {
-	u32		start_cid;
-	u32		max_count;
-	unsigned long	*cid_map;
-};
-
-struct qed_cxt_mngr {
-	/* Per protocl configuration */
-	struct qed_conn_type_cfg	conn_cfg[MAX_CONN_TYPES];
-
-	/* computed ILT structure */
-	struct qed_ilt_client_cfg	clients[ILT_CLI_MAX];
-
-	/* Task type sizes */
-	u32 task_type_size[NUM_TASK_TYPES];
-
-	/* total number of VFs for this hwfn -
-	 * ALL VFs are symmetric in terms of HW resources
-	 */
-	u32				vf_count;
-
-	/* Acquired CIDs */
-	struct qed_cid_acquired_map	acquired[MAX_CONN_TYPES];
-
-	struct qed_cid_acquired_map
-	acquired_vf[MAX_CONN_TYPES][MAX_NUM_VFS];
-
-	/* ILT  shadow table */
-	struct qed_dma_mem		*ilt_shadow;
-	u32				pf_start_line;
-
-	/* Mutex for a dynamic ILT allocation */
-	struct mutex mutex;
-
-	/* SRC T2 */
-	struct qed_dma_mem *t2;
-	u32 t2_num_pages;
-	u64 first_free;
-	u64 last_free;
-
-	/* total number of SRQ's for this hwfn */
-	u32 srq_count;
-
-	/* Maximal number of L2 steering filters */
-	u32 arfs_count;
-};
 static bool src_proto(enum protocol_type type)
 {
 	return type == PROTOCOLID_ISCSI ||
@@ -880,30 +754,60 @@ u32 qed_cxt_cfg_ilt_compute_excess(struct qed_hwfn *p_hwfn, u32 used_lines)
 
 static void qed_cxt_src_t2_free(struct qed_hwfn *p_hwfn)
 {
-	struct qed_cxt_mngr *p_mngr = p_hwfn->p_cxt_mngr;
+	struct qed_src_t2 *p_t2 = &p_hwfn->p_cxt_mngr->src_t2;
 	u32 i;
 
-	if (!p_mngr->t2)
+	if (!p_t2 || !p_t2->dma_mem)
 		return;
 
-	for (i = 0; i < p_mngr->t2_num_pages; i++)
-		if (p_mngr->t2[i].p_virt)
+	for (i = 0; i < p_t2->num_pages; i++)
+		if (p_t2->dma_mem[i].virt_addr)
 			dma_free_coherent(&p_hwfn->cdev->pdev->dev,
-					  p_mngr->t2[i].size,
-					  p_mngr->t2[i].p_virt,
-					  p_mngr->t2[i].p_phys);
+					  p_t2->dma_mem[i].size,
+					  p_t2->dma_mem[i].virt_addr,
+					  p_t2->dma_mem[i].phys_addr);
+
+	kfree(p_t2->dma_mem);
+	p_t2->dma_mem = NULL;
+}
+
+static int
+qed_cxt_t2_alloc_pages(struct qed_hwfn *p_hwfn,
+		       struct qed_src_t2 *p_t2, u32 total_size, u32 page_size)
+{
+	void **p_virt;
+	u32 size, i;
+
+	if (!p_t2 || !p_t2->dma_mem)
+		return -EINVAL;
 
-	kfree(p_mngr->t2);
-	p_mngr->t2 = NULL;
+	for (i = 0; i < p_t2->num_pages; i++) {
+		size = min_t(u32, total_size, page_size);
+		p_virt = &p_t2->dma_mem[i].virt_addr;
+
+		*p_virt = dma_alloc_coherent(&p_hwfn->cdev->pdev->dev,
+					     size,
+					     &p_t2->dma_mem[i].phys_addr,
+					     GFP_KERNEL);
+		if (!p_t2->dma_mem[i].virt_addr)
+			return -ENOMEM;
+
+		memset(*p_virt, 0, size);
+		p_t2->dma_mem[i].size = size;
+		total_size -= size;
+	}
+
+	return 0;
 }
 
 static int qed_cxt_src_t2_alloc(struct qed_hwfn *p_hwfn)
 {
 	struct qed_cxt_mngr *p_mngr = p_hwfn->p_cxt_mngr;
 	u32 conn_num, total_size, ent_per_page, psz, i;
+	struct phys_mem_desc *p_t2_last_page;
 	struct qed_ilt_client_cfg *p_src;
 	struct qed_src_iids src_iids;
-	struct qed_dma_mem *p_t2;
+	struct qed_src_t2 *p_t2;
 	int rc;
 
 	memset(&src_iids, 0, sizeof(src_iids));
@@ -921,49 +825,39 @@ static int qed_cxt_src_t2_alloc(struct qed_hwfn *p_hwfn)
 
 	/* use the same page size as the SRC ILT client */
 	psz = ILT_PAGE_IN_BYTES(p_src->p_size.val);
-	p_mngr->t2_num_pages = DIV_ROUND_UP(total_size, psz);
+	p_t2 = &p_mngr->src_t2;
+	p_t2->num_pages = DIV_ROUND_UP(total_size, psz);
 
 	/* allocate t2 */
-	p_mngr->t2 = kcalloc(p_mngr->t2_num_pages, sizeof(struct qed_dma_mem),
-			     GFP_KERNEL);
-	if (!p_mngr->t2) {
+	p_t2->dma_mem = kcalloc(p_t2->num_pages, sizeof(struct phys_mem_desc),
+				GFP_KERNEL);
+	if (!p_t2->dma_mem) {
+		DP_NOTICE(p_hwfn, "Failed to allocate t2 table\n");
 		rc = -ENOMEM;
 		goto t2_fail;
 	}
 
-	/* allocate t2 pages */
-	for (i = 0; i < p_mngr->t2_num_pages; i++) {
-		u32 size = min_t(u32, total_size, psz);
-		void **p_virt = &p_mngr->t2[i].p_virt;
-
-		*p_virt = dma_alloc_coherent(&p_hwfn->cdev->pdev->dev, size,
-					     &p_mngr->t2[i].p_phys,
-					     GFP_KERNEL);
-		if (!p_mngr->t2[i].p_virt) {
-			rc = -ENOMEM;
-			goto t2_fail;
-		}
-		p_mngr->t2[i].size = size;
-		total_size -= size;
-	}
+	rc = qed_cxt_t2_alloc_pages(p_hwfn, p_t2, total_size, psz);
+	if (rc)
+		goto t2_fail;
 
 	/* Set the t2 pointers */
 
 	/* entries per page - must be a power of two */
 	ent_per_page = psz / sizeof(struct src_ent);
 
-	p_mngr->first_free = (u64) p_mngr->t2[0].p_phys;
+	p_t2->first_free = (u64)p_t2->dma_mem[0].phys_addr;
 
-	p_t2 = &p_mngr->t2[(conn_num - 1) / ent_per_page];
-	p_mngr->last_free = (u64) p_t2->p_phys +
+	p_t2_last_page = &p_t2->dma_mem[(conn_num - 1) / ent_per_page];
+	p_t2->last_free = (u64)p_t2_last_page->phys_addr +
 	    ((conn_num - 1) & (ent_per_page - 1)) * sizeof(struct src_ent);
 
-	for (i = 0; i < p_mngr->t2_num_pages; i++) {
+	for (i = 0; i < p_t2->num_pages; i++) {
 		u32 ent_num = min_t(u32,
 				    ent_per_page,
 				    conn_num);
-		struct src_ent *entries = p_mngr->t2[i].p_virt;
-		u64 p_ent_phys = (u64) p_mngr->t2[i].p_phys, val;
+		struct src_ent *entries = p_t2->dma_mem[i].virt_addr;
+		u64 p_ent_phys = (u64)p_t2->dma_mem[i].phys_addr, val;
 		u32 j;
 
 		for (j = 0; j < ent_num - 1; j++) {
@@ -971,8 +865,8 @@ static int qed_cxt_src_t2_alloc(struct qed_hwfn *p_hwfn)
 			entries[j].next = cpu_to_be64(val);
 		}
 
-		if (i < p_mngr->t2_num_pages - 1)
-			val = (u64) p_mngr->t2[i + 1].p_phys;
+		if (i < p_t2->num_pages - 1)
+			val = (u64)p_t2->dma_mem[i + 1].phys_addr;
 		else
 			val = 0;
 		entries[j].next = cpu_to_be64(val);
@@ -988,7 +882,7 @@ t2_fail:
 }
 
 #define for_each_ilt_valid_client(pos, clients)	\
-	for (pos = 0; pos < ILT_CLI_MAX; pos++)	\
+	for (pos = 0; pos < MAX_ILT_CLIENTS; pos++)	\
 		if (!clients[pos].active) {	\
 			continue;		\
 		} else				\
@@ -1014,13 +908,13 @@ static void qed_ilt_shadow_free(struct qed_hwfn *p_hwfn)
 	ilt_size = qed_cxt_ilt_shadow_size(p_cli);
 
 	for (i = 0; p_mngr->ilt_shadow && i < ilt_size; i++) {
-		struct qed_dma_mem *p_dma = &p_mngr->ilt_shadow[i];
+		struct phys_mem_desc *p_dma = &p_mngr->ilt_shadow[i];
 
-		if (p_dma->p_virt)
+		if (p_dma->virt_addr)
 			dma_free_coherent(&p_hwfn->cdev->pdev->dev,
-					  p_dma->size, p_dma->p_virt,
-					  p_dma->p_phys);
-		p_dma->p_virt = NULL;
+					  p_dma->size, p_dma->virt_addr,
+					  p_dma->phys_addr);
+		p_dma->virt_addr = NULL;
 	}
 	kfree(p_mngr->ilt_shadow);
 }
@@ -1030,7 +924,7 @@ static int qed_ilt_blk_alloc(struct qed_hwfn *p_hwfn,
 			     enum ilt_clients ilt_client,
 			     u32 start_line_offset)
 {
-	struct qed_dma_mem *ilt_shadow = p_hwfn->p_cxt_mngr->ilt_shadow;
+	struct phys_mem_desc *ilt_shadow = p_hwfn->p_cxt_mngr->ilt_shadow;
 	u32 lines, line, sz_left, lines_to_skip = 0;
 
 	/* Special handling for RoCE that supports dynamic allocation */
@@ -1059,8 +953,8 @@ static int qed_ilt_blk_alloc(struct qed_hwfn *p_hwfn,
 		if (!p_virt)
 			return -ENOMEM;
 
-		ilt_shadow[line].p_phys = p_phys;
-		ilt_shadow[line].p_virt = p_virt;
+		ilt_shadow[line].phys_addr = p_phys;
+		ilt_shadow[line].virt_addr = p_virt;
 		ilt_shadow[line].size = size;
 
 		DP_VERBOSE(p_hwfn, QED_MSG_ILT,
@@ -1083,7 +977,7 @@ static int qed_ilt_shadow_alloc(struct qed_hwfn *p_hwfn)
 	int rc;
 
 	size = qed_cxt_ilt_shadow_size(clients);
-	p_mngr->ilt_shadow = kcalloc(size, sizeof(struct qed_dma_mem),
+	p_mngr->ilt_shadow = kcalloc(size, sizeof(struct phys_mem_desc),
 				     GFP_KERNEL);
 	if (!p_mngr->ilt_shadow) {
 		rc = -ENOMEM;
@@ -1092,7 +986,7 @@ static int qed_ilt_shadow_alloc(struct qed_hwfn *p_hwfn)
 
 	DP_VERBOSE(p_hwfn, QED_MSG_ILT,
 		   "Allocated 0x%x bytes for ilt shadow\n",
-		   (u32)(size * sizeof(struct qed_dma_mem)));
+		   (u32)(size * sizeof(struct phys_mem_desc)));
 
 	for_each_ilt_valid_client(i, clients) {
 		for (j = 0; j < ILT_CLI_PF_BLOCKS; j++) {
@@ -1238,15 +1132,20 @@ int qed_cxt_mngr_alloc(struct qed_hwfn *p_hwfn)
 	clients[ILT_CLI_TSDM].last.reg = ILT_CFG_REG(TSDM, LAST_ILT);
 	clients[ILT_CLI_TSDM].p_size.reg = ILT_CFG_REG(TSDM, P_SIZE);
 	/* default ILT page size for all clients is 64K */
-	for (i = 0; i < ILT_CLI_MAX; i++)
+	for (i = 0; i < MAX_ILT_CLIENTS; i++)
 		p_mngr->clients[i].p_size.val = ILT_DEFAULT_HW_P_SIZE;
 
+	p_mngr->conn_ctx_size = CONN_CXT_SIZE(p_hwfn);
+
 	/* Initialize task sizes */
 	p_mngr->task_type_size[0] = TYPE0_TASK_CXT_SIZE(p_hwfn);
 	p_mngr->task_type_size[1] = TYPE1_TASK_CXT_SIZE(p_hwfn);
 
-	if (p_hwfn->cdev->p_iov_info)
+	if (p_hwfn->cdev->p_iov_info) {
 		p_mngr->vf_count = p_hwfn->cdev->p_iov_info->total_vfs;
+		p_mngr->first_vf_in_pf =
+			p_hwfn->cdev->p_iov_info->first_vf_in_pf;
+	}
 	/* Initialize the dynamic ILT allocation mutex */
 	mutex_init(&p_mngr->mutex);
 
@@ -1673,7 +1572,7 @@ static void qed_ilt_init_pf(struct qed_hwfn *p_hwfn)
 {
 	struct qed_ilt_client_cfg *clients;
 	struct qed_cxt_mngr *p_mngr;
-	struct qed_dma_mem *p_shdw;
+	struct phys_mem_desc *p_shdw;
 	u32 line, rt_offst, i;
 
 	qed_ilt_bounds_init(p_hwfn);
@@ -1698,15 +1597,15 @@ static void qed_ilt_init_pf(struct qed_hwfn *p_hwfn)
 			/** p_virt could be NULL incase of dynamic
 			 *  allocation
 			 */
-			if (p_shdw[line].p_virt) {
+			if (p_shdw[line].virt_addr) {
 				SET_FIELD(ilt_hw_entry, ILT_ENTRY_VALID, 1ULL);
 				SET_FIELD(ilt_hw_entry, ILT_ENTRY_PHY_ADDR,
-					  (p_shdw[line].p_phys >> 12));
+					  (p_shdw[line].phys_addr >> 12));
 
 				DP_VERBOSE(p_hwfn, QED_MSG_ILT,
 					   "Setting RT[0x%08x] from ILT[0x%08x] [Client is %d] to Physical addr: 0x%llx\n",
 					   rt_offst, line, i,
-					   (u64)(p_shdw[line].p_phys >> 12));
+					   (u64)(p_shdw[line].phys_addr >> 12));
 			}
 
 			STORE_RT_REG_AGG(p_hwfn, rt_offst, ilt_hw_entry);
@@ -2049,10 +1948,10 @@ int qed_cxt_get_cid_info(struct qed_hwfn *p_hwfn, struct qed_cxt_info *p_info)
 	line = p_info->iid / cxts_per_p;
 
 	/* Make sure context is allocated (dynamic allocation) */
-	if (!p_mngr->ilt_shadow[line].p_virt)
+	if (!p_mngr->ilt_shadow[line].virt_addr)
 		return -EINVAL;
 
-	p_info->p_cxt = p_mngr->ilt_shadow[line].p_virt +
+	p_info->p_cxt = p_mngr->ilt_shadow[line].virt_addr +
 			p_info->iid % cxts_per_p * conn_cxt_size;
 
 	DP_VERBOSE(p_hwfn, (QED_MSG_ILT | QED_MSG_CXT),
@@ -2233,7 +2132,7 @@ int qed_cxt_get_tid_mem_info(struct qed_hwfn *p_hwfn,
 	for (i = 0; i < total_lines; i++) {
 		shadow_line = i + p_fl_seg->start_line -
 		    p_hwfn->p_cxt_mngr->pf_start_line;
-		p_info->blocks[i] = p_mngr->ilt_shadow[shadow_line].p_virt;
+		p_info->blocks[i] = p_mngr->ilt_shadow[shadow_line].virt_addr;
 	}
 	p_info->waste = ILT_PAGE_IN_BYTES(p_cli->p_size.val) -
 	    p_fl_seg->real_size_in_page;
@@ -2295,7 +2194,7 @@ qed_cxt_dynamic_ilt_alloc(struct qed_hwfn *p_hwfn,
 
 	mutex_lock(&p_hwfn->p_cxt_mngr->mutex);
 
-	if (p_hwfn->p_cxt_mngr->ilt_shadow[shadow_line].p_virt)
+	if (p_hwfn->p_cxt_mngr->ilt_shadow[shadow_line].virt_addr)
 		goto out0;
 
 	p_ptt = qed_ptt_acquire(p_hwfn);
@@ -2333,8 +2232,8 @@ qed_cxt_dynamic_ilt_alloc(struct qed_hwfn *p_hwfn,
 		}
 	}
 
-	p_hwfn->p_cxt_mngr->ilt_shadow[shadow_line].p_virt = p_virt;
-	p_hwfn->p_cxt_mngr->ilt_shadow[shadow_line].p_phys = p_phys;
+	p_hwfn->p_cxt_mngr->ilt_shadow[shadow_line].virt_addr = p_virt;
+	p_hwfn->p_cxt_mngr->ilt_shadow[shadow_line].phys_addr = p_phys;
 	p_hwfn->p_cxt_mngr->ilt_shadow[shadow_line].size =
 	    p_blk->real_size_in_page;
 
@@ -2344,9 +2243,9 @@ qed_cxt_dynamic_ilt_alloc(struct qed_hwfn *p_hwfn,
 
 	ilt_hw_entry = 0;
 	SET_FIELD(ilt_hw_entry, ILT_ENTRY_VALID, 1ULL);
-	SET_FIELD(ilt_hw_entry,
-		  ILT_ENTRY_PHY_ADDR,
-		  (p_hwfn->p_cxt_mngr->ilt_shadow[shadow_line].p_phys >> 12));
+	SET_FIELD(ilt_hw_entry, ILT_ENTRY_PHY_ADDR,
+		  (p_hwfn->p_cxt_mngr->ilt_shadow[shadow_line].phys_addr
+		   >> 12));
 
 	/* Write via DMAE since the PSWRQ2_REG_ILT_MEMORY line is a wide-bus */
 	qed_dmae_host2grc(p_hwfn, p_ptt, (u64) (uintptr_t)&ilt_hw_entry,
@@ -2433,16 +2332,16 @@ qed_cxt_free_ilt_range(struct qed_hwfn *p_hwfn,
 	}
 
 	for (i = shadow_start_line; i < shadow_end_line; i++) {
-		if (!p_hwfn->p_cxt_mngr->ilt_shadow[i].p_virt)
+		if (!p_hwfn->p_cxt_mngr->ilt_shadow[i].virt_addr)
 			continue;
 
 		dma_free_coherent(&p_hwfn->cdev->pdev->dev,
 				  p_hwfn->p_cxt_mngr->ilt_shadow[i].size,
-				  p_hwfn->p_cxt_mngr->ilt_shadow[i].p_virt,
-				  p_hwfn->p_cxt_mngr->ilt_shadow[i].p_phys);
+				  p_hwfn->p_cxt_mngr->ilt_shadow[i].virt_addr,
+				  p_hwfn->p_cxt_mngr->ilt_shadow[i].phys_addr);
 
-		p_hwfn->p_cxt_mngr->ilt_shadow[i].p_virt = NULL;
-		p_hwfn->p_cxt_mngr->ilt_shadow[i].p_phys = 0;
+		p_hwfn->p_cxt_mngr->ilt_shadow[i].virt_addr = NULL;
+		p_hwfn->p_cxt_mngr->ilt_shadow[i].phys_addr = 0;
 		p_hwfn->p_cxt_mngr->ilt_shadow[i].size = 0;
 
 		/* compute absolute offset */
@@ -2546,8 +2445,76 @@ int qed_cxt_get_task_ctx(struct qed_hwfn *p_hwfn,
 
 	ilt_idx = tid / num_tids_per_block + p_seg->start_line -
 		  p_mngr->pf_start_line;
-	*pp_task_ctx = (u8 *)p_mngr->ilt_shadow[ilt_idx].p_virt +
+	*pp_task_ctx = (u8 *)p_mngr->ilt_shadow[ilt_idx].virt_addr +
 		       (tid % num_tids_per_block) * tid_size;
 
 	return 0;
 }
+
+static u16 qed_blk_calculate_pages(struct qed_ilt_cli_blk *p_blk)
+{
+	if (p_blk->real_size_in_page == 0)
+		return 0;
+
+	return DIV_ROUND_UP(p_blk->total_size, p_blk->real_size_in_page);
+}
+
+u16 qed_get_cdut_num_pf_init_pages(struct qed_hwfn *p_hwfn)
+{
+	struct qed_ilt_client_cfg *p_cli;
+	struct qed_ilt_cli_blk *p_blk;
+	u16 i, pages = 0;
+
+	p_cli = &p_hwfn->p_cxt_mngr->clients[ILT_CLI_CDUT];
+	for (i = 0; i < NUM_TASK_PF_SEGMENTS; i++) {
+		p_blk = &p_cli->pf_blks[CDUT_FL_SEG_BLK(i, PF)];
+		pages += qed_blk_calculate_pages(p_blk);
+	}
+
+	return pages;
+}
+
+u16 qed_get_cdut_num_vf_init_pages(struct qed_hwfn *p_hwfn)
+{
+	struct qed_ilt_client_cfg *p_cli;
+	struct qed_ilt_cli_blk *p_blk;
+	u16 i, pages = 0;
+
+	p_cli = &p_hwfn->p_cxt_mngr->clients[ILT_CLI_CDUT];
+	for (i = 0; i < NUM_TASK_VF_SEGMENTS; i++) {
+		p_blk = &p_cli->vf_blks[CDUT_FL_SEG_BLK(i, VF)];
+		pages += qed_blk_calculate_pages(p_blk);
+	}
+
+	return pages;
+}
+
+u16 qed_get_cdut_num_pf_work_pages(struct qed_hwfn *p_hwfn)
+{
+	struct qed_ilt_client_cfg *p_cli;
+	struct qed_ilt_cli_blk *p_blk;
+	u16 i, pages = 0;
+
+	p_cli = &p_hwfn->p_cxt_mngr->clients[ILT_CLI_CDUT];
+	for (i = 0; i < NUM_TASK_PF_SEGMENTS; i++) {
+		p_blk = &p_cli->pf_blks[CDUT_SEG_BLK(i)];
+		pages += qed_blk_calculate_pages(p_blk);
+	}
+
+	return pages;
+}
+
+u16 qed_get_cdut_num_vf_work_pages(struct qed_hwfn *p_hwfn)
+{
+	struct qed_ilt_client_cfg *p_cli;
+	struct qed_ilt_cli_blk *p_blk;
+	u16 pages = 0, i;
+
+	p_cli = &p_hwfn->p_cxt_mngr->clients[ILT_CLI_CDUT];
+	for (i = 0; i < NUM_TASK_VF_SEGMENTS; i++) {
+		p_blk = &p_cli->vf_blks[CDUT_SEG_BLK(i)];
+		pages += qed_blk_calculate_pages(p_blk);
+	}
+
+	return pages;
+}
diff --git a/drivers/net/ethernet/qlogic/qed/qed_cxt.h b/drivers/net/ethernet/qlogic/qed/qed_cxt.h
index 758a8b4c0de8..c4e815f6cabd 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_cxt.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_cxt.h
@@ -242,4 +242,134 @@ int qed_cxt_free_proto_ilt(struct qed_hwfn *p_hwfn, enum protocol_type proto);
 #define QED_CTX_FL_MEM 1
 int qed_cxt_get_task_ctx(struct qed_hwfn *p_hwfn,
 			 u32 tid, u8 ctx_type, void **task_ctx);
+
+/* Max number of connection types in HW (DQ/CDU etc.) */
+#define MAX_CONN_TYPES          PROTOCOLID_COMMON
+#define NUM_TASK_TYPES          2
+#define NUM_TASK_PF_SEGMENTS    4
+#define NUM_TASK_VF_SEGMENTS    1
+
+/* PF per protocl configuration object */
+#define TASK_SEGMENTS   (NUM_TASK_PF_SEGMENTS + NUM_TASK_VF_SEGMENTS)
+#define TASK_SEGMENT_VF (NUM_TASK_PF_SEGMENTS)
+
+struct qed_tid_seg {
+	u32 count;
+	u8 type;
+	bool has_fl_mem;
+};
+
+struct qed_conn_type_cfg {
+	u32 cid_count;
+	u32 cids_per_vf;
+	struct qed_tid_seg tid_seg[TASK_SEGMENTS];
+};
+
+/* ILT Client configuration,
+ * Per connection type (protocol) resources (cids, tis, vf cids etc.)
+ * 1 - for connection context (CDUC) and for each task context we need two
+ * values, for regular task context and for force load memory
+ */
+#define ILT_CLI_PF_BLOCKS       (1 + NUM_TASK_PF_SEGMENTS * 2)
+#define ILT_CLI_VF_BLOCKS       (1 + NUM_TASK_VF_SEGMENTS * 2)
+#define CDUC_BLK                (0)
+#define SRQ_BLK                 (0)
+#define CDUT_SEG_BLK(n)         (1 + (u8)(n))
+#define CDUT_FL_SEG_BLK(n, X)   (1 + (n) + NUM_TASK_ ## X ## _SEGMENTS)
+
+struct ilt_cfg_pair {
+	u32 reg;
+	u32 val;
+};
+
+struct qed_ilt_cli_blk {
+	u32 total_size;		/* 0 means not active */
+	u32 real_size_in_page;
+	u32 start_line;
+	u32 dynamic_line_offset;
+	u32 dynamic_line_cnt;
+};
+
+struct qed_ilt_client_cfg {
+	bool active;
+
+	/* ILT boundaries */
+	struct ilt_cfg_pair first;
+	struct ilt_cfg_pair last;
+	struct ilt_cfg_pair p_size;
+
+	/* ILT client blocks for PF */
+	struct qed_ilt_cli_blk pf_blks[ILT_CLI_PF_BLOCKS];
+	u32 pf_total_lines;
+
+	/* ILT client blocks for VFs */
+	struct qed_ilt_cli_blk vf_blks[ILT_CLI_VF_BLOCKS];
+	u32 vf_total_lines;
+};
+
+struct qed_cid_acquired_map {
+	u32		start_cid;
+	u32		max_count;
+	unsigned long	*cid_map;
+};
+
+struct qed_src_t2 {
+	struct phys_mem_desc *dma_mem;
+	u32 num_pages;
+	u64 first_free;
+	u64 last_free;
+};
+
+struct qed_cxt_mngr {
+	/* Per protocl configuration */
+	struct qed_conn_type_cfg	conn_cfg[MAX_CONN_TYPES];
+
+	/* computed ILT structure */
+	struct qed_ilt_client_cfg	clients[MAX_ILT_CLIENTS];
+
+	/* Task type sizes */
+	u32 task_type_size[NUM_TASK_TYPES];
+
+	/* total number of VFs for this hwfn -
+	 * ALL VFs are symmetric in terms of HW resources
+	 */
+	u32 vf_count;
+	u32 first_vf_in_pf;
+
+	/* Acquired CIDs */
+	struct qed_cid_acquired_map	acquired[MAX_CONN_TYPES];
+
+	struct qed_cid_acquired_map
+	acquired_vf[MAX_CONN_TYPES][MAX_NUM_VFS];
+
+	/* ILT  shadow table */
+	struct phys_mem_desc *ilt_shadow;
+	u32 ilt_shadow_size;
+	u32 pf_start_line;
+
+	/* Mutex for a dynamic ILT allocation */
+	struct mutex mutex;
+
+	/* SRC T2 */
+	struct qed_src_t2 src_t2;
+	u32 t2_num_pages;
+	u64 first_free;
+	u64 last_free;
+
+	/* total number of SRQ's for this hwfn */
+	u32 srq_count;
+
+	/* Maximal number of L2 steering filters */
+	u32 arfs_count;
+
+	u8 task_type_id;
+	u16 task_ctx_size;
+	u16 conn_ctx_size;
+};
+
+u16 qed_get_cdut_num_pf_init_pages(struct qed_hwfn *p_hwfn);
+u16 qed_get_cdut_num_vf_init_pages(struct qed_hwfn *p_hwfn);
+u16 qed_get_cdut_num_pf_work_pages(struct qed_hwfn *p_hwfn);
+u16 qed_get_cdut_num_vf_work_pages(struct qed_hwfn *p_hwfn);
+
 #endif
diff --git a/drivers/net/ethernet/qlogic/qed/qed_debug.c b/drivers/net/ethernet/qlogic/qed/qed_debug.c
index 859caa6c1a1f..17a9b30e6c62 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_debug.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_debug.c
@@ -7,6 +7,7 @@
 #include <linux/vmalloc.h>
 #include <linux/crc32.h>
 #include "qed.h"
+#include "qed_cxt.h"
 #include "qed_hsi.h"
 #include "qed_hw.h"
 #include "qed_mcp.h"
@@ -183,6 +184,7 @@ enum platform_ids {
 /* Chip constant definitions */
 struct chip_defs {
 	const char *name;
+	u32 num_ilt_pages;
 };
 
 /* Platform constant definitions */
@@ -380,6 +382,9 @@ struct split_type_defs {
 #define IDLE_CHK_RESULT_REG_HDR_DWORDS \
 	BYTES_TO_DWORDS(sizeof(struct dbg_idle_chk_result_reg_hdr))
 
+#define PAGE_MEM_DESC_SIZE_DWORDS \
+	BYTES_TO_DWORDS(sizeof(struct phys_mem_desc))
+
 #define IDLE_CHK_MAX_ENTRIES_SIZE	32
 
 /* The sizes and offsets below are specified in bits */
@@ -464,9 +469,8 @@ static struct dbg_array s_dbg_arrays[MAX_BIN_DBG_BUFFER_TYPE] = { {NULL} };
 
 /* Chip constant definitions array */
 static struct chip_defs s_chip_defs[MAX_CHIP_IDS] = {
-	{"bb"},
-	{"ah"},
-	{"reserved"},
+	{"bb", PSWRQ2_REG_ILT_MEMORY_SIZE_BB / 2},
+	{"ah", PSWRQ2_REG_ILT_MEMORY_SIZE_K2 / 2}
 };
 
 /* Storm constant definitions array */
@@ -1627,7 +1631,22 @@ static struct grc_param_defs s_grc_param_defs[] = {
 	{{0, 0, 0}, 0, 1, false, false, 0, 0},
 
 	/* DBG_GRC_PARAM_NO_FW_VER */
-	{{0, 0, 0}, 0, 1, false, false, 0, 0}
+	{{0, 0, 0}, 0, 1, false, false, 0, 0},
+
+	/* DBG_GRC_PARAM_RESERVED3 */
+	{{0, 0, 0}, 0, 1, false, false, 0, 0},
+
+	/* DBG_GRC_PARAM_DUMP_MCP_HW_DUMP */
+	{{0, 1, 1}, 0, 1, false, false, 0, 0},
+
+	/* DBG_GRC_PARAM_DUMP_ILT_CDUC */
+	{{1, 1, 1}, 0, 1, false, false, 0, 0},
+
+	/* DBG_GRC_PARAM_DUMP_ILT_CDUT */
+	{{1, 1, 1}, 0, 1, false, false, 0, 0},
+
+	/* DBG_GRC_PARAM_DUMP_CAU_EXT */
+	{{0, 0, 0}, 0, 1, false, false, 0, 1}
 };
 
 static struct rss_mem_defs s_rss_mem_defs[] = {
@@ -4992,16 +5011,18 @@ static enum dbg_status qed_protection_override_dump(struct qed_hwfn *p_hwfn,
 	override_window_dwords =
 		qed_rd(p_hwfn, p_ptt, GRC_REG_NUMBER_VALID_OVERRIDE_WINDOW) *
 		PROTECTION_OVERRIDE_ELEMENT_DWORDS;
-	addr = BYTES_TO_DWORDS(GRC_REG_PROTECTION_OVERRIDE_WINDOW);
-	offset += qed_grc_dump_addr_range(p_hwfn,
-					  p_ptt,
-					  dump_buf + offset,
-					  true,
-					  addr,
-					  override_window_dwords,
-					  true, SPLIT_TYPE_NONE, 0);
-	qed_dump_num_param(dump_buf + size_param_offset, dump, "size",
-			   override_window_dwords);
+	if (override_window_dwords) {
+		addr = BYTES_TO_DWORDS(GRC_REG_PROTECTION_OVERRIDE_WINDOW);
+		offset += qed_grc_dump_addr_range(p_hwfn,
+						  p_ptt,
+						  dump_buf + offset,
+						  true,
+						  addr,
+						  override_window_dwords,
+						  true, SPLIT_TYPE_NONE, 0);
+		qed_dump_num_param(dump_buf + size_param_offset, dump, "size",
+				   override_window_dwords);
+	}
 out:
 	/* Dump last section */
 	offset += qed_dump_last_section(dump_buf, offset, dump);
@@ -5088,6 +5109,348 @@ static u32 qed_fw_asserts_dump(struct qed_hwfn *p_hwfn,
 	return offset;
 }
 
+/* Dumps the specified ILT pages to the specified buffer.
+ * Returns the dumped size in dwords.
+ */
+static u32 qed_ilt_dump_pages_range(u32 *dump_buf,
+				    bool dump,
+				    u32 start_page_id,
+				    u32 num_pages,
+				    struct phys_mem_desc *ilt_pages,
+				    bool dump_page_ids)
+{
+	u32 page_id, end_page_id, offset = 0;
+
+	if (num_pages == 0)
+		return offset;
+
+	end_page_id = start_page_id + num_pages - 1;
+
+	for (page_id = start_page_id; page_id <= end_page_id; page_id++) {
+		struct phys_mem_desc *mem_desc = &ilt_pages[page_id];
+
+		/**
+		 *
+		 * if (page_id >= ->p_cxt_mngr->ilt_shadow_size)
+		 *     break;
+		 */
+
+		if (!ilt_pages[page_id].virt_addr)
+			continue;
+
+		if (dump_page_ids) {
+			/* Copy page ID to dump buffer */
+			if (dump)
+				*(dump_buf + offset) = page_id;
+			offset++;
+		} else {
+			/* Copy page memory to dump buffer */
+			if (dump)
+				memcpy(dump_buf + offset,
+				       mem_desc->virt_addr, mem_desc->size);
+			offset += BYTES_TO_DWORDS(mem_desc->size);
+		}
+	}
+
+	return offset;
+}
+
+/* Dumps a section containing the dumped ILT pages.
+ * Returns the dumped size in dwords.
+ */
+static u32 qed_ilt_dump_pages_section(struct qed_hwfn *p_hwfn,
+				      u32 *dump_buf,
+				      bool dump,
+				      u32 valid_conn_pf_pages,
+				      u32 valid_conn_vf_pages,
+				      struct phys_mem_desc *ilt_pages,
+				      bool dump_page_ids)
+{
+	struct qed_ilt_client_cfg *clients = p_hwfn->p_cxt_mngr->clients;
+	u32 pf_start_line, start_page_id, offset = 0;
+	u32 cdut_pf_init_pages, cdut_vf_init_pages;
+	u32 cdut_pf_work_pages, cdut_vf_work_pages;
+	u32 base_data_offset, size_param_offset;
+	u32 cdut_pf_pages, cdut_vf_pages;
+	const char *section_name;
+	u8 i;
+
+	section_name = dump_page_ids ? "ilt_page_ids" : "ilt_page_mem";
+	cdut_pf_init_pages = qed_get_cdut_num_pf_init_pages(p_hwfn);
+	cdut_vf_init_pages = qed_get_cdut_num_vf_init_pages(p_hwfn);
+	cdut_pf_work_pages = qed_get_cdut_num_pf_work_pages(p_hwfn);
+	cdut_vf_work_pages = qed_get_cdut_num_vf_work_pages(p_hwfn);
+	cdut_pf_pages = cdut_pf_init_pages + cdut_pf_work_pages;
+	cdut_vf_pages = cdut_vf_init_pages + cdut_vf_work_pages;
+	pf_start_line = p_hwfn->p_cxt_mngr->pf_start_line;
+
+	offset +=
+	    qed_dump_section_hdr(dump_buf + offset, dump, section_name, 1);
+
+	/* Dump size parameter (0 for now, overwritten with real size later) */
+	size_param_offset = offset;
+	offset += qed_dump_num_param(dump_buf + offset, dump, "size", 0);
+	base_data_offset = offset;
+
+	/* CDUC pages are ordered as follows:
+	 * - PF pages - valid section (included in PF connection type mapping)
+	 * - PF pages - invalid section (not dumped)
+	 * - For each VF in the PF:
+	 *   - VF pages - valid section (included in VF connection type mapping)
+	 *   - VF pages - invalid section (not dumped)
+	 */
+	if (qed_grc_get_param(p_hwfn, DBG_GRC_PARAM_DUMP_ILT_CDUC)) {
+		/* Dump connection PF pages */
+		start_page_id = clients[ILT_CLI_CDUC].first.val - pf_start_line;
+		offset += qed_ilt_dump_pages_range(dump_buf + offset,
+						   dump,
+						   start_page_id,
+						   valid_conn_pf_pages,
+						   ilt_pages, dump_page_ids);
+
+		/* Dump connection VF pages */
+		start_page_id += clients[ILT_CLI_CDUC].pf_total_lines;
+		for (i = 0; i < p_hwfn->p_cxt_mngr->vf_count;
+		     i++, start_page_id += clients[ILT_CLI_CDUC].vf_total_lines)
+			offset += qed_ilt_dump_pages_range(dump_buf + offset,
+							   dump,
+							   start_page_id,
+							   valid_conn_vf_pages,
+							   ilt_pages,
+							   dump_page_ids);
+	}
+
+	/* CDUT pages are ordered as follows:
+	 * - PF init pages (not dumped)
+	 * - PF work pages
+	 * - For each VF in the PF:
+	 *   - VF init pages (not dumped)
+	 *   - VF work pages
+	 */
+	if (qed_grc_get_param(p_hwfn, DBG_GRC_PARAM_DUMP_ILT_CDUT)) {
+		/* Dump task PF pages */
+		start_page_id = clients[ILT_CLI_CDUT].first.val +
+		    cdut_pf_init_pages - pf_start_line;
+		offset += qed_ilt_dump_pages_range(dump_buf + offset,
+						   dump,
+						   start_page_id,
+						   cdut_pf_work_pages,
+						   ilt_pages, dump_page_ids);
+
+		/* Dump task VF pages */
+		start_page_id = clients[ILT_CLI_CDUT].first.val +
+		    cdut_pf_pages + cdut_vf_init_pages - pf_start_line;
+		for (i = 0; i < p_hwfn->p_cxt_mngr->vf_count;
+		     i++, start_page_id += cdut_vf_pages)
+			offset += qed_ilt_dump_pages_range(dump_buf + offset,
+							   dump,
+							   start_page_id,
+							   cdut_vf_work_pages,
+							   ilt_pages,
+							   dump_page_ids);
+	}
+
+	/* Overwrite size param */
+	if (dump)
+		qed_dump_num_param(dump_buf + size_param_offset,
+				   dump, "size", offset - base_data_offset);
+
+	return offset;
+}
+
+/* Performs ILT Dump to the specified buffer.
+ * Returns the dumped size in dwords.
+ */
+static u32 qed_ilt_dump(struct qed_hwfn *p_hwfn,
+			struct qed_ptt *p_ptt, u32 *dump_buf, bool dump)
+{
+	struct qed_ilt_client_cfg *clients = p_hwfn->p_cxt_mngr->clients;
+	u32 valid_conn_vf_cids, valid_conn_vf_pages, offset = 0;
+	u32 valid_conn_pf_cids, valid_conn_pf_pages, num_pages;
+	u32 num_cids_per_page, conn_ctx_size;
+	u32 cduc_page_size, cdut_page_size;
+	struct phys_mem_desc *ilt_pages;
+	u8 conn_type;
+
+	cduc_page_size = 1 <<
+	    (clients[ILT_CLI_CDUC].p_size.val + PXP_ILT_PAGE_SIZE_NUM_BITS_MIN);
+	cdut_page_size = 1 <<
+	    (clients[ILT_CLI_CDUT].p_size.val + PXP_ILT_PAGE_SIZE_NUM_BITS_MIN);
+	conn_ctx_size = p_hwfn->p_cxt_mngr->conn_ctx_size;
+	num_cids_per_page = (int)(cduc_page_size / conn_ctx_size);
+	ilt_pages = p_hwfn->p_cxt_mngr->ilt_shadow;
+
+	/* Dump global params - 22 must match number of params below */
+	offset += qed_dump_common_global_params(p_hwfn, p_ptt,
+						dump_buf + offset, dump, 22);
+	offset += qed_dump_str_param(dump_buf + offset,
+				     dump, "dump-type", "ilt-dump");
+	offset += qed_dump_num_param(dump_buf + offset,
+				     dump,
+				     "cduc-page-size", cduc_page_size);
+	offset += qed_dump_num_param(dump_buf + offset,
+				     dump,
+				     "cduc-first-page-id",
+				     clients[ILT_CLI_CDUC].first.val);
+	offset += qed_dump_num_param(dump_buf + offset,
+				     dump,
+				     "cduc-last-page-id",
+				     clients[ILT_CLI_CDUC].last.val);
+	offset += qed_dump_num_param(dump_buf + offset,
+				     dump,
+				     "cduc-num-pf-pages",
+				     clients
+				     [ILT_CLI_CDUC].pf_total_lines);
+	offset += qed_dump_num_param(dump_buf + offset,
+				     dump,
+				     "cduc-num-vf-pages",
+				     clients
+				     [ILT_CLI_CDUC].vf_total_lines);
+	offset += qed_dump_num_param(dump_buf + offset,
+				     dump,
+				     "max-conn-ctx-size",
+				     conn_ctx_size);
+	offset += qed_dump_num_param(dump_buf + offset,
+				     dump,
+				     "cdut-page-size", cdut_page_size);
+	offset += qed_dump_num_param(dump_buf + offset,
+				     dump,
+				     "cdut-first-page-id",
+				     clients[ILT_CLI_CDUT].first.val);
+	offset += qed_dump_num_param(dump_buf + offset,
+				     dump,
+				     "cdut-last-page-id",
+				     clients[ILT_CLI_CDUT].last.val);
+	offset += qed_dump_num_param(dump_buf + offset,
+				     dump,
+				     "cdut-num-pf-init-pages",
+				     qed_get_cdut_num_pf_init_pages(p_hwfn));
+	offset += qed_dump_num_param(dump_buf + offset,
+				     dump,
+				     "cdut-num-vf-init-pages",
+				     qed_get_cdut_num_vf_init_pages(p_hwfn));
+	offset += qed_dump_num_param(dump_buf + offset,
+				     dump,
+				     "cdut-num-pf-work-pages",
+				     qed_get_cdut_num_pf_work_pages(p_hwfn));
+	offset += qed_dump_num_param(dump_buf + offset,
+				     dump,
+				     "cdut-num-vf-work-pages",
+				     qed_get_cdut_num_vf_work_pages(p_hwfn));
+	offset += qed_dump_num_param(dump_buf + offset,
+				     dump,
+				     "max-task-ctx-size",
+				     p_hwfn->p_cxt_mngr->task_ctx_size);
+	offset += qed_dump_num_param(dump_buf + offset,
+				     dump,
+				     "task-type-id",
+				     p_hwfn->p_cxt_mngr->task_type_id);
+	offset += qed_dump_num_param(dump_buf + offset,
+				     dump,
+				     "first-vf-id-in-pf",
+				     p_hwfn->p_cxt_mngr->first_vf_in_pf);
+	offset += /* 18 */ qed_dump_num_param(dump_buf + offset,
+					      dump,
+					      "num-vfs-in-pf",
+					      p_hwfn->p_cxt_mngr->vf_count);
+	offset += qed_dump_num_param(dump_buf + offset,
+				     dump,
+				     "ptr-size-bytes", sizeof(void *));
+	offset += qed_dump_num_param(dump_buf + offset,
+				     dump,
+				     "pf-start-line",
+				     p_hwfn->p_cxt_mngr->pf_start_line);
+	offset += qed_dump_num_param(dump_buf + offset,
+				     dump,
+				     "page-mem-desc-size-dwords",
+				     PAGE_MEM_DESC_SIZE_DWORDS);
+	offset += qed_dump_num_param(dump_buf + offset,
+				     dump,
+				     "ilt-shadow-size",
+				     p_hwfn->p_cxt_mngr->ilt_shadow_size);
+	/* Additional/Less parameters require matching of number in call to
+	 * dump_common_global_params()
+	 */
+
+	/* Dump section containing number of PF CIDs per connection type */
+	offset += qed_dump_section_hdr(dump_buf + offset,
+				       dump, "num_pf_cids_per_conn_type", 1);
+	offset += qed_dump_num_param(dump_buf + offset,
+				     dump, "size", NUM_OF_CONNECTION_TYPES_E4);
+	for (conn_type = 0, valid_conn_pf_cids = 0;
+	     conn_type < NUM_OF_CONNECTION_TYPES_E4; conn_type++, offset++) {
+		u32 num_pf_cids =
+		    p_hwfn->p_cxt_mngr->conn_cfg[conn_type].cid_count;
+
+		if (dump)
+			*(dump_buf + offset) = num_pf_cids;
+		valid_conn_pf_cids += num_pf_cids;
+	}
+
+	/* Dump section containing number of VF CIDs per connection type */
+	offset += qed_dump_section_hdr(dump_buf + offset,
+				       dump, "num_vf_cids_per_conn_type", 1);
+	offset += qed_dump_num_param(dump_buf + offset,
+				     dump, "size", NUM_OF_CONNECTION_TYPES_E4);
+	for (conn_type = 0, valid_conn_vf_cids = 0;
+	     conn_type < NUM_OF_CONNECTION_TYPES_E4; conn_type++, offset++) {
+		u32 num_vf_cids =
+		    p_hwfn->p_cxt_mngr->conn_cfg[conn_type].cids_per_vf;
+
+		if (dump)
+			*(dump_buf + offset) = num_vf_cids;
+		valid_conn_vf_cids += num_vf_cids;
+	}
+
+	/* Dump section containing physical memory descs for each ILT page */
+	num_pages = p_hwfn->p_cxt_mngr->ilt_shadow_size;
+	offset += qed_dump_section_hdr(dump_buf + offset,
+				       dump, "ilt_page_desc", 1);
+	offset += qed_dump_num_param(dump_buf + offset,
+				     dump,
+				     "size",
+				     num_pages * PAGE_MEM_DESC_SIZE_DWORDS);
+
+	/* Copy memory descriptors to dump buffer */
+	if (dump) {
+		u32 page_id;
+
+		for (page_id = 0; page_id < num_pages;
+		     page_id++, offset += PAGE_MEM_DESC_SIZE_DWORDS)
+			memcpy(dump_buf + offset,
+			       &ilt_pages[page_id],
+			       DWORDS_TO_BYTES(PAGE_MEM_DESC_SIZE_DWORDS));
+	} else {
+		offset += num_pages * PAGE_MEM_DESC_SIZE_DWORDS;
+	}
+
+	valid_conn_pf_pages = DIV_ROUND_UP(valid_conn_pf_cids,
+					   num_cids_per_page);
+	valid_conn_vf_pages = DIV_ROUND_UP(valid_conn_vf_cids,
+					   num_cids_per_page);
+
+	/* Dump ILT pages IDs */
+	offset += qed_ilt_dump_pages_section(p_hwfn,
+					     dump_buf + offset,
+					     dump,
+					     valid_conn_pf_pages,
+					     valid_conn_vf_pages,
+					     ilt_pages, true);
+
+	/* Dump ILT pages memory */
+	offset += qed_ilt_dump_pages_section(p_hwfn,
+					     dump_buf + offset,
+					     dump,
+					     valid_conn_pf_pages,
+					     valid_conn_vf_pages,
+					     ilt_pages, false);
+
+	/* Dump last section */
+	offset += qed_dump_last_section(dump_buf, offset, dump);
+
+	return offset;
+}
+
 /***************************** Public Functions *******************************/
 
 enum dbg_status qed_dbg_set_bin_ptr(const u8 * const bin_ptr)
@@ -5554,6 +5917,50 @@ enum dbg_status qed_dbg_fw_asserts_dump(struct qed_hwfn *p_hwfn,
 	return DBG_STATUS_OK;
 }
 
+static enum dbg_status qed_dbg_ilt_get_dump_buf_size(struct qed_hwfn *p_hwfn,
+						     struct qed_ptt *p_ptt,
+						     u32 *buf_size)
+{
+	enum dbg_status status = qed_dbg_dev_init(p_hwfn, p_ptt);
+
+	*buf_size = 0;
+
+	if (status != DBG_STATUS_OK)
+		return status;
+
+	*buf_size = qed_ilt_dump(p_hwfn, p_ptt, NULL, false);
+
+	return DBG_STATUS_OK;
+}
+
+static enum dbg_status qed_dbg_ilt_dump(struct qed_hwfn *p_hwfn,
+					struct qed_ptt *p_ptt,
+					u32 *dump_buf,
+					u32 buf_size_in_dwords,
+					u32 *num_dumped_dwords)
+{
+	u32 needed_buf_size_in_dwords;
+	enum dbg_status status;
+
+	*num_dumped_dwords = 0;
+
+	status = qed_dbg_ilt_get_dump_buf_size(p_hwfn,
+					       p_ptt,
+					       &needed_buf_size_in_dwords);
+	if (status != DBG_STATUS_OK)
+		return status;
+
+	if (buf_size_in_dwords < needed_buf_size_in_dwords)
+		return DBG_STATUS_DUMP_BUF_TOO_SMALL;
+
+	*num_dumped_dwords = qed_ilt_dump(p_hwfn, p_ptt, dump_buf, true);
+
+	/* Reveret GRC params to their default */
+	qed_dbg_grc_set_params_default(p_hwfn);
+
+	return DBG_STATUS_OK;
+}
+
 enum dbg_status qed_dbg_read_attn(struct qed_hwfn *p_hwfn,
 				  struct qed_ptt *p_ptt,
 				  enum block_id block_id,
@@ -7763,7 +8170,10 @@ static struct {
 		    qed_dbg_fw_asserts_get_dump_buf_size,
 		    qed_dbg_fw_asserts_dump,
 		    qed_print_fw_asserts_results,
-		    qed_get_fw_asserts_results_buf_size},};
+		    qed_get_fw_asserts_results_buf_size}, {
+	"ilt",
+		    qed_dbg_ilt_get_dump_buf_size,
+		    qed_dbg_ilt_dump, NULL, NULL},};
 
 static void qed_dbg_print_feature(u8 *p_text_buf, u32 text_size)
 {
@@ -7846,6 +8256,8 @@ static enum dbg_status format_feature(struct qed_hwfn *p_hwfn,
 	return rc;
 }
 
+#define MAX_DBG_FEATURE_SIZE_DWORDS	0x3FFFFFFF
+
 /* Generic function for performing the dump of a debug feature. */
 static enum dbg_status qed_dbg_dump(struct qed_hwfn *p_hwfn,
 				    struct qed_ptt *p_ptt,
@@ -8021,6 +8433,16 @@ int qed_dbg_fw_asserts_size(struct qed_dev *cdev)
 	return qed_dbg_feature_size(cdev, DBG_FEATURE_FW_ASSERTS);
 }
 
+int qed_dbg_ilt(struct qed_dev *cdev, void *buffer, u32 *num_dumped_bytes)
+{
+	return qed_dbg_feature(cdev, buffer, DBG_FEATURE_ILT, num_dumped_bytes);
+}
+
+int qed_dbg_ilt_size(struct qed_dev *cdev)
+{
+	return qed_dbg_feature_size(cdev, DBG_FEATURE_ILT);
+}
+
 int qed_dbg_mcp_trace(struct qed_dev *cdev, void *buffer,
 		      u32 *num_dumped_bytes)
 {
@@ -8037,9 +8459,17 @@ int qed_dbg_mcp_trace_size(struct qed_dev *cdev)
  * feature buffer.
  */
 #define REGDUMP_HEADER_SIZE			sizeof(u32)
+#define REGDUMP_HEADER_SIZE_SHIFT		0
+#define REGDUMP_HEADER_SIZE_MASK		0xffffff
 #define REGDUMP_HEADER_FEATURE_SHIFT		24
-#define REGDUMP_HEADER_ENGINE_SHIFT		31
+#define REGDUMP_HEADER_FEATURE_MASK		0x3f
 #define REGDUMP_HEADER_OMIT_ENGINE_SHIFT	30
+#define REGDUMP_HEADER_OMIT_ENGINE_MASK		0x1
+#define REGDUMP_HEADER_ENGINE_SHIFT		31
+#define REGDUMP_HEADER_ENGINE_MASK		0x1
+#define REGDUMP_MAX_SIZE			0x1000000
+#define ILT_DUMP_MAX_SIZE			(1024 * 1024 * 15)
+
 enum debug_print_features {
 	OLD_MODE = 0,
 	IDLE_CHK = 1,
@@ -8053,6 +8483,8 @@ enum debug_print_features {
 	NVM_CFG1 = 9,
 	DEFAULT_CFG = 10,
 	NVM_META = 11,
+	MDUMP = 12,
+	ILT_DUMP = 13,
 };
 
 static u32 qed_calc_regdump_header(enum debug_print_features feature,
@@ -8166,8 +8598,23 @@ int qed_dbg_all_data(struct qed_dev *cdev, void *buffer)
 			       rc);
 		}
 
-		for (i = 0; i < MAX_DBG_GRC_PARAMS; i++)
-			dev_data->grc.param_val[i] = grc_params[i];
+		feature_size = qed_dbg_ilt_size(cdev);
+		if (!cdev->disable_ilt_dump &&
+		    feature_size < ILT_DUMP_MAX_SIZE) {
+			rc = qed_dbg_ilt(cdev, (u8 *)buffer + offset +
+					 REGDUMP_HEADER_SIZE, &feature_size);
+			if (!rc) {
+				*(u32 *)((u8 *)buffer + offset) =
+				    qed_calc_regdump_header(ILT_DUMP,
+							    cur_engine,
+							    feature_size,
+							    omit_engine);
+				offset += feature_size + REGDUMP_HEADER_SIZE;
+			} else {
+				DP_ERR(cdev, "qed_dbg_ilt failed. rc = %d\n",
+				       rc);
+			}
+		}
 
 		/* GRC dump - must be last because when mcp stuck it will
 		 * clutter idle_chk, reg_fifo, ...
@@ -8243,6 +8690,21 @@ int qed_dbg_all_data(struct qed_dev *cdev, void *buffer)
 		       QED_NVM_IMAGE_NVM_META, "QED_NVM_IMAGE_NVM_META", rc);
 	}
 
+	/* nvm mdump */
+	rc = qed_dbg_nvm_image(cdev, (u8 *)buffer + offset +
+			       REGDUMP_HEADER_SIZE, &feature_size,
+			       QED_NVM_IMAGE_MDUMP);
+	if (!rc) {
+		*(u32 *)((u8 *)buffer + offset) =
+			qed_calc_regdump_header(MDUMP, cur_engine,
+						feature_size, omit_engine);
+		offset += (feature_size + REGDUMP_HEADER_SIZE);
+	} else if (rc != -ENOENT) {
+		DP_ERR(cdev,
+		       "qed_dbg_nvm_image failed for image %d (%s), rc = %d\n",
+		       QED_NVM_IMAGE_MDUMP, "QED_NVM_IMAGE_MDUMP", rc);
+	}
+
 	return 0;
 }
 
@@ -8250,7 +8712,7 @@ int qed_dbg_all_data_size(struct qed_dev *cdev)
 {
 	struct qed_hwfn *p_hwfn =
 		&cdev->hwfns[cdev->dbg_params.engine_for_debug];
-	u32 regs_len = 0, image_len = 0;
+	u32 regs_len = 0, image_len = 0, ilt_len = 0, total_ilt_len = 0;
 	u8 cur_engine, org_engine;
 
 	org_engine = qed_get_debug_engine(cdev);
@@ -8267,6 +8729,12 @@ int qed_dbg_all_data_size(struct qed_dev *cdev)
 			    REGDUMP_HEADER_SIZE +
 			    qed_dbg_protection_override_size(cdev) +
 			    REGDUMP_HEADER_SIZE + qed_dbg_fw_asserts_size(cdev);
+
+		ilt_len = REGDUMP_HEADER_SIZE + qed_dbg_ilt_size(cdev);
+		if (ilt_len < ILT_DUMP_MAX_SIZE) {
+			total_ilt_len += ilt_len;
+			regs_len += ilt_len;
+		}
 	}
 
 	qed_set_debug_engine(cdev, org_engine);
@@ -8282,6 +8750,17 @@ int qed_dbg_all_data_size(struct qed_dev *cdev)
 	qed_dbg_nvm_image_length(p_hwfn, QED_NVM_IMAGE_NVM_META, &image_len);
 	if (image_len)
 		regs_len += REGDUMP_HEADER_SIZE + image_len;
+	qed_dbg_nvm_image_length(p_hwfn, QED_NVM_IMAGE_MDUMP, &image_len);
+	if (image_len)
+		regs_len += REGDUMP_HEADER_SIZE + image_len;
+
+	if (regs_len > REGDUMP_MAX_SIZE) {
+		DP_VERBOSE(cdev, QED_MSG_DEBUG,
+			   "Dump exceeds max size 0x%x, disable ILT dump\n",
+			   REGDUMP_MAX_SIZE);
+		cdev->disable_ilt_dump = true;
+		regs_len -= total_ilt_len;
+	}
 
 	return regs_len;
 }
@@ -8341,6 +8820,10 @@ int qed_dbg_feature_size(struct qed_dev *cdev, enum qed_dbg_features feature)
 	if (rc != DBG_STATUS_OK)
 		buf_size_dwords = 0;
 
+	/* Feature will not be dumped if it exceeds maximum size */
+	if (buf_size_dwords > MAX_DBG_FEATURE_SIZE_DWORDS)
+		buf_size_dwords = 0;
+
 	qed_ptt_release(p_hwfn, p_ptt);
 	qed_feature->buf_size = buf_size_dwords * sizeof(u32);
 	return qed_feature->buf_size;
diff --git a/drivers/net/ethernet/qlogic/qed/qed_debug.h b/drivers/net/ethernet/qlogic/qed/qed_debug.h
index e47e0e8d75b0..edf99d296bd1 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_debug.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_debug.h
@@ -14,11 +14,13 @@ enum qed_dbg_features {
 	DBG_FEATURE_IGU_FIFO,
 	DBG_FEATURE_PROTECTION_OVERRIDE,
 	DBG_FEATURE_FW_ASSERTS,
+	DBG_FEATURE_ILT,
 	DBG_FEATURE_NUM
 };
 
 /* Forward Declaration */
 struct qed_dev;
+struct qed_hwfn;
 
 int qed_dbg_grc(struct qed_dev *cdev, void *buffer, u32 *num_dumped_bytes);
 int qed_dbg_grc_size(struct qed_dev *cdev);
@@ -37,6 +39,8 @@ int qed_dbg_protection_override_size(struct qed_dev *cdev);
 int qed_dbg_fw_asserts(struct qed_dev *cdev, void *buffer,
 		       u32 *num_dumped_bytes);
 int qed_dbg_fw_asserts_size(struct qed_dev *cdev);
+int qed_dbg_ilt(struct qed_dev *cdev, void *buffer, u32 *num_dumped_bytes);
+int qed_dbg_ilt_size(struct qed_dev *cdev);
 int qed_dbg_mcp_trace(struct qed_dev *cdev, void *buffer,
 		      u32 *num_dumped_bytes);
 int qed_dbg_mcp_trace_size(struct qed_dev *cdev);
diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
index 67c048415d3b..e7e3f7b31ee0 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
@@ -2575,6 +2575,11 @@ enum dbg_grc_params {
 	DBG_GRC_PARAM_DUMP_PHY,
 	DBG_GRC_PARAM_NO_MCP,
 	DBG_GRC_PARAM_NO_FW_VER,
+	DBG_GRC_PARAM_RESERVED3,
+	DBG_GRC_PARAM_DUMP_MCP_HW_DUMP,
+	DBG_GRC_PARAM_DUMP_ILT_CDUC,
+	DBG_GRC_PARAM_DUMP_ILT_CDUT,
+	DBG_GRC_PARAM_DUMP_CAU_EXT,
 	MAX_DBG_GRC_PARAMS
 };
 
@@ -2695,6 +2700,19 @@ struct dbg_tools_data {
 	u32 num_regs_read;
 };
 
+/* ILT Clients */
+enum ilt_clients {
+	ILT_CLI_CDUC,
+	ILT_CLI_CDUT,
+	ILT_CLI_QM,
+	ILT_CLI_TM,
+	ILT_CLI_SRC,
+	ILT_CLI_TSDM,
+	ILT_CLI_RGFS,
+	ILT_CLI_TGFS,
+	MAX_ILT_CLIENTS
+};
+
 /********************************/
 /* HSI Init Functions constants */
 /********************************/
diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.c b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
index af981ff5595e..280527cc0578 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_mcp.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
@@ -3167,6 +3167,9 @@ qed_mcp_get_nvm_image_att(struct qed_hwfn *p_hwfn,
 	case QED_NVM_IMAGE_FCOE_CFG:
 		type = NVM_TYPE_FCOE_CFG;
 		break;
+	case QED_NVM_IMAGE_MDUMP:
+		type = NVM_TYPE_MDUMP;
+		break;
 	case QED_NVM_IMAGE_NVM_CFG1:
 		type = NVM_TYPE_NVM_CFG1;
 		break;
diff --git a/include/linux/qed/qed_if.h b/include/linux/qed/qed_if.h
index 9bcb2f419004..1b27c22d39af 100644
--- a/include/linux/qed/qed_if.h
+++ b/include/linux/qed/qed_if.h
@@ -159,6 +159,7 @@ struct qed_dcbx_get {
 enum qed_nvm_images {
 	QED_NVM_IMAGE_ISCSI_CFG,
 	QED_NVM_IMAGE_FCOE_CFG,
+	QED_NVM_IMAGE_MDUMP,
 	QED_NVM_IMAGE_NVM_CFG1,
 	QED_NVM_IMAGE_DEFAULT_CFG,
 	QED_NVM_IMAGE_NVM_META,
-- 
cgit v1.2.3-59-g8ed1b


From 6bc82d9b7e6371673992ed5e3897cf7fb8cc4f41 Mon Sep 17 00:00:00 2001
From: Michal Kalderon <michal.kalderon@marvell.com>
Date: Mon, 27 Jan 2020 15:26:18 +0200
Subject: qed: rt init valid initialization changed

The QM phase init tool can be invoked multiple times during
the driver lifetime. Part of the init comes from the runtime array.
The logic for setting the values did not init all values, basically
assuming the runtime array was all zeroes. But if it was invoked
multiple times, nobody was zeroing it after the first time.

In this change we zero the runtime array right after using it.

Signed-off-by: Ariel Elior <ariel.elior@marvell.com>
Signed-off-by: Michal Kalderon <michal.kalderon@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/qlogic/qed/qed_dev.c      |  3 ---
 drivers/net/ethernet/qlogic/qed/qed_init_ops.c | 16 ++++++----------
 drivers/net/ethernet/qlogic/qed/qed_init_ops.h |  8 --------
 3 files changed, 6 insertions(+), 21 deletions(-)

diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c
index df97810b09b9..3fb73ce8c1d6 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_dev.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c
@@ -2106,9 +2106,6 @@ int qed_qm_reconf(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt)
 	if (!b_rc)
 		return -EINVAL;
 
-	/* clear the QM_PF runtime phase leftovers from previous init */
-	qed_init_clear_rt_data(p_hwfn);
-
 	/* prepare QM portion of runtime array */
 	qed_qm_init_pf(p_hwfn, p_ptt, false);
 
diff --git a/drivers/net/ethernet/qlogic/qed/qed_init_ops.c b/drivers/net/ethernet/qlogic/qed/qed_init_ops.c
index 36f998c89e74..5a6e4ac4fef4 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_init_ops.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_init_ops.c
@@ -74,15 +74,6 @@ void qed_init_iro_array(struct qed_dev *cdev)
 	cdev->iro_arr = iro_arr;
 }
 
-/* Runtime configuration helpers */
-void qed_init_clear_rt_data(struct qed_hwfn *p_hwfn)
-{
-	int i;
-
-	for (i = 0; i < RUNTIME_ARRAY_SIZE; i++)
-		p_hwfn->rt_data.b_valid[i] = false;
-}
-
 void qed_init_store_rt_reg(struct qed_hwfn *p_hwfn, u32 rt_offset, u32 val)
 {
 	p_hwfn->rt_data.init_val[rt_offset] = val;
@@ -106,7 +97,7 @@ static int qed_init_rt(struct qed_hwfn	*p_hwfn,
 {
 	u32 *p_init_val = &p_hwfn->rt_data.init_val[rt_offset];
 	bool *p_valid = &p_hwfn->rt_data.b_valid[rt_offset];
-	u16 i, segment;
+	u16 i, j, segment;
 	int rc = 0;
 
 	/* Since not all RT entries are initialized, go over the RT and
@@ -121,6 +112,7 @@ static int qed_init_rt(struct qed_hwfn	*p_hwfn,
 		 */
 		if (!b_must_dmae) {
 			qed_wr(p_hwfn, p_ptt, addr + (i << 2), p_init_val[i]);
+			p_valid[i] = false;
 			continue;
 		}
 
@@ -135,6 +127,10 @@ static int qed_init_rt(struct qed_hwfn	*p_hwfn,
 		if (rc)
 			return rc;
 
+		/* invalidate after writing */
+		for (j = i; j < i + segment; j++)
+			p_valid[j] = false;
+
 		/* Jump over the entire segment, including invalid entry */
 		i += segment;
 	}
diff --git a/drivers/net/ethernet/qlogic/qed/qed_init_ops.h b/drivers/net/ethernet/qlogic/qed/qed_init_ops.h
index 555dd086796d..e9e8ade50ed3 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_init_ops.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_init_ops.h
@@ -80,14 +80,6 @@ int qed_init_alloc(struct qed_hwfn *p_hwfn);
  */
 void qed_init_free(struct qed_hwfn *p_hwfn);
 
-/**
- * @brief qed_init_clear_rt_data - Clears the runtime init array.
- *
- *
- * @param p_hwfn
- */
-void qed_init_clear_rt_data(struct qed_hwfn *p_hwfn);
-
 /**
  * @brief qed_init_store_rt_reg - Store a configuration value in the RT array.
  *
-- 
cgit v1.2.3-59-g8ed1b


From 2d22bc8354b15abe413dff76cfe0f7aeb88ef9aa Mon Sep 17 00:00:00 2001
From: Michal Kalderon <michal.kalderon@marvell.com>
Date: Mon, 27 Jan 2020 15:26:19 +0200
Subject: qed: FW 8.42.2.0 debug features

Add to debug dump more information on the platform it was collected
from (pci func, path id).
Provide human readable reg fifo erros.

Removed static debug arrays from HSI Functions, and move them to
the hwfn.

Some structures were slightly changed (removing reserved chip id
for example) which lead to many long initializations being modified
with one parameter less during initialization. This leads to
some long diffs that don't really change anything.

Signed-off-by: Ariel Elior <ariel.elior@marvell.com>
Signed-off-by: Michal Kalderon <michal.kalderon@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/qlogic/qed/qed.h       |    2 +
 drivers/net/ethernet/qlogic/qed/qed_debug.c | 3392 +++++++++++----------------
 drivers/net/ethernet/qlogic/qed/qed_dev.c   |    2 +-
 drivers/net/ethernet/qlogic/qed/qed_hsi.h   |  661 ++----
 drivers/net/ethernet/qlogic/qed/qed_main.c  |    2 +-
 include/linux/qed/qed_if.h                  |    9 +
 6 files changed, 1530 insertions(+), 2538 deletions(-)

diff --git a/drivers/net/ethernet/qlogic/qed/qed.h b/drivers/net/ethernet/qlogic/qed/qed.h
index a6dc40681a32..fa41bf08a589 100644
--- a/drivers/net/ethernet/qlogic/qed/qed.h
+++ b/drivers/net/ethernet/qlogic/qed/qed.h
@@ -666,6 +666,7 @@ struct qed_hwfn {
 
 	struct dbg_tools_data		dbg_info;
 	void				*dbg_user_info;
+	struct virt_mem_desc		dbg_arrays[MAX_BIN_DBG_BUFFER_TYPE];
 
 	/* PWM region specific data */
 	u16				wid_count;
@@ -877,6 +878,7 @@ struct qed_dev {
 	struct qed_cb_ll2_info		*ll2;
 	u8				ll2_mac_address[ETH_ALEN];
 #endif
+	struct qed_dbg_feature dbg_features[DBG_FEATURE_NUM];
 	bool disable_ilt_dump;
 	DECLARE_HASHTABLE(connections, 10);
 	const struct firmware		*firmware;
diff --git a/drivers/net/ethernet/qlogic/qed/qed_debug.c b/drivers/net/ethernet/qlogic/qed/qed_debug.c
index 17a9b30e6c62..f4eebaabb6d0 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_debug.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_debug.c
@@ -23,27 +23,28 @@ enum mem_groups {
 	MEM_GROUP_BRB_RAM,
 	MEM_GROUP_BRB_MEM,
 	MEM_GROUP_PRS_MEM,
+	MEM_GROUP_SDM_MEM,
+	MEM_GROUP_PBUF,
 	MEM_GROUP_IOR,
+	MEM_GROUP_RAM,
 	MEM_GROUP_BTB_RAM,
+	MEM_GROUP_RDIF_CTX,
+	MEM_GROUP_TDIF_CTX,
+	MEM_GROUP_CFC_MEM,
 	MEM_GROUP_CONN_CFC_MEM,
-	MEM_GROUP_TASK_CFC_MEM,
 	MEM_GROUP_CAU_PI,
 	MEM_GROUP_CAU_MEM,
+	MEM_GROUP_CAU_MEM_EXT,
 	MEM_GROUP_PXP_ILT,
-	MEM_GROUP_TM_MEM,
-	MEM_GROUP_SDM_MEM,
-	MEM_GROUP_PBUF,
-	MEM_GROUP_RAM,
 	MEM_GROUP_MULD_MEM,
 	MEM_GROUP_BTB_MEM,
-	MEM_GROUP_RDIF_CTX,
-	MEM_GROUP_TDIF_CTX,
-	MEM_GROUP_CFC_MEM,
 	MEM_GROUP_IGU_MEM,
 	MEM_GROUP_IGU_MSIX,
 	MEM_GROUP_CAU_SB,
 	MEM_GROUP_BMB_RAM,
 	MEM_GROUP_BMB_MEM,
+	MEM_GROUP_TM_MEM,
+	MEM_GROUP_TASK_CFC_MEM,
 	MEM_GROUPS_NUM
 };
 
@@ -57,27 +58,28 @@ static const char * const s_mem_group_names[] = {
 	"BRB_RAM",
 	"BRB_MEM",
 	"PRS_MEM",
+	"SDM_MEM",
+	"PBUF",
 	"IOR",
+	"RAM",
 	"BTB_RAM",
+	"RDIF_CTX",
+	"TDIF_CTX",
+	"CFC_MEM",
 	"CONN_CFC_MEM",
-	"TASK_CFC_MEM",
 	"CAU_PI",
 	"CAU_MEM",
+	"CAU_MEM_EXT",
 	"PXP_ILT",
-	"TM_MEM",
-	"SDM_MEM",
-	"PBUF",
-	"RAM",
 	"MULD_MEM",
 	"BTB_MEM",
-	"RDIF_CTX",
-	"TDIF_CTX",
-	"CFC_MEM",
 	"IGU_MEM",
 	"IGU_MSIX",
 	"CAU_SB",
 	"BMB_RAM",
 	"BMB_MEM",
+	"TM_MEM",
+	"TASK_CFC_MEM",
 };
 
 /* Idle check conditions */
@@ -171,14 +173,38 @@ static u32(*cond_arr[]) (const u32 *r, const u32 *imm) = {
 	cond13,
 };
 
+#define NUM_PHYS_BLOCKS 84
+
+#define NUM_DBG_RESET_REGS 8
+
 /******************************* Data Types **********************************/
 
-enum platform_ids {
-	PLATFORM_ASIC,
+enum hw_types {
+	HW_TYPE_ASIC,
 	PLATFORM_RESERVED,
 	PLATFORM_RESERVED2,
 	PLATFORM_RESERVED3,
-	MAX_PLATFORM_IDS
+	PLATFORM_RESERVED4,
+	MAX_HW_TYPES
+};
+
+/* CM context types */
+enum cm_ctx_types {
+	CM_CTX_CONN_AG,
+	CM_CTX_CONN_ST,
+	CM_CTX_TASK_AG,
+	CM_CTX_TASK_ST,
+	NUM_CM_CTX_TYPES
+};
+
+/* Debug bus frame modes */
+enum dbg_bus_frame_modes {
+	DBG_BUS_FRAME_MODE_4ST = 0,	/* 4 Storm dwords (no HW) */
+	DBG_BUS_FRAME_MODE_2ST_2HW = 1,	/* 2 Storm dwords, 2 HW dwords */
+	DBG_BUS_FRAME_MODE_1ST_3HW = 2,	/* 1 Storm dwords, 3 HW dwords */
+	DBG_BUS_FRAME_MODE_4HW = 3,	/* 4 HW dwords (no Storms) */
+	DBG_BUS_FRAME_MODE_8HW = 4,	/* 8 HW dwords (no Storms) */
+	DBG_BUS_NUM_FRAME_MODES
 };
 
 /* Chip constant definitions */
@@ -187,20 +213,26 @@ struct chip_defs {
 	u32 num_ilt_pages;
 };
 
-/* Platform constant definitions */
-struct platform_defs {
+/* HW type constant definitions */
+struct hw_type_defs {
 	const char *name;
 	u32 delay_factor;
 	u32 dmae_thresh;
 	u32 log_thresh;
 };
 
+/* RBC reset definitions */
+struct rbc_reset_defs {
+	u32 reset_reg_addr;
+	u32 reset_val[MAX_CHIP_IDS];
+};
+
 /* Storm constant definitions.
  * Addresses are in bytes, sizes are in quad-regs.
  */
 struct storm_defs {
 	char letter;
-	enum block_id block_id;
+	enum block_id sem_block_id;
 	enum dbg_bus_clients dbg_client_id[MAX_CHIP_IDS];
 	bool has_vfc;
 	u32 sem_fast_mem_addr;
@@ -209,47 +241,26 @@ struct storm_defs {
 	u32 sem_slow_mode_addr;
 	u32 sem_slow_mode1_conf_addr;
 	u32 sem_sync_dbg_empty_addr;
-	u32 sem_slow_dbg_empty_addr;
+	u32 sem_gpre_vect_addr;
 	u32 cm_ctx_wr_addr;
-	u32 cm_conn_ag_ctx_lid_size;
-	u32 cm_conn_ag_ctx_rd_addr;
-	u32 cm_conn_st_ctx_lid_size;
-	u32 cm_conn_st_ctx_rd_addr;
-	u32 cm_task_ag_ctx_lid_size;
-	u32 cm_task_ag_ctx_rd_addr;
-	u32 cm_task_st_ctx_lid_size;
-	u32 cm_task_st_ctx_rd_addr;
+	u32 cm_ctx_rd_addr[NUM_CM_CTX_TYPES];
+	u32 cm_ctx_lid_sizes[MAX_CHIP_IDS][NUM_CM_CTX_TYPES];
 };
 
-/* Block constant definitions */
-struct block_defs {
-	const char *name;
-	bool exists[MAX_CHIP_IDS];
-	bool associated_to_storm;
-
-	/* Valid only if associated_to_storm is true */
-	u32 storm_id;
-	enum dbg_bus_clients dbg_client_id[MAX_CHIP_IDS];
-	u32 dbg_select_addr;
-	u32 dbg_enable_addr;
-	u32 dbg_shift_addr;
-	u32 dbg_force_valid_addr;
-	u32 dbg_force_frame_addr;
-	bool has_reset_bit;
-
-	/* If true, block is taken out of reset before dump */
-	bool unreset;
-	enum dbg_reset_regs reset_reg;
-
-	/* Bit offset in reset register */
-	u8 reset_bit_offset;
+/* Debug Bus Constraint operation constant definitions */
+struct dbg_bus_constraint_op_defs {
+	u8 hw_op_val;
+	bool is_cyclic;
 };
 
-/* Reset register definitions */
-struct reset_reg_defs {
-	u32 addr;
+/* Storm Mode definitions */
+struct storm_mode_defs {
+	const char *name;
+	bool is_fast_dbg;
+	u8 id_in_hw;
+	u32 src_disable_reg_addr;
+	u32 src_enable_val;
 	bool exists[MAX_CHIP_IDS];
-	u32 unreset_val[MAX_CHIP_IDS];
 };
 
 struct grc_param_defs {
@@ -259,7 +270,7 @@ struct grc_param_defs {
 	bool is_preset;
 	bool is_persistent;
 	u32 exclude_all_preset_val;
-	u32 crash_preset_val;
+	u32 crash_preset_val[MAX_CHIP_IDS];
 };
 
 /* Address is in 128b units. Width is in bits. */
@@ -316,15 +327,7 @@ struct split_type_defs {
 
 /******************************** Constants **********************************/
 
-#define MAX_LCIDS			320
-#define MAX_LTIDS			320
-
-#define NUM_IOR_SETS			2
-#define IORS_PER_SET			176
-#define IOR_SET_OFFSET(set_id)		((set_id) * 256)
-
 #define BYTES_IN_DWORD			sizeof(u32)
-
 /* In the macros below, size and offset are specified in bits */
 #define CEIL_DWORDS(size)		DIV_ROUND_UP(size, 32)
 #define FIELD_BIT_OFFSET(type, field)	type ## _ ## field ## _ ## OFFSET
@@ -350,20 +353,17 @@ struct split_type_defs {
 			qed_wr(dev, ptt, addr,	(arr)[i]); \
 	} while (0)
 
-#define ARR_REG_RD(dev, ptt, addr, arr, arr_size) \
-	do { \
-		for (i = 0; i < (arr_size); i++) \
-			(arr)[i] = qed_rd(dev, ptt, addr); \
-	} while (0)
-
 #define DWORDS_TO_BYTES(dwords)		((dwords) * BYTES_IN_DWORD)
 #define BYTES_TO_DWORDS(bytes)		((bytes) / BYTES_IN_DWORD)
 
-/* Extra lines include a signature line + optional latency events line */
-#define NUM_EXTRA_DBG_LINES(block_desc) \
-	(1 + ((block_desc)->has_latency_events ? 1 : 0))
-#define NUM_DBG_LINES(block_desc) \
-	((block_desc)->num_of_lines + NUM_EXTRA_DBG_LINES(block_desc))
+/* extra lines include a signature line + optional latency events line */
+#define NUM_EXTRA_DBG_LINES(block) \
+	(GET_FIELD((block)->flags, DBG_BLOCK_CHIP_HAS_LATENCY_EVENTS) ? 2 : 1)
+#define NUM_DBG_LINES(block) \
+	((block)->num_of_dbg_bus_lines + NUM_EXTRA_DBG_LINES(block))
+
+#define USE_DMAE			true
+#define PROTECT_WIDE_BUS		true
 
 #define RAM_LINES_TO_DWORDS(lines)	((lines) * 2)
 #define RAM_LINES_TO_BYTES(lines) \
@@ -430,7 +430,9 @@ struct split_type_defs {
 
 #define STATIC_DEBUG_LINE_DWORDS	9
 
-#define NUM_COMMON_GLOBAL_PARAMS	8
+#define NUM_COMMON_GLOBAL_PARAMS	9
+
+#define MAX_RECURSION_DEPTH		10
 
 #define FW_IMG_MAIN			1
 
@@ -454,19 +456,13 @@ struct split_type_defs {
 	(MCP_REG_SCRATCH + \
 	 offsetof(struct static_init, sections[SPAD_SECTION_TRACE]))
 
+#define MAX_SW_PLTAFORM_STR_SIZE	64
+
 #define EMPTY_FW_VERSION_STR		"???_???_???_???"
 #define EMPTY_FW_IMAGE_STR		"???????????????"
 
 /***************************** Constant Arrays *******************************/
 
-struct dbg_array {
-	const u32 *ptr;
-	u32 size_in_dwords;
-};
-
-/* Debug arrays */
-static struct dbg_array s_dbg_arrays[MAX_BIN_DBG_BUFFER_TYPE] = { {NULL} };
-
 /* Chip constant definitions array */
 static struct chip_defs s_chip_defs[MAX_CHIP_IDS] = {
 	{"bb", PSWRQ2_REG_ILT_MEMORY_SIZE_BB / 2},
@@ -477,1030 +473,104 @@ static struct chip_defs s_chip_defs[MAX_CHIP_IDS] = {
 static struct storm_defs s_storm_defs[] = {
 	/* Tstorm */
 	{'T', BLOCK_TSEM,
-	 {DBG_BUS_CLIENT_RBCT, DBG_BUS_CLIENT_RBCT,
-	  DBG_BUS_CLIENT_RBCT}, true,
-	 TSEM_REG_FAST_MEMORY,
-	 TSEM_REG_DBG_FRAME_MODE_BB_K2, TSEM_REG_SLOW_DBG_ACTIVE_BB_K2,
-	 TSEM_REG_SLOW_DBG_MODE_BB_K2, TSEM_REG_DBG_MODE1_CFG_BB_K2,
-	 TSEM_REG_SYNC_DBG_EMPTY, TSEM_REG_SLOW_DBG_EMPTY_BB_K2,
-	 TCM_REG_CTX_RBC_ACCS,
-	 4, TCM_REG_AGG_CON_CTX,
-	 16, TCM_REG_SM_CON_CTX,
-	 2, TCM_REG_AGG_TASK_CTX,
-	 4, TCM_REG_SM_TASK_CTX},
+		{DBG_BUS_CLIENT_RBCT, DBG_BUS_CLIENT_RBCT},
+		true,
+		TSEM_REG_FAST_MEMORY,
+		TSEM_REG_DBG_FRAME_MODE_BB_K2, TSEM_REG_SLOW_DBG_ACTIVE_BB_K2,
+		TSEM_REG_SLOW_DBG_MODE_BB_K2, TSEM_REG_DBG_MODE1_CFG_BB_K2,
+		TSEM_REG_SYNC_DBG_EMPTY, TSEM_REG_DBG_GPRE_VECT,
+		TCM_REG_CTX_RBC_ACCS,
+		{TCM_REG_AGG_CON_CTX, TCM_REG_SM_CON_CTX, TCM_REG_AGG_TASK_CTX,
+		 TCM_REG_SM_TASK_CTX},
+		{{4, 16, 2, 4}, {4, 16, 2, 4}} /* {bb} {k2} */
+	},
 
 	/* Mstorm */
 	{'M', BLOCK_MSEM,
-	 {DBG_BUS_CLIENT_RBCT, DBG_BUS_CLIENT_RBCM,
-	  DBG_BUS_CLIENT_RBCM}, false,
-	 MSEM_REG_FAST_MEMORY,
-	 MSEM_REG_DBG_FRAME_MODE_BB_K2, MSEM_REG_SLOW_DBG_ACTIVE_BB_K2,
-	 MSEM_REG_SLOW_DBG_MODE_BB_K2, MSEM_REG_DBG_MODE1_CFG_BB_K2,
-	 MSEM_REG_SYNC_DBG_EMPTY, MSEM_REG_SLOW_DBG_EMPTY_BB_K2,
-	 MCM_REG_CTX_RBC_ACCS,
-	 1, MCM_REG_AGG_CON_CTX,
-	 10, MCM_REG_SM_CON_CTX,
-	 2, MCM_REG_AGG_TASK_CTX,
-	 7, MCM_REG_SM_TASK_CTX},
+		{DBG_BUS_CLIENT_RBCT, DBG_BUS_CLIENT_RBCM},
+		false,
+		MSEM_REG_FAST_MEMORY,
+		MSEM_REG_DBG_FRAME_MODE_BB_K2,
+		MSEM_REG_SLOW_DBG_ACTIVE_BB_K2,
+		MSEM_REG_SLOW_DBG_MODE_BB_K2,
+		MSEM_REG_DBG_MODE1_CFG_BB_K2,
+		MSEM_REG_SYNC_DBG_EMPTY,
+		MSEM_REG_DBG_GPRE_VECT,
+		MCM_REG_CTX_RBC_ACCS,
+		{MCM_REG_AGG_CON_CTX, MCM_REG_SM_CON_CTX, MCM_REG_AGG_TASK_CTX,
+		 MCM_REG_SM_TASK_CTX },
+		{{1, 10, 2, 7}, {1, 10, 2, 7}} /* {bb} {k2}*/
+	},
 
 	/* Ustorm */
 	{'U', BLOCK_USEM,
-	 {DBG_BUS_CLIENT_RBCU, DBG_BUS_CLIENT_RBCU,
-	  DBG_BUS_CLIENT_RBCU}, false,
-	 USEM_REG_FAST_MEMORY,
-	 USEM_REG_DBG_FRAME_MODE_BB_K2, USEM_REG_SLOW_DBG_ACTIVE_BB_K2,
-	 USEM_REG_SLOW_DBG_MODE_BB_K2, USEM_REG_DBG_MODE1_CFG_BB_K2,
-	 USEM_REG_SYNC_DBG_EMPTY, USEM_REG_SLOW_DBG_EMPTY_BB_K2,
-	 UCM_REG_CTX_RBC_ACCS,
-	 2, UCM_REG_AGG_CON_CTX,
-	 13, UCM_REG_SM_CON_CTX,
-	 3, UCM_REG_AGG_TASK_CTX,
-	 3, UCM_REG_SM_TASK_CTX},
+		{DBG_BUS_CLIENT_RBCU, DBG_BUS_CLIENT_RBCU},
+		false,
+		USEM_REG_FAST_MEMORY,
+		USEM_REG_DBG_FRAME_MODE_BB_K2,
+		USEM_REG_SLOW_DBG_ACTIVE_BB_K2,
+		USEM_REG_SLOW_DBG_MODE_BB_K2,
+		USEM_REG_DBG_MODE1_CFG_BB_K2,
+		USEM_REG_SYNC_DBG_EMPTY,
+		USEM_REG_DBG_GPRE_VECT,
+		UCM_REG_CTX_RBC_ACCS,
+		{UCM_REG_AGG_CON_CTX, UCM_REG_SM_CON_CTX, UCM_REG_AGG_TASK_CTX,
+		 UCM_REG_SM_TASK_CTX},
+		{{2, 13, 3, 3}, {2, 13, 3, 3}} /* {bb} {k2} */
+	},
 
 	/* Xstorm */
 	{'X', BLOCK_XSEM,
-	 {DBG_BUS_CLIENT_RBCX, DBG_BUS_CLIENT_RBCX,
-	  DBG_BUS_CLIENT_RBCX}, false,
-	 XSEM_REG_FAST_MEMORY,
-	 XSEM_REG_DBG_FRAME_MODE_BB_K2, XSEM_REG_SLOW_DBG_ACTIVE_BB_K2,
-	 XSEM_REG_SLOW_DBG_MODE_BB_K2, XSEM_REG_DBG_MODE1_CFG_BB_K2,
-	 XSEM_REG_SYNC_DBG_EMPTY, XSEM_REG_SLOW_DBG_EMPTY_BB_K2,
-	 XCM_REG_CTX_RBC_ACCS,
-	 9, XCM_REG_AGG_CON_CTX,
-	 15, XCM_REG_SM_CON_CTX,
-	 0, 0,
-	 0, 0},
+		{DBG_BUS_CLIENT_RBCX, DBG_BUS_CLIENT_RBCX},
+		false,
+		XSEM_REG_FAST_MEMORY,
+		XSEM_REG_DBG_FRAME_MODE_BB_K2,
+		XSEM_REG_SLOW_DBG_ACTIVE_BB_K2,
+		XSEM_REG_SLOW_DBG_MODE_BB_K2,
+		XSEM_REG_DBG_MODE1_CFG_BB_K2,
+		XSEM_REG_SYNC_DBG_EMPTY,
+		XSEM_REG_DBG_GPRE_VECT,
+		XCM_REG_CTX_RBC_ACCS,
+		{XCM_REG_AGG_CON_CTX, XCM_REG_SM_CON_CTX, 0, 0},
+		{{9, 15, 0, 0}, {9, 15,	0, 0}} /* {bb} {k2} */
+	},
 
 	/* Ystorm */
 	{'Y', BLOCK_YSEM,
-	 {DBG_BUS_CLIENT_RBCX, DBG_BUS_CLIENT_RBCY,
-	  DBG_BUS_CLIENT_RBCY}, false,
-	 YSEM_REG_FAST_MEMORY,
-	 YSEM_REG_DBG_FRAME_MODE_BB_K2, YSEM_REG_SLOW_DBG_ACTIVE_BB_K2,
-	 YSEM_REG_SLOW_DBG_MODE_BB_K2, YSEM_REG_DBG_MODE1_CFG_BB_K2,
-	 YSEM_REG_SYNC_DBG_EMPTY, TSEM_REG_SLOW_DBG_EMPTY_BB_K2,
-	 YCM_REG_CTX_RBC_ACCS,
-	 2, YCM_REG_AGG_CON_CTX,
-	 3, YCM_REG_SM_CON_CTX,
-	 2, YCM_REG_AGG_TASK_CTX,
-	 12, YCM_REG_SM_TASK_CTX},
+		{DBG_BUS_CLIENT_RBCX, DBG_BUS_CLIENT_RBCY},
+		false,
+		YSEM_REG_FAST_MEMORY,
+		YSEM_REG_DBG_FRAME_MODE_BB_K2,
+		YSEM_REG_SLOW_DBG_ACTIVE_BB_K2,
+		YSEM_REG_SLOW_DBG_MODE_BB_K2,
+		YSEM_REG_DBG_MODE1_CFG_BB_K2,
+		YSEM_REG_SYNC_DBG_EMPTY,
+		YSEM_REG_DBG_GPRE_VECT,
+		YCM_REG_CTX_RBC_ACCS,
+		{YCM_REG_AGG_CON_CTX, YCM_REG_SM_CON_CTX, YCM_REG_AGG_TASK_CTX,
+		 YCM_REG_SM_TASK_CTX},
+		{{2, 3, 2, 12}, {2, 3, 2, 12}} /* {bb} {k2} */
+	},
 
 	/* Pstorm */
 	{'P', BLOCK_PSEM,
-	 {DBG_BUS_CLIENT_RBCS, DBG_BUS_CLIENT_RBCS,
-	  DBG_BUS_CLIENT_RBCS}, true,
-	 PSEM_REG_FAST_MEMORY,
-	 PSEM_REG_DBG_FRAME_MODE_BB_K2, PSEM_REG_SLOW_DBG_ACTIVE_BB_K2,
-	 PSEM_REG_SLOW_DBG_MODE_BB_K2, PSEM_REG_DBG_MODE1_CFG_BB_K2,
-	 PSEM_REG_SYNC_DBG_EMPTY, PSEM_REG_SLOW_DBG_EMPTY_BB_K2,
-	 PCM_REG_CTX_RBC_ACCS,
-	 0, 0,
-	 10, PCM_REG_SM_CON_CTX,
-	 0, 0,
-	 0, 0}
-};
-
-/* Block definitions array */
-
-static struct block_defs block_grc_defs = {
-	"grc",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCN, DBG_BUS_CLIENT_RBCN, DBG_BUS_CLIENT_RBCN},
-	GRC_REG_DBG_SELECT, GRC_REG_DBG_DWORD_ENABLE,
-	GRC_REG_DBG_SHIFT, GRC_REG_DBG_FORCE_VALID,
-	GRC_REG_DBG_FORCE_FRAME,
-	true, false, DBG_RESET_REG_MISC_PL_UA, 1
-};
-
-static struct block_defs block_miscs_defs = {
-	"miscs", {true, true, true}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS},
-	0, 0, 0, 0, 0,
-	false, false, MAX_DBG_RESET_REGS, 0
-};
-
-static struct block_defs block_misc_defs = {
-	"misc", {true, true, true}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS},
-	0, 0, 0, 0, 0,
-	false, false, MAX_DBG_RESET_REGS, 0
-};
-
-static struct block_defs block_dbu_defs = {
-	"dbu", {true, true, true}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS},
-	0, 0, 0, 0, 0,
-	false, false, MAX_DBG_RESET_REGS, 0
-};
-
-static struct block_defs block_pglue_b_defs = {
-	"pglue_b",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCH, DBG_BUS_CLIENT_RBCH, DBG_BUS_CLIENT_RBCH},
-	PGLUE_B_REG_DBG_SELECT, PGLUE_B_REG_DBG_DWORD_ENABLE,
-	PGLUE_B_REG_DBG_SHIFT, PGLUE_B_REG_DBG_FORCE_VALID,
-	PGLUE_B_REG_DBG_FORCE_FRAME,
-	true, false, DBG_RESET_REG_MISCS_PL_HV, 1
-};
-
-static struct block_defs block_cnig_defs = {
-	"cnig",
-	{true, true, true}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, DBG_BUS_CLIENT_RBCW,
-	 DBG_BUS_CLIENT_RBCW},
-	CNIG_REG_DBG_SELECT_K2_E5, CNIG_REG_DBG_DWORD_ENABLE_K2_E5,
-	CNIG_REG_DBG_SHIFT_K2_E5, CNIG_REG_DBG_FORCE_VALID_K2_E5,
-	CNIG_REG_DBG_FORCE_FRAME_K2_E5,
-	true, false, DBG_RESET_REG_MISCS_PL_HV, 0
-};
-
-static struct block_defs block_cpmu_defs = {
-	"cpmu", {true, true, true}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS},
-	0, 0, 0, 0, 0,
-	true, false, DBG_RESET_REG_MISCS_PL_HV, 8
-};
-
-static struct block_defs block_ncsi_defs = {
-	"ncsi",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCZ, DBG_BUS_CLIENT_RBCZ, DBG_BUS_CLIENT_RBCZ},
-	NCSI_REG_DBG_SELECT, NCSI_REG_DBG_DWORD_ENABLE,
-	NCSI_REG_DBG_SHIFT, NCSI_REG_DBG_FORCE_VALID,
-	NCSI_REG_DBG_FORCE_FRAME,
-	true, false, DBG_RESET_REG_MISCS_PL_HV, 5
-};
-
-static struct block_defs block_opte_defs = {
-	"opte", {true, true, false}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS},
-	0, 0, 0, 0, 0,
-	true, false, DBG_RESET_REG_MISCS_PL_HV, 4
-};
-
-static struct block_defs block_bmb_defs = {
-	"bmb",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCZ, DBG_BUS_CLIENT_RBCB, DBG_BUS_CLIENT_RBCB},
-	BMB_REG_DBG_SELECT, BMB_REG_DBG_DWORD_ENABLE,
-	BMB_REG_DBG_SHIFT, BMB_REG_DBG_FORCE_VALID,
-	BMB_REG_DBG_FORCE_FRAME,
-	true, false, DBG_RESET_REG_MISCS_PL_UA, 7
-};
-
-static struct block_defs block_pcie_defs = {
-	"pcie",
-	{true, true, true}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, DBG_BUS_CLIENT_RBCH,
-	 DBG_BUS_CLIENT_RBCH},
-	PCIE_REG_DBG_COMMON_SELECT_K2_E5,
-	PCIE_REG_DBG_COMMON_DWORD_ENABLE_K2_E5,
-	PCIE_REG_DBG_COMMON_SHIFT_K2_E5,
-	PCIE_REG_DBG_COMMON_FORCE_VALID_K2_E5,
-	PCIE_REG_DBG_COMMON_FORCE_FRAME_K2_E5,
-	false, false, MAX_DBG_RESET_REGS, 0
-};
-
-static struct block_defs block_mcp_defs = {
-	"mcp", {true, true, true}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS},
-	0, 0, 0, 0, 0,
-	false, false, MAX_DBG_RESET_REGS, 0
-};
-
-static struct block_defs block_mcp2_defs = {
-	"mcp2",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCZ, DBG_BUS_CLIENT_RBCZ, DBG_BUS_CLIENT_RBCZ},
-	MCP2_REG_DBG_SELECT, MCP2_REG_DBG_DWORD_ENABLE,
-	MCP2_REG_DBG_SHIFT, MCP2_REG_DBG_FORCE_VALID,
-	MCP2_REG_DBG_FORCE_FRAME,
-	false, false, MAX_DBG_RESET_REGS, 0
-};
-
-static struct block_defs block_pswhst_defs = {
-	"pswhst",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCP, DBG_BUS_CLIENT_RBCP, DBG_BUS_CLIENT_RBCP},
-	PSWHST_REG_DBG_SELECT, PSWHST_REG_DBG_DWORD_ENABLE,
-	PSWHST_REG_DBG_SHIFT, PSWHST_REG_DBG_FORCE_VALID,
-	PSWHST_REG_DBG_FORCE_FRAME,
-	true, false, DBG_RESET_REG_MISC_PL_HV, 0
-};
-
-static struct block_defs block_pswhst2_defs = {
-	"pswhst2",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCP, DBG_BUS_CLIENT_RBCP, DBG_BUS_CLIENT_RBCP},
-	PSWHST2_REG_DBG_SELECT, PSWHST2_REG_DBG_DWORD_ENABLE,
-	PSWHST2_REG_DBG_SHIFT, PSWHST2_REG_DBG_FORCE_VALID,
-	PSWHST2_REG_DBG_FORCE_FRAME,
-	true, false, DBG_RESET_REG_MISC_PL_HV, 0
-};
-
-static struct block_defs block_pswrd_defs = {
-	"pswrd",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCP, DBG_BUS_CLIENT_RBCP, DBG_BUS_CLIENT_RBCP},
-	PSWRD_REG_DBG_SELECT, PSWRD_REG_DBG_DWORD_ENABLE,
-	PSWRD_REG_DBG_SHIFT, PSWRD_REG_DBG_FORCE_VALID,
-	PSWRD_REG_DBG_FORCE_FRAME,
-	true, false, DBG_RESET_REG_MISC_PL_HV, 2
-};
-
-static struct block_defs block_pswrd2_defs = {
-	"pswrd2",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCP, DBG_BUS_CLIENT_RBCP, DBG_BUS_CLIENT_RBCP},
-	PSWRD2_REG_DBG_SELECT, PSWRD2_REG_DBG_DWORD_ENABLE,
-	PSWRD2_REG_DBG_SHIFT, PSWRD2_REG_DBG_FORCE_VALID,
-	PSWRD2_REG_DBG_FORCE_FRAME,
-	true, false, DBG_RESET_REG_MISC_PL_HV, 2
-};
-
-static struct block_defs block_pswwr_defs = {
-	"pswwr",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCP, DBG_BUS_CLIENT_RBCP, DBG_BUS_CLIENT_RBCP},
-	PSWWR_REG_DBG_SELECT, PSWWR_REG_DBG_DWORD_ENABLE,
-	PSWWR_REG_DBG_SHIFT, PSWWR_REG_DBG_FORCE_VALID,
-	PSWWR_REG_DBG_FORCE_FRAME,
-	true, false, DBG_RESET_REG_MISC_PL_HV, 3
-};
-
-static struct block_defs block_pswwr2_defs = {
-	"pswwr2", {true, true, true}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS},
-	0, 0, 0, 0, 0,
-	true, false, DBG_RESET_REG_MISC_PL_HV, 3
-};
-
-static struct block_defs block_pswrq_defs = {
-	"pswrq",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCP, DBG_BUS_CLIENT_RBCP, DBG_BUS_CLIENT_RBCP},
-	PSWRQ_REG_DBG_SELECT, PSWRQ_REG_DBG_DWORD_ENABLE,
-	PSWRQ_REG_DBG_SHIFT, PSWRQ_REG_DBG_FORCE_VALID,
-	PSWRQ_REG_DBG_FORCE_FRAME,
-	true, false, DBG_RESET_REG_MISC_PL_HV, 1
-};
-
-static struct block_defs block_pswrq2_defs = {
-	"pswrq2",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCP, DBG_BUS_CLIENT_RBCP, DBG_BUS_CLIENT_RBCP},
-	PSWRQ2_REG_DBG_SELECT, PSWRQ2_REG_DBG_DWORD_ENABLE,
-	PSWRQ2_REG_DBG_SHIFT, PSWRQ2_REG_DBG_FORCE_VALID,
-	PSWRQ2_REG_DBG_FORCE_FRAME,
-	true, false, DBG_RESET_REG_MISC_PL_HV, 1
-};
-
-static struct block_defs block_pglcs_defs = {
-	"pglcs",
-	{true, true, true}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, DBG_BUS_CLIENT_RBCH,
-	 DBG_BUS_CLIENT_RBCH},
-	PGLCS_REG_DBG_SELECT_K2_E5, PGLCS_REG_DBG_DWORD_ENABLE_K2_E5,
-	PGLCS_REG_DBG_SHIFT_K2_E5, PGLCS_REG_DBG_FORCE_VALID_K2_E5,
-	PGLCS_REG_DBG_FORCE_FRAME_K2_E5,
-	true, false, DBG_RESET_REG_MISCS_PL_HV, 2
-};
-
-static struct block_defs block_ptu_defs = {
-	"ptu",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCP, DBG_BUS_CLIENT_RBCP, DBG_BUS_CLIENT_RBCP},
-	PTU_REG_DBG_SELECT, PTU_REG_DBG_DWORD_ENABLE,
-	PTU_REG_DBG_SHIFT, PTU_REG_DBG_FORCE_VALID,
-	PTU_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_2, 20
-};
-
-static struct block_defs block_dmae_defs = {
-	"dmae",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCP, DBG_BUS_CLIENT_RBCP, DBG_BUS_CLIENT_RBCP},
-	DMAE_REG_DBG_SELECT, DMAE_REG_DBG_DWORD_ENABLE,
-	DMAE_REG_DBG_SHIFT, DMAE_REG_DBG_FORCE_VALID,
-	DMAE_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_1, 28
-};
-
-static struct block_defs block_tcm_defs = {
-	"tcm",
-	{true, true, true}, true, DBG_TSTORM_ID,
-	{DBG_BUS_CLIENT_RBCT, DBG_BUS_CLIENT_RBCT, DBG_BUS_CLIENT_RBCT},
-	TCM_REG_DBG_SELECT, TCM_REG_DBG_DWORD_ENABLE,
-	TCM_REG_DBG_SHIFT, TCM_REG_DBG_FORCE_VALID,
-	TCM_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_1, 5
-};
-
-static struct block_defs block_mcm_defs = {
-	"mcm",
-	{true, true, true}, true, DBG_MSTORM_ID,
-	{DBG_BUS_CLIENT_RBCT, DBG_BUS_CLIENT_RBCM, DBG_BUS_CLIENT_RBCM},
-	MCM_REG_DBG_SELECT, MCM_REG_DBG_DWORD_ENABLE,
-	MCM_REG_DBG_SHIFT, MCM_REG_DBG_FORCE_VALID,
-	MCM_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_2, 3
-};
-
-static struct block_defs block_ucm_defs = {
-	"ucm",
-	{true, true, true}, true, DBG_USTORM_ID,
-	{DBG_BUS_CLIENT_RBCU, DBG_BUS_CLIENT_RBCU, DBG_BUS_CLIENT_RBCU},
-	UCM_REG_DBG_SELECT, UCM_REG_DBG_DWORD_ENABLE,
-	UCM_REG_DBG_SHIFT, UCM_REG_DBG_FORCE_VALID,
-	UCM_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_1, 8
-};
-
-static struct block_defs block_xcm_defs = {
-	"xcm",
-	{true, true, true}, true, DBG_XSTORM_ID,
-	{DBG_BUS_CLIENT_RBCX, DBG_BUS_CLIENT_RBCX, DBG_BUS_CLIENT_RBCX},
-	XCM_REG_DBG_SELECT, XCM_REG_DBG_DWORD_ENABLE,
-	XCM_REG_DBG_SHIFT, XCM_REG_DBG_FORCE_VALID,
-	XCM_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_1, 19
-};
-
-static struct block_defs block_ycm_defs = {
-	"ycm",
-	{true, true, true}, true, DBG_YSTORM_ID,
-	{DBG_BUS_CLIENT_RBCX, DBG_BUS_CLIENT_RBCY, DBG_BUS_CLIENT_RBCY},
-	YCM_REG_DBG_SELECT, YCM_REG_DBG_DWORD_ENABLE,
-	YCM_REG_DBG_SHIFT, YCM_REG_DBG_FORCE_VALID,
-	YCM_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_2, 5
-};
-
-static struct block_defs block_pcm_defs = {
-	"pcm",
-	{true, true, true}, true, DBG_PSTORM_ID,
-	{DBG_BUS_CLIENT_RBCS, DBG_BUS_CLIENT_RBCS, DBG_BUS_CLIENT_RBCS},
-	PCM_REG_DBG_SELECT, PCM_REG_DBG_DWORD_ENABLE,
-	PCM_REG_DBG_SHIFT, PCM_REG_DBG_FORCE_VALID,
-	PCM_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_2, 4
-};
-
-static struct block_defs block_qm_defs = {
-	"qm",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCP, DBG_BUS_CLIENT_RBCQ, DBG_BUS_CLIENT_RBCQ},
-	QM_REG_DBG_SELECT, QM_REG_DBG_DWORD_ENABLE,
-	QM_REG_DBG_SHIFT, QM_REG_DBG_FORCE_VALID,
-	QM_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_1, 16
-};
-
-static struct block_defs block_tm_defs = {
-	"tm",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCS, DBG_BUS_CLIENT_RBCS, DBG_BUS_CLIENT_RBCS},
-	TM_REG_DBG_SELECT, TM_REG_DBG_DWORD_ENABLE,
-	TM_REG_DBG_SHIFT, TM_REG_DBG_FORCE_VALID,
-	TM_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_1, 17
-};
-
-static struct block_defs block_dorq_defs = {
-	"dorq",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCX, DBG_BUS_CLIENT_RBCY, DBG_BUS_CLIENT_RBCY},
-	DORQ_REG_DBG_SELECT, DORQ_REG_DBG_DWORD_ENABLE,
-	DORQ_REG_DBG_SHIFT, DORQ_REG_DBG_FORCE_VALID,
-	DORQ_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_1, 18
-};
-
-static struct block_defs block_brb_defs = {
-	"brb",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCR, DBG_BUS_CLIENT_RBCR, DBG_BUS_CLIENT_RBCR},
-	BRB_REG_DBG_SELECT, BRB_REG_DBG_DWORD_ENABLE,
-	BRB_REG_DBG_SHIFT, BRB_REG_DBG_FORCE_VALID,
-	BRB_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_1, 0
-};
-
-static struct block_defs block_src_defs = {
-	"src",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCF, DBG_BUS_CLIENT_RBCF, DBG_BUS_CLIENT_RBCF},
-	SRC_REG_DBG_SELECT, SRC_REG_DBG_DWORD_ENABLE,
-	SRC_REG_DBG_SHIFT, SRC_REG_DBG_FORCE_VALID,
-	SRC_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_1, 2
-};
-
-static struct block_defs block_prs_defs = {
-	"prs",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCR, DBG_BUS_CLIENT_RBCR, DBG_BUS_CLIENT_RBCR},
-	PRS_REG_DBG_SELECT, PRS_REG_DBG_DWORD_ENABLE,
-	PRS_REG_DBG_SHIFT, PRS_REG_DBG_FORCE_VALID,
-	PRS_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_1, 1
-};
-
-static struct block_defs block_tsdm_defs = {
-	"tsdm",
-	{true, true, true}, true, DBG_TSTORM_ID,
-	{DBG_BUS_CLIENT_RBCT, DBG_BUS_CLIENT_RBCT, DBG_BUS_CLIENT_RBCT},
-	TSDM_REG_DBG_SELECT, TSDM_REG_DBG_DWORD_ENABLE,
-	TSDM_REG_DBG_SHIFT, TSDM_REG_DBG_FORCE_VALID,
-	TSDM_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_1, 3
-};
-
-static struct block_defs block_msdm_defs = {
-	"msdm",
-	{true, true, true}, true, DBG_MSTORM_ID,
-	{DBG_BUS_CLIENT_RBCT, DBG_BUS_CLIENT_RBCM, DBG_BUS_CLIENT_RBCM},
-	MSDM_REG_DBG_SELECT, MSDM_REG_DBG_DWORD_ENABLE,
-	MSDM_REG_DBG_SHIFT, MSDM_REG_DBG_FORCE_VALID,
-	MSDM_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_2, 6
-};
-
-static struct block_defs block_usdm_defs = {
-	"usdm",
-	{true, true, true}, true, DBG_USTORM_ID,
-	{DBG_BUS_CLIENT_RBCU, DBG_BUS_CLIENT_RBCU, DBG_BUS_CLIENT_RBCU},
-	USDM_REG_DBG_SELECT, USDM_REG_DBG_DWORD_ENABLE,
-	USDM_REG_DBG_SHIFT, USDM_REG_DBG_FORCE_VALID,
-	USDM_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_1, 7
-};
-
-static struct block_defs block_xsdm_defs = {
-	"xsdm",
-	{true, true, true}, true, DBG_XSTORM_ID,
-	{DBG_BUS_CLIENT_RBCX, DBG_BUS_CLIENT_RBCX, DBG_BUS_CLIENT_RBCX},
-	XSDM_REG_DBG_SELECT, XSDM_REG_DBG_DWORD_ENABLE,
-	XSDM_REG_DBG_SHIFT, XSDM_REG_DBG_FORCE_VALID,
-	XSDM_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_1, 20
-};
-
-static struct block_defs block_ysdm_defs = {
-	"ysdm",
-	{true, true, true}, true, DBG_YSTORM_ID,
-	{DBG_BUS_CLIENT_RBCX, DBG_BUS_CLIENT_RBCY, DBG_BUS_CLIENT_RBCY},
-	YSDM_REG_DBG_SELECT, YSDM_REG_DBG_DWORD_ENABLE,
-	YSDM_REG_DBG_SHIFT, YSDM_REG_DBG_FORCE_VALID,
-	YSDM_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_2, 8
-};
-
-static struct block_defs block_psdm_defs = {
-	"psdm",
-	{true, true, true}, true, DBG_PSTORM_ID,
-	{DBG_BUS_CLIENT_RBCS, DBG_BUS_CLIENT_RBCS, DBG_BUS_CLIENT_RBCS},
-	PSDM_REG_DBG_SELECT, PSDM_REG_DBG_DWORD_ENABLE,
-	PSDM_REG_DBG_SHIFT, PSDM_REG_DBG_FORCE_VALID,
-	PSDM_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_2, 7
-};
-
-static struct block_defs block_tsem_defs = {
-	"tsem",
-	{true, true, true}, true, DBG_TSTORM_ID,
-	{DBG_BUS_CLIENT_RBCT, DBG_BUS_CLIENT_RBCT, DBG_BUS_CLIENT_RBCT},
-	TSEM_REG_DBG_SELECT, TSEM_REG_DBG_DWORD_ENABLE,
-	TSEM_REG_DBG_SHIFT, TSEM_REG_DBG_FORCE_VALID,
-	TSEM_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_1, 4
-};
-
-static struct block_defs block_msem_defs = {
-	"msem",
-	{true, true, true}, true, DBG_MSTORM_ID,
-	{DBG_BUS_CLIENT_RBCT, DBG_BUS_CLIENT_RBCM, DBG_BUS_CLIENT_RBCM},
-	MSEM_REG_DBG_SELECT, MSEM_REG_DBG_DWORD_ENABLE,
-	MSEM_REG_DBG_SHIFT, MSEM_REG_DBG_FORCE_VALID,
-	MSEM_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_2, 9
-};
-
-static struct block_defs block_usem_defs = {
-	"usem",
-	{true, true, true}, true, DBG_USTORM_ID,
-	{DBG_BUS_CLIENT_RBCU, DBG_BUS_CLIENT_RBCU, DBG_BUS_CLIENT_RBCU},
-	USEM_REG_DBG_SELECT, USEM_REG_DBG_DWORD_ENABLE,
-	USEM_REG_DBG_SHIFT, USEM_REG_DBG_FORCE_VALID,
-	USEM_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_1, 9
-};
-
-static struct block_defs block_xsem_defs = {
-	"xsem",
-	{true, true, true}, true, DBG_XSTORM_ID,
-	{DBG_BUS_CLIENT_RBCX, DBG_BUS_CLIENT_RBCX, DBG_BUS_CLIENT_RBCX},
-	XSEM_REG_DBG_SELECT, XSEM_REG_DBG_DWORD_ENABLE,
-	XSEM_REG_DBG_SHIFT, XSEM_REG_DBG_FORCE_VALID,
-	XSEM_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_1, 21
-};
-
-static struct block_defs block_ysem_defs = {
-	"ysem",
-	{true, true, true}, true, DBG_YSTORM_ID,
-	{DBG_BUS_CLIENT_RBCX, DBG_BUS_CLIENT_RBCY, DBG_BUS_CLIENT_RBCY},
-	YSEM_REG_DBG_SELECT, YSEM_REG_DBG_DWORD_ENABLE,
-	YSEM_REG_DBG_SHIFT, YSEM_REG_DBG_FORCE_VALID,
-	YSEM_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_2, 11
-};
-
-static struct block_defs block_psem_defs = {
-	"psem",
-	{true, true, true}, true, DBG_PSTORM_ID,
-	{DBG_BUS_CLIENT_RBCS, DBG_BUS_CLIENT_RBCS, DBG_BUS_CLIENT_RBCS},
-	PSEM_REG_DBG_SELECT, PSEM_REG_DBG_DWORD_ENABLE,
-	PSEM_REG_DBG_SHIFT, PSEM_REG_DBG_FORCE_VALID,
-	PSEM_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_2, 10
-};
-
-static struct block_defs block_rss_defs = {
-	"rss",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCT, DBG_BUS_CLIENT_RBCT, DBG_BUS_CLIENT_RBCT},
-	RSS_REG_DBG_SELECT, RSS_REG_DBG_DWORD_ENABLE,
-	RSS_REG_DBG_SHIFT, RSS_REG_DBG_FORCE_VALID,
-	RSS_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_2, 18
-};
-
-static struct block_defs block_tmld_defs = {
-	"tmld",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCT, DBG_BUS_CLIENT_RBCM, DBG_BUS_CLIENT_RBCM},
-	TMLD_REG_DBG_SELECT, TMLD_REG_DBG_DWORD_ENABLE,
-	TMLD_REG_DBG_SHIFT, TMLD_REG_DBG_FORCE_VALID,
-	TMLD_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_2, 13
-};
-
-static struct block_defs block_muld_defs = {
-	"muld",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCU, DBG_BUS_CLIENT_RBCU, DBG_BUS_CLIENT_RBCU},
-	MULD_REG_DBG_SELECT, MULD_REG_DBG_DWORD_ENABLE,
-	MULD_REG_DBG_SHIFT, MULD_REG_DBG_FORCE_VALID,
-	MULD_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_2, 14
-};
-
-static struct block_defs block_yuld_defs = {
-	"yuld",
-	{true, true, false}, false, 0,
-	{DBG_BUS_CLIENT_RBCU, DBG_BUS_CLIENT_RBCU,
-	 MAX_DBG_BUS_CLIENTS},
-	YULD_REG_DBG_SELECT_BB_K2, YULD_REG_DBG_DWORD_ENABLE_BB_K2,
-	YULD_REG_DBG_SHIFT_BB_K2, YULD_REG_DBG_FORCE_VALID_BB_K2,
-	YULD_REG_DBG_FORCE_FRAME_BB_K2,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_2,
-	15
-};
-
-static struct block_defs block_xyld_defs = {
-	"xyld",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCX, DBG_BUS_CLIENT_RBCX, DBG_BUS_CLIENT_RBCX},
-	XYLD_REG_DBG_SELECT, XYLD_REG_DBG_DWORD_ENABLE,
-	XYLD_REG_DBG_SHIFT, XYLD_REG_DBG_FORCE_VALID,
-	XYLD_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_2, 12
-};
-
-static struct block_defs block_ptld_defs = {
-	"ptld",
-	{false, false, true}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS, DBG_BUS_CLIENT_RBCT},
-	PTLD_REG_DBG_SELECT_E5, PTLD_REG_DBG_DWORD_ENABLE_E5,
-	PTLD_REG_DBG_SHIFT_E5, PTLD_REG_DBG_FORCE_VALID_E5,
-	PTLD_REG_DBG_FORCE_FRAME_E5,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_2,
-	28
-};
-
-static struct block_defs block_ypld_defs = {
-	"ypld",
-	{false, false, true}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS, DBG_BUS_CLIENT_RBCS},
-	YPLD_REG_DBG_SELECT_E5, YPLD_REG_DBG_DWORD_ENABLE_E5,
-	YPLD_REG_DBG_SHIFT_E5, YPLD_REG_DBG_FORCE_VALID_E5,
-	YPLD_REG_DBG_FORCE_FRAME_E5,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_2,
-	27
-};
-
-static struct block_defs block_prm_defs = {
-	"prm",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCT, DBG_BUS_CLIENT_RBCM, DBG_BUS_CLIENT_RBCM},
-	PRM_REG_DBG_SELECT, PRM_REG_DBG_DWORD_ENABLE,
-	PRM_REG_DBG_SHIFT, PRM_REG_DBG_FORCE_VALID,
-	PRM_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_2, 21
-};
-
-static struct block_defs block_pbf_pb1_defs = {
-	"pbf_pb1",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCS, DBG_BUS_CLIENT_RBCV, DBG_BUS_CLIENT_RBCV},
-	PBF_PB1_REG_DBG_SELECT, PBF_PB1_REG_DBG_DWORD_ENABLE,
-	PBF_PB1_REG_DBG_SHIFT, PBF_PB1_REG_DBG_FORCE_VALID,
-	PBF_PB1_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_1,
-	11
-};
-
-static struct block_defs block_pbf_pb2_defs = {
-	"pbf_pb2",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCS, DBG_BUS_CLIENT_RBCV, DBG_BUS_CLIENT_RBCV},
-	PBF_PB2_REG_DBG_SELECT, PBF_PB2_REG_DBG_DWORD_ENABLE,
-	PBF_PB2_REG_DBG_SHIFT, PBF_PB2_REG_DBG_FORCE_VALID,
-	PBF_PB2_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_1,
-	12
-};
-
-static struct block_defs block_rpb_defs = {
-	"rpb",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCT, DBG_BUS_CLIENT_RBCM, DBG_BUS_CLIENT_RBCM},
-	RPB_REG_DBG_SELECT, RPB_REG_DBG_DWORD_ENABLE,
-	RPB_REG_DBG_SHIFT, RPB_REG_DBG_FORCE_VALID,
-	RPB_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_1, 13
-};
-
-static struct block_defs block_btb_defs = {
-	"btb",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCR, DBG_BUS_CLIENT_RBCV, DBG_BUS_CLIENT_RBCV},
-	BTB_REG_DBG_SELECT, BTB_REG_DBG_DWORD_ENABLE,
-	BTB_REG_DBG_SHIFT, BTB_REG_DBG_FORCE_VALID,
-	BTB_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_1, 10
-};
-
-static struct block_defs block_pbf_defs = {
-	"pbf",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCS, DBG_BUS_CLIENT_RBCV, DBG_BUS_CLIENT_RBCV},
-	PBF_REG_DBG_SELECT, PBF_REG_DBG_DWORD_ENABLE,
-	PBF_REG_DBG_SHIFT, PBF_REG_DBG_FORCE_VALID,
-	PBF_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_1, 15
-};
-
-static struct block_defs block_rdif_defs = {
-	"rdif",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCT, DBG_BUS_CLIENT_RBCM, DBG_BUS_CLIENT_RBCM},
-	RDIF_REG_DBG_SELECT, RDIF_REG_DBG_DWORD_ENABLE,
-	RDIF_REG_DBG_SHIFT, RDIF_REG_DBG_FORCE_VALID,
-	RDIF_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_2, 16
-};
-
-static struct block_defs block_tdif_defs = {
-	"tdif",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCS, DBG_BUS_CLIENT_RBCS, DBG_BUS_CLIENT_RBCS},
-	TDIF_REG_DBG_SELECT, TDIF_REG_DBG_DWORD_ENABLE,
-	TDIF_REG_DBG_SHIFT, TDIF_REG_DBG_FORCE_VALID,
-	TDIF_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_2, 17
-};
-
-static struct block_defs block_cdu_defs = {
-	"cdu",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCF, DBG_BUS_CLIENT_RBCF, DBG_BUS_CLIENT_RBCF},
-	CDU_REG_DBG_SELECT, CDU_REG_DBG_DWORD_ENABLE,
-	CDU_REG_DBG_SHIFT, CDU_REG_DBG_FORCE_VALID,
-	CDU_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_1, 23
-};
-
-static struct block_defs block_ccfc_defs = {
-	"ccfc",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCF, DBG_BUS_CLIENT_RBCF, DBG_BUS_CLIENT_RBCF},
-	CCFC_REG_DBG_SELECT, CCFC_REG_DBG_DWORD_ENABLE,
-	CCFC_REG_DBG_SHIFT, CCFC_REG_DBG_FORCE_VALID,
-	CCFC_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_1, 24
-};
-
-static struct block_defs block_tcfc_defs = {
-	"tcfc",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCF, DBG_BUS_CLIENT_RBCF, DBG_BUS_CLIENT_RBCF},
-	TCFC_REG_DBG_SELECT, TCFC_REG_DBG_DWORD_ENABLE,
-	TCFC_REG_DBG_SHIFT, TCFC_REG_DBG_FORCE_VALID,
-	TCFC_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_1, 25
-};
-
-static struct block_defs block_igu_defs = {
-	"igu",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCP, DBG_BUS_CLIENT_RBCP, DBG_BUS_CLIENT_RBCP},
-	IGU_REG_DBG_SELECT, IGU_REG_DBG_DWORD_ENABLE,
-	IGU_REG_DBG_SHIFT, IGU_REG_DBG_FORCE_VALID,
-	IGU_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_1, 27
-};
-
-static struct block_defs block_cau_defs = {
-	"cau",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCP, DBG_BUS_CLIENT_RBCP, DBG_BUS_CLIENT_RBCP},
-	CAU_REG_DBG_SELECT, CAU_REG_DBG_DWORD_ENABLE,
-	CAU_REG_DBG_SHIFT, CAU_REG_DBG_FORCE_VALID,
-	CAU_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_2, 19
-};
-
-static struct block_defs block_rgfs_defs = {
-	"rgfs", {false, false, true}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS},
-	0, 0, 0, 0, 0,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_2, 29
-};
-
-static struct block_defs block_rgsrc_defs = {
-	"rgsrc",
-	{false, false, true}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS, DBG_BUS_CLIENT_RBCH},
-	RGSRC_REG_DBG_SELECT_E5, RGSRC_REG_DBG_DWORD_ENABLE_E5,
-	RGSRC_REG_DBG_SHIFT_E5, RGSRC_REG_DBG_FORCE_VALID_E5,
-	RGSRC_REG_DBG_FORCE_FRAME_E5,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_1,
-	30
-};
-
-static struct block_defs block_tgfs_defs = {
-	"tgfs", {false, false, true}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS},
-	0, 0, 0, 0, 0,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_2, 30
-};
-
-static struct block_defs block_tgsrc_defs = {
-	"tgsrc",
-	{false, false, true}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS, DBG_BUS_CLIENT_RBCV},
-	TGSRC_REG_DBG_SELECT_E5, TGSRC_REG_DBG_DWORD_ENABLE_E5,
-	TGSRC_REG_DBG_SHIFT_E5, TGSRC_REG_DBG_FORCE_VALID_E5,
-	TGSRC_REG_DBG_FORCE_FRAME_E5,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VMAIN_1,
-	31
-};
-
-static struct block_defs block_umac_defs = {
-	"umac",
-	{true, true, true}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, DBG_BUS_CLIENT_RBCZ,
-	 DBG_BUS_CLIENT_RBCZ},
-	UMAC_REG_DBG_SELECT_K2_E5, UMAC_REG_DBG_DWORD_ENABLE_K2_E5,
-	UMAC_REG_DBG_SHIFT_K2_E5, UMAC_REG_DBG_FORCE_VALID_K2_E5,
-	UMAC_REG_DBG_FORCE_FRAME_K2_E5,
-	true, false, DBG_RESET_REG_MISCS_PL_HV, 6
-};
-
-static struct block_defs block_xmac_defs = {
-	"xmac", {true, false, false}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS},
-	0, 0, 0, 0, 0,
-	false, false, MAX_DBG_RESET_REGS, 0
-};
-
-static struct block_defs block_dbg_defs = {
-	"dbg", {true, true, true}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS},
-	0, 0, 0, 0, 0,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VAUX, 3
-};
-
-static struct block_defs block_nig_defs = {
-	"nig",
-	{true, true, true}, false, 0,
-	{DBG_BUS_CLIENT_RBCN, DBG_BUS_CLIENT_RBCN, DBG_BUS_CLIENT_RBCN},
-	NIG_REG_DBG_SELECT, NIG_REG_DBG_DWORD_ENABLE,
-	NIG_REG_DBG_SHIFT, NIG_REG_DBG_FORCE_VALID,
-	NIG_REG_DBG_FORCE_FRAME,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VAUX, 0
-};
-
-static struct block_defs block_wol_defs = {
-	"wol",
-	{false, true, true}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, DBG_BUS_CLIENT_RBCZ, DBG_BUS_CLIENT_RBCZ},
-	WOL_REG_DBG_SELECT_K2_E5, WOL_REG_DBG_DWORD_ENABLE_K2_E5,
-	WOL_REG_DBG_SHIFT_K2_E5, WOL_REG_DBG_FORCE_VALID_K2_E5,
-	WOL_REG_DBG_FORCE_FRAME_K2_E5,
-	true, true, DBG_RESET_REG_MISC_PL_PDA_VAUX, 7
-};
-
-static struct block_defs block_bmbn_defs = {
-	"bmbn",
-	{false, true, true}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, DBG_BUS_CLIENT_RBCB,
-	 DBG_BUS_CLIENT_RBCB},
-	BMBN_REG_DBG_SELECT_K2_E5, BMBN_REG_DBG_DWORD_ENABLE_K2_E5,
-	BMBN_REG_DBG_SHIFT_K2_E5, BMBN_REG_DBG_FORCE_VALID_K2_E5,
-	BMBN_REG_DBG_FORCE_FRAME_K2_E5,
-	false, false, MAX_DBG_RESET_REGS, 0
-};
-
-static struct block_defs block_ipc_defs = {
-	"ipc", {true, true, true}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS},
-	0, 0, 0, 0, 0,
-	true, false, DBG_RESET_REG_MISCS_PL_UA, 8
-};
-
-static struct block_defs block_nwm_defs = {
-	"nwm",
-	{false, true, true}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, DBG_BUS_CLIENT_RBCW, DBG_BUS_CLIENT_RBCW},
-	NWM_REG_DBG_SELECT_K2_E5, NWM_REG_DBG_DWORD_ENABLE_K2_E5,
-	NWM_REG_DBG_SHIFT_K2_E5, NWM_REG_DBG_FORCE_VALID_K2_E5,
-	NWM_REG_DBG_FORCE_FRAME_K2_E5,
-	true, false, DBG_RESET_REG_MISCS_PL_HV_2, 0
-};
-
-static struct block_defs block_nws_defs = {
-	"nws",
-	{false, true, true}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, DBG_BUS_CLIENT_RBCW, DBG_BUS_CLIENT_RBCW},
-	NWS_REG_DBG_SELECT_K2_E5, NWS_REG_DBG_DWORD_ENABLE_K2_E5,
-	NWS_REG_DBG_SHIFT_K2_E5, NWS_REG_DBG_FORCE_VALID_K2_E5,
-	NWS_REG_DBG_FORCE_FRAME_K2_E5,
-	true, false, DBG_RESET_REG_MISCS_PL_HV, 12
-};
-
-static struct block_defs block_ms_defs = {
-	"ms",
-	{false, true, true}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, DBG_BUS_CLIENT_RBCZ, DBG_BUS_CLIENT_RBCZ},
-	MS_REG_DBG_SELECT_K2_E5, MS_REG_DBG_DWORD_ENABLE_K2_E5,
-	MS_REG_DBG_SHIFT_K2_E5, MS_REG_DBG_FORCE_VALID_K2_E5,
-	MS_REG_DBG_FORCE_FRAME_K2_E5,
-	true, false, DBG_RESET_REG_MISCS_PL_HV, 13
-};
-
-static struct block_defs block_phy_pcie_defs = {
-	"phy_pcie",
-	{false, true, true}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, DBG_BUS_CLIENT_RBCH,
-	 DBG_BUS_CLIENT_RBCH},
-	PCIE_REG_DBG_COMMON_SELECT_K2_E5,
-	PCIE_REG_DBG_COMMON_DWORD_ENABLE_K2_E5,
-	PCIE_REG_DBG_COMMON_SHIFT_K2_E5,
-	PCIE_REG_DBG_COMMON_FORCE_VALID_K2_E5,
-	PCIE_REG_DBG_COMMON_FORCE_FRAME_K2_E5,
-	false, false, MAX_DBG_RESET_REGS, 0
-};
-
-static struct block_defs block_led_defs = {
-	"led", {false, true, true}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS},
-	0, 0, 0, 0, 0,
-	true, false, DBG_RESET_REG_MISCS_PL_HV, 14
-};
-
-static struct block_defs block_avs_wrap_defs = {
-	"avs_wrap", {false, true, false}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS},
-	0, 0, 0, 0, 0,
-	true, false, DBG_RESET_REG_MISCS_PL_UA, 11
-};
-
-static struct block_defs block_pxpreqbus_defs = {
-	"pxpreqbus", {false, false, false}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS},
-	0, 0, 0, 0, 0,
-	false, false, MAX_DBG_RESET_REGS, 0
-};
-
-static struct block_defs block_misc_aeu_defs = {
-	"misc_aeu", {true, true, true}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS},
-	0, 0, 0, 0, 0,
-	false, false, MAX_DBG_RESET_REGS, 0
-};
-
-static struct block_defs block_bar0_map_defs = {
-	"bar0_map", {true, true, true}, false, 0,
-	{MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS, MAX_DBG_BUS_CLIENTS},
-	0, 0, 0, 0, 0,
-	false, false, MAX_DBG_RESET_REGS, 0
-};
-
-static struct block_defs *s_block_defs[MAX_BLOCK_ID] = {
-	&block_grc_defs,
-	&block_miscs_defs,
-	&block_misc_defs,
-	&block_dbu_defs,
-	&block_pglue_b_defs,
-	&block_cnig_defs,
-	&block_cpmu_defs,
-	&block_ncsi_defs,
-	&block_opte_defs,
-	&block_bmb_defs,
-	&block_pcie_defs,
-	&block_mcp_defs,
-	&block_mcp2_defs,
-	&block_pswhst_defs,
-	&block_pswhst2_defs,
-	&block_pswrd_defs,
-	&block_pswrd2_defs,
-	&block_pswwr_defs,
-	&block_pswwr2_defs,
-	&block_pswrq_defs,
-	&block_pswrq2_defs,
-	&block_pglcs_defs,
-	&block_dmae_defs,
-	&block_ptu_defs,
-	&block_tcm_defs,
-	&block_mcm_defs,
-	&block_ucm_defs,
-	&block_xcm_defs,
-	&block_ycm_defs,
-	&block_pcm_defs,
-	&block_qm_defs,
-	&block_tm_defs,
-	&block_dorq_defs,
-	&block_brb_defs,
-	&block_src_defs,
-	&block_prs_defs,
-	&block_tsdm_defs,
-	&block_msdm_defs,
-	&block_usdm_defs,
-	&block_xsdm_defs,
-	&block_ysdm_defs,
-	&block_psdm_defs,
-	&block_tsem_defs,
-	&block_msem_defs,
-	&block_usem_defs,
-	&block_xsem_defs,
-	&block_ysem_defs,
-	&block_psem_defs,
-	&block_rss_defs,
-	&block_tmld_defs,
-	&block_muld_defs,
-	&block_yuld_defs,
-	&block_xyld_defs,
-	&block_ptld_defs,
-	&block_ypld_defs,
-	&block_prm_defs,
-	&block_pbf_pb1_defs,
-	&block_pbf_pb2_defs,
-	&block_rpb_defs,
-	&block_btb_defs,
-	&block_pbf_defs,
-	&block_rdif_defs,
-	&block_tdif_defs,
-	&block_cdu_defs,
-	&block_ccfc_defs,
-	&block_tcfc_defs,
-	&block_igu_defs,
-	&block_cau_defs,
-	&block_rgfs_defs,
-	&block_rgsrc_defs,
-	&block_tgfs_defs,
-	&block_tgsrc_defs,
-	&block_umac_defs,
-	&block_xmac_defs,
-	&block_dbg_defs,
-	&block_nig_defs,
-	&block_wol_defs,
-	&block_bmbn_defs,
-	&block_ipc_defs,
-	&block_nwm_defs,
-	&block_nws_defs,
-	&block_ms_defs,
-	&block_phy_pcie_defs,
-	&block_led_defs,
-	&block_avs_wrap_defs,
-	&block_pxpreqbus_defs,
-	&block_misc_aeu_defs,
-	&block_bar0_map_defs,
-};
-
-static struct platform_defs s_platform_defs[] = {
+		{DBG_BUS_CLIENT_RBCS, DBG_BUS_CLIENT_RBCS},
+		true,
+		PSEM_REG_FAST_MEMORY,
+		PSEM_REG_DBG_FRAME_MODE_BB_K2,
+		PSEM_REG_SLOW_DBG_ACTIVE_BB_K2,
+		PSEM_REG_SLOW_DBG_MODE_BB_K2,
+		PSEM_REG_DBG_MODE1_CFG_BB_K2,
+		PSEM_REG_SYNC_DBG_EMPTY,
+		PSEM_REG_DBG_GPRE_VECT,
+		PCM_REG_CTX_RBC_ACCS,
+		{0, PCM_REG_SM_CON_CTX, 0, 0},
+		{{0, 10, 0, 0}, {0, 10, 0, 0}} /* {bb} {k2} */
+	},
+};
+
+static struct hw_type_defs s_hw_type_defs[] = {
+	/* HW_TYPE_ASIC */
 	{"asic", 1, 256, 32768},
 	{"reserved", 0, 0, 0},
 	{"reserved2", 0, 0, 0},
@@ -1509,161 +579,159 @@ static struct platform_defs s_platform_defs[] = {
 
 static struct grc_param_defs s_grc_param_defs[] = {
 	/* DBG_GRC_PARAM_DUMP_TSTORM */
-	{{1, 1, 1}, 0, 1, false, false, 1, 1},
+	{{1, 1}, 0, 1, false, false, 1, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_MSTORM */
-	{{1, 1, 1}, 0, 1, false, false, 1, 1},
+	{{1, 1}, 0, 1, false, false, 1, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_USTORM */
-	{{1, 1, 1}, 0, 1, false, false, 1, 1},
+	{{1, 1}, 0, 1, false, false, 1, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_XSTORM */
-	{{1, 1, 1}, 0, 1, false, false, 1, 1},
+	{{1, 1}, 0, 1, false, false, 1, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_YSTORM */
-	{{1, 1, 1}, 0, 1, false, false, 1, 1},
+	{{1, 1}, 0, 1, false, false, 1, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_PSTORM */
-	{{1, 1, 1}, 0, 1, false, false, 1, 1},
+	{{1, 1}, 0, 1, false, false, 1, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_REGS */
-	{{1, 1, 1}, 0, 1, false, false, 0, 1},
+	{{1, 1}, 0, 1, false, false, 0, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_RAM */
-	{{1, 1, 1}, 0, 1, false, false, 0, 1},
+	{{1, 1}, 0, 1, false, false, 0, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_PBUF */
-	{{1, 1, 1}, 0, 1, false, false, 0, 1},
+	{{1, 1}, 0, 1, false, false, 0, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_IOR */
-	{{0, 0, 0}, 0, 1, false, false, 0, 1},
+	{{0, 0}, 0, 1, false, false, 0, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_VFC */
-	{{0, 0, 0}, 0, 1, false, false, 0, 1},
+	{{0, 0}, 0, 1, false, false, 0, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_CM_CTX */
-	{{1, 1, 1}, 0, 1, false, false, 0, 1},
+	{{1, 1}, 0, 1, false, false, 0, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_ILT */
-	{{1, 1, 1}, 0, 1, false, false, 0, 1},
+	{{1, 1}, 0, 1, false, false, 0, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_RSS */
-	{{1, 1, 1}, 0, 1, false, false, 0, 1},
+	{{1, 1}, 0, 1, false, false, 0, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_CAU */
-	{{1, 1, 1}, 0, 1, false, false, 0, 1},
+	{{1, 1}, 0, 1, false, false, 0, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_QM */
-	{{1, 1, 1}, 0, 1, false, false, 0, 1},
+	{{1, 1}, 0, 1, false, false, 0, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_MCP */
-	{{1, 1, 1}, 0, 1, false, false, 0, 1},
+	{{1, 1}, 0, 1, false, false, 0, {1, 1}},
 
-	/* DBG_GRC_PARAM_MCP_TRACE_META_SIZE */
-	{{1, 1, 1}, 1, 0xffffffff, false, true, 0, 1},
+	/* DBG_GRC_PARAM_DUMP_DORQ */
+	{{1, 1}, 0, 1, false, false, 0, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_CFC */
-	{{1, 1, 1}, 0, 1, false, false, 0, 1},
+	{{1, 1}, 0, 1, false, false, 0, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_IGU */
-	{{1, 1, 1}, 0, 1, false, false, 0, 1},
+	{{1, 1}, 0, 1, false, false, 0, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_BRB */
-	{{0, 0, 0}, 0, 1, false, false, 0, 1},
+	{{0, 0}, 0, 1, false, false, 0, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_BTB */
-	{{0, 0, 0}, 0, 1, false, false, 0, 1},
+	{{0, 0}, 0, 1, false, false, 0, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_BMB */
-	{{0, 0, 0}, 0, 1, false, false, 0, 0},
+	{{0, 0}, 0, 1, false, false, 0, {0, 0}},
 
-	/* DBG_GRC_PARAM_DUMP_NIG */
-	{{1, 1, 1}, 0, 1, false, false, 0, 1},
+	/* DBG_GRC_PARAM_RESERVED1 */
+	{{0, 0}, 0, 1, false, false, 0, {0, 0}},
 
 	/* DBG_GRC_PARAM_DUMP_MULD */
-	{{1, 1, 1}, 0, 1, false, false, 0, 1},
+	{{1, 1}, 0, 1, false, false, 0, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_PRS */
-	{{1, 1, 1}, 0, 1, false, false, 0, 1},
+	{{1, 1}, 0, 1, false, false, 0, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_DMAE */
-	{{1, 1, 1}, 0, 1, false, false, 0, 1},
+	{{1, 1}, 0, 1, false, false, 0, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_TM */
-	{{1, 1, 1}, 0, 1, false, false, 0, 1},
+	{{1, 1}, 0, 1, false, false, 0, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_SDM */
-	{{1, 1, 1}, 0, 1, false, false, 0, 1},
+	{{1, 1}, 0, 1, false, false, 0, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_DIF */
-	{{1, 1, 1}, 0, 1, false, false, 0, 1},
+	{{1, 1}, 0, 1, false, false, 0, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_STATIC */
-	{{1, 1, 1}, 0, 1, false, false, 0, 1},
+	{{1, 1}, 0, 1, false, false, 0, {1, 1}},
 
 	/* DBG_GRC_PARAM_UNSTALL */
-	{{0, 0, 0}, 0, 1, false, false, 0, 0},
+	{{0, 0}, 0, 1, false, false, 0, {0, 0}},
 
-	/* DBG_GRC_PARAM_NUM_LCIDS */
-	{{MAX_LCIDS, MAX_LCIDS, MAX_LCIDS}, 1, MAX_LCIDS, false, false,
-	 MAX_LCIDS, MAX_LCIDS},
+	/* DBG_GRC_PARAM_RESERVED2 */
+	{{0, 0}, 0, 1, false, false, 0, {0, 0}},
 
-	/* DBG_GRC_PARAM_NUM_LTIDS */
-	{{MAX_LTIDS, MAX_LTIDS, MAX_LTIDS}, 1, MAX_LTIDS, false, false,
-	 MAX_LTIDS, MAX_LTIDS},
+	/* DBG_GRC_PARAM_MCP_TRACE_META_SIZE */
+	{{0, 0}, 1, 0xffffffff, false, true, 0, {0, 0}},
 
 	/* DBG_GRC_PARAM_EXCLUDE_ALL */
-	{{0, 0, 0}, 0, 1, true, false, 0, 0},
+	{{0, 0}, 0, 1, true, false, 0, {0, 0}},
 
 	/* DBG_GRC_PARAM_CRASH */
-	{{0, 0, 0}, 0, 1, true, false, 0, 0},
+	{{0, 0}, 0, 1, true, false, 0, {0, 0}},
 
 	/* DBG_GRC_PARAM_PARITY_SAFE */
-	{{0, 0, 0}, 0, 1, false, false, 1, 0},
+	{{0, 0}, 0, 1, false, false, 0, {0, 0}},
 
 	/* DBG_GRC_PARAM_DUMP_CM */
-	{{1, 1, 1}, 0, 1, false, false, 0, 1},
+	{{1, 1}, 0, 1, false, false, 0, {1, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_PHY */
-	{{1, 1, 1}, 0, 1, false, false, 0, 1},
+	{{0, 0}, 0, 1, false, false, 0, {0, 0}},
 
 	/* DBG_GRC_PARAM_NO_MCP */
-	{{0, 0, 0}, 0, 1, false, false, 0, 0},
+	{{0, 0}, 0, 1, false, false, 0, {0, 0}},
 
 	/* DBG_GRC_PARAM_NO_FW_VER */
-	{{0, 0, 0}, 0, 1, false, false, 0, 0},
+	{{0, 0}, 0, 1, false, false, 0, {0, 0}},
 
 	/* DBG_GRC_PARAM_RESERVED3 */
-	{{0, 0, 0}, 0, 1, false, false, 0, 0},
+	{{0, 0}, 0, 1, false, false, 0, {0, 0}},
 
 	/* DBG_GRC_PARAM_DUMP_MCP_HW_DUMP */
-	{{0, 1, 1}, 0, 1, false, false, 0, 0},
+	{{0, 1}, 0, 1, false, false, 0, {0, 1}},
 
 	/* DBG_GRC_PARAM_DUMP_ILT_CDUC */
-	{{1, 1, 1}, 0, 1, false, false, 0, 0},
+	{{1, 1}, 0, 1, false, false, 0, {0, 0}},
 
 	/* DBG_GRC_PARAM_DUMP_ILT_CDUT */
-	{{1, 1, 1}, 0, 1, false, false, 0, 0},
+	{{1, 1}, 0, 1, false, false, 0, {0, 0}},
 
 	/* DBG_GRC_PARAM_DUMP_CAU_EXT */
-	{{0, 0, 0}, 0, 1, false, false, 0, 1}
+	{{0, 0}, 0, 1, false, false, 0, {1, 1}}
 };
 
 static struct rss_mem_defs s_rss_mem_defs[] = {
-	{ "rss_mem_cid", "rss_cid", 0, 32,
-	  {256, 320, 512} },
+	{"rss_mem_cid", "rss_cid", 0, 32,
+	 {256, 320}},
 
-	{ "rss_mem_key_msb", "rss_key", 1024, 256,
-	  {128, 208, 257} },
+	{"rss_mem_key_msb", "rss_key", 1024, 256,
+	 {128, 208}},
 
-	{ "rss_mem_key_lsb", "rss_key", 2048, 64,
-	  {128, 208, 257} },
+	{"rss_mem_key_lsb", "rss_key", 2048, 64,
+	 {128, 208}},
 
-	{ "rss_mem_info", "rss_info", 3072, 16,
-	  {128, 208, 256} },
+	{"rss_mem_info", "rss_info", 3072, 16,
+	 {128, 208}},
 
-	{ "rss_mem_ind", "rss_ind", 4096, 16,
-	  {16384, 26624, 32768} }
+	{"rss_mem_ind", "rss_ind", 4096, 16,
+	 {16384, 26624}}
 };
 
 static struct vfc_ram_defs s_vfc_ram_defs[] = {
@@ -1674,54 +742,31 @@ static struct vfc_ram_defs s_vfc_ram_defs[] = {
 };
 
 static struct big_ram_defs s_big_ram_defs[] = {
-	{ "BRB", MEM_GROUP_BRB_MEM, MEM_GROUP_BRB_RAM, DBG_GRC_PARAM_DUMP_BRB,
-	  BRB_REG_BIG_RAM_ADDRESS, BRB_REG_BIG_RAM_DATA,
-	  MISC_REG_BLOCK_256B_EN, {0, 0, 0},
-	  {153600, 180224, 282624} },
-
-	{ "BTB", MEM_GROUP_BTB_MEM, MEM_GROUP_BTB_RAM, DBG_GRC_PARAM_DUMP_BTB,
-	  BTB_REG_BIG_RAM_ADDRESS, BTB_REG_BIG_RAM_DATA,
-	  MISC_REG_BLOCK_256B_EN, {0, 1, 1},
-	  {92160, 117760, 168960} },
-
-	{ "BMB", MEM_GROUP_BMB_MEM, MEM_GROUP_BMB_RAM, DBG_GRC_PARAM_DUMP_BMB,
-	  BMB_REG_BIG_RAM_ADDRESS, BMB_REG_BIG_RAM_DATA,
-	  MISCS_REG_BLOCK_256B_EN, {0, 0, 0},
-	  {36864, 36864, 36864} }
-};
-
-static struct reset_reg_defs s_reset_regs_defs[] = {
-	/* DBG_RESET_REG_MISCS_PL_UA */
-	{ MISCS_REG_RESET_PL_UA,
-	  {true, true, true}, {0x0, 0x0, 0x0} },
-
-	/* DBG_RESET_REG_MISCS_PL_HV */
-	{ MISCS_REG_RESET_PL_HV,
-	  {true, true, true}, {0x0, 0x400, 0x600} },
+	{"BRB", MEM_GROUP_BRB_MEM, MEM_GROUP_BRB_RAM, DBG_GRC_PARAM_DUMP_BRB,
+	 BRB_REG_BIG_RAM_ADDRESS, BRB_REG_BIG_RAM_DATA,
+	 MISC_REG_BLOCK_256B_EN, {0, 0},
+	 {153600, 180224}},
 
-	/* DBG_RESET_REG_MISCS_PL_HV_2 */
-	{ MISCS_REG_RESET_PL_HV_2_K2_E5,
-	  {false, true, true}, {0x0, 0x0, 0x0} },
+	{"BTB", MEM_GROUP_BTB_MEM, MEM_GROUP_BTB_RAM, DBG_GRC_PARAM_DUMP_BTB,
+	 BTB_REG_BIG_RAM_ADDRESS, BTB_REG_BIG_RAM_DATA,
+	 MISC_REG_BLOCK_256B_EN, {0, 1},
+	 {92160, 117760}},
 
-	/* DBG_RESET_REG_MISC_PL_UA */
-	{ MISC_REG_RESET_PL_UA,
-	  {true, true, true}, {0x0, 0x0, 0x0} },
-
-	/* DBG_RESET_REG_MISC_PL_HV */
-	{ MISC_REG_RESET_PL_HV,
-	  {true, true, true}, {0x0, 0x0, 0x0} },
-
-	/* DBG_RESET_REG_MISC_PL_PDA_VMAIN_1 */
-	{ MISC_REG_RESET_PL_PDA_VMAIN_1,
-	  {true, true, true}, {0x4404040, 0x4404040, 0x404040} },
-
-	/* DBG_RESET_REG_MISC_PL_PDA_VMAIN_2 */
-	{ MISC_REG_RESET_PL_PDA_VMAIN_2,
-	  {true, true, true}, {0x7, 0x7c00007, 0x5c08007} },
+	{"BMB", MEM_GROUP_BMB_MEM, MEM_GROUP_BMB_RAM, DBG_GRC_PARAM_DUMP_BMB,
+	 BMB_REG_BIG_RAM_ADDRESS, BMB_REG_BIG_RAM_DATA,
+	 MISCS_REG_BLOCK_256B_EN, {0, 0},
+	 {36864, 36864}}
+};
 
-	/* DBG_RESET_REG_MISC_PL_PDA_VAUX */
-	{ MISC_REG_RESET_PL_PDA_VAUX,
-	  {true, true, true}, {0x2, 0x2, 0x2} },
+static struct rbc_reset_defs s_rbc_reset_defs[] = {
+	{MISCS_REG_RESET_PL_HV,
+	 {0x0, 0x400}},
+	{MISC_REG_RESET_PL_PDA_VMAIN_1,
+	 {0x4404040, 0x4404040}},
+	{MISC_REG_RESET_PL_PDA_VMAIN_2,
+	 {0x7, 0x7c00007}},
+	{MISC_REG_RESET_PL_PDA_VAUX,
+	 {0x2, 0x2}},
 };
 
 static struct phy_defs s_phy_defs[] = {
@@ -1804,9 +849,19 @@ static void qed_dbg_grc_init_params(struct qed_hwfn *p_hwfn)
 	}
 }
 
+/* Sets pointer and size for the specified binary buffer type */
+static void qed_set_dbg_bin_buf(struct qed_hwfn *p_hwfn,
+				enum bin_dbg_buffer_type buf_type,
+				const u32 *ptr, u32 size)
+{
+	struct virt_mem_desc *buf = &p_hwfn->dbg_arrays[buf_type];
+
+	buf->ptr = (void *)ptr;
+	buf->size = size;
+}
+
 /* Initializes debug data for the specified device */
-static enum dbg_status qed_dbg_dev_init(struct qed_hwfn *p_hwfn,
-					struct qed_ptt *p_ptt)
+static enum dbg_status qed_dbg_dev_init(struct qed_hwfn *p_hwfn)
 {
 	struct dbg_tools_data *dev_data = &p_hwfn->dbg_info;
 	u8 num_pfs = 0, max_pfs_per_port = 0;
@@ -1831,26 +886,25 @@ static enum dbg_status qed_dbg_dev_init(struct qed_hwfn *p_hwfn,
 		return DBG_STATUS_UNKNOWN_CHIP;
 	}
 
-	/* Set platofrm */
-	dev_data->platform_id = PLATFORM_ASIC;
+	/* Set HW type */
+	dev_data->hw_type = HW_TYPE_ASIC;
 	dev_data->mode_enable[MODE_ASIC] = 1;
 
 	/* Set port mode */
-	switch (qed_rd(p_hwfn, p_ptt, MISC_REG_PORT_MODE)) {
-	case 0:
+	switch (p_hwfn->cdev->num_ports_in_engine) {
+	case 1:
 		dev_data->mode_enable[MODE_PORTS_PER_ENG_1] = 1;
 		break;
-	case 1:
+	case 2:
 		dev_data->mode_enable[MODE_PORTS_PER_ENG_2] = 1;
 		break;
-	case 2:
+	case 4:
 		dev_data->mode_enable[MODE_PORTS_PER_ENG_4] = 1;
 		break;
 	}
 
 	/* Set 100G mode */
-	if (dev_data->chip_id == CHIP_BB &&
-	    qed_rd(p_hwfn, p_ptt, CNIG_REG_NW_PORT_MODE_BB) == 2)
+	if (QED_IS_CMT(p_hwfn->cdev))
 		dev_data->mode_enable[MODE_100G] = 1;
 
 	/* Set number of ports */
@@ -1876,14 +930,36 @@ static enum dbg_status qed_dbg_dev_init(struct qed_hwfn *p_hwfn,
 	return DBG_STATUS_OK;
 }
 
-static struct dbg_bus_block *get_dbg_bus_block_desc(struct qed_hwfn *p_hwfn,
-						    enum block_id block_id)
+static const struct dbg_block *get_dbg_block(struct qed_hwfn *p_hwfn,
+					     enum block_id block_id)
+{
+	const struct dbg_block *dbg_block;
+
+	dbg_block = p_hwfn->dbg_arrays[BIN_BUF_DBG_BLOCKS].ptr;
+	return dbg_block + block_id;
+}
+
+static const struct dbg_block_chip *qed_get_dbg_block_per_chip(struct qed_hwfn
+							       *p_hwfn,
+							       enum block_id
+							       block_id)
+{
+	struct dbg_tools_data *dev_data = &p_hwfn->dbg_info;
+
+	return (const struct dbg_block_chip *)
+	    p_hwfn->dbg_arrays[BIN_BUF_DBG_BLOCKS_CHIP_DATA].ptr +
+	    block_id * MAX_CHIP_IDS + dev_data->chip_id;
+}
+
+static const struct dbg_reset_reg *qed_get_dbg_reset_reg(struct qed_hwfn
+							 *p_hwfn,
+							 u8 reset_reg_id)
 {
 	struct dbg_tools_data *dev_data = &p_hwfn->dbg_info;
 
-	return (struct dbg_bus_block *)&dbg_bus_blocks[block_id *
-						       MAX_CHIP_IDS +
-						       dev_data->chip_id];
+	return (const struct dbg_reset_reg *)
+	    p_hwfn->dbg_arrays[BIN_BUF_DBG_RESET_REGS].ptr +
+	    reset_reg_id * MAX_CHIP_IDS + dev_data->chip_id;
 }
 
 /* Reads the FW info structure for the specified Storm from the chip,
@@ -1904,8 +980,9 @@ static void qed_read_storm_fw_info(struct qed_hwfn *p_hwfn,
 	 * The address is located in the last line of the Storm RAM.
 	 */
 	addr = storm->sem_fast_mem_addr + SEM_FAST_REG_INT_RAM +
-	       DWORDS_TO_BYTES(SEM_FAST_REG_INT_RAM_SIZE_BB_K2) -
-	       sizeof(fw_info_location);
+	    DWORDS_TO_BYTES(SEM_FAST_REG_INT_RAM_SIZE) -
+	    sizeof(fw_info_location);
+
 	dest = (u32 *)&fw_info_location;
 
 	for (i = 0; i < BYTES_TO_DWORDS(sizeof(fw_info_location));
@@ -2100,6 +1177,29 @@ static u32 qed_dump_mfw_ver_param(struct qed_hwfn *p_hwfn,
 	return qed_dump_str_param(dump_buf, dump, "mfw-version", mfw_ver_str);
 }
 
+/* Reads the chip revision from the chip and writes it as a param to the
+ * specified buffer. Returns the dumped size in dwords.
+ */
+static u32 qed_dump_chip_revision_param(struct qed_hwfn *p_hwfn,
+					struct qed_ptt *p_ptt,
+					u32 *dump_buf, bool dump)
+{
+	struct dbg_tools_data *dev_data = &p_hwfn->dbg_info;
+	char param_str[3] = "??";
+
+	if (dev_data->hw_type == HW_TYPE_ASIC) {
+		u32 chip_rev, chip_metal;
+
+		chip_rev = qed_rd(p_hwfn, p_ptt, MISCS_REG_CHIP_REV);
+		chip_metal = qed_rd(p_hwfn, p_ptt, MISCS_REG_CHIP_METAL);
+
+		param_str[0] = 'a' + (u8)chip_rev;
+		param_str[1] = '0' + (u8)chip_metal;
+	}
+
+	return qed_dump_str_param(dump_buf, dump, "chip-revision", param_str);
+}
+
 /* Writes a section header to the specified buffer.
  * Returns the dumped size in dwords.
  */
@@ -2123,7 +1223,8 @@ static u32 qed_dump_common_global_params(struct qed_hwfn *p_hwfn,
 	u8 num_params;
 
 	/* Dump global params section header */
-	num_params = NUM_COMMON_GLOBAL_PARAMS + num_specific_global_params;
+	num_params = NUM_COMMON_GLOBAL_PARAMS + num_specific_global_params +
+		(dev_data->chip_id == CHIP_BB ? 1 : 0);
 	offset += qed_dump_section_hdr(dump_buf + offset,
 				       dump, "global_params", num_params);
 
@@ -2131,6 +1232,8 @@ static u32 qed_dump_common_global_params(struct qed_hwfn *p_hwfn,
 	offset += qed_dump_fw_ver_param(p_hwfn, p_ptt, dump_buf + offset, dump);
 	offset += qed_dump_mfw_ver_param(p_hwfn,
 					 p_ptt, dump_buf + offset, dump);
+	offset += qed_dump_chip_revision_param(p_hwfn,
+					       p_ptt, dump_buf + offset, dump);
 	offset += qed_dump_num_param(dump_buf + offset,
 				     dump, "tools-version", TOOLS_VERSION);
 	offset += qed_dump_str_param(dump_buf + offset,
@@ -2140,11 +1243,12 @@ static u32 qed_dump_common_global_params(struct qed_hwfn *p_hwfn,
 	offset += qed_dump_str_param(dump_buf + offset,
 				     dump,
 				     "platform",
-				     s_platform_defs[dev_data->platform_id].
-				     name);
-	offset +=
-	    qed_dump_num_param(dump_buf + offset, dump, "pci-func",
-			       p_hwfn->abs_pf_id);
+				     s_hw_type_defs[dev_data->hw_type].name);
+	offset += qed_dump_num_param(dump_buf + offset,
+				     dump, "pci-func", p_hwfn->abs_pf_id);
+	if (dev_data->chip_id == CHIP_BB)
+		offset += qed_dump_num_param(dump_buf + offset,
+					     dump, "path", QED_PATH_ID(p_hwfn));
 
 	return offset;
 }
@@ -2175,24 +1279,87 @@ static void qed_update_blocks_reset_state(struct qed_hwfn *p_hwfn,
 					  struct qed_ptt *p_ptt)
 {
 	struct dbg_tools_data *dev_data = &p_hwfn->dbg_info;
-	u32 reg_val[MAX_DBG_RESET_REGS] = { 0 };
-	u32 i;
+	u32 reg_val[NUM_DBG_RESET_REGS] = { 0 };
+	u8 rst_reg_id;
+	u32 blk_id;
 
 	/* Read reset registers */
-	for (i = 0; i < MAX_DBG_RESET_REGS; i++)
-		if (s_reset_regs_defs[i].exists[dev_data->chip_id])
-			reg_val[i] = qed_rd(p_hwfn,
-					    p_ptt, s_reset_regs_defs[i].addr);
+	for (rst_reg_id = 0; rst_reg_id < NUM_DBG_RESET_REGS; rst_reg_id++) {
+		const struct dbg_reset_reg *rst_reg;
+		bool rst_reg_removed;
+		u32 rst_reg_addr;
+
+		rst_reg = qed_get_dbg_reset_reg(p_hwfn, rst_reg_id);
+		rst_reg_removed = GET_FIELD(rst_reg->data,
+					    DBG_RESET_REG_IS_REMOVED);
+		rst_reg_addr = DWORDS_TO_BYTES(GET_FIELD(rst_reg->data,
+							 DBG_RESET_REG_ADDR));
+
+		if (!rst_reg_removed)
+			reg_val[rst_reg_id] = qed_rd(p_hwfn, p_ptt,
+						     rst_reg_addr);
+	}
 
 	/* Check if blocks are in reset */
-	for (i = 0; i < MAX_BLOCK_ID; i++) {
-		struct block_defs *block = s_block_defs[i];
+	for (blk_id = 0; blk_id < NUM_PHYS_BLOCKS; blk_id++) {
+		const struct dbg_block_chip *blk;
+		bool has_rst_reg;
+		bool is_removed;
+
+		blk = qed_get_dbg_block_per_chip(p_hwfn, (enum block_id)blk_id);
+		is_removed = GET_FIELD(blk->flags, DBG_BLOCK_CHIP_IS_REMOVED);
+		has_rst_reg = GET_FIELD(blk->flags,
+					DBG_BLOCK_CHIP_HAS_RESET_REG);
+
+		if (!is_removed && has_rst_reg)
+			dev_data->block_in_reset[blk_id] =
+			    !(reg_val[blk->reset_reg_id] &
+			      BIT(blk->reset_reg_bit_offset));
+	}
+}
+
+/* is_mode_match recursive function */
+static bool qed_is_mode_match_rec(struct qed_hwfn *p_hwfn,
+				  u16 *modes_buf_offset, u8 rec_depth)
+{
+	struct dbg_tools_data *dev_data = &p_hwfn->dbg_info;
+	u8 *dbg_array;
+	bool arg1, arg2;
+	u8 tree_val;
+
+	if (rec_depth > MAX_RECURSION_DEPTH) {
+		DP_NOTICE(p_hwfn,
+			  "Unexpected error: is_mode_match_rec exceeded the max recursion depth. This is probably due to a corrupt init/debug buffer.\n");
+		return false;
+	}
+
+	/* Get next element from modes tree buffer */
+	dbg_array = p_hwfn->dbg_arrays[BIN_BUF_DBG_MODE_TREE].ptr;
+	tree_val = dbg_array[(*modes_buf_offset)++];
 
-		dev_data->block_in_reset[i] = block->has_reset_bit &&
-		    !(reg_val[block->reset_reg] & BIT(block->reset_bit_offset));
+	switch (tree_val) {
+	case INIT_MODE_OP_NOT:
+		return !qed_is_mode_match_rec(p_hwfn,
+					      modes_buf_offset, rec_depth + 1);
+	case INIT_MODE_OP_OR:
+	case INIT_MODE_OP_AND:
+		arg1 = qed_is_mode_match_rec(p_hwfn,
+					     modes_buf_offset, rec_depth + 1);
+		arg2 = qed_is_mode_match_rec(p_hwfn,
+					     modes_buf_offset, rec_depth + 1);
+		return (tree_val == INIT_MODE_OP_OR) ? (arg1 ||
+							arg2) : (arg1 && arg2);
+	default:
+		return dev_data->mode_enable[tree_val - MAX_INIT_MODE_OPS] > 0;
 	}
 }
 
+/* Returns true if the mode (specified using modes_buf_offset) is enabled */
+static bool qed_is_mode_match(struct qed_hwfn *p_hwfn, u16 *modes_buf_offset)
+{
+	return qed_is_mode_match_rec(p_hwfn, modes_buf_offset, 0);
+}
+
 /* Enable / disable the Debug block */
 static void qed_bus_enable_dbg_block(struct qed_hwfn *p_hwfn,
 				     struct qed_ptt *p_ptt, bool enable)
@@ -2204,23 +1371,21 @@ static void qed_bus_enable_dbg_block(struct qed_hwfn *p_hwfn,
 static void qed_bus_reset_dbg_block(struct qed_hwfn *p_hwfn,
 				    struct qed_ptt *p_ptt)
 {
-	u32 dbg_reset_reg_addr, old_reset_reg_val, new_reset_reg_val;
-	struct block_defs *dbg_block = s_block_defs[BLOCK_DBG];
+	u32 reset_reg_addr, old_reset_reg_val, new_reset_reg_val;
+	const struct dbg_reset_reg *reset_reg;
+	const struct dbg_block_chip *block;
 
-	dbg_reset_reg_addr = s_reset_regs_defs[dbg_block->reset_reg].addr;
-	old_reset_reg_val = qed_rd(p_hwfn, p_ptt, dbg_reset_reg_addr);
-	new_reset_reg_val =
-	    old_reset_reg_val & ~BIT(dbg_block->reset_bit_offset);
+	block = qed_get_dbg_block_per_chip(p_hwfn, BLOCK_DBG);
+	reset_reg = qed_get_dbg_reset_reg(p_hwfn, block->reset_reg_id);
+	reset_reg_addr =
+	    DWORDS_TO_BYTES(GET_FIELD(reset_reg->data, DBG_RESET_REG_ADDR));
 
-	qed_wr(p_hwfn, p_ptt, dbg_reset_reg_addr, new_reset_reg_val);
-	qed_wr(p_hwfn, p_ptt, dbg_reset_reg_addr, old_reset_reg_val);
-}
+	old_reset_reg_val = qed_rd(p_hwfn, p_ptt, reset_reg_addr);
+	new_reset_reg_val =
+	    old_reset_reg_val & ~BIT(block->reset_reg_bit_offset);
 
-static void qed_bus_set_framing_mode(struct qed_hwfn *p_hwfn,
-				     struct qed_ptt *p_ptt,
-				     enum dbg_bus_frame_modes mode)
-{
-	qed_wr(p_hwfn, p_ptt, DBG_REG_FRAMING_MODE, (u8)mode);
+	qed_wr(p_hwfn, p_ptt, reset_reg_addr, new_reset_reg_val);
+	qed_wr(p_hwfn, p_ptt, reset_reg_addr, old_reset_reg_val);
 }
 
 /* Enable / disable Debug Bus clients according to the specified mask
@@ -2232,28 +1397,65 @@ static void qed_bus_enable_clients(struct qed_hwfn *p_hwfn,
 	qed_wr(p_hwfn, p_ptt, DBG_REG_CLIENT_ENABLE, client_mask);
 }
 
-static bool qed_is_mode_match(struct qed_hwfn *p_hwfn, u16 *modes_buf_offset)
+static void qed_bus_config_dbg_line(struct qed_hwfn *p_hwfn,
+				    struct qed_ptt *p_ptt,
+				    enum block_id block_id,
+				    u8 line_id,
+				    u8 enable_mask,
+				    u8 right_shift,
+				    u8 force_valid_mask, u8 force_frame_mask)
+{
+	const struct dbg_block_chip *block =
+		qed_get_dbg_block_per_chip(p_hwfn, block_id);
+
+	qed_wr(p_hwfn, p_ptt, DWORDS_TO_BYTES(block->dbg_select_reg_addr),
+	       line_id);
+	qed_wr(p_hwfn, p_ptt, DWORDS_TO_BYTES(block->dbg_dword_enable_reg_addr),
+	       enable_mask);
+	qed_wr(p_hwfn, p_ptt, DWORDS_TO_BYTES(block->dbg_shift_reg_addr),
+	       right_shift);
+	qed_wr(p_hwfn, p_ptt, DWORDS_TO_BYTES(block->dbg_force_valid_reg_addr),
+	       force_valid_mask);
+	qed_wr(p_hwfn, p_ptt, DWORDS_TO_BYTES(block->dbg_force_frame_reg_addr),
+	       force_frame_mask);
+}
+
+/* Disable debug bus in all blocks */
+static void qed_bus_disable_blocks(struct qed_hwfn *p_hwfn,
+				   struct qed_ptt *p_ptt)
 {
 	struct dbg_tools_data *dev_data = &p_hwfn->dbg_info;
-	bool arg1, arg2;
-	const u32 *ptr;
-	u8 tree_val;
+	u32 block_id;
 
-	/* Get next element from modes tree buffer */
-	ptr = s_dbg_arrays[BIN_BUF_DBG_MODE_TREE].ptr;
-	tree_val = ((u8 *)ptr)[(*modes_buf_offset)++];
+	/* Disable all blocks */
+	for (block_id = 0; block_id < MAX_BLOCK_ID; block_id++) {
+		const struct dbg_block_chip *block_per_chip =
+		    qed_get_dbg_block_per_chip(p_hwfn,
+					       (enum block_id)block_id);
 
-	switch (tree_val) {
-	case INIT_MODE_OP_NOT:
-		return !qed_is_mode_match(p_hwfn, modes_buf_offset);
-	case INIT_MODE_OP_OR:
-	case INIT_MODE_OP_AND:
-		arg1 = qed_is_mode_match(p_hwfn, modes_buf_offset);
-		arg2 = qed_is_mode_match(p_hwfn, modes_buf_offset);
-		return (tree_val == INIT_MODE_OP_OR) ? (arg1 ||
-							arg2) : (arg1 && arg2);
-	default:
-		return dev_data->mode_enable[tree_val - MAX_INIT_MODE_OPS] > 0;
+		if (GET_FIELD(block_per_chip->flags,
+			      DBG_BLOCK_CHIP_IS_REMOVED) ||
+		    dev_data->block_in_reset[block_id])
+			continue;
+
+		/* Disable debug bus */
+		if (GET_FIELD(block_per_chip->flags,
+			      DBG_BLOCK_CHIP_HAS_DBG_BUS)) {
+			u32 dbg_en_addr =
+				block_per_chip->dbg_dword_enable_reg_addr;
+			u16 modes_buf_offset =
+			    GET_FIELD(block_per_chip->dbg_bus_mode.data,
+				      DBG_MODE_HDR_MODES_BUF_OFFSET);
+			bool eval_mode =
+			    GET_FIELD(block_per_chip->dbg_bus_mode.data,
+				      DBG_MODE_HDR_EVAL_MODE) > 0;
+
+			if (!eval_mode ||
+			    qed_is_mode_match(p_hwfn, &modes_buf_offset))
+				qed_wr(p_hwfn, p_ptt,
+				       DWORDS_TO_BYTES(dbg_en_addr),
+				       0);
+		}
 	}
 }
 
@@ -2266,6 +1468,20 @@ static bool qed_grc_is_included(struct qed_hwfn *p_hwfn,
 	return qed_grc_get_param(p_hwfn, grc_param) > 0;
 }
 
+/* Returns the storm_id that matches the specified Storm letter,
+ * or MAX_DBG_STORMS if invalid storm letter.
+ */
+static enum dbg_storms qed_get_id_from_letter(char storm_letter)
+{
+	u8 storm_id;
+
+	for (storm_id = 0; storm_id < MAX_DBG_STORMS; storm_id++)
+		if (s_storm_defs[storm_id].letter == storm_letter)
+			return (enum dbg_storms)storm_id;
+
+	return MAX_DBG_STORMS;
+}
+
 /* Returns true of the specified Storm should be included in the dump, false
  * otherwise.
  */
@@ -2281,14 +1497,20 @@ static bool qed_grc_is_storm_included(struct qed_hwfn *p_hwfn,
 static bool qed_grc_is_mem_included(struct qed_hwfn *p_hwfn,
 				    enum block_id block_id, u8 mem_group_id)
 {
-	struct block_defs *block = s_block_defs[block_id];
+	const struct dbg_block *block;
 	u8 i;
 
-	/* Check Storm match */
-	if (block->associated_to_storm &&
-	    !qed_grc_is_storm_included(p_hwfn,
-				       (enum dbg_storms)block->storm_id))
-		return false;
+	block = get_dbg_block(p_hwfn, block_id);
+
+	/* If the block is associated with a Storm, check Storm match */
+	if (block->associated_storm_letter) {
+		enum dbg_storms associated_storm_id =
+		    qed_get_id_from_letter(block->associated_storm_letter);
+
+		if (associated_storm_id == MAX_DBG_STORMS ||
+		    !qed_grc_is_storm_included(p_hwfn, associated_storm_id))
+			return false;
+	}
 
 	for (i = 0; i < NUM_BIG_RAM_TYPES; i++) {
 		struct big_ram_defs *big_ram = &s_big_ram_defs[i];
@@ -2310,6 +1532,8 @@ static bool qed_grc_is_mem_included(struct qed_hwfn *p_hwfn,
 	case MEM_GROUP_CAU_SB:
 	case MEM_GROUP_CAU_PI:
 		return qed_grc_is_included(p_hwfn, DBG_GRC_PARAM_DUMP_CAU);
+	case MEM_GROUP_CAU_MEM_EXT:
+		return qed_grc_is_included(p_hwfn, DBG_GRC_PARAM_DUMP_CAU_EXT);
 	case MEM_GROUP_QM_MEM:
 		return qed_grc_is_included(p_hwfn, DBG_GRC_PARAM_DUMP_QM);
 	case MEM_GROUP_CFC_MEM:
@@ -2317,6 +1541,8 @@ static bool qed_grc_is_mem_included(struct qed_hwfn *p_hwfn,
 	case MEM_GROUP_TASK_CFC_MEM:
 		return qed_grc_is_included(p_hwfn, DBG_GRC_PARAM_DUMP_CFC) ||
 		       qed_grc_is_included(p_hwfn, DBG_GRC_PARAM_DUMP_CM_CTX);
+	case MEM_GROUP_DORQ_MEM:
+		return qed_grc_is_included(p_hwfn, DBG_GRC_PARAM_DUMP_DORQ);
 	case MEM_GROUP_IGU_MEM:
 	case MEM_GROUP_IGU_MSIX:
 		return qed_grc_is_included(p_hwfn, DBG_GRC_PARAM_DUMP_IGU);
@@ -2362,64 +1588,104 @@ static void qed_grc_stall_storms(struct qed_hwfn *p_hwfn,
 	msleep(STALL_DELAY_MS);
 }
 
-/* Takes all blocks out of reset */
+/* Takes all blocks out of reset. If rbc_only is true, only RBC clients are
+ * taken out of reset.
+ */
 static void qed_grc_unreset_blocks(struct qed_hwfn *p_hwfn,
-				   struct qed_ptt *p_ptt)
+				   struct qed_ptt *p_ptt, bool rbc_only)
 {
 	struct dbg_tools_data *dev_data = &p_hwfn->dbg_info;
-	u32 reg_val[MAX_DBG_RESET_REGS] = { 0 };
-	u32 block_id, i;
+	u8 chip_id = dev_data->chip_id;
+	u32 i;
 
-	/* Fill reset regs values */
-	for (block_id = 0; block_id < MAX_BLOCK_ID; block_id++) {
-		struct block_defs *block = s_block_defs[block_id];
+	/* Take RBCs out of reset */
+	for (i = 0; i < ARRAY_SIZE(s_rbc_reset_defs); i++)
+		if (s_rbc_reset_defs[i].reset_val[dev_data->chip_id])
+			qed_wr(p_hwfn,
+			       p_ptt,
+			       s_rbc_reset_defs[i].reset_reg_addr +
+			       RESET_REG_UNRESET_OFFSET,
+			       s_rbc_reset_defs[i].reset_val[chip_id]);
 
-		if (block->exists[dev_data->chip_id] && block->has_reset_bit &&
-		    block->unreset)
-			reg_val[block->reset_reg] |=
-			    BIT(block->reset_bit_offset);
-	}
+	if (!rbc_only) {
+		u32 reg_val[NUM_DBG_RESET_REGS] = { 0 };
+		u8 reset_reg_id;
+		u32 block_id;
 
-	/* Write reset registers */
-	for (i = 0; i < MAX_DBG_RESET_REGS; i++) {
-		if (!s_reset_regs_defs[i].exists[dev_data->chip_id])
-			continue;
+		/* Fill reset regs values */
+		for (block_id = 0; block_id < NUM_PHYS_BLOCKS; block_id++) {
+			bool is_removed, has_reset_reg, unreset_before_dump;
+			const struct dbg_block_chip *block;
+
+			block = qed_get_dbg_block_per_chip(p_hwfn,
+							   (enum block_id)
+							   block_id);
+			is_removed =
+			    GET_FIELD(block->flags, DBG_BLOCK_CHIP_IS_REMOVED);
+			has_reset_reg =
+			    GET_FIELD(block->flags,
+				      DBG_BLOCK_CHIP_HAS_RESET_REG);
+			unreset_before_dump =
+			    GET_FIELD(block->flags,
+				      DBG_BLOCK_CHIP_UNRESET_BEFORE_DUMP);
+
+			if (!is_removed && has_reset_reg && unreset_before_dump)
+				reg_val[block->reset_reg_id] |=
+				    BIT(block->reset_reg_bit_offset);
+		}
 
-		reg_val[i] |=
-			s_reset_regs_defs[i].unreset_val[dev_data->chip_id];
+		/* Write reset registers */
+		for (reset_reg_id = 0; reset_reg_id < NUM_DBG_RESET_REGS;
+		     reset_reg_id++) {
+			const struct dbg_reset_reg *reset_reg;
+			u32 reset_reg_addr;
 
-		if (reg_val[i])
-			qed_wr(p_hwfn,
-			       p_ptt,
-			       s_reset_regs_defs[i].addr +
-			       RESET_REG_UNRESET_OFFSET, reg_val[i]);
+			reset_reg = qed_get_dbg_reset_reg(p_hwfn, reset_reg_id);
+
+			if (GET_FIELD
+			    (reset_reg->data, DBG_RESET_REG_IS_REMOVED))
+				continue;
+
+			if (reg_val[reset_reg_id]) {
+				reset_reg_addr =
+				    GET_FIELD(reset_reg->data,
+					      DBG_RESET_REG_ADDR);
+				qed_wr(p_hwfn,
+				       p_ptt,
+				       DWORDS_TO_BYTES(reset_reg_addr) +
+				       RESET_REG_UNRESET_OFFSET,
+				       reg_val[reset_reg_id]);
+			}
+		}
 	}
 }
 
 /* Returns the attention block data of the specified block */
 static const struct dbg_attn_block_type_data *
-qed_get_block_attn_data(enum block_id block_id, enum dbg_attn_type attn_type)
+qed_get_block_attn_data(struct qed_hwfn *p_hwfn,
+			enum block_id block_id, enum dbg_attn_type attn_type)
 {
 	const struct dbg_attn_block *base_attn_block_arr =
-		(const struct dbg_attn_block *)
-		s_dbg_arrays[BIN_BUF_DBG_ATTN_BLOCKS].ptr;
+	    (const struct dbg_attn_block *)
+	    p_hwfn->dbg_arrays[BIN_BUF_DBG_ATTN_BLOCKS].ptr;
 
 	return &base_attn_block_arr[block_id].per_type_data[attn_type];
 }
 
 /* Returns the attention registers of the specified block */
 static const struct dbg_attn_reg *
-qed_get_block_attn_regs(enum block_id block_id, enum dbg_attn_type attn_type,
+qed_get_block_attn_regs(struct qed_hwfn *p_hwfn,
+			enum block_id block_id, enum dbg_attn_type attn_type,
 			u8 *num_attn_regs)
 {
 	const struct dbg_attn_block_type_data *block_type_data =
-		qed_get_block_attn_data(block_id, attn_type);
+	    qed_get_block_attn_data(p_hwfn, block_id, attn_type);
 
 	*num_attn_regs = block_type_data->num_regs;
 
-	return &((const struct dbg_attn_reg *)
-		 s_dbg_arrays[BIN_BUF_DBG_ATTN_REGS].ptr)[block_type_data->
-							  regs_offset];
+	return (const struct dbg_attn_reg *)
+		p_hwfn->dbg_arrays[BIN_BUF_DBG_ATTN_REGS].ptr +
+		block_type_data->regs_offset;
 }
 
 /* For each block, clear the status of all parities */
@@ -2431,11 +1697,12 @@ static void qed_grc_clear_all_prty(struct qed_hwfn *p_hwfn,
 	u8 reg_idx, num_attn_regs;
 	u32 block_id;
 
-	for (block_id = 0; block_id < MAX_BLOCK_ID; block_id++) {
+	for (block_id = 0; block_id < NUM_PHYS_BLOCKS; block_id++) {
 		if (dev_data->block_in_reset[block_id])
 			continue;
 
-		attn_reg_arr = qed_get_block_attn_regs((enum block_id)block_id,
+		attn_reg_arr = qed_get_block_attn_regs(p_hwfn,
+						       (enum block_id)block_id,
 						       ATTN_TYPE_PARITY,
 						       &num_attn_regs);
 
@@ -2463,22 +1730,20 @@ static void qed_grc_clear_all_prty(struct qed_hwfn *p_hwfn,
 }
 
 /* Dumps GRC registers section header. Returns the dumped size in dwords.
- * The following parameters are dumped:
+ * the following parameters are dumped:
  * - count: no. of dumped entries
  * - split_type: split type
  * - split_id: split ID (dumped only if split_id != SPLIT_TYPE_NONE)
- * - param_name: user parameter value (dumped only if param_name != NULL
- *		 and param_val != NULL).
+ * - reg_type_name: register type name (dumped only if reg_type_name != NULL)
  */
 static u32 qed_grc_dump_regs_hdr(u32 *dump_buf,
 				 bool dump,
 				 u32 num_reg_entries,
 				 enum init_split_types split_type,
-				 u8 split_id,
-				 const char *param_name, const char *param_val)
+				 u8 split_id, const char *reg_type_name)
 {
 	u8 num_params = 2 +
-	    (split_type != SPLIT_TYPE_NONE ? 1 : 0) + (param_name ? 1 : 0);
+	    (split_type != SPLIT_TYPE_NONE ? 1 : 0) + (reg_type_name ? 1 : 0);
 	u32 offset = 0;
 
 	offset += qed_dump_section_hdr(dump_buf + offset,
@@ -2491,9 +1756,9 @@ static u32 qed_grc_dump_regs_hdr(u32 *dump_buf,
 	if (split_type != SPLIT_TYPE_NONE)
 		offset += qed_dump_num_param(dump_buf + offset,
 					     dump, "id", split_id);
-	if (param_name && param_val)
+	if (reg_type_name)
 		offset += qed_dump_str_param(dump_buf + offset,
-					     dump, param_name, param_val);
+					     dump, "type", reg_type_name);
 
 	return offset;
 }
@@ -2523,20 +1788,11 @@ static u32 qed_grc_dump_addr_range(struct qed_hwfn *p_hwfn,
 {
 	struct dbg_tools_data *dev_data = &p_hwfn->dbg_info;
 	u8 port_id = 0, pf_id = 0, vf_id = 0, fid = 0;
+	bool read_using_dmae = false;
+	u32 thresh;
 
-	if (!dump)
-		return len;
-
-	/* Print log if needed */
-	dev_data->num_regs_read += len;
-	if (dev_data->num_regs_read >=
-	    s_platform_defs[dev_data->platform_id].log_thresh) {
-		DP_VERBOSE(p_hwfn,
-			   QED_MSG_DEBUG,
-			   "Dumping %d registers...\n",
-			   dev_data->num_regs_read);
-		dev_data->num_regs_read = 0;
-	}
+	if (!dump)
+		return len;
 
 	switch (split_type) {
 	case SPLIT_TYPE_PORT:
@@ -2558,38 +1814,77 @@ static u32 qed_grc_dump_addr_range(struct qed_hwfn *p_hwfn,
 	}
 
 	/* Try reading using DMAE */
-	if (dev_data->use_dmae && split_type == SPLIT_TYPE_NONE &&
-	    (len >= s_platform_defs[dev_data->platform_id].dmae_thresh ||
-	     wide_bus)) {
-		if (!qed_dmae_grc2host(p_hwfn, p_ptt, DWORDS_TO_BYTES(addr),
-				       (u64)(uintptr_t)(dump_buf), len, NULL))
-			return len;
-		dev_data->use_dmae = 0;
-		DP_VERBOSE(p_hwfn,
-			   QED_MSG_DEBUG,
-			   "Failed reading from chip using DMAE, using GRC instead\n");
+	if (dev_data->use_dmae && split_type != SPLIT_TYPE_VF &&
+	    (len >= s_hw_type_defs[dev_data->hw_type].dmae_thresh ||
+	     (PROTECT_WIDE_BUS && wide_bus))) {
+		struct qed_dmae_params dmae_params;
+
+		/* Set DMAE params */
+		memset(&dmae_params, 0, sizeof(dmae_params));
+		SET_FIELD(dmae_params.flags, QED_DMAE_PARAMS_COMPLETION_DST, 1);
+		switch (split_type) {
+		case SPLIT_TYPE_PORT:
+			SET_FIELD(dmae_params.flags, QED_DMAE_PARAMS_PORT_VALID,
+				  1);
+			dmae_params.port_id = port_id;
+			break;
+		case SPLIT_TYPE_PF:
+			SET_FIELD(dmae_params.flags,
+				  QED_DMAE_PARAMS_SRC_PF_VALID, 1);
+			dmae_params.src_pfid = pf_id;
+			break;
+		case SPLIT_TYPE_PORT_PF:
+			SET_FIELD(dmae_params.flags, QED_DMAE_PARAMS_PORT_VALID,
+				  1);
+			SET_FIELD(dmae_params.flags,
+				  QED_DMAE_PARAMS_SRC_PF_VALID, 1);
+			dmae_params.port_id = port_id;
+			dmae_params.src_pfid = pf_id;
+			break;
+		default:
+			break;
+		}
+
+		/* Execute DMAE command */
+		read_using_dmae = !qed_dmae_grc2host(p_hwfn,
+						     p_ptt,
+						     DWORDS_TO_BYTES(addr),
+						     (u64)(uintptr_t)(dump_buf),
+						     len, &dmae_params);
+		if (!read_using_dmae) {
+			dev_data->use_dmae = 0;
+			DP_VERBOSE(p_hwfn,
+				   QED_MSG_DEBUG,
+				   "Failed reading from chip using DMAE, using GRC instead\n");
+		}
 	}
 
+	if (read_using_dmae)
+		goto print_log;
+
 	/* If not read using DMAE, read using GRC */
 
 	/* Set pretend */
-	if (split_type != dev_data->pretend.split_type || split_id !=
-	    dev_data->pretend.split_id) {
+	if (split_type != dev_data->pretend.split_type ||
+	    split_id != dev_data->pretend.split_id) {
 		switch (split_type) {
 		case SPLIT_TYPE_PORT:
 			qed_port_pretend(p_hwfn, p_ptt, port_id);
 			break;
 		case SPLIT_TYPE_PF:
-			fid = pf_id << PXP_PRETEND_CONCRETE_FID_PFID_SHIFT;
+			fid = FIELD_VALUE(PXP_PRETEND_CONCRETE_FID_PFID,
+					  pf_id);
 			qed_fid_pretend(p_hwfn, p_ptt, fid);
 			break;
 		case SPLIT_TYPE_PORT_PF:
-			fid = pf_id << PXP_PRETEND_CONCRETE_FID_PFID_SHIFT;
+			fid = FIELD_VALUE(PXP_PRETEND_CONCRETE_FID_PFID,
+					  pf_id);
 			qed_port_fid_pretend(p_hwfn, p_ptt, port_id, fid);
 			break;
 		case SPLIT_TYPE_VF:
-			fid = BIT(PXP_PRETEND_CONCRETE_FID_VFVALID_SHIFT) |
-			      (vf_id << PXP_PRETEND_CONCRETE_FID_VFID_SHIFT);
+			fid = FIELD_VALUE(PXP_PRETEND_CONCRETE_FID_VFVALID, 1)
+			      | FIELD_VALUE(PXP_PRETEND_CONCRETE_FID_VFID,
+					  vf_id);
 			qed_fid_pretend(p_hwfn, p_ptt, fid);
 			break;
 		default:
@@ -2603,6 +1898,16 @@ static u32 qed_grc_dump_addr_range(struct qed_hwfn *p_hwfn,
 	/* Read registers using GRC */
 	qed_read_regs(p_hwfn, p_ptt, dump_buf, addr, len);
 
+print_log:
+	/* Print log */
+	dev_data->num_regs_read += len;
+	thresh = s_hw_type_defs[dev_data->hw_type].log_thresh;
+	if ((dev_data->num_regs_read / thresh) >
+	    ((dev_data->num_regs_read - len) / thresh))
+		DP_VERBOSE(p_hwfn,
+			   QED_MSG_DEBUG,
+			   "Dumped %d registers...\n", dev_data->num_regs_read);
+
 	return len;
 }
 
@@ -2687,7 +1992,7 @@ static u32 qed_grc_dump_reg_entry_skip(struct qed_hwfn *p_hwfn,
 /* Dumps GRC registers entries. Returns the dumped size in dwords. */
 static u32 qed_grc_dump_regs_entries(struct qed_hwfn *p_hwfn,
 				     struct qed_ptt *p_ptt,
-				     struct dbg_array input_regs_arr,
+				     struct virt_mem_desc input_regs_arr,
 				     u32 *dump_buf,
 				     bool dump,
 				     enum init_split_types split_type,
@@ -2700,10 +2005,10 @@ static u32 qed_grc_dump_regs_entries(struct qed_hwfn *p_hwfn,
 
 	*num_dumped_reg_entries = 0;
 
-	while (input_offset < input_regs_arr.size_in_dwords) {
+	while (input_offset < BYTES_TO_DWORDS(input_regs_arr.size)) {
 		const struct dbg_dump_cond_hdr *cond_hdr =
 		    (const struct dbg_dump_cond_hdr *)
-		    &input_regs_arr.ptr[input_offset++];
+		    input_regs_arr.ptr + input_offset++;
 		u16 modes_buf_offset;
 		bool eval_mode;
 
@@ -2726,7 +2031,7 @@ static u32 qed_grc_dump_regs_entries(struct qed_hwfn *p_hwfn,
 		for (i = 0; i < cond_hdr->data_size; i++, input_offset++) {
 			const struct dbg_dump_reg *reg =
 			    (const struct dbg_dump_reg *)
-			    &input_regs_arr.ptr[input_offset];
+			    input_regs_arr.ptr + input_offset;
 			u32 addr, len;
 			bool wide_bus;
 
@@ -2751,14 +2056,12 @@ static u32 qed_grc_dump_regs_entries(struct qed_hwfn *p_hwfn,
 /* Dumps GRC registers entries. Returns the dumped size in dwords. */
 static u32 qed_grc_dump_split_data(struct qed_hwfn *p_hwfn,
 				   struct qed_ptt *p_ptt,
-				   struct dbg_array input_regs_arr,
+				   struct virt_mem_desc input_regs_arr,
 				   u32 *dump_buf,
 				   bool dump,
 				   bool block_enable[MAX_BLOCK_ID],
 				   enum init_split_types split_type,
-				   u8 split_id,
-				   const char *param_name,
-				   const char *param_val)
+				   u8 split_id, const char *reg_type_name)
 {
 	struct dbg_tools_data *dev_data = &p_hwfn->dbg_info;
 	enum init_split_types hdr_split_type = split_type;
@@ -2776,7 +2079,7 @@ static u32 qed_grc_dump_split_data(struct qed_hwfn *p_hwfn,
 				       false,
 				       0,
 				       hdr_split_type,
-				       hdr_split_id, param_name, param_val);
+				       hdr_split_id, reg_type_name);
 
 	/* Dump registers */
 	offset += qed_grc_dump_regs_entries(p_hwfn,
@@ -2795,7 +2098,7 @@ static u32 qed_grc_dump_split_data(struct qed_hwfn *p_hwfn,
 				      dump,
 				      num_dumped_reg_entries,
 				      hdr_split_type,
-				      hdr_split_id, param_name, param_val);
+				      hdr_split_id, reg_type_name);
 
 	return num_dumped_reg_entries > 0 ? offset : 0;
 }
@@ -2808,32 +2111,33 @@ static u32 qed_grc_dump_registers(struct qed_hwfn *p_hwfn,
 				  u32 *dump_buf,
 				  bool dump,
 				  bool block_enable[MAX_BLOCK_ID],
-				  const char *param_name, const char *param_val)
+				  const char *reg_type_name)
 {
+	struct virt_mem_desc *dbg_buf =
+	    &p_hwfn->dbg_arrays[BIN_BUF_DBG_DUMP_REG];
 	struct dbg_tools_data *dev_data = &p_hwfn->dbg_info;
 	u32 offset = 0, input_offset = 0;
-	u16 fid;
-	while (input_offset <
-	       s_dbg_arrays[BIN_BUF_DBG_DUMP_REG].size_in_dwords) {
+
+	while (input_offset < BYTES_TO_DWORDS(dbg_buf->size)) {
 		const struct dbg_dump_split_hdr *split_hdr;
-		struct dbg_array curr_input_regs_arr;
+		struct virt_mem_desc curr_input_regs_arr;
 		enum init_split_types split_type;
 		u16 split_count = 0;
 		u32 split_data_size;
 		u8 split_id;
 
 		split_hdr =
-			(const struct dbg_dump_split_hdr *)
-			&s_dbg_arrays[BIN_BUF_DBG_DUMP_REG].ptr[input_offset++];
+		    (const struct dbg_dump_split_hdr *)
+		    dbg_buf->ptr + input_offset++;
 		split_type =
-			GET_FIELD(split_hdr->hdr,
-				  DBG_DUMP_SPLIT_HDR_SPLIT_TYPE_ID);
-		split_data_size =
-			GET_FIELD(split_hdr->hdr,
-				  DBG_DUMP_SPLIT_HDR_DATA_SIZE);
+		    GET_FIELD(split_hdr->hdr,
+			      DBG_DUMP_SPLIT_HDR_SPLIT_TYPE_ID);
+		split_data_size = GET_FIELD(split_hdr->hdr,
+					    DBG_DUMP_SPLIT_HDR_DATA_SIZE);
 		curr_input_regs_arr.ptr =
-			&s_dbg_arrays[BIN_BUF_DBG_DUMP_REG].ptr[input_offset];
-		curr_input_regs_arr.size_in_dwords = split_data_size;
+		    (u32 *)p_hwfn->dbg_arrays[BIN_BUF_DBG_DUMP_REG].ptr +
+		    input_offset;
+		curr_input_regs_arr.size = DWORDS_TO_BYTES(split_data_size);
 
 		switch (split_type) {
 		case SPLIT_TYPE_NONE:
@@ -2861,16 +2165,16 @@ static u32 qed_grc_dump_registers(struct qed_hwfn *p_hwfn,
 							  dump, block_enable,
 							  split_type,
 							  split_id,
-							  param_name,
-							  param_val);
+							  reg_type_name);
 
 		input_offset += split_data_size;
 	}
 
 	/* Cancel pretends (pretend to original PF) */
 	if (dump) {
-		fid = p_hwfn->rel_pf_id << PXP_PRETEND_CONCRETE_FID_PFID_SHIFT;
-		qed_fid_pretend(p_hwfn, p_ptt, fid);
+		qed_fid_pretend(p_hwfn, p_ptt,
+				FIELD_VALUE(PXP_PRETEND_CONCRETE_FID_PFID,
+					    p_hwfn->rel_pf_id));
 		dev_data->pretend.split_type = SPLIT_TYPE_NONE;
 		dev_data->pretend.split_id = 0;
 	}
@@ -2883,26 +2187,32 @@ static u32 qed_grc_dump_reset_regs(struct qed_hwfn *p_hwfn,
 				   struct qed_ptt *p_ptt,
 				   u32 *dump_buf, bool dump)
 {
-	struct dbg_tools_data *dev_data = &p_hwfn->dbg_info;
-	u32 i, offset = 0, num_regs = 0;
+	u32 offset = 0, num_regs = 0;
+	u8 reset_reg_id;
 
 	/* Calculate header size */
 	offset += qed_grc_dump_regs_hdr(dump_buf,
-					false, 0,
-					SPLIT_TYPE_NONE, 0, NULL, NULL);
+					false,
+					0, SPLIT_TYPE_NONE, 0, "RESET_REGS");
 
 	/* Write reset registers */
-	for (i = 0; i < MAX_DBG_RESET_REGS; i++) {
-		if (!s_reset_regs_defs[i].exists[dev_data->chip_id])
+	for (reset_reg_id = 0; reset_reg_id < NUM_DBG_RESET_REGS;
+	     reset_reg_id++) {
+		const struct dbg_reset_reg *reset_reg;
+		u32 reset_reg_addr;
+
+		reset_reg = qed_get_dbg_reset_reg(p_hwfn, reset_reg_id);
+
+		if (GET_FIELD(reset_reg->data, DBG_RESET_REG_IS_REMOVED))
 			continue;
 
+		reset_reg_addr = GET_FIELD(reset_reg->data, DBG_RESET_REG_ADDR);
 		offset += qed_grc_dump_reg_entry(p_hwfn,
 						 p_ptt,
 						 dump_buf + offset,
 						 dump,
-						 BYTES_TO_DWORDS
-						 (s_reset_regs_defs[i].addr), 1,
-						 false, SPLIT_TYPE_NONE, 0);
+						 reset_reg_addr,
+						 1, false, SPLIT_TYPE_NONE, 0);
 		num_regs++;
 	}
 
@@ -2910,7 +2220,7 @@ static u32 qed_grc_dump_reset_regs(struct qed_hwfn *p_hwfn,
 	if (dump)
 		qed_grc_dump_regs_hdr(dump_buf,
 				      true, num_regs, SPLIT_TYPE_NONE,
-				      0, NULL, NULL);
+				      0, "RESET_REGS");
 
 	return offset;
 }
@@ -2923,21 +2233,23 @@ static u32 qed_grc_dump_modified_regs(struct qed_hwfn *p_hwfn,
 				      u32 *dump_buf, bool dump)
 {
 	struct dbg_tools_data *dev_data = &p_hwfn->dbg_info;
-	u32 block_id, offset = 0, num_reg_entries = 0;
+	u32 block_id, offset = 0, stall_regs_offset;
 	const struct dbg_attn_reg *attn_reg_arr;
 	u8 storm_id, reg_idx, num_attn_regs;
+	u32 num_reg_entries = 0;
 
-	/* Calculate header size */
+	/* Write empty header for attention registers */
 	offset += qed_grc_dump_regs_hdr(dump_buf,
-					false, 0, SPLIT_TYPE_NONE,
-					0, NULL, NULL);
+					false,
+					0, SPLIT_TYPE_NONE, 0, "ATTN_REGS");
 
 	/* Write parity registers */
-	for (block_id = 0; block_id < MAX_BLOCK_ID; block_id++) {
+	for (block_id = 0; block_id < NUM_PHYS_BLOCKS; block_id++) {
 		if (dev_data->block_in_reset[block_id] && dump)
 			continue;
 
-		attn_reg_arr = qed_get_block_attn_regs((enum block_id)block_id,
+		attn_reg_arr = qed_get_block_attn_regs(p_hwfn,
+						       (enum block_id)block_id,
 						       ATTN_TYPE_PARITY,
 						       &num_attn_regs);
 
@@ -2980,16 +2292,29 @@ static u32 qed_grc_dump_modified_regs(struct qed_hwfn *p_hwfn,
 		}
 	}
 
+	/* Overwrite header for attention registers */
+	if (dump)
+		qed_grc_dump_regs_hdr(dump_buf,
+				      true,
+				      num_reg_entries,
+				      SPLIT_TYPE_NONE, 0, "ATTN_REGS");
+
+	/* Write empty header for stall registers */
+	stall_regs_offset = offset;
+	offset += qed_grc_dump_regs_hdr(dump_buf,
+					false, 0, SPLIT_TYPE_NONE, 0, "REGS");
+
 	/* Write Storm stall status registers */
-	for (storm_id = 0; storm_id < MAX_DBG_STORMS; storm_id++) {
+	for (storm_id = 0, num_reg_entries = 0; storm_id < MAX_DBG_STORMS;
+	     storm_id++) {
 		struct storm_defs *storm = &s_storm_defs[storm_id];
 		u32 addr;
 
-		if (dev_data->block_in_reset[storm->block_id] && dump)
+		if (dev_data->block_in_reset[storm->sem_block_id] && dump)
 			continue;
 
 		addr =
-		    BYTES_TO_DWORDS(s_storm_defs[storm_id].sem_fast_mem_addr +
+		    BYTES_TO_DWORDS(storm->sem_fast_mem_addr +
 				    SEM_FAST_REG_STALLED);
 		offset += qed_grc_dump_reg_entry(p_hwfn,
 						 p_ptt,
@@ -3001,12 +2326,12 @@ static u32 qed_grc_dump_modified_regs(struct qed_hwfn *p_hwfn,
 		num_reg_entries++;
 	}
 
-	/* Write header */
+	/* Overwrite header for stall registers */
 	if (dump)
-		qed_grc_dump_regs_hdr(dump_buf,
+		qed_grc_dump_regs_hdr(dump_buf + stall_regs_offset,
 				      true,
-				      num_reg_entries, SPLIT_TYPE_NONE,
-				      0, NULL, NULL);
+				      num_reg_entries,
+				      SPLIT_TYPE_NONE, 0, "REGS");
 
 	return offset;
 }
@@ -3019,8 +2344,7 @@ static u32 qed_grc_dump_special_regs(struct qed_hwfn *p_hwfn,
 	u32 offset = 0, addr;
 
 	offset += qed_grc_dump_regs_hdr(dump_buf,
-					dump, 2, SPLIT_TYPE_NONE, 0,
-					NULL, NULL);
+					dump, 2, SPLIT_TYPE_NONE, 0, "REGS");
 
 	/* Dump R/TDIF_REG_DEBUG_ERROR_INFO_SIZE (every 8'th register should be
 	 * skipped).
@@ -3068,8 +2392,7 @@ static u32 qed_grc_dump_mem_hdr(struct qed_hwfn *p_hwfn,
 				u32 len,
 				u32 bit_width,
 				bool packed,
-				const char *mem_group,
-				bool is_storm, char storm_letter)
+				const char *mem_group, char storm_letter)
 {
 	u8 num_params = 3;
 	u32 offset = 0;
@@ -3090,7 +2413,7 @@ static u32 qed_grc_dump_mem_hdr(struct qed_hwfn *p_hwfn,
 
 	if (name) {
 		/* Dump name */
-		if (is_storm) {
+		if (storm_letter) {
 			strcpy(buf, "?STORM_");
 			buf[0] = storm_letter;
 			strcpy(buf + strlen(buf), name);
@@ -3122,7 +2445,7 @@ static u32 qed_grc_dump_mem_hdr(struct qed_hwfn *p_hwfn,
 					     dump, "packed", 1);
 
 	/* Dump reg type */
-	if (is_storm) {
+	if (storm_letter) {
 		strcpy(buf, "?STORM_");
 		buf[0] = storm_letter;
 		strcpy(buf + strlen(buf), mem_group);
@@ -3149,8 +2472,7 @@ static u32 qed_grc_dump_mem(struct qed_hwfn *p_hwfn,
 			    bool wide_bus,
 			    u32 bit_width,
 			    bool packed,
-			    const char *mem_group,
-			    bool is_storm, char storm_letter)
+			    const char *mem_group, char storm_letter)
 {
 	u32 offset = 0;
 
@@ -3161,8 +2483,7 @@ static u32 qed_grc_dump_mem(struct qed_hwfn *p_hwfn,
 				       addr,
 				       len,
 				       bit_width,
-				       packed,
-				       mem_group, is_storm, storm_letter);
+				       packed, mem_group, storm_letter);
 	offset += qed_grc_dump_addr_range(p_hwfn,
 					  p_ptt,
 					  dump_buf + offset,
@@ -3175,20 +2496,21 @@ static u32 qed_grc_dump_mem(struct qed_hwfn *p_hwfn,
 /* Dumps GRC memories entries. Returns the dumped size in dwords. */
 static u32 qed_grc_dump_mem_entries(struct qed_hwfn *p_hwfn,
 				    struct qed_ptt *p_ptt,
-				    struct dbg_array input_mems_arr,
+				    struct virt_mem_desc input_mems_arr,
 				    u32 *dump_buf, bool dump)
 {
 	u32 i, offset = 0, input_offset = 0;
 	bool mode_match = true;
 
-	while (input_offset < input_mems_arr.size_in_dwords) {
+	while (input_offset < BYTES_TO_DWORDS(input_mems_arr.size)) {
 		const struct dbg_dump_cond_hdr *cond_hdr;
 		u16 modes_buf_offset;
 		u32 num_entries;
 		bool eval_mode;
 
-		cond_hdr = (const struct dbg_dump_cond_hdr *)
-			   &input_mems_arr.ptr[input_offset++];
+		cond_hdr =
+		    (const struct dbg_dump_cond_hdr *)input_mems_arr.ptr +
+		    input_offset++;
 		num_entries = cond_hdr->data_size / MEM_DUMP_ENTRY_SIZE_DWORDS;
 
 		/* Check required mode */
@@ -3210,24 +2532,25 @@ static u32 qed_grc_dump_mem_entries(struct qed_hwfn *p_hwfn,
 		for (i = 0; i < num_entries;
 		     i++, input_offset += MEM_DUMP_ENTRY_SIZE_DWORDS) {
 			const struct dbg_dump_mem *mem =
-				(const struct dbg_dump_mem *)
-				&input_mems_arr.ptr[input_offset];
-			u8 mem_group_id = GET_FIELD(mem->dword0,
-						    DBG_DUMP_MEM_MEM_GROUP_ID);
-			bool is_storm = false, mem_wide_bus;
-			enum dbg_grc_params grc_param;
-			char storm_letter = 'a';
-			enum block_id block_id;
+			    (const struct dbg_dump_mem *)((u32 *)
+							  input_mems_arr.ptr
+							  + input_offset);
+			const struct dbg_block *block;
+			char storm_letter = 0;
 			u32 mem_addr, mem_len;
+			bool mem_wide_bus;
+			u8 mem_group_id;
 
+			mem_group_id = GET_FIELD(mem->dword0,
+						 DBG_DUMP_MEM_MEM_GROUP_ID);
 			if (mem_group_id >= MEM_GROUPS_NUM) {
 				DP_NOTICE(p_hwfn, "Invalid mem_group_id\n");
 				return 0;
 			}
 
-			block_id = (enum block_id)cond_hdr->block_id;
 			if (!qed_grc_is_mem_included(p_hwfn,
-						     block_id,
+						     (enum block_id)
+						     cond_hdr->block_id,
 						     mem_group_id))
 				continue;
 
@@ -3236,42 +2559,14 @@ static u32 qed_grc_dump_mem_entries(struct qed_hwfn *p_hwfn,
 			mem_wide_bus = GET_FIELD(mem->dword1,
 						 DBG_DUMP_MEM_WIDE_BUS);
 
-			/* Update memory length for CCFC/TCFC memories
-			 * according to number of LCIDs/LTIDs.
-			 */
-			if (mem_group_id == MEM_GROUP_CONN_CFC_MEM) {
-				if (mem_len % MAX_LCIDS) {
-					DP_NOTICE(p_hwfn,
-						  "Invalid CCFC connection memory size\n");
-					return 0;
-				}
-
-				grc_param = DBG_GRC_PARAM_NUM_LCIDS;
-				mem_len = qed_grc_get_param(p_hwfn, grc_param) *
-					  (mem_len / MAX_LCIDS);
-			} else if (mem_group_id == MEM_GROUP_TASK_CFC_MEM) {
-				if (mem_len % MAX_LTIDS) {
-					DP_NOTICE(p_hwfn,
-						  "Invalid TCFC task memory size\n");
-					return 0;
-				}
-
-				grc_param = DBG_GRC_PARAM_NUM_LTIDS;
-				mem_len = qed_grc_get_param(p_hwfn, grc_param) *
-					  (mem_len / MAX_LTIDS);
-			}
+			block = get_dbg_block(p_hwfn,
+					      cond_hdr->block_id);
 
-			/* If memory is associated with Storm, update Storm
-			 * details.
+			/* If memory is associated with Storm,
+			 * update storm details
 			 */
-			if (s_block_defs
-			    [cond_hdr->block_id]->associated_to_storm) {
-				is_storm = true;
-				storm_letter =
-				    s_storm_defs[s_block_defs
-						 [cond_hdr->block_id]->
-						 storm_id].letter;
-			}
+			if (block->associated_storm_letter)
+				storm_letter = block->associated_storm_letter;
 
 			/* Dump memory */
 			offset += qed_grc_dump_mem(p_hwfn,
@@ -3285,7 +2580,6 @@ static u32 qed_grc_dump_mem_entries(struct qed_hwfn *p_hwfn,
 						0,
 						false,
 						s_mem_group_names[mem_group_id],
-						is_storm,
 						storm_letter);
 		}
 	}
@@ -3300,26 +2594,25 @@ static u32 qed_grc_dump_memories(struct qed_hwfn *p_hwfn,
 				 struct qed_ptt *p_ptt,
 				 u32 *dump_buf, bool dump)
 {
+	struct virt_mem_desc *dbg_buf =
+	    &p_hwfn->dbg_arrays[BIN_BUF_DBG_DUMP_MEM];
 	u32 offset = 0, input_offset = 0;
 
-	while (input_offset <
-	       s_dbg_arrays[BIN_BUF_DBG_DUMP_MEM].size_in_dwords) {
+	while (input_offset < BYTES_TO_DWORDS(dbg_buf->size)) {
 		const struct dbg_dump_split_hdr *split_hdr;
-		struct dbg_array curr_input_mems_arr;
+		struct virt_mem_desc curr_input_mems_arr;
 		enum init_split_types split_type;
 		u32 split_data_size;
 
-		split_hdr = (const struct dbg_dump_split_hdr *)
-			&s_dbg_arrays[BIN_BUF_DBG_DUMP_MEM].ptr[input_offset++];
-		split_type =
-			GET_FIELD(split_hdr->hdr,
-				  DBG_DUMP_SPLIT_HDR_SPLIT_TYPE_ID);
-		split_data_size =
-			GET_FIELD(split_hdr->hdr,
-				  DBG_DUMP_SPLIT_HDR_DATA_SIZE);
-		curr_input_mems_arr.ptr =
-			&s_dbg_arrays[BIN_BUF_DBG_DUMP_MEM].ptr[input_offset];
-		curr_input_mems_arr.size_in_dwords = split_data_size;
+		split_hdr =
+		    (const struct dbg_dump_split_hdr *)dbg_buf->ptr +
+		    input_offset++;
+		split_type = GET_FIELD(split_hdr->hdr,
+				       DBG_DUMP_SPLIT_HDR_SPLIT_TYPE_ID);
+		split_data_size = GET_FIELD(split_hdr->hdr,
+					    DBG_DUMP_SPLIT_HDR_DATA_SIZE);
+		curr_input_mems_arr.ptr = (u32 *)dbg_buf->ptr + input_offset;
+		curr_input_mems_arr.size = DWORDS_TO_BYTES(split_data_size);
 
 		if (split_type == SPLIT_TYPE_NONE)
 			offset += qed_grc_dump_mem_entries(p_hwfn,
@@ -3347,17 +2640,19 @@ static u32 qed_grc_dump_ctx_data(struct qed_hwfn *p_hwfn,
 				 bool dump,
 				 const char *name,
 				 u32 num_lids,
-				 u32 lid_size,
-				 u32 rd_reg_addr,
-				 u8 storm_id)
+				 enum cm_ctx_types ctx_type, u8 storm_id)
 {
+	struct dbg_tools_data *dev_data = &p_hwfn->dbg_info;
 	struct storm_defs *storm = &s_storm_defs[storm_id];
-	u32 i, lid, total_size, offset = 0;
+	u32 i, lid, lid_size, total_size;
+	u32 rd_reg_addr, offset = 0;
+
+	/* Convert quad-regs to dwords */
+	lid_size = storm->cm_ctx_lid_sizes[dev_data->chip_id][ctx_type] * 4;
 
 	if (!lid_size)
 		return 0;
 
-	lid_size *= BYTES_IN_DWORD;
 	total_size = num_lids * lid_size;
 
 	offset += qed_grc_dump_mem_hdr(p_hwfn,
@@ -3367,18 +2662,26 @@ static u32 qed_grc_dump_ctx_data(struct qed_hwfn *p_hwfn,
 				       0,
 				       total_size,
 				       lid_size * 32,
-				       false, name, true, storm->letter);
+				       false, name, storm->letter);
 
 	if (!dump)
 		return offset + total_size;
 
+	rd_reg_addr = BYTES_TO_DWORDS(storm->cm_ctx_rd_addr[ctx_type]);
+
 	/* Dump context data */
 	for (lid = 0; lid < num_lids; lid++) {
-		for (i = 0; i < lid_size; i++, offset++) {
+		for (i = 0; i < lid_size; i++) {
 			qed_wr(p_hwfn,
 			       p_ptt, storm->cm_ctx_wr_addr, (i << 9) | lid);
-			*(dump_buf + offset) = qed_rd(p_hwfn,
-						      p_ptt, rd_reg_addr);
+			offset += qed_grc_dump_addr_range(p_hwfn,
+							  p_ptt,
+							  dump_buf + offset,
+							  dump,
+							  rd_reg_addr,
+							  1,
+							  false,
+							  SPLIT_TYPE_NONE, 0);
 		}
 	}
 
@@ -3389,115 +2692,126 @@ static u32 qed_grc_dump_ctx_data(struct qed_hwfn *p_hwfn,
 static u32 qed_grc_dump_ctx(struct qed_hwfn *p_hwfn,
 			    struct qed_ptt *p_ptt, u32 *dump_buf, bool dump)
 {
-	enum dbg_grc_params grc_param;
 	u32 offset = 0;
 	u8 storm_id;
 
 	for (storm_id = 0; storm_id < MAX_DBG_STORMS; storm_id++) {
-		struct storm_defs *storm = &s_storm_defs[storm_id];
-
 		if (!qed_grc_is_storm_included(p_hwfn,
 					       (enum dbg_storms)storm_id))
 			continue;
 
 		/* Dump Conn AG context size */
-		grc_param = DBG_GRC_PARAM_NUM_LCIDS;
-		offset +=
-			qed_grc_dump_ctx_data(p_hwfn,
-					      p_ptt,
-					      dump_buf + offset,
-					      dump,
-					      "CONN_AG_CTX",
-					      qed_grc_get_param(p_hwfn,
-								grc_param),
-					      storm->cm_conn_ag_ctx_lid_size,
-					      storm->cm_conn_ag_ctx_rd_addr,
-					      storm_id);
+		offset += qed_grc_dump_ctx_data(p_hwfn,
+						p_ptt,
+						dump_buf + offset,
+						dump,
+						"CONN_AG_CTX",
+						NUM_OF_LCIDS,
+						CM_CTX_CONN_AG, storm_id);
 
 		/* Dump Conn ST context size */
-		grc_param = DBG_GRC_PARAM_NUM_LCIDS;
-		offset +=
-			qed_grc_dump_ctx_data(p_hwfn,
-					      p_ptt,
-					      dump_buf + offset,
-					      dump,
-					      "CONN_ST_CTX",
-					      qed_grc_get_param(p_hwfn,
-								grc_param),
-					      storm->cm_conn_st_ctx_lid_size,
-					      storm->cm_conn_st_ctx_rd_addr,
-					      storm_id);
+		offset += qed_grc_dump_ctx_data(p_hwfn,
+						p_ptt,
+						dump_buf + offset,
+						dump,
+						"CONN_ST_CTX",
+						NUM_OF_LCIDS,
+						CM_CTX_CONN_ST, storm_id);
 
 		/* Dump Task AG context size */
-		grc_param = DBG_GRC_PARAM_NUM_LTIDS;
-		offset +=
-			qed_grc_dump_ctx_data(p_hwfn,
-					      p_ptt,
-					      dump_buf + offset,
-					      dump,
-					      "TASK_AG_CTX",
-					      qed_grc_get_param(p_hwfn,
-								grc_param),
-					      storm->cm_task_ag_ctx_lid_size,
-					      storm->cm_task_ag_ctx_rd_addr,
-					      storm_id);
+		offset += qed_grc_dump_ctx_data(p_hwfn,
+						p_ptt,
+						dump_buf + offset,
+						dump,
+						"TASK_AG_CTX",
+						NUM_OF_LTIDS,
+						CM_CTX_TASK_AG, storm_id);
 
 		/* Dump Task ST context size */
-		grc_param = DBG_GRC_PARAM_NUM_LTIDS;
-		offset +=
-			qed_grc_dump_ctx_data(p_hwfn,
-					      p_ptt,
-					      dump_buf + offset,
-					      dump,
-					      "TASK_ST_CTX",
-					      qed_grc_get_param(p_hwfn,
-								grc_param),
-					      storm->cm_task_st_ctx_lid_size,
-					      storm->cm_task_st_ctx_rd_addr,
-					      storm_id);
+		offset += qed_grc_dump_ctx_data(p_hwfn,
+						p_ptt,
+						dump_buf + offset,
+						dump,
+						"TASK_ST_CTX",
+						NUM_OF_LTIDS,
+						CM_CTX_TASK_ST, storm_id);
 	}
 
 	return offset;
 }
 
-/* Dumps GRC IORs data. Returns the dumped size in dwords. */
-static u32 qed_grc_dump_iors(struct qed_hwfn *p_hwfn,
-			     struct qed_ptt *p_ptt, u32 *dump_buf, bool dump)
-{
-	char buf[10] = "IOR_SET_?";
-	u32 addr, offset = 0;
-	u8 storm_id, set_id;
+#define VFC_STATUS_RESP_READY_BIT	0
+#define VFC_STATUS_BUSY_BIT		1
+#define VFC_STATUS_SENDING_CMD_BIT	2
 
-	for (storm_id = 0; storm_id < MAX_DBG_STORMS; storm_id++) {
-		struct storm_defs *storm = &s_storm_defs[storm_id];
+#define VFC_POLLING_DELAY_MS	1
+#define VFC_POLLING_COUNT		20
 
-		if (!qed_grc_is_storm_included(p_hwfn,
-					       (enum dbg_storms)storm_id))
-			continue;
+/* Reads data from VFC. Returns the number of dwords read (0 on error).
+ * Sizes are specified in dwords.
+ */
+static u32 qed_grc_dump_read_from_vfc(struct qed_hwfn *p_hwfn,
+				      struct qed_ptt *p_ptt,
+				      struct storm_defs *storm,
+				      u32 *cmd_data,
+				      u32 cmd_size,
+				      u32 *addr_data,
+				      u32 addr_size,
+				      u32 resp_size, u32 *dump_buf)
+{
+	struct dbg_tools_data *dev_data = &p_hwfn->dbg_info;
+	u32 vfc_status, polling_ms, polling_count = 0, i;
+	u32 reg_addr, sem_base;
+	bool is_ready = false;
+
+	sem_base = storm->sem_fast_mem_addr;
+	polling_ms = VFC_POLLING_DELAY_MS *
+	    s_hw_type_defs[dev_data->hw_type].delay_factor;
+
+	/* Write VFC command */
+	ARR_REG_WR(p_hwfn,
+		   p_ptt,
+		   sem_base + SEM_FAST_REG_VFC_DATA_WR,
+		   cmd_data, cmd_size);
+
+	/* Write VFC address */
+	ARR_REG_WR(p_hwfn,
+		   p_ptt,
+		   sem_base + SEM_FAST_REG_VFC_ADDR,
+		   addr_data, addr_size);
+
+	/* Read response */
+	for (i = 0; i < resp_size; i++) {
+		/* Poll until ready */
+		do {
+			reg_addr = sem_base + SEM_FAST_REG_VFC_STATUS;
+			qed_grc_dump_addr_range(p_hwfn,
+						p_ptt,
+						&vfc_status,
+						true,
+						BYTES_TO_DWORDS(reg_addr),
+						1,
+						false, SPLIT_TYPE_NONE, 0);
+			is_ready = vfc_status & BIT(VFC_STATUS_RESP_READY_BIT);
+
+			if (!is_ready) {
+				if (polling_count++ == VFC_POLLING_COUNT)
+					return 0;
 
-		for (set_id = 0; set_id < NUM_IOR_SETS; set_id++) {
-			addr = BYTES_TO_DWORDS(storm->sem_fast_mem_addr +
-					       SEM_FAST_REG_STORM_REG_FILE) +
-			       IOR_SET_OFFSET(set_id);
-			if (strlen(buf) > 0)
-				buf[strlen(buf) - 1] = '0' + set_id;
-			offset += qed_grc_dump_mem(p_hwfn,
-						   p_ptt,
-						   dump_buf + offset,
-						   dump,
-						   buf,
-						   addr,
-						   IORS_PER_SET,
-						   false,
-						   32,
-						   false,
-						   "ior",
-						   true,
-						   storm->letter);
-		}
+				msleep(polling_ms);
+			}
+		} while (!is_ready);
+
+		reg_addr = sem_base + SEM_FAST_REG_VFC_DATA_RD;
+		qed_grc_dump_addr_range(p_hwfn,
+					p_ptt,
+					dump_buf + i,
+					true,
+					BYTES_TO_DWORDS(reg_addr),
+					1, false, SPLIT_TYPE_NONE, 0);
 	}
 
-	return offset;
+	return resp_size;
 }
 
 /* Dump VFC CAM. Returns the dumped size in dwords. */
@@ -3509,7 +2823,7 @@ static u32 qed_grc_dump_vfc_cam(struct qed_hwfn *p_hwfn,
 	struct storm_defs *storm = &s_storm_defs[storm_id];
 	u32 cam_addr[VFC_CAM_ADDR_DWORDS] = { 0 };
 	u32 cam_cmd[VFC_CAM_CMD_DWORDS] = { 0 };
-	u32 row, i, offset = 0;
+	u32 row, offset = 0;
 
 	offset += qed_grc_dump_mem_hdr(p_hwfn,
 				       dump_buf + offset,
@@ -3518,7 +2832,7 @@ static u32 qed_grc_dump_vfc_cam(struct qed_hwfn *p_hwfn,
 				       0,
 				       total_size,
 				       256,
-				       false, "vfc_cam", true, storm->letter);
+				       false, "vfc_cam", storm->letter);
 
 	if (!dump)
 		return offset + total_size;
@@ -3526,26 +2840,18 @@ static u32 qed_grc_dump_vfc_cam(struct qed_hwfn *p_hwfn,
 	/* Prepare CAM address */
 	SET_VAR_FIELD(cam_addr, VFC_CAM_ADDR, OP, VFC_OPCODE_CAM_RD);
 
-	for (row = 0; row < VFC_CAM_NUM_ROWS;
-	     row++, offset += VFC_CAM_RESP_DWORDS) {
-		/* Write VFC CAM command */
+	/* Read VFC CAM data */
+	for (row = 0; row < VFC_CAM_NUM_ROWS; row++) {
 		SET_VAR_FIELD(cam_cmd, VFC_CAM_CMD, ROW, row);
-		ARR_REG_WR(p_hwfn,
-			   p_ptt,
-			   storm->sem_fast_mem_addr + SEM_FAST_REG_VFC_DATA_WR,
-			   cam_cmd, VFC_CAM_CMD_DWORDS);
-
-		/* Write VFC CAM address */
-		ARR_REG_WR(p_hwfn,
-			   p_ptt,
-			   storm->sem_fast_mem_addr + SEM_FAST_REG_VFC_ADDR,
-			   cam_addr, VFC_CAM_ADDR_DWORDS);
-
-		/* Read VFC CAM read response */
-		ARR_REG_RD(p_hwfn,
-			   p_ptt,
-			   storm->sem_fast_mem_addr + SEM_FAST_REG_VFC_DATA_RD,
-			   dump_buf + offset, VFC_CAM_RESP_DWORDS);
+		offset += qed_grc_dump_read_from_vfc(p_hwfn,
+						     p_ptt,
+						     storm,
+						     cam_cmd,
+						     VFC_CAM_CMD_DWORDS,
+						     cam_addr,
+						     VFC_CAM_ADDR_DWORDS,
+						     VFC_CAM_RESP_DWORDS,
+						     dump_buf + offset);
 	}
 
 	return offset;
@@ -3562,7 +2868,7 @@ static u32 qed_grc_dump_vfc_ram(struct qed_hwfn *p_hwfn,
 	struct storm_defs *storm = &s_storm_defs[storm_id];
 	u32 ram_addr[VFC_RAM_ADDR_DWORDS] = { 0 };
 	u32 ram_cmd[VFC_RAM_CMD_DWORDS] = { 0 };
-	u32 row, i, offset = 0;
+	u32 row, offset = 0;
 
 	offset += qed_grc_dump_mem_hdr(p_hwfn,
 				       dump_buf + offset,
@@ -3573,35 +2879,27 @@ static u32 qed_grc_dump_vfc_ram(struct qed_hwfn *p_hwfn,
 				       256,
 				       false,
 				       ram_defs->type_name,
-				       true, storm->letter);
-
-	/* Prepare RAM address */
-	SET_VAR_FIELD(ram_addr, VFC_RAM_ADDR, OP, VFC_OPCODE_RAM_RD);
+				       storm->letter);
 
 	if (!dump)
 		return offset + total_size;
 
+	/* Prepare RAM address */
+	SET_VAR_FIELD(ram_addr, VFC_RAM_ADDR, OP, VFC_OPCODE_RAM_RD);
+
+	/* Read VFC RAM data */
 	for (row = ram_defs->base_row;
-	     row < ram_defs->base_row + ram_defs->num_rows;
-	     row++, offset += VFC_RAM_RESP_DWORDS) {
-		/* Write VFC RAM command */
-		ARR_REG_WR(p_hwfn,
-			   p_ptt,
-			   storm->sem_fast_mem_addr + SEM_FAST_REG_VFC_DATA_WR,
-			   ram_cmd, VFC_RAM_CMD_DWORDS);
-
-		/* Write VFC RAM address */
+	     row < ram_defs->base_row + ram_defs->num_rows; row++) {
 		SET_VAR_FIELD(ram_addr, VFC_RAM_ADDR, ROW, row);
-		ARR_REG_WR(p_hwfn,
-			   p_ptt,
-			   storm->sem_fast_mem_addr + SEM_FAST_REG_VFC_ADDR,
-			   ram_addr, VFC_RAM_ADDR_DWORDS);
-
-		/* Read VFC RAM read response */
-		ARR_REG_RD(p_hwfn,
-			   p_ptt,
-			   storm->sem_fast_mem_addr + SEM_FAST_REG_VFC_DATA_RD,
-			   dump_buf + offset, VFC_RAM_RESP_DWORDS);
+		offset += qed_grc_dump_read_from_vfc(p_hwfn,
+						     p_ptt,
+						     storm,
+						     ram_cmd,
+						     VFC_RAM_CMD_DWORDS,
+						     ram_addr,
+						     VFC_RAM_ADDR_DWORDS,
+						     VFC_RAM_RESP_DWORDS,
+						     dump_buf + offset);
 	}
 
 	return offset;
@@ -3611,16 +2909,13 @@ static u32 qed_grc_dump_vfc_ram(struct qed_hwfn *p_hwfn,
 static u32 qed_grc_dump_vfc(struct qed_hwfn *p_hwfn,
 			    struct qed_ptt *p_ptt, u32 *dump_buf, bool dump)
 {
-	struct dbg_tools_data *dev_data = &p_hwfn->dbg_info;
 	u8 storm_id, i;
 	u32 offset = 0;
 
 	for (storm_id = 0; storm_id < MAX_DBG_STORMS; storm_id++) {
 		if (!qed_grc_is_storm_included(p_hwfn,
 					       (enum dbg_storms)storm_id) ||
-		    !s_storm_defs[storm_id].has_vfc ||
-		    (storm_id == DBG_PSTORM_ID && dev_data->platform_id !=
-		     PLATFORM_ASIC))
+		    !s_storm_defs[storm_id].has_vfc)
 			continue;
 
 		/* Read CAM */
@@ -3670,7 +2965,7 @@ static u32 qed_grc_dump_rss(struct qed_hwfn *p_hwfn,
 					       total_dwords,
 					       rss_defs->entry_width,
 					       packed,
-					       rss_defs->type_name, false, 0);
+					       rss_defs->type_name, 0);
 
 		/* Dump RSS data */
 		if (!dump) {
@@ -3730,7 +3025,7 @@ static u32 qed_grc_dump_big_ram(struct qed_hwfn *p_hwfn,
 				       0,
 				       ram_size,
 				       block_size * 8,
-				       false, type_name, false, 0);
+				       false, type_name, 0);
 
 	/* Read and dump Big RAM data */
 	if (!dump)
@@ -3756,6 +3051,7 @@ static u32 qed_grc_dump_big_ram(struct qed_hwfn *p_hwfn,
 	return offset;
 }
 
+/* Dumps MCP scratchpad. Returns the dumped size in dwords. */
 static u32 qed_grc_dump_mcp(struct qed_hwfn *p_hwfn,
 			    struct qed_ptt *p_ptt, u32 *dump_buf, bool dump)
 {
@@ -3777,8 +3073,8 @@ static u32 qed_grc_dump_mcp(struct qed_hwfn *p_hwfn,
 				   dump,
 				   NULL,
 				   BYTES_TO_DWORDS(MCP_REG_SCRATCH),
-				   MCP_REG_SCRATCH_SIZE_BB_K2,
-				   false, 0, false, "MCP", false, 0);
+				   MCP_REG_SCRATCH_SIZE,
+				   false, 0, false, "MCP", 0);
 
 	/* Dump MCP cpu_reg_file */
 	offset += qed_grc_dump_mem(p_hwfn,
@@ -3788,19 +3084,19 @@ static u32 qed_grc_dump_mcp(struct qed_hwfn *p_hwfn,
 				   NULL,
 				   BYTES_TO_DWORDS(MCP_REG_CPU_REG_FILE),
 				   MCP_REG_CPU_REG_FILE_SIZE,
-				   false, 0, false, "MCP", false, 0);
+				   false, 0, false, "MCP", 0);
 
 	/* Dump MCP registers */
 	block_enable[BLOCK_MCP] = true;
 	offset += qed_grc_dump_registers(p_hwfn,
 					 p_ptt,
 					 dump_buf + offset,
-					 dump, block_enable, "block", "MCP");
+					 dump, block_enable, "MCP");
 
 	/* Dump required non-MCP registers */
 	offset += qed_grc_dump_regs_hdr(dump_buf + offset,
 					dump, 1, SPLIT_TYPE_NONE, 0,
-					"block", "MCP");
+					"MCP");
 	addr = BYTES_TO_DWORDS(MISC_REG_SHARED_MEM_ADDR);
 	offset += qed_grc_dump_reg_entry(p_hwfn,
 					 p_ptt,
@@ -3817,7 +3113,9 @@ static u32 qed_grc_dump_mcp(struct qed_hwfn *p_hwfn,
 	return offset;
 }
 
-/* Dumps the tbus indirect memory for all PHYs. */
+/* Dumps the tbus indirect memory for all PHYs.
+ * Returns the dumped size in dwords.
+ */
 static u32 qed_grc_dump_phy(struct qed_hwfn *p_hwfn,
 			    struct qed_ptt *p_ptt, u32 *dump_buf, bool dump)
 {
@@ -3851,7 +3149,7 @@ static u32 qed_grc_dump_phy(struct qed_hwfn *p_hwfn,
 					       mem_name,
 					       0,
 					       PHY_DUMP_SIZE_DWORDS,
-					       16, true, mem_name, false, 0);
+					       16, true, mem_name, 0);
 
 		if (!dump) {
 			offset += PHY_DUMP_SIZE_DWORDS;
@@ -3882,21 +3180,58 @@ static u32 qed_grc_dump_phy(struct qed_hwfn *p_hwfn,
 	return offset;
 }
 
-static void qed_config_dbg_line(struct qed_hwfn *p_hwfn,
-				struct qed_ptt *p_ptt,
-				enum block_id block_id,
-				u8 line_id,
-				u8 enable_mask,
-				u8 right_shift,
-				u8 force_valid_mask, u8 force_frame_mask)
+static enum dbg_status qed_find_nvram_image(struct qed_hwfn *p_hwfn,
+					    struct qed_ptt *p_ptt,
+					    u32 image_type,
+					    u32 *nvram_offset_bytes,
+					    u32 *nvram_size_bytes);
+
+static enum dbg_status qed_nvram_read(struct qed_hwfn *p_hwfn,
+				      struct qed_ptt *p_ptt,
+				      u32 nvram_offset_bytes,
+				      u32 nvram_size_bytes, u32 *ret_buf);
+
+/* Dumps the MCP HW dump from NVRAM. Returns the dumped size in dwords. */
+static u32 qed_grc_dump_mcp_hw_dump(struct qed_hwfn *p_hwfn,
+				    struct qed_ptt *p_ptt,
+				    u32 *dump_buf, bool dump)
 {
-	struct block_defs *block = s_block_defs[block_id];
+	u32 hw_dump_offset_bytes = 0, hw_dump_size_bytes = 0;
+	u32 hw_dump_size_dwords = 0, offset = 0;
+	enum dbg_status status;
+
+	/* Read HW dump image from NVRAM */
+	status = qed_find_nvram_image(p_hwfn,
+				      p_ptt,
+				      NVM_TYPE_HW_DUMP_OUT,
+				      &hw_dump_offset_bytes,
+				      &hw_dump_size_bytes);
+	if (status != DBG_STATUS_OK)
+		return 0;
+
+	hw_dump_size_dwords = BYTES_TO_DWORDS(hw_dump_size_bytes);
+
+	/* Dump HW dump image section */
+	offset += qed_dump_section_hdr(dump_buf + offset,
+				       dump, "mcp_hw_dump", 1);
+	offset += qed_dump_num_param(dump_buf + offset,
+				     dump, "size", hw_dump_size_dwords);
+
+	/* Read MCP HW dump image into dump buffer */
+	if (dump && hw_dump_size_dwords) {
+		status = qed_nvram_read(p_hwfn,
+					p_ptt,
+					hw_dump_offset_bytes,
+					hw_dump_size_bytes, dump_buf + offset);
+		if (status != DBG_STATUS_OK) {
+			DP_NOTICE(p_hwfn,
+				  "Failed to read MCP HW Dump image from NVRAM\n");
+			return 0;
+		}
+	}
+	offset += hw_dump_size_dwords;
 
-	qed_wr(p_hwfn, p_ptt, block->dbg_select_addr, line_id);
-	qed_wr(p_hwfn, p_ptt, block->dbg_enable_addr, enable_mask);
-	qed_wr(p_hwfn, p_ptt, block->dbg_shift_addr, right_shift);
-	qed_wr(p_hwfn, p_ptt, block->dbg_force_valid_addr, force_valid_mask);
-	qed_wr(p_hwfn, p_ptt, block->dbg_force_frame_addr, force_frame_mask);
+	return offset;
 }
 
 /* Dumps Static Debug data. Returns the dumped size in dwords. */
@@ -3905,26 +3240,19 @@ static u32 qed_grc_dump_static_debug(struct qed_hwfn *p_hwfn,
 				     u32 *dump_buf, bool dump)
 {
 	struct dbg_tools_data *dev_data = &p_hwfn->dbg_info;
-	u32 block_id, line_id, offset = 0;
+	u32 block_id, line_id, offset = 0, addr, len;
 
 	/* Don't dump static debug if a debug bus recording is in progress */
 	if (dump && qed_rd(p_hwfn, p_ptt, DBG_REG_DBG_BLOCK_ON))
 		return 0;
 
 	if (dump) {
-		/* Disable all blocks debug output */
-		for (block_id = 0; block_id < MAX_BLOCK_ID; block_id++) {
-			struct block_defs *block = s_block_defs[block_id];
-
-			if (block->dbg_client_id[dev_data->chip_id] !=
-			    MAX_DBG_BUS_CLIENTS)
-				qed_wr(p_hwfn, p_ptt, block->dbg_enable_addr,
-				       0);
-		}
+		/* Disable debug bus in all blocks */
+		qed_bus_disable_blocks(p_hwfn, p_ptt);
 
 		qed_bus_reset_dbg_block(p_hwfn, p_ptt);
-		qed_bus_set_framing_mode(p_hwfn,
-					 p_ptt, DBG_BUS_FRAME_MODE_8HW_0ST);
+		qed_wr(p_hwfn,
+		       p_ptt, DBG_REG_FRAMING_MODE, DBG_BUS_FRAME_MODE_8HW);
 		qed_wr(p_hwfn,
 		       p_ptt, DBG_REG_DEBUG_TARGET, DBG_BUS_TARGET_ID_INT_BUF);
 		qed_wr(p_hwfn, p_ptt, DBG_REG_FULL_MODE, 1);
@@ -3933,28 +3261,48 @@ static u32 qed_grc_dump_static_debug(struct qed_hwfn *p_hwfn,
 
 	/* Dump all static debug lines for each relevant block */
 	for (block_id = 0; block_id < MAX_BLOCK_ID; block_id++) {
-		struct block_defs *block = s_block_defs[block_id];
-		struct dbg_bus_block *block_desc;
-		u32 block_dwords, addr, len;
-		u8 dbg_client_id;
+		const struct dbg_block_chip *block_per_chip;
+		const struct dbg_block *block;
+		bool is_removed, has_dbg_bus;
+		u16 modes_buf_offset;
+		u32 block_dwords;
+
+		block_per_chip =
+		    qed_get_dbg_block_per_chip(p_hwfn, (enum block_id)block_id);
+		is_removed = GET_FIELD(block_per_chip->flags,
+				       DBG_BLOCK_CHIP_IS_REMOVED);
+		has_dbg_bus = GET_FIELD(block_per_chip->flags,
+					DBG_BLOCK_CHIP_HAS_DBG_BUS);
+
+		/* read+clear for NWS parity is not working, skip NWS block */
+		if (block_id == BLOCK_NWS)
+			continue;
+
+		if (!is_removed && has_dbg_bus &&
+		    GET_FIELD(block_per_chip->dbg_bus_mode.data,
+			      DBG_MODE_HDR_EVAL_MODE) > 0) {
+			modes_buf_offset =
+			    GET_FIELD(block_per_chip->dbg_bus_mode.data,
+				      DBG_MODE_HDR_MODES_BUF_OFFSET);
+			if (!qed_is_mode_match(p_hwfn, &modes_buf_offset))
+				has_dbg_bus = false;
+		}
 
-		if (block->dbg_client_id[dev_data->chip_id] ==
-		    MAX_DBG_BUS_CLIENTS)
+		if (is_removed || !has_dbg_bus)
 			continue;
 
-		block_desc = get_dbg_bus_block_desc(p_hwfn,
-						    (enum block_id)block_id);
-		block_dwords = NUM_DBG_LINES(block_desc) *
+		block_dwords = NUM_DBG_LINES(block_per_chip) *
 			       STATIC_DEBUG_LINE_DWORDS;
 
 		/* Dump static section params */
+		block = get_dbg_block(p_hwfn, (enum block_id)block_id);
 		offset += qed_grc_dump_mem_hdr(p_hwfn,
 					       dump_buf + offset,
 					       dump,
 					       block->name,
 					       0,
 					       block_dwords,
-					       32, false, "STATIC", false, 0);
+					       32, false, "STATIC", 0);
 
 		if (!dump) {
 			offset += block_dwords;
@@ -3970,20 +3318,19 @@ static u32 qed_grc_dump_static_debug(struct qed_hwfn *p_hwfn,
 		}
 
 		/* Enable block's client */
-		dbg_client_id = block->dbg_client_id[dev_data->chip_id];
 		qed_bus_enable_clients(p_hwfn,
 				       p_ptt,
-				       BIT(dbg_client_id));
+				       BIT(block_per_chip->dbg_client_id));
 
 		addr = BYTES_TO_DWORDS(DBG_REG_CALENDAR_OUT_DATA);
 		len = STATIC_DEBUG_LINE_DWORDS;
-		for (line_id = 0; line_id < (u32)NUM_DBG_LINES(block_desc);
+		for (line_id = 0; line_id < (u32)NUM_DBG_LINES(block_per_chip);
 		     line_id++) {
 			/* Configure debug line ID */
-			qed_config_dbg_line(p_hwfn,
-					    p_ptt,
-					    (enum block_id)block_id,
-					    (u8)line_id, 0xf, 0, 0, 0);
+			qed_bus_config_dbg_line(p_hwfn,
+						p_ptt,
+						(enum block_id)block_id,
+						(u8)line_id, 0xf, 0, 0, 0);
 
 			/* Read debug line info */
 			offset += qed_grc_dump_addr_range(p_hwfn,
@@ -3998,7 +3345,8 @@ static u32 qed_grc_dump_static_debug(struct qed_hwfn *p_hwfn,
 
 		/* Disable block's client and debug output */
 		qed_bus_enable_clients(p_hwfn, p_ptt, 0);
-		qed_wr(p_hwfn, p_ptt, block->dbg_enable_addr, 0);
+		qed_bus_config_dbg_line(p_hwfn, p_ptt,
+					(enum block_id)block_id, 0, 0, 0, 0, 0);
 	}
 
 	if (dump) {
@@ -4018,8 +3366,8 @@ static enum dbg_status qed_grc_dump(struct qed_hwfn *p_hwfn,
 				    bool dump, u32 *num_dumped_dwords)
 {
 	struct dbg_tools_data *dev_data = &p_hwfn->dbg_info;
+	u32 dwords_read, offset = 0;
 	bool parities_masked = false;
-	u32 offset = 0;
 	u8 i;
 
 	*num_dumped_dwords = 0;
@@ -4038,13 +3386,11 @@ static enum dbg_status qed_grc_dump(struct qed_hwfn *p_hwfn,
 	offset += qed_dump_num_param(dump_buf + offset,
 				     dump,
 				     "num-lcids",
-				     qed_grc_get_param(p_hwfn,
-						DBG_GRC_PARAM_NUM_LCIDS));
+				     NUM_OF_LCIDS);
 	offset += qed_dump_num_param(dump_buf + offset,
 				     dump,
 				     "num-ltids",
-				     qed_grc_get_param(p_hwfn,
-						DBG_GRC_PARAM_NUM_LTIDS));
+				     NUM_OF_LTIDS);
 	offset += qed_dump_num_param(dump_buf + offset,
 				     dump, "num-ports", dev_data->num_ports);
 
@@ -4056,7 +3402,7 @@ static enum dbg_status qed_grc_dump(struct qed_hwfn *p_hwfn,
 
 	/* Take all blocks out of reset (using reset registers) */
 	if (dump) {
-		qed_grc_unreset_blocks(p_hwfn, p_ptt);
+		qed_grc_unreset_blocks(p_hwfn, p_ptt, false);
 		qed_update_blocks_reset_state(p_hwfn, p_ptt);
 	}
 
@@ -4099,7 +3445,7 @@ static enum dbg_status qed_grc_dump(struct qed_hwfn *p_hwfn,
 						 dump_buf +
 						 offset,
 						 dump,
-						 block_enable, NULL, NULL);
+						 block_enable, NULL);
 
 		/* Dump special registers */
 		offset += qed_grc_dump_special_regs(p_hwfn,
@@ -4133,23 +3479,29 @@ static enum dbg_status qed_grc_dump(struct qed_hwfn *p_hwfn,
 						       dump_buf + offset,
 						       dump, i);
 
-	/* Dump IORs */
-	if (qed_grc_is_included(p_hwfn, DBG_GRC_PARAM_DUMP_IOR))
-		offset += qed_grc_dump_iors(p_hwfn,
-					    p_ptt, dump_buf + offset, dump);
-
 	/* Dump VFC */
-	if (qed_grc_is_included(p_hwfn, DBG_GRC_PARAM_DUMP_VFC))
-		offset += qed_grc_dump_vfc(p_hwfn,
-					   p_ptt, dump_buf + offset, dump);
+	if (qed_grc_is_included(p_hwfn, DBG_GRC_PARAM_DUMP_VFC)) {
+		dwords_read = qed_grc_dump_vfc(p_hwfn,
+					       p_ptt, dump_buf + offset, dump);
+		offset += dwords_read;
+		if (!dwords_read)
+			return DBG_STATUS_VFC_READ_ERROR;
+	}
 
 	/* Dump PHY tbus */
 	if (qed_grc_is_included(p_hwfn,
 				DBG_GRC_PARAM_DUMP_PHY) && dev_data->chip_id ==
-	    CHIP_K2 && dev_data->platform_id == PLATFORM_ASIC)
+	    CHIP_K2 && dev_data->hw_type == HW_TYPE_ASIC)
 		offset += qed_grc_dump_phy(p_hwfn,
 					   p_ptt, dump_buf + offset, dump);
 
+	/* Dump MCP HW Dump */
+	if (qed_grc_is_included(p_hwfn, DBG_GRC_PARAM_DUMP_MCP_HW_DUMP) &&
+	    !qed_grc_get_param(p_hwfn, DBG_GRC_PARAM_NO_MCP) && 1)
+		offset += qed_grc_dump_mcp_hw_dump(p_hwfn,
+						   p_ptt,
+						   dump_buf + offset, dump);
+
 	/* Dump static debug data (only if not during debug bus recording) */
 	if (qed_grc_is_included(p_hwfn,
 				DBG_GRC_PARAM_DUMP_STATIC) &&
@@ -4200,8 +3552,9 @@ static u32 qed_idle_chk_dump_failure(struct qed_hwfn *p_hwfn,
 	u8 reg_id;
 
 	hdr = (struct dbg_idle_chk_result_hdr *)dump_buf;
-	regs = &((const union dbg_idle_chk_reg *)
-		 s_dbg_arrays[BIN_BUF_DBG_IDLE_CHK_REGS].ptr)[rule->reg_offset];
+	regs = (const union dbg_idle_chk_reg *)
+		p_hwfn->dbg_arrays[BIN_BUF_DBG_IDLE_CHK_REGS].ptr +
+		rule->reg_offset;
 	cond_regs = &regs[0].cond_reg;
 	info_regs = &regs[rule->num_cond_regs].info_reg;
 
@@ -4221,8 +3574,8 @@ static u32 qed_idle_chk_dump_failure(struct qed_hwfn *p_hwfn,
 		const struct dbg_idle_chk_cond_reg *reg = &cond_regs[reg_id];
 		struct dbg_idle_chk_result_reg_hdr *reg_hdr;
 
-		reg_hdr = (struct dbg_idle_chk_result_reg_hdr *)
-			  (dump_buf + offset);
+		reg_hdr =
+		    (struct dbg_idle_chk_result_reg_hdr *)(dump_buf + offset);
 
 		/* Write register header */
 		if (!dump) {
@@ -4339,12 +3692,13 @@ qed_idle_chk_dump_rule_entries(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt,
 		const u32 *imm_values;
 
 		rule = &input_rules[i];
-		regs = &((const union dbg_idle_chk_reg *)
-			 s_dbg_arrays[BIN_BUF_DBG_IDLE_CHK_REGS].ptr)
-			[rule->reg_offset];
+		regs = (const union dbg_idle_chk_reg *)
+			p_hwfn->dbg_arrays[BIN_BUF_DBG_IDLE_CHK_REGS].ptr +
+			rule->reg_offset;
 		cond_regs = &regs[0].cond_reg;
-		imm_values = &s_dbg_arrays[BIN_BUF_DBG_IDLE_CHK_IMMS].ptr
-			     [rule->imm_offset];
+		imm_values =
+		    (u32 *)p_hwfn->dbg_arrays[BIN_BUF_DBG_IDLE_CHK_IMMS].ptr +
+		    rule->imm_offset;
 
 		/* Check if all condition register blocks are out of reset, and
 		 * find maximal number of entries (all condition registers that
@@ -4462,10 +3816,12 @@ qed_idle_chk_dump_rule_entries(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt,
 static u32 qed_idle_chk_dump(struct qed_hwfn *p_hwfn,
 			     struct qed_ptt *p_ptt, u32 *dump_buf, bool dump)
 {
-	u32 num_failing_rules_offset, offset = 0, input_offset = 0;
-	u32 num_failing_rules = 0;
+	struct virt_mem_desc *dbg_buf =
+	    &p_hwfn->dbg_arrays[BIN_BUF_DBG_IDLE_CHK_RULES];
+	u32 num_failing_rules_offset, offset = 0,
+	    input_offset = 0, num_failing_rules = 0;
 
-	/* Dump global params */
+	/* Dump global params  - 1 must match below amount of params */
 	offset += qed_dump_common_global_params(p_hwfn,
 						p_ptt,
 						dump_buf + offset, dump, 1);
@@ -4477,12 +3833,10 @@ static u32 qed_idle_chk_dump(struct qed_hwfn *p_hwfn,
 	num_failing_rules_offset = offset;
 	offset += qed_dump_num_param(dump_buf + offset, dump, "num_rules", 0);
 
-	while (input_offset <
-	       s_dbg_arrays[BIN_BUF_DBG_IDLE_CHK_RULES].size_in_dwords) {
+	while (input_offset < BYTES_TO_DWORDS(dbg_buf->size)) {
 		const struct dbg_idle_chk_cond_hdr *cond_hdr =
-			(const struct dbg_idle_chk_cond_hdr *)
-			&s_dbg_arrays[BIN_BUF_DBG_IDLE_CHK_RULES].ptr
-			[input_offset++];
+		    (const struct dbg_idle_chk_cond_hdr *)dbg_buf->ptr +
+		    input_offset++;
 		bool eval_mode, mode_match = true;
 		u32 curr_failing_rules;
 		u16 modes_buf_offset;
@@ -4499,16 +3853,21 @@ static u32 qed_idle_chk_dump(struct qed_hwfn *p_hwfn,
 		}
 
 		if (mode_match) {
+			const struct dbg_idle_chk_rule *rule =
+			    (const struct dbg_idle_chk_rule *)((u32 *)
+							       dbg_buf->ptr
+							       + input_offset);
+			u32 num_input_rules =
+				cond_hdr->data_size / IDLE_CHK_RULE_SIZE_DWORDS;
 			offset +=
 			    qed_idle_chk_dump_rule_entries(p_hwfn,
-				p_ptt,
-				dump_buf + offset,
-				dump,
-				(const struct dbg_idle_chk_rule *)
-				&s_dbg_arrays[BIN_BUF_DBG_IDLE_CHK_RULES].
-				ptr[input_offset],
-				cond_hdr->data_size / IDLE_CHK_RULE_SIZE_DWORDS,
-				&curr_failing_rules);
+							   p_ptt,
+							   dump_buf +
+							   offset,
+							   dump,
+							   rule,
+							   num_input_rules,
+							   &curr_failing_rules);
 			num_failing_rules += curr_failing_rules;
 		}
 
@@ -4575,7 +3934,7 @@ static enum dbg_status qed_nvram_read(struct qed_hwfn *p_hwfn,
 {
 	u32 ret_mcp_resp, ret_mcp_param, ret_read_size, bytes_to_copy;
 	s32 bytes_left = nvram_size_bytes;
-	u32 read_offset = 0;
+	u32 read_offset = 0, param = 0;
 
 	DP_VERBOSE(p_hwfn,
 		   QED_MSG_DEBUG,
@@ -4588,14 +3947,14 @@ static enum dbg_status qed_nvram_read(struct qed_hwfn *p_hwfn,
 		     MCP_DRV_NVM_BUF_LEN) ? MCP_DRV_NVM_BUF_LEN : bytes_left;
 
 		/* Call NVRAM read command */
+		SET_MFW_FIELD(param,
+			      DRV_MB_PARAM_NVM_OFFSET,
+			      nvram_offset_bytes + read_offset);
+		SET_MFW_FIELD(param, DRV_MB_PARAM_NVM_LEN, bytes_to_copy);
 		if (qed_mcp_nvm_rd_cmd(p_hwfn, p_ptt,
-				       DRV_MSG_CODE_NVM_READ_NVRAM,
-				       (nvram_offset_bytes +
-					read_offset) |
-				       (bytes_to_copy <<
-					DRV_MB_PARAM_NVM_LEN_OFFSET),
-				       &ret_mcp_resp, &ret_mcp_param,
-				       &ret_read_size,
+				       DRV_MSG_CODE_NVM_READ_NVRAM, param,
+				       &ret_mcp_resp,
+				       &ret_mcp_param, &ret_read_size,
 				       (u32 *)((u8 *)ret_buf + read_offset)))
 			return DBG_STATUS_NVRAM_READ_FAILED;
 
@@ -4733,12 +4092,12 @@ static enum dbg_status qed_mcp_trace_dump(struct qed_hwfn *p_hwfn,
 	u32 trace_meta_size_dwords = 0, running_bundle_id, offset = 0;
 	u32 trace_meta_offset_bytes = 0, trace_meta_size_bytes = 0;
 	enum dbg_status status;
-	bool mcp_access;
 	int halted = 0;
+	bool use_mfw;
 
 	*num_dumped_dwords = 0;
 
-	mcp_access = !qed_grc_get_param(p_hwfn, DBG_GRC_PARAM_NO_MCP);
+	use_mfw = !qed_grc_get_param(p_hwfn, DBG_GRC_PARAM_NO_MCP);
 
 	/* Get trace data info */
 	status = qed_mcp_trace_get_data_info(p_hwfn,
@@ -4759,7 +4118,7 @@ static enum dbg_status qed_mcp_trace_dump(struct qed_hwfn *p_hwfn,
 	 * consistent. if halt fails, MCP trace is taken anyway, with a small
 	 * risk that it may be corrupt.
 	 */
-	if (dump && mcp_access) {
+	if (dump && use_mfw) {
 		halted = !qed_mcp_halt(p_hwfn, p_ptt);
 		if (!halted)
 			DP_NOTICE(p_hwfn, "MCP halt failed!\n");
@@ -4799,17 +4158,15 @@ static enum dbg_status qed_mcp_trace_dump(struct qed_hwfn *p_hwfn,
 	 */
 	trace_meta_size_bytes =
 		qed_grc_get_param(p_hwfn, DBG_GRC_PARAM_MCP_TRACE_META_SIZE);
-	if ((!trace_meta_size_bytes || dump) && mcp_access) {
+	if ((!trace_meta_size_bytes || dump) && use_mfw)
 		status = qed_mcp_trace_get_meta_info(p_hwfn,
 						     p_ptt,
 						     trace_data_size_bytes,
 						     &running_bundle_id,
 						     &trace_meta_offset_bytes,
 						     &trace_meta_size_bytes);
-		if (status == DBG_STATUS_OK)
-			trace_meta_size_dwords =
-				BYTES_TO_DWORDS(trace_meta_size_bytes);
-	}
+	if (status == DBG_STATUS_OK)
+		trace_meta_size_dwords = BYTES_TO_DWORDS(trace_meta_size_bytes);
 
 	/* Dump trace meta size param */
 	offset += qed_dump_num_param(dump_buf + offset,
@@ -4833,7 +4190,7 @@ static enum dbg_status qed_mcp_trace_dump(struct qed_hwfn *p_hwfn,
 	/* If no mcp access, indicate that the dump doesn't contain the meta
 	 * data from NVRAM.
 	 */
-	return mcp_access ? status : DBG_STATUS_NVRAM_GET_IMAGE_FAILED;
+	return use_mfw ? status : DBG_STATUS_NVRAM_GET_IMAGE_FAILED;
 }
 
 /* Dump GRC FIFO */
@@ -5058,7 +4415,7 @@ static u32 qed_fw_asserts_dump(struct qed_hwfn *p_hwfn,
 		struct storm_defs *storm = &s_storm_defs[storm_id];
 		u32 last_list_idx, addr;
 
-		if (dev_data->block_in_reset[storm->block_id])
+		if (dev_data->block_in_reset[storm->sem_block_id])
 			continue;
 
 		/* Read FW info for the current Storm */
@@ -5453,18 +4810,18 @@ static u32 qed_ilt_dump(struct qed_hwfn *p_hwfn,
 
 /***************************** Public Functions *******************************/
 
-enum dbg_status qed_dbg_set_bin_ptr(const u8 * const bin_ptr)
+enum dbg_status qed_dbg_set_bin_ptr(struct qed_hwfn *p_hwfn,
+				    const u8 * const bin_ptr)
 {
-	struct bin_buffer_hdr *buf_array = (struct bin_buffer_hdr *)bin_ptr;
+	struct bin_buffer_hdr *buf_hdrs = (struct bin_buffer_hdr *)bin_ptr;
 	u8 buf_id;
 
-	/* convert binary data to debug arrays */
-	for (buf_id = 0; buf_id < MAX_BIN_DBG_BUFFER_TYPE; buf_id++) {
-		s_dbg_arrays[buf_id].ptr =
-		    (u32 *)(bin_ptr + buf_array[buf_id].offset);
-		s_dbg_arrays[buf_id].size_in_dwords =
-		    BYTES_TO_DWORDS(buf_array[buf_id].length);
-	}
+	/* Convert binary data to debug arrays */
+	for (buf_id = 0; buf_id < MAX_BIN_DBG_BUFFER_TYPE; buf_id++)
+		qed_set_dbg_bin_buf(p_hwfn,
+				    buf_id,
+				    (u32 *)(bin_ptr + buf_hdrs[buf_id].offset),
+				    buf_hdrs[buf_id].length);
 
 	return DBG_STATUS_OK;
 }
@@ -5479,7 +4836,7 @@ bool qed_read_fw_info(struct qed_hwfn *p_hwfn,
 		struct storm_defs *storm = &s_storm_defs[storm_id];
 
 		/* Skip Storm if it's in reset */
-		if (dev_data->block_in_reset[storm->block_id])
+		if (dev_data->block_in_reset[storm->sem_block_id])
 			continue;
 
 		/* Read FW info for the current Storm */
@@ -5492,16 +4849,17 @@ bool qed_read_fw_info(struct qed_hwfn *p_hwfn,
 }
 
 enum dbg_status qed_dbg_grc_config(struct qed_hwfn *p_hwfn,
-				   struct qed_ptt *p_ptt,
 				   enum dbg_grc_params grc_param, u32 val)
 {
+	struct dbg_tools_data *dev_data = &p_hwfn->dbg_info;
 	enum dbg_status status;
 	int i;
 
-	DP_VERBOSE(p_hwfn, QED_MSG_DEBUG,
+	DP_VERBOSE(p_hwfn,
+		   QED_MSG_DEBUG,
 		   "dbg_grc_config: paramId = %d, val = %d\n", grc_param, val);
 
-	status = qed_dbg_dev_init(p_hwfn, p_ptt);
+	status = qed_dbg_dev_init(p_hwfn);
 	if (status != DBG_STATUS_OK)
 		return status;
 
@@ -5527,24 +4885,23 @@ enum dbg_status qed_dbg_grc_config(struct qed_hwfn *p_hwfn,
 
 		/* Update all params with the preset values */
 		for (i = 0; i < MAX_DBG_GRC_PARAMS; i++) {
+			struct grc_param_defs *defs = &s_grc_param_defs[i];
 			u32 preset_val;
-
 			/* Skip persistent params */
-			if (s_grc_param_defs[i].is_persistent)
+			if (defs->is_persistent)
 				continue;
 
 			/* Find preset value */
 			if (grc_param == DBG_GRC_PARAM_EXCLUDE_ALL)
 				preset_val =
-				    s_grc_param_defs[i].exclude_all_preset_val;
+				    defs->exclude_all_preset_val;
 			else if (grc_param == DBG_GRC_PARAM_CRASH)
 				preset_val =
-				    s_grc_param_defs[i].crash_preset_val;
+				    defs->crash_preset_val[dev_data->chip_id];
 			else
 				return DBG_STATUS_INVALID_ARGS;
 
-			qed_grc_set_param(p_hwfn,
-					  (enum dbg_grc_params)i, preset_val);
+			qed_grc_set_param(p_hwfn, i, preset_val);
 		}
 	} else {
 		/* Regular param - set its value */
@@ -5570,18 +4927,18 @@ enum dbg_status qed_dbg_grc_get_dump_buf_size(struct qed_hwfn *p_hwfn,
 					      struct qed_ptt *p_ptt,
 					      u32 *buf_size)
 {
-	enum dbg_status status = qed_dbg_dev_init(p_hwfn, p_ptt);
+	enum dbg_status status = qed_dbg_dev_init(p_hwfn);
 
 	*buf_size = 0;
 
 	if (status != DBG_STATUS_OK)
 		return status;
 
-	if (!s_dbg_arrays[BIN_BUF_DBG_MODE_TREE].ptr ||
-	    !s_dbg_arrays[BIN_BUF_DBG_DUMP_REG].ptr ||
-	    !s_dbg_arrays[BIN_BUF_DBG_DUMP_MEM].ptr ||
-	    !s_dbg_arrays[BIN_BUF_DBG_ATTN_BLOCKS].ptr ||
-	    !s_dbg_arrays[BIN_BUF_DBG_ATTN_REGS].ptr)
+	if (!p_hwfn->dbg_arrays[BIN_BUF_DBG_MODE_TREE].ptr ||
+	    !p_hwfn->dbg_arrays[BIN_BUF_DBG_DUMP_REG].ptr ||
+	    !p_hwfn->dbg_arrays[BIN_BUF_DBG_DUMP_MEM].ptr ||
+	    !p_hwfn->dbg_arrays[BIN_BUF_DBG_ATTN_BLOCKS].ptr ||
+	    !p_hwfn->dbg_arrays[BIN_BUF_DBG_ATTN_REGS].ptr)
 		return DBG_STATUS_DBG_ARRAY_NOT_SET;
 
 	return qed_grc_dump(p_hwfn, p_ptt, NULL, false, buf_size);
@@ -5621,20 +4978,19 @@ enum dbg_status qed_dbg_idle_chk_get_dump_buf_size(struct qed_hwfn *p_hwfn,
 						   u32 *buf_size)
 {
 	struct dbg_tools_data *dev_data = &p_hwfn->dbg_info;
-	struct idle_chk_data *idle_chk;
+	struct idle_chk_data *idle_chk = &dev_data->idle_chk;
 	enum dbg_status status;
 
-	idle_chk = &dev_data->idle_chk;
 	*buf_size = 0;
 
-	status = qed_dbg_dev_init(p_hwfn, p_ptt);
+	status = qed_dbg_dev_init(p_hwfn);
 	if (status != DBG_STATUS_OK)
 		return status;
 
-	if (!s_dbg_arrays[BIN_BUF_DBG_MODE_TREE].ptr ||
-	    !s_dbg_arrays[BIN_BUF_DBG_IDLE_CHK_REGS].ptr ||
-	    !s_dbg_arrays[BIN_BUF_DBG_IDLE_CHK_IMMS].ptr ||
-	    !s_dbg_arrays[BIN_BUF_DBG_IDLE_CHK_RULES].ptr)
+	if (!p_hwfn->dbg_arrays[BIN_BUF_DBG_MODE_TREE].ptr ||
+	    !p_hwfn->dbg_arrays[BIN_BUF_DBG_IDLE_CHK_REGS].ptr ||
+	    !p_hwfn->dbg_arrays[BIN_BUF_DBG_IDLE_CHK_IMMS].ptr ||
+	    !p_hwfn->dbg_arrays[BIN_BUF_DBG_IDLE_CHK_RULES].ptr)
 		return DBG_STATUS_DBG_ARRAY_NOT_SET;
 
 	if (!idle_chk->buf_size_set) {
@@ -5669,6 +5025,7 @@ enum dbg_status qed_dbg_idle_chk_dump(struct qed_hwfn *p_hwfn,
 		return DBG_STATUS_DUMP_BUF_TOO_SMALL;
 
 	/* Update reset state */
+	qed_grc_unreset_blocks(p_hwfn, p_ptt, true);
 	qed_update_blocks_reset_state(p_hwfn, p_ptt);
 
 	/* Idle Check Dump */
@@ -5684,7 +5041,7 @@ enum dbg_status qed_dbg_mcp_trace_get_dump_buf_size(struct qed_hwfn *p_hwfn,
 						    struct qed_ptt *p_ptt,
 						    u32 *buf_size)
 {
-	enum dbg_status status = qed_dbg_dev_init(p_hwfn, p_ptt);
+	enum dbg_status status = qed_dbg_dev_init(p_hwfn);
 
 	*buf_size = 0;
 
@@ -5731,7 +5088,7 @@ enum dbg_status qed_dbg_reg_fifo_get_dump_buf_size(struct qed_hwfn *p_hwfn,
 						   struct qed_ptt *p_ptt,
 						   u32 *buf_size)
 {
-	enum dbg_status status = qed_dbg_dev_init(p_hwfn, p_ptt);
+	enum dbg_status status = qed_dbg_dev_init(p_hwfn);
 
 	*buf_size = 0;
 
@@ -5777,7 +5134,7 @@ enum dbg_status qed_dbg_igu_fifo_get_dump_buf_size(struct qed_hwfn *p_hwfn,
 						   struct qed_ptt *p_ptt,
 						   u32 *buf_size)
 {
-	enum dbg_status status = qed_dbg_dev_init(p_hwfn, p_ptt);
+	enum dbg_status status = qed_dbg_dev_init(p_hwfn);
 
 	*buf_size = 0;
 
@@ -5823,7 +5180,7 @@ qed_dbg_protection_override_get_dump_buf_size(struct qed_hwfn *p_hwfn,
 					      struct qed_ptt *p_ptt,
 					      u32 *buf_size)
 {
-	enum dbg_status status = qed_dbg_dev_init(p_hwfn, p_ptt);
+	enum dbg_status status = qed_dbg_dev_init(p_hwfn);
 
 	*buf_size = 0;
 
@@ -5873,7 +5230,7 @@ enum dbg_status qed_dbg_fw_asserts_get_dump_buf_size(struct qed_hwfn *p_hwfn,
 						     struct qed_ptt *p_ptt,
 						     u32 *buf_size)
 {
-	enum dbg_status status = qed_dbg_dev_init(p_hwfn, p_ptt);
+	enum dbg_status status = qed_dbg_dev_init(p_hwfn);
 
 	*buf_size = 0;
 
@@ -5921,7 +5278,7 @@ static enum dbg_status qed_dbg_ilt_get_dump_buf_size(struct qed_hwfn *p_hwfn,
 						     struct qed_ptt *p_ptt,
 						     u32 *buf_size)
 {
-	enum dbg_status status = qed_dbg_dev_init(p_hwfn, p_ptt);
+	enum dbg_status status = qed_dbg_dev_init(p_hwfn);
 
 	*buf_size = 0;
 
@@ -5968,19 +5325,20 @@ enum dbg_status qed_dbg_read_attn(struct qed_hwfn *p_hwfn,
 				  bool clear_status,
 				  struct dbg_attn_block_result *results)
 {
-	enum dbg_status status = qed_dbg_dev_init(p_hwfn, p_ptt);
+	enum dbg_status status = qed_dbg_dev_init(p_hwfn);
 	u8 reg_idx, num_attn_regs, num_result_regs = 0;
 	const struct dbg_attn_reg *attn_reg_arr;
 
 	if (status != DBG_STATUS_OK)
 		return status;
 
-	if (!s_dbg_arrays[BIN_BUF_DBG_MODE_TREE].ptr ||
-	    !s_dbg_arrays[BIN_BUF_DBG_ATTN_BLOCKS].ptr ||
-	    !s_dbg_arrays[BIN_BUF_DBG_ATTN_REGS].ptr)
+	if (!p_hwfn->dbg_arrays[BIN_BUF_DBG_MODE_TREE].ptr ||
+	    !p_hwfn->dbg_arrays[BIN_BUF_DBG_ATTN_BLOCKS].ptr ||
+	    !p_hwfn->dbg_arrays[BIN_BUF_DBG_ATTN_REGS].ptr)
 		return DBG_STATUS_DBG_ARRAY_NOT_SET;
 
-	attn_reg_arr = qed_get_block_attn_regs(block_id,
+	attn_reg_arr = qed_get_block_attn_regs(p_hwfn,
+					       block_id,
 					       attn_type, &num_attn_regs);
 
 	for (reg_idx = 0; reg_idx < num_attn_regs; reg_idx++) {
@@ -6025,7 +5383,7 @@ enum dbg_status qed_dbg_read_attn(struct qed_hwfn *p_hwfn,
 
 	results->block_id = (u8)block_id;
 	results->names_offset =
-	    qed_get_block_attn_data(block_id, attn_type)->names_offset;
+	    qed_get_block_attn_data(p_hwfn, block_id, attn_type)->names_offset;
 	SET_FIELD(results->data, DBG_ATTN_BLOCK_RESULT_ATTN_TYPE, attn_type);
 	SET_FIELD(results->data,
 		  DBG_ATTN_BLOCK_RESULT_NUM_REGS, num_result_regs);
@@ -6035,11 +5393,6 @@ enum dbg_status qed_dbg_read_attn(struct qed_hwfn *p_hwfn,
 
 /******************************* Data Types **********************************/
 
-struct block_info {
-	const char *name;
-	enum block_id id;
-};
-
 /* REG fifo element */
 struct reg_fifo_element {
 	u64 data;
@@ -6063,6 +5416,12 @@ struct reg_fifo_element {
 #define REG_FIFO_ELEMENT_ERROR_MASK		0x1f
 };
 
+/* REG fifo error element */
+struct reg_fifo_err {
+	u32 err_code;
+	const char *err_msg;
+};
+
 /* IGU fifo element */
 struct igu_fifo_element {
 	u32 dword0;
@@ -6162,20 +5521,6 @@ struct igu_fifo_addr_data {
 	enum igu_fifo_addr_types type;
 };
 
-struct mcp_trace_meta {
-	u32 modules_num;
-	char **modules;
-	u32 formats_num;
-	struct mcp_trace_format *formats;
-	bool is_allocated;
-};
-
-/* Debug Tools user data */
-struct dbg_tools_user_data {
-	struct mcp_trace_meta mcp_trace_meta;
-	const u32 *mcp_trace_user_meta_buf;
-};
-
 /******************************** Constants **********************************/
 
 #define MAX_MSG_LEN				1024
@@ -6183,7 +5528,7 @@ struct dbg_tools_user_data {
 #define MCP_TRACE_MAX_MODULE_LEN		8
 #define MCP_TRACE_FORMAT_MAX_PARAMS		3
 #define MCP_TRACE_FORMAT_PARAM_WIDTH \
-	(MCP_TRACE_FORMAT_P2_SIZE_SHIFT - MCP_TRACE_FORMAT_P1_SIZE_SHIFT)
+	(MCP_TRACE_FORMAT_P2_SIZE_OFFSET - MCP_TRACE_FORMAT_P1_SIZE_OFFSET)
 
 #define REG_FIFO_ELEMENT_ADDR_FACTOR		4
 #define REG_FIFO_ELEMENT_IS_PF_VF_VAL		127
@@ -6192,107 +5537,6 @@ struct dbg_tools_user_data {
 
 /***************************** Constant Arrays *******************************/
 
-struct user_dbg_array {
-	const u32 *ptr;
-	u32 size_in_dwords;
-};
-
-/* Debug arrays */
-static struct user_dbg_array
-s_user_dbg_arrays[MAX_BIN_DBG_BUFFER_TYPE] = { {NULL} };
-
-/* Block names array */
-static struct block_info s_block_info_arr[] = {
-	{"grc", BLOCK_GRC},
-	{"miscs", BLOCK_MISCS},
-	{"misc", BLOCK_MISC},
-	{"dbu", BLOCK_DBU},
-	{"pglue_b", BLOCK_PGLUE_B},
-	{"cnig", BLOCK_CNIG},
-	{"cpmu", BLOCK_CPMU},
-	{"ncsi", BLOCK_NCSI},
-	{"opte", BLOCK_OPTE},
-	{"bmb", BLOCK_BMB},
-	{"pcie", BLOCK_PCIE},
-	{"mcp", BLOCK_MCP},
-	{"mcp2", BLOCK_MCP2},
-	{"pswhst", BLOCK_PSWHST},
-	{"pswhst2", BLOCK_PSWHST2},
-	{"pswrd", BLOCK_PSWRD},
-	{"pswrd2", BLOCK_PSWRD2},
-	{"pswwr", BLOCK_PSWWR},
-	{"pswwr2", BLOCK_PSWWR2},
-	{"pswrq", BLOCK_PSWRQ},
-	{"pswrq2", BLOCK_PSWRQ2},
-	{"pglcs", BLOCK_PGLCS},
-	{"ptu", BLOCK_PTU},
-	{"dmae", BLOCK_DMAE},
-	{"tcm", BLOCK_TCM},
-	{"mcm", BLOCK_MCM},
-	{"ucm", BLOCK_UCM},
-	{"xcm", BLOCK_XCM},
-	{"ycm", BLOCK_YCM},
-	{"pcm", BLOCK_PCM},
-	{"qm", BLOCK_QM},
-	{"tm", BLOCK_TM},
-	{"dorq", BLOCK_DORQ},
-	{"brb", BLOCK_BRB},
-	{"src", BLOCK_SRC},
-	{"prs", BLOCK_PRS},
-	{"tsdm", BLOCK_TSDM},
-	{"msdm", BLOCK_MSDM},
-	{"usdm", BLOCK_USDM},
-	{"xsdm", BLOCK_XSDM},
-	{"ysdm", BLOCK_YSDM},
-	{"psdm", BLOCK_PSDM},
-	{"tsem", BLOCK_TSEM},
-	{"msem", BLOCK_MSEM},
-	{"usem", BLOCK_USEM},
-	{"xsem", BLOCK_XSEM},
-	{"ysem", BLOCK_YSEM},
-	{"psem", BLOCK_PSEM},
-	{"rss", BLOCK_RSS},
-	{"tmld", BLOCK_TMLD},
-	{"muld", BLOCK_MULD},
-	{"yuld", BLOCK_YULD},
-	{"xyld", BLOCK_XYLD},
-	{"ptld", BLOCK_PTLD},
-	{"ypld", BLOCK_YPLD},
-	{"prm", BLOCK_PRM},
-	{"pbf_pb1", BLOCK_PBF_PB1},
-	{"pbf_pb2", BLOCK_PBF_PB2},
-	{"rpb", BLOCK_RPB},
-	{"btb", BLOCK_BTB},
-	{"pbf", BLOCK_PBF},
-	{"rdif", BLOCK_RDIF},
-	{"tdif", BLOCK_TDIF},
-	{"cdu", BLOCK_CDU},
-	{"ccfc", BLOCK_CCFC},
-	{"tcfc", BLOCK_TCFC},
-	{"igu", BLOCK_IGU},
-	{"cau", BLOCK_CAU},
-	{"rgfs", BLOCK_RGFS},
-	{"rgsrc", BLOCK_RGSRC},
-	{"tgfs", BLOCK_TGFS},
-	{"tgsrc", BLOCK_TGSRC},
-	{"umac", BLOCK_UMAC},
-	{"xmac", BLOCK_XMAC},
-	{"dbg", BLOCK_DBG},
-	{"nig", BLOCK_NIG},
-	{"wol", BLOCK_WOL},
-	{"bmbn", BLOCK_BMBN},
-	{"ipc", BLOCK_IPC},
-	{"nwm", BLOCK_NWM},
-	{"nws", BLOCK_NWS},
-	{"ms", BLOCK_MS},
-	{"phy_pcie", BLOCK_PHY_PCIE},
-	{"led", BLOCK_LED},
-	{"avs_wrap", BLOCK_AVS_WRAP},
-	{"pxpreqbus", BLOCK_PXPREQBUS},
-	{"misc_aeu", BLOCK_MISC_AEU},
-	{"bar0_map", BLOCK_BAR0_MAP}
-};
-
 /* Status string array */
 static const char * const s_status_str[] = {
 	/* DBG_STATUS_OK */
@@ -6322,14 +5566,12 @@ static const char * const s_status_str[] = {
 	/* DBG_STATUS_PCI_BUF_NOT_ALLOCATED */
 	"A PCI buffer wasn't allocated",
 
-	/* DBG_STATUS_TOO_MANY_INPUTS */
-	"Too many inputs were enabled. Enabled less inputs, or set 'unifyInputs' to true",
+	/* DBG_STATUS_INVALID_FILTER_TRIGGER_DWORDS */
+	"The filter/trigger constraint dword offsets are not enabled for recording",
 
-	/* DBG_STATUS_INPUT_OVERLAP */
-	"Overlapping debug bus inputs",
 
-	/* DBG_STATUS_HW_ONLY_RECORDING */
-	"Cannot record Storm data since the entire recording cycle is used by HW",
+	/* DBG_STATUS_VFC_READ_ERROR */
+	"Error reading from VFC",
 
 	/* DBG_STATUS_STORM_ALREADY_ENABLED */
 	"The Storm was already enabled",
@@ -6346,8 +5588,8 @@ static const char * const s_status_str[] = {
 	/* DBG_STATUS_NO_INPUT_ENABLED */
 	"No input was enabled for recording",
 
-	/* DBG_STATUS_NO_FILTER_TRIGGER_64B */
-	"Filters and triggers are not allowed when recording in 64b units",
+	/* DBG_STATUS_NO_FILTER_TRIGGER_256B */
+	"Filters and triggers are not allowed in E4 256-bit mode",
 
 	/* DBG_STATUS_FILTER_ALREADY_ENABLED */
 	"The filter was already enabled",
@@ -6421,8 +5663,8 @@ static const char * const s_status_str[] = {
 	/* DBG_STATUS_MCP_COULD_NOT_RESUME */
 	"Failed to resume MCP after halt",
 
-	/* DBG_STATUS_RESERVED2 */
-	"Reserved debug status - shouldn't be returned",
+	/* DBG_STATUS_RESERVED0 */
+	"",
 
 	/* DBG_STATUS_SEMI_FIFO_NOT_EMPTY */
 	"Failed to empty SEMI sync FIFO",
@@ -6445,17 +5687,32 @@ static const char * const s_status_str[] = {
 	/* DBG_STATUS_DBG_ARRAY_NOT_SET */
 	"Debug arrays were not set (when using binary files, dbg_set_bin_ptr must be called)",
 
-	/* DBG_STATUS_FILTER_BUG */
-	"Debug Bus filtering requires the -unifyInputs option (due to a HW bug)",
+	/* DBG_STATUS_RESERVED1 */
+	"",
 
 	/* DBG_STATUS_NON_MATCHING_LINES */
-	"Non-matching debug lines - all lines must be of the same type (either 128b or 256b)",
+	"Non-matching debug lines - in E4, all lines must be of the same type (either 128b or 256b)",
 
-	/* DBG_STATUS_INVALID_TRIGGER_DWORD_OFFSET */
-	"The selected trigger dword offset wasn't enabled in the recorded HW block",
+	/* DBG_STATUS_INSUFFICIENT_HW_IDS */
+	"Insufficient HW IDs. Try to record less Storms/blocks",
 
 	/* DBG_STATUS_DBG_BUS_IN_USE */
-	"The debug bus is in use"
+	"The debug bus is in use",
+
+	/* DBG_STATUS_INVALID_STORM_DBG_MODE */
+	"The storm debug mode is not supported in the current chip",
+
+	/* DBG_STATUS_OTHER_ENGINE_BB_ONLY */
+	"Other engine is supported only in BB",
+
+	/* DBG_STATUS_FILTER_SINGLE_HW_ID */
+	"The configured filter mode requires a single Storm/block input",
+
+	/* DBG_STATUS_TRIGGER_SINGLE_HW_ID */
+	"The configured filter mode requires that all the constraints of a single trigger state will be defined on a single Storm/block input",
+
+	/* DBG_STATUS_MISSING_TRIGGER_STATE_STORM */
+	"When triggering on Storm data, the Storm to trigger on must be specified"
 };
 
 /* Idle check severity names array */
@@ -6511,7 +5768,7 @@ static const char * const s_master_strs[] = {
 	"xsdm",
 	"dbu",
 	"dmae",
-	"???",
+	"jdap",
 	"???",
 	"???",
 	"???",
@@ -6519,12 +5776,13 @@ static const char * const s_master_strs[] = {
 };
 
 /* REG FIFO error messages array */
-static const char * const s_reg_fifo_error_strs[] = {
-	"grc timeout",
-	"address doesn't belong to any block",
-	"reserved address in block or write to read-only address",
-	"privilege/protection mismatch",
-	"path isolation error"
+static struct reg_fifo_err s_reg_fifo_errors[] = {
+	{1, "grc timeout"},
+	{2, "address doesn't belong to any block"},
+	{4, "reserved address in block or write to read-only address"},
+	{8, "privilege/protection mismatch"},
+	{16, "path isolation error"},
+	{17, "RSL error"}
 };
 
 /* IGU FIFO sources array */
@@ -6764,8 +6022,21 @@ static u32 qed_print_section_params(u32 *dump_buf,
 	return dump_offset;
 }
 
-static struct dbg_tools_user_data *
-qed_dbg_get_user_data(struct qed_hwfn *p_hwfn)
+/* Returns the block name that matches the specified block ID,
+ * or NULL if not found.
+ */
+static const char *qed_dbg_get_block_name(struct qed_hwfn *p_hwfn,
+					  enum block_id block_id)
+{
+	const struct dbg_block_user *block =
+	    (const struct dbg_block_user *)
+	    p_hwfn->dbg_arrays[BIN_BUF_DBG_BLOCKS_USER_DATA].ptr + block_id;
+
+	return (const char *)block->name;
+}
+
+static struct dbg_tools_user_data *qed_dbg_get_user_data(struct qed_hwfn
+							 *p_hwfn)
 {
 	return (struct dbg_tools_user_data *)p_hwfn->dbg_user_info;
 }
@@ -6773,7 +6044,8 @@ qed_dbg_get_user_data(struct qed_hwfn *p_hwfn)
 /* Parses the idle check rules and returns the number of characters printed.
  * In case of parsing error, returns 0.
  */
-static u32 qed_parse_idle_chk_dump_rules(u32 *dump_buf,
+static u32 qed_parse_idle_chk_dump_rules(struct qed_hwfn *p_hwfn,
+					 u32 *dump_buf,
 					 u32 *dump_buf_end,
 					 u32 num_rules,
 					 bool print_fw_idle_chk,
@@ -6801,19 +6073,18 @@ static u32 qed_parse_idle_chk_dump_rules(u32 *dump_buf,
 
 		hdr = (struct dbg_idle_chk_result_hdr *)dump_buf;
 		rule_parsing_data =
-			(const struct dbg_idle_chk_rule_parsing_data *)
-			&s_user_dbg_arrays[BIN_BUF_DBG_IDLE_CHK_PARSING_DATA].
-			ptr[hdr->rule_id];
+		    (const struct dbg_idle_chk_rule_parsing_data *)
+		    p_hwfn->dbg_arrays[BIN_BUF_DBG_IDLE_CHK_PARSING_DATA].ptr +
+		    hdr->rule_id;
 		parsing_str_offset =
-			GET_FIELD(rule_parsing_data->data,
-				  DBG_IDLE_CHK_RULE_PARSING_DATA_STR_OFFSET);
+		    GET_FIELD(rule_parsing_data->data,
+			      DBG_IDLE_CHK_RULE_PARSING_DATA_STR_OFFSET);
 		has_fw_msg =
-			GET_FIELD(rule_parsing_data->data,
-				DBG_IDLE_CHK_RULE_PARSING_DATA_HAS_FW_MSG) > 0;
-		parsing_str =
-			&((const char *)
-			s_user_dbg_arrays[BIN_BUF_DBG_PARSING_STRINGS].ptr)
-			[parsing_str_offset];
+		    GET_FIELD(rule_parsing_data->data,
+			      DBG_IDLE_CHK_RULE_PARSING_DATA_HAS_FW_MSG) > 0;
+		parsing_str = (const char *)
+		    p_hwfn->dbg_arrays[BIN_BUF_DBG_PARSING_STRINGS].ptr +
+		    parsing_str_offset;
 		lsi_msg = parsing_str;
 		curr_reg_id = 0;
 
@@ -6917,7 +6188,8 @@ static u32 qed_parse_idle_chk_dump_rules(u32 *dump_buf,
  * parsed_results_bytes.
  * The parsing status is returned.
  */
-static enum dbg_status qed_parse_idle_chk_dump(u32 *dump_buf,
+static enum dbg_status qed_parse_idle_chk_dump(struct qed_hwfn *p_hwfn,
+					       u32 *dump_buf,
 					       u32 num_dumped_dwords,
 					       char *results_buf,
 					       u32 *parsed_results_bytes,
@@ -6935,8 +6207,8 @@ static enum dbg_status qed_parse_idle_chk_dump(u32 *dump_buf,
 	*num_errors = 0;
 	*num_warnings = 0;
 
-	if (!s_user_dbg_arrays[BIN_BUF_DBG_PARSING_STRINGS].ptr ||
-	    !s_user_dbg_arrays[BIN_BUF_DBG_IDLE_CHK_PARSING_DATA].ptr)
+	if (!p_hwfn->dbg_arrays[BIN_BUF_DBG_PARSING_STRINGS].ptr ||
+	    !p_hwfn->dbg_arrays[BIN_BUF_DBG_IDLE_CHK_PARSING_DATA].ptr)
 		return DBG_STATUS_DBG_ARRAY_NOT_SET;
 
 	/* Read global_params section */
@@ -6969,7 +6241,8 @@ static enum dbg_status qed_parse_idle_chk_dump(u32 *dump_buf,
 					    results_offset),
 			    "FW_IDLE_CHECK:\n");
 		rules_print_size =
-			qed_parse_idle_chk_dump_rules(dump_buf,
+			qed_parse_idle_chk_dump_rules(p_hwfn,
+						      dump_buf,
 						      dump_buf_end,
 						      num_rules,
 						      true,
@@ -6989,7 +6262,8 @@ static enum dbg_status qed_parse_idle_chk_dump(u32 *dump_buf,
 					    results_offset),
 			    "\nLSI_IDLE_CHECK:\n");
 		rules_print_size =
-			qed_parse_idle_chk_dump_rules(dump_buf,
+			qed_parse_idle_chk_dump_rules(p_hwfn,
+						      dump_buf,
 						      dump_buf_end,
 						      num_rules,
 						      false,
@@ -7101,9 +6375,8 @@ qed_mcp_trace_alloc_meta_data(struct qed_hwfn *p_hwfn,
 
 		format_ptr->data = qed_read_dword_from_buf(meta_buf_bytes,
 							   &offset);
-		format_len =
-		    (format_ptr->data &
-		     MCP_TRACE_FORMAT_LEN_MASK) >> MCP_TRACE_FORMAT_LEN_SHIFT;
+		format_len = GET_MFW_FIELD(format_ptr->data,
+					   MCP_TRACE_FORMAT_LEN);
 		format_ptr->format_str = kzalloc(format_len, GFP_KERNEL);
 		if (!format_ptr->format_str) {
 			/* Update number of modules to be released */
@@ -7126,7 +6399,7 @@ qed_mcp_trace_alloc_meta_data(struct qed_hwfn *p_hwfn,
  * trace_buf - MCP trace cyclic buffer
  * trace_buf_size - MCP trace cyclic buffer size in bytes
  * data_offset - offset in bytes of the data to parse in the MCP trace cyclic
- *               buffer.
+ *		 buffer.
  * data_size - size in bytes of data to parse.
  * parsed_buf - destination buffer for parsed data.
  * parsed_results_bytes - size of parsed data in bytes.
@@ -7171,9 +6444,8 @@ static enum dbg_status qed_parse_mcp_trace_buf(struct qed_hwfn *p_hwfn,
 
 		/* Skip message if its index doesn't exist in the meta data */
 		if (format_idx >= meta->formats_num) {
-			u8 format_size =
-				(u8)((header & MFW_TRACE_PRM_SIZE_MASK) >>
-				     MFW_TRACE_PRM_SIZE_SHIFT);
+			u8 format_size = (u8)GET_MFW_FIELD(header,
+							   MFW_TRACE_PRM_SIZE);
 
 			if (data_size < format_size)
 				return DBG_STATUS_MCP_TRACE_BAD_DATA;
@@ -7188,11 +6460,10 @@ static enum dbg_status qed_parse_mcp_trace_buf(struct qed_hwfn *p_hwfn,
 		format_ptr = &meta->formats[format_idx];
 
 		for (i = 0,
-		     param_mask = MCP_TRACE_FORMAT_P1_SIZE_MASK,
-		     param_shift = MCP_TRACE_FORMAT_P1_SIZE_SHIFT;
+		     param_mask = MCP_TRACE_FORMAT_P1_SIZE_MASK, param_shift =
+		     MCP_TRACE_FORMAT_P1_SIZE_OFFSET;
 		     i < MCP_TRACE_FORMAT_MAX_PARAMS;
-		     i++,
-		     param_mask <<= MCP_TRACE_FORMAT_PARAM_WIDTH,
+		     i++, param_mask <<= MCP_TRACE_FORMAT_PARAM_WIDTH,
 		     param_shift += MCP_TRACE_FORMAT_PARAM_WIDTH) {
 			/* Extract param size (0..3) */
 			u8 param_size = (u8)((format_ptr->data & param_mask) >>
@@ -7220,12 +6491,10 @@ static enum dbg_status qed_parse_mcp_trace_buf(struct qed_hwfn *p_hwfn,
 			data_size -= param_size;
 		}
 
-		format_level = (u8)((format_ptr->data &
-				     MCP_TRACE_FORMAT_LEVEL_MASK) >>
-				    MCP_TRACE_FORMAT_LEVEL_SHIFT);
-		format_module = (u8)((format_ptr->data &
-				      MCP_TRACE_FORMAT_MODULE_MASK) >>
-				     MCP_TRACE_FORMAT_MODULE_SHIFT);
+		format_level = (u8)GET_MFW_FIELD(format_ptr->data,
+						 MCP_TRACE_FORMAT_LEVEL);
+		format_module = (u8)GET_MFW_FIELD(format_ptr->data,
+						  MCP_TRACE_FORMAT_MODULE);
 		if (format_level >= ARRAY_SIZE(s_mcp_trace_level_str))
 			return DBG_STATUS_MCP_TRACE_BAD_DATA;
 
@@ -7367,7 +6636,7 @@ static enum dbg_status qed_parse_reg_fifo_dump(u32 *dump_buf,
 	const char *section_name, *param_name, *param_str_val;
 	u32 param_num_val, num_section_params, num_elements;
 	struct reg_fifo_element *elements;
-	u8 i, j, err_val, vf_val;
+	u8 i, j, err_code, vf_val;
 	u32 results_offset = 0;
 	char vf_str[4];
 
@@ -7398,7 +6667,7 @@ static enum dbg_status qed_parse_reg_fifo_dump(u32 *dump_buf,
 
 	/* Decode elements */
 	for (i = 0; i < num_elements; i++) {
-		bool err_printed = false;
+		const char *err_msg = NULL;
 
 		/* Discover if element belongs to a VF or a PF */
 		vf_val = GET_FIELD(elements[i].data, REG_FIFO_ELEMENT_VF);
@@ -7407,11 +6676,17 @@ static enum dbg_status qed_parse_reg_fifo_dump(u32 *dump_buf,
 		else
 			sprintf(vf_str, "%d", vf_val);
 
+		/* Find error message */
+		err_code = GET_FIELD(elements[i].data, REG_FIFO_ELEMENT_ERROR);
+		for (j = 0; j < ARRAY_SIZE(s_reg_fifo_errors) && !err_msg; j++)
+			if (err_code == s_reg_fifo_errors[j].err_code)
+				err_msg = s_reg_fifo_errors[j].err_msg;
+
 		/* Add parsed element to parsed buffer */
 		results_offset +=
 		    sprintf(qed_get_buf_ptr(results_buf,
 					    results_offset),
-			    "raw: 0x%016llx, address: 0x%07x, access: %-5s, pf: %2d, vf: %s, port: %d, privilege: %-3s, protection: %-12s, master: %-4s, errors: ",
+			    "raw: 0x%016llx, address: 0x%07x, access: %-5s, pf: %2d, vf: %s, port: %d, privilege: %-3s, protection: %-12s, master: %-4s, error: %s\n",
 			    elements[i].data,
 			    (u32)GET_FIELD(elements[i].data,
 					   REG_FIFO_ELEMENT_ADDRESS) *
@@ -7428,30 +6703,8 @@ static enum dbg_status qed_parse_reg_fifo_dump(u32 *dump_buf,
 			    s_protection_strs[GET_FIELD(elements[i].data,
 						REG_FIFO_ELEMENT_PROTECTION)],
 			    s_master_strs[GET_FIELD(elements[i].data,
-						REG_FIFO_ELEMENT_MASTER)]);
-
-		/* Print errors */
-		for (j = 0,
-		     err_val = GET_FIELD(elements[i].data,
-					 REG_FIFO_ELEMENT_ERROR);
-		     j < ARRAY_SIZE(s_reg_fifo_error_strs);
-		     j++, err_val >>= 1) {
-			if (err_val & 0x1) {
-				if (err_printed)
-					results_offset +=
-					    sprintf(qed_get_buf_ptr
-						    (results_buf,
-						     results_offset), ", ");
-				results_offset +=
-				    sprintf(qed_get_buf_ptr
-					    (results_buf, results_offset), "%s",
-					    s_reg_fifo_error_strs[j]);
-				err_printed = true;
-			}
-		}
-
-		results_offset +=
-		    sprintf(qed_get_buf_ptr(results_buf, results_offset), "\n");
+						    REG_FIFO_ELEMENT_MASTER)],
+			    err_msg ? err_msg : "unknown error code");
 	}
 
 	results_offset += sprintf(qed_get_buf_ptr(results_buf,
@@ -7805,27 +7058,28 @@ static enum dbg_status qed_parse_fw_asserts_dump(u32 *dump_buf,
 
 /***************************** Public Functions *******************************/
 
-enum dbg_status qed_dbg_user_set_bin_ptr(const u8 * const bin_ptr)
+enum dbg_status qed_dbg_user_set_bin_ptr(struct qed_hwfn *p_hwfn,
+					 const u8 * const bin_ptr)
 {
-	struct bin_buffer_hdr *buf_array = (struct bin_buffer_hdr *)bin_ptr;
+	struct bin_buffer_hdr *buf_hdrs = (struct bin_buffer_hdr *)bin_ptr;
 	u8 buf_id;
 
 	/* Convert binary data to debug arrays */
-	for (buf_id = 0; buf_id < MAX_BIN_DBG_BUFFER_TYPE; buf_id++) {
-		s_user_dbg_arrays[buf_id].ptr =
-			(u32 *)(bin_ptr + buf_array[buf_id].offset);
-		s_user_dbg_arrays[buf_id].size_in_dwords =
-			BYTES_TO_DWORDS(buf_array[buf_id].length);
-	}
+	for (buf_id = 0; buf_id < MAX_BIN_DBG_BUFFER_TYPE; buf_id++)
+		qed_set_dbg_bin_buf(p_hwfn,
+				    (enum bin_dbg_buffer_type)buf_id,
+				    (u32 *)(bin_ptr + buf_hdrs[buf_id].offset),
+				    buf_hdrs[buf_id].length);
 
 	return DBG_STATUS_OK;
 }
 
-enum dbg_status qed_dbg_alloc_user_data(struct qed_hwfn *p_hwfn)
+enum dbg_status qed_dbg_alloc_user_data(struct qed_hwfn *p_hwfn,
+					void **user_data_ptr)
 {
-	p_hwfn->dbg_user_info = kzalloc(sizeof(struct dbg_tools_user_data),
-					GFP_KERNEL);
-	if (!p_hwfn->dbg_user_info)
+	*user_data_ptr = kzalloc(sizeof(struct dbg_tools_user_data),
+				 GFP_KERNEL);
+	if (!(*user_data_ptr))
 		return DBG_STATUS_VIRT_MEM_ALLOC_FAILED;
 
 	return DBG_STATUS_OK;
@@ -7844,7 +7098,8 @@ enum dbg_status qed_get_idle_chk_results_buf_size(struct qed_hwfn *p_hwfn,
 {
 	u32 num_errors, num_warnings;
 
-	return qed_parse_idle_chk_dump(dump_buf,
+	return qed_parse_idle_chk_dump(p_hwfn,
+				       dump_buf,
 				       num_dumped_dwords,
 				       NULL,
 				       results_buf_size,
@@ -7860,7 +7115,8 @@ enum dbg_status qed_print_idle_chk_results(struct qed_hwfn *p_hwfn,
 {
 	u32 parsed_buf_size;
 
-	return qed_parse_idle_chk_dump(dump_buf,
+	return qed_parse_idle_chk_dump(p_hwfn,
+				       dump_buf,
 				       num_dumped_dwords,
 				       results_buf,
 				       &parsed_buf_size,
@@ -8031,25 +7287,28 @@ enum dbg_status qed_print_fw_asserts_results(struct qed_hwfn *p_hwfn,
 enum dbg_status qed_dbg_parse_attn(struct qed_hwfn *p_hwfn,
 				   struct dbg_attn_block_result *results)
 {
-	struct user_dbg_array *block_attn, *pstrings;
 	const u32 *block_attn_name_offsets;
-	enum dbg_attn_type attn_type;
+	const char *attn_name_base;
 	const char *block_name;
+	enum dbg_attn_type attn_type;
 	u8 num_regs, i, j;
 
 	num_regs = GET_FIELD(results->data, DBG_ATTN_BLOCK_RESULT_NUM_REGS);
-	attn_type = (enum dbg_attn_type)
-		    GET_FIELD(results->data,
-			      DBG_ATTN_BLOCK_RESULT_ATTN_TYPE);
-	block_name = s_block_info_arr[results->block_id].name;
-
-	if (!s_user_dbg_arrays[BIN_BUF_DBG_ATTN_INDEXES].ptr ||
-	    !s_user_dbg_arrays[BIN_BUF_DBG_ATTN_NAME_OFFSETS].ptr ||
-	    !s_user_dbg_arrays[BIN_BUF_DBG_PARSING_STRINGS].ptr)
+	attn_type = GET_FIELD(results->data, DBG_ATTN_BLOCK_RESULT_ATTN_TYPE);
+	block_name = qed_dbg_get_block_name(p_hwfn, results->block_id);
+	if (!block_name)
+		return DBG_STATUS_INVALID_ARGS;
+
+	if (!p_hwfn->dbg_arrays[BIN_BUF_DBG_ATTN_INDEXES].ptr ||
+	    !p_hwfn->dbg_arrays[BIN_BUF_DBG_ATTN_NAME_OFFSETS].ptr ||
+	    !p_hwfn->dbg_arrays[BIN_BUF_DBG_PARSING_STRINGS].ptr)
 		return DBG_STATUS_DBG_ARRAY_NOT_SET;
 
-	block_attn = &s_user_dbg_arrays[BIN_BUF_DBG_ATTN_NAME_OFFSETS];
-	block_attn_name_offsets = &block_attn->ptr[results->names_offset];
+	block_attn_name_offsets =
+	    (u32 *)p_hwfn->dbg_arrays[BIN_BUF_DBG_ATTN_NAME_OFFSETS].ptr +
+	    results->names_offset;
+
+	attn_name_base = p_hwfn->dbg_arrays[BIN_BUF_DBG_PARSING_STRINGS].ptr;
 
 	/* Go over registers with a non-zero attention status */
 	for (i = 0; i < num_regs; i++) {
@@ -8060,18 +7319,17 @@ enum dbg_status qed_dbg_parse_attn(struct qed_hwfn *p_hwfn,
 		reg_result = &results->reg_results[i];
 		num_reg_attn = GET_FIELD(reg_result->data,
 					 DBG_ATTN_REG_RESULT_NUM_REG_ATTN);
-		block_attn = &s_user_dbg_arrays[BIN_BUF_DBG_ATTN_INDEXES];
-		bit_mapping = &((struct dbg_attn_bit_mapping *)
-				block_attn->ptr)[reg_result->block_attn_offset];
-
-		pstrings = &s_user_dbg_arrays[BIN_BUF_DBG_PARSING_STRINGS];
+		bit_mapping = (struct dbg_attn_bit_mapping *)
+		    p_hwfn->dbg_arrays[BIN_BUF_DBG_ATTN_INDEXES].ptr +
+		    reg_result->block_attn_offset;
 
 		/* Go over attention status bits */
-		for (j = 0; j < num_reg_attn; j++) {
+		for (j = 0; j < num_reg_attn; j++, bit_idx++) {
 			u16 attn_idx_val = GET_FIELD(bit_mapping[j].data,
 						     DBG_ATTN_BIT_MAPPING_VAL);
 			const char *attn_name, *attn_type_str, *masked_str;
-			u32 attn_name_offset, sts_addr;
+			u32 attn_name_offset;
+			u32 sts_addr;
 
 			/* Check if bit mask should be advanced (due to unused
 			 * bits).
@@ -8083,18 +7341,19 @@ enum dbg_status qed_dbg_parse_attn(struct qed_hwfn *p_hwfn,
 			}
 
 			/* Check current bit index */
-			if (!(reg_result->sts_val & BIT(bit_idx))) {
-				bit_idx++;
+			if (!(reg_result->sts_val & BIT(bit_idx)))
 				continue;
-			}
 
-			/* Find attention name */
+			/* An attention bit with value=1 was found
+			 * Find attention name
+			 */
 			attn_name_offset =
 				block_attn_name_offsets[attn_idx_val];
-			attn_name = &((const char *)
-				      pstrings->ptr)[attn_name_offset];
-			attn_type_str = attn_type == ATTN_TYPE_INTERRUPT ?
-					"Interrupt" : "Parity";
+			attn_name = attn_name_base + attn_name_offset;
+			attn_type_str =
+				(attn_type ==
+				 ATTN_TYPE_INTERRUPT ? "Interrupt" :
+				 "Parity");
 			masked_str = reg_result->mask_val & BIT(bit_idx) ?
 				     " [masked]" : "";
 			sts_addr = GET_FIELD(reg_result->data,
@@ -8102,15 +7361,15 @@ enum dbg_status qed_dbg_parse_attn(struct qed_hwfn *p_hwfn,
 			DP_NOTICE(p_hwfn,
 				  "%s (%s) : %s [address 0x%08x, bit %d]%s\n",
 				  block_name, attn_type_str, attn_name,
-				  sts_addr, bit_idx, masked_str);
-
-			bit_idx++;
+				  sts_addr * 4, bit_idx, masked_str);
 		}
 	}
 
 	return DBG_STATUS_OK;
 }
 
+static DEFINE_MUTEX(qed_dbg_lock);
+
 /* Wrapper for unifying the idle_chk and mcp_trace api */
 static enum dbg_status
 qed_print_idle_chk_results_wrapper(struct qed_hwfn *p_hwfn,
@@ -8287,6 +7546,17 @@ static enum dbg_status qed_dbg_dump(struct qed_hwfn *p_hwfn,
 						       &buf_size_dwords);
 	if (rc != DBG_STATUS_OK && rc != DBG_STATUS_NVRAM_GET_IMAGE_FAILED)
 		return rc;
+
+	if (buf_size_dwords > MAX_DBG_FEATURE_SIZE_DWORDS) {
+		feature->buf_size = 0;
+		DP_NOTICE(p_hwfn->cdev,
+			  "Debug feature [\"%s\"] size (0x%x dwords) exceeds maximum size (0x%x dwords)\n",
+			  qed_features_lookup[feature_idx].name,
+			  buf_size_dwords, MAX_DBG_FEATURE_SIZE_DWORDS);
+
+		return DBG_STATUS_OK;
+	}
+
 	feature->buf_size = buf_size_dwords * sizeof(u32);
 	feature->dump_buf = vmalloc(feature->buf_size);
 	if (!feature->dump_buf)
@@ -8487,15 +7757,23 @@ enum debug_print_features {
 	ILT_DUMP = 13,
 };
 
-static u32 qed_calc_regdump_header(enum debug_print_features feature,
+static u32 qed_calc_regdump_header(struct qed_dev *cdev,
+				   enum debug_print_features feature,
 				   int engine, u32 feature_size, u8 omit_engine)
 {
-	/* Insert the engine, feature and mode inside the header and combine it
-	 * with feature size.
-	 */
-	return feature_size | (feature << REGDUMP_HEADER_FEATURE_SHIFT) |
-	       (omit_engine << REGDUMP_HEADER_OMIT_ENGINE_SHIFT) |
-	       (engine << REGDUMP_HEADER_ENGINE_SHIFT);
+	u32 res = 0;
+
+	SET_FIELD(res, REGDUMP_HEADER_SIZE, feature_size);
+	if (res != feature_size)
+		DP_NOTICE(cdev,
+			  "Feature %d is too large (size 0x%x) and will corrupt the dump\n",
+			  feature, feature_size);
+
+	SET_FIELD(res, REGDUMP_HEADER_FEATURE, feature);
+	SET_FIELD(res, REGDUMP_HEADER_OMIT_ENGINE, omit_engine);
+	SET_FIELD(res, REGDUMP_HEADER_ENGINE, engine);
+
+	return res;
 }
 
 int qed_dbg_all_data(struct qed_dev *cdev, void *buffer)
@@ -8511,9 +7789,11 @@ int qed_dbg_all_data(struct qed_dev *cdev, void *buffer)
 	for (i = 0; i < MAX_DBG_GRC_PARAMS; i++)
 		grc_params[i] = dev_data->grc.param_val[i];
 
-	if (cdev->num_hwfns == 1)
+	if (!QED_IS_CMT(cdev))
 		omit_engine = 1;
 
+	mutex_lock(&qed_dbg_lock);
+
 	org_engine = qed_get_debug_engine(cdev);
 	for (cur_engine = 0; cur_engine < cdev->num_hwfns; cur_engine++) {
 		/* Collect idle_chks and grcDump for each hw function */
@@ -8526,7 +7806,7 @@ int qed_dbg_all_data(struct qed_dev *cdev, void *buffer)
 				      REGDUMP_HEADER_SIZE, &feature_size);
 		if (!rc) {
 			*(u32 *)((u8 *)buffer + offset) =
-			    qed_calc_regdump_header(IDLE_CHK, cur_engine,
+			    qed_calc_regdump_header(cdev, IDLE_CHK, cur_engine,
 						    feature_size, omit_engine);
 			offset += (feature_size + REGDUMP_HEADER_SIZE);
 		} else {
@@ -8538,7 +7818,7 @@ int qed_dbg_all_data(struct qed_dev *cdev, void *buffer)
 				      REGDUMP_HEADER_SIZE, &feature_size);
 		if (!rc) {
 			*(u32 *)((u8 *)buffer + offset) =
-			    qed_calc_regdump_header(IDLE_CHK, cur_engine,
+			    qed_calc_regdump_header(cdev, IDLE_CHK, cur_engine,
 						    feature_size, omit_engine);
 			offset += (feature_size + REGDUMP_HEADER_SIZE);
 		} else {
@@ -8550,7 +7830,7 @@ int qed_dbg_all_data(struct qed_dev *cdev, void *buffer)
 				      REGDUMP_HEADER_SIZE, &feature_size);
 		if (!rc) {
 			*(u32 *)((u8 *)buffer + offset) =
-			    qed_calc_regdump_header(REG_FIFO, cur_engine,
+			    qed_calc_regdump_header(cdev, REG_FIFO, cur_engine,
 						    feature_size, omit_engine);
 			offset += (feature_size + REGDUMP_HEADER_SIZE);
 		} else {
@@ -8562,7 +7842,7 @@ int qed_dbg_all_data(struct qed_dev *cdev, void *buffer)
 				      REGDUMP_HEADER_SIZE, &feature_size);
 		if (!rc) {
 			*(u32 *)((u8 *)buffer + offset) =
-			    qed_calc_regdump_header(IGU_FIFO, cur_engine,
+			    qed_calc_regdump_header(cdev, IGU_FIFO, cur_engine,
 						    feature_size, omit_engine);
 			offset += (feature_size + REGDUMP_HEADER_SIZE);
 		} else {
@@ -8575,7 +7855,7 @@ int qed_dbg_all_data(struct qed_dev *cdev, void *buffer)
 						 &feature_size);
 		if (!rc) {
 			*(u32 *)((u8 *)buffer + offset) =
-			    qed_calc_regdump_header(PROTECTION_OVERRIDE,
+			    qed_calc_regdump_header(cdev, PROTECTION_OVERRIDE,
 						    cur_engine,
 						    feature_size, omit_engine);
 			offset += (feature_size + REGDUMP_HEADER_SIZE);
@@ -8590,8 +7870,9 @@ int qed_dbg_all_data(struct qed_dev *cdev, void *buffer)
 					REGDUMP_HEADER_SIZE, &feature_size);
 		if (!rc) {
 			*(u32 *)((u8 *)buffer + offset) =
-			    qed_calc_regdump_header(FW_ASSERTS, cur_engine,
-						    feature_size, omit_engine);
+			    qed_calc_regdump_header(cdev, FW_ASSERTS,
+						    cur_engine, feature_size,
+						    omit_engine);
 			offset += (feature_size + REGDUMP_HEADER_SIZE);
 		} else {
 			DP_ERR(cdev, "qed_dbg_fw_asserts failed. rc = %d\n",
@@ -8605,7 +7886,7 @@ int qed_dbg_all_data(struct qed_dev *cdev, void *buffer)
 					 REGDUMP_HEADER_SIZE, &feature_size);
 			if (!rc) {
 				*(u32 *)((u8 *)buffer + offset) =
-				    qed_calc_regdump_header(ILT_DUMP,
+				    qed_calc_regdump_header(cdev, ILT_DUMP,
 							    cur_engine,
 							    feature_size,
 							    omit_engine);
@@ -8619,11 +7900,15 @@ int qed_dbg_all_data(struct qed_dev *cdev, void *buffer)
 		/* GRC dump - must be last because when mcp stuck it will
 		 * clutter idle_chk, reg_fifo, ...
 		 */
+		for (i = 0; i < MAX_DBG_GRC_PARAMS; i++)
+			dev_data->grc.param_val[i] = grc_params[i];
+
 		rc = qed_dbg_grc(cdev, (u8 *)buffer + offset +
 				 REGDUMP_HEADER_SIZE, &feature_size);
 		if (!rc) {
 			*(u32 *)((u8 *)buffer + offset) =
-			    qed_calc_regdump_header(GRC_DUMP, cur_engine,
+			    qed_calc_regdump_header(cdev, GRC_DUMP,
+						    cur_engine,
 						    feature_size, omit_engine);
 			offset += (feature_size + REGDUMP_HEADER_SIZE);
 		} else {
@@ -8632,12 +7917,13 @@ int qed_dbg_all_data(struct qed_dev *cdev, void *buffer)
 	}
 
 	qed_set_debug_engine(cdev, org_engine);
+
 	/* mcp_trace */
 	rc = qed_dbg_mcp_trace(cdev, (u8 *)buffer + offset +
 			       REGDUMP_HEADER_SIZE, &feature_size);
 	if (!rc) {
 		*(u32 *)((u8 *)buffer + offset) =
-		    qed_calc_regdump_header(MCP_TRACE, cur_engine,
+		    qed_calc_regdump_header(cdev, MCP_TRACE, cur_engine,
 					    feature_size, omit_engine);
 		offset += (feature_size + REGDUMP_HEADER_SIZE);
 	} else {
@@ -8646,11 +7932,12 @@ int qed_dbg_all_data(struct qed_dev *cdev, void *buffer)
 
 	/* nvm cfg1 */
 	rc = qed_dbg_nvm_image(cdev,
-			       (u8 *)buffer + offset + REGDUMP_HEADER_SIZE,
-			       &feature_size, QED_NVM_IMAGE_NVM_CFG1);
+			       (u8 *)buffer + offset +
+			       REGDUMP_HEADER_SIZE, &feature_size,
+			       QED_NVM_IMAGE_NVM_CFG1);
 	if (!rc) {
 		*(u32 *)((u8 *)buffer + offset) =
-		    qed_calc_regdump_header(NVM_CFG1, cur_engine,
+		    qed_calc_regdump_header(cdev, NVM_CFG1, cur_engine,
 					    feature_size, omit_engine);
 		offset += (feature_size + REGDUMP_HEADER_SIZE);
 	} else if (rc != -ENOENT) {
@@ -8665,7 +7952,7 @@ int qed_dbg_all_data(struct qed_dev *cdev, void *buffer)
 			       &feature_size, QED_NVM_IMAGE_DEFAULT_CFG);
 	if (!rc) {
 		*(u32 *)((u8 *)buffer + offset) =
-		    qed_calc_regdump_header(DEFAULT_CFG, cur_engine,
+		    qed_calc_regdump_header(cdev, DEFAULT_CFG, cur_engine,
 					    feature_size, omit_engine);
 		offset += (feature_size + REGDUMP_HEADER_SIZE);
 	} else if (rc != -ENOENT) {
@@ -8681,8 +7968,8 @@ int qed_dbg_all_data(struct qed_dev *cdev, void *buffer)
 			       &feature_size, QED_NVM_IMAGE_NVM_META);
 	if (!rc) {
 		*(u32 *)((u8 *)buffer + offset) =
-		    qed_calc_regdump_header(NVM_META, cur_engine,
-					    feature_size, omit_engine);
+			qed_calc_regdump_header(cdev, NVM_META, cur_engine,
+						feature_size, omit_engine);
 		offset += (feature_size + REGDUMP_HEADER_SIZE);
 	} else if (rc != -ENOENT) {
 		DP_ERR(cdev,
@@ -8696,7 +7983,7 @@ int qed_dbg_all_data(struct qed_dev *cdev, void *buffer)
 			       QED_NVM_IMAGE_MDUMP);
 	if (!rc) {
 		*(u32 *)((u8 *)buffer + offset) =
-			qed_calc_regdump_header(MDUMP, cur_engine,
+			qed_calc_regdump_header(cdev, MDUMP, cur_engine,
 						feature_size, omit_engine);
 		offset += (feature_size + REGDUMP_HEADER_SIZE);
 	} else if (rc != -ENOENT) {
@@ -8705,6 +7992,8 @@ int qed_dbg_all_data(struct qed_dev *cdev, void *buffer)
 		       QED_NVM_IMAGE_MDUMP, "QED_NVM_IMAGE_MDUMP", rc);
 	}
 
+	mutex_unlock(&qed_dbg_lock);
+
 	return 0;
 }
 
@@ -8715,6 +8004,7 @@ int qed_dbg_all_data_size(struct qed_dev *cdev)
 	u32 regs_len = 0, image_len = 0, ilt_len = 0, total_ilt_len = 0;
 	u8 cur_engine, org_engine;
 
+	cdev->disable_ilt_dump = false;
 	org_engine = qed_get_debug_engine(cdev);
 	for (cur_engine = 0; cur_engine < cdev->num_hwfns; cur_engine++) {
 		/* Engine specific */
@@ -8806,9 +8096,8 @@ int qed_dbg_feature_size(struct qed_dev *cdev, enum qed_dbg_features feature)
 {
 	struct qed_hwfn *p_hwfn =
 		&cdev->hwfns[cdev->dbg_params.engine_for_debug];
+	struct qed_dbg_feature *qed_feature = &cdev->dbg_features[feature];
 	struct qed_ptt *p_ptt = qed_ptt_acquire(p_hwfn);
-	struct qed_dbg_feature *qed_feature =
-		&cdev->dbg_params.features[feature];
 	u32 buf_size_dwords;
 	enum dbg_status rc;
 
@@ -8843,14 +8132,21 @@ void qed_set_debug_engine(struct qed_dev *cdev, int engine_number)
 
 void qed_dbg_pf_init(struct qed_dev *cdev)
 {
-	const u8 *dbg_values;
+	const u8 *dbg_values = NULL;
+	int i;
 
 	/* Debug values are after init values.
 	 * The offset is the first dword of the file.
 	 */
 	dbg_values = cdev->firmware->data + *(u32 *)cdev->firmware->data;
-	qed_dbg_set_bin_ptr((u8 *)dbg_values);
-	qed_dbg_user_set_bin_ptr((u8 *)dbg_values);
+
+	for_each_hwfn(cdev, i) {
+		qed_dbg_set_bin_ptr(&cdev->hwfns[i], dbg_values);
+		qed_dbg_user_set_bin_ptr(&cdev->hwfns[i], dbg_values);
+	}
+
+	/* Set the hwfn to be 0 as default */
+	cdev->dbg_params.engine_for_debug = 0;
 }
 
 void qed_dbg_pf_exit(struct qed_dev *cdev)
@@ -8858,11 +8154,11 @@ void qed_dbg_pf_exit(struct qed_dev *cdev)
 	struct qed_dbg_feature *feature = NULL;
 	enum qed_dbg_features feature_idx;
 
-	/* Debug features' buffers may be allocated if debug feature was used
-	 * but dump wasn't called.
+	/* debug features' buffers may be allocated if debug feature was used
+	 * but dump wasn't called
 	 */
 	for (feature_idx = 0; feature_idx < DBG_FEATURE_NUM; feature_idx++) {
-		feature = &cdev->dbg_params.features[feature_idx];
+		feature = &cdev->dbg_features[feature_idx];
 		if (feature->dump_buf) {
 			vfree(feature->dump_buf);
 			feature->dump_buf = NULL;
diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c
index 3fb73ce8c1d6..7912911337d4 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_dev.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c
@@ -2346,7 +2346,7 @@ int qed_resc_alloc(struct qed_dev *cdev)
 		if (rc)
 			goto alloc_err;
 
-		rc = qed_dbg_alloc_user_data(p_hwfn);
+		rc = qed_dbg_alloc_user_data(p_hwfn, &p_hwfn->dbg_user_info);
 		if (rc)
 			goto alloc_err;
 	}
diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
index e7e3f7b31ee0..4597015b8bff 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
@@ -1860,98 +1860,6 @@ struct virt_mem_desc {
 /* Debug Tools HSI constants and macros */
 /****************************************/
 
-enum block_addr {
-	GRCBASE_GRC = 0x50000,
-	GRCBASE_MISCS = 0x9000,
-	GRCBASE_MISC = 0x8000,
-	GRCBASE_DBU = 0xa000,
-	GRCBASE_PGLUE_B = 0x2a8000,
-	GRCBASE_CNIG = 0x218000,
-	GRCBASE_CPMU = 0x30000,
-	GRCBASE_NCSI = 0x40000,
-	GRCBASE_OPTE = 0x53000,
-	GRCBASE_BMB = 0x540000,
-	GRCBASE_PCIE = 0x54000,
-	GRCBASE_MCP = 0xe00000,
-	GRCBASE_MCP2 = 0x52000,
-	GRCBASE_PSWHST = 0x2a0000,
-	GRCBASE_PSWHST2 = 0x29e000,
-	GRCBASE_PSWRD = 0x29c000,
-	GRCBASE_PSWRD2 = 0x29d000,
-	GRCBASE_PSWWR = 0x29a000,
-	GRCBASE_PSWWR2 = 0x29b000,
-	GRCBASE_PSWRQ = 0x280000,
-	GRCBASE_PSWRQ2 = 0x240000,
-	GRCBASE_PGLCS = 0x0,
-	GRCBASE_DMAE = 0xc000,
-	GRCBASE_PTU = 0x560000,
-	GRCBASE_TCM = 0x1180000,
-	GRCBASE_MCM = 0x1200000,
-	GRCBASE_UCM = 0x1280000,
-	GRCBASE_XCM = 0x1000000,
-	GRCBASE_YCM = 0x1080000,
-	GRCBASE_PCM = 0x1100000,
-	GRCBASE_QM = 0x2f0000,
-	GRCBASE_TM = 0x2c0000,
-	GRCBASE_DORQ = 0x100000,
-	GRCBASE_BRB = 0x340000,
-	GRCBASE_SRC = 0x238000,
-	GRCBASE_PRS = 0x1f0000,
-	GRCBASE_TSDM = 0xfb0000,
-	GRCBASE_MSDM = 0xfc0000,
-	GRCBASE_USDM = 0xfd0000,
-	GRCBASE_XSDM = 0xf80000,
-	GRCBASE_YSDM = 0xf90000,
-	GRCBASE_PSDM = 0xfa0000,
-	GRCBASE_TSEM = 0x1700000,
-	GRCBASE_MSEM = 0x1800000,
-	GRCBASE_USEM = 0x1900000,
-	GRCBASE_XSEM = 0x1400000,
-	GRCBASE_YSEM = 0x1500000,
-	GRCBASE_PSEM = 0x1600000,
-	GRCBASE_RSS = 0x238800,
-	GRCBASE_TMLD = 0x4d0000,
-	GRCBASE_MULD = 0x4e0000,
-	GRCBASE_YULD = 0x4c8000,
-	GRCBASE_XYLD = 0x4c0000,
-	GRCBASE_PTLD = 0x5a0000,
-	GRCBASE_YPLD = 0x5c0000,
-	GRCBASE_PRM = 0x230000,
-	GRCBASE_PBF_PB1 = 0xda0000,
-	GRCBASE_PBF_PB2 = 0xda4000,
-	GRCBASE_RPB = 0x23c000,
-	GRCBASE_BTB = 0xdb0000,
-	GRCBASE_PBF = 0xd80000,
-	GRCBASE_RDIF = 0x300000,
-	GRCBASE_TDIF = 0x310000,
-	GRCBASE_CDU = 0x580000,
-	GRCBASE_CCFC = 0x2e0000,
-	GRCBASE_TCFC = 0x2d0000,
-	GRCBASE_IGU = 0x180000,
-	GRCBASE_CAU = 0x1c0000,
-	GRCBASE_RGFS = 0xf00000,
-	GRCBASE_RGSRC = 0x320000,
-	GRCBASE_TGFS = 0xd00000,
-	GRCBASE_TGSRC = 0x322000,
-	GRCBASE_UMAC = 0x51000,
-	GRCBASE_XMAC = 0x210000,
-	GRCBASE_DBG = 0x10000,
-	GRCBASE_NIG = 0x500000,
-	GRCBASE_WOL = 0x600000,
-	GRCBASE_BMBN = 0x610000,
-	GRCBASE_IPC = 0x20000,
-	GRCBASE_NWM = 0x800000,
-	GRCBASE_NWS = 0x700000,
-	GRCBASE_MS = 0x6a0000,
-	GRCBASE_PHY_PCIE = 0x620000,
-	GRCBASE_LED = 0x6b8000,
-	GRCBASE_AVS_WRAP = 0x6b0000,
-	GRCBASE_PXPREQBUS = 0x56000,
-	GRCBASE_MISC_AEU = 0x8000,
-	GRCBASE_BAR0_MAP = 0x1c00000,
-	MAX_BLOCK_ADDR
-};
-
 enum block_id {
 	BLOCK_GRC,
 	BLOCK_MISCS,
@@ -2006,8 +1914,6 @@ enum block_id {
 	BLOCK_MULD,
 	BLOCK_YULD,
 	BLOCK_XYLD,
-	BLOCK_PTLD,
-	BLOCK_YPLD,
 	BLOCK_PRM,
 	BLOCK_PBF_PB1,
 	BLOCK_PBF_PB2,
@@ -2021,12 +1927,9 @@ enum block_id {
 	BLOCK_TCFC,
 	BLOCK_IGU,
 	BLOCK_CAU,
-	BLOCK_RGFS,
-	BLOCK_RGSRC,
-	BLOCK_TGFS,
-	BLOCK_TGSRC,
 	BLOCK_UMAC,
 	BLOCK_XMAC,
+	BLOCK_MSTAT,
 	BLOCK_DBG,
 	BLOCK_NIG,
 	BLOCK_WOL,
@@ -2039,8 +1942,17 @@ enum block_id {
 	BLOCK_LED,
 	BLOCK_AVS_WRAP,
 	BLOCK_PXPREQBUS,
-	BLOCK_MISC_AEU,
 	BLOCK_BAR0_MAP,
+	BLOCK_MCP_FIO,
+	BLOCK_LAST_INIT,
+	BLOCK_PRS_FC,
+	BLOCK_PBF_FC,
+	BLOCK_NIG_LB_FC,
+	BLOCK_NIG_LB_FC_PLLH,
+	BLOCK_NIG_TX_FC_PLLH,
+	BLOCK_NIG_TX_FC,
+	BLOCK_NIG_RX_FC_PLLH,
+	BLOCK_NIG_RX_FC,
 	MAX_BLOCK_ID
 };
 
@@ -2057,10 +1969,13 @@ enum bin_dbg_buffer_type {
 	BIN_BUF_DBG_ATTN_REGS,
 	BIN_BUF_DBG_ATTN_INDEXES,
 	BIN_BUF_DBG_ATTN_NAME_OFFSETS,
-	BIN_BUF_DBG_BUS_BLOCKS,
+	BIN_BUF_DBG_BLOCKS,
+	BIN_BUF_DBG_BLOCKS_CHIP_DATA,
 	BIN_BUF_DBG_BUS_LINES,
-	BIN_BUF_DBG_BUS_BLOCKS_USER_DATA,
+	BIN_BUF_DBG_BLOCKS_USER_DATA,
+	BIN_BUF_DBG_BLOCKS_CHIP_USER_DATA,
 	BIN_BUF_DBG_BUS_LINE_NAME_OFFSETS,
+	BIN_BUF_DBG_RESET_REGS,
 	BIN_BUF_DBG_PARSING_STRINGS,
 	MAX_BIN_DBG_BUFFER_TYPE
 };
@@ -2144,20 +2059,54 @@ enum dbg_attn_type {
 	MAX_DBG_ATTN_TYPE
 };
 
-/* Debug Bus block data */
-struct dbg_bus_block {
-	u8 num_of_lines;
-	u8 has_latency_events;
-	u16 lines_offset;
+/* Block debug data */
+struct dbg_block {
+	u8 name[15];
+	u8 associated_storm_letter;
 };
 
-/* Debug Bus block user data */
-struct dbg_bus_block_user_data {
-	u8 num_of_lines;
+/* Chip-specific block debug data */
+struct dbg_block_chip {
+	u8 flags;
+#define DBG_BLOCK_CHIP_IS_REMOVED_MASK		 0x1
+#define DBG_BLOCK_CHIP_IS_REMOVED_SHIFT		 0
+#define DBG_BLOCK_CHIP_HAS_RESET_REG_MASK	 0x1
+#define DBG_BLOCK_CHIP_HAS_RESET_REG_SHIFT	 1
+#define DBG_BLOCK_CHIP_UNRESET_BEFORE_DUMP_MASK  0x1
+#define DBG_BLOCK_CHIP_UNRESET_BEFORE_DUMP_SHIFT 2
+#define DBG_BLOCK_CHIP_HAS_DBG_BUS_MASK		 0x1
+#define DBG_BLOCK_CHIP_HAS_DBG_BUS_SHIFT	 3
+#define DBG_BLOCK_CHIP_HAS_LATENCY_EVENTS_MASK	 0x1
+#define DBG_BLOCK_CHIP_HAS_LATENCY_EVENTS_SHIFT  4
+#define DBG_BLOCK_CHIP_RESERVED0_MASK		 0x7
+#define DBG_BLOCK_CHIP_RESERVED0_SHIFT		 5
+	u8 dbg_client_id;
+	u8 reset_reg_id;
+	u8 reset_reg_bit_offset;
+	struct dbg_mode_hdr dbg_bus_mode;
+	u16 reserved1;
+	u8 reserved2;
+	u8 num_of_dbg_bus_lines;
+	u16 dbg_bus_lines_offset;
+	u32 dbg_select_reg_addr;
+	u32 dbg_dword_enable_reg_addr;
+	u32 dbg_shift_reg_addr;
+	u32 dbg_force_valid_reg_addr;
+	u32 dbg_force_frame_reg_addr;
+};
+
+/* Chip-specific block user debug data */
+struct dbg_block_chip_user {
+	u8 num_of_dbg_bus_lines;
 	u8 has_latency_events;
 	u16 names_offset;
 };
 
+/* Block user debug data */
+struct dbg_block_user {
+	u8 name[16];
+};
+
 /* Block Debug line data */
 struct dbg_bus_line {
 	u8 data;
@@ -2310,22 +2259,33 @@ enum dbg_idle_chk_severity_types {
 	MAX_DBG_IDLE_CHK_SEVERITY_TYPES
 };
 
+/* Reset register */
+struct dbg_reset_reg {
+	u32 data;
+#define DBG_RESET_REG_ADDR_MASK        0xFFFFFF
+#define DBG_RESET_REG_ADDR_SHIFT       0
+#define DBG_RESET_REG_IS_REMOVED_MASK  0x1
+#define DBG_RESET_REG_IS_REMOVED_SHIFT 24
+#define DBG_RESET_REG_RESERVED_MASK    0x7F
+#define DBG_RESET_REG_RESERVED_SHIFT   25
+};
+
 /* Debug Bus block data */
 struct dbg_bus_block_data {
-	u16 data;
-#define DBG_BUS_BLOCK_DATA_ENABLE_MASK_MASK		0xF
-#define DBG_BUS_BLOCK_DATA_ENABLE_MASK_SHIFT		0
-#define DBG_BUS_BLOCK_DATA_RIGHT_SHIFT_MASK		0xF
-#define DBG_BUS_BLOCK_DATA_RIGHT_SHIFT_SHIFT		4
-#define DBG_BUS_BLOCK_DATA_FORCE_VALID_MASK_MASK	0xF
-#define DBG_BUS_BLOCK_DATA_FORCE_VALID_MASK_SHIFT	8
-#define DBG_BUS_BLOCK_DATA_FORCE_FRAME_MASK_MASK	0xF
-#define DBG_BUS_BLOCK_DATA_FORCE_FRAME_MASK_SHIFT	12
+	u8 enable_mask;
+	u8 right_shift;
+	u8 force_valid_mask;
+	u8 force_frame_mask;
+	u8 dword_mask;
 	u8 line_num;
 	u8 hw_id;
+	u8 flags;
+#define DBG_BUS_BLOCK_DATA_IS_256B_LINE_MASK  0x1
+#define DBG_BUS_BLOCK_DATA_IS_256B_LINE_SHIFT 0
+#define DBG_BUS_BLOCK_DATA_RESERVED_MASK      0x7F
+#define DBG_BUS_BLOCK_DATA_RESERVED_SHIFT     1
 };
 
-/* Debug Bus Clients */
 enum dbg_bus_clients {
 	DBG_BUS_CLIENT_RBCN,
 	DBG_BUS_CLIENT_RBCP,
@@ -2366,11 +2326,10 @@ enum dbg_bus_constraint_ops {
 
 /* Debug Bus trigger state data */
 struct dbg_bus_trigger_state_data {
-	u8 data;
-#define DBG_BUS_TRIGGER_STATE_DATA_BLOCK_SHIFTED_ENABLE_MASK_MASK	0xF
-#define DBG_BUS_TRIGGER_STATE_DATA_BLOCK_SHIFTED_ENABLE_MASK_SHIFT	0
-#define DBG_BUS_TRIGGER_STATE_DATA_CONSTRAINT_DWORD_MASK_MASK		0xF
-#define DBG_BUS_TRIGGER_STATE_DATA_CONSTRAINT_DWORD_MASK_SHIFT		4
+	u8 msg_len;
+	u8 constraint_dword_mask;
+	u8 storm_id;
+	u8 reserved;
 };
 
 /* Debug Bus memory address */
@@ -2420,8 +2379,7 @@ struct dbg_bus_storm_data {
 struct dbg_bus_data {
 	u32 app_version;
 	u8 state;
-	u8 hw_dwords;
-	u16 hw_id_mask;
+	u8 mode_256b_en;
 	u8 num_enabled_blocks;
 	u8 num_enabled_storms;
 	u8 target;
@@ -2432,67 +2390,21 @@ struct dbg_bus_data {
 	u8 adding_filter;
 	u8 filter_pre_trigger;
 	u8 filter_post_trigger;
-	u16 reserved;
 	u8 trigger_en;
-	struct dbg_bus_trigger_state_data trigger_states[3];
+	u8 filter_constraint_dword_mask;
 	u8 next_trigger_state;
 	u8 next_constraint_id;
-	u8 unify_inputs;
+	struct dbg_bus_trigger_state_data trigger_states[3];
+	u8 filter_msg_len;
 	u8 rcv_from_other_engine;
+	u8 blocks_dword_mask;
+	u8 blocks_dword_overlap;
+	u32 hw_id_mask;
 	struct dbg_bus_pci_buf_data pci_buf;
-	struct dbg_bus_block_data blocks[88];
+	struct dbg_bus_block_data blocks[132];
 	struct dbg_bus_storm_data storms[6];
 };
 
-/* Debug bus filter types */
-enum dbg_bus_filter_types {
-	DBG_BUS_FILTER_TYPE_OFF,
-	DBG_BUS_FILTER_TYPE_PRE,
-	DBG_BUS_FILTER_TYPE_POST,
-	DBG_BUS_FILTER_TYPE_ON,
-	MAX_DBG_BUS_FILTER_TYPES
-};
-
-/* Debug bus frame modes */
-enum dbg_bus_frame_modes {
-	DBG_BUS_FRAME_MODE_0HW_4ST = 0, /* 0 HW dwords, 4 Storm dwords */
-	DBG_BUS_FRAME_MODE_4HW_0ST = 3, /* 4 HW dwords, 0 Storm dwords */
-	DBG_BUS_FRAME_MODE_8HW_0ST = 4, /* 8 HW dwords, 0 Storm dwords */
-	MAX_DBG_BUS_FRAME_MODES
-};
-
-/* Debug bus other engine mode */
-enum dbg_bus_other_engine_modes {
-	DBG_BUS_OTHER_ENGINE_MODE_NONE,
-	DBG_BUS_OTHER_ENGINE_MODE_DOUBLE_BW_TX,
-	DBG_BUS_OTHER_ENGINE_MODE_DOUBLE_BW_RX,
-	DBG_BUS_OTHER_ENGINE_MODE_CROSS_ENGINE_TX,
-	DBG_BUS_OTHER_ENGINE_MODE_CROSS_ENGINE_RX,
-	MAX_DBG_BUS_OTHER_ENGINE_MODES
-};
-
-/* Debug bus post-trigger recording types */
-enum dbg_bus_post_trigger_types {
-	DBG_BUS_POST_TRIGGER_RECORD,
-	DBG_BUS_POST_TRIGGER_DROP,
-	MAX_DBG_BUS_POST_TRIGGER_TYPES
-};
-
-/* Debug bus pre-trigger recording types */
-enum dbg_bus_pre_trigger_types {
-	DBG_BUS_PRE_TRIGGER_START_FROM_ZERO,
-	DBG_BUS_PRE_TRIGGER_NUM_CHUNKS,
-	DBG_BUS_PRE_TRIGGER_DROP,
-	MAX_DBG_BUS_PRE_TRIGGER_TYPES
-};
-
-/* Debug bus SEMI frame modes */
-enum dbg_bus_semi_frame_modes {
-	DBG_BUS_SEMI_FRAME_MODE_0SLOW_4FAST = 0,
-	DBG_BUS_SEMI_FRAME_MODE_4SLOW_0FAST = 3,
-	MAX_DBG_BUS_SEMI_FRAME_MODES
-};
-
 /* Debug bus states */
 enum dbg_bus_states {
 	DBG_BUS_STATE_IDLE,
@@ -2510,7 +2422,9 @@ enum dbg_bus_storm_modes {
 	DBG_BUS_STORM_MODE_DRA_W,
 	DBG_BUS_STORM_MODE_LD_ST_ADDR,
 	DBG_BUS_STORM_MODE_DRA_FSM,
+	DBG_BUS_STORM_MODE_FAST_DBGMUX,
 	DBG_BUS_STORM_MODE_RH,
+	DBG_BUS_STORM_MODE_RH_WITH_STORE,
 	DBG_BUS_STORM_MODE_FOC,
 	DBG_BUS_STORM_MODE_EXT_STORE,
 	MAX_DBG_BUS_STORM_MODES
@@ -2551,13 +2465,13 @@ enum dbg_grc_params {
 	DBG_GRC_PARAM_DUMP_CAU,
 	DBG_GRC_PARAM_DUMP_QM,
 	DBG_GRC_PARAM_DUMP_MCP,
-	DBG_GRC_PARAM_MCP_TRACE_META_SIZE,
+	DBG_GRC_PARAM_DUMP_DORQ,
 	DBG_GRC_PARAM_DUMP_CFC,
 	DBG_GRC_PARAM_DUMP_IGU,
 	DBG_GRC_PARAM_DUMP_BRB,
 	DBG_GRC_PARAM_DUMP_BTB,
 	DBG_GRC_PARAM_DUMP_BMB,
-	DBG_GRC_PARAM_DUMP_NIG,
+	DBG_GRC_PARAM_RESERVD1,
 	DBG_GRC_PARAM_DUMP_MULD,
 	DBG_GRC_PARAM_DUMP_PRS,
 	DBG_GRC_PARAM_DUMP_DMAE,
@@ -2566,8 +2480,8 @@ enum dbg_grc_params {
 	DBG_GRC_PARAM_DUMP_DIF,
 	DBG_GRC_PARAM_DUMP_STATIC,
 	DBG_GRC_PARAM_UNSTALL,
-	DBG_GRC_PARAM_NUM_LCIDS,
-	DBG_GRC_PARAM_NUM_LTIDS,
+	DBG_GRC_PARAM_RESERVED2,
+	DBG_GRC_PARAM_MCP_TRACE_META_SIZE,
 	DBG_GRC_PARAM_EXCLUDE_ALL,
 	DBG_GRC_PARAM_CRASH,
 	DBG_GRC_PARAM_PARITY_SAFE,
@@ -2583,19 +2497,6 @@ enum dbg_grc_params {
 	MAX_DBG_GRC_PARAMS
 };
 
-/* Debug reset registers */
-enum dbg_reset_regs {
-	DBG_RESET_REG_MISCS_PL_UA,
-	DBG_RESET_REG_MISCS_PL_HV,
-	DBG_RESET_REG_MISCS_PL_HV_2,
-	DBG_RESET_REG_MISC_PL_UA,
-	DBG_RESET_REG_MISC_PL_HV,
-	DBG_RESET_REG_MISC_PL_PDA_VMAIN_1,
-	DBG_RESET_REG_MISC_PL_PDA_VMAIN_2,
-	DBG_RESET_REG_MISC_PL_PDA_VAUX,
-	MAX_DBG_RESET_REGS
-};
-
 /* Debug status codes */
 enum dbg_status {
 	DBG_STATUS_OK,
@@ -2607,15 +2508,15 @@ enum dbg_status {
 	DBG_STATUS_INVALID_PCI_BUF_SIZE,
 	DBG_STATUS_PCI_BUF_ALLOC_FAILED,
 	DBG_STATUS_PCI_BUF_NOT_ALLOCATED,
-	DBG_STATUS_TOO_MANY_INPUTS,
-	DBG_STATUS_INPUT_OVERLAP,
-	DBG_STATUS_HW_ONLY_RECORDING,
+	DBG_STATUS_INVALID_FILTER_TRIGGER_DWORDS,
+	DBG_STATUS_NO_MATCHING_FRAMING_MODE,
+	DBG_STATUS_VFC_READ_ERROR,
 	DBG_STATUS_STORM_ALREADY_ENABLED,
 	DBG_STATUS_STORM_NOT_ENABLED,
 	DBG_STATUS_BLOCK_ALREADY_ENABLED,
 	DBG_STATUS_BLOCK_NOT_ENABLED,
 	DBG_STATUS_NO_INPUT_ENABLED,
-	DBG_STATUS_NO_FILTER_TRIGGER_64B,
+	DBG_STATUS_NO_FILTER_TRIGGER_256B,
 	DBG_STATUS_FILTER_ALREADY_ENABLED,
 	DBG_STATUS_TRIGGER_ALREADY_ENABLED,
 	DBG_STATUS_TRIGGER_NOT_ENABLED,
@@ -2640,7 +2541,7 @@ enum dbg_status {
 	DBG_STATUS_MCP_TRACE_NO_META,
 	DBG_STATUS_MCP_COULD_NOT_HALT,
 	DBG_STATUS_MCP_COULD_NOT_RESUME,
-	DBG_STATUS_RESERVED2,
+	DBG_STATUS_RESERVED0,
 	DBG_STATUS_SEMI_FIFO_NOT_EMPTY,
 	DBG_STATUS_IGU_FIFO_BAD_DATA,
 	DBG_STATUS_MCP_COULD_NOT_MASK_PRTY,
@@ -2648,10 +2549,15 @@ enum dbg_status {
 	DBG_STATUS_REG_FIFO_BAD_DATA,
 	DBG_STATUS_PROTECTION_OVERRIDE_BAD_DATA,
 	DBG_STATUS_DBG_ARRAY_NOT_SET,
-	DBG_STATUS_FILTER_BUG,
+	DBG_STATUS_RESERVED1,
 	DBG_STATUS_NON_MATCHING_LINES,
-	DBG_STATUS_INVALID_TRIGGER_DWORD_OFFSET,
+	DBG_STATUS_INSUFFICIENT_HW_IDS,
 	DBG_STATUS_DBG_BUS_IN_USE,
+	DBG_STATUS_INVALID_STORM_DBG_MODE,
+	DBG_STATUS_OTHER_ENGINE_BB_ONLY,
+	DBG_STATUS_FILTER_SINGLE_HW_ID,
+	DBG_STATUS_TRIGGER_SINGLE_HW_ID,
+	DBG_STATUS_MISSING_TRIGGER_STATE_STORM,
 	MAX_DBG_STATUS
 };
 
@@ -2687,9 +2593,9 @@ struct dbg_tools_data {
 	struct dbg_bus_data bus;
 	struct idle_chk_data idle_chk;
 	u8 mode_enable[40];
-	u8 block_in_reset[88];
+	u8 block_in_reset[132];
 	u8 chip_id;
-	u8 platform_id;
+	u8 hw_type;
 	u8 num_ports;
 	u8 num_pfs_per_port;
 	u8 num_vfs;
@@ -2808,7 +2714,6 @@ struct init_qm_vport_params {
 enum chip_ids {
 	CHIP_BB,
 	CHIP_K2,
-	CHIP_RESERVED,
 	MAX_CHIP_IDS
 };
 
@@ -3135,9 +3040,11 @@ struct iro {
  * @brief qed_dbg_set_bin_ptr - Sets a pointer to the binary data with debug
  *	arrays.
  *
+ * @param p_hwfn -	    HW device data
  * @param bin_ptr - a pointer to the binary data with debug arrays.
  */
-enum dbg_status qed_dbg_set_bin_ptr(const u8 * const bin_ptr);
+enum dbg_status qed_dbg_set_bin_ptr(struct qed_hwfn *p_hwfn,
+				    const u8 * const bin_ptr);
 
 /**
  * @brief qed_read_regs - Reads registers into a buffer (using GRC).
@@ -3181,7 +3088,6 @@ bool qed_read_fw_info(struct qed_hwfn *p_hwfn,
  *	- val is outside the allowed boundaries
  */
 enum dbg_status qed_dbg_grc_config(struct qed_hwfn *p_hwfn,
-				   struct qed_ptt *p_ptt,
 				   enum dbg_grc_params grc_param, u32 val);
 
 /**
@@ -3502,20 +3408,36 @@ enum dbg_status qed_dbg_print_attn(struct qed_hwfn *p_hwfn,
 struct mcp_trace_format {
 	u32 data;
 #define MCP_TRACE_FORMAT_MODULE_MASK	0x0000ffff
-#define MCP_TRACE_FORMAT_MODULE_SHIFT	0
+#define MCP_TRACE_FORMAT_MODULE_OFFSET	0
 #define MCP_TRACE_FORMAT_LEVEL_MASK	0x00030000
-#define MCP_TRACE_FORMAT_LEVEL_SHIFT	16
+#define MCP_TRACE_FORMAT_LEVEL_OFFSET	16
 #define MCP_TRACE_FORMAT_P1_SIZE_MASK	0x000c0000
-#define MCP_TRACE_FORMAT_P1_SIZE_SHIFT	18
+#define MCP_TRACE_FORMAT_P1_SIZE_OFFSET 18
 #define MCP_TRACE_FORMAT_P2_SIZE_MASK	0x00300000
-#define MCP_TRACE_FORMAT_P2_SIZE_SHIFT	20
+#define MCP_TRACE_FORMAT_P2_SIZE_OFFSET 20
 #define MCP_TRACE_FORMAT_P3_SIZE_MASK	0x00c00000
-#define MCP_TRACE_FORMAT_P3_SIZE_SHIFT	22
+#define MCP_TRACE_FORMAT_P3_SIZE_OFFSET 22
 #define MCP_TRACE_FORMAT_LEN_MASK	0xff000000
-#define MCP_TRACE_FORMAT_LEN_SHIFT	24
+#define MCP_TRACE_FORMAT_LEN_OFFSET	24
+
 	char *format_str;
 };
 
+/* MCP Trace Meta data structure */
+struct mcp_trace_meta {
+	u32 modules_num;
+	char **modules;
+	u32 formats_num;
+	struct mcp_trace_format *formats;
+	bool is_allocated;
+};
+
+/* Debug Tools user data */
+struct dbg_tools_user_data {
+	struct mcp_trace_meta mcp_trace_meta;
+	const u32 *mcp_trace_user_meta_buf;
+};
+
 /******************************** Constants **********************************/
 
 #define MAX_NAME_LEN	16
@@ -3526,16 +3448,20 @@ struct mcp_trace_format {
  * @brief qed_dbg_user_set_bin_ptr - Sets a pointer to the binary data with
  *	debug arrays.
  *
+ * @param p_hwfn - HW device data
  * @param bin_ptr - a pointer to the binary data with debug arrays.
  */
-enum dbg_status qed_dbg_user_set_bin_ptr(const u8 * const bin_ptr);
+enum dbg_status qed_dbg_user_set_bin_ptr(struct qed_hwfn *p_hwfn,
+					 const u8 * const bin_ptr);
 
 /**
  * @brief qed_dbg_alloc_user_data - Allocates user debug data.
  *
  * @param p_hwfn -		 HW device data
+ * @param user_data_ptr - OUT: a pointer to the allocated memory.
  */
-enum dbg_status qed_dbg_alloc_user_data(struct qed_hwfn *p_hwfn);
+enum dbg_status qed_dbg_alloc_user_data(struct qed_hwfn *p_hwfn,
+					void **user_data_ptr);
 
 /**
  * @brief qed_dbg_get_status_str - Returns a string for the specified status.
@@ -3808,271 +3734,6 @@ enum dbg_status qed_print_fw_asserts_results(struct qed_hwfn *p_hwfn,
 enum dbg_status qed_dbg_parse_attn(struct qed_hwfn *p_hwfn,
 				   struct dbg_attn_block_result *results);
 
-/* Debug Bus blocks */
-static const u32 dbg_bus_blocks[] = {
-	0x0000000f,		/* grc, bb, 15 lines */
-	0x0000000f,		/* grc, k2, 15 lines */
-	0x00000000,
-	0x00000000,		/* miscs, bb, 0 lines */
-	0x00000000,		/* miscs, k2, 0 lines */
-	0x00000000,
-	0x00000000,		/* misc, bb, 0 lines */
-	0x00000000,		/* misc, k2, 0 lines */
-	0x00000000,
-	0x00000000,		/* dbu, bb, 0 lines */
-	0x00000000,		/* dbu, k2, 0 lines */
-	0x00000000,
-	0x000f0127,		/* pglue_b, bb, 39 lines */
-	0x0036012a,		/* pglue_b, k2, 42 lines */
-	0x00000000,
-	0x00000000,		/* cnig, bb, 0 lines */
-	0x00120102,		/* cnig, k2, 2 lines */
-	0x00000000,
-	0x00000000,		/* cpmu, bb, 0 lines */
-	0x00000000,		/* cpmu, k2, 0 lines */
-	0x00000000,
-	0x00000001,		/* ncsi, bb, 1 lines */
-	0x00000001,		/* ncsi, k2, 1 lines */
-	0x00000000,
-	0x00000000,		/* opte, bb, 0 lines */
-	0x00000000,		/* opte, k2, 0 lines */
-	0x00000000,
-	0x00600085,		/* bmb, bb, 133 lines */
-	0x00600085,		/* bmb, k2, 133 lines */
-	0x00000000,
-	0x00000000,		/* pcie, bb, 0 lines */
-	0x00e50033,		/* pcie, k2, 51 lines */
-	0x00000000,
-	0x00000000,		/* mcp, bb, 0 lines */
-	0x00000000,		/* mcp, k2, 0 lines */
-	0x00000000,
-	0x01180009,		/* mcp2, bb, 9 lines */
-	0x01180009,		/* mcp2, k2, 9 lines */
-	0x00000000,
-	0x01210104,		/* pswhst, bb, 4 lines */
-	0x01210104,		/* pswhst, k2, 4 lines */
-	0x00000000,
-	0x01250103,		/* pswhst2, bb, 3 lines */
-	0x01250103,		/* pswhst2, k2, 3 lines */
-	0x00000000,
-	0x00340101,		/* pswrd, bb, 1 lines */
-	0x00340101,		/* pswrd, k2, 1 lines */
-	0x00000000,
-	0x01280119,		/* pswrd2, bb, 25 lines */
-	0x01280119,		/* pswrd2, k2, 25 lines */
-	0x00000000,
-	0x01410109,		/* pswwr, bb, 9 lines */
-	0x01410109,		/* pswwr, k2, 9 lines */
-	0x00000000,
-	0x00000000,		/* pswwr2, bb, 0 lines */
-	0x00000000,		/* pswwr2, k2, 0 lines */
-	0x00000000,
-	0x001c0001,		/* pswrq, bb, 1 lines */
-	0x001c0001,		/* pswrq, k2, 1 lines */
-	0x00000000,
-	0x014a0015,		/* pswrq2, bb, 21 lines */
-	0x014a0015,		/* pswrq2, k2, 21 lines */
-	0x00000000,
-	0x00000000,		/* pglcs, bb, 0 lines */
-	0x00120006,		/* pglcs, k2, 6 lines */
-	0x00000000,
-	0x00100001,		/* dmae, bb, 1 lines */
-	0x00100001,		/* dmae, k2, 1 lines */
-	0x00000000,
-	0x015f0105,		/* ptu, bb, 5 lines */
-	0x015f0105,		/* ptu, k2, 5 lines */
-	0x00000000,
-	0x01640120,		/* tcm, bb, 32 lines */
-	0x01640120,		/* tcm, k2, 32 lines */
-	0x00000000,
-	0x01640120,		/* mcm, bb, 32 lines */
-	0x01640120,		/* mcm, k2, 32 lines */
-	0x00000000,
-	0x01640120,		/* ucm, bb, 32 lines */
-	0x01640120,		/* ucm, k2, 32 lines */
-	0x00000000,
-	0x01640120,		/* xcm, bb, 32 lines */
-	0x01640120,		/* xcm, k2, 32 lines */
-	0x00000000,
-	0x01640120,		/* ycm, bb, 32 lines */
-	0x01640120,		/* ycm, k2, 32 lines */
-	0x00000000,
-	0x01640120,		/* pcm, bb, 32 lines */
-	0x01640120,		/* pcm, k2, 32 lines */
-	0x00000000,
-	0x01840062,		/* qm, bb, 98 lines */
-	0x01840062,		/* qm, k2, 98 lines */
-	0x00000000,
-	0x01e60021,		/* tm, bb, 33 lines */
-	0x01e60021,		/* tm, k2, 33 lines */
-	0x00000000,
-	0x02070107,		/* dorq, bb, 7 lines */
-	0x02070107,		/* dorq, k2, 7 lines */
-	0x00000000,
-	0x00600185,		/* brb, bb, 133 lines */
-	0x00600185,		/* brb, k2, 133 lines */
-	0x00000000,
-	0x020e0019,		/* src, bb, 25 lines */
-	0x020c001a,		/* src, k2, 26 lines */
-	0x00000000,
-	0x02270104,		/* prs, bb, 4 lines */
-	0x02270104,		/* prs, k2, 4 lines */
-	0x00000000,
-	0x022b0133,		/* tsdm, bb, 51 lines */
-	0x022b0133,		/* tsdm, k2, 51 lines */
-	0x00000000,
-	0x022b0133,		/* msdm, bb, 51 lines */
-	0x022b0133,		/* msdm, k2, 51 lines */
-	0x00000000,
-	0x022b0133,		/* usdm, bb, 51 lines */
-	0x022b0133,		/* usdm, k2, 51 lines */
-	0x00000000,
-	0x022b0133,		/* xsdm, bb, 51 lines */
-	0x022b0133,		/* xsdm, k2, 51 lines */
-	0x00000000,
-	0x022b0133,		/* ysdm, bb, 51 lines */
-	0x022b0133,		/* ysdm, k2, 51 lines */
-	0x00000000,
-	0x022b0133,		/* psdm, bb, 51 lines */
-	0x022b0133,		/* psdm, k2, 51 lines */
-	0x00000000,
-	0x025e010c,		/* tsem, bb, 12 lines */
-	0x025e010c,		/* tsem, k2, 12 lines */
-	0x00000000,
-	0x025e010c,		/* msem, bb, 12 lines */
-	0x025e010c,		/* msem, k2, 12 lines */
-	0x00000000,
-	0x025e010c,		/* usem, bb, 12 lines */
-	0x025e010c,		/* usem, k2, 12 lines */
-	0x00000000,
-	0x025e010c,		/* xsem, bb, 12 lines */
-	0x025e010c,		/* xsem, k2, 12 lines */
-	0x00000000,
-	0x025e010c,		/* ysem, bb, 12 lines */
-	0x025e010c,		/* ysem, k2, 12 lines */
-	0x00000000,
-	0x025e010c,		/* psem, bb, 12 lines */
-	0x025e010c,		/* psem, k2, 12 lines */
-	0x00000000,
-	0x026a000d,		/* rss, bb, 13 lines */
-	0x026a000d,		/* rss, k2, 13 lines */
-	0x00000000,
-	0x02770106,		/* tmld, bb, 6 lines */
-	0x02770106,		/* tmld, k2, 6 lines */
-	0x00000000,
-	0x027d0106,		/* muld, bb, 6 lines */
-	0x027d0106,		/* muld, k2, 6 lines */
-	0x00000000,
-	0x02770005,		/* yuld, bb, 5 lines */
-	0x02770005,		/* yuld, k2, 5 lines */
-	0x00000000,
-	0x02830107,		/* xyld, bb, 7 lines */
-	0x027d0107,		/* xyld, k2, 7 lines */
-	0x00000000,
-	0x00000000,		/* ptld, bb, 0 lines */
-	0x00000000,		/* ptld, k2, 0 lines */
-	0x00000000,
-	0x00000000,		/* ypld, bb, 0 lines */
-	0x00000000,		/* ypld, k2, 0 lines */
-	0x00000000,
-	0x028a010e,		/* prm, bb, 14 lines */
-	0x02980110,		/* prm, k2, 16 lines */
-	0x00000000,
-	0x02a8000d,		/* pbf_pb1, bb, 13 lines */
-	0x02a8000d,		/* pbf_pb1, k2, 13 lines */
-	0x00000000,
-	0x02a8000d,		/* pbf_pb2, bb, 13 lines */
-	0x02a8000d,		/* pbf_pb2, k2, 13 lines */
-	0x00000000,
-	0x02a8000d,		/* rpb, bb, 13 lines */
-	0x02a8000d,		/* rpb, k2, 13 lines */
-	0x00000000,
-	0x00600185,		/* btb, bb, 133 lines */
-	0x00600185,		/* btb, k2, 133 lines */
-	0x00000000,
-	0x02b50117,		/* pbf, bb, 23 lines */
-	0x02b50117,		/* pbf, k2, 23 lines */
-	0x00000000,
-	0x02cc0006,		/* rdif, bb, 6 lines */
-	0x02cc0006,		/* rdif, k2, 6 lines */
-	0x00000000,
-	0x02d20006,		/* tdif, bb, 6 lines */
-	0x02d20006,		/* tdif, k2, 6 lines */
-	0x00000000,
-	0x02d80003,		/* cdu, bb, 3 lines */
-	0x02db000e,		/* cdu, k2, 14 lines */
-	0x00000000,
-	0x02e9010d,		/* ccfc, bb, 13 lines */
-	0x02f60117,		/* ccfc, k2, 23 lines */
-	0x00000000,
-	0x02e9010d,		/* tcfc, bb, 13 lines */
-	0x02f60117,		/* tcfc, k2, 23 lines */
-	0x00000000,
-	0x030d0133,		/* igu, bb, 51 lines */
-	0x030d0133,		/* igu, k2, 51 lines */
-	0x00000000,
-	0x03400106,		/* cau, bb, 6 lines */
-	0x03400106,		/* cau, k2, 6 lines */
-	0x00000000,
-	0x00000000,		/* rgfs, bb, 0 lines */
-	0x00000000,		/* rgfs, k2, 0 lines */
-	0x00000000,
-	0x00000000,		/* rgsrc, bb, 0 lines */
-	0x00000000,		/* rgsrc, k2, 0 lines */
-	0x00000000,
-	0x00000000,		/* tgfs, bb, 0 lines */
-	0x00000000,		/* tgfs, k2, 0 lines */
-	0x00000000,
-	0x00000000,		/* tgsrc, bb, 0 lines */
-	0x00000000,		/* tgsrc, k2, 0 lines */
-	0x00000000,
-	0x00000000,		/* umac, bb, 0 lines */
-	0x00120006,		/* umac, k2, 6 lines */
-	0x00000000,
-	0x00000000,		/* xmac, bb, 0 lines */
-	0x00000000,		/* xmac, k2, 0 lines */
-	0x00000000,
-	0x00000000,		/* dbg, bb, 0 lines */
-	0x00000000,		/* dbg, k2, 0 lines */
-	0x00000000,
-	0x0346012b,		/* nig, bb, 43 lines */
-	0x0346011d,		/* nig, k2, 29 lines */
-	0x00000000,
-	0x00000000,		/* wol, bb, 0 lines */
-	0x001c0002,		/* wol, k2, 2 lines */
-	0x00000000,
-	0x00000000,		/* bmbn, bb, 0 lines */
-	0x00210008,		/* bmbn, k2, 8 lines */
-	0x00000000,
-	0x00000000,		/* ipc, bb, 0 lines */
-	0x00000000,		/* ipc, k2, 0 lines */
-	0x00000000,
-	0x00000000,		/* nwm, bb, 0 lines */
-	0x0371000b,		/* nwm, k2, 11 lines */
-	0x00000000,
-	0x00000000,		/* nws, bb, 0 lines */
-	0x037c0009,		/* nws, k2, 9 lines */
-	0x00000000,
-	0x00000000,		/* ms, bb, 0 lines */
-	0x00120004,		/* ms, k2, 4 lines */
-	0x00000000,
-	0x00000000,		/* phy_pcie, bb, 0 lines */
-	0x00e5001a,		/* phy_pcie, k2, 26 lines */
-	0x00000000,
-	0x00000000,		/* led, bb, 0 lines */
-	0x00000000,		/* led, k2, 0 lines */
-	0x00000000,
-	0x00000000,		/* avs_wrap, bb, 0 lines */
-	0x00000000,		/* avs_wrap, k2, 0 lines */
-	0x00000000,
-	0x00000000,		/* bar0_map, bb, 0 lines */
-	0x00000000,		/* bar0_map, k2, 0 lines */
-	0x00000000,
-	0x00000000,		/* bar0_map, bb, 0 lines */
-	0x00000000,		/* bar0_map, k2, 0 lines */
-	0x00000000,
-};
-
 /* Win 2 */
 #define GTT_BAR0_MAP_REG_IGU_CMD	0x00f000UL
 
@@ -4086,22 +3747,28 @@ static const u32 dbg_bus_blocks[] = {
 #define GTT_BAR0_MAP_REG_MSDM_RAM_1024	0x012000UL
 
 /* Win 6 */
-#define GTT_BAR0_MAP_REG_USDM_RAM	0x013000UL
+#define GTT_BAR0_MAP_REG_MSDM_RAM_2048	0x013000UL
 
 /* Win 7 */
-#define GTT_BAR0_MAP_REG_USDM_RAM_1024	0x014000UL
+#define GTT_BAR0_MAP_REG_USDM_RAM	0x014000UL
 
 /* Win 8 */
-#define GTT_BAR0_MAP_REG_USDM_RAM_2048	0x015000UL
+#define GTT_BAR0_MAP_REG_USDM_RAM_1024	0x015000UL
 
 /* Win 9 */
-#define GTT_BAR0_MAP_REG_XSDM_RAM	0x016000UL
+#define GTT_BAR0_MAP_REG_USDM_RAM_2048	0x016000UL
 
 /* Win 10 */
-#define GTT_BAR0_MAP_REG_YSDM_RAM	0x017000UL
+#define GTT_BAR0_MAP_REG_XSDM_RAM	0x017000UL
 
 /* Win 11 */
-#define GTT_BAR0_MAP_REG_PSDM_RAM	0x018000UL
+#define GTT_BAR0_MAP_REG_XSDM_RAM_1024	0x018000UL
+
+/* Win 12 */
+#define GTT_BAR0_MAP_REG_YSDM_RAM	0x019000UL
+
+/* Win 13 */
+#define GTT_BAR0_MAP_REG_PSDM_RAM	0x01a000UL
 
 /**
  * @brief qed_qm_pf_mem_size - prepare QM ILT sizes
@@ -11913,7 +11580,7 @@ struct e4_ystorm_iscsi_conn_ag_ctx {
 /* The trace in the buffer */
 #define MFW_TRACE_EVENTID_MASK          0x00ffff
 #define MFW_TRACE_PRM_SIZE_MASK         0x0f0000
-#define MFW_TRACE_PRM_SIZE_SHIFT        16
+#define MFW_TRACE_PRM_SIZE_OFFSET	16
 #define MFW_TRACE_ENTRY_SIZE            3
 
 struct mcp_trace {
@@ -12866,7 +12533,10 @@ struct public_drv_mb {
 #define DRV_MB_PARAM_DCBX_NOTIFY_SHIFT		3
 
 #define DRV_MB_PARAM_NVM_PUT_FILE_BEGIN_MBI     0x3
+#define DRV_MB_PARAM_NVM_OFFSET_OFFSET          0
+#define DRV_MB_PARAM_NVM_OFFSET_MASK            0x00FFFFFF
 #define DRV_MB_PARAM_NVM_LEN_OFFSET		24
+#define DRV_MB_PARAM_NVM_LEN_MASK               0xFF000000
 
 #define DRV_MB_PARAM_CFG_VF_MSIX_VF_ID_SHIFT	0
 #define DRV_MB_PARAM_CFG_VF_MSIX_VF_ID_MASK	0x000000FF
@@ -13627,6 +13297,21 @@ enum nvm_image_type {
 	NVM_TYPE_FCOE_CFG = 0x1f,
 	NVM_TYPE_ETH_PHY_FW1 = 0x20,
 	NVM_TYPE_ETH_PHY_FW2 = 0x21,
+	NVM_TYPE_BDN = 0x22,
+	NVM_TYPE_8485X_PHY_FW = 0x23,
+	NVM_TYPE_PUB_KEY = 0x24,
+	NVM_TYPE_RECOVERY = 0x25,
+	NVM_TYPE_PLDM = 0x26,
+	NVM_TYPE_UPK1 = 0x27,
+	NVM_TYPE_UPK2 = 0x28,
+	NVM_TYPE_MASTER_KC = 0x29,
+	NVM_TYPE_BACKUP_KC = 0x2a,
+	NVM_TYPE_HW_DUMP = 0x2b,
+	NVM_TYPE_HW_DUMP_OUT = 0x2c,
+	NVM_TYPE_BIN_NVM_META = 0x30,
+	NVM_TYPE_ROM_TEST = 0xf0,
+	NVM_TYPE_88X33X0_PHY_FW = 0x31,
+	NVM_TYPE_88X33X0_PHY_SLAVE_FW = 0x32,
 	NVM_TYPE_MAX,
 };
 
diff --git a/drivers/net/ethernet/qlogic/qed/qed_main.c b/drivers/net/ethernet/qlogic/qed/qed_main.c
index 38f7f40b3a4d..2c189c637cca 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_main.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_main.c
@@ -2637,7 +2637,7 @@ static int qed_set_grc_config(struct qed_dev *cdev, u32 cfg_id, u32 val)
 	if (!ptt)
 		return -EAGAIN;
 
-	rc = qed_dbg_grc_config(hwfn, ptt, cfg_id, val);
+	rc = qed_dbg_grc_config(hwfn, cfg_id, val);
 
 	qed_ptt_release(hwfn, ptt);
 
diff --git a/include/linux/qed/qed_if.h b/include/linux/qed/qed_if.h
index 1b27c22d39af..8f29e0d8a7b3 100644
--- a/include/linux/qed/qed_if.h
+++ b/include/linux/qed/qed_if.h
@@ -1178,6 +1178,15 @@ struct qed_common_ops {
 #define GET_FIELD(value, name) \
 	(((value) >> (name ## _SHIFT)) & name ## _MASK)
 
+#define GET_MFW_FIELD(name, field) \
+	(((name) & (field ## _MASK)) >> (field ## _OFFSET))
+
+#define SET_MFW_FIELD(name, field, value)				 \
+	do {								 \
+		(name) &= ~(field ## _MASK);				 \
+		(name) |= (((value) << (field ## _OFFSET)) & (field ## _MASK));\
+	} while (0)
+
 #define DB_ADDR_SHIFT(addr) ((addr) << DB_PWM_ADDR_OFFSET_SHIFT)
 
 /* Debug print definitions */
-- 
cgit v1.2.3-59-g8ed1b


From 3127642dc1d16124cdf175f8235e89bc2b63f424 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <stephen@networkplumber.org>
Date: Mon, 27 Jan 2020 06:40:36 -0800
Subject: netem: change mailing list

The old netem mailing list was inactive and recently was targeted by
spammers. Switch to just using netdev mailing list which is where all
the real change happens.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 MAINTAINERS | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index da8a77384dd4..d7542d8407b2 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -11397,7 +11397,7 @@ F:	Documentation/networking/net_failover.rst
 
 NETEM NETWORK EMULATOR
 M:	Stephen Hemminger <stephen@networkplumber.org>
-L:	netem@lists.linux-foundation.org (moderated for non-subscribers)
+L:	netdev@vger.kernel.org
 S:	Maintained
 F:	net/sched/sch_netem.c
 
-- 
cgit v1.2.3-59-g8ed1b


From 6cd021a58c18a1731f7e47f83e172c0c302d65e5 Mon Sep 17 00:00:00 2001
From: Willem de Bruijn <willemb@google.com>
Date: Mon, 27 Jan 2020 15:40:31 -0500
Subject: udp: segment looped gso packets correctly

Multicast and broadcast packets can be looped from egress to ingress
pre segmentation with dev_loopback_xmit. That function unconditionally
sets ip_summed to CHECKSUM_UNNECESSARY.

udp_rcv_segment segments gso packets in the udp rx path. Segmentation
usually executes on egress, and does not expect packets of this type.
__udp_gso_segment interprets !CHECKSUM_PARTIAL as CHECKSUM_NONE. But
the offsets are not correct for gso_make_checksum.

UDP GSO packets are of type CHECKSUM_PARTIAL, with their uh->check set
to the correct pseudo header checksum. Reset ip_summed to this type.
(CHECKSUM_PARTIAL is allowed on ingress, see comments in skbuff.h)

Reported-by: syzbot <syzkaller@googlegroups.com>
Fixes: cf329aa42b66 ("udp: cope with UDP GRO packet misdirection")
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/udp.h | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/include/net/udp.h b/include/net/udp.h
index 44e0e52b585c..4a180f2a13e3 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -476,6 +476,9 @@ static inline struct sk_buff *udp_rcv_segment(struct sock *sk,
 	if (!inet_get_convert_csum(sk))
 		features |= NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM;
 
+	if (skb->pkt_type == PACKET_LOOPBACK)
+		skb->ip_summed = CHECKSUM_PARTIAL;
+
 	/* the GSO CB lays after the UDP one, no need to save and restore any
 	 * CB fragment
 	 */
-- 
cgit v1.2.3-59-g8ed1b


From f76e4c167ea2212e23c15ee7e601a865e822c291 Mon Sep 17 00:00:00 2001
From: Scott Branden <scott.branden@broadcom.com>
Date: Mon, 27 Jan 2020 16:38:28 -0800
Subject: net: phy: add default ARCH_BCM_IPROC for MDIO_BCM_IPROC

Add default MDIO_BCM_IPROC Kconfig setting such that it is default
on for IPROC family of devices.

Signed-off-by: Scott Branden <scott.branden@broadcom.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/phy/Kconfig b/drivers/net/phy/Kconfig
index 71fc778ce398..9dabe03a668c 100644
--- a/drivers/net/phy/Kconfig
+++ b/drivers/net/phy/Kconfig
@@ -38,6 +38,7 @@ config MDIO_BCM_IPROC
 	tristate "Broadcom iProc MDIO bus controller"
 	depends on ARCH_BCM_IPROC || COMPILE_TEST
 	depends on HAS_IOMEM && OF_MDIO
+	default ARCH_BCM_IPROC
 	help
 	  This module provides a driver for the MDIO busses found in the
 	  Broadcom iProc SoC's.
-- 
cgit v1.2.3-59-g8ed1b